From 8f09588b2cdc2dfb6e6076fd10708791d6c6bedd Mon Sep 17 00:00:00 2001 From: Vasanthakumar Thiagarajan Date: Mon, 23 May 2016 23:12:43 +0300 Subject: ath10k: move rx_location_info out of struct rx_pkt_end Define rx_location_info in struct rx_ppdu_end_qca99x0 after rx_pkt_end. This is to prepare rx_ppdu_end for QCA9984 chip. Signed-off-by: Vasanthakumar Thiagarajan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/rx_desc.h b/drivers/net/wireless/ath/ath10k/rx_desc.h index ca8d168..3e7dfaa 100644 --- a/drivers/net/wireless/ath/ath10k/rx_desc.h +++ b/drivers/net/wireless/ath/ath10k/rx_desc.h @@ -994,7 +994,6 @@ struct rx_pkt_end { __le32 info0; /* %RX_PKT_END_INFO0_ */ __le32 phy_timestamp_1; __le32 phy_timestamp_2; - __le32 rx_location_info; /* %RX_LOCATION_INFO_ */ } __packed; enum rx_phy_ppdu_end_info0 { @@ -1067,6 +1066,7 @@ struct rx_phy_ppdu_end { struct rx_ppdu_end_qca99x0 { struct rx_pkt_end rx_pkt_end; + __le32 rx_location_info; /* %RX_LOCATION_INFO_ */ struct rx_phy_ppdu_end rx_phy_ppdu_end; __le32 rx_timing_offset; /* %RX_PPDU_END_RX_TIMING_OFFSET_ */ __le32 rx_info; /* %RX_PPDU_END_RX_INFO_ */ -- cgit v0.10.2 From 0fc7e270523bf3757687e930c02bb46e3dcedde9 Mon Sep 17 00:00:00 2001 From: Vasanthakumar Thiagarajan Date: Mon, 23 May 2016 23:12:43 +0300 Subject: ath10k: clean up growing hw checks during safe and full reset Store pci chip secific reset funtions in struct ath10k_pci as callbacks during early ath10k_pci_probe() and use the callback to perform chip specific resets. This patch essentially adds two callback in ath10k_pci, one for doing soft reset and the other for hard reset. By using callbacks we can get rid of those hw revision checks in ath10k_pci_safe_chip_reset() and ath10k_pci_chip_reset(). As such this patch does not fix any issue. Signed-off-by: Vasanthakumar Thiagarajan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c index 8133d7b..81d6bad 100644 --- a/drivers/net/wireless/ath/ath10k/pci.c +++ b/drivers/net/wireless/ath/ath10k/pci.c @@ -2293,16 +2293,20 @@ static int ath10k_pci_warm_reset(struct ath10k *ar) return 0; } +static int ath10k_pci_qca99x0_soft_chip_reset(struct ath10k *ar) +{ + ath10k_pci_irq_disable(ar); + return ath10k_pci_qca99x0_chip_reset(ar); +} + static int ath10k_pci_safe_chip_reset(struct ath10k *ar) { - if (QCA_REV_988X(ar) || QCA_REV_6174(ar)) { - return ath10k_pci_warm_reset(ar); - } else if (QCA_REV_99X0(ar)) { - ath10k_pci_irq_disable(ar); - return ath10k_pci_qca99x0_chip_reset(ar); - } else { + struct ath10k_pci *ar_pci = ath10k_pci_priv(ar); + + if (!ar_pci->pci_soft_reset) return -ENOTSUPP; - } + + return ar_pci->pci_soft_reset(ar); } static int ath10k_pci_qca988x_chip_reset(struct ath10k *ar) @@ -2437,16 +2441,12 @@ static int ath10k_pci_qca99x0_chip_reset(struct ath10k *ar) static int ath10k_pci_chip_reset(struct ath10k *ar) { - if (QCA_REV_988X(ar)) - return ath10k_pci_qca988x_chip_reset(ar); - else if (QCA_REV_6174(ar)) - return ath10k_pci_qca6174_chip_reset(ar); - else if (QCA_REV_9377(ar)) - return ath10k_pci_qca6174_chip_reset(ar); - else if (QCA_REV_99X0(ar)) - return ath10k_pci_qca99x0_chip_reset(ar); - else + struct ath10k_pci *ar_pci = ath10k_pci_priv(ar); + + if (WARN_ON(!ar_pci->pci_hard_reset)) return -ENOTSUPP; + + return ar_pci->pci_hard_reset(ar); } static int ath10k_pci_hif_power_up(struct ath10k *ar) @@ -2976,24 +2976,34 @@ static int ath10k_pci_probe(struct pci_dev *pdev, enum ath10k_hw_rev hw_rev; u32 chip_id; bool pci_ps; + int (*pci_soft_reset)(struct ath10k *ar); + int (*pci_hard_reset)(struct ath10k *ar); switch (pci_dev->device) { case QCA988X_2_0_DEVICE_ID: hw_rev = ATH10K_HW_QCA988X; pci_ps = false; + pci_soft_reset = ath10k_pci_warm_reset; + pci_hard_reset = ath10k_pci_qca988x_chip_reset; break; case QCA6164_2_1_DEVICE_ID: case QCA6174_2_1_DEVICE_ID: hw_rev = ATH10K_HW_QCA6174; pci_ps = true; + pci_soft_reset = ath10k_pci_warm_reset; + pci_hard_reset = ath10k_pci_qca6174_chip_reset; break; case QCA99X0_2_0_DEVICE_ID: hw_rev = ATH10K_HW_QCA99X0; pci_ps = false; + pci_soft_reset = ath10k_pci_qca99x0_soft_chip_reset; + pci_hard_reset = ath10k_pci_qca99x0_chip_reset; break; case QCA9377_1_0_DEVICE_ID: hw_rev = ATH10K_HW_QCA9377; pci_ps = true; + pci_soft_reset = NULL; + pci_hard_reset = ath10k_pci_qca6174_chip_reset; break; default: WARN_ON(1); @@ -3018,6 +3028,8 @@ static int ath10k_pci_probe(struct pci_dev *pdev, ar->dev_id = pci_dev->device; ar_pci->pci_ps = pci_ps; ar_pci->bus_ops = &ath10k_pci_bus_ops; + ar_pci->pci_soft_reset = pci_soft_reset; + ar_pci->pci_hard_reset = pci_hard_reset; ar->id.vendor = pdev->vendor; ar->id.device = pdev->device; diff --git a/drivers/net/wireless/ath/ath10k/pci.h b/drivers/net/wireless/ath/ath10k/pci.h index 959dc32..6eca1df 100644 --- a/drivers/net/wireless/ath/ath10k/pci.h +++ b/drivers/net/wireless/ath/ath10k/pci.h @@ -234,6 +234,12 @@ struct ath10k_pci { const struct ath10k_bus_ops *bus_ops; + /* Chip specific pci reset routine used to do a safe reset */ + int (*pci_soft_reset)(struct ath10k *ar); + + /* Chip specific pci full reset function */ + int (*pci_hard_reset)(struct ath10k *ar); + /* Keep this entry in the last, memory for struct ath10k_ahb is * allocated (ahb support enabled case) in the continuation of * this struct. -- cgit v0.10.2 From acc6b5593eee9386b209dd29da1e4bdbba99cf9a Mon Sep 17 00:00:00 2001 From: Vasanthakumar Thiagarajan Date: Mon, 23 May 2016 23:12:44 +0300 Subject: ath10k: define rx_ppdu_end for QCA9984 QCA9984 Rx descriptor has two 32-bit words of location information when compared to one 32-bit word in QCA99X0. To handle this difference in rx descriptor ppdu_end, define a new ppdu_end for QCA9984 descriptor which has the new structure to represent rx_location_info. Signed-off-by: Vasanthakumar Thiagarajan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/rx_desc.h b/drivers/net/wireless/ath/ath10k/rx_desc.h index 3e7dfaa..9ceebea 100644 --- a/drivers/net/wireless/ath/ath10k/rx_desc.h +++ b/drivers/net/wireless/ath/ath10k/rx_desc.h @@ -996,6 +996,41 @@ struct rx_pkt_end { __le32 phy_timestamp_2; } __packed; +#define RX_LOCATION_INFO0_RTT_FAC_LEGACY_MASK 0x00003fff +#define RX_LOCATION_INFO0_RTT_FAC_LEGACY_LSB 0 +#define RX_LOCATION_INFO0_RTT_FAC_VHT_MASK 0x1fff8000 +#define RX_LOCATION_INFO0_RTT_FAC_VHT_LSB 15 +#define RX_LOCATION_INFO0_RTT_STRONGEST_CHAIN_MASK 0xc0000000 +#define RX_LOCATION_INFO0_RTT_STRONGEST_CHAIN_LSB 30 +#define RX_LOCATION_INFO0_RTT_FAC_LEGACY_STATUS BIT(14) +#define RX_LOCATION_INFO0_RTT_FAC_VHT_STATUS BIT(29) + +#define RX_LOCATION_INFO1_RTT_PREAMBLE_TYPE_MASK 0x0000000c +#define RX_LOCATION_INFO1_RTT_PREAMBLE_TYPE_LSB 2 +#define RX_LOCATION_INFO1_PKT_BW_MASK 0x00000030 +#define RX_LOCATION_INFO1_PKT_BW_LSB 4 +#define RX_LOCATION_INFO1_SKIP_P_SKIP_BTCF_MASK 0x0000ff00 +#define RX_LOCATION_INFO1_SKIP_P_SKIP_BTCF_LSB 8 +#define RX_LOCATION_INFO1_RTT_MSC_RATE_MASK 0x000f0000 +#define RX_LOCATION_INFO1_RTT_MSC_RATE_LSB 16 +#define RX_LOCATION_INFO1_RTT_PBD_LEG_BW_MASK 0x00300000 +#define RX_LOCATION_INFO1_RTT_PBD_LEG_BW_LSB 20 +#define RX_LOCATION_INFO1_TIMING_BACKOFF_MASK 0x07c00000 +#define RX_LOCATION_INFO1_TIMING_BACKOFF_LSB 22 +#define RX_LOCATION_INFO1_RTT_TX_FRAME_PHASE_MASK 0x18000000 +#define RX_LOCATION_INFO1_RTT_TX_FRAME_PHASE_LSB 27 +#define RX_LOCATION_INFO1_RTT_CFR_STATUS BIT(0) +#define RX_LOCATION_INFO1_RTT_CIR_STATUS BIT(1) +#define RX_LOCATION_INFO1_RTT_GI_TYPE BIT(7) +#define RX_LOCATION_INFO1_RTT_MAC_PHY_PHASE BIT(29) +#define RX_LOCATION_INFO1_RTT_TX_DATA_START_X_PHASE BIT(30) +#define RX_LOCATION_INFO1_RX_LOCATION_VALID BIT(31) + +struct rx_location_info { + __le32 rx_location_info0; /* %RX_LOCATION_INFO0_ */ + __le32 rx_location_info1; /* %RX_LOCATION_INFO1_ */ +} __packed; + enum rx_phy_ppdu_end_info0 { RX_PHY_PPDU_END_INFO0_ERR_RADAR = BIT(2), RX_PHY_PPDU_END_INFO0_ERR_RX_ABORT = BIT(3), @@ -1074,12 +1109,23 @@ struct rx_ppdu_end_qca99x0 { __le16 info1; /* %RX_PPDU_END_INFO1_ */ } __packed; +struct rx_ppdu_end_qca9984 { + struct rx_pkt_end rx_pkt_end; + struct rx_location_info rx_location_info; + struct rx_phy_ppdu_end rx_phy_ppdu_end; + __le32 rx_timing_offset; /* %RX_PPDU_END_RX_TIMING_OFFSET_ */ + __le32 rx_info; /* %RX_PPDU_END_RX_INFO_ */ + __le16 bb_length; + __le16 info1; /* %RX_PPDU_END_INFO1_ */ +} __packed; + struct rx_ppdu_end { struct rx_ppdu_end_common common; union { struct rx_ppdu_end_qca988x qca988x; struct rx_ppdu_end_qca6174 qca6174; struct rx_ppdu_end_qca99x0 qca99x0; + struct rx_ppdu_end_qca9984 qca9984; } __packed; } __packed; -- cgit v0.10.2 From 651b4cdcf97e75f6346784b75ca7bf3c85187143 Mon Sep 17 00:00:00 2001 From: Vasanthakumar Thiagarajan Date: Mon, 23 May 2016 23:12:45 +0300 Subject: ath10k: enable support for QCA9984 QCA9984 shares the same configuration with QCA99X0. Signed-off-by: Vasanthakumar Thiagarajan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index 49af624..a003980 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -163,6 +163,28 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = { }, }, { + .id = QCA9984_HW_1_0_DEV_VERSION, + .dev_id = QCA9984_1_0_DEVICE_ID, + .name = "qca9984/qca9994 hw1.0", + .patch_load_addr = QCA9984_HW_1_0_PATCH_LOAD_ADDR, + .uart_pin = 7, + .otp_exe_param = 0x00000700, + .continuous_frag_desc = true, + .channel_counters_freq_hz = 150000, + .max_probe_resp_desc_thres = 24, + .hw_4addr_pad = ATH10K_HW_4ADDR_PAD_BEFORE, + .tx_chain_mask = 0xf, + .rx_chain_mask = 0xf, + .max_spatial_stream = 4, + .cal_data_len = 12064, + .fw = { + .dir = QCA9984_HW_1_0_FW_DIR, + .board = QCA9984_HW_1_0_BOARD_DATA_FILE, + .board_size = QCA99X0_BOARD_DATA_SZ, + .board_ext_size = QCA99X0_BOARD_EXT_DATA_SZ, + }, + }, + { .id = QCA9377_HW_1_0_DEV_VERSION, .dev_id = QCA9377_1_0_DEVICE_ID, .name = "qca9377 hw1.0", @@ -2071,6 +2093,7 @@ struct ath10k *ath10k_core_create(size_t priv_size, struct device *dev, ar->hw_values = &qca6174_values; break; case ATH10K_HW_QCA99X0: + case ATH10K_HW_QCA9984: ar->regs = &qca99x0_regs; ar->hw_values = &qca99x0_values; break; diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h index aedd898..f41c91c 100644 --- a/drivers/net/wireless/ath/ath10k/hw.h +++ b/drivers/net/wireless/ath/ath10k/hw.h @@ -26,6 +26,7 @@ #define QCA6164_2_1_DEVICE_ID (0x0041) #define QCA6174_2_1_DEVICE_ID (0x003e) #define QCA99X0_2_0_DEVICE_ID (0x0040) +#define QCA9984_1_0_DEVICE_ID (0x0046) #define QCA9377_1_0_DEVICE_ID (0x0042) /* QCA988X 1.0 definitions (unsupported) */ @@ -91,6 +92,14 @@ enum qca9377_chip_id_rev { #define QCA99X0_HW_2_0_BOARD_DATA_FILE "board.bin" #define QCA99X0_HW_2_0_PATCH_LOAD_ADDR 0x1234 +/* QCA9984 1.0 defines */ +#define QCA9984_HW_1_0_DEV_VERSION 0x1000000 +#define QCA9984_HW_DEV_TYPE 0xa +#define QCA9984_HW_1_0_CHIP_ID_REV 0x0 +#define QCA9984_HW_1_0_FW_DIR ATH10K_FW_DIR "/QCA9984/hw1.0" +#define QCA9984_HW_1_0_BOARD_DATA_FILE "board.bin" +#define QCA9984_HW_1_0_PATCH_LOAD_ADDR 0x1234 + /* QCA9377 1.0 definitions */ #define QCA9377_HW_1_0_FW_DIR ATH10K_FW_DIR "/QCA9377/hw1.0" #define QCA9377_HW_1_0_BOARD_DATA_FILE "board.bin" @@ -193,6 +202,7 @@ enum ath10k_hw_rev { ATH10K_HW_QCA988X, ATH10K_HW_QCA6174, ATH10K_HW_QCA99X0, + ATH10K_HW_QCA9984, ATH10K_HW_QCA9377, ATH10K_HW_QCA4019, }; @@ -249,6 +259,7 @@ void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey, #define QCA_REV_988X(ar) ((ar)->hw_rev == ATH10K_HW_QCA988X) #define QCA_REV_6174(ar) ((ar)->hw_rev == ATH10K_HW_QCA6174) #define QCA_REV_99X0(ar) ((ar)->hw_rev == ATH10K_HW_QCA99X0) +#define QCA_REV_9984(ar) ((ar)->hw_rev == ATH10K_HW_QCA9984) #define QCA_REV_9377(ar) ((ar)->hw_rev == ATH10K_HW_QCA9377) #define QCA_REV_40XX(ar) ((ar)->hw_rev == ATH10K_HW_QCA4019) diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c index 81d6bad..8e8e1eb 100644 --- a/drivers/net/wireless/ath/ath10k/pci.c +++ b/drivers/net/wireless/ath/ath10k/pci.c @@ -56,6 +56,7 @@ static const struct pci_device_id ath10k_pci_id_table[] = { { PCI_VDEVICE(ATHEROS, QCA6164_2_1_DEVICE_ID) }, /* PCI-E QCA6164 V2.1 */ { PCI_VDEVICE(ATHEROS, QCA6174_2_1_DEVICE_ID) }, /* PCI-E QCA6174 V2.1 */ { PCI_VDEVICE(ATHEROS, QCA99X0_2_0_DEVICE_ID) }, /* PCI-E QCA99X0 V2 */ + { PCI_VDEVICE(ATHEROS, QCA9984_1_0_DEVICE_ID) }, /* PCI-E QCA9984 V1 */ { PCI_VDEVICE(ATHEROS, QCA9377_1_0_DEVICE_ID) }, /* PCI-E QCA9377 V1 */ {0} }; @@ -81,8 +82,11 @@ static const struct ath10k_pci_supp_chip ath10k_pci_supp_chips[] = { { QCA99X0_2_0_DEVICE_ID, QCA99X0_HW_2_0_CHIP_ID_REV }, + { QCA9984_1_0_DEVICE_ID, QCA9984_HW_1_0_CHIP_ID_REV }, + { QCA9377_1_0_DEVICE_ID, QCA9377_HW_1_0_CHIP_ID_REV }, { QCA9377_1_0_DEVICE_ID, QCA9377_HW_1_1_CHIP_ID_REV }, + }; static void ath10k_pci_buffer_cleanup(struct ath10k *ar); @@ -844,6 +848,7 @@ static u32 ath10k_pci_targ_cpu_to_ce_addr(struct ath10k *ar, u32 addr) 0x7ff) << 21; break; case ATH10K_HW_QCA99X0: + case ATH10K_HW_QCA9984: case ATH10K_HW_QCA4019: val = ath10k_pci_read32(ar, PCIE_BAR_REG_ADDRESS); break; @@ -1569,6 +1574,7 @@ static void ath10k_pci_irq_msi_fw_mask(struct ath10k *ar) CORE_CTRL_ADDRESS, val); break; case ATH10K_HW_QCA99X0: + case ATH10K_HW_QCA9984: case ATH10K_HW_QCA4019: /* TODO: Find appropriate register configuration for QCA99X0 * to mask irq/MSI. @@ -1592,6 +1598,7 @@ static void ath10k_pci_irq_msi_fw_unmask(struct ath10k *ar) CORE_CTRL_ADDRESS, val); break; case ATH10K_HW_QCA99X0: + case ATH10K_HW_QCA9984: case ATH10K_HW_QCA4019: /* TODO: Find appropriate register configuration for QCA99X0 * to unmask irq/MSI. @@ -1932,6 +1939,7 @@ static int ath10k_pci_get_num_banks(struct ath10k *ar) switch (ar_pci->pdev->device) { case QCA988X_2_0_DEVICE_ID: case QCA99X0_2_0_DEVICE_ID: + case QCA9984_1_0_DEVICE_ID: return 1; case QCA6164_2_1_DEVICE_ID: case QCA6174_2_1_DEVICE_ID: @@ -2999,6 +3007,12 @@ static int ath10k_pci_probe(struct pci_dev *pdev, pci_soft_reset = ath10k_pci_qca99x0_soft_chip_reset; pci_hard_reset = ath10k_pci_qca99x0_chip_reset; break; + case QCA9984_1_0_DEVICE_ID: + hw_rev = ATH10K_HW_QCA9984; + pci_ps = false; + pci_soft_reset = ath10k_pci_qca99x0_soft_chip_reset; + pci_hard_reset = ath10k_pci_qca99x0_chip_reset; + break; case QCA9377_1_0_DEVICE_ID: hw_rev = ATH10K_HW_QCA9377; pci_ps = true; -- cgit v0.10.2 From 7a0adc83f34ddbb974b01e41890498af4f9a5725 Mon Sep 17 00:00:00 2001 From: Michal Kazior Date: Mon, 23 May 2016 23:12:45 +0300 Subject: ath10k: improve tx scheduling Recent changes revolving around implementing wake_tx_queue support introduced a significant performance regressions on some (slower, uni-proc) systems. Signed-off-by: Michal Kazior Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c index cc979a4..7411b69 100644 --- a/drivers/net/wireless/ath/ath10k/htt_rx.c +++ b/drivers/net/wireless/ath/ath10k/htt_rx.c @@ -2291,7 +2291,6 @@ bool ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb) ath10k_htt_tx_mgmt_dec_pending(htt); spin_unlock_bh(&htt->tx_lock); } - ath10k_mac_tx_push_pending(ar); break; } case HTT_T2H_MSG_TYPE_TX_COMPL_IND: @@ -2442,8 +2441,6 @@ static void ath10k_htt_txrx_compl_task(unsigned long ptr) dev_kfree_skb_any(skb); } - ath10k_mac_tx_push_pending(ar); - num_mpdus = atomic_read(&htt->num_mpdus_ready); while (num_mpdus) { diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index 6dd1d26..1dd415d 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -3781,6 +3781,9 @@ void ath10k_mac_tx_push_pending(struct ath10k *ar) int ret; int max; + if (ar->htt.num_pending_tx >= (ar->htt.max_num_pending_tx / 2)) + return; + spin_lock_bh(&ar->txqs_lock); rcu_read_lock(); @@ -4051,9 +4054,7 @@ static void ath10k_mac_op_wake_tx_queue(struct ieee80211_hw *hw, list_add_tail(&artxq->list, &ar->txqs); spin_unlock_bh(&ar->txqs_lock); - if (ath10k_mac_tx_can_push(hw, txq)) - tasklet_schedule(&ar->htt.txrx_compl_task); - + ath10k_mac_tx_push_pending(ar); ath10k_htt_tx_txq_update(hw, txq); } diff --git a/drivers/net/wireless/ath/ath10k/txrx.c b/drivers/net/wireless/ath/ath10k/txrx.c index 576e7c4..1966c78 100644 --- a/drivers/net/wireless/ath/ath10k/txrx.c +++ b/drivers/net/wireless/ath/ath10k/txrx.c @@ -117,6 +117,9 @@ int ath10k_txrx_tx_unref(struct ath10k_htt *htt, ieee80211_tx_status(htt->ar->hw, msdu); /* we do not own the msdu anymore */ + + ath10k_mac_tx_push_pending(ar); + return 0; } -- cgit v0.10.2 From 3b0499e9ce420e8841d5007130b5e074eb2e674b Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Mon, 23 May 2016 23:12:46 +0300 Subject: ath10k: reduce warning messages during rx without proper channel context WARN_ON_ONCE when we receive packets for self peer when mac80211 had not assigned a proper channel context. This scenario happens in QCA4019 when we start the AP via hostapd in background and start it once again in the background without killing the previous instance! This happens intermittently when we start / stop hostapd in a while loop (incase the hostapd is not properly killed). This results in mac80211 chancontext to be unassigned, while the self peer continuous receive packets in target operating channel. This results in lot of call traces in the rx path. Make this as a WARN_ON_ONCE to avoid flooding the console which result in rebooting low memory systems, while still reporting the warning once that we are receiving packets in target operating channel and to indicate that something is happening which is not the expected result. WARNING: CPU: 0 PID: 0 at ath/ath10k/htt_rx.c:803 [] (warn_slowpath_null) from [] (ath10k_htt_rx_h_channel+0xe0/0x1b8 [ath10k_core]) [] (ath10k_htt_rx_h_channel [ath10k_core]) from [] (ath10k_htt_rx_h_ppdu+0x80/0x288 [ath10k_core]) [] (ath10k_htt_rx_h_ppdu [ath10k_core]) from [] (ath10k_htt_txrx_compl_task+0x724/0x9d4 [ath10k_core]) [] (ath10k_htt_txrx_compl_task [ath10k_core]) Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c index 7411b69..3e556e0 100644 --- a/drivers/net/wireless/ath/ath10k/htt_rx.c +++ b/drivers/net/wireless/ath/ath10k/htt_rx.c @@ -748,7 +748,7 @@ ath10k_htt_rx_h_peer_channel(struct ath10k *ar, struct htt_rx_desc *rxd) if (WARN_ON_ONCE(!arvif)) return NULL; - if (WARN_ON(ath10k_mac_vif_chan(arvif->vif, &def))) + if (WARN_ON_ONCE(ath10k_mac_vif_chan(arvif->vif, &def))) return NULL; return def.chan; -- cgit v0.10.2 From 026441c9d4ff2651106d0e57e738a5afb742c338 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Mon, 23 May 2016 23:12:47 +0300 Subject: ath10k: fix legacy rate packet debug messages Legacy rate packets may not necessarily be having a rx status flag of '0' always, for example management frame have flags like RX_FLAG_ONLY_MONITOR / RX_FLAG_MACTIME_END also set Just check 'VHT' and 'HT' flags are not set , and simply clasify it as legacy rate packets Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c index 3e556e0..7bf1909 100644 --- a/drivers/net/wireless/ath/ath10k/htt_rx.c +++ b/drivers/net/wireless/ath/ath10k/htt_rx.c @@ -939,7 +939,8 @@ static void ath10k_process_rx(struct ath10k *ar, is_multicast_ether_addr(ieee80211_get_DA(hdr)) ? "mcast" : "ucast", (__le16_to_cpu(hdr->seq_ctrl) & IEEE80211_SCTL_SEQ) >> 4, - status->flag == 0 ? "legacy" : "", + (status->flag & (RX_FLAG_HT | RX_FLAG_VHT)) == 0 ? + "legacy" : "", status->flag & RX_FLAG_HT ? "ht" : "", status->flag & RX_FLAG_VHT ? "vht" : "", status->flag & RX_FLAG_40MHZ ? "40" : "", -- cgit v0.10.2 From ceda5153d8cde4b05df1db33e8d76394949d616c Mon Sep 17 00:00:00 2001 From: Eduardo Abinader Date: Mon, 2 May 2016 17:44:11 +0200 Subject: ath9k: Remove empty test condition Just some code cleanup to remove an empty if clause. Signed-off-by: Eduardo Abinader Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c b/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c index dec1a317a..d0224fc 100644 --- a/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c +++ b/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c @@ -3202,8 +3202,7 @@ static int ar9300_compress_decision(struct ath_hw *ah, it, length); break; case _CompressBlock: - if (reference == 0) { - } else { + if (reference != 0) { eep = ar9003_eeprom_struct_find_by_id(reference); if (eep == NULL) { ath_dbg(common, EEPROM, -- cgit v0.10.2 From e94610cc1c8f2cc1154c941cbc2d91723f85b60c Mon Sep 17 00:00:00 2001 From: Eduardo Abinader Date: Mon, 9 May 2016 14:38:40 +0200 Subject: ath9k: allow tx99 for ar9002 based cards As there is current support for ar9002 tx99 mode, just allow to init debugfs and enable tx99. Signed-off-by: Eduardo Abinader Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/tx99.c b/drivers/net/wireless/ath/ath9k/tx99.c index ac4781f..6161536 100644 --- a/drivers/net/wireless/ath/ath9k/tx99.c +++ b/drivers/net/wireless/ath/ath9k/tx99.c @@ -266,7 +266,7 @@ static const struct file_operations fops_tx99_power = { void ath9k_tx99_init_debug(struct ath_softc *sc) { - if (!AR_SREV_9300_20_OR_LATER(sc->sc_ah)) + if (!AR_SREV_9280_20_OR_LATER(sc->sc_ah)) return; debugfs_create_file("tx99", S_IRUSR | S_IWUSR, -- cgit v0.10.2 From eb26cff148f5449972121e46e403f549d71f6f49 Mon Sep 17 00:00:00 2001 From: Maya Erez Date: Mon, 16 May 2016 22:23:30 +0300 Subject: wil6210: fix race conditions between TX send and completion There are 2 possible race conditions, both are solved by addition of memory barrier: 1. wil_tx_complete reads the swhead to determine if the vring is empty. In case the swhead was updated before the descriptor update was performed in __wil_tx_vring/__wil_tx_vring_tso, the completion loop will not end and as the DU bit may still be set from a previous run, this skb can be handled as completed before it was sent, which will lead to double free of the same SKB. 2. __wil_tx_vring/__wil_tx_vring_tso calculate the number of available descriptors according to the swtail. In case the swtail is updated before memset of ctx to zero is completed, we can handle this descriptor while later on ctx is zeroed. Signed-off-by: Maya Erez Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c index a4e4379..fa6ea24 100644 --- a/drivers/net/wireless/ath/wil6210/txrx.c +++ b/drivers/net/wireless/ath/wil6210/txrx.c @@ -1551,6 +1551,13 @@ static int __wil_tx_vring_tso(struct wil6210_priv *wil, struct vring *vring, vring_index, used, used + descs_used); } + /* Make sure to advance the head only after descriptor update is done. + * This will prevent a race condition where the completion thread + * will see the DU bit set from previous run and will handle the + * skb before it was completed. + */ + wmb(); + /* advance swhead */ wil_vring_advance_head(vring, descs_used); wil_dbg_txrx(wil, "TSO: Tx swhead %d -> %d\n", swhead, vring->swhead); @@ -1691,6 +1698,13 @@ static int __wil_tx_vring(struct wil6210_priv *wil, struct vring *vring, vring_index, used, used + nr_frags + 1); } + /* Make sure to advance the head only after descriptor update is done. + * This will prevent a race condition where the completion thread + * will see the DU bit set from previous run and will handle the + * skb before it was completed. + */ + wmb(); + /* advance swhead */ wil_vring_advance_head(vring, nr_frags + 1); wil_dbg_txrx(wil, "Tx[%2d] swhead %d -> %d\n", vring_index, swhead, @@ -1914,6 +1928,12 @@ int wil_tx_complete(struct wil6210_priv *wil, int ringid) wil_consume_skb(skb, d->dma.error == 0); } memset(ctx, 0, sizeof(*ctx)); + /* Make sure the ctx is zeroed before updating the tail + * to prevent a case where wil_tx_vring will see + * this descriptor as used and handle it before ctx zero + * is completed. + */ + wmb(); /* There is no need to touch HW descriptor: * - ststus bit TX_DMA_STATUS_DU is set by design, * so hardware will not try to process this desc., -- cgit v0.10.2 From ab6d7cc3eab4093caf91ba8b27590c4080d7d01c Mon Sep 17 00:00:00 2001 From: Maya Erez Date: Mon, 16 May 2016 22:23:31 +0300 Subject: wil6210: guarantee safe access to rx descriptors shared memory add memory barrier after allocating new rx descriptors, before updating the hwtail. This will guarantee that all writes to descriptors (shared memory) are done before committing them to HW. Signed-off-by: Maya Erez Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c index fa6ea24..3909af1 100644 --- a/drivers/net/wireless/ath/wil6210/txrx.c +++ b/drivers/net/wireless/ath/wil6210/txrx.c @@ -544,6 +544,12 @@ static int wil_rx_refill(struct wil6210_priv *wil, int count) break; } } + + /* make sure all writes to descriptors (shared memory) are done before + * committing them to HW + */ + wmb(); + wil_w(wil, v->hwtail, v->swtail); return rc; -- cgit v0.10.2 From 34b8886e502a62d1355ccc0420044aa2749a24cd Mon Sep 17 00:00:00 2001 From: Maya Erez Date: Mon, 16 May 2016 22:23:32 +0300 Subject: wil6210: protect wil_vring_fini_tx in parallel to tx completions napi_synchronize is called before releasing the vring, with the assumption that setting txdata->enabled to 0 will prevent handling of this vring in the next scheduled napi. To guarantee this assumption, a memory barrier is added after disabling the txdata. In addition, as the ctx is zeroed in wil_tx_complete after this descriptor is handled (protected by wmb), ctx needs to be checked before releasing this descriptor in wil_vring_free. Signed-off-by: Maya Erez Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c index 3909af1..483e063 100644 --- a/drivers/net/wireless/ath/wil6210/txrx.c +++ b/drivers/net/wireless/ath/wil6210/txrx.c @@ -184,6 +184,13 @@ static void wil_vring_free(struct wil6210_priv *wil, struct vring *vring, &vring->va[vring->swtail].tx; ctx = &vring->ctx[vring->swtail]; + if (!ctx) { + wil_dbg_txrx(wil, + "ctx(%d) was already completed\n", + vring->swtail); + vring->swtail = wil_vring_next_tail(vring); + continue; + } *d = *_d; wil_txdesc_unmap(dev, d, ctx); if (ctx->skb) @@ -975,6 +982,13 @@ void wil_vring_fini_tx(struct wil6210_priv *wil, int id) txdata->dot1x_open = false; txdata->enabled = 0; /* no Tx can be in progress or start anew */ spin_unlock_bh(&txdata->lock); + /* napi_synchronize waits for completion of the current NAPI but will + * not prevent the next NAPI run. + * Add a memory barrier to guarantee that txdata->enabled is zeroed + * before napi_synchronize so that the next scheduled NAPI will not + * handle this vring + */ + wmb(); /* make sure NAPI won't touch this vring */ if (test_bit(wil_status_napi_en, wil->status)) napi_synchronize(&wil->napi_tx); -- cgit v0.10.2 From a1526f7eafa434f756579e2bc784b7605f96bf3e Mon Sep 17 00:00:00 2001 From: Maya Erez Date: Mon, 16 May 2016 22:23:33 +0300 Subject: wil6210: fix dma mapping error cleanup in __wil_tx_vring_tso In case we fail to map one of the TSO SKB fragments, we need to clear all the mapped descriptors, from swhead to swhead+descs_used-1. Change the desc index calculation to i = (swhead + descs_used - 1) % vring->size; to prevent unmpping of (swhead + descs_used) descriptor that wasn't mapped. Signed-off-by: Maya Erez Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wil6210/txrx.c b/drivers/net/wireless/ath/wil6210/txrx.c index 483e063..f2f6a40 100644 --- a/drivers/net/wireless/ath/wil6210/txrx.c +++ b/drivers/net/wireless/ath/wil6210/txrx.c @@ -1594,7 +1594,7 @@ mem_error: while (descs_used > 0) { struct wil_ctx *ctx; - i = (swhead + descs_used) % vring->size; + i = (swhead + descs_used - 1) % vring->size; d = (struct vring_tx_desc *)&vring->va[i].tx; _desc = &vring->va[i].tx; *d = *_desc; -- cgit v0.10.2 From e34dc6475a7b25d1aec5de8652a321672904c686 Mon Sep 17 00:00:00 2001 From: Maya Erez Date: Mon, 16 May 2016 22:23:34 +0300 Subject: wil6210: add pm_notify handling Adding pm_notify to allow the following: 1. Check if suspend is allowed in an earlier stage to prevent starting the suspend procedure in case it is not allowed 2. Notify the platform driver on the suspend request Signed-off-by: Maya Erez Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wil6210/pcie_bus.c b/drivers/net/wireless/ath/wil6210/pcie_bus.c index aeb72c4..7b5c422 100644 --- a/drivers/net/wireless/ath/wil6210/pcie_bus.c +++ b/drivers/net/wireless/ath/wil6210/pcie_bus.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2012-2015 Qualcomm Atheros, Inc. + * Copyright (c) 2012-2016 Qualcomm Atheros, Inc. * * Permission to use, copy, modify, and/or distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -18,13 +18,20 @@ #include #include #include - +#include #include "wil6210.h" static bool use_msi = true; module_param(use_msi, bool, S_IRUGO); MODULE_PARM_DESC(use_msi, " Use MSI interrupt, default - true"); +#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP +static int wil6210_pm_notify(struct notifier_block *notify_block, + unsigned long mode, void *unused); +#endif /* CONFIG_PM_SLEEP */ +#endif /* CONFIG_PM */ + static void wil_set_capabilities(struct wil6210_priv *wil) { @@ -238,6 +245,18 @@ static int wil_pcie_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto bus_disable; } +#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP + wil->pm_notify.notifier_call = wil6210_pm_notify; + rc = register_pm_notifier(&wil->pm_notify); + if (rc) + /* Do not fail the driver initialization, as suspend can + * be prevented in a later phase if needed + */ + wil_err(wil, "register_pm_notifier failed: %d\n", rc); +#endif /* CONFIG_PM_SLEEP */ +#endif /* CONFIG_PM */ + wil6210_debugfs_init(wil); @@ -267,6 +286,12 @@ static void wil_pcie_remove(struct pci_dev *pdev) wil_dbg_misc(wil, "%s()\n", __func__); +#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP + unregister_pm_notifier(&wil->pm_notify); +#endif /* CONFIG_PM_SLEEP */ +#endif /* CONFIG_PM */ + wil6210_debugfs_remove(wil); wil_if_remove(wil); wil_if_pcie_disable(wil); @@ -335,6 +360,45 @@ static int wil6210_resume(struct device *dev, bool is_runtime) return rc; } +static int wil6210_pm_notify(struct notifier_block *notify_block, + unsigned long mode, void *unused) +{ + struct wil6210_priv *wil = container_of( + notify_block, struct wil6210_priv, pm_notify); + int rc = 0; + enum wil_platform_event evt; + + wil_dbg_pm(wil, "%s: mode (%ld)\n", __func__, mode); + + switch (mode) { + case PM_HIBERNATION_PREPARE: + case PM_SUSPEND_PREPARE: + case PM_RESTORE_PREPARE: + rc = wil_can_suspend(wil, false); + if (rc) + break; + evt = WIL_PLATFORM_EVT_PRE_SUSPEND; + if (wil->platform_ops.notify) + rc = wil->platform_ops.notify(wil->platform_handle, + evt); + break; + case PM_POST_SUSPEND: + case PM_POST_HIBERNATION: + case PM_POST_RESTORE: + evt = WIL_PLATFORM_EVT_POST_SUSPEND; + if (wil->platform_ops.notify) + rc = wil->platform_ops.notify(wil->platform_handle, + evt); + break; + default: + wil_dbg_pm(wil, "unhandled notify mode %ld\n", mode); + break; + } + + wil_dbg_pm(wil, "notification mode %ld: rc (%d)\n", mode, rc); + return rc; +} + static int wil6210_pm_suspend(struct device *dev) { return wil6210_suspend(dev, false); diff --git a/drivers/net/wireless/ath/wil6210/pm.c b/drivers/net/wireless/ath/wil6210/pm.c index 0b7ecbc..11ee24d 100644 --- a/drivers/net/wireless/ath/wil6210/pm.c +++ b/drivers/net/wireless/ath/wil6210/pm.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2014 Qualcomm Atheros, Inc. + * Copyright (c) 2014,2016 Qualcomm Atheros, Inc. * * Permission to use, copy, modify, and/or distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -24,10 +24,32 @@ int wil_can_suspend(struct wil6210_priv *wil, bool is_runtime) wil_dbg_pm(wil, "%s(%s)\n", __func__, is_runtime ? "runtime" : "system"); + if (!netif_running(wil_to_ndev(wil))) { + /* can always sleep when down */ + wil_dbg_pm(wil, "Interface is down\n"); + goto out; + } + if (test_bit(wil_status_resetting, wil->status)) { + wil_dbg_pm(wil, "Delay suspend when resetting\n"); + rc = -EBUSY; + goto out; + } + if (wil->recovery_state != fw_recovery_idle) { + wil_dbg_pm(wil, "Delay suspend during recovery\n"); + rc = -EBUSY; + goto out; + } + + /* interface is running */ switch (wdev->iftype) { case NL80211_IFTYPE_MONITOR: case NL80211_IFTYPE_STATION: case NL80211_IFTYPE_P2P_CLIENT: + if (test_bit(wil_status_fwconnecting, wil->status)) { + wil_dbg_pm(wil, "Delay suspend when connecting\n"); + rc = -EBUSY; + goto out; + } break; /* AP-like interface - can't suspend */ default: @@ -36,6 +58,7 @@ int wil_can_suspend(struct wil6210_priv *wil, bool is_runtime) break; } +out: wil_dbg_pm(wil, "%s(%s) => %s (%d)\n", __func__, is_runtime ? "runtime" : "system", rc ? "No" : "Yes", rc); diff --git a/drivers/net/wireless/ath/wil6210/wil6210.h b/drivers/net/wireless/ath/wil6210/wil6210.h index aa09cbc..1feaf5a 100644 --- a/drivers/net/wireless/ath/wil6210/wil6210.h +++ b/drivers/net/wireless/ath/wil6210/wil6210.h @@ -662,6 +662,11 @@ struct wil6210_priv { /* High Access Latency Policy voting */ struct wil_halp halp; +#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP + struct notifier_block pm_notify; +#endif /* CONFIG_PM_SLEEP */ +#endif /* CONFIG_PM */ }; #define wil_to_wiphy(i) (i->wdev->wiphy) diff --git a/drivers/net/wireless/ath/wil6210/wil_platform.h b/drivers/net/wireless/ath/wil6210/wil_platform.h index 33d4a34..f8c4117 100644 --- a/drivers/net/wireless/ath/wil6210/wil_platform.h +++ b/drivers/net/wireless/ath/wil6210/wil_platform.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2014-2015 Qualcomm Atheros, Inc. + * Copyright (c) 2014-2016 Qualcomm Atheros, Inc. * * Permission to use, copy, modify, and/or distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -23,6 +23,8 @@ enum wil_platform_event { WIL_PLATFORM_EVT_FW_CRASH = 0, WIL_PLATFORM_EVT_PRE_RESET = 1, WIL_PLATFORM_EVT_FW_RDY = 2, + WIL_PLATFORM_EVT_PRE_SUSPEND = 3, + WIL_PLATFORM_EVT_POST_SUSPEND = 4, }; /** -- cgit v0.10.2 From 8fe2a5f9f9b5a2f26d9d31d350b2323aee13067a Mon Sep 17 00:00:00 2001 From: Maya Erez Date: Mon, 16 May 2016 22:23:35 +0300 Subject: wil6210: align wil log functions to wil_dbg_ratelimited implementation Change the implementation of wil log functions for consistency with __wil_dbg_ratelimited. Signed-off-by: Maya Erez Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wil6210/debug.c b/drivers/net/wireless/ath/wil6210/debug.c index c312a66..217a459 100644 --- a/drivers/net/wireless/ath/wil6210/debug.c +++ b/drivers/net/wireless/ath/wil6210/debug.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2013 Qualcomm Atheros, Inc. + * Copyright (c) 2013,2016 Qualcomm Atheros, Inc. * * Permission to use, copy, modify, and/or distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -19,34 +19,31 @@ void __wil_err(struct wil6210_priv *wil, const char *fmt, ...) { - struct net_device *ndev = wil_to_ndev(wil); - struct va_format vaf = { - .fmt = fmt, - }; + struct va_format vaf; va_list args; va_start(args, fmt); + vaf.fmt = fmt; vaf.va = &args; - netdev_err(ndev, "%pV", &vaf); + netdev_err(wil_to_ndev(wil), "%pV", &vaf); trace_wil6210_log_err(&vaf); va_end(args); } void __wil_err_ratelimited(struct wil6210_priv *wil, const char *fmt, ...) { - if (net_ratelimit()) { - struct net_device *ndev = wil_to_ndev(wil); - struct va_format vaf = { - .fmt = fmt, - }; - va_list args; + struct va_format vaf; + va_list args; - va_start(args, fmt); - vaf.va = &args; - netdev_err(ndev, "%pV", &vaf); - trace_wil6210_log_err(&vaf); - va_end(args); - } + if (!net_ratelimit()) + return; + + va_start(args, fmt); + vaf.fmt = fmt; + vaf.va = &args; + netdev_err(wil_to_ndev(wil), "%pV", &vaf); + trace_wil6210_log_err(&vaf); + va_end(args); } void wil_dbg_ratelimited(const struct wil6210_priv *wil, const char *fmt, ...) @@ -67,27 +64,24 @@ void wil_dbg_ratelimited(const struct wil6210_priv *wil, const char *fmt, ...) void __wil_info(struct wil6210_priv *wil, const char *fmt, ...) { - struct net_device *ndev = wil_to_ndev(wil); - struct va_format vaf = { - .fmt = fmt, - }; + struct va_format vaf; va_list args; va_start(args, fmt); + vaf.fmt = fmt; vaf.va = &args; - netdev_info(ndev, "%pV", &vaf); + netdev_info(wil_to_ndev(wil), "%pV", &vaf); trace_wil6210_log_info(&vaf); va_end(args); } void wil_dbg_trace(struct wil6210_priv *wil, const char *fmt, ...) { - struct va_format vaf = { - .fmt = fmt, - }; + struct va_format vaf; va_list args; va_start(args, fmt); + vaf.fmt = fmt; vaf.va = &args; trace_wil6210_log_dbg(&vaf); va_end(args); -- cgit v0.10.2 From 8a9a3efa9e777c47a2cc10caa7473ca205279ae9 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Tue, 17 May 2016 16:38:45 +0200 Subject: ath6kl: fix typo firmare -> firmware Signed-off-by: Julia Lawall Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath6kl/core.h b/drivers/net/wireless/ath/ath6kl/core.h index 7a1970e..ac25f17 100644 --- a/drivers/net/wireless/ath/ath6kl/core.h +++ b/drivers/net/wireless/ath/ath6kl/core.h @@ -148,7 +148,7 @@ enum ath6kl_fw_capability { /* ratetable is the 2 stream version (max MCS15) */ ATH6KL_FW_CAPABILITY_RATETABLE_MCS15, - /* firmare doesn't support IP checksumming */ + /* firmware doesn't support IP checksumming */ ATH6KL_FW_CAPABILITY_NO_IP_CHECKSUM, /* this needs to be last */ -- cgit v0.10.2 From 93b4a09f0f3068e3190548393f39262e5295960e Mon Sep 17 00:00:00 2001 From: Heinrich Schuchardt Date: Wed, 18 May 2016 02:24:17 +0200 Subject: ath6kl: simplify logical condition x <= 7 implies x < 8. Signed-off-by: Heinrich Schuchardt Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath6kl/wmi.c b/drivers/net/wireless/ath/ath6kl/wmi.c index 631c3a0..b8cf04d 100644 --- a/drivers/net/wireless/ath/ath6kl/wmi.c +++ b/drivers/net/wireless/ath/ath6kl/wmi.c @@ -2544,8 +2544,7 @@ int ath6kl_wmi_create_pstream_cmd(struct wmi *wmi, u8 if_idx, s32 nominal_phy = 0; int ret; - if (!((params->user_pri < 8) && - (params->user_pri <= 0x7) && + if (!((params->user_pri <= 0x7) && (up_to_ac[params->user_pri & 0x7] == params->traffic_class) && (params->traffic_direc == UPLINK_TRAFFIC || params->traffic_direc == DNLINK_TRAFFIC || -- cgit v0.10.2 From 8a0a36cf98dbe689c31568b4f0ed883b43815cae Mon Sep 17 00:00:00 2001 From: Eduardo Abinader Date: Thu, 19 May 2016 17:15:29 +0200 Subject: ath9k: Proper TX99 interrupt ref count On TX99 mode, instead of assuming interrupt mask non ATH9K_INT_GLOBAL, let ath9k_hw_disable_interrupts proper set interrupt ref count. This prevents some PCI PERR occurring specialy when setting 11b and n rates. Signed-off-by: Eduardo Abinader Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/tx99.c b/drivers/net/wireless/ath/ath9k/tx99.c index 6161536..16aca9e 100644 --- a/drivers/net/wireless/ath/ath9k/tx99.c +++ b/drivers/net/wireless/ath/ath9k/tx99.c @@ -132,7 +132,6 @@ static int ath9k_tx99_init(struct ath_softc *sc) ath9k_ps_wakeup(sc); ath9k_hw_disable_interrupts(ah); - atomic_set(&ah->intr_ref_cnt, -1); ath_drain_all_txq(sc); ath_stoprecv(sc); -- cgit v0.10.2 From 5617c6cd6f844eaa2f4d61f165b7e6664a658865 Mon Sep 17 00:00:00 2001 From: Martin Willi Date: Mon, 9 May 2016 18:33:58 +0200 Subject: nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index d759901..bf75afa 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -10945,7 +10945,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_WIPHY, .doit = nl80211_set_wiphy, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_RTNL, }, { @@ -10961,7 +10961,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_INTERFACE, .doit = nl80211_set_interface, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV | NL80211_FLAG_NEED_RTNL, }, @@ -10969,7 +10969,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_NEW_INTERFACE, .doit = nl80211_new_interface, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WIPHY | NL80211_FLAG_NEED_RTNL, }, @@ -10977,7 +10977,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_DEL_INTERFACE, .doit = nl80211_del_interface, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV | NL80211_FLAG_NEED_RTNL, }, @@ -10985,7 +10985,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_GET_KEY, .doit = nl80211_get_key, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -10993,7 +10993,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_KEY, .doit = nl80211_set_key, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL | NL80211_FLAG_CLEAR_SKB, @@ -11002,7 +11002,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_NEW_KEY, .doit = nl80211_new_key, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL | NL80211_FLAG_CLEAR_SKB, @@ -11011,14 +11011,14 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_DEL_KEY, .doit = nl80211_del_key, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, { .cmd = NL80211_CMD_SET_BEACON, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .doit = nl80211_set_beacon, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, @@ -11026,7 +11026,7 @@ static const struct genl_ops nl80211_ops[] = { { .cmd = NL80211_CMD_START_AP, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .doit = nl80211_start_ap, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, @@ -11034,7 +11034,7 @@ static const struct genl_ops nl80211_ops[] = { { .cmd = NL80211_CMD_STOP_AP, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .doit = nl80211_stop_ap, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, @@ -11051,7 +11051,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_STATION, .doit = nl80211_set_station, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11059,7 +11059,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_NEW_STATION, .doit = nl80211_new_station, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11067,7 +11067,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_DEL_STATION, .doit = nl80211_del_station, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11076,7 +11076,7 @@ static const struct genl_ops nl80211_ops[] = { .doit = nl80211_get_mpath, .dumpit = nl80211_dump_mpath, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11085,7 +11085,7 @@ static const struct genl_ops nl80211_ops[] = { .doit = nl80211_get_mpp, .dumpit = nl80211_dump_mpp, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11093,7 +11093,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_MPATH, .doit = nl80211_set_mpath, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11101,7 +11101,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_NEW_MPATH, .doit = nl80211_new_mpath, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11109,7 +11109,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_DEL_MPATH, .doit = nl80211_del_mpath, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11117,7 +11117,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_BSS, .doit = nl80211_set_bss, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11156,7 +11156,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_MESH_CONFIG, .doit = nl80211_update_mesh_config, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11164,7 +11164,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_TRIGGER_SCAN, .doit = nl80211_trigger_scan, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11172,7 +11172,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_ABORT_SCAN, .doit = nl80211_abort_scan, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11185,7 +11185,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_START_SCHED_SCAN, .doit = nl80211_start_sched_scan, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11193,7 +11193,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_STOP_SCHED_SCAN, .doit = nl80211_stop_sched_scan, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11201,7 +11201,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_AUTHENTICATE, .doit = nl80211_authenticate, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL | NL80211_FLAG_CLEAR_SKB, @@ -11210,7 +11210,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_ASSOCIATE, .doit = nl80211_associate, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11218,7 +11218,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_DEAUTHENTICATE, .doit = nl80211_deauthenticate, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11226,7 +11226,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_DISASSOCIATE, .doit = nl80211_disassociate, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11234,7 +11234,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_JOIN_IBSS, .doit = nl80211_join_ibss, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11242,7 +11242,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_LEAVE_IBSS, .doit = nl80211_leave_ibss, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11252,7 +11252,7 @@ static const struct genl_ops nl80211_ops[] = { .doit = nl80211_testmode_do, .dumpit = nl80211_testmode_dump, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WIPHY | NL80211_FLAG_NEED_RTNL, }, @@ -11261,7 +11261,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_CONNECT, .doit = nl80211_connect, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11269,7 +11269,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_DISCONNECT, .doit = nl80211_disconnect, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11277,7 +11277,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_WIPHY_NETNS, .doit = nl80211_wiphy_netns, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WIPHY | NL80211_FLAG_NEED_RTNL, }, @@ -11290,7 +11290,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_PMKSA, .doit = nl80211_setdel_pmksa, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11298,7 +11298,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_DEL_PMKSA, .doit = nl80211_setdel_pmksa, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11306,7 +11306,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_FLUSH_PMKSA, .doit = nl80211_flush_pmksa, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11314,7 +11314,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_REMAIN_ON_CHANNEL, .doit = nl80211_remain_on_channel, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11322,7 +11322,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_CANCEL_REMAIN_ON_CHANNEL, .doit = nl80211_cancel_remain_on_channel, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11330,7 +11330,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_TX_BITRATE_MASK, .doit = nl80211_set_tx_bitrate_mask, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11338,7 +11338,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_REGISTER_FRAME, .doit = nl80211_register_mgmt, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11346,7 +11346,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_FRAME, .doit = nl80211_tx_mgmt, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11354,7 +11354,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_FRAME_WAIT_CANCEL, .doit = nl80211_tx_mgmt_cancel_wait, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11362,7 +11362,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_POWER_SAVE, .doit = nl80211_set_power_save, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11378,7 +11378,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_CQM, .doit = nl80211_set_cqm, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11386,7 +11386,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_CHANNEL, .doit = nl80211_set_channel, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11394,7 +11394,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_WDS_PEER, .doit = nl80211_set_wds_peer, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11402,7 +11402,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_JOIN_MESH, .doit = nl80211_join_mesh, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11410,7 +11410,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_LEAVE_MESH, .doit = nl80211_leave_mesh, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11418,7 +11418,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_JOIN_OCB, .doit = nl80211_join_ocb, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11426,7 +11426,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_LEAVE_OCB, .doit = nl80211_leave_ocb, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11443,7 +11443,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_WOWLAN, .doit = nl80211_set_wowlan, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WIPHY | NL80211_FLAG_NEED_RTNL, }, @@ -11452,7 +11452,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_REKEY_OFFLOAD, .doit = nl80211_set_rekey_data, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL | NL80211_FLAG_CLEAR_SKB, @@ -11461,7 +11461,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_TDLS_MGMT, .doit = nl80211_tdls_mgmt, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11469,7 +11469,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_TDLS_OPER, .doit = nl80211_tdls_oper, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11477,7 +11477,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_UNEXPECTED_FRAME, .doit = nl80211_register_unexpected_frame, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11485,7 +11485,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_PROBE_CLIENT, .doit = nl80211_probe_client, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11493,7 +11493,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_REGISTER_BEACONS, .doit = nl80211_register_beacons, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WIPHY | NL80211_FLAG_NEED_RTNL, }, @@ -11501,7 +11501,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_NOACK_MAP, .doit = nl80211_set_noack_map, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11509,7 +11509,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_START_P2P_DEVICE, .doit = nl80211_start_p2p_device, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11517,7 +11517,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_STOP_P2P_DEVICE, .doit = nl80211_stop_p2p_device, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11525,7 +11525,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_MCAST_RATE, .doit = nl80211_set_mcast_rate, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11533,7 +11533,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_MAC_ACL, .doit = nl80211_set_mac_acl, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV | NL80211_FLAG_NEED_RTNL, }, @@ -11541,7 +11541,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_RADAR_DETECT, .doit = nl80211_start_radar_detection, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11554,7 +11554,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_UPDATE_FT_IES, .doit = nl80211_update_ft_ies, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11562,7 +11562,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_CRIT_PROTOCOL_START, .doit = nl80211_crit_protocol_start, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11570,7 +11570,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_CRIT_PROTOCOL_STOP, .doit = nl80211_crit_protocol_stop, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11585,7 +11585,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_COALESCE, .doit = nl80211_set_coalesce, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WIPHY | NL80211_FLAG_NEED_RTNL, }, @@ -11593,7 +11593,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_CHANNEL_SWITCH, .doit = nl80211_channel_switch, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11602,7 +11602,7 @@ static const struct genl_ops nl80211_ops[] = { .doit = nl80211_vendor_cmd, .dumpit = nl80211_vendor_cmd_dump, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WIPHY | NL80211_FLAG_NEED_RTNL, }, @@ -11610,7 +11610,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_SET_QOS_MAP, .doit = nl80211_set_qos_map, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11618,7 +11618,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_ADD_TX_TS, .doit = nl80211_add_tx_ts, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11626,7 +11626,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_DEL_TX_TS, .doit = nl80211_del_tx_ts, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11634,7 +11634,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_TDLS_CHANNEL_SWITCH, .doit = nl80211_tdls_channel_switch, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, @@ -11642,7 +11642,7 @@ static const struct genl_ops nl80211_ops[] = { .cmd = NL80211_CMD_TDLS_CANCEL_CHANNEL_SWITCH, .doit = nl80211_tdls_cancel_channel_switch, .policy = nl80211_policy, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_NETDEV_UP | NL80211_FLAG_NEED_RTNL, }, -- cgit v0.10.2 From 100cb9ff40e0c9eba1d88e3c9c21e9670faeb896 Mon Sep 17 00:00:00 2001 From: Martin Willi Date: Mon, 9 May 2016 18:33:59 +0200 Subject: mac80211_hwsim: Allow managing radios from non-initial namespaces While wiphys can be moved into network namespaces over nl80211, the creation and removal of hwsim radios is currently limited to the initial namespace. This patch allows management of namespaced radios from the owning namespace by setting genetlink netnsok. To prevent two arbitrary namespaces to communicate over the simulated shared medium, radios are separated by netgroups. Each radio created in the same namespace lives in the same netgroup and hence can communicate with other radios in that group. When moving radios to other namespaces, the netgroup is preserved, so two radios having the same netgroup can communicate even if not in the same namespace; This allows a controlling namespace to create radios and move them to other namespaces for communication. When a net namespace owning a radio exits, the radio is destroyed unless it was created in the initial network namespace. This keeps the previous behavior by returning them to the init namespace, but prevents unprivileged users from creating radios in the initial namespace. Signed-off-by: Martin Willi Signed-off-by: Johannes Berg diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 9ed0ed1..0c75e8d 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -30,6 +30,8 @@ #include #include #include +#include +#include #include "mac80211_hwsim.h" #define WARN_QUEUE 100 @@ -250,6 +252,28 @@ static inline void hwsim_clear_chanctx_magic(struct ieee80211_chanctx_conf *c) cp->magic = 0; } +static unsigned int hwsim_net_id; + +static int hwsim_netgroup; + +struct hwsim_net { + int netgroup; +}; + +static inline int hwsim_net_get_netgroup(struct net *net) +{ + struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id); + + return hwsim_net->netgroup; +} + +static inline void hwsim_net_set_netgroup(struct net *net) +{ + struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id); + + hwsim_net->netgroup = hwsim_netgroup++; +} + static struct class *hwsim_class; static struct net_device *hwsim_mon; /* global monitor netdev */ @@ -526,6 +550,9 @@ struct mac80211_hwsim_data { */ u64 group; + /* group shared by radios created in the same netns */ + int netgroup; + int power_level; /* difference between this hw's clock and the real clock, in usecs */ @@ -568,6 +595,7 @@ static struct genl_family hwsim_genl_family = { .name = "MAC80211_HWSIM", .version = 1, .maxattr = HWSIM_ATTR_MAX, + .netnsok = true, }; enum hwsim_multicast_groups { @@ -1202,6 +1230,9 @@ static bool mac80211_hwsim_tx_frame_no_nl(struct ieee80211_hw *hw, if (!(data->group & data2->group)) continue; + if (data->netgroup != data2->netgroup) + continue; + if (!hwsim_chans_compat(chan, data2->tmp_chan) && !hwsim_chans_compat(chan, data2->channel)) { ieee80211_iterate_active_interfaces_atomic( @@ -2349,6 +2380,7 @@ static int mac80211_hwsim_new_radio(struct genl_info *info, struct ieee80211_hw *hw; enum nl80211_band band; const struct ieee80211_ops *ops = &mac80211_hwsim_ops; + struct net *net; int idx; if (WARN_ON(param->channels > 1 && !param->use_chanctx)) @@ -2366,6 +2398,13 @@ static int mac80211_hwsim_new_radio(struct genl_info *info, err = -ENOMEM; goto failed; } + + if (info) + net = genl_info_net(info); + else + net = &init_net; + wiphy_net_set(hw->wiphy, net); + data = hw->priv; data->hw = hw; @@ -2541,6 +2580,8 @@ static int mac80211_hwsim_new_radio(struct genl_info *info, data->group = 1; mutex_init(&data->mutex); + data->netgroup = hwsim_net_get_netgroup(net); + /* Enable frame retransmissions for lossy channels */ hw->max_rates = 4; hw->max_rate_tries = 11; @@ -3013,6 +3054,9 @@ static int hwsim_del_radio_nl(struct sk_buff *msg, struct genl_info *info) continue; } + if (!net_eq(wiphy_net(data->hw->wiphy), genl_info_net(info))) + continue; + list_del(&data->list); spin_unlock_bh(&hwsim_radio_lock); mac80211_hwsim_del_radio(data, wiphy_name(data->hw->wiphy), @@ -3039,6 +3083,9 @@ static int hwsim_get_radio_nl(struct sk_buff *msg, struct genl_info *info) if (data->idx != idx) continue; + if (!net_eq(wiphy_net(data->hw->wiphy), genl_info_net(info))) + continue; + skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!skb) { res = -ENOMEM; @@ -3078,6 +3125,9 @@ static int hwsim_dump_radio_nl(struct sk_buff *skb, if (data->idx < idx) continue; + if (!net_eq(wiphy_net(data->hw->wiphy), sock_net(skb->sk))) + continue; + res = mac80211_hwsim_get_radio(skb, data, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, cb, @@ -3117,13 +3167,13 @@ static const struct genl_ops hwsim_ops[] = { .cmd = HWSIM_CMD_NEW_RADIO, .policy = hwsim_genl_policy, .doit = hwsim_new_radio_nl, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, }, { .cmd = HWSIM_CMD_DEL_RADIO, .policy = hwsim_genl_policy, .doit = hwsim_del_radio_nl, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, }, { .cmd = HWSIM_CMD_GET_RADIO, @@ -3205,6 +3255,40 @@ failure: return -EINVAL; } +static __net_init int hwsim_init_net(struct net *net) +{ + hwsim_net_set_netgroup(net); + + return 0; +} + +static void __net_exit hwsim_exit_net(struct net *net) +{ + struct mac80211_hwsim_data *data, *tmp; + + spin_lock_bh(&hwsim_radio_lock); + list_for_each_entry_safe(data, tmp, &hwsim_radios, list) { + if (!net_eq(wiphy_net(data->hw->wiphy), net)) + continue; + + /* Radios created in init_net are returned to init_net. */ + if (data->netgroup == hwsim_net_get_netgroup(&init_net)) + continue; + + list_del(&data->list); + INIT_WORK(&data->destroy_work, destroy_radio); + schedule_work(&data->destroy_work); + } + spin_unlock_bh(&hwsim_radio_lock); +} + +static struct pernet_operations hwsim_net_ops = { + .init = hwsim_init_net, + .exit = hwsim_exit_net, + .id = &hwsim_net_id, + .size = sizeof(struct hwsim_net), +}; + static void hwsim_exit_netlink(void) { /* unregister the notifier */ @@ -3241,10 +3325,14 @@ static int __init init_mac80211_hwsim(void) spin_lock_init(&hwsim_radio_lock); INIT_LIST_HEAD(&hwsim_radios); - err = platform_driver_register(&mac80211_hwsim_driver); + err = register_pernet_device(&hwsim_net_ops); if (err) return err; + err = platform_driver_register(&mac80211_hwsim_driver); + if (err) + goto out_unregister_pernet; + hwsim_class = class_create(THIS_MODULE, "mac80211_hwsim"); if (IS_ERR(hwsim_class)) { err = PTR_ERR(hwsim_class); @@ -3362,6 +3450,8 @@ out_free_radios: mac80211_hwsim_free(); out_unregister_driver: platform_driver_unregister(&mac80211_hwsim_driver); +out_unregister_pernet: + unregister_pernet_device(&hwsim_net_ops); return err; } module_init(init_mac80211_hwsim); @@ -3375,5 +3465,6 @@ static void __exit exit_mac80211_hwsim(void) mac80211_hwsim_free(); unregister_netdev(hwsim_mon); platform_driver_unregister(&mac80211_hwsim_driver); + unregister_pernet_device(&hwsim_net_ops); } module_exit(exit_mac80211_hwsim); -- cgit v0.10.2 From 0bb7ed426373cebf406ff3ae7b6d2344f2f4364c Mon Sep 17 00:00:00 2001 From: Ben Greear Date: Fri, 13 May 2016 11:29:33 -0700 Subject: mac80211: add vht cap decode to debugfs This makes it a lot easier to understand the capabilities used by the station: VHT supported cap: 0x300819b2 MAX-MPDU-11454 80Mhz RXLDPC SHORT-GI-80 TXSTBC RXSTBC_1 SU-BEAMFORMER-CAPABLE SU-BEAMFORMEE-CAPABLE BEAMFORMEE-STS: 0x0 SOUNDING-DIMENSIONS: 0x0 MU-BEAMFORMER-CAPABLE MPDU-LENGTH-EXPONENT: 0x0 LINK-ADAPTATION-VHT-MRQ-MFB: 0x0 RX-ANTENNA-PATTERN TX-ANTENNA-PATTERN RX MCS: fffe TX MCS: fffe Signed-off-by: Ben Greear Signed-off-by: Johannes Berg diff --git a/net/mac80211/debugfs_sta.c b/net/mac80211/debugfs_sta.c index 33dfcbc..fd33413 100644 --- a/net/mac80211/debugfs_sta.c +++ b/net/mac80211/debugfs_sta.c @@ -328,14 +328,88 @@ STA_OPS(ht_capa); static ssize_t sta_vht_capa_read(struct file *file, char __user *userbuf, size_t count, loff_t *ppos) { - char buf[128], *p = buf; + char buf[512], *p = buf; struct sta_info *sta = file->private_data; struct ieee80211_sta_vht_cap *vhtc = &sta->sta.vht_cap; p += scnprintf(p, sizeof(buf) + buf - p, "VHT %ssupported\n", vhtc->vht_supported ? "" : "not "); if (vhtc->vht_supported) { - p += scnprintf(p, sizeof(buf)+buf-p, "cap: %#.8x\n", vhtc->cap); + p += scnprintf(p, sizeof(buf) + buf - p, "cap: %#.8x\n", + vhtc->cap); +#define PFLAG(a, b) \ + do { \ + if (vhtc->cap & IEEE80211_VHT_CAP_ ## a) \ + p += scnprintf(p, sizeof(buf) + buf - p, \ + "\t\t%s\n", b); \ + } while (0) + + switch (vhtc->cap & 0x3) { + case IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_3895: + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\tMAX-MPDU-3895\n"); + break; + case IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_7991: + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\tMAX-MPDU-7991\n"); + break; + case IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454: + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\tMAX-MPDU-11454\n"); + break; + default: + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\tMAX-MPDU-UNKNOWN\n"); + }; + switch (vhtc->cap & IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_MASK) { + case 0: + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\t80Mhz\n"); + break; + case IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160MHZ: + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\t160Mhz\n"); + break; + case IEEE80211_VHT_CAP_SUPP_CHAN_WIDTH_160_80PLUS80MHZ: + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\t80+80Mhz\n"); + break; + default: + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\tUNKNOWN-MHZ: 0x%x\n", + (vhtc->cap >> 2) & 0x3); + }; + PFLAG(RXLDPC, "RXLDPC"); + PFLAG(SHORT_GI_80, "SHORT-GI-80"); + PFLAG(SHORT_GI_160, "SHORT-GI-160"); + PFLAG(TXSTBC, "TXSTBC"); + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\tRXSTBC_%d\n", (vhtc->cap >> 8) & 0x7); + PFLAG(SU_BEAMFORMER_CAPABLE, "SU-BEAMFORMER-CAPABLE"); + PFLAG(SU_BEAMFORMEE_CAPABLE, "SU-BEAMFORMEE-CAPABLE"); + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\tBEAMFORMEE-STS: 0x%x\n", + (vhtc->cap & IEEE80211_VHT_CAP_BEAMFORMEE_STS_MASK) >> + IEEE80211_VHT_CAP_BEAMFORMEE_STS_SHIFT); + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\tSOUNDING-DIMENSIONS: 0x%x\n", + (vhtc->cap & IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_MASK) + >> IEEE80211_VHT_CAP_SOUNDING_DIMENSIONS_SHIFT); + PFLAG(MU_BEAMFORMER_CAPABLE, "MU-BEAMFORMER-CAPABLE"); + PFLAG(MU_BEAMFORMEE_CAPABLE, "MU-BEAMFORMEE-CAPABLE"); + PFLAG(VHT_TXOP_PS, "TXOP-PS"); + PFLAG(HTC_VHT, "HTC-VHT"); + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\tMPDU-LENGTH-EXPONENT: 0x%x\n", + (vhtc->cap & IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_MASK) >> + IEEE80211_VHT_CAP_MAX_A_MPDU_LENGTH_EXPONENT_SHIFT); + PFLAG(VHT_LINK_ADAPTATION_VHT_UNSOL_MFB, + "LINK-ADAPTATION-VHT-UNSOL-MFB"); + p += scnprintf(p, sizeof(buf) + buf - p, + "\t\tLINK-ADAPTATION-VHT-MRQ-MFB: 0x%x\n", + (vhtc->cap & IEEE80211_VHT_CAP_VHT_LINK_ADAPTATION_VHT_MRQ_MFB) >> 26); + PFLAG(RX_ANTENNA_PATTERN, "RX-ANTENNA-PATTERN"); + PFLAG(TX_ANTENNA_PATTERN, "TX-ANTENNA-PATTERN"); p += scnprintf(p, sizeof(buf)+buf-p, "RX MCS: %.4x\n", le16_to_cpu(vhtc->vht_mcs.rx_mcs_map)); -- cgit v0.10.2 From bf1ecd210541ef5f3a110e88e8ca5d33b4aa5c23 Mon Sep 17 00:00:00 2001 From: Jouni Malinen Date: Tue, 31 May 2016 00:16:50 +0300 Subject: cfg80211: Allow cfg80211_connect_result() errors to be distinguished Previously, the status parameter to cfg80211_connect_result() was documented as using WLAN_STATUS_UNSPECIFIED_FAILURE (1) when the real status code for the failure is not known. This value can be used by an AP (and often is) and as such, user space cannot distinguish between explicitly rejected authentication/association and not being able to even try to associate or not receiving a response from the AP. Add a new inline function, cfg80211_connect_timeout(), to be used when the driver knows that the connection attempt failed due to a reason where connection could not be attempt or no response was received from the AP. The internal functions now allow a negative status value (-1) to be used as an indication of this special case. This results in the NL80211_ATTR_TIMED_OUT to be added to the NL80211_CMD_CONNECT event to allow user space to determine this case was hit. For backwards compatibility, NL80211_STATUS_CODE with the value WLAN_STATUS_UNSPECIFIED_FAILURE is still indicated in the event in such a case. Signed-off-by: Jouni Malinen [johannes: fix cfg80211_connect_bss() prototype to use int for status, add cfg80211_connect_timeout() to docbook, fix docbook] Signed-off-by: Johannes Berg diff --git a/Documentation/DocBook/80211.tmpl b/Documentation/DocBook/80211.tmpl index 5f7c559..800fe7a 100644 --- a/Documentation/DocBook/80211.tmpl +++ b/Documentation/DocBook/80211.tmpl @@ -136,6 +136,7 @@ !Finclude/net/cfg80211.h cfg80211_ibss_joined !Finclude/net/cfg80211.h cfg80211_connect_result !Finclude/net/cfg80211.h cfg80211_connect_bss +!Finclude/net/cfg80211.h cfg80211_connect_timeout !Finclude/net/cfg80211.h cfg80211_roamed !Finclude/net/cfg80211.h cfg80211_disconnected !Finclude/net/cfg80211.h cfg80211_ready_on_channel diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 6392167..537f010 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -2367,19 +2367,23 @@ struct cfg80211_qos_map { * (invoked with the wireless_dev mutex held) * * @connect: Connect to the ESS with the specified parameters. When connected, - * call cfg80211_connect_result() with status code %WLAN_STATUS_SUCCESS. - * If the connection fails for some reason, call cfg80211_connect_result() - * with the status from the AP. The driver is allowed to roam to other - * BSSes within the ESS when the other BSS matches the connect parameters. - * When such roaming is initiated by the driver, the driver is expected to - * verify that the target matches the configured security parameters and - * to use Reassociation Request frame instead of Association Request frame. - * The connect function can also be used to request the driver to perform - * a specific roam when connected to an ESS. In that case, the prev_bssid + * call cfg80211_connect_result()/cfg80211_connect_bss() with status code + * %WLAN_STATUS_SUCCESS. If the connection fails for some reason, call + * cfg80211_connect_result()/cfg80211_connect_bss() with the status code + * from the AP or cfg80211_connect_timeout() if no frame with status code + * was received. + * The driver is allowed to roam to other BSSes within the ESS when the + * other BSS matches the connect parameters. When such roaming is initiated + * by the driver, the driver is expected to verify that the target matches + * the configured security parameters and to use Reassociation Request + * frame instead of Association Request frame. + * The connect function can also be used to request the driver to perform a + * specific roam when connected to an ESS. In that case, the prev_bssid * parameter is set to the BSSID of the currently associated BSS as an - * indication of requesting reassociation. In both the driver-initiated and - * new connect() call initiated roaming cases, the result of roaming is - * indicated with a call to cfg80211_roamed() or cfg80211_roamed_bss(). + * indication of requesting reassociation. + * In both the driver-initiated and new connect() call initiated roaming + * cases, the result of roaming is indicated with a call to + * cfg80211_roamed() or cfg80211_roamed_bss(). * (invoked with the wireless_dev mutex held) * @disconnect: Disconnect from the BSS/ESS. * (invoked with the wireless_dev mutex held) @@ -4680,7 +4684,7 @@ static inline void cfg80211_testmode_event(struct sk_buff *skb, gfp_t gfp) void cfg80211_connect_bss(struct net_device *dev, const u8 *bssid, struct cfg80211_bss *bss, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, - size_t resp_ie_len, u16 status, gfp_t gfp); + size_t resp_ie_len, int status, gfp_t gfp); /** * cfg80211_connect_result - notify cfg80211 of connection result @@ -4710,6 +4714,29 @@ cfg80211_connect_result(struct net_device *dev, const u8 *bssid, } /** + * cfg80211_connect_timeout - notify cfg80211 of connection timeout + * + * @dev: network device + * @bssid: the BSSID of the AP + * @req_ie: association request IEs (maybe be %NULL) + * @req_ie_len: association request IEs length + * @gfp: allocation flags + * + * It should be called by the underlying driver whenever connect() has failed + * in a sequence where no explicit authentication/association rejection was + * received from the AP. This could happen, e.g., due to not being able to send + * out the Authentication or Association Request frame or timing out while + * waiting for the response. + */ +static inline void +cfg80211_connect_timeout(struct net_device *dev, const u8 *bssid, + const u8 *req_ie, size_t req_ie_len, gfp_t gfp) +{ + cfg80211_connect_bss(dev, bssid, NULL, req_ie, req_ie_len, NULL, 0, -1, + gfp); +} + +/** * cfg80211_roamed - notify cfg80211 of roaming * * @dev: network device diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index e23d786..8d99531 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -493,7 +493,12 @@ * This attribute is ignored if driver does not support roam scan. * It is also sent as an event, with the BSSID and response IEs when the * connection is established or failed to be established. This can be - * determined by the STATUS_CODE attribute. + * determined by the %NL80211_ATTR_STATUS_CODE attribute (0 = success, + * non-zero = failure). If %NL80211_ATTR_TIMED_OUT is included in the + * event, the connection attempt failed due to not being able to initiate + * authentication/association or not receiving a response from the AP. + * Non-zero %NL80211_ATTR_STATUS_CODE value is indicated in that case as + * well to remain backwards compatible. * @NL80211_CMD_ROAM: request that the card roam (currently not implemented), * sent as an event when the card/driver roamed by itself. * @NL80211_CMD_DISCONNECT: drop a given connection; also used to notify diff --git a/net/wireless/core.h b/net/wireless/core.h index 025b7a5..a4d547f 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -214,7 +214,7 @@ struct cfg80211_event { size_t req_ie_len; size_t resp_ie_len; struct cfg80211_bss *bss; - u16 status; + int status; /* -1 = failed; 0..65535 = status code */ } cr; struct { const u8 *req_ie; @@ -374,7 +374,7 @@ int cfg80211_connect(struct cfg80211_registered_device *rdev, void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, size_t resp_ie_len, - u16 status, bool wextev, + int status, bool wextev, struct cfg80211_bss *bss); void __cfg80211_disconnected(struct net_device *dev, const u8 *ie, size_t ie_len, u16 reason, bool from_ap); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index bf75afa..03ac2ba 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -12092,7 +12092,7 @@ void nl80211_send_connect_result(struct cfg80211_registered_device *rdev, struct net_device *netdev, const u8 *bssid, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, size_t resp_ie_len, - u16 status, gfp_t gfp) + int status, gfp_t gfp) { struct sk_buff *msg; void *hdr; @@ -12110,7 +12110,10 @@ void nl80211_send_connect_result(struct cfg80211_registered_device *rdev, if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) || nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) || (bssid && nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, bssid)) || - nla_put_u16(msg, NL80211_ATTR_STATUS_CODE, status) || + nla_put_u16(msg, NL80211_ATTR_STATUS_CODE, + status < 0 ? WLAN_STATUS_UNSPECIFIED_FAILURE : + status) || + (status < 0 && nla_put_flag(msg, NL80211_ATTR_TIMED_OUT)) || (req_ie && nla_put(msg, NL80211_ATTR_REQ_IE, req_ie_len, req_ie)) || (resp_ie && diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h index 84d4edf..a63f402 100644 --- a/net/wireless/nl80211.h +++ b/net/wireless/nl80211.h @@ -55,7 +55,7 @@ void nl80211_send_connect_result(struct cfg80211_registered_device *rdev, struct net_device *netdev, const u8 *bssid, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, size_t resp_ie_len, - u16 status, gfp_t gfp); + int status, gfp_t gfp); void nl80211_send_roamed(struct cfg80211_registered_device *rdev, struct net_device *netdev, const u8 *bssid, const u8 *req_ie, size_t req_ie_len, diff --git a/net/wireless/sme.c b/net/wireless/sme.c index 584fdc3..add6824 100644 --- a/net/wireless/sme.c +++ b/net/wireless/sme.c @@ -244,9 +244,7 @@ void cfg80211_conn_work(struct work_struct *work) if (cfg80211_conn_do_work(wdev)) { __cfg80211_connect_result( wdev->netdev, bssid, - NULL, 0, NULL, 0, - WLAN_STATUS_UNSPECIFIED_FAILURE, - false, NULL); + NULL, 0, NULL, 0, -1, false, NULL); } wdev_unlock(wdev); } @@ -648,7 +646,7 @@ static DECLARE_WORK(cfg80211_disconnect_work, disconnect_work); void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, size_t resp_ie_len, - u16 status, bool wextev, + int status, bool wextev, struct cfg80211_bss *bss) { struct wireless_dev *wdev = dev->ieee80211_ptr; @@ -757,7 +755,7 @@ void __cfg80211_connect_result(struct net_device *dev, const u8 *bssid, void cfg80211_connect_bss(struct net_device *dev, const u8 *bssid, struct cfg80211_bss *bss, const u8 *req_ie, size_t req_ie_len, const u8 *resp_ie, - size_t resp_ie_len, u16 status, gfp_t gfp) + size_t resp_ie_len, int status, gfp_t gfp) { struct wireless_dev *wdev = dev->ieee80211_ptr; struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy); -- cgit v0.10.2 From 019ae3a918811715192b22c400ac78d54acc26a9 Mon Sep 17 00:00:00 2001 From: "Kanchanapally, Vidyullatha" Date: Mon, 16 May 2016 10:41:04 +0530 Subject: cfg80211: Advertise extended capabilities per interface type to userspace The driver extended capabilities may differ for different interface types which the userspace needs to know (for example the fine timing measurement initiator and responder bits might differ for a station and AP). Add a new nl80211 attribute to provide extended capabilities per interface type to userspace. Signed-off-by: Vidyullatha Kanchanapally Reviewed-by: Jouni Malinen Signed-off-by: Johannes Berg diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 537f010..7bbb00d 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -3084,6 +3084,24 @@ struct wiphy_vendor_command { }; /** + * struct wiphy_iftype_ext_capab - extended capabilities per interface type + * @iftype: interface type + * @extended_capabilities: extended capabilities supported by the driver, + * additional capabilities might be supported by userspace; these are the + * 802.11 extended capabilities ("Extended Capabilities element") and are + * in the same format as in the information element. See IEEE Std + * 802.11-2012 8.4.2.29 for the defined fields. + * @extended_capabilities_mask: mask of the valid values + * @extended_capabilities_len: length of the extended capabilities + */ +struct wiphy_iftype_ext_capab { + enum nl80211_iftype iftype; + const u8 *extended_capabilities; + const u8 *extended_capabilities_mask; + u8 extended_capabilities_len; +}; + +/** * struct wiphy - wireless hardware description * @reg_notifier: the driver's regulatory notification callback, * note that if your driver uses wiphy_apply_custom_regulatory() @@ -3203,9 +3221,14 @@ struct wiphy_vendor_command { * additional capabilities might be supported by userspace; these are * the 802.11 extended capabilities ("Extended Capabilities element") * and are in the same format as in the information element. See - * 802.11-2012 8.4.2.29 for the defined fields. + * 802.11-2012 8.4.2.29 for the defined fields. These are the default + * extended capabilities to be used if the capabilities are not specified + * for a specific interface type in iftype_ext_capab. * @extended_capabilities_mask: mask of the valid values * @extended_capabilities_len: length of the extended capabilities + * @iftype_ext_capab: array of extended capabilities per interface type + * @num_iftype_ext_capab: number of interface types for which extended + * capabilities are specified separately. * @coalesce: packet coalescing support information * * @vendor_commands: array of vendor commands supported by the hardware @@ -3305,6 +3328,9 @@ struct wiphy { const u8 *extended_capabilities, *extended_capabilities_mask; u8 extended_capabilities_len; + const struct wiphy_iftype_ext_capab *iftype_ext_capab; + unsigned int num_iftype_ext_capab; + /* If multiple wiphys are registered and you're handed e.g. * a regular netdev with assigned ieee80211_ptr, you won't * know whether it points to a wiphy your driver has registered diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 8d99531..53c8278 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1824,6 +1824,11 @@ enum nl80211_commands { * * @NL80211_ATTR_PAD: attribute used for padding for 64-bit alignment * + * @NL80211_ATTR_IFTYPE_EXT_CAPA: Nested attribute of the following attributes: + * %NL80211_ATTR_IFTYPE, %NL80211_ATTR_EXT_CAPA, + * %NL80211_ATTR_EXT_CAPA_MASK, to specify the extended capabilities per + * interface type. + * * @NUM_NL80211_ATTR: total number of nl80211_attrs available * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use @@ -2206,6 +2211,8 @@ enum nl80211_attrs { NL80211_ATTR_PAD, + NL80211_ATTR_IFTYPE_EXT_CAPA, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, diff --git a/net/wireless/core.c b/net/wireless/core.c index d25c82b..b8e10a9 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -750,6 +750,36 @@ int wiphy_register(struct wiphy *wiphy) nl80211_send_reg_change_event(&request); } + /* Check that nobody globally advertises any capabilities they do not + * advertise on all possible interface types. + */ + if (wiphy->extended_capabilities_len && + wiphy->num_iftype_ext_capab && + wiphy->iftype_ext_capab) { + u8 supported_on_all, j; + const struct wiphy_iftype_ext_capab *capab; + + capab = wiphy->iftype_ext_capab; + for (j = 0; j < wiphy->extended_capabilities_len; j++) { + if (capab[0].extended_capabilities_len > j) + supported_on_all = + capab[0].extended_capabilities[j]; + else + supported_on_all = 0x00; + for (i = 1; i < wiphy->num_iftype_ext_capab; i++) { + if (j >= capab[i].extended_capabilities_len) { + supported_on_all = 0x00; + break; + } + supported_on_all &= + capab[i].extended_capabilities[j]; + } + if (WARN_ON(wiphy->extended_capabilities[j] & + ~supported_on_all)) + break; + } + } + rdev->wiphy.registered = true; rtnl_unlock(); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 03ac2ba..d120449 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -1264,7 +1264,7 @@ nl80211_send_mgmt_stypes(struct sk_buff *msg, struct nl80211_dump_wiphy_state { s64 filter_wiphy; long start; - long split_start, band_start, chan_start; + long split_start, band_start, chan_start, capa_start; bool split; }; @@ -1761,6 +1761,47 @@ static int nl80211_send_wiphy(struct cfg80211_registered_device *rdev, nla_nest_end(msg, nested); } + state->split_start++; + break; + case 13: + if (rdev->wiphy.num_iftype_ext_capab && + rdev->wiphy.iftype_ext_capab) { + struct nlattr *nested_ext_capab, *nested; + + nested = nla_nest_start(msg, + NL80211_ATTR_IFTYPE_EXT_CAPA); + if (!nested) + goto nla_put_failure; + + for (i = state->capa_start; + i < rdev->wiphy.num_iftype_ext_capab; i++) { + const struct wiphy_iftype_ext_capab *capab; + + capab = &rdev->wiphy.iftype_ext_capab[i]; + + nested_ext_capab = nla_nest_start(msg, i); + if (!nested_ext_capab || + nla_put_u32(msg, NL80211_ATTR_IFTYPE, + capab->iftype) || + nla_put(msg, NL80211_ATTR_EXT_CAPA, + capab->extended_capabilities_len, + capab->extended_capabilities) || + nla_put(msg, NL80211_ATTR_EXT_CAPA_MASK, + capab->extended_capabilities_len, + capab->extended_capabilities_mask)) + goto nla_put_failure; + + nla_nest_end(msg, nested_ext_capab); + if (state->split) + break; + } + nla_nest_end(msg, nested); + if (i < rdev->wiphy.num_iftype_ext_capab) { + state->capa_start = i + 1; + break; + } + } + /* done */ state->split_start = 0; break; -- cgit v0.10.2 From 8d9c418b865cf4b45b9b1f63bdf7cce837773a6c Mon Sep 17 00:00:00 2001 From: Masaru Nagai Date: Wed, 1 Jun 2016 03:01:28 +0900 Subject: ravb: Add ESF in RCR for enabling separation filter The H/W manual recommends B'10 or B'11 in a value of the separation filtering select bits in the receive configuration register (RCR.ESF). When B'10 is set, frames from non-matching streams are discarded. When B'11 is set, frames from non-matching streams are processed in reception queue 0 (best effort). This patch sets B'11 in ESF. [ykaneko0929@gmail.com: revised this commit message] Signed-off-by: Masaru Nagai Signed-off-by: Kazuya Mizuguchi Signed-off-by: Yoshihiro Kaneko Acked-by: Sergei Shtylyov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 867caf6..1ceadbf 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -402,7 +402,8 @@ static int ravb_dmac_init(struct net_device *ndev) #endif /* Set AVB RX */ - ravb_write(ndev, RCR_EFFS | RCR_ENCF | RCR_ETS0 | 0x18000000, RCR); + ravb_write(ndev, + RCR_EFFS | RCR_ENCF | RCR_ETS0 | RCR_ESF | 0x18000000, RCR); /* Set FIFO size */ ravb_write(ndev, TGC_TQP_AVBMODE1 | 0x00222200, TGC); -- cgit v0.10.2 From 524c6f691b99065577b245b250efe93fb0cda5c4 Mon Sep 17 00:00:00 2001 From: Kazuya Mizuguchi Date: Mon, 30 May 2016 05:25:43 +0900 Subject: ravb: Add SET_RUNTIME_PM_OPS macro Use SET_RUNTIME_PM_OPS macro instead of assigning a member of dev_pm_ops directly. Signed-off-by: Kazuya Mizuguchi Signed-off-by: Yoshihiro Kaneko Acked-by: Sergei Shtylyov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 1ceadbf..aa82131 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -2112,8 +2112,7 @@ static int ravb_runtime_nop(struct device *dev) } static const struct dev_pm_ops ravb_dev_pm_ops = { - .runtime_suspend = ravb_runtime_nop, - .runtime_resume = ravb_runtime_nop, + SET_RUNTIME_PM_OPS(ravb_runtime_nop, ravb_runtime_nop, NULL) }; #define RAVB_PM_OPS (&ravb_dev_pm_ops) -- cgit v0.10.2 From 595d0b29463343c3be995d3948930b8231e5b8cd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 31 May 2016 15:22:41 -0700 Subject: udp: avoid csum_partial() for validated skb In commit e6afc8ace6dd5 ("udp: remove headers from UDP packets before queueing"), udp_csum_pull_header() helper was added but missed fact that CHECKSUM_UNNECESSARY packets were now converted to CHECKSUM_NONE and skb->csum_valid was set to 1 for them. Since csum_partial() is quite expensive, even for 8-byte area, it is worth adding a test. We also can use skb->data instead of udp_hdr() as we are pulling UDP headers, as it is sightly faster. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/net/udp.h b/include/net/udp.h index ae07f37..8894d71 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -160,8 +160,8 @@ void udp_set_csum(bool nocheck, struct sk_buff *skb, static inline void udp_csum_pull_header(struct sk_buff *skb) { - if (skb->ip_summed == CHECKSUM_NONE) - skb->csum = csum_partial(udp_hdr(skb), sizeof(struct udphdr), + if (!skb->csum_valid && skb->ip_summed == CHECKSUM_NONE) + skb->csum = csum_partial(skb->data, sizeof(struct udphdr), skb->csum); skb_pull_rcsum(skb, sizeof(struct udphdr)); UDP_SKB_CB(skb)->cscov -= sizeof(struct udphdr); -- cgit v0.10.2 From 260916dfb48c374f7840f3b86e69afd3afdb6e96 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 1 Jun 2016 11:43:00 +0800 Subject: macvlan: Fix potential use-after free for broadcasts When we postpone a broadcast packet we save the source port in the skb if it is local. However, the source port can disappear before we get a chance to process the packet. This patch fixes this by holding a ref count on the netdev. It also delays the skb->cb modification until after we allocate the new skb as you should not modify shared skbs. Fixes: 412ca1550cbe ("macvlan: Move broadcasts into a work queue") Signed-off-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index cb01023..a71fa59 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -305,11 +305,14 @@ static void macvlan_process_broadcast(struct work_struct *w) rcu_read_unlock(); + if (src) + dev_put(src->dev); kfree_skb(skb); } } static void macvlan_broadcast_enqueue(struct macvlan_port *port, + const struct macvlan_dev *src, struct sk_buff *skb) { struct sk_buff *nskb; @@ -319,8 +322,12 @@ static void macvlan_broadcast_enqueue(struct macvlan_port *port, if (!nskb) goto err; + MACVLAN_SKB_CB(nskb)->src = src; + spin_lock(&port->bc_queue.lock); if (skb_queue_len(&port->bc_queue) < MACVLAN_BC_QUEUE_LEN) { + if (src) + dev_hold(src->dev); __skb_queue_tail(&port->bc_queue, nskb); err = 0; } @@ -429,8 +436,7 @@ static rx_handler_result_t macvlan_handle_frame(struct sk_buff **pskb) goto out; } - MACVLAN_SKB_CB(skb)->src = src; - macvlan_broadcast_enqueue(port, skb); + macvlan_broadcast_enqueue(port, src, skb); return RX_HANDLER_PASS; } -- cgit v0.10.2 From 9c127a016e66a85edaad6f9a674d0d1dce93d251 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 1 Jun 2016 11:45:44 +0800 Subject: macvlan: Avoid unnecessary multicast cloning Currently we always queue a multicast packet for further processing, even if none of the macvlan devices are subscribed to the address. This patch optimises this by adding a global multicast filter for a macvlan_port. Note that this patch doesn't handle the broadcast addresses of the individual macvlan devices correctly, if they are not all identical to vlan->lowerdev. However, this is already broken because there is no mechanism in place to update the individual multicast filters when you change the broadcast address. If someone cares enough they should fix this by collecting all broadcast addresses for a macvlan as we do for multicast and unicast. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index a71fa59..0c65bd9 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -49,6 +49,7 @@ struct macvlan_port { bool passthru; int count; struct hlist_head vlan_source_hash[MACVLAN_HASH_SIZE]; + DECLARE_BITMAP(mc_filter, MACVLAN_MC_FILTER_SZ); }; struct macvlan_source_entry { @@ -419,6 +420,8 @@ static rx_handler_result_t macvlan_handle_frame(struct sk_buff **pskb) port = macvlan_port_get_rcu(skb->dev); if (is_multicast_ether_addr(eth->h_dest)) { + unsigned int hash; + skb = ip_check_defrag(dev_net(skb->dev), skb, IP_DEFRAG_MACVLAN); if (!skb) return RX_HANDLER_CONSUMED; @@ -436,7 +439,9 @@ static rx_handler_result_t macvlan_handle_frame(struct sk_buff **pskb) goto out; } - macvlan_broadcast_enqueue(port, src, skb); + hash = mc_hash(NULL, eth->h_dest); + if (test_bit(hash, port->mc_filter)) + macvlan_broadcast_enqueue(port, src, skb); return RX_HANDLER_PASS; } @@ -722,12 +727,12 @@ static void macvlan_change_rx_flags(struct net_device *dev, int change) } } -static void macvlan_set_mac_lists(struct net_device *dev) +static void macvlan_compute_filter(unsigned long *mc_filter, + struct net_device *dev, + struct macvlan_dev *vlan) { - struct macvlan_dev *vlan = netdev_priv(dev); - if (dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) { - bitmap_fill(vlan->mc_filter, MACVLAN_MC_FILTER_SZ); + bitmap_fill(mc_filter, MACVLAN_MC_FILTER_SZ); } else { struct netdev_hw_addr *ha; DECLARE_BITMAP(filter, MACVLAN_MC_FILTER_SZ); @@ -739,10 +744,33 @@ static void macvlan_set_mac_lists(struct net_device *dev) __set_bit(mc_hash(vlan, dev->broadcast), filter); - bitmap_copy(vlan->mc_filter, filter, MACVLAN_MC_FILTER_SZ); + bitmap_copy(mc_filter, filter, MACVLAN_MC_FILTER_SZ); } +} + +static void macvlan_set_mac_lists(struct net_device *dev) +{ + struct macvlan_dev *vlan = netdev_priv(dev); + + macvlan_compute_filter(vlan->mc_filter, dev, vlan); + dev_uc_sync(vlan->lowerdev, dev); dev_mc_sync(vlan->lowerdev, dev); + + /* This is slightly inaccurate as we're including the subscription + * list of vlan->lowerdev too. + * + * Bug alert: This only works if everyone has the same broadcast + * address as lowerdev. As soon as someone changes theirs this + * will break. + * + * However, this is already broken as when you change your broadcast + * address we don't get called. + * + * The solution is to maintain a list of broadcast addresses like + * we do for uc/mc, if you care. + */ + macvlan_compute_filter(vlan->port->mc_filter, vlan->lowerdev, NULL); } static int macvlan_change_mtu(struct net_device *dev, int new_mtu) -- cgit v0.10.2 From 2e280c188f06b190475d115e5a7e60f546196d34 Mon Sep 17 00:00:00 2001 From: Peter Robinson Date: Wed, 1 Jun 2016 13:28:58 +0100 Subject: stmmac: make platform drivers depend on their associated SoC There's not much point, except compile test, enabling the stmmac platform drivers unless their actual SoC is enabled. They're not useful without it. Signed-off-by: Peter Robinson Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig index cec147d..8f06a66 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Kconfig +++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig @@ -40,7 +40,7 @@ config DWMAC_GENERIC config DWMAC_IPQ806X tristate "QCA IPQ806x DWMAC support" default ARCH_QCOM - depends on OF + depends on OF && (ARCH_QCOM || COMPILE_TEST) select MFD_SYSCON help Support for QCA IPQ806X DWMAC Ethernet. @@ -53,7 +53,7 @@ config DWMAC_IPQ806X config DWMAC_LPC18XX tristate "NXP LPC18xx/43xx DWMAC support" default ARCH_LPC18XX - depends on OF + depends on OF && (ARCH_LPC18XX || COMPILE_TEST) select MFD_SYSCON ---help--- Support for NXP LPC18xx/43xx DWMAC Ethernet. @@ -61,7 +61,7 @@ config DWMAC_LPC18XX config DWMAC_MESON tristate "Amlogic Meson dwmac support" default ARCH_MESON - depends on OF + depends on OF && (ARCH_MESON || COMPILE_TEST) help Support for Ethernet controller on Amlogic Meson SoCs. @@ -72,7 +72,7 @@ config DWMAC_MESON config DWMAC_ROCKCHIP tristate "Rockchip dwmac support" default ARCH_ROCKCHIP - depends on OF + depends on OF && (ARCH_ROCKCHIP || COMPILE_TEST) select MFD_SYSCON help Support for Ethernet controller on Rockchip RK3288 SoC. @@ -83,7 +83,7 @@ config DWMAC_ROCKCHIP config DWMAC_SOCFPGA tristate "SOCFPGA dwmac support" default ARCH_SOCFPGA - depends on OF + depends on OF && (ARCH_SOCFPGA || COMPILE_TEST) select MFD_SYSCON help Support for ethernet controller on Altera SOCFPGA @@ -95,7 +95,7 @@ config DWMAC_SOCFPGA config DWMAC_STI tristate "STi GMAC support" default ARCH_STI - depends on OF + depends on OF && (ARCH_STI || COMPILE_TEST) select MFD_SYSCON ---help--- Support for ethernet controller on STi SOCs. @@ -107,7 +107,7 @@ config DWMAC_STI config DWMAC_SUNXI tristate "Allwinner GMAC support" default ARCH_SUNXI - depends on OF + depends on OF && (ARCH_SUNXI || COMPILE_TEST) ---help--- Support for Allwinner A20/A31 GMAC ethernet controllers. -- cgit v0.10.2 From c8296b9aa79bd3e4ac580e273684d8ca137eb005 Mon Sep 17 00:00:00 2001 From: Masaru Nagai Date: Thu, 2 Jun 2016 01:41:05 +0900 Subject: ravb: Remove manual pause frame transmit Writing a non-zero value to the manual PAUSE frame register (MPR) starts the transmission of a PAUSE frame. A PAUSE frame is sent in ravb_emac_init(), but it is not expected behavior. Signed-off-by: Masaru Nagai Signed-off-by: Kazuya Mizuguchi Signed-off-by: Yoshihiro Kaneko Acked-by: Sergei Shtylyov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index aa82131..5349284 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -362,8 +362,6 @@ static void ravb_emac_init(struct net_device *ndev) ravb_write(ndev, (ndev->dev_addr[4] << 8) | (ndev->dev_addr[5]), MALR); - ravb_write(ndev, 1, MPR); - /* E-MAC status register clear */ ravb_write(ndev, ECSR_ICD | ECSR_MPD, ECSR); -- cgit v0.10.2 From 1e56d5127d5c2142fcd26b93f4a9abacbd8f975d Mon Sep 17 00:00:00 2001 From: Ashok Raj Nagarajan Date: Sat, 28 May 2016 11:25:40 +0300 Subject: ath10k: fix diag_read to collect data for larger memory diag_read uses dma_alloc_coherent to allocate memory requested by the caller. If this memory requested is larger, more than DIAG_TRANSFER_LIMIT (2K), then it is likely that we may not get the requested memory and we would fail. To solve this, request dma_alloc_coherent for only DIAG_TRANSFER_LIMIT, and reuse this buffer multiple times as needed to copy the data requested in smaller chunks of size not more than DIAG_TRANSFER_LIMIT. Previously we were reading into the caller's only after getting the complete requested data. Fixes: 68c03249f388 ('ath10k: convert pci_alloc_consistent() to dma_alloc_coherent()') Signed-off-by: Ashok Raj Nagarajan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c index 8e8e1eb..6109293 100644 --- a/drivers/net/wireless/ath/ath10k/pci.c +++ b/drivers/net/wireless/ath/ath10k/pci.c @@ -869,7 +869,7 @@ static int ath10k_pci_diag_read_mem(struct ath10k *ar, u32 address, void *data, struct ath10k_pci *ar_pci = ath10k_pci_priv(ar); int ret = 0; u32 *buf; - unsigned int completed_nbytes, orig_nbytes, remaining_bytes; + unsigned int completed_nbytes, alloc_nbytes, remaining_bytes; struct ath10k_ce_pipe *ce_diag; /* Host buffer address in CE space */ u32 ce_data; @@ -887,9 +887,10 @@ static int ath10k_pci_diag_read_mem(struct ath10k *ar, u32 address, void *data, * 1) 4-byte alignment * 2) Buffer in DMA-able space */ - orig_nbytes = nbytes; + alloc_nbytes = min_t(unsigned int, nbytes, DIAG_TRANSFER_LIMIT); + data_buf = (unsigned char *)dma_alloc_coherent(ar->dev, - orig_nbytes, + alloc_nbytes, &ce_data_base, GFP_ATOMIC); @@ -897,9 +898,9 @@ static int ath10k_pci_diag_read_mem(struct ath10k *ar, u32 address, void *data, ret = -ENOMEM; goto done; } - memset(data_buf, 0, orig_nbytes); + memset(data_buf, 0, alloc_nbytes); - remaining_bytes = orig_nbytes; + remaining_bytes = nbytes; ce_data = ce_data_base; while (remaining_bytes) { nbytes = min_t(unsigned int, remaining_bytes, @@ -959,19 +960,22 @@ static int ath10k_pci_diag_read_mem(struct ath10k *ar, u32 address, void *data, } remaining_bytes -= nbytes; + + if (ret) { + ath10k_warn(ar, "failed to read diag value at 0x%x: %d\n", + address, ret); + break; + } + memcpy(data, data_buf, nbytes); + address += nbytes; - ce_data += nbytes; + data += nbytes; } done: - if (ret == 0) - memcpy(data, data_buf, orig_nbytes); - else - ath10k_warn(ar, "failed to read diag value at 0x%x: %d\n", - address, ret); if (data_buf) - dma_free_coherent(ar->dev, orig_nbytes, data_buf, + dma_free_coherent(ar->dev, alloc_nbytes, data_buf, ce_data_base); spin_unlock_bh(&ar_pci->ce_lock); -- cgit v0.10.2 From f5e307515b2b9994f5e30fa8311ef38fd71c52a3 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Sat, 28 May 2016 11:25:41 +0300 Subject: ath10k: fix error while writing 'simulate_fw_crash' debugfs Fix invalid argument error while writing 'simulate_fw_crash', though the funcionality is working fine we get an error 'invalid argument' because 'count' value is not returned properly (no reason to reduce the count value for removing the newline) Fixes the below write error: /sys/kernel/debug/ieee80211/phy0/ath10k# echo hw-restart > simulate_fw_crash -bash: echo: write error: Invalid argument Also move the 'conf_mutex' as it is really not required for fetching the userspace buffer. Reported-by: Maharaja Kennadyrajan Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Maharaja Kennadyrajan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c index e251155..54cb629 100644 --- a/drivers/net/wireless/ath/ath10k/debug.c +++ b/drivers/net/wireless/ath/ath10k/debug.c @@ -609,25 +609,23 @@ static ssize_t ath10k_write_simulate_fw_crash(struct file *file, char buf[32]; int ret; - mutex_lock(&ar->conf_mutex); - simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, user_buf, count); /* make sure that buf is null terminated */ buf[sizeof(buf) - 1] = 0; + /* drop the possible '\n' from the end */ + if (buf[count - 1] == '\n') + buf[count - 1] = 0; + + mutex_lock(&ar->conf_mutex); + if (ar->state != ATH10K_STATE_ON && ar->state != ATH10K_STATE_RESTARTED) { ret = -ENETDOWN; goto exit; } - /* drop the possible '\n' from the end */ - if (buf[count - 1] == '\n') { - buf[count - 1] = 0; - count--; - } - if (!strcmp(buf, "soft")) { ath10k_info(ar, "simulating soft firmware crash\n"); ret = ath10k_wmi_force_fw_hang(ar, WMI_FORCE_FW_HANG_ASSERT, 0); -- cgit v0.10.2 From 2e550b6d277bdf5daf579bb7952d1b183ff0948a Mon Sep 17 00:00:00 2001 From: Rajkumar Manoharan Date: Fri, 27 May 2016 20:15:55 +0530 Subject: ath10k: fix operating irq mode for ahb device Earlier when operating irq mode is legacy, interrupts are disabled and re-enabled based on num_msi_intrs. commit cfe9011a05a8 ("ath10k: remove MSI range support") replaced num_msi_intrs by oper_irq_mode. Since oper_irq_mode is not initialized for ahb devices (i.e qca4019), device boot up is failed during probe. Fixes: cfe9011a05a8 ("ath10k: remove MSI range support") Signed-off-by: Rajkumar Manoharan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c index bd62bc1..37ff36f 100644 --- a/drivers/net/wireless/ath/ath10k/ahb.c +++ b/drivers/net/wireless/ath/ath10k/ahb.c @@ -476,6 +476,7 @@ static irqreturn_t ath10k_ahb_interrupt_handler(int irq, void *arg) static int ath10k_ahb_request_irq_legacy(struct ath10k *ar) { + struct ath10k_pci *ar_pci = ath10k_pci_priv(ar); struct ath10k_ahb *ar_ahb = ath10k_ahb_priv(ar); int ret; @@ -487,6 +488,7 @@ static int ath10k_ahb_request_irq_legacy(struct ath10k *ar) ar_ahb->irq, ret); return ret; } + ar_pci->oper_irq_mode = ATH10K_PCI_IRQ_LEGACY; return 0; } -- cgit v0.10.2 From 10f8ec64a2a68960a87c08c08f0dfef97331076f Mon Sep 17 00:00:00 2001 From: Rajkumar Manoharan Date: Fri, 27 May 2016 20:15:56 +0530 Subject: ath10k: remove unused phy_mode_to_band Remove unused inline function phy_mode_to_band. Signed-off-by: Rajkumar Manoharan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c index 7bf1909..3b35c7a 100644 --- a/drivers/net/wireless/ath/ath10k/htt_rx.c +++ b/drivers/net/wireless/ath/ath10k/htt_rx.c @@ -2183,34 +2183,6 @@ static void ath10k_htt_rx_tx_mode_switch_ind(struct ath10k *ar, ath10k_mac_tx_push_pending(ar); } -static inline enum nl80211_band phy_mode_to_band(u32 phy_mode) -{ - enum nl80211_band band; - - switch (phy_mode) { - case MODE_11A: - case MODE_11NA_HT20: - case MODE_11NA_HT40: - case MODE_11AC_VHT20: - case MODE_11AC_VHT40: - case MODE_11AC_VHT80: - band = NL80211_BAND_5GHZ; - break; - case MODE_11G: - case MODE_11B: - case MODE_11GONLY: - case MODE_11NG_HT20: - case MODE_11NG_HT40: - case MODE_11AC_VHT20_2G: - case MODE_11AC_VHT40_2G: - case MODE_11AC_VHT80_2G: - default: - band = NL80211_BAND_2GHZ; - } - - return band; -} - void ath10k_htt_htc_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb) { bool release; -- cgit v0.10.2 From b855de0f57920e11695f2e9051a7511f43a0eb86 Mon Sep 17 00:00:00 2001 From: Rajkumar Manoharan Date: Fri, 27 May 2016 20:15:57 +0530 Subject: ath10k: update module description Update module description to advertise all supported QCA 802.11ac devices. Signed-off-by: Rajkumar Manoharan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index a003980..21b8c8c 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -2182,5 +2182,5 @@ void ath10k_core_destroy(struct ath10k *ar) EXPORT_SYMBOL(ath10k_core_destroy); MODULE_AUTHOR("Qualcomm Atheros"); -MODULE_DESCRIPTION("Core module for QCA988X PCIe devices."); +MODULE_DESCRIPTION("Core module for Qualcomm Atheros 802.11ac wireless LAN cards."); MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c index 6109293..4150189 100644 --- a/drivers/net/wireless/ath/ath10k/pci.c +++ b/drivers/net/wireless/ath/ath10k/pci.c @@ -3199,7 +3199,7 @@ static void __exit ath10k_pci_exit(void) module_exit(ath10k_pci_exit); MODULE_AUTHOR("Qualcomm Atheros"); -MODULE_DESCRIPTION("Driver support for Atheros QCA988X PCIe devices"); +MODULE_DESCRIPTION("Driver support for Qualcomm Atheros 802.11ac WLAN PCIe/AHB devices"); MODULE_LICENSE("Dual BSD/GPL"); /* QCA988x 2.0 firmware files */ -- cgit v0.10.2 From 64e001f41676bd2455bef8895a6e8eee333ede93 Mon Sep 17 00:00:00 2001 From: Rajkumar Manoharan Date: Fri, 27 May 2016 20:15:58 +0530 Subject: ath10k: add new ATH10K_FW_FEATURE_BTCOEX_PARAM This feature flag will be used for firmware to support BT-Coex feature without reloading firmware via WMI pdev param. To support Bluetooth coexistence pdev param, WMI_COEX_GPIO_SUPPORT of extended resource config should be enabled always. This firmware IE is used to configure WMI_COEX_GPIO_SUPPORT. Signed-off-by: Rajkumar Manoharan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index 21b8c8c..88b5122 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -258,6 +258,7 @@ static const char *const ath10k_core_fw_feature_str[] = { [ATH10K_FW_FEATURE_SUPPORTS_ADAPTIVE_CCA] = "adaptive-cca", [ATH10K_FW_FEATURE_MFP_SUPPORT] = "mfp", [ATH10K_FW_FEATURE_PEER_FLOW_CONTROL] = "peer-flow-ctrl", + [ATH10K_FW_FEATURE_BTCOEX_PARAM] = "btcoex-param", }; static unsigned int ath10k_core_get_fw_feature_str(char *buf, diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h index 1852e0e..4462c3f 100644 --- a/drivers/net/wireless/ath/ath10k/core.h +++ b/drivers/net/wireless/ath/ath10k/core.h @@ -535,6 +535,13 @@ enum ath10k_fw_features { */ ATH10K_FW_FEATURE_PEER_FLOW_CONTROL = 13, + /* Firmware supports BT-Coex without reloading firmware via pdev param. + * To support Bluetooth coexistence pdev param, WMI_COEX_GPIO_SUPPORT of + * extended resource config should be enabled always. This firmware IE + * is used to configure WMI_COEX_GPIO_SUPPORT. + */ + ATH10K_FW_FEATURE_BTCOEX_PARAM = 14, + /* keep last */ ATH10K_FW_FEATURE_COUNT, }; -- cgit v0.10.2 From 39136248cf8d2627fb5925f44aa7752e36b004f6 Mon Sep 17 00:00:00 2001 From: Rajkumar Manoharan Date: Fri, 27 May 2016 20:15:59 +0530 Subject: ath10k: add pdev param support to enable/disable btcoex 10.4 firmware has support to enable or disable btcoex functionality without reloading firmware via wmi pdev param. Add provision to send pdev param command via existing btcoex knob. Signed-off-by: Rajkumar Manoharan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index 88b5122..cedf127 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -1756,6 +1756,16 @@ int ath10k_core_start(struct ath10k *ar, enum ath10k_firmware_mode mode, if (test_bit(WMI_SERVICE_BSS_CHANNEL_INFO_64, ar->wmi.svc_map)) val |= WMI_10_4_BSS_CHANNEL_INFO_64; + /* 10.4 firmware supports BT-Coex without reloading firmware + * via pdev param. To support Bluetooth coexistence pdev param, + * WMI_COEX_GPIO_SUPPORT of extended resource config should be + * enabled always. + */ + if (test_bit(WMI_SERVICE_COEX_GPIO, ar->wmi.svc_map) && + test_bit(ATH10K_FW_FEATURE_BTCOEX_PARAM, + ar->running_fw->fw_file.fw_features)) + val |= WMI_10_4_COEX_GPIO_SUPPORT; + status = ath10k_mac_ext_resource_config(ar, val); if (status) { ath10k_err(ar, diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c index 54cb629..8fbb8f2 100644 --- a/drivers/net/wireless/ath/ath10k/debug.c +++ b/drivers/net/wireless/ath/ath10k/debug.c @@ -2125,6 +2125,7 @@ static ssize_t ath10k_write_btcoex(struct file *file, size_t buf_size; int ret; bool val; + u32 pdev_param; buf_size = min(count, (sizeof(buf) - 1)); if (copy_from_user(buf, ubuf, buf_size)) @@ -2148,14 +2149,25 @@ static ssize_t ath10k_write_btcoex(struct file *file, goto exit; } + pdev_param = ar->wmi.pdev_param->enable_btcoex; + if (test_bit(ATH10K_FW_FEATURE_BTCOEX_PARAM, + ar->running_fw->fw_file.fw_features)) { + ret = ath10k_wmi_pdev_set_param(ar, pdev_param, val); + if (ret) { + ath10k_warn(ar, "failed to enable btcoex: %d\n", ret); + ret = count; + goto exit; + } + } else { + ath10k_info(ar, "restarting firmware due to btcoex change"); + queue_work(ar->workqueue, &ar->restart_work); + } + if (val) set_bit(ATH10K_FLAG_BTCOEX, &ar->dev_flags); else clear_bit(ATH10K_FLAG_BTCOEX, &ar->dev_flags); - ath10k_info(ar, "restarting firmware due to btcoex change"); - - queue_work(ar->workqueue, &ar->restart_work); ret = count; exit: diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index 1dd415d..84a3e49 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -4468,6 +4468,19 @@ static int ath10k_start(struct ieee80211_hw *hw) } } + param = ar->wmi.pdev_param->enable_btcoex; + if (test_bit(WMI_SERVICE_COEX_GPIO, ar->wmi.svc_map) && + test_bit(ATH10K_FW_FEATURE_BTCOEX_PARAM, + ar->running_fw->fw_file.fw_features)) { + ret = ath10k_wmi_pdev_set_param(ar, param, 0); + if (ret) { + ath10k_warn(ar, + "failed to set btcoex param: %d\n", ret); + goto err_core_stop; + } + clear_bit(ATH10K_FLAG_BTCOEX, &ar->dev_flags); + } + ar->num_started_vdevs = 0; ath10k_regd_update(ar); diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c index 2c30032..6279ab4 100644 --- a/drivers/net/wireless/ath/ath10k/wmi.c +++ b/drivers/net/wireless/ath/ath10k/wmi.c @@ -1104,6 +1104,7 @@ static struct wmi_pdev_param_map wmi_pdev_param_map = { .wapi_mbssid_offset = WMI_PDEV_PARAM_UNSUPPORTED, .arp_srcaddr = WMI_PDEV_PARAM_UNSUPPORTED, .arp_dstaddr = WMI_PDEV_PARAM_UNSUPPORTED, + .enable_btcoex = WMI_PDEV_PARAM_UNSUPPORTED, }; static struct wmi_pdev_param_map wmi_10x_pdev_param_map = { @@ -1199,6 +1200,7 @@ static struct wmi_pdev_param_map wmi_10x_pdev_param_map = { .wapi_mbssid_offset = WMI_PDEV_PARAM_UNSUPPORTED, .arp_srcaddr = WMI_PDEV_PARAM_UNSUPPORTED, .arp_dstaddr = WMI_PDEV_PARAM_UNSUPPORTED, + .enable_btcoex = WMI_PDEV_PARAM_UNSUPPORTED, }; static struct wmi_pdev_param_map wmi_10_2_4_pdev_param_map = { @@ -1294,6 +1296,7 @@ static struct wmi_pdev_param_map wmi_10_2_4_pdev_param_map = { .wapi_mbssid_offset = WMI_PDEV_PARAM_UNSUPPORTED, .arp_srcaddr = WMI_PDEV_PARAM_UNSUPPORTED, .arp_dstaddr = WMI_PDEV_PARAM_UNSUPPORTED, + .enable_btcoex = WMI_PDEV_PARAM_UNSUPPORTED, }; /* firmware 10.2 specific mappings */ @@ -1550,6 +1553,7 @@ static struct wmi_pdev_param_map wmi_10_4_pdev_param_map = { .wapi_mbssid_offset = WMI_10_4_PDEV_PARAM_WAPI_MBSSID_OFFSET, .arp_srcaddr = WMI_10_4_PDEV_PARAM_ARP_SRCADDR, .arp_dstaddr = WMI_10_4_PDEV_PARAM_ARP_DSTADDR, + .enable_btcoex = WMI_10_4_PDEV_PARAM_ENABLE_BTCOEX, }; static const struct wmi_peer_flags_map wmi_peer_flags_map = { diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h index 9fdf47e..90f594e 100644 --- a/drivers/net/wireless/ath/ath10k/wmi.h +++ b/drivers/net/wireless/ath/ath10k/wmi.h @@ -3447,6 +3447,7 @@ struct wmi_pdev_param_map { u32 wapi_mbssid_offset; u32 arp_srcaddr; u32 arp_dstaddr; + u32 enable_btcoex; }; #define WMI_PDEV_PARAM_UNSUPPORTED 0 @@ -3760,6 +3761,9 @@ enum wmi_10_4_pdev_param { WMI_10_4_PDEV_PARAM_ATF_OBSS_NOISE_SCH, WMI_10_4_PDEV_PARAM_ATF_OBSS_NOISE_SCALING_FACTOR, WMI_10_4_PDEV_PARAM_CUST_TXPOWER_SCALE, + WMI_10_4_PDEV_PARAM_ATF_DYNAMIC_ENABLE, + WMI_10_4_PDEV_PARAM_ATF_SSID_GROUP_POLICY, + WMI_10_4_PDEV_PARAM_ENABLE_BTCOEX, }; struct wmi_pdev_set_param_cmd { -- cgit v0.10.2 From 280e762e9c72ea1292fcc5321d5b6e31ad05967f Mon Sep 17 00:00:00 2001 From: Raja Mani Date: Mon, 30 May 2016 18:41:23 +0530 Subject: ath10k: enable ipq4019 device probe in ahb module All the necessary patches to make wifi running (over AHB) on ipq4019 SoC are ready now. It's good to enable ipq4019 wifi device probing in ahb module and remove work in progress debug print. Device tree change is there in the public review by below commit message "qcom: ipq4019: add wifi nodes to ipq4019 SoC device tree" Signed-off-by: Tamizh chelvam Signed-off-by: Raja Mani Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c index 37ff36f..acec16b 100644 --- a/drivers/net/wireless/ath/ath10k/ahb.c +++ b/drivers/net/wireless/ath/ath10k/ahb.c @@ -25,10 +25,9 @@ #include "ahb.h" static const struct of_device_id ath10k_ahb_of_match[] = { - /* TODO: enable this entry once everything in place. - * { .compatible = "qcom,ipq4019-wifi", - * .data = (void *)ATH10K_HW_QCA4019 }, - */ + { .compatible = "qcom,ipq4019-wifi", + .data = (void *)ATH10K_HW_QCA4019 + }, { } }; @@ -920,8 +919,6 @@ int ath10k_ahb_init(void) { int ret; - printk(KERN_ERR "AHB support is still work in progress\n"); - ret = platform_driver_register(&ath10k_ahb_driver); if (ret) printk(KERN_ERR "failed to register ath10k ahb driver: %d\n", -- cgit v0.10.2 From 41cae08c76c60281a5a089e9c9dcc84803c3f43d Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Wed, 1 Jun 2016 23:29:15 +0530 Subject: net: ethernet: wiznet: Remove create_workqueue alloc_workqueue replaces deprecated create_workqueue(). A dedicated workqueue has been used since the workitems are involved in normal device operation. Workitems &priv->rx_work and &priv->tx_work, map to w5100_rx_work and w5100_tx_work respectively and are involved in receiving and transmitting packets. Forward progress under memory pressure is a requirement here. create_workqueue has been replaced with alloc_workqueue with max_active as 0 since there is no need for throttling the number of active work items. Since the driver may be used in memory reclaim path, WQ_MEM_RECLAIM has been set to guarantee forward progress. flush_workqueue is unnecessary since destroy_workqueue() itself calls drain_workqueue() which flushes repeatedly till the workqueue becomes empty. Hence the call to flush_workqueue() has been dropped. Signed-off-by: Bhaktipriya Shridhar Acked-by: Tejun Heo Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/wiznet/w5100.c b/drivers/net/ethernet/wiznet/w5100.c index 4f6255c..37ab46c 100644 --- a/drivers/net/ethernet/wiznet/w5100.c +++ b/drivers/net/ethernet/wiznet/w5100.c @@ -1154,7 +1154,7 @@ int w5100_probe(struct device *dev, const struct w5100_ops *ops, if (err < 0) goto err_register; - priv->xfer_wq = create_workqueue(netdev_name(ndev)); + priv->xfer_wq = alloc_workqueue(netdev_name(ndev), WQ_MEM_RECLAIM, 0); if (!priv->xfer_wq) { err = -ENOMEM; goto err_wq; @@ -1233,7 +1233,6 @@ int w5100_remove(struct device *dev) flush_work(&priv->setrx_work); flush_work(&priv->restart_work); - flush_workqueue(priv->xfer_wq); destroy_workqueue(priv->xfer_wq); unregister_netdev(ndev); -- cgit v0.10.2 From 684ff4ef5edd758c47929b852b4ea79be56f8bc0 Mon Sep 17 00:00:00 2001 From: Zhang Shengju Date: Tue, 31 May 2016 13:41:02 +0000 Subject: ovs: set name assign type of internal port Set name_assign_type of internal port to NET_NAME_USER. Signed-off-by: Zhang Shengju Acked-by: Pravin B Shelar Signed-off-by: David S. Miller diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c index 2ee48e4..434e04c 100644 --- a/net/openvswitch/vport-internal_dev.c +++ b/net/openvswitch/vport-internal_dev.c @@ -195,7 +195,7 @@ static struct vport *internal_dev_create(const struct vport_parms *parms) } vport->dev = alloc_netdev(sizeof(struct internal_dev), - parms->name, NET_NAME_UNKNOWN, do_setup); + parms->name, NET_NAME_USER, do_setup); if (!vport->dev) { err = -ENOMEM; goto error_free_vport; -- cgit v0.10.2 From b87ab6b8e517563be372b69bdabd96c672458547 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 1 Jun 2016 21:16:39 -0700 Subject: net: vrf: set operstate and mtu at link create The VRF device exists to define L3 domains and guide FIB lookups. As such its operstate is not relevant. Seeing 'state UNKNOWN' in the output of 'ip link show' can be confusing, so set operstate at link create. Similarly, the MTU for a VRF device is not used; any fragmentation of the payload is done on the output path based on the real egress device. An MTU of 1500 on the VRF device while enslaved devices have a higher MTU can lead to confusion. Since the VRF MTU is not relevant set to 64k similar to what is done for loopback. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index dff0884..d356f5d 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -504,6 +504,12 @@ static int vrf_dev_init(struct net_device *dev) dev->flags = IFF_MASTER | IFF_NOARP; + /* MTU is irrelevant for VRF device; set to 64k similar to lo */ + dev->mtu = 64 * 1024; + + /* similarly, oper state is irrelevant; set to up to avoid confusion */ + dev->operstate = IF_OPER_UP; + return 0; out_rth: -- cgit v0.10.2 From 351a4dedb34cbeb9f747f0e2309e891b6fb906cb Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Thu, 2 Jun 2016 10:23:29 +0300 Subject: qed: Utilize FW 8.10.3.0 The New QED firmware contains several fixes, including: - Wrong classification of packets in 4-port devices. - Anti-spoof interoperability with encapsulated packets. - Tx-switching of encapsulated packets. It also slightly improves Tx performance of the device. In addition, this firmware contains the necessary logic for supporting iscsi & rdma, for which we plan on pushing protocol drivers in the imminent future. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index 2d89e8c..e9ce6a7 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -244,6 +244,7 @@ static int qed_init_qm_info(struct qed_hwfn *p_hwfn, bool b_sleepable) qm_info->qm_pq_params[curr_queue].tc_id = p_hwfn->hw_info.non_offload_tc; qm_info->qm_pq_params[curr_queue].wrr_group = 1; + qm_info->qm_pq_params[curr_queue].rl_valid = 1; curr_queue++; } @@ -256,7 +257,10 @@ static int qed_init_qm_info(struct qed_hwfn *p_hwfn, bool b_sleepable) for (i = 0; i < num_ports; i++) { p_qm_port = &qm_info->qm_port_params[i]; p_qm_port->active = 1; - p_qm_port->num_active_phys_tcs = 4; + if (num_ports == 4) + p_qm_port->active_phys_tcs = 0x7; + else + p_qm_port->active_phys_tcs = 0x9f; p_qm_port->num_pbf_cmd_lines = PBF_MAX_CMD_LINES / num_ports; p_qm_port->num_btb_blocks = BTB_MAX_BLOCKS / num_ports; } @@ -703,8 +707,31 @@ static int qed_hw_init_port(struct qed_hwfn *p_hwfn, { int rc = 0; - rc = qed_init_run(p_hwfn, p_ptt, PHASE_PORT, p_hwfn->port_id, - hw_mode); + rc = qed_init_run(p_hwfn, p_ptt, PHASE_PORT, p_hwfn->port_id, hw_mode); + if (rc != 0) + return rc; + + if (hw_mode & (1 << MODE_MF_SI)) { + u8 pf_id = 0; + + if (!qed_hw_init_first_eth(p_hwfn, p_ptt, &pf_id)) { + DP_VERBOSE(p_hwfn, NETIF_MSG_IFUP, + "PF[%08x] is first eth on engine\n", pf_id); + + /* We should have configured BIT for ppfid, i.e., the + * relative function number in the port. But there's a + * bug in LLH in BB where the ppfid is actually engine + * based, so we need to take this into account. + */ + qed_wr(p_hwfn, p_ptt, + NIG_REG_LLH_TAGMAC_DEF_PF_VECTOR, 1 << pf_id); + } + + /* Take the protocol-based hit vector if there is a hit, + * otherwise take the other vector. + */ + qed_wr(p_hwfn, p_ptt, NIG_REG_LLH_CLS_TYPE_DUALMODE, 0x2); + } return rc; } @@ -773,6 +800,21 @@ static int qed_hw_init_pf(struct qed_hwfn *p_hwfn, /* Pure runtime initializations - directly to the HW */ qed_int_igu_init_pure_rt(p_hwfn, p_ptt, true, true); + if (hw_mode & (1 << MODE_MF_SI)) { + u8 pf_id = 0; + u32 val; + + if (!qed_hw_init_first_eth(p_hwfn, p_ptt, &pf_id)) { + if (p_hwfn->rel_pf_id == pf_id) { + DP_VERBOSE(p_hwfn, NETIF_MSG_IFUP, + "PF[%d] is first ETH on engine\n", + pf_id); + val = 1; + } + qed_wr(p_hwfn, p_ptt, PRS_REG_MSG_INFO, val); + } + } + if (b_hw_start) { /* enable interrupts */ qed_int_igu_enable(p_hwfn, p_ptt, int_mode); @@ -1304,31 +1346,31 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, switch ((core_cfg & NVM_CFG1_GLOB_NETWORK_PORT_MODE_MASK) >> NVM_CFG1_GLOB_NETWORK_PORT_MODE_OFFSET) { - case NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_2X40G: + case NVM_CFG1_GLOB_NETWORK_PORT_MODE_BB_2X40G: p_hwfn->hw_info.port_mode = QED_PORT_MODE_DE_2X40G; break; - case NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_2X50G: + case NVM_CFG1_GLOB_NETWORK_PORT_MODE_2X50G: p_hwfn->hw_info.port_mode = QED_PORT_MODE_DE_2X50G; break; - case NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_1X100G: + case NVM_CFG1_GLOB_NETWORK_PORT_MODE_BB_1X100G: p_hwfn->hw_info.port_mode = QED_PORT_MODE_DE_1X100G; break; - case NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_4X10G_F: + case NVM_CFG1_GLOB_NETWORK_PORT_MODE_4X10G_F: p_hwfn->hw_info.port_mode = QED_PORT_MODE_DE_4X10G_F; break; - case NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_4X10G_E: + case NVM_CFG1_GLOB_NETWORK_PORT_MODE_BB_4X10G_E: p_hwfn->hw_info.port_mode = QED_PORT_MODE_DE_4X10G_E; break; - case NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_4X20G: + case NVM_CFG1_GLOB_NETWORK_PORT_MODE_BB_4X20G: p_hwfn->hw_info.port_mode = QED_PORT_MODE_DE_4X20G; break; - case NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_1X40G: + case NVM_CFG1_GLOB_NETWORK_PORT_MODE_1X40G: p_hwfn->hw_info.port_mode = QED_PORT_MODE_DE_1X40G; break; - case NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_2X25G: + case NVM_CFG1_GLOB_NETWORK_PORT_MODE_2X25G: p_hwfn->hw_info.port_mode = QED_PORT_MODE_DE_2X25G; break; - case NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_1X25G: + case NVM_CFG1_GLOB_NETWORK_PORT_MODE_1X25G: p_hwfn->hw_info.port_mode = QED_PORT_MODE_DE_1X25G; break; default: @@ -1373,7 +1415,7 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, case NVM_CFG1_PORT_DRV_LINK_SPEED_50G: link->speed.forced_speed = 50000; break; - case NVM_CFG1_PORT_DRV_LINK_SPEED_100G: + case NVM_CFG1_PORT_DRV_LINK_SPEED_BB_100G: link->speed.forced_speed = 100000; break; default: diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index 9afc15f..d104ca7 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -21,9 +21,6 @@ struct qed_hwfn; struct qed_ptt; -/********************************/ -/* Add include to common target */ -/********************************/ /* opcodes for the event ring */ enum common_event_opcode { @@ -32,9 +29,10 @@ enum common_event_opcode { COMMON_EVENT_VF_START, COMMON_EVENT_VF_STOP, COMMON_EVENT_VF_PF_CHANNEL, - COMMON_EVENT_RESERVED4, - COMMON_EVENT_RESERVED5, - COMMON_EVENT_RESERVED6, + COMMON_EVENT_VF_FLR, + COMMON_EVENT_PF_UPDATE, + COMMON_EVENT_MALICIOUS_VF, + COMMON_EVENT_RL_UPDATE, COMMON_EVENT_EMPTY, MAX_COMMON_EVENT_OPCODE }; @@ -42,11 +40,12 @@ enum common_event_opcode { /* Common Ramrod Command IDs */ enum common_ramrod_cmd_id { COMMON_RAMROD_UNUSED, - COMMON_RAMROD_PF_START /* PF Function Start Ramrod */, - COMMON_RAMROD_PF_STOP /* PF Function Stop Ramrod */, + COMMON_RAMROD_PF_START, + COMMON_RAMROD_PF_STOP, COMMON_RAMROD_VF_START, COMMON_RAMROD_VF_STOP, COMMON_RAMROD_PF_UPDATE, + COMMON_RAMROD_RL_UPDATE, COMMON_RAMROD_EMPTY, MAX_COMMON_RAMROD_CMD_ID }; @@ -63,448 +62,448 @@ struct pstorm_core_conn_st_ctx { /* Core Slowpath Connection storm context of Xstorm */ struct xstorm_core_conn_st_ctx { - __le32 spq_base_lo /* SPQ Ring Base Address low dword */; - __le32 spq_base_hi /* SPQ Ring Base Address high dword */; - struct regpair consolid_base_addr; - __le16 spq_cons /* SPQ Ring Consumer */; - __le16 consolid_cons /* Consolidation Ring Consumer */; - __le32 reserved0[55] /* Pad to 15 cycles */; + __le32 spq_base_lo; + __le32 spq_base_hi; + struct regpair consolid_base_addr; + __le16 spq_cons; + __le16 consolid_cons; + __le32 reserved0[55]; }; struct xstorm_core_conn_ag_ctx { - u8 reserved0 /* cdu_validation */; - u8 core_state /* state */; - u8 flags0; -#define XSTORM_CORE_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 -#define XSTORM_CORE_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED1_MASK 0x1 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED1_SHIFT 1 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED2_MASK 0x1 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED2_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 -#define XSTORM_CORE_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED3_MASK 0x1 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED3_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED4_MASK 0x1 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED4_SHIFT 5 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED5_MASK 0x1 /* bit6 */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED5_SHIFT 6 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED6_MASK 0x1 /* bit7 */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED6_SHIFT 7 + u8 reserved0; + u8 core_state; + u8 flags0; +#define XSTORM_CORE_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED1_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED1_SHIFT 1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED2_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED2_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED3_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED3_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED4_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED4_SHIFT 5 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED5_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED5_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED6_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED6_SHIFT 7 u8 flags1; -#define XSTORM_CORE_CONN_AG_CTX_RESERVED7_MASK 0x1 /* bit8 */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED7_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED8_MASK 0x1 /* bit9 */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED8_SHIFT 1 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED9_MASK 0x1 /* bit10 */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED9_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_BIT11_MASK 0x1 /* bit11 */ -#define XSTORM_CORE_CONN_AG_CTX_BIT11_SHIFT 3 -#define XSTORM_CORE_CONN_AG_CTX_BIT12_MASK 0x1 /* bit12 */ -#define XSTORM_CORE_CONN_AG_CTX_BIT12_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_BIT13_MASK 0x1 /* bit13 */ -#define XSTORM_CORE_CONN_AG_CTX_BIT13_SHIFT 5 -#define XSTORM_CORE_CONN_AG_CTX_TX_RULE_ACTIVE_MASK 0x1 /* bit14 */ -#define XSTORM_CORE_CONN_AG_CTX_TX_RULE_ACTIVE_SHIFT 6 -#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_ACTIVE_MASK 0x1 /* bit15 */ -#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_ACTIVE_SHIFT 7 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED7_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED7_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED8_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED8_SHIFT 1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED9_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED9_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_BIT11_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_BIT11_SHIFT 3 +#define XSTORM_CORE_CONN_AG_CTX_BIT12_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_BIT12_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_BIT13_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_BIT13_SHIFT 5 +#define XSTORM_CORE_CONN_AG_CTX_TX_RULE_ACTIVE_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_TX_RULE_ACTIVE_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_ACTIVE_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_ACTIVE_SHIFT 7 u8 flags2; -#define XSTORM_CORE_CONN_AG_CTX_CF0_MASK 0x3 /* timer0cf */ -#define XSTORM_CORE_CONN_AG_CTX_CF0_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_CF1_MASK 0x3 /* timer1cf */ -#define XSTORM_CORE_CONN_AG_CTX_CF1_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_CF2_MASK 0x3 /* timer2cf */ -#define XSTORM_CORE_CONN_AG_CTX_CF2_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_CF3_MASK 0x3 -#define XSTORM_CORE_CONN_AG_CTX_CF3_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_CF0_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF0_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_CF1_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF1_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_CF2_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF2_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_CF3_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF3_SHIFT 6 u8 flags3; -#define XSTORM_CORE_CONN_AG_CTX_CF4_MASK 0x3 /* cf4 */ -#define XSTORM_CORE_CONN_AG_CTX_CF4_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_CF5_MASK 0x3 /* cf5 */ -#define XSTORM_CORE_CONN_AG_CTX_CF5_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_CF6_MASK 0x3 /* cf6 */ -#define XSTORM_CORE_CONN_AG_CTX_CF6_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_CF7_MASK 0x3 /* cf7 */ -#define XSTORM_CORE_CONN_AG_CTX_CF7_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_CF4_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF4_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_CF5_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF5_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_CF6_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF6_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_CF7_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF7_SHIFT 6 u8 flags4; -#define XSTORM_CORE_CONN_AG_CTX_CF8_MASK 0x3 /* cf8 */ -#define XSTORM_CORE_CONN_AG_CTX_CF8_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_CF9_MASK 0x3 /* cf9 */ -#define XSTORM_CORE_CONN_AG_CTX_CF9_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_CF10_MASK 0x3 /* cf10 */ -#define XSTORM_CORE_CONN_AG_CTX_CF10_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_CF11_MASK 0x3 /* cf11 */ -#define XSTORM_CORE_CONN_AG_CTX_CF11_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_CF8_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF8_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_CF9_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF9_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_CF10_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF10_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_CF11_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF11_SHIFT 6 u8 flags5; -#define XSTORM_CORE_CONN_AG_CTX_CF12_MASK 0x3 /* cf12 */ -#define XSTORM_CORE_CONN_AG_CTX_CF12_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_CF13_MASK 0x3 /* cf13 */ -#define XSTORM_CORE_CONN_AG_CTX_CF13_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_CF14_MASK 0x3 /* cf14 */ -#define XSTORM_CORE_CONN_AG_CTX_CF14_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_CF15_MASK 0x3 /* cf15 */ -#define XSTORM_CORE_CONN_AG_CTX_CF15_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_CF12_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF12_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_CF13_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF13_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_CF14_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF14_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_CF15_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF15_SHIFT 6 u8 flags6; -#define XSTORM_CORE_CONN_AG_CTX_CONSOLID_PROD_CF_MASK 0x3 /* cf16 */ -#define XSTORM_CORE_CONN_AG_CTX_CONSOLID_PROD_CF_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_CF17_MASK 0x3 -#define XSTORM_CORE_CONN_AG_CTX_CF17_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_MASK 0x3 /* cf18 */ -#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_TERMINATE_CF_MASK 0x3 /* cf19 */ -#define XSTORM_CORE_CONN_AG_CTX_TERMINATE_CF_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_CONSOLID_PROD_CF_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CONSOLID_PROD_CF_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_CF17_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF17_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_TERMINATE_CF_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_TERMINATE_CF_SHIFT 6 u8 flags7; -#define XSTORM_CORE_CONN_AG_CTX_FLUSH_Q0_MASK 0x3 /* cf20 */ -#define XSTORM_CORE_CONN_AG_CTX_FLUSH_Q0_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED10_MASK 0x3 /* cf21 */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED10_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_SLOW_PATH_MASK 0x3 /* cf22 */ -#define XSTORM_CORE_CONN_AG_CTX_SLOW_PATH_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_CF0EN_MASK 0x1 /* cf0en */ -#define XSTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT 6 -#define XSTORM_CORE_CONN_AG_CTX_CF1EN_MASK 0x1 /* cf1en */ -#define XSTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT 7 +#define XSTORM_CORE_CONN_AG_CTX_FLUSH_Q0_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_FLUSH_Q0_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED10_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED10_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_SLOW_PATH_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_SLOW_PATH_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_CF0EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_CF1EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT 7 u8 flags8; -#define XSTORM_CORE_CONN_AG_CTX_CF2EN_MASK 0x1 /* cf2en */ -#define XSTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_CF3EN_MASK 0x1 /* cf3en */ -#define XSTORM_CORE_CONN_AG_CTX_CF3EN_SHIFT 1 -#define XSTORM_CORE_CONN_AG_CTX_CF4EN_MASK 0x1 /* cf4en */ -#define XSTORM_CORE_CONN_AG_CTX_CF4EN_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_CF5EN_MASK 0x1 /* cf5en */ -#define XSTORM_CORE_CONN_AG_CTX_CF5EN_SHIFT 3 -#define XSTORM_CORE_CONN_AG_CTX_CF6EN_MASK 0x1 /* cf6en */ -#define XSTORM_CORE_CONN_AG_CTX_CF6EN_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_CF7EN_MASK 0x1 /* cf7en */ -#define XSTORM_CORE_CONN_AG_CTX_CF7EN_SHIFT 5 -#define XSTORM_CORE_CONN_AG_CTX_CF8EN_MASK 0x1 /* cf8en */ -#define XSTORM_CORE_CONN_AG_CTX_CF8EN_SHIFT 6 -#define XSTORM_CORE_CONN_AG_CTX_CF9EN_MASK 0x1 /* cf9en */ -#define XSTORM_CORE_CONN_AG_CTX_CF9EN_SHIFT 7 +#define XSTORM_CORE_CONN_AG_CTX_CF2EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_CF3EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF3EN_SHIFT 1 +#define XSTORM_CORE_CONN_AG_CTX_CF4EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF4EN_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_CF5EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF5EN_SHIFT 3 +#define XSTORM_CORE_CONN_AG_CTX_CF6EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF6EN_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_CF7EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF7EN_SHIFT 5 +#define XSTORM_CORE_CONN_AG_CTX_CF8EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF8EN_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_CF9EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF9EN_SHIFT 7 u8 flags9; -#define XSTORM_CORE_CONN_AG_CTX_CF10EN_MASK 0x1 /* cf10en */ -#define XSTORM_CORE_CONN_AG_CTX_CF10EN_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_CF11EN_MASK 0x1 /* cf11en */ -#define XSTORM_CORE_CONN_AG_CTX_CF11EN_SHIFT 1 -#define XSTORM_CORE_CONN_AG_CTX_CF12EN_MASK 0x1 /* cf12en */ -#define XSTORM_CORE_CONN_AG_CTX_CF12EN_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_CF13EN_MASK 0x1 /* cf13en */ -#define XSTORM_CORE_CONN_AG_CTX_CF13EN_SHIFT 3 -#define XSTORM_CORE_CONN_AG_CTX_CF14EN_MASK 0x1 /* cf14en */ -#define XSTORM_CORE_CONN_AG_CTX_CF14EN_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_CF15EN_MASK 0x1 /* cf15en */ -#define XSTORM_CORE_CONN_AG_CTX_CF15EN_SHIFT 5 -#define XSTORM_CORE_CONN_AG_CTX_CONSOLID_PROD_CF_EN_MASK 0x1 /* cf16en */ -#define XSTORM_CORE_CONN_AG_CTX_CONSOLID_PROD_CF_EN_SHIFT 6 -#define XSTORM_CORE_CONN_AG_CTX_CF17EN_MASK 0x1 -#define XSTORM_CORE_CONN_AG_CTX_CF17EN_SHIFT 7 +#define XSTORM_CORE_CONN_AG_CTX_CF10EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF10EN_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_CF11EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF11EN_SHIFT 1 +#define XSTORM_CORE_CONN_AG_CTX_CF12EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF12EN_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_CF13EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF13EN_SHIFT 3 +#define XSTORM_CORE_CONN_AG_CTX_CF14EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF14EN_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_CF15EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF15EN_SHIFT 5 +#define XSTORM_CORE_CONN_AG_CTX_CONSOLID_PROD_CF_EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CONSOLID_PROD_CF_EN_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_CF17EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF17EN_SHIFT 7 u8 flags10; -#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_EN_MASK 0x1 /* cf18en */ -#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_EN_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_TERMINATE_CF_EN_MASK 0x1 /* cf19en */ -#define XSTORM_CORE_CONN_AG_CTX_TERMINATE_CF_EN_SHIFT 1 -#define XSTORM_CORE_CONN_AG_CTX_FLUSH_Q0_EN_MASK 0x1 /* cf20en */ -#define XSTORM_CORE_CONN_AG_CTX_FLUSH_Q0_EN_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED11_MASK 0x1 /* cf21en */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED11_SHIFT 3 -#define XSTORM_CORE_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 /* cf22en */ -#define XSTORM_CORE_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_CF23EN_MASK 0x1 /* cf23en */ -#define XSTORM_CORE_CONN_AG_CTX_CF23EN_SHIFT 5 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED12_MASK 0x1 /* rule0en */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED12_SHIFT 6 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED13_MASK 0x1 /* rule1en */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED13_SHIFT 7 +#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_DQ_CF_EN_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_TERMINATE_CF_EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_TERMINATE_CF_EN_SHIFT 1 +#define XSTORM_CORE_CONN_AG_CTX_FLUSH_Q0_EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_FLUSH_Q0_EN_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED11_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED11_SHIFT 3 +#define XSTORM_CORE_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_CF23EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_CF23EN_SHIFT 5 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED12_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED12_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED13_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED13_SHIFT 7 u8 flags11; -#define XSTORM_CORE_CONN_AG_CTX_RESERVED14_MASK 0x1 /* rule2en */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED14_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_RESERVED15_MASK 0x1 /* rule3en */ -#define XSTORM_CORE_CONN_AG_CTX_RESERVED15_SHIFT 1 -#define XSTORM_CORE_CONN_AG_CTX_TX_DEC_RULE_EN_MASK 0x1 /* rule4en */ -#define XSTORM_CORE_CONN_AG_CTX_TX_DEC_RULE_EN_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_RULE5EN_MASK 0x1 /* rule5en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE5EN_SHIFT 3 -#define XSTORM_CORE_CONN_AG_CTX_RULE6EN_MASK 0x1 /* rule6en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE6EN_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_RULE7EN_MASK 0x1 /* rule7en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE7EN_SHIFT 5 -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 /* rule8en */ -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 -#define XSTORM_CORE_CONN_AG_CTX_RULE9EN_MASK 0x1 /* rule9en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE9EN_SHIFT 7 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED14_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED14_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED15_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RESERVED15_SHIFT 1 +#define XSTORM_CORE_CONN_AG_CTX_TX_DEC_RULE_EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_TX_DEC_RULE_EN_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE5EN_SHIFT 3 +#define XSTORM_CORE_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE6EN_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE7EN_SHIFT 5 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_RULE9EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE9EN_SHIFT 7 u8 flags12; -#define XSTORM_CORE_CONN_AG_CTX_RULE10EN_MASK 0x1 /* rule10en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE10EN_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_RULE11EN_MASK 0x1 /* rule11en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE11EN_SHIFT 1 -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 /* rule12en */ -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 /* rule13en */ -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 -#define XSTORM_CORE_CONN_AG_CTX_RULE14EN_MASK 0x1 /* rule14en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE14EN_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_RULE15EN_MASK 0x1 /* rule15en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE15EN_SHIFT 5 -#define XSTORM_CORE_CONN_AG_CTX_RULE16EN_MASK 0x1 /* rule16en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE16EN_SHIFT 6 -#define XSTORM_CORE_CONN_AG_CTX_RULE17EN_MASK 0x1 /* rule17en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE17EN_SHIFT 7 +#define XSTORM_CORE_CONN_AG_CTX_RULE10EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE10EN_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_RULE11EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE11EN_SHIFT 1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 +#define XSTORM_CORE_CONN_AG_CTX_RULE14EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE14EN_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_RULE15EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE15EN_SHIFT 5 +#define XSTORM_CORE_CONN_AG_CTX_RULE16EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE16EN_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_RULE17EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE17EN_SHIFT 7 u8 flags13; -#define XSTORM_CORE_CONN_AG_CTX_RULE18EN_MASK 0x1 /* rule18en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE18EN_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_RULE19EN_MASK 0x1 /* rule19en */ -#define XSTORM_CORE_CONN_AG_CTX_RULE19EN_SHIFT 1 -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 /* rule20en */ -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 /* rule21en */ -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 /* rule22en */ -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 /* rule23en */ -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 /* rule24en */ -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 /* rule25en */ -#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 +#define XSTORM_CORE_CONN_AG_CTX_RULE18EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE18EN_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_RULE19EN_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_RULE19EN_SHIFT 1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 u8 flags14; -#define XSTORM_CORE_CONN_AG_CTX_BIT16_MASK 0x1 /* bit16 */ -#define XSTORM_CORE_CONN_AG_CTX_BIT16_SHIFT 0 -#define XSTORM_CORE_CONN_AG_CTX_BIT17_MASK 0x1 /* bit17 */ -#define XSTORM_CORE_CONN_AG_CTX_BIT17_SHIFT 1 -#define XSTORM_CORE_CONN_AG_CTX_BIT18_MASK 0x1 /* bit18 */ -#define XSTORM_CORE_CONN_AG_CTX_BIT18_SHIFT 2 -#define XSTORM_CORE_CONN_AG_CTX_BIT19_MASK 0x1 /* bit19 */ -#define XSTORM_CORE_CONN_AG_CTX_BIT19_SHIFT 3 -#define XSTORM_CORE_CONN_AG_CTX_BIT20_MASK 0x1 /* bit20 */ -#define XSTORM_CORE_CONN_AG_CTX_BIT20_SHIFT 4 -#define XSTORM_CORE_CONN_AG_CTX_BIT21_MASK 0x1 /* bit21 */ -#define XSTORM_CORE_CONN_AG_CTX_BIT21_SHIFT 5 -#define XSTORM_CORE_CONN_AG_CTX_CF23_MASK 0x3 /* cf23 */ -#define XSTORM_CORE_CONN_AG_CTX_CF23_SHIFT 6 - u8 byte2 /* byte2 */; - __le16 physical_q0 /* physical_q0 */; - __le16 consolid_prod /* physical_q1 */; - __le16 reserved16 /* physical_q2 */; - __le16 tx_bd_cons /* word3 */; - __le16 tx_bd_or_spq_prod /* word4 */; - __le16 word5 /* word5 */; - __le16 conn_dpi /* conn_dpi */; - u8 byte3 /* byte3 */; - u8 byte4 /* byte4 */; - u8 byte5 /* byte5 */; - u8 byte6 /* byte6 */; - __le32 reg0 /* reg0 */; - __le32 reg1 /* reg1 */; - __le32 reg2 /* reg2 */; - __le32 reg3 /* reg3 */; - __le32 reg4 /* reg4 */; - __le32 reg5 /* cf_array0 */; - __le32 reg6 /* cf_array1 */; - __le16 word7 /* word7 */; - __le16 word8 /* word8 */; - __le16 word9 /* word9 */; - __le16 word10 /* word10 */; - __le32 reg7 /* reg7 */; - __le32 reg8 /* reg8 */; - __le32 reg9 /* reg9 */; - u8 byte7 /* byte7 */; - u8 byte8 /* byte8 */; - u8 byte9 /* byte9 */; - u8 byte10 /* byte10 */; - u8 byte11 /* byte11 */; - u8 byte12 /* byte12 */; - u8 byte13 /* byte13 */; - u8 byte14 /* byte14 */; - u8 byte15 /* byte15 */; - u8 byte16 /* byte16 */; - __le16 word11 /* word11 */; - __le32 reg10 /* reg10 */; - __le32 reg11 /* reg11 */; - __le32 reg12 /* reg12 */; - __le32 reg13 /* reg13 */; - __le32 reg14 /* reg14 */; - __le32 reg15 /* reg15 */; - __le32 reg16 /* reg16 */; - __le32 reg17 /* reg17 */; - __le32 reg18 /* reg18 */; - __le32 reg19 /* reg19 */; - __le16 word12 /* word12 */; - __le16 word13 /* word13 */; - __le16 word14 /* word14 */; - __le16 word15 /* word15 */; +#define XSTORM_CORE_CONN_AG_CTX_BIT16_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_BIT16_SHIFT 0 +#define XSTORM_CORE_CONN_AG_CTX_BIT17_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_BIT17_SHIFT 1 +#define XSTORM_CORE_CONN_AG_CTX_BIT18_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_BIT18_SHIFT 2 +#define XSTORM_CORE_CONN_AG_CTX_BIT19_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_BIT19_SHIFT 3 +#define XSTORM_CORE_CONN_AG_CTX_BIT20_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_BIT20_SHIFT 4 +#define XSTORM_CORE_CONN_AG_CTX_BIT21_MASK 0x1 +#define XSTORM_CORE_CONN_AG_CTX_BIT21_SHIFT 5 +#define XSTORM_CORE_CONN_AG_CTX_CF23_MASK 0x3 +#define XSTORM_CORE_CONN_AG_CTX_CF23_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 consolid_prod; + __le16 reserved16; + __le16 tx_bd_cons; + __le16 tx_bd_or_spq_prod; + __le16 word5; + __le16 conn_dpi; + u8 byte3; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le16 word7; + __le16 word8; + __le16 word9; + __le16 word10; + __le32 reg7; + __le32 reg8; + __le32 reg9; + u8 byte7; + u8 byte8; + u8 byte9; + u8 byte10; + u8 byte11; + u8 byte12; + u8 byte13; + u8 byte14; + u8 byte15; + u8 byte16; + __le16 word11; + __le32 reg10; + __le32 reg11; + __le32 reg12; + __le32 reg13; + __le32 reg14; + __le32 reg15; + __le32 reg16; + __le32 reg17; + __le32 reg18; + __le32 reg19; + __le16 word12; + __le16 word13; + __le16 word14; + __le16 word15; }; struct tstorm_core_conn_ag_ctx { - u8 byte0 /* cdu_validation */; - u8 byte1 /* state */; - u8 flags0; -#define TSTORM_CORE_CONN_AG_CTX_BIT0_MASK 0x1 /* exist_in_qm0 */ -#define TSTORM_CORE_CONN_AG_CTX_BIT0_SHIFT 0 -#define TSTORM_CORE_CONN_AG_CTX_BIT1_MASK 0x1 /* exist_in_qm1 */ -#define TSTORM_CORE_CONN_AG_CTX_BIT1_SHIFT 1 -#define TSTORM_CORE_CONN_AG_CTX_BIT2_MASK 0x1 /* bit2 */ -#define TSTORM_CORE_CONN_AG_CTX_BIT2_SHIFT 2 -#define TSTORM_CORE_CONN_AG_CTX_BIT3_MASK 0x1 /* bit3 */ -#define TSTORM_CORE_CONN_AG_CTX_BIT3_SHIFT 3 -#define TSTORM_CORE_CONN_AG_CTX_BIT4_MASK 0x1 /* bit4 */ -#define TSTORM_CORE_CONN_AG_CTX_BIT4_SHIFT 4 -#define TSTORM_CORE_CONN_AG_CTX_BIT5_MASK 0x1 /* bit5 */ -#define TSTORM_CORE_CONN_AG_CTX_BIT5_SHIFT 5 -#define TSTORM_CORE_CONN_AG_CTX_CF0_MASK 0x3 /* timer0cf */ -#define TSTORM_CORE_CONN_AG_CTX_CF0_SHIFT 6 + u8 byte0; + u8 byte1; + u8 flags0; +#define TSTORM_CORE_CONN_AG_CTX_BIT0_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_BIT0_SHIFT 0 +#define TSTORM_CORE_CONN_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_BIT1_SHIFT 1 +#define TSTORM_CORE_CONN_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_BIT2_SHIFT 2 +#define TSTORM_CORE_CONN_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_BIT3_SHIFT 3 +#define TSTORM_CORE_CONN_AG_CTX_BIT4_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_BIT4_SHIFT 4 +#define TSTORM_CORE_CONN_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_BIT5_SHIFT 5 +#define TSTORM_CORE_CONN_AG_CTX_CF0_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF0_SHIFT 6 u8 flags1; -#define TSTORM_CORE_CONN_AG_CTX_CF1_MASK 0x3 /* timer1cf */ -#define TSTORM_CORE_CONN_AG_CTX_CF1_SHIFT 0 -#define TSTORM_CORE_CONN_AG_CTX_CF2_MASK 0x3 /* timer2cf */ -#define TSTORM_CORE_CONN_AG_CTX_CF2_SHIFT 2 -#define TSTORM_CORE_CONN_AG_CTX_CF3_MASK 0x3 /* timer_stop_all */ -#define TSTORM_CORE_CONN_AG_CTX_CF3_SHIFT 4 -#define TSTORM_CORE_CONN_AG_CTX_CF4_MASK 0x3 /* cf4 */ -#define TSTORM_CORE_CONN_AG_CTX_CF4_SHIFT 6 +#define TSTORM_CORE_CONN_AG_CTX_CF1_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF1_SHIFT 0 +#define TSTORM_CORE_CONN_AG_CTX_CF2_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF2_SHIFT 2 +#define TSTORM_CORE_CONN_AG_CTX_CF3_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF3_SHIFT 4 +#define TSTORM_CORE_CONN_AG_CTX_CF4_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF4_SHIFT 6 u8 flags2; -#define TSTORM_CORE_CONN_AG_CTX_CF5_MASK 0x3 /* cf5 */ -#define TSTORM_CORE_CONN_AG_CTX_CF5_SHIFT 0 -#define TSTORM_CORE_CONN_AG_CTX_CF6_MASK 0x3 /* cf6 */ -#define TSTORM_CORE_CONN_AG_CTX_CF6_SHIFT 2 -#define TSTORM_CORE_CONN_AG_CTX_CF7_MASK 0x3 /* cf7 */ -#define TSTORM_CORE_CONN_AG_CTX_CF7_SHIFT 4 -#define TSTORM_CORE_CONN_AG_CTX_CF8_MASK 0x3 /* cf8 */ -#define TSTORM_CORE_CONN_AG_CTX_CF8_SHIFT 6 +#define TSTORM_CORE_CONN_AG_CTX_CF5_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF5_SHIFT 0 +#define TSTORM_CORE_CONN_AG_CTX_CF6_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF6_SHIFT 2 +#define TSTORM_CORE_CONN_AG_CTX_CF7_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF7_SHIFT 4 +#define TSTORM_CORE_CONN_AG_CTX_CF8_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF8_SHIFT 6 u8 flags3; -#define TSTORM_CORE_CONN_AG_CTX_CF9_MASK 0x3 /* cf9 */ -#define TSTORM_CORE_CONN_AG_CTX_CF9_SHIFT 0 -#define TSTORM_CORE_CONN_AG_CTX_CF10_MASK 0x3 /* cf10 */ -#define TSTORM_CORE_CONN_AG_CTX_CF10_SHIFT 2 -#define TSTORM_CORE_CONN_AG_CTX_CF0EN_MASK 0x1 /* cf0en */ -#define TSTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT 4 -#define TSTORM_CORE_CONN_AG_CTX_CF1EN_MASK 0x1 /* cf1en */ -#define TSTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT 5 -#define TSTORM_CORE_CONN_AG_CTX_CF2EN_MASK 0x1 /* cf2en */ -#define TSTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT 6 -#define TSTORM_CORE_CONN_AG_CTX_CF3EN_MASK 0x1 /* cf3en */ -#define TSTORM_CORE_CONN_AG_CTX_CF3EN_SHIFT 7 +#define TSTORM_CORE_CONN_AG_CTX_CF9_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF9_SHIFT 0 +#define TSTORM_CORE_CONN_AG_CTX_CF10_MASK 0x3 +#define TSTORM_CORE_CONN_AG_CTX_CF10_SHIFT 2 +#define TSTORM_CORE_CONN_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT 4 +#define TSTORM_CORE_CONN_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT 5 +#define TSTORM_CORE_CONN_AG_CTX_CF2EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT 6 +#define TSTORM_CORE_CONN_AG_CTX_CF3EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF3EN_SHIFT 7 u8 flags4; -#define TSTORM_CORE_CONN_AG_CTX_CF4EN_MASK 0x1 /* cf4en */ -#define TSTORM_CORE_CONN_AG_CTX_CF4EN_SHIFT 0 -#define TSTORM_CORE_CONN_AG_CTX_CF5EN_MASK 0x1 /* cf5en */ -#define TSTORM_CORE_CONN_AG_CTX_CF5EN_SHIFT 1 -#define TSTORM_CORE_CONN_AG_CTX_CF6EN_MASK 0x1 /* cf6en */ -#define TSTORM_CORE_CONN_AG_CTX_CF6EN_SHIFT 2 -#define TSTORM_CORE_CONN_AG_CTX_CF7EN_MASK 0x1 /* cf7en */ -#define TSTORM_CORE_CONN_AG_CTX_CF7EN_SHIFT 3 -#define TSTORM_CORE_CONN_AG_CTX_CF8EN_MASK 0x1 /* cf8en */ -#define TSTORM_CORE_CONN_AG_CTX_CF8EN_SHIFT 4 -#define TSTORM_CORE_CONN_AG_CTX_CF9EN_MASK 0x1 /* cf9en */ -#define TSTORM_CORE_CONN_AG_CTX_CF9EN_SHIFT 5 -#define TSTORM_CORE_CONN_AG_CTX_CF10EN_MASK 0x1 /* cf10en */ -#define TSTORM_CORE_CONN_AG_CTX_CF10EN_SHIFT 6 -#define TSTORM_CORE_CONN_AG_CTX_RULE0EN_MASK 0x1 /* rule0en */ -#define TSTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 7 +#define TSTORM_CORE_CONN_AG_CTX_CF4EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF4EN_SHIFT 0 +#define TSTORM_CORE_CONN_AG_CTX_CF5EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF5EN_SHIFT 1 +#define TSTORM_CORE_CONN_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF6EN_SHIFT 2 +#define TSTORM_CORE_CONN_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF7EN_SHIFT 3 +#define TSTORM_CORE_CONN_AG_CTX_CF8EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF8EN_SHIFT 4 +#define TSTORM_CORE_CONN_AG_CTX_CF9EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF9EN_SHIFT 5 +#define TSTORM_CORE_CONN_AG_CTX_CF10EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_CF10EN_SHIFT 6 +#define TSTORM_CORE_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 7 u8 flags5; -#define TSTORM_CORE_CONN_AG_CTX_RULE1EN_MASK 0x1 /* rule1en */ -#define TSTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 0 -#define TSTORM_CORE_CONN_AG_CTX_RULE2EN_MASK 0x1 /* rule2en */ -#define TSTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 1 -#define TSTORM_CORE_CONN_AG_CTX_RULE3EN_MASK 0x1 /* rule3en */ -#define TSTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 2 -#define TSTORM_CORE_CONN_AG_CTX_RULE4EN_MASK 0x1 /* rule4en */ -#define TSTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 3 -#define TSTORM_CORE_CONN_AG_CTX_RULE5EN_MASK 0x1 /* rule5en */ -#define TSTORM_CORE_CONN_AG_CTX_RULE5EN_SHIFT 4 -#define TSTORM_CORE_CONN_AG_CTX_RULE6EN_MASK 0x1 /* rule6en */ -#define TSTORM_CORE_CONN_AG_CTX_RULE6EN_SHIFT 5 -#define TSTORM_CORE_CONN_AG_CTX_RULE7EN_MASK 0x1 /* rule7en */ -#define TSTORM_CORE_CONN_AG_CTX_RULE7EN_SHIFT 6 -#define TSTORM_CORE_CONN_AG_CTX_RULE8EN_MASK 0x1 /* rule8en */ -#define TSTORM_CORE_CONN_AG_CTX_RULE8EN_SHIFT 7 - __le32 reg0 /* reg0 */; - __le32 reg1 /* reg1 */; - __le32 reg2 /* reg2 */; - __le32 reg3 /* reg3 */; - __le32 reg4 /* reg4 */; - __le32 reg5 /* reg5 */; - __le32 reg6 /* reg6 */; - __le32 reg7 /* reg7 */; - __le32 reg8 /* reg8 */; - u8 byte2 /* byte2 */; - u8 byte3 /* byte3 */; - __le16 word0 /* word0 */; - u8 byte4 /* byte4 */; - u8 byte5 /* byte5 */; - __le16 word1 /* word1 */; - __le16 word2 /* conn_dpi */; - __le16 word3 /* word3 */; - __le32 reg9 /* reg9 */; - __le32 reg10 /* reg10 */; +#define TSTORM_CORE_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define TSTORM_CORE_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define TSTORM_CORE_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define TSTORM_CORE_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define TSTORM_CORE_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define TSTORM_CORE_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define TSTORM_CORE_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define TSTORM_CORE_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define TSTORM_CORE_CONN_AG_CTX_RULE8EN_SHIFT 7 + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le32 reg7; + __le32 reg8; + u8 byte2; + u8 byte3; + __le16 word0; + u8 byte4; + u8 byte5; + __le16 word1; + __le16 word2; + __le16 word3; + __le32 reg9; + __le32 reg10; }; struct ustorm_core_conn_ag_ctx { - u8 reserved /* cdu_validation */; - u8 byte1 /* state */; - u8 flags0; -#define USTORM_CORE_CONN_AG_CTX_BIT0_MASK 0x1 /* exist_in_qm0 */ -#define USTORM_CORE_CONN_AG_CTX_BIT0_SHIFT 0 -#define USTORM_CORE_CONN_AG_CTX_BIT1_MASK 0x1 /* exist_in_qm1 */ -#define USTORM_CORE_CONN_AG_CTX_BIT1_SHIFT 1 -#define USTORM_CORE_CONN_AG_CTX_CF0_MASK 0x3 /* timer0cf */ -#define USTORM_CORE_CONN_AG_CTX_CF0_SHIFT 2 -#define USTORM_CORE_CONN_AG_CTX_CF1_MASK 0x3 /* timer1cf */ -#define USTORM_CORE_CONN_AG_CTX_CF1_SHIFT 4 -#define USTORM_CORE_CONN_AG_CTX_CF2_MASK 0x3 /* timer2cf */ -#define USTORM_CORE_CONN_AG_CTX_CF2_SHIFT 6 + u8 reserved; + u8 byte1; + u8 flags0; +#define USTORM_CORE_CONN_AG_CTX_BIT0_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_BIT0_SHIFT 0 +#define USTORM_CORE_CONN_AG_CTX_BIT1_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_BIT1_SHIFT 1 +#define USTORM_CORE_CONN_AG_CTX_CF0_MASK 0x3 +#define USTORM_CORE_CONN_AG_CTX_CF0_SHIFT 2 +#define USTORM_CORE_CONN_AG_CTX_CF1_MASK 0x3 +#define USTORM_CORE_CONN_AG_CTX_CF1_SHIFT 4 +#define USTORM_CORE_CONN_AG_CTX_CF2_MASK 0x3 +#define USTORM_CORE_CONN_AG_CTX_CF2_SHIFT 6 u8 flags1; -#define USTORM_CORE_CONN_AG_CTX_CF3_MASK 0x3 /* timer_stop_all */ -#define USTORM_CORE_CONN_AG_CTX_CF3_SHIFT 0 -#define USTORM_CORE_CONN_AG_CTX_CF4_MASK 0x3 /* cf4 */ -#define USTORM_CORE_CONN_AG_CTX_CF4_SHIFT 2 -#define USTORM_CORE_CONN_AG_CTX_CF5_MASK 0x3 /* cf5 */ -#define USTORM_CORE_CONN_AG_CTX_CF5_SHIFT 4 -#define USTORM_CORE_CONN_AG_CTX_CF6_MASK 0x3 /* cf6 */ -#define USTORM_CORE_CONN_AG_CTX_CF6_SHIFT 6 +#define USTORM_CORE_CONN_AG_CTX_CF3_MASK 0x3 +#define USTORM_CORE_CONN_AG_CTX_CF3_SHIFT 0 +#define USTORM_CORE_CONN_AG_CTX_CF4_MASK 0x3 +#define USTORM_CORE_CONN_AG_CTX_CF4_SHIFT 2 +#define USTORM_CORE_CONN_AG_CTX_CF5_MASK 0x3 +#define USTORM_CORE_CONN_AG_CTX_CF5_SHIFT 4 +#define USTORM_CORE_CONN_AG_CTX_CF6_MASK 0x3 +#define USTORM_CORE_CONN_AG_CTX_CF6_SHIFT 6 u8 flags2; -#define USTORM_CORE_CONN_AG_CTX_CF0EN_MASK 0x1 /* cf0en */ -#define USTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT 0 -#define USTORM_CORE_CONN_AG_CTX_CF1EN_MASK 0x1 /* cf1en */ -#define USTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT 1 -#define USTORM_CORE_CONN_AG_CTX_CF2EN_MASK 0x1 /* cf2en */ -#define USTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT 2 -#define USTORM_CORE_CONN_AG_CTX_CF3EN_MASK 0x1 /* cf3en */ -#define USTORM_CORE_CONN_AG_CTX_CF3EN_SHIFT 3 -#define USTORM_CORE_CONN_AG_CTX_CF4EN_MASK 0x1 /* cf4en */ -#define USTORM_CORE_CONN_AG_CTX_CF4EN_SHIFT 4 -#define USTORM_CORE_CONN_AG_CTX_CF5EN_MASK 0x1 /* cf5en */ -#define USTORM_CORE_CONN_AG_CTX_CF5EN_SHIFT 5 -#define USTORM_CORE_CONN_AG_CTX_CF6EN_MASK 0x1 /* cf6en */ -#define USTORM_CORE_CONN_AG_CTX_CF6EN_SHIFT 6 -#define USTORM_CORE_CONN_AG_CTX_RULE0EN_MASK 0x1 /* rule0en */ -#define USTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 7 +#define USTORM_CORE_CONN_AG_CTX_CF0EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT 0 +#define USTORM_CORE_CONN_AG_CTX_CF1EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT 1 +#define USTORM_CORE_CONN_AG_CTX_CF2EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT 2 +#define USTORM_CORE_CONN_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_CORE_CONN_AG_CTX_CF4EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_CF4EN_SHIFT 4 +#define USTORM_CORE_CONN_AG_CTX_CF5EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_CF5EN_SHIFT 5 +#define USTORM_CORE_CONN_AG_CTX_CF6EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_CF6EN_SHIFT 6 +#define USTORM_CORE_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 7 u8 flags3; -#define USTORM_CORE_CONN_AG_CTX_RULE1EN_MASK 0x1 /* rule1en */ -#define USTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 0 -#define USTORM_CORE_CONN_AG_CTX_RULE2EN_MASK 0x1 /* rule2en */ -#define USTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 1 -#define USTORM_CORE_CONN_AG_CTX_RULE3EN_MASK 0x1 /* rule3en */ -#define USTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 2 -#define USTORM_CORE_CONN_AG_CTX_RULE4EN_MASK 0x1 /* rule4en */ -#define USTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 3 -#define USTORM_CORE_CONN_AG_CTX_RULE5EN_MASK 0x1 /* rule5en */ -#define USTORM_CORE_CONN_AG_CTX_RULE5EN_SHIFT 4 -#define USTORM_CORE_CONN_AG_CTX_RULE6EN_MASK 0x1 /* rule6en */ -#define USTORM_CORE_CONN_AG_CTX_RULE6EN_SHIFT 5 -#define USTORM_CORE_CONN_AG_CTX_RULE7EN_MASK 0x1 /* rule7en */ -#define USTORM_CORE_CONN_AG_CTX_RULE7EN_SHIFT 6 -#define USTORM_CORE_CONN_AG_CTX_RULE8EN_MASK 0x1 /* rule8en */ -#define USTORM_CORE_CONN_AG_CTX_RULE8EN_SHIFT 7 - u8 byte2 /* byte2 */; - u8 byte3 /* byte3 */; - __le16 word0 /* conn_dpi */; - __le16 word1 /* word1 */; - __le32 rx_producers /* reg0 */; - __le32 reg1 /* reg1 */; - __le32 reg2 /* reg2 */; - __le32 reg3 /* reg3 */; - __le16 word2 /* word2 */; - __le16 word3 /* word3 */; +#define USTORM_CORE_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define USTORM_CORE_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define USTORM_CORE_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define USTORM_CORE_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define USTORM_CORE_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define USTORM_CORE_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define USTORM_CORE_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define USTORM_CORE_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define USTORM_CORE_CONN_AG_CTX_RULE8EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le16 word1; + __le32 rx_producers; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le16 word2; + __le16 word3; }; /* The core storm context for the Mstorm */ @@ -519,122 +518,186 @@ struct ustorm_core_conn_st_ctx { /* core connection context */ struct core_conn_context { - struct ystorm_core_conn_st_ctx ystorm_st_context; - struct regpair ystorm_st_padding[2] /* padding */; - struct pstorm_core_conn_st_ctx pstorm_st_context; - struct regpair pstorm_st_padding[2]; - struct xstorm_core_conn_st_ctx xstorm_st_context; - struct xstorm_core_conn_ag_ctx xstorm_ag_context; - struct tstorm_core_conn_ag_ctx tstorm_ag_context; - struct ustorm_core_conn_ag_ctx ustorm_ag_context; - struct mstorm_core_conn_st_ctx mstorm_st_context; - struct ustorm_core_conn_st_ctx ustorm_st_context; - struct regpair ustorm_st_padding[2] /* padding */; + struct ystorm_core_conn_st_ctx ystorm_st_context; + struct regpair ystorm_st_padding[2]; + struct pstorm_core_conn_st_ctx pstorm_st_context; + struct regpair pstorm_st_padding[2]; + struct xstorm_core_conn_st_ctx xstorm_st_context; + struct xstorm_core_conn_ag_ctx xstorm_ag_context; + struct tstorm_core_conn_ag_ctx tstorm_ag_context; + struct ustorm_core_conn_ag_ctx ustorm_ag_context; + struct mstorm_core_conn_st_ctx mstorm_st_context; + struct ustorm_core_conn_st_ctx ustorm_st_context; + struct regpair ustorm_st_padding[2]; +}; + +struct eth_mstorm_per_pf_stat { + struct regpair gre_discard_pkts; + struct regpair vxlan_discard_pkts; + struct regpair geneve_discard_pkts; + struct regpair lb_discard_pkts; }; struct eth_mstorm_per_queue_stat { - struct regpair ttl0_discard; - struct regpair packet_too_big_discard; - struct regpair no_buff_discard; - struct regpair not_active_discard; - struct regpair tpa_coalesced_pkts; - struct regpair tpa_coalesced_events; - struct regpair tpa_aborts_num; - struct regpair tpa_coalesced_bytes; + struct regpair ttl0_discard; + struct regpair packet_too_big_discard; + struct regpair no_buff_discard; + struct regpair not_active_discard; + struct regpair tpa_coalesced_pkts; + struct regpair tpa_coalesced_events; + struct regpair tpa_aborts_num; + struct regpair tpa_coalesced_bytes; +}; + +/* Ethernet TX Per PF */ +struct eth_pstorm_per_pf_stat { + struct regpair sent_lb_ucast_bytes; + struct regpair sent_lb_mcast_bytes; + struct regpair sent_lb_bcast_bytes; + struct regpair sent_lb_ucast_pkts; + struct regpair sent_lb_mcast_pkts; + struct regpair sent_lb_bcast_pkts; + struct regpair sent_gre_bytes; + struct regpair sent_vxlan_bytes; + struct regpair sent_geneve_bytes; + struct regpair sent_gre_pkts; + struct regpair sent_vxlan_pkts; + struct regpair sent_geneve_pkts; + struct regpair gre_drop_pkts; + struct regpair vxlan_drop_pkts; + struct regpair geneve_drop_pkts; +}; + +/* Ethernet TX Per Queue Stats */ +struct eth_pstorm_per_queue_stat { + struct regpair sent_ucast_bytes; + struct regpair sent_mcast_bytes; + struct regpair sent_bcast_bytes; + struct regpair sent_ucast_pkts; + struct regpair sent_mcast_pkts; + struct regpair sent_bcast_pkts; + struct regpair error_drop_pkts; +}; + +/* ETH Rx producers data */ +struct eth_rx_rate_limit { + __le16 mult; + __le16 cnst; + u8 add_sub_cnst; + u8 reserved0; + __le16 reserved1; }; -struct eth_pstorm_per_queue_stat { - struct regpair sent_ucast_bytes; - struct regpair sent_mcast_bytes; - struct regpair sent_bcast_bytes; - struct regpair sent_ucast_pkts; - struct regpair sent_mcast_pkts; - struct regpair sent_bcast_pkts; - struct regpair error_drop_pkts; +struct eth_ustorm_per_pf_stat { + struct regpair rcv_lb_ucast_bytes; + struct regpair rcv_lb_mcast_bytes; + struct regpair rcv_lb_bcast_bytes; + struct regpair rcv_lb_ucast_pkts; + struct regpair rcv_lb_mcast_pkts; + struct regpair rcv_lb_bcast_pkts; + struct regpair rcv_gre_bytes; + struct regpair rcv_vxlan_bytes; + struct regpair rcv_geneve_bytes; + struct regpair rcv_gre_pkts; + struct regpair rcv_vxlan_pkts; + struct regpair rcv_geneve_pkts; }; struct eth_ustorm_per_queue_stat { - struct regpair rcv_ucast_bytes; - struct regpair rcv_mcast_bytes; - struct regpair rcv_bcast_bytes; - struct regpair rcv_ucast_pkts; - struct regpair rcv_mcast_pkts; - struct regpair rcv_bcast_pkts; + struct regpair rcv_ucast_bytes; + struct regpair rcv_mcast_bytes; + struct regpair rcv_bcast_bytes; + struct regpair rcv_ucast_pkts; + struct regpair rcv_mcast_pkts; + struct regpair rcv_bcast_pkts; }; /* Event Ring Next Page Address */ struct event_ring_next_addr { - struct regpair addr /* Next Page Address */; - __le32 reserved[2] /* Reserved */; + struct regpair addr; + __le32 reserved[2]; }; +/* Event Ring Element */ union event_ring_element { - struct event_ring_entry entry /* Event Ring Entry */; - struct event_ring_next_addr next_addr; + struct event_ring_entry entry; + struct event_ring_next_addr next_addr; +}; + +/* Major and Minor hsi Versions */ +struct hsi_fp_ver_struct { + u8 minor_ver_arr[2]; + u8 major_ver_arr[2]; }; +/* Mstorm non-triggering VF zone */ struct mstorm_non_trigger_vf_zone { struct eth_mstorm_per_queue_stat eth_queue_stat; + struct eth_rx_prod_data eth_rx_queue_producers[ETH_MAX_NUM_RX_QUEUES_PER_VF]; }; +/* Mstorm VF zone */ struct mstorm_vf_zone { struct mstorm_non_trigger_vf_zone non_trigger; + }; +/* personality per PF */ enum personality_type { BAD_PERSONALITY_TYP, PERSONALITY_RESERVED, PERSONALITY_RESERVED2, - PERSONALITY_RDMA_AND_ETH /* Roce or Iwarp */, + PERSONALITY_RDMA_AND_ETH, PERSONALITY_RESERVED3, PERSONALITY_CORE, - PERSONALITY_ETH /* Ethernet */, + PERSONALITY_ETH, PERSONALITY_RESERVED4, MAX_PERSONALITY_TYPE }; +/* tunnel configuration */ struct pf_start_tunnel_config { - u8 set_vxlan_udp_port_flg; - u8 set_geneve_udp_port_flg; - u8 tx_enable_vxlan /* If set, enable VXLAN tunnel in TX path. */; - u8 tx_enable_l2geneve; - u8 tx_enable_ipgeneve; - u8 tx_enable_l2gre /* If set, enable l2 GRE tunnel in TX path. */; - u8 tx_enable_ipgre /* If set, enable IP GRE tunnel in TX path. */; - u8 tunnel_clss_vxlan /* Classification scheme for VXLAN tunnel. */; - u8 tunnel_clss_l2geneve; - u8 tunnel_clss_ipgeneve; - u8 tunnel_clss_l2gre; - u8 tunnel_clss_ipgre; - __le16 vxlan_udp_port /* VXLAN tunnel UDP destination port. */; - __le16 geneve_udp_port /* GENEVE tunnel UDP destination port. */; + u8 set_vxlan_udp_port_flg; + u8 set_geneve_udp_port_flg; + u8 tx_enable_vxlan; + u8 tx_enable_l2geneve; + u8 tx_enable_ipgeneve; + u8 tx_enable_l2gre; + u8 tx_enable_ipgre; + u8 tunnel_clss_vxlan; + u8 tunnel_clss_l2geneve; + u8 tunnel_clss_ipgeneve; + u8 tunnel_clss_l2gre; + u8 tunnel_clss_ipgre; + __le16 vxlan_udp_port; + __le16 geneve_udp_port; }; /* Ramrod data for PF start ramrod */ struct pf_start_ramrod_data { - struct regpair event_ring_pbl_addr; - struct regpair consolid_q_pbl_addr; - struct pf_start_tunnel_config tunnel_config; - __le16 event_ring_sb_id; - u8 base_vf_id; - u8 num_vfs; - u8 event_ring_num_pages; - u8 event_ring_sb_index; - u8 path_id; - u8 warning_as_error; - u8 dont_log_ramrods; - u8 personality; - __le16 log_type_mask; - u8 mf_mode /* Multi function mode */; - u8 integ_phase /* Integration phase */; - u8 allow_npar_tx_switching; - u8 inner_to_outer_pri_map[8]; - u8 pri_map_valid; - u32 outer_tag; - u8 reserved0[4]; -}; - -/* Data for port update ramrod */ + struct regpair event_ring_pbl_addr; + struct regpair consolid_q_pbl_addr; + struct pf_start_tunnel_config tunnel_config; + __le16 event_ring_sb_id; + u8 base_vf_id; + u8 num_vfs; + u8 event_ring_num_pages; + u8 event_ring_sb_index; + u8 path_id; + u8 warning_as_error; + u8 dont_log_ramrods; + u8 personality; + __le16 log_type_mask; + u8 mf_mode; + u8 integ_phase; + u8 allow_npar_tx_switching; + u8 inner_to_outer_pri_map[8]; + u8 pri_map_valid; + __le32 outer_tag; + struct hsi_fp_ver_struct hsi_fp_ver; + +}; + struct protocol_dcb_data { u8 dcb_enable_flag; u8 dcb_priority; @@ -642,25 +705,24 @@ struct protocol_dcb_data { u8 reserved; }; -/* tunnel configuration */ struct pf_update_tunnel_config { - u8 update_rx_pf_clss; - u8 update_tx_pf_clss; - u8 set_vxlan_udp_port_flg; - u8 set_geneve_udp_port_flg; - u8 tx_enable_vxlan; - u8 tx_enable_l2geneve; - u8 tx_enable_ipgeneve; - u8 tx_enable_l2gre; - u8 tx_enable_ipgre; - u8 tunnel_clss_vxlan; - u8 tunnel_clss_l2geneve; - u8 tunnel_clss_ipgeneve; - u8 tunnel_clss_l2gre; - u8 tunnel_clss_ipgre; - __le16 vxlan_udp_port; - __le16 geneve_udp_port; - __le16 reserved[3]; + u8 update_rx_pf_clss; + u8 update_tx_pf_clss; + u8 set_vxlan_udp_port_flg; + u8 set_geneve_udp_port_flg; + u8 tx_enable_vxlan; + u8 tx_enable_l2geneve; + u8 tx_enable_ipgeneve; + u8 tx_enable_l2gre; + u8 tx_enable_ipgre; + u8 tunnel_clss_vxlan; + u8 tunnel_clss_l2geneve; + u8 tunnel_clss_ipgeneve; + u8 tunnel_clss_l2gre; + u8 tunnel_clss_ipgre; + __le16 vxlan_udp_port; + __le16 geneve_udp_port; + __le16 reserved[3]; }; struct pf_update_ramrod_data { @@ -669,38 +731,43 @@ struct pf_update_ramrod_data { u8 update_fcoe_dcb_data_flag; u8 update_iscsi_dcb_data_flag; u8 update_roce_dcb_data_flag; + u8 update_iwarp_dcb_data_flag; u8 update_mf_vlan_flag; - __le16 mf_vlan; + u8 reserved; struct protocol_dcb_data eth_dcb_data; struct protocol_dcb_data fcoe_dcb_data; struct protocol_dcb_data iscsi_dcb_data; struct protocol_dcb_data roce_dcb_data; - struct pf_update_tunnel_config tunnel_config; -}; - -/* Tunnel classification scheme */ -enum tunnel_clss { - TUNNEL_CLSS_MAC_VLAN = 0, - TUNNEL_CLSS_MAC_VNI, - TUNNEL_CLSS_INNER_MAC_VLAN, - TUNNEL_CLSS_INNER_MAC_VNI, - MAX_TUNNEL_CLSS + struct protocol_dcb_data iwarp_dcb_data; + __le16 mf_vlan; + __le16 reserved2; + struct pf_update_tunnel_config tunnel_config; }; +/* Ports mode */ enum ports_mode { - ENGX2_PORTX1 /* 2 engines x 1 port */, - ENGX2_PORTX2 /* 2 engines x 2 ports */, - ENGX1_PORTX1 /* 1 engine x 1 port */, - ENGX1_PORTX2 /* 1 engine x 2 ports */, - ENGX1_PORTX4 /* 1 engine x 4 ports */, + ENGX2_PORTX1, + ENGX2_PORTX2, + ENGX1_PORTX1, + ENGX1_PORTX2, + ENGX1_PORTX4, MAX_PORTS_MODE }; +/* use to index in hsi_fp_[major|minor]_ver_arr per protocol */ +enum protocol_version_array_key { + ETH_VER_KEY = 0, + ROCE_VER_KEY, + MAX_PROTOCOL_VERSION_ARRAY_KEY +}; + +/* Pstorm non-triggering VF zone */ struct pstorm_non_trigger_vf_zone { struct eth_pstorm_per_queue_stat eth_queue_stat; struct regpair reserved[2]; }; +/* Pstorm VF zone */ struct pstorm_vf_zone { struct pstorm_non_trigger_vf_zone non_trigger; struct regpair reserved[7]; @@ -708,56 +775,89 @@ struct pstorm_vf_zone { /* Ramrod Header of SPQE */ struct ramrod_header { - __le32 cid /* Slowpath Connection CID */; - u8 cmd_id /* Ramrod Cmd (Per Protocol Type) */; - u8 protocol_id /* Ramrod Protocol ID */; - __le16 echo /* Ramrod echo */; + __le32 cid; + u8 cmd_id; + u8 protocol_id; + __le16 echo; }; /* Slowpath Element (SPQE) */ struct slow_path_element { - struct ramrod_header hdr /* Ramrod Header */; - struct regpair data_ptr; + struct ramrod_header hdr; + struct regpair data_ptr; +}; + +/* Tstorm non-triggering VF zone */ +struct tstorm_non_trigger_vf_zone { + struct regpair reserved[2]; }; struct tstorm_per_port_stat { - struct regpair trunc_error_discard; - struct regpair mac_error_discard; - struct regpair mftag_filter_discard; - struct regpair eth_mac_filter_discard; - struct regpair ll2_mac_filter_discard; - struct regpair ll2_conn_disabled_discard; - struct regpair iscsi_irregular_pkt; - struct regpair fcoe_irregular_pkt; - struct regpair roce_irregular_pkt; - struct regpair eth_irregular_pkt; - struct regpair toe_irregular_pkt; - struct regpair preroce_irregular_pkt; + struct regpair trunc_error_discard; + struct regpair mac_error_discard; + struct regpair mftag_filter_discard; + struct regpair eth_mac_filter_discard; + struct regpair reserved[5]; + struct regpair eth_irregular_pkt; + struct regpair reserved1[2]; + struct regpair eth_gre_tunn_filter_discard; + struct regpair eth_vxlan_tunn_filter_discard; + struct regpair eth_geneve_tunn_filter_discard; }; +/* Tstorm VF zone */ +struct tstorm_vf_zone { + struct tstorm_non_trigger_vf_zone non_trigger; +}; + +/* Tunnel classification scheme */ +enum tunnel_clss { + TUNNEL_CLSS_MAC_VLAN = 0, + TUNNEL_CLSS_MAC_VNI, + TUNNEL_CLSS_INNER_MAC_VLAN, + TUNNEL_CLSS_INNER_MAC_VNI, + TUNNEL_CLSS_MAC_VLAN_DUAL_STAGE, + MAX_TUNNEL_CLSS +}; + +/* Ustorm non-triggering VF zone */ struct ustorm_non_trigger_vf_zone { struct eth_ustorm_per_queue_stat eth_queue_stat; struct regpair vf_pf_msg_addr; }; +/* Ustorm triggering VF zone */ struct ustorm_trigger_vf_zone { u8 vf_pf_msg_valid; u8 reserved[7]; }; +/* Ustorm VF zone */ struct ustorm_vf_zone { struct ustorm_non_trigger_vf_zone non_trigger; struct ustorm_trigger_vf_zone trigger; }; +/* VF-PF channel data */ +struct vf_pf_channel_data { + __le32 ready; + u8 valid; + u8 reserved0; + __le16 reserved1; +}; + +/* Ramrod data for VF start ramrod */ struct vf_start_ramrod_data { u8 vf_id; u8 enable_flr_ack; __le16 opaque_fid; u8 personality; - u8 reserved[3]; + u8 reserved[7]; + struct hsi_fp_ver_struct hsi_fp_ver; + }; +/* Ramrod data for VF start ramrod */ struct vf_stop_ramrod_data { u8 vf_id; u8 reserved0; @@ -765,94 +865,474 @@ struct vf_stop_ramrod_data { __le32 reserved2; }; +/* Attentions status block */ struct atten_status_block { - __le32 atten_bits; - __le32 atten_ack; - __le16 reserved0; - __le16 sb_index /* status block running index */; - __le32 reserved1; + __le32 atten_bits; + __le32 atten_ack; + __le16 reserved0; + __le16 sb_index; + __le32 reserved1; +}; + +enum command_type_bit { + IGU_COMMAND_TYPE_NOP = 0, + IGU_COMMAND_TYPE_SET = 1, + MAX_COMMAND_TYPE_BIT +}; + +/* DMAE command */ +struct dmae_cmd { + __le32 opcode; +#define DMAE_CMD_SRC_MASK 0x1 +#define DMAE_CMD_SRC_SHIFT 0 +#define DMAE_CMD_DST_MASK 0x3 +#define DMAE_CMD_DST_SHIFT 1 +#define DMAE_CMD_C_DST_MASK 0x1 +#define DMAE_CMD_C_DST_SHIFT 3 +#define DMAE_CMD_CRC_RESET_MASK 0x1 +#define DMAE_CMD_CRC_RESET_SHIFT 4 +#define DMAE_CMD_SRC_ADDR_RESET_MASK 0x1 +#define DMAE_CMD_SRC_ADDR_RESET_SHIFT 5 +#define DMAE_CMD_DST_ADDR_RESET_MASK 0x1 +#define DMAE_CMD_DST_ADDR_RESET_SHIFT 6 +#define DMAE_CMD_COMP_FUNC_MASK 0x1 +#define DMAE_CMD_COMP_FUNC_SHIFT 7 +#define DMAE_CMD_COMP_WORD_EN_MASK 0x1 +#define DMAE_CMD_COMP_WORD_EN_SHIFT 8 +#define DMAE_CMD_COMP_CRC_EN_MASK 0x1 +#define DMAE_CMD_COMP_CRC_EN_SHIFT 9 +#define DMAE_CMD_COMP_CRC_OFFSET_MASK 0x7 +#define DMAE_CMD_COMP_CRC_OFFSET_SHIFT 10 +#define DMAE_CMD_RESERVED1_MASK 0x1 +#define DMAE_CMD_RESERVED1_SHIFT 13 +#define DMAE_CMD_ENDIANITY_MODE_MASK 0x3 +#define DMAE_CMD_ENDIANITY_MODE_SHIFT 14 +#define DMAE_CMD_ERR_HANDLING_MASK 0x3 +#define DMAE_CMD_ERR_HANDLING_SHIFT 16 +#define DMAE_CMD_PORT_ID_MASK 0x3 +#define DMAE_CMD_PORT_ID_SHIFT 18 +#define DMAE_CMD_SRC_PF_ID_MASK 0xF +#define DMAE_CMD_SRC_PF_ID_SHIFT 20 +#define DMAE_CMD_DST_PF_ID_MASK 0xF +#define DMAE_CMD_DST_PF_ID_SHIFT 24 +#define DMAE_CMD_SRC_VF_ID_VALID_MASK 0x1 +#define DMAE_CMD_SRC_VF_ID_VALID_SHIFT 28 +#define DMAE_CMD_DST_VF_ID_VALID_MASK 0x1 +#define DMAE_CMD_DST_VF_ID_VALID_SHIFT 29 +#define DMAE_CMD_RESERVED2_MASK 0x3 +#define DMAE_CMD_RESERVED2_SHIFT 30 + __le32 src_addr_lo; + __le32 src_addr_hi; + __le32 dst_addr_lo; + __le32 dst_addr_hi; + __le16 length_dw; + __le16 opcode_b; +#define DMAE_CMD_SRC_VF_ID_MASK 0xFF +#define DMAE_CMD_SRC_VF_ID_SHIFT 0 +#define DMAE_CMD_DST_VF_ID_MASK 0xFF +#define DMAE_CMD_DST_VF_ID_SHIFT 8 + __le32 comp_addr_lo; + __le32 comp_addr_hi; + __le32 comp_val; + __le32 crc32; + __le32 crc_32_c; + __le16 crc16; + __le16 crc16_c; + __le16 crc10; + __le16 reserved; + __le16 xsum16; + __le16 xsum8; +}; + +enum dmae_cmd_comp_crc_en_enum { + dmae_cmd_comp_crc_disabled, + dmae_cmd_comp_crc_enabled, + MAX_DMAE_CMD_COMP_CRC_EN_ENUM +}; + +enum dmae_cmd_comp_func_enum { + dmae_cmd_comp_func_to_src, + dmae_cmd_comp_func_to_dst, + MAX_DMAE_CMD_COMP_FUNC_ENUM +}; + +enum dmae_cmd_comp_word_en_enum { + dmae_cmd_comp_word_disabled, + dmae_cmd_comp_word_enabled, + MAX_DMAE_CMD_COMP_WORD_EN_ENUM +}; + +enum dmae_cmd_c_dst_enum { + dmae_cmd_c_dst_pcie, + dmae_cmd_c_dst_grc, + MAX_DMAE_CMD_C_DST_ENUM +}; + +enum dmae_cmd_dst_enum { + dmae_cmd_dst_none_0, + dmae_cmd_dst_pcie, + dmae_cmd_dst_grc, + dmae_cmd_dst_none_3, + MAX_DMAE_CMD_DST_ENUM +}; + +enum dmae_cmd_error_handling_enum { + dmae_cmd_error_handling_send_regular_comp, + dmae_cmd_error_handling_send_comp_with_err, + dmae_cmd_error_handling_dont_send_comp, + MAX_DMAE_CMD_ERROR_HANDLING_ENUM +}; + +enum dmae_cmd_src_enum { + dmae_cmd_src_pcie, + dmae_cmd_src_grc, + MAX_DMAE_CMD_SRC_ENUM +}; + +/* IGU cleanup command */ +struct igu_cleanup { + __le32 sb_id_and_flags; +#define IGU_CLEANUP_RESERVED0_MASK 0x7FFFFFF +#define IGU_CLEANUP_RESERVED0_SHIFT 0 +#define IGU_CLEANUP_CLEANUP_SET_MASK 0x1 +#define IGU_CLEANUP_CLEANUP_SET_SHIFT 27 +#define IGU_CLEANUP_CLEANUP_TYPE_MASK 0x7 +#define IGU_CLEANUP_CLEANUP_TYPE_SHIFT 28 +#define IGU_CLEANUP_COMMAND_TYPE_MASK 0x1 +#define IGU_CLEANUP_COMMAND_TYPE_SHIFT 31 + __le32 reserved1; +}; + +/* IGU firmware driver command */ +union igu_command { + struct igu_prod_cons_update prod_cons_update; + struct igu_cleanup cleanup; +}; + +/* IGU firmware driver command */ +struct igu_command_reg_ctrl { + __le16 opaque_fid; + __le16 igu_command_reg_ctrl_fields; +#define IGU_COMMAND_REG_CTRL_PXP_BAR_ADDR_MASK 0xFFF +#define IGU_COMMAND_REG_CTRL_PXP_BAR_ADDR_SHIFT 0 +#define IGU_COMMAND_REG_CTRL_RESERVED_MASK 0x7 +#define IGU_COMMAND_REG_CTRL_RESERVED_SHIFT 12 +#define IGU_COMMAND_REG_CTRL_COMMAND_TYPE_MASK 0x1 +#define IGU_COMMAND_REG_CTRL_COMMAND_TYPE_SHIFT 15 +}; + +/* IGU mapping line structure */ +struct igu_mapping_line { + __le32 igu_mapping_line_fields; +#define IGU_MAPPING_LINE_VALID_MASK 0x1 +#define IGU_MAPPING_LINE_VALID_SHIFT 0 +#define IGU_MAPPING_LINE_VECTOR_NUMBER_MASK 0xFF +#define IGU_MAPPING_LINE_VECTOR_NUMBER_SHIFT 1 +#define IGU_MAPPING_LINE_FUNCTION_NUMBER_MASK 0xFF +#define IGU_MAPPING_LINE_FUNCTION_NUMBER_SHIFT 9 +#define IGU_MAPPING_LINE_PF_VALID_MASK 0x1 +#define IGU_MAPPING_LINE_PF_VALID_SHIFT 17 +#define IGU_MAPPING_LINE_IPS_GROUP_MASK 0x3F +#define IGU_MAPPING_LINE_IPS_GROUP_SHIFT 18 +#define IGU_MAPPING_LINE_RESERVED_MASK 0xFF +#define IGU_MAPPING_LINE_RESERVED_SHIFT 24 +}; + +/* IGU MSIX line structure */ +struct igu_msix_vector { + struct regpair address; + __le32 data; + __le32 msix_vector_fields; +#define IGU_MSIX_VECTOR_MASK_BIT_MASK 0x1 +#define IGU_MSIX_VECTOR_MASK_BIT_SHIFT 0 +#define IGU_MSIX_VECTOR_RESERVED0_MASK 0x7FFF +#define IGU_MSIX_VECTOR_RESERVED0_SHIFT 1 +#define IGU_MSIX_VECTOR_STEERING_TAG_MASK 0xFF +#define IGU_MSIX_VECTOR_STEERING_TAG_SHIFT 16 +#define IGU_MSIX_VECTOR_RESERVED1_MASK 0xFF +#define IGU_MSIX_VECTOR_RESERVED1_SHIFT 24 +}; + +struct mstorm_core_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define MSTORM_CORE_CONN_AG_CTX_BIT0_MASK 0x1 +#define MSTORM_CORE_CONN_AG_CTX_BIT0_SHIFT 0 +#define MSTORM_CORE_CONN_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_CORE_CONN_AG_CTX_BIT1_SHIFT 1 +#define MSTORM_CORE_CONN_AG_CTX_CF0_MASK 0x3 +#define MSTORM_CORE_CONN_AG_CTX_CF0_SHIFT 2 +#define MSTORM_CORE_CONN_AG_CTX_CF1_MASK 0x3 +#define MSTORM_CORE_CONN_AG_CTX_CF1_SHIFT 4 +#define MSTORM_CORE_CONN_AG_CTX_CF2_MASK 0x3 +#define MSTORM_CORE_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define MSTORM_CORE_CONN_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT 0 +#define MSTORM_CORE_CONN_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT 1 +#define MSTORM_CORE_CONN_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT 2 +#define MSTORM_CORE_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define MSTORM_CORE_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define MSTORM_CORE_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define MSTORM_CORE_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define MSTORM_CORE_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 7 + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; }; +/* per encapsulation type enabling flags */ +struct prs_reg_encapsulation_type_en { + u8 flags; +#define PRS_REG_ENCAPSULATION_TYPE_EN_ETH_OVER_GRE_ENABLE_MASK 0x1 +#define PRS_REG_ENCAPSULATION_TYPE_EN_ETH_OVER_GRE_ENABLE_SHIFT 0 +#define PRS_REG_ENCAPSULATION_TYPE_EN_IP_OVER_GRE_ENABLE_MASK 0x1 +#define PRS_REG_ENCAPSULATION_TYPE_EN_IP_OVER_GRE_ENABLE_SHIFT 1 +#define PRS_REG_ENCAPSULATION_TYPE_EN_VXLAN_ENABLE_MASK 0x1 +#define PRS_REG_ENCAPSULATION_TYPE_EN_VXLAN_ENABLE_SHIFT 2 +#define PRS_REG_ENCAPSULATION_TYPE_EN_T_TAG_ENABLE_MASK 0x1 +#define PRS_REG_ENCAPSULATION_TYPE_EN_T_TAG_ENABLE_SHIFT 3 +#define PRS_REG_ENCAPSULATION_TYPE_EN_ETH_OVER_GENEVE_ENABLE_MASK 0x1 +#define PRS_REG_ENCAPSULATION_TYPE_EN_ETH_OVER_GENEVE_ENABLE_SHIFT 4 +#define PRS_REG_ENCAPSULATION_TYPE_EN_IP_OVER_GENEVE_ENABLE_MASK 0x1 +#define PRS_REG_ENCAPSULATION_TYPE_EN_IP_OVER_GENEVE_ENABLE_SHIFT 5 +#define PRS_REG_ENCAPSULATION_TYPE_EN_RESERVED_MASK 0x3 +#define PRS_REG_ENCAPSULATION_TYPE_EN_RESERVED_SHIFT 6 +}; + +enum pxp_tph_st_hint { + TPH_ST_HINT_BIDIR, + TPH_ST_HINT_REQUESTER, + TPH_ST_HINT_TARGET, + TPH_ST_HINT_TARGET_PRIO, + MAX_PXP_TPH_ST_HINT +}; + +/* QM hardware structure of enable bypass credit mask */ +struct qm_rf_bypass_mask { + u8 flags; +#define QM_RF_BYPASS_MASK_LINEVOQ_MASK 0x1 +#define QM_RF_BYPASS_MASK_LINEVOQ_SHIFT 0 +#define QM_RF_BYPASS_MASK_RESERVED0_MASK 0x1 +#define QM_RF_BYPASS_MASK_RESERVED0_SHIFT 1 +#define QM_RF_BYPASS_MASK_PFWFQ_MASK 0x1 +#define QM_RF_BYPASS_MASK_PFWFQ_SHIFT 2 +#define QM_RF_BYPASS_MASK_VPWFQ_MASK 0x1 +#define QM_RF_BYPASS_MASK_VPWFQ_SHIFT 3 +#define QM_RF_BYPASS_MASK_PFRL_MASK 0x1 +#define QM_RF_BYPASS_MASK_PFRL_SHIFT 4 +#define QM_RF_BYPASS_MASK_VPQCNRL_MASK 0x1 +#define QM_RF_BYPASS_MASK_VPQCNRL_SHIFT 5 +#define QM_RF_BYPASS_MASK_FWPAUSE_MASK 0x1 +#define QM_RF_BYPASS_MASK_FWPAUSE_SHIFT 6 +#define QM_RF_BYPASS_MASK_RESERVED1_MASK 0x1 +#define QM_RF_BYPASS_MASK_RESERVED1_SHIFT 7 +}; + +/* QM hardware structure of opportunistic credit mask */ +struct qm_rf_opportunistic_mask { + __le16 flags; +#define QM_RF_OPPORTUNISTIC_MASK_LINEVOQ_MASK 0x1 +#define QM_RF_OPPORTUNISTIC_MASK_LINEVOQ_SHIFT 0 +#define QM_RF_OPPORTUNISTIC_MASK_BYTEVOQ_MASK 0x1 +#define QM_RF_OPPORTUNISTIC_MASK_BYTEVOQ_SHIFT 1 +#define QM_RF_OPPORTUNISTIC_MASK_PFWFQ_MASK 0x1 +#define QM_RF_OPPORTUNISTIC_MASK_PFWFQ_SHIFT 2 +#define QM_RF_OPPORTUNISTIC_MASK_VPWFQ_MASK 0x1 +#define QM_RF_OPPORTUNISTIC_MASK_VPWFQ_SHIFT 3 +#define QM_RF_OPPORTUNISTIC_MASK_PFRL_MASK 0x1 +#define QM_RF_OPPORTUNISTIC_MASK_PFRL_SHIFT 4 +#define QM_RF_OPPORTUNISTIC_MASK_VPQCNRL_MASK 0x1 +#define QM_RF_OPPORTUNISTIC_MASK_VPQCNRL_SHIFT 5 +#define QM_RF_OPPORTUNISTIC_MASK_FWPAUSE_MASK 0x1 +#define QM_RF_OPPORTUNISTIC_MASK_FWPAUSE_SHIFT 6 +#define QM_RF_OPPORTUNISTIC_MASK_RESERVED0_MASK 0x1 +#define QM_RF_OPPORTUNISTIC_MASK_RESERVED0_SHIFT 7 +#define QM_RF_OPPORTUNISTIC_MASK_QUEUEEMPTY_MASK 0x1 +#define QM_RF_OPPORTUNISTIC_MASK_QUEUEEMPTY_SHIFT 8 +#define QM_RF_OPPORTUNISTIC_MASK_RESERVED1_MASK 0x7F +#define QM_RF_OPPORTUNISTIC_MASK_RESERVED1_SHIFT 9 +}; + +/* QM hardware structure of QM map memory */ +struct qm_rf_pq_map { + __le32 reg; +#define QM_RF_PQ_MAP_PQ_VALID_MASK 0x1 +#define QM_RF_PQ_MAP_PQ_VALID_SHIFT 0 +#define QM_RF_PQ_MAP_RL_ID_MASK 0xFF +#define QM_RF_PQ_MAP_RL_ID_SHIFT 1 +#define QM_RF_PQ_MAP_VP_PQ_ID_MASK 0x1FF +#define QM_RF_PQ_MAP_VP_PQ_ID_SHIFT 9 +#define QM_RF_PQ_MAP_VOQ_MASK 0x1F +#define QM_RF_PQ_MAP_VOQ_SHIFT 18 +#define QM_RF_PQ_MAP_WRR_WEIGHT_GROUP_MASK 0x3 +#define QM_RF_PQ_MAP_WRR_WEIGHT_GROUP_SHIFT 23 +#define QM_RF_PQ_MAP_RL_VALID_MASK 0x1 +#define QM_RF_PQ_MAP_RL_VALID_SHIFT 25 +#define QM_RF_PQ_MAP_RESERVED_MASK 0x3F +#define QM_RF_PQ_MAP_RESERVED_SHIFT 26 +}; + +/* Completion params for aggregated interrupt completion */ +struct sdm_agg_int_comp_params { + __le16 params; +#define SDM_AGG_INT_COMP_PARAMS_AGG_INT_INDEX_MASK 0x3F +#define SDM_AGG_INT_COMP_PARAMS_AGG_INT_INDEX_SHIFT 0 +#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_ENABLE_MASK 0x1 +#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_ENABLE_SHIFT 6 +#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_BIT_MASK 0x1FF +#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_BIT_SHIFT 7 +}; + +/* SDM operation gen command (generate aggregative interrupt) */ +struct sdm_op_gen { + __le32 command; +#define SDM_OP_GEN_COMP_PARAM_MASK 0xFFFF +#define SDM_OP_GEN_COMP_PARAM_SHIFT 0 +#define SDM_OP_GEN_COMP_TYPE_MASK 0xF +#define SDM_OP_GEN_COMP_TYPE_SHIFT 16 +#define SDM_OP_GEN_RESERVED_MASK 0xFFF +#define SDM_OP_GEN_RESERVED_SHIFT 20 +}; + +struct ystorm_core_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define YSTORM_CORE_CONN_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_CORE_CONN_AG_CTX_BIT0_SHIFT 0 +#define YSTORM_CORE_CONN_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_CORE_CONN_AG_CTX_BIT1_SHIFT 1 +#define YSTORM_CORE_CONN_AG_CTX_CF0_MASK 0x3 +#define YSTORM_CORE_CONN_AG_CTX_CF0_SHIFT 2 +#define YSTORM_CORE_CONN_AG_CTX_CF1_MASK 0x3 +#define YSTORM_CORE_CONN_AG_CTX_CF1_SHIFT 4 +#define YSTORM_CORE_CONN_AG_CTX_CF2_MASK 0x3 +#define YSTORM_CORE_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define YSTORM_CORE_CONN_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_CORE_CONN_AG_CTX_CF0EN_SHIFT 0 +#define YSTORM_CORE_CONN_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_CORE_CONN_AG_CTX_CF1EN_SHIFT 1 +#define YSTORM_CORE_CONN_AG_CTX_CF2EN_MASK 0x1 +#define YSTORM_CORE_CONN_AG_CTX_CF2EN_SHIFT 2 +#define YSTORM_CORE_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_CORE_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define YSTORM_CORE_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_CORE_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define YSTORM_CORE_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_CORE_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define YSTORM_CORE_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_CORE_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define YSTORM_CORE_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_CORE_CONN_AG_CTX_RULE4EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le32 reg0; + __le32 reg1; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg2; + __le32 reg3; +}; + +/****************************************/ +/* Debug Tools HSI constants and macros */ +/****************************************/ + enum block_addr { - GRCBASE_GRC = 0x50000, - GRCBASE_MISCS = 0x9000, - GRCBASE_MISC = 0x8000, - GRCBASE_DBU = 0xa000, - GRCBASE_PGLUE_B = 0x2a8000, - GRCBASE_CNIG = 0x218000, - GRCBASE_CPMU = 0x30000, - GRCBASE_NCSI = 0x40000, - GRCBASE_OPTE = 0x53000, - GRCBASE_BMB = 0x540000, - GRCBASE_PCIE = 0x54000, - GRCBASE_MCP = 0xe00000, - GRCBASE_MCP2 = 0x52000, - GRCBASE_PSWHST = 0x2a0000, - GRCBASE_PSWHST2 = 0x29e000, - GRCBASE_PSWRD = 0x29c000, - GRCBASE_PSWRD2 = 0x29d000, - GRCBASE_PSWWR = 0x29a000, - GRCBASE_PSWWR2 = 0x29b000, - GRCBASE_PSWRQ = 0x280000, - GRCBASE_PSWRQ2 = 0x240000, - GRCBASE_PGLCS = 0x0, - GRCBASE_PTU = 0x560000, - GRCBASE_DMAE = 0xc000, - GRCBASE_TCM = 0x1180000, - GRCBASE_MCM = 0x1200000, - GRCBASE_UCM = 0x1280000, - GRCBASE_XCM = 0x1000000, - GRCBASE_YCM = 0x1080000, - GRCBASE_PCM = 0x1100000, - GRCBASE_QM = 0x2f0000, - GRCBASE_TM = 0x2c0000, - GRCBASE_DORQ = 0x100000, - GRCBASE_BRB = 0x340000, - GRCBASE_SRC = 0x238000, - GRCBASE_PRS = 0x1f0000, - GRCBASE_TSDM = 0xfb0000, - GRCBASE_MSDM = 0xfc0000, - GRCBASE_USDM = 0xfd0000, - GRCBASE_XSDM = 0xf80000, - GRCBASE_YSDM = 0xf90000, - GRCBASE_PSDM = 0xfa0000, - GRCBASE_TSEM = 0x1700000, - GRCBASE_MSEM = 0x1800000, - GRCBASE_USEM = 0x1900000, - GRCBASE_XSEM = 0x1400000, - GRCBASE_YSEM = 0x1500000, - GRCBASE_PSEM = 0x1600000, - GRCBASE_RSS = 0x238800, - GRCBASE_TMLD = 0x4d0000, - GRCBASE_MULD = 0x4e0000, - GRCBASE_YULD = 0x4c8000, - GRCBASE_XYLD = 0x4c0000, - GRCBASE_PRM = 0x230000, - GRCBASE_PBF_PB1 = 0xda0000, - GRCBASE_PBF_PB2 = 0xda4000, - GRCBASE_RPB = 0x23c000, - GRCBASE_BTB = 0xdb0000, - GRCBASE_PBF = 0xd80000, - GRCBASE_RDIF = 0x300000, - GRCBASE_TDIF = 0x310000, - GRCBASE_CDU = 0x580000, - GRCBASE_CCFC = 0x2e0000, - GRCBASE_TCFC = 0x2d0000, - GRCBASE_IGU = 0x180000, - GRCBASE_CAU = 0x1c0000, - GRCBASE_UMAC = 0x51000, - GRCBASE_XMAC = 0x210000, - GRCBASE_DBG = 0x10000, - GRCBASE_NIG = 0x500000, - GRCBASE_WOL = 0x600000, - GRCBASE_BMBN = 0x610000, - GRCBASE_IPC = 0x20000, - GRCBASE_NWM = 0x800000, - GRCBASE_NWS = 0x700000, - GRCBASE_MS = 0x6a0000, - GRCBASE_PHY_PCIE = 0x620000, - GRCBASE_MISC_AEU = 0x8000, - GRCBASE_BAR0_MAP = 0x1c00000, + GRCBASE_GRC = 0x50000, + GRCBASE_MISCS = 0x9000, + GRCBASE_MISC = 0x8000, + GRCBASE_DBU = 0xa000, + GRCBASE_PGLUE_B = 0x2a8000, + GRCBASE_CNIG = 0x218000, + GRCBASE_CPMU = 0x30000, + GRCBASE_NCSI = 0x40000, + GRCBASE_OPTE = 0x53000, + GRCBASE_BMB = 0x540000, + GRCBASE_PCIE = 0x54000, + GRCBASE_MCP = 0xe00000, + GRCBASE_MCP2 = 0x52000, + GRCBASE_PSWHST = 0x2a0000, + GRCBASE_PSWHST2 = 0x29e000, + GRCBASE_PSWRD = 0x29c000, + GRCBASE_PSWRD2 = 0x29d000, + GRCBASE_PSWWR = 0x29a000, + GRCBASE_PSWWR2 = 0x29b000, + GRCBASE_PSWRQ = 0x280000, + GRCBASE_PSWRQ2 = 0x240000, + GRCBASE_PGLCS = 0x0, + GRCBASE_DMAE = 0xc000, + GRCBASE_PTU = 0x560000, + GRCBASE_TCM = 0x1180000, + GRCBASE_MCM = 0x1200000, + GRCBASE_UCM = 0x1280000, + GRCBASE_XCM = 0x1000000, + GRCBASE_YCM = 0x1080000, + GRCBASE_PCM = 0x1100000, + GRCBASE_QM = 0x2f0000, + GRCBASE_TM = 0x2c0000, + GRCBASE_DORQ = 0x100000, + GRCBASE_BRB = 0x340000, + GRCBASE_SRC = 0x238000, + GRCBASE_PRS = 0x1f0000, + GRCBASE_TSDM = 0xfb0000, + GRCBASE_MSDM = 0xfc0000, + GRCBASE_USDM = 0xfd0000, + GRCBASE_XSDM = 0xf80000, + GRCBASE_YSDM = 0xf90000, + GRCBASE_PSDM = 0xfa0000, + GRCBASE_TSEM = 0x1700000, + GRCBASE_MSEM = 0x1800000, + GRCBASE_USEM = 0x1900000, + GRCBASE_XSEM = 0x1400000, + GRCBASE_YSEM = 0x1500000, + GRCBASE_PSEM = 0x1600000, + GRCBASE_RSS = 0x238800, + GRCBASE_TMLD = 0x4d0000, + GRCBASE_MULD = 0x4e0000, + GRCBASE_YULD = 0x4c8000, + GRCBASE_XYLD = 0x4c0000, + GRCBASE_PRM = 0x230000, + GRCBASE_PBF_PB1 = 0xda0000, + GRCBASE_PBF_PB2 = 0xda4000, + GRCBASE_RPB = 0x23c000, + GRCBASE_BTB = 0xdb0000, + GRCBASE_PBF = 0xd80000, + GRCBASE_RDIF = 0x300000, + GRCBASE_TDIF = 0x310000, + GRCBASE_CDU = 0x580000, + GRCBASE_CCFC = 0x2e0000, + GRCBASE_TCFC = 0x2d0000, + GRCBASE_IGU = 0x180000, + GRCBASE_CAU = 0x1c0000, + GRCBASE_UMAC = 0x51000, + GRCBASE_XMAC = 0x210000, + GRCBASE_DBG = 0x10000, + GRCBASE_NIG = 0x500000, + GRCBASE_WOL = 0x600000, + GRCBASE_BMBN = 0x610000, + GRCBASE_IPC = 0x20000, + GRCBASE_NWM = 0x800000, + GRCBASE_NWS = 0x700000, + GRCBASE_MS = 0x6a0000, + GRCBASE_PHY_PCIE = 0x620000, + GRCBASE_LED = 0x6b8000, + GRCBASE_MISC_AEU = 0x8000, + GRCBASE_BAR0_MAP = 0x1c00000, MAX_BLOCK_ADDR }; @@ -879,8 +1359,8 @@ enum block_id { BLOCK_PSWRQ, BLOCK_PSWRQ2, BLOCK_PGLCS, - BLOCK_PTU, BLOCK_DMAE, + BLOCK_PTU, BLOCK_TCM, BLOCK_MCM, BLOCK_UCM, @@ -934,141 +1414,216 @@ enum block_id { BLOCK_NWS, BLOCK_MS, BLOCK_PHY_PCIE, + BLOCK_LED, BLOCK_MISC_AEU, BLOCK_BAR0_MAP, MAX_BLOCK_ID }; -enum command_type_bit { - IGU_COMMAND_TYPE_NOP = 0, - IGU_COMMAND_TYPE_SET = 1, - MAX_COMMAND_TYPE_BIT +/* binary debug buffer types */ +enum bin_dbg_buffer_type { + BIN_BUF_DBG_MODE_TREE, + BIN_BUF_DBG_DUMP_REG, + BIN_BUF_DBG_DUMP_MEM, + BIN_BUF_DBG_IDLE_CHK_REGS, + BIN_BUF_DBG_IDLE_CHK_IMMS, + BIN_BUF_DBG_IDLE_CHK_RULES, + BIN_BUF_DBG_IDLE_CHK_PARSING_DATA, + BIN_BUF_DBG_ATTN_BLOCKS, + BIN_BUF_DBG_ATTN_REGS, + BIN_BUF_DBG_ATTN_INDEXES, + BIN_BUF_DBG_ATTN_NAME_OFFSETS, + BIN_BUF_DBG_PARSING_STRINGS, + MAX_BIN_DBG_BUFFER_TYPE }; -struct dmae_cmd { - __le32 opcode; -#define DMAE_CMD_SRC_MASK 0x1 -#define DMAE_CMD_SRC_SHIFT 0 -#define DMAE_CMD_DST_MASK 0x3 -#define DMAE_CMD_DST_SHIFT 1 -#define DMAE_CMD_C_DST_MASK 0x1 -#define DMAE_CMD_C_DST_SHIFT 3 -#define DMAE_CMD_CRC_RESET_MASK 0x1 -#define DMAE_CMD_CRC_RESET_SHIFT 4 -#define DMAE_CMD_SRC_ADDR_RESET_MASK 0x1 -#define DMAE_CMD_SRC_ADDR_RESET_SHIFT 5 -#define DMAE_CMD_DST_ADDR_RESET_MASK 0x1 -#define DMAE_CMD_DST_ADDR_RESET_SHIFT 6 -#define DMAE_CMD_COMP_FUNC_MASK 0x1 -#define DMAE_CMD_COMP_FUNC_SHIFT 7 -#define DMAE_CMD_COMP_WORD_EN_MASK 0x1 -#define DMAE_CMD_COMP_WORD_EN_SHIFT 8 -#define DMAE_CMD_COMP_CRC_EN_MASK 0x1 -#define DMAE_CMD_COMP_CRC_EN_SHIFT 9 -#define DMAE_CMD_COMP_CRC_OFFSET_MASK 0x7 -#define DMAE_CMD_COMP_CRC_OFFSET_SHIFT 10 -#define DMAE_CMD_RESERVED1_MASK 0x1 -#define DMAE_CMD_RESERVED1_SHIFT 13 -#define DMAE_CMD_ENDIANITY_MODE_MASK 0x3 -#define DMAE_CMD_ENDIANITY_MODE_SHIFT 14 -#define DMAE_CMD_ERR_HANDLING_MASK 0x3 -#define DMAE_CMD_ERR_HANDLING_SHIFT 16 -#define DMAE_CMD_PORT_ID_MASK 0x3 -#define DMAE_CMD_PORT_ID_SHIFT 18 -#define DMAE_CMD_SRC_PF_ID_MASK 0xF -#define DMAE_CMD_SRC_PF_ID_SHIFT 20 -#define DMAE_CMD_DST_PF_ID_MASK 0xF -#define DMAE_CMD_DST_PF_ID_SHIFT 24 -#define DMAE_CMD_SRC_VF_ID_VALID_MASK 0x1 -#define DMAE_CMD_SRC_VF_ID_VALID_SHIFT 28 -#define DMAE_CMD_DST_VF_ID_VALID_MASK 0x1 -#define DMAE_CMD_DST_VF_ID_VALID_SHIFT 29 -#define DMAE_CMD_RESERVED2_MASK 0x3 -#define DMAE_CMD_RESERVED2_SHIFT 30 - __le32 src_addr_lo; - __le32 src_addr_hi; - __le32 dst_addr_lo; - __le32 dst_addr_hi; - __le16 length /* Length in DW */; - __le16 opcode_b; -#define DMAE_CMD_SRC_VF_ID_MASK 0xFF /* Source VF id */ -#define DMAE_CMD_SRC_VF_ID_SHIFT 0 -#define DMAE_CMD_DST_VF_ID_MASK 0xFF /* Destination VF id */ -#define DMAE_CMD_DST_VF_ID_SHIFT 8 - __le32 comp_addr_lo /* PCIe completion address low or grc address */; - __le32 comp_addr_hi; - __le32 comp_val /* Value to write to copmletion address */; - __le32 crc32 /* crc16 result */; - __le32 crc_32_c /* crc32_c result */; - __le16 crc16 /* crc16 result */; - __le16 crc16_c /* crc16_c result */; - __le16 crc10 /* crc_t10 result */; - __le16 reserved; - __le16 xsum16 /* checksum16 result */; - __le16 xsum8 /* checksum8 result */; +/* Chip IDs */ +enum chip_ids { + CHIP_RESERVED, + CHIP_BB_B0, + CHIP_RESERVED2, + MAX_CHIP_IDS }; -struct igu_cleanup { - __le32 sb_id_and_flags; -#define IGU_CLEANUP_RESERVED0_MASK 0x7FFFFFF -#define IGU_CLEANUP_RESERVED0_SHIFT 0 -#define IGU_CLEANUP_CLEANUP_SET_MASK 0x1 /* cleanup clear - 0, set - 1 */ -#define IGU_CLEANUP_CLEANUP_SET_SHIFT 27 -#define IGU_CLEANUP_CLEANUP_TYPE_MASK 0x7 -#define IGU_CLEANUP_CLEANUP_TYPE_SHIFT 28 -#define IGU_CLEANUP_COMMAND_TYPE_MASK 0x1 -#define IGU_CLEANUP_COMMAND_TYPE_SHIFT 31 - __le32 reserved1; +/* Attention bit mapping */ +struct dbg_attn_bit_mapping { + __le16 data; +#define DBG_ATTN_BIT_MAPPING_VAL_MASK 0x7FFF +#define DBG_ATTN_BIT_MAPPING_VAL_SHIFT 0 +#define DBG_ATTN_BIT_MAPPING_IS_UNUSED_BIT_CNT_MASK 0x1 +#define DBG_ATTN_BIT_MAPPING_IS_UNUSED_BIT_CNT_SHIFT 15 }; -union igu_command { - struct igu_prod_cons_update prod_cons_update; - struct igu_cleanup cleanup; +/* Attention block per-type data */ +struct dbg_attn_block_type_data { + __le16 names_offset; + __le16 reserved1; + u8 num_regs; + u8 reserved2; + __le16 regs_offset; }; -struct igu_command_reg_ctrl { - __le16 opaque_fid; - __le16 igu_command_reg_ctrl_fields; -#define IGU_COMMAND_REG_CTRL_PXP_BAR_ADDR_MASK 0xFFF -#define IGU_COMMAND_REG_CTRL_PXP_BAR_ADDR_SHIFT 0 -#define IGU_COMMAND_REG_CTRL_RESERVED_MASK 0x7 -#define IGU_COMMAND_REG_CTRL_RESERVED_SHIFT 12 -#define IGU_COMMAND_REG_CTRL_COMMAND_TYPE_MASK 0x1 -#define IGU_COMMAND_REG_CTRL_COMMAND_TYPE_SHIFT 15 +/* Block attentions */ +struct dbg_attn_block { + struct dbg_attn_block_type_data per_type_data[2]; }; -struct igu_mapping_line { - __le32 igu_mapping_line_fields; -#define IGU_MAPPING_LINE_VALID_MASK 0x1 -#define IGU_MAPPING_LINE_VALID_SHIFT 0 -#define IGU_MAPPING_LINE_VECTOR_NUMBER_MASK 0xFF -#define IGU_MAPPING_LINE_VECTOR_NUMBER_SHIFT 1 -#define IGU_MAPPING_LINE_FUNCTION_NUMBER_MASK 0xFF -#define IGU_MAPPING_LINE_FUNCTION_NUMBER_SHIFT 9 -#define IGU_MAPPING_LINE_PF_VALID_MASK 0x1 /* PF-1, VF-0 */ -#define IGU_MAPPING_LINE_PF_VALID_SHIFT 17 -#define IGU_MAPPING_LINE_IPS_GROUP_MASK 0x3F -#define IGU_MAPPING_LINE_IPS_GROUP_SHIFT 18 -#define IGU_MAPPING_LINE_RESERVED_MASK 0xFF -#define IGU_MAPPING_LINE_RESERVED_SHIFT 24 +/* Attention register result */ +struct dbg_attn_reg_result { + __le32 data; +#define DBG_ATTN_REG_RESULT_STS_ADDRESS_MASK 0xFFFFFF +#define DBG_ATTN_REG_RESULT_STS_ADDRESS_SHIFT 0 +#define DBG_ATTN_REG_RESULT_NUM_ATTN_IDX_MASK 0xFF +#define DBG_ATTN_REG_RESULT_NUM_ATTN_IDX_SHIFT 24 + __le16 attn_idx_offset; + __le16 reserved; + __le32 sts_val; + __le32 mask_val; +}; + +/* Attention block result */ +struct dbg_attn_block_result { + u8 block_id; + u8 data; +#define DBG_ATTN_BLOCK_RESULT_ATTN_TYPE_MASK 0x3 +#define DBG_ATTN_BLOCK_RESULT_ATTN_TYPE_SHIFT 0 +#define DBG_ATTN_BLOCK_RESULT_NUM_REGS_MASK 0x3F +#define DBG_ATTN_BLOCK_RESULT_NUM_REGS_SHIFT 2 + __le16 names_offset; + struct dbg_attn_reg_result reg_results[15]; +}; + +/* mode header */ +struct dbg_mode_hdr { + __le16 data; +#define DBG_MODE_HDR_EVAL_MODE_MASK 0x1 +#define DBG_MODE_HDR_EVAL_MODE_SHIFT 0 +#define DBG_MODE_HDR_MODES_BUF_OFFSET_MASK 0x7FFF +#define DBG_MODE_HDR_MODES_BUF_OFFSET_SHIFT 1 +}; + +/* Attention register */ +struct dbg_attn_reg { + struct dbg_mode_hdr mode; + __le16 attn_idx_offset; + __le32 data; +#define DBG_ATTN_REG_STS_ADDRESS_MASK 0xFFFFFF +#define DBG_ATTN_REG_STS_ADDRESS_SHIFT 0 +#define DBG_ATTN_REG_NUM_ATTN_IDX_MASK 0xFF +#define DBG_ATTN_REG_NUM_ATTN_IDX_SHIFT 24 + __le32 sts_clr_address; + __le32 mask_address; +}; + +/* attention types */ +enum dbg_attn_type { + ATTN_TYPE_INTERRUPT, + ATTN_TYPE_PARITY, + MAX_DBG_ATTN_TYPE +}; + +/* Debug status codes */ +enum dbg_status { + DBG_STATUS_OK, + DBG_STATUS_APP_VERSION_NOT_SET, + DBG_STATUS_UNSUPPORTED_APP_VERSION, + DBG_STATUS_DBG_BLOCK_NOT_RESET, + DBG_STATUS_INVALID_ARGS, + DBG_STATUS_OUTPUT_ALREADY_SET, + DBG_STATUS_INVALID_PCI_BUF_SIZE, + DBG_STATUS_PCI_BUF_ALLOC_FAILED, + DBG_STATUS_PCI_BUF_NOT_ALLOCATED, + DBG_STATUS_TOO_MANY_INPUTS, + DBG_STATUS_INPUT_OVERLAP, + DBG_STATUS_HW_ONLY_RECORDING, + DBG_STATUS_STORM_ALREADY_ENABLED, + DBG_STATUS_STORM_NOT_ENABLED, + DBG_STATUS_BLOCK_ALREADY_ENABLED, + DBG_STATUS_BLOCK_NOT_ENABLED, + DBG_STATUS_NO_INPUT_ENABLED, + DBG_STATUS_NO_FILTER_TRIGGER_64B, + DBG_STATUS_FILTER_ALREADY_ENABLED, + DBG_STATUS_TRIGGER_ALREADY_ENABLED, + DBG_STATUS_TRIGGER_NOT_ENABLED, + DBG_STATUS_CANT_ADD_CONSTRAINT, + DBG_STATUS_TOO_MANY_TRIGGER_STATES, + DBG_STATUS_TOO_MANY_CONSTRAINTS, + DBG_STATUS_RECORDING_NOT_STARTED, + DBG_STATUS_DATA_DIDNT_TRIGGER, + DBG_STATUS_NO_DATA_RECORDED, + DBG_STATUS_DUMP_BUF_TOO_SMALL, + DBG_STATUS_DUMP_NOT_CHUNK_ALIGNED, + DBG_STATUS_UNKNOWN_CHIP, + DBG_STATUS_VIRT_MEM_ALLOC_FAILED, + DBG_STATUS_BLOCK_IN_RESET, + DBG_STATUS_INVALID_TRACE_SIGNATURE, + DBG_STATUS_INVALID_NVRAM_BUNDLE, + DBG_STATUS_NVRAM_GET_IMAGE_FAILED, + DBG_STATUS_NON_ALIGNED_NVRAM_IMAGE, + DBG_STATUS_NVRAM_READ_FAILED, + DBG_STATUS_IDLE_CHK_PARSE_FAILED, + DBG_STATUS_MCP_TRACE_BAD_DATA, + DBG_STATUS_MCP_TRACE_NO_META, + DBG_STATUS_MCP_COULD_NOT_HALT, + DBG_STATUS_MCP_COULD_NOT_RESUME, + DBG_STATUS_DMAE_FAILED, + DBG_STATUS_SEMI_FIFO_NOT_EMPTY, + DBG_STATUS_IGU_FIFO_BAD_DATA, + DBG_STATUS_MCP_COULD_NOT_MASK_PRTY, + DBG_STATUS_FW_ASSERTS_PARSE_FAILED, + DBG_STATUS_REG_FIFO_BAD_DATA, + DBG_STATUS_PROTECTION_OVERRIDE_BAD_DATA, + DBG_STATUS_DBG_ARRAY_NOT_SET, + MAX_DBG_STATUS }; -struct igu_msix_vector { - struct regpair address; - __le32 data; - __le32 msix_vector_fields; -#define IGU_MSIX_VECTOR_MASK_BIT_MASK 0x1 -#define IGU_MSIX_VECTOR_MASK_BIT_SHIFT 0 -#define IGU_MSIX_VECTOR_RESERVED0_MASK 0x7FFF -#define IGU_MSIX_VECTOR_RESERVED0_SHIFT 1 -#define IGU_MSIX_VECTOR_STEERING_TAG_MASK 0xFF -#define IGU_MSIX_VECTOR_STEERING_TAG_SHIFT 16 -#define IGU_MSIX_VECTOR_RESERVED1_MASK 0xFF -#define IGU_MSIX_VECTOR_RESERVED1_SHIFT 24 +/********************************/ +/* HSI Init Functions constants */ +/********************************/ + +/* Number of VLAN priorities */ +#define NUM_OF_VLAN_PRIORITIES 8 + +/* QM per-port init parameters */ +struct init_qm_port_params { + u8 active; + u8 active_phys_tcs; + __le16 num_pbf_cmd_lines; + __le16 num_btb_blocks; + __le16 reserved; }; +/* QM per-PQ init parameters */ +struct init_qm_pq_params { + u8 vport_id; + u8 tc_id; + u8 wrr_group; + u8 rl_valid; +}; + +/* QM per-vport init parameters */ +struct init_qm_vport_params { + __le32 vport_rl; + __le16 vport_wfq; + __le16 first_tx_pq_id[NUM_OF_TCS]; +}; + +/**************************************/ +/* Init Tool HSI constants and macros */ +/**************************************/ + +/* Width of GRC address in bits (addresses are specified in dwords) */ +#define GRC_ADDR_BITS 23 +#define MAX_GRC_ADDR ((1 << GRC_ADDR_BITS) - 1) + +/* indicates an init that should be applied to any phase ID */ +#define ANY_PHASE_ID 0xffff + +/* Max size in dwords of a zipped array */ +#define MAX_ZIPPED_SIZE 8192 + enum init_modes { - MODE_BB_A0, + MODE_RESERVED, MODE_BB_B0, MODE_RESERVED2, MODE_ASIC, @@ -1083,7 +1638,8 @@ enum init_modes { MODE_PORTS_PER_ENG_2, MODE_PORTS_PER_ENG_4, MODE_100G, - MODE_EAGLE_ENG1_WORKAROUND, + MODE_40G, + MODE_RESERVED7, MAX_INIT_MODES }; @@ -1096,484 +1652,302 @@ enum init_phases { MAX_INIT_PHASES }; -/* per encapsulation type enabling flags */ -struct prs_reg_encapsulation_type_en { - u8 flags; -#define PRS_REG_ENCAPSULATION_TYPE_EN_ETH_OVER_GRE_ENABLE_MASK 0x1 -#define PRS_REG_ENCAPSULATION_TYPE_EN_ETH_OVER_GRE_ENABLE_SHIFT 0 -#define PRS_REG_ENCAPSULATION_TYPE_EN_IP_OVER_GRE_ENABLE_MASK 0x1 -#define PRS_REG_ENCAPSULATION_TYPE_EN_IP_OVER_GRE_ENABLE_SHIFT 1 -#define PRS_REG_ENCAPSULATION_TYPE_EN_VXLAN_ENABLE_MASK 0x1 -#define PRS_REG_ENCAPSULATION_TYPE_EN_VXLAN_ENABLE_SHIFT 2 -#define PRS_REG_ENCAPSULATION_TYPE_EN_T_TAG_ENABLE_MASK 0x1 -#define PRS_REG_ENCAPSULATION_TYPE_EN_T_TAG_ENABLE_SHIFT 3 -#define PRS_REG_ENCAPSULATION_TYPE_EN_ETH_OVER_GENEVE_ENABLE_MASK 0x1 -#define PRS_REG_ENCAPSULATION_TYPE_EN_ETH_OVER_GENEVE_ENABLE_SHIFT 4 -#define PRS_REG_ENCAPSULATION_TYPE_EN_IP_OVER_GENEVE_ENABLE_MASK 0x1 -#define PRS_REG_ENCAPSULATION_TYPE_EN_IP_OVER_GENEVE_ENABLE_SHIFT 5 -#define PRS_REG_ENCAPSULATION_TYPE_EN_RESERVED_MASK 0x3 -#define PRS_REG_ENCAPSULATION_TYPE_EN_RESERVED_SHIFT 6 -}; - -enum pxp_tph_st_hint { - TPH_ST_HINT_BIDIR /* Read/Write access by Host and Device */, - TPH_ST_HINT_REQUESTER /* Read/Write access by Device */, - TPH_ST_HINT_TARGET, - TPH_ST_HINT_TARGET_PRIO, - MAX_PXP_TPH_ST_HINT +enum init_split_types { + SPLIT_TYPE_NONE, + SPLIT_TYPE_PORT, + SPLIT_TYPE_PF, + SPLIT_TYPE_PORT_PF, + SPLIT_TYPE_VF, + MAX_INIT_SPLIT_TYPES }; -/* QM hardware structure of enable bypass credit mask */ -struct qm_rf_bypass_mask { - u8 flags; -#define QM_RF_BYPASS_MASK_LINEVOQ_MASK 0x1 -#define QM_RF_BYPASS_MASK_LINEVOQ_SHIFT 0 -#define QM_RF_BYPASS_MASK_RESERVED0_MASK 0x1 -#define QM_RF_BYPASS_MASK_RESERVED0_SHIFT 1 -#define QM_RF_BYPASS_MASK_PFWFQ_MASK 0x1 -#define QM_RF_BYPASS_MASK_PFWFQ_SHIFT 2 -#define QM_RF_BYPASS_MASK_VPWFQ_MASK 0x1 -#define QM_RF_BYPASS_MASK_VPWFQ_SHIFT 3 -#define QM_RF_BYPASS_MASK_PFRL_MASK 0x1 -#define QM_RF_BYPASS_MASK_PFRL_SHIFT 4 -#define QM_RF_BYPASS_MASK_VPQCNRL_MASK 0x1 -#define QM_RF_BYPASS_MASK_VPQCNRL_SHIFT 5 -#define QM_RF_BYPASS_MASK_FWPAUSE_MASK 0x1 -#define QM_RF_BYPASS_MASK_FWPAUSE_SHIFT 6 -#define QM_RF_BYPASS_MASK_RESERVED1_MASK 0x1 -#define QM_RF_BYPASS_MASK_RESERVED1_SHIFT 7 -}; - -/* QM hardware structure of opportunistic credit mask */ -struct qm_rf_opportunistic_mask { - __le16 flags; -#define QM_RF_OPPORTUNISTIC_MASK_LINEVOQ_MASK 0x1 -#define QM_RF_OPPORTUNISTIC_MASK_LINEVOQ_SHIFT 0 -#define QM_RF_OPPORTUNISTIC_MASK_BYTEVOQ_MASK 0x1 -#define QM_RF_OPPORTUNISTIC_MASK_BYTEVOQ_SHIFT 1 -#define QM_RF_OPPORTUNISTIC_MASK_PFWFQ_MASK 0x1 -#define QM_RF_OPPORTUNISTIC_MASK_PFWFQ_SHIFT 2 -#define QM_RF_OPPORTUNISTIC_MASK_VPWFQ_MASK 0x1 -#define QM_RF_OPPORTUNISTIC_MASK_VPWFQ_SHIFT 3 -#define QM_RF_OPPORTUNISTIC_MASK_PFRL_MASK 0x1 -#define QM_RF_OPPORTUNISTIC_MASK_PFRL_SHIFT 4 -#define QM_RF_OPPORTUNISTIC_MASK_VPQCNRL_MASK 0x1 -#define QM_RF_OPPORTUNISTIC_MASK_VPQCNRL_SHIFT 5 -#define QM_RF_OPPORTUNISTIC_MASK_FWPAUSE_MASK 0x1 -#define QM_RF_OPPORTUNISTIC_MASK_FWPAUSE_SHIFT 6 -#define QM_RF_OPPORTUNISTIC_MASK_RESERVED0_MASK 0x1 -#define QM_RF_OPPORTUNISTIC_MASK_RESERVED0_SHIFT 7 -#define QM_RF_OPPORTUNISTIC_MASK_QUEUEEMPTY_MASK 0x1 -#define QM_RF_OPPORTUNISTIC_MASK_QUEUEEMPTY_SHIFT 8 -#define QM_RF_OPPORTUNISTIC_MASK_RESERVED1_MASK 0x7F -#define QM_RF_OPPORTUNISTIC_MASK_RESERVED1_SHIFT 9 -}; - -/* QM hardware structure of QM map memory */ -struct qm_rf_pq_map { - u32 reg; -#define QM_RF_PQ_MAP_PQ_VALID_MASK 0x1 /* PQ active */ -#define QM_RF_PQ_MAP_PQ_VALID_SHIFT 0 -#define QM_RF_PQ_MAP_RL_ID_MASK 0xFF /* RL ID */ -#define QM_RF_PQ_MAP_RL_ID_SHIFT 1 -#define QM_RF_PQ_MAP_VP_PQ_ID_MASK 0x1FF -#define QM_RF_PQ_MAP_VP_PQ_ID_SHIFT 9 -#define QM_RF_PQ_MAP_VOQ_MASK 0x1F /* VOQ */ -#define QM_RF_PQ_MAP_VOQ_SHIFT 18 -#define QM_RF_PQ_MAP_WRR_WEIGHT_GROUP_MASK 0x3 /* WRR weight */ -#define QM_RF_PQ_MAP_WRR_WEIGHT_GROUP_SHIFT 23 -#define QM_RF_PQ_MAP_RL_VALID_MASK 0x1 /* RL active */ -#define QM_RF_PQ_MAP_RL_VALID_SHIFT 25 -#define QM_RF_PQ_MAP_RESERVED_MASK 0x3F -#define QM_RF_PQ_MAP_RESERVED_SHIFT 26 -}; - -/* Completion params for aggregated interrupt completion */ -struct sdm_agg_int_comp_params { - __le16 params; -#define SDM_AGG_INT_COMP_PARAMS_AGG_INT_INDEX_MASK 0x3F -#define SDM_AGG_INT_COMP_PARAMS_AGG_INT_INDEX_SHIFT 0 -#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_ENABLE_MASK 0x1 -#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_ENABLE_SHIFT 6 -#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_BIT_MASK 0x1FF -#define SDM_AGG_INT_COMP_PARAMS_AGG_VECTOR_BIT_SHIFT 7 -}; - -/* SDM operation gen command (generate aggregative interrupt) */ -struct sdm_op_gen { - __le32 command; -#define SDM_OP_GEN_COMP_PARAM_MASK 0xFFFF /* completion parameters 0-15 */ -#define SDM_OP_GEN_COMP_PARAM_SHIFT 0 -#define SDM_OP_GEN_COMP_TYPE_MASK 0xF /* completion type 16-19 */ -#define SDM_OP_GEN_COMP_TYPE_SHIFT 16 -#define SDM_OP_GEN_RESERVED_MASK 0xFFF /* reserved 20-31 */ -#define SDM_OP_GEN_RESERVED_SHIFT 20 -}; - -/*********************************** Init ************************************/ - -/* Width of GRC address in bits (addresses are specified in dwords) */ -#define GRC_ADDR_BITS 23 -#define MAX_GRC_ADDR ((1 << GRC_ADDR_BITS) - 1) - -/* indicates an init that should be applied to any phase ID */ -#define ANY_PHASE_ID 0xffff - -/* init pattern size in bytes */ -#define INIT_PATTERN_SIZE_BITS 4 -#define MAX_INIT_PATTERN_SIZE BIT(INIT_PATTERN_SIZE_BITS) - -/* Max size in dwords of a zipped array */ -#define MAX_ZIPPED_SIZE 8192 - -/* Global PXP window */ -#define NUM_OF_PXP_WIN 19 -#define PXP_WIN_DWORD_SIZE_BITS 10 -#define PXP_WIN_DWORD_SIZE BIT(PXP_WIN_DWORD_SIZE_BITS) -#define PXP_WIN_BYTE_SIZE_BITS (PXP_WIN_DWORD_SIZE_BITS + 2) -#define PXP_WIN_BYTE_SIZE (PXP_WIN_DWORD_SIZE * 4) - -/********************************* GRC Dump **********************************/ - -/* width of GRC dump register sequence length in bits */ -#define DUMP_SEQ_LEN_BITS 8 -#define DUMP_SEQ_LEN_MAX_VAL ((1 << DUMP_SEQ_LEN_BITS) - 1) - -/* width of GRC dump memory length in bits */ -#define DUMP_MEM_LEN_BITS 18 -#define DUMP_MEM_LEN_MAX_VAL ((1 << DUMP_MEM_LEN_BITS) - 1) - -/* width of register type ID in bits */ -#define REG_TYPE_ID_BITS 6 -#define REG_TYPE_ID_MAX_VAL ((1 << REG_TYPE_ID_BITS) - 1) - -/* width of block ID in bits */ -#define BLOCK_ID_BITS 8 -#define BLOCK_ID_MAX_VAL ((1 << BLOCK_ID_BITS) - 1) - -/******************************** Idle Check *********************************/ - -/* max number of idle check predicate immediates */ -#define MAX_IDLE_CHK_PRED_IMM 3 - -/* max number of idle check argument registers */ -#define MAX_IDLE_CHK_READ_REGS 3 - -/* max number of idle check loops */ -#define MAX_IDLE_CHK_LOOPS 0x10000 - -/* max idle check address increment */ -#define MAX_IDLE_CHK_INCREMENT 0x10000 - -/* inicates an undefined idle check line index */ -#define IDLE_CHK_UNDEFINED_LINE_IDX 0xffffff - -/* max number of register values following the idle check header */ -#define IDLE_CHK_MAX_DUMP_REGS 2 - -/* arguments for IDLE_CHK_MACRO_TYPE_QM_RD_WR */ -#define IDLE_CHK_QM_RD_WR_PTR 0 -#define IDLE_CHK_QM_RD_WR_BANK 1 - -/**************************************/ -/* HSI Functions constants and macros */ -/**************************************/ - -/* Number of VLAN priorities */ -#define NUM_OF_VLAN_PRIORITIES 8 - -/* the MCP Trace meta data signautre is duplicated in the perl script that - * generats the NVRAM images. - */ -#define MCP_TRACE_META_IMAGE_SIGNATURE 0x669955aa - /* Binary buffer header */ struct bin_buffer_hdr { - u32 offset; - u32 length /* buffer length in bytes */; + __le32 offset; + __le32 length; }; -/* binary buffer types */ -enum bin_buffer_type { - BIN_BUF_FW_VER_INFO /* fw_ver_info struct */, - BIN_BUF_INIT_CMD /* init commands */, - BIN_BUF_INIT_VAL /* init data */, - BIN_BUF_INIT_MODE_TREE /* init modes tree */, - BIN_BUF_IRO /* internal RAM offsets array */, - MAX_BIN_BUFFER_TYPE -}; - -/* Chip IDs */ -enum chip_ids { - CHIP_BB_A0 /* BB A0 chip ID */, - CHIP_BB_B0 /* BB B0 chip ID */, - CHIP_K2 /* AH chip ID */, - MAX_CHIP_IDS +/* binary init buffer types */ +enum bin_init_buffer_type { + BIN_BUF_FW_VER_INFO, + BIN_BUF_INIT_CMD, + BIN_BUF_INIT_VAL, + BIN_BUF_INIT_MODE_TREE, + BIN_BUF_IRO, + MAX_BIN_INIT_BUFFER_TYPE }; +/* init array header: raw */ struct init_array_raw_hdr { __le32 data; -#define INIT_ARRAY_RAW_HDR_TYPE_MASK 0xF -#define INIT_ARRAY_RAW_HDR_TYPE_SHIFT 0 -#define INIT_ARRAY_RAW_HDR_PARAMS_MASK 0xFFFFFFF /* init array params */ -#define INIT_ARRAY_RAW_HDR_PARAMS_SHIFT 4 +#define INIT_ARRAY_RAW_HDR_TYPE_MASK 0xF +#define INIT_ARRAY_RAW_HDR_TYPE_SHIFT 0 +#define INIT_ARRAY_RAW_HDR_PARAMS_MASK 0xFFFFFFF +#define INIT_ARRAY_RAW_HDR_PARAMS_SHIFT 4 }; +/* init array header: standard */ struct init_array_standard_hdr { __le32 data; -#define INIT_ARRAY_STANDARD_HDR_TYPE_MASK 0xF -#define INIT_ARRAY_STANDARD_HDR_TYPE_SHIFT 0 -#define INIT_ARRAY_STANDARD_HDR_SIZE_MASK 0xFFFFFFF -#define INIT_ARRAY_STANDARD_HDR_SIZE_SHIFT 4 +#define INIT_ARRAY_STANDARD_HDR_TYPE_MASK 0xF +#define INIT_ARRAY_STANDARD_HDR_TYPE_SHIFT 0 +#define INIT_ARRAY_STANDARD_HDR_SIZE_MASK 0xFFFFFFF +#define INIT_ARRAY_STANDARD_HDR_SIZE_SHIFT 4 }; +/* init array header: zipped */ struct init_array_zipped_hdr { __le32 data; -#define INIT_ARRAY_ZIPPED_HDR_TYPE_MASK 0xF -#define INIT_ARRAY_ZIPPED_HDR_TYPE_SHIFT 0 -#define INIT_ARRAY_ZIPPED_HDR_ZIPPED_SIZE_MASK 0xFFFFFFF -#define INIT_ARRAY_ZIPPED_HDR_ZIPPED_SIZE_SHIFT 4 +#define INIT_ARRAY_ZIPPED_HDR_TYPE_MASK 0xF +#define INIT_ARRAY_ZIPPED_HDR_TYPE_SHIFT 0 +#define INIT_ARRAY_ZIPPED_HDR_ZIPPED_SIZE_MASK 0xFFFFFFF +#define INIT_ARRAY_ZIPPED_HDR_ZIPPED_SIZE_SHIFT 4 }; +/* init array header: pattern */ struct init_array_pattern_hdr { __le32 data; -#define INIT_ARRAY_PATTERN_HDR_TYPE_MASK 0xF -#define INIT_ARRAY_PATTERN_HDR_TYPE_SHIFT 0 -#define INIT_ARRAY_PATTERN_HDR_PATTERN_SIZE_MASK 0xF -#define INIT_ARRAY_PATTERN_HDR_PATTERN_SIZE_SHIFT 4 -#define INIT_ARRAY_PATTERN_HDR_REPETITIONS_MASK 0xFFFFFF -#define INIT_ARRAY_PATTERN_HDR_REPETITIONS_SHIFT 8 +#define INIT_ARRAY_PATTERN_HDR_TYPE_MASK 0xF +#define INIT_ARRAY_PATTERN_HDR_TYPE_SHIFT 0 +#define INIT_ARRAY_PATTERN_HDR_PATTERN_SIZE_MASK 0xF +#define INIT_ARRAY_PATTERN_HDR_PATTERN_SIZE_SHIFT 4 +#define INIT_ARRAY_PATTERN_HDR_REPETITIONS_MASK 0xFFFFFF +#define INIT_ARRAY_PATTERN_HDR_REPETITIONS_SHIFT 8 }; +/* init array header union */ union init_array_hdr { - struct init_array_raw_hdr raw /* raw init array header */; - struct init_array_standard_hdr standard; - struct init_array_zipped_hdr zipped /* zipped init array header */; - struct init_array_pattern_hdr pattern /* pattern init array header */; + struct init_array_raw_hdr raw; + struct init_array_standard_hdr standard; + struct init_array_zipped_hdr zipped; + struct init_array_pattern_hdr pattern; }; +/* init array types */ enum init_array_types { - INIT_ARR_STANDARD /* standard init array */, - INIT_ARR_ZIPPED /* zipped init array */, - INIT_ARR_PATTERN /* a repeated pattern */, + INIT_ARR_STANDARD, + INIT_ARR_ZIPPED, + INIT_ARR_PATTERN, MAX_INIT_ARRAY_TYPES }; /* init operation: callback */ struct init_callback_op { - __le32 op_data; -#define INIT_CALLBACK_OP_OP_MASK 0xF -#define INIT_CALLBACK_OP_OP_SHIFT 0 -#define INIT_CALLBACK_OP_RESERVED_MASK 0xFFFFFFF -#define INIT_CALLBACK_OP_RESERVED_SHIFT 4 - __le16 callback_id /* Callback ID */; - __le16 block_id /* Blocks ID */; + __le32 op_data; +#define INIT_CALLBACK_OP_OP_MASK 0xF +#define INIT_CALLBACK_OP_OP_SHIFT 0 +#define INIT_CALLBACK_OP_RESERVED_MASK 0xFFFFFFF +#define INIT_CALLBACK_OP_RESERVED_SHIFT 4 + __le16 callback_id; + __le16 block_id; }; /* init operation: delay */ struct init_delay_op { - __le32 op_data; -#define INIT_DELAY_OP_OP_MASK 0xF -#define INIT_DELAY_OP_OP_SHIFT 0 -#define INIT_DELAY_OP_RESERVED_MASK 0xFFFFFFF -#define INIT_DELAY_OP_RESERVED_SHIFT 4 - __le32 delay /* delay in us */; + __le32 op_data; +#define INIT_DELAY_OP_OP_MASK 0xF +#define INIT_DELAY_OP_OP_SHIFT 0 +#define INIT_DELAY_OP_RESERVED_MASK 0xFFFFFFF +#define INIT_DELAY_OP_RESERVED_SHIFT 4 + __le32 delay; }; /* init operation: if_mode */ struct init_if_mode_op { __le32 op_data; -#define INIT_IF_MODE_OP_OP_MASK 0xF -#define INIT_IF_MODE_OP_OP_SHIFT 0 -#define INIT_IF_MODE_OP_RESERVED1_MASK 0xFFF -#define INIT_IF_MODE_OP_RESERVED1_SHIFT 4 -#define INIT_IF_MODE_OP_CMD_OFFSET_MASK 0xFFFF -#define INIT_IF_MODE_OP_CMD_OFFSET_SHIFT 16 - __le16 reserved2; - __le16 modes_buf_offset; +#define INIT_IF_MODE_OP_OP_MASK 0xF +#define INIT_IF_MODE_OP_OP_SHIFT 0 +#define INIT_IF_MODE_OP_RESERVED1_MASK 0xFFF +#define INIT_IF_MODE_OP_RESERVED1_SHIFT 4 +#define INIT_IF_MODE_OP_CMD_OFFSET_MASK 0xFFFF +#define INIT_IF_MODE_OP_CMD_OFFSET_SHIFT 16 + __le16 reserved2; + __le16 modes_buf_offset; }; -/* init operation: if_phase */ +/* init operation: if_phase */ struct init_if_phase_op { __le32 op_data; -#define INIT_IF_PHASE_OP_OP_MASK 0xF -#define INIT_IF_PHASE_OP_OP_SHIFT 0 -#define INIT_IF_PHASE_OP_DMAE_ENABLE_MASK 0x1 -#define INIT_IF_PHASE_OP_DMAE_ENABLE_SHIFT 4 -#define INIT_IF_PHASE_OP_RESERVED1_MASK 0x7FF -#define INIT_IF_PHASE_OP_RESERVED1_SHIFT 5 -#define INIT_IF_PHASE_OP_CMD_OFFSET_MASK 0xFFFF -#define INIT_IF_PHASE_OP_CMD_OFFSET_SHIFT 16 +#define INIT_IF_PHASE_OP_OP_MASK 0xF +#define INIT_IF_PHASE_OP_OP_SHIFT 0 +#define INIT_IF_PHASE_OP_DMAE_ENABLE_MASK 0x1 +#define INIT_IF_PHASE_OP_DMAE_ENABLE_SHIFT 4 +#define INIT_IF_PHASE_OP_RESERVED1_MASK 0x7FF +#define INIT_IF_PHASE_OP_RESERVED1_SHIFT 5 +#define INIT_IF_PHASE_OP_CMD_OFFSET_MASK 0xFFFF +#define INIT_IF_PHASE_OP_CMD_OFFSET_SHIFT 16 __le32 phase_data; -#define INIT_IF_PHASE_OP_PHASE_MASK 0xFF /* Init phase */ -#define INIT_IF_PHASE_OP_PHASE_SHIFT 0 -#define INIT_IF_PHASE_OP_RESERVED2_MASK 0xFF -#define INIT_IF_PHASE_OP_RESERVED2_SHIFT 8 -#define INIT_IF_PHASE_OP_PHASE_ID_MASK 0xFFFF /* Init phase ID */ -#define INIT_IF_PHASE_OP_PHASE_ID_SHIFT 16 +#define INIT_IF_PHASE_OP_PHASE_MASK 0xFF +#define INIT_IF_PHASE_OP_PHASE_SHIFT 0 +#define INIT_IF_PHASE_OP_RESERVED2_MASK 0xFF +#define INIT_IF_PHASE_OP_RESERVED2_SHIFT 8 +#define INIT_IF_PHASE_OP_PHASE_ID_MASK 0xFFFF +#define INIT_IF_PHASE_OP_PHASE_ID_SHIFT 16 }; /* init mode operators */ enum init_mode_ops { - INIT_MODE_OP_NOT /* init mode not operator */, - INIT_MODE_OP_OR /* init mode or operator */, - INIT_MODE_OP_AND /* init mode and operator */, + INIT_MODE_OP_NOT, + INIT_MODE_OP_OR, + INIT_MODE_OP_AND, MAX_INIT_MODE_OPS }; /* init operation: raw */ struct init_raw_op { - __le32 op_data; -#define INIT_RAW_OP_OP_MASK 0xF -#define INIT_RAW_OP_OP_SHIFT 0 -#define INIT_RAW_OP_PARAM1_MASK 0xFFFFFFF /* init param 1 */ -#define INIT_RAW_OP_PARAM1_SHIFT 4 - __le32 param2 /* Init param 2 */; + __le32 op_data; +#define INIT_RAW_OP_OP_MASK 0xF +#define INIT_RAW_OP_OP_SHIFT 0 +#define INIT_RAW_OP_PARAM1_MASK 0xFFFFFFF +#define INIT_RAW_OP_PARAM1_SHIFT 4 + __le32 param2; }; /* init array params */ struct init_op_array_params { - __le16 size /* array size in dwords */; - __le16 offset /* array start offset in dwords */; + __le16 size; + __le16 offset; }; /* Write init operation arguments */ union init_write_args { - __le32 inline_val; - __le32 zeros_count; - __le32 array_offset; - struct init_op_array_params runtime; + __le32 inline_val; + __le32 zeros_count; + __le32 array_offset; + struct init_op_array_params runtime; }; /* init operation: write */ struct init_write_op { __le32 data; -#define INIT_WRITE_OP_OP_MASK 0xF -#define INIT_WRITE_OP_OP_SHIFT 0 -#define INIT_WRITE_OP_SOURCE_MASK 0x7 -#define INIT_WRITE_OP_SOURCE_SHIFT 4 -#define INIT_WRITE_OP_RESERVED_MASK 0x1 -#define INIT_WRITE_OP_RESERVED_SHIFT 7 -#define INIT_WRITE_OP_WIDE_BUS_MASK 0x1 -#define INIT_WRITE_OP_WIDE_BUS_SHIFT 8 -#define INIT_WRITE_OP_ADDRESS_MASK 0x7FFFFF -#define INIT_WRITE_OP_ADDRESS_SHIFT 9 - union init_write_args args /* Write init operation arguments */; +#define INIT_WRITE_OP_OP_MASK 0xF +#define INIT_WRITE_OP_OP_SHIFT 0 +#define INIT_WRITE_OP_SOURCE_MASK 0x7 +#define INIT_WRITE_OP_SOURCE_SHIFT 4 +#define INIT_WRITE_OP_RESERVED_MASK 0x1 +#define INIT_WRITE_OP_RESERVED_SHIFT 7 +#define INIT_WRITE_OP_WIDE_BUS_MASK 0x1 +#define INIT_WRITE_OP_WIDE_BUS_SHIFT 8 +#define INIT_WRITE_OP_ADDRESS_MASK 0x7FFFFF +#define INIT_WRITE_OP_ADDRESS_SHIFT 9 + union init_write_args args; }; /* init operation: read */ struct init_read_op { __le32 op_data; -#define INIT_READ_OP_OP_MASK 0xF -#define INIT_READ_OP_OP_SHIFT 0 -#define INIT_READ_OP_POLL_TYPE_MASK 0xF -#define INIT_READ_OP_POLL_TYPE_SHIFT 4 -#define INIT_READ_OP_RESERVED_MASK 0x1 -#define INIT_READ_OP_RESERVED_SHIFT 8 -#define INIT_READ_OP_ADDRESS_MASK 0x7FFFFF -#define INIT_READ_OP_ADDRESS_SHIFT 9 +#define INIT_READ_OP_OP_MASK 0xF +#define INIT_READ_OP_OP_SHIFT 0 +#define INIT_READ_OP_POLL_TYPE_MASK 0xF +#define INIT_READ_OP_POLL_TYPE_SHIFT 4 +#define INIT_READ_OP_RESERVED_MASK 0x1 +#define INIT_READ_OP_RESERVED_SHIFT 8 +#define INIT_READ_OP_ADDRESS_MASK 0x7FFFFF +#define INIT_READ_OP_ADDRESS_SHIFT 9 __le32 expected_val; + }; /* Init operations union */ union init_op { - struct init_raw_op raw /* raw init operation */; - struct init_write_op write /* write init operation */; - struct init_read_op read /* read init operation */; - struct init_if_mode_op if_mode /* if_mode init operation */; - struct init_if_phase_op if_phase /* if_phase init operation */; - struct init_callback_op callback /* callback init operation */; - struct init_delay_op delay /* delay init operation */; + struct init_raw_op raw; + struct init_write_op write; + struct init_read_op read; + struct init_if_mode_op if_mode; + struct init_if_phase_op if_phase; + struct init_callback_op callback; + struct init_delay_op delay; }; /* Init command operation types */ enum init_op_types { - INIT_OP_READ /* GRC read init command */, - INIT_OP_WRITE /* GRC write init command */, + INIT_OP_READ, + INIT_OP_WRITE, INIT_OP_IF_MODE, INIT_OP_IF_PHASE, - INIT_OP_DELAY /* delay init command */, - INIT_OP_CALLBACK /* callback init command */, + INIT_OP_DELAY, + INIT_OP_CALLBACK, MAX_INIT_OP_TYPES }; +/* init polling types */ enum init_poll_types { - INIT_POLL_NONE /* No polling */, - INIT_POLL_EQ /* init value is included in the init command */, - INIT_POLL_OR /* init value is all zeros */, - INIT_POLL_AND /* init value is an array of values */, + INIT_POLL_NONE, + INIT_POLL_EQ, + INIT_POLL_OR, + INIT_POLL_AND, MAX_INIT_POLL_TYPES }; /* init source types */ enum init_source_types { - INIT_SRC_INLINE /* init value is included in the init command */, - INIT_SRC_ZEROS /* init value is all zeros */, - INIT_SRC_ARRAY /* init value is an array of values */, - INIT_SRC_RUNTIME /* init value is provided during runtime */, + INIT_SRC_INLINE, + INIT_SRC_ZEROS, + INIT_SRC_ARRAY, + INIT_SRC_RUNTIME, MAX_INIT_SOURCE_TYPES }; /* Internal RAM Offsets macro data */ struct iro { - u32 base /* RAM field offset */; - u16 m1 /* multiplier 1 */; - u16 m2 /* multiplier 2 */; - u16 m3 /* multiplier 3 */; - u16 size /* RAM field size */; + __le32 base; + __le16 m1; + __le16 m2; + __le16 m3; + __le16 size; }; -/* QM per-port init parameters */ -struct init_qm_port_params { - u8 active /* Indicates if this port is active */; - u8 num_active_phys_tcs; - u16 num_pbf_cmd_lines; - u16 num_btb_blocks; - __le16 reserved; -}; - -/* QM per-PQ init parameters */ -struct init_qm_pq_params { - u8 vport_id /* VPORT ID */; - u8 tc_id /* TC ID */; - u8 wrr_group /* WRR group */; - u8 reserved; -}; +/** + * @brief qed_dbg_print_attn - Prints attention registers values in the specified results struct. + * + * @param p_hwfn + * @param results - Pointer to the attention read results + * + * @return error if one of the following holds: + * - the version wasn't set + * Otherwise, returns ok. + */ +enum dbg_status qed_dbg_print_attn(struct qed_hwfn *p_hwfn, + struct dbg_attn_block_result *results); -/* QM per-vport init parameters */ -struct init_qm_vport_params { - u32 vport_rl; - u16 vport_wfq; - u16 first_tx_pq_id[NUM_OF_TCS]; -}; +#define MAX_NAME_LEN 16 /* Win 2 */ #define GTT_BAR0_MAP_REG_IGU_CMD \ 0x00f000UL + /* Win 3 */ #define GTT_BAR0_MAP_REG_TSDM_RAM \ 0x010000UL + /* Win 4 */ #define GTT_BAR0_MAP_REG_MSDM_RAM \ 0x011000UL + /* Win 5 */ #define GTT_BAR0_MAP_REG_MSDM_RAM_1024 \ 0x012000UL + /* Win 6 */ #define GTT_BAR0_MAP_REG_USDM_RAM \ 0x013000UL + /* Win 7 */ #define GTT_BAR0_MAP_REG_USDM_RAM_1024 \ 0x014000UL + /* Win 8 */ #define GTT_BAR0_MAP_REG_USDM_RAM_2048 \ 0x015000UL + /* Win 9 */ #define GTT_BAR0_MAP_REG_XSDM_RAM \ 0x016000UL + /* Win 10 */ #define GTT_BAR0_MAP_REG_YSDM_RAM \ 0x017000UL + /* Win 11 */ #define GTT_BAR0_MAP_REG_PSDM_RAM \ 0x018000UL @@ -1584,785 +1958,718 @@ struct init_qm_vport_params { * Returns the required host memory size in 4KB units. * Must be called before all QM init HSI functions. * - * @param pf_id - physical function ID - * @param num_pf_cids - number of connections used by this PF - * @param num_vf_cids - number of connections used by VFs of this PF - * @param num_tids - number of tasks used by this PF - * @param num_pf_pqs - number of PQs used by this PF - * @param num_vf_pqs - number of PQs used by VFs of this PF + * @param pf_id - physical function ID + * @param num_pf_cids - number of connections used by this PF + * @param num_vf_cids - number of connections used by VFs of this PF + * @param num_tids - number of tasks used by this PF + * @param num_pf_pqs - number of PQs used by this PF + * @param num_vf_pqs - number of PQs used by VFs of this PF * * @return The required host memory size in 4KB units. */ -u32 qed_qm_pf_mem_size(u8 pf_id, - u32 num_pf_cids, - u32 num_vf_cids, - u32 num_tids, - u16 num_pf_pqs, - u16 num_vf_pqs); +u32 qed_qm_pf_mem_size(u8 pf_id, + u32 num_pf_cids, + u32 num_vf_cids, + u32 num_tids, u16 num_pf_pqs, u16 num_vf_pqs); struct qed_qm_common_rt_init_params { - u8 max_ports_per_engine; - u8 max_phys_tcs_per_port; - bool pf_rl_en; - bool pf_wfq_en; - bool vport_rl_en; - bool vport_wfq_en; - struct init_qm_port_params *port_params; + u8 max_ports_per_engine; + u8 max_phys_tcs_per_port; + bool pf_rl_en; + bool pf_wfq_en; + bool vport_rl_en; + bool vport_wfq_en; + struct init_qm_port_params *port_params; }; +int qed_qm_common_rt_init(struct qed_hwfn *p_hwfn, + struct qed_qm_common_rt_init_params *p_params); + +struct qed_qm_pf_rt_init_params { + u8 port_id; + u8 pf_id; + u8 max_phys_tcs_per_port; + bool is_first_pf; + u32 num_pf_cids; + u32 num_vf_cids; + u32 num_tids; + u16 start_pq; + u16 num_pf_pqs; + u16 num_vf_pqs; + u8 start_vport; + u8 num_vports; + u8 pf_wfq; + u32 pf_rl; + struct init_qm_pq_params *pq_params; + struct init_qm_vport_params *vport_params; +}; + +int qed_qm_pf_rt_init(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_qm_pf_rt_init_params *p_params); + /** - * @brief qed_qm_common_rt_init - Prepare QM runtime init values for the - * engine phase. + * @brief qed_init_pf_wfq - Initializes the WFQ weight of the specified PF * * @param p_hwfn - * @param max_ports_per_engine - max number of ports per engine in HW - * @param max_phys_tcs_per_port - max number of physical TCs per port in HW - * @param pf_rl_en - enable per-PF rate limiters - * @param pf_wfq_en - enable per-PF WFQ - * @param vport_rl_en - enable per-VPORT rate limiters - * @param vport_wfq_en - enable per-VPORT WFQ - * @param port_params - array of size MAX_NUM_PORTS with - * arameters for each port + * @param p_ptt - ptt window used for writing the registers + * @param pf_id - PF ID + * @param pf_wfq - WFQ weight. Must be non-zero. * * @return 0 on success, -1 on error. */ -int qed_qm_common_rt_init( - struct qed_hwfn *p_hwfn, - struct qed_qm_common_rt_init_params *p_params); - -struct qed_qm_pf_rt_init_params { - u8 port_id; - u8 pf_id; - u8 max_phys_tcs_per_port; - bool is_first_pf; - u32 num_pf_cids; - u32 num_vf_cids; - u32 num_tids; - u16 start_pq; - u16 num_pf_pqs; - u16 num_vf_pqs; - u8 start_vport; - u8 num_vports; - u8 pf_wfq; - u32 pf_rl; - struct init_qm_pq_params *pq_params; - struct init_qm_vport_params *vport_params; -}; - -int qed_qm_pf_rt_init(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - struct qed_qm_pf_rt_init_params *p_params); +int qed_init_pf_wfq(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, u8 pf_id, u16 pf_wfq); /** - * @brief qed_init_pf_rl Initializes the rate limit of the specified PF + * @brief qed_init_pf_rl - Initializes the rate limit of the specified PF * * @param p_hwfn - * @param p_ptt - ptt window used for writing the registers - * @param pf_id - PF ID - * @param pf_rl - rate limit in Mb/sec units + * @param p_ptt - ptt window used for writing the registers + * @param pf_id - PF ID + * @param pf_rl - rate limit in Mb/sec units * * @return 0 on success, -1 on error. */ -int qed_init_pf_rl(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u8 pf_id, - u32 pf_rl); +int qed_init_pf_rl(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, u8 pf_id, u32 pf_rl); /** - * @brief qed_init_vport_rl Initializes the rate limit of the specified VPORT + * @brief qed_init_vport_wfq Initializes the WFQ weight of the specified VPORT * * @param p_hwfn - * @param p_ptt - ptt window used for writing the registers - * @param vport_id - VPORT ID - * @param vport_rl - rate limit in Mb/sec units + * @param p_ptt - ptt window used for writing the registers + * @param first_tx_pq_id- An array containing the first Tx PQ ID associated + * with the VPORT for each TC. This array is filled by + * qed_qm_pf_rt_init + * @param vport_wfq - WFQ weight. Must be non-zero. * * @return 0 on success, -1 on error. */ +int qed_init_vport_wfq(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + u16 first_tx_pq_id[NUM_OF_TCS], u16 vport_wfq); -int qed_init_vport_rl(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u8 vport_id, - u32 vport_rl); +/** + * @brief qed_init_vport_rl - Initializes the rate limit of the specified VPORT + * + * @param p_hwfn + * @param p_ptt - ptt window used for writing the registers + * @param vport_id - VPORT ID + * @param vport_rl - rate limit in Mb/sec units + * + * @return 0 on success, -1 on error. + */ +int qed_init_vport_rl(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, u8 vport_id, u32 vport_rl); /** * @brief qed_send_qm_stop_cmd Sends a stop command to the QM * * @param p_hwfn - * @param p_ptt - ptt window used for writing the registers + * @param p_ptt * @param is_release_cmd - true for release, false for stop. - * @param is_tx_pq - true for Tx PQs, false for Other PQs. - * @param start_pq - first PQ ID to stop - * @param num_pqs - Number of PQs to stop, starting from start_pq. + * @param is_tx_pq - true for Tx PQs, false for Other PQs. + * @param start_pq - first PQ ID to stop + * @param num_pqs - Number of PQs to stop, starting from start_pq. * - * @return bool, true if successful, false if timeout occurred while waiting - * for QM command done. + * @return bool, true if successful, false if timeout occured while waiting for QM command done. */ +bool qed_send_qm_stop_cmd(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + bool is_release_cmd, + bool is_tx_pq, u16 start_pq, u16 num_pqs); -bool qed_send_qm_stop_cmd(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - bool is_release_cmd, - bool is_tx_pq, - u16 start_pq, - u16 num_pqs); - +/** + * @brief qed_set_vxlan_dest_port - initializes vxlan tunnel destination udp port + * + * @param p_ptt - ptt window used for writing the registers. + * @param dest_port - vxlan destination udp port. + */ void qed_set_vxlan_dest_port(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, u16 dest_port); + struct qed_ptt *p_ptt, u16 dest_port); + +/** + * @brief qed_set_vxlan_enable - enable or disable VXLAN tunnel in HW + * + * @param p_ptt - ptt window used for writing the registers. + * @param vxlan_enable - vxlan enable flag. + */ void qed_set_vxlan_enable(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, bool vxlan_enable); + +/** + * @brief qed_set_gre_enable - enable or disable GRE tunnel in HW + * + * @param p_ptt - ptt window used for writing the registers. + * @param eth_gre_enable - eth GRE enable enable flag. + * @param ip_gre_enable - IP GRE enable enable flag. + */ void qed_set_gre_enable(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, bool eth_gre_enable, - bool ip_gre_enable); + struct qed_ptt *p_ptt, + bool eth_gre_enable, bool ip_gre_enable); + +/** + * @brief qed_set_geneve_dest_port - initializes geneve tunnel destination udp port + * + * @param p_ptt - ptt window used for writing the registers. + * @param dest_port - geneve destination udp port. + */ void qed_set_geneve_dest_port(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u16 dest_port); + +/** + * @brief qed_set_gre_enable - enable or disable GRE tunnel in HW + * + * @param p_ptt - ptt window used for writing the registers. + * @param eth_geneve_enable - eth GENEVE enable enable flag. + * @param ip_geneve_enable - IP GENEVE enable enable flag. + */ void qed_set_geneve_enable(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, bool eth_geneve_enable, - bool ip_geneve_enable); - -/* Ystorm flow control mode. Use enum fw_flow_ctrl_mode */ -#define YSTORM_FLOW_CONTROL_MODE_OFFSET (IRO[0].base) -#define YSTORM_FLOW_CONTROL_MODE_SIZE (IRO[0].size) -/* Tstorm port statistics */ -#define TSTORM_PORT_STAT_OFFSET(port_id) (IRO[1].base + ((port_id) * IRO[1].m1)) -#define TSTORM_PORT_STAT_SIZE (IRO[1].size) -/* Tstorm ll2 port statistics */ -#define TSTORM_LL2_PORT_STAT_OFFSET(port_id) \ - (IRO[2].base + ((port_id) * IRO[2].m1)) -#define TSTORM_LL2_PORT_STAT_SIZE (IRO[2].size) -/* Ustorm VF-PF Channel ready flag */ -#define USTORM_VF_PF_CHANNEL_READY_OFFSET(vf_id) \ - (IRO[3].base + ((vf_id) * IRO[3].m1)) -#define USTORM_VF_PF_CHANNEL_READY_SIZE (IRO[3].size) -/* Ustorm Final flr cleanup ack */ -#define USTORM_FLR_FINAL_ACK_OFFSET(pf_id) (IRO[4].base + ((pf_id) * IRO[4].m1)) -#define USTORM_FLR_FINAL_ACK_SIZE (IRO[4].size) -/* Ustorm Event ring consumer */ -#define USTORM_EQE_CONS_OFFSET(pf_id) (IRO[5].base + ((pf_id) * IRO[5].m1)) -#define USTORM_EQE_CONS_SIZE (IRO[5].size) -/* Ustorm Common Queue ring consumer */ -#define USTORM_COMMON_QUEUE_CONS_OFFSET(global_queue_id) \ - (IRO[6].base + ((global_queue_id) * IRO[6].m1)) -#define USTORM_COMMON_QUEUE_CONS_SIZE (IRO[6].size) -/* Xstorm Integration Test Data */ -#define XSTORM_INTEG_TEST_DATA_OFFSET (IRO[7].base) -#define XSTORM_INTEG_TEST_DATA_SIZE (IRO[7].size) -/* Ystorm Integration Test Data */ -#define YSTORM_INTEG_TEST_DATA_OFFSET (IRO[8].base) -#define YSTORM_INTEG_TEST_DATA_SIZE (IRO[8].size) -/* Pstorm Integration Test Data */ -#define PSTORM_INTEG_TEST_DATA_OFFSET (IRO[9].base) -#define PSTORM_INTEG_TEST_DATA_SIZE (IRO[9].size) -/* Tstorm Integration Test Data */ -#define TSTORM_INTEG_TEST_DATA_OFFSET (IRO[10].base) -#define TSTORM_INTEG_TEST_DATA_SIZE (IRO[10].size) -/* Mstorm Integration Test Data */ -#define MSTORM_INTEG_TEST_DATA_OFFSET (IRO[11].base) -#define MSTORM_INTEG_TEST_DATA_SIZE (IRO[11].size) -/* Ustorm Integration Test Data */ -#define USTORM_INTEG_TEST_DATA_OFFSET (IRO[12].base) -#define USTORM_INTEG_TEST_DATA_SIZE (IRO[12].size) -/* Tstorm producers */ -#define TSTORM_LL2_RX_PRODS_OFFSET(core_rx_queue_id) \ - (IRO[13].base + ((core_rx_queue_id) * IRO[13].m1)) -#define TSTORM_LL2_RX_PRODS_SIZE (IRO[13].size) -/* Tstorm LightL2 queue statistics */ -#define CORE_LL2_TSTORM_PER_QUEUE_STAT_OFFSET(core_rx_queue_id) \ - (IRO[14].base + ((core_rx_queue_id) * IRO[14].m1)) -#define CORE_LL2_TSTORM_PER_QUEUE_STAT_SIZE (IRO[14].size) -/* Ustorm LiteL2 queue statistics */ -#define CORE_LL2_USTORM_PER_QUEUE_STAT_OFFSET(core_rx_queue_id) \ - (IRO[15].base + ((core_rx_queue_id) * IRO[15].m1)) -#define CORE_LL2_USTORM_PER_QUEUE_STAT_SIZE (IRO[15].size) -/* Pstorm LiteL2 queue statistics */ -#define CORE_LL2_PSTORM_PER_QUEUE_STAT_OFFSET(core_tx_stats_id) \ - (IRO[16].base + ((core_tx_stats_id) * IRO[16].m1)) -#define CORE_LL2_PSTORM_PER_QUEUE_STAT_SIZE (IRO[16].size) -/* Mstorm queue statistics */ -#define MSTORM_QUEUE_STAT_OFFSET(stat_counter_id) \ - (IRO[17].base + ((stat_counter_id) * IRO[17].m1)) -#define MSTORM_QUEUE_STAT_SIZE (IRO[17].size) -/* Mstorm producers */ -#define MSTORM_PRODS_OFFSET(queue_id) (IRO[18].base + ((queue_id) * IRO[18].m1)) -#define MSTORM_PRODS_SIZE (IRO[18].size) -/* TPA agregation timeout in us resolution (on ASIC) */ -#define MSTORM_TPA_TIMEOUT_US_OFFSET (IRO[19].base) -#define MSTORM_TPA_TIMEOUT_US_SIZE (IRO[19].size) -/* Ustorm queue statistics */ -#define USTORM_QUEUE_STAT_OFFSET(stat_counter_id) \ - (IRO[20].base + ((stat_counter_id) * IRO[20].m1)) -#define USTORM_QUEUE_STAT_SIZE (IRO[20].size) -/* Ustorm queue zone */ -#define USTORM_ETH_QUEUE_ZONE_OFFSET(queue_id) \ - (IRO[21].base + ((queue_id) * IRO[21].m1)) -#define USTORM_ETH_QUEUE_ZONE_SIZE (IRO[21].size) -/* Pstorm queue statistics */ -#define PSTORM_QUEUE_STAT_OFFSET(stat_counter_id) \ - (IRO[22].base + ((stat_counter_id) * IRO[22].m1)) -#define PSTORM_QUEUE_STAT_SIZE (IRO[22].size) -/* Tstorm last parser message */ -#define TSTORM_ETH_PRS_INPUT_OFFSET (IRO[23].base) -#define TSTORM_ETH_PRS_INPUT_SIZE (IRO[23].size) -/* Tstorm Eth limit Rx rate */ -#define ETH_RX_RATE_LIMIT_OFFSET(pf_id) (IRO[24].base + ((pf_id) * IRO[24].m1)) -#define ETH_RX_RATE_LIMIT_SIZE (IRO[24].size) -/* Ystorm queue zone */ -#define YSTORM_ETH_QUEUE_ZONE_OFFSET(queue_id) \ - (IRO[25].base + ((queue_id) * IRO[25].m1)) -#define YSTORM_ETH_QUEUE_ZONE_SIZE (IRO[25].size) -/* Ystorm cqe producer */ -#define YSTORM_TOE_CQ_PROD_OFFSET(rss_id) \ - (IRO[26].base + ((rss_id) * IRO[26].m1)) -#define YSTORM_TOE_CQ_PROD_SIZE (IRO[26].size) -/* Ustorm cqe producer */ -#define USTORM_TOE_CQ_PROD_OFFSET(rss_id) \ - (IRO[27].base + ((rss_id) * IRO[27].m1)) -#define USTORM_TOE_CQ_PROD_SIZE (IRO[27].size) -/* Ustorm grq producer */ -#define USTORM_TOE_GRQ_PROD_OFFSET(pf_id) \ - (IRO[28].base + ((pf_id) * IRO[28].m1)) -#define USTORM_TOE_GRQ_PROD_SIZE (IRO[28].size) -/* Tstorm cmdq-cons of given command queue-id */ -#define TSTORM_SCSI_CMDQ_CONS_OFFSET(cmdq_queue_id) \ - (IRO[29].base + ((cmdq_queue_id) * IRO[29].m1)) -#define TSTORM_SCSI_CMDQ_CONS_SIZE (IRO[29].size) -/* Mstorm rq-cons of given queue-id */ -#define MSTORM_SCSI_RQ_CONS_OFFSET(rq_queue_id) \ - (IRO[30].base + ((rq_queue_id) * IRO[30].m1)) -#define MSTORM_SCSI_RQ_CONS_SIZE (IRO[30].size) -/* Mstorm bdq-external-producer of given BDQ function ID, BDqueue-id */ -#define MSTORM_SCSI_BDQ_EXT_PROD_OFFSET(func_id, bdq_id) \ - (IRO[31].base + ((func_id) * IRO[31].m1) + ((bdq_id) * IRO[31].m2)) -#define MSTORM_SCSI_BDQ_EXT_PROD_SIZE (IRO[31].size) -/* Tstorm (reflects M-Storm) bdq-external-producer of given fn ID, BDqueue-id */ -#define TSTORM_SCSI_BDQ_EXT_PROD_OFFSET(func_id, bdq_id) \ - (IRO[32].base + ((func_id) * IRO[32].m1) + ((bdq_id) * IRO[32].m2)) -#define TSTORM_SCSI_BDQ_EXT_PROD_SIZE (IRO[32].size) -/* Tstorm iSCSI RX stats */ -#define TSTORM_ISCSI_RX_STATS_OFFSET(pf_id) \ - (IRO[33].base + ((pf_id) * IRO[33].m1)) -#define TSTORM_ISCSI_RX_STATS_SIZE (IRO[33].size) -/* Mstorm iSCSI RX stats */ -#define MSTORM_ISCSI_RX_STATS_OFFSET(pf_id) \ - (IRO[34].base + ((pf_id) * IRO[34].m1)) -#define MSTORM_ISCSI_RX_STATS_SIZE (IRO[34].size) -/* Ustorm iSCSI RX stats */ -#define USTORM_ISCSI_RX_STATS_OFFSET(pf_id) \ - (IRO[35].base + ((pf_id) * IRO[35].m1)) -#define USTORM_ISCSI_RX_STATS_SIZE (IRO[35].size) -/* Xstorm iSCSI TX stats */ -#define XSTORM_ISCSI_TX_STATS_OFFSET(pf_id) \ - (IRO[36].base + ((pf_id) * IRO[36].m1)) -#define XSTORM_ISCSI_TX_STATS_SIZE (IRO[36].size) -/* Ystorm iSCSI TX stats */ -#define YSTORM_ISCSI_TX_STATS_OFFSET(pf_id) \ - (IRO[37].base + ((pf_id) * IRO[37].m1)) -#define YSTORM_ISCSI_TX_STATS_SIZE (IRO[37].size) -/* Pstorm iSCSI TX stats */ -#define PSTORM_ISCSI_TX_STATS_OFFSET(pf_id) \ - (IRO[38].base + ((pf_id) * IRO[38].m1)) -#define PSTORM_ISCSI_TX_STATS_SIZE (IRO[38].size) -/* Tstorm FCoE RX stats */ -#define TSTORM_FCOE_RX_STATS_OFFSET(pf_id) \ - (IRO[39].base + ((pf_id) * IRO[39].m1)) -#define TSTORM_FCOE_RX_STATS_SIZE (IRO[39].size) -/* Mstorm FCoE RX stats */ -#define MSTORM_FCOE_RX_STATS_OFFSET(pf_id) \ - (IRO[40].base + ((pf_id) * IRO[40].m1)) -#define MSTORM_FCOE_RX_STATS_SIZE (IRO[40].size) -/* Pstorm FCoE TX stats */ -#define PSTORM_FCOE_TX_STATS_OFFSET(pf_id) \ - (IRO[41].base + ((pf_id) * IRO[41].m1)) -#define PSTORM_FCOE_TX_STATS_SIZE (IRO[41].size) -/* Pstorm RoCE statistics */ -#define PSTORM_ROCE_STAT_OFFSET(stat_counter_id) \ - (IRO[42].base + ((stat_counter_id) * IRO[42].m1)) -#define PSTORM_ROCE_STAT_SIZE (IRO[42].size) -/* Tstorm RoCE statistics */ -#define TSTORM_ROCE_STAT_OFFSET(stat_counter_id) \ - (IRO[43].base + ((stat_counter_id) * IRO[43].m1)) -#define TSTORM_ROCE_STAT_SIZE (IRO[43].size) - -static const struct iro iro_arr[44] = { - { 0x10, 0x0, 0x0, 0x0, 0x8 }, - { 0x47c8, 0x60, 0x0, 0x0, 0x60 }, - { 0x5e30, 0x20, 0x0, 0x0, 0x20 }, - { 0x510, 0x8, 0x0, 0x0, 0x4 }, - { 0x490, 0x8, 0x0, 0x0, 0x4 }, - { 0x10, 0x8, 0x0, 0x0, 0x2 }, - { 0x90, 0x8, 0x0, 0x0, 0x2 }, - { 0x4940, 0x0, 0x0, 0x0, 0x78 }, - { 0x3de0, 0x0, 0x0, 0x0, 0x78 }, - { 0x2998, 0x0, 0x0, 0x0, 0x78 }, - { 0x4750, 0x0, 0x0, 0x0, 0x78 }, - { 0x56d0, 0x0, 0x0, 0x0, 0x78 }, - { 0x7e50, 0x0, 0x0, 0x0, 0x78 }, - { 0x100, 0x8, 0x0, 0x0, 0x8 }, - { 0x5c10, 0x10, 0x0, 0x0, 0x10 }, - { 0xb508, 0x30, 0x0, 0x0, 0x30 }, - { 0x95c0, 0x30, 0x0, 0x0, 0x30 }, - { 0x58a0, 0x40, 0x0, 0x0, 0x40 }, - { 0x200, 0x10, 0x0, 0x0, 0x8 }, - { 0xa230, 0x0, 0x0, 0x0, 0x4 }, - { 0x8058, 0x40, 0x0, 0x0, 0x30 }, - { 0xd00, 0x8, 0x0, 0x0, 0x8 }, - { 0x2b30, 0x80, 0x0, 0x0, 0x38 }, - { 0xa808, 0x0, 0x0, 0x0, 0xf0 }, - { 0xa8f8, 0x8, 0x0, 0x0, 0x8 }, - { 0x80, 0x8, 0x0, 0x0, 0x8 }, - { 0xac0, 0x8, 0x0, 0x0, 0x8 }, - { 0x2580, 0x8, 0x0, 0x0, 0x8 }, - { 0x2500, 0x8, 0x0, 0x0, 0x8 }, - { 0x440, 0x8, 0x0, 0x0, 0x2 }, - { 0x1800, 0x8, 0x0, 0x0, 0x2 }, - { 0x1a00, 0x10, 0x8, 0x0, 0x2 }, - { 0x640, 0x10, 0x8, 0x0, 0x2 }, - { 0xd9b8, 0x38, 0x0, 0x0, 0x24 }, - { 0x11048, 0x10, 0x0, 0x0, 0x8 }, - { 0x11678, 0x38, 0x0, 0x0, 0x18 }, - { 0xaec0, 0x30, 0x0, 0x0, 0x10 }, - { 0x8700, 0x28, 0x0, 0x0, 0x18 }, - { 0xec00, 0x10, 0x0, 0x0, 0x10 }, - { 0xde38, 0x40, 0x0, 0x0, 0x30 }, - { 0x121a8, 0x38, 0x0, 0x0, 0x8 }, - { 0xf068, 0x20, 0x0, 0x0, 0x20 }, - { 0x2b68, 0x80, 0x0, 0x0, 0x10 }, - { 0x4ab8, 0x10, 0x0, 0x0, 0x10 }, + struct qed_ptt *p_ptt, + bool eth_geneve_enable, bool ip_geneve_enable); + +#define YSTORM_FLOW_CONTROL_MODE_OFFSET (IRO[0].base) +#define YSTORM_FLOW_CONTROL_MODE_SIZE (IRO[0].size) +#define TSTORM_PORT_STAT_OFFSET(port_id) \ + (IRO[1].base + ((port_id) * IRO[1].m1)) +#define TSTORM_PORT_STAT_SIZE (IRO[1].size) +#define USTORM_VF_PF_CHANNEL_READY_OFFSET(vf_id) \ + (IRO[3].base + ((vf_id) * IRO[3].m1)) +#define USTORM_VF_PF_CHANNEL_READY_SIZE (IRO[3].size) +#define USTORM_FLR_FINAL_ACK_OFFSET(pf_id) \ + (IRO[4].base + (pf_id) * IRO[4].m1) +#define USTORM_FLR_FINAL_ACK_SIZE (IRO[4].size) +#define USTORM_EQE_CONS_OFFSET(pf_id) \ + (IRO[5].base + ((pf_id) * IRO[5].m1)) +#define USTORM_EQE_CONS_SIZE (IRO[5].size) +#define USTORM_ETH_QUEUE_ZONE_OFFSET(queue_zone_id) \ + (IRO[6].base + ((queue_zone_id) * IRO[6].m1)) +#define USTORM_ETH_QUEUE_ZONE_SIZE (IRO[6].size) +#define USTORM_COMMON_QUEUE_CONS_OFFSET(queue_zone_id) \ + (IRO[7].base + ((queue_zone_id) * IRO[7].m1)) +#define USTORM_COMMON_QUEUE_CONS_SIZE (IRO[7].size) +#define MSTORM_QUEUE_STAT_OFFSET(stat_counter_id) \ + (IRO[18].base + ((stat_counter_id) * IRO[18].m1)) +#define MSTORM_QUEUE_STAT_SIZE (IRO[18].size) +#define MSTORM_ETH_PF_PRODS_OFFSET(queue_id) \ + (IRO[19].base + ((queue_id) * IRO[19].m1)) +#define MSTORM_ETH_PF_PRODS_SIZE (IRO[19].size) +#define MSTORM_TPA_TIMEOUT_US_OFFSET (IRO[20].base) +#define MSTORM_TPA_TIMEOUT_US_SIZE (IRO[20].size) +#define MSTORM_ETH_PF_STAT_OFFSET(pf_id) \ + (IRO[21].base + ((pf_id) * IRO[21].m1)) +#define MSTORM_ETH_PF_STAT_SIZE (IRO[21].size) +#define USTORM_QUEUE_STAT_OFFSET(stat_counter_id) \ + (IRO[22].base + ((stat_counter_id) * IRO[22].m1)) +#define USTORM_QUEUE_STAT_SIZE (IRO[22].size) +#define USTORM_ETH_PF_STAT_OFFSET(pf_id) \ + (IRO[23].base + ((pf_id) * IRO[23].m1)) +#define USTORM_ETH_PF_STAT_SIZE (IRO[23].size) +#define PSTORM_QUEUE_STAT_OFFSET(stat_counter_id) \ + (IRO[24].base + ((stat_counter_id) * IRO[24].m1)) +#define PSTORM_QUEUE_STAT_SIZE (IRO[24].size) +#define PSTORM_ETH_PF_STAT_OFFSET(pf_id) \ + (IRO[25].base + ((pf_id) * IRO[25].m1)) +#define PSTORM_ETH_PF_STAT_SIZE (IRO[25].size) +#define PSTORM_CTL_FRAME_ETHTYPE_OFFSET(ethtype) \ + (IRO[26].base + ((ethtype) * IRO[26].m1)) +#define PSTORM_CTL_FRAME_ETHTYPE_SIZE (IRO[26].size) +#define TSTORM_ETH_PRS_INPUT_OFFSET (IRO[27].base) +#define TSTORM_ETH_PRS_INPUT_SIZE (IRO[27].size) +#define ETH_RX_RATE_LIMIT_OFFSET(pf_id) \ + (IRO[28].base + ((pf_id) * IRO[28].m1)) +#define ETH_RX_RATE_LIMIT_SIZE (IRO[28].size) +#define XSTORM_ETH_QUEUE_ZONE_OFFSET(queue_id) \ + (IRO[29].base + ((queue_id) * IRO[29].m1)) +#define XSTORM_ETH_QUEUE_ZONE_SIZE (IRO[29].size) + +static const struct iro iro_arr[46] = { + {0x0, 0x0, 0x0, 0x0, 0x8}, + {0x4cb0, 0x78, 0x0, 0x0, 0x78}, + {0x6318, 0x20, 0x0, 0x0, 0x20}, + {0xb00, 0x8, 0x0, 0x0, 0x4}, + {0xa80, 0x8, 0x0, 0x0, 0x4}, + {0x0, 0x8, 0x0, 0x0, 0x2}, + {0x80, 0x8, 0x0, 0x0, 0x4}, + {0x84, 0x8, 0x0, 0x0, 0x2}, + {0x4bc0, 0x0, 0x0, 0x0, 0x78}, + {0x3df0, 0x0, 0x0, 0x0, 0x78}, + {0x29b0, 0x0, 0x0, 0x0, 0x78}, + {0x4c38, 0x0, 0x0, 0x0, 0x78}, + {0x4a48, 0x0, 0x0, 0x0, 0x78}, + {0x7e48, 0x0, 0x0, 0x0, 0x78}, + {0xa28, 0x8, 0x0, 0x0, 0x8}, + {0x60f8, 0x10, 0x0, 0x0, 0x10}, + {0xb820, 0x30, 0x0, 0x0, 0x30}, + {0x95b8, 0x30, 0x0, 0x0, 0x30}, + {0x4c18, 0x80, 0x0, 0x0, 0x40}, + {0x1f8, 0x4, 0x0, 0x0, 0x4}, + {0xc9a8, 0x0, 0x0, 0x0, 0x4}, + {0x4c58, 0x80, 0x0, 0x0, 0x20}, + {0x8050, 0x40, 0x0, 0x0, 0x30}, + {0xe770, 0x60, 0x0, 0x0, 0x60}, + {0x2b48, 0x80, 0x0, 0x0, 0x38}, + {0xdf88, 0x78, 0x0, 0x0, 0x78}, + {0x1f8, 0x4, 0x0, 0x0, 0x4}, + {0xacf0, 0x0, 0x0, 0x0, 0xf0}, + {0xade0, 0x8, 0x0, 0x0, 0x8}, + {0x1f8, 0x8, 0x0, 0x0, 0x8}, + {0xac0, 0x8, 0x0, 0x0, 0x8}, + {0x2578, 0x8, 0x0, 0x0, 0x8}, + {0x24f8, 0x8, 0x0, 0x0, 0x8}, + {0x0, 0x8, 0x0, 0x0, 0x8}, + {0x200, 0x10, 0x8, 0x0, 0x8}, + {0xb78, 0x10, 0x8, 0x0, 0x2}, + {0xd888, 0x38, 0x0, 0x0, 0x24}, + {0x12120, 0x10, 0x0, 0x0, 0x8}, + {0x11b20, 0x38, 0x0, 0x0, 0x18}, + {0xa8c0, 0x30, 0x0, 0x0, 0x10}, + {0x86f8, 0x28, 0x0, 0x0, 0x18}, + {0xeff8, 0x10, 0x0, 0x0, 0x10}, + {0xdd08, 0x48, 0x0, 0x0, 0x38}, + {0xf460, 0x20, 0x0, 0x0, 0x20}, + {0x2b80, 0x80, 0x0, 0x0, 0x10}, + {0x5000, 0x10, 0x0, 0x0, 0x10}, }; /* Runtime array offsets */ -#define DORQ_REG_PF_MAX_ICID_0_RT_OFFSET 0 -#define DORQ_REG_PF_MAX_ICID_1_RT_OFFSET 1 -#define DORQ_REG_PF_MAX_ICID_2_RT_OFFSET 2 -#define DORQ_REG_PF_MAX_ICID_3_RT_OFFSET 3 -#define DORQ_REG_PF_MAX_ICID_4_RT_OFFSET 4 -#define DORQ_REG_PF_MAX_ICID_5_RT_OFFSET 5 -#define DORQ_REG_PF_MAX_ICID_6_RT_OFFSET 6 -#define DORQ_REG_PF_MAX_ICID_7_RT_OFFSET 7 -#define DORQ_REG_VF_MAX_ICID_0_RT_OFFSET 8 -#define DORQ_REG_VF_MAX_ICID_1_RT_OFFSET 9 -#define DORQ_REG_VF_MAX_ICID_2_RT_OFFSET 10 -#define DORQ_REG_VF_MAX_ICID_3_RT_OFFSET 11 -#define DORQ_REG_VF_MAX_ICID_4_RT_OFFSET 12 -#define DORQ_REG_VF_MAX_ICID_5_RT_OFFSET 13 -#define DORQ_REG_VF_MAX_ICID_6_RT_OFFSET 14 -#define DORQ_REG_VF_MAX_ICID_7_RT_OFFSET 15 -#define DORQ_REG_PF_WAKE_ALL_RT_OFFSET 16 -#define DORQ_REG_TAG1_ETHERTYPE_RT_OFFSET 17 -#define IGU_REG_PF_CONFIGURATION_RT_OFFSET 18 -#define IGU_REG_VF_CONFIGURATION_RT_OFFSET 19 -#define IGU_REG_ATTN_MSG_ADDR_L_RT_OFFSET 20 -#define IGU_REG_ATTN_MSG_ADDR_H_RT_OFFSET 21 -#define IGU_REG_LEADING_EDGE_LATCH_RT_OFFSET 22 -#define IGU_REG_TRAILING_EDGE_LATCH_RT_OFFSET 23 -#define CAU_REG_CQE_AGG_UNIT_SIZE_RT_OFFSET 24 -#define CAU_REG_SB_VAR_MEMORY_RT_OFFSET 761 -#define CAU_REG_SB_VAR_MEMORY_RT_SIZE 736 -#define CAU_REG_SB_VAR_MEMORY_RT_OFFSET 761 -#define CAU_REG_SB_VAR_MEMORY_RT_SIZE 736 -#define CAU_REG_SB_ADDR_MEMORY_RT_OFFSET 1497 -#define CAU_REG_SB_ADDR_MEMORY_RT_SIZE 736 -#define CAU_REG_PI_MEMORY_RT_OFFSET 2233 -#define CAU_REG_PI_MEMORY_RT_SIZE 4416 -#define PRS_REG_SEARCH_RESP_INITIATOR_TYPE_RT_OFFSET 6649 -#define PRS_REG_TASK_ID_MAX_INITIATOR_PF_RT_OFFSET 6650 -#define PRS_REG_TASK_ID_MAX_INITIATOR_VF_RT_OFFSET 6651 -#define PRS_REG_TASK_ID_MAX_TARGET_PF_RT_OFFSET 6652 -#define PRS_REG_TASK_ID_MAX_TARGET_VF_RT_OFFSET 6653 -#define PRS_REG_SEARCH_TCP_RT_OFFSET 6654 -#define PRS_REG_SEARCH_FCOE_RT_OFFSET 6655 -#define PRS_REG_SEARCH_ROCE_RT_OFFSET 6656 -#define PRS_REG_ROCE_DEST_QP_MAX_VF_RT_OFFSET 6657 -#define PRS_REG_ROCE_DEST_QP_MAX_PF_RT_OFFSET 6658 -#define PRS_REG_SEARCH_OPENFLOW_RT_OFFSET 6659 -#define PRS_REG_SEARCH_NON_IP_AS_OPENFLOW_RT_OFFSET 6660 -#define PRS_REG_OPENFLOW_SUPPORT_ONLY_KNOWN_OVER_IP_RT_OFFSET 6661 -#define PRS_REG_OPENFLOW_SEARCH_KEY_MASK_RT_OFFSET 6662 -#define PRS_REG_TAG_ETHERTYPE_0_RT_OFFSET 6663 -#define PRS_REG_LIGHT_L2_ETHERTYPE_EN_RT_OFFSET 6664 -#define SRC_REG_FIRSTFREE_RT_OFFSET 6665 -#define SRC_REG_FIRSTFREE_RT_SIZE 2 -#define SRC_REG_LASTFREE_RT_OFFSET 6667 -#define SRC_REG_LASTFREE_RT_SIZE 2 -#define SRC_REG_COUNTFREE_RT_OFFSET 6669 -#define SRC_REG_NUMBER_HASH_BITS_RT_OFFSET 6670 -#define PSWRQ2_REG_CDUT_P_SIZE_RT_OFFSET 6671 -#define PSWRQ2_REG_CDUC_P_SIZE_RT_OFFSET 6672 -#define PSWRQ2_REG_TM_P_SIZE_RT_OFFSET 6673 -#define PSWRQ2_REG_QM_P_SIZE_RT_OFFSET 6674 -#define PSWRQ2_REG_SRC_P_SIZE_RT_OFFSET 6675 -#define PSWRQ2_REG_TM_FIRST_ILT_RT_OFFSET 6676 -#define PSWRQ2_REG_TM_LAST_ILT_RT_OFFSET 6677 -#define PSWRQ2_REG_QM_FIRST_ILT_RT_OFFSET 6678 -#define PSWRQ2_REG_QM_LAST_ILT_RT_OFFSET 6679 -#define PSWRQ2_REG_SRC_FIRST_ILT_RT_OFFSET 6680 -#define PSWRQ2_REG_SRC_LAST_ILT_RT_OFFSET 6681 -#define PSWRQ2_REG_CDUC_FIRST_ILT_RT_OFFSET 6682 -#define PSWRQ2_REG_CDUC_LAST_ILT_RT_OFFSET 6683 -#define PSWRQ2_REG_CDUT_FIRST_ILT_RT_OFFSET 6684 -#define PSWRQ2_REG_CDUT_LAST_ILT_RT_OFFSET 6685 -#define PSWRQ2_REG_TSDM_FIRST_ILT_RT_OFFSET 6686 -#define PSWRQ2_REG_TSDM_LAST_ILT_RT_OFFSET 6687 -#define PSWRQ2_REG_TM_NUMBER_OF_PF_BLOCKS_RT_OFFSET 6688 -#define PSWRQ2_REG_CDUT_NUMBER_OF_PF_BLOCKS_RT_OFFSET 6689 -#define PSWRQ2_REG_CDUC_NUMBER_OF_PF_BLOCKS_RT_OFFSET 6690 -#define PSWRQ2_REG_TM_VF_BLOCKS_RT_OFFSET 6691 -#define PSWRQ2_REG_CDUT_VF_BLOCKS_RT_OFFSET 6692 -#define PSWRQ2_REG_CDUC_VF_BLOCKS_RT_OFFSET 6693 -#define PSWRQ2_REG_TM_BLOCKS_FACTOR_RT_OFFSET 6694 -#define PSWRQ2_REG_CDUT_BLOCKS_FACTOR_RT_OFFSET 6695 -#define PSWRQ2_REG_CDUC_BLOCKS_FACTOR_RT_OFFSET 6696 -#define PSWRQ2_REG_VF_BASE_RT_OFFSET 6697 -#define PSWRQ2_REG_VF_LAST_ILT_RT_OFFSET 6698 -#define PSWRQ2_REG_WR_MBS0_RT_OFFSET 6699 -#define PSWRQ2_REG_RD_MBS0_RT_OFFSET 6700 -#define PSWRQ2_REG_DRAM_ALIGN_WR_RT_OFFSET 6701 -#define PSWRQ2_REG_DRAM_ALIGN_RD_RT_OFFSET 6702 -#define PSWRQ2_REG_ILT_MEMORY_RT_OFFSET 6703 -#define PSWRQ2_REG_ILT_MEMORY_RT_SIZE 22000 -#define PGLUE_REG_B_VF_BASE_RT_OFFSET 28703 -#define PGLUE_REG_B_CACHE_LINE_SIZE_RT_OFFSET 28704 -#define PGLUE_REG_B_PF_BAR0_SIZE_RT_OFFSET 28705 -#define PGLUE_REG_B_PF_BAR1_SIZE_RT_OFFSET 28706 -#define PGLUE_REG_B_VF_BAR1_SIZE_RT_OFFSET 28707 -#define TM_REG_VF_ENABLE_CONN_RT_OFFSET 28708 -#define TM_REG_PF_ENABLE_CONN_RT_OFFSET 28709 -#define TM_REG_PF_ENABLE_TASK_RT_OFFSET 28710 -#define TM_REG_GROUP_SIZE_RESOLUTION_CONN_RT_OFFSET 28711 -#define TM_REG_GROUP_SIZE_RESOLUTION_TASK_RT_OFFSET 28712 -#define TM_REG_CONFIG_CONN_MEM_RT_OFFSET 28713 -#define TM_REG_CONFIG_CONN_MEM_RT_SIZE 416 -#define TM_REG_CONFIG_TASK_MEM_RT_OFFSET 29129 -#define TM_REG_CONFIG_TASK_MEM_RT_SIZE 512 -#define QM_REG_MAXPQSIZE_0_RT_OFFSET 29641 -#define QM_REG_MAXPQSIZE_1_RT_OFFSET 29642 -#define QM_REG_MAXPQSIZE_2_RT_OFFSET 29643 -#define QM_REG_MAXPQSIZETXSEL_0_RT_OFFSET 29644 -#define QM_REG_MAXPQSIZETXSEL_1_RT_OFFSET 29645 -#define QM_REG_MAXPQSIZETXSEL_2_RT_OFFSET 29646 -#define QM_REG_MAXPQSIZETXSEL_3_RT_OFFSET 29647 -#define QM_REG_MAXPQSIZETXSEL_4_RT_OFFSET 29648 -#define QM_REG_MAXPQSIZETXSEL_5_RT_OFFSET 29649 -#define QM_REG_MAXPQSIZETXSEL_6_RT_OFFSET 29650 -#define QM_REG_MAXPQSIZETXSEL_7_RT_OFFSET 29651 -#define QM_REG_MAXPQSIZETXSEL_8_RT_OFFSET 29652 -#define QM_REG_MAXPQSIZETXSEL_9_RT_OFFSET 29653 -#define QM_REG_MAXPQSIZETXSEL_10_RT_OFFSET 29654 -#define QM_REG_MAXPQSIZETXSEL_11_RT_OFFSET 29655 -#define QM_REG_MAXPQSIZETXSEL_12_RT_OFFSET 29656 -#define QM_REG_MAXPQSIZETXSEL_13_RT_OFFSET 29657 -#define QM_REG_MAXPQSIZETXSEL_14_RT_OFFSET 29658 -#define QM_REG_MAXPQSIZETXSEL_15_RT_OFFSET 29659 -#define QM_REG_MAXPQSIZETXSEL_16_RT_OFFSET 29660 -#define QM_REG_MAXPQSIZETXSEL_17_RT_OFFSET 29661 -#define QM_REG_MAXPQSIZETXSEL_18_RT_OFFSET 29662 -#define QM_REG_MAXPQSIZETXSEL_19_RT_OFFSET 29663 -#define QM_REG_MAXPQSIZETXSEL_20_RT_OFFSET 29664 -#define QM_REG_MAXPQSIZETXSEL_21_RT_OFFSET 29665 -#define QM_REG_MAXPQSIZETXSEL_22_RT_OFFSET 29666 -#define QM_REG_MAXPQSIZETXSEL_23_RT_OFFSET 29667 -#define QM_REG_MAXPQSIZETXSEL_24_RT_OFFSET 29668 -#define QM_REG_MAXPQSIZETXSEL_25_RT_OFFSET 29669 -#define QM_REG_MAXPQSIZETXSEL_26_RT_OFFSET 29670 -#define QM_REG_MAXPQSIZETXSEL_27_RT_OFFSET 29671 -#define QM_REG_MAXPQSIZETXSEL_28_RT_OFFSET 29672 -#define QM_REG_MAXPQSIZETXSEL_29_RT_OFFSET 29673 -#define QM_REG_MAXPQSIZETXSEL_30_RT_OFFSET 29674 -#define QM_REG_MAXPQSIZETXSEL_31_RT_OFFSET 29675 -#define QM_REG_MAXPQSIZETXSEL_32_RT_OFFSET 29676 -#define QM_REG_MAXPQSIZETXSEL_33_RT_OFFSET 29677 -#define QM_REG_MAXPQSIZETXSEL_34_RT_OFFSET 29678 -#define QM_REG_MAXPQSIZETXSEL_35_RT_OFFSET 29679 -#define QM_REG_MAXPQSIZETXSEL_36_RT_OFFSET 29680 -#define QM_REG_MAXPQSIZETXSEL_37_RT_OFFSET 29681 -#define QM_REG_MAXPQSIZETXSEL_38_RT_OFFSET 29682 -#define QM_REG_MAXPQSIZETXSEL_39_RT_OFFSET 29683 -#define QM_REG_MAXPQSIZETXSEL_40_RT_OFFSET 29684 -#define QM_REG_MAXPQSIZETXSEL_41_RT_OFFSET 29685 -#define QM_REG_MAXPQSIZETXSEL_42_RT_OFFSET 29686 -#define QM_REG_MAXPQSIZETXSEL_43_RT_OFFSET 29687 -#define QM_REG_MAXPQSIZETXSEL_44_RT_OFFSET 29688 -#define QM_REG_MAXPQSIZETXSEL_45_RT_OFFSET 29689 -#define QM_REG_MAXPQSIZETXSEL_46_RT_OFFSET 29690 -#define QM_REG_MAXPQSIZETXSEL_47_RT_OFFSET 29691 -#define QM_REG_MAXPQSIZETXSEL_48_RT_OFFSET 29692 -#define QM_REG_MAXPQSIZETXSEL_49_RT_OFFSET 29693 -#define QM_REG_MAXPQSIZETXSEL_50_RT_OFFSET 29694 -#define QM_REG_MAXPQSIZETXSEL_51_RT_OFFSET 29695 -#define QM_REG_MAXPQSIZETXSEL_52_RT_OFFSET 29696 -#define QM_REG_MAXPQSIZETXSEL_53_RT_OFFSET 29697 -#define QM_REG_MAXPQSIZETXSEL_54_RT_OFFSET 29698 -#define QM_REG_MAXPQSIZETXSEL_55_RT_OFFSET 29699 -#define QM_REG_MAXPQSIZETXSEL_56_RT_OFFSET 29700 -#define QM_REG_MAXPQSIZETXSEL_57_RT_OFFSET 29701 -#define QM_REG_MAXPQSIZETXSEL_58_RT_OFFSET 29702 -#define QM_REG_MAXPQSIZETXSEL_59_RT_OFFSET 29703 -#define QM_REG_MAXPQSIZETXSEL_60_RT_OFFSET 29704 -#define QM_REG_MAXPQSIZETXSEL_61_RT_OFFSET 29705 -#define QM_REG_MAXPQSIZETXSEL_62_RT_OFFSET 29706 -#define QM_REG_MAXPQSIZETXSEL_63_RT_OFFSET 29707 -#define QM_REG_BASEADDROTHERPQ_RT_OFFSET 29708 -#define QM_REG_BASEADDROTHERPQ_RT_SIZE 128 -#define QM_REG_VOQCRDLINE_RT_OFFSET 29836 -#define QM_REG_VOQCRDLINE_RT_SIZE 20 -#define QM_REG_VOQINITCRDLINE_RT_OFFSET 29856 -#define QM_REG_VOQINITCRDLINE_RT_SIZE 20 -#define QM_REG_AFULLQMBYPTHRPFWFQ_RT_OFFSET 29876 -#define QM_REG_AFULLQMBYPTHRVPWFQ_RT_OFFSET 29877 -#define QM_REG_AFULLQMBYPTHRPFRL_RT_OFFSET 29878 -#define QM_REG_AFULLQMBYPTHRGLBLRL_RT_OFFSET 29879 -#define QM_REG_AFULLOPRTNSTCCRDMASK_RT_OFFSET 29880 -#define QM_REG_WRROTHERPQGRP_0_RT_OFFSET 29881 -#define QM_REG_WRROTHERPQGRP_1_RT_OFFSET 29882 -#define QM_REG_WRROTHERPQGRP_2_RT_OFFSET 29883 -#define QM_REG_WRROTHERPQGRP_3_RT_OFFSET 29884 -#define QM_REG_WRROTHERPQGRP_4_RT_OFFSET 29885 -#define QM_REG_WRROTHERPQGRP_5_RT_OFFSET 29886 -#define QM_REG_WRROTHERPQGRP_6_RT_OFFSET 29887 -#define QM_REG_WRROTHERPQGRP_7_RT_OFFSET 29888 -#define QM_REG_WRROTHERPQGRP_8_RT_OFFSET 29889 -#define QM_REG_WRROTHERPQGRP_9_RT_OFFSET 29890 -#define QM_REG_WRROTHERPQGRP_10_RT_OFFSET 29891 -#define QM_REG_WRROTHERPQGRP_11_RT_OFFSET 29892 -#define QM_REG_WRROTHERPQGRP_12_RT_OFFSET 29893 -#define QM_REG_WRROTHERPQGRP_13_RT_OFFSET 29894 -#define QM_REG_WRROTHERPQGRP_14_RT_OFFSET 29895 -#define QM_REG_WRROTHERPQGRP_15_RT_OFFSET 29896 -#define QM_REG_WRROTHERGRPWEIGHT_0_RT_OFFSET 29897 -#define QM_REG_WRROTHERGRPWEIGHT_1_RT_OFFSET 29898 -#define QM_REG_WRROTHERGRPWEIGHT_2_RT_OFFSET 29899 -#define QM_REG_WRROTHERGRPWEIGHT_3_RT_OFFSET 29900 -#define QM_REG_WRRTXGRPWEIGHT_0_RT_OFFSET 29901 -#define QM_REG_WRRTXGRPWEIGHT_1_RT_OFFSET 29902 -#define QM_REG_PQTX2PF_0_RT_OFFSET 29903 -#define QM_REG_PQTX2PF_1_RT_OFFSET 29904 -#define QM_REG_PQTX2PF_2_RT_OFFSET 29905 -#define QM_REG_PQTX2PF_3_RT_OFFSET 29906 -#define QM_REG_PQTX2PF_4_RT_OFFSET 29907 -#define QM_REG_PQTX2PF_5_RT_OFFSET 29908 -#define QM_REG_PQTX2PF_6_RT_OFFSET 29909 -#define QM_REG_PQTX2PF_7_RT_OFFSET 29910 -#define QM_REG_PQTX2PF_8_RT_OFFSET 29911 -#define QM_REG_PQTX2PF_9_RT_OFFSET 29912 -#define QM_REG_PQTX2PF_10_RT_OFFSET 29913 -#define QM_REG_PQTX2PF_11_RT_OFFSET 29914 -#define QM_REG_PQTX2PF_12_RT_OFFSET 29915 -#define QM_REG_PQTX2PF_13_RT_OFFSET 29916 -#define QM_REG_PQTX2PF_14_RT_OFFSET 29917 -#define QM_REG_PQTX2PF_15_RT_OFFSET 29918 -#define QM_REG_PQTX2PF_16_RT_OFFSET 29919 -#define QM_REG_PQTX2PF_17_RT_OFFSET 29920 -#define QM_REG_PQTX2PF_18_RT_OFFSET 29921 -#define QM_REG_PQTX2PF_19_RT_OFFSET 29922 -#define QM_REG_PQTX2PF_20_RT_OFFSET 29923 -#define QM_REG_PQTX2PF_21_RT_OFFSET 29924 -#define QM_REG_PQTX2PF_22_RT_OFFSET 29925 -#define QM_REG_PQTX2PF_23_RT_OFFSET 29926 -#define QM_REG_PQTX2PF_24_RT_OFFSET 29927 -#define QM_REG_PQTX2PF_25_RT_OFFSET 29928 -#define QM_REG_PQTX2PF_26_RT_OFFSET 29929 -#define QM_REG_PQTX2PF_27_RT_OFFSET 29930 -#define QM_REG_PQTX2PF_28_RT_OFFSET 29931 -#define QM_REG_PQTX2PF_29_RT_OFFSET 29932 -#define QM_REG_PQTX2PF_30_RT_OFFSET 29933 -#define QM_REG_PQTX2PF_31_RT_OFFSET 29934 -#define QM_REG_PQTX2PF_32_RT_OFFSET 29935 -#define QM_REG_PQTX2PF_33_RT_OFFSET 29936 -#define QM_REG_PQTX2PF_34_RT_OFFSET 29937 -#define QM_REG_PQTX2PF_35_RT_OFFSET 29938 -#define QM_REG_PQTX2PF_36_RT_OFFSET 29939 -#define QM_REG_PQTX2PF_37_RT_OFFSET 29940 -#define QM_REG_PQTX2PF_38_RT_OFFSET 29941 -#define QM_REG_PQTX2PF_39_RT_OFFSET 29942 -#define QM_REG_PQTX2PF_40_RT_OFFSET 29943 -#define QM_REG_PQTX2PF_41_RT_OFFSET 29944 -#define QM_REG_PQTX2PF_42_RT_OFFSET 29945 -#define QM_REG_PQTX2PF_43_RT_OFFSET 29946 -#define QM_REG_PQTX2PF_44_RT_OFFSET 29947 -#define QM_REG_PQTX2PF_45_RT_OFFSET 29948 -#define QM_REG_PQTX2PF_46_RT_OFFSET 29949 -#define QM_REG_PQTX2PF_47_RT_OFFSET 29950 -#define QM_REG_PQTX2PF_48_RT_OFFSET 29951 -#define QM_REG_PQTX2PF_49_RT_OFFSET 29952 -#define QM_REG_PQTX2PF_50_RT_OFFSET 29953 -#define QM_REG_PQTX2PF_51_RT_OFFSET 29954 -#define QM_REG_PQTX2PF_52_RT_OFFSET 29955 -#define QM_REG_PQTX2PF_53_RT_OFFSET 29956 -#define QM_REG_PQTX2PF_54_RT_OFFSET 29957 -#define QM_REG_PQTX2PF_55_RT_OFFSET 29958 -#define QM_REG_PQTX2PF_56_RT_OFFSET 29959 -#define QM_REG_PQTX2PF_57_RT_OFFSET 29960 -#define QM_REG_PQTX2PF_58_RT_OFFSET 29961 -#define QM_REG_PQTX2PF_59_RT_OFFSET 29962 -#define QM_REG_PQTX2PF_60_RT_OFFSET 29963 -#define QM_REG_PQTX2PF_61_RT_OFFSET 29964 -#define QM_REG_PQTX2PF_62_RT_OFFSET 29965 -#define QM_REG_PQTX2PF_63_RT_OFFSET 29966 -#define QM_REG_PQOTHER2PF_0_RT_OFFSET 29967 -#define QM_REG_PQOTHER2PF_1_RT_OFFSET 29968 -#define QM_REG_PQOTHER2PF_2_RT_OFFSET 29969 -#define QM_REG_PQOTHER2PF_3_RT_OFFSET 29970 -#define QM_REG_PQOTHER2PF_4_RT_OFFSET 29971 -#define QM_REG_PQOTHER2PF_5_RT_OFFSET 29972 -#define QM_REG_PQOTHER2PF_6_RT_OFFSET 29973 -#define QM_REG_PQOTHER2PF_7_RT_OFFSET 29974 -#define QM_REG_PQOTHER2PF_8_RT_OFFSET 29975 -#define QM_REG_PQOTHER2PF_9_RT_OFFSET 29976 -#define QM_REG_PQOTHER2PF_10_RT_OFFSET 29977 -#define QM_REG_PQOTHER2PF_11_RT_OFFSET 29978 -#define QM_REG_PQOTHER2PF_12_RT_OFFSET 29979 -#define QM_REG_PQOTHER2PF_13_RT_OFFSET 29980 -#define QM_REG_PQOTHER2PF_14_RT_OFFSET 29981 -#define QM_REG_PQOTHER2PF_15_RT_OFFSET 29982 -#define QM_REG_RLGLBLPERIOD_0_RT_OFFSET 29983 -#define QM_REG_RLGLBLPERIOD_1_RT_OFFSET 29984 -#define QM_REG_RLGLBLPERIODTIMER_0_RT_OFFSET 29985 -#define QM_REG_RLGLBLPERIODTIMER_1_RT_OFFSET 29986 -#define QM_REG_RLGLBLPERIODSEL_0_RT_OFFSET 29987 -#define QM_REG_RLGLBLPERIODSEL_1_RT_OFFSET 29988 -#define QM_REG_RLGLBLPERIODSEL_2_RT_OFFSET 29989 -#define QM_REG_RLGLBLPERIODSEL_3_RT_OFFSET 29990 -#define QM_REG_RLGLBLPERIODSEL_4_RT_OFFSET 29991 -#define QM_REG_RLGLBLPERIODSEL_5_RT_OFFSET 29992 -#define QM_REG_RLGLBLPERIODSEL_6_RT_OFFSET 29993 -#define QM_REG_RLGLBLPERIODSEL_7_RT_OFFSET 29994 -#define QM_REG_RLGLBLINCVAL_RT_OFFSET 29995 -#define QM_REG_RLGLBLINCVAL_RT_SIZE 256 -#define QM_REG_RLGLBLUPPERBOUND_RT_OFFSET 30251 -#define QM_REG_RLGLBLUPPERBOUND_RT_SIZE 256 -#define QM_REG_RLGLBLCRD_RT_OFFSET 30507 -#define QM_REG_RLGLBLCRD_RT_SIZE 256 -#define QM_REG_RLGLBLENABLE_RT_OFFSET 30763 -#define QM_REG_RLPFPERIOD_RT_OFFSET 30764 -#define QM_REG_RLPFPERIODTIMER_RT_OFFSET 30765 -#define QM_REG_RLPFINCVAL_RT_OFFSET 30766 -#define QM_REG_RLPFINCVAL_RT_SIZE 16 -#define QM_REG_RLPFUPPERBOUND_RT_OFFSET 30782 -#define QM_REG_RLPFUPPERBOUND_RT_SIZE 16 -#define QM_REG_RLPFCRD_RT_OFFSET 30798 -#define QM_REG_RLPFCRD_RT_SIZE 16 -#define QM_REG_RLPFENABLE_RT_OFFSET 30814 -#define QM_REG_RLPFVOQENABLE_RT_OFFSET 30815 -#define QM_REG_WFQPFWEIGHT_RT_OFFSET 30816 -#define QM_REG_WFQPFWEIGHT_RT_SIZE 16 -#define QM_REG_WFQPFUPPERBOUND_RT_OFFSET 30832 -#define QM_REG_WFQPFUPPERBOUND_RT_SIZE 16 -#define QM_REG_WFQPFCRD_RT_OFFSET 30848 -#define QM_REG_WFQPFCRD_RT_SIZE 160 -#define QM_REG_WFQPFENABLE_RT_OFFSET 31008 -#define QM_REG_WFQVPENABLE_RT_OFFSET 31009 -#define QM_REG_BASEADDRTXPQ_RT_OFFSET 31010 -#define QM_REG_BASEADDRTXPQ_RT_SIZE 512 -#define QM_REG_TXPQMAP_RT_OFFSET 31522 -#define QM_REG_TXPQMAP_RT_SIZE 512 -#define QM_REG_WFQVPWEIGHT_RT_OFFSET 32034 -#define QM_REG_WFQVPWEIGHT_RT_SIZE 512 -#define QM_REG_WFQVPCRD_RT_OFFSET 32546 -#define QM_REG_WFQVPCRD_RT_SIZE 512 -#define QM_REG_WFQVPMAP_RT_OFFSET 33058 -#define QM_REG_WFQVPMAP_RT_SIZE 512 -#define QM_REG_WFQPFCRD_MSB_RT_OFFSET 33570 -#define QM_REG_WFQPFCRD_MSB_RT_SIZE 160 -#define NIG_REG_TAG_ETHERTYPE_0_RT_OFFSET 33730 -#define NIG_REG_OUTER_TAG_VALUE_LIST0_RT_OFFSET 33731 -#define NIG_REG_OUTER_TAG_VALUE_LIST1_RT_OFFSET 33732 -#define NIG_REG_OUTER_TAG_VALUE_LIST2_RT_OFFSET 33733 -#define NIG_REG_OUTER_TAG_VALUE_LIST3_RT_OFFSET 33734 -#define NIG_REG_OUTER_TAG_VALUE_MASK_RT_OFFSET 33735 -#define NIG_REG_LLH_FUNC_TAGMAC_CLS_TYPE_RT_OFFSET 33736 -#define NIG_REG_LLH_FUNC_TAG_EN_RT_OFFSET 33737 -#define NIG_REG_LLH_FUNC_TAG_EN_RT_SIZE 4 -#define NIG_REG_LLH_FUNC_TAG_HDR_SEL_RT_OFFSET 33741 -#define NIG_REG_LLH_FUNC_TAG_HDR_SEL_RT_SIZE 4 -#define NIG_REG_LLH_FUNC_TAG_VALUE_RT_OFFSET 33745 -#define NIG_REG_LLH_FUNC_TAG_VALUE_RT_SIZE 4 -#define NIG_REG_LLH_FUNC_NO_TAG_RT_OFFSET 33749 -#define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_OFFSET 33750 -#define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_SIZE 32 -#define NIG_REG_LLH_FUNC_FILTER_EN_RT_OFFSET 33782 -#define NIG_REG_LLH_FUNC_FILTER_EN_RT_SIZE 16 -#define NIG_REG_LLH_FUNC_FILTER_MODE_RT_OFFSET 33798 -#define NIG_REG_LLH_FUNC_FILTER_MODE_RT_SIZE 16 -#define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_OFFSET 33814 -#define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_SIZE 16 -#define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_OFFSET 33830 -#define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_SIZE 16 -#define NIG_REG_TX_EDPM_CTRL_RT_OFFSET 33846 -#define CDU_REG_CID_ADDR_PARAMS_RT_OFFSET 33847 -#define CDU_REG_SEGMENT0_PARAMS_RT_OFFSET 33848 -#define CDU_REG_SEGMENT1_PARAMS_RT_OFFSET 33849 -#define CDU_REG_PF_SEG0_TYPE_OFFSET_RT_OFFSET 33850 -#define CDU_REG_PF_SEG1_TYPE_OFFSET_RT_OFFSET 33851 -#define CDU_REG_PF_SEG2_TYPE_OFFSET_RT_OFFSET 33852 -#define CDU_REG_PF_SEG3_TYPE_OFFSET_RT_OFFSET 33853 -#define CDU_REG_PF_FL_SEG0_TYPE_OFFSET_RT_OFFSET 33854 -#define CDU_REG_PF_FL_SEG1_TYPE_OFFSET_RT_OFFSET 33855 -#define CDU_REG_PF_FL_SEG2_TYPE_OFFSET_RT_OFFSET 33856 -#define CDU_REG_PF_FL_SEG3_TYPE_OFFSET_RT_OFFSET 33857 -#define CDU_REG_VF_SEG_TYPE_OFFSET_RT_OFFSET 33858 -#define CDU_REG_VF_FL_SEG_TYPE_OFFSET_RT_OFFSET 33859 -#define PBF_REG_TAG_ETHERTYPE_0_RT_OFFSET 33860 -#define PBF_REG_BTB_SHARED_AREA_SIZE_RT_OFFSET 33861 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ0_RT_OFFSET 33862 -#define PBF_REG_BTB_GUARANTEED_VOQ0_RT_OFFSET 33863 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ0_RT_OFFSET 33864 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ1_RT_OFFSET 33865 -#define PBF_REG_BTB_GUARANTEED_VOQ1_RT_OFFSET 33866 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ1_RT_OFFSET 33867 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ2_RT_OFFSET 33868 -#define PBF_REG_BTB_GUARANTEED_VOQ2_RT_OFFSET 33869 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ2_RT_OFFSET 33870 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ3_RT_OFFSET 33871 -#define PBF_REG_BTB_GUARANTEED_VOQ3_RT_OFFSET 33872 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ3_RT_OFFSET 33873 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ4_RT_OFFSET 33874 -#define PBF_REG_BTB_GUARANTEED_VOQ4_RT_OFFSET 33875 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ4_RT_OFFSET 33876 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ5_RT_OFFSET 33877 -#define PBF_REG_BTB_GUARANTEED_VOQ5_RT_OFFSET 33878 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ5_RT_OFFSET 33879 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ6_RT_OFFSET 33880 -#define PBF_REG_BTB_GUARANTEED_VOQ6_RT_OFFSET 33881 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ6_RT_OFFSET 33882 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ7_RT_OFFSET 33883 -#define PBF_REG_BTB_GUARANTEED_VOQ7_RT_OFFSET 33884 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ7_RT_OFFSET 33885 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ8_RT_OFFSET 33886 -#define PBF_REG_BTB_GUARANTEED_VOQ8_RT_OFFSET 33887 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ8_RT_OFFSET 33888 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ9_RT_OFFSET 33889 -#define PBF_REG_BTB_GUARANTEED_VOQ9_RT_OFFSET 33890 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ9_RT_OFFSET 33891 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ10_RT_OFFSET 33892 -#define PBF_REG_BTB_GUARANTEED_VOQ10_RT_OFFSET 33893 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ10_RT_OFFSET 33894 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ11_RT_OFFSET 33895 -#define PBF_REG_BTB_GUARANTEED_VOQ11_RT_OFFSET 33896 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ11_RT_OFFSET 33897 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ12_RT_OFFSET 33898 -#define PBF_REG_BTB_GUARANTEED_VOQ12_RT_OFFSET 33899 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ12_RT_OFFSET 33900 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ13_RT_OFFSET 33901 -#define PBF_REG_BTB_GUARANTEED_VOQ13_RT_OFFSET 33902 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ13_RT_OFFSET 33903 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ14_RT_OFFSET 33904 -#define PBF_REG_BTB_GUARANTEED_VOQ14_RT_OFFSET 33905 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ14_RT_OFFSET 33906 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ15_RT_OFFSET 33907 -#define PBF_REG_BTB_GUARANTEED_VOQ15_RT_OFFSET 33908 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ15_RT_OFFSET 33909 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ16_RT_OFFSET 33910 -#define PBF_REG_BTB_GUARANTEED_VOQ16_RT_OFFSET 33911 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ16_RT_OFFSET 33912 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ17_RT_OFFSET 33913 -#define PBF_REG_BTB_GUARANTEED_VOQ17_RT_OFFSET 33914 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ17_RT_OFFSET 33915 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ18_RT_OFFSET 33916 -#define PBF_REG_BTB_GUARANTEED_VOQ18_RT_OFFSET 33917 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ18_RT_OFFSET 33918 -#define PBF_REG_YCMD_QS_NUM_LINES_VOQ19_RT_OFFSET 33919 -#define PBF_REG_BTB_GUARANTEED_VOQ19_RT_OFFSET 33920 -#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ19_RT_OFFSET 33921 -#define XCM_REG_CON_PHY_Q3_RT_OFFSET 33922 - -#define RUNTIME_ARRAY_SIZE 33923 +#define DORQ_REG_PF_MAX_ICID_0_RT_OFFSET 0 +#define DORQ_REG_PF_MAX_ICID_1_RT_OFFSET 1 +#define DORQ_REG_PF_MAX_ICID_2_RT_OFFSET 2 +#define DORQ_REG_PF_MAX_ICID_3_RT_OFFSET 3 +#define DORQ_REG_PF_MAX_ICID_4_RT_OFFSET 4 +#define DORQ_REG_PF_MAX_ICID_5_RT_OFFSET 5 +#define DORQ_REG_PF_MAX_ICID_6_RT_OFFSET 6 +#define DORQ_REG_PF_MAX_ICID_7_RT_OFFSET 7 +#define DORQ_REG_VF_MAX_ICID_0_RT_OFFSET 8 +#define DORQ_REG_VF_MAX_ICID_1_RT_OFFSET 9 +#define DORQ_REG_VF_MAX_ICID_2_RT_OFFSET 10 +#define DORQ_REG_VF_MAX_ICID_3_RT_OFFSET 11 +#define DORQ_REG_VF_MAX_ICID_4_RT_OFFSET 12 +#define DORQ_REG_VF_MAX_ICID_5_RT_OFFSET 13 +#define DORQ_REG_VF_MAX_ICID_6_RT_OFFSET 14 +#define DORQ_REG_VF_MAX_ICID_7_RT_OFFSET 15 +#define DORQ_REG_PF_WAKE_ALL_RT_OFFSET 16 +#define DORQ_REG_TAG1_ETHERTYPE_RT_OFFSET 17 +#define IGU_REG_PF_CONFIGURATION_RT_OFFSET 18 +#define IGU_REG_VF_CONFIGURATION_RT_OFFSET 19 +#define IGU_REG_ATTN_MSG_ADDR_L_RT_OFFSET 20 +#define IGU_REG_ATTN_MSG_ADDR_H_RT_OFFSET 21 +#define IGU_REG_LEADING_EDGE_LATCH_RT_OFFSET 22 +#define IGU_REG_TRAILING_EDGE_LATCH_RT_OFFSET 23 +#define CAU_REG_CQE_AGG_UNIT_SIZE_RT_OFFSET 24 +#define CAU_REG_SB_VAR_MEMORY_RT_OFFSET 761 +#define CAU_REG_SB_VAR_MEMORY_RT_SIZE 736 +#define CAU_REG_SB_VAR_MEMORY_RT_OFFSET 761 +#define CAU_REG_SB_VAR_MEMORY_RT_SIZE 736 +#define CAU_REG_SB_ADDR_MEMORY_RT_OFFSET 1497 +#define CAU_REG_SB_ADDR_MEMORY_RT_SIZE 736 +#define CAU_REG_PI_MEMORY_RT_OFFSET 2233 +#define CAU_REG_PI_MEMORY_RT_SIZE 4416 +#define PRS_REG_SEARCH_RESP_INITIATOR_TYPE_RT_OFFSET 6649 +#define PRS_REG_TASK_ID_MAX_INITIATOR_PF_RT_OFFSET 6650 +#define PRS_REG_TASK_ID_MAX_INITIATOR_VF_RT_OFFSET 6651 +#define PRS_REG_TASK_ID_MAX_TARGET_PF_RT_OFFSET 6652 +#define PRS_REG_TASK_ID_MAX_TARGET_VF_RT_OFFSET 6653 +#define PRS_REG_SEARCH_TCP_RT_OFFSET 6654 +#define PRS_REG_SEARCH_FCOE_RT_OFFSET 6655 +#define PRS_REG_SEARCH_ROCE_RT_OFFSET 6656 +#define PRS_REG_ROCE_DEST_QP_MAX_VF_RT_OFFSET 6657 +#define PRS_REG_ROCE_DEST_QP_MAX_PF_RT_OFFSET 6658 +#define PRS_REG_SEARCH_OPENFLOW_RT_OFFSET 6659 +#define PRS_REG_SEARCH_NON_IP_AS_OPENFLOW_RT_OFFSET 6660 +#define PRS_REG_OPENFLOW_SUPPORT_ONLY_KNOWN_OVER_IP_RT_OFFSET 6661 +#define PRS_REG_OPENFLOW_SEARCH_KEY_MASK_RT_OFFSET 6662 +#define PRS_REG_TAG_ETHERTYPE_0_RT_OFFSET 6663 +#define PRS_REG_LIGHT_L2_ETHERTYPE_EN_RT_OFFSET 6664 +#define SRC_REG_FIRSTFREE_RT_OFFSET 6665 +#define SRC_REG_FIRSTFREE_RT_SIZE 2 +#define SRC_REG_LASTFREE_RT_OFFSET 6667 +#define SRC_REG_LASTFREE_RT_SIZE 2 +#define SRC_REG_COUNTFREE_RT_OFFSET 6669 +#define SRC_REG_NUMBER_HASH_BITS_RT_OFFSET 6670 +#define PSWRQ2_REG_CDUT_P_SIZE_RT_OFFSET 6671 +#define PSWRQ2_REG_CDUC_P_SIZE_RT_OFFSET 6672 +#define PSWRQ2_REG_TM_P_SIZE_RT_OFFSET 6673 +#define PSWRQ2_REG_QM_P_SIZE_RT_OFFSET 6674 +#define PSWRQ2_REG_SRC_P_SIZE_RT_OFFSET 6675 +#define PSWRQ2_REG_TSDM_P_SIZE_RT_OFFSET 6676 +#define PSWRQ2_REG_TM_FIRST_ILT_RT_OFFSET 6677 +#define PSWRQ2_REG_TM_LAST_ILT_RT_OFFSET 6678 +#define PSWRQ2_REG_QM_FIRST_ILT_RT_OFFSET 6679 +#define PSWRQ2_REG_QM_LAST_ILT_RT_OFFSET 6680 +#define PSWRQ2_REG_SRC_FIRST_ILT_RT_OFFSET 6681 +#define PSWRQ2_REG_SRC_LAST_ILT_RT_OFFSET 6682 +#define PSWRQ2_REG_CDUC_FIRST_ILT_RT_OFFSET 6683 +#define PSWRQ2_REG_CDUC_LAST_ILT_RT_OFFSET 6684 +#define PSWRQ2_REG_CDUT_FIRST_ILT_RT_OFFSET 6685 +#define PSWRQ2_REG_CDUT_LAST_ILT_RT_OFFSET 6686 +#define PSWRQ2_REG_TSDM_FIRST_ILT_RT_OFFSET 6687 +#define PSWRQ2_REG_TSDM_LAST_ILT_RT_OFFSET 6688 +#define PSWRQ2_REG_TM_NUMBER_OF_PF_BLOCKS_RT_OFFSET 6689 +#define PSWRQ2_REG_CDUT_NUMBER_OF_PF_BLOCKS_RT_OFFSET 6690 +#define PSWRQ2_REG_CDUC_NUMBER_OF_PF_BLOCKS_RT_OFFSET 6691 +#define PSWRQ2_REG_TM_VF_BLOCKS_RT_OFFSET 6692 +#define PSWRQ2_REG_CDUT_VF_BLOCKS_RT_OFFSET 6693 +#define PSWRQ2_REG_CDUC_VF_BLOCKS_RT_OFFSET 6694 +#define PSWRQ2_REG_TM_BLOCKS_FACTOR_RT_OFFSET 6695 +#define PSWRQ2_REG_CDUT_BLOCKS_FACTOR_RT_OFFSET 6696 +#define PSWRQ2_REG_CDUC_BLOCKS_FACTOR_RT_OFFSET 6697 +#define PSWRQ2_REG_VF_BASE_RT_OFFSET 6698 +#define PSWRQ2_REG_VF_LAST_ILT_RT_OFFSET 6699 +#define PSWRQ2_REG_WR_MBS0_RT_OFFSET 6700 +#define PSWRQ2_REG_RD_MBS0_RT_OFFSET 6701 +#define PSWRQ2_REG_DRAM_ALIGN_WR_RT_OFFSET 6702 +#define PSWRQ2_REG_DRAM_ALIGN_RD_RT_OFFSET 6703 +#define PSWRQ2_REG_ILT_MEMORY_RT_OFFSET 6704 +#define PSWRQ2_REG_ILT_MEMORY_RT_SIZE 22000 +#define PGLUE_REG_B_VF_BASE_RT_OFFSET 28704 +#define PGLUE_REG_B_CACHE_LINE_SIZE_RT_OFFSET 28705 +#define PGLUE_REG_B_PF_BAR0_SIZE_RT_OFFSET 28706 +#define PGLUE_REG_B_PF_BAR1_SIZE_RT_OFFSET 28707 +#define PGLUE_REG_B_VF_BAR1_SIZE_RT_OFFSET 28708 +#define TM_REG_VF_ENABLE_CONN_RT_OFFSET 28709 +#define TM_REG_PF_ENABLE_CONN_RT_OFFSET 28710 +#define TM_REG_PF_ENABLE_TASK_RT_OFFSET 28711 +#define TM_REG_GROUP_SIZE_RESOLUTION_CONN_RT_OFFSET 28712 +#define TM_REG_GROUP_SIZE_RESOLUTION_TASK_RT_OFFSET 28713 +#define TM_REG_CONFIG_CONN_MEM_RT_OFFSET 28714 +#define TM_REG_CONFIG_CONN_MEM_RT_SIZE 416 +#define TM_REG_CONFIG_TASK_MEM_RT_OFFSET 29130 +#define TM_REG_CONFIG_TASK_MEM_RT_SIZE 512 +#define QM_REG_MAXPQSIZE_0_RT_OFFSET 29642 +#define QM_REG_MAXPQSIZE_1_RT_OFFSET 29643 +#define QM_REG_MAXPQSIZE_2_RT_OFFSET 29644 +#define QM_REG_MAXPQSIZETXSEL_0_RT_OFFSET 29645 +#define QM_REG_MAXPQSIZETXSEL_1_RT_OFFSET 29646 +#define QM_REG_MAXPQSIZETXSEL_2_RT_OFFSET 29647 +#define QM_REG_MAXPQSIZETXSEL_3_RT_OFFSET 29648 +#define QM_REG_MAXPQSIZETXSEL_4_RT_OFFSET 29649 +#define QM_REG_MAXPQSIZETXSEL_5_RT_OFFSET 29650 +#define QM_REG_MAXPQSIZETXSEL_6_RT_OFFSET 29651 +#define QM_REG_MAXPQSIZETXSEL_7_RT_OFFSET 29652 +#define QM_REG_MAXPQSIZETXSEL_8_RT_OFFSET 29653 +#define QM_REG_MAXPQSIZETXSEL_9_RT_OFFSET 29654 +#define QM_REG_MAXPQSIZETXSEL_10_RT_OFFSET 29655 +#define QM_REG_MAXPQSIZETXSEL_11_RT_OFFSET 29656 +#define QM_REG_MAXPQSIZETXSEL_12_RT_OFFSET 29657 +#define QM_REG_MAXPQSIZETXSEL_13_RT_OFFSET 29658 +#define QM_REG_MAXPQSIZETXSEL_14_RT_OFFSET 29659 +#define QM_REG_MAXPQSIZETXSEL_15_RT_OFFSET 29660 +#define QM_REG_MAXPQSIZETXSEL_16_RT_OFFSET 29661 +#define QM_REG_MAXPQSIZETXSEL_17_RT_OFFSET 29662 +#define QM_REG_MAXPQSIZETXSEL_18_RT_OFFSET 29663 +#define QM_REG_MAXPQSIZETXSEL_19_RT_OFFSET 29664 +#define QM_REG_MAXPQSIZETXSEL_20_RT_OFFSET 29665 +#define QM_REG_MAXPQSIZETXSEL_21_RT_OFFSET 29666 +#define QM_REG_MAXPQSIZETXSEL_22_RT_OFFSET 29667 +#define QM_REG_MAXPQSIZETXSEL_23_RT_OFFSET 29668 +#define QM_REG_MAXPQSIZETXSEL_24_RT_OFFSET 29669 +#define QM_REG_MAXPQSIZETXSEL_25_RT_OFFSET 29670 +#define QM_REG_MAXPQSIZETXSEL_26_RT_OFFSET 29671 +#define QM_REG_MAXPQSIZETXSEL_27_RT_OFFSET 29672 +#define QM_REG_MAXPQSIZETXSEL_28_RT_OFFSET 29673 +#define QM_REG_MAXPQSIZETXSEL_29_RT_OFFSET 29674 +#define QM_REG_MAXPQSIZETXSEL_30_RT_OFFSET 29675 +#define QM_REG_MAXPQSIZETXSEL_31_RT_OFFSET 29676 +#define QM_REG_MAXPQSIZETXSEL_32_RT_OFFSET 29677 +#define QM_REG_MAXPQSIZETXSEL_33_RT_OFFSET 29678 +#define QM_REG_MAXPQSIZETXSEL_34_RT_OFFSET 29679 +#define QM_REG_MAXPQSIZETXSEL_35_RT_OFFSET 29680 +#define QM_REG_MAXPQSIZETXSEL_36_RT_OFFSET 29681 +#define QM_REG_MAXPQSIZETXSEL_37_RT_OFFSET 29682 +#define QM_REG_MAXPQSIZETXSEL_38_RT_OFFSET 29683 +#define QM_REG_MAXPQSIZETXSEL_39_RT_OFFSET 29684 +#define QM_REG_MAXPQSIZETXSEL_40_RT_OFFSET 29685 +#define QM_REG_MAXPQSIZETXSEL_41_RT_OFFSET 29686 +#define QM_REG_MAXPQSIZETXSEL_42_RT_OFFSET 29687 +#define QM_REG_MAXPQSIZETXSEL_43_RT_OFFSET 29688 +#define QM_REG_MAXPQSIZETXSEL_44_RT_OFFSET 29689 +#define QM_REG_MAXPQSIZETXSEL_45_RT_OFFSET 29690 +#define QM_REG_MAXPQSIZETXSEL_46_RT_OFFSET 29691 +#define QM_REG_MAXPQSIZETXSEL_47_RT_OFFSET 29692 +#define QM_REG_MAXPQSIZETXSEL_48_RT_OFFSET 29693 +#define QM_REG_MAXPQSIZETXSEL_49_RT_OFFSET 29694 +#define QM_REG_MAXPQSIZETXSEL_50_RT_OFFSET 29695 +#define QM_REG_MAXPQSIZETXSEL_51_RT_OFFSET 29696 +#define QM_REG_MAXPQSIZETXSEL_52_RT_OFFSET 29697 +#define QM_REG_MAXPQSIZETXSEL_53_RT_OFFSET 29698 +#define QM_REG_MAXPQSIZETXSEL_54_RT_OFFSET 29699 +#define QM_REG_MAXPQSIZETXSEL_55_RT_OFFSET 29700 +#define QM_REG_MAXPQSIZETXSEL_56_RT_OFFSET 29701 +#define QM_REG_MAXPQSIZETXSEL_57_RT_OFFSET 29702 +#define QM_REG_MAXPQSIZETXSEL_58_RT_OFFSET 29703 +#define QM_REG_MAXPQSIZETXSEL_59_RT_OFFSET 29704 +#define QM_REG_MAXPQSIZETXSEL_60_RT_OFFSET 29705 +#define QM_REG_MAXPQSIZETXSEL_61_RT_OFFSET 29706 +#define QM_REG_MAXPQSIZETXSEL_62_RT_OFFSET 29707 +#define QM_REG_MAXPQSIZETXSEL_63_RT_OFFSET 29708 +#define QM_REG_BASEADDROTHERPQ_RT_OFFSET 29709 +#define QM_REG_BASEADDROTHERPQ_RT_SIZE 128 +#define QM_REG_VOQCRDLINE_RT_OFFSET 29837 +#define QM_REG_VOQCRDLINE_RT_SIZE 20 +#define QM_REG_VOQINITCRDLINE_RT_OFFSET 29857 +#define QM_REG_VOQINITCRDLINE_RT_SIZE 20 +#define QM_REG_AFULLQMBYPTHRPFWFQ_RT_OFFSET 29877 +#define QM_REG_AFULLQMBYPTHRVPWFQ_RT_OFFSET 29878 +#define QM_REG_AFULLQMBYPTHRPFRL_RT_OFFSET 29879 +#define QM_REG_AFULLQMBYPTHRGLBLRL_RT_OFFSET 29880 +#define QM_REG_AFULLOPRTNSTCCRDMASK_RT_OFFSET 29881 +#define QM_REG_WRROTHERPQGRP_0_RT_OFFSET 29882 +#define QM_REG_WRROTHERPQGRP_1_RT_OFFSET 29883 +#define QM_REG_WRROTHERPQGRP_2_RT_OFFSET 29884 +#define QM_REG_WRROTHERPQGRP_3_RT_OFFSET 29885 +#define QM_REG_WRROTHERPQGRP_4_RT_OFFSET 29886 +#define QM_REG_WRROTHERPQGRP_5_RT_OFFSET 29887 +#define QM_REG_WRROTHERPQGRP_6_RT_OFFSET 29888 +#define QM_REG_WRROTHERPQGRP_7_RT_OFFSET 29889 +#define QM_REG_WRROTHERPQGRP_8_RT_OFFSET 29890 +#define QM_REG_WRROTHERPQGRP_9_RT_OFFSET 29891 +#define QM_REG_WRROTHERPQGRP_10_RT_OFFSET 29892 +#define QM_REG_WRROTHERPQGRP_11_RT_OFFSET 29893 +#define QM_REG_WRROTHERPQGRP_12_RT_OFFSET 29894 +#define QM_REG_WRROTHERPQGRP_13_RT_OFFSET 29895 +#define QM_REG_WRROTHERPQGRP_14_RT_OFFSET 29896 +#define QM_REG_WRROTHERPQGRP_15_RT_OFFSET 29897 +#define QM_REG_WRROTHERGRPWEIGHT_0_RT_OFFSET 29898 +#define QM_REG_WRROTHERGRPWEIGHT_1_RT_OFFSET 29899 +#define QM_REG_WRROTHERGRPWEIGHT_2_RT_OFFSET 29900 +#define QM_REG_WRROTHERGRPWEIGHT_3_RT_OFFSET 29901 +#define QM_REG_WRRTXGRPWEIGHT_0_RT_OFFSET 29902 +#define QM_REG_WRRTXGRPWEIGHT_1_RT_OFFSET 29903 +#define QM_REG_PQTX2PF_0_RT_OFFSET 29904 +#define QM_REG_PQTX2PF_1_RT_OFFSET 29905 +#define QM_REG_PQTX2PF_2_RT_OFFSET 29906 +#define QM_REG_PQTX2PF_3_RT_OFFSET 29907 +#define QM_REG_PQTX2PF_4_RT_OFFSET 29908 +#define QM_REG_PQTX2PF_5_RT_OFFSET 29909 +#define QM_REG_PQTX2PF_6_RT_OFFSET 29910 +#define QM_REG_PQTX2PF_7_RT_OFFSET 29911 +#define QM_REG_PQTX2PF_8_RT_OFFSET 29912 +#define QM_REG_PQTX2PF_9_RT_OFFSET 29913 +#define QM_REG_PQTX2PF_10_RT_OFFSET 29914 +#define QM_REG_PQTX2PF_11_RT_OFFSET 29915 +#define QM_REG_PQTX2PF_12_RT_OFFSET 29916 +#define QM_REG_PQTX2PF_13_RT_OFFSET 29917 +#define QM_REG_PQTX2PF_14_RT_OFFSET 29918 +#define QM_REG_PQTX2PF_15_RT_OFFSET 29919 +#define QM_REG_PQTX2PF_16_RT_OFFSET 29920 +#define QM_REG_PQTX2PF_17_RT_OFFSET 29921 +#define QM_REG_PQTX2PF_18_RT_OFFSET 29922 +#define QM_REG_PQTX2PF_19_RT_OFFSET 29923 +#define QM_REG_PQTX2PF_20_RT_OFFSET 29924 +#define QM_REG_PQTX2PF_21_RT_OFFSET 29925 +#define QM_REG_PQTX2PF_22_RT_OFFSET 29926 +#define QM_REG_PQTX2PF_23_RT_OFFSET 29927 +#define QM_REG_PQTX2PF_24_RT_OFFSET 29928 +#define QM_REG_PQTX2PF_25_RT_OFFSET 29929 +#define QM_REG_PQTX2PF_26_RT_OFFSET 29930 +#define QM_REG_PQTX2PF_27_RT_OFFSET 29931 +#define QM_REG_PQTX2PF_28_RT_OFFSET 29932 +#define QM_REG_PQTX2PF_29_RT_OFFSET 29933 +#define QM_REG_PQTX2PF_30_RT_OFFSET 29934 +#define QM_REG_PQTX2PF_31_RT_OFFSET 29935 +#define QM_REG_PQTX2PF_32_RT_OFFSET 29936 +#define QM_REG_PQTX2PF_33_RT_OFFSET 29937 +#define QM_REG_PQTX2PF_34_RT_OFFSET 29938 +#define QM_REG_PQTX2PF_35_RT_OFFSET 29939 +#define QM_REG_PQTX2PF_36_RT_OFFSET 29940 +#define QM_REG_PQTX2PF_37_RT_OFFSET 29941 +#define QM_REG_PQTX2PF_38_RT_OFFSET 29942 +#define QM_REG_PQTX2PF_39_RT_OFFSET 29943 +#define QM_REG_PQTX2PF_40_RT_OFFSET 29944 +#define QM_REG_PQTX2PF_41_RT_OFFSET 29945 +#define QM_REG_PQTX2PF_42_RT_OFFSET 29946 +#define QM_REG_PQTX2PF_43_RT_OFFSET 29947 +#define QM_REG_PQTX2PF_44_RT_OFFSET 29948 +#define QM_REG_PQTX2PF_45_RT_OFFSET 29949 +#define QM_REG_PQTX2PF_46_RT_OFFSET 29950 +#define QM_REG_PQTX2PF_47_RT_OFFSET 29951 +#define QM_REG_PQTX2PF_48_RT_OFFSET 29952 +#define QM_REG_PQTX2PF_49_RT_OFFSET 29953 +#define QM_REG_PQTX2PF_50_RT_OFFSET 29954 +#define QM_REG_PQTX2PF_51_RT_OFFSET 29955 +#define QM_REG_PQTX2PF_52_RT_OFFSET 29956 +#define QM_REG_PQTX2PF_53_RT_OFFSET 29957 +#define QM_REG_PQTX2PF_54_RT_OFFSET 29958 +#define QM_REG_PQTX2PF_55_RT_OFFSET 29959 +#define QM_REG_PQTX2PF_56_RT_OFFSET 29960 +#define QM_REG_PQTX2PF_57_RT_OFFSET 29961 +#define QM_REG_PQTX2PF_58_RT_OFFSET 29962 +#define QM_REG_PQTX2PF_59_RT_OFFSET 29963 +#define QM_REG_PQTX2PF_60_RT_OFFSET 29964 +#define QM_REG_PQTX2PF_61_RT_OFFSET 29965 +#define QM_REG_PQTX2PF_62_RT_OFFSET 29966 +#define QM_REG_PQTX2PF_63_RT_OFFSET 29967 +#define QM_REG_PQOTHER2PF_0_RT_OFFSET 29968 +#define QM_REG_PQOTHER2PF_1_RT_OFFSET 29969 +#define QM_REG_PQOTHER2PF_2_RT_OFFSET 29970 +#define QM_REG_PQOTHER2PF_3_RT_OFFSET 29971 +#define QM_REG_PQOTHER2PF_4_RT_OFFSET 29972 +#define QM_REG_PQOTHER2PF_5_RT_OFFSET 29973 +#define QM_REG_PQOTHER2PF_6_RT_OFFSET 29974 +#define QM_REG_PQOTHER2PF_7_RT_OFFSET 29975 +#define QM_REG_PQOTHER2PF_8_RT_OFFSET 29976 +#define QM_REG_PQOTHER2PF_9_RT_OFFSET 29977 +#define QM_REG_PQOTHER2PF_10_RT_OFFSET 29978 +#define QM_REG_PQOTHER2PF_11_RT_OFFSET 29979 +#define QM_REG_PQOTHER2PF_12_RT_OFFSET 29980 +#define QM_REG_PQOTHER2PF_13_RT_OFFSET 29981 +#define QM_REG_PQOTHER2PF_14_RT_OFFSET 29982 +#define QM_REG_PQOTHER2PF_15_RT_OFFSET 29983 +#define QM_REG_RLGLBLPERIOD_0_RT_OFFSET 29984 +#define QM_REG_RLGLBLPERIOD_1_RT_OFFSET 29985 +#define QM_REG_RLGLBLPERIODTIMER_0_RT_OFFSET 29986 +#define QM_REG_RLGLBLPERIODTIMER_1_RT_OFFSET 29987 +#define QM_REG_RLGLBLPERIODSEL_0_RT_OFFSET 29988 +#define QM_REG_RLGLBLPERIODSEL_1_RT_OFFSET 29989 +#define QM_REG_RLGLBLPERIODSEL_2_RT_OFFSET 29990 +#define QM_REG_RLGLBLPERIODSEL_3_RT_OFFSET 29991 +#define QM_REG_RLGLBLPERIODSEL_4_RT_OFFSET 29992 +#define QM_REG_RLGLBLPERIODSEL_5_RT_OFFSET 29993 +#define QM_REG_RLGLBLPERIODSEL_6_RT_OFFSET 29994 +#define QM_REG_RLGLBLPERIODSEL_7_RT_OFFSET 29995 +#define QM_REG_RLGLBLINCVAL_RT_OFFSET 29996 +#define QM_REG_RLGLBLINCVAL_RT_SIZE 256 +#define QM_REG_RLGLBLUPPERBOUND_RT_OFFSET 30252 +#define QM_REG_RLGLBLUPPERBOUND_RT_SIZE 256 +#define QM_REG_RLGLBLCRD_RT_OFFSET 30508 +#define QM_REG_RLGLBLCRD_RT_SIZE 256 +#define QM_REG_RLGLBLENABLE_RT_OFFSET 30764 +#define QM_REG_RLPFPERIOD_RT_OFFSET 30765 +#define QM_REG_RLPFPERIODTIMER_RT_OFFSET 30766 +#define QM_REG_RLPFINCVAL_RT_OFFSET 30767 +#define QM_REG_RLPFINCVAL_RT_SIZE 16 +#define QM_REG_RLPFUPPERBOUND_RT_OFFSET 30783 +#define QM_REG_RLPFUPPERBOUND_RT_SIZE 16 +#define QM_REG_RLPFCRD_RT_OFFSET 30799 +#define QM_REG_RLPFCRD_RT_SIZE 16 +#define QM_REG_RLPFENABLE_RT_OFFSET 30815 +#define QM_REG_RLPFVOQENABLE_RT_OFFSET 30816 +#define QM_REG_WFQPFWEIGHT_RT_OFFSET 30817 +#define QM_REG_WFQPFWEIGHT_RT_SIZE 16 +#define QM_REG_WFQPFUPPERBOUND_RT_OFFSET 30833 +#define QM_REG_WFQPFUPPERBOUND_RT_SIZE 16 +#define QM_REG_WFQPFCRD_RT_OFFSET 30849 +#define QM_REG_WFQPFCRD_RT_SIZE 160 +#define QM_REG_WFQPFENABLE_RT_OFFSET 31009 +#define QM_REG_WFQVPENABLE_RT_OFFSET 31010 +#define QM_REG_BASEADDRTXPQ_RT_OFFSET 31011 +#define QM_REG_BASEADDRTXPQ_RT_SIZE 512 +#define QM_REG_TXPQMAP_RT_OFFSET 31523 +#define QM_REG_TXPQMAP_RT_SIZE 512 +#define QM_REG_WFQVPWEIGHT_RT_OFFSET 32035 +#define QM_REG_WFQVPWEIGHT_RT_SIZE 512 +#define QM_REG_WFQVPCRD_RT_OFFSET 32547 +#define QM_REG_WFQVPCRD_RT_SIZE 512 +#define QM_REG_WFQVPMAP_RT_OFFSET 33059 +#define QM_REG_WFQVPMAP_RT_SIZE 512 +#define QM_REG_WFQPFCRD_MSB_RT_OFFSET 33571 +#define QM_REG_WFQPFCRD_MSB_RT_SIZE 160 +#define NIG_REG_TAG_ETHERTYPE_0_RT_OFFSET 33731 +#define NIG_REG_OUTER_TAG_VALUE_LIST0_RT_OFFSET 33732 +#define NIG_REG_OUTER_TAG_VALUE_LIST1_RT_OFFSET 33733 +#define NIG_REG_OUTER_TAG_VALUE_LIST2_RT_OFFSET 33734 +#define NIG_REG_OUTER_TAG_VALUE_LIST3_RT_OFFSET 33735 +#define NIG_REG_OUTER_TAG_VALUE_MASK_RT_OFFSET 33736 +#define NIG_REG_LLH_FUNC_TAGMAC_CLS_TYPE_RT_OFFSET 33737 +#define NIG_REG_LLH_FUNC_TAG_EN_RT_OFFSET 33738 +#define NIG_REG_LLH_FUNC_TAG_EN_RT_SIZE 4 +#define NIG_REG_LLH_FUNC_TAG_HDR_SEL_RT_OFFSET 33742 +#define NIG_REG_LLH_FUNC_TAG_HDR_SEL_RT_SIZE 4 +#define NIG_REG_LLH_FUNC_TAG_VALUE_RT_OFFSET 33746 +#define NIG_REG_LLH_FUNC_TAG_VALUE_RT_SIZE 4 +#define NIG_REG_LLH_FUNC_NO_TAG_RT_OFFSET 33750 +#define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_OFFSET 33751 +#define NIG_REG_LLH_FUNC_FILTER_VALUE_RT_SIZE 32 +#define NIG_REG_LLH_FUNC_FILTER_EN_RT_OFFSET 33783 +#define NIG_REG_LLH_FUNC_FILTER_EN_RT_SIZE 16 +#define NIG_REG_LLH_FUNC_FILTER_MODE_RT_OFFSET 33799 +#define NIG_REG_LLH_FUNC_FILTER_MODE_RT_SIZE 16 +#define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_OFFSET 33815 +#define NIG_REG_LLH_FUNC_FILTER_PROTOCOL_TYPE_RT_SIZE 16 +#define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_OFFSET 33831 +#define NIG_REG_LLH_FUNC_FILTER_HDR_SEL_RT_SIZE 16 +#define NIG_REG_TX_EDPM_CTRL_RT_OFFSET 33847 +#define NIG_REG_ROCE_DUPLICATE_TO_HOST_RT_OFFSET 33848 +#define CDU_REG_CID_ADDR_PARAMS_RT_OFFSET 33849 +#define CDU_REG_SEGMENT0_PARAMS_RT_OFFSET 33850 +#define CDU_REG_SEGMENT1_PARAMS_RT_OFFSET 33851 +#define CDU_REG_PF_SEG0_TYPE_OFFSET_RT_OFFSET 33852 +#define CDU_REG_PF_SEG1_TYPE_OFFSET_RT_OFFSET 33853 +#define CDU_REG_PF_SEG2_TYPE_OFFSET_RT_OFFSET 33854 +#define CDU_REG_PF_SEG3_TYPE_OFFSET_RT_OFFSET 33855 +#define CDU_REG_PF_FL_SEG0_TYPE_OFFSET_RT_OFFSET 33856 +#define CDU_REG_PF_FL_SEG1_TYPE_OFFSET_RT_OFFSET 33857 +#define CDU_REG_PF_FL_SEG2_TYPE_OFFSET_RT_OFFSET 33858 +#define CDU_REG_PF_FL_SEG3_TYPE_OFFSET_RT_OFFSET 33859 +#define CDU_REG_VF_SEG_TYPE_OFFSET_RT_OFFSET 33860 +#define CDU_REG_VF_FL_SEG_TYPE_OFFSET_RT_OFFSET 33861 +#define PBF_REG_TAG_ETHERTYPE_0_RT_OFFSET 33862 +#define PBF_REG_BTB_SHARED_AREA_SIZE_RT_OFFSET 33863 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ0_RT_OFFSET 33864 +#define PBF_REG_BTB_GUARANTEED_VOQ0_RT_OFFSET 33865 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ0_RT_OFFSET 33866 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ1_RT_OFFSET 33867 +#define PBF_REG_BTB_GUARANTEED_VOQ1_RT_OFFSET 33868 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ1_RT_OFFSET 33869 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ2_RT_OFFSET 33870 +#define PBF_REG_BTB_GUARANTEED_VOQ2_RT_OFFSET 33871 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ2_RT_OFFSET 33872 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ3_RT_OFFSET 33873 +#define PBF_REG_BTB_GUARANTEED_VOQ3_RT_OFFSET 33874 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ3_RT_OFFSET 33875 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ4_RT_OFFSET 33876 +#define PBF_REG_BTB_GUARANTEED_VOQ4_RT_OFFSET 33877 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ4_RT_OFFSET 33878 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ5_RT_OFFSET 33879 +#define PBF_REG_BTB_GUARANTEED_VOQ5_RT_OFFSET 33880 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ5_RT_OFFSET 33881 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ6_RT_OFFSET 33882 +#define PBF_REG_BTB_GUARANTEED_VOQ6_RT_OFFSET 33883 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ6_RT_OFFSET 33884 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ7_RT_OFFSET 33885 +#define PBF_REG_BTB_GUARANTEED_VOQ7_RT_OFFSET 33886 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ7_RT_OFFSET 33887 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ8_RT_OFFSET 33888 +#define PBF_REG_BTB_GUARANTEED_VOQ8_RT_OFFSET 33889 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ8_RT_OFFSET 33890 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ9_RT_OFFSET 33891 +#define PBF_REG_BTB_GUARANTEED_VOQ9_RT_OFFSET 33892 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ9_RT_OFFSET 33893 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ10_RT_OFFSET 33894 +#define PBF_REG_BTB_GUARANTEED_VOQ10_RT_OFFSET 33895 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ10_RT_OFFSET 33896 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ11_RT_OFFSET 33897 +#define PBF_REG_BTB_GUARANTEED_VOQ11_RT_OFFSET 33898 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ11_RT_OFFSET 33899 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ12_RT_OFFSET 33900 +#define PBF_REG_BTB_GUARANTEED_VOQ12_RT_OFFSET 33901 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ12_RT_OFFSET 33902 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ13_RT_OFFSET 33903 +#define PBF_REG_BTB_GUARANTEED_VOQ13_RT_OFFSET 33904 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ13_RT_OFFSET 33905 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ14_RT_OFFSET 33906 +#define PBF_REG_BTB_GUARANTEED_VOQ14_RT_OFFSET 33907 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ14_RT_OFFSET 33908 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ15_RT_OFFSET 33909 +#define PBF_REG_BTB_GUARANTEED_VOQ15_RT_OFFSET 33910 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ15_RT_OFFSET 33911 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ16_RT_OFFSET 33912 +#define PBF_REG_BTB_GUARANTEED_VOQ16_RT_OFFSET 33913 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ16_RT_OFFSET 33914 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ17_RT_OFFSET 33915 +#define PBF_REG_BTB_GUARANTEED_VOQ17_RT_OFFSET 33916 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ17_RT_OFFSET 33917 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ18_RT_OFFSET 33918 +#define PBF_REG_BTB_GUARANTEED_VOQ18_RT_OFFSET 33919 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ18_RT_OFFSET 33920 +#define PBF_REG_YCMD_QS_NUM_LINES_VOQ19_RT_OFFSET 33921 +#define PBF_REG_BTB_GUARANTEED_VOQ19_RT_OFFSET 33922 +#define PBF_REG_BTB_SHARED_AREA_SETUP_VOQ19_RT_OFFSET 33923 +#define XCM_REG_CON_PHY_Q3_RT_OFFSET 33924 + +#define RUNTIME_ARRAY_SIZE 33925 /* The eth storm context for the Tstorm */ struct tstorm_eth_conn_st_ctx { @@ -2380,266 +2687,266 @@ struct xstorm_eth_conn_st_ctx { }; struct xstorm_eth_conn_ag_ctx { - u8 reserved0 /* cdu_validation */; - u8 eth_state /* state */; - u8 flags0; -#define XSTORM_ETH_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 -#define XSTORM_ETH_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED1_MASK 0x1 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED1_SHIFT 1 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED2_MASK 0x1 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED2_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 -#define XSTORM_ETH_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED3_MASK 0x1 /* bit4 */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED3_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED4_MASK 0x1 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED4_SHIFT 5 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED5_MASK 0x1 /* bit6 */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED5_SHIFT 6 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED6_MASK 0x1 /* bit7 */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED6_SHIFT 7 - u8 flags1; -#define XSTORM_ETH_CONN_AG_CTX_RESERVED7_MASK 0x1 /* bit8 */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED7_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED8_MASK 0x1 /* bit9 */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED8_SHIFT 1 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED9_MASK 0x1 /* bit10 */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED9_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_BIT11_MASK 0x1 /* bit11 */ -#define XSTORM_ETH_CONN_AG_CTX_BIT11_SHIFT 3 -#define XSTORM_ETH_CONN_AG_CTX_BIT12_MASK 0x1 /* bit12 */ -#define XSTORM_ETH_CONN_AG_CTX_BIT12_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_BIT13_MASK 0x1 /* bit13 */ -#define XSTORM_ETH_CONN_AG_CTX_BIT13_SHIFT 5 -#define XSTORM_ETH_CONN_AG_CTX_TX_RULE_ACTIVE_MASK 0x1 /* bit14 */ -#define XSTORM_ETH_CONN_AG_CTX_TX_RULE_ACTIVE_SHIFT 6 -#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_ACTIVE_MASK 0x1 /* bit15 */ -#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_ACTIVE_SHIFT 7 + u8 reserved0; + u8 eth_state; + u8 flags0; +#define XSTORM_ETH_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED1_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED1_SHIFT 1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED2_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED2_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED3_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED3_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED4_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED4_SHIFT 5 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED5_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED5_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED6_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED6_SHIFT 7 + u8 flags1; +#define XSTORM_ETH_CONN_AG_CTX_RESERVED7_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED7_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED8_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED8_SHIFT 1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED9_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED9_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_BIT11_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_BIT11_SHIFT 3 +#define XSTORM_ETH_CONN_AG_CTX_BIT12_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_BIT12_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_BIT13_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_BIT13_SHIFT 5 +#define XSTORM_ETH_CONN_AG_CTX_TX_RULE_ACTIVE_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_TX_RULE_ACTIVE_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_ACTIVE_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_ACTIVE_SHIFT 7 u8 flags2; -#define XSTORM_ETH_CONN_AG_CTX_CF0_MASK 0x3 /* timer0cf */ -#define XSTORM_ETH_CONN_AG_CTX_CF0_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_CF1_MASK 0x3 /* timer1cf */ -#define XSTORM_ETH_CONN_AG_CTX_CF1_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_CF2_MASK 0x3 /* timer2cf */ -#define XSTORM_ETH_CONN_AG_CTX_CF2_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_CF3_MASK 0x3 -#define XSTORM_ETH_CONN_AG_CTX_CF3_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_CF0_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF0_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_CF1_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF1_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_CF2_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF2_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_CF3_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF3_SHIFT 6 u8 flags3; -#define XSTORM_ETH_CONN_AG_CTX_CF4_MASK 0x3 /* cf4 */ -#define XSTORM_ETH_CONN_AG_CTX_CF4_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_CF5_MASK 0x3 /* cf5 */ -#define XSTORM_ETH_CONN_AG_CTX_CF5_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_CF6_MASK 0x3 /* cf6 */ -#define XSTORM_ETH_CONN_AG_CTX_CF6_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_CF7_MASK 0x3 /* cf7 */ -#define XSTORM_ETH_CONN_AG_CTX_CF7_SHIFT 6 - u8 flags4; -#define XSTORM_ETH_CONN_AG_CTX_CF8_MASK 0x3 /* cf8 */ -#define XSTORM_ETH_CONN_AG_CTX_CF8_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_CF9_MASK 0x3 /* cf9 */ -#define XSTORM_ETH_CONN_AG_CTX_CF9_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_CF10_MASK 0x3 /* cf10 */ -#define XSTORM_ETH_CONN_AG_CTX_CF10_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_CF11_MASK 0x3 /* cf11 */ -#define XSTORM_ETH_CONN_AG_CTX_CF11_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_CF4_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF4_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_CF5_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF5_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_CF6_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF6_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_CF7_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF7_SHIFT 6 + u8 flags4; +#define XSTORM_ETH_CONN_AG_CTX_CF8_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF8_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_CF9_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF9_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_CF10_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF10_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_CF11_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF11_SHIFT 6 u8 flags5; -#define XSTORM_ETH_CONN_AG_CTX_CF12_MASK 0x3 /* cf12 */ -#define XSTORM_ETH_CONN_AG_CTX_CF12_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_CF13_MASK 0x3 /* cf13 */ -#define XSTORM_ETH_CONN_AG_CTX_CF13_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_CF14_MASK 0x3 /* cf14 */ -#define XSTORM_ETH_CONN_AG_CTX_CF14_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_CF15_MASK 0x3 /* cf15 */ -#define XSTORM_ETH_CONN_AG_CTX_CF15_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_CF12_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF12_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_CF13_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF13_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_CF14_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF14_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_CF15_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_CF15_SHIFT 6 u8 flags6; -#define XSTORM_ETH_CONN_AG_CTX_GO_TO_BD_CONS_CF_MASK 0x3 /* cf16 */ -#define XSTORM_ETH_CONN_AG_CTX_GO_TO_BD_CONS_CF_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_MULTI_UNICAST_CF_MASK 0x3 -#define XSTORM_ETH_CONN_AG_CTX_MULTI_UNICAST_CF_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_MASK 0x3 /* cf18 */ -#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_TERMINATE_CF_MASK 0x3 /* cf19 */ -#define XSTORM_ETH_CONN_AG_CTX_TERMINATE_CF_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_GO_TO_BD_CONS_CF_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_GO_TO_BD_CONS_CF_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_MULTI_UNICAST_CF_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_MULTI_UNICAST_CF_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_TERMINATE_CF_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_TERMINATE_CF_SHIFT 6 u8 flags7; -#define XSTORM_ETH_CONN_AG_CTX_FLUSH_Q0_MASK 0x3 /* cf20 */ -#define XSTORM_ETH_CONN_AG_CTX_FLUSH_Q0_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED10_MASK 0x3 /* cf21 */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED10_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_SLOW_PATH_MASK 0x3 /* cf22 */ -#define XSTORM_ETH_CONN_AG_CTX_SLOW_PATH_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_CF0EN_MASK 0x1 /* cf0en */ -#define XSTORM_ETH_CONN_AG_CTX_CF0EN_SHIFT 6 -#define XSTORM_ETH_CONN_AG_CTX_CF1EN_MASK 0x1 /* cf1en */ -#define XSTORM_ETH_CONN_AG_CTX_CF1EN_SHIFT 7 +#define XSTORM_ETH_CONN_AG_CTX_FLUSH_Q0_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_FLUSH_Q0_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED10_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED10_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_SLOW_PATH_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_SLOW_PATH_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_CF0EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF0EN_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_CF1EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF1EN_SHIFT 7 u8 flags8; -#define XSTORM_ETH_CONN_AG_CTX_CF2EN_MASK 0x1 /* cf2en */ -#define XSTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_CF3EN_MASK 0x1 /* cf3en */ -#define XSTORM_ETH_CONN_AG_CTX_CF3EN_SHIFT 1 -#define XSTORM_ETH_CONN_AG_CTX_CF4EN_MASK 0x1 /* cf4en */ -#define XSTORM_ETH_CONN_AG_CTX_CF4EN_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_CF5EN_MASK 0x1 /* cf5en */ -#define XSTORM_ETH_CONN_AG_CTX_CF5EN_SHIFT 3 -#define XSTORM_ETH_CONN_AG_CTX_CF6EN_MASK 0x1 /* cf6en */ -#define XSTORM_ETH_CONN_AG_CTX_CF6EN_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_CF7EN_MASK 0x1 /* cf7en */ -#define XSTORM_ETH_CONN_AG_CTX_CF7EN_SHIFT 5 -#define XSTORM_ETH_CONN_AG_CTX_CF8EN_MASK 0x1 /* cf8en */ -#define XSTORM_ETH_CONN_AG_CTX_CF8EN_SHIFT 6 -#define XSTORM_ETH_CONN_AG_CTX_CF9EN_MASK 0x1 /* cf9en */ -#define XSTORM_ETH_CONN_AG_CTX_CF9EN_SHIFT 7 +#define XSTORM_ETH_CONN_AG_CTX_CF2EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_CF3EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF3EN_SHIFT 1 +#define XSTORM_ETH_CONN_AG_CTX_CF4EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF4EN_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_CF5EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF5EN_SHIFT 3 +#define XSTORM_ETH_CONN_AG_CTX_CF6EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF6EN_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_CF7EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF7EN_SHIFT 5 +#define XSTORM_ETH_CONN_AG_CTX_CF8EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF8EN_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_CF9EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF9EN_SHIFT 7 u8 flags9; -#define XSTORM_ETH_CONN_AG_CTX_CF10EN_MASK 0x1 /* cf10en */ -#define XSTORM_ETH_CONN_AG_CTX_CF10EN_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_CF11EN_MASK 0x1 /* cf11en */ -#define XSTORM_ETH_CONN_AG_CTX_CF11EN_SHIFT 1 -#define XSTORM_ETH_CONN_AG_CTX_CF12EN_MASK 0x1 /* cf12en */ -#define XSTORM_ETH_CONN_AG_CTX_CF12EN_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_CF13EN_MASK 0x1 /* cf13en */ -#define XSTORM_ETH_CONN_AG_CTX_CF13EN_SHIFT 3 -#define XSTORM_ETH_CONN_AG_CTX_CF14EN_MASK 0x1 /* cf14en */ -#define XSTORM_ETH_CONN_AG_CTX_CF14EN_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_CF15EN_MASK 0x1 /* cf15en */ -#define XSTORM_ETH_CONN_AG_CTX_CF15EN_SHIFT 5 -#define XSTORM_ETH_CONN_AG_CTX_GO_TO_BD_CONS_CF_EN_MASK 0x1 /* cf16en */ -#define XSTORM_ETH_CONN_AG_CTX_GO_TO_BD_CONS_CF_EN_SHIFT 6 -#define XSTORM_ETH_CONN_AG_CTX_MULTI_UNICAST_CF_EN_MASK 0x1 -#define XSTORM_ETH_CONN_AG_CTX_MULTI_UNICAST_CF_EN_SHIFT 7 +#define XSTORM_ETH_CONN_AG_CTX_CF10EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF10EN_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_CF11EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF11EN_SHIFT 1 +#define XSTORM_ETH_CONN_AG_CTX_CF12EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF12EN_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_CF13EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF13EN_SHIFT 3 +#define XSTORM_ETH_CONN_AG_CTX_CF14EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF14EN_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_CF15EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_CF15EN_SHIFT 5 +#define XSTORM_ETH_CONN_AG_CTX_GO_TO_BD_CONS_CF_EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_GO_TO_BD_CONS_CF_EN_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_MULTI_UNICAST_CF_EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_MULTI_UNICAST_CF_EN_SHIFT 7 u8 flags10; -#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_EN_MASK 0x1 /* cf18en */ -#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_EN_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_TERMINATE_CF_EN_MASK 0x1 /* cf19en */ -#define XSTORM_ETH_CONN_AG_CTX_TERMINATE_CF_EN_SHIFT 1 -#define XSTORM_ETH_CONN_AG_CTX_FLUSH_Q0_EN_MASK 0x1 /* cf20en */ -#define XSTORM_ETH_CONN_AG_CTX_FLUSH_Q0_EN_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED11_MASK 0x1 /* cf21en */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED11_SHIFT 3 -#define XSTORM_ETH_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 /* cf22en */ -#define XSTORM_ETH_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_TPH_ENABLE_EN_RESERVED_MASK 0x1 /* cf23en */ +#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_DQ_CF_EN_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_TERMINATE_CF_EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_TERMINATE_CF_EN_SHIFT 1 +#define XSTORM_ETH_CONN_AG_CTX_FLUSH_Q0_EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_FLUSH_Q0_EN_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED11_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED11_SHIFT 3 +#define XSTORM_ETH_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_TPH_ENABLE_EN_RESERVED_MASK 0x1 #define XSTORM_ETH_CONN_AG_CTX_TPH_ENABLE_EN_RESERVED_SHIFT 5 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED12_MASK 0x1 /* rule0en */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED12_SHIFT 6 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED13_MASK 0x1 /* rule1en */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED13_SHIFT 7 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED12_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED12_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED13_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED13_SHIFT 7 u8 flags11; -#define XSTORM_ETH_CONN_AG_CTX_RESERVED14_MASK 0x1 /* rule2en */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED14_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_RESERVED15_MASK 0x1 /* rule3en */ -#define XSTORM_ETH_CONN_AG_CTX_RESERVED15_SHIFT 1 -#define XSTORM_ETH_CONN_AG_CTX_TX_DEC_RULE_EN_MASK 0x1 /* rule4en */ -#define XSTORM_ETH_CONN_AG_CTX_TX_DEC_RULE_EN_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_RULE5EN_MASK 0x1 /* rule5en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE5EN_SHIFT 3 -#define XSTORM_ETH_CONN_AG_CTX_RULE6EN_MASK 0x1 /* rule6en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE6EN_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_RULE7EN_MASK 0x1 /* rule7en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE7EN_SHIFT 5 -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 /* rule8en */ -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 -#define XSTORM_ETH_CONN_AG_CTX_RULE9EN_MASK 0x1 /* rule9en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE9EN_SHIFT 7 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED14_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED14_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED15_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RESERVED15_SHIFT 1 +#define XSTORM_ETH_CONN_AG_CTX_TX_DEC_RULE_EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_TX_DEC_RULE_EN_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE5EN_SHIFT 3 +#define XSTORM_ETH_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE6EN_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE7EN_SHIFT 5 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_RULE9EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE9EN_SHIFT 7 u8 flags12; -#define XSTORM_ETH_CONN_AG_CTX_RULE10EN_MASK 0x1 /* rule10en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE10EN_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_RULE11EN_MASK 0x1 /* rule11en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE11EN_SHIFT 1 -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 /* rule12en */ -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 /* rule13en */ -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 -#define XSTORM_ETH_CONN_AG_CTX_RULE14EN_MASK 0x1 /* rule14en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE14EN_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_RULE15EN_MASK 0x1 /* rule15en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE15EN_SHIFT 5 -#define XSTORM_ETH_CONN_AG_CTX_RULE16EN_MASK 0x1 /* rule16en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE16EN_SHIFT 6 -#define XSTORM_ETH_CONN_AG_CTX_RULE17EN_MASK 0x1 /* rule17en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE17EN_SHIFT 7 +#define XSTORM_ETH_CONN_AG_CTX_RULE10EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE10EN_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_RULE11EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE11EN_SHIFT 1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 +#define XSTORM_ETH_CONN_AG_CTX_RULE14EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE14EN_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_RULE15EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE15EN_SHIFT 5 +#define XSTORM_ETH_CONN_AG_CTX_RULE16EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE16EN_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_RULE17EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE17EN_SHIFT 7 u8 flags13; -#define XSTORM_ETH_CONN_AG_CTX_RULE18EN_MASK 0x1 /* rule18en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE18EN_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_RULE19EN_MASK 0x1 /* rule19en */ -#define XSTORM_ETH_CONN_AG_CTX_RULE19EN_SHIFT 1 -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 /* rule20en */ -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 /* rule21en */ -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 /* rule22en */ -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 /* rule23en */ -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 /* rule24en */ -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 /* rule25en */ -#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 +#define XSTORM_ETH_CONN_AG_CTX_RULE18EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE18EN_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_RULE19EN_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_RULE19EN_SHIFT 1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 u8 flags14; -#define XSTORM_ETH_CONN_AG_CTX_EDPM_USE_EXT_HDR_MASK 0x1 /* bit16 */ -#define XSTORM_ETH_CONN_AG_CTX_EDPM_USE_EXT_HDR_SHIFT 0 -#define XSTORM_ETH_CONN_AG_CTX_EDPM_SEND_RAW_L3L4_MASK 0x1 /* bit17 */ -#define XSTORM_ETH_CONN_AG_CTX_EDPM_SEND_RAW_L3L4_SHIFT 1 -#define XSTORM_ETH_CONN_AG_CTX_EDPM_INBAND_PROP_HDR_MASK 0x1 /* bit18 */ -#define XSTORM_ETH_CONN_AG_CTX_EDPM_INBAND_PROP_HDR_SHIFT 2 -#define XSTORM_ETH_CONN_AG_CTX_EDPM_SEND_EXT_TUNNEL_MASK 0x1 /* bit19 */ -#define XSTORM_ETH_CONN_AG_CTX_EDPM_SEND_EXT_TUNNEL_SHIFT 3 -#define XSTORM_ETH_CONN_AG_CTX_L2_EDPM_ENABLE_MASK 0x1 /* bit20 */ -#define XSTORM_ETH_CONN_AG_CTX_L2_EDPM_ENABLE_SHIFT 4 -#define XSTORM_ETH_CONN_AG_CTX_ROCE_EDPM_ENABLE_MASK 0x1 /* bit21 */ -#define XSTORM_ETH_CONN_AG_CTX_ROCE_EDPM_ENABLE_SHIFT 5 -#define XSTORM_ETH_CONN_AG_CTX_TPH_ENABLE_MASK 0x3 /* cf23 */ -#define XSTORM_ETH_CONN_AG_CTX_TPH_ENABLE_SHIFT 6 - u8 edpm_event_id /* byte2 */; - __le16 physical_q0 /* physical_q0 */; - __le16 word1 /* physical_q1 */; - __le16 edpm_num_bds /* physical_q2 */; - __le16 tx_bd_cons /* word3 */; - __le16 tx_bd_prod /* word4 */; - __le16 go_to_bd_cons /* word5 */; - __le16 conn_dpi /* conn_dpi */; - u8 byte3 /* byte3 */; - u8 byte4 /* byte4 */; - u8 byte5 /* byte5 */; - u8 byte6 /* byte6 */; - __le32 reg0 /* reg0 */; - __le32 reg1 /* reg1 */; - __le32 reg2 /* reg2 */; - __le32 reg3 /* reg3 */; - __le32 reg4 /* reg4 */; - __le32 reg5 /* cf_array0 */; - __le32 reg6 /* cf_array1 */; - __le16 word7 /* word7 */; - __le16 word8 /* word8 */; - __le16 word9 /* word9 */; - __le16 word10 /* word10 */; - __le32 reg7 /* reg7 */; - __le32 reg8 /* reg8 */; - __le32 reg9 /* reg9 */; - u8 byte7 /* byte7 */; - u8 byte8 /* byte8 */; - u8 byte9 /* byte9 */; - u8 byte10 /* byte10 */; - u8 byte11 /* byte11 */; - u8 byte12 /* byte12 */; - u8 byte13 /* byte13 */; - u8 byte14 /* byte14 */; - u8 byte15 /* byte15 */; - u8 byte16 /* byte16 */; - __le16 word11 /* word11 */; - __le32 reg10 /* reg10 */; - __le32 reg11 /* reg11 */; - __le32 reg12 /* reg12 */; - __le32 reg13 /* reg13 */; - __le32 reg14 /* reg14 */; - __le32 reg15 /* reg15 */; - __le32 reg16 /* reg16 */; - __le32 reg17 /* reg17 */; - __le32 reg18 /* reg18 */; - __le32 reg19 /* reg19 */; - __le16 word12 /* word12 */; - __le16 word13 /* word13 */; - __le16 word14 /* word14 */; - __le16 word15 /* word15 */; +#define XSTORM_ETH_CONN_AG_CTX_EDPM_USE_EXT_HDR_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_EDPM_USE_EXT_HDR_SHIFT 0 +#define XSTORM_ETH_CONN_AG_CTX_EDPM_SEND_RAW_L3L4_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_EDPM_SEND_RAW_L3L4_SHIFT 1 +#define XSTORM_ETH_CONN_AG_CTX_EDPM_INBAND_PROP_HDR_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_EDPM_INBAND_PROP_HDR_SHIFT 2 +#define XSTORM_ETH_CONN_AG_CTX_EDPM_SEND_EXT_TUNNEL_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_EDPM_SEND_EXT_TUNNEL_SHIFT 3 +#define XSTORM_ETH_CONN_AG_CTX_L2_EDPM_ENABLE_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_L2_EDPM_ENABLE_SHIFT 4 +#define XSTORM_ETH_CONN_AG_CTX_ROCE_EDPM_ENABLE_MASK 0x1 +#define XSTORM_ETH_CONN_AG_CTX_ROCE_EDPM_ENABLE_SHIFT 5 +#define XSTORM_ETH_CONN_AG_CTX_TPH_ENABLE_MASK 0x3 +#define XSTORM_ETH_CONN_AG_CTX_TPH_ENABLE_SHIFT 6 + u8 edpm_event_id; + __le16 physical_q0; + __le16 quota; + __le16 edpm_num_bds; + __le16 tx_bd_cons; + __le16 tx_bd_prod; + __le16 tx_class; + __le16 conn_dpi; + u8 byte3; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le16 word7; + __le16 word8; + __le16 word9; + __le16 word10; + __le32 reg7; + __le32 reg8; + __le32 reg9; + u8 byte7; + u8 byte8; + u8 byte9; + u8 byte10; + u8 byte11; + u8 byte12; + u8 byte13; + u8 byte14; + u8 byte15; + u8 byte16; + __le16 word11; + __le32 reg10; + __le32 reg11; + __le32 reg12; + __le32 reg13; + __le32 reg14; + __le32 reg15; + __le32 reg16; + __le32 reg17; + __le32 reg18; + __le32 reg19; + __le16 word12; + __le16 word13; + __le16 word14; + __le16 word15; }; /* The eth storm context for the Ystorm */ @@ -2648,220 +2955,220 @@ struct ystorm_eth_conn_st_ctx { }; struct ystorm_eth_conn_ag_ctx { - u8 byte0 /* cdu_validation */; - u8 byte1 /* state */; - u8 flags0; -#define YSTORM_ETH_CONN_AG_CTX_BIT0_MASK 0x1 /* exist_in_qm0 */ -#define YSTORM_ETH_CONN_AG_CTX_BIT0_SHIFT 0 -#define YSTORM_ETH_CONN_AG_CTX_BIT1_MASK 0x1 /* exist_in_qm1 */ -#define YSTORM_ETH_CONN_AG_CTX_BIT1_SHIFT 1 -#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_MASK 0x3 /* cf0 */ -#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_SHIFT 2 -#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_MASK 0x3 /* cf1 */ -#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_SHIFT 4 -#define YSTORM_ETH_CONN_AG_CTX_CF2_MASK 0x3 /* cf2 */ -#define YSTORM_ETH_CONN_AG_CTX_CF2_SHIFT 6 + u8 byte0; + u8 state; + u8 flags0; +#define YSTORM_ETH_CONN_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_ETH_CONN_AG_CTX_BIT0_SHIFT 0 +#define YSTORM_ETH_CONN_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_ETH_CONN_AG_CTX_BIT1_SHIFT 1 +#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_MASK 0x3 +#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_SHIFT 2 +#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_MASK 0x3 +#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_SHIFT 4 +#define YSTORM_ETH_CONN_AG_CTX_CF2_MASK 0x3 +#define YSTORM_ETH_CONN_AG_CTX_CF2_SHIFT 6 u8 flags1; -#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_MASK 0x1 /* cf0en */ -#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_SHIFT 0 -#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_EN_MASK 0x1 /* cf1en */ -#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_EN_SHIFT 1 -#define YSTORM_ETH_CONN_AG_CTX_CF2EN_MASK 0x1 /* cf2en */ -#define YSTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT 2 -#define YSTORM_ETH_CONN_AG_CTX_RULE0EN_MASK 0x1 /* rule0en */ -#define YSTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT 3 -#define YSTORM_ETH_CONN_AG_CTX_RULE1EN_MASK 0x1 /* rule1en */ -#define YSTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT 4 -#define YSTORM_ETH_CONN_AG_CTX_RULE2EN_MASK 0x1 /* rule2en */ -#define YSTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT 5 -#define YSTORM_ETH_CONN_AG_CTX_RULE3EN_MASK 0x1 /* rule3en */ -#define YSTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT 6 -#define YSTORM_ETH_CONN_AG_CTX_RULE4EN_MASK 0x1 /* rule4en */ -#define YSTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT 7 - u8 byte2 /* byte2 */; - u8 byte3 /* byte3 */; - __le16 word0 /* word0 */; - __le32 terminate_spqe /* reg0 */; - __le32 reg1 /* reg1 */; - __le16 tx_bd_cons_upd /* word1 */; - __le16 word2 /* word2 */; - __le16 word3 /* word3 */; - __le16 word4 /* word4 */; - __le32 reg2 /* reg2 */; - __le32 reg3 /* reg3 */; +#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_MASK 0x1 +#define YSTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_SHIFT 0 +#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_EN_MASK 0x1 +#define YSTORM_ETH_CONN_AG_CTX_PMD_TERMINATE_CF_EN_SHIFT 1 +#define YSTORM_ETH_CONN_AG_CTX_CF2EN_MASK 0x1 +#define YSTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT 2 +#define YSTORM_ETH_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define YSTORM_ETH_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define YSTORM_ETH_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define YSTORM_ETH_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define YSTORM_ETH_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT 7 + u8 tx_q0_int_coallecing_timeset; + u8 byte3; + __le16 word0; + __le32 terminate_spqe; + __le32 reg1; + __le16 tx_bd_cons_upd; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg2; + __le32 reg3; }; struct tstorm_eth_conn_ag_ctx { - u8 byte0 /* cdu_validation */; - u8 byte1 /* state */; - u8 flags0; -#define TSTORM_ETH_CONN_AG_CTX_BIT0_MASK 0x1 /* exist_in_qm0 */ -#define TSTORM_ETH_CONN_AG_CTX_BIT0_SHIFT 0 -#define TSTORM_ETH_CONN_AG_CTX_BIT1_MASK 0x1 /* exist_in_qm1 */ -#define TSTORM_ETH_CONN_AG_CTX_BIT1_SHIFT 1 -#define TSTORM_ETH_CONN_AG_CTX_BIT2_MASK 0x1 /* bit2 */ -#define TSTORM_ETH_CONN_AG_CTX_BIT2_SHIFT 2 -#define TSTORM_ETH_CONN_AG_CTX_BIT3_MASK 0x1 /* bit3 */ -#define TSTORM_ETH_CONN_AG_CTX_BIT3_SHIFT 3 -#define TSTORM_ETH_CONN_AG_CTX_BIT4_MASK 0x1 /* bit4 */ -#define TSTORM_ETH_CONN_AG_CTX_BIT4_SHIFT 4 -#define TSTORM_ETH_CONN_AG_CTX_BIT5_MASK 0x1 /* bit5 */ -#define TSTORM_ETH_CONN_AG_CTX_BIT5_SHIFT 5 -#define TSTORM_ETH_CONN_AG_CTX_CF0_MASK 0x3 /* timer0cf */ -#define TSTORM_ETH_CONN_AG_CTX_CF0_SHIFT 6 + u8 byte0; + u8 byte1; + u8 flags0; +#define TSTORM_ETH_CONN_AG_CTX_BIT0_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_BIT0_SHIFT 0 +#define TSTORM_ETH_CONN_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_BIT1_SHIFT 1 +#define TSTORM_ETH_CONN_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_BIT2_SHIFT 2 +#define TSTORM_ETH_CONN_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_BIT3_SHIFT 3 +#define TSTORM_ETH_CONN_AG_CTX_BIT4_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_BIT4_SHIFT 4 +#define TSTORM_ETH_CONN_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_BIT5_SHIFT 5 +#define TSTORM_ETH_CONN_AG_CTX_CF0_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF0_SHIFT 6 u8 flags1; -#define TSTORM_ETH_CONN_AG_CTX_CF1_MASK 0x3 /* timer1cf */ -#define TSTORM_ETH_CONN_AG_CTX_CF1_SHIFT 0 -#define TSTORM_ETH_CONN_AG_CTX_CF2_MASK 0x3 /* timer2cf */ -#define TSTORM_ETH_CONN_AG_CTX_CF2_SHIFT 2 -#define TSTORM_ETH_CONN_AG_CTX_CF3_MASK 0x3 /* timer_stop_all */ -#define TSTORM_ETH_CONN_AG_CTX_CF3_SHIFT 4 -#define TSTORM_ETH_CONN_AG_CTX_CF4_MASK 0x3 /* cf4 */ -#define TSTORM_ETH_CONN_AG_CTX_CF4_SHIFT 6 +#define TSTORM_ETH_CONN_AG_CTX_CF1_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF1_SHIFT 0 +#define TSTORM_ETH_CONN_AG_CTX_CF2_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF2_SHIFT 2 +#define TSTORM_ETH_CONN_AG_CTX_CF3_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF3_SHIFT 4 +#define TSTORM_ETH_CONN_AG_CTX_CF4_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF4_SHIFT 6 u8 flags2; -#define TSTORM_ETH_CONN_AG_CTX_CF5_MASK 0x3 /* cf5 */ -#define TSTORM_ETH_CONN_AG_CTX_CF5_SHIFT 0 -#define TSTORM_ETH_CONN_AG_CTX_CF6_MASK 0x3 /* cf6 */ -#define TSTORM_ETH_CONN_AG_CTX_CF6_SHIFT 2 -#define TSTORM_ETH_CONN_AG_CTX_CF7_MASK 0x3 /* cf7 */ -#define TSTORM_ETH_CONN_AG_CTX_CF7_SHIFT 4 -#define TSTORM_ETH_CONN_AG_CTX_CF8_MASK 0x3 /* cf8 */ -#define TSTORM_ETH_CONN_AG_CTX_CF8_SHIFT 6 +#define TSTORM_ETH_CONN_AG_CTX_CF5_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF5_SHIFT 0 +#define TSTORM_ETH_CONN_AG_CTX_CF6_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF6_SHIFT 2 +#define TSTORM_ETH_CONN_AG_CTX_CF7_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF7_SHIFT 4 +#define TSTORM_ETH_CONN_AG_CTX_CF8_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF8_SHIFT 6 u8 flags3; -#define TSTORM_ETH_CONN_AG_CTX_CF9_MASK 0x3 /* cf9 */ -#define TSTORM_ETH_CONN_AG_CTX_CF9_SHIFT 0 -#define TSTORM_ETH_CONN_AG_CTX_CF10_MASK 0x3 /* cf10 */ -#define TSTORM_ETH_CONN_AG_CTX_CF10_SHIFT 2 -#define TSTORM_ETH_CONN_AG_CTX_CF0EN_MASK 0x1 /* cf0en */ -#define TSTORM_ETH_CONN_AG_CTX_CF0EN_SHIFT 4 -#define TSTORM_ETH_CONN_AG_CTX_CF1EN_MASK 0x1 /* cf1en */ -#define TSTORM_ETH_CONN_AG_CTX_CF1EN_SHIFT 5 -#define TSTORM_ETH_CONN_AG_CTX_CF2EN_MASK 0x1 /* cf2en */ -#define TSTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT 6 -#define TSTORM_ETH_CONN_AG_CTX_CF3EN_MASK 0x1 /* cf3en */ -#define TSTORM_ETH_CONN_AG_CTX_CF3EN_SHIFT 7 +#define TSTORM_ETH_CONN_AG_CTX_CF9_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF9_SHIFT 0 +#define TSTORM_ETH_CONN_AG_CTX_CF10_MASK 0x3 +#define TSTORM_ETH_CONN_AG_CTX_CF10_SHIFT 2 +#define TSTORM_ETH_CONN_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF0EN_SHIFT 4 +#define TSTORM_ETH_CONN_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF1EN_SHIFT 5 +#define TSTORM_ETH_CONN_AG_CTX_CF2EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT 6 +#define TSTORM_ETH_CONN_AG_CTX_CF3EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF3EN_SHIFT 7 u8 flags4; -#define TSTORM_ETH_CONN_AG_CTX_CF4EN_MASK 0x1 /* cf4en */ -#define TSTORM_ETH_CONN_AG_CTX_CF4EN_SHIFT 0 -#define TSTORM_ETH_CONN_AG_CTX_CF5EN_MASK 0x1 /* cf5en */ -#define TSTORM_ETH_CONN_AG_CTX_CF5EN_SHIFT 1 -#define TSTORM_ETH_CONN_AG_CTX_CF6EN_MASK 0x1 /* cf6en */ -#define TSTORM_ETH_CONN_AG_CTX_CF6EN_SHIFT 2 -#define TSTORM_ETH_CONN_AG_CTX_CF7EN_MASK 0x1 /* cf7en */ -#define TSTORM_ETH_CONN_AG_CTX_CF7EN_SHIFT 3 -#define TSTORM_ETH_CONN_AG_CTX_CF8EN_MASK 0x1 /* cf8en */ -#define TSTORM_ETH_CONN_AG_CTX_CF8EN_SHIFT 4 -#define TSTORM_ETH_CONN_AG_CTX_CF9EN_MASK 0x1 /* cf9en */ -#define TSTORM_ETH_CONN_AG_CTX_CF9EN_SHIFT 5 -#define TSTORM_ETH_CONN_AG_CTX_CF10EN_MASK 0x1 /* cf10en */ -#define TSTORM_ETH_CONN_AG_CTX_CF10EN_SHIFT 6 -#define TSTORM_ETH_CONN_AG_CTX_RULE0EN_MASK 0x1 /* rule0en */ -#define TSTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT 7 +#define TSTORM_ETH_CONN_AG_CTX_CF4EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF4EN_SHIFT 0 +#define TSTORM_ETH_CONN_AG_CTX_CF5EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF5EN_SHIFT 1 +#define TSTORM_ETH_CONN_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF6EN_SHIFT 2 +#define TSTORM_ETH_CONN_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF7EN_SHIFT 3 +#define TSTORM_ETH_CONN_AG_CTX_CF8EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF8EN_SHIFT 4 +#define TSTORM_ETH_CONN_AG_CTX_CF9EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF9EN_SHIFT 5 +#define TSTORM_ETH_CONN_AG_CTX_CF10EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_CF10EN_SHIFT 6 +#define TSTORM_ETH_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT 7 u8 flags5; -#define TSTORM_ETH_CONN_AG_CTX_RULE1EN_MASK 0x1 /* rule1en */ -#define TSTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT 0 -#define TSTORM_ETH_CONN_AG_CTX_RULE2EN_MASK 0x1 /* rule2en */ -#define TSTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT 1 -#define TSTORM_ETH_CONN_AG_CTX_RULE3EN_MASK 0x1 /* rule3en */ -#define TSTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT 2 -#define TSTORM_ETH_CONN_AG_CTX_RULE4EN_MASK 0x1 /* rule4en */ -#define TSTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT 3 -#define TSTORM_ETH_CONN_AG_CTX_RULE5EN_MASK 0x1 /* rule5en */ -#define TSTORM_ETH_CONN_AG_CTX_RULE5EN_SHIFT 4 -#define TSTORM_ETH_CONN_AG_CTX_RX_BD_EN_MASK 0x1 /* rule6en */ -#define TSTORM_ETH_CONN_AG_CTX_RX_BD_EN_SHIFT 5 -#define TSTORM_ETH_CONN_AG_CTX_RULE7EN_MASK 0x1 /* rule7en */ -#define TSTORM_ETH_CONN_AG_CTX_RULE7EN_SHIFT 6 -#define TSTORM_ETH_CONN_AG_CTX_RULE8EN_MASK 0x1 /* rule8en */ -#define TSTORM_ETH_CONN_AG_CTX_RULE8EN_SHIFT 7 - __le32 reg0 /* reg0 */; - __le32 reg1 /* reg1 */; - __le32 reg2 /* reg2 */; - __le32 reg3 /* reg3 */; - __le32 reg4 /* reg4 */; - __le32 reg5 /* reg5 */; - __le32 reg6 /* reg6 */; - __le32 reg7 /* reg7 */; - __le32 reg8 /* reg8 */; - u8 byte2 /* byte2 */; - u8 byte3 /* byte3 */; - __le16 rx_bd_cons /* word0 */; - u8 byte4 /* byte4 */; - u8 byte5 /* byte5 */; - __le16 rx_bd_prod /* word1 */; - __le16 word2 /* conn_dpi */; - __le16 word3 /* word3 */; - __le32 reg9 /* reg9 */; - __le32 reg10 /* reg10 */; +#define TSTORM_ETH_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define TSTORM_ETH_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define TSTORM_ETH_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define TSTORM_ETH_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define TSTORM_ETH_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define TSTORM_ETH_CONN_AG_CTX_RX_BD_EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_RX_BD_EN_SHIFT 5 +#define TSTORM_ETH_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define TSTORM_ETH_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define TSTORM_ETH_CONN_AG_CTX_RULE8EN_SHIFT 7 + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le32 reg7; + __le32 reg8; + u8 byte2; + u8 byte3; + __le16 rx_bd_cons; + u8 byte4; + u8 byte5; + __le16 rx_bd_prod; + __le16 word2; + __le16 word3; + __le32 reg9; + __le32 reg10; }; struct ustorm_eth_conn_ag_ctx { - u8 byte0 /* cdu_validation */; - u8 byte1 /* state */; - u8 flags0; -#define USTORM_ETH_CONN_AG_CTX_BIT0_MASK 0x1 /* exist_in_qm0 */ -#define USTORM_ETH_CONN_AG_CTX_BIT0_SHIFT 0 -#define USTORM_ETH_CONN_AG_CTX_BIT1_MASK 0x1 /* exist_in_qm1 */ -#define USTORM_ETH_CONN_AG_CTX_BIT1_SHIFT 1 -#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_MASK 0x3 /* timer0cf */ -#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_SHIFT 2 -#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_MASK 0x3 /* timer1cf */ -#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_SHIFT 4 -#define USTORM_ETH_CONN_AG_CTX_CF2_MASK 0x3 /* timer2cf */ -#define USTORM_ETH_CONN_AG_CTX_CF2_SHIFT 6 + u8 byte0; + u8 byte1; + u8 flags0; +#define USTORM_ETH_CONN_AG_CTX_BIT0_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_BIT0_SHIFT 0 +#define USTORM_ETH_CONN_AG_CTX_BIT1_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_BIT1_SHIFT 1 +#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_MASK 0x3 +#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_SHIFT 2 +#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_MASK 0x3 +#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_SHIFT 4 +#define USTORM_ETH_CONN_AG_CTX_CF2_MASK 0x3 +#define USTORM_ETH_CONN_AG_CTX_CF2_SHIFT 6 u8 flags1; -#define USTORM_ETH_CONN_AG_CTX_CF3_MASK 0x3 /* timer_stop_all */ -#define USTORM_ETH_CONN_AG_CTX_CF3_SHIFT 0 -#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_MASK 0x3 /* cf4 */ -#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_SHIFT 2 -#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_MASK 0x3 /* cf5 */ -#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_SHIFT 4 -#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_MASK 0x3 /* cf6 */ -#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_SHIFT 6 +#define USTORM_ETH_CONN_AG_CTX_CF3_MASK 0x3 +#define USTORM_ETH_CONN_AG_CTX_CF3_SHIFT 0 +#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_MASK 0x3 +#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_SHIFT 2 +#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_MASK 0x3 +#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_SHIFT 4 +#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_MASK 0x3 +#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_SHIFT 6 u8 flags2; -#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_EN_MASK 0x1 /* cf0en */ -#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_EN_SHIFT 0 -#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_EN_MASK 0x1 /* cf1en */ -#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_EN_SHIFT 1 -#define USTORM_ETH_CONN_AG_CTX_CF2EN_MASK 0x1 /* cf2en */ -#define USTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT 2 -#define USTORM_ETH_CONN_AG_CTX_CF3EN_MASK 0x1 /* cf3en */ -#define USTORM_ETH_CONN_AG_CTX_CF3EN_SHIFT 3 -#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_EN_MASK 0x1 /* cf4en */ -#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_EN_SHIFT 4 -#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_EN_MASK 0x1 /* cf5en */ -#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_EN_SHIFT 5 -#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_MASK 0x1 /* cf6en */ -#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_SHIFT 6 -#define USTORM_ETH_CONN_AG_CTX_RULE0EN_MASK 0x1 /* rule0en */ -#define USTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT 7 +#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_TX_PMD_TERMINATE_CF_EN_SHIFT 0 +#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RX_PMD_TERMINATE_CF_EN_SHIFT 1 +#define USTORM_ETH_CONN_AG_CTX_CF2EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_CF2EN_SHIFT 2 +#define USTORM_ETH_CONN_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_TX_ARM_CF_EN_SHIFT 4 +#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RX_ARM_CF_EN_SHIFT 5 +#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_TX_BD_CONS_UPD_CF_EN_SHIFT 6 +#define USTORM_ETH_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RULE0EN_SHIFT 7 u8 flags3; -#define USTORM_ETH_CONN_AG_CTX_RULE1EN_MASK 0x1 /* rule1en */ -#define USTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT 0 -#define USTORM_ETH_CONN_AG_CTX_RULE2EN_MASK 0x1 /* rule2en */ -#define USTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT 1 -#define USTORM_ETH_CONN_AG_CTX_RULE3EN_MASK 0x1 /* rule3en */ -#define USTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT 2 -#define USTORM_ETH_CONN_AG_CTX_RULE4EN_MASK 0x1 /* rule4en */ -#define USTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT 3 -#define USTORM_ETH_CONN_AG_CTX_RULE5EN_MASK 0x1 /* rule5en */ -#define USTORM_ETH_CONN_AG_CTX_RULE5EN_SHIFT 4 -#define USTORM_ETH_CONN_AG_CTX_RULE6EN_MASK 0x1 /* rule6en */ -#define USTORM_ETH_CONN_AG_CTX_RULE6EN_SHIFT 5 -#define USTORM_ETH_CONN_AG_CTX_RULE7EN_MASK 0x1 /* rule7en */ -#define USTORM_ETH_CONN_AG_CTX_RULE7EN_SHIFT 6 -#define USTORM_ETH_CONN_AG_CTX_RULE8EN_MASK 0x1 /* rule8en */ -#define USTORM_ETH_CONN_AG_CTX_RULE8EN_SHIFT 7 - u8 byte2 /* byte2 */; - u8 byte3 /* byte3 */; - __le16 word0 /* conn_dpi */; - __le16 tx_bd_cons /* word1 */; - __le32 reg0 /* reg0 */; - __le32 reg1 /* reg1 */; - __le32 reg2 /* reg2 */; - __le32 tx_int_coallecing_timeset /* reg3 */; - __le16 tx_drv_bd_cons /* word2 */; - __le16 rx_drv_cqe_cons /* word3 */; +#define USTORM_ETH_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define USTORM_ETH_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define USTORM_ETH_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define USTORM_ETH_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define USTORM_ETH_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define USTORM_ETH_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define USTORM_ETH_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define USTORM_ETH_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define USTORM_ETH_CONN_AG_CTX_RULE8EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le16 tx_bd_cons; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 tx_int_coallecing_timeset; + __le16 tx_drv_bd_cons; + __le16 rx_drv_cqe_cons; }; /* The eth storm context for the Ustorm */ @@ -2876,47 +3183,75 @@ struct mstorm_eth_conn_st_ctx { /* eth connection context */ struct eth_conn_context { - struct tstorm_eth_conn_st_ctx tstorm_st_context; - struct regpair tstorm_st_padding[2]; - struct pstorm_eth_conn_st_ctx pstorm_st_context; - struct xstorm_eth_conn_st_ctx xstorm_st_context; - struct xstorm_eth_conn_ag_ctx xstorm_ag_context; - struct ystorm_eth_conn_st_ctx ystorm_st_context; - struct ystorm_eth_conn_ag_ctx ystorm_ag_context; - struct tstorm_eth_conn_ag_ctx tstorm_ag_context; - struct ustorm_eth_conn_ag_ctx ustorm_ag_context; - struct ustorm_eth_conn_st_ctx ustorm_st_context; - struct mstorm_eth_conn_st_ctx mstorm_st_context; + struct tstorm_eth_conn_st_ctx tstorm_st_context; + struct regpair tstorm_st_padding[2]; + struct pstorm_eth_conn_st_ctx pstorm_st_context; + struct xstorm_eth_conn_st_ctx xstorm_st_context; + struct xstorm_eth_conn_ag_ctx xstorm_ag_context; + struct ystorm_eth_conn_st_ctx ystorm_st_context; + struct ystorm_eth_conn_ag_ctx ystorm_ag_context; + struct tstorm_eth_conn_ag_ctx tstorm_ag_context; + struct ustorm_eth_conn_ag_ctx ustorm_ag_context; + struct ustorm_eth_conn_st_ctx ustorm_st_context; + struct mstorm_eth_conn_st_ctx mstorm_st_context; }; +/* opcodes for the event ring */ +enum eth_event_opcode { + ETH_EVENT_UNUSED, + ETH_EVENT_VPORT_START, + ETH_EVENT_VPORT_UPDATE, + ETH_EVENT_VPORT_STOP, + ETH_EVENT_TX_QUEUE_START, + ETH_EVENT_TX_QUEUE_STOP, + ETH_EVENT_RX_QUEUE_START, + ETH_EVENT_RX_QUEUE_UPDATE, + ETH_EVENT_RX_QUEUE_STOP, + ETH_EVENT_FILTERS_UPDATE, + ETH_EVENT_RESERVED, + ETH_EVENT_RESERVED2, + ETH_EVENT_RESERVED3, + ETH_EVENT_RX_ADD_UDP_FILTER, + ETH_EVENT_RX_DELETE_UDP_FILTER, + ETH_EVENT_RESERVED4, + ETH_EVENT_RESERVED5, + MAX_ETH_EVENT_OPCODE +}; + +/* Classify rule types in E2/E3 */ enum eth_filter_action { + ETH_FILTER_ACTION_UNUSED, ETH_FILTER_ACTION_REMOVE, ETH_FILTER_ACTION_ADD, ETH_FILTER_ACTION_REMOVE_ALL, MAX_ETH_FILTER_ACTION }; +/* Command for adding/removing a classification rule $$KEEP_ENDIANNESS$$ */ struct eth_filter_cmd { - u8 type /* Filter Type (MAC/VLAN/Pair/VNI) */; - u8 vport_id /* the vport id */; - u8 action /* filter command action: add/remove/replace */; - u8 reserved0; - __le32 vni; - __le16 mac_lsb; - __le16 mac_mid; - __le16 mac_msb; - __le16 vlan_id; + u8 type; + u8 vport_id; + u8 action; + u8 reserved0; + __le32 vni; + __le16 mac_lsb; + __le16 mac_mid; + __le16 mac_msb; + __le16 vlan_id; }; +/* $$KEEP_ENDIANNESS$$ */ struct eth_filter_cmd_header { - u8 rx; - u8 tx; - u8 cmd_cnt; - u8 assert_on_error; - u8 reserved1[4]; + u8 rx; + u8 tx; + u8 cmd_cnt; + u8 assert_on_error; + u8 reserved1[4]; }; +/* Ethernet filter types: mac/vlan/pair */ enum eth_filter_type { + ETH_FILTER_TYPE_UNUSED, ETH_FILTER_TYPE_MAC, ETH_FILTER_TYPE_VLAN, ETH_FILTER_TYPE_PAIR, @@ -2929,463 +3264,512 @@ enum eth_filter_type { MAX_ETH_FILTER_TYPE }; +/* Ethernet Ramrod Command IDs */ enum eth_ramrod_cmd_id { ETH_RAMROD_UNUSED, - ETH_RAMROD_VPORT_START /* VPort Start Ramrod */, - ETH_RAMROD_VPORT_UPDATE /* VPort Update Ramrod */, - ETH_RAMROD_VPORT_STOP /* VPort Stop Ramrod */, - ETH_RAMROD_RX_QUEUE_START /* RX Queue Start Ramrod */, - ETH_RAMROD_RX_QUEUE_STOP /* RX Queue Stop Ramrod */, - ETH_RAMROD_TX_QUEUE_START /* TX Queue Start Ramrod */, - ETH_RAMROD_TX_QUEUE_STOP /* TX Queue Stop Ramrod */, - ETH_RAMROD_FILTERS_UPDATE /* Add or Remove Mac/Vlan/Pair filters */, - ETH_RAMROD_RX_QUEUE_UPDATE /* RX Queue Update Ramrod */, - ETH_RAMROD_RESERVED, - ETH_RAMROD_RESERVED2, - ETH_RAMROD_RESERVED3, - ETH_RAMROD_RESERVED4, - ETH_RAMROD_RESERVED5, - ETH_RAMROD_RESERVED6, - ETH_RAMROD_RESERVED7, - ETH_RAMROD_RESERVED8, + ETH_RAMROD_VPORT_START, + ETH_RAMROD_VPORT_UPDATE, + ETH_RAMROD_VPORT_STOP, + ETH_RAMROD_RX_QUEUE_START, + ETH_RAMROD_RX_QUEUE_STOP, + ETH_RAMROD_TX_QUEUE_START, + ETH_RAMROD_TX_QUEUE_STOP, + ETH_RAMROD_FILTERS_UPDATE, + ETH_RAMROD_RX_QUEUE_UPDATE, + ETH_RAMROD_RX_CREATE_OPENFLOW_ACTION, + ETH_RAMROD_RX_ADD_OPENFLOW_FILTER, + ETH_RAMROD_RX_DELETE_OPENFLOW_FILTER, + ETH_RAMROD_RX_ADD_UDP_FILTER, + ETH_RAMROD_RX_DELETE_UDP_FILTER, + ETH_RAMROD_RX_CREATE_GFT_ACTION, + ETH_RAMROD_GFT_UPDATE_FILTER, MAX_ETH_RAMROD_CMD_ID }; +/* return code from eth sp ramrods */ +struct eth_return_code { + u8 value; +#define ETH_RETURN_CODE_ERR_CODE_MASK 0x1F +#define ETH_RETURN_CODE_ERR_CODE_SHIFT 0 +#define ETH_RETURN_CODE_RESERVED_MASK 0x3 +#define ETH_RETURN_CODE_RESERVED_SHIFT 5 +#define ETH_RETURN_CODE_RX_TX_MASK 0x1 +#define ETH_RETURN_CODE_RX_TX_SHIFT 7 +}; + +/* What to do in case an error occurs */ enum eth_tx_err { - ETH_TX_ERR_DROP /* Drop erronous packet. */, + ETH_TX_ERR_DROP, ETH_TX_ERR_ASSERT_MALICIOUS, MAX_ETH_TX_ERR }; +/* Array of the different error type behaviors */ struct eth_tx_err_vals { __le16 values; -#define ETH_TX_ERR_VALS_ILLEGAL_VLAN_MODE_MASK 0x1 -#define ETH_TX_ERR_VALS_ILLEGAL_VLAN_MODE_SHIFT 0 -#define ETH_TX_ERR_VALS_PACKET_TOO_SMALL_MASK 0x1 -#define ETH_TX_ERR_VALS_PACKET_TOO_SMALL_SHIFT 1 -#define ETH_TX_ERR_VALS_ANTI_SPOOFING_ERR_MASK 0x1 -#define ETH_TX_ERR_VALS_ANTI_SPOOFING_ERR_SHIFT 2 -#define ETH_TX_ERR_VALS_ILLEGAL_INBAND_TAGS_MASK 0x1 -#define ETH_TX_ERR_VALS_ILLEGAL_INBAND_TAGS_SHIFT 3 -#define ETH_TX_ERR_VALS_VLAN_INSERTION_W_INBAND_TAG_MASK 0x1 -#define ETH_TX_ERR_VALS_VLAN_INSERTION_W_INBAND_TAG_SHIFT 4 -#define ETH_TX_ERR_VALS_MTU_VIOLATION_MASK 0x1 -#define ETH_TX_ERR_VALS_MTU_VIOLATION_SHIFT 5 -#define ETH_TX_ERR_VALS_ILLEGAL_CONTROL_FRAME_MASK 0x1 -#define ETH_TX_ERR_VALS_ILLEGAL_CONTROL_FRAME_SHIFT 6 -#define ETH_TX_ERR_VALS_RESERVED_MASK 0x1FF -#define ETH_TX_ERR_VALS_RESERVED_SHIFT 7 -}; - +#define ETH_TX_ERR_VALS_ILLEGAL_VLAN_MODE_MASK 0x1 +#define ETH_TX_ERR_VALS_ILLEGAL_VLAN_MODE_SHIFT 0 +#define ETH_TX_ERR_VALS_PACKET_TOO_SMALL_MASK 0x1 +#define ETH_TX_ERR_VALS_PACKET_TOO_SMALL_SHIFT 1 +#define ETH_TX_ERR_VALS_ANTI_SPOOFING_ERR_MASK 0x1 +#define ETH_TX_ERR_VALS_ANTI_SPOOFING_ERR_SHIFT 2 +#define ETH_TX_ERR_VALS_ILLEGAL_INBAND_TAGS_MASK 0x1 +#define ETH_TX_ERR_VALS_ILLEGAL_INBAND_TAGS_SHIFT 3 +#define ETH_TX_ERR_VALS_VLAN_INSERTION_W_INBAND_TAG_MASK 0x1 +#define ETH_TX_ERR_VALS_VLAN_INSERTION_W_INBAND_TAG_SHIFT 4 +#define ETH_TX_ERR_VALS_MTU_VIOLATION_MASK 0x1 +#define ETH_TX_ERR_VALS_MTU_VIOLATION_SHIFT 5 +#define ETH_TX_ERR_VALS_ILLEGAL_CONTROL_FRAME_MASK 0x1 +#define ETH_TX_ERR_VALS_ILLEGAL_CONTROL_FRAME_SHIFT 6 +#define ETH_TX_ERR_VALS_RESERVED_MASK 0x1FF +#define ETH_TX_ERR_VALS_RESERVED_SHIFT 7 +}; + +/* vport rss configuration data */ struct eth_vport_rss_config { __le16 capabilities; -#define ETH_VPORT_RSS_CONFIG_IPV4_CAPABILITY_MASK 0x1 -#define ETH_VPORT_RSS_CONFIG_IPV4_CAPABILITY_SHIFT 0 -#define ETH_VPORT_RSS_CONFIG_IPV6_CAPABILITY_MASK 0x1 -#define ETH_VPORT_RSS_CONFIG_IPV6_CAPABILITY_SHIFT 1 -#define ETH_VPORT_RSS_CONFIG_IPV4_TCP_CAPABILITY_MASK 0x1 -#define ETH_VPORT_RSS_CONFIG_IPV4_TCP_CAPABILITY_SHIFT 2 -#define ETH_VPORT_RSS_CONFIG_IPV6_TCP_CAPABILITY_MASK 0x1 -#define ETH_VPORT_RSS_CONFIG_IPV6_TCP_CAPABILITY_SHIFT 3 -#define ETH_VPORT_RSS_CONFIG_IPV4_UDP_CAPABILITY_MASK 0x1 -#define ETH_VPORT_RSS_CONFIG_IPV4_UDP_CAPABILITY_SHIFT 4 -#define ETH_VPORT_RSS_CONFIG_IPV6_UDP_CAPABILITY_MASK 0x1 -#define ETH_VPORT_RSS_CONFIG_IPV6_UDP_CAPABILITY_SHIFT 5 -#define ETH_VPORT_RSS_CONFIG_EN_5_TUPLE_CAPABILITY_MASK 0x1 -#define ETH_VPORT_RSS_CONFIG_EN_5_TUPLE_CAPABILITY_SHIFT 6 -#define ETH_VPORT_RSS_CONFIG_RESERVED0_MASK 0x1FF -#define ETH_VPORT_RSS_CONFIG_RESERVED0_SHIFT 7 - u8 rss_id; - u8 rss_mode; - u8 update_rss_key; - u8 update_rss_ind_table; - u8 update_rss_capabilities; - u8 tbl_size; - __le32 reserved2[2]; - __le16 indirection_table[ETH_RSS_IND_TABLE_ENTRIES_NUM]; - __le32 rss_key[ETH_RSS_KEY_SIZE_REGS]; - __le32 reserved3[2]; -}; - +#define ETH_VPORT_RSS_CONFIG_IPV4_CAPABILITY_MASK 0x1 +#define ETH_VPORT_RSS_CONFIG_IPV4_CAPABILITY_SHIFT 0 +#define ETH_VPORT_RSS_CONFIG_IPV6_CAPABILITY_MASK 0x1 +#define ETH_VPORT_RSS_CONFIG_IPV6_CAPABILITY_SHIFT 1 +#define ETH_VPORT_RSS_CONFIG_IPV4_TCP_CAPABILITY_MASK 0x1 +#define ETH_VPORT_RSS_CONFIG_IPV4_TCP_CAPABILITY_SHIFT 2 +#define ETH_VPORT_RSS_CONFIG_IPV6_TCP_CAPABILITY_MASK 0x1 +#define ETH_VPORT_RSS_CONFIG_IPV6_TCP_CAPABILITY_SHIFT 3 +#define ETH_VPORT_RSS_CONFIG_IPV4_UDP_CAPABILITY_MASK 0x1 +#define ETH_VPORT_RSS_CONFIG_IPV4_UDP_CAPABILITY_SHIFT 4 +#define ETH_VPORT_RSS_CONFIG_IPV6_UDP_CAPABILITY_MASK 0x1 +#define ETH_VPORT_RSS_CONFIG_IPV6_UDP_CAPABILITY_SHIFT 5 +#define ETH_VPORT_RSS_CONFIG_EN_5_TUPLE_CAPABILITY_MASK 0x1 +#define ETH_VPORT_RSS_CONFIG_EN_5_TUPLE_CAPABILITY_SHIFT 6 +#define ETH_VPORT_RSS_CONFIG_RESERVED0_MASK 0x1FF +#define ETH_VPORT_RSS_CONFIG_RESERVED0_SHIFT 7 + u8 rss_id; + u8 rss_mode; + u8 update_rss_key; + u8 update_rss_ind_table; + u8 update_rss_capabilities; + u8 tbl_size; + __le32 reserved2[2]; + __le16 indirection_table[ETH_RSS_IND_TABLE_ENTRIES_NUM]; + + __le32 rss_key[ETH_RSS_KEY_SIZE_REGS]; + __le32 reserved3[2]; +}; + +/* eth vport RSS mode */ enum eth_vport_rss_mode { ETH_VPORT_RSS_MODE_DISABLED, ETH_VPORT_RSS_MODE_REGULAR, MAX_ETH_VPORT_RSS_MODE }; +/* Command for setting classification flags for a vport $$KEEP_ENDIANNESS$$ */ struct eth_vport_rx_mode { __le16 state; -#define ETH_VPORT_RX_MODE_UCAST_DROP_ALL_MASK 0x1 -#define ETH_VPORT_RX_MODE_UCAST_DROP_ALL_SHIFT 0 -#define ETH_VPORT_RX_MODE_UCAST_ACCEPT_ALL_MASK 0x1 -#define ETH_VPORT_RX_MODE_UCAST_ACCEPT_ALL_SHIFT 1 -#define ETH_VPORT_RX_MODE_UCAST_ACCEPT_UNMATCHED_MASK 0x1 -#define ETH_VPORT_RX_MODE_UCAST_ACCEPT_UNMATCHED_SHIFT 2 -#define ETH_VPORT_RX_MODE_MCAST_DROP_ALL_MASK 0x1 -#define ETH_VPORT_RX_MODE_MCAST_DROP_ALL_SHIFT 3 -#define ETH_VPORT_RX_MODE_MCAST_ACCEPT_ALL_MASK 0x1 -#define ETH_VPORT_RX_MODE_MCAST_ACCEPT_ALL_SHIFT 4 -#define ETH_VPORT_RX_MODE_BCAST_ACCEPT_ALL_MASK 0x1 -#define ETH_VPORT_RX_MODE_BCAST_ACCEPT_ALL_SHIFT 5 -#define ETH_VPORT_RX_MODE_RESERVED1_MASK 0x3FF -#define ETH_VPORT_RX_MODE_RESERVED1_SHIFT 6 +#define ETH_VPORT_RX_MODE_UCAST_DROP_ALL_MASK 0x1 +#define ETH_VPORT_RX_MODE_UCAST_DROP_ALL_SHIFT 0 +#define ETH_VPORT_RX_MODE_UCAST_ACCEPT_ALL_MASK 0x1 +#define ETH_VPORT_RX_MODE_UCAST_ACCEPT_ALL_SHIFT 1 +#define ETH_VPORT_RX_MODE_UCAST_ACCEPT_UNMATCHED_MASK 0x1 +#define ETH_VPORT_RX_MODE_UCAST_ACCEPT_UNMATCHED_SHIFT 2 +#define ETH_VPORT_RX_MODE_MCAST_DROP_ALL_MASK 0x1 +#define ETH_VPORT_RX_MODE_MCAST_DROP_ALL_SHIFT 3 +#define ETH_VPORT_RX_MODE_MCAST_ACCEPT_ALL_MASK 0x1 +#define ETH_VPORT_RX_MODE_MCAST_ACCEPT_ALL_SHIFT 4 +#define ETH_VPORT_RX_MODE_BCAST_ACCEPT_ALL_MASK 0x1 +#define ETH_VPORT_RX_MODE_BCAST_ACCEPT_ALL_SHIFT 5 +#define ETH_VPORT_RX_MODE_RESERVED1_MASK 0x3FF +#define ETH_VPORT_RX_MODE_RESERVED1_SHIFT 6 __le16 reserved2[3]; }; +/* Command for setting tpa parameters */ struct eth_vport_tpa_param { - u8 tpa_ipv4_en_flg; - u8 tpa_ipv6_en_flg; - u8 tpa_ipv4_tunn_en_flg; - u8 tpa_ipv6_tunn_en_flg; - u8 tpa_pkt_split_flg; - u8 tpa_hdr_data_split_flg; - u8 tpa_gro_consistent_flg; - u8 tpa_max_aggs_num; - u16 tpa_max_size; - u16 tpa_min_size_to_start; - u16 tpa_min_size_to_cont; - u8 max_buff_num; - u8 reserved; + u8 tpa_ipv4_en_flg; + u8 tpa_ipv6_en_flg; + u8 tpa_ipv4_tunn_en_flg; + u8 tpa_ipv6_tunn_en_flg; + u8 tpa_pkt_split_flg; + u8 tpa_hdr_data_split_flg; + u8 tpa_gro_consistent_flg; + + u8 tpa_max_aggs_num; + + __le16 tpa_max_size; + __le16 tpa_min_size_to_start; + + __le16 tpa_min_size_to_cont; + u8 max_buff_num; + u8 reserved; }; +/* Command for setting classification flags for a vport $$KEEP_ENDIANNESS$$ */ struct eth_vport_tx_mode { __le16 state; -#define ETH_VPORT_TX_MODE_UCAST_DROP_ALL_MASK 0x1 -#define ETH_VPORT_TX_MODE_UCAST_DROP_ALL_SHIFT 0 -#define ETH_VPORT_TX_MODE_UCAST_ACCEPT_ALL_MASK 0x1 -#define ETH_VPORT_TX_MODE_UCAST_ACCEPT_ALL_SHIFT 1 -#define ETH_VPORT_TX_MODE_MCAST_DROP_ALL_MASK 0x1 -#define ETH_VPORT_TX_MODE_MCAST_DROP_ALL_SHIFT 2 -#define ETH_VPORT_TX_MODE_MCAST_ACCEPT_ALL_MASK 0x1 -#define ETH_VPORT_TX_MODE_MCAST_ACCEPT_ALL_SHIFT 3 -#define ETH_VPORT_TX_MODE_BCAST_ACCEPT_ALL_MASK 0x1 -#define ETH_VPORT_TX_MODE_BCAST_ACCEPT_ALL_SHIFT 4 -#define ETH_VPORT_TX_MODE_RESERVED1_MASK 0x7FF -#define ETH_VPORT_TX_MODE_RESERVED1_SHIFT 5 +#define ETH_VPORT_TX_MODE_UCAST_DROP_ALL_MASK 0x1 +#define ETH_VPORT_TX_MODE_UCAST_DROP_ALL_SHIFT 0 +#define ETH_VPORT_TX_MODE_UCAST_ACCEPT_ALL_MASK 0x1 +#define ETH_VPORT_TX_MODE_UCAST_ACCEPT_ALL_SHIFT 1 +#define ETH_VPORT_TX_MODE_MCAST_DROP_ALL_MASK 0x1 +#define ETH_VPORT_TX_MODE_MCAST_DROP_ALL_SHIFT 2 +#define ETH_VPORT_TX_MODE_MCAST_ACCEPT_ALL_MASK 0x1 +#define ETH_VPORT_TX_MODE_MCAST_ACCEPT_ALL_SHIFT 3 +#define ETH_VPORT_TX_MODE_BCAST_ACCEPT_ALL_MASK 0x1 +#define ETH_VPORT_TX_MODE_BCAST_ACCEPT_ALL_SHIFT 4 +#define ETH_VPORT_TX_MODE_RESERVED1_MASK 0x7FF +#define ETH_VPORT_TX_MODE_RESERVED1_SHIFT 5 __le16 reserved2[3]; }; +/* Ramrod data for rx queue start ramrod */ struct rx_queue_start_ramrod_data { - __le16 rx_queue_id; - __le16 num_of_pbl_pages; - __le16 bd_max_bytes; - __le16 sb_id; - u8 sb_index; - u8 vport_id; - u8 default_rss_queue_flg; - u8 complete_cqe_flg; - u8 complete_event_flg; - u8 stats_counter_id; - u8 pin_context; - u8 pxp_tph_valid_bd; - u8 pxp_tph_valid_pkt; - u8 pxp_st_hint; - __le16 pxp_st_index; - u8 pmd_mode; - u8 notify_en; - u8 toggle_val; - u8 reserved[7]; - __le16 reserved1; - struct regpair cqe_pbl_addr; - struct regpair bd_base; - struct regpair reserved2; + __le16 rx_queue_id; + __le16 num_of_pbl_pages; + __le16 bd_max_bytes; + __le16 sb_id; + u8 sb_index; + u8 vport_id; + u8 default_rss_queue_flg; + u8 complete_cqe_flg; + u8 complete_event_flg; + u8 stats_counter_id; + u8 pin_context; + u8 pxp_tph_valid_bd; + u8 pxp_tph_valid_pkt; + u8 pxp_st_hint; + + __le16 pxp_st_index; + u8 pmd_mode; + + u8 notify_en; + u8 toggle_val; + + u8 vf_rx_prod_index; + + u8 reserved[6]; + __le16 reserved1; + struct regpair cqe_pbl_addr; + struct regpair bd_base; + struct regpair reserved2; }; +/* Ramrod data for rx queue start ramrod */ struct rx_queue_stop_ramrod_data { - __le16 rx_queue_id; - u8 complete_cqe_flg; - u8 complete_event_flg; - u8 vport_id; - u8 reserved[3]; + __le16 rx_queue_id; + u8 complete_cqe_flg; + u8 complete_event_flg; + u8 vport_id; + u8 reserved[3]; }; +/* Ramrod data for rx queue update ramrod */ struct rx_queue_update_ramrod_data { - __le16 rx_queue_id; - u8 complete_cqe_flg; - u8 complete_event_flg; - u8 vport_id; - u8 reserved[4]; - u8 reserved1; - u8 reserved2; - u8 reserved3; - __le16 reserved4; - __le16 reserved5; + __le16 rx_queue_id; + u8 complete_cqe_flg; + u8 complete_event_flg; + u8 vport_id; + u8 reserved[4]; + u8 reserved1; + u8 reserved2; + u8 reserved3; + __le16 reserved4; + __le16 reserved5; struct regpair reserved6; }; -struct tx_queue_start_ramrod_data { - __le16 sb_id; - u8 sb_index; - u8 vport_id; - u8 reserved0; - u8 stats_counter_id; - __le16 qm_pq_id; - u8 flags; -#define TX_QUEUE_START_RAMROD_DATA_DISABLE_OPPORTUNISTIC_MASK 0x1 -#define TX_QUEUE_START_RAMROD_DATA_DISABLE_OPPORTUNISTIC_SHIFT 0 -#define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_PKT_DUP_MASK 0x1 -#define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_PKT_DUP_SHIFT 1 -#define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_TX_DEST_MASK 0x1 -#define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_TX_DEST_SHIFT 2 -#define TX_QUEUE_START_RAMROD_DATA_PMD_MODE_MASK 0x1 -#define TX_QUEUE_START_RAMROD_DATA_PMD_MODE_SHIFT 3 -#define TX_QUEUE_START_RAMROD_DATA_NOTIFY_EN_MASK 0x1 -#define TX_QUEUE_START_RAMROD_DATA_NOTIFY_EN_SHIFT 4 -#define TX_QUEUE_START_RAMROD_DATA_PIN_CONTEXT_MASK 0x1 -#define TX_QUEUE_START_RAMROD_DATA_PIN_CONTEXT_SHIFT 5 -#define TX_QUEUE_START_RAMROD_DATA_RESERVED1_MASK 0x3 -#define TX_QUEUE_START_RAMROD_DATA_RESERVED1_SHIFT 6 - u8 pxp_st_hint; - u8 pxp_tph_valid_bd; - u8 pxp_tph_valid_pkt; - __le16 pxp_st_index; - __le16 comp_agg_size; - __le16 queue_zone_id; - __le16 test_dup_count; - __le16 pbl_size; - __le16 tx_queue_id; - struct regpair pbl_base_addr; - struct regpair bd_cons_address; +/* Ramrod data for rx Add UDP Filter */ +struct rx_udp_filter_data { + __le16 action_icid; + __le16 vlan_id; + u8 ip_type; + u8 tenant_id_exists; + __le16 reserved1; + __le32 ip_dst_addr[4]; + __le32 ip_src_addr[4]; + __le16 udp_dst_port; + __le16 udp_src_port; + __le32 tenant_id; }; +/* Ramrod data for rx queue start ramrod */ +struct tx_queue_start_ramrod_data { + __le16 sb_id; + u8 sb_index; + u8 vport_id; + u8 reserved0; + u8 stats_counter_id; + __le16 qm_pq_id; + u8 flags; +#define TX_QUEUE_START_RAMROD_DATA_DISABLE_OPPORTUNISTIC_MASK 0x1 +#define TX_QUEUE_START_RAMROD_DATA_DISABLE_OPPORTUNISTIC_SHIFT 0 +#define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_PKT_DUP_MASK 0x1 +#define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_PKT_DUP_SHIFT 1 +#define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_TX_DEST_MASK 0x1 +#define TX_QUEUE_START_RAMROD_DATA_TEST_MODE_TX_DEST_SHIFT 2 +#define TX_QUEUE_START_RAMROD_DATA_PMD_MODE_MASK 0x1 +#define TX_QUEUE_START_RAMROD_DATA_PMD_MODE_SHIFT 3 +#define TX_QUEUE_START_RAMROD_DATA_NOTIFY_EN_MASK 0x1 +#define TX_QUEUE_START_RAMROD_DATA_NOTIFY_EN_SHIFT 4 +#define TX_QUEUE_START_RAMROD_DATA_PIN_CONTEXT_MASK 0x1 +#define TX_QUEUE_START_RAMROD_DATA_PIN_CONTEXT_SHIFT 5 +#define TX_QUEUE_START_RAMROD_DATA_RESERVED1_MASK 0x3 +#define TX_QUEUE_START_RAMROD_DATA_RESERVED1_SHIFT 6 + u8 pxp_st_hint; + u8 pxp_tph_valid_bd; + u8 pxp_tph_valid_pkt; + __le16 pxp_st_index; + __le16 comp_agg_size; + __le16 queue_zone_id; + __le16 test_dup_count; + __le16 pbl_size; + __le16 tx_queue_id; + + struct regpair pbl_base_addr; + struct regpair bd_cons_address; +}; + +/* Ramrod data for tx queue stop ramrod */ struct tx_queue_stop_ramrod_data { __le16 reserved[4]; }; +/* Ramrod data for vport update ramrod */ struct vport_filter_update_ramrod_data { - struct eth_filter_cmd_header filter_cmd_hdr; - struct eth_filter_cmd filter_cmds[ETH_FILTER_RULES_COUNT]; + struct eth_filter_cmd_header filter_cmd_hdr; + struct eth_filter_cmd filter_cmds[ETH_FILTER_RULES_COUNT]; }; +/* Ramrod data for vport start ramrod */ struct vport_start_ramrod_data { - u8 vport_id; - u8 sw_fid; - __le16 mtu; - u8 drop_ttl0_en; - u8 inner_vlan_removal_en; - struct eth_vport_rx_mode rx_mode; - struct eth_vport_tx_mode tx_mode; - struct eth_vport_tpa_param tpa_param; - __le16 default_vlan; - u8 tx_switching_en; - u8 anti_spoofing_en; - u8 default_vlan_en; - u8 handle_ptp_pkts; - u8 silent_vlan_removal_en; - u8 untagged; - struct eth_tx_err_vals tx_err_behav; - u8 zero_placement_offset; - u8 reserved[7]; -}; - + u8 vport_id; + u8 sw_fid; + __le16 mtu; + u8 drop_ttl0_en; + u8 inner_vlan_removal_en; + struct eth_vport_rx_mode rx_mode; + struct eth_vport_tx_mode tx_mode; + struct eth_vport_tpa_param tpa_param; + __le16 default_vlan; + u8 tx_switching_en; + u8 anti_spoofing_en; + + u8 default_vlan_en; + + u8 handle_ptp_pkts; + u8 silent_vlan_removal_en; + u8 untagged; + struct eth_tx_err_vals tx_err_behav; + + u8 zero_placement_offset; + u8 ctl_frame_mac_check_en; + u8 ctl_frame_ethtype_check_en; + u8 reserved[5]; +}; + +/* Ramrod data for vport stop ramrod */ struct vport_stop_ramrod_data { - u8 vport_id; - u8 reserved[7]; + u8 vport_id; + u8 reserved[7]; }; +/* Ramrod data for vport update ramrod */ struct vport_update_ramrod_data_cmn { - u8 vport_id; - u8 update_rx_active_flg; - u8 rx_active_flg; - u8 update_tx_active_flg; - u8 tx_active_flg; - u8 update_rx_mode_flg; - u8 update_tx_mode_flg; - u8 update_approx_mcast_flg; - u8 update_rss_flg; - u8 update_inner_vlan_removal_en_flg; - u8 inner_vlan_removal_en; - u8 update_tpa_param_flg; - u8 update_tpa_en_flg; - u8 update_tx_switching_en_flg; - u8 tx_switching_en; - u8 update_anti_spoofing_en_flg; - u8 anti_spoofing_en; - u8 update_handle_ptp_pkts; - u8 handle_ptp_pkts; - u8 update_default_vlan_en_flg; - u8 default_vlan_en; - u8 update_default_vlan_flg; - __le16 default_vlan; - u8 update_accept_any_vlan_flg; - u8 accept_any_vlan; - u8 silent_vlan_removal_en; - u8 update_mtu_flg; - __le16 mtu; - u8 reserved[2]; + u8 vport_id; + u8 update_rx_active_flg; + u8 rx_active_flg; + u8 update_tx_active_flg; + u8 tx_active_flg; + u8 update_rx_mode_flg; + u8 update_tx_mode_flg; + u8 update_approx_mcast_flg; + + u8 update_rss_flg; + u8 update_inner_vlan_removal_en_flg; + + u8 inner_vlan_removal_en; + u8 update_tpa_param_flg; + u8 update_tpa_en_flg; + u8 update_tx_switching_en_flg; + + u8 tx_switching_en; + u8 update_anti_spoofing_en_flg; + + u8 anti_spoofing_en; + u8 update_handle_ptp_pkts; + + u8 handle_ptp_pkts; + u8 update_default_vlan_en_flg; + + u8 default_vlan_en; + + u8 update_default_vlan_flg; + + __le16 default_vlan; + u8 update_accept_any_vlan_flg; + + u8 accept_any_vlan; + u8 silent_vlan_removal_en; + u8 update_mtu_flg; + + __le16 mtu; + u8 reserved[2]; }; struct vport_update_ramrod_mcast { __le32 bins[ETH_MULTICAST_MAC_BINS_IN_REGS]; }; +/* Ramrod data for vport update ramrod */ struct vport_update_ramrod_data { - struct vport_update_ramrod_data_cmn common; - struct eth_vport_rx_mode rx_mode; - struct eth_vport_tx_mode tx_mode; - struct eth_vport_tpa_param tpa_param; - struct vport_update_ramrod_mcast approx_mcast; - struct eth_vport_rss_config rss_config; + struct vport_update_ramrod_data_cmn common; + + struct eth_vport_rx_mode rx_mode; + struct eth_vport_tx_mode tx_mode; + struct eth_vport_tpa_param tpa_param; + struct vport_update_ramrod_mcast approx_mcast; + struct eth_vport_rss_config rss_config; }; -#define VF_MAX_STATIC 192 /* In case of K2 */ +#define VF_MAX_STATIC 192 -#define MCP_GLOB_PATH_MAX 2 -#define MCP_PORT_MAX 2 /* Global */ -#define MCP_GLOB_PORT_MAX 4 /* Global */ -#define MCP_GLOB_FUNC_MAX 16 /* Global */ +#define MCP_GLOB_PATH_MAX 2 +#define MCP_PORT_MAX 2 +#define MCP_GLOB_PORT_MAX 4 +#define MCP_GLOB_FUNC_MAX 16 -typedef u32 offsize_t; /* In DWORDS !!! */ /* Offset from the beginning of the MCP scratchpad */ -#define OFFSIZE_OFFSET_SHIFT 0 -#define OFFSIZE_OFFSET_MASK 0x0000ffff +#define OFFSIZE_OFFSET_SHIFT 0 +#define OFFSIZE_OFFSET_MASK 0x0000ffff /* Size of specific element (not the whole array if any) */ -#define OFFSIZE_SIZE_SHIFT 16 -#define OFFSIZE_SIZE_MASK 0xffff0000 +#define OFFSIZE_SIZE_SHIFT 16 +#define OFFSIZE_SIZE_MASK 0xffff0000 -/* SECTION_OFFSET is calculating the offset in bytes out of offsize */ -#define SECTION_OFFSET(_offsize) ((((_offsize & \ - OFFSIZE_OFFSET_MASK) >> \ - OFFSIZE_OFFSET_SHIFT) << 2)) +#define SECTION_OFFSET(_offsize) ((((_offsize & \ + OFFSIZE_OFFSET_MASK) >> \ + OFFSIZE_OFFSET_SHIFT) << 2)) -/* QED_SECTION_SIZE is calculating the size in bytes out of offsize */ -#define QED_SECTION_SIZE(_offsize) (((_offsize & \ - OFFSIZE_SIZE_MASK) >> \ - OFFSIZE_SIZE_SHIFT) << 2) +#define QED_SECTION_SIZE(_offsize) (((_offsize & \ + OFFSIZE_SIZE_MASK) >> \ + OFFSIZE_SIZE_SHIFT) << 2) -/* SECTION_ADDR returns the GRC addr of a section, given offsize and index - * within section. - */ -#define SECTION_ADDR(_offsize, idx) (MCP_REG_SCRATCH + \ - SECTION_OFFSET(_offsize) + \ - (QED_SECTION_SIZE(_offsize) * idx)) +#define SECTION_ADDR(_offsize, idx) (MCP_REG_SCRATCH + \ + SECTION_OFFSET(_offsize) + \ + (QED_SECTION_SIZE(_offsize) * idx)) + +#define SECTION_OFFSIZE_ADDR(_pub_base, _section) \ + (_pub_base + offsetof(struct mcp_public_data, sections[_section])) -/* SECTION_OFFSIZE_ADDR returns the GRC addr to the offsize address. - * Use offsetof, since the OFFSETUP collide with the firmware definition - */ -#define SECTION_OFFSIZE_ADDR(_pub_base, _section) (_pub_base + \ - offsetof(struct \ - mcp_public_data, \ - sections[_section])) /* PHY configuration */ -struct pmm_phy_cfg { - u32 speed; -#define PMM_SPEED_AUTONEG 0 - - u32 pause; /* bitmask */ -#define PMM_PAUSE_NONE 0x0 -#define PMM_PAUSE_AUTONEG 0x1 -#define PMM_PAUSE_RX 0x2 -#define PMM_PAUSE_TX 0x4 - - u32 adv_speed; /* Default should be the speed_cap_mask */ - u32 loopback_mode; -#define PMM_LOOPBACK_NONE 0 -#define PMM_LOOPBACK_INT_PHY 1 -#define PMM_LOOPBACK_EXT_PHY 2 -#define PMM_LOOPBACK_EXT 3 -#define PMM_LOOPBACK_MAC 4 - - /* features */ +struct eth_phy_cfg { + u32 speed; +#define ETH_SPEED_AUTONEG 0 +#define ETH_SPEED_SMARTLINQ 0x8 + + u32 pause; +#define ETH_PAUSE_NONE 0x0 +#define ETH_PAUSE_AUTONEG 0x1 +#define ETH_PAUSE_RX 0x2 +#define ETH_PAUSE_TX 0x4 + + u32 adv_speed; + u32 loopback_mode; +#define ETH_LOOPBACK_NONE (0) +#define ETH_LOOPBACK_INT_PHY (1) +#define ETH_LOOPBACK_EXT_PHY (2) +#define ETH_LOOPBACK_EXT (3) +#define ETH_LOOPBACK_MAC (4) + u32 feature_config_flags; +#define ETH_EEE_MODE_ADV_LPI (1 << 0) }; struct port_mf_cfg { - u32 dynamic_cfg; /* device control channel */ -#define PORT_MF_CFG_OV_TAG_MASK 0x0000ffff -#define PORT_MF_CFG_OV_TAG_SHIFT 0 -#define PORT_MF_CFG_OV_TAG_DEFAULT PORT_MF_CFG_OV_TAG_MASK - - u32 reserved[1]; -}; - -/* DO NOT add new fields in the middle - * MUST be synced with struct pmm_stats_map - */ -struct pmm_stats { - u64 r64; /* 0x00 (Offset 0x00 ) RX 64-byte frame counter*/ - u64 r127; /* 0x01 (Offset 0x08 ) RX 65 to 127 byte frame counter*/ - u64 r255; - u64 r511; - u64 r1023; - u64 r1518; - u64 r1522; - u64 r2047; - u64 r4095; - u64 r9216; - u64 r16383; - u64 rfcs; /* 0x0F (Offset 0x58 ) RX FCS error frame counter*/ - u64 rxcf; /* 0x10 (Offset 0x60 ) RX control frame counter*/ - u64 rxpf; /* 0x11 (Offset 0x68 ) RX pause frame counter*/ - u64 rxpp; /* 0x12 (Offset 0x70 ) RX PFC frame counter*/ - u64 raln; /* 0x16 (Offset 0x78 ) RX alignment error counter*/ - u64 rfcr; /* 0x19 (Offset 0x80 ) RX false carrier counter */ - u64 rovr; /* 0x1A (Offset 0x88 ) RX oversized frame counter*/ - u64 rjbr; /* 0x1B (Offset 0x90 ) RX jabber frame counter */ - u64 rund; /* 0x34 (Offset 0x98 ) RX undersized frame counter */ - u64 rfrg; /* 0x35 (Offset 0xa0 ) RX fragment counter */ - u64 t64; /* 0x40 (Offset 0xa8 ) TX 64-byte frame counter */ - u64 t127; - u64 t255; - u64 t511; - u64 t1023; - u64 t1518; - u64 t2047; - u64 t4095; - u64 t9216; - u64 t16383; - u64 txpf; /* 0x50 (Offset 0xf8 ) TX pause frame counter */ - u64 txpp; /* 0x51 (Offset 0x100) TX PFC frame counter */ - u64 tlpiec; - u64 tncl; - u64 rbyte; /* 0x3d (Offset 0x118) RX byte counter */ - u64 rxuca; /* 0x0c (Offset 0x120) RX UC frame counter */ - u64 rxmca; /* 0x0d (Offset 0x128) RX MC frame counter */ - u64 rxbca; /* 0x0e (Offset 0x130) RX BC frame counter */ - u64 rxpok; - u64 tbyte; /* 0x6f (Offset 0x140) TX byte counter */ - u64 txuca; /* 0x4d (Offset 0x148) TX UC frame counter */ - u64 txmca; /* 0x4e (Offset 0x150) TX MC frame counter */ - u64 txbca; /* 0x4f (Offset 0x158) TX BC frame counter */ - u64 txcf; /* 0x54 (Offset 0x160) TX control frame counter */ + u32 dynamic_cfg; +#define PORT_MF_CFG_OV_TAG_MASK 0x0000ffff +#define PORT_MF_CFG_OV_TAG_SHIFT 0 +#define PORT_MF_CFG_OV_TAG_DEFAULT PORT_MF_CFG_OV_TAG_MASK + + u32 reserved[1]; +}; + +struct eth_stats { + u64 r64; + u64 r127; + u64 r255; + u64 r511; + u64 r1023; + u64 r1518; + u64 r1522; + u64 r2047; + u64 r4095; + u64 r9216; + u64 r16383; + u64 rfcs; + u64 rxcf; + u64 rxpf; + u64 rxpp; + u64 raln; + u64 rfcr; + u64 rovr; + u64 rjbr; + u64 rund; + u64 rfrg; + u64 t64; + u64 t127; + u64 t255; + u64 t511; + u64 t1023; + u64 t1518; + u64 t2047; + u64 t4095; + u64 t9216; + u64 t16383; + u64 txpf; + u64 txpp; + u64 tlpiec; + u64 tncl; + u64 rbyte; + u64 rxuca; + u64 rxmca; + u64 rxbca; + u64 rxpok; + u64 tbyte; + u64 txuca; + u64 txmca; + u64 txbca; + u64 txcf; }; struct brb_stats { - u64 brb_truncate[8]; - u64 brb_discard[8]; + u64 brb_truncate[8]; + u64 brb_discard[8]; }; struct port_stats { - struct brb_stats brb; - struct pmm_stats pmm; + struct brb_stats brb; + struct eth_stats eth; }; -#define CMT_TEAM0 0 -#define CMT_TEAM1 1 -#define CMT_TEAM_MAX 2 - struct couple_mode_teaming { u8 port_cmt[MCP_GLOB_PORT_MAX]; -#define PORT_CMT_IN_TEAM BIT(0) +#define PORT_CMT_IN_TEAM (1 << 0) -#define PORT_CMT_PORT_ROLE BIT(1) -#define PORT_CMT_PORT_INACTIVE (0 << 1) -#define PORT_CMT_PORT_ACTIVE BIT(1) +#define PORT_CMT_PORT_ROLE (1 << 1) +#define PORT_CMT_PORT_INACTIVE (0 << 1) +#define PORT_CMT_PORT_ACTIVE (1 << 1) -#define PORT_CMT_TEAM_MASK BIT(2) -#define PORT_CMT_TEAM0 (0 << 2) -#define PORT_CMT_TEAM1 BIT(2) +#define PORT_CMT_TEAM_MASK (1 << 2) +#define PORT_CMT_TEAM0 (0 << 2) +#define PORT_CMT_TEAM1 (1 << 2) }; -/************************************** -* LLDP and DCBX HSI structures -**************************************/ -#define LLDP_CHASSIS_ID_STAT_LEN 4 -#define LLDP_PORT_ID_STAT_LEN 4 -#define DCBX_MAX_APP_PROTOCOL 32 -#define MAX_SYSTEM_LLDP_TLV_DATA 32 +#define LLDP_CHASSIS_ID_STAT_LEN 4 +#define LLDP_PORT_ID_STAT_LEN 4 +#define DCBX_MAX_APP_PROTOCOL 32 +#define MAX_SYSTEM_LLDP_TLV_DATA 32 -enum lldp_agent_e { +enum _lldp_agent { LLDP_NEAREST_BRIDGE = 0, LLDP_NEAREST_NON_TPMR_BRIDGE, LLDP_NEAREST_CUSTOMER_BRIDGE, @@ -3394,689 +3778,512 @@ enum lldp_agent_e { struct lldp_config_params_s { u32 config; -#define LLDP_CONFIG_TX_INTERVAL_MASK 0x000000ff -#define LLDP_CONFIG_TX_INTERVAL_SHIFT 0 -#define LLDP_CONFIG_HOLD_MASK 0x00000f00 -#define LLDP_CONFIG_HOLD_SHIFT 8 -#define LLDP_CONFIG_MAX_CREDIT_MASK 0x0000f000 -#define LLDP_CONFIG_MAX_CREDIT_SHIFT 12 -#define LLDP_CONFIG_ENABLE_RX_MASK 0x40000000 -#define LLDP_CONFIG_ENABLE_RX_SHIFT 30 -#define LLDP_CONFIG_ENABLE_TX_MASK 0x80000000 -#define LLDP_CONFIG_ENABLE_TX_SHIFT 31 - u32 local_chassis_id[LLDP_CHASSIS_ID_STAT_LEN]; - u32 local_port_id[LLDP_PORT_ID_STAT_LEN]; +#define LLDP_CONFIG_TX_INTERVAL_MASK 0x000000ff +#define LLDP_CONFIG_TX_INTERVAL_SHIFT 0 +#define LLDP_CONFIG_HOLD_MASK 0x00000f00 +#define LLDP_CONFIG_HOLD_SHIFT 8 +#define LLDP_CONFIG_MAX_CREDIT_MASK 0x0000f000 +#define LLDP_CONFIG_MAX_CREDIT_SHIFT 12 +#define LLDP_CONFIG_ENABLE_RX_MASK 0x40000000 +#define LLDP_CONFIG_ENABLE_RX_SHIFT 30 +#define LLDP_CONFIG_ENABLE_TX_MASK 0x80000000 +#define LLDP_CONFIG_ENABLE_TX_SHIFT 31 + u32 local_chassis_id[LLDP_CHASSIS_ID_STAT_LEN]; + u32 local_port_id[LLDP_PORT_ID_STAT_LEN]; }; struct lldp_status_params_s { - u32 prefix_seq_num; - u32 status; /* TBD */ - - /* Holds remote Chassis ID TLV header, subtype and 9B of payload. */ - u32 peer_chassis_id[LLDP_CHASSIS_ID_STAT_LEN]; - - /* Holds remote Port ID TLV header, subtype and 9B of payload. */ - u32 peer_port_id[LLDP_PORT_ID_STAT_LEN]; - u32 suffix_seq_num; + u32 prefix_seq_num; + u32 status; + u32 peer_chassis_id[LLDP_CHASSIS_ID_STAT_LEN]; + u32 peer_port_id[LLDP_PORT_ID_STAT_LEN]; + u32 suffix_seq_num; }; struct dcbx_ets_feature { u32 flags; -#define DCBX_ETS_ENABLED_MASK 0x00000001 -#define DCBX_ETS_ENABLED_SHIFT 0 -#define DCBX_ETS_WILLING_MASK 0x00000002 -#define DCBX_ETS_WILLING_SHIFT 1 -#define DCBX_ETS_ERROR_MASK 0x00000004 -#define DCBX_ETS_ERROR_SHIFT 2 -#define DCBX_ETS_CBS_MASK 0x00000008 -#define DCBX_ETS_CBS_SHIFT 3 -#define DCBX_ETS_MAX_TCS_MASK 0x000000f0 -#define DCBX_ETS_MAX_TCS_SHIFT 4 - u32 pri_tc_tbl[1]; -#define DCBX_ISCSI_OOO_TC 4 -#define NIG_ETS_ISCSI_OOO_CLIENT_OFFSET (DCBX_ISCSI_OOO_TC + 1) - u32 tc_bw_tbl[2]; - u32 tc_tsa_tbl[2]; -#define DCBX_ETS_TSA_STRICT 0 -#define DCBX_ETS_TSA_CBS 1 -#define DCBX_ETS_TSA_ETS 2 +#define DCBX_ETS_ENABLED_MASK 0x00000001 +#define DCBX_ETS_ENABLED_SHIFT 0 +#define DCBX_ETS_WILLING_MASK 0x00000002 +#define DCBX_ETS_WILLING_SHIFT 1 +#define DCBX_ETS_ERROR_MASK 0x00000004 +#define DCBX_ETS_ERROR_SHIFT 2 +#define DCBX_ETS_CBS_MASK 0x00000008 +#define DCBX_ETS_CBS_SHIFT 3 +#define DCBX_ETS_MAX_TCS_MASK 0x000000f0 +#define DCBX_ETS_MAX_TCS_SHIFT 4 +#define DCBX_ISCSI_OOO_TC_MASK 0x00000f00 +#define DCBX_ISCSI_OOO_TC_SHIFT 8 + u32 pri_tc_tbl[1]; +#define DCBX_ISCSI_OOO_TC (4) + +#define NIG_ETS_ISCSI_OOO_CLIENT_OFFSET (DCBX_ISCSI_OOO_TC + 1) +#define DCBX_CEE_STRICT_PRIORITY 0xf + u32 tc_bw_tbl[2]; + u32 tc_tsa_tbl[2]; +#define DCBX_ETS_TSA_STRICT 0 +#define DCBX_ETS_TSA_CBS 1 +#define DCBX_ETS_TSA_ETS 2 }; struct dcbx_app_priority_entry { u32 entry; -#define DCBX_APP_PRI_MAP_MASK 0x000000ff -#define DCBX_APP_PRI_MAP_SHIFT 0 -#define DCBX_APP_PRI_0 0x01 -#define DCBX_APP_PRI_1 0x02 -#define DCBX_APP_PRI_2 0x04 -#define DCBX_APP_PRI_3 0x08 -#define DCBX_APP_PRI_4 0x10 -#define DCBX_APP_PRI_5 0x20 -#define DCBX_APP_PRI_6 0x40 -#define DCBX_APP_PRI_7 0x80 -#define DCBX_APP_SF_MASK 0x00000300 -#define DCBX_APP_SF_SHIFT 8 -#define DCBX_APP_SF_ETHTYPE 0 -#define DCBX_APP_SF_PORT 1 -#define DCBX_APP_PROTOCOL_ID_MASK 0xffff0000 -#define DCBX_APP_PROTOCOL_ID_SHIFT 16 -}; - -/* FW structure in BE */ +#define DCBX_APP_PRI_MAP_MASK 0x000000ff +#define DCBX_APP_PRI_MAP_SHIFT 0 +#define DCBX_APP_PRI_0 0x01 +#define DCBX_APP_PRI_1 0x02 +#define DCBX_APP_PRI_2 0x04 +#define DCBX_APP_PRI_3 0x08 +#define DCBX_APP_PRI_4 0x10 +#define DCBX_APP_PRI_5 0x20 +#define DCBX_APP_PRI_6 0x40 +#define DCBX_APP_PRI_7 0x80 +#define DCBX_APP_SF_MASK 0x00000300 +#define DCBX_APP_SF_SHIFT 8 +#define DCBX_APP_SF_ETHTYPE 0 +#define DCBX_APP_SF_PORT 1 +#define DCBX_APP_PROTOCOL_ID_MASK 0xffff0000 +#define DCBX_APP_PROTOCOL_ID_SHIFT 16 +}; + struct dcbx_app_priority_feature { u32 flags; -#define DCBX_APP_ENABLED_MASK 0x00000001 -#define DCBX_APP_ENABLED_SHIFT 0 -#define DCBX_APP_WILLING_MASK 0x00000002 -#define DCBX_APP_WILLING_SHIFT 1 -#define DCBX_APP_ERROR_MASK 0x00000004 -#define DCBX_APP_ERROR_SHIFT 2 -/* Not in use - * #define DCBX_APP_DEFAULT_PRI_MASK 0x00000f00 - * #define DCBX_APP_DEFAULT_PRI_SHIFT 8 - */ -#define DCBX_APP_MAX_TCS_MASK 0x0000f000 -#define DCBX_APP_MAX_TCS_SHIFT 12 -#define DCBX_APP_NUM_ENTRIES_MASK 0x00ff0000 -#define DCBX_APP_NUM_ENTRIES_SHIFT 16 +#define DCBX_APP_ENABLED_MASK 0x00000001 +#define DCBX_APP_ENABLED_SHIFT 0 +#define DCBX_APP_WILLING_MASK 0x00000002 +#define DCBX_APP_WILLING_SHIFT 1 +#define DCBX_APP_ERROR_MASK 0x00000004 +#define DCBX_APP_ERROR_SHIFT 2 +#define DCBX_APP_MAX_TCS_MASK 0x0000f000 +#define DCBX_APP_MAX_TCS_SHIFT 12 +#define DCBX_APP_NUM_ENTRIES_MASK 0x00ff0000 +#define DCBX_APP_NUM_ENTRIES_SHIFT 16 struct dcbx_app_priority_entry app_pri_tbl[DCBX_MAX_APP_PROTOCOL]; }; -/* FW structure in BE */ struct dcbx_features { - /* PG feature */ struct dcbx_ets_feature ets; + u32 pfc; +#define DCBX_PFC_PRI_EN_BITMAP_MASK 0x000000ff +#define DCBX_PFC_PRI_EN_BITMAP_SHIFT 0 +#define DCBX_PFC_PRI_EN_BITMAP_PRI_0 0x01 +#define DCBX_PFC_PRI_EN_BITMAP_PRI_1 0x02 +#define DCBX_PFC_PRI_EN_BITMAP_PRI_2 0x04 +#define DCBX_PFC_PRI_EN_BITMAP_PRI_3 0x08 +#define DCBX_PFC_PRI_EN_BITMAP_PRI_4 0x10 +#define DCBX_PFC_PRI_EN_BITMAP_PRI_5 0x20 +#define DCBX_PFC_PRI_EN_BITMAP_PRI_6 0x40 +#define DCBX_PFC_PRI_EN_BITMAP_PRI_7 0x80 + +#define DCBX_PFC_FLAGS_MASK 0x0000ff00 +#define DCBX_PFC_FLAGS_SHIFT 8 +#define DCBX_PFC_CAPS_MASK 0x00000f00 +#define DCBX_PFC_CAPS_SHIFT 8 +#define DCBX_PFC_MBC_MASK 0x00004000 +#define DCBX_PFC_MBC_SHIFT 14 +#define DCBX_PFC_WILLING_MASK 0x00008000 +#define DCBX_PFC_WILLING_SHIFT 15 +#define DCBX_PFC_ENABLED_MASK 0x00010000 +#define DCBX_PFC_ENABLED_SHIFT 16 +#define DCBX_PFC_ERROR_MASK 0x00020000 +#define DCBX_PFC_ERROR_SHIFT 17 - /* PFC feature */ - u32 pfc; -#define DCBX_PFC_PRI_EN_BITMAP_MASK 0x000000ff -#define DCBX_PFC_PRI_EN_BITMAP_SHIFT 0 -#define DCBX_PFC_PRI_EN_BITMAP_PRI_0 0x01 -#define DCBX_PFC_PRI_EN_BITMAP_PRI_1 0x02 -#define DCBX_PFC_PRI_EN_BITMAP_PRI_2 0x04 -#define DCBX_PFC_PRI_EN_BITMAP_PRI_3 0x08 -#define DCBX_PFC_PRI_EN_BITMAP_PRI_4 0x10 -#define DCBX_PFC_PRI_EN_BITMAP_PRI_5 0x20 -#define DCBX_PFC_PRI_EN_BITMAP_PRI_6 0x40 -#define DCBX_PFC_PRI_EN_BITMAP_PRI_7 0x80 - -#define DCBX_PFC_FLAGS_MASK 0x0000ff00 -#define DCBX_PFC_FLAGS_SHIFT 8 -#define DCBX_PFC_CAPS_MASK 0x00000f00 -#define DCBX_PFC_CAPS_SHIFT 8 -#define DCBX_PFC_MBC_MASK 0x00004000 -#define DCBX_PFC_MBC_SHIFT 14 -#define DCBX_PFC_WILLING_MASK 0x00008000 -#define DCBX_PFC_WILLING_SHIFT 15 -#define DCBX_PFC_ENABLED_MASK 0x00010000 -#define DCBX_PFC_ENABLED_SHIFT 16 -#define DCBX_PFC_ERROR_MASK 0x00020000 -#define DCBX_PFC_ERROR_SHIFT 17 - - /* APP feature */ struct dcbx_app_priority_feature app; }; struct dcbx_local_params { u32 config; -#define DCBX_CONFIG_VERSION_MASK 0x00000003 -#define DCBX_CONFIG_VERSION_SHIFT 0 -#define DCBX_CONFIG_VERSION_DISABLED 0 -#define DCBX_CONFIG_VERSION_IEEE 1 -#define DCBX_CONFIG_VERSION_CEE 2 +#define DCBX_CONFIG_VERSION_MASK 0x00000007 +#define DCBX_CONFIG_VERSION_SHIFT 0 +#define DCBX_CONFIG_VERSION_DISABLED 0 +#define DCBX_CONFIG_VERSION_IEEE 1 +#define DCBX_CONFIG_VERSION_CEE 2 +#define DCBX_CONFIG_VERSION_STATIC 4 - u32 flags; - struct dcbx_features features; + u32 flags; + struct dcbx_features features; }; struct dcbx_mib { - u32 prefix_seq_num; - u32 flags; - struct dcbx_features features; - u32 suffix_seq_num; + u32 prefix_seq_num; + u32 flags; + struct dcbx_features features; + u32 suffix_seq_num; }; struct lldp_system_tlvs_buffer_s { - u16 valid; - u16 length; - u32 data[MAX_SYSTEM_LLDP_TLV_DATA]; + u16 valid; + u16 length; + u32 data[MAX_SYSTEM_LLDP_TLV_DATA]; }; -/**************************************/ -/* */ -/* P U B L I C G L O B A L */ -/* */ -/**************************************/ -struct public_global { - u32 max_path; -#define MAX_PATH_BIG_BEAR 2 -#define MAX_PATH_K2 1 - u32 max_ports; -#define MODE_1P 1 -#define MODE_2P 2 -#define MODE_3P 3 -#define MODE_4P 4 - u32 debug_mb_offset; - u32 phymod_dbg_mb_offset; - struct couple_mode_teaming cmt; - s32 internal_temperature; - u32 mfw_ver; - u32 running_bundle_id; +struct dcb_dscp_map { + u32 flags; +#define DCB_DSCP_ENABLE_MASK 0x1 +#define DCB_DSCP_ENABLE_SHIFT 0 +#define DCB_DSCP_ENABLE 1 + u32 dscp_pri_map[8]; }; -/**************************************/ -/* */ -/* P U B L I C P A T H */ -/* */ -/**************************************/ +struct public_global { + u32 max_path; + u32 max_ports; + u32 debug_mb_offset; + u32 phymod_dbg_mb_offset; + struct couple_mode_teaming cmt; + s32 internal_temperature; + u32 mfw_ver; + u32 running_bundle_id; + s32 external_temperature; + u32 mdump_reason; +}; -/**************************************************************************** -* Shared Memory 2 Region * -****************************************************************************/ -/* The fw_flr_ack is actually built in the following way: */ -/* 8 bit: PF ack */ -/* 128 bit: VF ack */ -/* 8 bit: ios_dis_ack */ -/* In order to maintain endianity in the mailbox hsi, we want to keep using */ -/* u32. The fw must have the VF right after the PF since this is how it */ -/* access arrays(it expects always the VF to reside after the PF, and that */ -/* makes the calculation much easier for it. ) */ -/* In order to answer both limitations, and keep the struct small, the code */ -/* will abuse the structure defined here to achieve the actual partition */ -/* above */ -/****************************************************************************/ struct fw_flr_mb { - u32 aggint; - u32 opgen_addr; - u32 accum_ack; /* 0..15:PF, 16..207:VF, 256..271:IOV_DIS */ -#define ACCUM_ACK_PF_BASE 0 -#define ACCUM_ACK_PF_SHIFT 0 - -#define ACCUM_ACK_VF_BASE 8 -#define ACCUM_ACK_VF_SHIFT 3 - -#define ACCUM_ACK_IOV_DIS_BASE 256 -#define ACCUM_ACK_IOV_DIS_SHIFT 8 + u32 aggint; + u32 opgen_addr; + u32 accum_ack; }; struct public_path { - struct fw_flr_mb flr_mb; - u32 mcp_vf_disabled[VF_MAX_STATIC / 32]; - - u32 process_kill; -#define PROCESS_KILL_COUNTER_MASK 0x0000ffff -#define PROCESS_KILL_COUNTER_SHIFT 0 -#define PROCESS_KILL_GLOB_AEU_BIT_MASK 0xffff0000 -#define PROCESS_KILL_GLOB_AEU_BIT_SHIFT 16 + struct fw_flr_mb flr_mb; + u32 mcp_vf_disabled[VF_MAX_STATIC / 32]; + + u32 process_kill; +#define PROCESS_KILL_COUNTER_MASK 0x0000ffff +#define PROCESS_KILL_COUNTER_SHIFT 0 +#define PROCESS_KILL_GLOB_AEU_BIT_MASK 0xffff0000 +#define PROCESS_KILL_GLOB_AEU_BIT_SHIFT 16 #define GLOBAL_AEU_BIT(aeu_reg_id, aeu_bit) (aeu_reg_id * 32 + aeu_bit) }; -/**************************************/ -/* */ -/* P U B L I C P O R T */ -/* */ -/**************************************/ - -/**************************************************************************** -* Driver <-> FW Mailbox * -****************************************************************************/ - struct public_port { - u32 validity_map; /* 0x0 (4*2 = 0x8) */ - - /* validity bits */ -#define MCP_VALIDITY_PCI_CFG 0x00100000 -#define MCP_VALIDITY_MB 0x00200000 -#define MCP_VALIDITY_DEV_INFO 0x00400000 -#define MCP_VALIDITY_RESERVED 0x00000007 - - /* One licensing bit should be set */ -#define MCP_VALIDITY_LIC_KEY_IN_EFFECT_MASK 0x00000038 -#define MCP_VALIDITY_LIC_MANUF_KEY_IN_EFFECT 0x00000008 -#define MCP_VALIDITY_LIC_UPGRADE_KEY_IN_EFFECT 0x00000010 -#define MCP_VALIDITY_LIC_NO_KEY_IN_EFFECT 0x00000020 - - /* Active MFW */ -#define MCP_VALIDITY_ACTIVE_MFW_UNKNOWN 0x00000000 -#define MCP_VALIDITY_ACTIVE_MFW_MASK 0x000001c0 -#define MCP_VALIDITY_ACTIVE_MFW_NCSI 0x00000040 -#define MCP_VALIDITY_ACTIVE_MFW_NONE 0x000001c0 + u32 validity_map; u32 link_status; -#define LINK_STATUS_LINK_UP \ - 0x00000001 -#define LINK_STATUS_SPEED_AND_DUPLEX_MASK 0x0000001e -#define LINK_STATUS_SPEED_AND_DUPLEX_1000THD BIT(1) -#define LINK_STATUS_SPEED_AND_DUPLEX_1000TFD (2 << 1) -#define LINK_STATUS_SPEED_AND_DUPLEX_10G (3 << 1) -#define LINK_STATUS_SPEED_AND_DUPLEX_20G (4 << 1) -#define LINK_STATUS_SPEED_AND_DUPLEX_40G (5 << 1) -#define LINK_STATUS_SPEED_AND_DUPLEX_50G (6 << 1) -#define LINK_STATUS_SPEED_AND_DUPLEX_100G (7 << 1) -#define LINK_STATUS_SPEED_AND_DUPLEX_25G (8 << 1) - -#define LINK_STATUS_AUTO_NEGOTIATE_ENABLED 0x00000020 - -#define LINK_STATUS_AUTO_NEGOTIATE_COMPLETE 0x00000040 -#define LINK_STATUS_PARALLEL_DETECTION_USED 0x00000080 - -#define LINK_STATUS_PFC_ENABLED \ - 0x00000100 -#define LINK_STATUS_LINK_PARTNER_1000TFD_CAPABLE 0x00000200 -#define LINK_STATUS_LINK_PARTNER_1000THD_CAPABLE 0x00000400 -#define LINK_STATUS_LINK_PARTNER_10G_CAPABLE 0x00000800 -#define LINK_STATUS_LINK_PARTNER_20G_CAPABLE 0x00001000 -#define LINK_STATUS_LINK_PARTNER_40G_CAPABLE 0x00002000 -#define LINK_STATUS_LINK_PARTNER_50G_CAPABLE 0x00004000 -#define LINK_STATUS_LINK_PARTNER_100G_CAPABLE 0x00008000 -#define LINK_STATUS_LINK_PARTNER_25G_CAPABLE 0x00010000 - -#define LINK_STATUS_LINK_PARTNER_FLOW_CONTROL_MASK 0x000C0000 -#define LINK_STATUS_LINK_PARTNER_NOT_PAUSE_CAPABLE (0 << 18) -#define LINK_STATUS_LINK_PARTNER_SYMMETRIC_PAUSE BIT(18) -#define LINK_STATUS_LINK_PARTNER_ASYMMETRIC_PAUSE (2 << 18) -#define LINK_STATUS_LINK_PARTNER_BOTH_PAUSE (3 << 18) - -#define LINK_STATUS_SFP_TX_FAULT \ - 0x00100000 -#define LINK_STATUS_TX_FLOW_CONTROL_ENABLED 0x00200000 -#define LINK_STATUS_RX_FLOW_CONTROL_ENABLED 0x00400000 - - u32 link_status1; - u32 ext_phy_fw_version; - u32 drv_phy_cfg_addr; - - u32 port_stx; - - u32 stat_nig_timer; - - struct port_mf_cfg port_mf_config; - struct port_stats stats; - - u32 media_type; -#define MEDIA_UNSPECIFIED 0x0 -#define MEDIA_SFPP_10G_FIBER 0x1 -#define MEDIA_XFP_FIBER 0x2 -#define MEDIA_DA_TWINAX 0x3 -#define MEDIA_BASE_T 0x4 -#define MEDIA_SFP_1G_FIBER 0x5 -#define MEDIA_KR 0xf0 -#define MEDIA_NOT_PRESENT 0xff +#define LINK_STATUS_LINK_UP 0x00000001 +#define LINK_STATUS_SPEED_AND_DUPLEX_MASK 0x0000001e +#define LINK_STATUS_SPEED_AND_DUPLEX_1000THD (1 << 1) +#define LINK_STATUS_SPEED_AND_DUPLEX_1000TFD (2 << 1) +#define LINK_STATUS_SPEED_AND_DUPLEX_10G (3 << 1) +#define LINK_STATUS_SPEED_AND_DUPLEX_20G (4 << 1) +#define LINK_STATUS_SPEED_AND_DUPLEX_40G (5 << 1) +#define LINK_STATUS_SPEED_AND_DUPLEX_50G (6 << 1) +#define LINK_STATUS_SPEED_AND_DUPLEX_100G (7 << 1) +#define LINK_STATUS_SPEED_AND_DUPLEX_25G (8 << 1) + +#define LINK_STATUS_AUTO_NEGOTIATE_ENABLED 0x00000020 + +#define LINK_STATUS_AUTO_NEGOTIATE_COMPLETE 0x00000040 +#define LINK_STATUS_PARALLEL_DETECTION_USED 0x00000080 + +#define LINK_STATUS_PFC_ENABLED 0x00000100 +#define LINK_STATUS_LINK_PARTNER_1000TFD_CAPABLE 0x00000200 +#define LINK_STATUS_LINK_PARTNER_1000THD_CAPABLE 0x00000400 +#define LINK_STATUS_LINK_PARTNER_10G_CAPABLE 0x00000800 +#define LINK_STATUS_LINK_PARTNER_20G_CAPABLE 0x00001000 +#define LINK_STATUS_LINK_PARTNER_40G_CAPABLE 0x00002000 +#define LINK_STATUS_LINK_PARTNER_50G_CAPABLE 0x00004000 +#define LINK_STATUS_LINK_PARTNER_100G_CAPABLE 0x00008000 +#define LINK_STATUS_LINK_PARTNER_25G_CAPABLE 0x00010000 + +#define LINK_STATUS_LINK_PARTNER_FLOW_CONTROL_MASK 0x000C0000 +#define LINK_STATUS_LINK_PARTNER_NOT_PAUSE_CAPABLE (0 << 18) +#define LINK_STATUS_LINK_PARTNER_SYMMETRIC_PAUSE (1 << 18) +#define LINK_STATUS_LINK_PARTNER_ASYMMETRIC_PAUSE (2 << 18) +#define LINK_STATUS_LINK_PARTNER_BOTH_PAUSE (3 << 18) + +#define LINK_STATUS_SFP_TX_FAULT 0x00100000 +#define LINK_STATUS_TX_FLOW_CONTROL_ENABLED 0x00200000 +#define LINK_STATUS_RX_FLOW_CONTROL_ENABLED 0x00400000 +#define LINK_STATUS_RX_SIGNAL_PRESENT 0x00800000 +#define LINK_STATUS_MAC_LOCAL_FAULT 0x01000000 +#define LINK_STATUS_MAC_REMOTE_FAULT 0x02000000 +#define LINK_STATUS_UNSUPPORTED_SPD_REQ 0x04000000 + + u32 link_status1; + u32 ext_phy_fw_version; + u32 drv_phy_cfg_addr; + + u32 port_stx; + + u32 stat_nig_timer; + + struct port_mf_cfg port_mf_config; + struct port_stats stats; + + u32 media_type; +#define MEDIA_UNSPECIFIED 0x0 +#define MEDIA_SFPP_10G_FIBER 0x1 +#define MEDIA_XFP_FIBER 0x2 +#define MEDIA_DA_TWINAX 0x3 +#define MEDIA_BASE_T 0x4 +#define MEDIA_SFP_1G_FIBER 0x5 +#define MEDIA_MODULE_FIBER 0x6 +#define MEDIA_KR 0xf0 +#define MEDIA_NOT_PRESENT 0xff u32 lfa_status; -#define LFA_LINK_FLAP_REASON_OFFSET 0 -#define LFA_LINK_FLAP_REASON_MASK 0x000000ff -#define LFA_NO_REASON (0 << 0) -#define LFA_LINK_DOWN BIT(0) -#define LFA_FORCE_INIT BIT(1) -#define LFA_LOOPBACK_MISMATCH BIT(2) -#define LFA_SPEED_MISMATCH BIT(3) -#define LFA_FLOW_CTRL_MISMATCH BIT(4) -#define LFA_ADV_SPEED_MISMATCH BIT(5) -#define LINK_FLAP_AVOIDANCE_COUNT_OFFSET 8 -#define LINK_FLAP_AVOIDANCE_COUNT_MASK 0x0000ff00 -#define LINK_FLAP_COUNT_OFFSET 16 -#define LINK_FLAP_COUNT_MASK 0x00ff0000 - - u32 link_change_count; - - /* LLDP params */ - struct lldp_config_params_s lldp_config_params[ - LLDP_MAX_LLDP_AGENTS]; - struct lldp_status_params_s lldp_status_params[ - LLDP_MAX_LLDP_AGENTS]; - struct lldp_system_tlvs_buffer_s system_lldp_tlvs_buf; + u32 link_change_count; + + struct lldp_config_params_s lldp_config_params[LLDP_MAX_LLDP_AGENTS]; + struct lldp_status_params_s lldp_status_params[LLDP_MAX_LLDP_AGENTS]; + struct lldp_system_tlvs_buffer_s system_lldp_tlvs_buf; /* DCBX related MIB */ - struct dcbx_local_params local_admin_dcbx_mib; - struct dcbx_mib remote_dcbx_mib; - struct dcbx_mib operational_dcbx_mib; + struct dcbx_local_params local_admin_dcbx_mib; + struct dcbx_mib remote_dcbx_mib; + struct dcbx_mib operational_dcbx_mib; - u32 fc_npiv_nvram_tbl_addr; - u32 fc_npiv_nvram_tbl_size; - u32 transceiver_data; -#define PMM_TRANSCEIVER_STATE_MASK 0x000000FF -#define PMM_TRANSCEIVER_STATE_SHIFT 0x00000000 -#define PMM_TRANSCEIVER_STATE_PRESENT 0x00000001 -}; + u32 reserved[2]; + u32 transceiver_data; +#define ETH_TRANSCEIVER_STATE_MASK 0x000000FF +#define ETH_TRANSCEIVER_STATE_SHIFT 0x00000000 +#define ETH_TRANSCEIVER_STATE_UNPLUGGED 0x00000000 +#define ETH_TRANSCEIVER_STATE_PRESENT 0x00000001 +#define ETH_TRANSCEIVER_STATE_VALID 0x00000003 +#define ETH_TRANSCEIVER_STATE_UPDATING 0x00000008 -/**************************************/ -/* */ -/* P U B L I C F U N C */ -/* */ -/**************************************/ + u32 wol_info; + u32 wol_pkt_len; + u32 wol_pkt_details; + struct dcb_dscp_map dcb_dscp_map; +}; struct public_func { - u32 iscsi_boot_signature; - u32 iscsi_boot_block_offset; - - u32 mtu_size; - u32 c2s_pcp_map_lower; - u32 c2s_pcp_map_upper; - u32 c2s_pcp_map_default; - u32 reserved[4]; - - u32 config; - - /* E/R/I/D */ - /* function 0 of each port cannot be hidden */ -#define FUNC_MF_CFG_FUNC_HIDE 0x00000001 -#define FUNC_MF_CFG_PAUSE_ON_HOST_RING 0x00000002 -#define FUNC_MF_CFG_PAUSE_ON_HOST_RING_SHIFT 0x00000001 - -#define FUNC_MF_CFG_PROTOCOL_MASK 0x000000f0 -#define FUNC_MF_CFG_PROTOCOL_SHIFT 4 -#define FUNC_MF_CFG_PROTOCOL_ETHERNET 0x00000000 -#define FUNC_MF_CFG_PROTOCOL_ISCSI 0x00000010 -#define FUNC_MF_CFG_PROTOCOL_FCOE 0x00000020 -#define FUNC_MF_CFG_PROTOCOL_ROCE 0x00000030 -#define FUNC_MF_CFG_PROTOCOL_MAX 0x00000030 - - /* MINBW, MAXBW */ - /* value range - 0..100, increments in 1 % */ -#define FUNC_MF_CFG_MIN_BW_MASK 0x0000ff00 -#define FUNC_MF_CFG_MIN_BW_SHIFT 8 -#define FUNC_MF_CFG_MIN_BW_DEFAULT 0x00000000 -#define FUNC_MF_CFG_MAX_BW_MASK 0x00ff0000 -#define FUNC_MF_CFG_MAX_BW_SHIFT 16 -#define FUNC_MF_CFG_MAX_BW_DEFAULT 0x00640000 - - u32 status; -#define FUNC_STATUS_VLINK_DOWN 0x00000001 - - u32 mac_upper; /* MAC */ -#define FUNC_MF_CFG_UPPERMAC_MASK 0x0000ffff -#define FUNC_MF_CFG_UPPERMAC_SHIFT 0 -#define FUNC_MF_CFG_UPPERMAC_DEFAULT FUNC_MF_CFG_UPPERMAC_MASK - u32 mac_lower; -#define FUNC_MF_CFG_LOWERMAC_DEFAULT 0xffffffff - - u32 fcoe_wwn_port_name_upper; - u32 fcoe_wwn_port_name_lower; - - u32 fcoe_wwn_node_name_upper; - u32 fcoe_wwn_node_name_lower; - - u32 ovlan_stag; /* tags */ -#define FUNC_MF_CFG_OV_STAG_MASK 0x0000ffff -#define FUNC_MF_CFG_OV_STAG_SHIFT 0 -#define FUNC_MF_CFG_OV_STAG_DEFAULT FUNC_MF_CFG_OV_STAG_MASK - - u32 pf_allocation; /* vf per pf */ - - u32 preserve_data; /* Will be used bt CCM */ - - u32 driver_last_activity_ts; - - u32 drv_ack_vf_disabled[VF_MAX_STATIC / 32]; /* 0x0044 */ - - u32 drv_id; -#define DRV_ID_PDA_COMP_VER_MASK 0x0000ffff -#define DRV_ID_PDA_COMP_VER_SHIFT 0 - -#define DRV_ID_MCP_HSI_VER_MASK 0x00ff0000 -#define DRV_ID_MCP_HSI_VER_SHIFT 16 -#define DRV_ID_MCP_HSI_VER_CURRENT BIT(DRV_ID_MCP_HSI_VER_SHIFT) - -#define DRV_ID_DRV_TYPE_MASK 0x7f000000 -#define DRV_ID_DRV_TYPE_SHIFT 24 -#define DRV_ID_DRV_TYPE_UNKNOWN (0 << DRV_ID_DRV_TYPE_SHIFT) -#define DRV_ID_DRV_TYPE_LINUX (1 << DRV_ID_DRV_TYPE_SHIFT) -#define DRV_ID_DRV_TYPE_WINDOWS (2 << DRV_ID_DRV_TYPE_SHIFT) -#define DRV_ID_DRV_TYPE_DIAG (3 << DRV_ID_DRV_TYPE_SHIFT) -#define DRV_ID_DRV_TYPE_PREBOOT (4 << DRV_ID_DRV_TYPE_SHIFT) -#define DRV_ID_DRV_TYPE_SOLARIS (5 << DRV_ID_DRV_TYPE_SHIFT) -#define DRV_ID_DRV_TYPE_VMWARE (6 << DRV_ID_DRV_TYPE_SHIFT) -#define DRV_ID_DRV_TYPE_FREEBSD (7 << DRV_ID_DRV_TYPE_SHIFT) -#define DRV_ID_DRV_TYPE_AIX (8 << DRV_ID_DRV_TYPE_SHIFT) - -#define DRV_ID_DRV_INIT_HW_MASK 0x80000000 -#define DRV_ID_DRV_INIT_HW_SHIFT 31 -#define DRV_ID_DRV_INIT_HW_FLAG BIT(DRV_ID_DRV_INIT_HW_SHIFT) -}; + u32 reserved0[2]; -/**************************************/ -/* */ -/* P U B L I C M B */ -/* */ -/**************************************/ -/* This is the only section that the driver can write to, and each */ -/* Basically each driver request to set feature parameters, - * will be done using a different command, which will be linked - * to a specific data structure from the union below. - * For huge strucuture, the common blank structure should be used. - */ + u32 mtu_size; + + u32 reserved[7]; + + u32 config; +#define FUNC_MF_CFG_FUNC_HIDE 0x00000001 +#define FUNC_MF_CFG_PAUSE_ON_HOST_RING 0x00000002 +#define FUNC_MF_CFG_PAUSE_ON_HOST_RING_SHIFT 0x00000001 + +#define FUNC_MF_CFG_PROTOCOL_MASK 0x000000f0 +#define FUNC_MF_CFG_PROTOCOL_SHIFT 4 +#define FUNC_MF_CFG_PROTOCOL_ETHERNET 0x00000000 +#define FUNC_MF_CFG_PROTOCOL_MAX 0x00000030 + +#define FUNC_MF_CFG_MIN_BW_MASK 0x0000ff00 +#define FUNC_MF_CFG_MIN_BW_SHIFT 8 +#define FUNC_MF_CFG_MIN_BW_DEFAULT 0x00000000 +#define FUNC_MF_CFG_MAX_BW_MASK 0x00ff0000 +#define FUNC_MF_CFG_MAX_BW_SHIFT 16 +#define FUNC_MF_CFG_MAX_BW_DEFAULT 0x00640000 + + u32 status; +#define FUNC_STATUS_VLINK_DOWN 0x00000001 + + u32 mac_upper; +#define FUNC_MF_CFG_UPPERMAC_MASK 0x0000ffff +#define FUNC_MF_CFG_UPPERMAC_SHIFT 0 +#define FUNC_MF_CFG_UPPERMAC_DEFAULT FUNC_MF_CFG_UPPERMAC_MASK + u32 mac_lower; +#define FUNC_MF_CFG_LOWERMAC_DEFAULT 0xffffffff + + u32 fcoe_wwn_port_name_upper; + u32 fcoe_wwn_port_name_lower; + + u32 fcoe_wwn_node_name_upper; + u32 fcoe_wwn_node_name_lower; + + u32 ovlan_stag; +#define FUNC_MF_CFG_OV_STAG_MASK 0x0000ffff +#define FUNC_MF_CFG_OV_STAG_SHIFT 0 +#define FUNC_MF_CFG_OV_STAG_DEFAULT FUNC_MF_CFG_OV_STAG_MASK + + u32 pf_allocation; + + u32 preserve_data; + + u32 driver_last_activity_ts; + + u32 drv_ack_vf_disabled[VF_MAX_STATIC / 32]; + + u32 drv_id; +#define DRV_ID_PDA_COMP_VER_MASK 0x0000ffff +#define DRV_ID_PDA_COMP_VER_SHIFT 0 + +#define DRV_ID_MCP_HSI_VER_MASK 0x00ff0000 +#define DRV_ID_MCP_HSI_VER_SHIFT 16 +#define DRV_ID_MCP_HSI_VER_CURRENT (1 << DRV_ID_MCP_HSI_VER_SHIFT) + +#define DRV_ID_DRV_TYPE_MASK 0x7f000000 +#define DRV_ID_DRV_TYPE_SHIFT 24 +#define DRV_ID_DRV_TYPE_UNKNOWN (0 << DRV_ID_DRV_TYPE_SHIFT) +#define DRV_ID_DRV_TYPE_LINUX (1 << DRV_ID_DRV_TYPE_SHIFT) + +#define DRV_ID_DRV_INIT_HW_MASK 0x80000000 +#define DRV_ID_DRV_INIT_HW_SHIFT 31 +#define DRV_ID_DRV_INIT_HW_FLAG (1 << DRV_ID_DRV_INIT_HW_SHIFT) +}; struct mcp_mac { - u32 mac_upper; /* Upper 16 bits are always zeroes */ - u32 mac_lower; + u32 mac_upper; + u32 mac_lower; }; struct mcp_val64 { - u32 lo; - u32 hi; + u32 lo; + u32 hi; }; struct mcp_file_att { - u32 nvm_start_addr; - u32 len; + u32 nvm_start_addr; + u32 len; +}; + +struct bist_nvm_image_att { + u32 return_code; + u32 image_type; + u32 nvm_start_addr; + u32 len; }; #define MCP_DRV_VER_STR_SIZE 16 #define MCP_DRV_VER_STR_SIZE_DWORD (MCP_DRV_VER_STR_SIZE / sizeof(u32)) #define MCP_DRV_NVM_BUF_LEN 32 struct drv_version_stc { - u32 version; - u8 name[MCP_DRV_VER_STR_SIZE - 4]; + u32 version; + u8 name[MCP_DRV_VER_STR_SIZE - 4]; +}; + +struct lan_stats_stc { + u64 ucast_rx_pkts; + u64 ucast_tx_pkts; + u32 fcs_err; + u32 rserved; +}; + +struct ocbb_data_stc { + u32 ocbb_host_addr; + u32 ocsd_host_addr; + u32 ocsd_req_update_interval; +}; + +#define MAX_NUM_OF_SENSORS 7 +struct temperature_status_stc { + u32 num_of_sensors; + u32 sensor[MAX_NUM_OF_SENSORS]; +}; + +/* crash dump configuration header */ +struct mdump_config_stc { + u32 version; + u32 config; + u32 epoc; + u32 num_of_logs; + u32 valid_logs; }; union drv_union_data { - u32 ver_str[MCP_DRV_VER_STR_SIZE_DWORD]; - struct mcp_mac wol_mac; + u32 ver_str[MCP_DRV_VER_STR_SIZE_DWORD]; + struct mcp_mac wol_mac; + + struct eth_phy_cfg drv_phy_cfg; - struct pmm_phy_cfg drv_phy_cfg; + struct mcp_val64 val64; - struct mcp_val64 val64; /* For PHY / AVS commands */ + u8 raw_data[MCP_DRV_NVM_BUF_LEN]; - u8 raw_data[MCP_DRV_NVM_BUF_LEN]; + struct mcp_file_att file_att; - struct mcp_file_att file_att; + u32 ack_vf_disabled[VF_MAX_STATIC / 32]; - u32 ack_vf_disabled[VF_MAX_STATIC / 32]; + struct drv_version_stc drv_version; - struct drv_version_stc drv_version; + struct lan_stats_stc lan_stats; + u64 reserved_stats[11]; + struct ocbb_data_stc ocbb_info; + struct temperature_status_stc temp_info; + struct bist_nvm_image_att nvm_image_att; + struct mdump_config_stc mdump_config; }; struct public_drv_mb { u32 drv_mb_header; -#define DRV_MSG_CODE_MASK 0xffff0000 -#define DRV_MSG_CODE_LOAD_REQ 0x10000000 -#define DRV_MSG_CODE_LOAD_DONE 0x11000000 -#define DRV_MSG_CODE_INIT_HW 0x12000000 -#define DRV_MSG_CODE_UNLOAD_REQ 0x20000000 -#define DRV_MSG_CODE_UNLOAD_DONE 0x21000000 -#define DRV_MSG_CODE_INIT_PHY 0x22000000 - /* Params - FORCE - Reinitialize the link regardless of LFA */ - /* - DONT_CARE - Don't flap the link if up */ -#define DRV_MSG_CODE_LINK_RESET 0x23000000 - -#define DRV_MSG_CODE_SET_LLDP 0x24000000 -#define DRV_MSG_CODE_SET_DCBX 0x25000000 +#define DRV_MSG_CODE_MASK 0xffff0000 +#define DRV_MSG_CODE_LOAD_REQ 0x10000000 +#define DRV_MSG_CODE_LOAD_DONE 0x11000000 +#define DRV_MSG_CODE_INIT_HW 0x12000000 +#define DRV_MSG_CODE_UNLOAD_REQ 0x20000000 +#define DRV_MSG_CODE_UNLOAD_DONE 0x21000000 +#define DRV_MSG_CODE_INIT_PHY 0x22000000 +#define DRV_MSG_CODE_LINK_RESET 0x23000000 +#define DRV_MSG_CODE_SET_DCBX 0x25000000 + #define DRV_MSG_CODE_BW_UPDATE_ACK 0x32000000 -#define DRV_MSG_CODE_NIG_DRAIN 0x30000000 - -#define DRV_MSG_CODE_INITIATE_FLR 0x02000000 -#define DRV_MSG_CODE_VF_DISABLED_DONE 0xc0000000 -#define DRV_MSG_CODE_CFG_VF_MSIX 0xc0010000 -#define DRV_MSG_CODE_NVM_PUT_FILE_BEGIN 0x00010000 -#define DRV_MSG_CODE_NVM_PUT_FILE_DATA 0x00020000 -#define DRV_MSG_CODE_NVM_GET_FILE_ATT 0x00030000 -#define DRV_MSG_CODE_NVM_READ_NVRAM 0x00050000 -#define DRV_MSG_CODE_NVM_WRITE_NVRAM 0x00060000 -#define DRV_MSG_CODE_NVM_DEL_FILE 0x00080000 -#define DRV_MSG_CODE_MCP_RESET 0x00090000 -#define DRV_MSG_CODE_SET_SECURE_MODE 0x000a0000 -#define DRV_MSG_CODE_PHY_RAW_READ 0x000b0000 -#define DRV_MSG_CODE_PHY_RAW_WRITE 0x000c0000 -#define DRV_MSG_CODE_PHY_CORE_READ 0x000d0000 -#define DRV_MSG_CODE_PHY_CORE_WRITE 0x000e0000 -#define DRV_MSG_CODE_SET_VERSION 0x000f0000 - -#define DRV_MSG_CODE_BIST_TEST 0x001e0000 -#define DRV_MSG_CODE_SET_LED_MODE 0x00200000 - -#define DRV_MSG_SEQ_NUMBER_MASK 0x0000ffff +#define DRV_MSG_CODE_NIG_DRAIN 0x30000000 +#define DRV_MSG_CODE_VF_DISABLED_DONE 0xc0000000 +#define DRV_MSG_CODE_CFG_VF_MSIX 0xc0010000 +#define DRV_MSG_CODE_MCP_RESET 0x00090000 +#define DRV_MSG_CODE_SET_VERSION 0x000f0000 + +#define DRV_MSG_CODE_BIST_TEST 0x001e0000 +#define DRV_MSG_CODE_SET_LED_MODE 0x00200000 + +#define DRV_MSG_SEQ_NUMBER_MASK 0x0000ffff u32 drv_mb_param; +#define DRV_MB_PARAM_UNLOAD_WOL_MCP 0x00000001 +#define DRV_MB_PARAM_DCBX_NOTIFY_MASK 0x000000FF +#define DRV_MB_PARAM_DCBX_NOTIFY_SHIFT 3 +#define DRV_MB_PARAM_CFG_VF_MSIX_VF_ID_SHIFT 0 +#define DRV_MB_PARAM_CFG_VF_MSIX_VF_ID_MASK 0x000000FF +#define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_SHIFT 8 +#define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_MASK 0x0000FF00 + +#define DRV_MB_PARAM_SET_LED_MODE_OPER 0x0 +#define DRV_MB_PARAM_SET_LED_MODE_ON 0x1 +#define DRV_MB_PARAM_SET_LED_MODE_OFF 0x2 - /* UNLOAD_REQ params */ -#define DRV_MB_PARAM_UNLOAD_WOL_UNKNOWN 0x00000000 -#define DRV_MB_PARAM_UNLOAD_WOL_MCP 0x00000001 -#define DRV_MB_PARAM_UNLOAD_WOL_DISABLED 0x00000002 -#define DRV_MB_PARAM_UNLOAD_WOL_ENABLED 0x00000003 - - /* UNLOAD_DONE_params */ -#define DRV_MB_PARAM_UNLOAD_NON_D3_POWER 0x00000001 - - /* INIT_PHY params */ -#define DRV_MB_PARAM_INIT_PHY_FORCE 0x00000001 -#define DRV_MB_PARAM_INIT_PHY_DONT_CARE 0x00000002 - - /* LLDP / DCBX params*/ -#define DRV_MB_PARAM_LLDP_SEND_MASK 0x00000001 -#define DRV_MB_PARAM_LLDP_SEND_SHIFT 0 -#define DRV_MB_PARAM_LLDP_AGENT_MASK 0x00000006 -#define DRV_MB_PARAM_LLDP_AGENT_SHIFT 1 -#define DRV_MB_PARAM_DCBX_NOTIFY_MASK 0x00000008 -#define DRV_MB_PARAM_DCBX_NOTIFY_SHIFT 3 - -#define DRV_MB_PARAM_NIG_DRAIN_PERIOD_MS_MASK 0x000000FF -#define DRV_MB_PARAM_NIG_DRAIN_PERIOD_MS_SHIFT 0 - -#define DRV_MB_PARAM_NVM_PUT_FILE_BEGIN_MFW 0x1 -#define DRV_MB_PARAM_NVM_PUT_FILE_BEGIN_IMAGE 0x2 - -#define DRV_MB_PARAM_NVM_OFFSET_SHIFT 0 -#define DRV_MB_PARAM_NVM_OFFSET_MASK 0x00FFFFFF -#define DRV_MB_PARAM_NVM_LEN_SHIFT 24 -#define DRV_MB_PARAM_NVM_LEN_MASK 0xFF000000 - -#define DRV_MB_PARAM_PHY_ADDR_SHIFT 0 -#define DRV_MB_PARAM_PHY_ADDR_MASK 0x1FF0FFFF -#define DRV_MB_PARAM_PHY_LANE_SHIFT 16 -#define DRV_MB_PARAM_PHY_LANE_MASK 0x000F0000 -#define DRV_MB_PARAM_PHY_SELECT_PORT_SHIFT 29 -#define DRV_MB_PARAM_PHY_SELECT_PORT_MASK 0x20000000 -#define DRV_MB_PARAM_PHY_PORT_SHIFT 30 -#define DRV_MB_PARAM_PHY_PORT_MASK 0xc0000000 - -/* configure vf MSIX params*/ -#define DRV_MB_PARAM_CFG_VF_MSIX_VF_ID_SHIFT 0 -#define DRV_MB_PARAM_CFG_VF_MSIX_VF_ID_MASK 0x000000FF -#define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_SHIFT 8 -#define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_MASK 0x0000FF00 - -#define DRV_MB_PARAM_SET_LED_MODE_OPER 0x0 -#define DRV_MB_PARAM_SET_LED_MODE_ON 0x1 -#define DRV_MB_PARAM_SET_LED_MODE_OFF 0x2 - -#define DRV_MB_PARAM_BIST_UNKNOWN_TEST 0 -#define DRV_MB_PARAM_BIST_REGISTER_TEST 1 -#define DRV_MB_PARAM_BIST_CLOCK_TEST 2 - -#define DRV_MB_PARAM_BIST_RC_UNKNOWN 0 -#define DRV_MB_PARAM_BIST_RC_PASSED 1 -#define DRV_MB_PARAM_BIST_RC_FAILED 2 -#define DRV_MB_PARAM_BIST_RC_INVALID_PARAMETER 3 - -#define DRV_MB_PARAM_BIST_TEST_INDEX_SHIFT 0 -#define DRV_MB_PARAM_BIST_TEST_INDEX_MASK 0x000000FF +#define DRV_MB_PARAM_BIST_REGISTER_TEST 1 +#define DRV_MB_PARAM_BIST_CLOCK_TEST 2 + +#define DRV_MB_PARAM_BIST_RC_UNKNOWN 0 +#define DRV_MB_PARAM_BIST_RC_PASSED 1 +#define DRV_MB_PARAM_BIST_RC_FAILED 2 +#define DRV_MB_PARAM_BIST_RC_INVALID_PARAMETER 3 + +#define DRV_MB_PARAM_BIST_TEST_INDEX_SHIFT 0 +#define DRV_MB_PARAM_BIST_TEST_INDEX_MASK 0x000000FF u32 fw_mb_header; -#define FW_MSG_CODE_MASK 0xffff0000 -#define FW_MSG_CODE_DRV_LOAD_ENGINE 0x10100000 -#define FW_MSG_CODE_DRV_LOAD_PORT 0x10110000 -#define FW_MSG_CODE_DRV_LOAD_FUNCTION 0x10120000 -#define FW_MSG_CODE_DRV_LOAD_REFUSED_PDA 0x10200000 -#define FW_MSG_CODE_DRV_LOAD_REFUSED_HSI 0x10210000 -#define FW_MSG_CODE_DRV_LOAD_REFUSED_DIAG 0x10220000 -#define FW_MSG_CODE_DRV_LOAD_DONE 0x11100000 -#define FW_MSG_CODE_DRV_UNLOAD_ENGINE 0x20110000 -#define FW_MSG_CODE_DRV_UNLOAD_PORT 0x20120000 -#define FW_MSG_CODE_DRV_UNLOAD_FUNCTION 0x20130000 -#define FW_MSG_CODE_DRV_UNLOAD_DONE 0x21100000 -#define FW_MSG_CODE_INIT_PHY_DONE 0x21200000 -#define FW_MSG_CODE_INIT_PHY_ERR_INVALID_ARGS 0x21300000 -#define FW_MSG_CODE_LINK_RESET_DONE 0x23000000 -#define FW_MSG_CODE_SET_LLDP_DONE 0x24000000 -#define FW_MSG_CODE_SET_LLDP_UNSUPPORTED_AGENT 0x24010000 -#define FW_MSG_CODE_SET_DCBX_DONE 0x25000000 -#define FW_MSG_CODE_NIG_DRAIN_DONE 0x30000000 -#define FW_MSG_CODE_VF_DISABLED_DONE 0xb0000000 -#define FW_MSG_CODE_DRV_CFG_VF_MSIX_DONE 0xb0010000 -#define FW_MSG_CODE_FLR_ACK 0x02000000 -#define FW_MSG_CODE_FLR_NACK 0x02100000 - -#define FW_MSG_CODE_NVM_OK 0x00010000 -#define FW_MSG_CODE_NVM_INVALID_MODE 0x00020000 -#define FW_MSG_CODE_NVM_PREV_CMD_WAS_NOT_FINISHED 0x00030000 -#define FW_MSG_CODE_NVM_FAILED_TO_ALLOCATE_PAGE 0x00040000 -#define FW_MSG_CODE_NVM_INVALID_DIR_FOUND 0x00050000 -#define FW_MSG_CODE_NVM_PAGE_NOT_FOUND 0x00060000 -#define FW_MSG_CODE_NVM_FAILED_PARSING_BNDLE_HEADER 0x00070000 -#define FW_MSG_CODE_NVM_FAILED_PARSING_IMAGE_HEADER 0x00080000 -#define FW_MSG_CODE_NVM_PARSING_OUT_OF_SYNC 0x00090000 -#define FW_MSG_CODE_NVM_FAILED_UPDATING_DIR 0x000a0000 -#define FW_MSG_CODE_NVM_FAILED_TO_FREE_PAGE 0x000b0000 -#define FW_MSG_CODE_NVM_FILE_NOT_FOUND 0x000c0000 -#define FW_MSG_CODE_NVM_OPERATION_FAILED 0x000d0000 -#define FW_MSG_CODE_NVM_FAILED_UNALIGNED 0x000e0000 -#define FW_MSG_CODE_NVM_BAD_OFFSET 0x000f0000 -#define FW_MSG_CODE_NVM_BAD_SIGNATURE 0x00100000 -#define FW_MSG_CODE_NVM_FILE_READ_ONLY 0x00200000 -#define FW_MSG_CODE_NVM_UNKNOWN_FILE 0x00300000 -#define FW_MSG_CODE_NVM_PUT_FILE_FINISH_OK 0x00400000 -#define FW_MSG_CODE_MCP_RESET_REJECT 0x00600000 -#define FW_MSG_CODE_PHY_OK 0x00110000 -#define FW_MSG_CODE_PHY_ERROR 0x00120000 -#define FW_MSG_CODE_SET_SECURE_MODE_ERROR 0x00130000 -#define FW_MSG_CODE_SET_SECURE_MODE_OK 0x00140000 -#define FW_MSG_MODE_PHY_PRIVILEGE_ERROR 0x00150000 -#define FW_MSG_CODE_OK 0x00160000 - -#define FW_MSG_SEQ_NUMBER_MASK 0x0000ffff - - u32 fw_mb_param; - - u32 drv_pulse_mb; -#define DRV_PULSE_SEQ_MASK 0x00007fff -#define DRV_PULSE_SYSTEM_TIME_MASK 0xffff0000 -#define DRV_PULSE_ALWAYS_ALIVE 0x00008000 +#define FW_MSG_CODE_MASK 0xffff0000 +#define FW_MSG_CODE_DRV_LOAD_ENGINE 0x10100000 +#define FW_MSG_CODE_DRV_LOAD_PORT 0x10110000 +#define FW_MSG_CODE_DRV_LOAD_FUNCTION 0x10120000 +#define FW_MSG_CODE_DRV_LOAD_REFUSED_PDA 0x10200000 +#define FW_MSG_CODE_DRV_LOAD_REFUSED_HSI 0x10210000 +#define FW_MSG_CODE_DRV_LOAD_REFUSED_DIAG 0x10220000 +#define FW_MSG_CODE_DRV_LOAD_DONE 0x11100000 +#define FW_MSG_CODE_DRV_UNLOAD_ENGINE 0x20110000 +#define FW_MSG_CODE_DRV_UNLOAD_PORT 0x20120000 +#define FW_MSG_CODE_DRV_UNLOAD_FUNCTION 0x20130000 +#define FW_MSG_CODE_DRV_UNLOAD_DONE 0x21100000 +#define FW_MSG_CODE_DRV_CFG_VF_MSIX_DONE 0xb0010000 +#define FW_MSG_CODE_OK 0x00160000 + +#define FW_MSG_SEQ_NUMBER_MASK 0x0000ffff + + u32 fw_mb_param; + + u32 drv_pulse_mb; +#define DRV_PULSE_SEQ_MASK 0x00007fff +#define DRV_PULSE_SYSTEM_TIME_MASK 0xffff0000 +#define DRV_PULSE_ALWAYS_ALIVE 0x00008000 + u32 mcp_pulse_mb; -#define MCP_PULSE_SEQ_MASK 0x00007fff -#define MCP_PULSE_ALWAYS_ALIVE 0x00008000 -#define MCP_EVENT_MASK 0xffff0000 -#define MCP_EVENT_OTHER_DRIVER_RESET_REQ 0x00010000 +#define MCP_PULSE_SEQ_MASK 0x00007fff +#define MCP_PULSE_ALWAYS_ALIVE 0x00008000 +#define MCP_EVENT_MASK 0xffff0000 +#define MCP_EVENT_OTHER_DRIVER_RESET_REQ 0x00010000 union drv_union_data union_data; }; -/* MFW - DRV MB */ -/********************************************************************** -* Description -* Incremental Aggregative -* 8-bit MFW counter per message -* 8-bit ack-counter per message -* Capabilities -* Provides up to 256 aggregative message per type -* Provides 4 message types in dword -* Message type pointers to byte offset -* Backward Compatibility by using sizeof for the counters. -* No lock requires for 32bit messages -* Limitations: -* In case of messages greater than 32bit, a dedicated mechanism(e.g lock) -* is required to prevent data corruption. -**********************************************************************/ enum MFW_DRV_MSG_TYPE { MFW_DRV_MSG_LINK_CHANGE, MFW_DRV_MSG_FLR_FW_ACK_FAILED, @@ -4084,37 +4291,33 @@ enum MFW_DRV_MSG_TYPE { MFW_DRV_MSG_LLDP_DATA_UPDATED, MFW_DRV_MSG_DCBX_REMOTE_MIB_UPDATED, MFW_DRV_MSG_DCBX_OPERATIONAL_MIB_UPDATED, - MFW_DRV_MSG_ERROR_RECOVERY, + MFW_DRV_MSG_RESERVED4, MFW_DRV_MSG_BW_UPDATE, - MFW_DRV_MSG_S_TAG_UPDATE, - MFW_DRV_MSG_GET_LAN_STATS, - MFW_DRV_MSG_GET_FCOE_STATS, - MFW_DRV_MSG_GET_ISCSI_STATS, - MFW_DRV_MSG_GET_RDMA_STATS, - MFW_DRV_MSG_FAILURE_DETECTED, + MFW_DRV_MSG_BW_UPDATE5, + MFW_DRV_MSG_BW_UPDATE6, + MFW_DRV_MSG_BW_UPDATE7, + MFW_DRV_MSG_BW_UPDATE8, + MFW_DRV_MSG_BW_UPDATE9, + MFW_DRV_MSG_BW_UPDATE10, MFW_DRV_MSG_TRANSCEIVER_STATE_CHANGE, + MFW_DRV_MSG_BW_UPDATE11, MFW_DRV_MSG_MAX }; -#define MFW_DRV_MSG_MAX_DWORDS(msgs) (((msgs - 1) >> 2) + 1) -#define MFW_DRV_MSG_DWORD(msg_id) (msg_id >> 2) -#define MFW_DRV_MSG_OFFSET(msg_id) ((msg_id & 0x3) << 3) -#define MFW_DRV_MSG_MASK(msg_id) (0xff << MFW_DRV_MSG_OFFSET(msg_id)) +#define MFW_DRV_MSG_MAX_DWORDS(msgs) (((msgs - 1) >> 2) + 1) +#define MFW_DRV_MSG_DWORD(msg_id) (msg_id >> 2) +#define MFW_DRV_MSG_OFFSET(msg_id) ((msg_id & 0x3) << 3) +#define MFW_DRV_MSG_MASK(msg_id) (0xff << MFW_DRV_MSG_OFFSET(msg_id)) struct public_mfw_mb { - u32 sup_msgs; - u32 msg[MFW_DRV_MSG_MAX_DWORDS(MFW_DRV_MSG_MAX)]; - u32 ack[MFW_DRV_MSG_MAX_DWORDS(MFW_DRV_MSG_MAX)]; + u32 sup_msgs; + u32 msg[MFW_DRV_MSG_MAX_DWORDS(MFW_DRV_MSG_MAX)]; + u32 ack[MFW_DRV_MSG_MAX_DWORDS(MFW_DRV_MSG_MAX)]; }; -/**************************************/ -/* */ -/* P U B L I C D A T A */ -/* */ -/**************************************/ enum public_sections { - PUBLIC_DRV_MB, /* Points to the first drv_mb of path0 */ - PUBLIC_MFW_MB, /* Points to the first mfw_mb of path0 */ + PUBLIC_DRV_MB, + PUBLIC_MFW_MB, PUBLIC_GLOBAL, PUBLIC_PATH, PUBLIC_PORT, @@ -4122,1080 +4325,177 @@ enum public_sections { PUBLIC_MAX_SECTIONS }; -struct drv_ver_info_stc { - u32 ver; - u8 name[32]; -}; - struct mcp_public_data { - /* The sections fields is an array */ - u32 num_sections; - offsize_t sections[PUBLIC_MAX_SECTIONS]; - struct public_drv_mb drv_mb[MCP_GLOB_FUNC_MAX]; - struct public_mfw_mb mfw_mb[MCP_GLOB_FUNC_MAX]; - struct public_global global; - struct public_path path[MCP_GLOB_PATH_MAX]; - struct public_port port[MCP_GLOB_PORT_MAX]; - struct public_func func[MCP_GLOB_FUNC_MAX]; - struct drv_ver_info_stc drv_info; + u32 num_sections; + u32 sections[PUBLIC_MAX_SECTIONS]; + struct public_drv_mb drv_mb[MCP_GLOB_FUNC_MAX]; + struct public_mfw_mb mfw_mb[MCP_GLOB_FUNC_MAX]; + struct public_global global; + struct public_path path[MCP_GLOB_PATH_MAX]; + struct public_port port[MCP_GLOB_PORT_MAX]; + struct public_func func[MCP_GLOB_FUNC_MAX]; }; struct nvm_cfg_mac_address { - u32 mac_addr_hi; -#define NVM_CFG_MAC_ADDRESS_HI_MASK 0x0000FFFF -#define NVM_CFG_MAC_ADDRESS_HI_OFFSET 0 - - u32 mac_addr_lo; + u32 mac_addr_hi; +#define NVM_CFG_MAC_ADDRESS_HI_MASK 0x0000FFFF +#define NVM_CFG_MAC_ADDRESS_HI_OFFSET 0 + u32 mac_addr_lo; }; -/****************************************** -* nvm_cfg1 structs -******************************************/ - struct nvm_cfg1_glob { - u32 generic_cont0; /* 0x0 */ -#define NVM_CFG1_GLOB_BOARD_SWAP_MASK 0x0000000F -#define NVM_CFG1_GLOB_BOARD_SWAP_OFFSET 0 -#define NVM_CFG1_GLOB_BOARD_SWAP_NONE 0x0 -#define NVM_CFG1_GLOB_BOARD_SWAP_PATH 0x1 -#define NVM_CFG1_GLOB_BOARD_SWAP_PORT 0x2 -#define NVM_CFG1_GLOB_BOARD_SWAP_BOTH 0x3 -#define NVM_CFG1_GLOB_MF_MODE_MASK 0x00000FF0 -#define NVM_CFG1_GLOB_MF_MODE_OFFSET 4 -#define NVM_CFG1_GLOB_MF_MODE_MF_ALLOWED 0x0 -#define NVM_CFG1_GLOB_MF_MODE_DEFAULT 0x1 -#define NVM_CFG1_GLOB_MF_MODE_SPIO4 0x2 -#define NVM_CFG1_GLOB_MF_MODE_NPAR1_0 0x3 -#define NVM_CFG1_GLOB_MF_MODE_NPAR1_5 0x4 -#define NVM_CFG1_GLOB_MF_MODE_NPAR2_0 0x5 -#define NVM_CFG1_GLOB_MF_MODE_BD 0x6 -#define NVM_CFG1_GLOB_MF_MODE_UFP 0x7 -#define NVM_CFG1_GLOB_FAN_FAILURE_ENFORCEMENT_MASK 0x00001000 -#define NVM_CFG1_GLOB_FAN_FAILURE_ENFORCEMENT_OFFSET 12 -#define NVM_CFG1_GLOB_FAN_FAILURE_ENFORCEMENT_DISABLED 0x0 -#define NVM_CFG1_GLOB_FAN_FAILURE_ENFORCEMENT_ENABLED 0x1 -#define NVM_CFG1_GLOB_AVS_MARGIN_LOW_MASK 0x001FE000 -#define NVM_CFG1_GLOB_AVS_MARGIN_LOW_OFFSET 13 -#define NVM_CFG1_GLOB_AVS_MARGIN_HIGH_MASK 0x1FE00000 -#define NVM_CFG1_GLOB_AVS_MARGIN_HIGH_OFFSET 21 -#define NVM_CFG1_GLOB_ENABLE_SRIOV_MASK 0x20000000 -#define NVM_CFG1_GLOB_ENABLE_SRIOV_OFFSET 29 -#define NVM_CFG1_GLOB_ENABLE_SRIOV_DISABLED 0x0 -#define NVM_CFG1_GLOB_ENABLE_SRIOV_ENABLED 0x1 -#define NVM_CFG1_GLOB_ENABLE_ATC_MASK 0x40000000 -#define NVM_CFG1_GLOB_ENABLE_ATC_OFFSET 30 -#define NVM_CFG1_GLOB_ENABLE_ATC_DISABLED 0x0 -#define NVM_CFG1_GLOB_ENABLE_ATC_ENABLED 0x1 -#define NVM_CFG1_GLOB_CLOCK_SLOWDOWN_MASK 0x80000000 -#define NVM_CFG1_GLOB_CLOCK_SLOWDOWN_OFFSET 31 -#define NVM_CFG1_GLOB_CLOCK_SLOWDOWN_DISABLED 0x0 -#define NVM_CFG1_GLOB_CLOCK_SLOWDOWN_ENABLED 0x1 - - u32 engineering_change[3]; /* 0x4 */ - - u32 manufacturing_id; /* 0x10 */ - - u32 serial_number[4]; /* 0x14 */ - - u32 pcie_cfg; /* 0x24 */ -#define NVM_CFG1_GLOB_PCI_GEN_MASK 0x00000003 -#define NVM_CFG1_GLOB_PCI_GEN_OFFSET 0 -#define NVM_CFG1_GLOB_PCI_GEN_PCI_GEN1 0x0 -#define NVM_CFG1_GLOB_PCI_GEN_PCI_GEN2 0x1 -#define NVM_CFG1_GLOB_PCI_GEN_PCI_GEN3 0x2 -#define NVM_CFG1_GLOB_BEACON_WOL_ENABLED_MASK 0x00000004 -#define NVM_CFG1_GLOB_BEACON_WOL_ENABLED_OFFSET 2 -#define NVM_CFG1_GLOB_BEACON_WOL_ENABLED_DISABLED 0x0 -#define NVM_CFG1_GLOB_BEACON_WOL_ENABLED_ENABLED 0x1 -#define NVM_CFG1_GLOB_ASPM_SUPPORT_MASK 0x00000018 -#define NVM_CFG1_GLOB_ASPM_SUPPORT_OFFSET 3 -#define NVM_CFG1_GLOB_ASPM_SUPPORT_L0S_L1_ENABLED 0x0 -#define NVM_CFG1_GLOB_ASPM_SUPPORT_L0S_DISABLED 0x1 -#define NVM_CFG1_GLOB_ASPM_SUPPORT_L1_DISABLED 0x2 -#define NVM_CFG1_GLOB_ASPM_SUPPORT_L0S_L1_DISABLED 0x3 -#define NVM_CFG1_GLOB_PREVENT_PCIE_L1_MENTRY_MASK 0x00000020 -#define NVM_CFG1_GLOB_PREVENT_PCIE_L1_MENTRY_OFFSET 5 -#define NVM_CFG1_GLOB_PREVENT_PCIE_L1_MENTRY_DISABLED 0x0 -#define NVM_CFG1_GLOB_PREVENT_PCIE_L1_MENTRY_ENABLED 0x1 -#define NVM_CFG1_GLOB_PCIE_G2_TX_AMPLITUDE_MASK 0x000003C0 -#define NVM_CFG1_GLOB_PCIE_G2_TX_AMPLITUDE_OFFSET 6 -#define NVM_CFG1_GLOB_PCIE_PREEMPHASIS_MASK 0x00001C00 -#define NVM_CFG1_GLOB_PCIE_PREEMPHASIS_OFFSET 10 -#define NVM_CFG1_GLOB_PCIE_PREEMPHASIS_HW 0x0 -#define NVM_CFG1_GLOB_PCIE_PREEMPHASIS_0DB 0x1 -#define NVM_CFG1_GLOB_PCIE_PREEMPHASIS_3_5DB 0x2 -#define NVM_CFG1_GLOB_PCIE_PREEMPHASIS_6_0DB 0x3 -#define NVM_CFG1_GLOB_WWN_NODE_PREFIX0_MASK 0x001FE000 -#define NVM_CFG1_GLOB_WWN_NODE_PREFIX0_OFFSET 13 -#define NVM_CFG1_GLOB_WWN_NODE_PREFIX1_MASK 0x1FE00000 -#define NVM_CFG1_GLOB_WWN_NODE_PREFIX1_OFFSET 21 -#define NVM_CFG1_GLOB_NCSI_PACKAGE_ID_MASK 0x60000000 -#define NVM_CFG1_GLOB_NCSI_PACKAGE_ID_OFFSET 29 - - u32 mgmt_traffic; /* 0x28 */ -#define NVM_CFG1_GLOB_RESERVED60_MASK 0x00000001 -#define NVM_CFG1_GLOB_RESERVED60_OFFSET 0 -#define NVM_CFG1_GLOB_RESERVED60_100KHZ 0x0 -#define NVM_CFG1_GLOB_RESERVED60_400KHZ 0x1 -#define NVM_CFG1_GLOB_WWN_PORT_PREFIX0_MASK 0x000001FE -#define NVM_CFG1_GLOB_WWN_PORT_PREFIX0_OFFSET 1 -#define NVM_CFG1_GLOB_WWN_PORT_PREFIX1_MASK 0x0001FE00 -#define NVM_CFG1_GLOB_WWN_PORT_PREFIX1_OFFSET 9 -#define NVM_CFG1_GLOB_SMBUS_ADDRESS_MASK 0x01FE0000 -#define NVM_CFG1_GLOB_SMBUS_ADDRESS_OFFSET 17 -#define NVM_CFG1_GLOB_SIDEBAND_MODE_MASK 0x06000000 -#define NVM_CFG1_GLOB_SIDEBAND_MODE_OFFSET 25 -#define NVM_CFG1_GLOB_SIDEBAND_MODE_DISABLED 0x0 -#define NVM_CFG1_GLOB_SIDEBAND_MODE_RMII 0x1 -#define NVM_CFG1_GLOB_SIDEBAND_MODE_SGMII 0x2 - - u32 core_cfg; /* 0x2C */ -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_MASK 0x000000FF -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_OFFSET 0 -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_2X40G 0x0 -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_2X50G 0x1 -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_1X100G 0x2 -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_4X10G_F 0x3 -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_4X10G_E 0x4 -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_4X20G 0x5 -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_1X40G 0xB -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_2X25G 0xC -#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_DE_1X25G 0xD -#define NVM_CFG1_GLOB_EAGLE_ENFORCE_TX_FIR_CFG_MASK 0x00000100 -#define NVM_CFG1_GLOB_EAGLE_ENFORCE_TX_FIR_CFG_OFFSET 8 -#define NVM_CFG1_GLOB_EAGLE_ENFORCE_TX_FIR_CFG_DISABLED 0x0 -#define NVM_CFG1_GLOB_EAGLE_ENFORCE_TX_FIR_CFG_ENABLED 0x1 -#define NVM_CFG1_GLOB_FALCON_ENFORCE_TX_FIR_CFG_MASK 0x00000200 -#define NVM_CFG1_GLOB_FALCON_ENFORCE_TX_FIR_CFG_OFFSET 9 -#define NVM_CFG1_GLOB_FALCON_ENFORCE_TX_FIR_CFG_DISABLED 0x0 -#define NVM_CFG1_GLOB_FALCON_ENFORCE_TX_FIR_CFG_ENABLED 0x1 -#define NVM_CFG1_GLOB_EAGLE_CORE_ADDR_MASK 0x0003FC00 -#define NVM_CFG1_GLOB_EAGLE_CORE_ADDR_OFFSET 10 -#define NVM_CFG1_GLOB_FALCON_CORE_ADDR_MASK 0x03FC0000 -#define NVM_CFG1_GLOB_FALCON_CORE_ADDR_OFFSET 18 -#define NVM_CFG1_GLOB_AVS_MODE_MASK 0x1C000000 -#define NVM_CFG1_GLOB_AVS_MODE_OFFSET 26 -#define NVM_CFG1_GLOB_AVS_MODE_CLOSE_LOOP 0x0 -#define NVM_CFG1_GLOB_AVS_MODE_OPEN_LOOP 0x1 -#define NVM_CFG1_GLOB_AVS_MODE_DISABLED 0x3 -#define NVM_CFG1_GLOB_OVERRIDE_SECURE_MODE_MASK 0x60000000 -#define NVM_CFG1_GLOB_OVERRIDE_SECURE_MODE_OFFSET 29 -#define NVM_CFG1_GLOB_OVERRIDE_SECURE_MODE_DISABLED 0x0 -#define NVM_CFG1_GLOB_OVERRIDE_SECURE_MODE_ENABLED 0x1 - - u32 e_lane_cfg1; /* 0x30 */ -#define NVM_CFG1_GLOB_RX_LANE0_SWAP_MASK 0x0000000F -#define NVM_CFG1_GLOB_RX_LANE0_SWAP_OFFSET 0 -#define NVM_CFG1_GLOB_RX_LANE1_SWAP_MASK 0x000000F0 -#define NVM_CFG1_GLOB_RX_LANE1_SWAP_OFFSET 4 -#define NVM_CFG1_GLOB_RX_LANE2_SWAP_MASK 0x00000F00 -#define NVM_CFG1_GLOB_RX_LANE2_SWAP_OFFSET 8 -#define NVM_CFG1_GLOB_RX_LANE3_SWAP_MASK 0x0000F000 -#define NVM_CFG1_GLOB_RX_LANE3_SWAP_OFFSET 12 -#define NVM_CFG1_GLOB_TX_LANE0_SWAP_MASK 0x000F0000 -#define NVM_CFG1_GLOB_TX_LANE0_SWAP_OFFSET 16 -#define NVM_CFG1_GLOB_TX_LANE1_SWAP_MASK 0x00F00000 -#define NVM_CFG1_GLOB_TX_LANE1_SWAP_OFFSET 20 -#define NVM_CFG1_GLOB_TX_LANE2_SWAP_MASK 0x0F000000 -#define NVM_CFG1_GLOB_TX_LANE2_SWAP_OFFSET 24 -#define NVM_CFG1_GLOB_TX_LANE3_SWAP_MASK 0xF0000000 -#define NVM_CFG1_GLOB_TX_LANE3_SWAP_OFFSET 28 - - u32 e_lane_cfg2; /* 0x34 */ -#define NVM_CFG1_GLOB_RX_LANE0_POL_FLIP_MASK 0x00000001 -#define NVM_CFG1_GLOB_RX_LANE0_POL_FLIP_OFFSET 0 -#define NVM_CFG1_GLOB_RX_LANE1_POL_FLIP_MASK 0x00000002 -#define NVM_CFG1_GLOB_RX_LANE1_POL_FLIP_OFFSET 1 -#define NVM_CFG1_GLOB_RX_LANE2_POL_FLIP_MASK 0x00000004 -#define NVM_CFG1_GLOB_RX_LANE2_POL_FLIP_OFFSET 2 -#define NVM_CFG1_GLOB_RX_LANE3_POL_FLIP_MASK 0x00000008 -#define NVM_CFG1_GLOB_RX_LANE3_POL_FLIP_OFFSET 3 -#define NVM_CFG1_GLOB_TX_LANE0_POL_FLIP_MASK 0x00000010 -#define NVM_CFG1_GLOB_TX_LANE0_POL_FLIP_OFFSET 4 -#define NVM_CFG1_GLOB_TX_LANE1_POL_FLIP_MASK 0x00000020 -#define NVM_CFG1_GLOB_TX_LANE1_POL_FLIP_OFFSET 5 -#define NVM_CFG1_GLOB_TX_LANE2_POL_FLIP_MASK 0x00000040 -#define NVM_CFG1_GLOB_TX_LANE2_POL_FLIP_OFFSET 6 -#define NVM_CFG1_GLOB_TX_LANE3_POL_FLIP_MASK 0x00000080 -#define NVM_CFG1_GLOB_TX_LANE3_POL_FLIP_OFFSET 7 -#define NVM_CFG1_GLOB_SMBUS_MODE_MASK 0x00000F00 -#define NVM_CFG1_GLOB_SMBUS_MODE_OFFSET 8 -#define NVM_CFG1_GLOB_SMBUS_MODE_DISABLED 0x0 -#define NVM_CFG1_GLOB_SMBUS_MODE_100KHZ 0x1 -#define NVM_CFG1_GLOB_SMBUS_MODE_400KHZ 0x2 -#define NVM_CFG1_GLOB_NCSI_MASK 0x0000F000 -#define NVM_CFG1_GLOB_NCSI_OFFSET 12 -#define NVM_CFG1_GLOB_NCSI_DISABLED 0x0 -#define NVM_CFG1_GLOB_NCSI_ENABLED 0x1 - - u32 f_lane_cfg1; /* 0x38 */ -#define NVM_CFG1_GLOB_RX_LANE0_SWAP_MASK 0x0000000F -#define NVM_CFG1_GLOB_RX_LANE0_SWAP_OFFSET 0 -#define NVM_CFG1_GLOB_RX_LANE1_SWAP_MASK 0x000000F0 -#define NVM_CFG1_GLOB_RX_LANE1_SWAP_OFFSET 4 -#define NVM_CFG1_GLOB_RX_LANE2_SWAP_MASK 0x00000F00 -#define NVM_CFG1_GLOB_RX_LANE2_SWAP_OFFSET 8 -#define NVM_CFG1_GLOB_RX_LANE3_SWAP_MASK 0x0000F000 -#define NVM_CFG1_GLOB_RX_LANE3_SWAP_OFFSET 12 -#define NVM_CFG1_GLOB_TX_LANE0_SWAP_MASK 0x000F0000 -#define NVM_CFG1_GLOB_TX_LANE0_SWAP_OFFSET 16 -#define NVM_CFG1_GLOB_TX_LANE1_SWAP_MASK 0x00F00000 -#define NVM_CFG1_GLOB_TX_LANE1_SWAP_OFFSET 20 -#define NVM_CFG1_GLOB_TX_LANE2_SWAP_MASK 0x0F000000 -#define NVM_CFG1_GLOB_TX_LANE2_SWAP_OFFSET 24 -#define NVM_CFG1_GLOB_TX_LANE3_SWAP_MASK 0xF0000000 -#define NVM_CFG1_GLOB_TX_LANE3_SWAP_OFFSET 28 - - u32 f_lane_cfg2; /* 0x3C */ -#define NVM_CFG1_GLOB_RX_LANE0_POL_FLIP_MASK 0x00000001 -#define NVM_CFG1_GLOB_RX_LANE0_POL_FLIP_OFFSET 0 -#define NVM_CFG1_GLOB_RX_LANE1_POL_FLIP_MASK 0x00000002 -#define NVM_CFG1_GLOB_RX_LANE1_POL_FLIP_OFFSET 1 -#define NVM_CFG1_GLOB_RX_LANE2_POL_FLIP_MASK 0x00000004 -#define NVM_CFG1_GLOB_RX_LANE2_POL_FLIP_OFFSET 2 -#define NVM_CFG1_GLOB_RX_LANE3_POL_FLIP_MASK 0x00000008 -#define NVM_CFG1_GLOB_RX_LANE3_POL_FLIP_OFFSET 3 -#define NVM_CFG1_GLOB_TX_LANE0_POL_FLIP_MASK 0x00000010 -#define NVM_CFG1_GLOB_TX_LANE0_POL_FLIP_OFFSET 4 -#define NVM_CFG1_GLOB_TX_LANE1_POL_FLIP_MASK 0x00000020 -#define NVM_CFG1_GLOB_TX_LANE1_POL_FLIP_OFFSET 5 -#define NVM_CFG1_GLOB_TX_LANE2_POL_FLIP_MASK 0x00000040 -#define NVM_CFG1_GLOB_TX_LANE2_POL_FLIP_OFFSET 6 -#define NVM_CFG1_GLOB_TX_LANE3_POL_FLIP_MASK 0x00000080 -#define NVM_CFG1_GLOB_TX_LANE3_POL_FLIP_OFFSET 7 - - u32 eagle_preemphasis; /* 0x40 */ -#define NVM_CFG1_GLOB_LANE0_PREEMP_MASK 0x000000FF -#define NVM_CFG1_GLOB_LANE0_PREEMP_OFFSET 0 -#define NVM_CFG1_GLOB_LANE1_PREEMP_MASK 0x0000FF00 -#define NVM_CFG1_GLOB_LANE1_PREEMP_OFFSET 8 -#define NVM_CFG1_GLOB_LANE2_PREEMP_MASK 0x00FF0000 -#define NVM_CFG1_GLOB_LANE2_PREEMP_OFFSET 16 -#define NVM_CFG1_GLOB_LANE3_PREEMP_MASK 0xFF000000 -#define NVM_CFG1_GLOB_LANE3_PREEMP_OFFSET 24 - - u32 eagle_driver_current; /* 0x44 */ -#define NVM_CFG1_GLOB_LANE0_AMP_MASK 0x000000FF -#define NVM_CFG1_GLOB_LANE0_AMP_OFFSET 0 -#define NVM_CFG1_GLOB_LANE1_AMP_MASK 0x0000FF00 -#define NVM_CFG1_GLOB_LANE1_AMP_OFFSET 8 -#define NVM_CFG1_GLOB_LANE2_AMP_MASK 0x00FF0000 -#define NVM_CFG1_GLOB_LANE2_AMP_OFFSET 16 -#define NVM_CFG1_GLOB_LANE3_AMP_MASK 0xFF000000 -#define NVM_CFG1_GLOB_LANE3_AMP_OFFSET 24 - - u32 falcon_preemphasis; /* 0x48 */ -#define NVM_CFG1_GLOB_LANE0_PREEMP_MASK 0x000000FF -#define NVM_CFG1_GLOB_LANE0_PREEMP_OFFSET 0 -#define NVM_CFG1_GLOB_LANE1_PREEMP_MASK 0x0000FF00 -#define NVM_CFG1_GLOB_LANE1_PREEMP_OFFSET 8 -#define NVM_CFG1_GLOB_LANE2_PREEMP_MASK 0x00FF0000 -#define NVM_CFG1_GLOB_LANE2_PREEMP_OFFSET 16 -#define NVM_CFG1_GLOB_LANE3_PREEMP_MASK 0xFF000000 -#define NVM_CFG1_GLOB_LANE3_PREEMP_OFFSET 24 - - u32 falcon_driver_current; /* 0x4C */ -#define NVM_CFG1_GLOB_LANE0_AMP_MASK 0x000000FF -#define NVM_CFG1_GLOB_LANE0_AMP_OFFSET 0 -#define NVM_CFG1_GLOB_LANE1_AMP_MASK 0x0000FF00 -#define NVM_CFG1_GLOB_LANE1_AMP_OFFSET 8 -#define NVM_CFG1_GLOB_LANE2_AMP_MASK 0x00FF0000 -#define NVM_CFG1_GLOB_LANE2_AMP_OFFSET 16 -#define NVM_CFG1_GLOB_LANE3_AMP_MASK 0xFF000000 -#define NVM_CFG1_GLOB_LANE3_AMP_OFFSET 24 - - u32 pci_id; /* 0x50 */ -#define NVM_CFG1_GLOB_VENDOR_ID_MASK 0x0000FFFF -#define NVM_CFG1_GLOB_VENDOR_ID_OFFSET 0 - - u32 pci_subsys_id; /* 0x54 */ -#define NVM_CFG1_GLOB_SUBSYSTEM_VENDOR_ID_MASK 0x0000FFFF -#define NVM_CFG1_GLOB_SUBSYSTEM_VENDOR_ID_OFFSET 0 -#define NVM_CFG1_GLOB_SUBSYSTEM_DEVICE_ID_MASK 0xFFFF0000 -#define NVM_CFG1_GLOB_SUBSYSTEM_DEVICE_ID_OFFSET 16 - - u32 bar; /* 0x58 */ -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_MASK 0x0000000F -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_OFFSET 0 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_DISABLED 0x0 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_2K 0x1 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_4K 0x2 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_8K 0x3 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_16K 0x4 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_32K 0x5 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_64K 0x6 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_128K 0x7 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_256K 0x8 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_512K 0x9 -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_1M 0xA -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_2M 0xB -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_4M 0xC -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_8M 0xD -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_16M 0xE -#define NVM_CFG1_GLOB_EXPANSION_ROM_SIZE_32M 0xF -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_MASK 0x000000F0 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_OFFSET 4 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_DISABLED 0x0 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_4K 0x1 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_8K 0x2 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_16K 0x3 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_32K 0x4 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_64K 0x5 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_128K 0x6 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_256K 0x7 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_512K 0x8 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_1M 0x9 -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_2M 0xA -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_4M 0xB -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_8M 0xC -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_16M 0xD -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_32M 0xE -#define NVM_CFG1_GLOB_VF_PCI_BAR2_SIZE_64M 0xF -#define NVM_CFG1_GLOB_BAR2_SIZE_MASK 0x00000F00 -#define NVM_CFG1_GLOB_BAR2_SIZE_OFFSET 8 -#define NVM_CFG1_GLOB_BAR2_SIZE_DISABLED 0x0 -#define NVM_CFG1_GLOB_BAR2_SIZE_64K 0x1 -#define NVM_CFG1_GLOB_BAR2_SIZE_128K 0x2 -#define NVM_CFG1_GLOB_BAR2_SIZE_256K 0x3 -#define NVM_CFG1_GLOB_BAR2_SIZE_512K 0x4 -#define NVM_CFG1_GLOB_BAR2_SIZE_1M 0x5 -#define NVM_CFG1_GLOB_BAR2_SIZE_2M 0x6 -#define NVM_CFG1_GLOB_BAR2_SIZE_4M 0x7 -#define NVM_CFG1_GLOB_BAR2_SIZE_8M 0x8 -#define NVM_CFG1_GLOB_BAR2_SIZE_16M 0x9 -#define NVM_CFG1_GLOB_BAR2_SIZE_32M 0xA -#define NVM_CFG1_GLOB_BAR2_SIZE_64M 0xB -#define NVM_CFG1_GLOB_BAR2_SIZE_128M 0xC -#define NVM_CFG1_GLOB_BAR2_SIZE_256M 0xD -#define NVM_CFG1_GLOB_BAR2_SIZE_512M 0xE -#define NVM_CFG1_GLOB_BAR2_SIZE_1G 0xF - - u32 eagle_txfir_main; /* 0x5C */ -#define NVM_CFG1_GLOB_LANE0_TXFIR_MAIN_MASK 0x000000FF -#define NVM_CFG1_GLOB_LANE0_TXFIR_MAIN_OFFSET 0 -#define NVM_CFG1_GLOB_LANE1_TXFIR_MAIN_MASK 0x0000FF00 -#define NVM_CFG1_GLOB_LANE1_TXFIR_MAIN_OFFSET 8 -#define NVM_CFG1_GLOB_LANE2_TXFIR_MAIN_MASK 0x00FF0000 -#define NVM_CFG1_GLOB_LANE2_TXFIR_MAIN_OFFSET 16 -#define NVM_CFG1_GLOB_LANE3_TXFIR_MAIN_MASK 0xFF000000 -#define NVM_CFG1_GLOB_LANE3_TXFIR_MAIN_OFFSET 24 - - u32 eagle_txfir_post; /* 0x60 */ -#define NVM_CFG1_GLOB_LANE0_TXFIR_POST_MASK 0x000000FF -#define NVM_CFG1_GLOB_LANE0_TXFIR_POST_OFFSET 0 -#define NVM_CFG1_GLOB_LANE1_TXFIR_POST_MASK 0x0000FF00 -#define NVM_CFG1_GLOB_LANE1_TXFIR_POST_OFFSET 8 -#define NVM_CFG1_GLOB_LANE2_TXFIR_POST_MASK 0x00FF0000 -#define NVM_CFG1_GLOB_LANE2_TXFIR_POST_OFFSET 16 -#define NVM_CFG1_GLOB_LANE3_TXFIR_POST_MASK 0xFF000000 -#define NVM_CFG1_GLOB_LANE3_TXFIR_POST_OFFSET 24 - - u32 falcon_txfir_main; /* 0x64 */ -#define NVM_CFG1_GLOB_LANE0_TXFIR_MAIN_MASK 0x000000FF -#define NVM_CFG1_GLOB_LANE0_TXFIR_MAIN_OFFSET 0 -#define NVM_CFG1_GLOB_LANE1_TXFIR_MAIN_MASK 0x0000FF00 -#define NVM_CFG1_GLOB_LANE1_TXFIR_MAIN_OFFSET 8 -#define NVM_CFG1_GLOB_LANE2_TXFIR_MAIN_MASK 0x00FF0000 -#define NVM_CFG1_GLOB_LANE2_TXFIR_MAIN_OFFSET 16 -#define NVM_CFG1_GLOB_LANE3_TXFIR_MAIN_MASK 0xFF000000 -#define NVM_CFG1_GLOB_LANE3_TXFIR_MAIN_OFFSET 24 - - u32 falcon_txfir_post; /* 0x68 */ -#define NVM_CFG1_GLOB_LANE0_TXFIR_POST_MASK 0x000000FF -#define NVM_CFG1_GLOB_LANE0_TXFIR_POST_OFFSET 0 -#define NVM_CFG1_GLOB_LANE1_TXFIR_POST_MASK 0x0000FF00 -#define NVM_CFG1_GLOB_LANE1_TXFIR_POST_OFFSET 8 -#define NVM_CFG1_GLOB_LANE2_TXFIR_POST_MASK 0x00FF0000 -#define NVM_CFG1_GLOB_LANE2_TXFIR_POST_OFFSET 16 -#define NVM_CFG1_GLOB_LANE3_TXFIR_POST_MASK 0xFF000000 -#define NVM_CFG1_GLOB_LANE3_TXFIR_POST_OFFSET 24 - - u32 manufacture_ver; /* 0x6C */ -#define NVM_CFG1_GLOB_MANUF0_VER_MASK 0x0000003F -#define NVM_CFG1_GLOB_MANUF0_VER_OFFSET 0 -#define NVM_CFG1_GLOB_MANUF1_VER_MASK 0x00000FC0 -#define NVM_CFG1_GLOB_MANUF1_VER_OFFSET 6 -#define NVM_CFG1_GLOB_MANUF2_VER_MASK 0x0003F000 -#define NVM_CFG1_GLOB_MANUF2_VER_OFFSET 12 -#define NVM_CFG1_GLOB_MANUF3_VER_MASK 0x00FC0000 -#define NVM_CFG1_GLOB_MANUF3_VER_OFFSET 18 -#define NVM_CFG1_GLOB_MANUF4_VER_MASK 0x3F000000 -#define NVM_CFG1_GLOB_MANUF4_VER_OFFSET 24 - - u32 manufacture_time; /* 0x70 */ -#define NVM_CFG1_GLOB_MANUF0_TIME_MASK 0x0000003F -#define NVM_CFG1_GLOB_MANUF0_TIME_OFFSET 0 -#define NVM_CFG1_GLOB_MANUF1_TIME_MASK 0x00000FC0 -#define NVM_CFG1_GLOB_MANUF1_TIME_OFFSET 6 -#define NVM_CFG1_GLOB_MANUF2_TIME_MASK 0x0003F000 -#define NVM_CFG1_GLOB_MANUF2_TIME_OFFSET 12 - - u32 led_global_settings; /* 0x74 */ -#define NVM_CFG1_GLOB_LED_SWAP_0_MASK 0x0000000F -#define NVM_CFG1_GLOB_LED_SWAP_0_OFFSET 0 -#define NVM_CFG1_GLOB_LED_SWAP_1_MASK 0x000000F0 -#define NVM_CFG1_GLOB_LED_SWAP_1_OFFSET 4 -#define NVM_CFG1_GLOB_LED_SWAP_2_MASK 0x00000F00 -#define NVM_CFG1_GLOB_LED_SWAP_2_OFFSET 8 -#define NVM_CFG1_GLOB_LED_SWAP_3_MASK 0x0000F000 -#define NVM_CFG1_GLOB_LED_SWAP_3_OFFSET 12 - - u32 generic_cont1; /* 0x78 */ -#define NVM_CFG1_GLOB_AVS_DAC_CODE_MASK 0x000003FF -#define NVM_CFG1_GLOB_AVS_DAC_CODE_OFFSET 0 - - u32 mbi_version; /* 0x7C */ -#define NVM_CFG1_GLOB_MBI_VERSION_0_MASK 0x000000FF -#define NVM_CFG1_GLOB_MBI_VERSION_0_OFFSET 0 -#define NVM_CFG1_GLOB_MBI_VERSION_1_MASK 0x0000FF00 -#define NVM_CFG1_GLOB_MBI_VERSION_1_OFFSET 8 -#define NVM_CFG1_GLOB_MBI_VERSION_2_MASK 0x00FF0000 -#define NVM_CFG1_GLOB_MBI_VERSION_2_OFFSET 16 - - u32 mbi_date; /* 0x80 */ - - u32 misc_sig; /* 0x84 */ - - /* Define the GPIO mapping to switch i2c mux */ -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO_0_MASK 0x000000FF -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO_0_OFFSET 0 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO_1_MASK 0x0000FF00 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO_1_OFFSET 8 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__NA 0x0 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO0 0x1 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO1 0x2 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO2 0x3 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO3 0x4 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO4 0x5 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO5 0x6 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO6 0x7 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO7 0x8 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO8 0x9 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO9 0xA -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO10 0xB -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO11 0xC -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO12 0xD -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO13 0xE -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO14 0xF -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO15 0x10 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO16 0x11 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO17 0x12 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO18 0x13 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO19 0x14 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO20 0x15 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO21 0x16 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO22 0x17 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO23 0x18 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO24 0x19 -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO25 0x1A -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO26 0x1B -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO27 0x1C -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO28 0x1D -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO29 0x1E -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO30 0x1F -#define NVM_CFG1_GLOB_I2C_MUX_SEL_GPIO__GPIO31 0x20 - u32 device_capabilities; /* 0x88 */ -#define NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ETHERNET 0x1 - u32 power_dissipated; /* 0x8C */ - u32 power_consumed; /* 0x90 */ - u32 efi_version; /* 0x94 */ - u32 reserved[42]; /* 0x98 */ + u32 generic_cont0; +#define NVM_CFG1_GLOB_MF_MODE_MASK 0x00000FF0 +#define NVM_CFG1_GLOB_MF_MODE_OFFSET 4 +#define NVM_CFG1_GLOB_MF_MODE_MF_ALLOWED 0x0 +#define NVM_CFG1_GLOB_MF_MODE_DEFAULT 0x1 +#define NVM_CFG1_GLOB_MF_MODE_SPIO4 0x2 +#define NVM_CFG1_GLOB_MF_MODE_NPAR1_0 0x3 +#define NVM_CFG1_GLOB_MF_MODE_NPAR1_5 0x4 +#define NVM_CFG1_GLOB_MF_MODE_NPAR2_0 0x5 +#define NVM_CFG1_GLOB_MF_MODE_BD 0x6 +#define NVM_CFG1_GLOB_MF_MODE_UFP 0x7 + u32 engineering_change[3]; + u32 manufacturing_id; + u32 serial_number[4]; + u32 pcie_cfg; + u32 mgmt_traffic; + u32 core_cfg; +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_MASK 0x000000FF +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_OFFSET 0 +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_BB_2X40G 0x0 +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_2X50G 0x1 +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_BB_1X100G 0x2 +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_4X10G_F 0x3 +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_BB_4X10G_E 0x4 +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_BB_4X20G 0x5 +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_1X40G 0xB +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_2X25G 0xC +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_1X25G 0xD +#define NVM_CFG1_GLOB_NETWORK_PORT_MODE_4X25G 0xE + u32 e_lane_cfg1; + u32 e_lane_cfg2; + u32 f_lane_cfg1; + u32 f_lane_cfg2; + u32 mps10_preemphasis; + u32 mps10_driver_current; + u32 mps25_preemphasis; + u32 mps25_driver_current; + u32 pci_id; + u32 pci_subsys_id; + u32 bar; + u32 mps10_txfir_main; + u32 mps10_txfir_post; + u32 mps25_txfir_main; + u32 mps25_txfir_post; + u32 manufacture_ver; + u32 manufacture_time; + u32 led_global_settings; + u32 generic_cont1; + u32 mbi_version; + u32 mbi_date; + u32 misc_sig; + u32 device_capabilities; +#define NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ETHERNET 0x1 + u32 power_dissipated; + u32 power_consumed; + u32 efi_version; + u32 multi_network_modes_capability; + u32 reserved[41]; }; struct nvm_cfg1_path { - u32 reserved[30]; /* 0x0 */ + u32 reserved[30]; }; struct nvm_cfg1_port { - u32 reserved__m_relocated_to_option_123; /* 0x0 */ - u32 reserved__m_relocated_to_option_124; /* 0x4 */ - u32 generic_cont0; /* 0x8 */ -#define NVM_CFG1_PORT_LED_MODE_MASK 0x000000FF -#define NVM_CFG1_PORT_LED_MODE_OFFSET 0 -#define NVM_CFG1_PORT_LED_MODE_MAC1 0x0 -#define NVM_CFG1_PORT_LED_MODE_PHY1 0x1 -#define NVM_CFG1_PORT_LED_MODE_PHY2 0x2 -#define NVM_CFG1_PORT_LED_MODE_PHY3 0x3 -#define NVM_CFG1_PORT_LED_MODE_MAC2 0x4 -#define NVM_CFG1_PORT_LED_MODE_PHY4 0x5 -#define NVM_CFG1_PORT_LED_MODE_PHY5 0x6 -#define NVM_CFG1_PORT_LED_MODE_PHY6 0x7 -#define NVM_CFG1_PORT_LED_MODE_MAC3 0x8 -#define NVM_CFG1_PORT_LED_MODE_PHY7 0x9 -#define NVM_CFG1_PORT_LED_MODE_PHY8 0xA -#define NVM_CFG1_PORT_LED_MODE_PHY9 0xB -#define NVM_CFG1_PORT_LED_MODE_MAC4 0xC -#define NVM_CFG1_PORT_LED_MODE_PHY10 0xD -#define NVM_CFG1_PORT_LED_MODE_PHY11 0xE -#define NVM_CFG1_PORT_LED_MODE_PHY12 0xF -#define NVM_CFG1_PORT_ROCE_PRIORITY_MASK 0x0000FF00 -#define NVM_CFG1_PORT_ROCE_PRIORITY_OFFSET 8 -#define NVM_CFG1_PORT_DCBX_MODE_MASK 0x000F0000 -#define NVM_CFG1_PORT_DCBX_MODE_OFFSET 16 -#define NVM_CFG1_PORT_DCBX_MODE_DISABLED 0x0 -#define NVM_CFG1_PORT_DCBX_MODE_IEEE 0x1 -#define NVM_CFG1_PORT_DCBX_MODE_CEE 0x2 -#define NVM_CFG1_PORT_DCBX_MODE_DYNAMIC 0x3 -#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_MASK 0x00F00000 -#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_OFFSET 20 -#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_ETHERNET 0x1 - u32 pcie_cfg; /* 0xC */ -#define NVM_CFG1_PORT_RESERVED15_MASK 0x00000007 -#define NVM_CFG1_PORT_RESERVED15_OFFSET 0 - - u32 features; /* 0x10 */ -#define NVM_CFG1_PORT_ENABLE_WOL_ON_ACPI_PATTERN_MASK 0x00000001 -#define NVM_CFG1_PORT_ENABLE_WOL_ON_ACPI_PATTERN_OFFSET 0 -#define NVM_CFG1_PORT_ENABLE_WOL_ON_ACPI_PATTERN_DISABLED 0x0 -#define NVM_CFG1_PORT_ENABLE_WOL_ON_ACPI_PATTERN_ENABLED 0x1 -#define NVM_CFG1_PORT_MAGIC_PACKET_WOL_MASK 0x00000002 -#define NVM_CFG1_PORT_MAGIC_PACKET_WOL_OFFSET 1 -#define NVM_CFG1_PORT_MAGIC_PACKET_WOL_DISABLED 0x0 -#define NVM_CFG1_PORT_MAGIC_PACKET_WOL_ENABLED 0x1 - - u32 speed_cap_mask; /* 0x14 */ -#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_MASK 0x0000FFFF -#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_OFFSET 0 -#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_1G 0x1 -#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_10G 0x2 -#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_25G 0x8 -#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_40G 0x10 -#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G 0x20 -#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_100G 0x40 -#define NVM_CFG1_PORT_MFW_SPEED_CAPABILITY_MASK_MASK 0xFFFF0000 -#define NVM_CFG1_PORT_MFW_SPEED_CAPABILITY_MASK_OFFSET 16 -#define NVM_CFG1_PORT_MFW_SPEED_CAPABILITY_MASK_1G 0x1 -#define NVM_CFG1_PORT_MFW_SPEED_CAPABILITY_MASK_10G 0x2 -#define NVM_CFG1_PORT_MFW_SPEED_CAPABILITY_MASK_25G 0x8 -#define NVM_CFG1_PORT_MFW_SPEED_CAPABILITY_MASK_40G 0x10 -#define NVM_CFG1_PORT_MFW_SPEED_CAPABILITY_MASK_50G 0x20 -#define NVM_CFG1_PORT_MFW_SPEED_CAPABILITY_MASK_100G 0x40 - - u32 link_settings; /* 0x18 */ -#define NVM_CFG1_PORT_DRV_LINK_SPEED_MASK 0x0000000F -#define NVM_CFG1_PORT_DRV_LINK_SPEED_OFFSET 0 -#define NVM_CFG1_PORT_DRV_LINK_SPEED_AUTONEG 0x0 -#define NVM_CFG1_PORT_DRV_LINK_SPEED_1G 0x1 -#define NVM_CFG1_PORT_DRV_LINK_SPEED_10G 0x2 -#define NVM_CFG1_PORT_DRV_LINK_SPEED_25G 0x4 -#define NVM_CFG1_PORT_DRV_LINK_SPEED_40G 0x5 -#define NVM_CFG1_PORT_DRV_LINK_SPEED_50G 0x6 -#define NVM_CFG1_PORT_DRV_LINK_SPEED_100G 0x7 -#define NVM_CFG1_PORT_DRV_FLOW_CONTROL_MASK 0x00000070 -#define NVM_CFG1_PORT_DRV_FLOW_CONTROL_OFFSET 4 -#define NVM_CFG1_PORT_DRV_FLOW_CONTROL_AUTONEG 0x1 -#define NVM_CFG1_PORT_DRV_FLOW_CONTROL_RX 0x2 -#define NVM_CFG1_PORT_DRV_FLOW_CONTROL_TX 0x4 -#define NVM_CFG1_PORT_MFW_LINK_SPEED_MASK 0x00000780 -#define NVM_CFG1_PORT_MFW_LINK_SPEED_OFFSET 7 -#define NVM_CFG1_PORT_MFW_LINK_SPEED_AUTONEG 0x0 -#define NVM_CFG1_PORT_MFW_LINK_SPEED_1G 0x1 -#define NVM_CFG1_PORT_MFW_LINK_SPEED_10G 0x2 -#define NVM_CFG1_PORT_MFW_LINK_SPEED_25G 0x4 -#define NVM_CFG1_PORT_MFW_LINK_SPEED_40G 0x5 -#define NVM_CFG1_PORT_MFW_LINK_SPEED_50G 0x6 -#define NVM_CFG1_PORT_MFW_LINK_SPEED_100G 0x7 -#define NVM_CFG1_PORT_MFW_FLOW_CONTROL_MASK 0x00003800 -#define NVM_CFG1_PORT_MFW_FLOW_CONTROL_OFFSET 11 -#define NVM_CFG1_PORT_MFW_FLOW_CONTROL_AUTONEG 0x1 -#define NVM_CFG1_PORT_MFW_FLOW_CONTROL_RX 0x2 -#define NVM_CFG1_PORT_MFW_FLOW_CONTROL_TX 0x4 -#define NVM_CFG1_PORT_OPTIC_MODULE_VENDOR_ENFORCEMENT_MASK 0x00004000 -#define NVM_CFG1_PORT_OPTIC_MODULE_VENDOR_ENFORCEMENT_OFFSET 14 -#define NVM_CFG1_PORT_OPTIC_MODULE_VENDOR_ENFORCEMENT_DISABLED 0x0 -#define NVM_CFG1_PORT_OPTIC_MODULE_VENDOR_ENFORCEMENT_ENABLED 0x1 - - u32 phy_cfg; /* 0x1C */ -#define NVM_CFG1_PORT_OPTIONAL_LINK_MODES_MASK 0x0000FFFF -#define NVM_CFG1_PORT_OPTIONAL_LINK_MODES_OFFSET 0 -#define NVM_CFG1_PORT_OPTIONAL_LINK_MODES_HIGIG 0x1 -#define NVM_CFG1_PORT_OPTIONAL_LINK_MODES_SCRAMBLER 0x2 -#define NVM_CFG1_PORT_OPTIONAL_LINK_MODES_FIBER 0x4 -#define NVM_CFG1_PORT_OPTIONAL_LINK_MODES_DISABLE_CL72_AN 0x8 -#define NVM_CFG1_PORT_OPTIONAL_LINK_MODES_DISABLE_FEC_AN 0x10 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_MASK 0x00FF0000 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_OFFSET 16 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_BYPASS 0x0 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_KR 0x2 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_KR2 0x3 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_KR4 0x4 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_XFI 0x8 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_SFI 0x9 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_1000X 0xB -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_SGMII 0xC -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_XLAUI 0x11 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_XLPPI 0x12 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_CAUI 0x21 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_CPPI 0x22 -#define NVM_CFG1_PORT_SERDES_NET_INTERFACE_25GAUI 0x31 -#define NVM_CFG1_PORT_AN_MODE_MASK 0xFF000000 -#define NVM_CFG1_PORT_AN_MODE_OFFSET 24 -#define NVM_CFG1_PORT_AN_MODE_NONE 0x0 -#define NVM_CFG1_PORT_AN_MODE_CL73 0x1 -#define NVM_CFG1_PORT_AN_MODE_CL37 0x2 -#define NVM_CFG1_PORT_AN_MODE_CL73_BAM 0x3 -#define NVM_CFG1_PORT_AN_MODE_CL37_BAM 0x4 -#define NVM_CFG1_PORT_AN_MODE_HPAM 0x5 -#define NVM_CFG1_PORT_AN_MODE_SGMII 0x6 - - u32 mgmt_traffic; /* 0x20 */ -#define NVM_CFG1_PORT_RESERVED61_MASK 0x0000000F -#define NVM_CFG1_PORT_RESERVED61_OFFSET 0 - - u32 ext_phy; /* 0x24 */ -#define NVM_CFG1_PORT_EXTERNAL_PHY_TYPE_MASK 0x000000FF -#define NVM_CFG1_PORT_EXTERNAL_PHY_TYPE_OFFSET 0 -#define NVM_CFG1_PORT_EXTERNAL_PHY_TYPE_NONE 0x0 -#define NVM_CFG1_PORT_EXTERNAL_PHY_TYPE_BCM84844 0x1 -#define NVM_CFG1_PORT_EXTERNAL_PHY_ADDRESS_MASK 0x0000FF00 -#define NVM_CFG1_PORT_EXTERNAL_PHY_ADDRESS_OFFSET 8 - - u32 mba_cfg1; /* 0x28 */ -#define NVM_CFG1_PORT_PREBOOT_OPROM_MASK 0x00000001 -#define NVM_CFG1_PORT_PREBOOT_OPROM_OFFSET 0 -#define NVM_CFG1_PORT_PREBOOT_OPROM_DISABLED 0x0 -#define NVM_CFG1_PORT_PREBOOT_OPROM_ENABLED 0x1 -#define NVM_CFG1_PORT_RESERVED__M_MBA_BOOT_TYPE_MASK 0x00000006 -#define NVM_CFG1_PORT_RESERVED__M_MBA_BOOT_TYPE_OFFSET 1 -#define NVM_CFG1_PORT_MBA_DELAY_TIME_MASK 0x00000078 -#define NVM_CFG1_PORT_MBA_DELAY_TIME_OFFSET 3 -#define NVM_CFG1_PORT_MBA_SETUP_HOT_KEY_MASK 0x00000080 -#define NVM_CFG1_PORT_MBA_SETUP_HOT_KEY_OFFSET 7 -#define NVM_CFG1_PORT_MBA_SETUP_HOT_KEY_CTRL_S 0x0 -#define NVM_CFG1_PORT_MBA_SETUP_HOT_KEY_CTRL_B 0x1 -#define NVM_CFG1_PORT_MBA_HIDE_SETUP_PROMPT_MASK 0x00000100 -#define NVM_CFG1_PORT_MBA_HIDE_SETUP_PROMPT_OFFSET 8 -#define NVM_CFG1_PORT_MBA_HIDE_SETUP_PROMPT_DISABLED 0x0 -#define NVM_CFG1_PORT_MBA_HIDE_SETUP_PROMPT_ENABLED 0x1 -#define NVM_CFG1_PORT_RESERVED5_MASK 0x0001FE00 -#define NVM_CFG1_PORT_RESERVED5_OFFSET 9 -#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_MASK 0x001E0000 -#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_OFFSET 17 -#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_AUTONEG 0x0 -#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_1G 0x1 -#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_10G 0x2 -#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_25G 0x4 -#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_40G 0x5 -#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_50G 0x6 -#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_100G 0x7 -#define NVM_CFG1_PORT_PREBOOT_LINK_SPEED_SMARTLINQ 0x8 -#define NVM_CFG1_PORT_RESERVED__M_MBA_BOOT_RETRY_COUNT_MASK 0x00E00000 -#define NVM_CFG1_PORT_RESERVED__M_MBA_BOOT_RETRY_COUNT_OFFSET 21 - - u32 mba_cfg2; /* 0x2C */ -#define NVM_CFG1_PORT_RESERVED65_MASK 0x0000FFFF -#define NVM_CFG1_PORT_RESERVED65_OFFSET 0 -#define NVM_CFG1_PORT_RESERVED66_MASK 0x00010000 -#define NVM_CFG1_PORT_RESERVED66_OFFSET 16 - - u32 vf_cfg; /* 0x30 */ -#define NVM_CFG1_PORT_RESERVED8_MASK 0x0000FFFF -#define NVM_CFG1_PORT_RESERVED8_OFFSET 0 -#define NVM_CFG1_PORT_RESERVED6_MASK 0x000F0000 -#define NVM_CFG1_PORT_RESERVED6_OFFSET 16 - - struct nvm_cfg_mac_address lldp_mac_address; /* 0x34 */ - - u32 led_port_settings; /* 0x3C */ -#define NVM_CFG1_PORT_LANE_LED_SPD_0_SEL_MASK 0x000000FF -#define NVM_CFG1_PORT_LANE_LED_SPD_0_SEL_OFFSET 0 -#define NVM_CFG1_PORT_LANE_LED_SPD_1_SEL_MASK 0x0000FF00 -#define NVM_CFG1_PORT_LANE_LED_SPD_1_SEL_OFFSET 8 -#define NVM_CFG1_PORT_LANE_LED_SPD_2_SEL_MASK 0x00FF0000 -#define NVM_CFG1_PORT_LANE_LED_SPD_2_SEL_OFFSET 16 -#define NVM_CFG1_PORT_LANE_LED_SPD__SEL_1G 0x1 -#define NVM_CFG1_PORT_LANE_LED_SPD__SEL_10G 0x2 -#define NVM_CFG1_PORT_LANE_LED_SPD__SEL_25G 0x8 -#define NVM_CFG1_PORT_LANE_LED_SPD__SEL_40G 0x10 -#define NVM_CFG1_PORT_LANE_LED_SPD__SEL_50G 0x20 -#define NVM_CFG1_PORT_LANE_LED_SPD__SEL_100G 0x40 - - u32 transceiver_00; /* 0x40 */ - - /* Define for mapping of transceiver signal module absent */ -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_MASK 0x000000FF -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_OFFSET 0 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_NA 0x0 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO0 0x1 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO1 0x2 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO2 0x3 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO3 0x4 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO4 0x5 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO5 0x6 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO6 0x7 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO7 0x8 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO8 0x9 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO9 0xA -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO10 0xB -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO11 0xC -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO12 0xD -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO13 0xE -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO14 0xF -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO15 0x10 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO16 0x11 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO17 0x12 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO18 0x13 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO19 0x14 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO20 0x15 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO21 0x16 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO22 0x17 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO23 0x18 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO24 0x19 -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO25 0x1A -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO26 0x1B -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO27 0x1C -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO28 0x1D -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO29 0x1E -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO30 0x1F -#define NVM_CFG1_PORT_TRANS_MODULE_ABS_GPIO31 0x20 - /* Define the GPIO mux settings to switch i2c mux to this port */ -#define NVM_CFG1_PORT_I2C_MUX_SEL_VALUE_0_MASK 0x00000F00 -#define NVM_CFG1_PORT_I2C_MUX_SEL_VALUE_0_OFFSET 8 -#define NVM_CFG1_PORT_I2C_MUX_SEL_VALUE_1_MASK 0x0000F000 -#define NVM_CFG1_PORT_I2C_MUX_SEL_VALUE_1_OFFSET 12 - - u32 reserved[133]; /* 0x44 */ + u32 reserved__m_relocated_to_option_123; + u32 reserved__m_relocated_to_option_124; + u32 generic_cont0; +#define NVM_CFG1_PORT_DCBX_MODE_MASK 0x000F0000 +#define NVM_CFG1_PORT_DCBX_MODE_OFFSET 16 +#define NVM_CFG1_PORT_DCBX_MODE_DISABLED 0x0 +#define NVM_CFG1_PORT_DCBX_MODE_IEEE 0x1 +#define NVM_CFG1_PORT_DCBX_MODE_CEE 0x2 +#define NVM_CFG1_PORT_DCBX_MODE_DYNAMIC 0x3 +#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_MASK 0x00F00000 +#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_OFFSET 20 +#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_ETHERNET 0x1 +#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_FCOE 0x2 +#define NVM_CFG1_PORT_DEFAULT_ENABLED_PROTOCOLS_ISCSI 0x4 + u32 pcie_cfg; + u32 features; + u32 speed_cap_mask; +#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_MASK 0x0000FFFF +#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_OFFSET 0 +#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_1G 0x1 +#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_10G 0x2 +#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_25G 0x8 +#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_40G 0x10 +#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G 0x20 +#define NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_BB_100G 0x40 + u32 link_settings; +#define NVM_CFG1_PORT_DRV_LINK_SPEED_MASK 0x0000000F +#define NVM_CFG1_PORT_DRV_LINK_SPEED_OFFSET 0 +#define NVM_CFG1_PORT_DRV_LINK_SPEED_AUTONEG 0x0 +#define NVM_CFG1_PORT_DRV_LINK_SPEED_1G 0x1 +#define NVM_CFG1_PORT_DRV_LINK_SPEED_10G 0x2 +#define NVM_CFG1_PORT_DRV_LINK_SPEED_25G 0x4 +#define NVM_CFG1_PORT_DRV_LINK_SPEED_40G 0x5 +#define NVM_CFG1_PORT_DRV_LINK_SPEED_50G 0x6 +#define NVM_CFG1_PORT_DRV_LINK_SPEED_BB_100G 0x7 +#define NVM_CFG1_PORT_DRV_LINK_SPEED_SMARTLINQ 0x8 +#define NVM_CFG1_PORT_DRV_FLOW_CONTROL_MASK 0x00000070 +#define NVM_CFG1_PORT_DRV_FLOW_CONTROL_OFFSET 4 +#define NVM_CFG1_PORT_DRV_FLOW_CONTROL_AUTONEG 0x1 +#define NVM_CFG1_PORT_DRV_FLOW_CONTROL_RX 0x2 +#define NVM_CFG1_PORT_DRV_FLOW_CONTROL_TX 0x4 + u32 phy_cfg; + u32 mgmt_traffic; + u32 ext_phy; + u32 mba_cfg1; + u32 mba_cfg2; + u32 vf_cfg; + struct nvm_cfg_mac_address lldp_mac_address; + u32 led_port_settings; + u32 transceiver_00; + u32 device_ids; + u32 board_cfg; + u32 mnm_10g_cap; + u32 mnm_10g_ctrl; + u32 mnm_10g_misc; + u32 mnm_25g_cap; + u32 mnm_25g_ctrl; + u32 mnm_25g_misc; + u32 mnm_40g_cap; + u32 mnm_40g_ctrl; + u32 mnm_40g_misc; + u32 mnm_50g_cap; + u32 mnm_50g_ctrl; + u32 mnm_50g_misc; + u32 mnm_100g_cap; + u32 mnm_100g_ctrl; + u32 mnm_100g_misc; + u32 reserved[116]; }; struct nvm_cfg1_func { - struct nvm_cfg_mac_address mac_address; /* 0x0 */ - - u32 rsrv1; /* 0x8 */ -#define NVM_CFG1_FUNC_RESERVED1_MASK 0x0000FFFF -#define NVM_CFG1_FUNC_RESERVED1_OFFSET 0 -#define NVM_CFG1_FUNC_RESERVED2_MASK 0xFFFF0000 -#define NVM_CFG1_FUNC_RESERVED2_OFFSET 16 - - u32 rsrv2; /* 0xC */ -#define NVM_CFG1_FUNC_RESERVED3_MASK 0x0000FFFF -#define NVM_CFG1_FUNC_RESERVED3_OFFSET 0 -#define NVM_CFG1_FUNC_RESERVED4_MASK 0xFFFF0000 -#define NVM_CFG1_FUNC_RESERVED4_OFFSET 16 - - u32 device_id; /* 0x10 */ -#define NVM_CFG1_FUNC_MF_VENDOR_DEVICE_ID_MASK 0x0000FFFF -#define NVM_CFG1_FUNC_MF_VENDOR_DEVICE_ID_OFFSET 0 -#define NVM_CFG1_FUNC_RESERVED77_MASK 0xFFFF0000 -#define NVM_CFG1_FUNC_RESERVED77_OFFSET 16 - - u32 cmn_cfg; /* 0x14 */ -#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_MASK 0x00000007 -#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_OFFSET 0 -#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_PXE 0x0 -#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_ISCSI_BOOT 0x3 -#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_FCOE_BOOT 0x4 -#define NVM_CFG1_FUNC_PREBOOT_BOOT_PROTOCOL_NONE 0x7 -#define NVM_CFG1_FUNC_VF_PCI_DEVICE_ID_MASK 0x0007FFF8 -#define NVM_CFG1_FUNC_VF_PCI_DEVICE_ID_OFFSET 3 -#define NVM_CFG1_FUNC_PERSONALITY_MASK 0x00780000 -#define NVM_CFG1_FUNC_PERSONALITY_OFFSET 19 -#define NVM_CFG1_FUNC_PERSONALITY_ETHERNET 0x0 -#define NVM_CFG1_FUNC_PERSONALITY_ISCSI 0x1 -#define NVM_CFG1_FUNC_PERSONALITY_FCOE 0x2 -#define NVM_CFG1_FUNC_PERSONALITY_ROCE 0x3 -#define NVM_CFG1_FUNC_BANDWIDTH_WEIGHT_MASK 0x7F800000 -#define NVM_CFG1_FUNC_BANDWIDTH_WEIGHT_OFFSET 23 -#define NVM_CFG1_FUNC_PAUSE_ON_HOST_RING_MASK 0x80000000 -#define NVM_CFG1_FUNC_PAUSE_ON_HOST_RING_OFFSET 31 -#define NVM_CFG1_FUNC_PAUSE_ON_HOST_RING_DISABLED 0x0 -#define NVM_CFG1_FUNC_PAUSE_ON_HOST_RING_ENABLED 0x1 - - u32 pci_cfg; /* 0x18 */ -#define NVM_CFG1_FUNC_NUMBER_OF_VFS_PER_PF_MASK 0x0000007F -#define NVM_CFG1_FUNC_NUMBER_OF_VFS_PER_PF_OFFSET 0 -#define NVM_CFG1_FUNC_RESERVESD12_MASK 0x00003F80 -#define NVM_CFG1_FUNC_RESERVESD12_OFFSET 7 -#define NVM_CFG1_FUNC_BAR1_SIZE_MASK 0x0003C000 -#define NVM_CFG1_FUNC_BAR1_SIZE_OFFSET 14 -#define NVM_CFG1_FUNC_BAR1_SIZE_DISABLED 0x0 -#define NVM_CFG1_FUNC_BAR1_SIZE_64K 0x1 -#define NVM_CFG1_FUNC_BAR1_SIZE_128K 0x2 -#define NVM_CFG1_FUNC_BAR1_SIZE_256K 0x3 -#define NVM_CFG1_FUNC_BAR1_SIZE_512K 0x4 -#define NVM_CFG1_FUNC_BAR1_SIZE_1M 0x5 -#define NVM_CFG1_FUNC_BAR1_SIZE_2M 0x6 -#define NVM_CFG1_FUNC_BAR1_SIZE_4M 0x7 -#define NVM_CFG1_FUNC_BAR1_SIZE_8M 0x8 -#define NVM_CFG1_FUNC_BAR1_SIZE_16M 0x9 -#define NVM_CFG1_FUNC_BAR1_SIZE_32M 0xA -#define NVM_CFG1_FUNC_BAR1_SIZE_64M 0xB -#define NVM_CFG1_FUNC_BAR1_SIZE_128M 0xC -#define NVM_CFG1_FUNC_BAR1_SIZE_256M 0xD -#define NVM_CFG1_FUNC_BAR1_SIZE_512M 0xE -#define NVM_CFG1_FUNC_BAR1_SIZE_1G 0xF -#define NVM_CFG1_FUNC_MAX_BANDWIDTH_MASK 0x03FC0000 -#define NVM_CFG1_FUNC_MAX_BANDWIDTH_OFFSET 18 - - struct nvm_cfg_mac_address fcoe_node_wwn_mac_addr; /* 0x1C */ - - struct nvm_cfg_mac_address fcoe_port_wwn_mac_addr; /* 0x24 */ - u32 preboot_generic_cfg; /* 0x2C */ - u32 reserved[8]; /* 0x30 */ + struct nvm_cfg_mac_address mac_address; + u32 rsrv1; + u32 rsrv2; + u32 device_id; + u32 cmn_cfg; + u32 pci_cfg; + struct nvm_cfg_mac_address fcoe_node_wwn_mac_addr; + struct nvm_cfg_mac_address fcoe_port_wwn_mac_addr; + u32 preboot_generic_cfg; + u32 reserved[8]; }; struct nvm_cfg1 { - struct nvm_cfg1_glob glob; /* 0x0 */ - - struct nvm_cfg1_path path[MCP_GLOB_PATH_MAX]; /* 0x140 */ - - struct nvm_cfg1_port port[MCP_GLOB_PORT_MAX]; /* 0x230 */ - - struct nvm_cfg1_func func[MCP_GLOB_FUNC_MAX]; /* 0xB90 */ -}; - -/****************************************** -* nvm_cfg structs -******************************************/ - -enum nvm_cfg_sections { - NVM_CFG_SECTION_NVM_CFG1, - NVM_CFG_SECTION_MAX -}; - -struct nvm_cfg { - u32 num_sections; - u32 sections_offset[NVM_CFG_SECTION_MAX]; - struct nvm_cfg1 cfg1; -}; - -#define PORT_0 0 -#define PORT_1 1 -#define PORT_2 2 -#define PORT_3 3 - -extern struct spad_layout g_spad; - -#define MCP_SPAD_SIZE 0x00028000 /* 160 KB */ - -#define SPAD_OFFSET(addr) (((u32)addr - (u32)CPU_SPAD_BASE)) - -#define TO_OFFSIZE(_offset, _size) \ - (u32)((((u32)(_offset) >> 2) << OFFSIZE_OFFSET_SHIFT) | \ - (((u32)(_size) >> 2) << OFFSIZE_SIZE_SHIFT)) - -enum spad_sections { - SPAD_SECTION_TRACE, - SPAD_SECTION_NVM_CFG, - SPAD_SECTION_PUBLIC, - SPAD_SECTION_PRIVATE, - SPAD_SECTION_MAX -}; - -struct spad_layout { - struct nvm_cfg nvm_cfg; - struct mcp_public_data public_data; + struct nvm_cfg1_glob glob; + struct nvm_cfg1_path path[MCP_GLOB_PATH_MAX]; + struct nvm_cfg1_port port[MCP_GLOB_PORT_MAX]; + struct nvm_cfg1_func func[MCP_GLOB_FUNC_MAX]; }; - -#define CRC_MAGIC_VALUE 0xDEBB20E3 -#define CRC32_POLYNOMIAL 0xEDB88320 -#define NVM_CRC_SIZE (sizeof(u32)) - -enum nvm_sw_arbitrator { - NVM_SW_ARB_HOST, - NVM_SW_ARB_MCP, - NVM_SW_ARB_UART, - NVM_SW_ARB_RESERVED -}; - -/**************************************************************************** -* Boot Strap Region * -****************************************************************************/ -struct legacy_bootstrap_region { - u32 magic_value; -#define NVM_MAGIC_VALUE 0x669955aa - u32 sram_start_addr; - u32 code_len; /* boot code length (in dwords) */ - u32 code_start_addr; - u32 crc; /* 32-bit CRC */ -}; - -/**************************************************************************** -* Directories Region * -****************************************************************************/ -struct nvm_code_entry { - u32 image_type; /* Image type */ - u32 nvm_start_addr; /* NVM address of the image */ - u32 len; /* Include CRC */ - u32 sram_start_addr; - u32 sram_run_addr; /* Relevant in case of MIM only */ -}; - -enum nvm_image_type { - NVM_TYPE_TIM1 = 0x01, - NVM_TYPE_TIM2 = 0x02, - NVM_TYPE_MIM1 = 0x03, - NVM_TYPE_MIM2 = 0x04, - NVM_TYPE_MBA = 0x05, - NVM_TYPE_MODULES_PN = 0x06, - NVM_TYPE_VPD = 0x07, - NVM_TYPE_MFW_TRACE1 = 0x08, - NVM_TYPE_MFW_TRACE2 = 0x09, - NVM_TYPE_NVM_CFG1 = 0x0a, - NVM_TYPE_L2B = 0x0b, - NVM_TYPE_DIR1 = 0x0c, - NVM_TYPE_EAGLE_FW1 = 0x0d, - NVM_TYPE_FALCON_FW1 = 0x0e, - NVM_TYPE_PCIE_FW1 = 0x0f, - NVM_TYPE_HW_SET = 0x10, - NVM_TYPE_LIM = 0x11, - NVM_TYPE_AVS_FW1 = 0x12, - NVM_TYPE_DIR2 = 0x13, - NVM_TYPE_CCM = 0x14, - NVM_TYPE_EAGLE_FW2 = 0x15, - NVM_TYPE_FALCON_FW2 = 0x16, - NVM_TYPE_PCIE_FW2 = 0x17, - NVM_TYPE_AVS_FW2 = 0x18, - - NVM_TYPE_MAX, -}; - -#define MAX_NVM_DIR_ENTRIES 200 - -struct nvm_dir { - s32 seq; -#define NVM_DIR_NEXT_MFW_MASK 0x00000001 -#define NVM_DIR_SEQ_MASK 0xfffffffe -#define NVM_DIR_NEXT_MFW(seq) ((seq) & NVM_DIR_NEXT_MFW_MASK) - -#define IS_DIR_SEQ_VALID(seq) ((seq & NVM_DIR_SEQ_MASK) != NVM_DIR_SEQ_MASK) - - u32 num_images; - u32 rsrv; - struct nvm_code_entry code[1]; /* Up to MAX_NVM_DIR_ENTRIES */ -}; - -#define NVM_DIR_SIZE(_num_images) (sizeof(struct nvm_dir) + \ - (_num_images - \ - 1) * sizeof(struct nvm_code_entry) + \ - NVM_CRC_SIZE) - -struct nvm_vpd_image { - u32 format_revision; -#define VPD_IMAGE_VERSION 1 - - /* This array length depends on the number of VPD fields */ - u8 vpd_data[1]; -}; - -/**************************************************************************** -* NVRAM FULL MAP * -****************************************************************************/ -#define DIR_ID_1 (0) -#define DIR_ID_2 (1) -#define MAX_DIR_IDS (2) - -#define MFW_BUNDLE_1 (0) -#define MFW_BUNDLE_2 (1) -#define MAX_MFW_BUNDLES (2) - -#define FLASH_PAGE_SIZE 0x1000 -#define NVM_DIR_MAX_SIZE (FLASH_PAGE_SIZE) /* 4Kb */ -#define ASIC_MIM_MAX_SIZE (300 * FLASH_PAGE_SIZE) /* 1.2Mb */ -#define FPGA_MIM_MAX_SIZE (25 * FLASH_PAGE_SIZE) /* 60Kb */ - -#define LIM_MAX_SIZE ((2 * \ - FLASH_PAGE_SIZE) - \ - sizeof(struct legacy_bootstrap_region) - \ - NVM_RSV_SIZE) -#define LIM_OFFSET (NVM_OFFSET(lim_image)) -#define NVM_RSV_SIZE (44) -#define MIM_MAX_SIZE(is_asic) ((is_asic) ? ASIC_MIM_MAX_SIZE : \ - FPGA_MIM_MAX_SIZE) -#define MIM_OFFSET(idx, is_asic) (NVM_OFFSET(dir[MAX_MFW_BUNDLES]) + \ - ((idx == \ - NVM_TYPE_MIM2) ? MIM_MAX_SIZE(is_asic) : 0)) -#define NVM_FIXED_AREA_SIZE(is_asic) (sizeof(struct nvm_image) + \ - MIM_MAX_SIZE(is_asic) * 2) - -union nvm_dir_union { - struct nvm_dir dir; - u8 page[FLASH_PAGE_SIZE]; -}; - -/* Address - * +-------------------+ 0x000000 - * | Bootstrap: | - * | magic_number | - * | sram_start_addr | - * | code_len | - * | code_start_addr | - * | crc | - * +-------------------+ 0x000014 - * | rsrv | - * +-------------------+ 0x000040 - * | LIM | - * +-------------------+ 0x002000 - * | Dir1 | - * +-------------------+ 0x003000 - * | Dir2 | - * +-------------------+ 0x004000 - * | MIM1 | - * +-------------------+ 0x130000 - * | MIM2 | - * +-------------------+ 0x25C000 - * | Rest Images: | - * | TIM1/2 | - * | MFW_TRACE1/2 | - * | Eagle/Falcon FW | - * | PCIE/AVS FW | - * | MBA/CCM/L2B | - * | VPD | - * | optic_modules | - * | ... | - * +-------------------+ 0x400000 - */ -struct nvm_image { -/*********** !!! FIXED SECTIONS !!! DO NOT MODIFY !!! **********************/ - /* NVM Offset (size) */ - struct legacy_bootstrap_region bootstrap; - u8 rsrv[NVM_RSV_SIZE]; - u8 lim_image[LIM_MAX_SIZE]; - union nvm_dir_union dir[MAX_MFW_BUNDLES]; - - /* MIM1_IMAGE 0x004000 (0x12c000) */ - /* MIM2_IMAGE 0x130000 (0x12c000) */ -/*********** !!! FIXED SECTIONS !!! DO NOT MODIFY !!! **********************/ -}; /* 0x134 */ - -#define NVM_OFFSET(f) ((u32_t)((int_ptr_t)(&(((struct nvm_image *)0)->f)))) - -struct hw_set_info { - u32 reg_type; -#define GRC_REG_TYPE 1 -#define PHY_REG_TYPE 2 -#define PCI_REG_TYPE 4 - - u32 bank_num; - u32 pf_num; - u32 operation; -#define READ_OP 1 -#define WRITE_OP 2 -#define RMW_SET_OP 3 -#define RMW_CLR_OP 4 - - u32 reg_addr; - u32 reg_data; - - u32 reset_type; -#define POR_RESET_TYPE BIT(0) -#define HARD_RESET_TYPE BIT(1) -#define CORE_RESET_TYPE BIT(2) -#define MCP_RESET_TYPE BIT(3) -#define PERSET_ASSERT BIT(4) -#define PERSET_DEASSERT BIT(5) -}; - -struct hw_set_image { - u32 format_version; -#define HW_SET_IMAGE_VERSION 1 - u32 no_hw_sets; - - /* This array length depends on the no_hw_sets */ - struct hw_set_info hw_sets[1]; -}; - -int qed_init_pf_wfq(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - u8 pf_id, u16 pf_wfq); -int qed_init_vport_wfq(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - u16 first_tx_pq_id[NUM_OF_TCS], u16 vport_wfq); #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c index 0ada7fd..7363d2b 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.c +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c @@ -446,7 +446,7 @@ qed_dmae_post_command(struct qed_hwfn *p_hwfn, idx_cmd, le32_to_cpu(command->opcode), le16_to_cpu(command->opcode_b), - le16_to_cpu(command->length), + le16_to_cpu(command->length_dw), le32_to_cpu(command->src_addr_hi), le32_to_cpu(command->src_addr_lo), le32_to_cpu(command->dst_addr_hi), @@ -461,7 +461,7 @@ qed_dmae_post_command(struct qed_hwfn *p_hwfn, idx_cmd, le32_to_cpu(command->opcode), le16_to_cpu(command->opcode_b), - le16_to_cpu(command->length), + le16_to_cpu(command->length_dw), le32_to_cpu(command->src_addr_hi), le32_to_cpu(command->src_addr_lo), le32_to_cpu(command->dst_addr_hi), @@ -645,7 +645,7 @@ static int qed_dmae_execute_sub_operation(struct qed_hwfn *p_hwfn, return -EINVAL; } - cmd->length = cpu_to_le16((u16)length); + cmd->length_dw = cpu_to_le16((u16)length); qed_dmae_post_command(p_hwfn, p_ptt); diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c index e8a3b9d..23e455f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c +++ b/drivers/net/ethernet/qlogic/qed/qed_init_fw_funcs.c @@ -31,7 +31,6 @@ enum cminterface { }; /* general constants */ -#define QM_PQ_ELEMENT_SIZE 4 /* in bytes */ #define QM_PQ_MEM_4KB(pq_size) (pq_size ? DIV_ROUND_UP((pq_size + 1) * \ QM_PQ_ELEMENT_SIZE, \ 0x1000) : 0) @@ -44,28 +43,28 @@ enum cminterface { /* other PQ constants */ #define QM_OTHER_PQS_PER_PF 4 /* WFQ constants */ -#define QM_WFQ_UPPER_BOUND 6250000 +#define QM_WFQ_UPPER_BOUND 62500000 #define QM_WFQ_VP_PQ_VOQ_SHIFT 0 #define QM_WFQ_VP_PQ_PF_SHIFT 5 #define QM_WFQ_INC_VAL(weight) ((weight) * 0x9000) -#define QM_WFQ_MAX_INC_VAL 4375000 -#define QM_WFQ_INIT_CRD(inc_val) (2 * (inc_val)) +#define QM_WFQ_MAX_INC_VAL 43750000 + /* RL constants */ -#define QM_RL_UPPER_BOUND 6250000 +#define QM_RL_UPPER_BOUND 62500000 #define QM_RL_PERIOD 5 /* in us */ #define QM_RL_PERIOD_CLK_25M (25 * QM_RL_PERIOD) +#define QM_RL_MAX_INC_VAL 43750000 #define QM_RL_INC_VAL(rate) max_t(u32, \ - (((rate ? rate : 1000000) \ - * QM_RL_PERIOD) / 8), 1) -#define QM_RL_MAX_INC_VAL 4375000 + (u32)(((rate ? rate : \ + 1000000) * \ + QM_RL_PERIOD * \ + 101) / (8 * 100)), 1) /* AFullOprtnstcCrdMask constants */ #define QM_OPPOR_LINE_VOQ_DEF 1 #define QM_OPPOR_FW_STOP_DEF 0 #define QM_OPPOR_PQ_EMPTY_DEF 1 -#define EAGLE_WORKAROUND_TC 7 /* Command Queue constants */ #define PBF_CMDQ_PURE_LB_LINES 150 -#define PBF_CMDQ_EAGLE_WORKAROUND_LINES 8 #define PBF_CMDQ_LINES_RT_OFFSET(voq) ( \ PBF_REG_YCMD_QS_NUM_LINES_VOQ0_RT_OFFSET + voq * \ (PBF_REG_YCMD_QS_NUM_LINES_VOQ1_RT_OFFSET - \ @@ -80,7 +79,6 @@ enum cminterface { /* BTB: blocks constants (block size = 256B) */ #define BTB_JUMBO_PKT_BLOCKS 38 #define BTB_HEADROOM_BLOCKS BTB_JUMBO_PKT_BLOCKS -#define BTB_EAGLE_WORKAROUND_BLOCKS 4 #define BTB_PURE_LB_FACTOR 10 #define BTB_PURE_LB_RATIO 7 /* QM stop command constants */ @@ -107,9 +105,9 @@ enum cminterface { cmd ## _ ## field, \ value) /* QM: VOQ macros */ -#define PHYS_VOQ(port, tc, max_phy_tcs_pr_port) ((port) * \ - (max_phy_tcs_pr_port) \ - + (tc)) +#define PHYS_VOQ(port, tc, max_phys_tcs_per_port) ((port) * \ + (max_phys_tcs_per_port) + \ + (tc)) #define LB_VOQ(port) ( \ MAX_PHYS_VOQS + (port)) #define VOQ(port, tc, max_phy_tcs_pr_port) \ @@ -120,8 +118,7 @@ enum cminterface { : LB_VOQ(port)) /******************** INTERNAL IMPLEMENTATION *********************/ /* Prepare PF RL enable/disable runtime init values */ -static void qed_enable_pf_rl(struct qed_hwfn *p_hwfn, - bool pf_rl_en) +static void qed_enable_pf_rl(struct qed_hwfn *p_hwfn, bool pf_rl_en) { STORE_RT_REG(p_hwfn, QM_REG_RLPFENABLE_RT_OFFSET, pf_rl_en ? 1 : 0); if (pf_rl_en) { @@ -130,8 +127,7 @@ static void qed_enable_pf_rl(struct qed_hwfn *p_hwfn, (1 << MAX_NUM_VOQS) - 1); /* write RL period */ STORE_RT_REG(p_hwfn, - QM_REG_RLPFPERIOD_RT_OFFSET, - QM_RL_PERIOD_CLK_25M); + QM_REG_RLPFPERIOD_RT_OFFSET, QM_RL_PERIOD_CLK_25M); STORE_RT_REG(p_hwfn, QM_REG_RLPFPERIODTIMER_RT_OFFSET, QM_RL_PERIOD_CLK_25M); @@ -144,8 +140,7 @@ static void qed_enable_pf_rl(struct qed_hwfn *p_hwfn, } /* Prepare PF WFQ enable/disable runtime init values */ -static void qed_enable_pf_wfq(struct qed_hwfn *p_hwfn, - bool pf_wfq_en) +static void qed_enable_pf_wfq(struct qed_hwfn *p_hwfn, bool pf_wfq_en) { STORE_RT_REG(p_hwfn, QM_REG_WFQPFENABLE_RT_OFFSET, pf_wfq_en ? 1 : 0); /* set credit threshold for QM bypass flow */ @@ -156,8 +151,7 @@ static void qed_enable_pf_wfq(struct qed_hwfn *p_hwfn, } /* Prepare VPORT RL enable/disable runtime init values */ -static void qed_enable_vport_rl(struct qed_hwfn *p_hwfn, - bool vport_rl_en) +static void qed_enable_vport_rl(struct qed_hwfn *p_hwfn, bool vport_rl_en) { STORE_RT_REG(p_hwfn, QM_REG_RLGLBLENABLE_RT_OFFSET, vport_rl_en ? 1 : 0); @@ -178,8 +172,7 @@ static void qed_enable_vport_rl(struct qed_hwfn *p_hwfn, } /* Prepare VPORT WFQ enable/disable runtime init values */ -static void qed_enable_vport_wfq(struct qed_hwfn *p_hwfn, - bool vport_wfq_en) +static void qed_enable_vport_wfq(struct qed_hwfn *p_hwfn, bool vport_wfq_en) { STORE_RT_REG(p_hwfn, QM_REG_WFQVPENABLE_RT_OFFSET, vport_wfq_en ? 1 : 0); @@ -194,8 +187,7 @@ static void qed_enable_vport_wfq(struct qed_hwfn *p_hwfn, * the specified VOQ */ static void qed_cmdq_lines_voq_rt_init(struct qed_hwfn *p_hwfn, - u8 voq, - u16 cmdq_lines) + u8 voq, u16 cmdq_lines) { u32 qm_line_crd; @@ -221,7 +213,7 @@ static void qed_cmdq_lines_rt_init( u8 max_phys_tcs_per_port, struct init_qm_port_params port_params[MAX_NUM_PORTS]) { - u8 tc, voq, port_id; + u8 tc, voq, port_id, num_tcs_in_port; /* clear PBF lines for all VOQs */ for (voq = 0; voq < MAX_NUM_VOQS; voq++) @@ -229,22 +221,31 @@ static void qed_cmdq_lines_rt_init( for (port_id = 0; port_id < max_ports_per_engine; port_id++) { if (port_params[port_id].active) { u16 phys_lines, phys_lines_per_tc; - u8 phys_tcs = port_params[port_id].num_active_phys_tcs; - /* find #lines to divide between the active - * physical TCs. - */ + /* find #lines to divide between active phys TCs */ phys_lines = port_params[port_id].num_pbf_cmd_lines - PBF_CMDQ_PURE_LB_LINES; /* find #lines per active physical TC */ - phys_lines_per_tc = phys_lines / phys_tcs; + num_tcs_in_port = 0; + for (tc = 0; tc < NUM_OF_PHYS_TCS; tc++) { + if (((port_params[port_id].active_phys_tcs >> + tc) & 0x1) == 1) + num_tcs_in_port++; + } + + phys_lines_per_tc = phys_lines / num_tcs_in_port; /* init registers per active TC */ - for (tc = 0; tc < phys_tcs; tc++) { + for (tc = 0; tc < NUM_OF_PHYS_TCS; tc++) { + if (((port_params[port_id].active_phys_tcs >> + tc) & 0x1) != 1) + continue; + voq = PHYS_VOQ(port_id, tc, max_phys_tcs_per_port); qed_cmdq_lines_voq_rt_init(p_hwfn, voq, phys_lines_per_tc); } + /* init registers for pure LB TC */ qed_cmdq_lines_voq_rt_init(p_hwfn, LB_VOQ(port_id), PBF_CMDQ_PURE_LB_LINES); @@ -259,34 +260,42 @@ static void qed_btb_blocks_rt_init( struct init_qm_port_params port_params[MAX_NUM_PORTS]) { u32 usable_blocks, pure_lb_blocks, phys_blocks; - u8 tc, voq, port_id; + u8 tc, voq, port_id, num_tcs_in_port; for (port_id = 0; port_id < max_ports_per_engine; port_id++) { u32 temp; - u8 phys_tcs; if (!port_params[port_id].active) continue; - phys_tcs = port_params[port_id].num_active_phys_tcs; - /* subtract headroom blocks */ usable_blocks = port_params[port_id].num_btb_blocks - BTB_HEADROOM_BLOCKS; - /* find blocks per physical TC. use factor to avoid - * floating arithmethic. - */ + /* find blocks per physical TC */ + num_tcs_in_port = 0; + for (tc = 0; tc < NUM_OF_PHYS_TCS; tc++) { + if (((port_params[port_id].active_phys_tcs >> + tc) & 0x1) == 1) + num_tcs_in_port++; + } + pure_lb_blocks = (usable_blocks * BTB_PURE_LB_FACTOR) / - (phys_tcs * BTB_PURE_LB_FACTOR + + (num_tcs_in_port * BTB_PURE_LB_FACTOR + BTB_PURE_LB_RATIO); pure_lb_blocks = max_t(u32, BTB_JUMBO_PKT_BLOCKS, pure_lb_blocks / BTB_PURE_LB_FACTOR); - phys_blocks = (usable_blocks - pure_lb_blocks) / phys_tcs; + phys_blocks = (usable_blocks - pure_lb_blocks) / + num_tcs_in_port; /* init physical TCs */ - for (tc = 0; tc < phys_tcs; tc++) { - voq = PHYS_VOQ(port_id, tc, max_phys_tcs_per_port); + for (tc = 0; tc < NUM_OF_PHYS_TCS; tc++) { + if (((port_params[port_id].active_phys_tcs >> + tc) & 0x1) != 1) + continue; + + voq = PHYS_VOQ(port_id, tc, + max_phys_tcs_per_port); STORE_RT_REG(p_hwfn, PBF_BTB_GUARANTEED_RT_OFFSET(voq), phys_blocks); } @@ -360,10 +369,11 @@ static void qed_tx_pq_map_rt_init( memset(&tx_pq_map, 0, sizeof(tx_pq_map)); SET_FIELD(tx_pq_map.reg, QM_RF_PQ_MAP_PQ_VALID, 1); SET_FIELD(tx_pq_map.reg, QM_RF_PQ_MAP_RL_VALID, - is_vf_pq ? 1 : 0); + p_params->pq_params[i].rl_valid ? 1 : 0); SET_FIELD(tx_pq_map.reg, QM_RF_PQ_MAP_VP_PQ_ID, first_tx_pq_id); SET_FIELD(tx_pq_map.reg, QM_RF_PQ_MAP_RL_ID, - is_vf_pq ? p_params->pq_params[i].vport_id : 0); + p_params->pq_params[i].rl_valid ? + p_params->pq_params[i].vport_id : 0); SET_FIELD(tx_pq_map.reg, QM_RF_PQ_MAP_VOQ, voq); SET_FIELD(tx_pq_map.reg, QM_RF_PQ_MAP_WRR_WEIGHT_GROUP, p_params->pq_params[i].wrr_group); @@ -390,25 +400,11 @@ static void qed_tx_pq_map_rt_init( /* store Tx PQ VF mask to size select register */ for (i = 0; i < num_tx_pq_vf_masks; i++) { if (tx_pq_vf_mask[i]) { - if (is_bb_a0) { - u32 curr_mask = 0, addr; - - addr = QM_REG_MAXPQSIZETXSEL_0 + (i * 4); - if (!p_params->is_first_pf) - curr_mask = qed_rd(p_hwfn, p_ptt, - addr); - - addr = QM_REG_MAXPQSIZETXSEL_0_RT_OFFSET + i; - - STORE_RT_REG(p_hwfn, addr, - curr_mask | tx_pq_vf_mask[i]); - } else { - u32 addr; + u32 addr; - addr = QM_REG_MAXPQSIZETXSEL_0_RT_OFFSET + i; - STORE_RT_REG(p_hwfn, addr, - tx_pq_vf_mask[i]); - } + addr = QM_REG_MAXPQSIZETXSEL_0_RT_OFFSET + i; + STORE_RT_REG(p_hwfn, addr, + tx_pq_vf_mask[i]); } } } @@ -418,8 +414,7 @@ static void qed_other_pq_map_rt_init(struct qed_hwfn *p_hwfn, u8 port_id, u8 pf_id, u32 num_pf_cids, - u32 num_tids, - u32 base_mem_addr_4kb) + u32 num_tids, u32 base_mem_addr_4kb) { u16 i, pq_id; @@ -465,15 +460,10 @@ static int qed_pf_wfq_rt_init(struct qed_hwfn *p_hwfn, (p_params->pf_id % MAX_NUM_PFS_BB); inc_val = QM_WFQ_INC_VAL(p_params->pf_wfq); - if (inc_val > QM_WFQ_MAX_INC_VAL) { + if (!inc_val || inc_val > QM_WFQ_MAX_INC_VAL) { DP_NOTICE(p_hwfn, "Invalid PF WFQ weight configuration"); return -1; } - STORE_RT_REG(p_hwfn, QM_REG_WFQPFWEIGHT_RT_OFFSET + p_params->pf_id, - inc_val); - STORE_RT_REG(p_hwfn, - QM_REG_WFQPFUPPERBOUND_RT_OFFSET + p_params->pf_id, - QM_WFQ_UPPER_BOUND | QM_WFQ_CRD_REG_SIGN_BIT); for (i = 0; i < num_tx_pqs; i++) { u8 voq = VOQ(p_params->port_id, p_params->pq_params[i].tc_id, @@ -481,19 +471,21 @@ static int qed_pf_wfq_rt_init(struct qed_hwfn *p_hwfn, OVERWRITE_RT_REG(p_hwfn, crd_reg_offset + voq * MAX_NUM_PFS_BB, - QM_WFQ_INIT_CRD(inc_val) | QM_WFQ_CRD_REG_SIGN_BIT); } + STORE_RT_REG(p_hwfn, QM_REG_WFQPFWEIGHT_RT_OFFSET + p_params->pf_id, + inc_val); + STORE_RT_REG(p_hwfn, + QM_REG_WFQPFUPPERBOUND_RT_OFFSET + p_params->pf_id, + QM_WFQ_UPPER_BOUND | QM_WFQ_CRD_REG_SIGN_BIT); return 0; } /* Prepare PF RL runtime init values for the specified PF. * Return -1 on error. */ -static int qed_pf_rl_rt_init(struct qed_hwfn *p_hwfn, - u8 pf_id, - u32 pf_rl) +static int qed_pf_rl_rt_init(struct qed_hwfn *p_hwfn, u8 pf_id, u32 pf_rl) { u32 inc_val = QM_RL_INC_VAL(pf_rl); @@ -607,9 +599,7 @@ static bool qed_poll_on_qm_cmd_ready(struct qed_hwfn *p_hwfn, static bool qed_send_qm_cmd(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - u32 cmd_addr, - u32 cmd_data_lsb, - u32 cmd_data_msb) + u32 cmd_addr, u32 cmd_data_lsb, u32 cmd_data_msb) { if (!qed_poll_on_qm_cmd_ready(p_hwfn, p_ptt)) return false; @@ -627,9 +617,7 @@ static bool qed_send_qm_cmd(struct qed_hwfn *p_hwfn, u32 qed_qm_pf_mem_size(u8 pf_id, u32 num_pf_cids, u32 num_vf_cids, - u32 num_tids, - u16 num_pf_pqs, - u16 num_vf_pqs) + u32 num_tids, u16 num_pf_pqs, u16 num_vf_pqs) { return QM_PQ_MEM_4KB(num_pf_cids) * num_pf_pqs + QM_PQ_MEM_4KB(num_vf_cids) * num_vf_pqs + @@ -713,8 +701,7 @@ int qed_qm_pf_rt_init(struct qed_hwfn *p_hwfn, } int qed_init_pf_wfq(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u8 pf_id, u16 pf_wfq) + struct qed_ptt *p_ptt, u8 pf_id, u16 pf_wfq) { u32 inc_val = QM_WFQ_INC_VAL(pf_wfq); @@ -728,9 +715,7 @@ int qed_init_pf_wfq(struct qed_hwfn *p_hwfn, } int qed_init_pf_rl(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u8 pf_id, - u32 pf_rl) + struct qed_ptt *p_ptt, u8 pf_id, u32 pf_rl) { u32 inc_val = QM_RL_INC_VAL(pf_rl); @@ -749,8 +734,7 @@ int qed_init_pf_rl(struct qed_hwfn *p_hwfn, int qed_init_vport_wfq(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - u16 first_tx_pq_id[NUM_OF_TCS], - u16 vport_wfq) + u16 first_tx_pq_id[NUM_OF_TCS], u16 vport_wfq) { u32 inc_val = QM_WFQ_INC_VAL(vport_wfq); u8 tc; @@ -773,9 +757,7 @@ int qed_init_vport_wfq(struct qed_hwfn *p_hwfn, } int qed_init_vport_rl(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u8 vport_id, - u32 vport_rl) + struct qed_ptt *p_ptt, u8 vport_id, u32 vport_rl) { u32 inc_val = QM_RL_INC_VAL(vport_rl); @@ -795,9 +777,7 @@ int qed_init_vport_rl(struct qed_hwfn *p_hwfn, bool qed_send_qm_stop_cmd(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, bool is_release_cmd, - bool is_tx_pq, - u16 start_pq, - u16 num_pqs) + bool is_tx_pq, u16 start_pq, u16 num_pqs) { u32 cmd_arr[QM_CMD_STRUCT_SIZE(QM_STOP_CMD)] = { 0 }; u32 pq_mask = 0, last_pq = start_pq + num_pqs - 1, pq_id; @@ -841,17 +821,15 @@ qed_set_tunnel_type_enable_bit(unsigned long *var, int bit, bool enable) #define PRS_ETH_TUNN_FIC_FORMAT -188897008 void qed_set_vxlan_dest_port(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u16 dest_port) + struct qed_ptt *p_ptt, u16 dest_port) { qed_wr(p_hwfn, p_ptt, PRS_REG_VXLAN_PORT, dest_port); - qed_wr(p_hwfn, p_ptt, NIG_REG_VXLAN_PORT, dest_port); + qed_wr(p_hwfn, p_ptt, NIG_REG_VXLAN_CTRL, dest_port); qed_wr(p_hwfn, p_ptt, PBF_REG_VXLAN_PORT, dest_port); } void qed_set_vxlan_enable(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - bool vxlan_enable) + struct qed_ptt *p_ptt, bool vxlan_enable) { unsigned long reg_val = 0; u8 shift; @@ -908,8 +886,7 @@ void qed_set_gre_enable(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, } void qed_set_geneve_dest_port(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - u16 dest_port) + struct qed_ptt *p_ptt, u16 dest_port) { qed_wr(p_hwfn, p_ptt, PRS_REG_NGE_PORT, dest_port); qed_wr(p_hwfn, p_ptt, NIG_REG_NGE_PORT, dest_port); @@ -918,8 +895,7 @@ void qed_set_geneve_dest_port(struct qed_hwfn *p_hwfn, void qed_set_geneve_enable(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, - bool eth_geneve_enable, - bool ip_geneve_enable) + bool eth_geneve_enable, bool ip_geneve_enable) { unsigned long reg_val = 0; u8 shift; diff --git a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c index d358c3b..9866a20 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_init_ops.c +++ b/drivers/net/ethernet/qlogic/qed/qed_init_ops.c @@ -543,8 +543,7 @@ void qed_gtt_init(struct qed_hwfn *p_hwfn) pxp_global_win[i]); } -int qed_init_fw_data(struct qed_dev *cdev, - const u8 *data) +int qed_init_fw_data(struct qed_dev *cdev, const u8 *data) { struct qed_fw_data *fw = cdev->fw_data; struct bin_buffer_hdr *buf_hdr; @@ -555,7 +554,11 @@ int qed_init_fw_data(struct qed_dev *cdev, return -EINVAL; } - buf_hdr = (struct bin_buffer_hdr *)data; + /* First Dword contains metadata and should be skipped */ + buf_hdr = (struct bin_buffer_hdr *)(data + sizeof(u32)); + + offset = buf_hdr[BIN_BUF_FW_VER_INFO].offset; + fw->fw_ver_info = (struct fw_ver_info *)(data + offset); offset = buf_hdr[BIN_BUF_INIT_CMD].offset; fw->init_ops = (union init_op *)(data + offset); diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c index 8fba87dd..2ee496e 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_l2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c @@ -575,9 +575,12 @@ int qed_sp_eth_rxq_start_ramrod(struct qed_hwfn *p_hwfn, p_ramrod->num_of_pbl_pages = cpu_to_le16(cqe_pbl_size); DMA_REGPAIR_LE(p_ramrod->cqe_pbl_addr, cqe_pbl_addr); - rc = qed_spq_post(p_hwfn, p_ent, NULL); + p_ramrod->vf_rx_prod_index = params->vf_qid; + if (params->vf_qid) + DP_VERBOSE(p_hwfn, QED_MSG_SP, + "Queue is meant for VF rxq[%04x]\n", params->vf_qid); - return rc; + return qed_spq_post(p_hwfn, p_ent, NULL); } static int @@ -615,7 +618,7 @@ qed_sp_eth_rx_queue_start(struct qed_hwfn *p_hwfn, *pp_prod = (u8 __iomem *)p_hwfn->regview + GTT_BAR0_MAP_REG_MSDM_RAM + - MSTORM_PRODS_OFFSET(abs_l2_queue); + MSTORM_ETH_PF_PRODS_OFFSET(abs_l2_queue); /* Init the rcq, rx bd and rx sge (if valid) producers to 0 */ __internal_ram_wr(p_hwfn, *pp_prod, sizeof(u64), @@ -759,9 +762,9 @@ int qed_sp_eth_txq_start_ramrod(struct qed_hwfn *p_hwfn, struct qed_spq_entry *p_ent = NULL; struct qed_sp_init_data init_data; struct qed_hw_cid_data *p_tx_cid; - u8 abs_vport_id; + u16 pq_id, abs_tx_q_id = 0; int rc = -EINVAL; - u16 pq_id; + u8 abs_vport_id; /* Store information for the stop */ p_tx_cid = &p_hwfn->p_tx_cids[p_params->queue_id]; @@ -772,6 +775,10 @@ int qed_sp_eth_txq_start_ramrod(struct qed_hwfn *p_hwfn, if (rc) return rc; + rc = qed_fw_l2_queue(p_hwfn, p_params->queue_id, &abs_tx_q_id); + if (rc) + return rc; + /* Get SPQ entry */ memset(&init_data, 0, sizeof(init_data)); init_data.cid = cid; @@ -791,6 +798,7 @@ int qed_sp_eth_txq_start_ramrod(struct qed_hwfn *p_hwfn, p_ramrod->sb_index = p_params->sb_idx; p_ramrod->stats_counter_id = stats_id; + p_ramrod->queue_zone_id = cpu_to_le16(abs_tx_q_id); p_ramrod->pbl_size = cpu_to_le16(pbl_size); DMA_REGPAIR_LE(p_ramrod->pbl_base_addr, pbl_addr); @@ -1485,51 +1493,51 @@ static void __qed_get_vport_port_stats(struct qed_hwfn *p_hwfn, offsetof(struct public_port, stats), sizeof(port_stats)); - p_stats->rx_64_byte_packets += port_stats.pmm.r64; - p_stats->rx_65_to_127_byte_packets += port_stats.pmm.r127; - p_stats->rx_128_to_255_byte_packets += port_stats.pmm.r255; - p_stats->rx_256_to_511_byte_packets += port_stats.pmm.r511; - p_stats->rx_512_to_1023_byte_packets += port_stats.pmm.r1023; - p_stats->rx_1024_to_1518_byte_packets += port_stats.pmm.r1518; - p_stats->rx_1519_to_1522_byte_packets += port_stats.pmm.r1522; - p_stats->rx_1519_to_2047_byte_packets += port_stats.pmm.r2047; - p_stats->rx_2048_to_4095_byte_packets += port_stats.pmm.r4095; - p_stats->rx_4096_to_9216_byte_packets += port_stats.pmm.r9216; - p_stats->rx_9217_to_16383_byte_packets += port_stats.pmm.r16383; - p_stats->rx_crc_errors += port_stats.pmm.rfcs; - p_stats->rx_mac_crtl_frames += port_stats.pmm.rxcf; - p_stats->rx_pause_frames += port_stats.pmm.rxpf; - p_stats->rx_pfc_frames += port_stats.pmm.rxpp; - p_stats->rx_align_errors += port_stats.pmm.raln; - p_stats->rx_carrier_errors += port_stats.pmm.rfcr; - p_stats->rx_oversize_packets += port_stats.pmm.rovr; - p_stats->rx_jabbers += port_stats.pmm.rjbr; - p_stats->rx_undersize_packets += port_stats.pmm.rund; - p_stats->rx_fragments += port_stats.pmm.rfrg; - p_stats->tx_64_byte_packets += port_stats.pmm.t64; - p_stats->tx_65_to_127_byte_packets += port_stats.pmm.t127; - p_stats->tx_128_to_255_byte_packets += port_stats.pmm.t255; - p_stats->tx_256_to_511_byte_packets += port_stats.pmm.t511; - p_stats->tx_512_to_1023_byte_packets += port_stats.pmm.t1023; - p_stats->tx_1024_to_1518_byte_packets += port_stats.pmm.t1518; - p_stats->tx_1519_to_2047_byte_packets += port_stats.pmm.t2047; - p_stats->tx_2048_to_4095_byte_packets += port_stats.pmm.t4095; - p_stats->tx_4096_to_9216_byte_packets += port_stats.pmm.t9216; - p_stats->tx_9217_to_16383_byte_packets += port_stats.pmm.t16383; - p_stats->tx_pause_frames += port_stats.pmm.txpf; - p_stats->tx_pfc_frames += port_stats.pmm.txpp; - p_stats->tx_lpi_entry_count += port_stats.pmm.tlpiec; - p_stats->tx_total_collisions += port_stats.pmm.tncl; - p_stats->rx_mac_bytes += port_stats.pmm.rbyte; - p_stats->rx_mac_uc_packets += port_stats.pmm.rxuca; - p_stats->rx_mac_mc_packets += port_stats.pmm.rxmca; - p_stats->rx_mac_bc_packets += port_stats.pmm.rxbca; - p_stats->rx_mac_frames_ok += port_stats.pmm.rxpok; - p_stats->tx_mac_bytes += port_stats.pmm.tbyte; - p_stats->tx_mac_uc_packets += port_stats.pmm.txuca; - p_stats->tx_mac_mc_packets += port_stats.pmm.txmca; - p_stats->tx_mac_bc_packets += port_stats.pmm.txbca; - p_stats->tx_mac_ctrl_frames += port_stats.pmm.txcf; + p_stats->rx_64_byte_packets += port_stats.eth.r64; + p_stats->rx_65_to_127_byte_packets += port_stats.eth.r127; + p_stats->rx_128_to_255_byte_packets += port_stats.eth.r255; + p_stats->rx_256_to_511_byte_packets += port_stats.eth.r511; + p_stats->rx_512_to_1023_byte_packets += port_stats.eth.r1023; + p_stats->rx_1024_to_1518_byte_packets += port_stats.eth.r1518; + p_stats->rx_1519_to_1522_byte_packets += port_stats.eth.r1522; + p_stats->rx_1519_to_2047_byte_packets += port_stats.eth.r2047; + p_stats->rx_2048_to_4095_byte_packets += port_stats.eth.r4095; + p_stats->rx_4096_to_9216_byte_packets += port_stats.eth.r9216; + p_stats->rx_9217_to_16383_byte_packets += port_stats.eth.r16383; + p_stats->rx_crc_errors += port_stats.eth.rfcs; + p_stats->rx_mac_crtl_frames += port_stats.eth.rxcf; + p_stats->rx_pause_frames += port_stats.eth.rxpf; + p_stats->rx_pfc_frames += port_stats.eth.rxpp; + p_stats->rx_align_errors += port_stats.eth.raln; + p_stats->rx_carrier_errors += port_stats.eth.rfcr; + p_stats->rx_oversize_packets += port_stats.eth.rovr; + p_stats->rx_jabbers += port_stats.eth.rjbr; + p_stats->rx_undersize_packets += port_stats.eth.rund; + p_stats->rx_fragments += port_stats.eth.rfrg; + p_stats->tx_64_byte_packets += port_stats.eth.t64; + p_stats->tx_65_to_127_byte_packets += port_stats.eth.t127; + p_stats->tx_128_to_255_byte_packets += port_stats.eth.t255; + p_stats->tx_256_to_511_byte_packets += port_stats.eth.t511; + p_stats->tx_512_to_1023_byte_packets += port_stats.eth.t1023; + p_stats->tx_1024_to_1518_byte_packets += port_stats.eth.t1518; + p_stats->tx_1519_to_2047_byte_packets += port_stats.eth.t2047; + p_stats->tx_2048_to_4095_byte_packets += port_stats.eth.t4095; + p_stats->tx_4096_to_9216_byte_packets += port_stats.eth.t9216; + p_stats->tx_9217_to_16383_byte_packets += port_stats.eth.t16383; + p_stats->tx_pause_frames += port_stats.eth.txpf; + p_stats->tx_pfc_frames += port_stats.eth.txpp; + p_stats->tx_lpi_entry_count += port_stats.eth.tlpiec; + p_stats->tx_total_collisions += port_stats.eth.tncl; + p_stats->rx_mac_bytes += port_stats.eth.rbyte; + p_stats->rx_mac_uc_packets += port_stats.eth.rxuca; + p_stats->rx_mac_mc_packets += port_stats.eth.rxmca; + p_stats->rx_mac_bc_packets += port_stats.eth.rxbca; + p_stats->rx_mac_frames_ok += port_stats.eth.rxpok; + p_stats->tx_mac_bytes += port_stats.eth.tbyte; + p_stats->tx_mac_uc_packets += port_stats.eth.txuca; + p_stats->tx_mac_mc_packets += port_stats.eth.txmca; + p_stats->tx_mac_bc_packets += port_stats.eth.txbca; + p_stats->tx_mac_ctrl_frames += port_stats.eth.txcf; for (j = 0; j < 8; j++) { p_stats->brb_truncates += port_stats.brb.brb_truncate[j]; p_stats->brb_discards += port_stats.brb.brb_discard[j]; diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 7530646..c807f67 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -832,7 +832,8 @@ static int qed_slowpath_start(struct qed_dev *cdev, goto err2; } - data = cdev->firmware->data; + /* First Dword used to diffrentiate between various sources */ + data = cdev->firmware->data + sizeof(u32); } memset(&tunn_info, 0, sizeof(tunn_info)); @@ -991,8 +992,7 @@ static bool qed_can_link_change(struct qed_dev *cdev) return true; } -static int qed_set_link(struct qed_dev *cdev, - struct qed_link_params *params) +static int qed_set_link(struct qed_dev *cdev, struct qed_link_params *params) { struct qed_hwfn *hwfn; struct qed_mcp_link_params *link_params; @@ -1032,7 +1032,7 @@ static int qed_set_link(struct qed_dev *cdev, NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G; if (params->adv_speeds & 0) link_params->speed.advertised_speeds |= - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_100G; + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_BB_100G; } if (params->override_flags & QED_LINK_OVERRIDE_SPEED_FORCED_SPEED) link_params->speed.forced_speed = params->forced_speed; @@ -1053,19 +1053,19 @@ static int qed_set_link(struct qed_dev *cdev, if (params->override_flags & QED_LINK_OVERRIDE_LOOPBACK_MODE) { switch (params->loopback_mode) { case QED_LINK_LOOPBACK_INT_PHY: - link_params->loopback_mode = PMM_LOOPBACK_INT_PHY; + link_params->loopback_mode = ETH_LOOPBACK_INT_PHY; break; case QED_LINK_LOOPBACK_EXT_PHY: - link_params->loopback_mode = PMM_LOOPBACK_EXT_PHY; + link_params->loopback_mode = ETH_LOOPBACK_EXT_PHY; break; case QED_LINK_LOOPBACK_EXT: - link_params->loopback_mode = PMM_LOOPBACK_EXT; + link_params->loopback_mode = ETH_LOOPBACK_EXT; break; case QED_LINK_LOOPBACK_MAC: - link_params->loopback_mode = PMM_LOOPBACK_MAC; + link_params->loopback_mode = ETH_LOOPBACK_MAC; break; default: - link_params->loopback_mode = PMM_LOOPBACK_NONE; + link_params->loopback_mode = ETH_LOOPBACK_NONE; break; } } @@ -1157,7 +1157,7 @@ static void qed_fill_link(struct qed_hwfn *hwfn, NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G) if_link->advertised_caps |= 0; if (params.speed.advertised_speeds & - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_100G) + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_BB_100G) if_link->advertised_caps |= 0; if (link_caps.speed_capabilities & @@ -1174,7 +1174,7 @@ static void qed_fill_link(struct qed_hwfn *hwfn, NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_50G) if_link->supported_caps |= 0; if (link_caps.speed_capabilities & - NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_100G) + NVM_CFG1_PORT_DRV_SPEED_CAPABILITY_MASK_BB_100G) if_link->supported_caps |= 0; if (link.link_up) diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c index 1182361..2c143b3 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c @@ -531,9 +531,9 @@ static void qed_mcp_handle_transceiver_change(struct qed_hwfn *p_hwfn, transceiver_data))); transceiver_state = GET_FIELD(transceiver_state, - PMM_TRANSCEIVER_STATE); + ETH_TRANSCEIVER_STATE); - if (transceiver_state == PMM_TRANSCEIVER_STATE_PRESENT) + if (transceiver_state == ETH_TRANSCEIVER_STATE_PRESENT) DP_NOTICE(p_hwfn, "Transceiver is present.\n"); else DP_NOTICE(p_hwfn, "Transceiver is unplugged.\n"); @@ -668,14 +668,12 @@ static void qed_mcp_handle_link_change(struct qed_hwfn *p_hwfn, qed_link_update(p_hwfn); } -int qed_mcp_set_link(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, - bool b_up) +int qed_mcp_set_link(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, bool b_up) { struct qed_mcp_link_params *params = &p_hwfn->mcp_info->link_input; struct qed_mcp_mb_params mb_params; union drv_union_data union_data; - struct pmm_phy_cfg *phy_cfg; + struct eth_phy_cfg *phy_cfg; int rc = 0; u32 cmd; @@ -685,9 +683,9 @@ int qed_mcp_set_link(struct qed_hwfn *p_hwfn, cmd = b_up ? DRV_MSG_CODE_INIT_PHY : DRV_MSG_CODE_LINK_RESET; if (!params->speed.autoneg) phy_cfg->speed = params->speed.forced_speed; - phy_cfg->pause |= (params->pause.autoneg) ? PMM_PAUSE_AUTONEG : 0; - phy_cfg->pause |= (params->pause.forced_rx) ? PMM_PAUSE_RX : 0; - phy_cfg->pause |= (params->pause.forced_tx) ? PMM_PAUSE_TX : 0; + phy_cfg->pause |= (params->pause.autoneg) ? ETH_PAUSE_AUTONEG : 0; + phy_cfg->pause |= (params->pause.forced_rx) ? ETH_PAUSE_RX : 0; + phy_cfg->pause |= (params->pause.forced_tx) ? ETH_PAUSE_TX : 0; phy_cfg->adv_speed = params->speed.advertised_speeds; phy_cfg->loopback_mode = params->loopback_mode; @@ -773,6 +771,34 @@ static u32 qed_mcp_get_shmem_func(struct qed_hwfn *p_hwfn, return size; } +int qed_hw_init_first_eth(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, u8 *p_pf) +{ + struct public_func shmem_info; + int i; + + /* Find first Ethernet interface in port */ + for (i = 0; i < NUM_OF_ENG_PFS(p_hwfn->cdev); + i += p_hwfn->cdev->num_ports_in_engines) { + qed_mcp_get_shmem_func(p_hwfn, p_ptt, &shmem_info, + MCP_PF_ID_BY_REL(p_hwfn, i)); + + if (shmem_info.config & FUNC_MF_CFG_FUNC_HIDE) + continue; + + if ((shmem_info.config & FUNC_MF_CFG_PROTOCOL_MASK) == + FUNC_MF_CFG_PROTOCOL_ETHERNET) { + *p_pf = (u8)i; + return 0; + } + } + + DP_NOTICE(p_hwfn, + "Failed to find on port an ethernet interface in MF_SI mode\n"); + + return -EINVAL; +} + static void qed_mcp_update_bw(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h index 6dd59eb..7f319aa 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h @@ -457,4 +457,7 @@ int __qed_configure_pf_min_bandwidth(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_mcp_link_state *p_link, u8 min_bw); + +int qed_hw_init_first_eth(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, u8 *p_pf); #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h index 3a6c506..b889585 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h +++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h @@ -167,6 +167,10 @@ 0x1800004UL #define NIG_REG_CM_HDR \ 0x500840UL +#define NIG_REG_LLH_TAGMAC_DEF_PF_VECTOR \ + 0x50196cUL +#define NIG_REG_LLH_CLS_TYPE_DUALMODE \ + 0x501964UL #define NCSI_REG_CONFIG \ 0x040200UL #define PBF_REG_INIT \ @@ -219,6 +223,8 @@ 0x230000UL #define PRS_REG_SOFT_RST \ 0x1f0000UL +#define PRS_REG_MSG_INFO \ + 0x1f0a1cUL #define PSDM_REG_ENABLE_IN1 \ 0xfa0004UL #define PSEM_REG_ENABLE_IN \ @@ -460,7 +466,7 @@ #define NIG_REG_ENC_TYPE_ENABLE_VXLAN_ENABLE (0x1 << 2) #define NIG_REG_ENC_TYPE_ENABLE_VXLAN_ENABLE_SHIFT 2 -#define NIG_REG_VXLAN_PORT 0x50105cUL +#define NIG_REG_VXLAN_CTRL 0x50105cUL #define PBF_REG_VXLAN_PORT 0xd80518UL #define PBF_REG_NGE_PORT 0xd8051cUL #define PRS_REG_NGE_PORT 0x1f086cUL diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c index 67f6ce3..1225064 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c @@ -332,7 +332,7 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, p_ramrod->path_id = QED_PATH_ID(p_hwfn); p_ramrod->dont_log_ramrods = 0; p_ramrod->log_type_mask = cpu_to_le16(0xf); - p_ramrod->mf_mode = mode; + switch (mode) { case QED_MF_DEFAULT: case QED_MF_NPAR: @@ -368,6 +368,8 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, p_ramrod->base_vf_id = (u8) p_iov->first_vf_in_pf; p_ramrod->num_vfs = (u8) p_iov->total_vfs; } + p_ramrod->hsi_fp_ver.major_ver_arr[ETH_VER_KEY] = ETH_HSI_VER_MAJOR; + p_ramrod->hsi_fp_ver.minor_ver_arr[ETH_VER_KEY] = ETH_HSI_VER_MINOR; DP_VERBOSE(p_hwfn, QED_MSG_SPQ, "Setting event_ring_sb [id %04x index %02x], outer_tag [%d]\n", diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index c325ee8..eb75b82 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -47,6 +47,8 @@ static int qed_sp_vf_start(struct qed_hwfn *p_hwfn, p_ramrod->opaque_fid = cpu_to_le16(opaque_vfid); p_ramrod->personality = PERSONALITY_ETH; + p_ramrod->hsi_fp_ver.major_ver_arr[ETH_VER_KEY] = ETH_HSI_VER_MAJOR; + p_ramrod->hsi_fp_ver.minor_ver_arr[ETH_VER_KEY] = ETH_HSI_VER_MINOR; return qed_spq_post(p_hwfn, p_ent, NULL); } @@ -1585,10 +1587,6 @@ static void qed_iov_vf_mbx_stop_vport(struct qed_hwfn *p_hwfn, sizeof(struct pfvf_def_resp_tlv), status); } -#define TSTORM_QZONE_START PXP_VF_BAR0_START_SDM_ZONE_A -#define MSTORM_QZONE_START(dev) (TSTORM_QZONE_START + \ - (TSTORM_QZONE_SIZE * NUM_OF_L2_QUEUES(dev))) - static void qed_iov_vf_mbx_start_rxq_resp(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_vf_info *vf, u8 status) @@ -1606,16 +1604,11 @@ static void qed_iov_vf_mbx_start_rxq_resp(struct qed_hwfn *p_hwfn, /* Update the TLV with the response */ if (status == PFVF_STATUS_SUCCESS) { - u16 hw_qid = 0; - req = &mbx->req_virt->start_rxq; - qed_fw_l2_queue(p_hwfn, vf->vf_queues[req->rx_qid].fw_rx_qid, - &hw_qid); - - p_tlv->offset = MSTORM_QZONE_START(p_hwfn->cdev) + - hw_qid * MSTORM_QZONE_SIZE + - offsetof(struct mstorm_eth_queue_zone, - rx_producers); + p_tlv->offset = PXP_VF_BAR0_START_MSDM_ZONE_B + + offsetof(struct mstorm_vf_zone, + non_trigger.eth_rx_queue_producers) + + sizeof(struct eth_rx_prod_data) * req->rx_qid; } qed_iov_send_response(p_hwfn, p_ptt, vf, sizeof(*p_tlv), status); @@ -1634,6 +1627,7 @@ static void qed_iov_vf_mbx_start_rxq(struct qed_hwfn *p_hwfn, memset(¶ms, 0, sizeof(params)); req = &mbx->req_virt->start_rxq; params.queue_id = vf->vf_queues[req->rx_qid].fw_rx_qid; + params.vf_qid = req->rx_qid; params.vport_id = vf->vport_id; params.sb = req->hw_sb; params.sb_idx = req->sb_index; diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index ad3cae3..6836d44 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -910,6 +910,8 @@ static int qede_selftest_transmit_traffic(struct qede_dev *edev, memset(first_bd, 0, sizeof(*first_bd)); val = 1 << ETH_TX_1ST_BD_FLAGS_START_BD_SHIFT; first_bd->data.bd_flags.bitfields = val; + val = skb->len & ETH_TX_DATA_1ST_BD_PKT_LEN_MASK; + first_bd->data.bitfields |= (val << ETH_TX_DATA_1ST_BD_PKT_LEN_SHIFT); /* Map skb linear data for DMA and set in the first BD */ mapping = dma_map_single(&edev->pdev->dev, skb->data, diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 5d00d14..4f35247 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -577,8 +577,6 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, /* Fill the parsing flags & params according to the requested offload */ if (xmit_type & XMIT_L4_CSUM) { - u16 temp = 1 << ETH_TX_DATA_1ST_BD_TUNN_CFG_OVERRIDE_SHIFT; - /* We don't re-calculate IP checksum as it is already done by * the upper stack */ @@ -588,14 +586,8 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, if (xmit_type & XMIT_ENC) { first_bd->data.bd_flags.bitfields |= 1 << ETH_TX_1ST_BD_FLAGS_IP_CSUM_SHIFT; - } else { - /* In cases when OS doesn't indicate for inner offloads - * when packet is tunnelled, we need to override the HW - * tunnel configuration so that packets are treated as - * regular non tunnelled packets and no inner offloads - * are done by the hardware. - */ - first_bd->data.bitfields |= cpu_to_le16(temp); + first_bd->data.bitfields |= + 1 << ETH_TX_DATA_1ST_BD_TUNN_FLAG_SHIFT; } /* If the packet is IPv6 with extension header, indicate that @@ -653,6 +645,10 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, tx_data_bd = (struct eth_tx_bd *)third_bd; data_split = true; } + } else { + first_bd->data.bitfields |= + (skb->len & ETH_TX_DATA_1ST_BD_PKT_LEN_MASK) << + ETH_TX_DATA_1ST_BD_PKT_LEN_SHIFT; } /* Handle fragmented skb */ diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h index 3f14c7e..285189a 100644 --- a/include/linux/qed/common_hsi.h +++ b/include/linux/qed/common_hsi.h @@ -13,9 +13,19 @@ #define X_FINAL_CLEANUP_AGG_INT 1 +/* Queue Zone sizes in bytes */ +#define TSTORM_QZONE_SIZE 8 +#define MSTORM_QZONE_SIZE 0 +#define USTORM_QZONE_SIZE 8 +#define XSTORM_QZONE_SIZE 8 +#define YSTORM_QZONE_SIZE 0 +#define PSTORM_QZONE_SIZE 0 + +#define ETH_MAX_NUM_RX_QUEUES_PER_VF 16 + #define FW_MAJOR_VERSION 8 -#define FW_MINOR_VERSION 7 -#define FW_REVISION_VERSION 3 +#define FW_MINOR_VERSION 10 +#define FW_REVISION_VERSION 5 #define FW_ENGINEERING_VERSION 0 /***********************/ @@ -97,45 +107,86 @@ #define DQ_XCM_AGG_VAL_SEL_REG6 7 /* XCM agg val selection */ -#define DQ_XCM_ETH_EDPM_NUM_BDS_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD2 -#define DQ_XCM_ETH_TX_BD_CONS_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD3 -#define DQ_XCM_CORE_TX_BD_CONS_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD3 -#define DQ_XCM_ETH_TX_BD_PROD_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD4 -#define DQ_XCM_CORE_TX_BD_PROD_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD4 -#define DQ_XCM_CORE_SPQ_PROD_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD4 -#define DQ_XCM_ETH_GO_TO_BD_CONS_CMD DQ_XCM_AGG_VAL_SEL_WORD5 +#define DQ_XCM_CORE_TX_BD_CONS_CMD DQ_XCM_AGG_VAL_SEL_WORD3 +#define DQ_XCM_CORE_TX_BD_PROD_CMD DQ_XCM_AGG_VAL_SEL_WORD4 +#define DQ_XCM_CORE_SPQ_PROD_CMD DQ_XCM_AGG_VAL_SEL_WORD4 +#define DQ_XCM_ETH_EDPM_NUM_BDS_CMD DQ_XCM_AGG_VAL_SEL_WORD2 +#define DQ_XCM_ETH_TX_BD_CONS_CMD DQ_XCM_AGG_VAL_SEL_WORD3 +#define DQ_XCM_ETH_TX_BD_PROD_CMD DQ_XCM_AGG_VAL_SEL_WORD4 +#define DQ_XCM_ETH_GO_TO_BD_CONS_CMD DQ_XCM_AGG_VAL_SEL_WORD5 + +/* UCM agg val selection (HW) */ +#define DQ_UCM_AGG_VAL_SEL_WORD0 0 +#define DQ_UCM_AGG_VAL_SEL_WORD1 1 +#define DQ_UCM_AGG_VAL_SEL_WORD2 2 +#define DQ_UCM_AGG_VAL_SEL_WORD3 3 +#define DQ_UCM_AGG_VAL_SEL_REG0 4 +#define DQ_UCM_AGG_VAL_SEL_REG1 5 +#define DQ_UCM_AGG_VAL_SEL_REG2 6 +#define DQ_UCM_AGG_VAL_SEL_REG3 7 + +/* UCM agg val selection (FW) */ +#define DQ_UCM_ETH_PMD_TX_CONS_CMD DQ_UCM_AGG_VAL_SEL_WORD2 +#define DQ_UCM_ETH_PMD_RX_CONS_CMD DQ_UCM_AGG_VAL_SEL_WORD3 +#define DQ_UCM_ROCE_CQ_CONS_CMD DQ_UCM_AGG_VAL_SEL_REG0 +#define DQ_UCM_ROCE_CQ_PROD_CMD DQ_UCM_AGG_VAL_SEL_REG2 + +/* TCM agg val selection (HW) */ +#define DQ_TCM_AGG_VAL_SEL_WORD0 0 +#define DQ_TCM_AGG_VAL_SEL_WORD1 1 +#define DQ_TCM_AGG_VAL_SEL_WORD2 2 +#define DQ_TCM_AGG_VAL_SEL_WORD3 3 +#define DQ_TCM_AGG_VAL_SEL_REG1 4 +#define DQ_TCM_AGG_VAL_SEL_REG2 5 +#define DQ_TCM_AGG_VAL_SEL_REG6 6 +#define DQ_TCM_AGG_VAL_SEL_REG9 7 + +/* TCM agg val selection (FW) */ +#define DQ_TCM_L2B_BD_PROD_CMD \ + DQ_TCM_AGG_VAL_SEL_WORD1 +#define DQ_TCM_ROCE_RQ_PROD_CMD \ + DQ_TCM_AGG_VAL_SEL_WORD0 /* XCM agg counter flag selection */ -#define DQ_XCM_AGG_FLG_SHIFT_BIT14 0 -#define DQ_XCM_AGG_FLG_SHIFT_BIT15 1 -#define DQ_XCM_AGG_FLG_SHIFT_CF12 2 -#define DQ_XCM_AGG_FLG_SHIFT_CF13 3 -#define DQ_XCM_AGG_FLG_SHIFT_CF18 4 -#define DQ_XCM_AGG_FLG_SHIFT_CF19 5 -#define DQ_XCM_AGG_FLG_SHIFT_CF22 6 -#define DQ_XCM_AGG_FLG_SHIFT_CF23 7 +#define DQ_XCM_AGG_FLG_SHIFT_BIT14 0 +#define DQ_XCM_AGG_FLG_SHIFT_BIT15 1 +#define DQ_XCM_AGG_FLG_SHIFT_CF12 2 +#define DQ_XCM_AGG_FLG_SHIFT_CF13 3 +#define DQ_XCM_AGG_FLG_SHIFT_CF18 4 +#define DQ_XCM_AGG_FLG_SHIFT_CF19 5 +#define DQ_XCM_AGG_FLG_SHIFT_CF22 6 +#define DQ_XCM_AGG_FLG_SHIFT_CF23 7 /* XCM agg counter flag selection */ -#define DQ_XCM_ETH_DQ_CF_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF18) -#define DQ_XCM_CORE_DQ_CF_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF18) -#define DQ_XCM_ETH_TERMINATE_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF19) -#define DQ_XCM_CORE_TERMINATE_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF19) -#define DQ_XCM_ETH_SLOW_PATH_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF22) -#define DQ_XCM_CORE_SLOW_PATH_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF22) -#define DQ_XCM_ETH_TPH_EN_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF23) +#define DQ_XCM_CORE_DQ_CF_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF18) +#define DQ_XCM_CORE_TERMINATE_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF19) +#define DQ_XCM_CORE_SLOW_PATH_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF22) +#define DQ_XCM_ETH_DQ_CF_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF18) +#define DQ_XCM_ETH_TERMINATE_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF19) +#define DQ_XCM_ETH_SLOW_PATH_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF22) +#define DQ_XCM_ETH_TPH_EN_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF23) + +/* UCM agg counter flag selection (HW) */ +#define DQ_UCM_AGG_FLG_SHIFT_CF0 0 +#define DQ_UCM_AGG_FLG_SHIFT_CF1 1 +#define DQ_UCM_AGG_FLG_SHIFT_CF3 2 +#define DQ_UCM_AGG_FLG_SHIFT_CF4 3 +#define DQ_UCM_AGG_FLG_SHIFT_CF5 4 +#define DQ_UCM_AGG_FLG_SHIFT_CF6 5 +#define DQ_UCM_AGG_FLG_SHIFT_RULE0EN 6 +#define DQ_UCM_AGG_FLG_SHIFT_RULE1EN 7 + +/* UCM agg counter flag selection (FW) */ +#define DQ_UCM_ETH_PMD_TX_ARM_CMD (1 << DQ_UCM_AGG_FLG_SHIFT_CF4) +#define DQ_UCM_ETH_PMD_RX_ARM_CMD (1 << DQ_UCM_AGG_FLG_SHIFT_CF5) + +#define DQ_REGION_SHIFT (12) + +/* DPM */ +#define DQ_DPM_WQE_BUFF_SIZE (320) + +/* Conn type ranges */ +#define DQ_CONN_TYPE_RANGE_SHIFT (4) /*****************/ /* QM CONSTANTS */ @@ -282,8 +333,6 @@ (PXP_EXTERNAL_BAR_GLOBAL_WINDOW_START + \ PXP_EXTERNAL_BAR_GLOBAL_WINDOW_LENGTH - 1) -#define PXP_ILT_PAGE_SIZE_NUM_BITS_MIN 12 -#define PXP_ILT_BLOCK_FACTOR_MULTIPLIER 1024 #define PXP_VF_BAR0_START_IGU 0 #define PXP_VF_BAR0_IGU_LENGTH 0x3000 @@ -342,6 +391,9 @@ #define PXP_VF_BAR0_GRC_WINDOW_LENGTH 32 +#define PXP_ILT_PAGE_SIZE_NUM_BITS_MIN 12 +#define PXP_ILT_BLOCK_FACTOR_MULTIPLIER 1024 + /* ILT Records */ #define PXP_NUM_ILT_RECORDS_BB 7600 #define PXP_NUM_ILT_RECORDS_K2 11000 @@ -379,6 +431,38 @@ struct async_data { u8 fw_debug_param; }; +struct coalescing_timeset { + u8 value; +#define COALESCING_TIMESET_TIMESET_MASK 0x7F +#define COALESCING_TIMESET_TIMESET_SHIFT 0 +#define COALESCING_TIMESET_VALID_MASK 0x1 +#define COALESCING_TIMESET_VALID_SHIFT 7 +}; + +struct common_prs_pf_msg_info { + __le32 value; +#define COMMON_PRS_PF_MSG_INFO_NPAR_DEFAULT_PF_MASK 0x1 +#define COMMON_PRS_PF_MSG_INFO_NPAR_DEFAULT_PF_SHIFT 0 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_1_MASK 0x1 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_1_SHIFT 1 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_2_MASK 0x1 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_2_SHIFT 2 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_3_MASK 0x1 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_3_SHIFT 3 +#define COMMON_PRS_PF_MSG_INFO_RESERVED_MASK 0xFFFFFFF +#define COMMON_PRS_PF_MSG_INFO_RESERVED_SHIFT 4 +}; + +struct common_queue_zone { + __le16 ring_drv_data_consumer; + __le16 reserved; +}; + +struct eth_rx_prod_data { + __le16 bd_prod; + __le16 cqe_prod; +}; + struct regpair { __le32 lo; __le32 hi; @@ -388,11 +472,23 @@ struct vf_pf_channel_eqe_data { struct regpair msg_addr; }; +struct malicious_vf_eqe_data { + u8 vf_id; + u8 err_id; + __le16 reserved[3]; +}; + +struct initial_cleanup_eqe_data { + u8 vf_id; + u8 reserved[7]; +}; + /* Event Data Union */ union event_ring_data { - u8 bytes[8]; - struct vf_pf_channel_eqe_data vf_pf_channel; - struct async_data async_info; + u8 bytes[8]; + struct vf_pf_channel_eqe_data vf_pf_channel; + struct malicious_vf_eqe_data malicious_vf; + struct initial_cleanup_eqe_data vf_init_cleanup; }; /* Event Ring Entry */ @@ -433,6 +529,16 @@ enum protocol_type { MAX_PROTOCOL_TYPE }; +struct ustorm_eth_queue_zone { + struct coalescing_timeset int_coalescing_timeset; + u8 reserved[3]; +}; + +struct ustorm_queue_zone { + struct ustorm_eth_queue_zone eth; + struct common_queue_zone common; +}; + /* status block structure */ struct cau_pi_entry { u32 prod; @@ -683,19 +789,4 @@ struct status_block { #define STATUS_BLOCK_ZERO_PAD3_SHIFT 24 }; -struct tunnel_parsing_flags { - u8 flags; -#define TUNNEL_PARSING_FLAGS_TYPE_MASK 0x3 -#define TUNNEL_PARSING_FLAGS_TYPE_SHIFT 0 -#define TUNNEL_PARSING_FLAGS_TENNANT_ID_EXIST_MASK 0x1 -#define TUNNEL_PARSING_FLAGS_TENNANT_ID_EXIST_SHIFT 2 -#define TUNNEL_PARSING_FLAGS_NEXT_PROTOCOL_MASK 0x3 -#define TUNNEL_PARSING_FLAGS_NEXT_PROTOCOL_SHIFT 3 -#define TUNNEL_PARSING_FLAGS_FIRSTHDRIPMATCH_MASK 0x1 -#define TUNNEL_PARSING_FLAGS_FIRSTHDRIPMATCH_SHIFT 5 -#define TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_MASK 0x1 -#define TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_SHIFT 6 -#define TUNNEL_PARSING_FLAGS_IPV4_OPTIONS_MASK 0x1 -#define TUNNEL_PARSING_FLAGS_IPV4_OPTIONS_SHIFT 7 -}; #endif /* __COMMON_HSI__ */ diff --git a/include/linux/qed/eth_common.h b/include/linux/qed/eth_common.h index 092cb0c..b5ebc69 100644 --- a/include/linux/qed/eth_common.h +++ b/include/linux/qed/eth_common.h @@ -12,6 +12,8 @@ /********************/ /* ETH FW CONSTANTS */ /********************/ +#define ETH_HSI_VER_MAJOR 3 +#define ETH_HSI_VER_MINOR 0 #define ETH_CACHE_LINE_SIZE 64 #define ETH_MAX_RAMROD_PER_CON 8 @@ -57,19 +59,6 @@ #define ETH_TPA_CQE_CONT_LEN_LIST_SIZE 6 #define ETH_TPA_CQE_END_LEN_LIST_SIZE 4 -/* Queue Zone sizes */ -#define TSTORM_QZONE_SIZE 0 -#define MSTORM_QZONE_SIZE sizeof(struct mstorm_eth_queue_zone) -#define USTORM_QZONE_SIZE sizeof(struct ustorm_eth_queue_zone) -#define XSTORM_QZONE_SIZE 0 -#define YSTORM_QZONE_SIZE sizeof(struct ystorm_eth_queue_zone) -#define PSTORM_QZONE_SIZE 0 - -/* Interrupt coalescing TimeSet */ -struct coalescing_timeset { - u8 timeset; - u8 valid; -}; struct eth_tx_1st_bd_flags { u8 bitfields; @@ -97,12 +86,12 @@ struct eth_tx_data_1st_bd { u8 nbds; struct eth_tx_1st_bd_flags bd_flags; __le16 bitfields; -#define ETH_TX_DATA_1ST_BD_TUNN_CFG_OVERRIDE_MASK 0x1 -#define ETH_TX_DATA_1ST_BD_TUNN_CFG_OVERRIDE_SHIFT 0 +#define ETH_TX_DATA_1ST_BD_TUNN_FLAG_MASK 0x1 +#define ETH_TX_DATA_1ST_BD_TUNN_FLAG_SHIFT 0 #define ETH_TX_DATA_1ST_BD_RESERVED0_MASK 0x1 #define ETH_TX_DATA_1ST_BD_RESERVED0_SHIFT 1 -#define ETH_TX_DATA_1ST_BD_FW_USE_ONLY_MASK 0x3FFF -#define ETH_TX_DATA_1ST_BD_FW_USE_ONLY_SHIFT 2 +#define ETH_TX_DATA_1ST_BD_PKT_LEN_MASK 0x3FFF +#define ETH_TX_DATA_1ST_BD_PKT_LEN_SHIFT 2 }; /* The parsing information data for the second tx bd of a given packet. */ @@ -136,28 +125,51 @@ struct eth_tx_data_2nd_bd { #define ETH_TX_DATA_2ND_BD_RESERVED0_SHIFT 13 }; +struct eth_fast_path_cqe_fw_debug { + u8 reserved0; + u8 reserved1; + __le16 reserved2; +}; + +/* tunneling parsing flags */ +struct eth_tunnel_parsing_flags { + u8 flags; +#define ETH_TUNNEL_PARSING_FLAGS_TYPE_MASK 0x3 +#define ETH_TUNNEL_PARSING_FLAGS_TYPE_SHIFT 0 +#define ETH_TUNNEL_PARSING_FLAGS_TENNANT_ID_EXIST_MASK 0x1 +#define ETH_TUNNEL_PARSING_FLAGS_TENNANT_ID_EXIST_SHIFT 2 +#define ETH_TUNNEL_PARSING_FLAGS_NEXT_PROTOCOL_MASK 0x3 +#define ETH_TUNNEL_PARSING_FLAGS_NEXT_PROTOCOL_SHIFT 3 +#define ETH_TUNNEL_PARSING_FLAGS_FIRSTHDRIPMATCH_MASK 0x1 +#define ETH_TUNNEL_PARSING_FLAGS_FIRSTHDRIPMATCH_SHIFT 5 +#define ETH_TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_MASK 0x1 +#define ETH_TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_SHIFT 6 +#define ETH_TUNNEL_PARSING_FLAGS_IPV4_OPTIONS_MASK 0x1 +#define ETH_TUNNEL_PARSING_FLAGS_IPV4_OPTIONS_SHIFT 7 +}; + /* Regular ETH Rx FP CQE. */ struct eth_fast_path_rx_reg_cqe { - u8 type; - u8 bitfields; + u8 type; + u8 bitfields; #define ETH_FAST_PATH_RX_REG_CQE_RSS_HASH_TYPE_MASK 0x7 #define ETH_FAST_PATH_RX_REG_CQE_RSS_HASH_TYPE_SHIFT 0 #define ETH_FAST_PATH_RX_REG_CQE_TC_MASK 0xF #define ETH_FAST_PATH_RX_REG_CQE_TC_SHIFT 3 #define ETH_FAST_PATH_RX_REG_CQE_RESERVED0_MASK 0x1 #define ETH_FAST_PATH_RX_REG_CQE_RESERVED0_SHIFT 7 - __le16 pkt_len; - struct parsing_and_err_flags pars_flags; - __le16 vlan_tag; - __le32 rss_hash; - __le16 len_on_first_bd; - u8 placement_offset; - struct tunnel_parsing_flags tunnel_pars_flags; - u8 bd_num; - u8 reserved[7]; - u32 fw_debug; - u8 reserved1[3]; - u8 flags; + __le16 pkt_len; + struct parsing_and_err_flags pars_flags; + __le16 vlan_tag; + __le32 rss_hash; + __le16 len_on_first_bd; + u8 placement_offset; + struct eth_tunnel_parsing_flags tunnel_pars_flags; + u8 bd_num; + u8 reserved[7]; + struct eth_fast_path_cqe_fw_debug fw_debug; + u8 reserved1[3]; + u8 flags; #define ETH_FAST_PATH_RX_REG_CQE_VALID_MASK 0x1 #define ETH_FAST_PATH_RX_REG_CQE_VALID_SHIFT 0 #define ETH_FAST_PATH_RX_REG_CQE_VALID_TOGGLE_MASK 0x1 @@ -207,11 +219,11 @@ struct eth_fast_path_rx_tpa_start_cqe { __le32 rss_hash; __le16 len_on_first_bd; u8 placement_offset; - struct tunnel_parsing_flags tunnel_pars_flags; + struct eth_tunnel_parsing_flags tunnel_pars_flags; u8 tpa_agg_index; u8 header_len; __le16 ext_bd_len_list[ETH_TPA_CQE_START_LEN_LIST_SIZE]; - u32 fw_debug; + struct eth_fast_path_cqe_fw_debug fw_debug; }; /* The L4 pseudo checksum mode for Ethernet */ @@ -264,12 +276,25 @@ enum eth_rx_cqe_type { MAX_ETH_RX_CQE_TYPE }; -/* ETH Rx producers data */ -struct eth_rx_prod_data { - __le16 bd_prod; - __le16 cqe_prod; - __le16 reserved; - __le16 reserved1; +enum eth_rx_tunn_type { + ETH_RX_NO_TUNN, + ETH_RX_TUNN_GENEVE, + ETH_RX_TUNN_GRE, + ETH_RX_TUNN_VXLAN, + MAX_ETH_RX_TUNN_TYPE +}; + +/* Aggregation end reason. */ +enum eth_tpa_end_reason { + ETH_AGG_END_UNUSED, + ETH_AGG_END_SP_UPDATE, + ETH_AGG_END_MAX_LEN, + ETH_AGG_END_LAST_SEG, + ETH_AGG_END_TIMEOUT, + ETH_AGG_END_NOT_CONSISTENT, + ETH_AGG_END_OUT_OF_ORDER, + ETH_AGG_END_NON_TPA_SEG, + MAX_ETH_TPA_END_REASON }; /* The first tx bd of a given packet */ @@ -337,21 +362,18 @@ union eth_tx_bd_types { }; /* Mstorm Queue Zone */ -struct mstorm_eth_queue_zone { - struct eth_rx_prod_data rx_producers; - __le32 reserved[2]; -}; - -/* Ustorm Queue Zone */ -struct ustorm_eth_queue_zone { - struct coalescing_timeset int_coalescing_timeset; - __le16 reserved[3]; +enum eth_tx_tunn_type { + ETH_TX_TUNN_GENEVE, + ETH_TX_TUNN_TTAG, + ETH_TX_TUNN_GRE, + ETH_TX_TUNN_VXLAN, + MAX_ETH_TX_TUNN_TYPE }; /* Ystorm Queue Zone */ -struct ystorm_eth_queue_zone { - struct coalescing_timeset int_coalescing_timeset; - __le16 reserved[3]; +struct xstorm_eth_queue_zone { + struct coalescing_timeset int_coalescing_timeset; + u8 reserved[7]; }; /* ETH doorbell data */ diff --git a/include/linux/qed/qed_eth_if.h b/include/linux/qed/qed_eth_if.h index 6ae8cb4..f8ff711 100644 --- a/include/linux/qed/qed_eth_if.h +++ b/include/linux/qed/qed_eth_if.h @@ -113,6 +113,7 @@ struct qed_queue_start_common_params { u8 vport_id; u16 sb; u16 sb_idx; + u16 vf_qid; }; struct qed_tunn_params { -- cgit v0.10.2 From f2edc4e1b0786a285d4eb312cc052b029faa813d Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Thu, 2 Jun 2016 15:00:57 +0530 Subject: net: fjes: fjes_main: Remove create_workqueue alloc_workqueue replaces deprecated create_workqueue(). The workqueue adapter->txrx_wq has workitem &adapter->raise_intr_rxdata_task per adapter. Extended Socket Network Device is shared memory based, so someone's transmission denotes other's reception. raise_intr_rxdata_task raises interruption of receivers from the sender in order to notify receivers. The workqueue adapter->control_wq has workitem &adapter->interrupt_watch_task per adapter. interrupt_watch_task is used to prevent delay of interrupts. Dedicated workqueues have been used in both cases since the workitems on the workqueues are involved in normal device operation and require forward progress under memory pressure. max_active has been set to 0 since there is no need for throttling the number of active work items. Since network devices may be used for memory reclaim, WQ_MEM_RECLAIM has been set to guarantee forward progress. Signed-off-by: Bhaktipriya Shridhar Signed-off-by: David S. Miller diff --git a/drivers/net/fjes/fjes_main.c b/drivers/net/fjes/fjes_main.c index 86c331b..9006877 100644 --- a/drivers/net/fjes/fjes_main.c +++ b/drivers/net/fjes/fjes_main.c @@ -1187,8 +1187,9 @@ static int fjes_probe(struct platform_device *plat_dev) adapter->force_reset = false; adapter->open_guard = false; - adapter->txrx_wq = create_workqueue(DRV_NAME "/txrx"); - adapter->control_wq = create_workqueue(DRV_NAME "/control"); + adapter->txrx_wq = alloc_workqueue(DRV_NAME "/txrx", WQ_MEM_RECLAIM, 0); + adapter->control_wq = alloc_workqueue(DRV_NAME "/control", + WQ_MEM_RECLAIM, 0); INIT_WORK(&adapter->tx_stall_task, fjes_tx_stall_task); INIT_WORK(&adapter->raise_intr_rxdata_task, -- cgit v0.10.2 From f6c382fc553b49bc2dfaf705e34fde32171b5db4 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 2 Jun 2016 15:05:38 -0300 Subject: loopback: make use of NETIF_F_GSO_SOFTWARE NETIF_F_GSO_SOFTWARE was defined to list all GSO software types, so lets make use of it in loopback code. Note that veth/vxlan/others already uses it. Within this patch series, this patch causes lo to pick up SCTP GSO feature automatically (as it's added to NETIF_F_GSO_SOFTWARE) and thus avoiding segmentation if possible. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long Signed-off-by: David S. Miller diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index a400288..6255973 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -169,10 +169,9 @@ static void loopback_setup(struct net_device *dev) dev->flags = IFF_LOOPBACK; dev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_NO_QUEUE; netif_keep_dst(dev); - dev->hw_features = NETIF_F_ALL_TSO | NETIF_F_UFO; + dev->hw_features = NETIF_F_GSO_SOFTWARE; dev->features = NETIF_F_SG | NETIF_F_FRAGLIST - | NETIF_F_ALL_TSO - | NETIF_F_UFO + | NETIF_F_GSO_SOFTWARE | NETIF_F_HW_CSUM | NETIF_F_RXCSUM | NETIF_F_SCTP_CRC -- cgit v0.10.2 From 57c05650394b384605f5183747991d19ee543759 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 2 Jun 2016 15:05:39 -0300 Subject: skbuff: export skb_gro_receive sctp GSO requires it and sctp can be compiled as a module, so we need to export this function. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long Signed-off-by: David S. Miller diff --git a/net/core/skbuff.c b/net/core/skbuff.c index f2b77e5..4724bcf 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -3438,6 +3438,7 @@ done: NAPI_GRO_CB(skb)->same_flow = 1; return 0; } +EXPORT_SYMBOL_GPL(skb_gro_receive); void __init skb_init(void) { -- cgit v0.10.2 From 3953c46c3ac7eef31a9935427371c6f54a22f1ba Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 2 Jun 2016 15:05:40 -0300 Subject: sk_buff: allow segmenting based on frag sizes This patch allows segmenting a skb based on its frags sizes instead of based on a fixed value. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long Signed-off-by: David S. Miller diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index ee38a41..329a0a9 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -301,6 +301,11 @@ struct sk_buff; #endif extern int sysctl_max_skb_frags; +/* Set skb_shinfo(skb)->gso_size to this in case you want skb_segment to + * segment using its current segmentation instead. + */ +#define GSO_BY_FRAGS 0xFFFF + typedef struct skb_frag_struct skb_frag_t; struct skb_frag_struct { diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 4724bcf..97c32c7 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -3116,9 +3116,13 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb, int hsize; int size; - len = head_skb->len - offset; - if (len > mss) - len = mss; + if (unlikely(mss == GSO_BY_FRAGS)) { + len = list_skb->len; + } else { + len = head_skb->len - offset; + if (len > mss) + len = mss; + } hsize = skb_headlen(head_skb) - offset; if (hsize < 0) -- cgit v0.10.2 From ae7ef81ef000adeee7a87585b9135ff8a8064acc Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 2 Jun 2016 15:05:41 -0300 Subject: skbuff: introduce skb_gso_validate_mtu skb_gso_network_seglen is not enough for checking fragment sizes if skb is using GSO_BY_FRAGS as we have to check frag per frag. This patch introduces skb_gso_validate_mtu, based on the former, which will wrap the use case inside it as all calls to skb_gso_network_seglen were to validate if it fits on a given TMU, and improve the check. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long Signed-off-by: David S. Miller diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 329a0a9..aa3f9d7 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -2992,6 +2992,7 @@ void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len); int skb_shift(struct sk_buff *tgt, struct sk_buff *skb, int shiftlen); void skb_scrub_packet(struct sk_buff *skb, bool xnet); unsigned int skb_gso_transport_seglen(const struct sk_buff *skb); +bool skb_gso_validate_mtu(const struct sk_buff *skb, unsigned int mtu); struct sk_buff *skb_segment(struct sk_buff *skb, netdev_features_t features); struct sk_buff *skb_vlan_untag(struct sk_buff *skb); int skb_ensure_writable(struct sk_buff *skb, int write_len); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 97c32c7..5ca562b 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -4392,6 +4392,37 @@ unsigned int skb_gso_transport_seglen(const struct sk_buff *skb) } EXPORT_SYMBOL_GPL(skb_gso_transport_seglen); +/** + * skb_gso_validate_mtu - Return in case such skb fits a given MTU + * + * @skb: GSO skb + * + * skb_gso_validate_mtu validates if a given skb will fit a wanted MTU + * once split. + */ +bool skb_gso_validate_mtu(const struct sk_buff *skb, unsigned int mtu) +{ + const struct skb_shared_info *shinfo = skb_shinfo(skb); + const struct sk_buff *iter; + unsigned int hlen; + + hlen = skb_gso_network_seglen(skb); + + if (shinfo->gso_size != GSO_BY_FRAGS) + return hlen <= mtu; + + /* Undo this so we can re-use header sizes */ + hlen -= GSO_BY_FRAGS; + + skb_walk_frags(skb, iter) { + if (hlen + skb_headlen(iter) > mtu) + return false; + } + + return true; +} +EXPORT_SYMBOL_GPL(skb_gso_validate_mtu); + static struct sk_buff *skb_reorder_vlan_header(struct sk_buff *skb) { if (skb_cow(skb, skb_headroom(skb)) < 0) { diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index cbfb180..9f0a7b9 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -54,7 +54,7 @@ static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu) if (skb->ignore_df) return false; - if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) + if (skb_is_gso(skb) && skb_gso_validate_mtu(skb, mtu)) return false; return true; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 124bf0a..cbac493 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -225,7 +225,7 @@ static int ip_finish_output_gso(struct net *net, struct sock *sk, /* common case: locally created skb or seglen is <= mtu */ if (((IPCB(skb)->flags & IPSKB_FORWARDED) == 0) || - skb_gso_network_seglen(skb) <= mtu) + skb_gso_validate_mtu(skb, mtu)) return ip_finish_output2(net, sk, skb); /* Slowpath - GSO segment length is exceeding the dst MTU. diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index cbf127a..6b2f60a 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -368,7 +368,7 @@ static bool ip6_pkt_too_big(const struct sk_buff *skb, unsigned int mtu) if (skb->ignore_df) return false; - if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) + if (skb_is_gso(skb) && skb_gso_validate_mtu(skb, mtu)) return false; return true; diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 0b80a71..7a4aa34 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -91,7 +91,7 @@ bool mpls_pkt_too_big(const struct sk_buff *skb, unsigned int mtu) if (skb->len <= mtu) return false; - if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) + if (skb_is_gso(skb) && skb_gso_validate_mtu(skb, mtu)) return false; return true; -- cgit v0.10.2 From 3acb50c18d8d6650f10919464ade4dcdaf41d62f Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 2 Jun 2016 15:05:42 -0300 Subject: sctp: delay as much as possible skb_linearize This patch is a preparation for the GSO one. In order to successfully handle GSO packets on rx path we must not call skb_linearize, otherwise it defeats any gain GSO may have had. This patch thus delays as much as possible the call to skb_linearize, leaving it to sctp_inq_pop() moment. For that the sanity checks performed now know how to deal with fragments. One positive side-effect of this is that if the socket is backlogged it will have the chance of doing it on backlog processing instead of during softirq. With this move, it's evident that a check for non-linearity in sctp_inq_pop was ineffective and is now removed. Note that a similar check is performed a bit below this one. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long Signed-off-by: David S. Miller diff --git a/net/sctp/input.c b/net/sctp/input.c index a701527..5cff254 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -112,7 +112,6 @@ int sctp_rcv(struct sk_buff *skb) struct sctp_ep_common *rcvr; struct sctp_transport *transport = NULL; struct sctp_chunk *chunk; - struct sctphdr *sh; union sctp_addr src; union sctp_addr dest; int family; @@ -124,15 +123,18 @@ int sctp_rcv(struct sk_buff *skb) __SCTP_INC_STATS(net, SCTP_MIB_INSCTPPACKS); - if (skb_linearize(skb)) + /* If packet is too small to contain a single chunk, let's not + * waste time on it anymore. + */ + if (skb->len < sizeof(struct sctphdr) + sizeof(struct sctp_chunkhdr) + + skb_transport_offset(skb)) goto discard_it; - sh = sctp_hdr(skb); + if (!pskb_may_pull(skb, sizeof(struct sctphdr))) + goto discard_it; - /* Pull up the IP and SCTP headers. */ + /* Pull up the IP header. */ __skb_pull(skb, skb_transport_offset(skb)); - if (skb->len < sizeof(struct sctphdr)) - goto discard_it; skb->csum_valid = 0; /* Previous value not applicable */ if (skb_csum_unnecessary(skb)) @@ -141,11 +143,7 @@ int sctp_rcv(struct sk_buff *skb) goto discard_it; skb->csum_valid = 1; - skb_pull(skb, sizeof(struct sctphdr)); - - /* Make sure we at least have chunk headers worth of data left. */ - if (skb->len < sizeof(struct sctp_chunkhdr)) - goto discard_it; + __skb_pull(skb, sizeof(struct sctphdr)); family = ipver2af(ip_hdr(skb)->version); af = sctp_get_af_specific(family); @@ -230,7 +228,7 @@ int sctp_rcv(struct sk_buff *skb) chunk->rcvr = rcvr; /* Remember the SCTP header. */ - chunk->sctp_hdr = sh; + chunk->sctp_hdr = sctp_hdr(skb); /* Set the source and destination addresses of the incoming chunk. */ sctp_init_addrs(chunk, &src, &dest); @@ -660,19 +658,23 @@ out_unlock: */ static int sctp_rcv_ootb(struct sk_buff *skb) { - sctp_chunkhdr_t *ch; - __u8 *ch_end; - - ch = (sctp_chunkhdr_t *) skb->data; + sctp_chunkhdr_t *ch, _ch; + int ch_end, offset = 0; /* Scan through all the chunks in the packet. */ do { + /* Make sure we have at least the header there */ + if (offset + sizeof(sctp_chunkhdr_t) > skb->len) + break; + + ch = skb_header_pointer(skb, offset, sizeof(*ch), &_ch); + /* Break out if chunk length is less then minimal. */ if (ntohs(ch->length) < sizeof(sctp_chunkhdr_t)) break; - ch_end = ((__u8 *)ch) + WORD_ROUND(ntohs(ch->length)); - if (ch_end > skb_tail_pointer(skb)) + ch_end = offset + WORD_ROUND(ntohs(ch->length)); + if (ch_end > skb->len) break; /* RFC 8.4, 2) If the OOTB packet contains an ABORT chunk, the @@ -697,8 +699,8 @@ static int sctp_rcv_ootb(struct sk_buff *skb) if (SCTP_CID_INIT == ch->type && (void *)ch != skb->data) goto discard; - ch = (sctp_chunkhdr_t *) ch_end; - } while (ch_end < skb_tail_pointer(skb)); + offset = ch_end; + } while (ch_end < skb->len); return 0; @@ -1173,6 +1175,9 @@ static struct sctp_association *__sctp_rcv_lookup_harder(struct net *net, { sctp_chunkhdr_t *ch; + if (skb_linearize(skb)) + return NULL; + ch = (sctp_chunkhdr_t *) skb->data; /* The code below will attempt to walk the chunk and extract diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index 9d87bba..5ba08ce 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -130,7 +130,8 @@ struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) * at this time. */ - if ((chunk = queue->in_progress)) { + chunk = queue->in_progress; + if (chunk) { /* There is a packet that we have been working on. * Any post processing work to do before we move on? */ @@ -152,15 +153,29 @@ struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) if (!chunk) { struct list_head *entry; +next_chunk: /* Is the queue empty? */ if (list_empty(&queue->in_chunk_list)) return NULL; entry = queue->in_chunk_list.next; - chunk = queue->in_progress = - list_entry(entry, struct sctp_chunk, list); + chunk = list_entry(entry, struct sctp_chunk, list); list_del_init(entry); + /* Linearize if it's not GSO */ + if (skb_is_nonlinear(chunk->skb)) { + if (skb_linearize(chunk->skb)) { + __SCTP_INC_STATS(dev_net(chunk->skb->dev), SCTP_MIB_IN_PKT_DISCARDS); + sctp_chunk_free(chunk); + goto next_chunk; + } + + /* Update sctp_hdr as it probably changed */ + chunk->sctp_hdr = sctp_hdr(chunk->skb); + } + + queue->in_progress = chunk; + /* This is the first chunk in the packet. */ chunk->singleton = 1; ch = (sctp_chunkhdr_t *) chunk->skb->data; @@ -172,14 +187,6 @@ struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) chunk->chunk_hdr = ch; chunk->chunk_end = ((__u8 *)ch) + WORD_ROUND(ntohs(ch->length)); - /* In the unlikely case of an IP reassembly, the skb could be - * non-linear. If so, update chunk_end so that it doesn't go past - * the skb->tail. - */ - if (unlikely(skb_is_nonlinear(chunk->skb))) { - if (chunk->chunk_end > skb_tail_pointer(chunk->skb)) - chunk->chunk_end = skb_tail_pointer(chunk->skb); - } skb_pull(chunk->skb, sizeof(sctp_chunkhdr_t)); chunk->subh.v = NULL; /* Subheader is no longer valid. */ -- cgit v0.10.2 From 90017accff61ae89283ad9a51f9ac46ca01633fb Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 2 Jun 2016 15:05:43 -0300 Subject: sctp: Add GSO support SCTP has this pecualiarity that its packets cannot be just segmented to (P)MTU. Its chunks must be contained in IP segments, padding respected. So we can't just generate a big skb, set gso_size to the fragmentation point and deliver it to IP layer. This patch takes a different approach. SCTP will now build a skb as it would be if it was received using GRO. That is, there will be a cover skb with protocol headers and children ones containing the actual segments, already segmented to a way that respects SCTP RFCs. With that, we can tell skb_segment() to just split based on frag_list, trusting its sizes are already in accordance. This way SCTP can benefit from GSO and instead of passing several packets through the stack, it can pass a single large packet. v2: - Added support for receiving GSO frames, as requested by Dave Miller. - Clear skb->cb if packet is GSO (otherwise it's not used by SCTP) - Added heuristics similar to what we have in TCP for not generating single GSO packets that fills cwnd. v3: - consider sctphdr size in skb_gso_transport_seglen() - rebased due to 5c7cdf339af5 ("gso: Remove arbitrary checks for unsupported GSO") Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long Signed-off-by: David S. Miller diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h index aa7b240..9c6c8ef 100644 --- a/include/linux/netdev_features.h +++ b/include/linux/netdev_features.h @@ -53,8 +53,9 @@ enum { * headers in software. */ NETIF_F_GSO_TUNNEL_REMCSUM_BIT, /* ... TUNNEL with TSO & REMCSUM */ + NETIF_F_GSO_SCTP_BIT, /* ... SCTP fragmentation */ /**/NETIF_F_GSO_LAST = /* last bit, see GSO_MASK */ - NETIF_F_GSO_TUNNEL_REMCSUM_BIT, + NETIF_F_GSO_SCTP_BIT, NETIF_F_FCOE_CRC_BIT, /* FCoE CRC32 */ NETIF_F_SCTP_CRC_BIT, /* SCTP checksum offload */ @@ -128,6 +129,7 @@ enum { #define NETIF_F_TSO_MANGLEID __NETIF_F(TSO_MANGLEID) #define NETIF_F_GSO_PARTIAL __NETIF_F(GSO_PARTIAL) #define NETIF_F_GSO_TUNNEL_REMCSUM __NETIF_F(GSO_TUNNEL_REMCSUM) +#define NETIF_F_GSO_SCTP __NETIF_F(GSO_SCTP) #define NETIF_F_HW_VLAN_STAG_FILTER __NETIF_F(HW_VLAN_STAG_FILTER) #define NETIF_F_HW_VLAN_STAG_RX __NETIF_F(HW_VLAN_STAG_RX) #define NETIF_F_HW_VLAN_STAG_TX __NETIF_F(HW_VLAN_STAG_TX) @@ -166,7 +168,8 @@ enum { NETIF_F_FSO) /* List of features with software fallbacks. */ -#define NETIF_F_GSO_SOFTWARE (NETIF_F_ALL_TSO | NETIF_F_UFO) +#define NETIF_F_GSO_SOFTWARE (NETIF_F_ALL_TSO | NETIF_F_UFO | \ + NETIF_F_GSO_SCTP) /* * If one device supports one of these features, then enable them diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index f45929c..fa6df26 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -4012,6 +4012,7 @@ static inline bool net_gso_ok(netdev_features_t features, int gso_type) BUILD_BUG_ON(SKB_GSO_UDP_TUNNEL_CSUM != (NETIF_F_GSO_UDP_TUNNEL_CSUM >> NETIF_F_GSO_SHIFT)); BUILD_BUG_ON(SKB_GSO_PARTIAL != (NETIF_F_GSO_PARTIAL >> NETIF_F_GSO_SHIFT)); BUILD_BUG_ON(SKB_GSO_TUNNEL_REMCSUM != (NETIF_F_GSO_TUNNEL_REMCSUM >> NETIF_F_GSO_SHIFT)); + BUILD_BUG_ON(SKB_GSO_SCTP != (NETIF_F_GSO_SCTP >> NETIF_F_GSO_SHIFT)); return (features & feature) == feature; } diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index aa3f9d7..dc0fca7 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -487,6 +487,8 @@ enum { SKB_GSO_PARTIAL = 1 << 13, SKB_GSO_TUNNEL_REMCSUM = 1 << 14, + + SKB_GSO_SCTP = 1 << 15, }; #if BITS_PER_LONG > 32 diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index b392ac8..632e205 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -186,6 +186,10 @@ void sctp_assocs_proc_exit(struct net *net); int sctp_remaddr_proc_init(struct net *net); void sctp_remaddr_proc_exit(struct net *net); +/* + * sctp/offload.c + */ +int sctp_offload_init(void); /* * Module global variables diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 16b013a..83c5ec5 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -566,6 +566,9 @@ struct sctp_chunk { /* This points to the sk_buff containing the actual data. */ struct sk_buff *skb; + /* In case of GSO packets, this will store the head one */ + struct sk_buff *head_skb; + /* These are the SCTP headers by reverse order in a packet. * Note that some of these may happen more than once. In that * case, we point at the "current" one, whatever that means @@ -696,6 +699,8 @@ struct sctp_packet { size_t overhead; /* This is the total size of all chunks INCLUDING padding. */ size_t size; + /* This is the maximum size this packet may have */ + size_t max_size; /* The packet is destined for this transport address. * The function we finally use to pass down to the next lower diff --git a/net/core/ethtool.c b/net/core/ethtool.c index f403481..9774898 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -89,6 +89,7 @@ static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] [NETIF_F_GSO_UDP_TUNNEL_BIT] = "tx-udp_tnl-segmentation", [NETIF_F_GSO_UDP_TUNNEL_CSUM_BIT] = "tx-udp_tnl-csum-segmentation", [NETIF_F_GSO_PARTIAL_BIT] = "tx-gso-partial", + [NETIF_F_GSO_SCTP_BIT] = "tx-sctp-segmentation", [NETIF_F_FCOE_CRC_BIT] = "tx-checksum-fcoe-crc", [NETIF_F_SCTP_CRC_BIT] = "tx-checksum-sctp", diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 5ca562b..b6e0f95 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -49,6 +49,7 @@ #include #include #include +#include #include #ifdef CONFIG_NET_CLS_ACT #include @@ -4383,6 +4384,8 @@ unsigned int skb_gso_transport_seglen(const struct sk_buff *skb) thlen += inner_tcp_hdrlen(skb); } else if (likely(shinfo->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))) { thlen = tcp_hdrlen(skb); + } else if (unlikely(shinfo->gso_type & SKB_GSO_SCTP)) { + thlen = sizeof(struct sctphdr); } /* UFO sets gso_size to the size of the fragmentation * payload, i.e. the size of the L4 (UDP) header is already diff --git a/net/sctp/Makefile b/net/sctp/Makefile index 0fca582..6c4f749 100644 --- a/net/sctp/Makefile +++ b/net/sctp/Makefile @@ -11,7 +11,8 @@ sctp-y := sm_statetable.o sm_statefuns.o sm_sideeffect.o \ transport.o chunk.o sm_make_chunk.o ulpevent.o \ inqueue.o outqueue.o ulpqueue.o \ tsnmap.o bind_addr.o socket.o primitive.o \ - output.o input.o debug.o ssnmap.o auth.o + output.o input.o debug.o ssnmap.o auth.o \ + offload.o sctp_probe-y := probe.o diff --git a/net/sctp/input.c b/net/sctp/input.c index 5cff254..6f8e676 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -139,7 +139,9 @@ int sctp_rcv(struct sk_buff *skb) skb->csum_valid = 0; /* Previous value not applicable */ if (skb_csum_unnecessary(skb)) __skb_decr_checksum_unnecessary(skb); - else if (!sctp_checksum_disable && sctp_rcv_checksum(net, skb) < 0) + else if (!sctp_checksum_disable && + !(skb_shinfo(skb)->gso_type & SKB_GSO_SCTP) && + sctp_rcv_checksum(net, skb) < 0) goto discard_it; skb->csum_valid = 1; @@ -1175,6 +1177,14 @@ static struct sctp_association *__sctp_rcv_lookup_harder(struct net *net, { sctp_chunkhdr_t *ch; + /* We do not allow GSO frames here as we need to linearize and + * then cannot guarantee frame boundaries. This shouldn't be an + * issue as packets hitting this are mostly INIT or INIT-ACK and + * those cannot be on GSO-style anyway. + */ + if ((skb_shinfo(skb)->gso_type & SKB_GSO_SCTP) == SKB_GSO_SCTP) + return NULL; + if (skb_linearize(skb)) return NULL; diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index 5ba08ce..edabbbd 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -138,6 +138,17 @@ struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) if (chunk->singleton || chunk->end_of_packet || chunk->pdiscard) { + if (chunk->head_skb == chunk->skb) { + chunk->skb = skb_shinfo(chunk->skb)->frag_list; + goto new_skb; + } + if (chunk->skb->next) { + chunk->skb = chunk->skb->next; + goto new_skb; + } + + if (chunk->head_skb) + chunk->skb = chunk->head_skb; sctp_chunk_free(chunk); chunk = queue->in_progress = NULL; } else { @@ -155,15 +166,15 @@ struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) next_chunk: /* Is the queue empty? */ - if (list_empty(&queue->in_chunk_list)) + entry = sctp_list_dequeue(&queue->in_chunk_list); + if (!entry) return NULL; - entry = queue->in_chunk_list.next; chunk = list_entry(entry, struct sctp_chunk, list); - list_del_init(entry); /* Linearize if it's not GSO */ - if (skb_is_nonlinear(chunk->skb)) { + if ((skb_shinfo(chunk->skb)->gso_type & SKB_GSO_SCTP) != SKB_GSO_SCTP && + skb_is_nonlinear(chunk->skb)) { if (skb_linearize(chunk->skb)) { __SCTP_INC_STATS(dev_net(chunk->skb->dev), SCTP_MIB_IN_PKT_DISCARDS); sctp_chunk_free(chunk); @@ -174,15 +185,39 @@ next_chunk: chunk->sctp_hdr = sctp_hdr(chunk->skb); } + if ((skb_shinfo(chunk->skb)->gso_type & SKB_GSO_SCTP) == SKB_GSO_SCTP) { + /* GSO-marked skbs but without frags, handle + * them normally + */ + if (skb_shinfo(chunk->skb)->frag_list) + chunk->head_skb = chunk->skb; + + /* skbs with "cover letter" */ + if (chunk->head_skb && chunk->skb->data_len == chunk->skb->len) + chunk->skb = skb_shinfo(chunk->skb)->frag_list; + + if (WARN_ON(!chunk->skb)) { + __SCTP_INC_STATS(dev_net(chunk->skb->dev), SCTP_MIB_IN_PKT_DISCARDS); + sctp_chunk_free(chunk); + goto next_chunk; + } + } + + if (chunk->asoc) + sock_rps_save_rxhash(chunk->asoc->base.sk, chunk->skb); + queue->in_progress = chunk; +new_skb: /* This is the first chunk in the packet. */ - chunk->singleton = 1; ch = (sctp_chunkhdr_t *) chunk->skb->data; + chunk->singleton = 1; chunk->data_accepted = 0; - - if (chunk->asoc) - sock_rps_save_rxhash(chunk->asoc->base.sk, chunk->skb); + chunk->pdiscard = 0; + chunk->auth = 0; + chunk->has_asconf = 0; + chunk->end_of_packet = 0; + chunk->ecn_ce_done = 0; } chunk->chunk_hdr = ch; diff --git a/net/sctp/offload.c b/net/sctp/offload.c new file mode 100644 index 0000000..a37887b --- /dev/null +++ b/net/sctp/offload.c @@ -0,0 +1,98 @@ +/* + * sctp_offload - GRO/GSO Offloading for SCTP + * + * Copyright (C) 2015, Marcelo Ricardo Leitner + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +static __le32 sctp_gso_make_checksum(struct sk_buff *skb) +{ + skb->ip_summed = CHECKSUM_NONE; + return sctp_compute_cksum(skb, skb_transport_offset(skb)); +} + +static struct sk_buff *sctp_gso_segment(struct sk_buff *skb, + netdev_features_t features) +{ + struct sk_buff *segs = ERR_PTR(-EINVAL); + struct sctphdr *sh; + + sh = sctp_hdr(skb); + if (!pskb_may_pull(skb, sizeof(*sh))) + goto out; + + __skb_pull(skb, sizeof(*sh)); + + if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) { + /* Packet is from an untrusted source, reset gso_segs. */ + struct skb_shared_info *pinfo = skb_shinfo(skb); + struct sk_buff *frag_iter; + + pinfo->gso_segs = 0; + if (skb->len != skb->data_len) { + /* Means we have chunks in here too */ + pinfo->gso_segs++; + } + + skb_walk_frags(skb, frag_iter) + pinfo->gso_segs++; + + segs = NULL; + goto out; + } + + segs = skb_segment(skb, features | NETIF_F_HW_CSUM); + if (IS_ERR(segs)) + goto out; + + /* All that is left is update SCTP CRC if necessary */ + if (!(features & NETIF_F_SCTP_CRC)) { + for (skb = segs; skb; skb = skb->next) { + if (skb->ip_summed == CHECKSUM_PARTIAL) { + sh = sctp_hdr(skb); + sh->checksum = sctp_gso_make_checksum(skb); + } + } + } + +out: + return segs; +} + +static const struct net_offload sctp_offload = { + .callbacks = { + .gso_segment = sctp_gso_segment, + }, +}; + +int __init sctp_offload_init(void) +{ + return inet_add_offload(&sctp_offload, IPPROTO_SCTP); +} diff --git a/net/sctp/output.c b/net/sctp/output.c index 9844fe5..60499a6 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -84,18 +84,42 @@ static void sctp_packet_reset(struct sctp_packet *packet) struct sctp_packet *sctp_packet_config(struct sctp_packet *packet, __u32 vtag, int ecn_capable) { - struct sctp_chunk *chunk = NULL; + struct sctp_transport *tp = packet->transport; + struct sctp_association *asoc = tp->asoc; pr_debug("%s: packet:%p vtag:0x%x\n", __func__, packet, vtag); packet->vtag = vtag; + if (asoc && tp->dst) { + struct sock *sk = asoc->base.sk; + + rcu_read_lock(); + if (__sk_dst_get(sk) != tp->dst) { + dst_hold(tp->dst); + sk_setup_caps(sk, tp->dst); + } + + if (sk_can_gso(sk)) { + struct net_device *dev = tp->dst->dev; + + packet->max_size = dev->gso_max_size; + } else { + packet->max_size = asoc->pathmtu; + } + rcu_read_unlock(); + + } else { + packet->max_size = tp->pathmtu; + } + if (ecn_capable && sctp_packet_empty(packet)) { - chunk = sctp_get_ecne_prepend(packet->transport->asoc); + struct sctp_chunk *chunk; /* If there a is a prepend chunk stick it on the list before * any other chunks get appended. */ + chunk = sctp_get_ecne_prepend(asoc); if (chunk) sctp_packet_append_chunk(packet, chunk); } @@ -381,12 +405,15 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) struct sctp_transport *tp = packet->transport; struct sctp_association *asoc = tp->asoc; struct sctphdr *sh; - struct sk_buff *nskb; + struct sk_buff *nskb = NULL, *head = NULL; struct sctp_chunk *chunk, *tmp; struct sock *sk; int err = 0; int padding; /* How much padding do we need? */ + int pkt_size; __u8 has_data = 0; + int gso = 0; + int pktcount = 0; struct dst_entry *dst; unsigned char *auth = NULL; /* pointer to auth in skb data */ @@ -400,18 +427,37 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) chunk = list_entry(packet->chunk_list.next, struct sctp_chunk, list); sk = chunk->skb->sk; - /* Allocate the new skb. */ - nskb = alloc_skb(packet->size + MAX_HEADER, gfp); - if (!nskb) + /* Allocate the head skb, or main one if not in GSO */ + if (packet->size > tp->pathmtu && !packet->ipfragok) { + if (sk_can_gso(sk)) { + gso = 1; + pkt_size = packet->overhead; + } else { + /* If this happens, we trash this packet and try + * to build a new one, hopefully correct this + * time. Application may notice this error. + */ + pr_err_once("Trying to GSO but underlying device doesn't support it."); + goto nomem; + } + } else { + pkt_size = packet->size; + } + head = alloc_skb(pkt_size + MAX_HEADER, gfp); + if (!head) goto nomem; + if (gso) { + NAPI_GRO_CB(head)->last = head; + skb_shinfo(head)->gso_type = sk->sk_gso_type; + } /* Make sure the outbound skb has enough header room reserved. */ - skb_reserve(nskb, packet->overhead + MAX_HEADER); + skb_reserve(head, packet->overhead + MAX_HEADER); /* Set the owning socket so that we know where to get the * destination IP address. */ - sctp_packet_set_owner_w(nskb, sk); + sctp_packet_set_owner_w(head, sk); if (!sctp_transport_dst_check(tp)) { sctp_transport_route(tp, NULL, sctp_sk(sk)); @@ -422,11 +468,11 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) dst = dst_clone(tp->dst); if (!dst) goto no_route; - skb_dst_set(nskb, dst); + skb_dst_set(head, dst); /* Build the SCTP header. */ - sh = (struct sctphdr *)skb_push(nskb, sizeof(struct sctphdr)); - skb_reset_transport_header(nskb); + sh = (struct sctphdr *)skb_push(head, sizeof(struct sctphdr)); + skb_reset_transport_header(head); sh->source = htons(packet->source_port); sh->dest = htons(packet->destination_port); @@ -441,90 +487,133 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) sh->vtag = htonl(packet->vtag); sh->checksum = 0; - /** - * 6.10 Bundling - * - * An endpoint bundles chunks by simply including multiple - * chunks in one outbound SCTP packet. ... - */ - - /** - * 3.2 Chunk Field Descriptions - * - * The total length of a chunk (including Type, Length and - * Value fields) MUST be a multiple of 4 bytes. If the length - * of the chunk is not a multiple of 4 bytes, the sender MUST - * pad the chunk with all zero bytes and this padding is not - * included in the chunk length field. The sender should - * never pad with more than 3 bytes. - * - * [This whole comment explains WORD_ROUND() below.] - */ - pr_debug("***sctp_transmit_packet***\n"); - list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) { - list_del_init(&chunk->list); - if (sctp_chunk_is_data(chunk)) { - /* 6.3.1 C4) When data is in flight and when allowed - * by rule C5, a new RTT measurement MUST be made each - * round trip. Furthermore, new RTT measurements - * SHOULD be made no more than once per round-trip - * for a given destination transport address. - */ + do { + /* Set up convenience variables... */ + chunk = list_entry(packet->chunk_list.next, struct sctp_chunk, list); + pktcount++; - if (!chunk->resent && !tp->rto_pending) { - chunk->rtt_in_progress = 1; - tp->rto_pending = 1; + /* Calculate packet size, so it fits in PMTU. Leave + * other chunks for the next packets. + */ + if (gso) { + pkt_size = packet->overhead; + list_for_each_entry(chunk, &packet->chunk_list, list) { + int padded = WORD_ROUND(chunk->skb->len); + + if (pkt_size + padded > tp->pathmtu) + break; + pkt_size += padded; } - has_data = 1; - } + /* Allocate a new skb. */ + nskb = alloc_skb(pkt_size + MAX_HEADER, gfp); + if (!nskb) + goto nomem; - padding = WORD_ROUND(chunk->skb->len) - chunk->skb->len; - if (padding) - memset(skb_put(chunk->skb, padding), 0, padding); + /* Make sure the outbound skb has enough header + * room reserved. + */ + skb_reserve(nskb, packet->overhead + MAX_HEADER); + } else { + nskb = head; + } - /* if this is the auth chunk that we are adding, - * store pointer where it will be added and put - * the auth into the packet. + /** + * 3.2 Chunk Field Descriptions + * + * The total length of a chunk (including Type, Length and + * Value fields) MUST be a multiple of 4 bytes. If the length + * of the chunk is not a multiple of 4 bytes, the sender MUST + * pad the chunk with all zero bytes and this padding is not + * included in the chunk length field. The sender should + * never pad with more than 3 bytes. + * + * [This whole comment explains WORD_ROUND() below.] */ - if (chunk == packet->auth) - auth = skb_tail_pointer(nskb); - memcpy(skb_put(nskb, chunk->skb->len), + pkt_size -= packet->overhead; + list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) { + list_del_init(&chunk->list); + if (sctp_chunk_is_data(chunk)) { + /* 6.3.1 C4) When data is in flight and when allowed + * by rule C5, a new RTT measurement MUST be made each + * round trip. Furthermore, new RTT measurements + * SHOULD be made no more than once per round-trip + * for a given destination transport address. + */ + + if (!chunk->resent && !tp->rto_pending) { + chunk->rtt_in_progress = 1; + tp->rto_pending = 1; + } + + has_data = 1; + } + + padding = WORD_ROUND(chunk->skb->len) - chunk->skb->len; + if (padding) + memset(skb_put(chunk->skb, padding), 0, padding); + + /* if this is the auth chunk that we are adding, + * store pointer where it will be added and put + * the auth into the packet. + */ + if (chunk == packet->auth) + auth = skb_tail_pointer(nskb); + + memcpy(skb_put(nskb, chunk->skb->len), chunk->skb->data, chunk->skb->len); - pr_debug("*** Chunk:%p[%s] %s 0x%x, length:%d, chunk->skb->len:%d, " - "rtt_in_progress:%d\n", chunk, - sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)), - chunk->has_tsn ? "TSN" : "No TSN", - chunk->has_tsn ? ntohl(chunk->subh.data_hdr->tsn) : 0, - ntohs(chunk->chunk_hdr->length), chunk->skb->len, - chunk->rtt_in_progress); - - /* - * If this is a control chunk, this is our last - * reference. Free data chunks after they've been - * acknowledged or have failed. - */ - if (!sctp_chunk_is_data(chunk)) - sctp_chunk_free(chunk); - } + pr_debug("*** Chunk:%p[%s] %s 0x%x, length:%d, chunk->skb->len:%d, rtt_in_progress:%d\n", + chunk, + sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)), + chunk->has_tsn ? "TSN" : "No TSN", + chunk->has_tsn ? ntohl(chunk->subh.data_hdr->tsn) : 0, + ntohs(chunk->chunk_hdr->length), chunk->skb->len, + chunk->rtt_in_progress); + + /* If this is a control chunk, this is our last + * reference. Free data chunks after they've been + * acknowledged or have failed. + * Re-queue auth chunks if needed. + */ + pkt_size -= WORD_ROUND(chunk->skb->len); - /* SCTP-AUTH, Section 6.2 - * The sender MUST calculate the MAC as described in RFC2104 [2] - * using the hash function H as described by the MAC Identifier and - * the shared association key K based on the endpoint pair shared key - * described by the shared key identifier. The 'data' used for the - * computation of the AUTH-chunk is given by the AUTH chunk with its - * HMAC field set to zero (as shown in Figure 6) followed by all - * chunks that are placed after the AUTH chunk in the SCTP packet. - */ - if (auth) - sctp_auth_calculate_hmac(asoc, nskb, - (struct sctp_auth_chunk *)auth, - gfp); + if (chunk == packet->auth && !list_empty(&packet->chunk_list)) + list_add(&chunk->list, &packet->chunk_list); + else if (!sctp_chunk_is_data(chunk)) + sctp_chunk_free(chunk); + + if (!pkt_size) + break; + } + + /* SCTP-AUTH, Section 6.2 + * The sender MUST calculate the MAC as described in RFC2104 [2] + * using the hash function H as described by the MAC Identifier and + * the shared association key K based on the endpoint pair shared key + * described by the shared key identifier. The 'data' used for the + * computation of the AUTH-chunk is given by the AUTH chunk with its + * HMAC field set to zero (as shown in Figure 6) followed by all + * chunks that are placed after the AUTH chunk in the SCTP packet. + */ + if (auth) + sctp_auth_calculate_hmac(asoc, nskb, + (struct sctp_auth_chunk *)auth, + gfp); + + if (!gso) + break; + + if (skb_gro_receive(&head, nskb)) + goto nomem; + nskb = NULL; + if (WARN_ON_ONCE(skb_shinfo(head)->gso_segs >= + sk->sk_gso_max_segs)) + goto nomem; + } while (!list_empty(&packet->chunk_list)); /* 2) Calculate the Adler-32 checksum of the whole packet, * including the SCTP common header and all the @@ -532,16 +621,18 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) * * Note: Adler-32 is no longer applicable, as has been replaced * by CRC32-C as described in . + * + * If it's a GSO packet, it's postponed to sctp_skb_segment. */ - if (!sctp_checksum_disable) { - if (!(dst->dev->features & NETIF_F_SCTP_CRC) || - (dst_xfrm(dst) != NULL) || packet->ipfragok) { - sh->checksum = sctp_compute_cksum(nskb, 0); + if (!sctp_checksum_disable || gso) { + if (!gso && (!(dst->dev->features & NETIF_F_SCTP_CRC) || + dst_xfrm(dst) || packet->ipfragok)) { + sh->checksum = sctp_compute_cksum(head, 0); } else { /* no need to seed pseudo checksum for SCTP */ - nskb->ip_summed = CHECKSUM_PARTIAL; - nskb->csum_start = skb_transport_header(nskb) - nskb->head; - nskb->csum_offset = offsetof(struct sctphdr, checksum); + head->ip_summed = CHECKSUM_PARTIAL; + head->csum_start = skb_transport_header(head) - head->head; + head->csum_offset = offsetof(struct sctphdr, checksum); } } @@ -557,7 +648,7 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) * Note: The works for IPv6 layer checks this bit too later * in transmission. See IP6_ECN_flow_xmit(). */ - tp->af_specific->ecn_capable(nskb->sk); + tp->af_specific->ecn_capable(sk); /* Set up the IP options. */ /* BUG: not implemented @@ -566,7 +657,7 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) /* Dump that on IP! */ if (asoc) { - asoc->stats.opackets++; + asoc->stats.opackets += pktcount; if (asoc->peer.last_sent_to != tp) /* Considering the multiple CPU scenario, this is a * "correcter" place for last_sent_to. --xguo @@ -589,16 +680,36 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) } } - pr_debug("***sctp_transmit_packet*** skb->len:%d\n", nskb->len); + pr_debug("***sctp_transmit_packet*** skb->len:%d\n", head->len); + + if (gso) { + /* Cleanup our debris for IP stacks */ + memset(head->cb, 0, max(sizeof(struct inet_skb_parm), + sizeof(struct inet6_skb_parm))); - nskb->ignore_df = packet->ipfragok; - tp->af_specific->sctp_xmit(nskb, tp); + skb_shinfo(head)->gso_segs = pktcount; + skb_shinfo(head)->gso_size = GSO_BY_FRAGS; + + /* We have to refresh this in case we are xmiting to + * more than one transport at a time + */ + rcu_read_lock(); + if (__sk_dst_get(sk) != tp->dst) { + dst_hold(tp->dst); + sk_setup_caps(sk, tp->dst); + } + rcu_read_unlock(); + } + head->ignore_df = packet->ipfragok; + tp->af_specific->sctp_xmit(head, tp); out: sctp_packet_reset(packet); return err; no_route: - kfree_skb(nskb); + kfree_skb(head); + if (nskb != head) + kfree_skb(nskb); if (asoc) IP_INC_STATS(sock_net(asoc->base.sk), IPSTATS_MIB_OUTNOROUTES); @@ -751,39 +862,63 @@ static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet, struct sctp_chunk *chunk, u16 chunk_len) { - size_t psize; - size_t pmtu; - int too_big; + size_t psize, pmtu; sctp_xmit_t retval = SCTP_XMIT_OK; psize = packet->size; - pmtu = ((packet->transport->asoc) ? - (packet->transport->asoc->pathmtu) : - (packet->transport->pathmtu)); - - too_big = (psize + chunk_len > pmtu); + if (packet->transport->asoc) + pmtu = packet->transport->asoc->pathmtu; + else + pmtu = packet->transport->pathmtu; /* Decide if we need to fragment or resubmit later. */ - if (too_big) { - /* It's OK to fragmet at IP level if any one of the following + if (psize + chunk_len > pmtu) { + /* It's OK to fragment at IP level if any one of the following * is true: - * 1. The packet is empty (meaning this chunk is greater - * the MTU) - * 2. The chunk we are adding is a control chunk - * 3. The packet doesn't have any data in it yet and data - * requires authentication. + * 1. The packet is empty (meaning this chunk is greater + * the MTU) + * 2. The packet doesn't have any data in it yet and data + * requires authentication. */ - if (sctp_packet_empty(packet) || !sctp_chunk_is_data(chunk) || + if (sctp_packet_empty(packet) || (!packet->has_data && chunk->auth)) { /* We no longer do re-fragmentation. * Just fragment at the IP layer, if we * actually hit this condition */ packet->ipfragok = 1; - } else { - retval = SCTP_XMIT_PMTU_FULL; + goto out; } + + /* It is also okay to fragment if the chunk we are + * adding is a control chunk, but only if current packet + * is not a GSO one otherwise it causes fragmentation of + * a large frame. So in this case we allow the + * fragmentation by forcing it to be in a new packet. + */ + if (!sctp_chunk_is_data(chunk) && packet->has_data) + retval = SCTP_XMIT_PMTU_FULL; + + if (psize + chunk_len > packet->max_size) + /* Hit GSO/PMTU limit, gotta flush */ + retval = SCTP_XMIT_PMTU_FULL; + + if (!packet->transport->burst_limited && + psize + chunk_len > (packet->transport->cwnd >> 1)) + /* Do not allow a single GSO packet to use more + * than half of cwnd. + */ + retval = SCTP_XMIT_PMTU_FULL; + + if (packet->transport->burst_limited && + psize + chunk_len > (packet->transport->burst_limited >> 1)) + /* Do not allow a single GSO packet to use more + * than half of original cwnd. + */ + retval = SCTP_XMIT_PMTU_FULL; + /* Otherwise it will fit in the GSO packet */ } +out: return retval; } diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index d3d50da..40022ee 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1516,6 +1516,9 @@ static __init int sctp_init(void) if (status) goto err_v6_add_protocol; + if (sctp_offload_init() < 0) + pr_crit("%s: Cannot add SCTP protocol offload\n", __func__); + out: return status; err_v6_add_protocol: diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 67154b8..712fb23 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4003,6 +4003,8 @@ static int sctp_init_sock(struct sock *sk) return -ESOCKTNOSUPPORT; } + sk->sk_gso_type = SKB_GSO_SCTP; + /* Initialize default send parameters. These parameters can be * modified with the SCTP_DEFAULT_SEND_PARAM socket option. */ -- cgit v0.10.2 From 942b3235bf77e5600a05d6e85f0415bdeb8068bb Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 2 Jun 2016 15:05:44 -0300 Subject: sctp: improve debug message to also log curr pkt and new chunk size This is useful for debugging packet sizes. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long Signed-off-by: David S. Miller diff --git a/net/sctp/output.c b/net/sctp/output.c index 60499a6..90d2e12 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -182,7 +182,8 @@ sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet, sctp_xmit_t retval; int error = 0; - pr_debug("%s: packet:%p chunk:%p\n", __func__, packet, chunk); + pr_debug("%s: packet:%p size:%lu chunk:%p size:%d\n", __func__, + packet, packet->size, chunk, chunk->skb ? chunk->skb->len : -1); switch ((retval = (sctp_packet_append_chunk(packet, chunk)))) { case SCTP_XMIT_PMTU_FULL: -- cgit v0.10.2 From cb2911fed61497e4d0383355f1c865fcdaa94061 Mon Sep 17 00:00:00 2001 From: Haiyang Zhang Date: Thu, 2 Jun 2016 12:02:04 -0700 Subject: hv_netvsc: Fix VF register on vlan devices Added a condition to avoid vlan devices with same MAC registering as VF. Signed-off-by: Haiyang Zhang Reviewed-by: K. Y. Srinivasan Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 6a69b5c..5ac1267 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -1500,6 +1500,10 @@ static int netvsc_netdev_event(struct notifier_block *this, { struct net_device *event_dev = netdev_notifier_info_to_dev(ptr); + /* Avoid Vlan dev with same MAC registering as VF */ + if (event_dev->priv_flags & IFF_802_1Q_VLAN) + return NOTIFY_DONE; + switch (event) { case NETDEV_REGISTER: return netvsc_register_vf(event_dev); -- cgit v0.10.2 From 9b6d53985fd130c24ad2260c2edb0df50449f020 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Thu, 2 Jun 2016 12:08:52 -0700 Subject: rxrpc: Use pr_ and pr_fmt, reduce object size a few KB Use the more common kernel logging style and reduce object size. The logging message prefix changes from a mixture of "RxRPC:" and "RXRPC:" to "af_rxrpc: ". $ size net/rxrpc/built-in.o* text data bss dec hex filename 64172 1972 8304 74448 122d0 net/rxrpc/built-in.o.new 67512 1972 8304 77788 12fdc net/rxrpc/built-in.o.old Miscellanea: o Consolidate the ASSERT macros to use a single pr_err call with decimal and hexadecimal output and a stringified #OP argument Signed-off-by: Joe Perches Signed-off-by: David S. Miller diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index e45e94c..7840b8e 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include @@ -796,49 +798,49 @@ static int __init af_rxrpc_init(void) "rxrpc_call_jar", sizeof(struct rxrpc_call), 0, SLAB_HWCACHE_ALIGN, NULL); if (!rxrpc_call_jar) { - printk(KERN_NOTICE "RxRPC: Failed to allocate call jar\n"); + pr_notice("Failed to allocate call jar\n"); goto error_call_jar; } rxrpc_workqueue = alloc_workqueue("krxrpcd", 0, 1); if (!rxrpc_workqueue) { - printk(KERN_NOTICE "RxRPC: Failed to allocate work queue\n"); + pr_notice("Failed to allocate work queue\n"); goto error_work_queue; } ret = rxrpc_init_security(); if (ret < 0) { - printk(KERN_CRIT "RxRPC: Cannot initialise security\n"); + pr_crit("Cannot initialise security\n"); goto error_security; } ret = proto_register(&rxrpc_proto, 1); if (ret < 0) { - printk(KERN_CRIT "RxRPC: Cannot register protocol\n"); + pr_crit("Cannot register protocol\n"); goto error_proto; } ret = sock_register(&rxrpc_family_ops); if (ret < 0) { - printk(KERN_CRIT "RxRPC: Cannot register socket family\n"); + pr_crit("Cannot register socket family\n"); goto error_sock; } ret = register_key_type(&key_type_rxrpc); if (ret < 0) { - printk(KERN_CRIT "RxRPC: Cannot register client key type\n"); + pr_crit("Cannot register client key type\n"); goto error_key_type; } ret = register_key_type(&key_type_rxrpc_s); if (ret < 0) { - printk(KERN_CRIT "RxRPC: Cannot register server key type\n"); + pr_crit("Cannot register server key type\n"); goto error_key_type_s; } ret = rxrpc_sysctl_init(); if (ret < 0) { - printk(KERN_CRIT "RxRPC: Cannot register sysctls\n"); + pr_crit("Cannot register sysctls\n"); goto error_sysctls; } diff --git a/net/rxrpc/ar-accept.c b/net/rxrpc/ar-accept.c index e7a7f05..eea5f4a 100644 --- a/net/rxrpc/ar-accept.c +++ b/net/rxrpc/ar-accept.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include diff --git a/net/rxrpc/ar-ack.c b/net/rxrpc/ar-ack.c index 374478e..1838178 100644 --- a/net/rxrpc/ar-ack.c +++ b/net/rxrpc/ar-ack.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include diff --git a/net/rxrpc/ar-call.c b/net/rxrpc/ar-call.c index 571a41f..1fbaae1 100644 --- a/net/rxrpc/ar-call.c +++ b/net/rxrpc/ar-call.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include @@ -669,8 +671,7 @@ void rxrpc_release_call(struct rxrpc_call *call) conn->channels[3] == NULL); break; default: - printk(KERN_ERR "RxRPC: conn->avail_calls=%d\n", - conn->avail_calls); + pr_err("conn->avail_calls=%d\n", conn->avail_calls); BUG(); } } @@ -935,16 +936,15 @@ void __exit rxrpc_destroy_all_calls(void) if (call->state != RXRPC_CALL_DEAD) break; default: - printk(KERN_ERR "RXRPC:" - " Call %p still in use (%d,%d,%s,%lx,%lx)!\n", + pr_err("Call %p still in use (%d,%d,%s,%lx,%lx)!\n", call, atomic_read(&call->usage), atomic_read(&call->ackr_not_idle), rxrpc_call_states[call->state], call->flags, call->events); if (!skb_queue_empty(&call->rx_queue)) - printk(KERN_ERR"RXRPC: Rx queue occupied\n"); + pr_err("Rx queue occupied\n"); if (!skb_queue_empty(&call->rx_oos_queue)) - printk(KERN_ERR"RXRPC: OOS queue occupied\n"); + pr_err("OOS queue occupied\n"); break; } diff --git a/net/rxrpc/ar-connection.c b/net/rxrpc/ar-connection.c index 97f4fae..d67b1f1 100644 --- a/net/rxrpc/ar-connection.c +++ b/net/rxrpc/ar-connection.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include diff --git a/net/rxrpc/ar-connevent.c b/net/rxrpc/ar-connevent.c index 5f95639..8bdd692 100644 --- a/net/rxrpc/ar-connevent.c +++ b/net/rxrpc/ar-connevent.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include diff --git a/net/rxrpc/ar-input.c b/net/rxrpc/ar-input.c index 6ff9741..d7c2a0b 100644 --- a/net/rxrpc/ar-input.c +++ b/net/rxrpc/ar-input.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index f0b807a..18ab5c5 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -744,21 +744,18 @@ do { \ #define ASSERT(X) \ do { \ if (unlikely(!(X))) { \ - printk(KERN_ERR "\n"); \ - printk(KERN_ERR "RxRPC: Assertion failed\n"); \ + pr_err("Assertion failed\n"); \ BUG(); \ } \ } while (0) #define ASSERTCMP(X, OP, Y) \ do { \ - if (unlikely(!((X) OP (Y)))) { \ - printk(KERN_ERR "\n"); \ - printk(KERN_ERR "RxRPC: Assertion failed\n"); \ - printk(KERN_ERR "%lu " #OP " %lu is false\n", \ - (unsigned long)(X), (unsigned long)(Y)); \ - printk(KERN_ERR "0x%lx " #OP " 0x%lx is false\n", \ - (unsigned long)(X), (unsigned long)(Y)); \ + unsigned long _x = (unsigned long)(X); \ + unsigned long _y = (unsigned long)(Y); \ + if (unlikely(!(_x OP _y))) { \ + pr_err("Assertion failed - %lu(0x%lx) %s %lu(0x%lx) is false\n", \ + _x, _x, #OP, _y, _y); \ BUG(); \ } \ } while (0) @@ -766,21 +763,18 @@ do { \ #define ASSERTIF(C, X) \ do { \ if (unlikely((C) && !(X))) { \ - printk(KERN_ERR "\n"); \ - printk(KERN_ERR "RxRPC: Assertion failed\n"); \ + pr_err("Assertion failed\n"); \ BUG(); \ } \ } while (0) #define ASSERTIFCMP(C, X, OP, Y) \ do { \ - if (unlikely((C) && !((X) OP (Y)))) { \ - printk(KERN_ERR "\n"); \ - printk(KERN_ERR "RxRPC: Assertion failed\n"); \ - printk(KERN_ERR "%lu " #OP " %lu is false\n", \ - (unsigned long)(X), (unsigned long)(Y)); \ - printk(KERN_ERR "0x%lx " #OP " 0x%lx is false\n", \ - (unsigned long)(X), (unsigned long)(Y)); \ + unsigned long _x = (unsigned long)(X); \ + unsigned long _y = (unsigned long)(Y); \ + if (unlikely((C) && !(_x OP _y))) { \ + pr_err("Assertion failed - %lu(0x%lx) %s %lu(0x%lx) is false\n", \ + _x, _x, #OP, _y, _y); \ BUG(); \ } \ } while (0) diff --git a/net/rxrpc/ar-key.c b/net/rxrpc/ar-key.c index 1021b4c..4ad56fa 100644 --- a/net/rxrpc/ar-key.c +++ b/net/rxrpc/ar-key.c @@ -12,6 +12,8 @@ * "afs@CAMBRIDGE.REDHAT.COM> */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include @@ -800,7 +802,7 @@ static void rxrpc_free_token_list(struct rxrpc_key_token *token) rxrpc_rxk5_free(token->k5); break; default: - printk(KERN_ERR "Unknown token type %x on rxrpc key\n", + pr_err("Unknown token type %x on rxrpc key\n", token->security_index); BUG(); } diff --git a/net/rxrpc/ar-local.c b/net/rxrpc/ar-local.c index 4e1e6db..701c42b 100644 --- a/net/rxrpc/ar-local.c +++ b/net/rxrpc/ar-local.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c index 51cb100..ea61953 100644 --- a/net/rxrpc/ar-output.c +++ b/net/rxrpc/ar-output.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include diff --git a/net/rxrpc/ar-peer.c b/net/rxrpc/ar-peer.c index dc089b1..0b54cda 100644 --- a/net/rxrpc/ar-peer.c +++ b/net/rxrpc/ar-peer.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include diff --git a/net/rxrpc/ar-recvmsg.c b/net/rxrpc/ar-recvmsg.c index 160f092..59706b9 100644 --- a/net/rxrpc/ar-recvmsg.c +++ b/net/rxrpc/ar-recvmsg.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include @@ -307,7 +309,7 @@ receive_non_data_message: &abort_code); break; default: - pr_err("RxRPC: Unknown packet mark %u\n", skb->mark); + pr_err("Unknown packet mark %u\n", skb->mark); BUG(); break; } diff --git a/net/rxrpc/ar-skbuff.c b/net/rxrpc/ar-skbuff.c index 62a2674..eee0cfd9 100644 --- a/net/rxrpc/ar-skbuff.c +++ b/net/rxrpc/ar-skbuff.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include diff --git a/net/rxrpc/ar-transport.c b/net/rxrpc/ar-transport.c index 66a1a56..a1b6518 100644 --- a/net/rxrpc/ar-transport.c +++ b/net/rxrpc/ar-transport.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index 6b726a0..d4da538 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -9,6 +9,8 @@ * 2 of the License, or (at your option) any later version. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include -- cgit v0.10.2 From 330348d94223346f855357fab2517f6c903001c7 Mon Sep 17 00:00:00 2001 From: Ivan Khoronzhuk Date: Fri, 3 Jun 2016 01:37:08 +0300 Subject: net: ethernet: ti: cpsw: remove unused priv lock There is no reason in this lock. At least for now. Signed-off-by: Ivan Khoronzhuk Reviewed-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 4b08a2f..635979b 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -364,7 +364,6 @@ static inline void slave_write(struct cpsw_slave *slave, u32 val, u32 offset) } struct cpsw_priv { - spinlock_t lock; struct platform_device *pdev; struct net_device *ndev; struct napi_struct napi_rx; @@ -2124,7 +2123,6 @@ static int cpsw_probe_dual_emac(struct platform_device *pdev, } priv_sl2 = netdev_priv(ndev); - spin_lock_init(&priv_sl2->lock); priv_sl2->data = *data; priv_sl2->pdev = pdev; priv_sl2->ndev = ndev; @@ -2243,7 +2241,6 @@ static int cpsw_probe(struct platform_device *pdev) platform_set_drvdata(pdev, ndev); priv = netdev_priv(ndev); - spin_lock_init(&priv->lock); priv->pdev = pdev; priv->ndev = ndev; priv->dev = &ndev->dev; -- cgit v0.10.2 From a91eb52abb504a1dd3248a5d07b54e7f95d5fcf1 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Fri, 3 Jun 2016 14:35:32 +0300 Subject: qed: Revisit chain implementation RoCE driver is going to need a 32-bit chain [current chain implementation for qed* currently supports only 16-bit producer/consumer chains]. This patch adds said support, as well as doing other slight tweaks and modifications to qed's chain API. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index e9ce6a7..151173d 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -1779,92 +1780,285 @@ void qed_hw_remove(struct qed_dev *cdev) qed_iov_free_hw_info(cdev); } -int qed_chain_alloc(struct qed_dev *cdev, - enum qed_chain_use_mode intended_use, - enum qed_chain_mode mode, - u16 num_elems, - size_t elem_size, - struct qed_chain *p_chain) +static void qed_chain_free_next_ptr(struct qed_dev *cdev, + struct qed_chain *p_chain) +{ + void *p_virt = p_chain->p_virt_addr, *p_virt_next = NULL; + dma_addr_t p_phys = p_chain->p_phys_addr, p_phys_next = 0; + struct qed_chain_next *p_next; + u32 size, i; + + if (!p_virt) + return; + + size = p_chain->elem_size * p_chain->usable_per_page; + + for (i = 0; i < p_chain->page_cnt; i++) { + if (!p_virt) + break; + + p_next = (struct qed_chain_next *)((u8 *)p_virt + size); + p_virt_next = p_next->next_virt; + p_phys_next = HILO_DMA_REGPAIR(p_next->next_phys); + + dma_free_coherent(&cdev->pdev->dev, + QED_CHAIN_PAGE_SIZE, p_virt, p_phys); + + p_virt = p_virt_next; + p_phys = p_phys_next; + } +} + +static void qed_chain_free_single(struct qed_dev *cdev, + struct qed_chain *p_chain) +{ + if (!p_chain->p_virt_addr) + return; + + dma_free_coherent(&cdev->pdev->dev, + QED_CHAIN_PAGE_SIZE, + p_chain->p_virt_addr, p_chain->p_phys_addr); +} + +static void qed_chain_free_pbl(struct qed_dev *cdev, struct qed_chain *p_chain) +{ + void **pp_virt_addr_tbl = p_chain->pbl.pp_virt_addr_tbl; + u32 page_cnt = p_chain->page_cnt, i, pbl_size; + u8 *p_pbl_virt = p_chain->pbl.p_virt_table; + + if (!pp_virt_addr_tbl) + return; + + if (!p_chain->pbl.p_virt_table) + goto out; + + for (i = 0; i < page_cnt; i++) { + if (!pp_virt_addr_tbl[i]) + break; + + dma_free_coherent(&cdev->pdev->dev, + QED_CHAIN_PAGE_SIZE, + pp_virt_addr_tbl[i], + *(dma_addr_t *)p_pbl_virt); + + p_pbl_virt += QED_CHAIN_PBL_ENTRY_SIZE; + } + + pbl_size = page_cnt * QED_CHAIN_PBL_ENTRY_SIZE; + dma_free_coherent(&cdev->pdev->dev, + pbl_size, + p_chain->pbl.p_virt_table, p_chain->pbl.p_phys_table); +out: + vfree(p_chain->pbl.pp_virt_addr_tbl); +} + +void qed_chain_free(struct qed_dev *cdev, struct qed_chain *p_chain) +{ + switch (p_chain->mode) { + case QED_CHAIN_MODE_NEXT_PTR: + qed_chain_free_next_ptr(cdev, p_chain); + break; + case QED_CHAIN_MODE_SINGLE: + qed_chain_free_single(cdev, p_chain); + break; + case QED_CHAIN_MODE_PBL: + qed_chain_free_pbl(cdev, p_chain); + break; + } +} + +static int +qed_chain_alloc_sanity_check(struct qed_dev *cdev, + enum qed_chain_cnt_type cnt_type, + size_t elem_size, u32 page_cnt) +{ + u64 chain_size = ELEMS_PER_PAGE(elem_size) * page_cnt; + + /* The actual chain size can be larger than the maximal possible value + * after rounding up the requested elements number to pages, and after + * taking into acount the unusuable elements (next-ptr elements). + * The size of a "u16" chain can be (U16_MAX + 1) since the chain + * size/capacity fields are of a u32 type. + */ + if ((cnt_type == QED_CHAIN_CNT_TYPE_U16 && + chain_size > 0x10000) || + (cnt_type == QED_CHAIN_CNT_TYPE_U32 && + chain_size > 0x100000000ULL)) { + DP_NOTICE(cdev, + "The actual chain size (0x%llx) is larger than the maximal possible value\n", + chain_size); + return -EINVAL; + } + + return 0; +} + +static int +qed_chain_alloc_next_ptr(struct qed_dev *cdev, struct qed_chain *p_chain) { - dma_addr_t p_pbl_phys = 0; - void *p_pbl_virt = NULL; + void *p_virt = NULL, *p_virt_prev = NULL; dma_addr_t p_phys = 0; - void *p_virt = NULL; - u16 page_cnt = 0; - size_t size; + u32 i; - if (mode == QED_CHAIN_MODE_SINGLE) - page_cnt = 1; - else - page_cnt = QED_CHAIN_PAGE_CNT(num_elems, elem_size, mode); + for (i = 0; i < p_chain->page_cnt; i++) { + p_virt = dma_alloc_coherent(&cdev->pdev->dev, + QED_CHAIN_PAGE_SIZE, + &p_phys, GFP_KERNEL); + if (!p_virt) { + DP_NOTICE(cdev, "Failed to allocate chain memory\n"); + return -ENOMEM; + } + + if (i == 0) { + qed_chain_init_mem(p_chain, p_virt, p_phys); + qed_chain_reset(p_chain); + } else { + qed_chain_init_next_ptr_elem(p_chain, p_virt_prev, + p_virt, p_phys); + } + + p_virt_prev = p_virt; + } + /* Last page's next element should point to the beginning of the + * chain. + */ + qed_chain_init_next_ptr_elem(p_chain, p_virt_prev, + p_chain->p_virt_addr, + p_chain->p_phys_addr); + + return 0; +} + +static int +qed_chain_alloc_single(struct qed_dev *cdev, struct qed_chain *p_chain) +{ + dma_addr_t p_phys = 0; + void *p_virt = NULL; - size = page_cnt * QED_CHAIN_PAGE_SIZE; p_virt = dma_alloc_coherent(&cdev->pdev->dev, - size, &p_phys, GFP_KERNEL); + QED_CHAIN_PAGE_SIZE, &p_phys, GFP_KERNEL); if (!p_virt) { - DP_NOTICE(cdev, "Failed to allocate chain mem\n"); - goto nomem; + DP_NOTICE(cdev, "Failed to allocate chain memory\n"); + return -ENOMEM; } - if (mode == QED_CHAIN_MODE_PBL) { - size = page_cnt * QED_CHAIN_PBL_ENTRY_SIZE; - p_pbl_virt = dma_alloc_coherent(&cdev->pdev->dev, - size, &p_pbl_phys, - GFP_KERNEL); - if (!p_pbl_virt) { - DP_NOTICE(cdev, "Failed to allocate chain pbl mem\n"); - goto nomem; - } + qed_chain_init_mem(p_chain, p_virt, p_phys); + qed_chain_reset(p_chain); - qed_chain_pbl_init(p_chain, p_virt, p_phys, page_cnt, - (u8)elem_size, intended_use, - p_pbl_phys, p_pbl_virt); - } else { - qed_chain_init(p_chain, p_virt, p_phys, page_cnt, - (u8)elem_size, intended_use, mode); + return 0; +} + +static int qed_chain_alloc_pbl(struct qed_dev *cdev, struct qed_chain *p_chain) +{ + u32 page_cnt = p_chain->page_cnt, size, i; + dma_addr_t p_phys = 0, p_pbl_phys = 0; + void **pp_virt_addr_tbl = NULL; + u8 *p_pbl_virt = NULL; + void *p_virt = NULL; + + size = page_cnt * sizeof(*pp_virt_addr_tbl); + pp_virt_addr_tbl = vmalloc(size); + if (!pp_virt_addr_tbl) { + DP_NOTICE(cdev, + "Failed to allocate memory for the chain virtual addresses table\n"); + return -ENOMEM; } + memset(pp_virt_addr_tbl, 0, size); - return 0; + /* The allocation of the PBL table is done with its full size, since it + * is expected to be successive. + * qed_chain_init_pbl_mem() is called even in a case of an allocation + * failure, since pp_virt_addr_tbl was previously allocated, and it + * should be saved to allow its freeing during the error flow. + */ + size = page_cnt * QED_CHAIN_PBL_ENTRY_SIZE; + p_pbl_virt = dma_alloc_coherent(&cdev->pdev->dev, + size, &p_pbl_phys, GFP_KERNEL); + qed_chain_init_pbl_mem(p_chain, p_pbl_virt, p_pbl_phys, + pp_virt_addr_tbl); + if (!p_pbl_virt) { + DP_NOTICE(cdev, "Failed to allocate chain pbl memory\n"); + return -ENOMEM; + } -nomem: - dma_free_coherent(&cdev->pdev->dev, - page_cnt * QED_CHAIN_PAGE_SIZE, - p_virt, p_phys); - dma_free_coherent(&cdev->pdev->dev, - page_cnt * QED_CHAIN_PBL_ENTRY_SIZE, - p_pbl_virt, p_pbl_phys); + for (i = 0; i < page_cnt; i++) { + p_virt = dma_alloc_coherent(&cdev->pdev->dev, + QED_CHAIN_PAGE_SIZE, + &p_phys, GFP_KERNEL); + if (!p_virt) { + DP_NOTICE(cdev, "Failed to allocate chain memory\n"); + return -ENOMEM; + } - return -ENOMEM; + if (i == 0) { + qed_chain_init_mem(p_chain, p_virt, p_phys); + qed_chain_reset(p_chain); + } + + /* Fill the PBL table with the physical address of the page */ + *(dma_addr_t *)p_pbl_virt = p_phys; + /* Keep the virtual address of the page */ + p_chain->pbl.pp_virt_addr_tbl[i] = p_virt; + + p_pbl_virt += QED_CHAIN_PBL_ENTRY_SIZE; + } + + return 0; } -void qed_chain_free(struct qed_dev *cdev, - struct qed_chain *p_chain) +int qed_chain_alloc(struct qed_dev *cdev, + enum qed_chain_use_mode intended_use, + enum qed_chain_mode mode, + enum qed_chain_cnt_type cnt_type, + u32 num_elems, size_t elem_size, struct qed_chain *p_chain) { - size_t size; + u32 page_cnt; + int rc = 0; - if (!p_chain->p_virt_addr) - return; + if (mode == QED_CHAIN_MODE_SINGLE) + page_cnt = 1; + else + page_cnt = QED_CHAIN_PAGE_CNT(num_elems, elem_size, mode); - if (p_chain->mode == QED_CHAIN_MODE_PBL) { - size = p_chain->page_cnt * QED_CHAIN_PBL_ENTRY_SIZE; - dma_free_coherent(&cdev->pdev->dev, size, - p_chain->pbl.p_virt_table, - p_chain->pbl.p_phys_table); + rc = qed_chain_alloc_sanity_check(cdev, cnt_type, elem_size, page_cnt); + if (rc) { + DP_NOTICE(cdev, + "Cannot allocate a chain with the given arguments:\n" + "[use_mode %d, mode %d, cnt_type %d, num_elems %d, elem_size %zu]\n", + intended_use, mode, cnt_type, num_elems, elem_size); + return rc; } - size = p_chain->page_cnt * QED_CHAIN_PAGE_SIZE; - dma_free_coherent(&cdev->pdev->dev, size, - p_chain->p_virt_addr, - p_chain->p_phys_addr); + qed_chain_init_params(p_chain, page_cnt, (u8) elem_size, intended_use, + mode, cnt_type); + + switch (mode) { + case QED_CHAIN_MODE_NEXT_PTR: + rc = qed_chain_alloc_next_ptr(cdev, p_chain); + break; + case QED_CHAIN_MODE_SINGLE: + rc = qed_chain_alloc_single(cdev, p_chain); + break; + case QED_CHAIN_MODE_PBL: + rc = qed_chain_alloc_pbl(cdev, p_chain); + break; + } + if (rc) + goto nomem; + + return 0; + +nomem: + qed_chain_free(cdev, p_chain); + return rc; } -int qed_fw_l2_queue(struct qed_hwfn *p_hwfn, - u16 src_id, u16 *dst_id) +int qed_fw_l2_queue(struct qed_hwfn *p_hwfn, u16 src_id, u16 *dst_id) { if (src_id >= RESC_NUM(p_hwfn, QED_L2_QUEUE)) { u16 min, max; - min = (u16)RESC_START(p_hwfn, QED_L2_QUEUE); + min = (u16) RESC_START(p_hwfn, QED_L2_QUEUE); max = min + RESC_NUM(p_hwfn, QED_L2_QUEUE); DP_NOTICE(p_hwfn, "l2_queue id [%d] is not valid, available indices [%d - %d]\n", diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h index dde364d..f810ce4 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h +++ b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h @@ -245,9 +245,8 @@ int qed_chain_alloc(struct qed_dev *cdev, enum qed_chain_use_mode intended_use, enum qed_chain_mode mode, - u16 num_elems, - size_t elem_size, - struct qed_chain *p_chain); + enum qed_chain_cnt_type cnt_type, + u32 num_elems, size_t elem_size, struct qed_chain *p_chain); /** * @brief qed_chain_free - Free chain DMA memory @@ -255,8 +254,7 @@ qed_chain_alloc(struct qed_dev *cdev, * @param p_hwfn * @param p_chain */ -void qed_chain_free(struct qed_dev *cdev, - struct qed_chain *p_chain); +void qed_chain_free(struct qed_dev *cdev, struct qed_chain *p_chain); /** * @@brief qed_fw_l2_queue - Get absolute L2 queue ID diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c index 1225064..63b93fb 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c @@ -308,6 +308,7 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, struct qed_spq_entry *p_ent = NULL; struct qed_sp_init_data init_data; int rc = -EINVAL; + u8 page_cnt; /* update initial eq producer */ qed_eq_prod_update(p_hwfn, @@ -350,8 +351,8 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, /* Place EQ address in RAMROD */ DMA_REGPAIR_LE(p_ramrod->event_ring_pbl_addr, p_hwfn->p_eq->chain.pbl.p_phys_table); - p_ramrod->event_ring_num_pages = (u8)p_hwfn->p_eq->chain.page_cnt; - + page_cnt = (u8)qed_chain_get_page_cnt(&p_hwfn->p_eq->chain); + p_ramrod->event_ring_num_pages = page_cnt; DMA_REGPAIR_LE(p_ramrod->consolid_q_pbl_addr, p_hwfn->p_consq->chain.pbl.p_phys_table); diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index acac662..ad9bf5c 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -343,6 +343,7 @@ struct qed_eq *qed_eq_alloc(struct qed_hwfn *p_hwfn, if (qed_chain_alloc(p_hwfn->cdev, QED_CHAIN_USE_TO_PRODUCE, QED_CHAIN_MODE_PBL, + QED_CHAIN_CNT_TYPE_U16, num_elem, sizeof(union event_ring_element), &p_eq->chain)) { @@ -416,10 +417,10 @@ int qed_eth_cqe_completion(struct qed_hwfn *p_hwfn, ***************************************************************************/ void qed_spq_setup(struct qed_hwfn *p_hwfn) { - struct qed_spq *p_spq = p_hwfn->p_spq; - struct qed_spq_entry *p_virt = NULL; - dma_addr_t p_phys = 0; - unsigned int i = 0; + struct qed_spq *p_spq = p_hwfn->p_spq; + struct qed_spq_entry *p_virt = NULL; + dma_addr_t p_phys = 0; + u32 i, capacity; INIT_LIST_HEAD(&p_spq->pending); INIT_LIST_HEAD(&p_spq->completion_pending); @@ -431,7 +432,8 @@ void qed_spq_setup(struct qed_hwfn *p_hwfn) p_phys = p_spq->p_phys + offsetof(struct qed_spq_entry, ramrod); p_virt = p_spq->p_virt; - for (i = 0; i < p_spq->chain.capacity; i++) { + capacity = qed_chain_get_capacity(&p_spq->chain); + for (i = 0; i < capacity; i++) { DMA_REGPAIR_LE(p_virt->elem.data_ptr, p_phys); list_add_tail(&p_virt->list, &p_spq->free_pool); @@ -459,9 +461,10 @@ void qed_spq_setup(struct qed_hwfn *p_hwfn) int qed_spq_alloc(struct qed_hwfn *p_hwfn) { - struct qed_spq *p_spq = NULL; - dma_addr_t p_phys = 0; - struct qed_spq_entry *p_virt = NULL; + struct qed_spq_entry *p_virt = NULL; + struct qed_spq *p_spq = NULL; + dma_addr_t p_phys = 0; + u32 capacity; /* SPQ struct */ p_spq = @@ -475,6 +478,7 @@ int qed_spq_alloc(struct qed_hwfn *p_hwfn) if (qed_chain_alloc(p_hwfn->cdev, QED_CHAIN_USE_TO_PRODUCE, QED_CHAIN_MODE_SINGLE, + QED_CHAIN_CNT_TYPE_U16, 0, /* N/A when the mode is SINGLE */ sizeof(struct slow_path_element), &p_spq->chain)) { @@ -483,11 +487,11 @@ int qed_spq_alloc(struct qed_hwfn *p_hwfn) } /* allocate and fill the SPQ elements (incl. ramrod data list) */ + capacity = qed_chain_get_capacity(&p_spq->chain); p_virt = dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, - p_spq->chain.capacity * + capacity * sizeof(struct qed_spq_entry), - &p_phys, - GFP_KERNEL); + &p_phys, GFP_KERNEL); if (!p_virt) goto spq_allocate_fail; @@ -507,16 +511,18 @@ spq_allocate_fail: void qed_spq_free(struct qed_hwfn *p_hwfn) { struct qed_spq *p_spq = p_hwfn->p_spq; + u32 capacity; if (!p_spq) return; - if (p_spq->p_virt) + if (p_spq->p_virt) { + capacity = qed_chain_get_capacity(&p_spq->chain); dma_free_coherent(&p_hwfn->cdev->pdev->dev, - p_spq->chain.capacity * + capacity * sizeof(struct qed_spq_entry), - p_spq->p_virt, - p_spq->p_phys); + p_spq->p_virt, p_spq->p_phys); + } qed_chain_free(p_hwfn->cdev, &p_spq->chain); ; @@ -871,9 +877,9 @@ struct qed_consq *qed_consq_alloc(struct qed_hwfn *p_hwfn) if (qed_chain_alloc(p_hwfn->cdev, QED_CHAIN_USE_TO_PRODUCE, QED_CHAIN_MODE_PBL, + QED_CHAIN_CNT_TYPE_U16, QED_CHAIN_PAGE_SIZE / 0x80, - 0x80, - &p_consq->chain)) { + 0x80, &p_consq->chain)) { DP_NOTICE(p_hwfn, "Failed to allocate consq chain"); goto consq_allocate_fail; } diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 4f35247..edd2d90 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -2817,6 +2817,7 @@ static int qede_alloc_mem_rxq(struct qede_dev *edev, rc = edev->ops->common->chain_alloc(edev->cdev, QED_CHAIN_USE_TO_CONSUME_PRODUCE, QED_CHAIN_MODE_NEXT_PTR, + QED_CHAIN_CNT_TYPE_U16, RX_RING_SIZE, sizeof(struct eth_rx_bd), &rxq->rx_bd_ring); @@ -2828,6 +2829,7 @@ static int qede_alloc_mem_rxq(struct qede_dev *edev, rc = edev->ops->common->chain_alloc(edev->cdev, QED_CHAIN_USE_TO_CONSUME, QED_CHAIN_MODE_PBL, + QED_CHAIN_CNT_TYPE_U16, RX_RING_SIZE, sizeof(union eth_rx_cqe), &rxq->rx_comp_ring); @@ -2879,9 +2881,9 @@ static int qede_alloc_mem_txq(struct qede_dev *edev, rc = edev->ops->common->chain_alloc(edev->cdev, QED_CHAIN_USE_TO_CONSUME_PRODUCE, QED_CHAIN_MODE_PBL, + QED_CHAIN_CNT_TYPE_U16, NUM_TX_BDS_MAX, - sizeof(*p_virt), - &txq->tx_pbl); + sizeof(*p_virt), &txq->tx_pbl); if (rc) goto err; diff --git a/include/linux/qed/qed_chain.h b/include/linux/qed/qed_chain.h index 5f8fcaa..eceaa9e 100644 --- a/include/linux/qed/qed_chain.h +++ b/include/linux/qed/qed_chain.h @@ -47,16 +47,56 @@ enum qed_chain_use_mode { QED_CHAIN_USE_TO_CONSUME_PRODUCE, /* Chain starts empty */ }; +enum qed_chain_cnt_type { + /* The chain's size/prod/cons are kept in 16-bit variables */ + QED_CHAIN_CNT_TYPE_U16, + + /* The chain's size/prod/cons are kept in 32-bit variables */ + QED_CHAIN_CNT_TYPE_U32, +}; + struct qed_chain_next { struct regpair next_phys; void *next_virt; }; +struct qed_chain_pbl_u16 { + u16 prod_page_idx; + u16 cons_page_idx; +}; + +struct qed_chain_pbl_u32 { + u32 prod_page_idx; + u32 cons_page_idx; +}; + struct qed_chain_pbl { + /* Base address of a pre-allocated buffer for pbl */ dma_addr_t p_phys_table; void *p_virt_table; - u16 prod_page_idx; - u16 cons_page_idx; + + /* Table for keeping the virtual addresses of the chain pages, + * respectively to the physical addresses in the pbl table. + */ + void **pp_virt_addr_tbl; + + /* Index to current used page by producer/consumer */ + union { + struct qed_chain_pbl_u16 pbl16; + struct qed_chain_pbl_u32 pbl32; + } u; +}; + +struct qed_chain_u16 { + /* Cyclic index of next element to produce/consme */ + u16 prod_idx; + u16 cons_idx; +}; + +struct qed_chain_u32 { + /* Cyclic index of next element to produce/consme */ + u32 prod_idx; + u32 cons_idx; }; struct qed_chain { @@ -64,13 +104,25 @@ struct qed_chain { dma_addr_t p_phys_addr; void *p_prod_elem; void *p_cons_elem; - u16 page_cnt; + enum qed_chain_mode mode; enum qed_chain_use_mode intended_use; /* used to produce/consume */ - u16 capacity; /*< number of _usable_ elements */ - u16 size; /* number of elements */ - u16 prod_idx; - u16 cons_idx; + enum qed_chain_cnt_type cnt_type; + + union { + struct qed_chain_u16 chain16; + struct qed_chain_u32 chain32; + } u; + + u32 page_cnt; + + /* Number of elements - capacity is for usable elements only, + * while size will contain total number of elements [for entire chain]. + */ + u32 capacity; + u32 size; + + /* Elements information for fast calculations */ u16 elem_per_page; u16 elem_per_page_mask; u16 elem_unusable; @@ -96,66 +148,69 @@ struct qed_chain { #define QED_CHAIN_PAGE_CNT(elem_cnt, elem_size, mode) \ DIV_ROUND_UP(elem_cnt, USABLE_ELEMS_PER_PAGE(elem_size, mode)) +#define is_chain_u16(p) ((p)->cnt_type == QED_CHAIN_CNT_TYPE_U16) +#define is_chain_u32(p) ((p)->cnt_type == QED_CHAIN_CNT_TYPE_U32) + /* Accessors */ static inline u16 qed_chain_get_prod_idx(struct qed_chain *p_chain) { - return p_chain->prod_idx; + return p_chain->u.chain16.prod_idx; } static inline u16 qed_chain_get_cons_idx(struct qed_chain *p_chain) { - return p_chain->cons_idx; + return p_chain->u.chain16.cons_idx; +} + +static inline u32 qed_chain_get_cons_idx_u32(struct qed_chain *p_chain) +{ + return p_chain->u.chain32.cons_idx; } static inline u16 qed_chain_get_elem_left(struct qed_chain *p_chain) { u16 used; - /* we don't need to trancate upon assignmet, as we assign u32->u16 */ - used = ((u32)0x10000u + (u32)(p_chain->prod_idx)) - - (u32)p_chain->cons_idx; + used = (u16) (((u32)0x10000 + + (u32)p_chain->u.chain16.prod_idx) - + (u32)p_chain->u.chain16.cons_idx); if (p_chain->mode == QED_CHAIN_MODE_NEXT_PTR) - used -= p_chain->prod_idx / p_chain->elem_per_page - - p_chain->cons_idx / p_chain->elem_per_page; + used -= p_chain->u.chain16.prod_idx / p_chain->elem_per_page - + p_chain->u.chain16.cons_idx / p_chain->elem_per_page; - return p_chain->capacity - used; + return (u16)(p_chain->capacity - used); } -static inline u8 qed_chain_is_full(struct qed_chain *p_chain) +static inline u32 qed_chain_get_elem_left_u32(struct qed_chain *p_chain) { - return qed_chain_get_elem_left(p_chain) == p_chain->capacity; -} + u32 used; -static inline u8 qed_chain_is_empty(struct qed_chain *p_chain) -{ - return qed_chain_get_elem_left(p_chain) == 0; -} + used = (u32) (((u64)0x100000000ULL + + (u64)p_chain->u.chain32.prod_idx) - + (u64)p_chain->u.chain32.cons_idx); + if (p_chain->mode == QED_CHAIN_MODE_NEXT_PTR) + used -= p_chain->u.chain32.prod_idx / p_chain->elem_per_page - + p_chain->u.chain32.cons_idx / p_chain->elem_per_page; -static inline u16 qed_chain_get_elem_per_page( - struct qed_chain *p_chain) -{ - return p_chain->elem_per_page; + return p_chain->capacity - used; } -static inline u16 qed_chain_get_usable_per_page( - struct qed_chain *p_chain) +static inline u16 qed_chain_get_usable_per_page(struct qed_chain *p_chain) { return p_chain->usable_per_page; } -static inline u16 qed_chain_get_unusable_per_page( - struct qed_chain *p_chain) +static inline u16 qed_chain_get_unusable_per_page(struct qed_chain *p_chain) { return p_chain->elem_unusable; } -static inline u16 qed_chain_get_size(struct qed_chain *p_chain) +static inline u32 qed_chain_get_page_cnt(struct qed_chain *p_chain) { - return p_chain->size; + return p_chain->page_cnt; } -static inline dma_addr_t -qed_chain_get_pbl_phys(struct qed_chain *p_chain) +static inline dma_addr_t qed_chain_get_pbl_phys(struct qed_chain *p_chain) { return p_chain->pbl.p_phys_table; } @@ -172,65 +227,63 @@ qed_chain_get_pbl_phys(struct qed_chain *p_chain) */ static inline void qed_chain_advance_page(struct qed_chain *p_chain, - void **p_next_elem, - u16 *idx_to_inc, - u16 *page_to_inc) + void **p_next_elem, void *idx_to_inc, void *page_to_inc) { + struct qed_chain_next *p_next = NULL; + u32 page_index = 0; switch (p_chain->mode) { case QED_CHAIN_MODE_NEXT_PTR: - { - struct qed_chain_next *p_next = *p_next_elem; + p_next = *p_next_elem; *p_next_elem = p_next->next_virt; - *idx_to_inc += p_chain->elem_unusable; + if (is_chain_u16(p_chain)) + *(u16 *)idx_to_inc += p_chain->elem_unusable; + else + *(u32 *)idx_to_inc += p_chain->elem_unusable; break; - } case QED_CHAIN_MODE_SINGLE: *p_next_elem = p_chain->p_virt_addr; break; case QED_CHAIN_MODE_PBL: - /* It is assumed pages are sequential, next element needs - * to change only when passing going back to first from last. - */ - if (++(*page_to_inc) == p_chain->page_cnt) { - *page_to_inc = 0; - *p_next_elem = p_chain->p_virt_addr; + if (is_chain_u16(p_chain)) { + if (++(*(u16 *)page_to_inc) == p_chain->page_cnt) + *(u16 *)page_to_inc = 0; + page_index = *(u16 *)page_to_inc; + } else { + if (++(*(u32 *)page_to_inc) == p_chain->page_cnt) + *(u32 *)page_to_inc = 0; + page_index = *(u32 *)page_to_inc; } + *p_next_elem = p_chain->pbl.pp_virt_addr_tbl[page_index]; } } #define is_unusable_idx(p, idx) \ - (((p)->idx & (p)->elem_per_page_mask) == (p)->usable_per_page) + (((p)->u.chain16.idx & (p)->elem_per_page_mask) == (p)->usable_per_page) + +#define is_unusable_idx_u32(p, idx) \ + (((p)->u.chain32.idx & (p)->elem_per_page_mask) == (p)->usable_per_page) +#define is_unusable_next_idx(p, idx) \ + ((((p)->u.chain16.idx + 1) & (p)->elem_per_page_mask) == \ + (p)->usable_per_page) -#define is_unusable_next_idx(p, idx) \ - ((((p)->idx + 1) & (p)->elem_per_page_mask) == (p)->usable_per_page) +#define is_unusable_next_idx_u32(p, idx) \ + ((((p)->u.chain32.idx + 1) & (p)->elem_per_page_mask) == \ + (p)->usable_per_page) -#define test_ans_skip(p, idx) \ +#define test_and_skip(p, idx) \ do { \ - if (is_unusable_idx(p, idx)) { \ - (p)->idx += (p)->elem_unusable; \ + if (is_chain_u16(p)) { \ + if (is_unusable_idx(p, idx)) \ + (p)->u.chain16.idx += (p)->elem_unusable; \ + } else { \ + if (is_unusable_idx_u32(p, idx)) \ + (p)->u.chain32.idx += (p)->elem_unusable; \ } \ } while (0) /** - * @brief qed_chain_return_multi_produced - - * - * A chain in which the driver "Produces" elements should use this API - * to indicate previous produced elements are now consumed. - * - * @param p_chain - * @param num - */ -static inline void -qed_chain_return_multi_produced(struct qed_chain *p_chain, - u16 num) -{ - p_chain->cons_idx += num; - test_ans_skip(p_chain, cons_idx); -} - -/** * @brief qed_chain_return_produced - * * A chain in which the driver "Produces" elements should use this API @@ -240,8 +293,11 @@ qed_chain_return_multi_produced(struct qed_chain *p_chain, */ static inline void qed_chain_return_produced(struct qed_chain *p_chain) { - p_chain->cons_idx++; - test_ans_skip(p_chain, cons_idx); + if (is_chain_u16(p_chain)) + p_chain->u.chain16.cons_idx++; + else + p_chain->u.chain32.cons_idx++; + test_and_skip(p_chain, cons_idx); } /** @@ -257,21 +313,33 @@ static inline void qed_chain_return_produced(struct qed_chain *p_chain) */ static inline void *qed_chain_produce(struct qed_chain *p_chain) { - void *ret = NULL; - - if ((p_chain->prod_idx & p_chain->elem_per_page_mask) == - p_chain->next_page_mask) { - qed_chain_advance_page(p_chain, &p_chain->p_prod_elem, - &p_chain->prod_idx, - &p_chain->pbl.prod_page_idx); + void *p_ret = NULL, *p_prod_idx, *p_prod_page_idx; + + if (is_chain_u16(p_chain)) { + if ((p_chain->u.chain16.prod_idx & + p_chain->elem_per_page_mask) == p_chain->next_page_mask) { + p_prod_idx = &p_chain->u.chain16.prod_idx; + p_prod_page_idx = &p_chain->pbl.u.pbl16.prod_page_idx; + qed_chain_advance_page(p_chain, &p_chain->p_prod_elem, + p_prod_idx, p_prod_page_idx); + } + p_chain->u.chain16.prod_idx++; + } else { + if ((p_chain->u.chain32.prod_idx & + p_chain->elem_per_page_mask) == p_chain->next_page_mask) { + p_prod_idx = &p_chain->u.chain32.prod_idx; + p_prod_page_idx = &p_chain->pbl.u.pbl32.prod_page_idx; + qed_chain_advance_page(p_chain, &p_chain->p_prod_elem, + p_prod_idx, p_prod_page_idx); + } + p_chain->u.chain32.prod_idx++; } - ret = p_chain->p_prod_elem; - p_chain->prod_idx++; + p_ret = p_chain->p_prod_elem; p_chain->p_prod_elem = (void *)(((u8 *)p_chain->p_prod_elem) + p_chain->elem_size); - return ret; + return p_ret; } /** @@ -282,9 +350,9 @@ static inline void *qed_chain_produce(struct qed_chain *p_chain) * @param p_chain * @param num * - * @return u16, number of unusable BDs + * @return number of unusable BDs */ -static inline u16 qed_chain_get_capacity(struct qed_chain *p_chain) +static inline u32 qed_chain_get_capacity(struct qed_chain *p_chain) { return p_chain->capacity; } @@ -297,11 +365,13 @@ static inline u16 qed_chain_get_capacity(struct qed_chain *p_chain) * * @param p_chain */ -static inline void -qed_chain_recycle_consumed(struct qed_chain *p_chain) +static inline void qed_chain_recycle_consumed(struct qed_chain *p_chain) { - test_ans_skip(p_chain, prod_idx); - p_chain->prod_idx++; + test_and_skip(p_chain, prod_idx); + if (is_chain_u16(p_chain)) + p_chain->u.chain16.prod_idx++; + else + p_chain->u.chain32.prod_idx++; } /** @@ -316,21 +386,33 @@ qed_chain_recycle_consumed(struct qed_chain *p_chain) */ static inline void *qed_chain_consume(struct qed_chain *p_chain) { - void *ret = NULL; - - if ((p_chain->cons_idx & p_chain->elem_per_page_mask) == - p_chain->next_page_mask) { + void *p_ret = NULL, *p_cons_idx, *p_cons_page_idx; + + if (is_chain_u16(p_chain)) { + if ((p_chain->u.chain16.cons_idx & + p_chain->elem_per_page_mask) == p_chain->next_page_mask) { + p_cons_idx = &p_chain->u.chain16.cons_idx; + p_cons_page_idx = &p_chain->pbl.u.pbl16.cons_page_idx; + qed_chain_advance_page(p_chain, &p_chain->p_cons_elem, + p_cons_idx, p_cons_page_idx); + } + p_chain->u.chain16.cons_idx++; + } else { + if ((p_chain->u.chain32.cons_idx & + p_chain->elem_per_page_mask) == p_chain->next_page_mask) { + p_cons_idx = &p_chain->u.chain32.cons_idx; + p_cons_page_idx = &p_chain->pbl.u.pbl32.cons_page_idx; qed_chain_advance_page(p_chain, &p_chain->p_cons_elem, - &p_chain->cons_idx, - &p_chain->pbl.cons_page_idx); + p_cons_idx, p_cons_page_idx); + } + p_chain->u.chain32.cons_idx++; } - ret = p_chain->p_cons_elem; - p_chain->cons_idx++; + p_ret = p_chain->p_cons_elem; p_chain->p_cons_elem = (void *)(((u8 *)p_chain->p_cons_elem) + p_chain->elem_size); - return ret; + return p_ret; } /** @@ -340,16 +422,33 @@ static inline void *qed_chain_consume(struct qed_chain *p_chain) */ static inline void qed_chain_reset(struct qed_chain *p_chain) { - int i; - - p_chain->prod_idx = 0; - p_chain->cons_idx = 0; - p_chain->p_cons_elem = p_chain->p_virt_addr; - p_chain->p_prod_elem = p_chain->p_virt_addr; + u32 i; + + if (is_chain_u16(p_chain)) { + p_chain->u.chain16.prod_idx = 0; + p_chain->u.chain16.cons_idx = 0; + } else { + p_chain->u.chain32.prod_idx = 0; + p_chain->u.chain32.cons_idx = 0; + } + p_chain->p_cons_elem = p_chain->p_virt_addr; + p_chain->p_prod_elem = p_chain->p_virt_addr; if (p_chain->mode == QED_CHAIN_MODE_PBL) { - p_chain->pbl.prod_page_idx = p_chain->page_cnt - 1; - p_chain->pbl.cons_page_idx = p_chain->page_cnt - 1; + /* Use (page_cnt - 1) as a reset value for the prod/cons page's + * indices, to avoid unnecessary page advancing on the first + * call to qed_chain_produce/consume. Instead, the indices + * will be advanced to page_cnt and then will be wrapped to 0. + */ + u32 reset_val = p_chain->page_cnt - 1; + + if (is_chain_u16(p_chain)) { + p_chain->pbl.u.pbl16.prod_page_idx = (u16)reset_val; + p_chain->pbl.u.pbl16.cons_page_idx = (u16)reset_val; + } else { + p_chain->pbl.u.pbl32.prod_page_idx = reset_val; + p_chain->pbl.u.pbl32.cons_page_idx = reset_val; + } } switch (p_chain->intended_use) { @@ -377,168 +476,184 @@ static inline void qed_chain_reset(struct qed_chain *p_chain) * @param intended_use * @param mode */ -static inline void qed_chain_init(struct qed_chain *p_chain, - void *p_virt_addr, - dma_addr_t p_phys_addr, - u16 page_cnt, - u8 elem_size, - enum qed_chain_use_mode intended_use, - enum qed_chain_mode mode) +static inline void qed_chain_init_params(struct qed_chain *p_chain, + u32 page_cnt, + u8 elem_size, + enum qed_chain_use_mode intended_use, + enum qed_chain_mode mode, + enum qed_chain_cnt_type cnt_type) { /* chain fixed parameters */ - p_chain->p_virt_addr = p_virt_addr; - p_chain->p_phys_addr = p_phys_addr; + p_chain->p_virt_addr = NULL; + p_chain->p_phys_addr = 0; p_chain->elem_size = elem_size; - p_chain->page_cnt = page_cnt; + p_chain->intended_use = intended_use; p_chain->mode = mode; + p_chain->cnt_type = cnt_type; - p_chain->intended_use = intended_use; p_chain->elem_per_page = ELEMS_PER_PAGE(elem_size); - p_chain->usable_per_page = - USABLE_ELEMS_PER_PAGE(elem_size, mode); - p_chain->capacity = p_chain->usable_per_page * page_cnt; - p_chain->size = p_chain->elem_per_page * page_cnt; + p_chain->usable_per_page = USABLE_ELEMS_PER_PAGE(elem_size, mode); p_chain->elem_per_page_mask = p_chain->elem_per_page - 1; - p_chain->elem_unusable = UNUSABLE_ELEMS_PER_PAGE(elem_size, mode); - p_chain->next_page_mask = (p_chain->usable_per_page & p_chain->elem_per_page_mask); - if (mode == QED_CHAIN_MODE_NEXT_PTR) { - struct qed_chain_next *p_next; - u16 i; - - for (i = 0; i < page_cnt - 1; i++) { - /* Increment mem_phy to the next page. */ - p_phys_addr += QED_CHAIN_PAGE_SIZE; - - /* Initialize the physical address of the next page. */ - p_next = (struct qed_chain_next *)((u8 *)p_virt_addr + - elem_size * - p_chain-> - usable_per_page); - - p_next->next_phys.lo = DMA_LO_LE(p_phys_addr); - p_next->next_phys.hi = DMA_HI_LE(p_phys_addr); - - /* Initialize the virtual address of the next page. */ - p_next->next_virt = (void *)((u8 *)p_virt_addr + - QED_CHAIN_PAGE_SIZE); - - /* Move to the next page. */ - p_virt_addr = p_next->next_virt; - } - - /* Last page's next should point to beginning of the chain */ - p_next = (struct qed_chain_next *)((u8 *)p_virt_addr + - elem_size * - p_chain->usable_per_page); + p_chain->page_cnt = page_cnt; + p_chain->capacity = p_chain->usable_per_page * page_cnt; + p_chain->size = p_chain->elem_per_page * page_cnt; - p_next->next_phys.lo = DMA_LO_LE(p_chain->p_phys_addr); - p_next->next_phys.hi = DMA_HI_LE(p_chain->p_phys_addr); - p_next->next_virt = p_chain->p_virt_addr; - } - qed_chain_reset(p_chain); + p_chain->pbl.p_phys_table = 0; + p_chain->pbl.p_virt_table = NULL; + p_chain->pbl.pp_virt_addr_tbl = NULL; } /** - * @brief qed_chain_pbl_init - Initalizes a basic pbl chain - * struct + * @brief qed_chain_init_mem - + * + * Initalizes a basic chain struct with its chain buffers + * * @param p_chain * @param p_virt_addr virtual address of allocated buffer's beginning * @param p_phys_addr physical address of allocated buffer's beginning - * @param page_cnt number of pages in the allocated buffer - * @param elem_size size of each element in the chain - * @param use_mode - * @param p_phys_pbl pointer to a pre-allocated side table - * which will hold physical page addresses. - * @param p_virt_pbl pointer to a pre allocated side table - * which will hold virtual page addresses. + * */ -static inline void -qed_chain_pbl_init(struct qed_chain *p_chain, - void *p_virt_addr, - dma_addr_t p_phys_addr, - u16 page_cnt, - u8 elem_size, - enum qed_chain_use_mode use_mode, - dma_addr_t p_phys_pbl, - dma_addr_t *p_virt_pbl) +static inline void qed_chain_init_mem(struct qed_chain *p_chain, + void *p_virt_addr, dma_addr_t p_phys_addr) { - dma_addr_t *p_pbl_dma = p_virt_pbl; - int i; - - qed_chain_init(p_chain, p_virt_addr, p_phys_addr, page_cnt, - elem_size, use_mode, QED_CHAIN_MODE_PBL); + p_chain->p_virt_addr = p_virt_addr; + p_chain->p_phys_addr = p_phys_addr; +} +/** + * @brief qed_chain_init_pbl_mem - + * + * Initalizes a basic chain struct with its pbl buffers + * + * @param p_chain + * @param p_virt_pbl pointer to a pre allocated side table which will hold + * virtual page addresses. + * @param p_phys_pbl pointer to a pre-allocated side table which will hold + * physical page addresses. + * @param pp_virt_addr_tbl + * pointer to a pre-allocated side table which will hold + * the virtual addresses of the chain pages. + * + */ +static inline void qed_chain_init_pbl_mem(struct qed_chain *p_chain, + void *p_virt_pbl, + dma_addr_t p_phys_pbl, + void **pp_virt_addr_tbl) +{ p_chain->pbl.p_phys_table = p_phys_pbl; p_chain->pbl.p_virt_table = p_virt_pbl; - - /* Fill the PBL with physical addresses*/ - for (i = 0; i < page_cnt; i++) { - *p_pbl_dma = p_phys_addr; - p_phys_addr += QED_CHAIN_PAGE_SIZE; - p_pbl_dma++; - } + p_chain->pbl.pp_virt_addr_tbl = pp_virt_addr_tbl; } /** - * @brief qed_chain_set_prod - sets the prod to the given - * value + * @brief qed_chain_init_next_ptr_elem - + * + * Initalizes a next pointer element + * + * @param p_chain + * @param p_virt_curr virtual address of a chain page of which the next + * pointer element is initialized + * @param p_virt_next virtual address of the next chain page + * @param p_phys_next physical address of the next chain page * - * @param prod_idx - * @param p_prod_elem */ -static inline void qed_chain_set_prod(struct qed_chain *p_chain, - u16 prod_idx, - void *p_prod_elem) +static inline void +qed_chain_init_next_ptr_elem(struct qed_chain *p_chain, + void *p_virt_curr, + void *p_virt_next, dma_addr_t p_phys_next) { - p_chain->prod_idx = prod_idx; - p_chain->p_prod_elem = p_prod_elem; + struct qed_chain_next *p_next; + u32 size; + + size = p_chain->elem_size * p_chain->usable_per_page; + p_next = (struct qed_chain_next *)((u8 *)p_virt_curr + size); + + DMA_REGPAIR_LE(p_next->next_phys, p_phys_next); + + p_next->next_virt = p_virt_next; } /** - * @brief qed_chain_get_elem - + * @brief qed_chain_get_last_elem - * - * get a pointer to an element represented by absolute idx + * Returns a pointer to the last element of the chain * * @param p_chain - * @assumption p_chain->size is a power of 2 * - * @return void*, a pointer to next element + * @return void* */ -static inline void *qed_chain_sge_get_elem(struct qed_chain *p_chain, - u16 idx) +static inline void *qed_chain_get_last_elem(struct qed_chain *p_chain) { - void *ret = NULL; - - if (idx >= p_chain->size) - return NULL; + struct qed_chain_next *p_next = NULL; + void *p_virt_addr = NULL; + u32 size, last_page_idx; - ret = (u8 *)p_chain->p_virt_addr + p_chain->elem_size * idx; + if (!p_chain->p_virt_addr) + goto out; - return ret; + switch (p_chain->mode) { + case QED_CHAIN_MODE_NEXT_PTR: + size = p_chain->elem_size * p_chain->usable_per_page; + p_virt_addr = p_chain->p_virt_addr; + p_next = (struct qed_chain_next *)((u8 *)p_virt_addr + size); + while (p_next->next_virt != p_chain->p_virt_addr) { + p_virt_addr = p_next->next_virt; + p_next = (struct qed_chain_next *)((u8 *)p_virt_addr + + size); + } + break; + case QED_CHAIN_MODE_SINGLE: + p_virt_addr = p_chain->p_virt_addr; + break; + case QED_CHAIN_MODE_PBL: + last_page_idx = p_chain->page_cnt - 1; + p_virt_addr = p_chain->pbl.pp_virt_addr_tbl[last_page_idx]; + break; + } + /* p_virt_addr points at this stage to the last page of the chain */ + size = p_chain->elem_size * (p_chain->usable_per_page - 1); + p_virt_addr = (u8 *)p_virt_addr + size; +out: + return p_virt_addr; } /** - * @brief qed_chain_sge_inc_cons_prod + * @brief qed_chain_set_prod - sets the prod to the given value * - * for sge chains, producer isn't increased serially, the ring - * is expected to be full at all times. Once elements are - * consumed, they are immediately produced. + * @param prod_idx + * @param p_prod_elem + */ +static inline void qed_chain_set_prod(struct qed_chain *p_chain, + u32 prod_idx, void *p_prod_elem) +{ + if (is_chain_u16(p_chain)) + p_chain->u.chain16.prod_idx = (u16) prod_idx; + else + p_chain->u.chain32.prod_idx = prod_idx; + p_chain->p_prod_elem = p_prod_elem; +} + +/** + * @brief qed_chain_pbl_zero_mem - set chain memory to 0 * * @param p_chain - * @param cnt - * - * @return inline void */ -static inline void -qed_chain_sge_inc_cons_prod(struct qed_chain *p_chain, - u16 cnt) +static inline void qed_chain_pbl_zero_mem(struct qed_chain *p_chain) { - p_chain->prod_idx += cnt; - p_chain->cons_idx += cnt; + u32 i, page_cnt; + + if (p_chain->mode != QED_CHAIN_MODE_PBL) + return; + + page_cnt = qed_chain_get_page_cnt(p_chain); + + for (i = 0; i < page_cnt; i++) + memset(p_chain->pbl.pp_virt_addr_tbl[i], 0, + QED_CHAIN_PAGE_SIZE); } #endif diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index 4c29439..15efccf 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -325,7 +325,8 @@ struct qed_common_ops { int (*chain_alloc)(struct qed_dev *cdev, enum qed_chain_use_mode intended_use, enum qed_chain_mode mode, - u16 num_elems, + enum qed_chain_cnt_type cnt_type, + u32 num_elems, size_t elem_size, struct qed_chain *p_chain); -- cgit v0.10.2 From 7a9b6b8f6e4c52b31830d32570c2a226e27651f9 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Fri, 3 Jun 2016 14:35:33 +0300 Subject: qed: Add common HSI for new protocols This adds the qed portion of the RoCE & iSCSI firmware HSI, as well as adding several new common HSI files which would be required by both qed and qed* protocols. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index d104ca7..4cd70bf 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -17,7 +17,12 @@ #include #include #include +#include +#include #include +#include +#include +#include struct qed_hwfn; struct qed_ptt; @@ -3631,6 +3636,3009 @@ struct vport_update_ramrod_data { struct eth_vport_rss_config rss_config; }; +struct mstorm_rdma_task_st_ctx { + struct regpair temp[4]; +}; + +struct rdma_close_func_ramrod_data { + u8 cnq_start_offset; + u8 num_cnqs; + u8 vf_id; + u8 vf_valid; + u8 reserved[4]; +}; + +struct rdma_cnq_params { + __le16 sb_num; + u8 sb_index; + u8 num_pbl_pages; + __le32 reserved; + struct regpair pbl_base_addr; + __le16 queue_zone_num; + u8 reserved1[6]; +}; + +struct rdma_create_cq_ramrod_data { + struct regpair cq_handle; + struct regpair pbl_addr; + __le32 max_cqes; + __le16 pbl_num_pages; + __le16 dpi; + u8 is_two_level_pbl; + u8 cnq_id; + u8 pbl_log_page_size; + u8 toggle_bit; + __le16 int_timeout; + __le16 reserved1; +}; + +struct rdma_deregister_tid_ramrod_data { + __le32 itid; + __le32 reserved; +}; + +struct rdma_destroy_cq_output_params { + __le16 cnq_num; + __le16 reserved0; + __le32 reserved1; +}; + +struct rdma_destroy_cq_ramrod_data { + struct regpair output_params_addr; +}; + +enum rdma_event_opcode { + RDMA_EVENT_UNUSED, + RDMA_EVENT_FUNC_INIT, + RDMA_EVENT_FUNC_CLOSE, + RDMA_EVENT_REGISTER_MR, + RDMA_EVENT_DEREGISTER_MR, + RDMA_EVENT_CREATE_CQ, + RDMA_EVENT_RESIZE_CQ, + RDMA_EVENT_DESTROY_CQ, + RDMA_EVENT_CREATE_SRQ, + RDMA_EVENT_MODIFY_SRQ, + RDMA_EVENT_DESTROY_SRQ, + MAX_RDMA_EVENT_OPCODE +}; + +enum rdma_fw_return_code { + RDMA_RETURN_OK = 0, + RDMA_RETURN_REGISTER_MR_BAD_STATE_ERR, + RDMA_RETURN_DEREGISTER_MR_BAD_STATE_ERR, + RDMA_RETURN_RESIZE_CQ_ERR, + RDMA_RETURN_NIG_DRAIN_REQ, + MAX_RDMA_FW_RETURN_CODE +}; + +struct rdma_init_func_hdr { + u8 cnq_start_offset; + u8 num_cnqs; + u8 cq_ring_mode; + u8 cnp_vlan_priority; + __le32 cnp_send_timeout; + u8 cnp_dscp; + u8 vf_id; + u8 vf_valid; + u8 reserved[5]; +}; + +struct rdma_init_func_ramrod_data { + struct rdma_init_func_hdr params_header; + struct rdma_cnq_params cnq_params[NUM_OF_GLOBAL_QUEUES]; +}; + +enum rdma_ramrod_cmd_id { + RDMA_RAMROD_UNUSED, + RDMA_RAMROD_FUNC_INIT, + RDMA_RAMROD_FUNC_CLOSE, + RDMA_RAMROD_REGISTER_MR, + RDMA_RAMROD_DEREGISTER_MR, + RDMA_RAMROD_CREATE_CQ, + RDMA_RAMROD_RESIZE_CQ, + RDMA_RAMROD_DESTROY_CQ, + RDMA_RAMROD_CREATE_SRQ, + RDMA_RAMROD_MODIFY_SRQ, + RDMA_RAMROD_DESTROY_SRQ, + MAX_RDMA_RAMROD_CMD_ID +}; + +struct rdma_register_tid_ramrod_data { + __le32 flags; +#define RDMA_REGISTER_TID_RAMROD_DATA_MAX_ID_MASK 0x3FFFF +#define RDMA_REGISTER_TID_RAMROD_DATA_MAX_ID_SHIFT 0 +#define RDMA_REGISTER_TID_RAMROD_DATA_PAGE_SIZE_LOG_MASK 0x1F +#define RDMA_REGISTER_TID_RAMROD_DATA_PAGE_SIZE_LOG_SHIFT 18 +#define RDMA_REGISTER_TID_RAMROD_DATA_TWO_LEVEL_PBL_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_TWO_LEVEL_PBL_SHIFT 23 +#define RDMA_REGISTER_TID_RAMROD_DATA_ZERO_BASED_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_ZERO_BASED_SHIFT 24 +#define RDMA_REGISTER_TID_RAMROD_DATA_PHY_MR_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_PHY_MR_SHIFT 25 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_READ_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_READ_SHIFT 26 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_WRITE_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_WRITE_SHIFT 27 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_ATOMIC_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_ATOMIC_SHIFT 28 +#define RDMA_REGISTER_TID_RAMROD_DATA_LOCAL_WRITE_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_LOCAL_WRITE_SHIFT 29 +#define RDMA_REGISTER_TID_RAMROD_DATA_LOCAL_READ_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_LOCAL_READ_SHIFT 30 +#define RDMA_REGISTER_TID_RAMROD_DATA_ENABLE_MW_BIND_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_ENABLE_MW_BIND_SHIFT 31 + u8 flags1; +#define RDMA_REGISTER_TID_RAMROD_DATA_PBL_PAGE_SIZE_LOG_MASK 0x1F +#define RDMA_REGISTER_TID_RAMROD_DATA_PBL_PAGE_SIZE_LOG_SHIFT 0 +#define RDMA_REGISTER_TID_RAMROD_DATA_TID_TYPE_MASK 0x7 +#define RDMA_REGISTER_TID_RAMROD_DATA_TID_TYPE_SHIFT 5 + u8 flags2; +#define RDMA_REGISTER_TID_RAMROD_DATA_DMA_MR_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_DMA_MR_SHIFT 0 +#define RDMA_REGISTER_TID_RAMROD_DATA_DIF_ON_HOST_FLG_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_DIF_ON_HOST_FLG_SHIFT 1 +#define RDMA_REGISTER_TID_RAMROD_DATA_RESERVED1_MASK 0x3F +#define RDMA_REGISTER_TID_RAMROD_DATA_RESERVED1_SHIFT 2 + u8 key; + u8 length_hi; + u8 vf_id; + u8 vf_valid; + __le16 pd; + __le32 length_lo; + __le32 itid; + __le32 reserved2; + struct regpair va; + struct regpair pbl_base; + struct regpair dif_error_addr; + struct regpair dif_runt_addr; + __le32 reserved3[2]; +}; + +struct rdma_resize_cq_output_params { + __le32 old_cq_cons; + __le32 old_cq_prod; +}; + +struct rdma_resize_cq_ramrod_data { + u8 flags; +#define RDMA_RESIZE_CQ_RAMROD_DATA_TOGGLE_BIT_MASK 0x1 +#define RDMA_RESIZE_CQ_RAMROD_DATA_TOGGLE_BIT_SHIFT 0 +#define RDMA_RESIZE_CQ_RAMROD_DATA_IS_TWO_LEVEL_PBL_MASK 0x1 +#define RDMA_RESIZE_CQ_RAMROD_DATA_IS_TWO_LEVEL_PBL_SHIFT 1 +#define RDMA_RESIZE_CQ_RAMROD_DATA_RESERVED_MASK 0x3F +#define RDMA_RESIZE_CQ_RAMROD_DATA_RESERVED_SHIFT 2 + u8 pbl_log_page_size; + __le16 pbl_num_pages; + __le32 max_cqes; + struct regpair pbl_addr; + struct regpair output_params_addr; +}; + +struct rdma_srq_context { + struct regpair temp[8]; +}; + +struct rdma_srq_create_ramrod_data { + struct regpair pbl_base_addr; + __le16 pages_in_srq_pbl; + __le16 pd_id; + struct rdma_srq_id srq_id; + __le16 page_size; + __le16 reserved1; + __le32 reserved2; + struct regpair producers_addr; +}; + +struct rdma_srq_destroy_ramrod_data { + struct rdma_srq_id srq_id; + __le32 reserved; +}; + +struct rdma_srq_modify_ramrod_data { + struct rdma_srq_id srq_id; + __le32 wqe_limit; +}; + +struct ystorm_rdma_task_st_ctx { + struct regpair temp[4]; +}; + +struct ystorm_rdma_task_ag_ctx { + u8 reserved; + u8 byte1; + __le16 msem_ctx_upd_seq; + u8 flags0; +#define YSTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_MASK 0xF +#define YSTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_SHIFT 0 +#define YSTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_SHIFT 4 +#define YSTORM_RDMA_TASK_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_BIT1_SHIFT 5 +#define YSTORM_RDMA_TASK_AG_CTX_VALID_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_VALID_SHIFT 6 +#define YSTORM_RDMA_TASK_AG_CTX_BIT3_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_BIT3_SHIFT 7 + u8 flags1; +#define YSTORM_RDMA_TASK_AG_CTX_CF0_MASK 0x3 +#define YSTORM_RDMA_TASK_AG_CTX_CF0_SHIFT 0 +#define YSTORM_RDMA_TASK_AG_CTX_CF1_MASK 0x3 +#define YSTORM_RDMA_TASK_AG_CTX_CF1_SHIFT 2 +#define YSTORM_RDMA_TASK_AG_CTX_CF2SPECIAL_MASK 0x3 +#define YSTORM_RDMA_TASK_AG_CTX_CF2SPECIAL_SHIFT 4 +#define YSTORM_RDMA_TASK_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_CF0EN_SHIFT 6 +#define YSTORM_RDMA_TASK_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_CF1EN_SHIFT 7 + u8 flags2; +#define YSTORM_RDMA_TASK_AG_CTX_BIT4_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_BIT4_SHIFT 0 +#define YSTORM_RDMA_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE0EN_SHIFT 1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE1EN_SHIFT 2 +#define YSTORM_RDMA_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE2EN_SHIFT 3 +#define YSTORM_RDMA_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE3EN_SHIFT 4 +#define YSTORM_RDMA_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE4EN_SHIFT 5 +#define YSTORM_RDMA_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE5EN_SHIFT 6 +#define YSTORM_RDMA_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE6EN_SHIFT 7 + u8 key; + __le32 mw_cnt; + u8 ref_cnt_seq; + u8 ctx_upd_seq; + __le16 dif_flags; + __le16 tx_ref_count; + __le16 last_used_ltid; + __le16 parent_mr_lo; + __le16 parent_mr_hi; + __le32 fbo_lo; + __le32 fbo_hi; +}; + +struct mstorm_rdma_task_ag_ctx { + u8 reserved; + u8 byte1; + __le16 icid; + u8 flags0; +#define MSTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_MASK 0xF +#define MSTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_SHIFT 0 +#define MSTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_SHIFT 4 +#define MSTORM_RDMA_TASK_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_BIT1_SHIFT 5 +#define MSTORM_RDMA_TASK_AG_CTX_BIT2_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_BIT2_SHIFT 6 +#define MSTORM_RDMA_TASK_AG_CTX_BIT3_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_BIT3_SHIFT 7 + u8 flags1; +#define MSTORM_RDMA_TASK_AG_CTX_CF0_MASK 0x3 +#define MSTORM_RDMA_TASK_AG_CTX_CF0_SHIFT 0 +#define MSTORM_RDMA_TASK_AG_CTX_CF1_MASK 0x3 +#define MSTORM_RDMA_TASK_AG_CTX_CF1_SHIFT 2 +#define MSTORM_RDMA_TASK_AG_CTX_CF2_MASK 0x3 +#define MSTORM_RDMA_TASK_AG_CTX_CF2_SHIFT 4 +#define MSTORM_RDMA_TASK_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_CF0EN_SHIFT 6 +#define MSTORM_RDMA_TASK_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_CF1EN_SHIFT 7 + u8 flags2; +#define MSTORM_RDMA_TASK_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_CF2EN_SHIFT 0 +#define MSTORM_RDMA_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE0EN_SHIFT 1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE1EN_SHIFT 2 +#define MSTORM_RDMA_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE2EN_SHIFT 3 +#define MSTORM_RDMA_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE3EN_SHIFT 4 +#define MSTORM_RDMA_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE4EN_SHIFT 5 +#define MSTORM_RDMA_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE5EN_SHIFT 6 +#define MSTORM_RDMA_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE6EN_SHIFT 7 + u8 key; + __le32 mw_cnt; + u8 ref_cnt_seq; + u8 ctx_upd_seq; + __le16 dif_flags; + __le16 tx_ref_count; + __le16 last_used_ltid; + __le16 parent_mr_lo; + __le16 parent_mr_hi; + __le32 fbo_lo; + __le32 fbo_hi; +}; + +struct ustorm_rdma_task_st_ctx { + struct regpair temp[2]; +}; + +struct ustorm_rdma_task_ag_ctx { + u8 reserved; + u8 byte1; + __le16 icid; + u8 flags0; +#define USTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_MASK 0xF +#define USTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_SHIFT 0 +#define USTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_SHIFT 4 +#define USTORM_RDMA_TASK_AG_CTX_DIF_RUNT_VALID_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_DIF_RUNT_VALID_SHIFT 5 +#define USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_RESULT_CF_MASK 0x3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_RESULT_CF_SHIFT 6 + u8 flags1; +#define USTORM_RDMA_TASK_AG_CTX_DIF_RESULT_TOGGLE_BIT_MASK 0x3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_RESULT_TOGGLE_BIT_SHIFT 0 +#define USTORM_RDMA_TASK_AG_CTX_DIF_TX_IO_FLG_MASK 0x3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_TX_IO_FLG_SHIFT 2 +#define USTORM_RDMA_TASK_AG_CTX_CF3_MASK 0x3 +#define USTORM_RDMA_TASK_AG_CTX_CF3_SHIFT 4 +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_CF_MASK 0x3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_CF_SHIFT 6 + u8 flags2; +#define USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_RESULT_CF_EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_RESULT_CF_EN_SHIFT 0 +#define USTORM_RDMA_TASK_AG_CTX_RESERVED2_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RESERVED2_SHIFT 1 +#define USTORM_RDMA_TASK_AG_CTX_RESERVED3_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RESERVED3_SHIFT 2 +#define USTORM_RDMA_TASK_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_CF_EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_CF_EN_SHIFT 4 +#define USTORM_RDMA_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE0EN_SHIFT 5 +#define USTORM_RDMA_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE1EN_SHIFT 6 +#define USTORM_RDMA_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE2EN_SHIFT 7 + u8 flags3; +#define USTORM_RDMA_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE3EN_SHIFT 0 +#define USTORM_RDMA_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE4EN_SHIFT 1 +#define USTORM_RDMA_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE5EN_SHIFT 2 +#define USTORM_RDMA_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE6EN_SHIFT 3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_TYPE_MASK 0xF +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_TYPE_SHIFT 4 + __le32 dif_err_intervals; + __le32 dif_error_1st_interval; + __le32 reg2; + __le32 dif_runt_value; + __le32 reg4; + __le32 reg5; +}; + +struct rdma_task_context { + struct ystorm_rdma_task_st_ctx ystorm_st_context; + struct ystorm_rdma_task_ag_ctx ystorm_ag_context; + struct tdif_task_context tdif_context; + struct mstorm_rdma_task_ag_ctx mstorm_ag_context; + struct mstorm_rdma_task_st_ctx mstorm_st_context; + struct rdif_task_context rdif_context; + struct ustorm_rdma_task_st_ctx ustorm_st_context; + struct regpair ustorm_st_padding[2]; + struct ustorm_rdma_task_ag_ctx ustorm_ag_context; +}; + +enum rdma_tid_type { + RDMA_TID_REGISTERED_MR, + RDMA_TID_FMR, + RDMA_TID_MW_TYPE1, + RDMA_TID_MW_TYPE2A, + MAX_RDMA_TID_TYPE +}; + +struct mstorm_rdma_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define MSTORM_RDMA_CONN_AG_CTX_BIT0_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_BIT0_SHIFT 0 +#define MSTORM_RDMA_CONN_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_BIT1_SHIFT 1 +#define MSTORM_RDMA_CONN_AG_CTX_CF0_MASK 0x3 +#define MSTORM_RDMA_CONN_AG_CTX_CF0_SHIFT 2 +#define MSTORM_RDMA_CONN_AG_CTX_CF1_MASK 0x3 +#define MSTORM_RDMA_CONN_AG_CTX_CF1_SHIFT 4 +#define MSTORM_RDMA_CONN_AG_CTX_CF2_MASK 0x3 +#define MSTORM_RDMA_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define MSTORM_RDMA_CONN_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_CF0EN_SHIFT 0 +#define MSTORM_RDMA_CONN_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_CF1EN_SHIFT 1 +#define MSTORM_RDMA_CONN_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_CF2EN_SHIFT 2 +#define MSTORM_RDMA_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define MSTORM_RDMA_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define MSTORM_RDMA_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define MSTORM_RDMA_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define MSTORM_RDMA_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_RULE4EN_SHIFT 7 + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; +}; + +struct tstorm_rdma_conn_ag_ctx { + u8 reserved0; + u8 byte1; + u8 flags0; +#define TSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT1_SHIFT 1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT2_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT3_SHIFT 3 +#define TSTORM_RDMA_CONN_AG_CTX_BIT4_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT4_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT5_SHIFT 5 +#define TSTORM_RDMA_CONN_AG_CTX_CF0_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF0_SHIFT 6 + u8 flags1; +#define TSTORM_RDMA_CONN_AG_CTX_CF1_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF1_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_CF2_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF2_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_TIMER_STOP_ALL_CF_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_TIMER_STOP_ALL_CF_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags2; +#define TSTORM_RDMA_CONN_AG_CTX_MSTORM_FLUSH_CF_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_MSTORM_FLUSH_CF_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_CF6_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF6_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_CF7_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF7_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_CF8_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF8_SHIFT 6 + u8 flags3; +#define TSTORM_RDMA_CONN_AG_CTX_CF9_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF9_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_CF10_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF10_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF0EN_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF1EN_SHIFT 5 +#define TSTORM_RDMA_CONN_AG_CTX_CF2EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF2EN_SHIFT 6 +#define TSTORM_RDMA_CONN_AG_CTX_TIMER_STOP_ALL_CF_EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_TIMER_STOP_ALL_CF_EN_SHIFT 7 + u8 flags4; +#define TSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_SHIFT 1 +#define TSTORM_RDMA_CONN_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF6EN_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF7EN_SHIFT 3 +#define TSTORM_RDMA_CONN_AG_CTX_CF8EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF8EN_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_CF9EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF9EN_SHIFT 5 +#define TSTORM_RDMA_CONN_AG_CTX_CF10EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF10EN_SHIFT 6 +#define TSTORM_RDMA_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags5; +#define TSTORM_RDMA_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define TSTORM_RDMA_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define TSTORM_RDMA_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define TSTORM_RDMA_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE8EN_SHIFT 7 + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le32 reg7; + __le32 reg8; + u8 byte2; + u8 byte3; + __le16 word0; + u8 byte4; + u8 byte5; + __le16 word1; + __le16 word2; + __le16 word3; + __le32 reg9; + __le32 reg10; +}; + +struct tstorm_rdma_task_ag_ctx { + u8 byte0; + u8 byte1; + __le16 word0; + u8 flags0; +#define TSTORM_RDMA_TASK_AG_CTX_NIBBLE0_MASK 0xF +#define TSTORM_RDMA_TASK_AG_CTX_NIBBLE0_SHIFT 0 +#define TSTORM_RDMA_TASK_AG_CTX_BIT0_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT0_SHIFT 4 +#define TSTORM_RDMA_TASK_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT1_SHIFT 5 +#define TSTORM_RDMA_TASK_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT2_SHIFT 6 +#define TSTORM_RDMA_TASK_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT3_SHIFT 7 + u8 flags1; +#define TSTORM_RDMA_TASK_AG_CTX_BIT4_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT4_SHIFT 0 +#define TSTORM_RDMA_TASK_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT5_SHIFT 1 +#define TSTORM_RDMA_TASK_AG_CTX_CF0_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF0_SHIFT 2 +#define TSTORM_RDMA_TASK_AG_CTX_CF1_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF1_SHIFT 4 +#define TSTORM_RDMA_TASK_AG_CTX_CF2_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF2_SHIFT 6 + u8 flags2; +#define TSTORM_RDMA_TASK_AG_CTX_CF3_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF3_SHIFT 0 +#define TSTORM_RDMA_TASK_AG_CTX_CF4_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF4_SHIFT 2 +#define TSTORM_RDMA_TASK_AG_CTX_CF5_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF5_SHIFT 4 +#define TSTORM_RDMA_TASK_AG_CTX_CF6_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF6_SHIFT 6 + u8 flags3; +#define TSTORM_RDMA_TASK_AG_CTX_CF7_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF7_SHIFT 0 +#define TSTORM_RDMA_TASK_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF0EN_SHIFT 2 +#define TSTORM_RDMA_TASK_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF1EN_SHIFT 3 +#define TSTORM_RDMA_TASK_AG_CTX_CF2EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF2EN_SHIFT 4 +#define TSTORM_RDMA_TASK_AG_CTX_CF3EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF3EN_SHIFT 5 +#define TSTORM_RDMA_TASK_AG_CTX_CF4EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF4EN_SHIFT 6 +#define TSTORM_RDMA_TASK_AG_CTX_CF5EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF5EN_SHIFT 7 + u8 flags4; +#define TSTORM_RDMA_TASK_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF6EN_SHIFT 0 +#define TSTORM_RDMA_TASK_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF7EN_SHIFT 1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE0EN_SHIFT 2 +#define TSTORM_RDMA_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE1EN_SHIFT 3 +#define TSTORM_RDMA_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE2EN_SHIFT 4 +#define TSTORM_RDMA_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE3EN_SHIFT 5 +#define TSTORM_RDMA_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE4EN_SHIFT 6 +#define TSTORM_RDMA_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE5EN_SHIFT 7 + u8 byte2; + __le16 word1; + __le32 reg0; + u8 byte3; + u8 byte4; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg1; + __le32 reg2; +}; + +struct ustorm_rdma_conn_ag_ctx { + u8 reserved; + u8 byte1; + u8 flags0; +#define USTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define USTORM_RDMA_CONN_AG_CTX_BIT1_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_BIT1_SHIFT 1 +#define USTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 2 +#define USTORM_RDMA_CONN_AG_CTX_CF1_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CF1_SHIFT 4 +#define USTORM_RDMA_CONN_AG_CTX_CF2_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define USTORM_RDMA_CONN_AG_CTX_CF3_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CF3_SHIFT 0 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_SE_CF_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_SE_CF_SHIFT 2 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_CF_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_CF_SHIFT 4 +#define USTORM_RDMA_CONN_AG_CTX_CF6_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CF6_SHIFT 6 + u8 flags2; +#define USTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 0 +#define USTORM_RDMA_CONN_AG_CTX_CF1EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CF1EN_SHIFT 1 +#define USTORM_RDMA_CONN_AG_CTX_CF2EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CF2EN_SHIFT 2 +#define USTORM_RDMA_CONN_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_SE_CF_EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_SE_CF_EN_SHIFT 4 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_CF_EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_CF_EN_SHIFT 5 +#define USTORM_RDMA_CONN_AG_CTX_CF6EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CF6EN_SHIFT 6 +#define USTORM_RDMA_CONN_AG_CTX_CQ_SE_EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CQ_SE_EN_SHIFT 7 + u8 flags3; +#define USTORM_RDMA_CONN_AG_CTX_CQ_EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CQ_EN_SHIFT 0 +#define USTORM_RDMA_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define USTORM_RDMA_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define USTORM_RDMA_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define USTORM_RDMA_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define USTORM_RDMA_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define USTORM_RDMA_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define USTORM_RDMA_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE8EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 conn_dpi; + __le16 word1; + __le32 cq_cons; + __le32 cq_se_prod; + __le32 cq_prod; + __le32 reg3; + __le16 int_timeout; + __le16 word3; +}; + +struct xstorm_roce_conn_ag_ctx_dq_ext_ld_part { + u8 reserved0; + u8 state; + u8 flags0; +#define XSTORMROCECONNAGCTXDQEXTLDPART_EXIST_IN_QM0_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_EXIST_IN_QM0_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT1_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT1_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT2_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT2_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_EXIST_IN_QM3_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_EXIST_IN_QM3_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT4_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT4_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT5_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT5_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT6_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT6_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT7_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT7_SHIFT 7 + u8 flags1; +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT8_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT8_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT9_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT9_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT10_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT10_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT11_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT11_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT12_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT12_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT13_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT13_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT14_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT14_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_YSTORM_FLUSH_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_YSTORM_FLUSH_SHIFT 7 + u8 flags2; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF0_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF0_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF1_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF1_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF2_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF2_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF3_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF3_SHIFT 6 + u8 flags3; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF4_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF4_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF5_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF5_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF6_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF6_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_FLUSH_Q0_CF_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_FLUSH_Q0_CF_SHIFT 6 + u8 flags4; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF8_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF8_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF9_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF9_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF10_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF10_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF11_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF11_SHIFT 6 + u8 flags5; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF12_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF12_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF13_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF13_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF14_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF14_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF15_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF15_SHIFT 6 + u8 flags6; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF16_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF16_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF17_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF17_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF18_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF18_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF19_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF19_SHIFT 6 + u8 flags7; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF20_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF20_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF21_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF21_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_SLOW_PATH_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_SLOW_PATH_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF0EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF0EN_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF1EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF1EN_SHIFT 7 + u8 flags8; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF2EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF2EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF3EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF3EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF4EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF4EN_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF5EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF5EN_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF6EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF6EN_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_FLUSH_Q0_CF_EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_FLUSH_Q0_CF_EN_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF8EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF8EN_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF9EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF9EN_SHIFT 7 + u8 flags9; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF10EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF10EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF11EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF11EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF12EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF12EN_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF13EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF13EN_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF14EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF14EN_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF15EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF15EN_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF16EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF16EN_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF17EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF17EN_SHIFT 7 + u8 flags10; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF18EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF18EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF19EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF19EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF20EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF20EN_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF21EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF21EN_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_SLOW_PATH_EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_SLOW_PATH_EN_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF23EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF23EN_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE0EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE0EN_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE1EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE1EN_SHIFT 7 + u8 flags11; +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE2EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE2EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE3EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE3EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE4EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE4EN_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE5EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE5EN_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE6EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE6EN_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE7EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE7EN_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED1_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED1_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE9EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE9EN_SHIFT 7 + u8 flags12; +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE10EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE10EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE11EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE11EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED2_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED2_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED3_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED3_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE14EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE14EN_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE15EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE15EN_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE16EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE16EN_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE17EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE17EN_SHIFT 7 + u8 flags13; +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE18EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE18EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE19EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE19EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED4_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED4_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED5_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED5_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED6_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED6_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED7_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED7_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED8_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED8_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED9_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED9_SHIFT 7 + u8 flags14; +#define XSTORMROCECONNAGCTXDQEXTLDPART_MIGRATION_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_MIGRATION_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT17_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT17_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_DPM_PORT_NUM_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_DPM_PORT_NUM_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RESERVED_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RESERVED_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_ROCE_EDPM_ENABLE_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_ROCE_EDPM_ENABLE_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF23_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF23_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le16 word5; + __le16 conn_dpi; + u8 byte3; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 snd_nxt_psn; + __le32 reg4; +}; + +struct xstorm_rdma_conn_ag_ctx { + u8 reserved0; + u8 state; + u8 flags0; +#define XSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_BIT1_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT1_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT2_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT2_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_BIT4_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT4_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_BIT5_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT5_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_BIT6_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT6_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_BIT7_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT7_SHIFT 7 + u8 flags1; +#define XSTORM_RDMA_CONN_AG_CTX_BIT8_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT8_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_BIT9_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT9_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT10_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT10_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_BIT11_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT11_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_BIT12_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT12_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_BIT13_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT13_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_BIT14_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT14_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_YSTORM_FLUSH_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_YSTORM_FLUSH_SHIFT 7 + u8 flags2; +#define XSTORM_RDMA_CONN_AG_CTX_CF0_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF0_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF1_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF1_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF2_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF2_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF3_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF3_SHIFT 6 + u8 flags3; +#define XSTORM_RDMA_CONN_AG_CTX_CF4_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF4_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF5_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF5_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF6_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF6_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags4; +#define XSTORM_RDMA_CONN_AG_CTX_CF8_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF8_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF9_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF9_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF10_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF10_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF11_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF11_SHIFT 6 + u8 flags5; +#define XSTORM_RDMA_CONN_AG_CTX_CF12_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF12_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF13_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF13_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF14_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF14_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF15_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF15_SHIFT 6 + u8 flags6; +#define XSTORM_RDMA_CONN_AG_CTX_CF16_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF16_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF17_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF17_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF18_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF18_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF19_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF19_SHIFT 6 + u8 flags7; +#define XSTORM_RDMA_CONN_AG_CTX_CF20_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF20_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF21_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF21_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_SLOW_PATH_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_SLOW_PATH_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF0EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF0EN_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_CF1EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF1EN_SHIFT 7 + u8 flags8; +#define XSTORM_RDMA_CONN_AG_CTX_CF2EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF2EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF3EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF3EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_CF4EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF4EN_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF5EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF5EN_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_CF6EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF6EN_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_CF8EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF8EN_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_CF9EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF9EN_SHIFT 7 + u8 flags9; +#define XSTORM_RDMA_CONN_AG_CTX_CF10EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF10EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF11EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF11EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_CF12EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF12EN_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF13EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF13EN_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_CF14EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF14EN_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF15EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF15EN_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_CF16EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF16EN_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_CF17EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF17EN_SHIFT 7 + u8 flags10; +#define XSTORM_RDMA_CONN_AG_CTX_CF18EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF18EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF19EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF19EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_CF20EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF20EN_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF21EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF21EN_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF23EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF23EN_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE0EN_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE1EN_SHIFT 7 + u8 flags11; +#define XSTORM_RDMA_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE2EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE3EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE4EN_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE5EN_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE6EN_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE7EN_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_RULE9EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE9EN_SHIFT 7 + u8 flags12; +#define XSTORM_RDMA_CONN_AG_CTX_RULE10EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE10EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_RULE11EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE11EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_RULE14EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE14EN_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_RULE15EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE15EN_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_RULE16EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE16EN_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_RULE17EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE17EN_SHIFT 7 + u8 flags13; +#define XSTORM_RDMA_CONN_AG_CTX_RULE18EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE18EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_RULE19EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE19EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 + u8 flags14; +#define XSTORM_RDMA_CONN_AG_CTX_MIGRATION_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_MIGRATION_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_BIT17_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT17_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_DPM_PORT_NUM_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_DPM_PORT_NUM_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_RESERVED_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RESERVED_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_ROCE_EDPM_ENABLE_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_ROCE_EDPM_ENABLE_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_CF23_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF23_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le16 word5; + __le16 conn_dpi; + u8 byte3; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 snd_nxt_psn; + __le32 reg4; + __le32 reg5; + __le32 reg6; +}; + +struct ystorm_rdma_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define YSTORM_RDMA_CONN_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_BIT0_SHIFT 0 +#define YSTORM_RDMA_CONN_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_BIT1_SHIFT 1 +#define YSTORM_RDMA_CONN_AG_CTX_CF0_MASK 0x3 +#define YSTORM_RDMA_CONN_AG_CTX_CF0_SHIFT 2 +#define YSTORM_RDMA_CONN_AG_CTX_CF1_MASK 0x3 +#define YSTORM_RDMA_CONN_AG_CTX_CF1_SHIFT 4 +#define YSTORM_RDMA_CONN_AG_CTX_CF2_MASK 0x3 +#define YSTORM_RDMA_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define YSTORM_RDMA_CONN_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_CF0EN_SHIFT 0 +#define YSTORM_RDMA_CONN_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_CF1EN_SHIFT 1 +#define YSTORM_RDMA_CONN_AG_CTX_CF2EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_CF2EN_SHIFT 2 +#define YSTORM_RDMA_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define YSTORM_RDMA_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define YSTORM_RDMA_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define YSTORM_RDMA_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define YSTORM_RDMA_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_RULE4EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le32 reg0; + __le32 reg1; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg2; + __le32 reg3; +}; + +struct mstorm_roce_conn_st_ctx { + struct regpair temp[6]; +}; + +struct pstorm_roce_conn_st_ctx { + struct regpair temp[16]; +}; + +struct ystorm_roce_conn_st_ctx { + struct regpair temp[2]; +}; + +struct xstorm_roce_conn_st_ctx { + struct regpair temp[22]; +}; + +struct tstorm_roce_conn_st_ctx { + struct regpair temp[30]; +}; + +struct ustorm_roce_conn_st_ctx { + struct regpair temp[12]; +}; + +struct roce_conn_context { + struct ystorm_roce_conn_st_ctx ystorm_st_context; + struct regpair ystorm_st_padding[2]; + struct pstorm_roce_conn_st_ctx pstorm_st_context; + struct xstorm_roce_conn_st_ctx xstorm_st_context; + struct regpair xstorm_st_padding[2]; + struct xstorm_rdma_conn_ag_ctx xstorm_ag_context; + struct tstorm_rdma_conn_ag_ctx tstorm_ag_context; + struct timers_context timer_context; + struct ustorm_rdma_conn_ag_ctx ustorm_ag_context; + struct tstorm_roce_conn_st_ctx tstorm_st_context; + struct mstorm_roce_conn_st_ctx mstorm_st_context; + struct ustorm_roce_conn_st_ctx ustorm_st_context; + struct regpair ustorm_st_padding[2]; +}; + +struct roce_create_qp_req_ramrod_data { + __le16 flags; +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_ROCE_FLAVOR_MASK 0x3 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_ROCE_FLAVOR_SHIFT 0 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_FMR_AND_RESERVED_EN_MASK 0x1 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_FMR_AND_RESERVED_EN_SHIFT 2 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_SIGNALED_COMP_MASK 0x1 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_SIGNALED_COMP_SHIFT 3 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_PRI_MASK 0x7 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_PRI_SHIFT 4 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RESERVED_MASK 0x1 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RESERVED_SHIFT 7 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_MASK 0xF +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_SHIFT 8 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_MASK 0xF +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_SHIFT 12 + u8 max_ord; + u8 traffic_class; + u8 hop_limit; + u8 orq_num_pages; + __le16 p_key; + __le32 flow_label; + __le32 dst_qp_id; + __le32 ack_timeout_val; + __le32 initial_psn; + __le16 mtu; + __le16 pd; + __le16 sq_num_pages; + __le16 reseved2; + struct regpair sq_pbl_addr; + struct regpair orq_pbl_addr; + __le16 local_mac_addr[3]; + __le16 remote_mac_addr[3]; + __le16 vlan_id; + __le16 udp_src_port; + __le32 src_gid[4]; + __le32 dst_gid[4]; + struct regpair qp_handle_for_cqe; + struct regpair qp_handle_for_async; + u8 stats_counter_id; + u8 reserved3[7]; + __le32 cq_cid; + __le16 physical_queue0; + __le16 dpi; +}; + +struct roce_create_qp_resp_ramrod_data { + __le16 flags; +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_ROCE_FLAVOR_MASK 0x3 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_ROCE_FLAVOR_SHIFT 0 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RDMA_RD_EN_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RDMA_RD_EN_SHIFT 2 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RDMA_WR_EN_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RDMA_WR_EN_SHIFT 3 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_ATOMIC_EN_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_ATOMIC_EN_SHIFT 4 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_SRQ_FLG_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_SRQ_FLG_SHIFT 5 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_E2E_FLOW_CONTROL_EN_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_E2E_FLOW_CONTROL_EN_SHIFT 6 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RESERVED0_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RESERVED0_SHIFT 7 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_PRI_MASK 0x7 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_PRI_SHIFT 8 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_MASK 0x1F +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_SHIFT 11 + u8 max_ird; + u8 traffic_class; + u8 hop_limit; + u8 irq_num_pages; + __le16 p_key; + __le32 flow_label; + __le32 dst_qp_id; + u8 stats_counter_id; + u8 reserved1; + __le16 mtu; + __le32 initial_psn; + __le16 pd; + __le16 rq_num_pages; + struct rdma_srq_id srq_id; + struct regpair rq_pbl_addr; + struct regpair irq_pbl_addr; + __le16 local_mac_addr[3]; + __le16 remote_mac_addr[3]; + __le16 vlan_id; + __le16 udp_src_port; + __le32 src_gid[4]; + __le32 dst_gid[4]; + struct regpair qp_handle_for_cqe; + struct regpair qp_handle_for_async; + __le32 reserved2[2]; + __le32 cq_cid; + __le16 physical_queue0; + __le16 dpi; +}; + +struct roce_destroy_qp_req_output_params { + __le32 num_bound_mw; + __le32 reserved; +}; + +struct roce_destroy_qp_req_ramrod_data { + struct regpair output_params_addr; +}; + +struct roce_destroy_qp_resp_output_params { + __le32 num_invalidated_mw; + __le32 reserved; +}; + +struct roce_destroy_qp_resp_ramrod_data { + struct regpair output_params_addr; +}; + +enum roce_event_opcode { + ROCE_EVENT_CREATE_QP = 11, + ROCE_EVENT_MODIFY_QP, + ROCE_EVENT_QUERY_QP, + ROCE_EVENT_DESTROY_QP, + MAX_ROCE_EVENT_OPCODE +}; + +struct roce_modify_qp_req_ramrod_data { + __le16 flags; +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MOVE_TO_ERR_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MOVE_TO_ERR_FLG_SHIFT 0 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MOVE_TO_SQD_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MOVE_TO_SQD_FLG_SHIFT 1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_EN_SQD_ASYNC_NOTIFY_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_EN_SQD_ASYNC_NOTIFY_SHIFT 2 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_P_KEY_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_P_KEY_FLG_SHIFT 3 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ADDRESS_VECTOR_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ADDRESS_VECTOR_FLG_SHIFT 4 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MAX_ORD_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MAX_ORD_FLG_SHIFT 5 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_FLG_SHIFT 6 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_FLG_SHIFT 7 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ACK_TIMEOUT_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ACK_TIMEOUT_FLG_SHIFT 8 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_FLG_SHIFT 9 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_MASK 0x7 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_SHIFT 10 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RESERVED1_MASK 0x7 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RESERVED1_SHIFT 13 + u8 fields; +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_MASK 0xF +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_SHIFT 0 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_MASK 0xF +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_SHIFT 4 + u8 max_ord; + u8 traffic_class; + u8 hop_limit; + __le16 p_key; + __le32 flow_label; + __le32 ack_timeout_val; + __le16 mtu; + __le16 reserved2; + __le32 reserved3[3]; + __le32 src_gid[4]; + __le32 dst_gid[4]; +}; + +struct roce_modify_qp_resp_ramrod_data { + __le16 flags; +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MOVE_TO_ERR_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MOVE_TO_ERR_FLG_SHIFT 0 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_RD_EN_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_RD_EN_SHIFT 1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_WR_EN_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_WR_EN_SHIFT 2 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_ATOMIC_EN_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_ATOMIC_EN_SHIFT 3 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_P_KEY_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_P_KEY_FLG_SHIFT 4 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_ADDRESS_VECTOR_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_ADDRESS_VECTOR_FLG_SHIFT 5 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MAX_IRD_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MAX_IRD_FLG_SHIFT 6 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PRI_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PRI_FLG_SHIFT 7 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_FLG_SHIFT 8 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_OPS_EN_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_OPS_EN_FLG_SHIFT 9 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RESERVED1_MASK 0x3F +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RESERVED1_SHIFT 10 + u8 fields; +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PRI_MASK 0x7 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PRI_SHIFT 0 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_MASK 0x1F +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_SHIFT 3 + u8 max_ird; + u8 traffic_class; + u8 hop_limit; + __le16 p_key; + __le32 flow_label; + __le16 mtu; + __le16 reserved2; + __le32 src_gid[4]; + __le32 dst_gid[4]; +}; + +struct roce_query_qp_req_output_params { + __le32 psn; + __le32 flags; +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_ERR_FLG_MASK 0x1 +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_ERR_FLG_SHIFT 0 +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_SQ_DRAINING_FLG_MASK 0x1 +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_SQ_DRAINING_FLG_SHIFT 1 +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_RESERVED0_MASK 0x3FFFFFFF +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_RESERVED0_SHIFT 2 +}; + +struct roce_query_qp_req_ramrod_data { + struct regpair output_params_addr; +}; + +struct roce_query_qp_resp_output_params { + __le32 psn; + __le32 err_flag; +#define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_ERROR_FLG_MASK 0x1 +#define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_ERROR_FLG_SHIFT 0 +#define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_RESERVED0_MASK 0x7FFFFFFF +#define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_RESERVED0_SHIFT 1 +}; + +struct roce_query_qp_resp_ramrod_data { + struct regpair output_params_addr; +}; + +enum roce_ramrod_cmd_id { + ROCE_RAMROD_CREATE_QP = 11, + ROCE_RAMROD_MODIFY_QP, + ROCE_RAMROD_QUERY_QP, + ROCE_RAMROD_DESTROY_QP, + MAX_ROCE_RAMROD_CMD_ID +}; + +struct mstorm_roce_req_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define MSTORM_ROCE_REQ_CONN_AG_CTX_BIT0_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_BIT0_SHIFT 0 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_BIT1_SHIFT 1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF0_MASK 0x3 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF0_SHIFT 2 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF1_MASK 0x3 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF1_SHIFT 4 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF2_MASK 0x3 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_SHIFT 0 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_SHIFT 1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_SHIFT 2 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_SHIFT 7 + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; +}; + +struct mstorm_roce_resp_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define MSTORM_ROCE_RESP_CONN_AG_CTX_BIT0_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_BIT0_SHIFT 0 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_SHIFT 1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF0_MASK 0x3 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF0_SHIFT 2 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF1_MASK 0x3 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF1_SHIFT 4 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF2_MASK 0x3 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_SHIFT 0 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_SHIFT 1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_SHIFT 2 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_SHIFT 7 + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; +}; + +enum roce_flavor { + PLAIN_ROCE /* RoCE v1 */ , + RROCE_IPV4 /* RoCE v2 (Routable RoCE) over ipv4 */ , + RROCE_IPV6 /* RoCE v2 (Routable RoCE) over ipv6 */ , + MAX_ROCE_FLAVOR +}; + +struct tstorm_roce_req_conn_ag_ctx { + u8 reserved0; + u8 state; + u8 flags0; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_OCCURED_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_OCCURED_SHIFT 1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_CQE_ERROR_OCCURED_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_CQE_ERROR_OCCURED_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_BIT3_SHIFT 3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CACHED_ORQ_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CACHED_ORQ_SHIFT 5 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_CF_SHIFT 6 + u8 flags1; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CF1_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CF1_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_SQ_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_SQ_CF_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_STOP_ALL_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_STOP_ALL_CF_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags2; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_CF_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SET_TIMER_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SET_TIMER_CF_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_ASYNC_ERROR_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_ASYNC_ERROR_CF_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RXMIT_DONE_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RXMIT_DONE_CF_SHIFT 6 + u8 flags3; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_SCAN_COMPLETED_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_SCAN_COMPLETED_CF_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SQ_DRAIN_COMPLETED_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SQ_DRAIN_COMPLETED_CF_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_CF_EN_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_SHIFT 5 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_SQ_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_SQ_CF_EN_SHIFT 6 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_STOP_ALL_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_STOP_ALL_CF_EN_SHIFT 7 + u8 flags4; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_SHIFT 1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SET_TIMER_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SET_TIMER_CF_EN_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_ASYNC_ERROR_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_ASYNC_ERROR_CF_EN_SHIFT 3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RXMIT_DONE_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RXMIT_DONE_CF_EN_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_SCAN_COMPLETED_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_SCAN_COMPLETED_CF_EN_SHIFT 5 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SQ_DRAIN_COMPLETED_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SQ_DRAIN_COMPLETED_CF_EN_SHIFT 6 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags5; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SND_SQ_CONS_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SND_SQ_CONS_EN_SHIFT 5 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE8EN_SHIFT 7 + __le32 reg0; + __le32 snd_nxt_psn; + __le32 snd_max_psn; + __le32 orq_prod; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le32 reg7; + __le32 reg8; + u8 tx_cqe_error_type; + u8 orq_cache_idx; + __le16 snd_sq_cons_th; + u8 byte4; + u8 byte5; + __le16 snd_sq_cons; + __le16 word2; + __le16 word3; + __le32 reg9; + __le32 reg10; +}; + +struct tstorm_roce_resp_conn_ag_ctx { + u8 byte0; + u8 state; + u8 flags0; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_SHIFT 1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT2_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT3_SHIFT 3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT5_SHIFT 5 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF0_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF0_SHIFT 6 + u8 flags1; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_TX_ERROR_CF_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_TX_ERROR_CF_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF3_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF3_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags2; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_CF_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_CF_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF6_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF6_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF7_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF7_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF8_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF8_SHIFT 6 + u8 flags3; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF9_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF9_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF10_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF10_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_EN_SHIFT 5 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_TX_ERROR_CF_EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_TX_ERROR_CF_EN_SHIFT 6 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_SHIFT 7 + u8 flags4; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_SHIFT 1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF6EN_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF7EN_SHIFT 3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF8EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF8EN_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF9EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF9EN_SHIFT 5 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF10EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF10EN_SHIFT 6 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags5; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RQ_RULE_EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RQ_RULE_EN_SHIFT 5 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE8EN_SHIFT 7 + __le32 psn_and_rxmit_id_echo; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le32 reg7; + __le32 reg8; + u8 tx_async_error_type; + u8 byte3; + __le16 rq_cons; + u8 byte4; + u8 byte5; + __le16 rq_prod; + __le16 conn_dpi; + __le16 irq_cons; + __le32 num_invlidated_mw; + __le32 reg10; +}; + +struct ustorm_roce_req_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define USTORM_ROCE_REQ_CONN_AG_CTX_BIT0_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_BIT0_SHIFT 0 +#define USTORM_ROCE_REQ_CONN_AG_CTX_BIT1_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_BIT1_SHIFT 1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF0_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF0_SHIFT 2 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF1_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF1_SHIFT 4 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF2_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF3_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF3_SHIFT 0 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF4_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF4_SHIFT 2 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF5_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF5_SHIFT 4 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF6_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF6_SHIFT 6 + u8 flags2; +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_SHIFT 0 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_SHIFT 1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_SHIFT 2 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF4EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF4EN_SHIFT 4 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF5EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF5EN_SHIFT 5 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF6EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF6EN_SHIFT 6 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags3; +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE8EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le16 word2; + __le16 word3; +}; + +struct ustorm_roce_resp_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define USTORM_ROCE_RESP_CONN_AG_CTX_BIT0_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_BIT0_SHIFT 0 +#define USTORM_ROCE_RESP_CONN_AG_CTX_BIT1_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_BIT1_SHIFT 1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF0_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF0_SHIFT 2 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF1_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF1_SHIFT 4 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF2_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF3_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF3_SHIFT 0 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF4_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF4_SHIFT 2 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF5_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF5_SHIFT 4 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF6_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF6_SHIFT 6 + u8 flags2; +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_SHIFT 0 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_SHIFT 1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_SHIFT 2 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF4EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF4EN_SHIFT 4 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF5EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF5EN_SHIFT 5 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF6EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF6EN_SHIFT 6 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags3; +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE8EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le16 word2; + __le16 word3; +}; + +struct xstorm_roce_req_conn_ag_ctx { + u8 reserved0; + u8 state; + u8 flags0; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED1_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED1_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED2_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED2_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED3_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED3_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED4_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED4_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED5_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED5_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED6_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED6_SHIFT 7 + u8 flags1; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED7_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED7_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED8_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED8_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT10_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT10_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT11_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT11_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT12_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT12_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT13_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT13_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_STATE_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_STATE_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_YSTORM_FLUSH_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_YSTORM_FLUSH_SHIFT 7 + u8 flags2; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF0_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF0_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF1_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF1_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF2_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF2_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF3_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF3_SHIFT 6 + u8 flags3; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_FLUSH_CF_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_FLUSH_CF_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_CF_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_CF_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SND_RXMIT_CF_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SND_RXMIT_CF_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags4; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF8_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF8_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF9_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF9_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF10_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF10_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF11_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF11_SHIFT 6 + u8 flags5; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF12_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF12_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF13_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF13_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FMR_ENDED_CF_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FMR_ENDED_CF_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF15_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF15_SHIFT 6 + u8 flags6; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF16_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF16_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF17_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF17_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF18_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF18_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF19_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF19_SHIFT 6 + u8 flags7; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF20_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF20_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF21_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF21_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SLOW_PATH_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SLOW_PATH_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_SHIFT 7 + u8 flags8; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF3EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF3EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_FLUSH_CF_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_FLUSH_CF_EN_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_CF_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_CF_EN_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SND_RXMIT_CF_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SND_RXMIT_CF_EN_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF8EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF8EN_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF9EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF9EN_SHIFT 7 + u8 flags9; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF10EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF10EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF11EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF11EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF12EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF12EN_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF13EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF13EN_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FME_ENDED_CF_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FME_ENDED_CF_EN_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF15EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF15EN_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF16EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF16EN_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF17EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF17EN_SHIFT 7 + u8 flags10; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF18EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF18EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF19EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF19EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF20EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF20EN_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF21EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF21EN_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF23EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF23EN_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT 7 + u8 flags11; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE6EN_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_E2E_CREDIT_RULE_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_E2E_CREDIT_RULE_EN_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE9EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE9EN_SHIFT 7 + u8 flags12; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_PROD_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_PROD_EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE11EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE11EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_INV_FENCE_RULE_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_INV_FENCE_RULE_EN_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE15EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE15EN_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ORQ_FENCE_RULE_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ORQ_FENCE_RULE_EN_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_MAX_ORD_RULE_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_MAX_ORD_RULE_EN_SHIFT 7 + u8 flags13; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE18EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE18EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE19EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE19EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 + u8 flags14; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_MIGRATION_FLAG_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_MIGRATION_FLAG_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT17_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT17_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_DPM_PORT_NUM_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_DPM_PORT_NUM_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ROCE_EDPM_ENABLE_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ROCE_EDPM_ENABLE_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF23_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF23_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 word1; + __le16 sq_cmp_cons; + __le16 sq_cons; + __le16 sq_prod; + __le16 word5; + __le16 conn_dpi; + u8 byte3; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 lsn; + __le32 ssn; + __le32 snd_una_psn; + __le32 snd_nxt_psn; + __le32 reg4; + __le32 orq_cons_th; + __le32 orq_cons; +}; + +struct xstorm_roce_resp_conn_ag_ctx { + u8 reserved0; + u8 state; + u8 flags0; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED1_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED1_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED2_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED2_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED3_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED3_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED4_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED4_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED5_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED5_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED6_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED6_SHIFT 7 + u8 flags1; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED7_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED7_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED8_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED8_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT10_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT10_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT11_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT11_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT12_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT12_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT13_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT13_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_ERROR_STATE_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_ERROR_STATE_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_YSTORM_FLUSH_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_YSTORM_FLUSH_SHIFT 7 + u8 flags2; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF0_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF0_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF1_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF1_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF2_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF2_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF3_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF3_SHIFT 6 + u8 flags3; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RXMIT_CF_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RXMIT_CF_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FORCE_ACK_CF_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FORCE_ACK_CF_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags4; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF8_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF8_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF9_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF9_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF10_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF10_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF11_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF11_SHIFT 6 + u8 flags5; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF12_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF12_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF13_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF13_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF14_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF14_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF15_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF15_SHIFT 6 + u8 flags6; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF16_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF16_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF17_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF17_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF18_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF18_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF19_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF19_SHIFT 6 + u8 flags7; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF20_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF20_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF21_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF21_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_SLOW_PATH_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_SLOW_PATH_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_SHIFT 7 + u8 flags8; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RXMIT_CF_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RXMIT_CF_EN_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_EN_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FORCE_ACK_CF_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FORCE_ACK_CF_EN_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF8EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF8EN_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF9EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF9EN_SHIFT 7 + u8 flags9; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF10EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF10EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF11EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF11EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF12EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF12EN_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF13EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF13EN_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF14EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF14EN_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF15EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF15EN_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF16EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF16EN_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF17EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF17EN_SHIFT 7 + u8 flags10; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF18EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF18EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF19EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF19EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF20EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF20EN_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF21EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF21EN_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF23EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF23EN_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_SHIFT 7 + u8 flags11; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE6EN_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE9EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE9EN_SHIFT 7 + u8 flags12; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE10EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE10EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_IRQ_PROD_RULE_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_IRQ_PROD_RULE_EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE14EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE14EN_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE15EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE15EN_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE16EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE16EN_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE17EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE17EN_SHIFT 7 + u8 flags13; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE18EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE18EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE19EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE19EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 + u8 flags14; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT16_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT16_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT17_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT17_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT18_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT18_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT19_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT19_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT20_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT20_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT21_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT21_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF23_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF23_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 word1; + __le16 irq_prod; + __le16 word3; + __le16 word4; + __le16 word5; + __le16 irq_cons; + u8 rxmit_opcode; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 rxmit_psn_and_id; + __le32 rxmit_bytes_length; + __le32 psn; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 msn_and_syndrome; +}; + +struct ystorm_roce_req_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define YSTORM_ROCE_REQ_CONN_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_BIT0_SHIFT 0 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_BIT1_SHIFT 1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF0_MASK 0x3 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF0_SHIFT 2 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF1_MASK 0x3 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF1_SHIFT 4 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF2_MASK 0x3 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_SHIFT 0 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_SHIFT 1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_SHIFT 2 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le32 reg0; + __le32 reg1; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg2; + __le32 reg3; +}; + +struct ystorm_roce_resp_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define YSTORM_ROCE_RESP_CONN_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_BIT0_SHIFT 0 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_SHIFT 1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF0_MASK 0x3 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF0_SHIFT 2 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF1_MASK 0x3 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF1_SHIFT 4 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF2_MASK 0x3 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_SHIFT 0 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_SHIFT 1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_SHIFT 2 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le32 reg0; + __le32 reg1; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg2; + __le32 reg3; +}; + +struct ystorm_iscsi_conn_st_ctx { + __le32 reserved[4]; +}; + +struct pstorm_iscsi_tcp_conn_st_ctx { + __le32 tcp[32]; + __le32 iscsi[4]; +}; + +struct xstorm_iscsi_tcp_conn_st_ctx { + __le32 reserved_iscsi[40]; + __le32 reserved_tcp[4]; +}; + +struct xstorm_iscsi_conn_ag_ctx { + u8 cdu_validation; + u8 state; + u8 flags0; +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM1_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM1_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED1_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED1_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT4_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT4_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED2_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED2_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT6_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT6_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT7_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT7_SHIFT 7 + u8 flags1; +#define XSTORM_ISCSI_CONN_AG_CTX_BIT8_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT8_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT9_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT9_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT10_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT10_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT11_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT11_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT12_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT12_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT13_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT13_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT14_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT14_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_TX_TRUNCATE_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_TX_TRUNCATE_SHIFT 7 + u8 flags2; +#define XSTORM_ISCSI_CONN_AG_CTX_CF0_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF0_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF1_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF1_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF2_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF2_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_SHIFT 6 + u8 flags3; +#define XSTORM_ISCSI_CONN_AG_CTX_CF4_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF4_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF5_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF5_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF6_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF6_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_CF7_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF7_SHIFT 6 + u8 flags4; +#define XSTORM_ISCSI_CONN_AG_CTX_CF8_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF8_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF9_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF9_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF10_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF10_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_CF11_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF11_SHIFT 6 + u8 flags5; +#define XSTORM_ISCSI_CONN_AG_CTX_CF12_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF12_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF13_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF13_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF14_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF14_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_UPDATE_STATE_TO_BASE_CF_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_UPDATE_STATE_TO_BASE_CF_SHIFT 6 + u8 flags6; +#define XSTORM_ISCSI_CONN_AG_CTX_CF16_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF16_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF17_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF17_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF18_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF18_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_DQ_FLUSH_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_DQ_FLUSH_SHIFT 6 + u8 flags7; +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q1_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q1_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_SLOW_PATH_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_SLOW_PATH_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_CF1EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF1EN_SHIFT 7 + u8 flags8; +#define XSTORM_ISCSI_CONN_AG_CTX_CF2EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF2EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF4EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF4EN_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF5EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF5EN_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF6EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF6EN_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_CF7EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF7EN_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_CF8EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF8EN_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_CF9EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF9EN_SHIFT 7 + u8 flags9; +#define XSTORM_ISCSI_CONN_AG_CTX_CF10EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF10EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF11EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF11EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF12EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF12EN_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF13EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF13EN_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF14EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF14EN_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_UPDATE_STATE_TO_BASE_CF_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_UPDATE_STATE_TO_BASE_CF_EN_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_CF16EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF16EN_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_CF17EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF17EN_SHIFT 7 + u8 flags10; +#define XSTORM_ISCSI_CONN_AG_CTX_CF18EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF18EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_DQ_FLUSH_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_DQ_FLUSH_EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_EN_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q1_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q1_EN_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_PROC_ONLY_CLEANUP_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_PROC_ONLY_CLEANUP_EN_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_MORE_TO_SEND_DEC_RULE_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_MORE_TO_SEND_DEC_RULE_EN_SHIFT 7 + u8 flags11; +#define XSTORM_ISCSI_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE2EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE3EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED3_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED3_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE5EN_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE6EN_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE7EN_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE9EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE9EN_SHIFT 7 + u8 flags12; +#define XSTORM_ISCSI_CONN_AG_CTX_SQ_DEC_RULE_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_SQ_DEC_RULE_EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE11EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE11EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE14EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE14EN_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE15EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE15EN_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE16EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE16EN_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE17EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE17EN_SHIFT 7 + u8 flags13; +#define XSTORM_ISCSI_CONN_AG_CTX_R2TQ_DEC_RULE_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_R2TQ_DEC_RULE_EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_HQ_DEC_RULE_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_HQ_DEC_RULE_EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 + u8 flags14; +#define XSTORM_ISCSI_CONN_AG_CTX_BIT16_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT16_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT17_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT17_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT18_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT18_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT19_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT19_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT20_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT20_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_DUMMY_READ_DONE_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_DUMMY_READ_DONE_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_PROC_ONLY_CLEANUP_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_PROC_ONLY_CLEANUP_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 physical_q1; + __le16 dummy_dorq_var; + __le16 sq_cons; + __le16 sq_prod; + __le16 word5; + __le16 slow_io_total_data_tx_update; + u8 byte3; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 more_to_send_seq; + __le32 reg4; + __le32 reg5; + __le32 hq_scan_next_relevant_ack; + __le16 r2tq_prod; + __le16 r2tq_cons; + __le16 hq_prod; + __le16 hq_cons; + __le32 remain_seq; + __le32 bytes_to_next_pdu; + __le32 hq_tcp_seq; + u8 byte7; + u8 byte8; + u8 byte9; + u8 byte10; + u8 byte11; + u8 byte12; + u8 byte13; + u8 byte14; + u8 byte15; + u8 byte16; + __le16 word11; + __le32 reg10; + __le32 reg11; + __le32 exp_stat_sn; + __le32 reg13; + __le32 reg14; + __le32 reg15; + __le32 reg16; + __le32 reg17; +}; + +struct tstorm_iscsi_conn_ag_ctx { + u8 reserved0; + u8 state; + u8 flags0; +#define TSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT1_SHIFT 1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT2_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT3_SHIFT 3 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT4_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT4_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT5_SHIFT 5 +#define TSTORM_ISCSI_CONN_AG_CTX_CF0_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF0_SHIFT 6 + u8 flags1; +#define TSTORM_ISCSI_CONN_AG_CTX_CF1_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF1_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_CF2_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF2_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_CF4_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF4_SHIFT 6 + u8 flags2; +#define TSTORM_ISCSI_CONN_AG_CTX_CF5_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF5_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_CF6_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF6_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_CF7_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF7_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_CF8_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF8_SHIFT 6 + u8 flags3; +#define TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_CF10_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF10_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF1EN_SHIFT 5 +#define TSTORM_ISCSI_CONN_AG_CTX_CF2EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF2EN_SHIFT 6 +#define TSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_EN_SHIFT 7 + u8 flags4; +#define TSTORM_ISCSI_CONN_AG_CTX_CF4EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF4EN_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_CF5EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF5EN_SHIFT 1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF6EN_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF7EN_SHIFT 3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF8EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF8EN_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_EN_SHIFT 5 +#define TSTORM_ISCSI_CONN_AG_CTX_CF10EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF10EN_SHIFT 6 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags5; +#define TSTORM_ISCSI_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE8EN_SHIFT 7 + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le32 reg7; + __le32 reg8; + u8 byte2; + u8 byte3; + __le16 word0; +}; + +struct ustorm_iscsi_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define USTORM_ISCSI_CONN_AG_CTX_BIT0_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_BIT0_SHIFT 0 +#define USTORM_ISCSI_CONN_AG_CTX_BIT1_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_BIT1_SHIFT 1 +#define USTORM_ISCSI_CONN_AG_CTX_CF0_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF0_SHIFT 2 +#define USTORM_ISCSI_CONN_AG_CTX_CF1_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF1_SHIFT 4 +#define USTORM_ISCSI_CONN_AG_CTX_CF2_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define USTORM_ISCSI_CONN_AG_CTX_CF3_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF3_SHIFT 0 +#define USTORM_ISCSI_CONN_AG_CTX_CF4_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF4_SHIFT 2 +#define USTORM_ISCSI_CONN_AG_CTX_CF5_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF5_SHIFT 4 +#define USTORM_ISCSI_CONN_AG_CTX_CF6_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF6_SHIFT 6 + u8 flags2; +#define USTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT 0 +#define USTORM_ISCSI_CONN_AG_CTX_CF1EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF1EN_SHIFT 1 +#define USTORM_ISCSI_CONN_AG_CTX_CF2EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF2EN_SHIFT 2 +#define USTORM_ISCSI_CONN_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_ISCSI_CONN_AG_CTX_CF4EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF4EN_SHIFT 4 +#define USTORM_ISCSI_CONN_AG_CTX_CF5EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF5EN_SHIFT 5 +#define USTORM_ISCSI_CONN_AG_CTX_CF6EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF6EN_SHIFT 6 +#define USTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags3; +#define USTORM_ISCSI_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define USTORM_ISCSI_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define USTORM_ISCSI_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define USTORM_ISCSI_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define USTORM_ISCSI_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define USTORM_ISCSI_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define USTORM_ISCSI_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE8EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le16 word2; + __le16 word3; +}; + +struct tstorm_iscsi_conn_st_ctx { + __le32 reserved[40]; +}; + +struct mstorm_iscsi_conn_ag_ctx { + u8 reserved; + u8 state; + u8 flags0; +#define MSTORM_ISCSI_CONN_AG_CTX_BIT0_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_BIT0_SHIFT 0 +#define MSTORM_ISCSI_CONN_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_BIT1_SHIFT 1 +#define MSTORM_ISCSI_CONN_AG_CTX_CF0_MASK 0x3 +#define MSTORM_ISCSI_CONN_AG_CTX_CF0_SHIFT 2 +#define MSTORM_ISCSI_CONN_AG_CTX_CF1_MASK 0x3 +#define MSTORM_ISCSI_CONN_AG_CTX_CF1_SHIFT 4 +#define MSTORM_ISCSI_CONN_AG_CTX_CF2_MASK 0x3 +#define MSTORM_ISCSI_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define MSTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT 0 +#define MSTORM_ISCSI_CONN_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_CF1EN_SHIFT 1 +#define MSTORM_ISCSI_CONN_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_CF2EN_SHIFT 2 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE4EN_SHIFT 7 + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; +}; + +struct mstorm_iscsi_tcp_conn_st_ctx { + __le32 reserved_tcp[20]; + __le32 reserved_iscsi[8]; +}; + +struct ustorm_iscsi_conn_st_ctx { + __le32 reserved[52]; +}; + +struct iscsi_conn_context { + struct ystorm_iscsi_conn_st_ctx ystorm_st_context; + struct regpair ystorm_st_padding[2]; + struct pstorm_iscsi_tcp_conn_st_ctx pstorm_st_context; + struct regpair pstorm_st_padding[2]; + struct pb_context xpb2_context; + struct xstorm_iscsi_tcp_conn_st_ctx xstorm_st_context; + struct regpair xstorm_st_padding[2]; + struct xstorm_iscsi_conn_ag_ctx xstorm_ag_context; + struct tstorm_iscsi_conn_ag_ctx tstorm_ag_context; + struct regpair tstorm_ag_padding[2]; + struct timers_context timer_context; + struct ustorm_iscsi_conn_ag_ctx ustorm_ag_context; + struct pb_context upb_context; + struct tstorm_iscsi_conn_st_ctx tstorm_st_context; + struct regpair tstorm_st_padding[2]; + struct mstorm_iscsi_conn_ag_ctx mstorm_ag_context; + struct mstorm_iscsi_tcp_conn_st_ctx mstorm_st_context; + struct ustorm_iscsi_conn_st_ctx ustorm_st_context; +}; + +struct iscsi_init_ramrod_params { + struct iscsi_spe_func_init iscsi_init_spe; + struct tcp_init_params tcp_init; +}; + +struct ystorm_iscsi_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define YSTORM_ISCSI_CONN_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_BIT0_SHIFT 0 +#define YSTORM_ISCSI_CONN_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_BIT1_SHIFT 1 +#define YSTORM_ISCSI_CONN_AG_CTX_CF0_MASK 0x3 +#define YSTORM_ISCSI_CONN_AG_CTX_CF0_SHIFT 2 +#define YSTORM_ISCSI_CONN_AG_CTX_CF1_MASK 0x3 +#define YSTORM_ISCSI_CONN_AG_CTX_CF1_SHIFT 4 +#define YSTORM_ISCSI_CONN_AG_CTX_CF2_MASK 0x3 +#define YSTORM_ISCSI_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define YSTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT 0 +#define YSTORM_ISCSI_CONN_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_CF1EN_SHIFT 1 +#define YSTORM_ISCSI_CONN_AG_CTX_CF2EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_CF2EN_SHIFT 2 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE4EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le32 reg0; + __le32 reg1; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg2; + __le32 reg3; +}; #define VF_MAX_STATIC 192 #define MCP_GLOB_PATH_MAX 2 diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp.h b/drivers/net/ethernet/qlogic/qed/qed_sp.h index ea4e9ce..a548504 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_sp.h @@ -63,6 +63,32 @@ union ramrod_data { struct vport_update_ramrod_data vport_update; struct vport_filter_update_ramrod_data vport_filter_update; + struct rdma_init_func_ramrod_data rdma_init_func; + struct rdma_close_func_ramrod_data rdma_close_func; + struct rdma_register_tid_ramrod_data rdma_register_tid; + struct rdma_deregister_tid_ramrod_data rdma_deregister_tid; + struct roce_create_qp_resp_ramrod_data roce_create_qp_resp; + struct roce_create_qp_req_ramrod_data roce_create_qp_req; + struct roce_modify_qp_resp_ramrod_data roce_modify_qp_resp; + struct roce_modify_qp_req_ramrod_data roce_modify_qp_req; + struct roce_query_qp_resp_ramrod_data roce_query_qp_resp; + struct roce_query_qp_req_ramrod_data roce_query_qp_req; + struct roce_destroy_qp_resp_ramrod_data roce_destroy_qp_resp; + struct roce_destroy_qp_req_ramrod_data roce_destroy_qp_req; + struct rdma_create_cq_ramrod_data rdma_create_cq; + struct rdma_resize_cq_ramrod_data rdma_resize_cq; + struct rdma_destroy_cq_ramrod_data rdma_destroy_cq; + struct rdma_srq_create_ramrod_data rdma_create_srq; + struct rdma_srq_destroy_ramrod_data rdma_destroy_srq; + struct rdma_srq_modify_ramrod_data rdma_modify_srq; + + struct iscsi_slow_path_hdr iscsi_empty; + struct iscsi_init_ramrod_params iscsi_init; + struct iscsi_spe_func_dstry iscsi_destroy; + struct iscsi_spe_conn_offload iscsi_conn_offload; + struct iscsi_conn_update_ramrod_params iscsi_conn_update; + struct iscsi_spe_conn_termination iscsi_conn_terminate; + struct vf_start_ramrod_data vf_start; struct vf_stop_ramrod_data vf_stop; }; diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h index 285189a..077a3b6 100644 --- a/include/linux/qed/common_hsi.h +++ b/include/linux/qed/common_hsi.h @@ -12,6 +12,7 @@ #define CORE_SPQE_PAGE_SIZE_BYTES 4096 #define X_FINAL_CLEANUP_AGG_INT 1 +#define NUM_OF_GLOBAL_QUEUES 128 /* Queue Zone sizes in bytes */ #define TSTORM_QZONE_SIZE 8 @@ -694,7 +695,10 @@ struct parsing_and_err_flags { #define PARSING_AND_ERR_FLAGS_TUNNELL4CHKSMERROR_SHIFT 15 }; -/* Concrete Function ID. */ +struct pb_context { + __le32 crc[4]; +}; + struct pxp_concrete_fid { __le16 fid; #define PXP_CONCRETE_FID_PFID_MASK 0xF @@ -761,6 +765,72 @@ struct pxp_ptt_entry { }; /* RSS hash type */ +struct rdif_task_context { + __le32 initial_ref_tag; + __le16 app_tag_value; + __le16 app_tag_mask; + u8 flags0; +#define RDIF_TASK_CONTEXT_IGNOREAPPTAG_MASK 0x1 +#define RDIF_TASK_CONTEXT_IGNOREAPPTAG_SHIFT 0 +#define RDIF_TASK_CONTEXT_INITIALREFTAGVALID_MASK 0x1 +#define RDIF_TASK_CONTEXT_INITIALREFTAGVALID_SHIFT 1 +#define RDIF_TASK_CONTEXT_HOSTGUARDTYPE_MASK 0x1 +#define RDIF_TASK_CONTEXT_HOSTGUARDTYPE_SHIFT 2 +#define RDIF_TASK_CONTEXT_SETERRORWITHEOP_MASK 0x1 +#define RDIF_TASK_CONTEXT_SETERRORWITHEOP_SHIFT 3 +#define RDIF_TASK_CONTEXT_PROTECTIONTYPE_MASK 0x3 +#define RDIF_TASK_CONTEXT_PROTECTIONTYPE_SHIFT 4 +#define RDIF_TASK_CONTEXT_CRC_SEED_MASK 0x1 +#define RDIF_TASK_CONTEXT_CRC_SEED_SHIFT 6 +#define RDIF_TASK_CONTEXT_KEEPREFTAGCONST_MASK 0x1 +#define RDIF_TASK_CONTEXT_KEEPREFTAGCONST_SHIFT 7 + u8 partial_dif_data[7]; + __le16 partial_crc_value; + __le16 partial_checksum_value; + __le32 offset_in_io; + __le16 flags1; +#define RDIF_TASK_CONTEXT_VALIDATEGUARD_MASK 0x1 +#define RDIF_TASK_CONTEXT_VALIDATEGUARD_SHIFT 0 +#define RDIF_TASK_CONTEXT_VALIDATEAPPTAG_MASK 0x1 +#define RDIF_TASK_CONTEXT_VALIDATEAPPTAG_SHIFT 1 +#define RDIF_TASK_CONTEXT_VALIDATEREFTAG_MASK 0x1 +#define RDIF_TASK_CONTEXT_VALIDATEREFTAG_SHIFT 2 +#define RDIF_TASK_CONTEXT_FORWARDGUARD_MASK 0x1 +#define RDIF_TASK_CONTEXT_FORWARDGUARD_SHIFT 3 +#define RDIF_TASK_CONTEXT_FORWARDAPPTAG_MASK 0x1 +#define RDIF_TASK_CONTEXT_FORWARDAPPTAG_SHIFT 4 +#define RDIF_TASK_CONTEXT_FORWARDREFTAG_MASK 0x1 +#define RDIF_TASK_CONTEXT_FORWARDREFTAG_SHIFT 5 +#define RDIF_TASK_CONTEXT_INTERVALSIZE_MASK 0x7 +#define RDIF_TASK_CONTEXT_INTERVALSIZE_SHIFT 6 +#define RDIF_TASK_CONTEXT_HOSTINTERFACE_MASK 0x3 +#define RDIF_TASK_CONTEXT_HOSTINTERFACE_SHIFT 9 +#define RDIF_TASK_CONTEXT_DIFBEFOREDATA_MASK 0x1 +#define RDIF_TASK_CONTEXT_DIFBEFOREDATA_SHIFT 11 +#define RDIF_TASK_CONTEXT_RESERVED0_MASK 0x1 +#define RDIF_TASK_CONTEXT_RESERVED0_SHIFT 12 +#define RDIF_TASK_CONTEXT_NETWORKINTERFACE_MASK 0x1 +#define RDIF_TASK_CONTEXT_NETWORKINTERFACE_SHIFT 13 +#define RDIF_TASK_CONTEXT_FORWARDAPPTAGWITHMASK_MASK 0x1 +#define RDIF_TASK_CONTEXT_FORWARDAPPTAGWITHMASK_SHIFT 14 +#define RDIF_TASK_CONTEXT_FORWARDREFTAGWITHMASK_MASK 0x1 +#define RDIF_TASK_CONTEXT_FORWARDREFTAGWITHMASK_SHIFT 15 + __le16 state; +#define RDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFT_MASK 0xF +#define RDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFT_SHIFT 0 +#define RDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFT_MASK 0xF +#define RDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFT_SHIFT 4 +#define RDIF_TASK_CONTEXT_ERRORINIO_MASK 0x1 +#define RDIF_TASK_CONTEXT_ERRORINIO_SHIFT 8 +#define RDIF_TASK_CONTEXT_CHECKSUMOVERFLOW_MASK 0x1 +#define RDIF_TASK_CONTEXT_CHECKSUMOVERFLOW_SHIFT 9 +#define RDIF_TASK_CONTEXT_REFTAGMASK_MASK 0xF +#define RDIF_TASK_CONTEXT_REFTAGMASK_SHIFT 10 +#define RDIF_TASK_CONTEXT_RESERVED1_MASK 0x3 +#define RDIF_TASK_CONTEXT_RESERVED1_SHIFT 14 + __le32 reserved2; +}; + enum rss_hash_type { RSS_HASH_TYPE_DEFAULT = 0, RSS_HASH_TYPE_IPV4 = 1, @@ -789,4 +859,122 @@ struct status_block { #define STATUS_BLOCK_ZERO_PAD3_SHIFT 24 }; +struct tdif_task_context { + __le32 initial_ref_tag; + __le16 app_tag_value; + __le16 app_tag_mask; + __le16 partial_crc_valueB; + __le16 partial_checksum_valueB; + __le16 stateB; +#define TDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFTB_MASK 0xF +#define TDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFTB_SHIFT 0 +#define TDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFTB_MASK 0xF +#define TDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFTB_SHIFT 4 +#define TDIF_TASK_CONTEXT_ERRORINIOB_MASK 0x1 +#define TDIF_TASK_CONTEXT_ERRORINIOB_SHIFT 8 +#define TDIF_TASK_CONTEXT_CHECKSUMOVERFLOW_MASK 0x1 +#define TDIF_TASK_CONTEXT_CHECKSUMOVERFLOW_SHIFT 9 +#define TDIF_TASK_CONTEXT_RESERVED0_MASK 0x3F +#define TDIF_TASK_CONTEXT_RESERVED0_SHIFT 10 + u8 reserved1; + u8 flags0; +#define TDIF_TASK_CONTEXT_IGNOREAPPTAG_MASK 0x1 +#define TDIF_TASK_CONTEXT_IGNOREAPPTAG_SHIFT 0 +#define TDIF_TASK_CONTEXT_INITIALREFTAGVALID_MASK 0x1 +#define TDIF_TASK_CONTEXT_INITIALREFTAGVALID_SHIFT 1 +#define TDIF_TASK_CONTEXT_HOSTGUARDTYPE_MASK 0x1 +#define TDIF_TASK_CONTEXT_HOSTGUARDTYPE_SHIFT 2 +#define TDIF_TASK_CONTEXT_SETERRORWITHEOP_MASK 0x1 +#define TDIF_TASK_CONTEXT_SETERRORWITHEOP_SHIFT 3 +#define TDIF_TASK_CONTEXT_PROTECTIONTYPE_MASK 0x3 +#define TDIF_TASK_CONTEXT_PROTECTIONTYPE_SHIFT 4 +#define TDIF_TASK_CONTEXT_CRC_SEED_MASK 0x1 +#define TDIF_TASK_CONTEXT_CRC_SEED_SHIFT 6 +#define TDIF_TASK_CONTEXT_RESERVED2_MASK 0x1 +#define TDIF_TASK_CONTEXT_RESERVED2_SHIFT 7 + __le32 flags1; +#define TDIF_TASK_CONTEXT_VALIDATEGUARD_MASK 0x1 +#define TDIF_TASK_CONTEXT_VALIDATEGUARD_SHIFT 0 +#define TDIF_TASK_CONTEXT_VALIDATEAPPTAG_MASK 0x1 +#define TDIF_TASK_CONTEXT_VALIDATEAPPTAG_SHIFT 1 +#define TDIF_TASK_CONTEXT_VALIDATEREFTAG_MASK 0x1 +#define TDIF_TASK_CONTEXT_VALIDATEREFTAG_SHIFT 2 +#define TDIF_TASK_CONTEXT_FORWARDGUARD_MASK 0x1 +#define TDIF_TASK_CONTEXT_FORWARDGUARD_SHIFT 3 +#define TDIF_TASK_CONTEXT_FORWARDAPPTAG_MASK 0x1 +#define TDIF_TASK_CONTEXT_FORWARDAPPTAG_SHIFT 4 +#define TDIF_TASK_CONTEXT_FORWARDREFTAG_MASK 0x1 +#define TDIF_TASK_CONTEXT_FORWARDREFTAG_SHIFT 5 +#define TDIF_TASK_CONTEXT_INTERVALSIZE_MASK 0x7 +#define TDIF_TASK_CONTEXT_INTERVALSIZE_SHIFT 6 +#define TDIF_TASK_CONTEXT_HOSTINTERFACE_MASK 0x3 +#define TDIF_TASK_CONTEXT_HOSTINTERFACE_SHIFT 9 +#define TDIF_TASK_CONTEXT_DIFBEFOREDATA_MASK 0x1 +#define TDIF_TASK_CONTEXT_DIFBEFOREDATA_SHIFT 11 +#define TDIF_TASK_CONTEXT_RESERVED3_MASK 0x1 +#define TDIF_TASK_CONTEXT_RESERVED3_SHIFT 12 +#define TDIF_TASK_CONTEXT_NETWORKINTERFACE_MASK 0x1 +#define TDIF_TASK_CONTEXT_NETWORKINTERFACE_SHIFT 13 +#define TDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFTA_MASK 0xF +#define TDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFTA_SHIFT 14 +#define TDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFTA_MASK 0xF +#define TDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFTA_SHIFT 18 +#define TDIF_TASK_CONTEXT_ERRORINIOA_MASK 0x1 +#define TDIF_TASK_CONTEXT_ERRORINIOA_SHIFT 22 +#define TDIF_TASK_CONTEXT_CHECKSUMOVERFLOWA_MASK 0x1 +#define TDIF_TASK_CONTEXT_CHECKSUMOVERFLOWA_SHIFT 23 +#define TDIF_TASK_CONTEXT_REFTAGMASK_MASK 0xF +#define TDIF_TASK_CONTEXT_REFTAGMASK_SHIFT 24 +#define TDIF_TASK_CONTEXT_FORWARDAPPTAGWITHMASK_MASK 0x1 +#define TDIF_TASK_CONTEXT_FORWARDAPPTAGWITHMASK_SHIFT 28 +#define TDIF_TASK_CONTEXT_FORWARDREFTAGWITHMASK_MASK 0x1 +#define TDIF_TASK_CONTEXT_FORWARDREFTAGWITHMASK_SHIFT 29 +#define TDIF_TASK_CONTEXT_KEEPREFTAGCONST_MASK 0x1 +#define TDIF_TASK_CONTEXT_KEEPREFTAGCONST_SHIFT 30 +#define TDIF_TASK_CONTEXT_RESERVED4_MASK 0x1 +#define TDIF_TASK_CONTEXT_RESERVED4_SHIFT 31 + __le32 offset_in_iob; + __le16 partial_crc_value_a; + __le16 partial_checksum_valuea_; + __le32 offset_in_ioa; + u8 partial_dif_data_a[8]; + u8 partial_dif_data_b[8]; +}; + +struct timers_context { + __le32 logical_client0; +#define TIMERS_CONTEXT_EXPIRATIONTIMELC0_MASK 0xFFFFFFF +#define TIMERS_CONTEXT_EXPIRATIONTIMELC0_SHIFT 0 +#define TIMERS_CONTEXT_VALIDLC0_MASK 0x1 +#define TIMERS_CONTEXT_VALIDLC0_SHIFT 28 +#define TIMERS_CONTEXT_ACTIVELC0_MASK 0x1 +#define TIMERS_CONTEXT_ACTIVELC0_SHIFT 29 +#define TIMERS_CONTEXT_RESERVED0_MASK 0x3 +#define TIMERS_CONTEXT_RESERVED0_SHIFT 30 + __le32 logical_client1; +#define TIMERS_CONTEXT_EXPIRATIONTIMELC1_MASK 0xFFFFFFF +#define TIMERS_CONTEXT_EXPIRATIONTIMELC1_SHIFT 0 +#define TIMERS_CONTEXT_VALIDLC1_MASK 0x1 +#define TIMERS_CONTEXT_VALIDLC1_SHIFT 28 +#define TIMERS_CONTEXT_ACTIVELC1_MASK 0x1 +#define TIMERS_CONTEXT_ACTIVELC1_SHIFT 29 +#define TIMERS_CONTEXT_RESERVED1_MASK 0x3 +#define TIMERS_CONTEXT_RESERVED1_SHIFT 30 + __le32 logical_client2; +#define TIMERS_CONTEXT_EXPIRATIONTIMELC2_MASK 0xFFFFFFF +#define TIMERS_CONTEXT_EXPIRATIONTIMELC2_SHIFT 0 +#define TIMERS_CONTEXT_VALIDLC2_MASK 0x1 +#define TIMERS_CONTEXT_VALIDLC2_SHIFT 28 +#define TIMERS_CONTEXT_ACTIVELC2_MASK 0x1 +#define TIMERS_CONTEXT_ACTIVELC2_SHIFT 29 +#define TIMERS_CONTEXT_RESERVED2_MASK 0x3 +#define TIMERS_CONTEXT_RESERVED2_SHIFT 30 + __le32 host_expiration_fields; +#define TIMERS_CONTEXT_HOSTEXPRIRATIONVALUE_MASK 0xFFFFFFF +#define TIMERS_CONTEXT_HOSTEXPRIRATIONVALUE_SHIFT 0 +#define TIMERS_CONTEXT_HOSTEXPRIRATIONVALID_MASK 0x1 +#define TIMERS_CONTEXT_HOSTEXPRIRATIONVALID_SHIFT 28 +#define TIMERS_CONTEXT_RESERVED3_MASK 0x7 +#define TIMERS_CONTEXT_RESERVED3_SHIFT 29 +}; #endif /* __COMMON_HSI__ */ diff --git a/include/linux/qed/iscsi_common.h b/include/linux/qed/iscsi_common.h new file mode 100644 index 0000000..b3c0feb --- /dev/null +++ b/include/linux/qed/iscsi_common.h @@ -0,0 +1,1439 @@ +/* QLogic qed NIC Driver + * Copyright (c) 2015 QLogic Corporation + * + * This software is available under the terms of the GNU General Public License + * (GPL) Version 2, available from the file COPYING in the main directory of + * this source tree. + */ + +#ifndef __ISCSI_COMMON__ +#define __ISCSI_COMMON__ +/**********************/ +/* ISCSI FW CONSTANTS */ +/**********************/ + +/* iSCSI HSI constants */ +#define ISCSI_DEFAULT_MTU (1500) + +/* Current iSCSI HSI version number composed of two fields (16 bit) */ +#define ISCSI_HSI_MAJOR_VERSION (0) +#define ISCSI_HSI_MINOR_VERSION (0) + +/* KWQ (kernel work queue) layer codes */ +#define ISCSI_SLOW_PATH_LAYER_CODE (6) + +/* CQE completion status */ +#define ISCSI_EQE_COMPLETION_SUCCESS (0x0) +#define ISCSI_EQE_RST_CONN_RCVD (0x1) + +/* iSCSI parameter defaults */ +#define ISCSI_DEFAULT_HEADER_DIGEST (0) +#define ISCSI_DEFAULT_DATA_DIGEST (0) +#define ISCSI_DEFAULT_INITIAL_R2T (1) +#define ISCSI_DEFAULT_IMMEDIATE_DATA (1) +#define ISCSI_DEFAULT_MAX_PDU_LENGTH (0x2000) +#define ISCSI_DEFAULT_FIRST_BURST_LENGTH (0x10000) +#define ISCSI_DEFAULT_MAX_BURST_LENGTH (0x40000) +#define ISCSI_DEFAULT_MAX_OUTSTANDING_R2T (1) + +/* iSCSI parameter limits */ +#define ISCSI_MIN_VAL_MAX_PDU_LENGTH (0x200) +#define ISCSI_MAX_VAL_MAX_PDU_LENGTH (0xffffff) +#define ISCSI_MIN_VAL_BURST_LENGTH (0x200) +#define ISCSI_MAX_VAL_BURST_LENGTH (0xffffff) +#define ISCSI_MIN_VAL_MAX_OUTSTANDING_R2T (1) +#define ISCSI_MAX_VAL_MAX_OUTSTANDING_R2T (0xff) + +/* iSCSI reserved params */ +#define ISCSI_ITT_ALL_ONES (0xffffffff) +#define ISCSI_TTT_ALL_ONES (0xffffffff) + +#define ISCSI_OPTION_1_OFF_CHIP_TCP 1 +#define ISCSI_OPTION_2_ON_CHIP_TCP 2 + +#define ISCSI_INITIATOR_MODE 0 +#define ISCSI_TARGET_MODE 1 + +/* iSCSI request op codes */ +#define ISCSI_OPCODE_NOP_OUT_NO_IMM (0) +#define ISCSI_OPCODE_NOP_OUT ( \ + ISCSI_OPCODE_NOP_OUT_NO_IMM | 0x40) +#define ISCSI_OPCODE_SCSI_CMD_NO_IMM (1) +#define ISCSI_OPCODE_SCSI_CMD ( \ + ISCSI_OPCODE_SCSI_CMD_NO_IMM | 0x40) +#define ISCSI_OPCODE_TMF_REQUEST_NO_IMM (2) +#define ISCSI_OPCODE_TMF_REQUEST ( \ + ISCSI_OPCODE_TMF_REQUEST_NO_IMM | 0x40) +#define ISCSI_OPCODE_LOGIN_REQUEST_NO_IMM (3) +#define ISCSI_OPCODE_LOGIN_REQUEST ( \ + ISCSI_OPCODE_LOGIN_REQUEST_NO_IMM | 0x40) +#define ISCSI_OPCODE_TEXT_REQUEST_NO_IMM (4) +#define ISCSI_OPCODE_TEXT_REQUEST ( \ + ISCSI_OPCODE_TEXT_REQUEST_NO_IMM | 0x40) +#define ISCSI_OPCODE_DATA_OUT (5) +#define ISCSI_OPCODE_LOGOUT_REQUEST_NO_IMM (6) +#define ISCSI_OPCODE_LOGOUT_REQUEST ( \ + ISCSI_OPCODE_LOGOUT_REQUEST_NO_IMM | 0x40) + +/* iSCSI response/messages op codes */ +#define ISCSI_OPCODE_NOP_IN (0x20) +#define ISCSI_OPCODE_SCSI_RESPONSE (0x21) +#define ISCSI_OPCODE_TMF_RESPONSE (0x22) +#define ISCSI_OPCODE_LOGIN_RESPONSE (0x23) +#define ISCSI_OPCODE_TEXT_RESPONSE (0x24) +#define ISCSI_OPCODE_DATA_IN (0x25) +#define ISCSI_OPCODE_LOGOUT_RESPONSE (0x26) +#define ISCSI_OPCODE_R2T (0x31) +#define ISCSI_OPCODE_ASYNC_MSG (0x32) +#define ISCSI_OPCODE_REJECT (0x3f) + +/* iSCSI stages */ +#define ISCSI_STAGE_SECURITY_NEGOTIATION (0) +#define ISCSI_STAGE_LOGIN_OPERATIONAL_NEGOTIATION (1) +#define ISCSI_STAGE_FULL_FEATURE_PHASE (3) + +/* iSCSI CQE errors */ +#define CQE_ERROR_BITMAP_DATA_DIGEST (0x08) +#define CQE_ERROR_BITMAP_RCV_ON_INVALID_CONN (0x10) +#define CQE_ERROR_BITMAP_DATA_TRUNCATED (0x20) + +struct cqe_error_bitmap { + u8 cqe_error_status_bits; +#define CQE_ERROR_BITMAP_DIF_ERR_BITS_MASK 0x7 +#define CQE_ERROR_BITMAP_DIF_ERR_BITS_SHIFT 0 +#define CQE_ERROR_BITMAP_DATA_DIGEST_ERR_MASK 0x1 +#define CQE_ERROR_BITMAP_DATA_DIGEST_ERR_SHIFT 3 +#define CQE_ERROR_BITMAP_RCV_ON_INVALID_CONN_MASK 0x1 +#define CQE_ERROR_BITMAP_RCV_ON_INVALID_CONN_SHIFT 4 +#define CQE_ERROR_BITMAP_DATA_TRUNCATED_ERR_MASK 0x1 +#define CQE_ERROR_BITMAP_DATA_TRUNCATED_ERR_SHIFT 5 +#define CQE_ERROR_BITMAP_UNDER_RUN_ERR_MASK 0x1 +#define CQE_ERROR_BITMAP_UNDER_RUN_ERR_SHIFT 6 +#define CQE_ERROR_BITMAP_RESERVED2_MASK 0x1 +#define CQE_ERROR_BITMAP_RESERVED2_SHIFT 7 +}; + +union cqe_error_status { + u8 error_status; + struct cqe_error_bitmap error_bits; +}; + +struct data_hdr { + __le32 data[12]; +}; + +struct iscsi_async_msg_hdr { + __le16 reserved0; + u8 flags_attr; +#define ISCSI_ASYNC_MSG_HDR_RSRV_MASK 0x7F +#define ISCSI_ASYNC_MSG_HDR_RSRV_SHIFT 0 +#define ISCSI_ASYNC_MSG_HDR_CONST1_MASK 0x1 +#define ISCSI_ASYNC_MSG_HDR_CONST1_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_ASYNC_MSG_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_ASYNC_MSG_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_ASYNC_MSG_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_ASYNC_MSG_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 all_ones; + __le32 reserved1; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le16 param1_rsrv; + u8 async_vcode; + u8 async_event; + __le16 param3_rsrv; + __le16 param2_rsrv; + __le32 reserved7; +}; + +struct iscsi_sge { + struct regpair sge_addr; + __le16 sge_len; + __le16 reserved0; + __le32 reserved1; +}; + +struct iscsi_cached_sge_ctx { + struct iscsi_sge sge; + struct regpair reserved; + __le32 dsgl_curr_offset[2]; +}; + +struct iscsi_cmd_hdr { + __le16 reserved1; + u8 flags_attr; +#define ISCSI_CMD_HDR_ATTR_MASK 0x7 +#define ISCSI_CMD_HDR_ATTR_SHIFT 0 +#define ISCSI_CMD_HDR_RSRV_MASK 0x3 +#define ISCSI_CMD_HDR_RSRV_SHIFT 3 +#define ISCSI_CMD_HDR_WRITE_MASK 0x1 +#define ISCSI_CMD_HDR_WRITE_SHIFT 5 +#define ISCSI_CMD_HDR_READ_MASK 0x1 +#define ISCSI_CMD_HDR_READ_SHIFT 6 +#define ISCSI_CMD_HDR_FINAL_MASK 0x1 +#define ISCSI_CMD_HDR_FINAL_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_CMD_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_CMD_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_CMD_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_CMD_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 expected_transfer_length; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 cdb[4]; +}; + +struct iscsi_common_hdr { + u8 hdr_status; + u8 hdr_response; + u8 hdr_flags; + u8 hdr_first_byte; +#define ISCSI_COMMON_HDR_OPCODE_MASK 0x3F +#define ISCSI_COMMON_HDR_OPCODE_SHIFT 0 +#define ISCSI_COMMON_HDR_IMM_MASK 0x1 +#define ISCSI_COMMON_HDR_IMM_SHIFT 6 +#define ISCSI_COMMON_HDR_RSRV_MASK 0x1 +#define ISCSI_COMMON_HDR_RSRV_SHIFT 7 + __le32 hdr_second_dword; +#define ISCSI_COMMON_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_COMMON_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_COMMON_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_COMMON_HDR_TOTAL_AHS_LEN_SHIFT 24 + __le32 lun_reserved[4]; + __le32 data[6]; +}; + +struct iscsi_conn_offload_params { + struct regpair sq_pbl_addr; + struct regpair r2tq_pbl_addr; + struct regpair xhq_pbl_addr; + struct regpair uhq_pbl_addr; + __le32 initial_ack; + __le16 physical_q0; + __le16 physical_q1; + u8 flags; +#define ISCSI_CONN_OFFLOAD_PARAMS_TCP_ON_CHIP_1B_MASK 0x1 +#define ISCSI_CONN_OFFLOAD_PARAMS_TCP_ON_CHIP_1B_SHIFT 0 +#define ISCSI_CONN_OFFLOAD_PARAMS_TARGET_MODE_MASK 0x1 +#define ISCSI_CONN_OFFLOAD_PARAMS_TARGET_MODE_SHIFT 1 +#define ISCSI_CONN_OFFLOAD_PARAMS_RESERVED1_MASK 0x3F +#define ISCSI_CONN_OFFLOAD_PARAMS_RESERVED1_SHIFT 2 + u8 pbl_page_size_log; + u8 pbe_page_size_log; + u8 default_cq; + __le32 stat_sn; +}; + +struct iscsi_slow_path_hdr { + u8 op_code; + u8 flags; +#define ISCSI_SLOW_PATH_HDR_RESERVED0_MASK 0xF +#define ISCSI_SLOW_PATH_HDR_RESERVED0_SHIFT 0 +#define ISCSI_SLOW_PATH_HDR_LAYER_CODE_MASK 0x7 +#define ISCSI_SLOW_PATH_HDR_LAYER_CODE_SHIFT 4 +#define ISCSI_SLOW_PATH_HDR_RESERVED1_MASK 0x1 +#define ISCSI_SLOW_PATH_HDR_RESERVED1_SHIFT 7 +}; + +struct iscsi_conn_update_ramrod_params { + struct iscsi_slow_path_hdr hdr; + __le16 conn_id; + __le32 fw_cid; + u8 flags; +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_HD_EN_MASK 0x1 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_HD_EN_SHIFT 0 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_DD_EN_MASK 0x1 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_DD_EN_SHIFT 1 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_INITIAL_R2T_MASK 0x1 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_INITIAL_R2T_SHIFT 2 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_IMMEDIATE_DATA_MASK 0x1 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_IMMEDIATE_DATA_SHIFT 3 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_RESERVED1_MASK 0xF +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_RESERVED1_SHIFT 4 + u8 reserved0[3]; + __le32 max_seq_size; + __le32 max_send_pdu_length; + __le32 max_recv_pdu_length; + __le32 first_seq_length; + __le32 exp_stat_sn; +}; + +struct iscsi_ext_cdb_cmd_hdr { + __le16 reserved1; + u8 flags_attr; +#define ISCSI_EXT_CDB_CMD_HDR_ATTR_MASK 0x7 +#define ISCSI_EXT_CDB_CMD_HDR_ATTR_SHIFT 0 +#define ISCSI_EXT_CDB_CMD_HDR_RSRV_MASK 0x3 +#define ISCSI_EXT_CDB_CMD_HDR_RSRV_SHIFT 3 +#define ISCSI_EXT_CDB_CMD_HDR_WRITE_MASK 0x1 +#define ISCSI_EXT_CDB_CMD_HDR_WRITE_SHIFT 5 +#define ISCSI_EXT_CDB_CMD_HDR_READ_MASK 0x1 +#define ISCSI_EXT_CDB_CMD_HDR_READ_SHIFT 6 +#define ISCSI_EXT_CDB_CMD_HDR_FINAL_MASK 0x1 +#define ISCSI_EXT_CDB_CMD_HDR_FINAL_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_EXT_CDB_CMD_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_EXT_CDB_CMD_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_EXT_CDB_CMD_HDR_CDB_SIZE_MASK 0xFF +#define ISCSI_EXT_CDB_CMD_HDR_CDB_SIZE_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 expected_transfer_length; + __le32 cmd_sn; + __le32 exp_stat_sn; + struct iscsi_sge cdb_sge; +}; + +struct iscsi_login_req_hdr { + u8 version_min; + u8 version_max; + u8 flags_attr; +#define ISCSI_LOGIN_REQ_HDR_NSG_MASK 0x3 +#define ISCSI_LOGIN_REQ_HDR_NSG_SHIFT 0 +#define ISCSI_LOGIN_REQ_HDR_CSG_MASK 0x3 +#define ISCSI_LOGIN_REQ_HDR_CSG_SHIFT 2 +#define ISCSI_LOGIN_REQ_HDR_RSRV_MASK 0x3 +#define ISCSI_LOGIN_REQ_HDR_RSRV_SHIFT 4 +#define ISCSI_LOGIN_REQ_HDR_C_MASK 0x1 +#define ISCSI_LOGIN_REQ_HDR_C_SHIFT 6 +#define ISCSI_LOGIN_REQ_HDR_T_MASK 0x1 +#define ISCSI_LOGIN_REQ_HDR_T_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_LOGIN_REQ_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_LOGIN_REQ_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_LOGIN_REQ_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_LOGIN_REQ_HDR_TOTAL_AHS_LEN_SHIFT 24 + __le32 isid_TABC; + __le16 tsih; + __le16 isid_d; + __le32 itt; + __le16 reserved1; + __le16 cid; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 reserved2[4]; +}; + +struct iscsi_logout_req_hdr { + __le16 reserved0; + u8 reason_code; + u8 opcode; + __le32 reserved1; + __le32 reserved2[2]; + __le32 itt; + __le16 reserved3; + __le16 cid; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 reserved4[4]; +}; + +struct iscsi_data_out_hdr { + __le16 reserved1; + u8 flags_attr; +#define ISCSI_DATA_OUT_HDR_RSRV_MASK 0x7F +#define ISCSI_DATA_OUT_HDR_RSRV_SHIFT 0 +#define ISCSI_DATA_OUT_HDR_FINAL_MASK 0x1 +#define ISCSI_DATA_OUT_HDR_FINAL_SHIFT 7 + u8 opcode; + __le32 reserved2; + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 reserved3; + __le32 exp_stat_sn; + __le32 reserved4; + __le32 data_sn; + __le32 buffer_offset; + __le32 reserved5; +}; + +struct iscsi_data_in_hdr { + u8 status_rsvd; + u8 reserved1; + u8 flags; +#define ISCSI_DATA_IN_HDR_STATUS_MASK 0x1 +#define ISCSI_DATA_IN_HDR_STATUS_SHIFT 0 +#define ISCSI_DATA_IN_HDR_UNDERFLOW_MASK 0x1 +#define ISCSI_DATA_IN_HDR_UNDERFLOW_SHIFT 1 +#define ISCSI_DATA_IN_HDR_OVERFLOW_MASK 0x1 +#define ISCSI_DATA_IN_HDR_OVERFLOW_SHIFT 2 +#define ISCSI_DATA_IN_HDR_RSRV_MASK 0x7 +#define ISCSI_DATA_IN_HDR_RSRV_SHIFT 3 +#define ISCSI_DATA_IN_HDR_ACK_MASK 0x1 +#define ISCSI_DATA_IN_HDR_ACK_SHIFT 6 +#define ISCSI_DATA_IN_HDR_FINAL_MASK 0x1 +#define ISCSI_DATA_IN_HDR_FINAL_SHIFT 7 + u8 opcode; + __le32 reserved2; + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 data_sn; + __le32 buffer_offset; + __le32 residual_count; +}; + +struct iscsi_r2t_hdr { + u8 reserved0[3]; + u8 opcode; + __le32 reserved2; + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 r2t_sn; + __le32 buffer_offset; + __le32 desired_data_trns_len; +}; + +struct iscsi_nop_out_hdr { + __le16 reserved1; + u8 flags_attr; +#define ISCSI_NOP_OUT_HDR_RSRV_MASK 0x7F +#define ISCSI_NOP_OUT_HDR_RSRV_SHIFT 0 +#define ISCSI_NOP_OUT_HDR_CONST1_MASK 0x1 +#define ISCSI_NOP_OUT_HDR_CONST1_SHIFT 7 + u8 opcode; + __le32 reserved2; + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 reserved3; + __le32 reserved4; + __le32 reserved5; + __le32 reserved6; +}; + +struct iscsi_nop_in_hdr { + __le16 reserved0; + u8 flags_attr; +#define ISCSI_NOP_IN_HDR_RSRV_MASK 0x7F +#define ISCSI_NOP_IN_HDR_RSRV_SHIFT 0 +#define ISCSI_NOP_IN_HDR_CONST1_MASK 0x1 +#define ISCSI_NOP_IN_HDR_CONST1_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_NOP_IN_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_NOP_IN_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_NOP_IN_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_NOP_IN_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 reserved5; + __le32 reserved6; + __le32 reserved7; +}; + +struct iscsi_login_response_hdr { + u8 version_active; + u8 version_max; + u8 flags_attr; +#define ISCSI_LOGIN_RESPONSE_HDR_NSG_MASK 0x3 +#define ISCSI_LOGIN_RESPONSE_HDR_NSG_SHIFT 0 +#define ISCSI_LOGIN_RESPONSE_HDR_CSG_MASK 0x3 +#define ISCSI_LOGIN_RESPONSE_HDR_CSG_SHIFT 2 +#define ISCSI_LOGIN_RESPONSE_HDR_RSRV_MASK 0x3 +#define ISCSI_LOGIN_RESPONSE_HDR_RSRV_SHIFT 4 +#define ISCSI_LOGIN_RESPONSE_HDR_C_MASK 0x1 +#define ISCSI_LOGIN_RESPONSE_HDR_C_SHIFT 6 +#define ISCSI_LOGIN_RESPONSE_HDR_T_MASK 0x1 +#define ISCSI_LOGIN_RESPONSE_HDR_T_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_LOGIN_RESPONSE_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_LOGIN_RESPONSE_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_LOGIN_RESPONSE_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_LOGIN_RESPONSE_HDR_TOTAL_AHS_LEN_SHIFT 24 + __le32 isid_TABC; + __le16 tsih; + __le16 isid_d; + __le32 itt; + __le32 reserved1; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le16 reserved2; + u8 status_detail; + u8 status_class; + __le32 reserved4[2]; +}; + +struct iscsi_logout_response_hdr { + u8 reserved1; + u8 response; + u8 flags; + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_LOGOUT_RESPONSE_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_LOGOUT_RESPONSE_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_LOGOUT_RESPONSE_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_LOGOUT_RESPONSE_HDR_TOTAL_AHS_LEN_SHIFT 24 + __le32 reserved2[2]; + __le32 itt; + __le32 reserved3; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 reserved4; + __le16 time2retain; + __le16 time2wait; + __le32 reserved5[1]; +}; + +struct iscsi_text_request_hdr { + __le16 reserved0; + u8 flags_attr; +#define ISCSI_TEXT_REQUEST_HDR_RSRV_MASK 0x3F +#define ISCSI_TEXT_REQUEST_HDR_RSRV_SHIFT 0 +#define ISCSI_TEXT_REQUEST_HDR_C_MASK 0x1 +#define ISCSI_TEXT_REQUEST_HDR_C_SHIFT 6 +#define ISCSI_TEXT_REQUEST_HDR_F_MASK 0x1 +#define ISCSI_TEXT_REQUEST_HDR_F_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_TEXT_REQUEST_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_TEXT_REQUEST_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_TEXT_REQUEST_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_TEXT_REQUEST_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 reserved4[4]; +}; + +struct iscsi_text_response_hdr { + __le16 reserved1; + u8 flags; +#define ISCSI_TEXT_RESPONSE_HDR_RSRV_MASK 0x3F +#define ISCSI_TEXT_RESPONSE_HDR_RSRV_SHIFT 0 +#define ISCSI_TEXT_RESPONSE_HDR_C_MASK 0x1 +#define ISCSI_TEXT_RESPONSE_HDR_C_SHIFT 6 +#define ISCSI_TEXT_RESPONSE_HDR_F_MASK 0x1 +#define ISCSI_TEXT_RESPONSE_HDR_F_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_TEXT_RESPONSE_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_TEXT_RESPONSE_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_TEXT_RESPONSE_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_TEXT_RESPONSE_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 reserved4[3]; +}; + +struct iscsi_tmf_request_hdr { + __le16 reserved0; + u8 function; + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_TMF_REQUEST_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_TMF_REQUEST_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_TMF_REQUEST_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_TMF_REQUEST_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 rtt; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 ref_cmd_sn; + __le32 exp_data_sn; + __le32 reserved4[2]; +}; + +struct iscsi_tmf_response_hdr { + u8 reserved2; + u8 hdr_response; + u8 hdr_flags; + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_TMF_RESPONSE_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_TMF_RESPONSE_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_TMF_RESPONSE_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_TMF_RESPONSE_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair reserved0; + __le32 itt; + __le32 rtt; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 reserved4[3]; +}; + +struct iscsi_response_hdr { + u8 hdr_status; + u8 hdr_response; + u8 hdr_flags; + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_RESPONSE_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_RESPONSE_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_RESPONSE_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_RESPONSE_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 snack_tag; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 exp_data_sn; + __le32 bi_residual_count; + __le32 residual_count; +}; + +struct iscsi_reject_hdr { + u8 reserved4; + u8 hdr_reason; + u8 hdr_flags; + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_REJECT_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_REJECT_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_REJECT_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_REJECT_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair reserved0; + __le32 reserved1; + __le32 reserved2; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 data_sn; + __le32 reserved3[2]; +}; + +union iscsi_task_hdr { + struct iscsi_common_hdr common; + struct data_hdr data; + struct iscsi_cmd_hdr cmd; + struct iscsi_ext_cdb_cmd_hdr ext_cdb_cmd; + struct iscsi_login_req_hdr login_req; + struct iscsi_logout_req_hdr logout_req; + struct iscsi_data_out_hdr data_out; + struct iscsi_data_in_hdr data_in; + struct iscsi_r2t_hdr r2t; + struct iscsi_nop_out_hdr nop_out; + struct iscsi_nop_in_hdr nop_in; + struct iscsi_login_response_hdr login_response; + struct iscsi_logout_response_hdr logout_response; + struct iscsi_text_request_hdr text_request; + struct iscsi_text_response_hdr text_response; + struct iscsi_tmf_request_hdr tmf_request; + struct iscsi_tmf_response_hdr tmf_response; + struct iscsi_response_hdr response; + struct iscsi_reject_hdr reject; + struct iscsi_async_msg_hdr async_msg; +}; + +struct iscsi_cqe_common { + __le16 conn_id; + u8 cqe_type; + union cqe_error_status error_bitmap; + __le32 reserved[3]; + union iscsi_task_hdr iscsi_hdr; +}; + +struct iscsi_cqe_solicited { + __le16 conn_id; + u8 cqe_type; + union cqe_error_status error_bitmap; + __le16 itid; + u8 task_type; + u8 fw_dbg_field; + __le32 reserved1[2]; + union iscsi_task_hdr iscsi_hdr; +}; + +struct iscsi_cqe_unsolicited { + __le16 conn_id; + u8 cqe_type; + union cqe_error_status error_bitmap; + __le16 reserved0; + u8 reserved1; + u8 unsol_cqe_type; + struct regpair rqe_opaque; + union iscsi_task_hdr iscsi_hdr; +}; + +union iscsi_cqe { + struct iscsi_cqe_common cqe_common; + struct iscsi_cqe_solicited cqe_solicited; + struct iscsi_cqe_unsolicited cqe_unsolicited; +}; + +enum iscsi_cqes_type { + ISCSI_CQE_TYPE_SOLICITED = 1, + ISCSI_CQE_TYPE_UNSOLICITED, + ISCSI_CQE_TYPE_SOLICITED_WITH_SENSE + , + ISCSI_CQE_TYPE_TASK_CLEANUP, + ISCSI_CQE_TYPE_DUMMY, + MAX_ISCSI_CQES_TYPE +}; + +enum iscsi_cqe_unsolicited_type { + ISCSI_CQE_UNSOLICITED_NONE, + ISCSI_CQE_UNSOLICITED_SINGLE, + ISCSI_CQE_UNSOLICITED_FIRST, + ISCSI_CQE_UNSOLICITED_MIDDLE, + ISCSI_CQE_UNSOLICITED_LAST, + MAX_ISCSI_CQE_UNSOLICITED_TYPE +}; + +struct iscsi_virt_sgl_ctx { + struct regpair sgl_base; + struct regpair dsgl_base; + __le32 sgl_initial_offset; + __le32 dsgl_initial_offset; + __le32 dsgl_curr_offset[2]; +}; + +struct iscsi_sgl_var_params { + u8 sgl_ptr; + u8 dsgl_ptr; + __le16 sge_offset; + __le16 dsge_offset; +}; + +struct iscsi_phys_sgl_ctx { + struct regpair sgl_base; + struct regpair dsgl_base; + u8 sgl_size; + u8 dsgl_size; + __le16 reserved; + struct iscsi_sgl_var_params var_params[2]; +}; + +union iscsi_data_desc_ctx { + struct iscsi_virt_sgl_ctx virt_sgl; + struct iscsi_phys_sgl_ctx phys_sgl; + struct iscsi_cached_sge_ctx cached_sge; +}; + +struct iscsi_debug_modes { + u8 flags; +#define ISCSI_DEBUG_MODES_ASSERT_IF_RX_CONN_ERROR_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RX_CONN_ERROR_SHIFT 0 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_RESET_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_RESET_SHIFT 1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_FIN_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_FIN_SHIFT 2 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_CLEANUP_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_CLEANUP_SHIFT 3 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_REJECT_OR_ASYNC_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_REJECT_OR_ASYNC_SHIFT 4 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_NOP_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_NOP_SHIFT 5 +#define ISCSI_DEBUG_MODES_RESERVED0_MASK 0x3 +#define ISCSI_DEBUG_MODES_RESERVED0_SHIFT 6 +}; + +struct iscsi_dif_flags { + u8 flags; +#define ISCSI_DIF_FLAGS_PROT_INTERVAL_SIZE_LOG_MASK 0xF +#define ISCSI_DIF_FLAGS_PROT_INTERVAL_SIZE_LOG_SHIFT 0 +#define ISCSI_DIF_FLAGS_DIF_TO_PEER_MASK 0x1 +#define ISCSI_DIF_FLAGS_DIF_TO_PEER_SHIFT 4 +#define ISCSI_DIF_FLAGS_HOST_INTERFACE_MASK 0x7 +#define ISCSI_DIF_FLAGS_HOST_INTERFACE_SHIFT 5 +}; + +enum iscsi_eqe_opcode { + ISCSI_EVENT_TYPE_INIT_FUNC = 0, + ISCSI_EVENT_TYPE_DESTROY_FUNC, + ISCSI_EVENT_TYPE_OFFLOAD_CONN, + ISCSI_EVENT_TYPE_UPDATE_CONN, + ISCSI_EVENT_TYPE_CLEAR_SQ, + ISCSI_EVENT_TYPE_TERMINATE_CONN, + ISCSI_EVENT_TYPE_ASYN_CONNECT_COMPLETE, + ISCSI_EVENT_TYPE_ASYN_TERMINATE_DONE, + RESERVED8, + RESERVED9, + ISCSI_EVENT_TYPE_START_OF_ERROR_TYPES = 10, + ISCSI_EVENT_TYPE_ASYN_ABORT_RCVD, + ISCSI_EVENT_TYPE_ASYN_CLOSE_RCVD, + ISCSI_EVENT_TYPE_ASYN_SYN_RCVD, + ISCSI_EVENT_TYPE_ASYN_MAX_RT_TIME, + ISCSI_EVENT_TYPE_ASYN_MAX_RT_CNT, + ISCSI_EVENT_TYPE_ASYN_MAX_KA_PROBES_CNT, + ISCSI_EVENT_TYPE_ASYN_FIN_WAIT2, + ISCSI_EVENT_TYPE_ISCSI_CONN_ERROR, + ISCSI_EVENT_TYPE_TCP_CONN_ERROR, + ISCSI_EVENT_TYPE_ASYN_DELETE_OOO_ISLES, + MAX_ISCSI_EQE_OPCODE +}; + +enum iscsi_error_types { + ISCSI_STATUS_NONE = 0, + ISCSI_CQE_ERROR_UNSOLICITED_RCV_ON_INVALID_CONN = 1, + ISCSI_CONN_ERROR_TASK_CID_MISMATCH, + ISCSI_CONN_ERROR_TASK_NOT_VALID, + ISCSI_CONN_ERROR_RQ_RING_IS_FULL, + ISCSI_CONN_ERROR_CMDQ_RING_IS_FULL, + ISCSI_CONN_ERROR_HQE_CACHING_FAILED, + ISCSI_CONN_ERROR_HEADER_DIGEST_ERROR, + ISCSI_CONN_ERROR_LOCAL_COMPLETION_ERROR, + ISCSI_CONN_ERROR_DATA_OVERRUN, + ISCSI_CONN_ERROR_OUT_OF_SGES_ERROR, + ISCSI_CONN_ERROR_TCP_SEG_PROC_URG_ERROR, + ISCSI_CONN_ERROR_TCP_SEG_PROC_IP_OPTIONS_ERROR, + ISCSI_CONN_ERROR_TCP_SEG_PROC_CONNECT_INVALID_WS_OPTION, + ISCSI_CONN_ERROR_TCP_IP_FRAGMENT_ERROR, + ISCSI_CONN_ERROR_PROTOCOL_ERR_AHS_LEN, + ISCSI_CONN_ERROR_PROTOCOL_ERR_AHS_TYPE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_ITT_OUT_OF_RANGE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_TTT_OUT_OF_RANGE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DATA_SEG_LEN_EXCEEDS_PDU_SIZE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_INVALID_OPCODE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_INVALID_OPCODE_BEFORE_UPDATE, + ISCSI_CONN_ERROR_UNVALID_NOPIN_DSL, + ISCSI_CONN_ERROR_PROTOCOL_ERR_R2T_CARRIES_NO_DATA, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DATA_SN, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DATA_IN_TTT, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DATA_OUT_ITT, + ISCSI_CONN_ERROR_PROTOCOL_ERR_R2T_TTT, + ISCSI_CONN_ERROR_PROTOCOL_ERR_R2T_BUFFER_OFFSET, + ISCSI_CONN_ERROR_PROTOCOL_ERR_BUFFER_OFFSET_OOO, + ISCSI_CONN_ERROR_PROTOCOL_ERR_R2T_SN, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DESIRED_DATA_TRNS_LEN_0, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DESIRED_DATA_TRNS_LEN_1, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DESIRED_DATA_TRNS_LEN_2, + ISCSI_CONN_ERROR_PROTOCOL_ERR_LUN, + ISCSI_CONN_ERROR_PROTOCOL_ERR_F_BIT_ZERO, + ISCSI_CONN_ERROR_PROTOCOL_ERR_F_BIT_ZERO_S_BIT_ONE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_EXP_STAT_SN, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DSL_NOT_ZERO, + ISCSI_CONN_ERROR_PROTOCOL_ERR_INVALID_DSL, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DATA_SEG_LEN_TOO_BIG, + ISCSI_CONN_ERROR_PROTOCOL_ERR_OUTSTANDING_R2T_COUNT, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DIF_TX, + ISCSI_CONN_ERROR_SENSE_DATA_LENGTH, + ISCSI_CONN_ERROR_DATA_PLACEMENT_ERROR, + ISCSI_ERROR_UNKNOWN, + MAX_ISCSI_ERROR_TYPES +}; + +struct iscsi_mflags { + u8 mflags; +#define ISCSI_MFLAGS_SLOW_IO_MASK 0x1 +#define ISCSI_MFLAGS_SLOW_IO_SHIFT 0 +#define ISCSI_MFLAGS_SINGLE_SGE_MASK 0x1 +#define ISCSI_MFLAGS_SINGLE_SGE_SHIFT 1 +#define ISCSI_MFLAGS_RESERVED_MASK 0x3F +#define ISCSI_MFLAGS_RESERVED_SHIFT 2 +}; + +struct iscsi_sgl { + struct regpair sgl_addr; + __le16 updated_sge_size; + __le16 updated_sge_offset; + __le32 byte_offset; +}; + +union iscsi_mstorm_sgl { + struct iscsi_sgl sgl_struct; + struct iscsi_sge single_sge; +}; + +enum iscsi_ramrod_cmd_id { + ISCSI_RAMROD_CMD_ID_UNUSED = 0, + ISCSI_RAMROD_CMD_ID_INIT_FUNC = 1, + ISCSI_RAMROD_CMD_ID_DESTROY_FUNC = 2, + ISCSI_RAMROD_CMD_ID_OFFLOAD_CONN = 3, + ISCSI_RAMROD_CMD_ID_UPDATE_CONN = 4, + ISCSI_RAMROD_CMD_ID_TERMINATION_CONN = 5, + ISCSI_RAMROD_CMD_ID_CLEAR_SQ = 6, + MAX_ISCSI_RAMROD_CMD_ID +}; + +struct iscsi_reg1 { + __le32 reg1_map; +#define ISCSI_REG1_NUM_FAST_SGES_MASK 0x7 +#define ISCSI_REG1_NUM_FAST_SGES_SHIFT 0 +#define ISCSI_REG1_RESERVED1_MASK 0x1FFFFFFF +#define ISCSI_REG1_RESERVED1_SHIFT 3 +}; + +union iscsi_seq_num { + __le16 data_sn; + __le16 r2t_sn; +}; + +struct iscsi_spe_conn_offload { + struct iscsi_slow_path_hdr hdr; + __le16 conn_id; + __le32 fw_cid; + struct iscsi_conn_offload_params iscsi; + struct tcp_offload_params tcp; +}; + +struct iscsi_spe_conn_offload_option2 { + struct iscsi_slow_path_hdr hdr; + __le16 conn_id; + __le32 fw_cid; + struct iscsi_conn_offload_params iscsi; + struct tcp_offload_params_opt2 tcp; +}; + +struct iscsi_spe_conn_termination { + struct iscsi_slow_path_hdr hdr; + __le16 conn_id; + __le32 fw_cid; + u8 abortive; + u8 reserved0[7]; + struct regpair queue_cnts_addr; + struct regpair query_params_addr; +}; + +struct iscsi_spe_func_dstry { + struct iscsi_slow_path_hdr hdr; + __le16 reserved0; + __le32 reserved1; +}; + +struct iscsi_spe_func_init { + struct iscsi_slow_path_hdr hdr; + __le16 half_way_close_timeout; + u8 num_sq_pages_in_ring; + u8 num_r2tq_pages_in_ring; + u8 num_uhq_pages_in_ring; + u8 ll2_rx_queue_id; + u8 ooo_enable; + struct iscsi_debug_modes debug_mode; + __le16 reserved1; + __le32 reserved2; + __le32 reserved3; + __le32 reserved4; + struct scsi_init_func_params func_params; + struct scsi_init_func_queues q_params; +}; + +struct ystorm_iscsi_task_state { + union iscsi_data_desc_ctx sgl_ctx_union; + __le32 buffer_offset[2]; + __le16 bytes_nxt_dif; + __le16 rxmit_bytes_nxt_dif; + union iscsi_seq_num seq_num_union; + u8 dif_bytes_leftover; + u8 rxmit_dif_bytes_leftover; + __le16 reuse_count; + struct iscsi_dif_flags dif_flags; + u8 local_comp; + __le32 exp_r2t_sn; + __le32 sgl_offset[2]; +}; + +struct ystorm_iscsi_task_st_ctx { + struct ystorm_iscsi_task_state state; + union iscsi_task_hdr pdu_hdr; +}; + +struct ystorm_iscsi_task_ag_ctx { + u8 reserved; + u8 byte1; + __le16 word0; + u8 flags0; +#define YSTORM_ISCSI_TASK_AG_CTX_NIBBLE0_MASK 0xF +#define YSTORM_ISCSI_TASK_AG_CTX_NIBBLE0_SHIFT 0 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT0_SHIFT 4 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT1_SHIFT 5 +#define YSTORM_ISCSI_TASK_AG_CTX_VALID_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_VALID_SHIFT 6 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT3_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT3_SHIFT 7 + u8 flags1; +#define YSTORM_ISCSI_TASK_AG_CTX_CF0_MASK 0x3 +#define YSTORM_ISCSI_TASK_AG_CTX_CF0_SHIFT 0 +#define YSTORM_ISCSI_TASK_AG_CTX_CF1_MASK 0x3 +#define YSTORM_ISCSI_TASK_AG_CTX_CF1_SHIFT 2 +#define YSTORM_ISCSI_TASK_AG_CTX_CF2SPECIAL_MASK 0x3 +#define YSTORM_ISCSI_TASK_AG_CTX_CF2SPECIAL_SHIFT 4 +#define YSTORM_ISCSI_TASK_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_CF0EN_SHIFT 6 +#define YSTORM_ISCSI_TASK_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_CF1EN_SHIFT 7 + u8 flags2; +#define YSTORM_ISCSI_TASK_AG_CTX_BIT4_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT4_SHIFT 0 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE0EN_SHIFT 1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE1EN_SHIFT 2 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE2EN_SHIFT 3 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE3EN_SHIFT 4 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE4EN_SHIFT 5 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE5EN_SHIFT 6 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE6EN_SHIFT 7 + u8 byte2; + __le32 TTT; + u8 byte3; + u8 byte4; + __le16 word1; +}; + +struct mstorm_iscsi_task_ag_ctx { + u8 cdu_validation; + u8 byte1; + __le16 task_cid; + u8 flags0; +#define MSTORM_ISCSI_TASK_AG_CTX_CONNECTION_TYPE_MASK 0xF +#define MSTORM_ISCSI_TASK_AG_CTX_CONNECTION_TYPE_SHIFT 0 +#define MSTORM_ISCSI_TASK_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_EXIST_IN_QM0_SHIFT 4 +#define MSTORM_ISCSI_TASK_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_BIT1_SHIFT 5 +#define MSTORM_ISCSI_TASK_AG_CTX_VALID_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_VALID_SHIFT 6 +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_FLAG_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_FLAG_SHIFT 7 + u8 flags1; +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_CF_MASK 0x3 +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_CF_SHIFT 0 +#define MSTORM_ISCSI_TASK_AG_CTX_CF1_MASK 0x3 +#define MSTORM_ISCSI_TASK_AG_CTX_CF1_SHIFT 2 +#define MSTORM_ISCSI_TASK_AG_CTX_CF2_MASK 0x3 +#define MSTORM_ISCSI_TASK_AG_CTX_CF2_SHIFT 4 +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_CF_EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_CF_EN_SHIFT 6 +#define MSTORM_ISCSI_TASK_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_CF1EN_SHIFT 7 + u8 flags2; +#define MSTORM_ISCSI_TASK_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_CF2EN_SHIFT 0 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE0EN_SHIFT 1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE1EN_SHIFT 2 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE2EN_SHIFT 3 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE3EN_SHIFT 4 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE4EN_SHIFT 5 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE5EN_SHIFT 6 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE6EN_SHIFT 7 + u8 byte2; + __le32 reg0; + u8 byte3; + u8 byte4; + __le16 word1; +}; + +struct ustorm_iscsi_task_ag_ctx { + u8 reserved; + u8 state; + __le16 icid; + u8 flags0; +#define USTORM_ISCSI_TASK_AG_CTX_CONNECTION_TYPE_MASK 0xF +#define USTORM_ISCSI_TASK_AG_CTX_CONNECTION_TYPE_SHIFT 0 +#define USTORM_ISCSI_TASK_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_EXIST_IN_QM0_SHIFT 4 +#define USTORM_ISCSI_TASK_AG_CTX_BIT1_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_BIT1_SHIFT 5 +#define USTORM_ISCSI_TASK_AG_CTX_HQ_SCANNED_CF_MASK 0x3 +#define USTORM_ISCSI_TASK_AG_CTX_HQ_SCANNED_CF_SHIFT 6 + u8 flags1; +#define USTORM_ISCSI_TASK_AG_CTX_RESERVED1_MASK 0x3 +#define USTORM_ISCSI_TASK_AG_CTX_RESERVED1_SHIFT 0 +#define USTORM_ISCSI_TASK_AG_CTX_R2T2RECV_MASK 0x3 +#define USTORM_ISCSI_TASK_AG_CTX_R2T2RECV_SHIFT 2 +#define USTORM_ISCSI_TASK_AG_CTX_CF3_MASK 0x3 +#define USTORM_ISCSI_TASK_AG_CTX_CF3_SHIFT 4 +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_CF_MASK 0x3 +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_CF_SHIFT 6 + u8 flags2; +#define USTORM_ISCSI_TASK_AG_CTX_HQ_SCANNED_CF_EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_HQ_SCANNED_CF_EN_SHIFT 0 +#define USTORM_ISCSI_TASK_AG_CTX_DISABLE_DATA_ACKED_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_DISABLE_DATA_ACKED_SHIFT 1 +#define USTORM_ISCSI_TASK_AG_CTX_R2T2RECV_EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_R2T2RECV_EN_SHIFT 2 +#define USTORM_ISCSI_TASK_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_CF_EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_CF_EN_SHIFT 4 +#define USTORM_ISCSI_TASK_AG_CTX_CMP_DATA_TOTAL_EXP_EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_CMP_DATA_TOTAL_EXP_EN_SHIFT 5 +#define USTORM_ISCSI_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE1EN_SHIFT 6 +#define USTORM_ISCSI_TASK_AG_CTX_CMP_CONT_RCV_EXP_EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_CMP_CONT_RCV_EXP_EN_SHIFT 7 + u8 flags3; +#define USTORM_ISCSI_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE3EN_SHIFT 0 +#define USTORM_ISCSI_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE4EN_SHIFT 1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE5EN_SHIFT 2 +#define USTORM_ISCSI_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE6EN_SHIFT 3 +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_TYPE_MASK 0xF +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_TYPE_SHIFT 4 + __le32 dif_err_intervals; + __le32 dif_error_1st_interval; + __le32 rcv_cont_len; + __le32 exp_cont_len; + __le32 total_data_acked; + __le32 exp_data_acked; + u8 next_tid_valid; + u8 byte3; + __le16 word1; + __le16 next_tid; + __le16 word3; + __le32 hdr_residual_count; + __le32 exp_r2t_sn; +}; + +struct mstorm_iscsi_task_st_ctx { + union iscsi_mstorm_sgl sgl_union; + struct iscsi_dif_flags dif_flags; + struct iscsi_mflags flags; + u8 sgl_size; + u8 host_sge_index; + __le16 dix_cur_sge_offset; + __le16 dix_cur_sge_size; + __le32 data_offset_rtid; + u8 dif_offset; + u8 dix_sgl_size; + u8 dix_sge_index; + u8 task_type; + struct regpair sense_db; + struct regpair dix_sgl_cur_sge; + __le32 rem_task_size; + __le16 reuse_count; + __le16 dif_data_residue; + u8 reserved0[4]; + __le32 reserved1[1]; +}; + +struct ustorm_iscsi_task_st_ctx { + __le32 rem_rcv_len; + __le32 exp_data_transfer_len; + __le32 exp_data_sn; + struct regpair lun; + struct iscsi_reg1 reg1; + u8 flags2; +#define USTORM_ISCSI_TASK_ST_CTX_AHS_EXIST_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_AHS_EXIST_SHIFT 0 +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED1_MASK 0x7F +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED1_SHIFT 1 + u8 reserved2; + __le16 reserved3; + __le32 reserved4; + __le32 reserved5; + __le32 reserved6; + __le32 reserved7; + u8 task_type; + u8 error_flags; +#define USTORM_ISCSI_TASK_ST_CTX_DATA_DIGEST_ERROR_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_DATA_DIGEST_ERROR_SHIFT 0 +#define USTORM_ISCSI_TASK_ST_CTX_DATA_TRUNCATED_ERROR_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_DATA_TRUNCATED_ERROR_SHIFT 1 +#define USTORM_ISCSI_TASK_ST_CTX_UNDER_RUN_ERROR_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_UNDER_RUN_ERROR_SHIFT 2 +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED8_MASK 0x1F +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED8_SHIFT 3 + u8 flags; +#define USTORM_ISCSI_TASK_ST_CTX_CQE_WRITE_MASK 0x3 +#define USTORM_ISCSI_TASK_ST_CTX_CQE_WRITE_SHIFT 0 +#define USTORM_ISCSI_TASK_ST_CTX_LOCAL_COMP_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_LOCAL_COMP_SHIFT 2 +#define USTORM_ISCSI_TASK_ST_CTX_Q0_R2TQE_WRITE_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_Q0_R2TQE_WRITE_SHIFT 3 +#define USTORM_ISCSI_TASK_ST_CTX_TOTALDATAACKED_DONE_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_TOTALDATAACKED_DONE_SHIFT 4 +#define USTORM_ISCSI_TASK_ST_CTX_HQSCANNED_DONE_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_HQSCANNED_DONE_SHIFT 5 +#define USTORM_ISCSI_TASK_ST_CTX_R2T2RECV_DONE_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_R2T2RECV_DONE_SHIFT 6 +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED0_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED0_SHIFT 7 + u8 cq_rss_number; +}; + +struct iscsi_task_context { + struct ystorm_iscsi_task_st_ctx ystorm_st_context; + struct regpair ystorm_st_padding[2]; + struct ystorm_iscsi_task_ag_ctx ystorm_ag_context; + struct regpair ystorm_ag_padding[2]; + struct tdif_task_context tdif_context; + struct mstorm_iscsi_task_ag_ctx mstorm_ag_context; + struct regpair mstorm_ag_padding[2]; + struct ustorm_iscsi_task_ag_ctx ustorm_ag_context; + struct mstorm_iscsi_task_st_ctx mstorm_st_context; + struct ustorm_iscsi_task_st_ctx ustorm_st_context; + struct rdif_task_context rdif_context; +}; + +enum iscsi_task_type { + ISCSI_TASK_TYPE_INITIATOR_WRITE, + ISCSI_TASK_TYPE_INITIATOR_READ, + ISCSI_TASK_TYPE_MIDPATH, + ISCSI_TASK_TYPE_UNSOLIC, + ISCSI_TASK_TYPE_EXCHCLEANUP, + ISCSI_TASK_TYPE_IRRELEVANT, + ISCSI_TASK_TYPE_TARGET_WRITE, + ISCSI_TASK_TYPE_TARGET_READ, + ISCSI_TASK_TYPE_TARGET_RESPONSE, + ISCSI_TASK_TYPE_LOGIN_RESPONSE, + MAX_ISCSI_TASK_TYPE +}; + +union iscsi_ttt_txlen_union { + __le32 desired_tx_len; + __le32 ttt; +}; + +struct iscsi_uhqe { + __le32 reg1; +#define ISCSI_UHQE_PDU_PAYLOAD_LEN_MASK 0xFFFFF +#define ISCSI_UHQE_PDU_PAYLOAD_LEN_SHIFT 0 +#define ISCSI_UHQE_LOCAL_COMP_MASK 0x1 +#define ISCSI_UHQE_LOCAL_COMP_SHIFT 20 +#define ISCSI_UHQE_TOGGLE_BIT_MASK 0x1 +#define ISCSI_UHQE_TOGGLE_BIT_SHIFT 21 +#define ISCSI_UHQE_PURE_PAYLOAD_MASK 0x1 +#define ISCSI_UHQE_PURE_PAYLOAD_SHIFT 22 +#define ISCSI_UHQE_LOGIN_RESPONSE_PDU_MASK 0x1 +#define ISCSI_UHQE_LOGIN_RESPONSE_PDU_SHIFT 23 +#define ISCSI_UHQE_TASK_ID_HI_MASK 0xFF +#define ISCSI_UHQE_TASK_ID_HI_SHIFT 24 + __le32 reg2; +#define ISCSI_UHQE_BUFFER_OFFSET_MASK 0xFFFFFF +#define ISCSI_UHQE_BUFFER_OFFSET_SHIFT 0 +#define ISCSI_UHQE_TASK_ID_LO_MASK 0xFF +#define ISCSI_UHQE_TASK_ID_LO_SHIFT 24 +}; + +struct iscsi_wqe_field { + __le32 contlen_cdbsize_field; +#define ISCSI_WQE_FIELD_CONT_LEN_MASK 0xFFFFFF +#define ISCSI_WQE_FIELD_CONT_LEN_SHIFT 0 +#define ISCSI_WQE_FIELD_CDB_SIZE_MASK 0xFF +#define ISCSI_WQE_FIELD_CDB_SIZE_SHIFT 24 +}; + +union iscsi_wqe_field_union { + struct iscsi_wqe_field cont_field; + __le32 prev_tid; +}; + +struct iscsi_wqe { + __le16 task_id; + u8 flags; +#define ISCSI_WQE_WQE_TYPE_MASK 0x7 +#define ISCSI_WQE_WQE_TYPE_SHIFT 0 +#define ISCSI_WQE_NUM_FAST_SGES_MASK 0x7 +#define ISCSI_WQE_NUM_FAST_SGES_SHIFT 3 +#define ISCSI_WQE_PTU_INVALIDATE_MASK 0x1 +#define ISCSI_WQE_PTU_INVALIDATE_SHIFT 6 +#define ISCSI_WQE_RESPONSE_MASK 0x1 +#define ISCSI_WQE_RESPONSE_SHIFT 7 + struct iscsi_dif_flags prot_flags; + union iscsi_wqe_field_union cont_prevtid_union; +}; + +enum iscsi_wqe_type { + ISCSI_WQE_TYPE_NORMAL, + ISCSI_WQE_TYPE_TASK_CLEANUP, + ISCSI_WQE_TYPE_MIDDLE_PATH, + ISCSI_WQE_TYPE_LOGIN, + ISCSI_WQE_TYPE_FIRST_R2T_CONT, + ISCSI_WQE_TYPE_NONFIRST_R2T_CONT, + ISCSI_WQE_TYPE_RESPONSE, + MAX_ISCSI_WQE_TYPE +}; + +struct iscsi_xhqe { + union iscsi_ttt_txlen_union ttt_or_txlen; + __le32 exp_stat_sn; + struct iscsi_dif_flags prot_flags; + u8 total_ahs_length; + u8 opcode; + u8 flags; +#define ISCSI_XHQE_NUM_FAST_SGES_MASK 0x7 +#define ISCSI_XHQE_NUM_FAST_SGES_SHIFT 0 +#define ISCSI_XHQE_FINAL_MASK 0x1 +#define ISCSI_XHQE_FINAL_SHIFT 3 +#define ISCSI_XHQE_SUPER_IO_MASK 0x1 +#define ISCSI_XHQE_SUPER_IO_SHIFT 4 +#define ISCSI_XHQE_STATUS_BIT_MASK 0x1 +#define ISCSI_XHQE_STATUS_BIT_SHIFT 5 +#define ISCSI_XHQE_RESERVED_MASK 0x3 +#define ISCSI_XHQE_RESERVED_SHIFT 6 + union iscsi_seq_num seq_num_union; + __le16 reserved1; +}; + +struct mstorm_iscsi_stats_drv { + struct regpair iscsi_rx_dropped_pdus_task_not_valid; +}; + +struct ooo_opaque { + __le32 cid; + u8 drop_isle; + u8 drop_size; + u8 ooo_opcode; + u8 ooo_isle; +}; + +struct pstorm_iscsi_stats_drv { + struct regpair iscsi_tx_bytes_cnt; + struct regpair iscsi_tx_packet_cnt; +}; + +struct tstorm_iscsi_stats_drv { + struct regpair iscsi_rx_bytes_cnt; + struct regpair iscsi_rx_packet_cnt; + struct regpair iscsi_rx_new_ooo_isle_events_cnt; + __le32 iscsi_cmdq_threshold_cnt; + __le32 iscsi_rq_threshold_cnt; + __le32 iscsi_immq_threshold_cnt; +}; + +struct ustorm_iscsi_stats_drv { + struct regpair iscsi_rx_data_pdu_cnt; + struct regpair iscsi_rx_r2t_pdu_cnt; + struct regpair iscsi_rx_total_pdu_cnt; +}; + +struct xstorm_iscsi_stats_drv { + struct regpair iscsi_tx_go_to_slow_start_event_cnt; + struct regpair iscsi_tx_fast_retransmit_event_cnt; +}; + +struct ystorm_iscsi_stats_drv { + struct regpair iscsi_tx_data_pdu_cnt; + struct regpair iscsi_tx_r2t_pdu_cnt; + struct regpair iscsi_tx_total_pdu_cnt; +}; + +struct iscsi_db_data { + u8 params; +#define ISCSI_DB_DATA_DEST_MASK 0x3 +#define ISCSI_DB_DATA_DEST_SHIFT 0 +#define ISCSI_DB_DATA_AGG_CMD_MASK 0x3 +#define ISCSI_DB_DATA_AGG_CMD_SHIFT 2 +#define ISCSI_DB_DATA_BYPASS_EN_MASK 0x1 +#define ISCSI_DB_DATA_BYPASS_EN_SHIFT 4 +#define ISCSI_DB_DATA_RESERVED_MASK 0x1 +#define ISCSI_DB_DATA_RESERVED_SHIFT 5 +#define ISCSI_DB_DATA_AGG_VAL_SEL_MASK 0x3 +#define ISCSI_DB_DATA_AGG_VAL_SEL_SHIFT 6 + u8 agg_flags; + __le16 sq_prod; +}; + +struct tstorm_iscsi_task_ag_ctx { + u8 byte0; + u8 byte1; + __le16 word0; + u8 flags0; +#define TSTORM_ISCSI_TASK_AG_CTX_NIBBLE0_MASK 0xF +#define TSTORM_ISCSI_TASK_AG_CTX_NIBBLE0_SHIFT 0 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT0_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT0_SHIFT 4 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT1_SHIFT 5 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT2_SHIFT 6 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT3_SHIFT 7 + u8 flags1; +#define TSTORM_ISCSI_TASK_AG_CTX_BIT4_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT4_SHIFT 0 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT5_SHIFT 1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF0_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF0_SHIFT 2 +#define TSTORM_ISCSI_TASK_AG_CTX_CF1_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF1_SHIFT 4 +#define TSTORM_ISCSI_TASK_AG_CTX_CF2_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF2_SHIFT 6 + u8 flags2; +#define TSTORM_ISCSI_TASK_AG_CTX_CF3_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF3_SHIFT 0 +#define TSTORM_ISCSI_TASK_AG_CTX_CF4_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF4_SHIFT 2 +#define TSTORM_ISCSI_TASK_AG_CTX_CF5_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF5_SHIFT 4 +#define TSTORM_ISCSI_TASK_AG_CTX_CF6_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF6_SHIFT 6 + u8 flags3; +#define TSTORM_ISCSI_TASK_AG_CTX_CF7_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF7_SHIFT 0 +#define TSTORM_ISCSI_TASK_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF0EN_SHIFT 2 +#define TSTORM_ISCSI_TASK_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF1EN_SHIFT 3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF2EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF2EN_SHIFT 4 +#define TSTORM_ISCSI_TASK_AG_CTX_CF3EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF3EN_SHIFT 5 +#define TSTORM_ISCSI_TASK_AG_CTX_CF4EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF4EN_SHIFT 6 +#define TSTORM_ISCSI_TASK_AG_CTX_CF5EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF5EN_SHIFT 7 + u8 flags4; +#define TSTORM_ISCSI_TASK_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF6EN_SHIFT 0 +#define TSTORM_ISCSI_TASK_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF7EN_SHIFT 1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE0EN_SHIFT 2 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE1EN_SHIFT 3 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE2EN_SHIFT 4 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE3EN_SHIFT 5 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE4EN_SHIFT 6 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE5EN_SHIFT 7 + u8 byte2; + __le16 word1; + __le32 reg0; + u8 byte3; + u8 byte4; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg1; + __le32 reg2; +}; + +#endif /* __ISCSI_COMMON__ */ diff --git a/include/linux/qed/rdma_common.h b/include/linux/qed/rdma_common.h new file mode 100644 index 0000000..187991c --- /dev/null +++ b/include/linux/qed/rdma_common.h @@ -0,0 +1,44 @@ +/* QLogic qed NIC Driver + * Copyright (c) 2015 QLogic Corporation + * + * This software is available under the terms of the GNU General Public License + * (GPL) Version 2, available from the file COPYING in the main directory of + * this source tree. + */ + +#ifndef __RDMA_COMMON__ +#define __RDMA_COMMON__ +/************************/ +/* RDMA FW CONSTANTS */ +/************************/ + +#define RDMA_RESERVED_LKEY (0) +#define RDMA_RING_PAGE_SIZE (0x1000) + +#define RDMA_MAX_SGE_PER_SQ_WQE (4) +#define RDMA_MAX_SGE_PER_RQ_WQE (4) + +#define RDMA_MAX_DATA_SIZE_IN_WQE (0x7FFFFFFF) + +#define RDMA_REQ_RD_ATOMIC_ELM_SIZE (0x50) +#define RDMA_RESP_RD_ATOMIC_ELM_SIZE (0x20) + +#define RDMA_MAX_CQS (64 * 1024) +#define RDMA_MAX_TIDS (128 * 1024 - 1) +#define RDMA_MAX_PDS (64 * 1024) + +#define RDMA_NUM_STATISTIC_COUNTERS MAX_NUM_VPORTS + +#define RDMA_TASK_TYPE (PROTOCOLID_ROCE) + +struct rdma_srq_id { + __le16 srq_idx; + __le16 opaque_fid; +}; + +struct rdma_srq_producers { + __le32 sge_prod; + __le32 wqe_prod; +}; + +#endif /* __RDMA_COMMON__ */ diff --git a/include/linux/qed/roce_common.h b/include/linux/qed/roce_common.h new file mode 100644 index 0000000..2eeaf3d --- /dev/null +++ b/include/linux/qed/roce_common.h @@ -0,0 +1,17 @@ +/* QLogic qed NIC Driver + * Copyright (c) 2015 QLogic Corporation + * + * This software is available under the terms of the GNU General Public License + * (GPL) Version 2, available from the file COPYING in the main directory of + * this source tree. + */ + +#ifndef __ROCE_COMMON__ +#define __ROCE_COMMON__ + +#define ROCE_REQ_MAX_INLINE_DATA_SIZE (256) +#define ROCE_REQ_MAX_SINGLE_SQ_WQE_SIZE (288) + +#define ROCE_MAX_QPS (32 * 1024) + +#endif /* __ROCE_COMMON__ */ diff --git a/include/linux/qed/storage_common.h b/include/linux/qed/storage_common.h new file mode 100644 index 0000000..3b8e1ef --- /dev/null +++ b/include/linux/qed/storage_common.h @@ -0,0 +1,91 @@ +/* QLogic qed NIC Driver + * Copyright (c) 2015 QLogic Corporation + * + * This software is available under the terms of the GNU General Public License + * (GPL) Version 2, available from the file COPYING in the main directory of + * this source tree. + */ + +#ifndef __STORAGE_COMMON__ +#define __STORAGE_COMMON__ + +#define NUM_OF_CMDQS_CQS (NUM_OF_GLOBAL_QUEUES / 2) +#define BDQ_NUM_RESOURCES (4) + +#define BDQ_ID_RQ (0) +#define BDQ_ID_IMM_DATA (1) +#define BDQ_NUM_IDS (2) + +#define BDQ_MAX_EXTERNAL_RING_SIZE (1 << 15) + +struct scsi_bd { + struct regpair address; + struct regpair opaque; +}; + +struct scsi_bdq_ram_drv_data { + __le16 external_producer; + __le16 reserved0[3]; +}; + +struct scsi_drv_cmdq { + __le16 cmdq_cons; + __le16 reserved0; + __le32 reserved1; +}; + +struct scsi_init_func_params { + __le16 num_tasks; + u8 log_page_size; + u8 debug_mode; + u8 reserved2[12]; +}; + +struct scsi_init_func_queues { + struct regpair glbl_q_params_addr; + __le16 rq_buffer_size; + __le16 cq_num_entries; + __le16 cmdq_num_entries; + u8 bdq_resource_id; + u8 q_validity; +#define SCSI_INIT_FUNC_QUEUES_RQ_VALID_MASK 0x1 +#define SCSI_INIT_FUNC_QUEUES_RQ_VALID_SHIFT 0 +#define SCSI_INIT_FUNC_QUEUES_IMM_DATA_VALID_MASK 0x1 +#define SCSI_INIT_FUNC_QUEUES_IMM_DATA_VALID_SHIFT 1 +#define SCSI_INIT_FUNC_QUEUES_CMD_VALID_MASK 0x1 +#define SCSI_INIT_FUNC_QUEUES_CMD_VALID_SHIFT 2 +#define SCSI_INIT_FUNC_QUEUES_RESERVED_VALID_MASK 0x1F +#define SCSI_INIT_FUNC_QUEUES_RESERVED_VALID_SHIFT 3 + u8 num_queues; + u8 queue_relative_offset; + u8 cq_sb_pi; + u8 cmdq_sb_pi; + __le16 cq_cmdq_sb_num_arr[NUM_OF_CMDQS_CQS]; + __le16 reserved0; + u8 bdq_pbl_num_entries[BDQ_NUM_IDS]; + struct regpair bdq_pbl_base_address[BDQ_NUM_IDS]; + __le16 bdq_xoff_threshold[BDQ_NUM_IDS]; + __le16 bdq_xon_threshold[BDQ_NUM_IDS]; + __le16 cmdq_xoff_threshold; + __le16 cmdq_xon_threshold; + __le32 reserved1; +}; + +struct scsi_ram_per_bdq_resource_drv_data { + struct scsi_bdq_ram_drv_data drv_data_per_bdq_id[BDQ_NUM_IDS]; +}; + +struct scsi_sge { + struct regpair sge_addr; + __le16 sge_len; + __le16 reserved0; + __le32 reserved1; +}; + +struct scsi_terminate_extra_params { + __le16 unsolicited_cq_count; + __le16 cmdq_count; + u8 reserved[4]; +}; + +#endif /* __STORAGE_COMMON__ */ diff --git a/include/linux/qed/tcp_common.h b/include/linux/qed/tcp_common.h new file mode 100644 index 0000000..accba0e --- /dev/null +++ b/include/linux/qed/tcp_common.h @@ -0,0 +1,226 @@ +/* QLogic qed NIC Driver + * Copyright (c) 2015 QLogic Corporation + * + * This software is available under the terms of the GNU General Public License + * (GPL) Version 2, available from the file COPYING in the main directory of + * this source tree. + */ + +#ifndef __TCP_COMMON__ +#define __TCP_COMMON__ + +#define TCP_INVALID_TIMEOUT_VAL -1 + +enum tcp_connect_mode { + TCP_CONNECT_ACTIVE, + TCP_CONNECT_PASSIVE, + MAX_TCP_CONNECT_MODE +}; + +struct tcp_init_params { + __le32 max_cwnd; + __le16 dup_ack_threshold; + __le16 tx_sws_timer; + __le16 min_rto; + __le16 min_rto_rt; + __le16 max_rto; + u8 maxfinrt; + u8 reserved[1]; +}; + +enum tcp_ip_version { + TCP_IPV4, + TCP_IPV6, + MAX_TCP_IP_VERSION +}; + +struct tcp_offload_params { + __le16 local_mac_addr_lo; + __le16 local_mac_addr_mid; + __le16 local_mac_addr_hi; + __le16 remote_mac_addr_lo; + __le16 remote_mac_addr_mid; + __le16 remote_mac_addr_hi; + __le16 vlan_id; + u8 flags; +#define TCP_OFFLOAD_PARAMS_TS_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_TS_EN_SHIFT 0 +#define TCP_OFFLOAD_PARAMS_DA_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_DA_EN_SHIFT 1 +#define TCP_OFFLOAD_PARAMS_KA_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_KA_EN_SHIFT 2 +#define TCP_OFFLOAD_PARAMS_NAGLE_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_NAGLE_EN_SHIFT 3 +#define TCP_OFFLOAD_PARAMS_DA_CNT_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_DA_CNT_EN_SHIFT 4 +#define TCP_OFFLOAD_PARAMS_FIN_SENT_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_FIN_SENT_SHIFT 5 +#define TCP_OFFLOAD_PARAMS_FIN_RECEIVED_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_FIN_RECEIVED_SHIFT 6 +#define TCP_OFFLOAD_PARAMS_RESERVED0_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_RESERVED0_SHIFT 7 + u8 ip_version; + __le32 remote_ip[4]; + __le32 local_ip[4]; + __le32 flow_label; + u8 ttl; + u8 tos_or_tc; + __le16 remote_port; + __le16 local_port; + __le16 mss; + u8 rcv_wnd_scale; + u8 connect_mode; + __le16 srtt; + __le32 cwnd; + __le32 ss_thresh; + __le16 reserved1; + u8 ka_max_probe_cnt; + u8 dup_ack_theshold; + __le32 rcv_next; + __le32 snd_una; + __le32 snd_next; + __le32 snd_max; + __le32 snd_wnd; + __le32 rcv_wnd; + __le32 snd_wl1; + __le32 ts_time; + __le32 ts_recent; + __le32 ts_recent_age; + __le32 total_rt; + __le32 ka_timeout_delta; + __le32 rt_timeout_delta; + u8 dup_ack_cnt; + u8 snd_wnd_probe_cnt; + u8 ka_probe_cnt; + u8 rt_cnt; + __le16 rtt_var; + __le16 reserved2; + __le32 ka_timeout; + __le32 ka_interval; + __le32 max_rt_time; + __le32 initial_rcv_wnd; + u8 snd_wnd_scale; + u8 ack_frequency; + __le16 da_timeout_value; + __le32 ts_ticks_per_second; +}; + +struct tcp_offload_params_opt2 { + __le16 local_mac_addr_lo; + __le16 local_mac_addr_mid; + __le16 local_mac_addr_hi; + __le16 remote_mac_addr_lo; + __le16 remote_mac_addr_mid; + __le16 remote_mac_addr_hi; + __le16 vlan_id; + u8 flags; +#define TCP_OFFLOAD_PARAMS_OPT2_TS_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_OPT2_TS_EN_SHIFT 0 +#define TCP_OFFLOAD_PARAMS_OPT2_DA_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_OPT2_DA_EN_SHIFT 1 +#define TCP_OFFLOAD_PARAMS_OPT2_KA_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_OPT2_KA_EN_SHIFT 2 +#define TCP_OFFLOAD_PARAMS_OPT2_RESERVED0_MASK 0x1F +#define TCP_OFFLOAD_PARAMS_OPT2_RESERVED0_SHIFT 3 + u8 ip_version; + __le32 remote_ip[4]; + __le32 local_ip[4]; + __le32 flow_label; + u8 ttl; + u8 tos_or_tc; + __le16 remote_port; + __le16 local_port; + __le16 mss; + u8 rcv_wnd_scale; + u8 connect_mode; + __le16 syn_ip_payload_length; + __le32 syn_phy_addr_lo; + __le32 syn_phy_addr_hi; + __le32 reserved1[22]; +}; + +enum tcp_seg_placement_event { + TCP_EVENT_ADD_PEN, + TCP_EVENT_ADD_NEW_ISLE, + TCP_EVENT_ADD_ISLE_RIGHT, + TCP_EVENT_ADD_ISLE_LEFT, + TCP_EVENT_JOIN, + TCP_EVENT_NOP, + MAX_TCP_SEG_PLACEMENT_EVENT +}; + +struct tcp_update_params { + __le16 flags; +#define TCP_UPDATE_PARAMS_REMOTE_MAC_ADDR_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_REMOTE_MAC_ADDR_CHANGED_SHIFT 0 +#define TCP_UPDATE_PARAMS_MSS_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_MSS_CHANGED_SHIFT 1 +#define TCP_UPDATE_PARAMS_TTL_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_TTL_CHANGED_SHIFT 2 +#define TCP_UPDATE_PARAMS_TOS_OR_TC_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_TOS_OR_TC_CHANGED_SHIFT 3 +#define TCP_UPDATE_PARAMS_KA_TIMEOUT_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_TIMEOUT_CHANGED_SHIFT 4 +#define TCP_UPDATE_PARAMS_KA_INTERVAL_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_INTERVAL_CHANGED_SHIFT 5 +#define TCP_UPDATE_PARAMS_MAX_RT_TIME_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_MAX_RT_TIME_CHANGED_SHIFT 6 +#define TCP_UPDATE_PARAMS_FLOW_LABEL_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_FLOW_LABEL_CHANGED_SHIFT 7 +#define TCP_UPDATE_PARAMS_INITIAL_RCV_WND_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_INITIAL_RCV_WND_CHANGED_SHIFT 8 +#define TCP_UPDATE_PARAMS_KA_MAX_PROBE_CNT_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_MAX_PROBE_CNT_CHANGED_SHIFT 9 +#define TCP_UPDATE_PARAMS_KA_EN_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_EN_CHANGED_SHIFT 10 +#define TCP_UPDATE_PARAMS_NAGLE_EN_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_NAGLE_EN_CHANGED_SHIFT 11 +#define TCP_UPDATE_PARAMS_KA_EN_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_EN_SHIFT 12 +#define TCP_UPDATE_PARAMS_NAGLE_EN_MASK 0x1 +#define TCP_UPDATE_PARAMS_NAGLE_EN_SHIFT 13 +#define TCP_UPDATE_PARAMS_KA_RESTART_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_RESTART_SHIFT 14 +#define TCP_UPDATE_PARAMS_RETRANSMIT_RESTART_MASK 0x1 +#define TCP_UPDATE_PARAMS_RETRANSMIT_RESTART_SHIFT 15 + __le16 remote_mac_addr_lo; + __le16 remote_mac_addr_mid; + __le16 remote_mac_addr_hi; + __le16 mss; + u8 ttl; + u8 tos_or_tc; + __le32 ka_timeout; + __le32 ka_interval; + __le32 max_rt_time; + __le32 flow_label; + __le32 initial_rcv_wnd; + u8 ka_max_probe_cnt; + u8 reserved1[7]; +}; + +struct tcp_upload_params { + __le32 rcv_next; + __le32 snd_una; + __le32 snd_next; + __le32 snd_max; + __le32 snd_wnd; + __le32 rcv_wnd; + __le32 snd_wl1; + __le32 cwnd; + __le32 ss_thresh; + __le16 srtt; + __le16 rtt_var; + __le32 ts_time; + __le32 ts_recent; + __le32 ts_recent_age; + __le32 total_rt; + __le32 ka_timeout_delta; + __le32 rt_timeout_delta; + u8 dup_ack_cnt; + u8 snd_wnd_probe_cnt; + u8 ka_probe_cnt; + u8 rt_cnt; + __le32 reserved; +}; + +#endif /* __TCP_COMMON__ */ -- cgit v0.10.2 From c5ac93191d7e6977c5c3465ac94c73ebb8a8ecba Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Fri, 3 Jun 2016 14:35:34 +0300 Subject: qed: Add iscsi/rdma personalities This patch adds in the ecore 2 new personalities in addition to QED_PCI_ETH - QED_PCI_ISCSI and QED_PCI_ETH_ROCE. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h index 1042f2a..4417f12 100644 --- a/drivers/net/ethernet/qlogic/qed/qed.h +++ b/drivers/net/ethernet/qlogic/qed/qed.h @@ -127,6 +127,8 @@ struct qed_tunn_update_params { */ enum qed_pci_personality { QED_PCI_ETH, + QED_PCI_ISCSI, + QED_PCI_ETH_ROCE, QED_PCI_DEFAULT /* default in shmem */ }; @@ -170,6 +172,8 @@ enum QED_PORT_MODE { enum qed_dev_cap { QED_DEV_CAP_ETH, + QED_DEV_CAP_ISCSI, + QED_DEV_CAP_ROCE, }; struct qed_hw_info { diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index 151173d..30c6d2e 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -1472,6 +1472,12 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, if (device_capabilities & NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ETHERNET) __set_bit(QED_DEV_CAP_ETH, &p_hwfn->hw_info.device_capabilities); + if (device_capabilities & NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ISCSI) + __set_bit(QED_DEV_CAP_ISCSI, + &p_hwfn->hw_info.device_capabilities); + if (device_capabilities & NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ROCE) + __set_bit(QED_DEV_CAP_ROCE, + &p_hwfn->hw_info.device_capabilities); return qed_mcp_fill_shmem_func_info(p_hwfn, p_ptt); } diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index 4cd70bf..72e0ad6 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -650,7 +650,7 @@ struct mstorm_vf_zone { /* personality per PF */ enum personality_type { BAD_PERSONALITY_TYP, - PERSONALITY_RESERVED, + PERSONALITY_ISCSI, PERSONALITY_RESERVED2, PERSONALITY_RDMA_AND_ETH, PERSONALITY_RESERVED3, @@ -7072,6 +7072,8 @@ struct public_func { #define FUNC_MF_CFG_PROTOCOL_MASK 0x000000f0 #define FUNC_MF_CFG_PROTOCOL_SHIFT 4 #define FUNC_MF_CFG_PROTOCOL_ETHERNET 0x00000000 +#define FUNC_MF_CFG_PROTOCOL_ISCSI 0x00000010 +#define FUNC_MF_CFG_PROTOCOL_ROCE 0x00000030 #define FUNC_MF_CFG_PROTOCOL_MAX 0x00000030 #define FUNC_MF_CFG_MIN_BW_MASK 0x0000ff00 @@ -7405,6 +7407,8 @@ struct nvm_cfg1_glob { u32 misc_sig; u32 device_capabilities; #define NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ETHERNET 0x1 +#define NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ISCSI 0x4 +#define NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ROCE 0x8 u32 power_dissipated; u32 power_consumed; u32 efi_version; diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index c807f67..f41f78d 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -207,6 +207,8 @@ int qed_fill_dev_info(struct qed_dev *cdev, dev_info->pci_mem_start = cdev->pci_params.mem_start; dev_info->pci_mem_end = cdev->pci_params.mem_end; dev_info->pci_irq = cdev->pci_params.irq; + dev_info->rdma_supported = + (cdev->hwfns[0].hw_info.personality == QED_PCI_ETH_ROCE); dev_info->is_mf_default = IS_MF_DEFAULT(&cdev->hwfns[0]); ether_addr_copy(dev_info->hw_mac, cdev->hwfns[0].hw_info.hw_mac_addr); @@ -901,7 +903,8 @@ static int qed_slowpath_stop(struct qed_dev *cdev) if (IS_PF(cdev)) { qed_free_stream_mem(cdev); - qed_sriov_disable(cdev, true); + if (IS_QED_ETH_IF(cdev)) + qed_sriov_disable(cdev, true); qed_nic_stop(cdev); qed_slowpath_irq_free(cdev); diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c index 2c143b3..a240f26 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c @@ -977,7 +977,18 @@ qed_mcp_get_shmem_proto(struct qed_hwfn *p_hwfn, switch (p_info->config & FUNC_MF_CFG_PROTOCOL_MASK) { case FUNC_MF_CFG_PROTOCOL_ETHERNET: - *p_proto = QED_PCI_ETH; + if (test_bit(QED_DEV_CAP_ROCE, + &p_hwfn->hw_info.device_capabilities)) + *p_proto = QED_PCI_ETH_ROCE; + else + *p_proto = QED_PCI_ETH; + break; + case FUNC_MF_CFG_PROTOCOL_ISCSI: + *p_proto = QED_PCI_ISCSI; + break; + case FUNC_MF_CFG_PROTOCOL_ROCE: + DP_NOTICE(p_hwfn, "RoCE personality is not a valid value!\n"); + rc = -EINVAL; break; default: rc = -EINVAL; diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c index 63b93fb..a52f3fc 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c @@ -358,11 +358,26 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, qed_tunn_set_pf_start_params(p_hwfn, p_tunn, &p_ramrod->tunnel_config); - p_hwfn->hw_info.personality = PERSONALITY_ETH; if (IS_MF_SI(p_hwfn)) p_ramrod->allow_npar_tx_switching = allow_npar_tx_switch; + switch (p_hwfn->hw_info.personality) { + case QED_PCI_ETH: + p_ramrod->personality = PERSONALITY_ETH; + break; + case QED_PCI_ISCSI: + p_ramrod->personality = PERSONALITY_ISCSI; + break; + case QED_PCI_ETH_ROCE: + p_ramrod->personality = PERSONALITY_RDMA_AND_ETH; + break; + default: + DP_NOTICE(p_hwfn, "Unkown personality %d\n", + p_hwfn->hw_info.personality); + p_ramrod->personality = PERSONALITY_ETH; + } + if (p_hwfn->cdev->p_iov_info) { struct qed_hw_sriov_info *p_iov = p_hwfn->cdev->p_iov_info; diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h index 077a3b6..40c0ada 100644 --- a/include/linux/qed/common_hsi.h +++ b/include/linux/qed/common_hsi.h @@ -517,9 +517,9 @@ enum mf_mode { /* Per-protocol connection types */ enum protocol_type { - PROTOCOLID_RESERVED1, + PROTOCOLID_ISCSI, PROTOCOLID_RESERVED2, - PROTOCOLID_RESERVED3, + PROTOCOLID_ROCE, PROTOCOLID_CORE, PROTOCOLID_ETH, PROTOCOLID_RESERVED4, diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index 15efccf..e8cc49f 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -58,8 +58,70 @@ struct qed_eth_pf_params { u16 num_cons; }; +/* Most of the the parameters below are described in the FW iSCSI / TCP HSI */ +struct qed_iscsi_pf_params { + u64 glbl_q_params_addr; + u64 bdq_pbl_base_addr[2]; + u32 max_cwnd; + u16 cq_num_entries; + u16 cmdq_num_entries; + u16 dup_ack_threshold; + u16 tx_sws_timer; + u16 min_rto; + u16 min_rto_rt; + u16 max_rto; + + /* The following parameters are used during HW-init + * and these parameters need to be passed as arguments + * to update_pf_params routine invoked before slowpath start + */ + u16 num_cons; + u16 num_tasks; + + /* The following parameters are used during protocol-init */ + u16 half_way_close_timeout; + u16 bdq_xoff_threshold[2]; + u16 bdq_xon_threshold[2]; + u16 cmdq_xoff_threshold; + u16 cmdq_xon_threshold; + u16 rq_buffer_size; + + u8 num_sq_pages_in_ring; + u8 num_r2tq_pages_in_ring; + u8 num_uhq_pages_in_ring; + u8 num_queues; + u8 log_page_size; + u8 rqe_log_size; + u8 max_fin_rt; + u8 gl_rq_pi; + u8 gl_cmd_pi; + u8 debug_mode; + u8 ll2_ooo_queue_id; + u8 ooo_enable; + + u8 is_target; + u8 bdq_pbl_num_entries[2]; +}; + +struct qed_rdma_pf_params { + /* Supplied to QED during resource allocation (may affect the ILT and + * the doorbell BAR). + */ + u32 min_dpis; /* number of requested DPIs */ + u32 num_mrs; /* number of requested memory regions */ + u32 num_qps; /* number of requested Queue Pairs */ + u32 num_srqs; /* number of requested SRQ */ + u8 roce_edpm_mode; /* see QED_ROCE_EDPM_MODE_ENABLE */ + u8 gl_pi; /* protocol index */ + + /* Will allocate rate limiters to be used with QPs */ + u8 enable_dcqcn; +}; + struct qed_pf_params { struct qed_eth_pf_params eth_pf_params; + struct qed_iscsi_pf_params iscsi_pf_params; + struct qed_rdma_pf_params rdma_pf_params; }; enum qed_int_mode { @@ -100,6 +162,8 @@ struct qed_dev_info { /* MFW version */ u32 mfw_rev; + bool rdma_supported; + u32 flash_size; u8 mf_mode; bool tx_switching; @@ -111,6 +175,7 @@ enum qed_sb_type { enum qed_protocol { QED_PROTOCOL_ETH, + QED_PROTOCOL_ISCSI, }; struct qed_link_params { -- cgit v0.10.2 From dbb799c39717e7b71a386b04b015ddca2dcb1ecd Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Fri, 3 Jun 2016 14:35:35 +0300 Subject: qed: Initialize hardware for new protocols RoCE and iSCSI would require some added/changed hw configuration in order to properly run; The biggest single change being the requirement of allocating and mapping host memory for several HW blocks that aren't being used by qede [SRC, QM, TM, etc.]. In addition, whereas qede is only using context memory for HW blocks, the new protocol would also require task memories to be added. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h index 4417f12..9a63df1 100644 --- a/drivers/net/ethernet/qlogic/qed/qed.h +++ b/drivers/net/ethernet/qlogic/qed/qed.h @@ -187,6 +187,8 @@ struct qed_hw_info { #define RESC_START(_p_hwfn, resc) ((_p_hwfn)->hw_info.resc_start[resc]) #define RESC_NUM(_p_hwfn, resc) ((_p_hwfn)->hw_info.resc_num[resc]) +#define RESC_END(_p_hwfn, resc) (RESC_START(_p_hwfn, resc) + \ + RESC_NUM(_p_hwfn, resc)) #define FEAT_NUM(_p_hwfn, resc) ((_p_hwfn)->hw_info.feat_num[resc]) u8 num_tc; @@ -259,6 +261,7 @@ struct qed_qm_info { u8 pure_lb_pq; u8 offload_pq; u8 pure_ack_pq; + u8 ooo_pq; u8 vf_queues_offset; u16 num_pqs; u16 num_vf_pqs; @@ -271,6 +274,7 @@ struct qed_qm_info { u8 pf_wfq; u32 pf_rl; struct qed_wfq_data *wfq_data; + u8 num_pf_rls; }; struct storm_stats { @@ -316,6 +320,7 @@ struct qed_hwfn { bool hw_init_done; u8 num_funcs_on_engine; + u8 enabled_func_idx; /* BAR access */ void __iomem *regview; @@ -354,6 +359,9 @@ struct qed_hwfn { /* Protocol related */ struct qed_pf_params pf_params; + bool b_rdma_enabled_in_prs; + u32 rdma_prs_search_reg; + /* Array of sb_info of all status blocks */ struct qed_sb_info *sbs_info[MAX_SB_PER_PF_MIMD]; u16 num_sbs; @@ -559,6 +567,7 @@ static inline u8 qed_concrete_to_sw_fid(struct qed_dev *cdev, } #define PURE_LB_TC 8 +#define OOO_LB_TC 9 int qed_configure_vport_wfq(struct qed_dev *cdev, u16 vp_id, u32 rate); void qed_configure_vp_wfq_on_link_change(struct qed_dev *cdev, u32 min_pf_rate); diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c index ac284c5..d85b7ba 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c +++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c @@ -39,6 +39,14 @@ #define DQ_RANGE_SHIFT 4 #define DQ_RANGE_ALIGN BIT(DQ_RANGE_SHIFT) +/* Searcher constants */ +#define SRC_MIN_NUM_ELEMS 256 + +/* Timers constants */ +#define TM_SHIFT 7 +#define TM_ALIGN BIT(TM_SHIFT) +#define TM_ELEM_SIZE 4 + /* ILT constants */ #define ILT_DEFAULT_HW_P_SIZE 3 #define ILT_PAGE_IN_BYTES(hw_p_size) (1U << ((hw_p_size) + 12)) @@ -56,26 +64,71 @@ union conn_context { struct core_conn_context core_ctx; struct eth_conn_context eth_ctx; + struct iscsi_conn_context iscsi_ctx; + struct roce_conn_context roce_ctx; +}; + +/* TYPE-0 task context - iSCSI */ +union type0_task_context { + struct iscsi_task_context iscsi_ctx; }; +/* TYPE-1 task context - ROCE */ +union type1_task_context { + struct rdma_task_context roce_ctx; +}; + +struct src_ent { + u8 opaque[56]; + u64 next; +}; + +#define CDUT_SEG_ALIGNMET 3 /* in 4k chunks */ +#define CDUT_SEG_ALIGNMET_IN_BYTES (1 << (CDUT_SEG_ALIGNMET + 12)) + #define CONN_CXT_SIZE(p_hwfn) \ ALIGNED_TYPE_SIZE(union conn_context, p_hwfn) +#define SRQ_CXT_SIZE (sizeof(struct rdma_srq_context)) + +#define TYPE0_TASK_CXT_SIZE(p_hwfn) \ + ALIGNED_TYPE_SIZE(union type0_task_context, p_hwfn) + +/* Alignment is inherent to the type1_task_context structure */ +#define TYPE1_TASK_CXT_SIZE(p_hwfn) sizeof(union type1_task_context) + /* PF per protocl configuration object */ +#define TASK_SEGMENTS (NUM_TASK_PF_SEGMENTS + NUM_TASK_VF_SEGMENTS) +#define TASK_SEGMENT_VF (NUM_TASK_PF_SEGMENTS) + +struct qed_tid_seg { + u32 count; + u8 type; + bool has_fl_mem; +}; + struct qed_conn_type_cfg { u32 cid_count; u32 cid_start; u32 cids_per_vf; + struct qed_tid_seg tid_seg[TASK_SEGMENTS]; }; /* ILT Client configuration, Per connection type (protocol) resources. */ #define ILT_CLI_PF_BLOCKS (1 + NUM_TASK_PF_SEGMENTS * 2) #define ILT_CLI_VF_BLOCKS (1 + NUM_TASK_VF_SEGMENTS * 2) #define CDUC_BLK (0) +#define SRQ_BLK (0) +#define CDUT_SEG_BLK(n) (1 + (u8)(n)) +#define CDUT_FL_SEG_BLK(n, X) (1 + (n) + NUM_TASK_ ## X ## _SEGMENTS) enum ilt_clients { ILT_CLI_CDUC, + ILT_CLI_CDUT, ILT_CLI_QM, + ILT_CLI_TM, + ILT_CLI_SRC, + ILT_CLI_TSDM, ILT_CLI_MAX }; @@ -88,6 +141,7 @@ struct qed_ilt_cli_blk { u32 total_size; /* 0 means not active */ u32 real_size_in_page; u32 start_line; + u32 dynamic_line_cnt; }; struct qed_ilt_client_cfg { @@ -131,18 +185,44 @@ struct qed_cxt_mngr { /* computed ILT structure */ struct qed_ilt_client_cfg clients[ILT_CLI_MAX]; + /* Task type sizes */ + u32 task_type_size[NUM_TASK_TYPES]; + /* total number of VFs for this hwfn - * ALL VFs are symmetric in terms of HW resources */ u32 vf_count; + /* total number of SRQ's for this hwfn */ + u32 srq_count; + /* Acquired CIDs */ struct qed_cid_acquired_map acquired[MAX_CONN_TYPES]; /* ILT shadow table */ struct qed_dma_mem *ilt_shadow; u32 pf_start_line; + + /* Mutex for a dynamic ILT allocation */ + struct mutex mutex; + + /* SRC T2 */ + struct qed_dma_mem *t2; + u32 t2_num_pages; + u64 first_free; + u64 last_free; }; +static bool src_proto(enum protocol_type type) +{ + return type == PROTOCOLID_ISCSI || + type == PROTOCOLID_ROCE; +} + +static bool tm_cid_proto(enum protocol_type type) +{ + return type == PROTOCOLID_ISCSI || + type == PROTOCOLID_ROCE; +} /* counts the iids for the CDU/CDUC ILT client configuration */ struct qed_cdu_iids { @@ -161,21 +241,120 @@ static void qed_cxt_cdu_iids(struct qed_cxt_mngr *p_mngr, } } +/* counts the iids for the Searcher block configuration */ +struct qed_src_iids { + u32 pf_cids; + u32 per_vf_cids; +}; + +static void qed_cxt_src_iids(struct qed_cxt_mngr *p_mngr, + struct qed_src_iids *iids) +{ + u32 i; + + for (i = 0; i < MAX_CONN_TYPES; i++) { + if (!src_proto(i)) + continue; + + iids->pf_cids += p_mngr->conn_cfg[i].cid_count; + iids->per_vf_cids += p_mngr->conn_cfg[i].cids_per_vf; + } +} + +/* counts the iids for the Timers block configuration */ +struct qed_tm_iids { + u32 pf_cids; + u32 pf_tids[NUM_TASK_PF_SEGMENTS]; /* per segment */ + u32 pf_tids_total; + u32 per_vf_cids; + u32 per_vf_tids; +}; + +static void qed_cxt_tm_iids(struct qed_cxt_mngr *p_mngr, + struct qed_tm_iids *iids) +{ + u32 i, j; + + for (i = 0; i < MAX_CONN_TYPES; i++) { + struct qed_conn_type_cfg *p_cfg = &p_mngr->conn_cfg[i]; + + if (tm_cid_proto(i)) { + iids->pf_cids += p_cfg->cid_count; + iids->per_vf_cids += p_cfg->cids_per_vf; + } + } + + iids->pf_cids = roundup(iids->pf_cids, TM_ALIGN); + iids->per_vf_cids = roundup(iids->per_vf_cids, TM_ALIGN); + iids->per_vf_tids = roundup(iids->per_vf_tids, TM_ALIGN); + + for (iids->pf_tids_total = 0, j = 0; j < NUM_TASK_PF_SEGMENTS; j++) { + iids->pf_tids[j] = roundup(iids->pf_tids[j], TM_ALIGN); + iids->pf_tids_total += iids->pf_tids[j]; + } +} + static void qed_cxt_qm_iids(struct qed_hwfn *p_hwfn, struct qed_qm_iids *iids) { struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; - u32 vf_cids = 0, type; + struct qed_tid_seg *segs; + u32 vf_cids = 0, type, j; + u32 vf_tids = 0; for (type = 0; type < MAX_CONN_TYPES; type++) { iids->cids += p_mngr->conn_cfg[type].cid_count; vf_cids += p_mngr->conn_cfg[type].cids_per_vf; + + segs = p_mngr->conn_cfg[type].tid_seg; + /* for each segment there is at most one + * protocol for which count is not 0. + */ + for (j = 0; j < NUM_TASK_PF_SEGMENTS; j++) + iids->tids += segs[j].count; + + /* The last array elelment is for the VFs. As for PF + * segments there can be only one protocol for + * which this value is not 0. + */ + vf_tids += segs[NUM_TASK_PF_SEGMENTS].count; } iids->vf_cids += vf_cids * p_mngr->vf_count; + iids->tids += vf_tids * p_mngr->vf_count; + DP_VERBOSE(p_hwfn, QED_MSG_ILT, - "iids: CIDS %08x vf_cids %08x\n", - iids->cids, iids->vf_cids); + "iids: CIDS %08x vf_cids %08x tids %08x vf_tids %08x\n", + iids->cids, iids->vf_cids, iids->tids, vf_tids); +} + +static struct qed_tid_seg *qed_cxt_tid_seg_info(struct qed_hwfn *p_hwfn, + u32 seg) +{ + struct qed_cxt_mngr *p_cfg = p_hwfn->p_cxt_mngr; + u32 i; + + /* Find the protocol with tid count > 0 for this segment. + * Note: there can only be one and this is already validated. + */ + for (i = 0; i < MAX_CONN_TYPES; i++) + if (p_cfg->conn_cfg[i].tid_seg[seg].count) + return &p_cfg->conn_cfg[i].tid_seg[seg]; + return NULL; +} + +void qed_cxt_set_srq_count(struct qed_hwfn *p_hwfn, u32 num_srqs) +{ + struct qed_cxt_mngr *p_mgr = p_hwfn->p_cxt_mngr; + + p_mgr->srq_count = num_srqs; +} + +u32 qed_cxt_get_srq_count(struct qed_hwfn *p_hwfn) +{ + struct qed_cxt_mngr *p_mgr = p_hwfn->p_cxt_mngr; + + return p_mgr->srq_count; } /* set the iids count per protocol */ @@ -188,6 +367,14 @@ static void qed_cxt_set_proto_cid_count(struct qed_hwfn *p_hwfn, p_conn->cid_count = roundup(cid_count, DQ_RANGE_ALIGN); p_conn->cids_per_vf = roundup(vf_cid_cnt, DQ_RANGE_ALIGN); + + if (type == PROTOCOLID_ROCE) { + u32 page_sz = p_mgr->clients[ILT_CLI_CDUC].p_size.val; + u32 cxt_size = CONN_CXT_SIZE(p_hwfn); + u32 elems_per_page = ILT_PAGE_IN_BYTES(page_sz) / cxt_size; + + p_conn->cid_count = roundup(p_conn->cid_count, elems_per_page); + } } u32 qed_cxt_get_proto_cid_count(struct qed_hwfn *p_hwfn, @@ -200,6 +387,37 @@ u32 qed_cxt_get_proto_cid_count(struct qed_hwfn *p_hwfn, return p_hwfn->p_cxt_mngr->conn_cfg[type].cid_count; } +u32 qed_cxt_get_proto_cid_start(struct qed_hwfn *p_hwfn, + enum protocol_type type) +{ + return p_hwfn->p_cxt_mngr->acquired[type].start_cid; +} + +u32 qed_cxt_get_proto_tid_count(struct qed_hwfn *p_hwfn, + enum protocol_type type) +{ + u32 cnt = 0; + int i; + + for (i = 0; i < TASK_SEGMENTS; i++) + cnt += p_hwfn->p_cxt_mngr->conn_cfg[type].tid_seg[i].count; + + return cnt; +} + +static void +qed_cxt_set_proto_tid_count(struct qed_hwfn *p_hwfn, + enum protocol_type proto, + u8 seg, u8 seg_type, u32 count, bool has_fl) +{ + struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; + struct qed_tid_seg *p_seg = &p_mngr->conn_cfg[proto].tid_seg[seg]; + + p_seg->count = count; + p_seg->has_fl_mem = has_fl; + p_seg->type = seg_type; +} + static void qed_ilt_cli_blk_fill(struct qed_ilt_client_cfg *p_cli, struct qed_ilt_cli_blk *p_blk, u32 start_line, u32 total_size, @@ -241,17 +459,42 @@ static void qed_ilt_cli_adv_line(struct qed_hwfn *p_hwfn, p_blk->real_size_in_page, p_blk->start_line); } +static u32 qed_ilt_get_dynamic_line_cnt(struct qed_hwfn *p_hwfn, + enum ilt_clients ilt_client) +{ + u32 cid_count = p_hwfn->p_cxt_mngr->conn_cfg[PROTOCOLID_ROCE].cid_count; + struct qed_ilt_client_cfg *p_cli; + u32 lines_to_skip = 0; + u32 cxts_per_p; + + if (ilt_client == ILT_CLI_CDUC) { + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUC]; + + cxts_per_p = ILT_PAGE_IN_BYTES(p_cli->p_size.val) / + (u32) CONN_CXT_SIZE(p_hwfn); + + lines_to_skip = cid_count / cxts_per_p; + } + + return lines_to_skip; +} + int qed_cxt_cfg_ilt_compute(struct qed_hwfn *p_hwfn) { struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; + u32 curr_line, total, i, task_size, line; struct qed_ilt_client_cfg *p_cli; struct qed_ilt_cli_blk *p_blk; struct qed_cdu_iids cdu_iids; + struct qed_src_iids src_iids; struct qed_qm_iids qm_iids; - u32 curr_line, total, i; + struct qed_tm_iids tm_iids; + struct qed_tid_seg *p_seg; memset(&qm_iids, 0, sizeof(qm_iids)); memset(&cdu_iids, 0, sizeof(cdu_iids)); + memset(&src_iids, 0, sizeof(src_iids)); + memset(&tm_iids, 0, sizeof(tm_iids)); p_mngr->pf_start_line = RESC_START(p_hwfn, QED_ILT); @@ -279,6 +522,9 @@ int qed_cxt_cfg_ilt_compute(struct qed_hwfn *p_hwfn) qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, ILT_CLI_CDUC); p_cli->pf_total_lines = curr_line - p_blk->start_line; + p_blk->dynamic_line_cnt = qed_ilt_get_dynamic_line_cnt(p_hwfn, + ILT_CLI_CDUC); + /* CDUC VF */ p_blk = &p_cli->vf_blks[CDUC_BLK]; total = cdu_iids.per_vf_cids * CONN_CXT_SIZE(p_hwfn); @@ -293,21 +539,128 @@ int qed_cxt_cfg_ilt_compute(struct qed_hwfn *p_hwfn) qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, ILT_CLI_CDUC); + /* CDUT PF */ + p_cli = &p_mngr->clients[ILT_CLI_CDUT]; + p_cli->first.val = curr_line; + + /* first the 'working' task memory */ + for (i = 0; i < NUM_TASK_PF_SEGMENTS; i++) { + p_seg = qed_cxt_tid_seg_info(p_hwfn, i); + if (!p_seg || p_seg->count == 0) + continue; + + p_blk = &p_cli->pf_blks[CDUT_SEG_BLK(i)]; + total = p_seg->count * p_mngr->task_type_size[p_seg->type]; + qed_ilt_cli_blk_fill(p_cli, p_blk, curr_line, total, + p_mngr->task_type_size[p_seg->type]); + + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_CDUT); + } + + /* next the 'init' task memory (forced load memory) */ + for (i = 0; i < NUM_TASK_PF_SEGMENTS; i++) { + p_seg = qed_cxt_tid_seg_info(p_hwfn, i); + if (!p_seg || p_seg->count == 0) + continue; + + p_blk = &p_cli->pf_blks[CDUT_FL_SEG_BLK(i, PF)]; + + if (!p_seg->has_fl_mem) { + /* The segment is active (total size pf 'working' + * memory is > 0) but has no FL (forced-load, Init) + * memory. Thus: + * + * 1. The total-size in the corrsponding FL block of + * the ILT client is set to 0 - No ILT line are + * provisioned and no ILT memory allocated. + * + * 2. The start-line of said block is set to the + * start line of the matching working memory + * block in the ILT client. This is later used to + * configure the CDU segment offset registers and + * results in an FL command for TIDs of this + * segement behaves as regular load commands + * (loading TIDs from the working memory). + */ + line = p_cli->pf_blks[CDUT_SEG_BLK(i)].start_line; + + qed_ilt_cli_blk_fill(p_cli, p_blk, line, 0, 0); + continue; + } + total = p_seg->count * p_mngr->task_type_size[p_seg->type]; + + qed_ilt_cli_blk_fill(p_cli, p_blk, + curr_line, total, + p_mngr->task_type_size[p_seg->type]); + + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_CDUT); + } + p_cli->pf_total_lines = curr_line - p_cli->pf_blks[0].start_line; + + /* CDUT VF */ + p_seg = qed_cxt_tid_seg_info(p_hwfn, TASK_SEGMENT_VF); + if (p_seg && p_seg->count) { + /* Stricly speaking we need to iterate over all VF + * task segment types, but a VF has only 1 segment + */ + + /* 'working' memory */ + total = p_seg->count * p_mngr->task_type_size[p_seg->type]; + + p_blk = &p_cli->vf_blks[CDUT_SEG_BLK(0)]; + qed_ilt_cli_blk_fill(p_cli, p_blk, + curr_line, total, + p_mngr->task_type_size[p_seg->type]); + + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_CDUT); + + /* 'init' memory */ + p_blk = &p_cli->vf_blks[CDUT_FL_SEG_BLK(0, VF)]; + if (!p_seg->has_fl_mem) { + /* see comment above */ + line = p_cli->vf_blks[CDUT_SEG_BLK(0)].start_line; + qed_ilt_cli_blk_fill(p_cli, p_blk, line, 0, 0); + } else { + task_size = p_mngr->task_type_size[p_seg->type]; + qed_ilt_cli_blk_fill(p_cli, p_blk, + curr_line, total, task_size); + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_CDUT); + } + p_cli->vf_total_lines = curr_line - + p_cli->vf_blks[0].start_line; + + /* Now for the rest of the VFs */ + for (i = 1; i < p_mngr->vf_count; i++) { + p_blk = &p_cli->vf_blks[CDUT_SEG_BLK(0)]; + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_CDUT); + + p_blk = &p_cli->vf_blks[CDUT_FL_SEG_BLK(0, VF)]; + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_CDUT); + } + } + /* QM */ p_cli = &p_mngr->clients[ILT_CLI_QM]; p_blk = &p_cli->pf_blks[0]; qed_cxt_qm_iids(p_hwfn, &qm_iids); total = qed_qm_pf_mem_size(p_hwfn->rel_pf_id, qm_iids.cids, - qm_iids.vf_cids, 0, + qm_iids.vf_cids, qm_iids.tids, p_hwfn->qm_info.num_pqs, p_hwfn->qm_info.num_vf_pqs); DP_VERBOSE(p_hwfn, QED_MSG_ILT, - "QM ILT Info, (cids=%d, vf_cids=%d, num_pqs=%d, num_vf_pqs=%d, memory_size=%d)\n", + "QM ILT Info, (cids=%d, vf_cids=%d, tids=%d, num_pqs=%d, num_vf_pqs=%d, memory_size=%d)\n", qm_iids.cids, qm_iids.vf_cids, + qm_iids.tids, p_hwfn->qm_info.num_pqs, p_hwfn->qm_info.num_vf_pqs, total); qed_ilt_cli_blk_fill(p_cli, p_blk, @@ -317,6 +670,75 @@ int qed_cxt_cfg_ilt_compute(struct qed_hwfn *p_hwfn) qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, ILT_CLI_QM); p_cli->pf_total_lines = curr_line - p_blk->start_line; + /* SRC */ + p_cli = &p_mngr->clients[ILT_CLI_SRC]; + qed_cxt_src_iids(p_mngr, &src_iids); + + /* Both the PF and VFs searcher connections are stored in the per PF + * database. Thus sum the PF searcher cids and all the VFs searcher + * cids. + */ + total = src_iids.pf_cids + src_iids.per_vf_cids * p_mngr->vf_count; + if (total) { + u32 local_max = max_t(u32, total, + SRC_MIN_NUM_ELEMS); + + total = roundup_pow_of_two(local_max); + + p_blk = &p_cli->pf_blks[0]; + qed_ilt_cli_blk_fill(p_cli, p_blk, curr_line, + total * sizeof(struct src_ent), + sizeof(struct src_ent)); + + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_SRC); + p_cli->pf_total_lines = curr_line - p_blk->start_line; + } + + /* TM PF */ + p_cli = &p_mngr->clients[ILT_CLI_TM]; + qed_cxt_tm_iids(p_mngr, &tm_iids); + total = tm_iids.pf_cids + tm_iids.pf_tids_total; + if (total) { + p_blk = &p_cli->pf_blks[0]; + qed_ilt_cli_blk_fill(p_cli, p_blk, curr_line, + total * TM_ELEM_SIZE, TM_ELEM_SIZE); + + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_TM); + p_cli->pf_total_lines = curr_line - p_blk->start_line; + } + + /* TM VF */ + total = tm_iids.per_vf_cids + tm_iids.per_vf_tids; + if (total) { + p_blk = &p_cli->vf_blks[0]; + qed_ilt_cli_blk_fill(p_cli, p_blk, curr_line, + total * TM_ELEM_SIZE, TM_ELEM_SIZE); + + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_TM); + p_cli->pf_total_lines = curr_line - p_blk->start_line; + + for (i = 1; i < p_mngr->vf_count; i++) + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_TM); + } + + /* TSDM (SRQ CONTEXT) */ + total = qed_cxt_get_srq_count(p_hwfn); + + if (total) { + p_cli = &p_mngr->clients[ILT_CLI_TSDM]; + p_blk = &p_cli->pf_blks[SRQ_BLK]; + qed_ilt_cli_blk_fill(p_cli, p_blk, curr_line, + total * SRQ_CXT_SIZE, SRQ_CXT_SIZE); + + qed_ilt_cli_adv_line(p_hwfn, p_cli, p_blk, &curr_line, + ILT_CLI_TSDM); + p_cli->pf_total_lines = curr_line - p_blk->start_line; + } + if (curr_line - p_hwfn->p_cxt_mngr->pf_start_line > RESC_NUM(p_hwfn, QED_ILT)) { DP_ERR(p_hwfn, "too many ilt lines...#lines=%d\n", @@ -327,8 +749,122 @@ int qed_cxt_cfg_ilt_compute(struct qed_hwfn *p_hwfn) return 0; } +static void qed_cxt_src_t2_free(struct qed_hwfn *p_hwfn) +{ + struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; + u32 i; + + if (!p_mngr->t2) + return; + + for (i = 0; i < p_mngr->t2_num_pages; i++) + if (p_mngr->t2[i].p_virt) + dma_free_coherent(&p_hwfn->cdev->pdev->dev, + p_mngr->t2[i].size, + p_mngr->t2[i].p_virt, + p_mngr->t2[i].p_phys); + + kfree(p_mngr->t2); + p_mngr->t2 = NULL; +} + +static int qed_cxt_src_t2_alloc(struct qed_hwfn *p_hwfn) +{ + struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; + u32 conn_num, total_size, ent_per_page, psz, i; + struct qed_ilt_client_cfg *p_src; + struct qed_src_iids src_iids; + struct qed_dma_mem *p_t2; + int rc; + + memset(&src_iids, 0, sizeof(src_iids)); + + /* if the SRC ILT client is inactive - there are no connection + * requiring the searcer, leave. + */ + p_src = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_SRC]; + if (!p_src->active) + return 0; + + qed_cxt_src_iids(p_mngr, &src_iids); + conn_num = src_iids.pf_cids + src_iids.per_vf_cids * p_mngr->vf_count; + total_size = conn_num * sizeof(struct src_ent); + + /* use the same page size as the SRC ILT client */ + psz = ILT_PAGE_IN_BYTES(p_src->p_size.val); + p_mngr->t2_num_pages = DIV_ROUND_UP(total_size, psz); + + /* allocate t2 */ + p_mngr->t2 = kzalloc(p_mngr->t2_num_pages * sizeof(struct qed_dma_mem), + GFP_KERNEL); + if (!p_mngr->t2) { + DP_NOTICE(p_hwfn, "Failed to allocate t2 table\n"); + rc = -ENOMEM; + goto t2_fail; + } + + /* allocate t2 pages */ + for (i = 0; i < p_mngr->t2_num_pages; i++) { + u32 size = min_t(u32, total_size, psz); + void **p_virt = &p_mngr->t2[i].p_virt; + + *p_virt = dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, + size, + &p_mngr->t2[i].p_phys, GFP_KERNEL); + if (!p_mngr->t2[i].p_virt) { + rc = -ENOMEM; + goto t2_fail; + } + memset(*p_virt, 0, size); + p_mngr->t2[i].size = size; + total_size -= size; + } + + /* Set the t2 pointers */ + + /* entries per page - must be a power of two */ + ent_per_page = psz / sizeof(struct src_ent); + + p_mngr->first_free = (u64) p_mngr->t2[0].p_phys; + + p_t2 = &p_mngr->t2[(conn_num - 1) / ent_per_page]; + p_mngr->last_free = (u64) p_t2->p_phys + + ((conn_num - 1) & (ent_per_page - 1)) * sizeof(struct src_ent); + + for (i = 0; i < p_mngr->t2_num_pages; i++) { + u32 ent_num = min_t(u32, + ent_per_page, + conn_num); + struct src_ent *entries = p_mngr->t2[i].p_virt; + u64 p_ent_phys = (u64) p_mngr->t2[i].p_phys, val; + u32 j; + + for (j = 0; j < ent_num - 1; j++) { + val = p_ent_phys + (j + 1) * sizeof(struct src_ent); + entries[j].next = cpu_to_be64(val); + } + + if (i < p_mngr->t2_num_pages - 1) + val = (u64) p_mngr->t2[i + 1].p_phys; + else + val = 0; + entries[j].next = cpu_to_be64(val); + + conn_num -= ent_per_page; + } + + return 0; + +t2_fail: + qed_cxt_src_t2_free(p_hwfn); + return rc; +} + #define for_each_ilt_valid_client(pos, clients) \ - for (pos = 0; pos < ILT_CLI_MAX; pos++) + for (pos = 0; pos < ILT_CLI_MAX; pos++) \ + if (!clients[pos].active) { \ + continue; \ + } else \ /* Total number of ILT lines used by this PF */ static u32 qed_cxt_ilt_shadow_size(struct qed_ilt_client_cfg *ilt_clients) @@ -336,12 +872,8 @@ static u32 qed_cxt_ilt_shadow_size(struct qed_ilt_client_cfg *ilt_clients) u32 size = 0; u32 i; - for_each_ilt_valid_client(i, ilt_clients) { - if (!ilt_clients[i].active) - continue; - size += (ilt_clients[i].last.val - - ilt_clients[i].first.val + 1); - } + for_each_ilt_valid_client(i, ilt_clients) + size += (ilt_clients[i].last.val - ilt_clients[i].first.val + 1); return size; } @@ -372,15 +904,22 @@ static int qed_ilt_blk_alloc(struct qed_hwfn *p_hwfn, u32 start_line_offset) { struct qed_dma_mem *ilt_shadow = p_hwfn->p_cxt_mngr->ilt_shadow; - u32 lines, line, sz_left; + u32 lines, line, sz_left, lines_to_skip = 0; + + /* Special handling for RoCE that supports dynamic allocation */ + if ((p_hwfn->hw_info.personality == QED_PCI_ETH_ROCE) && + ((ilt_client == ILT_CLI_CDUT) || ilt_client == ILT_CLI_TSDM)) + return 0; + + lines_to_skip = p_blk->dynamic_line_cnt; if (!p_blk->total_size) return 0; sz_left = p_blk->total_size; - lines = DIV_ROUND_UP(sz_left, p_blk->real_size_in_page); + lines = DIV_ROUND_UP(sz_left, p_blk->real_size_in_page) - lines_to_skip; line = p_blk->start_line + start_line_offset - - p_hwfn->p_cxt_mngr->pf_start_line; + p_hwfn->p_cxt_mngr->pf_start_line + lines_to_skip; for (; lines; lines--) { dma_addr_t p_phys; @@ -434,8 +973,6 @@ static int qed_ilt_shadow_alloc(struct qed_hwfn *p_hwfn) (u32)(size * sizeof(struct qed_dma_mem))); for_each_ilt_valid_client(i, clients) { - if (!clients[i].active) - continue; for (j = 0; j < ILT_CLI_PF_BLOCKS; j++) { p_blk = &clients[i].pf_blks[j]; rc = qed_ilt_blk_alloc(p_hwfn, p_blk, i, 0); @@ -514,6 +1051,7 @@ cid_map_fail: int qed_cxt_mngr_alloc(struct qed_hwfn *p_hwfn) { + struct qed_ilt_client_cfg *clients; struct qed_cxt_mngr *p_mngr; u32 i; @@ -524,20 +1062,42 @@ int qed_cxt_mngr_alloc(struct qed_hwfn *p_hwfn) } /* Initialize ILT client registers */ - p_mngr->clients[ILT_CLI_CDUC].first.reg = ILT_CFG_REG(CDUC, FIRST_ILT); - p_mngr->clients[ILT_CLI_CDUC].last.reg = ILT_CFG_REG(CDUC, LAST_ILT); - p_mngr->clients[ILT_CLI_CDUC].p_size.reg = ILT_CFG_REG(CDUC, P_SIZE); - - p_mngr->clients[ILT_CLI_QM].first.reg = ILT_CFG_REG(QM, FIRST_ILT); - p_mngr->clients[ILT_CLI_QM].last.reg = ILT_CFG_REG(QM, LAST_ILT); - p_mngr->clients[ILT_CLI_QM].p_size.reg = ILT_CFG_REG(QM, P_SIZE); - + clients = p_mngr->clients; + clients[ILT_CLI_CDUC].first.reg = ILT_CFG_REG(CDUC, FIRST_ILT); + clients[ILT_CLI_CDUC].last.reg = ILT_CFG_REG(CDUC, LAST_ILT); + clients[ILT_CLI_CDUC].p_size.reg = ILT_CFG_REG(CDUC, P_SIZE); + + clients[ILT_CLI_QM].first.reg = ILT_CFG_REG(QM, FIRST_ILT); + clients[ILT_CLI_QM].last.reg = ILT_CFG_REG(QM, LAST_ILT); + clients[ILT_CLI_QM].p_size.reg = ILT_CFG_REG(QM, P_SIZE); + + clients[ILT_CLI_TM].first.reg = ILT_CFG_REG(TM, FIRST_ILT); + clients[ILT_CLI_TM].last.reg = ILT_CFG_REG(TM, LAST_ILT); + clients[ILT_CLI_TM].p_size.reg = ILT_CFG_REG(TM, P_SIZE); + + clients[ILT_CLI_SRC].first.reg = ILT_CFG_REG(SRC, FIRST_ILT); + clients[ILT_CLI_SRC].last.reg = ILT_CFG_REG(SRC, LAST_ILT); + clients[ILT_CLI_SRC].p_size.reg = ILT_CFG_REG(SRC, P_SIZE); + + clients[ILT_CLI_CDUT].first.reg = ILT_CFG_REG(CDUT, FIRST_ILT); + clients[ILT_CLI_CDUT].last.reg = ILT_CFG_REG(CDUT, LAST_ILT); + clients[ILT_CLI_CDUT].p_size.reg = ILT_CFG_REG(CDUT, P_SIZE); + + clients[ILT_CLI_TSDM].first.reg = ILT_CFG_REG(TSDM, FIRST_ILT); + clients[ILT_CLI_TSDM].last.reg = ILT_CFG_REG(TSDM, LAST_ILT); + clients[ILT_CLI_TSDM].p_size.reg = ILT_CFG_REG(TSDM, P_SIZE); /* default ILT page size for all clients is 32K */ for (i = 0; i < ILT_CLI_MAX; i++) p_mngr->clients[i].p_size.val = ILT_DEFAULT_HW_P_SIZE; + /* Initialize task sizes */ + p_mngr->task_type_size[0] = TYPE0_TASK_CXT_SIZE(p_hwfn); + p_mngr->task_type_size[1] = TYPE1_TASK_CXT_SIZE(p_hwfn); + if (p_hwfn->cdev->p_iov_info) p_mngr->vf_count = p_hwfn->cdev->p_iov_info->total_vfs; + /* Initialize the dynamic ILT allocation mutex */ + mutex_init(&p_mngr->mutex); /* Set the cxt mangr pointer priori to further allocations */ p_hwfn->p_cxt_mngr = p_mngr; @@ -556,6 +1116,13 @@ int qed_cxt_tables_alloc(struct qed_hwfn *p_hwfn) goto tables_alloc_fail; } + /* Allocate the T2 table */ + rc = qed_cxt_src_t2_alloc(p_hwfn); + if (rc) { + DP_NOTICE(p_hwfn, "Failed to allocate T2 memory\n"); + goto tables_alloc_fail; + } + /* Allocate and initialize the acquired cids bitmaps */ rc = qed_cid_map_alloc(p_hwfn); if (rc) { @@ -576,6 +1143,7 @@ void qed_cxt_mngr_free(struct qed_hwfn *p_hwfn) return; qed_cid_map_free(p_hwfn); + qed_cxt_src_t2_free(p_hwfn); qed_ilt_shadow_free(p_hwfn); kfree(p_hwfn->p_cxt_mngr); @@ -620,6 +1188,48 @@ void qed_cxt_mngr_setup(struct qed_hwfn *p_hwfn) #define CDUC_NCIB_MASK \ (CDU_REG_CID_ADDR_PARAMS_NCIB >> CDUC_NCIB_SHIFT) +#define CDUT_TYPE0_CXT_SIZE_SHIFT \ + CDU_REG_SEGMENT0_PARAMS_T0_TID_SIZE_SHIFT + +#define CDUT_TYPE0_CXT_SIZE_MASK \ + (CDU_REG_SEGMENT0_PARAMS_T0_TID_SIZE >> \ + CDUT_TYPE0_CXT_SIZE_SHIFT) + +#define CDUT_TYPE0_BLOCK_WASTE_SHIFT \ + CDU_REG_SEGMENT0_PARAMS_T0_TID_BLOCK_WASTE_SHIFT + +#define CDUT_TYPE0_BLOCK_WASTE_MASK \ + (CDU_REG_SEGMENT0_PARAMS_T0_TID_BLOCK_WASTE >> \ + CDUT_TYPE0_BLOCK_WASTE_SHIFT) + +#define CDUT_TYPE0_NCIB_SHIFT \ + CDU_REG_SEGMENT0_PARAMS_T0_NUM_TIDS_IN_BLOCK_SHIFT + +#define CDUT_TYPE0_NCIB_MASK \ + (CDU_REG_SEGMENT0_PARAMS_T0_NUM_TIDS_IN_BLOCK >> \ + CDUT_TYPE0_NCIB_SHIFT) + +#define CDUT_TYPE1_CXT_SIZE_SHIFT \ + CDU_REG_SEGMENT1_PARAMS_T1_TID_SIZE_SHIFT + +#define CDUT_TYPE1_CXT_SIZE_MASK \ + (CDU_REG_SEGMENT1_PARAMS_T1_TID_SIZE >> \ + CDUT_TYPE1_CXT_SIZE_SHIFT) + +#define CDUT_TYPE1_BLOCK_WASTE_SHIFT \ + CDU_REG_SEGMENT1_PARAMS_T1_TID_BLOCK_WASTE_SHIFT + +#define CDUT_TYPE1_BLOCK_WASTE_MASK \ + (CDU_REG_SEGMENT1_PARAMS_T1_TID_BLOCK_WASTE >> \ + CDUT_TYPE1_BLOCK_WASTE_SHIFT) + +#define CDUT_TYPE1_NCIB_SHIFT \ + CDU_REG_SEGMENT1_PARAMS_T1_NUM_TIDS_IN_BLOCK_SHIFT + +#define CDUT_TYPE1_NCIB_MASK \ + (CDU_REG_SEGMENT1_PARAMS_T1_NUM_TIDS_IN_BLOCK >> \ + CDUT_TYPE1_NCIB_SHIFT) + static void qed_cdu_init_common(struct qed_hwfn *p_hwfn) { u32 page_sz, elems_per_page, block_waste, cxt_size, cdu_params = 0; @@ -634,6 +1244,92 @@ static void qed_cdu_init_common(struct qed_hwfn *p_hwfn) SET_FIELD(cdu_params, CDUC_BLOCK_WASTE, block_waste); SET_FIELD(cdu_params, CDUC_NCIB, elems_per_page); STORE_RT_REG(p_hwfn, CDU_REG_CID_ADDR_PARAMS_RT_OFFSET, cdu_params); + + /* CDUT - type-0 tasks configuration */ + page_sz = p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUT].p_size.val; + cxt_size = p_hwfn->p_cxt_mngr->task_type_size[0]; + elems_per_page = ILT_PAGE_IN_BYTES(page_sz) / cxt_size; + block_waste = ILT_PAGE_IN_BYTES(page_sz) - elems_per_page * cxt_size; + + /* cxt size and block-waste are multipes of 8 */ + cdu_params = 0; + SET_FIELD(cdu_params, CDUT_TYPE0_CXT_SIZE, (cxt_size >> 3)); + SET_FIELD(cdu_params, CDUT_TYPE0_BLOCK_WASTE, (block_waste >> 3)); + SET_FIELD(cdu_params, CDUT_TYPE0_NCIB, elems_per_page); + STORE_RT_REG(p_hwfn, CDU_REG_SEGMENT0_PARAMS_RT_OFFSET, cdu_params); + + /* CDUT - type-1 tasks configuration */ + cxt_size = p_hwfn->p_cxt_mngr->task_type_size[1]; + elems_per_page = ILT_PAGE_IN_BYTES(page_sz) / cxt_size; + block_waste = ILT_PAGE_IN_BYTES(page_sz) - elems_per_page * cxt_size; + + /* cxt size and block-waste are multipes of 8 */ + cdu_params = 0; + SET_FIELD(cdu_params, CDUT_TYPE1_CXT_SIZE, (cxt_size >> 3)); + SET_FIELD(cdu_params, CDUT_TYPE1_BLOCK_WASTE, (block_waste >> 3)); + SET_FIELD(cdu_params, CDUT_TYPE1_NCIB, elems_per_page); + STORE_RT_REG(p_hwfn, CDU_REG_SEGMENT1_PARAMS_RT_OFFSET, cdu_params); +} + +/* CDU PF */ +#define CDU_SEG_REG_TYPE_SHIFT CDU_SEG_TYPE_OFFSET_REG_TYPE_SHIFT +#define CDU_SEG_REG_TYPE_MASK 0x1 +#define CDU_SEG_REG_OFFSET_SHIFT 0 +#define CDU_SEG_REG_OFFSET_MASK CDU_SEG_TYPE_OFFSET_REG_OFFSET_MASK + +static void qed_cdu_init_pf(struct qed_hwfn *p_hwfn) +{ + struct qed_ilt_client_cfg *p_cli; + struct qed_tid_seg *p_seg; + u32 cdu_seg_params, offset; + int i; + + static const u32 rt_type_offset_arr[] = { + CDU_REG_PF_SEG0_TYPE_OFFSET_RT_OFFSET, + CDU_REG_PF_SEG1_TYPE_OFFSET_RT_OFFSET, + CDU_REG_PF_SEG2_TYPE_OFFSET_RT_OFFSET, + CDU_REG_PF_SEG3_TYPE_OFFSET_RT_OFFSET + }; + + static const u32 rt_type_offset_fl_arr[] = { + CDU_REG_PF_FL_SEG0_TYPE_OFFSET_RT_OFFSET, + CDU_REG_PF_FL_SEG1_TYPE_OFFSET_RT_OFFSET, + CDU_REG_PF_FL_SEG2_TYPE_OFFSET_RT_OFFSET, + CDU_REG_PF_FL_SEG3_TYPE_OFFSET_RT_OFFSET + }; + + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUT]; + + /* There are initializations only for CDUT during pf Phase */ + for (i = 0; i < NUM_TASK_PF_SEGMENTS; i++) { + /* Segment 0 */ + p_seg = qed_cxt_tid_seg_info(p_hwfn, i); + if (!p_seg) + continue; + + /* Note: start_line is already adjusted for the CDU + * segment register granularity, so we just need to + * divide. Adjustment is implicit as we assume ILT + * Page size is larger than 32K! + */ + offset = (ILT_PAGE_IN_BYTES(p_cli->p_size.val) * + (p_cli->pf_blks[CDUT_SEG_BLK(i)].start_line - + p_cli->first.val)) / CDUT_SEG_ALIGNMET_IN_BYTES; + + cdu_seg_params = 0; + SET_FIELD(cdu_seg_params, CDU_SEG_REG_TYPE, p_seg->type); + SET_FIELD(cdu_seg_params, CDU_SEG_REG_OFFSET, offset); + STORE_RT_REG(p_hwfn, rt_type_offset_arr[i], cdu_seg_params); + + offset = (ILT_PAGE_IN_BYTES(p_cli->p_size.val) * + (p_cli->pf_blks[CDUT_FL_SEG_BLK(i, PF)].start_line - + p_cli->first.val)) / CDUT_SEG_ALIGNMET_IN_BYTES; + + cdu_seg_params = 0; + SET_FIELD(cdu_seg_params, CDU_SEG_REG_TYPE, p_seg->type); + SET_FIELD(cdu_seg_params, CDU_SEG_REG_OFFSET, offset); + STORE_RT_REG(p_hwfn, rt_type_offset_fl_arr[i], cdu_seg_params); + } } void qed_qm_init_pf(struct qed_hwfn *p_hwfn) @@ -742,14 +1438,11 @@ static void qed_ilt_bounds_init(struct qed_hwfn *p_hwfn) ilt_clients = p_hwfn->p_cxt_mngr->clients; for_each_ilt_valid_client(i, ilt_clients) { - if (!ilt_clients[i].active) - continue; STORE_RT_REG(p_hwfn, ilt_clients[i].first.reg, ilt_clients[i].first.val); STORE_RT_REG(p_hwfn, - ilt_clients[i].last.reg, - ilt_clients[i].last.val); + ilt_clients[i].last.reg, ilt_clients[i].last.val); STORE_RT_REG(p_hwfn, ilt_clients[i].p_size.reg, ilt_clients[i].p_size.val); @@ -786,6 +1479,33 @@ static void qed_ilt_vf_bounds_init(struct qed_hwfn *p_hwfn) PSWRQ2_REG_CDUC_VF_BLOCKS_RT_OFFSET, p_cli->vf_total_lines); } + + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUT]; + blk_factor = ilog2(ILT_PAGE_IN_BYTES(p_cli->p_size.val) >> 10); + if (p_cli->active) { + STORE_RT_REG(p_hwfn, + PSWRQ2_REG_CDUT_BLOCKS_FACTOR_RT_OFFSET, + blk_factor); + STORE_RT_REG(p_hwfn, + PSWRQ2_REG_CDUT_NUMBER_OF_PF_BLOCKS_RT_OFFSET, + p_cli->pf_total_lines); + STORE_RT_REG(p_hwfn, + PSWRQ2_REG_CDUT_VF_BLOCKS_RT_OFFSET, + p_cli->vf_total_lines); + } + + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_TM]; + blk_factor = ilog2(ILT_PAGE_IN_BYTES(p_cli->p_size.val) >> 10); + if (p_cli->active) { + STORE_RT_REG(p_hwfn, + PSWRQ2_REG_TM_BLOCKS_FACTOR_RT_OFFSET, blk_factor); + STORE_RT_REG(p_hwfn, + PSWRQ2_REG_TM_NUMBER_OF_PF_BLOCKS_RT_OFFSET, + p_cli->pf_total_lines); + STORE_RT_REG(p_hwfn, + PSWRQ2_REG_TM_VF_BLOCKS_RT_OFFSET, + p_cli->vf_total_lines); + } } /* ILT (PSWRQ2) PF */ @@ -804,9 +1524,6 @@ static void qed_ilt_init_pf(struct qed_hwfn *p_hwfn) clients = p_hwfn->p_cxt_mngr->clients; for_each_ilt_valid_client(i, clients) { - if (!clients[i].active) - continue; - /** Client's 1st val and RT array are absolute, ILT shadows' * lines are relative. */ @@ -837,6 +1554,137 @@ static void qed_ilt_init_pf(struct qed_hwfn *p_hwfn) } } +/* SRC (Searcher) PF */ +static void qed_src_init_pf(struct qed_hwfn *p_hwfn) +{ + struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; + u32 rounded_conn_num, conn_num, conn_max; + struct qed_src_iids src_iids; + + memset(&src_iids, 0, sizeof(src_iids)); + qed_cxt_src_iids(p_mngr, &src_iids); + conn_num = src_iids.pf_cids + src_iids.per_vf_cids * p_mngr->vf_count; + if (!conn_num) + return; + + conn_max = max_t(u32, conn_num, SRC_MIN_NUM_ELEMS); + rounded_conn_num = roundup_pow_of_two(conn_max); + + STORE_RT_REG(p_hwfn, SRC_REG_COUNTFREE_RT_OFFSET, conn_num); + STORE_RT_REG(p_hwfn, SRC_REG_NUMBER_HASH_BITS_RT_OFFSET, + ilog2(rounded_conn_num)); + + STORE_RT_REG_AGG(p_hwfn, SRC_REG_FIRSTFREE_RT_OFFSET, + p_hwfn->p_cxt_mngr->first_free); + STORE_RT_REG_AGG(p_hwfn, SRC_REG_LASTFREE_RT_OFFSET, + p_hwfn->p_cxt_mngr->last_free); +} + +/* Timers PF */ +#define TM_CFG_NUM_IDS_SHIFT 0 +#define TM_CFG_NUM_IDS_MASK 0xFFFFULL +#define TM_CFG_PRE_SCAN_OFFSET_SHIFT 16 +#define TM_CFG_PRE_SCAN_OFFSET_MASK 0x1FFULL +#define TM_CFG_PARENT_PF_SHIFT 25 +#define TM_CFG_PARENT_PF_MASK 0x7ULL + +#define TM_CFG_CID_PRE_SCAN_ROWS_SHIFT 30 +#define TM_CFG_CID_PRE_SCAN_ROWS_MASK 0x1FFULL + +#define TM_CFG_TID_OFFSET_SHIFT 30 +#define TM_CFG_TID_OFFSET_MASK 0x7FFFFULL +#define TM_CFG_TID_PRE_SCAN_ROWS_SHIFT 49 +#define TM_CFG_TID_PRE_SCAN_ROWS_MASK 0x1FFULL + +static void qed_tm_init_pf(struct qed_hwfn *p_hwfn) +{ + struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; + u32 active_seg_mask = 0, tm_offset, rt_reg; + struct qed_tm_iids tm_iids; + u64 cfg_word; + u8 i; + + memset(&tm_iids, 0, sizeof(tm_iids)); + qed_cxt_tm_iids(p_mngr, &tm_iids); + + /* @@@TBD No pre-scan for now */ + + /* Note: We assume consecutive VFs for a PF */ + for (i = 0; i < p_mngr->vf_count; i++) { + cfg_word = 0; + SET_FIELD(cfg_word, TM_CFG_NUM_IDS, tm_iids.per_vf_cids); + SET_FIELD(cfg_word, TM_CFG_PRE_SCAN_OFFSET, 0); + SET_FIELD(cfg_word, TM_CFG_PARENT_PF, p_hwfn->rel_pf_id); + SET_FIELD(cfg_word, TM_CFG_CID_PRE_SCAN_ROWS, 0); + rt_reg = TM_REG_CONFIG_CONN_MEM_RT_OFFSET + + (sizeof(cfg_word) / sizeof(u32)) * + (p_hwfn->cdev->p_iov_info->first_vf_in_pf + i); + STORE_RT_REG_AGG(p_hwfn, rt_reg, cfg_word); + } + + cfg_word = 0; + SET_FIELD(cfg_word, TM_CFG_NUM_IDS, tm_iids.pf_cids); + SET_FIELD(cfg_word, TM_CFG_PRE_SCAN_OFFSET, 0); + SET_FIELD(cfg_word, TM_CFG_PARENT_PF, 0); /* n/a for PF */ + SET_FIELD(cfg_word, TM_CFG_CID_PRE_SCAN_ROWS, 0); /* scan all */ + + rt_reg = TM_REG_CONFIG_CONN_MEM_RT_OFFSET + + (sizeof(cfg_word) / sizeof(u32)) * + (NUM_OF_VFS(p_hwfn->cdev) + p_hwfn->rel_pf_id); + STORE_RT_REG_AGG(p_hwfn, rt_reg, cfg_word); + + /* enale scan */ + STORE_RT_REG(p_hwfn, TM_REG_PF_ENABLE_CONN_RT_OFFSET, + tm_iids.pf_cids ? 0x1 : 0x0); + + /* @@@TBD how to enable the scan for the VFs */ + + tm_offset = tm_iids.per_vf_cids; + + /* Note: We assume consecutive VFs for a PF */ + for (i = 0; i < p_mngr->vf_count; i++) { + cfg_word = 0; + SET_FIELD(cfg_word, TM_CFG_NUM_IDS, tm_iids.per_vf_tids); + SET_FIELD(cfg_word, TM_CFG_PRE_SCAN_OFFSET, 0); + SET_FIELD(cfg_word, TM_CFG_PARENT_PF, p_hwfn->rel_pf_id); + SET_FIELD(cfg_word, TM_CFG_TID_OFFSET, tm_offset); + SET_FIELD(cfg_word, TM_CFG_TID_PRE_SCAN_ROWS, (u64) 0); + + rt_reg = TM_REG_CONFIG_TASK_MEM_RT_OFFSET + + (sizeof(cfg_word) / sizeof(u32)) * + (p_hwfn->cdev->p_iov_info->first_vf_in_pf + i); + + STORE_RT_REG_AGG(p_hwfn, rt_reg, cfg_word); + } + + tm_offset = tm_iids.pf_cids; + for (i = 0; i < NUM_TASK_PF_SEGMENTS; i++) { + cfg_word = 0; + SET_FIELD(cfg_word, TM_CFG_NUM_IDS, tm_iids.pf_tids[i]); + SET_FIELD(cfg_word, TM_CFG_PRE_SCAN_OFFSET, 0); + SET_FIELD(cfg_word, TM_CFG_PARENT_PF, 0); + SET_FIELD(cfg_word, TM_CFG_TID_OFFSET, tm_offset); + SET_FIELD(cfg_word, TM_CFG_TID_PRE_SCAN_ROWS, (u64) 0); + + rt_reg = TM_REG_CONFIG_TASK_MEM_RT_OFFSET + + (sizeof(cfg_word) / sizeof(u32)) * + (NUM_OF_VFS(p_hwfn->cdev) + + p_hwfn->rel_pf_id * NUM_TASK_PF_SEGMENTS + i); + + STORE_RT_REG_AGG(p_hwfn, rt_reg, cfg_word); + active_seg_mask |= (tm_iids.pf_tids[i] ? (1 << i) : 0); + + tm_offset += tm_iids.pf_tids[i]; + } + + if (p_hwfn->hw_info.personality == QED_PCI_ETH_ROCE) + active_seg_mask = 0; + + STORE_RT_REG(p_hwfn, TM_REG_PF_ENABLE_TASK_RT_OFFSET, active_seg_mask); + + /* @@@TBD how to enable the scan for the VFs */ +} + void qed_cxt_hw_init_common(struct qed_hwfn *p_hwfn) { qed_cdu_init_common(p_hwfn); @@ -847,7 +1695,10 @@ void qed_cxt_hw_init_pf(struct qed_hwfn *p_hwfn) qed_qm_init_pf(p_hwfn); qed_cm_init_pf(p_hwfn); qed_dq_init_pf(p_hwfn); + qed_cdu_init_pf(p_hwfn); qed_ilt_init_pf(p_hwfn); + qed_src_init_pf(p_hwfn); + qed_tm_init_pf(p_hwfn); } int qed_cxt_acquire_cid(struct qed_hwfn *p_hwfn, @@ -968,17 +1819,439 @@ int qed_cxt_get_cid_info(struct qed_hwfn *p_hwfn, return 0; } -int qed_cxt_set_pf_params(struct qed_hwfn *p_hwfn) +void qed_rdma_set_pf_params(struct qed_hwfn *p_hwfn, + struct qed_rdma_pf_params *p_params) { - struct qed_eth_pf_params *p_params = &p_hwfn->pf_params.eth_pf_params; + u32 num_cons, num_tasks, num_qps, num_mrs, num_srqs; + enum protocol_type proto; + + num_mrs = min_t(u32, RDMA_MAX_TIDS, p_params->num_mrs); + num_tasks = num_mrs; /* each mr uses a single task id */ + num_srqs = min_t(u32, 32 * 1024, p_params->num_srqs); + + switch (p_hwfn->hw_info.personality) { + case QED_PCI_ETH_ROCE: + num_qps = min_t(u32, ROCE_MAX_QPS, p_params->num_qps); + num_cons = num_qps * 2; /* each QP requires two connections */ + proto = PROTOCOLID_ROCE; + break; + default: + return; + } + + if (num_cons && num_tasks) { + qed_cxt_set_proto_cid_count(p_hwfn, proto, num_cons, 0); + + /* Deliberatly passing ROCE for tasks id. This is because + * iWARP / RoCE share the task id. + */ + qed_cxt_set_proto_tid_count(p_hwfn, PROTOCOLID_ROCE, + QED_CXT_ROCE_TID_SEG, 1, + num_tasks, false); + qed_cxt_set_srq_count(p_hwfn, num_srqs); + } else { + DP_INFO(p_hwfn->cdev, + "RDMA personality used without setting params!\n"); + } +} +int qed_cxt_set_pf_params(struct qed_hwfn *p_hwfn) +{ /* Set the number of required CORE connections */ u32 core_cids = 1; /* SPQ */ qed_cxt_set_proto_cid_count(p_hwfn, PROTOCOLID_CORE, core_cids, 0); - qed_cxt_set_proto_cid_count(p_hwfn, PROTOCOLID_ETH, - p_params->num_cons, 1); + switch (p_hwfn->hw_info.personality) { + case QED_PCI_ETH_ROCE: + { + qed_rdma_set_pf_params(p_hwfn, + &p_hwfn-> + pf_params.rdma_pf_params); + /* no need for break since RoCE coexist with Ethernet */ + } + case QED_PCI_ETH: + { + struct qed_eth_pf_params *p_params = + &p_hwfn->pf_params.eth_pf_params; + + qed_cxt_set_proto_cid_count(p_hwfn, PROTOCOLID_ETH, + p_params->num_cons, 1); + break; + } + case QED_PCI_ISCSI: + { + struct qed_iscsi_pf_params *p_params; + + p_params = &p_hwfn->pf_params.iscsi_pf_params; + + if (p_params->num_cons && p_params->num_tasks) { + qed_cxt_set_proto_cid_count(p_hwfn, + PROTOCOLID_ISCSI, + p_params->num_cons, + 0); + + qed_cxt_set_proto_tid_count(p_hwfn, + PROTOCOLID_ISCSI, + QED_CXT_ISCSI_TID_SEG, + 0, + p_params->num_tasks, + true); + } else { + DP_INFO(p_hwfn->cdev, + "Iscsi personality used without setting params!\n"); + } + break; + } + default: + return -EINVAL; + } + + return 0; +} + +int qed_cxt_get_tid_mem_info(struct qed_hwfn *p_hwfn, + struct qed_tid_mem *p_info) +{ + struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; + u32 proto, seg, total_lines, i, shadow_line; + struct qed_ilt_client_cfg *p_cli; + struct qed_ilt_cli_blk *p_fl_seg; + struct qed_tid_seg *p_seg_info; + + /* Verify the personality */ + switch (p_hwfn->hw_info.personality) { + case QED_PCI_ISCSI: + proto = PROTOCOLID_ISCSI; + seg = QED_CXT_ISCSI_TID_SEG; + break; + default: + return -EINVAL; + } + + p_cli = &p_mngr->clients[ILT_CLI_CDUT]; + if (!p_cli->active) + return -EINVAL; + + p_seg_info = &p_mngr->conn_cfg[proto].tid_seg[seg]; + if (!p_seg_info->has_fl_mem) + return -EINVAL; + + p_fl_seg = &p_cli->pf_blks[CDUT_FL_SEG_BLK(seg, PF)]; + total_lines = DIV_ROUND_UP(p_fl_seg->total_size, + p_fl_seg->real_size_in_page); + + for (i = 0; i < total_lines; i++) { + shadow_line = i + p_fl_seg->start_line - + p_hwfn->p_cxt_mngr->pf_start_line; + p_info->blocks[i] = p_mngr->ilt_shadow[shadow_line].p_virt; + } + p_info->waste = ILT_PAGE_IN_BYTES(p_cli->p_size.val) - + p_fl_seg->real_size_in_page; + p_info->tid_size = p_mngr->task_type_size[p_seg_info->type]; + p_info->num_tids_per_block = p_fl_seg->real_size_in_page / + p_info->tid_size; + + return 0; +} + +/* This function is very RoCE oriented, if another protocol in the future + * will want this feature we'll need to modify the function to be more generic + */ +int +qed_cxt_dynamic_ilt_alloc(struct qed_hwfn *p_hwfn, + enum qed_cxt_elem_type elem_type, u32 iid) +{ + u32 reg_offset, shadow_line, elem_size, hw_p_size, elems_per_p, line; + struct qed_ilt_client_cfg *p_cli; + struct qed_ilt_cli_blk *p_blk; + struct qed_ptt *p_ptt; + dma_addr_t p_phys; + u64 ilt_hw_entry; + void *p_virt; + int rc = 0; + + switch (elem_type) { + case QED_ELEM_CXT: + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUC]; + elem_size = CONN_CXT_SIZE(p_hwfn); + p_blk = &p_cli->pf_blks[CDUC_BLK]; + break; + case QED_ELEM_SRQ: + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_TSDM]; + elem_size = SRQ_CXT_SIZE; + p_blk = &p_cli->pf_blks[SRQ_BLK]; + break; + case QED_ELEM_TASK: + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUT]; + elem_size = TYPE1_TASK_CXT_SIZE(p_hwfn); + p_blk = &p_cli->pf_blks[CDUT_SEG_BLK(QED_CXT_ROCE_TID_SEG)]; + break; + default: + DP_NOTICE(p_hwfn, "-EINVALID elem type = %d", elem_type); + return -EINVAL; + } + + /* Calculate line in ilt */ + hw_p_size = p_cli->p_size.val; + elems_per_p = ILT_PAGE_IN_BYTES(hw_p_size) / elem_size; + line = p_blk->start_line + (iid / elems_per_p); + shadow_line = line - p_hwfn->p_cxt_mngr->pf_start_line; + + /* If line is already allocated, do nothing, otherwise allocate it and + * write it to the PSWRQ2 registers. + * This section can be run in parallel from different contexts and thus + * a mutex protection is needed. + */ + + mutex_lock(&p_hwfn->p_cxt_mngr->mutex); + + if (p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].p_virt) + goto out0; + + p_ptt = qed_ptt_acquire(p_hwfn); + if (!p_ptt) { + DP_NOTICE(p_hwfn, + "QED_TIME_OUT on ptt acquire - dynamic allocation"); + rc = -EBUSY; + goto out0; + } + + p_virt = dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, + p_blk->real_size_in_page, + &p_phys, GFP_KERNEL); + if (!p_virt) { + rc = -ENOMEM; + goto out1; + } + memset(p_virt, 0, p_blk->real_size_in_page); + + /* configuration of refTagMask to 0xF is required for RoCE DIF MR only, + * to compensate for a HW bug, but it is configured even if DIF is not + * enabled. This is harmless and allows us to avoid a dedicated API. We + * configure the field for all of the contexts on the newly allocated + * page. + */ + if (elem_type == QED_ELEM_TASK) { + u32 elem_i; + u8 *elem_start = (u8 *)p_virt; + union type1_task_context *elem; + + for (elem_i = 0; elem_i < elems_per_p; elem_i++) { + elem = (union type1_task_context *)elem_start; + SET_FIELD(elem->roce_ctx.tdif_context.flags1, + TDIF_TASK_CONTEXT_REFTAGMASK, 0xf); + elem_start += TYPE1_TASK_CXT_SIZE(p_hwfn); + } + } + + p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].p_virt = p_virt; + p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].p_phys = p_phys; + p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].size = + p_blk->real_size_in_page; + + /* compute absolute offset */ + reg_offset = PSWRQ2_REG_ILT_MEMORY + + (line * ILT_REG_SIZE_IN_BYTES * ILT_ENTRY_IN_REGS); + + ilt_hw_entry = 0; + SET_FIELD(ilt_hw_entry, ILT_ENTRY_VALID, 1ULL); + SET_FIELD(ilt_hw_entry, + ILT_ENTRY_PHY_ADDR, + (p_hwfn->p_cxt_mngr->ilt_shadow[shadow_line].p_phys >> 12)); + + /* Write via DMAE since the PSWRQ2_REG_ILT_MEMORY line is a wide-bus */ + qed_dmae_host2grc(p_hwfn, p_ptt, (u64) (uintptr_t)&ilt_hw_entry, + reg_offset, sizeof(ilt_hw_entry) / sizeof(u32), 0); + + if (elem_type == QED_ELEM_CXT) { + u32 last_cid_allocated = (1 + (iid / elems_per_p)) * + elems_per_p; + + /* Update the relevant register in the parser */ + qed_wr(p_hwfn, p_ptt, PRS_REG_ROCE_DEST_QP_MAX_PF, + last_cid_allocated - 1); + + if (!p_hwfn->b_rdma_enabled_in_prs) { + /* Enable RoCE search */ + qed_wr(p_hwfn, p_ptt, p_hwfn->rdma_prs_search_reg, 1); + p_hwfn->b_rdma_enabled_in_prs = true; + } + } + +out1: + qed_ptt_release(p_hwfn, p_ptt); +out0: + mutex_unlock(&p_hwfn->p_cxt_mngr->mutex); + + return rc; +} + +/* This function is very RoCE oriented, if another protocol in the future + * will want this feature we'll need to modify the function to be more generic + */ +static int +qed_cxt_free_ilt_range(struct qed_hwfn *p_hwfn, + enum qed_cxt_elem_type elem_type, + u32 start_iid, u32 count) +{ + u32 start_line, end_line, shadow_start_line, shadow_end_line; + u32 reg_offset, elem_size, hw_p_size, elems_per_p; + struct qed_ilt_client_cfg *p_cli; + struct qed_ilt_cli_blk *p_blk; + u32 end_iid = start_iid + count; + struct qed_ptt *p_ptt; + u64 ilt_hw_entry = 0; + u32 i; + + switch (elem_type) { + case QED_ELEM_CXT: + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUC]; + elem_size = CONN_CXT_SIZE(p_hwfn); + p_blk = &p_cli->pf_blks[CDUC_BLK]; + break; + case QED_ELEM_SRQ: + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_TSDM]; + elem_size = SRQ_CXT_SIZE; + p_blk = &p_cli->pf_blks[SRQ_BLK]; + break; + case QED_ELEM_TASK: + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUT]; + elem_size = TYPE1_TASK_CXT_SIZE(p_hwfn); + p_blk = &p_cli->pf_blks[CDUT_SEG_BLK(QED_CXT_ROCE_TID_SEG)]; + break; + default: + DP_NOTICE(p_hwfn, "-EINVALID elem type = %d", elem_type); + return -EINVAL; + } + + /* Calculate line in ilt */ + hw_p_size = p_cli->p_size.val; + elems_per_p = ILT_PAGE_IN_BYTES(hw_p_size) / elem_size; + start_line = p_blk->start_line + (start_iid / elems_per_p); + end_line = p_blk->start_line + (end_iid / elems_per_p); + if (((end_iid + 1) / elems_per_p) != (end_iid / elems_per_p)) + end_line--; + + shadow_start_line = start_line - p_hwfn->p_cxt_mngr->pf_start_line; + shadow_end_line = end_line - p_hwfn->p_cxt_mngr->pf_start_line; + + p_ptt = qed_ptt_acquire(p_hwfn); + if (!p_ptt) { + DP_NOTICE(p_hwfn, + "QED_TIME_OUT on ptt acquire - dynamic allocation"); + return -EBUSY; + } + + for (i = shadow_start_line; i < shadow_end_line; i++) { + if (!p_hwfn->p_cxt_mngr->ilt_shadow[i].p_virt) + continue; + + dma_free_coherent(&p_hwfn->cdev->pdev->dev, + p_hwfn->p_cxt_mngr->ilt_shadow[i].size, + p_hwfn->p_cxt_mngr->ilt_shadow[i].p_virt, + p_hwfn->p_cxt_mngr->ilt_shadow[i].p_phys); + + p_hwfn->p_cxt_mngr->ilt_shadow[i].p_virt = NULL; + p_hwfn->p_cxt_mngr->ilt_shadow[i].p_phys = 0; + p_hwfn->p_cxt_mngr->ilt_shadow[i].size = 0; + + /* compute absolute offset */ + reg_offset = PSWRQ2_REG_ILT_MEMORY + + ((start_line++) * ILT_REG_SIZE_IN_BYTES * + ILT_ENTRY_IN_REGS); + + /* Write via DMAE since the PSWRQ2_REG_ILT_MEMORY line is a + * wide-bus. + */ + qed_dmae_host2grc(p_hwfn, p_ptt, + (u64) (uintptr_t) &ilt_hw_entry, + reg_offset, + sizeof(ilt_hw_entry) / sizeof(u32), + 0); + } + + qed_ptt_release(p_hwfn, p_ptt); + + return 0; +} + +int qed_cxt_free_proto_ilt(struct qed_hwfn *p_hwfn, enum protocol_type proto) +{ + int rc; + u32 cid; + + /* Free Connection CXT */ + rc = qed_cxt_free_ilt_range(p_hwfn, QED_ELEM_CXT, + qed_cxt_get_proto_cid_start(p_hwfn, + proto), + qed_cxt_get_proto_cid_count(p_hwfn, + proto, &cid)); + + if (rc) + return rc; + + /* Free Task CXT */ + rc = qed_cxt_free_ilt_range(p_hwfn, QED_ELEM_TASK, 0, + qed_cxt_get_proto_tid_count(p_hwfn, proto)); + if (rc) + return rc; + + /* Free TSDM CXT */ + rc = qed_cxt_free_ilt_range(p_hwfn, QED_ELEM_SRQ, 0, + qed_cxt_get_srq_count(p_hwfn)); + + return rc; +} + +int qed_cxt_get_task_ctx(struct qed_hwfn *p_hwfn, + u32 tid, u8 ctx_type, void **pp_task_ctx) +{ + struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; + struct qed_ilt_client_cfg *p_cli; + struct qed_ilt_cli_blk *p_seg; + struct qed_tid_seg *p_seg_info; + u32 proto, seg; + u32 total_lines; + u32 tid_size, ilt_idx; + u32 num_tids_per_block; + + /* Verify the personality */ + switch (p_hwfn->hw_info.personality) { + case QED_PCI_ISCSI: + proto = PROTOCOLID_ISCSI; + seg = QED_CXT_ISCSI_TID_SEG; + break; + default: + return -EINVAL; + } + + p_cli = &p_mngr->clients[ILT_CLI_CDUT]; + if (!p_cli->active) + return -EINVAL; + + p_seg_info = &p_mngr->conn_cfg[proto].tid_seg[seg]; + + if (ctx_type == QED_CTX_WORKING_MEM) { + p_seg = &p_cli->pf_blks[CDUT_SEG_BLK(seg)]; + } else if (ctx_type == QED_CTX_FL_MEM) { + if (!p_seg_info->has_fl_mem) + return -EINVAL; + p_seg = &p_cli->pf_blks[CDUT_FL_SEG_BLK(seg, PF)]; + } else { + return -EINVAL; + } + total_lines = DIV_ROUND_UP(p_seg->total_size, p_seg->real_size_in_page); + tid_size = p_mngr->task_type_size[p_seg_info->type]; + num_tids_per_block = p_seg->real_size_in_page / tid_size; + + if (total_lines < tid / num_tids_per_block) + return -EINVAL; + + ilt_idx = tid / num_tids_per_block + p_seg->start_line - + p_mngr->pf_start_line; + *pp_task_ctx = (u8 *)p_mngr->ilt_shadow[ilt_idx].p_virt + + (tid % num_tids_per_block) * tid_size; return 0; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.h b/drivers/net/ethernet/qlogic/qed/qed_cxt.h index 234c0fa..c6f6f2e 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_cxt.h +++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.h @@ -21,6 +21,14 @@ struct qed_cxt_info { enum protocol_type type; }; +#define MAX_TID_BLOCKS 512 +struct qed_tid_mem { + u32 tid_size; + u32 num_tids_per_block; + u32 waste; + u8 *blocks[MAX_TID_BLOCKS]; /* 4K */ +}; + /** * @brief qed_cxt_acquire - Acquire a new cid of a specific protocol type * @@ -46,8 +54,22 @@ int qed_cxt_acquire_cid(struct qed_hwfn *p_hwfn, int qed_cxt_get_cid_info(struct qed_hwfn *p_hwfn, struct qed_cxt_info *p_info); +/** + * @brief qed_cxt_get_tid_mem_info + * + * @param p_hwfn + * @param p_info + * + * @return int + */ +int qed_cxt_get_tid_mem_info(struct qed_hwfn *p_hwfn, + struct qed_tid_mem *p_info); + +#define QED_CXT_ISCSI_TID_SEG PROTOCOLID_ISCSI +#define QED_CXT_ROCE_TID_SEG PROTOCOLID_ROCE enum qed_cxt_elem_type { QED_ELEM_CXT, + QED_ELEM_SRQ, QED_ELEM_TASK }; @@ -149,4 +171,6 @@ int qed_qm_reconf(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt); void qed_cxt_release_cid(struct qed_hwfn *p_hwfn, u32 cid); +#define QED_CTX_WORKING_MEM 0 +#define QED_CTX_FL_MEM 1 #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index 30c6d2e..e45cff4 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -161,9 +161,13 @@ static int qed_init_qm_info(struct qed_hwfn *p_hwfn, bool b_sleepable) u8 num_vports, vf_offset = 0, i, vport_id, num_ports, curr_queue = 0; struct qed_qm_info *qm_info = &p_hwfn->qm_info; struct init_qm_port_params *p_qm_port; + bool init_rdma_offload_pq = false; + bool init_pure_ack_pq = false; + bool init_ooo_pq = false; u16 num_pqs, multi_cos_tcs = 1; u8 pf_wfq = qm_info->pf_wfq; u32 pf_rl = qm_info->pf_rl; + u16 num_pf_rls = 0; u16 num_vfs = 0; #ifdef CONFIG_QED_SRIOV @@ -175,6 +179,25 @@ static int qed_init_qm_info(struct qed_hwfn *p_hwfn, bool b_sleepable) num_pqs = multi_cos_tcs + num_vfs + 1; /* The '1' is for pure-LB */ num_vports = (u8)RESC_NUM(p_hwfn, QED_VPORT); + if (p_hwfn->hw_info.personality == QED_PCI_ETH_ROCE) { + num_pqs++; /* for RoCE queue */ + init_rdma_offload_pq = true; + /* we subtract num_vfs because each require a rate limiter, + * and one default rate limiter + */ + if (p_hwfn->pf_params.rdma_pf_params.enable_dcqcn) + num_pf_rls = RESC_NUM(p_hwfn, QED_RL) - num_vfs - 1; + + num_pqs += num_pf_rls; + qm_info->num_pf_rls = (u8) num_pf_rls; + } + + if (p_hwfn->hw_info.personality == QED_PCI_ISCSI) { + num_pqs += 2; /* for iSCSI pure-ACK / OOO queue */ + init_pure_ack_pq = true; + init_ooo_pq = true; + } + /* Sanity checking that setup requires legal number of resources */ if (num_pqs > RESC_NUM(p_hwfn, QED_PQ)) { DP_ERR(p_hwfn, @@ -212,12 +235,22 @@ static int qed_init_qm_info(struct qed_hwfn *p_hwfn, bool b_sleepable) vport_id = (u8)RESC_START(p_hwfn, QED_VPORT); + /* First init rate limited queues */ + for (curr_queue = 0; curr_queue < num_pf_rls; curr_queue++) { + qm_info->qm_pq_params[curr_queue].vport_id = vport_id++; + qm_info->qm_pq_params[curr_queue].tc_id = + p_hwfn->hw_info.non_offload_tc; + qm_info->qm_pq_params[curr_queue].wrr_group = 1; + qm_info->qm_pq_params[curr_queue].rl_valid = 1; + } + /* First init per-TC PQs */ for (i = 0; i < multi_cos_tcs; i++) { struct init_qm_pq_params *params = &qm_info->qm_pq_params[curr_queue++]; - if (p_hwfn->hw_info.personality == QED_PCI_ETH) { + if (p_hwfn->hw_info.personality == QED_PCI_ETH_ROCE || + p_hwfn->hw_info.personality == QED_PCI_ETH) { params->vport_id = vport_id; params->tc_id = p_hwfn->hw_info.non_offload_tc; params->wrr_group = 1; @@ -237,6 +270,32 @@ static int qed_init_qm_info(struct qed_hwfn *p_hwfn, bool b_sleepable) curr_queue++; qm_info->offload_pq = 0; + if (init_rdma_offload_pq) { + qm_info->offload_pq = curr_queue; + qm_info->qm_pq_params[curr_queue].vport_id = vport_id; + qm_info->qm_pq_params[curr_queue].tc_id = + p_hwfn->hw_info.offload_tc; + qm_info->qm_pq_params[curr_queue].wrr_group = 1; + curr_queue++; + } + + if (init_pure_ack_pq) { + qm_info->pure_ack_pq = curr_queue; + qm_info->qm_pq_params[curr_queue].vport_id = vport_id; + qm_info->qm_pq_params[curr_queue].tc_id = + p_hwfn->hw_info.offload_tc; + qm_info->qm_pq_params[curr_queue].wrr_group = 1; + curr_queue++; + } + + if (init_ooo_pq) { + qm_info->ooo_pq = curr_queue; + qm_info->qm_pq_params[curr_queue].vport_id = vport_id; + qm_info->qm_pq_params[curr_queue].tc_id = DCBX_ISCSI_OOO_TC; + qm_info->qm_pq_params[curr_queue].wrr_group = 1; + curr_queue++; + } + /* Then init per-VF PQs */ vf_offset = curr_queue; for (i = 0; i < num_vfs; i++) { @@ -371,21 +430,20 @@ int qed_resc_alloc(struct qed_dev *cdev) if (!p_hwfn->p_tx_cids) { DP_NOTICE(p_hwfn, "Failed to allocate memory for Tx Cids\n"); - rc = -ENOMEM; - goto alloc_err; + goto alloc_no_mem; } p_hwfn->p_rx_cids = kzalloc(rx_size, GFP_KERNEL); if (!p_hwfn->p_rx_cids) { DP_NOTICE(p_hwfn, "Failed to allocate memory for Rx Cids\n"); - rc = -ENOMEM; - goto alloc_err; + goto alloc_no_mem; } } for_each_hwfn(cdev, i) { struct qed_hwfn *p_hwfn = &cdev->hwfns[i]; + u32 n_eqes, num_cons; /* First allocate the context manager structure */ rc = qed_cxt_mngr_alloc(p_hwfn); @@ -434,18 +492,34 @@ int qed_resc_alloc(struct qed_dev *cdev) goto alloc_err; /* EQ */ - p_eq = qed_eq_alloc(p_hwfn, 256); - if (!p_eq) { - rc = -ENOMEM; + n_eqes = qed_chain_get_capacity(&p_hwfn->p_spq->chain); + if (p_hwfn->hw_info.personality == QED_PCI_ETH_ROCE) { + num_cons = qed_cxt_get_proto_cid_count(p_hwfn, + PROTOCOLID_ROCE, + 0) * 2; + n_eqes += num_cons + 2 * MAX_NUM_VFS_BB; + } else if (p_hwfn->hw_info.personality == QED_PCI_ISCSI) { + num_cons = + qed_cxt_get_proto_cid_count(p_hwfn, + PROTOCOLID_ISCSI, 0); + n_eqes += 2 * num_cons; + } + + if (n_eqes > 0xFFFF) { + DP_ERR(p_hwfn, + "Cannot allocate 0x%x EQ elements. The maximum of a u16 chain is 0x%x\n", + n_eqes, 0xFFFF); goto alloc_err; } + + p_eq = qed_eq_alloc(p_hwfn, (u16) n_eqes); + if (!p_eq) + goto alloc_no_mem; p_hwfn->p_eq = p_eq; p_consq = qed_consq_alloc(p_hwfn); - if (!p_consq) { - rc = -ENOMEM; - goto alloc_err; - } + if (!p_consq) + goto alloc_no_mem; p_hwfn->p_consq = p_consq; /* DMA info initialization */ @@ -474,6 +548,8 @@ int qed_resc_alloc(struct qed_dev *cdev) return 0; +alloc_no_mem: + rc = -ENOMEM; alloc_err: qed_resc_free(cdev); return rc; @@ -639,6 +715,7 @@ static int qed_hw_init_common(struct qed_hwfn *p_hwfn, struct qed_qm_info *qm_info = &p_hwfn->qm_info; struct qed_qm_common_rt_init_params params; struct qed_dev *cdev = p_hwfn->cdev; + u16 num_pfs, pf_id; u32 concrete_fid; int rc = 0; u8 vf_id; @@ -687,9 +764,16 @@ static int qed_hw_init_common(struct qed_hwfn *p_hwfn, qed_wr(p_hwfn, p_ptt, PSWRQ2_REG_L2P_VALIDATE_VFID, 0); qed_wr(p_hwfn, p_ptt, PGLUE_B_REG_USE_CLIENTID_IN_TAG, 1); - /* Disable relaxed ordering in the PCI config space */ - qed_wr(p_hwfn, p_ptt, 0x20b4, - qed_rd(p_hwfn, p_ptt, 0x20b4) & ~0x10); + if (QED_IS_BB(p_hwfn->cdev)) { + num_pfs = NUM_OF_ENG_PFS(p_hwfn->cdev); + for (pf_id = 0; pf_id < num_pfs; pf_id++) { + qed_fid_pretend(p_hwfn, p_ptt, pf_id); + qed_wr(p_hwfn, p_ptt, PRS_REG_SEARCH_ROCE, 0x0); + qed_wr(p_hwfn, p_ptt, PRS_REG_SEARCH_TCP, 0x0); + } + /* pretend to original PF */ + qed_fid_pretend(p_hwfn, p_ptt, p_hwfn->rel_pf_id); + } for (vf_id = 0; vf_id < MAX_NUM_VFS_BB; vf_id++) { concrete_fid = qed_vfid_to_concrete(p_hwfn, vf_id); @@ -779,7 +863,8 @@ static int qed_hw_init_pf(struct qed_hwfn *p_hwfn, } /* Protocl Configuration */ - STORE_RT_REG(p_hwfn, PRS_REG_SEARCH_TCP_RT_OFFSET, 0); + STORE_RT_REG(p_hwfn, PRS_REG_SEARCH_TCP_RT_OFFSET, + (p_hwfn->hw_info.personality == QED_PCI_ISCSI) ? 1 : 0); STORE_RT_REG(p_hwfn, PRS_REG_SEARCH_FCOE_RT_OFFSET, 0); STORE_RT_REG(p_hwfn, PRS_REG_SEARCH_ROCE_RT_OFFSET, 0); @@ -1256,8 +1341,9 @@ static void qed_hw_set_feat(struct qed_hwfn *p_hwfn) num_features); } -static void qed_hw_get_resc(struct qed_hwfn *p_hwfn) +static int qed_hw_get_resc(struct qed_hwfn *p_hwfn) { + u8 enabled_func_idx = p_hwfn->enabled_func_idx; u32 *resc_start = p_hwfn->hw_info.resc_start; u8 num_funcs = p_hwfn->num_funcs_on_engine; u32 *resc_num = p_hwfn->hw_info.resc_num; @@ -1281,14 +1367,22 @@ static void qed_hw_get_resc(struct qed_hwfn *p_hwfn) resc_num[QED_VPORT] = MAX_NUM_VPORTS_BB / num_funcs; resc_num[QED_RSS_ENG] = ETH_RSS_ENGINE_NUM_BB / num_funcs; resc_num[QED_PQ] = MAX_QM_TX_QUEUES_BB / num_funcs; - resc_num[QED_RL] = 8; + resc_num[QED_RL] = min_t(u32, 64, resc_num[QED_VPORT]); resc_num[QED_MAC] = ETH_NUM_MAC_FILTERS / num_funcs; resc_num[QED_VLAN] = (ETH_NUM_VLAN_FILTERS - 1 /*For vlan0*/) / num_funcs; - resc_num[QED_ILT] = 950; + resc_num[QED_ILT] = PXP_NUM_ILT_RECORDS_BB / num_funcs; for (i = 0; i < QED_MAX_RESC; i++) - resc_start[i] = resc_num[i] * p_hwfn->rel_pf_id; + resc_start[i] = resc_num[i] * enabled_func_idx; + + /* Sanity for ILT */ + if (RESC_END(p_hwfn, QED_ILT) > PXP_NUM_ILT_RECORDS_BB) { + DP_NOTICE(p_hwfn, "Can't assign ILT pages [%08x,...,%08x]\n", + RESC_START(p_hwfn, QED_ILT), + RESC_END(p_hwfn, QED_ILT) - 1); + return -EINVAL; + } qed_hw_set_feat(p_hwfn); @@ -1318,6 +1412,8 @@ static void qed_hw_get_resc(struct qed_hwfn *p_hwfn) p_hwfn->hw_info.resc_start[QED_VLAN], p_hwfn->hw_info.resc_num[QED_ILT], p_hwfn->hw_info.resc_start[QED_ILT]); + + return 0; } static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, @@ -1484,8 +1580,8 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, static void qed_get_num_funcs(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { - u32 reg_function_hide, tmp, eng_mask; - u8 num_funcs; + u8 num_funcs, enabled_func_idx = p_hwfn->rel_pf_id; + u32 reg_function_hide, tmp, eng_mask, low_pfs_mask; num_funcs = MAX_NUM_PFS_BB; @@ -1515,9 +1611,19 @@ static void qed_get_num_funcs(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) num_funcs++; tmp >>= 0x1; } + + /* Get the PF index within the enabled functions */ + low_pfs_mask = (0x1 << p_hwfn->abs_pf_id) - 1; + tmp = reg_function_hide & eng_mask & low_pfs_mask; + while (tmp) { + if (tmp & 0x1) + enabled_func_idx--; + tmp >>= 0x1; + } } p_hwfn->num_funcs_on_engine = num_funcs; + p_hwfn->enabled_func_idx = enabled_func_idx; DP_VERBOSE(p_hwfn, NETIF_MSG_PROBE, @@ -1587,9 +1693,7 @@ qed_get_hw_info(struct qed_hwfn *p_hwfn, qed_get_num_funcs(p_hwfn, p_ptt); - qed_hw_get_resc(p_hwfn); - - return rc; + return qed_hw_get_resc(p_hwfn); } static int qed_get_dev_info(struct qed_dev *cdev) diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c index 7363d2b..2693c30 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.c +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c @@ -791,16 +791,16 @@ qed_dmae_host2host(struct qed_hwfn *p_hwfn, } u16 qed_get_qm_pq(struct qed_hwfn *p_hwfn, - enum protocol_type proto, - union qed_qm_pq_params *p_params) + enum protocol_type proto, union qed_qm_pq_params *p_params) { u16 pq_id = 0; - if ((proto == PROTOCOLID_CORE || proto == PROTOCOLID_ETH) && - !p_params) { + if ((proto == PROTOCOLID_CORE || + proto == PROTOCOLID_ETH || + proto == PROTOCOLID_ISCSI || + proto == PROTOCOLID_ROCE) && !p_params) { DP_NOTICE(p_hwfn, - "Protocol %d received NULL PQ params\n", - proto); + "Protocol %d received NULL PQ params\n", proto); return 0; } @@ -808,6 +808,8 @@ u16 qed_get_qm_pq(struct qed_hwfn *p_hwfn, case PROTOCOLID_CORE: if (p_params->core.tc == LB_TC) pq_id = p_hwfn->qm_info.pure_lb_pq; + else if (p_params->core.tc == OOO_LB_TC) + pq_id = p_hwfn->qm_info.ooo_pq; else pq_id = p_hwfn->qm_info.offload_pq; break; @@ -817,6 +819,18 @@ u16 qed_get_qm_pq(struct qed_hwfn *p_hwfn, pq_id += p_hwfn->qm_info.vf_queues_offset + p_params->eth.vf_id; break; + case PROTOCOLID_ISCSI: + if (p_params->iscsi.q_idx == 1) + pq_id = p_hwfn->qm_info.pure_ack_pq; + break; + case PROTOCOLID_ROCE: + if (p_params->roce.dcqcn) + pq_id = p_params->roce.qpid; + else + pq_id = p_hwfn->qm_info.offload_pq; + if (pq_id > p_hwfn->qm_info.num_pf_rls) + pq_id = p_hwfn->qm_info.offload_pq; + break; default: pq_id = 0; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.h b/drivers/net/ethernet/qlogic/qed/qed_hw.h index 4367363..d015570 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.h @@ -254,6 +254,10 @@ void qed_dmae_info_free(struct qed_hwfn *p_hwfn); union qed_qm_pq_params { struct { + u8 q_idx; + } iscsi; + + struct { u8 tc; } core; @@ -262,11 +266,15 @@ union qed_qm_pq_params { u8 vf_id; u8 tc; } eth; + + struct { + u8 dcqcn; + u8 qpid; /* roce relative */ + } roce; }; u16 qed_get_qm_pq(struct qed_hwfn *p_hwfn, - enum protocol_type proto, - union qed_qm_pq_params *params); + enum protocol_type proto, union qed_qm_pq_params *params); int qed_init_fw_data(struct qed_dev *cdev, const u8 *fw_data); diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h index b889585..aa08ddb 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h +++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h @@ -27,6 +27,35 @@ #define CDU_REG_CID_ADDR_PARAMS_NCIB ( \ 0xff << 24) +#define CDU_REG_SEGMENT0_PARAMS \ + 0x580904UL +#define CDU_REG_SEGMENT0_PARAMS_T0_NUM_TIDS_IN_BLOCK \ + (0xfff << 0) +#define CDU_REG_SEGMENT0_PARAMS_T0_NUM_TIDS_IN_BLOCK_SHIFT \ + 0 +#define CDU_REG_SEGMENT0_PARAMS_T0_TID_BLOCK_WASTE \ + (0xff << 16) +#define CDU_REG_SEGMENT0_PARAMS_T0_TID_BLOCK_WASTE_SHIFT \ + 16 +#define CDU_REG_SEGMENT0_PARAMS_T0_TID_SIZE \ + (0xff << 24) +#define CDU_REG_SEGMENT0_PARAMS_T0_TID_SIZE_SHIFT \ + 24 +#define CDU_REG_SEGMENT1_PARAMS \ + 0x580908UL +#define CDU_REG_SEGMENT1_PARAMS_T1_NUM_TIDS_IN_BLOCK \ + (0xfff << 0) +#define CDU_REG_SEGMENT1_PARAMS_T1_NUM_TIDS_IN_BLOCK_SHIFT \ + 0 +#define CDU_REG_SEGMENT1_PARAMS_T1_TID_BLOCK_WASTE \ + (0xff << 16) +#define CDU_REG_SEGMENT1_PARAMS_T1_TID_BLOCK_WASTE_SHIFT \ + 16 +#define CDU_REG_SEGMENT1_PARAMS_T1_TID_SIZE \ + (0xff << 24) +#define CDU_REG_SEGMENT1_PARAMS_T1_TID_SIZE_SHIFT \ + 24 + #define XSDM_REG_OPERATION_GEN \ 0xf80408UL #define NIG_REG_RX_BRB_OUT_EN \ @@ -225,6 +254,8 @@ 0x1f0000UL #define PRS_REG_MSG_INFO \ 0x1f0a1cUL +#define PRS_REG_ROCE_DEST_QP_MAX_PF \ + 0x1f0430UL #define PSDM_REG_ENABLE_IN1 \ 0xfa0004UL #define PSEM_REG_ENABLE_IN \ @@ -233,6 +264,8 @@ 0x280020UL #define PSWRQ2_REG_CDUT_P_SIZE \ 0x24000cUL +#define PSWRQ2_REG_ILT_MEMORY \ + 0x260000UL #define PSWHST_REG_DISCARD_INTERNAL_WRITES \ 0x2a0040UL #define PSWHST2_REG_DBGSYN_ALMOST_FULL_THR \ -- cgit v0.10.2 From 81b1251d3b761b303955408dd6a49618b5683c2b Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Sat, 4 Jun 2016 08:20:16 +0300 Subject: qed: Fix next-ptr chains for BE / 32-bit Commit a91eb52abb50 ("qed: Revisit chain implementation") contains an incorrect implementation for BE platforms, as device's regpairs containing addresses are LE and they're not converted correctly when read back. In addition, it raises a compilation warning for 32-bit platforms where dma_addr_t is a 32-bit variable. Reported-by: kbuild test robot Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/include/linux/qed/qed_chain.h b/include/linux/qed/qed_chain.h index eceaa9e..7e441bd 100644 --- a/include/linux/qed/qed_chain.h +++ b/include/linux/qed/qed_chain.h @@ -25,10 +25,9 @@ } while (0) #define HILO_GEN(hi, lo, type) ((((type)(hi)) << 32) + (lo)) -#define HILO_DMA(hi, lo) HILO_GEN(hi, lo, dma_addr_t) #define HILO_64(hi, lo) HILO_GEN((le32_to_cpu(hi)), (le32_to_cpu(lo)), u64) -#define HILO_DMA_REGPAIR(regpair) (HILO_DMA(regpair.hi, regpair.lo)) #define HILO_64_REGPAIR(regpair) (HILO_64(regpair.hi, regpair.lo)) +#define HILO_DMA_REGPAIR(regpair) ((dma_addr_t)HILO_64_REGPAIR(regpair)) enum qed_chain_mode { /* Each Page contains a next pointer at its end */ -- cgit v0.10.2 From 3b55a537d028ccc3e423e74a2037476318918341 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 3 Jun 2016 22:53:26 -0700 Subject: sctp: Fix warning in sctp_packet_transmit_chunk() size_t objects should be printed with %Z printf format. Reported-by: kbuild test robot Signed-off-by: David S. Miller diff --git a/net/sctp/output.c b/net/sctp/output.c index 90d2e12..1541a91 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -182,7 +182,7 @@ sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet, sctp_xmit_t retval; int error = 0; - pr_debug("%s: packet:%p size:%lu chunk:%p size:%d\n", __func__, + pr_debug("%s: packet:%p size:%Zu chunk:%p size:%d\n", __func__, packet, packet->size, chunk, chunk->skb ? chunk->skb->len : -1); switch ((retval = (sctp_packet_append_chunk(packet, chunk)))) { -- cgit v0.10.2 From 76f21b99004ef1f16be6184678f660eab911b8b8 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 3 Jun 2016 22:56:28 -0700 Subject: net: Add docbook description for 'mtu' arg to skb_gso_validate_mtu() Signed-off-by: David S. Miller diff --git a/net/core/skbuff.c b/net/core/skbuff.c index b6e0f95..e7ec6d3 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -4399,6 +4399,7 @@ EXPORT_SYMBOL_GPL(skb_gso_transport_seglen); * skb_gso_validate_mtu - Return in case such skb fits a given MTU * * @skb: GSO skb + * @mtu: MTU to validate against * * skb_gso_validate_mtu validates if a given skb will fit a wanted MTU * once split. -- cgit v0.10.2 From 83e41e77819b642a084f5e9e54ba4ad5d4e17466 Mon Sep 17 00:00:00 2001 From: Lauri Kasanen Date: Sat, 16 Apr 2016 17:18:56 +0300 Subject: carl9170: Clarify kconfig text The previous text was confusing, leading readers to think this driver was a duplicate, and so didn't need to be enabled. After the removal of the older staging driver, this is the only driver in mainline for these devices. Signed-off-by: Lauri Kasanen Acked-by: Christian Lamparter Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/carl9170/Kconfig b/drivers/net/wireless/ath/carl9170/Kconfig index 1a796e5..2e34bae 100644 --- a/drivers/net/wireless/ath/carl9170/Kconfig +++ b/drivers/net/wireless/ath/carl9170/Kconfig @@ -5,12 +5,10 @@ config CARL9170 select FW_LOADER select CRC32 help - This is another driver for the Atheros "otus" 802.11n USB devices. + This is the mainline driver for the Atheros "otus" 802.11n USB devices. - This driver provides more features than the original, - but it needs a special firmware (carl9170-1.fw) to do that. - - The firmware can be downloaded from our wiki here: + It needs a special firmware (carl9170-1.fw), which can be downloaded + from our wiki here: If you choose to build a module, it'll be called carl9170. -- cgit v0.10.2 From b88a2e80396ba463a4800c62c96e86954cb0f4f7 Mon Sep 17 00:00:00 2001 From: Christian Daudt Date: Wed, 11 May 2016 15:06:48 -0700 Subject: brcmfmac: Fix kernel oops in failed chip_attach When chip attach fails, brcmf_sdiod_intr_unregister is being called but that is too early as sdiodev->settings has not been set yet nor has brcmf_sdiod_intr_register been called. Change to use oob_irq_requested + newly created sd_irq_requested to decide on what to unregister at intr_unregister time. Steps to reproduce problem: - modprobe brcmfmac using buggy FW - rmmod brcmfmac - modprobe brcmfmac again. If done with a buggy firmware, brcm_chip_attach will fail on the 2nd modprobe triggering the call to intr_unregister and the kernel oops when attempting to de-reference sdiodev->settings->bus.sdio which has not yet been set. Signed-off-by: Christian Daudt Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c index c7550da..7abcf67 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c @@ -166,6 +166,7 @@ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev) sdio_claim_irq(sdiodev->func[1], brcmf_sdiod_ib_irqhandler); sdio_claim_irq(sdiodev->func[2], brcmf_sdiod_dummy_irqhandler); sdio_release_host(sdiodev->func[1]); + sdiodev->sd_irq_requested = true; } return 0; @@ -173,27 +174,30 @@ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev) int brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev) { - struct brcmfmac_sdio_pd *pdata; - brcmf_dbg(SDIO, "Entering\n"); + brcmf_dbg(SDIO, "Entering oob=%d sd=%d\n", + sdiodev->oob_irq_requested, + sdiodev->sd_irq_requested); - pdata = &sdiodev->settings->bus.sdio; - if (pdata->oob_irq_supported) { + if (sdiodev->oob_irq_requested) { + struct brcmfmac_sdio_pd *pdata; + + pdata = &sdiodev->settings->bus.sdio; sdio_claim_host(sdiodev->func[1]); brcmf_sdiod_regwb(sdiodev, SDIO_CCCR_BRCM_SEPINT, 0, NULL); brcmf_sdiod_regwb(sdiodev, SDIO_CCCR_IENx, 0, NULL); sdio_release_host(sdiodev->func[1]); - if (sdiodev->oob_irq_requested) { - sdiodev->oob_irq_requested = false; - if (sdiodev->irq_wake) { - disable_irq_wake(pdata->oob_irq_nr); - sdiodev->irq_wake = false; - } - free_irq(pdata->oob_irq_nr, &sdiodev->func[1]->dev); - sdiodev->irq_en = false; + sdiodev->oob_irq_requested = false; + if (sdiodev->irq_wake) { + disable_irq_wake(pdata->oob_irq_nr); + sdiodev->irq_wake = false; } - } else { + free_irq(pdata->oob_irq_nr, &sdiodev->func[1]->dev); + sdiodev->irq_en = false; + } + + if (sdiodev->sd_irq_requested) { sdio_claim_host(sdiodev->func[1]); sdio_release_irq(sdiodev->func[2]); sdio_release_irq(sdiodev->func[1]); diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h index dcf0ce8..c07ad25 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h @@ -186,6 +186,7 @@ struct brcmf_sdio_dev { struct brcmf_bus *bus_if; struct brcmf_mp_device *settings; bool oob_irq_requested; + bool sd_irq_requested; bool irq_en; /* irq enable flags */ spinlock_t irq_en_lock; bool irq_wake; /* irq wake enable flags */ -- cgit v0.10.2 From b746740147dcfe1eb635f76f0a73f7df2b3e37c0 Mon Sep 17 00:00:00 2001 From: Christian Daudt Date: Wed, 11 May 2016 15:06:49 -0700 Subject: brcmfmac: Fix 'did not remove int handler' warning brcmf_sdiod_intr_unregister call that removes both func1 and func2 interrupt handlers only called when brcmf_ops_sdio_remove is called for func 1 (which is the 2nd call) but sdio is expecting it to be removed at the end of each sdio_remove call. This is causing 'rmmod bcmrfmac' on a 4356-sdio chip to complain with: WARNING: driver brcmfmac did not remove its interrupt handler! The modification makes calling brcmf_sdiod_intr_unregister multiple times harmless by clearing the variables that track if interrupt handlers have been installed, and then calls it on every brcmf_ops_sdio_remove call instead of just remove for func 1. Signed-off-by: Christian Daudt Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c index 7abcf67..c4b89d2 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c @@ -172,7 +172,7 @@ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev) return 0; } -int brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev) +void brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev) { brcmf_dbg(SDIO, "Entering oob=%d sd=%d\n", @@ -195,6 +195,7 @@ int brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev) } free_irq(pdata->oob_irq_nr, &sdiodev->func[1]->dev); sdiodev->irq_en = false; + sdiodev->oob_irq_requested = false; } if (sdiodev->sd_irq_requested) { @@ -202,9 +203,8 @@ int brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev) sdio_release_irq(sdiodev->func[2]); sdio_release_irq(sdiodev->func[1]); sdio_release_host(sdiodev->func[1]); + sdiodev->sd_irq_requested = false; } - - return 0; } void brcmf_sdiod_change_state(struct brcmf_sdio_dev *sdiodev, @@ -1201,12 +1201,17 @@ static void brcmf_ops_sdio_remove(struct sdio_func *func) brcmf_dbg(SDIO, "sdio device ID: 0x%04x\n", func->device); brcmf_dbg(SDIO, "Function: %d\n", func->num); - if (func->num != 1) - return; - bus_if = dev_get_drvdata(&func->dev); if (bus_if) { sdiodev = bus_if->bus_priv.sdio; + + /* start by unregistering irqs */ + brcmf_sdiod_intr_unregister(sdiodev); + + if (func->num != 1) + return; + + /* only proceed with rest of cleanup if func 1 */ brcmf_sdiod_remove(sdiodev); dev_set_drvdata(&sdiodev->func[1]->dev, NULL); diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h index c07ad25..f3da32f 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h @@ -294,7 +294,7 @@ struct sdpcmd_regs { /* Register/deregister interrupt handler. */ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev *sdiodev); -int brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev); +void brcmf_sdiod_intr_unregister(struct brcmf_sdio_dev *sdiodev); /* sdio device register access interface */ u8 brcmf_sdiod_regrb(struct brcmf_sdio_dev *sdiodev, u32 addr, int *ret); -- cgit v0.10.2 From d0b03439f7845ad9a72cc0f0cbb1ed34a5de5a08 Mon Sep 17 00:00:00 2001 From: Adrian Chadd Date: Mon, 16 May 2016 20:25:34 -0700 Subject: b43: don't unconditionally fall back to CCK if the rate is 6MB OFDM. Check the current PHY operating mode (gmode) to see if we should fall back from 6MB OFDM to 11MB CCK. For 5GHz operation this isn't allowed. Note, the fallback lookup is only done for RTS rates; normal fallback rates are done via mac80211 and aren't affected by this change. Signed-off-by: Adrian Chadd Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/b43/xmit.c b/drivers/net/wireless/broadcom/b43/xmit.c index f620126..7edbcdb 100644 --- a/drivers/net/wireless/broadcom/b43/xmit.c +++ b/drivers/net/wireless/broadcom/b43/xmit.c @@ -205,7 +205,7 @@ static u16 b43_generate_tx_phy_ctl1(struct b43_wldev *dev, u8 bitrate) return control; } -static u8 b43_calc_fallback_rate(u8 bitrate) +static u8 b43_calc_fallback_rate(u8 bitrate, int gmode) { switch (bitrate) { case B43_CCK_RATE_1MB: @@ -216,8 +216,15 @@ static u8 b43_calc_fallback_rate(u8 bitrate) return B43_CCK_RATE_2MB; case B43_CCK_RATE_11MB: return B43_CCK_RATE_5MB; + /* + * Don't just fallback to CCK; it may be in 5GHz operation + * and falling back to CCK won't work out very well. + */ case B43_OFDM_RATE_6MB: - return B43_CCK_RATE_5MB; + if (gmode) + return B43_CCK_RATE_5MB; + else + return B43_OFDM_RATE_6MB; case B43_OFDM_RATE_9MB: return B43_OFDM_RATE_6MB; case B43_OFDM_RATE_12MB: @@ -438,7 +445,7 @@ int b43_generate_txhdr(struct b43_wldev *dev, rts_rate = rts_cts_rate ? rts_cts_rate->hw_value : B43_CCK_RATE_1MB; rts_rate_ofdm = b43_is_ofdm_rate(rts_rate); - rts_rate_fb = b43_calc_fallback_rate(rts_rate); + rts_rate_fb = b43_calc_fallback_rate(rts_rate, phy->gmode); rts_rate_fb_ofdm = b43_is_ofdm_rate(rts_rate_fb); if (rates[0].flags & IEEE80211_TX_RC_USE_CTS_PROTECT) { -- cgit v0.10.2 From 47ce90f9f08ae282dee4beecc5c736a1e41f61c1 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Tue, 17 May 2016 16:38:46 +0200 Subject: mwifiex: fix typo firmare -> firmware Signed-off-by: Julia Lawall Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c index 0c7937e..9246ce8 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.c +++ b/drivers/net/wireless/marvell/mwifiex/pcie.c @@ -2804,7 +2804,7 @@ static int mwifiex_pcie_request_irq(struct mwifiex_adapter *adapter) } /* - * This function get firmare name for downloading by revision id + * This function gets the firmware name for downloading by revision id * * Read revision id register to get revision id */ -- cgit v0.10.2 From 0e5760440842eee57ff573251e95289e2ee7b15f Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:16:52 +0200 Subject: net: dsa: slave: chip data is optional, don't dereference NULL The new binding does not make use of dsa_chip_data, a.k.a cd. When retrieving the size of the EEPROM attached to a switch, don't assume there is a cd attached to the switch structure. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 152436c..135a917 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -615,7 +615,7 @@ static int dsa_slave_get_eeprom_len(struct net_device *dev) struct dsa_slave_priv *p = netdev_priv(dev); struct dsa_switch *ds = p->parent; - if (ds->cd->eeprom_len) + if (ds->cd && ds->cd->eeprom_len) return ds->cd->eeprom_len; if (ds->drv->get_eeprom_len) -- cgit v0.10.2 From 762eb67bc2dd704e6a9c4fe44b6b0d72dcc5642e Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Sat, 4 Jun 2016 21:16:54 +0200 Subject: net: dsa: mv88e6xxx: fix circular lock in PPU work Lock debugging shows that there is a possible circular lock in the PPU work code. Switch the lock order of smi_mutex and ppu_mutex to fix this. Here's the full trace: [ 4.341325] ====================================================== [ 4.347519] [ INFO: possible circular locking dependency detected ] [ 4.353800] 4.6.0 #4 Not tainted [ 4.357039] ------------------------------------------------------- [ 4.363315] kworker/0:1/328 is trying to acquire lock: [ 4.368463] (&ps->smi_mutex){+.+.+.}, at: [<8049c758>] mv88e6xxx_reg_read+0x30/0x54 [ 4.376313] [ 4.376313] but task is already holding lock: [ 4.382160] (&ps->ppu_mutex){+.+...}, at: [<8049cac0>] mv88e6xxx_ppu_reenable_work+0x28/0xd4 [ 4.390772] [ 4.390772] which lock already depends on the new lock. [ 4.390772] [ 4.398963] [ 4.398963] the existing dependency chain (in reverse order) is: [ 4.406461] [ 4.406461] -> #1 (&ps->ppu_mutex){+.+...}: [ 4.410897] [<806d86bc>] mutex_lock_nested+0x54/0x360 [ 4.416606] [<8049a800>] mv88e6xxx_ppu_access_get+0x28/0x100 [ 4.422906] [<8049b778>] mv88e6xxx_phy_read+0x90/0xdc [ 4.428599] [<806a4534>] dsa_slave_phy_read+0x3c/0x40 [ 4.434300] [<804943ec>] mdiobus_read+0x68/0x80 [ 4.439481] [<804939d4>] get_phy_device+0x58/0x1d8 [ 4.444914] [<80493ed0>] mdiobus_scan+0x24/0xf4 [ 4.450078] [<8049409c>] __mdiobus_register+0xfc/0x1ac [ 4.455857] [<806a40b0>] dsa_probe+0x860/0xca8 [ 4.460934] [<8043246c>] platform_drv_probe+0x5c/0xc0 [ 4.466627] [<804305a0>] driver_probe_device+0x118/0x450 [ 4.472589] [<80430b00>] __device_attach_driver+0xac/0x128 [ 4.478724] [<8042e350>] bus_for_each_drv+0x74/0xa8 [ 4.484235] [<804302d8>] __device_attach+0xc4/0x154 [ 4.489755] [<80430cec>] device_initial_probe+0x1c/0x20 [ 4.495612] [<8042f620>] bus_probe_device+0x98/0xa0 [ 4.501123] [<8042fbd0>] deferred_probe_work_func+0x4c/0xd4 [ 4.507328] [<8013a794>] process_one_work+0x1a8/0x604 [ 4.513030] [<8013ac54>] worker_thread+0x64/0x528 [ 4.518367] [<801409e8>] kthread+0xec/0x100 [ 4.523201] [<80108f30>] ret_from_fork+0x14/0x24 [ 4.528462] [ 4.528462] -> #0 (&ps->smi_mutex){+.+.+.}: [ 4.532895] [<8015ad5c>] lock_acquire+0xb4/0x1dc [ 4.538154] [<806d86bc>] mutex_lock_nested+0x54/0x360 [ 4.543856] [<8049c758>] mv88e6xxx_reg_read+0x30/0x54 [ 4.549549] [<8049cad8>] mv88e6xxx_ppu_reenable_work+0x40/0xd4 [ 4.556022] [<8013a794>] process_one_work+0x1a8/0x604 [ 4.561707] [<8013ac54>] worker_thread+0x64/0x528 [ 4.567053] [<801409e8>] kthread+0xec/0x100 [ 4.571878] [<80108f30>] ret_from_fork+0x14/0x24 [ 4.577139] [ 4.577139] other info that might help us debug this: [ 4.577139] [ 4.585159] Possible unsafe locking scenario: [ 4.585159] [ 4.591093] CPU0 CPU1 [ 4.595631] ---- ---- [ 4.600169] lock(&ps->ppu_mutex); [ 4.603693] lock(&ps->smi_mutex); [ 4.609742] lock(&ps->ppu_mutex); [ 4.615790] lock(&ps->smi_mutex); [ 4.619314] [ 4.619314] *** DEADLOCK *** [ 4.619314] [ 4.625256] 3 locks held by kworker/0:1/328: [ 4.629537] #0: ("events"){.+.+..}, at: [<8013a704>] process_one_work+0x118/0x604 [ 4.637288] #1: ((&ps->ppu_work)){+.+...}, at: [<8013a704>] process_one_work+0x118/0x604 [ 4.645653] #2: (&ps->ppu_mutex){+.+...}, at: [<8049cac0>] mv88e6xxx_ppu_reenable_work+0x28/0xd4 [ 4.654714] [ 4.654714] stack backtrace: [ 4.659098] CPU: 0 PID: 328 Comm: kworker/0:1 Not tainted 4.6.0 #4 [ 4.665286] Hardware name: Freescale Vybrid VF5xx/VF6xx (Device Tree) [ 4.671748] Workqueue: events mv88e6xxx_ppu_reenable_work [ 4.677174] Backtrace: [ 4.679674] [<8010d354>] (dump_backtrace) from [<8010d5a0>] (show_stack+0x20/0x24) [ 4.687252] r6:80fb3c88 r5:80fb3c88 r4:80fb4728 r3:00000002 [ 4.693003] [<8010d580>] (show_stack) from [<803b45e8>] (dump_stack+0x24/0x28) [ 4.700246] [<803b45c4>] (dump_stack) from [<80157398>] (print_circular_bug+0x208/0x32c) [ 4.708361] [<80157190>] (print_circular_bug) from [<8015a630>] (__lock_acquire+0x185c/0x1b80) [ 4.716982] r10:9ec22a00 r9:00000060 r8:8164b6bc r7:00000040 r6:00000003 r5:8163a5b4 [ 4.724905] r4:00000003 r3:9ec22de8 [ 4.728537] [<80158dd4>] (__lock_acquire) from [<8015ad5c>] (lock_acquire+0xb4/0x1dc) [ 4.736378] r10:60000013 r9:00000000 r8:00000000 r7:00000000 r6:9e5e9c50 r5:80e618e0 [ 4.744301] r4:00000000 [ 4.746879] [<8015aca8>] (lock_acquire) from [<806d86bc>] (mutex_lock_nested+0x54/0x360) [ 4.754976] r10:9e5e9c1c r9:80e616c4 r8:9f685ea0 r7:0000001b r6:9ec22a00 r5:8163a5b4 [ 4.762899] r4:9e5e9c1c [ 4.765477] [<806d8668>] (mutex_lock_nested) from [<8049c758>] (mv88e6xxx_reg_read+0x30/0x54) [ 4.774008] r10:80e60c5b r9:80e616c4 r8:9f685ea0 r7:0000001b r6:00000004 r5:9e5e9c10 [ 4.781930] r4:9e5e9c1c [ 4.784507] [<8049c728>] (mv88e6xxx_reg_read) from [<8049cad8>] (mv88e6xxx_ppu_reenable_work+0x40/0xd4) [ 4.793907] r7:9ffd5400 r6:9e5e9c68 r5:9e5e9cb0 r4:9e5e9c10 [ 4.799659] [<8049ca98>] (mv88e6xxx_ppu_reenable_work) from [<8013a794>] (process_one_work+0x1a8/0x604) [ 4.809059] r9:80e616c4 r8:9f685ea0 r7:9ffd5400 r6:80e0a1c8 r5:9f5f2e80 r4:9e5e9cb0 [ 4.816910] [<8013a5ec>] (process_one_work) from [<8013ac54>] (worker_thread+0x64/0x528) [ 4.825010] r10:9f5f2e80 r9:00000008 r8:80e0dc80 r7:80e0a1fc r6:80e0a1c8 r5:9f5f2e98 [ 4.832933] r4:80e0a1c8 [ 4.835510] [<8013abf0>] (worker_thread) from [<801409e8>] (kthread+0xec/0x100) [ 4.842827] r10:00000000 r9:00000000 r8:00000000 r7:8013abf0 r6:9f5f2e80 r5:9ec15740 [ 4.850749] r4:00000000 [ 4.853327] [<801408fc>] (kthread) from [<80108f30>] (ret_from_fork+0x14/0x24) [ 4.860557] r7:00000000 r6:00000000 r5:801408fc r4:9ec15740 Signed-off-by: Vivien Didelot Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index ba9dfc9..c361036 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -288,18 +288,18 @@ static int mv88e6xxx_ppu_enable(struct mv88e6xxx_priv_state *ps) int ret, err; unsigned long timeout; - ret = mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_CONTROL); + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_CONTROL); if (ret < 0) return ret; - err = mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, - ret | GLOBAL_CONTROL_PPU_ENABLE); + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, + ret | GLOBAL_CONTROL_PPU_ENABLE); if (err) return err; timeout = jiffies + 1 * HZ; while (time_before(jiffies, timeout)) { - ret = mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATUS); + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATUS); if (ret < 0) return ret; @@ -317,11 +317,16 @@ static void mv88e6xxx_ppu_reenable_work(struct work_struct *ugly) struct mv88e6xxx_priv_state *ps; ps = container_of(ugly, struct mv88e6xxx_priv_state, ppu_work); + + mutex_lock(&ps->smi_mutex); + if (mutex_trylock(&ps->ppu_mutex)) { if (mv88e6xxx_ppu_enable(ps) == 0) ps->ppu_disabled = 0; mutex_unlock(&ps->ppu_mutex); } + + mutex_unlock(&ps->smi_mutex); } static void mv88e6xxx_ppu_reenable_timer(unsigned long _ps) -- cgit v0.10.2 From 6e8e862ded41bd966b088960ad6f4dc7a1a3ce36 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:16:55 +0200 Subject: net: dsa: slave: Remove MDIO address from switch MDIO bus name The DSA layer should no longer assume the switch is connected to an MDIO bus. As a result, we cannot use the address on the MDIO bus when forming the name of the switches internal MDIO bus for its builtin and possibly external PHYs. The switch index is sufficient to make the name unique, so drop the MDIO address. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 135a917..f640a48 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -49,8 +49,7 @@ void dsa_slave_mii_bus_init(struct dsa_switch *ds) ds->slave_mii_bus->name = "dsa slave smi"; ds->slave_mii_bus->read = dsa_slave_phy_read; ds->slave_mii_bus->write = dsa_slave_phy_write; - snprintf(ds->slave_mii_bus->id, MII_BUS_ID_SIZE, "dsa-%d:%.2x", - ds->index, ds->cd->sw_addr); + snprintf(ds->slave_mii_bus->id, MII_BUS_ID_SIZE, "dsa-%d", ds->index); ds->slave_mii_bus->parent = ds->dev; ds->slave_mii_bus->phy_mask = ~ds->phys_mii_mask; } -- cgit v0.10.2 From 149cafd790573294752787f5ce1dc0f99e4088d3 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:16:56 +0200 Subject: net: dsa: tag_{e}dsa.c: Remove dependency on platform data The platform data nr_chips is used when validating a received packet, to ensure it comes from a know switch chip. The number of possible switches is limited to DSA_MAX_SWITCHES, so use this as the first validation step. The new binding allows holes in the dst->ds[] array, so also ensure ensure there is a valid dsa_switch for this packet. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/net/dsa/tag_dsa.c b/net/dsa/tag_dsa.c index aa780e4..f9832f0 100644 --- a/net/dsa/tag_dsa.c +++ b/net/dsa/tag_dsa.c @@ -107,9 +107,13 @@ static int dsa_rcv(struct sk_buff *skb, struct net_device *dev, * Check that the source device exists and that the source * port is a registered DSA port. */ - if (source_device >= dst->pd->nr_chips) + if (source_device >= DSA_MAX_SWITCHES) goto out_drop; + ds = dst->ds[source_device]; + if (!ds) + goto out_drop; + if (source_port >= DSA_MAX_PORTS || ds->ports[source_port] == NULL) goto out_drop; diff --git a/net/dsa/tag_edsa.c b/net/dsa/tag_edsa.c index 2288c80..3890aac 100644 --- a/net/dsa/tag_edsa.c +++ b/net/dsa/tag_edsa.c @@ -120,9 +120,13 @@ static int edsa_rcv(struct sk_buff *skb, struct net_device *dev, * Check that the source device exists and that the source * port is a registered DSA port. */ - if (source_device >= dst->pd->nr_chips) + if (source_device >= DSA_MAX_SWITCHES) goto out_drop; + ds = dst->ds[source_device]; + if (!ds) + goto out_drop; + if (source_port >= DSA_MAX_PORTS || ds->ports[source_port] == NULL) goto out_drop; -- cgit v0.10.2 From c8b098086b4c744084350d2757a637ad756adf34 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:16:57 +0200 Subject: net: dsa: Add a ports structure and use it in the switch structure There are going to be more per-port members added to the switch structure. So add a port structure and move the netdev into it. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index 10ddd5a..73df91bb 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -804,7 +804,7 @@ static int bcm_sf2_sw_fdb_dump(struct dsa_switch *ds, int port, int (*cb)(struct switchdev_obj *obj)) { struct bcm_sf2_priv *priv = ds_to_priv(ds); - struct net_device *dev = ds->ports[port]; + struct net_device *dev = ds->ports[port].netdev; struct bcm_sf2_arl_entry results[2]; unsigned int count = 0; int ret; @@ -1248,7 +1248,7 @@ static void bcm_sf2_sw_fixed_link_update(struct dsa_switch *ds, int port, * state machine and make it go in PHY_FORCING state instead. */ if (!status->link) - netif_carrier_off(ds->ports[port]); + netif_carrier_off(ds->ports[port].netdev); status->duplex = 1; } else { status->link = 1; diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index c361036..85332d9 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -1327,7 +1327,7 @@ static int _mv88e6xxx_port_state(struct mv88e6xxx_priv_state *ps, int port, if (ret) return ret; - netdev_dbg(ds->ports[port], "PortState %s (was %s)\n", + netdev_dbg(ds->ports[port].netdev, "PortState %s (was %s)\n", mv88e6xxx_port_state_names[state], mv88e6xxx_port_state_names[oldstate]); } @@ -1405,7 +1405,8 @@ static void mv88e6xxx_port_stp_state_set(struct dsa_switch *ds, int port, mutex_unlock(&ps->smi_mutex); if (err) - netdev_err(ds->ports[port], "failed to update state to %s\n", + netdev_err(ds->ports[port].netdev, + "failed to update state to %s\n", mv88e6xxx_port_state_names[stp_state]); } @@ -1431,8 +1432,8 @@ static int _mv88e6xxx_port_pvid(struct mv88e6xxx_priv_state *ps, int port, if (ret < 0) return ret; - netdev_dbg(ds->ports[port], "DefaultVID %d (was %d)\n", *new, - pvid); + netdev_dbg(ds->ports[port].netdev, + "DefaultVID %d (was %d)\n", *new, pvid); } if (old) @@ -1847,7 +1848,8 @@ static int _mv88e6xxx_port_fid(struct mv88e6xxx_priv_state *ps, int port, if (ret < 0) return ret; - netdev_dbg(ds->ports[port], "FID %d (was %d)\n", *new, fid); + netdev_dbg(ds->ports[port].netdev, + "FID %d (was %d)\n", *new, fid); } if (old) @@ -2028,7 +2030,7 @@ static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port, ps->ports[port].bridge_dev) break; /* same bridge, check next VLAN */ - netdev_warn(ds->ports[port], + netdev_warn(ds->ports[port].netdev, "hardware VLAN %d already used by %s\n", vlan.vid, netdev_name(ps->ports[i].bridge_dev)); @@ -2078,7 +2080,7 @@ static int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port, if (ret < 0) goto unlock; - netdev_dbg(ds->ports[port], "802.1Q Mode %s (was %s)\n", + netdev_dbg(ds->ports[port].netdev, "802.1Q Mode %s (was %s)\n", mv88e6xxx_port_8021q_mode_names[new], mv88e6xxx_port_8021q_mode_names[old]); } @@ -2147,11 +2149,12 @@ static void mv88e6xxx_port_vlan_add(struct dsa_switch *ds, int port, for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) if (_mv88e6xxx_port_vlan_add(ps, port, vid, untagged)) - netdev_err(ds->ports[port], "failed to add VLAN %d%c\n", + netdev_err(ds->ports[port].netdev, + "failed to add VLAN %d%c\n", vid, untagged ? 'u' : 't'); if (pvid && _mv88e6xxx_port_pvid_set(ps, port, vlan->vid_end)) - netdev_err(ds->ports[port], "failed to set PVID %d\n", + netdev_err(ds->ports[port].netdev, "failed to set PVID %d\n", vlan->vid_end); mutex_unlock(&ps->smi_mutex); @@ -2336,7 +2339,8 @@ static void mv88e6xxx_port_fdb_add(struct dsa_switch *ds, int port, mutex_lock(&ps->smi_mutex); if (_mv88e6xxx_port_fdb_load(ps, port, fdb->addr, fdb->vid, state)) - netdev_err(ds->ports[port], "failed to load MAC address\n"); + netdev_err(ds->ports[port].netdev, + "failed to load MAC address\n"); mutex_unlock(&ps->smi_mutex); } @@ -2537,7 +2541,8 @@ static void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port) for (i = 0; i < ps->info->num_ports; ++i) if (i == port || ps->ports[i].bridge_dev == bridge) if (_mv88e6xxx_port_based_vlan_map(ps, i)) - netdev_warn(ds->ports[i], "failed to remap\n"); + netdev_warn(ds->ports[i].netdev, + "failed to remap\n"); mutex_unlock(&ps->smi_mutex); } diff --git a/include/net/dsa.h b/include/net/dsa.h index 17c3d37..9aed857 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -119,6 +119,10 @@ struct dsa_switch_tree { struct dsa_switch *ds[DSA_MAX_SWITCHES]; }; +struct dsa_port { + struct net_device *netdev; +}; + struct dsa_switch { struct device *dev; @@ -158,8 +162,8 @@ struct dsa_switch { u32 dsa_port_mask; u32 enabled_port_mask; u32 phys_mii_mask; + struct dsa_port ports[DSA_MAX_PORTS]; struct mii_bus *slave_mii_bus; - struct net_device *ports[DSA_MAX_PORTS]; }; static inline bool dsa_is_cpu_port(struct dsa_switch *ds, int p) @@ -174,7 +178,7 @@ static inline bool dsa_is_dsa_port(struct dsa_switch *ds, int p) static inline bool dsa_is_port_initialized(struct dsa_switch *ds, int p) { - return ds->enabled_port_mask & (1 << p) && ds->ports[p]; + return ds->enabled_port_mask & (1 << p) && ds->ports[p].netdev; } static inline u8 dsa_upstream_port(struct dsa_switch *ds) diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index eff5dfc..18086e0 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -437,10 +437,10 @@ static void dsa_switch_destroy(struct dsa_switch *ds) if (!(ds->enabled_port_mask & (1 << port))) continue; - if (!ds->ports[port]) + if (!ds->ports[port].netdev) continue; - dsa_slave_destroy(ds->ports[port]); + dsa_slave_destroy(ds->ports[port].netdev); } /* Remove any fixed link PHYs */ @@ -469,7 +469,7 @@ static int dsa_switch_suspend(struct dsa_switch *ds) if (!dsa_is_port_initialized(ds, i)) continue; - ret = dsa_slave_suspend(ds->ports[i]); + ret = dsa_slave_suspend(ds->ports[i].netdev); if (ret) return ret; } @@ -495,7 +495,7 @@ static int dsa_switch_resume(struct dsa_switch *ds) if (!dsa_is_port_initialized(ds, i)) continue; - ret = dsa_slave_resume(ds->ports[i]); + ret = dsa_slave_resume(ds->ports[i].netdev); if (ret) return ret; } diff --git a/net/dsa/slave.c b/net/dsa/slave.c index f640a48..169abac 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -1183,12 +1183,12 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent, p->old_link = -1; p->old_duplex = -1; - ds->ports[port] = slave_dev; + ds->ports[port].netdev = slave_dev; ret = register_netdev(slave_dev); if (ret) { netdev_err(master, "error %d registering interface %s\n", ret, slave_dev->name); - ds->ports[port] = NULL; + ds->ports[port].netdev = NULL; free_netdev(slave_dev); return ret; } diff --git a/net/dsa/tag_brcm.c b/net/dsa/tag_brcm.c index e2aadb7..21bffde 100644 --- a/net/dsa/tag_brcm.c +++ b/net/dsa/tag_brcm.c @@ -127,7 +127,7 @@ static int brcm_tag_rcv(struct sk_buff *skb, struct net_device *dev, source_port = brcm_tag[3] & BRCM_EG_PID_MASK; /* Validate port against switch setup, either the port is totally */ - if (source_port >= DSA_MAX_PORTS || ds->ports[source_port] == NULL) + if (source_port >= DSA_MAX_PORTS || !ds->ports[source_port].netdev) goto out_drop; /* Remove Broadcom tag and update checksum */ @@ -140,7 +140,7 @@ static int brcm_tag_rcv(struct sk_buff *skb, struct net_device *dev, skb_push(skb, ETH_HLEN); skb->pkt_type = PACKET_HOST; - skb->dev = ds->ports[source_port]; + skb->dev = ds->ports[source_port].netdev; skb->protocol = eth_type_trans(skb, skb->dev); skb->dev->stats.rx_packets++; diff --git a/net/dsa/tag_dsa.c b/net/dsa/tag_dsa.c index f9832f0..bce79ff 100644 --- a/net/dsa/tag_dsa.c +++ b/net/dsa/tag_dsa.c @@ -114,7 +114,7 @@ static int dsa_rcv(struct sk_buff *skb, struct net_device *dev, if (!ds) goto out_drop; - if (source_port >= DSA_MAX_PORTS || ds->ports[source_port] == NULL) + if (source_port >= DSA_MAX_PORTS || !ds->ports[source_port].netdev) goto out_drop; /* @@ -163,7 +163,7 @@ static int dsa_rcv(struct sk_buff *skb, struct net_device *dev, 2 * ETH_ALEN); } - skb->dev = ds->ports[source_port]; + skb->dev = ds->ports[source_port].netdev; skb_push(skb, ETH_HLEN); skb->pkt_type = PACKET_HOST; skb->protocol = eth_type_trans(skb, skb->dev); diff --git a/net/dsa/tag_edsa.c b/net/dsa/tag_edsa.c index 3890aac..6c1720e 100644 --- a/net/dsa/tag_edsa.c +++ b/net/dsa/tag_edsa.c @@ -127,7 +127,7 @@ static int edsa_rcv(struct sk_buff *skb, struct net_device *dev, if (!ds) goto out_drop; - if (source_port >= DSA_MAX_PORTS || ds->ports[source_port] == NULL) + if (source_port >= DSA_MAX_PORTS || !ds->ports[source_port].netdev) goto out_drop; /* @@ -182,7 +182,7 @@ static int edsa_rcv(struct sk_buff *skb, struct net_device *dev, 2 * ETH_ALEN); } - skb->dev = ds->ports[source_port]; + skb->dev = ds->ports[source_port].netdev; skb_push(skb, ETH_HLEN); skb->pkt_type = PACKET_HOST; skb->protocol = eth_type_trans(skb, skb->dev); diff --git a/net/dsa/tag_trailer.c b/net/dsa/tag_trailer.c index b6ca089..5e3903e 100644 --- a/net/dsa/tag_trailer.c +++ b/net/dsa/tag_trailer.c @@ -82,12 +82,12 @@ static int trailer_rcv(struct sk_buff *skb, struct net_device *dev, goto out_drop; source_port = trailer[1] & 7; - if (source_port >= DSA_MAX_PORTS || ds->ports[source_port] == NULL) + if (source_port >= DSA_MAX_PORTS || !ds->ports[source_port].netdev) goto out_drop; pskb_trim_rcsum(skb, skb->len - 4); - skb->dev = ds->ports[source_port]; + skb->dev = ds->ports[source_port].netdev; skb_push(skb, ETH_HLEN); skb->pkt_type = PACKET_HOST; skb->protocol = eth_type_trans(skb, skb->dev); -- cgit v0.10.2 From 189b0d93ec61e1f991e96d7bc03b03cf929d164c Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:16:58 +0200 Subject: net: dsa: Move port device node into port structure Move the port device node structure into the port structure, from the chip data. This information is needed in the next step of implementing the new binding. The chip data structure is used while parsing the whole old binding, before the individual switch structures exist. With the new bindings, this is reversed, the switches exist first, and the interconnections between the switches is derived from the individual switch bindings. Thus this chip data structure becomes unneeded. Signed-off-by: Andrew Lunn eviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/include/net/dsa.h b/include/net/dsa.h index 9aed857..8314197 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -121,6 +121,7 @@ struct dsa_switch_tree { struct dsa_port { struct net_device *netdev; + struct device_node *dn; }; struct dsa_switch { diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 18086e0..5907f8c 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -182,7 +182,6 @@ __ATTRIBUTE_GROUPS(dsa_hwmon); /* basic switch operations **************************************************/ static int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct net_device *master) { - struct dsa_chip_data *cd = ds->cd; struct device_node *port_dn; struct phy_device *phydev; int ret, port, mode; @@ -191,7 +190,7 @@ static int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct net_device *master) if (!(dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port))) continue; - port_dn = cd->port_dn[port]; + port_dn = ds->ports[port].dn; if (of_phy_is_fixed_link(port_dn)) { ret = of_phy_register_fixed_link(port_dn); if (ret) { @@ -325,6 +324,8 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, struct device *parent) * Create network devices for physical switch ports. */ for (i = 0; i < DSA_MAX_PORTS; i++) { + ds->ports[i].dn = cd->port_dn[i]; + if (!(ds->enabled_port_mask & (1 << i))) continue; @@ -424,7 +425,6 @@ static void dsa_switch_destroy(struct dsa_switch *ds) { struct device_node *port_dn; struct phy_device *phydev; - struct dsa_chip_data *cd = ds->cd; int port; #ifdef CONFIG_NET_DSA_HWMON @@ -445,7 +445,7 @@ static void dsa_switch_destroy(struct dsa_switch *ds) /* Remove any fixed link PHYs */ for (port = 0; port < DSA_MAX_PORTS; port++) { - port_dn = cd->port_dn[port]; + port_dn = ds->ports[port].dn; if (of_phy_is_fixed_link(port_dn)) { phydev = of_phy_find_device(port_dn); if (phydev) { diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 169abac..52f1183 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -998,13 +998,12 @@ static int dsa_slave_phy_setup(struct dsa_slave_priv *p, struct net_device *slave_dev) { struct dsa_switch *ds = p->parent; - struct dsa_chip_data *cd = ds->cd; struct device_node *phy_dn, *port_dn; bool phy_is_fixed = false; u32 phy_flags = 0; int mode, ret; - port_dn = cd->port_dn[p->port]; + port_dn = ds->ports[p->port].dn; mode = of_get_phy_mode(port_dn); if (mode < 0) mode = PHY_INTERFACE_MODE_NA; @@ -1146,7 +1145,7 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent, NULL); SET_NETDEV_DEV(slave_dev, parent); - slave_dev->dev.of_node = ds->cd->port_dn[port]; + slave_dev->dev.of_node = ds->ports[port].dn; slave_dev->vlan_features = master->vlan_features; p = netdev_priv(slave_dev); -- cgit v0.10.2 From 4a7704ffa86705b0580b6473c407b7b7618e072d Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:16:59 +0200 Subject: net: dsa: Remove dynamic allocate of routing table With a maximum of four switches, the size of the routing table is the same as the pointer to it. Removing it makes the code simpler. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 85332d9..d622c0f 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3024,8 +3024,7 @@ static int mv88e6xxx_setup_global(struct mv88e6xxx_priv_state *ps) for (i = 0; i < 32; i++) { int nexthop = 0x1f; - if (ps->ds->cd->rtable && - i != ps->ds->index && i < ps->ds->dst->pd->nr_chips) + if (i != ps->ds->index && i < ps->ds->dst->pd->nr_chips) nexthop = ps->ds->cd->rtable[i] & 0x1f; err = _mv88e6xxx_reg_write( diff --git a/include/net/dsa.h b/include/net/dsa.h index 8314197..4e3afa9 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -58,12 +58,11 @@ struct dsa_chip_data { struct device_node *port_dn[DSA_MAX_PORTS]; /* - * An array (with nr_chips elements) of which element [a] - * indicates which port on this switch should be used to - * send packets to that are destined for switch a. Can be - * NULL if there is only one switch chip. + * An array of which element [a] indicates which port on this + * switch should be used to send packets to that are destined + * for switch a. Can be NULL if there is only one switch chip. */ - s8 *rtable; + s8 rtable[DSA_MAX_SWITCHES]; }; struct dsa_platform_data { diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 5907f8c..6177dd7 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -587,17 +587,6 @@ static int dsa_of_setup_routing_table(struct dsa_platform_data *pd, if (link_sw_addr >= pd->nr_chips) return -EINVAL; - /* First time routing table allocation */ - if (!cd->rtable) { - cd->rtable = kmalloc_array(pd->nr_chips, sizeof(s8), - GFP_KERNEL); - if (!cd->rtable) - return -ENOMEM; - - /* default to no valid uplink/downlink */ - memset(cd->rtable, -1, pd->nr_chips * sizeof(s8)); - } - cd->rtable[link_sw_addr] = port_index; return 0; @@ -639,7 +628,6 @@ static void dsa_of_free_platform_data(struct dsa_platform_data *pd) kfree(pd->chip[i].port_names[port_index]); port_index++; } - kfree(pd->chip[i].rtable); /* Drop our reference to the MDIO bus device */ if (pd->chip[i].host_dev) -- cgit v0.10.2 From 66472fc04e8be62858f29c7798ed17e984c1ab3b Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:17:00 +0200 Subject: net: dsa: Copy the routing table into the switch structure The new binding will not have a chip data structure, it will place the routing directly into the switch structure. To enable backwards compatibility, copy the routing from the chip data into the switch structure. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index d622c0f..492801a 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3024,8 +3024,8 @@ static int mv88e6xxx_setup_global(struct mv88e6xxx_priv_state *ps) for (i = 0; i < 32; i++) { int nexthop = 0x1f; - if (i != ps->ds->index && i < ps->ds->dst->pd->nr_chips) - nexthop = ps->ds->cd->rtable[i] & 0x1f; + if (i != ds->index && i < DSA_MAX_SWITCHES) + nexthop = ds->rtable[i] & 0x1f; err = _mv88e6xxx_reg_write( ps, REG_GLOBAL2, diff --git a/include/net/dsa.h b/include/net/dsa.h index 4e3afa9..b666f27 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -148,6 +148,13 @@ struct dsa_switch { */ struct dsa_switch_driver *drv; + /* + * An array of which element [a] indicates which port on this + * switch should be used to send packets to that are destined + * for switch a. Can be NULL if there is only one switch chip. + */ + s8 rtable[DSA_MAX_SWITCHES]; + #ifdef CONFIG_NET_DSA_HWMON /* * Hardware monitoring information @@ -194,7 +201,7 @@ static inline u8 dsa_upstream_port(struct dsa_switch *ds) if (dst->cpu_switch == ds->index) return dst->cpu_port; else - return ds->cd->rtable[dst->cpu_switch]; + return ds->rtable[dst->cpu_switch]; } struct switchdev_trans; diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 6177dd7..bfe1d03 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -297,6 +297,8 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, struct device *parent) dst->tag_protocol = drv->tag_protocol; } + memcpy(ds->rtable, cd->rtable, sizeof(ds->rtable)); + /* * Do basic register setup. */ -- cgit v0.10.2 From 9b8e895c4e9d217dfa0e48aafa072258e2a3480e Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:17:01 +0200 Subject: net: dsa: Split up creating/destroying of DSA and CPU ports Refactor the code to setup a single DSA/CPU port into a function of its own, and export it, so it can be used by the new binding. Similarly, refactor the destroy code into a function. When destroying the ports, don't put the of node. They should be released at the end along with the normal ports. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index bfe1d03..7140de4 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -180,36 +180,47 @@ __ATTRIBUTE_GROUPS(dsa_hwmon); #endif /* CONFIG_NET_DSA_HWMON */ /* basic switch operations **************************************************/ -static int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct net_device *master) +int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct device *dev, + struct device_node *port_dn, int port) { - struct device_node *port_dn; struct phy_device *phydev; - int ret, port, mode; + int ret, mode; + + if (of_phy_is_fixed_link(port_dn)) { + ret = of_phy_register_fixed_link(port_dn); + if (ret) { + dev_err(dev, "failed to register fixed PHY\n"); + return ret; + } + phydev = of_phy_find_device(port_dn); + + mode = of_get_phy_mode(port_dn); + if (mode < 0) + mode = PHY_INTERFACE_MODE_NA; + phydev->interface = mode; + + genphy_config_init(phydev); + genphy_read_status(phydev); + if (ds->drv->adjust_link) + ds->drv->adjust_link(ds, port, phydev); + } + + return 0; +} + +static int dsa_cpu_dsa_setups(struct dsa_switch *ds, struct device *dev) +{ + struct device_node *port_dn; + int ret, port; for (port = 0; port < DSA_MAX_PORTS; port++) { if (!(dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port))) continue; port_dn = ds->ports[port].dn; - if (of_phy_is_fixed_link(port_dn)) { - ret = of_phy_register_fixed_link(port_dn); - if (ret) { - netdev_err(master, - "failed to register fixed PHY\n"); - return ret; - } - phydev = of_phy_find_device(port_dn); - - mode = of_get_phy_mode(port_dn); - if (mode < 0) - mode = PHY_INTERFACE_MODE_NA; - phydev->interface = mode; - - genphy_config_init(phydev); - genphy_read_status(phydev); - if (ds->drv->adjust_link) - ds->drv->adjust_link(ds, port, phydev); - } + ret = dsa_cpu_dsa_setup(ds, dev, port_dn, port); + if (ret) + return ret; } return 0; } @@ -340,7 +351,7 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, struct device *parent) } /* Perform configuration of the CPU and DSA ports */ - ret = dsa_cpu_dsa_setup(ds, dst->master_netdev); + ret = dsa_cpu_dsa_setups(ds, parent); if (ret < 0) { netdev_err(dst->master_netdev, "[%d] : can't configure CPU and DSA ports\n", index); @@ -423,10 +434,21 @@ dsa_switch_setup(struct dsa_switch_tree *dst, int index, return ds; } -static void dsa_switch_destroy(struct dsa_switch *ds) +void dsa_cpu_dsa_destroy(struct device_node *port_dn) { - struct device_node *port_dn; struct phy_device *phydev; + + if (of_phy_is_fixed_link(port_dn)) { + phydev = of_phy_find_device(port_dn); + if (phydev) { + phy_device_free(phydev); + fixed_phy_unregister(phydev); + } + } +} + +static void dsa_switch_destroy(struct dsa_switch *ds) +{ int port; #ifdef CONFIG_NET_DSA_HWMON @@ -445,17 +467,11 @@ static void dsa_switch_destroy(struct dsa_switch *ds) dsa_slave_destroy(ds->ports[port].netdev); } - /* Remove any fixed link PHYs */ + /* Disable configuration of the CPU and DSA ports */ for (port = 0; port < DSA_MAX_PORTS; port++) { - port_dn = ds->ports[port].dn; - if (of_phy_is_fixed_link(port_dn)) { - phydev = of_phy_find_device(port_dn); - if (phydev) { - phy_device_free(phydev); - of_node_put(port_dn); - fixed_phy_unregister(phydev); - } - } + if (!(dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port))) + continue; + dsa_cpu_dsa_destroy(ds->ports[port].dn); } mdiobus_unregister(ds->slave_mii_bus); diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index dfa3377..dbea5d9 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -50,6 +50,9 @@ struct dsa_slave_priv { /* dsa.c */ extern char dsa_driver_version[]; +int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct device *dev, + struct device_node *port_dn, int port); +void dsa_cpu_dsa_destroy(struct device_node *port_dn); /* slave.c */ extern const struct dsa_device_ops notag_netdev_ops; -- cgit v0.10.2 From 5377b802fc9fde9442dcaba571edefcb73765056 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:17:02 +0200 Subject: net: dsa: mv88e6xxx: Only support EDSA tagging The merged driver no longer offers the option to use DSA tagging. So remove the code to setup the switch to do DSA tagging and hard code the use of EDSA. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot y Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 492801a..11845ec 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -2725,11 +2725,8 @@ static int mv88e6xxx_setup_port(struct mv88e6xxx_priv_state *ps, int port) if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || mv88e6xxx_6320_family(ps)) { - if (ds->dst->tag_protocol == DSA_TAG_PROTO_EDSA) - reg |= PORT_CONTROL_FRAME_ETHER_TYPE_DSA; - else - reg |= PORT_CONTROL_FRAME_MODE_DSA; - reg |= PORT_CONTROL_FORWARD_UNKNOWN | + reg |= PORT_CONTROL_FRAME_ETHER_TYPE_DSA | + PORT_CONTROL_FORWARD_UNKNOWN | PORT_CONTROL_FORWARD_UNKNOWN_MC; } @@ -2737,7 +2734,6 @@ static int mv88e6xxx_setup_port(struct mv88e6xxx_priv_state *ps, int port) mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || mv88e6xxx_6095_family(ps) || mv88e6xxx_6065_family(ps) || mv88e6xxx_6185_family(ps) || mv88e6xxx_6320_family(ps)) { - if (ds->dst->tag_protocol == DSA_TAG_PROTO_EDSA) reg |= PORT_CONTROL_EGRESS_ADD_TAG; } } -- cgit v0.10.2 From 39a7f2a4eb496c0c68cc93fcb403190b48605168 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:17:03 +0200 Subject: net: dsa: Refactor selection of tag ops into a function Replace the two switch statements with an array lookup, and store the result in the dsa tree structure. The drivers no longer need to know the selected tag protocol, so remove it from the dsa switch structure. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/include/net/dsa.h b/include/net/dsa.h index b666f27..bd6ecaa 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -26,6 +26,7 @@ enum dsa_tag_protocol { DSA_TAG_PROTO_TRAILER, DSA_TAG_PROTO_EDSA, DSA_TAG_PROTO_BRCM, + DSA_TAG_LAST, /* MUST BE LAST */ }; #define DSA_MAX_SWITCHES 4 @@ -99,7 +100,6 @@ struct dsa_switch_tree { struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev); - enum dsa_tag_protocol tag_protocol; /* * Original copy of the master netdev ethtool_ops @@ -116,6 +116,12 @@ struct dsa_switch_tree { * Data for the individual switch chips. */ struct dsa_switch *ds[DSA_MAX_SWITCHES]; + + /* + * Tagging protocol operations for adding and removing an + * encapsulation tag. + */ + const struct dsa_device_ops *tag_ops; }; struct dsa_port { diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 7140de4..221ebde 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -29,6 +29,33 @@ char dsa_driver_version[] = "0.1"; +static struct sk_buff *dsa_slave_notag_xmit(struct sk_buff *skb, + struct net_device *dev) +{ + /* Just return the original SKB */ + return skb; +} + +static const struct dsa_device_ops none_ops = { + .xmit = dsa_slave_notag_xmit, + .rcv = NULL, +}; + +const struct dsa_device_ops *dsa_device_ops[DSA_TAG_LAST] = { +#ifdef CONFIG_NET_DSA_TAG_DSA + [DSA_TAG_PROTO_DSA] = &dsa_netdev_ops, +#endif +#ifdef CONFIG_NET_DSA_TAG_EDSA + [DSA_TAG_PROTO_EDSA] = &edsa_netdev_ops, +#endif +#ifdef CONFIG_NET_DSA_TAG_TRAILER + [DSA_TAG_PROTO_TRAILER] = &trailer_netdev_ops, +#endif +#ifdef CONFIG_NET_DSA_TAG_BRCM + [DSA_TAG_PROTO_BRCM] = &brcm_netdev_ops, +#endif + [DSA_TAG_PROTO_NONE] = &none_ops, +}; /* switch driver registration ***********************************************/ static DEFINE_MUTEX(dsa_switch_drivers_mutex); @@ -225,6 +252,20 @@ static int dsa_cpu_dsa_setups(struct dsa_switch *ds, struct device *dev) return 0; } +const struct dsa_device_ops *dsa_resolve_tag_protocol(int tag_protocol) +{ + const struct dsa_device_ops *ops; + + if (tag_protocol >= DSA_TAG_LAST) + return ERR_PTR(-EINVAL); + ops = dsa_device_ops[tag_protocol]; + + if (!ops) + return ERR_PTR(-ENOPROTOOPT); + + return ops; +} + static int dsa_switch_setup_one(struct dsa_switch *ds, struct device *parent) { struct dsa_switch_driver *drv = ds->drv; @@ -277,35 +318,13 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, struct device *parent) * switch. */ if (dst->cpu_switch == index) { - switch (drv->tag_protocol) { -#ifdef CONFIG_NET_DSA_TAG_DSA - case DSA_TAG_PROTO_DSA: - dst->rcv = dsa_netdev_ops.rcv; - break; -#endif -#ifdef CONFIG_NET_DSA_TAG_EDSA - case DSA_TAG_PROTO_EDSA: - dst->rcv = edsa_netdev_ops.rcv; - break; -#endif -#ifdef CONFIG_NET_DSA_TAG_TRAILER - case DSA_TAG_PROTO_TRAILER: - dst->rcv = trailer_netdev_ops.rcv; - break; -#endif -#ifdef CONFIG_NET_DSA_TAG_BRCM - case DSA_TAG_PROTO_BRCM: - dst->rcv = brcm_netdev_ops.rcv; - break; -#endif - case DSA_TAG_PROTO_NONE: - break; - default: - ret = -ENOPROTOOPT; + dst->tag_ops = dsa_resolve_tag_protocol(drv->tag_protocol); + if (IS_ERR(dst->tag_ops)) { + ret = PTR_ERR(dst->tag_ops); goto out; } - dst->tag_protocol = drv->tag_protocol; + dst->rcv = dst->tag_ops->rcv; } memcpy(ds->rtable, cd->rtable, sizeof(ds->rtable)); diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index dbea5d9..72f7b89 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -53,6 +53,7 @@ extern char dsa_driver_version[]; int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct device *dev, struct device_node *port_dn, int port); void dsa_cpu_dsa_destroy(struct device_node *port_dn); +const struct dsa_device_ops *dsa_resolve_tag_protocol(int tag_protocol); /* slave.c */ extern const struct dsa_device_ops notag_netdev_ops; diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 52f1183..35e5f0f 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -521,14 +521,6 @@ static netdev_tx_t dsa_slave_xmit(struct sk_buff *skb, struct net_device *dev) return NETDEV_TX_OK; } -static struct sk_buff *dsa_slave_notag_xmit(struct sk_buff *skb, - struct net_device *dev) -{ - /* Just return the original SKB */ - return skb; -} - - /* ethtool operations *******************************************************/ static int dsa_slave_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) @@ -1151,32 +1143,7 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent, p = netdev_priv(slave_dev); p->parent = ds; p->port = port; - - switch (ds->dst->tag_protocol) { -#ifdef CONFIG_NET_DSA_TAG_DSA - case DSA_TAG_PROTO_DSA: - p->xmit = dsa_netdev_ops.xmit; - break; -#endif -#ifdef CONFIG_NET_DSA_TAG_EDSA - case DSA_TAG_PROTO_EDSA: - p->xmit = edsa_netdev_ops.xmit; - break; -#endif -#ifdef CONFIG_NET_DSA_TAG_TRAILER - case DSA_TAG_PROTO_TRAILER: - p->xmit = trailer_netdev_ops.xmit; - break; -#endif -#ifdef CONFIG_NET_DSA_TAG_BRCM - case DSA_TAG_PROTO_BRCM: - p->xmit = brcm_netdev_ops.xmit; - break; -#endif - default: - p->xmit = dsa_slave_notag_xmit; - break; - } + p->xmit = dst->tag_ops->xmit; p->old_pause = -1; p->old_link = -1; -- cgit v0.10.2 From e755e49eb3ea925834006c294e989df52f592580 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:17:04 +0200 Subject: net: dsa: Make mdio bus optional The switch may want to instantiate its own MDIO bus. Only do it centrally if the switch has not already created one, and the read op is implemented. Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 221ebde..6c314f3 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -340,17 +340,18 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, struct device *parent) if (ret < 0) goto out; - ds->slave_mii_bus = devm_mdiobus_alloc(parent); - if (ds->slave_mii_bus == NULL) { - ret = -ENOMEM; - goto out; - } - dsa_slave_mii_bus_init(ds); - - ret = mdiobus_register(ds->slave_mii_bus); - if (ret < 0) - goto out; + if (!ds->slave_mii_bus && drv->phy_read) { + ds->slave_mii_bus = devm_mdiobus_alloc(parent); + if (!ds->slave_mii_bus) { + ret = -ENOMEM; + goto out; + } + dsa_slave_mii_bus_init(ds); + ret = mdiobus_register(ds->slave_mii_bus); + if (ret < 0) + goto out; + } /* * Create network devices for physical switch ports. @@ -493,7 +494,8 @@ static void dsa_switch_destroy(struct dsa_switch *ds) dsa_cpu_dsa_destroy(ds->ports[port].dn); } - mdiobus_unregister(ds->slave_mii_bus); + if (ds->slave_mii_bus && ds->drv->phy_read) + mdiobus_unregister(ds->slave_mii_bus); } #ifdef CONFIG_PM_SLEEP -- cgit v0.10.2 From 03a4a5408b674576e8f7e54dd6ed86398217a98e Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:17:05 +0200 Subject: net: dsa: mv88e6xxx: Rename _phy_ to _mdio_ The switch implements a generic MDIO bus, which could host more than PHYs. It is conventional to use _mdio_ or _mii_ in the function name, so rename them. Also postfix make the historically first read/write function with _direct, to help distinguish it from _indirect and _ppu. While touching these functions, remove some of the _ prefixes, which we are deprecating. Signed-off-by: Andrew Lunn Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 11845ec..b8e65e6 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -238,16 +238,16 @@ int mv88e6xxx_set_addr(struct dsa_switch *ds, u8 *addr) return mv88e6xxx_set_addr_direct(ds, addr); } -static int _mv88e6xxx_phy_read(struct mv88e6xxx_priv_state *ps, int addr, - int regnum) +static int mv88e6xxx_mdio_read_direct(struct mv88e6xxx_priv_state *ps, + int addr, int regnum) { if (addr >= 0) return _mv88e6xxx_reg_read(ps, addr, regnum); return 0xffff; } -static int _mv88e6xxx_phy_write(struct mv88e6xxx_priv_state *ps, int addr, - int regnum, u16 val) +static int mv88e6xxx_mdio_write_direct(struct mv88e6xxx_priv_state *ps, + int addr, int regnum, u16 val) { if (addr >= 0) return _mv88e6xxx_reg_write(ps, addr, regnum, val); @@ -378,8 +378,8 @@ void mv88e6xxx_ppu_state_init(struct mv88e6xxx_priv_state *ps) ps->ppu_timer.function = mv88e6xxx_ppu_reenable_timer; } -static int mv88e6xxx_phy_read_ppu(struct mv88e6xxx_priv_state *ps, int addr, - int regnum) +static int mv88e6xxx_mdio_read_ppu(struct mv88e6xxx_priv_state *ps, int addr, + int regnum) { int ret; @@ -392,8 +392,8 @@ static int mv88e6xxx_phy_read_ppu(struct mv88e6xxx_priv_state *ps, int addr, return ret; } -static int mv88e6xxx_phy_write_ppu(struct mv88e6xxx_priv_state *ps, int addr, - int regnum, u16 val) +static int mv88e6xxx_mdio_write_ppu(struct mv88e6xxx_priv_state *ps, int addr, + int regnum, u16 val) { int ret; @@ -829,7 +829,7 @@ static int mv88e6xxx_wait(struct mv88e6xxx_priv_state *ps, int reg, return ret; } -static int _mv88e6xxx_phy_wait(struct mv88e6xxx_priv_state *ps) +static int mv88e6xxx_mdio_wait(struct mv88e6xxx_priv_state *ps) { return _mv88e6xxx_wait(ps, REG_GLOBAL2, GLOBAL2_SMI_OP, GLOBAL2_SMI_OP_BUSY); @@ -1076,7 +1076,7 @@ static int _mv88e6xxx_atu_wait(struct mv88e6xxx_priv_state *ps) GLOBAL_ATU_OP_BUSY); } -static int _mv88e6xxx_phy_read_indirect(struct mv88e6xxx_priv_state *ps, +static int mv88e6xxx_mdio_read_indirect(struct mv88e6xxx_priv_state *ps, int addr, int regnum) { int ret; @@ -1087,7 +1087,7 @@ static int _mv88e6xxx_phy_read_indirect(struct mv88e6xxx_priv_state *ps, if (ret < 0) return ret; - ret = _mv88e6xxx_phy_wait(ps); + ret = mv88e6xxx_mdio_wait(ps); if (ret < 0) return ret; @@ -1096,7 +1096,7 @@ static int _mv88e6xxx_phy_read_indirect(struct mv88e6xxx_priv_state *ps, return ret; } -static int _mv88e6xxx_phy_write_indirect(struct mv88e6xxx_priv_state *ps, +static int mv88e6xxx_mdio_write_indirect(struct mv88e6xxx_priv_state *ps, int addr, int regnum, u16 val) { int ret; @@ -1109,7 +1109,7 @@ static int _mv88e6xxx_phy_write_indirect(struct mv88e6xxx_priv_state *ps, GLOBAL2_SMI_OP_22_WRITE | (addr << 5) | regnum); - return _mv88e6xxx_phy_wait(ps); + return mv88e6xxx_mdio_wait(ps); } static int mv88e6xxx_get_eee(struct dsa_switch *ds, int port, @@ -1123,7 +1123,7 @@ static int mv88e6xxx_get_eee(struct dsa_switch *ds, int port, mutex_lock(&ps->smi_mutex); - reg = _mv88e6xxx_phy_read_indirect(ps, port, 16); + reg = mv88e6xxx_mdio_read_indirect(ps, port, 16); if (reg < 0) goto out; @@ -1154,7 +1154,7 @@ static int mv88e6xxx_set_eee(struct dsa_switch *ds, int port, mutex_lock(&ps->smi_mutex); - ret = _mv88e6xxx_phy_read_indirect(ps, port, 16); + ret = mv88e6xxx_mdio_read_indirect(ps, port, 16); if (ret < 0) goto out; @@ -1164,7 +1164,7 @@ static int mv88e6xxx_set_eee(struct dsa_switch *ds, int port, if (e->tx_lpi_enabled) reg |= 0x0100; - ret = _mv88e6xxx_phy_write_indirect(ps, port, 16, reg); + ret = mv88e6xxx_mdio_write_indirect(ps, port, 16, reg); out: mutex_unlock(&ps->smi_mutex); @@ -2547,34 +2547,34 @@ static void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port) mutex_unlock(&ps->smi_mutex); } -static int _mv88e6xxx_phy_page_write(struct mv88e6xxx_priv_state *ps, - int port, int page, int reg, int val) +static int _mv88e6xxx_mdio_page_write(struct mv88e6xxx_priv_state *ps, + int port, int page, int reg, int val) { int ret; - ret = _mv88e6xxx_phy_write_indirect(ps, port, 0x16, page); + ret = mv88e6xxx_mdio_write_indirect(ps, port, 0x16, page); if (ret < 0) goto restore_page_0; - ret = _mv88e6xxx_phy_write_indirect(ps, port, reg, val); + ret = mv88e6xxx_mdio_write_indirect(ps, port, reg, val); restore_page_0: - _mv88e6xxx_phy_write_indirect(ps, port, 0x16, 0x0); + mv88e6xxx_mdio_write_indirect(ps, port, 0x16, 0x0); return ret; } -static int _mv88e6xxx_phy_page_read(struct mv88e6xxx_priv_state *ps, - int port, int page, int reg) +static int _mv88e6xxx_mdio_page_read(struct mv88e6xxx_priv_state *ps, + int port, int page, int reg) { int ret; - ret = _mv88e6xxx_phy_write_indirect(ps, port, 0x16, page); + ret = mv88e6xxx_mdio_write_indirect(ps, port, 0x16, page); if (ret < 0) goto restore_page_0; - ret = _mv88e6xxx_phy_read_indirect(ps, port, reg); + ret = mv88e6xxx_mdio_read_indirect(ps, port, reg); restore_page_0: - _mv88e6xxx_phy_write_indirect(ps, port, 0x16, 0x0); + mv88e6xxx_mdio_write_indirect(ps, port, 0x16, 0x0); return ret; } @@ -2645,16 +2645,16 @@ static int mv88e6xxx_power_on_serdes(struct mv88e6xxx_priv_state *ps) { int ret; - ret = _mv88e6xxx_phy_page_read(ps, REG_FIBER_SERDES, PAGE_FIBER_SERDES, - MII_BMCR); + ret = _mv88e6xxx_mdio_page_read(ps, REG_FIBER_SERDES, + PAGE_FIBER_SERDES, MII_BMCR); if (ret < 0) return ret; if (ret & BMCR_PDOWN) { ret &= ~BMCR_PDOWN; - ret = _mv88e6xxx_phy_page_write(ps, REG_FIBER_SERDES, - PAGE_FIBER_SERDES, MII_BMCR, - ret); + ret = _mv88e6xxx_mdio_page_write(ps, REG_FIBER_SERDES, + PAGE_FIBER_SERDES, MII_BMCR, + ret); } return ret; @@ -3159,43 +3159,43 @@ unlock: return err; } -int mv88e6xxx_phy_page_read(struct dsa_switch *ds, int port, int page, int reg) +int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, int reg) { struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); int ret; mutex_lock(&ps->smi_mutex); - ret = _mv88e6xxx_phy_page_read(ps, port, page, reg); + ret = _mv88e6xxx_mdio_page_read(ps, port, page, reg); mutex_unlock(&ps->smi_mutex); return ret; } -int mv88e6xxx_phy_page_write(struct dsa_switch *ds, int port, int page, - int reg, int val) +int mv88e6xxx_mdio_page_write(struct dsa_switch *ds, int port, int page, + int reg, int val) { struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); int ret; mutex_lock(&ps->smi_mutex); - ret = _mv88e6xxx_phy_page_write(ps, port, page, reg, val); + ret = _mv88e6xxx_mdio_page_write(ps, port, page, reg, val); mutex_unlock(&ps->smi_mutex); return ret; } -static int mv88e6xxx_port_to_phy_addr(struct mv88e6xxx_priv_state *ps, - int port) +static int mv88e6xxx_port_to_mdio_addr(struct mv88e6xxx_priv_state *ps, + int port) { if (port >= 0 && port < ps->info->num_ports) return port; return -EINVAL; } -static int mv88e6xxx_phy_read(struct dsa_switch *ds, int port, int regnum) +static int mv88e6xxx_mdio_read(struct dsa_switch *ds, int port, int regnum) { struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int addr = mv88e6xxx_port_to_phy_addr(ps, port); + int addr = mv88e6xxx_port_to_mdio_addr(ps, port); int ret; if (addr < 0) @@ -3204,21 +3204,21 @@ static int mv88e6xxx_phy_read(struct dsa_switch *ds, int port, int regnum) mutex_lock(&ps->smi_mutex); if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) - ret = mv88e6xxx_phy_read_ppu(ps, addr, regnum); + ret = mv88e6xxx_mdio_read_ppu(ps, addr, regnum); else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SMI_PHY)) - ret = _mv88e6xxx_phy_read_indirect(ps, addr, regnum); + ret = mv88e6xxx_mdio_read_indirect(ps, addr, regnum); else - ret = _mv88e6xxx_phy_read(ps, addr, regnum); + ret = mv88e6xxx_mdio_read_direct(ps, addr, regnum); mutex_unlock(&ps->smi_mutex); return ret; } -static int mv88e6xxx_phy_write(struct dsa_switch *ds, int port, int regnum, - u16 val) +static int mv88e6xxx_mdio_write(struct dsa_switch *ds, int port, int regnum, + u16 val) { struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int addr = mv88e6xxx_port_to_phy_addr(ps, port); + int addr = mv88e6xxx_port_to_mdio_addr(ps, port); int ret; if (addr < 0) @@ -3227,11 +3227,11 @@ static int mv88e6xxx_phy_write(struct dsa_switch *ds, int port, int regnum, mutex_lock(&ps->smi_mutex); if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) - ret = mv88e6xxx_phy_write_ppu(ps, addr, regnum, val); + ret = mv88e6xxx_mdio_write_ppu(ps, addr, regnum, val); else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SMI_PHY)) - ret = _mv88e6xxx_phy_write_indirect(ps, addr, regnum, val); + ret = mv88e6xxx_mdio_write_indirect(ps, addr, regnum, val); else - ret = _mv88e6xxx_phy_write(ps, addr, regnum, val); + ret = mv88e6xxx_mdio_write_direct(ps, addr, regnum, val); mutex_unlock(&ps->smi_mutex); return ret; @@ -3249,37 +3249,37 @@ static int mv88e61xx_get_temp(struct dsa_switch *ds, int *temp) mutex_lock(&ps->smi_mutex); - ret = _mv88e6xxx_phy_write(ps, 0x0, 0x16, 0x6); + ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x16, 0x6); if (ret < 0) goto error; /* Enable temperature sensor */ - ret = _mv88e6xxx_phy_read(ps, 0x0, 0x1a); + ret = mv88e6xxx_mdio_read_direct(ps, 0x0, 0x1a); if (ret < 0) goto error; - ret = _mv88e6xxx_phy_write(ps, 0x0, 0x1a, ret | (1 << 5)); + ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x1a, ret | (1 << 5)); if (ret < 0) goto error; /* Wait for temperature to stabilize */ usleep_range(10000, 12000); - val = _mv88e6xxx_phy_read(ps, 0x0, 0x1a); + val = mv88e6xxx_mdio_read_direct(ps, 0x0, 0x1a); if (val < 0) { ret = val; goto error; } /* Disable temperature sensor */ - ret = _mv88e6xxx_phy_write(ps, 0x0, 0x1a, ret & ~(1 << 5)); + ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x1a, ret & ~(1 << 5)); if (ret < 0) goto error; *temp = ((val & 0x1f) - 5) * 5; error: - _mv88e6xxx_phy_write(ps, 0x0, 0x16, 0x0); + mv88e6xxx_mdio_write_direct(ps, 0x0, 0x16, 0x0); mutex_unlock(&ps->smi_mutex); return ret; } @@ -3292,7 +3292,7 @@ static int mv88e63xx_get_temp(struct dsa_switch *ds, int *temp) *temp = 0; - ret = mv88e6xxx_phy_page_read(ds, phy, 6, 27); + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 27); if (ret < 0) return ret; @@ -3325,7 +3325,7 @@ static int mv88e6xxx_get_temp_limit(struct dsa_switch *ds, int *temp) *temp = 0; - ret = mv88e6xxx_phy_page_read(ds, phy, 6, 26); + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); if (ret < 0) return ret; @@ -3343,12 +3343,12 @@ static int mv88e6xxx_set_temp_limit(struct dsa_switch *ds, int temp) if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP_LIMIT)) return -EOPNOTSUPP; - ret = mv88e6xxx_phy_page_read(ds, phy, 6, 26); + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); if (ret < 0) return ret; temp = clamp_val(DIV_ROUND_CLOSEST(temp, 5) + 5, 0, 0x1f); - return mv88e6xxx_phy_page_write(ds, phy, 6, 26, - (ret & 0xe0ff) | (temp << 8)); + return mv88e6xxx_mdio_page_write(ds, phy, 6, 26, + (ret & 0xe0ff) | (temp << 8)); } static int mv88e6xxx_get_temp_alarm(struct dsa_switch *ds, bool *alarm) @@ -3362,7 +3362,7 @@ static int mv88e6xxx_get_temp_alarm(struct dsa_switch *ds, bool *alarm) *alarm = false; - ret = mv88e6xxx_phy_page_read(ds, phy, 6, 26); + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); if (ret < 0) return ret; @@ -3590,8 +3590,8 @@ struct dsa_switch_driver mv88e6xxx_switch_driver = { .probe = mv88e6xxx_drv_probe, .setup = mv88e6xxx_setup, .set_addr = mv88e6xxx_set_addr, - .phy_read = mv88e6xxx_phy_read, - .phy_write = mv88e6xxx_phy_write, + .phy_read = mv88e6xxx_mdio_read, + .phy_write = mv88e6xxx_mdio_write, .adjust_link = mv88e6xxx_adjust_link, .get_strings = mv88e6xxx_get_strings, .get_ethtool_stats = mv88e6xxx_get_ethtool_stats, -- cgit v0.10.2 From b516d453239551d7916d5e35bc68823ed5b55f98 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:17:06 +0200 Subject: net: dsa: mv88e6xxx: Refactor MDIO so driver registers mdio bus Have the switch driver register its own MDIO bus. This allows for an mdio property in the device tree, with child nodes for phys, which can be referenced via phandles, etc. Signed-off-by: Andrew Lunn Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index b8e65e6..192b39c 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include @@ -3130,13 +3131,11 @@ static int mv88e6xxx_setup(struct dsa_switch *ds) int i; ps->ds = ds; + ds->slave_mii_bus = ps->mdio_bus; if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) mutex_init(&ps->eeprom_mutex); - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) - mv88e6xxx_ppu_state_init(ps); - mutex_lock(&ps->smi_mutex); err = mv88e6xxx_switch_reset(ps); @@ -3192,9 +3191,9 @@ static int mv88e6xxx_port_to_mdio_addr(struct mv88e6xxx_priv_state *ps, return -EINVAL; } -static int mv88e6xxx_mdio_read(struct dsa_switch *ds, int port, int regnum) +static int mv88e6xxx_mdio_read(struct mii_bus *bus, int port, int regnum) { - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + struct mv88e6xxx_priv_state *ps = bus->priv; int addr = mv88e6xxx_port_to_mdio_addr(ps, port); int ret; @@ -3214,10 +3213,10 @@ static int mv88e6xxx_mdio_read(struct dsa_switch *ds, int port, int regnum) return ret; } -static int mv88e6xxx_mdio_write(struct dsa_switch *ds, int port, int regnum, +static int mv88e6xxx_mdio_write(struct mii_bus *bus, int port, int regnum, u16 val) { - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + struct mv88e6xxx_priv_state *ps = bus->priv; int addr = mv88e6xxx_port_to_mdio_addr(ps, port); int ret; @@ -3237,6 +3236,66 @@ static int mv88e6xxx_mdio_write(struct dsa_switch *ds, int port, int regnum, return ret; } +static int mv88e6xxx_mdio_register(struct mv88e6xxx_priv_state *ps, + struct device_node *np) +{ + static int index; + struct mii_bus *bus; + int err; + + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) + mv88e6xxx_ppu_state_init(ps); + + if (np) + ps->mdio_np = of_get_child_by_name(np, "mdio"); + + bus = devm_mdiobus_alloc(ps->dev); + if (!bus) + return -ENOMEM; + + bus->priv = (void *)ps; + if (np) { + bus->name = np->full_name; + snprintf(bus->id, MII_BUS_ID_SIZE, "%s", np->full_name); + } else { + bus->name = "mv88e6xxx SMI"; + snprintf(bus->id, MII_BUS_ID_SIZE, "mv88e6xxx-%d", index++); + } + + bus->read = mv88e6xxx_mdio_read; + bus->write = mv88e6xxx_mdio_write; + bus->parent = ps->dev; + + if (ps->mdio_np) + err = of_mdiobus_register(bus, ps->mdio_np); + else + err = mdiobus_register(bus); + if (err) { + dev_err(ps->dev, "Cannot register MDIO bus (%d)\n", err); + goto out; + } + ps->mdio_bus = bus; + + return 0; + +out: + if (ps->mdio_np) + of_node_put(ps->mdio_np); + + return err; +} + +static void mv88e6xxx_mdio_unregister(struct mv88e6xxx_priv_state *ps) + +{ + struct mii_bus *bus = ps->mdio_bus; + + mdiobus_unregister(bus); + + if (ps->mdio_np) + of_node_put(ps->mdio_np); +} + #ifdef CONFIG_NET_DSA_HWMON static int mv88e61xx_get_temp(struct dsa_switch *ds, int *temp) @@ -3549,6 +3608,7 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, struct mii_bus *bus; const char *name; int id, prod_num, rev; + int err; bus = dsa_host_dev_to_mii_bus(host_dev); if (!bus) @@ -3575,8 +3635,13 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, ps->bus = bus; ps->sw_addr = sw_addr; ps->info = info; + ps->dev = dsa_dev; mutex_init(&ps->smi_mutex); + err = mv88e6xxx_mdio_register(ps, NULL); + if (err) + return NULL; + *priv = ps; dev_info(&ps->bus->dev, "switch 0x%x probed: %s, revision %u\n", @@ -3590,8 +3655,6 @@ struct dsa_switch_driver mv88e6xxx_switch_driver = { .probe = mv88e6xxx_drv_probe, .setup = mv88e6xxx_setup, .set_addr = mv88e6xxx_set_addr, - .phy_read = mv88e6xxx_mdio_read, - .phy_write = mv88e6xxx_mdio_write, .adjust_link = mv88e6xxx_adjust_link, .get_strings = mv88e6xxx_get_strings, .get_ethtool_stats = mv88e6xxx_get_ethtool_stats, @@ -3677,6 +3740,12 @@ int mv88e6xxx_probe(struct mdio_device *mdiodev) !of_property_read_u32(np, "eeprom-length", &eeprom_len)) ps->eeprom_len = eeprom_len; + err = mv88e6xxx_mdio_register(ps, mdiodev->dev.of_node); + if (err) + return err; + + ds->slave_mii_bus = ps->mdio_bus; + dev_set_drvdata(dev, ds); dev_info(dev, "switch 0x%x probed: %s, revision %u\n", @@ -3691,6 +3760,8 @@ static void mv88e6xxx_remove(struct mdio_device *mdiodev) struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); put_device(&ps->bus->dev); + + mv88e6xxx_mdio_unregister(ps); } static const struct of_device_id mv88e6xxx_of_match[] = { diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h index 36d0e15..8221c3c 100644 --- a/drivers/net/dsa/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx.h @@ -600,6 +600,12 @@ struct mv88e6xxx_priv_state { /* set to size of eeprom if supported by the switch */ int eeprom_len; + + /* Device node for the MDIO bus */ + struct device_node *mdio_np; + + /* And the MDIO bus itself */ + struct mii_bus *mdio_bus; }; enum stat_type { -- cgit v0.10.2 From 83c0afaec7b730b16c518aecc8e6246ec91b265e Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:17:07 +0200 Subject: net: dsa: Add new binding implementation The existing DSA binding has a number of limitations and problems. The main problem is that it cannot represent a switch as a linux device, hanging off some bus. It is limited to one CPU port. The DSA platform device is artificial, and does not really represent hardware. Implement a new binding which can be embedded into any type of node on a bus to represent one switch device, and its links to other switches. Signed-off-by: Andrew Lunn Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 192b39c..ee06055 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3748,6 +3748,12 @@ int mv88e6xxx_probe(struct mdio_device *mdiodev) dev_set_drvdata(dev, ds); + err = dsa_register_switch(ds, mdiodev->dev.of_node); + if (err) { + mv88e6xxx_mdio_unregister(ps); + return err; + } + dev_info(dev, "switch 0x%x probed: %s, revision %u\n", prod_num, ps->info->name, rev); @@ -3759,6 +3765,7 @@ static void mv88e6xxx_remove(struct mdio_device *mdiodev) struct dsa_switch *ds = dev_get_drvdata(&mdiodev->dev); struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + dsa_unregister_switch(ds); put_device(&ps->bus->dev); mv88e6xxx_mdio_unregister(ps); diff --git a/include/net/dsa.h b/include/net/dsa.h index bd6ecaa..cca7ef2 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -85,6 +85,17 @@ struct dsa_platform_data { struct packet_type; struct dsa_switch_tree { + struct list_head list; + + /* Tree identifier */ + u32 tree; + + /* Number of switches attached to this tree */ + struct kref refcount; + + /* Has this tree been applied to the hardware? */ + bool applied; + /* * Configuration data for the platform device that owns * this dsa switch tree instance. @@ -170,9 +181,15 @@ struct dsa_switch { #endif /* + * The lower device this switch uses to talk to the host + */ + struct net_device *master_netdev; + + /* * Slave mii_bus and devices for the individual ports. */ u32 dsa_port_mask; + u32 cpu_port_mask; u32 enabled_port_mask; u32 phys_mii_mask; struct dsa_port ports[DSA_MAX_PORTS]; @@ -361,4 +378,7 @@ static inline bool dsa_uses_tagged_protocol(struct dsa_switch_tree *dst) { return dst->rcv != NULL; } + +void dsa_unregister_switch(struct dsa_switch *ds); +int dsa_register_switch(struct dsa_switch *ds, struct device_node *np); #endif diff --git a/net/dsa/Makefile b/net/dsa/Makefile index da06ed1..8af4ded 100644 --- a/net/dsa/Makefile +++ b/net/dsa/Makefile @@ -1,6 +1,6 @@ # the core obj-$(CONFIG_NET_DSA) += dsa_core.o -dsa_core-y += dsa.o slave.o +dsa_core-y += dsa.o slave.o dsa2.o # tagging formats dsa_core-$(CONFIG_NET_DSA_TAG_BRCM) += tag_brcm.o diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 6c314f3..ce3b942 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -294,6 +294,7 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, struct device *parent) } dst->cpu_switch = index; dst->cpu_port = i; + ds->cpu_port_mask |= 1 << i; } else if (!strcmp(name, "dsa")) { ds->dsa_port_mask |= 1 << i; } else { @@ -492,6 +493,10 @@ static void dsa_switch_destroy(struct dsa_switch *ds) if (!(dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port))) continue; dsa_cpu_dsa_destroy(ds->ports[port].dn); + + /* Clearing a bit which is not set does no harm */ + ds->cpu_port_mask |= ~(1 << port); + ds->dsa_port_mask |= ~(1 << port); } if (ds->slave_mii_bus && ds->drv->phy_read) diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c new file mode 100644 index 0000000..80dfe08 --- /dev/null +++ b/net/dsa/dsa2.c @@ -0,0 +1,654 @@ +/* + * net/dsa/dsa2.c - Hardware switch handling, binding version 2 + * Copyright (c) 2008-2009 Marvell Semiconductor + * Copyright (c) 2013 Florian Fainelli + * Copyright (c) 2016 Andrew Lunn + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include "dsa_priv.h" + +static LIST_HEAD(dsa_switch_trees); +static DEFINE_MUTEX(dsa2_mutex); + +static struct dsa_switch_tree *dsa_get_dst(u32 tree) +{ + struct dsa_switch_tree *dst; + + list_for_each_entry(dst, &dsa_switch_trees, list) + if (dst->tree == tree) + return dst; + return NULL; +} + +static void dsa_free_dst(struct kref *ref) +{ + struct dsa_switch_tree *dst = container_of(ref, struct dsa_switch_tree, + refcount); + + list_del(&dst->list); + kfree(dst); +} + +static void dsa_put_dst(struct dsa_switch_tree *dst) +{ + kref_put(&dst->refcount, dsa_free_dst); +} + +static struct dsa_switch_tree *dsa_add_dst(u32 tree) +{ + struct dsa_switch_tree *dst; + + dst = kzalloc(sizeof(*dst), GFP_KERNEL); + if (!dst) + return NULL; + dst->tree = tree; + dst->cpu_switch = -1; + INIT_LIST_HEAD(&dst->list); + list_add_tail(&dsa_switch_trees, &dst->list); + kref_init(&dst->refcount); + + return dst; +} + +static void dsa_dst_add_ds(struct dsa_switch_tree *dst, + struct dsa_switch *ds, u32 index) +{ + kref_get(&dst->refcount); + dst->ds[index] = ds; +} + +static void dsa_dst_del_ds(struct dsa_switch_tree *dst, + struct dsa_switch *ds, u32 index) +{ + dst->ds[index] = NULL; + kref_put(&dst->refcount, dsa_free_dst); +} + +static bool dsa_port_is_dsa(struct device_node *port) +{ + const char *name; + + name = of_get_property(port, "label", NULL); + if (!name) + return false; + + if (!strcmp(name, "dsa")) + return true; + + return false; +} + +static bool dsa_port_is_cpu(struct device_node *port) +{ + const char *name; + + name = of_get_property(port, "label", NULL); + if (!name) + return false; + + if (!strcmp(name, "cpu")) + return true; + + return false; +} + +static bool dsa_ds_find_port(struct dsa_switch *ds, + struct device_node *port) +{ + u32 index; + + for (index = 0; index < DSA_MAX_PORTS; index++) + if (ds->ports[index].dn == port) + return true; + return false; +} + +static struct dsa_switch *dsa_dst_find_port(struct dsa_switch_tree *dst, + struct device_node *port) +{ + struct dsa_switch *ds; + u32 index; + + for (index = 0; index < DSA_MAX_SWITCHES; index++) { + ds = dst->ds[index]; + if (!ds) + continue; + + if (dsa_ds_find_port(ds, port)) + return ds; + } + + return NULL; +} + +static int dsa_port_complete(struct dsa_switch_tree *dst, + struct dsa_switch *src_ds, + struct device_node *port, + u32 src_port) +{ + struct device_node *link; + int index; + struct dsa_switch *dst_ds; + + for (index = 0;; index++) { + link = of_parse_phandle(port, "link", index); + if (!link) + break; + + dst_ds = dsa_dst_find_port(dst, link); + of_node_put(link); + + if (!dst_ds) + return 1; + + src_ds->rtable[dst_ds->index] = src_port; + } + + return 0; +} + +/* A switch is complete if all the DSA ports phandles point to ports + * known in the tree. A return value of 1 means the tree is not + * complete. This is not an error condition. A value of 0 is + * success. + */ +static int dsa_ds_complete(struct dsa_switch_tree *dst, struct dsa_switch *ds) +{ + struct device_node *port; + u32 index; + int err; + + for (index = 0; index < DSA_MAX_PORTS; index++) { + port = ds->ports[index].dn; + if (!port) + continue; + + if (!dsa_port_is_dsa(port)) + continue; + + err = dsa_port_complete(dst, ds, port, index); + if (err != 0) + return err; + + ds->dsa_port_mask |= BIT(index); + } + + return 0; +} + +/* A tree is complete if all the DSA ports phandles point to ports + * known in the tree. A return value of 1 means the tree is not + * complete. This is not an error condition. A value of 0 is + * success. + */ +static int dsa_dst_complete(struct dsa_switch_tree *dst) +{ + struct dsa_switch *ds; + u32 index; + int err; + + for (index = 0; index < DSA_MAX_SWITCHES; index++) { + ds = dst->ds[index]; + if (!ds) + continue; + + err = dsa_ds_complete(dst, ds); + if (err != 0) + return err; + } + + return 0; +} + +static int dsa_dsa_port_apply(struct device_node *port, u32 index, + struct dsa_switch *ds) +{ + int err; + + err = dsa_cpu_dsa_setup(ds, ds->dev, port, index); + if (err) { + dev_warn(ds->dev, "Failed to setup dsa port %d: %d\n", + index, err); + return err; + } + + return 0; +} + +static void dsa_dsa_port_unapply(struct device_node *port, u32 index, + struct dsa_switch *ds) +{ + dsa_cpu_dsa_destroy(port); +} + +static int dsa_cpu_port_apply(struct device_node *port, u32 index, + struct dsa_switch *ds) +{ + int err; + + err = dsa_cpu_dsa_setup(ds, ds->dev, port, index); + if (err) { + dev_warn(ds->dev, "Failed to setup cpu port %d: %d\n", + index, err); + return err; + } + + ds->cpu_port_mask |= BIT(index); + + return 0; +} + +static void dsa_cpu_port_unapply(struct device_node *port, u32 index, + struct dsa_switch *ds) +{ + dsa_cpu_dsa_destroy(port); + ds->cpu_port_mask &= ~BIT(index); + +} + +static int dsa_user_port_apply(struct device_node *port, u32 index, + struct dsa_switch *ds) +{ + const char *name; + int err; + + name = of_get_property(port, "label", NULL); + + err = dsa_slave_create(ds, ds->dev, index, name); + if (err) { + dev_warn(ds->dev, "Failed to create slave %d: %d\n", + index, err); + return err; + } + + return 0; +} + +static void dsa_user_port_unapply(struct device_node *port, u32 index, + struct dsa_switch *ds) +{ + if (ds->ports[index].netdev) { + dsa_slave_destroy(ds->ports[index].netdev); + ds->ports[index].netdev = NULL; + } +} + +static int dsa_ds_apply(struct dsa_switch_tree *dst, struct dsa_switch *ds) +{ + struct device_node *port; + u32 index; + int err; + + err = ds->drv->setup(ds); + if (err < 0) + return err; + + err = ds->drv->set_addr(ds, dst->master_netdev->dev_addr); + if (err < 0) + return err; + + err = ds->drv->set_addr(ds, dst->master_netdev->dev_addr); + if (err < 0) + return err; + + for (index = 0; index < DSA_MAX_PORTS; index++) { + port = ds->ports[index].dn; + if (!port) + continue; + + if (dsa_port_is_dsa(port)) { + err = dsa_dsa_port_apply(port, index, ds); + if (err) + return err; + continue; + } + + if (dsa_port_is_cpu(port)) { + err = dsa_cpu_port_apply(port, index, ds); + if (err) + return err; + continue; + } + + err = dsa_user_port_apply(port, index, ds); + if (err) + continue; + } + + return 0; +} + +static void dsa_ds_unapply(struct dsa_switch_tree *dst, struct dsa_switch *ds) +{ + struct device_node *port; + u32 index; + + for (index = 0; index < DSA_MAX_PORTS; index++) { + port = ds->ports[index].dn; + if (!port) + continue; + + if (dsa_port_is_dsa(port)) { + dsa_dsa_port_unapply(port, index, ds); + continue; + } + + if (dsa_port_is_cpu(port)) { + dsa_cpu_port_unapply(port, index, ds); + continue; + } + + dsa_user_port_unapply(port, index, ds); + } +} + +static int dsa_dst_apply(struct dsa_switch_tree *dst) +{ + struct dsa_switch *ds; + u32 index; + int err; + + for (index = 0; index < DSA_MAX_SWITCHES; index++) { + ds = dst->ds[index]; + if (!ds) + continue; + + err = dsa_ds_apply(dst, ds); + if (err) + return err; + } + + /* If we use a tagging format that doesn't have an ethertype + * field, make sure that all packets from this point on get + * sent to the tag format's receive function. + */ + wmb(); + dst->master_netdev->dsa_ptr = (void *)dst; + dst->applied = true; + + return 0; +} + +static void dsa_dst_unapply(struct dsa_switch_tree *dst) +{ + struct dsa_switch *ds; + u32 index; + + if (!dst->applied) + return; + + dst->master_netdev->dsa_ptr = NULL; + + /* If we used a tagging format that doesn't have an ethertype + * field, make sure that all packets from this point get sent + * without the tag and go through the regular receive path. + */ + wmb(); + + for (index = 0; index < DSA_MAX_SWITCHES; index++) { + ds = dst->ds[index]; + if (!ds) + continue; + + dsa_ds_unapply(dst, ds); + } + + pr_info("DSA: tree %d unapplied\n", dst->tree); + dst->applied = false; +} + +static int dsa_cpu_parse(struct device_node *port, u32 index, + struct dsa_switch_tree *dst, + struct dsa_switch *ds) +{ + struct net_device *ethernet_dev; + struct device_node *ethernet; + + ethernet = of_parse_phandle(port, "ethernet", 0); + if (!ethernet) + return -EINVAL; + + ethernet_dev = of_find_net_device_by_node(ethernet); + if (!ethernet_dev) + return -EPROBE_DEFER; + + if (!ds->master_netdev) + ds->master_netdev = ethernet_dev; + + if (!dst->master_netdev) + dst->master_netdev = ethernet_dev; + + if (dst->cpu_switch == -1) { + dst->cpu_switch = ds->index; + dst->cpu_port = index; + } + + dst->tag_ops = dsa_resolve_tag_protocol(ds->drv->tag_protocol); + if (IS_ERR(dst->tag_ops)) { + dev_warn(ds->dev, "No tagger for this switch\n"); + return PTR_ERR(dst->tag_ops); + } + + dst->rcv = dst->tag_ops->rcv; + + return 0; +} + +static int dsa_ds_parse(struct dsa_switch_tree *dst, struct dsa_switch *ds) +{ + struct device_node *port; + u32 index; + int err; + + for (index = 0; index < DSA_MAX_PORTS; index++) { + port = ds->ports[index].dn; + if (!port) + continue; + + if (dsa_port_is_cpu(port)) { + err = dsa_cpu_parse(port, index, dst, ds); + if (err) + return err; + } + } + + pr_info("DSA: switch %d %d parsed\n", dst->tree, ds->index); + + return 0; +} + +static int dsa_dst_parse(struct dsa_switch_tree *dst) +{ + struct dsa_switch *ds; + u32 index; + int err; + + for (index = 0; index < DSA_MAX_SWITCHES; index++) { + ds = dst->ds[index]; + if (!ds) + continue; + + err = dsa_ds_parse(dst, ds); + if (err) + return err; + } + + if (!dst->master_netdev) { + pr_warn("Tree has no master device\n"); + return -EINVAL; + } + + pr_info("DSA: tree %d parsed\n", dst->tree); + + return 0; +} + +static int dsa_parse_ports_dn(struct device_node *ports, struct dsa_switch *ds) +{ + struct device_node *port; + int err; + u32 reg; + + for_each_available_child_of_node(ports, port) { + err = of_property_read_u32(port, "reg", ®); + if (err) + return err; + + if (reg >= DSA_MAX_PORTS) + return -EINVAL; + + ds->ports[reg].dn = port; + } + + return 0; +} + +static int dsa_parse_member(struct device_node *np, u32 *tree, u32 *index) +{ + int err; + + *tree = *index = 0; + + err = of_property_read_u32_index(np, "dsa,member", 0, tree); + if (err) { + /* Does not exist, but it is optional */ + if (err == -EINVAL) + return 0; + return err; + } + + err = of_property_read_u32_index(np, "dsa,member", 1, index); + if (err) + return err; + + if (*index >= DSA_MAX_SWITCHES) + return -EINVAL; + + return 0; +} + +static struct device_node *dsa_get_ports(struct dsa_switch *ds, + struct device_node *np) +{ + struct device_node *ports; + + ports = of_get_child_by_name(np, "ports"); + if (!ports) { + dev_err(ds->dev, "no ports child node found\n"); + return ERR_PTR(-EINVAL); + } + + return ports; +} + +static int _dsa_register_switch(struct dsa_switch *ds, struct device_node *np) +{ + struct device_node *ports = dsa_get_ports(ds, np); + struct dsa_switch_tree *dst; + u32 tree, index; + int err; + + err = dsa_parse_member(np, &tree, &index); + if (err) + return err; + + if (IS_ERR(ports)) + return PTR_ERR(ports); + + err = dsa_parse_ports_dn(ports, ds); + if (err) + return err; + + dst = dsa_get_dst(tree); + if (!dst) { + dst = dsa_add_dst(tree); + if (!dst) + return -ENOMEM; + } + + if (dst->ds[index]) { + err = -EBUSY; + goto out; + } + + ds->dst = dst; + ds->index = index; + dsa_dst_add_ds(dst, ds, index); + + err = dsa_dst_complete(dst); + if (err < 0) + goto out_del_dst; + + if (err == 1) { + /* Not all switches registered yet */ + err = 0; + goto out; + } + + if (dst->applied) { + pr_info("DSA: Disjoint trees?\n"); + return -EINVAL; + } + + err = dsa_dst_parse(dst); + if (err) + goto out_del_dst; + + err = dsa_dst_apply(dst); + if (err) { + dsa_dst_unapply(dst); + goto out_del_dst; + } + + dsa_put_dst(dst); + return 0; + +out_del_dst: + dsa_dst_del_ds(dst, ds, ds->index); +out: + dsa_put_dst(dst); + + return err; +} + +int dsa_register_switch(struct dsa_switch *ds, struct device_node *np) +{ + int err; + + mutex_lock(&dsa2_mutex); + err = _dsa_register_switch(ds, np); + mutex_unlock(&dsa2_mutex); + + return err; +} +EXPORT_SYMBOL_GPL(dsa_register_switch); + +void _dsa_unregister_switch(struct dsa_switch *ds) +{ + struct dsa_switch_tree *dst = ds->dst; + + dsa_dst_unapply(dst); + + dsa_dst_del_ds(dst, ds, ds->index); +} + +void dsa_unregister_switch(struct dsa_switch *ds) +{ + mutex_lock(&dsa2_mutex); + _dsa_unregister_switch(ds); + mutex_unlock(&dsa2_mutex); +} +EXPORT_SYMBOL_GPL(dsa_unregister_switch); diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index 72f7b89..b42f1a5 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -59,7 +59,7 @@ const struct dsa_device_ops *dsa_resolve_tag_protocol(int tag_protocol); extern const struct dsa_device_ops notag_netdev_ops; void dsa_slave_mii_bus_init(struct dsa_switch *ds); int dsa_slave_create(struct dsa_switch *ds, struct device *parent, - int port, char *name); + int port, const char *name); void dsa_slave_destroy(struct net_device *slave_dev); int dsa_slave_suspend(struct net_device *slave_dev); int dsa_slave_resume(struct net_device *slave_dev); diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 35e5f0f..15a4922 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -1099,14 +1099,18 @@ int dsa_slave_resume(struct net_device *slave_dev) } int dsa_slave_create(struct dsa_switch *ds, struct device *parent, - int port, char *name) + int port, const char *name) { - struct net_device *master = ds->dst->master_netdev; struct dsa_switch_tree *dst = ds->dst; + struct net_device *master; struct net_device *slave_dev; struct dsa_slave_priv *p; int ret; + master = ds->dst->master_netdev; + if (ds->master_netdev) + master = ds->master_netdev; + slave_dev = alloc_netdev(sizeof(struct dsa_slave_priv), name, NET_NAME_UNKNOWN, ether_setup); if (slave_dev == NULL) -- cgit v0.10.2 From 9dff67333af8b384e37a72fce5c251451f2e75c9 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:17:08 +0200 Subject: arm: dt: vf610-zii-devel-b: Make use of new DSA binding Hang the three switches of the three MDIO busses using the new DSA binding. Also, make use of the mdio-bus and explicitly list the phys on one device. This is not required, but good for testing. Signed-off-by: Andrew Lunn Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/arch/arm/boot/dts/vf610-zii-dev-rev-b.dts b/arch/arm/boot/dts/vf610-zii-dev-rev-b.dts index 6c60b7f..5c1fcab 100644 --- a/arch/arm/boot/dts/vf610-zii-dev-rev-b.dts +++ b/arch/arm/boot/dts/vf610-zii-dev-rev-b.dts @@ -85,187 +85,199 @@ reg = <1>; #address-cells = <1>; #size-cells = <0>; + + switch0: switch0@0 { + compatible = "marvell,mv88e6085"; + #address-cells = <1>; + #size-cells = <0>; + reg = <0>; + dsa,member = <0 0>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + port@0 { + reg = <0>; + label = "lan0"; + }; + + port@1 { + reg = <1>; + label = "lan1"; + }; + + port@2 { + reg = <2>; + label = "lan2"; + }; + + switch0port5: port@5 { + reg = <5>; + label = "dsa"; + phy-mode = "rgmii-txid"; + link = <&switch1port6 + &switch2port9>; + fixed-link { + speed = <1000>; + full-duplex; + }; + }; + + port@6 { + reg = <6>; + label = "cpu"; + ethernet = <&fec1>; + fixed-link { + speed = <100>; + full-duplex; + }; + }; + }; + }; }; mdio_mux_2: mdio@2 { reg = <2>; #address-cells = <1>; #size-cells = <0>; - }; - - mdio_mux_4: mdio@4 { - reg = <4>; - #address-cells = <1>; - #size-cells = <0>; - }; - - mdio_mux_8: mdio@8 { - reg = <8>; - #address-cells = <1>; - #size-cells = <0>; - }; - }; - - dsa { - compatible = "marvell,dsa"; - #address-cells = <2>; - #size-cells = <0>; - dsa,ethernet = <&fec1>; - dsa,mii-bus = <&mdio_mux_1>; - - /* 6352 - Primary - 7 ports */ - switch0: switch@0-0 { - #address-cells = <1>; - #size-cells = <0>; - reg = <0x00 0>; - eeprom-length = <512>; - port@0 { + switch1: switch1@0 { + compatible = "marvell,mv88e6085"; + #address-cells = <1>; + #size-cells = <0>; reg = <0>; - label = "lan0"; - }; - - port@1 { - reg = <1>; - label = "lan1"; - }; - - port@2 { - reg = <2>; - label = "lan2"; - }; - - switch0port5: port@5 { - reg = <5>; - label = "dsa"; - phy-mode = "rgmii-txid"; - link = <&switch1port6 - &switch2port9>; - - fixed-link { - speed = <1000>; - full-duplex; + dsa,member = <0 1>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + port@0 { + reg = <0>; + label = "lan3"; + phy-handle = <&switch1phy0>; + }; + + port@1 { + reg = <1>; + label = "lan4"; + phy-handle = <&switch1phy1>; + }; + + port@2 { + reg = <2>; + label = "lan5"; + phy-handle = <&switch1phy2>; + }; + + switch1port5: port@5 { + reg = <5>; + label = "dsa"; + link = <&switch2port9>; + phy-mode = "rgmii-txid"; + fixed-link { + speed = <1000>; + full-duplex; + }; + }; + + switch1port6: port@6 { + reg = <6>; + label = "dsa"; + phy-mode = "rgmii-txid"; + link = <&switch0port5>; + fixed-link { + speed = <1000>; + full-duplex; + }; + }; }; - }; - - port@6 { - reg = <6>; - label = "cpu"; - - fixed-link { - speed = <100>; - full-duplex; + mdio { + #address-cells = <1>; + #size-cells = <0>; + switch1phy0: switch1phy0@0 { + reg = <0>; + }; + switch1phy1: switch1phy0@1 { + reg = <1>; + }; + switch1phy2: switch1phy0@2 { + reg = <2>; + }; }; }; - }; - /* 6352 - Secondary - 7 ports */ - switch1: switch@0-1 { + mdio_mux_4: mdio@4 { #address-cells = <1>; #size-cells = <0>; - reg = <0x00 1>; - eeprom-length = <512>; - mii-bus = <&mdio_mux_2>; + reg = <4>; - port@0 { + switch2: switch2@0 { + compatible = "marvell,mv88e6085"; + #address-cells = <1>; + #size-cells = <0>; reg = <0>; - label = "lan3"; - }; - - port@1 { - reg = <1>; - label = "lan4"; - }; - - port@2 { - reg = <2>; - label = "lan5"; - }; - - switch1port5: port@5 { - reg = <5>; - label = "dsa"; - link = <&switch2port9>; - phy-mode = "rgmii-txid"; - - fixed-link { - speed = <1000>; - full-duplex; - }; - }; - - switch1port6: port@6 { - reg = <6>; - label = "dsa"; - phy-mode = "rgmii-txid"; - link = <&switch0port5>; - - fixed-link { - speed = <1000>; - full-duplex; + dsa,member = <0 2>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + port@0 { + reg = <0>; + label = "lan6"; + }; + + port@1 { + reg = <1>; + label = "lan7"; + }; + + port@2 { + reg = <2>; + label = "lan8"; + }; + + port@3 { + reg = <3>; + label = "optical3"; + fixed-link { + speed = <1000>; + full-duplex; + link-gpios = <&gpio6 2 + GPIO_ACTIVE_HIGH>; + }; + }; + + port@4 { + reg = <4>; + label = "optical4"; + fixed-link { + speed = <1000>; + full-duplex; + link-gpios = <&gpio6 3 + GPIO_ACTIVE_HIGH>; + }; + }; + + switch2port9: port@9 { + reg = <9>; + label = "dsa"; + phy-mode = "rgmii-txid"; + link = <&switch1port5 + &switch0port5>; + fixed-link { + speed = <1000>; + full-duplex; + }; + }; }; }; }; - /* 6185 - 10 ports */ - switch2: switch@0-2 { + mdio_mux_8: mdio@8 { + reg = <8>; #address-cells = <1>; #size-cells = <0>; - reg = <0x00 2>; - mii-bus = <&mdio_mux_4>; - - port@0 { - reg = <0>; - label = "lan6"; - }; - - port@1 { - reg = <1>; - label = "lan7"; - }; - - port@2 { - reg = <2>; - label = "lan8"; - }; - - port@3 { - reg = <3>; - label = "optical3"; - - fixed-link { - speed = <1000>; - full-duplex; - link-gpios = <&gpio6 2 - GPIO_ACTIVE_HIGH>; - }; - }; - - port@4 { - reg = <4>; - label = "optical4"; - - fixed-link { - speed = <1000>; - full-duplex; - link-gpios = <&gpio6 3 - GPIO_ACTIVE_HIGH>; - }; - }; - - switch2port9: port@9 { - reg = <9>; - label = "dsa"; - phy-mode = "rgmii-txid"; - link = <&switch1port5 - &switch0port5>; - - fixed-link { - speed = <1000>; - full-duplex; - }; - }; }; }; -- cgit v0.10.2 From 8c5ad1d6179df31eea1b736c88296db6e3ffd25d Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Sat, 4 Jun 2016 21:17:09 +0200 Subject: net: dsa: Document new binding Add the new binding to the documentation of the existing binding. Mark the old binding as deprecated. Signed-off-by: Andrew Lunn Signed-off-by: Florian Fainelli Reviewed-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/dsa/dsa.txt b/Documentation/devicetree/bindings/net/dsa/dsa.txt index 9f4807f..9bbbe7f 100644 --- a/Documentation/devicetree/bindings/net/dsa/dsa.txt +++ b/Documentation/devicetree/bindings/net/dsa/dsa.txt @@ -1,5 +1,279 @@ -Marvell Distributed Switch Architecture Device Tree Bindings ------------------------------------------------------------- +Distributed Switch Architecture Device Tree Bindings +---------------------------------------------------- + +Two bindings exist, one of which has been deprecated due to +limitations. + +Current Binding +--------------- + +Switches are true Linux devices and can be probes by any means. Once +probed, they register to the DSA framework, passing a node +pointer. This node is expected to fulfil the following binding, and +may contain additional properties as required by the device it is +embedded within. + +Required properties: + +- ports : A container for child nodes representing switch ports. + +Optional properties: + +- dsa,member : A two element list indicates which DSA cluster, and position + within the cluster a switch takes. <0 0> is cluster 0, + switch 0. <0 1> is cluster 0, switch 1. <1 0> is cluster 1, + switch 0. A switch not part of any cluster (single device + hanging off a CPU port) must not specify this property + +The ports container has the following properties + +Required properties: + +- #address-cells : Must be 1 +- #size-cells : Must be 0 + +Each port children node must have the following mandatory properties: +- reg : Describes the port address in the switch +- label : Describes the label associated with this port, which + will become the netdev name. Special labels are + "cpu" to indicate a CPU port and "dsa" to + indicate an uplink/downlink port between switches in + the cluster. + +A port labelled "dsa" has the following mandatory property: + +- link : Should be a list of phandles to other switch's DSA + port. This port is used as the outgoing port + towards the phandle ports. The full routing + information must be given, not just the one hop + routes to neighbouring switches. + +A port labelled "cpu" has the following mandatory property: + +- ethernet : Should be a phandle to a valid Ethernet device node. + This host device is what the switch port is + connected to. + +Port child nodes may also contain the following optional standardised +properties, described in binding documents: + +- phy-handle : Phandle to a PHY on an MDIO bus. See + Documentation/devicetree/bindings/net/ethernet.txt + for details. + +- phy-mode : See + Documentation/devicetree/bindings/net/ethernet.txt + for details. + +- fixed-link : Fixed-link subnode describing a link to a non-MDIO + managed entity. See + Documentation/devicetree/bindings/net/fixed-link.txt + for details. + +Example + +The following example shows three switches on three MDIO busses, +linked into one DSA cluster. + +&mdio1 { + #address-cells = <1>; + #size-cells = <0>; + + switch0: switch0@0 { + compatible = "marvell,mv88e6085"; + #address-cells = <1>; + #size-cells = <0>; + reg = <0>; + + dsa,member = <0 0>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + port@0 { + reg = <0>; + label = "lan0"; + }; + + port@1 { + reg = <1>; + label = "lan1"; + }; + + port@2 { + reg = <2>; + label = "lan2"; + }; + + switch0port5: port@5 { + reg = <5>; + label = "dsa"; + phy-mode = "rgmii-txid"; + link = <&switch1port6 + &switch2port9>; + fixed-link { + speed = <1000>; + full-duplex; + }; + }; + + port@6 { + reg = <6>; + label = "cpu"; + ethernet = <&fec1>; + fixed-link { + speed = <100>; + full-duplex; + }; + }; + }; + }; +}; + +&mdio2 { + #address-cells = <1>; + #size-cells = <0>; + + switch1: switch1@0 { + compatible = "marvell,mv88e6085"; + #address-cells = <1>; + #size-cells = <0>; + reg = <0>; + + dsa,member = <0 1>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + port@0 { + reg = <0>; + label = "lan3"; + phy-handle = <&switch1phy0>; + }; + + port@1 { + reg = <1>; + label = "lan4"; + phy-handle = <&switch1phy1>; + }; + + port@2 { + reg = <2>; + label = "lan5"; + phy-handle = <&switch1phy2>; + }; + + switch1port5: port@5 { + reg = <5>; + label = "dsa"; + link = <&switch2port9>; + phy-mode = "rgmii-txid"; + fixed-link { + speed = <1000>; + full-duplex; + }; + }; + + switch1port6: port@6 { + reg = <6>; + label = "dsa"; + phy-mode = "rgmii-txid"; + link = <&switch0port5>; + fixed-link { + speed = <1000>; + full-duplex; + }; + }; + }; + mdio-bus { + #address-cells = <1>; + #size-cells = <0>; + switch1phy0: switch1phy0@0 { + reg = <0>; + }; + switch1phy1: switch1phy0@1 { + reg = <1>; + }; + switch1phy2: switch1phy0@2 { + reg = <2>; + }; + }; + }; +}; + +&mdio4 { + #address-cells = <1>; + #size-cells = <0>; + + switch2: switch2@0 { + compatible = "marvell,mv88e6085"; + #address-cells = <1>; + #size-cells = <0>; + reg = <0>; + + dsa,member = <0 2>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + port@0 { + reg = <0>; + label = "lan6"; + }; + + port@1 { + reg = <1>; + label = "lan7"; + }; + + port@2 { + reg = <2>; + label = "lan8"; + }; + + port@3 { + reg = <3>; + label = "optical3"; + fixed-link { + speed = <1000>; + full-duplex; + link-gpios = <&gpio6 2 + GPIO_ACTIVE_HIGH>; + }; + }; + + port@4 { + reg = <4>; + label = "optical4"; + fixed-link { + speed = <1000>; + full-duplex; + link-gpios = <&gpio6 3 + GPIO_ACTIVE_HIGH>; + }; + }; + + switch2port9: port@9 { + reg = <9>; + label = "dsa"; + phy-mode = "rgmii-txid"; + link = <&switch1port5 + &switch0port5>; + fixed-link { + speed = <1000>; + full-duplex; + }; + }; + }; + }; +}; + +Deprecated Binding +------------------ + +The deprecated binding makes use of a platform device to represent the +switches. The switches themselves are not Linux devices, and make use +of an MDIO bus for management. Required properties: - compatible : Should be "marvell,dsa" -- cgit v0.10.2 From 6eb17e0df74c036eba7548915b37f009403fe09e Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:09 +0800 Subject: ACPI: bus: add stub acpi_dev_found() to linux/acpi.h acpi_dev_found() will be used to detect if a given ACPI device is in the system. It will be compiled in non-ACPI case, but the function is in acpi_bus.h and acpi_bus.h can only be used in ACPI case, so this patch add the stub function to linux/acpi.h to make compiled successfully in non-ACPI cases. Cc: Rafael J. Wysocki Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 288fac5..3025d19 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -543,6 +543,11 @@ struct platform_device *acpi_create_platform_device(struct acpi_device *); struct fwnode_handle; +static inline bool acpi_dev_found(const char *hid) +{ + return false; +} + static inline bool is_acpi_node(struct fwnode_handle *fwnode) { return false; -- cgit v0.10.2 From 4ae399241adba66ad72e5973a1004f37ffbe67cd Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:10 +0800 Subject: ACPI: bus: add stub acpi_evaluate_dsm() to linux/acpi.h acpi_evaluate_dsm() will be used to handle the _DSM method in ACPI case. It will be compiled in non-ACPI case, but the function is in acpi_bus.h and acpi_bus.h can only be used in ACPI case, so this patch add the stub function to linux/acpi.h to make compiled successfully in non-ACPI cases. Cc: Rafael J. Wysocki Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 3025d19..4d4bb49 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -659,6 +659,14 @@ static inline bool acpi_driver_match_device(struct device *dev, return false; } +static inline union acpi_object *acpi_evaluate_dsm(acpi_handle handle, + const u8 *uuid, + int rev, int func, + union acpi_object *argv4) +{ + return NULL; +} + static inline int acpi_device_uevent_modalias(struct device *dev, struct kobj_uevent_env *env) { -- cgit v0.10.2 From 16a9f36150aa3cc516e8df0d1a979b5ca0c907ba Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:11 +0800 Subject: net: hisilicon: cleanup to prepare for other cases Hns-mdio only supports DT case now. do some cleanup to prepare for introducing other cases later, no functional change. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns_mdio.c b/drivers/net/ethernet/hisilicon/hns_mdio.c index 765ddb3..297edc4 100644 --- a/drivers/net/ethernet/hisilicon/hns_mdio.c +++ b/drivers/net/ethernet/hisilicon/hns_mdio.c @@ -354,6 +354,9 @@ static int hns_mdio_reset(struct mii_bus *bus) struct hns_mdio_device *mdio_dev = (struct hns_mdio_device *)bus->priv; int ret; + if (!dev_of_node(bus->parent)) + return -ENOTSUPP; + if (!mdio_dev->subctrl_vbase) { dev_err(&bus->dev, "mdio sys ctl reg has not maped\n"); return -ENODEV; @@ -397,24 +400,6 @@ static int hns_mdio_reset(struct mii_bus *bus) } /** - * hns_mdio_bus_name - get mdio bus name - * @name: mdio bus name - * @np: mdio device node pointer - */ -static void hns_mdio_bus_name(char *name, struct device_node *np) -{ - const u32 *addr; - u64 taddr = OF_BAD_ADDR; - - addr = of_get_address(np, 0, NULL, NULL); - if (addr) - taddr = of_translate_address(np, addr); - - snprintf(name, MII_BUS_ID_SIZE, "%s@%llx", np->name, - (unsigned long long)taddr); -} - -/** * hns_mdio_probe - probe mdio device * @pdev: mdio platform device * @@ -422,17 +407,16 @@ static void hns_mdio_bus_name(char *name, struct device_node *np) */ static int hns_mdio_probe(struct platform_device *pdev) { - struct device_node *np; struct hns_mdio_device *mdio_dev; struct mii_bus *new_bus; struct resource *res; - int ret; + int ret = -ENODEV; if (!pdev) { dev_err(NULL, "pdev is NULL!\r\n"); return -ENODEV; } - np = pdev->dev.of_node; + mdio_dev = devm_kzalloc(&pdev->dev, sizeof(*mdio_dev), GFP_KERNEL); if (!mdio_dev) return -ENOMEM; @@ -448,7 +432,7 @@ static int hns_mdio_probe(struct platform_device *pdev) new_bus->write = hns_mdio_write; new_bus->reset = hns_mdio_reset; new_bus->priv = mdio_dev; - hns_mdio_bus_name(new_bus->id, np); + new_bus->parent = &pdev->dev; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); mdio_dev->vbase = devm_ioremap_resource(&pdev->dev, res); @@ -457,16 +441,20 @@ static int hns_mdio_probe(struct platform_device *pdev) return ret; } - mdio_dev->subctrl_vbase = - syscon_node_to_regmap(of_parse_phandle(np, "subctrl-vbase", 0)); - if (IS_ERR(mdio_dev->subctrl_vbase)) { - dev_warn(&pdev->dev, "no syscon hisilicon,peri-c-subctrl\n"); - mdio_dev->subctrl_vbase = NULL; - } - new_bus->parent = &pdev->dev; platform_set_drvdata(pdev, new_bus); + snprintf(new_bus->id, MII_BUS_ID_SIZE, "%s-%s", "Mii", + dev_name(&pdev->dev)); + if (dev_of_node(&pdev->dev)) { + mdio_dev->subctrl_vbase = syscon_node_to_regmap( + of_parse_phandle(pdev->dev.of_node, + "subctrl-vbase", 0)); + if (IS_ERR(mdio_dev->subctrl_vbase)) { + dev_warn(&pdev->dev, "no syscon hisilicon,peri-c-subctrl\n"); + mdio_dev->subctrl_vbase = NULL; + } + ret = of_mdiobus_register(new_bus, pdev->dev.of_node); + } - ret = of_mdiobus_register(new_bus, np); if (ret) { dev_err(&pdev->dev, "Cannot register as MDIO bus!\n"); platform_set_drvdata(pdev, NULL); -- cgit v0.10.2 From 8a99ff5ab4f4f209358b177b376660066c9c4400 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:12 +0800 Subject: net: hisilicon: add support of acpi for hns-mdio hns-mdio needs to register itself to mii-bus. The info of the device can be read by both DT and ACPI. HNS tries to call Linux PHY driver to help access PHY-devices, the HNS hardware topology is as below. The MDIO controller may control several PHY-devices, and each PHY-device connects to a MAC device. The MDIO will be registered to mdiobus, then PHY-devices will register when each mac find PHY device. cpu | | ------------------------------------------- | | | | | | | dsaf | MDIO | MDIO | --------------------------- | | | | | | | | | | | | | | MAC MAC MAC MAC | | | | | | | ---- |-------- |-------- | | -------- || || || || PHY PHY PHY PHY And the driver can handle reset sequence by _RST method in DSDT in ACPI case. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns_mdio.c b/drivers/net/ethernet/hisilicon/hns_mdio.c index 297edc4..761a32f 100644 --- a/drivers/net/ethernet/hisilicon/hns_mdio.c +++ b/drivers/net/ethernet/hisilicon/hns_mdio.c @@ -7,6 +7,7 @@ * (at your option) any later version. */ +#include #include #include #include @@ -354,48 +355,60 @@ static int hns_mdio_reset(struct mii_bus *bus) struct hns_mdio_device *mdio_dev = (struct hns_mdio_device *)bus->priv; int ret; - if (!dev_of_node(bus->parent)) - return -ENOTSUPP; + if (dev_of_node(bus->parent)) { + if (!mdio_dev->subctrl_vbase) { + dev_err(&bus->dev, "mdio sys ctl reg has not maped\n"); + return -ENODEV; + } - if (!mdio_dev->subctrl_vbase) { - dev_err(&bus->dev, "mdio sys ctl reg has not maped\n"); - return -ENODEV; - } + /* 1. reset req, and read reset st check */ + ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_RESET_REQ, 0x1, + MDIO_SC_RESET_ST, 0x1, + MDIO_CHECK_SET_ST); + if (ret) { + dev_err(&bus->dev, "MDIO reset fail\n"); + return ret; + } - /*1. reset req, and read reset st check*/ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_RESET_REQ, 0x1, - MDIO_SC_RESET_ST, 0x1, - MDIO_CHECK_SET_ST); - if (ret) { - dev_err(&bus->dev, "MDIO reset fail\n"); - return ret; - } + /* 2. dis clk, and read clk st check */ + ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_CLK_DIS, + 0x1, MDIO_SC_CLK_ST, 0x1, + MDIO_CHECK_CLR_ST); + if (ret) { + dev_err(&bus->dev, "MDIO dis clk fail\n"); + return ret; + } - /*2. dis clk, and read clk st check*/ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_CLK_DIS, - 0x1, MDIO_SC_CLK_ST, 0x1, - MDIO_CHECK_CLR_ST); - if (ret) { - dev_err(&bus->dev, "MDIO dis clk fail\n"); - return ret; - } + /* 3. reset dreq, and read reset st check */ + ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_RESET_DREQ, 0x1, + MDIO_SC_RESET_ST, 0x1, + MDIO_CHECK_CLR_ST); + if (ret) { + dev_err(&bus->dev, "MDIO dis clk fail\n"); + return ret; + } - /*3. reset dreq, and read reset st check*/ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_RESET_DREQ, 0x1, - MDIO_SC_RESET_ST, 0x1, - MDIO_CHECK_CLR_ST); - if (ret) { - dev_err(&bus->dev, "MDIO dis clk fail\n"); - return ret; + /* 4. en clk, and read clk st check */ + ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_CLK_EN, + 0x1, MDIO_SC_CLK_ST, 0x1, + MDIO_CHECK_SET_ST); + if (ret) + dev_err(&bus->dev, "MDIO en clk fail\n"); + } else if (is_acpi_node(bus->parent->fwnode)) { + acpi_status s; + + s = acpi_evaluate_object(ACPI_HANDLE(bus->parent), + "_RST", NULL, NULL); + if (ACPI_FAILURE(s)) { + dev_err(&bus->dev, "Reset failed, return:%#x\n", s); + ret = -EBUSY; + } else { + ret = 0; + } + } else { + dev_err(&bus->dev, "Can not get cfg data from DT or ACPI\n"); + ret = -ENXIO; } - - /*4. en clk, and read clk st check*/ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_CLK_EN, - 0x1, MDIO_SC_CLK_ST, 0x1, - MDIO_CHECK_SET_ST); - if (ret) - dev_err(&bus->dev, "MDIO en clk fail\n"); - return ret; } @@ -453,6 +466,18 @@ static int hns_mdio_probe(struct platform_device *pdev) mdio_dev->subctrl_vbase = NULL; } ret = of_mdiobus_register(new_bus, pdev->dev.of_node); + } else if (is_acpi_node(pdev->dev.fwnode)) { + /* Clear all the IRQ properties */ + memset(new_bus->irq, PHY_POLL, 4 * PHY_MAX_ADDR); + + /* Mask out all PHYs from auto probing. */ + new_bus->phy_mask = ~0; + + /* Register the MDIO bus */ + ret = mdiobus_register(new_bus); + } else { + dev_err(&pdev->dev, "Can not get cfg data from DT or ACPI\n"); + ret = -ENXIO; } if (ret) { @@ -487,12 +512,19 @@ static const struct of_device_id hns_mdio_match[] = { {} }; +static const struct acpi_device_id hns_mdio_acpi_match[] = { + { "HISI0141", 0 }, + { }, +}; +MODULE_DEVICE_TABLE(acpi, hns_mdio_acpi_match); + static struct platform_driver hns_mdio_driver = { .probe = hns_mdio_probe, .remove = hns_mdio_remove, .driver = { .name = MDIO_DRV_NAME, .of_match_table = hns_mdio_match, + .acpi_match_table = ACPI_PTR(hns_mdio_acpi_match), }, }; -- cgit v0.10.2 From 6162928c76dcba2504c88eff19cbc190995401d7 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:13 +0800 Subject: net: hns: use device_* APIs instead of of_* APIs OF series functions can be used only for DT case. Use unified device property function instead to support both DT and ACPI. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index 1c2ddb2..9afc5e6 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -50,7 +50,7 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) else dsaf_dev->dsaf_ver = AE_VERSION_2; - ret = of_property_read_string(np, "mode", &mode_str); + ret = device_property_read_string(dsaf_dev->dev, "mode", &mode_str); if (ret) { dev_err(dsaf_dev->dev, "get dsaf mode fail, ret=%d!\n", ret); return ret; @@ -142,7 +142,7 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) } } - ret = of_property_read_u32(np, "desc-num", &desc_num); + ret = device_property_read_u32(dsaf_dev->dev, "desc-num", &desc_num); if (ret < 0 || desc_num < HNS_DSAF_MIN_DESC_CNT || desc_num > HNS_DSAF_MAX_DESC_CNT) { dev_err(dsaf_dev->dev, "get desc-num(%d) fail, ret=%d!\n", @@ -151,14 +151,15 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) } dsaf_dev->desc_num = desc_num; - ret = of_property_read_u32(np, "reset-field-offset", &reset_offset); + ret = device_property_read_u32(dsaf_dev->dev, "reset-field-offset", + &reset_offset); if (ret < 0) { dev_dbg(dsaf_dev->dev, "get reset-field-offset fail, ret=%d!\r\n", ret); } dsaf_dev->reset_offset = reset_offset; - ret = of_property_read_u32(np, "buf-size", &buf_size); + ret = device_property_read_u32(dsaf_dev->dev, "buf-size", &buf_size); if (ret < 0) { dev_err(dsaf_dev->dev, "get buf-size fail, ret=%d!\r\n", ret); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index e621636..8851420 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -1067,13 +1067,8 @@ void hns_nic_update_stats(struct net_device *netdev) static void hns_init_mac_addr(struct net_device *ndev) { struct hns_nic_priv *priv = netdev_priv(ndev); - struct device_node *node = priv->dev->of_node; - const void *mac_addr_temp; - mac_addr_temp = of_get_mac_address(node); - if (mac_addr_temp && is_valid_ether_addr(mac_addr_temp)) { - memcpy(ndev->dev_addr, mac_addr_temp, ndev->addr_len); - } else { + if (!device_get_mac_address(priv->dev, ndev->dev_addr, ETH_ALEN)) { eth_hw_addr_random(ndev); dev_warn(priv->dev, "No valid mac, use random mac %pM", ndev->dev_addr); @@ -1898,10 +1893,10 @@ static int hns_nic_dev_probe(struct platform_device *pdev) goto out_read_prop_fail; } /* try to find port-idx-in-ae first */ - ret = of_property_read_u32(node, "port-idx-in-ae", &port_id); + ret = device_property_read_u32(dev, "port-idx-in-ae", &port_id); if (ret) { /* only for old code compatible */ - ret = of_property_read_u32(node, "port-id", &port_id); + ret = device_property_read_u32(dev, "port-id", &port_id); if (ret) goto out_read_prop_fail; /* for old dts, we need to caculate the port offset */ -- cgit v0.10.2 From 72fdddcefe603b9b7de3f1d6177dd413eceacb93 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:14 +0800 Subject: net: hns: use platform_get_irq instead of irq_of_parse_and_map As irq_of_parse_and_map is only used by DT case, it is excepted to use a uniform interface. So it is used platform_get_irq() instead. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c index 4ef6d23..3ce2409 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c @@ -458,7 +458,6 @@ void hns_rcb_get_cfg(struct rcb_common_cb *rcb_common) u32 i; u32 ring_num = rcb_common->ring_num; int base_irq_idx = hns_rcb_get_base_irq_idx(rcb_common); - struct device_node *np = rcb_common->dsaf_dev->dev->of_node; struct platform_device *pdev = to_platform_device(rcb_common->dsaf_dev->dev); bool is_ver1 = AE_IS_VER1(rcb_common->dsaf_dev->dsaf_ver); @@ -473,10 +472,10 @@ void hns_rcb_get_cfg(struct rcb_common_cb *rcb_common) ring_pair_cb->port_id_in_comm = hns_rcb_get_port_in_comm(rcb_common, i); ring_pair_cb->virq[HNS_RCB_IRQ_IDX_TX] = - is_ver1 ? irq_of_parse_and_map(np, base_irq_idx + i * 2) : + is_ver1 ? platform_get_irq(pdev, base_irq_idx + i * 2) : platform_get_irq(pdev, base_irq_idx + i * 3 + 1); ring_pair_cb->virq[HNS_RCB_IRQ_IDX_RX] = - is_ver1 ? irq_of_parse_and_map(np, base_irq_idx + i * 2 + 1) : + is_ver1 ? platform_get_irq(pdev, base_irq_idx + i * 2 + 1) : platform_get_irq(pdev, base_irq_idx + i * 3); ring_pair_cb->q.phy_base = RCB_COMM_BASE_TO_RING_BASE(rcb_common->phy_base, i); -- cgit v0.10.2 From 7b2acae6fb28f0b3d32c0620f8045c4a1bf5bb92 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:15 +0800 Subject: net: hns: enet specify a reference to dsaf by fwnode_handle As device_node is only used by DT case, it is expected to find uniform ways. So fwnode_handle is the suitable method. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.c b/drivers/net/ethernet/hisilicon/hns/hnae.c index 3bfe36f..d630acd 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.c +++ b/drivers/net/ethernet/hisilicon/hns/hnae.c @@ -96,16 +96,16 @@ static int __ae_match(struct device *dev, const void *data) { struct hnae_ae_dev *hdev = cls_to_ae_dev(dev); - return hdev->dev->of_node == data; + return (data == &hdev->dev->of_node->fwnode); } -static struct hnae_ae_dev *find_ae(const struct device_node *ae_node) +static struct hnae_ae_dev *find_ae(const struct fwnode_handle *fwnode) { struct device *dev; - WARN_ON(!ae_node); + WARN_ON(!fwnode); - dev = class_find_device(hnae_class, NULL, ae_node, __ae_match); + dev = class_find_device(hnae_class, NULL, fwnode, __ae_match); return dev ? cls_to_ae_dev(dev) : NULL; } @@ -312,7 +312,7 @@ EXPORT_SYMBOL(hnae_reinit_handle); * return handle ptr or ERR_PTR */ struct hnae_handle *hnae_get_handle(struct device *owner_dev, - const struct device_node *ae_node, + const struct fwnode_handle *fwnode, u32 port_id, struct hnae_buf_ops *bops) { @@ -321,7 +321,7 @@ struct hnae_handle *hnae_get_handle(struct device *owner_dev, int i, j; int ret; - dev = find_ae(ae_node); + dev = find_ae(fwnode); if (!dev) return ERR_PTR(-ENODEV); diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.h b/drivers/net/ethernet/hisilicon/hns/hnae.h index e8d36aa..f5f8140 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.h +++ b/drivers/net/ethernet/hisilicon/hns/hnae.h @@ -528,7 +528,7 @@ struct hnae_handle { #define ring_to_dev(ring) ((ring)->q->dev->dev) struct hnae_handle *hnae_get_handle(struct device *owner_dev, - const struct device_node *ae_node, + const struct fwnode_handle *fwnode, u32 port_id, struct hnae_buf_ops *bops); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index 8851420..93f6ccb 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -1807,7 +1807,7 @@ static int hns_nic_try_get_ae(struct net_device *ndev) int ret; h = hnae_get_handle(&priv->netdev->dev, - priv->ae_node, priv->port_id, NULL); + priv->fwnode, priv->port_id, NULL); if (IS_ERR_OR_NULL(h)) { ret = -ENODEV; dev_dbg(priv->dev, "has not handle, register notifier!\n"); @@ -1867,7 +1867,7 @@ static int hns_nic_dev_probe(struct platform_device *pdev) struct device *dev = &pdev->dev; struct net_device *ndev; struct hns_nic_priv *priv; - struct device_node *node = dev->of_node; + struct device_node *ae_node; u32 port_id; int ret; @@ -1881,17 +1881,19 @@ static int hns_nic_dev_probe(struct platform_device *pdev) priv->dev = dev; priv->netdev = ndev; - if (of_device_is_compatible(node, "hisilicon,hns-nic-v1")) + if (of_device_is_compatible(dev->of_node, "hisilicon,hns-nic-v1")) priv->enet_ver = AE_VERSION_1; else priv->enet_ver = AE_VERSION_2; - priv->ae_node = (void *)of_parse_phandle(node, "ae-handle", 0); - if (IS_ERR_OR_NULL(priv->ae_node)) { - ret = PTR_ERR(priv->ae_node); + ae_node = of_parse_phandle(dev->of_node, "ae-handle", 0); + if (IS_ERR_OR_NULL(ae_node)) { + ret = PTR_ERR(ae_node); dev_err(dev, "not find ae-handle\n"); goto out_read_prop_fail; } + priv->fwnode = &ae_node->fwnode; + /* try to find port-idx-in-ae first */ ret = device_property_read_u32(dev, "port-idx-in-ae", &port_id); if (ret) { diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.h b/drivers/net/ethernet/hisilicon/hns/hns_enet.h index 337efa5..44bb301 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.h @@ -54,7 +54,7 @@ struct hns_nic_ops { }; struct hns_nic_priv { - const struct device_node *ae_node; + const struct fwnode_handle *fwnode; u32 enet_ver; u32 port_id; int phy_mode; -- cgit v0.10.2 From 652d39b0d50a6e99495442431c8dadeb8c05c986 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:16 +0800 Subject: net: hns: add uniform interface for phy connection As device_node is only used by DT case, HNS needs to treat the other cases including ACPI. It needs to use uniform ways to handle both of DT and ACPI. This patch chooses phy_device, and of_phy_connect and of_phy_attach are only used by DT case. It needs to use uniform interface to handle that sequence by both DT and ACPI. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.c b/drivers/net/ethernet/hisilicon/hns/hnae.c index d630acd..5d3047c 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.c +++ b/drivers/net/ethernet/hisilicon/hns/hnae.c @@ -96,7 +96,13 @@ static int __ae_match(struct device *dev, const void *data) { struct hnae_ae_dev *hdev = cls_to_ae_dev(dev); - return (data == &hdev->dev->of_node->fwnode); + if (dev_of_node(hdev->dev)) + return (data == &hdev->dev->of_node->fwnode); + else if (is_acpi_node(hdev->dev->fwnode)) + return (data == hdev->dev->fwnode); + + dev_err(dev, "__ae_match cannot read cfg data from OF or acpi\n"); + return 0; } static struct hnae_ae_dev *find_ae(const struct fwnode_handle *fwnode) diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.h b/drivers/net/ethernet/hisilicon/hns/hnae.h index f5f8140..529cb13 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.h +++ b/drivers/net/ethernet/hisilicon/hns/hnae.h @@ -27,6 +27,7 @@ * "cb" means control block */ +#include #include #include #include @@ -512,7 +513,7 @@ struct hnae_ae_dev { struct hnae_handle { struct device *owner_dev; /* the device which make use of this handle */ struct hnae_ae_dev *dev; /* the device who provides this handle */ - struct device_node *phy_node; + struct phy_device *phy_dev; phy_interface_t phy_if; u32 if_support; int q_num; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c index 7a757e8..8e009f4 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c @@ -131,7 +131,7 @@ struct hnae_handle *hns_ae_get_handle(struct hnae_ae_dev *dev, vf_cb->mac_cb = dsaf_dev->mac_cb[port_id]; ae_handle->phy_if = vf_cb->mac_cb->phy_if; - ae_handle->phy_node = vf_cb->mac_cb->phy_node; + ae_handle->phy_dev = vf_cb->mac_cb->phy_dev; ae_handle->if_support = vf_cb->mac_cb->if_support; ae_handle->port_type = vf_cb->mac_cb->mac_type; ae_handle->dport_id = port_id; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c index 611581f..527b49d 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c @@ -15,7 +15,8 @@ #include #include #include -#include +#include +#include #include #include "hns_dsaf_main.h" @@ -645,7 +646,7 @@ free_mac_drv: */ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) { - struct device_node *np = mac_cb->dev->of_node; + struct device_node *np; struct regmap *syscon; struct of_phandle_args cpld_args; u32 ret; @@ -672,21 +673,34 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) * from dsaf node */ if (!mac_cb->fw_port) { - mac_cb->phy_node = of_parse_phandle(np, "phy-handle", - mac_cb->mac_id); - if (mac_cb->phy_node) + np = of_parse_phandle(mac_cb->dev->of_node, "phy-handle", + mac_cb->mac_id); + mac_cb->phy_dev = of_phy_find_device(np); + if (mac_cb->phy_dev) { + /* refcount is held by of_phy_find_device() + * if the phy_dev is found + */ + put_device(&mac_cb->phy_dev->mdio.dev); + dev_dbg(mac_cb->dev, "mac%d phy_node: %s\n", - mac_cb->mac_id, mac_cb->phy_node->name); + mac_cb->mac_id, np->name); + } + return 0; } + if (!is_of_node(mac_cb->fw_port)) return -EINVAL; + /* parse property from port subnode in dsaf */ - mac_cb->phy_node = of_parse_phandle(to_of_node(mac_cb->fw_port), - "phy-handle", 0); - if (mac_cb->phy_node) + np = of_parse_phandle(to_of_node(mac_cb->fw_port), "phy-handle", 0); + mac_cb->phy_dev = of_phy_find_device(np); + if (mac_cb->phy_dev) { + put_device(&mac_cb->phy_dev->mdio.dev); dev_dbg(mac_cb->dev, "mac%d phy_node: %s\n", - mac_cb->mac_id, mac_cb->phy_node->name); + mac_cb->mac_id, np->name); + } + syscon = syscon_node_to_regmap( of_parse_phandle(to_of_node(mac_cb->fw_port), "serdes-syscon", 0)); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h index 97ce9a7..89b49d7 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h @@ -338,7 +338,7 @@ struct hns_mac_cb { phy_interface_t phy_if; enum hnae_loop loop_mode; - struct device_node *phy_node; + struct phy_device *phy_dev; struct mac_hw_stats hw_stats; }; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c index a837bb9..a843a86 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c @@ -332,7 +332,7 @@ int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, u8 en) int sfp_prsnt; int ret = hns_mac_get_sfp_prsnt(mac_cb, &sfp_prsnt); - if (!mac_cb->phy_node) { + if (!mac_cb->phy_dev) { if (ret) pr_info("please confirm sfp is present or not\n"); else diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index 93f6ccb..3ec3c27 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -996,19 +996,22 @@ static void hns_nic_adjust_link(struct net_device *ndev) int hns_nic_init_phy(struct net_device *ndev, struct hnae_handle *h) { struct hns_nic_priv *priv = netdev_priv(ndev); - struct phy_device *phy_dev = NULL; + struct phy_device *phy_dev = h->phy_dev; + int ret; - if (!h->phy_node) + if (!h->phy_dev) return 0; - if (h->phy_if != PHY_INTERFACE_MODE_XGMII) - phy_dev = of_phy_connect(ndev, h->phy_node, - hns_nic_adjust_link, 0, h->phy_if); - else - phy_dev = of_phy_attach(ndev, h->phy_node, 0, h->phy_if); + if (h->phy_if != PHY_INTERFACE_MODE_XGMII) { + phy_dev->dev_flags = 0; - if (unlikely(!phy_dev) || IS_ERR(phy_dev)) - return !phy_dev ? -ENODEV : PTR_ERR(phy_dev); + ret = phy_connect_direct(ndev, phy_dev, hns_nic_adjust_link, + h->phy_if); + } else { + ret = phy_attach_direct(ndev, phy_dev, 0, h->phy_if); + } + if (unlikely(ret)) + return -ENODEV; phy_dev->supported &= h->if_support; phy_dev->advertising = phy_dev->supported; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c index 67a648c..a809f52 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c @@ -596,7 +596,7 @@ static void hns_nic_self_test(struct net_device *ndev, st_param[1][0] = MAC_INTERNALLOOP_SERDES; st_param[1][1] = 1; /*serdes must exist*/ st_param[2][0] = MAC_INTERNALLOOP_PHY; /* only supporte phy node*/ - st_param[2][1] = ((!!(priv->ae_handle->phy_node)) && + st_param[2][1] = ((!!(priv->ae_handle->phy_dev)) && (priv->ae_handle->phy_if != PHY_INTERFACE_MODE_XGMII)); if (eth_test->flags == ETH_TEST_FL_OFFLINE) { -- cgit v0.10.2 From a24274aa5c2328a6ef4296d1ca8e81648cd0ddda Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:17 +0800 Subject: net: hns: add dsaf misc operation method The misc operation for different hw platform may be different, if using current implementation, it will add a new branch on each function for every new hw platform, so we add a method for this operation. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c index 8e009f4..d37b778 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c @@ -637,13 +637,15 @@ static int hns_ae_config_loopback(struct hnae_handle *handle, int ret; struct hnae_vf_cb *vf_cb = hns_ae_get_vf_cb(handle); struct hns_mac_cb *mac_cb = hns_get_mac_cb(handle); + struct dsaf_device *dsaf_dev = mac_cb->dsaf_dev; switch (loop) { case MAC_INTERNALLOOP_PHY: ret = 0; break; case MAC_INTERNALLOOP_SERDES: - ret = hns_mac_config_sds_loopback(vf_cb->mac_cb, en); + ret = dsaf_dev->misc_op->cfg_serdes_loopback(vf_cb->mac_cb, + !!en); break; case MAC_INTERNALLOOP_MAC: ret = hns_mac_config_mac_loopback(vf_cb->mac_cb, loop, en); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.c index 44abb08..1235c7f 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_gmac.c @@ -110,7 +110,7 @@ static void hns_gmac_free(void *mac_drv) u32 mac_id = drv->mac_id; - hns_dsaf_ge_srst_by_port(dsaf_dev, mac_id, 0); + dsaf_dev->misc_op->ge_srst(dsaf_dev, mac_id, 0); } static void hns_gmac_set_tx_auto_pause_frames(void *mac_drv, u16 newval) @@ -317,9 +317,9 @@ static void hns_gmac_init(void *mac_drv) port = drv->mac_id; - hns_dsaf_ge_srst_by_port(dsaf_dev, port, 0); + dsaf_dev->misc_op->ge_srst(dsaf_dev, port, 0); mdelay(10); - hns_dsaf_ge_srst_by_port(dsaf_dev, port, 1); + dsaf_dev->misc_op->ge_srst(dsaf_dev, port, 1); mdelay(10); hns_gmac_disable(mac_drv, MAC_COMM_MODE_RX_AND_TX); hns_gmac_tx_loop_pkt_dis(mac_drv); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c index 527b49d..2ebf14a 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c @@ -95,7 +95,7 @@ void hns_mac_get_link_status(struct hns_mac_cb *mac_cb, u32 *link_status) else *link_status = 0; - ret = hns_mac_get_sfp_prsnt(mac_cb, &sfp_prsnt); + ret = mac_cb->dsaf_dev->misc_op->get_sfp_prsnt(mac_cb, &sfp_prsnt); if (!ret) *link_status = *link_status && sfp_prsnt; @@ -512,7 +512,7 @@ void hns_mac_stop(struct hns_mac_cb *mac_cb) mac_ctrl_drv->mac_en_flg = 0; mac_cb->link = 0; - cpld_led_reset(mac_cb); + mac_cb->dsaf_dev->misc_op->cpld_reset_led(mac_cb); } /** @@ -804,7 +804,7 @@ int hns_mac_get_cfg(struct dsaf_device *dsaf_dev, struct hns_mac_cb *mac_cb) else mac_cb->mac_type = HNAE_PORT_DEBUG; - mac_cb->phy_if = hns_mac_get_phy_if(mac_cb); + mac_cb->phy_if = dsaf_dev->misc_op->get_phy_if(mac_cb); ret = hns_mac_get_mode(mac_cb->phy_if); if (ret < 0) { @@ -819,7 +819,7 @@ int hns_mac_get_cfg(struct dsaf_device *dsaf_dev, struct hns_mac_cb *mac_cb) if (ret) return ret; - cpld_led_reset(mac_cb); + mac_cb->dsaf_dev->misc_op->cpld_reset_led(mac_cb); mac_cb->vaddr = hns_mac_get_vaddr(dsaf_dev, mac_cb, mac_mode_idx); return 0; @@ -906,7 +906,7 @@ void hns_mac_uninit(struct dsaf_device *dsaf_dev) int max_port_num = hns_mac_get_max_port_num(dsaf_dev); for (i = 0; i < max_port_num; i++) { - cpld_led_reset(dsaf_dev->mac_cb[i]); + dsaf_dev->misc_op->cpld_reset_led(dsaf_dev->mac_cb[i]); dsaf_dev->mac_cb[i] = NULL; } } @@ -989,7 +989,7 @@ void hns_set_led_opt(struct hns_mac_cb *mac_cb) nic_data = 0; mac_cb->txpkt_for_led = mac_cb->hw_stats.tx_good_pkts; mac_cb->rxpkt_for_led = mac_cb->hw_stats.rx_good_pkts; - hns_cpld_set_led(mac_cb, (int)mac_cb->link, + mac_cb->dsaf_dev->misc_op->cpld_set_led(mac_cb, (int)mac_cb->link, mac_cb->speed, nic_data); } @@ -999,5 +999,5 @@ int hns_cpld_led_set_id(struct hns_mac_cb *mac_cb, if (!mac_cb || !mac_cb->cpld_ctrl) return 0; - return cpld_set_led_id(mac_cb, status); + return mac_cb->dsaf_dev->misc_op->cpld_set_led_id(mac_cb, status); } diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h index 89b49d7..05a6e8f 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h @@ -448,8 +448,6 @@ int hns_mac_set_pauseparam(struct hns_mac_cb *mac_cb, u32 rx_en, u32 tx_en); int hns_mac_set_mtu(struct hns_mac_cb *mac_cb, u32 new_mtu); int hns_mac_get_port_info(struct hns_mac_cb *mac_cb, u8 *auto_neg, u16 *speed, u8 *duplex); -phy_interface_t hns_mac_get_phy_if(struct hns_mac_cb *mac_cb); -int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, u8 en); int hns_mac_config_mac_loopback(struct hns_mac_cb *mac_cb, enum hnae_loop loop, int en); void hns_mac_update_stats(struct hns_mac_cb *mac_cb); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index 9afc5e6..28bde08 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -24,6 +24,7 @@ #include "hns_dsaf_main.h" #include "hns_dsaf_ppe.h" #include "hns_dsaf_rcb.h" +#include "hns_dsaf_misc.h" const char *g_dsaf_mode_match[DSAF_MODE_MAX] = { [DSAF_MODE_DISABLE_2PORT_64VM] = "2port-64vf", @@ -174,6 +175,10 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) goto unmap_base_addr; } + dsaf_dev->misc_op = hns_misc_op_get(dsaf_dev); + if (!dsaf_dev->misc_op) + return -ENOMEM; + if (!dma_set_mask_and_coherent(dsaf_dev->dev, DMA_BIT_MASK(64ULL))) dev_dbg(dsaf_dev->dev, "set mask to 64bit\n"); else @@ -1296,9 +1301,9 @@ static int hns_dsaf_init_hw(struct dsaf_device *dsaf_dev) dev_dbg(dsaf_dev->dev, "hns_dsaf_init_hw begin %s !\n", dsaf_dev->ae_dev.name); - hns_dsaf_rst(dsaf_dev, 0); + dsaf_dev->misc_op->dsaf_reset(dsaf_dev, 0); mdelay(10); - hns_dsaf_rst(dsaf_dev, 1); + dsaf_dev->misc_op->dsaf_reset(dsaf_dev, 1); hns_dsaf_comm_init(dsaf_dev); @@ -1326,7 +1331,7 @@ static int hns_dsaf_init_hw(struct dsaf_device *dsaf_dev) static void hns_dsaf_remove_hw(struct dsaf_device *dsaf_dev) { /*reset*/ - hns_dsaf_rst(dsaf_dev, 0); + dsaf_dev->misc_op->dsaf_reset(dsaf_dev, 0); } /** diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h index f0502ba..2e55b3c 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h @@ -268,6 +268,27 @@ struct dsaf_int_stat { }; +struct dsaf_misc_op { + void (*cpld_set_led)(struct hns_mac_cb *mac_cb, int link_status, + u16 speed, int data); + void (*cpld_reset_led)(struct hns_mac_cb *mac_cb); + int (*cpld_set_led_id)(struct hns_mac_cb *mac_cb, + enum hnae_led_state status); + /* reset seris function, it will be reset if the dereseet is 0 */ + void (*dsaf_reset)(struct dsaf_device *dsaf_dev, bool dereset); + void (*xge_srst)(struct dsaf_device *dsaf_dev, u32 port, bool dereset); + void (*xge_core_srst)(struct dsaf_device *dsaf_dev, u32 port, + bool dereset); + void (*ge_srst)(struct dsaf_device *dsaf_dev, u32 port, bool dereset); + void (*ppe_srst)(struct dsaf_device *dsaf_dev, u32 port, bool dereset); + void (*ppe_comm_srst)(struct dsaf_device *dsaf_dev, bool dereset); + + phy_interface_t (*get_phy_if)(struct hns_mac_cb *mac_cb); + int (*get_sfp_prsnt)(struct hns_mac_cb *mac_cb, int *sfp_prsnt); + + int (*cfg_serdes_loopback)(struct hns_mac_cb *mac_cb, bool en); +}; + /* Dsaf device struct define ,and mac -> dsaf */ struct dsaf_device { struct device *dev; @@ -292,6 +313,7 @@ struct dsaf_device { struct ppe_common_cb *ppe_common[DSAF_COMM_DEV_NUM]; struct rcb_common_cb *rcb_common[DSAF_COMM_DEV_NUM]; struct hns_mac_cb *mac_cb[DSAF_MAX_PORT_NUM]; + struct dsaf_misc_op *misc_op; struct dsaf_hw_stats hw_stats[DSAF_NODE_NUM]; struct dsaf_int_stat int_stat; @@ -388,22 +410,11 @@ int hns_dsaf_get_mac_entry_by_index( u16 entry_index, struct dsaf_drv_mac_multi_dest_entry *mac_entry); -void hns_dsaf_rst(struct dsaf_device *dsaf_dev, u32 val); - -void hns_ppe_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val); - -void hns_ppe_com_srst(struct ppe_common_cb *ppe_common, u32 val); - void hns_dsaf_fix_mac_mode(struct hns_mac_cb *mac_cb); int hns_dsaf_ae_init(struct dsaf_device *dsaf_dev); void hns_dsaf_ae_uninit(struct dsaf_device *dsaf_dev); -void hns_dsaf_xge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val); -void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val); -void hns_dsaf_xge_core_srst_by_port(struct dsaf_device *dsaf_dev, - u32 port, u32 val); - void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 inode_num); int hns_dsaf_get_sset_count(int stringset); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c index a843a86..e1c3e0d 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c @@ -32,8 +32,8 @@ static u32 dsaf_read_sub(struct dsaf_device *dsaf_dev, u32 reg) return ret; } -void hns_cpld_set_led(struct hns_mac_cb *mac_cb, int link_status, - u16 speed, int data) +static void hns_cpld_set_led(struct hns_mac_cb *mac_cb, int link_status, + u16 speed, int data) { int speed_reg = 0; u8 value; @@ -71,7 +71,7 @@ void hns_cpld_set_led(struct hns_mac_cb *mac_cb, int link_status, } } -void cpld_led_reset(struct hns_mac_cb *mac_cb) +static void cpld_led_reset(struct hns_mac_cb *mac_cb) { if (!mac_cb || !mac_cb->cpld_ctrl) return; @@ -81,8 +81,8 @@ void cpld_led_reset(struct hns_mac_cb *mac_cb) mac_cb->cpld_led_value = CPLD_LED_DEFAULT_VALUE; } -int cpld_set_led_id(struct hns_mac_cb *mac_cb, - enum hnae_led_state status) +static int cpld_set_led_id(struct hns_mac_cb *mac_cb, + enum hnae_led_state status) { switch (status) { case HNAE_LED_ACTIVE: @@ -109,12 +109,12 @@ int cpld_set_led_id(struct hns_mac_cb *mac_cb, #define RESET_REQ_OR_DREQ 1 -void hns_dsaf_rst(struct dsaf_device *dsaf_dev, u32 val) +static void hns_dsaf_rst(struct dsaf_device *dsaf_dev, bool dereset) { u32 xbar_reg_addr; u32 nt_reg_addr; - if (!val) { + if (!dereset) { xbar_reg_addr = DSAF_SUB_SC_XBAR_RESET_REQ_REG; nt_reg_addr = DSAF_SUB_SC_NT_RESET_REQ_REG; } else { @@ -126,7 +126,8 @@ void hns_dsaf_rst(struct dsaf_device *dsaf_dev, u32 val) dsaf_write_sub(dsaf_dev, nt_reg_addr, RESET_REQ_OR_DREQ); } -void hns_dsaf_xge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val) +static void hns_dsaf_xge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, + bool dereset) { u32 reg_val = 0; u32 reg_addr; @@ -137,7 +138,7 @@ void hns_dsaf_xge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val) reg_val |= RESET_REQ_OR_DREQ; reg_val |= 0x2082082 << dsaf_dev->mac_cb[port]->port_rst_off; - if (val == 0) + if (!dereset) reg_addr = DSAF_SUB_SC_XGE_RESET_REQ_REG; else reg_addr = DSAF_SUB_SC_XGE_RESET_DREQ_REG; @@ -145,8 +146,8 @@ void hns_dsaf_xge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val) dsaf_write_sub(dsaf_dev, reg_addr, reg_val); } -void hns_dsaf_xge_core_srst_by_port(struct dsaf_device *dsaf_dev, - u32 port, u32 val) +static void hns_dsaf_xge_core_srst_by_port(struct dsaf_device *dsaf_dev, + u32 port, bool dereset) { u32 reg_val = 0; u32 reg_addr; @@ -157,7 +158,7 @@ void hns_dsaf_xge_core_srst_by_port(struct dsaf_device *dsaf_dev, reg_val |= XGMAC_TRX_CORE_SRST_M << dsaf_dev->mac_cb[port]->port_rst_off; - if (val == 0) + if (!dereset) reg_addr = DSAF_SUB_SC_XGE_RESET_REQ_REG; else reg_addr = DSAF_SUB_SC_XGE_RESET_DREQ_REG; @@ -165,7 +166,8 @@ void hns_dsaf_xge_core_srst_by_port(struct dsaf_device *dsaf_dev, dsaf_write_sub(dsaf_dev, reg_addr, reg_val); } -void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val) +static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, + bool dereset) { u32 reg_val_1; u32 reg_val_2; @@ -183,7 +185,7 @@ void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val) else reg_val_2 = 0x2082082 << port_rst_off; - if (val == 0) { + if (!dereset) { dsaf_write_sub(dsaf_dev, DSAF_SUB_SC_GE_RESET_REQ1_REG, reg_val_1); @@ -200,7 +202,7 @@ void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val) reg_val_1 = 0x15540 << dsaf_dev->reset_offset; reg_val_2 = 0x100 << dsaf_dev->reset_offset; - if (val == 0) { + if (!dereset) { dsaf_write_sub(dsaf_dev, DSAF_SUB_SC_GE_RESET_REQ1_REG, reg_val_1); @@ -216,14 +218,15 @@ void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val) } } -void hns_ppe_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val) +static void hns_ppe_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, + bool dereset) { u32 reg_val = 0; u32 reg_addr; reg_val |= RESET_REQ_OR_DREQ << dsaf_dev->mac_cb[port]->port_rst_off; - if (val == 0) + if (!dereset) reg_addr = DSAF_SUB_SC_PPE_RESET_REQ_REG; else reg_addr = DSAF_SUB_SC_PPE_RESET_DREQ_REG; @@ -231,15 +234,14 @@ void hns_ppe_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, u32 val) dsaf_write_sub(dsaf_dev, reg_addr, reg_val); } -void hns_ppe_com_srst(struct ppe_common_cb *ppe_common, u32 val) +static void hns_ppe_com_srst(struct dsaf_device *dsaf_dev, bool dereset) { - struct dsaf_device *dsaf_dev = ppe_common->dsaf_dev; u32 reg_val; u32 reg_addr; if (!HNS_DSAF_IS_DEBUG(dsaf_dev)) { reg_val = RESET_REQ_OR_DREQ; - if (val == 0) + if (!dereset) reg_addr = DSAF_SUB_SC_RCB_PPE_COM_RESET_REQ_REG; else reg_addr = DSAF_SUB_SC_RCB_PPE_COM_RESET_DREQ_REG; @@ -247,7 +249,7 @@ void hns_ppe_com_srst(struct ppe_common_cb *ppe_common, u32 val) } else { reg_val = 0x100 << dsaf_dev->reset_offset; - if (val == 0) + if (!dereset) reg_addr = DSAF_SUB_SC_PPE_RESET_REQ_REG; else reg_addr = DSAF_SUB_SC_PPE_RESET_DREQ_REG; @@ -261,7 +263,7 @@ void hns_ppe_com_srst(struct ppe_common_cb *ppe_common, u32 val) * @mac_cb: mac control block * retuen phy interface */ -phy_interface_t hns_mac_get_phy_if(struct hns_mac_cb *mac_cb) +static phy_interface_t hns_mac_get_phy_if(struct hns_mac_cb *mac_cb) { u32 mode; u32 reg; @@ -309,7 +311,7 @@ int hns_mac_get_sfp_prsnt(struct hns_mac_cb *mac_cb, int *sfp_prsnt) * @mac_cb: mac control block * retuen 0 == success */ -int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, u8 en) +static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en) { /* port 0-3 hilink4 base is serdes_vaddr + 0x00280000 * port 4-7 hilink3 base is serdes_vaddr + 0x00200000 @@ -343,11 +345,38 @@ int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, u8 en) if (mac_cb->serdes_ctrl) { u32 origin = dsaf_read_syscon(mac_cb->serdes_ctrl, reg_offset); - dsaf_set_field(origin, 1ull << 10, 10, !!en); + dsaf_set_field(origin, 1ull << 10, 10, en); dsaf_write_syscon(mac_cb->serdes_ctrl, reg_offset, origin); } else { - dsaf_set_reg_field(base_addr, reg_offset, 1ull << 10, 10, !!en); + dsaf_set_reg_field(base_addr, reg_offset, 1ull << 10, 10, en); } return 0; } + +struct dsaf_misc_op *hns_misc_op_get(struct dsaf_device *dsaf_dev) +{ + struct dsaf_misc_op *misc_op; + + misc_op = devm_kzalloc(dsaf_dev->dev, sizeof(*misc_op), GFP_KERNEL); + if (!misc_op) + return NULL; + + misc_op->cpld_set_led = hns_cpld_set_led; + misc_op->cpld_reset_led = cpld_led_reset; + misc_op->cpld_set_led_id = cpld_set_led_id; + + misc_op->dsaf_reset = hns_dsaf_rst; + misc_op->xge_srst = hns_dsaf_xge_srst_by_port; + misc_op->xge_core_srst = hns_dsaf_xge_core_srst_by_port; + misc_op->ge_srst = hns_dsaf_ge_srst_by_port; + misc_op->ppe_srst = hns_ppe_srst_by_port; + misc_op->ppe_comm_srst = hns_ppe_com_srst; + + misc_op->get_phy_if = hns_mac_get_phy_if; + misc_op->get_sfp_prsnt = hns_mac_get_sfp_prsnt; + + misc_op->cfg_serdes_loopback = hns_mac_config_sds_loopback; + + return (void *)misc_op; +} diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.h index 419f07a..f06bb03 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.h @@ -33,11 +33,6 @@ #define DSAF_LED_DATA_B 4 #define DSAF_LED_ANCHOR_B 5 -void hns_cpld_set_led(struct hns_mac_cb *mac_cb, int link_status, - u16 speed, int data); -void cpld_led_reset(struct hns_mac_cb *mac_cb); -int cpld_set_led_id(struct hns_mac_cb *mac_cb, - enum hnae_led_state status); -int hns_mac_get_sfp_prsnt(struct hns_mac_cb *mac_cb, int *sfp_prsnt); +struct dsaf_misc_op *hns_misc_op_get(struct dsaf_device *dsaf_dev); #endif diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c index 8cd151a..ff8b6a4 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_ppe.c @@ -112,7 +112,6 @@ void hns_ppe_common_free_cfg(struct dsaf_device *dsaf_dev, u32 comm_index) static void __iomem *hns_ppe_get_iobase(struct ppe_common_cb *ppe_common, int ppe_idx) { - return ppe_common->dsaf_dev->ppe_base + ppe_idx * PPE_REG_OFFSET; } @@ -200,11 +199,12 @@ static void hns_ppe_set_port_mode(struct hns_ppe_cb *ppe_cb, static int hns_ppe_common_init_hw(struct ppe_common_cb *ppe_common) { enum ppe_qid_mode qid_mode; - enum dsaf_mode dsaf_mode = ppe_common->dsaf_dev->dsaf_mode; + struct dsaf_device *dsaf_dev = ppe_common->dsaf_dev; + enum dsaf_mode dsaf_mode = dsaf_dev->dsaf_mode; - hns_ppe_com_srst(ppe_common, 0); + dsaf_dev->misc_op->ppe_comm_srst(dsaf_dev, 0); mdelay(100); - hns_ppe_com_srst(ppe_common, 1); + dsaf_dev->misc_op->ppe_comm_srst(dsaf_dev, 1); mdelay(100); if (ppe_common->ppe_mode == PPE_COMMON_MODE_SERVICE) { @@ -288,9 +288,9 @@ static void hns_ppe_init_hw(struct hns_ppe_cb *ppe_cb) /* get default RSS key */ netdev_rss_key_fill(ppe_cb->rss_key, HNS_PPEV2_RSS_KEY_SIZE); - hns_ppe_srst_by_port(dsaf_dev, port, 0); + dsaf_dev->misc_op->ppe_srst(dsaf_dev, port, 0); mdelay(10); - hns_ppe_srst_by_port(dsaf_dev, port, 1); + dsaf_dev->misc_op->ppe_srst(dsaf_dev, port, 1); /* clr and msk except irq*/ hns_ppe_exc_irq_en(ppe_cb, 0); @@ -328,10 +328,11 @@ static void hns_ppe_init_hw(struct hns_ppe_cb *ppe_cb) static void hns_ppe_uninit_hw(struct hns_ppe_cb *ppe_cb) { u32 port; + struct dsaf_device *dsaf_dev = ppe_cb->ppe_common_cb->dsaf_dev; if (ppe_cb->ppe_common_cb) { port = ppe_cb->index; - hns_ppe_srst_by_port(ppe_cb->ppe_common_cb->dsaf_dev, port, 0); + dsaf_dev->misc_op->ppe_srst(dsaf_dev, port, 0); } } diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c index fd90f37..8f4f0e8 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_xgmac.c @@ -119,7 +119,7 @@ static void hns_xgmac_enable(void *mac_drv, enum mac_commom_mode mode) = (struct dsaf_device *)dev_get_drvdata(drv->dev); u32 port = drv->mac_id; - hns_dsaf_xge_core_srst_by_port(dsaf_dev, port, 1); + dsaf_dev->misc_op->xge_core_srst(dsaf_dev, port, 1); mdelay(10); /*enable XGE rX/tX */ @@ -157,7 +157,7 @@ static void hns_xgmac_disable(void *mac_drv, enum mac_commom_mode mode) } mdelay(10); - hns_dsaf_xge_core_srst_by_port(dsaf_dev, port, 0); + dsaf_dev->misc_op->xge_core_srst(dsaf_dev, port, 0); } /** @@ -198,9 +198,9 @@ static void hns_xgmac_init(void *mac_drv) = (struct dsaf_device *)dev_get_drvdata(drv->dev); u32 port = drv->mac_id; - hns_dsaf_xge_srst_by_port(dsaf_dev, port, 0); + dsaf_dev->misc_op->xge_srst(dsaf_dev, port, 0); mdelay(100); - hns_dsaf_xge_srst_by_port(dsaf_dev, port, 1); + dsaf_dev->misc_op->xge_srst(dsaf_dev, port, 1); mdelay(100); hns_xgmac_exc_irq_en(drv, 0); @@ -425,7 +425,7 @@ static void hns_xgmac_free(void *mac_drv) u32 mac_id = drv->mac_id; - hns_dsaf_xge_srst_by_port(dsaf_dev, mac_id, 0); + dsaf_dev->misc_op->xge_srst(dsaf_dev, mac_id, 0); } /** -- cgit v0.10.2 From 8413b3be4d77dae0d0203cc8b6f2b928dda66e7e Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:18 +0800 Subject: net: hns: dsaf adds support of acpi Dsaf needs to get configuration parameter by ACPI, so this patch add support of ACPI. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c index 2ebf14a..3ef0c9b 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c @@ -689,9 +689,7 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) return 0; } - if (!is_of_node(mac_cb->fw_port)) - return -EINVAL; - + if (is_of_node(mac_cb->fw_port)) { /* parse property from port subnode in dsaf */ np = of_parse_phandle(to_of_node(mac_cb->fw_port), "phy-handle", 0); mac_cb->phy_dev = of_phy_find_device(np); @@ -701,47 +699,49 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) mac_cb->mac_id, np->name); } - syscon = syscon_node_to_regmap( - of_parse_phandle(to_of_node(mac_cb->fw_port), - "serdes-syscon", 0)); - if (IS_ERR_OR_NULL(syscon)) { - dev_err(mac_cb->dev, "serdes-syscon is needed!\n"); - return -EINVAL; - } - mac_cb->serdes_ctrl = syscon; - - ret = fwnode_property_read_u32(mac_cb->fw_port, - "port-rst-offset", - &mac_cb->port_rst_off); - if (ret) { - dev_dbg(mac_cb->dev, - "mac%d port-rst-offset not found, use default value.\n", - mac_cb->mac_id); - } + syscon = syscon_node_to_regmap( + of_parse_phandle(to_of_node(mac_cb->fw_port), + "serdes-syscon", 0)); + if (IS_ERR_OR_NULL(syscon)) { + dev_err(mac_cb->dev, "serdes-syscon is needed!\n"); + return -EINVAL; + } + mac_cb->serdes_ctrl = syscon; - ret = fwnode_property_read_u32(mac_cb->fw_port, - "port-mode-offset", - &mac_cb->port_mode_off); - if (ret) { - dev_dbg(mac_cb->dev, - "mac%d port-mode-offset not found, use default value.\n", - mac_cb->mac_id); - } + ret = fwnode_property_read_u32(mac_cb->fw_port, + "port-rst-offset", + &mac_cb->port_rst_off); + if (ret) { + dev_dbg(mac_cb->dev, + "mac%d port-rst-offset not found, use default value.\n", + mac_cb->mac_id); + } - ret = of_parse_phandle_with_fixed_args(to_of_node(mac_cb->fw_port), - "cpld-syscon", 1, 0, &cpld_args); - if (ret) { - dev_dbg(mac_cb->dev, "mac%d no cpld-syscon found.\n", - mac_cb->mac_id); - mac_cb->cpld_ctrl = NULL; - } else { - syscon = syscon_node_to_regmap(cpld_args.np); - if (IS_ERR_OR_NULL(syscon)) { - dev_dbg(mac_cb->dev, "no cpld-syscon found!\n"); + ret = fwnode_property_read_u32(mac_cb->fw_port, + "port-mode-offset", + &mac_cb->port_mode_off); + if (ret) { + dev_dbg(mac_cb->dev, + "mac%d port-mode-offset not found, use default value.\n", + mac_cb->mac_id); + } + + ret = of_parse_phandle_with_fixed_args( + to_of_node(mac_cb->fw_port), "cpld-syscon", 1, 0, + &cpld_args); + if (ret) { + dev_dbg(mac_cb->dev, "mac%d no cpld-syscon found.\n", + mac_cb->mac_id); mac_cb->cpld_ctrl = NULL; } else { - mac_cb->cpld_ctrl = syscon; - mac_cb->cpld_ctrl_reg = cpld_args.args[0]; + syscon = syscon_node_to_regmap(cpld_args.np); + if (IS_ERR_OR_NULL(syscon)) { + dev_dbg(mac_cb->dev, "no cpld-syscon found!\n"); + mac_cb->cpld_ctrl = NULL; + } else { + mac_cb->cpld_ctrl = syscon; + mac_cb->cpld_ctrl_reg = cpld_args.args[0]; + } } } diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index 28bde08..ac03c4a 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -7,6 +7,7 @@ * (at your option) any later version. */ +#include #include #include #include @@ -33,6 +34,13 @@ const char *g_dsaf_mode_match[DSAF_MODE_MAX] = { [DSAF_MODE_DISABLE_SP] = "single-port", }; +static const struct acpi_device_id hns_dsaf_acpi_match[] = { + { "HISI00B1", 0 }, + { "HISI00B2", 0 }, + { }, +}; +MODULE_DEVICE_TABLE(acpi, hns_dsaf_acpi_match); + int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) { int ret, i; @@ -46,10 +54,22 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) struct device_node *np = dsaf_dev->dev->of_node; struct platform_device *pdev = to_platform_device(dsaf_dev->dev); - if (of_device_is_compatible(np, "hisilicon,hns-dsaf-v1")) - dsaf_dev->dsaf_ver = AE_VERSION_1; - else - dsaf_dev->dsaf_ver = AE_VERSION_2; + if (dev_of_node(dsaf_dev->dev)) { + if (of_device_is_compatible(np, "hisilicon,hns-dsaf-v1")) + dsaf_dev->dsaf_ver = AE_VERSION_1; + else + dsaf_dev->dsaf_ver = AE_VERSION_2; + } else if (is_acpi_node(dsaf_dev->dev->fwnode)) { + if (acpi_dev_found(hns_dsaf_acpi_match[0].id)) + dsaf_dev->dsaf_ver = AE_VERSION_1; + else if (acpi_dev_found(hns_dsaf_acpi_match[1].id)) + dsaf_dev->dsaf_ver = AE_VERSION_2; + else + return -ENXIO; + } else { + dev_err(dsaf_dev->dev, "cannot get cfg data from of or acpi\n"); + return -ENXIO; + } ret = device_property_read_string(dsaf_dev->dev, "mode", &mode_str); if (ret) { @@ -81,32 +101,40 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) else dsaf_dev->dsaf_tc_mode = HRD_DSAF_4TC_MODE; - syscon = syscon_node_to_regmap( - of_parse_phandle(np, "subctrl-syscon", 0)); - if (IS_ERR_OR_NULL(syscon)) { - res = platform_get_resource(pdev, IORESOURCE_MEM, res_idx++); - if (!res) { - dev_err(dsaf_dev->dev, "subctrl info is needed!\n"); - return -ENOMEM; - } - dsaf_dev->sc_base = devm_ioremap_resource(&pdev->dev, res); - if (!dsaf_dev->sc_base) { - dev_err(dsaf_dev->dev, "subctrl can not map!\n"); - return -ENOMEM; - } + if (dev_of_node(dsaf_dev->dev)) { + syscon = syscon_node_to_regmap( + of_parse_phandle(np, "subctrl-syscon", 0)); + if (IS_ERR_OR_NULL(syscon)) { + res = platform_get_resource(pdev, IORESOURCE_MEM, + res_idx++); + if (!res) { + dev_err(dsaf_dev->dev, "subctrl info is needed!\n"); + return -ENOMEM; + } - res = platform_get_resource(pdev, IORESOURCE_MEM, res_idx++); - if (!res) { - dev_err(dsaf_dev->dev, "serdes-ctrl info is needed!\n"); - return -ENOMEM; - } - dsaf_dev->sds_base = devm_ioremap_resource(&pdev->dev, res); - if (!dsaf_dev->sds_base) { - dev_err(dsaf_dev->dev, "serdes-ctrl can not map!\n"); - return -ENOMEM; + dsaf_dev->sc_base = devm_ioremap_resource(&pdev->dev, + res); + if (!dsaf_dev->sc_base) { + dev_err(dsaf_dev->dev, "subctrl can not map!\n"); + return -ENOMEM; + } + + res = platform_get_resource(pdev, IORESOURCE_MEM, + res_idx++); + if (!res) { + dev_err(dsaf_dev->dev, "serdes-ctrl info is needed!\n"); + return -ENOMEM; + } + + dsaf_dev->sds_base = devm_ioremap_resource(&pdev->dev, + res); + if (!dsaf_dev->sds_base) { + dev_err(dsaf_dev->dev, "serdes-ctrl can not map!\n"); + return -ENOMEM; + } + } else { + dsaf_dev->sub_ctrl = syscon; } - } else { - dsaf_dev->sub_ctrl = syscon; } res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "ppe-base"); @@ -2686,6 +2714,7 @@ static struct platform_driver g_dsaf_driver = { .driver = { .name = DSAF_DRV_NAME, .of_match_table = g_dsaf_match, + .acpi_match_table = hns_dsaf_acpi_match, }, }; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c index e1c3e0d..f21177b 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c @@ -362,21 +362,23 @@ struct dsaf_misc_op *hns_misc_op_get(struct dsaf_device *dsaf_dev) if (!misc_op) return NULL; - misc_op->cpld_set_led = hns_cpld_set_led; - misc_op->cpld_reset_led = cpld_led_reset; - misc_op->cpld_set_led_id = cpld_set_led_id; - - misc_op->dsaf_reset = hns_dsaf_rst; - misc_op->xge_srst = hns_dsaf_xge_srst_by_port; - misc_op->xge_core_srst = hns_dsaf_xge_core_srst_by_port; - misc_op->ge_srst = hns_dsaf_ge_srst_by_port; - misc_op->ppe_srst = hns_ppe_srst_by_port; - misc_op->ppe_comm_srst = hns_ppe_com_srst; - - misc_op->get_phy_if = hns_mac_get_phy_if; - misc_op->get_sfp_prsnt = hns_mac_get_sfp_prsnt; - - misc_op->cfg_serdes_loopback = hns_mac_config_sds_loopback; + if (dev_of_node(dsaf_dev->dev)) { + misc_op->cpld_set_led = hns_cpld_set_led; + misc_op->cpld_reset_led = cpld_led_reset; + misc_op->cpld_set_led_id = cpld_set_led_id; + + misc_op->dsaf_reset = hns_dsaf_rst; + misc_op->xge_srst = hns_dsaf_xge_srst_by_port; + misc_op->xge_core_srst = hns_dsaf_xge_core_srst_by_port; + misc_op->ge_srst = hns_dsaf_ge_srst_by_port; + misc_op->ppe_srst = hns_ppe_srst_by_port; + misc_op->ppe_comm_srst = hns_ppe_com_srst; + + misc_op->get_phy_if = hns_mac_get_phy_if; + misc_op->get_sfp_prsnt = hns_mac_get_sfp_prsnt; + + misc_op->cfg_serdes_loopback = hns_mac_config_sds_loopback; + } return (void *)misc_op; } -- cgit v0.10.2 From 1d1afa2ebf8234cbeafd719cf964d7d4f4d6ac96 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:19 +0800 Subject: net: hns: register phy device in each mac initial sequence In ACPI case, there is no interface to register phy device to mdio-bus. Phy device has to be registered itself to mdio-bus, and then enet can get the phy device's info so that it can config the phy-device to help to trasmit and receive data. HNS hardware topology is as below. The MDIO controller may control several PHY-devices, and each PHY-device connects to a MAC device. PHY-devices will register when each mac find PHY device in initial sequence. cpu | | ------------------------------------------- | | | | | | | dsaf | MDIO | MDIO | --------------------------- | | | | | | | | | | | | | | MAC MAC MAC MAC | | | | | | | ---- |-------- |-------- | | -------- || || || || PHY PHY PHY PHY Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c index 3ef0c9b..c526558 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c @@ -7,6 +7,7 @@ * (at your option) any later version. */ +#include #include #include #include @@ -638,6 +639,115 @@ free_mac_drv: return ret; } +static int +hns_mac_phy_parse_addr(struct device *dev, struct fwnode_handle *fwnode) +{ + u32 addr; + int ret; + + ret = fwnode_property_read_u32(fwnode, "phy-addr", &addr); + if (ret) { + dev_err(dev, "has invalid PHY address ret:%d\n", ret); + return ret; + } + + if (addr >= PHY_MAX_ADDR) { + dev_err(dev, "PHY address %i is too large\n", addr); + return -EINVAL; + } + + return addr; +} + +static int hns_mac_phydev_match(struct device *dev, void *fwnode) +{ + return dev->fwnode == fwnode; +} + +static struct +platform_device *hns_mac_find_platform_device(struct fwnode_handle *fwnode) +{ + struct device *dev; + + dev = bus_find_device(&platform_bus_type, NULL, + fwnode, hns_mac_phydev_match); + return dev ? to_platform_device(dev) : NULL; +} + +static int +hns_mac_register_phydev(struct mii_bus *mdio, struct hns_mac_cb *mac_cb, + u32 addr) +{ + struct phy_device *phy; + const char *phy_type; + bool is_c45; + int rc; + + rc = fwnode_property_read_string(mac_cb->fw_port, + "phy-mode", &phy_type); + if (rc < 0) + return rc; + + if (!strcmp(phy_type, phy_modes(PHY_INTERFACE_MODE_XGMII))) + is_c45 = 1; + else if (!strcmp(phy_type, phy_modes(PHY_INTERFACE_MODE_SGMII))) + is_c45 = 0; + else + return -ENODATA; + + phy = get_phy_device(mdio, addr, is_c45); + if (!phy || IS_ERR(phy)) + return -EIO; + + if (mdio->irq) + phy->irq = mdio->irq[addr]; + + /* All data is now stored in the phy struct; + * register it + */ + rc = phy_device_register(phy); + if (rc) { + phy_device_free(phy); + return -ENODEV; + } + + mac_cb->phy_dev = phy; + + dev_dbg(&mdio->dev, "registered phy at address %i\n", addr); + + return 0; +} + +static void hns_mac_register_phy(struct hns_mac_cb *mac_cb) +{ + struct acpi_reference_args args; + struct platform_device *pdev; + struct mii_bus *mii_bus; + int rc; + int addr; + + /* Loop over the child nodes and register a phy_device for each one */ + if (!to_acpi_device_node(mac_cb->fw_port)) + return; + + rc = acpi_node_get_property_reference( + mac_cb->fw_port, "mdio-node", 0, &args); + if (rc) + return; + + addr = hns_mac_phy_parse_addr(mac_cb->dev, mac_cb->fw_port); + if (addr < 0) + return; + + /* dev address in adev */ + pdev = hns_mac_find_platform_device(acpi_fwnode_handle(args.adev)); + mii_bus = platform_get_drvdata(pdev); + rc = hns_mac_register_phydev(mii_bus, mac_cb, addr); + if (!rc) + dev_dbg(mac_cb->dev, "mac%d register phy addr:%d\n", + mac_cb->mac_id, addr); +} + /** *hns_mac_get_info - get mac information from device node *@mac_cb: mac device @@ -690,13 +800,17 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) } if (is_of_node(mac_cb->fw_port)) { - /* parse property from port subnode in dsaf */ - np = of_parse_phandle(to_of_node(mac_cb->fw_port), "phy-handle", 0); - mac_cb->phy_dev = of_phy_find_device(np); - if (mac_cb->phy_dev) { - put_device(&mac_cb->phy_dev->mdio.dev); - dev_dbg(mac_cb->dev, "mac%d phy_node: %s\n", - mac_cb->mac_id, np->name); + /* parse property from port subnode in dsaf */ + np = of_parse_phandle(to_of_node(mac_cb->fw_port), + "phy-handle", 0); + mac_cb->phy_dev = of_phy_find_device(np); + if (mac_cb->phy_dev) { + /* refcount is held by of_phy_find_device() + * if the phy_dev is found + */ + put_device(&mac_cb->phy_dev->mdio.dev); + dev_dbg(mac_cb->dev, "mac%d phy_node: %s\n", + mac_cb->mac_id, np->name); } syscon = syscon_node_to_regmap( @@ -743,6 +857,11 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) mac_cb->cpld_ctrl_reg = cpld_args.args[0]; } } + } else if (is_acpi_node(mac_cb->fw_port)) { + hns_mac_register_phy(mac_cb); + } else { + dev_err(mac_cb->dev, "mac%d cannot find phy node\n", + mac_cb->mac_id); } return 0; -- cgit v0.10.2 From f00ef863da2b59313c466d2c826d2e8c28157d36 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:20 +0800 Subject: net: hns: implement the miscellaneous operation by asl The miscellaneous operation is implemented in BIOS, the kernel can call _DSM method help to call the implementation in ACPI case. Here is a patch to do that. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c index f21177b..96cb628 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c @@ -12,6 +12,27 @@ #include "hns_dsaf_ppe.h" #include "hns_dsaf_reg.h" +enum _dsm_op_index { + HNS_OP_RESET_FUNC = 0x1, + HNS_OP_SERDES_LP_FUNC = 0x2, + HNS_OP_LED_SET_FUNC = 0x3, + HNS_OP_GET_PORT_TYPE_FUNC = 0x4, + HNS_OP_GET_SFP_STAT_FUNC = 0x5, +}; + +enum _dsm_rst_type { + HNS_DSAF_RESET_FUNC = 0x1, + HNS_PPE_RESET_FUNC = 0x2, + HNS_XGE_CORE_RESET_FUNC = 0x3, + HNS_XGE_RESET_FUNC = 0x4, + HNS_GE_RESET_FUNC = 0x5, +}; + +const u8 hns_dsaf_acpi_dsm_uuid[] = { + 0x1A, 0xAA, 0x85, 0x1A, 0x93, 0xE2, 0x5E, 0x41, + 0x8E, 0x28, 0x8D, 0x69, 0x0A, 0x0F, 0x82, 0x0A +}; + static void dsaf_write_sub(struct dsaf_device *dsaf_dev, u32 reg, u32 val) { if (dsaf_dev->sub_ctrl) @@ -109,6 +130,34 @@ static int cpld_set_led_id(struct hns_mac_cb *mac_cb, #define RESET_REQ_OR_DREQ 1 +static void hns_dsaf_acpi_srst_by_port(struct dsaf_device *dsaf_dev, u8 op_type, + u32 port_type, u32 port, u32 val) +{ + union acpi_object *obj; + union acpi_object obj_args[3], argv4; + + obj_args[0].integer.type = ACPI_TYPE_INTEGER; + obj_args[0].integer.value = port_type; + obj_args[1].integer.type = ACPI_TYPE_INTEGER; + obj_args[1].integer.value = port; + obj_args[2].integer.type = ACPI_TYPE_INTEGER; + obj_args[2].integer.value = val; + + argv4.type = ACPI_TYPE_PACKAGE; + argv4.package.count = 3; + argv4.package.elements = obj_args; + + obj = acpi_evaluate_dsm(ACPI_HANDLE(dsaf_dev->dev), + hns_dsaf_acpi_dsm_uuid, 0, op_type, &argv4); + if (!obj) { + dev_warn(dsaf_dev->dev, "reset port_type%d port%d fail!", + port_type, port); + return; + } + + ACPI_FREE(obj); +} + static void hns_dsaf_rst(struct dsaf_device *dsaf_dev, bool dereset) { u32 xbar_reg_addr; @@ -126,6 +175,13 @@ static void hns_dsaf_rst(struct dsaf_device *dsaf_dev, bool dereset) dsaf_write_sub(dsaf_dev, nt_reg_addr, RESET_REQ_OR_DREQ); } +static void hns_dsaf_rst_acpi(struct dsaf_device *dsaf_dev, bool dereset) +{ + hns_dsaf_acpi_srst_by_port(dsaf_dev, HNS_OP_RESET_FUNC, + HNS_DSAF_RESET_FUNC, + 0, dereset); +} + static void hns_dsaf_xge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, bool dereset) { @@ -146,6 +202,13 @@ static void hns_dsaf_xge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, dsaf_write_sub(dsaf_dev, reg_addr, reg_val); } +static void hns_dsaf_xge_srst_by_port_acpi(struct dsaf_device *dsaf_dev, + u32 port, bool dereset) +{ + hns_dsaf_acpi_srst_by_port(dsaf_dev, HNS_OP_RESET_FUNC, + HNS_XGE_RESET_FUNC, port, dereset); +} + static void hns_dsaf_xge_core_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, bool dereset) { @@ -166,6 +229,14 @@ static void hns_dsaf_xge_core_srst_by_port(struct dsaf_device *dsaf_dev, dsaf_write_sub(dsaf_dev, reg_addr, reg_val); } +static void +hns_dsaf_xge_core_srst_by_port_acpi(struct dsaf_device *dsaf_dev, + u32 port, bool dereset) +{ + hns_dsaf_acpi_srst_by_port(dsaf_dev, HNS_OP_RESET_FUNC, + HNS_XGE_CORE_RESET_FUNC, port, dereset); +} + static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, bool dereset) { @@ -218,6 +289,13 @@ static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, } } +static void hns_dsaf_ge_srst_by_port_acpi(struct dsaf_device *dsaf_dev, + u32 port, bool dereset) +{ + hns_dsaf_acpi_srst_by_port(dsaf_dev, HNS_OP_RESET_FUNC, + HNS_GE_RESET_FUNC, port, dereset); +} + static void hns_ppe_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, bool dereset) { @@ -234,11 +312,21 @@ static void hns_ppe_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, dsaf_write_sub(dsaf_dev, reg_addr, reg_val); } +static void +hns_ppe_srst_by_port_acpi(struct dsaf_device *dsaf_dev, u32 port, bool dereset) +{ + hns_dsaf_acpi_srst_by_port(dsaf_dev, HNS_OP_RESET_FUNC, + HNS_PPE_RESET_FUNC, port, dereset); +} + static void hns_ppe_com_srst(struct dsaf_device *dsaf_dev, bool dereset) { u32 reg_val; u32 reg_addr; + if (!(dev_of_node(dsaf_dev->dev))) + return; + if (!HNS_DSAF_IS_DEBUG(dsaf_dev)) { reg_val = RESET_REQ_OR_DREQ; if (!dereset) @@ -295,6 +383,36 @@ static phy_interface_t hns_mac_get_phy_if(struct hns_mac_cb *mac_cb) return phy_if; } +static phy_interface_t hns_mac_get_phy_if_acpi(struct hns_mac_cb *mac_cb) +{ + phy_interface_t phy_if = PHY_INTERFACE_MODE_NA; + union acpi_object *obj; + union acpi_object obj_args, argv4; + + obj_args.integer.type = ACPI_TYPE_INTEGER; + obj_args.integer.value = mac_cb->mac_id; + + argv4.type = ACPI_TYPE_PACKAGE, + argv4.package.count = 1, + argv4.package.elements = &obj_args, + + obj = acpi_evaluate_dsm(ACPI_HANDLE(mac_cb->dev), + hns_dsaf_acpi_dsm_uuid, 0, + HNS_OP_GET_PORT_TYPE_FUNC, &argv4); + + if (!obj || obj->type != ACPI_TYPE_INTEGER) + return phy_if; + + phy_if = obj->integer.value ? + PHY_INTERFACE_MODE_XGMII : PHY_INTERFACE_MODE_SGMII; + + dev_dbg(mac_cb->dev, "mac_id=%d, phy_if=%d\n", mac_cb->mac_id, phy_if); + + ACPI_FREE(obj); + + return phy_if; +} + int hns_mac_get_sfp_prsnt(struct hns_mac_cb *mac_cb, int *sfp_prsnt) { if (!mac_cb->cpld_ctrl) @@ -354,6 +472,36 @@ static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en) return 0; } +static int +hns_mac_config_sds_loopback_acpi(struct hns_mac_cb *mac_cb, bool en) +{ + union acpi_object *obj; + union acpi_object obj_args[3], argv4; + + obj_args[0].integer.type = ACPI_TYPE_INTEGER; + obj_args[0].integer.value = mac_cb->mac_id; + obj_args[1].integer.type = ACPI_TYPE_INTEGER; + obj_args[1].integer.value = !!en; + + argv4.type = ACPI_TYPE_PACKAGE; + argv4.package.count = 2; + argv4.package.elements = obj_args; + + obj = acpi_evaluate_dsm(ACPI_HANDLE(mac_cb->dsaf_dev->dev), + hns_dsaf_acpi_dsm_uuid, 0, + HNS_OP_SERDES_LP_FUNC, &argv4); + if (!obj) { + dev_warn(mac_cb->dsaf_dev->dev, "set port%d serdes lp fail!", + mac_cb->mac_id); + + return -ENOTSUPP; + } + + ACPI_FREE(obj); + + return 0; +} + struct dsaf_misc_op *hns_misc_op_get(struct dsaf_device *dsaf_dev) { struct dsaf_misc_op *misc_op; @@ -378,6 +526,25 @@ struct dsaf_misc_op *hns_misc_op_get(struct dsaf_device *dsaf_dev) misc_op->get_sfp_prsnt = hns_mac_get_sfp_prsnt; misc_op->cfg_serdes_loopback = hns_mac_config_sds_loopback; + } else if (is_acpi_node(dsaf_dev->dev->fwnode)) { + misc_op->cpld_set_led = hns_cpld_set_led; + misc_op->cpld_reset_led = cpld_led_reset; + misc_op->cpld_set_led_id = cpld_set_led_id; + + misc_op->dsaf_reset = hns_dsaf_rst_acpi; + misc_op->xge_srst = hns_dsaf_xge_srst_by_port_acpi; + misc_op->xge_core_srst = hns_dsaf_xge_core_srst_by_port_acpi; + misc_op->ge_srst = hns_dsaf_ge_srst_by_port_acpi; + misc_op->ppe_srst = hns_ppe_srst_by_port_acpi; + misc_op->ppe_comm_srst = hns_ppe_com_srst; + + misc_op->get_phy_if = hns_mac_get_phy_if_acpi; + misc_op->get_sfp_prsnt = hns_mac_get_sfp_prsnt; + + misc_op->cfg_serdes_loopback = hns_mac_config_sds_loopback_acpi; + } else { + devm_kfree(dsaf_dev->dev, (void *)misc_op); + misc_op = NULL; } return (void *)misc_op; -- cgit v0.10.2 From 63434888aaf1b2e73780f3ada9dcf09bf0f206e1 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:21 +0800 Subject: net: hns: net: hns: enet adds support of acpi Enet needs to get configration parameter by acpi. This patch adds support of ACPI for enet. The configuration parameter will be configed in BIOS. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index 3ec3c27..ad742a6 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -132,6 +132,13 @@ static void fill_v2_desc(struct hnae_ring *ring, void *priv, ring_ptr_move_fw(ring, next_to_use); } +static const struct acpi_device_id hns_enet_acpi_match[] = { + { "HISI00C1", 0 }, + { "HISI00C2", 0 }, + { }, +}; +MODULE_DEVICE_TABLE(acpi, hns_enet_acpi_match); + static void fill_desc(struct hnae_ring *ring, void *priv, int size, dma_addr_t dma, int frag_end, int buf_num, enum hns_desc_type type, int mtu) @@ -1870,7 +1877,6 @@ static int hns_nic_dev_probe(struct platform_device *pdev) struct device *dev = &pdev->dev; struct net_device *ndev; struct hns_nic_priv *priv; - struct device_node *ae_node; u32 port_id; int ret; @@ -1884,20 +1890,45 @@ static int hns_nic_dev_probe(struct platform_device *pdev) priv->dev = dev; priv->netdev = ndev; - if (of_device_is_compatible(dev->of_node, "hisilicon,hns-nic-v1")) - priv->enet_ver = AE_VERSION_1; - else - priv->enet_ver = AE_VERSION_2; + if (dev_of_node(dev)) { + struct device_node *ae_node; - ae_node = of_parse_phandle(dev->of_node, "ae-handle", 0); - if (IS_ERR_OR_NULL(ae_node)) { - ret = PTR_ERR(ae_node); - dev_err(dev, "not find ae-handle\n"); - goto out_read_prop_fail; + if (of_device_is_compatible(dev->of_node, + "hisilicon,hns-nic-v1")) + priv->enet_ver = AE_VERSION_1; + else + priv->enet_ver = AE_VERSION_2; + + ae_node = of_parse_phandle(dev->of_node, "ae-handle", 0); + if (IS_ERR_OR_NULL(ae_node)) { + ret = PTR_ERR(ae_node); + dev_err(dev, "not find ae-handle\n"); + goto out_read_prop_fail; + } + priv->fwnode = &ae_node->fwnode; + } else if (is_acpi_node(dev->fwnode)) { + struct acpi_reference_args args; + + if (acpi_dev_found(hns_enet_acpi_match[0].id)) + priv->enet_ver = AE_VERSION_1; + else if (acpi_dev_found(hns_enet_acpi_match[1].id)) + priv->enet_ver = AE_VERSION_2; + else + return -ENXIO; + + /* try to find port-idx-in-ae first */ + ret = acpi_node_get_property_reference(dev->fwnode, + "ae-handle", 0, &args); + if (ret) { + dev_err(dev, "not find ae-handle\n"); + goto out_read_prop_fail; + } + priv->fwnode = acpi_fwnode_handle(args.adev); + } else { + dev_err(dev, "cannot read cfg data from OF or acpi\n"); + return -ENXIO; } - priv->fwnode = &ae_node->fwnode; - /* try to find port-idx-in-ae first */ ret = device_property_read_u32(dev, "port-idx-in-ae", &port_id); if (ret) { /* only for old code compatible */ @@ -2014,6 +2045,7 @@ static struct platform_driver hns_nic_dev_driver = { .driver = { .name = "hns-nic", .of_match_table = hns_enet_of_match, + .acpi_match_table = ACPI_PTR(hns_enet_acpi_match), }, .probe = hns_nic_dev_probe, .remove = hns_nic_dev_remove, -- cgit v0.10.2 From 30759219f562cfaaebe7b9c1d1c0e6b5445c69b0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michal=20Kube=C4=8Dek?= Date: Fri, 27 May 2016 17:53:52 +0200 Subject: net: disable fragment reassembly if high_thresh is zero Before commit 6d7b857d541e ("net: use lib/percpu_counter API for fragmentation mem accounting"), setting the reassembly high threshold to 0 prevented fragment reassembly as first fragment would be always evicted before second could be added to the queue. While inefficient, some users apparently relied on this method. Since the commit mentioned above, a percpu counter is used for reassembly memory accounting and high batch size avoids taking slow path in most common scenarios. As a result, a whole full sized packet can be reassembled without the percpu counter's main counter changing its value so that even with high_thresh set to 0, fragmented packets can be still reassembled and processed. Add explicit check preventing reassembly if high threshold is zero. Signed-off-by: Michal Kubecek Signed-off-by: David S. Miller diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index 3a88b0c..b5e9317 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -355,7 +355,7 @@ static struct inet_frag_queue *inet_frag_alloc(struct netns_frags *nf, { struct inet_frag_queue *q; - if (frag_mem_limit(nf) > nf->high_thresh) { + if (!nf->high_thresh || frag_mem_limit(nf) > nf->high_thresh) { inet_frag_schedule_worker(f); return NULL; } -- cgit v0.10.2 From 4baa994dc93287321c1b712d98cb6f721652891b Mon Sep 17 00:00:00 2001 From: Vitaly Kuznetsov Date: Fri, 3 Jun 2016 17:50:58 +0200 Subject: hv_netvsc: remove redundant assignment in netvsc_recv_callback() net_device_ctx is assigned in the very beginning of the function and 'net' pointer doesn't change. Signed-off-by: Vitaly Kuznetsov Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 5ac1267..58cb5fe 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -701,7 +701,6 @@ int netvsc_recv_callback(struct hv_device *device_obj, } vf_injection_done: - net_device_ctx = netdev_priv(net); rx_stats = this_cpu_ptr(net_device_ctx->rx_stats); /* Allocate a skb - TODO direct I/O to pages? */ -- cgit v0.10.2 From 2625466d6d92f056da970bee990c862c54188819 Mon Sep 17 00:00:00 2001 From: Vitaly Kuznetsov Date: Fri, 3 Jun 2016 17:50:59 +0200 Subject: hv_netvsc: introduce {net, hv}_device_to_netvsc_device() helpers Make it easier to get 'struct netvsc_device' from 'struct net_device' and 'struct hv_device' by introducing inline helpers. Signed-off-by: Vitaly Kuznetsov Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h index c270c5a..952cbc8 100644 --- a/drivers/net/hyperv/hyperv_net.h +++ b/drivers/net/hyperv/hyperv_net.h @@ -743,6 +743,18 @@ struct netvsc_device { atomic_t vf_use_cnt; }; +static inline struct netvsc_device * +net_device_to_netvsc_device(struct net_device *ndev) +{ + return ((struct net_device_context *)netdev_priv(ndev))->nvdev; +} + +static inline struct netvsc_device * +hv_device_to_netvsc_device(struct hv_device *device) +{ + return net_device_to_netvsc_device(hv_get_drvdata(device)); +} + /* NdisInitialize message */ struct rndis_initialize_request { u32 req_id; diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index 719cb35..96f00c0 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -95,9 +95,7 @@ static void free_netvsc_device(struct netvsc_device *nvdev) static struct netvsc_device *get_outbound_net_device(struct hv_device *device) { - struct net_device *ndev = hv_get_drvdata(device); - struct net_device_context *net_device_ctx = netdev_priv(ndev); - struct netvsc_device *net_device = net_device_ctx->nvdev; + struct netvsc_device *net_device = hv_device_to_netvsc_device(device); if (net_device && net_device->destroy) net_device = NULL; @@ -107,9 +105,7 @@ static struct netvsc_device *get_outbound_net_device(struct hv_device *device) static struct netvsc_device *get_inbound_net_device(struct hv_device *device) { - struct net_device *ndev = hv_get_drvdata(device); - struct net_device_context *net_device_ctx = netdev_priv(ndev); - struct netvsc_device *net_device = net_device_ctx->nvdev; + struct netvsc_device *net_device = hv_device_to_netvsc_device(device); if (!net_device) goto get_in_err; @@ -128,8 +124,7 @@ static int netvsc_destroy_buf(struct hv_device *device) struct nvsp_message *revoke_packet; int ret = 0; struct net_device *ndev = hv_get_drvdata(device); - struct net_device_context *net_device_ctx = netdev_priv(ndev); - struct netvsc_device *net_device = net_device_ctx->nvdev; + struct netvsc_device *net_device = net_device_to_netvsc_device(ndev); /* * If we got a section count, it means we received a diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c index 97c292b..42c652e 100644 --- a/drivers/net/hyperv/rndis_filter.c +++ b/drivers/net/hyperv/rndis_filter.c @@ -546,8 +546,7 @@ static int rndis_filter_query_device_mac(struct rndis_device *dev) int rndis_filter_set_device_mac(struct hv_device *hdev, char *mac) { struct net_device *ndev = hv_get_drvdata(hdev); - struct net_device_context *net_device_ctx = netdev_priv(ndev); - struct netvsc_device *nvdev = net_device_ctx->nvdev; + struct netvsc_device *nvdev = net_device_to_netvsc_device(ndev); struct rndis_device *rdev = nvdev->extension; struct rndis_request *request; struct rndis_set_request *set; @@ -626,8 +625,7 @@ rndis_filter_set_offload_params(struct hv_device *hdev, struct ndis_offload_params *req_offloads) { struct net_device *ndev = hv_get_drvdata(hdev); - struct net_device_context *net_device_ctx = netdev_priv(ndev); - struct netvsc_device *nvdev = net_device_ctx->nvdev; + struct netvsc_device *nvdev = net_device_to_netvsc_device(ndev); struct rndis_device *rdev = nvdev->extension; struct rndis_request *request; struct rndis_set_request *set; @@ -851,8 +849,7 @@ static int rndis_filter_init_device(struct rndis_device *dev) u32 status; int ret; unsigned long t; - struct net_device_context *net_device_ctx = netdev_priv(dev->ndev); - struct netvsc_device *nvdev = net_device_ctx->nvdev; + struct netvsc_device *nvdev = net_device_to_netvsc_device(dev->ndev); request = get_rndis_request(dev, RNDIS_MSG_INIT, RNDIS_MESSAGE_SIZE(struct rndis_initialize_request)); @@ -977,8 +974,7 @@ static void netvsc_sc_open(struct vmbus_channel *new_sc) { struct net_device *ndev = hv_get_drvdata(new_sc->primary_channel->device_obj); - struct net_device_context *net_device_ctx = netdev_priv(ndev); - struct netvsc_device *nvscdev = net_device_ctx->nvdev; + struct netvsc_device *nvscdev = net_device_to_netvsc_device(ndev); u16 chn_index = new_sc->offermsg.offer.sub_channel_index; int ret; unsigned long flags; @@ -1196,9 +1192,7 @@ err_dev_remv: void rndis_filter_device_remove(struct hv_device *dev) { - struct net_device *ndev = hv_get_drvdata(dev); - struct net_device_context *net_device_ctx = netdev_priv(ndev); - struct netvsc_device *net_dev = net_device_ctx->nvdev; + struct netvsc_device *net_dev = hv_device_to_netvsc_device(dev); struct rndis_device *rndis_dev = net_dev->extension; unsigned long t; @@ -1224,9 +1218,7 @@ void rndis_filter_device_remove(struct hv_device *dev) int rndis_filter_open(struct hv_device *dev) { - struct net_device *ndev = hv_get_drvdata(dev); - struct net_device_context *net_device_ctx = netdev_priv(ndev); - struct netvsc_device *net_device = net_device_ctx->nvdev; + struct netvsc_device *net_device = hv_device_to_netvsc_device(dev); if (!net_device) return -EINVAL; @@ -1239,9 +1231,7 @@ int rndis_filter_open(struct hv_device *dev) int rndis_filter_close(struct hv_device *dev) { - struct net_device *ndev = hv_get_drvdata(dev); - struct net_device_context *net_device_ctx = netdev_priv(ndev); - struct netvsc_device *nvdev = net_device_ctx->nvdev; + struct netvsc_device *nvdev = hv_device_to_netvsc_device(dev); if (!nvdev) return -EINVAL; -- cgit v0.10.2 From 2f5fa6c869e8f8c340dd05a2817eecbcea382c35 Mon Sep 17 00:00:00 2001 From: Vitaly Kuznetsov Date: Fri, 3 Jun 2016 17:51:00 +0200 Subject: hv_netvsc: pass struct netvsc_device to rndis_filter_{open, close}() Both rndis_filter_open()/rndis_filter_close() use struct hv_device to reach to struct netvsc_device only and all callers have it already. While on it, rename net_device to nvdev in rndis_filter_open() as net_device is misleading. Signed-off-by: Vitaly Kuznetsov Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h index 952cbc8..f650ec1 100644 --- a/drivers/net/hyperv/hyperv_net.h +++ b/drivers/net/hyperv/hyperv_net.h @@ -173,6 +173,7 @@ struct rndis_device { /* Interface */ struct rndis_message; +struct netvsc_device; int netvsc_device_add(struct hv_device *device, void *additional_info); int netvsc_device_remove(struct hv_device *device); int netvsc_send(struct hv_device *device, @@ -189,8 +190,8 @@ int netvsc_recv_callback(struct hv_device *device_obj, struct vmbus_channel *channel, u16 vlan_tci); void netvsc_channel_cb(void *context); -int rndis_filter_open(struct hv_device *dev); -int rndis_filter_close(struct hv_device *dev); +int rndis_filter_open(struct netvsc_device *nvdev); +int rndis_filter_close(struct netvsc_device *nvdev); int rndis_filter_device_add(struct hv_device *dev, void *additional_info); void rndis_filter_device_remove(struct hv_device *dev); diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 58cb5fe..2492f83 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -98,16 +98,14 @@ static void netvsc_set_multicast_list(struct net_device *net) static int netvsc_open(struct net_device *net) { - struct net_device_context *net_device_ctx = netdev_priv(net); - struct hv_device *device_obj = net_device_ctx->device_ctx; - struct netvsc_device *nvdev = net_device_ctx->nvdev; + struct netvsc_device *nvdev = net_device_to_netvsc_device(net); struct rndis_device *rdev; int ret = 0; netif_carrier_off(net); /* Open up the device */ - ret = rndis_filter_open(device_obj); + ret = rndis_filter_open(nvdev); if (ret != 0) { netdev_err(net, "unable to open device (ret %d).\n", ret); return ret; @@ -125,7 +123,6 @@ static int netvsc_open(struct net_device *net) static int netvsc_close(struct net_device *net) { struct net_device_context *net_device_ctx = netdev_priv(net); - struct hv_device *device_obj = net_device_ctx->device_ctx; struct netvsc_device *nvdev = net_device_ctx->nvdev; int ret; u32 aread, awrite, i, msec = 10, retry = 0, retry_max = 20; @@ -135,7 +132,7 @@ static int netvsc_close(struct net_device *net) /* Make sure netvsc_set_multicast_list doesn't re-enable filter! */ cancel_work_sync(&net_device_ctx->work); - ret = rndis_filter_close(device_obj); + ret = rndis_filter_close(nvdev); if (ret != 0) { netdev_err(net, "unable to close device (ret %d).\n", ret); return ret; @@ -1247,7 +1244,7 @@ static int netvsc_vf_up(struct net_device *vf_netdev) /* * Open the device before switching data path. */ - rndis_filter_open(net_device_ctx->device_ctx); + rndis_filter_open(netvsc_dev); /* * notify the host to switch the data path. @@ -1302,7 +1299,7 @@ static int netvsc_vf_down(struct net_device *vf_netdev) udelay(50); netvsc_switch_datapath(ndev, false); netdev_info(ndev, "Data path switched from VF: %s\n", vf_netdev->name); - rndis_filter_close(net_device_ctx->device_ctx); + rndis_filter_close(netvsc_dev); netif_carrier_on(ndev); /* * Notify peers. diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c index 42c652e..2c2f3b9 100644 --- a/drivers/net/hyperv/rndis_filter.c +++ b/drivers/net/hyperv/rndis_filter.c @@ -1216,23 +1216,19 @@ void rndis_filter_device_remove(struct hv_device *dev) } -int rndis_filter_open(struct hv_device *dev) +int rndis_filter_open(struct netvsc_device *nvdev) { - struct netvsc_device *net_device = hv_device_to_netvsc_device(dev); - - if (!net_device) + if (!nvdev) return -EINVAL; - if (atomic_inc_return(&net_device->open_cnt) != 1) + if (atomic_inc_return(&nvdev->open_cnt) != 1) return 0; - return rndis_filter_open_device(net_device->extension); + return rndis_filter_open_device(nvdev->extension); } -int rndis_filter_close(struct hv_device *dev) +int rndis_filter_close(struct netvsc_device *nvdev) { - struct netvsc_device *nvdev = hv_device_to_netvsc_device(dev); - if (!nvdev) return -EINVAL; -- cgit v0.10.2 From e834da9a40edd3117ef0a9b2a73d845fe6b622a8 Mon Sep 17 00:00:00 2001 From: Vitaly Kuznetsov Date: Fri, 3 Jun 2016 17:51:01 +0200 Subject: hv_netvsc: pass struct net_device to rndis_filter_set_device_mac() We unpack 'struct net_device' in netvsc_set_mac_addr() to get to 'struct hv_device' pointer which we use in rndis_filter_set_device_mac() to get back to 'struct net_device'. Signed-off-by: Vitaly Kuznetsov Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h index f650ec1..467fb8b 100644 --- a/drivers/net/hyperv/hyperv_net.h +++ b/drivers/net/hyperv/hyperv_net.h @@ -201,7 +201,7 @@ int rndis_filter_receive(struct hv_device *dev, struct vmbus_channel *channel); int rndis_filter_set_packet_filter(struct rndis_device *dev, u32 new_filter); -int rndis_filter_set_device_mac(struct hv_device *hdev, char *mac); +int rndis_filter_set_device_mac(struct net_device *ndev, char *mac); void netvsc_switch_datapath(struct net_device *nv_dev, bool vf); diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 2492f83..787a202 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -982,8 +982,6 @@ static struct rtnl_link_stats64 *netvsc_get_stats64(struct net_device *net, static int netvsc_set_mac_addr(struct net_device *ndev, void *p) { - struct net_device_context *ndevctx = netdev_priv(ndev); - struct hv_device *hdev = ndevctx->device_ctx; struct sockaddr *addr = p; char save_adr[ETH_ALEN]; unsigned char save_aatype; @@ -996,7 +994,7 @@ static int netvsc_set_mac_addr(struct net_device *ndev, void *p) if (err != 0) return err; - err = rndis_filter_set_device_mac(hdev, addr->sa_data); + err = rndis_filter_set_device_mac(ndev, addr->sa_data); if (err != 0) { /* roll back to saved MAC */ memcpy(ndev->dev_addr, save_adr, ETH_ALEN); diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c index 2c2f3b9..f1692bc 100644 --- a/drivers/net/hyperv/rndis_filter.c +++ b/drivers/net/hyperv/rndis_filter.c @@ -543,9 +543,8 @@ static int rndis_filter_query_device_mac(struct rndis_device *dev) #define NWADR_STR "NetworkAddress" #define NWADR_STRLEN 14 -int rndis_filter_set_device_mac(struct hv_device *hdev, char *mac) +int rndis_filter_set_device_mac(struct net_device *ndev, char *mac) { - struct net_device *ndev = hv_get_drvdata(hdev); struct netvsc_device *nvdev = net_device_to_netvsc_device(ndev); struct rndis_device *rdev = nvdev->extension; struct rndis_request *request; -- cgit v0.10.2 From 426d95417eebc0698e878708a8a44980e8b7570d Mon Sep 17 00:00:00 2001 From: Vitaly Kuznetsov Date: Fri, 3 Jun 2016 17:51:02 +0200 Subject: hv_netvsc: pass struct net_device to rndis_filter_set_offload_params() The only caller rndis_filter_device_add() has 'struct net_device' pointer already. Signed-off-by: Vitaly Kuznetsov Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c index f1692bc..979fa44 100644 --- a/drivers/net/hyperv/rndis_filter.c +++ b/drivers/net/hyperv/rndis_filter.c @@ -620,10 +620,9 @@ cleanup: } static int -rndis_filter_set_offload_params(struct hv_device *hdev, +rndis_filter_set_offload_params(struct net_device *ndev, struct ndis_offload_params *req_offloads) { - struct net_device *ndev = hv_get_drvdata(hdev); struct netvsc_device *nvdev = net_device_to_netvsc_device(ndev); struct rndis_device *rdev = nvdev->extension; struct rndis_request *request; @@ -1083,7 +1082,7 @@ int rndis_filter_device_add(struct hv_device *dev, offloads.lso_v2_ipv4 = NDIS_OFFLOAD_PARAMETERS_LSOV2_ENABLED; - ret = rndis_filter_set_offload_params(dev, &offloads); + ret = rndis_filter_set_offload_params(net, &offloads); if (ret) goto err_dev_remv; -- cgit v0.10.2 From 09fcf916620a7cfc9f2a74f315168e4770249534 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 2 Jun 2016 13:15:10 -0700 Subject: net: vrf: Minor refactoring for local address patches Move the stripping of the ethernet header from is_ip_tx_frame into the ipv4 and ipv6 outbound functions. If the packet is destined to a local address the header is retained since the packet is sent back to netif_rx. Collapse vrf_send_v4_prep into vrf_process_v4_outbound. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index d356f5d..ec0cb65 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -119,6 +119,9 @@ static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, skb_dst_drop(skb); skb_dst_set(skb, dst); + /* strip the ethernet header added for pass through VRF device */ + __skb_pull(skb, skb_network_offset(skb)); + ret = ip6_local_out(net, skb->sk, skb); if (unlikely(net_xmit_eval(ret))) dev->stats.tx_errors++; @@ -139,29 +142,6 @@ static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, } #endif -static int vrf_send_v4_prep(struct sk_buff *skb, struct flowi4 *fl4, - struct net_device *vrf_dev) -{ - struct rtable *rt; - int err = 1; - - rt = ip_route_output_flow(dev_net(vrf_dev), fl4, NULL); - if (IS_ERR(rt)) - goto out; - - /* TO-DO: what about broadcast ? */ - if (rt->rt_type != RTN_UNICAST && rt->rt_type != RTN_LOCAL) { - ip_rt_put(rt); - goto out; - } - - skb_dst_drop(skb); - skb_dst_set(skb, &rt->dst); - err = 0; -out: - return err; -} - static netdev_tx_t vrf_process_v4_outbound(struct sk_buff *skb, struct net_device *vrf_dev) { @@ -176,10 +156,24 @@ static netdev_tx_t vrf_process_v4_outbound(struct sk_buff *skb, FLOWI_FLAG_SKIP_NH_OIF, .daddr = ip4h->daddr, }; + struct net *net = dev_net(vrf_dev); + struct rtable *rt; - if (vrf_send_v4_prep(skb, &fl4, vrf_dev)) + rt = ip_route_output_flow(net, &fl4, NULL); + if (IS_ERR(rt)) goto err; + if (rt->rt_type != RTN_UNICAST && rt->rt_type != RTN_LOCAL) { + ip_rt_put(rt); + goto err; + } + + skb_dst_drop(skb); + skb_dst_set(skb, &rt->dst); + + /* strip the ethernet header added for pass through VRF device */ + __skb_pull(skb, skb_network_offset(skb)); + if (!ip4h->saddr) { ip4h->saddr = inet_select_addr(skb_dst(skb)->dev, 0, RT_SCOPE_LINK); @@ -200,9 +194,6 @@ err: static netdev_tx_t is_ip_tx_frame(struct sk_buff *skb, struct net_device *dev) { - /* strip the ethernet header added for pass through VRF device */ - __skb_pull(skb, skb_network_offset(skb)); - switch (skb->protocol) { case htons(ETH_P_IP): return vrf_process_v4_outbound(skb, dev); -- cgit v0.10.2 From 671cd19ade97e98046d0b0d1d470d4968f012401 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 2 Jun 2016 13:15:11 -0700 Subject: net: vrf: ipv4 support for local traffic to local addresses Add support for locally originated traffic to VRF-local addresses. If destination device for an skb is the loopback or VRF device then set its dst to a local version of the VRF cached dst_entry and call netif_rx to insert the packet onto the rx queue - similar to what is done for loopback. This patch handles IPv4 support; follow on patch handles IPv6. With this patch, ping, tcp and udp packets to a local IPv4 address are successfully routed: $ ip addr show dev eth1 4: eth1: mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000 link/ether 02:e0:f9:1c:b9:74 brd ff:ff:ff:ff:ff:ff inet 10.100.1.1/24 brd 10.100.1.255 scope global eth1 valid_lft forever preferred_lft forever inet6 2100:1::1/120 scope global valid_lft forever preferred_lft forever inet6 fe80::e0:f9ff:fe1c:b974/64 scope link valid_lft forever preferred_lft forever $ ping -c1 -I red 10.100.1.1 ping: Warning: source address might be selected on device other than red. PING 10.100.1.1 (10.100.1.1) from 10.100.1.1 red: 56(84) bytes of data. 64 bytes from 10.100.1.1: icmp_seq=1 ttl=64 time=0.057 ms This patch also enables use of IPv4 loopback address on the VRF device: $ ip addr add dev red 127.0.0.1/8 $ ping -c1 -I red 127.0.0.1 PING 127.0.0.1 (127.0.0.1) from 127.0.0.1 red: 56(84) bytes of data. 64 bytes from 127.0.0.1: icmp_seq=1 ttl=64 time=0.058 ms Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index ec0cb65..203d3c7 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -44,6 +44,7 @@ struct net_vrf { struct rtable __rcu *rth; + struct rtable __rcu *rth_local; struct rt6_info __rcu *rt6; u32 tb_id; }; @@ -54,9 +55,20 @@ struct pcpu_dstats { u64 tx_drps; u64 rx_pkts; u64 rx_bytes; + u64 rx_drps; struct u64_stats_sync syncp; }; +static void vrf_rx_stats(struct net_device *dev, int len) +{ + struct pcpu_dstats *dstats = this_cpu_ptr(dev->dstats); + + u64_stats_update_begin(&dstats->syncp); + dstats->rx_pkts++; + dstats->rx_bytes += len; + u64_stats_update_end(&dstats->syncp); +} + static void vrf_tx_error(struct net_device *vrf_dev, struct sk_buff *skb) { vrf_dev->stats.tx_errors++; @@ -91,6 +103,34 @@ static struct rtnl_link_stats64 *vrf_get_stats64(struct net_device *dev, return stats; } +/* Local traffic destined to local address. Reinsert the packet to rx + * path, similar to loopback handling. + */ +static int vrf_local_xmit(struct sk_buff *skb, struct net_device *dev, + struct dst_entry *dst) +{ + int len = skb->len; + + skb_orphan(skb); + + skb_dst_set(skb, dst); + skb_dst_force(skb); + + /* set pkt_type to avoid skb hitting packet taps twice - + * once on Tx and again in Rx processing + */ + skb->pkt_type = PACKET_LOOPBACK; + + skb->protocol = eth_type_trans(skb, dev); + + if (likely(netif_rx(skb) == NET_RX_SUCCESS)) + vrf_rx_stats(dev, len); + else + this_cpu_inc(dev->dstats->rx_drps); + + return NETDEV_TX_OK; +} + #if IS_ENABLED(CONFIG_IPV6) static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, struct net_device *dev) @@ -169,6 +209,34 @@ static netdev_tx_t vrf_process_v4_outbound(struct sk_buff *skb, } skb_dst_drop(skb); + + /* if dst.dev is loopback or the VRF device again this is locally + * originated traffic destined to a local address. Short circuit + * to Rx path using our local dst + */ + if (rt->dst.dev == net->loopback_dev || rt->dst.dev == vrf_dev) { + struct net_vrf *vrf = netdev_priv(vrf_dev); + struct rtable *rth_local; + struct dst_entry *dst = NULL; + + ip_rt_put(rt); + + rcu_read_lock(); + + rth_local = rcu_dereference(vrf->rth_local); + if (likely(rth_local)) { + dst = &rth_local->dst; + dst_hold(dst); + } + + rcu_read_unlock(); + + if (unlikely(!dst)) + goto err; + + return vrf_local_xmit(skb, vrf_dev, dst); + } + skb_dst_set(skb, &rt->dst); /* strip the ethernet header added for pass through VRF device */ @@ -375,29 +443,48 @@ static int vrf_output(struct net *net, struct sock *sk, struct sk_buff *skb) static void vrf_rtable_release(struct net_vrf *vrf) { struct rtable *rth = rtnl_dereference(vrf->rth); + struct rtable *rth_local = rtnl_dereference(vrf->rth_local); - rcu_assign_pointer(vrf->rth, NULL); + RCU_INIT_POINTER(vrf->rth, NULL); + RCU_INIT_POINTER(vrf->rth_local, NULL); + synchronize_rcu(); if (rth) dst_release(&rth->dst); + + if (rth_local) + dst_release(&rth_local->dst); } static int vrf_rtable_create(struct net_device *dev) { struct net_vrf *vrf = netdev_priv(dev); - struct rtable *rth; + struct rtable *rth, *rth_local; if (!fib_new_table(dev_net(dev), vrf->tb_id)) return -ENOMEM; + /* create a dst for routing packets out through a VRF device */ rth = rt_dst_alloc(dev, 0, RTN_UNICAST, 1, 1, 0); if (!rth) return -ENOMEM; + /* create a dst for local ingress routing - packets sent locally + * to local address via the VRF device as a loopback + */ + rth_local = rt_dst_alloc(dev, RTCF_LOCAL, RTN_LOCAL, 1, 1, 0); + if (!rth_local) { + dst_release(&rth->dst); + return -ENOMEM; + } + rth->dst.output = vrf_output; rth->rt_table_id = vrf->tb_id; + rth_local->rt_table_id = vrf->tb_id; + rcu_assign_pointer(vrf->rth, rth); + rcu_assign_pointer(vrf->rth_local, rth_local); return 0; } @@ -652,10 +739,19 @@ static struct sk_buff *vrf_ip_rcv(struct net_device *vrf_dev, skb->dev = vrf_dev; skb->skb_iif = vrf_dev->ifindex; + /* loopback traffic; do not push through packet taps again. + * Reset pkt_type for upper layers to process skb + */ + if (skb->pkt_type == PACKET_LOOPBACK) { + skb->pkt_type = PACKET_HOST; + goto out; + } + skb_push(skb, skb->mac_len); dev_queue_xmit_nit(skb, vrf_dev); skb_pull(skb, skb->mac_len); +out: return skb; } -- cgit v0.10.2 From 625b47b507329c4d12b256bca245e70467d3b096 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 2 Jun 2016 13:15:12 -0700 Subject: net: vrf: ipv6 support for local traffic to local addresses Add support for locally originated traffic to VRF-local IPv6 addresses. Similar to IPv4 a local dst is set on the skb and the packet is reinserted with a call to netif_rx. With this patch, ping, tcp and udp packets to a local IPv6 address are successfully routed: $ ip addr show dev eth1 4: eth1: mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000 link/ether 02:e0:f9:1c:b9:74 brd ff:ff:ff:ff:ff:ff inet 10.100.1.1/24 brd 10.100.1.255 scope global eth1 valid_lft forever preferred_lft forever inet6 2100:1::1/120 scope global valid_lft forever preferred_lft forever inet6 fe80::e0:f9ff:fe1c:b974/64 scope link valid_lft forever preferred_lft forever $ ping6 -c1 -I red 2100:1::1 ping6: Warning: source address might be selected on device other than red. PING 2100:1::1(2100:1::1) from 2100:1::1 red: 56 data bytes 64 bytes from 2100:1::1: icmp_seq=1 ttl=64 time=0.098 ms ip6_input is exported so the VRF driver can use it for the dst input function. The dst_alloc function for IPv4 defaults to setting the input and output functions; IPv6's does not. VRF does not need to duplicate the Rx path so just export the ipv6 input function. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 203d3c7..1b214ea 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -46,6 +46,7 @@ struct net_vrf { struct rtable __rcu *rth; struct rtable __rcu *rth_local; struct rt6_info __rcu *rt6; + struct rt6_info __rcu *rt6_local; u32 tb_id; }; @@ -157,6 +158,46 @@ static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, goto err; skb_dst_drop(skb); + + /* if dst.dev is loopback or the VRF device again this is locally + * originated traffic destined to a local address. Short circuit + * to Rx path using our local dst + */ + if (dst->dev == net->loopback_dev || dst->dev == dev) { + struct net_vrf *vrf = netdev_priv(dev); + struct rt6_info *rt6_local; + + /* release looked up dst and use cached local dst */ + dst_release(dst); + + rcu_read_lock(); + + rt6_local = rcu_dereference(vrf->rt6_local); + if (unlikely(!rt6_local)) { + rcu_read_unlock(); + goto err; + } + + /* Ordering issue: cached local dst is created on newlink + * before the IPv6 initialization. Using the local dst + * requires rt6i_idev to be set so make sure it is. + */ + if (unlikely(!rt6_local->rt6i_idev)) { + rt6_local->rt6i_idev = in6_dev_get(dev); + if (!rt6_local->rt6i_idev) { + rcu_read_unlock(); + goto err; + } + } + + dst = &rt6_local->dst; + dst_hold(dst); + + rcu_read_unlock(); + + return vrf_local_xmit(skb, dev, &rt6_local->dst); + } + skb_dst_set(skb, dst); /* strip the ethernet header added for pass through VRF device */ @@ -336,27 +377,38 @@ static int vrf_output6(struct net *net, struct sock *sk, struct sk_buff *skb) static void vrf_rt6_release(struct net_vrf *vrf) { struct rt6_info *rt6 = rtnl_dereference(vrf->rt6); + struct rt6_info *rt6_local = rtnl_dereference(vrf->rt6_local); - rcu_assign_pointer(vrf->rt6, NULL); + RCU_INIT_POINTER(vrf->rt6, NULL); + RCU_INIT_POINTER(vrf->rt6_local, NULL); + synchronize_rcu(); if (rt6) dst_release(&rt6->dst); + + if (rt6_local) { + if (rt6_local->rt6i_idev) + in6_dev_put(rt6_local->rt6i_idev); + + dst_release(&rt6_local->dst); + } } static int vrf_rt6_create(struct net_device *dev) { + int flags = DST_HOST | DST_NOPOLICY | DST_NOXFRM | DST_NOCACHE; struct net_vrf *vrf = netdev_priv(dev); struct net *net = dev_net(dev); struct fib6_table *rt6i_table; - struct rt6_info *rt6; + struct rt6_info *rt6, *rt6_local; int rc = -ENOMEM; rt6i_table = fib6_new_table(net, vrf->tb_id); if (!rt6i_table) goto out; - rt6 = ip6_dst_alloc(net, dev, - DST_HOST | DST_NOPOLICY | DST_NOXFRM | DST_NOCACHE); + /* create a dst for routing packets out a VRF device */ + rt6 = ip6_dst_alloc(net, dev, flags); if (!rt6) goto out; @@ -364,7 +416,25 @@ static int vrf_rt6_create(struct net_device *dev) rt6->rt6i_table = rt6i_table; rt6->dst.output = vrf_output6; + + /* create a dst for local routing - packets sent locally + * to local address via the VRF device as a loopback + */ + rt6_local = ip6_dst_alloc(net, dev, flags); + if (!rt6_local) { + dst_release(&rt6->dst); + goto out; + } + + dst_hold(&rt6_local->dst); + + rt6_local->rt6i_idev = in6_dev_get(dev); + rt6_local->rt6i_flags = RTF_UP | RTF_NONEXTHOP | RTF_LOCAL; + rt6_local->rt6i_table = rt6i_table; + rt6_local->dst.input = ip6_input; + rcu_assign_pointer(vrf->rt6, rt6); + rcu_assign_pointer(vrf->rt6_local, rt6_local); rc = 0; out: @@ -710,6 +780,16 @@ out: static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, struct sk_buff *skb) { + /* loopback traffic; do not push through packet taps again. + * Reset pkt_type for upper layers to process skb + */ + if (skb->pkt_type == PACKET_LOOPBACK) { + skb->dev = vrf_dev; + skb->skb_iif = vrf_dev->ifindex; + skb->pkt_type = PACKET_HOST; + goto out; + } + /* if packet is NDISC keep the ingress interface */ if (!ipv6_ndisc_frame(skb)) { skb->dev = vrf_dev; @@ -722,6 +802,7 @@ static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, IP6CB(skb)->flags |= IP6SKB_L3SLAVE; } +out: return skb; } -- cgit v0.10.2 From 3d9dc408fa6abf8dd1ea39e243e565891004a6f9 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 6 Jun 2016 15:58:34 -0700 Subject: net: Revert vrf-local changes. This reverts commit 2fb7ea455d57e22110c54fc2de0656b6f744263c. It results in build errors because ip6_input is not a symbol exported to modules. Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 1b214ea..d356f5d 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -44,9 +44,7 @@ struct net_vrf { struct rtable __rcu *rth; - struct rtable __rcu *rth_local; struct rt6_info __rcu *rt6; - struct rt6_info __rcu *rt6_local; u32 tb_id; }; @@ -56,20 +54,9 @@ struct pcpu_dstats { u64 tx_drps; u64 rx_pkts; u64 rx_bytes; - u64 rx_drps; struct u64_stats_sync syncp; }; -static void vrf_rx_stats(struct net_device *dev, int len) -{ - struct pcpu_dstats *dstats = this_cpu_ptr(dev->dstats); - - u64_stats_update_begin(&dstats->syncp); - dstats->rx_pkts++; - dstats->rx_bytes += len; - u64_stats_update_end(&dstats->syncp); -} - static void vrf_tx_error(struct net_device *vrf_dev, struct sk_buff *skb) { vrf_dev->stats.tx_errors++; @@ -104,34 +91,6 @@ static struct rtnl_link_stats64 *vrf_get_stats64(struct net_device *dev, return stats; } -/* Local traffic destined to local address. Reinsert the packet to rx - * path, similar to loopback handling. - */ -static int vrf_local_xmit(struct sk_buff *skb, struct net_device *dev, - struct dst_entry *dst) -{ - int len = skb->len; - - skb_orphan(skb); - - skb_dst_set(skb, dst); - skb_dst_force(skb); - - /* set pkt_type to avoid skb hitting packet taps twice - - * once on Tx and again in Rx processing - */ - skb->pkt_type = PACKET_LOOPBACK; - - skb->protocol = eth_type_trans(skb, dev); - - if (likely(netif_rx(skb) == NET_RX_SUCCESS)) - vrf_rx_stats(dev, len); - else - this_cpu_inc(dev->dstats->rx_drps); - - return NETDEV_TX_OK; -} - #if IS_ENABLED(CONFIG_IPV6) static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, struct net_device *dev) @@ -158,51 +117,8 @@ static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, goto err; skb_dst_drop(skb); - - /* if dst.dev is loopback or the VRF device again this is locally - * originated traffic destined to a local address. Short circuit - * to Rx path using our local dst - */ - if (dst->dev == net->loopback_dev || dst->dev == dev) { - struct net_vrf *vrf = netdev_priv(dev); - struct rt6_info *rt6_local; - - /* release looked up dst and use cached local dst */ - dst_release(dst); - - rcu_read_lock(); - - rt6_local = rcu_dereference(vrf->rt6_local); - if (unlikely(!rt6_local)) { - rcu_read_unlock(); - goto err; - } - - /* Ordering issue: cached local dst is created on newlink - * before the IPv6 initialization. Using the local dst - * requires rt6i_idev to be set so make sure it is. - */ - if (unlikely(!rt6_local->rt6i_idev)) { - rt6_local->rt6i_idev = in6_dev_get(dev); - if (!rt6_local->rt6i_idev) { - rcu_read_unlock(); - goto err; - } - } - - dst = &rt6_local->dst; - dst_hold(dst); - - rcu_read_unlock(); - - return vrf_local_xmit(skb, dev, &rt6_local->dst); - } - skb_dst_set(skb, dst); - /* strip the ethernet header added for pass through VRF device */ - __skb_pull(skb, skb_network_offset(skb)); - ret = ip6_local_out(net, skb->sk, skb); if (unlikely(net_xmit_eval(ret))) dev->stats.tx_errors++; @@ -223,6 +139,29 @@ static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, } #endif +static int vrf_send_v4_prep(struct sk_buff *skb, struct flowi4 *fl4, + struct net_device *vrf_dev) +{ + struct rtable *rt; + int err = 1; + + rt = ip_route_output_flow(dev_net(vrf_dev), fl4, NULL); + if (IS_ERR(rt)) + goto out; + + /* TO-DO: what about broadcast ? */ + if (rt->rt_type != RTN_UNICAST && rt->rt_type != RTN_LOCAL) { + ip_rt_put(rt); + goto out; + } + + skb_dst_drop(skb); + skb_dst_set(skb, &rt->dst); + err = 0; +out: + return err; +} + static netdev_tx_t vrf_process_v4_outbound(struct sk_buff *skb, struct net_device *vrf_dev) { @@ -237,51 +176,9 @@ static netdev_tx_t vrf_process_v4_outbound(struct sk_buff *skb, FLOWI_FLAG_SKIP_NH_OIF, .daddr = ip4h->daddr, }; - struct net *net = dev_net(vrf_dev); - struct rtable *rt; - - rt = ip_route_output_flow(net, &fl4, NULL); - if (IS_ERR(rt)) - goto err; - if (rt->rt_type != RTN_UNICAST && rt->rt_type != RTN_LOCAL) { - ip_rt_put(rt); + if (vrf_send_v4_prep(skb, &fl4, vrf_dev)) goto err; - } - - skb_dst_drop(skb); - - /* if dst.dev is loopback or the VRF device again this is locally - * originated traffic destined to a local address. Short circuit - * to Rx path using our local dst - */ - if (rt->dst.dev == net->loopback_dev || rt->dst.dev == vrf_dev) { - struct net_vrf *vrf = netdev_priv(vrf_dev); - struct rtable *rth_local; - struct dst_entry *dst = NULL; - - ip_rt_put(rt); - - rcu_read_lock(); - - rth_local = rcu_dereference(vrf->rth_local); - if (likely(rth_local)) { - dst = &rth_local->dst; - dst_hold(dst); - } - - rcu_read_unlock(); - - if (unlikely(!dst)) - goto err; - - return vrf_local_xmit(skb, vrf_dev, dst); - } - - skb_dst_set(skb, &rt->dst); - - /* strip the ethernet header added for pass through VRF device */ - __skb_pull(skb, skb_network_offset(skb)); if (!ip4h->saddr) { ip4h->saddr = inet_select_addr(skb_dst(skb)->dev, 0, @@ -303,6 +200,9 @@ err: static netdev_tx_t is_ip_tx_frame(struct sk_buff *skb, struct net_device *dev) { + /* strip the ethernet header added for pass through VRF device */ + __skb_pull(skb, skb_network_offset(skb)); + switch (skb->protocol) { case htons(ETH_P_IP): return vrf_process_v4_outbound(skb, dev); @@ -377,38 +277,27 @@ static int vrf_output6(struct net *net, struct sock *sk, struct sk_buff *skb) static void vrf_rt6_release(struct net_vrf *vrf) { struct rt6_info *rt6 = rtnl_dereference(vrf->rt6); - struct rt6_info *rt6_local = rtnl_dereference(vrf->rt6_local); - RCU_INIT_POINTER(vrf->rt6, NULL); - RCU_INIT_POINTER(vrf->rt6_local, NULL); - synchronize_rcu(); + rcu_assign_pointer(vrf->rt6, NULL); if (rt6) dst_release(&rt6->dst); - - if (rt6_local) { - if (rt6_local->rt6i_idev) - in6_dev_put(rt6_local->rt6i_idev); - - dst_release(&rt6_local->dst); - } } static int vrf_rt6_create(struct net_device *dev) { - int flags = DST_HOST | DST_NOPOLICY | DST_NOXFRM | DST_NOCACHE; struct net_vrf *vrf = netdev_priv(dev); struct net *net = dev_net(dev); struct fib6_table *rt6i_table; - struct rt6_info *rt6, *rt6_local; + struct rt6_info *rt6; int rc = -ENOMEM; rt6i_table = fib6_new_table(net, vrf->tb_id); if (!rt6i_table) goto out; - /* create a dst for routing packets out a VRF device */ - rt6 = ip6_dst_alloc(net, dev, flags); + rt6 = ip6_dst_alloc(net, dev, + DST_HOST | DST_NOPOLICY | DST_NOXFRM | DST_NOCACHE); if (!rt6) goto out; @@ -416,25 +305,7 @@ static int vrf_rt6_create(struct net_device *dev) rt6->rt6i_table = rt6i_table; rt6->dst.output = vrf_output6; - - /* create a dst for local routing - packets sent locally - * to local address via the VRF device as a loopback - */ - rt6_local = ip6_dst_alloc(net, dev, flags); - if (!rt6_local) { - dst_release(&rt6->dst); - goto out; - } - - dst_hold(&rt6_local->dst); - - rt6_local->rt6i_idev = in6_dev_get(dev); - rt6_local->rt6i_flags = RTF_UP | RTF_NONEXTHOP | RTF_LOCAL; - rt6_local->rt6i_table = rt6i_table; - rt6_local->dst.input = ip6_input; - rcu_assign_pointer(vrf->rt6, rt6); - rcu_assign_pointer(vrf->rt6_local, rt6_local); rc = 0; out: @@ -513,48 +384,29 @@ static int vrf_output(struct net *net, struct sock *sk, struct sk_buff *skb) static void vrf_rtable_release(struct net_vrf *vrf) { struct rtable *rth = rtnl_dereference(vrf->rth); - struct rtable *rth_local = rtnl_dereference(vrf->rth_local); - RCU_INIT_POINTER(vrf->rth, NULL); - RCU_INIT_POINTER(vrf->rth_local, NULL); - synchronize_rcu(); + rcu_assign_pointer(vrf->rth, NULL); if (rth) dst_release(&rth->dst); - - if (rth_local) - dst_release(&rth_local->dst); } static int vrf_rtable_create(struct net_device *dev) { struct net_vrf *vrf = netdev_priv(dev); - struct rtable *rth, *rth_local; + struct rtable *rth; if (!fib_new_table(dev_net(dev), vrf->tb_id)) return -ENOMEM; - /* create a dst for routing packets out through a VRF device */ rth = rt_dst_alloc(dev, 0, RTN_UNICAST, 1, 1, 0); if (!rth) return -ENOMEM; - /* create a dst for local ingress routing - packets sent locally - * to local address via the VRF device as a loopback - */ - rth_local = rt_dst_alloc(dev, RTCF_LOCAL, RTN_LOCAL, 1, 1, 0); - if (!rth_local) { - dst_release(&rth->dst); - return -ENOMEM; - } - rth->dst.output = vrf_output; rth->rt_table_id = vrf->tb_id; - rth_local->rt_table_id = vrf->tb_id; - rcu_assign_pointer(vrf->rth, rth); - rcu_assign_pointer(vrf->rth_local, rth_local); return 0; } @@ -780,16 +632,6 @@ out: static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, struct sk_buff *skb) { - /* loopback traffic; do not push through packet taps again. - * Reset pkt_type for upper layers to process skb - */ - if (skb->pkt_type == PACKET_LOOPBACK) { - skb->dev = vrf_dev; - skb->skb_iif = vrf_dev->ifindex; - skb->pkt_type = PACKET_HOST; - goto out; - } - /* if packet is NDISC keep the ingress interface */ if (!ipv6_ndisc_frame(skb)) { skb->dev = vrf_dev; @@ -802,7 +644,6 @@ static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, IP6CB(skb)->flags |= IP6SKB_L3SLAVE; } -out: return skb; } @@ -820,19 +661,10 @@ static struct sk_buff *vrf_ip_rcv(struct net_device *vrf_dev, skb->dev = vrf_dev; skb->skb_iif = vrf_dev->ifindex; - /* loopback traffic; do not push through packet taps again. - * Reset pkt_type for upper layers to process skb - */ - if (skb->pkt_type == PACKET_LOOPBACK) { - skb->pkt_type = PACKET_HOST; - goto out; - } - skb_push(skb, skb->mac_len); dev_queue_xmit_nit(skb, vrf_dev); skb_pull(skb, skb->mac_len); -out: return skb; } -- cgit v0.10.2 From 14de9d114a82a564b94388c95af79a701dc93134 Mon Sep 17 00:00:00 2001 From: Aaron Conole Date: Fri, 3 Jun 2016 16:57:12 -0400 Subject: virtio-net: Add initial MTU advice feature This commit adds the feature bit and associated mtu device entry for the virtio network device. When a virtio device comes up, it checks the feature bit for the VIRTIO_NET_F_MTU feature. If such feature bit is enabled, the driver will read the advised MTU and use it as the initial value. Signed-off-by: Aaron Conole Signed-off-by: David S. Miller diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index e0638e5..192f321 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -1780,6 +1780,7 @@ static int virtnet_probe(struct virtio_device *vdev) struct net_device *dev; struct virtnet_info *vi; u16 max_queue_pairs; + int mtu; if (!vdev->config->get) { dev_err(&vdev->dev, "%s failure: config access disabled\n", @@ -1896,6 +1897,14 @@ static int virtnet_probe(struct virtio_device *vdev) if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_VQ)) vi->has_cvq = true; + if (virtio_has_feature(vdev, VIRTIO_NET_F_MTU)) { + mtu = virtio_cread16(vdev, + offsetof(struct virtio_net_config, + mtu)); + if (virtnet_change_mtu(dev, mtu)) + __virtio_clear_bit(vdev, VIRTIO_NET_F_MTU); + } + if (vi->any_header_sg) dev->needed_headroom = vi->hdr_len; @@ -2067,6 +2076,7 @@ static unsigned int features[] = { VIRTIO_NET_F_GUEST_ANNOUNCE, VIRTIO_NET_F_MQ, VIRTIO_NET_F_CTRL_MAC_ADDR, VIRTIO_F_ANY_LAYOUT, + VIRTIO_NET_F_MTU, }; static struct virtio_driver virtio_net_driver = { diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h index ec32293..1ab4ea6 100644 --- a/include/uapi/linux/virtio_net.h +++ b/include/uapi/linux/virtio_net.h @@ -55,6 +55,7 @@ #define VIRTIO_NET_F_MQ 22 /* Device supports Receive Flow * Steering */ #define VIRTIO_NET_F_CTRL_MAC_ADDR 23 /* Set MAC address */ +#define VIRTIO_NET_F_MTU 25 /* Initial MTU advice */ #ifndef VIRTIO_NET_NO_LEGACY #define VIRTIO_NET_F_GSO 6 /* Host handles pkts w/ any GSO type */ @@ -73,6 +74,8 @@ struct virtio_net_config { * Legal values are between 1 and 0x8000 */ __u16 max_virtqueue_pairs; + /* Default maximum transmit unit advice */ + __u16 mtu; } __attribute__((packed)); /* -- cgit v0.10.2 From 523a61b488afb51af18729d8300b7f624761370f Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Sat, 4 Jun 2016 20:21:40 +0530 Subject: net: ethernet: cavium: liquidio: response_manager: Remove create_workqueue alloc_workqueue replaces deprecated create_workqueue(). A dedicated workqueue has been used since the workitem viz (&cwq->wk.work which maps to oct_poll_req_completion) is involved in normal device operation. WQ_MEM_RECLAIM has been set to guarantee forward progress under memory pressure, which is a requirement here. Since there are only a fixed number of work items, explicit concurrency limit is unnecessary. flush_workqueue is unnecessary since destroy_workqueue() itself calls drain_workqueue() which flushes repeatedly till the workqueue becomes empty. Hence the call to flush_workqueue() has been dropped. Signed-off-by: Bhaktipriya Shridhar Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.c b/drivers/net/ethernet/cavium/liquidio/response_manager.c index 091f537..6287a7c 100644 --- a/drivers/net/ethernet/cavium/liquidio/response_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/response_manager.c @@ -55,7 +55,7 @@ int octeon_setup_response_list(struct octeon_device *oct) atomic_set(&oct->response_list[i].pending_req_count, 0); } - oct->dma_comp_wq.wq = create_workqueue("dma-comp"); + oct->dma_comp_wq.wq = alloc_workqueue("dma-comp", WQ_MEM_RECLAIM, 0); if (!oct->dma_comp_wq.wq) { dev_err(&oct->pci_dev->dev, "failed to create wq thread\n"); return -ENOMEM; @@ -72,7 +72,6 @@ int octeon_setup_response_list(struct octeon_device *oct) void octeon_delete_response_list(struct octeon_device *oct) { cancel_delayed_work_sync(&oct->dma_comp_wq.wk.work); - flush_workqueue(oct->dma_comp_wq.wq); destroy_workqueue(oct->dma_comp_wq.wq); } -- cgit v0.10.2 From aaa76724d7ece87eb5f66e02fac9da036dd20667 Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Sat, 4 Jun 2016 20:54:00 +0530 Subject: net: ethernet: cavium: liquidio: request_manager: Remove create_workqueue alloc_workqueue replaces deprecated create_workqueue(). A dedicated workqueue has been used since the workitem viz (&db_wq->wk.work which maps to check_db_timeout) is involved in normal device operation. WQ_MEM_RECLAIM has been set to guarantee forward progress under memory pressure, which is a requirement here. Since there are only a fixed number of work items, explicit concurrency limit is unnecessary. flush_workqueue is unnecessary since destroy_workqueue() itself calls drain_workqueue() which flushes repeatedly till the workqueue becomes empty. Signed-off-by: Bhaktipriya Shridhar Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index a2a2465..9313915 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -144,7 +144,9 @@ int octeon_init_instr_queue(struct octeon_device *oct, oct->fn_list.setup_iq_regs(oct, iq_no); - oct->check_db_wq[iq_no].wq = create_workqueue("check_iq_db"); + oct->check_db_wq[iq_no].wq = alloc_workqueue("check_iq_db", + WQ_MEM_RECLAIM, + 0); if (!oct->check_db_wq[iq_no].wq) { lio_dma_free(oct, q_size, iq->base_addr, iq->base_addr_dma); dev_err(&oct->pci_dev->dev, "check db wq create failed for iq %d\n", @@ -168,7 +170,6 @@ int octeon_delete_instr_queue(struct octeon_device *oct, u32 iq_no) struct octeon_instr_queue *iq = oct->instr_queue[iq_no]; cancel_delayed_work_sync(&oct->check_db_wq[iq_no].wk.work); - flush_workqueue(oct->check_db_wq[iq_no].wq); destroy_workqueue(oct->check_db_wq[iq_no].wq); if (OCTEON_CN6XXX(oct)) -- cgit v0.10.2 From 6fd3dd7160918f528b0d6b7084d18b31cd5fced5 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Thu, 2 Jun 2016 17:59:49 +0300 Subject: ath10k: add QCA9887 chipset support Add the hardware name, revision, firmware names and update the pci_id table. QA9887 HW1.0 is supposed to be similar to QCA988X HW2.0 . Details about he firmware interface are currently unknown. Signed-off-by: Sven Eckelmann [kvalo@qca.qualcomm.com: add a warning about experimental support] Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index cedf127..ed4e52e 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -69,6 +69,25 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = { }, }, { + .id = QCA9887_HW_1_0_VERSION, + .dev_id = QCA9887_1_0_DEVICE_ID, + .name = "qca9887 hw1.0", + .patch_load_addr = QCA9887_HW_1_0_PATCH_LOAD_ADDR, + .uart_pin = 7, + .has_shifted_cc_wraparound = true, + .otp_exe_param = 0, + .channel_counters_freq_hz = 88000, + .max_probe_resp_desc_thres = 0, + .hw_4addr_pad = ATH10K_HW_4ADDR_PAD_AFTER, + .cal_data_len = 2116, + .fw = { + .dir = QCA9887_HW_1_0_FW_DIR, + .board = QCA9887_HW_1_0_BOARD_DATA_FILE, + .board_size = QCA9887_BOARD_DATA_SZ, + .board_ext_size = QCA9887_BOARD_EXT_DATA_SZ, + }, + }, + { .id = QCA6174_HW_2_1_VERSION, .dev_id = QCA6164_2_1_DEVICE_ID, .name = "qca6164 hw2.1", @@ -2095,6 +2114,7 @@ struct ath10k *ath10k_core_create(size_t priv_size, struct device *dev, switch (hw_rev) { case ATH10K_HW_QCA988X: + case ATH10K_HW_QCA9887: ar->regs = &qca988x_regs; ar->hw_values = &qca988x_values; break; diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h index f41c91c..5ef1fa0 100644 --- a/drivers/net/wireless/ath/ath10k/hw.h +++ b/drivers/net/wireless/ath/ath10k/hw.h @@ -28,6 +28,7 @@ #define QCA99X0_2_0_DEVICE_ID (0x0040) #define QCA9984_1_0_DEVICE_ID (0x0046) #define QCA9377_1_0_DEVICE_ID (0x0042) +#define QCA9887_1_0_DEVICE_ID (0x0050) /* QCA988X 1.0 definitions (unsupported) */ #define QCA988X_HW_1_0_CHIP_ID_REV 0x0 @@ -39,6 +40,13 @@ #define QCA988X_HW_2_0_BOARD_DATA_FILE "board.bin" #define QCA988X_HW_2_0_PATCH_LOAD_ADDR 0x1234 +/* QCA9887 1.0 definitions */ +#define QCA9887_HW_1_0_VERSION 0x4100016d +#define QCA9887_HW_1_0_CHIP_ID_REV 0 +#define QCA9887_HW_1_0_FW_DIR ATH10K_FW_DIR "/QCA9887/hw1.0" +#define QCA9887_HW_1_0_BOARD_DATA_FILE "board.bin" +#define QCA9887_HW_1_0_PATCH_LOAD_ADDR 0x1234 + /* QCA6174 target BMI version signatures */ #define QCA6174_HW_1_0_VERSION 0x05000000 #define QCA6174_HW_1_1_VERSION 0x05000001 @@ -205,6 +213,7 @@ enum ath10k_hw_rev { ATH10K_HW_QCA9984, ATH10K_HW_QCA9377, ATH10K_HW_QCA4019, + ATH10K_HW_QCA9887, }; struct ath10k_hw_regs { @@ -257,6 +266,7 @@ void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey, u32 cc, u32 rcc, u32 cc_prev, u32 rcc_prev); #define QCA_REV_988X(ar) ((ar)->hw_rev == ATH10K_HW_QCA988X) +#define QCA_REV_9887(ar) ((ar)->hw_rev == ATH10K_HW_QCA9887) #define QCA_REV_6174(ar) ((ar)->hw_rev == ATH10K_HW_QCA6174) #define QCA_REV_99X0(ar) ((ar)->hw_rev == ATH10K_HW_QCA99X0) #define QCA_REV_9984(ar) ((ar)->hw_rev == ATH10K_HW_QCA9984) diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c index 4150189..3e8e7ed 100644 --- a/drivers/net/wireless/ath/ath10k/pci.c +++ b/drivers/net/wireless/ath/ath10k/pci.c @@ -58,6 +58,7 @@ static const struct pci_device_id ath10k_pci_id_table[] = { { PCI_VDEVICE(ATHEROS, QCA99X0_2_0_DEVICE_ID) }, /* PCI-E QCA99X0 V2 */ { PCI_VDEVICE(ATHEROS, QCA9984_1_0_DEVICE_ID) }, /* PCI-E QCA9984 V1 */ { PCI_VDEVICE(ATHEROS, QCA9377_1_0_DEVICE_ID) }, /* PCI-E QCA9377 V1 */ + { PCI_VDEVICE(ATHEROS, QCA9887_1_0_DEVICE_ID) }, /* PCI-E QCA9887 */ {0} }; @@ -87,6 +88,7 @@ static const struct ath10k_pci_supp_chip ath10k_pci_supp_chips[] = { { QCA9377_1_0_DEVICE_ID, QCA9377_HW_1_0_CHIP_ID_REV }, { QCA9377_1_0_DEVICE_ID, QCA9377_HW_1_1_CHIP_ID_REV }, + { QCA9887_1_0_DEVICE_ID, QCA9887_HW_1_0_CHIP_ID_REV }, }; static void ath10k_pci_buffer_cleanup(struct ath10k *ar); @@ -841,6 +843,7 @@ static u32 ath10k_pci_targ_cpu_to_ce_addr(struct ath10k *ar, u32 addr) switch (ar->hw_rev) { case ATH10K_HW_QCA988X: + case ATH10K_HW_QCA9887: case ATH10K_HW_QCA6174: case ATH10K_HW_QCA9377: val = (ath10k_pci_read32(ar, SOC_CORE_BASE_ADDRESS + @@ -1569,6 +1572,7 @@ static void ath10k_pci_irq_msi_fw_mask(struct ath10k *ar) switch (ar->hw_rev) { case ATH10K_HW_QCA988X: + case ATH10K_HW_QCA9887: case ATH10K_HW_QCA6174: case ATH10K_HW_QCA9377: val = ath10k_pci_read32(ar, SOC_CORE_BASE_ADDRESS + @@ -1593,6 +1597,7 @@ static void ath10k_pci_irq_msi_fw_unmask(struct ath10k *ar) switch (ar->hw_rev) { case ATH10K_HW_QCA988X: + case ATH10K_HW_QCA9887: case ATH10K_HW_QCA6174: case ATH10K_HW_QCA9377: val = ath10k_pci_read32(ar, SOC_CORE_BASE_ADDRESS + @@ -1944,6 +1949,7 @@ static int ath10k_pci_get_num_banks(struct ath10k *ar) case QCA988X_2_0_DEVICE_ID: case QCA99X0_2_0_DEVICE_ID: case QCA9984_1_0_DEVICE_ID: + case QCA9887_1_0_DEVICE_ID: return 1; case QCA6164_2_1_DEVICE_ID: case QCA6174_2_1_DEVICE_ID: @@ -2998,6 +3004,13 @@ static int ath10k_pci_probe(struct pci_dev *pdev, pci_soft_reset = ath10k_pci_warm_reset; pci_hard_reset = ath10k_pci_qca988x_chip_reset; break; + case QCA9887_1_0_DEVICE_ID: + dev_warn(&pdev->dev, "QCA9887 support is still experimental, there are likely bugs. You have been warned.\n"); + hw_rev = ATH10K_HW_QCA9887; + pci_ps = false; + pci_soft_reset = ath10k_pci_warm_reset; + pci_hard_reset = ath10k_pci_qca988x_chip_reset; + break; case QCA6164_2_1_DEVICE_ID: case QCA6174_2_1_DEVICE_ID: hw_rev = ATH10K_HW_QCA6174; @@ -3210,6 +3223,11 @@ MODULE_FIRMWARE(QCA988X_HW_2_0_FW_DIR "/" ATH10K_FW_API5_FILE); MODULE_FIRMWARE(QCA988X_HW_2_0_FW_DIR "/" QCA988X_HW_2_0_BOARD_DATA_FILE); MODULE_FIRMWARE(QCA988X_HW_2_0_FW_DIR "/" ATH10K_BOARD_API2_FILE); +/* QCA9887 1.0 firmware files */ +MODULE_FIRMWARE(QCA9887_HW_1_0_FW_DIR "/" ATH10K_FW_API5_FILE); +MODULE_FIRMWARE(QCA9887_HW_1_0_FW_DIR "/" QCA9887_HW_1_0_BOARD_DATA_FILE); +MODULE_FIRMWARE(QCA9887_HW_1_0_FW_DIR "/" ATH10K_BOARD_API2_FILE); + /* QCA6174 2.1 firmware files */ MODULE_FIRMWARE(QCA6174_HW_2_1_FW_DIR "/" ATH10K_FW_API4_FILE); MODULE_FIRMWARE(QCA6174_HW_2_1_FW_DIR "/" ATH10K_FW_API5_FILE); diff --git a/drivers/net/wireless/ath/ath10k/targaddrs.h b/drivers/net/wireless/ath/ath10k/targaddrs.h index 8e24099..aaf53a8 100644 --- a/drivers/net/wireless/ath/ath10k/targaddrs.h +++ b/drivers/net/wireless/ath/ath10k/targaddrs.h @@ -447,6 +447,9 @@ Fw Mode/SubMode Mask #define QCA988X_BOARD_DATA_SZ 7168 #define QCA988X_BOARD_EXT_DATA_SZ 0 +#define QCA9887_BOARD_DATA_SZ 7168 +#define QCA9887_BOARD_EXT_DATA_SZ 0 + #define QCA6174_BOARD_DATA_SZ 8192 #define QCA6174_BOARD_EXT_DATA_SZ 0 -- cgit v0.10.2 From 6847f967339179b72eab690d46bc6e88ff847844 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Thu, 2 Jun 2016 17:59:50 +0300 Subject: ath10k: add board data download from target The QCA9887 stores its calibration data (board.bin) inside the EEPROM of the target. This has to be downloaded manually to allow the device to initialize correctly. Signed-off-by: Sven Eckelmann [kvalo@qca.qualcomm.com: handle -EOPNOTSUPP and s/fetch_board_data/fetch_cal_eeprom] Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index ed4e52e..1e88251 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -18,6 +18,7 @@ #include #include #include +#include #include "core.h" #include "mac.h" @@ -573,6 +574,35 @@ out: return ret; } +static int ath10k_download_cal_eeprom(struct ath10k *ar) +{ + size_t data_len; + void *data = NULL; + int ret; + + ret = ath10k_hif_fetch_cal_eeprom(ar, &data, &data_len); + if (ret) { + if (ret != -EOPNOTSUPP) + ath10k_warn(ar, "failed to read calibration data from EEPROM: %d\n", + ret); + goto out_free; + } + + ret = ath10k_download_board_data(ar, data, data_len); + if (ret) { + ath10k_warn(ar, "failed to download calibration data from EEPROM: %d\n", + ret); + goto out_free; + } + + ret = 0; + +out_free: + kfree(data); + + return ret; +} + static int ath10k_core_get_board_id_from_otp(struct ath10k *ar) { u32 result, address; @@ -1335,7 +1365,17 @@ static int ath10k_download_cal_data(struct ath10k *ar) } ath10k_dbg(ar, ATH10K_DBG_BOOT, - "boot did not find DT entry, try OTP next: %d\n", + "boot did not find DT entry, try target EEPROM next: %d\n", + ret); + + ret = ath10k_download_cal_eeprom(ar); + if (ret == 0) { + ar->cal_mode = ATH10K_CAL_MODE_EEPROM; + goto done; + } + + ath10k_dbg(ar, ATH10K_DBG_BOOT, + "boot did not find target EEPROM entry, try OTP next: %d\n", ret); ret = ath10k_download_and_run_otp(ar); diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h index 4462c3f..3387733 100644 --- a/drivers/net/wireless/ath/ath10k/core.h +++ b/drivers/net/wireless/ath/ath10k/core.h @@ -578,6 +578,7 @@ enum ath10k_cal_mode { ATH10K_CAL_MODE_DT, ATH10K_PRE_CAL_MODE_FILE, ATH10K_PRE_CAL_MODE_DT, + ATH10K_CAL_MODE_EEPROM, }; enum ath10k_crypt_mode { @@ -600,6 +601,8 @@ static inline const char *ath10k_cal_mode_str(enum ath10k_cal_mode mode) return "pre-cal-file"; case ATH10K_PRE_CAL_MODE_DT: return "pre-cal-dt"; + case ATH10K_CAL_MODE_EEPROM: + return "eeprom"; } return "unknown"; diff --git a/drivers/net/wireless/ath/ath10k/hif.h b/drivers/net/wireless/ath/ath10k/hif.h index 89e7076..b2566b0 100644 --- a/drivers/net/wireless/ath/ath10k/hif.h +++ b/drivers/net/wireless/ath/ath10k/hif.h @@ -87,6 +87,10 @@ struct ath10k_hif_ops { int (*suspend)(struct ath10k *ar); int (*resume)(struct ath10k *ar); + + /* fetch calibration data from target eeprom */ + int (*fetch_cal_eeprom)(struct ath10k *ar, void **data, + size_t *data_len); }; static inline int ath10k_hif_tx_sg(struct ath10k *ar, u8 pipe_id, @@ -202,4 +206,14 @@ static inline void ath10k_hif_write32(struct ath10k *ar, ar->hif.ops->write32(ar, address, data); } +static inline int ath10k_hif_fetch_cal_eeprom(struct ath10k *ar, + void **data, + size_t *data_len) +{ + if (!ar->hif.ops->fetch_cal_eeprom) + return -EOPNOTSUPP; + + return ar->hif.ops->fetch_cal_eeprom(ar, data, data_len); +} + #endif /* _HIF_H_ */ diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h index 5ef1fa0..49097af 100644 --- a/drivers/net/wireless/ath/ath10k/hw.h +++ b/drivers/net/wireless/ath/ath10k/hw.h @@ -568,7 +568,10 @@ enum ath10k_hw_4addr_pad { #define WLAN_SYSTEM_SLEEP_DISABLE_MASK 0x00000001 #define WLAN_GPIO_PIN0_ADDRESS 0x00000028 +#define WLAN_GPIO_PIN0_CONFIG_LSB 11 #define WLAN_GPIO_PIN0_CONFIG_MASK 0x00007800 +#define WLAN_GPIO_PIN0_PAD_PULL_LSB 5 +#define WLAN_GPIO_PIN0_PAD_PULL_MASK 0x00000060 #define WLAN_GPIO_PIN1_ADDRESS 0x0000002c #define WLAN_GPIO_PIN1_CONFIG_MASK 0x00007800 #define WLAN_GPIO_PIN10_ADDRESS 0x00000050 @@ -581,6 +584,8 @@ enum ath10k_hw_4addr_pad { #define CLOCK_GPIO_BT_CLK_OUT_EN_MASK 0 #define SI_CONFIG_OFFSET 0x00000000 +#define SI_CONFIG_ERR_INT_LSB 19 +#define SI_CONFIG_ERR_INT_MASK 0x00080000 #define SI_CONFIG_BIDIR_OD_DATA_LSB 18 #define SI_CONFIG_BIDIR_OD_DATA_MASK 0x00040000 #define SI_CONFIG_I2C_LSB 16 @@ -594,7 +599,9 @@ enum ath10k_hw_4addr_pad { #define SI_CONFIG_DIVIDER_LSB 0 #define SI_CONFIG_DIVIDER_MASK 0x0000000f #define SI_CS_OFFSET 0x00000004 +#define SI_CS_DONE_ERR_LSB 10 #define SI_CS_DONE_ERR_MASK 0x00000400 +#define SI_CS_DONE_INT_LSB 9 #define SI_CS_DONE_INT_MASK 0x00000200 #define SI_CS_START_LSB 8 #define SI_CS_START_MASK 0x00000100 @@ -645,7 +652,10 @@ enum ath10k_hw_4addr_pad { #define GPIO_BASE_ADDRESS WLAN_GPIO_BASE_ADDRESS #define GPIO_PIN0_OFFSET WLAN_GPIO_PIN0_ADDRESS #define GPIO_PIN1_OFFSET WLAN_GPIO_PIN1_ADDRESS +#define GPIO_PIN0_CONFIG_LSB WLAN_GPIO_PIN0_CONFIG_LSB #define GPIO_PIN0_CONFIG_MASK WLAN_GPIO_PIN0_CONFIG_MASK +#define GPIO_PIN0_PAD_PULL_LSB WLAN_GPIO_PIN0_PAD_PULL_LSB +#define GPIO_PIN0_PAD_PULL_MASK WLAN_GPIO_PIN0_PAD_PULL_MASK #define GPIO_PIN1_CONFIG_MASK WLAN_GPIO_PIN1_CONFIG_MASK #define SI_BASE_ADDRESS WLAN_SI_BASE_ADDRESS #define SCRATCH_BASE_ADDRESS SOC_CORE_BASE_ADDRESS @@ -700,6 +710,18 @@ enum ath10k_hw_4addr_pad { #define WINDOW_READ_ADDR_ADDRESS MISSING #define WINDOW_WRITE_ADDR_ADDRESS MISSING +#define QCA9887_1_0_I2C_SDA_GPIO_PIN 5 +#define QCA9887_1_0_I2C_SDA_PIN_CONFIG 3 +#define QCA9887_1_0_SI_CLK_GPIO_PIN 17 +#define QCA9887_1_0_SI_CLK_PIN_CONFIG 3 +#define QCA9887_1_0_GPIO_ENABLE_W1TS_LOW_ADDRESS 0x00000010 + +#define QCA9887_EEPROM_SELECT_READ 0xa10000a0 +#define QCA9887_EEPROM_ADDR_HI_MASK 0x0000ff00 +#define QCA9887_EEPROM_ADDR_HI_LSB 8 +#define QCA9887_EEPROM_ADDR_LO_MASK 0x00ff0000 +#define QCA9887_EEPROM_ADDR_LO_LSB 16 + #define RTC_STATE_V_GET(x) (((x) & RTC_STATE_V_MASK) >> RTC_STATE_V_LSB) #endif /* _HW_H_ */ diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c index 3e8e7ed..f06dd39 100644 --- a/drivers/net/wireless/ath/ath10k/pci.c +++ b/drivers/net/wireless/ath/ath10k/pci.c @@ -2577,6 +2577,144 @@ static int ath10k_pci_hif_resume(struct ath10k *ar) } #endif +static bool ath10k_pci_validate_cal(void *data, size_t size) +{ + __le16 *cal_words = data; + u16 checksum = 0; + size_t i; + + if (size % 2 != 0) + return false; + + for (i = 0; i < size / 2; i++) + checksum ^= le16_to_cpu(cal_words[i]); + + return checksum == 0xffff; +} + +static void ath10k_pci_enable_eeprom(struct ath10k *ar) +{ + /* Enable SI clock */ + ath10k_pci_soc_write32(ar, CLOCK_CONTROL_OFFSET, 0x0); + + /* Configure GPIOs for I2C operation */ + ath10k_pci_write32(ar, + GPIO_BASE_ADDRESS + GPIO_PIN0_OFFSET + + 4 * QCA9887_1_0_I2C_SDA_GPIO_PIN, + SM(QCA9887_1_0_I2C_SDA_PIN_CONFIG, + GPIO_PIN0_CONFIG) | + SM(1, GPIO_PIN0_PAD_PULL)); + + ath10k_pci_write32(ar, + GPIO_BASE_ADDRESS + GPIO_PIN0_OFFSET + + 4 * QCA9887_1_0_SI_CLK_GPIO_PIN, + SM(QCA9887_1_0_SI_CLK_PIN_CONFIG, GPIO_PIN0_CONFIG) | + SM(1, GPIO_PIN0_PAD_PULL)); + + ath10k_pci_write32(ar, + GPIO_BASE_ADDRESS + + QCA9887_1_0_GPIO_ENABLE_W1TS_LOW_ADDRESS, + 1u << QCA9887_1_0_SI_CLK_GPIO_PIN); + + /* In Swift ASIC - EEPROM clock will be (110MHz/512) = 214KHz */ + ath10k_pci_write32(ar, + SI_BASE_ADDRESS + SI_CONFIG_OFFSET, + SM(1, SI_CONFIG_ERR_INT) | + SM(1, SI_CONFIG_BIDIR_OD_DATA) | + SM(1, SI_CONFIG_I2C) | + SM(1, SI_CONFIG_POS_SAMPLE) | + SM(1, SI_CONFIG_INACTIVE_DATA) | + SM(1, SI_CONFIG_INACTIVE_CLK) | + SM(8, SI_CONFIG_DIVIDER)); +} + +static int ath10k_pci_read_eeprom(struct ath10k *ar, u16 addr, u8 *out) +{ + u32 reg; + int wait_limit; + + /* set device select byte and for the read operation */ + reg = QCA9887_EEPROM_SELECT_READ | + SM(addr, QCA9887_EEPROM_ADDR_LO) | + SM(addr >> 8, QCA9887_EEPROM_ADDR_HI); + ath10k_pci_write32(ar, SI_BASE_ADDRESS + SI_TX_DATA0_OFFSET, reg); + + /* write transmit data, transfer length, and START bit */ + ath10k_pci_write32(ar, SI_BASE_ADDRESS + SI_CS_OFFSET, + SM(1, SI_CS_START) | SM(1, SI_CS_RX_CNT) | + SM(4, SI_CS_TX_CNT)); + + /* wait max 1 sec */ + wait_limit = 100000; + + /* wait for SI_CS_DONE_INT */ + do { + reg = ath10k_pci_read32(ar, SI_BASE_ADDRESS + SI_CS_OFFSET); + if (MS(reg, SI_CS_DONE_INT)) + break; + + wait_limit--; + udelay(10); + } while (wait_limit > 0); + + if (!MS(reg, SI_CS_DONE_INT)) { + ath10k_err(ar, "timeout while reading device EEPROM at %04x\n", + addr); + return -ETIMEDOUT; + } + + /* clear SI_CS_DONE_INT */ + ath10k_pci_write32(ar, SI_BASE_ADDRESS + SI_CS_OFFSET, reg); + + if (MS(reg, SI_CS_DONE_ERR)) { + ath10k_err(ar, "failed to read device EEPROM at %04x\n", addr); + return -EIO; + } + + /* extract receive data */ + reg = ath10k_pci_read32(ar, SI_BASE_ADDRESS + SI_RX_DATA0_OFFSET); + *out = reg; + + return 0; +} + +static int ath10k_pci_hif_fetch_cal_eeprom(struct ath10k *ar, void **data, + size_t *data_len) +{ + u8 *caldata = NULL; + size_t calsize, i; + int ret; + + if (!QCA_REV_9887(ar)) + return -EOPNOTSUPP; + + calsize = ar->hw_params.cal_data_len; + caldata = kmalloc(calsize, GFP_KERNEL); + if (!caldata) + return -ENOMEM; + + ath10k_pci_enable_eeprom(ar); + + for (i = 0; i < calsize; i++) { + ret = ath10k_pci_read_eeprom(ar, i, &caldata[i]); + if (ret) + goto err_free; + } + + if (!ath10k_pci_validate_cal(caldata, calsize)) + goto err_free; + + *data = caldata; + *data_len = calsize; + + return 0; + +err_free: + kfree(data); + + return -EINVAL; +} + static const struct ath10k_hif_ops ath10k_pci_hif_ops = { .tx_sg = ath10k_pci_hif_tx_sg, .diag_read = ath10k_pci_hif_diag_read, @@ -2596,6 +2734,7 @@ static const struct ath10k_hif_ops ath10k_pci_hif_ops = { .suspend = ath10k_pci_hif_suspend, .resume = ath10k_pci_hif_resume, #endif + .fetch_cal_eeprom = ath10k_pci_hif_fetch_cal_eeprom, }; /* -- cgit v0.10.2 From 402f9030cb68d235cfa94b898e96f2d6f7da76ae Mon Sep 17 00:00:00 2001 From: Tobin C Harding Date: Tue, 10 May 2016 11:26:57 +1000 Subject: bridge: netfilter: checkpatch data type fixes checkpatch produces data type 'checks'. This patch amends them by changing, for example: uint8_t -> u8 Signed-off-by: Tobin C Harding Signed-off-by: Pablo Neira Ayuso diff --git a/net/bridge/netfilter/ebt_stp.c b/net/bridge/netfilter/ebt_stp.c index 6b731e1..e77f90b 100644 --- a/net/bridge/netfilter/ebt_stp.c +++ b/net/bridge/netfilter/ebt_stp.c @@ -17,24 +17,24 @@ #define BPDU_TYPE_TCN 0x80 struct stp_header { - uint8_t dsap; - uint8_t ssap; - uint8_t ctrl; - uint8_t pid; - uint8_t vers; - uint8_t type; + u8 dsap; + u8 ssap; + u8 ctrl; + u8 pid; + u8 vers; + u8 type; }; struct stp_config_pdu { - uint8_t flags; - uint8_t root[8]; - uint8_t root_cost[4]; - uint8_t sender[8]; - uint8_t port[2]; - uint8_t msg_age[2]; - uint8_t max_age[2]; - uint8_t hello_time[2]; - uint8_t forward_delay[2]; + u8 flags; + u8 root[8]; + u8 root_cost[4]; + u8 sender[8]; + u8 port[2]; + u8 msg_age[2]; + u8 max_age[2]; + u8 hello_time[2]; + u8 forward_delay[2]; }; #define NR16(p) (p[0] << 8 | p[1]) @@ -44,8 +44,8 @@ static bool ebt_filter_config(const struct ebt_stp_info *info, const struct stp_config_pdu *stpc) { const struct ebt_stp_config_info *c; - uint16_t v16; - uint32_t v32; + u16 v16; + u32 v32; int verdict, i; c = &info->config; @@ -125,7 +125,7 @@ ebt_stp_mt(const struct sk_buff *skb, struct xt_action_param *par) const struct ebt_stp_info *info = par->matchinfo; const struct stp_header *sp; struct stp_header _stph; - const uint8_t header[6] = {0x42, 0x42, 0x03, 0x00, 0x00, 0x00}; + const u8 header[6] = {0x42, 0x42, 0x03, 0x00, 0x00, 0x00}; sp = skb_header_pointer(skb, 0, sizeof(_stph), &_stph); if (sp == NULL) @@ -156,8 +156,8 @@ ebt_stp_mt(const struct sk_buff *skb, struct xt_action_param *par) static int ebt_stp_mt_check(const struct xt_mtchk_param *par) { const struct ebt_stp_info *info = par->matchinfo; - const uint8_t bridge_ula[6] = {0x01, 0x80, 0xc2, 0x00, 0x00, 0x00}; - const uint8_t msk[6] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; + const u8 bridge_ula[6] = {0x01, 0x80, 0xc2, 0x00, 0x00, 0x00}; + const u8 msk[6] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; const struct ebt_entry *e = par->entryinfo; if (info->bitmask & ~EBT_STP_MASK || info->invflags & ~EBT_STP_MASK || -- cgit v0.10.2 From 436a850dd9cac09bf88e12e20cc79408b1d29788 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sun, 15 May 2016 19:50:14 +0200 Subject: netfilter: helper: avoid extra expectation iterations on unregister The expectation table is not duplicated per net namespace anymore, so we can move the expectation table and conntrack table iteration out of the per-net loop. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index f703adb..7ba16e9 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -390,11 +390,38 @@ static void __nf_conntrack_helper_unregister(struct nf_conntrack_helper *me, struct net *net) { struct nf_conntrack_tuple_hash *h; + const struct hlist_nulls_node *nn; + int cpu; + + /* Get rid of expecteds, set helpers to NULL. */ + for_each_possible_cpu(cpu) { + struct ct_pcpu *pcpu = per_cpu_ptr(net->ct.pcpu_lists, cpu); + + spin_lock_bh(&pcpu->lock); + hlist_nulls_for_each_entry(h, nn, &pcpu->unconfirmed, hnnode) + unhelp(h, me); + spin_unlock_bh(&pcpu->lock); + } +} + +void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) +{ + struct nf_conntrack_tuple_hash *h; struct nf_conntrack_expect *exp; const struct hlist_node *next; const struct hlist_nulls_node *nn; + struct net *net; unsigned int i; - int cpu; + + mutex_lock(&nf_ct_helper_mutex); + hlist_del_rcu(&me->hnode); + nf_ct_helper_count--; + mutex_unlock(&nf_ct_helper_mutex); + + /* Make sure every nothing is still using the helper unless its a + * connection in the hash. + */ + synchronize_rcu(); /* Get rid of expectations */ spin_lock_bh(&nf_conntrack_expect_lock); @@ -414,15 +441,11 @@ static void __nf_conntrack_helper_unregister(struct nf_conntrack_helper *me, } spin_unlock_bh(&nf_conntrack_expect_lock); - /* Get rid of expecteds, set helpers to NULL. */ - for_each_possible_cpu(cpu) { - struct ct_pcpu *pcpu = per_cpu_ptr(net->ct.pcpu_lists, cpu); + rtnl_lock(); + for_each_net(net) + __nf_conntrack_helper_unregister(me, net); + rtnl_unlock(); - spin_lock_bh(&pcpu->lock); - hlist_nulls_for_each_entry(h, nn, &pcpu->unconfirmed, hnnode) - unhelp(h, me); - spin_unlock_bh(&pcpu->lock); - } local_bh_disable(); for (i = 0; i < nf_conntrack_htable_size; i++) { nf_conntrack_lock(&nf_conntrack_locks[i % CONNTRACK_LOCKS]); @@ -434,26 +457,6 @@ static void __nf_conntrack_helper_unregister(struct nf_conntrack_helper *me, } local_bh_enable(); } - -void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) -{ - struct net *net; - - mutex_lock(&nf_ct_helper_mutex); - hlist_del_rcu(&me->hnode); - nf_ct_helper_count--; - mutex_unlock(&nf_ct_helper_mutex); - - /* Make sure every nothing is still using the helper unless its a - * connection in the hash. - */ - synchronize_rcu(); - - rtnl_lock(); - for_each_net(net) - __nf_conntrack_helper_unregister(me, net); - rtnl_unlock(); -} EXPORT_SYMBOL_GPL(nf_conntrack_helper_unregister); static struct nf_ct_ext_type helper_extend __read_mostly = { -- cgit v0.10.2 From 8241a1e466cd56e6c10472cac9c1ad4e54bc65db Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Wed, 1 Jun 2016 01:56:33 -0400 Subject: vhost_net: stop polling socket during rx processing We don't stop rx polling socket during rx processing, this will lead unnecessary wakeups from under layer net devices (E.g sock_def_readable() form tun). Rx will be slowed down in this way. This patch avoids this by stop polling socket during rx processing. A small drawback is that this introduces some overheads in light load case because of the extra start/stop polling, but single netperf TCP_RR does not notice any change. In a super heavy load case, e.g using pktgen to inject packet to guest, we get about ~8.8% improvement on pps: before: ~1240000 pkt/s after: ~1350000 pkt/s Signed-off-by: Jason Wang Acked-by: Michael S. Tsirkin Signed-off-by: David S. Miller diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index f744eeb..1d3e45f 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -301,6 +301,32 @@ static bool vhost_can_busy_poll(struct vhost_dev *dev, !vhost_has_work(dev); } +static void vhost_net_disable_vq(struct vhost_net *n, + struct vhost_virtqueue *vq) +{ + struct vhost_net_virtqueue *nvq = + container_of(vq, struct vhost_net_virtqueue, vq); + struct vhost_poll *poll = n->poll + (nvq - n->vqs); + if (!vq->private_data) + return; + vhost_poll_stop(poll); +} + +static int vhost_net_enable_vq(struct vhost_net *n, + struct vhost_virtqueue *vq) +{ + struct vhost_net_virtqueue *nvq = + container_of(vq, struct vhost_net_virtqueue, vq); + struct vhost_poll *poll = n->poll + (nvq - n->vqs); + struct socket *sock; + + sock = vq->private_data; + if (!sock) + return 0; + + return vhost_poll_start(poll, sock->file); +} + static int vhost_net_tx_get_vq_desc(struct vhost_net *net, struct vhost_virtqueue *vq, struct iovec iov[], unsigned int iov_size, @@ -613,6 +639,7 @@ static void handle_rx(struct vhost_net *net) if (!sock) goto out; vhost_disable_notify(&net->dev, vq); + vhost_net_disable_vq(net, vq); vhost_hlen = nvq->vhost_hlen; sock_hlen = nvq->sock_hlen; @@ -629,7 +656,7 @@ static void handle_rx(struct vhost_net *net) likely(mergeable) ? UIO_MAXIOV : 1); /* On error, stop handling until the next kick. */ if (unlikely(headcount < 0)) - break; + goto out; /* On overrun, truncate and discard */ if (unlikely(headcount > UIO_MAXIOV)) { iov_iter_init(&msg.msg_iter, READ, vq->iov, 1, 1); @@ -648,7 +675,7 @@ static void handle_rx(struct vhost_net *net) } /* Nothing new? Wait for eventfd to tell us * they refilled. */ - break; + goto out; } /* We don't need to be notified again. */ iov_iter_init(&msg.msg_iter, READ, vq->iov, in, vhost_len); @@ -676,7 +703,7 @@ static void handle_rx(struct vhost_net *net) &fixup) != sizeof(hdr)) { vq_err(vq, "Unable to write vnet_hdr " "at addr %p\n", vq->iov->iov_base); - break; + goto out; } } else { /* Header came from socket; we'll need to patch @@ -692,7 +719,7 @@ static void handle_rx(struct vhost_net *net) &fixup) != sizeof num_buffers) { vq_err(vq, "Failed num_buffers write"); vhost_discard_vq_desc(vq, headcount); - break; + goto out; } vhost_add_used_and_signal_n(&net->dev, vq, vq->heads, headcount); @@ -701,9 +728,10 @@ static void handle_rx(struct vhost_net *net) total_len += vhost_len; if (unlikely(total_len >= VHOST_NET_WEIGHT)) { vhost_poll_queue(&vq->poll); - break; + goto out; } } + vhost_net_enable_vq(net, vq); out: mutex_unlock(&vq->mutex); } @@ -782,32 +810,6 @@ static int vhost_net_open(struct inode *inode, struct file *f) return 0; } -static void vhost_net_disable_vq(struct vhost_net *n, - struct vhost_virtqueue *vq) -{ - struct vhost_net_virtqueue *nvq = - container_of(vq, struct vhost_net_virtqueue, vq); - struct vhost_poll *poll = n->poll + (nvq - n->vqs); - if (!vq->private_data) - return; - vhost_poll_stop(poll); -} - -static int vhost_net_enable_vq(struct vhost_net *n, - struct vhost_virtqueue *vq) -{ - struct vhost_net_virtqueue *nvq = - container_of(vq, struct vhost_net_virtqueue, vq); - struct vhost_poll *poll = n->poll + (nvq - n->vqs); - struct socket *sock; - - sock = vq->private_data; - if (!sock) - return 0; - - return vhost_poll_start(poll, sock->file); -} - static struct socket *vhost_net_stop_vq(struct vhost_net *n, struct vhost_virtqueue *vq) { -- cgit v0.10.2 From 3bcb846ca4cf55415d3719e64bb45a124792c589 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 4 Jun 2016 20:02:28 -0700 Subject: net: get rid of spin_trylock() in net_tx_action() Note: Tom Herbert posted almost same patch 3 months back, but for different reasons. The reasons we want to get rid of this spin_trylock() are : 1) Under high qdisc pressure, the spin_trylock() has almost no chance to succeed. 2) We loop multiple times in softirq handler, eventually reaching the max retry count (10), and we schedule ksoftirqd. Since we want to adhere more strictly to ksoftirqd being waked up in the future (https://lwn.net/Articles/687617/), better avoid spurious wakeups. 3) calls to __netif_reschedule() dirty the cache line containing q->next_sched, slowing down the owner of qdisc. 4) RT kernels can not use the spin_trylock() here. With help of busylock, we get the qdisc spinlock fast enough, and the trylock trick brings only performance penalty. Depending on qdisc setup, I observed a gain of up to 19 % in qdisc performance (1016600 pps instead of 853400 pps, using prio+tbf+fq_codel) ("mpstat -I SCPU 1" is much happier now) Signed-off-by: Eric Dumazet Cc: Tom Herbert Acked-by: Tom Herbert Signed-off-by: David S. Miller diff --git a/net/core/dev.c b/net/core/dev.c index 904ff43..896b686 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2253,7 +2253,7 @@ int netif_get_num_default_rss_queues(void) } EXPORT_SYMBOL(netif_get_num_default_rss_queues); -static inline void __netif_reschedule(struct Qdisc *q) +static void __netif_reschedule(struct Qdisc *q) { struct softnet_data *sd; unsigned long flags; @@ -3898,22 +3898,14 @@ static void net_tx_action(struct softirq_action *h) head = head->next_sched; root_lock = qdisc_lock(q); - if (spin_trylock(root_lock)) { - smp_mb__before_atomic(); - clear_bit(__QDISC_STATE_SCHED, - &q->state); - qdisc_run(q); - spin_unlock(root_lock); - } else { - if (!test_bit(__QDISC_STATE_DEACTIVATED, - &q->state)) { - __netif_reschedule(q); - } else { - smp_mb__before_atomic(); - clear_bit(__QDISC_STATE_SCHED, - &q->state); - } - } + spin_lock(root_lock); + /* We need to make sure head->next_sched is read + * before clearing __QDISC_STATE_SCHED + */ + smp_mb__before_atomic(); + clear_bit(__QDISC_STATE_SCHED, &q->state); + qdisc_run(q); + spin_unlock(root_lock); } } } -- cgit v0.10.2 From 1fe614d10f45e4697a307bfff9eab66c25de0c72 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Sun, 5 Jun 2016 13:11:11 +0300 Subject: qed: Relax VF firmware requirements Current driver require an exact match between VF and PF storm firmware; Any difference would fail the VF acquire message, causing the VF probe to be aborted. While there's still dependencies between the two, the recent FW submission has relaxed the match requirement - instead of an exact match, there's now a 'fastpath' HSI major/minor scheme, where VFs and PFs that match in their major number can co-exist even if their minor is different. In order to accomadate this change some changes in the vf-start init flow had to be made, as the VF start ramrod now has to be sent only after PF learns which fastpath HSI its VF is requiring. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index eb75b82..9759a49 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -21,18 +21,18 @@ #include "qed_vf.h" /* IOV ramrods */ -static int qed_sp_vf_start(struct qed_hwfn *p_hwfn, - u32 concrete_vfid, u16 opaque_vfid) +static int qed_sp_vf_start(struct qed_hwfn *p_hwfn, struct qed_vf_info *p_vf) { struct vf_start_ramrod_data *p_ramrod = NULL; struct qed_spq_entry *p_ent = NULL; struct qed_sp_init_data init_data; int rc = -EINVAL; + u8 fp_minor; /* Get SPQ entry */ memset(&init_data, 0, sizeof(init_data)); init_data.cid = qed_spq_get_cid(p_hwfn); - init_data.opaque_fid = opaque_vfid; + init_data.opaque_fid = p_vf->opaque_fid; init_data.comp_mode = QED_SPQ_MODE_EBLOCK; rc = qed_sp_init_request(p_hwfn, &p_ent, @@ -43,12 +43,39 @@ static int qed_sp_vf_start(struct qed_hwfn *p_hwfn, p_ramrod = &p_ent->ramrod.vf_start; - p_ramrod->vf_id = GET_FIELD(concrete_vfid, PXP_CONCRETE_FID_VFID); - p_ramrod->opaque_fid = cpu_to_le16(opaque_vfid); + p_ramrod->vf_id = GET_FIELD(p_vf->concrete_fid, PXP_CONCRETE_FID_VFID); + p_ramrod->opaque_fid = cpu_to_le16(p_vf->opaque_fid); + + switch (p_hwfn->hw_info.personality) { + case QED_PCI_ETH: + p_ramrod->personality = PERSONALITY_ETH; + break; + case QED_PCI_ETH_ROCE: + p_ramrod->personality = PERSONALITY_RDMA_AND_ETH; + break; + default: + DP_NOTICE(p_hwfn, "Unknown VF personality %d\n", + p_hwfn->hw_info.personality); + return -EINVAL; + } + + fp_minor = p_vf->acquire.vfdev_info.eth_fp_hsi_minor; + if (fp_minor > ETH_HSI_VER_MINOR) { + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "VF [%d] - Requested fp hsi %02x.%02x which is slightly newer than PF's %02x.%02x; Configuring PFs version\n", + p_vf->abs_vf_id, + ETH_HSI_VER_MAJOR, + fp_minor, ETH_HSI_VER_MAJOR, ETH_HSI_VER_MINOR); + fp_minor = ETH_HSI_VER_MINOR; + } - p_ramrod->personality = PERSONALITY_ETH; p_ramrod->hsi_fp_ver.major_ver_arr[ETH_VER_KEY] = ETH_HSI_VER_MAJOR; - p_ramrod->hsi_fp_ver.minor_ver_arr[ETH_VER_KEY] = ETH_HSI_VER_MINOR; + p_ramrod->hsi_fp_ver.minor_ver_arr[ETH_VER_KEY] = fp_minor; + + DP_VERBOSE(p_hwfn, QED_MSG_IOV, + "VF[%d] - Starting using HSI %02x.%02x\n", + p_vf->abs_vf_id, ETH_HSI_VER_MAJOR, fp_minor); return qed_spq_post(p_hwfn, p_ent, NULL); } @@ -600,17 +627,6 @@ static int qed_iov_enable_vf_access(struct qed_hwfn *p_hwfn, /* unpretend */ qed_fid_pretend(p_hwfn, p_ptt, (u16) p_hwfn->hw_info.concrete_fid); - if (vf->state != VF_STOPPED) { - DP_NOTICE(p_hwfn, "VF[%02x] is already started\n", - vf->abs_vf_id); - return -EINVAL; - } - - /* Start VF */ - rc = qed_sp_vf_start(p_hwfn, vf->concrete_fid, vf->opaque_fid); - if (rc) - DP_NOTICE(p_hwfn, "Failed to start VF[%02x]\n", vf->abs_vf_id); - vf->state = VF_FREE; return rc; @@ -854,7 +870,6 @@ static int qed_iov_release_hw_for_vf(struct qed_hwfn *p_hwfn, struct qed_mcp_link_params params; struct qed_mcp_link_state link; struct qed_vf_info *vf = NULL; - int rc = 0; vf = qed_iov_get_vf_info(p_hwfn, rel_vf_id, true); if (!vf) { @@ -876,18 +891,8 @@ static int qed_iov_release_hw_for_vf(struct qed_hwfn *p_hwfn, memcpy(&caps, qed_mcp_get_link_capabilities(p_hwfn), sizeof(caps)); qed_iov_set_link(p_hwfn, rel_vf_id, ¶ms, &link, &caps); - if (vf->state != VF_STOPPED) { - /* Stopping the VF */ - rc = qed_sp_vf_stop(p_hwfn, vf->concrete_fid, vf->opaque_fid); - - if (rc != 0) { - DP_ERR(p_hwfn, "qed_sp_vf_stop returned error %d\n", - rc); - return rc; - } - - vf->state = VF_STOPPED; - } + /* Forget the VF's acquisition message */ + memset(&vf->acquire, 0, sizeof(vf->acquire)); /* disablng interrupts and resetting permission table was done during * vf-close, however, we could get here without going through vf_close @@ -1132,6 +1137,7 @@ static void qed_iov_vf_cleanup(struct qed_hwfn *p_hwfn, p_vf->vf_queues[i].rxq_active = 0; memset(&p_vf->shadow_config, 0, sizeof(p_vf->shadow_config)); + memset(&p_vf->acquire, 0, sizeof(p_vf->acquire)); qed_iov_clean_vf(p_hwfn, p_vf->relative_vf_id); } @@ -1143,25 +1149,27 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, struct pfvf_acquire_resp_tlv *resp = &mbx->reply_virt->acquire_resp; struct pf_vf_pfdev_info *pfdev_info = &resp->pfdev_info; struct vfpf_acquire_tlv *req = &mbx->req_virt->acquire; - u8 i, vfpf_status = PFVF_STATUS_SUCCESS; + u8 i, vfpf_status = PFVF_STATUS_NOT_SUPPORTED; struct pf_vf_resc *resc = &resp->resc; + int rc; + + memset(resp, 0, sizeof(*resp)); /* Validate FW compatibility */ - if (req->vfdev_info.fw_major != FW_MAJOR_VERSION || - req->vfdev_info.fw_minor != FW_MINOR_VERSION || - req->vfdev_info.fw_revision != FW_REVISION_VERSION || - req->vfdev_info.fw_engineering != FW_ENGINEERING_VERSION) { + if (req->vfdev_info.eth_fp_hsi_major != ETH_HSI_VER_MAJOR) { DP_INFO(p_hwfn, - "VF[%d] is running an incompatible driver [VF needs FW %02x:%02x:%02x:%02x but Hypervisor is using %02x:%02x:%02x:%02x]\n", + "VF[%d] needs fastpath HSI %02x.%02x, which is incompatible with loaded FW's faspath HSI %02x.%02x\n", vf->abs_vf_id, - req->vfdev_info.fw_major, - req->vfdev_info.fw_minor, - req->vfdev_info.fw_revision, - req->vfdev_info.fw_engineering, - FW_MAJOR_VERSION, - FW_MINOR_VERSION, - FW_REVISION_VERSION, FW_ENGINEERING_VERSION); - vfpf_status = PFVF_STATUS_NOT_SUPPORTED; + req->vfdev_info.eth_fp_hsi_major, + req->vfdev_info.eth_fp_hsi_minor, + ETH_HSI_VER_MAJOR, ETH_HSI_VER_MINOR); + + /* Write the PF version so that VF would know which version + * is supported. + */ + pfdev_info->major_fp_hsi = ETH_HSI_VER_MAJOR; + pfdev_info->minor_fp_hsi = ETH_HSI_VER_MINOR; + goto out; } @@ -1171,11 +1179,11 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, DP_INFO(p_hwfn, "VF[%d] is running an old driver that doesn't support 100g\n", vf->abs_vf_id); - vfpf_status = PFVF_STATUS_NOT_SUPPORTED; goto out; } - memset(resp, 0, sizeof(*resp)); + /* Store the acquire message */ + memcpy(&vf->acquire, req, sizeof(vf->acquire)); /* Fill in vf info stuff */ vf->opaque_fid = req->vfdev_info.opaque_fid; @@ -1223,6 +1231,9 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, pfdev_info->fw_minor = FW_MINOR_VERSION; pfdev_info->fw_rev = FW_REVISION_VERSION; pfdev_info->fw_eng = FW_ENGINEERING_VERSION; + pfdev_info->minor_fp_hsi = min_t(u8, + ETH_HSI_VER_MINOR, + req->vfdev_info.eth_fp_hsi_minor); pfdev_info->os_type = VFPF_ACQUIRE_OS_LINUX; qed_mcp_get_mfw_ver(p_hwfn, p_ptt, &pfdev_info->mfw_ver, NULL); @@ -1253,6 +1264,14 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, */ resc->num_mc_filters = req->resc_request.num_mc_filters; + /* Start the VF in FW */ + rc = qed_sp_vf_start(p_hwfn, vf); + if (rc) { + DP_NOTICE(p_hwfn, "Failed to start VF[%02x]\n", vf->abs_vf_id); + vfpf_status = PFVF_STATUS_FAILURE; + goto out; + } + /* Fill agreed size of bulletin board in response */ resp->bulletin_size = vf->bulletin.size; qed_iov_post_vf_bulletin(p_hwfn, vf->relative_vf_id, p_ptt); @@ -2360,11 +2379,27 @@ static void qed_iov_vf_mbx_release(struct qed_hwfn *p_hwfn, struct qed_vf_info *p_vf) { u16 length = sizeof(struct pfvf_def_resp_tlv); + u8 status = PFVF_STATUS_SUCCESS; + int rc = 0; qed_iov_vf_cleanup(p_hwfn, p_vf); + if (p_vf->state != VF_STOPPED && p_vf->state != VF_FREE) { + /* Stopping the VF */ + rc = qed_sp_vf_stop(p_hwfn, p_vf->concrete_fid, + p_vf->opaque_fid); + + if (rc) { + DP_ERR(p_hwfn, "qed_sp_vf_stop returned error %d\n", + rc); + status = PFVF_STATUS_FAILURE; + } + + p_vf->state = VF_STOPPED; + } + qed_iov_prepare_resp(p_hwfn, p_ptt, p_vf, CHANNEL_TLV_RELEASE, - length, PFVF_STATUS_SUCCESS); + length, status); } static int diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.h b/drivers/net/ethernet/qlogic/qed/qed_sriov.h index c8667c6..2bcaeb3 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.h +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.h @@ -131,6 +131,9 @@ struct qed_vf_info { struct qed_bulletin bulletin; dma_addr_t vf_bulletin; + /* PF saves a copy of the last VF acquire message */ + struct vfpf_acquire_tlv acquire; + u32 concrete_fid; u16 opaque_fid; u16 mtu; diff --git a/drivers/net/ethernet/qlogic/qed/qed_vf.c b/drivers/net/ethernet/qlogic/qed/qed_vf.c index 72e69c0..ce8aec3 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_vf.c +++ b/drivers/net/ethernet/qlogic/qed/qed_vf.c @@ -147,6 +147,8 @@ static int qed_vf_pf_acquire(struct qed_hwfn *p_hwfn) req->vfdev_info.fw_minor = FW_MINOR_VERSION; req->vfdev_info.fw_revision = FW_REVISION_VERSION; req->vfdev_info.fw_engineering = FW_ENGINEERING_VERSION; + req->vfdev_info.eth_fp_hsi_major = ETH_HSI_VER_MAJOR; + req->vfdev_info.eth_fp_hsi_minor = ETH_HSI_VER_MINOR; /* Fill capability field with any non-deprecated config we support */ req->vfdev_info.capabilities |= VFPF_ACQUIRE_CAP_100G; @@ -200,6 +202,16 @@ static int qed_vf_pf_acquire(struct qed_hwfn *p_hwfn) /* Clear response buffer */ memset(p_iov->pf2vf_reply, 0, sizeof(union pfvf_tlvs)); + } else if ((resp->hdr.status == PFVF_STATUS_NOT_SUPPORTED) && + pfdev_info->major_fp_hsi && + (pfdev_info->major_fp_hsi != ETH_HSI_VER_MAJOR)) { + DP_NOTICE(p_hwfn, + "PF uses an incompatible fastpath HSI %02x.%02x [VF requires %02x.%02x]. Please change to a VF driver using %02x.xx.\n", + pfdev_info->major_fp_hsi, + pfdev_info->minor_fp_hsi, + ETH_HSI_VER_MAJOR, + ETH_HSI_VER_MINOR, pfdev_info->major_fp_hsi); + return -EINVAL; } else { DP_ERR(p_hwfn, "PF returned error %d to VF acquisition request\n", @@ -225,6 +237,13 @@ static int qed_vf_pf_acquire(struct qed_hwfn *p_hwfn) } } + if (ETH_HSI_VER_MINOR && + (resp->pfdev_info.minor_fp_hsi < ETH_HSI_VER_MINOR)) { + DP_INFO(p_hwfn, + "PF is using older fastpath HSI; %02x.%02x is configured\n", + ETH_HSI_VER_MAJOR, resp->pfdev_info.minor_fp_hsi); + } + return 0; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_vf.h b/drivers/net/ethernet/qlogic/qed/qed_vf.h index b82fda9..b23ce58 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_vf.h +++ b/drivers/net/ethernet/qlogic/qed/qed_vf.h @@ -96,7 +96,9 @@ struct vfpf_acquire_tlv { u32 driver_version; u16 opaque_fid; /* ME register value */ u8 os_type; /* VFPF_ACQUIRE_OS_* value */ - u8 padding[5]; + u8 eth_fp_hsi_major; + u8 eth_fp_hsi_minor; + u8 padding[3]; } vfdev_info; struct vf_pf_resc_request resc_request; @@ -171,7 +173,14 @@ struct pfvf_acquire_resp_tlv { struct pfvf_stats_info stats_info; u8 port_mac[ETH_ALEN]; - u8 padding2[2]; + + /* It's possible PF had to configure an older fastpath HSI + * [in case VF is newer than PF]. This is communicated back + * to the VF. It can also be used in case of error due to + * non-matching versions to shed light in VF about failure. + */ + u8 major_fp_hsi; + u8 minor_fp_hsi; } pfdev_info; struct pf_vf_resc { -- cgit v0.10.2 From 1cf2b1a971d29d4435a5c462b91d4a22325ec07d Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Sun, 5 Jun 2016 13:11:12 +0300 Subject: qed: PF-VF resource negotiation One of the goals of the vf's first message to the PF [acquire] is to learn about the number of resources available to it [macs, vlans, etc.]. This is done via negotiation - the VF requires a set of resources, which the PF either approves or disaproves and sends a smaller set of resources as alternative. In this later case, the VF is then expected to either abort the probe or re-send the acquire message with less required resources. While this infrastructure exists since the initial submision of qed SRIOV support, it's in fact completely inoperational - PF isn't really looking into the resources the VF has asked for and is never going to reply to the VF that it lacks resources. This patch addresses this flow, fixing it and allowing the PF and VF to actually agree on a set of resources. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index 9759a49..74fc82b 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -322,6 +322,9 @@ static void qed_iov_setup_vfdb(struct qed_hwfn *p_hwfn) vf->opaque_fid = (p_hwfn->hw_info.opaque_fid & 0xff) | (vf->abs_vf_id << 8); vf->vport_id = idx + 1; + + vf->num_mac_filters = QED_ETH_VF_NUM_MAC_FILTERS; + vf->num_vlan_filters = QED_ETH_VF_NUM_VLAN_FILTERS; } } @@ -1123,8 +1126,6 @@ static void qed_iov_vf_cleanup(struct qed_hwfn *p_hwfn, p_vf->vf_bulletin = 0; p_vf->vport_instance = 0; - p_vf->num_mac_filters = 0; - p_vf->num_vlan_filters = 0; p_vf->configured_features = 0; /* If VF previously requested less resources, go back to default */ @@ -1141,6 +1142,91 @@ static void qed_iov_vf_cleanup(struct qed_hwfn *p_hwfn, qed_iov_clean_vf(p_hwfn, p_vf->relative_vf_id); } +static u8 qed_iov_vf_mbx_acquire_resc(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_vf_info *p_vf, + struct vf_pf_resc_request *p_req, + struct pf_vf_resc *p_resp) +{ + int i; + + /* Queue related information */ + p_resp->num_rxqs = p_vf->num_rxqs; + p_resp->num_txqs = p_vf->num_txqs; + p_resp->num_sbs = p_vf->num_sbs; + + for (i = 0; i < p_resp->num_sbs; i++) { + p_resp->hw_sbs[i].hw_sb_id = p_vf->igu_sbs[i]; + p_resp->hw_sbs[i].sb_qid = 0; + } + + /* These fields are filled for backward compatibility. + * Unused by modern vfs. + */ + for (i = 0; i < p_resp->num_rxqs; i++) { + qed_fw_l2_queue(p_hwfn, p_vf->vf_queues[i].fw_rx_qid, + (u16 *)&p_resp->hw_qid[i]); + p_resp->cid[i] = p_vf->vf_queues[i].fw_cid; + } + + /* Filter related information */ + p_resp->num_mac_filters = min_t(u8, p_vf->num_mac_filters, + p_req->num_mac_filters); + p_resp->num_vlan_filters = min_t(u8, p_vf->num_vlan_filters, + p_req->num_vlan_filters); + + /* This isn't really needed/enforced, but some legacy VFs might depend + * on the correct filling of this field. + */ + p_resp->num_mc_filters = QED_MAX_MC_ADDRS; + + /* Validate sufficient resources for VF */ + if (p_resp->num_rxqs < p_req->num_rxqs || + p_resp->num_txqs < p_req->num_txqs || + p_resp->num_sbs < p_req->num_sbs || + p_resp->num_mac_filters < p_req->num_mac_filters || + p_resp->num_vlan_filters < p_req->num_vlan_filters || + p_resp->num_mc_filters < p_req->num_mc_filters) { + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "VF[%d] - Insufficient resources: rxq [%02x/%02x] txq [%02x/%02x] sbs [%02x/%02x] mac [%02x/%02x] vlan [%02x/%02x] mc [%02x/%02x]\n", + p_vf->abs_vf_id, + p_req->num_rxqs, + p_resp->num_rxqs, + p_req->num_rxqs, + p_resp->num_txqs, + p_req->num_sbs, + p_resp->num_sbs, + p_req->num_mac_filters, + p_resp->num_mac_filters, + p_req->num_vlan_filters, + p_resp->num_vlan_filters, + p_req->num_mc_filters, p_resp->num_mc_filters); + return PFVF_STATUS_NO_RESOURCE; + } + + return PFVF_STATUS_SUCCESS; +} + +static void qed_iov_vf_mbx_acquire_stats(struct qed_hwfn *p_hwfn, + struct pfvf_stats_info *p_stats) +{ + p_stats->mstats.address = PXP_VF_BAR0_START_MSDM_ZONE_B + + offsetof(struct mstorm_vf_zone, + non_trigger.eth_queue_stat); + p_stats->mstats.len = sizeof(struct eth_mstorm_per_queue_stat); + p_stats->ustats.address = PXP_VF_BAR0_START_USDM_ZONE_B + + offsetof(struct ustorm_vf_zone, + non_trigger.eth_queue_stat); + p_stats->ustats.len = sizeof(struct eth_ustorm_per_queue_stat); + p_stats->pstats.address = PXP_VF_BAR0_START_PSDM_ZONE_B + + offsetof(struct pstorm_vf_zone, + non_trigger.eth_queue_stat); + p_stats->pstats.len = sizeof(struct eth_pstorm_per_queue_stat); + p_stats->tstats.address = 0; + p_stats->tstats.len = 0; +} + static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_vf_info *vf) @@ -1149,7 +1235,7 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, struct pfvf_acquire_resp_tlv *resp = &mbx->reply_virt->acquire_resp; struct pf_vf_pfdev_info *pfdev_info = &resp->pfdev_info; struct vfpf_acquire_tlv *req = &mbx->req_virt->acquire; - u8 i, vfpf_status = PFVF_STATUS_NOT_SUPPORTED; + u8 vfpf_status = PFVF_STATUS_NOT_SUPPORTED; struct pf_vf_resc *resc = &resp->resc; int rc; @@ -1185,10 +1271,7 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, /* Store the acquire message */ memcpy(&vf->acquire, req, sizeof(vf->acquire)); - /* Fill in vf info stuff */ vf->opaque_fid = req->vfdev_info.opaque_fid; - vf->num_mac_filters = 1; - vf->num_vlan_filters = QED_ETH_VF_NUM_VLAN_FILTERS; vf->vf_bulletin = req->bulletin_addr; vf->bulletin.size = (vf->bulletin.size < req->bulletin_size) ? @@ -1204,26 +1287,7 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, if (p_hwfn->cdev->num_hwfns > 1) pfdev_info->capabilities |= PFVF_ACQUIRE_CAP_100G; - pfdev_info->stats_info.mstats.address = - PXP_VF_BAR0_START_MSDM_ZONE_B + - offsetof(struct mstorm_vf_zone, non_trigger.eth_queue_stat); - pfdev_info->stats_info.mstats.len = - sizeof(struct eth_mstorm_per_queue_stat); - - pfdev_info->stats_info.ustats.address = - PXP_VF_BAR0_START_USDM_ZONE_B + - offsetof(struct ustorm_vf_zone, non_trigger.eth_queue_stat); - pfdev_info->stats_info.ustats.len = - sizeof(struct eth_ustorm_per_queue_stat); - - pfdev_info->stats_info.pstats.address = - PXP_VF_BAR0_START_PSDM_ZONE_B + - offsetof(struct pstorm_vf_zone, non_trigger.eth_queue_stat); - pfdev_info->stats_info.pstats.len = - sizeof(struct eth_pstorm_per_queue_stat); - - pfdev_info->stats_info.tstats.address = 0; - pfdev_info->stats_info.tstats.len = 0; + qed_iov_vf_mbx_acquire_stats(p_hwfn, &pfdev_info->stats_info); memcpy(pfdev_info->port_mac, p_hwfn->hw_info.hw_mac_addr, ETH_ALEN); @@ -1240,29 +1304,13 @@ static void qed_iov_vf_mbx_acquire(struct qed_hwfn *p_hwfn, pfdev_info->dev_type = p_hwfn->cdev->type; pfdev_info->chip_rev = p_hwfn->cdev->chip_rev; - resc->num_rxqs = vf->num_rxqs; - resc->num_txqs = vf->num_txqs; - resc->num_sbs = vf->num_sbs; - for (i = 0; i < resc->num_sbs; i++) { - resc->hw_sbs[i].hw_sb_id = vf->igu_sbs[i]; - resc->hw_sbs[i].sb_qid = 0; - } - - for (i = 0; i < resc->num_rxqs; i++) { - qed_fw_l2_queue(p_hwfn, vf->vf_queues[i].fw_rx_qid, - (u16 *)&resc->hw_qid[i]); - resc->cid[i] = vf->vf_queues[i].fw_cid; - } - - resc->num_mac_filters = min_t(u8, vf->num_mac_filters, - req->resc_request.num_mac_filters); - resc->num_vlan_filters = min_t(u8, vf->num_vlan_filters, - req->resc_request.num_vlan_filters); - - /* This isn't really required as VF isn't limited, but some VFs might - * actually test this value, so need to provide it. + /* Fill resources available to VF; Make sure there are enough to + * satisfy the VF's request. */ - resc->num_mc_filters = req->resc_request.num_mc_filters; + vfpf_status = qed_iov_vf_mbx_acquire_resc(p_hwfn, p_ptt, vf, + &req->resc_request, resc); + if (vfpf_status != PFVF_STATUS_SUCCESS) + goto out; /* Start the VF in FW */ rc = qed_sp_vf_start(p_hwfn, vf); diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.h b/drivers/net/ethernet/qlogic/qed/qed_sriov.h index 2bcaeb3..ea24795 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.h +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.h @@ -10,6 +10,9 @@ #define _QED_SRIOV_H #include #include "qed_vf.h" + +#define QED_ETH_VF_NUM_MAC_FILTERS 1 +#define QED_ETH_VF_NUM_VLAN_FILTERS 2 #define QED_VF_ARRAY_LENGTH (3) #define IS_VF(cdev) ((cdev)->b_is_vf) @@ -22,7 +25,6 @@ #define IS_PF_SRIOV_ALLOC(p_hwfn) (!!((p_hwfn)->pf_iov_info)) #define QED_MAX_VF_CHAINS_PER_PF 16 -#define QED_ETH_VF_NUM_VLAN_FILTERS 2 #define QED_ETH_MAX_VF_NUM_VLAN_FILTERS \ (MAX_NUM_VFS * QED_ETH_VF_NUM_VLAN_FILTERS) diff --git a/drivers/net/ethernet/qlogic/qed/qed_vf.c b/drivers/net/ethernet/qlogic/qed/qed_vf.c index ce8aec3..46751ce 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_vf.c +++ b/drivers/net/ethernet/qlogic/qed/qed_vf.c @@ -117,30 +117,56 @@ exit: } #define VF_ACQUIRE_THRESH 3 -#define VF_ACQUIRE_MAC_FILTERS 1 +static void qed_vf_pf_acquire_reduce_resc(struct qed_hwfn *p_hwfn, + struct vf_pf_resc_request *p_req, + struct pf_vf_resc *p_resp) +{ + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "PF unwilling to fullill resource request: rxq [%02x/%02x] txq [%02x/%02x] sbs [%02x/%02x] mac [%02x/%02x] vlan [%02x/%02x] mc [%02x/%02x]. Try PF recommended amount\n", + p_req->num_rxqs, + p_resp->num_rxqs, + p_req->num_rxqs, + p_resp->num_txqs, + p_req->num_sbs, + p_resp->num_sbs, + p_req->num_mac_filters, + p_resp->num_mac_filters, + p_req->num_vlan_filters, + p_resp->num_vlan_filters, + p_req->num_mc_filters, p_resp->num_mc_filters); + + /* humble our request */ + p_req->num_txqs = p_resp->num_txqs; + p_req->num_rxqs = p_resp->num_rxqs; + p_req->num_sbs = p_resp->num_sbs; + p_req->num_mac_filters = p_resp->num_mac_filters; + p_req->num_vlan_filters = p_resp->num_vlan_filters; + p_req->num_mc_filters = p_resp->num_mc_filters; +} static int qed_vf_pf_acquire(struct qed_hwfn *p_hwfn) { struct qed_vf_iov *p_iov = p_hwfn->vf_iov_info; struct pfvf_acquire_resp_tlv *resp = &p_iov->pf2vf_reply->acquire_resp; struct pf_vf_pfdev_info *pfdev_info = &resp->pfdev_info; - u8 rx_count = 1, tx_count = 1, num_sbs = 1; - u8 num_mac = VF_ACQUIRE_MAC_FILTERS; + struct vf_pf_resc_request *p_resc; bool resources_acquired = false; struct vfpf_acquire_tlv *req; int rc = 0, attempts = 0; /* clear mailbox and prep first tlv */ req = qed_vf_pf_prep(p_hwfn, CHANNEL_TLV_ACQUIRE, sizeof(*req)); + p_resc = &req->resc_request; /* starting filling the request */ req->vfdev_info.opaque_fid = p_hwfn->hw_info.opaque_fid; - req->resc_request.num_rxqs = rx_count; - req->resc_request.num_txqs = tx_count; - req->resc_request.num_sbs = num_sbs; - req->resc_request.num_mac_filters = num_mac; - req->resc_request.num_vlan_filters = QED_ETH_VF_NUM_VLAN_FILTERS; + p_resc->num_rxqs = QED_MAX_VF_CHAINS_PER_PF; + p_resc->num_txqs = QED_MAX_VF_CHAINS_PER_PF; + p_resc->num_sbs = QED_MAX_VF_CHAINS_PER_PF; + p_resc->num_mac_filters = QED_ETH_VF_NUM_MAC_FILTERS; + p_resc->num_vlan_filters = QED_ETH_VF_NUM_VLAN_FILTERS; req->vfdev_info.os_type = VFPF_ACQUIRE_OS_LINUX; req->vfdev_info.fw_major = FW_MAJOR_VERSION; @@ -187,18 +213,8 @@ static int qed_vf_pf_acquire(struct qed_hwfn *p_hwfn) resources_acquired = true; } else if (resp->hdr.status == PFVF_STATUS_NO_RESOURCE && attempts < VF_ACQUIRE_THRESH) { - DP_VERBOSE(p_hwfn, - QED_MSG_IOV, - "PF unwilling to fullfill resource request. Try PF recommended amount\n"); - - /* humble our request */ - req->resc_request.num_txqs = resp->resc.num_txqs; - req->resc_request.num_rxqs = resp->resc.num_rxqs; - req->resc_request.num_sbs = resp->resc.num_sbs; - req->resc_request.num_mac_filters = - resp->resc.num_mac_filters; - req->resc_request.num_vlan_filters = - resp->resc.num_vlan_filters; + qed_vf_pf_acquire_reduce_resc(p_hwfn, p_resc, + &resp->resc); /* Clear response buffer */ memset(p_iov->pf2vf_reply, 0, sizeof(union pfvf_tlvs)); -- cgit v0.10.2 From 41086467d6647e66325f4c95eab84faa307f81ee Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Sun, 5 Jun 2016 13:11:13 +0300 Subject: qed: Make PF more robust against malicious VF There are several requests the VF can make toward the PF which the driver would pass to firmware without checking the validity first - specifically, opening queues and updating vports. Such configurations might cause the firmware to assert. This adds validation of the legality of said configurations on the PF side before passing it onward via ramrod to firmware. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index 74fc82b..dce81f5 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -146,6 +146,45 @@ static struct qed_vf_info *qed_iov_get_vf_info(struct qed_hwfn *p_hwfn, return vf; } +static bool qed_iov_validate_rxq(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, u16 rx_qid) +{ + if (rx_qid >= p_vf->num_rxqs) + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "VF[0x%02x] - can't touch Rx queue[%04x]; Only 0x%04x are allocated\n", + p_vf->abs_vf_id, rx_qid, p_vf->num_rxqs); + return rx_qid < p_vf->num_rxqs; +} + +static bool qed_iov_validate_txq(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, u16 tx_qid) +{ + if (tx_qid >= p_vf->num_txqs) + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "VF[0x%02x] - can't touch Tx queue[%04x]; Only 0x%04x are allocated\n", + p_vf->abs_vf_id, tx_qid, p_vf->num_txqs); + return tx_qid < p_vf->num_txqs; +} + +static bool qed_iov_validate_sb(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, u16 sb_idx) +{ + int i; + + for (i = 0; i < p_vf->num_sbs; i++) + if (p_vf->igu_sbs[i] == sb_idx) + return true; + + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "VF[0%02x] - tried using sb_idx %04x which doesn't exist as one of its 0x%02x SBs\n", + p_vf->abs_vf_id, sb_idx, p_vf->num_sbs); + + return false; +} + int qed_iov_post_vf_bulletin(struct qed_hwfn *p_hwfn, int vfid, struct qed_ptt *p_ptt) { @@ -1687,12 +1726,17 @@ static void qed_iov_vf_mbx_start_rxq(struct qed_hwfn *p_hwfn, { struct qed_queue_start_common_params params; struct qed_iov_vf_mbx *mbx = &vf->vf_mbx; - u8 status = PFVF_STATUS_SUCCESS; + u8 status = PFVF_STATUS_NO_RESOURCE; struct vfpf_start_rxq_tlv *req; int rc; memset(¶ms, 0, sizeof(params)); req = &mbx->req_virt->start_rxq; + + if (!qed_iov_validate_rxq(p_hwfn, vf, req->rx_qid) || + !qed_iov_validate_sb(p_hwfn, vf, req->hw_sb)) + goto out; + params.queue_id = vf->vf_queues[req->rx_qid].fw_rx_qid; params.vf_qid = req->rx_qid; params.vport_id = vf->vport_id; @@ -1710,10 +1754,12 @@ static void qed_iov_vf_mbx_start_rxq(struct qed_hwfn *p_hwfn, if (rc) { status = PFVF_STATUS_FAILURE; } else { + status = PFVF_STATUS_SUCCESS; vf->vf_queues[req->rx_qid].rxq_active = true; vf->num_active_rxqs++; } +out: qed_iov_vf_mbx_start_rxq_resp(p_hwfn, p_ptt, vf, status); } @@ -1724,8 +1770,8 @@ static void qed_iov_vf_mbx_start_txq(struct qed_hwfn *p_hwfn, u16 length = sizeof(struct pfvf_def_resp_tlv); struct qed_queue_start_common_params params; struct qed_iov_vf_mbx *mbx = &vf->vf_mbx; + u8 status = PFVF_STATUS_NO_RESOURCE; union qed_qm_pq_params pq_params; - u8 status = PFVF_STATUS_SUCCESS; struct vfpf_start_txq_tlv *req; int rc; @@ -1736,6 +1782,11 @@ static void qed_iov_vf_mbx_start_txq(struct qed_hwfn *p_hwfn, memset(¶ms, 0, sizeof(params)); req = &mbx->req_virt->start_txq; + + if (!qed_iov_validate_txq(p_hwfn, vf, req->tx_qid) || + !qed_iov_validate_sb(p_hwfn, vf, req->hw_sb)) + goto out; + params.queue_id = vf->vf_queues[req->tx_qid].fw_tx_qid; params.vport_id = vf->vport_id; params.sb = req->hw_sb; @@ -1749,11 +1800,14 @@ static void qed_iov_vf_mbx_start_txq(struct qed_hwfn *p_hwfn, req->pbl_addr, req->pbl_size, &pq_params); - if (rc) + if (rc) { status = PFVF_STATUS_FAILURE; - else + } else { + status = PFVF_STATUS_SUCCESS; vf->vf_queues[req->tx_qid].txq_active = true; + } +out: qed_iov_prepare_resp(p_hwfn, p_ptt, vf, CHANNEL_TLV_START_TXQ, length, status); } @@ -2180,6 +2234,16 @@ static void qed_iov_vf_mbx_vport_update(struct qed_hwfn *p_hwfn, u16 length; int rc; + /* Valiate PF can send such a request */ + if (!vf->vport_instance) { + DP_VERBOSE(p_hwfn, + QED_MSG_IOV, + "No VPORT instance available for VF[%d], failing vport update\n", + vf->abs_vf_id); + status = PFVF_STATUS_FAILURE; + goto out; + } + memset(¶ms, 0, sizeof(params)); params.opaque_fid = vf->opaque_fid; params.vport_id = vf->vport_id; -- cgit v0.10.2 From 5040acf537ff716f1c8fe42a3814cd0c06ec44b9 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Sun, 5 Jun 2016 13:11:14 +0300 Subject: qed: Move doorbell calculation from VF to PF Today, the VF is aware of its queues context-ids, and calculates the doorbell address when opening its queues on its own. The configuration of doorbells in HW can sometime in the future be changed by the PF [hw has several configurable features that might affect doorbell addresses, e.g., dpm support], this would break compatibility with older VFs as their calculated doorbell addresses would be incorrect for such a configuration. In order to avoid such a backward compatibility failure, let the PF make the calculation of the doorbell offset based on the context-id, and pass that to the VF. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index dce81f5..0ecd522 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -1763,11 +1763,35 @@ out: qed_iov_vf_mbx_start_rxq_resp(p_hwfn, p_ptt, vf, status); } +static void qed_iov_vf_mbx_start_txq_resp(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_vf_info *p_vf, u8 status) +{ + struct qed_iov_vf_mbx *mbx = &p_vf->vf_mbx; + struct pfvf_start_queue_resp_tlv *p_tlv; + + mbx->offset = (u8 *)mbx->reply_virt; + + p_tlv = qed_add_tlv(p_hwfn, &mbx->offset, CHANNEL_TLV_START_TXQ, + sizeof(*p_tlv)); + qed_add_tlv(p_hwfn, &mbx->offset, CHANNEL_TLV_LIST_END, + sizeof(struct channel_list_end_tlv)); + + /* Update the TLV with the response */ + if (status == PFVF_STATUS_SUCCESS) { + u16 qid = mbx->req_virt->start_txq.tx_qid; + + p_tlv->offset = qed_db_addr(p_vf->vf_queues[qid].fw_cid, + DQ_DEMS_LEGACY); + } + + qed_iov_send_response(p_hwfn, p_ptt, p_vf, sizeof(*p_tlv), status); +} + static void qed_iov_vf_mbx_start_txq(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_vf_info *vf) { - u16 length = sizeof(struct pfvf_def_resp_tlv); struct qed_queue_start_common_params params; struct qed_iov_vf_mbx *mbx = &vf->vf_mbx; u8 status = PFVF_STATUS_NO_RESOURCE; @@ -1808,8 +1832,7 @@ static void qed_iov_vf_mbx_start_txq(struct qed_hwfn *p_hwfn, } out: - qed_iov_prepare_resp(p_hwfn, p_ptt, vf, CHANNEL_TLV_START_TXQ, - length, status); + qed_iov_vf_mbx_start_txq_resp(p_hwfn, p_ptt, vf, status); } static int qed_iov_vf_stop_rxqs(struct qed_hwfn *p_hwfn, diff --git a/drivers/net/ethernet/qlogic/qed/qed_vf.c b/drivers/net/ethernet/qlogic/qed/qed_vf.c index 46751ce..9819230 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_vf.c +++ b/drivers/net/ethernet/qlogic/qed/qed_vf.c @@ -440,8 +440,8 @@ int qed_vf_pf_txq_start(struct qed_hwfn *p_hwfn, u16 pbl_size, void __iomem **pp_doorbell) { struct qed_vf_iov *p_iov = p_hwfn->vf_iov_info; + struct pfvf_start_queue_resp_tlv *resp; struct vfpf_start_txq_tlv *req; - struct pfvf_def_resp_tlv *resp; int rc; /* clear mailbox and prep first tlv */ @@ -459,20 +459,24 @@ int qed_vf_pf_txq_start(struct qed_hwfn *p_hwfn, qed_add_tlv(p_hwfn, &p_iov->offset, CHANNEL_TLV_LIST_END, sizeof(struct channel_list_end_tlv)); - resp = &p_iov->pf2vf_reply->default_resp; + resp = &p_iov->pf2vf_reply->queue_start; rc = qed_send_msg2pf(p_hwfn, &resp->hdr.status, sizeof(*resp)); if (rc) - return rc; + goto exit; - if (resp->hdr.status != PFVF_STATUS_SUCCESS) - return -EINVAL; + if (resp->hdr.status != PFVF_STATUS_SUCCESS) { + rc = -EINVAL; + goto exit; + } if (pp_doorbell) { - u8 cid = p_iov->acquire_resp.resc.cid[tx_queue_id]; + *pp_doorbell = (u8 __iomem *)p_hwfn->doorbells + resp->offset; - *pp_doorbell = (u8 __iomem *)p_hwfn->doorbells + - qed_db_addr(cid, DQ_DEMS_LEGACY); + DP_VERBOSE(p_hwfn, QED_MSG_IOV, + "Txq[0x%02x]: doorbell at %p [offset 0x%08x]\n", + tx_queue_id, *pp_doorbell, resp->offset); } +exit: return rc; } -- cgit v0.10.2 From 8246d0b48bc27f43988e62ed26fac22e74b1d8e8 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Sun, 5 Jun 2016 13:11:15 +0300 Subject: qed: PF enforce MAC limitation of VFs The only limitation relating to MACs the PF enforce today on its VFs is in case it has a forced-unicast MAC address for them, in which case they can't configure other unicast addresses. Specifically, the PF isn't enforcing the number of MAC addresse a VF can configure regardless of the nubmer of such filters agreed upon by PF and VF during the acquisition process. PF's shadow-config is now extended to also contain information about its VFs' unicast addresses configuration, allowing such enforcement. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index 0ecd522..c204373 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -2309,15 +2309,12 @@ out: qed_iov_send_response(p_hwfn, p_ptt, vf, length, status); } -static int qed_iov_vf_update_unicast_shadow(struct qed_hwfn *p_hwfn, - struct qed_vf_info *p_vf, - struct qed_filter_ucast *p_params) +static int qed_iov_vf_update_vlan_shadow(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, + struct qed_filter_ucast *p_params) { int i; - if (p_params->type == QED_FILTER_MAC) - return 0; - /* First remove entries and then add new ones */ if (p_params->opcode == QED_FILTER_REMOVE) { for (i = 0; i < QED_ETH_VF_NUM_VLAN_FILTERS + 1; i++) @@ -2370,6 +2367,80 @@ static int qed_iov_vf_update_unicast_shadow(struct qed_hwfn *p_hwfn, return 0; } +static int qed_iov_vf_update_mac_shadow(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, + struct qed_filter_ucast *p_params) +{ + int i; + + /* If we're in forced-mode, we don't allow any change */ + if (p_vf->bulletin.p_virt->valid_bitmap & (1 << MAC_ADDR_FORCED)) + return 0; + + /* First remove entries and then add new ones */ + if (p_params->opcode == QED_FILTER_REMOVE) { + for (i = 0; i < QED_ETH_VF_NUM_MAC_FILTERS; i++) { + if (ether_addr_equal(p_vf->shadow_config.macs[i], + p_params->mac)) { + memset(p_vf->shadow_config.macs[i], 0, + ETH_ALEN); + break; + } + } + + if (i == QED_ETH_VF_NUM_MAC_FILTERS) { + DP_VERBOSE(p_hwfn, QED_MSG_IOV, + "MAC isn't configured\n"); + return -EINVAL; + } + } else if (p_params->opcode == QED_FILTER_REPLACE || + p_params->opcode == QED_FILTER_FLUSH) { + for (i = 0; i < QED_ETH_VF_NUM_MAC_FILTERS; i++) + memset(p_vf->shadow_config.macs[i], 0, ETH_ALEN); + } + + /* List the new MAC address */ + if (p_params->opcode != QED_FILTER_ADD && + p_params->opcode != QED_FILTER_REPLACE) + return 0; + + for (i = 0; i < QED_ETH_VF_NUM_MAC_FILTERS; i++) { + if (is_zero_ether_addr(p_vf->shadow_config.macs[i])) { + ether_addr_copy(p_vf->shadow_config.macs[i], + p_params->mac); + DP_VERBOSE(p_hwfn, QED_MSG_IOV, + "Added MAC at %d entry in shadow\n", i); + break; + } + } + + if (i == QED_ETH_VF_NUM_MAC_FILTERS) { + DP_VERBOSE(p_hwfn, QED_MSG_IOV, "No available place for MAC\n"); + return -EINVAL; + } + + return 0; +} + +static int +qed_iov_vf_update_unicast_shadow(struct qed_hwfn *p_hwfn, + struct qed_vf_info *p_vf, + struct qed_filter_ucast *p_params) +{ + int rc = 0; + + if (p_params->type == QED_FILTER_MAC) { + rc = qed_iov_vf_update_mac_shadow(p_hwfn, p_vf, p_params); + if (rc) + return rc; + } + + if (p_params->type == QED_FILTER_VLAN) + rc = qed_iov_vf_update_vlan_shadow(p_hwfn, p_vf, p_params); + + return rc; +} + int qed_iov_chk_ucast(struct qed_hwfn *hwfn, int vfid, struct qed_filter_ucast *params) { diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.h b/drivers/net/ethernet/qlogic/qed/qed_sriov.h index ea24795..96a7273 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.h +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.h @@ -120,6 +120,8 @@ struct qed_vf_shadow_config { /* Shadow copy of all guest vlans */ struct qed_vf_vlan_shadow vlans[QED_ETH_VF_NUM_VLAN_FILTERS + 1]; + /* Shadow copy of all configured MACs; Empty if forcing MACs */ + u8 macs[QED_ETH_VF_NUM_MAC_FILTERS][ETH_ALEN]; u8 inner_vlan_removal; }; -- cgit v0.10.2 From 54fdd80f6fe430381e6eeeef8894eb72b657d649 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Sun, 5 Jun 2016 13:11:16 +0300 Subject: qed: PF to reply to unknown messages If a future VF would send the PF an unknown message, the PF today would not send a reply. This would have 2 bad effects: a. VF would have to timeout on the request. b. If VF were to send an additional message to PF, firmware would mark it as malicious. Instead, if there's some valid reply-address on the message - let the PF answer and tell the VF it doesn't know the message. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index c204373..4d161c75 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -2857,7 +2857,6 @@ static void qed_iov_process_mbx_req(struct qed_hwfn *p_hwfn, { struct qed_iov_vf_mbx *mbx; struct qed_vf_info *p_vf; - int i; p_vf = qed_iov_get_vf_info(p_hwfn, (u16) vfid, true); if (!p_vf) @@ -2866,9 +2865,8 @@ static void qed_iov_process_mbx_req(struct qed_hwfn *p_hwfn, mbx = &p_vf->vf_mbx; /* qed_iov_process_mbx_request */ - DP_VERBOSE(p_hwfn, - QED_MSG_IOV, - "qed_iov_process_mbx_req vfid %d\n", p_vf->abs_vf_id); + DP_VERBOSE(p_hwfn, QED_MSG_IOV, + "VF[%02x]: Processing mailbox message\n", p_vf->abs_vf_id); mbx->first_tlv = mbx->req_virt->first_tlv; @@ -2922,15 +2920,28 @@ static void qed_iov_process_mbx_req(struct qed_hwfn *p_hwfn, * support them. Or this may be because someone wrote a crappy * VF driver and is sending garbage over the channel. */ - DP_ERR(p_hwfn, - "unknown TLV. type %d length %d. first 20 bytes of mailbox buffer:\n", - mbx->first_tlv.tl.type, mbx->first_tlv.tl.length); - - for (i = 0; i < 20; i++) { + DP_NOTICE(p_hwfn, + "VF[%02x]: unknown TLV. type %04x length %04x padding %08x reply address %llu\n", + p_vf->abs_vf_id, + mbx->first_tlv.tl.type, + mbx->first_tlv.tl.length, + mbx->first_tlv.padding, mbx->first_tlv.reply_address); + + /* Try replying in case reply address matches the acquisition's + * posted address. + */ + if (p_vf->acquire.first_tlv.reply_address && + (mbx->first_tlv.reply_address == + p_vf->acquire.first_tlv.reply_address)) { + qed_iov_prepare_resp(p_hwfn, p_ptt, p_vf, + mbx->first_tlv.tl.type, + sizeof(struct pfvf_def_resp_tlv), + PFVF_STATUS_NOT_SUPPORTED); + } else { DP_VERBOSE(p_hwfn, QED_MSG_IOV, - "%x ", - mbx->req_virt->tlv_buf_size.tlv_buffer[i]); + "VF[%02x]: Can't respond to TLV - no valid reply address\n", + p_vf->abs_vf_id); } } } -- cgit v0.10.2 From e69985c67c33f1d981a87986237366e83a8f0e13 Mon Sep 17 00:00:00 2001 From: Amir Vadai Date: Sun, 5 Jun 2016 17:11:18 +0300 Subject: net/sched: cls_flower: Introduce support in SKIP SW flag In order to make a filter processed only by hardware, skip_sw flag should be supplied. This is an addition to the already existing skip_hw flag (filter will be processed by software only). If no flag is specified, filter will be processed by both software and hardware. If only hardware offloaded filters exist, fl_classify() will return without doing anything. A following userspace patch will be sent once kernel patch is accepted. Example: tc filter add dev enp0s9 protocol ip prio 20 parent ffff: \ flower \ ip_proto 6 \ indev enp0s9 \ skip_sw \ action skbedit mark 0x1234 Signed-off-by: Amir Vadai Acked-by: Jiri Pirko Acked-by: John Fastabend Signed-off-by: David S. Miller diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 730aaca..d737492 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -66,6 +66,7 @@ struct cls_fl_filter { struct fl_flow_key key; struct list_head list; u32 handle; + u32 flags; struct rcu_head rcu; }; @@ -123,6 +124,9 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct fl_flow_key skb_key; struct fl_flow_key skb_mkey; + if (!atomic_read(&head->ht.nelems)) + return -1; + fl_clear_masked_range(&skb_key, &head->mask); skb_key.indev_ifindex = skb->skb_iif; /* skb_flow_dissect() does not set n_proto in case an unknown protocol, @@ -136,7 +140,7 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp, f = rhashtable_lookup_fast(&head->ht, fl_key_get_start(&skb_mkey, &head->mask), head->ht_params); - if (f) { + if (f && !(f->flags & TCA_CLS_FLAGS_SKIP_SW)) { *res = f->res; return tcf_exts_exec(skb, &f->exts, res); } @@ -524,7 +528,6 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, struct cls_fl_filter *fnew; struct nlattr *tb[TCA_FLOWER_MAX + 1]; struct fl_flow_mask mask = {}; - u32 flags = 0; int err; if (!tca[TCA_OPTIONS]) @@ -552,8 +555,14 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, } fnew->handle = handle; - if (tb[TCA_FLOWER_FLAGS]) - flags = nla_get_u32(tb[TCA_FLOWER_FLAGS]); + if (tb[TCA_FLOWER_FLAGS]) { + fnew->flags = nla_get_u32(tb[TCA_FLOWER_FLAGS]); + + if (!tc_flags_valid(fnew->flags)) { + err = -EINVAL; + goto errout; + } + } err = fl_set_parms(net, tp, fnew, &mask, base, tb, tca[TCA_RATE], ovr); if (err) @@ -563,10 +572,12 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, if (err) goto errout; - err = rhashtable_insert_fast(&head->ht, &fnew->ht_node, - head->ht_params); - if (err) - goto errout; + if (!(fnew->flags & TCA_CLS_FLAGS_SKIP_SW)) { + err = rhashtable_insert_fast(&head->ht, &fnew->ht_node, + head->ht_params); + if (err) + goto errout; + } fl_hw_replace_filter(tp, &head->dissector, @@ -574,7 +585,7 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, &fnew->key, &fnew->exts, (unsigned long)fnew, - flags); + fnew->flags); if (fold) { rhashtable_remove_fast(&head->ht, &fold->ht_node, @@ -734,6 +745,8 @@ static int fl_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, sizeof(key->tp.dst)))) goto nla_put_failure; + nla_put_u32(skb, TCA_FLOWER_FLAGS, f->flags); + if (tcf_exts_dump(skb, &f->exts)) goto nla_put_failure; -- cgit v0.10.2 From 9c4a4e488bc8f55dfc8782c7d7757fb058e9088e Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 6 Jun 2016 06:32:53 -0400 Subject: net sched: actions use tcf_lastuse_update for consistency Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c index 2ba700c..e0e6c68 100644 --- a/net/sched/act_connmark.c +++ b/net/sched/act_connmark.c @@ -44,7 +44,7 @@ static int tcf_connmark(struct sk_buff *skb, const struct tc_action *a, int proto; spin_lock(&ca->tcf_lock); - ca->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&ca->tcf_tm); bstats_update(&ca->tcf_bstats, skb); if (skb->protocol == htons(ETH_P_IP)) { diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c index 28e934e..065f716 100644 --- a/net/sched/act_csum.c +++ b/net/sched/act_csum.c @@ -501,7 +501,7 @@ static int tcf_csum(struct sk_buff *skb, u32 update_flags; spin_lock(&p->tcf_lock); - p->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&p->tcf_tm); bstats_update(&p->tcf_bstats, skb); action = p->tcf_action; update_flags = p->update_flags; diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c index 658046d..6491576 100644 --- a/net/sched/act_ife.c +++ b/net/sched/act_ife.c @@ -623,7 +623,7 @@ static int tcf_ife_decode(struct sk_buff *skb, const struct tc_action *a, spin_lock(&ife->tcf_lock); bstats_update(&ife->tcf_bstats, skb); - ife->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&ife->tcf_tm); spin_unlock(&ife->tcf_lock); ifehdrln = ntohs(ifehdrln); @@ -711,7 +711,7 @@ static int tcf_ife_encode(struct sk_buff *skb, const struct tc_action *a, spin_lock(&ife->tcf_lock); bstats_update(&ife->tcf_bstats, skb); - ife->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&ife->tcf_tm); if (!metalen) { /* no metadata to send */ /* abuse overlimits to count when we allow packet @@ -802,7 +802,7 @@ static int tcf_ife_act(struct sk_buff *skb, const struct tc_action *a, pr_info_ratelimited("unknown failure(policy neither de/encode\n"); spin_lock(&ife->tcf_lock); bstats_update(&ife->tcf_bstats, skb); - ife->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&ife->tcf_tm); ife->tcf_qstats.drops++; spin_unlock(&ife->tcf_lock); diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index 9f002ad..30e9087 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -212,7 +212,7 @@ static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a, spin_lock(&ipt->tcf_lock); - ipt->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&ipt->tcf_tm); bstats_update(&ipt->tcf_bstats, skb); /* yes, we have to worry about both in and out dev diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 128942b..d3ac73e 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -157,7 +157,6 @@ static int tcf_mirred(struct sk_buff *skb, const struct tc_action *a, u32 at; tcf_lastuse_update(&m->tcf_tm); - bstats_cpu_update(this_cpu_ptr(m->common.cpu_bstats), skb); rcu_read_lock(); diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c index c0a879f..9135aa8 100644 --- a/net/sched/act_nat.c +++ b/net/sched/act_nat.c @@ -103,7 +103,7 @@ static int tcf_nat(struct sk_buff *skb, const struct tc_action *a, spin_lock(&p->tcf_lock); - p->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&p->tcf_tm); old_addr = p->old_addr; new_addr = p->new_addr; mask = p->mask; diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index c6e18f2..67a1726 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -121,7 +121,7 @@ static int tcf_pedit(struct sk_buff *skb, const struct tc_action *a, spin_lock(&p->tcf_lock); - p->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&p->tcf_tm); if (p->tcfp_nkeys > 0) { struct tc_pedit_key *tkey = p->tcfp_keys; diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c index e42f8da..f95d1c5 100644 --- a/net/sched/act_simple.c +++ b/net/sched/act_simple.c @@ -35,7 +35,7 @@ static int tcf_simp(struct sk_buff *skb, const struct tc_action *a, struct tcf_defact *d = a->priv; spin_lock(&d->tcf_lock); - d->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&d->tcf_tm); bstats_update(&d->tcf_bstats, skb); /* print policy string followed by _ then packet count diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index e928802..82105c8 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -37,7 +37,7 @@ static int tcf_skbedit(struct sk_buff *skb, const struct tc_action *a, struct tcf_skbedit *d = a->priv; spin_lock(&d->tcf_lock); - d->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&d->tcf_tm); bstats_update(&d->tcf_bstats, skb); if (d->flags & SKBEDIT_F_PRIORITY) diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index ac4adc8..da5120f 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -31,7 +31,7 @@ static int tcf_vlan(struct sk_buff *skb, const struct tc_action *a, int err; spin_lock(&v->tcf_lock); - v->tcf_tm.lastuse = jiffies; + tcf_lastuse_update(&v->tcf_tm); bstats_update(&v->tcf_bstats, skb); action = v->tcf_action; -- cgit v0.10.2 From 53eb440f4ada034ea43b295891feec3df0fa7a29 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 6 Jun 2016 06:32:54 -0400 Subject: net sched actions: introduce timestamp for firsttime use Useful to know when the action was first used for accounting (and debugging) Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller diff --git a/include/net/act_api.h b/include/net/act_api.h index 9a9a8ed..8389c00 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -76,6 +76,8 @@ static inline void tcf_lastuse_update(struct tcf_t *tm) if (tm->lastuse != now) tm->lastuse = now; + if (unlikely(!tm->firstuse)) + tm->firstuse = now; } struct tc_action { diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h index f4297c8..9ba1410 100644 --- a/include/uapi/linux/pkt_cls.h +++ b/include/uapi/linux/pkt_cls.h @@ -124,6 +124,7 @@ struct tcf_t { __u64 install; __u64 lastuse; __u64 expires; + __u64 firstuse; }; struct tc_cnt { diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 336774a..5ebf6d6 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -283,6 +283,7 @@ err2: p->tcfc_index = index ? index : tcf_hash_new_index(tn); p->tcfc_tm.install = jiffies; p->tcfc_tm.lastuse = jiffies; + p->tcfc_tm.firstuse = 0; if (est) { err = gen_new_estimator(&p->tcfc_bstats, p->cpu_bstats, &p->tcfc_rate_est, diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index c7123e0..e4b877f 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -156,6 +156,7 @@ static int tcf_bpf_dump(struct sk_buff *skb, struct tc_action *act, tm.install = jiffies_to_clock_t(jiffies - prog->tcf_tm.install); tm.lastuse = jiffies_to_clock_t(jiffies - prog->tcf_tm.lastuse); + tm.firstuse = jiffies_to_clock_t(jiffies - prog->tcf_tm.firstuse); tm.expires = jiffies_to_clock_t(prog->tcf_tm.expires); if (nla_put_64bit(skb, TCA_ACT_BPF_TM, sizeof(tm), &tm, diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c index e0e6c68..e3f64f2 100644 --- a/net/sched/act_connmark.c +++ b/net/sched/act_connmark.c @@ -163,6 +163,7 @@ static inline int tcf_connmark_dump(struct sk_buff *skb, struct tc_action *a, t.install = jiffies_to_clock_t(jiffies - ci->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - ci->tcf_tm.lastuse); t.expires = jiffies_to_clock_t(ci->tcf_tm.expires); + t.firstuse = jiffies_to_clock_t(jiffies - ci->tcf_tm.firstuse); if (nla_put_64bit(skb, TCA_CONNMARK_TM, sizeof(t), &t, TCA_CONNMARK_PAD)) goto nla_put_failure; diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c index 065f716..7725eafb 100644 --- a/net/sched/act_csum.c +++ b/net/sched/act_csum.c @@ -548,6 +548,7 @@ static int tcf_csum_dump(struct sk_buff *skb, goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - p->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(p->tcf_tm.expires); if (nla_put_64bit(skb, TCA_CSUM_TM, sizeof(t), &t, TCA_CSUM_PAD)) goto nla_put_failure; diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c index ec5cc84..c9d59f3 100644 --- a/net/sched/act_gact.c +++ b/net/sched/act_gact.c @@ -190,6 +190,7 @@ static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int #endif t.install = jiffies_to_clock_t(jiffies - gact->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - gact->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - gact->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(gact->tcf_tm.expires); if (nla_put_64bit(skb, TCA_GACT_TM, sizeof(t), &t, TCA_GACT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index 30e9087..47525ee 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -279,6 +279,7 @@ static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int goto nla_put_failure; tm.install = jiffies_to_clock_t(jiffies - ipt->tcf_tm.install); tm.lastuse = jiffies_to_clock_t(jiffies - ipt->tcf_tm.lastuse); + tm.firstuse = jiffies_to_clock_t(jiffies - ipt->tcf_tm.firstuse); tm.expires = jiffies_to_clock_t(ipt->tcf_tm.expires); if (nla_put_64bit(skb, TCA_IPT_TM, sizeof(tm), &tm, TCA_IPT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index d3ac73e..1b06093 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -220,6 +220,7 @@ static int tcf_mirred_dump(struct sk_buff *skb, struct tc_action *a, int bind, i goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - m->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - m->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - m->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(m->tcf_tm.expires); if (nla_put_64bit(skb, TCA_MIRRED_TM, sizeof(t), &t, TCA_MIRRED_PAD)) goto nla_put_failure; diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c index 9135aa8..9fbf780 100644 --- a/net/sched/act_nat.c +++ b/net/sched/act_nat.c @@ -266,6 +266,7 @@ static int tcf_nat_dump(struct sk_buff *skb, struct tc_action *a, goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - p->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(p->tcf_tm.expires); if (nla_put_64bit(skb, TCA_NAT_TM, sizeof(t), &t, TCA_NAT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index 67a1726..fb89275 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -202,6 +202,7 @@ static int tcf_pedit_dump(struct sk_buff *skb, struct tc_action *a, goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - p->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(p->tcf_tm.expires); if (nla_put_64bit(skb, TCA_PEDIT_TM, sizeof(t), &t, TCA_PEDIT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_police.c b/net/sched/act_police.c index b884dae..820b116 100644 --- a/net/sched/act_police.c +++ b/net/sched/act_police.c @@ -241,6 +241,7 @@ override: tcf_hash_new_index(tn); police->tcf_tm.install = jiffies; police->tcf_tm.lastuse = jiffies; + police->tcf_tm.firstuse = 0; h = tcf_hash(police->tcf_index, POL_TAB_MASK); spin_lock_bh(&hinfo->lock); hlist_add_head(&police->tcf_head, &hinfo->htab[h]); @@ -347,6 +348,7 @@ tcf_act_police_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) t.install = jiffies_to_clock_t(jiffies - police->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - police->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - police->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(police->tcf_tm.expires); if (nla_put_64bit(skb, TCA_POLICE_TM, sizeof(t), &t, TCA_POLICE_PAD)) goto nla_put_failure; diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c index f95d1c5..81040f1 100644 --- a/net/sched/act_simple.c +++ b/net/sched/act_simple.c @@ -160,6 +160,7 @@ static int tcf_simp_dump(struct sk_buff *skb, struct tc_action *a, goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - d->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - d->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(d->tcf_tm.expires); if (nla_put_64bit(skb, TCA_DEF_TM, sizeof(t), &t, TCA_DEF_PAD)) goto nla_put_failure; diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index 82105c8..cf34f31 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -170,6 +170,7 @@ static int tcf_skbedit_dump(struct sk_buff *skb, struct tc_action *a, goto nla_put_failure; t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - d->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - d->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(d->tcf_tm.expires); if (nla_put_64bit(skb, TCA_SKBEDIT_TM, sizeof(t), &t, TCA_SKBEDIT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index da5120f..978ec4c 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -184,6 +184,7 @@ static int tcf_vlan_dump(struct sk_buff *skb, struct tc_action *a, t.install = jiffies_to_clock_t(jiffies - v->tcf_tm.install); t.lastuse = jiffies_to_clock_t(jiffies - v->tcf_tm.lastuse); + t.firstuse = jiffies_to_clock_t(jiffies - v->tcf_tm.firstuse); t.expires = jiffies_to_clock_t(v->tcf_tm.expires); if (nla_put_64bit(skb, TCA_VLAN_TM, sizeof(t), &t, TCA_VLAN_PAD)) goto nla_put_failure; -- cgit v0.10.2 From 48d8ee1694dd1ab25614b58f968123a4598f887e Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 6 Jun 2016 06:32:55 -0400 Subject: net sched actions: aggregate dumping of actions timeinfo Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller diff --git a/include/net/act_api.h b/include/net/act_api.h index 8389c00..a891978 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -80,6 +80,14 @@ static inline void tcf_lastuse_update(struct tcf_t *tm) tm->firstuse = now; } +static inline void tcf_tm_dump(struct tcf_t *dtm, const struct tcf_t *stm) +{ + dtm->install = jiffies_to_clock_t(jiffies - stm->install); + dtm->lastuse = jiffies_to_clock_t(jiffies - stm->lastuse); + dtm->firstuse = jiffies_to_clock_t(jiffies - stm->firstuse); + dtm->expires = jiffies_to_clock_t(stm->expires); +} + struct tc_action { void *priv; const struct tc_action_ops *ops; diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index e4b877f..ae0e7cb 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -154,11 +154,7 @@ static int tcf_bpf_dump(struct sk_buff *skb, struct tc_action *act, if (ret) goto nla_put_failure; - tm.install = jiffies_to_clock_t(jiffies - prog->tcf_tm.install); - tm.lastuse = jiffies_to_clock_t(jiffies - prog->tcf_tm.lastuse); - tm.firstuse = jiffies_to_clock_t(jiffies - prog->tcf_tm.firstuse); - tm.expires = jiffies_to_clock_t(prog->tcf_tm.expires); - + tcf_tm_dump(&tm, &prog->tcf_tm); if (nla_put_64bit(skb, TCA_ACT_BPF_TM, sizeof(tm), &tm, TCA_ACT_BPF_PAD)) goto nla_put_failure; diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c index e3f64f2..35a5270 100644 --- a/net/sched/act_connmark.c +++ b/net/sched/act_connmark.c @@ -160,10 +160,7 @@ static inline int tcf_connmark_dump(struct sk_buff *skb, struct tc_action *a, if (nla_put(skb, TCA_CONNMARK_PARMS, sizeof(opt), &opt)) goto nla_put_failure; - t.install = jiffies_to_clock_t(jiffies - ci->tcf_tm.install); - t.lastuse = jiffies_to_clock_t(jiffies - ci->tcf_tm.lastuse); - t.expires = jiffies_to_clock_t(ci->tcf_tm.expires); - t.firstuse = jiffies_to_clock_t(jiffies - ci->tcf_tm.firstuse); + tcf_tm_dump(&t, &ci->tcf_tm); if (nla_put_64bit(skb, TCA_CONNMARK_TM, sizeof(t), &t, TCA_CONNMARK_PAD)) goto nla_put_failure; diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c index 7725eafb..dcd9aba 100644 --- a/net/sched/act_csum.c +++ b/net/sched/act_csum.c @@ -546,10 +546,8 @@ static int tcf_csum_dump(struct sk_buff *skb, if (nla_put(skb, TCA_CSUM_PARMS, sizeof(opt), &opt)) goto nla_put_failure; - t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); - t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); - t.firstuse = jiffies_to_clock_t(jiffies - p->tcf_tm.firstuse); - t.expires = jiffies_to_clock_t(p->tcf_tm.expires); + + tcf_tm_dump(&t, &p->tcf_tm); if (nla_put_64bit(skb, TCA_CSUM_TM, sizeof(t), &t, TCA_CSUM_PAD)) goto nla_put_failure; diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c index c9d59f3..4c6e008 100644 --- a/net/sched/act_gact.c +++ b/net/sched/act_gact.c @@ -188,10 +188,7 @@ static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int goto nla_put_failure; } #endif - t.install = jiffies_to_clock_t(jiffies - gact->tcf_tm.install); - t.lastuse = jiffies_to_clock_t(jiffies - gact->tcf_tm.lastuse); - t.firstuse = jiffies_to_clock_t(jiffies - gact->tcf_tm.firstuse); - t.expires = jiffies_to_clock_t(gact->tcf_tm.expires); + tcf_tm_dump(&t, &gact->tcf_tm); if (nla_put_64bit(skb, TCA_GACT_TM, sizeof(t), &t, TCA_GACT_PAD)) goto nla_put_failure; return skb->len; diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c index 6491576..02f5a8b 100644 --- a/net/sched/act_ife.c +++ b/net/sched/act_ife.c @@ -553,9 +553,7 @@ static int tcf_ife_dump(struct sk_buff *skb, struct tc_action *a, int bind, if (nla_put(skb, TCA_IFE_PARMS, sizeof(opt), &opt)) goto nla_put_failure; - t.install = jiffies_to_clock_t(jiffies - ife->tcf_tm.install); - t.lastuse = jiffies_to_clock_t(jiffies - ife->tcf_tm.lastuse); - t.expires = jiffies_to_clock_t(ife->tcf_tm.expires); + tcf_tm_dump(&t, &ife->tcf_tm); if (nla_put_64bit(skb, TCA_IFE_TM, sizeof(t), &t, TCA_IFE_PAD)) goto nla_put_failure; diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index 47525ee..3fcde44 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -277,12 +277,11 @@ static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int nla_put(skb, TCA_IPT_CNT, sizeof(struct tc_cnt), &c) || nla_put_string(skb, TCA_IPT_TABLE, ipt->tcfi_tname)) goto nla_put_failure; - tm.install = jiffies_to_clock_t(jiffies - ipt->tcf_tm.install); - tm.lastuse = jiffies_to_clock_t(jiffies - ipt->tcf_tm.lastuse); - tm.firstuse = jiffies_to_clock_t(jiffies - ipt->tcf_tm.firstuse); - tm.expires = jiffies_to_clock_t(ipt->tcf_tm.expires); + + tcf_tm_dump(&tm, &ipt->tcf_tm); if (nla_put_64bit(skb, TCA_IPT_TM, sizeof(tm), &tm, TCA_IPT_PAD)) goto nla_put_failure; + kfree(t); return skb->len; diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 1b06093..787751a 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -218,10 +218,8 @@ static int tcf_mirred_dump(struct sk_buff *skb, struct tc_action *a, int bind, i if (nla_put(skb, TCA_MIRRED_PARMS, sizeof(opt), &opt)) goto nla_put_failure; - t.install = jiffies_to_clock_t(jiffies - m->tcf_tm.install); - t.lastuse = jiffies_to_clock_t(jiffies - m->tcf_tm.lastuse); - t.firstuse = jiffies_to_clock_t(jiffies - m->tcf_tm.firstuse); - t.expires = jiffies_to_clock_t(m->tcf_tm.expires); + + tcf_tm_dump(&t, &m->tcf_tm); if (nla_put_64bit(skb, TCA_MIRRED_TM, sizeof(t), &t, TCA_MIRRED_PAD)) goto nla_put_failure; return skb->len; diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c index 9fbf780..06ccb03 100644 --- a/net/sched/act_nat.c +++ b/net/sched/act_nat.c @@ -264,10 +264,8 @@ static int tcf_nat_dump(struct sk_buff *skb, struct tc_action *a, if (nla_put(skb, TCA_NAT_PARMS, sizeof(opt), &opt)) goto nla_put_failure; - t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); - t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); - t.firstuse = jiffies_to_clock_t(jiffies - p->tcf_tm.firstuse); - t.expires = jiffies_to_clock_t(p->tcf_tm.expires); + + tcf_tm_dump(&t, &p->tcf_tm); if (nla_put_64bit(skb, TCA_NAT_TM, sizeof(t), &t, TCA_NAT_PAD)) goto nla_put_failure; diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index fb89275..82d3c14 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -200,12 +200,11 @@ static int tcf_pedit_dump(struct sk_buff *skb, struct tc_action *a, if (nla_put(skb, TCA_PEDIT_PARMS, s, opt)) goto nla_put_failure; - t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install); - t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse); - t.firstuse = jiffies_to_clock_t(jiffies - p->tcf_tm.firstuse); - t.expires = jiffies_to_clock_t(p->tcf_tm.expires); + + tcf_tm_dump(&t, &p->tcf_tm); if (nla_put_64bit(skb, TCA_PEDIT_TM, sizeof(t), &t, TCA_PEDIT_PAD)) goto nla_put_failure; + kfree(opt); return skb->len; diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c index 81040f1..be5fbb5 100644 --- a/net/sched/act_simple.c +++ b/net/sched/act_simple.c @@ -158,10 +158,8 @@ static int tcf_simp_dump(struct sk_buff *skb, struct tc_action *a, if (nla_put(skb, TCA_DEF_PARMS, sizeof(opt), &opt) || nla_put_string(skb, TCA_DEF_DATA, d->tcfd_defdata)) goto nla_put_failure; - t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install); - t.lastuse = jiffies_to_clock_t(jiffies - d->tcf_tm.lastuse); - t.firstuse = jiffies_to_clock_t(jiffies - d->tcf_tm.firstuse); - t.expires = jiffies_to_clock_t(d->tcf_tm.expires); + + tcf_tm_dump(&t, &d->tcf_tm); if (nla_put_64bit(skb, TCA_DEF_TM, sizeof(t), &t, TCA_DEF_PAD)) goto nla_put_failure; return skb->len; diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index cf34f31..7e2bc3c 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -168,10 +168,8 @@ static int tcf_skbedit_dump(struct sk_buff *skb, struct tc_action *a, nla_put(skb, TCA_SKBEDIT_MARK, sizeof(d->mark), &d->mark)) goto nla_put_failure; - t.install = jiffies_to_clock_t(jiffies - d->tcf_tm.install); - t.lastuse = jiffies_to_clock_t(jiffies - d->tcf_tm.lastuse); - t.firstuse = jiffies_to_clock_t(jiffies - d->tcf_tm.firstuse); - t.expires = jiffies_to_clock_t(d->tcf_tm.expires); + + tcf_tm_dump(&t, &d->tcf_tm); if (nla_put_64bit(skb, TCA_SKBEDIT_TM, sizeof(t), &t, TCA_SKBEDIT_PAD)) goto nla_put_failure; return skb->len; diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index 978ec4c..f0a08a1 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -182,10 +182,7 @@ static int tcf_vlan_dump(struct sk_buff *skb, struct tc_action *a, nla_put_be16(skb, TCA_VLAN_PUSH_VLAN_PROTOCOL, v->tcfv_push_proto))) goto nla_put_failure; - t.install = jiffies_to_clock_t(jiffies - v->tcf_tm.install); - t.lastuse = jiffies_to_clock_t(jiffies - v->tcf_tm.lastuse); - t.firstuse = jiffies_to_clock_t(jiffies - v->tcf_tm.firstuse); - t.expires = jiffies_to_clock_t(v->tcf_tm.expires); + tcf_tm_dump(&t, &v->tcf_tm); if (nla_put_64bit(skb, TCA_VLAN_TM, sizeof(t), &t, TCA_VLAN_PAD)) goto nla_put_failure; return skb->len; -- cgit v0.10.2 From 0b0f43fe2e7291aa97b1febeaa5a0de453d007ca Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sun, 5 Jun 2016 10:41:32 -0400 Subject: net sched: indentation and other OCD stylistic fixes Signed-off-by: Jamal Hadi Salim Acked-by: Cong Wang diff --git a/include/net/act_api.h b/include/net/act_api.h index a891978..db218a1 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -2,8 +2,8 @@ #define __NET_ACT_API_H /* - * Public police action API for classifiers/qdiscs - */ + * Public action API for classifiers/qdiscs +*/ #include #include @@ -107,7 +107,8 @@ struct tc_action_ops { char kind[IFNAMSIZ]; __u32 type; /* TBD to match kind */ struct module *owner; - int (*act)(struct sk_buff *, const struct tc_action *, struct tcf_result *); + int (*act)(struct sk_buff *, const struct tc_action *, + struct tcf_result *); int (*dump)(struct sk_buff *, struct tc_action *, int, int); void (*cleanup)(struct tc_action *, int bind); int (*lookup)(struct net *, struct tc_action *, u32); @@ -125,8 +126,8 @@ struct tc_action_net { }; static inline -int tc_action_net_init(struct tc_action_net *tn, const struct tc_action_ops *ops, - unsigned int mask) +int tc_action_net_init(struct tc_action_net *tn, + const struct tc_action_ops *ops, unsigned int mask) { int err = 0; @@ -169,7 +170,8 @@ static inline int tcf_hash_release(struct tc_action *a, bool bind) } int tcf_register_action(struct tc_action_ops *a, struct pernet_operations *ops); -int tcf_unregister_action(struct tc_action_ops *a, struct pernet_operations *ops); +int tcf_unregister_action(struct tc_action_ops *a, + struct pernet_operations *ops); int tcf_action_destroy(struct list_head *actions, int bind); int tcf_action_exec(struct sk_buff *skb, const struct list_head *actions, struct tcf_result *res); diff --git a/include/net/tc_act/tc_defact.h b/include/net/tc_act/tc_defact.h index 9763dcb..ab9b5d6 100644 --- a/include/net/tc_act/tc_defact.h +++ b/include/net/tc_act/tc_defact.h @@ -5,8 +5,8 @@ struct tcf_defact { struct tcf_common common; - u32 tcfd_datalen; - void *tcfd_defdata; + u32 tcfd_datalen; + void *tcfd_defdata; }; #define to_defact(a) \ container_of(a->priv, struct tcf_defact, common) diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h index 9ba1410..5702e93 100644 --- a/include/uapi/linux/pkt_cls.h +++ b/include/uapi/linux/pkt_cls.h @@ -115,8 +115,8 @@ struct tc_police { __u32 mtu; struct tc_ratespec rate; struct tc_ratespec peakrate; - int refcnt; - int bindcnt; + int refcnt; + int bindcnt; __u32 capab; }; @@ -128,7 +128,7 @@ struct tcf_t { }; struct tc_cnt { - int refcnt; + int refcnt; int bindcnt; }; diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 5ebf6d6..719bc2e 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -504,8 +504,8 @@ nla_put_failure: } EXPORT_SYMBOL(tcf_action_dump_1); -int -tcf_action_dump(struct sk_buff *skb, struct list_head *actions, int bind, int ref) +int tcf_action_dump(struct sk_buff *skb, struct list_head *actions, + int bind, int ref) { struct tc_action *a; int err = -EINVAL; @@ -688,9 +688,9 @@ errout: return -1; } -static int -tca_get_fill(struct sk_buff *skb, struct list_head *actions, u32 portid, u32 seq, - u16 flags, int event, int bind, int ref) +static int tca_get_fill(struct sk_buff *skb, struct list_head *actions, + u32 portid, u32 seq, u16 flags, int event, int bind, + int ref) { struct tcamsg *t; struct nlmsghdr *nlh; @@ -731,7 +731,8 @@ act_get_notify(struct net *net, u32 portid, struct nlmsghdr *n, skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb) return -ENOBUFS; - if (tca_get_fill(skb, actions, portid, n->nlmsg_seq, 0, event, 0, 0) <= 0) { + if (tca_get_fill(skb, actions, portid, n->nlmsg_seq, 0, event, + 0, 0) <= 0) { kfree_skb(skb); return -EINVAL; } @@ -839,7 +840,8 @@ static int tca_action_flush(struct net *net, struct nlattr *nla, if (a.ops == NULL) /*some idjot trying to flush unknown action */ goto err_out; - nlh = nlmsg_put(skb, portid, n->nlmsg_seq, RTM_DELACTION, sizeof(*t), 0); + nlh = nlmsg_put(skb, portid, n->nlmsg_seq, RTM_DELACTION, + sizeof(*t), 0); if (!nlh) goto out_module_put; t = nlmsg_data(nlh); @@ -1002,7 +1004,8 @@ static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n) u32 portid = skb ? NETLINK_CB(skb).portid : 0; int ret = 0, ovr = 0; - if ((n->nlmsg_type != RTM_GETACTION) && !netlink_capable(skb, CAP_NET_ADMIN)) + if ((n->nlmsg_type != RTM_GETACTION) && + !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; ret = nlmsg_parse(n, sizeof(struct tcamsg), tca, TCA_ACT_MAX, NULL); diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index ae0e7cb..f7b6cf4 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -169,7 +169,8 @@ nla_put_failure: static const struct nla_policy act_bpf_policy[TCA_ACT_BPF_MAX + 1] = { [TCA_ACT_BPF_PARMS] = { .len = sizeof(struct tc_act_bpf) }, [TCA_ACT_BPF_FD] = { .type = NLA_U32 }, - [TCA_ACT_BPF_NAME] = { .type = NLA_NUL_STRING, .len = ACT_BPF_NAME_LEN }, + [TCA_ACT_BPF_NAME] = { .type = NLA_NUL_STRING, + .len = ACT_BPF_NAME_LEN }, [TCA_ACT_BPF_OPS_LEN] = { .type = NLA_U16 }, [TCA_ACT_BPF_OPS] = { .type = NLA_BINARY, .len = sizeof(struct sock_filter) * BPF_MAXINSNS }, diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c index 4c6e008..19058a7 100644 --- a/net/sched/act_gact.c +++ b/net/sched/act_gact.c @@ -162,7 +162,8 @@ static void tcf_gact_stats_update(struct tc_action *a, u64 bytes, u32 packets, tm->lastuse = lastuse; } -static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) +static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, + int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); struct tcf_gact *gact = a->priv; diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index 3fcde44..e7c0f4d 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -34,7 +34,8 @@ static int ipt_net_id; static int xt_net_id; -static int ipt_init_target(struct xt_entry_target *t, char *table, unsigned int hook) +static int ipt_init_target(struct xt_entry_target *t, char *table, + unsigned int hook) { struct xt_tgchk_param par; struct xt_target *target; @@ -250,7 +251,8 @@ static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a, } -static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) +static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, + int ref) { unsigned char *b = skb_tail_pointer(skb); struct tcf_ipt *ipt = a->priv; diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index f0a08a1..b075d50 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -179,7 +179,8 @@ static int tcf_vlan_dump(struct sk_buff *skb, struct tc_action *a, if (v->tcfv_action == TCA_VLAN_ACT_PUSH && (nla_put_u16(skb, TCA_VLAN_PUSH_VLAN_ID, v->tcfv_push_vid) || - nla_put_be16(skb, TCA_VLAN_PUSH_VLAN_PROTOCOL, v->tcfv_push_proto))) + nla_put_be16(skb, TCA_VLAN_PUSH_VLAN_PROTOCOL, + v->tcfv_push_proto))) goto nla_put_failure; tcf_tm_dump(&t, &v->tcf_tm); diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index a75864d..aafa6bce 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -351,8 +351,9 @@ errout: return err; } -static int tcf_fill_node(struct net *net, struct sk_buff *skb, struct tcf_proto *tp, - unsigned long fh, u32 portid, u32 seq, u16 flags, int event) +static int tcf_fill_node(struct net *net, struct sk_buff *skb, + struct tcf_proto *tp, unsigned long fh, u32 portid, + u32 seq, u16 flags, int event) { struct tcmsg *tcm; struct nlmsghdr *nlh; @@ -474,9 +475,11 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb) TC_H_MIN(tcm->tcm_info) != tp->protocol) continue; if (t > s_t) - memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(cb->args[0])); + memset(&cb->args[1], 0, + sizeof(cb->args)-sizeof(cb->args[0])); if (cb->args[1] == 0) { - if (tcf_fill_node(net, skb, tp, 0, NETLINK_CB(cb->skb).portid, + if (tcf_fill_node(net, skb, tp, 0, + NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, NLM_F_MULTI, RTM_NEWTFILTER) <= 0) break; -- cgit v0.10.2 From 68f047e3d62ebfac24ff9be551476cf30eafb00e Mon Sep 17 00:00:00 2001 From: Zhao Qiang Date: Mon, 6 Jun 2016 14:29:58 +0800 Subject: fsl/qe: add rx_sync and tx_sync for TDM mode Rx_sync and tx_sync are used by QE-TDM mode, add them to struct ucc_fast_info. Signed-off-by: Zhao Qiang Signed-off-by: David S. Miller diff --git a/drivers/soc/fsl/qe/qe.c b/drivers/soc/fsl/qe/qe.c index 709fc63..7026507 100644 --- a/drivers/soc/fsl/qe/qe.c +++ b/drivers/soc/fsl/qe/qe.c @@ -239,6 +239,12 @@ enum qe_clock qe_clock_source(const char *source) if (strcasecmp(source, "none") == 0) return QE_CLK_NONE; + if (strcmp(source, "tsync_pin") == 0) + return QE_TSYNC_PIN; + + if (strcmp(source, "rsync_pin") == 0) + return QE_RSYNC_PIN; + if (strncasecmp(source, "brg", 3) == 0) { i = simple_strtoul(source + 3, NULL, 10); if ((i >= 1) && (i <= 16)) diff --git a/include/soc/fsl/qe/qe.h b/include/soc/fsl/qe/qe.h index 33b29ea..f918745 100644 --- a/include/soc/fsl/qe/qe.h +++ b/include/soc/fsl/qe/qe.h @@ -80,6 +80,8 @@ enum qe_clock { QE_CLK22, /* Clock 22 */ QE_CLK23, /* Clock 23 */ QE_CLK24, /* Clock 24 */ + QE_RSYNC_PIN, /* RSYNC from pin */ + QE_TSYNC_PIN, /* TSYNC from pin */ QE_CLK_DUMMY }; diff --git a/include/soc/fsl/qe/ucc_fast.h b/include/soc/fsl/qe/ucc_fast.h index df8ea79..31548b7 100644 --- a/include/soc/fsl/qe/ucc_fast.h +++ b/include/soc/fsl/qe/ucc_fast.h @@ -120,6 +120,8 @@ struct ucc_fast_info { int ucc_num; enum qe_clock rx_clock; enum qe_clock tx_clock; + enum qe_clock rx_sync; + enum qe_clock tx_sync; u32 regs; int irq; u32 uccm_mask; -- cgit v0.10.2 From bb8b2062aff321af1fc58781cc07fbbea01cceb3 Mon Sep 17 00:00:00 2001 From: Zhao Qiang Date: Mon, 6 Jun 2016 14:29:59 +0800 Subject: fsl/qe: setup clock source for TDM mode Add tdm clock configuration in both qe clock system and ucc fast controller. Signed-off-by: Zhao Qiang Signed-off-by: David S. Miller diff --git a/drivers/soc/fsl/qe/ucc.c b/drivers/soc/fsl/qe/ucc.c index b59d335..c646d87 100644 --- a/drivers/soc/fsl/qe/ucc.c +++ b/drivers/soc/fsl/qe/ucc.c @@ -25,6 +25,12 @@ #include #include +#define UCC_TDM_NUM 8 +#define RX_SYNC_SHIFT_BASE 30 +#define TX_SYNC_SHIFT_BASE 14 +#define RX_CLK_SHIFT_BASE 28 +#define TX_CLK_SHIFT_BASE 12 + int ucc_set_qe_mux_mii_mng(unsigned int ucc_num) { unsigned long flags; @@ -210,3 +216,447 @@ int ucc_set_qe_mux_rxtx(unsigned int ucc_num, enum qe_clock clock, return 0; } + +static int ucc_get_tdm_common_clk(u32 tdm_num, enum qe_clock clock) +{ + int clock_bits = -EINVAL; + + /* + * for TDM[0, 1, 2, 3], TX and RX use common + * clock source BRG3,4 and CLK1,2 + * for TDM[4, 5, 6, 7], TX and RX use common + * clock source BRG12,13 and CLK23,24 + */ + switch (tdm_num) { + case 0: + case 1: + case 2: + case 3: + switch (clock) { + case QE_BRG3: + clock_bits = 1; + break; + case QE_BRG4: + clock_bits = 2; + break; + case QE_CLK1: + clock_bits = 4; + break; + case QE_CLK2: + clock_bits = 5; + break; + default: + break; + } + break; + case 4: + case 5: + case 6: + case 7: + switch (clock) { + case QE_BRG12: + clock_bits = 1; + break; + case QE_BRG13: + clock_bits = 2; + break; + case QE_CLK23: + clock_bits = 4; + break; + case QE_CLK24: + clock_bits = 5; + break; + default: + break; + } + break; + default: + break; + } + + return clock_bits; +} + +static int ucc_get_tdm_rx_clk(u32 tdm_num, enum qe_clock clock) +{ + int clock_bits = -EINVAL; + + switch (tdm_num) { + case 0: + switch (clock) { + case QE_CLK3: + clock_bits = 6; + break; + case QE_CLK8: + clock_bits = 7; + break; + default: + break; + } + break; + case 1: + switch (clock) { + case QE_CLK5: + clock_bits = 6; + break; + case QE_CLK10: + clock_bits = 7; + break; + default: + break; + } + break; + case 2: + switch (clock) { + case QE_CLK7: + clock_bits = 6; + break; + case QE_CLK12: + clock_bits = 7; + break; + default: + break; + } + break; + case 3: + switch (clock) { + case QE_CLK9: + clock_bits = 6; + break; + case QE_CLK14: + clock_bits = 7; + break; + default: + break; + } + break; + case 4: + switch (clock) { + case QE_CLK11: + clock_bits = 6; + break; + case QE_CLK16: + clock_bits = 7; + break; + default: + break; + } + break; + case 5: + switch (clock) { + case QE_CLK13: + clock_bits = 6; + break; + case QE_CLK18: + clock_bits = 7; + break; + default: + break; + } + break; + case 6: + switch (clock) { + case QE_CLK15: + clock_bits = 6; + break; + case QE_CLK20: + clock_bits = 7; + break; + default: + break; + } + break; + case 7: + switch (clock) { + case QE_CLK17: + clock_bits = 6; + break; + case QE_CLK22: + clock_bits = 7; + break; + default: + break; + } + break; + } + + return clock_bits; +} + +static int ucc_get_tdm_tx_clk(u32 tdm_num, enum qe_clock clock) +{ + int clock_bits = -EINVAL; + + switch (tdm_num) { + case 0: + switch (clock) { + case QE_CLK4: + clock_bits = 6; + break; + case QE_CLK9: + clock_bits = 7; + break; + default: + break; + } + break; + case 1: + switch (clock) { + case QE_CLK6: + clock_bits = 6; + break; + case QE_CLK11: + clock_bits = 7; + break; + default: + break; + } + break; + case 2: + switch (clock) { + case QE_CLK8: + clock_bits = 6; + break; + case QE_CLK13: + clock_bits = 7; + break; + default: + break; + } + break; + case 3: + switch (clock) { + case QE_CLK10: + clock_bits = 6; + break; + case QE_CLK15: + clock_bits = 7; + break; + default: + break; + } + break; + case 4: + switch (clock) { + case QE_CLK12: + clock_bits = 6; + break; + case QE_CLK17: + clock_bits = 7; + break; + default: + break; + } + break; + case 5: + switch (clock) { + case QE_CLK14: + clock_bits = 6; + break; + case QE_CLK19: + clock_bits = 7; + break; + default: + break; + } + break; + case 6: + switch (clock) { + case QE_CLK16: + clock_bits = 6; + break; + case QE_CLK21: + clock_bits = 7; + break; + default: + break; + } + break; + case 7: + switch (clock) { + case QE_CLK18: + clock_bits = 6; + break; + case QE_CLK3: + clock_bits = 7; + break; + default: + break; + } + break; + } + + return clock_bits; +} + +/* tdm_num: TDM A-H port num is 0-7 */ +static int ucc_get_tdm_rxtx_clk(enum comm_dir mode, u32 tdm_num, + enum qe_clock clock) +{ + int clock_bits; + + clock_bits = ucc_get_tdm_common_clk(tdm_num, clock); + if (clock_bits > 0) + return clock_bits; + if (mode == COMM_DIR_RX) + clock_bits = ucc_get_tdm_rx_clk(tdm_num, clock); + if (mode == COMM_DIR_TX) + clock_bits = ucc_get_tdm_tx_clk(tdm_num, clock); + return clock_bits; +} + +static u32 ucc_get_tdm_clk_shift(enum comm_dir mode, u32 tdm_num) +{ + u32 shift; + + shift = (mode == COMM_DIR_RX) ? RX_CLK_SHIFT_BASE : TX_CLK_SHIFT_BASE; + if (tdm_num < 4) + shift -= tdm_num * 4; + else + shift -= (tdm_num - 4) * 4; + + return shift; +} + +int ucc_set_tdm_rxtx_clk(u32 tdm_num, enum qe_clock clock, + enum comm_dir mode) +{ + int clock_bits; + u32 shift; + struct qe_mux __iomem *qe_mux_reg; + __be32 __iomem *cmxs1cr; + + qe_mux_reg = &qe_immr->qmx; + + if (tdm_num > 7 || tdm_num < 0) + return -EINVAL; + + /* The communications direction must be RX or TX */ + if (mode != COMM_DIR_RX && mode != COMM_DIR_TX) + return -EINVAL; + + clock_bits = ucc_get_tdm_rxtx_clk(mode, tdm_num, clock); + if (clock_bits < 0) + return -EINVAL; + + shift = ucc_get_tdm_clk_shift(mode, tdm_num); + + cmxs1cr = (tdm_num < 4) ? &qe_mux_reg->cmxsi1cr_l : + &qe_mux_reg->cmxsi1cr_h; + + qe_clrsetbits32(cmxs1cr, QE_CMXUCR_TX_CLK_SRC_MASK << shift, + clock_bits << shift); + + return 0; +} + +static int ucc_get_tdm_sync_source(u32 tdm_num, enum qe_clock clock, + enum comm_dir mode) +{ + int source = -EINVAL; + + if (mode == COMM_DIR_RX && clock == QE_RSYNC_PIN) { + source = 0; + return source; + } + if (mode == COMM_DIR_TX && clock == QE_TSYNC_PIN) { + source = 0; + return source; + } + + switch (tdm_num) { + case 0: + case 1: + switch (clock) { + case QE_BRG9: + source = 1; + break; + case QE_BRG10: + source = 2; + break; + default: + break; + } + break; + case 2: + case 3: + switch (clock) { + case QE_BRG9: + source = 1; + break; + case QE_BRG11: + source = 2; + break; + default: + break; + } + break; + case 4: + case 5: + switch (clock) { + case QE_BRG13: + source = 1; + break; + case QE_BRG14: + source = 2; + break; + default: + break; + } + break; + case 6: + case 7: + switch (clock) { + case QE_BRG13: + source = 1; + break; + case QE_BRG15: + source = 2; + break; + default: + break; + } + break; + } + + return source; +} + +static u32 ucc_get_tdm_sync_shift(enum comm_dir mode, u32 tdm_num) +{ + u32 shift; + + shift = (mode == COMM_DIR_RX) ? RX_SYNC_SHIFT_BASE : RX_SYNC_SHIFT_BASE; + shift -= tdm_num * 2; + + return shift; +} + +int ucc_set_tdm_rxtx_sync(u32 tdm_num, enum qe_clock clock, + enum comm_dir mode) +{ + int source; + u32 shift; + struct qe_mux *qe_mux_reg; + + qe_mux_reg = &qe_immr->qmx; + + if (tdm_num >= UCC_TDM_NUM) + return -EINVAL; + + /* The communications direction must be RX or TX */ + if (mode != COMM_DIR_RX && mode != COMM_DIR_TX) + return -EINVAL; + + source = ucc_get_tdm_sync_source(tdm_num, clock, mode); + if (source < 0) + return -EINVAL; + + shift = ucc_get_tdm_sync_shift(mode, tdm_num); + + qe_clrsetbits32(&qe_mux_reg->cmxsi1syr, + QE_CMXUCR_TX_CLK_SRC_MASK << shift, + source << shift); + + return 0; +} diff --git a/drivers/soc/fsl/qe/ucc_fast.c b/drivers/soc/fsl/qe/ucc_fast.c index a768931..83d8d16 100644 --- a/drivers/soc/fsl/qe/ucc_fast.c +++ b/drivers/soc/fsl/qe/ucc_fast.c @@ -327,6 +327,42 @@ int ucc_fast_init(struct ucc_fast_info * uf_info, struct ucc_fast_private ** ucc ucc_fast_free(uccf); return -EINVAL; } + } else { + /* tdm Rx clock routing */ + if ((uf_info->rx_clock != QE_CLK_NONE) && + ucc_set_tdm_rxtx_clk(uf_info->tdm_num, uf_info->rx_clock, + COMM_DIR_RX)) { + pr_err("%s: illegal value for RX clock", __func__); + ucc_fast_free(uccf); + return -EINVAL; + } + + /* tdm Tx clock routing */ + if ((uf_info->tx_clock != QE_CLK_NONE) && + ucc_set_tdm_rxtx_clk(uf_info->tdm_num, uf_info->tx_clock, + COMM_DIR_TX)) { + pr_err("%s: illegal value for TX clock", __func__); + ucc_fast_free(uccf); + return -EINVAL; + } + + /* tdm Rx sync clock routing */ + if ((uf_info->rx_sync != QE_CLK_NONE) && + ucc_set_tdm_rxtx_sync(uf_info->tdm_num, uf_info->rx_sync, + COMM_DIR_RX)) { + pr_err("%s: illegal value for RX clock", __func__); + ucc_fast_free(uccf); + return -EINVAL; + } + + /* tdm Tx sync clock routing */ + if ((uf_info->tx_sync != QE_CLK_NONE) && + ucc_set_tdm_rxtx_sync(uf_info->tdm_num, uf_info->tx_sync, + COMM_DIR_TX)) { + pr_err("%s: illegal value for TX clock", __func__); + ucc_fast_free(uccf); + return -EINVAL; + } } /* Set interrupt mask register at UCC level. */ diff --git a/include/soc/fsl/qe/qe.h b/include/soc/fsl/qe/qe.h index f918745..c3b1dc8 100644 --- a/include/soc/fsl/qe/qe.h +++ b/include/soc/fsl/qe/qe.h @@ -244,6 +244,22 @@ static inline int qe_alive_during_sleep(void) #define qe_muram_addr cpm_muram_addr #define qe_muram_offset cpm_muram_offset +#define qe_setbits32(_addr, _v) iowrite32be(ioread32be(_addr) | (_v), (_addr)) +#define qe_clrbits32(_addr, _v) iowrite32be(ioread32be(_addr) & ~(_v), (_addr)) + +#define qe_setbits16(_addr, _v) iowrite16be(ioread16be(_addr) | (_v), (_addr)) +#define qe_clrbits16(_addr, _v) iowrite16be(ioread16be(_addr) & ~(_v), (_addr)) + +#define qe_setbits8(_addr, _v) iowrite8(ioread8(_addr) | (_v), (_addr)) +#define qe_clrbits8(_addr, _v) iowrite8(ioread8(_addr) & ~(_v), (_addr)) + +#define qe_clrsetbits32(addr, clear, set) \ + iowrite32be((ioread32be(addr) & ~(clear)) | (set), (addr)) +#define qe_clrsetbits16(addr, clear, set) \ + iowrite16be((ioread16be(addr) & ~(clear)) | (set), (addr)) +#define qe_clrsetbits8(addr, clear, set) \ + iowrite8((ioread8(addr) & ~(clear)) | (set), (addr)) + /* Structure that defines QE firmware binary files. * * See Documentation/powerpc/qe_firmware.txt for a description of these diff --git a/include/soc/fsl/qe/ucc.h b/include/soc/fsl/qe/ucc.h index 894f14c..6bbbb59 100644 --- a/include/soc/fsl/qe/ucc.h +++ b/include/soc/fsl/qe/ucc.h @@ -41,6 +41,10 @@ int ucc_set_qe_mux_mii_mng(unsigned int ucc_num); int ucc_set_qe_mux_rxtx(unsigned int ucc_num, enum qe_clock clock, enum comm_dir mode); +int ucc_set_tdm_rxtx_clk(unsigned int tdm_num, enum qe_clock clock, + enum comm_dir mode); +int ucc_set_tdm_rxtx_sync(unsigned int tdm_num, enum qe_clock clock, + enum comm_dir mode); int ucc_mux_set_grant_tsa_bkpt(unsigned int ucc_num, int set, u32 mask); diff --git a/include/soc/fsl/qe/ucc_fast.h b/include/soc/fsl/qe/ucc_fast.h index 31548b7..b2633b7 100644 --- a/include/soc/fsl/qe/ucc_fast.h +++ b/include/soc/fsl/qe/ucc_fast.h @@ -118,6 +118,7 @@ enum ucc_fast_transparent_tcrc { /* Fast UCC initialization structure */ struct ucc_fast_info { int ucc_num; + int tdm_num; enum qe_clock rx_clock; enum qe_clock tx_clock; enum qe_clock rx_sync; -- cgit v0.10.2 From 19163ac3123e7fef8b1ecb2f1d4223f58ed5e884 Mon Sep 17 00:00:00 2001 From: Zhao Qiang Date: Mon, 6 Jun 2016 14:30:00 +0800 Subject: fsl/qe: Make regs resouce_size_t Signed-off-by: Zhao Qiang Signed-off-by: David S. Miller diff --git a/include/soc/fsl/qe/ucc_fast.h b/include/soc/fsl/qe/ucc_fast.h index b2633b7..e898895 100644 --- a/include/soc/fsl/qe/ucc_fast.h +++ b/include/soc/fsl/qe/ucc_fast.h @@ -123,7 +123,7 @@ struct ucc_fast_info { enum qe_clock tx_clock; enum qe_clock rx_sync; enum qe_clock tx_sync; - u32 regs; + resource_size_t regs; int irq; u32 uccm_mask; int bd_mem_part; -- cgit v0.10.2 From 35ef1c20fdb26779b6c3c4fd74bbdd5028e70005 Mon Sep 17 00:00:00 2001 From: Zhao Qiang Date: Mon, 6 Jun 2016 14:30:01 +0800 Subject: fsl/qe: Add QE TDM lib QE has module to support TDM, some other protocols supported by QE are based on TDM. add a qe-tdm lib, this lib provides functions to the protocols using TDM to configurate QE-TDM. Signed-off-by: Zhao Qiang Signed-off-by: David S. Miller diff --git a/drivers/soc/fsl/qe/Kconfig b/drivers/soc/fsl/qe/Kconfig index 20978f2..73a2e08 100644 --- a/drivers/soc/fsl/qe/Kconfig +++ b/drivers/soc/fsl/qe/Kconfig @@ -22,7 +22,7 @@ config UCC_SLOW config UCC_FAST bool - default y if UCC_GETH + default y if UCC_GETH || QE_TDM help This option provides qe_lib support to UCC fast protocols: HDLC, Ethernet, ATM, transparent @@ -31,6 +31,10 @@ config UCC bool default y if UCC_FAST || UCC_SLOW +config QE_TDM + bool + default y if FSL_UCC_HDLC + config QE_USB bool default y if USB_FSL_QE diff --git a/drivers/soc/fsl/qe/Makefile b/drivers/soc/fsl/qe/Makefile index ffac541..2031d38 100644 --- a/drivers/soc/fsl/qe/Makefile +++ b/drivers/soc/fsl/qe/Makefile @@ -6,5 +6,6 @@ obj-$(CONFIG_CPM) += qe_common.o obj-$(CONFIG_UCC) += ucc.o obj-$(CONFIG_UCC_SLOW) += ucc_slow.o obj-$(CONFIG_UCC_FAST) += ucc_fast.o +obj-$(CONFIG_QE_TDM) += qe_tdm.o obj-$(CONFIG_QE_USB) += usb.o obj-$(CONFIG_QE_GPIO) += gpio.o diff --git a/drivers/soc/fsl/qe/qe_tdm.c b/drivers/soc/fsl/qe/qe_tdm.c new file mode 100644 index 0000000..5e48b14 --- /dev/null +++ b/drivers/soc/fsl/qe/qe_tdm.c @@ -0,0 +1,276 @@ +/* + * Copyright (C) 2015 Freescale Semiconductor, Inc. All rights reserved. + * + * Authors: Zhao Qiang + * + * Description: + * QE TDM API Set - TDM specific routines implementations. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ +#include +#include +#include +#include +#include +#include + +static int set_tdm_framer(const char *tdm_framer_type) +{ + if (strcmp(tdm_framer_type, "e1") == 0) + return TDM_FRAMER_E1; + else if (strcmp(tdm_framer_type, "t1") == 0) + return TDM_FRAMER_T1; + else + return -EINVAL; +} + +static void set_si_param(struct ucc_tdm *utdm, struct ucc_tdm_info *ut_info) +{ + struct si_mode_info *si_info = &ut_info->si_info; + + if (utdm->tdm_mode == TDM_INTERNAL_LOOPBACK) { + si_info->simr_crt = 1; + si_info->simr_rfsd = 0; + } +} + +int ucc_of_parse_tdm(struct device_node *np, struct ucc_tdm *utdm, + struct ucc_tdm_info *ut_info) +{ + const char *sprop; + int ret = 0; + u32 val; + struct resource *res; + struct device_node *np2; + static int siram_init_flag; + struct platform_device *pdev; + + sprop = of_get_property(np, "fsl,rx-sync-clock", NULL); + if (sprop) { + ut_info->uf_info.rx_sync = qe_clock_source(sprop); + if ((ut_info->uf_info.rx_sync < QE_CLK_NONE) || + (ut_info->uf_info.rx_sync > QE_RSYNC_PIN)) { + pr_err("QE-TDM: Invalid rx-sync-clock property\n"); + return -EINVAL; + } + } else { + pr_err("QE-TDM: Invalid rx-sync-clock property\n"); + return -EINVAL; + } + + sprop = of_get_property(np, "fsl,tx-sync-clock", NULL); + if (sprop) { + ut_info->uf_info.tx_sync = qe_clock_source(sprop); + if ((ut_info->uf_info.tx_sync < QE_CLK_NONE) || + (ut_info->uf_info.tx_sync > QE_TSYNC_PIN)) { + pr_err("QE-TDM: Invalid tx-sync-clock property\n"); + return -EINVAL; + } + } else { + pr_err("QE-TDM: Invalid tx-sync-clock property\n"); + return -EINVAL; + } + + ret = of_property_read_u32_index(np, "fsl,tx-timeslot-mask", 0, &val); + if (ret) { + pr_err("QE-TDM: Invalid tx-timeslot-mask property\n"); + return -EINVAL; + } + utdm->tx_ts_mask = val; + + ret = of_property_read_u32_index(np, "fsl,rx-timeslot-mask", 0, &val); + if (ret) { + ret = -EINVAL; + pr_err("QE-TDM: Invalid rx-timeslot-mask property\n"); + return ret; + } + utdm->rx_ts_mask = val; + + ret = of_property_read_u32_index(np, "fsl,tdm-id", 0, &val); + if (ret) { + ret = -EINVAL; + pr_err("QE-TDM: No fsl,tdm-id property for this UCC\n"); + return ret; + } + utdm->tdm_port = val; + ut_info->uf_info.tdm_num = utdm->tdm_port; + + if (of_get_property(np, "fsl,tdm-internal-loopback", NULL)) + utdm->tdm_mode = TDM_INTERNAL_LOOPBACK; + else + utdm->tdm_mode = TDM_NORMAL; + + sprop = of_get_property(np, "fsl,tdm-framer-type", NULL); + if (!sprop) { + ret = -EINVAL; + pr_err("QE-TDM: No tdm-framer-type property for UCC\n"); + return ret; + } + ret = set_tdm_framer(sprop); + if (ret < 0) + return -EINVAL; + utdm->tdm_framer_type = ret; + + ret = of_property_read_u32_index(np, "fsl,siram-entry-id", 0, &val); + if (ret) { + ret = -EINVAL; + pr_err("QE-TDM: No siram entry id for UCC\n"); + return ret; + } + utdm->siram_entry_id = val; + + set_si_param(utdm, ut_info); + + np2 = of_find_compatible_node(NULL, NULL, "fsl,t1040-qe-si"); + if (!np2) + return -EINVAL; + + pdev = of_find_device_by_node(np2); + if (!pdev) { + pr_err("%s: failed to lookup pdev\n", np2->name); + of_node_put(np2); + return -EINVAL; + } + + of_node_put(np2); + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + utdm->si_regs = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(utdm->si_regs)) { + ret = PTR_ERR(utdm->si_regs); + goto err_miss_siram_property; + } + + np2 = of_find_compatible_node(NULL, NULL, "fsl,t1040-qe-siram"); + if (!np2) { + ret = -EINVAL; + goto err_miss_siram_property; + } + + pdev = of_find_device_by_node(np2); + if (!pdev) { + ret = -EINVAL; + pr_err("%s: failed to lookup pdev\n", np2->name); + of_node_put(np2); + goto err_miss_siram_property; + } + + of_node_put(np2); + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + utdm->siram = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(utdm->siram)) { + ret = PTR_ERR(utdm->siram); + goto err_miss_siram_property; + } + + if (siram_init_flag == 0) { + memset_io(utdm->siram, 0, res->end - res->start + 1); + siram_init_flag = 1; + } + + return ret; + +err_miss_siram_property: + devm_iounmap(&pdev->dev, utdm->si_regs); + return ret; +} + +void ucc_tdm_init(struct ucc_tdm *utdm, struct ucc_tdm_info *ut_info) +{ + struct si1 __iomem *si_regs; + u16 __iomem *siram; + u16 siram_entry_valid; + u16 siram_entry_closed; + u16 ucc_num; + u8 csel; + u16 sixmr; + u16 tdm_port; + u32 siram_entry_id; + u32 mask; + int i; + + si_regs = utdm->si_regs; + siram = utdm->siram; + ucc_num = ut_info->uf_info.ucc_num; + tdm_port = utdm->tdm_port; + siram_entry_id = utdm->siram_entry_id; + + if (utdm->tdm_framer_type == TDM_FRAMER_T1) + utdm->num_of_ts = 24; + if (utdm->tdm_framer_type == TDM_FRAMER_E1) + utdm->num_of_ts = 32; + + /* set siram table */ + csel = (ucc_num < 4) ? ucc_num + 9 : ucc_num - 3; + + siram_entry_valid = SIR_CSEL(csel) | SIR_BYTE | SIR_CNT(0); + siram_entry_closed = SIR_IDLE | SIR_BYTE | SIR_CNT(0); + + for (i = 0; i < utdm->num_of_ts; i++) { + mask = 0x01 << i; + + if (utdm->tx_ts_mask & mask) + iowrite16be(siram_entry_valid, + &siram[siram_entry_id * 32 + i]); + else + iowrite16be(siram_entry_closed, + &siram[siram_entry_id * 32 + i]); + + if (utdm->rx_ts_mask & mask) + iowrite16be(siram_entry_valid, + &siram[siram_entry_id * 32 + 0x200 + i]); + else + iowrite16be(siram_entry_closed, + &siram[siram_entry_id * 32 + 0x200 + i]); + } + + setbits16(&siram[(siram_entry_id * 32) + (utdm->num_of_ts - 1)], + SIR_LAST); + setbits16(&siram[(siram_entry_id * 32) + 0x200 + (utdm->num_of_ts - 1)], + SIR_LAST); + + /* Set SIxMR register */ + sixmr = SIMR_SAD(siram_entry_id); + + sixmr &= ~SIMR_SDM_MASK; + + if (utdm->tdm_mode == TDM_INTERNAL_LOOPBACK) + sixmr |= SIMR_SDM_INTERNAL_LOOPBACK; + else + sixmr |= SIMR_SDM_NORMAL; + + sixmr |= SIMR_RFSD(ut_info->si_info.simr_rfsd) | + SIMR_TFSD(ut_info->si_info.simr_tfsd); + + if (ut_info->si_info.simr_crt) + sixmr |= SIMR_CRT; + if (ut_info->si_info.simr_sl) + sixmr |= SIMR_SL; + if (ut_info->si_info.simr_ce) + sixmr |= SIMR_CE; + if (ut_info->si_info.simr_fe) + sixmr |= SIMR_FE; + if (ut_info->si_info.simr_gm) + sixmr |= SIMR_GM; + + switch (tdm_port) { + case 0: + iowrite16be(sixmr, &si_regs->sixmr1[0]); + break; + case 1: + iowrite16be(sixmr, &si_regs->sixmr1[1]); + break; + case 2: + iowrite16be(sixmr, &si_regs->sixmr1[2]); + break; + case 3: + iowrite16be(sixmr, &si_regs->sixmr1[3]); + break; + default: + pr_err("QE-TDM: can not find tdm sixmr reg\n"); + break; + } +} diff --git a/include/soc/fsl/qe/immap_qe.h b/include/soc/fsl/qe/immap_qe.h index bedbff8..c76ef30 100644 --- a/include/soc/fsl/qe/immap_qe.h +++ b/include/soc/fsl/qe/immap_qe.h @@ -159,10 +159,7 @@ struct spi { /* SI */ struct si1 { - __be16 siamr1; /* SI1 TDMA mode register */ - __be16 sibmr1; /* SI1 TDMB mode register */ - __be16 sicmr1; /* SI1 TDMC mode register */ - __be16 sidmr1; /* SI1 TDMD mode register */ + __be16 sixmr1[4]; /* SI1 TDMx (x = A B C D) mode register */ u8 siglmr1_h; /* SI1 global mode register high */ u8 res0[0x1]; u8 sicmdr1_h; /* SI1 command register high */ diff --git a/include/soc/fsl/qe/qe_tdm.h b/include/soc/fsl/qe/qe_tdm.h new file mode 100644 index 0000000..4c91498 --- /dev/null +++ b/include/soc/fsl/qe/qe_tdm.h @@ -0,0 +1,94 @@ +/* + * Internal header file for QE TDM mode routines. + * + * Copyright (C) 2016 Freescale Semiconductor, Inc. All rights reserved. + * + * Authors: Zhao Qiang + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version + */ + +#ifndef CONFIG_QE_TDM_H +#define CONFIG_QE_TDM_H + +#include +#include + +#include +#include + +#include +#include + +/* SI RAM entries */ +#define SIR_LAST 0x0001 +#define SIR_BYTE 0x0002 +#define SIR_CNT(x) ((x) << 2) +#define SIR_CSEL(x) ((x) << 5) +#define SIR_SGS 0x0200 +#define SIR_SWTR 0x4000 +#define SIR_MCC 0x8000 +#define SIR_IDLE 0 + +/* SIxMR fields */ +#define SIMR_SAD(x) ((x) << 12) +#define SIMR_SDM_NORMAL 0x0000 +#define SIMR_SDM_INTERNAL_LOOPBACK 0x0800 +#define SIMR_SDM_MASK 0x0c00 +#define SIMR_CRT 0x0040 +#define SIMR_SL 0x0020 +#define SIMR_CE 0x0010 +#define SIMR_FE 0x0008 +#define SIMR_GM 0x0004 +#define SIMR_TFSD(n) (n) +#define SIMR_RFSD(n) ((n) << 8) + +enum tdm_ts_t { + TDM_TX_TS, + TDM_RX_TS +}; + +enum tdm_framer_t { + TDM_FRAMER_T1, + TDM_FRAMER_E1 +}; + +enum tdm_mode_t { + TDM_INTERNAL_LOOPBACK, + TDM_NORMAL +}; + +struct si_mode_info { + u8 simr_rfsd; + u8 simr_tfsd; + u8 simr_crt; + u8 simr_sl; + u8 simr_ce; + u8 simr_fe; + u8 simr_gm; +}; + +struct ucc_tdm_info { + struct ucc_fast_info uf_info; + struct si_mode_info si_info; +}; + +struct ucc_tdm { + u16 tdm_port; /* port for this tdm:TDMA,TDMB */ + u32 siram_entry_id; + u16 __iomem *siram; + struct si1 __iomem *si_regs; + enum tdm_framer_t tdm_framer_type; + enum tdm_mode_t tdm_mode; + u8 num_of_ts; /* the number of timeslots in this tdm frame */ + u32 tx_ts_mask; /* tx time slot mask */ + u32 rx_ts_mask; /* rx time slot mask */ +}; + +int ucc_of_parse_tdm(struct device_node *np, struct ucc_tdm *utdm, + struct ucc_tdm_info *ut_info); +void ucc_tdm_init(struct ucc_tdm *utdm, struct ucc_tdm_info *ut_info); +#endif -- cgit v0.10.2 From c19b6d246a35627c3a69b2fa6bdece212b48214b Mon Sep 17 00:00:00 2001 From: Zhao Qiang Date: Mon, 6 Jun 2016 14:30:02 +0800 Subject: drivers/net: support hdlc function for QE-UCC The driver add hdlc support for Freescale QUICC Engine. It support NMSI and TSA mode. Signed-off-by: Zhao Qiang Signed-off-by: David S. Miller diff --git a/MAINTAINERS b/MAINTAINERS index f8d5a37..16e1500 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4870,6 +4870,13 @@ F: drivers/net/ethernet/freescale/gianfar* X: drivers/net/ethernet/freescale/gianfar_ptp.c F: Documentation/devicetree/bindings/net/fsl-tsec-phy.txt +FREESCALE QUICC ENGINE UCC HDLC DRIVER +M: Zhao Qiang +L: netdev@vger.kernel.org +L: linuxppc-dev@lists.ozlabs.org +S: Maintained +F: drivers/net/wan/fsl_ucc_hdlc* + FREESCALE QUICC ENGINE UCC UART DRIVER M: Timur Tabi L: linuxppc-dev@lists.ozlabs.org diff --git a/drivers/net/wan/Kconfig b/drivers/net/wan/Kconfig index a2fdd15..9e314b7 100644 --- a/drivers/net/wan/Kconfig +++ b/drivers/net/wan/Kconfig @@ -280,6 +280,17 @@ config DSCC4 To compile this driver as a module, choose M here: the module will be called dscc4. +config FSL_UCC_HDLC + tristate "Freescale QUICC Engine HDLC support" + depends on HDLC + depends on QUICC_ENGINE + help + Driver for Freescale QUICC Engine HDLC controller. The driver + supports HDLC in NMSI and TDM mode. + + To compile this driver as a module, choose M here: the + module will be called fsl_ucc_hdlc. + config DSCC4_PCISYNC bool "Etinc PCISYNC features" depends on DSCC4 diff --git a/drivers/net/wan/Makefile b/drivers/net/wan/Makefile index c135ef4..25fec40 100644 --- a/drivers/net/wan/Makefile +++ b/drivers/net/wan/Makefile @@ -32,6 +32,7 @@ obj-$(CONFIG_WANXL) += wanxl.o obj-$(CONFIG_PCI200SYN) += pci200syn.o obj-$(CONFIG_PC300TOO) += pc300too.o obj-$(CONFIG_IXP4XX_HSS) += ixp4xx_hss.o +obj-$(CONFIG_FSL_UCC_HDLC) += fsl_ucc_hdlc.o clean-files := wanxlfw.inc $(obj)/wanxl.o: $(obj)/wanxlfw.inc diff --git a/drivers/net/wan/fsl_ucc_hdlc.c b/drivers/net/wan/fsl_ucc_hdlc.c new file mode 100644 index 0000000..19174ac --- /dev/null +++ b/drivers/net/wan/fsl_ucc_hdlc.c @@ -0,0 +1,1192 @@ +/* Freescale QUICC Engine HDLC Device Driver + * + * Copyright 2016 Freescale Semiconductor Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "fsl_ucc_hdlc.h" + +#define DRV_DESC "Freescale QE UCC HDLC Driver" +#define DRV_NAME "ucc_hdlc" + +#define TDM_PPPOHT_SLIC_MAXIN +#define BROKEN_FRAME_INFO + +static struct ucc_tdm_info utdm_primary_info = { + .uf_info = { + .tsa = 0, + .cdp = 0, + .cds = 1, + .ctsp = 1, + .ctss = 1, + .revd = 0, + .urfs = 256, + .utfs = 256, + .urfet = 128, + .urfset = 192, + .utfet = 128, + .utftt = 0x40, + .ufpt = 256, + .mode = UCC_FAST_PROTOCOL_MODE_HDLC, + .ttx_trx = UCC_FAST_GUMR_TRANSPARENT_TTX_TRX_NORMAL, + .tenc = UCC_FAST_TX_ENCODING_NRZ, + .renc = UCC_FAST_RX_ENCODING_NRZ, + .tcrc = UCC_FAST_16_BIT_CRC, + .synl = UCC_FAST_SYNC_LEN_NOT_USED, + }, + + .si_info = { +#ifdef TDM_PPPOHT_SLIC_MAXIN + .simr_rfsd = 1, + .simr_tfsd = 2, +#else + .simr_rfsd = 0, + .simr_tfsd = 0, +#endif + .simr_crt = 0, + .simr_sl = 0, + .simr_ce = 1, + .simr_fe = 1, + .simr_gm = 0, + }, +}; + +static struct ucc_tdm_info utdm_info[MAX_HDLC_NUM]; + +static int uhdlc_init(struct ucc_hdlc_private *priv) +{ + struct ucc_tdm_info *ut_info; + struct ucc_fast_info *uf_info; + u32 cecr_subblock; + u16 bd_status; + int ret, i; + void *bd_buffer; + dma_addr_t bd_dma_addr; + u32 riptr; + u32 tiptr; + u32 gumr; + + ut_info = priv->ut_info; + uf_info = &ut_info->uf_info; + + if (priv->tsa) { + uf_info->tsa = 1; + uf_info->ctsp = 1; + } + uf_info->uccm_mask = ((UCC_HDLC_UCCE_RXB | UCC_HDLC_UCCE_RXF | + UCC_HDLC_UCCE_TXB) << 16); + + ret = ucc_fast_init(uf_info, &priv->uccf); + if (ret) { + dev_err(priv->dev, "Failed to init uccf."); + return ret; + } + + priv->uf_regs = priv->uccf->uf_regs; + ucc_fast_disable(priv->uccf, COMM_DIR_RX | COMM_DIR_TX); + + /* Loopback mode */ + if (priv->loopback) { + dev_info(priv->dev, "Loopback Mode\n"); + gumr = ioread32be(&priv->uf_regs->gumr); + gumr |= (UCC_FAST_GUMR_LOOPBACK | UCC_FAST_GUMR_CDS | + UCC_FAST_GUMR_TCI); + gumr &= ~(UCC_FAST_GUMR_CTSP | UCC_FAST_GUMR_RSYN); + iowrite32be(gumr, &priv->uf_regs->gumr); + } + + /* Initialize SI */ + if (priv->tsa) + ucc_tdm_init(priv->utdm, priv->ut_info); + + /* Write to QE CECR, UCCx channel to Stop Transmission */ + cecr_subblock = ucc_fast_get_qe_cr_subblock(uf_info->ucc_num); + ret = qe_issue_cmd(QE_STOP_TX, cecr_subblock, + QE_CR_PROTOCOL_UNSPECIFIED, 0); + + /* Set UPSMR normal mode (need fixed)*/ + iowrite32be(0, &priv->uf_regs->upsmr); + + priv->rx_ring_size = RX_BD_RING_LEN; + priv->tx_ring_size = TX_BD_RING_LEN; + /* Alloc Rx BD */ + priv->rx_bd_base = dma_alloc_coherent(priv->dev, + RX_BD_RING_LEN * sizeof(struct qe_bd *), + &priv->dma_rx_bd, GFP_KERNEL); + + if (!priv->rx_bd_base) { + dev_err(priv->dev, "Cannot allocate MURAM memory for RxBDs\n"); + ret = -ENOMEM; + goto rxbd_alloc_error; + } + + /* Alloc Tx BD */ + priv->tx_bd_base = dma_alloc_coherent(priv->dev, + TX_BD_RING_LEN * sizeof(struct qe_bd *), + &priv->dma_tx_bd, GFP_KERNEL); + + if (!priv->tx_bd_base) { + dev_err(priv->dev, "Cannot allocate MURAM memory for TxBDs\n"); + ret = -ENOMEM; + goto txbd_alloc_error; + } + + /* Alloc parameter ram for ucc hdlc */ + priv->ucc_pram_offset = qe_muram_alloc(sizeof(priv->ucc_pram), + ALIGNMENT_OF_UCC_HDLC_PRAM); + + if (priv->ucc_pram_offset < 0) { + dev_err(priv->dev, "Can not allocate MURAM for hdlc prameter.\n"); + ret = -ENOMEM; + goto pram_alloc_error; + } + + priv->rx_skbuff = kzalloc(priv->rx_ring_size * sizeof(*priv->rx_skbuff), + GFP_KERNEL); + if (!priv->rx_skbuff) + goto rx_skb_alloc_error; + + priv->tx_skbuff = kzalloc(priv->tx_ring_size * sizeof(*priv->tx_skbuff), + GFP_KERNEL); + if (!priv->tx_skbuff) + goto tx_skb_alloc_error; + + priv->skb_curtx = 0; + priv->skb_dirtytx = 0; + priv->curtx_bd = priv->tx_bd_base; + priv->dirty_tx = priv->tx_bd_base; + priv->currx_bd = priv->rx_bd_base; + priv->currx_bdnum = 0; + + /* init parameter base */ + cecr_subblock = ucc_fast_get_qe_cr_subblock(uf_info->ucc_num); + ret = qe_issue_cmd(QE_ASSIGN_PAGE_TO_DEVICE, cecr_subblock, + QE_CR_PROTOCOL_UNSPECIFIED, priv->ucc_pram_offset); + + priv->ucc_pram = (struct ucc_hdlc_param __iomem *) + qe_muram_addr(priv->ucc_pram_offset); + + /* Zero out parameter ram */ + memset_io(priv->ucc_pram, 0, sizeof(struct ucc_hdlc_param)); + + /* Alloc riptr, tiptr */ + riptr = qe_muram_alloc(32, 32); + if (riptr < 0) { + dev_err(priv->dev, "Cannot allocate MURAM mem for Receive internal temp data pointer\n"); + ret = -ENOMEM; + goto riptr_alloc_error; + } + + tiptr = qe_muram_alloc(32, 32); + if (tiptr < 0) { + dev_err(priv->dev, "Cannot allocate MURAM mem for Transmit internal temp data pointer\n"); + ret = -ENOMEM; + goto tiptr_alloc_error; + } + + /* Set RIPTR, TIPTR */ + iowrite16be(riptr, &priv->ucc_pram->riptr); + iowrite16be(tiptr, &priv->ucc_pram->tiptr); + + /* Set MRBLR */ + iowrite16be(MAX_RX_BUF_LENGTH, &priv->ucc_pram->mrblr); + + /* Set RBASE, TBASE */ + iowrite32be(priv->dma_rx_bd, &priv->ucc_pram->rbase); + iowrite32be(priv->dma_tx_bd, &priv->ucc_pram->tbase); + + /* Set RSTATE, TSTATE */ + iowrite32be(BMR_GBL | BMR_BIG_ENDIAN, &priv->ucc_pram->rstate); + iowrite32be(BMR_GBL | BMR_BIG_ENDIAN, &priv->ucc_pram->tstate); + + /* Set C_MASK, C_PRES for 16bit CRC */ + iowrite32be(CRC_16BIT_MASK, &priv->ucc_pram->c_mask); + iowrite32be(CRC_16BIT_PRES, &priv->ucc_pram->c_pres); + + iowrite16be(MAX_FRAME_LENGTH, &priv->ucc_pram->mflr); + iowrite16be(DEFAULT_RFTHR, &priv->ucc_pram->rfthr); + iowrite16be(DEFAULT_RFTHR, &priv->ucc_pram->rfcnt); + iowrite16be(DEFAULT_ADDR_MASK, &priv->ucc_pram->hmask); + iowrite16be(DEFAULT_HDLC_ADDR, &priv->ucc_pram->haddr1); + iowrite16be(DEFAULT_HDLC_ADDR, &priv->ucc_pram->haddr2); + iowrite16be(DEFAULT_HDLC_ADDR, &priv->ucc_pram->haddr3); + iowrite16be(DEFAULT_HDLC_ADDR, &priv->ucc_pram->haddr4); + + /* Get BD buffer */ + bd_buffer = dma_alloc_coherent(priv->dev, + (RX_BD_RING_LEN + TX_BD_RING_LEN) * + MAX_RX_BUF_LENGTH, + &bd_dma_addr, GFP_KERNEL); + + if (!bd_buffer) { + dev_err(priv->dev, "Could not allocate buffer descriptors\n"); + ret = -ENOMEM; + goto bd_alloc_error; + } + + memset(bd_buffer, 0, (RX_BD_RING_LEN + TX_BD_RING_LEN) + * MAX_RX_BUF_LENGTH); + + priv->rx_buffer = bd_buffer; + priv->tx_buffer = bd_buffer + RX_BD_RING_LEN * MAX_RX_BUF_LENGTH; + + priv->dma_rx_addr = bd_dma_addr; + priv->dma_tx_addr = bd_dma_addr + RX_BD_RING_LEN * MAX_RX_BUF_LENGTH; + + for (i = 0; i < RX_BD_RING_LEN; i++) { + if (i < (RX_BD_RING_LEN - 1)) + bd_status = R_E_S | R_I_S; + else + bd_status = R_E_S | R_I_S | R_W_S; + + iowrite16be(bd_status, &priv->rx_bd_base[i].status); + iowrite32be(priv->dma_rx_addr + i * MAX_RX_BUF_LENGTH, + &priv->rx_bd_base[i].buf); + } + + for (i = 0; i < TX_BD_RING_LEN; i++) { + if (i < (TX_BD_RING_LEN - 1)) + bd_status = T_I_S | T_TC_S; + else + bd_status = T_I_S | T_TC_S | T_W_S; + + iowrite16be(bd_status, &priv->tx_bd_base[i].status); + iowrite32be(priv->dma_tx_addr + i * MAX_RX_BUF_LENGTH, + &priv->tx_bd_base[i].buf); + } + + return 0; + +bd_alloc_error: + qe_muram_free(tiptr); +tiptr_alloc_error: + qe_muram_free(riptr); +riptr_alloc_error: + kfree(priv->tx_skbuff); +tx_skb_alloc_error: + kfree(priv->rx_skbuff); +rx_skb_alloc_error: + qe_muram_free(priv->ucc_pram_offset); +pram_alloc_error: + dma_free_coherent(priv->dev, + TX_BD_RING_LEN * sizeof(struct qe_bd), + priv->tx_bd_base, priv->dma_tx_bd); +txbd_alloc_error: + dma_free_coherent(priv->dev, + RX_BD_RING_LEN * sizeof(struct qe_bd), + priv->rx_bd_base, priv->dma_rx_bd); +rxbd_alloc_error: + ucc_fast_free(priv->uccf); + + return ret; +} + +static netdev_tx_t ucc_hdlc_tx(struct sk_buff *skb, struct net_device *dev) +{ + hdlc_device *hdlc = dev_to_hdlc(dev); + struct ucc_hdlc_private *priv = (struct ucc_hdlc_private *)hdlc->priv; + struct qe_bd __iomem *bd; + u16 bd_status; + unsigned long flags; + u8 *send_buf; + int i; + u16 *proto_head; + + switch (dev->type) { + case ARPHRD_RAWHDLC: + if (skb_headroom(skb) < HDLC_HEAD_LEN) { + dev->stats.tx_dropped++; + dev_kfree_skb(skb); + netdev_err(dev, "No enough space for hdlc head\n"); + return -ENOMEM; + } + + skb_push(skb, HDLC_HEAD_LEN); + + proto_head = (u16 *)skb->data; + *proto_head = htons(DEFAULT_HDLC_HEAD); + + dev->stats.tx_bytes += skb->len; + break; + + case ARPHRD_PPP: + proto_head = (u16 *)skb->data; + if (*proto_head != htons(DEFAULT_PPP_HEAD)) { + dev->stats.tx_dropped++; + dev_kfree_skb(skb); + netdev_err(dev, "Wrong ppp header\n"); + return -ENOMEM; + } + + dev->stats.tx_bytes += skb->len; + break; + + default: + dev->stats.tx_dropped++; + dev_kfree_skb(skb); + return -ENOMEM; + } + + pr_info("Tx data skb->len:%d ", skb->len); + send_buf = (u8 *)skb->data; + pr_info("\nTransmitted data:\n"); + for (i = 0; i < 16; i++) { + if (i == skb->len) + pr_info("++++"); + else + pr_info("%02x\n", send_buf[i]); + } + spin_lock_irqsave(&priv->lock, flags); + + /* Start from the next BD that should be filled */ + bd = priv->curtx_bd; + bd_status = ioread16be(&bd->status); + /* Save the skb pointer so we can free it later */ + priv->tx_skbuff[priv->skb_curtx] = skb; + + /* Update the current skb pointer (wrapping if this was the last) */ + priv->skb_curtx = + (priv->skb_curtx + 1) & TX_RING_MOD_MASK(TX_BD_RING_LEN); + + /* copy skb data to tx buffer for sdma processing */ + memcpy(priv->tx_buffer + (be32_to_cpu(bd->buf) - priv->dma_tx_addr), + skb->data, skb->len); + + /* set bd status and length */ + bd_status = (bd_status & T_W_S) | T_R_S | T_I_S | T_L_S | T_TC_S; + + iowrite16be(bd_status, &bd->status); + iowrite16be(skb->len, &bd->length); + + /* Move to next BD in the ring */ + if (!(bd_status & T_W_S)) + bd += 1; + else + bd = priv->tx_bd_base; + + if (bd == priv->dirty_tx) { + if (!netif_queue_stopped(dev)) + netif_stop_queue(dev); + } + + priv->curtx_bd = bd; + + spin_unlock_irqrestore(&priv->lock, flags); + + return NETDEV_TX_OK; +} + +static int hdlc_tx_done(struct ucc_hdlc_private *priv) +{ + /* Start from the next BD that should be filled */ + struct net_device *dev = priv->ndev; + struct qe_bd *bd; /* BD pointer */ + u16 bd_status; + + bd = priv->dirty_tx; + bd_status = ioread16be(&bd->status); + + /* Normal processing. */ + while ((bd_status & T_R_S) == 0) { + struct sk_buff *skb; + + /* BD contains already transmitted buffer. */ + /* Handle the transmitted buffer and release */ + /* the BD to be used with the current frame */ + + skb = priv->tx_skbuff[priv->skb_dirtytx]; + if (!skb) + break; + pr_info("TxBD: %x\n", bd_status); + dev->stats.tx_packets++; + memset(priv->tx_buffer + + (be32_to_cpu(bd->buf) - priv->dma_tx_addr), + 0, skb->len); + dev_kfree_skb_irq(skb); + + priv->tx_skbuff[priv->skb_dirtytx] = NULL; + priv->skb_dirtytx = + (priv->skb_dirtytx + + 1) & TX_RING_MOD_MASK(TX_BD_RING_LEN); + + /* We freed a buffer, so now we can restart transmission */ + if (netif_queue_stopped(dev)) + netif_wake_queue(dev); + + /* Advance the confirmation BD pointer */ + if (!(bd_status & T_W_S)) + bd += 1; + else + bd = priv->tx_bd_base; + bd_status = ioread16be(&bd->status); + } + priv->dirty_tx = bd; + + return 0; +} + +static int hdlc_rx_done(struct ucc_hdlc_private *priv, int rx_work_limit) +{ + struct net_device *dev = priv->ndev; + struct sk_buff *skb; + hdlc_device *hdlc = dev_to_hdlc(dev); + struct qe_bd *bd; + u32 bd_status; + u16 length, howmany = 0; + u8 *bdbuffer; + int i; + static int entry; + + bd = priv->currx_bd; + bd_status = ioread16be(&bd->status); + + /* while there are received buffers and BD is full (~R_E) */ + while (!((bd_status & (R_E_S)) || (--rx_work_limit < 0))) { + if (bd_status & R_OV_S) + dev->stats.rx_over_errors++; + if (bd_status & R_CR_S) { +#ifdef BROKEN_FRAME_INFO + pr_info("Broken Frame with RxBD: %x\n", bd_status); +#endif + dev->stats.rx_crc_errors++; + dev->stats.rx_dropped++; + goto recycle; + } + bdbuffer = priv->rx_buffer + + (priv->currx_bdnum * MAX_RX_BUF_LENGTH); + length = ioread16be(&bd->length); + + pr_info("Received data length:%d", length); + pr_info("while entry times:%d", entry++); + + pr_info("\nReceived data:\n"); + for (i = 0; (i < 16); i++) { + if (i == length) + pr_info("++++"); + else + pr_info("%02x\n", bdbuffer[i]); + } + + switch (dev->type) { + case ARPHRD_RAWHDLC: + bdbuffer += HDLC_HEAD_LEN; + length -= (HDLC_HEAD_LEN + HDLC_CRC_SIZE); + + skb = dev_alloc_skb(length); + if (!skb) { + dev->stats.rx_dropped++; + return -ENOMEM; + } + + skb_put(skb, length); + skb->len = length; + skb->dev = dev; + memcpy(skb->data, bdbuffer, length); + break; + + case ARPHRD_PPP: + length -= HDLC_CRC_SIZE; + + skb = dev_alloc_skb(length); + if (!skb) { + dev->stats.rx_dropped++; + return -ENOMEM; + } + + skb_put(skb, length); + skb->len = length; + skb->dev = dev; + memcpy(skb->data, bdbuffer, length); + break; + } + + dev->stats.rx_packets++; + dev->stats.rx_bytes += skb->len; + howmany++; + if (hdlc->proto) + skb->protocol = hdlc_type_trans(skb, dev); + pr_info("skb->protocol:%x\n", skb->protocol); + netif_receive_skb(skb); + +recycle: + iowrite16be(bd_status | R_E_S | R_I_S, &bd->status); + + /* update to point at the next bd */ + if (bd_status & R_W_S) { + priv->currx_bdnum = 0; + bd = priv->rx_bd_base; + } else { + if (priv->currx_bdnum < (RX_BD_RING_LEN - 1)) + priv->currx_bdnum += 1; + else + priv->currx_bdnum = RX_BD_RING_LEN - 1; + + bd += 1; + } + + bd_status = ioread16be(&bd->status); + } + + priv->currx_bd = bd; + return howmany; +} + +static int ucc_hdlc_poll(struct napi_struct *napi, int budget) +{ + struct ucc_hdlc_private *priv = container_of(napi, + struct ucc_hdlc_private, + napi); + int howmany; + + /* Tx event processing */ + spin_lock(&priv->lock); + hdlc_tx_done(priv); + spin_unlock(&priv->lock); + + howmany = 0; + howmany += hdlc_rx_done(priv, budget - howmany); + + if (howmany < budget) { + napi_complete(napi); + qe_setbits32(priv->uccf->p_uccm, + (UCCE_HDLC_RX_EVENTS | UCCE_HDLC_TX_EVENTS) << 16); + } + + return howmany; +} + +static irqreturn_t ucc_hdlc_irq_handler(int irq, void *dev_id) +{ + struct ucc_hdlc_private *priv = (struct ucc_hdlc_private *)dev_id; + struct net_device *dev = priv->ndev; + struct ucc_fast_private *uccf; + struct ucc_tdm_info *ut_info; + u32 ucce; + u32 uccm; + + ut_info = priv->ut_info; + uccf = priv->uccf; + + ucce = ioread32be(uccf->p_ucce); + uccm = ioread32be(uccf->p_uccm); + ucce &= uccm; + iowrite32be(ucce, uccf->p_ucce); + pr_info("irq ucce:%x\n", ucce); + if (!ucce) + return IRQ_NONE; + + if ((ucce >> 16) & (UCCE_HDLC_RX_EVENTS | UCCE_HDLC_TX_EVENTS)) { + if (napi_schedule_prep(&priv->napi)) { + uccm &= ~((UCCE_HDLC_RX_EVENTS | UCCE_HDLC_TX_EVENTS) + << 16); + iowrite32be(uccm, uccf->p_uccm); + __napi_schedule(&priv->napi); + } + } + + /* Errors and other events */ + if (ucce >> 16 & UCC_HDLC_UCCE_BSY) + dev->stats.rx_errors++; + if (ucce >> 16 & UCC_HDLC_UCCE_TXE) + dev->stats.tx_errors++; + + return IRQ_HANDLED; +} + +static int uhdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) +{ + const size_t size = sizeof(te1_settings); + te1_settings line; + struct ucc_hdlc_private *priv = netdev_priv(dev); + + if (cmd != SIOCWANDEV) + return hdlc_ioctl(dev, ifr, cmd); + + switch (ifr->ifr_settings.type) { + case IF_GET_IFACE: + ifr->ifr_settings.type = IF_IFACE_E1; + if (ifr->ifr_settings.size < size) { + ifr->ifr_settings.size = size; /* data size wanted */ + return -ENOBUFS; + } + line.clock_type = priv->clocking; + line.clock_rate = 0; + line.loopback = 0; + + if (copy_to_user(ifr->ifr_settings.ifs_ifsu.sync, &line, size)) + return -EFAULT; + return 0; + + default: + return hdlc_ioctl(dev, ifr, cmd); + } +} + +static int uhdlc_open(struct net_device *dev) +{ + u32 cecr_subblock; + hdlc_device *hdlc = dev_to_hdlc(dev); + struct ucc_hdlc_private *priv = hdlc->priv; + struct ucc_tdm *utdm = priv->utdm; + + if (priv->hdlc_busy != 1) { + if (request_irq(priv->ut_info->uf_info.irq, + ucc_hdlc_irq_handler, 0, "hdlc", priv)) + return -ENODEV; + + cecr_subblock = ucc_fast_get_qe_cr_subblock( + priv->ut_info->uf_info.ucc_num); + + qe_issue_cmd(QE_INIT_TX_RX, cecr_subblock, + QE_CR_PROTOCOL_UNSPECIFIED, 0); + + ucc_fast_enable(priv->uccf, COMM_DIR_RX | COMM_DIR_TX); + + /* Enable the TDM port */ + if (priv->tsa) + utdm->si_regs->siglmr1_h |= (0x1 << utdm->tdm_port); + + priv->hdlc_busy = 1; + netif_device_attach(priv->ndev); + napi_enable(&priv->napi); + netif_start_queue(dev); + hdlc_open(dev); + } + + return 0; +} + +static void uhdlc_memclean(struct ucc_hdlc_private *priv) +{ + qe_muram_free(priv->ucc_pram->riptr); + qe_muram_free(priv->ucc_pram->tiptr); + + if (priv->rx_bd_base) { + dma_free_coherent(priv->dev, + RX_BD_RING_LEN * sizeof(struct qe_bd), + priv->rx_bd_base, priv->dma_rx_bd); + + priv->rx_bd_base = NULL; + priv->dma_rx_bd = 0; + } + + if (priv->tx_bd_base) { + dma_free_coherent(priv->dev, + TX_BD_RING_LEN * sizeof(struct qe_bd), + priv->tx_bd_base, priv->dma_tx_bd); + + priv->tx_bd_base = NULL; + priv->dma_tx_bd = 0; + } + + if (priv->ucc_pram) { + qe_muram_free(priv->ucc_pram_offset); + priv->ucc_pram = NULL; + priv->ucc_pram_offset = 0; + } + + kfree(priv->rx_skbuff); + priv->rx_skbuff = NULL; + + kfree(priv->tx_skbuff); + priv->tx_skbuff = NULL; + + if (priv->uf_regs) { + iounmap(priv->uf_regs); + priv->uf_regs = NULL; + } + + if (priv->uccf) { + ucc_fast_free(priv->uccf); + priv->uccf = NULL; + } + + if (priv->rx_buffer) { + dma_free_coherent(priv->dev, + RX_BD_RING_LEN * MAX_RX_BUF_LENGTH, + priv->rx_buffer, priv->dma_rx_addr); + priv->rx_buffer = NULL; + priv->dma_rx_addr = 0; + } + + if (priv->tx_buffer) { + dma_free_coherent(priv->dev, + TX_BD_RING_LEN * MAX_RX_BUF_LENGTH, + priv->tx_buffer, priv->dma_tx_addr); + priv->tx_buffer = NULL; + priv->dma_tx_addr = 0; + } +} + +static int uhdlc_close(struct net_device *dev) +{ + struct ucc_hdlc_private *priv = dev_to_hdlc(dev)->priv; + struct ucc_tdm *utdm = priv->utdm; + u32 cecr_subblock; + + napi_disable(&priv->napi); + cecr_subblock = ucc_fast_get_qe_cr_subblock( + priv->ut_info->uf_info.ucc_num); + + qe_issue_cmd(QE_GRACEFUL_STOP_TX, cecr_subblock, + (u8)QE_CR_PROTOCOL_UNSPECIFIED, 0); + qe_issue_cmd(QE_CLOSE_RX_BD, cecr_subblock, + (u8)QE_CR_PROTOCOL_UNSPECIFIED, 0); + + if (priv->tsa) + utdm->si_regs->siglmr1_h &= ~(0x1 << utdm->tdm_port); + + ucc_fast_disable(priv->uccf, COMM_DIR_RX | COMM_DIR_TX); + + free_irq(priv->ut_info->uf_info.irq, priv); + netif_stop_queue(dev); + priv->hdlc_busy = 0; + + return 0; +} + +static int ucc_hdlc_attach(struct net_device *dev, unsigned short encoding, + unsigned short parity) +{ + struct ucc_hdlc_private *priv = dev_to_hdlc(dev)->priv; + + if (encoding != ENCODING_NRZ && + encoding != ENCODING_NRZI) + return -EINVAL; + + if (parity != PARITY_NONE && + parity != PARITY_CRC32_PR1_CCITT && + parity != PARITY_CRC16_PR1_CCITT) + return -EINVAL; + + priv->encoding = encoding; + priv->parity = parity; + + return 0; +} + +#ifdef CONFIG_PM +static void store_clk_config(struct ucc_hdlc_private *priv) +{ + struct qe_mux *qe_mux_reg = &qe_immr->qmx; + + /* store si clk */ + priv->cmxsi1cr_h = ioread32be(&qe_mux_reg->cmxsi1cr_h); + priv->cmxsi1cr_l = ioread32be(&qe_mux_reg->cmxsi1cr_l); + + /* store si sync */ + priv->cmxsi1syr = ioread32be(&qe_mux_reg->cmxsi1syr); + + /* store ucc clk */ + memcpy_fromio(priv->cmxucr, qe_mux_reg->cmxucr, 4 * sizeof(u32)); +} + +static void resume_clk_config(struct ucc_hdlc_private *priv) +{ + struct qe_mux *qe_mux_reg = &qe_immr->qmx; + + memcpy_toio(qe_mux_reg->cmxucr, priv->cmxucr, 4 * sizeof(u32)); + + iowrite32be(priv->cmxsi1cr_h, &qe_mux_reg->cmxsi1cr_h); + iowrite32be(priv->cmxsi1cr_l, &qe_mux_reg->cmxsi1cr_l); + + iowrite32be(priv->cmxsi1syr, &qe_mux_reg->cmxsi1syr); +} + +static int uhdlc_suspend(struct device *dev) +{ + struct ucc_hdlc_private *priv = dev_get_drvdata(dev); + struct ucc_tdm_info *ut_info; + struct ucc_fast __iomem *uf_regs; + + if (!priv) + return -EINVAL; + + if (!netif_running(priv->ndev)) + return 0; + + netif_device_detach(priv->ndev); + napi_disable(&priv->napi); + + ut_info = priv->ut_info; + uf_regs = priv->uf_regs; + + /* backup gumr guemr*/ + priv->gumr = ioread32be(&uf_regs->gumr); + priv->guemr = ioread8(&uf_regs->guemr); + + priv->ucc_pram_bak = kmalloc(sizeof(*priv->ucc_pram_bak), + GFP_KERNEL); + if (!priv->ucc_pram_bak) + return -ENOMEM; + + /* backup HDLC parameter */ + memcpy_fromio(priv->ucc_pram_bak, priv->ucc_pram, + sizeof(struct ucc_hdlc_param)); + + /* store the clk configuration */ + store_clk_config(priv); + + /* save power */ + ucc_fast_disable(priv->uccf, COMM_DIR_RX | COMM_DIR_TX); + + dev_dbg(dev, "ucc hdlc suspend\n"); + return 0; +} + +static int uhdlc_resume(struct device *dev) +{ + struct ucc_hdlc_private *priv = dev_get_drvdata(dev); + struct ucc_tdm *utdm = priv->utdm; + struct ucc_tdm_info *ut_info; + struct ucc_fast __iomem *uf_regs; + struct ucc_fast_private *uccf; + struct ucc_fast_info *uf_info; + int ret, i; + u32 cecr_subblock; + u16 bd_status; + + if (!priv) + return -EINVAL; + + if (!netif_running(priv->ndev)) + return 0; + + ut_info = priv->ut_info; + uf_info = &ut_info->uf_info; + uf_regs = priv->uf_regs; + uccf = priv->uccf; + + /* restore gumr guemr */ + iowrite8(priv->guemr, &uf_regs->guemr); + iowrite32be(priv->gumr, &uf_regs->gumr); + + /* Set Virtual Fifo registers */ + iowrite16be(uf_info->urfs, &uf_regs->urfs); + iowrite16be(uf_info->urfet, &uf_regs->urfet); + iowrite16be(uf_info->urfset, &uf_regs->urfset); + iowrite16be(uf_info->utfs, &uf_regs->utfs); + iowrite16be(uf_info->utfet, &uf_regs->utfet); + iowrite16be(uf_info->utftt, &uf_regs->utftt); + /* utfb, urfb are offsets from MURAM base */ + iowrite32be(uccf->ucc_fast_tx_virtual_fifo_base_offset, &uf_regs->utfb); + iowrite32be(uccf->ucc_fast_rx_virtual_fifo_base_offset, &uf_regs->urfb); + + /* Rx Tx and sync clock routing */ + resume_clk_config(priv); + + iowrite32be(uf_info->uccm_mask, &uf_regs->uccm); + iowrite32be(0xffffffff, &uf_regs->ucce); + + ucc_fast_disable(priv->uccf, COMM_DIR_RX | COMM_DIR_TX); + + /* rebuild SIRAM */ + if (priv->tsa) + ucc_tdm_init(priv->utdm, priv->ut_info); + + /* Write to QE CECR, UCCx channel to Stop Transmission */ + cecr_subblock = ucc_fast_get_qe_cr_subblock(uf_info->ucc_num); + ret = qe_issue_cmd(QE_STOP_TX, cecr_subblock, + (u8)QE_CR_PROTOCOL_UNSPECIFIED, 0); + + /* Set UPSMR normal mode */ + iowrite32be(0, &uf_regs->upsmr); + + /* init parameter base */ + cecr_subblock = ucc_fast_get_qe_cr_subblock(uf_info->ucc_num); + ret = qe_issue_cmd(QE_ASSIGN_PAGE_TO_DEVICE, cecr_subblock, + QE_CR_PROTOCOL_UNSPECIFIED, priv->ucc_pram_offset); + + priv->ucc_pram = (struct ucc_hdlc_param __iomem *) + qe_muram_addr(priv->ucc_pram_offset); + + /* restore ucc parameter */ + memcpy_toio(priv->ucc_pram, priv->ucc_pram_bak, + sizeof(struct ucc_hdlc_param)); + kfree(priv->ucc_pram_bak); + + /* rebuild BD entry */ + for (i = 0; i < RX_BD_RING_LEN; i++) { + if (i < (RX_BD_RING_LEN - 1)) + bd_status = R_E_S | R_I_S; + else + bd_status = R_E_S | R_I_S | R_W_S; + + iowrite16be(bd_status, &priv->rx_bd_base[i].status); + iowrite32be(priv->dma_rx_addr + i * MAX_RX_BUF_LENGTH, + &priv->rx_bd_base[i].buf); + } + + for (i = 0; i < TX_BD_RING_LEN; i++) { + if (i < (TX_BD_RING_LEN - 1)) + bd_status = T_I_S | T_TC_S; + else + bd_status = T_I_S | T_TC_S | T_W_S; + + iowrite16be(bd_status, &priv->tx_bd_base[i].status); + iowrite32be(priv->dma_tx_addr + i * MAX_RX_BUF_LENGTH, + &priv->tx_bd_base[i].buf); + } + + /* if hdlc is busy enable TX and RX */ + if (priv->hdlc_busy == 1) { + cecr_subblock = ucc_fast_get_qe_cr_subblock( + priv->ut_info->uf_info.ucc_num); + + qe_issue_cmd(QE_INIT_TX_RX, cecr_subblock, + (u8)QE_CR_PROTOCOL_UNSPECIFIED, 0); + + ucc_fast_enable(priv->uccf, COMM_DIR_RX | COMM_DIR_TX); + + /* Enable the TDM port */ + if (priv->tsa) + utdm->si_regs->siglmr1_h |= (0x1 << utdm->tdm_port); + } + + napi_enable(&priv->napi); + netif_device_attach(priv->ndev); + + return 0; +} + +static const struct dev_pm_ops uhdlc_pm_ops = { + .suspend = uhdlc_suspend, + .resume = uhdlc_resume, + .freeze = uhdlc_suspend, + .thaw = uhdlc_resume, +}; + +#define HDLC_PM_OPS (&uhdlc_pm_ops) + +#else + +#define HDLC_PM_OPS NULL + +#endif +static const struct net_device_ops uhdlc_ops = { + .ndo_open = uhdlc_open, + .ndo_stop = uhdlc_close, + .ndo_change_mtu = hdlc_change_mtu, + .ndo_start_xmit = hdlc_start_xmit, + .ndo_do_ioctl = uhdlc_ioctl, +}; + +static int ucc_hdlc_probe(struct platform_device *pdev) +{ + struct device_node *np = pdev->dev.of_node; + struct ucc_hdlc_private *uhdlc_priv = NULL; + struct ucc_tdm_info *ut_info; + struct ucc_tdm *utdm; + struct resource res; + struct net_device *dev; + hdlc_device *hdlc; + int ucc_num; + const char *sprop; + int ret; + u32 val; + + ret = of_property_read_u32_index(np, "cell-index", 0, &val); + if (ret) { + dev_err(&pdev->dev, "Invalid ucc property\n"); + return -ENODEV; + } + + ucc_num = val - 1; + if ((ucc_num > 3) || (ucc_num < 0)) { + dev_err(&pdev->dev, ": Invalid UCC num\n"); + return -EINVAL; + } + + memcpy(&utdm_info[ucc_num], &utdm_primary_info, + sizeof(utdm_primary_info)); + + ut_info = &utdm_info[ucc_num]; + ut_info->uf_info.ucc_num = ucc_num; + + sprop = of_get_property(np, "rx-clock-name", NULL); + if (sprop) { + ut_info->uf_info.rx_clock = qe_clock_source(sprop); + if ((ut_info->uf_info.rx_clock < QE_CLK_NONE) || + (ut_info->uf_info.rx_clock > QE_CLK24)) { + dev_err(&pdev->dev, "Invalid rx-clock-name property\n"); + return -EINVAL; + } + } else { + dev_err(&pdev->dev, "Invalid rx-clock-name property\n"); + return -EINVAL; + } + + sprop = of_get_property(np, "tx-clock-name", NULL); + if (sprop) { + ut_info->uf_info.tx_clock = qe_clock_source(sprop); + if ((ut_info->uf_info.tx_clock < QE_CLK_NONE) || + (ut_info->uf_info.tx_clock > QE_CLK24)) { + dev_err(&pdev->dev, "Invalid tx-clock-name property\n"); + return -EINVAL; + } + } else { + dev_err(&pdev->dev, "Invalid tx-clock-name property\n"); + return -EINVAL; + } + + /* use the same clock when work in loopback */ + if (ut_info->uf_info.rx_clock == ut_info->uf_info.tx_clock) + qe_setbrg(ut_info->uf_info.rx_clock, 20000000, 1); + + ret = of_address_to_resource(np, 0, &res); + if (ret) + return -EINVAL; + + ut_info->uf_info.regs = res.start; + ut_info->uf_info.irq = irq_of_parse_and_map(np, 0); + + uhdlc_priv = kzalloc(sizeof(*uhdlc_priv), GFP_KERNEL); + if (!uhdlc_priv) { + ret = -ENOMEM; + dev_err(&pdev->dev, "No mem to alloc hdlc private data\n"); + goto err_alloc_priv; + } + + dev_set_drvdata(&pdev->dev, uhdlc_priv); + uhdlc_priv->dev = &pdev->dev; + uhdlc_priv->ut_info = ut_info; + + if (of_get_property(np, "fsl,tdm-interface", NULL)) + uhdlc_priv->tsa = 1; + + if (of_get_property(np, "fsl,ucc-internal-loopback", NULL)) + uhdlc_priv->loopback = 1; + + if (uhdlc_priv->tsa == 1) { + utdm = kzalloc(sizeof(*utdm), GFP_KERNEL); + if (!utdm) { + ret = -ENOMEM; + dev_err(&pdev->dev, "No mem to alloc ucc tdm data\n"); + goto err_alloc_utdm; + } + uhdlc_priv->utdm = utdm; + ret = ucc_of_parse_tdm(np, utdm, ut_info); + if (ret) + goto err_miss_tsa_property; + } + + ret = uhdlc_init(uhdlc_priv); + if (ret) { + dev_err(&pdev->dev, "Failed to init uhdlc\n"); + goto err_hdlc_init; + } + + dev = alloc_hdlcdev(uhdlc_priv); + if (!dev) { + ret = -ENOMEM; + pr_err("ucc_hdlc: unable to allocate memory\n"); + goto err_hdlc_init; + } + + uhdlc_priv->ndev = dev; + hdlc = dev_to_hdlc(dev); + dev->tx_queue_len = 16; + dev->netdev_ops = &uhdlc_ops; + hdlc->attach = ucc_hdlc_attach; + hdlc->xmit = ucc_hdlc_tx; + netif_napi_add(dev, &uhdlc_priv->napi, ucc_hdlc_poll, 32); + if (register_hdlc_device(dev)) { + ret = -ENOBUFS; + pr_err("ucc_hdlc: unable to register hdlc device\n"); + free_netdev(dev); + goto err_hdlc_init; + } + + return 0; + +err_hdlc_init: +err_miss_tsa_property: + kfree(uhdlc_priv); + if (uhdlc_priv->tsa) + kfree(utdm); +err_alloc_utdm: + kfree(uhdlc_priv); +err_alloc_priv: + return ret; +} + +static int ucc_hdlc_remove(struct platform_device *pdev) +{ + struct ucc_hdlc_private *priv = dev_get_drvdata(&pdev->dev); + + uhdlc_memclean(priv); + + if (priv->utdm->si_regs) { + iounmap(priv->utdm->si_regs); + priv->utdm->si_regs = NULL; + } + + if (priv->utdm->siram) { + iounmap(priv->utdm->siram); + priv->utdm->siram = NULL; + } + kfree(priv); + + dev_info(&pdev->dev, "UCC based hdlc module removed\n"); + + return 0; +} + +static const struct of_device_id fsl_ucc_hdlc_of_match[] = { + { + .compatible = "fsl,ucc-hdlc", + }, + {}, +}; + +MODULE_DEVICE_TABLE(of, fsl_ucc_hdlc_of_match); + +static struct platform_driver ucc_hdlc_driver = { + .probe = ucc_hdlc_probe, + .remove = ucc_hdlc_remove, + .driver = { + .owner = THIS_MODULE, + .name = DRV_NAME, + .pm = HDLC_PM_OPS, + .of_match_table = fsl_ucc_hdlc_of_match, + }, +}; + +static int __init ucc_hdlc_init(void) +{ + return platform_driver_register(&ucc_hdlc_driver); +} + +static void __exit ucc_hdlc_exit(void) +{ + platform_driver_unregister(&ucc_hdlc_driver); +} + +module_init(ucc_hdlc_init); +module_exit(ucc_hdlc_exit); diff --git a/drivers/net/wan/fsl_ucc_hdlc.h b/drivers/net/wan/fsl_ucc_hdlc.h new file mode 100644 index 0000000..525786a --- /dev/null +++ b/drivers/net/wan/fsl_ucc_hdlc.h @@ -0,0 +1,147 @@ +/* Freescale QUICC Engine HDLC Device Driver + * + * Copyright 2014 Freescale Semiconductor Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +#ifndef CONFIG_UCC_HDLC_H +#define CONFIG_UCC_HDLC_H + +#include +#include + +#include +#include + +#include +#include + +/* UCC HDLC event register */ +#define UCCE_HDLC_RX_EVENTS \ +(UCC_HDLC_UCCE_RXF | UCC_HDLC_UCCE_RXB | UCC_HDLC_UCCE_BSY) +#define UCCE_HDLC_TX_EVENTS (UCC_HDLC_UCCE_TXB | UCC_HDLC_UCCE_TXE) + +struct ucc_hdlc_param { + __be16 riptr; + __be16 tiptr; + __be16 res0; + __be16 mrblr; + __be32 rstate; + __be32 rbase; + __be16 rbdstat; + __be16 rbdlen; + __be32 rdptr; + __be32 tstate; + __be32 tbase; + __be16 tbdstat; + __be16 tbdlen; + __be32 tdptr; + __be32 rbptr; + __be32 tbptr; + __be32 rcrc; + __be32 res1; + __be32 tcrc; + __be32 res2; + __be32 res3; + __be32 c_mask; + __be32 c_pres; + __be16 disfc; + __be16 crcec; + __be16 abtsc; + __be16 nmarc; + __be32 max_cnt; + __be16 mflr; + __be16 rfthr; + __be16 rfcnt; + __be16 hmask; + __be16 haddr1; + __be16 haddr2; + __be16 haddr3; + __be16 haddr4; + __be16 ts_tmp; + __be16 tmp_mb; +}; + +struct ucc_hdlc_private { + struct ucc_tdm *utdm; + struct ucc_tdm_info *ut_info; + struct ucc_fast_private *uccf; + struct device *dev; + struct net_device *ndev; + struct napi_struct napi; + struct ucc_fast __iomem *uf_regs; /* UCC Fast registers */ + struct ucc_hdlc_param __iomem *ucc_pram; + u16 tsa; + bool hdlc_busy; + bool loopback; + + u8 *tx_buffer; + u8 *rx_buffer; + dma_addr_t dma_tx_addr; + dma_addr_t dma_rx_addr; + + struct qe_bd *tx_bd_base; + struct qe_bd *rx_bd_base; + dma_addr_t dma_tx_bd; + dma_addr_t dma_rx_bd; + struct qe_bd *curtx_bd; + struct qe_bd *currx_bd; + struct qe_bd *dirty_tx; + u16 currx_bdnum; + + struct sk_buff **tx_skbuff; + struct sk_buff **rx_skbuff; + u16 skb_curtx; + u16 skb_currx; + unsigned short skb_dirtytx; + + unsigned short tx_ring_size; + unsigned short rx_ring_size; + u32 ucc_pram_offset; + + unsigned short encoding; + unsigned short parity; + u32 clocking; + spinlock_t lock; /* lock for Tx BD and Tx buffer */ +#ifdef CONFIG_PM + struct ucc_hdlc_param *ucc_pram_bak; + u32 gumr; + u8 guemr; + u32 cmxsi1cr_l, cmxsi1cr_h; + u32 cmxsi1syr; + u32 cmxucr[4]; +#endif +}; + +#define TX_BD_RING_LEN 0x10 +#define RX_BD_RING_LEN 0x20 +#define RX_CLEAN_MAX 0x10 +#define NUM_OF_BUF 4 +#define MAX_RX_BUF_LENGTH (48 * 0x20) +#define MAX_FRAME_LENGTH (MAX_RX_BUF_LENGTH + 8) +#define ALIGNMENT_OF_UCC_HDLC_PRAM 64 +#define SI_BANK_SIZE 128 +#define MAX_HDLC_NUM 4 +#define HDLC_HEAD_LEN 2 +#define HDLC_CRC_SIZE 2 +#define TX_RING_MOD_MASK(size) (size - 1) +#define RX_RING_MOD_MASK(size) (size - 1) + +#define HDLC_HEAD_MASK 0x0000 +#define DEFAULT_HDLC_HEAD 0xff44 +#define DEFAULT_ADDR_MASK 0x00ff +#define DEFAULT_HDLC_ADDR 0x00ff + +#define BMR_GBL 0x20000000 +#define BMR_BIG_ENDIAN 0x10000000 +#define CRC_16BIT_MASK 0x0000F0B8 +#define CRC_16BIT_PRES 0x0000FFFF +#define DEFAULT_RFTHR 1 + +#define DEFAULT_PPP_HEAD 0xff03 + +#endif diff --git a/include/soc/fsl/qe/qe.h b/include/soc/fsl/qe/qe.h index c3b1dc8..70339d7 100644 --- a/include/soc/fsl/qe/qe.h +++ b/include/soc/fsl/qe/qe.h @@ -657,6 +657,7 @@ struct ucc_slow_pram { #define UCC_SLOW_GUMR_L_MODE_QMC 0x00000002 /* General UCC FAST Mode Register */ +#define UCC_FAST_GUMR_LOOPBACK 0x40000000 #define UCC_FAST_GUMR_TCI 0x20000000 #define UCC_FAST_GUMR_TRX 0x10000000 #define UCC_FAST_GUMR_TTX 0x08000000 diff --git a/include/soc/fsl/qe/ucc_fast.h b/include/soc/fsl/qe/ucc_fast.h index e898895..3ee9e7c 100644 --- a/include/soc/fsl/qe/ucc_fast.h +++ b/include/soc/fsl/qe/ucc_fast.h @@ -21,19 +21,37 @@ #include -/* Receive BD's status */ +/* Receive BD's status and length*/ #define R_E 0x80000000 /* buffer empty */ #define R_W 0x20000000 /* wrap bit */ #define R_I 0x10000000 /* interrupt on reception */ #define R_L 0x08000000 /* last */ #define R_F 0x04000000 /* first */ -/* transmit BD's status */ +/* transmit BD's status and length*/ #define T_R 0x80000000 /* ready bit */ #define T_W 0x20000000 /* wrap bit */ #define T_I 0x10000000 /* interrupt on completion */ #define T_L 0x08000000 /* last */ +/* Receive BD's status */ +#define R_E_S 0x8000 /* buffer empty */ +#define R_W_S 0x2000 /* wrap bit */ +#define R_I_S 0x1000 /* interrupt on reception */ +#define R_L_S 0x0800 /* last */ +#define R_F_S 0x0400 /* first */ +#define R_CM_S 0x0200 /* continuous mode */ +#define R_CR_S 0x0004 /* crc */ +#define R_OV_S 0x0002 /* crc */ + +/* transmit BD's status */ +#define T_R_S 0x8000 /* ready bit */ +#define T_W_S 0x2000 /* wrap bit */ +#define T_I_S 0x1000 /* interrupt on completion */ +#define T_L_S 0x0800 /* last */ +#define T_TC_S 0x0400 /* crc */ +#define T_TM_S 0x0200 /* continuous mode */ + /* Rx Data buffer must be 4 bytes aligned in most cases */ #define UCC_FAST_RX_ALIGN 4 #define UCC_FAST_MRBLR_ALIGNMENT 4 -- cgit v0.10.2 From b9263cbf217433c1249c7341f8c0a9b4a5df6753 Mon Sep 17 00:00:00 2001 From: Suresh Reddy Date: Mon, 6 Jun 2016 07:22:08 -0400 Subject: be2net: use max-TXQs limit too while provisioning VF queue pairs When the PF driver provisions resources for VFs, it currently only looks at max RSS queues available to calculate the number of VF queue pairs. This logic breaks when there are less number of TX-queues than RSS-queues. This patch fixes this problem by using the max-TXQs available in the PF-pool in the calculations. As a part of this change the be_calculate_vf_qs() routine is renamed as be_calculate_vf_res() and the code that calculates limits on other related resources is moved here to contain all resource calculation code inside one routine. Signed-off-by: Suresh Reddy Signed-off-by: Sathya Perla Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index fe3763d..ee0bdb7 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -444,6 +444,7 @@ struct be_resources { u16 max_evt_qs; u32 if_cap_flags; u32 vf_if_cap_flags; /* VF if capability flags */ + u32 flags; }; #define be_is_os2bmc_enabled(adapter) (adapter->flags & BE_FLAGS_OS2BMC) diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index 22402db..bef92c4 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -4465,7 +4465,7 @@ static int be_cmd_set_profile_config(struct be_adapter *adapter, void *desc, } /* Mark all fields invalid */ -static void be_reset_nic_desc(struct be_nic_res_desc *nic) +void be_reset_nic_desc(struct be_nic_res_desc *nic) { memset(nic, 0, sizeof(*nic)); nic->unicast_mac_count = 0xFFFF; @@ -4534,73 +4534,9 @@ int be_cmd_config_qos(struct be_adapter *adapter, u32 max_rate, u16 link_speed, 1, version, domain); } -static void be_fill_vf_res_template(struct be_adapter *adapter, - struct be_resources pool_res, - u16 num_vfs, u16 num_vf_qs, - struct be_nic_res_desc *nic_vft) -{ - u32 vf_if_cap_flags = pool_res.vf_if_cap_flags; - struct be_resources res_mod = {0}; - - /* Resource with fields set to all '1's by GET_PROFILE_CONFIG cmd, - * which are modifiable using SET_PROFILE_CONFIG cmd. - */ - be_cmd_get_profile_config(adapter, &res_mod, RESOURCE_MODIFIABLE, 0); - - /* If RSS IFACE capability flags are modifiable for a VF, set the - * capability flag as valid and set RSS and DEFQ_RSS IFACE flags if - * more than 1 RSSQ is available for a VF. - * Otherwise, provision only 1 queue pair for VF. - */ - if (res_mod.vf_if_cap_flags & BE_IF_FLAGS_RSS) { - nic_vft->flags |= BIT(IF_CAPS_FLAGS_VALID_SHIFT); - if (num_vf_qs > 1) { - vf_if_cap_flags |= BE_IF_FLAGS_RSS; - if (pool_res.if_cap_flags & BE_IF_FLAGS_DEFQ_RSS) - vf_if_cap_flags |= BE_IF_FLAGS_DEFQ_RSS; - } else { - vf_if_cap_flags &= ~(BE_IF_FLAGS_RSS | - BE_IF_FLAGS_DEFQ_RSS); - } - } else { - num_vf_qs = 1; - } - - if (res_mod.vf_if_cap_flags & BE_IF_FLAGS_VLAN_PROMISCUOUS) { - nic_vft->flags |= BIT(IF_CAPS_FLAGS_VALID_SHIFT); - vf_if_cap_flags &= ~BE_IF_FLAGS_VLAN_PROMISCUOUS; - } - - nic_vft->cap_flags = cpu_to_le32(vf_if_cap_flags); - nic_vft->rq_count = cpu_to_le16(num_vf_qs); - nic_vft->txq_count = cpu_to_le16(num_vf_qs); - nic_vft->rssq_count = cpu_to_le16(num_vf_qs); - nic_vft->cq_count = cpu_to_le16(pool_res.max_cq_count / - (num_vfs + 1)); - - /* Distribute unicast MACs, VLANs, IFACE count and MCCQ count equally - * among the PF and it's VFs, if the fields are changeable - */ - if (res_mod.max_uc_mac == FIELD_MODIFIABLE) - nic_vft->unicast_mac_count = cpu_to_le16(pool_res.max_uc_mac / - (num_vfs + 1)); - - if (res_mod.max_vlans == FIELD_MODIFIABLE) - nic_vft->vlan_count = cpu_to_le16(pool_res.max_vlans / - (num_vfs + 1)); - - if (res_mod.max_iface_count == FIELD_MODIFIABLE) - nic_vft->iface_count = cpu_to_le16(pool_res.max_iface_count / - (num_vfs + 1)); - - if (res_mod.max_mcc_count == FIELD_MODIFIABLE) - nic_vft->mcc_count = cpu_to_le16(pool_res.max_mcc_count / - (num_vfs + 1)); -} - int be_cmd_set_sriov_config(struct be_adapter *adapter, struct be_resources pool_res, u16 num_vfs, - u16 num_vf_qs) + struct be_resources *vft_res) { struct { struct be_pcie_res_desc pcie; @@ -4620,12 +4556,26 @@ int be_cmd_set_sriov_config(struct be_adapter *adapter, be_reset_nic_desc(&desc.nic_vft); desc.nic_vft.hdr.desc_type = NIC_RESOURCE_DESC_TYPE_V1; desc.nic_vft.hdr.desc_len = RESOURCE_DESC_SIZE_V1; - desc.nic_vft.flags = BIT(VFT_SHIFT) | BIT(IMM_SHIFT) | BIT(NOSV_SHIFT); + desc.nic_vft.flags = vft_res->flags | BIT(VFT_SHIFT) | + BIT(IMM_SHIFT) | BIT(NOSV_SHIFT); desc.nic_vft.pf_num = adapter->pdev->devfn; desc.nic_vft.vf_num = 0; - - be_fill_vf_res_template(adapter, pool_res, num_vfs, num_vf_qs, - &desc.nic_vft); + desc.nic_vft.cap_flags = cpu_to_le32(vft_res->vf_if_cap_flags); + desc.nic_vft.rq_count = cpu_to_le16(vft_res->max_rx_qs); + desc.nic_vft.txq_count = cpu_to_le16(vft_res->max_tx_qs); + desc.nic_vft.rssq_count = cpu_to_le16(vft_res->max_rss_qs); + desc.nic_vft.cq_count = cpu_to_le16(vft_res->max_cq_count); + + if (vft_res->max_uc_mac) + desc.nic_vft.unicast_mac_count = + cpu_to_le16(vft_res->max_uc_mac); + if (vft_res->max_vlans) + desc.nic_vft.vlan_count = cpu_to_le16(vft_res->max_vlans); + if (vft_res->max_iface_count) + desc.nic_vft.iface_count = + cpu_to_le16(vft_res->max_iface_count); + if (vft_res->max_mcc_count) + desc.nic_vft.mcc_count = cpu_to_le16(vft_res->max_mcc_count); return be_cmd_set_profile_config(adapter, &desc, 2 * RESOURCE_DESC_SIZE_V1, 2, 1, 0); diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h index d8540ae..b63ae37 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.h +++ b/drivers/net/ethernet/emulex/benet/be_cmds.h @@ -2461,4 +2461,4 @@ int be_cmd_set_vxlan_port(struct be_adapter *adapter, __be16 port); int be_cmd_manage_iface(struct be_adapter *adapter, u32 iface, u8 op); int be_cmd_set_sriov_config(struct be_adapter *adapter, struct be_resources res, u16 num_vfs, - u16 num_vf_qs); + struct be_resources *vft_res); diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index ed98ef1..8601047 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -3810,17 +3810,20 @@ static void be_disable_vxlan_offloads(struct be_adapter *adapter) } #endif -static u16 be_calculate_vf_qs(struct be_adapter *adapter, u16 num_vfs) +static void be_calculate_vf_res(struct be_adapter *adapter, u16 num_vfs, + struct be_resources *vft_res) { struct be_resources res = adapter->pool_res; + u32 vf_if_cap_flags = res.vf_if_cap_flags; + struct be_resources res_mod = {0}; u16 num_vf_qs = 1; /* Distribute the queue resources among the PF and it's VFs * Do not distribute queue resources in multi-channel configuration. */ if (num_vfs && !be_is_mc(adapter)) { - /* Divide the qpairs evenly among the VFs and the PF, capped - * at VF-EQ-count. Any remainder qpairs belong to the PF. + /* Divide the rx queues evenly among the VFs and the PF, capped + * at VF-EQ-count. Any remainder queues belong to the PF. */ num_vf_qs = min(SH_VF_MAX_NIC_EQS, res.max_rss_qs / (num_vfs + 1)); @@ -3832,13 +3835,62 @@ static u16 be_calculate_vf_qs(struct be_adapter *adapter, u16 num_vfs) if (num_vfs >= MAX_RSS_IFACES) num_vf_qs = 1; } - return num_vf_qs; + + /* Resource with fields set to all '1's by GET_PROFILE_CONFIG cmd, + * which are modifiable using SET_PROFILE_CONFIG cmd. + */ + be_cmd_get_profile_config(adapter, &res_mod, RESOURCE_MODIFIABLE, 0); + + /* If RSS IFACE capability flags are modifiable for a VF, set the + * capability flag as valid and set RSS and DEFQ_RSS IFACE flags if + * more than 1 RSSQ is available for a VF. + * Otherwise, provision only 1 queue pair for VF. + */ + if (res_mod.vf_if_cap_flags & BE_IF_FLAGS_RSS) { + vft_res->flags |= BIT(IF_CAPS_FLAGS_VALID_SHIFT); + if (num_vf_qs > 1) { + vf_if_cap_flags |= BE_IF_FLAGS_RSS; + if (res.if_cap_flags & BE_IF_FLAGS_DEFQ_RSS) + vf_if_cap_flags |= BE_IF_FLAGS_DEFQ_RSS; + } else { + vf_if_cap_flags &= ~(BE_IF_FLAGS_RSS | + BE_IF_FLAGS_DEFQ_RSS); + } + } else { + num_vf_qs = 1; + } + + if (res_mod.vf_if_cap_flags & BE_IF_FLAGS_VLAN_PROMISCUOUS) { + vft_res->flags |= BIT(IF_CAPS_FLAGS_VALID_SHIFT); + vf_if_cap_flags &= ~BE_IF_FLAGS_VLAN_PROMISCUOUS; + } + + vft_res->vf_if_cap_flags = vf_if_cap_flags; + vft_res->max_rx_qs = num_vf_qs; + vft_res->max_rss_qs = num_vf_qs; + vft_res->max_tx_qs = res.max_tx_qs / (num_vfs + 1); + vft_res->max_cq_count = res.max_cq_count / (num_vfs + 1); + + /* Distribute unicast MACs, VLANs, IFACE count and MCCQ count equally + * among the PF and it's VFs, if the fields are changeable + */ + if (res_mod.max_uc_mac == FIELD_MODIFIABLE) + vft_res->max_uc_mac = res.max_uc_mac / (num_vfs + 1); + + if (res_mod.max_vlans == FIELD_MODIFIABLE) + vft_res->max_vlans = res.max_vlans / (num_vfs + 1); + + if (res_mod.max_iface_count == FIELD_MODIFIABLE) + vft_res->max_iface_count = res.max_iface_count / (num_vfs + 1); + + if (res_mod.max_mcc_count == FIELD_MODIFIABLE) + vft_res->max_mcc_count = res.max_mcc_count / (num_vfs + 1); } static int be_clear(struct be_adapter *adapter) { struct pci_dev *pdev = adapter->pdev; - u16 num_vf_qs; + struct be_resources vft_res = {0}; be_cancel_worker(adapter); @@ -3850,11 +3902,12 @@ static int be_clear(struct be_adapter *adapter) */ if (skyhawk_chip(adapter) && be_physfn(adapter) && !pci_vfs_assigned(pdev)) { - num_vf_qs = be_calculate_vf_qs(adapter, - pci_sriov_get_totalvfs(pdev)); + be_calculate_vf_res(adapter, + pci_sriov_get_totalvfs(pdev), + &vft_res); be_cmd_set_sriov_config(adapter, adapter->pool_res, pci_sriov_get_totalvfs(pdev), - num_vf_qs); + &vft_res); } #ifdef CONFIG_BE2NET_VXLAN @@ -4144,7 +4197,7 @@ static int be_get_sriov_config(struct be_adapter *adapter) static void be_alloc_sriov_res(struct be_adapter *adapter) { int old_vfs = pci_num_vf(adapter->pdev); - u16 num_vf_qs; + struct be_resources vft_res = {0}; int status; be_get_sriov_config(adapter); @@ -4158,9 +4211,9 @@ static void be_alloc_sriov_res(struct be_adapter *adapter) * Also, this is done by FW in Lancer chip. */ if (skyhawk_chip(adapter) && be_max_vfs(adapter) && !old_vfs) { - num_vf_qs = be_calculate_vf_qs(adapter, 0); + be_calculate_vf_res(adapter, 0, &vft_res); status = be_cmd_set_sriov_config(adapter, adapter->pool_res, 0, - num_vf_qs); + &vft_res); if (status) dev_err(&adapter->pdev->dev, "Failed to optimize SRIOV resources\n"); @@ -5552,7 +5605,7 @@ err: static int be_pci_sriov_configure(struct pci_dev *pdev, int num_vfs) { struct be_adapter *adapter = pci_get_drvdata(pdev); - u16 num_vf_qs; + struct be_resources vft_res = {0}; int status; if (!num_vfs) @@ -5575,9 +5628,10 @@ static int be_pci_sriov_configure(struct pci_dev *pdev, int num_vfs) * Also, this is done by FW in Lancer chip. */ if (skyhawk_chip(adapter) && !pci_num_vf(pdev)) { - num_vf_qs = be_calculate_vf_qs(adapter, adapter->num_vfs); + be_calculate_vf_res(adapter, adapter->num_vfs, + &vft_res); status = be_cmd_set_sriov_config(adapter, adapter->pool_res, - adapter->num_vfs, num_vf_qs); + adapter->num_vfs, &vft_res); if (status) dev_err(&pdev->dev, "Failed to optimize SR-IOV resources\n"); -- cgit v0.10.2 From 45f13df75f9044568af4789af1c52d0ffdc68155 Mon Sep 17 00:00:00 2001 From: Sriharsha Basavapatna Date: Mon, 6 Jun 2016 07:22:09 -0400 Subject: be2net: Enable Wake-On-LAN from shutdown for Skyhawk Skyhawk does support wake-up from ACPI shutdown state - S5, provided the platform supports it (like Auxiliary power source etc). The changes listed below are done to fix this. 1) There's no need to defer the HW configuration of WOL to be_suspend(). Remove this in be_suspend() and move it to be_set_wol() ethtool function so it is configured directly in the context of ethtool. This automatically takes care of the shutdown case. 2) The driver incorrectly uses WOL_CAP field in the FW response to get_acpi_wol_cap() command, to determine if WOL is enabled. Instead the driver must rely on the macaddr field in the response to infer WOL state. 3) In be_get_config() during init, if we find that WOL is enabled in FW, call pci_enable_wake() to enable pmcsr.pme_en bit. This is needed to support persistent WOL configuration provided by the FW in some platforms. 4) Remove code in be_set_wol() that writes to PCICFG_PM_CONTROL_OFFSET to set pme_en bit; pci_enable_wake() sets that. Fixes: 028991e49 ("Enabling Wake-on-LAN is not supported in S5 state") Signed-off-by: Sriharsha Basavapatna Signed-off-by: Sathya Perla Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index bef92c4..ddc8611 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -4023,7 +4023,10 @@ int be_cmd_get_acpi_wol_cap(struct be_adapter *adapter) resp = (struct be_cmd_resp_acpi_wol_magic_config_v1 *)cmd.va; adapter->wol_cap = resp->wol_settings; - if (adapter->wol_cap & BE_WOL_CAP) + + /* Non-zero macaddr indicates WOL is enabled */ + if (adapter->wol_cap & BE_WOL_CAP && + !is_zero_ether_addr(resp->magic_mac)) adapter->wol_en = true; } err: diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h index b63ae37..855a170 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.h +++ b/drivers/net/ethernet/emulex/benet/be_cmds.h @@ -1556,7 +1556,9 @@ struct be_cmd_resp_acpi_wol_magic_config_v1 { u8 rsvd0[2]; u8 wol_settings; u8 rsvd1[5]; - u32 rsvd2[295]; + u32 rsvd2[288]; + u8 magic_mac[6]; + u8 rsvd3[22]; } __packed; #define BE_GET_WOL_CAP 2 diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c index 2ff6916..c569cd7 100644 --- a/drivers/net/ethernet/emulex/benet/be_ethtool.c +++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c @@ -793,6 +793,11 @@ static void be_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) static int be_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) { struct be_adapter *adapter = netdev_priv(netdev); + struct device *dev = &adapter->pdev->dev; + struct be_dma_mem cmd; + u8 mac[ETH_ALEN]; + bool enable; + int status; if (wol->wolopts & ~WAKE_MAGIC) return -EOPNOTSUPP; @@ -802,12 +807,32 @@ static int be_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) return -EOPNOTSUPP; } - if (wol->wolopts & WAKE_MAGIC) - adapter->wol_en = true; - else - adapter->wol_en = false; + cmd.size = sizeof(struct be_cmd_req_acpi_wol_magic_config); + cmd.va = dma_zalloc_coherent(dev, cmd.size, &cmd.dma, GFP_KERNEL); + if (!cmd.va) + return -ENOMEM; - return 0; + eth_zero_addr(mac); + + enable = wol->wolopts & WAKE_MAGIC; + if (enable) + ether_addr_copy(mac, adapter->netdev->dev_addr); + + status = be_cmd_enable_magic_wol(adapter, mac, &cmd); + if (status) { + dev_err(dev, "Could not set Wake-on-lan mac address\n"); + status = be_cmd_status(status); + goto err; + } + + pci_enable_wake(adapter->pdev, PCI_D3hot, enable); + pci_enable_wake(adapter->pdev, PCI_D3cold, enable); + + adapter->wol_en = enable ? true : false; + +err: + dma_free_coherent(dev, cmd.size, cmd.va, cmd.dma); + return status; } static int be_test_ddr_dma(struct be_adapter *adapter) diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 8601047..32823a7 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -3636,40 +3636,6 @@ err: return -EIO; } -static int be_setup_wol(struct be_adapter *adapter, bool enable) -{ - struct device *dev = &adapter->pdev->dev; - struct be_dma_mem cmd; - u8 mac[ETH_ALEN]; - int status; - - eth_zero_addr(mac); - - cmd.size = sizeof(struct be_cmd_req_acpi_wol_magic_config); - cmd.va = dma_zalloc_coherent(dev, cmd.size, &cmd.dma, GFP_KERNEL); - if (!cmd.va) - return -ENOMEM; - - if (enable) { - status = pci_write_config_dword(adapter->pdev, - PCICFG_PM_CONTROL_OFFSET, - PCICFG_PM_CONTROL_MASK); - if (status) { - dev_err(dev, "Could not enable Wake-on-lan\n"); - goto err; - } - } else { - ether_addr_copy(mac, adapter->netdev->dev_addr); - } - - status = be_cmd_enable_magic_wol(adapter, mac, &cmd); - pci_enable_wake(adapter->pdev, PCI_D3hot, enable); - pci_enable_wake(adapter->pdev, PCI_D3cold, enable); -err: - dma_free_coherent(dev, cmd.size, cmd.va, cmd.dma); - return status; -} - static void be_vf_eth_addr_generate(struct be_adapter *adapter, u8 *mac) { u32 addr; @@ -4294,6 +4260,8 @@ static int be_get_config(struct be_adapter *adapter) } be_cmd_get_acpi_wol_cap(adapter); + pci_enable_wake(adapter->pdev, PCI_D3hot, adapter->wol_en); + pci_enable_wake(adapter->pdev, PCI_D3cold, adapter->wol_en); be_cmd_query_port_name(adapter); @@ -5463,9 +5431,6 @@ static int be_suspend(struct pci_dev *pdev, pm_message_t state) { struct be_adapter *adapter = pci_get_drvdata(pdev); - if (adapter->wol_en) - be_setup_wol(adapter, true); - be_intr_set(adapter, false); be_cancel_err_detection(adapter); @@ -5494,9 +5459,6 @@ static int be_pci_resume(struct pci_dev *pdev) be_schedule_err_detection(adapter, ERR_DETECTION_DELAY); - if (adapter->wol_en) - be_setup_wol(adapter, false); - return 0; } -- cgit v0.10.2 From de2b1e0366c8c332fa403e8bd8b6e2b22f557d2e Mon Sep 17 00:00:00 2001 From: Somnath Kotur Date: Mon, 6 Jun 2016 07:22:10 -0400 Subject: be2net: Fix provisioning of RSS for VFs in multi-partition configurations Currently, we do not distribute queue resources to enable RSS for VFs in multi-channel/partition configurations. Fix this by having each PF(SRIOV capable) calculate it's share of the 15 RSS Policy Tables available per port before provisioning resources for all the VFs. This proportional share calculation is done based on division of the PF's MAX VFs with the Total MAX VFs on that port. It also needs to learn about the no: of NIC PFs on the port and subtract that from the 15 RSS Policy Tables on the port. Signed-off-by: Somnath Kotur Signed-off-by: Sathya Perla Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index ee0bdb7..e1e6c40 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -97,7 +97,8 @@ * SURF/DPDK */ -#define MAX_RSS_IFACES 15 +#define MAX_PORT_RSS_TABLES 15 +#define MAX_NIC_FUNCS 16 #define MAX_RX_QS 32 #define MAX_EVT_QS 32 #define MAX_TX_QS 32 @@ -445,6 +446,16 @@ struct be_resources { u32 if_cap_flags; u32 vf_if_cap_flags; /* VF if capability flags */ u32 flags; + /* Calculated PF Pool's share of RSS Tables. This is not enforced by + * the FW, but is a self-imposed driver limitation. + */ + u16 max_rss_tables; +}; + +/* These are port-wide values */ +struct be_port_resources { + u16 max_vfs; + u16 nic_pfs; }; #define be_is_os2bmc_enabled(adapter) (adapter->flags & BE_FLAGS_OS2BMC) @@ -635,6 +646,8 @@ struct be_adapter { #define be_max_rxqs(adapter) (adapter->res.max_rx_qs) #define be_max_eqs(adapter) (adapter->res.max_evt_qs) #define be_if_cap_flags(adapter) (adapter->res.if_cap_flags) +#define be_max_pf_pool_rss_tables(adapter) \ + (adapter->pool_res.max_rss_tables) static inline u16 be_max_qs(struct be_adapter *adapter) { diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index ddc8611..29aeb91 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -4363,9 +4363,35 @@ err: return status; } +/* This routine returns a list of all the NIC PF_nums in the adapter */ +u16 be_get_nic_pf_num_list(u8 *buf, u32 desc_count, u16 *nic_pf_nums) +{ + struct be_res_desc_hdr *hdr = (struct be_res_desc_hdr *)buf; + struct be_pcie_res_desc *pcie = NULL; + int i; + u16 nic_pf_count = 0; + + for (i = 0; i < desc_count; i++) { + if (hdr->desc_type == PCIE_RESOURCE_DESC_TYPE_V0 || + hdr->desc_type == PCIE_RESOURCE_DESC_TYPE_V1) { + pcie = (struct be_pcie_res_desc *)hdr; + if (pcie->pf_state && (pcie->pf_type == MISSION_NIC || + pcie->pf_type == MISSION_RDMA)) { + nic_pf_nums[nic_pf_count++] = pcie->pf_num; + } + } + + hdr->desc_len = hdr->desc_len ? : RESOURCE_DESC_SIZE_V0; + hdr = (void *)hdr + hdr->desc_len; + } + return nic_pf_count; +} + /* Will use MBOX only if MCCQ has not been created */ int be_cmd_get_profile_config(struct be_adapter *adapter, - struct be_resources *res, u8 query, u8 domain) + struct be_resources *res, + struct be_port_resources *port_res, + u8 profile_type, u8 query, u8 domain) { struct be_cmd_resp_get_profile_config *resp; struct be_cmd_req_get_profile_config *req; @@ -4392,7 +4418,7 @@ int be_cmd_get_profile_config(struct be_adapter *adapter, if (!lancer_chip(adapter)) req->hdr.version = 1; - req->type = ACTIVE_PROFILE_TYPE; + req->type = profile_type; req->hdr.domain = domain; /* When QUERY_MODIFIABLE_FIELDS_TYPE bit is set, cmd returns the @@ -4409,6 +4435,28 @@ int be_cmd_get_profile_config(struct be_adapter *adapter, resp = cmd.va; desc_count = le16_to_cpu(resp->desc_count); + if (port_res) { + u16 nic_pf_cnt = 0, i; + u16 nic_pf_num_list[MAX_NIC_FUNCS]; + + nic_pf_cnt = be_get_nic_pf_num_list(resp->func_param, + desc_count, + nic_pf_num_list); + + for (i = 0; i < nic_pf_cnt; i++) { + nic = be_get_func_nic_desc(resp->func_param, desc_count, + nic_pf_num_list[i]); + if (nic->link_param == adapter->port_num) { + port_res->nic_pfs++; + pcie = be_get_pcie_desc(resp->func_param, + desc_count, + nic_pf_num_list[i]); + port_res->max_vfs += le16_to_cpu(pcie->num_vfs); + } + } + return status; + } + pcie = be_get_pcie_desc(resp->func_param, desc_count, adapter->pf_num); if (pcie) diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h index 855a170..cb96ddd 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.h +++ b/drivers/net/ethernet/emulex/benet/be_cmds.h @@ -2130,6 +2130,9 @@ struct be_cmd_req_set_ext_fat_caps { #define IMM_SHIFT 6 /* Immediate */ #define NOSV_SHIFT 7 /* No save */ +#define MISSION_NIC 1 +#define MISSION_RDMA 8 + struct be_res_desc_hdr { u8 desc_type; u8 desc_len; @@ -2246,6 +2249,7 @@ struct be_cmd_req_get_profile_config { struct be_cmd_req_hdr hdr; u8 rsvd; #define ACTIVE_PROFILE_TYPE 0x2 +#define SAVED_PROFILE_TYPE 0x0 #define QUERY_MODIFIABLE_FIELDS_TYPE BIT(3) u8 type; u16 rsvd1; @@ -2451,7 +2455,9 @@ int be_cmd_query_port_name(struct be_adapter *adapter); int be_cmd_get_func_config(struct be_adapter *adapter, struct be_resources *res); int be_cmd_get_profile_config(struct be_adapter *adapter, - struct be_resources *res, u8 query, u8 domain); + struct be_resources *res, + struct be_port_resources *port_res, + u8 profile_type, u8 query, u8 domain); int be_cmd_get_active_profile(struct be_adapter *adapter, u16 *profile); int be_cmd_get_if_id(struct be_adapter *adapter, struct be_vf_cfg *vf_cfg, int vf_num); diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 32823a7..2451a47 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -3784,28 +3784,27 @@ static void be_calculate_vf_res(struct be_adapter *adapter, u16 num_vfs, struct be_resources res_mod = {0}; u16 num_vf_qs = 1; - /* Distribute the queue resources among the PF and it's VFs - * Do not distribute queue resources in multi-channel configuration. - */ - if (num_vfs && !be_is_mc(adapter)) { - /* Divide the rx queues evenly among the VFs and the PF, capped - * at VF-EQ-count. Any remainder queues belong to the PF. - */ + /* Distribute the queue resources among the PF and it's VFs */ + if (num_vfs) { + /* Divide the rx queues evenly among the VFs and the PF, capped + * at VF-EQ-count. Any remainder queues belong to the PF. + */ num_vf_qs = min(SH_VF_MAX_NIC_EQS, res.max_rss_qs / (num_vfs + 1)); - /* Skyhawk-R chip supports only MAX_RSS_IFACES RSS capable - * interfaces per port. Provide RSS on VFs, only if number - * of VFs requested is less than MAX_RSS_IFACES limit. + /* Skyhawk-R chip supports only MAX_PORT_RSS_TABLES + * RSS Tables per port. Provide RSS on VFs, only if number of + * VFs requested is less than it's PF Pool's RSS Tables limit. */ - if (num_vfs >= MAX_RSS_IFACES) + if (num_vfs >= be_max_pf_pool_rss_tables(adapter)) num_vf_qs = 1; } /* Resource with fields set to all '1's by GET_PROFILE_CONFIG cmd, * which are modifiable using SET_PROFILE_CONFIG cmd. */ - be_cmd_get_profile_config(adapter, &res_mod, RESOURCE_MODIFIABLE, 0); + be_cmd_get_profile_config(adapter, &res_mod, NULL, ACTIVE_PROFILE_TYPE, + RESOURCE_MODIFIABLE, 0); /* If RSS IFACE capability flags are modifiable for a VF, set the * capability flag as valid and set RSS and DEFQ_RSS IFACE flags if @@ -3903,7 +3902,8 @@ static int be_vfs_if_create(struct be_adapter *adapter) for_all_vfs(adapter, vf_cfg, vf) { if (!BE3_chip(adapter)) { - status = be_cmd_get_profile_config(adapter, &res, + status = be_cmd_get_profile_config(adapter, &res, NULL, + ACTIVE_PROFILE_TYPE, RESOURCE_LIMITS, vf + 1); if (!status) { @@ -4088,8 +4088,9 @@ static void BEx_get_resources(struct be_adapter *adapter, /* On a SuperNIC profile, the driver needs to use the * GET_PROFILE_CONFIG cmd to query the per-function TXQ limits */ - be_cmd_get_profile_config(adapter, &super_nic_res, - RESOURCE_LIMITS, 0); + be_cmd_get_profile_config(adapter, &super_nic_res, NULL, + ACTIVE_PROFILE_TYPE, RESOURCE_LIMITS, + 0); /* Some old versions of BE3 FW don't report max_tx_qs value */ res->max_tx_qs = super_nic_res.max_tx_qs ? : BE3_MAX_TX_QS; } else { @@ -4128,12 +4129,38 @@ static void be_setup_init(struct be_adapter *adapter) adapter->cmd_privileges = MIN_PRIVILEGES; } +/* HW supports only MAX_PORT_RSS_TABLES RSS Policy Tables per port. + * However, this HW limitation is not exposed to the host via any SLI cmd. + * As a result, in the case of SRIOV and in particular multi-partition configs + * the driver needs to calcuate a proportional share of RSS Tables per PF-pool + * for distribution between the VFs. This self-imposed limit will determine the + * no: of VFs for which RSS can be enabled. + */ +void be_calculate_pf_pool_rss_tables(struct be_adapter *adapter) +{ + struct be_port_resources port_res = {0}; + u8 rss_tables_on_port; + u16 max_vfs = be_max_vfs(adapter); + + be_cmd_get_profile_config(adapter, NULL, &port_res, SAVED_PROFILE_TYPE, + RESOURCE_LIMITS, 0); + + rss_tables_on_port = MAX_PORT_RSS_TABLES - port_res.nic_pfs; + + /* Each PF Pool's RSS Tables limit = + * PF's Max VFs / Total_Max_VFs on Port * RSS Tables on Port + */ + adapter->pool_res.max_rss_tables = + max_vfs * rss_tables_on_port / port_res.max_vfs; +} + static int be_get_sriov_config(struct be_adapter *adapter) { struct be_resources res = {0}; int max_vfs, old_vfs; - be_cmd_get_profile_config(adapter, &res, RESOURCE_LIMITS, 0); + be_cmd_get_profile_config(adapter, &res, NULL, ACTIVE_PROFILE_TYPE, + RESOURCE_LIMITS, 0); /* Some old versions of BE3 FW don't report max_vfs value */ if (BE3_chip(adapter) && !res.max_vfs) { @@ -4157,6 +4184,12 @@ static int be_get_sriov_config(struct be_adapter *adapter) adapter->num_vfs = old_vfs; } + if (skyhawk_chip(adapter) && be_max_vfs(adapter) && !old_vfs) { + be_calculate_pf_pool_rss_tables(adapter); + dev_info(&adapter->pdev->dev, + "RSS can be enabled for all VFs if num_vfs <= %d\n", + be_max_pf_pool_rss_tables(adapter)); + } return 0; } @@ -4272,15 +4305,6 @@ static int be_get_config(struct be_adapter *adapter) "Using profile 0x%x\n", profile_id); } - status = be_get_resources(adapter); - if (status) - return status; - - adapter->pmac_id = kcalloc(be_max_uc(adapter), - sizeof(*adapter->pmac_id), GFP_KERNEL); - if (!adapter->pmac_id) - return -ENOMEM; - return 0; } @@ -4481,13 +4505,22 @@ static int be_setup(struct be_adapter *adapter) return status; } + status = be_get_config(adapter); + if (status) + goto err; + if (!BE2_chip(adapter) && be_physfn(adapter)) be_alloc_sriov_res(adapter); - status = be_get_config(adapter); + status = be_get_resources(adapter); if (status) goto err; + adapter->pmac_id = kcalloc(be_max_uc(adapter), + sizeof(*adapter->pmac_id), GFP_KERNEL); + if (!adapter->pmac_id) + return -ENOMEM; + status = be_msix_enable(adapter); if (status) goto err; -- cgit v0.10.2 From f9eb8aea2a1e12fc2f584d1627deeb957435a801 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 6 Jun 2016 09:37:15 -0700 Subject: net_sched: transform qdisc running bit into a seqcount Instead of using a single bit (__QDISC___STATE_RUNNING) in sch->__state, use a seqcount. This adds lockdep support, but more importantly it will allow us to sample qdisc/class statistics without having to grab qdisc root lock. Signed-off-by: Eric Dumazet Cc: Cong Wang Cc: Jamal Hadi Salim Signed-off-by: David S. Miller diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 941ec99..681af31 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4610,6 +4610,7 @@ static int bond_check_params(struct bond_params *params) static struct lock_class_key bonding_netdev_xmit_lock_key; static struct lock_class_key bonding_netdev_addr_lock_key; static struct lock_class_key bonding_tx_busylock_key; +static struct lock_class_key bonding_qdisc_running_key; static void bond_set_lockdep_class_one(struct net_device *dev, struct netdev_queue *txq, @@ -4625,6 +4626,7 @@ static void bond_set_lockdep_class(struct net_device *dev) &bonding_netdev_addr_lock_key); netdev_for_each_tx_queue(dev, bond_set_lockdep_class_one, NULL); dev->qdisc_tx_busylock = &bonding_tx_busylock_key; + dev->qdisc_running_key = &bonding_qdisc_running_key; } /* Called from registration process */ diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c index 8dedafa..aeabaa4 100644 --- a/drivers/net/ppp/ppp_generic.c +++ b/drivers/net/ppp/ppp_generic.c @@ -1313,9 +1313,12 @@ ppp_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats64) } static struct lock_class_key ppp_tx_busylock; +static struct lock_class_key ppp_qdisc_running_key; + static int ppp_dev_init(struct net_device *dev) { dev->qdisc_tx_busylock = &ppp_tx_busylock; + dev->qdisc_running_key = &ppp_qdisc_running_key; return 0; } diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index 2ace126..00eb389 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -1577,6 +1577,7 @@ static const struct team_option team_options[] = { static struct lock_class_key team_netdev_xmit_lock_key; static struct lock_class_key team_netdev_addr_lock_key; static struct lock_class_key team_tx_busylock_key; +static struct lock_class_key team_qdisc_running_key; static void team_set_lockdep_class_one(struct net_device *dev, struct netdev_queue *txq, @@ -1590,6 +1591,7 @@ static void team_set_lockdep_class(struct net_device *dev) lockdep_set_class(&dev->addr_list_lock, &team_netdev_addr_lock_key); netdev_for_each_tx_queue(dev, team_set_lockdep_class_one, NULL); dev->qdisc_tx_busylock = &team_tx_busylock_key; + dev->qdisc_running_key = &team_qdisc_running_key; } static int team_init(struct net_device *dev) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index fa6df26..59d7e06 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1862,6 +1862,7 @@ struct net_device { #endif struct phy_device *phydev; struct lock_class_key *qdisc_tx_busylock; + struct lock_class_key *qdisc_running_key; bool proto_down; }; #define to_net_dev(d) container_of(d, struct net_device, dev) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index a1fd76c..bff8d89 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -29,13 +29,6 @@ enum qdisc_state_t { __QDISC_STATE_THROTTLED, }; -/* - * following bits are only changed while qdisc lock is held - */ -enum qdisc___state_t { - __QDISC___STATE_RUNNING = 1, -}; - struct qdisc_size_table { struct rcu_head rcu; struct list_head list; @@ -93,7 +86,7 @@ struct Qdisc { unsigned long state; struct sk_buff_head q; struct gnet_stats_basic_packed bstats; - unsigned int __state; + seqcount_t running; struct gnet_stats_queue qstats; struct rcu_head rcu_head; int padded; @@ -104,20 +97,20 @@ struct Qdisc { static inline bool qdisc_is_running(const struct Qdisc *qdisc) { - return (qdisc->__state & __QDISC___STATE_RUNNING) ? true : false; + return (raw_read_seqcount(&qdisc->running) & 1) ? true : false; } static inline bool qdisc_run_begin(struct Qdisc *qdisc) { if (qdisc_is_running(qdisc)) return false; - qdisc->__state |= __QDISC___STATE_RUNNING; + write_seqcount_begin(&qdisc->running); return true; } static inline void qdisc_run_end(struct Qdisc *qdisc) { - qdisc->__state &= ~__QDISC___STATE_RUNNING; + write_seqcount_end(&qdisc->running); } static inline bool qdisc_may_bulk(const struct Qdisc *qdisc) diff --git a/net/bluetooth/6lowpan.c b/net/bluetooth/6lowpan.c index 780089d..977a11e 100644 --- a/net/bluetooth/6lowpan.c +++ b/net/bluetooth/6lowpan.c @@ -629,6 +629,7 @@ static netdev_tx_t bt_xmit(struct sk_buff *skb, struct net_device *netdev) static struct lock_class_key bt_tx_busylock; static struct lock_class_key bt_netdev_xmit_lock_key; +static struct lock_class_key bt_qdisc_running_key; static void bt_set_lockdep_class_one(struct net_device *dev, struct netdev_queue *txq, @@ -641,6 +642,7 @@ static int bt_dev_init(struct net_device *dev) { netdev_for_each_tx_queue(dev, bt_set_lockdep_class_one, NULL); dev->qdisc_tx_busylock = &bt_tx_busylock; + dev->qdisc_running_key = &bt_qdisc_running_key; return 0; } diff --git a/net/core/dev.c b/net/core/dev.c index 896b686..e0bcc39 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3075,7 +3075,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, /* * Heuristic to force contended enqueues to serialize on a * separate lock before trying to get qdisc main lock. - * This permits __QDISC___STATE_RUNNING owner to get the lock more + * This permits qdisc->running owner to get the lock more * often and dequeue packets faster. */ contended = qdisc_is_running(q); diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index dd085db..14aa5ef 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -60,6 +60,7 @@ static struct header_ops lowpan_header_ops = { static struct lock_class_key lowpan_tx_busylock; static struct lock_class_key lowpan_netdev_xmit_lock_key; +static struct lock_class_key lowpan_qdisc_running_key; static void lowpan_set_lockdep_class_one(struct net_device *ldev, struct netdev_queue *txq, @@ -73,6 +74,8 @@ static int lowpan_dev_init(struct net_device *ldev) { netdev_for_each_tx_queue(ldev, lowpan_set_lockdep_class_one, NULL); ldev->qdisc_tx_busylock = &lowpan_tx_busylock; + ldev->qdisc_running_key = &lowpan_qdisc_running_key; + return 0; } diff --git a/net/l2tp/l2tp_eth.c b/net/l2tp/l2tp_eth.c index e253c26..c00d72d 100644 --- a/net/l2tp/l2tp_eth.c +++ b/net/l2tp/l2tp_eth.c @@ -68,6 +68,8 @@ static inline struct l2tp_eth_net *l2tp_eth_pernet(struct net *net) } static struct lock_class_key l2tp_eth_tx_busylock; +static struct lock_class_key l2tp_qdisc_running_key; + static int l2tp_eth_dev_init(struct net_device *dev) { struct l2tp_eth *priv = netdev_priv(dev); @@ -76,6 +78,8 @@ static int l2tp_eth_dev_init(struct net_device *dev) eth_hw_addr_random(dev); eth_broadcast_addr(dev->broadcast); dev->qdisc_tx_busylock = &l2tp_eth_tx_busylock; + dev->qdisc_running_key = &l2tp_qdisc_running_key; + return 0; } diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 269dd71..cebea73 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -110,7 +110,7 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate, /* * Transmit possibly several skbs, and handle the return status as - * required. Holding the __QDISC___STATE_RUNNING bit guarantees that + * required. Owning running seqcount bit guarantees that * only one CPU can execute this function. * * Returns to the caller: @@ -137,10 +137,10 @@ int sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q, HARD_TX_UNLOCK(dev, txq); } else { - spin_lock(root_lock); + spin_lock_nested(root_lock, SINGLE_DEPTH_NESTING); return qdisc_qlen(q); } - spin_lock(root_lock); + spin_lock_nested(root_lock, SINGLE_DEPTH_NESTING); if (dev_xmit_complete(ret)) { /* Driver sent out skb successfully or skb was consumed */ @@ -163,7 +163,7 @@ int sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q, /* * NOTE: Called under qdisc_lock(q) with locally disabled BH. * - * __QDISC___STATE_RUNNING guarantees only one CPU can process + * running seqcount guarantees only one CPU can process * this qdisc at a time. qdisc_lock(q) serializes queue accesses for * this queue. * @@ -379,6 +379,7 @@ struct Qdisc noop_qdisc = { .list = LIST_HEAD_INIT(noop_qdisc.list), .q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock), .dev_queue = &noop_netdev_queue, + .running = SEQCNT_ZERO(noop_qdisc.running), .busylock = __SPIN_LOCK_UNLOCKED(noop_qdisc.busylock), }; EXPORT_SYMBOL(noop_qdisc); @@ -537,6 +538,7 @@ struct Qdisc_ops pfifo_fast_ops __read_mostly = { EXPORT_SYMBOL(pfifo_fast_ops); static struct lock_class_key qdisc_tx_busylock; +static struct lock_class_key qdisc_running_key; struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, const struct Qdisc_ops *ops) @@ -570,6 +572,10 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, lockdep_set_class(&sch->busylock, dev->qdisc_tx_busylock ?: &qdisc_tx_busylock); + seqcount_init(&sch->running); + lockdep_set_class(&sch->running, + dev->qdisc_running_key ?: &qdisc_running_key); + sch->ops = ops; sch->enqueue = ops->enqueue; sch->dequeue = ops->dequeue; -- cgit v0.10.2 From edb09eb17ed89eaa82a52dd306beac93e292b485 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 6 Jun 2016 09:37:16 -0700 Subject: net: sched: do not acquire qdisc spinlock in qdisc/class stats dump Large tc dumps (tc -s {qdisc|class} sh dev ethX) done by Google BwE host agent [1] are problematic at scale : For each qdisc/class found in the dump, we currently lock the root qdisc spinlock in order to get stats. Sampling stats every 5 seconds from thousands of HTB classes is a challenge when the root qdisc spinlock is under high pressure. Not only the dumps take time, they also slow down the fast path (queue/dequeue packets) by 10 % to 20 % in some cases. An audit of existing qdiscs showed that sch_fq_codel is the only qdisc that might need the qdisc lock in fq_codel_dump_stats() and fq_codel_dump_class_stats() In v2 of this patch, I now use the Qdisc running seqcount to provide consistent reads of packets/bytes counters, regardless of 32/64 bit arches. I also changed rate estimators to use the same infrastructure so that they no longer need to lock root qdisc lock. [1] http://static.googleusercontent.com/media/research.google.com/en//pubs/archive/43838.pdf Signed-off-by: Eric Dumazet Cc: Cong Wang Cc: Jamal Hadi Salim Cc: John Fastabend Cc: Kevin Athey Cc: Xiaotian Pei Signed-off-by: David S. Miller diff --git a/Documentation/networking/gen_stats.txt b/Documentation/networking/gen_stats.txt index ff630a8..179b18c 100644 --- a/Documentation/networking/gen_stats.txt +++ b/Documentation/networking/gen_stats.txt @@ -21,7 +21,7 @@ struct mystruct { ... }; -Update statistics: +Update statistics, in dequeue() methods only, (while owning qdisc->running) mystruct->tstats.packet++; mystruct->qstats.backlog += skb->pkt_len; diff --git a/include/net/gen_stats.h b/include/net/gen_stats.h index 610cd39..231e121 100644 --- a/include/net/gen_stats.h +++ b/include/net/gen_stats.h @@ -33,10 +33,12 @@ int gnet_stats_start_copy_compat(struct sk_buff *skb, int type, spinlock_t *lock, struct gnet_dump *d, int padattr); -int gnet_stats_copy_basic(struct gnet_dump *d, +int gnet_stats_copy_basic(const seqcount_t *running, + struct gnet_dump *d, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b); -void __gnet_stats_copy_basic(struct gnet_stats_basic_packed *bstats, +void __gnet_stats_copy_basic(const seqcount_t *running, + struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b); int gnet_stats_copy_rate_est(struct gnet_dump *d, @@ -52,13 +54,15 @@ int gnet_stats_finish_copy(struct gnet_dump *d); int gen_new_estimator(struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu_bstats, struct gnet_stats_rate_est64 *rate_est, - spinlock_t *stats_lock, struct nlattr *opt); + spinlock_t *stats_lock, + seqcount_t *running, struct nlattr *opt); void gen_kill_estimator(struct gnet_stats_basic_packed *bstats, struct gnet_stats_rate_est64 *rate_est); int gen_replace_estimator(struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu_bstats, struct gnet_stats_rate_est64 *rate_est, - spinlock_t *stats_lock, struct nlattr *opt); + spinlock_t *stats_lock, + seqcount_t *running, struct nlattr *opt); bool gen_estimator_active(const struct gnet_stats_basic_packed *bstats, const struct gnet_stats_rate_est64 *rate_est); #endif diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index bff8d89..c4f5749 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -314,6 +314,14 @@ static inline spinlock_t *qdisc_root_sleeping_lock(const struct Qdisc *qdisc) return qdisc_lock(root); } +static inline seqcount_t *qdisc_root_sleeping_running(const struct Qdisc *qdisc) +{ + struct Qdisc *root = qdisc_root_sleeping(qdisc); + + ASSERT_RTNL(); + return &root->running; +} + static inline struct net_device *qdisc_dev(const struct Qdisc *qdisc) { return qdisc->dev_queue->dev; diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c index 4573d81..cad8e79 100644 --- a/net/core/gen_estimator.c +++ b/net/core/gen_estimator.c @@ -84,6 +84,7 @@ struct gen_estimator struct gnet_stats_basic_packed *bstats; struct gnet_stats_rate_est64 *rate_est; spinlock_t *stats_lock; + seqcount_t *running; int ewma_log; u32 last_packets; unsigned long avpps; @@ -121,26 +122,28 @@ static void est_timer(unsigned long arg) unsigned long rate; u64 brate; - spin_lock(e->stats_lock); + if (e->stats_lock) + spin_lock(e->stats_lock); read_lock(&est_lock); if (e->bstats == NULL) goto skip; - __gnet_stats_copy_basic(&b, e->cpu_bstats, e->bstats); + __gnet_stats_copy_basic(e->running, &b, e->cpu_bstats, e->bstats); brate = (b.bytes - e->last_bytes)<<(7 - idx); e->last_bytes = b.bytes; e->avbps += (brate >> e->ewma_log) - (e->avbps >> e->ewma_log); - e->rate_est->bps = (e->avbps+0xF)>>5; + WRITE_ONCE(e->rate_est->bps, (e->avbps + 0xF) >> 5); rate = b.packets - e->last_packets; rate <<= (7 - idx); e->last_packets = b.packets; e->avpps += (rate >> e->ewma_log) - (e->avpps >> e->ewma_log); - e->rate_est->pps = (e->avpps + 0xF) >> 5; + WRITE_ONCE(e->rate_est->pps, (e->avpps + 0xF) >> 5); skip: read_unlock(&est_lock); - spin_unlock(e->stats_lock); + if (e->stats_lock) + spin_unlock(e->stats_lock); } if (!list_empty(&elist[idx].list)) @@ -194,6 +197,7 @@ struct gen_estimator *gen_find_node(const struct gnet_stats_basic_packed *bstats * @cpu_bstats: bstats per cpu * @rate_est: rate estimator statistics * @stats_lock: statistics lock + * @running: qdisc running seqcount * @opt: rate estimator configuration TLV * * Creates a new rate estimator with &bstats as source and &rate_est @@ -209,6 +213,7 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu_bstats, struct gnet_stats_rate_est64 *rate_est, spinlock_t *stats_lock, + seqcount_t *running, struct nlattr *opt) { struct gen_estimator *est; @@ -226,12 +231,13 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats, if (est == NULL) return -ENOBUFS; - __gnet_stats_copy_basic(&b, cpu_bstats, bstats); + __gnet_stats_copy_basic(running, &b, cpu_bstats, bstats); idx = parm->interval + 2; est->bstats = bstats; est->rate_est = rate_est; est->stats_lock = stats_lock; + est->running = running; est->ewma_log = parm->ewma_log; est->last_bytes = b.bytes; est->avbps = rate_est->bps<<5; @@ -291,6 +297,7 @@ EXPORT_SYMBOL(gen_kill_estimator); * @cpu_bstats: bstats per cpu * @rate_est: rate estimator statistics * @stats_lock: statistics lock + * @running: qdisc running seqcount (might be NULL) * @opt: rate estimator configuration TLV * * Replaces the configuration of a rate estimator by calling @@ -301,10 +308,11 @@ EXPORT_SYMBOL(gen_kill_estimator); int gen_replace_estimator(struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu_bstats, struct gnet_stats_rate_est64 *rate_est, - spinlock_t *stats_lock, struct nlattr *opt) + spinlock_t *stats_lock, + seqcount_t *running, struct nlattr *opt) { gen_kill_estimator(bstats, rate_est); - return gen_new_estimator(bstats, cpu_bstats, rate_est, stats_lock, opt); + return gen_new_estimator(bstats, cpu_bstats, rate_est, stats_lock, running, opt); } EXPORT_SYMBOL(gen_replace_estimator); diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c index f96ee8b..d9c210c 100644 --- a/net/core/gen_stats.c +++ b/net/core/gen_stats.c @@ -32,10 +32,11 @@ gnet_stats_copy(struct gnet_dump *d, int type, void *buf, int size, int padattr) return 0; nla_put_failure: + if (d->lock) + spin_unlock_bh(d->lock); kfree(d->xstats); d->xstats = NULL; d->xstats_len = 0; - spin_unlock_bh(d->lock); return -1; } @@ -65,15 +66,16 @@ gnet_stats_start_copy_compat(struct sk_buff *skb, int type, int tc_stats_type, { memset(d, 0, sizeof(*d)); - spin_lock_bh(lock); - d->lock = lock; if (type) d->tail = (struct nlattr *)skb_tail_pointer(skb); d->skb = skb; d->compat_tc_stats = tc_stats_type; d->compat_xstats = xstats_type; d->padattr = padattr; - + if (lock) { + d->lock = lock; + spin_lock_bh(lock); + } if (d->tail) return gnet_stats_copy(d, type, NULL, 0, padattr); @@ -126,16 +128,23 @@ __gnet_stats_copy_basic_cpu(struct gnet_stats_basic_packed *bstats, } void -__gnet_stats_copy_basic(struct gnet_stats_basic_packed *bstats, +__gnet_stats_copy_basic(const seqcount_t *running, + struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b) { + unsigned int seq; + if (cpu) { __gnet_stats_copy_basic_cpu(bstats, cpu); - } else { + return; + } + do { + if (running) + seq = read_seqcount_begin(running); bstats->bytes = b->bytes; bstats->packets = b->packets; - } + } while (running && read_seqcount_retry(running, seq)); } EXPORT_SYMBOL(__gnet_stats_copy_basic); @@ -152,13 +161,14 @@ EXPORT_SYMBOL(__gnet_stats_copy_basic); * if the room in the socket buffer was not sufficient. */ int -gnet_stats_copy_basic(struct gnet_dump *d, +gnet_stats_copy_basic(const seqcount_t *running, + struct gnet_dump *d, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b) { struct gnet_stats_basic_packed bstats = {0}; - __gnet_stats_copy_basic(&bstats, cpu, b); + __gnet_stats_copy_basic(running, &bstats, cpu, b); if (d->compat_tc_stats) { d->tc_stats.bytes = bstats.bytes; @@ -328,8 +338,9 @@ gnet_stats_copy_app(struct gnet_dump *d, void *st, int len) return 0; err_out: + if (d->lock) + spin_unlock_bh(d->lock); d->xstats_len = 0; - spin_unlock_bh(d->lock); return -1; } EXPORT_SYMBOL(gnet_stats_copy_app); @@ -363,10 +374,11 @@ gnet_stats_finish_copy(struct gnet_dump *d) return -1; } + if (d->lock) + spin_unlock_bh(d->lock); kfree(d->xstats); d->xstats = NULL; d->xstats_len = 0; - spin_unlock_bh(d->lock); return 0; } EXPORT_SYMBOL(gnet_stats_finish_copy); diff --git a/net/netfilter/xt_RATEEST.c b/net/netfilter/xt_RATEEST.c index 604df6f..515131f 100644 --- a/net/netfilter/xt_RATEEST.c +++ b/net/netfilter/xt_RATEEST.c @@ -137,7 +137,7 @@ static int xt_rateest_tg_checkentry(const struct xt_tgchk_param *par) cfg.est.ewma_log = info->ewma_log; ret = gen_new_estimator(&est->bstats, NULL, &est->rstats, - &est->lock, &cfg.opt); + &est->lock, NULL, &cfg.opt); if (ret < 0) goto err2; diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 719bc2e..b6db56e 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -287,7 +287,7 @@ err2: if (est) { err = gen_new_estimator(&p->tcfc_bstats, p->cpu_bstats, &p->tcfc_rate_est, - &p->tcfc_lock, est); + &p->tcfc_lock, NULL, est); if (err) { free_percpu(p->cpu_qstats); goto err2; @@ -671,7 +671,7 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a, if (err < 0) goto errout; - if (gnet_stats_copy_basic(&d, p->cpu_bstats, &p->tcfc_bstats) < 0 || + if (gnet_stats_copy_basic(NULL, &d, p->cpu_bstats, &p->tcfc_bstats) < 0 || gnet_stats_copy_rate_est(&d, &p->tcfc_bstats, &p->tcfc_rate_est) < 0 || gnet_stats_copy_queue(&d, p->cpu_qstats, diff --git a/net/sched/act_police.c b/net/sched/act_police.c index 820b116..bcb3114 100644 --- a/net/sched/act_police.c +++ b/net/sched/act_police.c @@ -185,7 +185,8 @@ override: if (est) { err = gen_replace_estimator(&police->tcf_bstats, NULL, &police->tcf_rate_est, - &police->tcf_lock, est); + &police->tcf_lock, + NULL, est); if (err) goto failure_unlock; } else if (tb[TCA_POLICE_AVRATE] && diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index ddf047d..d4a8bbf 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -982,7 +982,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, rcu_assign_pointer(sch->stab, stab); } if (tca[TCA_RATE]) { - spinlock_t *root_lock; + seqcount_t *running; err = -EOPNOTSUPP; if (sch->flags & TCQ_F_MQROOT) @@ -991,14 +991,15 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, if ((sch->parent != TC_H_ROOT) && !(sch->flags & TCQ_F_INGRESS) && (!p || !(p->flags & TCQ_F_MQROOT))) - root_lock = qdisc_root_sleeping_lock(sch); + running = qdisc_root_sleeping_running(sch); else - root_lock = qdisc_lock(sch); + running = &sch->running; err = gen_new_estimator(&sch->bstats, sch->cpu_bstats, &sch->rate_est, - root_lock, + NULL, + running, tca[TCA_RATE]); if (err) goto err_out4; @@ -1061,7 +1062,8 @@ static int qdisc_change(struct Qdisc *sch, struct nlattr **tca) gen_replace_estimator(&sch->bstats, sch->cpu_bstats, &sch->rate_est, - qdisc_root_sleeping_lock(sch), + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE]); } out: @@ -1369,8 +1371,7 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid, goto nla_put_failure; if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, - qdisc_root_sleeping_lock(q), &d, - TCA_PAD) < 0) + NULL, &d, TCA_PAD) < 0) goto nla_put_failure; if (q->ops->dump_stats && q->ops->dump_stats(q, &d) < 0) @@ -1381,7 +1382,8 @@ static int tc_fill_qdisc(struct sk_buff *skb, struct Qdisc *q, u32 clid, cpu_qstats = q->cpu_qstats; } - if (gnet_stats_copy_basic(&d, cpu_bstats, &q->bstats) < 0 || + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(q), + &d, cpu_bstats, &q->bstats) < 0 || gnet_stats_copy_rate_est(&d, &q->bstats, &q->rate_est) < 0 || gnet_stats_copy_queue(&d, cpu_qstats, &q->qstats, qlen) < 0) goto nla_put_failure; @@ -1684,8 +1686,7 @@ static int tc_fill_tclass(struct sk_buff *skb, struct Qdisc *q, goto nla_put_failure; if (gnet_stats_start_copy_compat(skb, TCA_STATS2, TCA_STATS, TCA_XSTATS, - qdisc_root_sleeping_lock(q), &d, - TCA_PAD) < 0) + NULL, &d, TCA_PAD) < 0) goto nla_put_failure; if (cl_ops->dump_stats && cl_ops->dump_stats(q, cl, &d) < 0) diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index 1911af3..34f8f79 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -637,7 +637,8 @@ atm_tc_dump_class_stats(struct Qdisc *sch, unsigned long arg, { struct atm_flow_data *flow = (struct atm_flow_data *)arg; - if (gnet_stats_copy_basic(d, NULL, &flow->bstats) < 0 || + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &flow->bstats) < 0 || gnet_stats_copy_queue(d, NULL, &flow->qstats, flow->q->q.qlen) < 0) return -1; diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index baafddf..1b8128f 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -1600,7 +1600,8 @@ cbq_dump_class_stats(struct Qdisc *sch, unsigned long arg, if (cl->undertime != PSCHED_PASTPERFECT) cl->xstats.undertime = cl->undertime - q->now; - if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 || + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &cl->bstats) < 0 || gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 || gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->q->q.qlen) < 0) return -1; @@ -1755,7 +1756,8 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t if (tca[TCA_RATE]) { err = gen_replace_estimator(&cl->bstats, NULL, &cl->rate_est, - qdisc_root_sleeping_lock(sch), + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE]); if (err) { qdisc_put_rtab(rtab); @@ -1848,7 +1850,8 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t if (tca[TCA_RATE]) { err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est, - qdisc_root_sleeping_lock(sch), + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE]); if (err) { kfree(cl); diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c index a63e879..1b7e1a2 100644 --- a/net/sched/sch_drr.c +++ b/net/sched/sch_drr.c @@ -91,7 +91,8 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (tca[TCA_RATE]) { err = gen_replace_estimator(&cl->bstats, NULL, &cl->rate_est, - qdisc_root_sleeping_lock(sch), + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE]); if (err) return err; @@ -119,7 +120,8 @@ static int drr_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (tca[TCA_RATE]) { err = gen_replace_estimator(&cl->bstats, NULL, &cl->rate_est, - qdisc_root_sleeping_lock(sch), + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE]); if (err) { qdisc_destroy(cl->qdisc); @@ -279,7 +281,8 @@ static int drr_dump_class_stats(struct Qdisc *sch, unsigned long arg, if (qlen) xstats.deficit = cl->deficit; - if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 || + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &cl->bstats) < 0 || gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 || gnet_stats_copy_queue(d, NULL, &cl->qdisc->qstats, qlen) < 0) return -1; diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index 6883a89..1daa542 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -566,11 +566,13 @@ static int fq_codel_dump_stats(struct Qdisc *sch, struct gnet_dump *d) st.qdisc_stats.memory_usage = q->memory_usage; st.qdisc_stats.drop_overmemory = q->drop_overmemory; + sch_tree_lock(sch); list_for_each(pos, &q->new_flows) st.qdisc_stats.new_flows_len++; list_for_each(pos, &q->old_flows) st.qdisc_stats.old_flows_len++; + sch_tree_unlock(sch); return gnet_stats_copy_app(d, &st, sizeof(st)); } @@ -624,7 +626,7 @@ static int fq_codel_dump_class_stats(struct Qdisc *sch, unsigned long cl, if (idx < q->flows_cnt) { const struct fq_codel_flow *flow = &q->flows[idx]; - const struct sk_buff *skb = flow->head; + const struct sk_buff *skb; memset(&xstats, 0, sizeof(xstats)); xstats.type = TCA_FQ_CODEL_XSTATS_CLASS; @@ -642,9 +644,14 @@ static int fq_codel_dump_class_stats(struct Qdisc *sch, unsigned long cl, codel_time_to_us(delta) : -codel_time_to_us(-delta); } - while (skb) { - qs.qlen++; - skb = skb->next; + if (flow->head) { + sch_tree_lock(sch); + skb = flow->head; + while (skb) { + qs.qlen++; + skb = skb->next; + } + sch_tree_unlock(sch); } qs.backlog = q->backlogs[idx]; qs.drops = flow->dropped; diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index d783d7c..74813dd 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1015,11 +1015,10 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, cur_time = psched_get_time(); if (tca[TCA_RATE]) { - spinlock_t *lock = qdisc_root_sleeping_lock(sch); - err = gen_replace_estimator(&cl->bstats, NULL, &cl->rate_est, - lock, + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE]); if (err) return err; @@ -1068,7 +1067,8 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (tca[TCA_RATE]) { err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est, - qdisc_root_sleeping_lock(sch), + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE]); if (err) { kfree(cl); @@ -1373,7 +1373,7 @@ hfsc_dump_class_stats(struct Qdisc *sch, unsigned long arg, xstats.work = cl->cl_total; xstats.rtwork = cl->cl_cumul; - if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 || + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), d, NULL, &cl->bstats) < 0 || gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 || gnet_stats_copy_queue(d, NULL, &cl->qstats, cl->qdisc->q.qlen) < 0) return -1; diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index d4b4218..2b05764 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1141,7 +1141,8 @@ htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d) cl->xstats.tokens = PSCHED_NS2TICKS(cl->tokens); cl->xstats.ctokens = PSCHED_NS2TICKS(cl->ctokens); - if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 || + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &cl->bstats) < 0 || gnet_stats_copy_rate_est(d, NULL, &cl->rate_est) < 0 || gnet_stats_copy_queue(d, NULL, &cl->qstats, qlen) < 0) return -1; @@ -1395,7 +1396,8 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, if (htb_rate_est || tca[TCA_RATE]) { err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est, - qdisc_root_sleeping_lock(sch), + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE] ? : &est.nla); if (err) { kfree(cl); @@ -1457,11 +1459,10 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, parent->children++; } else { if (tca[TCA_RATE]) { - spinlock_t *lock = qdisc_root_sleeping_lock(sch); - err = gen_replace_estimator(&cl->bstats, NULL, &cl->rate_est, - lock, + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE]); if (err) return err; diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c index 56a77b8..b943982 100644 --- a/net/sched/sch_mq.c +++ b/net/sched/sch_mq.c @@ -199,7 +199,7 @@ static int mq_dump_class_stats(struct Qdisc *sch, unsigned long cl, struct netdev_queue *dev_queue = mq_queue_get(sch, cl); sch = dev_queue->qdisc_sleeping; - if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 || + if (gnet_stats_copy_basic(&sch->running, d, NULL, &sch->bstats) < 0 || gnet_stats_copy_queue(d, NULL, &sch->qstats, sch->q.qlen) < 0) return -1; return 0; diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index b8002ce..549c663 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -342,7 +342,8 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl, * hold here is the look on dev_queue->qdisc_sleeping * also acquired below. */ - spin_unlock_bh(d->lock); + if (d->lock) + spin_unlock_bh(d->lock); for (i = tc.offset; i < tc.offset + tc.count; i++) { struct netdev_queue *q = netdev_get_tx_queue(dev, i); @@ -359,15 +360,17 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl, spin_unlock_bh(qdisc_lock(qdisc)); } /* Reclaim root sleeping lock before completing stats */ - spin_lock_bh(d->lock); - if (gnet_stats_copy_basic(d, NULL, &bstats) < 0 || + if (d->lock) + spin_lock_bh(d->lock); + if (gnet_stats_copy_basic(NULL, d, NULL, &bstats) < 0 || gnet_stats_copy_queue(d, NULL, &qstats, qlen) < 0) return -1; } else { struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl); sch = dev_queue->qdisc_sleeping; - if (gnet_stats_copy_basic(d, NULL, &sch->bstats) < 0 || + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &sch->bstats) < 0 || gnet_stats_copy_queue(d, NULL, &sch->qstats, sch->q.qlen) < 0) return -1; diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c index bcdd54b..21e69d2 100644 --- a/net/sched/sch_multiq.c +++ b/net/sched/sch_multiq.c @@ -356,7 +356,8 @@ static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl, struct Qdisc *cl_q; cl_q = q->queues[cl - 1]; - if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 || + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &cl_q->bstats) < 0 || gnet_stats_copy_queue(d, NULL, &cl_q->qstats, cl_q->q.qlen) < 0) return -1; diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index fee1b15..06eca70 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -319,7 +319,8 @@ static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl, struct Qdisc *cl_q; cl_q = q->queues[cl - 1]; - if (gnet_stats_copy_basic(d, NULL, &cl_q->bstats) < 0 || + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &cl_q->bstats) < 0 || gnet_stats_copy_queue(d, NULL, &cl_q->qstats, cl_q->q.qlen) < 0) return -1; diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 8d2d8d9..85d4197 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -460,7 +460,8 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (tca[TCA_RATE]) { err = gen_replace_estimator(&cl->bstats, NULL, &cl->rate_est, - qdisc_root_sleeping_lock(sch), + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE]); if (err) return err; @@ -486,7 +487,8 @@ static int qfq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (tca[TCA_RATE]) { err = gen_new_estimator(&cl->bstats, NULL, &cl->rate_est, - qdisc_root_sleeping_lock(sch), + NULL, + qdisc_root_sleeping_running(sch), tca[TCA_RATE]); if (err) goto destroy_class; @@ -663,7 +665,8 @@ static int qfq_dump_class_stats(struct Qdisc *sch, unsigned long arg, xstats.weight = cl->agg->class_weight; xstats.lmax = cl->agg->lmax; - if (gnet_stats_copy_basic(d, NULL, &cl->bstats) < 0 || + if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), + d, NULL, &cl->bstats) < 0 || gnet_stats_copy_rate_est(d, &cl->bstats, &cl->rate_est) < 0 || gnet_stats_copy_queue(d, NULL, &cl->qdisc->qstats, cl->qdisc->q.qlen) < 0) -- cgit v0.10.2 From c1e48af7960e93e1fbe54934be8f4a2fb66ef6fd Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 6 Jun 2016 16:06:02 -0700 Subject: gue: Implement direction IP encapsulation This patch implements direct encapsulation of IPv4 and IPv6 packets in UDP. This is done a version "1" of GUE and as explained in I-D draft-ietf-nvo3-gue-03. Changes here are only in the receive path, fou with IPxIPx already supports the transmit side. Both the normal receive path and GRO path are modified to check for GUE version and check for IP version in the case that GUE version is "1". Tested: IPIP with direct GUE encap 1 TCP_STREAM 4530 Mbps 200 TCP_RR 1297625 tps 135/232/444 90/95/99% latencies IP4IP6 with direct GUE encap 1 TCP_STREAM 4903 Mbps 200 TCP_RR 1184481 tps 149/253/473 90/95/99% latencies IP6IP6 direct GUE encap 1 TCP_STREAM 5146 Mbps 200 TCP_RR 1202879 tps 146/251/472 90/95/99% latencies SIT with direct GUE encap 1 TCP_STREAM 6111 Mbps 200 TCP_RR 1250337 tps 139/241/467 90/95/99% latencies Signed-off-by: Tom Herbert Signed-off-by: David S. Miller diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c index 5f9207c..321d57f 100644 --- a/net/ipv4/fou.c +++ b/net/ipv4/fou.c @@ -129,6 +129,36 @@ static int gue_udp_recv(struct sock *sk, struct sk_buff *skb) guehdr = (struct guehdr *)&udp_hdr(skb)[1]; + switch (guehdr->version) { + case 0: /* Full GUE header present */ + break; + + case 1: { + /* Direct encasulation of IPv4 or IPv6 */ + + int prot; + + switch (((struct iphdr *)guehdr)->version) { + case 4: + prot = IPPROTO_IPIP; + break; + case 6: + prot = IPPROTO_IPV6; + break; + default: + goto drop; + } + + if (fou_recv_pull(skb, fou, sizeof(struct udphdr))) + goto drop; + + return -prot; + } + + default: /* Undefined version */ + goto drop; + } + optlen = guehdr->hlen << 2; len += optlen; @@ -289,6 +319,7 @@ static struct sk_buff **gue_gro_receive(struct sock *sk, int flush = 1; struct fou *fou = fou_from_sock(sk); struct gro_remcsum grc; + u8 proto; skb_gro_remcsum_init(&grc); @@ -302,6 +333,25 @@ static struct sk_buff **gue_gro_receive(struct sock *sk, goto out; } + switch (guehdr->version) { + case 0: + break; + case 1: + switch (((struct iphdr *)guehdr)->version) { + case 4: + proto = IPPROTO_IPIP; + break; + case 6: + proto = IPPROTO_IPV6; + break; + default: + goto out; + } + goto next_proto; + default: + goto out; + } + optlen = guehdr->hlen << 2; len += optlen; @@ -370,6 +420,10 @@ static struct sk_buff **gue_gro_receive(struct sock *sk, } } + proto = guehdr->proto_ctype; + +next_proto: + /* We can clear the encap_mark for GUE as we are essentially doing * one of two possible things. We are either adding an L4 tunnel * header to the outer L3 tunnel header, or we are are simply @@ -383,7 +437,7 @@ static struct sk_buff **gue_gro_receive(struct sock *sk, rcu_read_lock(); offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads; - ops = rcu_dereference(offloads[guehdr->proto_ctype]); + ops = rcu_dereference(offloads[proto]); if (WARN_ON_ONCE(!ops || !ops->callbacks.gro_receive)) goto out_unlock; @@ -404,13 +458,30 @@ static int gue_gro_complete(struct sock *sk, struct sk_buff *skb, int nhoff) const struct net_offload **offloads; struct guehdr *guehdr = (struct guehdr *)(skb->data + nhoff); const struct net_offload *ops; - unsigned int guehlen; + unsigned int guehlen = 0; u8 proto; int err = -ENOENT; - proto = guehdr->proto_ctype; - - guehlen = sizeof(*guehdr) + (guehdr->hlen << 2); + switch (guehdr->version) { + case 0: + proto = guehdr->proto_ctype; + guehlen = sizeof(*guehdr) + (guehdr->hlen << 2); + break; + case 1: + switch (((struct iphdr *)guehdr)->version) { + case 4: + proto = IPPROTO_IPIP; + break; + case 6: + proto = IPPROTO_IPV6; + break; + default: + return err; + } + break; + default: + return err; + } rcu_read_lock(); offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads; -- cgit v0.10.2 From 911a66fbc83c7eeeac8872a5bfe284a240b3cc4a Mon Sep 17 00:00:00 2001 From: David Ahern Date: Mon, 6 Jun 2016 20:50:38 -0700 Subject: net: vrf: Minor refactoring for local address patches Move the stripping of the ethernet header from is_ip_tx_frame into the ipv4 and ipv6 outbound functions and collapse vrf_send_v4_prep into vrf_process_v4_outbound. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index d356f5d..ec0cb65 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -119,6 +119,9 @@ static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, skb_dst_drop(skb); skb_dst_set(skb, dst); + /* strip the ethernet header added for pass through VRF device */ + __skb_pull(skb, skb_network_offset(skb)); + ret = ip6_local_out(net, skb->sk, skb); if (unlikely(net_xmit_eval(ret))) dev->stats.tx_errors++; @@ -139,29 +142,6 @@ static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, } #endif -static int vrf_send_v4_prep(struct sk_buff *skb, struct flowi4 *fl4, - struct net_device *vrf_dev) -{ - struct rtable *rt; - int err = 1; - - rt = ip_route_output_flow(dev_net(vrf_dev), fl4, NULL); - if (IS_ERR(rt)) - goto out; - - /* TO-DO: what about broadcast ? */ - if (rt->rt_type != RTN_UNICAST && rt->rt_type != RTN_LOCAL) { - ip_rt_put(rt); - goto out; - } - - skb_dst_drop(skb); - skb_dst_set(skb, &rt->dst); - err = 0; -out: - return err; -} - static netdev_tx_t vrf_process_v4_outbound(struct sk_buff *skb, struct net_device *vrf_dev) { @@ -176,10 +156,24 @@ static netdev_tx_t vrf_process_v4_outbound(struct sk_buff *skb, FLOWI_FLAG_SKIP_NH_OIF, .daddr = ip4h->daddr, }; + struct net *net = dev_net(vrf_dev); + struct rtable *rt; - if (vrf_send_v4_prep(skb, &fl4, vrf_dev)) + rt = ip_route_output_flow(net, &fl4, NULL); + if (IS_ERR(rt)) goto err; + if (rt->rt_type != RTN_UNICAST && rt->rt_type != RTN_LOCAL) { + ip_rt_put(rt); + goto err; + } + + skb_dst_drop(skb); + skb_dst_set(skb, &rt->dst); + + /* strip the ethernet header added for pass through VRF device */ + __skb_pull(skb, skb_network_offset(skb)); + if (!ip4h->saddr) { ip4h->saddr = inet_select_addr(skb_dst(skb)->dev, 0, RT_SCOPE_LINK); @@ -200,9 +194,6 @@ err: static netdev_tx_t is_ip_tx_frame(struct sk_buff *skb, struct net_device *dev) { - /* strip the ethernet header added for pass through VRF device */ - __skb_pull(skb, skb_network_offset(skb)); - switch (skb->protocol) { case htons(ETH_P_IP): return vrf_process_v4_outbound(skb, dev); -- cgit v0.10.2 From afe80a4998efa17b4429ce9f151f957d3e6af317 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Mon, 6 Jun 2016 20:50:39 -0700 Subject: net: vrf: ipv4 support for local traffic to local addresses Add support for locally originated traffic to VRF-local addresses. If destination device for an skb is the loopback or VRF device then set its dst to a local version of the VRF cached dst_entry and call netif_rx to insert the packet onto the rx queue - similar to what is done for loopback. This patch handles IPv4 support; follow on patch handles IPv6. With this patch, ping, tcp and udp packets to a local IPv4 address are successfully routed: $ ip addr show dev eth1 4: eth1: mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000 link/ether 02:e0:f9:1c:b9:74 brd ff:ff:ff:ff:ff:ff inet 10.100.1.1/24 brd 10.100.1.255 scope global eth1 valid_lft forever preferred_lft forever inet6 2100:1::1/120 scope global valid_lft forever preferred_lft forever inet6 fe80::e0:f9ff:fe1c:b974/64 scope link valid_lft forever preferred_lft forever $ ping -c1 -I red 10.100.1.1 ping: Warning: source address might be selected on device other than red. PING 10.100.1.1 (10.100.1.1) from 10.100.1.1 red: 56(84) bytes of data. 64 bytes from 10.100.1.1: icmp_seq=1 ttl=64 time=0.057 ms This patch also enables use of IPv4 loopback address on the VRF device: $ ip addr add dev red 127.0.0.1/8 $ ping -c1 -I red 127.0.0.1 PING 127.0.0.1 (127.0.0.1) from 127.0.0.1 red: 56(84) bytes of data. 64 bytes from 127.0.0.1: icmp_seq=1 ttl=64 time=0.058 ms Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index ec0cb65..203d3c7 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -44,6 +44,7 @@ struct net_vrf { struct rtable __rcu *rth; + struct rtable __rcu *rth_local; struct rt6_info __rcu *rt6; u32 tb_id; }; @@ -54,9 +55,20 @@ struct pcpu_dstats { u64 tx_drps; u64 rx_pkts; u64 rx_bytes; + u64 rx_drps; struct u64_stats_sync syncp; }; +static void vrf_rx_stats(struct net_device *dev, int len) +{ + struct pcpu_dstats *dstats = this_cpu_ptr(dev->dstats); + + u64_stats_update_begin(&dstats->syncp); + dstats->rx_pkts++; + dstats->rx_bytes += len; + u64_stats_update_end(&dstats->syncp); +} + static void vrf_tx_error(struct net_device *vrf_dev, struct sk_buff *skb) { vrf_dev->stats.tx_errors++; @@ -91,6 +103,34 @@ static struct rtnl_link_stats64 *vrf_get_stats64(struct net_device *dev, return stats; } +/* Local traffic destined to local address. Reinsert the packet to rx + * path, similar to loopback handling. + */ +static int vrf_local_xmit(struct sk_buff *skb, struct net_device *dev, + struct dst_entry *dst) +{ + int len = skb->len; + + skb_orphan(skb); + + skb_dst_set(skb, dst); + skb_dst_force(skb); + + /* set pkt_type to avoid skb hitting packet taps twice - + * once on Tx and again in Rx processing + */ + skb->pkt_type = PACKET_LOOPBACK; + + skb->protocol = eth_type_trans(skb, dev); + + if (likely(netif_rx(skb) == NET_RX_SUCCESS)) + vrf_rx_stats(dev, len); + else + this_cpu_inc(dev->dstats->rx_drps); + + return NETDEV_TX_OK; +} + #if IS_ENABLED(CONFIG_IPV6) static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, struct net_device *dev) @@ -169,6 +209,34 @@ static netdev_tx_t vrf_process_v4_outbound(struct sk_buff *skb, } skb_dst_drop(skb); + + /* if dst.dev is loopback or the VRF device again this is locally + * originated traffic destined to a local address. Short circuit + * to Rx path using our local dst + */ + if (rt->dst.dev == net->loopback_dev || rt->dst.dev == vrf_dev) { + struct net_vrf *vrf = netdev_priv(vrf_dev); + struct rtable *rth_local; + struct dst_entry *dst = NULL; + + ip_rt_put(rt); + + rcu_read_lock(); + + rth_local = rcu_dereference(vrf->rth_local); + if (likely(rth_local)) { + dst = &rth_local->dst; + dst_hold(dst); + } + + rcu_read_unlock(); + + if (unlikely(!dst)) + goto err; + + return vrf_local_xmit(skb, vrf_dev, dst); + } + skb_dst_set(skb, &rt->dst); /* strip the ethernet header added for pass through VRF device */ @@ -375,29 +443,48 @@ static int vrf_output(struct net *net, struct sock *sk, struct sk_buff *skb) static void vrf_rtable_release(struct net_vrf *vrf) { struct rtable *rth = rtnl_dereference(vrf->rth); + struct rtable *rth_local = rtnl_dereference(vrf->rth_local); - rcu_assign_pointer(vrf->rth, NULL); + RCU_INIT_POINTER(vrf->rth, NULL); + RCU_INIT_POINTER(vrf->rth_local, NULL); + synchronize_rcu(); if (rth) dst_release(&rth->dst); + + if (rth_local) + dst_release(&rth_local->dst); } static int vrf_rtable_create(struct net_device *dev) { struct net_vrf *vrf = netdev_priv(dev); - struct rtable *rth; + struct rtable *rth, *rth_local; if (!fib_new_table(dev_net(dev), vrf->tb_id)) return -ENOMEM; + /* create a dst for routing packets out through a VRF device */ rth = rt_dst_alloc(dev, 0, RTN_UNICAST, 1, 1, 0); if (!rth) return -ENOMEM; + /* create a dst for local ingress routing - packets sent locally + * to local address via the VRF device as a loopback + */ + rth_local = rt_dst_alloc(dev, RTCF_LOCAL, RTN_LOCAL, 1, 1, 0); + if (!rth_local) { + dst_release(&rth->dst); + return -ENOMEM; + } + rth->dst.output = vrf_output; rth->rt_table_id = vrf->tb_id; + rth_local->rt_table_id = vrf->tb_id; + rcu_assign_pointer(vrf->rth, rth); + rcu_assign_pointer(vrf->rth_local, rth_local); return 0; } @@ -652,10 +739,19 @@ static struct sk_buff *vrf_ip_rcv(struct net_device *vrf_dev, skb->dev = vrf_dev; skb->skb_iif = vrf_dev->ifindex; + /* loopback traffic; do not push through packet taps again. + * Reset pkt_type for upper layers to process skb + */ + if (skb->pkt_type == PACKET_LOOPBACK) { + skb->pkt_type = PACKET_HOST; + goto out; + } + skb_push(skb, skb->mac_len); dev_queue_xmit_nit(skb, vrf_dev); skb_pull(skb, skb->mac_len); +out: return skb; } -- cgit v0.10.2 From b4869aa2f881ea4fcd36cd01ad591e4ed96eb33b Mon Sep 17 00:00:00 2001 From: David Ahern Date: Mon, 6 Jun 2016 20:50:40 -0700 Subject: net: vrf: ipv6 support for local traffic to local addresses Add support for locally originated traffic to VRF-local IPv6 addresses. Similar to IPv4 a local dst is set on the skb and the packet is reinserted with a call to netif_rx. With this patch, ping, tcp and udp packets to a local IPv6 address are successfully routed: $ ip addr show dev eth1 4: eth1: mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000 link/ether 02:e0:f9:1c:b9:74 brd ff:ff:ff:ff:ff:ff inet 10.100.1.1/24 brd 10.100.1.255 scope global eth1 valid_lft forever preferred_lft forever inet6 2100:1::1/120 scope global valid_lft forever preferred_lft forever inet6 fe80::e0:f9ff:fe1c:b974/64 scope link valid_lft forever preferred_lft forever $ ping6 -c1 -I red 2100:1::1 ping6: Warning: source address might be selected on device other than red. PING 2100:1::1(2100:1::1) from 2100:1::1 red: 56 data bytes 64 bytes from 2100:1::1: icmp_seq=1 ttl=64 time=0.098 ms ip6_input is exported so the VRF driver can use it for the dst input function. The dst_alloc function for IPv4 defaults to setting the input and output functions; IPv6's does not. VRF does not need to duplicate the Rx path so just export the ipv6 input function. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 203d3c7..1b214ea 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -46,6 +46,7 @@ struct net_vrf { struct rtable __rcu *rth; struct rtable __rcu *rth_local; struct rt6_info __rcu *rt6; + struct rt6_info __rcu *rt6_local; u32 tb_id; }; @@ -157,6 +158,46 @@ static netdev_tx_t vrf_process_v6_outbound(struct sk_buff *skb, goto err; skb_dst_drop(skb); + + /* if dst.dev is loopback or the VRF device again this is locally + * originated traffic destined to a local address. Short circuit + * to Rx path using our local dst + */ + if (dst->dev == net->loopback_dev || dst->dev == dev) { + struct net_vrf *vrf = netdev_priv(dev); + struct rt6_info *rt6_local; + + /* release looked up dst and use cached local dst */ + dst_release(dst); + + rcu_read_lock(); + + rt6_local = rcu_dereference(vrf->rt6_local); + if (unlikely(!rt6_local)) { + rcu_read_unlock(); + goto err; + } + + /* Ordering issue: cached local dst is created on newlink + * before the IPv6 initialization. Using the local dst + * requires rt6i_idev to be set so make sure it is. + */ + if (unlikely(!rt6_local->rt6i_idev)) { + rt6_local->rt6i_idev = in6_dev_get(dev); + if (!rt6_local->rt6i_idev) { + rcu_read_unlock(); + goto err; + } + } + + dst = &rt6_local->dst; + dst_hold(dst); + + rcu_read_unlock(); + + return vrf_local_xmit(skb, dev, &rt6_local->dst); + } + skb_dst_set(skb, dst); /* strip the ethernet header added for pass through VRF device */ @@ -336,27 +377,38 @@ static int vrf_output6(struct net *net, struct sock *sk, struct sk_buff *skb) static void vrf_rt6_release(struct net_vrf *vrf) { struct rt6_info *rt6 = rtnl_dereference(vrf->rt6); + struct rt6_info *rt6_local = rtnl_dereference(vrf->rt6_local); - rcu_assign_pointer(vrf->rt6, NULL); + RCU_INIT_POINTER(vrf->rt6, NULL); + RCU_INIT_POINTER(vrf->rt6_local, NULL); + synchronize_rcu(); if (rt6) dst_release(&rt6->dst); + + if (rt6_local) { + if (rt6_local->rt6i_idev) + in6_dev_put(rt6_local->rt6i_idev); + + dst_release(&rt6_local->dst); + } } static int vrf_rt6_create(struct net_device *dev) { + int flags = DST_HOST | DST_NOPOLICY | DST_NOXFRM | DST_NOCACHE; struct net_vrf *vrf = netdev_priv(dev); struct net *net = dev_net(dev); struct fib6_table *rt6i_table; - struct rt6_info *rt6; + struct rt6_info *rt6, *rt6_local; int rc = -ENOMEM; rt6i_table = fib6_new_table(net, vrf->tb_id); if (!rt6i_table) goto out; - rt6 = ip6_dst_alloc(net, dev, - DST_HOST | DST_NOPOLICY | DST_NOXFRM | DST_NOCACHE); + /* create a dst for routing packets out a VRF device */ + rt6 = ip6_dst_alloc(net, dev, flags); if (!rt6) goto out; @@ -364,7 +416,25 @@ static int vrf_rt6_create(struct net_device *dev) rt6->rt6i_table = rt6i_table; rt6->dst.output = vrf_output6; + + /* create a dst for local routing - packets sent locally + * to local address via the VRF device as a loopback + */ + rt6_local = ip6_dst_alloc(net, dev, flags); + if (!rt6_local) { + dst_release(&rt6->dst); + goto out; + } + + dst_hold(&rt6_local->dst); + + rt6_local->rt6i_idev = in6_dev_get(dev); + rt6_local->rt6i_flags = RTF_UP | RTF_NONEXTHOP | RTF_LOCAL; + rt6_local->rt6i_table = rt6i_table; + rt6_local->dst.input = ip6_input; + rcu_assign_pointer(vrf->rt6, rt6); + rcu_assign_pointer(vrf->rt6_local, rt6_local); rc = 0; out: @@ -710,6 +780,16 @@ out: static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, struct sk_buff *skb) { + /* loopback traffic; do not push through packet taps again. + * Reset pkt_type for upper layers to process skb + */ + if (skb->pkt_type == PACKET_LOOPBACK) { + skb->dev = vrf_dev; + skb->skb_iif = vrf_dev->ifindex; + skb->pkt_type = PACKET_HOST; + goto out; + } + /* if packet is NDISC keep the ingress interface */ if (!ipv6_ndisc_frame(skb)) { skb->dev = vrf_dev; @@ -722,6 +802,7 @@ static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, IP6CB(skb)->flags |= IP6SKB_L3SLAVE; } +out: return skb; } diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 94611e4..aacfb4b 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -323,6 +323,7 @@ int ip6_input(struct sk_buff *skb) dev_net(skb->dev), NULL, skb, skb->dev, NULL, ip6_input_finish); } +EXPORT_SYMBOL_GPL(ip6_input); int ip6_mc_input(struct sk_buff *skb) { -- cgit v0.10.2 From 01e517f16e38fc2345d4d555a7764b5f3f35af84 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 7 Jun 2016 15:04:16 +0300 Subject: qed: potential overflow in qed_cxt_src_t2_alloc() In the current code "ent_per_page" could be more than "conn_num" making "conn_num" negative after the subtraction. In the next iteration through the loop then the negative is treated as a very high positive meaning we don't put a limit on "ent_num". It could lead to memory corruption. Fixes: dbb799c39717 ('qed: Initialize hardware for new protocols') Signed-off-by: Dan Carpenter Acked-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c index d85b7ba..1c35f37 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c +++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c @@ -850,7 +850,7 @@ static int qed_cxt_src_t2_alloc(struct qed_hwfn *p_hwfn) val = 0; entries[j].next = cpu_to_be64(val); - conn_num -= ent_per_page; + conn_num -= ent_num; } return 0; -- cgit v0.10.2 From e00431bc93bb48c650273be4a00007b2a392d32a Mon Sep 17 00:00:00 2001 From: Pau Espin Pedrol Date: Tue, 7 Jun 2016 16:30:34 +0200 Subject: tcp: accept RST if SEQ matches right edge of right-most SACK block RFC 5961 advises to only accept RST packets containing a seq number matching the next expected seq number instead of the whole receive window in order to avoid spoofing attacks. However, this situation is not optimal in the case SACK is in use at the time the RST is sent. I recently run into a scenario in which packet losses were high while uploading data to a server, and userspace was willing to frequently terminate connections by sending a RST. In this case, the ACK sent on the receiver side (rcv_nxt) is frozen waiting for a lost packet retransmission and SACK blocks are used to let the client continue uploading data. At some point later on, the client sends the RST (snd_nxt), which matches the next expected seq number of the right-most SACK block on the receiver side which is going forward receiving data. In this scenario, as RFC 5961 defines, the RST SEQ doesn't match the frozen main ACK at receiver side and thus gets dropped and a challenge ACK is sent, which gets usually lost due to network conditions. The main consequence is that the connection stays alive for a while even if it made sense to accept the RST. This can get really bad if lots of connections like this one are created in few seconds, allocating all the resources of the server easily. For security reasons, not all SACK blocks are checked (there could be a big amount of SACK blocks => acceptable SEQ numbers). Furthermore, it wouldn't make sense to check for RST in blocks other than the right-most received one because the sender is not expected to be sending new data after the RST. For simplicity, only up to the 4 most recently updated SACK blocks (selective_acks[4] field) are compared to find the right-most block, as usually those are the ones with bigger probability to contain it. This patch was tested in a 3.18 kernel and probed to improve the situation in the scenario described above. Signed-off-by: Pau Espin Pedrol Acked-by: Eric Dumazet Acked-by: Neal Cardwell Tested-by: Neal Cardwell Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index d6c8f4cd0..89dd8d8 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5159,6 +5159,7 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, const struct tcphdr *th, int syn_inerr) { struct tcp_sock *tp = tcp_sk(sk); + bool rst_seq_match = false; /* RFC1323: H1. Apply PAWS check first. */ if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp && @@ -5195,13 +5196,32 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, /* Step 2: check RST bit */ if (th->rst) { - /* RFC 5961 3.2 : - * If sequence number exactly matches RCV.NXT, then + /* RFC 5961 3.2 (extend to match against SACK too if available): + * If seq num matches RCV.NXT or the right-most SACK block, + * then * RESET the connection * else * Send a challenge ACK */ - if (TCP_SKB_CB(skb)->seq == tp->rcv_nxt) + if (TCP_SKB_CB(skb)->seq == tp->rcv_nxt) { + rst_seq_match = true; + } else if (tcp_is_sack(tp) && tp->rx_opt.num_sacks > 0) { + struct tcp_sack_block *sp = &tp->selective_acks[0]; + int max_sack = sp[0].end_seq; + int this_sack; + + for (this_sack = 1; this_sack < tp->rx_opt.num_sacks; + ++this_sack) { + max_sack = after(sp[this_sack].end_seq, + max_sack) ? + sp[this_sack].end_seq : max_sack; + } + + if (TCP_SKB_CB(skb)->seq == max_sack) + rst_seq_match = true; + } + + if (rst_seq_match) tcp_reset(sk); else tcp_send_challenge_ack(sk, skb); -- cgit v0.10.2 From 707a2ca4870fcf6b5480cdfad563b940f56f0844 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Tue, 7 Jun 2016 16:09:44 -0700 Subject: ila: Perform only one translation in forwarding path When setting up ILA in a router we noticed that the the encapsulation is invoked twice: once in the route input path and again upon route output. To resolve this we add a flag set_csum_neutral for the ila_update_ipv6_locator. If this flag is set and the checksum neutral bit is also set we assume that checksum-neutral translation has already been performed and take no further action. The flag is set only in ila_output path. The flag is not set for ila_input and ila_xlat. Tested: Used 3 netns to set to emulate a router and two hosts. The router translates SIR addresses between the two destinations in other two netns. Verified ping and netperf are functional. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller diff --git a/net/ipv6/ila/ila.h b/net/ipv6/ila/ila.h index d08fd2d..e0170f6 100644 --- a/net/ipv6/ila/ila.h +++ b/net/ipv6/ila/ila.h @@ -109,7 +109,8 @@ static inline bool ila_csum_neutral_set(struct ila_identifier ident) return !!(ident.csum_neutral); } -void ila_update_ipv6_locator(struct sk_buff *skb, struct ila_params *p); +void ila_update_ipv6_locator(struct sk_buff *skb, struct ila_params *p, + bool set_csum_neutral); void ila_init_saved_csum(struct ila_params *p); diff --git a/net/ipv6/ila/ila_common.c b/net/ipv6/ila/ila_common.c index 0e94042..b3d00be 100644 --- a/net/ipv6/ila/ila_common.c +++ b/net/ipv6/ila/ila_common.c @@ -103,7 +103,8 @@ static void ila_csum_adjust_transport(struct sk_buff *skb, iaddr->loc = p->locator; } -void ila_update_ipv6_locator(struct sk_buff *skb, struct ila_params *p) +void ila_update_ipv6_locator(struct sk_buff *skb, struct ila_params *p, + bool set_csum_neutral) { struct ipv6hdr *ip6h = ipv6_hdr(skb); struct ila_addr *iaddr = ila_a2i(&ip6h->daddr); @@ -114,7 +115,8 @@ void ila_update_ipv6_locator(struct sk_buff *skb, struct ila_params *p) * is a locator being translated to a SIR address. * Perform (receiver) checksum-neutral translation. */ - ila_csum_do_neutral(iaddr, p); + if (!set_csum_neutral) + ila_csum_do_neutral(iaddr, p); } else { switch (p->csum_mode) { case ILA_CSUM_ADJUST_TRANSPORT: diff --git a/net/ipv6/ila/ila_lwt.c b/net/ipv6/ila/ila_lwt.c index 1dfb641..c8314c6 100644 --- a/net/ipv6/ila/ila_lwt.c +++ b/net/ipv6/ila/ila_lwt.c @@ -26,7 +26,7 @@ static int ila_output(struct net *net, struct sock *sk, struct sk_buff *skb) if (skb->protocol != htons(ETH_P_IPV6)) goto drop; - ila_update_ipv6_locator(skb, ila_params_lwtunnel(dst->lwtstate)); + ila_update_ipv6_locator(skb, ila_params_lwtunnel(dst->lwtstate), true); return dst->lwtstate->orig_output(net, sk, skb); @@ -42,7 +42,7 @@ static int ila_input(struct sk_buff *skb) if (skb->protocol != htons(ETH_P_IPV6)) goto drop; - ila_update_ipv6_locator(skb, ila_params_lwtunnel(dst->lwtstate)); + ila_update_ipv6_locator(skb, ila_params_lwtunnel(dst->lwtstate), false); return dst->lwtstate->orig_input(skb); diff --git a/net/ipv6/ila/ila_xlat.c b/net/ipv6/ila/ila_xlat.c index a90e572..e6eca5f 100644 --- a/net/ipv6/ila/ila_xlat.c +++ b/net/ipv6/ila/ila_xlat.c @@ -210,14 +210,14 @@ static void ila_free_cb(void *ptr, void *arg) } } -static int ila_xlat_addr(struct sk_buff *skb); +static int ila_xlat_addr(struct sk_buff *skb, bool set_csum_neutral); static unsigned int ila_nf_input(void *priv, struct sk_buff *skb, const struct nf_hook_state *state) { - ila_xlat_addr(skb); + ila_xlat_addr(skb, false); return NF_ACCEPT; } @@ -597,7 +597,7 @@ static struct pernet_operations ila_net_ops = { .size = sizeof(struct ila_net), }; -static int ila_xlat_addr(struct sk_buff *skb) +static int ila_xlat_addr(struct sk_buff *skb, bool set_csum_neutral) { struct ila_map *ila; struct ipv6hdr *ip6h = ipv6_hdr(skb); @@ -616,7 +616,7 @@ static int ila_xlat_addr(struct sk_buff *skb) ila = ila_lookup_wildcards(iaddr, skb->dev->ifindex, ilan); if (ila) - ila_update_ipv6_locator(skb, &ila->xp.ip); + ila_update_ipv6_locator(skb, &ila->xp.ip, set_csum_neutral); rcu_read_unlock(); -- cgit v0.10.2 From c57397670fe4430cc07572420d51220e9724543c Mon Sep 17 00:00:00 2001 From: Andreas Ziegler Date: Wed, 8 Jun 2016 11:40:28 +0200 Subject: drivers/net/fsl_ucc: Do not prefix header guard with CONFIG_ The CONFIG_ prefix should only be used for options which can be configured through Kconfig and not for guarding headers. Signed-off-by: Andreas Ziegler Signed-off-by: David S. Miller diff --git a/drivers/net/wan/fsl_ucc_hdlc.h b/drivers/net/wan/fsl_ucc_hdlc.h index 525786a..881ecde 100644 --- a/drivers/net/wan/fsl_ucc_hdlc.h +++ b/drivers/net/wan/fsl_ucc_hdlc.h @@ -8,8 +8,8 @@ * option) any later version. */ -#ifndef CONFIG_UCC_HDLC_H -#define CONFIG_UCC_HDLC_H +#ifndef _UCC_HDLC_H_ +#define _UCC_HDLC_H_ #include #include -- cgit v0.10.2 From 6f23d96cfa4fb68c4c9683f161f831057a5a134f Mon Sep 17 00:00:00 2001 From: Andreas Ziegler Date: Wed, 8 Jun 2016 11:36:56 +0200 Subject: fsl/qe: Do not prefix header guard with CONFIG_ The CONFIG_ prefix should only be used for options which can be configured through Kconfig and not for guarding headers. Signed-off-by: Andreas Ziegler Signed-off-by: David S. Miller diff --git a/include/soc/fsl/qe/qe_tdm.h b/include/soc/fsl/qe/qe_tdm.h index 4c91498..a1664b6 100644 --- a/include/soc/fsl/qe/qe_tdm.h +++ b/include/soc/fsl/qe/qe_tdm.h @@ -11,8 +11,8 @@ * option) any later version */ -#ifndef CONFIG_QE_TDM_H -#define CONFIG_QE_TDM_H +#ifndef _QE_TDM_H_ +#define _QE_TDM_H_ #include #include -- cgit v0.10.2 From 6ad8c632ee48ae099aa13704ef18a641220fe211 Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Wed, 8 Jun 2016 06:22:10 -0400 Subject: qed: Add support for query/config dcbx. Query API reads the dcbx data from the device shared memory and return it to the caller. The config API configures the user provided dcbx values on the device, and initiates the dcbx negotiation with the peer. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index 21ec1c2..e782484 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -252,7 +252,7 @@ qed_dcbx_process_tlv(struct qed_hwfn *p_hwfn, if (p_data->arr[type].update) continue; - enable = (type == DCBX_PROTOCOL_ETH) ? false : dcbx_enabled; + enable = !(type == DCBX_PROTOCOL_ETH); qed_dcbx_update_app_info(p_data, p_hwfn, enable, true, priority, tc, type); } @@ -351,6 +351,293 @@ qed_dcbx_copy_mib(struct qed_hwfn *p_hwfn, return rc; } +#ifdef CONFIG_DCB +static void +qed_dcbx_get_priority_info(struct qed_hwfn *p_hwfn, + struct qed_dcbx_app_prio *p_prio, + struct qed_dcbx_results *p_results) +{ + u8 val; + + p_prio->roce = QED_DCBX_INVALID_PRIORITY; + p_prio->roce_v2 = QED_DCBX_INVALID_PRIORITY; + p_prio->iscsi = QED_DCBX_INVALID_PRIORITY; + p_prio->fcoe = QED_DCBX_INVALID_PRIORITY; + + if (p_results->arr[DCBX_PROTOCOL_ROCE].update && + p_results->arr[DCBX_PROTOCOL_ROCE].enable) + p_prio->roce = p_results->arr[DCBX_PROTOCOL_ROCE].priority; + + if (p_results->arr[DCBX_PROTOCOL_ROCE_V2].update && + p_results->arr[DCBX_PROTOCOL_ROCE_V2].enable) { + val = p_results->arr[DCBX_PROTOCOL_ROCE_V2].priority; + p_prio->roce_v2 = val; + } + + if (p_results->arr[DCBX_PROTOCOL_ISCSI].update && + p_results->arr[DCBX_PROTOCOL_ISCSI].enable) + p_prio->iscsi = p_results->arr[DCBX_PROTOCOL_ISCSI].priority; + + if (p_results->arr[DCBX_PROTOCOL_FCOE].update && + p_results->arr[DCBX_PROTOCOL_FCOE].enable) + p_prio->fcoe = p_results->arr[DCBX_PROTOCOL_FCOE].priority; + + if (p_results->arr[DCBX_PROTOCOL_ETH].update && + p_results->arr[DCBX_PROTOCOL_ETH].enable) + p_prio->eth = p_results->arr[DCBX_PROTOCOL_ETH].priority; + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "Priorities: iscsi %d, roce %d, roce v2 %d, fcoe %d, eth %d\n", + p_prio->iscsi, p_prio->roce, p_prio->roce_v2, p_prio->fcoe, + p_prio->eth); +} + +static void +qed_dcbx_get_app_data(struct qed_hwfn *p_hwfn, + struct dcbx_app_priority_feature *p_app, + struct dcbx_app_priority_entry *p_tbl, + struct qed_dcbx_params *p_params) +{ + struct qed_app_entry *entry; + u8 pri_map; + int i; + + p_params->app_willing = QED_MFW_GET_FIELD(p_app->flags, + DCBX_APP_WILLING); + p_params->app_valid = QED_MFW_GET_FIELD(p_app->flags, DCBX_APP_ENABLED); + p_params->app_error = QED_MFW_GET_FIELD(p_app->flags, DCBX_APP_ERROR); + p_params->num_app_entries = QED_MFW_GET_FIELD(p_app->flags, + DCBX_APP_NUM_ENTRIES); + for (i = 0; i < DCBX_MAX_APP_PROTOCOL; i++) { + entry = &p_params->app_entry[i]; + entry->ethtype = !(QED_MFW_GET_FIELD(p_tbl[i].entry, + DCBX_APP_SF)); + pri_map = QED_MFW_GET_FIELD(p_tbl[i].entry, DCBX_APP_PRI_MAP); + entry->prio = ffs(pri_map) - 1; + entry->proto_id = QED_MFW_GET_FIELD(p_tbl[i].entry, + DCBX_APP_PROTOCOL_ID); + qed_dcbx_get_app_protocol_type(p_hwfn, p_tbl[i].entry, + entry->proto_id, + &entry->proto_type); + } + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "APP params: willing %d, valid %d error = %d\n", + p_params->app_willing, p_params->app_valid, + p_params->app_error); +} + +static void +qed_dcbx_get_pfc_data(struct qed_hwfn *p_hwfn, + u32 pfc, struct qed_dcbx_params *p_params) +{ + u8 pfc_map; + + p_params->pfc.willing = QED_MFW_GET_FIELD(pfc, DCBX_PFC_WILLING); + p_params->pfc.max_tc = QED_MFW_GET_FIELD(pfc, DCBX_PFC_CAPS); + p_params->pfc.enabled = QED_MFW_GET_FIELD(pfc, DCBX_PFC_ENABLED); + pfc_map = QED_MFW_GET_FIELD(pfc, DCBX_PFC_PRI_EN_BITMAP); + p_params->pfc.prio[0] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_0); + p_params->pfc.prio[1] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_1); + p_params->pfc.prio[2] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_2); + p_params->pfc.prio[3] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_3); + p_params->pfc.prio[4] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_4); + p_params->pfc.prio[5] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_5); + p_params->pfc.prio[6] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_6); + p_params->pfc.prio[7] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_7); + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "PFC params: willing %d, pfc_bitmap %d\n", + p_params->pfc.willing, pfc_map); +} + +static void +qed_dcbx_get_ets_data(struct qed_hwfn *p_hwfn, + struct dcbx_ets_feature *p_ets, + struct qed_dcbx_params *p_params) +{ + u32 bw_map[2], tsa_map[2], pri_map; + int i; + + p_params->ets_willing = QED_MFW_GET_FIELD(p_ets->flags, + DCBX_ETS_WILLING); + p_params->ets_enabled = QED_MFW_GET_FIELD(p_ets->flags, + DCBX_ETS_ENABLED); + p_params->ets_cbs = QED_MFW_GET_FIELD(p_ets->flags, DCBX_ETS_CBS); + p_params->max_ets_tc = QED_MFW_GET_FIELD(p_ets->flags, + DCBX_ETS_MAX_TCS); + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "ETS params: willing %d, ets_cbs %d pri_tc_tbl_0 %x max_ets_tc %d\n", + p_params->ets_willing, + p_params->ets_cbs, + p_ets->pri_tc_tbl[0], p_params->max_ets_tc); + + /* 8 bit tsa and bw data corresponding to each of the 8 TC's are + * encoded in a type u32 array of size 2. + */ + bw_map[0] = be32_to_cpu(p_ets->tc_bw_tbl[0]); + bw_map[1] = be32_to_cpu(p_ets->tc_bw_tbl[1]); + tsa_map[0] = be32_to_cpu(p_ets->tc_tsa_tbl[0]); + tsa_map[1] = be32_to_cpu(p_ets->tc_tsa_tbl[1]); + pri_map = be32_to_cpu(p_ets->pri_tc_tbl[0]); + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) { + p_params->ets_tc_bw_tbl[i] = ((u8 *)bw_map)[i]; + p_params->ets_tc_tsa_tbl[i] = ((u8 *)tsa_map)[i]; + p_params->ets_pri_tc_tbl[i] = QED_DCBX_PRIO2TC(pri_map, i); + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "elem %d bw_tbl %x tsa_tbl %x\n", + i, p_params->ets_tc_bw_tbl[i], + p_params->ets_tc_tsa_tbl[i]); + } +} + +static void +qed_dcbx_get_common_params(struct qed_hwfn *p_hwfn, + struct dcbx_app_priority_feature *p_app, + struct dcbx_app_priority_entry *p_tbl, + struct dcbx_ets_feature *p_ets, + u32 pfc, struct qed_dcbx_params *p_params) +{ + qed_dcbx_get_app_data(p_hwfn, p_app, p_tbl, p_params); + qed_dcbx_get_ets_data(p_hwfn, p_ets, p_params); + qed_dcbx_get_pfc_data(p_hwfn, pfc, p_params); +} + +static void +qed_dcbx_get_local_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, struct qed_dcbx_get *params) +{ + struct dcbx_features *p_feat; + + p_feat = &p_hwfn->p_dcbx_info->local_admin.features; + qed_dcbx_get_common_params(p_hwfn, &p_feat->app, + p_feat->app.app_pri_tbl, &p_feat->ets, + p_feat->pfc, ¶ms->local.params); + params->local.valid = true; +} + +static void +qed_dcbx_get_remote_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, struct qed_dcbx_get *params) +{ + struct dcbx_features *p_feat; + + p_feat = &p_hwfn->p_dcbx_info->remote.features; + qed_dcbx_get_common_params(p_hwfn, &p_feat->app, + p_feat->app.app_pri_tbl, &p_feat->ets, + p_feat->pfc, ¶ms->remote.params); + params->remote.valid = true; +} + +static void +qed_dcbx_get_operational_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_dcbx_get *params) +{ + struct qed_dcbx_operational_params *p_operational; + struct qed_dcbx_results *p_results; + struct dcbx_features *p_feat; + bool enabled, err; + u32 flags; + bool val; + + flags = p_hwfn->p_dcbx_info->operational.flags; + + /* If DCBx version is non zero, then negotiation + * was successfuly performed + */ + p_operational = ¶ms->operational; + enabled = !!(QED_MFW_GET_FIELD(flags, DCBX_CONFIG_VERSION) != + DCBX_CONFIG_VERSION_DISABLED); + if (!enabled) { + p_operational->enabled = enabled; + p_operational->valid = false; + return; + } + + p_feat = &p_hwfn->p_dcbx_info->operational.features; + p_results = &p_hwfn->p_dcbx_info->results; + + val = !!(QED_MFW_GET_FIELD(flags, DCBX_CONFIG_VERSION) == + DCBX_CONFIG_VERSION_IEEE); + p_operational->ieee = val; + val = !!(QED_MFW_GET_FIELD(flags, DCBX_CONFIG_VERSION) == + DCBX_CONFIG_VERSION_CEE); + p_operational->cee = val; + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, "Version support: ieee %d, cee %d\n", + p_operational->ieee, p_operational->cee); + + qed_dcbx_get_common_params(p_hwfn, &p_feat->app, + p_feat->app.app_pri_tbl, &p_feat->ets, + p_feat->pfc, ¶ms->operational.params); + qed_dcbx_get_priority_info(p_hwfn, &p_operational->app_prio, p_results); + err = QED_MFW_GET_FIELD(p_feat->app.flags, DCBX_APP_ERROR); + p_operational->err = err; + p_operational->enabled = enabled; + p_operational->valid = true; +} + +static void +qed_dcbx_get_local_lldp_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_dcbx_get *params) +{ + struct lldp_config_params_s *p_local; + + p_local = &p_hwfn->p_dcbx_info->lldp_local[LLDP_NEAREST_BRIDGE]; + + memcpy(params->lldp_local.local_chassis_id, p_local->local_chassis_id, + ARRAY_SIZE(p_local->local_chassis_id)); + memcpy(params->lldp_local.local_port_id, p_local->local_port_id, + ARRAY_SIZE(p_local->local_port_id)); +} + +static void +qed_dcbx_get_remote_lldp_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_dcbx_get *params) +{ + struct lldp_status_params_s *p_remote; + + p_remote = &p_hwfn->p_dcbx_info->lldp_remote[LLDP_NEAREST_BRIDGE]; + + memcpy(params->lldp_remote.peer_chassis_id, p_remote->peer_chassis_id, + ARRAY_SIZE(p_remote->peer_chassis_id)); + memcpy(params->lldp_remote.peer_port_id, p_remote->peer_port_id, + ARRAY_SIZE(p_remote->peer_port_id)); +} + +static int +qed_dcbx_get_params(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + struct qed_dcbx_get *p_params, + enum qed_mib_read_type type) +{ + switch (type) { + case QED_DCBX_REMOTE_MIB: + qed_dcbx_get_remote_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_LOCAL_MIB: + qed_dcbx_get_local_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_OPERATIONAL_MIB: + qed_dcbx_get_operational_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_REMOTE_LLDP_MIB: + qed_dcbx_get_remote_lldp_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_LOCAL_LLDP_MIB: + qed_dcbx_get_local_lldp_params(p_hwfn, p_ptt, p_params); + break; + default: + DP_ERR(p_hwfn, "MIB read err, unknown mib type %d\n", type); + return -EINVAL; + } + + return 0; +} +#endif + static int qed_dcbx_read_local_lldp_mib(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { @@ -561,3 +848,247 @@ void qed_dcbx_set_pf_update_params(struct qed_dcbx_results *p_src, p_dcb_data = &p_dest->eth_dcb_data; qed_dcbx_update_protocol_data(p_dcb_data, p_src, DCBX_PROTOCOL_ETH); } + +#ifdef CONFIG_DCB +static int qed_dcbx_query_params(struct qed_hwfn *p_hwfn, + struct qed_dcbx_get *p_get, + enum qed_mib_read_type type) +{ + struct qed_ptt *p_ptt; + int rc; + + p_ptt = qed_ptt_acquire(p_hwfn); + if (!p_ptt) + return -EBUSY; + + rc = qed_dcbx_read_mib(p_hwfn, p_ptt, type); + if (rc) + goto out; + + rc = qed_dcbx_get_params(p_hwfn, p_ptt, p_get, type); + +out: + qed_ptt_release(p_hwfn, p_ptt); + return rc; +} + +static void +qed_dcbx_set_pfc_data(struct qed_hwfn *p_hwfn, + u32 *pfc, struct qed_dcbx_params *p_params) +{ + u8 pfc_map = 0; + int i; + + if (p_params->pfc.willing) + *pfc |= DCBX_PFC_WILLING_MASK; + else + *pfc &= ~DCBX_PFC_WILLING_MASK; + + if (p_params->pfc.enabled) + *pfc |= DCBX_PFC_ENABLED_MASK; + else + *pfc &= ~DCBX_PFC_ENABLED_MASK; + + *pfc &= ~DCBX_PFC_CAPS_MASK; + *pfc |= (u32)p_params->pfc.max_tc << DCBX_PFC_CAPS_SHIFT; + + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + if (p_params->pfc.prio[i]) + pfc_map |= BIT(i); + + *pfc |= (pfc_map << DCBX_PFC_PRI_EN_BITMAP_SHIFT); + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, "pfc = 0x%x\n", *pfc); +} + +static void +qed_dcbx_set_ets_data(struct qed_hwfn *p_hwfn, + struct dcbx_ets_feature *p_ets, + struct qed_dcbx_params *p_params) +{ + u8 *bw_map, *tsa_map; + u32 val; + int i; + + if (p_params->ets_willing) + p_ets->flags |= DCBX_ETS_WILLING_MASK; + else + p_ets->flags &= ~DCBX_ETS_WILLING_MASK; + + if (p_params->ets_cbs) + p_ets->flags |= DCBX_ETS_CBS_MASK; + else + p_ets->flags &= ~DCBX_ETS_CBS_MASK; + + if (p_params->ets_enabled) + p_ets->flags |= DCBX_ETS_ENABLED_MASK; + else + p_ets->flags &= ~DCBX_ETS_ENABLED_MASK; + + p_ets->flags &= ~DCBX_ETS_MAX_TCS_MASK; + p_ets->flags |= (u32)p_params->max_ets_tc << DCBX_ETS_MAX_TCS_SHIFT; + + bw_map = (u8 *)&p_ets->tc_bw_tbl[0]; + tsa_map = (u8 *)&p_ets->tc_tsa_tbl[0]; + p_ets->pri_tc_tbl[0] = 0; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) { + bw_map[i] = p_params->ets_tc_bw_tbl[i]; + tsa_map[i] = p_params->ets_tc_tsa_tbl[i]; + /* Copy the priority value to the corresponding 4 bits in the + * traffic class table. + */ + val = (((u32)p_params->ets_pri_tc_tbl[i]) << ((7 - i) * 4)); + p_ets->pri_tc_tbl[0] |= val; + } + p_ets->pri_tc_tbl[0] = cpu_to_be32(p_ets->pri_tc_tbl[0]); + for (i = 0; i < 2; i++) { + p_ets->tc_bw_tbl[i] = cpu_to_be32(p_ets->tc_bw_tbl[i]); + p_ets->tc_tsa_tbl[i] = cpu_to_be32(p_ets->tc_tsa_tbl[i]); + } +} + +static void +qed_dcbx_set_app_data(struct qed_hwfn *p_hwfn, + struct dcbx_app_priority_feature *p_app, + struct qed_dcbx_params *p_params) +{ + u32 *entry; + int i; + + if (p_params->app_willing) + p_app->flags |= DCBX_APP_WILLING_MASK; + else + p_app->flags &= ~DCBX_APP_WILLING_MASK; + + if (p_params->app_valid) + p_app->flags |= DCBX_APP_ENABLED_MASK; + else + p_app->flags &= ~DCBX_APP_ENABLED_MASK; + + p_app->flags &= ~DCBX_APP_NUM_ENTRIES_MASK; + p_app->flags |= (u32)p_params->num_app_entries << + DCBX_APP_NUM_ENTRIES_SHIFT; + + for (i = 0; i < DCBX_MAX_APP_PROTOCOL; i++) { + entry = &p_app->app_pri_tbl[i].entry; + *entry &= ~DCBX_APP_SF_MASK; + if (p_params->app_entry[i].ethtype) + *entry |= ((u32)DCBX_APP_SF_ETHTYPE << + DCBX_APP_SF_SHIFT); + else + *entry |= ((u32)DCBX_APP_SF_PORT << DCBX_APP_SF_SHIFT); + *entry &= ~DCBX_APP_PROTOCOL_ID_MASK; + *entry |= ((u32)p_params->app_entry[i].proto_id << + DCBX_APP_PROTOCOL_ID_SHIFT); + *entry &= ~DCBX_APP_PRI_MAP_MASK; + *entry |= ((u32)(p_params->app_entry[i].prio) << + DCBX_APP_PRI_MAP_SHIFT); + } +} + +static void +qed_dcbx_set_local_params(struct qed_hwfn *p_hwfn, + struct dcbx_local_params *local_admin, + struct qed_dcbx_set *params) +{ + local_admin->flags = 0; + memcpy(&local_admin->features, + &p_hwfn->p_dcbx_info->operational.features, + sizeof(local_admin->features)); + + if (params->enabled) + local_admin->config = params->ver_num; + else + local_admin->config = DCBX_CONFIG_VERSION_DISABLED; + + if (params->override_flags & QED_DCBX_OVERRIDE_PFC_CFG) + qed_dcbx_set_pfc_data(p_hwfn, &local_admin->features.pfc, + ¶ms->config.params); + + if (params->override_flags & QED_DCBX_OVERRIDE_ETS_CFG) + qed_dcbx_set_ets_data(p_hwfn, &local_admin->features.ets, + ¶ms->config.params); + + if (params->override_flags & QED_DCBX_OVERRIDE_APP_CFG) + qed_dcbx_set_app_data(p_hwfn, &local_admin->features.app, + ¶ms->config.params); +} + +int qed_dcbx_config_params(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + struct qed_dcbx_set *params, bool hw_commit) +{ + struct dcbx_local_params local_admin; + struct qed_dcbx_mib_meta_data data; + u32 resp = 0, param = 0; + int rc = 0; + + if (!hw_commit) { + memcpy(&p_hwfn->p_dcbx_info->set, params, + sizeof(struct qed_dcbx_set)); + return 0; + } + + /* clear set-parmas cache */ + memset(&p_hwfn->p_dcbx_info->set, 0, sizeof(p_hwfn->p_dcbx_info->set)); + + memset(&local_admin, 0, sizeof(local_admin)); + qed_dcbx_set_local_params(p_hwfn, &local_admin, params); + + data.addr = p_hwfn->mcp_info->port_addr + + offsetof(struct public_port, local_admin_dcbx_mib); + data.local_admin = &local_admin; + data.size = sizeof(struct dcbx_local_params); + qed_memcpy_to(p_hwfn, p_ptt, data.addr, data.local_admin, data.size); + + rc = qed_mcp_cmd(p_hwfn, p_ptt, DRV_MSG_CODE_SET_DCBX, + 1 << DRV_MB_PARAM_LLDP_SEND_SHIFT, &resp, ¶m); + if (rc) + DP_NOTICE(p_hwfn, "Failed to send DCBX update request\n"); + + return rc; +} + +int qed_dcbx_get_config_params(struct qed_hwfn *p_hwfn, + struct qed_dcbx_set *params) +{ + struct qed_dcbx_get *dcbx_info; + int rc; + + if (p_hwfn->p_dcbx_info->set.config.valid) { + memcpy(params, &p_hwfn->p_dcbx_info->set, + sizeof(struct qed_dcbx_set)); + return 0; + } + + dcbx_info = kmalloc(sizeof(*dcbx_info), GFP_KERNEL); + if (!dcbx_info) { + DP_ERR(p_hwfn, "Failed to allocate struct qed_dcbx_info\n"); + return -ENOMEM; + } + + rc = qed_dcbx_query_params(p_hwfn, dcbx_info, QED_DCBX_OPERATIONAL_MIB); + if (rc) { + kfree(dcbx_info); + return rc; + } + + p_hwfn->p_dcbx_info->set.override_flags = 0; + p_hwfn->p_dcbx_info->set.ver_num = DCBX_CONFIG_VERSION_DISABLED; + if (dcbx_info->operational.cee) + p_hwfn->p_dcbx_info->set.ver_num |= DCBX_CONFIG_VERSION_CEE; + if (dcbx_info->operational.ieee) + p_hwfn->p_dcbx_info->set.ver_num |= DCBX_CONFIG_VERSION_IEEE; + + p_hwfn->p_dcbx_info->set.enabled = dcbx_info->operational.enabled; + memcpy(&p_hwfn->p_dcbx_info->set.config.params, + &dcbx_info->operational.params, + sizeof(struct qed_dcbx_admin_params)); + p_hwfn->p_dcbx_info->set.config.valid = true; + + memcpy(params, &p_hwfn->p_dcbx_info->set, sizeof(struct qed_dcbx_set)); + + kfree(dcbx_info); + + return 0; +} +#endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.h b/drivers/net/ethernet/qlogic/qed/qed_dcbx.h index e7f834d..9ba6816 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.h +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.h @@ -33,6 +33,24 @@ struct qed_dcbx_app_data { u8 tc; /* Traffic Class */ }; +#ifdef CONFIG_DCB +#define QED_DCBX_VERSION_DISABLED 0 +#define QED_DCBX_VERSION_IEEE 1 +#define QED_DCBX_VERSION_CEE 2 + +struct qed_dcbx_set { +#define QED_DCBX_OVERRIDE_STATE BIT(0) +#define QED_DCBX_OVERRIDE_PFC_CFG BIT(1) +#define QED_DCBX_OVERRIDE_ETS_CFG BIT(2) +#define QED_DCBX_OVERRIDE_APP_CFG BIT(3) +#define QED_DCBX_OVERRIDE_DSCP_CFG BIT(4) + u32 override_flags; + bool enabled; + struct qed_dcbx_admin_params config; + u32 ver_num; +}; +#endif + struct qed_dcbx_results { bool dcbx_enabled; u8 pf_id; @@ -55,6 +73,9 @@ struct qed_dcbx_info { struct qed_dcbx_results results; struct dcbx_mib operational; struct dcbx_mib remote; +#ifdef CONFIG_DCB + struct qed_dcbx_set set; +#endif u8 dcbx_cap; }; @@ -67,6 +88,13 @@ struct qed_dcbx_mib_meta_data { u32 addr; }; +#ifdef CONFIG_DCB +int qed_dcbx_get_config_params(struct qed_hwfn *, struct qed_dcbx_set *); + +int qed_dcbx_config_params(struct qed_hwfn *, + struct qed_ptt *, struct qed_dcbx_set *, bool); +#endif + /* QED local interface routines */ int qed_dcbx_mib_update_event(struct qed_hwfn *, diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index 72e0ad6..5927840 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -7244,6 +7244,9 @@ struct public_drv_mb { #define DRV_MB_PARAM_CFG_VF_MSIX_VF_ID_MASK 0x000000FF #define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_SHIFT 8 #define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_MASK 0x0000FF00 +#define DRV_MB_PARAM_LLDP_SEND_MASK 0x00000001 +#define DRV_MB_PARAM_LLDP_SEND_SHIFT 0 + #define DRV_MB_PARAM_SET_LED_MODE_OPER 0x0 #define DRV_MB_PARAM_SET_LED_MODE_ON 0x1 diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index e8cc49f..e1d5122 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -34,6 +34,96 @@ enum dcbx_protocol_type { DCBX_MAX_PROTOCOL_TYPE }; +#ifdef CONFIG_DCB +#define QED_LLDP_CHASSIS_ID_STAT_LEN 4 +#define QED_LLDP_PORT_ID_STAT_LEN 4 +#define QED_DCBX_MAX_APP_PROTOCOL 32 +#define QED_MAX_PFC_PRIORITIES 8 +#define QED_DCBX_DSCP_SIZE 64 + +struct qed_dcbx_lldp_remote { + u32 peer_chassis_id[QED_LLDP_CHASSIS_ID_STAT_LEN]; + u32 peer_port_id[QED_LLDP_PORT_ID_STAT_LEN]; + bool enable_rx; + bool enable_tx; + u32 tx_interval; + u32 max_credit; +}; + +struct qed_dcbx_lldp_local { + u32 local_chassis_id[QED_LLDP_CHASSIS_ID_STAT_LEN]; + u32 local_port_id[QED_LLDP_PORT_ID_STAT_LEN]; +}; + +struct qed_dcbx_app_prio { + u8 roce; + u8 roce_v2; + u8 fcoe; + u8 iscsi; + u8 eth; +}; + +struct qed_dbcx_pfc_params { + bool willing; + bool enabled; + u8 prio[QED_MAX_PFC_PRIORITIES]; + u8 max_tc; +}; + +struct qed_app_entry { + bool ethtype; + bool enabled; + u8 prio; + u16 proto_id; + enum dcbx_protocol_type proto_type; +}; + +struct qed_dcbx_params { + struct qed_app_entry app_entry[QED_DCBX_MAX_APP_PROTOCOL]; + u16 num_app_entries; + bool app_willing; + bool app_valid; + bool app_error; + bool ets_willing; + bool ets_enabled; + bool ets_cbs; + bool valid; + u8 ets_pri_tc_tbl[QED_MAX_PFC_PRIORITIES]; + u8 ets_tc_bw_tbl[QED_MAX_PFC_PRIORITIES]; + u8 ets_tc_tsa_tbl[QED_MAX_PFC_PRIORITIES]; + struct qed_dbcx_pfc_params pfc; + u8 max_ets_tc; +}; + +struct qed_dcbx_admin_params { + struct qed_dcbx_params params; + bool valid; +}; + +struct qed_dcbx_remote_params { + struct qed_dcbx_params params; + bool valid; +}; + +struct qed_dcbx_operational_params { + struct qed_dcbx_app_prio app_prio; + struct qed_dcbx_params params; + bool valid; + bool enabled; + bool ieee; + bool cee; + u32 err; +}; + +struct qed_dcbx_get { + struct qed_dcbx_operational_params operational; + struct qed_dcbx_lldp_remote lldp_remote; + struct qed_dcbx_lldp_local lldp_local; + struct qed_dcbx_remote_params remote; + struct qed_dcbx_admin_params local; +}; +#endif + enum qed_led_mode { QED_LED_MODE_OFF, QED_LED_MODE_ON, -- cgit v0.10.2 From a1d8d8a51e8317269dd127d94b9de14f67d9563f Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Wed, 8 Jun 2016 06:22:11 -0400 Subject: qed: Add dcbnl support. This patch adds the implementation for both cee/ieee dcbnl callbacks by using the qed query/config APIs. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index e782484..d0dc28f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -18,6 +19,9 @@ #include "qed_dcbx.h" #include "qed_hsi.h" #include "qed_sp.h" +#ifdef CONFIG_DCB +#include +#endif #define QED_DCBX_MAX_MIB_READ_TRY (100) #define QED_ETH_TYPE_DEFAULT (0) @@ -1091,4 +1095,1090 @@ int qed_dcbx_get_config_params(struct qed_hwfn *p_hwfn, return 0; } + +static struct qed_dcbx_get *qed_dcbnl_get_dcbx(struct qed_hwfn *hwfn, + enum qed_mib_read_type type) +{ + struct qed_dcbx_get *dcbx_info; + + dcbx_info = kmalloc(sizeof(*dcbx_info), GFP_KERNEL); + if (!dcbx_info) { + DP_ERR(hwfn->cdev, "Failed to allocate memory for dcbx_info\n"); + return NULL; + } + + if (qed_dcbx_query_params(hwfn, dcbx_info, type)) { + kfree(dcbx_info); + return NULL; + } + + if ((type == QED_DCBX_OPERATIONAL_MIB) && + !dcbx_info->operational.enabled) { + DP_INFO(hwfn, "DCBX is not enabled/operational\n"); + kfree(dcbx_info); + return NULL; + } + + return dcbx_info; +} + +static u8 qed_dcbnl_getstate(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + bool enabled; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 0; + + enabled = dcbx_info->operational.enabled; + DP_VERBOSE(hwfn, QED_MSG_DCB, "DCB state = %d\n", enabled); + kfree(dcbx_info); + + return enabled; +} + +static u8 qed_dcbnl_setstate(struct qed_dev *cdev, u8 state) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "DCB state = %d\n", state); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + dcbx_set.enabled = !!state; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc ? 1 : 0; +} + +static void qed_dcbnl_getpgtccfgtx(struct qed_dev *cdev, int tc, u8 *prio_type, + u8 *pgid, u8 *bw_pct, u8 *up_map) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "tc = %d\n", tc); + *prio_type = *pgid = *bw_pct = *up_map = 0; + if (tc < 0 || tc >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid tc %d\n", tc); + return; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return; + + *pgid = dcbx_info->operational.params.ets_pri_tc_tbl[tc]; + kfree(dcbx_info); +} + +static void qed_dcbnl_getpgbwgcfgtx(struct qed_dev *cdev, int pgid, u8 *bw_pct) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + *bw_pct = 0; + DP_VERBOSE(hwfn, QED_MSG_DCB, "pgid = %d\n", pgid); + if (pgid < 0 || pgid >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid pgid %d\n", pgid); + return; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return; + + *bw_pct = dcbx_info->operational.params.ets_tc_bw_tbl[pgid]; + DP_VERBOSE(hwfn, QED_MSG_DCB, "bw_pct = %d\n", *bw_pct); + kfree(dcbx_info); +} + +static void qed_dcbnl_getpgtccfgrx(struct qed_dev *cdev, int tc, u8 *prio, + u8 *bwg_id, u8 *bw_pct, u8 *up_map) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); + *prio = *bwg_id = *bw_pct = *up_map = 0; +} + +static void qed_dcbnl_getpgbwgcfgrx(struct qed_dev *cdev, + int bwg_id, u8 *bw_pct) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); + *bw_pct = 0; +} + +static void qed_dcbnl_getpfccfg(struct qed_dev *cdev, + int priority, u8 *setting) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "priority = %d\n", priority); + if (priority < 0 || priority >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid priority %d\n", priority); + return; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return; + + *setting = dcbx_info->operational.params.pfc.prio[priority]; + DP_VERBOSE(hwfn, QED_MSG_DCB, "setting = %d\n", *setting); + kfree(dcbx_info); +} + +static void qed_dcbnl_setpfccfg(struct qed_dev *cdev, int priority, u8 setting) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "priority = %d setting = %d\n", + priority, setting); + if (priority < 0 || priority >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid priority %d\n", priority); + return; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.prio[priority] = !!setting; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static u8 qed_dcbnl_getcap(struct qed_dev *cdev, int capid, u8 *cap) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int rc = 0; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "capid = %d\n", capid); + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 1; + + switch (capid) { + case DCB_CAP_ATTR_PG: + case DCB_CAP_ATTR_PFC: + case DCB_CAP_ATTR_UP2TC: + case DCB_CAP_ATTR_GSP: + *cap = true; + break; + case DCB_CAP_ATTR_PG_TCS: + case DCB_CAP_ATTR_PFC_TCS: + *cap = 0x80; + break; + case DCB_CAP_ATTR_DCBX: + *cap = (DCB_CAP_DCBX_LLD_MANAGED | DCB_CAP_DCBX_VER_CEE | + DCB_CAP_DCBX_VER_IEEE); + break; + default: + *cap = false; + rc = 1; + } + + DP_VERBOSE(hwfn, QED_MSG_DCB, "id = %d caps = %d\n", capid, *cap); + kfree(dcbx_info); + + return rc; +} + +static int qed_dcbnl_getnumtcs(struct qed_dev *cdev, int tcid, u8 *num) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int rc = 0; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "tcid = %d\n", tcid); + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + switch (tcid) { + case DCB_NUMTCS_ATTR_PG: + *num = dcbx_info->operational.params.max_ets_tc; + break; + case DCB_NUMTCS_ATTR_PFC: + *num = dcbx_info->operational.params.pfc.max_tc; + break; + default: + rc = -EINVAL; + } + + kfree(dcbx_info); + DP_VERBOSE(hwfn, QED_MSG_DCB, "numtcs = %d\n", *num); + + return rc; +} + +static u8 qed_dcbnl_getpfcstate(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + bool enabled; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 0; + + enabled = dcbx_info->operational.params.pfc.enabled; + DP_VERBOSE(hwfn, QED_MSG_DCB, "pfc state = %d\n", enabled); + kfree(dcbx_info); + + return enabled; +} + +static u8 qed_dcbnl_getdcbx(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + u8 mode = 0; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 0; + + if (dcbx_info->operational.enabled) + mode |= DCB_CAP_DCBX_LLD_MANAGED; + if (dcbx_info->operational.ieee) + mode |= DCB_CAP_DCBX_VER_IEEE; + if (dcbx_info->operational.cee) + mode |= DCB_CAP_DCBX_VER_CEE; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "dcb mode = %d\n", mode); + kfree(dcbx_info); + + return mode; +} + +static void qed_dcbnl_setpgtccfgtx(struct qed_dev *cdev, + int tc, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, + "tc = %d pri_type = %d pgid = %d bw_pct = %d up_map = %d\n", + tc, pri_type, pgid, bw_pct, up_map); + + if (tc < 0 || tc >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid tc %d\n", tc); + return; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.ets_pri_tc_tbl[tc] = pgid; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static void qed_dcbnl_setpgtccfgrx(struct qed_dev *cdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); +} + +static void qed_dcbnl_setpgbwgcfgtx(struct qed_dev *cdev, int pgid, u8 bw_pct) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "pgid = %d bw_pct = %d\n", pgid, bw_pct); + if (pgid < 0 || pgid >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid pgid %d\n", pgid); + return; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.ets_tc_bw_tbl[pgid] = bw_pct; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static void qed_dcbnl_setpgbwgcfgrx(struct qed_dev *cdev, int pgid, u8 bw_pct) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); +} + +static u8 qed_dcbnl_setall(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 1); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +static int qed_dcbnl_setnumtcs(struct qed_dev *cdev, int tcid, u8 num) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "tcid = %d num = %d\n", tcid, num); + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + switch (tcid) { + case DCB_NUMTCS_ATTR_PG: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.max_ets_tc = num; + break; + case DCB_NUMTCS_ATTR_PFC: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.max_tc = num; + break; + default: + DP_INFO(hwfn, "Invalid tcid %d\n", tcid); + return -EINVAL; + } + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EINVAL; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return 0; +} + +static void qed_dcbnl_setpfcstate(struct qed_dev *cdev, u8 state) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "new state = %d\n", state); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.enabled = !!state; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static int qed_dcbnl_getapp(struct qed_dev *cdev, u8 idtype, u16 idval) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_app_entry *entry; + bool ethtype; + u8 prio = 0; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + ethtype = !!(idtype == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_info->operational.params.app_entry[i]; + if ((entry->ethtype == ethtype) && (entry->proto_id == idval)) { + prio = entry->prio; + break; + } + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App entry (%d, %d) not found\n", idtype, idval); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + return prio; +} + +static int qed_dcbnl_setapp(struct qed_dev *cdev, + u8 idtype, u16 idval, u8 pri_map) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_app_entry *entry; + struct qed_ptt *ptt; + bool ethtype; + int rc, i; + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + ethtype = !!(idtype == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_set.config.params.app_entry[i]; + if ((entry->ethtype == ethtype) && (entry->proto_id == idval)) + break; + /* First empty slot */ + if (!entry->proto_id) + break; + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App table is full\n"); + return -EBUSY; + } + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_APP_CFG; + dcbx_set.config.params.app_entry[i].ethtype = ethtype; + dcbx_set.config.params.app_entry[i].proto_id = idval; + dcbx_set.config.params.app_entry[i].prio = pri_map; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EBUSY; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +static u8 qed_dcbnl_setdcbx(struct qed_dev *cdev, u8 mode) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "new mode = %x\n", mode); + + if (!(mode & DCB_CAP_DCBX_VER_IEEE) && !(mode & DCB_CAP_DCBX_VER_CEE)) { + DP_INFO(hwfn, "Allowed mode is cee, ieee or both\n"); + return 1; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + dcbx_set.ver_num = 0; + if (mode & DCB_CAP_DCBX_VER_CEE) { + dcbx_set.ver_num |= DCBX_CONFIG_VERSION_CEE; + dcbx_set.enabled = true; + } + + if (mode & DCB_CAP_DCBX_VER_IEEE) { + dcbx_set.ver_num |= DCBX_CONFIG_VERSION_IEEE; + dcbx_set.enabled = true; + } + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return 0; +} + +static u8 qed_dcbnl_getfeatcfg(struct qed_dev *cdev, int featid, u8 *flags) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "Feature id = %d\n", featid); + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 1; + + *flags = 0; + switch (featid) { + case DCB_FEATCFG_ATTR_PG: + if (dcbx_info->operational.params.ets_enabled) + *flags = DCB_FEATCFG_ENABLE; + else + *flags = DCB_FEATCFG_ERROR; + break; + case DCB_FEATCFG_ATTR_PFC: + if (dcbx_info->operational.params.pfc.enabled) + *flags = DCB_FEATCFG_ENABLE; + else + *flags = DCB_FEATCFG_ERROR; + break; + case DCB_FEATCFG_ATTR_APP: + if (dcbx_info->operational.params.app_valid) + *flags = DCB_FEATCFG_ENABLE; + else + *flags = DCB_FEATCFG_ERROR; + break; + default: + DP_INFO(hwfn, "Invalid feature-ID %d\n", featid); + kfree(dcbx_info); + return 1; + } + + DP_VERBOSE(hwfn, QED_MSG_DCB, "flags = %d\n", *flags); + kfree(dcbx_info); + + return 0; +} + +static u8 qed_dcbnl_setfeatcfg(struct qed_dev *cdev, int featid, u8 flags) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + bool enabled, willing; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "featid = %d flags = %d\n", + featid, flags); + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + enabled = !!(flags & DCB_FEATCFG_ENABLE); + willing = !!(flags & DCB_FEATCFG_WILLING); + switch (featid) { + case DCB_FEATCFG_ATTR_PG: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.ets_enabled = enabled; + dcbx_set.config.params.ets_willing = willing; + break; + case DCB_FEATCFG_ATTR_PFC: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.enabled = enabled; + dcbx_set.config.params.pfc.willing = willing; + break; + case DCB_FEATCFG_ATTR_APP: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_APP_CFG; + dcbx_set.config.params.app_willing = willing; + break; + default: + DP_INFO(hwfn, "Invalid feature-ID %d\n", featid); + return 1; + } + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return 0; +} + +static int qed_dcbnl_peer_getappinfo(struct qed_dev *cdev, + struct dcb_peer_app_info *info, + u16 *app_count) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + info->willing = dcbx_info->remote.params.app_willing; + info->error = dcbx_info->remote.params.app_error; + *app_count = dcbx_info->remote.params.num_app_entries; + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_peer_getapptable(struct qed_dev *cdev, + struct dcb_app *table) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + for (i = 0; i < dcbx_info->remote.params.num_app_entries; i++) { + if (dcbx_info->remote.params.app_entry[i].ethtype) + table[i].selector = DCB_APP_IDTYPE_ETHTYPE; + else + table[i].selector = DCB_APP_IDTYPE_PORTNUM; + table[i].priority = dcbx_info->remote.params.app_entry[i].prio; + table[i].protocol = + dcbx_info->remote.params.app_entry[i].proto_id; + } + + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_cee_peer_getpfc(struct qed_dev *cdev, struct cee_pfc *pfc) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + if (dcbx_info->remote.params.pfc.prio[i]) + pfc->pfc_en |= BIT(i); + + pfc->tcs_supported = dcbx_info->remote.params.pfc.max_tc; + DP_VERBOSE(hwfn, QED_MSG_DCB, "pfc state = %d tcs_supported = %d\n", + pfc->pfc_en, pfc->tcs_supported); + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_cee_peer_getpg(struct qed_dev *cdev, struct cee_pg *pg) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + pg->willing = dcbx_info->remote.params.ets_willing; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) { + pg->pg_bw[i] = dcbx_info->remote.params.ets_tc_bw_tbl[i]; + pg->prio_pg[i] = dcbx_info->remote.params.ets_pri_tc_tbl[i]; + } + + DP_VERBOSE(hwfn, QED_MSG_DCB, "willing = %d", pg->willing); + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_get_ieee_pfc(struct qed_dev *cdev, + struct ieee_pfc *pfc, bool remote) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_params *params; + struct qed_dcbx_get *dcbx_info; + int rc, i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + return -EINVAL; + } + + if (remote) { + memset(dcbx_info, 0, sizeof(*dcbx_info)); + rc = qed_dcbx_query_params(hwfn, dcbx_info, + QED_DCBX_REMOTE_MIB); + if (rc) { + kfree(dcbx_info); + return -EINVAL; + } + + params = &dcbx_info->remote.params; + } else { + params = &dcbx_info->operational.params; + } + + pfc->pfc_cap = params->pfc.max_tc; + pfc->pfc_en = 0; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + if (params->pfc.prio[i]) + pfc->pfc_en |= BIT(i); + + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_ieee_getpfc(struct qed_dev *cdev, struct ieee_pfc *pfc) +{ + return qed_dcbnl_get_ieee_pfc(cdev, pfc, false); +} + +static int qed_dcbnl_ieee_setpfc(struct qed_dev *cdev, struct ieee_pfc *pfc) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc, i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + dcbx_set.config.params.pfc.prio[i] = !!(pfc->pfc_en & BIT(i)); + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EINVAL; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +static int qed_dcbnl_get_ieee_ets(struct qed_dev *cdev, + struct ieee_ets *ets, bool remote) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_params *params; + int rc; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + if (remote) { + memset(dcbx_info, 0, sizeof(*dcbx_info)); + rc = qed_dcbx_query_params(hwfn, dcbx_info, + QED_DCBX_REMOTE_MIB); + if (rc) { + kfree(dcbx_info); + return -EINVAL; + } + + params = &dcbx_info->remote.params; + } else { + params = &dcbx_info->operational.params; + } + + ets->ets_cap = params->max_ets_tc; + ets->willing = params->ets_willing; + ets->cbs = params->ets_cbs; + memcpy(ets->tc_tx_bw, params->ets_tc_bw_tbl, sizeof(ets->tc_tx_bw)); + memcpy(ets->tc_tsa, params->ets_tc_tsa_tbl, sizeof(ets->tc_tsa)); + memcpy(ets->prio_tc, params->ets_pri_tc_tbl, sizeof(ets->prio_tc)); + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_ieee_getets(struct qed_dev *cdev, struct ieee_ets *ets) +{ + return qed_dcbnl_get_ieee_ets(cdev, ets, false); +} + +static int qed_dcbnl_ieee_setets(struct qed_dev *cdev, struct ieee_ets *ets) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.max_ets_tc = ets->ets_cap; + dcbx_set.config.params.ets_willing = ets->willing; + dcbx_set.config.params.ets_cbs = ets->cbs; + memcpy(dcbx_set.config.params.ets_tc_bw_tbl, ets->tc_tx_bw, + sizeof(ets->tc_tx_bw)); + memcpy(dcbx_set.config.params.ets_tc_tsa_tbl, ets->tc_tsa, + sizeof(ets->tc_tsa)); + memcpy(dcbx_set.config.params.ets_pri_tc_tbl, ets->prio_tc, + sizeof(ets->prio_tc)); + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EINVAL; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +int qed_dcbnl_ieee_peer_getets(struct qed_dev *cdev, struct ieee_ets *ets) +{ + return qed_dcbnl_get_ieee_ets(cdev, ets, true); +} + +int qed_dcbnl_ieee_peer_getpfc(struct qed_dev *cdev, struct ieee_pfc *pfc) +{ + return qed_dcbnl_get_ieee_pfc(cdev, pfc, true); +} + +int qed_dcbnl_ieee_getapp(struct qed_dev *cdev, struct dcb_app *app) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_app_entry *entry; + bool ethtype; + u8 prio = 0; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + /* ieee defines the selector field value for ethertype to be 1 */ + ethtype = !!((app->selector - 1) == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_info->operational.params.app_entry[i]; + if ((entry->ethtype == ethtype) && + (entry->proto_id == app->protocol)) { + prio = entry->prio; + break; + } + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App entry (%d, %d) not found\n", app->selector, + app->protocol); + kfree(dcbx_info); + return -EINVAL; + } + + app->priority = ffs(prio) - 1; + + kfree(dcbx_info); + + return 0; +} + +int qed_dcbnl_ieee_setapp(struct qed_dev *cdev, struct dcb_app *app) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_set dcbx_set; + struct qed_app_entry *entry; + struct qed_ptt *ptt; + bool ethtype; + int rc, i; + + if (app->priority < 0 || app->priority >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid priority %d\n", app->priority); + return -EINVAL; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + /* ieee defines the selector field value for ethertype to be 1 */ + ethtype = !!((app->selector - 1) == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_set.config.params.app_entry[i]; + if ((entry->ethtype == ethtype) && + (entry->proto_id == app->protocol)) + break; + /* First empty slot */ + if (!entry->proto_id) + break; + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App table is full\n"); + return -EBUSY; + } + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_APP_CFG; + dcbx_set.config.params.app_entry[i].ethtype = ethtype; + dcbx_set.config.params.app_entry[i].proto_id = app->protocol; + dcbx_set.config.params.app_entry[i].prio = BIT(app->priority); + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EBUSY; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +const struct qed_eth_dcbnl_ops qed_dcbnl_ops_pass = { + .getstate = qed_dcbnl_getstate, + .setstate = qed_dcbnl_setstate, + .getpgtccfgtx = qed_dcbnl_getpgtccfgtx, + .getpgbwgcfgtx = qed_dcbnl_getpgbwgcfgtx, + .getpgtccfgrx = qed_dcbnl_getpgtccfgrx, + .getpgbwgcfgrx = qed_dcbnl_getpgbwgcfgrx, + .getpfccfg = qed_dcbnl_getpfccfg, + .setpfccfg = qed_dcbnl_setpfccfg, + .getcap = qed_dcbnl_getcap, + .getnumtcs = qed_dcbnl_getnumtcs, + .getpfcstate = qed_dcbnl_getpfcstate, + .getdcbx = qed_dcbnl_getdcbx, + .setpgtccfgtx = qed_dcbnl_setpgtccfgtx, + .setpgtccfgrx = qed_dcbnl_setpgtccfgrx, + .setpgbwgcfgtx = qed_dcbnl_setpgbwgcfgtx, + .setpgbwgcfgrx = qed_dcbnl_setpgbwgcfgrx, + .setall = qed_dcbnl_setall, + .setnumtcs = qed_dcbnl_setnumtcs, + .setpfcstate = qed_dcbnl_setpfcstate, + .setapp = qed_dcbnl_setapp, + .setdcbx = qed_dcbnl_setdcbx, + .setfeatcfg = qed_dcbnl_setfeatcfg, + .getfeatcfg = qed_dcbnl_getfeatcfg, + .getapp = qed_dcbnl_getapp, + .peer_getappinfo = qed_dcbnl_peer_getappinfo, + .peer_getapptable = qed_dcbnl_peer_getapptable, + .cee_peer_getpfc = qed_dcbnl_cee_peer_getpfc, + .cee_peer_getpg = qed_dcbnl_cee_peer_getpg, + .ieee_getpfc = qed_dcbnl_ieee_getpfc, + .ieee_setpfc = qed_dcbnl_ieee_setpfc, + .ieee_getets = qed_dcbnl_ieee_getets, + .ieee_setets = qed_dcbnl_ieee_setets, + .ieee_peer_getpfc = qed_dcbnl_ieee_peer_getpfc, + .ieee_peer_getets = qed_dcbnl_ieee_peer_getets, + .ieee_getapp = qed_dcbnl_ieee_getapp, + .ieee_setapp = qed_dcbnl_ieee_setapp, +}; + #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c index 2ee496e..d121a8b 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_l2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c @@ -2166,11 +2166,18 @@ static int qed_fp_cqe_completion(struct qed_dev *dev, extern const struct qed_iov_hv_ops qed_iov_ops_pass; #endif +#ifdef CONFIG_DCB +extern const struct qed_eth_dcbnl_ops qed_dcbnl_ops_pass; +#endif + static const struct qed_eth_ops qed_eth_ops_pass = { .common = &qed_common_ops_pass, #ifdef CONFIG_QED_SRIOV .iov = &qed_iov_ops_pass, #endif +#ifdef CONFIG_DCB + .dcb = &qed_dcbnl_ops_pass, +#endif .fill_dev_info = &qed_fill_eth_dev_info, .register_ops = &qed_register_eth_ops, .check_mac = &qed_check_mac, diff --git a/include/linux/qed/qed_eth_if.h b/include/linux/qed/qed_eth_if.h index f8ff711..71d523b 100644 --- a/include/linux/qed/qed_eth_if.h +++ b/include/linux/qed/qed_eth_if.h @@ -128,11 +128,73 @@ struct qed_eth_cb_ops { void (*force_mac) (void *dev, u8 *mac); }; +#ifdef CONFIG_DCB +/* Prototype declaration of qed_eth_dcbnl_ops should match with the declaration + * of dcbnl_rtnl_ops structure. + */ +struct qed_eth_dcbnl_ops { + /* IEEE 802.1Qaz std */ + int (*ieee_getpfc)(struct qed_dev *cdev, struct ieee_pfc *pfc); + int (*ieee_setpfc)(struct qed_dev *cdev, struct ieee_pfc *pfc); + int (*ieee_getets)(struct qed_dev *cdev, struct ieee_ets *ets); + int (*ieee_setets)(struct qed_dev *cdev, struct ieee_ets *ets); + int (*ieee_peer_getets)(struct qed_dev *cdev, struct ieee_ets *ets); + int (*ieee_peer_getpfc)(struct qed_dev *cdev, struct ieee_pfc *pfc); + int (*ieee_getapp)(struct qed_dev *cdev, struct dcb_app *app); + int (*ieee_setapp)(struct qed_dev *cdev, struct dcb_app *app); + + /* CEE std */ + u8 (*getstate)(struct qed_dev *cdev); + u8 (*setstate)(struct qed_dev *cdev, u8 state); + void (*getpgtccfgtx)(struct qed_dev *cdev, int prio, u8 *prio_type, + u8 *pgid, u8 *bw_pct, u8 *up_map); + void (*getpgbwgcfgtx)(struct qed_dev *cdev, int pgid, u8 *bw_pct); + void (*getpgtccfgrx)(struct qed_dev *cdev, int prio, u8 *prio_type, + u8 *pgid, u8 *bw_pct, u8 *up_map); + void (*getpgbwgcfgrx)(struct qed_dev *cdev, int pgid, u8 *bw_pct); + void (*getpfccfg)(struct qed_dev *cdev, int prio, u8 *setting); + void (*setpfccfg)(struct qed_dev *cdev, int prio, u8 setting); + u8 (*getcap)(struct qed_dev *cdev, int capid, u8 *cap); + int (*getnumtcs)(struct qed_dev *cdev, int tcid, u8 *num); + u8 (*getpfcstate)(struct qed_dev *cdev); + int (*getapp)(struct qed_dev *cdev, u8 idtype, u16 id); + u8 (*getfeatcfg)(struct qed_dev *cdev, int featid, u8 *flags); + + /* DCBX configuration */ + u8 (*getdcbx)(struct qed_dev *cdev); + void (*setpgtccfgtx)(struct qed_dev *cdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map); + void (*setpgtccfgrx)(struct qed_dev *cdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map); + void (*setpgbwgcfgtx)(struct qed_dev *cdev, int pgid, u8 bw_pct); + void (*setpgbwgcfgrx)(struct qed_dev *cdev, int pgid, u8 bw_pct); + u8 (*setall)(struct qed_dev *cdev); + int (*setnumtcs)(struct qed_dev *cdev, int tcid, u8 num); + void (*setpfcstate)(struct qed_dev *cdev, u8 state); + int (*setapp)(struct qed_dev *cdev, u8 idtype, u16 idval, u8 up); + u8 (*setdcbx)(struct qed_dev *cdev, u8 state); + u8 (*setfeatcfg)(struct qed_dev *cdev, int featid, u8 flags); + + /* Peer apps */ + int (*peer_getappinfo)(struct qed_dev *cdev, + struct dcb_peer_app_info *info, + u16 *app_count); + int (*peer_getapptable)(struct qed_dev *cdev, struct dcb_app *table); + + /* CEE peer */ + int (*cee_peer_getpfc)(struct qed_dev *cdev, struct cee_pfc *pfc); + int (*cee_peer_getpg)(struct qed_dev *cdev, struct cee_pg *pg); +}; +#endif + struct qed_eth_ops { const struct qed_common_ops *common; #ifdef CONFIG_QED_SRIOV const struct qed_iov_hv_ops *iov; #endif +#ifdef CONFIG_DCB + const struct qed_eth_dcbnl_ops *dcb; +#endif int (*fill_dev_info)(struct qed_dev *cdev, struct qed_dev_eth_info *info); -- cgit v0.10.2 From 489e45ae42f000a5e045ac203ad5d6f08824cd56 Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Wed, 8 Jun 2016 06:22:12 -0400 Subject: qede: Add dcbnl support. This patch adds the interfaces for ieee/cee dcbnl callbacks and registers them with the kernel. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qede/Makefile b/drivers/net/ethernet/qlogic/qede/Makefile index 06ff90d..74a4985 100644 --- a/drivers/net/ethernet/qlogic/qede/Makefile +++ b/drivers/net/ethernet/qlogic/qede/Makefile @@ -1,3 +1,4 @@ obj-$(CONFIG_QEDE) := qede.o qede-y := qede_main.o qede_ethtool.o +qede-$(CONFIG_DCB) += qede_dcbnl.o diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 47d6b22..1441c8f 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -304,6 +304,9 @@ union qede_reload_args { u16 mtu; }; +#ifdef CONFIG_DCB +void qede_set_dcbnl_ops(struct net_device *ndev); +#endif void qede_config_debug(uint debug, u32 *p_dp_module, u8 *p_dp_level); void qede_set_ethtool_ops(struct net_device *netdev); void qede_reload(struct qede_dev *edev, diff --git a/drivers/net/ethernet/qlogic/qede/qede_dcbnl.c b/drivers/net/ethernet/qlogic/qede/qede_dcbnl.c new file mode 100644 index 0000000..03e8c02 --- /dev/null +++ b/drivers/net/ethernet/qlogic/qede/qede_dcbnl.c @@ -0,0 +1,348 @@ +/* QLogic qede NIC Driver +* Copyright (c) 2015 QLogic Corporation +* +* This software is available under the terms of the GNU General Public License +* (GPL) Version 2, available from the file COPYING in the main directory of +* this source tree. +*/ + +#include +#include +#include +#include +#include "qede.h" + +static u8 qede_dcbnl_getstate(struct net_device *netdev) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getstate(edev->cdev); +} + +static u8 qede_dcbnl_setstate(struct net_device *netdev, u8 state) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setstate(edev->cdev, state); +} + +static void qede_dcbnl_getpermhwaddr(struct net_device *netdev, + u8 *perm_addr) +{ + memcpy(perm_addr, netdev->dev_addr, netdev->addr_len); +} + +static void qede_dcbnl_getpgtccfgtx(struct net_device *netdev, int prio, + u8 *prio_type, u8 *pgid, u8 *bw_pct, + u8 *up_map) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->getpgtccfgtx(edev->cdev, prio, prio_type, + pgid, bw_pct, up_map); +} + +static void qede_dcbnl_getpgbwgcfgtx(struct net_device *netdev, + int pgid, u8 *bw_pct) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->getpgbwgcfgtx(edev->cdev, pgid, bw_pct); +} + +static void qede_dcbnl_getpgtccfgrx(struct net_device *netdev, int prio, + u8 *prio_type, u8 *pgid, u8 *bw_pct, + u8 *up_map) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->getpgtccfgrx(edev->cdev, prio, prio_type, pgid, bw_pct, + up_map); +} + +static void qede_dcbnl_getpgbwgcfgrx(struct net_device *netdev, + int pgid, u8 *bw_pct) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->getpgbwgcfgrx(edev->cdev, pgid, bw_pct); +} + +static void qede_dcbnl_getpfccfg(struct net_device *netdev, int prio, + u8 *setting) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->getpfccfg(edev->cdev, prio, setting); +} + +static void qede_dcbnl_setpfccfg(struct net_device *netdev, int prio, + u8 setting) +{ + struct qede_dev *edev = netdev_priv(netdev); + + edev->ops->dcb->setpfccfg(edev->cdev, prio, setting); +} + +static u8 qede_dcbnl_getcap(struct net_device *netdev, int capid, u8 *cap) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getcap(edev->cdev, capid, cap); +} + +static int qede_dcbnl_getnumtcs(struct net_device *netdev, int tcid, u8 *num) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getnumtcs(edev->cdev, tcid, num); +} + +static u8 qede_dcbnl_getpfcstate(struct net_device *netdev) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getpfcstate(edev->cdev); +} + +static int qede_dcbnl_getapp(struct net_device *netdev, u8 idtype, u16 id) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getapp(edev->cdev, idtype, id); +} + +static u8 qede_dcbnl_getdcbx(struct net_device *netdev) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getdcbx(edev->cdev); +} + +static void qede_dcbnl_setpgtccfgtx(struct net_device *netdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setpgtccfgtx(edev->cdev, prio, pri_type, pgid, + bw_pct, up_map); +} + +static void qede_dcbnl_setpgtccfgrx(struct net_device *netdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setpgtccfgrx(edev->cdev, prio, pri_type, pgid, + bw_pct, up_map); +} + +static void qede_dcbnl_setpgbwgcfgtx(struct net_device *netdev, int pgid, + u8 bw_pct) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setpgbwgcfgtx(edev->cdev, pgid, bw_pct); +} + +static void qede_dcbnl_setpgbwgcfgrx(struct net_device *netdev, int pgid, + u8 bw_pct) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setpgbwgcfgrx(edev->cdev, pgid, bw_pct); +} + +static u8 qede_dcbnl_setall(struct net_device *netdev) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setall(edev->cdev); +} + +static int qede_dcbnl_setnumtcs(struct net_device *netdev, int tcid, u8 num) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setnumtcs(edev->cdev, tcid, num); +} + +static void qede_dcbnl_setpfcstate(struct net_device *netdev, u8 state) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setpfcstate(edev->cdev, state); +} + +static int qede_dcbnl_setapp(struct net_device *netdev, u8 idtype, u16 idval, + u8 up) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setapp(edev->cdev, idtype, idval, up); +} + +static u8 qede_dcbnl_setdcbx(struct net_device *netdev, u8 state) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setdcbx(edev->cdev, state); +} + +static u8 qede_dcbnl_getfeatcfg(struct net_device *netdev, int featid, + u8 *flags) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->getfeatcfg(edev->cdev, featid, flags); +} + +static u8 qede_dcbnl_setfeatcfg(struct net_device *netdev, int featid, u8 flags) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->setfeatcfg(edev->cdev, featid, flags); +} + +static int qede_dcbnl_peer_getappinfo(struct net_device *netdev, + struct dcb_peer_app_info *info, + u16 *count) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->peer_getappinfo(edev->cdev, info, count); +} + +static int qede_dcbnl_peer_getapptable(struct net_device *netdev, + struct dcb_app *app) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->peer_getapptable(edev->cdev, app); +} + +static int qede_dcbnl_cee_peer_getpfc(struct net_device *netdev, + struct cee_pfc *pfc) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->cee_peer_getpfc(edev->cdev, pfc); +} + +static int qede_dcbnl_cee_peer_getpg(struct net_device *netdev, + struct cee_pg *pg) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->cee_peer_getpg(edev->cdev, pg); +} + +static int qede_dcbnl_ieee_getpfc(struct net_device *netdev, + struct ieee_pfc *pfc) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_getpfc(edev->cdev, pfc); +} + +static int qede_dcbnl_ieee_setpfc(struct net_device *netdev, + struct ieee_pfc *pfc) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_setpfc(edev->cdev, pfc); +} + +static int qede_dcbnl_ieee_getets(struct net_device *netdev, + struct ieee_ets *ets) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_getets(edev->cdev, ets); +} + +static int qede_dcbnl_ieee_setets(struct net_device *netdev, + struct ieee_ets *ets) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_setets(edev->cdev, ets); +} + +static int qede_dcbnl_ieee_getapp(struct net_device *netdev, + struct dcb_app *app) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_getapp(edev->cdev, app); +} + +static int qede_dcbnl_ieee_setapp(struct net_device *netdev, + struct dcb_app *app) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_setapp(edev->cdev, app); +} + +static int qede_dcbnl_ieee_peer_getpfc(struct net_device *netdev, + struct ieee_pfc *pfc) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_peer_getpfc(edev->cdev, pfc); +} + +static int qede_dcbnl_ieee_peer_getets(struct net_device *netdev, + struct ieee_ets *ets) +{ + struct qede_dev *edev = netdev_priv(netdev); + + return edev->ops->dcb->ieee_peer_getets(edev->cdev, ets); +} + +static const struct dcbnl_rtnl_ops qede_dcbnl_ops = { + .ieee_getpfc = qede_dcbnl_ieee_getpfc, + .ieee_setpfc = qede_dcbnl_ieee_setpfc, + .ieee_getets = qede_dcbnl_ieee_getets, + .ieee_setets = qede_dcbnl_ieee_setets, + .ieee_getapp = qede_dcbnl_ieee_getapp, + .ieee_setapp = qede_dcbnl_ieee_setapp, + .getdcbx = qede_dcbnl_getdcbx, + .ieee_peer_getpfc = qede_dcbnl_ieee_peer_getpfc, + .ieee_peer_getets = qede_dcbnl_ieee_peer_getets, + .getstate = qede_dcbnl_getstate, + .setstate = qede_dcbnl_setstate, + .getpermhwaddr = qede_dcbnl_getpermhwaddr, + .getpgtccfgtx = qede_dcbnl_getpgtccfgtx, + .getpgbwgcfgtx = qede_dcbnl_getpgbwgcfgtx, + .getpgtccfgrx = qede_dcbnl_getpgtccfgrx, + .getpgbwgcfgrx = qede_dcbnl_getpgbwgcfgrx, + .getpfccfg = qede_dcbnl_getpfccfg, + .setpfccfg = qede_dcbnl_setpfccfg, + .getcap = qede_dcbnl_getcap, + .getnumtcs = qede_dcbnl_getnumtcs, + .getpfcstate = qede_dcbnl_getpfcstate, + .getapp = qede_dcbnl_getapp, + .getdcbx = qede_dcbnl_getdcbx, + .setpgtccfgtx = qede_dcbnl_setpgtccfgtx, + .setpgtccfgrx = qede_dcbnl_setpgtccfgrx, + .setpgbwgcfgtx = qede_dcbnl_setpgbwgcfgtx, + .setpgbwgcfgrx = qede_dcbnl_setpgbwgcfgrx, + .setall = qede_dcbnl_setall, + .setnumtcs = qede_dcbnl_setnumtcs, + .setpfcstate = qede_dcbnl_setpfcstate, + .setapp = qede_dcbnl_setapp, + .setdcbx = qede_dcbnl_setdcbx, + .setfeatcfg = qede_dcbnl_setfeatcfg, + .getfeatcfg = qede_dcbnl_getfeatcfg, + .peer_getappinfo = qede_dcbnl_peer_getappinfo, + .peer_getapptable = qede_dcbnl_peer_getapptable, + .cee_peer_getpfc = qede_dcbnl_cee_peer_getpfc, + .cee_peer_getpg = qede_dcbnl_cee_peer_getpg, +}; + +void qede_set_dcbnl_ops(struct net_device *dev) +{ + dev->dcbnl_ops = &qede_dcbnl_ops; +} diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index edd2d90..f5962f6 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -2499,6 +2499,10 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, edev->ops->register_ops(cdev, &qede_ll_ops, edev); +#ifdef CONFIG_DCB + qede_set_dcbnl_ops(edev->ndev); +#endif + INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); mutex_init(&edev->qede_lock); -- cgit v0.10.2 From 40e4e713ebb279eb569584836d7cc6b799ed7f7f Mon Sep 17 00:00:00 2001 From: Hariprasad Shenai Date: Wed, 8 Jun 2016 18:09:08 +0530 Subject: net: Reduce queue allocation to one in kdump kernel When in kdump kernel, reduce memory usage by only using a single Queue Set for multiqueue devices. So make netif_get_num_default_rss_queues() return one, when in kdump kernel. Signed-off-by: Hariprasad Shenai Signed-off-by: David S. Miller diff --git a/net/core/dev.c b/net/core/dev.c index e0bcc39..c43c9d2 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -139,6 +139,7 @@ #include #include #include +#include #include "net-sysfs.h" @@ -2249,7 +2250,8 @@ EXPORT_SYMBOL(netif_set_real_num_rx_queues); */ int netif_get_num_default_rss_queues(void) { - return min_t(int, DEFAULT_MAX_NUM_RSS_QUEUES, num_online_cpus()); + return is_kdump_kernel() ? + 1 : min_t(int, DEFAULT_MAX_NUM_RSS_QUEUES, num_online_cpus()); } EXPORT_SYMBOL(netif_get_num_default_rss_queues); -- cgit v0.10.2 From 123b36526592f009bf8eccb7c8833aeda296d9cf Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 8 Jun 2016 07:22:49 -0700 Subject: net: sched: fix missing doc annotations "make htmldocs" complains otherwise: .//net/core/gen_stats.c:168: warning: No description found for parameter 'running' .//include/linux/netdevice.h:1867: warning: No description found for parameter 'qdisc_running_key' Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount") Fixes: edb09eb17ed8 ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump") Signed-off-by: Eric Dumazet Reported-by: kbuild test robot Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 59d7e06..5415623 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1594,7 +1594,8 @@ enum netdev_priv_flags { * @phydev: Physical device may attach itself * for hardware timestamping * - * @qdisc_tx_busylock: XXX: need comments on this one + * @qdisc_tx_busylock: lockdep class annotating Qdisc->busylock spinlock + * @qdisc_running_key: lockdep class annotating Qdisc->running seqcount * * @proto_down: protocol port state information can be sent to the * switch driver and used to set the phys state of the diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c index d9c210c..32207e6 100644 --- a/net/core/gen_stats.c +++ b/net/core/gen_stats.c @@ -150,6 +150,7 @@ EXPORT_SYMBOL(__gnet_stats_copy_basic); /** * gnet_stats_copy_basic - copy basic statistics into statistic TLV + * @running: seqcount_t pointer * @d: dumping handle * @cpu: copy statistic per cpu * @b: basic statistics -- cgit v0.10.2 From 0b7b498d41709d68bb4f520051f68d64f752beee Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 7 Jun 2016 16:32:38 -0700 Subject: net: dsa: Provide unique DSA slave MII bus names In case we have multiples trees and switches with the same index, we need to add another discriminating id: the switch tree. Reviewed-by: Andrew Lunn Reviewed-by: Vivien Didelot Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 15a4922..a51dfed 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -49,7 +49,8 @@ void dsa_slave_mii_bus_init(struct dsa_switch *ds) ds->slave_mii_bus->name = "dsa slave smi"; ds->slave_mii_bus->read = dsa_slave_phy_read; ds->slave_mii_bus->write = dsa_slave_phy_write; - snprintf(ds->slave_mii_bus->id, MII_BUS_ID_SIZE, "dsa-%d", ds->index); + snprintf(ds->slave_mii_bus->id, MII_BUS_ID_SIZE, "dsa-%d.%d", + ds->dst->tree, ds->index); ds->slave_mii_bus->parent = ds->dev; ds->slave_mii_bus->phy_mask = ~ds->phys_mii_mask; } -- cgit v0.10.2 From 6e830d8f0deca91fcb84d3156dcebb20384a9e2d Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 7 Jun 2016 16:32:39 -0700 Subject: net: dsa: Initialize ds->enabled_port_mask and ds->phys_mii_mask Some drivers rely on these two bitmasks to contain the correct values for them to successfully probe and initialize at drv->setup() time, calculate correct values to put in both masks as early as possible in dsa_get_ports_dn(). Reviewed-by: Andrew Lunn Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 80dfe08..921a36f 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -283,6 +283,7 @@ static void dsa_user_port_unapply(struct device_node *port, u32 index, if (ds->ports[index].netdev) { dsa_slave_destroy(ds->ports[index].netdev); ds->ports[index].netdev = NULL; + ds->enabled_port_mask &= ~(1 << index); } } @@ -292,6 +293,13 @@ static int dsa_ds_apply(struct dsa_switch_tree *dst, struct dsa_switch *ds) u32 index; int err; + /* Initialize ds->phys_mii_mask before registering the slave MDIO bus + * driver and before drv->setup() has run, since the switch drivers and + * the slave MDIO bus driver rely on these values for probing PHY + * devices or not + */ + ds->phys_mii_mask = ds->enabled_port_mask; + err = ds->drv->setup(ds); if (err < 0) return err; @@ -511,6 +519,13 @@ static int dsa_parse_ports_dn(struct device_node *ports, struct dsa_switch *ds) return -EINVAL; ds->ports[reg].dn = port; + + /* Initialize enabled_port_mask now for drv->setup() + * to have access to a correct value, just like what + * net/dsa/dsa.c::dsa_switch_setup_one does. + */ + if (!dsa_port_is_cpu(port)) + ds->enabled_port_mask |= 1 << reg; } return 0; -- cgit v0.10.2 From 1eb59443e72c69edbb836626f9f7f7e82427eeac Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 7 Jun 2016 16:32:40 -0700 Subject: net: dsa: Provide a slave MII bus if needed Mimic what net/dsa/dsa.c does and provide a slave MII bus by default which will be created if the driver implements a phy_read method. Reviewed-by: Andrew Lunn Reviewed-by: Vivien Didelot Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 921a36f..4e0f3c2 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -312,6 +312,18 @@ static int dsa_ds_apply(struct dsa_switch_tree *dst, struct dsa_switch *ds) if (err < 0) return err; + if (!ds->slave_mii_bus && ds->drv->phy_read) { + ds->slave_mii_bus = devm_mdiobus_alloc(ds->dev); + if (!ds->slave_mii_bus) + return -ENOMEM; + + dsa_slave_mii_bus_init(ds); + + err = mdiobus_register(ds->slave_mii_bus); + if (err < 0) + return err; + } + for (index = 0; index < DSA_MAX_PORTS; index++) { port = ds->ports[index].dn; if (!port) @@ -361,6 +373,9 @@ static void dsa_ds_unapply(struct dsa_switch_tree *dst, struct dsa_switch *ds) dsa_user_port_unapply(port, index, ds); } + + if (ds->slave_mii_bus && ds->drv->phy_read) + mdiobus_unregister(ds->slave_mii_bus); } static int dsa_dst_apply(struct dsa_switch_tree *dst) -- cgit v0.10.2 From af42192c47c41ec132bda736a78d6d5e0d2999a9 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 7 Jun 2016 16:32:41 -0700 Subject: net: dsa: Add initialization helper for CPU port ethtool_ops Add a helper function: dsa_cpu_port_ethtool_init() which initializes a custom ethtool_ops structure with custom DSA ethtool operations for CPU ports. This is a preliminary change to move the initialization outside of net/dsa/slave.c. Reviewed-by: Vivien Didelot Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index b42f1a5..106a9f0 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -58,6 +58,7 @@ const struct dsa_device_ops *dsa_resolve_tag_protocol(int tag_protocol); /* slave.c */ extern const struct dsa_device_ops notag_netdev_ops; void dsa_slave_mii_bus_init(struct dsa_switch *ds); +void dsa_cpu_port_ethtool_init(struct ethtool_ops *ops); int dsa_slave_create(struct dsa_switch *ds, struct device *parent, int port, const char *name); void dsa_slave_destroy(struct net_device *slave_dev); diff --git a/net/dsa/slave.c b/net/dsa/slave.c index a51dfed..8d15993 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -865,6 +865,13 @@ static void dsa_slave_poll_controller(struct net_device *dev) } #endif +void dsa_cpu_port_ethtool_init(struct ethtool_ops *ops) +{ + ops->get_sset_count = dsa_cpu_port_get_sset_count; + ops->get_ethtool_stats = dsa_cpu_port_get_ethtool_stats; + ops->get_strings = dsa_cpu_port_get_strings; +} + static const struct ethtool_ops dsa_slave_ethtool_ops = { .get_settings = dsa_slave_get_settings, .set_settings = dsa_slave_set_settings, @@ -1124,12 +1131,7 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent, sizeof(struct ethtool_ops)); memcpy(&dsa_cpu_port_ethtool_ops, &dst->master_ethtool_ops, sizeof(struct ethtool_ops)); - dsa_cpu_port_ethtool_ops.get_sset_count = - dsa_cpu_port_get_sset_count; - dsa_cpu_port_ethtool_ops.get_ethtool_stats = - dsa_cpu_port_get_ethtool_stats; - dsa_cpu_port_ethtool_ops.get_strings = - dsa_cpu_port_get_strings; + dsa_cpu_port_ethtool_init(&dsa_cpu_port_ethtool_ops); master->ethtool_ops = &dsa_cpu_port_ethtool_ops; } eth_hw_addr_inherit(slave_dev, master); -- cgit v0.10.2 From 0c73c523cf737b5d446705392e0e14ee0411a351 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 7 Jun 2016 16:32:42 -0700 Subject: net: dsa: Initialize CPU port ethtool ops per tree Now that we can properly support multiple distinct trees in the system, using a global variable: dsa_cpu_port_ethtool_ops is getting clobbered as soon as the second switch tree gets probed, and we don't want that. We need to move this to be dynamically allocated, and since we can't really be comparing addresses anymore to determine first time initialization versus any other times, just move this to dsa.c and dsa2.c where the remainder of the dst/ds initialization happens. The operations teardown restores the master netdev's ethtool_ops to its original ethtool_ops pointer (typically within the Ethernet driver) Signed-off-by: Florian Fainelli Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/include/net/dsa.h b/include/net/dsa.h index cca7ef2..20b3087 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -116,6 +116,7 @@ struct dsa_switch_tree { * Original copy of the master netdev ethtool_ops */ struct ethtool_ops master_ethtool_ops; + const struct ethtool_ops *master_orig_ethtool_ops; /* * The switch and port to which the CPU is attached. diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index ce3b942..766d2a5 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -266,6 +266,41 @@ const struct dsa_device_ops *dsa_resolve_tag_protocol(int tag_protocol) return ops; } +int dsa_cpu_port_ethtool_setup(struct dsa_switch *ds) +{ + struct net_device *master; + struct ethtool_ops *cpu_ops; + + master = ds->dst->master_netdev; + if (ds->master_netdev) + master = ds->master_netdev; + + cpu_ops = devm_kzalloc(ds->dev, sizeof(*cpu_ops), GFP_KERNEL); + if (!cpu_ops) + return -ENOMEM; + + memcpy(&ds->dst->master_ethtool_ops, master->ethtool_ops, + sizeof(struct ethtool_ops)); + ds->dst->master_orig_ethtool_ops = master->ethtool_ops; + memcpy(cpu_ops, &ds->dst->master_ethtool_ops, + sizeof(struct ethtool_ops)); + dsa_cpu_port_ethtool_init(cpu_ops); + master->ethtool_ops = cpu_ops; + + return 0; +} + +void dsa_cpu_port_ethtool_restore(struct dsa_switch *ds) +{ + struct net_device *master; + + master = ds->dst->master_netdev; + if (ds->master_netdev) + master = ds->master_netdev; + + master->ethtool_ops = ds->dst->master_orig_ethtool_ops; +} + static int dsa_switch_setup_one(struct dsa_switch *ds, struct device *parent) { struct dsa_switch_driver *drv = ds->drv; @@ -379,6 +414,10 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, struct device *parent) ret = 0; } + ret = dsa_cpu_port_ethtool_setup(ds); + if (ret) + return ret; + #ifdef CONFIG_NET_DSA_HWMON /* If the switch provides a temperature sensor, * register with hardware monitoring subsystem. @@ -963,6 +1002,8 @@ static void dsa_remove_dst(struct dsa_switch_tree *dst) dsa_switch_destroy(ds); } + dsa_cpu_port_ethtool_restore(dst->ds[0]); + dev_put(dst->master_netdev); } diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 4e0f3c2..83b95fc 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -394,6 +394,10 @@ static int dsa_dst_apply(struct dsa_switch_tree *dst) return err; } + err = dsa_cpu_port_ethtool_setup(dst->ds[0]); + if (err) + return err; + /* If we use a tagging format that doesn't have an ethertype * field, make sure that all packets from this point on get * sent to the tag format's receive function. @@ -429,6 +433,8 @@ static void dsa_dst_unapply(struct dsa_switch_tree *dst) dsa_ds_unapply(dst, ds); } + dsa_cpu_port_ethtool_restore(dst->ds[0]); + pr_info("DSA: tree %d unapplied\n", dst->tree); dst->applied = false; } diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index 106a9f0..00077a9 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -54,6 +54,8 @@ int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct device *dev, struct device_node *port_dn, int port); void dsa_cpu_dsa_destroy(struct device_node *port_dn); const struct dsa_device_ops *dsa_resolve_tag_protocol(int tag_protocol); +int dsa_cpu_port_ethtool_setup(struct dsa_switch *ds); +void dsa_cpu_port_ethtool_restore(struct dsa_switch *ds); /* slave.c */ extern const struct dsa_device_ops notag_netdev_ops; diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 8d15993..7236eb2 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -892,8 +892,6 @@ static const struct ethtool_ops dsa_slave_ethtool_ops = { .get_eee = dsa_slave_get_eee, }; -static struct ethtool_ops dsa_cpu_port_ethtool_ops; - static const struct net_device_ops dsa_slave_netdev_ops = { .ndo_open = dsa_slave_open, .ndo_stop = dsa_slave_close, @@ -1126,14 +1124,6 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent, slave_dev->features = master->vlan_features; slave_dev->ethtool_ops = &dsa_slave_ethtool_ops; - if (master->ethtool_ops != &dsa_cpu_port_ethtool_ops) { - memcpy(&dst->master_ethtool_ops, master->ethtool_ops, - sizeof(struct ethtool_ops)); - memcpy(&dsa_cpu_port_ethtool_ops, &dst->master_ethtool_ops, - sizeof(struct ethtool_ops)); - dsa_cpu_port_ethtool_init(&dsa_cpu_port_ethtool_ops); - master->ethtool_ops = &dsa_cpu_port_ethtool_ops; - } eth_hw_addr_inherit(slave_dev, master); slave_dev->priv_flags |= IFF_NO_QUEUE; slave_dev->netdev_ops = &dsa_slave_netdev_ops; -- cgit v0.10.2 From 461cd1b03e321e9c4ed1b831c80965c69a7b761b Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 7 Jun 2016 16:32:43 -0700 Subject: net: dsa: bcm_sf2: Register our slave MDIO bus Register a slave MDIO bus which allows us to divert problematic read/writes towards conflicting pseudo-PHY address (30). Do no longer rely on DSA's slave_mii_bus, but instead provide our own implementation which offers more flexibility as to what to do, and when to register it. We need to register it by the time we are able to get access to our memory mapped registers, which is not until drv->setup() time. In order to avoid forward declarations, we need to re-order the function bodies a bit. Reviewed-by: Andrew Lunn Reviewed-by: Vivien Didelot Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index 73df91bb..d662578 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -836,6 +837,66 @@ static int bcm_sf2_sw_fdb_dump(struct dsa_switch *ds, int port, return 0; } +static int bcm_sf2_sw_indir_rw(struct bcm_sf2_priv *priv, int op, int addr, + int regnum, u16 val) +{ + int ret = 0; + u32 reg; + + reg = reg_readl(priv, REG_SWITCH_CNTRL); + reg |= MDIO_MASTER_SEL; + reg_writel(priv, reg, REG_SWITCH_CNTRL); + + /* Page << 8 | offset */ + reg = 0x70; + reg <<= 2; + core_writel(priv, addr, reg); + + /* Page << 8 | offset */ + reg = 0x80 << 8 | regnum << 1; + reg <<= 2; + + if (op) + ret = core_readl(priv, reg); + else + core_writel(priv, val, reg); + + reg = reg_readl(priv, REG_SWITCH_CNTRL); + reg &= ~MDIO_MASTER_SEL; + reg_writel(priv, reg, REG_SWITCH_CNTRL); + + return ret & 0xffff; +} + +static int bcm_sf2_sw_mdio_read(struct mii_bus *bus, int addr, int regnum) +{ + struct bcm_sf2_priv *priv = bus->priv; + + /* Intercept reads from Broadcom pseudo-PHY address, else, send + * them to our master MDIO bus controller + */ + if (addr == BRCM_PSEUDO_PHY_ADDR && priv->indir_phy_mask & BIT(addr)) + return bcm_sf2_sw_indir_rw(priv, 1, addr, regnum, 0); + else + return mdiobus_read(priv->master_mii_bus, addr, regnum); +} + +static int bcm_sf2_sw_mdio_write(struct mii_bus *bus, int addr, int regnum, + u16 val) +{ + struct bcm_sf2_priv *priv = bus->priv; + + /* Intercept writes to the Broadcom pseudo-PHY address, else, + * send them to our master MDIO bus controller + */ + if (addr == BRCM_PSEUDO_PHY_ADDR && priv->indir_phy_mask & BIT(addr)) + bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val); + else + mdiobus_write(priv->master_mii_bus, addr, regnum, val); + + return 0; +} + static irqreturn_t bcm_sf2_switch_0_isr(int irq, void *dev_id) { struct bcm_sf2_priv *priv = dev_id; @@ -932,6 +993,72 @@ static void bcm_sf2_identify_ports(struct bcm_sf2_priv *priv, } } +static int bcm_sf2_mdio_register(struct dsa_switch *ds) +{ + struct bcm_sf2_priv *priv = ds_to_priv(ds); + struct device_node *dn; + static int index; + int err; + + /* Find our integrated MDIO bus node */ + dn = of_find_compatible_node(NULL, NULL, "brcm,unimac-mdio"); + priv->master_mii_bus = of_mdio_find_bus(dn); + if (!priv->master_mii_bus) + return -EPROBE_DEFER; + + get_device(&priv->master_mii_bus->dev); + priv->master_mii_dn = dn; + + priv->slave_mii_bus = devm_mdiobus_alloc(ds->dev); + if (!priv->slave_mii_bus) + return -ENOMEM; + + priv->slave_mii_bus->priv = priv; + priv->slave_mii_bus->name = "sf2 slave mii"; + priv->slave_mii_bus->read = bcm_sf2_sw_mdio_read; + priv->slave_mii_bus->write = bcm_sf2_sw_mdio_write; + snprintf(priv->slave_mii_bus->id, MII_BUS_ID_SIZE, "sf2-%d", + index++); + priv->slave_mii_bus->dev.of_node = dn; + + /* Include the pseudo-PHY address to divert reads towards our + * workaround. This is only required for 7445D0, since 7445E0 + * disconnects the internal switch pseudo-PHY such that we can use the + * regular SWITCH_MDIO master controller instead. + * + * Here we flag the pseudo PHY as needing special treatment and would + * otherwise make all other PHY read/writes go to the master MDIO bus + * controller that comes with this switch backed by the "mdio-unimac" + * driver. + */ + if (of_machine_is_compatible("brcm,bcm7445d0")) + priv->indir_phy_mask |= (1 << BRCM_PSEUDO_PHY_ADDR); + else + priv->indir_phy_mask = 0; + + ds->phys_mii_mask = priv->indir_phy_mask; + ds->slave_mii_bus = priv->slave_mii_bus; + priv->slave_mii_bus->parent = ds->dev->parent; + priv->slave_mii_bus->phy_mask = ~priv->indir_phy_mask; + + if (dn) + err = of_mdiobus_register(priv->slave_mii_bus, dn); + else + err = mdiobus_register(priv->slave_mii_bus); + + if (err) + of_node_put(dn); + + return err; +} + +static void bcm_sf2_mdio_unregister(struct bcm_sf2_priv *priv) +{ + mdiobus_unregister(priv->slave_mii_bus); + if (priv->master_mii_dn) + of_node_put(priv->master_mii_dn); +} + static int bcm_sf2_sw_setup(struct dsa_switch *ds) { const char *reg_names[BCM_SF2_REGS_NUM] = BCM_SF2_REGS_NAME; @@ -972,6 +1099,12 @@ static int bcm_sf2_sw_setup(struct dsa_switch *ds) goto out_unmap; } + ret = bcm_sf2_mdio_register(ds); + if (ret) { + pr_err("failed to register MDIO bus\n"); + goto out_unmap; + } + /* Disable all interrupts and request them */ bcm_sf2_intr_disable(priv); @@ -1017,23 +1150,6 @@ static int bcm_sf2_sw_setup(struct dsa_switch *ds) bcm_sf2_port_disable(ds, port, NULL); } - /* Include the pseudo-PHY address and the broadcast PHY address to - * divert reads towards our workaround. This is only required for - * 7445D0, since 7445E0 disconnects the internal switch pseudo-PHY such - * that we can use the regular SWITCH_MDIO master controller instead. - * - * By default, DSA initializes ds->phys_mii_mask to - * ds->enabled_port_mask to have a 1:1 mapping between Port address - * and PHY address in order to utilize the slave_mii_bus instance to - * read from Port PHYs. This is not what we want here, so we - * initialize phys_mii_mask 0 to always utilize the "master" MDIO - * bus backed by the "mdio-unimac" driver. - */ - if (of_machine_is_compatible("brcm,bcm7445d0")) - ds->phys_mii_mask |= ((1 << BRCM_PSEUDO_PHY_ADDR) | (1 << 0)); - else - ds->phys_mii_mask = 0; - rev = reg_readl(priv, REG_SWITCH_REVISION); priv->hw_params.top_rev = (rev >> SWITCH_TOP_REV_SHIFT) & SWITCH_TOP_REV_MASK; @@ -1058,6 +1174,7 @@ out_unmap: iounmap(*base); base++; } + bcm_sf2_mdio_unregister(priv); return ret; } @@ -1078,68 +1195,6 @@ static u32 bcm_sf2_sw_get_phy_flags(struct dsa_switch *ds, int port) return priv->hw_params.gphy_rev; } -static int bcm_sf2_sw_indir_rw(struct dsa_switch *ds, int op, int addr, - int regnum, u16 val) -{ - struct bcm_sf2_priv *priv = ds_to_priv(ds); - int ret = 0; - u32 reg; - - reg = reg_readl(priv, REG_SWITCH_CNTRL); - reg |= MDIO_MASTER_SEL; - reg_writel(priv, reg, REG_SWITCH_CNTRL); - - /* Page << 8 | offset */ - reg = 0x70; - reg <<= 2; - core_writel(priv, addr, reg); - - /* Page << 8 | offset */ - reg = 0x80 << 8 | regnum << 1; - reg <<= 2; - - if (op) - ret = core_readl(priv, reg); - else - core_writel(priv, val, reg); - - reg = reg_readl(priv, REG_SWITCH_CNTRL); - reg &= ~MDIO_MASTER_SEL; - reg_writel(priv, reg, REG_SWITCH_CNTRL); - - return ret & 0xffff; -} - -static int bcm_sf2_sw_phy_read(struct dsa_switch *ds, int addr, int regnum) -{ - /* Intercept reads from the MDIO broadcast address or Broadcom - * pseudo-PHY address - */ - switch (addr) { - case 0: - case BRCM_PSEUDO_PHY_ADDR: - return bcm_sf2_sw_indir_rw(ds, 1, addr, regnum, 0); - default: - return 0xffff; - } -} - -static int bcm_sf2_sw_phy_write(struct dsa_switch *ds, int addr, int regnum, - u16 val) -{ - /* Intercept writes to the MDIO broadcast address or Broadcom - * pseudo-PHY address - */ - switch (addr) { - case 0: - case BRCM_PSEUDO_PHY_ADDR: - bcm_sf2_sw_indir_rw(ds, 0, addr, regnum, val); - break; - } - - return 0; -} - static void bcm_sf2_sw_adjust_link(struct dsa_switch *ds, int port, struct phy_device *phydev) { @@ -1376,8 +1431,6 @@ static struct dsa_switch_driver bcm_sf2_switch_driver = { .setup = bcm_sf2_sw_setup, .set_addr = bcm_sf2_sw_set_addr, .get_phy_flags = bcm_sf2_sw_get_phy_flags, - .phy_read = bcm_sf2_sw_phy_read, - .phy_write = bcm_sf2_sw_phy_write, .get_strings = bcm_sf2_sw_get_strings, .get_ethtool_stats = bcm_sf2_sw_get_ethtool_stats, .get_sset_count = bcm_sf2_sw_get_sset_count, diff --git a/drivers/net/dsa/bcm_sf2.h b/drivers/net/dsa/bcm_sf2.h index 200b1f5..bde11eb 100644 --- a/drivers/net/dsa/bcm_sf2.h +++ b/drivers/net/dsa/bcm_sf2.h @@ -142,6 +142,12 @@ struct bcm_sf2_priv { /* Bitmask of ports having an integrated PHY */ unsigned int int_phy_mask; + + /* Master and slave MDIO bus controller */ + unsigned int indir_phy_mask; + struct device_node *master_mii_dn; + struct mii_bus *slave_mii_bus; + struct mii_bus *master_mii_bus; }; struct bcm_sf2_hw_stats { -- cgit v0.10.2 From c4282ca76c5b81ed73ef4c5eb5c07ee397e51642 Mon Sep 17 00:00:00 2001 From: Jon Paul Maloy Date: Wed, 8 Jun 2016 12:00:04 -0400 Subject: tipc: correct error in node fsm commit 88e8ac7000dc ("tipc: reduce transmission rate of reset messages when link is down") revealed a flaw in the node FSM, as defined in the log of commit 66996b6c47ed ("tipc: extend node FSM"). We see the following scenario: 1: Node B receives a RESET message from node A before its link endpoint is fully up, i.e., the node FSM is in state SELF_UP_PEER_COMING. This event will not change the node FSM state, but the (distinct) link FSM will move to state RESETTING. 2: As an effect of the previous event, the local endpoint on B will declare node A lost, and post the event SELF_DOWN to the its node FSM. This moves the FSM state to SELF_DOWN_PEER_LEAVING, meaning that no messages will be accepted from A until it receives another RESET message that confirms that A's endpoint has been reset. This is wasteful, since we know this as a fact already from the first received RESET, but worse is that the link instance's FSM has not wasted this information, but instead moved on to state ESTABLISHING, meaning that it repeatedly sends out ACTIVATE messages to the reset peer A. 3: Node A will receive one of the ACTIVATE messages, move its link FSM to state ESTABLISHED, and start repeatedly sending out STATE messages to node B. 4: Node B will consistently drop these messages, since it can only accept accept a RESET according to its node FSM. 5: After four lost STATE messages node A will reset its link and start repeatedly sending out RESET messages to B. 6: Because of the reduced send rate for RESET messages, it is very likely that A will receive an ACTIVATE (which is sent out at a much higher frequency) before it gets the chance to send a RESET, and A may hence quickly move back to state ESTABLISHED and continue sending out STATE messages, which will again be dropped by B. 7: GOTO 5. 8: After having repeated the cycle 5-7 a number of times, node A will by chance get in between with sending a RESET, and the situation is resolved. Unfortunately, we have seen that it may take a substantial amount of time before this vicious loop is broken, sometimes in the order of minutes. We correct this by making a small correction to the node FSM: When a node in state SELF_UP_PEER_COMING receives a SELF_DOWN event, it now moves directly back to state SELF_DOWN_PEER_DOWN, instead of as now SELF_DOWN_PEER_LEAVING. This is logically consistent, since we don't need to wait for RESET confirmation from of an endpoint that we alread know has been reset. It also means that node B in the scenario above will not be dropping incoming STATE messages, and the link can come up immediately. Finally, a symmetry comparison reveals that the FSM has a similar error when receiving the event PEER_DOWN in state PEER_UP_SELF_COMING. Instead of moving to PERR_DOWN_SELF_LEAVING, it should move directly to SELF_DOWN_PEER_DOWN. Although we have never seen any negative effect of this logical error, we choose fix this one, too. The node FSM looks as follows after those changes: +----------------------------------------+ | PEER_DOWN_EVT| | | +------------------------+----------------+ | |SELF_DOWN_EVT | | | | | | | | +-----------+ +-----------+ | | |NODE_ | |NODE_ | | | +----------|FAILINGOVER|<---------|SYNCHING |-----------+ | | |SELF_ +-----------+ FAILOVER_+-----------+ PEER_ | | | |DOWN_EVT | A BEGIN_EVT A | DOWN_EVT| | | | | | | | | | | | | | | | | | | | |FAILOVER_ |FAILOVER_ |SYNCH_ |SYNCH_ | | | | |END_EVT |BEGIN_EVT |BEGIN_EVT|END_EVT | | | | | | | | | | | | | | | | | | | | | +--------------+ | | | | | +-------->| SELF_UP_ |<-------+ | | | | +-----------------| PEER_UP |----------------+ | | | | |SELF_DOWN_EVT +--------------+ PEER_DOWN_EVT| | | | | | A A | | | | | | | | | | | | | | PEER_UP_EVT| |SELF_UP_EVT | | | | | | | | | | | V V V | | V V V +------------+ +-----------+ +-----------+ +------------+ |SELF_DOWN_ | |SELF_UP_ | |PEER_UP_ | |PEER_DOWN | |PEER_LEAVING| |PEER_COMING| |SELF_COMING| |SELF_LEAVING| +------------+ +-----------+ +-----------+ +------------+ | | A A | | | | | | | | | SELF_ | |SELF_ |PEER_ |PEER_ | | DOWN_EVT| |UP_EVT |UP_EVT |DOWN_EVT | | | | | | | | | | | | | | | +--------------+ | | |PEER_DOWN_EVT +--->| SELF_DOWN_ |<---+ SELF_DOWN_EVT| +------------------->| PEER_DOWN |<--------------------+ +--------------+ Acked-by: Ying Xue Signed-off-by: Jon Maloy Signed-off-by: David S. Miller diff --git a/net/tipc/node.c b/net/tipc/node.c index e01e2c71..c7985b2 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -950,7 +950,7 @@ static void tipc_node_fsm_evt(struct tipc_node *n, int evt) state = SELF_UP_PEER_UP; break; case SELF_LOST_CONTACT_EVT: - state = SELF_DOWN_PEER_LEAVING; + state = SELF_DOWN_PEER_DOWN; break; case SELF_ESTABL_CONTACT_EVT: case PEER_LOST_CONTACT_EVT: @@ -969,7 +969,7 @@ static void tipc_node_fsm_evt(struct tipc_node *n, int evt) state = SELF_UP_PEER_UP; break; case PEER_LOST_CONTACT_EVT: - state = SELF_LEAVING_PEER_DOWN; + state = SELF_DOWN_PEER_DOWN; break; case SELF_LOST_CONTACT_EVT: case PEER_ESTABL_CONTACT_EVT: -- cgit v0.10.2 From 5ca509fc0b6bcfeccf03c8c4bb5e4d1a62720c03 Mon Sep 17 00:00:00 2001 From: Jon Paul Maloy Date: Wed, 8 Jun 2016 12:00:05 -0400 Subject: tipc: change node timer unit from jiffies to ms The node keepalive interval is recalculated at each timer expiration to catch any changes in the link tolerance, and stored in a field in struct tipc_node. We use jiffies as unit for the stored value. This is suboptimal, because it makes the calculation unnecessary complex, including two unit conversions. The conversions also lead to a rounding error that causes the link "abort limit" to be 3 in the normal case, instead of 4, as intended. This again leads to unnecessary link resets when the network is pushed close to its limit, e.g., in an environment with hundreds of nodes or namesapces. In this commit, we do instead let the keepalive value be calculated and stored in milliseconds, so that there is only one conversion and the rounding error is eliminated. We also remove a redundant "keepalive" field in struct tipc_link. This is remnant from the previous implementation. Acked-by: Ying Xue Signed-off-by: Jon Maloy Signed-off-by: David S. Miller diff --git a/net/tipc/link.c b/net/tipc/link.c index 7059c94..a904ccd 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -87,7 +87,6 @@ struct tipc_stats { * @peer_bearer_id: bearer id used by link's peer endpoint * @bearer_id: local bearer id used by link * @tolerance: minimum link continuity loss needed to reset link [in ms] - * @keepalive_intv: link keepalive timer interval * @abort_limit: # of unacknowledged continuity probes needed to reset link * @state: current state of link FSM * @peer_caps: bitmap describing capabilities of peer node @@ -131,7 +130,6 @@ struct tipc_link { u32 peer_bearer_id; u32 bearer_id; u32 tolerance; - unsigned long keepalive_intv; u32 abort_limit; u32 state; u16 peer_caps; diff --git a/net/tipc/node.c b/net/tipc/node.c index c7985b2..d6a490f 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -378,14 +378,13 @@ static void tipc_node_calculate_timer(struct tipc_node *n, struct tipc_link *l) { unsigned long tol = tipc_link_tolerance(l); unsigned long intv = ((tol / 4) > 500) ? 500 : tol / 4; - unsigned long keepalive_intv = msecs_to_jiffies(intv); /* Link with lowest tolerance determines timer interval */ - if (keepalive_intv < n->keepalive_intv) - n->keepalive_intv = keepalive_intv; + if (intv < n->keepalive_intv) + n->keepalive_intv = intv; - /* Ensure link's abort limit corresponds to current interval */ - tipc_link_set_abort_limit(l, tol / jiffies_to_msecs(n->keepalive_intv)); + /* Ensure link's abort limit corresponds to current tolerance */ + tipc_link_set_abort_limit(l, tol / n->keepalive_intv); } static void tipc_node_delete(struct tipc_node *node) @@ -526,7 +525,7 @@ static void tipc_node_timeout(unsigned long data) if (rc & TIPC_LINK_DOWN_EVT) tipc_node_link_down(n, bearer_id, false); } - mod_timer(&n->timer, jiffies + n->keepalive_intv); + mod_timer(&n->timer, jiffies + msecs_to_jiffies(n->keepalive_intv)); } /** @@ -735,6 +734,7 @@ void tipc_node_check_dest(struct net *net, u32 onode, bool accept_addr = false; bool reset = true; char *if_name; + unsigned long intv; *dupl_addr = false; *respond = false; @@ -840,9 +840,11 @@ void tipc_node_check_dest(struct net *net, u32 onode, le->link = l; n->link_cnt++; tipc_node_calculate_timer(n, l); - if (n->link_cnt == 1) - if (!mod_timer(&n->timer, jiffies + n->keepalive_intv)) + if (n->link_cnt == 1) { + intv = jiffies + msecs_to_jiffies(n->keepalive_intv); + if (!mod_timer(&n->timer, intv)) tipc_node_get(n); + } } memcpy(&le->maddr, maddr, sizeof(*maddr)); exit: -- cgit v0.10.2 From 96c63fa7393d0a346acfe5a91e0c7d4c7782641b Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 8 Jun 2016 10:55:39 -0700 Subject: net: Add l3mdev rule Currently, VRFs require 1 oif and 1 iif rule per address family per VRF. As the number of VRF devices increases it brings scalability issues with the increasing rule list. All of the VRF rules have the same format with the exception of the specific table id to direct the lookup. Since the table id is available from the oif or iif in the loopup, the VRF rules can be consolidated to a single rule that pulls the table from the VRF device. This patch introduces a new rule attribute l3mdev. The l3mdev rule means the table id used for the lookup is pulled from the L3 master device (e.g., VRF) rather than being statically defined. With the l3mdev rule all of the basic VRF FIB rules are reduced to 1 l3mdev rule per address family (IPv4 and IPv6). If an admin wishes to insert higher priority rules for specific VRFs those rules will co-exist with the l3mdev rule. This capability means current VRF scripts will co-exist with this new simpler implementation. Currently, the rules list for both ipv4 and ipv6 look like this: $ ip ru ls 1000: from all oif vrf1 lookup 1001 1000: from all iif vrf1 lookup 1001 1000: from all oif vrf2 lookup 1002 1000: from all iif vrf2 lookup 1002 1000: from all oif vrf3 lookup 1003 1000: from all iif vrf3 lookup 1003 1000: from all oif vrf4 lookup 1004 1000: from all iif vrf4 lookup 1004 1000: from all oif vrf5 lookup 1005 1000: from all iif vrf5 lookup 1005 1000: from all oif vrf6 lookup 1006 1000: from all iif vrf6 lookup 1006 1000: from all oif vrf7 lookup 1007 1000: from all iif vrf7 lookup 1007 1000: from all oif vrf8 lookup 1008 1000: from all iif vrf8 lookup 1008 ... 32765: from all lookup local 32766: from all lookup main 32767: from all lookup default With the l3mdev rule the list is just the following regardless of the number of VRFs: $ ip ru ls 1000: from all lookup [l3mdev table] 32765: from all lookup local 32766: from all lookup main 32767: from all lookup default (Note: the above pretty print of the rule is based on an iproute2 prototype. Actual verbage may change) Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index 59160de..456e4a6 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -17,7 +17,8 @@ struct fib_rule { u32 flags; u32 table; u8 action; - /* 3 bytes hole, try to use */ + u8 l3mdev; + /* 2 bytes hole, try to use */ u32 target; __be64 tun_id; struct fib_rule __rcu *ctarget; @@ -36,6 +37,7 @@ struct fib_lookup_arg { void *lookup_ptr; void *result; struct fib_rule *rule; + u32 table; int flags; #define FIB_LOOKUP_NOREF 1 #define FIB_LOOKUP_IGNORE_LINKSTATE 2 @@ -89,7 +91,8 @@ struct fib_rules_ops { [FRA_TABLE] = { .type = NLA_U32 }, \ [FRA_SUPPRESS_PREFIXLEN] = { .type = NLA_U32 }, \ [FRA_SUPPRESS_IFGROUP] = { .type = NLA_U32 }, \ - [FRA_GOTO] = { .type = NLA_U32 } + [FRA_GOTO] = { .type = NLA_U32 }, \ + [FRA_L3MDEV] = { .type = NLA_U8 } static inline void fib_rule_get(struct fib_rule *rule) { @@ -102,6 +105,20 @@ static inline void fib_rule_put(struct fib_rule *rule) kfree_rcu(rule, rcu); } +#ifdef CONFIG_NET_L3_MASTER_DEV +static inline u32 fib_rule_get_table(struct fib_rule *rule, + struct fib_lookup_arg *arg) +{ + return rule->l3mdev ? arg->table : rule->table; +} +#else +static inline u32 fib_rule_get_table(struct fib_rule *rule, + struct fib_lookup_arg *arg) +{ + return rule->table; +} +#endif + static inline u32 frh_get_table(struct fib_rule_hdr *frh, struct nlattr **nla) { if (nla[FRA_TABLE]) @@ -117,4 +134,7 @@ int fib_rules_lookup(struct fib_rules_ops *, struct flowi *, int flags, struct fib_lookup_arg *); int fib_default_rule_add(struct fib_rules_ops *, u32 pref, u32 table, u32 flags); + +int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh); +int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr *nlh); #endif diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h index 374388d..34f33eb 100644 --- a/include/net/l3mdev.h +++ b/include/net/l3mdev.h @@ -11,6 +11,8 @@ #ifndef _NET_L3MDEV_H_ #define _NET_L3MDEV_H_ +#include + /** * struct l3mdev_ops - l3mdev operations * @@ -41,6 +43,9 @@ struct l3mdev_ops { #ifdef CONFIG_NET_L3_MASTER_DEV +int l3mdev_fib_rule_match(struct net *net, struct flowi *fl, + struct fib_lookup_arg *arg); + int l3mdev_master_ifindex_rcu(const struct net_device *dev); static inline int l3mdev_master_ifindex(struct net_device *dev) { @@ -236,6 +241,13 @@ struct sk_buff *l3mdev_ip6_rcv(struct sk_buff *skb) { return skb; } + +static inline +int l3mdev_fib_rule_match(struct net *net, struct flowi *fl, + struct fib_lookup_arg *arg) +{ + return 1; +} #endif #endif /* _NET_L3MDEV_H_ */ diff --git a/include/uapi/linux/fib_rules.h b/include/uapi/linux/fib_rules.h index 620c8a5..14404b3 100644 --- a/include/uapi/linux/fib_rules.h +++ b/include/uapi/linux/fib_rules.h @@ -50,6 +50,7 @@ enum { FRA_FWMASK, /* mask for netfilter mark */ FRA_OIFNAME, FRA_PAD, + FRA_L3MDEV, /* iif or oif is l3mdev goto its table */ __FRA_MAX }; diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 840aceb..98298b1 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -173,7 +173,8 @@ void fib_rules_unregister(struct fib_rules_ops *ops) EXPORT_SYMBOL_GPL(fib_rules_unregister); static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, - struct flowi *fl, int flags) + struct flowi *fl, int flags, + struct fib_lookup_arg *arg) { int ret = 0; @@ -189,6 +190,9 @@ static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, if (rule->tun_id && (rule->tun_id != fl->flowi_tun_key.tun_id)) goto out; + if (rule->l3mdev && !l3mdev_fib_rule_match(rule->fr_net, fl, arg)) + goto out; + ret = ops->match(rule, fl, flags); out: return (rule->flags & FIB_RULE_INVERT) ? !ret : ret; @@ -204,7 +208,7 @@ int fib_rules_lookup(struct fib_rules_ops *ops, struct flowi *fl, list_for_each_entry_rcu(rule, &ops->rules_list, list) { jumped: - if (!fib_rule_match(rule, ops, fl, flags)) + if (!fib_rule_match(rule, ops, fl, flags, arg)) continue; if (rule->action == FR_ACT_GOTO) { @@ -265,7 +269,7 @@ errout: return err; } -static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh) +int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh) { struct net *net = sock_net(skb->sk); struct fib_rule_hdr *frh = nlmsg_data(nlh); @@ -336,6 +340,14 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh) if (tb[FRA_TUN_ID]) rule->tun_id = nla_get_be64(tb[FRA_TUN_ID]); + if (tb[FRA_L3MDEV]) { +#ifdef CONFIG_NET_L3_MASTER_DEV + rule->l3mdev = nla_get_u8(tb[FRA_L3MDEV]); + if (rule->l3mdev != 1) +#endif + goto errout_free; + } + rule->action = frh->action; rule->flags = frh->flags; rule->table = frh_get_table(frh, tb); @@ -371,6 +383,9 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh) } else if (rule->action == FR_ACT_GOTO) goto errout_free; + if (rule->l3mdev && rule->table) + goto errout_free; + err = ops->configure(rule, skb, frh, tb); if (err < 0) goto errout_free; @@ -424,8 +439,9 @@ errout: rules_ops_put(ops); return err; } +EXPORT_SYMBOL_GPL(fib_nl_newrule); -static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh) +int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr *nlh) { struct net *net = sock_net(skb->sk); struct fib_rule_hdr *frh = nlmsg_data(nlh); @@ -483,6 +499,10 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh) (rule->tun_id != nla_get_be64(tb[FRA_TUN_ID]))) continue; + if (tb[FRA_L3MDEV] && + (rule->l3mdev != nla_get_u8(tb[FRA_L3MDEV]))) + continue; + if (!ops->compare(rule, frh, tb)) continue; @@ -536,6 +556,7 @@ errout: rules_ops_put(ops); return err; } +EXPORT_SYMBOL_GPL(fib_nl_delrule); static inline size_t fib_rule_nlmsg_size(struct fib_rules_ops *ops, struct fib_rule *rule) @@ -607,7 +628,9 @@ static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule, (rule->target && nla_put_u32(skb, FRA_GOTO, rule->target)) || (rule->tun_id && - nla_put_be64(skb, FRA_TUN_ID, rule->tun_id, FRA_PAD))) + nla_put_be64(skb, FRA_TUN_ID, rule->tun_id, FRA_PAD)) || + (rule->l3mdev && + nla_put_u8(skb, FRA_L3MDEV, rule->l3mdev))) goto nla_put_failure; if (rule->suppress_ifgroup != -1) { diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index f2bda9e..6e9ea69 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -76,6 +76,7 @@ static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp, { int err = -EAGAIN; struct fib_table *tbl; + u32 tb_id; switch (rule->action) { case FR_ACT_TO_TBL: @@ -94,7 +95,8 @@ static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp, rcu_read_lock(); - tbl = fib_get_table(rule->fr_net, rule->table); + tb_id = fib_rule_get_table(rule, arg); + tbl = fib_get_table(rule->fr_net, tb_id); if (tbl) err = fib_table_lookup(tbl, &flp->u.ip4, (struct fib_result *)arg->result, @@ -180,7 +182,7 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (err) goto errout; - if (rule->table == RT_TABLE_UNSPEC) { + if (rule->table == RT_TABLE_UNSPEC && !rule->l3mdev) { if (rule->action == FR_ACT_TO_TBL) { struct fib_table *table; diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index ed33abf..5857c1f 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -67,6 +67,7 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, struct net *net = rule->fr_net; pol_lookup_t lookup = arg->lookup_ptr; int err = 0; + u32 tb_id; switch (rule->action) { case FR_ACT_TO_TBL: @@ -86,7 +87,8 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, goto discard_pkt; } - table = fib6_get_table(net, rule->table); + tb_id = fib_rule_get_table(rule, arg); + table = fib6_get_table(net, tb_id); if (!table) { err = -EAGAIN; goto out; @@ -199,7 +201,7 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, struct net *net = sock_net(skb->sk); struct fib6_rule *rule6 = (struct fib6_rule *) rule; - if (rule->action == FR_ACT_TO_TBL) { + if (rule->action == FR_ACT_TO_TBL && !rule->l3mdev) { if (rule->table == RT6_TABLE_UNSPEC) goto errout; diff --git a/net/l3mdev/l3mdev.c b/net/l3mdev/l3mdev.c index 6651a78..7da9780 100644 --- a/net/l3mdev/l3mdev.c +++ b/net/l3mdev/l3mdev.c @@ -10,6 +10,7 @@ */ #include +#include #include /** @@ -160,3 +161,40 @@ int l3mdev_get_saddr(struct net *net, int ifindex, struct flowi4 *fl4) return rc; } EXPORT_SYMBOL_GPL(l3mdev_get_saddr); + +/** + * l3mdev_fib_rule_match - Determine if flowi references an + * L3 master device + * @net: network namespace for device index lookup + * @fl: flow struct + */ + +int l3mdev_fib_rule_match(struct net *net, struct flowi *fl, + struct fib_lookup_arg *arg) +{ + struct net_device *dev; + int rc = 0; + + rcu_read_lock(); + + dev = dev_get_by_index_rcu(net, fl->flowi_oif); + if (dev && netif_is_l3_master(dev) && + dev->l3mdev_ops->l3mdev_fib_table) { + arg->table = dev->l3mdev_ops->l3mdev_fib_table(dev); + rc = 1; + goto out; + } + + dev = dev_get_by_index_rcu(net, fl->flowi_iif); + if (dev && netif_is_l3_master(dev) && + dev->l3mdev_ops->l3mdev_fib_table) { + arg->table = dev->l3mdev_ops->l3mdev_fib_table(dev); + rc = 1; + goto out; + } + +out: + rcu_read_unlock(); + + return rc; +} -- cgit v0.10.2 From 1aa6c4f6b8cd84b8b36ebf43c6861ca87eab4da0 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 8 Jun 2016 10:55:40 -0700 Subject: net: vrf: Add l3mdev rules on first device create Add l3mdev rule per address family when the first VRF device is created. The rules are installed with a default preference of 1000. Users can replace the default rule as desired. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 1b214ea..e3e693b 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -35,6 +35,7 @@ #include #include #include +#include #define RT_FL_TOS(oldflp4) \ ((oldflp4)->flowi4_tos & (IPTOS_RT_MASK | RTO_ONLINK)) @@ -42,6 +43,9 @@ #define DRV_NAME "vrf" #define DRV_VERSION "1.0" +#define FIB_RULE_PREF 1000 /* default preference for FIB rules */ +static bool add_fib_rules = true; + struct net_vrf { struct rtable __rcu *rth; struct rtable __rcu *rth_local; @@ -897,6 +901,91 @@ static const struct ethtool_ops vrf_ethtool_ops = { .get_drvinfo = vrf_get_drvinfo, }; +static inline size_t vrf_fib_rule_nl_size(void) +{ + size_t sz; + + sz = NLMSG_ALIGN(sizeof(struct fib_rule_hdr)); + sz += nla_total_size(sizeof(u8)); /* FRA_L3MDEV */ + sz += nla_total_size(sizeof(u32)); /* FRA_PRIORITY */ + + return sz; +} + +static int vrf_fib_rule(const struct net_device *dev, __u8 family, bool add_it) +{ + struct fib_rule_hdr *frh; + struct nlmsghdr *nlh; + struct sk_buff *skb; + int err; + + skb = nlmsg_new(vrf_fib_rule_nl_size(), GFP_KERNEL); + if (!skb) + return -ENOMEM; + + nlh = nlmsg_put(skb, 0, 0, 0, sizeof(*frh), 0); + if (!nlh) + goto nla_put_failure; + + /* rule only needs to appear once */ + nlh->nlmsg_flags &= NLM_F_EXCL; + + frh = nlmsg_data(nlh); + memset(frh, 0, sizeof(*frh)); + frh->family = family; + frh->action = FR_ACT_TO_TBL; + + if (nla_put_u32(skb, FRA_L3MDEV, 1)) + goto nla_put_failure; + + if (nla_put_u32(skb, FRA_PRIORITY, FIB_RULE_PREF)) + goto nla_put_failure; + + nlmsg_end(skb, nlh); + + /* fib_nl_{new,del}rule handling looks for net from skb->sk */ + skb->sk = dev_net(dev)->rtnl; + if (add_it) { + err = fib_nl_newrule(skb, nlh); + if (err == -EEXIST) + err = 0; + } else { + err = fib_nl_delrule(skb, nlh); + if (err == -ENOENT) + err = 0; + } + nlmsg_free(skb); + + return err; + +nla_put_failure: + nlmsg_free(skb); + + return -EMSGSIZE; +} + +static int vrf_add_fib_rules(const struct net_device *dev) +{ + int err; + + err = vrf_fib_rule(dev, AF_INET, true); + if (err < 0) + goto out_err; + + err = vrf_fib_rule(dev, AF_INET6, true); + if (err < 0) + goto ipv6_err; + + return 0; + +ipv6_err: + vrf_fib_rule(dev, AF_INET, false); + +out_err: + netdev_err(dev, "Failed to add FIB rules.\n"); + return err; +} + static void vrf_setup(struct net_device *dev) { ether_setup(dev); @@ -937,6 +1026,7 @@ static int vrf_newlink(struct net *src_net, struct net_device *dev, struct nlattr *tb[], struct nlattr *data[]) { struct net_vrf *vrf = netdev_priv(dev); + int err; if (!data || !data[IFLA_VRF_TABLE]) return -EINVAL; @@ -945,7 +1035,21 @@ static int vrf_newlink(struct net *src_net, struct net_device *dev, dev->priv_flags |= IFF_L3MDEV_MASTER; - return register_netdevice(dev); + err = register_netdevice(dev); + if (err) + goto out; + + if (add_fib_rules) { + err = vrf_add_fib_rules(dev); + if (err) { + unregister_netdevice(dev); + goto out; + } + add_fib_rules = false; + } + +out: + return err; } static size_t vrf_nl_getsize(const struct net_device *dev) -- cgit v0.10.2 From 76e48f9fbe3b0d1279868eef0543725577525e97 Mon Sep 17 00:00:00 2001 From: Shweta Choudaha Date: Wed, 8 Jun 2016 20:15:43 +0100 Subject: ip6gre: Allow live link address change The ip6 GRE tap device should not be forced to down state to change the mac address and should allow live address change for tap device similar to ipv4 gre. Signed-off-by: Shweta Choudaha Signed-off-by: David S. Miller diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index f4ac284..fdc9de2 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -1256,6 +1256,8 @@ static int ip6gre_tap_init(struct net_device *dev) if (ret) return ret; + dev->priv_flags |= IFF_LIVE_ADDR_CHANGE; + tunnel = netdev_priv(dev); ip6gre_tnl_link_config(tunnel, 1); @@ -1289,6 +1291,7 @@ static void ip6gre_tap_setup(struct net_device *dev) dev->features |= NETIF_F_NETNS_LOCAL; dev->priv_flags &= ~IFF_TX_SKB_SHARING; + dev->priv_flags |= IFF_LIVE_ADDR_CHANGE; } static bool ip6gre_netlink_encap_parms(struct nlattr *data[], -- cgit v0.10.2 From c3498d34dd369115a06e5bb862b90b06fde3d114 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 9 Jun 2016 00:27:39 +0200 Subject: cbq: remove TCA_CBQ_OVL_STRATEGY support since initial revision of cbq in 2004 iproute 2 has never implemented support for TCA_CBQ_OVL_STRATEGY, which is what needs to be set to activate the class->drop() call (TC_CBQ_OVL_DROP strategy must be set by userspace value must be set by userspace). David Miller says: It seems really safe to kill this thing off, flag an error if someone tries to set the attribute, and therefore kill off all of the non-default cbq_ovl_*() functions. A followup commit can then remove all .drop qdisc methods since this removed the only caller. Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 1b8128f..fdca45e 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -80,7 +80,6 @@ struct cbq_class { unsigned char priority; /* class priority */ unsigned char priority2; /* priority to be used after overlimit */ unsigned char ewma_log; /* time constant for idle time calculation */ - unsigned char ovl_strategy; #ifdef CONFIG_NET_CLS_ACT unsigned char police; #endif @@ -94,10 +93,6 @@ struct cbq_class { u32 avpkt; struct qdisc_rate_table *R_tab; - /* Overlimit strategy parameters */ - void (*overlimit)(struct cbq_class *cl); - psched_tdiff_t penalty; - /* General scheduler (WRR) parameters */ long allot; long quantum; /* Allotment per WRR round */ @@ -402,11 +397,8 @@ cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch) return ret; } -/* Overlimit actions */ - -/* TC_CBQ_OVL_CLASSIC: (default) penalize leaf class by adding offtime */ - -static void cbq_ovl_classic(struct cbq_class *cl) +/* Overlimit action: penalize leaf class by adding offtime */ +static void cbq_overlimit(struct cbq_class *cl) { struct cbq_sched_data *q = qdisc_priv(cl->qdisc); psched_tdiff_t delay = cl->undertime - q->now; @@ -456,99 +448,6 @@ static void cbq_ovl_classic(struct cbq_class *cl) } } -/* TC_CBQ_OVL_RCLASSIC: penalize by offtime classes in hierarchy, when - * they go overlimit - */ - -static void cbq_ovl_rclassic(struct cbq_class *cl) -{ - struct cbq_sched_data *q = qdisc_priv(cl->qdisc); - struct cbq_class *this = cl; - - do { - if (cl->level > q->toplevel) { - cl = NULL; - break; - } - } while ((cl = cl->borrow) != NULL); - - if (cl == NULL) - cl = this; - cbq_ovl_classic(cl); -} - -/* TC_CBQ_OVL_DELAY: delay until it will go to underlimit */ - -static void cbq_ovl_delay(struct cbq_class *cl) -{ - struct cbq_sched_data *q = qdisc_priv(cl->qdisc); - psched_tdiff_t delay = cl->undertime - q->now; - - if (test_bit(__QDISC_STATE_DEACTIVATED, - &qdisc_root_sleeping(cl->qdisc)->state)) - return; - - if (!cl->delayed) { - psched_time_t sched = q->now; - ktime_t expires; - - delay += cl->offtime; - if (cl->avgidle < 0) - delay -= (-cl->avgidle) - ((-cl->avgidle) >> cl->ewma_log); - if (cl->avgidle < cl->minidle) - cl->avgidle = cl->minidle; - cl->undertime = q->now + delay; - - if (delay > 0) { - sched += delay + cl->penalty; - cl->penalized = sched; - cl->cpriority = TC_CBQ_MAXPRIO; - q->pmask |= (1<delay_timer) && - ktime_to_ns(ktime_sub( - hrtimer_get_expires(&q->delay_timer), - expires)) > 0) - hrtimer_set_expires(&q->delay_timer, expires); - hrtimer_restart(&q->delay_timer); - cl->delayed = 1; - cl->xstats.overactions++; - return; - } - delay = 1; - } - if (q->wd_expires == 0 || q->wd_expires > delay) - q->wd_expires = delay; -} - -/* TC_CBQ_OVL_LOWPRIO: penalize class by lowering its priority band */ - -static void cbq_ovl_lowprio(struct cbq_class *cl) -{ - struct cbq_sched_data *q = qdisc_priv(cl->qdisc); - - cl->penalized = q->now + cl->penalty; - - if (cl->cpriority != cl->priority2) { - cl->cpriority = cl->priority2; - q->pmask |= (1<cpriority); - cl->xstats.overactions++; - } - cbq_ovl_classic(cl); -} - -/* TC_CBQ_OVL_DROP: penalize class by dropping */ - -static void cbq_ovl_drop(struct cbq_class *cl) -{ - if (cl->q->ops->drop) - if (cl->q->ops->drop(cl->q)) - cl->qdisc->q.qlen--; - cl->xstats.overactions++; - cbq_ovl_classic(cl); -} - static psched_tdiff_t cbq_undelay_prio(struct cbq_sched_data *q, int prio, psched_time_t now) { @@ -807,7 +706,7 @@ cbq_under_limit(struct cbq_class *cl) cl = cl->borrow; if (!cl) { this_cl->qstats.overlimits++; - this_cl->overlimit(this_cl); + cbq_overlimit(this_cl); return NULL; } if (cl->level > q->toplevel) @@ -1280,35 +1179,6 @@ static int cbq_set_wrr(struct cbq_class *cl, struct tc_cbq_wrropt *wrr) return 0; } -static int cbq_set_overlimit(struct cbq_class *cl, struct tc_cbq_ovl *ovl) -{ - switch (ovl->strategy) { - case TC_CBQ_OVL_CLASSIC: - cl->overlimit = cbq_ovl_classic; - break; - case TC_CBQ_OVL_DELAY: - cl->overlimit = cbq_ovl_delay; - break; - case TC_CBQ_OVL_LOWPRIO: - if (ovl->priority2 - 1 >= TC_CBQ_MAXPRIO || - ovl->priority2 - 1 <= cl->priority) - return -EINVAL; - cl->priority2 = ovl->priority2 - 1; - cl->overlimit = cbq_ovl_lowprio; - break; - case TC_CBQ_OVL_DROP: - cl->overlimit = cbq_ovl_drop; - break; - case TC_CBQ_OVL_RCLASSIC: - cl->overlimit = cbq_ovl_rclassic; - break; - default: - return -EINVAL; - } - cl->penalty = ovl->penalty; - return 0; -} - #ifdef CONFIG_NET_CLS_ACT static int cbq_set_police(struct cbq_class *cl, struct tc_cbq_police *p) { @@ -1375,8 +1245,6 @@ static int cbq_init(struct Qdisc *sch, struct nlattr *opt) q->link.priority = TC_CBQ_MAXPRIO - 1; q->link.priority2 = TC_CBQ_MAXPRIO - 1; q->link.cpriority = TC_CBQ_MAXPRIO - 1; - q->link.ovl_strategy = TC_CBQ_OVL_CLASSIC; - q->link.overlimit = cbq_ovl_classic; q->link.allot = psched_mtu(qdisc_dev(sch)); q->link.quantum = q->link.allot; q->link.weight = q->link.R_tab->rate.rate; @@ -1463,24 +1331,6 @@ nla_put_failure: return -1; } -static int cbq_dump_ovl(struct sk_buff *skb, struct cbq_class *cl) -{ - unsigned char *b = skb_tail_pointer(skb); - struct tc_cbq_ovl opt; - - opt.strategy = cl->ovl_strategy; - opt.priority2 = cl->priority2 + 1; - opt.pad = 0; - opt.penalty = cl->penalty; - if (nla_put(skb, TCA_CBQ_OVL_STRATEGY, sizeof(opt), &opt)) - goto nla_put_failure; - return skb->len; - -nla_put_failure: - nlmsg_trim(skb, b); - return -1; -} - static int cbq_dump_fopt(struct sk_buff *skb, struct cbq_class *cl) { unsigned char *b = skb_tail_pointer(skb); @@ -1526,7 +1376,6 @@ static int cbq_dump_attr(struct sk_buff *skb, struct cbq_class *cl) if (cbq_dump_lss(skb, cl) < 0 || cbq_dump_rate(skb, cl) < 0 || cbq_dump_wrr(skb, cl) < 0 || - cbq_dump_ovl(skb, cl) < 0 || #ifdef CONFIG_NET_CLS_ACT cbq_dump_police(skb, cl) < 0 || #endif @@ -1736,6 +1585,9 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t if (err < 0) return err; + if (tb[TCA_CBQ_OVL_STRATEGY]) + return -EOPNOTSUPP; + if (cl) { /* Check parent */ if (parentid) { @@ -1784,9 +1636,6 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t cbq_set_wrr(cl, nla_data(tb[TCA_CBQ_WRROPT])); } - if (tb[TCA_CBQ_OVL_STRATEGY]) - cbq_set_overlimit(cl, nla_data(tb[TCA_CBQ_OVL_STRATEGY])); - #ifdef CONFIG_NET_CLS_ACT if (tb[TCA_CBQ_POLICE]) cbq_set_police(cl, nla_data(tb[TCA_CBQ_POLICE])); @@ -1887,9 +1736,6 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t cl->maxidle = q->link.maxidle; if (cl->avpkt == 0) cl->avpkt = q->link.avpkt; - cl->overlimit = cbq_ovl_classic; - if (tb[TCA_CBQ_OVL_STRATEGY]) - cbq_set_overlimit(cl, nla_data(tb[TCA_CBQ_OVL_STRATEGY])); #ifdef CONFIG_NET_CLS_ACT if (tb[TCA_CBQ_POLICE]) cbq_set_police(cl, nla_data(tb[TCA_CBQ_POLICE])); -- cgit v0.10.2 From dd47c1fa776cda48531b651c88341e951140b0a7 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 9 Jun 2016 00:27:40 +0200 Subject: cbq: remove TCA_CBQ_POLICE support iproute2 doesn't implement any cbq option that results in this attribute being sent to kernel. To make use of it, user would have to - patch iproute2 - add a class - attach a qdisc to the class (default pfifo doesn't work as q->handle is 0 and cbq_set_police() is a no-op in this case) - re-'add' the same class (tc class change ...) again - user must also specifiy a defmap (e.g. 'split 1:0 defmap 3f'), since this 'police' feature relies on its presence - the added qdisc must be one of bfifo, pfifo or netem If all of these conditions are met and _some_ leaf qdiscs, namely p/bfifo, netem, plug or tbf would drop a packet, kernel calls back into cbq, which will attempt to re-queue the skb into a different class as indicated by the parents' defmap entry for TC_PRIO_BESTEFFORT. [ i.e. we behave as if tc_classify returned TC_ACT_RECLASSIFY ]. This feature, which isn't documented or implemented in iproute2, and isn't implemented consistently (most qdiscs like sfq, codel, etc drop right away instead of attempting this reclassification) is the sole reason for the reshape_fail and __parent member in Qdisc struct. So remove TCA_CBQ_POLICE support from the kernel, reject it via EOPNOTSUPP so userspace knows we don't support it, and then remove no-longer needed infrastructure in followup commit. Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index c4f5749..c069ac1 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -68,10 +68,6 @@ struct Qdisc { void *u32_node; - /* This field is deprecated, but it is still used by CBQ - * and it will live until better solution will be invented. - */ - struct Qdisc *__parent; struct netdev_queue *dev_queue; struct gnet_stats_rate_est64 rate_est; diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index fdca45e..7f4d6c5 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -80,9 +80,6 @@ struct cbq_class { unsigned char priority; /* class priority */ unsigned char priority2; /* priority to be used after overlimit */ unsigned char ewma_log; /* time constant for idle time calculation */ -#ifdef CONFIG_NET_CLS_ACT - unsigned char police; -#endif u32 defmap; @@ -377,9 +374,6 @@ cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch) return ret; } -#ifdef CONFIG_NET_CLS_ACT - cl->q->__parent = sch; -#endif ret = qdisc_enqueue(skb, cl->q); if (ret == NET_XMIT_SUCCESS) { sch->q.qlen++; @@ -524,40 +518,6 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer) return HRTIMER_NORESTART; } -#ifdef CONFIG_NET_CLS_ACT -static int cbq_reshape_fail(struct sk_buff *skb, struct Qdisc *child) -{ - struct Qdisc *sch = child->__parent; - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *cl = q->rx_class; - - q->rx_class = NULL; - - if (cl && (cl = cbq_reclassify(skb, cl)) != NULL) { - int ret; - - cbq_mark_toplevel(q, cl); - - q->rx_class = cl; - cl->q->__parent = sch; - - ret = qdisc_enqueue(skb, cl->q); - if (ret == NET_XMIT_SUCCESS) { - sch->q.qlen++; - if (!cl->next_alive) - cbq_activate_class(cl); - return 0; - } - if (net_xmit_drop_count(ret)) - qdisc_qstats_drop(sch); - return 0; - } - - qdisc_qstats_drop(sch); - return -1; -} -#endif - /* * It is mission critical procedure. * @@ -1179,21 +1139,6 @@ static int cbq_set_wrr(struct cbq_class *cl, struct tc_cbq_wrropt *wrr) return 0; } -#ifdef CONFIG_NET_CLS_ACT -static int cbq_set_police(struct cbq_class *cl, struct tc_cbq_police *p) -{ - cl->police = p->police; - - if (cl->q->handle) { - if (p->police == TC_POLICE_RECLASSIFY) - cl->q->reshape_fail = cbq_reshape_fail; - else - cl->q->reshape_fail = NULL; - } - return 0; -} -#endif - static int cbq_set_fopt(struct cbq_class *cl, struct tc_cbq_fopt *fopt) { cbq_change_defmap(cl, fopt->split, fopt->defmap, fopt->defchange); @@ -1350,35 +1295,11 @@ nla_put_failure: return -1; } -#ifdef CONFIG_NET_CLS_ACT -static int cbq_dump_police(struct sk_buff *skb, struct cbq_class *cl) -{ - unsigned char *b = skb_tail_pointer(skb); - struct tc_cbq_police opt; - - if (cl->police) { - opt.police = cl->police; - opt.__res1 = 0; - opt.__res2 = 0; - if (nla_put(skb, TCA_CBQ_POLICE, sizeof(opt), &opt)) - goto nla_put_failure; - } - return skb->len; - -nla_put_failure: - nlmsg_trim(skb, b); - return -1; -} -#endif - static int cbq_dump_attr(struct sk_buff *skb, struct cbq_class *cl) { if (cbq_dump_lss(skb, cl) < 0 || cbq_dump_rate(skb, cl) < 0 || cbq_dump_wrr(skb, cl) < 0 || -#ifdef CONFIG_NET_CLS_ACT - cbq_dump_police(skb, cl) < 0 || -#endif cbq_dump_fopt(skb, cl) < 0) return -1; return 0; @@ -1468,11 +1389,6 @@ static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, &pfifo_qdisc_ops, cl->common.classid); if (new == NULL) return -ENOBUFS; - } else { -#ifdef CONFIG_NET_CLS_ACT - if (cl->police == TC_POLICE_RECLASSIFY) - new->reshape_fail = cbq_reshape_fail; -#endif } *old = qdisc_replace(sch, new, &cl->q); @@ -1585,7 +1501,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t if (err < 0) return err; - if (tb[TCA_CBQ_OVL_STRATEGY]) + if (tb[TCA_CBQ_OVL_STRATEGY] || tb[TCA_CBQ_POLICE]) return -EOPNOTSUPP; if (cl) { @@ -1636,11 +1552,6 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t cbq_set_wrr(cl, nla_data(tb[TCA_CBQ_WRROPT])); } -#ifdef CONFIG_NET_CLS_ACT - if (tb[TCA_CBQ_POLICE]) - cbq_set_police(cl, nla_data(tb[TCA_CBQ_POLICE])); -#endif - if (tb[TCA_CBQ_FOPT]) cbq_set_fopt(cl, nla_data(tb[TCA_CBQ_FOPT])); @@ -1736,10 +1647,6 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t cl->maxidle = q->link.maxidle; if (cl->avpkt == 0) cl->avpkt = q->link.avpkt; -#ifdef CONFIG_NET_CLS_ACT - if (tb[TCA_CBQ_POLICE]) - cbq_set_police(cl, nla_data(tb[TCA_CBQ_POLICE])); -#endif if (tb[TCA_CBQ_FOPT]) cbq_set_fopt(cl, nla_data(tb[TCA_CBQ_FOPT])); sch_tree_unlock(sch); -- cgit v0.10.2 From c3a173d7dba2d7c74dd4ab871b8f22bf56ac10b2 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 9 Jun 2016 00:27:41 +0200 Subject: sched: remove qdisc_rehape_fail After the removal of TCA_CBQ_POLICE in cbq scheduler qdisc->reshape_fail is always NULL, i.e. qdisc_rehape_fail is now the same as qdisc_drop. Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index c069ac1..a9aec63 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -63,9 +63,6 @@ struct Qdisc { struct list_head list; u32 handle; u32 parent; - int (*reshape_fail)(struct sk_buff *skb, - struct Qdisc *q); - void *u32_node; struct netdev_queue *dev_queue; @@ -771,22 +768,6 @@ static inline int qdisc_drop(struct sk_buff *skb, struct Qdisc *sch) return NET_XMIT_DROP; } -static inline int qdisc_reshape_fail(struct sk_buff *skb, struct Qdisc *sch) -{ - qdisc_qstats_drop(sch); - -#ifdef CONFIG_NET_CLS_ACT - if (sch->reshape_fail == NULL || sch->reshape_fail(skb, sch)) - goto drop; - - return NET_XMIT_SUCCESS; - -drop: -#endif - kfree_skb(skb); - return NET_XMIT_DROP; -} - /* Length to Time (L2T) lookup in a qdisc_rate_table, to determine how long it will take to send a packet given its size. */ diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c index 2177eac..7857f74 100644 --- a/net/sched/sch_fifo.c +++ b/net/sched/sch_fifo.c @@ -24,7 +24,7 @@ static int bfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (likely(sch->qstats.backlog + qdisc_pkt_len(skb) <= sch->limit)) return qdisc_enqueue_tail(skb, sch); - return qdisc_reshape_fail(skb, sch); + return qdisc_drop(skb, sch); } static int pfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch) @@ -32,7 +32,7 @@ static int pfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (likely(skb_queue_len(&sch->q) < sch->limit)) return qdisc_enqueue_tail(skb, sch); - return qdisc_reshape_fail(skb, sch); + return qdisc_drop(skb, sch); } static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc *sch) diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 205bed0..31984c7 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -407,7 +407,7 @@ static struct sk_buff *netem_segment(struct sk_buff *skb, struct Qdisc *sch) segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); if (IS_ERR_OR_NULL(segs)) { - qdisc_reshape_fail(skb, sch); + qdisc_drop(skb, sch); return NULL; } consume_skb(skb); @@ -499,7 +499,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) } if (unlikely(skb_queue_len(&sch->q) >= sch->limit)) - return qdisc_reshape_fail(skb, sch); + return qdisc_drop(skb, sch); qdisc_qstats_backlog_inc(sch, skb); diff --git a/net/sched/sch_plug.c b/net/sched/sch_plug.c index 5abfe44..ff0d968 100644 --- a/net/sched/sch_plug.c +++ b/net/sched/sch_plug.c @@ -96,7 +96,7 @@ static int plug_enqueue(struct sk_buff *skb, struct Qdisc *sch) return qdisc_enqueue_tail(skb, sch); } - return qdisc_reshape_fail(skb, sch); + return qdisc_drop(skb, sch); } static struct sk_buff *plug_dequeue(struct Qdisc *sch) diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index 83b90b5..801148c 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -166,7 +166,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch) segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); if (IS_ERR_OR_NULL(segs)) - return qdisc_reshape_fail(skb, sch); + return qdisc_drop(skb, sch); nb = 0; while (segs) { @@ -198,7 +198,7 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (qdisc_pkt_len(skb) > q->max_size) { if (skb_is_gso(skb) && skb_gso_mac_seglen(skb) <= q->max_size) return tbf_segment(skb, sch); - return qdisc_reshape_fail(skb, sch); + return qdisc_drop(skb, sch); } ret = qdisc_enqueue(skb, q->qdisc); if (ret != NET_XMIT_SUCCESS) { -- cgit v0.10.2 From a09ceb0e08140a1eec05b49b4c232d3481339cb0 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 9 Jun 2016 00:27:42 +0200 Subject: sched: remove qdisc->drop after removal of TCA_CBQ_OVL_STRATEGY from cbq scheduler, there are no more callers of ->drop() outside of other ->drop functions, i.e. nothing calls them. Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index a9aec63..4dedb7f 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -174,7 +174,6 @@ struct Qdisc_ops { int (*enqueue)(struct sk_buff *, struct Qdisc *); struct sk_buff * (*dequeue)(struct Qdisc *); struct sk_buff * (*peek)(struct Qdisc *); - unsigned int (*drop)(struct Qdisc *); int (*init)(struct Qdisc *, struct nlattr *arg); void (*reset)(struct Qdisc *); @@ -658,22 +657,6 @@ static inline unsigned int qdisc_queue_drop_head(struct Qdisc *sch) return __qdisc_queue_drop_head(sch, &sch->q); } -static inline struct sk_buff *__qdisc_dequeue_tail(struct Qdisc *sch, - struct sk_buff_head *list) -{ - struct sk_buff *skb = __skb_dequeue_tail(list); - - if (likely(skb != NULL)) - qdisc_qstats_backlog_dec(sch, skb); - - return skb; -} - -static inline struct sk_buff *qdisc_dequeue_tail(struct Qdisc *sch) -{ - return __qdisc_dequeue_tail(sch, &sch->q); -} - static inline struct sk_buff *qdisc_peek_head(struct Qdisc *sch) { return skb_peek(&sch->q); @@ -741,25 +724,6 @@ static inline struct Qdisc *qdisc_replace(struct Qdisc *sch, struct Qdisc *new, return old; } -static inline unsigned int __qdisc_queue_drop(struct Qdisc *sch, - struct sk_buff_head *list) -{ - struct sk_buff *skb = __qdisc_dequeue_tail(sch, list); - - if (likely(skb != NULL)) { - unsigned int len = qdisc_pkt_len(skb); - kfree_skb(skb); - return len; - } - - return 0; -} - -static inline unsigned int qdisc_queue_drop(struct Qdisc *sch) -{ - return __qdisc_queue_drop(sch, &sch->q); -} - static inline int qdisc_drop(struct sk_buff *skb, struct Qdisc *sch) { kfree_skb(skb); diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index 34f8f79..7e6c12d 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -519,20 +519,6 @@ static struct sk_buff *atm_tc_peek(struct Qdisc *sch) return p->link.q->ops->peek(p->link.q); } -static unsigned int atm_tc_drop(struct Qdisc *sch) -{ - struct atm_qdisc_data *p = qdisc_priv(sch); - struct atm_flow_data *flow; - unsigned int len; - - pr_debug("atm_tc_drop(sch %p,[qdisc %p])\n", sch, p); - list_for_each_entry(flow, &p->flows, list) { - if (flow->q->ops->drop && (len = flow->q->ops->drop(flow->q))) - return len; - } - return 0; -} - static int atm_tc_init(struct Qdisc *sch, struct nlattr *opt) { struct atm_qdisc_data *p = qdisc_priv(sch); @@ -672,7 +658,6 @@ static struct Qdisc_ops atm_qdisc_ops __read_mostly = { .enqueue = atm_tc_enqueue, .dequeue = atm_tc_dequeue, .peek = atm_tc_peek, - .drop = atm_tc_drop, .init = atm_tc_init, .reset = atm_tc_reset, .destroy = atm_tc_destroy, diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 7f4d6c5..f2af31b 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -1025,31 +1025,6 @@ static void cbq_link_class(struct cbq_class *this) } } -static unsigned int cbq_drop(struct Qdisc *sch) -{ - struct cbq_sched_data *q = qdisc_priv(sch); - struct cbq_class *cl, *cl_head; - int prio; - unsigned int len; - - for (prio = TC_CBQ_MAXPRIO; prio >= 0; prio--) { - cl_head = q->active[prio]; - if (!cl_head) - continue; - - cl = cl_head; - do { - if (cl->q->ops->drop && (len = cl->q->ops->drop(cl->q))) { - sch->q.qlen--; - if (!cl->q->q.qlen) - cbq_deactivate_class(cl); - return len; - } - } while ((cl = cl->next_alive) != cl_head); - } - return 0; -} - static void cbq_reset(struct Qdisc *sch) { @@ -1791,7 +1766,6 @@ static struct Qdisc_ops cbq_qdisc_ops __read_mostly = { .enqueue = cbq_enqueue, .dequeue = cbq_dequeue, .peek = qdisc_peek_dequeued, - .drop = cbq_drop, .init = cbq_init, .reset = cbq_reset, .destroy = cbq_destroy, diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c index 0a08c86..04e0b05 100644 --- a/net/sched/sch_choke.c +++ b/net/sched/sch_choke.c @@ -365,22 +365,6 @@ static struct sk_buff *choke_dequeue(struct Qdisc *sch) return skb; } -static unsigned int choke_drop(struct Qdisc *sch) -{ - struct choke_sched_data *q = qdisc_priv(sch); - unsigned int len; - - len = qdisc_queue_drop(sch); - if (len > 0) - q->stats.other++; - else { - if (!red_is_idling(&q->vars)) - red_start_of_idle_period(&q->vars); - } - - return len; -} - static void choke_reset(struct Qdisc *sch) { struct choke_sched_data *q = qdisc_priv(sch); @@ -569,7 +553,6 @@ static struct Qdisc_ops choke_qdisc_ops __read_mostly = { .enqueue = choke_enqueue, .dequeue = choke_dequeue, .peek = choke_peek_head, - .drop = choke_drop, .init = choke_init, .destroy = choke_destroy, .reset = choke_reset, diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c index 1b7e1a2..2bec94c 100644 --- a/net/sched/sch_drr.c +++ b/net/sched/sch_drr.c @@ -421,26 +421,6 @@ out: return NULL; } -static unsigned int drr_drop(struct Qdisc *sch) -{ - struct drr_sched *q = qdisc_priv(sch); - struct drr_class *cl; - unsigned int len; - - list_for_each_entry(cl, &q->active, alist) { - if (cl->qdisc->ops->drop) { - len = cl->qdisc->ops->drop(cl->qdisc); - if (len > 0) { - sch->q.qlen--; - if (cl->qdisc->q.qlen == 0) - list_del(&cl->alist); - return len; - } - } - } - return 0; -} - static int drr_init_qdisc(struct Qdisc *sch, struct nlattr *opt) { struct drr_sched *q = qdisc_priv(sch); @@ -509,7 +489,6 @@ static struct Qdisc_ops drr_qdisc_ops __read_mostly = { .enqueue = drr_enqueue, .dequeue = drr_dequeue, .peek = qdisc_peek_dequeued, - .drop = drr_drop, .init = drr_init_qdisc, .reset = drr_reset_qdisc, .destroy = drr_destroy_qdisc, diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c index 34b4dda..b9ba5f6 100644 --- a/net/sched/sch_dsmark.c +++ b/net/sched/sch_dsmark.c @@ -320,23 +320,6 @@ static struct sk_buff *dsmark_peek(struct Qdisc *sch) return p->q->ops->peek(p->q); } -static unsigned int dsmark_drop(struct Qdisc *sch) -{ - struct dsmark_qdisc_data *p = qdisc_priv(sch); - unsigned int len; - - pr_debug("%s(sch %p,[qdisc %p])\n", __func__, sch, p); - - if (p->q->ops->drop == NULL) - return 0; - - len = p->q->ops->drop(p->q); - if (len) - sch->q.qlen--; - - return len; -} - static int dsmark_init(struct Qdisc *sch, struct nlattr *opt) { struct dsmark_qdisc_data *p = qdisc_priv(sch); @@ -489,7 +472,6 @@ static struct Qdisc_ops dsmark_qdisc_ops __read_mostly = { .enqueue = dsmark_enqueue, .dequeue = dsmark_dequeue, .peek = dsmark_peek, - .drop = dsmark_drop, .init = dsmark_init, .reset = dsmark_reset, .destroy = dsmark_destroy, diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c index 7857f74..dea70e3 100644 --- a/net/sched/sch_fifo.c +++ b/net/sched/sch_fifo.c @@ -99,7 +99,6 @@ struct Qdisc_ops pfifo_qdisc_ops __read_mostly = { .enqueue = pfifo_enqueue, .dequeue = qdisc_dequeue_head, .peek = qdisc_peek_head, - .drop = qdisc_queue_drop, .init = fifo_init, .reset = qdisc_reset_queue, .change = fifo_init, @@ -114,7 +113,6 @@ struct Qdisc_ops bfifo_qdisc_ops __read_mostly = { .enqueue = bfifo_enqueue, .dequeue = qdisc_dequeue_head, .peek = qdisc_peek_head, - .drop = qdisc_queue_drop, .init = fifo_init, .reset = qdisc_reset_queue, .change = fifo_init, @@ -129,7 +127,6 @@ struct Qdisc_ops pfifo_head_drop_qdisc_ops __read_mostly = { .enqueue = pfifo_tail_enqueue, .dequeue = qdisc_dequeue_head, .peek = qdisc_peek_head, - .drop = qdisc_queue_drop_head, .init = fifo_init, .reset = qdisc_reset_queue, .change = fifo_init, diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index 1daa542..176a7e2 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -184,15 +184,6 @@ static unsigned int fq_codel_drop(struct Qdisc *sch, unsigned int max_packets) return idx; } -static unsigned int fq_codel_qdisc_drop(struct Qdisc *sch) -{ - unsigned int prev_backlog; - - prev_backlog = sch->qstats.backlog; - fq_codel_drop(sch, 1U); - return prev_backlog - sch->qstats.backlog; -} - static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch) { struct fq_codel_sched_data *q = qdisc_priv(sch); @@ -704,7 +695,6 @@ static struct Qdisc_ops fq_codel_qdisc_ops __read_mostly = { .enqueue = fq_codel_enqueue, .dequeue = fq_codel_dequeue, .peek = qdisc_peek_dequeued, - .drop = fq_codel_qdisc_drop, .init = fq_codel_init, .reset = fq_codel_reset, .destroy = fq_codel_destroy, diff --git a/net/sched/sch_gred.c b/net/sched/sch_gred.c index 8010510..b5fb63c7 100644 --- a/net/sched/sch_gred.c +++ b/net/sched/sch_gred.c @@ -276,40 +276,6 @@ static struct sk_buff *gred_dequeue(struct Qdisc *sch) return NULL; } -static unsigned int gred_drop(struct Qdisc *sch) -{ - struct sk_buff *skb; - struct gred_sched *t = qdisc_priv(sch); - - skb = qdisc_dequeue_tail(sch); - if (skb) { - unsigned int len = qdisc_pkt_len(skb); - struct gred_sched_data *q; - u16 dp = tc_index_to_dp(skb); - - if (dp >= t->DPs || (q = t->tab[dp]) == NULL) { - net_warn_ratelimited("GRED: Unable to relocate VQ 0x%x while dropping, screwing up backlog\n", - tc_index_to_dp(skb)); - } else { - q->backlog -= len; - q->stats.other++; - - if (gred_wred_mode(t)) { - if (!sch->qstats.backlog) - red_start_of_idle_period(&t->wred_set); - } else { - if (!q->backlog) - red_start_of_idle_period(&q->vars); - } - } - - qdisc_drop(skb, sch); - return len; - } - - return 0; -} - static void gred_reset(struct Qdisc *sch) { int i; @@ -623,7 +589,6 @@ static struct Qdisc_ops gred_qdisc_ops __read_mostly = { .enqueue = gred_enqueue, .dequeue = gred_dequeue, .peek = qdisc_peek_head, - .drop = gred_drop, .init = gred_init, .reset = gred_reset, .destroy = gred_destroy, diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index 74813dd..155e3ce 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1677,31 +1677,6 @@ hfsc_dequeue(struct Qdisc *sch) return skb; } -static unsigned int -hfsc_drop(struct Qdisc *sch) -{ - struct hfsc_sched *q = qdisc_priv(sch); - struct hfsc_class *cl; - unsigned int len; - - list_for_each_entry(cl, &q->droplist, dlist) { - if (cl->qdisc->ops->drop != NULL && - (len = cl->qdisc->ops->drop(cl->qdisc)) > 0) { - if (cl->qdisc->q.qlen == 0) { - update_vf(cl, 0, 0); - set_passive(cl); - } else { - list_move_tail(&cl->dlist, &q->droplist); - } - cl->qstats.drops++; - qdisc_qstats_drop(sch); - sch->q.qlen--; - return len; - } - } - return 0; -} - static const struct Qdisc_class_ops hfsc_class_ops = { .change = hfsc_change_class, .delete = hfsc_delete_class, @@ -1728,7 +1703,6 @@ static struct Qdisc_ops hfsc_qdisc_ops __read_mostly = { .enqueue = hfsc_enqueue, .dequeue = hfsc_dequeue, .peek = qdisc_peek_dequeued, - .drop = hfsc_drop, .cl_ops = &hfsc_class_ops, .priv_size = sizeof(struct hfsc_sched), .owner = THIS_MODULE diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c index 13d6f83..c517918 100644 --- a/net/sched/sch_hhf.c +++ b/net/sched/sch_hhf.c @@ -368,15 +368,6 @@ static unsigned int hhf_drop(struct Qdisc *sch) return bucket - q->buckets; } -static unsigned int hhf_qdisc_drop(struct Qdisc *sch) -{ - unsigned int prev_backlog; - - prev_backlog = sch->qstats.backlog; - hhf_drop(sch); - return prev_backlog - sch->qstats.backlog; -} - static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch) { struct hhf_sched_data *q = qdisc_priv(sch); @@ -709,7 +700,6 @@ static struct Qdisc_ops hhf_qdisc_ops __read_mostly = { .enqueue = hhf_enqueue, .dequeue = hhf_dequeue, .peek = qdisc_peek_dequeued, - .drop = hhf_qdisc_drop, .init = hhf_init, .reset = hhf_reset, .destroy = hhf_destroy, diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 2b05764..b74d066 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -936,31 +936,6 @@ fin: return skb; } -/* try to drop from each class (by prio) until one succeed */ -static unsigned int htb_drop(struct Qdisc *sch) -{ - struct htb_sched *q = qdisc_priv(sch); - int prio; - - for (prio = TC_HTB_NUMPRIO - 1; prio >= 0; prio--) { - struct list_head *p; - list_for_each(p, q->drops + prio) { - struct htb_class *cl = list_entry(p, struct htb_class, - un.leaf.drop_list); - unsigned int len; - if (cl->un.leaf.q->ops->drop && - (len = cl->un.leaf.q->ops->drop(cl->un.leaf.q))) { - sch->qstats.backlog -= len; - sch->q.qlen--; - if (!cl->un.leaf.q->q.qlen) - htb_deactivate(q, cl); - return len; - } - } - } - return 0; -} - /* reset all classes */ /* always caled under BH & queue lock */ static void htb_reset(struct Qdisc *sch) @@ -1600,7 +1575,6 @@ static struct Qdisc_ops htb_qdisc_ops __read_mostly = { .enqueue = htb_enqueue, .dequeue = htb_dequeue, .peek = qdisc_peek_dequeued, - .drop = htb_drop, .init = htb_init, .reset = htb_reset, .destroy = htb_destroy, diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c index 21e69d2..5ea9330 100644 --- a/net/sched/sch_multiq.c +++ b/net/sched/sch_multiq.c @@ -151,27 +151,6 @@ static struct sk_buff *multiq_peek(struct Qdisc *sch) } -static unsigned int multiq_drop(struct Qdisc *sch) -{ - struct multiq_sched_data *q = qdisc_priv(sch); - int band; - unsigned int len; - struct Qdisc *qdisc; - - for (band = q->bands - 1; band >= 0; band--) { - qdisc = q->queues[band]; - if (qdisc->ops->drop) { - len = qdisc->ops->drop(qdisc); - if (len != 0) { - sch->q.qlen--; - return len; - } - } - } - return 0; -} - - static void multiq_reset(struct Qdisc *sch) { @@ -416,7 +395,6 @@ static struct Qdisc_ops multiq_qdisc_ops __read_mostly = { .enqueue = multiq_enqueue, .dequeue = multiq_dequeue, .peek = multiq_peek, - .drop = multiq_drop, .init = multiq_init, .reset = multiq_reset, .destroy = multiq_destroy, diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 31984c7..9ca7947 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -576,35 +576,6 @@ finish_segs: return NET_XMIT_SUCCESS; } -static unsigned int netem_drop(struct Qdisc *sch) -{ - struct netem_sched_data *q = qdisc_priv(sch); - unsigned int len; - - len = qdisc_queue_drop(sch); - - if (!len) { - struct rb_node *p = rb_first(&q->t_root); - - if (p) { - struct sk_buff *skb = netem_rb_to_skb(p); - - rb_erase(p, &q->t_root); - sch->q.qlen--; - skb->next = NULL; - skb->prev = NULL; - qdisc_qstats_backlog_dec(sch, skb); - kfree_skb(skb); - } - } - if (!len && q->qdisc && q->qdisc->ops->drop) - len = q->qdisc->ops->drop(q->qdisc); - if (len) - qdisc_qstats_drop(sch); - - return len; -} - static struct sk_buff *netem_dequeue(struct Qdisc *sch) { struct netem_sched_data *q = qdisc_priv(sch); @@ -1143,7 +1114,6 @@ static struct Qdisc_ops netem_qdisc_ops __read_mostly = { .enqueue = netem_enqueue, .dequeue = netem_dequeue, .peek = qdisc_peek_dequeued, - .drop = netem_drop, .init = netem_init, .reset = netem_reset, .destroy = netem_destroy, diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index 06eca70..4c77780 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -125,24 +125,6 @@ static struct sk_buff *prio_dequeue(struct Qdisc *sch) } -static unsigned int prio_drop(struct Qdisc *sch) -{ - struct prio_sched_data *q = qdisc_priv(sch); - int prio; - unsigned int len; - struct Qdisc *qdisc; - - for (prio = q->bands-1; prio >= 0; prio--) { - qdisc = q->queues[prio]; - if (qdisc->ops->drop && (len = qdisc->ops->drop(qdisc)) != 0) { - sch->q.qlen--; - return len; - } - } - return 0; -} - - static void prio_reset(struct Qdisc *sch) { @@ -379,7 +361,6 @@ static struct Qdisc_ops prio_qdisc_ops __read_mostly = { .enqueue = prio_enqueue, .dequeue = prio_dequeue, .peek = prio_peek, - .drop = prio_drop, .init = prio_init, .reset = prio_reset, .destroy = prio_destroy, diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 85d4197..cdb3966 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -1423,52 +1423,6 @@ static void qfq_qlen_notify(struct Qdisc *sch, unsigned long arg) qfq_deactivate_class(q, cl); } -static unsigned int qfq_drop_from_slot(struct qfq_sched *q, - struct hlist_head *slot) -{ - struct qfq_aggregate *agg; - struct qfq_class *cl; - unsigned int len; - - hlist_for_each_entry(agg, slot, next) { - list_for_each_entry(cl, &agg->active, alist) { - - if (!cl->qdisc->ops->drop) - continue; - - len = cl->qdisc->ops->drop(cl->qdisc); - if (len > 0) { - if (cl->qdisc->q.qlen == 0) - qfq_deactivate_class(q, cl); - - return len; - } - } - } - return 0; -} - -static unsigned int qfq_drop(struct Qdisc *sch) -{ - struct qfq_sched *q = qdisc_priv(sch); - struct qfq_group *grp; - unsigned int i, j, len; - - for (i = 0; i <= QFQ_MAX_INDEX; i++) { - grp = &q->groups[i]; - for (j = 0; j < QFQ_MAX_SLOTS; j++) { - len = qfq_drop_from_slot(q, &grp->slots[j]); - if (len > 0) { - sch->q.qlen--; - return len; - } - } - - } - - return 0; -} - static int qfq_init_qdisc(struct Qdisc *sch, struct nlattr *opt) { struct qfq_sched *q = qdisc_priv(sch); @@ -1563,7 +1517,6 @@ static struct Qdisc_ops qfq_qdisc_ops __read_mostly = { .enqueue = qfq_enqueue, .dequeue = qfq_dequeue, .peek = qdisc_peek_dequeued, - .drop = qfq_drop, .init = qfq_init_qdisc, .reset = qfq_reset_qdisc, .destroy = qfq_destroy_qdisc, diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c index 8c0508c..235942f 100644 --- a/net/sched/sch_red.c +++ b/net/sched/sch_red.c @@ -134,25 +134,6 @@ static struct sk_buff *red_peek(struct Qdisc *sch) return child->ops->peek(child); } -static unsigned int red_drop(struct Qdisc *sch) -{ - struct red_sched_data *q = qdisc_priv(sch); - struct Qdisc *child = q->qdisc; - unsigned int len; - - if (child->ops->drop && (len = child->ops->drop(child)) > 0) { - q->stats.other++; - qdisc_qstats_drop(sch); - sch->q.qlen--; - return len; - } - - if (!red_is_idling(&q->vars)) - red_start_of_idle_period(&q->vars); - - return 0; -} - static void red_reset(struct Qdisc *sch) { struct red_sched_data *q = qdisc_priv(sch); @@ -361,7 +342,6 @@ static struct Qdisc_ops red_qdisc_ops __read_mostly = { .enqueue = red_enqueue, .dequeue = red_dequeue, .peek = red_peek, - .drop = red_drop, .init = red_init, .reset = red_reset, .destroy = red_destroy, diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 498f0a2..a2e0b85 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -896,7 +896,6 @@ static struct Qdisc_ops sfq_qdisc_ops __read_mostly = { .enqueue = sfq_enqueue, .dequeue = sfq_dequeue, .peek = qdisc_peek_dequeued, - .drop = sfq_drop, .init = sfq_init, .reset = sfq_reset, .destroy = sfq_destroy, diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index 801148c..1342e46 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -211,18 +211,6 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch) return NET_XMIT_SUCCESS; } -static unsigned int tbf_drop(struct Qdisc *sch) -{ - struct tbf_sched_data *q = qdisc_priv(sch); - unsigned int len = 0; - - if (q->qdisc->ops->drop && (len = q->qdisc->ops->drop(q->qdisc)) != 0) { - sch->q.qlen--; - qdisc_qstats_drop(sch); - } - return len; -} - static bool tbf_peak_present(const struct tbf_sched_data *q) { return q->peak.rate_bytes_ps; @@ -555,7 +543,6 @@ static struct Qdisc_ops tbf_qdisc_ops __read_mostly = { .enqueue = tbf_enqueue, .dequeue = tbf_dequeue, .peek = qdisc_peek_dequeued, - .drop = tbf_drop, .init = tbf_init, .reset = tbf_reset, .destroy = tbf_destroy, -- cgit v0.10.2 From c8945043cdc687388b7a43fc6f474bddd9607e80 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 9 Jun 2016 00:27:43 +0200 Subject: sched: place state, next_sched and gso_skb in same cacheline again Earlier commits removed two members from struct Qdisc which places next_sched/gso_skb into a different cacheline than ->state. This restores the struct layout to what it was before the removal. Move the two members, then add an annotation so they all reside in the same cacheline. This adds a 16 byte hole after cpu_qstats. The hole could be closed but as it doesn't decrease total struct size just do it this way. Reported-by: Eric Dumazet Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 4dedb7f..49534e2 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -71,11 +71,11 @@ struct Qdisc { struct gnet_stats_basic_cpu __percpu *cpu_bstats; struct gnet_stats_queue __percpu *cpu_qstats; - struct Qdisc *next_sched; - struct sk_buff *gso_skb; /* * For performance sake on SMP, we put highly modified fields at the end */ + struct Qdisc *next_sched ____cacheline_aligned_in_smp; + struct sk_buff *gso_skb; unsigned long state; struct sk_buff_head q; struct gnet_stats_basic_packed bstats; -- cgit v0.10.2 From 56ab364f17637a6e5a67623ff1d6ed4a505025c2 Mon Sep 17 00:00:00 2001 From: Kirtika Ruchandani Date: Sun, 29 May 2016 19:54:10 -0700 Subject: nl80211: Fix spelling Fix 'implementation' spelling, reported by checkpatch.pl Signed-off-by: Kirtika Ruchandani Reviewed-by: Julian Calaby Signed-off-by: Johannes Berg diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index d120449..cf588f1 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -6552,7 +6552,7 @@ nl80211_parse_sched_scan(struct wiphy *wiphy, struct wireless_dev *wdev, nla_data(ssid), nla_len(ssid)); request->match_sets[i].ssid.ssid_len = nla_len(ssid); - /* special attribute - old implemenation w/a */ + /* special attribute - old implementation w/a */ request->match_sets[i].rssi_thold = default_match_rssi; rssi = tb[NL80211_SCHED_SCAN_MATCH_ATTR_RSSI]; -- cgit v0.10.2 From 7a087e7484c9e40203c89b11aa1508ea8d5d649f Mon Sep 17 00:00:00 2001 From: Kirtika Ruchandani Date: Sun, 29 May 2016 19:51:23 -0700 Subject: nl80211: Fix checkpatch warnings about blank lines This patch fixes the following checkpatch.pl issues - - Please don't use multiple blank lines - Blank lines aren't necessary before a close brace - Missing a blank line after declarations Reviewed-by: Julian Calaby Signed-off-by: Kirtika Ruchandani Signed-off-by: Johannes Berg diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index cf588f1..0d7db10 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -167,6 +167,7 @@ __cfg80211_rdev_from_attrs(struct net *netns, struct nlattr **attrs) if (attrs[NL80211_ATTR_IFINDEX]) { int ifindex = nla_get_u32(attrs[NL80211_ATTR_IFINDEX]); + netdev = __dev_get_by_index(netns, ifindex); if (netdev) { if (netdev->ieee80211_ptr) @@ -731,6 +732,7 @@ static int nl80211_parse_key_new(struct nlattr *key, struct key_parse *k) if (tb[NL80211_KEY_DEFAULT_TYPES]) { struct nlattr *kdt[NUM_NL80211_KEY_DEFAULT_TYPES]; + err = nla_parse_nested(kdt, NUM_NL80211_KEY_DEFAULT_TYPES - 1, tb[NL80211_KEY_DEFAULT_TYPES], nl80211_key_default_policy); @@ -1382,6 +1384,7 @@ static int nl80211_send_wiphy(struct cfg80211_registered_device *rdev, rdev->ops->get_antenna) { u32 tx_ant = 0, rx_ant = 0; int res; + res = rdev_get_antenna(rdev, &tx_ant, &rx_ant); if (!res) { if (nla_put_u32(msg, @@ -2157,7 +2160,6 @@ static int nl80211_set_wds_peer(struct sk_buff *skb, struct genl_info *info) return rdev_set_wds_peer(rdev, dev, bssid); } - static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) { struct cfg80211_registered_device *rdev; @@ -2292,6 +2294,7 @@ static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info) if (info->attrs[NL80211_ATTR_WIPHY_ANTENNA_TX] && info->attrs[NL80211_ATTR_WIPHY_ANTENNA_RX]) { u32 tx_ant, rx_ant; + if ((!rdev->wiphy.available_antennas_tx && !rdev->wiphy.available_antennas_rx) || !rdev->ops->set_antenna) @@ -2960,6 +2963,7 @@ static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info) pairwise = !!mac_addr; if (info->attrs[NL80211_ATTR_KEY_TYPE]) { u32 kt = nla_get_u32(info->attrs[NL80211_ATTR_KEY_TYPE]); + if (kt >= NUM_NL80211_KEYTYPES) return -EINVAL; if (kt != NL80211_KEYTYPE_GROUP && @@ -4003,7 +4007,6 @@ static int nl80211_dump_station(struct sk_buff *skb, sta_idx++; } - out: cb->args[2] = sta_idx; err = skb->len; @@ -4804,7 +4807,6 @@ static int nl80211_dump_mpath(struct sk_buff *skb, path_idx++; } - out: cb->args[2] = path_idx; err = skb->len; @@ -5094,7 +5096,6 @@ static int nl80211_req_set_reg(struct sk_buff *skb, struct genl_info *info) enum nl80211_user_reg_hint_type user_reg_hint_type; u32 owner_nlportid; - /* * You should only get this when cfg80211 hasn't yet initialized * completely when built-in to the kernel right between the time @@ -5303,7 +5304,6 @@ do { \ } \ } while (0) - if (!info->attrs[NL80211_ATTR_MESH_CONFIG]) return -EINVAL; if (nla_parse_nested(tb, NL80211_MESHCONF_ATTR_MAX, @@ -5450,7 +5450,6 @@ static int nl80211_parse_mesh_setup(struct genl_info *info, IEEE80211_PATH_METRIC_VENDOR : IEEE80211_PATH_METRIC_AIRTIME; - if (tb[NL80211_MESH_SETUP_IE]) { struct nlattr *ieattr = tb[NL80211_MESH_SETUP_IE]; @@ -6074,6 +6073,7 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) /* all channels */ for (band = 0; band < NUM_NL80211_BANDS; band++) { int j; + if (!wiphy->bands[band]) continue; for (j = 0; j < wiphy->bands[band]->n_channels; j++) { @@ -6483,6 +6483,7 @@ nl80211_parse_sched_scan(struct wiphy *wiphy, struct wireless_dev *wdev, /* all channels */ for (band = 0; band < NUM_NL80211_BANDS; band++) { int j; + if (!wiphy->bands[band]) continue; for (j = 0; j < wiphy->bands[band]->n_channels; j++) { @@ -7245,6 +7246,7 @@ static int nl80211_authenticate(struct sk_buff *skb, struct genl_info *info) if (key.idx >= 0) { int i; bool ok = false; + for (i = 0; i < rdev->wiphy.n_cipher_suites; i++) { if (key.p.cipher == rdev->wiphy.cipher_suites[i]) { ok = true; @@ -7323,6 +7325,7 @@ static int nl80211_crypto_settings(struct cfg80211_registered_device *rdev, if (info->attrs[NL80211_ATTR_CONTROL_PORT_ETHERTYPE]) { u16 proto; + proto = nla_get_u16( info->attrs[NL80211_ATTR_CONTROL_PORT_ETHERTYPE]); settings->control_port_ethertype = cpu_to_be16(proto); @@ -8476,6 +8479,7 @@ static u32 rateset_to_mask(struct ieee80211_supported_band *sband, for (i = 0; i < rates_len; i++) { int rate = (rates[i] & 0x7f) * 5; int ridx; + for (ridx = 0; ridx < sband->n_bitrates; ridx++) { struct ieee80211_rate *srate = &sband->bitrates[ridx]; @@ -8784,7 +8788,6 @@ static int nl80211_tx_mgmt(struct sk_buff *skb, struct genl_info *info) if (params.wait < NL80211_MIN_REMAIN_ON_CHANNEL_TIME || params.wait > rdev->wiphy.max_remain_on_channel_duration) return -EINVAL; - } params.offchan = info->attrs[NL80211_ATTR_OFFCHANNEL_TX_OK]; @@ -10631,7 +10634,6 @@ int cfg80211_vendor_cmd_reply(struct sk_buff *skb) } EXPORT_SYMBOL_GPL(cfg80211_vendor_cmd_reply); - static int nl80211_set_qos_map(struct sk_buff *skb, struct genl_info *info) { @@ -12170,7 +12172,6 @@ void nl80211_send_connect_result(struct cfg80211_registered_device *rdev, nla_put_failure: genlmsg_cancel(msg, hdr); nlmsg_free(msg); - } void nl80211_send_roamed(struct cfg80211_registered_device *rdev, @@ -12209,7 +12210,6 @@ void nl80211_send_roamed(struct cfg80211_registered_device *rdev, nla_put_failure: genlmsg_cancel(msg, hdr); nlmsg_free(msg); - } void nl80211_send_disconnected(struct cfg80211_registered_device *rdev, @@ -12247,7 +12247,6 @@ void nl80211_send_disconnected(struct cfg80211_registered_device *rdev, nla_put_failure: genlmsg_cancel(msg, hdr); nlmsg_free(msg); - } void nl80211_send_ibss_bssid(struct cfg80211_registered_device *rdev, @@ -13589,7 +13588,6 @@ void cfg80211_crit_proto_stopped(struct wireless_dev *wdev, gfp_t gfp) if (hdr) genlmsg_cancel(msg, hdr); nlmsg_free(msg); - } EXPORT_SYMBOL(cfg80211_crit_proto_stopped); -- cgit v0.10.2 From 1dad640b9ef320e8de92c418bcc08448d67590a4 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 9 Jun 2016 10:14:39 +0200 Subject: wext: reformat struct/union declarations Everytime I need to look for these, my usual strategy fails because it assumes the right formatting. Fix the formatting here to make it consistent with the rest of the kernel. Signed-off-by: Johannes Berg diff --git a/include/uapi/linux/wireless.h b/include/uapi/linux/wireless.h index c1592e3..d9ecd7c 100644 --- a/include/uapi/linux/wireless.h +++ b/include/uapi/linux/wireless.h @@ -670,8 +670,7 @@ /* * Generic format for most parameters that fit in an int */ -struct iw_param -{ +struct iw_param { __s32 value; /* The value of the parameter itself */ __u8 fixed; /* Hardware should not use auto select */ __u8 disabled; /* Disable the feature */ @@ -682,8 +681,7 @@ struct iw_param * For all data larger than 16 octets, we need to use a * pointer to memory allocated in user space. */ -struct iw_point -{ +struct iw_point { void __user *pointer; /* Pointer to the data (in user space) */ __u16 length; /* number of fields or size in bytes */ __u16 flags; /* Optional params */ @@ -698,8 +696,7 @@ struct iw_point * of 10 to get 'm' lower than 10^9, with 'm'= f / (10^'e')... * The power of 10 is in 'e', the result of the division is in 'm'. */ -struct iw_freq -{ +struct iw_freq { __s32 m; /* Mantissa */ __s16 e; /* Exponent */ __u8 i; /* List index (when in range struct) */ @@ -709,8 +706,7 @@ struct iw_freq /* * Quality of the link */ -struct iw_quality -{ +struct iw_quality { __u8 qual; /* link quality (%retries, SNR, %missed beacons or better...) */ __u8 level; /* signal level (dBm) */ @@ -725,8 +721,7 @@ struct iw_quality * is already pretty exhaustive, and you should use that first. * This is only additional stats... */ -struct iw_discarded -{ +struct iw_discarded { __u32 nwid; /* Rx : Wrong nwid/essid */ __u32 code; /* Rx : Unable to code/decode (WEP) */ __u32 fragment; /* Rx : Can't perform MAC reassembly */ @@ -738,16 +733,14 @@ struct iw_discarded * Packet/Time period missed in the wireless adapter due to * "wireless" specific problems... */ -struct iw_missed -{ +struct iw_missed { __u32 beacon; /* Missed beacons/superframe */ }; /* * Quality range (for spy threshold) */ -struct iw_thrspy -{ +struct iw_thrspy { struct sockaddr addr; /* Source address (hw/mac) */ struct iw_quality qual; /* Quality of the link */ struct iw_quality low; /* Low threshold */ @@ -765,8 +758,7 @@ struct iw_thrspy * Especially, scan results are required to include an entry for the * current BSS if the driver is in Managed mode and associated with an AP. */ -struct iw_scan_req -{ +struct iw_scan_req { __u8 scan_type; /* IW_SCAN_TYPE_{ACTIVE,PASSIVE} */ __u8 essid_len; __u8 num_channels; /* num entries in channel_list; @@ -827,8 +819,7 @@ struct iw_scan_req * RX_SEQ_VALID for SIOCGIWENCODEEXT are optional, but can be useful for * debugging/testing. */ -struct iw_encode_ext -{ +struct iw_encode_ext { __u32 ext_flags; /* IW_ENCODE_EXT_* */ __u8 tx_seq[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ __u8 rx_seq[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ @@ -841,8 +832,7 @@ struct iw_encode_ext }; /* SIOCSIWMLME data */ -struct iw_mlme -{ +struct iw_mlme { __u16 cmd; /* IW_MLME_* */ __u16 reason_code; struct sockaddr addr; @@ -855,16 +845,14 @@ struct iw_mlme #define IW_PMKID_LEN 16 -struct iw_pmksa -{ +struct iw_pmksa { __u32 cmd; /* IW_PMKSA_* */ struct sockaddr bssid; __u8 pmkid[IW_PMKID_LEN]; }; /* IWEVMICHAELMICFAILURE data */ -struct iw_michaelmicfailure -{ +struct iw_michaelmicfailure { __u32 flags; struct sockaddr src_addr; __u8 tsc[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ @@ -872,8 +860,7 @@ struct iw_michaelmicfailure /* IWEVPMKIDCAND data */ #define IW_PMKID_CAND_PREAUTH 0x00000001 /* RNS pre-authentication enabled */ -struct iw_pmkid_cand -{ +struct iw_pmkid_cand { __u32 flags; /* IW_PMKID_CAND_* */ __u32 index; /* the smaller the index, the higher the * priority */ @@ -884,8 +871,7 @@ struct iw_pmkid_cand /* * Wireless statistics (used for /proc/net/wireless) */ -struct iw_statistics -{ +struct iw_statistics { __u16 status; /* Status * - device dependent for now */ @@ -897,7 +883,7 @@ struct iw_statistics /* ------------------------ IOCTL REQUEST ------------------------ */ /* - * This structure defines the payload of an ioctl, and is used + * This structure defines the payload of an ioctl, and is used * below. * * Note that this structure should fit on the memory footprint @@ -906,8 +892,7 @@ struct iw_statistics * You should check this when increasing the structures defined * above in this file... */ -union iwreq_data -{ +union iwreq_data { /* Config - generic */ char name[IFNAMSIZ]; /* Name : used to verify the presence of wireless extensions. @@ -944,15 +929,14 @@ union iwreq_data * convenience... * Do I need to remind you about structure size (32 octets) ? */ -struct iwreq -{ +struct iwreq { union { char ifrn_name[IFNAMSIZ]; /* if name, e.g. "eth0" */ } ifr_ifrn; /* Data part (defined just above) */ - union iwreq_data u; + union iwreq_data u; }; /* -------------------------- IOCTL DATA -------------------------- */ @@ -965,8 +949,7 @@ struct iwreq * Range of parameters */ -struct iw_range -{ +struct iw_range { /* Informative stuff (to choose between different interface) */ __u32 throughput; /* To give an idea... */ /* In theory this value should be the maximum benchmarked @@ -1069,9 +1052,8 @@ struct iw_range /* * Private ioctl interface information */ - -struct iw_priv_args -{ + +struct iw_priv_args { __u32 cmd; /* Number of the ioctl to issue */ __u16 set_args; /* Type and number of args */ __u16 get_args; /* Type and number of args */ @@ -1088,8 +1070,7 @@ struct iw_priv_args /* * A Wireless Event. Contains basically the same data as the ioctl... */ -struct iw_event -{ +struct iw_event { __u16 len; /* Real length of this stuff */ __u16 cmd; /* Wireless IOCTL */ union iwreq_data u; /* IOCTL fixed payload */ -- cgit v0.10.2 From 0662799023de4be686263b0a4f4b7910999172b9 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 9 Jun 2016 10:40:09 +0200 Subject: nl80211: clarify nl80211_set_reg() success path Setting rd to NULL to avoid freeing it, just to be able to return from the function in a single place, doesn't make much sense. Return the set_regdom() return value directly. Signed-off-by: Johannes Berg diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 0d7db10..c503e96 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -5836,10 +5836,8 @@ static int nl80211_set_reg(struct sk_buff *skb, struct genl_info *info) } } - r = set_regdom(rd, REGD_SOURCE_CRDA); - /* set_regdom took ownership */ - rd = NULL; - + /* set_regdom takes ownership of rd */ + return set_regdom(rd, REGD_SOURCE_CRDA); bad_reg: kfree(rd); return r; -- cgit v0.10.2 From 80a83cfc434b1e3afe38974570b460db4898bec6 Mon Sep 17 00:00:00 2001 From: Michal Kazior Date: Thu, 19 May 2016 10:37:48 +0200 Subject: mac80211: skip netdev queue control with software queuing Qdiscs are designed with no regard to 802.11 aggregation requirements and hand out packet-by-packet with no guarantee they are destined to the same tid. This does more bad than good no matter how fairly a given qdisc may behave on an ethernet interface. Software queuing used per-AC netdev subqueue congestion control whenever a global AC limit was hit. This meant in practice a single station or tid queue could starve others rather easily. This could resonate with qdiscs in a bad way or could just end up with poor aggregation performance. Increasing the AC limit would increase induced latency which is also bad. Disabling qdiscs by default and performing taildrop instead of netdev subqueue congestion control on the other hand makes it possible for tid queues to fill up "in the meantime" while preventing stations starving each other. This increases aggregation opportunities and should allow software queuing based drivers achieve better performance by utilizing airtime more efficiently with big aggregates. Signed-off-by: Michal Kazior Signed-off-by: Johannes Berg diff --git a/include/net/mac80211.h b/include/net/mac80211.h index be30b05..a8683ae 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -2147,9 +2147,6 @@ enum ieee80211_hw_flags { * @n_cipher_schemes: a size of an array of cipher schemes definitions. * @cipher_schemes: a pointer to an array of cipher scheme definitions * supported by HW. - * - * @txq_ac_max_pending: maximum number of frames per AC pending in all txq - * entries for a vif. */ struct ieee80211_hw { struct ieee80211_conf conf; @@ -2180,7 +2177,6 @@ struct ieee80211_hw { u8 uapsd_max_sp_len; u8 n_cipher_schemes; const struct ieee80211_cipher_scheme *cipher_schemes; - int txq_ac_max_pending; }; static inline bool _ieee80211_hw_check(struct ieee80211_hw *hw, diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 9438c94..6346033 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -856,7 +856,7 @@ struct ieee80211_sub_if_data { bool control_port_no_encrypt; int encrypt_headroom; - atomic_t txqs_len[IEEE80211_NUM_ACS]; + atomic_t num_tx_queued; struct ieee80211_tx_queue_params tx_conf[IEEE80211_NUM_ACS]; struct mac80211_qos_map __rcu *qos_map; diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index c59af3e..609c517 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -976,13 +976,13 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, if (sdata->vif.txq) { struct txq_info *txqi = to_txq_info(sdata->vif.txq); + int n = skb_queue_len(&txqi->queue); spin_lock_bh(&txqi->queue.lock); ieee80211_purge_tx_queue(&local->hw, &txqi->queue); + atomic_sub(n, &sdata->num_tx_queued); txqi->byte_cnt = 0; spin_unlock_bh(&txqi->queue.lock); - - atomic_set(&sdata->txqs_len[txqi->txq.ac], 0); } if (local->open_count == 0) @@ -1198,6 +1198,12 @@ static void ieee80211_if_setup(struct net_device *dev) dev->destructor = ieee80211_if_free; } +static void ieee80211_if_setup_no_queue(struct net_device *dev) +{ + ieee80211_if_setup(dev); + dev->priv_flags |= IFF_NO_QUEUE; +} + static void ieee80211_iface_work(struct work_struct *work) { struct ieee80211_sub_if_data *sdata = @@ -1707,6 +1713,7 @@ int ieee80211_if_add(struct ieee80211_local *local, const char *name, struct net_device *ndev = NULL; struct ieee80211_sub_if_data *sdata = NULL; struct txq_info *txqi; + void (*if_setup)(struct net_device *dev); int ret, i; int txqs = 1; @@ -1734,12 +1741,17 @@ int ieee80211_if_add(struct ieee80211_local *local, const char *name, txq_size += sizeof(struct txq_info) + local->hw.txq_data_size; + if (local->ops->wake_tx_queue) + if_setup = ieee80211_if_setup_no_queue; + else + if_setup = ieee80211_if_setup; + if (local->hw.queues >= IEEE80211_NUM_ACS) txqs = IEEE80211_NUM_ACS; ndev = alloc_netdev_mqs(size + txq_size, name, name_assign_type, - ieee80211_if_setup, txqs, 1); + if_setup, txqs, 1); if (!ndev) return -ENOMEM; dev_net_set(ndev, wiphy_net(local->hw.wiphy)); diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 7ee91d6..160ac6b 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -1055,9 +1055,6 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) local->dynamic_ps_forced_timeout = -1; - if (!local->hw.txq_ac_max_pending) - local->hw.txq_ac_max_pending = 64; - result = ieee80211_wep_init(local); if (result < 0) wiphy_debug(local->hw.wiphy, "Failed to initialize wep: %d\n", diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 5ccfdbd..177cc6c 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -116,7 +116,7 @@ static void __cleanup_single_sta(struct sta_info *sta) int n = skb_queue_len(&txqi->queue); ieee80211_purge_tx_queue(&local->hw, &txqi->queue); - atomic_sub(n, &sdata->txqs_len[txqi->txq.ac]); + atomic_sub(n, &sdata->num_tx_queued); txqi->byte_cnt = 0; } } diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 2030443..3e77da1 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1236,27 +1236,21 @@ ieee80211_tx_prepare(struct ieee80211_sub_if_data *sdata, return TX_CONTINUE; } -static void ieee80211_drv_tx(struct ieee80211_local *local, - struct ieee80211_vif *vif, - struct ieee80211_sta *pubsta, - struct sk_buff *skb) +static struct txq_info *ieee80211_get_txq(struct ieee80211_local *local, + struct ieee80211_vif *vif, + struct ieee80211_sta *pubsta, + struct sk_buff *skb) { struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data; - struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif); struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); - struct ieee80211_tx_control control = { - .sta = pubsta, - }; struct ieee80211_txq *txq = NULL; - struct txq_info *txqi; - u8 ac; if ((info->flags & IEEE80211_TX_CTL_SEND_AFTER_DTIM) || (info->control.flags & IEEE80211_TX_CTRL_PS_RESPONSE)) - goto tx_normal; + return NULL; if (!ieee80211_is_data(hdr->frame_control)) - goto tx_normal; + return NULL; if (pubsta) { u8 tid = skb->priority & IEEE80211_QOS_CTL_TID_MASK; @@ -1267,25 +1261,28 @@ static void ieee80211_drv_tx(struct ieee80211_local *local, } if (!txq) - goto tx_normal; + return NULL; - ac = txq->ac; - txqi = to_txq_info(txq); - atomic_inc(&sdata->txqs_len[ac]); - if (atomic_read(&sdata->txqs_len[ac]) >= local->hw.txq_ac_max_pending) - netif_stop_subqueue(sdata->dev, ac); + return to_txq_info(txq); +} - spin_lock_bh(&txqi->queue.lock); - txqi->byte_cnt += skb->len; - __skb_queue_tail(&txqi->queue, skb); - spin_unlock_bh(&txqi->queue.lock); +static void ieee80211_txq_enqueue(struct ieee80211_local *local, + struct txq_info *txqi, + struct sk_buff *skb) +{ + struct ieee80211_sub_if_data *sdata = vif_to_sdata(txqi->txq.vif); - drv_wake_tx_queue(local, txqi); + lockdep_assert_held(&txqi->queue.lock); - return; + if (atomic_read(&sdata->num_tx_queued) >= TOTAL_MAX_TX_BUFFER || + txqi->queue.qlen >= STA_MAX_TX_BUFFER) { + ieee80211_free_txskb(&local->hw, skb); + return; + } -tx_normal: - drv_tx(local, &control, skb); + atomic_inc(&sdata->num_tx_queued); + txqi->byte_cnt += skb->len; + __skb_queue_tail(&txqi->queue, skb); } struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, @@ -1296,7 +1293,6 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, struct txq_info *txqi = container_of(txq, struct txq_info, txq); struct ieee80211_hdr *hdr; struct sk_buff *skb = NULL; - u8 ac = txq->ac; spin_lock_bh(&txqi->queue.lock); @@ -1307,12 +1303,9 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, if (!skb) goto out; + atomic_dec(&sdata->num_tx_queued); txqi->byte_cnt -= skb->len; - atomic_dec(&sdata->txqs_len[ac]); - if (__netif_subqueue_stopped(sdata->dev, ac)) - ieee80211_propagate_queue_wake(local, sdata->vif.hw_queue[ac]); - hdr = (struct ieee80211_hdr *)skb->data; if (txq->sta && ieee80211_is_data_qos(hdr->frame_control)) { struct sta_info *sta = container_of(txq->sta, struct sta_info, @@ -1343,7 +1336,9 @@ static bool ieee80211_tx_frags(struct ieee80211_local *local, struct sk_buff_head *skbs, bool txpending) { + struct ieee80211_tx_control control = {}; struct sk_buff *skb, *tmp; + struct txq_info *txqi; unsigned long flags; skb_queue_walk_safe(skbs, skb, tmp) { @@ -1358,6 +1353,21 @@ static bool ieee80211_tx_frags(struct ieee80211_local *local, } #endif + txqi = ieee80211_get_txq(local, vif, sta, skb); + if (txqi) { + info->control.vif = vif; + + __skb_unlink(skb, skbs); + + spin_lock_bh(&txqi->queue.lock); + ieee80211_txq_enqueue(local, txqi, skb); + spin_unlock_bh(&txqi->queue.lock); + + drv_wake_tx_queue(local, txqi); + + continue; + } + spin_lock_irqsave(&local->queue_stop_reason_lock, flags); if (local->queue_stop_reasons[q] || (!txpending && !skb_queue_empty(&local->pending[q]))) { @@ -1400,9 +1410,10 @@ static bool ieee80211_tx_frags(struct ieee80211_local *local, spin_unlock_irqrestore(&local->queue_stop_reason_lock, flags); info->control.vif = vif; + control.sta = sta; __skb_unlink(skb, skbs); - ieee80211_drv_tx(local, vif, sta, skb); + drv_tx(local, &control, skb); } return true; diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 905003f..0db4644 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -244,6 +244,9 @@ void ieee80211_propagate_queue_wake(struct ieee80211_local *local, int queue) struct ieee80211_sub_if_data *sdata; int n_acs = IEEE80211_NUM_ACS; + if (local->ops->wake_tx_queue) + return; + if (local->hw.queues < IEEE80211_NUM_ACS) n_acs = 1; @@ -260,11 +263,6 @@ void ieee80211_propagate_queue_wake(struct ieee80211_local *local, int queue) for (ac = 0; ac < n_acs; ac++) { int ac_queue = sdata->vif.hw_queue[ac]; - if (local->ops->wake_tx_queue && - (atomic_read(&sdata->txqs_len[ac]) > - local->hw.txq_ac_max_pending)) - continue; - if (ac_queue == queue || (sdata->vif.cab_queue == queue && local->queue_stop_reasons[ac_queue] == 0 && @@ -352,6 +350,9 @@ static void __ieee80211_stop_queue(struct ieee80211_hw *hw, int queue, if (__test_and_set_bit(reason, &local->queue_stop_reasons[queue])) return; + if (local->ops->wake_tx_queue) + return; + if (local->hw.queues < IEEE80211_NUM_ACS) n_acs = 1; -- cgit v0.10.2 From fa962b92120bb70693a4db545f89067eb3373294 Mon Sep 17 00:00:00 2001 From: Michal Kazior Date: Thu, 19 May 2016 10:37:49 +0200 Subject: mac80211: implement fair queueing per txq mac80211's software queues were designed to work very closely with device tx queues. They are required to make use of 802.11 packet aggregation easily and efficiently. Due to the way 802.11 aggregation is designed it only makes sense to keep fair queuing as close to hardware as possible to reduce induced latency and inertia and provide the best flow responsiveness. This change doesn't translate directly to immediate and significant gains. End result depends on driver's induced latency. Best results can be achieved if driver keeps its own tx queue/fifo fill level to a minimum. Signed-off-by: Michal Kazior Signed-off-by: Johannes Berg diff --git a/net/mac80211/agg-tx.c b/net/mac80211/agg-tx.c index 42fa810..5650c46 100644 --- a/net/mac80211/agg-tx.c +++ b/net/mac80211/agg-tx.c @@ -194,17 +194,21 @@ static void ieee80211_agg_stop_txq(struct sta_info *sta, int tid) { struct ieee80211_txq *txq = sta->sta.txq[tid]; + struct ieee80211_sub_if_data *sdata; + struct fq *fq; struct txq_info *txqi; if (!txq) return; txqi = to_txq_info(txq); + sdata = vif_to_sdata(txq->vif); + fq = &sdata->local->fq; /* Lock here to protect against further seqno updates on dequeue */ - spin_lock_bh(&txqi->queue.lock); + spin_lock_bh(&fq->lock); set_bit(IEEE80211_TXQ_STOP, &txqi->flags); - spin_unlock_bh(&txqi->queue.lock); + spin_unlock_bh(&fq->lock); } static void diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 6346033..6f8375f 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -30,6 +30,7 @@ #include #include #include +#include #include "key.h" #include "sta_info.h" #include "debug.h" @@ -805,10 +806,17 @@ enum txq_info_flags { IEEE80211_TXQ_NO_AMSDU, }; +/** + * struct txq_info - per tid queue + * + * @tin: contains packets split into multiple flows + * @def_flow: used as a fallback flow when a packet destined to @tin hashes to + * a fq_flow which is already owned by a different tin + */ struct txq_info { - struct sk_buff_head queue; + struct fq_tin tin; + struct fq_flow def_flow; unsigned long flags; - unsigned long byte_cnt; /* keep last! */ struct ieee80211_txq txq; @@ -1099,6 +1107,8 @@ struct ieee80211_local { * it first anyway so they become a no-op */ struct ieee80211_hw hw; + struct fq fq; + const struct ieee80211_ops *ops; /* @@ -1931,9 +1941,13 @@ static inline bool ieee80211_can_run_worker(struct ieee80211_local *local) return true; } -void ieee80211_init_tx_queue(struct ieee80211_sub_if_data *sdata, - struct sta_info *sta, - struct txq_info *txq, int tid); +int ieee80211_txq_setup_flows(struct ieee80211_local *local); +void ieee80211_txq_teardown_flows(struct ieee80211_local *local); +void ieee80211_txq_init(struct ieee80211_sub_if_data *sdata, + struct sta_info *sta, + struct txq_info *txq, int tid); +void ieee80211_txq_purge(struct ieee80211_local *local, + struct txq_info *txqi); void ieee80211_send_auth(struct ieee80211_sub_if_data *sdata, u16 transaction, u16 auth_alg, u16 status, const u8 *extra, size_t extra_len, const u8 *bssid, diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index 609c517..b123a9e 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -779,6 +779,7 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, bool going_down) { struct ieee80211_local *local = sdata->local; + struct fq *fq = &local->fq; unsigned long flags; struct sk_buff *skb, *tmp; u32 hw_reconf_flags = 0; @@ -976,13 +977,10 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata, if (sdata->vif.txq) { struct txq_info *txqi = to_txq_info(sdata->vif.txq); - int n = skb_queue_len(&txqi->queue); - spin_lock_bh(&txqi->queue.lock); - ieee80211_purge_tx_queue(&local->hw, &txqi->queue); - atomic_sub(n, &sdata->num_tx_queued); - txqi->byte_cnt = 0; - spin_unlock_bh(&txqi->queue.lock); + spin_lock_bh(&fq->lock); + ieee80211_txq_purge(local, txqi); + spin_unlock_bh(&fq->lock); } if (local->open_count == 0) @@ -1792,7 +1790,7 @@ int ieee80211_if_add(struct ieee80211_local *local, const char *name, if (txq_size) { txqi = netdev_priv(ndev) + size; - ieee80211_init_tx_queue(sdata, NULL, txqi, 0); + ieee80211_txq_init(sdata, NULL, txqi, 0); } sdata->dev = ndev; diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 160ac6b..d00ea9b 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -1086,6 +1086,10 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) rtnl_unlock(); + result = ieee80211_txq_setup_flows(local); + if (result) + goto fail_flows; + #ifdef CONFIG_INET local->ifa_notifier.notifier_call = ieee80211_ifa_changed; result = register_inetaddr_notifier(&local->ifa_notifier); @@ -1111,6 +1115,8 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) #if defined(CONFIG_INET) || defined(CONFIG_IPV6) fail_ifa: #endif + ieee80211_txq_teardown_flows(local); + fail_flows: rtnl_lock(); rate_control_deinitialize(local); ieee80211_remove_interfaces(local); @@ -1169,6 +1175,7 @@ void ieee80211_unregister_hw(struct ieee80211_hw *hw) skb_queue_purge(&local->skb_queue); skb_queue_purge(&local->skb_queue_unreliable); skb_queue_purge(&local->skb_queue_tdls_chsw); + ieee80211_txq_teardown_flows(local); destroy_workqueue(local->workqueue); wiphy_unregister(local->hw.wiphy); diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 5e65e83..9a1eb70 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -1268,7 +1268,7 @@ static void sta_ps_start(struct sta_info *sta) for (tid = 0; tid < ARRAY_SIZE(sta->sta.txq); tid++) { struct txq_info *txqi = to_txq_info(sta->sta.txq[tid]); - if (!skb_queue_len(&txqi->queue)) + if (!txqi->tin.backlog_packets) set_bit(tid, &sta->txq_buffered_tids); else clear_bit(tid, &sta->txq_buffered_tids); diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 177cc6c..76b737d 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -90,6 +90,7 @@ static void __cleanup_single_sta(struct sta_info *sta) struct tid_ampdu_tx *tid_tx; struct ieee80211_sub_if_data *sdata = sta->sdata; struct ieee80211_local *local = sdata->local; + struct fq *fq = &local->fq; struct ps_data *ps; if (test_sta_flag(sta, WLAN_STA_PS_STA) || @@ -113,11 +114,10 @@ static void __cleanup_single_sta(struct sta_info *sta) if (sta->sta.txq[0]) { for (i = 0; i < ARRAY_SIZE(sta->sta.txq); i++) { struct txq_info *txqi = to_txq_info(sta->sta.txq[i]); - int n = skb_queue_len(&txqi->queue); - ieee80211_purge_tx_queue(&local->hw, &txqi->queue); - atomic_sub(n, &sdata->num_tx_queued); - txqi->byte_cnt = 0; + spin_lock_bh(&fq->lock); + ieee80211_txq_purge(local, txqi); + spin_unlock_bh(&fq->lock); } } @@ -368,7 +368,7 @@ struct sta_info *sta_info_alloc(struct ieee80211_sub_if_data *sdata, for (i = 0; i < ARRAY_SIZE(sta->sta.txq); i++) { struct txq_info *txq = txq_data + i * size; - ieee80211_init_tx_queue(sdata, sta, txq, i); + ieee80211_txq_init(sdata, sta, txq, i); } } @@ -1211,7 +1211,7 @@ void ieee80211_sta_ps_deliver_wakeup(struct sta_info *sta) for (i = 0; i < ARRAY_SIZE(sta->sta.txq); i++) { struct txq_info *txqi = to_txq_info(sta->sta.txq[i]); - if (!skb_queue_len(&txqi->queue)) + if (!txqi->tin.backlog_packets) continue; drv_wake_tx_queue(local, txqi); @@ -1648,7 +1648,7 @@ ieee80211_sta_ps_deliver_response(struct sta_info *sta, for (tid = 0; tid < ARRAY_SIZE(sta->sta.txq); tid++) { struct txq_info *txqi = to_txq_info(sta->sta.txq[tid]); - if (!(tids & BIT(tid)) || skb_queue_len(&txqi->queue)) + if (!(tids & BIT(tid)) || txqi->tin.backlog_packets) continue; sta_info_recalc_tim(sta); diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 3e77da1..1d8343f 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -25,6 +25,7 @@ #include #include #include +#include #include "ieee80211_i.h" #include "driver-ops.h" @@ -1266,46 +1267,121 @@ static struct txq_info *ieee80211_get_txq(struct ieee80211_local *local, return to_txq_info(txq); } +static struct sk_buff *fq_tin_dequeue_func(struct fq *fq, + struct fq_tin *tin, + struct fq_flow *flow) +{ + return fq_flow_dequeue(fq, flow); +} + +static void fq_skb_free_func(struct fq *fq, + struct fq_tin *tin, + struct fq_flow *flow, + struct sk_buff *skb) +{ + struct ieee80211_local *local; + + local = container_of(fq, struct ieee80211_local, fq); + ieee80211_free_txskb(&local->hw, skb); +} + +static struct fq_flow *fq_flow_get_default_func(struct fq *fq, + struct fq_tin *tin, + int idx, + struct sk_buff *skb) +{ + struct txq_info *txqi; + + txqi = container_of(tin, struct txq_info, tin); + return &txqi->def_flow; +} + static void ieee80211_txq_enqueue(struct ieee80211_local *local, struct txq_info *txqi, struct sk_buff *skb) { - struct ieee80211_sub_if_data *sdata = vif_to_sdata(txqi->txq.vif); + struct fq *fq = &local->fq; + struct fq_tin *tin = &txqi->tin; - lockdep_assert_held(&txqi->queue.lock); + fq_tin_enqueue(fq, tin, skb, + fq_skb_free_func, + fq_flow_get_default_func); +} - if (atomic_read(&sdata->num_tx_queued) >= TOTAL_MAX_TX_BUFFER || - txqi->queue.qlen >= STA_MAX_TX_BUFFER) { - ieee80211_free_txskb(&local->hw, skb); - return; +void ieee80211_txq_init(struct ieee80211_sub_if_data *sdata, + struct sta_info *sta, + struct txq_info *txqi, int tid) +{ + fq_tin_init(&txqi->tin); + fq_flow_init(&txqi->def_flow); + + txqi->txq.vif = &sdata->vif; + + if (sta) { + txqi->txq.sta = &sta->sta; + sta->sta.txq[tid] = &txqi->txq; + txqi->txq.tid = tid; + txqi->txq.ac = ieee802_1d_to_ac[tid & 7]; + } else { + sdata->vif.txq = &txqi->txq; + txqi->txq.tid = 0; + txqi->txq.ac = IEEE80211_AC_BE; } +} - atomic_inc(&sdata->num_tx_queued); - txqi->byte_cnt += skb->len; - __skb_queue_tail(&txqi->queue, skb); +void ieee80211_txq_purge(struct ieee80211_local *local, + struct txq_info *txqi) +{ + struct fq *fq = &local->fq; + struct fq_tin *tin = &txqi->tin; + + fq_tin_reset(fq, tin, fq_skb_free_func); +} + +int ieee80211_txq_setup_flows(struct ieee80211_local *local) +{ + struct fq *fq = &local->fq; + int ret; + + if (!local->ops->wake_tx_queue) + return 0; + + ret = fq_init(fq, 4096); + if (ret) + return ret; + + return 0; +} + +void ieee80211_txq_teardown_flows(struct ieee80211_local *local) +{ + struct fq *fq = &local->fq; + + if (!local->ops->wake_tx_queue) + return; + + fq_reset(fq, fq_skb_free_func); } struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, struct ieee80211_txq *txq) { struct ieee80211_local *local = hw_to_local(hw); - struct ieee80211_sub_if_data *sdata = vif_to_sdata(txq->vif); struct txq_info *txqi = container_of(txq, struct txq_info, txq); struct ieee80211_hdr *hdr; struct sk_buff *skb = NULL; + struct fq *fq = &local->fq; + struct fq_tin *tin = &txqi->tin; - spin_lock_bh(&txqi->queue.lock); + spin_lock_bh(&fq->lock); if (test_bit(IEEE80211_TXQ_STOP, &txqi->flags)) goto out; - skb = __skb_dequeue(&txqi->queue); + skb = fq_tin_dequeue(fq, tin, fq_tin_dequeue_func); if (!skb) goto out; - atomic_dec(&sdata->num_tx_queued); - txqi->byte_cnt -= skb->len; - hdr = (struct ieee80211_hdr *)skb->data; if (txq->sta && ieee80211_is_data_qos(hdr->frame_control)) { struct sta_info *sta = container_of(txq->sta, struct sta_info, @@ -1320,7 +1396,7 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, } out: - spin_unlock_bh(&txqi->queue.lock); + spin_unlock_bh(&fq->lock); if (skb && skb_has_frag_list(skb) && !ieee80211_hw_check(&local->hw, TX_FRAG_LIST)) @@ -1337,6 +1413,7 @@ static bool ieee80211_tx_frags(struct ieee80211_local *local, bool txpending) { struct ieee80211_tx_control control = {}; + struct fq *fq = &local->fq; struct sk_buff *skb, *tmp; struct txq_info *txqi; unsigned long flags; @@ -1359,9 +1436,9 @@ static bool ieee80211_tx_frags(struct ieee80211_local *local, __skb_unlink(skb, skbs); - spin_lock_bh(&txqi->queue.lock); + spin_lock_bh(&fq->lock); ieee80211_txq_enqueue(local, txqi, skb); - spin_unlock_bh(&txqi->queue.lock); + spin_unlock_bh(&fq->lock); drv_wake_tx_queue(local, txqi); @@ -2893,6 +2970,9 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb) { struct ieee80211_local *local = sdata->local; + struct fq *fq = &local->fq; + struct fq_tin *tin; + struct fq_flow *flow; u8 tid = skb->priority & IEEE80211_QOS_CTL_TAG1D_MASK; struct ieee80211_txq *txq = sta->sta.txq[tid]; struct txq_info *txqi; @@ -2904,6 +2984,7 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata, __be16 len; void *data; bool ret = false; + unsigned int orig_len; int n = 1, nfrags; if (!ieee80211_hw_check(&local->hw, TX_AMSDU)) @@ -2920,12 +3001,20 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata, max_amsdu_len = min_t(int, max_amsdu_len, sta->sta.max_rc_amsdu_len); - spin_lock_bh(&txqi->queue.lock); + spin_lock_bh(&fq->lock); - head = skb_peek_tail(&txqi->queue); + /* TODO: Ideally aggregation should be done on dequeue to remain + * responsive to environment changes. + */ + + tin = &txqi->tin; + flow = fq_flow_classify(fq, tin, skb, fq_flow_get_default_func); + head = skb_peek_tail(&flow->queue); if (!head) goto out; + orig_len = head->len; + if (skb->len + head->len > max_amsdu_len) goto out; @@ -2964,8 +3053,13 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata, head->data_len += skb->len; *frag_tail = skb; + flow->backlog += head->len - orig_len; + tin->backlog_bytes += head->len - orig_len; + + fq_recalc_backlog(fq, tin, flow); + out: - spin_unlock_bh(&txqi->queue.lock); + spin_unlock_bh(&fq->lock); return ret; } diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 0db4644..42bf0b6 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -3389,25 +3389,6 @@ u8 *ieee80211_add_wmm_info_ie(u8 *buf, u8 qosinfo) return buf; } -void ieee80211_init_tx_queue(struct ieee80211_sub_if_data *sdata, - struct sta_info *sta, - struct txq_info *txqi, int tid) -{ - skb_queue_head_init(&txqi->queue); - txqi->txq.vif = &sdata->vif; - - if (sta) { - txqi->txq.sta = &sta->sta; - sta->sta.txq[tid] = &txqi->txq; - txqi->txq.tid = tid; - txqi->txq.ac = ieee802_1d_to_ac[tid & 7]; - } else { - sdata->vif.txq = &txqi->txq; - txqi->txq.tid = 0; - txqi->txq.ac = IEEE80211_AC_BE; - } -} - void ieee80211_txq_get_depth(struct ieee80211_txq *txq, unsigned long *frame_cnt, unsigned long *byte_cnt) @@ -3415,9 +3396,9 @@ void ieee80211_txq_get_depth(struct ieee80211_txq *txq, struct txq_info *txqi = to_txq_info(txq); if (frame_cnt) - *frame_cnt = txqi->queue.qlen; + *frame_cnt = txqi->tin.backlog_packets; if (byte_cnt) - *byte_cnt = txqi->byte_cnt; + *byte_cnt = txqi->tin.backlog_bytes; } EXPORT_SYMBOL(ieee80211_txq_get_depth); -- cgit v0.10.2 From 9399b86c0e9a058928e8b5f1e695056714814873 Mon Sep 17 00:00:00 2001 From: Michal Kazior Date: Thu, 19 May 2016 10:37:50 +0200 Subject: mac80211: add debug knobs for fair queuing This adds a debugfs entry to read and modify some fq parameters. This makes it easy to debug, test and experiment. Signed-off-by: Michal Kazior [remove module parameter for now] Signed-off-by: Johannes Berg diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c index b251b2f..2906c10 100644 --- a/net/mac80211/debugfs.c +++ b/net/mac80211/debugfs.c @@ -10,6 +10,7 @@ #include #include +#include #include "ieee80211_i.h" #include "driver-ops.h" #include "rate.h" @@ -70,6 +71,177 @@ DEBUGFS_READONLY_FILE(wep_iv, "%#08x", DEBUGFS_READONLY_FILE(rate_ctrl_alg, "%s", local->rate_ctrl ? local->rate_ctrl->ops->name : "hw/driver"); +struct aqm_info { + struct ieee80211_local *local; + size_t size; + size_t len; + unsigned char buf[0]; +}; + +#define AQM_HDR_LEN 200 +#define AQM_HW_ENTRY_LEN 40 +#define AQM_TXQ_ENTRY_LEN 110 + +static int aqm_open(struct inode *inode, struct file *file) +{ + struct ieee80211_local *local = inode->i_private; + struct ieee80211_sub_if_data *sdata; + struct sta_info *sta; + struct txq_info *txqi; + struct fq *fq = &local->fq; + struct aqm_info *info = NULL; + int len = 0; + int i; + + if (!local->ops->wake_tx_queue) + return -EOPNOTSUPP; + + len += AQM_HDR_LEN; + len += 6 * AQM_HW_ENTRY_LEN; + + rcu_read_lock(); + list_for_each_entry_rcu(sdata, &local->interfaces, list) + len += AQM_TXQ_ENTRY_LEN; + list_for_each_entry_rcu(sta, &local->sta_list, list) + len += AQM_TXQ_ENTRY_LEN * ARRAY_SIZE(sta->sta.txq); + rcu_read_unlock(); + + info = vmalloc(len); + if (!info) + return -ENOMEM; + + spin_lock_bh(&local->fq.lock); + rcu_read_lock(); + + file->private_data = info; + info->local = local; + info->size = len; + len = 0; + + len += scnprintf(info->buf + len, info->size - len, + "* hw\n" + "access name value\n" + "R fq_flows_cnt %u\n" + "R fq_backlog %u\n" + "R fq_overlimit %u\n" + "R fq_collisions %u\n" + "RW fq_limit %u\n" + "RW fq_quantum %u\n", + fq->flows_cnt, + fq->backlog, + fq->overlimit, + fq->collisions, + fq->limit, + fq->quantum); + + len += scnprintf(info->buf + len, + info->size - len, + "* vif\n" + "ifname addr ac backlog-bytes backlog-packets flows overlimit collisions tx-bytes tx-packets\n"); + + list_for_each_entry_rcu(sdata, &local->interfaces, list) { + txqi = to_txq_info(sdata->vif.txq); + len += scnprintf(info->buf + len, info->size - len, + "%s %pM %u %u %u %u %u %u %u %u\n", + sdata->name, + sdata->vif.addr, + txqi->txq.ac, + txqi->tin.backlog_bytes, + txqi->tin.backlog_packets, + txqi->tin.flows, + txqi->tin.overlimit, + txqi->tin.collisions, + txqi->tin.tx_bytes, + txqi->tin.tx_packets); + } + + len += scnprintf(info->buf + len, + info->size - len, + "* sta\n" + "ifname addr tid ac backlog-bytes backlog-packets flows overlimit collisions tx-bytes tx-packets\n"); + + list_for_each_entry_rcu(sta, &local->sta_list, list) { + sdata = sta->sdata; + for (i = 0; i < ARRAY_SIZE(sta->sta.txq); i++) { + txqi = to_txq_info(sta->sta.txq[i]); + len += scnprintf(info->buf + len, info->size - len, + "%s %pM %d %d %u %u %u %u %u %u %u\n", + sdata->name, + sta->sta.addr, + txqi->txq.tid, + txqi->txq.ac, + txqi->tin.backlog_bytes, + txqi->tin.backlog_packets, + txqi->tin.flows, + txqi->tin.overlimit, + txqi->tin.collisions, + txqi->tin.tx_bytes, + txqi->tin.tx_packets); + } + } + + info->len = len; + + rcu_read_unlock(); + spin_unlock_bh(&local->fq.lock); + + return 0; +} + +static int aqm_release(struct inode *inode, struct file *file) +{ + vfree(file->private_data); + return 0; +} + +static ssize_t aqm_read(struct file *file, + char __user *user_buf, + size_t count, + loff_t *ppos) +{ + struct aqm_info *info = file->private_data; + + return simple_read_from_buffer(user_buf, count, ppos, + info->buf, info->len); +} + +static ssize_t aqm_write(struct file *file, + const char __user *user_buf, + size_t count, + loff_t *ppos) +{ + struct aqm_info *info = file->private_data; + struct ieee80211_local *local = info->local; + char buf[100]; + size_t len; + + if (count > sizeof(buf)) + return -EINVAL; + + if (copy_from_user(buf, user_buf, count)) + return -EFAULT; + + buf[sizeof(buf) - 1] = '\0'; + len = strlen(buf); + if (len > 0 && buf[len-1] == '\n') + buf[len-1] = 0; + + if (sscanf(buf, "fq_limit %u", &local->fq.limit) == 1) + return count; + else if (sscanf(buf, "fq_quantum %u", &local->fq.quantum) == 1) + return count; + + return -EINVAL; +} + +static const struct file_operations aqm_ops = { + .write = aqm_write, + .read = aqm_read, + .open = aqm_open, + .release = aqm_release, + .llseek = default_llseek, +}; + #ifdef CONFIG_PM static ssize_t reset_write(struct file *file, const char __user *user_buf, size_t count, loff_t *ppos) @@ -256,6 +428,7 @@ void debugfs_hw_add(struct ieee80211_local *local) DEBUGFS_ADD(hwflags); DEBUGFS_ADD(user_power); DEBUGFS_ADD(power); + DEBUGFS_ADD_MODE(aqm, 0600); statsd = debugfs_create_dir("statistics", phyd); -- cgit v0.10.2 From 5caa328e3811b7cfa33fd02c93280ffa622deb0e Mon Sep 17 00:00:00 2001 From: Michal Kazior Date: Thu, 19 May 2016 10:37:51 +0200 Subject: mac80211: implement codel on fair queuing flows There is no other limit other than a global packet count limit when using software queuing. This means a single flow queue can grow insanely long. This is particularly bad for TCP congestion algorithms which requires a little more sophisticated frame dropping scheme than a mere headdrop on limit overflow. Hence apply (a slighly modified, to fit the knobs) CoDel5 on flow queues. This improves TCP convergence and stability when combined with wireless driver which keeps its own tx queue/fifo at a minimum fill level for given link conditions. Signed-off-by: Michal Kazior Signed-off-by: Johannes Berg diff --git a/include/net/mac80211.h b/include/net/mac80211.h index a8683ae..a52009f 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -21,6 +21,7 @@ #include #include #include +#include #include /** @@ -895,7 +896,18 @@ struct ieee80211_tx_info { unsigned long jiffies; }; /* NB: vif can be NULL for injected frames */ - struct ieee80211_vif *vif; + union { + /* NB: vif can be NULL for injected frames */ + struct ieee80211_vif *vif; + + /* When packets are enqueued on txq it's easy + * to re-construct the vif pointer. There's no + * more space in tx_info so it can be used to + * store the necessary enqueue time for packet + * sojourn time computation. + */ + codel_time_t enqueue_time; + }; struct ieee80211_key_conf *hw_key; u32 flags; /* 4 bytes free */ diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 6f8375f..54edfb6 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -812,10 +812,12 @@ enum txq_info_flags { * @tin: contains packets split into multiple flows * @def_flow: used as a fallback flow when a packet destined to @tin hashes to * a fq_flow which is already owned by a different tin + * @def_cvars: codel vars for @def_flow */ struct txq_info { struct fq_tin tin; struct fq_flow def_flow; + struct codel_vars def_cvars; unsigned long flags; /* keep last! */ @@ -1108,6 +1110,9 @@ struct ieee80211_local { struct ieee80211_hw hw; struct fq fq; + struct codel_vars *cvars; + struct codel_params cparams; + struct codel_stats cstats; const struct ieee80211_ops *ops; diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 1d8343f..44ec605 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -24,6 +24,8 @@ #include #include #include +#include +#include #include #include @@ -1267,11 +1269,92 @@ static struct txq_info *ieee80211_get_txq(struct ieee80211_local *local, return to_txq_info(txq); } +static void ieee80211_set_skb_enqueue_time(struct sk_buff *skb) +{ + IEEE80211_SKB_CB(skb)->control.enqueue_time = codel_get_time(); +} + +static void ieee80211_set_skb_vif(struct sk_buff *skb, struct txq_info *txqi) +{ + IEEE80211_SKB_CB(skb)->control.vif = txqi->txq.vif; +} + +static u32 codel_skb_len_func(const struct sk_buff *skb) +{ + return skb->len; +} + +static codel_time_t codel_skb_time_func(const struct sk_buff *skb) +{ + const struct ieee80211_tx_info *info; + + info = (const struct ieee80211_tx_info *)skb->cb; + return info->control.enqueue_time; +} + +static struct sk_buff *codel_dequeue_func(struct codel_vars *cvars, + void *ctx) +{ + struct ieee80211_local *local; + struct txq_info *txqi; + struct fq *fq; + struct fq_flow *flow; + + txqi = ctx; + local = vif_to_sdata(txqi->txq.vif)->local; + fq = &local->fq; + + if (cvars == &txqi->def_cvars) + flow = &txqi->def_flow; + else + flow = &fq->flows[cvars - local->cvars]; + + return fq_flow_dequeue(fq, flow); +} + +static void codel_drop_func(struct sk_buff *skb, + void *ctx) +{ + struct ieee80211_local *local; + struct ieee80211_hw *hw; + struct txq_info *txqi; + + txqi = ctx; + local = vif_to_sdata(txqi->txq.vif)->local; + hw = &local->hw; + + ieee80211_free_txskb(hw, skb); +} + static struct sk_buff *fq_tin_dequeue_func(struct fq *fq, struct fq_tin *tin, struct fq_flow *flow) { - return fq_flow_dequeue(fq, flow); + struct ieee80211_local *local; + struct txq_info *txqi; + struct codel_vars *cvars; + struct codel_params *cparams; + struct codel_stats *cstats; + + local = container_of(fq, struct ieee80211_local, fq); + txqi = container_of(tin, struct txq_info, tin); + cparams = &local->cparams; + cstats = &local->cstats; + + if (flow == &txqi->def_flow) + cvars = &txqi->def_cvars; + else + cvars = &local->cvars[flow - fq->flows]; + + return codel_dequeue(txqi, + &flow->backlog, + cparams, + cvars, + cstats, + codel_skb_len_func, + codel_skb_time_func, + codel_drop_func, + codel_dequeue_func); } static void fq_skb_free_func(struct fq *fq, @@ -1303,6 +1386,7 @@ static void ieee80211_txq_enqueue(struct ieee80211_local *local, struct fq *fq = &local->fq; struct fq_tin *tin = &txqi->tin; + ieee80211_set_skb_enqueue_time(skb); fq_tin_enqueue(fq, tin, skb, fq_skb_free_func, fq_flow_get_default_func); @@ -1314,6 +1398,7 @@ void ieee80211_txq_init(struct ieee80211_sub_if_data *sdata, { fq_tin_init(&txqi->tin); fq_flow_init(&txqi->def_flow); + codel_vars_init(&txqi->def_cvars); txqi->txq.vif = &sdata->vif; @@ -1342,6 +1427,7 @@ int ieee80211_txq_setup_flows(struct ieee80211_local *local) { struct fq *fq = &local->fq; int ret; + int i; if (!local->ops->wake_tx_queue) return 0; @@ -1350,6 +1436,22 @@ int ieee80211_txq_setup_flows(struct ieee80211_local *local) if (ret) return ret; + codel_params_init(&local->cparams); + codel_stats_init(&local->cstats); + local->cparams.interval = MS2TIME(100); + local->cparams.target = MS2TIME(20); + local->cparams.ecn = true; + + local->cvars = kcalloc(fq->flows_cnt, sizeof(local->cvars[0]), + GFP_KERNEL); + if (!local->cvars) { + fq_reset(fq, fq_skb_free_func); + return -ENOMEM; + } + + for (i = 0; i < fq->flows_cnt; i++) + codel_vars_init(&local->cvars[i]); + return 0; } @@ -1360,6 +1462,9 @@ void ieee80211_txq_teardown_flows(struct ieee80211_local *local) if (!local->ops->wake_tx_queue) return; + kfree(local->cvars); + local->cvars = NULL; + fq_reset(fq, fq_skb_free_func); } @@ -1382,6 +1487,8 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, if (!skb) goto out; + ieee80211_set_skb_vif(skb, txqi); + hdr = (struct ieee80211_hdr *)skb->data; if (txq->sta && ieee80211_is_data_qos(hdr->frame_control)) { struct sta_info *sta = container_of(txq->sta, struct sta_info, -- cgit v0.10.2 From d02fb8f14b2d1779870744ff0ab435141491bf5f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Thu, 26 May 2016 01:44:27 +0200 Subject: brcmfmac: rework function picking free BSS index MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The old implementation was overcomplicated and slightly bugged in some corner cases. Consider following state of BSS-es (limited to 6 for simplification): drvr->iflist[0]: { bsscfgidx:0, ndev->name:wlan1, } drvr->iflist[1]: (null) drvr->iflist[2]: { bsscfgidx:2, ndev->name:wlan1-1, } drvr->iflist[3]: { bsscfgidx:3, ndev->name:wlan1-2, } drvr->iflist[4]: (null) drvr->iflist[5]: (null) In such case the next AP interface should bsscfgidx 4 (we don't use 1 as it's reserved for P2P). With old code the loop iterations were following: [ifidx = 0] [bsscfgidx = 2] [highest = 2] [ifidx = 1] [bsscfgidx = 2] [highest = 2] available = true [ifidx = 2] [bsscfgidx = 2] [highest = 2] bsscfgidx = highest + 1 [ifidx = 3] [bsscfgidx = 3] [highest = 2] bsscfgidx = highest + 1 [ifidx = 4] [bsscfgidx = 3] [highest = 2] available = true [ifidx = 5] [bsscfgidx = 3] [highest = 2] available = true There were 2 obvious problems: 1) Having empty BSS at index 1 was resulting in available being always set to true, even if we would run out of BSS-es. 2) Calculated bsscfgidx was invalid (3 instead of 4) resulting in driver not being able to create the 4th AP interface. New code is simpler, placed in file where it's really used, handles running out of free BSS-es and allows using 4 interfaces at the same time. It also looks for the first free BSS instead of one after the last in use. It works well with current driver (which doesn't allow deleting interfaces) and should be future proof (if we ever allow deleting). Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index d0631b6..87819b3 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -541,6 +541,21 @@ brcmf_cfg80211_update_proto_addr_mode(struct wireless_dev *wdev) ADDR_INDIRECT); } +static int brcmf_get_first_free_bsscfgidx(struct brcmf_pub *drvr) +{ + int bsscfgidx; + + for (bsscfgidx = 0; bsscfgidx < BRCMF_MAX_IFS; bsscfgidx++) { + /* bsscfgidx 1 is reserved for legacy P2P */ + if (bsscfgidx == 1) + continue; + if (!drvr->iflist[bsscfgidx]) + return bsscfgidx; + } + + return -ENOMEM; +} + static int brcmf_cfg80211_request_ap_if(struct brcmf_if *ifp) { struct brcmf_mbss_ssid_le mbss_ssid_le; @@ -548,7 +563,7 @@ static int brcmf_cfg80211_request_ap_if(struct brcmf_if *ifp) int err; memset(&mbss_ssid_le, 0, sizeof(mbss_ssid_le)); - bsscfgidx = brcmf_get_next_free_bsscfgidx(ifp->drvr); + bsscfgidx = brcmf_get_first_free_bsscfgidx(ifp->drvr); if (bsscfgidx < 0) return bsscfgidx; diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c index b590499..6a76480 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c @@ -753,30 +753,6 @@ void brcmf_remove_interface(struct brcmf_if *ifp) brcmf_del_if(ifp->drvr, ifp->bsscfgidx); } -int brcmf_get_next_free_bsscfgidx(struct brcmf_pub *drvr) -{ - int ifidx; - int bsscfgidx; - bool available; - int highest; - - available = false; - bsscfgidx = 2; - highest = 2; - for (ifidx = 0; ifidx < BRCMF_MAX_IFS; ifidx++) { - if (drvr->iflist[ifidx]) { - if (drvr->iflist[ifidx]->bsscfgidx == bsscfgidx) - bsscfgidx = highest + 1; - else if (drvr->iflist[ifidx]->bsscfgidx > highest) - highest = drvr->iflist[ifidx]->bsscfgidx; - } else { - available = true; - } - } - - return available ? bsscfgidx : -ENOMEM; -} - #ifdef CONFIG_INET #define ARPOL_MAX_ENTRIES 8 static int brcmf_inetaddr_changed(struct notifier_block *nb, diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h index 647d3cc..2a075c5 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h @@ -217,7 +217,6 @@ int brcmf_net_attach(struct brcmf_if *ifp, bool rtnl_locked); struct brcmf_if *brcmf_add_if(struct brcmf_pub *drvr, s32 bsscfgidx, s32 ifidx, bool is_p2pdev, char *name, u8 *mac_addr); void brcmf_remove_interface(struct brcmf_if *ifp); -int brcmf_get_next_free_bsscfgidx(struct brcmf_pub *drvr); void brcmf_txflowblock_if(struct brcmf_if *ifp, enum brcmf_netif_stop_reason reason, bool state); void brcmf_txfinalize(struct brcmf_if *ifp, struct sk_buff *txp, bool success); -- cgit v0.10.2 From adba931fbc825efca7c821f0d76baed0a8dc9189 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 9 Jun 2016 18:03:34 +0900 Subject: sit: remove unnecessary protocol check in ipip6_tunnel_xmit() ipip6_tunnel_xmit() is called immediately after checking that skb->protocol is htons(ETH_P_IPV6) so there is no need to check it a second time. Found by inspection. Signed-off-by: Simon Horman Reviewed-by: Dinan Gunawardena Signed-off-by: David S. Miller diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 0a5a255..d9f2bd6 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -825,9 +825,6 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb, u8 protocol = IPPROTO_IPV6; int t_hlen = tunnel->hlen + sizeof(struct iphdr); - if (skb->protocol != htons(ETH_P_IPV6)) - goto tx_error; - if (tos == 1) tos = ipv6_get_dsfield(iph6); -- cgit v0.10.2 From 5362855aba7159aab8f7c6573eb675d9da317914 Mon Sep 17 00:00:00 2001 From: Vitaly Kuznetsov Date: Thu, 9 Jun 2016 12:44:03 +0200 Subject: netvsc: get rid of completion timeouts I'm hitting 5 second timeout in rndis_filter_set_rss_param() while setting RSS parameters for the device. When this happens we end up returning -ETIMEDOUT from the function and rndis_filter_device_add() falls back to setting net_device->max_chn = 1; net_device->num_chn = 1; net_device->num_sc_offered = 0; but after a moment the rndis request succeeds and subchannels start to appear. netvsc_sc_open() does unconditional nvscdev->num_sc_offered-- and it becomes U32_MAX-1. Consequent rndis_filter_device_remove() will hang while waiting for all U32_MAX-1 subchannels to appear and this is not going to happen. The immediate issue could be solved by adding num_sc_offered > 0 check to netvsc_sc_open() but we're getting out of sync with the host and it's not easy to adjust things later, e.g. in this particular case we'll be creating queues without a user request for it and races are expected. Same applies to other parts of the driver which have the same completion timeout. Following the trend in drivers/hv/* code I suggest we remove all these timeouts completely. As a guest we can always trust the host we're running on and if the host screws things up there is no easy way to recover anyway. Signed-off-by: Vitaly Kuznetsov Acked-by: Haiyang Zhang Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index 96f00c0..6909c32 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -244,7 +244,6 @@ static int netvsc_destroy_buf(struct hv_device *device) static int netvsc_init_buf(struct hv_device *device) { int ret = 0; - unsigned long t; struct netvsc_device *net_device; struct nvsp_message *init_packet; struct net_device *ndev; @@ -305,9 +304,7 @@ static int netvsc_init_buf(struct hv_device *device) goto cleanup; } - t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ); - BUG_ON(t == 0); - + wait_for_completion(&net_device->channel_init_wait); /* Check the response */ if (init_packet->msg.v1_msg. @@ -390,8 +387,7 @@ static int netvsc_init_buf(struct hv_device *device) goto cleanup; } - t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ); - BUG_ON(t == 0); + wait_for_completion(&net_device->channel_init_wait); /* Check the response */ if (init_packet->msg.v1_msg. @@ -445,7 +441,6 @@ static int negotiate_nvsp_ver(struct hv_device *device, { struct net_device *ndev = hv_get_drvdata(device); int ret; - unsigned long t; memset(init_packet, 0, sizeof(struct nvsp_message)); init_packet->hdr.msg_type = NVSP_MSG_TYPE_INIT; @@ -462,10 +457,7 @@ static int negotiate_nvsp_ver(struct hv_device *device, if (ret != 0) return ret; - t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ); - - if (t == 0) - return -ETIMEDOUT; + wait_for_completion(&net_device->channel_init_wait); if (init_packet->msg.init_msg.init_complete.status != NVSP_STAT_SUCCESS) diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c index 979fa44..8e830f7 100644 --- a/drivers/net/hyperv/rndis_filter.c +++ b/drivers/net/hyperv/rndis_filter.c @@ -466,7 +466,6 @@ static int rndis_filter_query_device(struct rndis_device *dev, u32 oid, struct rndis_query_request *query; struct rndis_query_complete *query_complete; int ret = 0; - unsigned long t; if (!result) return -EINVAL; @@ -503,11 +502,7 @@ static int rndis_filter_query_device(struct rndis_device *dev, u32 oid, if (ret != 0) goto cleanup; - t = wait_for_completion_timeout(&request->wait_event, 5*HZ); - if (t == 0) { - ret = -ETIMEDOUT; - goto cleanup; - } + wait_for_completion(&request->wait_event); /* Copy the response back */ query_complete = &request->response_msg.msg.query_complete; @@ -556,7 +551,6 @@ int rndis_filter_set_device_mac(struct net_device *ndev, char *mac) u32 extlen = sizeof(struct rndis_config_parameter_info) + 2*NWADR_STRLEN + 4*ETH_ALEN; int ret; - unsigned long t; request = get_rndis_request(rdev, RNDIS_MSG_SET, RNDIS_MESSAGE_SIZE(struct rndis_set_request) + extlen); @@ -597,21 +591,13 @@ int rndis_filter_set_device_mac(struct net_device *ndev, char *mac) if (ret != 0) goto cleanup; - t = wait_for_completion_timeout(&request->wait_event, 5*HZ); - if (t == 0) { - netdev_err(ndev, "timeout before we got a set response...\n"); - /* - * can't put_rndis_request, since we may still receive a - * send-completion. - */ - return -EBUSY; - } else { - set_complete = &request->response_msg.msg.set_complete; - if (set_complete->status != RNDIS_STATUS_SUCCESS) { - netdev_err(ndev, "Fail to set MAC on host side:0x%x\n", - set_complete->status); - ret = -EINVAL; - } + wait_for_completion(&request->wait_event); + + set_complete = &request->response_msg.msg.set_complete; + if (set_complete->status != RNDIS_STATUS_SUCCESS) { + netdev_err(ndev, "Fail to set MAC on host side:0x%x\n", + set_complete->status); + ret = -EINVAL; } cleanup: @@ -631,7 +617,6 @@ rndis_filter_set_offload_params(struct net_device *ndev, struct rndis_set_complete *set_complete; u32 extlen = sizeof(struct ndis_offload_params); int ret; - unsigned long t; u32 vsp_version = nvdev->nvsp_version; if (vsp_version <= NVSP_PROTOCOL_VERSION_4) { @@ -665,20 +650,12 @@ rndis_filter_set_offload_params(struct net_device *ndev, if (ret != 0) goto cleanup; - t = wait_for_completion_timeout(&request->wait_event, 5*HZ); - if (t == 0) { - netdev_err(ndev, "timeout before we got aOFFLOAD set response...\n"); - /* can't put_rndis_request, since we may still receive a - * send-completion. - */ - return -EBUSY; - } else { - set_complete = &request->response_msg.msg.set_complete; - if (set_complete->status != RNDIS_STATUS_SUCCESS) { - netdev_err(ndev, "Fail to set offload on host side:0x%x\n", - set_complete->status); - ret = -EINVAL; - } + wait_for_completion(&request->wait_event); + set_complete = &request->response_msg.msg.set_complete; + if (set_complete->status != RNDIS_STATUS_SUCCESS) { + netdev_err(ndev, "Fail to set offload on host side:0x%x\n", + set_complete->status); + ret = -EINVAL; } cleanup: @@ -706,7 +683,6 @@ static int rndis_filter_set_rss_param(struct rndis_device *rdev, int num_queue) u32 *itab; u8 *keyp; int i, ret; - unsigned long t; request = get_rndis_request( rdev, RNDIS_MSG_SET, @@ -749,20 +725,12 @@ static int rndis_filter_set_rss_param(struct rndis_device *rdev, int num_queue) if (ret != 0) goto cleanup; - t = wait_for_completion_timeout(&request->wait_event, 5*HZ); - if (t == 0) { - netdev_err(ndev, "timeout before we got a set response...\n"); - /* can't put_rndis_request, since we may still receive a - * send-completion. - */ - return -ETIMEDOUT; - } else { - set_complete = &request->response_msg.msg.set_complete; - if (set_complete->status != RNDIS_STATUS_SUCCESS) { - netdev_err(ndev, "Fail to set RSS parameters:0x%x\n", - set_complete->status); - ret = -EINVAL; - } + wait_for_completion(&request->wait_event); + set_complete = &request->response_msg.msg.set_complete; + if (set_complete->status != RNDIS_STATUS_SUCCESS) { + netdev_err(ndev, "Fail to set RSS parameters:0x%x\n", + set_complete->status); + ret = -EINVAL; } cleanup: @@ -791,8 +759,6 @@ int rndis_filter_set_packet_filter(struct rndis_device *dev, u32 new_filter) struct rndis_set_complete *set_complete; u32 status; int ret; - unsigned long t; - struct net_device *ndev = dev->ndev; request = get_rndis_request(dev, RNDIS_MSG_SET, RNDIS_MESSAGE_SIZE(struct rndis_set_request) + @@ -815,26 +781,14 @@ int rndis_filter_set_packet_filter(struct rndis_device *dev, u32 new_filter) if (ret != 0) goto cleanup; - t = wait_for_completion_timeout(&request->wait_event, 5*HZ); + wait_for_completion(&request->wait_event); - if (t == 0) { - netdev_err(ndev, - "timeout before we got a set response...\n"); - ret = -ETIMEDOUT; - /* - * We can't deallocate the request since we may still receive a - * send completion for it. - */ - goto exit; - } else { - set_complete = &request->response_msg.msg.set_complete; - status = set_complete->status; - } + set_complete = &request->response_msg.msg.set_complete; + status = set_complete->status; cleanup: if (request) put_rndis_request(dev, request); -exit: return ret; } @@ -846,7 +800,6 @@ static int rndis_filter_init_device(struct rndis_device *dev) struct rndis_initialize_complete *init_complete; u32 status; int ret; - unsigned long t; struct netvsc_device *nvdev = net_device_to_netvsc_device(dev->ndev); request = get_rndis_request(dev, RNDIS_MSG_INIT, @@ -870,12 +823,7 @@ static int rndis_filter_init_device(struct rndis_device *dev) goto cleanup; } - t = wait_for_completion_timeout(&request->wait_event, 5*HZ); - - if (t == 0) { - ret = -ETIMEDOUT; - goto cleanup; - } + wait_for_completion(&request->wait_event); init_complete = &request->response_msg.msg.init_complete; status = init_complete->status; @@ -1008,7 +956,6 @@ int rndis_filter_device_add(struct hv_device *dev, struct netvsc_device_info *device_info = additional_info; struct ndis_offload_params offloads; struct nvsp_message *init_packet; - unsigned long t; struct ndis_recv_scale_cap rsscap; u32 rsscap_size = sizeof(struct ndis_recv_scale_cap); u32 mtu, size; @@ -1151,11 +1098,8 @@ int rndis_filter_device_add(struct hv_device *dev, VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED); if (ret) goto out; - t = wait_for_completion_timeout(&net_device->channel_init_wait, 5*HZ); - if (t == 0) { - ret = -ETIMEDOUT; - goto out; - } + wait_for_completion(&net_device->channel_init_wait); + if (init_packet->msg.v5_msg.subchn_comp.status != NVSP_STAT_SUCCESS) { ret = -ENODEV; @@ -1192,17 +1136,12 @@ void rndis_filter_device_remove(struct hv_device *dev) { struct netvsc_device *net_dev = hv_device_to_netvsc_device(dev); struct rndis_device *rndis_dev = net_dev->extension; - unsigned long t; /* If not all subchannel offers are complete, wait for them until * completion to avoid race. */ - while (net_dev->num_sc_offered > 0) { - t = wait_for_completion_timeout(&net_dev->channel_init_wait, - 10 * HZ); - if (t == 0) - WARN(1, "Netvsc: Waiting for sub-channel processing"); - } + if (net_dev->num_sc_offered > 0) + wait_for_completion(&net_dev->channel_init_wait); /* Halt and release the rndis device */ rndis_filter_halt_device(rndis_dev); -- cgit v0.10.2 From 52fbb2907988aa0583c6d9d53a56aee090b2df7e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 9 Jun 2016 07:45:11 -0700 Subject: net: sched: fix qdisc->running lockdep annotations 1) qdisc_run_begin() is really using the equivalent of a trylock. Instead of using write_seqcount_begin(), use a combination of raw_write_seqcount_begin() and correct lockdep annotation. 2) sch_direct_xmit() should use regular spin_lock(root_lock) Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount") Signed-off-by: Eric Dumazet Reported-by: David Ahern Signed-off-by: David S. Miller diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 49534e2..a4c0f16 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -97,7 +97,11 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc) { if (qdisc_is_running(qdisc)) return false; - write_seqcount_begin(&qdisc->running); + /* Variant of write_seqcount_begin() telling lockdep a trylock + * was attempted. + */ + raw_write_seqcount_begin(&qdisc->running); + seqcount_acquire(&qdisc->running.dep_map, 0, 1, _RET_IP_); return true; } diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index cebea73..cad810b 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -137,10 +137,10 @@ int sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q, HARD_TX_UNLOCK(dev, txq); } else { - spin_lock_nested(root_lock, SINGLE_DEPTH_NESTING); + spin_lock(root_lock); return qdisc_qlen(q); } - spin_lock_nested(root_lock, SINGLE_DEPTH_NESTING); + spin_lock(root_lock); if (dev_xmit_complete(ret)) { /* Driver sent out skb successfully or skb was consumed */ -- cgit v0.10.2 From d3fff6c443fe8f8a5ef2bdcea45e2ff39db948c7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 9 Jun 2016 07:45:12 -0700 Subject: net: add netdev_lockdep_set_classes() helper It is time to add netdev_lockdep_set_classes() helper so that lockdep annotations per device type are easier to manage. This removes a lot of copies and missing annotations. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 681af31..90157e2 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4607,28 +4607,6 @@ static int bond_check_params(struct bond_params *params) return 0; } -static struct lock_class_key bonding_netdev_xmit_lock_key; -static struct lock_class_key bonding_netdev_addr_lock_key; -static struct lock_class_key bonding_tx_busylock_key; -static struct lock_class_key bonding_qdisc_running_key; - -static void bond_set_lockdep_class_one(struct net_device *dev, - struct netdev_queue *txq, - void *_unused) -{ - lockdep_set_class(&txq->_xmit_lock, - &bonding_netdev_xmit_lock_key); -} - -static void bond_set_lockdep_class(struct net_device *dev) -{ - lockdep_set_class(&dev->addr_list_lock, - &bonding_netdev_addr_lock_key); - netdev_for_each_tx_queue(dev, bond_set_lockdep_class_one, NULL); - dev->qdisc_tx_busylock = &bonding_tx_busylock_key; - dev->qdisc_running_key = &bonding_qdisc_running_key; -} - /* Called from registration process */ static int bond_init(struct net_device *bond_dev) { @@ -4641,7 +4619,7 @@ static int bond_init(struct net_device *bond_dev) if (!bond->wq) return -ENOMEM; - bond_set_lockdep_class(bond_dev); + netdev_lockdep_set_classes(bond_dev); list_add_tail(&bond->bond_list, &bn->dev_list); diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c index aeabaa4..17953ab 100644 --- a/drivers/net/ppp/ppp_generic.c +++ b/drivers/net/ppp/ppp_generic.c @@ -1312,13 +1312,9 @@ ppp_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats64) return stats64; } -static struct lock_class_key ppp_tx_busylock; -static struct lock_class_key ppp_qdisc_running_key; - static int ppp_dev_init(struct net_device *dev) { - dev->qdisc_tx_busylock = &ppp_tx_busylock; - dev->qdisc_running_key = &ppp_qdisc_running_key; + netdev_lockdep_set_classes(dev); return 0; } diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index 00eb389..0a1bb83 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -1574,25 +1574,6 @@ static const struct team_option team_options[] = { }, }; -static struct lock_class_key team_netdev_xmit_lock_key; -static struct lock_class_key team_netdev_addr_lock_key; -static struct lock_class_key team_tx_busylock_key; -static struct lock_class_key team_qdisc_running_key; - -static void team_set_lockdep_class_one(struct net_device *dev, - struct netdev_queue *txq, - void *unused) -{ - lockdep_set_class(&txq->_xmit_lock, &team_netdev_xmit_lock_key); -} - -static void team_set_lockdep_class(struct net_device *dev) -{ - lockdep_set_class(&dev->addr_list_lock, &team_netdev_addr_lock_key); - netdev_for_each_tx_queue(dev, team_set_lockdep_class_one, NULL); - dev->qdisc_tx_busylock = &team_tx_busylock_key; - dev->qdisc_running_key = &team_qdisc_running_key; -} static int team_init(struct net_device *dev) { @@ -1628,7 +1609,7 @@ static int team_init(struct net_device *dev) goto err_options_register; netif_carrier_off(dev); - team_set_lockdep_class(dev); + netdev_lockdep_set_classes(dev); return 0; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 5415623..4f234b1 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1946,6 +1946,23 @@ static inline void netdev_for_each_tx_queue(struct net_device *dev, f(dev, &dev->_tx[i], arg); } +#define netdev_lockdep_set_classes(dev) \ +{ \ + static struct lock_class_key qdisc_tx_busylock_key; \ + static struct lock_class_key qdisc_running_key; \ + static struct lock_class_key qdisc_xmit_lock_key; \ + static struct lock_class_key dev_addr_list_lock_key; \ + unsigned int i; \ + \ + (dev)->qdisc_tx_busylock = &qdisc_tx_busylock_key; \ + (dev)->qdisc_running_key = &qdisc_running_key; \ + lockdep_set_class(&(dev)->addr_list_lock, \ + &dev_addr_list_lock_key); \ + for (i = 0; i < (dev)->num_tx_queues; i++) \ + lockdep_set_class(&(dev)->_tx[i]._xmit_lock, \ + &qdisc_xmit_lock_key); \ +} + struct netdev_queue *netdev_pick_tx(struct net_device *dev, struct sk_buff *skb, void *accel_priv); diff --git a/net/bluetooth/6lowpan.c b/net/bluetooth/6lowpan.c index 977a11e..d020299 100644 --- a/net/bluetooth/6lowpan.c +++ b/net/bluetooth/6lowpan.c @@ -627,22 +627,9 @@ static netdev_tx_t bt_xmit(struct sk_buff *skb, struct net_device *netdev) return err < 0 ? NET_XMIT_DROP : err; } -static struct lock_class_key bt_tx_busylock; -static struct lock_class_key bt_netdev_xmit_lock_key; -static struct lock_class_key bt_qdisc_running_key; - -static void bt_set_lockdep_class_one(struct net_device *dev, - struct netdev_queue *txq, - void *_unused) -{ - lockdep_set_class(&txq->_xmit_lock, &bt_netdev_xmit_lock_key); -} - static int bt_dev_init(struct net_device *dev) { - netdev_for_each_tx_queue(dev, bt_set_lockdep_class_one, NULL); - dev->qdisc_tx_busylock = &bt_tx_busylock; - dev->qdisc_running_key = &bt_qdisc_running_key; + netdev_lockdep_set_classes(dev); return 0; } diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index 14aa5ef..4e2b308 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -58,23 +58,9 @@ static struct header_ops lowpan_header_ops = { .create = lowpan_header_create, }; -static struct lock_class_key lowpan_tx_busylock; -static struct lock_class_key lowpan_netdev_xmit_lock_key; -static struct lock_class_key lowpan_qdisc_running_key; - -static void lowpan_set_lockdep_class_one(struct net_device *ldev, - struct netdev_queue *txq, - void *_unused) -{ - lockdep_set_class(&txq->_xmit_lock, - &lowpan_netdev_xmit_lock_key); -} - static int lowpan_dev_init(struct net_device *ldev) { - netdev_for_each_tx_queue(ldev, lowpan_set_lockdep_class_one, NULL); - ldev->qdisc_tx_busylock = &lowpan_tx_busylock; - ldev->qdisc_running_key = &lowpan_qdisc_running_key; + netdev_lockdep_set_classes(ldev); return 0; } diff --git a/net/l2tp/l2tp_eth.c b/net/l2tp/l2tp_eth.c index c00d72d..57fc5a4 100644 --- a/net/l2tp/l2tp_eth.c +++ b/net/l2tp/l2tp_eth.c @@ -67,9 +67,6 @@ static inline struct l2tp_eth_net *l2tp_eth_pernet(struct net *net) return net_generic(net, l2tp_eth_net_id); } -static struct lock_class_key l2tp_eth_tx_busylock; -static struct lock_class_key l2tp_qdisc_running_key; - static int l2tp_eth_dev_init(struct net_device *dev) { struct l2tp_eth *priv = netdev_priv(dev); @@ -77,8 +74,7 @@ static int l2tp_eth_dev_init(struct net_device *dev) priv->dev = dev; eth_hw_addr_random(dev); eth_broadcast_addr(dev->broadcast); - dev->qdisc_tx_busylock = &l2tp_eth_tx_busylock; - dev->qdisc_running_key = &l2tp_qdisc_running_key; + netdev_lockdep_set_classes(dev); return 0; } -- cgit v0.10.2 From 78e7a2ae872770e5f26bf794b97b7b8e3e30ccfd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 9 Jun 2016 07:45:13 -0700 Subject: net: vrf: call netdev_lockdep_set_classes() In case a qdisc is used on a vrf device, we need to use different lockdep classes to avoid false positives. Use the new netdev_lockdep_set_classes() generic helper. Reported-by: David Ahern Signed-off-by: Eric Dumazet Tested-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index e3e693b..b82e3527 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -661,7 +661,7 @@ static int vrf_dev_init(struct net_device *dev) /* similarly, oper state is irrelevant; set to up to avoid confusion */ dev->operstate = IF_OPER_UP; - + netdev_lockdep_set_classes(dev); return 0; out_rth: -- cgit v0.10.2 From 24ffd752007fb04e8efb560029c6076f6f3c0c5e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 9 Jun 2016 07:45:14 -0700 Subject: net: macvlan: call netdev_lockdep_set_classes() In case a qdisc is used on a macvlan device, we need to use different lockdep classes to avoid false positives. Use the new netdev_lockdep_set_classes() generic helper. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 0c65bd9..cd9b538 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -788,7 +788,6 @@ static int macvlan_change_mtu(struct net_device *dev, int new_mtu) * "super class" of normal network devices; split their locks off into a * separate class since they always nest. */ -static struct lock_class_key macvlan_netdev_xmit_lock_key; static struct lock_class_key macvlan_netdev_addr_lock_key; #define ALWAYS_ON_FEATURES \ @@ -809,20 +808,12 @@ static int macvlan_get_nest_level(struct net_device *dev) return ((struct macvlan_dev *)netdev_priv(dev))->nest_level; } -static void macvlan_set_lockdep_class_one(struct net_device *dev, - struct netdev_queue *txq, - void *_unused) -{ - lockdep_set_class(&txq->_xmit_lock, - &macvlan_netdev_xmit_lock_key); -} - static void macvlan_set_lockdep_class(struct net_device *dev) { + netdev_lockdep_set_classes(dev); lockdep_set_class_and_subclass(&dev->addr_list_lock, &macvlan_netdev_addr_lock_key, macvlan_get_nest_level(dev)); - netdev_for_each_tx_queue(dev, macvlan_set_lockdep_class_one, NULL); } static int macvlan_init(struct net_device *dev) -- cgit v0.10.2 From 0d7dd798fd89d986a6c59030b704827a4e6a13b4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 9 Jun 2016 07:45:15 -0700 Subject: net: ipvlan: call netdev_lockdep_set_classes() In case a qdisc is used on a ipvlan device, we need to use different lockdep classes to avoid false positives. Use the new netdev_lockdep_set_classes() generic helper. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c index 1c4d395..18b4e8c 100644 --- a/drivers/net/ipvlan/ipvlan_main.c +++ b/drivers/net/ipvlan/ipvlan_main.c @@ -80,13 +80,6 @@ static void ipvlan_port_destroy(struct net_device *dev) kfree_rcu(port, rcu); } -/* ipvlan network devices have devices nesting below it and are a special - * "super class" of normal network devices; split their locks off into a - * separate class since they always nest. - */ -static struct lock_class_key ipvlan_netdev_xmit_lock_key; -static struct lock_class_key ipvlan_netdev_addr_lock_key; - #define IPVLAN_FEATURES \ (NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST | \ NETIF_F_GSO | NETIF_F_TSO | NETIF_F_UFO | NETIF_F_GSO_ROBUST | \ @@ -96,19 +89,6 @@ static struct lock_class_key ipvlan_netdev_addr_lock_key; #define IPVLAN_STATE_MASK \ ((1<<__LINK_STATE_NOCARRIER) | (1<<__LINK_STATE_DORMANT)) -static void ipvlan_set_lockdep_class_one(struct net_device *dev, - struct netdev_queue *txq, - void *_unused) -{ - lockdep_set_class(&txq->_xmit_lock, &ipvlan_netdev_xmit_lock_key); -} - -static void ipvlan_set_lockdep_class(struct net_device *dev) -{ - lockdep_set_class(&dev->addr_list_lock, &ipvlan_netdev_addr_lock_key); - netdev_for_each_tx_queue(dev, ipvlan_set_lockdep_class_one, NULL); -} - static int ipvlan_init(struct net_device *dev) { struct ipvl_dev *ipvlan = netdev_priv(dev); @@ -123,7 +103,7 @@ static int ipvlan_init(struct net_device *dev) dev->gso_max_segs = phy_dev->gso_max_segs; dev->hard_header_len = phy_dev->hard_header_len; - ipvlan_set_lockdep_class(dev); + netdev_lockdep_set_classes(dev); ipvlan->pcpu_stats = alloc_percpu(struct ipvl_pcpu_stats); if (!ipvlan->pcpu_stats) -- cgit v0.10.2 From a468ef452ab91907215e29e98cfadff8cb1c15f7 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 9 Jun 2016 17:42:05 -0700 Subject: net: dsa: bcm_sf2: Split fast age into a helper function Add a helper function to fast age something that is controlled by the caller: port, VLAN. We will use this to implement a VLAN fast age operation. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index d662578..ad22cab 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -461,17 +461,11 @@ static int bcm_sf2_sw_set_eee(struct dsa_switch *ds, int port, return 0; } -/* Fast-ageing of ARL entries for a given port, equivalent to an ARL - * flush for that port. - */ -static int bcm_sf2_sw_fast_age_port(struct dsa_switch *ds, int port) +static int bcm_sf2_fast_age_op(struct bcm_sf2_priv *priv) { - struct bcm_sf2_priv *priv = ds_to_priv(ds); unsigned int timeout = 1000; u32 reg; - core_writel(priv, port, CORE_FAST_AGE_PORT); - reg = core_readl(priv, CORE_FAST_AGE_CTRL); reg |= EN_AGE_PORT | EN_AGE_DYNAMIC | FAST_AGE_STR_DONE; core_writel(priv, reg, CORE_FAST_AGE_CTRL); @@ -492,6 +486,18 @@ static int bcm_sf2_sw_fast_age_port(struct dsa_switch *ds, int port) return 0; } +/* Fast-ageing of ARL entries for a given port, equivalent to an ARL + * flush for that port. + */ +static int bcm_sf2_sw_fast_age_port(struct dsa_switch *ds, int port) +{ + struct bcm_sf2_priv *priv = ds_to_priv(ds); + + core_writel(priv, port, CORE_FAST_AGE_PORT); + + return bcm_sf2_fast_age_op(priv); +} + static int bcm_sf2_sw_br_join(struct dsa_switch *ds, int port, struct net_device *bridge) { -- cgit v0.10.2 From 7fbb1a92ef928bd04e357422bb27b24a1c25fb42 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 9 Jun 2016 17:42:06 -0700 Subject: net: dsa: bcm_sf2: Move setup function at the far end Re-order the bcm_sf2_sw_setup() function so that it is at the far end of the driver to avoid any kind of forward declarations. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index ad22cab..d726f59 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -1065,125 +1065,6 @@ static void bcm_sf2_mdio_unregister(struct bcm_sf2_priv *priv) of_node_put(priv->master_mii_dn); } -static int bcm_sf2_sw_setup(struct dsa_switch *ds) -{ - const char *reg_names[BCM_SF2_REGS_NUM] = BCM_SF2_REGS_NAME; - struct bcm_sf2_priv *priv = ds_to_priv(ds); - struct device_node *dn; - void __iomem **base; - unsigned int port; - unsigned int i; - u32 reg, rev; - int ret; - - spin_lock_init(&priv->indir_lock); - mutex_init(&priv->stats_mutex); - - /* All the interesting properties are at the parent device_node - * level - */ - dn = ds->cd->of_node->parent; - bcm_sf2_identify_ports(priv, ds->cd->of_node); - - priv->irq0 = irq_of_parse_and_map(dn, 0); - priv->irq1 = irq_of_parse_and_map(dn, 1); - - base = &priv->core; - for (i = 0; i < BCM_SF2_REGS_NUM; i++) { - *base = of_iomap(dn, i); - if (*base == NULL) { - pr_err("unable to find register: %s\n", reg_names[i]); - ret = -ENOMEM; - goto out_unmap; - } - base++; - } - - ret = bcm_sf2_sw_rst(priv); - if (ret) { - pr_err("unable to software reset switch: %d\n", ret); - goto out_unmap; - } - - ret = bcm_sf2_mdio_register(ds); - if (ret) { - pr_err("failed to register MDIO bus\n"); - goto out_unmap; - } - - /* Disable all interrupts and request them */ - bcm_sf2_intr_disable(priv); - - ret = request_irq(priv->irq0, bcm_sf2_switch_0_isr, 0, - "switch_0", priv); - if (ret < 0) { - pr_err("failed to request switch_0 IRQ\n"); - goto out_unmap; - } - - ret = request_irq(priv->irq1, bcm_sf2_switch_1_isr, 0, - "switch_1", priv); - if (ret < 0) { - pr_err("failed to request switch_1 IRQ\n"); - goto out_free_irq0; - } - - /* Reset the MIB counters */ - reg = core_readl(priv, CORE_GMNCFGCFG); - reg |= RST_MIB_CNT; - core_writel(priv, reg, CORE_GMNCFGCFG); - reg &= ~RST_MIB_CNT; - core_writel(priv, reg, CORE_GMNCFGCFG); - - /* Get the maximum number of ports for this switch */ - priv->hw_params.num_ports = core_readl(priv, CORE_IMP0_PRT_ID) + 1; - if (priv->hw_params.num_ports > DSA_MAX_PORTS) - priv->hw_params.num_ports = DSA_MAX_PORTS; - - /* Assume a single GPHY setup if we can't read that property */ - if (of_property_read_u32(dn, "brcm,num-gphy", - &priv->hw_params.num_gphy)) - priv->hw_params.num_gphy = 1; - - /* Enable all valid ports and disable those unused */ - for (port = 0; port < priv->hw_params.num_ports; port++) { - /* IMP port receives special treatment */ - if ((1 << port) & ds->enabled_port_mask) - bcm_sf2_port_setup(ds, port, NULL); - else if (dsa_is_cpu_port(ds, port)) - bcm_sf2_imp_setup(ds, port); - else - bcm_sf2_port_disable(ds, port, NULL); - } - - rev = reg_readl(priv, REG_SWITCH_REVISION); - priv->hw_params.top_rev = (rev >> SWITCH_TOP_REV_SHIFT) & - SWITCH_TOP_REV_MASK; - priv->hw_params.core_rev = (rev & SF2_REV_MASK); - - rev = reg_readl(priv, REG_PHY_REVISION); - priv->hw_params.gphy_rev = rev & PHY_REVISION_MASK; - - pr_info("Starfighter 2 top: %x.%02x, core: %x.%02x base: 0x%p, IRQs: %d, %d\n", - priv->hw_params.top_rev >> 8, priv->hw_params.top_rev & 0xff, - priv->hw_params.core_rev >> 8, priv->hw_params.core_rev & 0xff, - priv->core, priv->irq0, priv->irq1); - - return 0; - -out_free_irq0: - free_irq(priv->irq0, priv); -out_unmap: - base = &priv->core; - for (i = 0; i < BCM_SF2_REGS_NUM; i++) { - if (*base) - iounmap(*base); - base++; - } - bcm_sf2_mdio_unregister(priv); - return ret; -} - static int bcm_sf2_sw_set_addr(struct dsa_switch *ds, u8 *addr) { return 0; @@ -1431,6 +1312,125 @@ static int bcm_sf2_sw_set_wol(struct dsa_switch *ds, int port, return p->ethtool_ops->set_wol(p, wol); } +static int bcm_sf2_sw_setup(struct dsa_switch *ds) +{ + const char *reg_names[BCM_SF2_REGS_NUM] = BCM_SF2_REGS_NAME; + struct bcm_sf2_priv *priv = ds_to_priv(ds); + struct device_node *dn; + void __iomem **base; + unsigned int port; + unsigned int i; + u32 reg, rev; + int ret; + + spin_lock_init(&priv->indir_lock); + mutex_init(&priv->stats_mutex); + + /* All the interesting properties are at the parent device_node + * level + */ + dn = ds->cd->of_node->parent; + bcm_sf2_identify_ports(priv, ds->cd->of_node); + + priv->irq0 = irq_of_parse_and_map(dn, 0); + priv->irq1 = irq_of_parse_and_map(dn, 1); + + base = &priv->core; + for (i = 0; i < BCM_SF2_REGS_NUM; i++) { + *base = of_iomap(dn, i); + if (*base == NULL) { + pr_err("unable to find register: %s\n", reg_names[i]); + ret = -ENOMEM; + goto out_unmap; + } + base++; + } + + ret = bcm_sf2_sw_rst(priv); + if (ret) { + pr_err("unable to software reset switch: %d\n", ret); + goto out_unmap; + } + + ret = bcm_sf2_mdio_register(ds); + if (ret) { + pr_err("failed to register MDIO bus\n"); + goto out_unmap; + } + + /* Disable all interrupts and request them */ + bcm_sf2_intr_disable(priv); + + ret = request_irq(priv->irq0, bcm_sf2_switch_0_isr, 0, + "switch_0", priv); + if (ret < 0) { + pr_err("failed to request switch_0 IRQ\n"); + goto out_unmap; + } + + ret = request_irq(priv->irq1, bcm_sf2_switch_1_isr, 0, + "switch_1", priv); + if (ret < 0) { + pr_err("failed to request switch_1 IRQ\n"); + goto out_free_irq0; + } + + /* Reset the MIB counters */ + reg = core_readl(priv, CORE_GMNCFGCFG); + reg |= RST_MIB_CNT; + core_writel(priv, reg, CORE_GMNCFGCFG); + reg &= ~RST_MIB_CNT; + core_writel(priv, reg, CORE_GMNCFGCFG); + + /* Get the maximum number of ports for this switch */ + priv->hw_params.num_ports = core_readl(priv, CORE_IMP0_PRT_ID) + 1; + if (priv->hw_params.num_ports > DSA_MAX_PORTS) + priv->hw_params.num_ports = DSA_MAX_PORTS; + + /* Assume a single GPHY setup if we can't read that property */ + if (of_property_read_u32(dn, "brcm,num-gphy", + &priv->hw_params.num_gphy)) + priv->hw_params.num_gphy = 1; + + /* Enable all valid ports and disable those unused */ + for (port = 0; port < priv->hw_params.num_ports; port++) { + /* IMP port receives special treatment */ + if ((1 << port) & ds->enabled_port_mask) + bcm_sf2_port_setup(ds, port, NULL); + else if (dsa_is_cpu_port(ds, port)) + bcm_sf2_imp_setup(ds, port); + else + bcm_sf2_port_disable(ds, port, NULL); + } + + rev = reg_readl(priv, REG_SWITCH_REVISION); + priv->hw_params.top_rev = (rev >> SWITCH_TOP_REV_SHIFT) & + SWITCH_TOP_REV_MASK; + priv->hw_params.core_rev = (rev & SF2_REV_MASK); + + rev = reg_readl(priv, REG_PHY_REVISION); + priv->hw_params.gphy_rev = rev & PHY_REVISION_MASK; + + pr_info("Starfighter 2 top: %x.%02x, core: %x.%02x base: 0x%p, IRQs: %d, %d\n", + priv->hw_params.top_rev >> 8, priv->hw_params.top_rev & 0xff, + priv->hw_params.core_rev >> 8, priv->hw_params.core_rev & 0xff, + priv->core, priv->irq0, priv->irq1); + + return 0; + +out_free_irq0: + free_irq(priv->irq0, priv); +out_unmap: + base = &priv->core; + for (i = 0; i < BCM_SF2_REGS_NUM; i++) { + if (*base) + iounmap(*base); + base++; + } + bcm_sf2_mdio_unregister(priv); + return ret; +} + static struct dsa_switch_driver bcm_sf2_switch_driver = { .tag_protocol = DSA_TAG_PROTO_BRCM, .probe = bcm_sf2_sw_drv_probe, -- cgit v0.10.2 From 064523ff786093d81ae967959196c723d30f3da5 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 9 Jun 2016 17:42:07 -0700 Subject: net: dsa: bcm_sf2: Add VLAN registers definitions Add the definitions for the VLAN registers that we are going to manipulate in subsequent patches. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/bcm_sf2_regs.h b/drivers/net/dsa/bcm_sf2_regs.h index 97780d4..9f2a9cb 100644 --- a/drivers/net/dsa/bcm_sf2_regs.h +++ b/drivers/net/dsa/bcm_sf2_regs.h @@ -274,6 +274,23 @@ #define CORE_ARLA_SRCH_RSLT_MACVID(x) (CORE_ARLA_SRCH_RSLT_0_MACVID + ((x) * 0x40)) #define CORE_ARLA_SRCH_RSLT(x) (CORE_ARLA_SRCH_RSLT_0 + ((x) * 0x40)) +#define CORE_ARLA_VTBL_RWCTRL 0x1600 +#define ARLA_VTBL_CMD_WRITE 0 +#define ARLA_VTBL_CMD_READ 1 +#define ARLA_VTBL_CMD_CLEAR 2 +#define ARLA_VTBL_STDN (1 << 7) + +#define CORE_ARLA_VTBL_ADDR 0x1604 +#define VTBL_ADDR_INDEX_MASK 0xfff + +#define CORE_ARLA_VTBL_ENTRY 0x160c +#define FWD_MAP_MASK 0x1ff +#define UNTAG_MAP_MASK 0x1ff +#define UNTAG_MAP_SHIFT 9 +#define MSTP_INDEX_MASK 0x7 +#define MSTP_INDEX_SHIFT 18 +#define FWD_MODE (1 << 21) + #define CORE_MEM_PSM_VDD_CTRL 0x2380 #define P_TXQ_PSM_VDD_SHIFT 2 #define P_TXQ_PSM_VDD_MASK 0x3 @@ -287,6 +304,59 @@ #define CORE_PORT_VLAN_CTL_PORT(x) (0xc400 + ((x) * 0x8)) #define PORT_VLAN_CTRL_MASK 0x1ff +#define CORE_VLAN_CTRL0 0xd000 +#define CHANGE_1P_VID_INNER (1 << 0) +#define CHANGE_1P_VID_OUTER (1 << 1) +#define CHANGE_1Q_VID (1 << 3) +#define VLAN_LEARN_MODE_SVL (0 << 5) +#define VLAN_LEARN_MODE_IVL (3 << 5) +#define VLAN_EN (1 << 7) + +#define CORE_VLAN_CTRL1 0xd004 +#define EN_RSV_MCAST_FWDMAP (1 << 2) +#define EN_RSV_MCAST_UNTAG (1 << 3) +#define EN_IPMC_BYPASS_FWDMAP (1 << 5) +#define EN_IPMC_BYPASS_UNTAG (1 << 6) + +#define CORE_VLAN_CTRL2 0xd008 +#define EN_MIIM_BYPASS_V_FWDMAP (1 << 2) +#define EN_GMRP_GVRP_V_FWDMAP (1 << 5) +#define EN_GMRP_GVRP_UNTAG_MAP (1 << 6) + +#define CORE_VLAN_CTRL3 0xd00c +#define EN_DROP_NON1Q_MASK 0x1ff + +#define CORE_VLAN_CTRL4 0xd014 +#define RESV_MCAST_FLOOD (1 << 1) +#define EN_DOUBLE_TAG_MASK 0x3 +#define EN_DOUBLE_TAG_SHIFT 2 +#define EN_MGE_REV_GMRP (1 << 4) +#define EN_MGE_REV_GVRP (1 << 5) +#define INGR_VID_CHK_SHIFT 6 +#define INGR_VID_CHK_MASK 0x3 +#define INGR_VID_CHK_FWD (0 << INGR_VID_CHK_SHIFT) +#define INGR_VID_CHK_DROP (1 << INGR_VID_CHK_SHIFT) +#define INGR_VID_CHK_NO_CHK (2 << INGR_VID_CHK_SHIFT) +#define INGR_VID_CHK_VID_VIOL_IMP (3 << INGR_VID_CHK_SHIFT) + +#define CORE_VLAN_CTRL5 0xd018 +#define EN_CPU_RX_BYP_INNER_CRCCHCK (1 << 0) +#define EN_VID_FFF_FWD (1 << 2) +#define DROP_VTABLE_MISS (1 << 3) +#define EGRESS_DIR_FRM_BYP_TRUNK_EN (1 << 4) +#define PRESV_NON1Q (1 << 6) + +#define CORE_VLAN_CTRL6 0xd01c +#define STRICT_SFD_DETECT (1 << 0) +#define DIS_ARL_BUST_LMIT (1 << 4) + +#define CORE_DEFAULT_1Q_TAG_P(x) (0xd040 + ((x) * 8)) +#define CFI_SHIFT 12 +#define PRI_SHIFT 13 +#define PRI_MASK 0x7 + +#define CORE_JOIN_ALL_VLAN_EN 0xd140 + #define CORE_EEE_EN_CTRL 0x24800 #define CORE_EEE_LPI_INDICATE 0x24810 -- cgit v0.10.2 From 9c57a77182c89e1bf773008f904f4a2e9ea30bd5 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 9 Jun 2016 17:42:08 -0700 Subject: net: dsa: bcm_sf2: Add VLAN support Add support for configuring VLANs on the Broadcom Starfigther2 switch. This is all done through the bridge vlan facility just like other DSA drivers. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index d726f59..cd1d630 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -467,7 +467,7 @@ static int bcm_sf2_fast_age_op(struct bcm_sf2_priv *priv) u32 reg; reg = core_readl(priv, CORE_FAST_AGE_CTRL); - reg |= EN_AGE_PORT | EN_AGE_DYNAMIC | FAST_AGE_STR_DONE; + reg |= EN_AGE_PORT | EN_AGE_VLAN | EN_AGE_DYNAMIC | FAST_AGE_STR_DONE; core_writel(priv, reg, CORE_FAST_AGE_CTRL); do { @@ -498,13 +498,86 @@ static int bcm_sf2_sw_fast_age_port(struct dsa_switch *ds, int port) return bcm_sf2_fast_age_op(priv); } +static int bcm_sf2_sw_fast_age_vlan(struct bcm_sf2_priv *priv, u16 vid) +{ + core_writel(priv, vid, CORE_FAST_AGE_VID); + + return bcm_sf2_fast_age_op(priv); +} + +static int bcm_sf2_vlan_op_wait(struct bcm_sf2_priv *priv) +{ + unsigned int timeout = 10; + u32 reg; + + do { + reg = core_readl(priv, CORE_ARLA_VTBL_RWCTRL); + if (!(reg & ARLA_VTBL_STDN)) + return 0; + + usleep_range(1000, 2000); + } while (timeout--); + + return -ETIMEDOUT; +} + +static int bcm_sf2_vlan_op(struct bcm_sf2_priv *priv, u8 op) +{ + core_writel(priv, ARLA_VTBL_STDN | op, CORE_ARLA_VTBL_RWCTRL); + + return bcm_sf2_vlan_op_wait(priv); +} + +static void bcm_sf2_set_vlan_entry(struct bcm_sf2_priv *priv, u16 vid, + struct bcm_sf2_vlan *vlan) +{ + int ret; + + core_writel(priv, vid & VTBL_ADDR_INDEX_MASK, CORE_ARLA_VTBL_ADDR); + core_writel(priv, vlan->untag << UNTAG_MAP_SHIFT | vlan->members, + CORE_ARLA_VTBL_ENTRY); + + ret = bcm_sf2_vlan_op(priv, ARLA_VTBL_CMD_WRITE); + if (ret) + pr_err("failed to write VLAN entry\n"); +} + +static int bcm_sf2_get_vlan_entry(struct bcm_sf2_priv *priv, u16 vid, + struct bcm_sf2_vlan *vlan) +{ + u32 entry; + int ret; + + core_writel(priv, vid & VTBL_ADDR_INDEX_MASK, CORE_ARLA_VTBL_ADDR); + + ret = bcm_sf2_vlan_op(priv, ARLA_VTBL_CMD_READ); + if (ret) + return ret; + + entry = core_readl(priv, CORE_ARLA_VTBL_ENTRY); + vlan->members = entry & FWD_MAP_MASK; + vlan->untag = (entry >> UNTAG_MAP_SHIFT) & UNTAG_MAP_MASK; + + return 0; +} + static int bcm_sf2_sw_br_join(struct dsa_switch *ds, int port, struct net_device *bridge) { struct bcm_sf2_priv *priv = ds_to_priv(ds); + s8 cpu_port = ds->dst->cpu_port; unsigned int i; u32 reg, p_ctl; + /* Make this port leave the all VLANs join since we will have proper + * VLAN entries from now on + */ + reg = core_readl(priv, CORE_JOIN_ALL_VLAN_EN); + reg &= ~BIT(port); + if ((reg & BIT(cpu_port)) == BIT(cpu_port)) + reg &= ~BIT(cpu_port); + core_writel(priv, reg, CORE_JOIN_ALL_VLAN_EN); + priv->port_sts[port].bridge_dev = bridge; p_ctl = core_readl(priv, CORE_PORT_VLAN_CTL_PORT(port)); @@ -536,6 +609,7 @@ static void bcm_sf2_sw_br_leave(struct dsa_switch *ds, int port) { struct bcm_sf2_priv *priv = ds_to_priv(ds); struct net_device *bridge = priv->port_sts[port].bridge_dev; + s8 cpu_port = ds->dst->cpu_port; unsigned int i; u32 reg, p_ctl; @@ -559,6 +633,13 @@ static void bcm_sf2_sw_br_leave(struct dsa_switch *ds, int port) core_writel(priv, p_ctl, CORE_PORT_VLAN_CTL_PORT(port)); priv->port_sts[port].vlan_ctl_mask = p_ctl; priv->port_sts[port].bridge_dev = NULL; + + /* Make this port join all VLANs without VLAN entries */ + reg = core_readl(priv, CORE_JOIN_ALL_VLAN_EN); + reg |= BIT(port); + if (!(reg & BIT(cpu_port))) + reg |= BIT(cpu_port); + core_writel(priv, reg, CORE_JOIN_ALL_VLAN_EN); } static void bcm_sf2_sw_br_set_stp_state(struct dsa_switch *ds, int port, @@ -1312,6 +1393,182 @@ static int bcm_sf2_sw_set_wol(struct dsa_switch *ds, int port, return p->ethtool_ops->set_wol(p, wol); } +static void bcm_sf2_enable_vlan(struct bcm_sf2_priv *priv, bool enable) +{ + u32 mgmt, vc0, vc1, vc4, vc5; + + mgmt = core_readl(priv, CORE_SWMODE); + vc0 = core_readl(priv, CORE_VLAN_CTRL0); + vc1 = core_readl(priv, CORE_VLAN_CTRL1); + vc4 = core_readl(priv, CORE_VLAN_CTRL4); + vc5 = core_readl(priv, CORE_VLAN_CTRL5); + + mgmt &= ~SW_FWDG_MODE; + + if (enable) { + vc0 |= VLAN_EN | VLAN_LEARN_MODE_IVL; + vc1 |= EN_RSV_MCAST_UNTAG | EN_RSV_MCAST_FWDMAP; + vc4 &= ~(INGR_VID_CHK_MASK << INGR_VID_CHK_SHIFT); + vc4 |= INGR_VID_CHK_DROP; + vc5 |= DROP_VTABLE_MISS | EN_VID_FFF_FWD; + } else { + vc0 &= ~(VLAN_EN | VLAN_LEARN_MODE_IVL); + vc1 &= ~(EN_RSV_MCAST_UNTAG | EN_RSV_MCAST_FWDMAP); + vc4 &= ~(INGR_VID_CHK_MASK << INGR_VID_CHK_SHIFT); + vc5 &= ~(DROP_VTABLE_MISS | EN_VID_FFF_FWD); + vc4 |= INGR_VID_CHK_VID_VIOL_IMP; + } + + core_writel(priv, vc0, CORE_VLAN_CTRL0); + core_writel(priv, vc1, CORE_VLAN_CTRL1); + core_writel(priv, 0, CORE_VLAN_CTRL3); + core_writel(priv, vc4, CORE_VLAN_CTRL4); + core_writel(priv, vc5, CORE_VLAN_CTRL5); + core_writel(priv, mgmt, CORE_SWMODE); +} + +static void bcm_sf2_sw_configure_vlan(struct dsa_switch *ds) +{ + struct bcm_sf2_priv *priv = ds_to_priv(ds); + unsigned int port; + + /* Clear all VLANs */ + bcm_sf2_vlan_op(priv, ARLA_VTBL_CMD_CLEAR); + + for (port = 0; port < priv->hw_params.num_ports; port++) { + if (!((1 << port) & ds->enabled_port_mask)) + continue; + + core_writel(priv, 1, CORE_DEFAULT_1Q_TAG_P(port)); + } +} + +static int bcm_sf2_sw_vlan_filtering(struct dsa_switch *ds, int port, + bool vlan_filtering) +{ + return 0; +} + +static int bcm_sf2_sw_vlan_prepare(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans) +{ + struct bcm_sf2_priv *priv = ds_to_priv(ds); + + bcm_sf2_enable_vlan(priv, true); + + return 0; +} + +static void bcm_sf2_sw_vlan_add(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans) +{ + struct bcm_sf2_priv *priv = ds_to_priv(ds); + bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; + bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; + s8 cpu_port = ds->dst->cpu_port; + struct bcm_sf2_vlan *vl; + u16 vid; + + for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) { + vl = &priv->vlans[vid]; + + bcm_sf2_get_vlan_entry(priv, vid, vl); + + vl->members |= BIT(port) | BIT(cpu_port); + if (untagged) + vl->untag |= BIT(port) | BIT(cpu_port); + else + vl->untag &= ~(BIT(port) | BIT(cpu_port)); + + bcm_sf2_set_vlan_entry(priv, vid, vl); + bcm_sf2_sw_fast_age_vlan(priv, vid); + } + + if (pvid) { + core_writel(priv, vlan->vid_end, CORE_DEFAULT_1Q_TAG_P(port)); + core_writel(priv, vlan->vid_end, + CORE_DEFAULT_1Q_TAG_P(cpu_port)); + bcm_sf2_sw_fast_age_vlan(priv, vid); + } +} + +static int bcm_sf2_sw_vlan_del(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan) +{ + struct bcm_sf2_priv *priv = ds_to_priv(ds); + bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; + s8 cpu_port = ds->dst->cpu_port; + struct bcm_sf2_vlan *vl; + u16 vid, pvid; + int ret; + + pvid = core_readl(priv, CORE_DEFAULT_1Q_TAG_P(port)); + + for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) { + vl = &priv->vlans[vid]; + + ret = bcm_sf2_get_vlan_entry(priv, vid, vl); + if (ret) + return ret; + + vl->members &= ~BIT(port); + if ((vl->members & BIT(cpu_port)) == BIT(cpu_port)) + vl->members = 0; + if (pvid == vid) + pvid = 0; + if (untagged) { + vl->untag &= ~BIT(port); + if ((vl->untag & BIT(port)) == BIT(cpu_port)) + vl->untag = 0; + } + + bcm_sf2_set_vlan_entry(priv, vid, vl); + bcm_sf2_sw_fast_age_vlan(priv, vid); + } + + core_writel(priv, pvid, CORE_DEFAULT_1Q_TAG_P(port)); + core_writel(priv, pvid, CORE_DEFAULT_1Q_TAG_P(cpu_port)); + bcm_sf2_sw_fast_age_vlan(priv, vid); + + return 0; +} + +static int bcm_sf2_sw_vlan_dump(struct dsa_switch *ds, int port, + struct switchdev_obj_port_vlan *vlan, + int (*cb)(struct switchdev_obj *obj)) +{ + struct bcm_sf2_priv *priv = ds_to_priv(ds); + struct bcm_sf2_port_status *p = &priv->port_sts[port]; + struct bcm_sf2_vlan *vl; + u16 vid, pvid; + int err = 0; + + pvid = core_readl(priv, CORE_DEFAULT_1Q_TAG_P(port)); + + for (vid = 0; vid < VLAN_N_VID; vid++) { + vl = &priv->vlans[vid]; + + if (!(vl->members & BIT(port))) + continue; + + vlan->vid_begin = vlan->vid_end = vid; + vlan->flags = 0; + + if (vl->untag & BIT(port)) + vlan->flags |= BRIDGE_VLAN_INFO_UNTAGGED; + if (p->pvid == vid) + vlan->flags |= BRIDGE_VLAN_INFO_PVID; + + err = cb(&vlan->obj); + if (err) + break; + } + + return err; +} + static int bcm_sf2_sw_setup(struct dsa_switch *ds) { const char *reg_names[BCM_SF2_REGS_NUM] = BCM_SF2_REGS_NAME; @@ -1403,6 +1660,8 @@ static int bcm_sf2_sw_setup(struct dsa_switch *ds) bcm_sf2_port_disable(ds, port, NULL); } + bcm_sf2_sw_configure_vlan(ds); + rev = reg_readl(priv, REG_SWITCH_REVISION); priv->hw_params.top_rev = (rev >> SWITCH_TOP_REV_SHIFT) & SWITCH_TOP_REV_MASK; @@ -1457,6 +1716,11 @@ static struct dsa_switch_driver bcm_sf2_switch_driver = { .port_fdb_add = bcm_sf2_sw_fdb_add, .port_fdb_del = bcm_sf2_sw_fdb_del, .port_fdb_dump = bcm_sf2_sw_fdb_dump, + .port_vlan_filtering = bcm_sf2_sw_vlan_filtering, + .port_vlan_prepare = bcm_sf2_sw_vlan_prepare, + .port_vlan_add = bcm_sf2_sw_vlan_add, + .port_vlan_del = bcm_sf2_sw_vlan_del, + .port_vlan_dump = bcm_sf2_sw_vlan_dump, }; static int __init bcm_sf2_init(void) diff --git a/drivers/net/dsa/bcm_sf2.h b/drivers/net/dsa/bcm_sf2.h index bde11eb..463bed8 100644 --- a/drivers/net/dsa/bcm_sf2.h +++ b/drivers/net/dsa/bcm_sf2.h @@ -21,6 +21,7 @@ #include #include #include +#include #include @@ -50,6 +51,7 @@ struct bcm_sf2_port_status { struct ethtool_eee eee; u32 vlan_ctl_mask; + u16 pvid; struct net_device *bridge_dev; }; @@ -63,6 +65,11 @@ struct bcm_sf2_arl_entry { u8 is_static:1; }; +struct bcm_sf2_vlan { + u16 members; + u16 untag; +}; + static inline void bcm_sf2_mac_from_u64(u64 src, u8 *dst) { unsigned int i; @@ -148,6 +155,9 @@ struct bcm_sf2_priv { struct device_node *master_mii_dn; struct mii_bus *slave_mii_bus; struct mii_bus *master_mii_bus; + + /* Cache of programmed VLANs */ + struct bcm_sf2_vlan vlans[VLAN_N_VID]; }; struct bcm_sf2_hw_stats { -- cgit v0.10.2 From 967dd82ffc52e9d8ea0defde094f9a39a3f4eeed Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 9 Jun 2016 18:23:53 -0700 Subject: net: dsa: b53: Add support for Broadcom RoboSwitch This patch adds support for Broadcom's BCM53xx switch family, also known as RoboSwitch. Some of these switches are ubiquituous, found in home routers, Wi-Fi routers, DSL and cable modem gateways and other networking related products. This drivers adds the library driver (b53_common.c) as well as a few bus glue drivers for MDIO, SPI, Switch Register Access Block (SRAB) and memory-mapped I/O into a SoC's address space (Broadcom BCM63xx/33xx). Basic operations are supported to bring the Layer 1/2 up and running, but not much more at this point, subsequent patches add the remaining features. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/dsa/b53.txt b/Documentation/devicetree/bindings/net/dsa/b53.txt new file mode 100644 index 0000000..ca752db --- /dev/null +++ b/Documentation/devicetree/bindings/net/dsa/b53.txt @@ -0,0 +1,88 @@ +Broadcom BCM53xx Ethernet switches +================================== + +Required properties: + +- compatible: For external switch chips, compatible string must be exactly one + of: "brcm,bcm5325" + "brcm,bcm53115" + "brcm,bcm53125" + "brcm,bcm53128" + "brcm,bcm5365" + "brcm,bcm5395" + "brcm,bcm5397" + "brcm,bcm5398" + + For the BCM5310x SoCs with an integrated switch, must be one of: + "brcm,bcm53010-srab" + "brcm,bcm53011-srab" + "brcm,bcm53012-srab" + "brcm,bcm53018-srab" + "brcm,bcm53019-srab" and the mandatory "brcm,bcm5301x-srab" string + + For the BCM63xx/33xx SoCs with an integrated switch, must be one of: + "brcm,bcm3384-switch" + "brcm,bcm6328-switch" + "brcm,bcm6368-switch" and the mandatory "brcm,bcm63xx-switch" + +See Documentation/devicetree/bindings/dsa/dsa.txt for a list of additional +required and optional properties. + +Examples: + +Ethernet switch connected via MDIO to the host, CPU port wired to eth0: + + eth0: ethernet@10001000 { + compatible = "brcm,unimac"; + reg = <0x10001000 0x1000>; + + fixed-link { + speed = <1000>; + duplex-full; + }; + }; + + mdio0: mdio@10000000 { + compatible = "brcm,unimac-mdio"; + #address-cells = <1>; + #size-cells = <0>; + + switch0: ethernet-switch@30 { + compatible = "brcm,bcm53125"; + #address-cells = <1>; + #size-cells = <0>; + + ports { + port0@0 { + reg = <0>; + label = "lan1"; + }; + + port1@1 { + reg = <1>; + label = "lan2"; + }; + + port5@5 { + reg = <5>; + label = "cable-modem"; + fixed-link { + speed = <1000>; + duplex-full; + }; + phy-mode = "rgmii-txid"; + }; + + port8@8 { + reg = <8>; + label = "cpu"; + fixed-link { + speed = <1000>; + duplex-full; + }; + phy-mode = "rgmii-txid"; + ethernet = <ð0>; + }; + }; + }; + }; diff --git a/MAINTAINERS b/MAINTAINERS index 16e1500..b29a088 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2454,6 +2454,14 @@ L: netdev@vger.kernel.org S: Supported F: drivers/net/ethernet/broadcom/b44.* +BROADCOM B53 ETHERNET SWITCH DRIVER +M: Florian Fainelli +L: netdev@vger.kernel.org +L: openwrt-devel@lists.openwrt.org (subscribers-only) +S: Supported +F: drivers/net/dsa/b53/* +F: include/linux/platform_data/b53.h + BROADCOM GENET ETHERNET DRIVER M: Florian Fainelli L: netdev@vger.kernel.org diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig index 200663c..be481e1 100644 --- a/drivers/net/dsa/Kconfig +++ b/drivers/net/dsa/Kconfig @@ -28,4 +28,6 @@ config NET_DSA_BCM_SF2 This enables support for the Broadcom Starfighter 2 Ethernet switch chips. +source "drivers/net/dsa/b53/Kconfig" + endmenu diff --git a/drivers/net/dsa/Makefile b/drivers/net/dsa/Makefile index 76b751d..97bc70a 100644 --- a/drivers/net/dsa/Makefile +++ b/drivers/net/dsa/Makefile @@ -1,3 +1,5 @@ obj-$(CONFIG_NET_DSA_MV88E6060) += mv88e6060.o obj-$(CONFIG_NET_DSA_MV88E6XXX) += mv88e6xxx.o obj-$(CONFIG_NET_DSA_BCM_SF2) += bcm_sf2.o + +obj-y += b53/ diff --git a/drivers/net/dsa/b53/Kconfig b/drivers/net/dsa/b53/Kconfig new file mode 100644 index 0000000..27f32a5 --- /dev/null +++ b/drivers/net/dsa/b53/Kconfig @@ -0,0 +1,33 @@ +menuconfig B53 + tristate "Broadcom BCM53xx managed switch support" + depends on NET_DSA + help + This driver adds support for Broadcom managed switch chips. It supports + BCM5325E, BCM5365, BCM539x, BCM53115 and BCM53125 as well as BCM63XX + integrated switches. + +config B53_SPI_DRIVER + tristate "B53 SPI connected switch driver" + depends on B53 && SPI + help + Select to enable support for registering switches configured through SPI. + +config B53_MDIO_DRIVER + tristate "B53 MDIO connected switch driver" + depends on B53 + help + Select to enable support for registering switches configured through MDIO. + +config B53_MMAP_DRIVER + tristate "B53 MMAP connected switch driver" + depends on B53 && HAS_IOMEM + help + Select to enable support for memory-mapped switches like the BCM63XX + integrated switches. + +config B53_SRAB_DRIVER + tristate "B53 SRAB connected switch driver" + depends on B53 && HAS_IOMEM + help + Select to enable support for memory-mapped Switch Register Access + Bridge Registers (SRAB) like it is found on the BCM53010 diff --git a/drivers/net/dsa/b53/Makefile b/drivers/net/dsa/b53/Makefile new file mode 100644 index 0000000..7e6f9a8 --- /dev/null +++ b/drivers/net/dsa/b53/Makefile @@ -0,0 +1,6 @@ +obj-$(CONFIG_B53) += b53_common.o + +obj-$(CONFIG_B53_SPI_DRIVER) += b53_spi.o +obj-$(CONFIG_B53_MDIO_DRIVER) += b53_mdio.o +obj-$(CONFIG_B53_MMAP_DRIVER) += b53_mmap.o +obj-$(CONFIG_B53_SRAB_DRIVER) += b53_srab.o diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c new file mode 100644 index 0000000..6f0337d --- /dev/null +++ b/drivers/net/dsa/b53/b53_common.c @@ -0,0 +1,1158 @@ +/* + * B53 switch driver main logic + * + * Copyright (C) 2011-2013 Jonas Gorski + * Copyright (C) 2016 Florian Fainelli + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "b53_regs.h" +#include "b53_priv.h" + +struct b53_mib_desc { + u8 size; + u8 offset; + const char *name; +}; + +/* BCM5365 MIB counters */ +static const struct b53_mib_desc b53_mibs_65[] = { + { 8, 0x00, "TxOctets" }, + { 4, 0x08, "TxDropPkts" }, + { 4, 0x10, "TxBroadcastPkts" }, + { 4, 0x14, "TxMulticastPkts" }, + { 4, 0x18, "TxUnicastPkts" }, + { 4, 0x1c, "TxCollisions" }, + { 4, 0x20, "TxSingleCollision" }, + { 4, 0x24, "TxMultipleCollision" }, + { 4, 0x28, "TxDeferredTransmit" }, + { 4, 0x2c, "TxLateCollision" }, + { 4, 0x30, "TxExcessiveCollision" }, + { 4, 0x38, "TxPausePkts" }, + { 8, 0x44, "RxOctets" }, + { 4, 0x4c, "RxUndersizePkts" }, + { 4, 0x50, "RxPausePkts" }, + { 4, 0x54, "Pkts64Octets" }, + { 4, 0x58, "Pkts65to127Octets" }, + { 4, 0x5c, "Pkts128to255Octets" }, + { 4, 0x60, "Pkts256to511Octets" }, + { 4, 0x64, "Pkts512to1023Octets" }, + { 4, 0x68, "Pkts1024to1522Octets" }, + { 4, 0x6c, "RxOversizePkts" }, + { 4, 0x70, "RxJabbers" }, + { 4, 0x74, "RxAlignmentErrors" }, + { 4, 0x78, "RxFCSErrors" }, + { 8, 0x7c, "RxGoodOctets" }, + { 4, 0x84, "RxDropPkts" }, + { 4, 0x88, "RxUnicastPkts" }, + { 4, 0x8c, "RxMulticastPkts" }, + { 4, 0x90, "RxBroadcastPkts" }, + { 4, 0x94, "RxSAChanges" }, + { 4, 0x98, "RxFragments" }, +}; + +#define B53_MIBS_65_SIZE ARRAY_SIZE(b53_mibs_65) + +/* BCM63xx MIB counters */ +static const struct b53_mib_desc b53_mibs_63xx[] = { + { 8, 0x00, "TxOctets" }, + { 4, 0x08, "TxDropPkts" }, + { 4, 0x0c, "TxQoSPkts" }, + { 4, 0x10, "TxBroadcastPkts" }, + { 4, 0x14, "TxMulticastPkts" }, + { 4, 0x18, "TxUnicastPkts" }, + { 4, 0x1c, "TxCollisions" }, + { 4, 0x20, "TxSingleCollision" }, + { 4, 0x24, "TxMultipleCollision" }, + { 4, 0x28, "TxDeferredTransmit" }, + { 4, 0x2c, "TxLateCollision" }, + { 4, 0x30, "TxExcessiveCollision" }, + { 4, 0x38, "TxPausePkts" }, + { 8, 0x3c, "TxQoSOctets" }, + { 8, 0x44, "RxOctets" }, + { 4, 0x4c, "RxUndersizePkts" }, + { 4, 0x50, "RxPausePkts" }, + { 4, 0x54, "Pkts64Octets" }, + { 4, 0x58, "Pkts65to127Octets" }, + { 4, 0x5c, "Pkts128to255Octets" }, + { 4, 0x60, "Pkts256to511Octets" }, + { 4, 0x64, "Pkts512to1023Octets" }, + { 4, 0x68, "Pkts1024to1522Octets" }, + { 4, 0x6c, "RxOversizePkts" }, + { 4, 0x70, "RxJabbers" }, + { 4, 0x74, "RxAlignmentErrors" }, + { 4, 0x78, "RxFCSErrors" }, + { 8, 0x7c, "RxGoodOctets" }, + { 4, 0x84, "RxDropPkts" }, + { 4, 0x88, "RxUnicastPkts" }, + { 4, 0x8c, "RxMulticastPkts" }, + { 4, 0x90, "RxBroadcastPkts" }, + { 4, 0x94, "RxSAChanges" }, + { 4, 0x98, "RxFragments" }, + { 4, 0xa0, "RxSymbolErrors" }, + { 4, 0xa4, "RxQoSPkts" }, + { 8, 0xa8, "RxQoSOctets" }, + { 4, 0xb0, "Pkts1523to2047Octets" }, + { 4, 0xb4, "Pkts2048to4095Octets" }, + { 4, 0xb8, "Pkts4096to8191Octets" }, + { 4, 0xbc, "Pkts8192to9728Octets" }, + { 4, 0xc0, "RxDiscarded" }, +}; + +#define B53_MIBS_63XX_SIZE ARRAY_SIZE(b53_mibs_63xx) + +/* MIB counters */ +static const struct b53_mib_desc b53_mibs[] = { + { 8, 0x00, "TxOctets" }, + { 4, 0x08, "TxDropPkts" }, + { 4, 0x10, "TxBroadcastPkts" }, + { 4, 0x14, "TxMulticastPkts" }, + { 4, 0x18, "TxUnicastPkts" }, + { 4, 0x1c, "TxCollisions" }, + { 4, 0x20, "TxSingleCollision" }, + { 4, 0x24, "TxMultipleCollision" }, + { 4, 0x28, "TxDeferredTransmit" }, + { 4, 0x2c, "TxLateCollision" }, + { 4, 0x30, "TxExcessiveCollision" }, + { 4, 0x38, "TxPausePkts" }, + { 8, 0x50, "RxOctets" }, + { 4, 0x58, "RxUndersizePkts" }, + { 4, 0x5c, "RxPausePkts" }, + { 4, 0x60, "Pkts64Octets" }, + { 4, 0x64, "Pkts65to127Octets" }, + { 4, 0x68, "Pkts128to255Octets" }, + { 4, 0x6c, "Pkts256to511Octets" }, + { 4, 0x70, "Pkts512to1023Octets" }, + { 4, 0x74, "Pkts1024to1522Octets" }, + { 4, 0x78, "RxOversizePkts" }, + { 4, 0x7c, "RxJabbers" }, + { 4, 0x80, "RxAlignmentErrors" }, + { 4, 0x84, "RxFCSErrors" }, + { 8, 0x88, "RxGoodOctets" }, + { 4, 0x90, "RxDropPkts" }, + { 4, 0x94, "RxUnicastPkts" }, + { 4, 0x98, "RxMulticastPkts" }, + { 4, 0x9c, "RxBroadcastPkts" }, + { 4, 0xa0, "RxSAChanges" }, + { 4, 0xa4, "RxFragments" }, + { 4, 0xa8, "RxJumboPkts" }, + { 4, 0xac, "RxSymbolErrors" }, + { 4, 0xc0, "RxDiscarded" }, +}; + +#define B53_MIBS_SIZE ARRAY_SIZE(b53_mibs) + +static int b53_do_vlan_op(struct b53_device *dev, u8 op) +{ + unsigned int i; + + b53_write8(dev, B53_ARLIO_PAGE, dev->vta_regs[0], VTA_START_CMD | op); + + for (i = 0; i < 10; i++) { + u8 vta; + + b53_read8(dev, B53_ARLIO_PAGE, dev->vta_regs[0], &vta); + if (!(vta & VTA_START_CMD)) + return 0; + + usleep_range(100, 200); + } + + return -EIO; +} + +static void b53_set_vlan_entry(struct b53_device *dev, u16 vid, u16 members, + u16 untag) +{ + if (is5325(dev)) { + u32 entry = 0; + + if (members) { + entry = ((untag & VA_UNTAG_MASK_25) << VA_UNTAG_S_25) | + members; + if (dev->core_rev >= 3) + entry |= VA_VALID_25_R4 | vid << VA_VID_HIGH_S; + else + entry |= VA_VALID_25; + } + + b53_write32(dev, B53_VLAN_PAGE, B53_VLAN_WRITE_25, entry); + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_25, vid | + VTA_RW_STATE_WR | VTA_RW_OP_EN); + } else if (is5365(dev)) { + u16 entry = 0; + + if (members) + entry = ((untag & VA_UNTAG_MASK_65) << VA_UNTAG_S_65) | + members | VA_VALID_65; + + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_WRITE_65, entry); + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_65, vid | + VTA_RW_STATE_WR | VTA_RW_OP_EN); + } else { + b53_write16(dev, B53_ARLIO_PAGE, dev->vta_regs[1], vid); + b53_write32(dev, B53_ARLIO_PAGE, dev->vta_regs[2], + (untag << VTE_UNTAG_S) | members); + + b53_do_vlan_op(dev, VTA_CMD_WRITE); + } +} + +void b53_set_forwarding(struct b53_device *dev, int enable) +{ + u8 mgmt; + + b53_read8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, &mgmt); + + if (enable) + mgmt |= SM_SW_FWD_EN; + else + mgmt &= ~SM_SW_FWD_EN; + + b53_write8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, mgmt); +} + +static void b53_enable_vlan(struct b53_device *dev, int enable) +{ + u8 mgmt, vc0, vc1, vc4 = 0, vc5; + + b53_read8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, &mgmt); + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL0, &vc0); + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL1, &vc1); + + if (is5325(dev) || is5365(dev)) { + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4_25, &vc4); + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5_25, &vc5); + } else if (is63xx(dev)) { + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4_63XX, &vc4); + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5_63XX, &vc5); + } else { + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4, &vc4); + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5, &vc5); + } + + mgmt &= ~SM_SW_FWD_MODE; + + if (enable) { + vc0 |= VC0_VLAN_EN | VC0_VID_CHK_EN | VC0_VID_HASH_VID; + vc1 |= VC1_RX_MCST_UNTAG_EN | VC1_RX_MCST_FWD_EN; + vc4 &= ~VC4_ING_VID_CHECK_MASK; + vc4 |= VC4_ING_VID_VIO_DROP << VC4_ING_VID_CHECK_S; + vc5 |= VC5_DROP_VTABLE_MISS; + + if (is5325(dev)) + vc0 &= ~VC0_RESERVED_1; + + if (is5325(dev) || is5365(dev)) + vc1 |= VC1_RX_MCST_TAG_EN; + + if (!is5325(dev) && !is5365(dev)) { + if (dev->allow_vid_4095) + vc5 |= VC5_VID_FFF_EN; + else + vc5 &= ~VC5_VID_FFF_EN; + } + } else { + vc0 &= ~(VC0_VLAN_EN | VC0_VID_CHK_EN | VC0_VID_HASH_VID); + vc1 &= ~(VC1_RX_MCST_UNTAG_EN | VC1_RX_MCST_FWD_EN); + vc4 &= ~VC4_ING_VID_CHECK_MASK; + vc5 &= ~VC5_DROP_VTABLE_MISS; + + if (is5325(dev) || is5365(dev)) + vc4 |= VC4_ING_VID_VIO_FWD << VC4_ING_VID_CHECK_S; + else + vc4 |= VC4_ING_VID_VIO_TO_IMP << VC4_ING_VID_CHECK_S; + + if (is5325(dev) || is5365(dev)) + vc1 &= ~VC1_RX_MCST_TAG_EN; + + if (!is5325(dev) && !is5365(dev)) + vc5 &= ~VC5_VID_FFF_EN; + } + + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL0, vc0); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL1, vc1); + + if (is5325(dev) || is5365(dev)) { + /* enable the high 8 bit vid check on 5325 */ + if (is5325(dev) && enable) + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL3, + VC3_HIGH_8BIT_EN); + else + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL3, 0); + + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4_25, vc4); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5_25, vc5); + } else if (is63xx(dev)) { + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_CTRL3_63XX, 0); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4_63XX, vc4); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5_63XX, vc5); + } else { + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_CTRL3, 0); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4, vc4); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5, vc5); + } + + b53_write8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, mgmt); +} + +static int b53_set_jumbo(struct b53_device *dev, bool enable, bool allow_10_100) +{ + u32 port_mask = 0; + u16 max_size = JMS_MIN_SIZE; + + if (is5325(dev) || is5365(dev)) + return -EINVAL; + + if (enable) { + port_mask = dev->enabled_ports; + max_size = JMS_MAX_SIZE; + if (allow_10_100) + port_mask |= JPM_10_100_JUMBO_EN; + } + + b53_write32(dev, B53_JUMBO_PAGE, dev->jumbo_pm_reg, port_mask); + return b53_write16(dev, B53_JUMBO_PAGE, dev->jumbo_size_reg, max_size); +} + +static int b53_flush_arl(struct b53_device *dev) +{ + unsigned int i; + + b53_write8(dev, B53_CTRL_PAGE, B53_FAST_AGE_CTRL, + FAST_AGE_DONE | FAST_AGE_DYNAMIC | FAST_AGE_STATIC); + + for (i = 0; i < 10; i++) { + u8 fast_age_ctrl; + + b53_read8(dev, B53_CTRL_PAGE, B53_FAST_AGE_CTRL, + &fast_age_ctrl); + + if (!(fast_age_ctrl & FAST_AGE_DONE)) + goto out; + + msleep(1); + } + + return -ETIMEDOUT; +out: + /* Only age dynamic entries (default behavior) */ + b53_write8(dev, B53_CTRL_PAGE, B53_FAST_AGE_CTRL, FAST_AGE_DYNAMIC); + return 0; +} + +static int b53_enable_port(struct dsa_switch *ds, int port, + struct phy_device *phy) +{ + struct b53_device *dev = ds_to_priv(ds); + + /* Clear the Rx and Tx disable bits and set to no spanning tree */ + b53_write8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), 0); + + return 0; +} + +static void b53_disable_port(struct dsa_switch *ds, int port, + struct phy_device *phy) +{ + struct b53_device *dev = ds_to_priv(ds); + u8 reg; + + /* Disable Tx/Rx for the port */ + b53_read8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), ®); + reg |= PORT_CTRL_RX_DISABLE | PORT_CTRL_TX_DISABLE; + b53_write8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), reg); +} + +static void b53_enable_cpu_port(struct b53_device *dev) +{ + unsigned int cpu_port = dev->cpu_port; + u8 port_ctrl; + + /* BCM5325 CPU port is at 8 */ + if ((is5325(dev) || is5365(dev)) && cpu_port == B53_CPU_PORT_25) + cpu_port = B53_CPU_PORT; + + port_ctrl = PORT_CTRL_RX_BCST_EN | + PORT_CTRL_RX_MCST_EN | + PORT_CTRL_RX_UCST_EN; + b53_write8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(cpu_port), port_ctrl); +} + +static void b53_enable_mib(struct b53_device *dev) +{ + u8 gc; + + b53_read8(dev, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, &gc); + gc &= ~(GC_RESET_MIB | GC_MIB_AC_EN); + b53_write8(dev, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, gc); +} + +static int b53_configure_vlan(struct b53_device *dev) +{ + int i; + + /* clear all vlan entries */ + if (is5325(dev) || is5365(dev)) { + for (i = 1; i < dev->num_vlans; i++) + b53_set_vlan_entry(dev, i, 0, 0); + } else { + b53_do_vlan_op(dev, VTA_CMD_CLEAR); + } + + b53_enable_vlan(dev, false); + + b53_for_each_port(dev, i) + b53_write16(dev, B53_VLAN_PAGE, + B53_VLAN_PORT_DEF_TAG(i), 1); + + if (!is5325(dev) && !is5365(dev)) + b53_set_jumbo(dev, dev->enable_jumbo, false); + + return 0; +} + +static void b53_switch_reset_gpio(struct b53_device *dev) +{ + int gpio = dev->reset_gpio; + + if (gpio < 0) + return; + + /* Reset sequence: RESET low(50ms)->high(20ms) + */ + gpio_set_value(gpio, 0); + mdelay(50); + + gpio_set_value(gpio, 1); + mdelay(20); + + dev->current_page = 0xff; +} + +static int b53_switch_reset(struct b53_device *dev) +{ + u8 mgmt; + + b53_switch_reset_gpio(dev); + + if (is539x(dev)) { + b53_write8(dev, B53_CTRL_PAGE, B53_SOFTRESET, 0x83); + b53_write8(dev, B53_CTRL_PAGE, B53_SOFTRESET, 0x00); + } + + b53_read8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, &mgmt); + + if (!(mgmt & SM_SW_FWD_EN)) { + mgmt &= ~SM_SW_FWD_MODE; + mgmt |= SM_SW_FWD_EN; + + b53_write8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, mgmt); + b53_read8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, &mgmt); + + if (!(mgmt & SM_SW_FWD_EN)) { + dev_err(dev->dev, "Failed to enable switch!\n"); + return -EINVAL; + } + } + + b53_enable_mib(dev); + + return b53_flush_arl(dev); +} + +static int b53_phy_read16(struct dsa_switch *ds, int addr, int reg) +{ + struct b53_device *priv = ds_to_priv(ds); + u16 value = 0; + int ret; + + if (priv->ops->phy_read16) + ret = priv->ops->phy_read16(priv, addr, reg, &value); + else + ret = b53_read16(priv, B53_PORT_MII_PAGE(addr), + reg * 2, &value); + + return ret ? ret : value; +} + +static int b53_phy_write16(struct dsa_switch *ds, int addr, int reg, u16 val) +{ + struct b53_device *priv = ds_to_priv(ds); + + if (priv->ops->phy_write16) + return priv->ops->phy_write16(priv, addr, reg, val); + + return b53_write16(priv, B53_PORT_MII_PAGE(addr), reg * 2, val); +} + +static int b53_reset_switch(struct b53_device *priv) +{ + /* reset vlans */ + priv->enable_jumbo = false; + + memset(priv->ports, 0, sizeof(*priv->ports) * priv->num_ports); + + return b53_switch_reset(priv); +} + +static int b53_apply_config(struct b53_device *priv) +{ + /* disable switching */ + b53_set_forwarding(priv, 0); + + b53_configure_vlan(priv); + + /* enable switching */ + b53_set_forwarding(priv, 1); + + return 0; +} + +static void b53_reset_mib(struct b53_device *priv) +{ + u8 gc; + + b53_read8(priv, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, &gc); + + b53_write8(priv, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, gc | GC_RESET_MIB); + msleep(1); + b53_write8(priv, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, gc & ~GC_RESET_MIB); + msleep(1); +} + +static const struct b53_mib_desc *b53_get_mib(struct b53_device *dev) +{ + if (is5365(dev)) + return b53_mibs_65; + else if (is63xx(dev)) + return b53_mibs_63xx; + else + return b53_mibs; +} + +static unsigned int b53_get_mib_size(struct b53_device *dev) +{ + if (is5365(dev)) + return B53_MIBS_65_SIZE; + else if (is63xx(dev)) + return B53_MIBS_63XX_SIZE; + else + return B53_MIBS_SIZE; +} + +static void b53_get_strings(struct dsa_switch *ds, int port, uint8_t *data) +{ + struct b53_device *dev = ds_to_priv(ds); + const struct b53_mib_desc *mibs = b53_get_mib(dev); + unsigned int mib_size = b53_get_mib_size(dev); + unsigned int i; + + for (i = 0; i < mib_size; i++) + memcpy(data + i * ETH_GSTRING_LEN, + mibs[i].name, ETH_GSTRING_LEN); +} + +static void b53_get_ethtool_stats(struct dsa_switch *ds, int port, + uint64_t *data) +{ + struct b53_device *dev = ds_to_priv(ds); + const struct b53_mib_desc *mibs = b53_get_mib(dev); + unsigned int mib_size = b53_get_mib_size(dev); + const struct b53_mib_desc *s; + unsigned int i; + u64 val = 0; + + if (is5365(dev) && port == 5) + port = 8; + + mutex_lock(&dev->stats_mutex); + + for (i = 0; i < mib_size; i++) { + s = &mibs[i]; + + if (mibs->size == 8) { + b53_read64(dev, B53_MIB_PAGE(port), s->offset, &val); + } else { + u32 val32; + + b53_read32(dev, B53_MIB_PAGE(port), s->offset, + &val32); + val = val32; + } + data[i] = (u64)val; + } + + mutex_unlock(&dev->stats_mutex); +} + +static int b53_get_sset_count(struct dsa_switch *ds) +{ + struct b53_device *dev = ds_to_priv(ds); + + return b53_get_mib_size(dev); +} + +static int b53_set_addr(struct dsa_switch *ds, u8 *addr) +{ + return 0; +} + +static int b53_setup(struct dsa_switch *ds) +{ + struct b53_device *dev = ds_to_priv(ds); + unsigned int port; + int ret; + + ret = b53_reset_switch(dev); + if (ret) { + dev_err(ds->dev, "failed to reset switch\n"); + return ret; + } + + b53_reset_mib(dev); + + ret = b53_apply_config(dev); + if (ret) + dev_err(ds->dev, "failed to apply configuration\n"); + + for (port = 0; port < dev->num_ports; port++) { + if (BIT(port) & ds->enabled_port_mask) + b53_enable_port(ds, port, NULL); + else if (dsa_is_cpu_port(ds, port)) + b53_enable_cpu_port(dev); + else + b53_disable_port(ds, port, NULL); + } + + return ret; +} + +static void b53_adjust_link(struct dsa_switch *ds, int port, + struct phy_device *phydev) +{ + struct b53_device *dev = ds_to_priv(ds); + u8 rgmii_ctrl = 0, reg = 0, off; + + if (!phy_is_pseudo_fixed_link(phydev)) + return; + + /* Override the port settings */ + if (port == dev->cpu_port) { + off = B53_PORT_OVERRIDE_CTRL; + reg = PORT_OVERRIDE_EN; + } else { + off = B53_GMII_PORT_OVERRIDE_CTRL(port); + reg = GMII_PO_EN; + } + + /* Set the link UP */ + if (phydev->link) + reg |= PORT_OVERRIDE_LINK; + + if (phydev->duplex == DUPLEX_FULL) + reg |= PORT_OVERRIDE_FULL_DUPLEX; + + switch (phydev->speed) { + case 2000: + reg |= PORT_OVERRIDE_SPEED_2000M; + /* fallthrough */ + case SPEED_1000: + reg |= PORT_OVERRIDE_SPEED_1000M; + break; + case SPEED_100: + reg |= PORT_OVERRIDE_SPEED_100M; + break; + case SPEED_10: + reg |= PORT_OVERRIDE_SPEED_10M; + break; + default: + dev_err(ds->dev, "unknown speed: %d\n", phydev->speed); + return; + } + + /* Enable flow control on BCM5301x's CPU port */ + if (is5301x(dev) && port == dev->cpu_port) + reg |= PORT_OVERRIDE_RX_FLOW | PORT_OVERRIDE_TX_FLOW; + + if (phydev->pause) { + if (phydev->asym_pause) + reg |= PORT_OVERRIDE_TX_FLOW; + reg |= PORT_OVERRIDE_RX_FLOW; + } + + b53_write8(dev, B53_CTRL_PAGE, off, reg); + + if (is531x5(dev) && phy_interface_is_rgmii(phydev)) { + if (port == 8) + off = B53_RGMII_CTRL_IMP; + else + off = B53_RGMII_CTRL_P(port); + + /* Configure the port RGMII clock delay by DLL disabled and + * tx_clk aligned timing (restoring to reset defaults) + */ + b53_read8(dev, B53_CTRL_PAGE, off, &rgmii_ctrl); + rgmii_ctrl &= ~(RGMII_CTRL_DLL_RXC | RGMII_CTRL_DLL_TXC | + RGMII_CTRL_TIMING_SEL); + + /* PHY_INTERFACE_MODE_RGMII_TXID means TX internal delay, make + * sure that we enable the port TX clock internal delay to + * account for this internal delay that is inserted, otherwise + * the switch won't be able to receive correctly. + * + * PHY_INTERFACE_MODE_RGMII means that we are not introducing + * any delay neither on transmission nor reception, so the + * BCM53125 must also be configured accordingly to account for + * the lack of delay and introduce + * + * The BCM53125 switch has its RX clock and TX clock control + * swapped, hence the reason why we modify the TX clock path in + * the "RGMII" case + */ + if (phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID) + rgmii_ctrl |= RGMII_CTRL_DLL_TXC; + if (phydev->interface == PHY_INTERFACE_MODE_RGMII) + rgmii_ctrl |= RGMII_CTRL_DLL_TXC | RGMII_CTRL_DLL_RXC; + rgmii_ctrl |= RGMII_CTRL_TIMING_SEL; + b53_write8(dev, B53_CTRL_PAGE, off, rgmii_ctrl); + + dev_info(ds->dev, "Configured port %d for %s\n", port, + phy_modes(phydev->interface)); + } + + /* configure MII port if necessary */ + if (is5325(dev)) { + b53_read8(dev, B53_CTRL_PAGE, B53_PORT_OVERRIDE_CTRL, + ®); + + /* reverse mii needs to be enabled */ + if (!(reg & PORT_OVERRIDE_RV_MII_25)) { + b53_write8(dev, B53_CTRL_PAGE, B53_PORT_OVERRIDE_CTRL, + reg | PORT_OVERRIDE_RV_MII_25); + b53_read8(dev, B53_CTRL_PAGE, B53_PORT_OVERRIDE_CTRL, + ®); + + if (!(reg & PORT_OVERRIDE_RV_MII_25)) { + dev_err(ds->dev, + "Failed to enable reverse MII mode\n"); + return; + } + } + } else if (is5301x(dev)) { + if (port != dev->cpu_port) { + u8 po_reg = B53_GMII_PORT_OVERRIDE_CTRL(dev->cpu_port); + u8 gmii_po; + + b53_read8(dev, B53_CTRL_PAGE, po_reg, &gmii_po); + gmii_po |= GMII_PO_LINK | + GMII_PO_RX_FLOW | + GMII_PO_TX_FLOW | + GMII_PO_EN | + GMII_PO_SPEED_2000M; + b53_write8(dev, B53_CTRL_PAGE, po_reg, gmii_po); + } + } +} + +static struct dsa_switch_driver b53_switch_ops = { + .tag_protocol = DSA_TAG_PROTO_NONE, + .setup = b53_setup, + .set_addr = b53_set_addr, + .get_strings = b53_get_strings, + .get_ethtool_stats = b53_get_ethtool_stats, + .get_sset_count = b53_get_sset_count, + .phy_read = b53_phy_read16, + .phy_write = b53_phy_write16, + .adjust_link = b53_adjust_link, + .port_enable = b53_enable_port, + .port_disable = b53_disable_port, +}; + +struct b53_chip_data { + u32 chip_id; + const char *dev_name; + u16 vlans; + u16 enabled_ports; + u8 cpu_port; + u8 vta_regs[3]; + u8 duplex_reg; + u8 jumbo_pm_reg; + u8 jumbo_size_reg; +}; + +#define B53_VTA_REGS \ + { B53_VT_ACCESS, B53_VT_INDEX, B53_VT_ENTRY } +#define B53_VTA_REGS_9798 \ + { B53_VT_ACCESS_9798, B53_VT_INDEX_9798, B53_VT_ENTRY_9798 } +#define B53_VTA_REGS_63XX \ + { B53_VT_ACCESS_63XX, B53_VT_INDEX_63XX, B53_VT_ENTRY_63XX } + +static const struct b53_chip_data b53_switch_chips[] = { + { + .chip_id = BCM5325_DEVICE_ID, + .dev_name = "BCM5325", + .vlans = 16, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT_25, + .duplex_reg = B53_DUPLEX_STAT_FE, + }, + { + .chip_id = BCM5365_DEVICE_ID, + .dev_name = "BCM5365", + .vlans = 256, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT_25, + .duplex_reg = B53_DUPLEX_STAT_FE, + }, + { + .chip_id = BCM5395_DEVICE_ID, + .dev_name = "BCM5395", + .vlans = 4096, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM5397_DEVICE_ID, + .dev_name = "BCM5397", + .vlans = 4096, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS_9798, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM5398_DEVICE_ID, + .dev_name = "BCM5398", + .vlans = 4096, + .enabled_ports = 0x7f, + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS_9798, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53115_DEVICE_ID, + .dev_name = "BCM53115", + .vlans = 4096, + .enabled_ports = 0x1f, + .vta_regs = B53_VTA_REGS, + .cpu_port = B53_CPU_PORT, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53125_DEVICE_ID, + .dev_name = "BCM53125", + .vlans = 4096, + .enabled_ports = 0xff, + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53128_DEVICE_ID, + .dev_name = "BCM53128", + .vlans = 4096, + .enabled_ports = 0x1ff, + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM63XX_DEVICE_ID, + .dev_name = "BCM63xx", + .vlans = 4096, + .enabled_ports = 0, /* pdata must provide them */ + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS_63XX, + .duplex_reg = B53_DUPLEX_STAT_63XX, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK_63XX, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE_63XX, + }, + { + .chip_id = BCM53010_DEVICE_ID, + .dev_name = "BCM53010", + .vlans = 4096, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53011_DEVICE_ID, + .dev_name = "BCM53011", + .vlans = 4096, + .enabled_ports = 0x1bf, + .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53012_DEVICE_ID, + .dev_name = "BCM53012", + .vlans = 4096, + .enabled_ports = 0x1bf, + .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53018_DEVICE_ID, + .dev_name = "BCM53018", + .vlans = 4096, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53019_DEVICE_ID, + .dev_name = "BCM53019", + .vlans = 4096, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, +}; + +static int b53_switch_init(struct b53_device *dev) +{ + struct dsa_switch *ds = dev->ds; + unsigned int i; + int ret; + + for (i = 0; i < ARRAY_SIZE(b53_switch_chips); i++) { + const struct b53_chip_data *chip = &b53_switch_chips[i]; + + if (chip->chip_id == dev->chip_id) { + if (!dev->enabled_ports) + dev->enabled_ports = chip->enabled_ports; + dev->name = chip->dev_name; + dev->duplex_reg = chip->duplex_reg; + dev->vta_regs[0] = chip->vta_regs[0]; + dev->vta_regs[1] = chip->vta_regs[1]; + dev->vta_regs[2] = chip->vta_regs[2]; + dev->jumbo_pm_reg = chip->jumbo_pm_reg; + ds->drv = &b53_switch_ops; + dev->cpu_port = chip->cpu_port; + dev->num_vlans = chip->vlans; + break; + } + } + + /* check which BCM5325x version we have */ + if (is5325(dev)) { + u8 vc4; + + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4_25, &vc4); + + /* check reserved bits */ + switch (vc4 & 3) { + case 1: + /* BCM5325E */ + break; + case 3: + /* BCM5325F - do not use port 4 */ + dev->enabled_ports &= ~BIT(4); + break; + default: +/* On the BCM47XX SoCs this is the supported internal switch.*/ +#ifndef CONFIG_BCM47XX + /* BCM5325M */ + return -EINVAL; +#else + break; +#endif + } + } else if (dev->chip_id == BCM53115_DEVICE_ID) { + u64 strap_value; + + b53_read48(dev, B53_STAT_PAGE, B53_STRAP_VALUE, &strap_value); + /* use second IMP port if GMII is enabled */ + if (strap_value & SV_GMII_CTRL_115) + dev->cpu_port = 5; + } + + /* cpu port is always last */ + dev->num_ports = dev->cpu_port + 1; + dev->enabled_ports |= BIT(dev->cpu_port); + + dev->ports = devm_kzalloc(dev->dev, + sizeof(struct b53_port) * dev->num_ports, + GFP_KERNEL); + if (!dev->ports) + return -ENOMEM; + + dev->reset_gpio = b53_switch_get_reset_gpio(dev); + if (dev->reset_gpio >= 0) { + ret = devm_gpio_request_one(dev->dev, dev->reset_gpio, + GPIOF_OUT_INIT_HIGH, "robo_reset"); + if (ret) + return ret; + } + + return 0; +} + +struct b53_device *b53_switch_alloc(struct device *base, struct b53_io_ops *ops, + void *priv) +{ + struct dsa_switch *ds; + struct b53_device *dev; + + ds = devm_kzalloc(base, sizeof(*ds) + sizeof(*dev), GFP_KERNEL); + if (!ds) + return NULL; + + dev = (struct b53_device *)(ds + 1); + + ds->priv = dev; + ds->dev = base; + dev->dev = base; + + dev->ds = ds; + dev->priv = priv; + dev->ops = ops; + mutex_init(&dev->reg_mutex); + mutex_init(&dev->stats_mutex); + + return dev; +} +EXPORT_SYMBOL(b53_switch_alloc); + +int b53_switch_detect(struct b53_device *dev) +{ + u32 id32; + u16 tmp; + u8 id8; + int ret; + + ret = b53_read8(dev, B53_MGMT_PAGE, B53_DEVICE_ID, &id8); + if (ret) + return ret; + + switch (id8) { + case 0: + /* BCM5325 and BCM5365 do not have this register so reads + * return 0. But the read operation did succeed, so assume this + * is one of them. + * + * Next check if we can write to the 5325's VTA register; for + * 5365 it is read only. + */ + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_25, 0xf); + b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_25, &tmp); + + if (tmp == 0xf) + dev->chip_id = BCM5325_DEVICE_ID; + else + dev->chip_id = BCM5365_DEVICE_ID; + break; + case BCM5395_DEVICE_ID: + case BCM5397_DEVICE_ID: + case BCM5398_DEVICE_ID: + dev->chip_id = id8; + break; + default: + ret = b53_read32(dev, B53_MGMT_PAGE, B53_DEVICE_ID, &id32); + if (ret) + return ret; + + switch (id32) { + case BCM53115_DEVICE_ID: + case BCM53125_DEVICE_ID: + case BCM53128_DEVICE_ID: + case BCM53010_DEVICE_ID: + case BCM53011_DEVICE_ID: + case BCM53012_DEVICE_ID: + case BCM53018_DEVICE_ID: + case BCM53019_DEVICE_ID: + dev->chip_id = id32; + break; + default: + pr_err("unsupported switch detected (BCM53%02x/BCM%x)\n", + id8, id32); + return -ENODEV; + } + } + + if (dev->chip_id == BCM5325_DEVICE_ID) + return b53_read8(dev, B53_STAT_PAGE, B53_REV_ID_25, + &dev->core_rev); + else + return b53_read8(dev, B53_MGMT_PAGE, B53_REV_ID, + &dev->core_rev); +} +EXPORT_SYMBOL(b53_switch_detect); + +int b53_switch_register(struct b53_device *dev) +{ + int ret; + + if (dev->pdata) { + dev->chip_id = dev->pdata->chip_id; + dev->enabled_ports = dev->pdata->enabled_ports; + } + + if (!dev->chip_id && b53_switch_detect(dev)) + return -EINVAL; + + ret = b53_switch_init(dev); + if (ret) + return ret; + + pr_info("found switch: %s, rev %i\n", dev->name, dev->core_rev); + + return dsa_register_switch(dev->ds, dev->ds->dev->of_node); +} +EXPORT_SYMBOL(b53_switch_register); + +MODULE_AUTHOR("Jonas Gorski "); +MODULE_DESCRIPTION("B53 switch library"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/dsa/b53/b53_mdio.c b/drivers/net/dsa/b53/b53_mdio.c new file mode 100644 index 0000000..c6cf7cf --- /dev/null +++ b/drivers/net/dsa/b53/b53_mdio.c @@ -0,0 +1,381 @@ +/* + * B53 register access through MII registers + * + * Copyright (C) 2011-2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include +#include + +#include "b53_priv.h" + +/* MII registers */ +#define REG_MII_PAGE 0x10 /* MII Page register */ +#define REG_MII_ADDR 0x11 /* MII Address register */ +#define REG_MII_DATA0 0x18 /* MII Data register 0 */ +#define REG_MII_DATA1 0x19 /* MII Data register 1 */ +#define REG_MII_DATA2 0x1a /* MII Data register 2 */ +#define REG_MII_DATA3 0x1b /* MII Data register 3 */ + +#define REG_MII_PAGE_ENABLE BIT(0) +#define REG_MII_ADDR_WRITE BIT(0) +#define REG_MII_ADDR_READ BIT(1) + +static int b53_mdio_op(struct b53_device *dev, u8 page, u8 reg, u16 op) +{ + int i; + u16 v; + int ret; + struct mii_bus *bus = dev->priv; + + if (dev->current_page != page) { + /* set page number */ + v = (page << 8) | REG_MII_PAGE_ENABLE; + ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_PAGE, v); + if (ret) + return ret; + dev->current_page = page; + } + + /* set register address */ + v = (reg << 8) | op; + ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, REG_MII_ADDR, v); + if (ret) + return ret; + + /* check if operation completed */ + for (i = 0; i < 5; ++i) { + v = mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_ADDR); + if (!(v & (REG_MII_ADDR_WRITE | REG_MII_ADDR_READ))) + break; + usleep_range(10, 100); + } + + if (WARN_ON(i == 5)) + return -EIO; + + return 0; +} + +static int b53_mdio_read8(struct b53_device *dev, u8 page, u8 reg, u8 *val) +{ + struct mii_bus *bus = dev->priv; + int ret; + + ret = b53_mdio_op(dev, page, reg, REG_MII_ADDR_READ); + if (ret) + return ret; + + *val = mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0) & 0xff; + + return 0; +} + +static int b53_mdio_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) +{ + struct mii_bus *bus = dev->priv; + int ret; + + ret = b53_mdio_op(dev, page, reg, REG_MII_ADDR_READ); + if (ret) + return ret; + + *val = mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, REG_MII_DATA0); + + return 0; +} + +static int b53_mdio_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) +{ + struct mii_bus *bus = dev->priv; + int ret; + + ret = b53_mdio_op(dev, page, reg, REG_MII_ADDR_READ); + if (ret) + return ret; + + *val = mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, REG_MII_DATA0); + *val |= mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA1) << 16; + + return 0; +} + +static int b53_mdio_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + struct mii_bus *bus = dev->priv; + u64 temp = 0; + int i; + int ret; + + ret = b53_mdio_op(dev, page, reg, REG_MII_ADDR_READ); + if (ret) + return ret; + + for (i = 2; i >= 0; i--) { + temp <<= 16; + temp |= mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0 + i); + } + + *val = temp; + + return 0; +} + +static int b53_mdio_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + struct mii_bus *bus = dev->priv; + u64 temp = 0; + int i; + int ret; + + ret = b53_mdio_op(dev, page, reg, REG_MII_ADDR_READ); + if (ret) + return ret; + + for (i = 3; i >= 0; i--) { + temp <<= 16; + temp |= mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0 + i); + } + + *val = temp; + + return 0; +} + +static int b53_mdio_write8(struct b53_device *dev, u8 page, u8 reg, u8 value) +{ + struct mii_bus *bus = dev->priv; + int ret; + + ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0, value); + if (ret) + return ret; + + return b53_mdio_op(dev, page, reg, REG_MII_ADDR_WRITE); +} + +static int b53_mdio_write16(struct b53_device *dev, u8 page, u8 reg, + u16 value) +{ + struct mii_bus *bus = dev->priv; + int ret; + + ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0, value); + if (ret) + return ret; + + return b53_mdio_op(dev, page, reg, REG_MII_ADDR_WRITE); +} + +static int b53_mdio_write32(struct b53_device *dev, u8 page, u8 reg, + u32 value) +{ + struct mii_bus *bus = dev->priv; + unsigned int i; + u32 temp = value; + + for (i = 0; i < 2; i++) { + int ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0 + i, + temp & 0xffff); + if (ret) + return ret; + temp >>= 16; + } + + return b53_mdio_op(dev, page, reg, REG_MII_ADDR_WRITE); +} + +static int b53_mdio_write48(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + struct mii_bus *bus = dev->priv; + unsigned int i; + u64 temp = value; + + for (i = 0; i < 3; i++) { + int ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0 + i, + temp & 0xffff); + if (ret) + return ret; + temp >>= 16; + } + + return b53_mdio_op(dev, page, reg, REG_MII_ADDR_WRITE); +} + +static int b53_mdio_write64(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + struct mii_bus *bus = dev->priv; + unsigned int i; + u64 temp = value; + + for (i = 0; i < 4; i++) { + int ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0 + i, + temp & 0xffff); + if (ret) + return ret; + temp >>= 16; + } + + return b53_mdio_op(dev, page, reg, REG_MII_ADDR_WRITE); +} + +static int b53_mdio_phy_read16(struct b53_device *dev, int addr, int reg, + u16 *value) +{ + struct mii_bus *bus = dev->priv; + + *value = mdiobus_read_nested(bus, addr, reg); + + return 0; +} + +static int b53_mdio_phy_write16(struct b53_device *dev, int addr, int reg, + u16 value) +{ + struct mii_bus *bus = dev->bus; + + return mdiobus_write_nested(bus, addr, reg, value); +} + +static struct b53_io_ops b53_mdio_ops = { + .read8 = b53_mdio_read8, + .read16 = b53_mdio_read16, + .read32 = b53_mdio_read32, + .read48 = b53_mdio_read48, + .read64 = b53_mdio_read64, + .write8 = b53_mdio_write8, + .write16 = b53_mdio_write16, + .write32 = b53_mdio_write32, + .write48 = b53_mdio_write48, + .write64 = b53_mdio_write64, + .phy_read16 = b53_mdio_phy_read16, + .phy_write16 = b53_mdio_phy_write16, +}; + +#define B53_BRCM_OUI_1 0x0143bc00 +#define B53_BRCM_OUI_2 0x03625c00 +#define B53_BRCM_OUI_3 0x00406000 + +static int b53_mdio_probe(struct mdio_device *mdiodev) +{ + struct b53_device *dev; + u32 phy_id; + int ret; + + /* allow the generic PHY driver to take over the non-management MDIO + * addresses + */ + if (mdiodev->addr != BRCM_PSEUDO_PHY_ADDR && mdiodev->addr != 0) { + dev_err(&mdiodev->dev, "leaving address %d to PHY\n", + mdiodev->addr); + return -ENODEV; + } + + /* read the first port's id */ + phy_id = mdiobus_read(mdiodev->bus, 0, 2) << 16; + phy_id |= mdiobus_read(mdiodev->bus, 0, 3); + + /* BCM5325, BCM539x (OUI_1) + * BCM53125, BCM53128 (OUI_2) + * BCM5365 (OUI_3) + */ + if ((phy_id & 0xfffffc00) != B53_BRCM_OUI_1 && + (phy_id & 0xfffffc00) != B53_BRCM_OUI_2 && + (phy_id & 0xfffffc00) != B53_BRCM_OUI_3) { + dev_err(&mdiodev->dev, "Unsupported device: 0x%08x\n", phy_id); + return -ENODEV; + } + + dev = b53_switch_alloc(&mdiodev->dev, &b53_mdio_ops, mdiodev->bus); + if (!dev) + return -ENOMEM; + + /* we don't use page 0xff, so force a page set */ + dev->current_page = 0xff; + dev->bus = mdiodev->bus; + + dev_set_drvdata(&mdiodev->dev, dev); + + ret = b53_switch_register(dev); + if (ret) { + dev_err(&mdiodev->dev, "failed to register switch: %i\n", ret); + return ret; + } + + return ret; +} + +static void b53_mdio_remove(struct mdio_device *mdiodev) +{ + struct b53_device *dev = dev_get_drvdata(&mdiodev->dev); + struct dsa_switch *ds = dev->ds; + + dsa_unregister_switch(ds); +} + +static const struct of_device_id b53_of_match[] = { + { .compatible = "brcm,bcm5325" }, + { .compatible = "brcm,bcm53115" }, + { .compatible = "brcm,bcm53125" }, + { .compatible = "brcm,bcm53128" }, + { .compatible = "brcm,bcm5365" }, + { .compatible = "brcm,bcm5395" }, + { .compatible = "brcm,bcm5397" }, + { .compatible = "brcm,bcm5398" }, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, b53_of_match); + +static struct mdio_driver b53_mdio_driver = { + .probe = b53_mdio_probe, + .remove = b53_mdio_remove, + .mdiodrv.driver = { + .name = "bcm53xx", + .of_match_table = b53_of_match, + }, +}; + +static int __init b53_mdio_driver_register(void) +{ + return mdio_driver_register(&b53_mdio_driver); +} +module_init(b53_mdio_driver_register); + +static void __exit b53_mdio_driver_unregister(void) +{ + mdio_driver_unregister(&b53_mdio_driver); +} +module_exit(b53_mdio_driver_unregister); + +MODULE_DESCRIPTION("B53 MDIO access driver"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/dsa/b53/b53_mmap.c b/drivers/net/dsa/b53/b53_mmap.c new file mode 100644 index 0000000..f115ee2 --- /dev/null +++ b/drivers/net/dsa/b53/b53_mmap.c @@ -0,0 +1,260 @@ +/* + * B53 register access through memory mapped registers + * + * Copyright (C) 2012-2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include + +#include "b53_priv.h" + +struct b53_mmap_priv { + void __iomem *regs; +}; + +static int b53_mmap_read8(struct b53_device *dev, u8 page, u8 reg, u8 *val) +{ + u8 __iomem *regs = dev->priv; + + *val = readb(regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) +{ + u8 __iomem *regs = dev->priv; + + if (WARN_ON(reg % 2)) + return -EINVAL; + + if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && + dev->pdata->big_endian) + *val = __raw_readw(regs + (page << 8) + reg); + else + *val = readw(regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) +{ + u8 __iomem *regs = dev->priv; + + if (WARN_ON(reg % 4)) + return -EINVAL; + + if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && + dev->pdata->big_endian) + *val = __raw_readl(regs + (page << 8) + reg); + else + *val = readl(regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + if (WARN_ON(reg % 2)) + return -EINVAL; + + if (reg % 4) { + u16 lo; + u32 hi; + + b53_mmap_read16(dev, page, reg, &lo); + b53_mmap_read32(dev, page, reg + 2, &hi); + + *val = ((u64)hi << 16) | lo; + } else { + u32 lo; + u16 hi; + + b53_mmap_read32(dev, page, reg, &lo); + b53_mmap_read16(dev, page, reg + 4, &hi); + + *val = ((u64)hi << 32) | lo; + } + + return 0; +} + +static int b53_mmap_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + u32 hi, lo; + + if (WARN_ON(reg % 4)) + return -EINVAL; + + b53_mmap_read32(dev, page, reg, &lo); + b53_mmap_read32(dev, page, reg + 4, &hi); + + *val = ((u64)hi << 32) | lo; + + return 0; +} + +static int b53_mmap_write8(struct b53_device *dev, u8 page, u8 reg, u8 value) +{ + u8 __iomem *regs = dev->priv; + + writeb(value, regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_write16(struct b53_device *dev, u8 page, u8 reg, + u16 value) +{ + u8 __iomem *regs = dev->priv; + + if (WARN_ON(reg % 2)) + return -EINVAL; + + if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && + dev->pdata->big_endian) + __raw_writew(value, regs + (page << 8) + reg); + else + writew(value, regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_write32(struct b53_device *dev, u8 page, u8 reg, + u32 value) +{ + u8 __iomem *regs = dev->priv; + + if (WARN_ON(reg % 4)) + return -EINVAL; + + if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && + dev->pdata->big_endian) + __raw_writel(value, regs + (page << 8) + reg); + else + writel(value, regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_write48(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + if (WARN_ON(reg % 2)) + return -EINVAL; + + if (reg % 4) { + u32 hi = (u32)(value >> 16); + u16 lo = (u16)value; + + b53_mmap_write16(dev, page, reg, lo); + b53_mmap_write32(dev, page, reg + 2, hi); + } else { + u16 hi = (u16)(value >> 32); + u32 lo = (u32)value; + + b53_mmap_write32(dev, page, reg, lo); + b53_mmap_write16(dev, page, reg + 4, hi); + } + + return 0; +} + +static int b53_mmap_write64(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + u32 hi, lo; + + hi = upper_32_bits(value); + lo = lower_32_bits(value); + + if (WARN_ON(reg % 4)) + return -EINVAL; + + b53_mmap_write32(dev, page, reg, lo); + b53_mmap_write32(dev, page, reg + 4, hi); + + return 0; +} + +static struct b53_io_ops b53_mmap_ops = { + .read8 = b53_mmap_read8, + .read16 = b53_mmap_read16, + .read32 = b53_mmap_read32, + .read48 = b53_mmap_read48, + .read64 = b53_mmap_read64, + .write8 = b53_mmap_write8, + .write16 = b53_mmap_write16, + .write32 = b53_mmap_write32, + .write48 = b53_mmap_write48, + .write64 = b53_mmap_write64, +}; + +static int b53_mmap_probe(struct platform_device *pdev) +{ + struct b53_platform_data *pdata = pdev->dev.platform_data; + struct b53_device *dev; + + if (!pdata) + return -EINVAL; + + dev = b53_switch_alloc(&pdev->dev, &b53_mmap_ops, pdata->regs); + if (!dev) + return -ENOMEM; + + if (pdata) + dev->pdata = pdata; + + platform_set_drvdata(pdev, dev); + + return b53_switch_register(dev); +} + +static int b53_mmap_remove(struct platform_device *pdev) +{ + struct b53_device *dev = platform_get_drvdata(pdev); + + if (dev) + b53_switch_remove(dev); + + return 0; +} + +static const struct of_device_id b53_mmap_of_table[] = { + { .compatible = "brcm,bcm3384-switch" }, + { .compatible = "brcm,bcm6328-switch" }, + { .compatible = "brcm,bcm6368-switch" }, + { .compatible = "brcm,bcm63xx-switch" }, + { /* sentinel */ }, +}; + +static struct platform_driver b53_mmap_driver = { + .probe = b53_mmap_probe, + .remove = b53_mmap_remove, + .driver = { + .name = "b53-switch", + .of_match_table = b53_mmap_of_table, + }, +}; + +module_platform_driver(b53_mmap_driver); +MODULE_AUTHOR("Jonas Gorski "); +MODULE_DESCRIPTION("B53 MMAP access driver"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h new file mode 100644 index 0000000..c484415 --- /dev/null +++ b/drivers/net/dsa/b53/b53_priv.h @@ -0,0 +1,322 @@ +/* + * B53 common definitions + * + * Copyright (C) 2011-2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#ifndef __B53_PRIV_H +#define __B53_PRIV_H + +#include +#include +#include +#include + +struct b53_device; + +struct b53_io_ops { + int (*read8)(struct b53_device *dev, u8 page, u8 reg, u8 *value); + int (*read16)(struct b53_device *dev, u8 page, u8 reg, u16 *value); + int (*read32)(struct b53_device *dev, u8 page, u8 reg, u32 *value); + int (*read48)(struct b53_device *dev, u8 page, u8 reg, u64 *value); + int (*read64)(struct b53_device *dev, u8 page, u8 reg, u64 *value); + int (*write8)(struct b53_device *dev, u8 page, u8 reg, u8 value); + int (*write16)(struct b53_device *dev, u8 page, u8 reg, u16 value); + int (*write32)(struct b53_device *dev, u8 page, u8 reg, u32 value); + int (*write48)(struct b53_device *dev, u8 page, u8 reg, u64 value); + int (*write64)(struct b53_device *dev, u8 page, u8 reg, u64 value); + int (*phy_read16)(struct b53_device *dev, int addr, int reg, u16 *value); + int (*phy_write16)(struct b53_device *dev, int addr, int reg, u16 value); +}; + +enum { + BCM5325_DEVICE_ID = 0x25, + BCM5365_DEVICE_ID = 0x65, + BCM5395_DEVICE_ID = 0x95, + BCM5397_DEVICE_ID = 0x97, + BCM5398_DEVICE_ID = 0x98, + BCM53115_DEVICE_ID = 0x53115, + BCM53125_DEVICE_ID = 0x53125, + BCM53128_DEVICE_ID = 0x53128, + BCM63XX_DEVICE_ID = 0x6300, + BCM53010_DEVICE_ID = 0x53010, + BCM53011_DEVICE_ID = 0x53011, + BCM53012_DEVICE_ID = 0x53012, + BCM53018_DEVICE_ID = 0x53018, + BCM53019_DEVICE_ID = 0x53019, +}; + +#define B53_N_PORTS 9 +#define B53_N_PORTS_25 6 + +struct b53_port { +}; + +struct b53_device { + struct dsa_switch *ds; + struct b53_platform_data *pdata; + const char *name; + + struct mutex reg_mutex; + struct mutex stats_mutex; + const struct b53_io_ops *ops; + + /* chip specific data */ + u32 chip_id; + u8 core_rev; + u8 vta_regs[3]; + u8 duplex_reg; + u8 jumbo_pm_reg; + u8 jumbo_size_reg; + int reset_gpio; + + /* used ports mask */ + u16 enabled_ports; + unsigned int cpu_port; + + /* connect specific data */ + u8 current_page; + struct device *dev; + + /* Master MDIO bus we got probed from */ + struct mii_bus *bus; + + /* Slave MDIO bus we created */ + struct mii_bus *slave_bus; + void *priv; + + /* run time configuration */ + unsigned enable_jumbo:1; + unsigned allow_vid_4095:1; + unsigned int num_vlans; + unsigned int num_ports; + struct b53_port *ports; +}; + +#define b53_for_each_port(dev, i) \ + for (i = 0; i < B53_N_PORTS; i++) \ + if (dev->enabled_ports & BIT(i)) + + +static inline int is5325(struct b53_device *dev) +{ + return dev->chip_id == BCM5325_DEVICE_ID; +} + +static inline int is5365(struct b53_device *dev) +{ +#ifdef CONFIG_BCM47XX + return dev->chip_id == BCM5365_DEVICE_ID; +#else + return 0; +#endif +} + +static inline int is5397_98(struct b53_device *dev) +{ + return dev->chip_id == BCM5397_DEVICE_ID || + dev->chip_id == BCM5398_DEVICE_ID; +} + +static inline int is539x(struct b53_device *dev) +{ + return dev->chip_id == BCM5395_DEVICE_ID || + dev->chip_id == BCM5397_DEVICE_ID || + dev->chip_id == BCM5398_DEVICE_ID; +} + +static inline int is531x5(struct b53_device *dev) +{ + return dev->chip_id == BCM53115_DEVICE_ID || + dev->chip_id == BCM53125_DEVICE_ID || + dev->chip_id == BCM53128_DEVICE_ID; +} + +static inline int is63xx(struct b53_device *dev) +{ +#ifdef CONFIG_BCM63XX + return dev->chip_id == BCM63XX_DEVICE_ID; +#else + return 0; +#endif +} + +static inline int is5301x(struct b53_device *dev) +{ + return dev->chip_id == BCM53010_DEVICE_ID || + dev->chip_id == BCM53011_DEVICE_ID || + dev->chip_id == BCM53012_DEVICE_ID || + dev->chip_id == BCM53018_DEVICE_ID || + dev->chip_id == BCM53019_DEVICE_ID; +} + +#define B53_CPU_PORT_25 5 +#define B53_CPU_PORT 8 + +static inline int is_cpu_port(struct b53_device *dev, int port) +{ + return dev->cpu_port; +} + +struct b53_device *b53_switch_alloc(struct device *base, struct b53_io_ops *ops, + void *priv); + +int b53_switch_detect(struct b53_device *dev); + +int b53_switch_register(struct b53_device *dev); + +static inline void b53_switch_remove(struct b53_device *dev) +{ + dsa_unregister_switch(dev->ds); +} + +static inline int b53_read8(struct b53_device *dev, u8 page, u8 reg, u8 *val) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->read8(dev, page, reg, val); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->read16(dev, page, reg, val); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->read32(dev, page, reg, val); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->read48(dev, page, reg, val); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->read64(dev, page, reg, val); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_write8(struct b53_device *dev, u8 page, u8 reg, u8 value) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->write8(dev, page, reg, value); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_write16(struct b53_device *dev, u8 page, u8 reg, + u16 value) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->write16(dev, page, reg, value); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_write32(struct b53_device *dev, u8 page, u8 reg, + u32 value) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->write32(dev, page, reg, value); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_write48(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->write48(dev, page, reg, value); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_write64(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->write64(dev, page, reg, value); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +#ifdef CONFIG_BCM47XX + +#include +#include +#include +static inline int b53_switch_get_reset_gpio(struct b53_device *dev) +{ + enum bcm47xx_board board = bcm47xx_board_get(); + + switch (board) { + case BCM47XX_BOARD_LINKSYS_WRT300NV11: + case BCM47XX_BOARD_LINKSYS_WRT310NV1: + return 8; + default: + return bcm47xx_nvram_gpio_pin("robo_reset"); + } +} +#else +static inline int b53_switch_get_reset_gpio(struct b53_device *dev) +{ + return -ENOENT; +} +#endif +#endif diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h new file mode 100644 index 0000000..ccf8af7 --- /dev/null +++ b/drivers/net/dsa/b53/b53_regs.h @@ -0,0 +1,358 @@ +/* + * B53 register definitions + * + * Copyright (C) 2004 Broadcom Corporation + * Copyright (C) 2011-2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#ifndef __B53_REGS_H +#define __B53_REGS_H + +/* Management Port (SMP) Page offsets */ +#define B53_CTRL_PAGE 0x00 /* Control */ +#define B53_STAT_PAGE 0x01 /* Status */ +#define B53_MGMT_PAGE 0x02 /* Management Mode */ +#define B53_MIB_AC_PAGE 0x03 /* MIB Autocast */ +#define B53_ARLCTRL_PAGE 0x04 /* ARL Control */ +#define B53_ARLIO_PAGE 0x05 /* ARL Access */ +#define B53_FRAMEBUF_PAGE 0x06 /* Management frame access */ +#define B53_MEM_ACCESS_PAGE 0x08 /* Memory access */ + +/* PHY Registers */ +#define B53_PORT_MII_PAGE(i) (0x10 + (i)) /* Port i MII Registers */ +#define B53_IM_PORT_PAGE 0x18 /* Inverse MII Port (to EMAC) */ +#define B53_ALL_PORT_PAGE 0x19 /* All ports MII (broadcast) */ + +/* MIB registers */ +#define B53_MIB_PAGE(i) (0x20 + (i)) + +/* Quality of Service (QoS) Registers */ +#define B53_QOS_PAGE 0x30 + +/* Port VLAN Page */ +#define B53_PVLAN_PAGE 0x31 + +/* VLAN Registers */ +#define B53_VLAN_PAGE 0x34 + +/* Jumbo Frame Registers */ +#define B53_JUMBO_PAGE 0x40 + +/* CFP Configuration Registers Page */ +#define B53_CFP_PAGE 0xa1 + +/************************************************************************* + * Control Page registers + *************************************************************************/ + +/* Port Control Register (8 bit) */ +#define B53_PORT_CTRL(i) (0x00 + (i)) +#define PORT_CTRL_RX_DISABLE BIT(0) +#define PORT_CTRL_TX_DISABLE BIT(1) +#define PORT_CTRL_RX_BCST_EN BIT(2) /* Broadcast RX (P8 only) */ +#define PORT_CTRL_RX_MCST_EN BIT(3) /* Multicast RX (P8 only) */ +#define PORT_CTRL_RX_UCST_EN BIT(4) /* Unicast RX (P8 only) */ +#define PORT_CTRL_STP_STATE_S 5 +#define PORT_CTRL_STP_STATE_MASK (0x7 << PORT_CTRL_STP_STATE_S) + +/* SMP Control Register (8 bit) */ +#define B53_SMP_CTRL 0x0a + +/* Switch Mode Control Register (8 bit) */ +#define B53_SWITCH_MODE 0x0b +#define SM_SW_FWD_MODE BIT(0) /* 1 = Managed Mode */ +#define SM_SW_FWD_EN BIT(1) /* Forwarding Enable */ + +/* IMP Port state override register (8 bit) */ +#define B53_PORT_OVERRIDE_CTRL 0x0e +#define PORT_OVERRIDE_LINK BIT(0) +#define PORT_OVERRIDE_FULL_DUPLEX BIT(1) /* 0 = Half Duplex */ +#define PORT_OVERRIDE_SPEED_S 2 +#define PORT_OVERRIDE_SPEED_10M (0 << PORT_OVERRIDE_SPEED_S) +#define PORT_OVERRIDE_SPEED_100M (1 << PORT_OVERRIDE_SPEED_S) +#define PORT_OVERRIDE_SPEED_1000M (2 << PORT_OVERRIDE_SPEED_S) +#define PORT_OVERRIDE_RV_MII_25 BIT(4) /* BCM5325 only */ +#define PORT_OVERRIDE_RX_FLOW BIT(4) +#define PORT_OVERRIDE_TX_FLOW BIT(5) +#define PORT_OVERRIDE_SPEED_2000M BIT(6) /* BCM5301X only, requires setting 1000M */ +#define PORT_OVERRIDE_EN BIT(7) /* Use the register contents */ + +/* Power-down mode control */ +#define B53_PD_MODE_CTRL_25 0x0f + +/* IP Multicast control (8 bit) */ +#define B53_IP_MULTICAST_CTRL 0x21 +#define B53_IPMC_FWD_EN BIT(1) +#define B53_UC_FWD_EN BIT(6) +#define B53_MC_FWD_EN BIT(7) + +/* (16 bit) */ +#define B53_UC_FLOOD_MASK 0x32 +#define B53_MC_FLOOD_MASK 0x34 +#define B53_IPMC_FLOOD_MASK 0x36 + +/* + * Override Ports 0-7 State on devices with xMII interfaces (8 bit) + * + * For port 8 still use B53_PORT_OVERRIDE_CTRL + * Please note that not all ports are available on every hardware, e.g. BCM5301X + * don't include overriding port 6, BCM63xx also have some limitations. + */ +#define B53_GMII_PORT_OVERRIDE_CTRL(i) (0x58 + (i)) +#define GMII_PO_LINK BIT(0) +#define GMII_PO_FULL_DUPLEX BIT(1) /* 0 = Half Duplex */ +#define GMII_PO_SPEED_S 2 +#define GMII_PO_SPEED_10M (0 << GMII_PO_SPEED_S) +#define GMII_PO_SPEED_100M (1 << GMII_PO_SPEED_S) +#define GMII_PO_SPEED_1000M (2 << GMII_PO_SPEED_S) +#define GMII_PO_RX_FLOW BIT(4) +#define GMII_PO_TX_FLOW BIT(5) +#define GMII_PO_EN BIT(6) /* Use the register contents */ +#define GMII_PO_SPEED_2000M BIT(7) /* BCM5301X only, requires setting 1000M */ + +#define B53_RGMII_CTRL_IMP 0x60 +#define RGMII_CTRL_ENABLE_GMII BIT(7) +#define RGMII_CTRL_TIMING_SEL BIT(2) +#define RGMII_CTRL_DLL_RXC BIT(1) +#define RGMII_CTRL_DLL_TXC BIT(0) + +#define B53_RGMII_CTRL_P(i) (B53_RGMII_CTRL_IMP + (i)) + +/* Software reset register (8 bit) */ +#define B53_SOFTRESET 0x79 +#define SW_RST BIT(7) +#define EN_SW_RST BIT(4) + +/* Fast Aging Control register (8 bit) */ +#define B53_FAST_AGE_CTRL 0x88 +#define FAST_AGE_STATIC BIT(0) +#define FAST_AGE_DYNAMIC BIT(1) +#define FAST_AGE_PORT BIT(2) +#define FAST_AGE_VLAN BIT(3) +#define FAST_AGE_STP BIT(4) +#define FAST_AGE_MC BIT(5) +#define FAST_AGE_DONE BIT(7) + +/************************************************************************* + * Status Page registers + *************************************************************************/ + +/* Link Status Summary Register (16bit) */ +#define B53_LINK_STAT 0x00 + +/* Link Status Change Register (16 bit) */ +#define B53_LINK_STAT_CHANGE 0x02 + +/* Port Speed Summary Register (16 bit for FE, 32 bit for GE) */ +#define B53_SPEED_STAT 0x04 +#define SPEED_PORT_FE(reg, port) (((reg) >> (port)) & 1) +#define SPEED_PORT_GE(reg, port) (((reg) >> 2 * (port)) & 3) +#define SPEED_STAT_10M 0 +#define SPEED_STAT_100M 1 +#define SPEED_STAT_1000M 2 + +/* Duplex Status Summary (16 bit) */ +#define B53_DUPLEX_STAT_FE 0x06 +#define B53_DUPLEX_STAT_GE 0x08 +#define B53_DUPLEX_STAT_63XX 0x0c + +/* Revision ID register for BCM5325 */ +#define B53_REV_ID_25 0x50 + +/* Strap Value (48 bit) */ +#define B53_STRAP_VALUE 0x70 +#define SV_GMII_CTRL_115 BIT(27) + +/************************************************************************* + * Management Mode Page Registers + *************************************************************************/ + +/* Global Management Config Register (8 bit) */ +#define B53_GLOBAL_CONFIG 0x00 +#define GC_RESET_MIB 0x01 +#define GC_RX_BPDU_EN 0x02 +#define GC_MIB_AC_HDR_EN 0x10 +#define GC_MIB_AC_EN 0x20 +#define GC_FRM_MGMT_PORT_M 0xC0 +#define GC_FRM_MGMT_PORT_04 0x00 +#define GC_FRM_MGMT_PORT_MII 0x80 + +/* Broadcom Header control register (8 bit) */ +#define B53_BRCM_HDR 0x03 +#define BRCM_HDR_P8_EN BIT(0) /* Enable tagging on port 8 */ +#define BRCM_HDR_P5_EN BIT(1) /* Enable tagging on port 5 */ + +/* Device ID register (8 or 32 bit) */ +#define B53_DEVICE_ID 0x30 + +/* Revision ID register (8 bit) */ +#define B53_REV_ID 0x40 + +/************************************************************************* + * ARL Access Page Registers + *************************************************************************/ + +/* VLAN Table Access Register (8 bit) */ +#define B53_VT_ACCESS 0x80 +#define B53_VT_ACCESS_9798 0x60 /* for BCM5397/BCM5398 */ +#define B53_VT_ACCESS_63XX 0x60 /* for BCM6328/62/68 */ +#define VTA_CMD_WRITE 0 +#define VTA_CMD_READ 1 +#define VTA_CMD_CLEAR 2 +#define VTA_START_CMD BIT(7) + +/* VLAN Table Index Register (16 bit) */ +#define B53_VT_INDEX 0x81 +#define B53_VT_INDEX_9798 0x61 +#define B53_VT_INDEX_63XX 0x62 + +/* VLAN Table Entry Register (32 bit) */ +#define B53_VT_ENTRY 0x83 +#define B53_VT_ENTRY_9798 0x63 +#define B53_VT_ENTRY_63XX 0x64 +#define VTE_MEMBERS 0x1ff +#define VTE_UNTAG_S 9 +#define VTE_UNTAG (0x1ff << 9) + +/************************************************************************* + * Port VLAN Registers + *************************************************************************/ + +/* Port VLAN mask (16 bit) IMP port is always 8, also on 5325 & co */ +#define B53_PVLAN_PORT_MASK(i) ((i) * 2) + +/************************************************************************* + * 802.1Q Page Registers + *************************************************************************/ + +/* Global QoS Control (8 bit) */ +#define B53_QOS_GLOBAL_CTL 0x00 + +/* Enable 802.1Q for individual Ports (16 bit) */ +#define B53_802_1P_EN 0x04 + +/************************************************************************* + * VLAN Page Registers + *************************************************************************/ + +/* VLAN Control 0 (8 bit) */ +#define B53_VLAN_CTRL0 0x00 +#define VC0_8021PF_CTRL_MASK 0x3 +#define VC0_8021PF_CTRL_NONE 0x0 +#define VC0_8021PF_CTRL_CHANGE_PRI 0x1 +#define VC0_8021PF_CTRL_CHANGE_VID 0x2 +#define VC0_8021PF_CTRL_CHANGE_BOTH 0x3 +#define VC0_8021QF_CTRL_MASK 0xc +#define VC0_8021QF_CTRL_CHANGE_PRI 0x1 +#define VC0_8021QF_CTRL_CHANGE_VID 0x2 +#define VC0_8021QF_CTRL_CHANGE_BOTH 0x3 +#define VC0_RESERVED_1 BIT(1) +#define VC0_DROP_VID_MISS BIT(4) +#define VC0_VID_HASH_VID BIT(5) +#define VC0_VID_CHK_EN BIT(6) /* Use VID,DA or VID,SA */ +#define VC0_VLAN_EN BIT(7) /* 802.1Q VLAN Enabled */ + +/* VLAN Control 1 (8 bit) */ +#define B53_VLAN_CTRL1 0x01 +#define VC1_RX_MCST_TAG_EN BIT(1) +#define VC1_RX_MCST_FWD_EN BIT(2) +#define VC1_RX_MCST_UNTAG_EN BIT(3) + +/* VLAN Control 2 (8 bit) */ +#define B53_VLAN_CTRL2 0x02 + +/* VLAN Control 3 (8 bit when BCM5325, 16 bit else) */ +#define B53_VLAN_CTRL3 0x03 +#define B53_VLAN_CTRL3_63XX 0x04 +#define VC3_MAXSIZE_1532 BIT(6) /* 5325 only */ +#define VC3_HIGH_8BIT_EN BIT(7) /* 5325 only */ + +/* VLAN Control 4 (8 bit) */ +#define B53_VLAN_CTRL4 0x05 +#define B53_VLAN_CTRL4_25 0x04 +#define B53_VLAN_CTRL4_63XX 0x06 +#define VC4_ING_VID_CHECK_S 6 +#define VC4_ING_VID_CHECK_MASK (0x3 << VC4_ING_VID_CHECK_S) +#define VC4_ING_VID_VIO_FWD 0 /* forward, but do not learn */ +#define VC4_ING_VID_VIO_DROP 1 /* drop VID violations */ +#define VC4_NO_ING_VID_CHK 2 /* do not check */ +#define VC4_ING_VID_VIO_TO_IMP 3 /* redirect to MII port */ + +/* VLAN Control 5 (8 bit) */ +#define B53_VLAN_CTRL5 0x06 +#define B53_VLAN_CTRL5_25 0x05 +#define B53_VLAN_CTRL5_63XX 0x07 +#define VC5_VID_FFF_EN BIT(2) +#define VC5_DROP_VTABLE_MISS BIT(3) + +/* VLAN Control 6 (8 bit) */ +#define B53_VLAN_CTRL6 0x07 +#define B53_VLAN_CTRL6_63XX 0x08 + +/* VLAN Table Access Register (16 bit) */ +#define B53_VLAN_TABLE_ACCESS_25 0x06 /* BCM5325E/5350 */ +#define B53_VLAN_TABLE_ACCESS_65 0x08 /* BCM5365 */ +#define VTA_VID_LOW_MASK_25 0xf +#define VTA_VID_LOW_MASK_65 0xff +#define VTA_VID_HIGH_S_25 4 +#define VTA_VID_HIGH_S_65 8 +#define VTA_VID_HIGH_MASK_25 (0xff << VTA_VID_HIGH_S_25E) +#define VTA_VID_HIGH_MASK_65 (0xf << VTA_VID_HIGH_S_65) +#define VTA_RW_STATE BIT(12) +#define VTA_RW_STATE_RD 0 +#define VTA_RW_STATE_WR BIT(12) +#define VTA_RW_OP_EN BIT(13) + +/* VLAN Read/Write Registers for (16/32 bit) */ +#define B53_VLAN_WRITE_25 0x08 +#define B53_VLAN_WRITE_65 0x0a +#define B53_VLAN_READ 0x0c +#define VA_MEMBER_MASK 0x3f +#define VA_UNTAG_S_25 6 +#define VA_UNTAG_MASK_25 0x3f +#define VA_UNTAG_S_65 7 +#define VA_UNTAG_MASK_65 0x1f +#define VA_VID_HIGH_S 12 +#define VA_VID_HIGH_MASK (0xffff << VA_VID_HIGH_S) +#define VA_VALID_25 BIT(20) +#define VA_VALID_25_R4 BIT(24) +#define VA_VALID_65 BIT(14) + +/* VLAN Port Default Tag (16 bit) */ +#define B53_VLAN_PORT_DEF_TAG(i) (0x10 + 2 * (i)) + +/************************************************************************* + * Jumbo Frame Page Registers + *************************************************************************/ + +/* Jumbo Enable Port Mask (bit i == port i enabled) (32 bit) */ +#define B53_JUMBO_PORT_MASK 0x01 +#define B53_JUMBO_PORT_MASK_63XX 0x04 +#define JPM_10_100_JUMBO_EN BIT(24) /* GigE always enabled */ + +/* Good Frame Max Size without 802.1Q TAG (16 bit) */ +#define B53_JUMBO_MAX_SIZE 0x05 +#define B53_JUMBO_MAX_SIZE_63XX 0x08 +#define JMS_MIN_SIZE 1518 +#define JMS_MAX_SIZE 9724 + +/************************************************************************* + * CFP Configuration Page Registers + *************************************************************************/ + +/* CFP Control Register with ports map (8 bit) */ +#define B53_CFP_CTRL 0x00 + +#endif /* !__B53_REGS_H */ diff --git a/drivers/net/dsa/b53/b53_spi.c b/drivers/net/dsa/b53/b53_spi.c new file mode 100644 index 0000000..2bda0b5 --- /dev/null +++ b/drivers/net/dsa/b53/b53_spi.c @@ -0,0 +1,331 @@ +/* + * B53 register access through SPI + * + * Copyright (C) 2011-2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#include + +#include +#include +#include +#include +#include + +#include "b53_priv.h" + +#define B53_SPI_DATA 0xf0 + +#define B53_SPI_STATUS 0xfe +#define B53_SPI_CMD_SPIF BIT(7) +#define B53_SPI_CMD_RACK BIT(5) + +#define B53_SPI_CMD_READ 0x00 +#define B53_SPI_CMD_WRITE 0x01 +#define B53_SPI_CMD_NORMAL 0x60 +#define B53_SPI_CMD_FAST 0x10 + +#define B53_SPI_PAGE_SELECT 0xff + +static inline int b53_spi_read_reg(struct spi_device *spi, u8 reg, u8 *val, + unsigned int len) +{ + u8 txbuf[2]; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_READ; + txbuf[1] = reg; + + return spi_write_then_read(spi, txbuf, 2, val, len); +} + +static inline int b53_spi_clear_status(struct spi_device *spi) +{ + unsigned int i; + u8 rxbuf; + int ret; + + for (i = 0; i < 10; i++) { + ret = b53_spi_read_reg(spi, B53_SPI_STATUS, &rxbuf, 1); + if (ret) + return ret; + + if (!(rxbuf & B53_SPI_CMD_SPIF)) + break; + + mdelay(1); + } + + if (i == 10) + return -EIO; + + return 0; +} + +static inline int b53_spi_set_page(struct spi_device *spi, u8 page) +{ + u8 txbuf[3]; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = B53_SPI_PAGE_SELECT; + txbuf[2] = page; + + return spi_write(spi, txbuf, sizeof(txbuf)); +} + +static inline int b53_prepare_reg_access(struct spi_device *spi, u8 page) +{ + int ret = b53_spi_clear_status(spi); + + if (ret) + return ret; + + return b53_spi_set_page(spi, page); +} + +static int b53_spi_prepare_reg_read(struct spi_device *spi, u8 reg) +{ + u8 rxbuf; + int retry_count; + int ret; + + ret = b53_spi_read_reg(spi, reg, &rxbuf, 1); + if (ret) + return ret; + + for (retry_count = 0; retry_count < 10; retry_count++) { + ret = b53_spi_read_reg(spi, B53_SPI_STATUS, &rxbuf, 1); + if (ret) + return ret; + + if (rxbuf & B53_SPI_CMD_RACK) + break; + + mdelay(1); + } + + if (retry_count == 10) + return -EIO; + + return 0; +} + +static int b53_spi_read(struct b53_device *dev, u8 page, u8 reg, u8 *data, + unsigned int len) +{ + struct spi_device *spi = dev->priv; + int ret; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + ret = b53_spi_prepare_reg_read(spi, reg); + if (ret) + return ret; + + return b53_spi_read_reg(spi, B53_SPI_DATA, data, len); +} + +static int b53_spi_read8(struct b53_device *dev, u8 page, u8 reg, u8 *val) +{ + return b53_spi_read(dev, page, reg, val, 1); +} + +static int b53_spi_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) +{ + int ret = b53_spi_read(dev, page, reg, (u8 *)val, 2); + + if (!ret) + *val = le16_to_cpu(*val); + + return ret; +} + +static int b53_spi_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) +{ + int ret = b53_spi_read(dev, page, reg, (u8 *)val, 4); + + if (!ret) + *val = le32_to_cpu(*val); + + return ret; +} + +static int b53_spi_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + int ret; + + *val = 0; + ret = b53_spi_read(dev, page, reg, (u8 *)val, 6); + if (!ret) + *val = le64_to_cpu(*val); + + return ret; +} + +static int b53_spi_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + int ret = b53_spi_read(dev, page, reg, (u8 *)val, 8); + + if (!ret) + *val = le64_to_cpu(*val); + + return ret; +} + +static int b53_spi_write8(struct b53_device *dev, u8 page, u8 reg, u8 value) +{ + struct spi_device *spi = dev->priv; + int ret; + u8 txbuf[3]; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = reg; + txbuf[2] = value; + + return spi_write(spi, txbuf, sizeof(txbuf)); +} + +static int b53_spi_write16(struct b53_device *dev, u8 page, u8 reg, u16 value) +{ + struct spi_device *spi = dev->priv; + int ret; + u8 txbuf[4]; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = reg; + put_unaligned_le16(value, &txbuf[2]); + + return spi_write(spi, txbuf, sizeof(txbuf)); +} + +static int b53_spi_write32(struct b53_device *dev, u8 page, u8 reg, u32 value) +{ + struct spi_device *spi = dev->priv; + int ret; + u8 txbuf[6]; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = reg; + put_unaligned_le32(value, &txbuf[2]); + + return spi_write(spi, txbuf, sizeof(txbuf)); +} + +static int b53_spi_write48(struct b53_device *dev, u8 page, u8 reg, u64 value) +{ + struct spi_device *spi = dev->priv; + int ret; + u8 txbuf[10]; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = reg; + put_unaligned_le64(value, &txbuf[2]); + + return spi_write(spi, txbuf, sizeof(txbuf) - 2); +} + +static int b53_spi_write64(struct b53_device *dev, u8 page, u8 reg, u64 value) +{ + struct spi_device *spi = dev->priv; + int ret; + u8 txbuf[10]; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = reg; + put_unaligned_le64(value, &txbuf[2]); + + return spi_write(spi, txbuf, sizeof(txbuf)); +} + +static struct b53_io_ops b53_spi_ops = { + .read8 = b53_spi_read8, + .read16 = b53_spi_read16, + .read32 = b53_spi_read32, + .read48 = b53_spi_read48, + .read64 = b53_spi_read64, + .write8 = b53_spi_write8, + .write16 = b53_spi_write16, + .write32 = b53_spi_write32, + .write48 = b53_spi_write48, + .write64 = b53_spi_write64, +}; + +static int b53_spi_probe(struct spi_device *spi) +{ + struct b53_device *dev; + int ret; + + dev = b53_switch_alloc(&spi->dev, &b53_spi_ops, spi); + if (!dev) + return -ENOMEM; + + if (spi->dev.platform_data) + dev->pdata = spi->dev.platform_data; + + ret = b53_switch_register(dev); + if (ret) + return ret; + + spi_set_drvdata(spi, dev); + + return 0; +} + +static int b53_spi_remove(struct spi_device *spi) +{ + struct b53_device *dev = spi_get_drvdata(spi); + + if (dev) + b53_switch_remove(dev); + + return 0; +} + +static struct spi_driver b53_spi_driver = { + .driver = { + .name = "b53-switch", + .bus = &spi_bus_type, + .owner = THIS_MODULE, + }, + .probe = b53_spi_probe, + .remove = b53_spi_remove, +}; + +module_spi_driver(b53_spi_driver); + +MODULE_AUTHOR("Jonas Gorski "); +MODULE_DESCRIPTION("B53 SPI access driver"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/dsa/b53/b53_srab.c b/drivers/net/dsa/b53/b53_srab.c new file mode 100644 index 0000000..70fd472 --- /dev/null +++ b/drivers/net/dsa/b53/b53_srab.c @@ -0,0 +1,415 @@ +/* + * B53 register access through Switch Register Access Bridge Registers + * + * Copyright (C) 2013 Hauke Mehrtens + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#include +#include +#include +#include +#include + +#include "b53_priv.h" + +/* command and status register of the SRAB */ +#define B53_SRAB_CMDSTAT 0x2c +#define B53_SRAB_CMDSTAT_RST BIT(2) +#define B53_SRAB_CMDSTAT_WRITE BIT(1) +#define B53_SRAB_CMDSTAT_GORDYN BIT(0) +#define B53_SRAB_CMDSTAT_PAGE 24 +#define B53_SRAB_CMDSTAT_REG 16 + +/* high order word of write data to switch registe */ +#define B53_SRAB_WD_H 0x30 + +/* low order word of write data to switch registe */ +#define B53_SRAB_WD_L 0x34 + +/* high order word of read data from switch register */ +#define B53_SRAB_RD_H 0x38 + +/* low order word of read data from switch register */ +#define B53_SRAB_RD_L 0x3c + +/* command and status register of the SRAB */ +#define B53_SRAB_CTRLS 0x40 +#define B53_SRAB_CTRLS_RCAREQ BIT(3) +#define B53_SRAB_CTRLS_RCAGNT BIT(4) +#define B53_SRAB_CTRLS_SW_INIT_DONE BIT(6) + +/* the register captures interrupt pulses from the switch */ +#define B53_SRAB_INTR 0x44 +#define B53_SRAB_INTR_P(x) BIT(x) +#define B53_SRAB_SWITCH_PHY BIT(8) +#define B53_SRAB_1588_SYNC BIT(9) +#define B53_SRAB_IMP1_SLEEP_TIMER BIT(10) +#define B53_SRAB_P7_SLEEP_TIMER BIT(11) +#define B53_SRAB_IMP0_SLEEP_TIMER BIT(12) + +struct b53_srab_priv { + void __iomem *regs; +}; + +static int b53_srab_request_grant(struct b53_device *dev) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + u32 ctrls; + int i; + + ctrls = readl(regs + B53_SRAB_CTRLS); + ctrls |= B53_SRAB_CTRLS_RCAREQ; + writel(ctrls, regs + B53_SRAB_CTRLS); + + for (i = 0; i < 20; i++) { + ctrls = readl(regs + B53_SRAB_CTRLS); + if (ctrls & B53_SRAB_CTRLS_RCAGNT) + break; + usleep_range(10, 100); + } + if (WARN_ON(i == 5)) + return -EIO; + + return 0; +} + +static void b53_srab_release_grant(struct b53_device *dev) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + u32 ctrls; + + ctrls = readl(regs + B53_SRAB_CTRLS); + ctrls &= ~B53_SRAB_CTRLS_RCAREQ; + writel(ctrls, regs + B53_SRAB_CTRLS); +} + +static int b53_srab_op(struct b53_device *dev, u8 page, u8 reg, u32 op) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int i; + u32 cmdstat; + + /* set register address */ + cmdstat = (page << B53_SRAB_CMDSTAT_PAGE) | + (reg << B53_SRAB_CMDSTAT_REG) | + B53_SRAB_CMDSTAT_GORDYN | + op; + writel(cmdstat, regs + B53_SRAB_CMDSTAT); + + /* check if operation completed */ + for (i = 0; i < 5; ++i) { + cmdstat = readl(regs + B53_SRAB_CMDSTAT); + if (!(cmdstat & B53_SRAB_CMDSTAT_GORDYN)) + break; + usleep_range(10, 100); + } + + if (WARN_ON(i == 5)) + return -EIO; + + return 0; +} + +static int b53_srab_read8(struct b53_device *dev, u8 page, u8 reg, u8 *val) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + ret = b53_srab_op(dev, page, reg, 0); + if (ret) + goto err; + + *val = readl(regs + B53_SRAB_RD_L) & 0xff; + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + ret = b53_srab_op(dev, page, reg, 0); + if (ret) + goto err; + + *val = readl(regs + B53_SRAB_RD_L) & 0xffff; + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + ret = b53_srab_op(dev, page, reg, 0); + if (ret) + goto err; + + *val = readl(regs + B53_SRAB_RD_L); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + ret = b53_srab_op(dev, page, reg, 0); + if (ret) + goto err; + + *val = readl(regs + B53_SRAB_RD_L); + *val += ((u64)readl(regs + B53_SRAB_RD_H) & 0xffff) << 32; + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + ret = b53_srab_op(dev, page, reg, 0); + if (ret) + goto err; + + *val = readl(regs + B53_SRAB_RD_L); + *val += (u64)readl(regs + B53_SRAB_RD_H) << 32; + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_write8(struct b53_device *dev, u8 page, u8 reg, u8 value) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + writel(value, regs + B53_SRAB_WD_L); + + ret = b53_srab_op(dev, page, reg, B53_SRAB_CMDSTAT_WRITE); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_write16(struct b53_device *dev, u8 page, u8 reg, + u16 value) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + writel(value, regs + B53_SRAB_WD_L); + + ret = b53_srab_op(dev, page, reg, B53_SRAB_CMDSTAT_WRITE); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_write32(struct b53_device *dev, u8 page, u8 reg, + u32 value) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + writel(value, regs + B53_SRAB_WD_L); + + ret = b53_srab_op(dev, page, reg, B53_SRAB_CMDSTAT_WRITE); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_write48(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + writel((u32)value, regs + B53_SRAB_WD_L); + writel((u16)(value >> 32), regs + B53_SRAB_WD_H); + + ret = b53_srab_op(dev, page, reg, B53_SRAB_CMDSTAT_WRITE); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_write64(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + writel((u32)value, regs + B53_SRAB_WD_L); + writel((u32)(value >> 32), regs + B53_SRAB_WD_H); + + ret = b53_srab_op(dev, page, reg, B53_SRAB_CMDSTAT_WRITE); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static struct b53_io_ops b53_srab_ops = { + .read8 = b53_srab_read8, + .read16 = b53_srab_read16, + .read32 = b53_srab_read32, + .read48 = b53_srab_read48, + .read64 = b53_srab_read64, + .write8 = b53_srab_write8, + .write16 = b53_srab_write16, + .write32 = b53_srab_write32, + .write48 = b53_srab_write48, + .write64 = b53_srab_write64, +}; + +static int b53_srab_probe(struct platform_device *pdev) +{ + struct b53_srab_priv *priv; + struct b53_device *dev; + struct resource *r; + + priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + r = platform_get_resource(pdev, IORESOURCE_MEM, 0); + priv->regs = devm_ioremap_resource(&pdev->dev, r); + if (IS_ERR(priv->regs)) + return -ENOMEM; + + dev = b53_switch_alloc(&pdev->dev, &b53_srab_ops, priv); + if (!dev) + return -ENOMEM; + + platform_set_drvdata(pdev, dev); + + return b53_switch_register(dev); +} + +static int b53_srab_remove(struct platform_device *pdev) +{ + struct b53_device *dev = platform_get_drvdata(pdev); + + if (dev) + b53_switch_remove(dev); + + return 0; +} + +static const struct of_device_id b53_srab_of_match[] = { + { .compatible = "brcm,bcm53010-srab" }, + { .compatible = "brcm,bcm53011-srab" }, + { .compatible = "brcm,bcm53012-srab" }, + { .compatible = "brcm,bcm53018-srab" }, + { .compatible = "brcm,bcm53019-srab" }, + { .compatible = "brcm,bcm5301x-srab" }, + { /* sentinel */ }, +}; + +static struct platform_driver b53_srab_driver = { + .probe = b53_srab_probe, + .remove = b53_srab_remove, + .driver = { + .name = "b53-srab-switch", + .of_match_table = b53_srab_of_match, + }, +}; + +module_platform_driver(b53_srab_driver); +MODULE_AUTHOR("Hauke Mehrtens "); +MODULE_DESCRIPTION("B53 Switch Register Access Bridge Registers (SRAB) access driver"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/include/linux/platform_data/b53.h b/include/linux/platform_data/b53.h new file mode 100644 index 0000000..69d279c --- /dev/null +++ b/include/linux/platform_data/b53.h @@ -0,0 +1,33 @@ +/* + * B53 platform data + * + * Copyright (C) 2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#ifndef __B53_H +#define __B53_H + +#include + +struct b53_platform_data { + u32 chip_id; + u16 enabled_ports; + + /* only used by MMAP'd driver */ + unsigned big_endian:1; + void __iomem *regs; +}; + +#endif -- cgit v0.10.2 From 0830c9802e33d8c94b9b2f89b3db4aae92e55bf6 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 9 Jun 2016 18:23:54 -0700 Subject: net: dsa: b53: Add BCM7445 quirk The Broadcom BCM7445 STB chip has an issued in its revision D0 which was previously worked around in drivers/net/dsa/bcm_sf2.c where we may end-up double programming the integrated BCM7445 switch (bcm_sf2) and an external Broadcom switch such as BCM53125, since these are mostly register compatible. Add a small quirk which just defers probing until we are sitting on the slave DSA MDIO bus, which will allow us to intercept reads/writes and funnel them through the SF2 internal MDIO master (which happens to disconnect its pseudo PHY). Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/b53/b53_mdio.c b/drivers/net/dsa/b53/b53_mdio.c index c6cf7cf..aa87c3f 100644 --- a/drivers/net/dsa/b53/b53_mdio.c +++ b/drivers/net/dsa/b53/b53_mdio.c @@ -316,6 +316,17 @@ static int b53_mdio_probe(struct mdio_device *mdiodev) return -ENODEV; } + /* First probe will come from SWITCH_MDIO controller on the 7445D0 + * switch, which will conflict with the 7445 integrated switch + * pseudo-phy (we end-up programming both). In that case, we return + * -EPROBE_DEFER for the first time we get here, and wait until we come + * back with the slave MDIO bus which has the correct indirection + * layer setup + */ + if (of_machine_is_compatible("brcm,bcm7445d0") && + strcmp(mdiodev->bus->name, "sf2 slave mii")) + return -EPROBE_DEFER; + dev = b53_switch_alloc(&mdiodev->dev, &b53_mdio_ops, mdiodev->bus); if (!dev) return -ENOMEM; -- cgit v0.10.2 From 1da6df85c6fbed8f39c56eadcae7fa75a7a0c635 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 9 Jun 2016 18:23:55 -0700 Subject: net: dsa: b53: Implement ARL add/del/dump operations Adds support for FDB add/delete/dump using the ARL read/write logic and the ARL search logic for faster dumps. The code is made flexible enough it could support devices with a different register layout like BCM5325 and BCM5365 which have fewer number of entries or pack values into a single 64 bits register. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 6f0337d..a9f1de4 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -26,7 +26,9 @@ #include #include #include +#include #include +#include #include "b53_regs.h" #include "b53_priv.h" @@ -777,6 +779,246 @@ static void b53_adjust_link(struct dsa_switch *ds, int port, } } +/* Address Resolution Logic routines */ +static int b53_arl_op_wait(struct b53_device *dev) +{ + unsigned int timeout = 10; + u8 reg; + + do { + b53_read8(dev, B53_ARLIO_PAGE, B53_ARLTBL_RW_CTRL, ®); + if (!(reg & ARLTBL_START_DONE)) + return 0; + + usleep_range(1000, 2000); + } while (timeout--); + + dev_warn(dev->dev, "timeout waiting for ARL to finish: 0x%02x\n", reg); + + return -ETIMEDOUT; +} + +static int b53_arl_rw_op(struct b53_device *dev, unsigned int op) +{ + u8 reg; + + if (op > ARLTBL_RW) + return -EINVAL; + + b53_read8(dev, B53_ARLIO_PAGE, B53_ARLTBL_RW_CTRL, ®); + reg |= ARLTBL_START_DONE; + if (op) + reg |= ARLTBL_RW; + else + reg &= ~ARLTBL_RW; + b53_write8(dev, B53_ARLIO_PAGE, B53_ARLTBL_RW_CTRL, reg); + + return b53_arl_op_wait(dev); +} + +static int b53_arl_read(struct b53_device *dev, u64 mac, + u16 vid, struct b53_arl_entry *ent, u8 *idx, + bool is_valid) +{ + unsigned int i; + int ret; + + ret = b53_arl_op_wait(dev); + if (ret) + return ret; + + /* Read the bins */ + for (i = 0; i < dev->num_arl_entries; i++) { + u64 mac_vid; + u32 fwd_entry; + + b53_read64(dev, B53_ARLIO_PAGE, + B53_ARLTBL_MAC_VID_ENTRY(i), &mac_vid); + b53_read32(dev, B53_ARLIO_PAGE, + B53_ARLTBL_DATA_ENTRY(i), &fwd_entry); + b53_arl_to_entry(ent, mac_vid, fwd_entry); + + if (!(fwd_entry & ARLTBL_VALID)) + continue; + if ((mac_vid & ARLTBL_MAC_MASK) != mac) + continue; + *idx = i; + } + + return -ENOENT; +} + +static int b53_arl_op(struct b53_device *dev, int op, int port, + const unsigned char *addr, u16 vid, bool is_valid) +{ + struct b53_arl_entry ent; + u32 fwd_entry; + u64 mac, mac_vid = 0; + u8 idx = 0; + int ret; + + /* Convert the array into a 64-bit MAC */ + mac = b53_mac_to_u64(addr); + + /* Perform a read for the given MAC and VID */ + b53_write48(dev, B53_ARLIO_PAGE, B53_MAC_ADDR_IDX, mac); + b53_write16(dev, B53_ARLIO_PAGE, B53_VLAN_ID_IDX, vid); + + /* Issue a read operation for this MAC */ + ret = b53_arl_rw_op(dev, 1); + if (ret) + return ret; + + ret = b53_arl_read(dev, mac, vid, &ent, &idx, is_valid); + /* If this is a read, just finish now */ + if (op) + return ret; + + /* We could not find a matching MAC, so reset to a new entry */ + if (ret) { + fwd_entry = 0; + idx = 1; + } + + memset(&ent, 0, sizeof(ent)); + ent.port = port; + ent.is_valid = is_valid; + ent.vid = vid; + ent.is_static = true; + memcpy(ent.mac, addr, ETH_ALEN); + b53_arl_from_entry(&mac_vid, &fwd_entry, &ent); + + b53_write64(dev, B53_ARLIO_PAGE, + B53_ARLTBL_MAC_VID_ENTRY(idx), mac_vid); + b53_write32(dev, B53_ARLIO_PAGE, + B53_ARLTBL_DATA_ENTRY(idx), fwd_entry); + + return b53_arl_rw_op(dev, 0); +} + +static int b53_fdb_prepare(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_fdb *fdb, + struct switchdev_trans *trans) +{ + struct b53_device *priv = ds_to_priv(ds); + + /* 5325 and 5365 require some more massaging, but could + * be supported eventually + */ + if (is5325(priv) || is5365(priv)) + return -EOPNOTSUPP; + + return 0; +} + +static void b53_fdb_add(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_fdb *fdb, + struct switchdev_trans *trans) +{ + struct b53_device *priv = ds_to_priv(ds); + + if (b53_arl_op(priv, 0, port, fdb->addr, fdb->vid, true)) + pr_err("%s: failed to add MAC address\n", __func__); +} + +static int b53_fdb_del(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_fdb *fdb) +{ + struct b53_device *priv = ds_to_priv(ds); + + return b53_arl_op(priv, 0, port, fdb->addr, fdb->vid, false); +} + +static int b53_arl_search_wait(struct b53_device *dev) +{ + unsigned int timeout = 1000; + u8 reg; + + do { + b53_read8(dev, B53_ARLIO_PAGE, B53_ARL_SRCH_CTL, ®); + if (!(reg & ARL_SRCH_STDN)) + return 0; + + if (reg & ARL_SRCH_VLID) + return 0; + + usleep_range(1000, 2000); + } while (timeout--); + + return -ETIMEDOUT; +} + +static void b53_arl_search_rd(struct b53_device *dev, u8 idx, + struct b53_arl_entry *ent) +{ + u64 mac_vid; + u32 fwd_entry; + + b53_read64(dev, B53_ARLIO_PAGE, + B53_ARL_SRCH_RSTL_MACVID(idx), &mac_vid); + b53_read32(dev, B53_ARLIO_PAGE, + B53_ARL_SRCH_RSTL(idx), &fwd_entry); + b53_arl_to_entry(ent, mac_vid, fwd_entry); +} + +static int b53_fdb_copy(struct net_device *dev, int port, + const struct b53_arl_entry *ent, + struct switchdev_obj_port_fdb *fdb, + int (*cb)(struct switchdev_obj *obj)) +{ + if (!ent->is_valid) + return 0; + + if (port != ent->port) + return 0; + + ether_addr_copy(fdb->addr, ent->mac); + fdb->vid = ent->vid; + fdb->ndm_state = ent->is_static ? NUD_NOARP : NUD_REACHABLE; + + return cb(&fdb->obj); +} + +static int b53_fdb_dump(struct dsa_switch *ds, int port, + struct switchdev_obj_port_fdb *fdb, + int (*cb)(struct switchdev_obj *obj)) +{ + struct b53_device *priv = ds_to_priv(ds); + struct net_device *dev = ds->ports[port].netdev; + struct b53_arl_entry results[2]; + unsigned int count = 0; + int ret; + u8 reg; + + /* Start search operation */ + reg = ARL_SRCH_STDN; + b53_write8(priv, B53_ARLIO_PAGE, B53_ARL_SRCH_CTL, reg); + + do { + ret = b53_arl_search_wait(priv); + if (ret) + return ret; + + b53_arl_search_rd(priv, 0, &results[0]); + ret = b53_fdb_copy(dev, port, &results[0], fdb, cb); + if (ret) + return ret; + + if (priv->num_arl_entries > 2) { + b53_arl_search_rd(priv, 1, &results[1]); + ret = b53_fdb_copy(dev, port, &results[1], fdb, cb); + if (ret) + return ret; + + if (!results[0].is_valid && !results[1].is_valid) + break; + } + + } while (count++ < 1024); + + return 0; +} + static struct dsa_switch_driver b53_switch_ops = { .tag_protocol = DSA_TAG_PROTO_NONE, .setup = b53_setup, @@ -789,6 +1031,10 @@ static struct dsa_switch_driver b53_switch_ops = { .adjust_link = b53_adjust_link, .port_enable = b53_enable_port, .port_disable = b53_disable_port, + .port_fdb_prepare = b53_fdb_prepare, + .port_fdb_dump = b53_fdb_dump, + .port_fdb_add = b53_fdb_add, + .port_fdb_del = b53_fdb_del, }; struct b53_chip_data { @@ -798,6 +1044,7 @@ struct b53_chip_data { u16 enabled_ports; u8 cpu_port; u8 vta_regs[3]; + u8 arl_entries; u8 duplex_reg; u8 jumbo_pm_reg; u8 jumbo_size_reg; @@ -816,6 +1063,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM5325", .vlans = 16, .enabled_ports = 0x1f, + .arl_entries = 2, .cpu_port = B53_CPU_PORT_25, .duplex_reg = B53_DUPLEX_STAT_FE, }, @@ -824,6 +1072,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM5365", .vlans = 256, .enabled_ports = 0x1f, + .arl_entries = 2, .cpu_port = B53_CPU_PORT_25, .duplex_reg = B53_DUPLEX_STAT_FE, }, @@ -832,6 +1081,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM5395", .vlans = 4096, .enabled_ports = 0x1f, + .arl_entries = 4, .cpu_port = B53_CPU_PORT, .vta_regs = B53_VTA_REGS, .duplex_reg = B53_DUPLEX_STAT_GE, @@ -843,6 +1093,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM5397", .vlans = 4096, .enabled_ports = 0x1f, + .arl_entries = 4, .cpu_port = B53_CPU_PORT, .vta_regs = B53_VTA_REGS_9798, .duplex_reg = B53_DUPLEX_STAT_GE, @@ -854,6 +1105,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM5398", .vlans = 4096, .enabled_ports = 0x7f, + .arl_entries = 4, .cpu_port = B53_CPU_PORT, .vta_regs = B53_VTA_REGS_9798, .duplex_reg = B53_DUPLEX_STAT_GE, @@ -865,6 +1117,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM53115", .vlans = 4096, .enabled_ports = 0x1f, + .arl_entries = 4, .vta_regs = B53_VTA_REGS, .cpu_port = B53_CPU_PORT, .duplex_reg = B53_DUPLEX_STAT_GE, @@ -887,6 +1140,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM53128", .vlans = 4096, .enabled_ports = 0x1ff, + .arl_entries = 4, .cpu_port = B53_CPU_PORT, .vta_regs = B53_VTA_REGS, .duplex_reg = B53_DUPLEX_STAT_GE, @@ -898,6 +1152,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM63xx", .vlans = 4096, .enabled_ports = 0, /* pdata must provide them */ + .arl_entries = 4, .cpu_port = B53_CPU_PORT, .vta_regs = B53_VTA_REGS_63XX, .duplex_reg = B53_DUPLEX_STAT_63XX, @@ -909,6 +1164,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM53010", .vlans = 4096, .enabled_ports = 0x1f, + .arl_entries = 4, .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ .vta_regs = B53_VTA_REGS, .duplex_reg = B53_DUPLEX_STAT_GE, @@ -920,6 +1176,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM53011", .vlans = 4096, .enabled_ports = 0x1bf, + .arl_entries = 4, .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ .vta_regs = B53_VTA_REGS, .duplex_reg = B53_DUPLEX_STAT_GE, @@ -931,6 +1188,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM53012", .vlans = 4096, .enabled_ports = 0x1bf, + .arl_entries = 4, .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ .vta_regs = B53_VTA_REGS, .duplex_reg = B53_DUPLEX_STAT_GE, @@ -942,6 +1200,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM53018", .vlans = 4096, .enabled_ports = 0x1f, + .arl_entries = 4, .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ .vta_regs = B53_VTA_REGS, .duplex_reg = B53_DUPLEX_STAT_GE, @@ -953,6 +1212,7 @@ static const struct b53_chip_data b53_switch_chips[] = { .dev_name = "BCM53019", .vlans = 4096, .enabled_ports = 0x1f, + .arl_entries = 4, .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ .vta_regs = B53_VTA_REGS, .duplex_reg = B53_DUPLEX_STAT_GE, @@ -982,6 +1242,7 @@ static int b53_switch_init(struct b53_device *dev) ds->drv = &b53_switch_ops; dev->cpu_port = chip->cpu_port; dev->num_vlans = chip->vlans; + dev->num_arl_entries = chip->arl_entries; break; } } diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index c484415..1ee4e80 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -24,6 +24,8 @@ #include #include +#include "b53_regs.h" + struct b53_device; struct b53_io_ops { @@ -81,6 +83,7 @@ struct b53_device { u8 jumbo_pm_reg; u8 jumbo_size_reg; int reset_gpio; + u8 num_arl_entries; /* used ports mask */ u16 enabled_ports; @@ -296,6 +299,60 @@ static inline int b53_write64(struct b53_device *dev, u8 page, u8 reg, return ret; } +struct b53_arl_entry { + u8 port; + u8 mac[ETH_ALEN]; + u16 vid; + u8 is_valid:1; + u8 is_age:1; + u8 is_static:1; +}; + +static inline void b53_mac_from_u64(u64 src, u8 *dst) +{ + unsigned int i; + + for (i = 0; i < ETH_ALEN; i++) + dst[ETH_ALEN - 1 - i] = (src >> (8 * i)) & 0xff; +} + +static inline u64 b53_mac_to_u64(const u8 *src) +{ + unsigned int i; + u64 dst = 0; + + for (i = 0; i < ETH_ALEN; i++) + dst |= (u64)src[ETH_ALEN - 1 - i] << (8 * i); + + return dst; +} + +static inline void b53_arl_to_entry(struct b53_arl_entry *ent, + u64 mac_vid, u32 fwd_entry) +{ + memset(ent, 0, sizeof(*ent)); + ent->port = fwd_entry & ARLTBL_DATA_PORT_ID_MASK; + ent->is_valid = !!(fwd_entry & ARLTBL_VALID); + ent->is_age = !!(fwd_entry & ARLTBL_AGE); + ent->is_static = !!(fwd_entry & ARLTBL_STATIC); + b53_mac_from_u64(mac_vid, ent->mac); + ent->vid = mac_vid >> ARLTBL_VID_S; +} + +static inline void b53_arl_from_entry(u64 *mac_vid, u32 *fwd_entry, + const struct b53_arl_entry *ent) +{ + *mac_vid = b53_mac_to_u64(ent->mac); + *mac_vid |= (u64)(ent->vid & ARLTBL_VID_MASK) << ARLTBL_VID_S; + *fwd_entry = ent->port & ARLTBL_DATA_PORT_ID_MASK; + if (ent->is_valid) + *fwd_entry |= ARLTBL_VALID; + if (ent->is_static) + *fwd_entry |= ARLTBL_STATIC; + if (ent->is_age) + *fwd_entry |= ARLTBL_AGE; +} + #ifdef CONFIG_BCM47XX #include diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h index ccf8af7..441d2b4 100644 --- a/drivers/net/dsa/b53/b53_regs.h +++ b/drivers/net/dsa/b53/b53_regs.h @@ -227,6 +227,70 @@ #define VTE_UNTAG (0x1ff << 9) /************************************************************************* + * ARL I/O Registers + *************************************************************************/ + +/* ARL Table Read/Write Register (8 bit) */ +#define B53_ARLTBL_RW_CTRL 0x00 +#define ARLTBL_RW BIT(0) +#define ARLTBL_START_DONE BIT(7) + +/* MAC Address Index Register (48 bit) */ +#define B53_MAC_ADDR_IDX 0x02 + +/* VLAN ID Index Register (16 bit) */ +#define B53_VLAN_ID_IDX 0x08 + +/* ARL Table MAC/VID Entry N Registers (64 bit) + * + * BCM5325 and BCM5365 share most definitions below + */ +#define B53_ARLTBL_MAC_VID_ENTRY(n) (0x10 * (n)) +#define ARLTBL_MAC_MASK 0xffffffffffff +#define ARLTBL_VID_S 48 +#define ARLTBL_VID_MASK_25 0xff +#define ARLTBL_VID_MASK 0xfff +#define ARLTBL_DATA_PORT_ID_S_25 48 +#define ARLTBL_DATA_PORT_ID_MASK_25 0xf +#define ARLTBL_AGE_25 BIT(61) +#define ARLTBL_STATIC_25 BIT(62) +#define ARLTBL_VALID_25 BIT(63) + +/* ARL Table Data Entry N Registers (32 bit) */ +#define B53_ARLTBL_DATA_ENTRY(n) ((0x10 * (n)) + 0x08) +#define ARLTBL_DATA_PORT_ID_MASK 0x1ff +#define ARLTBL_TC(tc) ((3 & tc) << 11) +#define ARLTBL_AGE BIT(14) +#define ARLTBL_STATIC BIT(15) +#define ARLTBL_VALID BIT(16) + +/* ARL Search Control Register (8 bit) */ +#define B53_ARL_SRCH_CTL 0x50 +#define B53_ARL_SRCH_CTL_25 0x20 +#define ARL_SRCH_VLID BIT(0) +#define ARL_SRCH_STDN BIT(7) + +/* ARL Search Address Register (16 bit) */ +#define B53_ARL_SRCH_ADDR 0x51 +#define B53_ARL_SRCH_ADDR_25 0x22 +#define B53_ARL_SRCH_ADDR_65 0x24 +#define ARL_ADDR_MASK GENMASK(14, 0) + +/* ARL Search MAC/VID Result (64 bit) */ +#define B53_ARL_SRCH_RSTL_0_MACVID 0x60 + +/* Single register search result on 5325 */ +#define B53_ARL_SRCH_RSTL_0_MACVID_25 0x24 +/* Single register search result on 5365 */ +#define B53_ARL_SRCH_RSTL_0_MACVID_65 0x30 + +/* ARL Search Data Result (32 bit) */ +#define B53_ARL_SRCH_RSTL_0 0x68 + +#define B53_ARL_SRCH_RSTL_MACVID(x) (B53_ARL_SRCH_RSTL_0_MACVID + ((x) * 0x10)) +#define B53_ARL_SRCH_RSTL(x) (B53_ARL_SRCH_RSTL_0 + ((x) * 0x10)) + +/************************************************************************* * Port VLAN Registers *************************************************************************/ -- cgit v0.10.2 From ff39c2d68679c8d2d07d0915d6ddbdb6fb26837c Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 9 Jun 2016 18:23:56 -0700 Subject: net: dsa: b53: Add bridge support Add support for HW bridging by tying the ports together in the same port VLAN mask when they belong to the same bridge, and isolating them to be alone with the CPU port when they are not. Propagate STP states from the bridge layer to the switch's HW mapping when requested. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index a9f1de4..ad1d682 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include @@ -339,12 +340,12 @@ static int b53_set_jumbo(struct b53_device *dev, bool enable, bool allow_10_100) return b53_write16(dev, B53_JUMBO_PAGE, dev->jumbo_size_reg, max_size); } -static int b53_flush_arl(struct b53_device *dev) +static int b53_flush_arl(struct b53_device *dev, u8 mask) { unsigned int i; b53_write8(dev, B53_CTRL_PAGE, B53_FAST_AGE_CTRL, - FAST_AGE_DONE | FAST_AGE_DYNAMIC | FAST_AGE_STATIC); + FAST_AGE_DONE | FAST_AGE_DYNAMIC | mask); for (i = 0; i < 10; i++) { u8 fast_age_ctrl; @@ -365,14 +366,52 @@ out: return 0; } +static int b53_fast_age_port(struct b53_device *dev, int port) +{ + b53_write8(dev, B53_CTRL_PAGE, B53_FAST_AGE_PORT_CTRL, port); + + return b53_flush_arl(dev, FAST_AGE_PORT); +} + +static void b53_imp_vlan_setup(struct dsa_switch *ds, int cpu_port) +{ + struct b53_device *dev = ds_to_priv(ds); + unsigned int i; + u16 pvlan; + + /* Enable the IMP port to be in the same VLAN as the other ports + * on a per-port basis such that we only have Port i and IMP in + * the same VLAN. + */ + b53_for_each_port(dev, i) { + b53_read16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(i), &pvlan); + pvlan |= BIT(cpu_port); + b53_write16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(i), pvlan); + } +} + static int b53_enable_port(struct dsa_switch *ds, int port, struct phy_device *phy) { struct b53_device *dev = ds_to_priv(ds); + unsigned int cpu_port = dev->cpu_port; + u16 pvlan; /* Clear the Rx and Tx disable bits and set to no spanning tree */ b53_write8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), 0); + /* Set this port, and only this one to be in the default VLAN, + * if member of a bridge, restore its membership prior to + * bringing down this port. + */ + b53_read16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), &pvlan); + pvlan &= ~0x1ff; + pvlan |= BIT(port); + pvlan |= dev->ports[port].vlan_ctl_mask; + b53_write16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), pvlan); + + b53_imp_vlan_setup(ds, cpu_port); + return 0; } @@ -482,7 +521,7 @@ static int b53_switch_reset(struct b53_device *dev) b53_enable_mib(dev); - return b53_flush_arl(dev); + return b53_flush_arl(dev, FAST_AGE_STATIC); } static int b53_phy_read16(struct dsa_switch *ds, int addr, int reg) @@ -1019,6 +1058,120 @@ static int b53_fdb_dump(struct dsa_switch *ds, int port, return 0; } +static int b53_br_join(struct dsa_switch *ds, int port, + struct net_device *bridge) +{ + struct b53_device *dev = ds_to_priv(ds); + u16 pvlan, reg; + unsigned int i; + + dev->ports[port].bridge_dev = bridge; + b53_read16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), &pvlan); + + b53_for_each_port(dev, i) { + if (dev->ports[i].bridge_dev != bridge) + continue; + + /* Add this local port to the remote port VLAN control + * membership and update the remote port bitmask + */ + b53_read16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(i), ®); + reg |= BIT(port); + b53_write16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(i), reg); + dev->ports[i].vlan_ctl_mask = reg; + + pvlan |= BIT(i); + } + + /* Configure the local port VLAN control membership to include + * remote ports and update the local port bitmask + */ + b53_write16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), pvlan); + dev->ports[port].vlan_ctl_mask = pvlan; + + return 0; +} + +static void b53_br_leave(struct dsa_switch *ds, int port) +{ + struct b53_device *dev = ds_to_priv(ds); + struct net_device *bridge = dev->ports[port].bridge_dev; + unsigned int i; + u16 pvlan, reg; + + b53_read16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), &pvlan); + + b53_for_each_port(dev, i) { + /* Don't touch the remaining ports */ + if (dev->ports[i].bridge_dev != bridge) + continue; + + b53_read16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(i), ®); + reg &= ~BIT(port); + b53_write16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(i), reg); + dev->ports[port].vlan_ctl_mask = reg; + + /* Prevent self removal to preserve isolation */ + if (port != i) + pvlan &= ~BIT(i); + } + + b53_write16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), pvlan); + dev->ports[port].vlan_ctl_mask = pvlan; + dev->ports[port].bridge_dev = NULL; +} + +static void b53_br_set_stp_state(struct dsa_switch *ds, int port, + u8 state) +{ + struct b53_device *dev = ds_to_priv(ds); + u8 hw_state, cur_hw_state; + u8 reg; + + b53_read8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), ®); + cur_hw_state = reg & PORT_CTRL_STP_STATE_MASK; + + switch (state) { + case BR_STATE_DISABLED: + hw_state = PORT_CTRL_DIS_STATE; + break; + case BR_STATE_LISTENING: + hw_state = PORT_CTRL_LISTEN_STATE; + break; + case BR_STATE_LEARNING: + hw_state = PORT_CTRL_LEARN_STATE; + break; + case BR_STATE_FORWARDING: + hw_state = PORT_CTRL_FWD_STATE; + break; + case BR_STATE_BLOCKING: + hw_state = PORT_CTRL_BLOCK_STATE; + break; + default: + dev_err(ds->dev, "invalid STP state: %d\n", state); + return; + } + + /* Fast-age ARL entries if we are moving a port from Learning or + * Forwarding (cur_hw_state) state to Disabled, Blocking or Listening + * state (hw_state) + */ + if (cur_hw_state != hw_state) { + if (cur_hw_state >= PORT_CTRL_LEARN_STATE && + hw_state <= PORT_CTRL_LISTEN_STATE) { + if (b53_fast_age_port(dev, port)) { + dev_err(ds->dev, "fast ageing failed\n"); + return; + } + } + } + + b53_read8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), ®); + reg &= ~PORT_CTRL_STP_STATE_MASK; + reg |= hw_state; + b53_write8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), reg); +} + static struct dsa_switch_driver b53_switch_ops = { .tag_protocol = DSA_TAG_PROTO_NONE, .setup = b53_setup, @@ -1031,6 +1184,9 @@ static struct dsa_switch_driver b53_switch_ops = { .adjust_link = b53_adjust_link, .port_enable = b53_enable_port, .port_disable = b53_disable_port, + .port_bridge_join = b53_br_join, + .port_bridge_leave = b53_br_leave, + .port_stp_state_set = b53_br_set_stp_state, .port_fdb_prepare = b53_fdb_prepare, .port_fdb_dump = b53_fdb_dump, .port_fdb_add = b53_fdb_add, diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index 1ee4e80..c198429 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -27,6 +27,7 @@ #include "b53_regs.h" struct b53_device; +struct net_device; struct b53_io_ops { int (*read8)(struct b53_device *dev, u8 page, u8 reg, u8 *value); @@ -64,6 +65,8 @@ enum { #define B53_N_PORTS_25 6 struct b53_port { + u16 vlan_ctl_mask; + struct net_device *bridge_dev; }; struct b53_device { diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h index 441d2b4..8f12bdd 100644 --- a/drivers/net/dsa/b53/b53_regs.h +++ b/drivers/net/dsa/b53/b53_regs.h @@ -65,6 +65,12 @@ #define PORT_CTRL_RX_MCST_EN BIT(3) /* Multicast RX (P8 only) */ #define PORT_CTRL_RX_UCST_EN BIT(4) /* Unicast RX (P8 only) */ #define PORT_CTRL_STP_STATE_S 5 +#define PORT_CTRL_NO_STP (0 << PORT_CTRL_STP_STATE_S) +#define PORT_CTRL_DIS_STATE (1 << PORT_CTRL_STP_STATE_S) +#define PORT_CTRL_BLOCK_STATE (2 << PORT_CTRL_STP_STATE_S) +#define PORT_CTRL_LISTEN_STATE (3 << PORT_CTRL_STP_STATE_S) +#define PORT_CTRL_LEARN_STATE (4 << PORT_CTRL_STP_STATE_S) +#define PORT_CTRL_FWD_STATE (5 << PORT_CTRL_STP_STATE_S) #define PORT_CTRL_STP_STATE_MASK (0x7 << PORT_CTRL_STP_STATE_S) /* SMP Control Register (8 bit) */ @@ -145,6 +151,12 @@ #define FAST_AGE_MC BIT(5) #define FAST_AGE_DONE BIT(7) +/* Fast Aging Port Control register (8 bit) */ +#define B53_FAST_AGE_PORT_CTRL 0x89 + +/* Fast Aging VID Control register (16 bit) */ +#define B53_FAST_AGE_VID_CTRL 0x8a + /************************************************************************* * Status Page registers *************************************************************************/ -- cgit v0.10.2 From a2482d2ce3498642d180b9d7453d0d9c7452cb29 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 9 Jun 2016 18:23:57 -0700 Subject: net: dsa: b53: Plug in VLAN support Add support for configuration VLANs on B53 devices by implementing the port VLAN add/del/dump functions. We currently default to a behavior which is equivalent to having VLAN filtering turned on, where all VLANs not programmed into the VLAN port-based vector will be discarded on ingress. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index ad1d682..5321083 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -186,15 +186,15 @@ static int b53_do_vlan_op(struct b53_device *dev, u8 op) return -EIO; } -static void b53_set_vlan_entry(struct b53_device *dev, u16 vid, u16 members, - u16 untag) +static void b53_set_vlan_entry(struct b53_device *dev, u16 vid, + struct b53_vlan *vlan) { if (is5325(dev)) { u32 entry = 0; - if (members) { - entry = ((untag & VA_UNTAG_MASK_25) << VA_UNTAG_S_25) | - members; + if (vlan->members) { + entry = ((vlan->untag & VA_UNTAG_MASK_25) << + VA_UNTAG_S_25) | vlan->members; if (dev->core_rev >= 3) entry |= VA_VALID_25_R4 | vid << VA_VID_HIGH_S; else @@ -207,9 +207,9 @@ static void b53_set_vlan_entry(struct b53_device *dev, u16 vid, u16 members, } else if (is5365(dev)) { u16 entry = 0; - if (members) - entry = ((untag & VA_UNTAG_MASK_65) << VA_UNTAG_S_65) | - members | VA_VALID_65; + if (vlan->members) + entry = ((vlan->untag & VA_UNTAG_MASK_65) << + VA_UNTAG_S_65) | vlan->members | VA_VALID_65; b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_WRITE_65, entry); b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_65, vid | @@ -217,13 +217,55 @@ static void b53_set_vlan_entry(struct b53_device *dev, u16 vid, u16 members, } else { b53_write16(dev, B53_ARLIO_PAGE, dev->vta_regs[1], vid); b53_write32(dev, B53_ARLIO_PAGE, dev->vta_regs[2], - (untag << VTE_UNTAG_S) | members); + (vlan->untag << VTE_UNTAG_S) | vlan->members); b53_do_vlan_op(dev, VTA_CMD_WRITE); } + + dev_dbg(dev->ds->dev, "VID: %d, members: 0x%04x, untag: 0x%04x\n", + vid, vlan->members, vlan->untag); +} + +static void b53_get_vlan_entry(struct b53_device *dev, u16 vid, + struct b53_vlan *vlan) +{ + if (is5325(dev)) { + u32 entry = 0; + + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_25, vid | + VTA_RW_STATE_RD | VTA_RW_OP_EN); + b53_read32(dev, B53_VLAN_PAGE, B53_VLAN_WRITE_25, &entry); + + if (dev->core_rev >= 3) + vlan->valid = !!(entry & VA_VALID_25_R4); + else + vlan->valid = !!(entry & VA_VALID_25); + vlan->members = entry & VA_MEMBER_MASK; + vlan->untag = (entry >> VA_UNTAG_S_25) & VA_UNTAG_MASK_25; + + } else if (is5365(dev)) { + u16 entry = 0; + + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_65, vid | + VTA_RW_STATE_WR | VTA_RW_OP_EN); + b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_WRITE_65, &entry); + + vlan->valid = !!(entry & VA_VALID_65); + vlan->members = entry & VA_MEMBER_MASK; + vlan->untag = (entry >> VA_UNTAG_S_65) & VA_UNTAG_MASK_65; + } else { + u32 entry = 0; + + b53_write16(dev, B53_ARLIO_PAGE, dev->vta_regs[1], vid); + b53_do_vlan_op(dev, VTA_CMD_READ); + b53_read32(dev, B53_ARLIO_PAGE, dev->vta_regs[2], &entry); + vlan->members = entry & VTE_MEMBERS; + vlan->untag = (entry >> VTE_UNTAG_S) & VTE_MEMBERS; + vlan->valid = true; + } } -void b53_set_forwarding(struct b53_device *dev, int enable) +static void b53_set_forwarding(struct b53_device *dev, int enable) { u8 mgmt; @@ -237,7 +279,7 @@ void b53_set_forwarding(struct b53_device *dev, int enable) b53_write8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, mgmt); } -static void b53_enable_vlan(struct b53_device *dev, int enable) +static void b53_enable_vlan(struct b53_device *dev, bool enable) { u8 mgmt, vc0, vc1, vc4 = 0, vc5; @@ -271,12 +313,6 @@ static void b53_enable_vlan(struct b53_device *dev, int enable) if (is5325(dev) || is5365(dev)) vc1 |= VC1_RX_MCST_TAG_EN; - if (!is5325(dev) && !is5365(dev)) { - if (dev->allow_vid_4095) - vc5 |= VC5_VID_FFF_EN; - else - vc5 &= ~VC5_VID_FFF_EN; - } } else { vc0 &= ~(VC0_VLAN_EN | VC0_VID_CHK_EN | VC0_VID_HASH_VID); vc1 &= ~(VC1_RX_MCST_UNTAG_EN | VC1_RX_MCST_FWD_EN); @@ -290,11 +326,11 @@ static void b53_enable_vlan(struct b53_device *dev, int enable) if (is5325(dev) || is5365(dev)) vc1 &= ~VC1_RX_MCST_TAG_EN; - - if (!is5325(dev) && !is5365(dev)) - vc5 &= ~VC5_VID_FFF_EN; } + if (!is5325(dev) && !is5365(dev)) + vc5 &= ~VC5_VID_FFF_EN; + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL0, vc0); b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL1, vc1); @@ -373,6 +409,13 @@ static int b53_fast_age_port(struct b53_device *dev, int port) return b53_flush_arl(dev, FAST_AGE_PORT); } +static int b53_fast_age_vlan(struct b53_device *dev, u16 vid) +{ + b53_write16(dev, B53_CTRL_PAGE, B53_FAST_AGE_VID_CTRL, vid); + + return b53_flush_arl(dev, FAST_AGE_VLAN); +} + static void b53_imp_vlan_setup(struct dsa_switch *ds, int cpu_port) { struct b53_device *dev = ds_to_priv(ds); @@ -453,12 +496,13 @@ static void b53_enable_mib(struct b53_device *dev) static int b53_configure_vlan(struct b53_device *dev) { + struct b53_vlan vl = { 0 }; int i; /* clear all vlan entries */ if (is5325(dev) || is5365(dev)) { for (i = 1; i < dev->num_vlans; i++) - b53_set_vlan_entry(dev, i, 0, 0); + b53_set_vlan_entry(dev, i, &vl); } else { b53_do_vlan_op(dev, VTA_CMD_CLEAR); } @@ -554,6 +598,7 @@ static int b53_reset_switch(struct b53_device *priv) /* reset vlans */ priv->enable_jumbo = false; + memset(priv->vlans, 0, sizeof(*priv->vlans) * priv->num_vlans); memset(priv->ports, 0, sizeof(*priv->ports) * priv->num_ports); return b53_switch_reset(priv); @@ -818,6 +863,151 @@ static void b53_adjust_link(struct dsa_switch *ds, int port, } } +static int b53_vlan_filtering(struct dsa_switch *ds, int port, + bool vlan_filtering) +{ + return 0; +} + +static int b53_vlan_prepare(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans) +{ + struct b53_device *dev = ds_to_priv(ds); + + if ((is5325(dev) || is5365(dev)) && vlan->vid_begin == 0) + return -EOPNOTSUPP; + + if (vlan->vid_end > dev->num_vlans) + return -ERANGE; + + b53_enable_vlan(dev, true); + + return 0; +} + +static void b53_vlan_add(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans) +{ + struct b53_device *dev = ds_to_priv(ds); + bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; + bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; + unsigned int cpu_port = dev->cpu_port; + struct b53_vlan *vl; + u16 vid; + + for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) { + vl = &dev->vlans[vid]; + + b53_get_vlan_entry(dev, vid, vl); + + vl->members |= BIT(port) | BIT(cpu_port); + if (untagged) + vl->untag |= BIT(port) | BIT(cpu_port); + else + vl->untag &= ~(BIT(port) | BIT(cpu_port)); + + b53_set_vlan_entry(dev, vid, vl); + b53_fast_age_vlan(dev, vid); + } + + if (pvid) { + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), + vlan->vid_end); + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(cpu_port), + vlan->vid_end); + b53_fast_age_vlan(dev, vid); + } +} + +static int b53_vlan_del(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan) +{ + struct b53_device *dev = ds_to_priv(ds); + bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; + unsigned int cpu_port = dev->cpu_port; + struct b53_vlan *vl; + u16 vid; + u16 pvid; + + b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), &pvid); + + for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) { + vl = &dev->vlans[vid]; + + b53_get_vlan_entry(dev, vid, vl); + + vl->members &= ~BIT(port); + if ((vl->members & BIT(cpu_port)) == BIT(cpu_port)) + vl->members = 0; + + if (pvid == vid) { + if (is5325(dev) || is5365(dev)) + pvid = 1; + else + pvid = 0; + } + + if (untagged) { + vl->untag &= ~(BIT(port)); + if ((vl->untag & BIT(cpu_port)) == BIT(cpu_port)) + vl->untag = 0; + } + + b53_set_vlan_entry(dev, vid, vl); + b53_fast_age_vlan(dev, vid); + } + + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), pvid); + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(cpu_port), pvid); + b53_fast_age_vlan(dev, pvid); + + return 0; +} + +static int b53_vlan_dump(struct dsa_switch *ds, int port, + struct switchdev_obj_port_vlan *vlan, + int (*cb)(struct switchdev_obj *obj)) +{ + struct b53_device *dev = ds_to_priv(ds); + u16 vid, vid_start = 0, pvid; + struct b53_vlan *vl; + int err = 0; + + if (is5325(dev) || is5365(dev)) + vid_start = 1; + + b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_PORT_DEF_TAG(port), &pvid); + + /* Use our software cache for dumps, since we do not have any HW + * operation returning only the used/valid VLANs + */ + for (vid = vid_start; vid < dev->num_vlans; vid++) { + vl = &dev->vlans[vid]; + + if (!vl->valid) + continue; + + if (!(vl->members & BIT(port))) + continue; + + vlan->vid_begin = vlan->vid_end = vid; + vlan->flags = 0; + + if (vl->untag & BIT(port)) + vlan->flags |= BRIDGE_VLAN_INFO_UNTAGGED; + if (pvid == vid) + vlan->flags |= BRIDGE_VLAN_INFO_PVID; + + err = cb(&vlan->obj); + if (err) + break; + } + + return err; +} + /* Address Resolution Logic routines */ static int b53_arl_op_wait(struct b53_device *dev) { @@ -1096,8 +1286,9 @@ static void b53_br_leave(struct dsa_switch *ds, int port) { struct b53_device *dev = ds_to_priv(ds); struct net_device *bridge = dev->ports[port].bridge_dev; + struct b53_vlan *vl = &dev->vlans[0]; unsigned int i; - u16 pvlan, reg; + u16 pvlan, reg, pvid; b53_read16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), &pvlan); @@ -1119,6 +1310,16 @@ static void b53_br_leave(struct dsa_switch *ds, int port) b53_write16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), pvlan); dev->ports[port].vlan_ctl_mask = pvlan; dev->ports[port].bridge_dev = NULL; + + if (is5325(dev) || is5365(dev)) + pvid = 1; + else + pvid = 0; + + b53_get_vlan_entry(dev, pvid, vl); + vl->members |= BIT(port) | BIT(dev->cpu_port); + vl->untag |= BIT(port) | BIT(dev->cpu_port); + b53_set_vlan_entry(dev, pvid, vl); } static void b53_br_set_stp_state(struct dsa_switch *ds, int port, @@ -1187,6 +1388,11 @@ static struct dsa_switch_driver b53_switch_ops = { .port_bridge_join = b53_br_join, .port_bridge_leave = b53_br_leave, .port_stp_state_set = b53_br_set_stp_state, + .port_vlan_filtering = b53_vlan_filtering, + .port_vlan_prepare = b53_vlan_prepare, + .port_vlan_add = b53_vlan_add, + .port_vlan_del = b53_vlan_del, + .port_vlan_dump = b53_vlan_dump, .port_fdb_prepare = b53_fdb_prepare, .port_fdb_dump = b53_fdb_dump, .port_fdb_add = b53_fdb_add, @@ -1446,6 +1652,12 @@ static int b53_switch_init(struct b53_device *dev) if (!dev->ports) return -ENOMEM; + dev->vlans = devm_kzalloc(dev->dev, + sizeof(struct b53_vlan) * dev->num_vlans, + GFP_KERNEL); + if (!dev->vlans) + return -ENOMEM; + dev->reset_gpio = b53_switch_get_reset_gpio(dev); if (dev->reset_gpio >= 0) { ret = devm_gpio_request_one(dev->dev, dev->reset_gpio, diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index c198429..5d8c602 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -69,6 +69,12 @@ struct b53_port { struct net_device *bridge_dev; }; +struct b53_vlan { + u16 members; + u16 untag; + bool valid; +}; + struct b53_device { struct dsa_switch *ds; struct b53_platform_data *pdata; @@ -99,14 +105,13 @@ struct b53_device { /* Master MDIO bus we got probed from */ struct mii_bus *bus; - /* Slave MDIO bus we created */ - struct mii_bus *slave_bus; void *priv; /* run time configuration */ - unsigned enable_jumbo:1; - unsigned allow_vid_4095:1; + bool enable_jumbo; + unsigned int num_vlans; + struct b53_vlan *vlans; unsigned int num_ports; struct b53_port *ports; }; -- cgit v0.10.2 From 7d71e994cdef8055edf2d4dc1b7490e439602859 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 3 Jun 2016 11:52:49 -0700 Subject: net/mlx4_en: mlx4_en_netpoll() should schedule TX, not RX MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit I am not sure mlx4_en_netpoll() is doing anything useful right now. mlx4 has different NAPI structures for RX and TX, and netpoll only wants to drain TX queues. Lets schedule NAPI polls on TX, not RX. Signed-off-by: Eric Dumazet Cc: Maciej Żenczykowski Cc: Eric W. Biederman Acked-by: Tariq Toukan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 19ceced..973391b 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -1197,8 +1197,8 @@ static void mlx4_en_netpoll(struct net_device *dev) struct mlx4_en_cq *cq; int i; - for (i = 0; i < priv->rx_ring_num; i++) { - cq = priv->rx_cq[i]; + for (i = 0; i < priv->tx_ring_num; i++) { + cq = priv->tx_cq[i]; napi_schedule(&cq->napi); } } -- cgit v0.10.2 From 21aff3b905ad9e5e52b18a755c13fe755bd6ab3d Mon Sep 17 00:00:00 2001 From: Fabien Siron Date: Tue, 7 Jun 2016 13:02:04 +0000 Subject: net/netlink/af_netlink.h: Remove unused structure. Signed-off-by: Fabien Siron Signed-off-by: David S. Miller diff --git a/net/netlink/af_netlink.h b/net/netlink/af_netlink.h index e68ef9c..3cfd6cc 100644 --- a/net/netlink/af_netlink.h +++ b/net/netlink/af_netlink.h @@ -8,20 +8,6 @@ #define NLGRPSZ(x) (ALIGN(x, sizeof(unsigned long) * 8) / 8) #define NLGRPLONGS(x) (NLGRPSZ(x)/sizeof(unsigned long)) -struct netlink_ring { - void **pg_vec; - unsigned int head; - unsigned int frames_per_block; - unsigned int frame_size; - unsigned int frame_max; - - unsigned int pg_vec_order; - unsigned int pg_vec_pages; - unsigned int pg_vec_len; - - atomic_t pending; -}; - struct netlink_sock { /* struct sock has to be the first member of netlink_sock */ struct sock sk; -- cgit v0.10.2 From 2341e0775747864b684abe8627f3d45b167f2940 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 9 Jun 2016 23:02:51 +0100 Subject: rxrpc: Simplify connect() implementation and simplify sendmsg() op Simplify the RxRPC connect() implementation. It will just note the destination address it is given, and if a sendmsg() comes along with no address, this will be assigned as the address. No transport struct will be held internally, which will allow us to remove this later. Simplify sendmsg() also. Whilst a call is active, userspace refers to it by a private unique user ID specified in a control message. When sendmsg() sees a user ID that doesn't map to an extant call, it creates a new call for that user ID and attempts to add it. If, when we try to add it, the user ID is now registered, we now reject the message with -EEXIST. We should never see this situation unless two threads are racing, trying to create a call with the same ID - which would be an error. It also isn't required to provide sendmsg() with an address - provided the control message data holds a user ID that maps to a currently active call. Signed-off-by: David Howells Signed-off-by: David S. Miller diff --git a/include/linux/rxrpc.h b/include/linux/rxrpc.h index a53915c..1e8f216 100644 --- a/include/linux/rxrpc.h +++ b/include/linux/rxrpc.h @@ -40,16 +40,18 @@ struct sockaddr_rxrpc { /* * RxRPC control messages + * - If neither abort or accept are specified, the message is a data message. * - terminal messages mean that a user call ID tag can be recycled + * - s/r/- indicate whether these are applicable to sendmsg() and/or recvmsg() */ -#define RXRPC_USER_CALL_ID 1 /* user call ID specifier */ -#define RXRPC_ABORT 2 /* abort request / notification [terminal] */ -#define RXRPC_ACK 3 /* [Server] RPC op final ACK received [terminal] */ -#define RXRPC_NET_ERROR 5 /* network error received [terminal] */ -#define RXRPC_BUSY 6 /* server busy received [terminal] */ -#define RXRPC_LOCAL_ERROR 7 /* local error generated [terminal] */ -#define RXRPC_NEW_CALL 8 /* [Server] new incoming call notification */ -#define RXRPC_ACCEPT 9 /* [Server] accept request */ +#define RXRPC_USER_CALL_ID 1 /* sr: user call ID specifier */ +#define RXRPC_ABORT 2 /* sr: abort request / notification [terminal] */ +#define RXRPC_ACK 3 /* -r: [Service] RPC op final ACK received [terminal] */ +#define RXRPC_NET_ERROR 5 /* -r: network error received [terminal] */ +#define RXRPC_BUSY 6 /* -r: server busy received [terminal] */ +#define RXRPC_LOCAL_ERROR 7 /* -r: local error generated [terminal] */ +#define RXRPC_NEW_CALL 8 /* -r: [Service] new incoming call notification */ +#define RXRPC_ACCEPT 9 /* s-: [Service] accept request */ /* * RxRPC security levels diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 7840b8e..38512a2 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -139,33 +139,33 @@ static int rxrpc_bind(struct socket *sock, struct sockaddr *saddr, int len) lock_sock(&rx->sk); - if (rx->sk.sk_state != RXRPC_UNCONNECTED) { + if (rx->sk.sk_state != RXRPC_UNBOUND) { ret = -EINVAL; goto error_unlock; } memcpy(&rx->srx, srx, sizeof(rx->srx)); - /* Find or create a local transport endpoint to use */ local = rxrpc_lookup_local(&rx->srx); if (IS_ERR(local)) { ret = PTR_ERR(local); goto error_unlock; } - rx->local = local; - if (srx->srx_service) { + if (rx->srx.srx_service) { write_lock_bh(&local->services_lock); list_for_each_entry(prx, &local->services, listen_link) { - if (prx->srx.srx_service == srx->srx_service) + if (prx->srx.srx_service == rx->srx.srx_service) goto service_in_use; } + rx->local = local; list_add_tail(&rx->listen_link, &local->services); write_unlock_bh(&local->services_lock); rx->sk.sk_state = RXRPC_SERVER_BOUND; } else { + rx->local = local; rx->sk.sk_state = RXRPC_CLIENT_BOUND; } @@ -174,8 +174,9 @@ static int rxrpc_bind(struct socket *sock, struct sockaddr *saddr, int len) return 0; service_in_use: - ret = -EADDRINUSE; write_unlock_bh(&local->services_lock); + rxrpc_put_local(local); + ret = -EADDRINUSE; error_unlock: release_sock(&rx->sk); error: @@ -197,11 +198,11 @@ static int rxrpc_listen(struct socket *sock, int backlog) lock_sock(&rx->sk); switch (rx->sk.sk_state) { - case RXRPC_UNCONNECTED: + case RXRPC_UNBOUND: ret = -EADDRNOTAVAIL; break; + case RXRPC_CLIENT_UNBOUND: case RXRPC_CLIENT_BOUND: - case RXRPC_CLIENT_CONNECTED: default: ret = -EBUSY; break; @@ -221,20 +222,18 @@ static int rxrpc_listen(struct socket *sock, int backlog) /* * find a transport by address */ -static struct rxrpc_transport *rxrpc_name_to_transport(struct socket *sock, - struct sockaddr *addr, - int addr_len, int flags, - gfp_t gfp) +struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *rx, + struct sockaddr *addr, + int addr_len, int flags, + gfp_t gfp) { struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *) addr; struct rxrpc_transport *trans; - struct rxrpc_sock *rx = rxrpc_sk(sock->sk); struct rxrpc_peer *peer; _enter("%p,%p,%d,%d", rx, addr, addr_len, flags); ASSERT(rx->local != NULL); - ASSERT(rx->sk.sk_state > RXRPC_UNCONNECTED); if (rx->srx.transport_type != srx->transport_type) return ERR_PTR(-ESOCKTNOSUPPORT); @@ -256,7 +255,7 @@ static struct rxrpc_transport *rxrpc_name_to_transport(struct socket *sock, /** * rxrpc_kernel_begin_call - Allow a kernel service to begin a call * @sock: The socket on which to make the call - * @srx: The address of the peer to contact (defaults to socket setting) + * @srx: The address of the peer to contact * @key: The security context to use (defaults to socket setting) * @user_call_ID: The ID to use * @@ -282,25 +281,14 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, lock_sock(&rx->sk); - if (srx) { - trans = rxrpc_name_to_transport(sock, (struct sockaddr *) srx, - sizeof(*srx), 0, gfp); - if (IS_ERR(trans)) { - call = ERR_CAST(trans); - trans = NULL; - goto out_notrans; - } - } else { - trans = rx->trans; - if (!trans) { - call = ERR_PTR(-ENOTCONN); - goto out_notrans; - } - atomic_inc(&trans->usage); + trans = rxrpc_name_to_transport(rx, (struct sockaddr *)srx, + sizeof(*srx), 0, gfp); + if (IS_ERR(trans)) { + call = ERR_CAST(trans); + trans = NULL; + goto out_notrans; } - if (!srx) - srx = &rx->srx; if (!key) key = rx->key; if (key && !key->payload.data[0]) @@ -312,8 +300,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, goto out; } - call = rxrpc_get_client_call(rx, trans, bundle, user_call_ID, true, - gfp); + call = rxrpc_new_client_call(rx, trans, bundle, user_call_ID, gfp); rxrpc_put_bundle(trans, bundle); out: rxrpc_put_transport(trans); @@ -369,11 +356,8 @@ EXPORT_SYMBOL(rxrpc_kernel_intercept_rx_messages); static int rxrpc_connect(struct socket *sock, struct sockaddr *addr, int addr_len, int flags) { - struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *) addr; - struct sock *sk = sock->sk; - struct rxrpc_transport *trans; - struct rxrpc_local *local; - struct rxrpc_sock *rx = rxrpc_sk(sk); + struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *)addr; + struct rxrpc_sock *rx = rxrpc_sk(sock->sk); int ret; _enter("%p,%p,%d,%d", rx, addr, addr_len, flags); @@ -386,45 +370,28 @@ static int rxrpc_connect(struct socket *sock, struct sockaddr *addr, lock_sock(&rx->sk); + ret = -EISCONN; + if (test_bit(RXRPC_SOCK_CONNECTED, &rx->flags)) + goto error; + switch (rx->sk.sk_state) { - case RXRPC_UNCONNECTED: - /* find a local transport endpoint if we don't have one already */ - ASSERTCMP(rx->local, ==, NULL); - rx->srx.srx_family = AF_RXRPC; - rx->srx.srx_service = 0; - rx->srx.transport_type = srx->transport_type; - rx->srx.transport_len = sizeof(sa_family_t); - rx->srx.transport.family = srx->transport.family; - local = rxrpc_lookup_local(&rx->srx); - if (IS_ERR(local)) { - release_sock(&rx->sk); - return PTR_ERR(local); - } - rx->local = local; - rx->sk.sk_state = RXRPC_CLIENT_BOUND; + case RXRPC_UNBOUND: + rx->sk.sk_state = RXRPC_CLIENT_UNBOUND; + case RXRPC_CLIENT_UNBOUND: case RXRPC_CLIENT_BOUND: break; - case RXRPC_CLIENT_CONNECTED: - release_sock(&rx->sk); - return -EISCONN; default: - release_sock(&rx->sk); - return -EBUSY; /* server sockets can't connect as well */ - } - - trans = rxrpc_name_to_transport(sock, addr, addr_len, flags, - GFP_KERNEL); - if (IS_ERR(trans)) { - release_sock(&rx->sk); - _leave(" = %ld", PTR_ERR(trans)); - return PTR_ERR(trans); + ret = -EBUSY; + goto error; } - rx->trans = trans; - rx->sk.sk_state = RXRPC_CLIENT_CONNECTED; + rx->connect_srx = *srx; + set_bit(RXRPC_SOCK_CONNECTED, &rx->flags); + ret = 0; +error: release_sock(&rx->sk); - return 0; + return ret; } /* @@ -438,7 +405,7 @@ static int rxrpc_connect(struct socket *sock, struct sockaddr *addr, */ static int rxrpc_sendmsg(struct socket *sock, struct msghdr *m, size_t len) { - struct rxrpc_transport *trans; + struct rxrpc_local *local; struct rxrpc_sock *rx = rxrpc_sk(sock->sk); int ret; @@ -455,48 +422,38 @@ static int rxrpc_sendmsg(struct socket *sock, struct msghdr *m, size_t len) } } - trans = NULL; lock_sock(&rx->sk); - if (m->msg_name) { - ret = -EISCONN; - trans = rxrpc_name_to_transport(sock, m->msg_name, - m->msg_namelen, 0, GFP_KERNEL); - if (IS_ERR(trans)) { - ret = PTR_ERR(trans); - trans = NULL; - goto out; - } - } else { - trans = rx->trans; - if (trans) - atomic_inc(&trans->usage); - } - switch (rx->sk.sk_state) { - case RXRPC_SERVER_LISTENING: - if (!m->msg_name) { - ret = rxrpc_server_sendmsg(rx, m, len); - break; + case RXRPC_UNBOUND: + local = rxrpc_lookup_local(&rx->srx); + if (IS_ERR(local)) { + ret = PTR_ERR(local); + goto error_unlock; } - case RXRPC_SERVER_BOUND: + + rx->local = local; + rx->sk.sk_state = RXRPC_CLIENT_UNBOUND; + /* Fall through */ + + case RXRPC_CLIENT_UNBOUND: case RXRPC_CLIENT_BOUND: - if (!m->msg_name) { - ret = -ENOTCONN; - break; + if (!m->msg_name && + test_bit(RXRPC_SOCK_CONNECTED, &rx->flags)) { + m->msg_name = &rx->connect_srx; + m->msg_namelen = sizeof(rx->connect_srx); } - case RXRPC_CLIENT_CONNECTED: - ret = rxrpc_client_sendmsg(rx, trans, m, len); + case RXRPC_SERVER_BOUND: + case RXRPC_SERVER_LISTENING: + ret = rxrpc_do_sendmsg(rx, m, len); break; default: - ret = -ENOTCONN; + ret = -EINVAL; break; } -out: +error_unlock: release_sock(&rx->sk); - if (trans) - rxrpc_put_transport(trans); _leave(" = %d", ret); return ret; } @@ -523,7 +480,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname, if (optlen != 0) goto error; ret = -EISCONN; - if (rx->sk.sk_state != RXRPC_UNCONNECTED) + if (rx->sk.sk_state != RXRPC_UNBOUND) goto error; set_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags); goto success; @@ -533,7 +490,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname, if (rx->key) goto error; ret = -EISCONN; - if (rx->sk.sk_state != RXRPC_UNCONNECTED) + if (rx->sk.sk_state != RXRPC_UNBOUND) goto error; ret = rxrpc_request_key(rx, optval, optlen); goto error; @@ -543,7 +500,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname, if (rx->key) goto error; ret = -EISCONN; - if (rx->sk.sk_state != RXRPC_UNCONNECTED) + if (rx->sk.sk_state != RXRPC_UNBOUND) goto error; ret = rxrpc_server_keyring(rx, optval, optlen); goto error; @@ -553,7 +510,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname, if (optlen != sizeof(unsigned int)) goto error; ret = -EISCONN; - if (rx->sk.sk_state != RXRPC_UNCONNECTED) + if (rx->sk.sk_state != RXRPC_UNBOUND) goto error; ret = get_user(min_sec_level, (unsigned int __user *) optval); @@ -632,7 +589,7 @@ static int rxrpc_create(struct net *net, struct socket *sock, int protocol, return -ENOMEM; sock_init_data(sock, sk); - sk->sk_state = RXRPC_UNCONNECTED; + sk->sk_state = RXRPC_UNBOUND; sk->sk_write_space = rxrpc_write_space; sk->sk_max_ack_backlog = sysctl_rxrpc_max_qlen; sk->sk_destruct = rxrpc_sock_destructor; @@ -705,14 +662,6 @@ static int rxrpc_release_sock(struct sock *sk) rx->conn = NULL; } - if (rx->bundle) { - rxrpc_put_bundle(rx->trans, rx->bundle); - rx->bundle = NULL; - } - if (rx->trans) { - rxrpc_put_transport(rx->trans); - rx->trans = NULL; - } if (rx->local) { rxrpc_put_local(rx->local); rx->local = NULL; diff --git a/net/rxrpc/ar-call.c b/net/rxrpc/ar-call.c index 1fbaae1..68125dc 100644 --- a/net/rxrpc/ar-call.c +++ b/net/rxrpc/ar-call.c @@ -196,6 +196,43 @@ struct rxrpc_call *rxrpc_find_call_hash( } /* + * find an extant server call + * - called in process context with IRQs enabled + */ +struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *rx, + unsigned long user_call_ID) +{ + struct rxrpc_call *call; + struct rb_node *p; + + _enter("%p,%lx", rx, user_call_ID); + + read_lock(&rx->call_lock); + + p = rx->calls.rb_node; + while (p) { + call = rb_entry(p, struct rxrpc_call, sock_node); + + if (user_call_ID < call->user_call_ID) + p = p->rb_left; + else if (user_call_ID > call->user_call_ID) + p = p->rb_right; + else + goto found_extant_call; + } + + read_unlock(&rx->call_lock); + _leave(" = NULL"); + return NULL; + +found_extant_call: + rxrpc_get_call(call); + read_unlock(&rx->call_lock); + _leave(" = %p [%d]", call, atomic_read(&call->usage)); + return call; +} + +/* * allocate a new call */ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp) @@ -311,51 +348,27 @@ static struct rxrpc_call *rxrpc_alloc_client_call( * set up a call for the given data * - called in process context with IRQs enabled */ -struct rxrpc_call *rxrpc_get_client_call(struct rxrpc_sock *rx, +struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, struct rxrpc_transport *trans, struct rxrpc_conn_bundle *bundle, unsigned long user_call_ID, - int create, gfp_t gfp) { - struct rxrpc_call *call, *candidate; - struct rb_node *p, *parent, **pp; + struct rxrpc_call *call, *xcall; + struct rb_node *parent, **pp; - _enter("%p,%d,%d,%lx,%d", - rx, trans ? trans->debug_id : -1, bundle ? bundle->debug_id : -1, - user_call_ID, create); + _enter("%p,%d,%d,%lx", + rx, trans->debug_id, bundle ? bundle->debug_id : -1, + user_call_ID); - /* search the extant calls first for one that matches the specified - * user ID */ - read_lock(&rx->call_lock); - - p = rx->calls.rb_node; - while (p) { - call = rb_entry(p, struct rxrpc_call, sock_node); - - if (user_call_ID < call->user_call_ID) - p = p->rb_left; - else if (user_call_ID > call->user_call_ID) - p = p->rb_right; - else - goto found_extant_call; + call = rxrpc_alloc_client_call(rx, trans, bundle, gfp); + if (IS_ERR(call)) { + _leave(" = %ld", PTR_ERR(call)); + return call; } - read_unlock(&rx->call_lock); - - if (!create || !trans) - return ERR_PTR(-EBADSLT); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_client_call(rx, trans, bundle, gfp); - if (IS_ERR(candidate)) { - _leave(" = %ld", PTR_ERR(candidate)); - return candidate; - } - - candidate->user_call_ID = user_call_ID; - __set_bit(RXRPC_CALL_HAS_USERID, &candidate->flags); + call->user_call_ID = user_call_ID; + __set_bit(RXRPC_CALL_HAS_USERID, &call->flags); write_lock(&rx->call_lock); @@ -363,19 +376,16 @@ struct rxrpc_call *rxrpc_get_client_call(struct rxrpc_sock *rx, parent = NULL; while (*pp) { parent = *pp; - call = rb_entry(parent, struct rxrpc_call, sock_node); + xcall = rb_entry(parent, struct rxrpc_call, sock_node); - if (user_call_ID < call->user_call_ID) + if (user_call_ID < xcall->user_call_ID) pp = &(*pp)->rb_left; - else if (user_call_ID > call->user_call_ID) + else if (user_call_ID > xcall->user_call_ID) pp = &(*pp)->rb_right; else - goto found_extant_second; + goto found_user_ID_now_present; } - /* second search also failed; add the new call */ - call = candidate; - candidate = NULL; rxrpc_get_call(call); rb_link_node(&call->sock_node, parent, pp); @@ -391,20 +401,16 @@ struct rxrpc_call *rxrpc_get_client_call(struct rxrpc_sock *rx, _leave(" = %p [new]", call); return call; - /* we found the call in the list immediately */ -found_extant_call: - rxrpc_get_call(call); - read_unlock(&rx->call_lock); - _leave(" = %p [extant %d]", call, atomic_read(&call->usage)); - return call; - - /* we found the call on the second time through the list */ -found_extant_second: - rxrpc_get_call(call); + /* We unexpectedly found the user ID in the list after taking + * the call_lock. This shouldn't happen unless the user races + * with itself and tries to add the same user ID twice at the + * same time in different threads. + */ +found_user_ID_now_present: write_unlock(&rx->call_lock); - rxrpc_put_call(candidate); - _leave(" = %p [second %d]", call, atomic_read(&call->usage)); - return call; + rxrpc_put_call(call); + _leave(" = -EEXIST [%p]", call); + return ERR_PTR(-EEXIST); } /* @@ -566,46 +572,6 @@ old_call: } /* - * find an extant server call - * - called in process context with IRQs enabled - */ -struct rxrpc_call *rxrpc_find_server_call(struct rxrpc_sock *rx, - unsigned long user_call_ID) -{ - struct rxrpc_call *call; - struct rb_node *p; - - _enter("%p,%lx", rx, user_call_ID); - - /* search the extant calls for one that matches the specified user - * ID */ - read_lock(&rx->call_lock); - - p = rx->calls.rb_node; - while (p) { - call = rb_entry(p, struct rxrpc_call, sock_node); - - if (user_call_ID < call->user_call_ID) - p = p->rb_left; - else if (user_call_ID > call->user_call_ID) - p = p->rb_right; - else - goto found_extant_call; - } - - read_unlock(&rx->call_lock); - _leave(" = NULL"); - return NULL; - - /* we found the call in the list immediately */ -found_extant_call: - rxrpc_get_call(call); - read_unlock(&rx->call_lock); - _leave(" = %p [%d]", call, atomic_read(&call->usage)); - return call; -} - -/* * detach a call from a socket and set up for release */ void rxrpc_release_call(struct rxrpc_call *call) diff --git a/net/rxrpc/ar-connection.c b/net/rxrpc/ar-connection.c index d67b1f1..8ecde4b 100644 --- a/net/rxrpc/ar-connection.c +++ b/net/rxrpc/ar-connection.c @@ -80,11 +80,6 @@ struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *rx, _enter("%p{%x},%x,%hx,", rx, key_serial(key), trans->debug_id, service_id); - if (rx->trans == trans && rx->bundle) { - atomic_inc(&rx->bundle->usage); - return rx->bundle; - } - /* search the extant bundles first for one that matches the specified * user ID */ spin_lock(&trans->client_lock); @@ -138,10 +133,6 @@ struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *rx, rb_insert_color(&bundle->node, &trans->bundles); spin_unlock(&trans->client_lock); _net("BUNDLE new on trans %d", trans->debug_id); - if (!rx->bundle && rx->sk.sk_state == RXRPC_CLIENT_CONNECTED) { - atomic_inc(&bundle->usage); - rx->bundle = bundle; - } _leave(" = %p [new]", bundle); return bundle; @@ -150,10 +141,6 @@ found_extant_bundle: atomic_inc(&bundle->usage); spin_unlock(&trans->client_lock); _net("BUNDLE old on trans %d", trans->debug_id); - if (!rx->bundle && rx->sk.sk_state == RXRPC_CLIENT_CONNECTED) { - atomic_inc(&bundle->usage); - rx->bundle = bundle; - } _leave(" = %p [extant %d]", bundle, atomic_read(&bundle->usage)); return bundle; @@ -163,10 +150,6 @@ found_extant_second: spin_unlock(&trans->client_lock); kfree(candidate); _net("BUNDLE old2 on trans %d", trans->debug_id); - if (!rx->bundle && rx->sk.sk_state == RXRPC_CLIENT_CONNECTED) { - atomic_inc(&bundle->usage); - rx->bundle = bundle; - } _leave(" = %p [second %d]", bundle, atomic_read(&bundle->usage)); return bundle; } diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 18ab5c5..b89dcdc 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -39,9 +39,9 @@ struct rxrpc_crypt { * sk_state for RxRPC sockets */ enum { - RXRPC_UNCONNECTED = 0, + RXRPC_UNBOUND = 0, + RXRPC_CLIENT_UNBOUND, /* Unbound socket used as client */ RXRPC_CLIENT_BOUND, /* client local address bound */ - RXRPC_CLIENT_CONNECTED, /* client is connected */ RXRPC_SERVER_BOUND, /* server local address bound */ RXRPC_SERVER_LISTENING, /* server listening for connections */ RXRPC_CLOSE, /* socket is being closed */ @@ -55,8 +55,6 @@ struct rxrpc_sock { struct sock sk; rxrpc_interceptor_t interceptor; /* kernel service Rx interceptor function */ struct rxrpc_local *local; /* local endpoint */ - struct rxrpc_transport *trans; /* transport handler */ - struct rxrpc_conn_bundle *bundle; /* virtual connection bundle */ struct rxrpc_connection *conn; /* exclusive virtual connection */ struct list_head listen_link; /* link in the local endpoint's listen list */ struct list_head secureq; /* calls awaiting connection security clearance */ @@ -65,11 +63,13 @@ struct rxrpc_sock { struct key *securities; /* list of server security descriptors */ struct rb_root calls; /* outstanding calls on this socket */ unsigned long flags; +#define RXRPC_SOCK_CONNECTED 0 /* connect_srx is set */ #define RXRPC_SOCK_EXCLUSIVE_CONN 1 /* exclusive connection for a client socket */ rwlock_t call_lock; /* lock for calls */ u32 min_sec_level; /* minimum security level */ #define RXRPC_SECURITY_MAX RXRPC_SECURITY_ENCRYPT struct sockaddr_rxrpc srx; /* local address */ + struct sockaddr_rxrpc connect_srx; /* Default client address from connect() */ sa_family_t proto; /* protocol created with */ }; @@ -477,6 +477,10 @@ extern u32 rxrpc_epoch; extern atomic_t rxrpc_debug_id; extern struct workqueue_struct *rxrpc_workqueue; +extern struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *, + struct sockaddr *, + int, int, gfp_t); + /* * ar-accept.c */ @@ -502,14 +506,14 @@ extern rwlock_t rxrpc_call_lock; struct rxrpc_call *rxrpc_find_call_hash(struct rxrpc_host_header *, void *, sa_family_t, const void *); -struct rxrpc_call *rxrpc_get_client_call(struct rxrpc_sock *, +struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *, unsigned long); +struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *, struct rxrpc_transport *, struct rxrpc_conn_bundle *, - unsigned long, int, gfp_t); + unsigned long, gfp_t); struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *, struct rxrpc_connection *, struct rxrpc_host_header *); -struct rxrpc_call *rxrpc_find_server_call(struct rxrpc_sock *, unsigned long); void rxrpc_release_call(struct rxrpc_call *); void rxrpc_release_calls_on_socket(struct rxrpc_sock *); void __rxrpc_put_call(struct rxrpc_call *); @@ -581,9 +585,7 @@ int rxrpc_get_server_data_key(struct rxrpc_connection *, const void *, time_t, extern unsigned int rxrpc_resend_timeout; int rxrpc_send_packet(struct rxrpc_transport *, struct sk_buff *); -int rxrpc_client_sendmsg(struct rxrpc_sock *, struct rxrpc_transport *, - struct msghdr *, size_t); -int rxrpc_server_sendmsg(struct rxrpc_sock *, struct msghdr *, size_t); +int rxrpc_do_sendmsg(struct rxrpc_sock *, struct msghdr *, size_t); /* * ar-peer.c diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c index ea61953..2e3c406 100644 --- a/net/rxrpc/ar-output.c +++ b/net/rxrpc/ar-output.c @@ -32,13 +32,13 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, /* * extract control messages from the sendmsg() control buffer */ -static int rxrpc_sendmsg_cmsg(struct rxrpc_sock *rx, struct msghdr *msg, +static int rxrpc_sendmsg_cmsg(struct msghdr *msg, unsigned long *user_call_ID, enum rxrpc_command *command, - u32 *abort_code, - bool server) + u32 *abort_code) { struct cmsghdr *cmsg; + bool got_user_ID = false; int len; *command = RXRPC_CMD_SEND_DATA; @@ -70,6 +70,7 @@ static int rxrpc_sendmsg_cmsg(struct rxrpc_sock *rx, struct msghdr *msg, CMSG_DATA(cmsg); } _debug("User Call ID %lx", *user_call_ID); + got_user_ID = true; break; case RXRPC_ABORT: @@ -90,8 +91,6 @@ static int rxrpc_sendmsg_cmsg(struct rxrpc_sock *rx, struct msghdr *msg, *command = RXRPC_CMD_ACCEPT; if (len != 0) return -EINVAL; - if (!server) - return -EISCONN; break; default: @@ -99,6 +98,8 @@ static int rxrpc_sendmsg_cmsg(struct rxrpc_sock *rx, struct msghdr *msg, } } + if (!got_user_ID) + return -EINVAL; _leave(" = 0"); return 0; } @@ -126,55 +127,96 @@ static void rxrpc_send_abort(struct rxrpc_call *call, u32 abort_code) } /* + * Create a new client call for sendmsg(). + */ +static struct rxrpc_call * +rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, + unsigned long user_call_ID) +{ + struct rxrpc_conn_bundle *bundle; + struct rxrpc_transport *trans; + struct rxrpc_call *call; + struct key *key; + long ret; + + DECLARE_SOCKADDR(struct sockaddr_rxrpc *, srx, msg->msg_name); + + _enter(""); + + if (!msg->msg_name) + return ERR_PTR(-EDESTADDRREQ); + + trans = rxrpc_name_to_transport(rx, msg->msg_name, msg->msg_namelen, 0, + GFP_KERNEL); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + goto out; + } + + key = rx->key; + if (key && !rx->key->payload.data[0]) + key = NULL; + bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, GFP_KERNEL); + if (IS_ERR(bundle)) { + ret = PTR_ERR(bundle); + goto out_trans; + } + + call = rxrpc_new_client_call(rx, trans, bundle, user_call_ID, + GFP_KERNEL); + rxrpc_put_bundle(trans, bundle); + rxrpc_put_transport(trans); + if (IS_ERR(call)) { + ret = PTR_ERR(call); + goto out_trans; + } + + _leave(" = %p\n", call); + return call; + +out_trans: + rxrpc_put_transport(trans); +out: + _leave(" = %ld", ret); + return ERR_PTR(ret); +} + +/* * send a message forming part of a client call through an RxRPC socket * - caller holds the socket locked * - the socket may be either a client socket or a server socket */ -int rxrpc_client_sendmsg(struct rxrpc_sock *rx, struct rxrpc_transport *trans, - struct msghdr *msg, size_t len) +int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) { - struct rxrpc_conn_bundle *bundle; enum rxrpc_command cmd; struct rxrpc_call *call; unsigned long user_call_ID = 0; - struct key *key; - u16 service_id; u32 abort_code = 0; int ret; _enter(""); - ASSERT(trans != NULL); - - ret = rxrpc_sendmsg_cmsg(rx, msg, &user_call_ID, &cmd, &abort_code, - false); + ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code); if (ret < 0) return ret; - bundle = NULL; - if (trans) { - service_id = rx->srx.srx_service; - if (msg->msg_name) { - DECLARE_SOCKADDR(struct sockaddr_rxrpc *, srx, - msg->msg_name); - service_id = srx->srx_service; - } - key = rx->key; - if (key && !rx->key->payload.data[0]) - key = NULL; - bundle = rxrpc_get_bundle(rx, trans, key, service_id, - GFP_KERNEL); - if (IS_ERR(bundle)) - return PTR_ERR(bundle); + if (cmd == RXRPC_CMD_ACCEPT) { + if (rx->sk.sk_state != RXRPC_SERVER_LISTENING) + return -EINVAL; + call = rxrpc_accept_call(rx, user_call_ID); + if (IS_ERR(call)) + return PTR_ERR(call); + rxrpc_put_call(call); + return 0; } - call = rxrpc_get_client_call(rx, trans, bundle, user_call_ID, - abort_code == 0, GFP_KERNEL); - if (trans) - rxrpc_put_bundle(trans, bundle); - if (IS_ERR(call)) { - _leave(" = %ld", PTR_ERR(call)); - return PTR_ERR(call); + call = rxrpc_find_call_by_user_ID(rx, user_call_ID); + if (!call) { + if (cmd != RXRPC_CMD_SEND_DATA) + return -EBADSLT; + call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID); + if (IS_ERR(call)) + return PTR_ERR(call); } _debug("CALL %d USR %lx ST %d on CONN %p", @@ -182,14 +224,21 @@ int rxrpc_client_sendmsg(struct rxrpc_sock *rx, struct rxrpc_transport *trans, if (call->state >= RXRPC_CALL_COMPLETE) { /* it's too late for this call */ - ret = -ESHUTDOWN; + ret = -ECONNRESET; } else if (cmd == RXRPC_CMD_SEND_ABORT) { rxrpc_send_abort(call, abort_code); + ret = 0; } else if (cmd != RXRPC_CMD_SEND_DATA) { ret = -EINVAL; - } else if (call->state != RXRPC_CALL_CLIENT_SEND_REQUEST) { + } else if (!call->in_clientflag && + call->state != RXRPC_CALL_CLIENT_SEND_REQUEST) { /* request phase complete for this client call */ ret = -EPROTO; + } else if (call->in_clientflag && + call->state != RXRPC_CALL_SERVER_ACK_REQUEST && + call->state != RXRPC_CALL_SERVER_SEND_REPLY) { + /* Reply phase not begun or not complete for service call. */ + ret = -EPROTO; } else { ret = rxrpc_send_data(rx, call, msg, len); } @@ -268,67 +317,6 @@ void rxrpc_kernel_abort_call(struct rxrpc_call *call, u32 abort_code) EXPORT_SYMBOL(rxrpc_kernel_abort_call); /* - * send a message through a server socket - * - caller holds the socket locked - */ -int rxrpc_server_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) -{ - enum rxrpc_command cmd; - struct rxrpc_call *call; - unsigned long user_call_ID = 0; - u32 abort_code = 0; - int ret; - - _enter(""); - - ret = rxrpc_sendmsg_cmsg(rx, msg, &user_call_ID, &cmd, &abort_code, - true); - if (ret < 0) - return ret; - - if (cmd == RXRPC_CMD_ACCEPT) { - call = rxrpc_accept_call(rx, user_call_ID); - if (IS_ERR(call)) - return PTR_ERR(call); - rxrpc_put_call(call); - return 0; - } - - call = rxrpc_find_server_call(rx, user_call_ID); - if (!call) - return -EBADSLT; - if (call->state >= RXRPC_CALL_COMPLETE) { - ret = -ESHUTDOWN; - goto out; - } - - switch (cmd) { - case RXRPC_CMD_SEND_DATA: - if (call->state != RXRPC_CALL_CLIENT_SEND_REQUEST && - call->state != RXRPC_CALL_SERVER_ACK_REQUEST && - call->state != RXRPC_CALL_SERVER_SEND_REPLY) { - /* Tx phase not yet begun for this call */ - ret = -EPROTO; - break; - } - - ret = rxrpc_send_data(rx, call, msg, len); - break; - - case RXRPC_CMD_SEND_ABORT: - rxrpc_send_abort(call, abort_code); - break; - default: - BUG(); - } - - out: - rxrpc_put_call(call); - _leave(" = %d", ret); - return ret; -} - -/* * send a packet through the transport endpoint */ int rxrpc_send_packet(struct rxrpc_transport *trans, struct sk_buff *skb) -- cgit v0.10.2 From e434863718d4b99dd0d6e0cefd3c5e79e4fa2083 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 9 Jun 2016 10:21:00 -0700 Subject: net: vrf: Fix crash when IPv6 is disabled at boot time Frank Kellermann reported a kernel crash with 4.5.0 when IPv6 is disabled at boot using the kernel option ipv6.disable=1. Using current net-next with the boot option: $ ip link add red type vrf table 1001 Generates: [12210.919584] BUG: unable to handle kernel NULL pointer dereference at 0000000000000748 [12210.921341] IP: [] fib6_get_table+0x2c/0x5a [12210.922537] PGD b79e3067 PUD bb32b067 PMD 0 [12210.923479] Oops: 0000 [#1] SMP [12210.924001] Modules linked in: ipvlan 8021q garp mrp stp llc [12210.925130] CPU: 3 PID: 1177 Comm: ip Not tainted 4.7.0-rc1+ #235 [12210.926168] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [12210.928065] task: ffff8800b9ac4640 ti: ffff8800bacac000 task.ti: ffff8800bacac000 [12210.929328] RIP: 0010:[] [] fib6_get_table+0x2c/0x5a [12210.930697] RSP: 0018:ffff8800bacaf888 EFLAGS: 00010202 [12210.931563] RAX: 0000000000000748 RBX: ffffffff81a9e280 RCX: ffff8800b9ac4e28 [12210.932688] RDX: 00000000000000e9 RSI: 0000000000000002 RDI: 0000000000000286 [12210.933820] RBP: ffff8800bacaf898 R08: ffff8800b9ac4df0 R09: 000000000052001b [12210.934941] R10: 00000000657c0000 R11: 000000000000c649 R12: 00000000000003e9 [12210.936032] R13: 00000000000003e9 R14: ffff8800bace7800 R15: ffff8800bb3ec000 [12210.937103] FS: 00007faa1766c700(0000) GS:ffff88013ac00000(0000) knlGS:0000000000000000 [12210.938321] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [12210.939166] CR2: 0000000000000748 CR3: 00000000b79d6000 CR4: 00000000000406e0 [12210.940278] Stack: [12210.940603] ffff8800bb3ec000 ffffffff81a9e280 ffff8800bacaf8c8 ffffffff814b3135 [12210.941818] ffff8800bb3ec000 ffffffff81a9e280 ffffffff81a9e280 ffff8800bace7800 [12210.943040] ffff8800bacaf8f0 ffffffff81397c88 ffff8800bb3ec000 ffffffff81a9e280 [12210.944288] Call Trace: [12210.944688] [] fib6_new_table+0x24/0x8a [12210.945516] [] vrf_dev_init+0xd4/0x162 [12210.946328] [] register_netdevice+0x100/0x396 [12210.947209] [] vrf_newlink+0x40/0xb3 [12210.948001] [] rtnl_newlink+0x5d3/0x6d5 ... The problem above is due to the fact that the fib hash table is not allocated when IPv6 is disabled at boot. As for the VRF driver it should not do any IPv6 initializations if IPv6 is disabled, so it needs to know if IPv6 is disabled at boot. The disable parameter is private to the IPv6 module, so provide an accessor for modules to determine if IPv6 was disabled at boot time. Fixes: 35402e3136634 ("net: Add IPv6 support to VRF device") Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index b82e3527..b4d7469 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -407,6 +407,10 @@ static int vrf_rt6_create(struct net_device *dev) struct rt6_info *rt6, *rt6_local; int rc = -ENOMEM; + /* IPv6 can be CONFIG enabled and then disabled runtime */ + if (!ipv6_mod_enabled()) + return 0; + rt6i_table = fib6_new_table(net, vrf->tb_id); if (!rt6i_table) goto out; @@ -919,6 +923,9 @@ static int vrf_fib_rule(const struct net_device *dev, __u8 family, bool add_it) struct sk_buff *skb; int err; + if (family == AF_INET6 && !ipv6_mod_enabled()) + return 0; + skb = nlmsg_new(vrf_fib_rule_nl_size(), GFP_KERNEL); if (!skb) return -ENOMEM; diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 5c91b0b..c6dbcd8 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -283,6 +283,8 @@ struct tcp6_timewait_sock { }; #if IS_ENABLED(CONFIG_IPV6) +bool ipv6_mod_enabled(void); + static inline struct ipv6_pinfo *inet6_sk(const struct sock *__sk) { return sk_fullsock(__sk) ? inet_sk(__sk)->pinet6 : NULL; @@ -326,6 +328,11 @@ static inline int inet_v6_ipv6only(const struct sock *sk) #define ipv6_only_sock(sk) 0 #define ipv6_sk_rxinfo(sk) 0 +static inline bool ipv6_mod_enabled(void) +{ + return false; +} + static inline struct ipv6_pinfo * inet6_sk(const struct sock *__sk) { return NULL; diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index bfa86f0..2076c21 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -92,6 +92,12 @@ MODULE_PARM_DESC(disable_ipv6, "Disable IPv6 on all interfaces"); module_param_named(autoconf, ipv6_defaults.autoconf, int, 0444); MODULE_PARM_DESC(autoconf, "Enable IPv6 address autoconfiguration on all interfaces"); +bool ipv6_mod_enabled(void) +{ + return disable_ipv6_mod == 0; +} +EXPORT_SYMBOL_GPL(ipv6_mod_enabled); + static __inline__ struct ipv6_pinfo *inet6_sk_generic(struct sock *sk) { const int offset = sk->sk_prot->obj_size - sizeof(struct ipv6_pinfo); -- cgit v0.10.2 From 7d84e37e114215be0a0c80095891c8268a99352b Mon Sep 17 00:00:00 2001 From: Aaron Conole Date: Thu, 9 Jun 2016 13:41:15 -0400 Subject: virtio_net: Update the feature bit to comply with spec A draft version of the MTU Advice feature bit was specified as 25. This bit is not within the allowed range for network device feature bits, and should be changed to be feature bit 3 to fully comply with the spec. Fixes 14de9d114a82 ('virtio-net: Add initial MTU advice feature') Signed-off-by: Aaron Conole Suggested-by: "Michael S. Tsirkin" Signed-off-by: David S. Miller diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h index 1ab4ea6..0da0e3a 100644 --- a/include/uapi/linux/virtio_net.h +++ b/include/uapi/linux/virtio_net.h @@ -35,6 +35,7 @@ #define VIRTIO_NET_F_CSUM 0 /* Host handles pkts w/ partial csum */ #define VIRTIO_NET_F_GUEST_CSUM 1 /* Guest handles pkts w/ partial csum */ #define VIRTIO_NET_F_CTRL_GUEST_OFFLOADS 2 /* Dynamic offload configuration. */ +#define VIRTIO_NET_F_MTU 3 /* Initial MTU advice */ #define VIRTIO_NET_F_MAC 5 /* Host has given MAC address. */ #define VIRTIO_NET_F_GUEST_TSO4 7 /* Guest can handle TSOv4 in. */ #define VIRTIO_NET_F_GUEST_TSO6 8 /* Guest can handle TSOv6 in. */ @@ -55,7 +56,6 @@ #define VIRTIO_NET_F_MQ 22 /* Device supports Receive Flow * Steering */ #define VIRTIO_NET_F_CTRL_MAC_ADDR 23 /* Set MAC address */ -#define VIRTIO_NET_F_MTU 25 /* Initial MTU advice */ #ifndef VIRTIO_NET_NO_LEGACY #define VIRTIO_NET_F_GSO 6 /* Host handles pkts w/ any GSO type */ -- cgit v0.10.2 From 2022e9d50798aa592887ccb5a7d045e537f3855f Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 7 Jun 2016 15:06:13 -0700 Subject: bgmac: Bind net_device with backing device structure In preparation for allowing different helpers to be utilized against network devices created by the bgmac driver, make sure that we bind the net_device with core->dev. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index ee5f431..156fa63 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -1588,6 +1588,7 @@ static int bgmac_probe(struct bcma_device *core) bgmac->net_dev = net_dev; bgmac->core = core; bcma_set_drvdata(core, bgmac); + SET_NETDEV_DEV(net_dev, &core->dev); /* Defaults */ memcpy(bgmac->net_dev->dev_addr, mac, ETH_ALEN); -- cgit v0.10.2 From f6613d4fa937fa8388f2c1cb4e69ccc25e9e2336 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 7 Jun 2016 15:06:14 -0700 Subject: bgmac: Add support for ethtool statistics Read the statistics from the BGMAC's builtin MAC and return them to user-space using the standard ethtool helpers. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index 156fa63..133bf64 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -1382,6 +1382,127 @@ static const struct net_device_ops bgmac_netdev_ops = { * ethtool_ops **************************************************/ +struct bgmac_stat { + u8 size; + u32 offset; + const char *name; +}; + +static struct bgmac_stat bgmac_get_strings_stats[] = { + { 8, BGMAC_TX_GOOD_OCTETS, "tx_good_octets" }, + { 4, BGMAC_TX_GOOD_PKTS, "tx_good" }, + { 8, BGMAC_TX_OCTETS, "tx_octets" }, + { 4, BGMAC_TX_PKTS, "tx_pkts" }, + { 4, BGMAC_TX_BROADCAST_PKTS, "tx_broadcast" }, + { 4, BGMAC_TX_MULTICAST_PKTS, "tx_multicast" }, + { 4, BGMAC_TX_LEN_64, "tx_64" }, + { 4, BGMAC_TX_LEN_65_TO_127, "tx_65_127" }, + { 4, BGMAC_TX_LEN_128_TO_255, "tx_128_255" }, + { 4, BGMAC_TX_LEN_256_TO_511, "tx_256_511" }, + { 4, BGMAC_TX_LEN_512_TO_1023, "tx_512_1023" }, + { 4, BGMAC_TX_LEN_1024_TO_1522, "tx_1024_1522" }, + { 4, BGMAC_TX_LEN_1523_TO_2047, "tx_1523_2047" }, + { 4, BGMAC_TX_LEN_2048_TO_4095, "tx_2048_4095" }, + { 4, BGMAC_TX_LEN_4096_TO_8191, "tx_4096_8191" }, + { 4, BGMAC_TX_LEN_8192_TO_MAX, "tx_8192_max" }, + { 4, BGMAC_TX_JABBER_PKTS, "tx_jabber" }, + { 4, BGMAC_TX_OVERSIZE_PKTS, "tx_oversize" }, + { 4, BGMAC_TX_FRAGMENT_PKTS, "tx_fragment" }, + { 4, BGMAC_TX_UNDERRUNS, "tx_underruns" }, + { 4, BGMAC_TX_TOTAL_COLS, "tx_total_cols" }, + { 4, BGMAC_TX_SINGLE_COLS, "tx_single_cols" }, + { 4, BGMAC_TX_MULTIPLE_COLS, "tx_multiple_cols" }, + { 4, BGMAC_TX_EXCESSIVE_COLS, "tx_excessive_cols" }, + { 4, BGMAC_TX_LATE_COLS, "tx_late_cols" }, + { 4, BGMAC_TX_DEFERED, "tx_defered" }, + { 4, BGMAC_TX_CARRIER_LOST, "tx_carrier_lost" }, + { 4, BGMAC_TX_PAUSE_PKTS, "tx_pause" }, + { 4, BGMAC_TX_UNI_PKTS, "tx_unicast" }, + { 4, BGMAC_TX_Q0_PKTS, "tx_q0" }, + { 8, BGMAC_TX_Q0_OCTETS, "tx_q0_octets" }, + { 4, BGMAC_TX_Q1_PKTS, "tx_q1" }, + { 8, BGMAC_TX_Q1_OCTETS, "tx_q1_octets" }, + { 4, BGMAC_TX_Q2_PKTS, "tx_q2" }, + { 8, BGMAC_TX_Q2_OCTETS, "tx_q2_octets" }, + { 4, BGMAC_TX_Q3_PKTS, "tx_q3" }, + { 8, BGMAC_TX_Q3_OCTETS, "tx_q3_octets" }, + { 8, BGMAC_RX_GOOD_OCTETS, "rx_good_octets" }, + { 4, BGMAC_RX_GOOD_PKTS, "rx_good" }, + { 8, BGMAC_RX_OCTETS, "rx_octets" }, + { 4, BGMAC_RX_PKTS, "rx_pkts" }, + { 4, BGMAC_RX_BROADCAST_PKTS, "rx_broadcast" }, + { 4, BGMAC_RX_MULTICAST_PKTS, "rx_multicast" }, + { 4, BGMAC_RX_LEN_64, "rx_64" }, + { 4, BGMAC_RX_LEN_65_TO_127, "rx_65_127" }, + { 4, BGMAC_RX_LEN_128_TO_255, "rx_128_255" }, + { 4, BGMAC_RX_LEN_256_TO_511, "rx_256_511" }, + { 4, BGMAC_RX_LEN_512_TO_1023, "rx_512_1023" }, + { 4, BGMAC_RX_LEN_1024_TO_1522, "rx_1024_1522" }, + { 4, BGMAC_RX_LEN_1523_TO_2047, "rx_1523_2047" }, + { 4, BGMAC_RX_LEN_2048_TO_4095, "rx_2048_4095" }, + { 4, BGMAC_RX_LEN_4096_TO_8191, "rx_4096_8191" }, + { 4, BGMAC_RX_LEN_8192_TO_MAX, "rx_8192_max" }, + { 4, BGMAC_RX_JABBER_PKTS, "rx_jabber" }, + { 4, BGMAC_RX_OVERSIZE_PKTS, "rx_oversize" }, + { 4, BGMAC_RX_FRAGMENT_PKTS, "rx_fragment" }, + { 4, BGMAC_RX_MISSED_PKTS, "rx_missed" }, + { 4, BGMAC_RX_CRC_ALIGN_ERRS, "rx_crc_align" }, + { 4, BGMAC_RX_UNDERSIZE, "rx_undersize" }, + { 4, BGMAC_RX_CRC_ERRS, "rx_crc" }, + { 4, BGMAC_RX_ALIGN_ERRS, "rx_align" }, + { 4, BGMAC_RX_SYMBOL_ERRS, "rx_symbol" }, + { 4, BGMAC_RX_PAUSE_PKTS, "rx_pause" }, + { 4, BGMAC_RX_NONPAUSE_PKTS, "rx_nonpause" }, + { 4, BGMAC_RX_SACHANGES, "rx_sa_changes" }, + { 4, BGMAC_RX_UNI_PKTS, "rx_unicast" }, +}; + +#define BGMAC_STATS_LEN ARRAY_SIZE(bgmac_get_strings_stats) + +static int bgmac_get_sset_count(struct net_device *dev, int string_set) +{ + switch (string_set) { + case ETH_SS_STATS: + return BGMAC_STATS_LEN; + } + + return -EOPNOTSUPP; +} + +static void bgmac_get_strings(struct net_device *dev, u32 stringset, + u8 *data) +{ + int i; + + if (stringset != ETH_SS_STATS) + return; + + for (i = 0; i < BGMAC_STATS_LEN; i++) + strlcpy(data + i * ETH_GSTRING_LEN, + bgmac_get_strings_stats[i].name, ETH_GSTRING_LEN); +} + +static void bgmac_get_ethtool_stats(struct net_device *dev, + struct ethtool_stats *ss, uint64_t *data) +{ + struct bgmac *bgmac = netdev_priv(dev); + const struct bgmac_stat *s; + unsigned int i; + u64 val; + + if (!netif_running(dev)) + return; + + for (i = 0; i < BGMAC_STATS_LEN; i++) { + s = &bgmac_get_strings_stats[i]; + val = 0; + if (s->size == 8) + val = (u64)bgmac_read(bgmac, s->offset + 4) << 32; + val |= bgmac_read(bgmac, s->offset); + data[i] = val; + } +} + static int bgmac_get_settings(struct net_device *net_dev, struct ethtool_cmd *cmd) { @@ -1406,6 +1527,9 @@ static void bgmac_get_drvinfo(struct net_device *net_dev, } static const struct ethtool_ops bgmac_ethtool_ops = { + .get_strings = bgmac_get_strings, + .get_sset_count = bgmac_get_sset_count, + .get_ethtool_stats = bgmac_get_ethtool_stats, .get_settings = bgmac_get_settings, .set_settings = bgmac_set_settings, .get_drvinfo = bgmac_get_drvinfo, diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h index 9a03c14..853d72b 100644 --- a/drivers/net/ethernet/broadcom/bgmac.h +++ b/drivers/net/ethernet/broadcom/bgmac.h @@ -123,7 +123,7 @@ #define BGMAC_TX_LEN_1024_TO_1522 0x334 #define BGMAC_TX_LEN_1523_TO_2047 0x338 #define BGMAC_TX_LEN_2048_TO_4095 0x33c -#define BGMAC_TX_LEN_4095_TO_8191 0x340 +#define BGMAC_TX_LEN_4096_TO_8191 0x340 #define BGMAC_TX_LEN_8192_TO_MAX 0x344 #define BGMAC_TX_JABBER_PKTS 0x348 /* Error */ #define BGMAC_TX_OVERSIZE_PKTS 0x34c /* Error */ @@ -166,7 +166,7 @@ #define BGMAC_RX_LEN_1024_TO_1522 0x3e4 #define BGMAC_RX_LEN_1523_TO_2047 0x3e8 #define BGMAC_RX_LEN_2048_TO_4095 0x3ec -#define BGMAC_RX_LEN_4095_TO_8191 0x3f0 +#define BGMAC_RX_LEN_4096_TO_8191 0x3f0 #define BGMAC_RX_LEN_8192_TO_MAX 0x3f4 #define BGMAC_RX_JABBER_PKTS 0x3f8 /* Error */ #define BGMAC_RX_OVERSIZE_PKTS 0x3fc /* Error */ -- cgit v0.10.2 From 6d490f62a4c7f11c552591bdd08eda3636aa0db9 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 7 Jun 2016 15:06:15 -0700 Subject: bgmac: Maintain some netdev statistics Add a few netdev statistics to report transmitted and received bytes and packets and a few obvious errors. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index 133bf64..0ee34cc 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -246,6 +246,8 @@ err_dma_head: err_drop: dev_kfree_skb(skb); + net_dev->stats.tx_dropped++; + net_dev->stats.tx_errors++; return NETDEV_TX_OK; } @@ -284,6 +286,8 @@ static void bgmac_dma_tx_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring) DMA_TO_DEVICE); if (slot->skb) { + bgmac->net_dev->stats.tx_bytes += slot->skb->len; + bgmac->net_dev->stats.tx_packets++; bytes_compl += slot->skb->len; pkts_compl++; @@ -464,6 +468,7 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, bgmac_err(bgmac, "Found poisoned packet at slot %d, DMA issue!\n", ring->start); put_page(virt_to_head_page(buf)); + bgmac->net_dev->stats.rx_errors++; break; } @@ -471,6 +476,8 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, bgmac_err(bgmac, "Found oversized packet at slot %d, DMA issue!\n", ring->start); put_page(virt_to_head_page(buf)); + bgmac->net_dev->stats.rx_length_errors++; + bgmac->net_dev->stats.rx_errors++; break; } @@ -481,6 +488,7 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, if (unlikely(!skb)) { bgmac_err(bgmac, "build_skb failed\n"); put_page(virt_to_head_page(buf)); + bgmac->net_dev->stats.rx_errors++; break; } skb_put(skb, BGMAC_RX_FRAME_OFFSET + @@ -490,6 +498,8 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, skb_checksum_none_assert(skb); skb->protocol = eth_type_trans(skb, bgmac->net_dev); + bgmac->net_dev->stats.rx_bytes += len; + bgmac->net_dev->stats.rx_packets++; napi_gro_receive(&bgmac->napi, skb); handled++; } while (0); -- cgit v0.10.2 From c8d83963709ea69f7433c7b4fe3e8c3a4288458a Mon Sep 17 00:00:00 2001 From: hayeswang Date: Wed, 8 Jun 2016 14:52:33 +0800 Subject: r8152: replace netdev_alloc_skb_ip_align with napi_alloc_skb Replace netdev_alloc_skb_ip_align() with napi_alloc_skb() which can save several CPU cycles by avoiding having to disable and re-enable IRQs. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 3f9f6ed..161c25e 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -1742,7 +1742,7 @@ static int rx_bottom(struct r8152 *tp, int budget) pkt_len -= CRC_SIZE; rx_data += sizeof(struct rx_desc); - skb = netdev_alloc_skb_ip_align(netdev, pkt_len); + skb = napi_alloc_skb(&tp->napi, pkt_len); if (!skb) { stats->rx_dropped++; goto find_next_rx; -- cgit v0.10.2 From f7d3c1cbe34bfcb3eb94a80409ed2577394d0d28 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 Jun 2016 21:24:18 -0700 Subject: net/mlx4_en: fix ethtool -x MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit mlx4 RSS is limited to spread incoming packets to a power of two number of queues. An uniformly distibuted traffic would be split on queues 0 to N-1, N being a power of two, each queue having a 1/N weight. If number of RX queues is not a power of two, upper RX queues do not receive traffic. ethtool -x is lying, because it pretends some queues have higher weight. Before patch: lpaa24:~# ethtool -L eth1 rx 24 lpaa24:~# ethtool -x eth1 RX flow hash indirection table for eth1 with 24 RX ring(s): 0: 0 1 2 3 4 5 6 7 8: 8 9 10 11 12 13 14 15 16: 0 1 2 3 4 5 6 7 RSS hash key: e0:7c:3a:89:07:55:b6:58:69:cc:f4:e5:24:62:e3:25:88:6c:42:5b:d2:cb:9a:d2:e0:06:e1:dc:f9:09:a1:89:0f:a0:30:43:73:6f:0c:b6 If this information was correct, user space tools could expect queues 0 to 7 to receive twice more traffic than queues 8 to 15 After patch : lpaa24:~# ethtool -L eth1 rx 24 lpaa24:~# ethtool -x eth1 RX flow hash indirection table for eth1 with 24 RX ring(s): 0: 0 1 2 3 4 5 6 7 8: 8 9 10 11 12 13 14 15 RSS hash key: da:7b:09:60:f1:ac:67:b4:d0:72:d4:ec:a2:e5:80:0a:ad:50:22:1a:f8:f9:66:54:5f:22:45:c3:88:f4:57:82:c1:c1:90:ed:70:cb:40:ce lpaa24:~# ethtool -X eth1 equal 8 lpaa24:~# ethtool -x eth1 RX flow hash indirection table for eth1 with 24 RX ring(s): 0: 0 1 2 3 4 5 6 7 8: 0 1 2 3 4 5 6 7 RSS hash key: da:7b:09:60:f1:ac:67:b4:d0:72:d4:ec:a2:e5:80:0a:ad:50:22:1a:f8:f9:66:54:5f:22:45:c3:88:f4:57:82:c1:c1:90:ed:70:cb:40:ce Signed-off-by: Eric Dumazet Reported-by: Maciej Żenczykowski Cc: Eugenia Emantayev Cc: Wei Wang Cc: Willem de Bruijn Reviewed-by: Tariq Toukan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index fc95aff..51a2e82 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -1107,7 +1107,7 @@ static u32 mlx4_en_get_rxfh_indir_size(struct net_device *dev) { struct mlx4_en_priv *priv = netdev_priv(dev); - return priv->rx_ring_num; + return rounddown_pow_of_two(priv->rx_ring_num); } static u32 mlx4_en_get_rxfh_key_size(struct net_device *netdev) @@ -1141,19 +1141,17 @@ static int mlx4_en_get_rxfh(struct net_device *dev, u32 *ring_index, u8 *key, u8 *hfunc) { struct mlx4_en_priv *priv = netdev_priv(dev); - struct mlx4_en_rss_map *rss_map = &priv->rss_map; - int rss_rings; - size_t n = priv->rx_ring_num; + u32 n = mlx4_en_get_rxfh_indir_size(dev); + u32 i, rss_rings; int err = 0; - rss_rings = priv->prof->rss_rings ?: priv->rx_ring_num; - rss_rings = 1 << ilog2(rss_rings); + rss_rings = priv->prof->rss_rings ?: n; + rss_rings = rounddown_pow_of_two(rss_rings); - while (n--) { + for (i = 0; i < n; i++) { if (!ring_index) break; - ring_index[n] = rss_map->qps[n % rss_rings].qpn - - rss_map->base_qpn; + ring_index[i] = i % rss_rings; } if (key) memcpy(key, priv->rss_key, MLX4_EN_RSS_KEY_SIZE); @@ -1166,6 +1164,7 @@ static int mlx4_en_set_rxfh(struct net_device *dev, const u32 *ring_index, const u8 *key, const u8 hfunc) { struct mlx4_en_priv *priv = netdev_priv(dev); + u32 n = mlx4_en_get_rxfh_indir_size(dev); struct mlx4_en_dev *mdev = priv->mdev; int port_up = 0; int err = 0; @@ -1175,18 +1174,18 @@ static int mlx4_en_set_rxfh(struct net_device *dev, const u32 *ring_index, /* Calculate RSS table size and make sure flows are spread evenly * between rings */ - for (i = 0; i < priv->rx_ring_num; i++) { + for (i = 0; i < n; i++) { if (!ring_index) - continue; + break; if (i > 0 && !ring_index[i] && !rss_rings) rss_rings = i; - if (ring_index[i] != (i % (rss_rings ?: priv->rx_ring_num))) + if (ring_index[i] != (i % (rss_rings ?: n))) return -EINVAL; } if (!rss_rings) - rss_rings = priv->rx_ring_num; + rss_rings = n; /* RSS table size must be an order of 2 */ if (!is_power_of_2(rss_rings)) -- cgit v0.10.2 From 292b9dab403973a8644cdb84d8e46bd4e6b4baa2 Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Wed, 8 Jun 2016 01:47:59 +0530 Subject: net: cavium: liquidio: Remove deprecated create_workqueue alloc_workqueue replaces deprecated create_workqueue(). A dedicated workqueue has been used since the workitem viz (&lio->txq_status_wq.wk.work which maps to octnet_poll_check_txq_status) is involved in a brief poll routine for checking transmit queue status and is an intergral part of normal device operation. WQ_MEM_RECLAIM has been set to guarantee forward progress under memory pressure, which is a requirement here. Since there are only a fixed number of work items, explicit concurrency limit is unnecessary. flush_workqueue is unnecessary since destroy_workqueue() itself calls drain_workqueue() which flushes repeatedly till the workqueue becomes empty. Signed-off-by: Bhaktipriya Shridhar Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 8de79ae..655d89e 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -2036,7 +2036,8 @@ static inline void setup_tx_poll_fn(struct net_device *netdev) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; - lio->txq_status_wq.wq = create_workqueue("txq-status"); + lio->txq_status_wq.wq = alloc_workqueue("txq-status", + WQ_MEM_RECLAIM, 0); if (!lio->txq_status_wq.wq) { dev_err(&oct->pci_dev->dev, "unable to create cavium txq status wq\n"); return; @@ -2103,7 +2104,6 @@ static int liquidio_stop(struct net_device *netdev) send_rx_ctrl_cmd(lio, 0); cancel_delayed_work_sync(&lio->txq_status_wq.wk.work); - flush_workqueue(lio->txq_status_wq.wq); destroy_workqueue(lio->txq_status_wq.wq); if (lio->ptp_clock) { -- cgit v0.10.2 From 3d5479e92087f6249231e26a2d7327e86a8d0dfc Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Wed, 8 Jun 2016 01:29:46 +0530 Subject: mlxsw: core: Remove deprecated create_workqueue alloc_workqueue replaces deprecated create_workqueue(). A dedicated workqueue has been used since the workqueue mlxsw_wq is used for FDB notif. processing with workitems that are involved in normal device operation && because it's a network device which can be depended upon during memory reclaim. Workitems &trans->timeout_dw and &mlxsw_sp->fdb_notify.dw, map to mlxsw_sp_fdb_notify_work (processes FDB notifications from the underlying device and resolves the netdev to which the entry points to and notifies the bridge using the switchdev notifier) and mlxsw_emad_trans_timeout_work (provides async EMAD register access) respectively. They require forward progress under memory pressure and hence, WQ_MEM_RECLAIM has been set. Since there are only a fixed number of work items, explicit concurrency limit is unnecessary here. Signed-off-by: Bhaktipriya Shridhar Tested-by: Ido Schimmel Acked-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index b0a0b01..01ae548 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -1736,7 +1736,7 @@ static int __init mlxsw_core_module_init(void) { int err; - mlxsw_wq = create_workqueue(mlxsw_core_driver_name); + mlxsw_wq = alloc_workqueue(mlxsw_core_driver_name, WQ_MEM_RECLAIM, 0); if (!mlxsw_wq) return -ENOMEM; mlxsw_core_dbg_root = debugfs_create_dir(mlxsw_core_driver_name, NULL); -- cgit v0.10.2 From 7486216b3a0bd26375b17b2cc168a311106cea70 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Thu, 9 Jun 2016 15:11:34 +0300 Subject: {net,IB}/mlx5: mlx5_ifc updates Introducing mlx5_ifc updates for upcoming ConnectX-4 features. Needed bits and hardware structures for mlx5e netdev: - MLX5_CQ_PERIOD_NUM_MODES for adaptive moderation support - QoS rate limiting - SQ context rate limiting - Auto negotiation fields in PTYS register - Source SQN field in flow table entry match structure - DCBX parameters Needed bits and hardware structures for IB: - New XRQ opcodes, commands and capabilities layout - Extend q counters definition to support IB. Signed-off-by: Saeed Mahameed Signed-off-by: Leon Romanovsky Signed-off-by: Leon Romanovsky diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 9a05cd7..209add9 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -123,6 +123,10 @@ enum { MLX5_CMD_OP_DRAIN_DCT = 0x712, MLX5_CMD_OP_QUERY_DCT = 0x713, MLX5_CMD_OP_ARM_DCT_FOR_KEY_VIOLATION = 0x714, + MLX5_CMD_OP_CREATE_XRQ = 0x717, + MLX5_CMD_OP_DESTROY_XRQ = 0x718, + MLX5_CMD_OP_QUERY_XRQ = 0x719, + MLX5_CMD_OP_ARM_XRQ = 0x71a, MLX5_CMD_OP_QUERY_VPORT_STATE = 0x750, MLX5_CMD_OP_MODIFY_VPORT_STATE = 0x751, MLX5_CMD_OP_QUERY_ESW_VPORT_CONTEXT = 0x752, @@ -139,6 +143,8 @@ enum { MLX5_CMD_OP_ALLOC_Q_COUNTER = 0x771, MLX5_CMD_OP_DEALLOC_Q_COUNTER = 0x772, MLX5_CMD_OP_QUERY_Q_COUNTER = 0x773, + MLX5_CMD_OP_SET_RATE_LIMIT = 0x780, + MLX5_CMD_OP_QUERY_RATE_LIMIT = 0x781, MLX5_CMD_OP_ALLOC_PD = 0x800, MLX5_CMD_OP_DEALLOC_PD = 0x801, MLX5_CMD_OP_ALLOC_UAR = 0x802, @@ -361,7 +367,8 @@ struct mlx5_ifc_fte_match_set_lyr_2_4_bits { }; struct mlx5_ifc_fte_match_set_misc_bits { - u8 reserved_at_0[0x20]; + u8 reserved_at_0[0x8]; + u8 source_sqn[0x18]; u8 reserved_at_20[0x10]; u8 source_port[0x10]; @@ -505,6 +512,17 @@ struct mlx5_ifc_e_switch_cap_bits { u8 reserved_at_20[0x7e0]; }; +struct mlx5_ifc_qos_cap_bits { + u8 packet_pacing[0x1]; + u8 reserved_0[0x1f]; + u8 reserved_1[0x20]; + u8 packet_pacing_max_rate[0x20]; + u8 packet_pacing_min_rate[0x20]; + u8 reserved_2[0x10]; + u8 packet_pacing_rate_table_size[0x10]; + u8 reserved_3[0x760]; +}; + struct mlx5_ifc_per_protocol_networking_offload_caps_bits { u8 csum_cap[0x1]; u8 vlan_cap[0x1]; @@ -744,7 +762,8 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 out_of_seq_cnt[0x1]; u8 vport_counters[0x1]; - u8 reserved_at_182[0x4]; + u8 retransmission_q_counters[0x1]; + u8 reserved_at_183[0x3]; u8 max_qp_cnt[0xa]; u8 pkey_table_size[0x10]; @@ -771,7 +790,9 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 log_max_msg[0x5]; u8 reserved_at_1c8[0x4]; u8 max_tc[0x4]; - u8 reserved_at_1d0[0x6]; + u8 reserved_at_1d0[0x1]; + u8 dcbx[0x1]; + u8 reserved_at_1d2[0x4]; u8 rol_s[0x1]; u8 rol_g[0x1]; u8 reserved_at_1d8[0x1]; @@ -803,7 +824,7 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 tph[0x1]; u8 rf[0x1]; u8 dct[0x1]; - u8 reserved_at_21b[0x1]; + u8 qos[0x1]; u8 eth_net_offloads[0x1]; u8 roce[0x1]; u8 atomic[0x1]; @@ -929,7 +950,15 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 cqe_compression_timeout[0x10]; u8 cqe_compression_max_num[0x10]; - u8 reserved_at_5e0[0x220]; + u8 reserved_at_5e0[0x10]; + u8 tag_matching[0x1]; + u8 rndv_offload_rc[0x1]; + u8 rndv_offload_dc[0x1]; + u8 log_tag_matching_list_sz[0x5]; + u8 reserved_at_5e8[0x3]; + u8 log_max_xrq[0x5]; + + u8 reserved_at_5f0[0x200]; }; enum mlx5_flow_destination_type { @@ -1967,7 +1996,7 @@ struct mlx5_ifc_qpc_bits { u8 reserved_at_560[0x5]; u8 rq_type[0x3]; - u8 srqn_rmpn[0x18]; + u8 srqn_rmpn_xrqn[0x18]; u8 reserved_at_580[0x8]; u8 rmsn[0x18]; @@ -2018,6 +2047,7 @@ union mlx5_ifc_hca_cap_union_bits { struct mlx5_ifc_flow_table_eswitch_cap_bits flow_table_eswitch_cap; struct mlx5_ifc_e_switch_cap_bits e_switch_cap; struct mlx5_ifc_vector_calc_cap_bits vector_calc_cap; + struct mlx5_ifc_qos_cap_bits qos_cap; u8 reserved_at_0[0x8000]; }; @@ -2244,8 +2274,9 @@ struct mlx5_ifc_sqc_bits { u8 reserved_at_40[0x8]; u8 cqn[0x18]; - u8 reserved_at_60[0xa0]; + u8 reserved_at_60[0x90]; + u8 packet_pacing_rate_limit_index[0x10]; u8 tis_lst_sz[0x10]; u8 reserved_at_110[0x10]; @@ -2593,7 +2624,7 @@ struct mlx5_ifc_dctc_bits { u8 reserved_at_98[0x8]; u8 reserved_at_a0[0x8]; - u8 srqn[0x18]; + u8 srqn_xrqn[0x18]; u8 reserved_at_c0[0x8]; u8 pd[0x18]; @@ -2645,6 +2676,7 @@ enum { enum { MLX5_CQ_PERIOD_MODE_START_FROM_EQE = 0x0, MLX5_CQ_PERIOD_MODE_START_FROM_CQE = 0x1, + MLX5_CQ_PERIOD_NUM_MODES }; struct mlx5_ifc_cqc_bits { @@ -2722,6 +2754,54 @@ struct mlx5_ifc_query_adapter_param_block_bits { u8 vsd_contd_psid[16][0x8]; }; +enum { + MLX5_XRQC_STATE_GOOD = 0x0, + MLX5_XRQC_STATE_ERROR = 0x1, +}; + +enum { + MLX5_XRQC_TOPOLOGY_NO_SPECIAL_TOPOLOGY = 0x0, + MLX5_XRQC_TOPOLOGY_TAG_MATCHING = 0x1, +}; + +enum { + MLX5_XRQC_OFFLOAD_RNDV = 0x1, +}; + +struct mlx5_ifc_tag_matching_topology_context_bits { + u8 log_matching_list_sz[0x4]; + u8 reserved_at_4[0xc]; + u8 append_next_index[0x10]; + + u8 sw_phase_cnt[0x10]; + u8 hw_phase_cnt[0x10]; + + u8 reserved_at_40[0x40]; +}; + +struct mlx5_ifc_xrqc_bits { + u8 state[0x4]; + u8 rlkey[0x1]; + u8 reserved_at_5[0xf]; + u8 topology[0x4]; + u8 reserved_at_18[0x4]; + u8 offload[0x4]; + + u8 reserved_at_20[0x8]; + u8 user_index[0x18]; + + u8 reserved_at_40[0x8]; + u8 cqn[0x18]; + + u8 reserved_at_60[0xa0]; + + struct mlx5_ifc_tag_matching_topology_context_bits tag_matching_topology_context; + + u8 reserved_at_180[0x180]; + + struct mlx5_ifc_wq_bits wq; +}; + union mlx5_ifc_modify_field_select_resize_field_select_auto_bits { struct mlx5_ifc_modify_field_select_bits modify_field_select; struct mlx5_ifc_resize_field_select_bits resize_field_select; @@ -3144,6 +3224,30 @@ struct mlx5_ifc_rst2init_qp_in_bits { u8 reserved_at_800[0x80]; }; +struct mlx5_ifc_query_xrq_out_bits { + u8 status[0x8]; + u8 reserved_at_8[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_at_40[0x40]; + + struct mlx5_ifc_xrqc_bits xrq_context; +}; + +struct mlx5_ifc_query_xrq_in_bits { + u8 opcode[0x10]; + u8 reserved_at_10[0x10]; + + u8 reserved_at_20[0x10]; + u8 op_mod[0x10]; + + u8 reserved_at_40[0x8]; + u8 xrqn[0x18]; + + u8 reserved_at_60[0x20]; +}; + struct mlx5_ifc_query_xrc_srq_out_bits { u8 status[0x8]; u8 reserved_at_8[0x18]; @@ -3547,7 +3651,27 @@ struct mlx5_ifc_query_q_counter_out_bits { u8 out_of_sequence[0x20]; - u8 reserved_at_1e0[0x620]; + u8 reserved_at_1e0[0x20]; + + u8 duplicate_request[0x20]; + + u8 reserved_at_220[0x20]; + + u8 rnr_nak_retry_err[0x20]; + + u8 reserved_at_260[0x20]; + + u8 packet_seq_err[0x20]; + + u8 reserved_at_2a0[0x20]; + + u8 implied_nak_seq_err[0x20]; + + u8 reserved_at_2e0[0x20]; + + u8 local_ack_timeout_err[0x20]; + + u8 reserved_at_320[0x4e0]; }; struct mlx5_ifc_query_q_counter_in_bits { @@ -4998,6 +5122,28 @@ struct mlx5_ifc_detach_from_mcg_in_bits { u8 multicast_gid[16][0x8]; }; +struct mlx5_ifc_destroy_xrq_out_bits { + u8 status[0x8]; + u8 reserved_at_8[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_at_40[0x40]; +}; + +struct mlx5_ifc_destroy_xrq_in_bits { + u8 opcode[0x10]; + u8 reserved_at_10[0x10]; + + u8 reserved_at_20[0x10]; + u8 op_mod[0x10]; + + u8 reserved_at_40[0x8]; + u8 xrqn[0x18]; + + u8 reserved_at_60[0x20]; +}; + struct mlx5_ifc_destroy_xrc_srq_out_bits { u8 status[0x8]; u8 reserved_at_8[0x18]; @@ -5583,6 +5729,30 @@ struct mlx5_ifc_dealloc_flow_counter_in_bits { u8 reserved_at_60[0x20]; }; +struct mlx5_ifc_create_xrq_out_bits { + u8 status[0x8]; + u8 reserved_at_8[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_at_40[0x8]; + u8 xrqn[0x18]; + + u8 reserved_at_60[0x20]; +}; + +struct mlx5_ifc_create_xrq_in_bits { + u8 opcode[0x10]; + u8 reserved_at_10[0x10]; + + u8 reserved_at_20[0x10]; + u8 op_mod[0x10]; + + u8 reserved_at_40[0x40]; + + struct mlx5_ifc_xrqc_bits xrq_context; +}; + struct mlx5_ifc_create_xrc_srq_out_bits { u8 status[0x8]; u8 reserved_at_8[0x18]; @@ -6124,6 +6294,29 @@ struct mlx5_ifc_attach_to_mcg_in_bits { u8 multicast_gid[16][0x8]; }; +struct mlx5_ifc_arm_xrq_out_bits { + u8 status[0x8]; + u8 reserved_at_8[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_at_40[0x40]; +}; + +struct mlx5_ifc_arm_xrq_in_bits { + u8 opcode[0x10]; + u8 reserved_at_10[0x10]; + + u8 reserved_at_20[0x10]; + u8 op_mod[0x10]; + + u8 reserved_at_40[0x8]; + u8 xrqn[0x18]; + + u8 reserved_at_60[0x10]; + u8 lwm[0x10]; +}; + struct mlx5_ifc_arm_xrc_srq_out_bits { u8 status[0x8]; u8 reserved_at_8[0x18]; @@ -6161,7 +6354,8 @@ struct mlx5_ifc_arm_rq_out_bits { }; enum { - MLX5_ARM_RQ_IN_OP_MOD_SRQ_ = 0x1, + MLX5_ARM_RQ_IN_OP_MOD_SRQ = 0x1, + MLX5_ARM_RQ_IN_OP_MOD_XRQ = 0x2, }; struct mlx5_ifc_arm_rq_in_bits { @@ -6354,6 +6548,30 @@ struct mlx5_ifc_add_vxlan_udp_dport_in_bits { u8 vxlan_udp_port[0x10]; }; +struct mlx5_ifc_set_rate_limit_out_bits { + u8 status[0x8]; + u8 reserved_at_8[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_at_40[0x40]; +}; + +struct mlx5_ifc_set_rate_limit_in_bits { + u8 opcode[0x10]; + u8 reserved_at_10[0x10]; + + u8 reserved_at_20[0x10]; + u8 op_mod[0x10]; + + u8 reserved_at_40[0x10]; + u8 rate_limit_index[0x10]; + + u8 reserved_at_60[0x20]; + + u8 rate_limit[0x20]; +}; + struct mlx5_ifc_access_register_out_bits { u8 status[0x8]; u8 reserved_at_8[0x18]; @@ -6478,12 +6696,15 @@ struct mlx5_ifc_pude_reg_bits { }; struct mlx5_ifc_ptys_reg_bits { - u8 reserved_at_0[0x8]; + u8 an_disable_cap[0x1]; + u8 an_disable_admin[0x1]; + u8 reserved_at_2[0x6]; u8 local_port[0x8]; u8 reserved_at_10[0xd]; u8 proto_mask[0x3]; - u8 reserved_at_20[0x40]; + u8 an_status[0x4]; + u8 reserved_at_24[0x3c]; u8 eth_proto_capability[0x20]; @@ -7444,4 +7665,34 @@ struct mlx5_ifc_mcia_reg_bits { u8 dword_11[0x20]; }; +struct mlx5_ifc_dcbx_param_bits { + u8 dcbx_cee_cap[0x1]; + u8 dcbx_ieee_cap[0x1]; + u8 dcbx_standby_cap[0x1]; + u8 reserved_at_0[0x5]; + u8 port_number[0x8]; + u8 reserved_at_10[0xa]; + u8 max_application_table_size[6]; + u8 reserved_at_20[0x15]; + u8 version_oper[0x3]; + u8 reserved_at_38[5]; + u8 version_admin[0x3]; + u8 willing_admin[0x1]; + u8 reserved_at_41[0x3]; + u8 pfc_cap_oper[0x4]; + u8 reserved_at_48[0x4]; + u8 pfc_cap_admin[0x4]; + u8 reserved_at_50[0x4]; + u8 num_of_tc_oper[0x4]; + u8 reserved_at_58[0x4]; + u8 num_of_tc_admin[0x4]; + u8 remote_willing[0x1]; + u8 reserved_at_61[3]; + u8 remote_pfc_cap[4]; + u8 reserved_at_68[0x14]; + u8 remote_num_of_tc[0x4]; + u8 reserved_at_80[0x18]; + u8 error[0x8]; + u8 reserved_at_a0[0x160]; +}; #endif /* MLX5_IFC_H */ -- cgit v0.10.2 From f2a4d086ed4c588d32fe9b7aa67fead7280e7bf1 Mon Sep 17 00:00:00 2001 From: William Tu Date: Fri, 10 Jun 2016 11:49:33 -0700 Subject: openvswitch: Add packet truncation support. The patch adds a new OVS action, OVS_ACTION_ATTR_TRUNC, in order to truncate packets. A 'max_len' is added for setting up the maximum packet size, and a 'cutlen' field is to record the number of bytes to trim the packet when the packet is outputting to a port, or when the packet is sent to userspace. Signed-off-by: William Tu Cc: Pravin Shelar Acked-by: Pravin B Shelar Signed-off-by: David S. Miller diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h index bb0d515..8274675 100644 --- a/include/uapi/linux/openvswitch.h +++ b/include/uapi/linux/openvswitch.h @@ -580,6 +580,10 @@ enum ovs_userspace_attr { #define OVS_USERSPACE_ATTR_MAX (__OVS_USERSPACE_ATTR_MAX - 1) +struct ovs_action_trunc { + uint32_t max_len; /* Max packet size in bytes. */ +}; + /** * struct ovs_action_push_mpls - %OVS_ACTION_ATTR_PUSH_MPLS action argument. * @mpls_lse: MPLS label stack entry to push. @@ -703,6 +707,7 @@ enum ovs_nat_attr { * enum ovs_action_attr - Action types. * * @OVS_ACTION_ATTR_OUTPUT: Output packet to port. + * @OVS_ACTION_ATTR_TRUNC: Output packet to port with truncated packet size. * @OVS_ACTION_ATTR_USERSPACE: Send packet to userspace according to nested * %OVS_USERSPACE_ATTR_* attributes. * @OVS_ACTION_ATTR_SET: Replaces the contents of an existing header. The @@ -756,6 +761,7 @@ enum ovs_action_attr { * The data must be zero for the unmasked * bits. */ OVS_ACTION_ATTR_CT, /* Nested OVS_CT_ATTR_* . */ + OVS_ACTION_ATTR_TRUNC, /* u32 struct ovs_action_trunc. */ __OVS_ACTION_ATTR_MAX, /* Nothing past this will be accepted * from userspace. */ diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 9a3eb7a..1ecbd77 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -750,6 +750,14 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, if (likely(vport)) { u16 mru = OVS_CB(skb)->mru; + u32 cutlen = OVS_CB(skb)->cutlen; + + if (unlikely(cutlen > 0)) { + if (skb->len - cutlen > ETH_HLEN) + pskb_trim(skb, skb->len - cutlen); + else + pskb_trim(skb, ETH_HLEN); + } if (likely(!mru || (skb->len <= mru + ETH_HLEN))) { ovs_vport_send(vport, skb); @@ -775,7 +783,8 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, static int output_userspace(struct datapath *dp, struct sk_buff *skb, struct sw_flow_key *key, const struct nlattr *attr, - const struct nlattr *actions, int actions_len) + const struct nlattr *actions, int actions_len, + uint32_t cutlen) { struct dp_upcall_info upcall; const struct nlattr *a; @@ -822,7 +831,7 @@ static int output_userspace(struct datapath *dp, struct sk_buff *skb, } /* End of switch. */ } - return ovs_dp_upcall(dp, skb, key, &upcall); + return ovs_dp_upcall(dp, skb, key, &upcall, cutlen); } static int sample(struct datapath *dp, struct sk_buff *skb, @@ -832,6 +841,7 @@ static int sample(struct datapath *dp, struct sk_buff *skb, const struct nlattr *acts_list = NULL; const struct nlattr *a; int rem; + u32 cutlen = 0; for (a = nla_data(attr), rem = nla_len(attr); rem > 0; a = nla_next(a, &rem)) { @@ -858,13 +868,24 @@ static int sample(struct datapath *dp, struct sk_buff *skb, return 0; /* The only known usage of sample action is having a single user-space + * action, or having a truncate action followed by a single user-space * action. Treat this usage as a special case. * The output_userspace() should clone the skb to be sent to the * user space. This skb will be consumed by its caller. */ + if (unlikely(nla_type(a) == OVS_ACTION_ATTR_TRUNC)) { + struct ovs_action_trunc *trunc = nla_data(a); + + if (skb->len > trunc->max_len) + cutlen = skb->len - trunc->max_len; + + a = nla_next(a, &rem); + } + if (likely(nla_type(a) == OVS_ACTION_ATTR_USERSPACE && nla_is_last(a, rem))) - return output_userspace(dp, skb, key, a, actions, actions_len); + return output_userspace(dp, skb, key, a, actions, + actions_len, cutlen); skb = skb_clone(skb, GFP_ATOMIC); if (!skb) @@ -1051,6 +1072,7 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, if (out_skb) do_output(dp, out_skb, prev_port, key); + OVS_CB(skb)->cutlen = 0; prev_port = -1; } @@ -1059,8 +1081,18 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, prev_port = nla_get_u32(a); break; + case OVS_ACTION_ATTR_TRUNC: { + struct ovs_action_trunc *trunc = nla_data(a); + + if (skb->len > trunc->max_len) + OVS_CB(skb)->cutlen = skb->len - trunc->max_len; + break; + } + case OVS_ACTION_ATTR_USERSPACE: - output_userspace(dp, skb, key, a, attr, len); + output_userspace(dp, skb, key, a, attr, + len, OVS_CB(skb)->cutlen); + OVS_CB(skb)->cutlen = 0; break; case OVS_ACTION_ATTR_HASH: diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 856bd8d..6739342 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -137,10 +137,12 @@ EXPORT_SYMBOL_GPL(lockdep_ovsl_is_held); static struct vport *new_vport(const struct vport_parms *); static int queue_gso_packets(struct datapath *dp, struct sk_buff *, const struct sw_flow_key *, - const struct dp_upcall_info *); + const struct dp_upcall_info *, + uint32_t cutlen); static int queue_userspace_packet(struct datapath *dp, struct sk_buff *, const struct sw_flow_key *, - const struct dp_upcall_info *); + const struct dp_upcall_info *, + uint32_t cutlen); /* Must be called with rcu_read_lock. */ static struct datapath *get_dp_rcu(struct net *net, int dp_ifindex) @@ -275,7 +277,7 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key) upcall.cmd = OVS_PACKET_CMD_MISS; upcall.portid = ovs_vport_find_upcall_portid(p, skb); upcall.mru = OVS_CB(skb)->mru; - error = ovs_dp_upcall(dp, skb, key, &upcall); + error = ovs_dp_upcall(dp, skb, key, &upcall, 0); if (unlikely(error)) kfree_skb(skb); else @@ -300,7 +302,8 @@ out: int ovs_dp_upcall(struct datapath *dp, struct sk_buff *skb, const struct sw_flow_key *key, - const struct dp_upcall_info *upcall_info) + const struct dp_upcall_info *upcall_info, + uint32_t cutlen) { struct dp_stats_percpu *stats; int err; @@ -311,9 +314,9 @@ int ovs_dp_upcall(struct datapath *dp, struct sk_buff *skb, } if (!skb_is_gso(skb)) - err = queue_userspace_packet(dp, skb, key, upcall_info); + err = queue_userspace_packet(dp, skb, key, upcall_info, cutlen); else - err = queue_gso_packets(dp, skb, key, upcall_info); + err = queue_gso_packets(dp, skb, key, upcall_info, cutlen); if (err) goto err; @@ -331,7 +334,8 @@ err: static int queue_gso_packets(struct datapath *dp, struct sk_buff *skb, const struct sw_flow_key *key, - const struct dp_upcall_info *upcall_info) + const struct dp_upcall_info *upcall_info, + uint32_t cutlen) { unsigned short gso_type = skb_shinfo(skb)->gso_type; struct sw_flow_key later_key; @@ -360,7 +364,7 @@ static int queue_gso_packets(struct datapath *dp, struct sk_buff *skb, if (gso_type & SKB_GSO_UDP && skb != segs) key = &later_key; - err = queue_userspace_packet(dp, skb, key, upcall_info); + err = queue_userspace_packet(dp, skb, key, upcall_info, cutlen); if (err) break; @@ -416,7 +420,8 @@ static void pad_packet(struct datapath *dp, struct sk_buff *skb) static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, const struct sw_flow_key *key, - const struct dp_upcall_info *upcall_info) + const struct dp_upcall_info *upcall_info, + uint32_t cutlen) { struct ovs_header *upcall; struct sk_buff *nskb = NULL; @@ -461,7 +466,7 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, else hlen = skb->len; - len = upcall_msg_size(upcall_info, hlen); + len = upcall_msg_size(upcall_info, hlen - cutlen); user_skb = genlmsg_new(len, GFP_ATOMIC); if (!user_skb) { err = -ENOMEM; @@ -515,9 +520,9 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, err = -ENOBUFS; goto out; } - nla->nla_len = nla_attr_size(skb->len); + nla->nla_len = nla_attr_size(skb->len - cutlen); - err = skb_zerocopy(user_skb, skb, skb->len, hlen); + err = skb_zerocopy(user_skb, skb, skb->len - cutlen, hlen); if (err) goto out; diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index 427e39a..ab85c1c 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -100,11 +100,13 @@ struct datapath { * @input_vport: The original vport packet came in on. This value is cached * when a packet is received by OVS. * @mru: The maximum received fragement size; 0 if the packet is not + * @cutlen: The number of bytes from the packet end to be removed. * fragmented. */ struct ovs_skb_cb { struct vport *input_vport; u16 mru; + u32 cutlen; }; #define OVS_CB(skb) ((struct ovs_skb_cb *)(skb)->cb) @@ -194,7 +196,8 @@ extern struct genl_family dp_vport_genl_family; void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key); void ovs_dp_detach_port(struct vport *); int ovs_dp_upcall(struct datapath *, struct sk_buff *, - const struct sw_flow_key *, const struct dp_upcall_info *); + const struct sw_flow_key *, const struct dp_upcall_info *, + uint32_t cutlen); const char *ovs_dp_name(const struct datapath *dp); struct sk_buff *ovs_vport_cmd_build_info(struct vport *, u32 pid, u32 seq, diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index 0bb650f..c78a6a1 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -2229,6 +2229,7 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, [OVS_ACTION_ATTR_SAMPLE] = (u32)-1, [OVS_ACTION_ATTR_HASH] = sizeof(struct ovs_action_hash), [OVS_ACTION_ATTR_CT] = (u32)-1, + [OVS_ACTION_ATTR_TRUNC] = sizeof(struct ovs_action_trunc), }; const struct ovs_action_push_vlan *vlan; int type = nla_type(a); @@ -2255,6 +2256,14 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, return -EINVAL; break; + case OVS_ACTION_ATTR_TRUNC: { + const struct ovs_action_trunc *trunc = nla_data(a); + + if (trunc->max_len < ETH_HLEN) + return -EINVAL; + break; + } + case OVS_ACTION_ATTR_HASH: { const struct ovs_action_hash *act_hash = nla_data(a); diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index 31cbc8c..6b21fd0 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -444,6 +444,7 @@ int ovs_vport_receive(struct vport *vport, struct sk_buff *skb, OVS_CB(skb)->input_vport = vport; OVS_CB(skb)->mru = 0; + OVS_CB(skb)->cutlen = 0; if (unlikely(dev_net(skb->dev) != ovs_dp_get_net(vport->dp))) { u32 mark; -- cgit v0.10.2 From a70b506efe899dc8d650eafcc0b11fc9ee746627 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 10 Jun 2016 21:19:06 +0200 Subject: bpf: enforce recursion limit on redirects Respect the stack's xmit_recursion limit for calls into dev_queue_xmit(). Currently, they are not handeled by the limiter when attached to clsact's egress parent, for example, and a buggy program redirecting it to the same device again could run into stack overflow eventually. It would be good if we could notify an admin to give him a chance to react. We reuse xmit_recursion instead of having one private to eBPF, so that the stack's current recursion depth will be taken into account as well. Follow-up to commit 3896d655f4d4 ("bpf: introduce bpf_clone_redirect() helper") and 27b29f63058d ("bpf: add bpf_redirect() helper"). Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 4f234b1..94eef35 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2389,6 +2389,8 @@ void synchronize_net(void); int init_dummy_netdev(struct net_device *dev); DECLARE_PER_CPU(int, xmit_recursion); +#define XMIT_RECURSION_LIMIT 10 + static inline int dev_recursion_level(void) { return this_cpu_read(xmit_recursion); diff --git a/net/core/dev.c b/net/core/dev.c index c43c9d2..b148357 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3144,8 +3144,6 @@ static void skb_update_prio(struct sk_buff *skb) DEFINE_PER_CPU(int, xmit_recursion); EXPORT_SYMBOL(xmit_recursion); -#define RECURSION_LIMIT 10 - /** * dev_loopback_xmit - loop back @skb * @net: network namespace this loopback is happening in @@ -3388,8 +3386,8 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv) int cpu = smp_processor_id(); /* ok because BHs are off */ if (txq->xmit_lock_owner != cpu) { - - if (__this_cpu_read(xmit_recursion) > RECURSION_LIMIT) + if (unlikely(__this_cpu_read(xmit_recursion) > + XMIT_RECURSION_LIMIT)) goto recursion_alert; skb = validate_xmit_skb(skb, dev); diff --git a/net/core/filter.c b/net/core/filter.c index 68adb5f..d11744d 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1603,9 +1603,36 @@ static const struct bpf_func_proto bpf_csum_diff_proto = { .arg5_type = ARG_ANYTHING, }; +static inline int __bpf_rx_skb(struct net_device *dev, struct sk_buff *skb) +{ + if (skb_at_tc_ingress(skb)) + skb_postpush_rcsum(skb, skb_mac_header(skb), skb->mac_len); + + return dev_forward_skb(dev, skb); +} + +static inline int __bpf_tx_skb(struct net_device *dev, struct sk_buff *skb) +{ + int ret; + + if (unlikely(__this_cpu_read(xmit_recursion) > XMIT_RECURSION_LIMIT)) { + net_crit_ratelimited("bpf: recursion limit reached on datapath, buggy bpf program?\n"); + kfree_skb(skb); + return -ENETDOWN; + } + + skb->dev = dev; + + __this_cpu_inc(xmit_recursion); + ret = dev_queue_xmit(skb); + __this_cpu_dec(xmit_recursion); + + return ret; +} + static u64 bpf_clone_redirect(u64 r1, u64 ifindex, u64 flags, u64 r4, u64 r5) { - struct sk_buff *skb = (struct sk_buff *) (long) r1, *skb2; + struct sk_buff *skb = (struct sk_buff *) (long) r1; struct net_device *dev; if (unlikely(flags & ~(BPF_F_INGRESS))) @@ -1615,19 +1642,12 @@ static u64 bpf_clone_redirect(u64 r1, u64 ifindex, u64 flags, u64 r4, u64 r5) if (unlikely(!dev)) return -EINVAL; - skb2 = skb_clone(skb, GFP_ATOMIC); - if (unlikely(!skb2)) + skb = skb_clone(skb, GFP_ATOMIC); + if (unlikely(!skb)) return -ENOMEM; - if (flags & BPF_F_INGRESS) { - if (skb_at_tc_ingress(skb2)) - skb_postpush_rcsum(skb2, skb_mac_header(skb2), - skb2->mac_len); - return dev_forward_skb(dev, skb2); - } - - skb2->dev = dev; - return dev_queue_xmit(skb2); + return flags & BPF_F_INGRESS ? + __bpf_rx_skb(dev, skb) : __bpf_tx_skb(dev, skb); } static const struct bpf_func_proto bpf_clone_redirect_proto = { @@ -1671,15 +1691,8 @@ int skb_do_redirect(struct sk_buff *skb) return -EINVAL; } - if (ri->flags & BPF_F_INGRESS) { - if (skb_at_tc_ingress(skb)) - skb_postpush_rcsum(skb, skb_mac_header(skb), - skb->mac_len); - return dev_forward_skb(dev, skb); - } - - skb->dev = dev; - return dev_queue_xmit(skb); + return ri->flags & BPF_F_INGRESS ? + __bpf_rx_skb(dev, skb) : __bpf_tx_skb(dev, skb); } static const struct bpf_func_proto bpf_redirect_proto = { -- cgit v0.10.2 From f7bd9e36ee4a4ce38e1cddd7effe6c0d9943285b Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 10 Jun 2016 21:19:07 +0200 Subject: bpf: reject wrong sized filters earlier Add a bpf_check_basics_ok() and reject filters that are of invalid size much earlier, so we don't do any useless work such as invoking bpf_prog_alloc(). Currently, rejection happens in bpf_check_classic() only, but it's really unnecessarily late and they should be rejected at earliest point. While at it, also clean up one bpf_prog_size() to make it consistent with the remaining invocations. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/net/core/filter.c b/net/core/filter.c index d11744d..df6860c 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -748,6 +748,17 @@ static bool chk_code_allowed(u16 code_to_probe) return codes[code_to_probe]; } +static bool bpf_check_basics_ok(const struct sock_filter *filter, + unsigned int flen) +{ + if (filter == NULL) + return false; + if (flen == 0 || flen > BPF_MAXINSNS) + return false; + + return true; +} + /** * bpf_check_classic - verify socket filter code * @filter: filter to verify @@ -768,9 +779,6 @@ static int bpf_check_classic(const struct sock_filter *filter, bool anc_found; int pc; - if (flen == 0 || flen > BPF_MAXINSNS) - return -EINVAL; - /* Check the filter code now */ for (pc = 0; pc < flen; pc++) { const struct sock_filter *ftest = &filter[pc]; @@ -1065,7 +1073,7 @@ int bpf_prog_create(struct bpf_prog **pfp, struct sock_fprog_kern *fprog) struct bpf_prog *fp; /* Make sure new filter is there and in the right amounts. */ - if (fprog->filter == NULL) + if (!bpf_check_basics_ok(fprog->filter, fprog->len)) return -EINVAL; fp = bpf_prog_alloc(bpf_prog_size(fprog->len), 0); @@ -1112,7 +1120,7 @@ int bpf_prog_create_from_user(struct bpf_prog **pfp, struct sock_fprog *fprog, int err; /* Make sure new filter is there and in the right amounts. */ - if (fprog->filter == NULL) + if (!bpf_check_basics_ok(fprog->filter, fprog->len)) return -EINVAL; fp = bpf_prog_alloc(bpf_prog_size(fprog->len), 0); @@ -1207,7 +1215,6 @@ static struct bpf_prog *__get_filter(struct sock_fprog *fprog, struct sock *sk) { unsigned int fsize = bpf_classic_proglen(fprog); - unsigned int bpf_fsize = bpf_prog_size(fprog->len); struct bpf_prog *prog; int err; @@ -1215,10 +1222,10 @@ struct bpf_prog *__get_filter(struct sock_fprog *fprog, struct sock *sk) return ERR_PTR(-EPERM); /* Make sure new filter is there and in the right amounts. */ - if (fprog->filter == NULL) + if (!bpf_check_basics_ok(fprog->filter, fprog->len)) return ERR_PTR(-EINVAL); - prog = bpf_prog_alloc(bpf_fsize, 0); + prog = bpf_prog_alloc(bpf_prog_size(fprog->len), 0); if (!prog) return ERR_PTR(-ENOMEM); -- cgit v0.10.2 From ea7f8277f9076d71ed6a925e2835ef4b85d6f5e1 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 10 Jun 2016 23:10:22 +0200 Subject: net, cls: allow for deleting all filters for given parent Add a possibility where the user can just specify the parent and all filters under that parent are then being purged. Currently, for example for scripting, one needs to specify pref/prio to have a well-defined number for 'tc filter del' command for addressing the previously created instance or additionally filter handle in case of priorities being the same. Improve usage by allowing the option for tc to specify the parent and removing the whole chain for that given parent. Example usage after patch, no tc changes required: # tc qdisc replace dev foo clsact # tc filter add dev foo egress bpf da obj ./bpf.o # tc filter add dev foo egress bpf da obj ./bpf.o # tc filter show dev foo egress filter protocol all pref 49151 bpf filter protocol all pref 49151 bpf handle 0x1 bpf.o:[classifier] direct-action filter protocol all pref 49152 bpf filter protocol all pref 49152 bpf handle 0x1 bpf.o:[classifier] direct-action # tc filter del dev foo egress # tc filter show dev foo egress # Previously, RTM_DELTFILTER requests with invalid prio of 0 were rejected, so only netlink requests with RTM_NEWTFILTER and NLM_F_CREATE flag were allowed where the kernel would auto-generate a pref/prio. We can piggyback on that and use prio of 0 as a wildcard for requests of RTM_DELTFILTER. For notifying tc netlink monitoring users (e.g. libnl uses this for caching), there are two options, that is, sending individual tfilter_notify() notifications for each tcf_proto, or sending a single one indicating wildcard removal. I tried both and there are pros and cons for each, eventually I decided for sending individual tfilter_notify(), so that user space can support this seamlessly and there won't be a mess of changing each and every application to make sure expectations from the kernel won't break when they don't understand single notification. Since linear chains don't really scale, I expect only a handful of classifiers to be attached at max for a given parent anyway. Signed-off-by: Daniel Borkmann Acked-by: Jamal Hadi Salim Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index aafa6bce..cca1ef5 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -103,6 +103,17 @@ static int tfilter_notify(struct net *net, struct sk_buff *oskb, struct nlmsghdr *n, struct tcf_proto *tp, unsigned long fh, int event); +static void tfilter_notify_chain(struct net *net, struct sk_buff *oskb, + struct nlmsghdr *n, + struct tcf_proto __rcu **chain, int event) +{ + struct tcf_proto __rcu **it_chain; + struct tcf_proto *tp; + + for (it_chain = chain; (tp = rtnl_dereference(*it_chain)) != NULL; + it_chain = &tp->next) + tfilter_notify(net, oskb, n, tp, 0, event); +} /* Select new prio value from the range, managed by kernel. */ @@ -156,11 +167,23 @@ replay: cl = 0; if (prio == 0) { - /* If no priority is given, user wants we allocated it. */ - if (n->nlmsg_type != RTM_NEWTFILTER || - !(n->nlmsg_flags & NLM_F_CREATE)) + switch (n->nlmsg_type) { + case RTM_DELTFILTER: + if (protocol || t->tcm_handle) + return -ENOENT; + break; + case RTM_NEWTFILTER: + /* If no priority is provided by the user, + * we allocate one. + */ + if (n->nlmsg_flags & NLM_F_CREATE) { + prio = TC_H_MAKE(0x80000000U, 0U); + break; + } + /* fall-through */ + default: return -ENOENT; - prio = TC_H_MAKE(0x80000000U, 0U); + } } /* Find head of filter chain. */ @@ -200,6 +223,12 @@ replay: err = -EINVAL; if (chain == NULL) goto errout; + if (n->nlmsg_type == RTM_DELTFILTER && prio == 0) { + tfilter_notify_chain(net, skb, n, chain, RTM_DELTFILTER); + tcf_destroy_chain(chain); + err = 0; + goto errout; + } /* Check the chain for existence of proto-tcf with this priority */ for (back = chain; -- cgit v0.10.2 From bc6e1ea32c26ead06063a882e802fff7ab6535c2 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 10 Jun 2016 22:30:27 +0100 Subject: rxrpc: Trim line-terminal whitespace Trim line-terminal whitespace in net/rxrpc/ Signed-off-by: David Howells Signed-off-by: David S. Miller diff --git a/net/rxrpc/ar-input.c b/net/rxrpc/ar-input.c index d7c2a0b..e0815a0 100644 --- a/net/rxrpc/ar-input.c +++ b/net/rxrpc/ar-input.c @@ -734,7 +734,7 @@ void rxrpc_data_ready(struct sock *sk) rxrpc_post_packet_to_local(local, skb); goto out; } - + if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA && (sp->hdr.callNumber == 0 || sp->hdr.seq == 0)) goto bad_message; diff --git a/net/rxrpc/ar-local.c b/net/rxrpc/ar-local.c index 701c42b..111f250 100644 --- a/net/rxrpc/ar-local.c +++ b/net/rxrpc/ar-local.c @@ -388,7 +388,7 @@ static void rxrpc_process_local_events(struct work_struct *work) _enter(""); atomic_inc(&local->usage); - + while ((skb = skb_dequeue(&local->event_queue))) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); -- cgit v0.10.2 From 0e119b41b7f23e08799fa8b1c9c1360d7da75815 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 10 Jun 2016 22:30:37 +0100 Subject: rxrpc: Limit the listening backlog Limit the socket incoming call backlog queue size so that a remote client can't pump in sufficient new calls that the server runs out of memory. Note that this is partially theoretical at the moment since whilst the number of calls is limited, the number of packets trying to set up new calls is not. This will be addressed in a later patch. If the caller of listen() specifies a backlog INT_MAX, then they get the current maximum; anything else greater than max_backlog or anything negative incurs EINVAL. The limit on the maximum queue size can be set by: echo N >/proc/sys/net/rxrpc/max_backlog where 4<=N<=32. Further, set the default backlog to 0, requiring listen() to be called before we start actually queueing new calls. Whilst this kind of is a change in the UAPI, the caller can't actually *accept* new calls anyway unless they've first called listen() to put the socket into the LISTENING state - thus the aforementioned new calls would otherwise just sit there, eating up kernel memory. (Note that sockets that don't have a non-zero service ID bound don't get incoming calls anyway.) Given that the default backlog is now 0, make the AFS filesystem call kernel_listen() to set the maximum backlog for itself. Possible improvements include: (1) Trimming a too-large backlog to max_backlog when listen is called. (2) Trimming the backlog value whenever the value is used so that changes to max_backlog are applied to an open socket automatically. Note that the AFS filesystem opens one socket and keeps it open for extended periods, so would miss out on changes to max_backlog. (3) Having a separate setting for the AFS filesystem. Signed-off-by: David Howells Signed-off-by: David S. Miller diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index 63cd9f9..4832de84 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -85,18 +85,14 @@ int afs_open_socket(void) skb_queue_head_init(&afs_incoming_calls); + ret = -ENOMEM; afs_async_calls = create_singlethread_workqueue("kafsd"); - if (!afs_async_calls) { - _leave(" = -ENOMEM [wq]"); - return -ENOMEM; - } + if (!afs_async_calls) + goto error_0; ret = sock_create_kern(&init_net, AF_RXRPC, SOCK_DGRAM, PF_INET, &socket); - if (ret < 0) { - destroy_workqueue(afs_async_calls); - _leave(" = %d [socket]", ret); - return ret; - } + if (ret < 0) + goto error_1; socket->sk->sk_allocation = GFP_NOFS; @@ -111,18 +107,26 @@ int afs_open_socket(void) sizeof(srx.transport.sin.sin_addr)); ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx)); - if (ret < 0) { - sock_release(socket); - destroy_workqueue(afs_async_calls); - _leave(" = %d [bind]", ret); - return ret; - } + if (ret < 0) + goto error_2; + + ret = kernel_listen(socket, INT_MAX); + if (ret < 0) + goto error_2; rxrpc_kernel_intercept_rx_messages(socket, afs_rx_interceptor); afs_socket = socket; _leave(" = 0"); return 0; + +error_2: + sock_release(socket); +error_1: + destroy_workqueue(afs_async_calls); +error_0: + _leave(" = %d", ret); + return ret; } /* diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 38512a2..a1bcb0e 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -33,8 +33,6 @@ unsigned int rxrpc_debug; // = RXRPC_DEBUG_KPROTO; module_param_named(debug, rxrpc_debug, uint, S_IWUSR | S_IRUGO); MODULE_PARM_DESC(debug, "RxRPC debugging mask"); -static int sysctl_rxrpc_max_qlen __read_mostly = 10; - static struct proto rxrpc_proto; static const struct proto_ops rxrpc_rpc_ops; @@ -191,6 +189,7 @@ static int rxrpc_listen(struct socket *sock, int backlog) { struct sock *sk = sock->sk; struct rxrpc_sock *rx = rxrpc_sk(sk); + unsigned int max; int ret; _enter("%p,%d", rx, backlog); @@ -201,17 +200,21 @@ static int rxrpc_listen(struct socket *sock, int backlog) case RXRPC_UNBOUND: ret = -EADDRNOTAVAIL; break; - case RXRPC_CLIENT_UNBOUND: - case RXRPC_CLIENT_BOUND: - default: - ret = -EBUSY; - break; case RXRPC_SERVER_BOUND: ASSERT(rx->local != NULL); + max = READ_ONCE(rxrpc_max_backlog); + ret = -EINVAL; + if (backlog == INT_MAX) + backlog = max; + else if (backlog < 0 || backlog > max) + break; sk->sk_max_ack_backlog = backlog; rx->sk.sk_state = RXRPC_SERVER_LISTENING; ret = 0; break; + default: + ret = -EBUSY; + break; } release_sock(&rx->sk); @@ -591,7 +594,7 @@ static int rxrpc_create(struct net *net, struct socket *sock, int protocol, sock_init_data(sock, sk); sk->sk_state = RXRPC_UNBOUND; sk->sk_write_space = rxrpc_write_space; - sk->sk_max_ack_backlog = sysctl_rxrpc_max_qlen; + sk->sk_max_ack_backlog = 0; sk->sk_destruct = rxrpc_sock_destructor; rx = rxrpc_sk(sk); diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index b89dcdc..f715cca 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -641,6 +641,7 @@ extern const struct rxrpc_security rxrpc_no_security; /* * misc.c */ +extern unsigned int rxrpc_max_backlog __read_mostly; extern unsigned int rxrpc_requested_ack_delay; extern unsigned int rxrpc_soft_ack_delay; extern unsigned int rxrpc_idle_ack_delay; diff --git a/net/rxrpc/misc.c b/net/rxrpc/misc.c index 1afe987..bdc5e42 100644 --- a/net/rxrpc/misc.c +++ b/net/rxrpc/misc.c @@ -15,6 +15,12 @@ #include "ar-internal.h" /* + * The maximum listening backlog queue size that may be set on a socket by + * listen(). + */ +unsigned int rxrpc_max_backlog __read_mostly = 10; + +/* * How long to wait before scheduling ACK generation after seeing a * packet with RXRPC_REQUEST_ACK set (in jiffies). */ diff --git a/net/rxrpc/sysctl.c b/net/rxrpc/sysctl.c index d20ed57..a99690a 100644 --- a/net/rxrpc/sysctl.c +++ b/net/rxrpc/sysctl.c @@ -18,6 +18,7 @@ static struct ctl_table_header *rxrpc_sysctl_reg_table; static const unsigned int zero = 0; static const unsigned int one = 1; static const unsigned int four = 4; +static const unsigned int thirtytwo = 32; static const unsigned int n_65535 = 65535; static const unsigned int n_max_acks = RXRPC_MAXACKS; @@ -100,6 +101,15 @@ static struct ctl_table rxrpc_sysctl_table[] = { /* Non-time values */ { + .procname = "max_backlog", + .data = &rxrpc_max_backlog, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = (void *)&four, + .extra2 = (void *)&thirtytwo, + }, + { .procname = "rx_window_size", .data = &rxrpc_rx_window_size, .maxlen = sizeof(unsigned int), -- cgit v0.10.2 From 112b558d025712c0bbcefa3d07a4433dd3e32d27 Mon Sep 17 00:00:00 2001 From: Hauke Mehrtens Date: Sun, 5 Jun 2016 23:41:11 +0200 Subject: NET: PHY: adds driver for Intel XWAY PHY This adds support for the Intel (former Lantiq) XWAY 11G and 22E PHYs. These PHYs are also named PEF 7061, PEF 7071, PEF 7072. Signed-off-by: John Crispin Signed-off-by: Hauke Mehrtens Signed-off-by: David S. Miller diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig index 6dad9a9..37d40c1 100644 --- a/drivers/net/phy/Kconfig +++ b/drivers/net/phy/Kconfig @@ -271,6 +271,14 @@ config MDIO_BCM_IPROC This module provides a driver for the MDIO busses found in the Broadcom iProc SoC's. +config INTEL_XWAY_PHY + tristate "Driver for Intel XWAY PHYs" + ---help--- + Supports the Intel XWAY (former Lantiq) 11G and 22E PHYs. + These PHYs are marked as standalone chips under the names + PEF 7061, PEF 7071 and PEF 7072 or integrated into the Intel + SoCs xRX200, xRX300, xRX330, xRX350 and xRX550. + endif # PHYLIB config MICREL_KS8995MA diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile index fcdbb92..c26b651 100644 --- a/drivers/net/phy/Makefile +++ b/drivers/net/phy/Makefile @@ -44,3 +44,4 @@ obj-$(CONFIG_MDIO_MOXART) += mdio-moxart.o obj-$(CONFIG_MDIO_BCM_UNIMAC) += mdio-bcm-unimac.o obj-$(CONFIG_MICROCHIP_PHY) += microchip.o obj-$(CONFIG_MDIO_BCM_IPROC) += mdio-bcm-iproc.o +obj-$(CONFIG_INTEL_XWAY_PHY) += intel-xway.o diff --git a/drivers/net/phy/intel-xway.c b/drivers/net/phy/intel-xway.c new file mode 100644 index 0000000..c300ab5 --- /dev/null +++ b/drivers/net/phy/intel-xway.c @@ -0,0 +1,376 @@ +/* + * Copyright (C) 2012 Daniel Schwierzeck + * Copyright (C) 2016 Hauke Mehrtens + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include +#include +#include +#include + +#define XWAY_MDIO_IMASK 0x19 /* interrupt mask */ +#define XWAY_MDIO_ISTAT 0x1A /* interrupt status */ + +#define XWAY_MDIO_INIT_WOL BIT(15) /* Wake-On-LAN */ +#define XWAY_MDIO_INIT_MSRE BIT(14) +#define XWAY_MDIO_INIT_NPRX BIT(13) +#define XWAY_MDIO_INIT_NPTX BIT(12) +#define XWAY_MDIO_INIT_ANE BIT(11) /* Auto-Neg error */ +#define XWAY_MDIO_INIT_ANC BIT(10) /* Auto-Neg complete */ +#define XWAY_MDIO_INIT_ADSC BIT(5) /* Link auto-downspeed detect */ +#define XWAY_MDIO_INIT_MPIPC BIT(4) +#define XWAY_MDIO_INIT_MDIXC BIT(3) +#define XWAY_MDIO_INIT_DXMC BIT(2) /* Duplex mode change */ +#define XWAY_MDIO_INIT_LSPC BIT(1) /* Link speed change */ +#define XWAY_MDIO_INIT_LSTC BIT(0) /* Link state change */ +#define XWAY_MDIO_INIT_MASK (XWAY_MDIO_INIT_LSTC | \ + XWAY_MDIO_INIT_ADSC) + +#define ADVERTISED_MPD BIT(10) /* Multi-port device */ + +/* LED Configuration */ +#define XWAY_MMD_LEDCH 0x01E0 +/* Inverse of SCAN Function */ +#define XWAY_MMD_LEDCH_NACS_NONE 0x0000 +#define XWAY_MMD_LEDCH_NACS_LINK 0x0001 +#define XWAY_MMD_LEDCH_NACS_PDOWN 0x0002 +#define XWAY_MMD_LEDCH_NACS_EEE 0x0003 +#define XWAY_MMD_LEDCH_NACS_ANEG 0x0004 +#define XWAY_MMD_LEDCH_NACS_ABIST 0x0005 +#define XWAY_MMD_LEDCH_NACS_CDIAG 0x0006 +#define XWAY_MMD_LEDCH_NACS_TEST 0x0007 +/* Slow Blink Frequency */ +#define XWAY_MMD_LEDCH_SBF_F02HZ 0x0000 +#define XWAY_MMD_LEDCH_SBF_F04HZ 0x0010 +#define XWAY_MMD_LEDCH_SBF_F08HZ 0x0020 +#define XWAY_MMD_LEDCH_SBF_F16HZ 0x0030 +/* Fast Blink Frequency */ +#define XWAY_MMD_LEDCH_FBF_F02HZ 0x0000 +#define XWAY_MMD_LEDCH_FBF_F04HZ 0x0040 +#define XWAY_MMD_LEDCH_FBF_F08HZ 0x0080 +#define XWAY_MMD_LEDCH_FBF_F16HZ 0x00C0 +/* LED Configuration */ +#define XWAY_MMD_LEDCL 0x01E1 +/* Complex Blinking Configuration */ +#define XWAY_MMD_LEDCH_CBLINK_NONE 0x0000 +#define XWAY_MMD_LEDCH_CBLINK_LINK 0x0001 +#define XWAY_MMD_LEDCH_CBLINK_PDOWN 0x0002 +#define XWAY_MMD_LEDCH_CBLINK_EEE 0x0003 +#define XWAY_MMD_LEDCH_CBLINK_ANEG 0x0004 +#define XWAY_MMD_LEDCH_CBLINK_ABIST 0x0005 +#define XWAY_MMD_LEDCH_CBLINK_CDIAG 0x0006 +#define XWAY_MMD_LEDCH_CBLINK_TEST 0x0007 +/* Complex SCAN Configuration */ +#define XWAY_MMD_LEDCH_SCAN_NONE 0x0000 +#define XWAY_MMD_LEDCH_SCAN_LINK 0x0010 +#define XWAY_MMD_LEDCH_SCAN_PDOWN 0x0020 +#define XWAY_MMD_LEDCH_SCAN_EEE 0x0030 +#define XWAY_MMD_LEDCH_SCAN_ANEG 0x0040 +#define XWAY_MMD_LEDCH_SCAN_ABIST 0x0050 +#define XWAY_MMD_LEDCH_SCAN_CDIAG 0x0060 +#define XWAY_MMD_LEDCH_SCAN_TEST 0x0070 +/* Configuration for LED Pin x */ +#define XWAY_MMD_LED0H 0x01E2 +/* Fast Blinking Configuration */ +#define XWAY_MMD_LEDxH_BLINKF_MASK 0x000F +#define XWAY_MMD_LEDxH_BLINKF_NONE 0x0000 +#define XWAY_MMD_LEDxH_BLINKF_LINK10 0x0001 +#define XWAY_MMD_LEDxH_BLINKF_LINK100 0x0002 +#define XWAY_MMD_LEDxH_BLINKF_LINK10X 0x0003 +#define XWAY_MMD_LEDxH_BLINKF_LINK1000 0x0004 +#define XWAY_MMD_LEDxH_BLINKF_LINK10_0 0x0005 +#define XWAY_MMD_LEDxH_BLINKF_LINK100X 0x0006 +#define XWAY_MMD_LEDxH_BLINKF_LINK10XX 0x0007 +#define XWAY_MMD_LEDxH_BLINKF_PDOWN 0x0008 +#define XWAY_MMD_LEDxH_BLINKF_EEE 0x0009 +#define XWAY_MMD_LEDxH_BLINKF_ANEG 0x000A +#define XWAY_MMD_LEDxH_BLINKF_ABIST 0x000B +#define XWAY_MMD_LEDxH_BLINKF_CDIAG 0x000C +/* Constant On Configuration */ +#define XWAY_MMD_LEDxH_CON_MASK 0x00F0 +#define XWAY_MMD_LEDxH_CON_NONE 0x0000 +#define XWAY_MMD_LEDxH_CON_LINK10 0x0010 +#define XWAY_MMD_LEDxH_CON_LINK100 0x0020 +#define XWAY_MMD_LEDxH_CON_LINK10X 0x0030 +#define XWAY_MMD_LEDxH_CON_LINK1000 0x0040 +#define XWAY_MMD_LEDxH_CON_LINK10_0 0x0050 +#define XWAY_MMD_LEDxH_CON_LINK100X 0x0060 +#define XWAY_MMD_LEDxH_CON_LINK10XX 0x0070 +#define XWAY_MMD_LEDxH_CON_PDOWN 0x0080 +#define XWAY_MMD_LEDxH_CON_EEE 0x0090 +#define XWAY_MMD_LEDxH_CON_ANEG 0x00A0 +#define XWAY_MMD_LEDxH_CON_ABIST 0x00B0 +#define XWAY_MMD_LEDxH_CON_CDIAG 0x00C0 +#define XWAY_MMD_LEDxH_CON_COPPER 0x00D0 +#define XWAY_MMD_LEDxH_CON_FIBER 0x00E0 +/* Configuration for LED Pin x */ +#define XWAY_MMD_LED0L 0x01E3 +/* Pulsing Configuration */ +#define XWAY_MMD_LEDxL_PULSE_MASK 0x000F +#define XWAY_MMD_LEDxL_PULSE_NONE 0x0000 +#define XWAY_MMD_LEDxL_PULSE_TXACT 0x0001 +#define XWAY_MMD_LEDxL_PULSE_RXACT 0x0002 +#define XWAY_MMD_LEDxL_PULSE_COL 0x0004 +/* Slow Blinking Configuration */ +#define XWAY_MMD_LEDxL_BLINKS_MASK 0x00F0 +#define XWAY_MMD_LEDxL_BLINKS_NONE 0x0000 +#define XWAY_MMD_LEDxL_BLINKS_LINK10 0x0010 +#define XWAY_MMD_LEDxL_BLINKS_LINK100 0x0020 +#define XWAY_MMD_LEDxL_BLINKS_LINK10X 0x0030 +#define XWAY_MMD_LEDxL_BLINKS_LINK1000 0x0040 +#define XWAY_MMD_LEDxL_BLINKS_LINK10_0 0x0050 +#define XWAY_MMD_LEDxL_BLINKS_LINK100X 0x0060 +#define XWAY_MMD_LEDxL_BLINKS_LINK10XX 0x0070 +#define XWAY_MMD_LEDxL_BLINKS_PDOWN 0x0080 +#define XWAY_MMD_LEDxL_BLINKS_EEE 0x0090 +#define XWAY_MMD_LEDxL_BLINKS_ANEG 0x00A0 +#define XWAY_MMD_LEDxL_BLINKS_ABIST 0x00B0 +#define XWAY_MMD_LEDxL_BLINKS_CDIAG 0x00C0 +#define XWAY_MMD_LED1H 0x01E4 +#define XWAY_MMD_LED1L 0x01E5 +#define XWAY_MMD_LED2H 0x01E6 +#define XWAY_MMD_LED2L 0x01E7 +#define XWAY_MMD_LED3H 0x01E8 +#define XWAY_MMD_LED3L 0x01E9 + +#define PHY_ID_PHY11G_1_3 0x030260D1 +#define PHY_ID_PHY22F_1_3 0x030260E1 +#define PHY_ID_PHY11G_1_4 0xD565A400 +#define PHY_ID_PHY22F_1_4 0xD565A410 +#define PHY_ID_PHY11G_1_5 0xD565A401 +#define PHY_ID_PHY22F_1_5 0xD565A411 +#define PHY_ID_PHY11G_VR9 0xD565A409 +#define PHY_ID_PHY22F_VR9 0xD565A419 + +static int xway_gphy_config_init(struct phy_device *phydev) +{ + int err; + u32 ledxh; + u32 ledxl; + + /* Mask all interrupts */ + err = phy_write(phydev, XWAY_MDIO_IMASK, 0); + if (err) + return err; + + /* Clear all pending interrupts */ + phy_read(phydev, XWAY_MDIO_ISTAT); + + phy_write_mmd_indirect(phydev, XWAY_MMD_LEDCH, MDIO_MMD_VEND2, + XWAY_MMD_LEDCH_NACS_NONE | + XWAY_MMD_LEDCH_SBF_F02HZ | + XWAY_MMD_LEDCH_FBF_F16HZ); + phy_write_mmd_indirect(phydev, XWAY_MMD_LEDCL, MDIO_MMD_VEND2, + XWAY_MMD_LEDCH_CBLINK_NONE | + XWAY_MMD_LEDCH_SCAN_NONE); + + /** + * In most cases only one LED is connected to this phy, so + * configure them all to constant on and pulse mode. LED3 is + * only available in some packages, leave it in its reset + * configuration. + */ + ledxh = XWAY_MMD_LEDxH_BLINKF_NONE | XWAY_MMD_LEDxH_CON_LINK10XX; + ledxl = XWAY_MMD_LEDxL_PULSE_TXACT | XWAY_MMD_LEDxL_PULSE_RXACT | + XWAY_MMD_LEDxL_BLINKS_NONE; + phy_write_mmd_indirect(phydev, XWAY_MMD_LED0H, MDIO_MMD_VEND2, ledxh); + phy_write_mmd_indirect(phydev, XWAY_MMD_LED0L, MDIO_MMD_VEND2, ledxl); + phy_write_mmd_indirect(phydev, XWAY_MMD_LED1H, MDIO_MMD_VEND2, ledxh); + phy_write_mmd_indirect(phydev, XWAY_MMD_LED1L, MDIO_MMD_VEND2, ledxl); + phy_write_mmd_indirect(phydev, XWAY_MMD_LED2H, MDIO_MMD_VEND2, ledxh); + phy_write_mmd_indirect(phydev, XWAY_MMD_LED2L, MDIO_MMD_VEND2, ledxl); + + return 0; +} + +static int xway_gphy14_config_aneg(struct phy_device *phydev) +{ + int reg, err; + + /* Advertise as multi-port device, see IEEE802.3-2002 40.5.1.1 */ + /* This is a workaround for an errata in rev < 1.5 devices */ + reg = phy_read(phydev, MII_CTRL1000); + reg |= ADVERTISED_MPD; + err = phy_write(phydev, MII_CTRL1000, reg); + if (err) + return err; + + return genphy_config_aneg(phydev); +} + +static int xway_gphy_ack_interrupt(struct phy_device *phydev) +{ + int reg; + + reg = phy_read(phydev, XWAY_MDIO_ISTAT); + return (reg < 0) ? reg : 0; +} + +static int xway_gphy_did_interrupt(struct phy_device *phydev) +{ + int reg; + + reg = phy_read(phydev, XWAY_MDIO_ISTAT); + return reg & XWAY_MDIO_INIT_MASK; +} + +static int xway_gphy_config_intr(struct phy_device *phydev) +{ + u16 mask = 0; + + if (phydev->interrupts == PHY_INTERRUPT_ENABLED) + mask = XWAY_MDIO_INIT_MASK; + + return phy_write(phydev, XWAY_MDIO_IMASK, mask); +} + +static struct phy_driver xway_gphy[] = { + { + .phy_id = PHY_ID_PHY11G_1_3, + .phy_id_mask = 0xffffffff, + .name = "Intel XWAY PHY11G (PEF 7071/PEF 7072) v1.3", + .features = (PHY_GBIT_FEATURES | SUPPORTED_Pause | + SUPPORTED_Asym_Pause), + .flags = PHY_HAS_INTERRUPT, + .config_init = xway_gphy_config_init, + .config_aneg = xway_gphy14_config_aneg, + .read_status = genphy_read_status, + .ack_interrupt = xway_gphy_ack_interrupt, + .did_interrupt = xway_gphy_did_interrupt, + .config_intr = xway_gphy_config_intr, + .suspend = genphy_suspend, + .resume = genphy_resume, + }, { + .phy_id = PHY_ID_PHY22F_1_3, + .phy_id_mask = 0xffffffff, + .name = "Intel XWAY PHY22F (PEF 7061) v1.3", + .features = (PHY_BASIC_FEATURES | SUPPORTED_Pause | + SUPPORTED_Asym_Pause), + .flags = PHY_HAS_INTERRUPT, + .config_init = xway_gphy_config_init, + .config_aneg = xway_gphy14_config_aneg, + .read_status = genphy_read_status, + .ack_interrupt = xway_gphy_ack_interrupt, + .did_interrupt = xway_gphy_did_interrupt, + .config_intr = xway_gphy_config_intr, + .suspend = genphy_suspend, + .resume = genphy_resume, + }, { + .phy_id = PHY_ID_PHY11G_1_4, + .phy_id_mask = 0xffffffff, + .name = "Intel XWAY PHY11G (PEF 7071/PEF 7072) v1.4", + .features = (PHY_GBIT_FEATURES | SUPPORTED_Pause | + SUPPORTED_Asym_Pause), + .flags = PHY_HAS_INTERRUPT, + .config_init = xway_gphy_config_init, + .config_aneg = xway_gphy14_config_aneg, + .read_status = genphy_read_status, + .ack_interrupt = xway_gphy_ack_interrupt, + .did_interrupt = xway_gphy_did_interrupt, + .config_intr = xway_gphy_config_intr, + .suspend = genphy_suspend, + .resume = genphy_resume, + }, { + .phy_id = PHY_ID_PHY22F_1_4, + .phy_id_mask = 0xffffffff, + .name = "Intel XWAY PHY22F (PEF 7061) v1.4", + .features = (PHY_BASIC_FEATURES | SUPPORTED_Pause | + SUPPORTED_Asym_Pause), + .flags = PHY_HAS_INTERRUPT, + .config_init = xway_gphy_config_init, + .config_aneg = xway_gphy14_config_aneg, + .read_status = genphy_read_status, + .ack_interrupt = xway_gphy_ack_interrupt, + .did_interrupt = xway_gphy_did_interrupt, + .config_intr = xway_gphy_config_intr, + .suspend = genphy_suspend, + .resume = genphy_resume, + }, { + .phy_id = PHY_ID_PHY11G_1_5, + .phy_id_mask = 0xffffffff, + .name = "Intel XWAY PHY11G (PEF 7071/PEF 7072) v1.5 / v1.6", + .features = (PHY_GBIT_FEATURES | SUPPORTED_Pause | + SUPPORTED_Asym_Pause), + .flags = PHY_HAS_INTERRUPT, + .config_init = xway_gphy_config_init, + .config_aneg = genphy_config_aneg, + .read_status = genphy_read_status, + .ack_interrupt = xway_gphy_ack_interrupt, + .did_interrupt = xway_gphy_did_interrupt, + .config_intr = xway_gphy_config_intr, + .suspend = genphy_suspend, + .resume = genphy_resume, + }, { + .phy_id = PHY_ID_PHY22F_1_5, + .phy_id_mask = 0xffffffff, + .name = "Intel XWAY PHY22F (PEF 7061) v1.5 / v1.6", + .features = (PHY_BASIC_FEATURES | SUPPORTED_Pause | + SUPPORTED_Asym_Pause), + .flags = PHY_HAS_INTERRUPT, + .config_init = xway_gphy_config_init, + .config_aneg = genphy_config_aneg, + .read_status = genphy_read_status, + .ack_interrupt = xway_gphy_ack_interrupt, + .did_interrupt = xway_gphy_did_interrupt, + .config_intr = xway_gphy_config_intr, + .suspend = genphy_suspend, + .resume = genphy_resume, + }, { + .phy_id = PHY_ID_PHY11G_VR9, + .phy_id_mask = 0xffffffff, + .name = "Intel XWAY PHY11G (xRX integrated)", + .features = (PHY_GBIT_FEATURES | SUPPORTED_Pause | + SUPPORTED_Asym_Pause), + .flags = PHY_HAS_INTERRUPT, + .config_init = xway_gphy_config_init, + .config_aneg = genphy_config_aneg, + .read_status = genphy_read_status, + .ack_interrupt = xway_gphy_ack_interrupt, + .did_interrupt = xway_gphy_did_interrupt, + .config_intr = xway_gphy_config_intr, + .suspend = genphy_suspend, + .resume = genphy_resume, + }, { + .phy_id = PHY_ID_PHY22F_VR9, + .phy_id_mask = 0xffffffff, + .name = "Intel XWAY PHY22F (xRX integrated)", + .features = (PHY_BASIC_FEATURES | SUPPORTED_Pause | + SUPPORTED_Asym_Pause), + .flags = PHY_HAS_INTERRUPT, + .config_init = xway_gphy_config_init, + .config_aneg = genphy_config_aneg, + .read_status = genphy_read_status, + .ack_interrupt = xway_gphy_ack_interrupt, + .did_interrupt = xway_gphy_did_interrupt, + .config_intr = xway_gphy_config_intr, + .suspend = genphy_suspend, + .resume = genphy_resume, + }, +}; +module_phy_driver(xway_gphy); + +static struct mdio_device_id __maybe_unused xway_gphy_tbl[] = { + { PHY_ID_PHY11G_1_3, 0xffffffff }, + { PHY_ID_PHY22F_1_3, 0xffffffff }, + { PHY_ID_PHY11G_1_4, 0xffffffff }, + { PHY_ID_PHY22F_1_4, 0xffffffff }, + { PHY_ID_PHY11G_1_5, 0xffffffff }, + { PHY_ID_PHY22F_1_5, 0xffffffff }, + { PHY_ID_PHY11G_VR9, 0xffffffff }, + { PHY_ID_PHY22F_VR9, 0xffffffff }, + { } +}; +MODULE_DEVICE_TABLE(mdio, xway_gphy_tbl); + +MODULE_DESCRIPTION("Intel XWAY PHY driver"); +MODULE_LICENSE("GPL"); -- cgit v0.10.2 From 231edca97f947f50e18fa2f6d9c8285c3314232b Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Wed, 8 Jun 2016 01:03:45 +0530 Subject: RDS: IB: Remove deprecated create_workqueue alloc_workqueue replaces deprecated create_workqueue(). Since the driver is infiniband which can be used as block device and the workqueue seems involved in regular operation of the device, so a dedicated workqueue has been used with WQ_MEM_RECLAIM set to guarantee forward progress under memory pressure. Since there are only a fixed number of work items, explicit concurrency limit is unnecessary here. Signed-off-by: Bhaktipriya Shridhar Acked-by: Santosh Shilimkar Signed-off-by: David S. Miller diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c index f7164ac..a0f21b6 100644 --- a/net/rds/ib_rdma.c +++ b/net/rds/ib_rdma.c @@ -618,7 +618,7 @@ struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_ibdev, int rds_ib_mr_init(void) { - rds_ib_mr_wq = create_workqueue("rds_mr_flushd"); + rds_ib_mr_wq = alloc_workqueue("rds_mr_flushd", WQ_MEM_RECLAIM, 0); if (!rds_ib_mr_wq) return -ENOMEM; return 0; -- cgit v0.10.2 From 678ece30226ac05e95768d5f70db53a475569d37 Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Wed, 8 Jun 2016 16:09:17 +0300 Subject: virtio_net: add _UAPI prefix to virtio_net header guards This gives better namespacing and prevents conflicts with no-uapi version of virtio_net header that will be introduced in the following patch. Signed-off-by: Mike Rapoport Signed-off-by: David S. Miller diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h index 0da0e3a..fc353b5 100644 --- a/include/uapi/linux/virtio_net.h +++ b/include/uapi/linux/virtio_net.h @@ -1,5 +1,5 @@ -#ifndef _LINUX_VIRTIO_NET_H -#define _LINUX_VIRTIO_NET_H +#ifndef _UAPI_LINUX_VIRTIO_NET_H +#define _UAPI_LINUX_VIRTIO_NET_H /* This header is BSD licensed so anyone can use the definitions to implement * compatible drivers/servers. * @@ -245,4 +245,4 @@ struct virtio_net_ctrl_mq { #define VIRTIO_NET_CTRL_GUEST_OFFLOADS 5 #define VIRTIO_NET_CTRL_GUEST_OFFLOADS_SET 0 -#endif /* _LINUX_VIRTIO_NET_H */ +#endif /* _UAPI_LINUX_VIRTIO_NET_H */ -- cgit v0.10.2 From fd2a0437dc33b6425cabf74cc7fc7fdba6d5903b Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Wed, 8 Jun 2016 16:09:18 +0300 Subject: virtio_net: introduce virtio_net_hdr_{from,to}_skb The code for conversion between virtio_net_hdr and skb GSO info is duplicated at several places. Let's put it to a common place to allow reuse. Signed-off-by: Mike Rapoport Signed-off-by: David S. Miller diff --git a/include/linux/virtio_net.h b/include/linux/virtio_net.h new file mode 100644 index 0000000..1c912f8 --- /dev/null +++ b/include/linux/virtio_net.h @@ -0,0 +1,101 @@ +#ifndef _LINUX_VIRTIO_NET_H +#define _LINUX_VIRTIO_NET_H + +#include +#include + +static inline int virtio_net_hdr_to_skb(struct sk_buff *skb, + const struct virtio_net_hdr *hdr, + bool little_endian) +{ + unsigned short gso_type = 0; + + if (hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) { + switch (hdr->gso_type & ~VIRTIO_NET_HDR_GSO_ECN) { + case VIRTIO_NET_HDR_GSO_TCPV4: + gso_type = SKB_GSO_TCPV4; + break; + case VIRTIO_NET_HDR_GSO_TCPV6: + gso_type = SKB_GSO_TCPV6; + break; + case VIRTIO_NET_HDR_GSO_UDP: + gso_type = SKB_GSO_UDP; + break; + default: + return -EINVAL; + } + + if (hdr->gso_type & VIRTIO_NET_HDR_GSO_ECN) + gso_type |= SKB_GSO_TCP_ECN; + + if (hdr->gso_size == 0) + return -EINVAL; + } + + if (hdr->flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) { + u16 start = __virtio16_to_cpu(little_endian, hdr->csum_start); + u16 off = __virtio16_to_cpu(little_endian, hdr->csum_offset); + + if (!skb_partial_csum_set(skb, start, off)) + return -EINVAL; + } + + if (hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) { + u16 gso_size = __virtio16_to_cpu(little_endian, hdr->gso_size); + + skb_shinfo(skb)->gso_size = gso_size; + skb_shinfo(skb)->gso_type = gso_type; + + /* Header must be checked, and gso_segs computed. */ + skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; + skb_shinfo(skb)->gso_segs = 0; + } + + return 0; +} + +static inline int virtio_net_hdr_from_skb(const struct sk_buff *skb, + struct virtio_net_hdr *hdr, + bool little_endian) +{ + memset(hdr, 0, sizeof(*hdr)); + + if (skb_is_gso(skb)) { + struct skb_shared_info *sinfo = skb_shinfo(skb); + + /* This is a hint as to how much should be linear. */ + hdr->hdr_len = __cpu_to_virtio16(little_endian, + skb_headlen(skb)); + hdr->gso_size = __cpu_to_virtio16(little_endian, + sinfo->gso_size); + if (sinfo->gso_type & SKB_GSO_TCPV4) + hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV4; + else if (sinfo->gso_type & SKB_GSO_TCPV6) + hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV6; + else if (sinfo->gso_type & SKB_GSO_UDP) + hdr->gso_type = VIRTIO_NET_HDR_GSO_UDP; + else + return -EINVAL; + if (sinfo->gso_type & SKB_GSO_TCP_ECN) + hdr->gso_type |= VIRTIO_NET_HDR_GSO_ECN; + } else + hdr->gso_type = VIRTIO_NET_HDR_GSO_NONE; + + if (skb->ip_summed == CHECKSUM_PARTIAL) { + hdr->flags = VIRTIO_NET_HDR_F_NEEDS_CSUM; + if (skb_vlan_tag_present(skb)) + hdr->csum_start = __cpu_to_virtio16(little_endian, + skb_checksum_start_offset(skb) + VLAN_HLEN); + else + hdr->csum_start = __cpu_to_virtio16(little_endian, + skb_checksum_start_offset(skb)); + hdr->csum_offset = __cpu_to_virtio16(little_endian, + skb->csum_offset); + } else if (skb->ip_summed == CHECKSUM_UNNECESSARY) { + hdr->flags = VIRTIO_NET_HDR_F_DATA_VALID; + } /* else everything is zero */ + + return 0; +} + +#endif /* _LINUX_VIRTIO_BYTEORDER */ -- cgit v0.10.2 From fd88d68b3887e4ed1bcfb6a73eeac6f2063da34e Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Wed, 8 Jun 2016 16:09:19 +0300 Subject: macvtap: use common code for virtio_net_hdr and skb GSO conversion Replace open coded conversion between virtio_net_hdr to skb GSO info with virtio_net_hdr_{from,to}_skb Signed-off-by: Mike Rapoport Signed-off-by: David S. Miller diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index bd67209..95a1332 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -627,93 +627,6 @@ static inline struct sk_buff *macvtap_alloc_skb(struct sock *sk, size_t prepad, return skb; } -/* - * macvtap_skb_from_vnet_hdr and macvtap_skb_to_vnet_hdr should - * be shared with the tun/tap driver. - */ -static int macvtap_skb_from_vnet_hdr(struct macvtap_queue *q, - struct sk_buff *skb, - struct virtio_net_hdr *vnet_hdr) -{ - unsigned short gso_type = 0; - if (vnet_hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) { - switch (vnet_hdr->gso_type & ~VIRTIO_NET_HDR_GSO_ECN) { - case VIRTIO_NET_HDR_GSO_TCPV4: - gso_type = SKB_GSO_TCPV4; - break; - case VIRTIO_NET_HDR_GSO_TCPV6: - gso_type = SKB_GSO_TCPV6; - break; - case VIRTIO_NET_HDR_GSO_UDP: - gso_type = SKB_GSO_UDP; - break; - default: - return -EINVAL; - } - - if (vnet_hdr->gso_type & VIRTIO_NET_HDR_GSO_ECN) - gso_type |= SKB_GSO_TCP_ECN; - - if (vnet_hdr->gso_size == 0) - return -EINVAL; - } - - if (vnet_hdr->flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) { - if (!skb_partial_csum_set(skb, macvtap16_to_cpu(q, vnet_hdr->csum_start), - macvtap16_to_cpu(q, vnet_hdr->csum_offset))) - return -EINVAL; - } - - if (vnet_hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) { - skb_shinfo(skb)->gso_size = macvtap16_to_cpu(q, vnet_hdr->gso_size); - skb_shinfo(skb)->gso_type = gso_type; - - /* Header must be checked, and gso_segs computed. */ - skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; - skb_shinfo(skb)->gso_segs = 0; - } - return 0; -} - -static void macvtap_skb_to_vnet_hdr(struct macvtap_queue *q, - const struct sk_buff *skb, - struct virtio_net_hdr *vnet_hdr) -{ - memset(vnet_hdr, 0, sizeof(*vnet_hdr)); - - if (skb_is_gso(skb)) { - struct skb_shared_info *sinfo = skb_shinfo(skb); - - /* This is a hint as to how much should be linear. */ - vnet_hdr->hdr_len = cpu_to_macvtap16(q, skb_headlen(skb)); - vnet_hdr->gso_size = cpu_to_macvtap16(q, sinfo->gso_size); - if (sinfo->gso_type & SKB_GSO_TCPV4) - vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV4; - else if (sinfo->gso_type & SKB_GSO_TCPV6) - vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV6; - else if (sinfo->gso_type & SKB_GSO_UDP) - vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_UDP; - else - BUG(); - if (sinfo->gso_type & SKB_GSO_TCP_ECN) - vnet_hdr->gso_type |= VIRTIO_NET_HDR_GSO_ECN; - } else - vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_NONE; - - if (skb->ip_summed == CHECKSUM_PARTIAL) { - vnet_hdr->flags = VIRTIO_NET_HDR_F_NEEDS_CSUM; - if (skb_vlan_tag_present(skb)) - vnet_hdr->csum_start = cpu_to_macvtap16(q, - skb_checksum_start_offset(skb) + VLAN_HLEN); - else - vnet_hdr->csum_start = cpu_to_macvtap16(q, - skb_checksum_start_offset(skb)); - vnet_hdr->csum_offset = cpu_to_macvtap16(q, skb->csum_offset); - } else if (skb->ip_summed == CHECKSUM_UNNECESSARY) { - vnet_hdr->flags = VIRTIO_NET_HDR_F_DATA_VALID; - } /* else everything is zero */ -} - /* Neighbour code has some assumptions on HH_DATA_MOD alignment */ #define MACVTAP_RESERVE HH_DATA_OFF(ETH_HLEN) @@ -812,7 +725,8 @@ static ssize_t macvtap_get_user(struct macvtap_queue *q, struct msghdr *m, skb->protocol = eth_hdr(skb)->h_proto; if (vnet_hdr_len) { - err = macvtap_skb_from_vnet_hdr(q, skb, &vnet_hdr); + err = virtio_net_hdr_to_skb(skb, &vnet_hdr, + macvtap_is_little_endian(q)); if (err) goto err_kfree; } @@ -880,7 +794,10 @@ static ssize_t macvtap_put_user(struct macvtap_queue *q, if (iov_iter_count(iter) < vnet_hdr_len) return -EINVAL; - macvtap_skb_to_vnet_hdr(q, skb, &vnet_hdr); + ret = virtio_net_hdr_from_skb(skb, &vnet_hdr, + macvtap_is_little_endian(q)); + if (ret) + BUG(); if (copy_to_iter(&vnet_hdr, sizeof(vnet_hdr), iter) != sizeof(vnet_hdr)) -- cgit v0.10.2 From 34166093639b1fc359d364a0d8a30e58d7d95222 Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Wed, 8 Jun 2016 16:09:20 +0300 Subject: tuntap: use common code for virtio_net_hdr and skb GSO conversion Replace open coded conversion between virtio_net_hdr to skb GSO info with virtio_net_hdr_{from,to}_skb Signed-off-by: Mike Rapoport Signed-off-by: David S. Miller diff --git a/drivers/net/tun.c b/drivers/net/tun.c index e16487c..8cc6bf4 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1254,15 +1254,6 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, return -EFAULT; } - if (gso.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) { - if (!skb_partial_csum_set(skb, tun16_to_cpu(tun, gso.csum_start), - tun16_to_cpu(tun, gso.csum_offset))) { - this_cpu_inc(tun->pcpu_stats->rx_frame_errors); - kfree_skb(skb); - return -EINVAL; - } - } - switch (tun->flags & TUN_TYPE_MASK) { case IFF_TUN: if (tun->flags & IFF_NO_PI) { @@ -1289,37 +1280,11 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, break; } - if (gso.gso_type != VIRTIO_NET_HDR_GSO_NONE) { - pr_debug("GSO!\n"); - switch (gso.gso_type & ~VIRTIO_NET_HDR_GSO_ECN) { - case VIRTIO_NET_HDR_GSO_TCPV4: - skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4; - break; - case VIRTIO_NET_HDR_GSO_TCPV6: - skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6; - break; - case VIRTIO_NET_HDR_GSO_UDP: - skb_shinfo(skb)->gso_type = SKB_GSO_UDP; - break; - default: - this_cpu_inc(tun->pcpu_stats->rx_frame_errors); - kfree_skb(skb); - return -EINVAL; - } - - if (gso.gso_type & VIRTIO_NET_HDR_GSO_ECN) - skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN; - - skb_shinfo(skb)->gso_size = tun16_to_cpu(tun, gso.gso_size); - if (skb_shinfo(skb)->gso_size == 0) { - this_cpu_inc(tun->pcpu_stats->rx_frame_errors); - kfree_skb(skb); - return -EINVAL; - } - - /* Header must be checked, and gso_segs computed. */ - skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; - skb_shinfo(skb)->gso_segs = 0; + err = virtio_net_hdr_to_skb(skb, &gso, tun_is_little_endian(tun)); + if (err) { + this_cpu_inc(tun->pcpu_stats->rx_frame_errors); + kfree_skb(skb); + return -EINVAL; } /* copy skb_ubuf_info for callback when skb has no error */ @@ -1399,46 +1364,26 @@ static ssize_t tun_put_user(struct tun_struct *tun, if (vnet_hdr_sz) { struct virtio_net_hdr gso = { 0 }; /* no info leak */ + int ret; + if (iov_iter_count(iter) < vnet_hdr_sz) return -EINVAL; - if (skb_is_gso(skb)) { + ret = virtio_net_hdr_from_skb(skb, &gso, + tun_is_little_endian(tun)); + if (ret) { struct skb_shared_info *sinfo = skb_shinfo(skb); - - /* This is a hint as to how much should be linear. */ - gso.hdr_len = cpu_to_tun16(tun, skb_headlen(skb)); - gso.gso_size = cpu_to_tun16(tun, sinfo->gso_size); - if (sinfo->gso_type & SKB_GSO_TCPV4) - gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV4; - else if (sinfo->gso_type & SKB_GSO_TCPV6) - gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV6; - else if (sinfo->gso_type & SKB_GSO_UDP) - gso.gso_type = VIRTIO_NET_HDR_GSO_UDP; - else { - pr_err("unexpected GSO type: " - "0x%x, gso_size %d, hdr_len %d\n", - sinfo->gso_type, tun16_to_cpu(tun, gso.gso_size), - tun16_to_cpu(tun, gso.hdr_len)); - print_hex_dump(KERN_ERR, "tun: ", - DUMP_PREFIX_NONE, - 16, 1, skb->head, - min((int)tun16_to_cpu(tun, gso.hdr_len), 64), true); - WARN_ON_ONCE(1); - return -EINVAL; - } - if (sinfo->gso_type & SKB_GSO_TCP_ECN) - gso.gso_type |= VIRTIO_NET_HDR_GSO_ECN; - } else - gso.gso_type = VIRTIO_NET_HDR_GSO_NONE; - - if (skb->ip_summed == CHECKSUM_PARTIAL) { - gso.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM; - gso.csum_start = cpu_to_tun16(tun, skb_checksum_start_offset(skb) + - vlan_hlen); - gso.csum_offset = cpu_to_tun16(tun, skb->csum_offset); - } else if (skb->ip_summed == CHECKSUM_UNNECESSARY) { - gso.flags = VIRTIO_NET_HDR_F_DATA_VALID; - } /* else everything is zero */ + pr_err("unexpected GSO type: " + "0x%x, gso_size %d, hdr_len %d\n", + sinfo->gso_type, tun16_to_cpu(tun, gso.gso_size), + tun16_to_cpu(tun, gso.hdr_len)); + print_hex_dump(KERN_ERR, "tun: ", + DUMP_PREFIX_NONE, + 16, 1, skb->head, + min((int)tun16_to_cpu(tun, gso.hdr_len), 64), true); + WARN_ON_ONCE(1); + return -EINVAL; + } if (copy_to_iter(&gso, sizeof(gso), iter) != sizeof(gso)) return -EFAULT; -- cgit v0.10.2 From e858fae2b0b8f41f0bed2cdffde25e7c97da38a7 Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Wed, 8 Jun 2016 16:09:21 +0300 Subject: virtio_net: use common code for virtio_net_hdr and skb GSO conversion Replace open coded conversion between virtio_net_hdr to skb GSO info with virtio_net_hdr_{from,to}_skb Signed-off-by: Mike Rapoport Signed-off-by: David S. Miller diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 192f321..f9470f2 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -479,51 +479,19 @@ static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq, stats->rx_packets++; u64_stats_update_end(&stats->rx_syncp); - if (hdr->hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) { - pr_debug("Needs csum!\n"); - if (!skb_partial_csum_set(skb, - virtio16_to_cpu(vi->vdev, hdr->hdr.csum_start), - virtio16_to_cpu(vi->vdev, hdr->hdr.csum_offset))) - goto frame_err; - } else if (hdr->hdr.flags & VIRTIO_NET_HDR_F_DATA_VALID) { + if (hdr->hdr.flags & VIRTIO_NET_HDR_F_DATA_VALID) skb->ip_summed = CHECKSUM_UNNECESSARY; - } skb->protocol = eth_type_trans(skb, dev); pr_debug("Receiving skb proto 0x%04x len %i type %i\n", ntohs(skb->protocol), skb->len, skb->pkt_type); - if (hdr->hdr.gso_type != VIRTIO_NET_HDR_GSO_NONE) { - pr_debug("GSO!\n"); - switch (hdr->hdr.gso_type & ~VIRTIO_NET_HDR_GSO_ECN) { - case VIRTIO_NET_HDR_GSO_TCPV4: - skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4; - break; - case VIRTIO_NET_HDR_GSO_UDP: - skb_shinfo(skb)->gso_type = SKB_GSO_UDP; - break; - case VIRTIO_NET_HDR_GSO_TCPV6: - skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6; - break; - default: - net_warn_ratelimited("%s: bad gso type %u.\n", - dev->name, hdr->hdr.gso_type); - goto frame_err; - } - - if (hdr->hdr.gso_type & VIRTIO_NET_HDR_GSO_ECN) - skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN; - - skb_shinfo(skb)->gso_size = virtio16_to_cpu(vi->vdev, - hdr->hdr.gso_size); - if (skb_shinfo(skb)->gso_size == 0) { - net_warn_ratelimited("%s: zero gso size.\n", dev->name); - goto frame_err; - } - - /* Header must be checked, and gso_segs computed. */ - skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; - skb_shinfo(skb)->gso_segs = 0; + if (virtio_net_hdr_to_skb(skb, &hdr->hdr, + virtio_is_little_endian(vi->vdev))) { + net_warn_ratelimited("%s: bad gso: type: %u, size: %u\n", + dev->name, hdr->hdr.gso_type, + hdr->hdr.gso_size); + goto frame_err; } napi_gro_receive(&rq->napi, skb); @@ -868,35 +836,9 @@ static int xmit_skb(struct send_queue *sq, struct sk_buff *skb) else hdr = skb_vnet_hdr(skb); - if (skb->ip_summed == CHECKSUM_PARTIAL) { - hdr->hdr.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM; - hdr->hdr.csum_start = cpu_to_virtio16(vi->vdev, - skb_checksum_start_offset(skb)); - hdr->hdr.csum_offset = cpu_to_virtio16(vi->vdev, - skb->csum_offset); - } else { - hdr->hdr.flags = 0; - hdr->hdr.csum_offset = hdr->hdr.csum_start = 0; - } - - if (skb_is_gso(skb)) { - hdr->hdr.hdr_len = cpu_to_virtio16(vi->vdev, skb_headlen(skb)); - hdr->hdr.gso_size = cpu_to_virtio16(vi->vdev, - skb_shinfo(skb)->gso_size); - if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4) - hdr->hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV4; - else if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) - hdr->hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV6; - else if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP) - hdr->hdr.gso_type = VIRTIO_NET_HDR_GSO_UDP; - else - BUG(); - if (skb_shinfo(skb)->gso_type & SKB_GSO_TCP_ECN) - hdr->hdr.gso_type |= VIRTIO_NET_HDR_GSO_ECN; - } else { - hdr->hdr.gso_type = VIRTIO_NET_HDR_GSO_NONE; - hdr->hdr.gso_size = hdr->hdr.hdr_len = 0; - } + if (virtio_net_hdr_from_skb(skb, &hdr->hdr, + virtio_is_little_endian(vi->vdev))) + BUG(); if (vi->mergeable_rx_bufs) hdr->num_buffers = 0; -- cgit v0.10.2 From 1276f24eeef207e5572649d069ba1b531b2e2c3a Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Wed, 8 Jun 2016 16:09:22 +0300 Subject: packet: use common code for virtio_net_hdr and skb GSO conversion Replace open coded conversion between virtio_net_hdr to skb GSO info with virtio_net_hdr_from_skb Signed-off-by: Mike Rapoport Signed-off-by: David S. Miller diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 9bff6ef..d1f3b9e 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1979,40 +1979,8 @@ static int __packet_rcv_vnet(const struct sk_buff *skb, { *vnet_hdr = (const struct virtio_net_hdr) { 0 }; - if (skb_is_gso(skb)) { - struct skb_shared_info *sinfo = skb_shinfo(skb); - - /* This is a hint as to how much should be linear. */ - vnet_hdr->hdr_len = - __cpu_to_virtio16(vio_le(), skb_headlen(skb)); - vnet_hdr->gso_size = - __cpu_to_virtio16(vio_le(), sinfo->gso_size); - - if (sinfo->gso_type & SKB_GSO_TCPV4) - vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV4; - else if (sinfo->gso_type & SKB_GSO_TCPV6) - vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV6; - else if (sinfo->gso_type & SKB_GSO_UDP) - vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_UDP; - else if (sinfo->gso_type & SKB_GSO_FCOE) - return -EINVAL; - else - BUG(); - - if (sinfo->gso_type & SKB_GSO_TCP_ECN) - vnet_hdr->gso_type |= VIRTIO_NET_HDR_GSO_ECN; - } else - vnet_hdr->gso_type = VIRTIO_NET_HDR_GSO_NONE; - - if (skb->ip_summed == CHECKSUM_PARTIAL) { - vnet_hdr->flags = VIRTIO_NET_HDR_F_NEEDS_CSUM; - vnet_hdr->csum_start = __cpu_to_virtio16(vio_le(), - skb_checksum_start_offset(skb)); - vnet_hdr->csum_offset = __cpu_to_virtio16(vio_le(), - skb->csum_offset); - } else if (skb->ip_summed == CHECKSUM_UNNECESSARY) { - vnet_hdr->flags = VIRTIO_NET_HDR_F_DATA_VALID; - } /* else everything is zero */ + if (virtio_net_hdr_from_skb(skb, vnet_hdr, vio_le())) + BUG(); return 0; } -- cgit v0.10.2 From 6f094b9ec680209c5b7314feee983b2f4c910b1b Mon Sep 17 00:00:00 2001 From: Lawrence Brakmo Date: Wed, 8 Jun 2016 21:16:44 -0700 Subject: tcp: add in_flight to tcp_skb_cb Add in_flight (bytes in flight when packet was sent) field to tx component of tcp_skb_cb and make it available to congestion modules' pkts_acked() function through the ack_sample function argument. Signed-off-by: Lawrence Brakmo Acked-by: Yuchung Cheng Signed-off-by: David S. Miller diff --git a/include/net/tcp.h b/include/net/tcp.h index 0bcc70f..a79894b 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -767,6 +767,7 @@ struct tcp_skb_cb { union { struct { /* There is space for up to 20 bytes */ + __u32 in_flight;/* Bytes in flight when packet sent */ } tx; /* only used for outgoing skbs */ union { struct inet_skb_parm h4; @@ -859,6 +860,7 @@ union tcp_cc_info; struct ack_sample { u32 pkts_acked; s32 rtt_us; + u32 in_flight; }; struct tcp_congestion_ops { diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 89dd8d8..94d4aff 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3115,6 +3115,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, long ca_rtt_us = -1L; struct sk_buff *skb; u32 pkts_acked = 0; + u32 last_in_flight = 0; bool rtt_update; int flag = 0; @@ -3154,6 +3155,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, if (!first_ackt.v64) first_ackt = last_ackt; + last_in_flight = TCP_SKB_CB(skb)->tx.in_flight; reord = min(pkts_acked, reord); if (!after(scb->end_seq, tp->high_seq)) flag |= FLAG_ORIG_SACK_ACKED; @@ -3250,7 +3252,8 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, if (icsk->icsk_ca_ops->pkts_acked) { struct ack_sample sample = { .pkts_acked = pkts_acked, - .rtt_us = ca_rtt_us }; + .rtt_us = ca_rtt_us, + .in_flight = last_in_flight }; icsk->icsk_ca_ops->pkts_acked(sk, &sample); } diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 8bd9911..b1bcba0 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -911,9 +911,12 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, int err; BUG_ON(!skb || !tcp_skb_pcount(skb)); + tp = tcp_sk(sk); if (clone_it) { skb_mstamp_get(&skb->skb_mstamp); + TCP_SKB_CB(skb)->tx.in_flight = TCP_SKB_CB(skb)->end_seq + - tp->snd_una; if (unlikely(skb_cloned(skb))) skb = pskb_copy(skb, gfp_mask); @@ -924,7 +927,6 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, } inet = inet_sk(sk); - tp = tcp_sk(sk); tcb = TCP_SKB_CB(skb); memset(&opts, 0, sizeof(opts)); -- cgit v0.10.2 From 699fafafab6d765f12367b3ce0816e64ae19d1e8 Mon Sep 17 00:00:00 2001 From: Lawrence Brakmo Date: Wed, 8 Jun 2016 21:16:45 -0700 Subject: tcp: add NV congestion control TCP-NV (New Vegas) is a major update to TCP-Vegas. An earlier version of NV was presented at 2010's LPC. It is a delayed based congestion avoidance for the data center. This version has been tested within a 10G rack where the HW RTTs are 20-50us and with 1 to 400 flows. A description of TCP-NV, including implementation details as well as experimental results, can be found at: http://www.brakmo.org/networking/tcp-nv/TCPNV.html Signed-off-by: Lawrence Brakmo Signed-off-by: David S. Miller diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 238225b..50d6a9b 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -532,6 +532,22 @@ config TCP_CONG_VEGAS window. TCP Vegas should provide less packet loss, but it is not as aggressive as TCP Reno. +config TCP_CONG_NV + tristate "TCP NV" + default n + ---help--- + TCP NV is a follow up to TCP Vegas. It has been modified to deal with + 10G networks, measurement noise introduced by LRO, GRO and interrupt + coalescence. In addition, it will decrease its cwnd multiplicatively + instead of linearly. + + Note that in general congestion avoidance (cwnd decreased when # packets + queued grows) cannot coexist with congestion control (cwnd decreased only + when there is packet loss) due to fairness issues. One scenario when they + can coexist safely is when the CA flows have RTTs << CC flows RTTs. + + For further details see http://www.brakmo.org/networking/tcp-nv/ + config TCP_CONG_SCALABLE tristate "Scalable TCP" default n diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index bfa1336..24629b6 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -50,6 +50,7 @@ obj-$(CONFIG_TCP_CONG_HSTCP) += tcp_highspeed.o obj-$(CONFIG_TCP_CONG_HYBLA) += tcp_hybla.o obj-$(CONFIG_TCP_CONG_HTCP) += tcp_htcp.o obj-$(CONFIG_TCP_CONG_VEGAS) += tcp_vegas.o +obj-$(CONFIG_TCP_CONG_NV) += tcp_nv.o obj-$(CONFIG_TCP_CONG_VENO) += tcp_veno.o obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o diff --git a/net/ipv4/tcp_nv.c b/net/ipv4/tcp_nv.c new file mode 100644 index 0000000..5de82a8 --- /dev/null +++ b/net/ipv4/tcp_nv.c @@ -0,0 +1,476 @@ +/* + * TCP NV: TCP with Congestion Avoidance + * + * TCP-NV is a successor of TCP-Vegas that has been developed to + * deal with the issues that occur in modern networks. + * Like TCP-Vegas, TCP-NV supports true congestion avoidance, + * the ability to detect congestion before packet losses occur. + * When congestion (queue buildup) starts to occur, TCP-NV + * predicts what the cwnd size should be for the current + * throughput and it reduces the cwnd proportionally to + * the difference between the current cwnd and the predicted cwnd. + * + * NV is only recommeneded for traffic within a data center, and when + * all the flows are NV (at least those within the data center). This + * is due to the inherent unfairness between flows using losses to + * detect congestion (congestion control) and those that use queue + * buildup to detect congestion (congestion avoidance). + * + * Note: High NIC coalescence values may lower the performance of NV + * due to the increased noise in RTT values. In particular, we have + * seen issues with rx-frames values greater than 8. + * + * TODO: + * 1) Add mechanism to deal with reverse congestion. + */ + +#include +#include +#include +#include +#include + +/* TCP NV parameters + * + * nv_pad Max number of queued packets allowed in network + * nv_pad_buffer Do not grow cwnd if this closed to nv_pad + * nv_reset_period How often (in) seconds)to reset min_rtt + * nv_min_cwnd Don't decrease cwnd below this if there are no losses + * nv_cong_dec_mult Decrease cwnd by X% (30%) of congestion when detected + * nv_ssthresh_factor On congestion set ssthresh to this * / 8 + * nv_rtt_factor RTT averaging factor + * nv_loss_dec_factor Decrease cwnd by this (50%) when losses occur + * nv_dec_eval_min_calls Wait this many RTT measurements before dec cwnd + * nv_inc_eval_min_calls Wait this many RTT measurements before inc cwnd + * nv_ssthresh_eval_min_calls Wait this many RTT measurements before stopping + * slow-start due to congestion + * nv_stop_rtt_cnt Only grow cwnd for this many RTTs after non-congestion + * nv_rtt_min_cnt Wait these many RTTs before making congesion decision + * nv_cwnd_growth_rate_neg + * nv_cwnd_growth_rate_pos + * How quickly to double growth rate (not rate) of cwnd when not + * congested. One value (nv_cwnd_growth_rate_neg) for when + * rate < 1 pkt/RTT (after losses). The other (nv_cwnd_growth_rate_pos) + * otherwise. + */ + +static int nv_pad __read_mostly = 10; +static int nv_pad_buffer __read_mostly = 2; +static int nv_reset_period __read_mostly = 5; /* in seconds */ +static int nv_min_cwnd __read_mostly = 2; +static int nv_cong_dec_mult __read_mostly = 30 * 128 / 100; /* = 30% */ +static int nv_ssthresh_factor __read_mostly = 8; /* = 1 */ +static int nv_rtt_factor __read_mostly = 128; /* = 1/2*old + 1/2*new */ +static int nv_loss_dec_factor __read_mostly = 512; /* => 50% */ +static int nv_cwnd_growth_rate_neg __read_mostly = 8; +static int nv_cwnd_growth_rate_pos __read_mostly; /* 0 => fixed like Reno */ +static int nv_dec_eval_min_calls __read_mostly = 60; +static int nv_inc_eval_min_calls __read_mostly = 20; +static int nv_ssthresh_eval_min_calls __read_mostly = 30; +static int nv_stop_rtt_cnt __read_mostly = 10; +static int nv_rtt_min_cnt __read_mostly = 2; + +module_param(nv_pad, int, 0644); +MODULE_PARM_DESC(nv_pad, "max queued packets allowed in network"); +module_param(nv_reset_period, int, 0644); +MODULE_PARM_DESC(nv_reset_period, "nv_min_rtt reset period (secs)"); +module_param(nv_min_cwnd, int, 0644); +MODULE_PARM_DESC(nv_min_cwnd, "NV will not decrease cwnd below this value" + " without losses"); + +/* TCP NV Parameters */ +struct tcpnv { + unsigned long nv_min_rtt_reset_jiffies; /* when to switch to + * nv_min_rtt_new */ + s8 cwnd_growth_factor; /* Current cwnd growth factor, + * < 0 => less than 1 packet/RTT */ + u8 available8; + u16 available16; + u32 loss_cwnd; /* cwnd at last loss */ + u8 nv_allow_cwnd_growth:1, /* whether cwnd can grow */ + nv_reset:1, /* whether to reset values */ + nv_catchup:1; /* whether we are growing because + * of temporary cwnd decrease */ + u8 nv_eval_call_cnt; /* call count since last eval */ + u8 nv_min_cwnd; /* nv won't make a ca decision if cwnd is + * smaller than this. It may grow to handle + * TSO, LRO and interrupt coalescence because + * with these a small cwnd cannot saturate + * the link. Note that this is different from + * the file local nv_min_cwnd */ + u8 nv_rtt_cnt; /* RTTs without making ca decision */; + u32 nv_last_rtt; /* last rtt */ + u32 nv_min_rtt; /* active min rtt. Used to determine slope */ + u32 nv_min_rtt_new; /* min rtt for future use */ + u32 nv_rtt_max_rate; /* max rate seen during current RTT */ + u32 nv_rtt_start_seq; /* current RTT ends when packet arrives + * acking beyond nv_rtt_start_seq */ + u32 nv_last_snd_una; /* Previous value of tp->snd_una. It is + * used to determine bytes acked since last + * call to bictcp_acked */ + u32 nv_no_cong_cnt; /* Consecutive no congestion decisions */ +}; + +#define NV_INIT_RTT U32_MAX +#define NV_MIN_CWND 4 +#define NV_MIN_CWND_GROW 2 +#define NV_TSO_CWND_BOUND 80 + +static inline void tcpnv_reset(struct tcpnv *ca, struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + ca->nv_reset = 0; + ca->loss_cwnd = 0; + ca->nv_no_cong_cnt = 0; + ca->nv_rtt_cnt = 0; + ca->nv_last_rtt = 0; + ca->nv_rtt_max_rate = 0; + ca->nv_rtt_start_seq = tp->snd_una; + ca->nv_eval_call_cnt = 0; + ca->nv_last_snd_una = tp->snd_una; +} + +static void tcpnv_init(struct sock *sk) +{ + struct tcpnv *ca = inet_csk_ca(sk); + + tcpnv_reset(ca, sk); + + ca->nv_allow_cwnd_growth = 1; + ca->nv_min_rtt_reset_jiffies = jiffies + 2 * HZ; + ca->nv_min_rtt = NV_INIT_RTT; + ca->nv_min_rtt_new = NV_INIT_RTT; + ca->nv_min_cwnd = NV_MIN_CWND; + ca->nv_catchup = 0; + ca->cwnd_growth_factor = 0; +} + +static void tcpnv_cong_avoid(struct sock *sk, u32 ack, u32 acked) +{ + struct tcp_sock *tp = tcp_sk(sk); + struct tcpnv *ca = inet_csk_ca(sk); + u32 cnt; + + if (!tcp_is_cwnd_limited(sk)) + return; + + /* Only grow cwnd if NV has not detected congestion */ + if (!ca->nv_allow_cwnd_growth) + return; + + if (tcp_in_slow_start(tp)) { + acked = tcp_slow_start(tp, acked); + if (!acked) + return; + } + + if (ca->cwnd_growth_factor < 0) { + cnt = tp->snd_cwnd << -ca->cwnd_growth_factor; + tcp_cong_avoid_ai(tp, cnt, acked); + } else { + cnt = max(4U, tp->snd_cwnd >> ca->cwnd_growth_factor); + tcp_cong_avoid_ai(tp, cnt, acked); + } +} + +static u32 tcpnv_recalc_ssthresh(struct sock *sk) +{ + const struct tcp_sock *tp = tcp_sk(sk); + struct tcpnv *ca = inet_csk_ca(sk); + + ca->loss_cwnd = tp->snd_cwnd; + return max((tp->snd_cwnd * nv_loss_dec_factor) >> 10, 2U); +} + +static u32 tcpnv_undo_cwnd(struct sock *sk) +{ + struct tcpnv *ca = inet_csk_ca(sk); + + return max(tcp_sk(sk)->snd_cwnd, ca->loss_cwnd); +} + +static void tcpnv_state(struct sock *sk, u8 new_state) +{ + struct tcpnv *ca = inet_csk_ca(sk); + + if (new_state == TCP_CA_Open && ca->nv_reset) { + tcpnv_reset(ca, sk); + } else if (new_state == TCP_CA_Loss || new_state == TCP_CA_CWR || + new_state == TCP_CA_Recovery) { + ca->nv_reset = 1; + ca->nv_allow_cwnd_growth = 0; + if (new_state == TCP_CA_Loss) { + /* Reset cwnd growth factor to Reno value */ + if (ca->cwnd_growth_factor > 0) + ca->cwnd_growth_factor = 0; + /* Decrease growth rate if allowed */ + if (nv_cwnd_growth_rate_neg > 0 && + ca->cwnd_growth_factor > -8) + ca->cwnd_growth_factor--; + } + } +} + +/* Do congestion avoidance calculations for TCP-NV + */ +static void tcpnv_acked(struct sock *sk, const struct ack_sample *sample) +{ + const struct inet_connection_sock *icsk = inet_csk(sk); + struct tcp_sock *tp = tcp_sk(sk); + struct tcpnv *ca = inet_csk_ca(sk); + unsigned long now = jiffies; + s64 rate64 = 0; + u32 rate, max_win, cwnd_by_slope; + u32 avg_rtt; + u32 bytes_acked = 0; + + /* Some calls are for duplicates without timetamps */ + if (sample->rtt_us < 0) + return; + + /* If not in TCP_CA_Open or TCP_CA_Disorder states, skip. */ + if (icsk->icsk_ca_state != TCP_CA_Open && + icsk->icsk_ca_state != TCP_CA_Disorder) + return; + + /* Stop cwnd growth if we were in catch up mode */ + if (ca->nv_catchup && tp->snd_cwnd >= nv_min_cwnd) { + ca->nv_catchup = 0; + ca->nv_allow_cwnd_growth = 0; + } + + bytes_acked = tp->snd_una - ca->nv_last_snd_una; + ca->nv_last_snd_una = tp->snd_una; + + if (sample->in_flight == 0) + return; + + /* Calculate moving average of RTT */ + if (nv_rtt_factor > 0) { + if (ca->nv_last_rtt > 0) { + avg_rtt = (((u64)sample->rtt_us) * nv_rtt_factor + + ((u64)ca->nv_last_rtt) + * (256 - nv_rtt_factor)) >> 8; + } else { + avg_rtt = sample->rtt_us; + ca->nv_min_rtt = avg_rtt << 1; + } + ca->nv_last_rtt = avg_rtt; + } else { + avg_rtt = sample->rtt_us; + } + + /* rate in 100's bits per second */ + rate64 = ((u64)sample->in_flight) * 8000000; + rate = (u32)div64_u64(rate64, (u64)(avg_rtt * 100)); + + /* Remember the maximum rate seen during this RTT + * Note: It may be more than one RTT. This function should be + * called at least nv_dec_eval_min_calls times. + */ + if (ca->nv_rtt_max_rate < rate) + ca->nv_rtt_max_rate = rate; + + /* We have valid information, increment counter */ + if (ca->nv_eval_call_cnt < 255) + ca->nv_eval_call_cnt++; + + /* update min rtt if necessary */ + if (avg_rtt < ca->nv_min_rtt) + ca->nv_min_rtt = avg_rtt; + + /* update future min_rtt if necessary */ + if (avg_rtt < ca->nv_min_rtt_new) + ca->nv_min_rtt_new = avg_rtt; + + /* nv_min_rtt is updated with the minimum (possibley averaged) rtt + * seen in the last sysctl_tcp_nv_reset_period seconds (i.e. a + * warm reset). This new nv_min_rtt will be continued to be updated + * and be used for another sysctl_tcp_nv_reset_period seconds, + * when it will be updated again. + * In practice we introduce some randomness, so the actual period used + * is chosen randomly from the range: + * [sysctl_tcp_nv_reset_period*3/4, sysctl_tcp_nv_reset_period*5/4) + */ + if (time_after_eq(now, ca->nv_min_rtt_reset_jiffies)) { + unsigned char rand; + + ca->nv_min_rtt = ca->nv_min_rtt_new; + ca->nv_min_rtt_new = NV_INIT_RTT; + get_random_bytes(&rand, 1); + ca->nv_min_rtt_reset_jiffies = + now + ((nv_reset_period * (384 + rand) * HZ) >> 9); + /* Every so often we decrease ca->nv_min_cwnd in case previous + * value is no longer accurate. + */ + ca->nv_min_cwnd = max(ca->nv_min_cwnd / 2, NV_MIN_CWND); + } + + /* Once per RTT check if we need to do congestion avoidance */ + if (before(ca->nv_rtt_start_seq, tp->snd_una)) { + ca->nv_rtt_start_seq = tp->snd_nxt; + if (ca->nv_rtt_cnt < 0xff) + /* Increase counter for RTTs without CA decision */ + ca->nv_rtt_cnt++; + + /* If this function is only called once within an RTT + * the cwnd is probably too small (in some cases due to + * tso, lro or interrupt coalescence), so we increase + * ca->nv_min_cwnd. + */ + if (ca->nv_eval_call_cnt == 1 && + bytes_acked >= (ca->nv_min_cwnd - 1) * tp->mss_cache && + ca->nv_min_cwnd < (NV_TSO_CWND_BOUND + 1)) { + ca->nv_min_cwnd = min(ca->nv_min_cwnd + + NV_MIN_CWND_GROW, + NV_TSO_CWND_BOUND + 1); + ca->nv_rtt_start_seq = tp->snd_nxt + + ca->nv_min_cwnd * tp->mss_cache; + ca->nv_eval_call_cnt = 0; + ca->nv_allow_cwnd_growth = 1; + return; + } + + /* Find the ideal cwnd for current rate from slope + * slope = 80000.0 * mss / nv_min_rtt + * cwnd_by_slope = nv_rtt_max_rate / slope + */ + cwnd_by_slope = (u32) + div64_u64(((u64)ca->nv_rtt_max_rate) * ca->nv_min_rtt, + (u64)(80000 * tp->mss_cache)); + max_win = cwnd_by_slope + nv_pad; + + /* If cwnd > max_win, decrease cwnd + * if cwnd < max_win, grow cwnd + * else leave the same + */ + if (tp->snd_cwnd > max_win) { + /* there is congestion, check that it is ok + * to make a CA decision + * 1. We should have at least nv_dec_eval_min_calls + * data points before making a CA decision + * 2. We only make a congesion decision after + * nv_rtt_min_cnt RTTs + */ + if (ca->nv_rtt_cnt < nv_rtt_min_cnt) { + return; + } else if (tp->snd_ssthresh == TCP_INFINITE_SSTHRESH) { + if (ca->nv_eval_call_cnt < + nv_ssthresh_eval_min_calls) + return; + /* otherwise we will decrease cwnd */ + } else if (ca->nv_eval_call_cnt < + nv_dec_eval_min_calls) { + if (ca->nv_allow_cwnd_growth && + ca->nv_rtt_cnt > nv_stop_rtt_cnt) + ca->nv_allow_cwnd_growth = 0; + return; + } + + /* We have enough data to determine we are congested */ + ca->nv_allow_cwnd_growth = 0; + tp->snd_ssthresh = + (nv_ssthresh_factor * max_win) >> 3; + if (tp->snd_cwnd - max_win > 2) { + /* gap > 2, we do exponential cwnd decrease */ + int dec; + + dec = max(2U, ((tp->snd_cwnd - max_win) * + nv_cong_dec_mult) >> 7); + tp->snd_cwnd -= dec; + } else if (nv_cong_dec_mult > 0) { + tp->snd_cwnd = max_win; + } + if (ca->cwnd_growth_factor > 0) + ca->cwnd_growth_factor = 0; + ca->nv_no_cong_cnt = 0; + } else if (tp->snd_cwnd <= max_win - nv_pad_buffer) { + /* There is no congestion, grow cwnd if allowed*/ + if (ca->nv_eval_call_cnt < nv_inc_eval_min_calls) + return; + + ca->nv_allow_cwnd_growth = 1; + ca->nv_no_cong_cnt++; + if (ca->cwnd_growth_factor < 0 && + nv_cwnd_growth_rate_neg > 0 && + ca->nv_no_cong_cnt > nv_cwnd_growth_rate_neg) { + ca->cwnd_growth_factor++; + ca->nv_no_cong_cnt = 0; + } else if (ca->cwnd_growth_factor >= 0 && + nv_cwnd_growth_rate_pos > 0 && + ca->nv_no_cong_cnt > + nv_cwnd_growth_rate_pos) { + ca->cwnd_growth_factor++; + ca->nv_no_cong_cnt = 0; + } + } else { + /* cwnd is in-between, so do nothing */ + return; + } + + /* update state */ + ca->nv_eval_call_cnt = 0; + ca->nv_rtt_cnt = 0; + ca->nv_rtt_max_rate = 0; + + /* Don't want to make cwnd < nv_min_cwnd + * (it wasn't before, if it is now is because nv + * decreased it). + */ + if (tp->snd_cwnd < nv_min_cwnd) + tp->snd_cwnd = nv_min_cwnd; + } +} + +/* Extract info for Tcp socket info provided via netlink */ +size_t tcpnv_get_info(struct sock *sk, u32 ext, int *attr, + union tcp_cc_info *info) +{ + const struct tcpnv *ca = inet_csk_ca(sk); + + if (ext & (1 << (INET_DIAG_VEGASINFO - 1))) { + info->vegas.tcpv_enabled = 1; + info->vegas.tcpv_rttcnt = ca->nv_rtt_cnt; + info->vegas.tcpv_rtt = ca->nv_last_rtt; + info->vegas.tcpv_minrtt = ca->nv_min_rtt; + + *attr = INET_DIAG_VEGASINFO; + return sizeof(struct tcpvegas_info); + } + return 0; +} +EXPORT_SYMBOL_GPL(tcpnv_get_info); + +static struct tcp_congestion_ops tcpnv __read_mostly = { + .init = tcpnv_init, + .ssthresh = tcpnv_recalc_ssthresh, + .cong_avoid = tcpnv_cong_avoid, + .set_state = tcpnv_state, + .undo_cwnd = tcpnv_undo_cwnd, + .pkts_acked = tcpnv_acked, + .get_info = tcpnv_get_info, + + .owner = THIS_MODULE, + .name = "nv", +}; + +static int __init tcpnv_register(void) +{ + BUILD_BUG_ON(sizeof(struct tcpnv) > ICSK_CA_PRIV_SIZE); + + return tcp_register_congestion_control(&tcpnv); +} + +static void __exit tcpnv_unregister(void) +{ + tcp_unregister_congestion_control(&tcpnv); +} + +module_init(tcpnv_register); +module_exit(tcpnv_unregister); + +MODULE_AUTHOR("Lawrence Brakmo"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("TCP NV"); +MODULE_VERSION("1.0"); -- cgit v0.10.2 From 002245cc6407c8ff65b8024554080eb6de1a8e2c Mon Sep 17 00:00:00 2001 From: Zi Shen Lim Date: Wed, 8 Jun 2016 21:18:47 -0700 Subject: bpf: fix missing header inclusion Commit 0fc174dea545 ("ebpf: make internal bpf API independent of CONFIG_BPF_SYSCALL ifdefs") introduced usage of ERR_PTR() in bpf_prog_get(), however did not include linux/err.h. Without this patch, when compiling arm64 BPF without CONFIG_BPF_SYSCALL: ... In file included from arch/arm64/net/bpf_jit_comp.c:21:0: include/linux/bpf.h: In function 'bpf_prog_get': include/linux/bpf.h:235:9: error: implicit declaration of function 'ERR_PTR' [-Werror=implicit-function-declaration] return ERR_PTR(-EOPNOTSUPP); ^ include/linux/bpf.h:235:9: warning: return makes pointer from integer without a cast [-Wint-conversion] In file included from include/linux/rwsem.h:17:0, from include/linux/mm_types.h:10, from include/linux/sched.h:27, from arch/arm64/include/asm/compat.h:25, from arch/arm64/include/asm/stat.h:23, from include/linux/stat.h:5, from include/linux/compat.h:12, from include/linux/filter.h:10, from arch/arm64/net/bpf_jit_comp.c:22: include/linux/err.h: At top level: include/linux/err.h:23:35: error: conflicting types for 'ERR_PTR' static inline void * __must_check ERR_PTR(long error) ^ In file included from arch/arm64/net/bpf_jit_comp.c:21:0: include/linux/bpf.h:235:9: note: previous implicit declaration of 'ERR_PTR' was here return ERR_PTR(-EOPNOTSUPP); ^ ... Fixes: 0fc174dea545 ("ebpf: make internal bpf API independent of CONFIG_BPF_SYSCALL ifdefs") Suggested-by: Daniel Borkmann Signed-off-by: Zi Shen Lim Acked-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 8ee27b8..1bcae82 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -11,6 +11,7 @@ #include #include #include +#include struct bpf_map; -- cgit v0.10.2 From ddb55992b04d9749e7c00af7f855e4e13566a521 Mon Sep 17 00:00:00 2001 From: Zi Shen Lim Date: Wed, 8 Jun 2016 21:18:48 -0700 Subject: arm64: bpf: implement bpf_tail_call() helper Add support for JMP_CALL_X (tail call) introduced by commit 04fd61ab36ec ("bpf: allow bpf programs to tail-call other bpf programs"). bpf_tail_call() arguments: ctx - context pointer passed to next program array - pointer to map which type is BPF_MAP_TYPE_PROG_ARRAY index - index inside array that selects specific program to run In this implementation arm64 JIT jumps into callee program after prologue, so callee program reuses the same stack. For tail_call_cnt, we use the callee-saved R26 (which was already saved/restored but previously unused by JIT). With this patch a tail call generates the following code on arm64: if (index >= array->map.max_entries) goto out; 34: mov x10, #0x10 // #16 38: ldr w10, [x1,x10] 3c: cmp w2, w10 40: b.ge 0x0000000000000074 if (tail_call_cnt > MAX_TAIL_CALL_CNT) goto out; tail_call_cnt++; 44: mov x10, #0x20 // #32 48: cmp x26, x10 4c: b.gt 0x0000000000000074 50: add x26, x26, #0x1 prog = array->ptrs[index]; if (prog == NULL) goto out; 54: mov x10, #0x68 // #104 58: ldr x10, [x1,x10] 5c: ldr x11, [x10,x2] 60: cbz x11, 0x0000000000000074 goto *(prog->bpf_func + prologue_size); 64: mov x10, #0x20 // #32 68: ldr x10, [x11,x10] 6c: add x10, x10, #0x20 70: br x10 74: Signed-off-by: Zi Shen Lim Signed-off-by: David S. Miller diff --git a/arch/arm64/net/bpf_jit.h b/arch/arm64/net/bpf_jit.h index aee5637..7c16e54 100644 --- a/arch/arm64/net/bpf_jit.h +++ b/arch/arm64/net/bpf_jit.h @@ -1,7 +1,7 @@ /* * BPF JIT compiler for ARM64 * - * Copyright (C) 2014-2015 Zi Shen Lim + * Copyright (C) 2014-2016 Zi Shen Lim * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -55,6 +55,7 @@ #define A64_BL(imm26) A64_BRANCH((imm26) << 2, LINK) /* Unconditional branch (register) */ +#define A64_BR(Rn) aarch64_insn_gen_branch_reg(Rn, AARCH64_INSN_BRANCH_NOLINK) #define A64_BLR(Rn) aarch64_insn_gen_branch_reg(Rn, AARCH64_INSN_BRANCH_LINK) #define A64_RET(Rn) aarch64_insn_gen_branch_reg(Rn, AARCH64_INSN_BRANCH_RETURN) diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c index 49ba37e..51abc97 100644 --- a/arch/arm64/net/bpf_jit_comp.c +++ b/arch/arm64/net/bpf_jit_comp.c @@ -18,6 +18,7 @@ #define pr_fmt(fmt) "bpf_jit: " fmt +#include #include #include #include @@ -33,6 +34,7 @@ int bpf_jit_enable __read_mostly; #define TMP_REG_1 (MAX_BPF_JIT_REG + 0) #define TMP_REG_2 (MAX_BPF_JIT_REG + 1) +#define TCALL_CNT (MAX_BPF_JIT_REG + 2) /* Map BPF registers to A64 registers */ static const int bpf2a64[] = { @@ -54,6 +56,8 @@ static const int bpf2a64[] = { /* temporary registers for internal BPF JIT */ [TMP_REG_1] = A64_R(10), [TMP_REG_2] = A64_R(11), + /* tail_call_cnt */ + [TCALL_CNT] = A64_R(26), /* temporary register for blinding constants */ [BPF_REG_AX] = A64_R(9), }; @@ -146,13 +150,18 @@ static inline int epilogue_offset(const struct jit_ctx *ctx) #define STACK_SIZE STACK_ALIGN(_STACK_SIZE) -static void build_prologue(struct jit_ctx *ctx) +#define PROLOGUE_OFFSET 8 + +static int build_prologue(struct jit_ctx *ctx) { const u8 r6 = bpf2a64[BPF_REG_6]; const u8 r7 = bpf2a64[BPF_REG_7]; const u8 r8 = bpf2a64[BPF_REG_8]; const u8 r9 = bpf2a64[BPF_REG_9]; const u8 fp = bpf2a64[BPF_REG_FP]; + const u8 tcc = bpf2a64[TCALL_CNT]; + const int idx0 = ctx->idx; + int cur_offset; /* * BPF prog stack layout @@ -162,8 +171,6 @@ static void build_prologue(struct jit_ctx *ctx) * |FP/LR| * current A64_FP => -16:+-----+ * | ... | callee saved registers - * +-----+ - * | | x25/x26 * BPF fp register => -64:+-----+ <= (BPF_FP) * | | * | ... | BPF prog stack @@ -183,18 +190,90 @@ static void build_prologue(struct jit_ctx *ctx) emit(A64_PUSH(A64_FP, A64_LR, A64_SP), ctx); emit(A64_MOV(1, A64_FP, A64_SP), ctx); - /* Save callee-saved register */ + /* Save callee-saved registers */ emit(A64_PUSH(r6, r7, A64_SP), ctx); emit(A64_PUSH(r8, r9, A64_SP), ctx); + emit(A64_PUSH(fp, tcc, A64_SP), ctx); - /* Save fp (x25) and x26. SP requires 16 bytes alignment */ - emit(A64_PUSH(fp, A64_R(26), A64_SP), ctx); - - /* Set up BPF prog stack base register (x25) */ + /* Set up BPF prog stack base register */ emit(A64_MOV(1, fp, A64_SP), ctx); + /* Initialize tail_call_cnt */ + emit(A64_MOVZ(1, tcc, 0, 0), ctx); + /* Set up function call stack */ emit(A64_SUB_I(1, A64_SP, A64_SP, STACK_SIZE), ctx); + + cur_offset = ctx->idx - idx0; + if (cur_offset != PROLOGUE_OFFSET) { + pr_err_once("PROLOGUE_OFFSET = %d, expected %d!\n", + cur_offset, PROLOGUE_OFFSET); + return -1; + } + return 0; +} + +static int out_offset = -1; /* initialized on the first pass of build_body() */ +static int emit_bpf_tail_call(struct jit_ctx *ctx) +{ + /* bpf_tail_call(void *prog_ctx, struct bpf_array *array, u64 index) */ + const u8 r2 = bpf2a64[BPF_REG_2]; + const u8 r3 = bpf2a64[BPF_REG_3]; + + const u8 tmp = bpf2a64[TMP_REG_1]; + const u8 prg = bpf2a64[TMP_REG_2]; + const u8 tcc = bpf2a64[TCALL_CNT]; + const int idx0 = ctx->idx; +#define cur_offset (ctx->idx - idx0) +#define jmp_offset (out_offset - (cur_offset)) + size_t off; + + /* if (index >= array->map.max_entries) + * goto out; + */ + off = offsetof(struct bpf_array, map.max_entries); + emit_a64_mov_i64(tmp, off, ctx); + emit(A64_LDR32(tmp, r2, tmp), ctx); + emit(A64_CMP(0, r3, tmp), ctx); + emit(A64_B_(A64_COND_GE, jmp_offset), ctx); + + /* if (tail_call_cnt > MAX_TAIL_CALL_CNT) + * goto out; + * tail_call_cnt++; + */ + emit_a64_mov_i64(tmp, MAX_TAIL_CALL_CNT, ctx); + emit(A64_CMP(1, tcc, tmp), ctx); + emit(A64_B_(A64_COND_GT, jmp_offset), ctx); + emit(A64_ADD_I(1, tcc, tcc, 1), ctx); + + /* prog = array->ptrs[index]; + * if (prog == NULL) + * goto out; + */ + off = offsetof(struct bpf_array, ptrs); + emit_a64_mov_i64(tmp, off, ctx); + emit(A64_LDR64(tmp, r2, tmp), ctx); + emit(A64_LDR64(prg, tmp, r3), ctx); + emit(A64_CBZ(1, prg, jmp_offset), ctx); + + /* goto *(prog->bpf_func + prologue_size); */ + off = offsetof(struct bpf_prog, bpf_func); + emit_a64_mov_i64(tmp, off, ctx); + emit(A64_LDR64(tmp, prg, tmp), ctx); + emit(A64_ADD_I(1, tmp, tmp, sizeof(u32) * PROLOGUE_OFFSET), ctx); + emit(A64_BR(tmp), ctx); + + /* out: */ + if (out_offset == -1) + out_offset = cur_offset; + if (cur_offset != out_offset) { + pr_err_once("tail_call out_offset = %d, expected %d!\n", + cur_offset, out_offset); + return -1; + } + return 0; +#undef cur_offset +#undef jmp_offset } static void build_epilogue(struct jit_ctx *ctx) @@ -506,6 +585,11 @@ emit_cond_jmp: emit(A64_POP(A64_FP, A64_LR, A64_SP), ctx); break; } + /* tail call */ + case BPF_JMP | BPF_CALL | BPF_X: + if (emit_bpf_tail_call(ctx)) + return -EFAULT; + break; /* function return */ case BPF_JMP | BPF_EXIT: /* Optimization: when last instruction is EXIT, @@ -780,7 +864,10 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog) goto out_off; } - build_prologue(&ctx); + if (build_prologue(&ctx)) { + prog = orig_prog; + goto out_off; + } ctx.epilogue_offset = ctx.idx; build_epilogue(&ctx); -- cgit v0.10.2 From 997ce888324685a90fb5d0fa26293eb8826c767c Mon Sep 17 00:00:00 2001 From: Zi Shen Lim Date: Wed, 8 Jun 2016 21:18:49 -0700 Subject: arm64: bpf: optimize JMP_CALL Remove superfluous stack frame, saving us 3 instructions for every JMP_CALL. Signed-off-by: Zi Shen Lim Signed-off-by: David S. Miller diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c index 51abc97..7ae304e 100644 --- a/arch/arm64/net/bpf_jit_comp.c +++ b/arch/arm64/net/bpf_jit_comp.c @@ -578,11 +578,8 @@ emit_cond_jmp: const u64 func = (u64)__bpf_call_base + imm; emit_a64_mov_i64(tmp, func, ctx); - emit(A64_PUSH(A64_FP, A64_LR, A64_SP), ctx); - emit(A64_MOV(1, A64_FP, A64_SP), ctx); emit(A64_BLR(tmp), ctx); emit(A64_MOV(1, r0, A64_R(0)), ctx); - emit(A64_POP(A64_FP, A64_LR, A64_SP), ctx); break; } /* tail call */ -- cgit v0.10.2 From 643c332d519bdfbf80d21f40d1c0aa0ccf3ec1cb Mon Sep 17 00:00:00 2001 From: Zi Shen Lim Date: Wed, 8 Jun 2016 21:18:50 -0700 Subject: arm64: bpf: optimize LD_ABS, LD_IND Remove superfluous stack frame, saving us 3 instructions for every LD_ABS or LD_IND. Signed-off-by: Zi Shen Lim Signed-off-by: David S. Miller diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c index 7ae304e..b2fc97a 100644 --- a/arch/arm64/net/bpf_jit_comp.c +++ b/arch/arm64/net/bpf_jit_comp.c @@ -731,11 +731,8 @@ emit_cond_jmp: emit_a64_mov_i64(r3, size, ctx); emit(A64_SUB_I(1, r4, fp, STACK_SIZE), ctx); emit_a64_mov_i64(r5, (unsigned long)bpf_load_pointer, ctx); - emit(A64_PUSH(A64_FP, A64_LR, A64_SP), ctx); - emit(A64_MOV(1, A64_FP, A64_SP), ctx); emit(A64_BLR(r5), ctx); emit(A64_MOV(1, r0, A64_R(0)), ctx); - emit(A64_POP(A64_FP, A64_LR, A64_SP), ctx); jmp_offset = epilogue_offset(ctx); check_imm19(jmp_offset); -- cgit v0.10.2 From d46e416c11c88ef1deb5c7f19271806a5be597fe Mon Sep 17 00:00:00 2001 From: Xin Long Date: Thu, 9 Jun 2016 22:48:18 +0800 Subject: sctp: sctp should change socket state when shutdown is received Now sctp doesn't change socket state upon shutdown reception. It changes just the assoc state, even though it's a TCP-style socket. For some cases, if we really need to check sk->sk_state, it's necessary to fix this issue, at least when we use ss or netstat to dump, we can get a more exact information. As an improvement, we will change sk->sk_state when we change asoc->state to SHUTDOWN_RECEIVED, and also do it in sctp_shutdown to keep consistent with sctp_close. Signed-off-by: Xin Long Acked-by: Marcelo R. Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index aa37122..12d4519 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -806,8 +806,10 @@ static void sctp_cmd_new_state(sctp_cmd_seq_t *cmds, /* Set the RCV_SHUTDOWN flag when a SHUTDOWN is received. */ if (sctp_state(asoc, SHUTDOWN_RECEIVED) && - sctp_sstate(sk, ESTABLISHED)) + sctp_sstate(sk, ESTABLISHED)) { + sk->sk_state = SCTP_SS_CLOSING; sk->sk_shutdown |= RCV_SHUTDOWN; + } } if (sctp_state(asoc, COOKIE_WAIT)) { diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 712fb23..6cae4c6 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4195,6 +4195,7 @@ static void sctp_shutdown(struct sock *sk, int how) return; if (how & SEND_SHUTDOWN) { + sk->sk_state = SCTP_SS_CLOSING; ep = sctp_sk(sk)->ep; if (!list_empty(&ep->asocs)) { asoc = list_entry(ep->asocs.next, @@ -7566,10 +7567,13 @@ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, /* If the association on the newsk is already closed before accept() * is called, set RCV_SHUTDOWN flag. */ - if (sctp_state(assoc, CLOSED) && sctp_style(newsk, TCP)) + if (sctp_state(assoc, CLOSED) && sctp_style(newsk, TCP)) { + newsk->sk_state = SCTP_SS_CLOSING; newsk->sk_shutdown |= RCV_SHUTDOWN; + } else { + newsk->sk_state = SCTP_SS_ESTABLISHED; + } - newsk->sk_state = SCTP_SS_ESTABLISHED; release_sock(newsk); } -- cgit v0.10.2 From f20e6657a8758fe8d074889a6f1883674f01c7f2 Mon Sep 17 00:00:00 2001 From: Pramod Kumar Date: Fri, 10 Jun 2016 11:03:45 +0530 Subject: mdio: mux: Enhanced MDIO mux framework for integrated multiplexers An integrated multiplexer uses same address space for "muxed bus selection" and "generation of mdio transaction" hence its good to register parent bus from mux driver. Hence added a mechanism where mux driver could register a parent bus and pass it down to framework via mdio_mux_init api. Signed-off-by: Pramod Kumar Reviewed-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/phy/mdio-mux-gpio.c b/drivers/net/phy/mdio-mux-gpio.c index 7ddb1ab..9199499 100644 --- a/drivers/net/phy/mdio-mux-gpio.c +++ b/drivers/net/phy/mdio-mux-gpio.c @@ -55,7 +55,7 @@ static int mdio_mux_gpio_probe(struct platform_device *pdev) return PTR_ERR(s->gpios); r = mdio_mux_init(&pdev->dev, - mdio_mux_gpio_switch_fn, &s->mux_handle, s); + mdio_mux_gpio_switch_fn, &s->mux_handle, s, NULL); if (r != 0) { gpiod_put_array(s->gpios); diff --git a/drivers/net/phy/mdio-mux-mmioreg.c b/drivers/net/phy/mdio-mux-mmioreg.c index 7fde454..d0bed52 100644 --- a/drivers/net/phy/mdio-mux-mmioreg.c +++ b/drivers/net/phy/mdio-mux-mmioreg.c @@ -126,7 +126,7 @@ static int mdio_mux_mmioreg_probe(struct platform_device *pdev) } ret = mdio_mux_init(&pdev->dev, mdio_mux_mmioreg_switch_fn, - &s->mux_handle, s); + &s->mux_handle, s, NULL); if (ret) { dev_err(&pdev->dev, "failed to register mdio-mux bus %s\n", np->full_name); diff --git a/drivers/net/phy/mdio-mux.c b/drivers/net/phy/mdio-mux.c index 5c81d6f..dbd4ecc 100644 --- a/drivers/net/phy/mdio-mux.c +++ b/drivers/net/phy/mdio-mux.c @@ -89,7 +89,8 @@ static int parent_count; int mdio_mux_init(struct device *dev, int (*switch_fn)(int cur, int desired, void *data), void **mux_handle, - void *data) + void *data, + struct mii_bus *mux_bus) { struct device_node *parent_bus_node; struct device_node *child_bus_node; @@ -101,10 +102,21 @@ int mdio_mux_init(struct device *dev, if (!dev->of_node) return -ENODEV; - parent_bus_node = of_parse_phandle(dev->of_node, "mdio-parent-bus", 0); + if (!mux_bus) { + parent_bus_node = of_parse_phandle(dev->of_node, + "mdio-parent-bus", 0); - if (!parent_bus_node) - return -ENODEV; + if (!parent_bus_node) + return -ENODEV; + + parent_bus = of_mdio_find_bus(parent_bus_node); + if (!parent_bus) { + ret_val = -EPROBE_DEFER; + goto err_parent_bus; + } + } else { + parent_bus = mux_bus; + } pb = devm_kzalloc(dev, sizeof(*pb), GFP_KERNEL); if (pb == NULL) { @@ -112,11 +124,6 @@ int mdio_mux_init(struct device *dev, goto err_parent_bus; } - parent_bus = of_mdio_find_bus(parent_bus_node); - if (parent_bus == NULL) { - ret_val = -EPROBE_DEFER; - goto err_parent_bus; - } pb->switch_data = data; pb->switch_fn = switch_fn; @@ -177,7 +184,8 @@ int mdio_mux_init(struct device *dev, put_device(&pb->mii_bus->dev); err_parent_bus: - of_node_put(parent_bus_node); + if (!mux_bus) + of_node_put(parent_bus_node); return ret_val; } EXPORT_SYMBOL_GPL(mdio_mux_init); diff --git a/include/linux/mdio-mux.h b/include/linux/mdio-mux.h index a243dbb..61f5b21 100644 --- a/include/linux/mdio-mux.h +++ b/include/linux/mdio-mux.h @@ -10,11 +10,13 @@ #ifndef __LINUX_MDIO_MUX_H #define __LINUX_MDIO_MUX_H #include +#include int mdio_mux_init(struct device *dev, int (*switch_fn) (int cur, int desired, void *data), void **mux_handle, - void *data); + void *data, + struct mii_bus *mux_bus); void mdio_mux_uninit(void *mux_handle); -- cgit v0.10.2 From 6deb2087b8683cb64362b7ff0eeba8df0bf5a07d Mon Sep 17 00:00:00 2001 From: Pramod Kumar Date: Fri, 10 Jun 2016 11:03:46 +0530 Subject: binding: Make "mdio-parent-bus" property from mandatory to optional Change "mdio-parent-bus" from mandatory section to optional as it won't be required by integrated MDIO multiplexer which has bus selection and mdio transaction generation logic, integrated inside. Signed-off-by: Pramod Kumar Reviewed-by: Andrew Lunn Acked-by: Rob Herring Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/mdio-mux.txt b/Documentation/devicetree/bindings/net/mdio-mux.txt index 491f5bd..f58571f 100644 --- a/Documentation/devicetree/bindings/net/mdio-mux.txt +++ b/Documentation/devicetree/bindings/net/mdio-mux.txt @@ -5,11 +5,12 @@ numbered uniquely in a device dependent manner. The nodes for an MDIO bus multiplexer/switch will have one child node for each child bus. Required properties: -- mdio-parent-bus : phandle to the parent MDIO bus. - #address-cells = <1>; - #size-cells = <0>; Optional properties: +- mdio-parent-bus : phandle to the parent MDIO bus. + - Other properties specific to the multiplexer/switch hardware. Required properties for child nodes: -- cgit v0.10.2 From ce8d5dbfd64f6ca84e2fe55a0066ec8b22601567 Mon Sep 17 00:00:00 2001 From: Pramod Kumar Date: Fri, 10 Jun 2016 11:03:47 +0530 Subject: binding: mdio-mux: Add DT binding doc for Broadcom MDIO bus multiplexer Add DT binding doc for Broadcom MDIO bus multiplexer driver. Reviewed-by: Andrew Lunn Signed-off-by: Pramod Kumar Reviewed-by: Florian Fainelli Acked-by: Rob Herring Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/brcm,mdio-mux-iproc.txt b/Documentation/devicetree/bindings/net/brcm,mdio-mux-iproc.txt new file mode 100644 index 0000000..dfe287a --- /dev/null +++ b/Documentation/devicetree/bindings/net/brcm,mdio-mux-iproc.txt @@ -0,0 +1,59 @@ +Properties for an MDIO bus multiplexer found in Broadcom iProc based SoCs. + +This MDIO bus multiplexer defines buses that could be internal as well as +external to SoCs and could accept MDIO transaction compatible to C-22 or +C-45 Clause. When child bus is selected, one needs to select these two +properties as well to generate desired MDIO transaction on appropriate bus. + +Required properties in addition to the generic multiplexer properties: + +MDIO multiplexer node: +- compatible: brcm,mdio-mux-iproc. + +Every non-ethernet PHY requires a compatible so that it could be probed based +on this compatible string. + +Additional information regarding generic multiplexer properties can be found +at- Documentation/devicetree/bindings/net/mdio-mux.txt + + +for example: + mdio_mux_iproc: mdio-mux@6602023c { + compatible = "brcm,mdio-mux-iproc"; + reg = <0x6602023c 0x14>; + #address-cells = <1>; + #size-cells = <0>; + + mdio@0 { + reg = <0x0>; + #address-cells = <1>; + #size-cells = <0>; + + pci_phy0: pci-phy@0 { + compatible = "brcm,ns2-pcie-phy"; + reg = <0x0>; + #phy-cells = <0>; + }; + }; + + mdio@7 { + reg = <0x7>; + #address-cells = <1>; + #size-cells = <0>; + + pci_phy1: pci-phy@0 { + compatible = "brcm,ns2-pcie-phy"; + reg = <0x0>; + #phy-cells = <0>; + }; + }; + mdio@10 { + reg = <0x10>; + #address-cells = <1>; + #size-cells = <0>; + + gphy0: eth-phy@10 { + reg = <0x10>; + }; + }; + }; -- cgit v0.10.2 From 5f1a067bfa0ad29f944f772fbfd9ec7806260b54 Mon Sep 17 00:00:00 2001 From: Pramod Kumar Date: Fri, 10 Jun 2016 11:03:48 +0530 Subject: dt: mdio-mux: Add mdio multiplexer driver node Add integrated MDIO multiplexer driver node which contains two mux PCIe bus and one ethernet bus along with phys lying on these bus. Signed-off-by: Pramod Kumar Acked-by: Rob Herring Signed-off-by: David S. Miller diff --git a/arch/arm64/boot/dts/broadcom/ns2-svk.dts b/arch/arm64/boot/dts/broadcom/ns2-svk.dts index 54ca40c..ea5603f 100644 --- a/arch/arm64/boot/dts/broadcom/ns2-svk.dts +++ b/arch/arm64/boot/dts/broadcom/ns2-svk.dts @@ -52,6 +52,14 @@ }; }; +&pci_phy0 { + status = "ok"; +}; + +&pci_phy1 { + status = "ok"; +}; + &pcie0 { status = "ok"; }; @@ -132,3 +140,11 @@ #size-cells = <1>; }; }; + +&mdio_mux_iproc { + mdio@10 { + gphy0: eth-phy@10 { + reg = <0x10>; + }; + }; +}; diff --git a/arch/arm64/boot/dts/broadcom/ns2.dtsi b/arch/arm64/boot/dts/broadcom/ns2.dtsi index ec68ec1..46b78fa 100644 --- a/arch/arm64/boot/dts/broadcom/ns2.dtsi +++ b/arch/arm64/boot/dts/broadcom/ns2.dtsi @@ -263,6 +263,45 @@ IRQ_TYPE_LEVEL_HIGH)>; }; + mdio_mux_iproc: mdio-mux@6602023c { + compatible = "brcm,mdio-mux-iproc"; + reg = <0x6602023c 0x14>; + #address-cells = <1>; + #size-cells = <0>; + + mdio@0 { + reg = <0x0>; + #address-cells = <1>; + #size-cells = <0>; + + pci_phy0: pci-phy@0 { + compatible = "brcm,ns2-pcie-phy"; + reg = <0x0>; + #phy-cells = <0>; + status = "disabled"; + }; + }; + + mdio@7 { + reg = <0x7>; + #address-cells = <1>; + #size-cells = <0>; + + pci_phy1: pci-phy@0 { + compatible = "brcm,ns2-pcie-phy"; + reg = <0x0>; + #phy-cells = <0>; + status = "disabled"; + }; + }; + + mdio@10 { + reg = <0x10>; + #address-cells = <1>; + #size-cells = <0>; + }; + }; + timer0: timer@66030000 { compatible = "arm,sp804", "arm,primecell"; reg = <0x66030000 0x1000>; -- cgit v0.10.2 From 98bc865a1ec8074defd168b0feb9c466eeaeff33 Mon Sep 17 00:00:00 2001 From: Pramod Kumar Date: Fri, 10 Jun 2016 11:03:49 +0530 Subject: net: mdio-mux: Add MDIO mux driver for iProc SoCs iProc based SoCs supports the integrated mdio multiplexer which has the bus selection as well as mdio transaction generation logic inside. This multiplexer has child buses for PCIe, SATA, USB and ETH. These buses could be internal or external to SOC where PHYs are attached. These buses could use C-45 or C-22 mdio transaction. Signed-off-by: Pramod Kumar Reviewed-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig index 37d40c1..8dac88a 100644 --- a/drivers/net/phy/Kconfig +++ b/drivers/net/phy/Kconfig @@ -254,6 +254,17 @@ config MDIO_BUS_MUX_MMIOREG Currently, only 8-bit registers are supported. +config MDIO_BUS_MUX_BCM_IPROC + tristate "Support for iProc based MDIO bus multiplexers" + depends on OF && OF_MDIO && (ARCH_BCM_IPROC || COMPILE_TEST) + select MDIO_BUS_MUX + default ARCH_BCM_IPROC + help + This module provides a driver for MDIO bus multiplexers found in + iProc based Broadcom SoCs. This multiplexer connects one of several + child MDIO bus to a parent bus. Buses could be internal as well as + external and selection logic lies inside the same multiplexer. + config MDIO_BCM_UNIMAC tristate "Broadcom UniMAC MDIO bus controller" depends on HAS_IOMEM diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile index c26b651..4170642 100644 --- a/drivers/net/phy/Makefile +++ b/drivers/net/phy/Makefile @@ -39,6 +39,7 @@ obj-$(CONFIG_AMD_PHY) += amd.o obj-$(CONFIG_MDIO_BUS_MUX) += mdio-mux.o obj-$(CONFIG_MDIO_BUS_MUX_GPIO) += mdio-mux-gpio.o obj-$(CONFIG_MDIO_BUS_MUX_MMIOREG) += mdio-mux-mmioreg.o +obj-$(CONFIG_MDIO_BUS_MUX_BCM_IPROC) += mdio-mux-bcm-iproc.o obj-$(CONFIG_MDIO_SUN4I) += mdio-sun4i.o obj-$(CONFIG_MDIO_MOXART) += mdio-moxart.o obj-$(CONFIG_MDIO_BCM_UNIMAC) += mdio-bcm-unimac.o diff --git a/drivers/net/phy/mdio-mux-bcm-iproc.c b/drivers/net/phy/mdio-mux-bcm-iproc.c new file mode 100644 index 0000000..0a04125 --- /dev/null +++ b/drivers/net/phy/mdio-mux-bcm-iproc.c @@ -0,0 +1,248 @@ +/* + * Copyright 2016 Broadcom + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation (the "GPL"). + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License version 2 (GPLv2) for more details. + * + * You should have received a copy of the GNU General Public License + * version 2 (GPLv2) along with this source code. + */ + +#include +#include +#include +#include +#include +#include +#include + +#define MDIO_PARAM_OFFSET 0x00 +#define MDIO_PARAM_MIIM_CYCLE 29 +#define MDIO_PARAM_INTERNAL_SEL 25 +#define MDIO_PARAM_BUS_ID 22 +#define MDIO_PARAM_C45_SEL 21 +#define MDIO_PARAM_PHY_ID 16 +#define MDIO_PARAM_PHY_DATA 0 + +#define MDIO_READ_OFFSET 0x04 +#define MDIO_READ_DATA_MASK 0xffff +#define MDIO_ADDR_OFFSET 0x08 + +#define MDIO_CTRL_OFFSET 0x0C +#define MDIO_CTRL_WRITE_OP 0x1 +#define MDIO_CTRL_READ_OP 0x2 + +#define MDIO_STAT_OFFSET 0x10 +#define MDIO_STAT_DONE 1 + +#define BUS_MAX_ADDR 32 +#define EXT_BUS_START_ADDR 16 + +struct iproc_mdiomux_desc { + void *mux_handle; + void __iomem *base; + struct device *dev; + struct mii_bus *mii_bus; +}; + +static int iproc_mdio_wait_for_idle(void __iomem *base, bool result) +{ + unsigned int timeout = 1000; /* loop for 1s */ + u32 val; + + do { + val = readl(base + MDIO_STAT_OFFSET); + if ((val & MDIO_STAT_DONE) == result) + return 0; + + usleep_range(1000, 2000); + } while (timeout--); + + return -ETIMEDOUT; +} + +/* start_miim_ops- Program and start MDIO transaction over mdio bus. + * @base: Base address + * @phyid: phyid of the selected bus. + * @reg: register offset to be read/written. + * @val :0 if read op else value to be written in @reg; + * @op: Operation that need to be carried out. + * MDIO_CTRL_READ_OP: Read transaction. + * MDIO_CTRL_WRITE_OP: Write transaction. + * + * Return value: Successful Read operation returns read reg values and write + * operation returns 0. Failure operation returns negative error code. + */ +static int start_miim_ops(void __iomem *base, + u16 phyid, u32 reg, u16 val, u32 op) +{ + u32 param; + int ret; + + writel(0, base + MDIO_CTRL_OFFSET); + ret = iproc_mdio_wait_for_idle(base, 0); + if (ret) + goto err; + + param = readl(base + MDIO_PARAM_OFFSET); + param |= phyid << MDIO_PARAM_PHY_ID; + param |= val << MDIO_PARAM_PHY_DATA; + if (reg & MII_ADDR_C45) + param |= BIT(MDIO_PARAM_C45_SEL); + + writel(param, base + MDIO_PARAM_OFFSET); + + writel(reg, base + MDIO_ADDR_OFFSET); + + writel(op, base + MDIO_CTRL_OFFSET); + + ret = iproc_mdio_wait_for_idle(base, 1); + if (ret) + goto err; + + if (op == MDIO_CTRL_READ_OP) + ret = readl(base + MDIO_READ_OFFSET) & MDIO_READ_DATA_MASK; +err: + return ret; +} + +static int iproc_mdiomux_read(struct mii_bus *bus, int phyid, int reg) +{ + struct iproc_mdiomux_desc *md = bus->priv; + int ret; + + ret = start_miim_ops(md->base, phyid, reg, 0, MDIO_CTRL_READ_OP); + if (ret < 0) + dev_err(&bus->dev, "mdiomux read operation failed!!!"); + + return ret; +} + +static int iproc_mdiomux_write(struct mii_bus *bus, + int phyid, int reg, u16 val) +{ + struct iproc_mdiomux_desc *md = bus->priv; + int ret; + + /* Write val at reg offset */ + ret = start_miim_ops(md->base, phyid, reg, val, MDIO_CTRL_WRITE_OP); + if (ret < 0) + dev_err(&bus->dev, "mdiomux write operation failed!!!"); + + return ret; +} + +static int mdio_mux_iproc_switch_fn(int current_child, int desired_child, + void *data) +{ + struct iproc_mdiomux_desc *md = data; + u32 param, bus_id; + bool bus_dir; + + /* select bus and its properties */ + bus_dir = (desired_child < EXT_BUS_START_ADDR); + bus_id = bus_dir ? desired_child : (desired_child - EXT_BUS_START_ADDR); + + param = (bus_dir ? 1 : 0) << MDIO_PARAM_INTERNAL_SEL; + param |= (bus_id << MDIO_PARAM_BUS_ID); + + writel(param, md->base + MDIO_PARAM_OFFSET); + return 0; +} + +static int mdio_mux_iproc_probe(struct platform_device *pdev) +{ + struct iproc_mdiomux_desc *md; + struct mii_bus *bus; + struct resource *res; + int rc; + + md = devm_kzalloc(&pdev->dev, sizeof(*md), GFP_KERNEL); + if (!md) + return -ENOMEM; + md->dev = &pdev->dev; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + md->base = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(md->base)) { + dev_err(&pdev->dev, "failed to ioremap register\n"); + return PTR_ERR(md->base); + } + + md->mii_bus = mdiobus_alloc(); + if (!md->mii_bus) { + dev_err(&pdev->dev, "mdiomux bus alloc failed\n"); + return -ENOMEM; + } + + bus = md->mii_bus; + bus->priv = md; + bus->name = "iProc MDIO mux bus"; + snprintf(bus->id, MII_BUS_ID_SIZE, "%s-%d", pdev->name, pdev->id); + bus->parent = &pdev->dev; + bus->read = iproc_mdiomux_read; + bus->write = iproc_mdiomux_write; + + bus->phy_mask = ~0; + bus->dev.of_node = pdev->dev.of_node; + rc = mdiobus_register(bus); + if (rc) { + dev_err(&pdev->dev, "mdiomux registration failed\n"); + goto out; + } + + platform_set_drvdata(pdev, md); + + rc = mdio_mux_init(md->dev, mdio_mux_iproc_switch_fn, + &md->mux_handle, md, md->mii_bus); + if (rc) { + dev_info(md->dev, "mdiomux initialization failed\n"); + goto out; + } + + dev_info(md->dev, "iProc mdiomux registered\n"); + return 0; +out: + mdiobus_free(bus); + return rc; +} + +static int mdio_mux_iproc_remove(struct platform_device *pdev) +{ + struct iproc_mdiomux_desc *md = dev_get_platdata(&pdev->dev); + + mdio_mux_uninit(md->mux_handle); + mdiobus_unregister(md->mii_bus); + mdiobus_free(md->mii_bus); + + return 0; +} + +static const struct of_device_id mdio_mux_iproc_match[] = { + { + .compatible = "brcm,mdio-mux-iproc", + }, + {}, +}; +MODULE_DEVICE_TABLE(of, mdio_mux_iproc_match); + +static struct platform_driver mdiomux_iproc_driver = { + .driver = { + .name = "mdio-mux-iproc", + .of_match_table = mdio_mux_iproc_match, + }, + .probe = mdio_mux_iproc_probe, + .remove = mdio_mux_iproc_remove, +}; + +module_platform_driver(mdiomux_iproc_driver); + +MODULE_DESCRIPTION("iProc MDIO Mux Bus Driver"); +MODULE_AUTHOR("Pramod Kumar "); +MODULE_LICENSE("GPL v2"); -- cgit v0.10.2 From 464e3b4b118f5c7abfdb68f3628df0ca65d28948 Mon Sep 17 00:00:00 2001 From: Pramod Kumar Date: Fri, 10 Jun 2016 11:03:50 +0530 Subject: binding: PHY: Binding doc for NS2 PCIe PHYs. Binding doc for NS2 PCIe PHYs. Signed-off-by: Pramod Kumar Signed-off-by: Jon Mason Acked-by: Rob Herring Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/phy/brcm,mdio-mux-bus-pci.txt b/Documentation/devicetree/bindings/phy/brcm,mdio-mux-bus-pci.txt new file mode 100644 index 0000000..5b51007 --- /dev/null +++ b/Documentation/devicetree/bindings/phy/brcm,mdio-mux-bus-pci.txt @@ -0,0 +1,27 @@ +* Broadcom NS2 PCIe PHY binding document + +Required bus properties: +- reg: MDIO Bus number for the MDIO interface +- #address-cells: must be 1 +- #size-cells: must be 0 + +Required PHY properties: +- compatible: should be "brcm,ns2-pcie-phy" +- reg: MDIO Phy ID for the MDIO interface +- #phy-cells: must be 0 + +This is a child bus node of "brcm,mdio-mux-iproc" node. + +Example: + +mdio@0 { + reg = <0x0>; + #address-cells = <1>; + #size-cells = <0>; + + pci_phy0: pci-phy@0 { + compatible = "brcm,ns2-pcie-phy"; + reg = <0x0>; + #phy-cells = <0>; + }; +}; -- cgit v0.10.2 From 4484f730b3358363ca4b3b422354676581fab2cb Mon Sep 17 00:00:00 2001 From: Pramod Kumar Date: Fri, 10 Jun 2016 11:03:51 +0530 Subject: phy: Add Northstar2 PCI Phy support Add PCI Phy support for Broadcom Northstar2 SoCs. This driver uses the interface from the iproc mdio mux driver to enable the devices respective phys. Reviewed-by: Andrew Lunn Signed-off-by: Jon Mason Signed-off-by: Pramod Kumar Signed-off-by: David S. Miller diff --git a/drivers/phy/Kconfig b/drivers/phy/Kconfig index b869b98..01fb93b 100644 --- a/drivers/phy/Kconfig +++ b/drivers/phy/Kconfig @@ -434,4 +434,12 @@ config PHY_CYGNUS_PCIE source "drivers/phy/tegra/Kconfig" +config PHY_NS2_PCIE + tristate "Broadcom Northstar2 PCIe PHY driver" + depends on OF && MDIO_BUS_MUX_BCM_IPROC + select GENERIC_PHY + default ARCH_BCM_IPROC + help + Enable this to support the Broadcom Northstar2 PCIe PHY. + If unsure, say N. endmenu diff --git a/drivers/phy/Makefile b/drivers/phy/Makefile index 9c3e73c..7aea094 100644 --- a/drivers/phy/Makefile +++ b/drivers/phy/Makefile @@ -53,5 +53,5 @@ obj-$(CONFIG_PHY_TUSB1210) += phy-tusb1210.o obj-$(CONFIG_PHY_BRCM_SATA) += phy-brcm-sata.o obj-$(CONFIG_PHY_PISTACHIO_USB) += phy-pistachio-usb.o obj-$(CONFIG_PHY_CYGNUS_PCIE) += phy-bcm-cygnus-pcie.o - obj-$(CONFIG_ARCH_TEGRA) += tegra/ +obj-$(CONFIG_PHY_NS2_PCIE) += phy-bcm-ns2-pcie.o diff --git a/drivers/phy/phy-bcm-ns2-pcie.c b/drivers/phy/phy-bcm-ns2-pcie.c new file mode 100644 index 0000000..9513f7a --- /dev/null +++ b/drivers/phy/phy-bcm-ns2-pcie.c @@ -0,0 +1,115 @@ +/* + * Copyright (C) 2016 Broadcom + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation version 2. + * + * This program is distributed "as is" WITHOUT ANY WARRANTY of any + * kind, whether express or implied; without even the implied warranty + * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include +#include +#include +#include +#include +#include + +struct ns2_pci_phy { + struct mdio_device *mdiodev; + struct phy *phy; +}; + +#define BLK_ADDR_REG_OFFSET 0x1f +#define PLL_AFE1_100MHZ_BLK 0x2100 +#define PLL_CLK_AMP_OFFSET 0x03 +#define PLL_CLK_AMP_2P05V 0x2b18 + +static int ns2_pci_phy_init(struct phy *p) +{ + struct ns2_pci_phy *phy = phy_get_drvdata(p); + int rc; + + /* select the AFE 100MHz block page */ + rc = mdiobus_write(phy->mdiodev->bus, phy->mdiodev->addr, + BLK_ADDR_REG_OFFSET, PLL_AFE1_100MHZ_BLK); + if (rc) + goto err; + + /* set the 100 MHz reference clock amplitude to 2.05 v */ + rc = mdiobus_write(phy->mdiodev->bus, phy->mdiodev->addr, + PLL_CLK_AMP_OFFSET, PLL_CLK_AMP_2P05V); + if (rc) + goto err; + + return 0; + +err: + dev_err(&phy->mdiodev->dev, "Error %d writing to phy\n", rc); + return rc; +} + +static struct phy_ops ns2_pci_phy_ops = { + .init = ns2_pci_phy_init, +}; + +static int ns2_pci_phy_probe(struct mdio_device *mdiodev) +{ + struct device *dev = &mdiodev->dev; + struct phy_provider *provider; + struct ns2_pci_phy *p; + struct phy *phy; + + phy = devm_phy_create(dev, dev->of_node, &ns2_pci_phy_ops); + if (IS_ERR(phy)) { + dev_err(dev, "failed to create Phy\n"); + return PTR_ERR(phy); + } + + p = devm_kmalloc(dev, sizeof(struct ns2_pci_phy), + GFP_KERNEL); + if (!p) + return -ENOMEM; + + p->mdiodev = mdiodev; + dev_set_drvdata(dev, p); + + p->phy = phy; + phy_set_drvdata(phy, p); + + provider = devm_of_phy_provider_register(&phy->dev, + of_phy_simple_xlate); + if (IS_ERR(provider)) { + dev_err(dev, "failed to register Phy provider\n"); + return PTR_ERR(provider); + } + + dev_info(dev, "%s PHY registered\n", dev_name(dev)); + + return 0; +} + +static const struct of_device_id ns2_pci_phy_of_match[] = { + { .compatible = "brcm,ns2-pcie-phy", }, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, ns2_pci_phy_of_match); + +static struct mdio_driver ns2_pci_phy_driver = { + .mdiodrv = { + .driver = { + .name = "phy-bcm-ns2-pci", + .of_match_table = ns2_pci_phy_of_match, + }, + }, + .probe = ns2_pci_phy_probe, +}; +mdio_module_driver(ns2_pci_phy_driver); + +MODULE_AUTHOR("Broadcom"); +MODULE_DESCRIPTION("Broadcom Northstar2 PCI Phy driver"); +MODULE_LICENSE("GPL v2"); +MODULE_ALIAS("platform:phy-bcm-ns2-pci"); -- cgit v0.10.2 From 8fe6a79fb8088a759b3dc57eb641fc3183ad72b8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 10 Jun 2016 16:41:36 -0700 Subject: net_sched: sch_plug: use a private throttled status We want to get rid of generic qdisc throttled management, so this qdisc has to use a private flag. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_plug.c b/net/sched/sch_plug.c index ff0d968..a12cd37 100644 --- a/net/sched/sch_plug.c +++ b/net/sched/sch_plug.c @@ -64,6 +64,8 @@ struct plug_sched_data { */ bool unplug_indefinite; + bool throttled; + /* Queue Limit in bytes */ u32 limit; @@ -103,7 +105,7 @@ static struct sk_buff *plug_dequeue(struct Qdisc *sch) { struct plug_sched_data *q = qdisc_priv(sch); - if (qdisc_is_throttled(sch)) + if (q->throttled) return NULL; if (!q->unplug_indefinite) { @@ -111,7 +113,7 @@ static struct sk_buff *plug_dequeue(struct Qdisc *sch) /* No more packets to dequeue. Block the queue * and wait for the next release command. */ - qdisc_throttled(sch); + q->throttled = true; return NULL; } q->pkts_to_release--; @@ -141,7 +143,7 @@ static int plug_init(struct Qdisc *sch, struct nlattr *opt) q->limit = ctl->limit; } - qdisc_throttled(sch); + q->throttled = true; return 0; } @@ -173,7 +175,7 @@ static int plug_change(struct Qdisc *sch, struct nlattr *opt) q->pkts_last_epoch = q->pkts_current_epoch; q->pkts_current_epoch = 0; if (q->unplug_indefinite) - qdisc_throttled(sch); + q->throttled = true; q->unplug_indefinite = false; break; case TCQ_PLUG_RELEASE_ONE: @@ -182,7 +184,7 @@ static int plug_change(struct Qdisc *sch, struct nlattr *opt) */ q->pkts_to_release += q->pkts_last_epoch; q->pkts_last_epoch = 0; - qdisc_unthrottled(sch); + q->throttled = false; netif_schedule_queue(sch->dev_queue); break; case TCQ_PLUG_RELEASE_INDEFINITE: @@ -190,7 +192,7 @@ static int plug_change(struct Qdisc *sch, struct nlattr *opt) q->pkts_to_release = 0; q->pkts_last_epoch = 0; q->pkts_current_epoch = 0; - qdisc_unthrottled(sch); + q->throttled = false; netif_schedule_queue(sch->dev_queue); break; case TCQ_PLUG_LIMIT: -- cgit v0.10.2 From cca605dd4b3b2bfa381250b7dbbe16b124916f24 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 10 Jun 2016 16:41:37 -0700 Subject: net_sched: cbq: remove a flaky use of qdisc_is_throttled() So far no qdisc ever unset the throttled bit at enqueue() time, so CBQ usage of qdisc_is_throttled() was flaky. Since __QDISC_STATE_THROTTLED set/unset is way too expensive considering that only CBQ was eventually caring for this status, it would make sense to implement a Qdisc ops ->is_throttled() if we find that this is needed. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index f2af31b..6e61f9a 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -345,7 +345,7 @@ cbq_mark_toplevel(struct cbq_sched_data *q, struct cbq_class *cl) { int toplevel = q->toplevel; - if (toplevel > cl->level && !(qdisc_is_throttled(cl->q))) { + if (toplevel > cl->level) { psched_time_t now = psched_get_time(); do { -- cgit v0.10.2 From 42117927cab5a13192ecc227bea19da5059ffc6c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 10 Jun 2016 16:41:38 -0700 Subject: net_sched: netem: remove qdisc_is_throttled() use Looks like it is only there as some optimization attempt. Since __QDISC_STATE_THROTTLED set/unset is way too expensive, and netem is the last user, just remove this check. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 9ca7947..2dbe732 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -582,9 +582,6 @@ static struct sk_buff *netem_dequeue(struct Qdisc *sch) struct sk_buff *skb; struct rb_node *p; - if (qdisc_is_throttled(sch)) - return NULL; - tfifo_dequeue: skb = __skb_dequeue(&sch->q); if (skb) { -- cgit v0.10.2 From 45f50bed1d808794e514e9eed0e579a8756ce2ba Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 10 Jun 2016 16:41:39 -0700 Subject: net_sched: remove generic throttled management __QDISC_STATE_THROTTLED bit manipulation is rather expensive for HTB and few others. I already removed it for sch_fq in commit f2600cf02b5b ("net: sched: avoid costly atomic operation in fq_dequeue()") and so far nobody complained. When one ore more packets are stuck in one or more throttled HTB class, a htb dequeue() performs two atomic operations to clear/set __QDISC_STATE_THROTTLED bit, while root qdisc lock is held. Removing this pair of atomic operations bring me a 8 % performance increase on 200 TCP_RR tests, in presence of throttled classes. This patch has no side effect, since nothing actually uses disc_is_throttled() anymore. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index fea53f4..7caa99b 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -67,12 +67,12 @@ struct qdisc_watchdog { }; void qdisc_watchdog_init(struct qdisc_watchdog *wd, struct Qdisc *qdisc); -void qdisc_watchdog_schedule_ns(struct qdisc_watchdog *wd, u64 expires, bool throttle); +void qdisc_watchdog_schedule_ns(struct qdisc_watchdog *wd, u64 expires); static inline void qdisc_watchdog_schedule(struct qdisc_watchdog *wd, psched_time_t expires) { - qdisc_watchdog_schedule_ns(wd, PSCHED_TICKS2NS(expires), true); + qdisc_watchdog_schedule_ns(wd, PSCHED_TICKS2NS(expires)); } void qdisc_watchdog_cancel(struct qdisc_watchdog *wd); diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 9f35819..9a0d177 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -26,7 +26,6 @@ struct qdisc_rate_table { enum qdisc_state_t { __QDISC_STATE_SCHED, __QDISC_STATE_DEACTIVATED, - __QDISC_STATE_THROTTLED, }; struct qdisc_size_table { @@ -125,21 +124,6 @@ static inline int qdisc_avail_bulklimit(const struct netdev_queue *txq) #endif } -static inline bool qdisc_is_throttled(const struct Qdisc *qdisc) -{ - return test_bit(__QDISC_STATE_THROTTLED, &qdisc->state) ? true : false; -} - -static inline void qdisc_throttled(struct Qdisc *qdisc) -{ - set_bit(__QDISC_STATE_THROTTLED, &qdisc->state); -} - -static inline void qdisc_unthrottled(struct Qdisc *qdisc) -{ - clear_bit(__QDISC_STATE_THROTTLED, &qdisc->state); -} - struct Qdisc_class_ops { /* Child qdisc manipulation */ struct netdev_queue * (*select_queue)(struct Qdisc *, struct tcmsg *); diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index d4a8bbf..401eda6 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -583,7 +583,6 @@ static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer) timer); rcu_read_lock(); - qdisc_unthrottled(wd->qdisc); __netif_schedule(qdisc_root(wd->qdisc)); rcu_read_unlock(); @@ -598,15 +597,12 @@ void qdisc_watchdog_init(struct qdisc_watchdog *wd, struct Qdisc *qdisc) } EXPORT_SYMBOL(qdisc_watchdog_init); -void qdisc_watchdog_schedule_ns(struct qdisc_watchdog *wd, u64 expires, bool throttle) +void qdisc_watchdog_schedule_ns(struct qdisc_watchdog *wd, u64 expires) { if (test_bit(__QDISC_STATE_DEACTIVATED, &qdisc_root_sleeping(wd->qdisc)->state)) return; - if (throttle) - qdisc_throttled(wd->qdisc); - if (wd->last_expires == expires) return; @@ -620,7 +616,6 @@ EXPORT_SYMBOL(qdisc_watchdog_schedule_ns); void qdisc_watchdog_cancel(struct qdisc_watchdog *wd) { hrtimer_cancel(&wd->timer); - qdisc_unthrottled(wd->qdisc); } EXPORT_SYMBOL(qdisc_watchdog_cancel); diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 6e61f9a..a29fd81 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -513,7 +513,6 @@ static enum hrtimer_restart cbq_undelay(struct hrtimer *timer) hrtimer_start(&q->delay_timer, time, HRTIMER_MODE_ABS_PINNED); } - qdisc_unthrottled(sch); __netif_schedule(qdisc_root(sch)); return HRTIMER_NORESTART; } @@ -819,7 +818,6 @@ cbq_dequeue(struct Qdisc *sch) if (skb) { qdisc_bstats_update(sch, skb); sch->q.qlen--; - qdisc_unthrottled(sch); return skb; } diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index 3c6a47d..f49c81e 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -445,8 +445,7 @@ begin: if (!head->first) { if (q->time_next_delayed_flow != ~0ULL) qdisc_watchdog_schedule_ns(&q->watchdog, - q->time_next_delayed_flow, - false); + q->time_next_delayed_flow); return NULL; } } diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index eb3d3f5..bd08c36 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1664,7 +1664,6 @@ hfsc_dequeue(struct Qdisc *sch) set_passive(cl); } - qdisc_unthrottled(sch); qdisc_bstats_update(sch, skb); qdisc_qstats_backlog_dec(sch, skb); sch->q.qlen--; diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index b74d066..07dcd29 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -889,7 +889,6 @@ static struct sk_buff *htb_dequeue(struct Qdisc *sch) if (skb != NULL) { ok: qdisc_bstats_update(sch, skb); - qdisc_unthrottled(sch); qdisc_qstats_backlog_dec(sch, skb); sch->q.qlen--; return skb; @@ -929,7 +928,7 @@ ok: } qdisc_qstats_overlimit(sch); if (likely(next_event > q->now)) - qdisc_watchdog_schedule_ns(&q->watchdog, next_event, true); + qdisc_watchdog_schedule_ns(&q->watchdog, next_event); else schedule_work(&q->work); fin: diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 2dbe732..876df13 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -587,7 +587,6 @@ tfifo_dequeue: if (skb) { qdisc_qstats_backlog_dec(sch, skb); deliver: - qdisc_unthrottled(sch); qdisc_bstats_update(sch, skb); return skb; } diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index 7fa3d6e..c12df84 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -254,14 +254,12 @@ static struct sk_buff *tbf_dequeue(struct Qdisc *sch) q->ptokens = ptoks; qdisc_qstats_backlog_dec(sch, skb); sch->q.qlen--; - qdisc_unthrottled(sch); qdisc_bstats_update(sch, skb); return skb; } qdisc_watchdog_schedule_ns(&q->watchdog, - now + max_t(long, -toks, -ptoks), - true); + now + max_t(long, -toks, -ptoks)); /* Maybe we have a shorter packet in the queue, which can be sent now. It sounds cool, -- cgit v0.10.2 From 38b7097b55b6cf30adc5ac07cb1055683224393e Mon Sep 17 00:00:00 2001 From: Hannes Frederic Sowa Date: Sat, 11 Jun 2016 20:08:19 +0200 Subject: ipv6: use TOS marks from sockets for routing decision In IPv6 the ToS values are part of the flowlabel in flowi6 and get extracted during fib rule lookup, but we forgot to correctly initialize the flowlabel before the routing lookup. Reported-by: Signed-off-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 4527285..40454bf 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -502,12 +502,14 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) else if (!fl6.flowi6_oif) fl6.flowi6_oif = np->ucast_oif; + ipc6.tclass = np->tclass; + fl6.flowlabel = ip6_make_flowinfo(ipc6.tclass, fl6.flowlabel); + dst = icmpv6_route_lookup(net, skb, sk, &fl6); if (IS_ERR(dst)) goto out; ipc6.hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); - ipc6.tclass = np->tclass; ipc6.dontfrag = np->dontfrag; ipc6.opt = NULL; diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c index 3ee3e44..fed40d1 100644 --- a/net/ipv6/ping.c +++ b/net/ipv6/ping.c @@ -116,6 +116,9 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) else if (!fl6.flowi6_oif) fl6.flowi6_oif = np->ucast_oif; + ipc6.tclass = np->tclass; + fl6.flowlabel = ip6_make_flowinfo(ipc6.tclass, fl6.flowlabel); + dst = ip6_sk_dst_lookup_flow(sk, &fl6, daddr); if (IS_ERR(dst)) return PTR_ERR(dst); @@ -140,7 +143,6 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) pfh.family = AF_INET6; ipc6.hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); - ipc6.tclass = np->tclass; ipc6.dontfrag = np->dontfrag; ipc6.opt = NULL; diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 896350d..590dd1f 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -878,6 +878,11 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) if (inet->hdrincl) fl6.flowi6_flags |= FLOWI_FLAG_KNOWN_NH; + if (ipc6.tclass < 0) + ipc6.tclass = np->tclass; + + fl6.flowlabel = ip6_make_flowinfo(ipc6.tclass, fl6.flowlabel); + dst = ip6_dst_lookup_flow(sk, &fl6, final_p); if (IS_ERR(dst)) { err = PTR_ERR(dst); @@ -886,9 +891,6 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) if (ipc6.hlimit < 0) ipc6.hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); - if (ipc6.tclass < 0) - ipc6.tclass = np->tclass; - if (ipc6.dontfrag < 0) ipc6.dontfrag = np->dontfrag; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 969913d..c6ae6f9 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -3306,6 +3306,8 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh) err = -EINVAL; memset(&fl6, 0, sizeof(fl6)); + rtm = nlmsg_data(nlh); + fl6.flowlabel = ip6_make_flowinfo(rtm->rtm_tos, 0); if (tb[RTA_SRC]) { if (nla_len(tb[RTA_SRC]) < sizeof(struct in6_addr)) diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index f421c9f..4bb5c13 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -1246,6 +1246,11 @@ do_udp_sendmsg: security_sk_classify_flow(sk, flowi6_to_flowi(&fl6)); + if (ipc6.tclass < 0) + ipc6.tclass = np->tclass; + + fl6.flowlabel = ip6_make_flowinfo(ipc6.tclass, fl6.flowlabel); + dst = ip6_sk_dst_lookup_flow(sk, &fl6, final_p); if (IS_ERR(dst)) { err = PTR_ERR(dst); @@ -1256,9 +1261,6 @@ do_udp_sendmsg: if (ipc6.hlimit < 0) ipc6.hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); - if (ipc6.tclass < 0) - ipc6.tclass = np->tclass; - if (msg->msg_flags&MSG_CONFIRM) goto do_confirm; back_from_confirm: diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c index 6c54e03..ea2ae66 100644 --- a/net/l2tp/l2tp_ip6.c +++ b/net/l2tp/l2tp_ip6.c @@ -611,6 +611,11 @@ static int l2tp_ip6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) security_sk_classify_flow(sk, flowi6_to_flowi(&fl6)); + if (ipc6.tclass < 0) + ipc6.tclass = np->tclass; + + fl6.flowlabel = ip6_make_flowinfo(ipc6.tclass, fl6.flowlabel); + dst = ip6_dst_lookup_flow(sk, &fl6, final_p); if (IS_ERR(dst)) { err = PTR_ERR(dst); @@ -620,9 +625,6 @@ static int l2tp_ip6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) if (ipc6.hlimit < 0) ipc6.hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); - if (ipc6.tclass < 0) - ipc6.tclass = np->tclass; - if (ipc6.dontfrag < 0) ipc6.dontfrag = np->dontfrag; -- cgit v0.10.2 From eb37c56361df4a3e9705e808ce3f3bb483da3814 Mon Sep 17 00:00:00 2001 From: Stefan Wahren Date: Wed, 8 Jun 2016 20:42:46 +0000 Subject: net: fec: handle small PHY reset durations more precisely Since msleep is based on jiffies the PHY reset could take longer than expected. So use msleep for values greater than 20 msec otherwise usleep_range. Signed-off-by: Stefan Wahren Acked-by: Fugang Duan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index fea0f33..c36401e 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -3191,7 +3191,12 @@ static void fec_reset_phy(struct platform_device *pdev) dev_err(&pdev->dev, "failed to get phy-reset-gpios: %d\n", err); return; } - msleep(msec); + + if (msec > 20) + msleep(msec); + else + usleep_range(msec * 1000, msec * 1000 + 1000); + gpio_set_value_cansleep(phy_reset, !active_high); } #else /* CONFIG_OF */ -- cgit v0.10.2 From 99860208bc62d8ebd5c57495b84856506fe075bc Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sat, 11 Jun 2016 12:46:04 +0200 Subject: sched: remove NET_XMIT_POLICED sch_atm returns this when TC_ACT_SHOT classification occurs. But all other schedulers that use tc_classify (htb, hfsc, drr, fq_codel ...) return NET_XMIT_SUCCESS | __BYPASS in this case so just do that in atm. BATMAN uses it as an intermediate return value to signal forwarding vs. buffering, but it did not return POLICED to callers outside of BATMAN. Reviewed-by: Sven Eckelmann Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 94eef35..d101e4d 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -90,7 +90,6 @@ void netdev_set_default_ethtool_ops(struct net_device *dev, #define NET_XMIT_SUCCESS 0x00 #define NET_XMIT_DROP 0x01 /* skb dropped */ #define NET_XMIT_CN 0x02 /* congestion notification */ -#define NET_XMIT_POLICED 0x03 /* skb is shot by police */ #define NET_XMIT_MASK 0x0f /* qdisc flags in net/sch_generic.h */ /* NET_XMIT_CN is special. It does not guarantee that this packet is lost. It diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index e3857ed..f75091c 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -653,7 +653,7 @@ static int batadv_route_unicast_packet(struct sk_buff *skb, len + ETH_HLEN); ret = NET_RX_SUCCESS; - } else if (res == NET_XMIT_POLICED) { + } else if (res == -EINPROGRESS) { /* skb was buffered and consumed */ ret = NET_RX_SUCCESS; } diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c index f2f1256..b1a4e8a 100644 --- a/net/batman-adv/send.c +++ b/net/batman-adv/send.c @@ -156,7 +156,7 @@ int batadv_send_unicast_skb(struct sk_buff *skb, * attempted. * * Return: NET_XMIT_SUCCESS on success, NET_XMIT_DROP on failure, or - * NET_XMIT_POLICED if the skb is buffered for later transmit. + * -EINPROGRESS if the skb is buffered for later transmit. */ int batadv_send_skb_to_orig(struct sk_buff *skb, struct batadv_orig_node *orig_node, @@ -188,7 +188,7 @@ int batadv_send_skb_to_orig(struct sk_buff *skb, * network coding fails, then send the packet as usual. */ if (recv_if && batadv_nc_skb_forward(skb, neigh_node)) { - ret = NET_XMIT_POLICED; + ret = -EINPROGRESS; } else { batadv_send_unicast_skb(skb, neigh_node); ret = NET_XMIT_SUCCESS; diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 8b02df0..f74ab9c 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -3463,7 +3463,6 @@ xmit_more: break; case NET_XMIT_DROP: case NET_XMIT_CN: - case NET_XMIT_POLICED: /* skb has been consumed */ pkt_dev->errors++; break; diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 401eda6..12ebde8 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -95,8 +95,6 @@ static int tclass_notify(struct net *net, struct sk_buff *oskb, Expected action: do not backoff, but wait until queue will clear. NET_XMIT_CN - probably this packet enqueued, but another one dropped. Expected action: backoff or ignore - NET_XMIT_POLICED - dropped by police. - Expected action: backoff or error to real-time apps. Auxiliary routines: diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index 7e6c12d..0785b23 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -363,7 +363,7 @@ static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch) struct atm_flow_data *flow; struct tcf_result res; int result; - int ret = NET_XMIT_POLICED; + int ret = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; pr_debug("atm_tc_enqueue(skb %p,sch %p,[qdisc %p])\n", skb, sch, p); result = TC_POLICE_OK; /* be nice to gcc */ -- cgit v0.10.2 From 8c3e34a4ff85142ca5dba3f18cbc2061899e2612 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 13 Jun 2016 12:16:05 +0100 Subject: rxrpc: Rename files matching ar-*.c to git rid of the "ar-" prefix Rename files matching net/rxrpc/ar-*.c to get rid of the "ar-" prefix. This will aid splitting those files by making easier to come up with new names. Note that the not all files are simply renamed from ar-X.c to X.c. The following exceptions are made: (*) ar-call.c -> call_object.c ar-ack.c -> call_event.c call_object.c is going to contain the core of the call object handling. Call event handling is all going to be in call_event.c. (*) ar-accept.c -> call_accept.c Incoming call handling is going to be here. (*) ar-connection.c -> conn_object.c ar-connevent.c -> conn_event.c The former file is going to have the basic connection object handling, but there will likely be some differentiation between client connections and service connections in additional files later. The latter file will have all the connection-level event handling. (*) ar-local.c -> local_object.c This will have the local endpoint object handling code. The local endpoint event handling code will later be split out into local_event.c. (*) ar-peer.c -> peer_object.c This will have the peer endpoint object handling code. Peer event handling code will be placed in peer_event.c (for the moment, there is none). (*) ar-error.c -> peer_event.c This will become the peer event handling code, though for the moment it's actually driven from the local endpoint's perspective. Note that I haven't renamed ar-transport.c to transport_object.c as the intention is to delete it when the rxrpc_transport struct is excised. The only file that actually has its contents changed is net/rxrpc/Makefile. net/rxrpc/ar-internal.h will need its section marker comments updating, but I'll do that in a separate patch to make it easier for git to follow the history across the rename. I may also want to rename ar-internal.h at some point - but that would mean updating all the #includes and I'd rather do that in a separate step. Signed-off-by: David Howells -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * generate a connection-level abort - */ -static int rxrpc_busy(struct rxrpc_local *local, struct sockaddr_rxrpc *srx, - struct rxrpc_wire_header *whdr) -{ - struct msghdr msg; - struct kvec iov[1]; - size_t len; - int ret; - - _enter("%d,,", local->debug_id); - - whdr->type = RXRPC_PACKET_TYPE_BUSY; - whdr->serial = htonl(1); - - msg.msg_name = &srx->transport.sin; - msg.msg_namelen = sizeof(srx->transport.sin); - msg.msg_control = NULL; - msg.msg_controllen = 0; - msg.msg_flags = 0; - - iov[0].iov_base = whdr; - iov[0].iov_len = sizeof(*whdr); - - len = iov[0].iov_len; - - _proto("Tx BUSY %%1"); - - ret = kernel_sendmsg(local->socket, &msg, iov, 1, len); - if (ret < 0) { - _leave(" = -EAGAIN [sendmsg failed: %d]", ret); - return -EAGAIN; - } - - _leave(" = 0"); - return 0; -} - -/* - * accept an incoming call that needs peer, transport and/or connection setting - * up - */ -static int rxrpc_accept_incoming_call(struct rxrpc_local *local, - struct rxrpc_sock *rx, - struct sk_buff *skb, - struct sockaddr_rxrpc *srx) -{ - struct rxrpc_connection *conn; - struct rxrpc_transport *trans; - struct rxrpc_skb_priv *sp, *nsp; - struct rxrpc_peer *peer; - struct rxrpc_call *call; - struct sk_buff *notification; - int ret; - - _enter(""); - - sp = rxrpc_skb(skb); - - /* get a notification message to send to the server app */ - notification = alloc_skb(0, GFP_NOFS); - if (!notification) { - _debug("no memory"); - ret = -ENOMEM; - goto error_nofree; - } - rxrpc_new_skb(notification); - notification->mark = RXRPC_SKB_MARK_NEW_CALL; - - peer = rxrpc_get_peer(srx, GFP_NOIO); - if (IS_ERR(peer)) { - _debug("no peer"); - ret = -EBUSY; - goto error; - } - - trans = rxrpc_get_transport(local, peer, GFP_NOIO); - rxrpc_put_peer(peer); - if (IS_ERR(trans)) { - _debug("no trans"); - ret = -EBUSY; - goto error; - } - - conn = rxrpc_incoming_connection(trans, &sp->hdr); - rxrpc_put_transport(trans); - if (IS_ERR(conn)) { - _debug("no conn"); - ret = PTR_ERR(conn); - goto error; - } - - call = rxrpc_incoming_call(rx, conn, &sp->hdr); - rxrpc_put_connection(conn); - if (IS_ERR(call)) { - _debug("no call"); - ret = PTR_ERR(call); - goto error; - } - - /* attach the call to the socket */ - read_lock_bh(&local->services_lock); - if (rx->sk.sk_state == RXRPC_CLOSE) - goto invalid_service; - - write_lock(&rx->call_lock); - if (!test_and_set_bit(RXRPC_CALL_INIT_ACCEPT, &call->flags)) { - rxrpc_get_call(call); - - spin_lock(&call->conn->state_lock); - if (sp->hdr.securityIndex > 0 && - call->conn->state == RXRPC_CONN_SERVER_UNSECURED) { - _debug("await conn sec"); - list_add_tail(&call->accept_link, &rx->secureq); - call->conn->state = RXRPC_CONN_SERVER_CHALLENGING; - atomic_inc(&call->conn->usage); - set_bit(RXRPC_CONN_CHALLENGE, &call->conn->events); - rxrpc_queue_conn(call->conn); - } else { - _debug("conn ready"); - call->state = RXRPC_CALL_SERVER_ACCEPTING; - list_add_tail(&call->accept_link, &rx->acceptq); - rxrpc_get_call(call); - nsp = rxrpc_skb(notification); - nsp->call = call; - - ASSERTCMP(atomic_read(&call->usage), >=, 3); - - _debug("notify"); - spin_lock(&call->lock); - ret = rxrpc_queue_rcv_skb(call, notification, true, - false); - spin_unlock(&call->lock); - notification = NULL; - BUG_ON(ret < 0); - } - spin_unlock(&call->conn->state_lock); - - _debug("queued"); - } - write_unlock(&rx->call_lock); - - _debug("process"); - rxrpc_fast_process_packet(call, skb); - - _debug("done"); - read_unlock_bh(&local->services_lock); - rxrpc_free_skb(notification); - rxrpc_put_call(call); - _leave(" = 0"); - return 0; - -invalid_service: - _debug("invalid"); - read_unlock_bh(&local->services_lock); - - read_lock_bh(&call->state_lock); - if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && - !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) { - rxrpc_get_call(call); - rxrpc_queue_call(call); - } - read_unlock_bh(&call->state_lock); - rxrpc_put_call(call); - ret = -ECONNREFUSED; -error: - rxrpc_free_skb(notification); -error_nofree: - _leave(" = %d", ret); - return ret; -} - -/* - * accept incoming calls that need peer, transport and/or connection setting up - * - the packets we get are all incoming client DATA packets that have seq == 1 - */ -void rxrpc_accept_incoming_calls(struct work_struct *work) -{ - struct rxrpc_local *local = - container_of(work, struct rxrpc_local, acceptor); - struct rxrpc_skb_priv *sp; - struct sockaddr_rxrpc srx; - struct rxrpc_sock *rx; - struct rxrpc_wire_header whdr; - struct sk_buff *skb; - int ret; - - _enter("%d", local->debug_id); - - read_lock_bh(&rxrpc_local_lock); - if (atomic_read(&local->usage) > 0) - rxrpc_get_local(local); - else - local = NULL; - read_unlock_bh(&rxrpc_local_lock); - if (!local) { - _leave(" [local dead]"); - return; - } - -process_next_packet: - skb = skb_dequeue(&local->accept_queue); - if (!skb) { - rxrpc_put_local(local); - _leave("\n"); - return; - } - - _net("incoming call skb %p", skb); - - sp = rxrpc_skb(skb); - - /* Set up a response packet header in case we need it */ - whdr.epoch = htonl(sp->hdr.epoch); - whdr.cid = htonl(sp->hdr.cid); - whdr.callNumber = htonl(sp->hdr.callNumber); - whdr.seq = htonl(sp->hdr.seq); - whdr.serial = 0; - whdr.flags = 0; - whdr.type = 0; - whdr.userStatus = 0; - whdr.securityIndex = sp->hdr.securityIndex; - whdr._rsvd = 0; - whdr.serviceId = htons(sp->hdr.serviceId); - - /* determine the remote address */ - memset(&srx, 0, sizeof(srx)); - srx.srx_family = AF_RXRPC; - srx.transport.family = local->srx.transport.family; - srx.transport_type = local->srx.transport_type; - switch (srx.transport.family) { - case AF_INET: - srx.transport_len = sizeof(struct sockaddr_in); - srx.transport.sin.sin_port = udp_hdr(skb)->source; - srx.transport.sin.sin_addr.s_addr = ip_hdr(skb)->saddr; - break; - default: - goto busy; - } - - /* get the socket providing the service */ - read_lock_bh(&local->services_lock); - list_for_each_entry(rx, &local->services, listen_link) { - if (rx->srx.srx_service == sp->hdr.serviceId && - rx->sk.sk_state != RXRPC_CLOSE) - goto found_service; - } - read_unlock_bh(&local->services_lock); - goto invalid_service; - -found_service: - _debug("found service %hd", rx->srx.srx_service); - if (sk_acceptq_is_full(&rx->sk)) - goto backlog_full; - sk_acceptq_added(&rx->sk); - sock_hold(&rx->sk); - read_unlock_bh(&local->services_lock); - - ret = rxrpc_accept_incoming_call(local, rx, skb, &srx); - if (ret < 0) - sk_acceptq_removed(&rx->sk); - sock_put(&rx->sk); - switch (ret) { - case -ECONNRESET: /* old calls are ignored */ - case -ECONNABORTED: /* aborted calls are reaborted or ignored */ - case 0: - goto process_next_packet; - case -ECONNREFUSED: - goto invalid_service; - case -EBUSY: - goto busy; - case -EKEYREJECTED: - goto security_mismatch; - default: - BUG(); - } - -backlog_full: - read_unlock_bh(&local->services_lock); -busy: - rxrpc_busy(local, &srx, &whdr); - rxrpc_free_skb(skb); - goto process_next_packet; - -invalid_service: - skb->priority = RX_INVALID_OPERATION; - rxrpc_reject_packet(local, skb); - goto process_next_packet; - - /* can't change connection security type mid-flow */ -security_mismatch: - skb->priority = RX_PROTOCOL_ERROR; - rxrpc_reject_packet(local, skb); - goto process_next_packet; -} - -/* - * handle acceptance of a call by userspace - * - assign the user call ID to the call at the front of the queue - */ -struct rxrpc_call *rxrpc_accept_call(struct rxrpc_sock *rx, - unsigned long user_call_ID) -{ - struct rxrpc_call *call; - struct rb_node *parent, **pp; - int ret; - - _enter(",%lx", user_call_ID); - - ASSERT(!irqs_disabled()); - - write_lock(&rx->call_lock); - - ret = -ENODATA; - if (list_empty(&rx->acceptq)) - goto out; - - /* check the user ID isn't already in use */ - ret = -EBADSLT; - pp = &rx->calls.rb_node; - parent = NULL; - while (*pp) { - parent = *pp; - call = rb_entry(parent, struct rxrpc_call, sock_node); - - if (user_call_ID < call->user_call_ID) - pp = &(*pp)->rb_left; - else if (user_call_ID > call->user_call_ID) - pp = &(*pp)->rb_right; - else - goto out; - } - - /* dequeue the first call and check it's still valid */ - call = list_entry(rx->acceptq.next, struct rxrpc_call, accept_link); - list_del_init(&call->accept_link); - sk_acceptq_removed(&rx->sk); - - write_lock_bh(&call->state_lock); - switch (call->state) { - case RXRPC_CALL_SERVER_ACCEPTING: - call->state = RXRPC_CALL_SERVER_RECV_REQUEST; - break; - case RXRPC_CALL_REMOTELY_ABORTED: - case RXRPC_CALL_LOCALLY_ABORTED: - ret = -ECONNABORTED; - goto out_release; - case RXRPC_CALL_NETWORK_ERROR: - ret = call->conn->error; - goto out_release; - case RXRPC_CALL_DEAD: - ret = -ETIME; - goto out_discard; - default: - BUG(); - } - - /* formalise the acceptance */ - call->user_call_ID = user_call_ID; - rb_link_node(&call->sock_node, parent, pp); - rb_insert_color(&call->sock_node, &rx->calls); - if (test_and_set_bit(RXRPC_CALL_HAS_USERID, &call->flags)) - BUG(); - if (test_and_set_bit(RXRPC_CALL_EV_ACCEPTED, &call->events)) - BUG(); - rxrpc_queue_call(call); - - rxrpc_get_call(call); - write_unlock_bh(&call->state_lock); - write_unlock(&rx->call_lock); - _leave(" = %p{%d}", call, call->debug_id); - return call; - - /* if the call is already dying or dead, then we leave the socket's ref - * on it to be released by rxrpc_dead_call_expired() as induced by - * rxrpc_release_call() */ -out_release: - _debug("release %p", call); - if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && - !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) - rxrpc_queue_call(call); -out_discard: - write_unlock_bh(&call->state_lock); - _debug("discard %p", call); -out: - write_unlock(&rx->call_lock); - _leave(" = %d", ret); - return ERR_PTR(ret); -} - -/* - * Handle rejection of a call by userspace - * - reject the call at the front of the queue - */ -int rxrpc_reject_call(struct rxrpc_sock *rx) -{ - struct rxrpc_call *call; - int ret; - - _enter(""); - - ASSERT(!irqs_disabled()); - - write_lock(&rx->call_lock); - - ret = -ENODATA; - if (list_empty(&rx->acceptq)) - goto out; - - /* dequeue the first call and check it's still valid */ - call = list_entry(rx->acceptq.next, struct rxrpc_call, accept_link); - list_del_init(&call->accept_link); - sk_acceptq_removed(&rx->sk); - - write_lock_bh(&call->state_lock); - switch (call->state) { - case RXRPC_CALL_SERVER_ACCEPTING: - call->state = RXRPC_CALL_SERVER_BUSY; - if (test_and_set_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events)) - rxrpc_queue_call(call); - ret = 0; - goto out_release; - case RXRPC_CALL_REMOTELY_ABORTED: - case RXRPC_CALL_LOCALLY_ABORTED: - ret = -ECONNABORTED; - goto out_release; - case RXRPC_CALL_NETWORK_ERROR: - ret = call->conn->error; - goto out_release; - case RXRPC_CALL_DEAD: - ret = -ETIME; - goto out_discard; - default: - BUG(); - } - - /* if the call is already dying or dead, then we leave the socket's ref - * on it to be released by rxrpc_dead_call_expired() as induced by - * rxrpc_release_call() */ -out_release: - _debug("release %p", call); - if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && - !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) - rxrpc_queue_call(call); -out_discard: - write_unlock_bh(&call->state_lock); - _debug("discard %p", call); -out: - write_unlock(&rx->call_lock); - _leave(" = %d", ret); - return ret; -} - -/** - * rxrpc_kernel_accept_call - Allow a kernel service to accept an incoming call - * @sock: The socket on which the impending call is waiting - * @user_call_ID: The tag to attach to the call - * - * Allow a kernel service to accept an incoming call, assuming the incoming - * call is still valid. - */ -struct rxrpc_call *rxrpc_kernel_accept_call(struct socket *sock, - unsigned long user_call_ID) -{ - struct rxrpc_call *call; - - _enter(",%lx", user_call_ID); - call = rxrpc_accept_call(rxrpc_sk(sock->sk), user_call_ID); - _leave(" = %p", call); - return call; -} -EXPORT_SYMBOL(rxrpc_kernel_accept_call); - -/** - * rxrpc_kernel_reject_call - Allow a kernel service to reject an incoming call - * @sock: The socket on which the impending call is waiting - * - * Allow a kernel service to reject an incoming call with a BUSY message, - * assuming the incoming call is still valid. - */ -int rxrpc_kernel_reject_call(struct socket *sock) -{ - int ret; - - _enter(""); - ret = rxrpc_reject_call(rxrpc_sk(sock->sk)); - _leave(" = %d", ret); - return ret; -} -EXPORT_SYMBOL(rxrpc_kernel_reject_call); diff --git a/net/rxrpc/ar-ack.c b/net/rxrpc/ar-ack.c deleted file mode 100644 index 1838178..0000000 --- a/net/rxrpc/ar-ack.c +++ /dev/null @@ -1,1288 +0,0 @@ -/* Management of Tx window, Tx resend, ACKs and out-of-sequence reception - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * propose an ACK be sent - */ -void __rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason, - u32 serial, bool immediate) -{ - unsigned long expiry; - s8 prior = rxrpc_ack_priority[ack_reason]; - - ASSERTCMP(prior, >, 0); - - _enter("{%d},%s,%%%x,%u", - call->debug_id, rxrpc_acks(ack_reason), serial, immediate); - - if (prior < rxrpc_ack_priority[call->ackr_reason]) { - if (immediate) - goto cancel_timer; - return; - } - - /* update DELAY, IDLE, REQUESTED and PING_RESPONSE ACK serial - * numbers */ - if (prior == rxrpc_ack_priority[call->ackr_reason]) { - if (prior <= 4) - call->ackr_serial = serial; - if (immediate) - goto cancel_timer; - return; - } - - call->ackr_reason = ack_reason; - call->ackr_serial = serial; - - switch (ack_reason) { - case RXRPC_ACK_DELAY: - _debug("run delay timer"); - expiry = rxrpc_soft_ack_delay; - goto run_timer; - - case RXRPC_ACK_IDLE: - if (!immediate) { - _debug("run defer timer"); - expiry = rxrpc_idle_ack_delay; - goto run_timer; - } - goto cancel_timer; - - case RXRPC_ACK_REQUESTED: - expiry = rxrpc_requested_ack_delay; - if (!expiry) - goto cancel_timer; - if (!immediate || serial == 1) { - _debug("run defer timer"); - goto run_timer; - } - - default: - _debug("immediate ACK"); - goto cancel_timer; - } - -run_timer: - expiry += jiffies; - if (!timer_pending(&call->ack_timer) || - time_after(call->ack_timer.expires, expiry)) - mod_timer(&call->ack_timer, expiry); - return; - -cancel_timer: - _debug("cancel timer %%%u", serial); - try_to_del_timer_sync(&call->ack_timer); - read_lock_bh(&call->state_lock); - if (call->state <= RXRPC_CALL_COMPLETE && - !test_and_set_bit(RXRPC_CALL_EV_ACK, &call->events)) - rxrpc_queue_call(call); - read_unlock_bh(&call->state_lock); -} - -/* - * propose an ACK be sent, locking the call structure - */ -void rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason, - u32 serial, bool immediate) -{ - s8 prior = rxrpc_ack_priority[ack_reason]; - - if (prior > rxrpc_ack_priority[call->ackr_reason]) { - spin_lock_bh(&call->lock); - __rxrpc_propose_ACK(call, ack_reason, serial, immediate); - spin_unlock_bh(&call->lock); - } -} - -/* - * set the resend timer - */ -static void rxrpc_set_resend(struct rxrpc_call *call, u8 resend, - unsigned long resend_at) -{ - read_lock_bh(&call->state_lock); - if (call->state >= RXRPC_CALL_COMPLETE) - resend = 0; - - if (resend & 1) { - _debug("SET RESEND"); - set_bit(RXRPC_CALL_EV_RESEND, &call->events); - } - - if (resend & 2) { - _debug("MODIFY RESEND TIMER"); - set_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); - mod_timer(&call->resend_timer, resend_at); - } else { - _debug("KILL RESEND TIMER"); - del_timer_sync(&call->resend_timer); - clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events); - clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); - } - read_unlock_bh(&call->state_lock); -} - -/* - * resend packets - */ -static void rxrpc_resend(struct rxrpc_call *call) -{ - struct rxrpc_wire_header *whdr; - struct rxrpc_skb_priv *sp; - struct sk_buff *txb; - unsigned long *p_txb, resend_at; - bool stop; - int loop; - u8 resend; - - _enter("{%d,%d,%d,%d},", - call->acks_hard, call->acks_unacked, - atomic_read(&call->sequence), - CIRC_CNT(call->acks_head, call->acks_tail, call->acks_winsz)); - - stop = false; - resend = 0; - resend_at = 0; - - for (loop = call->acks_tail; - loop != call->acks_head || stop; - loop = (loop + 1) & (call->acks_winsz - 1) - ) { - p_txb = call->acks_window + loop; - smp_read_barrier_depends(); - if (*p_txb & 1) - continue; - - txb = (struct sk_buff *) *p_txb; - sp = rxrpc_skb(txb); - - if (sp->need_resend) { - sp->need_resend = false; - - /* each Tx packet has a new serial number */ - sp->hdr.serial = atomic_inc_return(&call->conn->serial); - - whdr = (struct rxrpc_wire_header *)txb->head; - whdr->serial = htonl(sp->hdr.serial); - - _proto("Tx DATA %%%u { #%d }", - sp->hdr.serial, sp->hdr.seq); - if (rxrpc_send_packet(call->conn->trans, txb) < 0) { - stop = true; - sp->resend_at = jiffies + 3; - } else { - sp->resend_at = - jiffies + rxrpc_resend_timeout; - } - } - - if (time_after_eq(jiffies + 1, sp->resend_at)) { - sp->need_resend = true; - resend |= 1; - } else if (resend & 2) { - if (time_before(sp->resend_at, resend_at)) - resend_at = sp->resend_at; - } else { - resend_at = sp->resend_at; - resend |= 2; - } - } - - rxrpc_set_resend(call, resend, resend_at); - _leave(""); -} - -/* - * handle resend timer expiry - */ -static void rxrpc_resend_timer(struct rxrpc_call *call) -{ - struct rxrpc_skb_priv *sp; - struct sk_buff *txb; - unsigned long *p_txb, resend_at; - int loop; - u8 resend; - - _enter("%d,%d,%d", - call->acks_tail, call->acks_unacked, call->acks_head); - - if (call->state >= RXRPC_CALL_COMPLETE) - return; - - resend = 0; - resend_at = 0; - - for (loop = call->acks_unacked; - loop != call->acks_head; - loop = (loop + 1) & (call->acks_winsz - 1) - ) { - p_txb = call->acks_window + loop; - smp_read_barrier_depends(); - txb = (struct sk_buff *) (*p_txb & ~1); - sp = rxrpc_skb(txb); - - ASSERT(!(*p_txb & 1)); - - if (sp->need_resend) { - ; - } else if (time_after_eq(jiffies + 1, sp->resend_at)) { - sp->need_resend = true; - resend |= 1; - } else if (resend & 2) { - if (time_before(sp->resend_at, resend_at)) - resend_at = sp->resend_at; - } else { - resend_at = sp->resend_at; - resend |= 2; - } - } - - rxrpc_set_resend(call, resend, resend_at); - _leave(""); -} - -/* - * process soft ACKs of our transmitted packets - * - these indicate packets the peer has or has not received, but hasn't yet - * given to the consumer, and so can still be discarded and re-requested - */ -static int rxrpc_process_soft_ACKs(struct rxrpc_call *call, - struct rxrpc_ackpacket *ack, - struct sk_buff *skb) -{ - struct rxrpc_skb_priv *sp; - struct sk_buff *txb; - unsigned long *p_txb, resend_at; - int loop; - u8 sacks[RXRPC_MAXACKS], resend; - - _enter("{%d,%d},{%d},", - call->acks_hard, - CIRC_CNT(call->acks_head, call->acks_tail, call->acks_winsz), - ack->nAcks); - - if (skb_copy_bits(skb, 0, sacks, ack->nAcks) < 0) - goto protocol_error; - - resend = 0; - resend_at = 0; - for (loop = 0; loop < ack->nAcks; loop++) { - p_txb = call->acks_window; - p_txb += (call->acks_tail + loop) & (call->acks_winsz - 1); - smp_read_barrier_depends(); - txb = (struct sk_buff *) (*p_txb & ~1); - sp = rxrpc_skb(txb); - - switch (sacks[loop]) { - case RXRPC_ACK_TYPE_ACK: - sp->need_resend = false; - *p_txb |= 1; - break; - case RXRPC_ACK_TYPE_NACK: - sp->need_resend = true; - *p_txb &= ~1; - resend = 1; - break; - default: - _debug("Unsupported ACK type %d", sacks[loop]); - goto protocol_error; - } - } - - smp_mb(); - call->acks_unacked = (call->acks_tail + loop) & (call->acks_winsz - 1); - - /* anything not explicitly ACK'd is implicitly NACK'd, but may just not - * have been received or processed yet by the far end */ - for (loop = call->acks_unacked; - loop != call->acks_head; - loop = (loop + 1) & (call->acks_winsz - 1) - ) { - p_txb = call->acks_window + loop; - smp_read_barrier_depends(); - txb = (struct sk_buff *) (*p_txb & ~1); - sp = rxrpc_skb(txb); - - if (*p_txb & 1) { - /* packet must have been discarded */ - sp->need_resend = true; - *p_txb &= ~1; - resend |= 1; - } else if (sp->need_resend) { - ; - } else if (time_after_eq(jiffies + 1, sp->resend_at)) { - sp->need_resend = true; - resend |= 1; - } else if (resend & 2) { - if (time_before(sp->resend_at, resend_at)) - resend_at = sp->resend_at; - } else { - resend_at = sp->resend_at; - resend |= 2; - } - } - - rxrpc_set_resend(call, resend, resend_at); - _leave(" = 0"); - return 0; - -protocol_error: - _leave(" = -EPROTO"); - return -EPROTO; -} - -/* - * discard hard-ACK'd packets from the Tx window - */ -static void rxrpc_rotate_tx_window(struct rxrpc_call *call, u32 hard) -{ - unsigned long _skb; - int tail = call->acks_tail, old_tail; - int win = CIRC_CNT(call->acks_head, tail, call->acks_winsz); - - _enter("{%u,%u},%u", call->acks_hard, win, hard); - - ASSERTCMP(hard - call->acks_hard, <=, win); - - while (call->acks_hard < hard) { - smp_read_barrier_depends(); - _skb = call->acks_window[tail] & ~1; - rxrpc_free_skb((struct sk_buff *) _skb); - old_tail = tail; - tail = (tail + 1) & (call->acks_winsz - 1); - call->acks_tail = tail; - if (call->acks_unacked == old_tail) - call->acks_unacked = tail; - call->acks_hard++; - } - - wake_up(&call->tx_waitq); -} - -/* - * clear the Tx window in the event of a failure - */ -static void rxrpc_clear_tx_window(struct rxrpc_call *call) -{ - rxrpc_rotate_tx_window(call, atomic_read(&call->sequence)); -} - -/* - * drain the out of sequence received packet queue into the packet Rx queue - */ -static int rxrpc_drain_rx_oos_queue(struct rxrpc_call *call) -{ - struct rxrpc_skb_priv *sp; - struct sk_buff *skb; - bool terminal; - int ret; - - _enter("{%d,%d}", call->rx_data_post, call->rx_first_oos); - - spin_lock_bh(&call->lock); - - ret = -ECONNRESET; - if (test_bit(RXRPC_CALL_RELEASED, &call->flags)) - goto socket_unavailable; - - skb = skb_dequeue(&call->rx_oos_queue); - if (skb) { - sp = rxrpc_skb(skb); - - _debug("drain OOS packet %d [%d]", - sp->hdr.seq, call->rx_first_oos); - - if (sp->hdr.seq != call->rx_first_oos) { - skb_queue_head(&call->rx_oos_queue, skb); - call->rx_first_oos = rxrpc_skb(skb)->hdr.seq; - _debug("requeue %p {%u}", skb, call->rx_first_oos); - } else { - skb->mark = RXRPC_SKB_MARK_DATA; - terminal = ((sp->hdr.flags & RXRPC_LAST_PACKET) && - !(sp->hdr.flags & RXRPC_CLIENT_INITIATED)); - ret = rxrpc_queue_rcv_skb(call, skb, true, terminal); - BUG_ON(ret < 0); - _debug("drain #%u", call->rx_data_post); - call->rx_data_post++; - - /* find out what the next packet is */ - skb = skb_peek(&call->rx_oos_queue); - if (skb) - call->rx_first_oos = rxrpc_skb(skb)->hdr.seq; - else - call->rx_first_oos = 0; - _debug("peek %p {%u}", skb, call->rx_first_oos); - } - } - - ret = 0; -socket_unavailable: - spin_unlock_bh(&call->lock); - _leave(" = %d", ret); - return ret; -} - -/* - * insert an out of sequence packet into the buffer - */ -static void rxrpc_insert_oos_packet(struct rxrpc_call *call, - struct sk_buff *skb) -{ - struct rxrpc_skb_priv *sp, *psp; - struct sk_buff *p; - u32 seq; - - sp = rxrpc_skb(skb); - seq = sp->hdr.seq; - _enter(",,{%u}", seq); - - skb->destructor = rxrpc_packet_destructor; - ASSERTCMP(sp->call, ==, NULL); - sp->call = call; - rxrpc_get_call(call); - - /* insert into the buffer in sequence order */ - spin_lock_bh(&call->lock); - - skb_queue_walk(&call->rx_oos_queue, p) { - psp = rxrpc_skb(p); - if (psp->hdr.seq > seq) { - _debug("insert oos #%u before #%u", seq, psp->hdr.seq); - skb_insert(p, skb, &call->rx_oos_queue); - goto inserted; - } - } - - _debug("append oos #%u", seq); - skb_queue_tail(&call->rx_oos_queue, skb); -inserted: - - /* we might now have a new front to the queue */ - if (call->rx_first_oos == 0 || seq < call->rx_first_oos) - call->rx_first_oos = seq; - - read_lock(&call->state_lock); - if (call->state < RXRPC_CALL_COMPLETE && - call->rx_data_post == call->rx_first_oos) { - _debug("drain rx oos now"); - set_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events); - } - read_unlock(&call->state_lock); - - spin_unlock_bh(&call->lock); - _leave(" [stored #%u]", call->rx_first_oos); -} - -/* - * clear the Tx window on final ACK reception - */ -static void rxrpc_zap_tx_window(struct rxrpc_call *call) -{ - struct rxrpc_skb_priv *sp; - struct sk_buff *skb; - unsigned long _skb, *acks_window; - u8 winsz = call->acks_winsz; - int tail; - - acks_window = call->acks_window; - call->acks_window = NULL; - - while (CIRC_CNT(call->acks_head, call->acks_tail, winsz) > 0) { - tail = call->acks_tail; - smp_read_barrier_depends(); - _skb = acks_window[tail] & ~1; - smp_mb(); - call->acks_tail = (call->acks_tail + 1) & (winsz - 1); - - skb = (struct sk_buff *) _skb; - sp = rxrpc_skb(skb); - _debug("+++ clear Tx %u", sp->hdr.seq); - rxrpc_free_skb(skb); - } - - kfree(acks_window); -} - -/* - * process the extra information that may be appended to an ACK packet - */ -static void rxrpc_extract_ackinfo(struct rxrpc_call *call, struct sk_buff *skb, - unsigned int latest, int nAcks) -{ - struct rxrpc_ackinfo ackinfo; - struct rxrpc_peer *peer; - unsigned int mtu; - - if (skb_copy_bits(skb, nAcks + 3, &ackinfo, sizeof(ackinfo)) < 0) { - _leave(" [no ackinfo]"); - return; - } - - _proto("Rx ACK %%%u Info { rx=%u max=%u rwin=%u jm=%u }", - latest, - ntohl(ackinfo.rxMTU), ntohl(ackinfo.maxMTU), - ntohl(ackinfo.rwind), ntohl(ackinfo.jumbo_max)); - - mtu = min(ntohl(ackinfo.rxMTU), ntohl(ackinfo.maxMTU)); - - peer = call->conn->trans->peer; - if (mtu < peer->maxdata) { - spin_lock_bh(&peer->lock); - peer->maxdata = mtu; - peer->mtu = mtu + peer->hdrsize; - spin_unlock_bh(&peer->lock); - _net("Net MTU %u (maxdata %u)", peer->mtu, peer->maxdata); - } -} - -/* - * process packets in the reception queue - */ -static int rxrpc_process_rx_queue(struct rxrpc_call *call, - u32 *_abort_code) -{ - struct rxrpc_ackpacket ack; - struct rxrpc_skb_priv *sp; - struct sk_buff *skb; - bool post_ACK; - int latest; - u32 hard, tx; - - _enter(""); - -process_further: - skb = skb_dequeue(&call->rx_queue); - if (!skb) - return -EAGAIN; - - _net("deferred skb %p", skb); - - sp = rxrpc_skb(skb); - - _debug("process %s [st %d]", rxrpc_pkts[sp->hdr.type], call->state); - - post_ACK = false; - - switch (sp->hdr.type) { - /* data packets that wind up here have been received out of - * order, need security processing or are jumbo packets */ - case RXRPC_PACKET_TYPE_DATA: - _proto("OOSQ DATA %%%u { #%u }", sp->hdr.serial, sp->hdr.seq); - - /* secured packets must be verified and possibly decrypted */ - if (call->conn->security->verify_packet(call, skb, - _abort_code) < 0) - goto protocol_error; - - rxrpc_insert_oos_packet(call, skb); - goto process_further; - - /* partial ACK to process */ - case RXRPC_PACKET_TYPE_ACK: - if (skb_copy_bits(skb, 0, &ack, sizeof(ack)) < 0) { - _debug("extraction failure"); - goto protocol_error; - } - if (!skb_pull(skb, sizeof(ack))) - BUG(); - - latest = sp->hdr.serial; - hard = ntohl(ack.firstPacket); - tx = atomic_read(&call->sequence); - - _proto("Rx ACK %%%u { m=%hu f=#%u p=#%u s=%%%u r=%s n=%u }", - latest, - ntohs(ack.maxSkew), - hard, - ntohl(ack.previousPacket), - ntohl(ack.serial), - rxrpc_acks(ack.reason), - ack.nAcks); - - rxrpc_extract_ackinfo(call, skb, latest, ack.nAcks); - - if (ack.reason == RXRPC_ACK_PING) { - _proto("Rx ACK %%%u PING Request", latest); - rxrpc_propose_ACK(call, RXRPC_ACK_PING_RESPONSE, - sp->hdr.serial, true); - } - - /* discard any out-of-order or duplicate ACKs */ - if (latest - call->acks_latest <= 0) { - _debug("discard ACK %d <= %d", - latest, call->acks_latest); - goto discard; - } - call->acks_latest = latest; - - if (call->state != RXRPC_CALL_CLIENT_SEND_REQUEST && - call->state != RXRPC_CALL_CLIENT_AWAIT_REPLY && - call->state != RXRPC_CALL_SERVER_SEND_REPLY && - call->state != RXRPC_CALL_SERVER_AWAIT_ACK) - goto discard; - - _debug("Tx=%d H=%u S=%d", tx, call->acks_hard, call->state); - - if (hard > 0) { - if (hard - 1 > tx) { - _debug("hard-ACK'd packet %d not transmitted" - " (%d top)", - hard - 1, tx); - goto protocol_error; - } - - if ((call->state == RXRPC_CALL_CLIENT_AWAIT_REPLY || - call->state == RXRPC_CALL_SERVER_AWAIT_ACK) && - hard > tx) { - call->acks_hard = tx; - goto all_acked; - } - - smp_rmb(); - rxrpc_rotate_tx_window(call, hard - 1); - } - - if (ack.nAcks > 0) { - if (hard - 1 + ack.nAcks > tx) { - _debug("soft-ACK'd packet %d+%d not" - " transmitted (%d top)", - hard - 1, ack.nAcks, tx); - goto protocol_error; - } - - if (rxrpc_process_soft_ACKs(call, &ack, skb) < 0) - goto protocol_error; - } - goto discard; - - /* complete ACK to process */ - case RXRPC_PACKET_TYPE_ACKALL: - goto all_acked; - - /* abort and busy are handled elsewhere */ - case RXRPC_PACKET_TYPE_BUSY: - case RXRPC_PACKET_TYPE_ABORT: - BUG(); - - /* connection level events - also handled elsewhere */ - case RXRPC_PACKET_TYPE_CHALLENGE: - case RXRPC_PACKET_TYPE_RESPONSE: - case RXRPC_PACKET_TYPE_DEBUG: - BUG(); - } - - /* if we've had a hard ACK that covers all the packets we've sent, then - * that ends that phase of the operation */ -all_acked: - write_lock_bh(&call->state_lock); - _debug("ack all %d", call->state); - - switch (call->state) { - case RXRPC_CALL_CLIENT_AWAIT_REPLY: - call->state = RXRPC_CALL_CLIENT_RECV_REPLY; - break; - case RXRPC_CALL_SERVER_AWAIT_ACK: - _debug("srv complete"); - call->state = RXRPC_CALL_COMPLETE; - post_ACK = true; - break; - case RXRPC_CALL_CLIENT_SEND_REQUEST: - case RXRPC_CALL_SERVER_RECV_REQUEST: - goto protocol_error_unlock; /* can't occur yet */ - default: - write_unlock_bh(&call->state_lock); - goto discard; /* assume packet left over from earlier phase */ - } - - write_unlock_bh(&call->state_lock); - - /* if all the packets we sent are hard-ACK'd, then we can discard - * whatever we've got left */ - _debug("clear Tx %d", - CIRC_CNT(call->acks_head, call->acks_tail, call->acks_winsz)); - - del_timer_sync(&call->resend_timer); - clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); - clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events); - - if (call->acks_window) - rxrpc_zap_tx_window(call); - - if (post_ACK) { - /* post the final ACK message for userspace to pick up */ - _debug("post ACK"); - skb->mark = RXRPC_SKB_MARK_FINAL_ACK; - sp->call = call; - rxrpc_get_call(call); - spin_lock_bh(&call->lock); - if (rxrpc_queue_rcv_skb(call, skb, true, true) < 0) - BUG(); - spin_unlock_bh(&call->lock); - goto process_further; - } - -discard: - rxrpc_free_skb(skb); - goto process_further; - -protocol_error_unlock: - write_unlock_bh(&call->state_lock); -protocol_error: - rxrpc_free_skb(skb); - _leave(" = -EPROTO"); - return -EPROTO; -} - -/* - * post a message to the socket Rx queue for recvmsg() to pick up - */ -static int rxrpc_post_message(struct rxrpc_call *call, u32 mark, u32 error, - bool fatal) -{ - struct rxrpc_skb_priv *sp; - struct sk_buff *skb; - int ret; - - _enter("{%d,%lx},%u,%u,%d", - call->debug_id, call->flags, mark, error, fatal); - - /* remove timers and things for fatal messages */ - if (fatal) { - del_timer_sync(&call->resend_timer); - del_timer_sync(&call->ack_timer); - clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); - } - - if (mark != RXRPC_SKB_MARK_NEW_CALL && - !test_bit(RXRPC_CALL_HAS_USERID, &call->flags)) { - _leave("[no userid]"); - return 0; - } - - if (!test_bit(RXRPC_CALL_TERMINAL_MSG, &call->flags)) { - skb = alloc_skb(0, GFP_NOFS); - if (!skb) - return -ENOMEM; - - rxrpc_new_skb(skb); - - skb->mark = mark; - - sp = rxrpc_skb(skb); - memset(sp, 0, sizeof(*sp)); - sp->error = error; - sp->call = call; - rxrpc_get_call(call); - - spin_lock_bh(&call->lock); - ret = rxrpc_queue_rcv_skb(call, skb, true, fatal); - spin_unlock_bh(&call->lock); - BUG_ON(ret < 0); - } - - return 0; -} - -/* - * handle background processing of incoming call packets and ACK / abort - * generation - */ -void rxrpc_process_call(struct work_struct *work) -{ - struct rxrpc_call *call = - container_of(work, struct rxrpc_call, processor); - struct rxrpc_wire_header whdr; - struct rxrpc_ackpacket ack; - struct rxrpc_ackinfo ackinfo; - struct msghdr msg; - struct kvec iov[5]; - enum rxrpc_call_event genbit; - unsigned long bits; - __be32 data, pad; - size_t len; - int loop, nbit, ioc, ret, mtu; - u32 serial, abort_code = RX_PROTOCOL_ERROR; - u8 *acks = NULL; - - //printk("\n--------------------\n"); - _enter("{%d,%s,%lx} [%lu]", - call->debug_id, rxrpc_call_states[call->state], call->events, - (jiffies - call->creation_jif) / (HZ / 10)); - - if (test_and_set_bit(RXRPC_CALL_PROC_BUSY, &call->flags)) { - _debug("XXXXXXXXXXXXX RUNNING ON MULTIPLE CPUS XXXXXXXXXXXXX"); - return; - } - - /* there's a good chance we're going to have to send a message, so set - * one up in advance */ - msg.msg_name = &call->conn->trans->peer->srx.transport; - msg.msg_namelen = call->conn->trans->peer->srx.transport_len; - msg.msg_control = NULL; - msg.msg_controllen = 0; - msg.msg_flags = 0; - - whdr.epoch = htonl(call->conn->epoch); - whdr.cid = htonl(call->cid); - whdr.callNumber = htonl(call->call_id); - whdr.seq = 0; - whdr.type = RXRPC_PACKET_TYPE_ACK; - whdr.flags = call->conn->out_clientflag; - whdr.userStatus = 0; - whdr.securityIndex = call->conn->security_ix; - whdr._rsvd = 0; - whdr.serviceId = htons(call->service_id); - - memset(iov, 0, sizeof(iov)); - iov[0].iov_base = &whdr; - iov[0].iov_len = sizeof(whdr); - - /* deal with events of a final nature */ - if (test_bit(RXRPC_CALL_EV_RELEASE, &call->events)) { - rxrpc_release_call(call); - clear_bit(RXRPC_CALL_EV_RELEASE, &call->events); - } - - if (test_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events)) { - int error; - - clear_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events); - clear_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events); - clear_bit(RXRPC_CALL_EV_ABORT, &call->events); - - error = call->conn->trans->peer->net_error; - _debug("post net error %d", error); - - if (rxrpc_post_message(call, RXRPC_SKB_MARK_NET_ERROR, - error, true) < 0) - goto no_mem; - clear_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events); - goto kill_ACKs; - } - - if (test_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events)) { - ASSERTCMP(call->state, >, RXRPC_CALL_COMPLETE); - - clear_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events); - clear_bit(RXRPC_CALL_EV_ABORT, &call->events); - - _debug("post conn abort"); - - if (rxrpc_post_message(call, RXRPC_SKB_MARK_LOCAL_ERROR, - call->conn->error, true) < 0) - goto no_mem; - clear_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events); - goto kill_ACKs; - } - - if (test_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events)) { - whdr.type = RXRPC_PACKET_TYPE_BUSY; - genbit = RXRPC_CALL_EV_REJECT_BUSY; - goto send_message; - } - - if (test_bit(RXRPC_CALL_EV_ABORT, &call->events)) { - ASSERTCMP(call->state, >, RXRPC_CALL_COMPLETE); - - if (rxrpc_post_message(call, RXRPC_SKB_MARK_LOCAL_ERROR, - ECONNABORTED, true) < 0) - goto no_mem; - whdr.type = RXRPC_PACKET_TYPE_ABORT; - data = htonl(call->local_abort); - iov[1].iov_base = &data; - iov[1].iov_len = sizeof(data); - genbit = RXRPC_CALL_EV_ABORT; - goto send_message; - } - - if (test_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events)) { - genbit = RXRPC_CALL_EV_ACK_FINAL; - - ack.bufferSpace = htons(8); - ack.maxSkew = 0; - ack.serial = 0; - ack.reason = RXRPC_ACK_IDLE; - ack.nAcks = 0; - call->ackr_reason = 0; - - spin_lock_bh(&call->lock); - ack.serial = htonl(call->ackr_serial); - ack.previousPacket = htonl(call->ackr_prev_seq); - ack.firstPacket = htonl(call->rx_data_eaten + 1); - spin_unlock_bh(&call->lock); - - pad = 0; - - iov[1].iov_base = &ack; - iov[1].iov_len = sizeof(ack); - iov[2].iov_base = &pad; - iov[2].iov_len = 3; - iov[3].iov_base = &ackinfo; - iov[3].iov_len = sizeof(ackinfo); - goto send_ACK; - } - - if (call->events & ((1 << RXRPC_CALL_EV_RCVD_BUSY) | - (1 << RXRPC_CALL_EV_RCVD_ABORT)) - ) { - u32 mark; - - if (test_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events)) - mark = RXRPC_SKB_MARK_REMOTE_ABORT; - else - mark = RXRPC_SKB_MARK_BUSY; - - _debug("post abort/busy"); - rxrpc_clear_tx_window(call); - if (rxrpc_post_message(call, mark, ECONNABORTED, true) < 0) - goto no_mem; - - clear_bit(RXRPC_CALL_EV_RCVD_BUSY, &call->events); - clear_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events); - goto kill_ACKs; - } - - if (test_and_clear_bit(RXRPC_CALL_EV_RCVD_ACKALL, &call->events)) { - _debug("do implicit ackall"); - rxrpc_clear_tx_window(call); - } - - if (test_bit(RXRPC_CALL_EV_LIFE_TIMER, &call->events)) { - write_lock_bh(&call->state_lock); - if (call->state <= RXRPC_CALL_COMPLETE) { - call->state = RXRPC_CALL_LOCALLY_ABORTED; - call->local_abort = RX_CALL_TIMEOUT; - set_bit(RXRPC_CALL_EV_ABORT, &call->events); - } - write_unlock_bh(&call->state_lock); - - _debug("post timeout"); - if (rxrpc_post_message(call, RXRPC_SKB_MARK_LOCAL_ERROR, - ETIME, true) < 0) - goto no_mem; - - clear_bit(RXRPC_CALL_EV_LIFE_TIMER, &call->events); - goto kill_ACKs; - } - - /* deal with assorted inbound messages */ - if (!skb_queue_empty(&call->rx_queue)) { - switch (rxrpc_process_rx_queue(call, &abort_code)) { - case 0: - case -EAGAIN: - break; - case -ENOMEM: - goto no_mem; - case -EKEYEXPIRED: - case -EKEYREJECTED: - case -EPROTO: - rxrpc_abort_call(call, abort_code); - goto kill_ACKs; - } - } - - /* handle resending */ - if (test_and_clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events)) - rxrpc_resend_timer(call); - if (test_and_clear_bit(RXRPC_CALL_EV_RESEND, &call->events)) - rxrpc_resend(call); - - /* consider sending an ordinary ACK */ - if (test_bit(RXRPC_CALL_EV_ACK, &call->events)) { - _debug("send ACK: window: %d - %d { %lx }", - call->rx_data_eaten, call->ackr_win_top, - call->ackr_window[0]); - - if (call->state > RXRPC_CALL_SERVER_ACK_REQUEST && - call->ackr_reason != RXRPC_ACK_PING_RESPONSE) { - /* ACK by sending reply DATA packet in this state */ - clear_bit(RXRPC_CALL_EV_ACK, &call->events); - goto maybe_reschedule; - } - - genbit = RXRPC_CALL_EV_ACK; - - acks = kzalloc(call->ackr_win_top - call->rx_data_eaten, - GFP_NOFS); - if (!acks) - goto no_mem; - - //hdr.flags = RXRPC_SLOW_START_OK; - ack.bufferSpace = htons(8); - ack.maxSkew = 0; - - spin_lock_bh(&call->lock); - ack.reason = call->ackr_reason; - ack.serial = htonl(call->ackr_serial); - ack.previousPacket = htonl(call->ackr_prev_seq); - ack.firstPacket = htonl(call->rx_data_eaten + 1); - - ack.nAcks = 0; - for (loop = 0; loop < RXRPC_ACKR_WINDOW_ASZ; loop++) { - nbit = loop * BITS_PER_LONG; - for (bits = call->ackr_window[loop]; bits; bits >>= 1 - ) { - _debug("- l=%d n=%d b=%lx", loop, nbit, bits); - if (bits & 1) { - acks[nbit] = RXRPC_ACK_TYPE_ACK; - ack.nAcks = nbit + 1; - } - nbit++; - } - } - call->ackr_reason = 0; - spin_unlock_bh(&call->lock); - - pad = 0; - - iov[1].iov_base = &ack; - iov[1].iov_len = sizeof(ack); - iov[2].iov_base = acks; - iov[2].iov_len = ack.nAcks; - iov[3].iov_base = &pad; - iov[3].iov_len = 3; - iov[4].iov_base = &ackinfo; - iov[4].iov_len = sizeof(ackinfo); - - switch (ack.reason) { - case RXRPC_ACK_REQUESTED: - case RXRPC_ACK_DUPLICATE: - case RXRPC_ACK_OUT_OF_SEQUENCE: - case RXRPC_ACK_EXCEEDS_WINDOW: - case RXRPC_ACK_NOSPACE: - case RXRPC_ACK_PING: - case RXRPC_ACK_PING_RESPONSE: - goto send_ACK_with_skew; - case RXRPC_ACK_DELAY: - case RXRPC_ACK_IDLE: - goto send_ACK; - } - } - - /* handle completion of security negotiations on an incoming - * connection */ - if (test_and_clear_bit(RXRPC_CALL_EV_SECURED, &call->events)) { - _debug("secured"); - spin_lock_bh(&call->lock); - - if (call->state == RXRPC_CALL_SERVER_SECURING) { - _debug("securing"); - write_lock(&call->conn->lock); - if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && - !test_bit(RXRPC_CALL_EV_RELEASE, &call->events)) { - _debug("not released"); - call->state = RXRPC_CALL_SERVER_ACCEPTING; - list_move_tail(&call->accept_link, - &call->socket->acceptq); - } - write_unlock(&call->conn->lock); - read_lock(&call->state_lock); - if (call->state < RXRPC_CALL_COMPLETE) - set_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events); - read_unlock(&call->state_lock); - } - - spin_unlock_bh(&call->lock); - if (!test_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events)) - goto maybe_reschedule; - } - - /* post a notification of an acceptable connection to the app */ - if (test_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events)) { - _debug("post accept"); - if (rxrpc_post_message(call, RXRPC_SKB_MARK_NEW_CALL, - 0, false) < 0) - goto no_mem; - clear_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events); - goto maybe_reschedule; - } - - /* handle incoming call acceptance */ - if (test_and_clear_bit(RXRPC_CALL_EV_ACCEPTED, &call->events)) { - _debug("accepted"); - ASSERTCMP(call->rx_data_post, ==, 0); - call->rx_data_post = 1; - read_lock_bh(&call->state_lock); - if (call->state < RXRPC_CALL_COMPLETE) - set_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events); - read_unlock_bh(&call->state_lock); - } - - /* drain the out of sequence received packet queue into the packet Rx - * queue */ - if (test_and_clear_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events)) { - while (call->rx_data_post == call->rx_first_oos) - if (rxrpc_drain_rx_oos_queue(call) < 0) - break; - goto maybe_reschedule; - } - - /* other events may have been raised since we started checking */ - goto maybe_reschedule; - -send_ACK_with_skew: - ack.maxSkew = htons(atomic_read(&call->conn->hi_serial) - - ntohl(ack.serial)); -send_ACK: - mtu = call->conn->trans->peer->if_mtu; - mtu -= call->conn->trans->peer->hdrsize; - ackinfo.maxMTU = htonl(mtu); - ackinfo.rwind = htonl(rxrpc_rx_window_size); - - /* permit the peer to send us jumbo packets if it wants to */ - ackinfo.rxMTU = htonl(rxrpc_rx_mtu); - ackinfo.jumbo_max = htonl(rxrpc_rx_jumbo_max); - - serial = atomic_inc_return(&call->conn->serial); - whdr.serial = htonl(serial); - _proto("Tx ACK %%%u { m=%hu f=#%u p=#%u s=%%%u r=%s n=%u }", - serial, - ntohs(ack.maxSkew), - ntohl(ack.firstPacket), - ntohl(ack.previousPacket), - ntohl(ack.serial), - rxrpc_acks(ack.reason), - ack.nAcks); - - del_timer_sync(&call->ack_timer); - if (ack.nAcks > 0) - set_bit(RXRPC_CALL_TX_SOFT_ACK, &call->flags); - goto send_message_2; - -send_message: - _debug("send message"); - - serial = atomic_inc_return(&call->conn->serial); - whdr.serial = htonl(serial); - _proto("Tx %s %%%u", rxrpc_pkts[whdr.type], serial); -send_message_2: - - len = iov[0].iov_len; - ioc = 1; - if (iov[4].iov_len) { - ioc = 5; - len += iov[4].iov_len; - len += iov[3].iov_len; - len += iov[2].iov_len; - len += iov[1].iov_len; - } else if (iov[3].iov_len) { - ioc = 4; - len += iov[3].iov_len; - len += iov[2].iov_len; - len += iov[1].iov_len; - } else if (iov[2].iov_len) { - ioc = 3; - len += iov[2].iov_len; - len += iov[1].iov_len; - } else if (iov[1].iov_len) { - ioc = 2; - len += iov[1].iov_len; - } - - ret = kernel_sendmsg(call->conn->trans->local->socket, - &msg, iov, ioc, len); - if (ret < 0) { - _debug("sendmsg failed: %d", ret); - read_lock_bh(&call->state_lock); - if (call->state < RXRPC_CALL_DEAD) - rxrpc_queue_call(call); - read_unlock_bh(&call->state_lock); - goto error; - } - - switch (genbit) { - case RXRPC_CALL_EV_ABORT: - clear_bit(genbit, &call->events); - clear_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events); - goto kill_ACKs; - - case RXRPC_CALL_EV_ACK_FINAL: - write_lock_bh(&call->state_lock); - if (call->state == RXRPC_CALL_CLIENT_FINAL_ACK) - call->state = RXRPC_CALL_COMPLETE; - write_unlock_bh(&call->state_lock); - goto kill_ACKs; - - default: - clear_bit(genbit, &call->events); - switch (call->state) { - case RXRPC_CALL_CLIENT_AWAIT_REPLY: - case RXRPC_CALL_CLIENT_RECV_REPLY: - case RXRPC_CALL_SERVER_RECV_REQUEST: - case RXRPC_CALL_SERVER_ACK_REQUEST: - _debug("start ACK timer"); - rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, - call->ackr_serial, false); - default: - break; - } - goto maybe_reschedule; - } - -kill_ACKs: - del_timer_sync(&call->ack_timer); - if (test_and_clear_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events)) - rxrpc_put_call(call); - clear_bit(RXRPC_CALL_EV_ACK, &call->events); - -maybe_reschedule: - if (call->events || !skb_queue_empty(&call->rx_queue)) { - read_lock_bh(&call->state_lock); - if (call->state < RXRPC_CALL_DEAD) - rxrpc_queue_call(call); - read_unlock_bh(&call->state_lock); - } - - /* don't leave aborted connections on the accept queue */ - if (call->state >= RXRPC_CALL_COMPLETE && - !list_empty(&call->accept_link)) { - _debug("X unlinking once-pending call %p { e=%lx f=%lx c=%x }", - call, call->events, call->flags, call->conn->cid); - - read_lock_bh(&call->state_lock); - if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && - !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) - rxrpc_queue_call(call); - read_unlock_bh(&call->state_lock); - } - -error: - clear_bit(RXRPC_CALL_PROC_BUSY, &call->flags); - kfree(acks); - - /* because we don't want two CPUs both processing the work item for one - * call at the same time, we use a flag to note when it's busy; however - * this means there's a race between clearing the flag and setting the - * work pending bit and the work item being processed again */ - if (call->events && !work_pending(&call->processor)) { - _debug("jumpstart %x", call->conn->cid); - rxrpc_queue_call(call); - } - - _leave(""); - return; - -no_mem: - _debug("out of memory"); - goto maybe_reschedule; -} diff --git a/net/rxrpc/ar-call.c b/net/rxrpc/ar-call.c deleted file mode 100644 index 68125dc..0000000 --- a/net/rxrpc/ar-call.c +++ /dev/null @@ -1,980 +0,0 @@ -/* RxRPC individual remote procedure call handling - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * Maximum lifetime of a call (in jiffies). - */ -unsigned int rxrpc_max_call_lifetime = 60 * HZ; - -/* - * Time till dead call expires after last use (in jiffies). - */ -unsigned int rxrpc_dead_call_expiry = 2 * HZ; - -const char *const rxrpc_call_states[NR__RXRPC_CALL_STATES] = { - [RXRPC_CALL_CLIENT_SEND_REQUEST] = "ClSndReq", - [RXRPC_CALL_CLIENT_AWAIT_REPLY] = "ClAwtRpl", - [RXRPC_CALL_CLIENT_RECV_REPLY] = "ClRcvRpl", - [RXRPC_CALL_CLIENT_FINAL_ACK] = "ClFnlACK", - [RXRPC_CALL_SERVER_SECURING] = "SvSecure", - [RXRPC_CALL_SERVER_ACCEPTING] = "SvAccept", - [RXRPC_CALL_SERVER_RECV_REQUEST] = "SvRcvReq", - [RXRPC_CALL_SERVER_ACK_REQUEST] = "SvAckReq", - [RXRPC_CALL_SERVER_SEND_REPLY] = "SvSndRpl", - [RXRPC_CALL_SERVER_AWAIT_ACK] = "SvAwtACK", - [RXRPC_CALL_COMPLETE] = "Complete", - [RXRPC_CALL_SERVER_BUSY] = "SvBusy ", - [RXRPC_CALL_REMOTELY_ABORTED] = "RmtAbort", - [RXRPC_CALL_LOCALLY_ABORTED] = "LocAbort", - [RXRPC_CALL_NETWORK_ERROR] = "NetError", - [RXRPC_CALL_DEAD] = "Dead ", -}; - -struct kmem_cache *rxrpc_call_jar; -LIST_HEAD(rxrpc_calls); -DEFINE_RWLOCK(rxrpc_call_lock); - -static void rxrpc_destroy_call(struct work_struct *work); -static void rxrpc_call_life_expired(unsigned long _call); -static void rxrpc_dead_call_expired(unsigned long _call); -static void rxrpc_ack_time_expired(unsigned long _call); -static void rxrpc_resend_time_expired(unsigned long _call); - -static DEFINE_SPINLOCK(rxrpc_call_hash_lock); -static DEFINE_HASHTABLE(rxrpc_call_hash, 10); - -/* - * Hash function for rxrpc_call_hash - */ -static unsigned long rxrpc_call_hashfunc( - u8 in_clientflag, - u32 cid, - u32 call_id, - u32 epoch, - u16 service_id, - sa_family_t proto, - void *localptr, - unsigned int addr_size, - const u8 *peer_addr) -{ - const u16 *p; - unsigned int i; - unsigned long key; - - _enter(""); - - key = (unsigned long)localptr; - /* We just want to add up the __be32 values, so forcing the - * cast should be okay. - */ - key += epoch; - key += service_id; - key += call_id; - key += (cid & RXRPC_CIDMASK) >> RXRPC_CIDSHIFT; - key += cid & RXRPC_CHANNELMASK; - key += in_clientflag; - key += proto; - /* Step through the peer address in 16-bit portions for speed */ - for (i = 0, p = (const u16 *)peer_addr; i < addr_size >> 1; i++, p++) - key += *p; - _leave(" key = 0x%lx", key); - return key; -} - -/* - * Add a call to the hashtable - */ -static void rxrpc_call_hash_add(struct rxrpc_call *call) -{ - unsigned long key; - unsigned int addr_size = 0; - - _enter(""); - switch (call->proto) { - case AF_INET: - addr_size = sizeof(call->peer_ip.ipv4_addr); - break; - case AF_INET6: - addr_size = sizeof(call->peer_ip.ipv6_addr); - break; - default: - break; - } - key = rxrpc_call_hashfunc(call->in_clientflag, call->cid, - call->call_id, call->epoch, - call->service_id, call->proto, - call->conn->trans->local, addr_size, - call->peer_ip.ipv6_addr); - /* Store the full key in the call */ - call->hash_key = key; - spin_lock(&rxrpc_call_hash_lock); - hash_add_rcu(rxrpc_call_hash, &call->hash_node, key); - spin_unlock(&rxrpc_call_hash_lock); - _leave(""); -} - -/* - * Remove a call from the hashtable - */ -static void rxrpc_call_hash_del(struct rxrpc_call *call) -{ - _enter(""); - spin_lock(&rxrpc_call_hash_lock); - hash_del_rcu(&call->hash_node); - spin_unlock(&rxrpc_call_hash_lock); - _leave(""); -} - -/* - * Find a call in the hashtable and return it, or NULL if it - * isn't there. - */ -struct rxrpc_call *rxrpc_find_call_hash( - struct rxrpc_host_header *hdr, - void *localptr, - sa_family_t proto, - const void *peer_addr) -{ - unsigned long key; - unsigned int addr_size = 0; - struct rxrpc_call *call = NULL; - struct rxrpc_call *ret = NULL; - u8 in_clientflag = hdr->flags & RXRPC_CLIENT_INITIATED; - - _enter(""); - switch (proto) { - case AF_INET: - addr_size = sizeof(call->peer_ip.ipv4_addr); - break; - case AF_INET6: - addr_size = sizeof(call->peer_ip.ipv6_addr); - break; - default: - break; - } - - key = rxrpc_call_hashfunc(in_clientflag, hdr->cid, hdr->callNumber, - hdr->epoch, hdr->serviceId, - proto, localptr, addr_size, - peer_addr); - hash_for_each_possible_rcu(rxrpc_call_hash, call, hash_node, key) { - if (call->hash_key == key && - call->call_id == hdr->callNumber && - call->cid == hdr->cid && - call->in_clientflag == in_clientflag && - call->service_id == hdr->serviceId && - call->proto == proto && - call->local == localptr && - memcmp(call->peer_ip.ipv6_addr, peer_addr, - addr_size) == 0 && - call->epoch == hdr->epoch) { - ret = call; - break; - } - } - _leave(" = %p", ret); - return ret; -} - -/* - * find an extant server call - * - called in process context with IRQs enabled - */ -struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *rx, - unsigned long user_call_ID) -{ - struct rxrpc_call *call; - struct rb_node *p; - - _enter("%p,%lx", rx, user_call_ID); - - read_lock(&rx->call_lock); - - p = rx->calls.rb_node; - while (p) { - call = rb_entry(p, struct rxrpc_call, sock_node); - - if (user_call_ID < call->user_call_ID) - p = p->rb_left; - else if (user_call_ID > call->user_call_ID) - p = p->rb_right; - else - goto found_extant_call; - } - - read_unlock(&rx->call_lock); - _leave(" = NULL"); - return NULL; - -found_extant_call: - rxrpc_get_call(call); - read_unlock(&rx->call_lock); - _leave(" = %p [%d]", call, atomic_read(&call->usage)); - return call; -} - -/* - * allocate a new call - */ -static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp) -{ - struct rxrpc_call *call; - - call = kmem_cache_zalloc(rxrpc_call_jar, gfp); - if (!call) - return NULL; - - call->acks_winsz = 16; - call->acks_window = kmalloc(call->acks_winsz * sizeof(unsigned long), - gfp); - if (!call->acks_window) { - kmem_cache_free(rxrpc_call_jar, call); - return NULL; - } - - setup_timer(&call->lifetimer, &rxrpc_call_life_expired, - (unsigned long) call); - setup_timer(&call->deadspan, &rxrpc_dead_call_expired, - (unsigned long) call); - setup_timer(&call->ack_timer, &rxrpc_ack_time_expired, - (unsigned long) call); - setup_timer(&call->resend_timer, &rxrpc_resend_time_expired, - (unsigned long) call); - INIT_WORK(&call->destroyer, &rxrpc_destroy_call); - INIT_WORK(&call->processor, &rxrpc_process_call); - INIT_LIST_HEAD(&call->accept_link); - skb_queue_head_init(&call->rx_queue); - skb_queue_head_init(&call->rx_oos_queue); - init_waitqueue_head(&call->tx_waitq); - spin_lock_init(&call->lock); - rwlock_init(&call->state_lock); - atomic_set(&call->usage, 1); - call->debug_id = atomic_inc_return(&rxrpc_debug_id); - call->state = RXRPC_CALL_CLIENT_SEND_REQUEST; - - memset(&call->sock_node, 0xed, sizeof(call->sock_node)); - - call->rx_data_expect = 1; - call->rx_data_eaten = 0; - call->rx_first_oos = 0; - call->ackr_win_top = call->rx_data_eaten + 1 + rxrpc_rx_window_size; - call->creation_jif = jiffies; - return call; -} - -/* - * allocate a new client call and attempt to get a connection slot for it - */ -static struct rxrpc_call *rxrpc_alloc_client_call( - struct rxrpc_sock *rx, - struct rxrpc_transport *trans, - struct rxrpc_conn_bundle *bundle, - gfp_t gfp) -{ - struct rxrpc_call *call; - int ret; - - _enter(""); - - ASSERT(rx != NULL); - ASSERT(trans != NULL); - ASSERT(bundle != NULL); - - call = rxrpc_alloc_call(gfp); - if (!call) - return ERR_PTR(-ENOMEM); - - sock_hold(&rx->sk); - call->socket = rx; - call->rx_data_post = 1; - - ret = rxrpc_connect_call(rx, trans, bundle, call, gfp); - if (ret < 0) { - kmem_cache_free(rxrpc_call_jar, call); - return ERR_PTR(ret); - } - - /* Record copies of information for hashtable lookup */ - call->proto = rx->proto; - call->local = trans->local; - switch (call->proto) { - case AF_INET: - call->peer_ip.ipv4_addr = - trans->peer->srx.transport.sin.sin_addr.s_addr; - break; - case AF_INET6: - memcpy(call->peer_ip.ipv6_addr, - trans->peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8, - sizeof(call->peer_ip.ipv6_addr)); - break; - } - call->epoch = call->conn->epoch; - call->service_id = call->conn->service_id; - call->in_clientflag = call->conn->in_clientflag; - /* Add the new call to the hashtable */ - rxrpc_call_hash_add(call); - - spin_lock(&call->conn->trans->peer->lock); - list_add(&call->error_link, &call->conn->trans->peer->error_targets); - spin_unlock(&call->conn->trans->peer->lock); - - call->lifetimer.expires = jiffies + rxrpc_max_call_lifetime; - add_timer(&call->lifetimer); - - _leave(" = %p", call); - return call; -} - -/* - * set up a call for the given data - * - called in process context with IRQs enabled - */ -struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, - struct rxrpc_transport *trans, - struct rxrpc_conn_bundle *bundle, - unsigned long user_call_ID, - gfp_t gfp) -{ - struct rxrpc_call *call, *xcall; - struct rb_node *parent, **pp; - - _enter("%p,%d,%d,%lx", - rx, trans->debug_id, bundle ? bundle->debug_id : -1, - user_call_ID); - - call = rxrpc_alloc_client_call(rx, trans, bundle, gfp); - if (IS_ERR(call)) { - _leave(" = %ld", PTR_ERR(call)); - return call; - } - - call->user_call_ID = user_call_ID; - __set_bit(RXRPC_CALL_HAS_USERID, &call->flags); - - write_lock(&rx->call_lock); - - pp = &rx->calls.rb_node; - parent = NULL; - while (*pp) { - parent = *pp; - xcall = rb_entry(parent, struct rxrpc_call, sock_node); - - if (user_call_ID < xcall->user_call_ID) - pp = &(*pp)->rb_left; - else if (user_call_ID > xcall->user_call_ID) - pp = &(*pp)->rb_right; - else - goto found_user_ID_now_present; - } - - rxrpc_get_call(call); - - rb_link_node(&call->sock_node, parent, pp); - rb_insert_color(&call->sock_node, &rx->calls); - write_unlock(&rx->call_lock); - - write_lock_bh(&rxrpc_call_lock); - list_add_tail(&call->link, &rxrpc_calls); - write_unlock_bh(&rxrpc_call_lock); - - _net("CALL new %d on CONN %d", call->debug_id, call->conn->debug_id); - - _leave(" = %p [new]", call); - return call; - - /* We unexpectedly found the user ID in the list after taking - * the call_lock. This shouldn't happen unless the user races - * with itself and tries to add the same user ID twice at the - * same time in different threads. - */ -found_user_ID_now_present: - write_unlock(&rx->call_lock); - rxrpc_put_call(call); - _leave(" = -EEXIST [%p]", call); - return ERR_PTR(-EEXIST); -} - -/* - * set up an incoming call - * - called in process context with IRQs enabled - */ -struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, - struct rxrpc_connection *conn, - struct rxrpc_host_header *hdr) -{ - struct rxrpc_call *call, *candidate; - struct rb_node **p, *parent; - u32 call_id; - - _enter(",%d", conn->debug_id); - - ASSERT(rx != NULL); - - candidate = rxrpc_alloc_call(GFP_NOIO); - if (!candidate) - return ERR_PTR(-EBUSY); - - candidate->socket = rx; - candidate->conn = conn; - candidate->cid = hdr->cid; - candidate->call_id = hdr->callNumber; - candidate->channel = hdr->cid & RXRPC_CHANNELMASK; - candidate->rx_data_post = 0; - candidate->state = RXRPC_CALL_SERVER_ACCEPTING; - if (conn->security_ix > 0) - candidate->state = RXRPC_CALL_SERVER_SECURING; - - write_lock_bh(&conn->lock); - - /* set the channel for this call */ - call = conn->channels[candidate->channel]; - _debug("channel[%u] is %p", candidate->channel, call); - if (call && call->call_id == hdr->callNumber) { - /* already set; must've been a duplicate packet */ - _debug("extant call [%d]", call->state); - ASSERTCMP(call->conn, ==, conn); - - read_lock(&call->state_lock); - switch (call->state) { - case RXRPC_CALL_LOCALLY_ABORTED: - if (!test_and_set_bit(RXRPC_CALL_EV_ABORT, &call->events)) - rxrpc_queue_call(call); - case RXRPC_CALL_REMOTELY_ABORTED: - read_unlock(&call->state_lock); - goto aborted_call; - default: - rxrpc_get_call(call); - read_unlock(&call->state_lock); - goto extant_call; - } - } - - if (call) { - /* it seems the channel is still in use from the previous call - * - ditch the old binding if its call is now complete */ - _debug("CALL: %u { %s }", - call->debug_id, rxrpc_call_states[call->state]); - - if (call->state >= RXRPC_CALL_COMPLETE) { - conn->channels[call->channel] = NULL; - } else { - write_unlock_bh(&conn->lock); - kmem_cache_free(rxrpc_call_jar, candidate); - _leave(" = -EBUSY"); - return ERR_PTR(-EBUSY); - } - } - - /* check the call number isn't duplicate */ - _debug("check dup"); - call_id = hdr->callNumber; - p = &conn->calls.rb_node; - parent = NULL; - while (*p) { - parent = *p; - call = rb_entry(parent, struct rxrpc_call, conn_node); - - /* The tree is sorted in order of the __be32 value without - * turning it into host order. - */ - if (call_id < call->call_id) - p = &(*p)->rb_left; - else if (call_id > call->call_id) - p = &(*p)->rb_right; - else - goto old_call; - } - - /* make the call available */ - _debug("new call"); - call = candidate; - candidate = NULL; - rb_link_node(&call->conn_node, parent, p); - rb_insert_color(&call->conn_node, &conn->calls); - conn->channels[call->channel] = call; - sock_hold(&rx->sk); - atomic_inc(&conn->usage); - write_unlock_bh(&conn->lock); - - spin_lock(&conn->trans->peer->lock); - list_add(&call->error_link, &conn->trans->peer->error_targets); - spin_unlock(&conn->trans->peer->lock); - - write_lock_bh(&rxrpc_call_lock); - list_add_tail(&call->link, &rxrpc_calls); - write_unlock_bh(&rxrpc_call_lock); - - /* Record copies of information for hashtable lookup */ - call->proto = rx->proto; - call->local = conn->trans->local; - switch (call->proto) { - case AF_INET: - call->peer_ip.ipv4_addr = - conn->trans->peer->srx.transport.sin.sin_addr.s_addr; - break; - case AF_INET6: - memcpy(call->peer_ip.ipv6_addr, - conn->trans->peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8, - sizeof(call->peer_ip.ipv6_addr)); - break; - default: - break; - } - call->epoch = conn->epoch; - call->service_id = conn->service_id; - call->in_clientflag = conn->in_clientflag; - /* Add the new call to the hashtable */ - rxrpc_call_hash_add(call); - - _net("CALL incoming %d on CONN %d", call->debug_id, call->conn->debug_id); - - call->lifetimer.expires = jiffies + rxrpc_max_call_lifetime; - add_timer(&call->lifetimer); - _leave(" = %p {%d} [new]", call, call->debug_id); - return call; - -extant_call: - write_unlock_bh(&conn->lock); - kmem_cache_free(rxrpc_call_jar, candidate); - _leave(" = %p {%d} [extant]", call, call ? call->debug_id : -1); - return call; - -aborted_call: - write_unlock_bh(&conn->lock); - kmem_cache_free(rxrpc_call_jar, candidate); - _leave(" = -ECONNABORTED"); - return ERR_PTR(-ECONNABORTED); - -old_call: - write_unlock_bh(&conn->lock); - kmem_cache_free(rxrpc_call_jar, candidate); - _leave(" = -ECONNRESET [old]"); - return ERR_PTR(-ECONNRESET); -} - -/* - * detach a call from a socket and set up for release - */ -void rxrpc_release_call(struct rxrpc_call *call) -{ - struct rxrpc_connection *conn = call->conn; - struct rxrpc_sock *rx = call->socket; - - _enter("{%d,%d,%d,%d}", - call->debug_id, atomic_read(&call->usage), - atomic_read(&call->ackr_not_idle), - call->rx_first_oos); - - spin_lock_bh(&call->lock); - if (test_and_set_bit(RXRPC_CALL_RELEASED, &call->flags)) - BUG(); - spin_unlock_bh(&call->lock); - - /* dissociate from the socket - * - the socket's ref on the call is passed to the death timer - */ - _debug("RELEASE CALL %p (%d CONN %p)", call, call->debug_id, conn); - - write_lock_bh(&rx->call_lock); - if (!list_empty(&call->accept_link)) { - _debug("unlinking once-pending call %p { e=%lx f=%lx }", - call, call->events, call->flags); - ASSERT(!test_bit(RXRPC_CALL_HAS_USERID, &call->flags)); - list_del_init(&call->accept_link); - sk_acceptq_removed(&rx->sk); - } else if (test_bit(RXRPC_CALL_HAS_USERID, &call->flags)) { - rb_erase(&call->sock_node, &rx->calls); - memset(&call->sock_node, 0xdd, sizeof(call->sock_node)); - clear_bit(RXRPC_CALL_HAS_USERID, &call->flags); - } - write_unlock_bh(&rx->call_lock); - - /* free up the channel for reuse */ - spin_lock(&conn->trans->client_lock); - write_lock_bh(&conn->lock); - write_lock(&call->state_lock); - - if (conn->channels[call->channel] == call) - conn->channels[call->channel] = NULL; - - if (conn->out_clientflag && conn->bundle) { - conn->avail_calls++; - switch (conn->avail_calls) { - case 1: - list_move_tail(&conn->bundle_link, - &conn->bundle->avail_conns); - case 2 ... RXRPC_MAXCALLS - 1: - ASSERT(conn->channels[0] == NULL || - conn->channels[1] == NULL || - conn->channels[2] == NULL || - conn->channels[3] == NULL); - break; - case RXRPC_MAXCALLS: - list_move_tail(&conn->bundle_link, - &conn->bundle->unused_conns); - ASSERT(conn->channels[0] == NULL && - conn->channels[1] == NULL && - conn->channels[2] == NULL && - conn->channels[3] == NULL); - break; - default: - pr_err("conn->avail_calls=%d\n", conn->avail_calls); - BUG(); - } - } - - spin_unlock(&conn->trans->client_lock); - - if (call->state < RXRPC_CALL_COMPLETE && - call->state != RXRPC_CALL_CLIENT_FINAL_ACK) { - _debug("+++ ABORTING STATE %d +++\n", call->state); - call->state = RXRPC_CALL_LOCALLY_ABORTED; - call->local_abort = RX_CALL_DEAD; - set_bit(RXRPC_CALL_EV_ABORT, &call->events); - rxrpc_queue_call(call); - } - write_unlock(&call->state_lock); - write_unlock_bh(&conn->lock); - - /* clean up the Rx queue */ - if (!skb_queue_empty(&call->rx_queue) || - !skb_queue_empty(&call->rx_oos_queue)) { - struct rxrpc_skb_priv *sp; - struct sk_buff *skb; - - _debug("purge Rx queues"); - - spin_lock_bh(&call->lock); - while ((skb = skb_dequeue(&call->rx_queue)) || - (skb = skb_dequeue(&call->rx_oos_queue))) { - sp = rxrpc_skb(skb); - if (sp->call) { - ASSERTCMP(sp->call, ==, call); - rxrpc_put_call(call); - sp->call = NULL; - } - skb->destructor = NULL; - spin_unlock_bh(&call->lock); - - _debug("- zap %s %%%u #%u", - rxrpc_pkts[sp->hdr.type], - sp->hdr.serial, sp->hdr.seq); - rxrpc_free_skb(skb); - spin_lock_bh(&call->lock); - } - spin_unlock_bh(&call->lock); - - ASSERTCMP(call->state, !=, RXRPC_CALL_COMPLETE); - } - - del_timer_sync(&call->resend_timer); - del_timer_sync(&call->ack_timer); - del_timer_sync(&call->lifetimer); - call->deadspan.expires = jiffies + rxrpc_dead_call_expiry; - add_timer(&call->deadspan); - - _leave(""); -} - -/* - * handle a dead call being ready for reaping - */ -static void rxrpc_dead_call_expired(unsigned long _call) -{ - struct rxrpc_call *call = (struct rxrpc_call *) _call; - - _enter("{%d}", call->debug_id); - - write_lock_bh(&call->state_lock); - call->state = RXRPC_CALL_DEAD; - write_unlock_bh(&call->state_lock); - rxrpc_put_call(call); -} - -/* - * mark a call as to be released, aborting it if it's still in progress - * - called with softirqs disabled - */ -static void rxrpc_mark_call_released(struct rxrpc_call *call) -{ - bool sched; - - write_lock(&call->state_lock); - if (call->state < RXRPC_CALL_DEAD) { - sched = false; - if (call->state < RXRPC_CALL_COMPLETE) { - _debug("abort call %p", call); - call->state = RXRPC_CALL_LOCALLY_ABORTED; - call->local_abort = RX_CALL_DEAD; - if (!test_and_set_bit(RXRPC_CALL_EV_ABORT, &call->events)) - sched = true; - } - if (!test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) - sched = true; - if (sched) - rxrpc_queue_call(call); - } - write_unlock(&call->state_lock); -} - -/* - * release all the calls associated with a socket - */ -void rxrpc_release_calls_on_socket(struct rxrpc_sock *rx) -{ - struct rxrpc_call *call; - struct rb_node *p; - - _enter("%p", rx); - - read_lock_bh(&rx->call_lock); - - /* mark all the calls as no longer wanting incoming packets */ - for (p = rb_first(&rx->calls); p; p = rb_next(p)) { - call = rb_entry(p, struct rxrpc_call, sock_node); - rxrpc_mark_call_released(call); - } - - /* kill the not-yet-accepted incoming calls */ - list_for_each_entry(call, &rx->secureq, accept_link) { - rxrpc_mark_call_released(call); - } - - list_for_each_entry(call, &rx->acceptq, accept_link) { - rxrpc_mark_call_released(call); - } - - read_unlock_bh(&rx->call_lock); - _leave(""); -} - -/* - * release a call - */ -void __rxrpc_put_call(struct rxrpc_call *call) -{ - ASSERT(call != NULL); - - _enter("%p{u=%d}", call, atomic_read(&call->usage)); - - ASSERTCMP(atomic_read(&call->usage), >, 0); - - if (atomic_dec_and_test(&call->usage)) { - _debug("call %d dead", call->debug_id); - ASSERTCMP(call->state, ==, RXRPC_CALL_DEAD); - rxrpc_queue_work(&call->destroyer); - } - _leave(""); -} - -/* - * clean up a call - */ -static void rxrpc_cleanup_call(struct rxrpc_call *call) -{ - _net("DESTROY CALL %d", call->debug_id); - - ASSERT(call->socket); - - memset(&call->sock_node, 0xcd, sizeof(call->sock_node)); - - del_timer_sync(&call->lifetimer); - del_timer_sync(&call->deadspan); - del_timer_sync(&call->ack_timer); - del_timer_sync(&call->resend_timer); - - ASSERT(test_bit(RXRPC_CALL_RELEASED, &call->flags)); - ASSERTCMP(call->events, ==, 0); - if (work_pending(&call->processor)) { - _debug("defer destroy"); - rxrpc_queue_work(&call->destroyer); - return; - } - - if (call->conn) { - spin_lock(&call->conn->trans->peer->lock); - list_del(&call->error_link); - spin_unlock(&call->conn->trans->peer->lock); - - write_lock_bh(&call->conn->lock); - rb_erase(&call->conn_node, &call->conn->calls); - write_unlock_bh(&call->conn->lock); - rxrpc_put_connection(call->conn); - } - - /* Remove the call from the hash */ - rxrpc_call_hash_del(call); - - if (call->acks_window) { - _debug("kill Tx window %d", - CIRC_CNT(call->acks_head, call->acks_tail, - call->acks_winsz)); - smp_mb(); - while (CIRC_CNT(call->acks_head, call->acks_tail, - call->acks_winsz) > 0) { - struct rxrpc_skb_priv *sp; - unsigned long _skb; - - _skb = call->acks_window[call->acks_tail] & ~1; - sp = rxrpc_skb((struct sk_buff *)_skb); - _debug("+++ clear Tx %u", sp->hdr.seq); - rxrpc_free_skb((struct sk_buff *)_skb); - call->acks_tail = - (call->acks_tail + 1) & (call->acks_winsz - 1); - } - - kfree(call->acks_window); - } - - rxrpc_free_skb(call->tx_pending); - - rxrpc_purge_queue(&call->rx_queue); - ASSERT(skb_queue_empty(&call->rx_oos_queue)); - sock_put(&call->socket->sk); - kmem_cache_free(rxrpc_call_jar, call); -} - -/* - * destroy a call - */ -static void rxrpc_destroy_call(struct work_struct *work) -{ - struct rxrpc_call *call = - container_of(work, struct rxrpc_call, destroyer); - - _enter("%p{%d,%d,%p}", - call, atomic_read(&call->usage), call->channel, call->conn); - - ASSERTCMP(call->state, ==, RXRPC_CALL_DEAD); - - write_lock_bh(&rxrpc_call_lock); - list_del_init(&call->link); - write_unlock_bh(&rxrpc_call_lock); - - rxrpc_cleanup_call(call); - _leave(""); -} - -/* - * preemptively destroy all the call records from a transport endpoint rather - * than waiting for them to time out - */ -void __exit rxrpc_destroy_all_calls(void) -{ - struct rxrpc_call *call; - - _enter(""); - write_lock_bh(&rxrpc_call_lock); - - while (!list_empty(&rxrpc_calls)) { - call = list_entry(rxrpc_calls.next, struct rxrpc_call, link); - _debug("Zapping call %p", call); - - list_del_init(&call->link); - - switch (atomic_read(&call->usage)) { - case 0: - ASSERTCMP(call->state, ==, RXRPC_CALL_DEAD); - break; - case 1: - if (del_timer_sync(&call->deadspan) != 0 && - call->state != RXRPC_CALL_DEAD) - rxrpc_dead_call_expired((unsigned long) call); - if (call->state != RXRPC_CALL_DEAD) - break; - default: - pr_err("Call %p still in use (%d,%d,%s,%lx,%lx)!\n", - call, atomic_read(&call->usage), - atomic_read(&call->ackr_not_idle), - rxrpc_call_states[call->state], - call->flags, call->events); - if (!skb_queue_empty(&call->rx_queue)) - pr_err("Rx queue occupied\n"); - if (!skb_queue_empty(&call->rx_oos_queue)) - pr_err("OOS queue occupied\n"); - break; - } - - write_unlock_bh(&rxrpc_call_lock); - cond_resched(); - write_lock_bh(&rxrpc_call_lock); - } - - write_unlock_bh(&rxrpc_call_lock); - _leave(""); -} - -/* - * handle call lifetime being exceeded - */ -static void rxrpc_call_life_expired(unsigned long _call) -{ - struct rxrpc_call *call = (struct rxrpc_call *) _call; - - if (call->state >= RXRPC_CALL_COMPLETE) - return; - - _enter("{%d}", call->debug_id); - read_lock_bh(&call->state_lock); - if (call->state < RXRPC_CALL_COMPLETE) { - set_bit(RXRPC_CALL_EV_LIFE_TIMER, &call->events); - rxrpc_queue_call(call); - } - read_unlock_bh(&call->state_lock); -} - -/* - * handle resend timer expiry - * - may not take call->state_lock as this can deadlock against del_timer_sync() - */ -static void rxrpc_resend_time_expired(unsigned long _call) -{ - struct rxrpc_call *call = (struct rxrpc_call *) _call; - - _enter("{%d}", call->debug_id); - - if (call->state >= RXRPC_CALL_COMPLETE) - return; - - clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); - if (!test_and_set_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events)) - rxrpc_queue_call(call); -} - -/* - * handle ACK timer expiry - */ -static void rxrpc_ack_time_expired(unsigned long _call) -{ - struct rxrpc_call *call = (struct rxrpc_call *) _call; - - _enter("{%d}", call->debug_id); - - if (call->state >= RXRPC_CALL_COMPLETE) - return; - - read_lock_bh(&call->state_lock); - if (call->state < RXRPC_CALL_COMPLETE && - !test_and_set_bit(RXRPC_CALL_EV_ACK, &call->events)) - rxrpc_queue_call(call); - read_unlock_bh(&call->state_lock); -} diff --git a/net/rxrpc/ar-connection.c b/net/rxrpc/ar-connection.c deleted file mode 100644 index 8ecde4b..0000000 --- a/net/rxrpc/ar-connection.c +++ /dev/null @@ -1,912 +0,0 @@ -/* RxRPC virtual connection handler - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * Time till a connection expires after last use (in seconds). - */ -unsigned int rxrpc_connection_expiry = 10 * 60; - -static void rxrpc_connection_reaper(struct work_struct *work); - -LIST_HEAD(rxrpc_connections); -DEFINE_RWLOCK(rxrpc_connection_lock); -static DECLARE_DELAYED_WORK(rxrpc_connection_reap, rxrpc_connection_reaper); - -/* - * allocate a new client connection bundle - */ -static struct rxrpc_conn_bundle *rxrpc_alloc_bundle(gfp_t gfp) -{ - struct rxrpc_conn_bundle *bundle; - - _enter(""); - - bundle = kzalloc(sizeof(struct rxrpc_conn_bundle), gfp); - if (bundle) { - INIT_LIST_HEAD(&bundle->unused_conns); - INIT_LIST_HEAD(&bundle->avail_conns); - INIT_LIST_HEAD(&bundle->busy_conns); - init_waitqueue_head(&bundle->chanwait); - atomic_set(&bundle->usage, 1); - } - - _leave(" = %p", bundle); - return bundle; -} - -/* - * compare bundle parameters with what we're looking for - * - return -ve, 0 or +ve - */ -static inline -int rxrpc_cmp_bundle(const struct rxrpc_conn_bundle *bundle, - struct key *key, u16 service_id) -{ - return (bundle->service_id - service_id) ?: - ((unsigned long)bundle->key - (unsigned long)key); -} - -/* - * get bundle of client connections that a client socket can make use of - */ -struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *rx, - struct rxrpc_transport *trans, - struct key *key, - u16 service_id, - gfp_t gfp) -{ - struct rxrpc_conn_bundle *bundle, *candidate; - struct rb_node *p, *parent, **pp; - - _enter("%p{%x},%x,%hx,", - rx, key_serial(key), trans->debug_id, service_id); - - /* search the extant bundles first for one that matches the specified - * user ID */ - spin_lock(&trans->client_lock); - - p = trans->bundles.rb_node; - while (p) { - bundle = rb_entry(p, struct rxrpc_conn_bundle, node); - - if (rxrpc_cmp_bundle(bundle, key, service_id) < 0) - p = p->rb_left; - else if (rxrpc_cmp_bundle(bundle, key, service_id) > 0) - p = p->rb_right; - else - goto found_extant_bundle; - } - - spin_unlock(&trans->client_lock); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_bundle(gfp); - if (!candidate) { - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } - - candidate->key = key_get(key); - candidate->service_id = service_id; - - spin_lock(&trans->client_lock); - - pp = &trans->bundles.rb_node; - parent = NULL; - while (*pp) { - parent = *pp; - bundle = rb_entry(parent, struct rxrpc_conn_bundle, node); - - if (rxrpc_cmp_bundle(bundle, key, service_id) < 0) - pp = &(*pp)->rb_left; - else if (rxrpc_cmp_bundle(bundle, key, service_id) > 0) - pp = &(*pp)->rb_right; - else - goto found_extant_second; - } - - /* second search also failed; add the new bundle */ - bundle = candidate; - candidate = NULL; - - rb_link_node(&bundle->node, parent, pp); - rb_insert_color(&bundle->node, &trans->bundles); - spin_unlock(&trans->client_lock); - _net("BUNDLE new on trans %d", trans->debug_id); - _leave(" = %p [new]", bundle); - return bundle; - - /* we found the bundle in the list immediately */ -found_extant_bundle: - atomic_inc(&bundle->usage); - spin_unlock(&trans->client_lock); - _net("BUNDLE old on trans %d", trans->debug_id); - _leave(" = %p [extant %d]", bundle, atomic_read(&bundle->usage)); - return bundle; - - /* we found the bundle on the second time through the list */ -found_extant_second: - atomic_inc(&bundle->usage); - spin_unlock(&trans->client_lock); - kfree(candidate); - _net("BUNDLE old2 on trans %d", trans->debug_id); - _leave(" = %p [second %d]", bundle, atomic_read(&bundle->usage)); - return bundle; -} - -/* - * release a bundle - */ -void rxrpc_put_bundle(struct rxrpc_transport *trans, - struct rxrpc_conn_bundle *bundle) -{ - _enter("%p,%p{%d}",trans, bundle, atomic_read(&bundle->usage)); - - if (atomic_dec_and_lock(&bundle->usage, &trans->client_lock)) { - _debug("Destroy bundle"); - rb_erase(&bundle->node, &trans->bundles); - spin_unlock(&trans->client_lock); - ASSERT(list_empty(&bundle->unused_conns)); - ASSERT(list_empty(&bundle->avail_conns)); - ASSERT(list_empty(&bundle->busy_conns)); - ASSERTCMP(bundle->num_conns, ==, 0); - key_put(bundle->key); - kfree(bundle); - } - - _leave(""); -} - -/* - * allocate a new connection - */ -static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) -{ - struct rxrpc_connection *conn; - - _enter(""); - - conn = kzalloc(sizeof(struct rxrpc_connection), gfp); - if (conn) { - INIT_WORK(&conn->processor, &rxrpc_process_connection); - INIT_LIST_HEAD(&conn->bundle_link); - conn->calls = RB_ROOT; - skb_queue_head_init(&conn->rx_queue); - conn->security = &rxrpc_no_security; - rwlock_init(&conn->lock); - spin_lock_init(&conn->state_lock); - atomic_set(&conn->usage, 1); - conn->debug_id = atomic_inc_return(&rxrpc_debug_id); - conn->avail_calls = RXRPC_MAXCALLS; - conn->size_align = 4; - conn->header_size = sizeof(struct rxrpc_wire_header); - } - - _leave(" = %p{%d}", conn, conn ? conn->debug_id : 0); - return conn; -} - -/* - * assign a connection ID to a connection and add it to the transport's - * connection lookup tree - * - called with transport client lock held - */ -static void rxrpc_assign_connection_id(struct rxrpc_connection *conn) -{ - struct rxrpc_connection *xconn; - struct rb_node *parent, **p; - __be32 epoch; - u32 cid; - - _enter(""); - - epoch = conn->epoch; - - write_lock_bh(&conn->trans->conn_lock); - - conn->trans->conn_idcounter += RXRPC_CID_INC; - if (conn->trans->conn_idcounter < RXRPC_CID_INC) - conn->trans->conn_idcounter = RXRPC_CID_INC; - cid = conn->trans->conn_idcounter; - -attempt_insertion: - parent = NULL; - p = &conn->trans->client_conns.rb_node; - - while (*p) { - parent = *p; - xconn = rb_entry(parent, struct rxrpc_connection, node); - - if (epoch < xconn->epoch) - p = &(*p)->rb_left; - else if (epoch > xconn->epoch) - p = &(*p)->rb_right; - else if (cid < xconn->cid) - p = &(*p)->rb_left; - else if (cid > xconn->cid) - p = &(*p)->rb_right; - else - goto id_exists; - } - - /* we've found a suitable hole - arrange for this connection to occupy - * it */ - rb_link_node(&conn->node, parent, p); - rb_insert_color(&conn->node, &conn->trans->client_conns); - - conn->cid = cid; - write_unlock_bh(&conn->trans->conn_lock); - _leave(" [CID %x]", cid); - return; - - /* we found a connection with the proposed ID - walk the tree from that - * point looking for the next unused ID */ -id_exists: - for (;;) { - cid += RXRPC_CID_INC; - if (cid < RXRPC_CID_INC) { - cid = RXRPC_CID_INC; - conn->trans->conn_idcounter = cid; - goto attempt_insertion; - } - - parent = rb_next(parent); - if (!parent) - goto attempt_insertion; - - xconn = rb_entry(parent, struct rxrpc_connection, node); - if (epoch < xconn->epoch || - cid < xconn->cid) - goto attempt_insertion; - } -} - -/* - * add a call to a connection's call-by-ID tree - */ -static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn, - struct rxrpc_call *call) -{ - struct rxrpc_call *xcall; - struct rb_node *parent, **p; - __be32 call_id; - - write_lock_bh(&conn->lock); - - call_id = call->call_id; - p = &conn->calls.rb_node; - parent = NULL; - while (*p) { - parent = *p; - xcall = rb_entry(parent, struct rxrpc_call, conn_node); - - if (call_id < xcall->call_id) - p = &(*p)->rb_left; - else if (call_id > xcall->call_id) - p = &(*p)->rb_right; - else - BUG(); - } - - rb_link_node(&call->conn_node, parent, p); - rb_insert_color(&call->conn_node, &conn->calls); - - write_unlock_bh(&conn->lock); -} - -/* - * connect a call on an exclusive connection - */ -static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, - struct rxrpc_transport *trans, - u16 service_id, - struct rxrpc_call *call, - gfp_t gfp) -{ - struct rxrpc_connection *conn; - int chan, ret; - - _enter(""); - - conn = rx->conn; - if (!conn) { - /* not yet present - create a candidate for a new connection - * and then redo the check */ - conn = rxrpc_alloc_connection(gfp); - if (!conn) { - _leave(" = -ENOMEM"); - return -ENOMEM; - } - - conn->trans = trans; - conn->bundle = NULL; - conn->service_id = service_id; - conn->epoch = rxrpc_epoch; - conn->in_clientflag = 0; - conn->out_clientflag = RXRPC_CLIENT_INITIATED; - conn->cid = 0; - conn->state = RXRPC_CONN_CLIENT; - conn->avail_calls = RXRPC_MAXCALLS - 1; - conn->security_level = rx->min_sec_level; - conn->key = key_get(rx->key); - - ret = rxrpc_init_client_conn_security(conn); - if (ret < 0) { - key_put(conn->key); - kfree(conn); - _leave(" = %d [key]", ret); - return ret; - } - - write_lock_bh(&rxrpc_connection_lock); - list_add_tail(&conn->link, &rxrpc_connections); - write_unlock_bh(&rxrpc_connection_lock); - - spin_lock(&trans->client_lock); - atomic_inc(&trans->usage); - - _net("CONNECT EXCL new %d on TRANS %d", - conn->debug_id, conn->trans->debug_id); - - rxrpc_assign_connection_id(conn); - rx->conn = conn; - } else { - spin_lock(&trans->client_lock); - } - - /* we've got a connection with a free channel and we can now attach the - * call to it - * - we're holding the transport's client lock - * - we're holding a reference on the connection - */ - for (chan = 0; chan < RXRPC_MAXCALLS; chan++) - if (!conn->channels[chan]) - goto found_channel; - goto no_free_channels; - -found_channel: - atomic_inc(&conn->usage); - conn->channels[chan] = call; - call->conn = conn; - call->channel = chan; - call->cid = conn->cid | chan; - call->call_id = ++conn->call_counter; - - _net("CONNECT client on conn %d chan %d as call %x", - conn->debug_id, chan, call->call_id); - - spin_unlock(&trans->client_lock); - - rxrpc_add_call_ID_to_conn(conn, call); - _leave(" = 0"); - return 0; - -no_free_channels: - spin_unlock(&trans->client_lock); - _leave(" = -ENOSR"); - return -ENOSR; -} - -/* - * find a connection for a call - * - called in process context with IRQs enabled - */ -int rxrpc_connect_call(struct rxrpc_sock *rx, - struct rxrpc_transport *trans, - struct rxrpc_conn_bundle *bundle, - struct rxrpc_call *call, - gfp_t gfp) -{ - struct rxrpc_connection *conn, *candidate; - int chan, ret; - - DECLARE_WAITQUEUE(myself, current); - - _enter("%p,%lx,", rx, call->user_call_ID); - - if (test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags)) - return rxrpc_connect_exclusive(rx, trans, bundle->service_id, - call, gfp); - - spin_lock(&trans->client_lock); - for (;;) { - /* see if the bundle has a call slot available */ - if (!list_empty(&bundle->avail_conns)) { - _debug("avail"); - conn = list_entry(bundle->avail_conns.next, - struct rxrpc_connection, - bundle_link); - if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) { - list_del_init(&conn->bundle_link); - bundle->num_conns--; - continue; - } - if (--conn->avail_calls == 0) - list_move(&conn->bundle_link, - &bundle->busy_conns); - ASSERTCMP(conn->avail_calls, <, RXRPC_MAXCALLS); - ASSERT(conn->channels[0] == NULL || - conn->channels[1] == NULL || - conn->channels[2] == NULL || - conn->channels[3] == NULL); - atomic_inc(&conn->usage); - break; - } - - if (!list_empty(&bundle->unused_conns)) { - _debug("unused"); - conn = list_entry(bundle->unused_conns.next, - struct rxrpc_connection, - bundle_link); - if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) { - list_del_init(&conn->bundle_link); - bundle->num_conns--; - continue; - } - ASSERTCMP(conn->avail_calls, ==, RXRPC_MAXCALLS); - conn->avail_calls = RXRPC_MAXCALLS - 1; - ASSERT(conn->channels[0] == NULL && - conn->channels[1] == NULL && - conn->channels[2] == NULL && - conn->channels[3] == NULL); - atomic_inc(&conn->usage); - list_move(&conn->bundle_link, &bundle->avail_conns); - break; - } - - /* need to allocate a new connection */ - _debug("get new conn [%d]", bundle->num_conns); - - spin_unlock(&trans->client_lock); - - if (signal_pending(current)) - goto interrupted; - - if (bundle->num_conns >= 20) { - _debug("too many conns"); - - if (!gfpflags_allow_blocking(gfp)) { - _leave(" = -EAGAIN"); - return -EAGAIN; - } - - add_wait_queue(&bundle->chanwait, &myself); - for (;;) { - set_current_state(TASK_INTERRUPTIBLE); - if (bundle->num_conns < 20 || - !list_empty(&bundle->unused_conns) || - !list_empty(&bundle->avail_conns)) - break; - if (signal_pending(current)) - goto interrupted_dequeue; - schedule(); - } - remove_wait_queue(&bundle->chanwait, &myself); - __set_current_state(TASK_RUNNING); - spin_lock(&trans->client_lock); - continue; - } - - /* not yet present - create a candidate for a new connection and then - * redo the check */ - candidate = rxrpc_alloc_connection(gfp); - if (!candidate) { - _leave(" = -ENOMEM"); - return -ENOMEM; - } - - candidate->trans = trans; - candidate->bundle = bundle; - candidate->service_id = bundle->service_id; - candidate->epoch = rxrpc_epoch; - candidate->in_clientflag = 0; - candidate->out_clientflag = RXRPC_CLIENT_INITIATED; - candidate->cid = 0; - candidate->state = RXRPC_CONN_CLIENT; - candidate->avail_calls = RXRPC_MAXCALLS; - candidate->security_level = rx->min_sec_level; - candidate->key = key_get(bundle->key); - - ret = rxrpc_init_client_conn_security(candidate); - if (ret < 0) { - key_put(candidate->key); - kfree(candidate); - _leave(" = %d [key]", ret); - return ret; - } - - write_lock_bh(&rxrpc_connection_lock); - list_add_tail(&candidate->link, &rxrpc_connections); - write_unlock_bh(&rxrpc_connection_lock); - - spin_lock(&trans->client_lock); - - list_add(&candidate->bundle_link, &bundle->unused_conns); - bundle->num_conns++; - atomic_inc(&bundle->usage); - atomic_inc(&trans->usage); - - _net("CONNECT new %d on TRANS %d", - candidate->debug_id, candidate->trans->debug_id); - - rxrpc_assign_connection_id(candidate); - candidate->security->prime_packet_security(candidate); - - /* leave the candidate lurking in zombie mode attached to the - * bundle until we're ready for it */ - rxrpc_put_connection(candidate); - candidate = NULL; - } - - /* we've got a connection with a free channel and we can now attach the - * call to it - * - we're holding the transport's client lock - * - we're holding a reference on the connection - * - we're holding a reference on the bundle - */ - for (chan = 0; chan < RXRPC_MAXCALLS; chan++) - if (!conn->channels[chan]) - goto found_channel; - ASSERT(conn->channels[0] == NULL || - conn->channels[1] == NULL || - conn->channels[2] == NULL || - conn->channels[3] == NULL); - BUG(); - -found_channel: - conn->channels[chan] = call; - call->conn = conn; - call->channel = chan; - call->cid = conn->cid | chan; - call->call_id = ++conn->call_counter; - - _net("CONNECT client on conn %d chan %d as call %x", - conn->debug_id, chan, call->call_id); - - ASSERTCMP(conn->avail_calls, <, RXRPC_MAXCALLS); - spin_unlock(&trans->client_lock); - - rxrpc_add_call_ID_to_conn(conn, call); - - _leave(" = 0"); - return 0; - -interrupted_dequeue: - remove_wait_queue(&bundle->chanwait, &myself); - __set_current_state(TASK_RUNNING); -interrupted: - _leave(" = -ERESTARTSYS"); - return -ERESTARTSYS; -} - -/* - * get a record of an incoming connection - */ -struct rxrpc_connection * -rxrpc_incoming_connection(struct rxrpc_transport *trans, - struct rxrpc_host_header *hdr) -{ - struct rxrpc_connection *conn, *candidate = NULL; - struct rb_node *p, **pp; - const char *new = "old"; - __be32 epoch; - u32 cid; - - _enter(""); - - ASSERT(hdr->flags & RXRPC_CLIENT_INITIATED); - - epoch = hdr->epoch; - cid = hdr->cid & RXRPC_CIDMASK; - - /* search the connection list first */ - read_lock_bh(&trans->conn_lock); - - p = trans->server_conns.rb_node; - while (p) { - conn = rb_entry(p, struct rxrpc_connection, node); - - _debug("maybe %x", conn->cid); - - if (epoch < conn->epoch) - p = p->rb_left; - else if (epoch > conn->epoch) - p = p->rb_right; - else if (cid < conn->cid) - p = p->rb_left; - else if (cid > conn->cid) - p = p->rb_right; - else - goto found_extant_connection; - } - read_unlock_bh(&trans->conn_lock); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_connection(GFP_NOIO); - if (!candidate) { - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } - - candidate->trans = trans; - candidate->epoch = hdr->epoch; - candidate->cid = hdr->cid & RXRPC_CIDMASK; - candidate->service_id = hdr->serviceId; - candidate->security_ix = hdr->securityIndex; - candidate->in_clientflag = RXRPC_CLIENT_INITIATED; - candidate->out_clientflag = 0; - candidate->state = RXRPC_CONN_SERVER; - if (candidate->service_id) - candidate->state = RXRPC_CONN_SERVER_UNSECURED; - - write_lock_bh(&trans->conn_lock); - - pp = &trans->server_conns.rb_node; - p = NULL; - while (*pp) { - p = *pp; - conn = rb_entry(p, struct rxrpc_connection, node); - - if (epoch < conn->epoch) - pp = &(*pp)->rb_left; - else if (epoch > conn->epoch) - pp = &(*pp)->rb_right; - else if (cid < conn->cid) - pp = &(*pp)->rb_left; - else if (cid > conn->cid) - pp = &(*pp)->rb_right; - else - goto found_extant_second; - } - - /* we can now add the new candidate to the list */ - conn = candidate; - candidate = NULL; - rb_link_node(&conn->node, p, pp); - rb_insert_color(&conn->node, &trans->server_conns); - atomic_inc(&conn->trans->usage); - - write_unlock_bh(&trans->conn_lock); - - write_lock_bh(&rxrpc_connection_lock); - list_add_tail(&conn->link, &rxrpc_connections); - write_unlock_bh(&rxrpc_connection_lock); - - new = "new"; - -success: - _net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->cid); - - _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); - return conn; - - /* we found the connection in the list immediately */ -found_extant_connection: - if (hdr->securityIndex != conn->security_ix) { - read_unlock_bh(&trans->conn_lock); - goto security_mismatch; - } - atomic_inc(&conn->usage); - read_unlock_bh(&trans->conn_lock); - goto success; - - /* we found the connection on the second time through the list */ -found_extant_second: - if (hdr->securityIndex != conn->security_ix) { - write_unlock_bh(&trans->conn_lock); - goto security_mismatch; - } - atomic_inc(&conn->usage); - write_unlock_bh(&trans->conn_lock); - kfree(candidate); - goto success; - -security_mismatch: - kfree(candidate); - _leave(" = -EKEYREJECTED"); - return ERR_PTR(-EKEYREJECTED); -} - -/* - * find a connection based on transport and RxRPC connection ID for an incoming - * packet - */ -struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans, - struct rxrpc_host_header *hdr) -{ - struct rxrpc_connection *conn; - struct rb_node *p; - u32 epoch, cid; - - _enter(",{%x,%x}", hdr->cid, hdr->flags); - - read_lock_bh(&trans->conn_lock); - - cid = hdr->cid & RXRPC_CIDMASK; - epoch = hdr->epoch; - - if (hdr->flags & RXRPC_CLIENT_INITIATED) - p = trans->server_conns.rb_node; - else - p = trans->client_conns.rb_node; - - while (p) { - conn = rb_entry(p, struct rxrpc_connection, node); - - _debug("maybe %x", conn->cid); - - if (epoch < conn->epoch) - p = p->rb_left; - else if (epoch > conn->epoch) - p = p->rb_right; - else if (cid < conn->cid) - p = p->rb_left; - else if (cid > conn->cid) - p = p->rb_right; - else - goto found; - } - - read_unlock_bh(&trans->conn_lock); - _leave(" = NULL"); - return NULL; - -found: - atomic_inc(&conn->usage); - read_unlock_bh(&trans->conn_lock); - _leave(" = %p", conn); - return conn; -} - -/* - * release a virtual connection - */ -void rxrpc_put_connection(struct rxrpc_connection *conn) -{ - _enter("%p{u=%d,d=%d}", - conn, atomic_read(&conn->usage), conn->debug_id); - - ASSERTCMP(atomic_read(&conn->usage), >, 0); - - conn->put_time = ktime_get_seconds(); - if (atomic_dec_and_test(&conn->usage)) { - _debug("zombie"); - rxrpc_queue_delayed_work(&rxrpc_connection_reap, 0); - } - - _leave(""); -} - -/* - * destroy a virtual connection - */ -static void rxrpc_destroy_connection(struct rxrpc_connection *conn) -{ - _enter("%p{%d}", conn, atomic_read(&conn->usage)); - - ASSERTCMP(atomic_read(&conn->usage), ==, 0); - - _net("DESTROY CONN %d", conn->debug_id); - - if (conn->bundle) - rxrpc_put_bundle(conn->trans, conn->bundle); - - ASSERT(RB_EMPTY_ROOT(&conn->calls)); - rxrpc_purge_queue(&conn->rx_queue); - - conn->security->clear(conn); - key_put(conn->key); - key_put(conn->server_key); - - rxrpc_put_transport(conn->trans); - kfree(conn); - _leave(""); -} - -/* - * reap dead connections - */ -static void rxrpc_connection_reaper(struct work_struct *work) -{ - struct rxrpc_connection *conn, *_p; - unsigned long now, earliest, reap_time; - - LIST_HEAD(graveyard); - - _enter(""); - - now = ktime_get_seconds(); - earliest = ULONG_MAX; - - write_lock_bh(&rxrpc_connection_lock); - list_for_each_entry_safe(conn, _p, &rxrpc_connections, link) { - _debug("reap CONN %d { u=%d,t=%ld }", - conn->debug_id, atomic_read(&conn->usage), - (long) now - (long) conn->put_time); - - if (likely(atomic_read(&conn->usage) > 0)) - continue; - - spin_lock(&conn->trans->client_lock); - write_lock(&conn->trans->conn_lock); - reap_time = conn->put_time + rxrpc_connection_expiry; - - if (atomic_read(&conn->usage) > 0) { - ; - } else if (reap_time <= now) { - list_move_tail(&conn->link, &graveyard); - if (conn->out_clientflag) - rb_erase(&conn->node, - &conn->trans->client_conns); - else - rb_erase(&conn->node, - &conn->trans->server_conns); - if (conn->bundle) { - list_del_init(&conn->bundle_link); - conn->bundle->num_conns--; - } - - } else if (reap_time < earliest) { - earliest = reap_time; - } - - write_unlock(&conn->trans->conn_lock); - spin_unlock(&conn->trans->client_lock); - } - write_unlock_bh(&rxrpc_connection_lock); - - if (earliest != ULONG_MAX) { - _debug("reschedule reaper %ld", (long) earliest - now); - ASSERTCMP(earliest, >, now); - rxrpc_queue_delayed_work(&rxrpc_connection_reap, - (earliest - now) * HZ); - } - - /* then destroy all those pulled out */ - while (!list_empty(&graveyard)) { - conn = list_entry(graveyard.next, struct rxrpc_connection, - link); - list_del_init(&conn->link); - - ASSERTCMP(atomic_read(&conn->usage), ==, 0); - rxrpc_destroy_connection(conn); - } - - _leave(""); -} - -/* - * preemptively destroy all the connection records rather than waiting for them - * to time out - */ -void __exit rxrpc_destroy_all_connections(void) -{ - _enter(""); - - rxrpc_connection_expiry = 0; - cancel_delayed_work(&rxrpc_connection_reap); - rxrpc_queue_delayed_work(&rxrpc_connection_reap, 0); - - _leave(""); -} diff --git a/net/rxrpc/ar-connevent.c b/net/rxrpc/ar-connevent.c deleted file mode 100644 index 8bdd692..0000000 --- a/net/rxrpc/ar-connevent.c +++ /dev/null @@ -1,403 +0,0 @@ -/* connection-level event handling - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * pass a connection-level abort onto all calls on that connection - */ -static void rxrpc_abort_calls(struct rxrpc_connection *conn, int state, - u32 abort_code) -{ - struct rxrpc_call *call; - struct rb_node *p; - - _enter("{%d},%x", conn->debug_id, abort_code); - - read_lock_bh(&conn->lock); - - for (p = rb_first(&conn->calls); p; p = rb_next(p)) { - call = rb_entry(p, struct rxrpc_call, conn_node); - write_lock(&call->state_lock); - if (call->state <= RXRPC_CALL_COMPLETE) { - call->state = state; - if (state == RXRPC_CALL_LOCALLY_ABORTED) { - call->local_abort = conn->local_abort; - set_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events); - } else { - call->remote_abort = conn->remote_abort; - set_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events); - } - rxrpc_queue_call(call); - } - write_unlock(&call->state_lock); - } - - read_unlock_bh(&conn->lock); - _leave(""); -} - -/* - * generate a connection-level abort - */ -static int rxrpc_abort_connection(struct rxrpc_connection *conn, - u32 error, u32 abort_code) -{ - struct rxrpc_wire_header whdr; - struct msghdr msg; - struct kvec iov[2]; - __be32 word; - size_t len; - u32 serial; - int ret; - - _enter("%d,,%u,%u", conn->debug_id, error, abort_code); - - /* generate a connection-level abort */ - spin_lock_bh(&conn->state_lock); - if (conn->state < RXRPC_CONN_REMOTELY_ABORTED) { - conn->state = RXRPC_CONN_LOCALLY_ABORTED; - conn->error = error; - spin_unlock_bh(&conn->state_lock); - } else { - spin_unlock_bh(&conn->state_lock); - _leave(" = 0 [already dead]"); - return 0; - } - - rxrpc_abort_calls(conn, RXRPC_CALL_LOCALLY_ABORTED, abort_code); - - msg.msg_name = &conn->trans->peer->srx.transport; - msg.msg_namelen = conn->trans->peer->srx.transport_len; - msg.msg_control = NULL; - msg.msg_controllen = 0; - msg.msg_flags = 0; - - whdr.epoch = htonl(conn->epoch); - whdr.cid = htonl(conn->cid); - whdr.callNumber = 0; - whdr.seq = 0; - whdr.type = RXRPC_PACKET_TYPE_ABORT; - whdr.flags = conn->out_clientflag; - whdr.userStatus = 0; - whdr.securityIndex = conn->security_ix; - whdr._rsvd = 0; - whdr.serviceId = htons(conn->service_id); - - word = htonl(conn->local_abort); - - iov[0].iov_base = &whdr; - iov[0].iov_len = sizeof(whdr); - iov[1].iov_base = &word; - iov[1].iov_len = sizeof(word); - - len = iov[0].iov_len + iov[1].iov_len; - - serial = atomic_inc_return(&conn->serial); - whdr.serial = htonl(serial); - _proto("Tx CONN ABORT %%%u { %d }", serial, conn->local_abort); - - ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 2, len); - if (ret < 0) { - _debug("sendmsg failed: %d", ret); - return -EAGAIN; - } - - _leave(" = 0"); - return 0; -} - -/* - * mark a call as being on a now-secured channel - * - must be called with softirqs disabled - */ -static void rxrpc_call_is_secure(struct rxrpc_call *call) -{ - _enter("%p", call); - if (call) { - read_lock(&call->state_lock); - if (call->state < RXRPC_CALL_COMPLETE && - !test_and_set_bit(RXRPC_CALL_EV_SECURED, &call->events)) - rxrpc_queue_call(call); - read_unlock(&call->state_lock); - } -} - -/* - * connection-level Rx packet processor - */ -static int rxrpc_process_event(struct rxrpc_connection *conn, - struct sk_buff *skb, - u32 *_abort_code) -{ - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - __be32 wtmp; - u32 abort_code; - int loop, ret; - - if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) { - kleave(" = -ECONNABORTED [%u]", conn->state); - return -ECONNABORTED; - } - - _enter("{%d},{%u,%%%u},", conn->debug_id, sp->hdr.type, sp->hdr.serial); - - switch (sp->hdr.type) { - case RXRPC_PACKET_TYPE_ABORT: - if (skb_copy_bits(skb, 0, &wtmp, sizeof(wtmp)) < 0) - return -EPROTO; - abort_code = ntohl(wtmp); - _proto("Rx ABORT %%%u { ac=%d }", sp->hdr.serial, abort_code); - - conn->state = RXRPC_CONN_REMOTELY_ABORTED; - rxrpc_abort_calls(conn, RXRPC_CALL_REMOTELY_ABORTED, - abort_code); - return -ECONNABORTED; - - case RXRPC_PACKET_TYPE_CHALLENGE: - return conn->security->respond_to_challenge(conn, skb, - _abort_code); - - case RXRPC_PACKET_TYPE_RESPONSE: - ret = conn->security->verify_response(conn, skb, _abort_code); - if (ret < 0) - return ret; - - ret = conn->security->init_connection_security(conn); - if (ret < 0) - return ret; - - conn->security->prime_packet_security(conn); - read_lock_bh(&conn->lock); - spin_lock(&conn->state_lock); - - if (conn->state == RXRPC_CONN_SERVER_CHALLENGING) { - conn->state = RXRPC_CONN_SERVER; - for (loop = 0; loop < RXRPC_MAXCALLS; loop++) - rxrpc_call_is_secure(conn->channels[loop]); - } - - spin_unlock(&conn->state_lock); - read_unlock_bh(&conn->lock); - return 0; - - default: - _leave(" = -EPROTO [%u]", sp->hdr.type); - return -EPROTO; - } -} - -/* - * set up security and issue a challenge - */ -static void rxrpc_secure_connection(struct rxrpc_connection *conn) -{ - u32 abort_code; - int ret; - - _enter("{%d}", conn->debug_id); - - ASSERT(conn->security_ix != 0); - - if (!conn->key) { - _debug("set up security"); - ret = rxrpc_init_server_conn_security(conn); - switch (ret) { - case 0: - break; - case -ENOENT: - abort_code = RX_CALL_DEAD; - goto abort; - default: - abort_code = RXKADNOAUTH; - goto abort; - } - } - - if (conn->security->issue_challenge(conn) < 0) { - abort_code = RX_CALL_DEAD; - ret = -ENOMEM; - goto abort; - } - - _leave(""); - return; - -abort: - _debug("abort %d, %d", ret, abort_code); - rxrpc_abort_connection(conn, -ret, abort_code); - _leave(" [aborted]"); -} - -/* - * connection-level event processor - */ -void rxrpc_process_connection(struct work_struct *work) -{ - struct rxrpc_connection *conn = - container_of(work, struct rxrpc_connection, processor); - struct sk_buff *skb; - u32 abort_code = RX_PROTOCOL_ERROR; - int ret; - - _enter("{%d}", conn->debug_id); - - atomic_inc(&conn->usage); - - if (test_and_clear_bit(RXRPC_CONN_CHALLENGE, &conn->events)) { - rxrpc_secure_connection(conn); - rxrpc_put_connection(conn); - } - - /* go through the conn-level event packets, releasing the ref on this - * connection that each one has when we've finished with it */ - while ((skb = skb_dequeue(&conn->rx_queue))) { - ret = rxrpc_process_event(conn, skb, &abort_code); - switch (ret) { - case -EPROTO: - case -EKEYEXPIRED: - case -EKEYREJECTED: - goto protocol_error; - case -EAGAIN: - goto requeue_and_leave; - case -ECONNABORTED: - default: - rxrpc_put_connection(conn); - rxrpc_free_skb(skb); - break; - } - } - -out: - rxrpc_put_connection(conn); - _leave(""); - return; - -requeue_and_leave: - skb_queue_head(&conn->rx_queue, skb); - goto out; - -protocol_error: - if (rxrpc_abort_connection(conn, -ret, abort_code) < 0) - goto requeue_and_leave; - rxrpc_put_connection(conn); - rxrpc_free_skb(skb); - _leave(" [EPROTO]"); - goto out; -} - -/* - * put a packet up for transport-level abort - */ -void rxrpc_reject_packet(struct rxrpc_local *local, struct sk_buff *skb) -{ - CHECK_SLAB_OKAY(&local->usage); - - if (!atomic_inc_not_zero(&local->usage)) { - printk("resurrected on reject\n"); - BUG(); - } - - skb_queue_tail(&local->reject_queue, skb); - rxrpc_queue_work(&local->rejecter); -} - -/* - * reject packets through the local endpoint - */ -void rxrpc_reject_packets(struct work_struct *work) -{ - union { - struct sockaddr sa; - struct sockaddr_in sin; - } sa; - struct rxrpc_skb_priv *sp; - struct rxrpc_wire_header whdr; - struct rxrpc_local *local; - struct sk_buff *skb; - struct msghdr msg; - struct kvec iov[2]; - size_t size; - __be32 code; - - local = container_of(work, struct rxrpc_local, rejecter); - rxrpc_get_local(local); - - _enter("%d", local->debug_id); - - iov[0].iov_base = &whdr; - iov[0].iov_len = sizeof(whdr); - iov[1].iov_base = &code; - iov[1].iov_len = sizeof(code); - size = sizeof(whdr) + sizeof(code); - - msg.msg_name = &sa; - msg.msg_control = NULL; - msg.msg_controllen = 0; - msg.msg_flags = 0; - - memset(&sa, 0, sizeof(sa)); - sa.sa.sa_family = local->srx.transport.family; - switch (sa.sa.sa_family) { - case AF_INET: - msg.msg_namelen = sizeof(sa.sin); - break; - default: - msg.msg_namelen = 0; - break; - } - - memset(&whdr, 0, sizeof(whdr)); - whdr.type = RXRPC_PACKET_TYPE_ABORT; - - while ((skb = skb_dequeue(&local->reject_queue))) { - sp = rxrpc_skb(skb); - switch (sa.sa.sa_family) { - case AF_INET: - sa.sin.sin_port = udp_hdr(skb)->source; - sa.sin.sin_addr.s_addr = ip_hdr(skb)->saddr; - code = htonl(skb->priority); - - whdr.epoch = htonl(sp->hdr.epoch); - whdr.cid = htonl(sp->hdr.cid); - whdr.callNumber = htonl(sp->hdr.callNumber); - whdr.serviceId = htons(sp->hdr.serviceId); - whdr.flags = sp->hdr.flags; - whdr.flags ^= RXRPC_CLIENT_INITIATED; - whdr.flags &= RXRPC_CLIENT_INITIATED; - - kernel_sendmsg(local->socket, &msg, iov, 2, size); - break; - - default: - break; - } - - rxrpc_free_skb(skb); - rxrpc_put_local(local); - } - - rxrpc_put_local(local); - _leave(""); -} diff --git a/net/rxrpc/ar-error.c b/net/rxrpc/ar-error.c deleted file mode 100644 index 3e82d6f..0000000 --- a/net/rxrpc/ar-error.c +++ /dev/null @@ -1,230 +0,0 @@ -/* Error message handling (ICMP) - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * handle an error received on the local endpoint - */ -void rxrpc_UDP_error_report(struct sock *sk) -{ - struct sock_exterr_skb *serr; - struct rxrpc_transport *trans; - struct rxrpc_local *local = sk->sk_user_data; - struct rxrpc_peer *peer; - struct sk_buff *skb; - __be32 addr; - __be16 port; - - _enter("%p{%d}", sk, local->debug_id); - - skb = sock_dequeue_err_skb(sk); - if (!skb) { - _leave("UDP socket errqueue empty"); - return; - } - serr = SKB_EXT_ERR(skb); - if (!skb->len && serr->ee.ee_origin == SO_EE_ORIGIN_TIMESTAMPING) { - _leave("UDP empty message"); - kfree_skb(skb); - return; - } - - rxrpc_new_skb(skb); - - addr = *(__be32 *)(skb_network_header(skb) + serr->addr_offset); - port = serr->port; - - _net("Rx UDP Error from %pI4:%hu", &addr, ntohs(port)); - _debug("Msg l:%d d:%d", skb->len, skb->data_len); - - peer = rxrpc_find_peer(local, addr, port); - if (IS_ERR(peer)) { - rxrpc_free_skb(skb); - _leave(" [no peer]"); - return; - } - - trans = rxrpc_find_transport(local, peer); - if (!trans) { - rxrpc_put_peer(peer); - rxrpc_free_skb(skb); - _leave(" [no trans]"); - return; - } - - if (serr->ee.ee_origin == SO_EE_ORIGIN_ICMP && - serr->ee.ee_type == ICMP_DEST_UNREACH && - serr->ee.ee_code == ICMP_FRAG_NEEDED - ) { - u32 mtu = serr->ee.ee_info; - - _net("Rx Received ICMP Fragmentation Needed (%d)", mtu); - - /* wind down the local interface MTU */ - if (mtu > 0 && peer->if_mtu == 65535 && mtu < peer->if_mtu) { - peer->if_mtu = mtu; - _net("I/F MTU %u", mtu); - } - - if (mtu == 0) { - /* they didn't give us a size, estimate one */ - mtu = peer->if_mtu; - if (mtu > 1500) { - mtu >>= 1; - if (mtu < 1500) - mtu = 1500; - } else { - mtu -= 100; - if (mtu < peer->hdrsize) - mtu = peer->hdrsize + 4; - } - } - - if (mtu < peer->mtu) { - spin_lock_bh(&peer->lock); - peer->mtu = mtu; - peer->maxdata = peer->mtu - peer->hdrsize; - spin_unlock_bh(&peer->lock); - _net("Net MTU %u (maxdata %u)", - peer->mtu, peer->maxdata); - } - } - - rxrpc_put_peer(peer); - - /* pass the transport ref to error_handler to release */ - skb_queue_tail(&trans->error_queue, skb); - rxrpc_queue_work(&trans->error_handler); - _leave(""); -} - -/* - * deal with UDP error messages - */ -void rxrpc_UDP_error_handler(struct work_struct *work) -{ - struct sock_extended_err *ee; - struct sock_exterr_skb *serr; - struct rxrpc_transport *trans = - container_of(work, struct rxrpc_transport, error_handler); - struct sk_buff *skb; - int err; - - _enter(""); - - skb = skb_dequeue(&trans->error_queue); - if (!skb) - return; - - serr = SKB_EXT_ERR(skb); - ee = &serr->ee; - - _net("Rx Error o=%d t=%d c=%d e=%d", - ee->ee_origin, ee->ee_type, ee->ee_code, ee->ee_errno); - - err = ee->ee_errno; - - switch (ee->ee_origin) { - case SO_EE_ORIGIN_ICMP: - switch (ee->ee_type) { - case ICMP_DEST_UNREACH: - switch (ee->ee_code) { - case ICMP_NET_UNREACH: - _net("Rx Received ICMP Network Unreachable"); - break; - case ICMP_HOST_UNREACH: - _net("Rx Received ICMP Host Unreachable"); - break; - case ICMP_PORT_UNREACH: - _net("Rx Received ICMP Port Unreachable"); - break; - case ICMP_NET_UNKNOWN: - _net("Rx Received ICMP Unknown Network"); - break; - case ICMP_HOST_UNKNOWN: - _net("Rx Received ICMP Unknown Host"); - break; - default: - _net("Rx Received ICMP DestUnreach code=%u", - ee->ee_code); - break; - } - break; - - case ICMP_TIME_EXCEEDED: - _net("Rx Received ICMP TTL Exceeded"); - break; - - default: - _proto("Rx Received ICMP error { type=%u code=%u }", - ee->ee_type, ee->ee_code); - break; - } - break; - - case SO_EE_ORIGIN_LOCAL: - _proto("Rx Received local error { error=%d }", - ee->ee_errno); - break; - - case SO_EE_ORIGIN_NONE: - case SO_EE_ORIGIN_ICMP6: - default: - _proto("Rx Received error report { orig=%u }", - ee->ee_origin); - break; - } - - /* terminate all the affected calls if there's an unrecoverable - * error */ - if (err) { - struct rxrpc_call *call, *_n; - - _debug("ISSUE ERROR %d", err); - - spin_lock_bh(&trans->peer->lock); - trans->peer->net_error = err; - - list_for_each_entry_safe(call, _n, &trans->peer->error_targets, - error_link) { - write_lock(&call->state_lock); - if (call->state != RXRPC_CALL_COMPLETE && - call->state < RXRPC_CALL_NETWORK_ERROR) { - call->state = RXRPC_CALL_NETWORK_ERROR; - set_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events); - rxrpc_queue_call(call); - } - write_unlock(&call->state_lock); - list_del_init(&call->error_link); - } - - spin_unlock_bh(&trans->peer->lock); - } - - if (!skb_queue_empty(&trans->error_queue)) - rxrpc_queue_work(&trans->error_handler); - - rxrpc_free_skb(skb); - rxrpc_put_transport(trans); - _leave(""); -} diff --git a/net/rxrpc/ar-input.c b/net/rxrpc/ar-input.c deleted file mode 100644 index e0815a0..0000000 --- a/net/rxrpc/ar-input.c +++ /dev/null @@ -1,800 +0,0 @@ -/* RxRPC packet reception - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * queue a packet for recvmsg to pass to userspace - * - the caller must hold a lock on call->lock - * - must not be called with interrupts disabled (sk_filter() disables BH's) - * - eats the packet whether successful or not - * - there must be just one reference to the packet, which the caller passes to - * this function - */ -int rxrpc_queue_rcv_skb(struct rxrpc_call *call, struct sk_buff *skb, - bool force, bool terminal) -{ - struct rxrpc_skb_priv *sp; - struct rxrpc_sock *rx = call->socket; - struct sock *sk; - int ret; - - _enter(",,%d,%d", force, terminal); - - ASSERT(!irqs_disabled()); - - sp = rxrpc_skb(skb); - ASSERTCMP(sp->call, ==, call); - - /* if we've already posted the terminal message for a call, then we - * don't post any more */ - if (test_bit(RXRPC_CALL_TERMINAL_MSG, &call->flags)) { - _debug("already terminated"); - ASSERTCMP(call->state, >=, RXRPC_CALL_COMPLETE); - skb->destructor = NULL; - sp->call = NULL; - rxrpc_put_call(call); - rxrpc_free_skb(skb); - return 0; - } - - sk = &rx->sk; - - if (!force) { - /* cast skb->rcvbuf to unsigned... It's pointless, but - * reduces number of warnings when compiling with -W - * --ANK */ -// ret = -ENOBUFS; -// if (atomic_read(&sk->sk_rmem_alloc) + skb->truesize >= -// (unsigned int) sk->sk_rcvbuf) -// goto out; - - ret = sk_filter(sk, skb); - if (ret < 0) - goto out; - } - - spin_lock_bh(&sk->sk_receive_queue.lock); - if (!test_bit(RXRPC_CALL_TERMINAL_MSG, &call->flags) && - !test_bit(RXRPC_CALL_RELEASED, &call->flags) && - call->socket->sk.sk_state != RXRPC_CLOSE) { - skb->destructor = rxrpc_packet_destructor; - skb->dev = NULL; - skb->sk = sk; - atomic_add(skb->truesize, &sk->sk_rmem_alloc); - - if (terminal) { - _debug("<<<< TERMINAL MESSAGE >>>>"); - set_bit(RXRPC_CALL_TERMINAL_MSG, &call->flags); - } - - /* allow interception by a kernel service */ - if (rx->interceptor) { - rx->interceptor(sk, call->user_call_ID, skb); - spin_unlock_bh(&sk->sk_receive_queue.lock); - } else { - _net("post skb %p", skb); - __skb_queue_tail(&sk->sk_receive_queue, skb); - spin_unlock_bh(&sk->sk_receive_queue.lock); - - if (!sock_flag(sk, SOCK_DEAD)) - sk->sk_data_ready(sk); - } - skb = NULL; - } else { - spin_unlock_bh(&sk->sk_receive_queue.lock); - } - ret = 0; - -out: - /* release the socket buffer */ - if (skb) { - skb->destructor = NULL; - sp->call = NULL; - rxrpc_put_call(call); - rxrpc_free_skb(skb); - } - - _leave(" = %d", ret); - return ret; -} - -/* - * process a DATA packet, posting the packet to the appropriate queue - * - eats the packet if successful - */ -static int rxrpc_fast_process_data(struct rxrpc_call *call, - struct sk_buff *skb, u32 seq) -{ - struct rxrpc_skb_priv *sp; - bool terminal; - int ret, ackbit, ack; - - _enter("{%u,%u},,{%u}", call->rx_data_post, call->rx_first_oos, seq); - - sp = rxrpc_skb(skb); - ASSERTCMP(sp->call, ==, NULL); - - spin_lock(&call->lock); - - if (call->state > RXRPC_CALL_COMPLETE) - goto discard; - - ASSERTCMP(call->rx_data_expect, >=, call->rx_data_post); - ASSERTCMP(call->rx_data_post, >=, call->rx_data_recv); - ASSERTCMP(call->rx_data_recv, >=, call->rx_data_eaten); - - if (seq < call->rx_data_post) { - _debug("dup #%u [-%u]", seq, call->rx_data_post); - ack = RXRPC_ACK_DUPLICATE; - ret = -ENOBUFS; - goto discard_and_ack; - } - - /* we may already have the packet in the out of sequence queue */ - ackbit = seq - (call->rx_data_eaten + 1); - ASSERTCMP(ackbit, >=, 0); - if (__test_and_set_bit(ackbit, call->ackr_window)) { - _debug("dup oos #%u [%u,%u]", - seq, call->rx_data_eaten, call->rx_data_post); - ack = RXRPC_ACK_DUPLICATE; - goto discard_and_ack; - } - - if (seq >= call->ackr_win_top) { - _debug("exceed #%u [%u]", seq, call->ackr_win_top); - __clear_bit(ackbit, call->ackr_window); - ack = RXRPC_ACK_EXCEEDS_WINDOW; - goto discard_and_ack; - } - - if (seq == call->rx_data_expect) { - clear_bit(RXRPC_CALL_EXPECT_OOS, &call->flags); - call->rx_data_expect++; - } else if (seq > call->rx_data_expect) { - _debug("oos #%u [%u]", seq, call->rx_data_expect); - call->rx_data_expect = seq + 1; - if (test_and_set_bit(RXRPC_CALL_EXPECT_OOS, &call->flags)) { - ack = RXRPC_ACK_OUT_OF_SEQUENCE; - goto enqueue_and_ack; - } - goto enqueue_packet; - } - - if (seq != call->rx_data_post) { - _debug("ahead #%u [%u]", seq, call->rx_data_post); - goto enqueue_packet; - } - - if (test_bit(RXRPC_CALL_RCVD_LAST, &call->flags)) - goto protocol_error; - - /* if the packet need security things doing to it, then it goes down - * the slow path */ - if (call->conn->security_ix) - goto enqueue_packet; - - sp->call = call; - rxrpc_get_call(call); - terminal = ((sp->hdr.flags & RXRPC_LAST_PACKET) && - !(sp->hdr.flags & RXRPC_CLIENT_INITIATED)); - ret = rxrpc_queue_rcv_skb(call, skb, false, terminal); - if (ret < 0) { - if (ret == -ENOMEM || ret == -ENOBUFS) { - __clear_bit(ackbit, call->ackr_window); - ack = RXRPC_ACK_NOSPACE; - goto discard_and_ack; - } - goto out; - } - - skb = NULL; - - _debug("post #%u", seq); - ASSERTCMP(call->rx_data_post, ==, seq); - call->rx_data_post++; - - if (sp->hdr.flags & RXRPC_LAST_PACKET) - set_bit(RXRPC_CALL_RCVD_LAST, &call->flags); - - /* if we've reached an out of sequence packet then we need to drain - * that queue into the socket Rx queue now */ - if (call->rx_data_post == call->rx_first_oos) { - _debug("drain rx oos now"); - read_lock(&call->state_lock); - if (call->state < RXRPC_CALL_COMPLETE && - !test_and_set_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events)) - rxrpc_queue_call(call); - read_unlock(&call->state_lock); - } - - spin_unlock(&call->lock); - atomic_inc(&call->ackr_not_idle); - rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, sp->hdr.serial, false); - _leave(" = 0 [posted]"); - return 0; - -protocol_error: - ret = -EBADMSG; -out: - spin_unlock(&call->lock); - _leave(" = %d", ret); - return ret; - -discard_and_ack: - _debug("discard and ACK packet %p", skb); - __rxrpc_propose_ACK(call, ack, sp->hdr.serial, true); -discard: - spin_unlock(&call->lock); - rxrpc_free_skb(skb); - _leave(" = 0 [discarded]"); - return 0; - -enqueue_and_ack: - __rxrpc_propose_ACK(call, ack, sp->hdr.serial, true); -enqueue_packet: - _net("defer skb %p", skb); - spin_unlock(&call->lock); - skb_queue_tail(&call->rx_queue, skb); - atomic_inc(&call->ackr_not_idle); - read_lock(&call->state_lock); - if (call->state < RXRPC_CALL_DEAD) - rxrpc_queue_call(call); - read_unlock(&call->state_lock); - _leave(" = 0 [queued]"); - return 0; -} - -/* - * assume an implicit ACKALL of the transmission phase of a client socket upon - * reception of the first reply packet - */ -static void rxrpc_assume_implicit_ackall(struct rxrpc_call *call, u32 serial) -{ - write_lock_bh(&call->state_lock); - - switch (call->state) { - case RXRPC_CALL_CLIENT_AWAIT_REPLY: - call->state = RXRPC_CALL_CLIENT_RECV_REPLY; - call->acks_latest = serial; - - _debug("implicit ACKALL %%%u", call->acks_latest); - set_bit(RXRPC_CALL_EV_RCVD_ACKALL, &call->events); - write_unlock_bh(&call->state_lock); - - if (try_to_del_timer_sync(&call->resend_timer) >= 0) { - clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events); - clear_bit(RXRPC_CALL_EV_RESEND, &call->events); - clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); - } - break; - - default: - write_unlock_bh(&call->state_lock); - break; - } -} - -/* - * post an incoming packet to the nominated call to deal with - * - must get rid of the sk_buff, either by freeing it or by queuing it - */ -void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb) -{ - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - __be32 wtmp; - u32 hi_serial, abort_code; - - _enter("%p,%p", call, skb); - - ASSERT(!irqs_disabled()); - -#if 0 // INJECT RX ERROR - if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA) { - static int skip = 0; - if (++skip == 3) { - printk("DROPPED 3RD PACKET!!!!!!!!!!!!!\n"); - skip = 0; - goto free_packet; - } - } -#endif - - /* track the latest serial number on this connection for ACK packet - * information */ - hi_serial = atomic_read(&call->conn->hi_serial); - while (sp->hdr.serial > hi_serial) - hi_serial = atomic_cmpxchg(&call->conn->hi_serial, hi_serial, - sp->hdr.serial); - - /* request ACK generation for any ACK or DATA packet that requests - * it */ - if (sp->hdr.flags & RXRPC_REQUEST_ACK) { - _proto("ACK Requested on %%%u", sp->hdr.serial); - rxrpc_propose_ACK(call, RXRPC_ACK_REQUESTED, sp->hdr.serial, false); - } - - switch (sp->hdr.type) { - case RXRPC_PACKET_TYPE_ABORT: - _debug("abort"); - - if (skb_copy_bits(skb, 0, &wtmp, sizeof(wtmp)) < 0) - goto protocol_error; - - abort_code = ntohl(wtmp); - _proto("Rx ABORT %%%u { %x }", sp->hdr.serial, abort_code); - - write_lock_bh(&call->state_lock); - if (call->state < RXRPC_CALL_COMPLETE) { - call->state = RXRPC_CALL_REMOTELY_ABORTED; - call->remote_abort = abort_code; - set_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events); - rxrpc_queue_call(call); - } - goto free_packet_unlock; - - case RXRPC_PACKET_TYPE_BUSY: - _proto("Rx BUSY %%%u", sp->hdr.serial); - - if (call->conn->out_clientflag) - goto protocol_error; - - write_lock_bh(&call->state_lock); - switch (call->state) { - case RXRPC_CALL_CLIENT_SEND_REQUEST: - call->state = RXRPC_CALL_SERVER_BUSY; - set_bit(RXRPC_CALL_EV_RCVD_BUSY, &call->events); - rxrpc_queue_call(call); - case RXRPC_CALL_SERVER_BUSY: - goto free_packet_unlock; - default: - goto protocol_error_locked; - } - - default: - _proto("Rx %s %%%u", rxrpc_pkts[sp->hdr.type], sp->hdr.serial); - goto protocol_error; - - case RXRPC_PACKET_TYPE_DATA: - _proto("Rx DATA %%%u { #%u }", sp->hdr.serial, sp->hdr.seq); - - if (sp->hdr.seq == 0) - goto protocol_error; - - call->ackr_prev_seq = sp->hdr.seq; - - /* received data implicitly ACKs all of the request packets we - * sent when we're acting as a client */ - if (call->state == RXRPC_CALL_CLIENT_AWAIT_REPLY) - rxrpc_assume_implicit_ackall(call, sp->hdr.serial); - - switch (rxrpc_fast_process_data(call, skb, sp->hdr.seq)) { - case 0: - skb = NULL; - goto done; - - default: - BUG(); - - /* data packet received beyond the last packet */ - case -EBADMSG: - goto protocol_error; - } - - case RXRPC_PACKET_TYPE_ACKALL: - case RXRPC_PACKET_TYPE_ACK: - /* ACK processing is done in process context */ - read_lock_bh(&call->state_lock); - if (call->state < RXRPC_CALL_DEAD) { - skb_queue_tail(&call->rx_queue, skb); - rxrpc_queue_call(call); - skb = NULL; - } - read_unlock_bh(&call->state_lock); - goto free_packet; - } - -protocol_error: - _debug("protocol error"); - write_lock_bh(&call->state_lock); -protocol_error_locked: - if (call->state <= RXRPC_CALL_COMPLETE) { - call->state = RXRPC_CALL_LOCALLY_ABORTED; - call->local_abort = RX_PROTOCOL_ERROR; - set_bit(RXRPC_CALL_EV_ABORT, &call->events); - rxrpc_queue_call(call); - } -free_packet_unlock: - write_unlock_bh(&call->state_lock); -free_packet: - rxrpc_free_skb(skb); -done: - _leave(""); -} - -/* - * split up a jumbo data packet - */ -static void rxrpc_process_jumbo_packet(struct rxrpc_call *call, - struct sk_buff *jumbo) -{ - struct rxrpc_jumbo_header jhdr; - struct rxrpc_skb_priv *sp; - struct sk_buff *part; - - _enter(",{%u,%u}", jumbo->data_len, jumbo->len); - - sp = rxrpc_skb(jumbo); - - do { - sp->hdr.flags &= ~RXRPC_JUMBO_PACKET; - - /* make a clone to represent the first subpacket in what's left - * of the jumbo packet */ - part = skb_clone(jumbo, GFP_ATOMIC); - if (!part) { - /* simply ditch the tail in the event of ENOMEM */ - pskb_trim(jumbo, RXRPC_JUMBO_DATALEN); - break; - } - rxrpc_new_skb(part); - - pskb_trim(part, RXRPC_JUMBO_DATALEN); - - if (!pskb_pull(jumbo, RXRPC_JUMBO_DATALEN)) - goto protocol_error; - - if (skb_copy_bits(jumbo, 0, &jhdr, sizeof(jhdr)) < 0) - goto protocol_error; - if (!pskb_pull(jumbo, sizeof(jhdr))) - BUG(); - - sp->hdr.seq += 1; - sp->hdr.serial += 1; - sp->hdr.flags = jhdr.flags; - sp->hdr._rsvd = jhdr._rsvd; - - _proto("Rx DATA Jumbo %%%u", sp->hdr.serial - 1); - - rxrpc_fast_process_packet(call, part); - part = NULL; - - } while (sp->hdr.flags & RXRPC_JUMBO_PACKET); - - rxrpc_fast_process_packet(call, jumbo); - _leave(""); - return; - -protocol_error: - _debug("protocol error"); - rxrpc_free_skb(part); - rxrpc_free_skb(jumbo); - write_lock_bh(&call->state_lock); - if (call->state <= RXRPC_CALL_COMPLETE) { - call->state = RXRPC_CALL_LOCALLY_ABORTED; - call->local_abort = RX_PROTOCOL_ERROR; - set_bit(RXRPC_CALL_EV_ABORT, &call->events); - rxrpc_queue_call(call); - } - write_unlock_bh(&call->state_lock); - _leave(""); -} - -/* - * post an incoming packet to the appropriate call/socket to deal with - * - must get rid of the sk_buff, either by freeing it or by queuing it - */ -static void rxrpc_post_packet_to_call(struct rxrpc_call *call, - struct sk_buff *skb) -{ - struct rxrpc_skb_priv *sp; - - _enter("%p,%p", call, skb); - - sp = rxrpc_skb(skb); - - _debug("extant call [%d]", call->state); - - read_lock(&call->state_lock); - switch (call->state) { - case RXRPC_CALL_LOCALLY_ABORTED: - if (!test_and_set_bit(RXRPC_CALL_EV_ABORT, &call->events)) { - rxrpc_queue_call(call); - goto free_unlock; - } - case RXRPC_CALL_REMOTELY_ABORTED: - case RXRPC_CALL_NETWORK_ERROR: - case RXRPC_CALL_DEAD: - goto dead_call; - case RXRPC_CALL_COMPLETE: - case RXRPC_CALL_CLIENT_FINAL_ACK: - /* complete server call */ - if (call->conn->in_clientflag) - goto dead_call; - /* resend last packet of a completed call */ - _debug("final ack again"); - rxrpc_get_call(call); - set_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events); - rxrpc_queue_call(call); - goto free_unlock; - default: - break; - } - - read_unlock(&call->state_lock); - rxrpc_get_call(call); - - if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA && - sp->hdr.flags & RXRPC_JUMBO_PACKET) - rxrpc_process_jumbo_packet(call, skb); - else - rxrpc_fast_process_packet(call, skb); - - rxrpc_put_call(call); - goto done; - -dead_call: - if (sp->hdr.type != RXRPC_PACKET_TYPE_ABORT) { - skb->priority = RX_CALL_DEAD; - rxrpc_reject_packet(call->conn->trans->local, skb); - goto unlock; - } -free_unlock: - rxrpc_free_skb(skb); -unlock: - read_unlock(&call->state_lock); -done: - _leave(""); -} - -/* - * post connection-level events to the connection - * - this includes challenges, responses and some aborts - */ -static void rxrpc_post_packet_to_conn(struct rxrpc_connection *conn, - struct sk_buff *skb) -{ - _enter("%p,%p", conn, skb); - - atomic_inc(&conn->usage); - skb_queue_tail(&conn->rx_queue, skb); - rxrpc_queue_conn(conn); -} - -/* - * post endpoint-level events to the local endpoint - * - this includes debug and version messages - */ -static void rxrpc_post_packet_to_local(struct rxrpc_local *local, - struct sk_buff *skb) -{ - _enter("%p,%p", local, skb); - - atomic_inc(&local->usage); - skb_queue_tail(&local->event_queue, skb); - rxrpc_queue_work(&local->event_processor); -} - -/* - * Extract the wire header from a packet and translate the byte order. - */ -static noinline -int rxrpc_extract_header(struct rxrpc_skb_priv *sp, struct sk_buff *skb) -{ - struct rxrpc_wire_header whdr; - - /* dig out the RxRPC connection details */ - if (skb_copy_bits(skb, 0, &whdr, sizeof(whdr)) < 0) - return -EBADMSG; - if (!pskb_pull(skb, sizeof(whdr))) - BUG(); - - memset(sp, 0, sizeof(*sp)); - sp->hdr.epoch = ntohl(whdr.epoch); - sp->hdr.cid = ntohl(whdr.cid); - sp->hdr.callNumber = ntohl(whdr.callNumber); - sp->hdr.seq = ntohl(whdr.seq); - sp->hdr.serial = ntohl(whdr.serial); - sp->hdr.flags = whdr.flags; - sp->hdr.type = whdr.type; - sp->hdr.userStatus = whdr.userStatus; - sp->hdr.securityIndex = whdr.securityIndex; - sp->hdr._rsvd = ntohs(whdr._rsvd); - sp->hdr.serviceId = ntohs(whdr.serviceId); - return 0; -} - -static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local, - struct sk_buff *skb, - struct rxrpc_skb_priv *sp) -{ - struct rxrpc_peer *peer; - struct rxrpc_transport *trans; - struct rxrpc_connection *conn; - - peer = rxrpc_find_peer(local, ip_hdr(skb)->saddr, - udp_hdr(skb)->source); - if (IS_ERR(peer)) - goto cant_find_conn; - - trans = rxrpc_find_transport(local, peer); - rxrpc_put_peer(peer); - if (!trans) - goto cant_find_conn; - - conn = rxrpc_find_connection(trans, &sp->hdr); - rxrpc_put_transport(trans); - if (!conn) - goto cant_find_conn; - - return conn; -cant_find_conn: - return NULL; -} - -/* - * handle data received on the local endpoint - * - may be called in interrupt context - */ -void rxrpc_data_ready(struct sock *sk) -{ - struct rxrpc_skb_priv *sp; - struct rxrpc_local *local; - struct sk_buff *skb; - int ret; - - _enter("%p", sk); - - ASSERT(!irqs_disabled()); - - read_lock_bh(&rxrpc_local_lock); - local = sk->sk_user_data; - if (local && atomic_read(&local->usage) > 0) - rxrpc_get_local(local); - else - local = NULL; - read_unlock_bh(&rxrpc_local_lock); - if (!local) { - _leave(" [local dead]"); - return; - } - - skb = skb_recv_datagram(sk, 0, 1, &ret); - if (!skb) { - rxrpc_put_local(local); - if (ret == -EAGAIN) - return; - _debug("UDP socket error %d", ret); - return; - } - - rxrpc_new_skb(skb); - - _net("recv skb %p", skb); - - /* we'll probably need to checksum it (didn't call sock_recvmsg) */ - if (skb_checksum_complete(skb)) { - rxrpc_free_skb(skb); - rxrpc_put_local(local); - __UDP_INC_STATS(&init_net, UDP_MIB_INERRORS, 0); - _leave(" [CSUM failed]"); - return; - } - - __UDP_INC_STATS(&init_net, UDP_MIB_INDATAGRAMS, 0); - - /* The socket buffer we have is owned by UDP, with UDP's data all over - * it, but we really want our own data there. - */ - skb_orphan(skb); - sp = rxrpc_skb(skb); - - _net("Rx UDP packet from %08x:%04hu", - ntohl(ip_hdr(skb)->saddr), ntohs(udp_hdr(skb)->source)); - - /* dig out the RxRPC connection details */ - if (rxrpc_extract_header(sp, skb) < 0) - goto bad_message; - - _net("Rx RxRPC %s ep=%x call=%x:%x", - sp->hdr.flags & RXRPC_CLIENT_INITIATED ? "ToServer" : "ToClient", - sp->hdr.epoch, sp->hdr.cid, sp->hdr.callNumber); - - if (sp->hdr.type >= RXRPC_N_PACKET_TYPES || - !((RXRPC_SUPPORTED_PACKET_TYPES >> sp->hdr.type) & 1)) { - _proto("Rx Bad Packet Type %u", sp->hdr.type); - goto bad_message; - } - - if (sp->hdr.type == RXRPC_PACKET_TYPE_VERSION) { - rxrpc_post_packet_to_local(local, skb); - goto out; - } - - if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA && - (sp->hdr.callNumber == 0 || sp->hdr.seq == 0)) - goto bad_message; - - if (sp->hdr.callNumber == 0) { - /* This is a connection-level packet. These should be - * fairly rare, so the extra overhead of looking them up the - * old-fashioned way doesn't really hurt */ - struct rxrpc_connection *conn; - - conn = rxrpc_conn_from_local(local, skb, sp); - if (!conn) - goto cant_route_call; - - _debug("CONN %p {%d}", conn, conn->debug_id); - rxrpc_post_packet_to_conn(conn, skb); - rxrpc_put_connection(conn); - } else { - struct rxrpc_call *call; - - call = rxrpc_find_call_hash(&sp->hdr, local, - AF_INET, &ip_hdr(skb)->saddr); - if (call) - rxrpc_post_packet_to_call(call, skb); - else - goto cant_route_call; - } - -out: - rxrpc_put_local(local); - return; - -cant_route_call: - _debug("can't route call"); - if (sp->hdr.flags & RXRPC_CLIENT_INITIATED && - sp->hdr.type == RXRPC_PACKET_TYPE_DATA) { - if (sp->hdr.seq == 1) { - _debug("first packet"); - skb_queue_tail(&local->accept_queue, skb); - rxrpc_queue_work(&local->acceptor); - rxrpc_put_local(local); - _leave(" [incoming]"); - return; - } - skb->priority = RX_INVALID_OPERATION; - } else { - skb->priority = RX_CALL_DEAD; - } - - if (sp->hdr.type != RXRPC_PACKET_TYPE_ABORT) { - _debug("reject type %d",sp->hdr.type); - rxrpc_reject_packet(local, skb); - } - rxrpc_put_local(local); - _leave(" [no call]"); - return; - -bad_message: - skb->priority = RX_PROTOCOL_ERROR; - rxrpc_reject_packet(local, skb); - rxrpc_put_local(local); - _leave(" [badmsg]"); -} diff --git a/net/rxrpc/ar-key.c b/net/rxrpc/ar-key.c deleted file mode 100644 index 4ad56fa..0000000 --- a/net/rxrpc/ar-key.c +++ /dev/null @@ -1,1237 +0,0 @@ -/* RxRPC key management - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * RxRPC keys should have a description of describing their purpose: - * "afs@CAMBRIDGE.REDHAT.COM> - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -static int rxrpc_vet_description_s(const char *); -static int rxrpc_preparse(struct key_preparsed_payload *); -static int rxrpc_preparse_s(struct key_preparsed_payload *); -static void rxrpc_free_preparse(struct key_preparsed_payload *); -static void rxrpc_free_preparse_s(struct key_preparsed_payload *); -static void rxrpc_destroy(struct key *); -static void rxrpc_destroy_s(struct key *); -static void rxrpc_describe(const struct key *, struct seq_file *); -static long rxrpc_read(const struct key *, char __user *, size_t); - -/* - * rxrpc defined keys take an arbitrary string as the description and an - * arbitrary blob of data as the payload - */ -struct key_type key_type_rxrpc = { - .name = "rxrpc", - .preparse = rxrpc_preparse, - .free_preparse = rxrpc_free_preparse, - .instantiate = generic_key_instantiate, - .destroy = rxrpc_destroy, - .describe = rxrpc_describe, - .read = rxrpc_read, -}; -EXPORT_SYMBOL(key_type_rxrpc); - -/* - * rxrpc server defined keys take ":" as the - * description and an 8-byte decryption key as the payload - */ -struct key_type key_type_rxrpc_s = { - .name = "rxrpc_s", - .vet_description = rxrpc_vet_description_s, - .preparse = rxrpc_preparse_s, - .free_preparse = rxrpc_free_preparse_s, - .instantiate = generic_key_instantiate, - .destroy = rxrpc_destroy_s, - .describe = rxrpc_describe, -}; - -/* - * Vet the description for an RxRPC server key - */ -static int rxrpc_vet_description_s(const char *desc) -{ - unsigned long num; - char *p; - - num = simple_strtoul(desc, &p, 10); - if (*p != ':' || num > 65535) - return -EINVAL; - num = simple_strtoul(p + 1, &p, 10); - if (*p || num < 1 || num > 255) - return -EINVAL; - return 0; -} - -/* - * parse an RxKAD type XDR format token - * - the caller guarantees we have at least 4 words - */ -static int rxrpc_preparse_xdr_rxkad(struct key_preparsed_payload *prep, - size_t datalen, - const __be32 *xdr, unsigned int toklen) -{ - struct rxrpc_key_token *token, **pptoken; - size_t plen; - u32 tktlen; - - _enter(",{%x,%x,%x,%x},%u", - ntohl(xdr[0]), ntohl(xdr[1]), ntohl(xdr[2]), ntohl(xdr[3]), - toklen); - - if (toklen <= 8 * 4) - return -EKEYREJECTED; - tktlen = ntohl(xdr[7]); - _debug("tktlen: %x", tktlen); - if (tktlen > AFSTOKEN_RK_TIX_MAX) - return -EKEYREJECTED; - if (toklen < 8 * 4 + tktlen) - return -EKEYREJECTED; - - plen = sizeof(*token) + sizeof(*token->kad) + tktlen; - prep->quotalen = datalen + plen; - - plen -= sizeof(*token); - token = kzalloc(sizeof(*token), GFP_KERNEL); - if (!token) - return -ENOMEM; - - token->kad = kzalloc(plen, GFP_KERNEL); - if (!token->kad) { - kfree(token); - return -ENOMEM; - } - - token->security_index = RXRPC_SECURITY_RXKAD; - token->kad->ticket_len = tktlen; - token->kad->vice_id = ntohl(xdr[0]); - token->kad->kvno = ntohl(xdr[1]); - token->kad->start = ntohl(xdr[4]); - token->kad->expiry = ntohl(xdr[5]); - token->kad->primary_flag = ntohl(xdr[6]); - memcpy(&token->kad->session_key, &xdr[2], 8); - memcpy(&token->kad->ticket, &xdr[8], tktlen); - - _debug("SCIX: %u", token->security_index); - _debug("TLEN: %u", token->kad->ticket_len); - _debug("EXPY: %x", token->kad->expiry); - _debug("KVNO: %u", token->kad->kvno); - _debug("PRIM: %u", token->kad->primary_flag); - _debug("SKEY: %02x%02x%02x%02x%02x%02x%02x%02x", - token->kad->session_key[0], token->kad->session_key[1], - token->kad->session_key[2], token->kad->session_key[3], - token->kad->session_key[4], token->kad->session_key[5], - token->kad->session_key[6], token->kad->session_key[7]); - if (token->kad->ticket_len >= 8) - _debug("TCKT: %02x%02x%02x%02x%02x%02x%02x%02x", - token->kad->ticket[0], token->kad->ticket[1], - token->kad->ticket[2], token->kad->ticket[3], - token->kad->ticket[4], token->kad->ticket[5], - token->kad->ticket[6], token->kad->ticket[7]); - - /* count the number of tokens attached */ - prep->payload.data[1] = (void *)((unsigned long)prep->payload.data[1] + 1); - - /* attach the data */ - for (pptoken = (struct rxrpc_key_token **)&prep->payload.data[0]; - *pptoken; - pptoken = &(*pptoken)->next) - continue; - *pptoken = token; - if (token->kad->expiry < prep->expiry) - prep->expiry = token->kad->expiry; - - _leave(" = 0"); - return 0; -} - -static void rxrpc_free_krb5_principal(struct krb5_principal *princ) -{ - int loop; - - if (princ->name_parts) { - for (loop = princ->n_name_parts - 1; loop >= 0; loop--) - kfree(princ->name_parts[loop]); - kfree(princ->name_parts); - } - kfree(princ->realm); -} - -static void rxrpc_free_krb5_tagged(struct krb5_tagged_data *td) -{ - kfree(td->data); -} - -/* - * free up an RxK5 token - */ -static void rxrpc_rxk5_free(struct rxk5_key *rxk5) -{ - int loop; - - rxrpc_free_krb5_principal(&rxk5->client); - rxrpc_free_krb5_principal(&rxk5->server); - rxrpc_free_krb5_tagged(&rxk5->session); - - if (rxk5->addresses) { - for (loop = rxk5->n_addresses - 1; loop >= 0; loop--) - rxrpc_free_krb5_tagged(&rxk5->addresses[loop]); - kfree(rxk5->addresses); - } - if (rxk5->authdata) { - for (loop = rxk5->n_authdata - 1; loop >= 0; loop--) - rxrpc_free_krb5_tagged(&rxk5->authdata[loop]); - kfree(rxk5->authdata); - } - - kfree(rxk5->ticket); - kfree(rxk5->ticket2); - kfree(rxk5); -} - -/* - * extract a krb5 principal - */ -static int rxrpc_krb5_decode_principal(struct krb5_principal *princ, - const __be32 **_xdr, - unsigned int *_toklen) -{ - const __be32 *xdr = *_xdr; - unsigned int toklen = *_toklen, n_parts, loop, tmp; - - /* there must be at least one name, and at least #names+1 length - * words */ - if (toklen <= 12) - return -EINVAL; - - _enter(",{%x,%x,%x},%u", - ntohl(xdr[0]), ntohl(xdr[1]), ntohl(xdr[2]), toklen); - - n_parts = ntohl(*xdr++); - toklen -= 4; - if (n_parts <= 0 || n_parts > AFSTOKEN_K5_COMPONENTS_MAX) - return -EINVAL; - princ->n_name_parts = n_parts; - - if (toklen <= (n_parts + 1) * 4) - return -EINVAL; - - princ->name_parts = kcalloc(n_parts, sizeof(char *), GFP_KERNEL); - if (!princ->name_parts) - return -ENOMEM; - - for (loop = 0; loop < n_parts; loop++) { - if (toklen < 4) - return -EINVAL; - tmp = ntohl(*xdr++); - toklen -= 4; - if (tmp <= 0 || tmp > AFSTOKEN_STRING_MAX) - return -EINVAL; - if (tmp > toklen) - return -EINVAL; - princ->name_parts[loop] = kmalloc(tmp + 1, GFP_KERNEL); - if (!princ->name_parts[loop]) - return -ENOMEM; - memcpy(princ->name_parts[loop], xdr, tmp); - princ->name_parts[loop][tmp] = 0; - tmp = (tmp + 3) & ~3; - toklen -= tmp; - xdr += tmp >> 2; - } - - if (toklen < 4) - return -EINVAL; - tmp = ntohl(*xdr++); - toklen -= 4; - if (tmp <= 0 || tmp > AFSTOKEN_K5_REALM_MAX) - return -EINVAL; - if (tmp > toklen) - return -EINVAL; - princ->realm = kmalloc(tmp + 1, GFP_KERNEL); - if (!princ->realm) - return -ENOMEM; - memcpy(princ->realm, xdr, tmp); - princ->realm[tmp] = 0; - tmp = (tmp + 3) & ~3; - toklen -= tmp; - xdr += tmp >> 2; - - _debug("%s/...@%s", princ->name_parts[0], princ->realm); - - *_xdr = xdr; - *_toklen = toklen; - _leave(" = 0 [toklen=%u]", toklen); - return 0; -} - -/* - * extract a piece of krb5 tagged data - */ -static int rxrpc_krb5_decode_tagged_data(struct krb5_tagged_data *td, - size_t max_data_size, - const __be32 **_xdr, - unsigned int *_toklen) -{ - const __be32 *xdr = *_xdr; - unsigned int toklen = *_toklen, len; - - /* there must be at least one tag and one length word */ - if (toklen <= 8) - return -EINVAL; - - _enter(",%zu,{%x,%x},%u", - max_data_size, ntohl(xdr[0]), ntohl(xdr[1]), toklen); - - td->tag = ntohl(*xdr++); - len = ntohl(*xdr++); - toklen -= 8; - if (len > max_data_size) - return -EINVAL; - td->data_len = len; - - if (len > 0) { - td->data = kmemdup(xdr, len, GFP_KERNEL); - if (!td->data) - return -ENOMEM; - len = (len + 3) & ~3; - toklen -= len; - xdr += len >> 2; - } - - _debug("tag %x len %x", td->tag, td->data_len); - - *_xdr = xdr; - *_toklen = toklen; - _leave(" = 0 [toklen=%u]", toklen); - return 0; -} - -/* - * extract an array of tagged data - */ -static int rxrpc_krb5_decode_tagged_array(struct krb5_tagged_data **_td, - u8 *_n_elem, - u8 max_n_elem, - size_t max_elem_size, - const __be32 **_xdr, - unsigned int *_toklen) -{ - struct krb5_tagged_data *td; - const __be32 *xdr = *_xdr; - unsigned int toklen = *_toklen, n_elem, loop; - int ret; - - /* there must be at least one count */ - if (toklen < 4) - return -EINVAL; - - _enter(",,%u,%zu,{%x},%u", - max_n_elem, max_elem_size, ntohl(xdr[0]), toklen); - - n_elem = ntohl(*xdr++); - toklen -= 4; - if (n_elem > max_n_elem) - return -EINVAL; - *_n_elem = n_elem; - if (n_elem > 0) { - if (toklen <= (n_elem + 1) * 4) - return -EINVAL; - - _debug("n_elem %d", n_elem); - - td = kcalloc(n_elem, sizeof(struct krb5_tagged_data), - GFP_KERNEL); - if (!td) - return -ENOMEM; - *_td = td; - - for (loop = 0; loop < n_elem; loop++) { - ret = rxrpc_krb5_decode_tagged_data(&td[loop], - max_elem_size, - &xdr, &toklen); - if (ret < 0) - return ret; - } - } - - *_xdr = xdr; - *_toklen = toklen; - _leave(" = 0 [toklen=%u]", toklen); - return 0; -} - -/* - * extract a krb5 ticket - */ -static int rxrpc_krb5_decode_ticket(u8 **_ticket, u16 *_tktlen, - const __be32 **_xdr, unsigned int *_toklen) -{ - const __be32 *xdr = *_xdr; - unsigned int toklen = *_toklen, len; - - /* there must be at least one length word */ - if (toklen <= 4) - return -EINVAL; - - _enter(",{%x},%u", ntohl(xdr[0]), toklen); - - len = ntohl(*xdr++); - toklen -= 4; - if (len > AFSTOKEN_K5_TIX_MAX) - return -EINVAL; - *_tktlen = len; - - _debug("ticket len %u", len); - - if (len > 0) { - *_ticket = kmemdup(xdr, len, GFP_KERNEL); - if (!*_ticket) - return -ENOMEM; - len = (len + 3) & ~3; - toklen -= len; - xdr += len >> 2; - } - - *_xdr = xdr; - *_toklen = toklen; - _leave(" = 0 [toklen=%u]", toklen); - return 0; -} - -/* - * parse an RxK5 type XDR format token - * - the caller guarantees we have at least 4 words - */ -static int rxrpc_preparse_xdr_rxk5(struct key_preparsed_payload *prep, - size_t datalen, - const __be32 *xdr, unsigned int toklen) -{ - struct rxrpc_key_token *token, **pptoken; - struct rxk5_key *rxk5; - const __be32 *end_xdr = xdr + (toklen >> 2); - int ret; - - _enter(",{%x,%x,%x,%x},%u", - ntohl(xdr[0]), ntohl(xdr[1]), ntohl(xdr[2]), ntohl(xdr[3]), - toklen); - - /* reserve some payload space for this subkey - the length of the token - * is a reasonable approximation */ - prep->quotalen = datalen + toklen; - - token = kzalloc(sizeof(*token), GFP_KERNEL); - if (!token) - return -ENOMEM; - - rxk5 = kzalloc(sizeof(*rxk5), GFP_KERNEL); - if (!rxk5) { - kfree(token); - return -ENOMEM; - } - - token->security_index = RXRPC_SECURITY_RXK5; - token->k5 = rxk5; - - /* extract the principals */ - ret = rxrpc_krb5_decode_principal(&rxk5->client, &xdr, &toklen); - if (ret < 0) - goto error; - ret = rxrpc_krb5_decode_principal(&rxk5->server, &xdr, &toklen); - if (ret < 0) - goto error; - - /* extract the session key and the encoding type (the tag field -> - * ENCTYPE_xxx) */ - ret = rxrpc_krb5_decode_tagged_data(&rxk5->session, AFSTOKEN_DATA_MAX, - &xdr, &toklen); - if (ret < 0) - goto error; - - if (toklen < 4 * 8 + 2 * 4) - goto inval; - rxk5->authtime = be64_to_cpup((const __be64 *) xdr); - xdr += 2; - rxk5->starttime = be64_to_cpup((const __be64 *) xdr); - xdr += 2; - rxk5->endtime = be64_to_cpup((const __be64 *) xdr); - xdr += 2; - rxk5->renew_till = be64_to_cpup((const __be64 *) xdr); - xdr += 2; - rxk5->is_skey = ntohl(*xdr++); - rxk5->flags = ntohl(*xdr++); - toklen -= 4 * 8 + 2 * 4; - - _debug("times: a=%llx s=%llx e=%llx rt=%llx", - rxk5->authtime, rxk5->starttime, rxk5->endtime, - rxk5->renew_till); - _debug("is_skey=%x flags=%x", rxk5->is_skey, rxk5->flags); - - /* extract the permitted client addresses */ - ret = rxrpc_krb5_decode_tagged_array(&rxk5->addresses, - &rxk5->n_addresses, - AFSTOKEN_K5_ADDRESSES_MAX, - AFSTOKEN_DATA_MAX, - &xdr, &toklen); - if (ret < 0) - goto error; - - ASSERTCMP((end_xdr - xdr) << 2, ==, toklen); - - /* extract the tickets */ - ret = rxrpc_krb5_decode_ticket(&rxk5->ticket, &rxk5->ticket_len, - &xdr, &toklen); - if (ret < 0) - goto error; - ret = rxrpc_krb5_decode_ticket(&rxk5->ticket2, &rxk5->ticket2_len, - &xdr, &toklen); - if (ret < 0) - goto error; - - ASSERTCMP((end_xdr - xdr) << 2, ==, toklen); - - /* extract the typed auth data */ - ret = rxrpc_krb5_decode_tagged_array(&rxk5->authdata, - &rxk5->n_authdata, - AFSTOKEN_K5_AUTHDATA_MAX, - AFSTOKEN_BDATALN_MAX, - &xdr, &toklen); - if (ret < 0) - goto error; - - ASSERTCMP((end_xdr - xdr) << 2, ==, toklen); - - if (toklen != 0) - goto inval; - - /* attach the payload */ - for (pptoken = (struct rxrpc_key_token **)&prep->payload.data[0]; - *pptoken; - pptoken = &(*pptoken)->next) - continue; - *pptoken = token; - if (token->kad->expiry < prep->expiry) - prep->expiry = token->kad->expiry; - - _leave(" = 0"); - return 0; - -inval: - ret = -EINVAL; -error: - rxrpc_rxk5_free(rxk5); - kfree(token); - _leave(" = %d", ret); - return ret; -} - -/* - * attempt to parse the data as the XDR format - * - the caller guarantees we have more than 7 words - */ -static int rxrpc_preparse_xdr(struct key_preparsed_payload *prep) -{ - const __be32 *xdr = prep->data, *token; - const char *cp; - unsigned int len, tmp, loop, ntoken, toklen, sec_ix; - size_t datalen = prep->datalen; - int ret; - - _enter(",{%x,%x,%x,%x},%zu", - ntohl(xdr[0]), ntohl(xdr[1]), ntohl(xdr[2]), ntohl(xdr[3]), - prep->datalen); - - if (datalen > AFSTOKEN_LENGTH_MAX) - goto not_xdr; - - /* XDR is an array of __be32's */ - if (datalen & 3) - goto not_xdr; - - /* the flags should be 0 (the setpag bit must be handled by - * userspace) */ - if (ntohl(*xdr++) != 0) - goto not_xdr; - datalen -= 4; - - /* check the cell name */ - len = ntohl(*xdr++); - if (len < 1 || len > AFSTOKEN_CELL_MAX) - goto not_xdr; - datalen -= 4; - tmp = (len + 3) & ~3; - if (tmp > datalen) - goto not_xdr; - - cp = (const char *) xdr; - for (loop = 0; loop < len; loop++) - if (!isprint(cp[loop])) - goto not_xdr; - if (len < tmp) - for (; loop < tmp; loop++) - if (cp[loop]) - goto not_xdr; - _debug("cellname: [%u/%u] '%*.*s'", - len, tmp, len, len, (const char *) xdr); - datalen -= tmp; - xdr += tmp >> 2; - - /* get the token count */ - if (datalen < 12) - goto not_xdr; - ntoken = ntohl(*xdr++); - datalen -= 4; - _debug("ntoken: %x", ntoken); - if (ntoken < 1 || ntoken > AFSTOKEN_MAX) - goto not_xdr; - - /* check each token wrapper */ - token = xdr; - loop = ntoken; - do { - if (datalen < 8) - goto not_xdr; - toklen = ntohl(*xdr++); - sec_ix = ntohl(*xdr); - datalen -= 4; - _debug("token: [%x/%zx] %x", toklen, datalen, sec_ix); - if (toklen < 20 || toklen > datalen) - goto not_xdr; - datalen -= (toklen + 3) & ~3; - xdr += (toklen + 3) >> 2; - - } while (--loop > 0); - - _debug("remainder: %zu", datalen); - if (datalen != 0) - goto not_xdr; - - /* okay: we're going to assume it's valid XDR format - * - we ignore the cellname, relying on the key to be correctly named - */ - do { - xdr = token; - toklen = ntohl(*xdr++); - token = xdr + ((toklen + 3) >> 2); - sec_ix = ntohl(*xdr++); - toklen -= 4; - - _debug("TOKEN type=%u [%p-%p]", sec_ix, xdr, token); - - switch (sec_ix) { - case RXRPC_SECURITY_RXKAD: - ret = rxrpc_preparse_xdr_rxkad(prep, datalen, xdr, toklen); - if (ret != 0) - goto error; - break; - - case RXRPC_SECURITY_RXK5: - ret = rxrpc_preparse_xdr_rxk5(prep, datalen, xdr, toklen); - if (ret != 0) - goto error; - break; - - default: - ret = -EPROTONOSUPPORT; - goto error; - } - - } while (--ntoken > 0); - - _leave(" = 0"); - return 0; - -not_xdr: - _leave(" = -EPROTO"); - return -EPROTO; -error: - _leave(" = %d", ret); - return ret; -} - -/* - * Preparse an rxrpc defined key. - * - * Data should be of the form: - * OFFSET LEN CONTENT - * 0 4 key interface version number - * 4 2 security index (type) - * 6 2 ticket length - * 8 4 key expiry time (time_t) - * 12 4 kvno - * 16 8 session key - * 24 [len] ticket - * - * if no data is provided, then a no-security key is made - */ -static int rxrpc_preparse(struct key_preparsed_payload *prep) -{ - const struct rxrpc_key_data_v1 *v1; - struct rxrpc_key_token *token, **pp; - size_t plen; - u32 kver; - int ret; - - _enter("%zu", prep->datalen); - - /* handle a no-security key */ - if (!prep->data && prep->datalen == 0) - return 0; - - /* determine if the XDR payload format is being used */ - if (prep->datalen > 7 * 4) { - ret = rxrpc_preparse_xdr(prep); - if (ret != -EPROTO) - return ret; - } - - /* get the key interface version number */ - ret = -EINVAL; - if (prep->datalen <= 4 || !prep->data) - goto error; - memcpy(&kver, prep->data, sizeof(kver)); - prep->data += sizeof(kver); - prep->datalen -= sizeof(kver); - - _debug("KEY I/F VERSION: %u", kver); - - ret = -EKEYREJECTED; - if (kver != 1) - goto error; - - /* deal with a version 1 key */ - ret = -EINVAL; - if (prep->datalen < sizeof(*v1)) - goto error; - - v1 = prep->data; - if (prep->datalen != sizeof(*v1) + v1->ticket_length) - goto error; - - _debug("SCIX: %u", v1->security_index); - _debug("TLEN: %u", v1->ticket_length); - _debug("EXPY: %x", v1->expiry); - _debug("KVNO: %u", v1->kvno); - _debug("SKEY: %02x%02x%02x%02x%02x%02x%02x%02x", - v1->session_key[0], v1->session_key[1], - v1->session_key[2], v1->session_key[3], - v1->session_key[4], v1->session_key[5], - v1->session_key[6], v1->session_key[7]); - if (v1->ticket_length >= 8) - _debug("TCKT: %02x%02x%02x%02x%02x%02x%02x%02x", - v1->ticket[0], v1->ticket[1], - v1->ticket[2], v1->ticket[3], - v1->ticket[4], v1->ticket[5], - v1->ticket[6], v1->ticket[7]); - - ret = -EPROTONOSUPPORT; - if (v1->security_index != RXRPC_SECURITY_RXKAD) - goto error; - - plen = sizeof(*token->kad) + v1->ticket_length; - prep->quotalen = plen + sizeof(*token); - - ret = -ENOMEM; - token = kzalloc(sizeof(*token), GFP_KERNEL); - if (!token) - goto error; - token->kad = kzalloc(plen, GFP_KERNEL); - if (!token->kad) - goto error_free; - - token->security_index = RXRPC_SECURITY_RXKAD; - token->kad->ticket_len = v1->ticket_length; - token->kad->expiry = v1->expiry; - token->kad->kvno = v1->kvno; - memcpy(&token->kad->session_key, &v1->session_key, 8); - memcpy(&token->kad->ticket, v1->ticket, v1->ticket_length); - - /* count the number of tokens attached */ - prep->payload.data[1] = (void *)((unsigned long)prep->payload.data[1] + 1); - - /* attach the data */ - pp = (struct rxrpc_key_token **)&prep->payload.data[0]; - while (*pp) - pp = &(*pp)->next; - *pp = token; - if (token->kad->expiry < prep->expiry) - prep->expiry = token->kad->expiry; - token = NULL; - ret = 0; - -error_free: - kfree(token); -error: - return ret; -} - -/* - * Free token list. - */ -static void rxrpc_free_token_list(struct rxrpc_key_token *token) -{ - struct rxrpc_key_token *next; - - for (; token; token = next) { - next = token->next; - switch (token->security_index) { - case RXRPC_SECURITY_RXKAD: - kfree(token->kad); - break; - case RXRPC_SECURITY_RXK5: - if (token->k5) - rxrpc_rxk5_free(token->k5); - break; - default: - pr_err("Unknown token type %x on rxrpc key\n", - token->security_index); - BUG(); - } - - kfree(token); - } -} - -/* - * Clean up preparse data. - */ -static void rxrpc_free_preparse(struct key_preparsed_payload *prep) -{ - rxrpc_free_token_list(prep->payload.data[0]); -} - -/* - * Preparse a server secret key. - * - * The data should be the 8-byte secret key. - */ -static int rxrpc_preparse_s(struct key_preparsed_payload *prep) -{ - struct crypto_skcipher *ci; - - _enter("%zu", prep->datalen); - - if (prep->datalen != 8) - return -EINVAL; - - memcpy(&prep->payload.data[2], prep->data, 8); - - ci = crypto_alloc_skcipher("pcbc(des)", 0, CRYPTO_ALG_ASYNC); - if (IS_ERR(ci)) { - _leave(" = %ld", PTR_ERR(ci)); - return PTR_ERR(ci); - } - - if (crypto_skcipher_setkey(ci, prep->data, 8) < 0) - BUG(); - - prep->payload.data[0] = ci; - _leave(" = 0"); - return 0; -} - -/* - * Clean up preparse data. - */ -static void rxrpc_free_preparse_s(struct key_preparsed_payload *prep) -{ - if (prep->payload.data[0]) - crypto_free_skcipher(prep->payload.data[0]); -} - -/* - * dispose of the data dangling from the corpse of a rxrpc key - */ -static void rxrpc_destroy(struct key *key) -{ - rxrpc_free_token_list(key->payload.data[0]); -} - -/* - * dispose of the data dangling from the corpse of a rxrpc key - */ -static void rxrpc_destroy_s(struct key *key) -{ - if (key->payload.data[0]) { - crypto_free_skcipher(key->payload.data[0]); - key->payload.data[0] = NULL; - } -} - -/* - * describe the rxrpc key - */ -static void rxrpc_describe(const struct key *key, struct seq_file *m) -{ - seq_puts(m, key->description); -} - -/* - * grab the security key for a socket - */ -int rxrpc_request_key(struct rxrpc_sock *rx, char __user *optval, int optlen) -{ - struct key *key; - char *description; - - _enter(""); - - if (optlen <= 0 || optlen > PAGE_SIZE - 1) - return -EINVAL; - - description = memdup_user_nul(optval, optlen); - if (IS_ERR(description)) - return PTR_ERR(description); - - key = request_key(&key_type_rxrpc, description, NULL); - if (IS_ERR(key)) { - kfree(description); - _leave(" = %ld", PTR_ERR(key)); - return PTR_ERR(key); - } - - rx->key = key; - kfree(description); - _leave(" = 0 [key %x]", key->serial); - return 0; -} - -/* - * grab the security keyring for a server socket - */ -int rxrpc_server_keyring(struct rxrpc_sock *rx, char __user *optval, - int optlen) -{ - struct key *key; - char *description; - - _enter(""); - - if (optlen <= 0 || optlen > PAGE_SIZE - 1) - return -EINVAL; - - description = memdup_user_nul(optval, optlen); - if (IS_ERR(description)) - return PTR_ERR(description); - - key = request_key(&key_type_keyring, description, NULL); - if (IS_ERR(key)) { - kfree(description); - _leave(" = %ld", PTR_ERR(key)); - return PTR_ERR(key); - } - - rx->securities = key; - kfree(description); - _leave(" = 0 [key %x]", key->serial); - return 0; -} - -/* - * generate a server data key - */ -int rxrpc_get_server_data_key(struct rxrpc_connection *conn, - const void *session_key, - time_t expiry, - u32 kvno) -{ - const struct cred *cred = current_cred(); - struct key *key; - int ret; - - struct { - u32 kver; - struct rxrpc_key_data_v1 v1; - } data; - - _enter(""); - - key = key_alloc(&key_type_rxrpc, "x", - GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, cred, 0, - KEY_ALLOC_NOT_IN_QUOTA, NULL); - if (IS_ERR(key)) { - _leave(" = -ENOMEM [alloc %ld]", PTR_ERR(key)); - return -ENOMEM; - } - - _debug("key %d", key_serial(key)); - - data.kver = 1; - data.v1.security_index = RXRPC_SECURITY_RXKAD; - data.v1.ticket_length = 0; - data.v1.expiry = expiry; - data.v1.kvno = 0; - - memcpy(&data.v1.session_key, session_key, sizeof(data.v1.session_key)); - - ret = key_instantiate_and_link(key, &data, sizeof(data), NULL, NULL); - if (ret < 0) - goto error; - - conn->key = key; - _leave(" = 0 [%d]", key_serial(key)); - return 0; - -error: - key_revoke(key); - key_put(key); - _leave(" = -ENOMEM [ins %d]", ret); - return -ENOMEM; -} -EXPORT_SYMBOL(rxrpc_get_server_data_key); - -/** - * rxrpc_get_null_key - Generate a null RxRPC key - * @keyname: The name to give the key. - * - * Generate a null RxRPC key that can be used to indicate anonymous security is - * required for a particular domain. - */ -struct key *rxrpc_get_null_key(const char *keyname) -{ - const struct cred *cred = current_cred(); - struct key *key; - int ret; - - key = key_alloc(&key_type_rxrpc, keyname, - GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, cred, - KEY_POS_SEARCH, KEY_ALLOC_NOT_IN_QUOTA, NULL); - if (IS_ERR(key)) - return key; - - ret = key_instantiate_and_link(key, NULL, 0, NULL, NULL); - if (ret < 0) { - key_revoke(key); - key_put(key); - return ERR_PTR(ret); - } - - return key; -} -EXPORT_SYMBOL(rxrpc_get_null_key); - -/* - * read the contents of an rxrpc key - * - this returns the result in XDR form - */ -static long rxrpc_read(const struct key *key, - char __user *buffer, size_t buflen) -{ - const struct rxrpc_key_token *token; - const struct krb5_principal *princ; - size_t size; - __be32 __user *xdr, *oldxdr; - u32 cnlen, toksize, ntoks, tok, zero; - u16 toksizes[AFSTOKEN_MAX]; - int loop; - - _enter(""); - - /* we don't know what form we should return non-AFS keys in */ - if (memcmp(key->description, "afs@", 4) != 0) - return -EOPNOTSUPP; - cnlen = strlen(key->description + 4); - -#define RND(X) (((X) + 3) & ~3) - - /* AFS keys we return in XDR form, so we need to work out the size of - * the XDR */ - size = 2 * 4; /* flags, cellname len */ - size += RND(cnlen); /* cellname */ - size += 1 * 4; /* token count */ - - ntoks = 0; - for (token = key->payload.data[0]; token; token = token->next) { - toksize = 4; /* sec index */ - - switch (token->security_index) { - case RXRPC_SECURITY_RXKAD: - toksize += 8 * 4; /* viceid, kvno, key*2, begin, - * end, primary, tktlen */ - toksize += RND(token->kad->ticket_len); - break; - - case RXRPC_SECURITY_RXK5: - princ = &token->k5->client; - toksize += 4 + princ->n_name_parts * 4; - for (loop = 0; loop < princ->n_name_parts; loop++) - toksize += RND(strlen(princ->name_parts[loop])); - toksize += 4 + RND(strlen(princ->realm)); - - princ = &token->k5->server; - toksize += 4 + princ->n_name_parts * 4; - for (loop = 0; loop < princ->n_name_parts; loop++) - toksize += RND(strlen(princ->name_parts[loop])); - toksize += 4 + RND(strlen(princ->realm)); - - toksize += 8 + RND(token->k5->session.data_len); - - toksize += 4 * 8 + 2 * 4; - - toksize += 4 + token->k5->n_addresses * 8; - for (loop = 0; loop < token->k5->n_addresses; loop++) - toksize += RND(token->k5->addresses[loop].data_len); - - toksize += 4 + RND(token->k5->ticket_len); - toksize += 4 + RND(token->k5->ticket2_len); - - toksize += 4 + token->k5->n_authdata * 8; - for (loop = 0; loop < token->k5->n_authdata; loop++) - toksize += RND(token->k5->authdata[loop].data_len); - break; - - default: /* we have a ticket we can't encode */ - BUG(); - continue; - } - - _debug("token[%u]: toksize=%u", ntoks, toksize); - ASSERTCMP(toksize, <=, AFSTOKEN_LENGTH_MAX); - - toksizes[ntoks++] = toksize; - size += toksize + 4; /* each token has a length word */ - } - -#undef RND - - if (!buffer || buflen < size) - return size; - - xdr = (__be32 __user *) buffer; - zero = 0; -#define ENCODE(x) \ - do { \ - __be32 y = htonl(x); \ - if (put_user(y, xdr++) < 0) \ - goto fault; \ - } while(0) -#define ENCODE_DATA(l, s) \ - do { \ - u32 _l = (l); \ - ENCODE(l); \ - if (copy_to_user(xdr, (s), _l) != 0) \ - goto fault; \ - if (_l & 3 && \ - copy_to_user((u8 __user *)xdr + _l, &zero, 4 - (_l & 3)) != 0) \ - goto fault; \ - xdr += (_l + 3) >> 2; \ - } while(0) -#define ENCODE64(x) \ - do { \ - __be64 y = cpu_to_be64(x); \ - if (copy_to_user(xdr, &y, 8) != 0) \ - goto fault; \ - xdr += 8 >> 2; \ - } while(0) -#define ENCODE_STR(s) \ - do { \ - const char *_s = (s); \ - ENCODE_DATA(strlen(_s), _s); \ - } while(0) - - ENCODE(0); /* flags */ - ENCODE_DATA(cnlen, key->description + 4); /* cellname */ - ENCODE(ntoks); - - tok = 0; - for (token = key->payload.data[0]; token; token = token->next) { - toksize = toksizes[tok++]; - ENCODE(toksize); - oldxdr = xdr; - ENCODE(token->security_index); - - switch (token->security_index) { - case RXRPC_SECURITY_RXKAD: - ENCODE(token->kad->vice_id); - ENCODE(token->kad->kvno); - ENCODE_DATA(8, token->kad->session_key); - ENCODE(token->kad->start); - ENCODE(token->kad->expiry); - ENCODE(token->kad->primary_flag); - ENCODE_DATA(token->kad->ticket_len, token->kad->ticket); - break; - - case RXRPC_SECURITY_RXK5: - princ = &token->k5->client; - ENCODE(princ->n_name_parts); - for (loop = 0; loop < princ->n_name_parts; loop++) - ENCODE_STR(princ->name_parts[loop]); - ENCODE_STR(princ->realm); - - princ = &token->k5->server; - ENCODE(princ->n_name_parts); - for (loop = 0; loop < princ->n_name_parts; loop++) - ENCODE_STR(princ->name_parts[loop]); - ENCODE_STR(princ->realm); - - ENCODE(token->k5->session.tag); - ENCODE_DATA(token->k5->session.data_len, - token->k5->session.data); - - ENCODE64(token->k5->authtime); - ENCODE64(token->k5->starttime); - ENCODE64(token->k5->endtime); - ENCODE64(token->k5->renew_till); - ENCODE(token->k5->is_skey); - ENCODE(token->k5->flags); - - ENCODE(token->k5->n_addresses); - for (loop = 0; loop < token->k5->n_addresses; loop++) { - ENCODE(token->k5->addresses[loop].tag); - ENCODE_DATA(token->k5->addresses[loop].data_len, - token->k5->addresses[loop].data); - } - - ENCODE_DATA(token->k5->ticket_len, token->k5->ticket); - ENCODE_DATA(token->k5->ticket2_len, token->k5->ticket2); - - ENCODE(token->k5->n_authdata); - for (loop = 0; loop < token->k5->n_authdata; loop++) { - ENCODE(token->k5->authdata[loop].tag); - ENCODE_DATA(token->k5->authdata[loop].data_len, - token->k5->authdata[loop].data); - } - break; - - default: - BUG(); - break; - } - - ASSERTCMP((unsigned long)xdr - (unsigned long)oldxdr, ==, - toksize); - } - -#undef ENCODE_STR -#undef ENCODE_DATA -#undef ENCODE64 -#undef ENCODE - - ASSERTCMP(tok, ==, ntoks); - ASSERTCMP((char __user *) xdr - buffer, ==, size); - _leave(" = %zu", size); - return size; - -fault: - _leave(" = -EFAULT"); - return -EFAULT; -} diff --git a/net/rxrpc/ar-local.c b/net/rxrpc/ar-local.c deleted file mode 100644 index 111f250..0000000 --- a/net/rxrpc/ar-local.c +++ /dev/null @@ -1,417 +0,0 @@ -/* AF_RXRPC local endpoint management - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -static const char rxrpc_version_string[65] = "linux-" UTS_RELEASE " AF_RXRPC"; - -static LIST_HEAD(rxrpc_locals); -DEFINE_RWLOCK(rxrpc_local_lock); -static DECLARE_RWSEM(rxrpc_local_sem); -static DECLARE_WAIT_QUEUE_HEAD(rxrpc_local_wq); - -static void rxrpc_destroy_local(struct work_struct *work); -static void rxrpc_process_local_events(struct work_struct *work); - -/* - * allocate a new local - */ -static -struct rxrpc_local *rxrpc_alloc_local(struct sockaddr_rxrpc *srx) -{ - struct rxrpc_local *local; - - local = kzalloc(sizeof(struct rxrpc_local), GFP_KERNEL); - if (local) { - INIT_WORK(&local->destroyer, &rxrpc_destroy_local); - INIT_WORK(&local->acceptor, &rxrpc_accept_incoming_calls); - INIT_WORK(&local->rejecter, &rxrpc_reject_packets); - INIT_WORK(&local->event_processor, &rxrpc_process_local_events); - INIT_LIST_HEAD(&local->services); - INIT_LIST_HEAD(&local->link); - init_rwsem(&local->defrag_sem); - skb_queue_head_init(&local->accept_queue); - skb_queue_head_init(&local->reject_queue); - skb_queue_head_init(&local->event_queue); - spin_lock_init(&local->lock); - rwlock_init(&local->services_lock); - atomic_set(&local->usage, 1); - local->debug_id = atomic_inc_return(&rxrpc_debug_id); - memcpy(&local->srx, srx, sizeof(*srx)); - } - - _leave(" = %p", local); - return local; -} - -/* - * create the local socket - * - must be called with rxrpc_local_sem writelocked - */ -static int rxrpc_create_local(struct rxrpc_local *local) -{ - struct sock *sock; - int ret, opt; - - _enter("%p{%d}", local, local->srx.transport_type); - - /* create a socket to represent the local endpoint */ - ret = sock_create_kern(&init_net, PF_INET, local->srx.transport_type, - IPPROTO_UDP, &local->socket); - if (ret < 0) { - _leave(" = %d [socket]", ret); - return ret; - } - - /* if a local address was supplied then bind it */ - if (local->srx.transport_len > sizeof(sa_family_t)) { - _debug("bind"); - ret = kernel_bind(local->socket, - (struct sockaddr *) &local->srx.transport, - local->srx.transport_len); - if (ret < 0) { - _debug("bind failed"); - goto error; - } - } - - /* we want to receive ICMP errors */ - opt = 1; - ret = kernel_setsockopt(local->socket, SOL_IP, IP_RECVERR, - (char *) &opt, sizeof(opt)); - if (ret < 0) { - _debug("setsockopt failed"); - goto error; - } - - /* we want to set the don't fragment bit */ - opt = IP_PMTUDISC_DO; - ret = kernel_setsockopt(local->socket, SOL_IP, IP_MTU_DISCOVER, - (char *) &opt, sizeof(opt)); - if (ret < 0) { - _debug("setsockopt failed"); - goto error; - } - - write_lock_bh(&rxrpc_local_lock); - list_add(&local->link, &rxrpc_locals); - write_unlock_bh(&rxrpc_local_lock); - - /* set the socket up */ - sock = local->socket->sk; - sock->sk_user_data = local; - sock->sk_data_ready = rxrpc_data_ready; - sock->sk_error_report = rxrpc_UDP_error_report; - _leave(" = 0"); - return 0; - -error: - kernel_sock_shutdown(local->socket, SHUT_RDWR); - local->socket->sk->sk_user_data = NULL; - sock_release(local->socket); - local->socket = NULL; - - _leave(" = %d", ret); - return ret; -} - -/* - * create a new local endpoint using the specified UDP address - */ -struct rxrpc_local *rxrpc_lookup_local(struct sockaddr_rxrpc *srx) -{ - struct rxrpc_local *local; - int ret; - - _enter("{%d,%u,%pI4+%hu}", - srx->transport_type, - srx->transport.family, - &srx->transport.sin.sin_addr, - ntohs(srx->transport.sin.sin_port)); - - down_write(&rxrpc_local_sem); - - /* see if we have a suitable local local endpoint already */ - read_lock_bh(&rxrpc_local_lock); - - list_for_each_entry(local, &rxrpc_locals, link) { - _debug("CMP {%d,%u,%pI4+%hu}", - local->srx.transport_type, - local->srx.transport.family, - &local->srx.transport.sin.sin_addr, - ntohs(local->srx.transport.sin.sin_port)); - - if (local->srx.transport_type != srx->transport_type || - local->srx.transport.family != srx->transport.family) - continue; - - switch (srx->transport.family) { - case AF_INET: - if (local->srx.transport.sin.sin_port != - srx->transport.sin.sin_port) - continue; - if (memcmp(&local->srx.transport.sin.sin_addr, - &srx->transport.sin.sin_addr, - sizeof(struct in_addr)) != 0) - continue; - goto found_local; - - default: - BUG(); - } - } - - read_unlock_bh(&rxrpc_local_lock); - - /* we didn't find one, so we need to create one */ - local = rxrpc_alloc_local(srx); - if (!local) { - up_write(&rxrpc_local_sem); - return ERR_PTR(-ENOMEM); - } - - ret = rxrpc_create_local(local); - if (ret < 0) { - up_write(&rxrpc_local_sem); - kfree(local); - _leave(" = %d", ret); - return ERR_PTR(ret); - } - - up_write(&rxrpc_local_sem); - - _net("LOCAL new %d {%d,%u,%pI4+%hu}", - local->debug_id, - local->srx.transport_type, - local->srx.transport.family, - &local->srx.transport.sin.sin_addr, - ntohs(local->srx.transport.sin.sin_port)); - - _leave(" = %p [new]", local); - return local; - -found_local: - rxrpc_get_local(local); - read_unlock_bh(&rxrpc_local_lock); - up_write(&rxrpc_local_sem); - - _net("LOCAL old %d {%d,%u,%pI4+%hu}", - local->debug_id, - local->srx.transport_type, - local->srx.transport.family, - &local->srx.transport.sin.sin_addr, - ntohs(local->srx.transport.sin.sin_port)); - - _leave(" = %p [reuse]", local); - return local; -} - -/* - * release a local endpoint - */ -void rxrpc_put_local(struct rxrpc_local *local) -{ - _enter("%p{u=%d}", local, atomic_read(&local->usage)); - - ASSERTCMP(atomic_read(&local->usage), >, 0); - - /* to prevent a race, the decrement and the dequeue must be effectively - * atomic */ - write_lock_bh(&rxrpc_local_lock); - if (unlikely(atomic_dec_and_test(&local->usage))) { - _debug("destroy local"); - rxrpc_queue_work(&local->destroyer); - } - write_unlock_bh(&rxrpc_local_lock); - _leave(""); -} - -/* - * destroy a local endpoint - */ -static void rxrpc_destroy_local(struct work_struct *work) -{ - struct rxrpc_local *local = - container_of(work, struct rxrpc_local, destroyer); - - _enter("%p{%d}", local, atomic_read(&local->usage)); - - down_write(&rxrpc_local_sem); - - write_lock_bh(&rxrpc_local_lock); - if (atomic_read(&local->usage) > 0) { - write_unlock_bh(&rxrpc_local_lock); - up_read(&rxrpc_local_sem); - _leave(" [resurrected]"); - return; - } - - list_del(&local->link); - local->socket->sk->sk_user_data = NULL; - write_unlock_bh(&rxrpc_local_lock); - - downgrade_write(&rxrpc_local_sem); - - ASSERT(list_empty(&local->services)); - ASSERT(!work_pending(&local->acceptor)); - ASSERT(!work_pending(&local->rejecter)); - ASSERT(!work_pending(&local->event_processor)); - - /* finish cleaning up the local descriptor */ - rxrpc_purge_queue(&local->accept_queue); - rxrpc_purge_queue(&local->reject_queue); - rxrpc_purge_queue(&local->event_queue); - kernel_sock_shutdown(local->socket, SHUT_RDWR); - sock_release(local->socket); - - up_read(&rxrpc_local_sem); - - _net("DESTROY LOCAL %d", local->debug_id); - kfree(local); - - if (list_empty(&rxrpc_locals)) - wake_up_all(&rxrpc_local_wq); - - _leave(""); -} - -/* - * preemptively destroy all local local endpoint rather than waiting for - * them to be destroyed - */ -void __exit rxrpc_destroy_all_locals(void) -{ - DECLARE_WAITQUEUE(myself,current); - - _enter(""); - - /* we simply have to wait for them to go away */ - if (!list_empty(&rxrpc_locals)) { - set_current_state(TASK_UNINTERRUPTIBLE); - add_wait_queue(&rxrpc_local_wq, &myself); - - while (!list_empty(&rxrpc_locals)) { - schedule(); - set_current_state(TASK_UNINTERRUPTIBLE); - } - - remove_wait_queue(&rxrpc_local_wq, &myself); - set_current_state(TASK_RUNNING); - } - - _leave(""); -} - -/* - * Reply to a version request - */ -static void rxrpc_send_version_request(struct rxrpc_local *local, - struct rxrpc_host_header *hdr, - struct sk_buff *skb) -{ - struct rxrpc_wire_header whdr; - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - struct sockaddr_in sin; - struct msghdr msg; - struct kvec iov[2]; - size_t len; - int ret; - - _enter(""); - - sin.sin_family = AF_INET; - sin.sin_port = udp_hdr(skb)->source; - sin.sin_addr.s_addr = ip_hdr(skb)->saddr; - - msg.msg_name = &sin; - msg.msg_namelen = sizeof(sin); - msg.msg_control = NULL; - msg.msg_controllen = 0; - msg.msg_flags = 0; - - whdr.epoch = htonl(sp->hdr.epoch); - whdr.cid = htonl(sp->hdr.cid); - whdr.callNumber = htonl(sp->hdr.callNumber); - whdr.seq = 0; - whdr.serial = 0; - whdr.type = RXRPC_PACKET_TYPE_VERSION; - whdr.flags = RXRPC_LAST_PACKET | (~hdr->flags & RXRPC_CLIENT_INITIATED); - whdr.userStatus = 0; - whdr.securityIndex = 0; - whdr._rsvd = 0; - whdr.serviceId = htons(sp->hdr.serviceId); - - iov[0].iov_base = &whdr; - iov[0].iov_len = sizeof(whdr); - iov[1].iov_base = (char *)rxrpc_version_string; - iov[1].iov_len = sizeof(rxrpc_version_string); - - len = iov[0].iov_len + iov[1].iov_len; - - _proto("Tx VERSION (reply)"); - - ret = kernel_sendmsg(local->socket, &msg, iov, 2, len); - if (ret < 0) - _debug("sendmsg failed: %d", ret); - - _leave(""); -} - -/* - * Process event packets targetted at a local endpoint. - */ -static void rxrpc_process_local_events(struct work_struct *work) -{ - struct rxrpc_local *local = container_of(work, struct rxrpc_local, event_processor); - struct sk_buff *skb; - char v; - - _enter(""); - - atomic_inc(&local->usage); - - while ((skb = skb_dequeue(&local->event_queue))) { - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - - _debug("{%d},{%u}", local->debug_id, sp->hdr.type); - - switch (sp->hdr.type) { - case RXRPC_PACKET_TYPE_VERSION: - if (skb_copy_bits(skb, 0, &v, 1) < 0) - return; - _proto("Rx VERSION { %02x }", v); - if (v == 0) - rxrpc_send_version_request(local, &sp->hdr, skb); - break; - - default: - /* Just ignore anything we don't understand */ - break; - } - - rxrpc_put_local(local); - rxrpc_free_skb(skb); - } - - rxrpc_put_local(local); - _leave(""); -} diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c deleted file mode 100644 index 2e3c406..0000000 --- a/net/rxrpc/ar-output.c +++ /dev/null @@ -1,724 +0,0 @@ -/* RxRPC packet transmission - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * Time till packet resend (in jiffies). - */ -unsigned int rxrpc_resend_timeout = 4 * HZ; - -static int rxrpc_send_data(struct rxrpc_sock *rx, - struct rxrpc_call *call, - struct msghdr *msg, size_t len); - -/* - * extract control messages from the sendmsg() control buffer - */ -static int rxrpc_sendmsg_cmsg(struct msghdr *msg, - unsigned long *user_call_ID, - enum rxrpc_command *command, - u32 *abort_code) -{ - struct cmsghdr *cmsg; - bool got_user_ID = false; - int len; - - *command = RXRPC_CMD_SEND_DATA; - - if (msg->msg_controllen == 0) - return -EINVAL; - - for_each_cmsghdr(cmsg, msg) { - if (!CMSG_OK(msg, cmsg)) - return -EINVAL; - - len = cmsg->cmsg_len - CMSG_ALIGN(sizeof(struct cmsghdr)); - _debug("CMSG %d, %d, %d", - cmsg->cmsg_level, cmsg->cmsg_type, len); - - if (cmsg->cmsg_level != SOL_RXRPC) - continue; - - switch (cmsg->cmsg_type) { - case RXRPC_USER_CALL_ID: - if (msg->msg_flags & MSG_CMSG_COMPAT) { - if (len != sizeof(u32)) - return -EINVAL; - *user_call_ID = *(u32 *) CMSG_DATA(cmsg); - } else { - if (len != sizeof(unsigned long)) - return -EINVAL; - *user_call_ID = *(unsigned long *) - CMSG_DATA(cmsg); - } - _debug("User Call ID %lx", *user_call_ID); - got_user_ID = true; - break; - - case RXRPC_ABORT: - if (*command != RXRPC_CMD_SEND_DATA) - return -EINVAL; - *command = RXRPC_CMD_SEND_ABORT; - if (len != sizeof(*abort_code)) - return -EINVAL; - *abort_code = *(unsigned int *) CMSG_DATA(cmsg); - _debug("Abort %x", *abort_code); - if (*abort_code == 0) - return -EINVAL; - break; - - case RXRPC_ACCEPT: - if (*command != RXRPC_CMD_SEND_DATA) - return -EINVAL; - *command = RXRPC_CMD_ACCEPT; - if (len != 0) - return -EINVAL; - break; - - default: - return -EINVAL; - } - } - - if (!got_user_ID) - return -EINVAL; - _leave(" = 0"); - return 0; -} - -/* - * abort a call, sending an ABORT packet to the peer - */ -static void rxrpc_send_abort(struct rxrpc_call *call, u32 abort_code) -{ - write_lock_bh(&call->state_lock); - - if (call->state <= RXRPC_CALL_COMPLETE) { - call->state = RXRPC_CALL_LOCALLY_ABORTED; - call->local_abort = abort_code; - set_bit(RXRPC_CALL_EV_ABORT, &call->events); - del_timer_sync(&call->resend_timer); - del_timer_sync(&call->ack_timer); - clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events); - clear_bit(RXRPC_CALL_EV_ACK, &call->events); - clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); - rxrpc_queue_call(call); - } - - write_unlock_bh(&call->state_lock); -} - -/* - * Create a new client call for sendmsg(). - */ -static struct rxrpc_call * -rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, - unsigned long user_call_ID) -{ - struct rxrpc_conn_bundle *bundle; - struct rxrpc_transport *trans; - struct rxrpc_call *call; - struct key *key; - long ret; - - DECLARE_SOCKADDR(struct sockaddr_rxrpc *, srx, msg->msg_name); - - _enter(""); - - if (!msg->msg_name) - return ERR_PTR(-EDESTADDRREQ); - - trans = rxrpc_name_to_transport(rx, msg->msg_name, msg->msg_namelen, 0, - GFP_KERNEL); - if (IS_ERR(trans)) { - ret = PTR_ERR(trans); - goto out; - } - - key = rx->key; - if (key && !rx->key->payload.data[0]) - key = NULL; - bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, GFP_KERNEL); - if (IS_ERR(bundle)) { - ret = PTR_ERR(bundle); - goto out_trans; - } - - call = rxrpc_new_client_call(rx, trans, bundle, user_call_ID, - GFP_KERNEL); - rxrpc_put_bundle(trans, bundle); - rxrpc_put_transport(trans); - if (IS_ERR(call)) { - ret = PTR_ERR(call); - goto out_trans; - } - - _leave(" = %p\n", call); - return call; - -out_trans: - rxrpc_put_transport(trans); -out: - _leave(" = %ld", ret); - return ERR_PTR(ret); -} - -/* - * send a message forming part of a client call through an RxRPC socket - * - caller holds the socket locked - * - the socket may be either a client socket or a server socket - */ -int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) -{ - enum rxrpc_command cmd; - struct rxrpc_call *call; - unsigned long user_call_ID = 0; - u32 abort_code = 0; - int ret; - - _enter(""); - - ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code); - if (ret < 0) - return ret; - - if (cmd == RXRPC_CMD_ACCEPT) { - if (rx->sk.sk_state != RXRPC_SERVER_LISTENING) - return -EINVAL; - call = rxrpc_accept_call(rx, user_call_ID); - if (IS_ERR(call)) - return PTR_ERR(call); - rxrpc_put_call(call); - return 0; - } - - call = rxrpc_find_call_by_user_ID(rx, user_call_ID); - if (!call) { - if (cmd != RXRPC_CMD_SEND_DATA) - return -EBADSLT; - call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID); - if (IS_ERR(call)) - return PTR_ERR(call); - } - - _debug("CALL %d USR %lx ST %d on CONN %p", - call->debug_id, call->user_call_ID, call->state, call->conn); - - if (call->state >= RXRPC_CALL_COMPLETE) { - /* it's too late for this call */ - ret = -ECONNRESET; - } else if (cmd == RXRPC_CMD_SEND_ABORT) { - rxrpc_send_abort(call, abort_code); - ret = 0; - } else if (cmd != RXRPC_CMD_SEND_DATA) { - ret = -EINVAL; - } else if (!call->in_clientflag && - call->state != RXRPC_CALL_CLIENT_SEND_REQUEST) { - /* request phase complete for this client call */ - ret = -EPROTO; - } else if (call->in_clientflag && - call->state != RXRPC_CALL_SERVER_ACK_REQUEST && - call->state != RXRPC_CALL_SERVER_SEND_REPLY) { - /* Reply phase not begun or not complete for service call. */ - ret = -EPROTO; - } else { - ret = rxrpc_send_data(rx, call, msg, len); - } - - rxrpc_put_call(call); - _leave(" = %d", ret); - return ret; -} - -/** - * rxrpc_kernel_send_data - Allow a kernel service to send data on a call - * @call: The call to send data through - * @msg: The data to send - * @len: The amount of data to send - * - * Allow a kernel service to send data on a call. The call must be in an state - * appropriate to sending data. No control data should be supplied in @msg, - * nor should an address be supplied. MSG_MORE should be flagged if there's - * more data to come, otherwise this data will end the transmission phase. - */ -int rxrpc_kernel_send_data(struct rxrpc_call *call, struct msghdr *msg, - size_t len) -{ - int ret; - - _enter("{%d,%s},", call->debug_id, rxrpc_call_states[call->state]); - - ASSERTCMP(msg->msg_name, ==, NULL); - ASSERTCMP(msg->msg_control, ==, NULL); - - lock_sock(&call->socket->sk); - - _debug("CALL %d USR %lx ST %d on CONN %p", - call->debug_id, call->user_call_ID, call->state, call->conn); - - if (call->state >= RXRPC_CALL_COMPLETE) { - ret = -ESHUTDOWN; /* it's too late for this call */ - } else if (call->state != RXRPC_CALL_CLIENT_SEND_REQUEST && - call->state != RXRPC_CALL_SERVER_ACK_REQUEST && - call->state != RXRPC_CALL_SERVER_SEND_REPLY) { - ret = -EPROTO; /* request phase complete for this client call */ - } else { - ret = rxrpc_send_data(call->socket, call, msg, len); - } - - release_sock(&call->socket->sk); - _leave(" = %d", ret); - return ret; -} - -EXPORT_SYMBOL(rxrpc_kernel_send_data); - -/** - * rxrpc_kernel_abort_call - Allow a kernel service to abort a call - * @call: The call to be aborted - * @abort_code: The abort code to stick into the ABORT packet - * - * Allow a kernel service to abort a call, if it's still in an abortable state. - */ -void rxrpc_kernel_abort_call(struct rxrpc_call *call, u32 abort_code) -{ - _enter("{%d},%d", call->debug_id, abort_code); - - lock_sock(&call->socket->sk); - - _debug("CALL %d USR %lx ST %d on CONN %p", - call->debug_id, call->user_call_ID, call->state, call->conn); - - if (call->state < RXRPC_CALL_COMPLETE) - rxrpc_send_abort(call, abort_code); - - release_sock(&call->socket->sk); - _leave(""); -} - -EXPORT_SYMBOL(rxrpc_kernel_abort_call); - -/* - * send a packet through the transport endpoint - */ -int rxrpc_send_packet(struct rxrpc_transport *trans, struct sk_buff *skb) -{ - struct kvec iov[1]; - struct msghdr msg; - int ret, opt; - - _enter(",{%d}", skb->len); - - iov[0].iov_base = skb->head; - iov[0].iov_len = skb->len; - - msg.msg_name = &trans->peer->srx.transport.sin; - msg.msg_namelen = sizeof(trans->peer->srx.transport.sin); - msg.msg_control = NULL; - msg.msg_controllen = 0; - msg.msg_flags = 0; - - /* send the packet with the don't fragment bit set if we currently - * think it's small enough */ - if (skb->len - sizeof(struct rxrpc_wire_header) < trans->peer->maxdata) { - down_read(&trans->local->defrag_sem); - /* send the packet by UDP - * - returns -EMSGSIZE if UDP would have to fragment the packet - * to go out of the interface - * - in which case, we'll have processed the ICMP error - * message and update the peer record - */ - ret = kernel_sendmsg(trans->local->socket, &msg, iov, 1, - iov[0].iov_len); - - up_read(&trans->local->defrag_sem); - if (ret == -EMSGSIZE) - goto send_fragmentable; - - _leave(" = %d [%u]", ret, trans->peer->maxdata); - return ret; - } - -send_fragmentable: - /* attempt to send this message with fragmentation enabled */ - _debug("send fragment"); - - down_write(&trans->local->defrag_sem); - opt = IP_PMTUDISC_DONT; - ret = kernel_setsockopt(trans->local->socket, SOL_IP, IP_MTU_DISCOVER, - (char *) &opt, sizeof(opt)); - if (ret == 0) { - ret = kernel_sendmsg(trans->local->socket, &msg, iov, 1, - iov[0].iov_len); - - opt = IP_PMTUDISC_DO; - kernel_setsockopt(trans->local->socket, SOL_IP, - IP_MTU_DISCOVER, (char *) &opt, sizeof(opt)); - } - - up_write(&trans->local->defrag_sem); - _leave(" = %d [frag %u]", ret, trans->peer->maxdata); - return ret; -} - -/* - * wait for space to appear in the transmit/ACK window - * - caller holds the socket locked - */ -static int rxrpc_wait_for_tx_window(struct rxrpc_sock *rx, - struct rxrpc_call *call, - long *timeo) -{ - DECLARE_WAITQUEUE(myself, current); - int ret; - - _enter(",{%d},%ld", - CIRC_SPACE(call->acks_head, ACCESS_ONCE(call->acks_tail), - call->acks_winsz), - *timeo); - - add_wait_queue(&call->tx_waitq, &myself); - - for (;;) { - set_current_state(TASK_INTERRUPTIBLE); - ret = 0; - if (CIRC_SPACE(call->acks_head, ACCESS_ONCE(call->acks_tail), - call->acks_winsz) > 0) - break; - if (signal_pending(current)) { - ret = sock_intr_errno(*timeo); - break; - } - - release_sock(&rx->sk); - *timeo = schedule_timeout(*timeo); - lock_sock(&rx->sk); - } - - remove_wait_queue(&call->tx_waitq, &myself); - set_current_state(TASK_RUNNING); - _leave(" = %d", ret); - return ret; -} - -/* - * attempt to schedule an instant Tx resend - */ -static inline void rxrpc_instant_resend(struct rxrpc_call *call) -{ - read_lock_bh(&call->state_lock); - if (try_to_del_timer_sync(&call->resend_timer) >= 0) { - clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); - if (call->state < RXRPC_CALL_COMPLETE && - !test_and_set_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events)) - rxrpc_queue_call(call); - } - read_unlock_bh(&call->state_lock); -} - -/* - * queue a packet for transmission, set the resend timer and attempt - * to send the packet immediately - */ -static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb, - bool last) -{ - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - int ret; - - _net("queue skb %p [%d]", skb, call->acks_head); - - ASSERT(call->acks_window != NULL); - call->acks_window[call->acks_head] = (unsigned long) skb; - smp_wmb(); - call->acks_head = (call->acks_head + 1) & (call->acks_winsz - 1); - - if (last || call->state == RXRPC_CALL_SERVER_ACK_REQUEST) { - _debug("________awaiting reply/ACK__________"); - write_lock_bh(&call->state_lock); - switch (call->state) { - case RXRPC_CALL_CLIENT_SEND_REQUEST: - call->state = RXRPC_CALL_CLIENT_AWAIT_REPLY; - break; - case RXRPC_CALL_SERVER_ACK_REQUEST: - call->state = RXRPC_CALL_SERVER_SEND_REPLY; - if (!last) - break; - case RXRPC_CALL_SERVER_SEND_REPLY: - call->state = RXRPC_CALL_SERVER_AWAIT_ACK; - break; - default: - break; - } - write_unlock_bh(&call->state_lock); - } - - _proto("Tx DATA %%%u { #%u }", sp->hdr.serial, sp->hdr.seq); - - sp->need_resend = false; - sp->resend_at = jiffies + rxrpc_resend_timeout; - if (!test_and_set_bit(RXRPC_CALL_RUN_RTIMER, &call->flags)) { - _debug("run timer"); - call->resend_timer.expires = sp->resend_at; - add_timer(&call->resend_timer); - } - - /* attempt to cancel the rx-ACK timer, deferring reply transmission if - * we're ACK'ing the request phase of an incoming call */ - ret = -EAGAIN; - if (try_to_del_timer_sync(&call->ack_timer) >= 0) { - /* the packet may be freed by rxrpc_process_call() before this - * returns */ - ret = rxrpc_send_packet(call->conn->trans, skb); - _net("sent skb %p", skb); - } else { - _debug("failed to delete ACK timer"); - } - - if (ret < 0) { - _debug("need instant resend %d", ret); - sp->need_resend = true; - rxrpc_instant_resend(call); - } - - _leave(""); -} - -/* - * Convert a host-endian header into a network-endian header. - */ -static void rxrpc_insert_header(struct sk_buff *skb) -{ - struct rxrpc_wire_header whdr; - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - - whdr.epoch = htonl(sp->hdr.epoch); - whdr.cid = htonl(sp->hdr.cid); - whdr.callNumber = htonl(sp->hdr.callNumber); - whdr.seq = htonl(sp->hdr.seq); - whdr.serial = htonl(sp->hdr.serial); - whdr.type = sp->hdr.type; - whdr.flags = sp->hdr.flags; - whdr.userStatus = sp->hdr.userStatus; - whdr.securityIndex = sp->hdr.securityIndex; - whdr._rsvd = htons(sp->hdr._rsvd); - whdr.serviceId = htons(sp->hdr.serviceId); - - memcpy(skb->head, &whdr, sizeof(whdr)); -} - -/* - * send data through a socket - * - must be called in process context - * - caller holds the socket locked - */ -static int rxrpc_send_data(struct rxrpc_sock *rx, - struct rxrpc_call *call, - struct msghdr *msg, size_t len) -{ - struct rxrpc_skb_priv *sp; - struct sk_buff *skb; - struct sock *sk = &rx->sk; - long timeo; - bool more; - int ret, copied; - - timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT); - - /* this should be in poll */ - sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk); - - if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN)) - return -EPIPE; - - more = msg->msg_flags & MSG_MORE; - - skb = call->tx_pending; - call->tx_pending = NULL; - - copied = 0; - do { - if (!skb) { - size_t size, chunk, max, space; - - _debug("alloc"); - - if (CIRC_SPACE(call->acks_head, - ACCESS_ONCE(call->acks_tail), - call->acks_winsz) <= 0) { - ret = -EAGAIN; - if (msg->msg_flags & MSG_DONTWAIT) - goto maybe_error; - ret = rxrpc_wait_for_tx_window(rx, call, - &timeo); - if (ret < 0) - goto maybe_error; - } - - max = call->conn->trans->peer->maxdata; - max -= call->conn->security_size; - max &= ~(call->conn->size_align - 1UL); - - chunk = max; - if (chunk > msg_data_left(msg) && !more) - chunk = msg_data_left(msg); - - space = chunk + call->conn->size_align; - space &= ~(call->conn->size_align - 1UL); - - size = space + call->conn->header_size; - - _debug("SIZE: %zu/%zu/%zu", chunk, space, size); - - /* create a buffer that we can retain until it's ACK'd */ - skb = sock_alloc_send_skb( - sk, size, msg->msg_flags & MSG_DONTWAIT, &ret); - if (!skb) - goto maybe_error; - - rxrpc_new_skb(skb); - - _debug("ALLOC SEND %p", skb); - - ASSERTCMP(skb->mark, ==, 0); - - _debug("HS: %u", call->conn->header_size); - skb_reserve(skb, call->conn->header_size); - skb->len += call->conn->header_size; - - sp = rxrpc_skb(skb); - sp->remain = chunk; - if (sp->remain > skb_tailroom(skb)) - sp->remain = skb_tailroom(skb); - - _net("skb: hr %d, tr %d, hl %d, rm %d", - skb_headroom(skb), - skb_tailroom(skb), - skb_headlen(skb), - sp->remain); - - skb->ip_summed = CHECKSUM_UNNECESSARY; - } - - _debug("append"); - sp = rxrpc_skb(skb); - - /* append next segment of data to the current buffer */ - if (msg_data_left(msg) > 0) { - int copy = skb_tailroom(skb); - ASSERTCMP(copy, >, 0); - if (copy > msg_data_left(msg)) - copy = msg_data_left(msg); - if (copy > sp->remain) - copy = sp->remain; - - _debug("add"); - ret = skb_add_data(skb, &msg->msg_iter, copy); - _debug("added"); - if (ret < 0) - goto efault; - sp->remain -= copy; - skb->mark += copy; - copied += copy; - } - - /* check for the far side aborting the call or a network error - * occurring */ - if (call->state > RXRPC_CALL_COMPLETE) - goto call_aborted; - - /* add the packet to the send queue if it's now full */ - if (sp->remain <= 0 || - (msg_data_left(msg) == 0 && !more)) { - struct rxrpc_connection *conn = call->conn; - uint32_t seq; - size_t pad; - - /* pad out if we're using security */ - if (conn->security_ix) { - pad = conn->security_size + skb->mark; - pad = conn->size_align - pad; - pad &= conn->size_align - 1; - _debug("pad %zu", pad); - if (pad) - memset(skb_put(skb, pad), 0, pad); - } - - seq = atomic_inc_return(&call->sequence); - - sp->hdr.epoch = conn->epoch; - sp->hdr.cid = call->cid; - sp->hdr.callNumber = call->call_id; - sp->hdr.seq = seq; - sp->hdr.serial = atomic_inc_return(&conn->serial); - sp->hdr.type = RXRPC_PACKET_TYPE_DATA; - sp->hdr.userStatus = 0; - sp->hdr.securityIndex = conn->security_ix; - sp->hdr._rsvd = 0; - sp->hdr.serviceId = call->service_id; - - sp->hdr.flags = conn->out_clientflag; - if (msg_data_left(msg) == 0 && !more) - sp->hdr.flags |= RXRPC_LAST_PACKET; - else if (CIRC_SPACE(call->acks_head, - ACCESS_ONCE(call->acks_tail), - call->acks_winsz) > 1) - sp->hdr.flags |= RXRPC_MORE_PACKETS; - if (more && seq & 1) - sp->hdr.flags |= RXRPC_REQUEST_ACK; - - ret = conn->security->secure_packet( - call, skb, skb->mark, - skb->head + sizeof(struct rxrpc_wire_header)); - if (ret < 0) - goto out; - - rxrpc_insert_header(skb); - rxrpc_queue_packet(call, skb, !msg_data_left(msg) && !more); - skb = NULL; - } - } while (msg_data_left(msg) > 0); - -success: - ret = copied; -out: - call->tx_pending = skb; - _leave(" = %d", ret); - return ret; - -call_aborted: - rxrpc_free_skb(skb); - if (call->state == RXRPC_CALL_NETWORK_ERROR) - ret = call->conn->trans->peer->net_error; - else - ret = -ECONNABORTED; - _leave(" = %d", ret); - return ret; - -maybe_error: - if (copied) - goto success; - goto out; - -efault: - ret = -EFAULT; - goto out; -} diff --git a/net/rxrpc/ar-peer.c b/net/rxrpc/ar-peer.c deleted file mode 100644 index 0b54cda..0000000 --- a/net/rxrpc/ar-peer.c +++ /dev/null @@ -1,305 +0,0 @@ -/* RxRPC remote transport endpoint management - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -static LIST_HEAD(rxrpc_peers); -static DEFINE_RWLOCK(rxrpc_peer_lock); -static DECLARE_WAIT_QUEUE_HEAD(rxrpc_peer_wq); - -static void rxrpc_destroy_peer(struct work_struct *work); - -/* - * assess the MTU size for the network interface through which this peer is - * reached - */ -static void rxrpc_assess_MTU_size(struct rxrpc_peer *peer) -{ - struct rtable *rt; - struct flowi4 fl4; - - peer->if_mtu = 1500; - - rt = ip_route_output_ports(&init_net, &fl4, NULL, - peer->srx.transport.sin.sin_addr.s_addr, 0, - htons(7000), htons(7001), - IPPROTO_UDP, 0, 0); - if (IS_ERR(rt)) { - _leave(" [route err %ld]", PTR_ERR(rt)); - return; - } - - peer->if_mtu = dst_mtu(&rt->dst); - dst_release(&rt->dst); - - _leave(" [if_mtu %u]", peer->if_mtu); -} - -/* - * allocate a new peer - */ -static struct rxrpc_peer *rxrpc_alloc_peer(struct sockaddr_rxrpc *srx, - gfp_t gfp) -{ - struct rxrpc_peer *peer; - - _enter(""); - - peer = kzalloc(sizeof(struct rxrpc_peer), gfp); - if (peer) { - INIT_WORK(&peer->destroyer, &rxrpc_destroy_peer); - INIT_LIST_HEAD(&peer->link); - INIT_LIST_HEAD(&peer->error_targets); - spin_lock_init(&peer->lock); - atomic_set(&peer->usage, 1); - peer->debug_id = atomic_inc_return(&rxrpc_debug_id); - memcpy(&peer->srx, srx, sizeof(*srx)); - - rxrpc_assess_MTU_size(peer); - peer->mtu = peer->if_mtu; - - if (srx->transport.family == AF_INET) { - peer->hdrsize = sizeof(struct iphdr); - switch (srx->transport_type) { - case SOCK_DGRAM: - peer->hdrsize += sizeof(struct udphdr); - break; - default: - BUG(); - break; - } - } else { - BUG(); - } - - peer->hdrsize += sizeof(struct rxrpc_wire_header); - peer->maxdata = peer->mtu - peer->hdrsize; - } - - _leave(" = %p", peer); - return peer; -} - -/* - * obtain a remote transport endpoint for the specified address - */ -struct rxrpc_peer *rxrpc_get_peer(struct sockaddr_rxrpc *srx, gfp_t gfp) -{ - struct rxrpc_peer *peer, *candidate; - const char *new = "old"; - int usage; - - _enter("{%d,%d,%pI4+%hu}", - srx->transport_type, - srx->transport_len, - &srx->transport.sin.sin_addr, - ntohs(srx->transport.sin.sin_port)); - - /* search the peer list first */ - read_lock_bh(&rxrpc_peer_lock); - list_for_each_entry(peer, &rxrpc_peers, link) { - _debug("check PEER %d { u=%d t=%d l=%d }", - peer->debug_id, - atomic_read(&peer->usage), - peer->srx.transport_type, - peer->srx.transport_len); - - if (atomic_read(&peer->usage) > 0 && - peer->srx.transport_type == srx->transport_type && - peer->srx.transport_len == srx->transport_len && - memcmp(&peer->srx.transport, - &srx->transport, - srx->transport_len) == 0) - goto found_extant_peer; - } - read_unlock_bh(&rxrpc_peer_lock); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_peer(srx, gfp); - if (!candidate) { - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } - - write_lock_bh(&rxrpc_peer_lock); - - list_for_each_entry(peer, &rxrpc_peers, link) { - if (atomic_read(&peer->usage) > 0 && - peer->srx.transport_type == srx->transport_type && - peer->srx.transport_len == srx->transport_len && - memcmp(&peer->srx.transport, - &srx->transport, - srx->transport_len) == 0) - goto found_extant_second; - } - - /* we can now add the new candidate to the list */ - peer = candidate; - candidate = NULL; - usage = atomic_read(&peer->usage); - - list_add_tail(&peer->link, &rxrpc_peers); - write_unlock_bh(&rxrpc_peer_lock); - new = "new"; - -success: - _net("PEER %s %d {%d,%u,%pI4+%hu}", - new, - peer->debug_id, - peer->srx.transport_type, - peer->srx.transport.family, - &peer->srx.transport.sin.sin_addr, - ntohs(peer->srx.transport.sin.sin_port)); - - _leave(" = %p {u=%d}", peer, usage); - return peer; - - /* we found the peer in the list immediately */ -found_extant_peer: - usage = atomic_inc_return(&peer->usage); - read_unlock_bh(&rxrpc_peer_lock); - goto success; - - /* we found the peer on the second time through the list */ -found_extant_second: - usage = atomic_inc_return(&peer->usage); - write_unlock_bh(&rxrpc_peer_lock); - kfree(candidate); - goto success; -} - -/* - * find the peer associated with a packet - */ -struct rxrpc_peer *rxrpc_find_peer(struct rxrpc_local *local, - __be32 addr, __be16 port) -{ - struct rxrpc_peer *peer; - - _enter(""); - - /* search the peer list */ - read_lock_bh(&rxrpc_peer_lock); - - if (local->srx.transport.family == AF_INET && - local->srx.transport_type == SOCK_DGRAM - ) { - list_for_each_entry(peer, &rxrpc_peers, link) { - if (atomic_read(&peer->usage) > 0 && - peer->srx.transport_type == SOCK_DGRAM && - peer->srx.transport.family == AF_INET && - peer->srx.transport.sin.sin_port == port && - peer->srx.transport.sin.sin_addr.s_addr == addr) - goto found_UDP_peer; - } - - goto new_UDP_peer; - } - - read_unlock_bh(&rxrpc_peer_lock); - _leave(" = -EAFNOSUPPORT"); - return ERR_PTR(-EAFNOSUPPORT); - -found_UDP_peer: - _net("Rx UDP DGRAM from peer %d", peer->debug_id); - atomic_inc(&peer->usage); - read_unlock_bh(&rxrpc_peer_lock); - _leave(" = %p", peer); - return peer; - -new_UDP_peer: - _net("Rx UDP DGRAM from NEW peer"); - read_unlock_bh(&rxrpc_peer_lock); - _leave(" = -EBUSY [new]"); - return ERR_PTR(-EBUSY); -} - -/* - * release a remote transport endpoint - */ -void rxrpc_put_peer(struct rxrpc_peer *peer) -{ - _enter("%p{u=%d}", peer, atomic_read(&peer->usage)); - - ASSERTCMP(atomic_read(&peer->usage), >, 0); - - if (likely(!atomic_dec_and_test(&peer->usage))) { - _leave(" [in use]"); - return; - } - - rxrpc_queue_work(&peer->destroyer); - _leave(""); -} - -/* - * destroy a remote transport endpoint - */ -static void rxrpc_destroy_peer(struct work_struct *work) -{ - struct rxrpc_peer *peer = - container_of(work, struct rxrpc_peer, destroyer); - - _enter("%p{%d}", peer, atomic_read(&peer->usage)); - - write_lock_bh(&rxrpc_peer_lock); - list_del(&peer->link); - write_unlock_bh(&rxrpc_peer_lock); - - _net("DESTROY PEER %d", peer->debug_id); - kfree(peer); - - if (list_empty(&rxrpc_peers)) - wake_up_all(&rxrpc_peer_wq); - _leave(""); -} - -/* - * preemptively destroy all the peer records from a transport endpoint rather - * than waiting for them to time out - */ -void __exit rxrpc_destroy_all_peers(void) -{ - DECLARE_WAITQUEUE(myself,current); - - _enter(""); - - /* we simply have to wait for them to go away */ - if (!list_empty(&rxrpc_peers)) { - set_current_state(TASK_UNINTERRUPTIBLE); - add_wait_queue(&rxrpc_peer_wq, &myself); - - while (!list_empty(&rxrpc_peers)) { - schedule(); - set_current_state(TASK_UNINTERRUPTIBLE); - } - - remove_wait_queue(&rxrpc_peer_wq, &myself); - set_current_state(TASK_RUNNING); - } - - _leave(""); -} diff --git a/net/rxrpc/ar-proc.c b/net/rxrpc/ar-proc.c deleted file mode 100644 index 225163b..0000000 --- a/net/rxrpc/ar-proc.c +++ /dev/null @@ -1,192 +0,0 @@ -/* /proc/net/ support for AF_RXRPC - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include -#include -#include -#include "ar-internal.h" - -static const char *const rxrpc_conn_states[] = { - [RXRPC_CONN_UNUSED] = "Unused ", - [RXRPC_CONN_CLIENT] = "Client ", - [RXRPC_CONN_SERVER_UNSECURED] = "SvUnsec ", - [RXRPC_CONN_SERVER_CHALLENGING] = "SvChall ", - [RXRPC_CONN_SERVER] = "SvSecure", - [RXRPC_CONN_REMOTELY_ABORTED] = "RmtAbort", - [RXRPC_CONN_LOCALLY_ABORTED] = "LocAbort", - [RXRPC_CONN_NETWORK_ERROR] = "NetError", -}; - -/* - * generate a list of extant and dead calls in /proc/net/rxrpc_calls - */ -static void *rxrpc_call_seq_start(struct seq_file *seq, loff_t *_pos) -{ - read_lock(&rxrpc_call_lock); - return seq_list_start_head(&rxrpc_calls, *_pos); -} - -static void *rxrpc_call_seq_next(struct seq_file *seq, void *v, loff_t *pos) -{ - return seq_list_next(v, &rxrpc_calls, pos); -} - -static void rxrpc_call_seq_stop(struct seq_file *seq, void *v) -{ - read_unlock(&rxrpc_call_lock); -} - -static int rxrpc_call_seq_show(struct seq_file *seq, void *v) -{ - struct rxrpc_transport *trans; - struct rxrpc_call *call; - char lbuff[4 + 4 + 4 + 4 + 5 + 1], rbuff[4 + 4 + 4 + 4 + 5 + 1]; - - if (v == &rxrpc_calls) { - seq_puts(seq, - "Proto Local Remote " - " SvID ConnID CallID End Use State Abort " - " UserID\n"); - return 0; - } - - call = list_entry(v, struct rxrpc_call, link); - trans = call->conn->trans; - - sprintf(lbuff, "%pI4:%u", - &trans->local->srx.transport.sin.sin_addr, - ntohs(trans->local->srx.transport.sin.sin_port)); - - sprintf(rbuff, "%pI4:%u", - &trans->peer->srx.transport.sin.sin_addr, - ntohs(trans->peer->srx.transport.sin.sin_port)); - - seq_printf(seq, - "UDP %-22.22s %-22.22s %4x %08x %08x %s %3u" - " %-8.8s %08x %lx\n", - lbuff, - rbuff, - call->conn->service_id, - call->cid, - call->call_id, - call->conn->in_clientflag ? "Svc" : "Clt", - atomic_read(&call->usage), - rxrpc_call_states[call->state], - call->remote_abort ?: call->local_abort, - call->user_call_ID); - - return 0; -} - -static const struct seq_operations rxrpc_call_seq_ops = { - .start = rxrpc_call_seq_start, - .next = rxrpc_call_seq_next, - .stop = rxrpc_call_seq_stop, - .show = rxrpc_call_seq_show, -}; - -static int rxrpc_call_seq_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &rxrpc_call_seq_ops); -} - -const struct file_operations rxrpc_call_seq_fops = { - .owner = THIS_MODULE, - .open = rxrpc_call_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; - -/* - * generate a list of extant virtual connections in /proc/net/rxrpc_conns - */ -static void *rxrpc_connection_seq_start(struct seq_file *seq, loff_t *_pos) -{ - read_lock(&rxrpc_connection_lock); - return seq_list_start_head(&rxrpc_connections, *_pos); -} - -static void *rxrpc_connection_seq_next(struct seq_file *seq, void *v, - loff_t *pos) -{ - return seq_list_next(v, &rxrpc_connections, pos); -} - -static void rxrpc_connection_seq_stop(struct seq_file *seq, void *v) -{ - read_unlock(&rxrpc_connection_lock); -} - -static int rxrpc_connection_seq_show(struct seq_file *seq, void *v) -{ - struct rxrpc_connection *conn; - struct rxrpc_transport *trans; - char lbuff[4 + 4 + 4 + 4 + 5 + 1], rbuff[4 + 4 + 4 + 4 + 5 + 1]; - - if (v == &rxrpc_connections) { - seq_puts(seq, - "Proto Local Remote " - " SvID ConnID Calls End Use State Key " - " Serial ISerial\n" - ); - return 0; - } - - conn = list_entry(v, struct rxrpc_connection, link); - trans = conn->trans; - - sprintf(lbuff, "%pI4:%u", - &trans->local->srx.transport.sin.sin_addr, - ntohs(trans->local->srx.transport.sin.sin_port)); - - sprintf(rbuff, "%pI4:%u", - &trans->peer->srx.transport.sin.sin_addr, - ntohs(trans->peer->srx.transport.sin.sin_port)); - - seq_printf(seq, - "UDP %-22.22s %-22.22s %4x %08x %08x %s %3u" - " %s %08x %08x %08x\n", - lbuff, - rbuff, - conn->service_id, - conn->cid, - conn->call_counter, - conn->in_clientflag ? "Svc" : "Clt", - atomic_read(&conn->usage), - rxrpc_conn_states[conn->state], - key_serial(conn->key), - atomic_read(&conn->serial), - atomic_read(&conn->hi_serial)); - - return 0; -} - -static const struct seq_operations rxrpc_connection_seq_ops = { - .start = rxrpc_connection_seq_start, - .next = rxrpc_connection_seq_next, - .stop = rxrpc_connection_seq_stop, - .show = rxrpc_connection_seq_show, -}; - - -static int rxrpc_connection_seq_open(struct inode *inode, struct file *file) -{ - return seq_open(file, &rxrpc_connection_seq_ops); -} - -const struct file_operations rxrpc_connection_seq_fops = { - .owner = THIS_MODULE, - .open = rxrpc_connection_seq_open, - .read = seq_read, - .llseek = seq_lseek, - .release = seq_release, -}; diff --git a/net/rxrpc/ar-recvmsg.c b/net/rxrpc/ar-recvmsg.c deleted file mode 100644 index 59706b9..0000000 --- a/net/rxrpc/ar-recvmsg.c +++ /dev/null @@ -1,436 +0,0 @@ -/* RxRPC recvmsg() implementation - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * removal a call's user ID from the socket tree to make the user ID available - * again and so that it won't be seen again in association with that call - */ -void rxrpc_remove_user_ID(struct rxrpc_sock *rx, struct rxrpc_call *call) -{ - _debug("RELEASE CALL %d", call->debug_id); - - if (test_bit(RXRPC_CALL_HAS_USERID, &call->flags)) { - write_lock_bh(&rx->call_lock); - rb_erase(&call->sock_node, &call->socket->calls); - clear_bit(RXRPC_CALL_HAS_USERID, &call->flags); - write_unlock_bh(&rx->call_lock); - } - - read_lock_bh(&call->state_lock); - if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && - !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) - rxrpc_queue_call(call); - read_unlock_bh(&call->state_lock); -} - -/* - * receive a message from an RxRPC socket - * - we need to be careful about two or more threads calling recvmsg - * simultaneously - */ -int rxrpc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, - int flags) -{ - struct rxrpc_skb_priv *sp; - struct rxrpc_call *call = NULL, *continue_call = NULL; - struct rxrpc_sock *rx = rxrpc_sk(sock->sk); - struct sk_buff *skb; - long timeo; - int copy, ret, ullen, offset, copied = 0; - u32 abort_code; - - DEFINE_WAIT(wait); - - _enter(",,,%zu,%d", len, flags); - - if (flags & (MSG_OOB | MSG_TRUNC)) - return -EOPNOTSUPP; - - ullen = msg->msg_flags & MSG_CMSG_COMPAT ? 4 : sizeof(unsigned long); - - timeo = sock_rcvtimeo(&rx->sk, flags & MSG_DONTWAIT); - msg->msg_flags |= MSG_MORE; - - lock_sock(&rx->sk); - - for (;;) { - /* return immediately if a client socket has no outstanding - * calls */ - if (RB_EMPTY_ROOT(&rx->calls)) { - if (copied) - goto out; - if (rx->sk.sk_state != RXRPC_SERVER_LISTENING) { - release_sock(&rx->sk); - if (continue_call) - rxrpc_put_call(continue_call); - return -ENODATA; - } - } - - /* get the next message on the Rx queue */ - skb = skb_peek(&rx->sk.sk_receive_queue); - if (!skb) { - /* nothing remains on the queue */ - if (copied && - (flags & MSG_PEEK || timeo == 0)) - goto out; - - /* wait for a message to turn up */ - release_sock(&rx->sk); - prepare_to_wait_exclusive(sk_sleep(&rx->sk), &wait, - TASK_INTERRUPTIBLE); - ret = sock_error(&rx->sk); - if (ret) - goto wait_error; - - if (skb_queue_empty(&rx->sk.sk_receive_queue)) { - if (signal_pending(current)) - goto wait_interrupted; - timeo = schedule_timeout(timeo); - } - finish_wait(sk_sleep(&rx->sk), &wait); - lock_sock(&rx->sk); - continue; - } - - peek_next_packet: - sp = rxrpc_skb(skb); - call = sp->call; - ASSERT(call != NULL); - - _debug("next pkt %s", rxrpc_pkts[sp->hdr.type]); - - /* make sure we wait for the state to be updated in this call */ - spin_lock_bh(&call->lock); - spin_unlock_bh(&call->lock); - - if (test_bit(RXRPC_CALL_RELEASED, &call->flags)) { - _debug("packet from released call"); - if (skb_dequeue(&rx->sk.sk_receive_queue) != skb) - BUG(); - rxrpc_free_skb(skb); - continue; - } - - /* determine whether to continue last data receive */ - if (continue_call) { - _debug("maybe cont"); - if (call != continue_call || - skb->mark != RXRPC_SKB_MARK_DATA) { - release_sock(&rx->sk); - rxrpc_put_call(continue_call); - _leave(" = %d [noncont]", copied); - return copied; - } - } - - rxrpc_get_call(call); - - /* copy the peer address and timestamp */ - if (!continue_call) { - if (msg->msg_name) { - size_t len = - sizeof(call->conn->trans->peer->srx); - memcpy(msg->msg_name, - &call->conn->trans->peer->srx, len); - msg->msg_namelen = len; - } - sock_recv_timestamp(msg, &rx->sk, skb); - } - - /* receive the message */ - if (skb->mark != RXRPC_SKB_MARK_DATA) - goto receive_non_data_message; - - _debug("recvmsg DATA #%u { %d, %d }", - sp->hdr.seq, skb->len, sp->offset); - - if (!continue_call) { - /* only set the control data once per recvmsg() */ - ret = put_cmsg(msg, SOL_RXRPC, RXRPC_USER_CALL_ID, - ullen, &call->user_call_ID); - if (ret < 0) - goto copy_error; - ASSERT(test_bit(RXRPC_CALL_HAS_USERID, &call->flags)); - } - - ASSERTCMP(sp->hdr.seq, >=, call->rx_data_recv); - ASSERTCMP(sp->hdr.seq, <=, call->rx_data_recv + 1); - call->rx_data_recv = sp->hdr.seq; - - ASSERTCMP(sp->hdr.seq, >, call->rx_data_eaten); - - offset = sp->offset; - copy = skb->len - offset; - if (copy > len - copied) - copy = len - copied; - - ret = skb_copy_datagram_msg(skb, offset, msg, copy); - - if (ret < 0) - goto copy_error; - - /* handle piecemeal consumption of data packets */ - _debug("copied %d+%d", copy, copied); - - offset += copy; - copied += copy; - - if (!(flags & MSG_PEEK)) - sp->offset = offset; - - if (sp->offset < skb->len) { - _debug("buffer full"); - ASSERTCMP(copied, ==, len); - break; - } - - /* we transferred the whole data packet */ - if (sp->hdr.flags & RXRPC_LAST_PACKET) { - _debug("last"); - if (call->conn->out_clientflag) { - /* last byte of reply received */ - ret = copied; - goto terminal_message; - } - - /* last bit of request received */ - if (!(flags & MSG_PEEK)) { - _debug("eat packet"); - if (skb_dequeue(&rx->sk.sk_receive_queue) != - skb) - BUG(); - rxrpc_free_skb(skb); - } - msg->msg_flags &= ~MSG_MORE; - break; - } - - /* move on to the next data message */ - _debug("next"); - if (!continue_call) - continue_call = sp->call; - else - rxrpc_put_call(call); - call = NULL; - - if (flags & MSG_PEEK) { - _debug("peek next"); - skb = skb->next; - if (skb == (struct sk_buff *) &rx->sk.sk_receive_queue) - break; - goto peek_next_packet; - } - - _debug("eat packet"); - if (skb_dequeue(&rx->sk.sk_receive_queue) != skb) - BUG(); - rxrpc_free_skb(skb); - } - - /* end of non-terminal data packet reception for the moment */ - _debug("end rcv data"); -out: - release_sock(&rx->sk); - if (call) - rxrpc_put_call(call); - if (continue_call) - rxrpc_put_call(continue_call); - _leave(" = %d [data]", copied); - return copied; - - /* handle non-DATA messages such as aborts, incoming connections and - * final ACKs */ -receive_non_data_message: - _debug("non-data"); - - if (skb->mark == RXRPC_SKB_MARK_NEW_CALL) { - _debug("RECV NEW CALL"); - ret = put_cmsg(msg, SOL_RXRPC, RXRPC_NEW_CALL, 0, &abort_code); - if (ret < 0) - goto copy_error; - if (!(flags & MSG_PEEK)) { - if (skb_dequeue(&rx->sk.sk_receive_queue) != skb) - BUG(); - rxrpc_free_skb(skb); - } - goto out; - } - - ret = put_cmsg(msg, SOL_RXRPC, RXRPC_USER_CALL_ID, - ullen, &call->user_call_ID); - if (ret < 0) - goto copy_error; - ASSERT(test_bit(RXRPC_CALL_HAS_USERID, &call->flags)); - - switch (skb->mark) { - case RXRPC_SKB_MARK_DATA: - BUG(); - case RXRPC_SKB_MARK_FINAL_ACK: - ret = put_cmsg(msg, SOL_RXRPC, RXRPC_ACK, 0, &abort_code); - break; - case RXRPC_SKB_MARK_BUSY: - ret = put_cmsg(msg, SOL_RXRPC, RXRPC_BUSY, 0, &abort_code); - break; - case RXRPC_SKB_MARK_REMOTE_ABORT: - abort_code = call->remote_abort; - ret = put_cmsg(msg, SOL_RXRPC, RXRPC_ABORT, 4, &abort_code); - break; - case RXRPC_SKB_MARK_LOCAL_ABORT: - abort_code = call->local_abort; - ret = put_cmsg(msg, SOL_RXRPC, RXRPC_ABORT, 4, &abort_code); - break; - case RXRPC_SKB_MARK_NET_ERROR: - _debug("RECV NET ERROR %d", sp->error); - abort_code = sp->error; - ret = put_cmsg(msg, SOL_RXRPC, RXRPC_NET_ERROR, 4, &abort_code); - break; - case RXRPC_SKB_MARK_LOCAL_ERROR: - _debug("RECV LOCAL ERROR %d", sp->error); - abort_code = sp->error; - ret = put_cmsg(msg, SOL_RXRPC, RXRPC_LOCAL_ERROR, 4, - &abort_code); - break; - default: - pr_err("Unknown packet mark %u\n", skb->mark); - BUG(); - break; - } - - if (ret < 0) - goto copy_error; - -terminal_message: - _debug("terminal"); - msg->msg_flags &= ~MSG_MORE; - msg->msg_flags |= MSG_EOR; - - if (!(flags & MSG_PEEK)) { - _net("free terminal skb %p", skb); - if (skb_dequeue(&rx->sk.sk_receive_queue) != skb) - BUG(); - rxrpc_free_skb(skb); - rxrpc_remove_user_ID(rx, call); - } - - release_sock(&rx->sk); - rxrpc_put_call(call); - if (continue_call) - rxrpc_put_call(continue_call); - _leave(" = %d", ret); - return ret; - -copy_error: - _debug("copy error"); - release_sock(&rx->sk); - rxrpc_put_call(call); - if (continue_call) - rxrpc_put_call(continue_call); - _leave(" = %d", ret); - return ret; - -wait_interrupted: - ret = sock_intr_errno(timeo); -wait_error: - finish_wait(sk_sleep(&rx->sk), &wait); - if (continue_call) - rxrpc_put_call(continue_call); - if (copied) - copied = ret; - _leave(" = %d [waitfail %d]", copied, ret); - return copied; - -} - -/** - * rxrpc_kernel_data_delivered - Record delivery of data message - * @skb: Message holding data - * - * Record the delivery of a data message. This permits RxRPC to keep its - * tracking correct. The socket buffer will be deleted. - */ -void rxrpc_kernel_data_delivered(struct sk_buff *skb) -{ - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - struct rxrpc_call *call = sp->call; - - ASSERTCMP(sp->hdr.seq, >=, call->rx_data_recv); - ASSERTCMP(sp->hdr.seq, <=, call->rx_data_recv + 1); - call->rx_data_recv = sp->hdr.seq; - - ASSERTCMP(sp->hdr.seq, >, call->rx_data_eaten); - rxrpc_free_skb(skb); -} - -EXPORT_SYMBOL(rxrpc_kernel_data_delivered); - -/** - * rxrpc_kernel_is_data_last - Determine if data message is last one - * @skb: Message holding data - * - * Determine if data message is last one for the parent call. - */ -bool rxrpc_kernel_is_data_last(struct sk_buff *skb) -{ - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - - ASSERTCMP(skb->mark, ==, RXRPC_SKB_MARK_DATA); - - return sp->hdr.flags & RXRPC_LAST_PACKET; -} - -EXPORT_SYMBOL(rxrpc_kernel_is_data_last); - -/** - * rxrpc_kernel_get_abort_code - Get the abort code from an RxRPC abort message - * @skb: Message indicating an abort - * - * Get the abort code from an RxRPC abort message. - */ -u32 rxrpc_kernel_get_abort_code(struct sk_buff *skb) -{ - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - - switch (skb->mark) { - case RXRPC_SKB_MARK_REMOTE_ABORT: - return sp->call->remote_abort; - case RXRPC_SKB_MARK_LOCAL_ABORT: - return sp->call->local_abort; - default: - BUG(); - } -} - -EXPORT_SYMBOL(rxrpc_kernel_get_abort_code); - -/** - * rxrpc_kernel_get_error - Get the error number from an RxRPC error message - * @skb: Message indicating an error - * - * Get the error number from an RxRPC error message. - */ -int rxrpc_kernel_get_error_number(struct sk_buff *skb) -{ - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - - return sp->error; -} - -EXPORT_SYMBOL(rxrpc_kernel_get_error_number); diff --git a/net/rxrpc/ar-security.c b/net/rxrpc/ar-security.c deleted file mode 100644 index d223253..0000000 --- a/net/rxrpc/ar-security.c +++ /dev/null @@ -1,168 +0,0 @@ -/* RxRPC security handling - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -static LIST_HEAD(rxrpc_security_methods); -static DECLARE_RWSEM(rxrpc_security_sem); - -static const struct rxrpc_security *rxrpc_security_types[] = { - [RXRPC_SECURITY_NONE] = &rxrpc_no_security, -#ifdef CONFIG_RXKAD - [RXRPC_SECURITY_RXKAD] = &rxkad, -#endif -}; - -int __init rxrpc_init_security(void) -{ - int i, ret; - - for (i = 0; i < ARRAY_SIZE(rxrpc_security_types); i++) { - if (rxrpc_security_types[i]) { - ret = rxrpc_security_types[i]->init(); - if (ret < 0) - goto failed; - } - } - - return 0; - -failed: - for (i--; i >= 0; i--) - if (rxrpc_security_types[i]) - rxrpc_security_types[i]->exit(); - return ret; -} - -void rxrpc_exit_security(void) -{ - int i; - - for (i = 0; i < ARRAY_SIZE(rxrpc_security_types); i++) - if (rxrpc_security_types[i]) - rxrpc_security_types[i]->exit(); -} - -/* - * look up an rxrpc security module - */ -static const struct rxrpc_security *rxrpc_security_lookup(u8 security_index) -{ - if (security_index >= ARRAY_SIZE(rxrpc_security_types)) - return NULL; - return rxrpc_security_types[security_index]; -} - -/* - * initialise the security on a client connection - */ -int rxrpc_init_client_conn_security(struct rxrpc_connection *conn) -{ - const struct rxrpc_security *sec; - struct rxrpc_key_token *token; - struct key *key = conn->key; - int ret; - - _enter("{%d},{%x}", conn->debug_id, key_serial(key)); - - if (!key) - return 0; - - ret = key_validate(key); - if (ret < 0) - return ret; - - token = key->payload.data[0]; - if (!token) - return -EKEYREJECTED; - - sec = rxrpc_security_lookup(token->security_index); - if (!sec) - return -EKEYREJECTED; - conn->security = sec; - - ret = conn->security->init_connection_security(conn); - if (ret < 0) { - conn->security = &rxrpc_no_security; - return ret; - } - - _leave(" = 0"); - return 0; -} - -/* - * initialise the security on a server connection - */ -int rxrpc_init_server_conn_security(struct rxrpc_connection *conn) -{ - const struct rxrpc_security *sec; - struct rxrpc_local *local = conn->trans->local; - struct rxrpc_sock *rx; - struct key *key; - key_ref_t kref; - char kdesc[5 + 1 + 3 + 1]; - - _enter(""); - - sprintf(kdesc, "%u:%u", conn->service_id, conn->security_ix); - - sec = rxrpc_security_lookup(conn->security_ix); - if (!sec) { - _leave(" = -ENOKEY [lookup]"); - return -ENOKEY; - } - - /* find the service */ - read_lock_bh(&local->services_lock); - list_for_each_entry(rx, &local->services, listen_link) { - if (rx->srx.srx_service == conn->service_id) - goto found_service; - } - - /* the service appears to have died */ - read_unlock_bh(&local->services_lock); - _leave(" = -ENOENT"); - return -ENOENT; - -found_service: - if (!rx->securities) { - read_unlock_bh(&local->services_lock); - _leave(" = -ENOKEY"); - return -ENOKEY; - } - - /* look through the service's keyring */ - kref = keyring_search(make_key_ref(rx->securities, 1UL), - &key_type_rxrpc_s, kdesc); - if (IS_ERR(kref)) { - read_unlock_bh(&local->services_lock); - _leave(" = %ld [search]", PTR_ERR(kref)); - return PTR_ERR(kref); - } - - key = key_ref_to_ptr(kref); - read_unlock_bh(&local->services_lock); - - conn->server_key = key; - conn->security = sec; - - _leave(" = 0"); - return 0; -} diff --git a/net/rxrpc/ar-skbuff.c b/net/rxrpc/ar-skbuff.c deleted file mode 100644 index eee0cfd9..0000000 --- a/net/rxrpc/ar-skbuff.c +++ /dev/null @@ -1,138 +0,0 @@ -/* ar-skbuff.c: socket buffer destruction handling - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * set up for the ACK at the end of the receive phase when we discard the final - * receive phase data packet - * - called with softirqs disabled - */ -static void rxrpc_request_final_ACK(struct rxrpc_call *call) -{ - /* the call may be aborted before we have a chance to ACK it */ - write_lock(&call->state_lock); - - switch (call->state) { - case RXRPC_CALL_CLIENT_RECV_REPLY: - call->state = RXRPC_CALL_CLIENT_FINAL_ACK; - _debug("request final ACK"); - - /* get an extra ref on the call for the final-ACK generator to - * release */ - rxrpc_get_call(call); - set_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events); - if (try_to_del_timer_sync(&call->ack_timer) >= 0) - rxrpc_queue_call(call); - break; - - case RXRPC_CALL_SERVER_RECV_REQUEST: - call->state = RXRPC_CALL_SERVER_ACK_REQUEST; - default: - break; - } - - write_unlock(&call->state_lock); -} - -/* - * drop the bottom ACK off of the call ACK window and advance the window - */ -static void rxrpc_hard_ACK_data(struct rxrpc_call *call, - struct rxrpc_skb_priv *sp) -{ - int loop; - u32 seq; - - spin_lock_bh(&call->lock); - - _debug("hard ACK #%u", sp->hdr.seq); - - for (loop = 0; loop < RXRPC_ACKR_WINDOW_ASZ; loop++) { - call->ackr_window[loop] >>= 1; - call->ackr_window[loop] |= - call->ackr_window[loop + 1] << (BITS_PER_LONG - 1); - } - - seq = sp->hdr.seq; - ASSERTCMP(seq, ==, call->rx_data_eaten + 1); - call->rx_data_eaten = seq; - - if (call->ackr_win_top < UINT_MAX) - call->ackr_win_top++; - - ASSERTIFCMP(call->state <= RXRPC_CALL_COMPLETE, - call->rx_data_post, >=, call->rx_data_recv); - ASSERTIFCMP(call->state <= RXRPC_CALL_COMPLETE, - call->rx_data_recv, >=, call->rx_data_eaten); - - if (sp->hdr.flags & RXRPC_LAST_PACKET) { - rxrpc_request_final_ACK(call); - } else if (atomic_dec_and_test(&call->ackr_not_idle) && - test_and_clear_bit(RXRPC_CALL_TX_SOFT_ACK, &call->flags)) { - /* We previously soft-ACK'd some received packets that have now - * been consumed, so send a hard-ACK if no more packets are - * immediately forthcoming to allow the transmitter to free up - * its Tx bufferage. - */ - _debug("send Rx idle ACK"); - __rxrpc_propose_ACK(call, RXRPC_ACK_IDLE, sp->hdr.serial, - false); - } - - spin_unlock_bh(&call->lock); -} - -/* - * destroy a packet that has an RxRPC control buffer - * - advance the hard-ACK state of the parent call (done here in case something - * in the kernel bypasses recvmsg() and steals the packet directly off of the - * socket receive queue) - */ -void rxrpc_packet_destructor(struct sk_buff *skb) -{ - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - struct rxrpc_call *call = sp->call; - - _enter("%p{%p}", skb, call); - - if (call) { - /* send the final ACK on a client call */ - if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA) - rxrpc_hard_ACK_data(call, sp); - rxrpc_put_call(call); - sp->call = NULL; - } - - if (skb->sk) - sock_rfree(skb); - _leave(""); -} - -/** - * rxrpc_kernel_free_skb - Free an RxRPC socket buffer - * @skb: The socket buffer to be freed - * - * Let RxRPC free its own socket buffer, permitting it to maintain debug - * accounting. - */ -void rxrpc_kernel_free_skb(struct sk_buff *skb) -{ - rxrpc_free_skb(skb); -} -EXPORT_SYMBOL(rxrpc_kernel_free_skb); diff --git a/net/rxrpc/ar-transport.c b/net/rxrpc/ar-transport.c deleted file mode 100644 index a1b6518..0000000 --- a/net/rxrpc/ar-transport.c +++ /dev/null @@ -1,286 +0,0 @@ -/* RxRPC point-to-point transport session management - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * Time after last use at which transport record is cleaned up. - */ -unsigned int rxrpc_transport_expiry = 3600 * 24; - -static void rxrpc_transport_reaper(struct work_struct *work); - -static LIST_HEAD(rxrpc_transports); -static DEFINE_RWLOCK(rxrpc_transport_lock); -static DECLARE_DELAYED_WORK(rxrpc_transport_reap, rxrpc_transport_reaper); - -/* - * allocate a new transport session manager - */ -static struct rxrpc_transport *rxrpc_alloc_transport(struct rxrpc_local *local, - struct rxrpc_peer *peer, - gfp_t gfp) -{ - struct rxrpc_transport *trans; - - _enter(""); - - trans = kzalloc(sizeof(struct rxrpc_transport), gfp); - if (trans) { - trans->local = local; - trans->peer = peer; - INIT_LIST_HEAD(&trans->link); - trans->bundles = RB_ROOT; - trans->client_conns = RB_ROOT; - trans->server_conns = RB_ROOT; - skb_queue_head_init(&trans->error_queue); - spin_lock_init(&trans->client_lock); - rwlock_init(&trans->conn_lock); - atomic_set(&trans->usage, 1); - trans->conn_idcounter = peer->srx.srx_service << 16; - trans->debug_id = atomic_inc_return(&rxrpc_debug_id); - - if (peer->srx.transport.family == AF_INET) { - switch (peer->srx.transport_type) { - case SOCK_DGRAM: - INIT_WORK(&trans->error_handler, - rxrpc_UDP_error_handler); - break; - default: - BUG(); - break; - } - } else { - BUG(); - } - } - - _leave(" = %p", trans); - return trans; -} - -/* - * obtain a transport session for the nominated endpoints - */ -struct rxrpc_transport *rxrpc_get_transport(struct rxrpc_local *local, - struct rxrpc_peer *peer, - gfp_t gfp) -{ - struct rxrpc_transport *trans, *candidate; - const char *new = "old"; - int usage; - - _enter("{%pI4+%hu},{%pI4+%hu},", - &local->srx.transport.sin.sin_addr, - ntohs(local->srx.transport.sin.sin_port), - &peer->srx.transport.sin.sin_addr, - ntohs(peer->srx.transport.sin.sin_port)); - - /* search the transport list first */ - read_lock_bh(&rxrpc_transport_lock); - list_for_each_entry(trans, &rxrpc_transports, link) { - if (trans->local == local && trans->peer == peer) - goto found_extant_transport; - } - read_unlock_bh(&rxrpc_transport_lock); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_transport(local, peer, gfp); - if (!candidate) { - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } - - write_lock_bh(&rxrpc_transport_lock); - - list_for_each_entry(trans, &rxrpc_transports, link) { - if (trans->local == local && trans->peer == peer) - goto found_extant_second; - } - - /* we can now add the new candidate to the list */ - trans = candidate; - candidate = NULL; - usage = atomic_read(&trans->usage); - - rxrpc_get_local(trans->local); - atomic_inc(&trans->peer->usage); - list_add_tail(&trans->link, &rxrpc_transports); - write_unlock_bh(&rxrpc_transport_lock); - new = "new"; - -success: - _net("TRANSPORT %s %d local %d -> peer %d", - new, - trans->debug_id, - trans->local->debug_id, - trans->peer->debug_id); - - _leave(" = %p {u=%d}", trans, usage); - return trans; - - /* we found the transport in the list immediately */ -found_extant_transport: - usage = atomic_inc_return(&trans->usage); - read_unlock_bh(&rxrpc_transport_lock); - goto success; - - /* we found the transport on the second time through the list */ -found_extant_second: - usage = atomic_inc_return(&trans->usage); - write_unlock_bh(&rxrpc_transport_lock); - kfree(candidate); - goto success; -} - -/* - * find the transport connecting two endpoints - */ -struct rxrpc_transport *rxrpc_find_transport(struct rxrpc_local *local, - struct rxrpc_peer *peer) -{ - struct rxrpc_transport *trans; - - _enter("{%pI4+%hu},{%pI4+%hu},", - &local->srx.transport.sin.sin_addr, - ntohs(local->srx.transport.sin.sin_port), - &peer->srx.transport.sin.sin_addr, - ntohs(peer->srx.transport.sin.sin_port)); - - /* search the transport list */ - read_lock_bh(&rxrpc_transport_lock); - - list_for_each_entry(trans, &rxrpc_transports, link) { - if (trans->local == local && trans->peer == peer) - goto found_extant_transport; - } - - read_unlock_bh(&rxrpc_transport_lock); - _leave(" = NULL"); - return NULL; - -found_extant_transport: - atomic_inc(&trans->usage); - read_unlock_bh(&rxrpc_transport_lock); - _leave(" = %p", trans); - return trans; -} - -/* - * release a transport session - */ -void rxrpc_put_transport(struct rxrpc_transport *trans) -{ - _enter("%p{u=%d}", trans, atomic_read(&trans->usage)); - - ASSERTCMP(atomic_read(&trans->usage), >, 0); - - trans->put_time = ktime_get_seconds(); - if (unlikely(atomic_dec_and_test(&trans->usage))) { - _debug("zombie"); - /* let the reaper determine the timeout to avoid a race with - * overextending the timeout if the reaper is running at the - * same time */ - rxrpc_queue_delayed_work(&rxrpc_transport_reap, 0); - } - _leave(""); -} - -/* - * clean up a transport session - */ -static void rxrpc_cleanup_transport(struct rxrpc_transport *trans) -{ - _net("DESTROY TRANS %d", trans->debug_id); - - rxrpc_purge_queue(&trans->error_queue); - - rxrpc_put_local(trans->local); - rxrpc_put_peer(trans->peer); - kfree(trans); -} - -/* - * reap dead transports that have passed their expiry date - */ -static void rxrpc_transport_reaper(struct work_struct *work) -{ - struct rxrpc_transport *trans, *_p; - unsigned long now, earliest, reap_time; - - LIST_HEAD(graveyard); - - _enter(""); - - now = ktime_get_seconds(); - earliest = ULONG_MAX; - - /* extract all the transports that have been dead too long */ - write_lock_bh(&rxrpc_transport_lock); - list_for_each_entry_safe(trans, _p, &rxrpc_transports, link) { - _debug("reap TRANS %d { u=%d t=%ld }", - trans->debug_id, atomic_read(&trans->usage), - (long) now - (long) trans->put_time); - - if (likely(atomic_read(&trans->usage) > 0)) - continue; - - reap_time = trans->put_time + rxrpc_transport_expiry; - if (reap_time <= now) - list_move_tail(&trans->link, &graveyard); - else if (reap_time < earliest) - earliest = reap_time; - } - write_unlock_bh(&rxrpc_transport_lock); - - if (earliest != ULONG_MAX) { - _debug("reschedule reaper %ld", (long) earliest - now); - ASSERTCMP(earliest, >, now); - rxrpc_queue_delayed_work(&rxrpc_transport_reap, - (earliest - now) * HZ); - } - - /* then destroy all those pulled out */ - while (!list_empty(&graveyard)) { - trans = list_entry(graveyard.next, struct rxrpc_transport, - link); - list_del_init(&trans->link); - - ASSERTCMP(atomic_read(&trans->usage), ==, 0); - rxrpc_cleanup_transport(trans); - } - - _leave(""); -} - -/* - * preemptively destroy all the transport session records rather than waiting - * for them to time out - */ -void __exit rxrpc_destroy_all_transports(void) -{ - _enter(""); - - rxrpc_transport_expiry = 0; - cancel_delayed_work(&rxrpc_transport_reap); - rxrpc_queue_delayed_work(&rxrpc_transport_reap, 0); - - _leave(""); -} diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c new file mode 100644 index 0000000..eea5f4a --- /dev/null +++ b/net/rxrpc/call_accept.c @@ -0,0 +1,518 @@ +/* incoming call handling + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * generate a connection-level abort + */ +static int rxrpc_busy(struct rxrpc_local *local, struct sockaddr_rxrpc *srx, + struct rxrpc_wire_header *whdr) +{ + struct msghdr msg; + struct kvec iov[1]; + size_t len; + int ret; + + _enter("%d,,", local->debug_id); + + whdr->type = RXRPC_PACKET_TYPE_BUSY; + whdr->serial = htonl(1); + + msg.msg_name = &srx->transport.sin; + msg.msg_namelen = sizeof(srx->transport.sin); + msg.msg_control = NULL; + msg.msg_controllen = 0; + msg.msg_flags = 0; + + iov[0].iov_base = whdr; + iov[0].iov_len = sizeof(*whdr); + + len = iov[0].iov_len; + + _proto("Tx BUSY %%1"); + + ret = kernel_sendmsg(local->socket, &msg, iov, 1, len); + if (ret < 0) { + _leave(" = -EAGAIN [sendmsg failed: %d]", ret); + return -EAGAIN; + } + + _leave(" = 0"); + return 0; +} + +/* + * accept an incoming call that needs peer, transport and/or connection setting + * up + */ +static int rxrpc_accept_incoming_call(struct rxrpc_local *local, + struct rxrpc_sock *rx, + struct sk_buff *skb, + struct sockaddr_rxrpc *srx) +{ + struct rxrpc_connection *conn; + struct rxrpc_transport *trans; + struct rxrpc_skb_priv *sp, *nsp; + struct rxrpc_peer *peer; + struct rxrpc_call *call; + struct sk_buff *notification; + int ret; + + _enter(""); + + sp = rxrpc_skb(skb); + + /* get a notification message to send to the server app */ + notification = alloc_skb(0, GFP_NOFS); + if (!notification) { + _debug("no memory"); + ret = -ENOMEM; + goto error_nofree; + } + rxrpc_new_skb(notification); + notification->mark = RXRPC_SKB_MARK_NEW_CALL; + + peer = rxrpc_get_peer(srx, GFP_NOIO); + if (IS_ERR(peer)) { + _debug("no peer"); + ret = -EBUSY; + goto error; + } + + trans = rxrpc_get_transport(local, peer, GFP_NOIO); + rxrpc_put_peer(peer); + if (IS_ERR(trans)) { + _debug("no trans"); + ret = -EBUSY; + goto error; + } + + conn = rxrpc_incoming_connection(trans, &sp->hdr); + rxrpc_put_transport(trans); + if (IS_ERR(conn)) { + _debug("no conn"); + ret = PTR_ERR(conn); + goto error; + } + + call = rxrpc_incoming_call(rx, conn, &sp->hdr); + rxrpc_put_connection(conn); + if (IS_ERR(call)) { + _debug("no call"); + ret = PTR_ERR(call); + goto error; + } + + /* attach the call to the socket */ + read_lock_bh(&local->services_lock); + if (rx->sk.sk_state == RXRPC_CLOSE) + goto invalid_service; + + write_lock(&rx->call_lock); + if (!test_and_set_bit(RXRPC_CALL_INIT_ACCEPT, &call->flags)) { + rxrpc_get_call(call); + + spin_lock(&call->conn->state_lock); + if (sp->hdr.securityIndex > 0 && + call->conn->state == RXRPC_CONN_SERVER_UNSECURED) { + _debug("await conn sec"); + list_add_tail(&call->accept_link, &rx->secureq); + call->conn->state = RXRPC_CONN_SERVER_CHALLENGING; + atomic_inc(&call->conn->usage); + set_bit(RXRPC_CONN_CHALLENGE, &call->conn->events); + rxrpc_queue_conn(call->conn); + } else { + _debug("conn ready"); + call->state = RXRPC_CALL_SERVER_ACCEPTING; + list_add_tail(&call->accept_link, &rx->acceptq); + rxrpc_get_call(call); + nsp = rxrpc_skb(notification); + nsp->call = call; + + ASSERTCMP(atomic_read(&call->usage), >=, 3); + + _debug("notify"); + spin_lock(&call->lock); + ret = rxrpc_queue_rcv_skb(call, notification, true, + false); + spin_unlock(&call->lock); + notification = NULL; + BUG_ON(ret < 0); + } + spin_unlock(&call->conn->state_lock); + + _debug("queued"); + } + write_unlock(&rx->call_lock); + + _debug("process"); + rxrpc_fast_process_packet(call, skb); + + _debug("done"); + read_unlock_bh(&local->services_lock); + rxrpc_free_skb(notification); + rxrpc_put_call(call); + _leave(" = 0"); + return 0; + +invalid_service: + _debug("invalid"); + read_unlock_bh(&local->services_lock); + + read_lock_bh(&call->state_lock); + if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && + !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) { + rxrpc_get_call(call); + rxrpc_queue_call(call); + } + read_unlock_bh(&call->state_lock); + rxrpc_put_call(call); + ret = -ECONNREFUSED; +error: + rxrpc_free_skb(notification); +error_nofree: + _leave(" = %d", ret); + return ret; +} + +/* + * accept incoming calls that need peer, transport and/or connection setting up + * - the packets we get are all incoming client DATA packets that have seq == 1 + */ +void rxrpc_accept_incoming_calls(struct work_struct *work) +{ + struct rxrpc_local *local = + container_of(work, struct rxrpc_local, acceptor); + struct rxrpc_skb_priv *sp; + struct sockaddr_rxrpc srx; + struct rxrpc_sock *rx; + struct rxrpc_wire_header whdr; + struct sk_buff *skb; + int ret; + + _enter("%d", local->debug_id); + + read_lock_bh(&rxrpc_local_lock); + if (atomic_read(&local->usage) > 0) + rxrpc_get_local(local); + else + local = NULL; + read_unlock_bh(&rxrpc_local_lock); + if (!local) { + _leave(" [local dead]"); + return; + } + +process_next_packet: + skb = skb_dequeue(&local->accept_queue); + if (!skb) { + rxrpc_put_local(local); + _leave("\n"); + return; + } + + _net("incoming call skb %p", skb); + + sp = rxrpc_skb(skb); + + /* Set up a response packet header in case we need it */ + whdr.epoch = htonl(sp->hdr.epoch); + whdr.cid = htonl(sp->hdr.cid); + whdr.callNumber = htonl(sp->hdr.callNumber); + whdr.seq = htonl(sp->hdr.seq); + whdr.serial = 0; + whdr.flags = 0; + whdr.type = 0; + whdr.userStatus = 0; + whdr.securityIndex = sp->hdr.securityIndex; + whdr._rsvd = 0; + whdr.serviceId = htons(sp->hdr.serviceId); + + /* determine the remote address */ + memset(&srx, 0, sizeof(srx)); + srx.srx_family = AF_RXRPC; + srx.transport.family = local->srx.transport.family; + srx.transport_type = local->srx.transport_type; + switch (srx.transport.family) { + case AF_INET: + srx.transport_len = sizeof(struct sockaddr_in); + srx.transport.sin.sin_port = udp_hdr(skb)->source; + srx.transport.sin.sin_addr.s_addr = ip_hdr(skb)->saddr; + break; + default: + goto busy; + } + + /* get the socket providing the service */ + read_lock_bh(&local->services_lock); + list_for_each_entry(rx, &local->services, listen_link) { + if (rx->srx.srx_service == sp->hdr.serviceId && + rx->sk.sk_state != RXRPC_CLOSE) + goto found_service; + } + read_unlock_bh(&local->services_lock); + goto invalid_service; + +found_service: + _debug("found service %hd", rx->srx.srx_service); + if (sk_acceptq_is_full(&rx->sk)) + goto backlog_full; + sk_acceptq_added(&rx->sk); + sock_hold(&rx->sk); + read_unlock_bh(&local->services_lock); + + ret = rxrpc_accept_incoming_call(local, rx, skb, &srx); + if (ret < 0) + sk_acceptq_removed(&rx->sk); + sock_put(&rx->sk); + switch (ret) { + case -ECONNRESET: /* old calls are ignored */ + case -ECONNABORTED: /* aborted calls are reaborted or ignored */ + case 0: + goto process_next_packet; + case -ECONNREFUSED: + goto invalid_service; + case -EBUSY: + goto busy; + case -EKEYREJECTED: + goto security_mismatch; + default: + BUG(); + } + +backlog_full: + read_unlock_bh(&local->services_lock); +busy: + rxrpc_busy(local, &srx, &whdr); + rxrpc_free_skb(skb); + goto process_next_packet; + +invalid_service: + skb->priority = RX_INVALID_OPERATION; + rxrpc_reject_packet(local, skb); + goto process_next_packet; + + /* can't change connection security type mid-flow */ +security_mismatch: + skb->priority = RX_PROTOCOL_ERROR; + rxrpc_reject_packet(local, skb); + goto process_next_packet; +} + +/* + * handle acceptance of a call by userspace + * - assign the user call ID to the call at the front of the queue + */ +struct rxrpc_call *rxrpc_accept_call(struct rxrpc_sock *rx, + unsigned long user_call_ID) +{ + struct rxrpc_call *call; + struct rb_node *parent, **pp; + int ret; + + _enter(",%lx", user_call_ID); + + ASSERT(!irqs_disabled()); + + write_lock(&rx->call_lock); + + ret = -ENODATA; + if (list_empty(&rx->acceptq)) + goto out; + + /* check the user ID isn't already in use */ + ret = -EBADSLT; + pp = &rx->calls.rb_node; + parent = NULL; + while (*pp) { + parent = *pp; + call = rb_entry(parent, struct rxrpc_call, sock_node); + + if (user_call_ID < call->user_call_ID) + pp = &(*pp)->rb_left; + else if (user_call_ID > call->user_call_ID) + pp = &(*pp)->rb_right; + else + goto out; + } + + /* dequeue the first call and check it's still valid */ + call = list_entry(rx->acceptq.next, struct rxrpc_call, accept_link); + list_del_init(&call->accept_link); + sk_acceptq_removed(&rx->sk); + + write_lock_bh(&call->state_lock); + switch (call->state) { + case RXRPC_CALL_SERVER_ACCEPTING: + call->state = RXRPC_CALL_SERVER_RECV_REQUEST; + break; + case RXRPC_CALL_REMOTELY_ABORTED: + case RXRPC_CALL_LOCALLY_ABORTED: + ret = -ECONNABORTED; + goto out_release; + case RXRPC_CALL_NETWORK_ERROR: + ret = call->conn->error; + goto out_release; + case RXRPC_CALL_DEAD: + ret = -ETIME; + goto out_discard; + default: + BUG(); + } + + /* formalise the acceptance */ + call->user_call_ID = user_call_ID; + rb_link_node(&call->sock_node, parent, pp); + rb_insert_color(&call->sock_node, &rx->calls); + if (test_and_set_bit(RXRPC_CALL_HAS_USERID, &call->flags)) + BUG(); + if (test_and_set_bit(RXRPC_CALL_EV_ACCEPTED, &call->events)) + BUG(); + rxrpc_queue_call(call); + + rxrpc_get_call(call); + write_unlock_bh(&call->state_lock); + write_unlock(&rx->call_lock); + _leave(" = %p{%d}", call, call->debug_id); + return call; + + /* if the call is already dying or dead, then we leave the socket's ref + * on it to be released by rxrpc_dead_call_expired() as induced by + * rxrpc_release_call() */ +out_release: + _debug("release %p", call); + if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && + !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) + rxrpc_queue_call(call); +out_discard: + write_unlock_bh(&call->state_lock); + _debug("discard %p", call); +out: + write_unlock(&rx->call_lock); + _leave(" = %d", ret); + return ERR_PTR(ret); +} + +/* + * Handle rejection of a call by userspace + * - reject the call at the front of the queue + */ +int rxrpc_reject_call(struct rxrpc_sock *rx) +{ + struct rxrpc_call *call; + int ret; + + _enter(""); + + ASSERT(!irqs_disabled()); + + write_lock(&rx->call_lock); + + ret = -ENODATA; + if (list_empty(&rx->acceptq)) + goto out; + + /* dequeue the first call and check it's still valid */ + call = list_entry(rx->acceptq.next, struct rxrpc_call, accept_link); + list_del_init(&call->accept_link); + sk_acceptq_removed(&rx->sk); + + write_lock_bh(&call->state_lock); + switch (call->state) { + case RXRPC_CALL_SERVER_ACCEPTING: + call->state = RXRPC_CALL_SERVER_BUSY; + if (test_and_set_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events)) + rxrpc_queue_call(call); + ret = 0; + goto out_release; + case RXRPC_CALL_REMOTELY_ABORTED: + case RXRPC_CALL_LOCALLY_ABORTED: + ret = -ECONNABORTED; + goto out_release; + case RXRPC_CALL_NETWORK_ERROR: + ret = call->conn->error; + goto out_release; + case RXRPC_CALL_DEAD: + ret = -ETIME; + goto out_discard; + default: + BUG(); + } + + /* if the call is already dying or dead, then we leave the socket's ref + * on it to be released by rxrpc_dead_call_expired() as induced by + * rxrpc_release_call() */ +out_release: + _debug("release %p", call); + if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && + !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) + rxrpc_queue_call(call); +out_discard: + write_unlock_bh(&call->state_lock); + _debug("discard %p", call); +out: + write_unlock(&rx->call_lock); + _leave(" = %d", ret); + return ret; +} + +/** + * rxrpc_kernel_accept_call - Allow a kernel service to accept an incoming call + * @sock: The socket on which the impending call is waiting + * @user_call_ID: The tag to attach to the call + * + * Allow a kernel service to accept an incoming call, assuming the incoming + * call is still valid. + */ +struct rxrpc_call *rxrpc_kernel_accept_call(struct socket *sock, + unsigned long user_call_ID) +{ + struct rxrpc_call *call; + + _enter(",%lx", user_call_ID); + call = rxrpc_accept_call(rxrpc_sk(sock->sk), user_call_ID); + _leave(" = %p", call); + return call; +} +EXPORT_SYMBOL(rxrpc_kernel_accept_call); + +/** + * rxrpc_kernel_reject_call - Allow a kernel service to reject an incoming call + * @sock: The socket on which the impending call is waiting + * + * Allow a kernel service to reject an incoming call with a BUSY message, + * assuming the incoming call is still valid. + */ +int rxrpc_kernel_reject_call(struct socket *sock) +{ + int ret; + + _enter(""); + ret = rxrpc_reject_call(rxrpc_sk(sock->sk)); + _leave(" = %d", ret); + return ret; +} +EXPORT_SYMBOL(rxrpc_kernel_reject_call); diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c new file mode 100644 index 0000000..1838178 --- /dev/null +++ b/net/rxrpc/call_event.c @@ -0,0 +1,1288 @@ +/* Management of Tx window, Tx resend, ACKs and out-of-sequence reception + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * propose an ACK be sent + */ +void __rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason, + u32 serial, bool immediate) +{ + unsigned long expiry; + s8 prior = rxrpc_ack_priority[ack_reason]; + + ASSERTCMP(prior, >, 0); + + _enter("{%d},%s,%%%x,%u", + call->debug_id, rxrpc_acks(ack_reason), serial, immediate); + + if (prior < rxrpc_ack_priority[call->ackr_reason]) { + if (immediate) + goto cancel_timer; + return; + } + + /* update DELAY, IDLE, REQUESTED and PING_RESPONSE ACK serial + * numbers */ + if (prior == rxrpc_ack_priority[call->ackr_reason]) { + if (prior <= 4) + call->ackr_serial = serial; + if (immediate) + goto cancel_timer; + return; + } + + call->ackr_reason = ack_reason; + call->ackr_serial = serial; + + switch (ack_reason) { + case RXRPC_ACK_DELAY: + _debug("run delay timer"); + expiry = rxrpc_soft_ack_delay; + goto run_timer; + + case RXRPC_ACK_IDLE: + if (!immediate) { + _debug("run defer timer"); + expiry = rxrpc_idle_ack_delay; + goto run_timer; + } + goto cancel_timer; + + case RXRPC_ACK_REQUESTED: + expiry = rxrpc_requested_ack_delay; + if (!expiry) + goto cancel_timer; + if (!immediate || serial == 1) { + _debug("run defer timer"); + goto run_timer; + } + + default: + _debug("immediate ACK"); + goto cancel_timer; + } + +run_timer: + expiry += jiffies; + if (!timer_pending(&call->ack_timer) || + time_after(call->ack_timer.expires, expiry)) + mod_timer(&call->ack_timer, expiry); + return; + +cancel_timer: + _debug("cancel timer %%%u", serial); + try_to_del_timer_sync(&call->ack_timer); + read_lock_bh(&call->state_lock); + if (call->state <= RXRPC_CALL_COMPLETE && + !test_and_set_bit(RXRPC_CALL_EV_ACK, &call->events)) + rxrpc_queue_call(call); + read_unlock_bh(&call->state_lock); +} + +/* + * propose an ACK be sent, locking the call structure + */ +void rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason, + u32 serial, bool immediate) +{ + s8 prior = rxrpc_ack_priority[ack_reason]; + + if (prior > rxrpc_ack_priority[call->ackr_reason]) { + spin_lock_bh(&call->lock); + __rxrpc_propose_ACK(call, ack_reason, serial, immediate); + spin_unlock_bh(&call->lock); + } +} + +/* + * set the resend timer + */ +static void rxrpc_set_resend(struct rxrpc_call *call, u8 resend, + unsigned long resend_at) +{ + read_lock_bh(&call->state_lock); + if (call->state >= RXRPC_CALL_COMPLETE) + resend = 0; + + if (resend & 1) { + _debug("SET RESEND"); + set_bit(RXRPC_CALL_EV_RESEND, &call->events); + } + + if (resend & 2) { + _debug("MODIFY RESEND TIMER"); + set_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); + mod_timer(&call->resend_timer, resend_at); + } else { + _debug("KILL RESEND TIMER"); + del_timer_sync(&call->resend_timer); + clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events); + clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); + } + read_unlock_bh(&call->state_lock); +} + +/* + * resend packets + */ +static void rxrpc_resend(struct rxrpc_call *call) +{ + struct rxrpc_wire_header *whdr; + struct rxrpc_skb_priv *sp; + struct sk_buff *txb; + unsigned long *p_txb, resend_at; + bool stop; + int loop; + u8 resend; + + _enter("{%d,%d,%d,%d},", + call->acks_hard, call->acks_unacked, + atomic_read(&call->sequence), + CIRC_CNT(call->acks_head, call->acks_tail, call->acks_winsz)); + + stop = false; + resend = 0; + resend_at = 0; + + for (loop = call->acks_tail; + loop != call->acks_head || stop; + loop = (loop + 1) & (call->acks_winsz - 1) + ) { + p_txb = call->acks_window + loop; + smp_read_barrier_depends(); + if (*p_txb & 1) + continue; + + txb = (struct sk_buff *) *p_txb; + sp = rxrpc_skb(txb); + + if (sp->need_resend) { + sp->need_resend = false; + + /* each Tx packet has a new serial number */ + sp->hdr.serial = atomic_inc_return(&call->conn->serial); + + whdr = (struct rxrpc_wire_header *)txb->head; + whdr->serial = htonl(sp->hdr.serial); + + _proto("Tx DATA %%%u { #%d }", + sp->hdr.serial, sp->hdr.seq); + if (rxrpc_send_packet(call->conn->trans, txb) < 0) { + stop = true; + sp->resend_at = jiffies + 3; + } else { + sp->resend_at = + jiffies + rxrpc_resend_timeout; + } + } + + if (time_after_eq(jiffies + 1, sp->resend_at)) { + sp->need_resend = true; + resend |= 1; + } else if (resend & 2) { + if (time_before(sp->resend_at, resend_at)) + resend_at = sp->resend_at; + } else { + resend_at = sp->resend_at; + resend |= 2; + } + } + + rxrpc_set_resend(call, resend, resend_at); + _leave(""); +} + +/* + * handle resend timer expiry + */ +static void rxrpc_resend_timer(struct rxrpc_call *call) +{ + struct rxrpc_skb_priv *sp; + struct sk_buff *txb; + unsigned long *p_txb, resend_at; + int loop; + u8 resend; + + _enter("%d,%d,%d", + call->acks_tail, call->acks_unacked, call->acks_head); + + if (call->state >= RXRPC_CALL_COMPLETE) + return; + + resend = 0; + resend_at = 0; + + for (loop = call->acks_unacked; + loop != call->acks_head; + loop = (loop + 1) & (call->acks_winsz - 1) + ) { + p_txb = call->acks_window + loop; + smp_read_barrier_depends(); + txb = (struct sk_buff *) (*p_txb & ~1); + sp = rxrpc_skb(txb); + + ASSERT(!(*p_txb & 1)); + + if (sp->need_resend) { + ; + } else if (time_after_eq(jiffies + 1, sp->resend_at)) { + sp->need_resend = true; + resend |= 1; + } else if (resend & 2) { + if (time_before(sp->resend_at, resend_at)) + resend_at = sp->resend_at; + } else { + resend_at = sp->resend_at; + resend |= 2; + } + } + + rxrpc_set_resend(call, resend, resend_at); + _leave(""); +} + +/* + * process soft ACKs of our transmitted packets + * - these indicate packets the peer has or has not received, but hasn't yet + * given to the consumer, and so can still be discarded and re-requested + */ +static int rxrpc_process_soft_ACKs(struct rxrpc_call *call, + struct rxrpc_ackpacket *ack, + struct sk_buff *skb) +{ + struct rxrpc_skb_priv *sp; + struct sk_buff *txb; + unsigned long *p_txb, resend_at; + int loop; + u8 sacks[RXRPC_MAXACKS], resend; + + _enter("{%d,%d},{%d},", + call->acks_hard, + CIRC_CNT(call->acks_head, call->acks_tail, call->acks_winsz), + ack->nAcks); + + if (skb_copy_bits(skb, 0, sacks, ack->nAcks) < 0) + goto protocol_error; + + resend = 0; + resend_at = 0; + for (loop = 0; loop < ack->nAcks; loop++) { + p_txb = call->acks_window; + p_txb += (call->acks_tail + loop) & (call->acks_winsz - 1); + smp_read_barrier_depends(); + txb = (struct sk_buff *) (*p_txb & ~1); + sp = rxrpc_skb(txb); + + switch (sacks[loop]) { + case RXRPC_ACK_TYPE_ACK: + sp->need_resend = false; + *p_txb |= 1; + break; + case RXRPC_ACK_TYPE_NACK: + sp->need_resend = true; + *p_txb &= ~1; + resend = 1; + break; + default: + _debug("Unsupported ACK type %d", sacks[loop]); + goto protocol_error; + } + } + + smp_mb(); + call->acks_unacked = (call->acks_tail + loop) & (call->acks_winsz - 1); + + /* anything not explicitly ACK'd is implicitly NACK'd, but may just not + * have been received or processed yet by the far end */ + for (loop = call->acks_unacked; + loop != call->acks_head; + loop = (loop + 1) & (call->acks_winsz - 1) + ) { + p_txb = call->acks_window + loop; + smp_read_barrier_depends(); + txb = (struct sk_buff *) (*p_txb & ~1); + sp = rxrpc_skb(txb); + + if (*p_txb & 1) { + /* packet must have been discarded */ + sp->need_resend = true; + *p_txb &= ~1; + resend |= 1; + } else if (sp->need_resend) { + ; + } else if (time_after_eq(jiffies + 1, sp->resend_at)) { + sp->need_resend = true; + resend |= 1; + } else if (resend & 2) { + if (time_before(sp->resend_at, resend_at)) + resend_at = sp->resend_at; + } else { + resend_at = sp->resend_at; + resend |= 2; + } + } + + rxrpc_set_resend(call, resend, resend_at); + _leave(" = 0"); + return 0; + +protocol_error: + _leave(" = -EPROTO"); + return -EPROTO; +} + +/* + * discard hard-ACK'd packets from the Tx window + */ +static void rxrpc_rotate_tx_window(struct rxrpc_call *call, u32 hard) +{ + unsigned long _skb; + int tail = call->acks_tail, old_tail; + int win = CIRC_CNT(call->acks_head, tail, call->acks_winsz); + + _enter("{%u,%u},%u", call->acks_hard, win, hard); + + ASSERTCMP(hard - call->acks_hard, <=, win); + + while (call->acks_hard < hard) { + smp_read_barrier_depends(); + _skb = call->acks_window[tail] & ~1; + rxrpc_free_skb((struct sk_buff *) _skb); + old_tail = tail; + tail = (tail + 1) & (call->acks_winsz - 1); + call->acks_tail = tail; + if (call->acks_unacked == old_tail) + call->acks_unacked = tail; + call->acks_hard++; + } + + wake_up(&call->tx_waitq); +} + +/* + * clear the Tx window in the event of a failure + */ +static void rxrpc_clear_tx_window(struct rxrpc_call *call) +{ + rxrpc_rotate_tx_window(call, atomic_read(&call->sequence)); +} + +/* + * drain the out of sequence received packet queue into the packet Rx queue + */ +static int rxrpc_drain_rx_oos_queue(struct rxrpc_call *call) +{ + struct rxrpc_skb_priv *sp; + struct sk_buff *skb; + bool terminal; + int ret; + + _enter("{%d,%d}", call->rx_data_post, call->rx_first_oos); + + spin_lock_bh(&call->lock); + + ret = -ECONNRESET; + if (test_bit(RXRPC_CALL_RELEASED, &call->flags)) + goto socket_unavailable; + + skb = skb_dequeue(&call->rx_oos_queue); + if (skb) { + sp = rxrpc_skb(skb); + + _debug("drain OOS packet %d [%d]", + sp->hdr.seq, call->rx_first_oos); + + if (sp->hdr.seq != call->rx_first_oos) { + skb_queue_head(&call->rx_oos_queue, skb); + call->rx_first_oos = rxrpc_skb(skb)->hdr.seq; + _debug("requeue %p {%u}", skb, call->rx_first_oos); + } else { + skb->mark = RXRPC_SKB_MARK_DATA; + terminal = ((sp->hdr.flags & RXRPC_LAST_PACKET) && + !(sp->hdr.flags & RXRPC_CLIENT_INITIATED)); + ret = rxrpc_queue_rcv_skb(call, skb, true, terminal); + BUG_ON(ret < 0); + _debug("drain #%u", call->rx_data_post); + call->rx_data_post++; + + /* find out what the next packet is */ + skb = skb_peek(&call->rx_oos_queue); + if (skb) + call->rx_first_oos = rxrpc_skb(skb)->hdr.seq; + else + call->rx_first_oos = 0; + _debug("peek %p {%u}", skb, call->rx_first_oos); + } + } + + ret = 0; +socket_unavailable: + spin_unlock_bh(&call->lock); + _leave(" = %d", ret); + return ret; +} + +/* + * insert an out of sequence packet into the buffer + */ +static void rxrpc_insert_oos_packet(struct rxrpc_call *call, + struct sk_buff *skb) +{ + struct rxrpc_skb_priv *sp, *psp; + struct sk_buff *p; + u32 seq; + + sp = rxrpc_skb(skb); + seq = sp->hdr.seq; + _enter(",,{%u}", seq); + + skb->destructor = rxrpc_packet_destructor; + ASSERTCMP(sp->call, ==, NULL); + sp->call = call; + rxrpc_get_call(call); + + /* insert into the buffer in sequence order */ + spin_lock_bh(&call->lock); + + skb_queue_walk(&call->rx_oos_queue, p) { + psp = rxrpc_skb(p); + if (psp->hdr.seq > seq) { + _debug("insert oos #%u before #%u", seq, psp->hdr.seq); + skb_insert(p, skb, &call->rx_oos_queue); + goto inserted; + } + } + + _debug("append oos #%u", seq); + skb_queue_tail(&call->rx_oos_queue, skb); +inserted: + + /* we might now have a new front to the queue */ + if (call->rx_first_oos == 0 || seq < call->rx_first_oos) + call->rx_first_oos = seq; + + read_lock(&call->state_lock); + if (call->state < RXRPC_CALL_COMPLETE && + call->rx_data_post == call->rx_first_oos) { + _debug("drain rx oos now"); + set_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events); + } + read_unlock(&call->state_lock); + + spin_unlock_bh(&call->lock); + _leave(" [stored #%u]", call->rx_first_oos); +} + +/* + * clear the Tx window on final ACK reception + */ +static void rxrpc_zap_tx_window(struct rxrpc_call *call) +{ + struct rxrpc_skb_priv *sp; + struct sk_buff *skb; + unsigned long _skb, *acks_window; + u8 winsz = call->acks_winsz; + int tail; + + acks_window = call->acks_window; + call->acks_window = NULL; + + while (CIRC_CNT(call->acks_head, call->acks_tail, winsz) > 0) { + tail = call->acks_tail; + smp_read_barrier_depends(); + _skb = acks_window[tail] & ~1; + smp_mb(); + call->acks_tail = (call->acks_tail + 1) & (winsz - 1); + + skb = (struct sk_buff *) _skb; + sp = rxrpc_skb(skb); + _debug("+++ clear Tx %u", sp->hdr.seq); + rxrpc_free_skb(skb); + } + + kfree(acks_window); +} + +/* + * process the extra information that may be appended to an ACK packet + */ +static void rxrpc_extract_ackinfo(struct rxrpc_call *call, struct sk_buff *skb, + unsigned int latest, int nAcks) +{ + struct rxrpc_ackinfo ackinfo; + struct rxrpc_peer *peer; + unsigned int mtu; + + if (skb_copy_bits(skb, nAcks + 3, &ackinfo, sizeof(ackinfo)) < 0) { + _leave(" [no ackinfo]"); + return; + } + + _proto("Rx ACK %%%u Info { rx=%u max=%u rwin=%u jm=%u }", + latest, + ntohl(ackinfo.rxMTU), ntohl(ackinfo.maxMTU), + ntohl(ackinfo.rwind), ntohl(ackinfo.jumbo_max)); + + mtu = min(ntohl(ackinfo.rxMTU), ntohl(ackinfo.maxMTU)); + + peer = call->conn->trans->peer; + if (mtu < peer->maxdata) { + spin_lock_bh(&peer->lock); + peer->maxdata = mtu; + peer->mtu = mtu + peer->hdrsize; + spin_unlock_bh(&peer->lock); + _net("Net MTU %u (maxdata %u)", peer->mtu, peer->maxdata); + } +} + +/* + * process packets in the reception queue + */ +static int rxrpc_process_rx_queue(struct rxrpc_call *call, + u32 *_abort_code) +{ + struct rxrpc_ackpacket ack; + struct rxrpc_skb_priv *sp; + struct sk_buff *skb; + bool post_ACK; + int latest; + u32 hard, tx; + + _enter(""); + +process_further: + skb = skb_dequeue(&call->rx_queue); + if (!skb) + return -EAGAIN; + + _net("deferred skb %p", skb); + + sp = rxrpc_skb(skb); + + _debug("process %s [st %d]", rxrpc_pkts[sp->hdr.type], call->state); + + post_ACK = false; + + switch (sp->hdr.type) { + /* data packets that wind up here have been received out of + * order, need security processing or are jumbo packets */ + case RXRPC_PACKET_TYPE_DATA: + _proto("OOSQ DATA %%%u { #%u }", sp->hdr.serial, sp->hdr.seq); + + /* secured packets must be verified and possibly decrypted */ + if (call->conn->security->verify_packet(call, skb, + _abort_code) < 0) + goto protocol_error; + + rxrpc_insert_oos_packet(call, skb); + goto process_further; + + /* partial ACK to process */ + case RXRPC_PACKET_TYPE_ACK: + if (skb_copy_bits(skb, 0, &ack, sizeof(ack)) < 0) { + _debug("extraction failure"); + goto protocol_error; + } + if (!skb_pull(skb, sizeof(ack))) + BUG(); + + latest = sp->hdr.serial; + hard = ntohl(ack.firstPacket); + tx = atomic_read(&call->sequence); + + _proto("Rx ACK %%%u { m=%hu f=#%u p=#%u s=%%%u r=%s n=%u }", + latest, + ntohs(ack.maxSkew), + hard, + ntohl(ack.previousPacket), + ntohl(ack.serial), + rxrpc_acks(ack.reason), + ack.nAcks); + + rxrpc_extract_ackinfo(call, skb, latest, ack.nAcks); + + if (ack.reason == RXRPC_ACK_PING) { + _proto("Rx ACK %%%u PING Request", latest); + rxrpc_propose_ACK(call, RXRPC_ACK_PING_RESPONSE, + sp->hdr.serial, true); + } + + /* discard any out-of-order or duplicate ACKs */ + if (latest - call->acks_latest <= 0) { + _debug("discard ACK %d <= %d", + latest, call->acks_latest); + goto discard; + } + call->acks_latest = latest; + + if (call->state != RXRPC_CALL_CLIENT_SEND_REQUEST && + call->state != RXRPC_CALL_CLIENT_AWAIT_REPLY && + call->state != RXRPC_CALL_SERVER_SEND_REPLY && + call->state != RXRPC_CALL_SERVER_AWAIT_ACK) + goto discard; + + _debug("Tx=%d H=%u S=%d", tx, call->acks_hard, call->state); + + if (hard > 0) { + if (hard - 1 > tx) { + _debug("hard-ACK'd packet %d not transmitted" + " (%d top)", + hard - 1, tx); + goto protocol_error; + } + + if ((call->state == RXRPC_CALL_CLIENT_AWAIT_REPLY || + call->state == RXRPC_CALL_SERVER_AWAIT_ACK) && + hard > tx) { + call->acks_hard = tx; + goto all_acked; + } + + smp_rmb(); + rxrpc_rotate_tx_window(call, hard - 1); + } + + if (ack.nAcks > 0) { + if (hard - 1 + ack.nAcks > tx) { + _debug("soft-ACK'd packet %d+%d not" + " transmitted (%d top)", + hard - 1, ack.nAcks, tx); + goto protocol_error; + } + + if (rxrpc_process_soft_ACKs(call, &ack, skb) < 0) + goto protocol_error; + } + goto discard; + + /* complete ACK to process */ + case RXRPC_PACKET_TYPE_ACKALL: + goto all_acked; + + /* abort and busy are handled elsewhere */ + case RXRPC_PACKET_TYPE_BUSY: + case RXRPC_PACKET_TYPE_ABORT: + BUG(); + + /* connection level events - also handled elsewhere */ + case RXRPC_PACKET_TYPE_CHALLENGE: + case RXRPC_PACKET_TYPE_RESPONSE: + case RXRPC_PACKET_TYPE_DEBUG: + BUG(); + } + + /* if we've had a hard ACK that covers all the packets we've sent, then + * that ends that phase of the operation */ +all_acked: + write_lock_bh(&call->state_lock); + _debug("ack all %d", call->state); + + switch (call->state) { + case RXRPC_CALL_CLIENT_AWAIT_REPLY: + call->state = RXRPC_CALL_CLIENT_RECV_REPLY; + break; + case RXRPC_CALL_SERVER_AWAIT_ACK: + _debug("srv complete"); + call->state = RXRPC_CALL_COMPLETE; + post_ACK = true; + break; + case RXRPC_CALL_CLIENT_SEND_REQUEST: + case RXRPC_CALL_SERVER_RECV_REQUEST: + goto protocol_error_unlock; /* can't occur yet */ + default: + write_unlock_bh(&call->state_lock); + goto discard; /* assume packet left over from earlier phase */ + } + + write_unlock_bh(&call->state_lock); + + /* if all the packets we sent are hard-ACK'd, then we can discard + * whatever we've got left */ + _debug("clear Tx %d", + CIRC_CNT(call->acks_head, call->acks_tail, call->acks_winsz)); + + del_timer_sync(&call->resend_timer); + clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); + clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events); + + if (call->acks_window) + rxrpc_zap_tx_window(call); + + if (post_ACK) { + /* post the final ACK message for userspace to pick up */ + _debug("post ACK"); + skb->mark = RXRPC_SKB_MARK_FINAL_ACK; + sp->call = call; + rxrpc_get_call(call); + spin_lock_bh(&call->lock); + if (rxrpc_queue_rcv_skb(call, skb, true, true) < 0) + BUG(); + spin_unlock_bh(&call->lock); + goto process_further; + } + +discard: + rxrpc_free_skb(skb); + goto process_further; + +protocol_error_unlock: + write_unlock_bh(&call->state_lock); +protocol_error: + rxrpc_free_skb(skb); + _leave(" = -EPROTO"); + return -EPROTO; +} + +/* + * post a message to the socket Rx queue for recvmsg() to pick up + */ +static int rxrpc_post_message(struct rxrpc_call *call, u32 mark, u32 error, + bool fatal) +{ + struct rxrpc_skb_priv *sp; + struct sk_buff *skb; + int ret; + + _enter("{%d,%lx},%u,%u,%d", + call->debug_id, call->flags, mark, error, fatal); + + /* remove timers and things for fatal messages */ + if (fatal) { + del_timer_sync(&call->resend_timer); + del_timer_sync(&call->ack_timer); + clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); + } + + if (mark != RXRPC_SKB_MARK_NEW_CALL && + !test_bit(RXRPC_CALL_HAS_USERID, &call->flags)) { + _leave("[no userid]"); + return 0; + } + + if (!test_bit(RXRPC_CALL_TERMINAL_MSG, &call->flags)) { + skb = alloc_skb(0, GFP_NOFS); + if (!skb) + return -ENOMEM; + + rxrpc_new_skb(skb); + + skb->mark = mark; + + sp = rxrpc_skb(skb); + memset(sp, 0, sizeof(*sp)); + sp->error = error; + sp->call = call; + rxrpc_get_call(call); + + spin_lock_bh(&call->lock); + ret = rxrpc_queue_rcv_skb(call, skb, true, fatal); + spin_unlock_bh(&call->lock); + BUG_ON(ret < 0); + } + + return 0; +} + +/* + * handle background processing of incoming call packets and ACK / abort + * generation + */ +void rxrpc_process_call(struct work_struct *work) +{ + struct rxrpc_call *call = + container_of(work, struct rxrpc_call, processor); + struct rxrpc_wire_header whdr; + struct rxrpc_ackpacket ack; + struct rxrpc_ackinfo ackinfo; + struct msghdr msg; + struct kvec iov[5]; + enum rxrpc_call_event genbit; + unsigned long bits; + __be32 data, pad; + size_t len; + int loop, nbit, ioc, ret, mtu; + u32 serial, abort_code = RX_PROTOCOL_ERROR; + u8 *acks = NULL; + + //printk("\n--------------------\n"); + _enter("{%d,%s,%lx} [%lu]", + call->debug_id, rxrpc_call_states[call->state], call->events, + (jiffies - call->creation_jif) / (HZ / 10)); + + if (test_and_set_bit(RXRPC_CALL_PROC_BUSY, &call->flags)) { + _debug("XXXXXXXXXXXXX RUNNING ON MULTIPLE CPUS XXXXXXXXXXXXX"); + return; + } + + /* there's a good chance we're going to have to send a message, so set + * one up in advance */ + msg.msg_name = &call->conn->trans->peer->srx.transport; + msg.msg_namelen = call->conn->trans->peer->srx.transport_len; + msg.msg_control = NULL; + msg.msg_controllen = 0; + msg.msg_flags = 0; + + whdr.epoch = htonl(call->conn->epoch); + whdr.cid = htonl(call->cid); + whdr.callNumber = htonl(call->call_id); + whdr.seq = 0; + whdr.type = RXRPC_PACKET_TYPE_ACK; + whdr.flags = call->conn->out_clientflag; + whdr.userStatus = 0; + whdr.securityIndex = call->conn->security_ix; + whdr._rsvd = 0; + whdr.serviceId = htons(call->service_id); + + memset(iov, 0, sizeof(iov)); + iov[0].iov_base = &whdr; + iov[0].iov_len = sizeof(whdr); + + /* deal with events of a final nature */ + if (test_bit(RXRPC_CALL_EV_RELEASE, &call->events)) { + rxrpc_release_call(call); + clear_bit(RXRPC_CALL_EV_RELEASE, &call->events); + } + + if (test_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events)) { + int error; + + clear_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events); + clear_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events); + clear_bit(RXRPC_CALL_EV_ABORT, &call->events); + + error = call->conn->trans->peer->net_error; + _debug("post net error %d", error); + + if (rxrpc_post_message(call, RXRPC_SKB_MARK_NET_ERROR, + error, true) < 0) + goto no_mem; + clear_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events); + goto kill_ACKs; + } + + if (test_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events)) { + ASSERTCMP(call->state, >, RXRPC_CALL_COMPLETE); + + clear_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events); + clear_bit(RXRPC_CALL_EV_ABORT, &call->events); + + _debug("post conn abort"); + + if (rxrpc_post_message(call, RXRPC_SKB_MARK_LOCAL_ERROR, + call->conn->error, true) < 0) + goto no_mem; + clear_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events); + goto kill_ACKs; + } + + if (test_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events)) { + whdr.type = RXRPC_PACKET_TYPE_BUSY; + genbit = RXRPC_CALL_EV_REJECT_BUSY; + goto send_message; + } + + if (test_bit(RXRPC_CALL_EV_ABORT, &call->events)) { + ASSERTCMP(call->state, >, RXRPC_CALL_COMPLETE); + + if (rxrpc_post_message(call, RXRPC_SKB_MARK_LOCAL_ERROR, + ECONNABORTED, true) < 0) + goto no_mem; + whdr.type = RXRPC_PACKET_TYPE_ABORT; + data = htonl(call->local_abort); + iov[1].iov_base = &data; + iov[1].iov_len = sizeof(data); + genbit = RXRPC_CALL_EV_ABORT; + goto send_message; + } + + if (test_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events)) { + genbit = RXRPC_CALL_EV_ACK_FINAL; + + ack.bufferSpace = htons(8); + ack.maxSkew = 0; + ack.serial = 0; + ack.reason = RXRPC_ACK_IDLE; + ack.nAcks = 0; + call->ackr_reason = 0; + + spin_lock_bh(&call->lock); + ack.serial = htonl(call->ackr_serial); + ack.previousPacket = htonl(call->ackr_prev_seq); + ack.firstPacket = htonl(call->rx_data_eaten + 1); + spin_unlock_bh(&call->lock); + + pad = 0; + + iov[1].iov_base = &ack; + iov[1].iov_len = sizeof(ack); + iov[2].iov_base = &pad; + iov[2].iov_len = 3; + iov[3].iov_base = &ackinfo; + iov[3].iov_len = sizeof(ackinfo); + goto send_ACK; + } + + if (call->events & ((1 << RXRPC_CALL_EV_RCVD_BUSY) | + (1 << RXRPC_CALL_EV_RCVD_ABORT)) + ) { + u32 mark; + + if (test_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events)) + mark = RXRPC_SKB_MARK_REMOTE_ABORT; + else + mark = RXRPC_SKB_MARK_BUSY; + + _debug("post abort/busy"); + rxrpc_clear_tx_window(call); + if (rxrpc_post_message(call, mark, ECONNABORTED, true) < 0) + goto no_mem; + + clear_bit(RXRPC_CALL_EV_RCVD_BUSY, &call->events); + clear_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events); + goto kill_ACKs; + } + + if (test_and_clear_bit(RXRPC_CALL_EV_RCVD_ACKALL, &call->events)) { + _debug("do implicit ackall"); + rxrpc_clear_tx_window(call); + } + + if (test_bit(RXRPC_CALL_EV_LIFE_TIMER, &call->events)) { + write_lock_bh(&call->state_lock); + if (call->state <= RXRPC_CALL_COMPLETE) { + call->state = RXRPC_CALL_LOCALLY_ABORTED; + call->local_abort = RX_CALL_TIMEOUT; + set_bit(RXRPC_CALL_EV_ABORT, &call->events); + } + write_unlock_bh(&call->state_lock); + + _debug("post timeout"); + if (rxrpc_post_message(call, RXRPC_SKB_MARK_LOCAL_ERROR, + ETIME, true) < 0) + goto no_mem; + + clear_bit(RXRPC_CALL_EV_LIFE_TIMER, &call->events); + goto kill_ACKs; + } + + /* deal with assorted inbound messages */ + if (!skb_queue_empty(&call->rx_queue)) { + switch (rxrpc_process_rx_queue(call, &abort_code)) { + case 0: + case -EAGAIN: + break; + case -ENOMEM: + goto no_mem; + case -EKEYEXPIRED: + case -EKEYREJECTED: + case -EPROTO: + rxrpc_abort_call(call, abort_code); + goto kill_ACKs; + } + } + + /* handle resending */ + if (test_and_clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events)) + rxrpc_resend_timer(call); + if (test_and_clear_bit(RXRPC_CALL_EV_RESEND, &call->events)) + rxrpc_resend(call); + + /* consider sending an ordinary ACK */ + if (test_bit(RXRPC_CALL_EV_ACK, &call->events)) { + _debug("send ACK: window: %d - %d { %lx }", + call->rx_data_eaten, call->ackr_win_top, + call->ackr_window[0]); + + if (call->state > RXRPC_CALL_SERVER_ACK_REQUEST && + call->ackr_reason != RXRPC_ACK_PING_RESPONSE) { + /* ACK by sending reply DATA packet in this state */ + clear_bit(RXRPC_CALL_EV_ACK, &call->events); + goto maybe_reschedule; + } + + genbit = RXRPC_CALL_EV_ACK; + + acks = kzalloc(call->ackr_win_top - call->rx_data_eaten, + GFP_NOFS); + if (!acks) + goto no_mem; + + //hdr.flags = RXRPC_SLOW_START_OK; + ack.bufferSpace = htons(8); + ack.maxSkew = 0; + + spin_lock_bh(&call->lock); + ack.reason = call->ackr_reason; + ack.serial = htonl(call->ackr_serial); + ack.previousPacket = htonl(call->ackr_prev_seq); + ack.firstPacket = htonl(call->rx_data_eaten + 1); + + ack.nAcks = 0; + for (loop = 0; loop < RXRPC_ACKR_WINDOW_ASZ; loop++) { + nbit = loop * BITS_PER_LONG; + for (bits = call->ackr_window[loop]; bits; bits >>= 1 + ) { + _debug("- l=%d n=%d b=%lx", loop, nbit, bits); + if (bits & 1) { + acks[nbit] = RXRPC_ACK_TYPE_ACK; + ack.nAcks = nbit + 1; + } + nbit++; + } + } + call->ackr_reason = 0; + spin_unlock_bh(&call->lock); + + pad = 0; + + iov[1].iov_base = &ack; + iov[1].iov_len = sizeof(ack); + iov[2].iov_base = acks; + iov[2].iov_len = ack.nAcks; + iov[3].iov_base = &pad; + iov[3].iov_len = 3; + iov[4].iov_base = &ackinfo; + iov[4].iov_len = sizeof(ackinfo); + + switch (ack.reason) { + case RXRPC_ACK_REQUESTED: + case RXRPC_ACK_DUPLICATE: + case RXRPC_ACK_OUT_OF_SEQUENCE: + case RXRPC_ACK_EXCEEDS_WINDOW: + case RXRPC_ACK_NOSPACE: + case RXRPC_ACK_PING: + case RXRPC_ACK_PING_RESPONSE: + goto send_ACK_with_skew; + case RXRPC_ACK_DELAY: + case RXRPC_ACK_IDLE: + goto send_ACK; + } + } + + /* handle completion of security negotiations on an incoming + * connection */ + if (test_and_clear_bit(RXRPC_CALL_EV_SECURED, &call->events)) { + _debug("secured"); + spin_lock_bh(&call->lock); + + if (call->state == RXRPC_CALL_SERVER_SECURING) { + _debug("securing"); + write_lock(&call->conn->lock); + if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && + !test_bit(RXRPC_CALL_EV_RELEASE, &call->events)) { + _debug("not released"); + call->state = RXRPC_CALL_SERVER_ACCEPTING; + list_move_tail(&call->accept_link, + &call->socket->acceptq); + } + write_unlock(&call->conn->lock); + read_lock(&call->state_lock); + if (call->state < RXRPC_CALL_COMPLETE) + set_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events); + read_unlock(&call->state_lock); + } + + spin_unlock_bh(&call->lock); + if (!test_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events)) + goto maybe_reschedule; + } + + /* post a notification of an acceptable connection to the app */ + if (test_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events)) { + _debug("post accept"); + if (rxrpc_post_message(call, RXRPC_SKB_MARK_NEW_CALL, + 0, false) < 0) + goto no_mem; + clear_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events); + goto maybe_reschedule; + } + + /* handle incoming call acceptance */ + if (test_and_clear_bit(RXRPC_CALL_EV_ACCEPTED, &call->events)) { + _debug("accepted"); + ASSERTCMP(call->rx_data_post, ==, 0); + call->rx_data_post = 1; + read_lock_bh(&call->state_lock); + if (call->state < RXRPC_CALL_COMPLETE) + set_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events); + read_unlock_bh(&call->state_lock); + } + + /* drain the out of sequence received packet queue into the packet Rx + * queue */ + if (test_and_clear_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events)) { + while (call->rx_data_post == call->rx_first_oos) + if (rxrpc_drain_rx_oos_queue(call) < 0) + break; + goto maybe_reschedule; + } + + /* other events may have been raised since we started checking */ + goto maybe_reschedule; + +send_ACK_with_skew: + ack.maxSkew = htons(atomic_read(&call->conn->hi_serial) - + ntohl(ack.serial)); +send_ACK: + mtu = call->conn->trans->peer->if_mtu; + mtu -= call->conn->trans->peer->hdrsize; + ackinfo.maxMTU = htonl(mtu); + ackinfo.rwind = htonl(rxrpc_rx_window_size); + + /* permit the peer to send us jumbo packets if it wants to */ + ackinfo.rxMTU = htonl(rxrpc_rx_mtu); + ackinfo.jumbo_max = htonl(rxrpc_rx_jumbo_max); + + serial = atomic_inc_return(&call->conn->serial); + whdr.serial = htonl(serial); + _proto("Tx ACK %%%u { m=%hu f=#%u p=#%u s=%%%u r=%s n=%u }", + serial, + ntohs(ack.maxSkew), + ntohl(ack.firstPacket), + ntohl(ack.previousPacket), + ntohl(ack.serial), + rxrpc_acks(ack.reason), + ack.nAcks); + + del_timer_sync(&call->ack_timer); + if (ack.nAcks > 0) + set_bit(RXRPC_CALL_TX_SOFT_ACK, &call->flags); + goto send_message_2; + +send_message: + _debug("send message"); + + serial = atomic_inc_return(&call->conn->serial); + whdr.serial = htonl(serial); + _proto("Tx %s %%%u", rxrpc_pkts[whdr.type], serial); +send_message_2: + + len = iov[0].iov_len; + ioc = 1; + if (iov[4].iov_len) { + ioc = 5; + len += iov[4].iov_len; + len += iov[3].iov_len; + len += iov[2].iov_len; + len += iov[1].iov_len; + } else if (iov[3].iov_len) { + ioc = 4; + len += iov[3].iov_len; + len += iov[2].iov_len; + len += iov[1].iov_len; + } else if (iov[2].iov_len) { + ioc = 3; + len += iov[2].iov_len; + len += iov[1].iov_len; + } else if (iov[1].iov_len) { + ioc = 2; + len += iov[1].iov_len; + } + + ret = kernel_sendmsg(call->conn->trans->local->socket, + &msg, iov, ioc, len); + if (ret < 0) { + _debug("sendmsg failed: %d", ret); + read_lock_bh(&call->state_lock); + if (call->state < RXRPC_CALL_DEAD) + rxrpc_queue_call(call); + read_unlock_bh(&call->state_lock); + goto error; + } + + switch (genbit) { + case RXRPC_CALL_EV_ABORT: + clear_bit(genbit, &call->events); + clear_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events); + goto kill_ACKs; + + case RXRPC_CALL_EV_ACK_FINAL: + write_lock_bh(&call->state_lock); + if (call->state == RXRPC_CALL_CLIENT_FINAL_ACK) + call->state = RXRPC_CALL_COMPLETE; + write_unlock_bh(&call->state_lock); + goto kill_ACKs; + + default: + clear_bit(genbit, &call->events); + switch (call->state) { + case RXRPC_CALL_CLIENT_AWAIT_REPLY: + case RXRPC_CALL_CLIENT_RECV_REPLY: + case RXRPC_CALL_SERVER_RECV_REQUEST: + case RXRPC_CALL_SERVER_ACK_REQUEST: + _debug("start ACK timer"); + rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, + call->ackr_serial, false); + default: + break; + } + goto maybe_reschedule; + } + +kill_ACKs: + del_timer_sync(&call->ack_timer); + if (test_and_clear_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events)) + rxrpc_put_call(call); + clear_bit(RXRPC_CALL_EV_ACK, &call->events); + +maybe_reschedule: + if (call->events || !skb_queue_empty(&call->rx_queue)) { + read_lock_bh(&call->state_lock); + if (call->state < RXRPC_CALL_DEAD) + rxrpc_queue_call(call); + read_unlock_bh(&call->state_lock); + } + + /* don't leave aborted connections on the accept queue */ + if (call->state >= RXRPC_CALL_COMPLETE && + !list_empty(&call->accept_link)) { + _debug("X unlinking once-pending call %p { e=%lx f=%lx c=%x }", + call, call->events, call->flags, call->conn->cid); + + read_lock_bh(&call->state_lock); + if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && + !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) + rxrpc_queue_call(call); + read_unlock_bh(&call->state_lock); + } + +error: + clear_bit(RXRPC_CALL_PROC_BUSY, &call->flags); + kfree(acks); + + /* because we don't want two CPUs both processing the work item for one + * call at the same time, we use a flag to note when it's busy; however + * this means there's a race between clearing the flag and setting the + * work pending bit and the work item being processed again */ + if (call->events && !work_pending(&call->processor)) { + _debug("jumpstart %x", call->conn->cid); + rxrpc_queue_call(call); + } + + _leave(""); + return; + +no_mem: + _debug("out of memory"); + goto maybe_reschedule; +} diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c new file mode 100644 index 0000000..68125dc --- /dev/null +++ b/net/rxrpc/call_object.c @@ -0,0 +1,980 @@ +/* RxRPC individual remote procedure call handling + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * Maximum lifetime of a call (in jiffies). + */ +unsigned int rxrpc_max_call_lifetime = 60 * HZ; + +/* + * Time till dead call expires after last use (in jiffies). + */ +unsigned int rxrpc_dead_call_expiry = 2 * HZ; + +const char *const rxrpc_call_states[NR__RXRPC_CALL_STATES] = { + [RXRPC_CALL_CLIENT_SEND_REQUEST] = "ClSndReq", + [RXRPC_CALL_CLIENT_AWAIT_REPLY] = "ClAwtRpl", + [RXRPC_CALL_CLIENT_RECV_REPLY] = "ClRcvRpl", + [RXRPC_CALL_CLIENT_FINAL_ACK] = "ClFnlACK", + [RXRPC_CALL_SERVER_SECURING] = "SvSecure", + [RXRPC_CALL_SERVER_ACCEPTING] = "SvAccept", + [RXRPC_CALL_SERVER_RECV_REQUEST] = "SvRcvReq", + [RXRPC_CALL_SERVER_ACK_REQUEST] = "SvAckReq", + [RXRPC_CALL_SERVER_SEND_REPLY] = "SvSndRpl", + [RXRPC_CALL_SERVER_AWAIT_ACK] = "SvAwtACK", + [RXRPC_CALL_COMPLETE] = "Complete", + [RXRPC_CALL_SERVER_BUSY] = "SvBusy ", + [RXRPC_CALL_REMOTELY_ABORTED] = "RmtAbort", + [RXRPC_CALL_LOCALLY_ABORTED] = "LocAbort", + [RXRPC_CALL_NETWORK_ERROR] = "NetError", + [RXRPC_CALL_DEAD] = "Dead ", +}; + +struct kmem_cache *rxrpc_call_jar; +LIST_HEAD(rxrpc_calls); +DEFINE_RWLOCK(rxrpc_call_lock); + +static void rxrpc_destroy_call(struct work_struct *work); +static void rxrpc_call_life_expired(unsigned long _call); +static void rxrpc_dead_call_expired(unsigned long _call); +static void rxrpc_ack_time_expired(unsigned long _call); +static void rxrpc_resend_time_expired(unsigned long _call); + +static DEFINE_SPINLOCK(rxrpc_call_hash_lock); +static DEFINE_HASHTABLE(rxrpc_call_hash, 10); + +/* + * Hash function for rxrpc_call_hash + */ +static unsigned long rxrpc_call_hashfunc( + u8 in_clientflag, + u32 cid, + u32 call_id, + u32 epoch, + u16 service_id, + sa_family_t proto, + void *localptr, + unsigned int addr_size, + const u8 *peer_addr) +{ + const u16 *p; + unsigned int i; + unsigned long key; + + _enter(""); + + key = (unsigned long)localptr; + /* We just want to add up the __be32 values, so forcing the + * cast should be okay. + */ + key += epoch; + key += service_id; + key += call_id; + key += (cid & RXRPC_CIDMASK) >> RXRPC_CIDSHIFT; + key += cid & RXRPC_CHANNELMASK; + key += in_clientflag; + key += proto; + /* Step through the peer address in 16-bit portions for speed */ + for (i = 0, p = (const u16 *)peer_addr; i < addr_size >> 1; i++, p++) + key += *p; + _leave(" key = 0x%lx", key); + return key; +} + +/* + * Add a call to the hashtable + */ +static void rxrpc_call_hash_add(struct rxrpc_call *call) +{ + unsigned long key; + unsigned int addr_size = 0; + + _enter(""); + switch (call->proto) { + case AF_INET: + addr_size = sizeof(call->peer_ip.ipv4_addr); + break; + case AF_INET6: + addr_size = sizeof(call->peer_ip.ipv6_addr); + break; + default: + break; + } + key = rxrpc_call_hashfunc(call->in_clientflag, call->cid, + call->call_id, call->epoch, + call->service_id, call->proto, + call->conn->trans->local, addr_size, + call->peer_ip.ipv6_addr); + /* Store the full key in the call */ + call->hash_key = key; + spin_lock(&rxrpc_call_hash_lock); + hash_add_rcu(rxrpc_call_hash, &call->hash_node, key); + spin_unlock(&rxrpc_call_hash_lock); + _leave(""); +} + +/* + * Remove a call from the hashtable + */ +static void rxrpc_call_hash_del(struct rxrpc_call *call) +{ + _enter(""); + spin_lock(&rxrpc_call_hash_lock); + hash_del_rcu(&call->hash_node); + spin_unlock(&rxrpc_call_hash_lock); + _leave(""); +} + +/* + * Find a call in the hashtable and return it, or NULL if it + * isn't there. + */ +struct rxrpc_call *rxrpc_find_call_hash( + struct rxrpc_host_header *hdr, + void *localptr, + sa_family_t proto, + const void *peer_addr) +{ + unsigned long key; + unsigned int addr_size = 0; + struct rxrpc_call *call = NULL; + struct rxrpc_call *ret = NULL; + u8 in_clientflag = hdr->flags & RXRPC_CLIENT_INITIATED; + + _enter(""); + switch (proto) { + case AF_INET: + addr_size = sizeof(call->peer_ip.ipv4_addr); + break; + case AF_INET6: + addr_size = sizeof(call->peer_ip.ipv6_addr); + break; + default: + break; + } + + key = rxrpc_call_hashfunc(in_clientflag, hdr->cid, hdr->callNumber, + hdr->epoch, hdr->serviceId, + proto, localptr, addr_size, + peer_addr); + hash_for_each_possible_rcu(rxrpc_call_hash, call, hash_node, key) { + if (call->hash_key == key && + call->call_id == hdr->callNumber && + call->cid == hdr->cid && + call->in_clientflag == in_clientflag && + call->service_id == hdr->serviceId && + call->proto == proto && + call->local == localptr && + memcmp(call->peer_ip.ipv6_addr, peer_addr, + addr_size) == 0 && + call->epoch == hdr->epoch) { + ret = call; + break; + } + } + _leave(" = %p", ret); + return ret; +} + +/* + * find an extant server call + * - called in process context with IRQs enabled + */ +struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *rx, + unsigned long user_call_ID) +{ + struct rxrpc_call *call; + struct rb_node *p; + + _enter("%p,%lx", rx, user_call_ID); + + read_lock(&rx->call_lock); + + p = rx->calls.rb_node; + while (p) { + call = rb_entry(p, struct rxrpc_call, sock_node); + + if (user_call_ID < call->user_call_ID) + p = p->rb_left; + else if (user_call_ID > call->user_call_ID) + p = p->rb_right; + else + goto found_extant_call; + } + + read_unlock(&rx->call_lock); + _leave(" = NULL"); + return NULL; + +found_extant_call: + rxrpc_get_call(call); + read_unlock(&rx->call_lock); + _leave(" = %p [%d]", call, atomic_read(&call->usage)); + return call; +} + +/* + * allocate a new call + */ +static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp) +{ + struct rxrpc_call *call; + + call = kmem_cache_zalloc(rxrpc_call_jar, gfp); + if (!call) + return NULL; + + call->acks_winsz = 16; + call->acks_window = kmalloc(call->acks_winsz * sizeof(unsigned long), + gfp); + if (!call->acks_window) { + kmem_cache_free(rxrpc_call_jar, call); + return NULL; + } + + setup_timer(&call->lifetimer, &rxrpc_call_life_expired, + (unsigned long) call); + setup_timer(&call->deadspan, &rxrpc_dead_call_expired, + (unsigned long) call); + setup_timer(&call->ack_timer, &rxrpc_ack_time_expired, + (unsigned long) call); + setup_timer(&call->resend_timer, &rxrpc_resend_time_expired, + (unsigned long) call); + INIT_WORK(&call->destroyer, &rxrpc_destroy_call); + INIT_WORK(&call->processor, &rxrpc_process_call); + INIT_LIST_HEAD(&call->accept_link); + skb_queue_head_init(&call->rx_queue); + skb_queue_head_init(&call->rx_oos_queue); + init_waitqueue_head(&call->tx_waitq); + spin_lock_init(&call->lock); + rwlock_init(&call->state_lock); + atomic_set(&call->usage, 1); + call->debug_id = atomic_inc_return(&rxrpc_debug_id); + call->state = RXRPC_CALL_CLIENT_SEND_REQUEST; + + memset(&call->sock_node, 0xed, sizeof(call->sock_node)); + + call->rx_data_expect = 1; + call->rx_data_eaten = 0; + call->rx_first_oos = 0; + call->ackr_win_top = call->rx_data_eaten + 1 + rxrpc_rx_window_size; + call->creation_jif = jiffies; + return call; +} + +/* + * allocate a new client call and attempt to get a connection slot for it + */ +static struct rxrpc_call *rxrpc_alloc_client_call( + struct rxrpc_sock *rx, + struct rxrpc_transport *trans, + struct rxrpc_conn_bundle *bundle, + gfp_t gfp) +{ + struct rxrpc_call *call; + int ret; + + _enter(""); + + ASSERT(rx != NULL); + ASSERT(trans != NULL); + ASSERT(bundle != NULL); + + call = rxrpc_alloc_call(gfp); + if (!call) + return ERR_PTR(-ENOMEM); + + sock_hold(&rx->sk); + call->socket = rx; + call->rx_data_post = 1; + + ret = rxrpc_connect_call(rx, trans, bundle, call, gfp); + if (ret < 0) { + kmem_cache_free(rxrpc_call_jar, call); + return ERR_PTR(ret); + } + + /* Record copies of information for hashtable lookup */ + call->proto = rx->proto; + call->local = trans->local; + switch (call->proto) { + case AF_INET: + call->peer_ip.ipv4_addr = + trans->peer->srx.transport.sin.sin_addr.s_addr; + break; + case AF_INET6: + memcpy(call->peer_ip.ipv6_addr, + trans->peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8, + sizeof(call->peer_ip.ipv6_addr)); + break; + } + call->epoch = call->conn->epoch; + call->service_id = call->conn->service_id; + call->in_clientflag = call->conn->in_clientflag; + /* Add the new call to the hashtable */ + rxrpc_call_hash_add(call); + + spin_lock(&call->conn->trans->peer->lock); + list_add(&call->error_link, &call->conn->trans->peer->error_targets); + spin_unlock(&call->conn->trans->peer->lock); + + call->lifetimer.expires = jiffies + rxrpc_max_call_lifetime; + add_timer(&call->lifetimer); + + _leave(" = %p", call); + return call; +} + +/* + * set up a call for the given data + * - called in process context with IRQs enabled + */ +struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, + struct rxrpc_transport *trans, + struct rxrpc_conn_bundle *bundle, + unsigned long user_call_ID, + gfp_t gfp) +{ + struct rxrpc_call *call, *xcall; + struct rb_node *parent, **pp; + + _enter("%p,%d,%d,%lx", + rx, trans->debug_id, bundle ? bundle->debug_id : -1, + user_call_ID); + + call = rxrpc_alloc_client_call(rx, trans, bundle, gfp); + if (IS_ERR(call)) { + _leave(" = %ld", PTR_ERR(call)); + return call; + } + + call->user_call_ID = user_call_ID; + __set_bit(RXRPC_CALL_HAS_USERID, &call->flags); + + write_lock(&rx->call_lock); + + pp = &rx->calls.rb_node; + parent = NULL; + while (*pp) { + parent = *pp; + xcall = rb_entry(parent, struct rxrpc_call, sock_node); + + if (user_call_ID < xcall->user_call_ID) + pp = &(*pp)->rb_left; + else if (user_call_ID > xcall->user_call_ID) + pp = &(*pp)->rb_right; + else + goto found_user_ID_now_present; + } + + rxrpc_get_call(call); + + rb_link_node(&call->sock_node, parent, pp); + rb_insert_color(&call->sock_node, &rx->calls); + write_unlock(&rx->call_lock); + + write_lock_bh(&rxrpc_call_lock); + list_add_tail(&call->link, &rxrpc_calls); + write_unlock_bh(&rxrpc_call_lock); + + _net("CALL new %d on CONN %d", call->debug_id, call->conn->debug_id); + + _leave(" = %p [new]", call); + return call; + + /* We unexpectedly found the user ID in the list after taking + * the call_lock. This shouldn't happen unless the user races + * with itself and tries to add the same user ID twice at the + * same time in different threads. + */ +found_user_ID_now_present: + write_unlock(&rx->call_lock); + rxrpc_put_call(call); + _leave(" = -EEXIST [%p]", call); + return ERR_PTR(-EEXIST); +} + +/* + * set up an incoming call + * - called in process context with IRQs enabled + */ +struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, + struct rxrpc_connection *conn, + struct rxrpc_host_header *hdr) +{ + struct rxrpc_call *call, *candidate; + struct rb_node **p, *parent; + u32 call_id; + + _enter(",%d", conn->debug_id); + + ASSERT(rx != NULL); + + candidate = rxrpc_alloc_call(GFP_NOIO); + if (!candidate) + return ERR_PTR(-EBUSY); + + candidate->socket = rx; + candidate->conn = conn; + candidate->cid = hdr->cid; + candidate->call_id = hdr->callNumber; + candidate->channel = hdr->cid & RXRPC_CHANNELMASK; + candidate->rx_data_post = 0; + candidate->state = RXRPC_CALL_SERVER_ACCEPTING; + if (conn->security_ix > 0) + candidate->state = RXRPC_CALL_SERVER_SECURING; + + write_lock_bh(&conn->lock); + + /* set the channel for this call */ + call = conn->channels[candidate->channel]; + _debug("channel[%u] is %p", candidate->channel, call); + if (call && call->call_id == hdr->callNumber) { + /* already set; must've been a duplicate packet */ + _debug("extant call [%d]", call->state); + ASSERTCMP(call->conn, ==, conn); + + read_lock(&call->state_lock); + switch (call->state) { + case RXRPC_CALL_LOCALLY_ABORTED: + if (!test_and_set_bit(RXRPC_CALL_EV_ABORT, &call->events)) + rxrpc_queue_call(call); + case RXRPC_CALL_REMOTELY_ABORTED: + read_unlock(&call->state_lock); + goto aborted_call; + default: + rxrpc_get_call(call); + read_unlock(&call->state_lock); + goto extant_call; + } + } + + if (call) { + /* it seems the channel is still in use from the previous call + * - ditch the old binding if its call is now complete */ + _debug("CALL: %u { %s }", + call->debug_id, rxrpc_call_states[call->state]); + + if (call->state >= RXRPC_CALL_COMPLETE) { + conn->channels[call->channel] = NULL; + } else { + write_unlock_bh(&conn->lock); + kmem_cache_free(rxrpc_call_jar, candidate); + _leave(" = -EBUSY"); + return ERR_PTR(-EBUSY); + } + } + + /* check the call number isn't duplicate */ + _debug("check dup"); + call_id = hdr->callNumber; + p = &conn->calls.rb_node; + parent = NULL; + while (*p) { + parent = *p; + call = rb_entry(parent, struct rxrpc_call, conn_node); + + /* The tree is sorted in order of the __be32 value without + * turning it into host order. + */ + if (call_id < call->call_id) + p = &(*p)->rb_left; + else if (call_id > call->call_id) + p = &(*p)->rb_right; + else + goto old_call; + } + + /* make the call available */ + _debug("new call"); + call = candidate; + candidate = NULL; + rb_link_node(&call->conn_node, parent, p); + rb_insert_color(&call->conn_node, &conn->calls); + conn->channels[call->channel] = call; + sock_hold(&rx->sk); + atomic_inc(&conn->usage); + write_unlock_bh(&conn->lock); + + spin_lock(&conn->trans->peer->lock); + list_add(&call->error_link, &conn->trans->peer->error_targets); + spin_unlock(&conn->trans->peer->lock); + + write_lock_bh(&rxrpc_call_lock); + list_add_tail(&call->link, &rxrpc_calls); + write_unlock_bh(&rxrpc_call_lock); + + /* Record copies of information for hashtable lookup */ + call->proto = rx->proto; + call->local = conn->trans->local; + switch (call->proto) { + case AF_INET: + call->peer_ip.ipv4_addr = + conn->trans->peer->srx.transport.sin.sin_addr.s_addr; + break; + case AF_INET6: + memcpy(call->peer_ip.ipv6_addr, + conn->trans->peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8, + sizeof(call->peer_ip.ipv6_addr)); + break; + default: + break; + } + call->epoch = conn->epoch; + call->service_id = conn->service_id; + call->in_clientflag = conn->in_clientflag; + /* Add the new call to the hashtable */ + rxrpc_call_hash_add(call); + + _net("CALL incoming %d on CONN %d", call->debug_id, call->conn->debug_id); + + call->lifetimer.expires = jiffies + rxrpc_max_call_lifetime; + add_timer(&call->lifetimer); + _leave(" = %p {%d} [new]", call, call->debug_id); + return call; + +extant_call: + write_unlock_bh(&conn->lock); + kmem_cache_free(rxrpc_call_jar, candidate); + _leave(" = %p {%d} [extant]", call, call ? call->debug_id : -1); + return call; + +aborted_call: + write_unlock_bh(&conn->lock); + kmem_cache_free(rxrpc_call_jar, candidate); + _leave(" = -ECONNABORTED"); + return ERR_PTR(-ECONNABORTED); + +old_call: + write_unlock_bh(&conn->lock); + kmem_cache_free(rxrpc_call_jar, candidate); + _leave(" = -ECONNRESET [old]"); + return ERR_PTR(-ECONNRESET); +} + +/* + * detach a call from a socket and set up for release + */ +void rxrpc_release_call(struct rxrpc_call *call) +{ + struct rxrpc_connection *conn = call->conn; + struct rxrpc_sock *rx = call->socket; + + _enter("{%d,%d,%d,%d}", + call->debug_id, atomic_read(&call->usage), + atomic_read(&call->ackr_not_idle), + call->rx_first_oos); + + spin_lock_bh(&call->lock); + if (test_and_set_bit(RXRPC_CALL_RELEASED, &call->flags)) + BUG(); + spin_unlock_bh(&call->lock); + + /* dissociate from the socket + * - the socket's ref on the call is passed to the death timer + */ + _debug("RELEASE CALL %p (%d CONN %p)", call, call->debug_id, conn); + + write_lock_bh(&rx->call_lock); + if (!list_empty(&call->accept_link)) { + _debug("unlinking once-pending call %p { e=%lx f=%lx }", + call, call->events, call->flags); + ASSERT(!test_bit(RXRPC_CALL_HAS_USERID, &call->flags)); + list_del_init(&call->accept_link); + sk_acceptq_removed(&rx->sk); + } else if (test_bit(RXRPC_CALL_HAS_USERID, &call->flags)) { + rb_erase(&call->sock_node, &rx->calls); + memset(&call->sock_node, 0xdd, sizeof(call->sock_node)); + clear_bit(RXRPC_CALL_HAS_USERID, &call->flags); + } + write_unlock_bh(&rx->call_lock); + + /* free up the channel for reuse */ + spin_lock(&conn->trans->client_lock); + write_lock_bh(&conn->lock); + write_lock(&call->state_lock); + + if (conn->channels[call->channel] == call) + conn->channels[call->channel] = NULL; + + if (conn->out_clientflag && conn->bundle) { + conn->avail_calls++; + switch (conn->avail_calls) { + case 1: + list_move_tail(&conn->bundle_link, + &conn->bundle->avail_conns); + case 2 ... RXRPC_MAXCALLS - 1: + ASSERT(conn->channels[0] == NULL || + conn->channels[1] == NULL || + conn->channels[2] == NULL || + conn->channels[3] == NULL); + break; + case RXRPC_MAXCALLS: + list_move_tail(&conn->bundle_link, + &conn->bundle->unused_conns); + ASSERT(conn->channels[0] == NULL && + conn->channels[1] == NULL && + conn->channels[2] == NULL && + conn->channels[3] == NULL); + break; + default: + pr_err("conn->avail_calls=%d\n", conn->avail_calls); + BUG(); + } + } + + spin_unlock(&conn->trans->client_lock); + + if (call->state < RXRPC_CALL_COMPLETE && + call->state != RXRPC_CALL_CLIENT_FINAL_ACK) { + _debug("+++ ABORTING STATE %d +++\n", call->state); + call->state = RXRPC_CALL_LOCALLY_ABORTED; + call->local_abort = RX_CALL_DEAD; + set_bit(RXRPC_CALL_EV_ABORT, &call->events); + rxrpc_queue_call(call); + } + write_unlock(&call->state_lock); + write_unlock_bh(&conn->lock); + + /* clean up the Rx queue */ + if (!skb_queue_empty(&call->rx_queue) || + !skb_queue_empty(&call->rx_oos_queue)) { + struct rxrpc_skb_priv *sp; + struct sk_buff *skb; + + _debug("purge Rx queues"); + + spin_lock_bh(&call->lock); + while ((skb = skb_dequeue(&call->rx_queue)) || + (skb = skb_dequeue(&call->rx_oos_queue))) { + sp = rxrpc_skb(skb); + if (sp->call) { + ASSERTCMP(sp->call, ==, call); + rxrpc_put_call(call); + sp->call = NULL; + } + skb->destructor = NULL; + spin_unlock_bh(&call->lock); + + _debug("- zap %s %%%u #%u", + rxrpc_pkts[sp->hdr.type], + sp->hdr.serial, sp->hdr.seq); + rxrpc_free_skb(skb); + spin_lock_bh(&call->lock); + } + spin_unlock_bh(&call->lock); + + ASSERTCMP(call->state, !=, RXRPC_CALL_COMPLETE); + } + + del_timer_sync(&call->resend_timer); + del_timer_sync(&call->ack_timer); + del_timer_sync(&call->lifetimer); + call->deadspan.expires = jiffies + rxrpc_dead_call_expiry; + add_timer(&call->deadspan); + + _leave(""); +} + +/* + * handle a dead call being ready for reaping + */ +static void rxrpc_dead_call_expired(unsigned long _call) +{ + struct rxrpc_call *call = (struct rxrpc_call *) _call; + + _enter("{%d}", call->debug_id); + + write_lock_bh(&call->state_lock); + call->state = RXRPC_CALL_DEAD; + write_unlock_bh(&call->state_lock); + rxrpc_put_call(call); +} + +/* + * mark a call as to be released, aborting it if it's still in progress + * - called with softirqs disabled + */ +static void rxrpc_mark_call_released(struct rxrpc_call *call) +{ + bool sched; + + write_lock(&call->state_lock); + if (call->state < RXRPC_CALL_DEAD) { + sched = false; + if (call->state < RXRPC_CALL_COMPLETE) { + _debug("abort call %p", call); + call->state = RXRPC_CALL_LOCALLY_ABORTED; + call->local_abort = RX_CALL_DEAD; + if (!test_and_set_bit(RXRPC_CALL_EV_ABORT, &call->events)) + sched = true; + } + if (!test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) + sched = true; + if (sched) + rxrpc_queue_call(call); + } + write_unlock(&call->state_lock); +} + +/* + * release all the calls associated with a socket + */ +void rxrpc_release_calls_on_socket(struct rxrpc_sock *rx) +{ + struct rxrpc_call *call; + struct rb_node *p; + + _enter("%p", rx); + + read_lock_bh(&rx->call_lock); + + /* mark all the calls as no longer wanting incoming packets */ + for (p = rb_first(&rx->calls); p; p = rb_next(p)) { + call = rb_entry(p, struct rxrpc_call, sock_node); + rxrpc_mark_call_released(call); + } + + /* kill the not-yet-accepted incoming calls */ + list_for_each_entry(call, &rx->secureq, accept_link) { + rxrpc_mark_call_released(call); + } + + list_for_each_entry(call, &rx->acceptq, accept_link) { + rxrpc_mark_call_released(call); + } + + read_unlock_bh(&rx->call_lock); + _leave(""); +} + +/* + * release a call + */ +void __rxrpc_put_call(struct rxrpc_call *call) +{ + ASSERT(call != NULL); + + _enter("%p{u=%d}", call, atomic_read(&call->usage)); + + ASSERTCMP(atomic_read(&call->usage), >, 0); + + if (atomic_dec_and_test(&call->usage)) { + _debug("call %d dead", call->debug_id); + ASSERTCMP(call->state, ==, RXRPC_CALL_DEAD); + rxrpc_queue_work(&call->destroyer); + } + _leave(""); +} + +/* + * clean up a call + */ +static void rxrpc_cleanup_call(struct rxrpc_call *call) +{ + _net("DESTROY CALL %d", call->debug_id); + + ASSERT(call->socket); + + memset(&call->sock_node, 0xcd, sizeof(call->sock_node)); + + del_timer_sync(&call->lifetimer); + del_timer_sync(&call->deadspan); + del_timer_sync(&call->ack_timer); + del_timer_sync(&call->resend_timer); + + ASSERT(test_bit(RXRPC_CALL_RELEASED, &call->flags)); + ASSERTCMP(call->events, ==, 0); + if (work_pending(&call->processor)) { + _debug("defer destroy"); + rxrpc_queue_work(&call->destroyer); + return; + } + + if (call->conn) { + spin_lock(&call->conn->trans->peer->lock); + list_del(&call->error_link); + spin_unlock(&call->conn->trans->peer->lock); + + write_lock_bh(&call->conn->lock); + rb_erase(&call->conn_node, &call->conn->calls); + write_unlock_bh(&call->conn->lock); + rxrpc_put_connection(call->conn); + } + + /* Remove the call from the hash */ + rxrpc_call_hash_del(call); + + if (call->acks_window) { + _debug("kill Tx window %d", + CIRC_CNT(call->acks_head, call->acks_tail, + call->acks_winsz)); + smp_mb(); + while (CIRC_CNT(call->acks_head, call->acks_tail, + call->acks_winsz) > 0) { + struct rxrpc_skb_priv *sp; + unsigned long _skb; + + _skb = call->acks_window[call->acks_tail] & ~1; + sp = rxrpc_skb((struct sk_buff *)_skb); + _debug("+++ clear Tx %u", sp->hdr.seq); + rxrpc_free_skb((struct sk_buff *)_skb); + call->acks_tail = + (call->acks_tail + 1) & (call->acks_winsz - 1); + } + + kfree(call->acks_window); + } + + rxrpc_free_skb(call->tx_pending); + + rxrpc_purge_queue(&call->rx_queue); + ASSERT(skb_queue_empty(&call->rx_oos_queue)); + sock_put(&call->socket->sk); + kmem_cache_free(rxrpc_call_jar, call); +} + +/* + * destroy a call + */ +static void rxrpc_destroy_call(struct work_struct *work) +{ + struct rxrpc_call *call = + container_of(work, struct rxrpc_call, destroyer); + + _enter("%p{%d,%d,%p}", + call, atomic_read(&call->usage), call->channel, call->conn); + + ASSERTCMP(call->state, ==, RXRPC_CALL_DEAD); + + write_lock_bh(&rxrpc_call_lock); + list_del_init(&call->link); + write_unlock_bh(&rxrpc_call_lock); + + rxrpc_cleanup_call(call); + _leave(""); +} + +/* + * preemptively destroy all the call records from a transport endpoint rather + * than waiting for them to time out + */ +void __exit rxrpc_destroy_all_calls(void) +{ + struct rxrpc_call *call; + + _enter(""); + write_lock_bh(&rxrpc_call_lock); + + while (!list_empty(&rxrpc_calls)) { + call = list_entry(rxrpc_calls.next, struct rxrpc_call, link); + _debug("Zapping call %p", call); + + list_del_init(&call->link); + + switch (atomic_read(&call->usage)) { + case 0: + ASSERTCMP(call->state, ==, RXRPC_CALL_DEAD); + break; + case 1: + if (del_timer_sync(&call->deadspan) != 0 && + call->state != RXRPC_CALL_DEAD) + rxrpc_dead_call_expired((unsigned long) call); + if (call->state != RXRPC_CALL_DEAD) + break; + default: + pr_err("Call %p still in use (%d,%d,%s,%lx,%lx)!\n", + call, atomic_read(&call->usage), + atomic_read(&call->ackr_not_idle), + rxrpc_call_states[call->state], + call->flags, call->events); + if (!skb_queue_empty(&call->rx_queue)) + pr_err("Rx queue occupied\n"); + if (!skb_queue_empty(&call->rx_oos_queue)) + pr_err("OOS queue occupied\n"); + break; + } + + write_unlock_bh(&rxrpc_call_lock); + cond_resched(); + write_lock_bh(&rxrpc_call_lock); + } + + write_unlock_bh(&rxrpc_call_lock); + _leave(""); +} + +/* + * handle call lifetime being exceeded + */ +static void rxrpc_call_life_expired(unsigned long _call) +{ + struct rxrpc_call *call = (struct rxrpc_call *) _call; + + if (call->state >= RXRPC_CALL_COMPLETE) + return; + + _enter("{%d}", call->debug_id); + read_lock_bh(&call->state_lock); + if (call->state < RXRPC_CALL_COMPLETE) { + set_bit(RXRPC_CALL_EV_LIFE_TIMER, &call->events); + rxrpc_queue_call(call); + } + read_unlock_bh(&call->state_lock); +} + +/* + * handle resend timer expiry + * - may not take call->state_lock as this can deadlock against del_timer_sync() + */ +static void rxrpc_resend_time_expired(unsigned long _call) +{ + struct rxrpc_call *call = (struct rxrpc_call *) _call; + + _enter("{%d}", call->debug_id); + + if (call->state >= RXRPC_CALL_COMPLETE) + return; + + clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); + if (!test_and_set_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events)) + rxrpc_queue_call(call); +} + +/* + * handle ACK timer expiry + */ +static void rxrpc_ack_time_expired(unsigned long _call) +{ + struct rxrpc_call *call = (struct rxrpc_call *) _call; + + _enter("{%d}", call->debug_id); + + if (call->state >= RXRPC_CALL_COMPLETE) + return; + + read_lock_bh(&call->state_lock); + if (call->state < RXRPC_CALL_COMPLETE && + !test_and_set_bit(RXRPC_CALL_EV_ACK, &call->events)) + rxrpc_queue_call(call); + read_unlock_bh(&call->state_lock); +} diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c new file mode 100644 index 0000000..8bdd692 --- /dev/null +++ b/net/rxrpc/conn_event.c @@ -0,0 +1,403 @@ +/* connection-level event handling + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * pass a connection-level abort onto all calls on that connection + */ +static void rxrpc_abort_calls(struct rxrpc_connection *conn, int state, + u32 abort_code) +{ + struct rxrpc_call *call; + struct rb_node *p; + + _enter("{%d},%x", conn->debug_id, abort_code); + + read_lock_bh(&conn->lock); + + for (p = rb_first(&conn->calls); p; p = rb_next(p)) { + call = rb_entry(p, struct rxrpc_call, conn_node); + write_lock(&call->state_lock); + if (call->state <= RXRPC_CALL_COMPLETE) { + call->state = state; + if (state == RXRPC_CALL_LOCALLY_ABORTED) { + call->local_abort = conn->local_abort; + set_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events); + } else { + call->remote_abort = conn->remote_abort; + set_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events); + } + rxrpc_queue_call(call); + } + write_unlock(&call->state_lock); + } + + read_unlock_bh(&conn->lock); + _leave(""); +} + +/* + * generate a connection-level abort + */ +static int rxrpc_abort_connection(struct rxrpc_connection *conn, + u32 error, u32 abort_code) +{ + struct rxrpc_wire_header whdr; + struct msghdr msg; + struct kvec iov[2]; + __be32 word; + size_t len; + u32 serial; + int ret; + + _enter("%d,,%u,%u", conn->debug_id, error, abort_code); + + /* generate a connection-level abort */ + spin_lock_bh(&conn->state_lock); + if (conn->state < RXRPC_CONN_REMOTELY_ABORTED) { + conn->state = RXRPC_CONN_LOCALLY_ABORTED; + conn->error = error; + spin_unlock_bh(&conn->state_lock); + } else { + spin_unlock_bh(&conn->state_lock); + _leave(" = 0 [already dead]"); + return 0; + } + + rxrpc_abort_calls(conn, RXRPC_CALL_LOCALLY_ABORTED, abort_code); + + msg.msg_name = &conn->trans->peer->srx.transport; + msg.msg_namelen = conn->trans->peer->srx.transport_len; + msg.msg_control = NULL; + msg.msg_controllen = 0; + msg.msg_flags = 0; + + whdr.epoch = htonl(conn->epoch); + whdr.cid = htonl(conn->cid); + whdr.callNumber = 0; + whdr.seq = 0; + whdr.type = RXRPC_PACKET_TYPE_ABORT; + whdr.flags = conn->out_clientflag; + whdr.userStatus = 0; + whdr.securityIndex = conn->security_ix; + whdr._rsvd = 0; + whdr.serviceId = htons(conn->service_id); + + word = htonl(conn->local_abort); + + iov[0].iov_base = &whdr; + iov[0].iov_len = sizeof(whdr); + iov[1].iov_base = &word; + iov[1].iov_len = sizeof(word); + + len = iov[0].iov_len + iov[1].iov_len; + + serial = atomic_inc_return(&conn->serial); + whdr.serial = htonl(serial); + _proto("Tx CONN ABORT %%%u { %d }", serial, conn->local_abort); + + ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 2, len); + if (ret < 0) { + _debug("sendmsg failed: %d", ret); + return -EAGAIN; + } + + _leave(" = 0"); + return 0; +} + +/* + * mark a call as being on a now-secured channel + * - must be called with softirqs disabled + */ +static void rxrpc_call_is_secure(struct rxrpc_call *call) +{ + _enter("%p", call); + if (call) { + read_lock(&call->state_lock); + if (call->state < RXRPC_CALL_COMPLETE && + !test_and_set_bit(RXRPC_CALL_EV_SECURED, &call->events)) + rxrpc_queue_call(call); + read_unlock(&call->state_lock); + } +} + +/* + * connection-level Rx packet processor + */ +static int rxrpc_process_event(struct rxrpc_connection *conn, + struct sk_buff *skb, + u32 *_abort_code) +{ + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + __be32 wtmp; + u32 abort_code; + int loop, ret; + + if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) { + kleave(" = -ECONNABORTED [%u]", conn->state); + return -ECONNABORTED; + } + + _enter("{%d},{%u,%%%u},", conn->debug_id, sp->hdr.type, sp->hdr.serial); + + switch (sp->hdr.type) { + case RXRPC_PACKET_TYPE_ABORT: + if (skb_copy_bits(skb, 0, &wtmp, sizeof(wtmp)) < 0) + return -EPROTO; + abort_code = ntohl(wtmp); + _proto("Rx ABORT %%%u { ac=%d }", sp->hdr.serial, abort_code); + + conn->state = RXRPC_CONN_REMOTELY_ABORTED; + rxrpc_abort_calls(conn, RXRPC_CALL_REMOTELY_ABORTED, + abort_code); + return -ECONNABORTED; + + case RXRPC_PACKET_TYPE_CHALLENGE: + return conn->security->respond_to_challenge(conn, skb, + _abort_code); + + case RXRPC_PACKET_TYPE_RESPONSE: + ret = conn->security->verify_response(conn, skb, _abort_code); + if (ret < 0) + return ret; + + ret = conn->security->init_connection_security(conn); + if (ret < 0) + return ret; + + conn->security->prime_packet_security(conn); + read_lock_bh(&conn->lock); + spin_lock(&conn->state_lock); + + if (conn->state == RXRPC_CONN_SERVER_CHALLENGING) { + conn->state = RXRPC_CONN_SERVER; + for (loop = 0; loop < RXRPC_MAXCALLS; loop++) + rxrpc_call_is_secure(conn->channels[loop]); + } + + spin_unlock(&conn->state_lock); + read_unlock_bh(&conn->lock); + return 0; + + default: + _leave(" = -EPROTO [%u]", sp->hdr.type); + return -EPROTO; + } +} + +/* + * set up security and issue a challenge + */ +static void rxrpc_secure_connection(struct rxrpc_connection *conn) +{ + u32 abort_code; + int ret; + + _enter("{%d}", conn->debug_id); + + ASSERT(conn->security_ix != 0); + + if (!conn->key) { + _debug("set up security"); + ret = rxrpc_init_server_conn_security(conn); + switch (ret) { + case 0: + break; + case -ENOENT: + abort_code = RX_CALL_DEAD; + goto abort; + default: + abort_code = RXKADNOAUTH; + goto abort; + } + } + + if (conn->security->issue_challenge(conn) < 0) { + abort_code = RX_CALL_DEAD; + ret = -ENOMEM; + goto abort; + } + + _leave(""); + return; + +abort: + _debug("abort %d, %d", ret, abort_code); + rxrpc_abort_connection(conn, -ret, abort_code); + _leave(" [aborted]"); +} + +/* + * connection-level event processor + */ +void rxrpc_process_connection(struct work_struct *work) +{ + struct rxrpc_connection *conn = + container_of(work, struct rxrpc_connection, processor); + struct sk_buff *skb; + u32 abort_code = RX_PROTOCOL_ERROR; + int ret; + + _enter("{%d}", conn->debug_id); + + atomic_inc(&conn->usage); + + if (test_and_clear_bit(RXRPC_CONN_CHALLENGE, &conn->events)) { + rxrpc_secure_connection(conn); + rxrpc_put_connection(conn); + } + + /* go through the conn-level event packets, releasing the ref on this + * connection that each one has when we've finished with it */ + while ((skb = skb_dequeue(&conn->rx_queue))) { + ret = rxrpc_process_event(conn, skb, &abort_code); + switch (ret) { + case -EPROTO: + case -EKEYEXPIRED: + case -EKEYREJECTED: + goto protocol_error; + case -EAGAIN: + goto requeue_and_leave; + case -ECONNABORTED: + default: + rxrpc_put_connection(conn); + rxrpc_free_skb(skb); + break; + } + } + +out: + rxrpc_put_connection(conn); + _leave(""); + return; + +requeue_and_leave: + skb_queue_head(&conn->rx_queue, skb); + goto out; + +protocol_error: + if (rxrpc_abort_connection(conn, -ret, abort_code) < 0) + goto requeue_and_leave; + rxrpc_put_connection(conn); + rxrpc_free_skb(skb); + _leave(" [EPROTO]"); + goto out; +} + +/* + * put a packet up for transport-level abort + */ +void rxrpc_reject_packet(struct rxrpc_local *local, struct sk_buff *skb) +{ + CHECK_SLAB_OKAY(&local->usage); + + if (!atomic_inc_not_zero(&local->usage)) { + printk("resurrected on reject\n"); + BUG(); + } + + skb_queue_tail(&local->reject_queue, skb); + rxrpc_queue_work(&local->rejecter); +} + +/* + * reject packets through the local endpoint + */ +void rxrpc_reject_packets(struct work_struct *work) +{ + union { + struct sockaddr sa; + struct sockaddr_in sin; + } sa; + struct rxrpc_skb_priv *sp; + struct rxrpc_wire_header whdr; + struct rxrpc_local *local; + struct sk_buff *skb; + struct msghdr msg; + struct kvec iov[2]; + size_t size; + __be32 code; + + local = container_of(work, struct rxrpc_local, rejecter); + rxrpc_get_local(local); + + _enter("%d", local->debug_id); + + iov[0].iov_base = &whdr; + iov[0].iov_len = sizeof(whdr); + iov[1].iov_base = &code; + iov[1].iov_len = sizeof(code); + size = sizeof(whdr) + sizeof(code); + + msg.msg_name = &sa; + msg.msg_control = NULL; + msg.msg_controllen = 0; + msg.msg_flags = 0; + + memset(&sa, 0, sizeof(sa)); + sa.sa.sa_family = local->srx.transport.family; + switch (sa.sa.sa_family) { + case AF_INET: + msg.msg_namelen = sizeof(sa.sin); + break; + default: + msg.msg_namelen = 0; + break; + } + + memset(&whdr, 0, sizeof(whdr)); + whdr.type = RXRPC_PACKET_TYPE_ABORT; + + while ((skb = skb_dequeue(&local->reject_queue))) { + sp = rxrpc_skb(skb); + switch (sa.sa.sa_family) { + case AF_INET: + sa.sin.sin_port = udp_hdr(skb)->source; + sa.sin.sin_addr.s_addr = ip_hdr(skb)->saddr; + code = htonl(skb->priority); + + whdr.epoch = htonl(sp->hdr.epoch); + whdr.cid = htonl(sp->hdr.cid); + whdr.callNumber = htonl(sp->hdr.callNumber); + whdr.serviceId = htons(sp->hdr.serviceId); + whdr.flags = sp->hdr.flags; + whdr.flags ^= RXRPC_CLIENT_INITIATED; + whdr.flags &= RXRPC_CLIENT_INITIATED; + + kernel_sendmsg(local->socket, &msg, iov, 2, size); + break; + + default: + break; + } + + rxrpc_free_skb(skb); + rxrpc_put_local(local); + } + + rxrpc_put_local(local); + _leave(""); +} diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c new file mode 100644 index 0000000..8ecde4b --- /dev/null +++ b/net/rxrpc/conn_object.c @@ -0,0 +1,912 @@ +/* RxRPC virtual connection handler + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * Time till a connection expires after last use (in seconds). + */ +unsigned int rxrpc_connection_expiry = 10 * 60; + +static void rxrpc_connection_reaper(struct work_struct *work); + +LIST_HEAD(rxrpc_connections); +DEFINE_RWLOCK(rxrpc_connection_lock); +static DECLARE_DELAYED_WORK(rxrpc_connection_reap, rxrpc_connection_reaper); + +/* + * allocate a new client connection bundle + */ +static struct rxrpc_conn_bundle *rxrpc_alloc_bundle(gfp_t gfp) +{ + struct rxrpc_conn_bundle *bundle; + + _enter(""); + + bundle = kzalloc(sizeof(struct rxrpc_conn_bundle), gfp); + if (bundle) { + INIT_LIST_HEAD(&bundle->unused_conns); + INIT_LIST_HEAD(&bundle->avail_conns); + INIT_LIST_HEAD(&bundle->busy_conns); + init_waitqueue_head(&bundle->chanwait); + atomic_set(&bundle->usage, 1); + } + + _leave(" = %p", bundle); + return bundle; +} + +/* + * compare bundle parameters with what we're looking for + * - return -ve, 0 or +ve + */ +static inline +int rxrpc_cmp_bundle(const struct rxrpc_conn_bundle *bundle, + struct key *key, u16 service_id) +{ + return (bundle->service_id - service_id) ?: + ((unsigned long)bundle->key - (unsigned long)key); +} + +/* + * get bundle of client connections that a client socket can make use of + */ +struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *rx, + struct rxrpc_transport *trans, + struct key *key, + u16 service_id, + gfp_t gfp) +{ + struct rxrpc_conn_bundle *bundle, *candidate; + struct rb_node *p, *parent, **pp; + + _enter("%p{%x},%x,%hx,", + rx, key_serial(key), trans->debug_id, service_id); + + /* search the extant bundles first for one that matches the specified + * user ID */ + spin_lock(&trans->client_lock); + + p = trans->bundles.rb_node; + while (p) { + bundle = rb_entry(p, struct rxrpc_conn_bundle, node); + + if (rxrpc_cmp_bundle(bundle, key, service_id) < 0) + p = p->rb_left; + else if (rxrpc_cmp_bundle(bundle, key, service_id) > 0) + p = p->rb_right; + else + goto found_extant_bundle; + } + + spin_unlock(&trans->client_lock); + + /* not yet present - create a candidate for a new record and then + * redo the search */ + candidate = rxrpc_alloc_bundle(gfp); + if (!candidate) { + _leave(" = -ENOMEM"); + return ERR_PTR(-ENOMEM); + } + + candidate->key = key_get(key); + candidate->service_id = service_id; + + spin_lock(&trans->client_lock); + + pp = &trans->bundles.rb_node; + parent = NULL; + while (*pp) { + parent = *pp; + bundle = rb_entry(parent, struct rxrpc_conn_bundle, node); + + if (rxrpc_cmp_bundle(bundle, key, service_id) < 0) + pp = &(*pp)->rb_left; + else if (rxrpc_cmp_bundle(bundle, key, service_id) > 0) + pp = &(*pp)->rb_right; + else + goto found_extant_second; + } + + /* second search also failed; add the new bundle */ + bundle = candidate; + candidate = NULL; + + rb_link_node(&bundle->node, parent, pp); + rb_insert_color(&bundle->node, &trans->bundles); + spin_unlock(&trans->client_lock); + _net("BUNDLE new on trans %d", trans->debug_id); + _leave(" = %p [new]", bundle); + return bundle; + + /* we found the bundle in the list immediately */ +found_extant_bundle: + atomic_inc(&bundle->usage); + spin_unlock(&trans->client_lock); + _net("BUNDLE old on trans %d", trans->debug_id); + _leave(" = %p [extant %d]", bundle, atomic_read(&bundle->usage)); + return bundle; + + /* we found the bundle on the second time through the list */ +found_extant_second: + atomic_inc(&bundle->usage); + spin_unlock(&trans->client_lock); + kfree(candidate); + _net("BUNDLE old2 on trans %d", trans->debug_id); + _leave(" = %p [second %d]", bundle, atomic_read(&bundle->usage)); + return bundle; +} + +/* + * release a bundle + */ +void rxrpc_put_bundle(struct rxrpc_transport *trans, + struct rxrpc_conn_bundle *bundle) +{ + _enter("%p,%p{%d}",trans, bundle, atomic_read(&bundle->usage)); + + if (atomic_dec_and_lock(&bundle->usage, &trans->client_lock)) { + _debug("Destroy bundle"); + rb_erase(&bundle->node, &trans->bundles); + spin_unlock(&trans->client_lock); + ASSERT(list_empty(&bundle->unused_conns)); + ASSERT(list_empty(&bundle->avail_conns)); + ASSERT(list_empty(&bundle->busy_conns)); + ASSERTCMP(bundle->num_conns, ==, 0); + key_put(bundle->key); + kfree(bundle); + } + + _leave(""); +} + +/* + * allocate a new connection + */ +static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) +{ + struct rxrpc_connection *conn; + + _enter(""); + + conn = kzalloc(sizeof(struct rxrpc_connection), gfp); + if (conn) { + INIT_WORK(&conn->processor, &rxrpc_process_connection); + INIT_LIST_HEAD(&conn->bundle_link); + conn->calls = RB_ROOT; + skb_queue_head_init(&conn->rx_queue); + conn->security = &rxrpc_no_security; + rwlock_init(&conn->lock); + spin_lock_init(&conn->state_lock); + atomic_set(&conn->usage, 1); + conn->debug_id = atomic_inc_return(&rxrpc_debug_id); + conn->avail_calls = RXRPC_MAXCALLS; + conn->size_align = 4; + conn->header_size = sizeof(struct rxrpc_wire_header); + } + + _leave(" = %p{%d}", conn, conn ? conn->debug_id : 0); + return conn; +} + +/* + * assign a connection ID to a connection and add it to the transport's + * connection lookup tree + * - called with transport client lock held + */ +static void rxrpc_assign_connection_id(struct rxrpc_connection *conn) +{ + struct rxrpc_connection *xconn; + struct rb_node *parent, **p; + __be32 epoch; + u32 cid; + + _enter(""); + + epoch = conn->epoch; + + write_lock_bh(&conn->trans->conn_lock); + + conn->trans->conn_idcounter += RXRPC_CID_INC; + if (conn->trans->conn_idcounter < RXRPC_CID_INC) + conn->trans->conn_idcounter = RXRPC_CID_INC; + cid = conn->trans->conn_idcounter; + +attempt_insertion: + parent = NULL; + p = &conn->trans->client_conns.rb_node; + + while (*p) { + parent = *p; + xconn = rb_entry(parent, struct rxrpc_connection, node); + + if (epoch < xconn->epoch) + p = &(*p)->rb_left; + else if (epoch > xconn->epoch) + p = &(*p)->rb_right; + else if (cid < xconn->cid) + p = &(*p)->rb_left; + else if (cid > xconn->cid) + p = &(*p)->rb_right; + else + goto id_exists; + } + + /* we've found a suitable hole - arrange for this connection to occupy + * it */ + rb_link_node(&conn->node, parent, p); + rb_insert_color(&conn->node, &conn->trans->client_conns); + + conn->cid = cid; + write_unlock_bh(&conn->trans->conn_lock); + _leave(" [CID %x]", cid); + return; + + /* we found a connection with the proposed ID - walk the tree from that + * point looking for the next unused ID */ +id_exists: + for (;;) { + cid += RXRPC_CID_INC; + if (cid < RXRPC_CID_INC) { + cid = RXRPC_CID_INC; + conn->trans->conn_idcounter = cid; + goto attempt_insertion; + } + + parent = rb_next(parent); + if (!parent) + goto attempt_insertion; + + xconn = rb_entry(parent, struct rxrpc_connection, node); + if (epoch < xconn->epoch || + cid < xconn->cid) + goto attempt_insertion; + } +} + +/* + * add a call to a connection's call-by-ID tree + */ +static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn, + struct rxrpc_call *call) +{ + struct rxrpc_call *xcall; + struct rb_node *parent, **p; + __be32 call_id; + + write_lock_bh(&conn->lock); + + call_id = call->call_id; + p = &conn->calls.rb_node; + parent = NULL; + while (*p) { + parent = *p; + xcall = rb_entry(parent, struct rxrpc_call, conn_node); + + if (call_id < xcall->call_id) + p = &(*p)->rb_left; + else if (call_id > xcall->call_id) + p = &(*p)->rb_right; + else + BUG(); + } + + rb_link_node(&call->conn_node, parent, p); + rb_insert_color(&call->conn_node, &conn->calls); + + write_unlock_bh(&conn->lock); +} + +/* + * connect a call on an exclusive connection + */ +static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, + struct rxrpc_transport *trans, + u16 service_id, + struct rxrpc_call *call, + gfp_t gfp) +{ + struct rxrpc_connection *conn; + int chan, ret; + + _enter(""); + + conn = rx->conn; + if (!conn) { + /* not yet present - create a candidate for a new connection + * and then redo the check */ + conn = rxrpc_alloc_connection(gfp); + if (!conn) { + _leave(" = -ENOMEM"); + return -ENOMEM; + } + + conn->trans = trans; + conn->bundle = NULL; + conn->service_id = service_id; + conn->epoch = rxrpc_epoch; + conn->in_clientflag = 0; + conn->out_clientflag = RXRPC_CLIENT_INITIATED; + conn->cid = 0; + conn->state = RXRPC_CONN_CLIENT; + conn->avail_calls = RXRPC_MAXCALLS - 1; + conn->security_level = rx->min_sec_level; + conn->key = key_get(rx->key); + + ret = rxrpc_init_client_conn_security(conn); + if (ret < 0) { + key_put(conn->key); + kfree(conn); + _leave(" = %d [key]", ret); + return ret; + } + + write_lock_bh(&rxrpc_connection_lock); + list_add_tail(&conn->link, &rxrpc_connections); + write_unlock_bh(&rxrpc_connection_lock); + + spin_lock(&trans->client_lock); + atomic_inc(&trans->usage); + + _net("CONNECT EXCL new %d on TRANS %d", + conn->debug_id, conn->trans->debug_id); + + rxrpc_assign_connection_id(conn); + rx->conn = conn; + } else { + spin_lock(&trans->client_lock); + } + + /* we've got a connection with a free channel and we can now attach the + * call to it + * - we're holding the transport's client lock + * - we're holding a reference on the connection + */ + for (chan = 0; chan < RXRPC_MAXCALLS; chan++) + if (!conn->channels[chan]) + goto found_channel; + goto no_free_channels; + +found_channel: + atomic_inc(&conn->usage); + conn->channels[chan] = call; + call->conn = conn; + call->channel = chan; + call->cid = conn->cid | chan; + call->call_id = ++conn->call_counter; + + _net("CONNECT client on conn %d chan %d as call %x", + conn->debug_id, chan, call->call_id); + + spin_unlock(&trans->client_lock); + + rxrpc_add_call_ID_to_conn(conn, call); + _leave(" = 0"); + return 0; + +no_free_channels: + spin_unlock(&trans->client_lock); + _leave(" = -ENOSR"); + return -ENOSR; +} + +/* + * find a connection for a call + * - called in process context with IRQs enabled + */ +int rxrpc_connect_call(struct rxrpc_sock *rx, + struct rxrpc_transport *trans, + struct rxrpc_conn_bundle *bundle, + struct rxrpc_call *call, + gfp_t gfp) +{ + struct rxrpc_connection *conn, *candidate; + int chan, ret; + + DECLARE_WAITQUEUE(myself, current); + + _enter("%p,%lx,", rx, call->user_call_ID); + + if (test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags)) + return rxrpc_connect_exclusive(rx, trans, bundle->service_id, + call, gfp); + + spin_lock(&trans->client_lock); + for (;;) { + /* see if the bundle has a call slot available */ + if (!list_empty(&bundle->avail_conns)) { + _debug("avail"); + conn = list_entry(bundle->avail_conns.next, + struct rxrpc_connection, + bundle_link); + if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) { + list_del_init(&conn->bundle_link); + bundle->num_conns--; + continue; + } + if (--conn->avail_calls == 0) + list_move(&conn->bundle_link, + &bundle->busy_conns); + ASSERTCMP(conn->avail_calls, <, RXRPC_MAXCALLS); + ASSERT(conn->channels[0] == NULL || + conn->channels[1] == NULL || + conn->channels[2] == NULL || + conn->channels[3] == NULL); + atomic_inc(&conn->usage); + break; + } + + if (!list_empty(&bundle->unused_conns)) { + _debug("unused"); + conn = list_entry(bundle->unused_conns.next, + struct rxrpc_connection, + bundle_link); + if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) { + list_del_init(&conn->bundle_link); + bundle->num_conns--; + continue; + } + ASSERTCMP(conn->avail_calls, ==, RXRPC_MAXCALLS); + conn->avail_calls = RXRPC_MAXCALLS - 1; + ASSERT(conn->channels[0] == NULL && + conn->channels[1] == NULL && + conn->channels[2] == NULL && + conn->channels[3] == NULL); + atomic_inc(&conn->usage); + list_move(&conn->bundle_link, &bundle->avail_conns); + break; + } + + /* need to allocate a new connection */ + _debug("get new conn [%d]", bundle->num_conns); + + spin_unlock(&trans->client_lock); + + if (signal_pending(current)) + goto interrupted; + + if (bundle->num_conns >= 20) { + _debug("too many conns"); + + if (!gfpflags_allow_blocking(gfp)) { + _leave(" = -EAGAIN"); + return -EAGAIN; + } + + add_wait_queue(&bundle->chanwait, &myself); + for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + if (bundle->num_conns < 20 || + !list_empty(&bundle->unused_conns) || + !list_empty(&bundle->avail_conns)) + break; + if (signal_pending(current)) + goto interrupted_dequeue; + schedule(); + } + remove_wait_queue(&bundle->chanwait, &myself); + __set_current_state(TASK_RUNNING); + spin_lock(&trans->client_lock); + continue; + } + + /* not yet present - create a candidate for a new connection and then + * redo the check */ + candidate = rxrpc_alloc_connection(gfp); + if (!candidate) { + _leave(" = -ENOMEM"); + return -ENOMEM; + } + + candidate->trans = trans; + candidate->bundle = bundle; + candidate->service_id = bundle->service_id; + candidate->epoch = rxrpc_epoch; + candidate->in_clientflag = 0; + candidate->out_clientflag = RXRPC_CLIENT_INITIATED; + candidate->cid = 0; + candidate->state = RXRPC_CONN_CLIENT; + candidate->avail_calls = RXRPC_MAXCALLS; + candidate->security_level = rx->min_sec_level; + candidate->key = key_get(bundle->key); + + ret = rxrpc_init_client_conn_security(candidate); + if (ret < 0) { + key_put(candidate->key); + kfree(candidate); + _leave(" = %d [key]", ret); + return ret; + } + + write_lock_bh(&rxrpc_connection_lock); + list_add_tail(&candidate->link, &rxrpc_connections); + write_unlock_bh(&rxrpc_connection_lock); + + spin_lock(&trans->client_lock); + + list_add(&candidate->bundle_link, &bundle->unused_conns); + bundle->num_conns++; + atomic_inc(&bundle->usage); + atomic_inc(&trans->usage); + + _net("CONNECT new %d on TRANS %d", + candidate->debug_id, candidate->trans->debug_id); + + rxrpc_assign_connection_id(candidate); + candidate->security->prime_packet_security(candidate); + + /* leave the candidate lurking in zombie mode attached to the + * bundle until we're ready for it */ + rxrpc_put_connection(candidate); + candidate = NULL; + } + + /* we've got a connection with a free channel and we can now attach the + * call to it + * - we're holding the transport's client lock + * - we're holding a reference on the connection + * - we're holding a reference on the bundle + */ + for (chan = 0; chan < RXRPC_MAXCALLS; chan++) + if (!conn->channels[chan]) + goto found_channel; + ASSERT(conn->channels[0] == NULL || + conn->channels[1] == NULL || + conn->channels[2] == NULL || + conn->channels[3] == NULL); + BUG(); + +found_channel: + conn->channels[chan] = call; + call->conn = conn; + call->channel = chan; + call->cid = conn->cid | chan; + call->call_id = ++conn->call_counter; + + _net("CONNECT client on conn %d chan %d as call %x", + conn->debug_id, chan, call->call_id); + + ASSERTCMP(conn->avail_calls, <, RXRPC_MAXCALLS); + spin_unlock(&trans->client_lock); + + rxrpc_add_call_ID_to_conn(conn, call); + + _leave(" = 0"); + return 0; + +interrupted_dequeue: + remove_wait_queue(&bundle->chanwait, &myself); + __set_current_state(TASK_RUNNING); +interrupted: + _leave(" = -ERESTARTSYS"); + return -ERESTARTSYS; +} + +/* + * get a record of an incoming connection + */ +struct rxrpc_connection * +rxrpc_incoming_connection(struct rxrpc_transport *trans, + struct rxrpc_host_header *hdr) +{ + struct rxrpc_connection *conn, *candidate = NULL; + struct rb_node *p, **pp; + const char *new = "old"; + __be32 epoch; + u32 cid; + + _enter(""); + + ASSERT(hdr->flags & RXRPC_CLIENT_INITIATED); + + epoch = hdr->epoch; + cid = hdr->cid & RXRPC_CIDMASK; + + /* search the connection list first */ + read_lock_bh(&trans->conn_lock); + + p = trans->server_conns.rb_node; + while (p) { + conn = rb_entry(p, struct rxrpc_connection, node); + + _debug("maybe %x", conn->cid); + + if (epoch < conn->epoch) + p = p->rb_left; + else if (epoch > conn->epoch) + p = p->rb_right; + else if (cid < conn->cid) + p = p->rb_left; + else if (cid > conn->cid) + p = p->rb_right; + else + goto found_extant_connection; + } + read_unlock_bh(&trans->conn_lock); + + /* not yet present - create a candidate for a new record and then + * redo the search */ + candidate = rxrpc_alloc_connection(GFP_NOIO); + if (!candidate) { + _leave(" = -ENOMEM"); + return ERR_PTR(-ENOMEM); + } + + candidate->trans = trans; + candidate->epoch = hdr->epoch; + candidate->cid = hdr->cid & RXRPC_CIDMASK; + candidate->service_id = hdr->serviceId; + candidate->security_ix = hdr->securityIndex; + candidate->in_clientflag = RXRPC_CLIENT_INITIATED; + candidate->out_clientflag = 0; + candidate->state = RXRPC_CONN_SERVER; + if (candidate->service_id) + candidate->state = RXRPC_CONN_SERVER_UNSECURED; + + write_lock_bh(&trans->conn_lock); + + pp = &trans->server_conns.rb_node; + p = NULL; + while (*pp) { + p = *pp; + conn = rb_entry(p, struct rxrpc_connection, node); + + if (epoch < conn->epoch) + pp = &(*pp)->rb_left; + else if (epoch > conn->epoch) + pp = &(*pp)->rb_right; + else if (cid < conn->cid) + pp = &(*pp)->rb_left; + else if (cid > conn->cid) + pp = &(*pp)->rb_right; + else + goto found_extant_second; + } + + /* we can now add the new candidate to the list */ + conn = candidate; + candidate = NULL; + rb_link_node(&conn->node, p, pp); + rb_insert_color(&conn->node, &trans->server_conns); + atomic_inc(&conn->trans->usage); + + write_unlock_bh(&trans->conn_lock); + + write_lock_bh(&rxrpc_connection_lock); + list_add_tail(&conn->link, &rxrpc_connections); + write_unlock_bh(&rxrpc_connection_lock); + + new = "new"; + +success: + _net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->cid); + + _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); + return conn; + + /* we found the connection in the list immediately */ +found_extant_connection: + if (hdr->securityIndex != conn->security_ix) { + read_unlock_bh(&trans->conn_lock); + goto security_mismatch; + } + atomic_inc(&conn->usage); + read_unlock_bh(&trans->conn_lock); + goto success; + + /* we found the connection on the second time through the list */ +found_extant_second: + if (hdr->securityIndex != conn->security_ix) { + write_unlock_bh(&trans->conn_lock); + goto security_mismatch; + } + atomic_inc(&conn->usage); + write_unlock_bh(&trans->conn_lock); + kfree(candidate); + goto success; + +security_mismatch: + kfree(candidate); + _leave(" = -EKEYREJECTED"); + return ERR_PTR(-EKEYREJECTED); +} + +/* + * find a connection based on transport and RxRPC connection ID for an incoming + * packet + */ +struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans, + struct rxrpc_host_header *hdr) +{ + struct rxrpc_connection *conn; + struct rb_node *p; + u32 epoch, cid; + + _enter(",{%x,%x}", hdr->cid, hdr->flags); + + read_lock_bh(&trans->conn_lock); + + cid = hdr->cid & RXRPC_CIDMASK; + epoch = hdr->epoch; + + if (hdr->flags & RXRPC_CLIENT_INITIATED) + p = trans->server_conns.rb_node; + else + p = trans->client_conns.rb_node; + + while (p) { + conn = rb_entry(p, struct rxrpc_connection, node); + + _debug("maybe %x", conn->cid); + + if (epoch < conn->epoch) + p = p->rb_left; + else if (epoch > conn->epoch) + p = p->rb_right; + else if (cid < conn->cid) + p = p->rb_left; + else if (cid > conn->cid) + p = p->rb_right; + else + goto found; + } + + read_unlock_bh(&trans->conn_lock); + _leave(" = NULL"); + return NULL; + +found: + atomic_inc(&conn->usage); + read_unlock_bh(&trans->conn_lock); + _leave(" = %p", conn); + return conn; +} + +/* + * release a virtual connection + */ +void rxrpc_put_connection(struct rxrpc_connection *conn) +{ + _enter("%p{u=%d,d=%d}", + conn, atomic_read(&conn->usage), conn->debug_id); + + ASSERTCMP(atomic_read(&conn->usage), >, 0); + + conn->put_time = ktime_get_seconds(); + if (atomic_dec_and_test(&conn->usage)) { + _debug("zombie"); + rxrpc_queue_delayed_work(&rxrpc_connection_reap, 0); + } + + _leave(""); +} + +/* + * destroy a virtual connection + */ +static void rxrpc_destroy_connection(struct rxrpc_connection *conn) +{ + _enter("%p{%d}", conn, atomic_read(&conn->usage)); + + ASSERTCMP(atomic_read(&conn->usage), ==, 0); + + _net("DESTROY CONN %d", conn->debug_id); + + if (conn->bundle) + rxrpc_put_bundle(conn->trans, conn->bundle); + + ASSERT(RB_EMPTY_ROOT(&conn->calls)); + rxrpc_purge_queue(&conn->rx_queue); + + conn->security->clear(conn); + key_put(conn->key); + key_put(conn->server_key); + + rxrpc_put_transport(conn->trans); + kfree(conn); + _leave(""); +} + +/* + * reap dead connections + */ +static void rxrpc_connection_reaper(struct work_struct *work) +{ + struct rxrpc_connection *conn, *_p; + unsigned long now, earliest, reap_time; + + LIST_HEAD(graveyard); + + _enter(""); + + now = ktime_get_seconds(); + earliest = ULONG_MAX; + + write_lock_bh(&rxrpc_connection_lock); + list_for_each_entry_safe(conn, _p, &rxrpc_connections, link) { + _debug("reap CONN %d { u=%d,t=%ld }", + conn->debug_id, atomic_read(&conn->usage), + (long) now - (long) conn->put_time); + + if (likely(atomic_read(&conn->usage) > 0)) + continue; + + spin_lock(&conn->trans->client_lock); + write_lock(&conn->trans->conn_lock); + reap_time = conn->put_time + rxrpc_connection_expiry; + + if (atomic_read(&conn->usage) > 0) { + ; + } else if (reap_time <= now) { + list_move_tail(&conn->link, &graveyard); + if (conn->out_clientflag) + rb_erase(&conn->node, + &conn->trans->client_conns); + else + rb_erase(&conn->node, + &conn->trans->server_conns); + if (conn->bundle) { + list_del_init(&conn->bundle_link); + conn->bundle->num_conns--; + } + + } else if (reap_time < earliest) { + earliest = reap_time; + } + + write_unlock(&conn->trans->conn_lock); + spin_unlock(&conn->trans->client_lock); + } + write_unlock_bh(&rxrpc_connection_lock); + + if (earliest != ULONG_MAX) { + _debug("reschedule reaper %ld", (long) earliest - now); + ASSERTCMP(earliest, >, now); + rxrpc_queue_delayed_work(&rxrpc_connection_reap, + (earliest - now) * HZ); + } + + /* then destroy all those pulled out */ + while (!list_empty(&graveyard)) { + conn = list_entry(graveyard.next, struct rxrpc_connection, + link); + list_del_init(&conn->link); + + ASSERTCMP(atomic_read(&conn->usage), ==, 0); + rxrpc_destroy_connection(conn); + } + + _leave(""); +} + +/* + * preemptively destroy all the connection records rather than waiting for them + * to time out + */ +void __exit rxrpc_destroy_all_connections(void) +{ + _enter(""); + + rxrpc_connection_expiry = 0; + cancel_delayed_work(&rxrpc_connection_reap); + rxrpc_queue_delayed_work(&rxrpc_connection_reap, 0); + + _leave(""); +} diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c new file mode 100644 index 0000000..e0815a0 --- /dev/null +++ b/net/rxrpc/input.c @@ -0,0 +1,800 @@ +/* RxRPC packet reception + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * queue a packet for recvmsg to pass to userspace + * - the caller must hold a lock on call->lock + * - must not be called with interrupts disabled (sk_filter() disables BH's) + * - eats the packet whether successful or not + * - there must be just one reference to the packet, which the caller passes to + * this function + */ +int rxrpc_queue_rcv_skb(struct rxrpc_call *call, struct sk_buff *skb, + bool force, bool terminal) +{ + struct rxrpc_skb_priv *sp; + struct rxrpc_sock *rx = call->socket; + struct sock *sk; + int ret; + + _enter(",,%d,%d", force, terminal); + + ASSERT(!irqs_disabled()); + + sp = rxrpc_skb(skb); + ASSERTCMP(sp->call, ==, call); + + /* if we've already posted the terminal message for a call, then we + * don't post any more */ + if (test_bit(RXRPC_CALL_TERMINAL_MSG, &call->flags)) { + _debug("already terminated"); + ASSERTCMP(call->state, >=, RXRPC_CALL_COMPLETE); + skb->destructor = NULL; + sp->call = NULL; + rxrpc_put_call(call); + rxrpc_free_skb(skb); + return 0; + } + + sk = &rx->sk; + + if (!force) { + /* cast skb->rcvbuf to unsigned... It's pointless, but + * reduces number of warnings when compiling with -W + * --ANK */ +// ret = -ENOBUFS; +// if (atomic_read(&sk->sk_rmem_alloc) + skb->truesize >= +// (unsigned int) sk->sk_rcvbuf) +// goto out; + + ret = sk_filter(sk, skb); + if (ret < 0) + goto out; + } + + spin_lock_bh(&sk->sk_receive_queue.lock); + if (!test_bit(RXRPC_CALL_TERMINAL_MSG, &call->flags) && + !test_bit(RXRPC_CALL_RELEASED, &call->flags) && + call->socket->sk.sk_state != RXRPC_CLOSE) { + skb->destructor = rxrpc_packet_destructor; + skb->dev = NULL; + skb->sk = sk; + atomic_add(skb->truesize, &sk->sk_rmem_alloc); + + if (terminal) { + _debug("<<<< TERMINAL MESSAGE >>>>"); + set_bit(RXRPC_CALL_TERMINAL_MSG, &call->flags); + } + + /* allow interception by a kernel service */ + if (rx->interceptor) { + rx->interceptor(sk, call->user_call_ID, skb); + spin_unlock_bh(&sk->sk_receive_queue.lock); + } else { + _net("post skb %p", skb); + __skb_queue_tail(&sk->sk_receive_queue, skb); + spin_unlock_bh(&sk->sk_receive_queue.lock); + + if (!sock_flag(sk, SOCK_DEAD)) + sk->sk_data_ready(sk); + } + skb = NULL; + } else { + spin_unlock_bh(&sk->sk_receive_queue.lock); + } + ret = 0; + +out: + /* release the socket buffer */ + if (skb) { + skb->destructor = NULL; + sp->call = NULL; + rxrpc_put_call(call); + rxrpc_free_skb(skb); + } + + _leave(" = %d", ret); + return ret; +} + +/* + * process a DATA packet, posting the packet to the appropriate queue + * - eats the packet if successful + */ +static int rxrpc_fast_process_data(struct rxrpc_call *call, + struct sk_buff *skb, u32 seq) +{ + struct rxrpc_skb_priv *sp; + bool terminal; + int ret, ackbit, ack; + + _enter("{%u,%u},,{%u}", call->rx_data_post, call->rx_first_oos, seq); + + sp = rxrpc_skb(skb); + ASSERTCMP(sp->call, ==, NULL); + + spin_lock(&call->lock); + + if (call->state > RXRPC_CALL_COMPLETE) + goto discard; + + ASSERTCMP(call->rx_data_expect, >=, call->rx_data_post); + ASSERTCMP(call->rx_data_post, >=, call->rx_data_recv); + ASSERTCMP(call->rx_data_recv, >=, call->rx_data_eaten); + + if (seq < call->rx_data_post) { + _debug("dup #%u [-%u]", seq, call->rx_data_post); + ack = RXRPC_ACK_DUPLICATE; + ret = -ENOBUFS; + goto discard_and_ack; + } + + /* we may already have the packet in the out of sequence queue */ + ackbit = seq - (call->rx_data_eaten + 1); + ASSERTCMP(ackbit, >=, 0); + if (__test_and_set_bit(ackbit, call->ackr_window)) { + _debug("dup oos #%u [%u,%u]", + seq, call->rx_data_eaten, call->rx_data_post); + ack = RXRPC_ACK_DUPLICATE; + goto discard_and_ack; + } + + if (seq >= call->ackr_win_top) { + _debug("exceed #%u [%u]", seq, call->ackr_win_top); + __clear_bit(ackbit, call->ackr_window); + ack = RXRPC_ACK_EXCEEDS_WINDOW; + goto discard_and_ack; + } + + if (seq == call->rx_data_expect) { + clear_bit(RXRPC_CALL_EXPECT_OOS, &call->flags); + call->rx_data_expect++; + } else if (seq > call->rx_data_expect) { + _debug("oos #%u [%u]", seq, call->rx_data_expect); + call->rx_data_expect = seq + 1; + if (test_and_set_bit(RXRPC_CALL_EXPECT_OOS, &call->flags)) { + ack = RXRPC_ACK_OUT_OF_SEQUENCE; + goto enqueue_and_ack; + } + goto enqueue_packet; + } + + if (seq != call->rx_data_post) { + _debug("ahead #%u [%u]", seq, call->rx_data_post); + goto enqueue_packet; + } + + if (test_bit(RXRPC_CALL_RCVD_LAST, &call->flags)) + goto protocol_error; + + /* if the packet need security things doing to it, then it goes down + * the slow path */ + if (call->conn->security_ix) + goto enqueue_packet; + + sp->call = call; + rxrpc_get_call(call); + terminal = ((sp->hdr.flags & RXRPC_LAST_PACKET) && + !(sp->hdr.flags & RXRPC_CLIENT_INITIATED)); + ret = rxrpc_queue_rcv_skb(call, skb, false, terminal); + if (ret < 0) { + if (ret == -ENOMEM || ret == -ENOBUFS) { + __clear_bit(ackbit, call->ackr_window); + ack = RXRPC_ACK_NOSPACE; + goto discard_and_ack; + } + goto out; + } + + skb = NULL; + + _debug("post #%u", seq); + ASSERTCMP(call->rx_data_post, ==, seq); + call->rx_data_post++; + + if (sp->hdr.flags & RXRPC_LAST_PACKET) + set_bit(RXRPC_CALL_RCVD_LAST, &call->flags); + + /* if we've reached an out of sequence packet then we need to drain + * that queue into the socket Rx queue now */ + if (call->rx_data_post == call->rx_first_oos) { + _debug("drain rx oos now"); + read_lock(&call->state_lock); + if (call->state < RXRPC_CALL_COMPLETE && + !test_and_set_bit(RXRPC_CALL_EV_DRAIN_RX_OOS, &call->events)) + rxrpc_queue_call(call); + read_unlock(&call->state_lock); + } + + spin_unlock(&call->lock); + atomic_inc(&call->ackr_not_idle); + rxrpc_propose_ACK(call, RXRPC_ACK_DELAY, sp->hdr.serial, false); + _leave(" = 0 [posted]"); + return 0; + +protocol_error: + ret = -EBADMSG; +out: + spin_unlock(&call->lock); + _leave(" = %d", ret); + return ret; + +discard_and_ack: + _debug("discard and ACK packet %p", skb); + __rxrpc_propose_ACK(call, ack, sp->hdr.serial, true); +discard: + spin_unlock(&call->lock); + rxrpc_free_skb(skb); + _leave(" = 0 [discarded]"); + return 0; + +enqueue_and_ack: + __rxrpc_propose_ACK(call, ack, sp->hdr.serial, true); +enqueue_packet: + _net("defer skb %p", skb); + spin_unlock(&call->lock); + skb_queue_tail(&call->rx_queue, skb); + atomic_inc(&call->ackr_not_idle); + read_lock(&call->state_lock); + if (call->state < RXRPC_CALL_DEAD) + rxrpc_queue_call(call); + read_unlock(&call->state_lock); + _leave(" = 0 [queued]"); + return 0; +} + +/* + * assume an implicit ACKALL of the transmission phase of a client socket upon + * reception of the first reply packet + */ +static void rxrpc_assume_implicit_ackall(struct rxrpc_call *call, u32 serial) +{ + write_lock_bh(&call->state_lock); + + switch (call->state) { + case RXRPC_CALL_CLIENT_AWAIT_REPLY: + call->state = RXRPC_CALL_CLIENT_RECV_REPLY; + call->acks_latest = serial; + + _debug("implicit ACKALL %%%u", call->acks_latest); + set_bit(RXRPC_CALL_EV_RCVD_ACKALL, &call->events); + write_unlock_bh(&call->state_lock); + + if (try_to_del_timer_sync(&call->resend_timer) >= 0) { + clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events); + clear_bit(RXRPC_CALL_EV_RESEND, &call->events); + clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); + } + break; + + default: + write_unlock_bh(&call->state_lock); + break; + } +} + +/* + * post an incoming packet to the nominated call to deal with + * - must get rid of the sk_buff, either by freeing it or by queuing it + */ +void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb) +{ + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + __be32 wtmp; + u32 hi_serial, abort_code; + + _enter("%p,%p", call, skb); + + ASSERT(!irqs_disabled()); + +#if 0 // INJECT RX ERROR + if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA) { + static int skip = 0; + if (++skip == 3) { + printk("DROPPED 3RD PACKET!!!!!!!!!!!!!\n"); + skip = 0; + goto free_packet; + } + } +#endif + + /* track the latest serial number on this connection for ACK packet + * information */ + hi_serial = atomic_read(&call->conn->hi_serial); + while (sp->hdr.serial > hi_serial) + hi_serial = atomic_cmpxchg(&call->conn->hi_serial, hi_serial, + sp->hdr.serial); + + /* request ACK generation for any ACK or DATA packet that requests + * it */ + if (sp->hdr.flags & RXRPC_REQUEST_ACK) { + _proto("ACK Requested on %%%u", sp->hdr.serial); + rxrpc_propose_ACK(call, RXRPC_ACK_REQUESTED, sp->hdr.serial, false); + } + + switch (sp->hdr.type) { + case RXRPC_PACKET_TYPE_ABORT: + _debug("abort"); + + if (skb_copy_bits(skb, 0, &wtmp, sizeof(wtmp)) < 0) + goto protocol_error; + + abort_code = ntohl(wtmp); + _proto("Rx ABORT %%%u { %x }", sp->hdr.serial, abort_code); + + write_lock_bh(&call->state_lock); + if (call->state < RXRPC_CALL_COMPLETE) { + call->state = RXRPC_CALL_REMOTELY_ABORTED; + call->remote_abort = abort_code; + set_bit(RXRPC_CALL_EV_RCVD_ABORT, &call->events); + rxrpc_queue_call(call); + } + goto free_packet_unlock; + + case RXRPC_PACKET_TYPE_BUSY: + _proto("Rx BUSY %%%u", sp->hdr.serial); + + if (call->conn->out_clientflag) + goto protocol_error; + + write_lock_bh(&call->state_lock); + switch (call->state) { + case RXRPC_CALL_CLIENT_SEND_REQUEST: + call->state = RXRPC_CALL_SERVER_BUSY; + set_bit(RXRPC_CALL_EV_RCVD_BUSY, &call->events); + rxrpc_queue_call(call); + case RXRPC_CALL_SERVER_BUSY: + goto free_packet_unlock; + default: + goto protocol_error_locked; + } + + default: + _proto("Rx %s %%%u", rxrpc_pkts[sp->hdr.type], sp->hdr.serial); + goto protocol_error; + + case RXRPC_PACKET_TYPE_DATA: + _proto("Rx DATA %%%u { #%u }", sp->hdr.serial, sp->hdr.seq); + + if (sp->hdr.seq == 0) + goto protocol_error; + + call->ackr_prev_seq = sp->hdr.seq; + + /* received data implicitly ACKs all of the request packets we + * sent when we're acting as a client */ + if (call->state == RXRPC_CALL_CLIENT_AWAIT_REPLY) + rxrpc_assume_implicit_ackall(call, sp->hdr.serial); + + switch (rxrpc_fast_process_data(call, skb, sp->hdr.seq)) { + case 0: + skb = NULL; + goto done; + + default: + BUG(); + + /* data packet received beyond the last packet */ + case -EBADMSG: + goto protocol_error; + } + + case RXRPC_PACKET_TYPE_ACKALL: + case RXRPC_PACKET_TYPE_ACK: + /* ACK processing is done in process context */ + read_lock_bh(&call->state_lock); + if (call->state < RXRPC_CALL_DEAD) { + skb_queue_tail(&call->rx_queue, skb); + rxrpc_queue_call(call); + skb = NULL; + } + read_unlock_bh(&call->state_lock); + goto free_packet; + } + +protocol_error: + _debug("protocol error"); + write_lock_bh(&call->state_lock); +protocol_error_locked: + if (call->state <= RXRPC_CALL_COMPLETE) { + call->state = RXRPC_CALL_LOCALLY_ABORTED; + call->local_abort = RX_PROTOCOL_ERROR; + set_bit(RXRPC_CALL_EV_ABORT, &call->events); + rxrpc_queue_call(call); + } +free_packet_unlock: + write_unlock_bh(&call->state_lock); +free_packet: + rxrpc_free_skb(skb); +done: + _leave(""); +} + +/* + * split up a jumbo data packet + */ +static void rxrpc_process_jumbo_packet(struct rxrpc_call *call, + struct sk_buff *jumbo) +{ + struct rxrpc_jumbo_header jhdr; + struct rxrpc_skb_priv *sp; + struct sk_buff *part; + + _enter(",{%u,%u}", jumbo->data_len, jumbo->len); + + sp = rxrpc_skb(jumbo); + + do { + sp->hdr.flags &= ~RXRPC_JUMBO_PACKET; + + /* make a clone to represent the first subpacket in what's left + * of the jumbo packet */ + part = skb_clone(jumbo, GFP_ATOMIC); + if (!part) { + /* simply ditch the tail in the event of ENOMEM */ + pskb_trim(jumbo, RXRPC_JUMBO_DATALEN); + break; + } + rxrpc_new_skb(part); + + pskb_trim(part, RXRPC_JUMBO_DATALEN); + + if (!pskb_pull(jumbo, RXRPC_JUMBO_DATALEN)) + goto protocol_error; + + if (skb_copy_bits(jumbo, 0, &jhdr, sizeof(jhdr)) < 0) + goto protocol_error; + if (!pskb_pull(jumbo, sizeof(jhdr))) + BUG(); + + sp->hdr.seq += 1; + sp->hdr.serial += 1; + sp->hdr.flags = jhdr.flags; + sp->hdr._rsvd = jhdr._rsvd; + + _proto("Rx DATA Jumbo %%%u", sp->hdr.serial - 1); + + rxrpc_fast_process_packet(call, part); + part = NULL; + + } while (sp->hdr.flags & RXRPC_JUMBO_PACKET); + + rxrpc_fast_process_packet(call, jumbo); + _leave(""); + return; + +protocol_error: + _debug("protocol error"); + rxrpc_free_skb(part); + rxrpc_free_skb(jumbo); + write_lock_bh(&call->state_lock); + if (call->state <= RXRPC_CALL_COMPLETE) { + call->state = RXRPC_CALL_LOCALLY_ABORTED; + call->local_abort = RX_PROTOCOL_ERROR; + set_bit(RXRPC_CALL_EV_ABORT, &call->events); + rxrpc_queue_call(call); + } + write_unlock_bh(&call->state_lock); + _leave(""); +} + +/* + * post an incoming packet to the appropriate call/socket to deal with + * - must get rid of the sk_buff, either by freeing it or by queuing it + */ +static void rxrpc_post_packet_to_call(struct rxrpc_call *call, + struct sk_buff *skb) +{ + struct rxrpc_skb_priv *sp; + + _enter("%p,%p", call, skb); + + sp = rxrpc_skb(skb); + + _debug("extant call [%d]", call->state); + + read_lock(&call->state_lock); + switch (call->state) { + case RXRPC_CALL_LOCALLY_ABORTED: + if (!test_and_set_bit(RXRPC_CALL_EV_ABORT, &call->events)) { + rxrpc_queue_call(call); + goto free_unlock; + } + case RXRPC_CALL_REMOTELY_ABORTED: + case RXRPC_CALL_NETWORK_ERROR: + case RXRPC_CALL_DEAD: + goto dead_call; + case RXRPC_CALL_COMPLETE: + case RXRPC_CALL_CLIENT_FINAL_ACK: + /* complete server call */ + if (call->conn->in_clientflag) + goto dead_call; + /* resend last packet of a completed call */ + _debug("final ack again"); + rxrpc_get_call(call); + set_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events); + rxrpc_queue_call(call); + goto free_unlock; + default: + break; + } + + read_unlock(&call->state_lock); + rxrpc_get_call(call); + + if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA && + sp->hdr.flags & RXRPC_JUMBO_PACKET) + rxrpc_process_jumbo_packet(call, skb); + else + rxrpc_fast_process_packet(call, skb); + + rxrpc_put_call(call); + goto done; + +dead_call: + if (sp->hdr.type != RXRPC_PACKET_TYPE_ABORT) { + skb->priority = RX_CALL_DEAD; + rxrpc_reject_packet(call->conn->trans->local, skb); + goto unlock; + } +free_unlock: + rxrpc_free_skb(skb); +unlock: + read_unlock(&call->state_lock); +done: + _leave(""); +} + +/* + * post connection-level events to the connection + * - this includes challenges, responses and some aborts + */ +static void rxrpc_post_packet_to_conn(struct rxrpc_connection *conn, + struct sk_buff *skb) +{ + _enter("%p,%p", conn, skb); + + atomic_inc(&conn->usage); + skb_queue_tail(&conn->rx_queue, skb); + rxrpc_queue_conn(conn); +} + +/* + * post endpoint-level events to the local endpoint + * - this includes debug and version messages + */ +static void rxrpc_post_packet_to_local(struct rxrpc_local *local, + struct sk_buff *skb) +{ + _enter("%p,%p", local, skb); + + atomic_inc(&local->usage); + skb_queue_tail(&local->event_queue, skb); + rxrpc_queue_work(&local->event_processor); +} + +/* + * Extract the wire header from a packet and translate the byte order. + */ +static noinline +int rxrpc_extract_header(struct rxrpc_skb_priv *sp, struct sk_buff *skb) +{ + struct rxrpc_wire_header whdr; + + /* dig out the RxRPC connection details */ + if (skb_copy_bits(skb, 0, &whdr, sizeof(whdr)) < 0) + return -EBADMSG; + if (!pskb_pull(skb, sizeof(whdr))) + BUG(); + + memset(sp, 0, sizeof(*sp)); + sp->hdr.epoch = ntohl(whdr.epoch); + sp->hdr.cid = ntohl(whdr.cid); + sp->hdr.callNumber = ntohl(whdr.callNumber); + sp->hdr.seq = ntohl(whdr.seq); + sp->hdr.serial = ntohl(whdr.serial); + sp->hdr.flags = whdr.flags; + sp->hdr.type = whdr.type; + sp->hdr.userStatus = whdr.userStatus; + sp->hdr.securityIndex = whdr.securityIndex; + sp->hdr._rsvd = ntohs(whdr._rsvd); + sp->hdr.serviceId = ntohs(whdr.serviceId); + return 0; +} + +static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local, + struct sk_buff *skb, + struct rxrpc_skb_priv *sp) +{ + struct rxrpc_peer *peer; + struct rxrpc_transport *trans; + struct rxrpc_connection *conn; + + peer = rxrpc_find_peer(local, ip_hdr(skb)->saddr, + udp_hdr(skb)->source); + if (IS_ERR(peer)) + goto cant_find_conn; + + trans = rxrpc_find_transport(local, peer); + rxrpc_put_peer(peer); + if (!trans) + goto cant_find_conn; + + conn = rxrpc_find_connection(trans, &sp->hdr); + rxrpc_put_transport(trans); + if (!conn) + goto cant_find_conn; + + return conn; +cant_find_conn: + return NULL; +} + +/* + * handle data received on the local endpoint + * - may be called in interrupt context + */ +void rxrpc_data_ready(struct sock *sk) +{ + struct rxrpc_skb_priv *sp; + struct rxrpc_local *local; + struct sk_buff *skb; + int ret; + + _enter("%p", sk); + + ASSERT(!irqs_disabled()); + + read_lock_bh(&rxrpc_local_lock); + local = sk->sk_user_data; + if (local && atomic_read(&local->usage) > 0) + rxrpc_get_local(local); + else + local = NULL; + read_unlock_bh(&rxrpc_local_lock); + if (!local) { + _leave(" [local dead]"); + return; + } + + skb = skb_recv_datagram(sk, 0, 1, &ret); + if (!skb) { + rxrpc_put_local(local); + if (ret == -EAGAIN) + return; + _debug("UDP socket error %d", ret); + return; + } + + rxrpc_new_skb(skb); + + _net("recv skb %p", skb); + + /* we'll probably need to checksum it (didn't call sock_recvmsg) */ + if (skb_checksum_complete(skb)) { + rxrpc_free_skb(skb); + rxrpc_put_local(local); + __UDP_INC_STATS(&init_net, UDP_MIB_INERRORS, 0); + _leave(" [CSUM failed]"); + return; + } + + __UDP_INC_STATS(&init_net, UDP_MIB_INDATAGRAMS, 0); + + /* The socket buffer we have is owned by UDP, with UDP's data all over + * it, but we really want our own data there. + */ + skb_orphan(skb); + sp = rxrpc_skb(skb); + + _net("Rx UDP packet from %08x:%04hu", + ntohl(ip_hdr(skb)->saddr), ntohs(udp_hdr(skb)->source)); + + /* dig out the RxRPC connection details */ + if (rxrpc_extract_header(sp, skb) < 0) + goto bad_message; + + _net("Rx RxRPC %s ep=%x call=%x:%x", + sp->hdr.flags & RXRPC_CLIENT_INITIATED ? "ToServer" : "ToClient", + sp->hdr.epoch, sp->hdr.cid, sp->hdr.callNumber); + + if (sp->hdr.type >= RXRPC_N_PACKET_TYPES || + !((RXRPC_SUPPORTED_PACKET_TYPES >> sp->hdr.type) & 1)) { + _proto("Rx Bad Packet Type %u", sp->hdr.type); + goto bad_message; + } + + if (sp->hdr.type == RXRPC_PACKET_TYPE_VERSION) { + rxrpc_post_packet_to_local(local, skb); + goto out; + } + + if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA && + (sp->hdr.callNumber == 0 || sp->hdr.seq == 0)) + goto bad_message; + + if (sp->hdr.callNumber == 0) { + /* This is a connection-level packet. These should be + * fairly rare, so the extra overhead of looking them up the + * old-fashioned way doesn't really hurt */ + struct rxrpc_connection *conn; + + conn = rxrpc_conn_from_local(local, skb, sp); + if (!conn) + goto cant_route_call; + + _debug("CONN %p {%d}", conn, conn->debug_id); + rxrpc_post_packet_to_conn(conn, skb); + rxrpc_put_connection(conn); + } else { + struct rxrpc_call *call; + + call = rxrpc_find_call_hash(&sp->hdr, local, + AF_INET, &ip_hdr(skb)->saddr); + if (call) + rxrpc_post_packet_to_call(call, skb); + else + goto cant_route_call; + } + +out: + rxrpc_put_local(local); + return; + +cant_route_call: + _debug("can't route call"); + if (sp->hdr.flags & RXRPC_CLIENT_INITIATED && + sp->hdr.type == RXRPC_PACKET_TYPE_DATA) { + if (sp->hdr.seq == 1) { + _debug("first packet"); + skb_queue_tail(&local->accept_queue, skb); + rxrpc_queue_work(&local->acceptor); + rxrpc_put_local(local); + _leave(" [incoming]"); + return; + } + skb->priority = RX_INVALID_OPERATION; + } else { + skb->priority = RX_CALL_DEAD; + } + + if (sp->hdr.type != RXRPC_PACKET_TYPE_ABORT) { + _debug("reject type %d",sp->hdr.type); + rxrpc_reject_packet(local, skb); + } + rxrpc_put_local(local); + _leave(" [no call]"); + return; + +bad_message: + skb->priority = RX_PROTOCOL_ERROR; + rxrpc_reject_packet(local, skb); + rxrpc_put_local(local); + _leave(" [badmsg]"); +} diff --git a/net/rxrpc/key.c b/net/rxrpc/key.c new file mode 100644 index 0000000..4ad56fa --- /dev/null +++ b/net/rxrpc/key.c @@ -0,0 +1,1237 @@ +/* RxRPC key management + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * RxRPC keys should have a description of describing their purpose: + * "afs@CAMBRIDGE.REDHAT.COM> + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +static int rxrpc_vet_description_s(const char *); +static int rxrpc_preparse(struct key_preparsed_payload *); +static int rxrpc_preparse_s(struct key_preparsed_payload *); +static void rxrpc_free_preparse(struct key_preparsed_payload *); +static void rxrpc_free_preparse_s(struct key_preparsed_payload *); +static void rxrpc_destroy(struct key *); +static void rxrpc_destroy_s(struct key *); +static void rxrpc_describe(const struct key *, struct seq_file *); +static long rxrpc_read(const struct key *, char __user *, size_t); + +/* + * rxrpc defined keys take an arbitrary string as the description and an + * arbitrary blob of data as the payload + */ +struct key_type key_type_rxrpc = { + .name = "rxrpc", + .preparse = rxrpc_preparse, + .free_preparse = rxrpc_free_preparse, + .instantiate = generic_key_instantiate, + .destroy = rxrpc_destroy, + .describe = rxrpc_describe, + .read = rxrpc_read, +}; +EXPORT_SYMBOL(key_type_rxrpc); + +/* + * rxrpc server defined keys take ":" as the + * description and an 8-byte decryption key as the payload + */ +struct key_type key_type_rxrpc_s = { + .name = "rxrpc_s", + .vet_description = rxrpc_vet_description_s, + .preparse = rxrpc_preparse_s, + .free_preparse = rxrpc_free_preparse_s, + .instantiate = generic_key_instantiate, + .destroy = rxrpc_destroy_s, + .describe = rxrpc_describe, +}; + +/* + * Vet the description for an RxRPC server key + */ +static int rxrpc_vet_description_s(const char *desc) +{ + unsigned long num; + char *p; + + num = simple_strtoul(desc, &p, 10); + if (*p != ':' || num > 65535) + return -EINVAL; + num = simple_strtoul(p + 1, &p, 10); + if (*p || num < 1 || num > 255) + return -EINVAL; + return 0; +} + +/* + * parse an RxKAD type XDR format token + * - the caller guarantees we have at least 4 words + */ +static int rxrpc_preparse_xdr_rxkad(struct key_preparsed_payload *prep, + size_t datalen, + const __be32 *xdr, unsigned int toklen) +{ + struct rxrpc_key_token *token, **pptoken; + size_t plen; + u32 tktlen; + + _enter(",{%x,%x,%x,%x},%u", + ntohl(xdr[0]), ntohl(xdr[1]), ntohl(xdr[2]), ntohl(xdr[3]), + toklen); + + if (toklen <= 8 * 4) + return -EKEYREJECTED; + tktlen = ntohl(xdr[7]); + _debug("tktlen: %x", tktlen); + if (tktlen > AFSTOKEN_RK_TIX_MAX) + return -EKEYREJECTED; + if (toklen < 8 * 4 + tktlen) + return -EKEYREJECTED; + + plen = sizeof(*token) + sizeof(*token->kad) + tktlen; + prep->quotalen = datalen + plen; + + plen -= sizeof(*token); + token = kzalloc(sizeof(*token), GFP_KERNEL); + if (!token) + return -ENOMEM; + + token->kad = kzalloc(plen, GFP_KERNEL); + if (!token->kad) { + kfree(token); + return -ENOMEM; + } + + token->security_index = RXRPC_SECURITY_RXKAD; + token->kad->ticket_len = tktlen; + token->kad->vice_id = ntohl(xdr[0]); + token->kad->kvno = ntohl(xdr[1]); + token->kad->start = ntohl(xdr[4]); + token->kad->expiry = ntohl(xdr[5]); + token->kad->primary_flag = ntohl(xdr[6]); + memcpy(&token->kad->session_key, &xdr[2], 8); + memcpy(&token->kad->ticket, &xdr[8], tktlen); + + _debug("SCIX: %u", token->security_index); + _debug("TLEN: %u", token->kad->ticket_len); + _debug("EXPY: %x", token->kad->expiry); + _debug("KVNO: %u", token->kad->kvno); + _debug("PRIM: %u", token->kad->primary_flag); + _debug("SKEY: %02x%02x%02x%02x%02x%02x%02x%02x", + token->kad->session_key[0], token->kad->session_key[1], + token->kad->session_key[2], token->kad->session_key[3], + token->kad->session_key[4], token->kad->session_key[5], + token->kad->session_key[6], token->kad->session_key[7]); + if (token->kad->ticket_len >= 8) + _debug("TCKT: %02x%02x%02x%02x%02x%02x%02x%02x", + token->kad->ticket[0], token->kad->ticket[1], + token->kad->ticket[2], token->kad->ticket[3], + token->kad->ticket[4], token->kad->ticket[5], + token->kad->ticket[6], token->kad->ticket[7]); + + /* count the number of tokens attached */ + prep->payload.data[1] = (void *)((unsigned long)prep->payload.data[1] + 1); + + /* attach the data */ + for (pptoken = (struct rxrpc_key_token **)&prep->payload.data[0]; + *pptoken; + pptoken = &(*pptoken)->next) + continue; + *pptoken = token; + if (token->kad->expiry < prep->expiry) + prep->expiry = token->kad->expiry; + + _leave(" = 0"); + return 0; +} + +static void rxrpc_free_krb5_principal(struct krb5_principal *princ) +{ + int loop; + + if (princ->name_parts) { + for (loop = princ->n_name_parts - 1; loop >= 0; loop--) + kfree(princ->name_parts[loop]); + kfree(princ->name_parts); + } + kfree(princ->realm); +} + +static void rxrpc_free_krb5_tagged(struct krb5_tagged_data *td) +{ + kfree(td->data); +} + +/* + * free up an RxK5 token + */ +static void rxrpc_rxk5_free(struct rxk5_key *rxk5) +{ + int loop; + + rxrpc_free_krb5_principal(&rxk5->client); + rxrpc_free_krb5_principal(&rxk5->server); + rxrpc_free_krb5_tagged(&rxk5->session); + + if (rxk5->addresses) { + for (loop = rxk5->n_addresses - 1; loop >= 0; loop--) + rxrpc_free_krb5_tagged(&rxk5->addresses[loop]); + kfree(rxk5->addresses); + } + if (rxk5->authdata) { + for (loop = rxk5->n_authdata - 1; loop >= 0; loop--) + rxrpc_free_krb5_tagged(&rxk5->authdata[loop]); + kfree(rxk5->authdata); + } + + kfree(rxk5->ticket); + kfree(rxk5->ticket2); + kfree(rxk5); +} + +/* + * extract a krb5 principal + */ +static int rxrpc_krb5_decode_principal(struct krb5_principal *princ, + const __be32 **_xdr, + unsigned int *_toklen) +{ + const __be32 *xdr = *_xdr; + unsigned int toklen = *_toklen, n_parts, loop, tmp; + + /* there must be at least one name, and at least #names+1 length + * words */ + if (toklen <= 12) + return -EINVAL; + + _enter(",{%x,%x,%x},%u", + ntohl(xdr[0]), ntohl(xdr[1]), ntohl(xdr[2]), toklen); + + n_parts = ntohl(*xdr++); + toklen -= 4; + if (n_parts <= 0 || n_parts > AFSTOKEN_K5_COMPONENTS_MAX) + return -EINVAL; + princ->n_name_parts = n_parts; + + if (toklen <= (n_parts + 1) * 4) + return -EINVAL; + + princ->name_parts = kcalloc(n_parts, sizeof(char *), GFP_KERNEL); + if (!princ->name_parts) + return -ENOMEM; + + for (loop = 0; loop < n_parts; loop++) { + if (toklen < 4) + return -EINVAL; + tmp = ntohl(*xdr++); + toklen -= 4; + if (tmp <= 0 || tmp > AFSTOKEN_STRING_MAX) + return -EINVAL; + if (tmp > toklen) + return -EINVAL; + princ->name_parts[loop] = kmalloc(tmp + 1, GFP_KERNEL); + if (!princ->name_parts[loop]) + return -ENOMEM; + memcpy(princ->name_parts[loop], xdr, tmp); + princ->name_parts[loop][tmp] = 0; + tmp = (tmp + 3) & ~3; + toklen -= tmp; + xdr += tmp >> 2; + } + + if (toklen < 4) + return -EINVAL; + tmp = ntohl(*xdr++); + toklen -= 4; + if (tmp <= 0 || tmp > AFSTOKEN_K5_REALM_MAX) + return -EINVAL; + if (tmp > toklen) + return -EINVAL; + princ->realm = kmalloc(tmp + 1, GFP_KERNEL); + if (!princ->realm) + return -ENOMEM; + memcpy(princ->realm, xdr, tmp); + princ->realm[tmp] = 0; + tmp = (tmp + 3) & ~3; + toklen -= tmp; + xdr += tmp >> 2; + + _debug("%s/...@%s", princ->name_parts[0], princ->realm); + + *_xdr = xdr; + *_toklen = toklen; + _leave(" = 0 [toklen=%u]", toklen); + return 0; +} + +/* + * extract a piece of krb5 tagged data + */ +static int rxrpc_krb5_decode_tagged_data(struct krb5_tagged_data *td, + size_t max_data_size, + const __be32 **_xdr, + unsigned int *_toklen) +{ + const __be32 *xdr = *_xdr; + unsigned int toklen = *_toklen, len; + + /* there must be at least one tag and one length word */ + if (toklen <= 8) + return -EINVAL; + + _enter(",%zu,{%x,%x},%u", + max_data_size, ntohl(xdr[0]), ntohl(xdr[1]), toklen); + + td->tag = ntohl(*xdr++); + len = ntohl(*xdr++); + toklen -= 8; + if (len > max_data_size) + return -EINVAL; + td->data_len = len; + + if (len > 0) { + td->data = kmemdup(xdr, len, GFP_KERNEL); + if (!td->data) + return -ENOMEM; + len = (len + 3) & ~3; + toklen -= len; + xdr += len >> 2; + } + + _debug("tag %x len %x", td->tag, td->data_len); + + *_xdr = xdr; + *_toklen = toklen; + _leave(" = 0 [toklen=%u]", toklen); + return 0; +} + +/* + * extract an array of tagged data + */ +static int rxrpc_krb5_decode_tagged_array(struct krb5_tagged_data **_td, + u8 *_n_elem, + u8 max_n_elem, + size_t max_elem_size, + const __be32 **_xdr, + unsigned int *_toklen) +{ + struct krb5_tagged_data *td; + const __be32 *xdr = *_xdr; + unsigned int toklen = *_toklen, n_elem, loop; + int ret; + + /* there must be at least one count */ + if (toklen < 4) + return -EINVAL; + + _enter(",,%u,%zu,{%x},%u", + max_n_elem, max_elem_size, ntohl(xdr[0]), toklen); + + n_elem = ntohl(*xdr++); + toklen -= 4; + if (n_elem > max_n_elem) + return -EINVAL; + *_n_elem = n_elem; + if (n_elem > 0) { + if (toklen <= (n_elem + 1) * 4) + return -EINVAL; + + _debug("n_elem %d", n_elem); + + td = kcalloc(n_elem, sizeof(struct krb5_tagged_data), + GFP_KERNEL); + if (!td) + return -ENOMEM; + *_td = td; + + for (loop = 0; loop < n_elem; loop++) { + ret = rxrpc_krb5_decode_tagged_data(&td[loop], + max_elem_size, + &xdr, &toklen); + if (ret < 0) + return ret; + } + } + + *_xdr = xdr; + *_toklen = toklen; + _leave(" = 0 [toklen=%u]", toklen); + return 0; +} + +/* + * extract a krb5 ticket + */ +static int rxrpc_krb5_decode_ticket(u8 **_ticket, u16 *_tktlen, + const __be32 **_xdr, unsigned int *_toklen) +{ + const __be32 *xdr = *_xdr; + unsigned int toklen = *_toklen, len; + + /* there must be at least one length word */ + if (toklen <= 4) + return -EINVAL; + + _enter(",{%x},%u", ntohl(xdr[0]), toklen); + + len = ntohl(*xdr++); + toklen -= 4; + if (len > AFSTOKEN_K5_TIX_MAX) + return -EINVAL; + *_tktlen = len; + + _debug("ticket len %u", len); + + if (len > 0) { + *_ticket = kmemdup(xdr, len, GFP_KERNEL); + if (!*_ticket) + return -ENOMEM; + len = (len + 3) & ~3; + toklen -= len; + xdr += len >> 2; + } + + *_xdr = xdr; + *_toklen = toklen; + _leave(" = 0 [toklen=%u]", toklen); + return 0; +} + +/* + * parse an RxK5 type XDR format token + * - the caller guarantees we have at least 4 words + */ +static int rxrpc_preparse_xdr_rxk5(struct key_preparsed_payload *prep, + size_t datalen, + const __be32 *xdr, unsigned int toklen) +{ + struct rxrpc_key_token *token, **pptoken; + struct rxk5_key *rxk5; + const __be32 *end_xdr = xdr + (toklen >> 2); + int ret; + + _enter(",{%x,%x,%x,%x},%u", + ntohl(xdr[0]), ntohl(xdr[1]), ntohl(xdr[2]), ntohl(xdr[3]), + toklen); + + /* reserve some payload space for this subkey - the length of the token + * is a reasonable approximation */ + prep->quotalen = datalen + toklen; + + token = kzalloc(sizeof(*token), GFP_KERNEL); + if (!token) + return -ENOMEM; + + rxk5 = kzalloc(sizeof(*rxk5), GFP_KERNEL); + if (!rxk5) { + kfree(token); + return -ENOMEM; + } + + token->security_index = RXRPC_SECURITY_RXK5; + token->k5 = rxk5; + + /* extract the principals */ + ret = rxrpc_krb5_decode_principal(&rxk5->client, &xdr, &toklen); + if (ret < 0) + goto error; + ret = rxrpc_krb5_decode_principal(&rxk5->server, &xdr, &toklen); + if (ret < 0) + goto error; + + /* extract the session key and the encoding type (the tag field -> + * ENCTYPE_xxx) */ + ret = rxrpc_krb5_decode_tagged_data(&rxk5->session, AFSTOKEN_DATA_MAX, + &xdr, &toklen); + if (ret < 0) + goto error; + + if (toklen < 4 * 8 + 2 * 4) + goto inval; + rxk5->authtime = be64_to_cpup((const __be64 *) xdr); + xdr += 2; + rxk5->starttime = be64_to_cpup((const __be64 *) xdr); + xdr += 2; + rxk5->endtime = be64_to_cpup((const __be64 *) xdr); + xdr += 2; + rxk5->renew_till = be64_to_cpup((const __be64 *) xdr); + xdr += 2; + rxk5->is_skey = ntohl(*xdr++); + rxk5->flags = ntohl(*xdr++); + toklen -= 4 * 8 + 2 * 4; + + _debug("times: a=%llx s=%llx e=%llx rt=%llx", + rxk5->authtime, rxk5->starttime, rxk5->endtime, + rxk5->renew_till); + _debug("is_skey=%x flags=%x", rxk5->is_skey, rxk5->flags); + + /* extract the permitted client addresses */ + ret = rxrpc_krb5_decode_tagged_array(&rxk5->addresses, + &rxk5->n_addresses, + AFSTOKEN_K5_ADDRESSES_MAX, + AFSTOKEN_DATA_MAX, + &xdr, &toklen); + if (ret < 0) + goto error; + + ASSERTCMP((end_xdr - xdr) << 2, ==, toklen); + + /* extract the tickets */ + ret = rxrpc_krb5_decode_ticket(&rxk5->ticket, &rxk5->ticket_len, + &xdr, &toklen); + if (ret < 0) + goto error; + ret = rxrpc_krb5_decode_ticket(&rxk5->ticket2, &rxk5->ticket2_len, + &xdr, &toklen); + if (ret < 0) + goto error; + + ASSERTCMP((end_xdr - xdr) << 2, ==, toklen); + + /* extract the typed auth data */ + ret = rxrpc_krb5_decode_tagged_array(&rxk5->authdata, + &rxk5->n_authdata, + AFSTOKEN_K5_AUTHDATA_MAX, + AFSTOKEN_BDATALN_MAX, + &xdr, &toklen); + if (ret < 0) + goto error; + + ASSERTCMP((end_xdr - xdr) << 2, ==, toklen); + + if (toklen != 0) + goto inval; + + /* attach the payload */ + for (pptoken = (struct rxrpc_key_token **)&prep->payload.data[0]; + *pptoken; + pptoken = &(*pptoken)->next) + continue; + *pptoken = token; + if (token->kad->expiry < prep->expiry) + prep->expiry = token->kad->expiry; + + _leave(" = 0"); + return 0; + +inval: + ret = -EINVAL; +error: + rxrpc_rxk5_free(rxk5); + kfree(token); + _leave(" = %d", ret); + return ret; +} + +/* + * attempt to parse the data as the XDR format + * - the caller guarantees we have more than 7 words + */ +static int rxrpc_preparse_xdr(struct key_preparsed_payload *prep) +{ + const __be32 *xdr = prep->data, *token; + const char *cp; + unsigned int len, tmp, loop, ntoken, toklen, sec_ix; + size_t datalen = prep->datalen; + int ret; + + _enter(",{%x,%x,%x,%x},%zu", + ntohl(xdr[0]), ntohl(xdr[1]), ntohl(xdr[2]), ntohl(xdr[3]), + prep->datalen); + + if (datalen > AFSTOKEN_LENGTH_MAX) + goto not_xdr; + + /* XDR is an array of __be32's */ + if (datalen & 3) + goto not_xdr; + + /* the flags should be 0 (the setpag bit must be handled by + * userspace) */ + if (ntohl(*xdr++) != 0) + goto not_xdr; + datalen -= 4; + + /* check the cell name */ + len = ntohl(*xdr++); + if (len < 1 || len > AFSTOKEN_CELL_MAX) + goto not_xdr; + datalen -= 4; + tmp = (len + 3) & ~3; + if (tmp > datalen) + goto not_xdr; + + cp = (const char *) xdr; + for (loop = 0; loop < len; loop++) + if (!isprint(cp[loop])) + goto not_xdr; + if (len < tmp) + for (; loop < tmp; loop++) + if (cp[loop]) + goto not_xdr; + _debug("cellname: [%u/%u] '%*.*s'", + len, tmp, len, len, (const char *) xdr); + datalen -= tmp; + xdr += tmp >> 2; + + /* get the token count */ + if (datalen < 12) + goto not_xdr; + ntoken = ntohl(*xdr++); + datalen -= 4; + _debug("ntoken: %x", ntoken); + if (ntoken < 1 || ntoken > AFSTOKEN_MAX) + goto not_xdr; + + /* check each token wrapper */ + token = xdr; + loop = ntoken; + do { + if (datalen < 8) + goto not_xdr; + toklen = ntohl(*xdr++); + sec_ix = ntohl(*xdr); + datalen -= 4; + _debug("token: [%x/%zx] %x", toklen, datalen, sec_ix); + if (toklen < 20 || toklen > datalen) + goto not_xdr; + datalen -= (toklen + 3) & ~3; + xdr += (toklen + 3) >> 2; + + } while (--loop > 0); + + _debug("remainder: %zu", datalen); + if (datalen != 0) + goto not_xdr; + + /* okay: we're going to assume it's valid XDR format + * - we ignore the cellname, relying on the key to be correctly named + */ + do { + xdr = token; + toklen = ntohl(*xdr++); + token = xdr + ((toklen + 3) >> 2); + sec_ix = ntohl(*xdr++); + toklen -= 4; + + _debug("TOKEN type=%u [%p-%p]", sec_ix, xdr, token); + + switch (sec_ix) { + case RXRPC_SECURITY_RXKAD: + ret = rxrpc_preparse_xdr_rxkad(prep, datalen, xdr, toklen); + if (ret != 0) + goto error; + break; + + case RXRPC_SECURITY_RXK5: + ret = rxrpc_preparse_xdr_rxk5(prep, datalen, xdr, toklen); + if (ret != 0) + goto error; + break; + + default: + ret = -EPROTONOSUPPORT; + goto error; + } + + } while (--ntoken > 0); + + _leave(" = 0"); + return 0; + +not_xdr: + _leave(" = -EPROTO"); + return -EPROTO; +error: + _leave(" = %d", ret); + return ret; +} + +/* + * Preparse an rxrpc defined key. + * + * Data should be of the form: + * OFFSET LEN CONTENT + * 0 4 key interface version number + * 4 2 security index (type) + * 6 2 ticket length + * 8 4 key expiry time (time_t) + * 12 4 kvno + * 16 8 session key + * 24 [len] ticket + * + * if no data is provided, then a no-security key is made + */ +static int rxrpc_preparse(struct key_preparsed_payload *prep) +{ + const struct rxrpc_key_data_v1 *v1; + struct rxrpc_key_token *token, **pp; + size_t plen; + u32 kver; + int ret; + + _enter("%zu", prep->datalen); + + /* handle a no-security key */ + if (!prep->data && prep->datalen == 0) + return 0; + + /* determine if the XDR payload format is being used */ + if (prep->datalen > 7 * 4) { + ret = rxrpc_preparse_xdr(prep); + if (ret != -EPROTO) + return ret; + } + + /* get the key interface version number */ + ret = -EINVAL; + if (prep->datalen <= 4 || !prep->data) + goto error; + memcpy(&kver, prep->data, sizeof(kver)); + prep->data += sizeof(kver); + prep->datalen -= sizeof(kver); + + _debug("KEY I/F VERSION: %u", kver); + + ret = -EKEYREJECTED; + if (kver != 1) + goto error; + + /* deal with a version 1 key */ + ret = -EINVAL; + if (prep->datalen < sizeof(*v1)) + goto error; + + v1 = prep->data; + if (prep->datalen != sizeof(*v1) + v1->ticket_length) + goto error; + + _debug("SCIX: %u", v1->security_index); + _debug("TLEN: %u", v1->ticket_length); + _debug("EXPY: %x", v1->expiry); + _debug("KVNO: %u", v1->kvno); + _debug("SKEY: %02x%02x%02x%02x%02x%02x%02x%02x", + v1->session_key[0], v1->session_key[1], + v1->session_key[2], v1->session_key[3], + v1->session_key[4], v1->session_key[5], + v1->session_key[6], v1->session_key[7]); + if (v1->ticket_length >= 8) + _debug("TCKT: %02x%02x%02x%02x%02x%02x%02x%02x", + v1->ticket[0], v1->ticket[1], + v1->ticket[2], v1->ticket[3], + v1->ticket[4], v1->ticket[5], + v1->ticket[6], v1->ticket[7]); + + ret = -EPROTONOSUPPORT; + if (v1->security_index != RXRPC_SECURITY_RXKAD) + goto error; + + plen = sizeof(*token->kad) + v1->ticket_length; + prep->quotalen = plen + sizeof(*token); + + ret = -ENOMEM; + token = kzalloc(sizeof(*token), GFP_KERNEL); + if (!token) + goto error; + token->kad = kzalloc(plen, GFP_KERNEL); + if (!token->kad) + goto error_free; + + token->security_index = RXRPC_SECURITY_RXKAD; + token->kad->ticket_len = v1->ticket_length; + token->kad->expiry = v1->expiry; + token->kad->kvno = v1->kvno; + memcpy(&token->kad->session_key, &v1->session_key, 8); + memcpy(&token->kad->ticket, v1->ticket, v1->ticket_length); + + /* count the number of tokens attached */ + prep->payload.data[1] = (void *)((unsigned long)prep->payload.data[1] + 1); + + /* attach the data */ + pp = (struct rxrpc_key_token **)&prep->payload.data[0]; + while (*pp) + pp = &(*pp)->next; + *pp = token; + if (token->kad->expiry < prep->expiry) + prep->expiry = token->kad->expiry; + token = NULL; + ret = 0; + +error_free: + kfree(token); +error: + return ret; +} + +/* + * Free token list. + */ +static void rxrpc_free_token_list(struct rxrpc_key_token *token) +{ + struct rxrpc_key_token *next; + + for (; token; token = next) { + next = token->next; + switch (token->security_index) { + case RXRPC_SECURITY_RXKAD: + kfree(token->kad); + break; + case RXRPC_SECURITY_RXK5: + if (token->k5) + rxrpc_rxk5_free(token->k5); + break; + default: + pr_err("Unknown token type %x on rxrpc key\n", + token->security_index); + BUG(); + } + + kfree(token); + } +} + +/* + * Clean up preparse data. + */ +static void rxrpc_free_preparse(struct key_preparsed_payload *prep) +{ + rxrpc_free_token_list(prep->payload.data[0]); +} + +/* + * Preparse a server secret key. + * + * The data should be the 8-byte secret key. + */ +static int rxrpc_preparse_s(struct key_preparsed_payload *prep) +{ + struct crypto_skcipher *ci; + + _enter("%zu", prep->datalen); + + if (prep->datalen != 8) + return -EINVAL; + + memcpy(&prep->payload.data[2], prep->data, 8); + + ci = crypto_alloc_skcipher("pcbc(des)", 0, CRYPTO_ALG_ASYNC); + if (IS_ERR(ci)) { + _leave(" = %ld", PTR_ERR(ci)); + return PTR_ERR(ci); + } + + if (crypto_skcipher_setkey(ci, prep->data, 8) < 0) + BUG(); + + prep->payload.data[0] = ci; + _leave(" = 0"); + return 0; +} + +/* + * Clean up preparse data. + */ +static void rxrpc_free_preparse_s(struct key_preparsed_payload *prep) +{ + if (prep->payload.data[0]) + crypto_free_skcipher(prep->payload.data[0]); +} + +/* + * dispose of the data dangling from the corpse of a rxrpc key + */ +static void rxrpc_destroy(struct key *key) +{ + rxrpc_free_token_list(key->payload.data[0]); +} + +/* + * dispose of the data dangling from the corpse of a rxrpc key + */ +static void rxrpc_destroy_s(struct key *key) +{ + if (key->payload.data[0]) { + crypto_free_skcipher(key->payload.data[0]); + key->payload.data[0] = NULL; + } +} + +/* + * describe the rxrpc key + */ +static void rxrpc_describe(const struct key *key, struct seq_file *m) +{ + seq_puts(m, key->description); +} + +/* + * grab the security key for a socket + */ +int rxrpc_request_key(struct rxrpc_sock *rx, char __user *optval, int optlen) +{ + struct key *key; + char *description; + + _enter(""); + + if (optlen <= 0 || optlen > PAGE_SIZE - 1) + return -EINVAL; + + description = memdup_user_nul(optval, optlen); + if (IS_ERR(description)) + return PTR_ERR(description); + + key = request_key(&key_type_rxrpc, description, NULL); + if (IS_ERR(key)) { + kfree(description); + _leave(" = %ld", PTR_ERR(key)); + return PTR_ERR(key); + } + + rx->key = key; + kfree(description); + _leave(" = 0 [key %x]", key->serial); + return 0; +} + +/* + * grab the security keyring for a server socket + */ +int rxrpc_server_keyring(struct rxrpc_sock *rx, char __user *optval, + int optlen) +{ + struct key *key; + char *description; + + _enter(""); + + if (optlen <= 0 || optlen > PAGE_SIZE - 1) + return -EINVAL; + + description = memdup_user_nul(optval, optlen); + if (IS_ERR(description)) + return PTR_ERR(description); + + key = request_key(&key_type_keyring, description, NULL); + if (IS_ERR(key)) { + kfree(description); + _leave(" = %ld", PTR_ERR(key)); + return PTR_ERR(key); + } + + rx->securities = key; + kfree(description); + _leave(" = 0 [key %x]", key->serial); + return 0; +} + +/* + * generate a server data key + */ +int rxrpc_get_server_data_key(struct rxrpc_connection *conn, + const void *session_key, + time_t expiry, + u32 kvno) +{ + const struct cred *cred = current_cred(); + struct key *key; + int ret; + + struct { + u32 kver; + struct rxrpc_key_data_v1 v1; + } data; + + _enter(""); + + key = key_alloc(&key_type_rxrpc, "x", + GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, cred, 0, + KEY_ALLOC_NOT_IN_QUOTA, NULL); + if (IS_ERR(key)) { + _leave(" = -ENOMEM [alloc %ld]", PTR_ERR(key)); + return -ENOMEM; + } + + _debug("key %d", key_serial(key)); + + data.kver = 1; + data.v1.security_index = RXRPC_SECURITY_RXKAD; + data.v1.ticket_length = 0; + data.v1.expiry = expiry; + data.v1.kvno = 0; + + memcpy(&data.v1.session_key, session_key, sizeof(data.v1.session_key)); + + ret = key_instantiate_and_link(key, &data, sizeof(data), NULL, NULL); + if (ret < 0) + goto error; + + conn->key = key; + _leave(" = 0 [%d]", key_serial(key)); + return 0; + +error: + key_revoke(key); + key_put(key); + _leave(" = -ENOMEM [ins %d]", ret); + return -ENOMEM; +} +EXPORT_SYMBOL(rxrpc_get_server_data_key); + +/** + * rxrpc_get_null_key - Generate a null RxRPC key + * @keyname: The name to give the key. + * + * Generate a null RxRPC key that can be used to indicate anonymous security is + * required for a particular domain. + */ +struct key *rxrpc_get_null_key(const char *keyname) +{ + const struct cred *cred = current_cred(); + struct key *key; + int ret; + + key = key_alloc(&key_type_rxrpc, keyname, + GLOBAL_ROOT_UID, GLOBAL_ROOT_GID, cred, + KEY_POS_SEARCH, KEY_ALLOC_NOT_IN_QUOTA, NULL); + if (IS_ERR(key)) + return key; + + ret = key_instantiate_and_link(key, NULL, 0, NULL, NULL); + if (ret < 0) { + key_revoke(key); + key_put(key); + return ERR_PTR(ret); + } + + return key; +} +EXPORT_SYMBOL(rxrpc_get_null_key); + +/* + * read the contents of an rxrpc key + * - this returns the result in XDR form + */ +static long rxrpc_read(const struct key *key, + char __user *buffer, size_t buflen) +{ + const struct rxrpc_key_token *token; + const struct krb5_principal *princ; + size_t size; + __be32 __user *xdr, *oldxdr; + u32 cnlen, toksize, ntoks, tok, zero; + u16 toksizes[AFSTOKEN_MAX]; + int loop; + + _enter(""); + + /* we don't know what form we should return non-AFS keys in */ + if (memcmp(key->description, "afs@", 4) != 0) + return -EOPNOTSUPP; + cnlen = strlen(key->description + 4); + +#define RND(X) (((X) + 3) & ~3) + + /* AFS keys we return in XDR form, so we need to work out the size of + * the XDR */ + size = 2 * 4; /* flags, cellname len */ + size += RND(cnlen); /* cellname */ + size += 1 * 4; /* token count */ + + ntoks = 0; + for (token = key->payload.data[0]; token; token = token->next) { + toksize = 4; /* sec index */ + + switch (token->security_index) { + case RXRPC_SECURITY_RXKAD: + toksize += 8 * 4; /* viceid, kvno, key*2, begin, + * end, primary, tktlen */ + toksize += RND(token->kad->ticket_len); + break; + + case RXRPC_SECURITY_RXK5: + princ = &token->k5->client; + toksize += 4 + princ->n_name_parts * 4; + for (loop = 0; loop < princ->n_name_parts; loop++) + toksize += RND(strlen(princ->name_parts[loop])); + toksize += 4 + RND(strlen(princ->realm)); + + princ = &token->k5->server; + toksize += 4 + princ->n_name_parts * 4; + for (loop = 0; loop < princ->n_name_parts; loop++) + toksize += RND(strlen(princ->name_parts[loop])); + toksize += 4 + RND(strlen(princ->realm)); + + toksize += 8 + RND(token->k5->session.data_len); + + toksize += 4 * 8 + 2 * 4; + + toksize += 4 + token->k5->n_addresses * 8; + for (loop = 0; loop < token->k5->n_addresses; loop++) + toksize += RND(token->k5->addresses[loop].data_len); + + toksize += 4 + RND(token->k5->ticket_len); + toksize += 4 + RND(token->k5->ticket2_len); + + toksize += 4 + token->k5->n_authdata * 8; + for (loop = 0; loop < token->k5->n_authdata; loop++) + toksize += RND(token->k5->authdata[loop].data_len); + break; + + default: /* we have a ticket we can't encode */ + BUG(); + continue; + } + + _debug("token[%u]: toksize=%u", ntoks, toksize); + ASSERTCMP(toksize, <=, AFSTOKEN_LENGTH_MAX); + + toksizes[ntoks++] = toksize; + size += toksize + 4; /* each token has a length word */ + } + +#undef RND + + if (!buffer || buflen < size) + return size; + + xdr = (__be32 __user *) buffer; + zero = 0; +#define ENCODE(x) \ + do { \ + __be32 y = htonl(x); \ + if (put_user(y, xdr++) < 0) \ + goto fault; \ + } while(0) +#define ENCODE_DATA(l, s) \ + do { \ + u32 _l = (l); \ + ENCODE(l); \ + if (copy_to_user(xdr, (s), _l) != 0) \ + goto fault; \ + if (_l & 3 && \ + copy_to_user((u8 __user *)xdr + _l, &zero, 4 - (_l & 3)) != 0) \ + goto fault; \ + xdr += (_l + 3) >> 2; \ + } while(0) +#define ENCODE64(x) \ + do { \ + __be64 y = cpu_to_be64(x); \ + if (copy_to_user(xdr, &y, 8) != 0) \ + goto fault; \ + xdr += 8 >> 2; \ + } while(0) +#define ENCODE_STR(s) \ + do { \ + const char *_s = (s); \ + ENCODE_DATA(strlen(_s), _s); \ + } while(0) + + ENCODE(0); /* flags */ + ENCODE_DATA(cnlen, key->description + 4); /* cellname */ + ENCODE(ntoks); + + tok = 0; + for (token = key->payload.data[0]; token; token = token->next) { + toksize = toksizes[tok++]; + ENCODE(toksize); + oldxdr = xdr; + ENCODE(token->security_index); + + switch (token->security_index) { + case RXRPC_SECURITY_RXKAD: + ENCODE(token->kad->vice_id); + ENCODE(token->kad->kvno); + ENCODE_DATA(8, token->kad->session_key); + ENCODE(token->kad->start); + ENCODE(token->kad->expiry); + ENCODE(token->kad->primary_flag); + ENCODE_DATA(token->kad->ticket_len, token->kad->ticket); + break; + + case RXRPC_SECURITY_RXK5: + princ = &token->k5->client; + ENCODE(princ->n_name_parts); + for (loop = 0; loop < princ->n_name_parts; loop++) + ENCODE_STR(princ->name_parts[loop]); + ENCODE_STR(princ->realm); + + princ = &token->k5->server; + ENCODE(princ->n_name_parts); + for (loop = 0; loop < princ->n_name_parts; loop++) + ENCODE_STR(princ->name_parts[loop]); + ENCODE_STR(princ->realm); + + ENCODE(token->k5->session.tag); + ENCODE_DATA(token->k5->session.data_len, + token->k5->session.data); + + ENCODE64(token->k5->authtime); + ENCODE64(token->k5->starttime); + ENCODE64(token->k5->endtime); + ENCODE64(token->k5->renew_till); + ENCODE(token->k5->is_skey); + ENCODE(token->k5->flags); + + ENCODE(token->k5->n_addresses); + for (loop = 0; loop < token->k5->n_addresses; loop++) { + ENCODE(token->k5->addresses[loop].tag); + ENCODE_DATA(token->k5->addresses[loop].data_len, + token->k5->addresses[loop].data); + } + + ENCODE_DATA(token->k5->ticket_len, token->k5->ticket); + ENCODE_DATA(token->k5->ticket2_len, token->k5->ticket2); + + ENCODE(token->k5->n_authdata); + for (loop = 0; loop < token->k5->n_authdata; loop++) { + ENCODE(token->k5->authdata[loop].tag); + ENCODE_DATA(token->k5->authdata[loop].data_len, + token->k5->authdata[loop].data); + } + break; + + default: + BUG(); + break; + } + + ASSERTCMP((unsigned long)xdr - (unsigned long)oldxdr, ==, + toksize); + } + +#undef ENCODE_STR +#undef ENCODE_DATA +#undef ENCODE64 +#undef ENCODE + + ASSERTCMP(tok, ==, ntoks); + ASSERTCMP((char __user *) xdr - buffer, ==, size); + _leave(" = %zu", size); + return size; + +fault: + _leave(" = -EFAULT"); + return -EFAULT; +} diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c new file mode 100644 index 0000000..111f250 --- /dev/null +++ b/net/rxrpc/local_object.c @@ -0,0 +1,417 @@ +/* AF_RXRPC local endpoint management + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +static const char rxrpc_version_string[65] = "linux-" UTS_RELEASE " AF_RXRPC"; + +static LIST_HEAD(rxrpc_locals); +DEFINE_RWLOCK(rxrpc_local_lock); +static DECLARE_RWSEM(rxrpc_local_sem); +static DECLARE_WAIT_QUEUE_HEAD(rxrpc_local_wq); + +static void rxrpc_destroy_local(struct work_struct *work); +static void rxrpc_process_local_events(struct work_struct *work); + +/* + * allocate a new local + */ +static +struct rxrpc_local *rxrpc_alloc_local(struct sockaddr_rxrpc *srx) +{ + struct rxrpc_local *local; + + local = kzalloc(sizeof(struct rxrpc_local), GFP_KERNEL); + if (local) { + INIT_WORK(&local->destroyer, &rxrpc_destroy_local); + INIT_WORK(&local->acceptor, &rxrpc_accept_incoming_calls); + INIT_WORK(&local->rejecter, &rxrpc_reject_packets); + INIT_WORK(&local->event_processor, &rxrpc_process_local_events); + INIT_LIST_HEAD(&local->services); + INIT_LIST_HEAD(&local->link); + init_rwsem(&local->defrag_sem); + skb_queue_head_init(&local->accept_queue); + skb_queue_head_init(&local->reject_queue); + skb_queue_head_init(&local->event_queue); + spin_lock_init(&local->lock); + rwlock_init(&local->services_lock); + atomic_set(&local->usage, 1); + local->debug_id = atomic_inc_return(&rxrpc_debug_id); + memcpy(&local->srx, srx, sizeof(*srx)); + } + + _leave(" = %p", local); + return local; +} + +/* + * create the local socket + * - must be called with rxrpc_local_sem writelocked + */ +static int rxrpc_create_local(struct rxrpc_local *local) +{ + struct sock *sock; + int ret, opt; + + _enter("%p{%d}", local, local->srx.transport_type); + + /* create a socket to represent the local endpoint */ + ret = sock_create_kern(&init_net, PF_INET, local->srx.transport_type, + IPPROTO_UDP, &local->socket); + if (ret < 0) { + _leave(" = %d [socket]", ret); + return ret; + } + + /* if a local address was supplied then bind it */ + if (local->srx.transport_len > sizeof(sa_family_t)) { + _debug("bind"); + ret = kernel_bind(local->socket, + (struct sockaddr *) &local->srx.transport, + local->srx.transport_len); + if (ret < 0) { + _debug("bind failed"); + goto error; + } + } + + /* we want to receive ICMP errors */ + opt = 1; + ret = kernel_setsockopt(local->socket, SOL_IP, IP_RECVERR, + (char *) &opt, sizeof(opt)); + if (ret < 0) { + _debug("setsockopt failed"); + goto error; + } + + /* we want to set the don't fragment bit */ + opt = IP_PMTUDISC_DO; + ret = kernel_setsockopt(local->socket, SOL_IP, IP_MTU_DISCOVER, + (char *) &opt, sizeof(opt)); + if (ret < 0) { + _debug("setsockopt failed"); + goto error; + } + + write_lock_bh(&rxrpc_local_lock); + list_add(&local->link, &rxrpc_locals); + write_unlock_bh(&rxrpc_local_lock); + + /* set the socket up */ + sock = local->socket->sk; + sock->sk_user_data = local; + sock->sk_data_ready = rxrpc_data_ready; + sock->sk_error_report = rxrpc_UDP_error_report; + _leave(" = 0"); + return 0; + +error: + kernel_sock_shutdown(local->socket, SHUT_RDWR); + local->socket->sk->sk_user_data = NULL; + sock_release(local->socket); + local->socket = NULL; + + _leave(" = %d", ret); + return ret; +} + +/* + * create a new local endpoint using the specified UDP address + */ +struct rxrpc_local *rxrpc_lookup_local(struct sockaddr_rxrpc *srx) +{ + struct rxrpc_local *local; + int ret; + + _enter("{%d,%u,%pI4+%hu}", + srx->transport_type, + srx->transport.family, + &srx->transport.sin.sin_addr, + ntohs(srx->transport.sin.sin_port)); + + down_write(&rxrpc_local_sem); + + /* see if we have a suitable local local endpoint already */ + read_lock_bh(&rxrpc_local_lock); + + list_for_each_entry(local, &rxrpc_locals, link) { + _debug("CMP {%d,%u,%pI4+%hu}", + local->srx.transport_type, + local->srx.transport.family, + &local->srx.transport.sin.sin_addr, + ntohs(local->srx.transport.sin.sin_port)); + + if (local->srx.transport_type != srx->transport_type || + local->srx.transport.family != srx->transport.family) + continue; + + switch (srx->transport.family) { + case AF_INET: + if (local->srx.transport.sin.sin_port != + srx->transport.sin.sin_port) + continue; + if (memcmp(&local->srx.transport.sin.sin_addr, + &srx->transport.sin.sin_addr, + sizeof(struct in_addr)) != 0) + continue; + goto found_local; + + default: + BUG(); + } + } + + read_unlock_bh(&rxrpc_local_lock); + + /* we didn't find one, so we need to create one */ + local = rxrpc_alloc_local(srx); + if (!local) { + up_write(&rxrpc_local_sem); + return ERR_PTR(-ENOMEM); + } + + ret = rxrpc_create_local(local); + if (ret < 0) { + up_write(&rxrpc_local_sem); + kfree(local); + _leave(" = %d", ret); + return ERR_PTR(ret); + } + + up_write(&rxrpc_local_sem); + + _net("LOCAL new %d {%d,%u,%pI4+%hu}", + local->debug_id, + local->srx.transport_type, + local->srx.transport.family, + &local->srx.transport.sin.sin_addr, + ntohs(local->srx.transport.sin.sin_port)); + + _leave(" = %p [new]", local); + return local; + +found_local: + rxrpc_get_local(local); + read_unlock_bh(&rxrpc_local_lock); + up_write(&rxrpc_local_sem); + + _net("LOCAL old %d {%d,%u,%pI4+%hu}", + local->debug_id, + local->srx.transport_type, + local->srx.transport.family, + &local->srx.transport.sin.sin_addr, + ntohs(local->srx.transport.sin.sin_port)); + + _leave(" = %p [reuse]", local); + return local; +} + +/* + * release a local endpoint + */ +void rxrpc_put_local(struct rxrpc_local *local) +{ + _enter("%p{u=%d}", local, atomic_read(&local->usage)); + + ASSERTCMP(atomic_read(&local->usage), >, 0); + + /* to prevent a race, the decrement and the dequeue must be effectively + * atomic */ + write_lock_bh(&rxrpc_local_lock); + if (unlikely(atomic_dec_and_test(&local->usage))) { + _debug("destroy local"); + rxrpc_queue_work(&local->destroyer); + } + write_unlock_bh(&rxrpc_local_lock); + _leave(""); +} + +/* + * destroy a local endpoint + */ +static void rxrpc_destroy_local(struct work_struct *work) +{ + struct rxrpc_local *local = + container_of(work, struct rxrpc_local, destroyer); + + _enter("%p{%d}", local, atomic_read(&local->usage)); + + down_write(&rxrpc_local_sem); + + write_lock_bh(&rxrpc_local_lock); + if (atomic_read(&local->usage) > 0) { + write_unlock_bh(&rxrpc_local_lock); + up_read(&rxrpc_local_sem); + _leave(" [resurrected]"); + return; + } + + list_del(&local->link); + local->socket->sk->sk_user_data = NULL; + write_unlock_bh(&rxrpc_local_lock); + + downgrade_write(&rxrpc_local_sem); + + ASSERT(list_empty(&local->services)); + ASSERT(!work_pending(&local->acceptor)); + ASSERT(!work_pending(&local->rejecter)); + ASSERT(!work_pending(&local->event_processor)); + + /* finish cleaning up the local descriptor */ + rxrpc_purge_queue(&local->accept_queue); + rxrpc_purge_queue(&local->reject_queue); + rxrpc_purge_queue(&local->event_queue); + kernel_sock_shutdown(local->socket, SHUT_RDWR); + sock_release(local->socket); + + up_read(&rxrpc_local_sem); + + _net("DESTROY LOCAL %d", local->debug_id); + kfree(local); + + if (list_empty(&rxrpc_locals)) + wake_up_all(&rxrpc_local_wq); + + _leave(""); +} + +/* + * preemptively destroy all local local endpoint rather than waiting for + * them to be destroyed + */ +void __exit rxrpc_destroy_all_locals(void) +{ + DECLARE_WAITQUEUE(myself,current); + + _enter(""); + + /* we simply have to wait for them to go away */ + if (!list_empty(&rxrpc_locals)) { + set_current_state(TASK_UNINTERRUPTIBLE); + add_wait_queue(&rxrpc_local_wq, &myself); + + while (!list_empty(&rxrpc_locals)) { + schedule(); + set_current_state(TASK_UNINTERRUPTIBLE); + } + + remove_wait_queue(&rxrpc_local_wq, &myself); + set_current_state(TASK_RUNNING); + } + + _leave(""); +} + +/* + * Reply to a version request + */ +static void rxrpc_send_version_request(struct rxrpc_local *local, + struct rxrpc_host_header *hdr, + struct sk_buff *skb) +{ + struct rxrpc_wire_header whdr; + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + struct sockaddr_in sin; + struct msghdr msg; + struct kvec iov[2]; + size_t len; + int ret; + + _enter(""); + + sin.sin_family = AF_INET; + sin.sin_port = udp_hdr(skb)->source; + sin.sin_addr.s_addr = ip_hdr(skb)->saddr; + + msg.msg_name = &sin; + msg.msg_namelen = sizeof(sin); + msg.msg_control = NULL; + msg.msg_controllen = 0; + msg.msg_flags = 0; + + whdr.epoch = htonl(sp->hdr.epoch); + whdr.cid = htonl(sp->hdr.cid); + whdr.callNumber = htonl(sp->hdr.callNumber); + whdr.seq = 0; + whdr.serial = 0; + whdr.type = RXRPC_PACKET_TYPE_VERSION; + whdr.flags = RXRPC_LAST_PACKET | (~hdr->flags & RXRPC_CLIENT_INITIATED); + whdr.userStatus = 0; + whdr.securityIndex = 0; + whdr._rsvd = 0; + whdr.serviceId = htons(sp->hdr.serviceId); + + iov[0].iov_base = &whdr; + iov[0].iov_len = sizeof(whdr); + iov[1].iov_base = (char *)rxrpc_version_string; + iov[1].iov_len = sizeof(rxrpc_version_string); + + len = iov[0].iov_len + iov[1].iov_len; + + _proto("Tx VERSION (reply)"); + + ret = kernel_sendmsg(local->socket, &msg, iov, 2, len); + if (ret < 0) + _debug("sendmsg failed: %d", ret); + + _leave(""); +} + +/* + * Process event packets targetted at a local endpoint. + */ +static void rxrpc_process_local_events(struct work_struct *work) +{ + struct rxrpc_local *local = container_of(work, struct rxrpc_local, event_processor); + struct sk_buff *skb; + char v; + + _enter(""); + + atomic_inc(&local->usage); + + while ((skb = skb_dequeue(&local->event_queue))) { + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + + _debug("{%d},{%u}", local->debug_id, sp->hdr.type); + + switch (sp->hdr.type) { + case RXRPC_PACKET_TYPE_VERSION: + if (skb_copy_bits(skb, 0, &v, 1) < 0) + return; + _proto("Rx VERSION { %02x }", v); + if (v == 0) + rxrpc_send_version_request(local, &sp->hdr, skb); + break; + + default: + /* Just ignore anything we don't understand */ + break; + } + + rxrpc_put_local(local); + rxrpc_free_skb(skb); + } + + rxrpc_put_local(local); + _leave(""); +} diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c new file mode 100644 index 0000000..2e3c406 --- /dev/null +++ b/net/rxrpc/output.c @@ -0,0 +1,724 @@ +/* RxRPC packet transmission + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * Time till packet resend (in jiffies). + */ +unsigned int rxrpc_resend_timeout = 4 * HZ; + +static int rxrpc_send_data(struct rxrpc_sock *rx, + struct rxrpc_call *call, + struct msghdr *msg, size_t len); + +/* + * extract control messages from the sendmsg() control buffer + */ +static int rxrpc_sendmsg_cmsg(struct msghdr *msg, + unsigned long *user_call_ID, + enum rxrpc_command *command, + u32 *abort_code) +{ + struct cmsghdr *cmsg; + bool got_user_ID = false; + int len; + + *command = RXRPC_CMD_SEND_DATA; + + if (msg->msg_controllen == 0) + return -EINVAL; + + for_each_cmsghdr(cmsg, msg) { + if (!CMSG_OK(msg, cmsg)) + return -EINVAL; + + len = cmsg->cmsg_len - CMSG_ALIGN(sizeof(struct cmsghdr)); + _debug("CMSG %d, %d, %d", + cmsg->cmsg_level, cmsg->cmsg_type, len); + + if (cmsg->cmsg_level != SOL_RXRPC) + continue; + + switch (cmsg->cmsg_type) { + case RXRPC_USER_CALL_ID: + if (msg->msg_flags & MSG_CMSG_COMPAT) { + if (len != sizeof(u32)) + return -EINVAL; + *user_call_ID = *(u32 *) CMSG_DATA(cmsg); + } else { + if (len != sizeof(unsigned long)) + return -EINVAL; + *user_call_ID = *(unsigned long *) + CMSG_DATA(cmsg); + } + _debug("User Call ID %lx", *user_call_ID); + got_user_ID = true; + break; + + case RXRPC_ABORT: + if (*command != RXRPC_CMD_SEND_DATA) + return -EINVAL; + *command = RXRPC_CMD_SEND_ABORT; + if (len != sizeof(*abort_code)) + return -EINVAL; + *abort_code = *(unsigned int *) CMSG_DATA(cmsg); + _debug("Abort %x", *abort_code); + if (*abort_code == 0) + return -EINVAL; + break; + + case RXRPC_ACCEPT: + if (*command != RXRPC_CMD_SEND_DATA) + return -EINVAL; + *command = RXRPC_CMD_ACCEPT; + if (len != 0) + return -EINVAL; + break; + + default: + return -EINVAL; + } + } + + if (!got_user_ID) + return -EINVAL; + _leave(" = 0"); + return 0; +} + +/* + * abort a call, sending an ABORT packet to the peer + */ +static void rxrpc_send_abort(struct rxrpc_call *call, u32 abort_code) +{ + write_lock_bh(&call->state_lock); + + if (call->state <= RXRPC_CALL_COMPLETE) { + call->state = RXRPC_CALL_LOCALLY_ABORTED; + call->local_abort = abort_code; + set_bit(RXRPC_CALL_EV_ABORT, &call->events); + del_timer_sync(&call->resend_timer); + del_timer_sync(&call->ack_timer); + clear_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events); + clear_bit(RXRPC_CALL_EV_ACK, &call->events); + clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); + rxrpc_queue_call(call); + } + + write_unlock_bh(&call->state_lock); +} + +/* + * Create a new client call for sendmsg(). + */ +static struct rxrpc_call * +rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, + unsigned long user_call_ID) +{ + struct rxrpc_conn_bundle *bundle; + struct rxrpc_transport *trans; + struct rxrpc_call *call; + struct key *key; + long ret; + + DECLARE_SOCKADDR(struct sockaddr_rxrpc *, srx, msg->msg_name); + + _enter(""); + + if (!msg->msg_name) + return ERR_PTR(-EDESTADDRREQ); + + trans = rxrpc_name_to_transport(rx, msg->msg_name, msg->msg_namelen, 0, + GFP_KERNEL); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + goto out; + } + + key = rx->key; + if (key && !rx->key->payload.data[0]) + key = NULL; + bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, GFP_KERNEL); + if (IS_ERR(bundle)) { + ret = PTR_ERR(bundle); + goto out_trans; + } + + call = rxrpc_new_client_call(rx, trans, bundle, user_call_ID, + GFP_KERNEL); + rxrpc_put_bundle(trans, bundle); + rxrpc_put_transport(trans); + if (IS_ERR(call)) { + ret = PTR_ERR(call); + goto out_trans; + } + + _leave(" = %p\n", call); + return call; + +out_trans: + rxrpc_put_transport(trans); +out: + _leave(" = %ld", ret); + return ERR_PTR(ret); +} + +/* + * send a message forming part of a client call through an RxRPC socket + * - caller holds the socket locked + * - the socket may be either a client socket or a server socket + */ +int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) +{ + enum rxrpc_command cmd; + struct rxrpc_call *call; + unsigned long user_call_ID = 0; + u32 abort_code = 0; + int ret; + + _enter(""); + + ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code); + if (ret < 0) + return ret; + + if (cmd == RXRPC_CMD_ACCEPT) { + if (rx->sk.sk_state != RXRPC_SERVER_LISTENING) + return -EINVAL; + call = rxrpc_accept_call(rx, user_call_ID); + if (IS_ERR(call)) + return PTR_ERR(call); + rxrpc_put_call(call); + return 0; + } + + call = rxrpc_find_call_by_user_ID(rx, user_call_ID); + if (!call) { + if (cmd != RXRPC_CMD_SEND_DATA) + return -EBADSLT; + call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID); + if (IS_ERR(call)) + return PTR_ERR(call); + } + + _debug("CALL %d USR %lx ST %d on CONN %p", + call->debug_id, call->user_call_ID, call->state, call->conn); + + if (call->state >= RXRPC_CALL_COMPLETE) { + /* it's too late for this call */ + ret = -ECONNRESET; + } else if (cmd == RXRPC_CMD_SEND_ABORT) { + rxrpc_send_abort(call, abort_code); + ret = 0; + } else if (cmd != RXRPC_CMD_SEND_DATA) { + ret = -EINVAL; + } else if (!call->in_clientflag && + call->state != RXRPC_CALL_CLIENT_SEND_REQUEST) { + /* request phase complete for this client call */ + ret = -EPROTO; + } else if (call->in_clientflag && + call->state != RXRPC_CALL_SERVER_ACK_REQUEST && + call->state != RXRPC_CALL_SERVER_SEND_REPLY) { + /* Reply phase not begun or not complete for service call. */ + ret = -EPROTO; + } else { + ret = rxrpc_send_data(rx, call, msg, len); + } + + rxrpc_put_call(call); + _leave(" = %d", ret); + return ret; +} + +/** + * rxrpc_kernel_send_data - Allow a kernel service to send data on a call + * @call: The call to send data through + * @msg: The data to send + * @len: The amount of data to send + * + * Allow a kernel service to send data on a call. The call must be in an state + * appropriate to sending data. No control data should be supplied in @msg, + * nor should an address be supplied. MSG_MORE should be flagged if there's + * more data to come, otherwise this data will end the transmission phase. + */ +int rxrpc_kernel_send_data(struct rxrpc_call *call, struct msghdr *msg, + size_t len) +{ + int ret; + + _enter("{%d,%s},", call->debug_id, rxrpc_call_states[call->state]); + + ASSERTCMP(msg->msg_name, ==, NULL); + ASSERTCMP(msg->msg_control, ==, NULL); + + lock_sock(&call->socket->sk); + + _debug("CALL %d USR %lx ST %d on CONN %p", + call->debug_id, call->user_call_ID, call->state, call->conn); + + if (call->state >= RXRPC_CALL_COMPLETE) { + ret = -ESHUTDOWN; /* it's too late for this call */ + } else if (call->state != RXRPC_CALL_CLIENT_SEND_REQUEST && + call->state != RXRPC_CALL_SERVER_ACK_REQUEST && + call->state != RXRPC_CALL_SERVER_SEND_REPLY) { + ret = -EPROTO; /* request phase complete for this client call */ + } else { + ret = rxrpc_send_data(call->socket, call, msg, len); + } + + release_sock(&call->socket->sk); + _leave(" = %d", ret); + return ret; +} + +EXPORT_SYMBOL(rxrpc_kernel_send_data); + +/** + * rxrpc_kernel_abort_call - Allow a kernel service to abort a call + * @call: The call to be aborted + * @abort_code: The abort code to stick into the ABORT packet + * + * Allow a kernel service to abort a call, if it's still in an abortable state. + */ +void rxrpc_kernel_abort_call(struct rxrpc_call *call, u32 abort_code) +{ + _enter("{%d},%d", call->debug_id, abort_code); + + lock_sock(&call->socket->sk); + + _debug("CALL %d USR %lx ST %d on CONN %p", + call->debug_id, call->user_call_ID, call->state, call->conn); + + if (call->state < RXRPC_CALL_COMPLETE) + rxrpc_send_abort(call, abort_code); + + release_sock(&call->socket->sk); + _leave(""); +} + +EXPORT_SYMBOL(rxrpc_kernel_abort_call); + +/* + * send a packet through the transport endpoint + */ +int rxrpc_send_packet(struct rxrpc_transport *trans, struct sk_buff *skb) +{ + struct kvec iov[1]; + struct msghdr msg; + int ret, opt; + + _enter(",{%d}", skb->len); + + iov[0].iov_base = skb->head; + iov[0].iov_len = skb->len; + + msg.msg_name = &trans->peer->srx.transport.sin; + msg.msg_namelen = sizeof(trans->peer->srx.transport.sin); + msg.msg_control = NULL; + msg.msg_controllen = 0; + msg.msg_flags = 0; + + /* send the packet with the don't fragment bit set if we currently + * think it's small enough */ + if (skb->len - sizeof(struct rxrpc_wire_header) < trans->peer->maxdata) { + down_read(&trans->local->defrag_sem); + /* send the packet by UDP + * - returns -EMSGSIZE if UDP would have to fragment the packet + * to go out of the interface + * - in which case, we'll have processed the ICMP error + * message and update the peer record + */ + ret = kernel_sendmsg(trans->local->socket, &msg, iov, 1, + iov[0].iov_len); + + up_read(&trans->local->defrag_sem); + if (ret == -EMSGSIZE) + goto send_fragmentable; + + _leave(" = %d [%u]", ret, trans->peer->maxdata); + return ret; + } + +send_fragmentable: + /* attempt to send this message with fragmentation enabled */ + _debug("send fragment"); + + down_write(&trans->local->defrag_sem); + opt = IP_PMTUDISC_DONT; + ret = kernel_setsockopt(trans->local->socket, SOL_IP, IP_MTU_DISCOVER, + (char *) &opt, sizeof(opt)); + if (ret == 0) { + ret = kernel_sendmsg(trans->local->socket, &msg, iov, 1, + iov[0].iov_len); + + opt = IP_PMTUDISC_DO; + kernel_setsockopt(trans->local->socket, SOL_IP, + IP_MTU_DISCOVER, (char *) &opt, sizeof(opt)); + } + + up_write(&trans->local->defrag_sem); + _leave(" = %d [frag %u]", ret, trans->peer->maxdata); + return ret; +} + +/* + * wait for space to appear in the transmit/ACK window + * - caller holds the socket locked + */ +static int rxrpc_wait_for_tx_window(struct rxrpc_sock *rx, + struct rxrpc_call *call, + long *timeo) +{ + DECLARE_WAITQUEUE(myself, current); + int ret; + + _enter(",{%d},%ld", + CIRC_SPACE(call->acks_head, ACCESS_ONCE(call->acks_tail), + call->acks_winsz), + *timeo); + + add_wait_queue(&call->tx_waitq, &myself); + + for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + ret = 0; + if (CIRC_SPACE(call->acks_head, ACCESS_ONCE(call->acks_tail), + call->acks_winsz) > 0) + break; + if (signal_pending(current)) { + ret = sock_intr_errno(*timeo); + break; + } + + release_sock(&rx->sk); + *timeo = schedule_timeout(*timeo); + lock_sock(&rx->sk); + } + + remove_wait_queue(&call->tx_waitq, &myself); + set_current_state(TASK_RUNNING); + _leave(" = %d", ret); + return ret; +} + +/* + * attempt to schedule an instant Tx resend + */ +static inline void rxrpc_instant_resend(struct rxrpc_call *call) +{ + read_lock_bh(&call->state_lock); + if (try_to_del_timer_sync(&call->resend_timer) >= 0) { + clear_bit(RXRPC_CALL_RUN_RTIMER, &call->flags); + if (call->state < RXRPC_CALL_COMPLETE && + !test_and_set_bit(RXRPC_CALL_EV_RESEND_TIMER, &call->events)) + rxrpc_queue_call(call); + } + read_unlock_bh(&call->state_lock); +} + +/* + * queue a packet for transmission, set the resend timer and attempt + * to send the packet immediately + */ +static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb, + bool last) +{ + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + int ret; + + _net("queue skb %p [%d]", skb, call->acks_head); + + ASSERT(call->acks_window != NULL); + call->acks_window[call->acks_head] = (unsigned long) skb; + smp_wmb(); + call->acks_head = (call->acks_head + 1) & (call->acks_winsz - 1); + + if (last || call->state == RXRPC_CALL_SERVER_ACK_REQUEST) { + _debug("________awaiting reply/ACK__________"); + write_lock_bh(&call->state_lock); + switch (call->state) { + case RXRPC_CALL_CLIENT_SEND_REQUEST: + call->state = RXRPC_CALL_CLIENT_AWAIT_REPLY; + break; + case RXRPC_CALL_SERVER_ACK_REQUEST: + call->state = RXRPC_CALL_SERVER_SEND_REPLY; + if (!last) + break; + case RXRPC_CALL_SERVER_SEND_REPLY: + call->state = RXRPC_CALL_SERVER_AWAIT_ACK; + break; + default: + break; + } + write_unlock_bh(&call->state_lock); + } + + _proto("Tx DATA %%%u { #%u }", sp->hdr.serial, sp->hdr.seq); + + sp->need_resend = false; + sp->resend_at = jiffies + rxrpc_resend_timeout; + if (!test_and_set_bit(RXRPC_CALL_RUN_RTIMER, &call->flags)) { + _debug("run timer"); + call->resend_timer.expires = sp->resend_at; + add_timer(&call->resend_timer); + } + + /* attempt to cancel the rx-ACK timer, deferring reply transmission if + * we're ACK'ing the request phase of an incoming call */ + ret = -EAGAIN; + if (try_to_del_timer_sync(&call->ack_timer) >= 0) { + /* the packet may be freed by rxrpc_process_call() before this + * returns */ + ret = rxrpc_send_packet(call->conn->trans, skb); + _net("sent skb %p", skb); + } else { + _debug("failed to delete ACK timer"); + } + + if (ret < 0) { + _debug("need instant resend %d", ret); + sp->need_resend = true; + rxrpc_instant_resend(call); + } + + _leave(""); +} + +/* + * Convert a host-endian header into a network-endian header. + */ +static void rxrpc_insert_header(struct sk_buff *skb) +{ + struct rxrpc_wire_header whdr; + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + + whdr.epoch = htonl(sp->hdr.epoch); + whdr.cid = htonl(sp->hdr.cid); + whdr.callNumber = htonl(sp->hdr.callNumber); + whdr.seq = htonl(sp->hdr.seq); + whdr.serial = htonl(sp->hdr.serial); + whdr.type = sp->hdr.type; + whdr.flags = sp->hdr.flags; + whdr.userStatus = sp->hdr.userStatus; + whdr.securityIndex = sp->hdr.securityIndex; + whdr._rsvd = htons(sp->hdr._rsvd); + whdr.serviceId = htons(sp->hdr.serviceId); + + memcpy(skb->head, &whdr, sizeof(whdr)); +} + +/* + * send data through a socket + * - must be called in process context + * - caller holds the socket locked + */ +static int rxrpc_send_data(struct rxrpc_sock *rx, + struct rxrpc_call *call, + struct msghdr *msg, size_t len) +{ + struct rxrpc_skb_priv *sp; + struct sk_buff *skb; + struct sock *sk = &rx->sk; + long timeo; + bool more; + int ret, copied; + + timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT); + + /* this should be in poll */ + sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk); + + if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN)) + return -EPIPE; + + more = msg->msg_flags & MSG_MORE; + + skb = call->tx_pending; + call->tx_pending = NULL; + + copied = 0; + do { + if (!skb) { + size_t size, chunk, max, space; + + _debug("alloc"); + + if (CIRC_SPACE(call->acks_head, + ACCESS_ONCE(call->acks_tail), + call->acks_winsz) <= 0) { + ret = -EAGAIN; + if (msg->msg_flags & MSG_DONTWAIT) + goto maybe_error; + ret = rxrpc_wait_for_tx_window(rx, call, + &timeo); + if (ret < 0) + goto maybe_error; + } + + max = call->conn->trans->peer->maxdata; + max -= call->conn->security_size; + max &= ~(call->conn->size_align - 1UL); + + chunk = max; + if (chunk > msg_data_left(msg) && !more) + chunk = msg_data_left(msg); + + space = chunk + call->conn->size_align; + space &= ~(call->conn->size_align - 1UL); + + size = space + call->conn->header_size; + + _debug("SIZE: %zu/%zu/%zu", chunk, space, size); + + /* create a buffer that we can retain until it's ACK'd */ + skb = sock_alloc_send_skb( + sk, size, msg->msg_flags & MSG_DONTWAIT, &ret); + if (!skb) + goto maybe_error; + + rxrpc_new_skb(skb); + + _debug("ALLOC SEND %p", skb); + + ASSERTCMP(skb->mark, ==, 0); + + _debug("HS: %u", call->conn->header_size); + skb_reserve(skb, call->conn->header_size); + skb->len += call->conn->header_size; + + sp = rxrpc_skb(skb); + sp->remain = chunk; + if (sp->remain > skb_tailroom(skb)) + sp->remain = skb_tailroom(skb); + + _net("skb: hr %d, tr %d, hl %d, rm %d", + skb_headroom(skb), + skb_tailroom(skb), + skb_headlen(skb), + sp->remain); + + skb->ip_summed = CHECKSUM_UNNECESSARY; + } + + _debug("append"); + sp = rxrpc_skb(skb); + + /* append next segment of data to the current buffer */ + if (msg_data_left(msg) > 0) { + int copy = skb_tailroom(skb); + ASSERTCMP(copy, >, 0); + if (copy > msg_data_left(msg)) + copy = msg_data_left(msg); + if (copy > sp->remain) + copy = sp->remain; + + _debug("add"); + ret = skb_add_data(skb, &msg->msg_iter, copy); + _debug("added"); + if (ret < 0) + goto efault; + sp->remain -= copy; + skb->mark += copy; + copied += copy; + } + + /* check for the far side aborting the call or a network error + * occurring */ + if (call->state > RXRPC_CALL_COMPLETE) + goto call_aborted; + + /* add the packet to the send queue if it's now full */ + if (sp->remain <= 0 || + (msg_data_left(msg) == 0 && !more)) { + struct rxrpc_connection *conn = call->conn; + uint32_t seq; + size_t pad; + + /* pad out if we're using security */ + if (conn->security_ix) { + pad = conn->security_size + skb->mark; + pad = conn->size_align - pad; + pad &= conn->size_align - 1; + _debug("pad %zu", pad); + if (pad) + memset(skb_put(skb, pad), 0, pad); + } + + seq = atomic_inc_return(&call->sequence); + + sp->hdr.epoch = conn->epoch; + sp->hdr.cid = call->cid; + sp->hdr.callNumber = call->call_id; + sp->hdr.seq = seq; + sp->hdr.serial = atomic_inc_return(&conn->serial); + sp->hdr.type = RXRPC_PACKET_TYPE_DATA; + sp->hdr.userStatus = 0; + sp->hdr.securityIndex = conn->security_ix; + sp->hdr._rsvd = 0; + sp->hdr.serviceId = call->service_id; + + sp->hdr.flags = conn->out_clientflag; + if (msg_data_left(msg) == 0 && !more) + sp->hdr.flags |= RXRPC_LAST_PACKET; + else if (CIRC_SPACE(call->acks_head, + ACCESS_ONCE(call->acks_tail), + call->acks_winsz) > 1) + sp->hdr.flags |= RXRPC_MORE_PACKETS; + if (more && seq & 1) + sp->hdr.flags |= RXRPC_REQUEST_ACK; + + ret = conn->security->secure_packet( + call, skb, skb->mark, + skb->head + sizeof(struct rxrpc_wire_header)); + if (ret < 0) + goto out; + + rxrpc_insert_header(skb); + rxrpc_queue_packet(call, skb, !msg_data_left(msg) && !more); + skb = NULL; + } + } while (msg_data_left(msg) > 0); + +success: + ret = copied; +out: + call->tx_pending = skb; + _leave(" = %d", ret); + return ret; + +call_aborted: + rxrpc_free_skb(skb); + if (call->state == RXRPC_CALL_NETWORK_ERROR) + ret = call->conn->trans->peer->net_error; + else + ret = -ECONNABORTED; + _leave(" = %d", ret); + return ret; + +maybe_error: + if (copied) + goto success; + goto out; + +efault: + ret = -EFAULT; + goto out; +} diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c new file mode 100644 index 0000000..3e82d6f --- /dev/null +++ b/net/rxrpc/peer_event.c @@ -0,0 +1,230 @@ +/* Error message handling (ICMP) + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * handle an error received on the local endpoint + */ +void rxrpc_UDP_error_report(struct sock *sk) +{ + struct sock_exterr_skb *serr; + struct rxrpc_transport *trans; + struct rxrpc_local *local = sk->sk_user_data; + struct rxrpc_peer *peer; + struct sk_buff *skb; + __be32 addr; + __be16 port; + + _enter("%p{%d}", sk, local->debug_id); + + skb = sock_dequeue_err_skb(sk); + if (!skb) { + _leave("UDP socket errqueue empty"); + return; + } + serr = SKB_EXT_ERR(skb); + if (!skb->len && serr->ee.ee_origin == SO_EE_ORIGIN_TIMESTAMPING) { + _leave("UDP empty message"); + kfree_skb(skb); + return; + } + + rxrpc_new_skb(skb); + + addr = *(__be32 *)(skb_network_header(skb) + serr->addr_offset); + port = serr->port; + + _net("Rx UDP Error from %pI4:%hu", &addr, ntohs(port)); + _debug("Msg l:%d d:%d", skb->len, skb->data_len); + + peer = rxrpc_find_peer(local, addr, port); + if (IS_ERR(peer)) { + rxrpc_free_skb(skb); + _leave(" [no peer]"); + return; + } + + trans = rxrpc_find_transport(local, peer); + if (!trans) { + rxrpc_put_peer(peer); + rxrpc_free_skb(skb); + _leave(" [no trans]"); + return; + } + + if (serr->ee.ee_origin == SO_EE_ORIGIN_ICMP && + serr->ee.ee_type == ICMP_DEST_UNREACH && + serr->ee.ee_code == ICMP_FRAG_NEEDED + ) { + u32 mtu = serr->ee.ee_info; + + _net("Rx Received ICMP Fragmentation Needed (%d)", mtu); + + /* wind down the local interface MTU */ + if (mtu > 0 && peer->if_mtu == 65535 && mtu < peer->if_mtu) { + peer->if_mtu = mtu; + _net("I/F MTU %u", mtu); + } + + if (mtu == 0) { + /* they didn't give us a size, estimate one */ + mtu = peer->if_mtu; + if (mtu > 1500) { + mtu >>= 1; + if (mtu < 1500) + mtu = 1500; + } else { + mtu -= 100; + if (mtu < peer->hdrsize) + mtu = peer->hdrsize + 4; + } + } + + if (mtu < peer->mtu) { + spin_lock_bh(&peer->lock); + peer->mtu = mtu; + peer->maxdata = peer->mtu - peer->hdrsize; + spin_unlock_bh(&peer->lock); + _net("Net MTU %u (maxdata %u)", + peer->mtu, peer->maxdata); + } + } + + rxrpc_put_peer(peer); + + /* pass the transport ref to error_handler to release */ + skb_queue_tail(&trans->error_queue, skb); + rxrpc_queue_work(&trans->error_handler); + _leave(""); +} + +/* + * deal with UDP error messages + */ +void rxrpc_UDP_error_handler(struct work_struct *work) +{ + struct sock_extended_err *ee; + struct sock_exterr_skb *serr; + struct rxrpc_transport *trans = + container_of(work, struct rxrpc_transport, error_handler); + struct sk_buff *skb; + int err; + + _enter(""); + + skb = skb_dequeue(&trans->error_queue); + if (!skb) + return; + + serr = SKB_EXT_ERR(skb); + ee = &serr->ee; + + _net("Rx Error o=%d t=%d c=%d e=%d", + ee->ee_origin, ee->ee_type, ee->ee_code, ee->ee_errno); + + err = ee->ee_errno; + + switch (ee->ee_origin) { + case SO_EE_ORIGIN_ICMP: + switch (ee->ee_type) { + case ICMP_DEST_UNREACH: + switch (ee->ee_code) { + case ICMP_NET_UNREACH: + _net("Rx Received ICMP Network Unreachable"); + break; + case ICMP_HOST_UNREACH: + _net("Rx Received ICMP Host Unreachable"); + break; + case ICMP_PORT_UNREACH: + _net("Rx Received ICMP Port Unreachable"); + break; + case ICMP_NET_UNKNOWN: + _net("Rx Received ICMP Unknown Network"); + break; + case ICMP_HOST_UNKNOWN: + _net("Rx Received ICMP Unknown Host"); + break; + default: + _net("Rx Received ICMP DestUnreach code=%u", + ee->ee_code); + break; + } + break; + + case ICMP_TIME_EXCEEDED: + _net("Rx Received ICMP TTL Exceeded"); + break; + + default: + _proto("Rx Received ICMP error { type=%u code=%u }", + ee->ee_type, ee->ee_code); + break; + } + break; + + case SO_EE_ORIGIN_LOCAL: + _proto("Rx Received local error { error=%d }", + ee->ee_errno); + break; + + case SO_EE_ORIGIN_NONE: + case SO_EE_ORIGIN_ICMP6: + default: + _proto("Rx Received error report { orig=%u }", + ee->ee_origin); + break; + } + + /* terminate all the affected calls if there's an unrecoverable + * error */ + if (err) { + struct rxrpc_call *call, *_n; + + _debug("ISSUE ERROR %d", err); + + spin_lock_bh(&trans->peer->lock); + trans->peer->net_error = err; + + list_for_each_entry_safe(call, _n, &trans->peer->error_targets, + error_link) { + write_lock(&call->state_lock); + if (call->state != RXRPC_CALL_COMPLETE && + call->state < RXRPC_CALL_NETWORK_ERROR) { + call->state = RXRPC_CALL_NETWORK_ERROR; + set_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events); + rxrpc_queue_call(call); + } + write_unlock(&call->state_lock); + list_del_init(&call->error_link); + } + + spin_unlock_bh(&trans->peer->lock); + } + + if (!skb_queue_empty(&trans->error_queue)) + rxrpc_queue_work(&trans->error_handler); + + rxrpc_free_skb(skb); + rxrpc_put_transport(trans); + _leave(""); +} diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c new file mode 100644 index 0000000..0b54cda --- /dev/null +++ b/net/rxrpc/peer_object.c @@ -0,0 +1,305 @@ +/* RxRPC remote transport endpoint management + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +static LIST_HEAD(rxrpc_peers); +static DEFINE_RWLOCK(rxrpc_peer_lock); +static DECLARE_WAIT_QUEUE_HEAD(rxrpc_peer_wq); + +static void rxrpc_destroy_peer(struct work_struct *work); + +/* + * assess the MTU size for the network interface through which this peer is + * reached + */ +static void rxrpc_assess_MTU_size(struct rxrpc_peer *peer) +{ + struct rtable *rt; + struct flowi4 fl4; + + peer->if_mtu = 1500; + + rt = ip_route_output_ports(&init_net, &fl4, NULL, + peer->srx.transport.sin.sin_addr.s_addr, 0, + htons(7000), htons(7001), + IPPROTO_UDP, 0, 0); + if (IS_ERR(rt)) { + _leave(" [route err %ld]", PTR_ERR(rt)); + return; + } + + peer->if_mtu = dst_mtu(&rt->dst); + dst_release(&rt->dst); + + _leave(" [if_mtu %u]", peer->if_mtu); +} + +/* + * allocate a new peer + */ +static struct rxrpc_peer *rxrpc_alloc_peer(struct sockaddr_rxrpc *srx, + gfp_t gfp) +{ + struct rxrpc_peer *peer; + + _enter(""); + + peer = kzalloc(sizeof(struct rxrpc_peer), gfp); + if (peer) { + INIT_WORK(&peer->destroyer, &rxrpc_destroy_peer); + INIT_LIST_HEAD(&peer->link); + INIT_LIST_HEAD(&peer->error_targets); + spin_lock_init(&peer->lock); + atomic_set(&peer->usage, 1); + peer->debug_id = atomic_inc_return(&rxrpc_debug_id); + memcpy(&peer->srx, srx, sizeof(*srx)); + + rxrpc_assess_MTU_size(peer); + peer->mtu = peer->if_mtu; + + if (srx->transport.family == AF_INET) { + peer->hdrsize = sizeof(struct iphdr); + switch (srx->transport_type) { + case SOCK_DGRAM: + peer->hdrsize += sizeof(struct udphdr); + break; + default: + BUG(); + break; + } + } else { + BUG(); + } + + peer->hdrsize += sizeof(struct rxrpc_wire_header); + peer->maxdata = peer->mtu - peer->hdrsize; + } + + _leave(" = %p", peer); + return peer; +} + +/* + * obtain a remote transport endpoint for the specified address + */ +struct rxrpc_peer *rxrpc_get_peer(struct sockaddr_rxrpc *srx, gfp_t gfp) +{ + struct rxrpc_peer *peer, *candidate; + const char *new = "old"; + int usage; + + _enter("{%d,%d,%pI4+%hu}", + srx->transport_type, + srx->transport_len, + &srx->transport.sin.sin_addr, + ntohs(srx->transport.sin.sin_port)); + + /* search the peer list first */ + read_lock_bh(&rxrpc_peer_lock); + list_for_each_entry(peer, &rxrpc_peers, link) { + _debug("check PEER %d { u=%d t=%d l=%d }", + peer->debug_id, + atomic_read(&peer->usage), + peer->srx.transport_type, + peer->srx.transport_len); + + if (atomic_read(&peer->usage) > 0 && + peer->srx.transport_type == srx->transport_type && + peer->srx.transport_len == srx->transport_len && + memcmp(&peer->srx.transport, + &srx->transport, + srx->transport_len) == 0) + goto found_extant_peer; + } + read_unlock_bh(&rxrpc_peer_lock); + + /* not yet present - create a candidate for a new record and then + * redo the search */ + candidate = rxrpc_alloc_peer(srx, gfp); + if (!candidate) { + _leave(" = -ENOMEM"); + return ERR_PTR(-ENOMEM); + } + + write_lock_bh(&rxrpc_peer_lock); + + list_for_each_entry(peer, &rxrpc_peers, link) { + if (atomic_read(&peer->usage) > 0 && + peer->srx.transport_type == srx->transport_type && + peer->srx.transport_len == srx->transport_len && + memcmp(&peer->srx.transport, + &srx->transport, + srx->transport_len) == 0) + goto found_extant_second; + } + + /* we can now add the new candidate to the list */ + peer = candidate; + candidate = NULL; + usage = atomic_read(&peer->usage); + + list_add_tail(&peer->link, &rxrpc_peers); + write_unlock_bh(&rxrpc_peer_lock); + new = "new"; + +success: + _net("PEER %s %d {%d,%u,%pI4+%hu}", + new, + peer->debug_id, + peer->srx.transport_type, + peer->srx.transport.family, + &peer->srx.transport.sin.sin_addr, + ntohs(peer->srx.transport.sin.sin_port)); + + _leave(" = %p {u=%d}", peer, usage); + return peer; + + /* we found the peer in the list immediately */ +found_extant_peer: + usage = atomic_inc_return(&peer->usage); + read_unlock_bh(&rxrpc_peer_lock); + goto success; + + /* we found the peer on the second time through the list */ +found_extant_second: + usage = atomic_inc_return(&peer->usage); + write_unlock_bh(&rxrpc_peer_lock); + kfree(candidate); + goto success; +} + +/* + * find the peer associated with a packet + */ +struct rxrpc_peer *rxrpc_find_peer(struct rxrpc_local *local, + __be32 addr, __be16 port) +{ + struct rxrpc_peer *peer; + + _enter(""); + + /* search the peer list */ + read_lock_bh(&rxrpc_peer_lock); + + if (local->srx.transport.family == AF_INET && + local->srx.transport_type == SOCK_DGRAM + ) { + list_for_each_entry(peer, &rxrpc_peers, link) { + if (atomic_read(&peer->usage) > 0 && + peer->srx.transport_type == SOCK_DGRAM && + peer->srx.transport.family == AF_INET && + peer->srx.transport.sin.sin_port == port && + peer->srx.transport.sin.sin_addr.s_addr == addr) + goto found_UDP_peer; + } + + goto new_UDP_peer; + } + + read_unlock_bh(&rxrpc_peer_lock); + _leave(" = -EAFNOSUPPORT"); + return ERR_PTR(-EAFNOSUPPORT); + +found_UDP_peer: + _net("Rx UDP DGRAM from peer %d", peer->debug_id); + atomic_inc(&peer->usage); + read_unlock_bh(&rxrpc_peer_lock); + _leave(" = %p", peer); + return peer; + +new_UDP_peer: + _net("Rx UDP DGRAM from NEW peer"); + read_unlock_bh(&rxrpc_peer_lock); + _leave(" = -EBUSY [new]"); + return ERR_PTR(-EBUSY); +} + +/* + * release a remote transport endpoint + */ +void rxrpc_put_peer(struct rxrpc_peer *peer) +{ + _enter("%p{u=%d}", peer, atomic_read(&peer->usage)); + + ASSERTCMP(atomic_read(&peer->usage), >, 0); + + if (likely(!atomic_dec_and_test(&peer->usage))) { + _leave(" [in use]"); + return; + } + + rxrpc_queue_work(&peer->destroyer); + _leave(""); +} + +/* + * destroy a remote transport endpoint + */ +static void rxrpc_destroy_peer(struct work_struct *work) +{ + struct rxrpc_peer *peer = + container_of(work, struct rxrpc_peer, destroyer); + + _enter("%p{%d}", peer, atomic_read(&peer->usage)); + + write_lock_bh(&rxrpc_peer_lock); + list_del(&peer->link); + write_unlock_bh(&rxrpc_peer_lock); + + _net("DESTROY PEER %d", peer->debug_id); + kfree(peer); + + if (list_empty(&rxrpc_peers)) + wake_up_all(&rxrpc_peer_wq); + _leave(""); +} + +/* + * preemptively destroy all the peer records from a transport endpoint rather + * than waiting for them to time out + */ +void __exit rxrpc_destroy_all_peers(void) +{ + DECLARE_WAITQUEUE(myself,current); + + _enter(""); + + /* we simply have to wait for them to go away */ + if (!list_empty(&rxrpc_peers)) { + set_current_state(TASK_UNINTERRUPTIBLE); + add_wait_queue(&rxrpc_peer_wq, &myself); + + while (!list_empty(&rxrpc_peers)) { + schedule(); + set_current_state(TASK_UNINTERRUPTIBLE); + } + + remove_wait_queue(&rxrpc_peer_wq, &myself); + set_current_state(TASK_RUNNING); + } + + _leave(""); +} diff --git a/net/rxrpc/proc.c b/net/rxrpc/proc.c new file mode 100644 index 0000000..225163b --- /dev/null +++ b/net/rxrpc/proc.c @@ -0,0 +1,192 @@ +/* /proc/net/ support for AF_RXRPC + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include "ar-internal.h" + +static const char *const rxrpc_conn_states[] = { + [RXRPC_CONN_UNUSED] = "Unused ", + [RXRPC_CONN_CLIENT] = "Client ", + [RXRPC_CONN_SERVER_UNSECURED] = "SvUnsec ", + [RXRPC_CONN_SERVER_CHALLENGING] = "SvChall ", + [RXRPC_CONN_SERVER] = "SvSecure", + [RXRPC_CONN_REMOTELY_ABORTED] = "RmtAbort", + [RXRPC_CONN_LOCALLY_ABORTED] = "LocAbort", + [RXRPC_CONN_NETWORK_ERROR] = "NetError", +}; + +/* + * generate a list of extant and dead calls in /proc/net/rxrpc_calls + */ +static void *rxrpc_call_seq_start(struct seq_file *seq, loff_t *_pos) +{ + read_lock(&rxrpc_call_lock); + return seq_list_start_head(&rxrpc_calls, *_pos); +} + +static void *rxrpc_call_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + return seq_list_next(v, &rxrpc_calls, pos); +} + +static void rxrpc_call_seq_stop(struct seq_file *seq, void *v) +{ + read_unlock(&rxrpc_call_lock); +} + +static int rxrpc_call_seq_show(struct seq_file *seq, void *v) +{ + struct rxrpc_transport *trans; + struct rxrpc_call *call; + char lbuff[4 + 4 + 4 + 4 + 5 + 1], rbuff[4 + 4 + 4 + 4 + 5 + 1]; + + if (v == &rxrpc_calls) { + seq_puts(seq, + "Proto Local Remote " + " SvID ConnID CallID End Use State Abort " + " UserID\n"); + return 0; + } + + call = list_entry(v, struct rxrpc_call, link); + trans = call->conn->trans; + + sprintf(lbuff, "%pI4:%u", + &trans->local->srx.transport.sin.sin_addr, + ntohs(trans->local->srx.transport.sin.sin_port)); + + sprintf(rbuff, "%pI4:%u", + &trans->peer->srx.transport.sin.sin_addr, + ntohs(trans->peer->srx.transport.sin.sin_port)); + + seq_printf(seq, + "UDP %-22.22s %-22.22s %4x %08x %08x %s %3u" + " %-8.8s %08x %lx\n", + lbuff, + rbuff, + call->conn->service_id, + call->cid, + call->call_id, + call->conn->in_clientflag ? "Svc" : "Clt", + atomic_read(&call->usage), + rxrpc_call_states[call->state], + call->remote_abort ?: call->local_abort, + call->user_call_ID); + + return 0; +} + +static const struct seq_operations rxrpc_call_seq_ops = { + .start = rxrpc_call_seq_start, + .next = rxrpc_call_seq_next, + .stop = rxrpc_call_seq_stop, + .show = rxrpc_call_seq_show, +}; + +static int rxrpc_call_seq_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &rxrpc_call_seq_ops); +} + +const struct file_operations rxrpc_call_seq_fops = { + .owner = THIS_MODULE, + .open = rxrpc_call_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +/* + * generate a list of extant virtual connections in /proc/net/rxrpc_conns + */ +static void *rxrpc_connection_seq_start(struct seq_file *seq, loff_t *_pos) +{ + read_lock(&rxrpc_connection_lock); + return seq_list_start_head(&rxrpc_connections, *_pos); +} + +static void *rxrpc_connection_seq_next(struct seq_file *seq, void *v, + loff_t *pos) +{ + return seq_list_next(v, &rxrpc_connections, pos); +} + +static void rxrpc_connection_seq_stop(struct seq_file *seq, void *v) +{ + read_unlock(&rxrpc_connection_lock); +} + +static int rxrpc_connection_seq_show(struct seq_file *seq, void *v) +{ + struct rxrpc_connection *conn; + struct rxrpc_transport *trans; + char lbuff[4 + 4 + 4 + 4 + 5 + 1], rbuff[4 + 4 + 4 + 4 + 5 + 1]; + + if (v == &rxrpc_connections) { + seq_puts(seq, + "Proto Local Remote " + " SvID ConnID Calls End Use State Key " + " Serial ISerial\n" + ); + return 0; + } + + conn = list_entry(v, struct rxrpc_connection, link); + trans = conn->trans; + + sprintf(lbuff, "%pI4:%u", + &trans->local->srx.transport.sin.sin_addr, + ntohs(trans->local->srx.transport.sin.sin_port)); + + sprintf(rbuff, "%pI4:%u", + &trans->peer->srx.transport.sin.sin_addr, + ntohs(trans->peer->srx.transport.sin.sin_port)); + + seq_printf(seq, + "UDP %-22.22s %-22.22s %4x %08x %08x %s %3u" + " %s %08x %08x %08x\n", + lbuff, + rbuff, + conn->service_id, + conn->cid, + conn->call_counter, + conn->in_clientflag ? "Svc" : "Clt", + atomic_read(&conn->usage), + rxrpc_conn_states[conn->state], + key_serial(conn->key), + atomic_read(&conn->serial), + atomic_read(&conn->hi_serial)); + + return 0; +} + +static const struct seq_operations rxrpc_connection_seq_ops = { + .start = rxrpc_connection_seq_start, + .next = rxrpc_connection_seq_next, + .stop = rxrpc_connection_seq_stop, + .show = rxrpc_connection_seq_show, +}; + + +static int rxrpc_connection_seq_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &rxrpc_connection_seq_ops); +} + +const struct file_operations rxrpc_connection_seq_fops = { + .owner = THIS_MODULE, + .open = rxrpc_connection_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c new file mode 100644 index 0000000..59706b9 --- /dev/null +++ b/net/rxrpc/recvmsg.c @@ -0,0 +1,436 @@ +/* RxRPC recvmsg() implementation + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * removal a call's user ID from the socket tree to make the user ID available + * again and so that it won't be seen again in association with that call + */ +void rxrpc_remove_user_ID(struct rxrpc_sock *rx, struct rxrpc_call *call) +{ + _debug("RELEASE CALL %d", call->debug_id); + + if (test_bit(RXRPC_CALL_HAS_USERID, &call->flags)) { + write_lock_bh(&rx->call_lock); + rb_erase(&call->sock_node, &call->socket->calls); + clear_bit(RXRPC_CALL_HAS_USERID, &call->flags); + write_unlock_bh(&rx->call_lock); + } + + read_lock_bh(&call->state_lock); + if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && + !test_and_set_bit(RXRPC_CALL_EV_RELEASE, &call->events)) + rxrpc_queue_call(call); + read_unlock_bh(&call->state_lock); +} + +/* + * receive a message from an RxRPC socket + * - we need to be careful about two or more threads calling recvmsg + * simultaneously + */ +int rxrpc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, + int flags) +{ + struct rxrpc_skb_priv *sp; + struct rxrpc_call *call = NULL, *continue_call = NULL; + struct rxrpc_sock *rx = rxrpc_sk(sock->sk); + struct sk_buff *skb; + long timeo; + int copy, ret, ullen, offset, copied = 0; + u32 abort_code; + + DEFINE_WAIT(wait); + + _enter(",,,%zu,%d", len, flags); + + if (flags & (MSG_OOB | MSG_TRUNC)) + return -EOPNOTSUPP; + + ullen = msg->msg_flags & MSG_CMSG_COMPAT ? 4 : sizeof(unsigned long); + + timeo = sock_rcvtimeo(&rx->sk, flags & MSG_DONTWAIT); + msg->msg_flags |= MSG_MORE; + + lock_sock(&rx->sk); + + for (;;) { + /* return immediately if a client socket has no outstanding + * calls */ + if (RB_EMPTY_ROOT(&rx->calls)) { + if (copied) + goto out; + if (rx->sk.sk_state != RXRPC_SERVER_LISTENING) { + release_sock(&rx->sk); + if (continue_call) + rxrpc_put_call(continue_call); + return -ENODATA; + } + } + + /* get the next message on the Rx queue */ + skb = skb_peek(&rx->sk.sk_receive_queue); + if (!skb) { + /* nothing remains on the queue */ + if (copied && + (flags & MSG_PEEK || timeo == 0)) + goto out; + + /* wait for a message to turn up */ + release_sock(&rx->sk); + prepare_to_wait_exclusive(sk_sleep(&rx->sk), &wait, + TASK_INTERRUPTIBLE); + ret = sock_error(&rx->sk); + if (ret) + goto wait_error; + + if (skb_queue_empty(&rx->sk.sk_receive_queue)) { + if (signal_pending(current)) + goto wait_interrupted; + timeo = schedule_timeout(timeo); + } + finish_wait(sk_sleep(&rx->sk), &wait); + lock_sock(&rx->sk); + continue; + } + + peek_next_packet: + sp = rxrpc_skb(skb); + call = sp->call; + ASSERT(call != NULL); + + _debug("next pkt %s", rxrpc_pkts[sp->hdr.type]); + + /* make sure we wait for the state to be updated in this call */ + spin_lock_bh(&call->lock); + spin_unlock_bh(&call->lock); + + if (test_bit(RXRPC_CALL_RELEASED, &call->flags)) { + _debug("packet from released call"); + if (skb_dequeue(&rx->sk.sk_receive_queue) != skb) + BUG(); + rxrpc_free_skb(skb); + continue; + } + + /* determine whether to continue last data receive */ + if (continue_call) { + _debug("maybe cont"); + if (call != continue_call || + skb->mark != RXRPC_SKB_MARK_DATA) { + release_sock(&rx->sk); + rxrpc_put_call(continue_call); + _leave(" = %d [noncont]", copied); + return copied; + } + } + + rxrpc_get_call(call); + + /* copy the peer address and timestamp */ + if (!continue_call) { + if (msg->msg_name) { + size_t len = + sizeof(call->conn->trans->peer->srx); + memcpy(msg->msg_name, + &call->conn->trans->peer->srx, len); + msg->msg_namelen = len; + } + sock_recv_timestamp(msg, &rx->sk, skb); + } + + /* receive the message */ + if (skb->mark != RXRPC_SKB_MARK_DATA) + goto receive_non_data_message; + + _debug("recvmsg DATA #%u { %d, %d }", + sp->hdr.seq, skb->len, sp->offset); + + if (!continue_call) { + /* only set the control data once per recvmsg() */ + ret = put_cmsg(msg, SOL_RXRPC, RXRPC_USER_CALL_ID, + ullen, &call->user_call_ID); + if (ret < 0) + goto copy_error; + ASSERT(test_bit(RXRPC_CALL_HAS_USERID, &call->flags)); + } + + ASSERTCMP(sp->hdr.seq, >=, call->rx_data_recv); + ASSERTCMP(sp->hdr.seq, <=, call->rx_data_recv + 1); + call->rx_data_recv = sp->hdr.seq; + + ASSERTCMP(sp->hdr.seq, >, call->rx_data_eaten); + + offset = sp->offset; + copy = skb->len - offset; + if (copy > len - copied) + copy = len - copied; + + ret = skb_copy_datagram_msg(skb, offset, msg, copy); + + if (ret < 0) + goto copy_error; + + /* handle piecemeal consumption of data packets */ + _debug("copied %d+%d", copy, copied); + + offset += copy; + copied += copy; + + if (!(flags & MSG_PEEK)) + sp->offset = offset; + + if (sp->offset < skb->len) { + _debug("buffer full"); + ASSERTCMP(copied, ==, len); + break; + } + + /* we transferred the whole data packet */ + if (sp->hdr.flags & RXRPC_LAST_PACKET) { + _debug("last"); + if (call->conn->out_clientflag) { + /* last byte of reply received */ + ret = copied; + goto terminal_message; + } + + /* last bit of request received */ + if (!(flags & MSG_PEEK)) { + _debug("eat packet"); + if (skb_dequeue(&rx->sk.sk_receive_queue) != + skb) + BUG(); + rxrpc_free_skb(skb); + } + msg->msg_flags &= ~MSG_MORE; + break; + } + + /* move on to the next data message */ + _debug("next"); + if (!continue_call) + continue_call = sp->call; + else + rxrpc_put_call(call); + call = NULL; + + if (flags & MSG_PEEK) { + _debug("peek next"); + skb = skb->next; + if (skb == (struct sk_buff *) &rx->sk.sk_receive_queue) + break; + goto peek_next_packet; + } + + _debug("eat packet"); + if (skb_dequeue(&rx->sk.sk_receive_queue) != skb) + BUG(); + rxrpc_free_skb(skb); + } + + /* end of non-terminal data packet reception for the moment */ + _debug("end rcv data"); +out: + release_sock(&rx->sk); + if (call) + rxrpc_put_call(call); + if (continue_call) + rxrpc_put_call(continue_call); + _leave(" = %d [data]", copied); + return copied; + + /* handle non-DATA messages such as aborts, incoming connections and + * final ACKs */ +receive_non_data_message: + _debug("non-data"); + + if (skb->mark == RXRPC_SKB_MARK_NEW_CALL) { + _debug("RECV NEW CALL"); + ret = put_cmsg(msg, SOL_RXRPC, RXRPC_NEW_CALL, 0, &abort_code); + if (ret < 0) + goto copy_error; + if (!(flags & MSG_PEEK)) { + if (skb_dequeue(&rx->sk.sk_receive_queue) != skb) + BUG(); + rxrpc_free_skb(skb); + } + goto out; + } + + ret = put_cmsg(msg, SOL_RXRPC, RXRPC_USER_CALL_ID, + ullen, &call->user_call_ID); + if (ret < 0) + goto copy_error; + ASSERT(test_bit(RXRPC_CALL_HAS_USERID, &call->flags)); + + switch (skb->mark) { + case RXRPC_SKB_MARK_DATA: + BUG(); + case RXRPC_SKB_MARK_FINAL_ACK: + ret = put_cmsg(msg, SOL_RXRPC, RXRPC_ACK, 0, &abort_code); + break; + case RXRPC_SKB_MARK_BUSY: + ret = put_cmsg(msg, SOL_RXRPC, RXRPC_BUSY, 0, &abort_code); + break; + case RXRPC_SKB_MARK_REMOTE_ABORT: + abort_code = call->remote_abort; + ret = put_cmsg(msg, SOL_RXRPC, RXRPC_ABORT, 4, &abort_code); + break; + case RXRPC_SKB_MARK_LOCAL_ABORT: + abort_code = call->local_abort; + ret = put_cmsg(msg, SOL_RXRPC, RXRPC_ABORT, 4, &abort_code); + break; + case RXRPC_SKB_MARK_NET_ERROR: + _debug("RECV NET ERROR %d", sp->error); + abort_code = sp->error; + ret = put_cmsg(msg, SOL_RXRPC, RXRPC_NET_ERROR, 4, &abort_code); + break; + case RXRPC_SKB_MARK_LOCAL_ERROR: + _debug("RECV LOCAL ERROR %d", sp->error); + abort_code = sp->error; + ret = put_cmsg(msg, SOL_RXRPC, RXRPC_LOCAL_ERROR, 4, + &abort_code); + break; + default: + pr_err("Unknown packet mark %u\n", skb->mark); + BUG(); + break; + } + + if (ret < 0) + goto copy_error; + +terminal_message: + _debug("terminal"); + msg->msg_flags &= ~MSG_MORE; + msg->msg_flags |= MSG_EOR; + + if (!(flags & MSG_PEEK)) { + _net("free terminal skb %p", skb); + if (skb_dequeue(&rx->sk.sk_receive_queue) != skb) + BUG(); + rxrpc_free_skb(skb); + rxrpc_remove_user_ID(rx, call); + } + + release_sock(&rx->sk); + rxrpc_put_call(call); + if (continue_call) + rxrpc_put_call(continue_call); + _leave(" = %d", ret); + return ret; + +copy_error: + _debug("copy error"); + release_sock(&rx->sk); + rxrpc_put_call(call); + if (continue_call) + rxrpc_put_call(continue_call); + _leave(" = %d", ret); + return ret; + +wait_interrupted: + ret = sock_intr_errno(timeo); +wait_error: + finish_wait(sk_sleep(&rx->sk), &wait); + if (continue_call) + rxrpc_put_call(continue_call); + if (copied) + copied = ret; + _leave(" = %d [waitfail %d]", copied, ret); + return copied; + +} + +/** + * rxrpc_kernel_data_delivered - Record delivery of data message + * @skb: Message holding data + * + * Record the delivery of a data message. This permits RxRPC to keep its + * tracking correct. The socket buffer will be deleted. + */ +void rxrpc_kernel_data_delivered(struct sk_buff *skb) +{ + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + struct rxrpc_call *call = sp->call; + + ASSERTCMP(sp->hdr.seq, >=, call->rx_data_recv); + ASSERTCMP(sp->hdr.seq, <=, call->rx_data_recv + 1); + call->rx_data_recv = sp->hdr.seq; + + ASSERTCMP(sp->hdr.seq, >, call->rx_data_eaten); + rxrpc_free_skb(skb); +} + +EXPORT_SYMBOL(rxrpc_kernel_data_delivered); + +/** + * rxrpc_kernel_is_data_last - Determine if data message is last one + * @skb: Message holding data + * + * Determine if data message is last one for the parent call. + */ +bool rxrpc_kernel_is_data_last(struct sk_buff *skb) +{ + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + + ASSERTCMP(skb->mark, ==, RXRPC_SKB_MARK_DATA); + + return sp->hdr.flags & RXRPC_LAST_PACKET; +} + +EXPORT_SYMBOL(rxrpc_kernel_is_data_last); + +/** + * rxrpc_kernel_get_abort_code - Get the abort code from an RxRPC abort message + * @skb: Message indicating an abort + * + * Get the abort code from an RxRPC abort message. + */ +u32 rxrpc_kernel_get_abort_code(struct sk_buff *skb) +{ + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + + switch (skb->mark) { + case RXRPC_SKB_MARK_REMOTE_ABORT: + return sp->call->remote_abort; + case RXRPC_SKB_MARK_LOCAL_ABORT: + return sp->call->local_abort; + default: + BUG(); + } +} + +EXPORT_SYMBOL(rxrpc_kernel_get_abort_code); + +/** + * rxrpc_kernel_get_error - Get the error number from an RxRPC error message + * @skb: Message indicating an error + * + * Get the error number from an RxRPC error message. + */ +int rxrpc_kernel_get_error_number(struct sk_buff *skb) +{ + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + + return sp->error; +} + +EXPORT_SYMBOL(rxrpc_kernel_get_error_number); diff --git a/net/rxrpc/security.c b/net/rxrpc/security.c new file mode 100644 index 0000000..d223253 --- /dev/null +++ b/net/rxrpc/security.c @@ -0,0 +1,168 @@ +/* RxRPC security handling + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +static LIST_HEAD(rxrpc_security_methods); +static DECLARE_RWSEM(rxrpc_security_sem); + +static const struct rxrpc_security *rxrpc_security_types[] = { + [RXRPC_SECURITY_NONE] = &rxrpc_no_security, +#ifdef CONFIG_RXKAD + [RXRPC_SECURITY_RXKAD] = &rxkad, +#endif +}; + +int __init rxrpc_init_security(void) +{ + int i, ret; + + for (i = 0; i < ARRAY_SIZE(rxrpc_security_types); i++) { + if (rxrpc_security_types[i]) { + ret = rxrpc_security_types[i]->init(); + if (ret < 0) + goto failed; + } + } + + return 0; + +failed: + for (i--; i >= 0; i--) + if (rxrpc_security_types[i]) + rxrpc_security_types[i]->exit(); + return ret; +} + +void rxrpc_exit_security(void) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(rxrpc_security_types); i++) + if (rxrpc_security_types[i]) + rxrpc_security_types[i]->exit(); +} + +/* + * look up an rxrpc security module + */ +static const struct rxrpc_security *rxrpc_security_lookup(u8 security_index) +{ + if (security_index >= ARRAY_SIZE(rxrpc_security_types)) + return NULL; + return rxrpc_security_types[security_index]; +} + +/* + * initialise the security on a client connection + */ +int rxrpc_init_client_conn_security(struct rxrpc_connection *conn) +{ + const struct rxrpc_security *sec; + struct rxrpc_key_token *token; + struct key *key = conn->key; + int ret; + + _enter("{%d},{%x}", conn->debug_id, key_serial(key)); + + if (!key) + return 0; + + ret = key_validate(key); + if (ret < 0) + return ret; + + token = key->payload.data[0]; + if (!token) + return -EKEYREJECTED; + + sec = rxrpc_security_lookup(token->security_index); + if (!sec) + return -EKEYREJECTED; + conn->security = sec; + + ret = conn->security->init_connection_security(conn); + if (ret < 0) { + conn->security = &rxrpc_no_security; + return ret; + } + + _leave(" = 0"); + return 0; +} + +/* + * initialise the security on a server connection + */ +int rxrpc_init_server_conn_security(struct rxrpc_connection *conn) +{ + const struct rxrpc_security *sec; + struct rxrpc_local *local = conn->trans->local; + struct rxrpc_sock *rx; + struct key *key; + key_ref_t kref; + char kdesc[5 + 1 + 3 + 1]; + + _enter(""); + + sprintf(kdesc, "%u:%u", conn->service_id, conn->security_ix); + + sec = rxrpc_security_lookup(conn->security_ix); + if (!sec) { + _leave(" = -ENOKEY [lookup]"); + return -ENOKEY; + } + + /* find the service */ + read_lock_bh(&local->services_lock); + list_for_each_entry(rx, &local->services, listen_link) { + if (rx->srx.srx_service == conn->service_id) + goto found_service; + } + + /* the service appears to have died */ + read_unlock_bh(&local->services_lock); + _leave(" = -ENOENT"); + return -ENOENT; + +found_service: + if (!rx->securities) { + read_unlock_bh(&local->services_lock); + _leave(" = -ENOKEY"); + return -ENOKEY; + } + + /* look through the service's keyring */ + kref = keyring_search(make_key_ref(rx->securities, 1UL), + &key_type_rxrpc_s, kdesc); + if (IS_ERR(kref)) { + read_unlock_bh(&local->services_lock); + _leave(" = %ld [search]", PTR_ERR(kref)); + return PTR_ERR(kref); + } + + key = key_ref_to_ptr(kref); + read_unlock_bh(&local->services_lock); + + conn->server_key = key; + conn->security = sec; + + _leave(" = 0"); + return 0; +} diff --git a/net/rxrpc/skbuff.c b/net/rxrpc/skbuff.c new file mode 100644 index 0000000..eee0cfd9 --- /dev/null +++ b/net/rxrpc/skbuff.c @@ -0,0 +1,138 @@ +/* ar-skbuff.c: socket buffer destruction handling + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * set up for the ACK at the end of the receive phase when we discard the final + * receive phase data packet + * - called with softirqs disabled + */ +static void rxrpc_request_final_ACK(struct rxrpc_call *call) +{ + /* the call may be aborted before we have a chance to ACK it */ + write_lock(&call->state_lock); + + switch (call->state) { + case RXRPC_CALL_CLIENT_RECV_REPLY: + call->state = RXRPC_CALL_CLIENT_FINAL_ACK; + _debug("request final ACK"); + + /* get an extra ref on the call for the final-ACK generator to + * release */ + rxrpc_get_call(call); + set_bit(RXRPC_CALL_EV_ACK_FINAL, &call->events); + if (try_to_del_timer_sync(&call->ack_timer) >= 0) + rxrpc_queue_call(call); + break; + + case RXRPC_CALL_SERVER_RECV_REQUEST: + call->state = RXRPC_CALL_SERVER_ACK_REQUEST; + default: + break; + } + + write_unlock(&call->state_lock); +} + +/* + * drop the bottom ACK off of the call ACK window and advance the window + */ +static void rxrpc_hard_ACK_data(struct rxrpc_call *call, + struct rxrpc_skb_priv *sp) +{ + int loop; + u32 seq; + + spin_lock_bh(&call->lock); + + _debug("hard ACK #%u", sp->hdr.seq); + + for (loop = 0; loop < RXRPC_ACKR_WINDOW_ASZ; loop++) { + call->ackr_window[loop] >>= 1; + call->ackr_window[loop] |= + call->ackr_window[loop + 1] << (BITS_PER_LONG - 1); + } + + seq = sp->hdr.seq; + ASSERTCMP(seq, ==, call->rx_data_eaten + 1); + call->rx_data_eaten = seq; + + if (call->ackr_win_top < UINT_MAX) + call->ackr_win_top++; + + ASSERTIFCMP(call->state <= RXRPC_CALL_COMPLETE, + call->rx_data_post, >=, call->rx_data_recv); + ASSERTIFCMP(call->state <= RXRPC_CALL_COMPLETE, + call->rx_data_recv, >=, call->rx_data_eaten); + + if (sp->hdr.flags & RXRPC_LAST_PACKET) { + rxrpc_request_final_ACK(call); + } else if (atomic_dec_and_test(&call->ackr_not_idle) && + test_and_clear_bit(RXRPC_CALL_TX_SOFT_ACK, &call->flags)) { + /* We previously soft-ACK'd some received packets that have now + * been consumed, so send a hard-ACK if no more packets are + * immediately forthcoming to allow the transmitter to free up + * its Tx bufferage. + */ + _debug("send Rx idle ACK"); + __rxrpc_propose_ACK(call, RXRPC_ACK_IDLE, sp->hdr.serial, + false); + } + + spin_unlock_bh(&call->lock); +} + +/* + * destroy a packet that has an RxRPC control buffer + * - advance the hard-ACK state of the parent call (done here in case something + * in the kernel bypasses recvmsg() and steals the packet directly off of the + * socket receive queue) + */ +void rxrpc_packet_destructor(struct sk_buff *skb) +{ + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + struct rxrpc_call *call = sp->call; + + _enter("%p{%p}", skb, call); + + if (call) { + /* send the final ACK on a client call */ + if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA) + rxrpc_hard_ACK_data(call, sp); + rxrpc_put_call(call); + sp->call = NULL; + } + + if (skb->sk) + sock_rfree(skb); + _leave(""); +} + +/** + * rxrpc_kernel_free_skb - Free an RxRPC socket buffer + * @skb: The socket buffer to be freed + * + * Let RxRPC free its own socket buffer, permitting it to maintain debug + * accounting. + */ +void rxrpc_kernel_free_skb(struct sk_buff *skb) +{ + rxrpc_free_skb(skb); +} +EXPORT_SYMBOL(rxrpc_kernel_free_skb); diff --git a/net/rxrpc/transport.c b/net/rxrpc/transport.c new file mode 100644 index 0000000..a1b6518 --- /dev/null +++ b/net/rxrpc/transport.c @@ -0,0 +1,286 @@ +/* RxRPC point-to-point transport session management + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +/* + * Time after last use at which transport record is cleaned up. + */ +unsigned int rxrpc_transport_expiry = 3600 * 24; + +static void rxrpc_transport_reaper(struct work_struct *work); + +static LIST_HEAD(rxrpc_transports); +static DEFINE_RWLOCK(rxrpc_transport_lock); +static DECLARE_DELAYED_WORK(rxrpc_transport_reap, rxrpc_transport_reaper); + +/* + * allocate a new transport session manager + */ +static struct rxrpc_transport *rxrpc_alloc_transport(struct rxrpc_local *local, + struct rxrpc_peer *peer, + gfp_t gfp) +{ + struct rxrpc_transport *trans; + + _enter(""); + + trans = kzalloc(sizeof(struct rxrpc_transport), gfp); + if (trans) { + trans->local = local; + trans->peer = peer; + INIT_LIST_HEAD(&trans->link); + trans->bundles = RB_ROOT; + trans->client_conns = RB_ROOT; + trans->server_conns = RB_ROOT; + skb_queue_head_init(&trans->error_queue); + spin_lock_init(&trans->client_lock); + rwlock_init(&trans->conn_lock); + atomic_set(&trans->usage, 1); + trans->conn_idcounter = peer->srx.srx_service << 16; + trans->debug_id = atomic_inc_return(&rxrpc_debug_id); + + if (peer->srx.transport.family == AF_INET) { + switch (peer->srx.transport_type) { + case SOCK_DGRAM: + INIT_WORK(&trans->error_handler, + rxrpc_UDP_error_handler); + break; + default: + BUG(); + break; + } + } else { + BUG(); + } + } + + _leave(" = %p", trans); + return trans; +} + +/* + * obtain a transport session for the nominated endpoints + */ +struct rxrpc_transport *rxrpc_get_transport(struct rxrpc_local *local, + struct rxrpc_peer *peer, + gfp_t gfp) +{ + struct rxrpc_transport *trans, *candidate; + const char *new = "old"; + int usage; + + _enter("{%pI4+%hu},{%pI4+%hu},", + &local->srx.transport.sin.sin_addr, + ntohs(local->srx.transport.sin.sin_port), + &peer->srx.transport.sin.sin_addr, + ntohs(peer->srx.transport.sin.sin_port)); + + /* search the transport list first */ + read_lock_bh(&rxrpc_transport_lock); + list_for_each_entry(trans, &rxrpc_transports, link) { + if (trans->local == local && trans->peer == peer) + goto found_extant_transport; + } + read_unlock_bh(&rxrpc_transport_lock); + + /* not yet present - create a candidate for a new record and then + * redo the search */ + candidate = rxrpc_alloc_transport(local, peer, gfp); + if (!candidate) { + _leave(" = -ENOMEM"); + return ERR_PTR(-ENOMEM); + } + + write_lock_bh(&rxrpc_transport_lock); + + list_for_each_entry(trans, &rxrpc_transports, link) { + if (trans->local == local && trans->peer == peer) + goto found_extant_second; + } + + /* we can now add the new candidate to the list */ + trans = candidate; + candidate = NULL; + usage = atomic_read(&trans->usage); + + rxrpc_get_local(trans->local); + atomic_inc(&trans->peer->usage); + list_add_tail(&trans->link, &rxrpc_transports); + write_unlock_bh(&rxrpc_transport_lock); + new = "new"; + +success: + _net("TRANSPORT %s %d local %d -> peer %d", + new, + trans->debug_id, + trans->local->debug_id, + trans->peer->debug_id); + + _leave(" = %p {u=%d}", trans, usage); + return trans; + + /* we found the transport in the list immediately */ +found_extant_transport: + usage = atomic_inc_return(&trans->usage); + read_unlock_bh(&rxrpc_transport_lock); + goto success; + + /* we found the transport on the second time through the list */ +found_extant_second: + usage = atomic_inc_return(&trans->usage); + write_unlock_bh(&rxrpc_transport_lock); + kfree(candidate); + goto success; +} + +/* + * find the transport connecting two endpoints + */ +struct rxrpc_transport *rxrpc_find_transport(struct rxrpc_local *local, + struct rxrpc_peer *peer) +{ + struct rxrpc_transport *trans; + + _enter("{%pI4+%hu},{%pI4+%hu},", + &local->srx.transport.sin.sin_addr, + ntohs(local->srx.transport.sin.sin_port), + &peer->srx.transport.sin.sin_addr, + ntohs(peer->srx.transport.sin.sin_port)); + + /* search the transport list */ + read_lock_bh(&rxrpc_transport_lock); + + list_for_each_entry(trans, &rxrpc_transports, link) { + if (trans->local == local && trans->peer == peer) + goto found_extant_transport; + } + + read_unlock_bh(&rxrpc_transport_lock); + _leave(" = NULL"); + return NULL; + +found_extant_transport: + atomic_inc(&trans->usage); + read_unlock_bh(&rxrpc_transport_lock); + _leave(" = %p", trans); + return trans; +} + +/* + * release a transport session + */ +void rxrpc_put_transport(struct rxrpc_transport *trans) +{ + _enter("%p{u=%d}", trans, atomic_read(&trans->usage)); + + ASSERTCMP(atomic_read(&trans->usage), >, 0); + + trans->put_time = ktime_get_seconds(); + if (unlikely(atomic_dec_and_test(&trans->usage))) { + _debug("zombie"); + /* let the reaper determine the timeout to avoid a race with + * overextending the timeout if the reaper is running at the + * same time */ + rxrpc_queue_delayed_work(&rxrpc_transport_reap, 0); + } + _leave(""); +} + +/* + * clean up a transport session + */ +static void rxrpc_cleanup_transport(struct rxrpc_transport *trans) +{ + _net("DESTROY TRANS %d", trans->debug_id); + + rxrpc_purge_queue(&trans->error_queue); + + rxrpc_put_local(trans->local); + rxrpc_put_peer(trans->peer); + kfree(trans); +} + +/* + * reap dead transports that have passed their expiry date + */ +static void rxrpc_transport_reaper(struct work_struct *work) +{ + struct rxrpc_transport *trans, *_p; + unsigned long now, earliest, reap_time; + + LIST_HEAD(graveyard); + + _enter(""); + + now = ktime_get_seconds(); + earliest = ULONG_MAX; + + /* extract all the transports that have been dead too long */ + write_lock_bh(&rxrpc_transport_lock); + list_for_each_entry_safe(trans, _p, &rxrpc_transports, link) { + _debug("reap TRANS %d { u=%d t=%ld }", + trans->debug_id, atomic_read(&trans->usage), + (long) now - (long) trans->put_time); + + if (likely(atomic_read(&trans->usage) > 0)) + continue; + + reap_time = trans->put_time + rxrpc_transport_expiry; + if (reap_time <= now) + list_move_tail(&trans->link, &graveyard); + else if (reap_time < earliest) + earliest = reap_time; + } + write_unlock_bh(&rxrpc_transport_lock); + + if (earliest != ULONG_MAX) { + _debug("reschedule reaper %ld", (long) earliest - now); + ASSERTCMP(earliest, >, now); + rxrpc_queue_delayed_work(&rxrpc_transport_reap, + (earliest - now) * HZ); + } + + /* then destroy all those pulled out */ + while (!list_empty(&graveyard)) { + trans = list_entry(graveyard.next, struct rxrpc_transport, + link); + list_del_init(&trans->link); + + ASSERTCMP(atomic_read(&trans->usage), ==, 0); + rxrpc_cleanup_transport(trans); + } + + _leave(""); +} + +/* + * preemptively destroy all the transport session records rather than waiting + * for them to time out + */ +void __exit rxrpc_destroy_all_transports(void) +{ + _enter(""); + + rxrpc_transport_expiry = 0; + cancel_delayed_work(&rxrpc_transport_reap); + rxrpc_queue_delayed_work(&rxrpc_transport_reap, 0); + + _leave(""); +} -- cgit v0.10.2 From 0d81a51ab94a536a9fb742a5546b6d011ed26c7f Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 13 Jun 2016 13:30:30 +0100 Subject: rxrpc: Update the comments in ar-internal.h to reflect renames Update the section comments in ar-internal.h that indicate the locations of the referenced items to reflect the renames done to the .c files in net/rxrpc/. This also involves some rearrangement to reflect keep the sections in order of filename. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index f715cca..03919b9 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -482,21 +482,21 @@ extern struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *, int, int, gfp_t); /* - * ar-accept.c + * call_accept.c */ void rxrpc_accept_incoming_calls(struct work_struct *); struct rxrpc_call *rxrpc_accept_call(struct rxrpc_sock *, unsigned long); int rxrpc_reject_call(struct rxrpc_sock *); /* - * ar-ack.c + * call_event.c */ void __rxrpc_propose_ACK(struct rxrpc_call *, u8, u32, bool); void rxrpc_propose_ACK(struct rxrpc_call *, u8, u32, bool); void rxrpc_process_call(struct work_struct *); /* - * ar-call.c + * call_object.c */ extern unsigned int rxrpc_max_call_lifetime; extern unsigned int rxrpc_dead_call_expiry; @@ -520,7 +520,14 @@ void __rxrpc_put_call(struct rxrpc_call *); void __exit rxrpc_destroy_all_calls(void); /* - * ar-connection.c + * conn_event.c + */ +void rxrpc_process_connection(struct work_struct *); +void rxrpc_reject_packet(struct rxrpc_local *, struct sk_buff *); +void rxrpc_reject_packets(struct work_struct *); + +/* + * conn_object.c */ extern unsigned int rxrpc_connection_expiry; extern struct list_head rxrpc_connections; @@ -540,27 +547,30 @@ extern struct rxrpc_connection * rxrpc_incoming_connection(struct rxrpc_transport *, struct rxrpc_host_header *); /* - * ar-connevent.c + * input.c */ -void rxrpc_process_connection(struct work_struct *); -void rxrpc_reject_packet(struct rxrpc_local *, struct sk_buff *); -void rxrpc_reject_packets(struct work_struct *); +void rxrpc_data_ready(struct sock *); +int rxrpc_queue_rcv_skb(struct rxrpc_call *, struct sk_buff *, bool, bool); +void rxrpc_fast_process_packet(struct rxrpc_call *, struct sk_buff *); /* - * ar-error.c + * insecure.c */ -void rxrpc_UDP_error_report(struct sock *); -void rxrpc_UDP_error_handler(struct work_struct *); +extern const struct rxrpc_security rxrpc_no_security; /* - * ar-input.c + * key.c */ -void rxrpc_data_ready(struct sock *); -int rxrpc_queue_rcv_skb(struct rxrpc_call *, struct sk_buff *, bool, bool); -void rxrpc_fast_process_packet(struct rxrpc_call *, struct sk_buff *); +extern struct key_type key_type_rxrpc; +extern struct key_type key_type_rxrpc_s; + +int rxrpc_request_key(struct rxrpc_sock *, char __user *, int); +int rxrpc_server_keyring(struct rxrpc_sock *, char __user *, int); +int rxrpc_get_server_data_key(struct rxrpc_connection *, const void *, time_t, + u32); /* - * ar-local.c + * local_object.c */ extern rwlock_t rxrpc_local_lock; @@ -569,18 +579,23 @@ void rxrpc_put_local(struct rxrpc_local *); void __exit rxrpc_destroy_all_locals(void); /* - * ar-key.c + * misc.c */ -extern struct key_type key_type_rxrpc; -extern struct key_type key_type_rxrpc_s; +extern unsigned int rxrpc_max_backlog __read_mostly; +extern unsigned int rxrpc_requested_ack_delay; +extern unsigned int rxrpc_soft_ack_delay; +extern unsigned int rxrpc_idle_ack_delay; +extern unsigned int rxrpc_rx_window_size; +extern unsigned int rxrpc_rx_mtu; +extern unsigned int rxrpc_rx_jumbo_max; -int rxrpc_request_key(struct rxrpc_sock *, char __user *, int); -int rxrpc_server_keyring(struct rxrpc_sock *, char __user *, int); -int rxrpc_get_server_data_key(struct rxrpc_connection *, const void *, time_t, - u32); +extern const char *const rxrpc_pkts[]; +extern const s8 rxrpc_ack_priority[]; + +extern const char *rxrpc_acks(u8 reason); /* - * ar-output.c + * output.c */ extern unsigned int rxrpc_resend_timeout; @@ -588,7 +603,13 @@ int rxrpc_send_packet(struct rxrpc_transport *, struct sk_buff *); int rxrpc_do_sendmsg(struct rxrpc_sock *, struct msghdr *, size_t); /* - * ar-peer.c + * peer_error.c + */ +void rxrpc_UDP_error_report(struct sock *); +void rxrpc_UDP_error_handler(struct work_struct *); + +/* + * peer_object.c */ struct rxrpc_peer *rxrpc_get_peer(struct sockaddr_rxrpc *, gfp_t); void rxrpc_put_peer(struct rxrpc_peer *); @@ -596,20 +617,27 @@ struct rxrpc_peer *rxrpc_find_peer(struct rxrpc_local *, __be32, __be16); void __exit rxrpc_destroy_all_peers(void); /* - * ar-proc.c + * proc.c */ extern const char *const rxrpc_call_states[]; extern const struct file_operations rxrpc_call_seq_fops; extern const struct file_operations rxrpc_connection_seq_fops; /* - * ar-recvmsg.c + * recvmsg.c */ void rxrpc_remove_user_ID(struct rxrpc_sock *, struct rxrpc_call *); int rxrpc_recvmsg(struct socket *, struct msghdr *, size_t, int); /* - * ar-security.c + * rxkad.c + */ +#ifdef CONFIG_RXKAD +extern const struct rxrpc_security rxkad; +#endif + +/* + * security.c */ int __init rxrpc_init_security(void); void rxrpc_exit_security(void); @@ -617,51 +645,11 @@ int rxrpc_init_client_conn_security(struct rxrpc_connection *); int rxrpc_init_server_conn_security(struct rxrpc_connection *); /* - * ar-skbuff.c + * skbuff.c */ void rxrpc_packet_destructor(struct sk_buff *); /* - * ar-transport.c - */ -extern unsigned int rxrpc_transport_expiry; - -struct rxrpc_transport *rxrpc_get_transport(struct rxrpc_local *, - struct rxrpc_peer *, gfp_t); -void rxrpc_put_transport(struct rxrpc_transport *); -void __exit rxrpc_destroy_all_transports(void); -struct rxrpc_transport *rxrpc_find_transport(struct rxrpc_local *, - struct rxrpc_peer *); - -/* - * insecure.c - */ -extern const struct rxrpc_security rxrpc_no_security; - -/* - * misc.c - */ -extern unsigned int rxrpc_max_backlog __read_mostly; -extern unsigned int rxrpc_requested_ack_delay; -extern unsigned int rxrpc_soft_ack_delay; -extern unsigned int rxrpc_idle_ack_delay; -extern unsigned int rxrpc_rx_window_size; -extern unsigned int rxrpc_rx_mtu; -extern unsigned int rxrpc_rx_jumbo_max; - -extern const char *const rxrpc_pkts[]; -extern const s8 rxrpc_ack_priority[]; - -extern const char *rxrpc_acks(u8 reason); - -/* - * rxkad.c - */ -#ifdef CONFIG_RXKAD -extern const struct rxrpc_security rxkad; -#endif - -/* * sysctl.c */ #ifdef CONFIG_SYSCTL @@ -673,6 +661,18 @@ static inline void rxrpc_sysctl_exit(void) {} #endif /* + * transport.c + */ +extern unsigned int rxrpc_transport_expiry; + +struct rxrpc_transport *rxrpc_get_transport(struct rxrpc_local *, + struct rxrpc_peer *, gfp_t); +void rxrpc_put_transport(struct rxrpc_transport *); +void __exit rxrpc_destroy_all_transports(void); +struct rxrpc_transport *rxrpc_find_transport(struct rxrpc_local *, + struct rxrpc_peer *); + +/* * debug tracing */ extern unsigned int rxrpc_debug; -- cgit v0.10.2 From d1dc06dcd0f8fc559e449322d2fa4d769c289e00 Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Tue, 14 Jun 2016 08:29:38 +0300 Subject: virtio_net: fix csum generation for virtio-net devices The commit e858fae2b0b8 ("virtio_net: use common code for virtio_net_hdr and skb GSO conversion") replaced the tun code for header manipulation with the generic helpers. While doing so, it implictly moved the skb_partial_csum_set() invocation after eth_type_trans(), which invalidate the current gso start/offset values. Fix it by moving the helper invocation before the mac pulling. Fixes: e858fae2b0b8 ("virtio_net: use common code for virtio_net_hdr and skb GSO conversion") Reported-by: David Ahern Signed-off-by: Mike Rapoport Signed-off-by: David S. Miller diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index f9470f2..1dd08d4 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -482,10 +482,6 @@ static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq, if (hdr->hdr.flags & VIRTIO_NET_HDR_F_DATA_VALID) skb->ip_summed = CHECKSUM_UNNECESSARY; - skb->protocol = eth_type_trans(skb, dev); - pr_debug("Receiving skb proto 0x%04x len %i type %i\n", - ntohs(skb->protocol), skb->len, skb->pkt_type); - if (virtio_net_hdr_to_skb(skb, &hdr->hdr, virtio_is_little_endian(vi->vdev))) { net_warn_ratelimited("%s: bad gso: type: %u, size: %u\n", @@ -494,6 +490,10 @@ static void receive_buf(struct virtnet_info *vi, struct receive_queue *rq, goto frame_err; } + skb->protocol = eth_type_trans(skb, dev); + pr_debug("Receiving skb proto 0x%04x len %i type %i\n", + ntohs(skb->protocol), skb->len, skb->pkt_type); + napi_gro_receive(&rq->napi, skb); return; -- cgit v0.10.2 From dc73787b8b8a78acbbbb81f0d7eaf80aaf9c3ce1 Mon Sep 17 00:00:00 2001 From: Vasanthakumar Thiagarajan Date: Tue, 7 Jun 2016 15:47:03 +0300 Subject: ath10k: fix some of the macro definitions of HTT_RX_IND message Only five bits are defined to pass tid information in HTT_RX_IND message, so the mask which can be used to extract tid should be 0x1f instead of the current 0x3f. Also, macros which can be used to extract flush_valid and release_valid bits have to be left shifted one bit less because these information follow the tid right after. This patch does not really fix anything functionally because these macros are not used currently. Signed-off-by: Vasanthakumar Thiagarajan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/htt.h b/drivers/net/wireless/ath/ath10k/htt.h index 911c535..430a83e 100644 --- a/drivers/net/wireless/ath/ath10k/htt.h +++ b/drivers/net/wireless/ath/ath10k/htt.h @@ -485,10 +485,10 @@ struct htt_mgmt_tx_completion { __le32 status; } __packed; -#define HTT_RX_INDICATION_INFO0_EXT_TID_MASK (0x3F) +#define HTT_RX_INDICATION_INFO0_EXT_TID_MASK (0x1F) #define HTT_RX_INDICATION_INFO0_EXT_TID_LSB (0) -#define HTT_RX_INDICATION_INFO0_FLUSH_VALID (1 << 6) -#define HTT_RX_INDICATION_INFO0_RELEASE_VALID (1 << 7) +#define HTT_RX_INDICATION_INFO0_FLUSH_VALID (1 << 5) +#define HTT_RX_INDICATION_INFO0_RELEASE_VALID (1 << 6) #define HTT_RX_INDICATION_INFO1_FLUSH_START_SEQNO_MASK 0x0000003F #define HTT_RX_INDICATION_INFO1_FLUSH_START_SEQNO_LSB 0 -- cgit v0.10.2 From 9cd24451859ae34a558cd04162cac0b01f2cfaef Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Tue, 7 Jun 2016 15:47:03 +0300 Subject: ath10k: remove duplicate and unused rx rate flags All these flags are not used and their use is completely covered by 'ath10k_hw_rate_ofdm', 'ath10k_hw_rate_cck', and RX_PPDU_START_RATE_FLAG Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/rx_desc.h b/drivers/net/wireless/ath/ath10k/rx_desc.h index 9ceebea..034e7a5 100644 --- a/drivers/net/wireless/ath/ath10k/rx_desc.h +++ b/drivers/net/wireless/ath/ath10k/rx_desc.h @@ -656,26 +656,6 @@ struct rx_msdu_end { * Reserved: HW should fill with zero. FW should ignore. */ -#define RX_PPDU_START_SIG_RATE_SELECT_OFDM 0 -#define RX_PPDU_START_SIG_RATE_SELECT_CCK 1 - -#define RX_PPDU_START_SIG_RATE_OFDM_48 0 -#define RX_PPDU_START_SIG_RATE_OFDM_24 1 -#define RX_PPDU_START_SIG_RATE_OFDM_12 2 -#define RX_PPDU_START_SIG_RATE_OFDM_6 3 -#define RX_PPDU_START_SIG_RATE_OFDM_54 4 -#define RX_PPDU_START_SIG_RATE_OFDM_36 5 -#define RX_PPDU_START_SIG_RATE_OFDM_18 6 -#define RX_PPDU_START_SIG_RATE_OFDM_9 7 - -#define RX_PPDU_START_SIG_RATE_CCK_LP_11 0 -#define RX_PPDU_START_SIG_RATE_CCK_LP_5_5 1 -#define RX_PPDU_START_SIG_RATE_CCK_LP_2 2 -#define RX_PPDU_START_SIG_RATE_CCK_LP_1 3 -#define RX_PPDU_START_SIG_RATE_CCK_SP_11 4 -#define RX_PPDU_START_SIG_RATE_CCK_SP_5_5 5 -#define RX_PPDU_START_SIG_RATE_CCK_SP_2 6 - #define HTT_RX_PPDU_START_PREAMBLE_LEGACY 0x04 #define HTT_RX_PPDU_START_PREAMBLE_HT 0x08 #define HTT_RX_PPDU_START_PREAMBLE_HT_WITH_TXBF 0x09 @@ -711,25 +691,6 @@ struct rx_msdu_end { /* No idea what this flag means. It seems to be always set in rate. */ #define RX_PPDU_START_RATE_FLAG BIT(3) -enum rx_ppdu_start_rate { - RX_PPDU_START_RATE_OFDM_48M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_OFDM_48M, - RX_PPDU_START_RATE_OFDM_24M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_OFDM_24M, - RX_PPDU_START_RATE_OFDM_12M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_OFDM_12M, - RX_PPDU_START_RATE_OFDM_6M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_OFDM_6M, - RX_PPDU_START_RATE_OFDM_54M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_OFDM_54M, - RX_PPDU_START_RATE_OFDM_36M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_OFDM_36M, - RX_PPDU_START_RATE_OFDM_18M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_OFDM_18M, - RX_PPDU_START_RATE_OFDM_9M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_OFDM_9M, - - RX_PPDU_START_RATE_CCK_LP_11M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_CCK_LP_11M, - RX_PPDU_START_RATE_CCK_LP_5_5M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_CCK_LP_5_5M, - RX_PPDU_START_RATE_CCK_LP_2M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_CCK_LP_2M, - RX_PPDU_START_RATE_CCK_LP_1M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_CCK_LP_1M, - RX_PPDU_START_RATE_CCK_SP_11M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_CCK_SP_11M, - RX_PPDU_START_RATE_CCK_SP_5_5M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_CCK_SP_5_5M, - RX_PPDU_START_RATE_CCK_SP_2M = RX_PPDU_START_RATE_FLAG | ATH10K_HW_RATE_CCK_SP_2M, -}; - struct rx_ppdu_start { struct { u8 pri20_mhz; -- cgit v0.10.2 From 5269c65900d9eef48a6380aba74777d77b8c9061 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Tue, 7 Jun 2016 15:47:04 +0300 Subject: ath10k: fix CCK h/w rates for QCA99X0 and newer chipsets CCK hardware table mapping from QCA99X0 onwards got revised. The CCK hardware rate values are in a proper order wrt. to rate and preamble as below ATH10K_HW_RATE_REV2_CCK_LP_1M = 1, ATH10K_HW_RATE_REV2_CCK_LP_2M = 2, ATH10K_HW_RATE_REV2_CCK_LP_5_5M = 3, ATH10K_HW_RATE_REV2_CCK_LP_11M = 4, ATH10K_HW_RATE_REV2_CCK_SP_2M = 5, ATH10K_HW_RATE_REV2_CCK_SP_5_5M = 6, ATH10K_HW_RATE_REV2_CCK_SP_11M = 7, This results in reporting of rx frames (with CCK rates) totally wrong for QCA99X0, QCA4019. Fix this by having separate CCK rate table for these chipsets with rev2 suffix and registering the correct rate mapping to mac80211 based on the new hw_param (introduced) 'cck_rate_map_rev2' which shall be true for any newchipsets from QCA99X0 onwards Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index 1e88251..2679d00 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -168,6 +168,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = { .uart_pin = 7, .otp_exe_param = 0x00000700, .continuous_frag_desc = true, + .cck_rate_map_rev2 = true, .channel_counters_freq_hz = 150000, .max_probe_resp_desc_thres = 24, .hw_4addr_pad = ATH10K_HW_4ADDR_PAD_BEFORE, @@ -190,6 +191,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = { .uart_pin = 7, .otp_exe_param = 0x00000700, .continuous_frag_desc = true, + .cck_rate_map_rev2 = true, .channel_counters_freq_hz = 150000, .max_probe_resp_desc_thres = 24, .hw_4addr_pad = ATH10K_HW_4ADDR_PAD_BEFORE, @@ -247,6 +249,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = { .has_shifted_cc_wraparound = true, .otp_exe_param = 0x0010000, .continuous_frag_desc = true, + .cck_rate_map_rev2 = true, .channel_counters_freq_hz = 125000, .max_probe_resp_desc_thres = 24, .hw_4addr_pad = ATH10K_HW_4ADDR_PAD_BEFORE, diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h index 3387733..bbc4e0f 100644 --- a/drivers/net/wireless/ath/ath10k/core.h +++ b/drivers/net/wireless/ath/ath10k/core.h @@ -726,6 +726,12 @@ struct ath10k { */ bool continuous_frag_desc; + /* CCK hardware rate table mapping for the newer chipsets + * like QCA99X0, QCA4019 got revised. The CCK h/w rate values + * are in a proper order with respect to the rate/preamble + */ + bool cck_rate_map_rev2; + u32 channel_counters_freq_hz; /* Mgmt tx descriptors threshold for limiting probe response diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h index 49097af..3dbe497 100644 --- a/drivers/net/wireless/ath/ath10k/hw.h +++ b/drivers/net/wireless/ath/ath10k/hw.h @@ -336,6 +336,16 @@ enum ath10k_hw_rate_cck { ATH10K_HW_RATE_CCK_SP_2M, }; +enum ath10k_hw_rate_rev2_cck { + ATH10K_HW_RATE_REV2_CCK_LP_1M = 1, + ATH10K_HW_RATE_REV2_CCK_LP_2M, + ATH10K_HW_RATE_REV2_CCK_LP_5_5M, + ATH10K_HW_RATE_REV2_CCK_LP_11M, + ATH10K_HW_RATE_REV2_CCK_SP_2M, + ATH10K_HW_RATE_REV2_CCK_SP_5_5M, + ATH10K_HW_RATE_REV2_CCK_SP_11M, +}; + enum ath10k_hw_4addr_pad { ATH10K_HW_4ADDR_PAD_AFTER, ATH10K_HW_4ADDR_PAD_BEFORE, diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index 84a3e49..3a170b1 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -62,6 +62,32 @@ static struct ieee80211_rate ath10k_rates[] = { { .bitrate = 540, .hw_value = ATH10K_HW_RATE_OFDM_54M }, }; +static struct ieee80211_rate ath10k_rates_rev2[] = { + { .bitrate = 10, + .hw_value = ATH10K_HW_RATE_REV2_CCK_LP_1M }, + { .bitrate = 20, + .hw_value = ATH10K_HW_RATE_REV2_CCK_LP_2M, + .hw_value_short = ATH10K_HW_RATE_REV2_CCK_SP_2M, + .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 55, + .hw_value = ATH10K_HW_RATE_REV2_CCK_LP_5_5M, + .hw_value_short = ATH10K_HW_RATE_REV2_CCK_SP_5_5M, + .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 110, + .hw_value = ATH10K_HW_RATE_REV2_CCK_LP_11M, + .hw_value_short = ATH10K_HW_RATE_REV2_CCK_SP_11M, + .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + + { .bitrate = 60, .hw_value = ATH10K_HW_RATE_OFDM_6M }, + { .bitrate = 90, .hw_value = ATH10K_HW_RATE_OFDM_9M }, + { .bitrate = 120, .hw_value = ATH10K_HW_RATE_OFDM_12M }, + { .bitrate = 180, .hw_value = ATH10K_HW_RATE_OFDM_18M }, + { .bitrate = 240, .hw_value = ATH10K_HW_RATE_OFDM_24M }, + { .bitrate = 360, .hw_value = ATH10K_HW_RATE_OFDM_36M }, + { .bitrate = 480, .hw_value = ATH10K_HW_RATE_OFDM_48M }, + { .bitrate = 540, .hw_value = ATH10K_HW_RATE_OFDM_54M }, +}; + #define ATH10K_MAC_FIRST_OFDM_RATE_IDX 4 #define ath10k_a_rates (ath10k_rates + ATH10K_MAC_FIRST_OFDM_RATE_IDX) @@ -70,6 +96,9 @@ static struct ieee80211_rate ath10k_rates[] = { #define ath10k_g_rates (ath10k_rates + 0) #define ath10k_g_rates_size (ARRAY_SIZE(ath10k_rates)) +#define ath10k_g_rates_rev2 (ath10k_rates_rev2 + 0) +#define ath10k_g_rates_rev2_size (ARRAY_SIZE(ath10k_rates_rev2)) + static bool ath10k_mac_bitrate_is_cck(int bitrate) { switch (bitrate) { @@ -7709,8 +7738,14 @@ int ath10k_mac_register(struct ath10k *ar) band = &ar->mac.sbands[NL80211_BAND_2GHZ]; band->n_channels = ARRAY_SIZE(ath10k_2ghz_channels); band->channels = channels; - band->n_bitrates = ath10k_g_rates_size; - band->bitrates = ath10k_g_rates; + + if (ar->hw_params.cck_rate_map_rev2) { + band->n_bitrates = ath10k_g_rates_rev2_size; + band->bitrates = ath10k_g_rates_rev2; + } else { + band->n_bitrates = ath10k_g_rates_size; + band->bitrates = ath10k_g_rates; + } ar->hw->wiphy->bands[NL80211_BAND_2GHZ] = band; } -- cgit v0.10.2 From 26c197600b4345f5143676d62260b4985da0b47b Mon Sep 17 00:00:00 2001 From: Vasanthakumar Thiagarajan Date: Tue, 7 Jun 2016 15:47:05 +0300 Subject: ath10k: define an enum to enable cycle counter wraparound logic QCA988X hw implements a different cycle counter wraparound behaviour when compared to QCA4019. To properly handle different wraparound logic for these chipsets replace already available bool hw_params member, has_shifted_cc_wraparound, with an enum which could be extended to handle different wraparound behaviour. This patch keeps the existing logic functionally same and a prepares cycle counter wraparound handling to extend for other chips. Signed-off-by: Vasanthakumar Thiagarajan [kvalo@qca.qualcomm.com: change also QCA9887 wrap type] Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index 2679d00..5427005 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -56,7 +56,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = { .name = "qca988x hw2.0", .patch_load_addr = QCA988X_HW_2_0_PATCH_LOAD_ADDR, .uart_pin = 7, - .has_shifted_cc_wraparound = true, + .cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_ALL, .otp_exe_param = 0, .channel_counters_freq_hz = 88000, .max_probe_resp_desc_thres = 0, @@ -75,7 +75,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = { .name = "qca9887 hw1.0", .patch_load_addr = QCA9887_HW_1_0_PATCH_LOAD_ADDR, .uart_pin = 7, - .has_shifted_cc_wraparound = true, + .cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_ALL, .otp_exe_param = 0, .channel_counters_freq_hz = 88000, .max_probe_resp_desc_thres = 0, @@ -246,7 +246,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = { .name = "qca4019 hw1.0", .patch_load_addr = QCA4019_HW_1_0_PATCH_LOAD_ADDR, .uart_pin = 7, - .has_shifted_cc_wraparound = true, + .cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_ALL, .otp_exe_param = 0x0010000, .continuous_frag_desc = true, .cck_rate_map_rev2 = true, diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h index bbc4e0f..3da18c9 100644 --- a/drivers/net/wireless/ath/ath10k/core.h +++ b/drivers/net/wireless/ath/ath10k/core.h @@ -713,12 +713,10 @@ struct ath10k { int uart_pin; u32 otp_exe_param; - /* This is true if given HW chip has a quirky Cycle Counter - * wraparound which resets to 0x7fffffff instead of 0. All - * other CC related counters (e.g. Rx Clear Count) are divided - * by 2 so they never wraparound themselves. + /* Type of hw cycle counter wraparound logic, for more info + * refer enum ath10k_hw_cc_wraparound_type. */ - bool has_shifted_cc_wraparound; + enum ath10k_hw_cc_wraparound_type cc_wraparound_type; /* Some of chip expects fragment descriptor to be continuous * memory for any TX operation. Set continuous_frag_desc flag diff --git a/drivers/net/wireless/ath/ath10k/hw.c b/drivers/net/wireless/ath/ath10k/hw.c index f544d48..31ec164 100644 --- a/drivers/net/wireless/ath/ath10k/hw.c +++ b/drivers/net/wireless/ath/ath10k/hw.c @@ -179,11 +179,13 @@ void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey, u32 cc, u32 rcc, u32 cc_prev, u32 rcc_prev) { u32 cc_fix = 0; + enum ath10k_hw_cc_wraparound_type wraparound_type; survey->filled |= SURVEY_INFO_TIME | SURVEY_INFO_TIME_BUSY; - if (ar->hw_params.has_shifted_cc_wraparound && cc < cc_prev) { + wraparound_type = ar->hw_params.cc_wraparound_type; + if (wraparound_type == ATH10K_HW_CC_WRAP_SHIFTED_ALL && cc < cc_prev) { cc_fix = 0x7fffffff; survey->filled &= ~SURVEY_INFO_TIME_BUSY; } diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h index 3dbe497..55038c76 100644 --- a/drivers/net/wireless/ath/ath10k/hw.h +++ b/drivers/net/wireless/ath/ath10k/hw.h @@ -351,6 +351,17 @@ enum ath10k_hw_4addr_pad { ATH10K_HW_4ADDR_PAD_BEFORE, }; +enum ath10k_hw_cc_wraparound_type { + ATH10K_HW_CC_WRAP_DISABLED = 0, + + /* This type is when the HW chip has a quirky Cycle Counter + * wraparound which resets to 0x7fffffff instead of 0. All + * other CC related counters (e.g. Rx Clear Count) are divided + * by 2 so they never wraparound themselves. + */ + ATH10K_HW_CC_WRAP_SHIFTED_ALL = 1, +}; + /* Target specific defines for MAIN firmware */ #define TARGET_NUM_VDEVS 8 #define TARGET_NUM_PEER_AST 2 -- cgit v0.10.2 From 8e100354a985e7e9d547684f1038a47e109a0158 Mon Sep 17 00:00:00 2001 From: Vasanthakumar Thiagarajan Date: Tue, 7 Jun 2016 15:47:06 +0300 Subject: ath10k: fix cycle counter wraparound handling for QCA4019 In QCA4019, cycle counter wraparound is not tied to rx clear counter. Each counter would wraparound individually and after wraparound the respective counter will be reset to 0x7fffffff while other counter still running unaffected. Define a new wraparound type for this behaviour and handle it separately so that rx clear counter wraparound is also handled just like cycle counter. With this type of wraparound we can accurately compute and report channel active/busy time when any of the counter overflows. Fixes: ee9ca147c59 ("ath10k: Fix survey reporting with QCA4019") Signed-off-by: Vasanthakumar Thiagarajan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index 5427005..c6291c2 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -246,7 +246,7 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = { .name = "qca4019 hw1.0", .patch_load_addr = QCA4019_HW_1_0_PATCH_LOAD_ADDR, .uart_pin = 7, - .cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_ALL, + .cc_wraparound_type = ATH10K_HW_CC_WRAP_SHIFTED_EACH, .otp_exe_param = 0x0010000, .continuous_frag_desc = true, .cck_rate_map_rev2 = true, diff --git a/drivers/net/wireless/ath/ath10k/hw.c b/drivers/net/wireless/ath/ath10k/hw.c index 31ec164..bd86e7a 100644 --- a/drivers/net/wireless/ath/ath10k/hw.c +++ b/drivers/net/wireless/ath/ath10k/hw.c @@ -179,19 +179,35 @@ void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey, u32 cc, u32 rcc, u32 cc_prev, u32 rcc_prev) { u32 cc_fix = 0; + u32 rcc_fix = 0; enum ath10k_hw_cc_wraparound_type wraparound_type; survey->filled |= SURVEY_INFO_TIME | SURVEY_INFO_TIME_BUSY; wraparound_type = ar->hw_params.cc_wraparound_type; - if (wraparound_type == ATH10K_HW_CC_WRAP_SHIFTED_ALL && cc < cc_prev) { - cc_fix = 0x7fffffff; - survey->filled &= ~SURVEY_INFO_TIME_BUSY; + + if (cc < cc_prev || rcc < rcc_prev) { + switch (wraparound_type) { + case ATH10K_HW_CC_WRAP_SHIFTED_ALL: + if (cc < cc_prev) { + cc_fix = 0x7fffffff; + survey->filled &= ~SURVEY_INFO_TIME_BUSY; + } + break; + case ATH10K_HW_CC_WRAP_SHIFTED_EACH: + if (cc < cc_prev) + cc_fix = 0x7fffffff; + else + rcc_fix = 0x7fffffff; + break; + case ATH10K_HW_CC_WRAP_DISABLED: + break; + } } cc -= cc_prev - cc_fix; - rcc -= rcc_prev; + rcc -= rcc_prev - rcc_fix; survey->time = CCNT_TO_MSEC(ar, cc); survey->time_busy = CCNT_TO_MSEC(ar, rcc); diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h index 55038c76..f31d3ce 100644 --- a/drivers/net/wireless/ath/ath10k/hw.h +++ b/drivers/net/wireless/ath/ath10k/hw.h @@ -360,6 +360,15 @@ enum ath10k_hw_cc_wraparound_type { * by 2 so they never wraparound themselves. */ ATH10K_HW_CC_WRAP_SHIFTED_ALL = 1, + + /* Each hw counter wrapsaround independently. When the + * counter overflows the repestive counter is right shifted + * by 1, i.e reset to 0x7fffffff, and other counters will be + * running unaffected. In this type of wraparound, it should + * be possible to report accurate Rx busy time unlike the + * first type. + */ + ATH10K_HW_CC_WRAP_SHIFTED_EACH = 2, }; /* Target specific defines for MAIN firmware */ -- cgit v0.10.2 From deb85bb1dadf8b4aaa5b33497766dc3f3e4b530e Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Thu, 2 Jun 2016 08:43:50 -0400 Subject: ath5k: fix misplaced default label in sifs switch In this switch statement, the default case does not always assign sifs. In practice, ah->ah_bwmode cannot take values besides the other labels, so this is not an actual problem, but it looks odd and smatch complains thus: ath5k_hw_get_default_sifs() warn: missing break? reassigning 'sifs' Silence the warning by moving default label up a line. Signed-off-by: Bob Copeland Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath5k/pcu.c b/drivers/net/wireless/ath/ath5k/pcu.c index fc47b70..f23c851 100644 --- a/drivers/net/wireless/ath/ath5k/pcu.c +++ b/drivers/net/wireless/ath/ath5k/pcu.c @@ -219,8 +219,8 @@ ath5k_hw_get_default_sifs(struct ath5k_hw *ah) sifs = AR5K_INIT_SIFS_QUARTER_RATE; break; case AR5K_BWMODE_DEFAULT: - sifs = AR5K_INIT_SIFS_DEFAULT_BG; default: + sifs = AR5K_INIT_SIFS_DEFAULT_BG; if (channel->band == NL80211_BAND_5GHZ) sifs = AR5K_INIT_SIFS_DEFAULT_A; break; -- cgit v0.10.2 From 9c830abe9194871d6d2846f37930ef373356a016 Mon Sep 17 00:00:00 2001 From: Maya Erez Date: Wed, 8 Jun 2016 20:07:47 +0300 Subject: wil6210: fix chan check in wil_p2p_listen In wil_p2p_listen chan is checked to protect against NULL pointer access only before setting channel = chan->hw_value. Add a global parameter check to cover all accesses to chan. Signed-off-by: Maya Erez Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wil6210/p2p.c b/drivers/net/wireless/ath/wil6210/p2p.c index 1c91538..213b825 100644 --- a/drivers/net/wireless/ath/wil6210/p2p.c +++ b/drivers/net/wireless/ath/wil6210/p2p.c @@ -114,8 +114,10 @@ int wil_p2p_listen(struct wil6210_priv *wil, unsigned int duration, u8 channel = P2P_DMG_SOCIAL_CHANNEL; int rc; - if (chan) - channel = chan->hw_value; + if (!chan) + return -EINVAL; + + channel = chan->hw_value; wil_dbg_misc(wil, "%s: duration %d\n", __func__, duration); -- cgit v0.10.2 From eb57a5b387d66be1f192a99b3707804e7f63715c Mon Sep 17 00:00:00 2001 From: Lior David Date: Wed, 8 Jun 2016 20:07:48 +0300 Subject: wil6210: abort P2P search when stopping P2P device The nl80211 layer expects P2P search operation to be aborted if needed when stopping P2P device. If the P2P search operation is still running after returning from stop_p2p_device it causes a WARN_ON and possibly a kernel crash. Fix this by aborting the P2P search in wil_cfg80211_stop_p2p_device and preventing P2P search from being started on a stopped P2P device. Note, the fix does not cover the case where a regular scan is started on the P2P device. It will be completed in the future when support is added for aborting a scan operation. Signed-off-by: Lior David Signed-off-by: Maya Erez Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wil6210/cfg80211.c b/drivers/net/wireless/ath/wil6210/cfg80211.c index 5769811..62bf933 100644 --- a/drivers/net/wireless/ath/wil6210/cfg80211.c +++ b/drivers/net/wireless/ath/wil6210/cfg80211.c @@ -378,6 +378,10 @@ static int wil_cfg80211_scan(struct wiphy *wiphy, /* social scan on P2P_DEVICE is handled as p2p search */ if (wdev->iftype == NL80211_IFTYPE_P2P_DEVICE && wil_p2p_is_social_scan(request)) { + if (!wil->p2p.p2p_dev_started) { + wil_err(wil, "P2P search requested on stopped P2P device\n"); + return -EIO; + } wil->scan_request = request; wil->radio_wdev = wdev; rc = wil_p2p_search(wil, request); @@ -1351,6 +1355,7 @@ static int wil_cfg80211_start_p2p_device(struct wiphy *wiphy, struct wil6210_priv *wil = wiphy_to_wil(wiphy); wil_dbg_misc(wil, "%s: entered\n", __func__); + wil->p2p.p2p_dev_started = 1; return 0; } @@ -1358,8 +1363,19 @@ static void wil_cfg80211_stop_p2p_device(struct wiphy *wiphy, struct wireless_dev *wdev) { struct wil6210_priv *wil = wiphy_to_wil(wiphy); + u8 started; wil_dbg_misc(wil, "%s: entered\n", __func__); + mutex_lock(&wil->mutex); + started = wil_p2p_stop_discovery(wil); + if (started && wil->scan_request) { + cfg80211_scan_done(wil->scan_request, 1); + wil->scan_request = NULL; + wil->radio_wdev = wil->wdev; + } + mutex_unlock(&wil->mutex); + + wil->p2p.p2p_dev_started = 0; } static struct cfg80211_ops wil_cfg80211_ops = { diff --git a/drivers/net/wireless/ath/wil6210/wil6210.h b/drivers/net/wireless/ath/wil6210/wil6210.h index 1feaf5a..ecab4af 100644 --- a/drivers/net/wireless/ath/wil6210/wil6210.h +++ b/drivers/net/wireless/ath/wil6210/wil6210.h @@ -458,6 +458,7 @@ struct wil_tid_crypto_rx { struct wil_p2p_info { struct ieee80211_channel listen_chan; u8 discovery_started; + u8 p2p_dev_started; u64 cookie; struct timer_list discovery_timer; /* listen/search duration */ struct work_struct discovery_expired_work; /* listen/search expire */ -- cgit v0.10.2 From 9b648d788d7841867eea220904dfb77df35f1e08 Mon Sep 17 00:00:00 2001 From: Heinrich Schuchardt Date: Wed, 18 May 2016 00:30:41 +0200 Subject: rsi: eliminate superfluous NULL check msg is dereferenced before checking against NULL, e.g. when assigning pad_bytes. Remove the superfluous check in function rsi_mgmt_pkt_to_core. Signed-off-by: Heinrich Schuchardt Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/rsi/rsi_91x_mgmt.c b/drivers/net/wireless/rsi/rsi_91x_mgmt.c index 40658b6..35c14cc 100644 --- a/drivers/net/wireless/rsi/rsi_91x_mgmt.c +++ b/drivers/net/wireless/rsi/rsi_91x_mgmt.c @@ -398,7 +398,7 @@ static int rsi_mgmt_pkt_to_core(struct rsi_common *common, return -ENOLINK; msg_len -= pad_bytes; - if ((msg_len <= 0) || (!msg)) { + if (msg_len <= 0) { rsi_dbg(MGMT_RX_ZONE, "%s: Invalid rx msg of len = %d\n", __func__, msg_len); -- cgit v0.10.2 From 3fdbda446fbcd7fb750179c01338e81cf04e46c7 Mon Sep 17 00:00:00 2001 From: Heinrich Schuchardt Date: Wed, 18 May 2016 01:01:58 +0200 Subject: mwifiex: illegal assignment Variable adapter is incorrectly initialized. Fixes: bf00dc22bc7a ("mwifiex: AMSDU Rx frame handling in AP mode") Signed-off-by: Heinrich Schuchardt Acked-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/uap_txrx.c b/drivers/net/wireless/marvell/mwifiex/uap_txrx.c index 666e91a..bf5660e 100644 --- a/drivers/net/wireless/marvell/mwifiex/uap_txrx.c +++ b/drivers/net/wireless/marvell/mwifiex/uap_txrx.c @@ -272,7 +272,7 @@ int mwifiex_handle_uap_rx_forward(struct mwifiex_private *priv, int mwifiex_uap_recv_packet(struct mwifiex_private *priv, struct sk_buff *skb) { - struct mwifiex_adapter *adapter = adapter; + struct mwifiex_adapter *adapter = priv->adapter; struct mwifiex_sta_node *src_node; struct ethhdr *p_ethhdr; struct sk_buff *skb_uap; -- cgit v0.10.2 From 6b3c33e985f20e7de07fc4b9b1a96dc452e37cb4 Mon Sep 17 00:00:00 2001 From: Heinrich Schuchardt Date: Wed, 18 May 2016 00:46:23 +0200 Subject: rtlwifi: rtl8723be: avoid undefined behavior Do not return undefined value for transmission power if the rate is invalid. Signed-off-by: Heinrich Schuchardt Acked-by: Larry Finger Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c index 445f681..c5ca9df 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c @@ -1019,7 +1019,7 @@ static u8 _rtl8723be_get_txpower_index(struct ieee80211_hw *hw, u8 path, struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); u8 index = (channel - 1); - u8 txpower; + u8 txpower = 0; u8 power_diff_byrate = 0; if (channel > 14 || channel < 1) { -- cgit v0.10.2 From 141bcf099076df1a74317a5b14dcd56c933b9de8 Mon Sep 17 00:00:00 2001 From: Heinrich Schuchardt Date: Wed, 18 May 2016 01:16:01 +0200 Subject: mwiflex: avoid possible null pointer dereference Do not dereference card before checking against NULL value. Signed-off-by: Heinrich Schuchardt Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c index 9246ce8..a35db02 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.c +++ b/drivers/net/wireless/marvell/mwifiex/pcie.c @@ -2901,10 +2901,11 @@ static void mwifiex_unregister_dev(struct mwifiex_adapter *adapter) { struct pcie_service_card *card = adapter->card; const struct mwifiex_pcie_card_reg *reg; - struct pci_dev *pdev = card->dev; + struct pci_dev *pdev; int i; if (card) { + pdev = card->dev; if (card->msix_enable) { for (i = 0; i < MWIFIEX_NUM_MSIX_VECTORS; i++) synchronize_irq(card->msix_entries[i].vector); -- cgit v0.10.2 From a81605b14942a408b7f465cf7197f79224aa4f94 Mon Sep 17 00:00:00 2001 From: Heinrich Schuchardt Date: Wed, 18 May 2016 02:43:58 +0200 Subject: rtlwifi: rtl8192ee: simplify coding Simplify _rtl92ee_phy_path_adda_on. Signed-off-by: Heinrich Schuchardt Acked-by: Larry Finger Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c index 018340a..c2bf8d1 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c @@ -2414,19 +2414,10 @@ static void _rtl92ee_phy_reload_mac_registers(struct ieee80211_hw *hw, static void _rtl92ee_phy_path_adda_on(struct ieee80211_hw *hw, u32 *addareg, bool is_patha_on, bool is2t) { - u32 pathon; u32 i; - pathon = is_patha_on ? 0x0fc01616 : 0x0fc01616; - if (!is2t) { - pathon = 0x0fc01616; - rtl_set_bbreg(hw, addareg[0], MASKDWORD, 0x0fc01616); - } else { - rtl_set_bbreg(hw, addareg[0], MASKDWORD, pathon); - } - - for (i = 1; i < IQK_ADDA_REG_NUM; i++) - rtl_set_bbreg(hw, addareg[i], MASKDWORD, pathon); + for (i = 0; i < IQK_ADDA_REG_NUM; i++) + rtl_set_bbreg(hw, addareg[i], MASKDWORD, 0x0fc01616); } static void _rtl92ee_phy_mac_setting_calibration(struct ieee80211_hw *hw, -- cgit v0.10.2 From 6b128a01c67301a4733138f94ca581c706277ef5 Mon Sep 17 00:00:00 2001 From: Heinrich Schuchardt Date: Wed, 18 May 2016 02:57:13 +0200 Subject: brcm80211: simplify assignment Simplify assignment in wlc_phy_rxcal_gainctrl_nphy_rev5. Signed-off-by: Heinrich Schuchardt Acked-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_n.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_n.c index 99dac9b..b3aab2f 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_n.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_n.c @@ -27017,7 +27017,7 @@ wlc_phy_rxcal_gainctrl_nphy_rev5(struct brcms_phy *pi, u8 rx_core, tx_core = 1 - rx_core; num_samps = 1024; - desired_log2_pwr = (cal_type == 0) ? 13 : 13; + desired_log2_pwr = 13; wlc_phy_rx_iq_coeffs_nphy(pi, 0, &save_comp); zero_comp.a0 = zero_comp.b0 = zero_comp.a1 = zero_comp.b1 = 0x0; -- cgit v0.10.2 From 65c71efe1c59c111a7b9e6d9540f111663b975b2 Mon Sep 17 00:00:00 2001 From: Wei-Ning Huang Date: Thu, 19 May 2016 11:53:43 +0800 Subject: mwifiex: fix racing condition when downloading firmware The action 'check for winner' and 'download firmware' should be an atomic action. This is true for btmrvl driver but not mwmfiex, which cause firmware download to fail when the following senerio happens: 1) mwifiex check winner status: true 2) btmrvl check winner status: true, and start downloading firmware 3) mwfieix tries to download firmware, but failed because btmrvl is already downloading. This won't happen if 1) and 3) is an atomic action. This patch adds sdio_claim/release_host call around those two actions to make sure it's atomic. Signed-off-by: Wei-Ning Huang Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/main.c b/drivers/net/wireless/marvell/mwifiex/main.c index 8b67a55..2b65334 100644 --- a/drivers/net/wireless/marvell/mwifiex/main.c +++ b/drivers/net/wireless/marvell/mwifiex/main.c @@ -21,6 +21,7 @@ #include "wmm.h" #include "cfg80211.h" #include "11n.h" +#include "sdio.h" #define VERSION "1.0" @@ -514,6 +515,7 @@ static void mwifiex_fw_dpc(const struct firmware *firmware, void *context) struct semaphore *sem = adapter->card_sem; bool init_failed = false; struct wireless_dev *wdev; + struct sdio_mmc_card *card = adapter->card; if (!firmware) { mwifiex_dbg(adapter, ERROR, @@ -526,10 +528,16 @@ static void mwifiex_fw_dpc(const struct firmware *firmware, void *context) fw.fw_buf = (u8 *) adapter->firmware->data; fw.fw_len = adapter->firmware->size; - if (adapter->if_ops.dnld_fw) + if (adapter->if_ops.dnld_fw) { ret = adapter->if_ops.dnld_fw(adapter, &fw); - else + } else { + if (adapter->iface_type == MWIFIEX_SDIO) + sdio_claim_host(card->func); ret = mwifiex_dnld_fw(adapter, &fw); + if (adapter->iface_type == MWIFIEX_SDIO) + sdio_release_host(card->func); + } + if (ret == -1) goto err_dnld_fw; -- cgit v0.10.2 From 2cce76c3fab410520610a7d2f52faebc3cfcf843 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 19 May 2016 09:58:49 +0200 Subject: iwlegacy: avoid warning about missing braces gcc-6 warns about code in il3945_hw_txq_ctx_free() being somewhat ambiguous: drivers/net/wireless/intel/iwlegacy/3945.c:1022:5: warning: suggest explicit braces to avoid ambiguous 'else' [-Wparentheses] This adds a set of curly braces to avoid the warning. Signed-off-by: Arnd Bergmann Acked-by: Stanislaw Gruszka Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/intel/iwlegacy/3945.c b/drivers/net/wireless/intel/iwlegacy/3945.c index 7bcedbb..209dc99 100644 --- a/drivers/net/wireless/intel/iwlegacy/3945.c +++ b/drivers/net/wireless/intel/iwlegacy/3945.c @@ -1019,12 +1019,13 @@ il3945_hw_txq_ctx_free(struct il_priv *il) int txq_id; /* Tx queues */ - if (il->txq) + if (il->txq) { for (txq_id = 0; txq_id < il->hw_params.max_txq_num; txq_id++) if (txq_id == IL39_CMD_QUEUE_NUM) il_cmd_queue_free(il); else il_tx_queue_free(il, txq_id); + } /* free tx queue structure */ il_free_txq_mem(il); -- cgit v0.10.2 From d464fd8b48f322223ba12e0d3eb0a2e007eaf03e Mon Sep 17 00:00:00 2001 From: Muhammad Falak R Wani Date: Thu, 19 May 2016 19:29:03 +0530 Subject: brcmfmac: use kmemdup Use kmemdup when some other buffer is immediately copied into allocated region. It replaces call to allocation followed by memcpy, by a single call to kmemdup. Signed-off-by: Muhammad Falak R Wani Acked-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index 87819b3..16271c2 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -6714,11 +6714,10 @@ struct brcmf_cfg80211_info *brcmf_cfg80211_attach(struct brcmf_pub *drvr, return NULL; } - ops = kzalloc(sizeof(*ops), GFP_KERNEL); + ops = kmemdup(&brcmf_cfg80211_ops, sizeof(*ops), GFP_KERNEL); if (!ops) return NULL; - memcpy(ops, &brcmf_cfg80211_ops, sizeof(*ops)); ifp = netdev_priv(ndev); #ifdef CONFIG_PM if (brcmf_feat_is_enabled(ifp, BRCMF_FEAT_WOWL_GTK)) -- cgit v0.10.2 From 4712d88a573272af6f4e45e92c7ec11c8c254ff4 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Fri, 20 May 2016 13:38:57 +0200 Subject: brcmutil: add field storing control channel to the struct brcmu_chan MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Our d11 code supports encoding/decoding channel info into/from chanspec format used by firmware. Current implementation is quite misleading because of the way "chnum" field is used. When encoding channel info, "chnum" has to be filled by a caller with *center* channel number. However when decoding chanspec the same field is filled with a *control* channel number. 1) This can be confusing. It's expected for information to be the same after encoding and decoding. 2) It doesn't allow accessing all info when decoding. Some functions may need to know both channel numbers, e.g. cfg80211 callback getting current channel. Solve this by adding a separated field for control channel. Signed-off-by: Rafał Miłecki Reviewed-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index 16271c2..b044c7e 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -2749,7 +2749,7 @@ static s32 brcmf_inform_single_bss(struct brcmf_cfg80211_info *cfg, if (!bi->ctl_ch) { ch.chspec = le16_to_cpu(bi->chanspec); cfg->d11inf.decchspec(&ch); - bi->ctl_ch = ch.chnum; + bi->ctl_ch = ch.control_ch_num; } channel = bi->ctl_ch; @@ -2867,7 +2867,7 @@ static s32 brcmf_inform_ibss(struct brcmf_cfg80211_info *cfg, else band = wiphy->bands[NL80211_BAND_5GHZ]; - freq = ieee80211_channel_to_frequency(ch.chnum, band->band); + freq = ieee80211_channel_to_frequency(ch.control_ch_num, band->band); cfg->channel = freq; notify_channel = ieee80211_get_channel(wiphy, freq); @@ -2877,7 +2877,7 @@ static s32 brcmf_inform_ibss(struct brcmf_cfg80211_info *cfg, notify_ielen = le32_to_cpu(bi->ie_length); notify_signal = (s16)le16_to_cpu(bi->RSSI) * 100; - brcmf_dbg(CONN, "channel: %d(%d)\n", ch.chnum, freq); + brcmf_dbg(CONN, "channel: %d(%d)\n", ch.control_ch_num, freq); brcmf_dbg(CONN, "capability: %X\n", notify_capability); brcmf_dbg(CONN, "beacon interval: %d\n", notify_interval); brcmf_dbg(CONN, "signal: %d\n", notify_signal); @@ -5295,7 +5295,7 @@ brcmf_bss_roaming_done(struct brcmf_cfg80211_info *cfg, else band = wiphy->bands[NL80211_BAND_5GHZ]; - freq = ieee80211_channel_to_frequency(ch.chnum, band->band); + freq = ieee80211_channel_to_frequency(ch.control_ch_num, band->band); notify_channel = ieee80211_get_channel(wiphy, freq); done: @@ -5817,14 +5817,15 @@ static int brcmf_construct_chaninfo(struct brcmf_cfg80211_info *cfg, channel = band->channels; index = band->n_channels; for (j = 0; j < band->n_channels; j++) { - if (channel[j].hw_value == ch.chnum) { + if (channel[j].hw_value == ch.control_ch_num) { index = j; break; } } channel[index].center_freq = - ieee80211_channel_to_frequency(ch.chnum, band->band); - channel[index].hw_value = ch.chnum; + ieee80211_channel_to_frequency(ch.control_ch_num, + band->band); + channel[index].hw_value = ch.control_ch_num; /* assuming the chanspecs order is HT20, * HT40 upper, HT40 lower, and VHT80. @@ -5926,7 +5927,7 @@ static int brcmf_enable_bw40_2g(struct brcmf_cfg80211_info *cfg) if (WARN_ON(ch.bw != BRCMU_CHAN_BW_40)) continue; for (j = 0; j < band->n_channels; j++) { - if (band->channels[j].hw_value == ch.chnum) + if (band->channels[j].hw_value == ch.control_ch_num) break; } if (WARN_ON(j == band->n_channels)) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c index a70cda6..1652a48 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c @@ -1246,7 +1246,7 @@ bool brcmf_p2p_scan_finding_common_channel(struct brcmf_cfg80211_info *cfg, if (!bi->ctl_ch) { ch.chspec = le16_to_cpu(bi->chanspec); cfg->d11inf.decchspec(&ch); - bi->ctl_ch = ch.chnum; + bi->ctl_ch = ch.control_ch_num; } afx_hdl->peer_chan = bi->ctl_ch; brcmf_dbg(TRACE, "ACTION FRAME SCAN : Peer %pM found, channel : %d\n", @@ -1385,7 +1385,7 @@ int brcmf_p2p_notify_action_frame_rx(struct brcmf_if *ifp, if (test_bit(BRCMF_P2P_STATUS_FINDING_COMMON_CHANNEL, &p2p->status) && (ether_addr_equal(afx_hdl->tx_dst_addr, e->addr))) { - afx_hdl->peer_chan = ch.chnum; + afx_hdl->peer_chan = ch.control_ch_num; brcmf_dbg(INFO, "GON request: Peer found, channel=%d\n", afx_hdl->peer_chan); complete(&afx_hdl->act_frm_scan); @@ -1428,7 +1428,7 @@ int brcmf_p2p_notify_action_frame_rx(struct brcmf_if *ifp, memcpy(&mgmt_frame->u, frame, mgmt_frame_len); mgmt_frame_len += offsetof(struct ieee80211_mgmt, u); - freq = ieee80211_channel_to_frequency(ch.chnum, + freq = ieee80211_channel_to_frequency(ch.control_ch_num, ch.band == BRCMU_CHAN_BAND_2G ? NL80211_BAND_2GHZ : NL80211_BAND_5GHZ); @@ -1873,7 +1873,7 @@ s32 brcmf_p2p_notify_rx_mgmt_p2p_probereq(struct brcmf_if *ifp, if (test_bit(BRCMF_P2P_STATUS_FINDING_COMMON_CHANNEL, &p2p->status) && (ether_addr_equal(afx_hdl->tx_dst_addr, e->addr))) { - afx_hdl->peer_chan = ch.chnum; + afx_hdl->peer_chan = ch.control_ch_num; brcmf_dbg(INFO, "PROBE REQUEST: Peer found, channel=%d\n", afx_hdl->peer_chan); complete(&afx_hdl->act_frm_scan); @@ -1898,7 +1898,7 @@ s32 brcmf_p2p_notify_rx_mgmt_p2p_probereq(struct brcmf_if *ifp, mgmt_frame = (u8 *)(rxframe + 1); mgmt_frame_len = e->datalen - sizeof(*rxframe); - freq = ieee80211_channel_to_frequency(ch.chnum, + freq = ieee80211_channel_to_frequency(ch.control_ch_num, ch.band == BRCMU_CHAN_BAND_2G ? NL80211_BAND_2GHZ : NL80211_BAND_5GHZ); diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmutil/d11.c b/drivers/net/wireless/broadcom/brcm80211/brcmutil/d11.c index 2b2522b..d8b79cb 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmutil/d11.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmutil/d11.c @@ -107,6 +107,7 @@ static void brcmu_d11n_decchspec(struct brcmu_chan *ch) u16 val; ch->chnum = (u8)(ch->chspec & BRCMU_CHSPEC_CH_MASK); + ch->control_ch_num = ch->chnum; switch (ch->chspec & BRCMU_CHSPEC_D11N_BW_MASK) { case BRCMU_CHSPEC_D11N_BW_20: @@ -118,10 +119,10 @@ static void brcmu_d11n_decchspec(struct brcmu_chan *ch) val = ch->chspec & BRCMU_CHSPEC_D11N_SB_MASK; if (val == BRCMU_CHSPEC_D11N_SB_L) { ch->sb = BRCMU_CHAN_SB_L; - ch->chnum -= CH_10MHZ_APART; + ch->control_ch_num -= CH_10MHZ_APART; } else { ch->sb = BRCMU_CHAN_SB_U; - ch->chnum += CH_10MHZ_APART; + ch->control_ch_num += CH_10MHZ_APART; } break; default: @@ -147,6 +148,7 @@ static void brcmu_d11ac_decchspec(struct brcmu_chan *ch) u16 val; ch->chnum = (u8)(ch->chspec & BRCMU_CHSPEC_CH_MASK); + ch->control_ch_num = ch->chnum; switch (ch->chspec & BRCMU_CHSPEC_D11AC_BW_MASK) { case BRCMU_CHSPEC_D11AC_BW_20: @@ -158,10 +160,10 @@ static void brcmu_d11ac_decchspec(struct brcmu_chan *ch) val = ch->chspec & BRCMU_CHSPEC_D11AC_SB_MASK; if (val == BRCMU_CHSPEC_D11AC_SB_L) { ch->sb = BRCMU_CHAN_SB_L; - ch->chnum -= CH_10MHZ_APART; + ch->control_ch_num -= CH_10MHZ_APART; } else if (val == BRCMU_CHSPEC_D11AC_SB_U) { ch->sb = BRCMU_CHAN_SB_U; - ch->chnum += CH_10MHZ_APART; + ch->control_ch_num += CH_10MHZ_APART; } else { WARN_ON_ONCE(1); } @@ -172,16 +174,16 @@ static void brcmu_d11ac_decchspec(struct brcmu_chan *ch) BRCMU_CHSPEC_D11AC_SB_SHIFT); switch (ch->sb) { case BRCMU_CHAN_SB_LL: - ch->chnum -= CH_30MHZ_APART; + ch->control_ch_num -= CH_30MHZ_APART; break; case BRCMU_CHAN_SB_LU: - ch->chnum -= CH_10MHZ_APART; + ch->control_ch_num -= CH_10MHZ_APART; break; case BRCMU_CHAN_SB_UL: - ch->chnum += CH_10MHZ_APART; + ch->control_ch_num += CH_10MHZ_APART; break; case BRCMU_CHAN_SB_UU: - ch->chnum += CH_30MHZ_APART; + ch->control_ch_num += CH_30MHZ_APART; break; default: WARN_ON_ONCE(1); diff --git a/drivers/net/wireless/broadcom/brcm80211/include/brcmu_d11.h b/drivers/net/wireless/broadcom/brcm80211/include/brcmu_d11.h index f9745ea..8b8b2ec 100644 --- a/drivers/net/wireless/broadcom/brcm80211/include/brcmu_d11.h +++ b/drivers/net/wireless/broadcom/brcm80211/include/brcmu_d11.h @@ -125,14 +125,36 @@ enum brcmu_chan_sb { BRCMU_CHAN_SB_UU = BRCMU_CHAN_SB_LUU, }; +/** + * struct brcmu_chan - stores channel formats + * + * This structure can be used with functions translating chanspec into generic + * channel info and the other way. + * + * @chspec: firmware specific format + * @chnum: center channel number + * @control_ch_num: control channel number + * @band: frequency band + * @bw: channel width + * @sb: control sideband (location of control channel against the center one) + */ struct brcmu_chan { u16 chspec; u8 chnum; + u8 control_ch_num; u8 band; enum brcmu_chan_bw bw; enum brcmu_chan_sb sb; }; +/** + * struct brcmu_d11inf - provides functions translating channel format + * + * @io_type: determines version of channel format used by firmware + * @encchspec: encodes channel info into a chanspec, requires center channel + * number, ignores control one + * @decchspec: decodes chanspec into generic info + */ struct brcmu_d11inf { u8 io_type; -- cgit v0.10.2 From ee6e7aa383944ce62860f35c86f1ac7da7dd27b6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Fri, 20 May 2016 13:38:58 +0200 Subject: brcmfmac: support get_channel cfg80211 callback MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is important for brcmfmac as some of released firmwares (e.g. brcmfmac4366b-pcie.bin) may pick different channel than requested. This has been tested with BCM4366B1 in D-Link DIR-885L. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index b044c7e..ba65a93 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -4907,6 +4907,68 @@ exit: return err; } +static int brcmf_cfg80211_get_channel(struct wiphy *wiphy, + struct wireless_dev *wdev, + struct cfg80211_chan_def *chandef) +{ + struct brcmf_cfg80211_info *cfg = wiphy_to_cfg(wiphy); + struct net_device *ndev = wdev->netdev; + struct brcmf_if *ifp; + struct brcmu_chan ch; + enum nl80211_band band = 0; + enum nl80211_chan_width width = 0; + u32 chanspec; + int freq, err; + + if (!ndev) + return -ENODEV; + ifp = netdev_priv(ndev); + + err = brcmf_fil_iovar_int_get(ifp, "chanspec", &chanspec); + if (err) { + brcmf_err("chanspec failed (%d)\n", err); + return err; + } + + ch.chspec = chanspec; + cfg->d11inf.decchspec(&ch); + + switch (ch.band) { + case BRCMU_CHAN_BAND_2G: + band = NL80211_BAND_2GHZ; + break; + case BRCMU_CHAN_BAND_5G: + band = NL80211_BAND_5GHZ; + break; + } + + switch (ch.bw) { + case BRCMU_CHAN_BW_80: + width = NL80211_CHAN_WIDTH_80; + break; + case BRCMU_CHAN_BW_40: + width = NL80211_CHAN_WIDTH_40; + break; + case BRCMU_CHAN_BW_20: + width = NL80211_CHAN_WIDTH_20; + break; + case BRCMU_CHAN_BW_80P80: + width = NL80211_CHAN_WIDTH_80P80; + break; + case BRCMU_CHAN_BW_160: + width = NL80211_CHAN_WIDTH_160; + break; + } + + freq = ieee80211_channel_to_frequency(ch.control_ch_num, band); + chandef->chan = ieee80211_get_channel(wiphy, freq); + chandef->width = width; + chandef->center_freq1 = ieee80211_channel_to_frequency(ch.chnum, band); + chandef->center_freq2 = 0; + + return 0; +} + static int brcmf_cfg80211_crit_proto_start(struct wiphy *wiphy, struct wireless_dev *wdev, enum nl80211_crit_proto_id proto, @@ -5069,6 +5131,7 @@ static struct cfg80211_ops brcmf_cfg80211_ops = { .mgmt_tx = brcmf_cfg80211_mgmt_tx, .remain_on_channel = brcmf_p2p_remain_on_channel, .cancel_remain_on_channel = brcmf_cfg80211_cancel_remain_on_channel, + .get_channel = brcmf_cfg80211_get_channel, .start_p2p_device = brcmf_p2p_start_device, .stop_p2p_device = brcmf_p2p_stop_device, .crit_proto_start = brcmf_cfg80211_crit_proto_start, -- cgit v0.10.2 From 5c87a55adbd5eb3536893c40086253e15ea53cd5 Mon Sep 17 00:00:00 2001 From: Mathias Krause Date: Sat, 21 May 2016 15:43:31 +0200 Subject: mwifiex: remove misleading GFP_DMA flag in buffer allocations The GFP_DMA flag is obviously misunderstood in the mwifiex driver. It's meant for legacy ISA DMA memory mappings only -- the lower 16MB on x86. That doesn't apply to PCIe or SDIO devices, I guess. Remove the GFP_DMA flag to reduce the need to place the socket buffer allocation into the low mem DMA area, which might already be in use by other drivers. This misuse was flagged by the PaX USERCOPY feature by chance, as it detected the user copy operation from a DMA buffer in the recvfrom() syscall path. Signed-off-by: Mathias Krause Tested-by: Dennis Wassenberg Cc: Amitkumar Karwar Cc: Nishant Sarmukadam Cc: Xinming Hu Cc: Kalle Valo Cc: Brad Spengler Cc: PaX Team Acked-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/11n_aggr.c b/drivers/net/wireless/marvell/mwifiex/11n_aggr.c index 1efef3b..dc49c3d 100644 --- a/drivers/net/wireless/marvell/mwifiex/11n_aggr.c +++ b/drivers/net/wireless/marvell/mwifiex/11n_aggr.c @@ -184,7 +184,7 @@ mwifiex_11n_aggregate_pkt(struct mwifiex_private *priv, tx_info_src = MWIFIEX_SKB_TXCB(skb_src); skb_aggr = mwifiex_alloc_dma_align_buf(adapter->tx_buf_size, - GFP_ATOMIC | GFP_DMA); + GFP_ATOMIC); if (!skb_aggr) { spin_unlock_irqrestore(&priv->wmm.ra_list_spinlock, ra_list_flags); diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c index a35db02..1b1e266 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.c +++ b/drivers/net/wireless/marvell/mwifiex/pcie.c @@ -507,7 +507,7 @@ static int mwifiex_init_rxq_ring(struct mwifiex_adapter *adapter) for (i = 0; i < MWIFIEX_MAX_TXRX_BD; i++) { /* Allocate skb here so that firmware can DMA data from it */ skb = mwifiex_alloc_dma_align_buf(MWIFIEX_RX_DATA_BUF_SIZE, - GFP_KERNEL | GFP_DMA); + GFP_KERNEL); if (!skb) { mwifiex_dbg(adapter, ERROR, "Unable to allocate skb for RX ring.\n"); @@ -1319,7 +1319,7 @@ static int mwifiex_pcie_process_recv_data(struct mwifiex_adapter *adapter) } skb_tmp = mwifiex_alloc_dma_align_buf(MWIFIEX_RX_DATA_BUF_SIZE, - GFP_KERNEL | GFP_DMA); + GFP_KERNEL); if (!skb_tmp) { mwifiex_dbg(adapter, ERROR, "Unable to allocate skb.\n"); diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c index bdc51ff..674465e 100644 --- a/drivers/net/wireless/marvell/mwifiex/sdio.c +++ b/drivers/net/wireless/marvell/mwifiex/sdio.c @@ -1492,7 +1492,7 @@ rx_curr_single: mwifiex_dbg(adapter, INFO, "info: RX: port: %d, rx_len: %d\n", port, rx_len); - skb = mwifiex_alloc_dma_align_buf(rx_len, GFP_KERNEL | GFP_DMA); + skb = mwifiex_alloc_dma_align_buf(rx_len, GFP_KERNEL); if (!skb) { mwifiex_dbg(adapter, ERROR, "single skb allocated fail,\t" @@ -1597,7 +1597,7 @@ static int mwifiex_process_int_status(struct mwifiex_adapter *adapter) rx_len = (u16) (rx_blocks * MWIFIEX_SDIO_BLOCK_SIZE); mwifiex_dbg(adapter, INFO, "info: rx_len = %d\n", rx_len); - skb = mwifiex_alloc_dma_align_buf(rx_len, GFP_KERNEL | GFP_DMA); + skb = mwifiex_alloc_dma_align_buf(rx_len, GFP_KERNEL); if (!skb) return -1; -- cgit v0.10.2 From c62d50a4062e586855a68c3acfbc2a2c299270cd Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Wed, 25 May 2016 08:37:35 -0700 Subject: mwifiex: inform disconnection initiator correctly. This patch ensures that 'locally_generated' parameter is correctly passed to cfg80211_disconnected() API. Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/join.c b/drivers/net/wireless/marvell/mwifiex/join.c index 62211fc..a4b773d 100644 --- a/drivers/net/wireless/marvell/mwifiex/join.c +++ b/drivers/net/wireless/marvell/mwifiex/join.c @@ -1281,7 +1281,7 @@ int mwifiex_ret_802_11_ad_hoc(struct mwifiex_private *priv, if (result) { mwifiex_dbg(priv->adapter, ERROR, "ADHOC_RESP: failed\n"); if (priv->media_connected) - mwifiex_reset_connect_state(priv, result); + mwifiex_reset_connect_state(priv, result, true); memset(&priv->curr_bss_params.bss_descriptor, 0x00, sizeof(struct mwifiex_bssdescriptor)); diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h index 0207af0..f0cd055 100644 --- a/drivers/net/wireless/marvell/mwifiex/main.h +++ b/drivers/net/wireless/marvell/mwifiex/main.h @@ -1128,7 +1128,8 @@ int mwifiex_cmd_802_11_associate(struct mwifiex_private *priv, struct mwifiex_bssdescriptor *bss_desc); int mwifiex_ret_802_11_associate(struct mwifiex_private *priv, struct host_cmd_ds_command *resp); -void mwifiex_reset_connect_state(struct mwifiex_private *priv, u16 reason); +void mwifiex_reset_connect_state(struct mwifiex_private *priv, u16 reason, + bool from_ap); u8 mwifiex_band_to_radio_type(u8 band); int mwifiex_deauthenticate(struct mwifiex_private *priv, u8 *mac); void mwifiex_deauthenticate_all(struct mwifiex_adapter *adapter); diff --git a/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c b/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c index d18c797..bcfd4b7 100644 --- a/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c +++ b/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c @@ -553,7 +553,8 @@ static int mwifiex_ret_802_11_deauthenticate(struct mwifiex_private *priv, if (!memcmp(resp->params.deauth.mac_addr, &priv->curr_bss_params.bss_descriptor.mac_address, sizeof(resp->params.deauth.mac_addr))) - mwifiex_reset_connect_state(priv, WLAN_REASON_DEAUTH_LEAVING); + mwifiex_reset_connect_state(priv, WLAN_REASON_DEAUTH_LEAVING, + false); return 0; } @@ -566,7 +567,7 @@ static int mwifiex_ret_802_11_deauthenticate(struct mwifiex_private *priv, static int mwifiex_ret_802_11_ad_hoc_stop(struct mwifiex_private *priv, struct host_cmd_ds_command *resp) { - mwifiex_reset_connect_state(priv, WLAN_REASON_DEAUTH_LEAVING); + mwifiex_reset_connect_state(priv, WLAN_REASON_DEAUTH_LEAVING, false); return 0; } diff --git a/drivers/net/wireless/marvell/mwifiex/sta_event.c b/drivers/net/wireless/marvell/mwifiex/sta_event.c index 0104108..0cefd40 100644 --- a/drivers/net/wireless/marvell/mwifiex/sta_event.c +++ b/drivers/net/wireless/marvell/mwifiex/sta_event.c @@ -40,8 +40,8 @@ * - Erases current SSID and BSSID information * - Sends a disconnect event to upper layers/applications. */ -void -mwifiex_reset_connect_state(struct mwifiex_private *priv, u16 reason_code) +void mwifiex_reset_connect_state(struct mwifiex_private *priv, u16 reason_code, + bool from_ap) { struct mwifiex_adapter *adapter = priv->adapter; @@ -140,7 +140,7 @@ mwifiex_reset_connect_state(struct mwifiex_private *priv, u16 reason_code) if (priv->bss_mode == NL80211_IFTYPE_STATION || priv->bss_mode == NL80211_IFTYPE_P2P_CLIENT) { cfg80211_disconnected(priv->netdev, reason_code, NULL, 0, - false, GFP_KERNEL); + !from_ap, GFP_KERNEL); } eth_zero_addr(priv->cfg_bssid); @@ -574,7 +574,7 @@ int mwifiex_process_sta_event(struct mwifiex_private *priv) if (priv->media_connected) { reason_code = le16_to_cpu(*(__le16 *)adapter->event_body); - mwifiex_reset_connect_state(priv, reason_code); + mwifiex_reset_connect_state(priv, reason_code, true); } break; @@ -589,7 +589,7 @@ int mwifiex_process_sta_event(struct mwifiex_private *priv) if (priv->media_connected) { reason_code = le16_to_cpu(*(__le16 *)adapter->event_body); - mwifiex_reset_connect_state(priv, reason_code); + mwifiex_reset_connect_state(priv, reason_code, true); } break; @@ -599,7 +599,7 @@ int mwifiex_process_sta_event(struct mwifiex_private *priv) if (priv->media_connected) { reason_code = le16_to_cpu(*(__le16 *)adapter->event_body); - mwifiex_reset_connect_state(priv, reason_code); + mwifiex_reset_connect_state(priv, reason_code, true); } break; -- cgit v0.10.2 From 7659f50c3c38f808af074dcb5b54247e0cc9989e Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 12 Jun 2016 23:30:54 +0200 Subject: net: ethernet: enic: move to new ethtool api {get|set}_link_ksettings The ethtool api {get|set}_settings is deprecated. We move the enic driver to new api {get|set}_link_ksettings. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cisco/enic/enic_ethtool.c b/drivers/net/ethernet/cisco/enic/enic_ethtool.c index f44a39c..fd3980c 100644 --- a/drivers/net/ethernet/cisco/enic/enic_ethtool.c +++ b/drivers/net/ethernet/cisco/enic/enic_ethtool.c @@ -103,25 +103,29 @@ static void enic_intr_coal_set_rx(struct enic *enic, u32 timer) } } -static int enic_get_settings(struct net_device *netdev, - struct ethtool_cmd *ecmd) +static int enic_get_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *ecmd) { struct enic *enic = netdev_priv(netdev); + struct ethtool_link_settings *base = &ecmd->base; - ecmd->supported = (SUPPORTED_10000baseT_Full | SUPPORTED_FIBRE); - ecmd->advertising = (ADVERTISED_10000baseT_Full | ADVERTISED_FIBRE); - ecmd->port = PORT_FIBRE; - ecmd->transceiver = XCVR_EXTERNAL; + ethtool_link_ksettings_add_link_mode(ecmd, supported, + 10000baseT_Full); + ethtool_link_ksettings_add_link_mode(ecmd, supported, FIBRE); + ethtool_link_ksettings_add_link_mode(ecmd, advertising, + 10000baseT_Full); + ethtool_link_ksettings_add_link_mode(ecmd, advertising, FIBRE); + base->port = PORT_FIBRE; if (netif_carrier_ok(netdev)) { - ethtool_cmd_speed_set(ecmd, vnic_dev_port_speed(enic->vdev)); - ecmd->duplex = DUPLEX_FULL; + base->speed = vnic_dev_port_speed(enic->vdev); + base->duplex = DUPLEX_FULL; } else { - ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN); - ecmd->duplex = DUPLEX_UNKNOWN; + base->speed = SPEED_UNKNOWN; + base->duplex = DUPLEX_UNKNOWN; } - ecmd->autoneg = AUTONEG_DISABLE; + base->autoneg = AUTONEG_DISABLE; return 0; } @@ -500,7 +504,6 @@ static int enic_set_rxfh(struct net_device *netdev, const u32 *indir, } static const struct ethtool_ops enic_ethtool_ops = { - .get_settings = enic_get_settings, .get_drvinfo = enic_get_drvinfo, .get_msglevel = enic_get_msglevel, .set_msglevel = enic_set_msglevel, @@ -516,6 +519,7 @@ static const struct ethtool_ops enic_ethtool_ops = { .get_rxfh_key_size = enic_get_rxfh_key_size, .get_rxfh = enic_get_rxfh, .set_rxfh = enic_set_rxfh, + .get_link_ksettings = enic_get_ksettings, }; void enic_set_ethtool_ops(struct net_device *netdev) -- cgit v0.10.2 From cf6645f8ebc69775a857b7c51928f3ad9e37aa66 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 13 Jun 2016 02:25:28 -0400 Subject: bnxt_en: Add function for VF driver to query default VLAN. The PF can setup a default VLAN for a VF. The default VLAN tag is automatically inserted and stripped without the knowledge of the stack running on the VF. The VF driver needs to know that default VLAN is enabled as VLAN acceleration on the RX side is no longer supported. Call netdev_update_features() to fix up the VLAN features as necessary. Also, VLAN strip mode must be enabled to strip out the default VLAN tag. Only allow VF default VLAN to be set if the firmware spec is >= 1.2.1. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index c777cde..67608d5 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -3277,6 +3277,7 @@ static int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id) unsigned int ring = 0, grp_idx; struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id]; struct hwrm_vnic_cfg_input req = {0}; + u16 def_vlan = 0; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_CFG, -1, -1); /* Only RSS support for now TBD: COS & LB */ @@ -3297,7 +3298,11 @@ static int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id) req.mru = cpu_to_le16(bp->dev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN); - if (bp->flags & BNXT_FLAG_STRIP_VLAN) +#ifdef CONFIG_BNXT_SRIOV + if (BNXT_VF(bp)) + def_vlan = bp->vf.vlan; +#endif + if ((bp->flags & BNXT_FLAG_STRIP_VLAN) || def_vlan) req.flags |= cpu_to_le32(VNIC_CFG_REQ_FLAGS_VLAN_STRIP_MODE); return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); @@ -3836,6 +3841,32 @@ static int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp) return 0; } +static int bnxt_hwrm_func_qcfg(struct bnxt *bp) +{ + struct hwrm_func_qcfg_input req = {0}; + int rc; + + bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_QCFG, -1, -1); + req.fid = cpu_to_le16(0xffff); + mutex_lock(&bp->hwrm_cmd_lock); + rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); + if (rc) + goto func_qcfg_exit; + +#ifdef CONFIG_BNXT_SRIOV + if (BNXT_VF(bp)) { + struct hwrm_func_qcfg_output *resp = bp->hwrm_cmd_resp_addr; + struct bnxt_vf_info *vf = &bp->vf; + + vf->vlan = le16_to_cpu(resp->vlan) & VLAN_VID_MASK; + } +#endif + +func_qcfg_exit: + mutex_unlock(&bp->hwrm_cmd_lock); + return rc; +} + int bnxt_hwrm_func_qcaps(struct bnxt *bp) { int rc = 0; @@ -4230,6 +4261,11 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init) netdev_warn(bp->dev, "HWRM set coalescing failure rc: %x\n", rc); + if (BNXT_VF(bp)) { + bnxt_hwrm_func_qcfg(bp); + netdev_update_features(bp->dev); + } + return 0; err_out: @@ -5469,7 +5505,14 @@ static netdev_features_t bnxt_fix_features(struct net_device *dev, features |= NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_STAG_RX; } - +#ifdef CONFIG_BNXT_SRIOV + if (BNXT_VF(bp)) { + if (bp->vf.vlan) { + features &= ~(NETIF_F_HW_VLAN_CTAG_RX | + NETIF_F_HW_VLAN_STAG_RX); + } + } +#endif return features; } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c index 363884d..50d2007 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c @@ -143,6 +143,9 @@ int bnxt_set_vf_vlan(struct net_device *dev, int vf_id, u16 vlan_id, u8 qos) u16 vlan_tag; int rc; + if (bp->hwrm_spec_code < 0x10201) + return -ENOTSUPP; + rc = bnxt_vf_ndo_prep(bp, vf_id); if (rc) return rc; -- cgit v0.10.2 From 6988bd920c6ea53497ed15db947408b7488c9e36 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 13 Jun 2016 02:25:29 -0400 Subject: bnxt_en: Add new function bnxt_reset(). When a default VLAN is added to the VF, the VF driver needs to reset to pick up the default VLAN ID. We can use the same tx timeout reset logic to do that, without the debug output. This new function, with the silent parameter to suppress debug output will now serve both purposes. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 67608d5..e30c43b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -5628,9 +5628,10 @@ static void bnxt_dbg_dump_states(struct bnxt *bp) } } -static void bnxt_reset_task(struct bnxt *bp) +static void bnxt_reset_task(struct bnxt *bp, bool silent) { - bnxt_dbg_dump_states(bp); + if (!silent) + bnxt_dbg_dump_states(bp); if (netif_running(bp->dev)) { bnxt_close_nic(bp, false, false); bnxt_open_nic(bp, false, false); @@ -5681,6 +5682,23 @@ bnxt_restart_timer: mod_timer(&bp->timer, jiffies + bp->current_interval); } +/* Only called from bnxt_sp_task() */ +static void bnxt_reset(struct bnxt *bp, bool silent) +{ + /* bnxt_reset_task() calls bnxt_close_nic() which waits + * for BNXT_STATE_IN_SP_TASK to clear. + * If there is a parallel dev_close(), bnxt_close() may be holding + * rtnl() and waiting for BNXT_STATE_IN_SP_TASK to clear. So we + * must clear BNXT_STATE_IN_SP_TASK before holding rtnl(). + */ + clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state); + rtnl_lock(); + if (test_bit(BNXT_STATE_OPEN, &bp->state)) + bnxt_reset_task(bp, silent); + set_bit(BNXT_STATE_IN_SP_TASK, &bp->state); + rtnl_unlock(); +} + static void bnxt_cfg_ntp_filters(struct bnxt *); static void bnxt_sp_task(struct work_struct *work) @@ -5717,16 +5735,8 @@ static void bnxt_sp_task(struct work_struct *work) bnxt_hwrm_tunnel_dst_port_free( bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN); } - if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event)) { - /* bnxt_reset_task() calls bnxt_close_nic() which waits - * for BNXT_STATE_IN_SP_TASK to clear. - */ - clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state); - rtnl_lock(); - bnxt_reset_task(bp); - set_bit(BNXT_STATE_IN_SP_TASK, &bp->state); - rtnl_unlock(); - } + if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event)) + bnxt_reset(bp, false); if (test_and_clear_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event)) bnxt_get_port_module_status(bp); -- cgit v0.10.2 From fc0f19294d1ffaf9366b10d966f86e6cf13335a4 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 13 Jun 2016 02:25:30 -0400 Subject: bnxt_en: Handle VF_CFG_CHANGE event from firmware. When the VF driver gets this event, the VF configuration has changed (such as default VLAN). The VF driver will initiate a silent reset to pick up the new configuration. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index e30c43b..51f9e38 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -125,6 +125,7 @@ static const u16 bnxt_async_events_arr[] = { HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_STATUS_CHANGE, HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD, HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED, + HWRM_ASYNC_EVENT_CMPL_EVENT_ID_VF_CFG_CHANGE, HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_CHANGE, }; @@ -1358,6 +1359,11 @@ static int bnxt_async_event_process(struct bnxt *bp, set_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event); break; } + case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_VF_CFG_CHANGE: + if (BNXT_PF(bp)) + goto async_event_process_exit; + set_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event); + break; default: netdev_err(bp->dev, "unhandled ASYNC event (id 0x%x)\n", event_id); @@ -5738,6 +5744,9 @@ static void bnxt_sp_task(struct work_struct *work) if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event)) bnxt_reset(bp, false); + if (test_and_clear_bit(BNXT_RESET_TASK_SILENT_SP_EVENT, &bp->sp_event)) + bnxt_reset(bp, true); + if (test_and_clear_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event)) bnxt_get_port_module_status(bp); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 2824d65..538eb1c 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -1018,6 +1018,7 @@ struct bnxt { #define BNXT_HWRM_PF_UNLOAD_SP_EVENT 8 #define BNXT_PERIODIC_STATS_SP_EVENT 9 #define BNXT_HWRM_PORT_MODULE_SP_EVENT 10 +#define BNXT_RESET_TASK_SILENT_SP_EVENT 11 struct bnxt_pf_info pf; #ifdef CONFIG_BNXT_SRIOV -- cgit v0.10.2 From 567b2abe68551781b725b3b739672da41cb92ef0 Mon Sep 17 00:00:00 2001 From: Satish Baddipadige Date: Mon, 13 Jun 2016 02:25:31 -0400 Subject: bnxt_en: Enable NPAR (NIC Partitioning) Support. NPAR type is read from bnxt_hwrm_func_qcfg. Do not allow changing link parameters if in NPAR mode sinc ethe port is shared among multiple partitions. The link parameters are set up by firmware. Signed-off-by: Satish Baddipadige Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 51f9e38..9d785e6 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -3850,6 +3850,7 @@ static int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp) static int bnxt_hwrm_func_qcfg(struct bnxt *bp) { struct hwrm_func_qcfg_input req = {0}; + struct hwrm_func_qcfg_output *resp = bp->hwrm_cmd_resp_addr; int rc; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_QCFG, -1, -1); @@ -3861,12 +3862,18 @@ static int bnxt_hwrm_func_qcfg(struct bnxt *bp) #ifdef CONFIG_BNXT_SRIOV if (BNXT_VF(bp)) { - struct hwrm_func_qcfg_output *resp = bp->hwrm_cmd_resp_addr; struct bnxt_vf_info *vf = &bp->vf; vf->vlan = le16_to_cpu(resp->vlan) & VLAN_VID_MASK; } #endif + switch (resp->port_partition_type) { + case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_0: + case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_5: + case FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR2_0: + bp->port_partition_type = resp->port_partition_type; + break; + } func_qcfg_exit: mutex_unlock(&bp->hwrm_cmd_lock); @@ -4965,7 +4972,7 @@ static int bnxt_hwrm_shutdown_link(struct bnxt *bp) { struct hwrm_port_phy_cfg_input req = {0}; - if (BNXT_VF(bp)) + if (!BNXT_SINGLE_PF(bp)) return 0; if (pci_num_vf(bp->pdev)) @@ -6427,6 +6434,8 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto init_err; } + bnxt_hwrm_func_qcfg(bp); + bnxt_set_tpa_flags(bp); bnxt_set_ring_params(bp); if (BNXT_PF(bp)) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 538eb1c..ec18a07 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -907,6 +907,8 @@ struct bnxt { #define BNXT_PF(bp) (!((bp)->flags & BNXT_FLAG_VF)) #define BNXT_VF(bp) ((bp)->flags & BNXT_FLAG_VF) +#define BNXT_NPAR(bp) ((bp)->port_partition_type) +#define BNXT_SINGLE_PF(bp) (BNXT_PF(bp) && !BNXT_NPAR(bp)) struct bnxt_napi **bnapi; @@ -993,6 +995,7 @@ struct bnxt { __le16 vxlan_fw_dst_port_id; u8 nge_port_cnt; __le16 nge_fw_dst_port_id; + u8 port_partition_type; u16 rx_coal_ticks; u16 rx_coal_ticks_irq; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index a38cb04..89050ed 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -823,7 +823,7 @@ static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) u32 speed, fw_advertising = 0; bool set_pause = false; - if (BNXT_VF(bp)) + if (!BNXT_SINGLE_PF(bp)) return rc; if (cmd->autoneg == AUTONEG_ENABLE) { @@ -911,7 +911,7 @@ static int bnxt_set_pauseparam(struct net_device *dev, struct bnxt *bp = netdev_priv(dev); struct bnxt_link_info *link_info = &bp->link_info; - if (BNXT_VF(bp)) + if (!BNXT_SINGLE_PF(bp)) return rc; if (epause->autoneg) { @@ -1433,7 +1433,7 @@ static int bnxt_set_eee(struct net_device *dev, struct ethtool_eee *edata) _bnxt_fw_to_ethtool_adv_spds(link_info->advertising, 0); int rc = 0; - if (BNXT_VF(bp)) + if (!BNXT_SINGLE_PF(bp)) return 0; if (!(bp->flags & BNXT_FLAG_EEE_CAP)) -- cgit v0.10.2 From ebcd4eeb2a0b4859d7aaa3308b222a30d51a643f Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 13 Jun 2016 02:25:32 -0400 Subject: bnxt_en: Add PCI device ID for 57404 NPAR devices. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 9d785e6..1cf885a 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -78,6 +78,7 @@ enum board_idx { BCM57402, BCM57404, BCM57406, + BCM57404_NPAR, BCM57314, BCM57304_VF, BCM57404_VF, @@ -93,6 +94,7 @@ static const struct { { "Broadcom BCM57402 NetXtreme-E Dual-port 10Gb Ethernet" }, { "Broadcom BCM57404 NetXtreme-E Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57406 NetXtreme-E Dual-port 10GBase-T Ethernet" }, + { "Broadcom BCM57404 NetXtreme-E Ethernet Partition" }, { "Broadcom BCM57314 NetXtreme-C Dual-port 10Gb/25Gb/40Gb/50Gb Ethernet" }, { "Broadcom BCM57304 NetXtreme-C Ethernet Virtual Function" }, { "Broadcom BCM57404 NetXtreme-E Ethernet Virtual Function" }, @@ -105,6 +107,7 @@ static const struct pci_device_id bnxt_pci_tbl[] = { { PCI_VDEVICE(BROADCOM, 0x16d0), .driver_data = BCM57402 }, { PCI_VDEVICE(BROADCOM, 0x16d1), .driver_data = BCM57404 }, { PCI_VDEVICE(BROADCOM, 0x16d2), .driver_data = BCM57406 }, + { PCI_VDEVICE(BROADCOM, 0x16d4), .driver_data = BCM57404_NPAR }, { PCI_VDEVICE(BROADCOM, 0x16df), .driver_data = BCM57314 }, #ifdef CONFIG_BNXT_SRIOV { PCI_VDEVICE(BROADCOM, 0x16cb), .driver_data = BCM57304_VF }, -- cgit v0.10.2 From 659c805cc01b3c5a6d972db0408164371a2bab4b Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 13 Jun 2016 02:25:33 -0400 Subject: bnxt_en: Define the supported chip numbers. Define all the supported chip numbers and chip categories. Store the chip_num returned by firmware. If the call to get the version and chip number fails, we should abort. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 1cf885a..0fd27b0 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -4037,6 +4037,8 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp) if (resp->hwrm_intf_maj >= 1) bp->hwrm_max_req_len = le16_to_cpu(resp->max_req_win_len); + bp->chip_num = le16_to_cpu(resp->chip_num); + hwrm_ver_get_exit: mutex_unlock(&bp->hwrm_cmd_lock); return rc; @@ -6414,7 +6416,9 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto init_err; mutex_init(&bp->hwrm_cmd_lock); - bnxt_hwrm_ver_get(bp); + rc = bnxt_hwrm_ver_get(bp); + if (rc) + goto init_err; rc = bnxt_hwrm_func_drv_rgtr(bp); if (rc) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index ec18a07..b754be6 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -873,6 +873,44 @@ struct bnxt { void __iomem *bar2; u32 reg_base; + u16 chip_num; +#define CHIP_NUM_57301 0x16c8 +#define CHIP_NUM_57302 0x16c9 +#define CHIP_NUM_57304 0x16ca +#define CHIP_NUM_57402 0x16d0 +#define CHIP_NUM_57404 0x16d1 +#define CHIP_NUM_57406 0x16d2 + +#define CHIP_NUM_57311 0x16ce +#define CHIP_NUM_57312 0x16cf +#define CHIP_NUM_57314 0x16df +#define CHIP_NUM_57412 0x16d6 +#define CHIP_NUM_57414 0x16d7 +#define CHIP_NUM_57416 0x16d8 +#define CHIP_NUM_57417 0x16d9 + +#define BNXT_CHIP_NUM_5730X(chip_num) \ + ((chip_num) >= CHIP_NUM_57301 && \ + (chip_num) <= CHIP_NUM_57304) + +#define BNXT_CHIP_NUM_5740X(chip_num) \ + ((chip_num) >= CHIP_NUM_57402 && \ + (chip_num) <= CHIP_NUM_57406) + +#define BNXT_CHIP_NUM_5731X(chip_num) \ + ((chip_num) == CHIP_NUM_57311 || \ + (chip_num) == CHIP_NUM_57312 || \ + (chip_num) == CHIP_NUM_57314) + +#define BNXT_CHIP_NUM_5741X(chip_num) \ + ((chip_num) >= CHIP_NUM_57412 && \ + (chip_num) <= CHIP_NUM_57417) + +#define BNXT_CHIP_NUM_57X0X(chip_num) \ + (BNXT_CHIP_NUM_5730X(chip_num) || BNXT_CHIP_NUM_5740X(chip_num)) + +#define BNXT_CHIP_NUM_57X1X(chip_num) \ + (BNXT_CHIP_NUM_5731X(chip_num) || BNXT_CHIP_NUM_5741X(chip_num)) struct net_device *dev; struct pci_dev *pdev; -- cgit v0.10.2 From 309369c9b3f6a8665e581d9014f222b602f6845a Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 13 Jun 2016 02:25:34 -0400 Subject: bnxt_en: Refactor bnxt_gro_skb(). Newer chips require different logic to handle GRO packets. So refactor the code so that we can call different functions depending on the chip. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 0fd27b0..4e3b9f5 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -945,29 +945,15 @@ static void bnxt_abort_tpa(struct bnxt *bp, struct bnxt_napi *bnapi, #define BNXT_IPV4_HDR_SIZE (sizeof(struct iphdr) + sizeof(struct tcphdr)) #define BNXT_IPV6_HDR_SIZE (sizeof(struct ipv6hdr) + sizeof(struct tcphdr)) -static inline struct sk_buff *bnxt_gro_skb(struct bnxt_tpa_info *tpa_info, - struct rx_tpa_end_cmp *tpa_end, - struct rx_tpa_end_cmp_ext *tpa_end1, +static struct sk_buff *bnxt_gro_func_5730x(struct bnxt_tpa_info *tpa_info, + int payload_off, int tcp_ts, struct sk_buff *skb) { #ifdef CONFIG_INET struct tcphdr *th; - int payload_off, tcp_opt_len = 0; - int len, nw_off; - u16 segs; - - segs = TPA_END_TPA_SEGS(tpa_end); - if (segs == 1) - return skb; + int len, nw_off, tcp_opt_len; - NAPI_GRO_CB(skb)->count = segs; - skb_shinfo(skb)->gso_size = - le32_to_cpu(tpa_end1->rx_tpa_end_cmp_seg_len); - skb_shinfo(skb)->gso_type = tpa_info->gso_type; - payload_off = (le32_to_cpu(tpa_end->rx_tpa_end_cmp_misc_v1) & - RX_TPA_END_CMP_PAYLOAD_OFFSET) >> - RX_TPA_END_CMP_PAYLOAD_OFFSET_SHIFT; - if (TPA_END_GRO_TS(tpa_end)) + if (tcp_ts) tcp_opt_len = 12; if (tpa_info->gso_type == SKB_GSO_TCPV4) { @@ -1024,6 +1010,32 @@ static inline struct sk_buff *bnxt_gro_skb(struct bnxt_tpa_info *tpa_info, return skb; } +static inline struct sk_buff *bnxt_gro_skb(struct bnxt *bp, + struct bnxt_tpa_info *tpa_info, + struct rx_tpa_end_cmp *tpa_end, + struct rx_tpa_end_cmp_ext *tpa_end1, + struct sk_buff *skb) +{ +#ifdef CONFIG_INET + int payload_off; + u16 segs; + + segs = TPA_END_TPA_SEGS(tpa_end); + if (segs == 1) + return skb; + + NAPI_GRO_CB(skb)->count = segs; + skb_shinfo(skb)->gso_size = + le32_to_cpu(tpa_end1->rx_tpa_end_cmp_seg_len); + skb_shinfo(skb)->gso_type = tpa_info->gso_type; + payload_off = (le32_to_cpu(tpa_end->rx_tpa_end_cmp_misc_v1) & + RX_TPA_END_CMP_PAYLOAD_OFFSET) >> + RX_TPA_END_CMP_PAYLOAD_OFFSET_SHIFT; + skb = bp->gro_func(tpa_info, payload_off, TPA_END_GRO_TS(tpa_end), skb); +#endif + return skb; +} + static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons, @@ -1134,7 +1146,7 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp, } if (TPA_END_GRO(tpa_end)) - skb = bnxt_gro_skb(tpa_info, tpa_end, tpa_end1, skb); + skb = bnxt_gro_skb(bp, tpa_info, tpa_end, tpa_end1, skb); return skb; } @@ -6420,6 +6432,8 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (rc) goto init_err; + bp->gro_func = bnxt_gro_func_5730x; + rc = bnxt_hwrm_func_drv_rgtr(bp); if (rc) goto init_err; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index b754be6..d62f4c2 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -953,6 +953,9 @@ struct bnxt { struct bnxt_rx_ring_info *rx_ring; struct bnxt_tx_ring_info *tx_ring; + struct sk_buff * (*gro_func)(struct bnxt_tpa_info *, int, int, + struct sk_buff *); + u32 rx_buf_size; u32 rx_buf_use_size; /* useable size */ u32 rx_ring_size; -- cgit v0.10.2 From 94758f8de037cf5c62eb56287f5d5e937cda8c9b Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 13 Jun 2016 02:25:35 -0400 Subject: bnxt_en: Add GRO logic for BCM5731X chips. Add bnxt_gro_func_5731x() to handle GRO packets for this chip. The completion structures used in the new chip have new data to help determine the header offsets. The offsets can be off by 4 if the packet is an internal loopback packet (e.g. from one VF to another VF). Some additional logic is added to adjust the offsets if it is a loopback packet. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 4e3b9f5..5f9285c 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -924,6 +924,7 @@ static void bnxt_tpa_start(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, } tpa_info->flags2 = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_flags2); tpa_info->metadata = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_metadata); + tpa_info->hdr_info = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_hdr_info); rxr->rx_prod = NEXT_RX(prod); cons = NEXT_RX(cons); @@ -942,6 +943,90 @@ static void bnxt_abort_tpa(struct bnxt *bp, struct bnxt_napi *bnapi, bnxt_reuse_rx_agg_bufs(bnapi, cp_cons, agg_bufs); } +static struct sk_buff *bnxt_gro_func_5731x(struct bnxt_tpa_info *tpa_info, + int payload_off, int tcp_ts, + struct sk_buff *skb) +{ +#ifdef CONFIG_INET + struct tcphdr *th; + int len, nw_off; + u16 outer_ip_off, inner_ip_off, inner_mac_off; + u32 hdr_info = tpa_info->hdr_info; + bool loopback = false; + + inner_ip_off = BNXT_TPA_INNER_L3_OFF(hdr_info); + inner_mac_off = BNXT_TPA_INNER_L2_OFF(hdr_info); + outer_ip_off = BNXT_TPA_OUTER_L3_OFF(hdr_info); + + /* If the packet is an internal loopback packet, the offsets will + * have an extra 4 bytes. + */ + if (inner_mac_off == 4) { + loopback = true; + } else if (inner_mac_off > 4) { + __be16 proto = *((__be16 *)(skb->data + inner_ip_off - + ETH_HLEN - 2)); + + /* We only support inner iPv4/ipv6. If we don't see the + * correct protocol ID, it must be a loopback packet where + * the offsets are off by 4. + */ + if (proto != htons(ETH_P_IP) && proto && htons(ETH_P_IPV6)) + loopback = true; + } + if (loopback) { + /* internal loopback packet, subtract all offsets by 4 */ + inner_ip_off -= 4; + inner_mac_off -= 4; + outer_ip_off -= 4; + } + + nw_off = inner_ip_off - ETH_HLEN; + skb_set_network_header(skb, nw_off); + if (tpa_info->flags2 & RX_TPA_START_CMP_FLAGS2_IP_TYPE) { + struct ipv6hdr *iph = ipv6_hdr(skb); + + skb_set_transport_header(skb, nw_off + sizeof(struct ipv6hdr)); + len = skb->len - skb_transport_offset(skb); + th = tcp_hdr(skb); + th->check = ~tcp_v6_check(len, &iph->saddr, &iph->daddr, 0); + } else { + struct iphdr *iph = ip_hdr(skb); + + skb_set_transport_header(skb, nw_off + sizeof(struct iphdr)); + len = skb->len - skb_transport_offset(skb); + th = tcp_hdr(skb); + th->check = ~tcp_v4_check(len, iph->saddr, iph->daddr, 0); + } + + if (inner_mac_off) { /* tunnel */ + struct udphdr *uh = NULL; + __be16 proto = *((__be16 *)(skb->data + outer_ip_off - + ETH_HLEN - 2)); + + if (proto == htons(ETH_P_IP)) { + struct iphdr *iph = (struct iphdr *)skb->data; + + if (iph->protocol == IPPROTO_UDP) + uh = (struct udphdr *)(iph + 1); + } else { + struct ipv6hdr *iph = (struct ipv6hdr *)skb->data; + + if (iph->nexthdr == IPPROTO_UDP) + uh = (struct udphdr *)(iph + 1); + } + if (uh) { + if (uh->check) + skb_shinfo(skb)->gso_type |= + SKB_GSO_UDP_TUNNEL_CSUM; + else + skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL; + } + } +#endif + return skb; +} + #define BNXT_IPV4_HDR_SIZE (sizeof(struct iphdr) + sizeof(struct tcphdr)) #define BNXT_IPV6_HDR_SIZE (sizeof(struct ipv6hdr) + sizeof(struct tcphdr)) @@ -2283,7 +2368,7 @@ static void bnxt_set_tpa_flags(struct bnxt *bp) bp->flags &= ~BNXT_FLAG_TPA; if (bp->dev->features & NETIF_F_LRO) bp->flags |= BNXT_FLAG_LRO; - if ((bp->dev->features & NETIF_F_GRO) && (bp->pdev->revision > 0)) + if (bp->dev->features & NETIF_F_GRO) bp->flags |= BNXT_FLAG_GRO; } @@ -6433,6 +6518,8 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) goto init_err; bp->gro_func = bnxt_gro_func_5730x; + if (BNXT_CHIP_NUM_57X1X(bp->chip_num)) + bp->gro_func = bnxt_gro_func_5731x; rc = bnxt_hwrm_func_drv_rgtr(bp); if (rc) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index d62f4c2..c1b41fb 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -298,13 +298,14 @@ struct rx_tpa_start_cmp_ext { #define RX_TPA_START_CMP_FLAGS2_L4_CS_CALC (0x1 << 1) #define RX_TPA_START_CMP_FLAGS2_T_IP_CS_CALC (0x1 << 2) #define RX_TPA_START_CMP_FLAGS2_T_L4_CS_CALC (0x1 << 3) + #define RX_TPA_START_CMP_FLAGS2_IP_TYPE (0x1 << 8) __le32 rx_tpa_start_cmp_metadata; __le32 rx_tpa_start_cmp_cfa_code_v2; #define RX_TPA_START_CMP_V2 (0x1 << 0) #define RX_TPA_START_CMP_CFA_CODE (0xffff << 16) #define RX_TPA_START_CMPL_CFA_CODE_SHIFT 16 - __le32 rx_tpa_start_cmp_unused5; + __le32 rx_tpa_start_cmp_hdr_info; }; struct rx_tpa_end_cmp { @@ -584,6 +585,19 @@ struct bnxt_tpa_info { u32 metadata; enum pkt_hash_types hash_type; u32 rss_hash; + u32 hdr_info; + +#define BNXT_TPA_L4_SIZE(hdr_info) \ + (((hdr_info) & 0xf8000000) ? ((hdr_info) >> 27) : 32) + +#define BNXT_TPA_INNER_L3_OFF(hdr_info) \ + (((hdr_info) >> 18) & 0x1ff) + +#define BNXT_TPA_INNER_L2_OFF(hdr_info) \ + (((hdr_info) >> 9) & 0x1ff) + +#define BNXT_TPA_OUTER_L3_OFF(hdr_info) \ + ((hdr_info) & 0x1ff) }; struct bnxt_rx_ring_info { -- cgit v0.10.2 From b24eb6ae7058ca1a42b0532489e5f5796c107d65 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 13 Jun 2016 02:25:36 -0400 Subject: bnxt_en: Add BCM5731X and BCM5741X device IDs. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 5f9285c..c275329 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -75,13 +75,22 @@ enum board_idx { BCM57301, BCM57302, BCM57304, + BCM57311, + BCM57312, BCM57402, BCM57404, BCM57406, BCM57404_NPAR, + BCM57412, + BCM57414, + BCM57416, + BCM57417, + BCM57414_NPAR, BCM57314, BCM57304_VF, BCM57404_VF, + BCM57414_VF, + BCM57314_VF, }; /* indexed by enum above */ @@ -91,27 +100,45 @@ static const struct { { "Broadcom BCM57301 NetXtreme-C Single-port 10Gb Ethernet" }, { "Broadcom BCM57302 NetXtreme-C Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57304 NetXtreme-C Dual-port 10Gb/25Gb/40Gb/50Gb Ethernet" }, + { "Broadcom BCM57311 NetXtreme-C Single-port 10Gb Ethernet" }, + { "Broadcom BCM57312 NetXtreme-C Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57402 NetXtreme-E Dual-port 10Gb Ethernet" }, { "Broadcom BCM57404 NetXtreme-E Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57406 NetXtreme-E Dual-port 10GBase-T Ethernet" }, { "Broadcom BCM57404 NetXtreme-E Ethernet Partition" }, + { "Broadcom BCM57412 NetXtreme-E Dual-port 10Gb Ethernet" }, + { "Broadcom BCM57414 NetXtreme-E Dual-port 10Gb/25Gb Ethernet" }, + { "Broadcom BCM57416 NetXtreme-E Dual-port 10GBase-T Ethernet" }, + { "Broadcom BCM57417 NetXtreme-E Dual-port 10GBase-T Ethernet" }, + { "Broadcom BCM57414 NetXtreme-E Ethernet Partition" }, { "Broadcom BCM57314 NetXtreme-C Dual-port 10Gb/25Gb/40Gb/50Gb Ethernet" }, { "Broadcom BCM57304 NetXtreme-C Ethernet Virtual Function" }, { "Broadcom BCM57404 NetXtreme-E Ethernet Virtual Function" }, + { "Broadcom BCM57414 NetXtreme-E Ethernet Virtual Function" }, + { "Broadcom BCM57314 NetXtreme-E Ethernet Virtual Function" }, }; static const struct pci_device_id bnxt_pci_tbl[] = { { PCI_VDEVICE(BROADCOM, 0x16c8), .driver_data = BCM57301 }, { PCI_VDEVICE(BROADCOM, 0x16c9), .driver_data = BCM57302 }, { PCI_VDEVICE(BROADCOM, 0x16ca), .driver_data = BCM57304 }, + { PCI_VDEVICE(BROADCOM, 0x16ce), .driver_data = BCM57311 }, + { PCI_VDEVICE(BROADCOM, 0x16cf), .driver_data = BCM57312 }, { PCI_VDEVICE(BROADCOM, 0x16d0), .driver_data = BCM57402 }, { PCI_VDEVICE(BROADCOM, 0x16d1), .driver_data = BCM57404 }, { PCI_VDEVICE(BROADCOM, 0x16d2), .driver_data = BCM57406 }, { PCI_VDEVICE(BROADCOM, 0x16d4), .driver_data = BCM57404_NPAR }, + { PCI_VDEVICE(BROADCOM, 0x16d6), .driver_data = BCM57412 }, + { PCI_VDEVICE(BROADCOM, 0x16d7), .driver_data = BCM57414 }, + { PCI_VDEVICE(BROADCOM, 0x16d8), .driver_data = BCM57416 }, + { PCI_VDEVICE(BROADCOM, 0x16d9), .driver_data = BCM57417 }, + { PCI_VDEVICE(BROADCOM, 0x16de), .driver_data = BCM57414_NPAR }, { PCI_VDEVICE(BROADCOM, 0x16df), .driver_data = BCM57314 }, #ifdef CONFIG_BNXT_SRIOV { PCI_VDEVICE(BROADCOM, 0x16cb), .driver_data = BCM57304_VF }, { PCI_VDEVICE(BROADCOM, 0x16d3), .driver_data = BCM57404_VF }, + { PCI_VDEVICE(BROADCOM, 0x16dc), .driver_data = BCM57414_VF }, + { PCI_VDEVICE(BROADCOM, 0x16e1), .driver_data = BCM57314_VF }, #endif { 0 } }; @@ -134,7 +161,8 @@ static const u16 bnxt_async_events_arr[] = { static bool bnxt_vf_pciid(enum board_idx idx) { - return (idx == BCM57304_VF || idx == BCM57404_VF); + return (idx == BCM57304_VF || idx == BCM57404_VF || + idx == BCM57314_VF || idx == BCM57414_VF); } #define DB_CP_REARM_FLAGS (DB_KEY_CP | DB_IDX_VALID) -- cgit v0.10.2 From 93ed8117336485af2cedb069d28f3d4270fb90a1 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 13 Jun 2016 02:25:37 -0400 Subject: bnxt_en: Don't allow autoneg on cards that don't support it. Some cards do not support autoneg. The current code does not prevent the user from enabling autoneg with ethtool on such cards, causing confusion. Firmware provides the autoneg capability information and we just need to store it in the support_auto_speeds field in bnxt_link_info struct. The ethtool set_settings() call will check this field before proceeding with autoneg. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index c275329..9aaa6a6 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -4823,6 +4823,7 @@ static int bnxt_hwrm_phy_qcaps(struct bnxt *bp) int rc = 0; struct hwrm_port_phy_qcaps_input req = {0}; struct hwrm_port_phy_qcaps_output *resp = bp->hwrm_cmd_resp_addr; + struct bnxt_link_info *link_info = &bp->link_info; if (bp->hwrm_spec_code < 0x10201) return 0; @@ -4845,6 +4846,8 @@ static int bnxt_hwrm_phy_qcaps(struct bnxt *bp) bp->lpi_tmr_hi = le32_to_cpu(resp->valid_tx_lpi_timer_high) & PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_HIGH_MASK; } + link_info->support_auto_speeds = + le16_to_cpu(resp->supported_speeds_auto_mode); hwrm_phy_qcaps_exit: mutex_unlock(&bp->hwrm_cmd_lock); @@ -6368,6 +6371,12 @@ static int bnxt_probe_phy(struct bnxt *bp) return rc; } + /* Older firmware does not have supported_auto_speeds, so assume + * that all supported speeds can be autonegotiated. + */ + if (link_info->auto_link_speeds && !link_info->support_auto_speeds) + link_info->support_auto_speeds = link_info->support_speeds; + /*initialize the ethool setting copy with NVM settings */ if (BNXT_AUTO_MODE(link_info->auto_mode)) { link_info->autoneg = BNXT_AUTONEG_SPEED; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index c1b41fb..04cc69b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -849,6 +849,7 @@ struct bnxt_link_info { #define BNXT_LINK_SPEED_MSK_25GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_25GB #define BNXT_LINK_SPEED_MSK_40GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_40GB #define BNXT_LINK_SPEED_MSK_50GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_50GB + u16 support_auto_speeds; u16 lp_auto_link_speeds; u16 force_link_speed; u32 preemphasis; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 89050ed..3362e90 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -659,6 +659,17 @@ static u32 bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info) return supported | SUPPORTED_Pause | SUPPORTED_Asym_Pause; } +static u32 bnxt_fw_to_ethtool_support_adv_spds(struct bnxt_link_info *link_info) +{ + u16 fw_speeds = link_info->support_auto_speeds; + u32 supported; + + supported = _bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0); + if (supported) + supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause; + return supported; +} + u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed) { switch (fw_link_speed) { @@ -691,7 +702,7 @@ static int bnxt_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) cmd->supported = bnxt_fw_to_ethtool_support_spds(link_info); - if (link_info->auto_link_speeds) + if (link_info->support_auto_speeds) cmd->supported |= SUPPORTED_Autoneg; if (link_info->autoneg) { @@ -827,8 +838,14 @@ static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) return rc; if (cmd->autoneg == AUTONEG_ENABLE) { - u32 supported_spds = bnxt_fw_to_ethtool_support_spds(link_info); + u32 supported_spds = + bnxt_fw_to_ethtool_support_adv_spds(link_info); + if (!supported_spds) { + netdev_err(dev, "Autoneg not supported\n"); + rc = -EINVAL; + goto set_setting_exit; + } if (cmd->advertising & ~(supported_spds | ADVERTISED_Autoneg | ADVERTISED_TP | ADVERTISED_FIBRE)) { netdev_err(dev, "Unsupported advertising mask (adv: 0x%x)\n", @@ -837,15 +854,9 @@ static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) goto set_setting_exit; } fw_advertising = bnxt_get_fw_auto_link_speeds(cmd->advertising); - if (fw_advertising & ~link_info->support_speeds) { - netdev_err(dev, "Advertising parameters are not supported! (adv: 0x%x)\n", - cmd->advertising); - rc = -EINVAL; - goto set_setting_exit; - } link_info->autoneg |= BNXT_AUTONEG_SPEED; if (!fw_advertising) - link_info->advertising = link_info->support_speeds; + link_info->advertising = link_info->support_auto_speeds; else link_info->advertising = fw_advertising; /* any change to autoneg will cause link change, therefore the -- cgit v0.10.2 From 00c04a928572991d30b2473a7e992c1be8e646f3 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 13 Jun 2016 02:25:38 -0400 Subject: bnxt_en: Support new ETHTOOL_{G|S}LINKSETTINGS API. To fully support 25G and 50G link settings. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 3362e90..d7ab2d79 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -628,7 +628,66 @@ u32 _bnxt_fw_to_ethtool_adv_spds(u16 fw_speeds, u8 fw_pause) return speed_mask; } -static u32 bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info) +#define BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings, name)\ +{ \ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_100MB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 100baseT_Full); \ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_1GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 1000baseT_Full); \ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_10GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 10000baseT_Full); \ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_25GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 25000baseCR_Full); \ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_40GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 40000baseCR4_Full);\ + if ((fw_speeds) & BNXT_LINK_SPEED_MSK_50GB) \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + 50000baseCR2_Full);\ + if ((fw_pause) & BNXT_LINK_PAUSE_RX) { \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + Pause); \ + if (!((fw_pause) & BNXT_LINK_PAUSE_TX)) \ + ethtool_link_ksettings_add_link_mode( \ + lk_ksettings, name, Asym_Pause);\ + } else if ((fw_pause) & BNXT_LINK_PAUSE_TX) { \ + ethtool_link_ksettings_add_link_mode(lk_ksettings, name,\ + Asym_Pause); \ + } \ +} + +#define BNXT_ETHTOOL_TO_FW_SPDS(fw_speeds, lk_ksettings, name) \ +{ \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 100baseT_Full) || \ + ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 100baseT_Half)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_100MB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 1000baseT_Full) || \ + ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 1000baseT_Half)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_1GB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 10000baseT_Full)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_10GB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 25000baseCR_Full)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_25GB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 40000baseCR4_Full)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_40GB; \ + if (ethtool_link_ksettings_test_link_mode(lk_ksettings, name, \ + 50000baseCR2_Full)) \ + (fw_speeds) |= BNXT_LINK_SPEED_MSK_50GB; \ +} + +static void bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info, + struct ethtool_link_ksettings *lk_ksettings) { u16 fw_speeds = link_info->auto_link_speeds; u8 fw_pause = 0; @@ -636,10 +695,11 @@ static u32 bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info) if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL) fw_pause = link_info->auto_pause_setting; - return _bnxt_fw_to_ethtool_adv_spds(fw_speeds, fw_pause); + BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings, advertising); } -static u32 bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info) +static void bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info, + struct ethtool_link_ksettings *lk_ksettings) { u16 fw_speeds = link_info->lp_auto_link_speeds; u8 fw_pause = 0; @@ -647,27 +707,24 @@ static u32 bnxt_fw_to_ethtool_lp_adv(struct bnxt_link_info *link_info) if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL) fw_pause = link_info->lp_pause; - return _bnxt_fw_to_ethtool_adv_spds(fw_speeds, fw_pause); + BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, fw_pause, lk_ksettings, + lp_advertising); } -static u32 bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info) +static void bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info, + struct ethtool_link_ksettings *lk_ksettings) { u16 fw_speeds = link_info->support_speeds; - u32 supported; - supported = _bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0); - return supported | SUPPORTED_Pause | SUPPORTED_Asym_Pause; -} + BNXT_FW_TO_ETHTOOL_SPDS(fw_speeds, 0, lk_ksettings, supported); -static u32 bnxt_fw_to_ethtool_support_adv_spds(struct bnxt_link_info *link_info) -{ - u16 fw_speeds = link_info->support_auto_speeds; - u32 supported; + ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, Pause); + ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, + Asym_Pause); - supported = _bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0); - if (supported) - supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause; - return supported; + if (link_info->support_auto_speeds) + ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, + Autoneg); } u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed) @@ -694,65 +751,62 @@ u32 bnxt_fw_to_ethtool_speed(u16 fw_link_speed) } } -static int bnxt_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) +static int bnxt_get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings *lk_ksettings) { struct bnxt *bp = netdev_priv(dev); struct bnxt_link_info *link_info = &bp->link_info; - u16 ethtool_speed; - - cmd->supported = bnxt_fw_to_ethtool_support_spds(link_info); + struct ethtool_link_settings *base = &lk_ksettings->base; + u32 ethtool_speed; - if (link_info->support_auto_speeds) - cmd->supported |= SUPPORTED_Autoneg; + ethtool_link_ksettings_zero_link_mode(lk_ksettings, supported); + bnxt_fw_to_ethtool_support_spds(link_info, lk_ksettings); + ethtool_link_ksettings_zero_link_mode(lk_ksettings, advertising); if (link_info->autoneg) { - cmd->advertising = - bnxt_fw_to_ethtool_advertised_spds(link_info); - cmd->advertising |= ADVERTISED_Autoneg; - cmd->autoneg = AUTONEG_ENABLE; + bnxt_fw_to_ethtool_advertised_spds(link_info, lk_ksettings); + ethtool_link_ksettings_add_link_mode(lk_ksettings, + advertising, Autoneg); + base->autoneg = AUTONEG_ENABLE; if (link_info->phy_link_status == BNXT_LINK_LINK) - cmd->lp_advertising = - bnxt_fw_to_ethtool_lp_adv(link_info); + bnxt_fw_to_ethtool_lp_adv(link_info, lk_ksettings); ethtool_speed = bnxt_fw_to_ethtool_speed(link_info->link_speed); if (!netif_carrier_ok(dev)) - cmd->duplex = DUPLEX_UNKNOWN; + base->duplex = DUPLEX_UNKNOWN; else if (link_info->duplex & BNXT_LINK_DUPLEX_FULL) - cmd->duplex = DUPLEX_FULL; + base->duplex = DUPLEX_FULL; else - cmd->duplex = DUPLEX_HALF; + base->duplex = DUPLEX_HALF; } else { - cmd->autoneg = AUTONEG_DISABLE; - cmd->advertising = 0; + base->autoneg = AUTONEG_DISABLE; ethtool_speed = bnxt_fw_to_ethtool_speed(link_info->req_link_speed); - cmd->duplex = DUPLEX_HALF; + base->duplex = DUPLEX_HALF; if (link_info->req_duplex == BNXT_LINK_DUPLEX_FULL) - cmd->duplex = DUPLEX_FULL; + base->duplex = DUPLEX_FULL; } - ethtool_cmd_speed_set(cmd, ethtool_speed); + base->speed = ethtool_speed; - cmd->port = PORT_NONE; + base->port = PORT_NONE; if (link_info->media_type == PORT_PHY_QCFG_RESP_MEDIA_TYPE_TP) { - cmd->port = PORT_TP; - cmd->supported |= SUPPORTED_TP; - cmd->advertising |= ADVERTISED_TP; + base->port = PORT_TP; + ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, + TP); + ethtool_link_ksettings_add_link_mode(lk_ksettings, advertising, + TP); } else { - cmd->supported |= SUPPORTED_FIBRE; - cmd->advertising |= ADVERTISED_FIBRE; + ethtool_link_ksettings_add_link_mode(lk_ksettings, supported, + FIBRE); + ethtool_link_ksettings_add_link_mode(lk_ksettings, advertising, + FIBRE); if (link_info->media_type == PORT_PHY_QCFG_RESP_MEDIA_TYPE_DAC) - cmd->port = PORT_DA; + base->port = PORT_DA; else if (link_info->media_type == PORT_PHY_QCFG_RESP_MEDIA_TYPE_FIBRE) - cmd->port = PORT_FIBRE; + base->port = PORT_FIBRE; } - - if (link_info->transceiver == - PORT_PHY_QCFG_RESP_XCVR_PKG_TYPE_XCVR_INTERNAL) - cmd->transceiver = XCVR_INTERNAL; - else - cmd->transceiver = XCVR_EXTERNAL; - cmd->phy_address = link_info->phy_addr; + base->phy_address = link_info->phy_addr; return 0; } @@ -826,34 +880,22 @@ u16 bnxt_get_fw_auto_link_speeds(u32 advertising) return fw_speed_mask; } -static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) +static int bnxt_set_link_ksettings(struct net_device *dev, + const struct ethtool_link_ksettings *lk_ksettings) { - int rc = 0; struct bnxt *bp = netdev_priv(dev); struct bnxt_link_info *link_info = &bp->link_info; + const struct ethtool_link_settings *base = &lk_ksettings->base; u32 speed, fw_advertising = 0; bool set_pause = false; + int rc = 0; if (!BNXT_SINGLE_PF(bp)) - return rc; - - if (cmd->autoneg == AUTONEG_ENABLE) { - u32 supported_spds = - bnxt_fw_to_ethtool_support_adv_spds(link_info); + return -EOPNOTSUPP; - if (!supported_spds) { - netdev_err(dev, "Autoneg not supported\n"); - rc = -EINVAL; - goto set_setting_exit; - } - if (cmd->advertising & ~(supported_spds | ADVERTISED_Autoneg | - ADVERTISED_TP | ADVERTISED_FIBRE)) { - netdev_err(dev, "Unsupported advertising mask (adv: 0x%x)\n", - cmd->advertising); - rc = -EINVAL; - goto set_setting_exit; - } - fw_advertising = bnxt_get_fw_auto_link_speeds(cmd->advertising); + if (base->autoneg == AUTONEG_ENABLE) { + BNXT_ETHTOOL_TO_FW_SPDS(fw_advertising, lk_ksettings, + advertising); link_info->autoneg |= BNXT_AUTONEG_SPEED; if (!fw_advertising) link_info->advertising = link_info->support_auto_speeds; @@ -874,16 +916,12 @@ static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) rc = -EINVAL; goto set_setting_exit; } - /* TODO: currently don't support half duplex */ - if (cmd->duplex == DUPLEX_HALF) { + if (base->duplex == DUPLEX_HALF) { netdev_err(dev, "HALF DUPLEX is not supported!\n"); rc = -EINVAL; goto set_setting_exit; } - /* If received a request for an unknown duplex, assume full*/ - if (cmd->duplex == DUPLEX_UNKNOWN) - cmd->duplex = DUPLEX_FULL; - speed = ethtool_cmd_speed(cmd); + speed = base->speed; fw_speed = bnxt_get_fw_speed(dev, speed); if (!fw_speed) { rc = -EINVAL; @@ -1629,8 +1667,8 @@ static int bnxt_get_module_eeprom(struct net_device *dev, } const struct ethtool_ops bnxt_ethtool_ops = { - .get_settings = bnxt_get_settings, - .set_settings = bnxt_set_settings, + .get_link_ksettings = bnxt_get_link_ksettings, + .set_link_ksettings = bnxt_set_link_ksettings, .get_pauseparam = bnxt_get_pauseparam, .set_pauseparam = bnxt_set_pauseparam, .get_drvinfo = bnxt_get_drvinfo, -- cgit v0.10.2 From e8eb36cd8ca93f52f738c6087073202c44ac7746 Mon Sep 17 00:00:00 2001 From: Amir Vadai Date: Mon, 13 Jun 2016 12:06:39 +0300 Subject: net/sched: flower: Return error when hw can't offload and skip_sw is set When skip_sw is set and hardware fails to apply filter, return error to user. This will make error propagation logic similar to the one currently used in u32 classifier. Also, changed code to use tc_skip_sw() utility function. Signed-off-by: Amir Vadai Acked-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 1ea6f76..5060801 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -140,7 +140,7 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp, f = rhashtable_lookup_fast(&head->ht, fl_key_get_start(&skb_mkey, &head->mask), head->ht_params); - if (f && !(f->flags & TCA_CLS_FLAGS_SKIP_SW)) { + if (f && !tc_skip_sw(f->flags)) { *res = f->res; return tcf_exts_exec(skb, &f->exts, res); } @@ -187,19 +187,20 @@ static void fl_hw_destroy_filter(struct tcf_proto *tp, unsigned long cookie) dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, &tc); } -static void fl_hw_replace_filter(struct tcf_proto *tp, - struct flow_dissector *dissector, - struct fl_flow_key *mask, - struct fl_flow_key *key, - struct tcf_exts *actions, - unsigned long cookie, u32 flags) +static int fl_hw_replace_filter(struct tcf_proto *tp, + struct flow_dissector *dissector, + struct fl_flow_key *mask, + struct fl_flow_key *key, + struct tcf_exts *actions, + unsigned long cookie, u32 flags) { struct net_device *dev = tp->q->dev_queue->dev; struct tc_cls_flower_offload offload = {0}; struct tc_to_netdev tc; + int err; if (!tc_should_offload(dev, tp, flags)) - return; + return tc_skip_sw(flags) ? -EINVAL : 0; offload.command = TC_CLSFLOWER_REPLACE; offload.cookie = cookie; @@ -211,7 +212,12 @@ static void fl_hw_replace_filter(struct tcf_proto *tp, tc.type = TC_SETUP_CLSFLOWER; tc.cls_flower = &offload; - dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, &tc); + err = dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, &tc); + + if (tc_skip_sw(flags)) + return err; + + return 0; } static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f) @@ -572,20 +578,22 @@ static int fl_change(struct net *net, struct sk_buff *in_skb, if (err) goto errout; - if (!(fnew->flags & TCA_CLS_FLAGS_SKIP_SW)) { + if (!tc_skip_sw(fnew->flags)) { err = rhashtable_insert_fast(&head->ht, &fnew->ht_node, head->ht_params); if (err) goto errout; } - fl_hw_replace_filter(tp, - &head->dissector, - &mask.key, - &fnew->key, - &fnew->exts, - (unsigned long)fnew, - fnew->flags); + err = fl_hw_replace_filter(tp, + &head->dissector, + &mask.key, + &fnew->key, + &fnew->exts, + (unsigned long)fnew, + fnew->flags); + if (err) + goto errout; if (fold) { rhashtable_remove_fast(&head->ht, &fold->ht_node, -- cgit v0.10.2 From a028a9e003f299cf427d98501861bbbb4f5ba7a5 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Mon, 13 Jun 2016 17:49:36 +0800 Subject: r8152: move the settings of PHY to a work queue Move the settings of PHY to a work queue and schedule it after rtl_ops.init(). There are some reasons for this. First, the settings are only needed for the first time initialization or after the power down occurs. Second, the settings are independent with the others. Last, the settings may take more time than the others. Leave they in probe() or open() may delay the following flows. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 161c25e..46fe9a7 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -602,7 +602,7 @@ struct r8152 { struct list_head rx_done, tx_free; struct sk_buff_head tx_queue, rx_queue; spinlock_t rx_lock, tx_lock; - struct delayed_work schedule; + struct delayed_work schedule, hw_phy_work; struct mii_if_info mii; struct mutex control; /* use for hw setting */ #ifdef CONFIG_PM_SLEEP @@ -619,6 +619,7 @@ struct r8152 { int (*eee_get)(struct r8152 *, struct ethtool_eee *); int (*eee_set)(struct r8152 *, struct ethtool_eee *); bool (*in_nway)(struct r8152 *); + void (*hw_phy_cfg)(struct r8152 *); } rtl_ops; int intr_interval; @@ -2499,8 +2500,6 @@ static void r8152b_exit_oob(struct r8152 *tp) rxdy_gated_en(tp, true); r8153_teredo_off(tp); - r8152b_hw_phy_cfg(tp); - ocp_write_byte(tp, MCU_TYPE_PLA, PLA_CRWECR, CRWECR_NORAML); ocp_write_byte(tp, MCU_TYPE_PLA, PLA_CR, 0x00); @@ -2678,8 +2677,6 @@ static void r8153_first_init(struct r8152 *tp) ocp_data &= ~RCR_ACPT_ALL; ocp_write_dword(tp, MCU_TYPE_PLA, PLA_RCR, ocp_data); - r8153_hw_phy_cfg(tp); - rtl8152_nic_reset(tp); ocp_data = ocp_read_byte(tp, MCU_TYPE_PLA, PLA_OOB_CTRL); @@ -3040,6 +3037,25 @@ out1: usb_autopm_put_interface(tp->intf); } +static void rtl_hw_phy_work_func_t(struct work_struct *work) +{ + struct r8152 *tp = container_of(work, struct r8152, hw_phy_work.work); + + if (test_bit(RTL8152_UNPLUG, &tp->flags)) + return; + + if (usb_autopm_get_interface(tp->intf) < 0) + return; + + mutex_lock(&tp->control); + + tp->rtl_ops.hw_phy_cfg(tp); + + mutex_unlock(&tp->control); + + usb_autopm_put_interface(tp->intf); +} + #ifdef CONFIG_PM_SLEEP static int rtl_notifier(struct notifier_block *nb, unsigned long action, void *data) @@ -3518,6 +3534,7 @@ static int rtl8152_resume(struct usb_interface *intf) if (!test_bit(SELECTIVE_SUSPEND, &tp->flags)) { tp->rtl_ops.init(tp); + queue_delayed_work(system_long_wq, &tp->hw_phy_work, 0); netif_device_attach(tp->netdev); } @@ -4122,6 +4139,7 @@ static int rtl_ops_init(struct r8152 *tp) ops->eee_get = r8152_get_eee; ops->eee_set = r8152_set_eee; ops->in_nway = rtl8152_in_nway; + ops->hw_phy_cfg = r8152b_hw_phy_cfg; break; case RTL_VER_03: @@ -4137,6 +4155,7 @@ static int rtl_ops_init(struct r8152 *tp) ops->eee_get = r8153_get_eee; ops->eee_set = r8153_set_eee; ops->in_nway = rtl8153_in_nway; + ops->hw_phy_cfg = r8153_hw_phy_cfg; break; default: @@ -4183,6 +4202,7 @@ static int rtl8152_probe(struct usb_interface *intf, mutex_init(&tp->control); INIT_DELAYED_WORK(&tp->schedule, rtl_work_func_t); + INIT_DELAYED_WORK(&tp->hw_phy_work, rtl_hw_phy_work_func_t); netdev->netdev_ops = &rtl8152_netdev_ops; netdev->watchdog_timeo = RTL8152_TX_TIMEOUT; @@ -4225,6 +4245,7 @@ static int rtl8152_probe(struct usb_interface *intf, intf->needs_remote_wakeup = 1; tp->rtl_ops.init(tp); + queue_delayed_work(system_long_wq, &tp->hw_phy_work, 0); set_ethernet_addr(tp); usb_set_intfdata(intf, tp); @@ -4270,6 +4291,7 @@ static void rtl8152_disconnect(struct usb_interface *intf) netif_napi_del(&tp->napi); unregister_netdev(tp->netdev); + cancel_delayed_work_sync(&tp->hw_phy_work); tp->rtl_ops.unload(tp); free_netdev(tp->netdev); } -- cgit v0.10.2 From 9d21c0d83ea3def89b9cd33222b1c44c1af44b4d Mon Sep 17 00:00:00 2001 From: hayeswang Date: Mon, 13 Jun 2016 17:49:37 +0800 Subject: r8152: move the setting for the default speed Move calling set_speed() from open() to rtl_hw_phy_work_func_t(). Then, we would set the default speed only for first initialization or after resuming. Besides, the set_speed() could handle the flag of PHY_RESET which would be set in rtl_ops.hw_phy_cfg(). Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 46fe9a7..dcdc4fa 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -3051,6 +3051,10 @@ static void rtl_hw_phy_work_func_t(struct work_struct *work) tp->rtl_ops.hw_phy_cfg(tp); + rtl8152_set_speed(tp, AUTONEG_ENABLE, + tp->mii.supports_gmii ? SPEED_1000 : SPEED_100, + DUPLEX_FULL); + mutex_unlock(&tp->control); usb_autopm_put_interface(tp->intf); @@ -3104,9 +3108,6 @@ static int rtl8152_open(struct net_device *netdev) tp->rtl_ops.up(tp); - rtl8152_set_speed(tp, AUTONEG_ENABLE, - tp->mii.supports_gmii ? SPEED_1000 : SPEED_100, - DUPLEX_FULL); netif_carrier_off(netdev); netif_start_queue(netdev); set_bit(WORK_ENABLE, &tp->flags); @@ -3549,10 +3550,6 @@ static int rtl8152_resume(struct usb_interface *intf) napi_enable(&tp->napi); } else { tp->rtl_ops.up(tp); - rtl8152_set_speed(tp, AUTONEG_ENABLE, - tp->mii.supports_gmii ? - SPEED_1000 : SPEED_100, - DUPLEX_FULL); netif_carrier_off(tp->netdev); set_bit(WORK_ENABLE, &tp->flags); } -- cgit v0.10.2 From aa7e26b66ad7e38371ca5a575f6581c645eef58a Mon Sep 17 00:00:00 2001 From: hayeswang Date: Mon, 13 Jun 2016 17:49:38 +0800 Subject: r8152: save the speed The user may change the speed. Use it to replace the default one. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index dcdc4fa..69d1bbfd 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -628,8 +628,11 @@ struct r8152 { u32 tx_qlen; u32 coalesce; u16 ocp_base; + u16 speed; u8 *intr_buff; u8 version; + u8 duplex; + u8 autoneg; }; enum rtl_version { @@ -3051,9 +3054,7 @@ static void rtl_hw_phy_work_func_t(struct work_struct *work) tp->rtl_ops.hw_phy_cfg(tp); - rtl8152_set_speed(tp, AUTONEG_ENABLE, - tp->mii.supports_gmii ? SPEED_1000 : SPEED_100, - DUPLEX_FULL); + rtl8152_set_speed(tp, tp->autoneg, tp->speed, tp->duplex); mutex_unlock(&tp->control); @@ -3679,6 +3680,11 @@ static int rtl8152_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) mutex_lock(&tp->control); ret = rtl8152_set_speed(tp, cmd->autoneg, cmd->speed, cmd->duplex); + if (!ret) { + tp->autoneg = cmd->autoneg; + tp->speed = cmd->speed; + tp->duplex = cmd->duplex; + } mutex_unlock(&tp->control); @@ -4239,6 +4245,10 @@ static int rtl8152_probe(struct usb_interface *intf, break; } + tp->autoneg = AUTONEG_ENABLE; + tp->speed = tp->mii.supports_gmii ? SPEED_1000 : SPEED_100; + tp->duplex = DUPLEX_FULL; + intf->needs_remote_wakeup = 1; tp->rtl_ops.init(tp); -- cgit v0.10.2 From 4a63538ef1ccfec7716e255b67b57080d11fa670 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Mon, 13 Jun 2016 16:56:20 +0800 Subject: net: hns: update the dependency After the patchset about adding support of ACPI (commit id is 6343488) being applied, HNS does not depend on OF. It depends on OF or ACPI, so the Kconfig file needs to be updated. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/Kconfig b/drivers/net/ethernet/hisilicon/Kconfig index 4ccc032..2e25662 100644 --- a/drivers/net/ethernet/hisilicon/Kconfig +++ b/drivers/net/ethernet/hisilicon/Kconfig @@ -5,7 +5,7 @@ config NET_VENDOR_HISILICON bool "Hisilicon devices" default y - depends on OF && HAS_DMA + depends on (OF || ACPI) && HAS_DMA depends on ARM || ARM64 || COMPILE_TEST ---help--- If you have a network (Ethernet) card belonging to this class, say Y. -- cgit v0.10.2 From a5e27d18fe64561a467b706f70cfc89ba6323f87 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Mon, 13 Jun 2016 23:08:26 +0800 Subject: sctp: fix error return code in sctp_init() Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function. Signed-off-by: Wei Yongjun Acked-by: Xin Long Acked-by: Neil Horman Signed-off-by: David S. Miller diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 40022ee..3b56ae5 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1479,7 +1479,8 @@ static __init int sctp_init(void) INIT_HLIST_HEAD(&sctp_port_hashtable[i].chain); } - if (sctp_transport_hashtable_init()) + status = sctp_transport_hashtable_init(); + if (status) goto err_thash_alloc; pr_info("Hash tables configured (bind %d/%d)\n", sctp_port_hashsize, -- cgit v0.10.2 From dcf1158b275f9d51d6a742cf7166edc764ee4718 Mon Sep 17 00:00:00 2001 From: Neal Cardwell Date: Mon, 13 Jun 2016 11:20:35 -0400 Subject: tcp: return sizeof tcp_dctcp_info in dctcp_get_info() Make sure that dctcp_get_info() returns only the size of the info->dctcp struct that it zeroes out and fills in. Previously it had been returning the size of the enclosing tcp_cc_info union, sizeof(*info). There is no problem yet, but that union that may one day be larger than struct tcp_dctcp_info, in which case the TCP_CC_INFO code might accidentally copy uninitialized bytes from the stack. Signed-off-by: Neal Cardwell Signed-off-by: Soheil Hassas Yeganeh Signed-off-by: Eric Dumazet Acked-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp_dctcp.c index 7e538f7..10d728b 100644 --- a/net/ipv4/tcp_dctcp.c +++ b/net/ipv4/tcp_dctcp.c @@ -293,7 +293,7 @@ static size_t dctcp_get_info(struct sock *sk, u32 ext, int *attr, */ if (ext & (1 << (INET_DIAG_DCTCPINFO - 1)) || ext & (1 << (INET_DIAG_VEGASINFO - 1))) { - memset(info, 0, sizeof(struct tcp_dctcp_info)); + memset(&info->dctcp, 0, sizeof(info->dctcp)); if (inet_csk(sk)->icsk_ca_ops != &dctcp_reno) { info->dctcp.dctcp_enabled = 1; info->dctcp.dctcp_ce_state = (u16) ca->ce_state; @@ -303,7 +303,7 @@ static size_t dctcp_get_info(struct sock *sk, u32 ext, int *attr, } *attr = INET_DIAG_DCTCPINFO; - return sizeof(*info); + return sizeof(info->dctcp); } return 0; } -- cgit v0.10.2 From 0cb43965d42a21a7af41f88f1021b478dc102425 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:26 -0700 Subject: RDS: split out connection specific state from rds_connection to rds_conn_path In preparation for multipath RDS, split the rds_connection structure into a base structure, and a per-path struct rds_conn_path. The base structure tracks information and locks common to all paths. The workqs for send/recv/shutdown etc are tracked per rds_conn_path. Thus the workq callbacks now work with rds_conn_path. This commit allows for one rds_conn_path per rds_connection, and will be extended into multiple conn_paths in subsequent commits. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/cong.c b/net/rds/cong.c index 6641bcf..8398fee 100644 --- a/net/rds/cong.c +++ b/net/rds/cong.c @@ -235,7 +235,8 @@ void rds_cong_queue_updates(struct rds_cong_map *map) * therefore trigger warnings. * Defer the xmit to rds_send_worker() instead. */ - queue_delayed_work(rds_wq, &conn->c_send_w, 0); + queue_delayed_work(rds_wq, + &conn->c_path[0].cp_send_w, 0); } } diff --git a/net/rds/connection.c b/net/rds/connection.c index e3b118c..6fa2074 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -36,6 +36,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "loop.h" @@ -155,6 +156,7 @@ static struct rds_connection *__rds_conn_create(struct net *net, conn->c_faddr = faddr; spin_lock_init(&conn->c_lock); conn->c_next_tx_seq = 1; + conn->c_path[0].cp_conn = conn; rds_conn_net_set(conn, net); init_waitqueue_head(&conn->c_waitq); @@ -197,7 +199,7 @@ static struct rds_connection *__rds_conn_create(struct net *net, atomic_set(&conn->c_state, RDS_CONN_DOWN); conn->c_send_gen = 0; - conn->c_outgoing = (is_outgoing ? 1 : 0); + conn->c_path[0].cp_outgoing = (is_outgoing ? 1 : 0); conn->c_reconnect_jiffies = 0; INIT_DELAYED_WORK(&conn->c_send_w, rds_send_worker); INIT_DELAYED_WORK(&conn->c_recv_w, rds_recv_worker); @@ -320,8 +322,8 @@ void rds_conn_shutdown(struct rds_connection *conn) if (!hlist_unhashed(&conn->c_hash_node)) { rcu_read_unlock(); if (conn->c_trans->t_type != RDS_TRANS_TCP || - conn->c_outgoing == 1) - rds_queue_reconnect(conn); + conn->c_path[0].cp_outgoing == 1) + rds_queue_reconnect(&conn->c_path[0]); } else { rcu_read_unlock(); } @@ -553,10 +555,16 @@ void rds_conn_exit(void) /* * Force a disconnect */ +void rds_conn_path_drop(struct rds_conn_path *cp) +{ + atomic_set(&cp->cp_state, RDS_CONN_ERROR); + queue_work(rds_wq, &cp->cp_down_w); +} +EXPORT_SYMBOL_GPL(rds_conn_path_drop); + void rds_conn_drop(struct rds_connection *conn) { - atomic_set(&conn->c_state, RDS_CONN_ERROR); - queue_work(rds_wq, &conn->c_down_w); + rds_conn_path_drop(&conn->c_path[0]); } EXPORT_SYMBOL_GPL(rds_conn_drop); diff --git a/net/rds/ib.c b/net/rds/ib.c index b5342fd..44946a6 100644 --- a/net/rds/ib.c +++ b/net/rds/ib.c @@ -40,6 +40,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "ib.h" #include "ib_mr.h" diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c index 310cabc..4de5a35 100644 --- a/net/rds/ib_cm.c +++ b/net/rds/ib_cm.c @@ -36,6 +36,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "ib.h" diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c index a0f21b6..977f698 100644 --- a/net/rds/ib_rdma.c +++ b/net/rds/ib_rdma.c @@ -35,6 +35,7 @@ #include #include +#include "rds_single_path.h" #include "ib_mr.h" struct workqueue_struct *rds_ib_mr_wq; diff --git a/net/rds/ib_recv.c b/net/rds/ib_recv.c index abc8cc8..4ea8cb1 100644 --- a/net/rds/ib_recv.c +++ b/net/rds/ib_recv.c @@ -36,6 +36,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "ib.h" diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c index f27d2c8..6e4110a 100644 --- a/net/rds/ib_send.c +++ b/net/rds/ib_send.c @@ -36,6 +36,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "ib.h" diff --git a/net/rds/loop.c b/net/rds/loop.c index 6b12b68..268f07f 100644 --- a/net/rds/loop.c +++ b/net/rds/loop.c @@ -34,6 +34,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "loop.h" diff --git a/net/rds/rdma_transport.c b/net/rds/rdma_transport.c index 7220beb..345f090 100644 --- a/net/rds/rdma_transport.c +++ b/net/rds/rdma_transport.c @@ -33,6 +33,7 @@ #include #include +#include "rds_single_path.h" #include "rdma_transport.h" #include "ib.h" diff --git a/net/rds/rds.h b/net/rds/rds.h index 387df5f..ca31a07 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -84,56 +84,69 @@ enum { #define RDS_IN_XMIT 2 #define RDS_RECV_REFILL 3 +/* Max number of multipaths per RDS connection. Must be a power of 2 */ +#define RDS_MPATH_WORKERS 1 + +/* Per mpath connection state */ +struct rds_conn_path { + struct rds_connection *cp_conn; + struct rds_message *cp_xmit_rm; + unsigned long cp_xmit_sg; + unsigned int cp_xmit_hdr_off; + unsigned int cp_xmit_data_off; + unsigned int cp_xmit_atomic_sent; + unsigned int cp_xmit_rdma_sent; + unsigned int cp_xmit_data_sent; + + spinlock_t cp_lock; /* protect msg queues */ + u64 cp_next_tx_seq; + struct list_head cp_send_queue; + struct list_head cp_retrans; + + u64 cp_next_rx_seq; + + void *cp_transport_data; + + atomic_t cp_state; + unsigned long cp_send_gen; + unsigned long cp_flags; + unsigned long cp_reconnect_jiffies; + struct delayed_work cp_send_w; + struct delayed_work cp_recv_w; + struct delayed_work cp_conn_w; + struct work_struct cp_down_w; + struct mutex cp_cm_lock; /* protect cp_state & cm */ + wait_queue_head_t cp_waitq; + + unsigned int cp_unacked_packets; + unsigned int cp_unacked_bytes; + unsigned int cp_outgoing:1, + cp_pad_to_32:31; + unsigned int cp_index; +}; + +/* One rds_connection per RDS address pair */ struct rds_connection { struct hlist_node c_hash_node; __be32 c_laddr; __be32 c_faddr; unsigned int c_loopback:1, - c_outgoing:1, - c_pad_to_32:30; + c_pad_to_32:31; + int c_npaths; struct rds_connection *c_passive; + struct rds_transport *c_trans; struct rds_cong_map *c_lcong; struct rds_cong_map *c_fcong; - struct rds_message *c_xmit_rm; - unsigned long c_xmit_sg; - unsigned int c_xmit_hdr_off; - unsigned int c_xmit_data_off; - unsigned int c_xmit_atomic_sent; - unsigned int c_xmit_rdma_sent; - unsigned int c_xmit_data_sent; - - spinlock_t c_lock; /* protect msg queues */ - u64 c_next_tx_seq; - struct list_head c_send_queue; - struct list_head c_retrans; - - u64 c_next_rx_seq; - - struct rds_transport *c_trans; - void *c_transport_data; - - atomic_t c_state; - unsigned long c_send_gen; - unsigned long c_flags; - unsigned long c_reconnect_jiffies; - struct delayed_work c_send_w; - struct delayed_work c_recv_w; - struct delayed_work c_conn_w; - struct work_struct c_down_w; - struct mutex c_cm_lock; /* protect conn state & cm */ - wait_queue_head_t c_waitq; + /* Protocol version */ + unsigned int c_version; + possible_net_t c_net; struct list_head c_map_item; unsigned long c_map_queued; - unsigned int c_unacked_packets; - unsigned int c_unacked_bytes; - - /* Protocol version */ - unsigned int c_version; - possible_net_t c_net; + struct rds_conn_path c_path[RDS_MPATH_WORKERS]; }; static inline @@ -639,6 +652,7 @@ struct rds_connection *rds_conn_create_outgoing(struct net *net, void rds_conn_shutdown(struct rds_connection *conn); void rds_conn_destroy(struct rds_connection *conn); void rds_conn_drop(struct rds_connection *conn); +void rds_conn_path_drop(struct rds_conn_path *cpath); void rds_conn_connect_if_down(struct rds_connection *conn); void rds_for_each_conn_info(struct socket *sock, unsigned int len, struct rds_info_iterator *iter, @@ -651,27 +665,51 @@ void __rds_conn_error(struct rds_connection *conn, const char *, ...); __rds_conn_error(conn, KERN_WARNING "RDS: " fmt) static inline int +rds_conn_path_transition(struct rds_conn_path *cp, int old, int new) +{ + return atomic_cmpxchg(&cp->cp_state, old, new) == old; +} + +static inline int rds_conn_transition(struct rds_connection *conn, int old, int new) { - return atomic_cmpxchg(&conn->c_state, old, new) == old; + return rds_conn_path_transition(&conn->c_path[0], old, new); +} + +static inline int +rds_conn_path_state(struct rds_conn_path *cp) +{ + return atomic_read(&cp->cp_state); } static inline int rds_conn_state(struct rds_connection *conn) { - return atomic_read(&conn->c_state); + return rds_conn_path_state(&conn->c_path[0]); +} + +static inline int +rds_conn_path_up(struct rds_conn_path *cp) +{ + return atomic_read(&cp->cp_state) == RDS_CONN_UP; } static inline int rds_conn_up(struct rds_connection *conn) { - return atomic_read(&conn->c_state) == RDS_CONN_UP; + return rds_conn_path_up(&conn->c_path[0]); +} + +static inline int +rds_conn_path_connecting(struct rds_conn_path *cp) +{ + return atomic_read(&cp->cp_state) == RDS_CONN_CONNECTING; } static inline int rds_conn_connecting(struct rds_connection *conn) { - return atomic_read(&conn->c_state) == RDS_CONN_CONNECTING; + return rds_conn_path_connecting(&conn->c_path[0]); } /* message.c */ @@ -809,12 +847,12 @@ extern unsigned int rds_sysctl_trace_level; int rds_threads_init(void); void rds_threads_exit(void); extern struct workqueue_struct *rds_wq; -void rds_queue_reconnect(struct rds_connection *conn); +void rds_queue_reconnect(struct rds_conn_path *cp); void rds_connect_worker(struct work_struct *); void rds_shutdown_worker(struct work_struct *); void rds_send_worker(struct work_struct *); void rds_recv_worker(struct work_struct *); -void rds_connect_path_complete(struct rds_connection *conn, int curr); +void rds_connect_path_complete(struct rds_conn_path *conn, int curr); void rds_connect_complete(struct rds_connection *conn); /* transport.c */ diff --git a/net/rds/rds_single_path.h b/net/rds/rds_single_path.h new file mode 100644 index 0000000..e1241af --- /dev/null +++ b/net/rds/rds_single_path.h @@ -0,0 +1,30 @@ +#ifndef _RDS_RDS_SINGLE_H +#define _RDS_RDS_SINGLE_H + +#define c_xmit_rm c_path[0].cp_xmit_rm +#define c_xmit_sg c_path[0].cp_xmit_sg +#define c_xmit_hdr_off c_path[0].cp_xmit_hdr_off +#define c_xmit_data_off c_path[0].cp_xmit_data_off +#define c_xmit_atomic_sent c_path[0].cp_xmit_atomic_sent +#define c_xmit_rdma_sent c_path[0].cp_xmit_rdma_sent +#define c_xmit_data_sent c_path[0].cp_xmit_data_sent +#define c_lock c_path[0].cp_lock +#define c_next_tx_seq c_path[0].cp_next_tx_seq +#define c_send_queue c_path[0].cp_send_queue +#define c_retrans c_path[0].cp_retrans +#define c_next_rx_seq c_path[0].cp_next_rx_seq +#define c_transport_data c_path[0].cp_transport_data +#define c_state c_path[0].cp_state +#define c_send_gen c_path[0].cp_send_gen +#define c_flags c_path[0].cp_flags +#define c_reconnect_jiffies c_path[0].cp_reconnect_jiffies +#define c_send_w c_path[0].cp_send_w +#define c_recv_w c_path[0].cp_recv_w +#define c_conn_w c_path[0].cp_conn_w +#define c_down_w c_path[0].cp_down_w +#define c_cm_lock c_path[0].cp_cm_lock +#define c_waitq c_path[0].cp_waitq +#define c_unacked_packets c_path[0].cp_unacked_packets +#define c_unacked_bytes c_path[0].cp_unacked_bytes + +#endif /* _RDS_RDS_SINGLE_H */ diff --git a/net/rds/recv.c b/net/rds/recv.c index 8413f6c..78b5c430 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -38,6 +38,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" void rds_inc_init(struct rds_incoming *inc, struct rds_connection *conn, diff --git a/net/rds/send.c b/net/rds/send.c index b1962f8..a3b3b35 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -40,6 +40,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" /* When transmitting messages in rds_send_xmit, we need to emerge from diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 74ee126..4bc1c15 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -38,6 +38,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "tcp.h" diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c index fba13d0..ba9ec67 100644 --- a/net/rds/tcp_connect.c +++ b/net/rds/tcp_connect.c @@ -34,6 +34,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "tcp.h" @@ -60,7 +61,8 @@ void rds_tcp_state_change(struct sock *sk) case TCP_SYN_RECV: break; case TCP_ESTABLISHED: - rds_connect_path_complete(conn, RDS_CONN_CONNECTING); + rds_connect_path_complete(&conn->c_path[0], + RDS_CONN_CONNECTING); break; case TCP_CLOSE_WAIT: case TCP_CLOSE: diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index 686b1d0..22d9bb1 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -35,6 +35,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "tcp.h" @@ -132,17 +133,19 @@ int rds_tcp_accept_one(struct socket *sock) * c_transport_data. */ if (ntohl(inet->inet_saddr) < ntohl(inet->inet_daddr) || - !conn->c_outgoing) { + !conn->c_path[0].cp_outgoing) { goto rst_nsk; } else { rds_tcp_reset_callbacks(new_sock, conn); - conn->c_outgoing = 0; + conn->c_path[0].cp_outgoing = 0; /* rds_connect_path_complete() marks RDS_CONN_UP */ - rds_connect_path_complete(conn, RDS_CONN_DISCONNECTING); + rds_connect_path_complete(&conn->c_path[0], + RDS_CONN_DISCONNECTING); } } else { rds_tcp_set_callbacks(new_sock, conn); - rds_connect_path_complete(conn, RDS_CONN_CONNECTING); + rds_connect_path_complete(&conn->c_path[0], + RDS_CONN_CONNECTING); } new_sock = NULL; ret = 0; diff --git a/net/rds/tcp_recv.c b/net/rds/tcp_recv.c index c3196f9..3f8fb38 100644 --- a/net/rds/tcp_recv.c +++ b/net/rds/tcp_recv.c @@ -34,6 +34,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "tcp.h" diff --git a/net/rds/tcp_send.c b/net/rds/tcp_send.c index 22d0f20..2b3414f 100644 --- a/net/rds/tcp_send.c +++ b/net/rds/tcp_send.c @@ -34,6 +34,7 @@ #include #include +#include "rds_single_path.h" #include "rds.h" #include "tcp.h" diff --git a/net/rds/threads.c b/net/rds/threads.c index 4a32304..6d0979b 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -71,30 +71,30 @@ struct workqueue_struct *rds_wq; EXPORT_SYMBOL_GPL(rds_wq); -void rds_connect_path_complete(struct rds_connection *conn, int curr) +void rds_connect_path_complete(struct rds_conn_path *cp, int curr) { - if (!rds_conn_transition(conn, curr, RDS_CONN_UP)) { + if (!rds_conn_path_transition(cp, curr, RDS_CONN_UP)) { printk(KERN_WARNING "%s: Cannot transition to state UP, " "current state is %d\n", __func__, - atomic_read(&conn->c_state)); - rds_conn_drop(conn); + atomic_read(&cp->cp_state)); + rds_conn_path_drop(cp); return; } rdsdebug("conn %p for %pI4 to %pI4 complete\n", - conn, &conn->c_laddr, &conn->c_faddr); + cp->cp_conn, &cp->cp_conn->c_laddr, &cp->cp_conn->c_faddr); - conn->c_reconnect_jiffies = 0; - set_bit(0, &conn->c_map_queued); - queue_delayed_work(rds_wq, &conn->c_send_w, 0); - queue_delayed_work(rds_wq, &conn->c_recv_w, 0); + cp->cp_reconnect_jiffies = 0; + set_bit(0, &cp->cp_conn->c_map_queued); + queue_delayed_work(rds_wq, &cp->cp_send_w, 0); + queue_delayed_work(rds_wq, &cp->cp_recv_w, 0); } EXPORT_SYMBOL_GPL(rds_connect_path_complete); void rds_connect_complete(struct rds_connection *conn) { - rds_connect_path_complete(conn, RDS_CONN_CONNECTING); + rds_connect_path_complete(&conn->c_path[0], RDS_CONN_CONNECTING); } EXPORT_SYMBOL_GPL(rds_connect_complete); @@ -116,46 +116,52 @@ EXPORT_SYMBOL_GPL(rds_connect_complete); * We should *always* start with a random backoff; otherwise a broken connection * will always take several iterations to be re-established. */ -void rds_queue_reconnect(struct rds_connection *conn) +void rds_queue_reconnect(struct rds_conn_path *cp) { unsigned long rand; + struct rds_connection *conn = cp->cp_conn; rdsdebug("conn %p for %pI4 to %pI4 reconnect jiffies %lu\n", conn, &conn->c_laddr, &conn->c_faddr, - conn->c_reconnect_jiffies); + cp->cp_reconnect_jiffies); - set_bit(RDS_RECONNECT_PENDING, &conn->c_flags); - if (conn->c_reconnect_jiffies == 0) { - conn->c_reconnect_jiffies = rds_sysctl_reconnect_min_jiffies; - queue_delayed_work(rds_wq, &conn->c_conn_w, 0); + set_bit(RDS_RECONNECT_PENDING, &cp->cp_flags); + if (cp->cp_reconnect_jiffies == 0) { + cp->cp_reconnect_jiffies = rds_sysctl_reconnect_min_jiffies; + queue_delayed_work(rds_wq, &cp->cp_conn_w, 0); return; } get_random_bytes(&rand, sizeof(rand)); rdsdebug("%lu delay %lu ceil conn %p for %pI4 -> %pI4\n", - rand % conn->c_reconnect_jiffies, conn->c_reconnect_jiffies, + rand % cp->cp_reconnect_jiffies, cp->cp_reconnect_jiffies, conn, &conn->c_laddr, &conn->c_faddr); - queue_delayed_work(rds_wq, &conn->c_conn_w, - rand % conn->c_reconnect_jiffies); + queue_delayed_work(rds_wq, &cp->cp_conn_w, + rand % cp->cp_reconnect_jiffies); - conn->c_reconnect_jiffies = min(conn->c_reconnect_jiffies * 2, + cp->cp_reconnect_jiffies = min(cp->cp_reconnect_jiffies * 2, rds_sysctl_reconnect_max_jiffies); } void rds_connect_worker(struct work_struct *work) { - struct rds_connection *conn = container_of(work, struct rds_connection, c_conn_w.work); + struct rds_conn_path *cp = container_of(work, + struct rds_conn_path, + cp_conn_w.work); + struct rds_connection *conn = cp->cp_conn; int ret; - clear_bit(RDS_RECONNECT_PENDING, &conn->c_flags); - if (rds_conn_transition(conn, RDS_CONN_DOWN, RDS_CONN_CONNECTING)) { + clear_bit(RDS_RECONNECT_PENDING, &cp->cp_flags); + if (rds_conn_path_transition(cp, RDS_CONN_DOWN, RDS_CONN_CONNECTING)) { ret = conn->c_trans->conn_connect(conn); rdsdebug("conn %p for %pI4 to %pI4 dispatched, ret %d\n", conn, &conn->c_laddr, &conn->c_faddr, ret); if (ret) { - if (rds_conn_transition(conn, RDS_CONN_CONNECTING, RDS_CONN_DOWN)) - rds_queue_reconnect(conn); + if (rds_conn_path_transition(cp, + RDS_CONN_CONNECTING, + RDS_CONN_DOWN)) + rds_queue_reconnect(cp); else rds_conn_error(conn, "RDS: connect failed\n"); } @@ -164,22 +170,24 @@ void rds_connect_worker(struct work_struct *work) void rds_send_worker(struct work_struct *work) { - struct rds_connection *conn = container_of(work, struct rds_connection, c_send_w.work); + struct rds_conn_path *cp = container_of(work, + struct rds_conn_path, + cp_send_w.work); int ret; - if (rds_conn_state(conn) == RDS_CONN_UP) { - clear_bit(RDS_LL_SEND_FULL, &conn->c_flags); - ret = rds_send_xmit(conn); + if (rds_conn_path_state(cp) == RDS_CONN_UP) { + clear_bit(RDS_LL_SEND_FULL, &cp->cp_flags); + ret = rds_send_xmit(cp->cp_conn); cond_resched(); - rdsdebug("conn %p ret %d\n", conn, ret); + rdsdebug("conn %p ret %d\n", cp->cp_conn, ret); switch (ret) { case -EAGAIN: rds_stats_inc(s_send_immediate_retry); - queue_delayed_work(rds_wq, &conn->c_send_w, 0); + queue_delayed_work(rds_wq, &cp->cp_send_w, 0); break; case -ENOMEM: rds_stats_inc(s_send_delayed_retry); - queue_delayed_work(rds_wq, &conn->c_send_w, 2); + queue_delayed_work(rds_wq, &cp->cp_send_w, 2); default: break; } @@ -188,20 +196,22 @@ void rds_send_worker(struct work_struct *work) void rds_recv_worker(struct work_struct *work) { - struct rds_connection *conn = container_of(work, struct rds_connection, c_recv_w.work); + struct rds_conn_path *cp = container_of(work, + struct rds_conn_path, + cp_recv_w.work); int ret; - if (rds_conn_state(conn) == RDS_CONN_UP) { - ret = conn->c_trans->recv(conn); - rdsdebug("conn %p ret %d\n", conn, ret); + if (rds_conn_path_state(cp) == RDS_CONN_UP) { + ret = cp->cp_conn->c_trans->recv(cp->cp_conn); + rdsdebug("conn %p ret %d\n", cp->cp_conn, ret); switch (ret) { case -EAGAIN: rds_stats_inc(s_recv_immediate_retry); - queue_delayed_work(rds_wq, &conn->c_recv_w, 0); + queue_delayed_work(rds_wq, &cp->cp_recv_w, 0); break; case -ENOMEM: rds_stats_inc(s_recv_delayed_retry); - queue_delayed_work(rds_wq, &conn->c_recv_w, 2); + queue_delayed_work(rds_wq, &cp->cp_recv_w, 2); default: break; } @@ -210,9 +220,11 @@ void rds_recv_worker(struct work_struct *work) void rds_shutdown_worker(struct work_struct *work) { - struct rds_connection *conn = container_of(work, struct rds_connection, c_down_w); + struct rds_conn_path *cp = container_of(work, + struct rds_conn_path, + cp_down_w); - rds_conn_shutdown(conn); + rds_conn_shutdown(cp->cp_conn); } void rds_threads_exit(void) -- cgit v0.10.2 From 7e8f4413d7861efcb332ebce8d9b000a17eaa0e5 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:27 -0700 Subject: RDS: add t_mp_capable bit to be set by MP capable transports The t_mp_capable bit will be used in the core rds module to support multipathing logic when the transport supports it. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/rds.h b/net/rds/rds.h index ca31a07..28f001c 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -446,7 +446,8 @@ struct rds_transport { char t_name[TRANSNAMSIZ]; struct list_head t_item; struct module *t_owner; - unsigned int t_prefer_loopback:1; + unsigned int t_prefer_loopback:1, + t_mp_capable:1; unsigned int t_type; int (*laddr_check)(struct net *net, __be32 addr); @@ -673,6 +674,7 @@ rds_conn_path_transition(struct rds_conn_path *cp, int old, int new) static inline int rds_conn_transition(struct rds_connection *conn, int old, int new) { + WARN_ON(conn->c_trans->t_mp_capable); return rds_conn_path_transition(&conn->c_path[0], old, new); } @@ -685,6 +687,7 @@ rds_conn_path_state(struct rds_conn_path *cp) static inline int rds_conn_state(struct rds_connection *conn) { + WARN_ON(conn->c_trans->t_mp_capable); return rds_conn_path_state(&conn->c_path[0]); } @@ -697,6 +700,7 @@ rds_conn_path_up(struct rds_conn_path *cp) static inline int rds_conn_up(struct rds_connection *conn) { + WARN_ON(conn->c_trans->t_mp_capable); return rds_conn_path_up(&conn->c_path[0]); } @@ -709,6 +713,7 @@ rds_conn_path_connecting(struct rds_conn_path *cp) static inline int rds_conn_connecting(struct rds_connection *conn) { + WARN_ON(conn->c_trans->t_mp_capable); return rds_conn_path_connecting(&conn->c_path[0]); } -- cgit v0.10.2 From ef9e62c2e5087cb9bc713e3d9776336e1bb40df1 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:28 -0700 Subject: RDS: recv path gets the conn_path from rds_incoming for MP capable transports Transports that are t_mp_capable should set the rds_conn_path on which the datagram was recived in the ->i_conn_path field of struct rds_incoming. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/rds.h b/net/rds/rds.h index 28f001c..7c85b2d 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -231,6 +231,7 @@ struct rds_incoming { atomic_t i_refcount; struct list_head i_item; struct rds_connection *i_conn; + struct rds_conn_path *i_conn_path; struct rds_header i_hdr; unsigned long i_rx_jiffies; __be32 i_saddr; diff --git a/net/rds/recv.c b/net/rds/recv.c index 78b5c430..e36652c 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -38,7 +38,6 @@ #include #include -#include "rds_single_path.h" #include "rds.h" void rds_inc_init(struct rds_incoming *inc, struct rds_connection *conn, @@ -165,13 +164,18 @@ void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr, struct rds_sock *rs = NULL; struct sock *sk; unsigned long flags; + struct rds_conn_path *cp; inc->i_conn = conn; inc->i_rx_jiffies = jiffies; + if (conn->c_trans->t_mp_capable) + cp = inc->i_conn_path; + else + cp = &conn->c_path[0]; rdsdebug("conn %p next %llu inc %p seq %llu len %u sport %u dport %u " "flags 0x%x rx_jiffies %lu\n", conn, - (unsigned long long)conn->c_next_rx_seq, + (unsigned long long)cp->cp_next_rx_seq, inc, (unsigned long long)be64_to_cpu(inc->i_hdr.h_sequence), be32_to_cpu(inc->i_hdr.h_len), @@ -200,12 +204,12 @@ void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr, * XXX we could spend more on the wire to get more robust failure * detection, arguably worth it to avoid data corruption. */ - if (be64_to_cpu(inc->i_hdr.h_sequence) < conn->c_next_rx_seq && + if (be64_to_cpu(inc->i_hdr.h_sequence) < cp->cp_next_rx_seq && (inc->i_hdr.h_flags & RDS_FLAG_RETRANSMITTED)) { rds_stats_inc(s_recv_drop_old_seq); goto out; } - conn->c_next_rx_seq = be64_to_cpu(inc->i_hdr.h_sequence) + 1; + cp->cp_next_rx_seq = be64_to_cpu(inc->i_hdr.h_sequence) + 1; if (rds_sysctl_ping_enable && inc->i_hdr.h_dport == 0) { rds_stats_inc(s_recv_ping); -- cgit v0.10.2 From 5e833e025d9dc3f61c04e74936a14419efb6a032 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:29 -0700 Subject: RDS: rds_inc_path_init() helper function for MP capable transports t_mp_capable transports can use rds_inc_path_init to initialize all fields in struct rds_incoming, including the i_conn_path. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/rds.h b/net/rds/rds.h index 7c85b2d..c3b14cc 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -764,6 +764,8 @@ void rds_page_exit(void); /* recv.c */ void rds_inc_init(struct rds_incoming *inc, struct rds_connection *conn, __be32 saddr); +void rds_inc_path_init(struct rds_incoming *inc, struct rds_conn_path *conn, + __be32 saddr); void rds_inc_put(struct rds_incoming *inc); void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr, struct rds_incoming *inc, gfp_t gfp); diff --git a/net/rds/recv.c b/net/rds/recv.c index e36652c..6d7bd63 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -53,6 +53,20 @@ void rds_inc_init(struct rds_incoming *inc, struct rds_connection *conn, } EXPORT_SYMBOL_GPL(rds_inc_init); +void rds_inc_path_init(struct rds_incoming *inc, struct rds_conn_path *cp, + __be32 saddr) +{ + atomic_set(&inc->i_refcount, 1); + INIT_LIST_HEAD(&inc->i_item); + inc->i_conn = cp->cp_conn; + inc->i_conn_path = cp; + inc->i_saddr = saddr; + inc->i_rdma_cookie = 0; + inc->i_rx_tstamp.tv_sec = 0; + inc->i_rx_tstamp.tv_usec = 0; +} +EXPORT_SYMBOL_GPL(rds_inc_path_init); + static void rds_inc_addref(struct rds_incoming *inc) { rdsdebug("addref inc %p ref %d\n", inc, atomic_read(&inc->i_refcount)); -- cgit v0.10.2 From 4e9b551c14560399776c05f4234650c6d3729458 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:30 -0700 Subject: RDS: Add rds_send_path_reset() rds_send_path_reset() is the path specific version of rds_send_reset() intended for MP capable callers. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/send.c b/net/rds/send.c index a3b3b35..bfb3e05 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -63,14 +63,14 @@ static void rds_send_remove_from_sock(struct list_head *messages, int status); * Reset the send state. Callers must ensure that this doesn't race with * rds_send_xmit(). */ -void rds_send_reset(struct rds_connection *conn) +static void rds_send_path_reset(struct rds_conn_path *cp) { struct rds_message *rm, *tmp; unsigned long flags; - if (conn->c_xmit_rm) { - rm = conn->c_xmit_rm; - conn->c_xmit_rm = NULL; + if (cp->cp_xmit_rm) { + rm = cp->cp_xmit_rm; + cp->cp_xmit_rm = NULL; /* Tell the user the RDMA op is no longer mapped by the * transport. This isn't entirely true (it's flushed out * independently) but as the connection is down, there's @@ -79,26 +79,31 @@ void rds_send_reset(struct rds_connection *conn) rds_message_put(rm); } - conn->c_xmit_sg = 0; - conn->c_xmit_hdr_off = 0; - conn->c_xmit_data_off = 0; - conn->c_xmit_atomic_sent = 0; - conn->c_xmit_rdma_sent = 0; - conn->c_xmit_data_sent = 0; + cp->cp_xmit_sg = 0; + cp->cp_xmit_hdr_off = 0; + cp->cp_xmit_data_off = 0; + cp->cp_xmit_atomic_sent = 0; + cp->cp_xmit_rdma_sent = 0; + cp->cp_xmit_data_sent = 0; - conn->c_map_queued = 0; + cp->cp_conn->c_map_queued = 0; - conn->c_unacked_packets = rds_sysctl_max_unacked_packets; - conn->c_unacked_bytes = rds_sysctl_max_unacked_bytes; + cp->cp_unacked_packets = rds_sysctl_max_unacked_packets; + cp->cp_unacked_bytes = rds_sysctl_max_unacked_bytes; /* Mark messages as retransmissions, and move them to the send q */ - spin_lock_irqsave(&conn->c_lock, flags); - list_for_each_entry_safe(rm, tmp, &conn->c_retrans, m_conn_item) { + spin_lock_irqsave(&cp->cp_lock, flags); + list_for_each_entry_safe(rm, tmp, &cp->cp_retrans, m_conn_item) { set_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags); set_bit(RDS_MSG_RETRANSMITTED, &rm->m_flags); } - list_splice_init(&conn->c_retrans, &conn->c_send_queue); - spin_unlock_irqrestore(&conn->c_lock, flags); + list_splice_init(&cp->cp_retrans, &cp->cp_send_queue); + spin_unlock_irqrestore(&cp->cp_lock, flags); +} + +void rds_send_reset(struct rds_connection *conn) +{ + rds_send_path_reset(&conn->c_path[0]); } EXPORT_SYMBOL_GPL(rds_send_reset); -- cgit v0.10.2 From 5c3d274c75fbcee8e1c919acf25c7feb19a31492 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:31 -0700 Subject: RDS: Add rds_send_path_drop_acked() rds_send_path_drop_acked() is the path-specific version of rds_send_drop_acked() to be invoked by MP capable callers. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/rds.h b/net/rds/rds.h index c3b14cc..d94aa36 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -786,6 +786,8 @@ void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest); typedef int (*is_acked_func)(struct rds_message *rm, uint64_t ack); void rds_send_drop_acked(struct rds_connection *conn, u64 ack, is_acked_func is_acked); +void rds_send_path_drop_acked(struct rds_conn_path *cp, u64 ack, + is_acked_func is_acked); int rds_send_pong(struct rds_connection *conn, __be16 dport); struct rds_message *rds_send_get_message(struct rds_connection *, struct rm_rdma_op *); diff --git a/net/rds/send.c b/net/rds/send.c index bfb3e05..3f6a96c 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -691,16 +691,16 @@ unlock_and_drop: * assigned the m_ack_seq yet - but that's fine as long as tcp_is_acked * checks the RDS_MSG_HAS_ACK_SEQ bit. */ -void rds_send_drop_acked(struct rds_connection *conn, u64 ack, - is_acked_func is_acked) +void rds_send_path_drop_acked(struct rds_conn_path *cp, u64 ack, + is_acked_func is_acked) { struct rds_message *rm, *tmp; unsigned long flags; LIST_HEAD(list); - spin_lock_irqsave(&conn->c_lock, flags); + spin_lock_irqsave(&cp->cp_lock, flags); - list_for_each_entry_safe(rm, tmp, &conn->c_retrans, m_conn_item) { + list_for_each_entry_safe(rm, tmp, &cp->cp_retrans, m_conn_item) { if (!rds_send_is_acked(rm, ack, is_acked)) break; @@ -712,11 +712,19 @@ void rds_send_drop_acked(struct rds_connection *conn, u64 ack, if (!list_empty(&list)) smp_mb__after_atomic(); - spin_unlock_irqrestore(&conn->c_lock, flags); + spin_unlock_irqrestore(&cp->cp_lock, flags); /* now remove the messages from the sock list as needed */ rds_send_remove_from_sock(&list, RDS_RDMA_SUCCESS); } +EXPORT_SYMBOL_GPL(rds_send_path_drop_acked); + +void rds_send_drop_acked(struct rds_connection *conn, u64 ack, + is_acked_func is_acked) +{ + WARN_ON(conn->c_trans->t_mp_capable); + rds_send_path_drop_acked(&conn->c_path[0], ack, is_acked); +} EXPORT_SYMBOL_GPL(rds_send_drop_acked); void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest) -- cgit v0.10.2 From 7d885d0fc69abe22382fae5dddd84684333ab29b Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:32 -0700 Subject: RDS: Remove stale function rds_send_get_message() The only caller of rds_send_get_message() was rds_iw_send_cq_comp_handler() which was removed as part of commit dcdede0406d3 ("RDS: Drop stale iWARP RDMA transport"), so remove rds_send_get_message() for the same reason. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/rds.h b/net/rds/rds.h index d94aa36..2cffd37 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -789,8 +789,6 @@ void rds_send_drop_acked(struct rds_connection *conn, u64 ack, void rds_send_path_drop_acked(struct rds_conn_path *cp, u64 ack, is_acked_func is_acked); int rds_send_pong(struct rds_connection *conn, __be16 dport); -struct rds_message *rds_send_get_message(struct rds_connection *, - struct rm_rdma_op *); /* rdma.c */ void rds_rdma_unuse(struct rds_sock *rs, u32 r_key, int force); diff --git a/net/rds/send.c b/net/rds/send.c index 3f6a96c..3fb280b 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -566,42 +566,6 @@ __rds_send_complete(struct rds_sock *rs, struct rds_message *rm, int status) } /* - * This is called from the IB send completion when we detect - * a RDMA operation that failed with remote access error. - * So speed is not an issue here. - */ -struct rds_message *rds_send_get_message(struct rds_connection *conn, - struct rm_rdma_op *op) -{ - struct rds_message *rm, *tmp, *found = NULL; - unsigned long flags; - - spin_lock_irqsave(&conn->c_lock, flags); - - list_for_each_entry_safe(rm, tmp, &conn->c_retrans, m_conn_item) { - if (&rm->rdma == op) { - atomic_inc(&rm->m_refcount); - found = rm; - goto out; - } - } - - list_for_each_entry_safe(rm, tmp, &conn->c_send_queue, m_conn_item) { - if (&rm->rdma == op) { - atomic_inc(&rm->m_refcount); - found = rm; - break; - } - } - -out: - spin_unlock_irqrestore(&conn->c_lock, flags); - - return found; -} -EXPORT_SYMBOL_GPL(rds_send_get_message); - -/* * This removes messages from the socket's list if they're on it. The list * argument must be private to the caller, we must be able to modify it * without locks. The messages must have a reference held for their -- cgit v0.10.2 From 780a6d9e16d1827eb97c2497d7814fe34d280c15 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:33 -0700 Subject: RDS: Make rds_send_queue_rm() rds_conn_path aware Pass the rds_conn_path to rds_send_queue_rm, and use it to initialize the i_conn_path field in struct rds_incoming. This commit also makes rds_send_queue_rm() MP capable, because it now takes locks specific to the rds_conn_path passed in, instead of defaulting to the c_path[0] based defines from rds_single_path.h Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/send.c b/net/rds/send.c index 3fb280b..076ee41 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -787,6 +787,7 @@ void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest) * message from the flow with RDS_CANCEL_SENT_TO. */ static int rds_send_queue_rm(struct rds_sock *rs, struct rds_connection *conn, + struct rds_conn_path *cp, struct rds_message *rm, __be16 sport, __be16 dport, int *queued) { @@ -830,13 +831,14 @@ static int rds_send_queue_rm(struct rds_sock *rs, struct rds_connection *conn, trying to minimize the time we hold c_lock */ rds_message_populate_header(&rm->m_inc.i_hdr, sport, dport, 0); rm->m_inc.i_conn = conn; + rm->m_inc.i_conn_path = cp; rds_message_addref(rm); - spin_lock(&conn->c_lock); - rm->m_inc.i_hdr.h_sequence = cpu_to_be64(conn->c_next_tx_seq++); - list_add_tail(&rm->m_conn_item, &conn->c_send_queue); + spin_lock(&cp->cp_lock); + rm->m_inc.i_hdr.h_sequence = cpu_to_be64(cp->cp_next_tx_seq++); + list_add_tail(&rm->m_conn_item, &cp->cp_send_queue); set_bit(RDS_MSG_ON_CONN, &rm->m_flags); - spin_unlock(&conn->c_lock); + spin_unlock(&cp->cp_lock); rdsdebug("queued msg %p len %d, rs %p bytes %d seq %llu\n", rm, len, rs, rs->rs_snd_bytes, @@ -968,6 +970,7 @@ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) int queued = 0, allocated_mr = 0; int nonblock = msg->msg_flags & MSG_DONTWAIT; long timeo = sock_sndtimeo(sk, nonblock); + struct rds_conn_path *cpath; /* Mirror Linux UDP mirror of BSD error message compatibility */ /* XXX: Perhaps MSG_MORE someday */ @@ -1074,7 +1077,9 @@ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) goto out; } - while (!rds_send_queue_rm(rs, conn, rm, rs->rs_bound_port, + cpath = &conn->c_path[0]; + + while (!rds_send_queue_rm(rs, conn, cpath, rm, rs->rs_bound_port, dport, &queued)) { rds_stats_inc(s_send_queue_full); @@ -1084,7 +1089,7 @@ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) } timeo = wait_event_interruptible_timeout(*sk_sleep(sk), - rds_send_queue_rm(rs, conn, rm, + rds_send_queue_rm(rs, conn, cpath, rm, rs->rs_bound_port, dport, &queued), -- cgit v0.10.2 From 1f9ecd7eacfd9ee52a114b87292bfe885aafdb1f Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:34 -0700 Subject: RDS: Pass rds_conn_path to rds_send_xmit() Pass a struct rds_conn_path to rds_send_xmit so that MP capable transports can transmit packets on something other than c_path[0]. The eventual goal for MP capable transports is to hash the rds socket to a path based on the bound local address/port, and use this path as the argument to rds_send_xmit() Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c index 4de5a35..3342876 100644 --- a/net/rds/ib_cm.c +++ b/net/rds/ib_cm.c @@ -274,7 +274,7 @@ static void rds_ib_tasklet_fn_send(unsigned long data) if (rds_conn_up(conn) && (!test_bit(RDS_LL_SEND_FULL, &conn->c_flags) || test_bit(0, &conn->c_map_queued))) - rds_send_xmit(ic->conn); + rds_send_xmit(&ic->conn->c_path[0]); } static void poll_rcq(struct rds_ib_connection *ic, struct ib_cq *cq, diff --git a/net/rds/rds.h b/net/rds/rds.h index 2cffd37..b6072eb0 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -457,7 +457,9 @@ struct rds_transport { int (*conn_connect)(struct rds_connection *conn); void (*conn_shutdown)(struct rds_connection *conn); void (*xmit_prepare)(struct rds_connection *conn); + void (*xmit_path_prepare)(struct rds_conn_path *cp); void (*xmit_complete)(struct rds_connection *conn); + void (*xmit_path_complete)(struct rds_conn_path *cp); int (*xmit)(struct rds_connection *conn, struct rds_message *rm, unsigned int hdr_off, unsigned int sg, unsigned int off); int (*xmit_rdma)(struct rds_connection *conn, struct rm_rdma_op *op); @@ -780,7 +782,7 @@ void rds_inc_info_copy(struct rds_incoming *inc, /* send.c */ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len); void rds_send_reset(struct rds_connection *conn); -int rds_send_xmit(struct rds_connection *conn); +int rds_send_xmit(struct rds_conn_path *cp); struct sockaddr_in; void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest); typedef int (*is_acked_func)(struct rds_message *rm, uint64_t ack); diff --git a/net/rds/send.c b/net/rds/send.c index 076ee41..966311d 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -107,14 +107,14 @@ void rds_send_reset(struct rds_connection *conn) } EXPORT_SYMBOL_GPL(rds_send_reset); -static int acquire_in_xmit(struct rds_connection *conn) +static int acquire_in_xmit(struct rds_conn_path *cp) { - return test_and_set_bit(RDS_IN_XMIT, &conn->c_flags) == 0; + return test_and_set_bit(RDS_IN_XMIT, &cp->cp_flags) == 0; } -static void release_in_xmit(struct rds_connection *conn) +static void release_in_xmit(struct rds_conn_path *cp) { - clear_bit(RDS_IN_XMIT, &conn->c_flags); + clear_bit(RDS_IN_XMIT, &cp->cp_flags); smp_mb__after_atomic(); /* * We don't use wait_on_bit()/wake_up_bit() because our waking is in a @@ -122,8 +122,8 @@ static void release_in_xmit(struct rds_connection *conn) * the system-wide hashed waitqueue buckets in the fast path only to * almost never find waiters. */ - if (waitqueue_active(&conn->c_waitq)) - wake_up_all(&conn->c_waitq); + if (waitqueue_active(&cp->cp_waitq)) + wake_up_all(&cp->cp_waitq); } /* @@ -140,8 +140,9 @@ static void release_in_xmit(struct rds_connection *conn) * - small message latency is higher behind queued large messages * - large message latency isn't starved by intervening small sends */ -int rds_send_xmit(struct rds_connection *conn) +int rds_send_xmit(struct rds_conn_path *cp) { + struct rds_connection *conn = cp->cp_conn; struct rds_message *rm; unsigned long flags; unsigned int tmp; @@ -161,7 +162,7 @@ restart: * avoids blocking the caller and trading per-connection data between * caches per message. */ - if (!acquire_in_xmit(conn)) { + if (!acquire_in_xmit(cp)) { rds_stats_inc(s_send_lock_contention); ret = -ENOMEM; goto out; @@ -175,21 +176,25 @@ restart: * The acquire_in_xmit() check above ensures that only one * caller can increment c_send_gen at any time. */ - conn->c_send_gen++; - send_gen = conn->c_send_gen; + cp->cp_send_gen++; + send_gen = cp->cp_send_gen; /* * rds_conn_shutdown() sets the conn state and then tests RDS_IN_XMIT, * we do the opposite to avoid races. */ - if (!rds_conn_up(conn)) { - release_in_xmit(conn); + if (!rds_conn_path_up(cp)) { + release_in_xmit(cp); ret = 0; goto out; } - if (conn->c_trans->xmit_prepare) + if (conn->c_trans->t_mp_capable) { + if (conn->c_trans->xmit_path_prepare) + conn->c_trans->xmit_path_prepare(cp); + } else if (conn->c_trans->xmit_prepare) { conn->c_trans->xmit_prepare(conn); + } /* * spin trying to push headers and data down the connection until @@ -197,7 +202,7 @@ restart: */ while (1) { - rm = conn->c_xmit_rm; + rm = cp->cp_xmit_rm; /* * If between sending messages, we can send a pending congestion @@ -210,14 +215,16 @@ restart: break; } rm->data.op_active = 1; + rm->m_inc.i_conn_path = cp; + rm->m_inc.i_conn = cp->cp_conn; - conn->c_xmit_rm = rm; + cp->cp_xmit_rm = rm; } /* * If not already working on one, grab the next message. * - * c_xmit_rm holds a ref while we're sending this message down + * cp_xmit_rm holds a ref while we're sending this message down * the connction. We can use this ref while holding the * send_sem.. rds_send_reset() is serialized with it. */ @@ -234,10 +241,10 @@ restart: if (batch_count >= send_batch_count) goto over_batch; - spin_lock_irqsave(&conn->c_lock, flags); + spin_lock_irqsave(&cp->cp_lock, flags); - if (!list_empty(&conn->c_send_queue)) { - rm = list_entry(conn->c_send_queue.next, + if (!list_empty(&cp->cp_send_queue)) { + rm = list_entry(cp->cp_send_queue.next, struct rds_message, m_conn_item); rds_message_addref(rm); @@ -246,10 +253,11 @@ restart: * Move the message from the send queue to the retransmit * list right away. */ - list_move_tail(&rm->m_conn_item, &conn->c_retrans); + list_move_tail(&rm->m_conn_item, + &cp->cp_retrans); } - spin_unlock_irqrestore(&conn->c_lock, flags); + spin_unlock_irqrestore(&cp->cp_lock, flags); if (!rm) break; @@ -263,32 +271,34 @@ restart: */ if (rm->rdma.op_active && test_bit(RDS_MSG_RETRANSMITTED, &rm->m_flags)) { - spin_lock_irqsave(&conn->c_lock, flags); + spin_lock_irqsave(&cp->cp_lock, flags); if (test_and_clear_bit(RDS_MSG_ON_CONN, &rm->m_flags)) list_move(&rm->m_conn_item, &to_be_dropped); - spin_unlock_irqrestore(&conn->c_lock, flags); + spin_unlock_irqrestore(&cp->cp_lock, flags); continue; } /* Require an ACK every once in a while */ len = ntohl(rm->m_inc.i_hdr.h_len); - if (conn->c_unacked_packets == 0 || - conn->c_unacked_bytes < len) { + if (cp->cp_unacked_packets == 0 || + cp->cp_unacked_bytes < len) { __set_bit(RDS_MSG_ACK_REQUIRED, &rm->m_flags); - conn->c_unacked_packets = rds_sysctl_max_unacked_packets; - conn->c_unacked_bytes = rds_sysctl_max_unacked_bytes; + cp->cp_unacked_packets = + rds_sysctl_max_unacked_packets; + cp->cp_unacked_bytes = + rds_sysctl_max_unacked_bytes; rds_stats_inc(s_send_ack_required); } else { - conn->c_unacked_bytes -= len; - conn->c_unacked_packets--; + cp->cp_unacked_bytes -= len; + cp->cp_unacked_packets--; } - conn->c_xmit_rm = rm; + cp->cp_xmit_rm = rm; } /* The transport either sends the whole rdma or none of it */ - if (rm->rdma.op_active && !conn->c_xmit_rdma_sent) { + if (rm->rdma.op_active && !cp->cp_xmit_rdma_sent) { rm->m_final_op = &rm->rdma; /* The transport owns the mapped memory for now. * You can't unmap it while it's on the send queue @@ -300,11 +310,11 @@ restart: wake_up_interruptible(&rm->m_flush_wait); break; } - conn->c_xmit_rdma_sent = 1; + cp->cp_xmit_rdma_sent = 1; } - if (rm->atomic.op_active && !conn->c_xmit_atomic_sent) { + if (rm->atomic.op_active && !cp->cp_xmit_atomic_sent) { rm->m_final_op = &rm->atomic; /* The transport owns the mapped memory for now. * You can't unmap it while it's on the send queue @@ -316,7 +326,7 @@ restart: wake_up_interruptible(&rm->m_flush_wait); break; } - conn->c_xmit_atomic_sent = 1; + cp->cp_xmit_atomic_sent = 1; } @@ -342,41 +352,42 @@ restart: rm->data.op_active = 0; } - if (rm->data.op_active && !conn->c_xmit_data_sent) { + if (rm->data.op_active && !cp->cp_xmit_data_sent) { rm->m_final_op = &rm->data; + ret = conn->c_trans->xmit(conn, rm, - conn->c_xmit_hdr_off, - conn->c_xmit_sg, - conn->c_xmit_data_off); + cp->cp_xmit_hdr_off, + cp->cp_xmit_sg, + cp->cp_xmit_data_off); if (ret <= 0) break; - if (conn->c_xmit_hdr_off < sizeof(struct rds_header)) { + if (cp->cp_xmit_hdr_off < sizeof(struct rds_header)) { tmp = min_t(int, ret, sizeof(struct rds_header) - - conn->c_xmit_hdr_off); - conn->c_xmit_hdr_off += tmp; + cp->cp_xmit_hdr_off); + cp->cp_xmit_hdr_off += tmp; ret -= tmp; } - sg = &rm->data.op_sg[conn->c_xmit_sg]; + sg = &rm->data.op_sg[cp->cp_xmit_sg]; while (ret) { tmp = min_t(int, ret, sg->length - - conn->c_xmit_data_off); - conn->c_xmit_data_off += tmp; + cp->cp_xmit_data_off); + cp->cp_xmit_data_off += tmp; ret -= tmp; - if (conn->c_xmit_data_off == sg->length) { - conn->c_xmit_data_off = 0; + if (cp->cp_xmit_data_off == sg->length) { + cp->cp_xmit_data_off = 0; sg++; - conn->c_xmit_sg++; - BUG_ON(ret != 0 && - conn->c_xmit_sg == rm->data.op_nents); + cp->cp_xmit_sg++; + BUG_ON(ret != 0 && cp->cp_xmit_sg == + rm->data.op_nents); } } - if (conn->c_xmit_hdr_off == sizeof(struct rds_header) && - (conn->c_xmit_sg == rm->data.op_nents)) - conn->c_xmit_data_sent = 1; + if (cp->cp_xmit_hdr_off == sizeof(struct rds_header) && + (cp->cp_xmit_sg == rm->data.op_nents)) + cp->cp_xmit_data_sent = 1; } /* @@ -384,23 +395,27 @@ restart: * if there is a data op. Thus, if the data is sent (or there was * none), then we're done with the rm. */ - if (!rm->data.op_active || conn->c_xmit_data_sent) { - conn->c_xmit_rm = NULL; - conn->c_xmit_sg = 0; - conn->c_xmit_hdr_off = 0; - conn->c_xmit_data_off = 0; - conn->c_xmit_rdma_sent = 0; - conn->c_xmit_atomic_sent = 0; - conn->c_xmit_data_sent = 0; + if (!rm->data.op_active || cp->cp_xmit_data_sent) { + cp->cp_xmit_rm = NULL; + cp->cp_xmit_sg = 0; + cp->cp_xmit_hdr_off = 0; + cp->cp_xmit_data_off = 0; + cp->cp_xmit_rdma_sent = 0; + cp->cp_xmit_atomic_sent = 0; + cp->cp_xmit_data_sent = 0; rds_message_put(rm); } } over_batch: - if (conn->c_trans->xmit_complete) + if (conn->c_trans->t_mp_capable) { + if (conn->c_trans->xmit_path_complete) + conn->c_trans->xmit_path_complete(cp); + } else if (conn->c_trans->xmit_complete) { conn->c_trans->xmit_complete(conn); - release_in_xmit(conn); + } + release_in_xmit(cp); /* Nuke any messages we decided not to retransmit. */ if (!list_empty(&to_be_dropped)) { @@ -428,12 +443,12 @@ over_batch: if (ret == 0) { smp_mb(); if ((test_bit(0, &conn->c_map_queued) || - !list_empty(&conn->c_send_queue)) && - send_gen == conn->c_send_gen) { + !list_empty(&cp->cp_send_queue)) && + send_gen == cp->cp_send_gen) { rds_stats_inc(s_send_lock_queue_raced); if (batch_count < send_batch_count) goto restart; - queue_delayed_work(rds_wq, &conn->c_send_w, 1); + queue_delayed_work(rds_wq, &cp->cp_send_w, 1); } } out: @@ -1110,9 +1125,9 @@ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) */ rds_stats_inc(s_send_queued); - ret = rds_send_xmit(conn); + ret = rds_send_xmit(cpath); if (ret == -ENOMEM || ret == -EAGAIN) - queue_delayed_work(rds_wq, &conn->c_send_w, 1); + queue_delayed_work(rds_wq, &cpath->cp_send_w, 1); rds_message_put(rm); return payload_len; diff --git a/net/rds/threads.c b/net/rds/threads.c index 6d0979b..50d2657 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -177,7 +177,7 @@ void rds_send_worker(struct work_struct *work) if (rds_conn_path_state(cp) == RDS_CONN_UP) { clear_bit(RDS_LL_SEND_FULL, &cp->cp_flags); - ret = rds_send_xmit(cp->cp_conn); + ret = rds_send_xmit(cp); cond_resched(); rdsdebug("conn %p ret %d\n", cp->cp_conn, ret); switch (ret) { -- cgit v0.10.2 From 01ff34ed44a48ed0ae875291b4b6b7dc9ebeea69 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:35 -0700 Subject: RDS: Extract rds_conn_path from i_conn_path in rds_send_drop_to() for MP-capable transports Explicitly set up rds_conn_path, either from i_conn_path (for MP capable transpots) or as c_path[0], and use this in rds_send_drop_to() Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/send.c b/net/rds/send.c index 966311d..9c34fd2 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -710,6 +710,7 @@ void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest) { struct rds_message *rm, *tmp; struct rds_connection *conn; + struct rds_conn_path *cp; unsigned long flags; LIST_HEAD(list); @@ -738,22 +739,26 @@ void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest) list_for_each_entry(rm, &list, m_sock_item) { conn = rm->m_inc.i_conn; + if (conn->c_trans->t_mp_capable) + cp = rm->m_inc.i_conn_path; + else + cp = &conn->c_path[0]; - spin_lock_irqsave(&conn->c_lock, flags); + spin_lock_irqsave(&cp->cp_lock, flags); /* * Maybe someone else beat us to removing rm from the conn. * If we race with their flag update we'll get the lock and * then really see that the flag has been cleared. */ if (!test_and_clear_bit(RDS_MSG_ON_CONN, &rm->m_flags)) { - spin_unlock_irqrestore(&conn->c_lock, flags); + spin_unlock_irqrestore(&cp->cp_lock, flags); spin_lock_irqsave(&rm->m_rs_lock, flags); rm->m_rs = NULL; spin_unlock_irqrestore(&rm->m_rs_lock, flags); continue; } list_del_init(&rm->m_conn_item); - spin_unlock_irqrestore(&conn->c_lock, flags); + spin_unlock_irqrestore(&cp->cp_lock, flags); /* * Couldn't grab m_rs_lock in top loop (lock ordering), -- cgit v0.10.2 From 45997e9e2e01d76607d70461414f66f51487bfe5 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:36 -0700 Subject: RDS: Make rds_send_pong() take a rds_conn_path argument This commit allows rds_send_pong() callers to send back the rds pong message on some path other than c_path[0] by passing in a struct rds_conn_path * argument. It also removes the last dependency on the #defines in rds_single.h from send.c Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/rds.h b/net/rds/rds.h index b6072eb0..e315151 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -790,7 +790,7 @@ void rds_send_drop_acked(struct rds_connection *conn, u64 ack, is_acked_func is_acked); void rds_send_path_drop_acked(struct rds_conn_path *cp, u64 ack, is_acked_func is_acked); -int rds_send_pong(struct rds_connection *conn, __be16 dport); +int rds_send_pong(struct rds_conn_path *cp, __be16 dport); /* rdma.c */ void rds_rdma_unuse(struct rds_sock *rs, u32 r_key, int force); diff --git a/net/rds/recv.c b/net/rds/recv.c index 6d7bd63..b58f505 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -227,7 +227,7 @@ void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr, if (rds_sysctl_ping_enable && inc->i_hdr.h_dport == 0) { rds_stats_inc(s_recv_ping); - rds_send_pong(conn, inc->i_hdr.h_sport); + rds_send_pong(cp, inc->i_hdr.h_sport); goto out; } diff --git a/net/rds/send.c b/net/rds/send.c index 9c34fd2..e614513 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -40,7 +40,6 @@ #include #include -#include "rds_single_path.h" #include "rds.h" /* When transmitting messages in rds_send_xmit, we need to emerge from @@ -1153,7 +1152,7 @@ out: * Reply to a ping packet. */ int -rds_send_pong(struct rds_connection *conn, __be16 dport) +rds_send_pong(struct rds_conn_path *cp, __be16 dport) { struct rds_message *rm; unsigned long flags; @@ -1165,31 +1164,32 @@ rds_send_pong(struct rds_connection *conn, __be16 dport) goto out; } - rm->m_daddr = conn->c_faddr; + rm->m_daddr = cp->cp_conn->c_faddr; rm->data.op_active = 1; - rds_conn_connect_if_down(conn); + rds_conn_connect_if_down(cp->cp_conn); - ret = rds_cong_wait(conn->c_fcong, dport, 1, NULL); + ret = rds_cong_wait(cp->cp_conn->c_fcong, dport, 1, NULL); if (ret) goto out; - spin_lock_irqsave(&conn->c_lock, flags); - list_add_tail(&rm->m_conn_item, &conn->c_send_queue); + spin_lock_irqsave(&cp->cp_lock, flags); + list_add_tail(&rm->m_conn_item, &cp->cp_send_queue); set_bit(RDS_MSG_ON_CONN, &rm->m_flags); rds_message_addref(rm); - rm->m_inc.i_conn = conn; + rm->m_inc.i_conn = cp->cp_conn; + rm->m_inc.i_conn_path = cp; rds_message_populate_header(&rm->m_inc.i_hdr, 0, dport, - conn->c_next_tx_seq); - conn->c_next_tx_seq++; - spin_unlock_irqrestore(&conn->c_lock, flags); + cp->cp_next_tx_seq); + cp->cp_next_tx_seq++; + spin_unlock_irqrestore(&cp->cp_lock, flags); rds_stats_inc(s_send_queued); rds_stats_inc(s_send_pong); /* schedule the send work on rds_wq */ - queue_delayed_work(rds_wq, &conn->c_send_w, 1); + queue_delayed_work(rds_wq, &cp->cp_send_w, 1); rds_message_put(rm); return 0; -- cgit v0.10.2 From 3c0a59001a416ec2a1c46576917732fe5b99336b Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:37 -0700 Subject: RDS: Add rds_conn_path_connect_if_down() for MP-aware callers rds_conn_path_connect_if_down() works on the rds_conn_path that it is passed. Callers who are not t_m_capable may continue calling rds_conn_connect_if_down, which will invoke rds_conn_path_connect_if_down() with the default c_path[0]. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/connection.c b/net/rds/connection.c index 6fa2074..953a426 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -572,11 +572,17 @@ EXPORT_SYMBOL_GPL(rds_conn_drop); * If the connection is down, trigger a connect. We may have scheduled a * delayed reconnect however - in this case we should not interfere. */ +void rds_conn_path_connect_if_down(struct rds_conn_path *cp) +{ + if (rds_conn_path_state(cp) == RDS_CONN_DOWN && + !test_and_set_bit(RDS_RECONNECT_PENDING, &cp->cp_flags)) + queue_delayed_work(rds_wq, &cp->cp_conn_w, 0); +} + void rds_conn_connect_if_down(struct rds_connection *conn) { - if (rds_conn_state(conn) == RDS_CONN_DOWN && - !test_and_set_bit(RDS_RECONNECT_PENDING, &conn->c_flags)) - queue_delayed_work(rds_wq, &conn->c_conn_w, 0); + WARN_ON(conn->c_trans->t_mp_capable); + rds_conn_path_connect_if_down(&conn->c_path[0]); } EXPORT_SYMBOL_GPL(rds_conn_connect_if_down); diff --git a/net/rds/rds.h b/net/rds/rds.h index e315151..74fcf5a 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -658,6 +658,7 @@ void rds_conn_destroy(struct rds_connection *conn); void rds_conn_drop(struct rds_connection *conn); void rds_conn_path_drop(struct rds_conn_path *cpath); void rds_conn_connect_if_down(struct rds_connection *conn); +void rds_conn_path_connect_if_down(struct rds_conn_path *cp); void rds_for_each_conn_info(struct socket *sock, unsigned int len, struct rds_info_iterator *iter, struct rds_info_lengths *lens, diff --git a/net/rds/send.c b/net/rds/send.c index e614513..369bd66 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -1088,16 +1088,15 @@ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) goto out; } - rds_conn_connect_if_down(conn); + cpath = &conn->c_path[0]; + + rds_conn_path_connect_if_down(cpath); ret = rds_cong_wait(conn->c_fcong, dport, nonblock, rs); if (ret) { rs->rs_seen_congestion = 1; goto out; } - - cpath = &conn->c_path[0]; - while (!rds_send_queue_rm(rs, conn, cpath, rm, rs->rs_bound_port, dport, &queued)) { rds_stats_inc(s_send_queue_full); @@ -1167,7 +1166,7 @@ rds_send_pong(struct rds_conn_path *cp, __be16 dport) rm->m_daddr = cp->cp_conn->c_faddr; rm->data.op_active = 1; - rds_conn_connect_if_down(cp->cp_conn); + rds_conn_path_connect_if_down(cp); ret = rds_cong_wait(cp->cp_conn->c_fcong, dport, 1, NULL); if (ret) -- cgit v0.10.2 From 992c9ec5fe1d0c7859e158ee22f293cbee95c6a3 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:38 -0700 Subject: RDS: update rds-info related functions to traverse multiple conn_paths This commit updates the callbacks related to the rds-info command so that they walk through all the rds_conn_path structures and report the requested info. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/connection.c b/net/rds/connection.c index 953a426..9e0b489 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -400,6 +400,7 @@ static void rds_conn_message_info(struct socket *sock, unsigned int len, unsigned int total = 0; unsigned long flags; size_t i; + int j; len /= sizeof(struct rds_info_message); @@ -408,23 +409,32 @@ static void rds_conn_message_info(struct socket *sock, unsigned int len, for (i = 0, head = rds_conn_hash; i < ARRAY_SIZE(rds_conn_hash); i++, head++) { hlist_for_each_entry_rcu(conn, head, c_hash_node) { - if (want_send) - list = &conn->c_send_queue; - else - list = &conn->c_retrans; - - spin_lock_irqsave(&conn->c_lock, flags); - - /* XXX too lazy to maintain counts.. */ - list_for_each_entry(rm, list, m_conn_item) { - total++; - if (total <= len) - rds_inc_info_copy(&rm->m_inc, iter, - conn->c_laddr, - conn->c_faddr, 0); + struct rds_conn_path *cp; + + for (j = 0; j < RDS_MPATH_WORKERS; j++) { + cp = &conn->c_path[j]; + if (want_send) + list = &cp->cp_send_queue; + else + list = &cp->cp_retrans; + + spin_lock_irqsave(&cp->cp_lock, flags); + + /* XXX too lazy to maintain counts.. */ + list_for_each_entry(rm, list, m_conn_item) { + total++; + if (total <= len) + rds_inc_info_copy(&rm->m_inc, + iter, + conn->c_laddr, + conn->c_faddr, + 0); + } + + spin_unlock_irqrestore(&cp->cp_lock, flags); + if (!conn->c_trans->t_mp_capable) + break; } - - spin_unlock_irqrestore(&conn->c_lock, flags); } } rcu_read_unlock(); @@ -486,27 +496,72 @@ void rds_for_each_conn_info(struct socket *sock, unsigned int len, } EXPORT_SYMBOL_GPL(rds_for_each_conn_info); -static int rds_conn_info_visitor(struct rds_connection *conn, - void *buffer) +void rds_walk_conn_path_info(struct socket *sock, unsigned int len, + struct rds_info_iterator *iter, + struct rds_info_lengths *lens, + int (*visitor)(struct rds_conn_path *, void *), + size_t item_len) +{ + u64 buffer[(item_len + 7) / 8]; + struct hlist_head *head; + struct rds_connection *conn; + size_t i; + int j; + + rcu_read_lock(); + + lens->nr = 0; + lens->each = item_len; + + for (i = 0, head = rds_conn_hash; i < ARRAY_SIZE(rds_conn_hash); + i++, head++) { + hlist_for_each_entry_rcu(conn, head, c_hash_node) { + struct rds_conn_path *cp; + + for (j = 0; j < RDS_MPATH_WORKERS; j++) { + cp = &conn->c_path[j]; + + /* XXX no cp_lock usage.. */ + if (!visitor(cp, buffer)) + continue; + if (!conn->c_trans->t_mp_capable) + break; + } + + /* We copy as much as we can fit in the buffer, + * but we count all items so that the caller + * can resize the buffer. + */ + if (len >= item_len) { + rds_info_copy(iter, buffer, item_len); + len -= item_len; + } + lens->nr++; + } + } + rcu_read_unlock(); +} + +static int rds_conn_info_visitor(struct rds_conn_path *cp, void *buffer) { struct rds_info_connection *cinfo = buffer; - cinfo->next_tx_seq = conn->c_next_tx_seq; - cinfo->next_rx_seq = conn->c_next_rx_seq; - cinfo->laddr = conn->c_laddr; - cinfo->faddr = conn->c_faddr; - strncpy(cinfo->transport, conn->c_trans->t_name, + cinfo->next_tx_seq = cp->cp_next_tx_seq; + cinfo->next_rx_seq = cp->cp_next_rx_seq; + cinfo->laddr = cp->cp_conn->c_laddr; + cinfo->faddr = cp->cp_conn->c_faddr; + strncpy(cinfo->transport, cp->cp_conn->c_trans->t_name, sizeof(cinfo->transport)); cinfo->flags = 0; - rds_conn_info_set(cinfo->flags, test_bit(RDS_IN_XMIT, &conn->c_flags), + rds_conn_info_set(cinfo->flags, test_bit(RDS_IN_XMIT, &cp->cp_flags), SENDING); /* XXX Future: return the state rather than these funky bits */ rds_conn_info_set(cinfo->flags, - atomic_read(&conn->c_state) == RDS_CONN_CONNECTING, + atomic_read(&cp->cp_state) == RDS_CONN_CONNECTING, CONNECTING); rds_conn_info_set(cinfo->flags, - atomic_read(&conn->c_state) == RDS_CONN_UP, + atomic_read(&cp->cp_state) == RDS_CONN_UP, CONNECTED); return 1; } @@ -515,7 +570,7 @@ static void rds_conn_info(struct socket *sock, unsigned int len, struct rds_info_iterator *iter, struct rds_info_lengths *lens) { - rds_for_each_conn_info(sock, len, iter, lens, + rds_walk_conn_path_info(sock, len, iter, lens, rds_conn_info_visitor, sizeof(struct rds_info_connection)); } -- cgit v0.10.2 From fb1b3dc43dabd4bf7b57b3d63fd2875a499c81f0 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:39 -0700 Subject: RDS: Add rds_conn_path_error() rds_conn_path_error() is the MP-aware analog of rds_conn_error, to be used by multipath-capable callers. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/connection.c b/net/rds/connection.c index 9e0b489..57556e6 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -655,3 +655,15 @@ __rds_conn_error(struct rds_connection *conn, const char *fmt, ...) rds_conn_drop(conn); } + +void +__rds_conn_path_error(struct rds_conn_path *cp, const char *fmt, ...) +{ + va_list ap; + + va_start(ap, fmt); + vprintk(fmt, ap); + va_end(ap); + + rds_conn_path_drop(cp); +} diff --git a/net/rds/rds.h b/net/rds/rds.h index 74fcf5a..85f98bd 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -669,6 +669,10 @@ void __rds_conn_error(struct rds_connection *conn, const char *, ...); #define rds_conn_error(conn, fmt...) \ __rds_conn_error(conn, KERN_WARNING "RDS: " fmt) +void __rds_conn_path_error(struct rds_conn_path *cp, const char *, ...); +#define rds_conn_path_error(cp, fmt...) \ + __rds_conn_path_error(cp, KERN_WARNING "RDS: " fmt) + static inline int rds_conn_path_transition(struct rds_conn_path *cp, int old, int new) { diff --git a/net/rds/threads.c b/net/rds/threads.c index 50d2657..94cca66 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -163,7 +163,8 @@ void rds_connect_worker(struct work_struct *work) RDS_CONN_DOWN)) rds_queue_reconnect(cp); else - rds_conn_error(conn, "RDS: connect failed\n"); + rds_conn_path_error(cp, + "RDS: connect failed\n"); } } } -- cgit v0.10.2 From 1c5113cf796bb730abc1798a3649b61e9e022be6 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:40 -0700 Subject: RDS: Initialize all RDS_MPATH_WORKERS in __rds_conn_create Add a for() loop in __rds_conn_create to initialize all the conn_paths, in preparate for MP capable transports. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/connection.c b/net/rds/connection.c index 57556e6..a99ac69 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -111,6 +111,32 @@ static void rds_conn_reset(struct rds_connection *conn) * reliability guarantees of RDS. */ } +static void __rds_conn_path_init(struct rds_connection *conn, + struct rds_conn_path *cp, bool is_outgoing) +{ + spin_lock_init(&cp->cp_lock); + cp->cp_next_tx_seq = 1; + init_waitqueue_head(&cp->cp_waitq); + INIT_LIST_HEAD(&cp->cp_send_queue); + INIT_LIST_HEAD(&cp->cp_retrans); + + cp->cp_conn = conn; + atomic_set(&cp->cp_state, RDS_CONN_DOWN); + cp->cp_send_gen = 0; + /* cp_outgoing is per-path. So we can only set it here + * for the single-path transports. + */ + if (!conn->c_trans->t_mp_capable) + cp->cp_outgoing = (is_outgoing ? 1 : 0); + cp->cp_reconnect_jiffies = 0; + INIT_DELAYED_WORK(&cp->cp_send_w, rds_send_worker); + INIT_DELAYED_WORK(&cp->cp_recv_w, rds_recv_worker); + INIT_DELAYED_WORK(&cp->cp_conn_w, rds_connect_worker); + INIT_WORK(&cp->cp_down_w, rds_shutdown_worker); + mutex_init(&cp->cp_cm_lock); + cp->cp_flags = 0; +} + /* * There is only every one 'conn' for a given pair of addresses in the * system at a time. They contain messages to be retransmitted and so @@ -154,14 +180,8 @@ static struct rds_connection *__rds_conn_create(struct net *net, INIT_HLIST_NODE(&conn->c_hash_node); conn->c_laddr = laddr; conn->c_faddr = faddr; - spin_lock_init(&conn->c_lock); - conn->c_next_tx_seq = 1; - conn->c_path[0].cp_conn = conn; - rds_conn_net_set(conn, net); - init_waitqueue_head(&conn->c_waitq); - INIT_LIST_HEAD(&conn->c_send_queue); - INIT_LIST_HEAD(&conn->c_retrans); + rds_conn_net_set(conn, net); ret = rds_cong_get_maps(conn); if (ret) { @@ -197,17 +217,6 @@ static struct rds_connection *__rds_conn_create(struct net *net, goto out; } - atomic_set(&conn->c_state, RDS_CONN_DOWN); - conn->c_send_gen = 0; - conn->c_path[0].cp_outgoing = (is_outgoing ? 1 : 0); - conn->c_reconnect_jiffies = 0; - INIT_DELAYED_WORK(&conn->c_send_w, rds_send_worker); - INIT_DELAYED_WORK(&conn->c_recv_w, rds_recv_worker); - INIT_DELAYED_WORK(&conn->c_conn_w, rds_connect_worker); - INIT_WORK(&conn->c_down_w, rds_shutdown_worker); - mutex_init(&conn->c_cm_lock); - conn->c_flags = 0; - rdsdebug("allocated conn %p for %pI4 -> %pI4 over %s %s\n", conn, &laddr, &faddr, trans->t_name ? trans->t_name : "[unknown]", @@ -224,7 +233,7 @@ static struct rds_connection *__rds_conn_create(struct net *net, if (parent) { /* Creating passive conn */ if (parent->c_passive) { - trans->conn_free(conn->c_transport_data); + trans->conn_free(conn->c_path[0].cp_transport_data); kmem_cache_free(rds_conn_slab, conn); conn = parent->c_passive; } else { @@ -238,10 +247,26 @@ static struct rds_connection *__rds_conn_create(struct net *net, found = rds_conn_lookup(net, head, laddr, faddr, trans); if (found) { - trans->conn_free(conn->c_transport_data); + struct rds_conn_path *cp; + int i; + + for (i = 0; i < RDS_MPATH_WORKERS; i++) { + cp = &conn->c_path[i]; + trans->conn_free(cp->cp_transport_data); + if (!trans->t_mp_capable) + break; + } kmem_cache_free(rds_conn_slab, conn); conn = found; } else { + int i; + + for (i = 0; i < RDS_MPATH_WORKERS; i++) { + __rds_conn_path_init(conn, &conn->c_path[i], + is_outgoing); + conn->c_path[i].cp_index = i; + } + hlist_add_head_rcu(&conn->c_hash_node, head); rds_cong_add_conn(conn); rds_conn_count++; -- cgit v0.10.2 From d769ef81d5b5932520fbefb02614a4380c132495 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:41 -0700 Subject: RDS: Update rds_conn_shutdown to work with rds_conn_path This commit changes rds_conn_shutdown to take a rds_conn_path * argument, allowing it to shutdown paths other than c_path[0] for MP-capable transports. Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/connection.c b/net/rds/connection.c index a99ac69..a88d26f 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -96,14 +96,16 @@ static struct rds_connection *rds_conn_lookup(struct net *net, * and receiving over this connection again in the future. It is up to * the transport to have serialized this call with its send and recv. */ -static void rds_conn_reset(struct rds_connection *conn) +static void rds_conn_path_reset(struct rds_conn_path *cp) { + struct rds_connection *conn = cp->cp_conn; + rdsdebug("connection %pI4 to %pI4 reset\n", &conn->c_laddr, &conn->c_faddr); rds_stats_inc(s_conn_reset); - rds_send_reset(conn); - conn->c_flags = 0; + rds_send_path_reset(cp); + cp->cp_flags = 0; /* Do not clear next_rx_seq here, else we cannot distinguish * retransmitted packets from new packets, and will hand all @@ -294,10 +296,12 @@ struct rds_connection *rds_conn_create_outgoing(struct net *net, } EXPORT_SYMBOL_GPL(rds_conn_create_outgoing); -void rds_conn_shutdown(struct rds_connection *conn) +void rds_conn_shutdown(struct rds_conn_path *cp) { + struct rds_connection *conn = cp->cp_conn; + /* shut it down unless it's down already */ - if (!rds_conn_transition(conn, RDS_CONN_DOWN, RDS_CONN_DOWN)) { + if (!rds_conn_path_transition(cp, RDS_CONN_DOWN, RDS_CONN_DOWN)) { /* * Quiesce the connection mgmt handlers before we start tearing * things down. We don't hold the mutex for the entire @@ -305,35 +309,41 @@ void rds_conn_shutdown(struct rds_connection *conn) * deadlocking with the CM handler. Instead, the CM event * handler is supposed to check for state DISCONNECTING */ - mutex_lock(&conn->c_cm_lock); - if (!rds_conn_transition(conn, RDS_CONN_UP, RDS_CONN_DISCONNECTING) - && !rds_conn_transition(conn, RDS_CONN_ERROR, RDS_CONN_DISCONNECTING)) { - rds_conn_error(conn, "shutdown called in state %d\n", - atomic_read(&conn->c_state)); - mutex_unlock(&conn->c_cm_lock); + mutex_lock(&cp->cp_cm_lock); + if (!rds_conn_path_transition(cp, RDS_CONN_UP, + RDS_CONN_DISCONNECTING) && + !rds_conn_path_transition(cp, RDS_CONN_ERROR, + RDS_CONN_DISCONNECTING)) { + rds_conn_path_error(cp, + "shutdown called in state %d\n", + atomic_read(&cp->cp_state)); + mutex_unlock(&cp->cp_cm_lock); return; } - mutex_unlock(&conn->c_cm_lock); + mutex_unlock(&cp->cp_cm_lock); - wait_event(conn->c_waitq, - !test_bit(RDS_IN_XMIT, &conn->c_flags)); - wait_event(conn->c_waitq, - !test_bit(RDS_RECV_REFILL, &conn->c_flags)); + wait_event(cp->cp_waitq, + !test_bit(RDS_IN_XMIT, &cp->cp_flags)); + wait_event(cp->cp_waitq, + !test_bit(RDS_RECV_REFILL, &cp->cp_flags)); - conn->c_trans->conn_shutdown(conn); - rds_conn_reset(conn); + if (!conn->c_trans->t_mp_capable) + conn->c_trans->conn_shutdown(conn); + else + conn->c_trans->conn_path_shutdown(cp); + rds_conn_path_reset(cp); - if (!rds_conn_transition(conn, RDS_CONN_DISCONNECTING, RDS_CONN_DOWN)) { + if (!rds_conn_path_transition(cp, RDS_CONN_DISCONNECTING, + RDS_CONN_DOWN)) { /* This can happen - eg when we're in the middle of tearing * down the connection, and someone unloads the rds module. * Quite reproduceable with loopback connections. * Mostly harmless. */ - rds_conn_error(conn, - "%s: failed to transition to state DOWN, " - "current state is %d\n", - __func__, - atomic_read(&conn->c_state)); + rds_conn_path_error(cp, "%s: failed to transition " + "to state DOWN, current state " + "is %d\n", __func__, + atomic_read(&cp->cp_state)); return; } } @@ -342,13 +352,13 @@ void rds_conn_shutdown(struct rds_connection *conn) * The passive side of an IB loopback connection is never added * to the conn hash, so we never trigger a reconnect on this * conn - the reconnect is always triggered by the active peer. */ - cancel_delayed_work_sync(&conn->c_conn_w); + cancel_delayed_work_sync(&cp->cp_conn_w); rcu_read_lock(); if (!hlist_unhashed(&conn->c_hash_node)) { rcu_read_unlock(); if (conn->c_trans->t_type != RDS_TRANS_TCP || - conn->c_path[0].cp_outgoing == 1) - rds_queue_reconnect(&conn->c_path[0]); + cp->cp_outgoing == 1) + rds_queue_reconnect(cp); } else { rcu_read_unlock(); } diff --git a/net/rds/rds.h b/net/rds/rds.h index 85f98bd..2e35b73 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -456,6 +456,7 @@ struct rds_transport { void (*conn_free)(void *data); int (*conn_connect)(struct rds_connection *conn); void (*conn_shutdown)(struct rds_connection *conn); + void (*conn_path_shutdown)(struct rds_conn_path *conn); void (*xmit_prepare)(struct rds_connection *conn); void (*xmit_path_prepare)(struct rds_conn_path *cp); void (*xmit_complete)(struct rds_connection *conn); @@ -653,7 +654,7 @@ struct rds_connection *rds_conn_create(struct net *net, struct rds_connection *rds_conn_create_outgoing(struct net *net, __be32 laddr, __be32 faddr, struct rds_transport *trans, gfp_t gfp); -void rds_conn_shutdown(struct rds_connection *conn); +void rds_conn_shutdown(struct rds_conn_path *cpath); void rds_conn_destroy(struct rds_connection *conn); void rds_conn_drop(struct rds_connection *conn); void rds_conn_path_drop(struct rds_conn_path *cpath); @@ -786,7 +787,7 @@ void rds_inc_info_copy(struct rds_incoming *inc, /* send.c */ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len); -void rds_send_reset(struct rds_connection *conn); +void rds_send_path_reset(struct rds_conn_path *conn); int rds_send_xmit(struct rds_conn_path *cp); struct sockaddr_in; void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest); diff --git a/net/rds/send.c b/net/rds/send.c index 369bd66..ee43d6b 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -62,7 +62,7 @@ static void rds_send_remove_from_sock(struct list_head *messages, int status); * Reset the send state. Callers must ensure that this doesn't race with * rds_send_xmit(). */ -static void rds_send_path_reset(struct rds_conn_path *cp) +void rds_send_path_reset(struct rds_conn_path *cp) { struct rds_message *rm, *tmp; unsigned long flags; @@ -99,12 +99,7 @@ static void rds_send_path_reset(struct rds_conn_path *cp) list_splice_init(&cp->cp_retrans, &cp->cp_send_queue); spin_unlock_irqrestore(&cp->cp_lock, flags); } - -void rds_send_reset(struct rds_connection *conn) -{ - rds_send_path_reset(&conn->c_path[0]); -} -EXPORT_SYMBOL_GPL(rds_send_reset); +EXPORT_SYMBOL_GPL(rds_send_path_reset); static int acquire_in_xmit(struct rds_conn_path *cp) { diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 4bc1c15..0e757a0 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -186,7 +186,7 @@ void rds_tcp_reset_callbacks(struct socket *sock, release_sock(osock->sk); sock_release(osock); newsock: - rds_send_reset(conn); + rds_send_path_reset(&conn->c_path[0]); lock_sock(sock->sk); write_lock_bh(&sock->sk->sk_callback_lock); tc->t_sock = sock; diff --git a/net/rds/threads.c b/net/rds/threads.c index 94cca66..9fbe95b 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -225,7 +225,7 @@ void rds_shutdown_worker(struct work_struct *work) struct rds_conn_path, cp_down_w); - rds_conn_shutdown(cp->cp_conn); + rds_conn_shutdown(cp); } void rds_threads_exit(void) -- cgit v0.10.2 From 3ecc5693c02bb154bc8609c640eb862804c4aabb Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Mon, 13 Jun 2016 09:44:42 -0700 Subject: RDS: Update rds_conn_destroy to be MP capable Refactor rds_conn_destroy() so that the per-path dismantling is done in rds_conn_path_destroy, and then iterate as needed over rds_conn_path_destroy(). Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/connection.c b/net/rds/connection.c index a88d26f..a4b07c8 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -36,7 +36,6 @@ #include #include -#include "rds_single_path.h" #include "rds.h" #include "loop.h" @@ -364,6 +363,34 @@ void rds_conn_shutdown(struct rds_conn_path *cp) } } +/* destroy a single rds_conn_path. rds_conn_destroy() iterates over + * all paths using rds_conn_path_destroy() + */ +static void rds_conn_path_destroy(struct rds_conn_path *cp) +{ + struct rds_message *rm, *rtmp; + + rds_conn_path_drop(cp); + flush_work(&cp->cp_down_w); + + /* make sure lingering queued work won't try to ref the conn */ + cancel_delayed_work_sync(&cp->cp_send_w); + cancel_delayed_work_sync(&cp->cp_recv_w); + + /* tear down queued messages */ + list_for_each_entry_safe(rm, rtmp, + &cp->cp_send_queue, + m_conn_item) { + list_del_init(&rm->m_conn_item); + BUG_ON(!list_empty(&rm->m_sock_item)); + rds_message_put(rm); + } + if (cp->cp_xmit_rm) + rds_message_put(cp->cp_xmit_rm); + + cp->cp_conn->c_trans->conn_free(cp->cp_transport_data); +} + /* * Stop and free a connection. * @@ -373,7 +400,6 @@ void rds_conn_shutdown(struct rds_conn_path *cp) */ void rds_conn_destroy(struct rds_connection *conn) { - struct rds_message *rm, *rtmp; unsigned long flags; rdsdebug("freeing conn %p for %pI4 -> " @@ -387,25 +413,19 @@ void rds_conn_destroy(struct rds_connection *conn) synchronize_rcu(); /* shut the connection down */ - rds_conn_drop(conn); - flush_work(&conn->c_down_w); - - /* make sure lingering queued work won't try to ref the conn */ - cancel_delayed_work_sync(&conn->c_send_w); - cancel_delayed_work_sync(&conn->c_recv_w); + if (!conn->c_trans->t_mp_capable) { + rds_conn_path_destroy(&conn->c_path[0]); + BUG_ON(!list_empty(&conn->c_path[0].cp_retrans)); + } else { + int i; + struct rds_conn_path *cp; - /* tear down queued messages */ - list_for_each_entry_safe(rm, rtmp, - &conn->c_send_queue, - m_conn_item) { - list_del_init(&rm->m_conn_item); - BUG_ON(!list_empty(&rm->m_sock_item)); - rds_message_put(rm); + for (i = 0; i < RDS_MPATH_WORKERS; i++) { + cp = &conn->c_path[i]; + rds_conn_path_destroy(cp); + BUG_ON(!list_empty(&cp->cp_retrans)); + } } - if (conn->c_xmit_rm) - rds_message_put(conn->c_xmit_rm); - - conn->c_trans->conn_free(conn->c_transport_data); /* * The congestion maps aren't freed up here. They're @@ -414,7 +434,6 @@ void rds_conn_destroy(struct rds_connection *conn) */ rds_cong_remove_conn(conn); - BUG_ON(!list_empty(&conn->c_retrans)); kmem_cache_free(rds_conn_slab, conn); spin_lock_irqsave(&rds_conn_lock, flags); -- cgit v0.10.2 From 95df1b16074ce1e5dc4129fa206afbac32663e06 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Mon, 13 Jun 2016 10:47:43 -0700 Subject: net_sched: remove internal use of TC_POLICE_* These should be gone when we removed CONFIG_NET_CLS_POLICE. We can not totally remove them since they are exposed to userspace. Cc: Jamal Hadi Salim Signed-off-by: Cong Wang Acked-by: Jamal Hadi Salim Signed-off-by: David S. Miller diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index e7c0f4d..8998a35 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -243,7 +243,7 @@ static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a, default: net_notice_ratelimited("tc filter: Bogus netfilter code %d assume ACCEPT\n", ret); - result = TC_POLICE_OK; + result = TC_ACT_OK; break; } spin_unlock(&ipt->tcf_lock); diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index 0785b23..e04ea69 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -366,7 +366,7 @@ static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch) int ret = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; pr_debug("atm_tc_enqueue(skb %p,sch %p,[qdisc %p])\n", skb, sch, p); - result = TC_POLICE_OK; /* be nice to gcc */ + result = TC_ACT_OK; /* be nice to gcc */ flow = NULL; if (TC_H_MAJ(skb->priority) != sch->handle || !(flow = (struct atm_flow_data *)atm_tc_get(sch, skb->priority))) { @@ -403,7 +403,7 @@ done: case TC_ACT_SHOT: kfree_skb(skb); goto drop; - case TC_POLICE_RECLASSIFY: + case TC_ACT_RECLASSIFY: if (flow->excess) flow = flow->excess; else -- cgit v0.10.2 From d9fa17ef9f084c755332898c8243a396ea02d73e Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Mon, 13 Jun 2016 10:47:44 -0700 Subject: act_police: rename tcf_act_police_locate() to tcf_act_police_init() This function is just ->init(), rename it to make it obvious. Cc: Jamal Hadi Salim Signed-off-by: Cong Wang Acked-by: Jamal Hadi Salim Signed-off-by: David S. Miller diff --git a/net/sched/act_police.c b/net/sched/act_police.c index ff34dd3..1e8ede3 100644 --- a/net/sched/act_police.c +++ b/net/sched/act_police.c @@ -115,9 +115,9 @@ static const struct nla_policy police_policy[TCA_POLICE_MAX + 1] = { [TCA_POLICE_RESULT] = { .type = NLA_U32 }, }; -static int tcf_act_police_locate(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, - int ovr, int bind) +static int tcf_act_police_init(struct net *net, struct nlattr *nla, + struct nlattr *est, struct tc_action *a, + int ovr, int bind) { int ret = 0, err; struct nlattr *tb[TCA_POLICE_MAX + 1]; @@ -366,7 +366,7 @@ static struct tc_action_ops act_police_ops = { .owner = THIS_MODULE, .act = tcf_act_police, .dump = tcf_act_police_dump, - .init = tcf_act_police_locate, + .init = tcf_act_police_init, .walk = tcf_act_police_walker, .lookup = tcf_police_search, }; -- cgit v0.10.2 From be6e6707f6eec2048d9be608bc0ceecde5bd4cef Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:32 +0100 Subject: rxrpc: Rework peer object handling to use hash table and RCU Rework peer object handling to use a hash table instead of a flat list and to use RCU. Peer objects are no longer destroyed by passing them to a workqueue to process, but rather are just passed to the RCU garbage collector as kfree'able objects. The hash function uses the local endpoint plus all the components of the remote address, except for the RxRPC service ID. Peers thus represent a UDP port on the remote machine as contacted by a UDP port on this machine. The RCU read lock is used to handle non-creating lookups so that they can be called from bottom half context in the sk_error_report handler without having to lock the hash table against modification. rxrpc_lookup_peer_rcu() *does* take a reference on the peer object as in the future, this will be passed to a work item for error distribution in the error_report path and this function will cease being used in the data_ready path. Creating lookups are done under spinlock rather than mutex as they might be set up due to an external stimulus if the local endpoint is a server. Captured network error messages (ICMP) are handled with respect to this struct and MTU size and RTT are cached here. Signed-off-by: David Howells diff --git a/net/rxrpc/Makefile b/net/rxrpc/Makefile index 7e1006a..a6f6f21 100644 --- a/net/rxrpc/Makefile +++ b/net/rxrpc/Makefile @@ -20,7 +20,8 @@ af-rxrpc-y := \ recvmsg.o \ security.o \ skbuff.o \ - transport.o + transport.o \ + utils.o af-rxrpc-$(CONFIG_PROC_FS) += proc.o af-rxrpc-$(CONFIG_RXKAD) += rxkad.o diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index a1bcb0e..ba373ca 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -244,7 +244,7 @@ struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *rx, return ERR_PTR(-EAFNOSUPPORT); /* find a remote transport endpoint from the local one */ - peer = rxrpc_get_peer(srx, gfp); + peer = rxrpc_lookup_peer(rx->local, srx, gfp); if (IS_ERR(peer)) return ERR_CAST(peer); @@ -835,7 +835,6 @@ static void __exit af_rxrpc_exit(void) rxrpc_destroy_all_calls(); rxrpc_destroy_all_connections(); rxrpc_destroy_all_transports(); - rxrpc_destroy_all_peers(); rxrpc_destroy_all_locals(); ASSERTCMP(atomic_read(&rxrpc_n_skbs), ==, 0); diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 03919b9..7dba667 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -9,7 +9,9 @@ * 2 of the License, or (at your option) any later version. */ +#include #include +#include #include #if 0 @@ -193,15 +195,16 @@ struct rxrpc_local { /* * RxRPC remote transport endpoint definition - * - matched by remote port, address and protocol type - * - holds the connection ID counter for connections between the two endpoints + * - matched by local endpoint, remote port, address and protocol type */ struct rxrpc_peer { - struct work_struct destroyer; /* peer destroyer */ - struct list_head link; /* link in master peer list */ + struct rcu_head rcu; /* This must be first */ + atomic_t usage; + unsigned long hash_key; + struct hlist_node hash_link; + struct rxrpc_local *local; struct list_head error_targets; /* targets for net error distribution */ spinlock_t lock; /* access lock */ - atomic_t usage; unsigned int if_mtu; /* interface MTU for this peer */ unsigned int mtu; /* network MTU for this peer */ unsigned int maxdata; /* data size (MTU - hdrsize) */ @@ -611,10 +614,29 @@ void rxrpc_UDP_error_handler(struct work_struct *); /* * peer_object.c */ -struct rxrpc_peer *rxrpc_get_peer(struct sockaddr_rxrpc *, gfp_t); -void rxrpc_put_peer(struct rxrpc_peer *); -struct rxrpc_peer *rxrpc_find_peer(struct rxrpc_local *, __be32, __be16); -void __exit rxrpc_destroy_all_peers(void); +struct rxrpc_peer *rxrpc_lookup_peer_rcu(struct rxrpc_local *, + const struct sockaddr_rxrpc *); +struct rxrpc_peer *rxrpc_lookup_peer(struct rxrpc_local *, + struct sockaddr_rxrpc *, gfp_t); +struct rxrpc_peer *rxrpc_alloc_peer(struct rxrpc_local *, gfp_t); + +static inline void rxrpc_get_peer(struct rxrpc_peer *peer) +{ + atomic_inc(&peer->usage); +} + +static inline +struct rxrpc_peer *rxrpc_get_peer_maybe(struct rxrpc_peer *peer) +{ + return atomic_inc_not_zero(&peer->usage) ? peer : NULL; +} + +extern void __rxrpc_put_peer(struct rxrpc_peer *peer); +static inline void rxrpc_put_peer(struct rxrpc_peer *peer) +{ + if (atomic_dec_and_test(&peer->usage)) + __rxrpc_put_peer(peer); +} /* * proc.c @@ -673,6 +695,12 @@ struct rxrpc_transport *rxrpc_find_transport(struct rxrpc_local *, struct rxrpc_peer *); /* + * utils.c + */ +void rxrpc_get_addr_from_skb(struct rxrpc_local *, const struct sk_buff *, + struct sockaddr_rxrpc *); + +/* * debug tracing */ extern unsigned int rxrpc_debug; diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index eea5f4a..e5723f4 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -95,7 +95,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, rxrpc_new_skb(notification); notification->mark = RXRPC_SKB_MARK_NEW_CALL; - peer = rxrpc_get_peer(srx, GFP_NOIO); + peer = rxrpc_lookup_peer(local, srx, GFP_NOIO); if (IS_ERR(peer)) { _debug("no peer"); ret = -EBUSY; diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index e0815a0..3b405db 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -635,14 +635,16 @@ static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local, struct rxrpc_peer *peer; struct rxrpc_transport *trans; struct rxrpc_connection *conn; + struct sockaddr_rxrpc srx; - peer = rxrpc_find_peer(local, ip_hdr(skb)->saddr, - udp_hdr(skb)->source); + rxrpc_get_addr_from_skb(local, skb, &srx); + rcu_read_lock(); + peer = rxrpc_lookup_peer_rcu(local, &srx); if (IS_ERR(peer)) - goto cant_find_conn; + goto cant_find_peer; trans = rxrpc_find_transport(local, peer); - rxrpc_put_peer(peer); + rcu_read_unlock(); if (!trans) goto cant_find_conn; @@ -652,6 +654,9 @@ static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local, goto cant_find_conn; return conn; + +cant_find_peer: + rcu_read_unlock(); cant_find_conn: return NULL; } diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 3e82d6f..24f5ec0 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -23,6 +23,55 @@ #include "ar-internal.h" /* + * Find the peer associated with an ICMP packet. + */ +static struct rxrpc_peer *rxrpc_lookup_peer_icmp_rcu(struct rxrpc_local *local, + const struct sk_buff *skb) +{ + struct sock_exterr_skb *serr = SKB_EXT_ERR(skb); + struct sockaddr_rxrpc srx; + + _enter(""); + + memset(&srx, 0, sizeof(srx)); + srx.transport_type = local->srx.transport_type; + srx.transport.family = local->srx.transport.family; + + /* Can we see an ICMP4 packet on an ICMP6 listening socket? and vice + * versa? + */ + switch (srx.transport.family) { + case AF_INET: + srx.transport.sin.sin_port = serr->port; + srx.transport_len = sizeof(struct sockaddr_in); + switch (serr->ee.ee_origin) { + case SO_EE_ORIGIN_ICMP: + _net("Rx ICMP"); + memcpy(&srx.transport.sin.sin_addr, + skb_network_header(skb) + serr->addr_offset, + sizeof(struct in_addr)); + break; + case SO_EE_ORIGIN_ICMP6: + _net("Rx ICMP6 on v4 sock"); + memcpy(&srx.transport.sin.sin_addr, + skb_network_header(skb) + serr->addr_offset + 12, + sizeof(struct in_addr)); + break; + default: + memcpy(&srx.transport.sin.sin_addr, &ip_hdr(skb)->saddr, + sizeof(struct in_addr)); + break; + } + break; + + default: + BUG(); + } + + return rxrpc_lookup_peer_rcu(local, &srx); +} + +/* * handle an error received on the local endpoint */ void rxrpc_UDP_error_report(struct sock *sk) @@ -57,8 +106,12 @@ void rxrpc_UDP_error_report(struct sock *sk) _net("Rx UDP Error from %pI4:%hu", &addr, ntohs(port)); _debug("Msg l:%d d:%d", skb->len, skb->data_len); - peer = rxrpc_find_peer(local, addr, port); - if (IS_ERR(peer)) { + rcu_read_lock(); + peer = rxrpc_lookup_peer_icmp_rcu(local, skb); + if (peer && !rxrpc_get_peer_maybe(peer)) + peer = NULL; + if (!peer) { + rcu_read_unlock(); rxrpc_free_skb(skb); _leave(" [no peer]"); return; @@ -66,6 +119,7 @@ void rxrpc_UDP_error_report(struct sock *sk) trans = rxrpc_find_transport(local, peer); if (!trans) { + rcu_read_unlock(); rxrpc_put_peer(peer); rxrpc_free_skb(skb); _leave(" [no trans]"); @@ -110,6 +164,7 @@ void rxrpc_UDP_error_report(struct sock *sk) } } + rcu_read_unlock(); rxrpc_put_peer(peer); /* pass the transport ref to error_handler to release */ diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c index 0b54cda..7fc50dc 100644 --- a/net/rxrpc/peer_object.c +++ b/net/rxrpc/peer_object.c @@ -1,6 +1,6 @@ -/* RxRPC remote transport endpoint management +/* RxRPC remote transport endpoint record management * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2007, 2016 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -16,20 +16,132 @@ #include #include #include -#include -#include #include +#include #include #include #include #include #include "ar-internal.h" -static LIST_HEAD(rxrpc_peers); -static DEFINE_RWLOCK(rxrpc_peer_lock); -static DECLARE_WAIT_QUEUE_HEAD(rxrpc_peer_wq); +static DEFINE_HASHTABLE(rxrpc_peer_hash, 10); +static DEFINE_SPINLOCK(rxrpc_peer_hash_lock); -static void rxrpc_destroy_peer(struct work_struct *work); +/* + * Hash a peer key. + */ +static unsigned long rxrpc_peer_hash_key(struct rxrpc_local *local, + const struct sockaddr_rxrpc *srx) +{ + const u16 *p; + unsigned int i, size; + unsigned long hash_key; + + _enter(""); + + hash_key = (unsigned long)local / __alignof__(*local); + hash_key += srx->transport_type; + hash_key += srx->transport_len; + hash_key += srx->transport.family; + + switch (srx->transport.family) { + case AF_INET: + hash_key += (u16 __force)srx->transport.sin.sin_port; + size = sizeof(srx->transport.sin.sin_addr); + p = (u16 *)&srx->transport.sin.sin_addr; + break; + } + + /* Step through the peer address in 16-bit portions for speed */ + for (i = 0; i < size; i += sizeof(*p), p++) + hash_key += *p; + + _leave(" 0x%lx", hash_key); + return hash_key; +} + +/* + * Compare a peer to a key. Return -ve, 0 or +ve to indicate less than, same + * or greater than. + * + * Unfortunately, the primitives in linux/hashtable.h don't allow for sorted + * buckets and mid-bucket insertion, so we don't make full use of this + * information at this point. + */ +static long rxrpc_peer_cmp_key(const struct rxrpc_peer *peer, + struct rxrpc_local *local, + const struct sockaddr_rxrpc *srx, + unsigned long hash_key) +{ + long diff; + + diff = ((peer->hash_key - hash_key) ?: + ((unsigned long)peer->local - (unsigned long)local) ?: + (peer->srx.transport_type - srx->transport_type) ?: + (peer->srx.transport_len - srx->transport_len) ?: + (peer->srx.transport.family - srx->transport.family)); + if (diff != 0) + return diff; + + switch (srx->transport.family) { + case AF_INET: + return ((u16 __force)peer->srx.transport.sin.sin_port - + (u16 __force)srx->transport.sin.sin_port) ?: + memcmp(&peer->srx.transport.sin.sin_addr, + &srx->transport.sin.sin_addr, + sizeof(struct in_addr)); + default: + BUG(); + } +} + +/* + * Look up a remote transport endpoint for the specified address using RCU. + */ +static struct rxrpc_peer *__rxrpc_lookup_peer_rcu( + struct rxrpc_local *local, + const struct sockaddr_rxrpc *srx, + unsigned long hash_key) +{ + struct rxrpc_peer *peer; + + hash_for_each_possible_rcu(rxrpc_peer_hash, peer, hash_link, hash_key) { + if (rxrpc_peer_cmp_key(peer, local, srx, hash_key) == 0) { + if (atomic_read(&peer->usage) == 0) + return NULL; + return peer; + } + } + + return NULL; +} + +/* + * Look up a remote transport endpoint for the specified address using RCU. + */ +struct rxrpc_peer *rxrpc_lookup_peer_rcu(struct rxrpc_local *local, + const struct sockaddr_rxrpc *srx) +{ + struct rxrpc_peer *peer; + unsigned long hash_key = rxrpc_peer_hash_key(local, srx); + + peer = __rxrpc_lookup_peer_rcu(local, srx, hash_key); + if (peer) { + switch (srx->transport.family) { + case AF_INET: + _net("PEER %d {%d,%u,%pI4+%hu}", + peer->debug_id, + peer->srx.transport_type, + peer->srx.transport.family, + &peer->srx.transport.sin.sin_addr, + ntohs(peer->srx.transport.sin.sin_port)); + break; + } + + _leave(" = %p {u=%d}", peer, atomic_read(&peer->usage)); + } + return peer; +} /* * assess the MTU size for the network interface through which this peer is @@ -58,10 +170,9 @@ static void rxrpc_assess_MTU_size(struct rxrpc_peer *peer) } /* - * allocate a new peer + * Allocate a peer. */ -static struct rxrpc_peer *rxrpc_alloc_peer(struct sockaddr_rxrpc *srx, - gfp_t gfp) +struct rxrpc_peer *rxrpc_alloc_peer(struct rxrpc_local *local, gfp_t gfp) { struct rxrpc_peer *peer; @@ -69,12 +180,32 @@ static struct rxrpc_peer *rxrpc_alloc_peer(struct sockaddr_rxrpc *srx, peer = kzalloc(sizeof(struct rxrpc_peer), gfp); if (peer) { - INIT_WORK(&peer->destroyer, &rxrpc_destroy_peer); - INIT_LIST_HEAD(&peer->link); + atomic_set(&peer->usage, 1); + peer->local = local; INIT_LIST_HEAD(&peer->error_targets); spin_lock_init(&peer->lock); - atomic_set(&peer->usage, 1); peer->debug_id = atomic_inc_return(&rxrpc_debug_id); + } + + _leave(" = %p", peer); + return peer; +} + +/* + * Set up a new peer. + */ +static struct rxrpc_peer *rxrpc_create_peer(struct rxrpc_local *local, + struct sockaddr_rxrpc *srx, + unsigned long hash_key, + gfp_t gfp) +{ + struct rxrpc_peer *peer; + + _enter(""); + + peer = rxrpc_alloc_peer(local, gfp); + if (peer) { + peer->hash_key = hash_key; memcpy(&peer->srx, srx, sizeof(*srx)); rxrpc_assess_MTU_size(peer); @@ -105,11 +236,11 @@ static struct rxrpc_peer *rxrpc_alloc_peer(struct sockaddr_rxrpc *srx, /* * obtain a remote transport endpoint for the specified address */ -struct rxrpc_peer *rxrpc_get_peer(struct sockaddr_rxrpc *srx, gfp_t gfp) +struct rxrpc_peer *rxrpc_lookup_peer(struct rxrpc_local *local, + struct sockaddr_rxrpc *srx, gfp_t gfp) { struct rxrpc_peer *peer, *candidate; - const char *new = "old"; - int usage; + unsigned long hash_key = rxrpc_peer_hash_key(local, srx); _enter("{%d,%d,%pI4+%hu}", srx->transport_type, @@ -118,188 +249,60 @@ struct rxrpc_peer *rxrpc_get_peer(struct sockaddr_rxrpc *srx, gfp_t gfp) ntohs(srx->transport.sin.sin_port)); /* search the peer list first */ - read_lock_bh(&rxrpc_peer_lock); - list_for_each_entry(peer, &rxrpc_peers, link) { - _debug("check PEER %d { u=%d t=%d l=%d }", - peer->debug_id, - atomic_read(&peer->usage), - peer->srx.transport_type, - peer->srx.transport_len); - - if (atomic_read(&peer->usage) > 0 && - peer->srx.transport_type == srx->transport_type && - peer->srx.transport_len == srx->transport_len && - memcmp(&peer->srx.transport, - &srx->transport, - srx->transport_len) == 0) - goto found_extant_peer; - } - read_unlock_bh(&rxrpc_peer_lock); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_peer(srx, gfp); - if (!candidate) { - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } + rcu_read_lock(); + peer = __rxrpc_lookup_peer_rcu(local, srx, hash_key); + if (peer && !rxrpc_get_peer_maybe(peer)) + peer = NULL; + rcu_read_unlock(); + + if (!peer) { + /* The peer is not yet present in hash - create a candidate + * for a new record and then redo the search. + */ + candidate = rxrpc_create_peer(local, srx, hash_key, gfp); + if (!candidate) { + _leave(" = NULL [nomem]"); + return NULL; + } - write_lock_bh(&rxrpc_peer_lock); + spin_lock(&rxrpc_peer_hash_lock); - list_for_each_entry(peer, &rxrpc_peers, link) { - if (atomic_read(&peer->usage) > 0 && - peer->srx.transport_type == srx->transport_type && - peer->srx.transport_len == srx->transport_len && - memcmp(&peer->srx.transport, - &srx->transport, - srx->transport_len) == 0) - goto found_extant_second; - } + /* Need to check that we aren't racing with someone else */ + peer = __rxrpc_lookup_peer_rcu(local, srx, hash_key); + if (peer && !rxrpc_get_peer_maybe(peer)) + peer = NULL; + if (!peer) + hash_add_rcu(rxrpc_peer_hash, + &candidate->hash_link, hash_key); - /* we can now add the new candidate to the list */ - peer = candidate; - candidate = NULL; - usage = atomic_read(&peer->usage); + spin_unlock(&rxrpc_peer_hash_lock); - list_add_tail(&peer->link, &rxrpc_peers); - write_unlock_bh(&rxrpc_peer_lock); - new = "new"; + if (peer) + kfree(candidate); + else + peer = candidate; + } -success: - _net("PEER %s %d {%d,%u,%pI4+%hu}", - new, + _net("PEER %d {%d,%pI4+%hu}", peer->debug_id, peer->srx.transport_type, - peer->srx.transport.family, &peer->srx.transport.sin.sin_addr, ntohs(peer->srx.transport.sin.sin_port)); - _leave(" = %p {u=%d}", peer, usage); - return peer; - - /* we found the peer in the list immediately */ -found_extant_peer: - usage = atomic_inc_return(&peer->usage); - read_unlock_bh(&rxrpc_peer_lock); - goto success; - - /* we found the peer on the second time through the list */ -found_extant_second: - usage = atomic_inc_return(&peer->usage); - write_unlock_bh(&rxrpc_peer_lock); - kfree(candidate); - goto success; -} - -/* - * find the peer associated with a packet - */ -struct rxrpc_peer *rxrpc_find_peer(struct rxrpc_local *local, - __be32 addr, __be16 port) -{ - struct rxrpc_peer *peer; - - _enter(""); - - /* search the peer list */ - read_lock_bh(&rxrpc_peer_lock); - - if (local->srx.transport.family == AF_INET && - local->srx.transport_type == SOCK_DGRAM - ) { - list_for_each_entry(peer, &rxrpc_peers, link) { - if (atomic_read(&peer->usage) > 0 && - peer->srx.transport_type == SOCK_DGRAM && - peer->srx.transport.family == AF_INET && - peer->srx.transport.sin.sin_port == port && - peer->srx.transport.sin.sin_addr.s_addr == addr) - goto found_UDP_peer; - } - - goto new_UDP_peer; - } - - read_unlock_bh(&rxrpc_peer_lock); - _leave(" = -EAFNOSUPPORT"); - return ERR_PTR(-EAFNOSUPPORT); - -found_UDP_peer: - _net("Rx UDP DGRAM from peer %d", peer->debug_id); - atomic_inc(&peer->usage); - read_unlock_bh(&rxrpc_peer_lock); - _leave(" = %p", peer); + _leave(" = %p {u=%d}", peer, atomic_read(&peer->usage)); return peer; - -new_UDP_peer: - _net("Rx UDP DGRAM from NEW peer"); - read_unlock_bh(&rxrpc_peer_lock); - _leave(" = -EBUSY [new]"); - return ERR_PTR(-EBUSY); } /* - * release a remote transport endpoint + * Discard a ref on a remote peer record. */ -void rxrpc_put_peer(struct rxrpc_peer *peer) +void __rxrpc_put_peer(struct rxrpc_peer *peer) { - _enter("%p{u=%d}", peer, atomic_read(&peer->usage)); + ASSERT(list_empty(&peer->error_targets)); - ASSERTCMP(atomic_read(&peer->usage), >, 0); - - if (likely(!atomic_dec_and_test(&peer->usage))) { - _leave(" [in use]"); - return; - } - - rxrpc_queue_work(&peer->destroyer); - _leave(""); -} - -/* - * destroy a remote transport endpoint - */ -static void rxrpc_destroy_peer(struct work_struct *work) -{ - struct rxrpc_peer *peer = - container_of(work, struct rxrpc_peer, destroyer); - - _enter("%p{%d}", peer, atomic_read(&peer->usage)); - - write_lock_bh(&rxrpc_peer_lock); - list_del(&peer->link); - write_unlock_bh(&rxrpc_peer_lock); - - _net("DESTROY PEER %d", peer->debug_id); - kfree(peer); - - if (list_empty(&rxrpc_peers)) - wake_up_all(&rxrpc_peer_wq); - _leave(""); -} - -/* - * preemptively destroy all the peer records from a transport endpoint rather - * than waiting for them to time out - */ -void __exit rxrpc_destroy_all_peers(void) -{ - DECLARE_WAITQUEUE(myself,current); - - _enter(""); - - /* we simply have to wait for them to go away */ - if (!list_empty(&rxrpc_peers)) { - set_current_state(TASK_UNINTERRUPTIBLE); - add_wait_queue(&rxrpc_peer_wq, &myself); - - while (!list_empty(&rxrpc_peers)) { - schedule(); - set_current_state(TASK_UNINTERRUPTIBLE); - } - - remove_wait_queue(&rxrpc_peer_wq, &myself); - set_current_state(TASK_RUNNING); - } + spin_lock(&rxrpc_peer_hash_lock); + hash_del_rcu(&peer->hash_link); + spin_unlock(&rxrpc_peer_hash_lock); - _leave(""); + kfree_rcu(peer, rcu); } diff --git a/net/rxrpc/transport.c b/net/rxrpc/transport.c index a1b6518..d33387d 100644 --- a/net/rxrpc/transport.c +++ b/net/rxrpc/transport.c @@ -121,7 +121,7 @@ struct rxrpc_transport *rxrpc_get_transport(struct rxrpc_local *local, usage = atomic_read(&trans->usage); rxrpc_get_local(trans->local); - atomic_inc(&trans->peer->usage); + rxrpc_get_peer(trans->peer); list_add_tail(&trans->link, &rxrpc_transports); write_unlock_bh(&rxrpc_transport_lock); new = "new"; diff --git a/net/rxrpc/utils.c b/net/rxrpc/utils.c new file mode 100644 index 0000000..f28122a --- /dev/null +++ b/net/rxrpc/utils.c @@ -0,0 +1,41 @@ +/* Utility routines + * + * Copyright (C) 2015 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#include +#include +#include "ar-internal.h" + +/* + * Set up an RxRPC address from a socket buffer. + */ +void rxrpc_get_addr_from_skb(struct rxrpc_local *local, + const struct sk_buff *skb, + struct sockaddr_rxrpc *srx) +{ + memset(srx, 0, sizeof(*srx)); + srx->transport_type = local->srx.transport_type; + srx->transport.family = local->srx.transport.family; + + /* Can we see an ipv4 UDP packet on an ipv6 UDP socket? and vice + * versa? + */ + switch (srx->transport.family) { + case AF_INET: + srx->transport.sin.sin_port = udp_hdr(skb)->source; + srx->transport_len = sizeof(struct sockaddr_in); + memcpy(&srx->transport.sin.sin_addr, &ip_hdr(skb)->saddr, + sizeof(struct in_addr)); + break; + + default: + BUG(); + } +} -- cgit v0.10.2 From abe89ef0ed1a50ef6186d9aee433b995641a1293 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:32 +0100 Subject: rxrpc: Rename rxrpc_UDP_error_report() to rxrpc_error_report() Rename rxrpc_UDP_error_report() to rxrpc_error_report() as it might get called for something other than UDP. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 7dba667..1e5c156 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -606,9 +606,9 @@ int rxrpc_send_packet(struct rxrpc_transport *, struct sk_buff *); int rxrpc_do_sendmsg(struct rxrpc_sock *, struct msghdr *, size_t); /* - * peer_error.c + * peer_event.c */ -void rxrpc_UDP_error_report(struct sock *); +void rxrpc_error_report(struct sock *); void rxrpc_UDP_error_handler(struct work_struct *); /* diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c index 111f250..28f9efb 100644 --- a/net/rxrpc/local_object.c +++ b/net/rxrpc/local_object.c @@ -120,7 +120,7 @@ static int rxrpc_create_local(struct rxrpc_local *local) sock = local->socket->sk; sock->sk_user_data = local; sock->sk_data_ready = rxrpc_data_ready; - sock->sk_error_report = rxrpc_UDP_error_report; + sock->sk_error_report = rxrpc_error_report; _leave(" = 0"); return 0; diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 24f5ec0..2c2df3a 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -74,7 +74,7 @@ static struct rxrpc_peer *rxrpc_lookup_peer_icmp_rcu(struct rxrpc_local *local, /* * handle an error received on the local endpoint */ -void rxrpc_UDP_error_report(struct sock *sk) +void rxrpc_error_report(struct sock *sk) { struct sock_exterr_skb *serr; struct rxrpc_transport *trans; -- cgit v0.10.2 From 1a70c05bad1383fdda95e713baee5f76c4726d24 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:33 +0100 Subject: rxrpc: Break MTU determination from ICMP into its own function Break MTU determination from ICMP out into its own function to reduce the complexity of the error report handler. Signed-off-by: David Howells diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 2c2df3a..80de842 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -72,6 +72,45 @@ static struct rxrpc_peer *rxrpc_lookup_peer_icmp_rcu(struct rxrpc_local *local, } /* + * Handle an MTU/fragmentation problem. + */ +static void rxrpc_adjust_mtu(struct rxrpc_peer *peer, struct sock_exterr_skb *serr) +{ + u32 mtu = serr->ee.ee_info; + + _net("Rx ICMP Fragmentation Needed (%d)", mtu); + + /* wind down the local interface MTU */ + if (mtu > 0 && peer->if_mtu == 65535 && mtu < peer->if_mtu) { + peer->if_mtu = mtu; + _net("I/F MTU %u", mtu); + } + + if (mtu == 0) { + /* they didn't give us a size, estimate one */ + mtu = peer->if_mtu; + if (mtu > 1500) { + mtu >>= 1; + if (mtu < 1500) + mtu = 1500; + } else { + mtu -= 100; + if (mtu < peer->hdrsize) + mtu = peer->hdrsize + 4; + } + } + + if (mtu < peer->mtu) { + spin_lock_bh(&peer->lock); + peer->mtu = mtu; + peer->maxdata = peer->mtu - peer->hdrsize; + spin_unlock_bh(&peer->lock); + _net("Net MTU %u (maxdata %u)", + peer->mtu, peer->maxdata); + } +} + +/* * handle an error received on the local endpoint */ void rxrpc_error_report(struct sock *sk) @@ -126,50 +165,26 @@ void rxrpc_error_report(struct sock *sk) return; } - if (serr->ee.ee_origin == SO_EE_ORIGIN_ICMP && - serr->ee.ee_type == ICMP_DEST_UNREACH && - serr->ee.ee_code == ICMP_FRAG_NEEDED - ) { - u32 mtu = serr->ee.ee_info; - - _net("Rx Received ICMP Fragmentation Needed (%d)", mtu); - - /* wind down the local interface MTU */ - if (mtu > 0 && peer->if_mtu == 65535 && mtu < peer->if_mtu) { - peer->if_mtu = mtu; - _net("I/F MTU %u", mtu); - } - - if (mtu == 0) { - /* they didn't give us a size, estimate one */ - mtu = peer->if_mtu; - if (mtu > 1500) { - mtu >>= 1; - if (mtu < 1500) - mtu = 1500; - } else { - mtu -= 100; - if (mtu < peer->hdrsize) - mtu = peer->hdrsize + 4; - } - } - - if (mtu < peer->mtu) { - spin_lock_bh(&peer->lock); - peer->mtu = mtu; - peer->maxdata = peer->mtu - peer->hdrsize; - spin_unlock_bh(&peer->lock); - _net("Net MTU %u (maxdata %u)", - peer->mtu, peer->maxdata); - } + if ((serr->ee.ee_origin == SO_EE_ORIGIN_ICMP && + serr->ee.ee_type == ICMP_DEST_UNREACH && + serr->ee.ee_code == ICMP_FRAG_NEEDED)) { + rxrpc_adjust_mtu(peer, serr); + rxrpc_free_skb(skb); + skb = NULL; + goto out; } +out: rcu_read_unlock(); rxrpc_put_peer(peer); - /* pass the transport ref to error_handler to release */ - skb_queue_tail(&trans->error_queue, skb); - rxrpc_queue_work(&trans->error_handler); + if (skb) { + /* pass the transport ref to error_handler to release */ + skb_queue_tail(&trans->error_queue, skb); + rxrpc_queue_work(&trans->error_handler); + } else { + rxrpc_put_transport(trans); + } _leave(""); } -- cgit v0.10.2 From 1c1df86fad68dd7188ea498e796c9d2ede679421 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:33 +0100 Subject: rxrpc: Don't assume anything about the address in an ICMP packet Don't assume anything about the address in an ICMP packet in rxrpc_error_report() as the address may not be IPv4 in future, especially since we're just printing these details. Signed-off-by: David Howells diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 80de842..6ba798d 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -120,8 +120,6 @@ void rxrpc_error_report(struct sock *sk) struct rxrpc_local *local = sk->sk_user_data; struct rxrpc_peer *peer; struct sk_buff *skb; - __be32 addr; - __be16 port; _enter("%p{%d}", sk, local->debug_id); @@ -139,12 +137,6 @@ void rxrpc_error_report(struct sock *sk) rxrpc_new_skb(skb); - addr = *(__be32 *)(skb_network_header(skb) + serr->addr_offset); - port = serr->port; - - _net("Rx UDP Error from %pI4:%hu", &addr, ntohs(port)); - _debug("Msg l:%d d:%d", skb->len, skb->data_len); - rcu_read_lock(); peer = rxrpc_lookup_peer_icmp_rcu(local, skb); if (peer && !rxrpc_get_peer_maybe(peer)) -- cgit v0.10.2 From fe77d5fc5ab33bb088cf8448767a77fdc32e08d1 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:34 +0100 Subject: rxrpc: Do a little bit of tidying in the ICMP processing Do a little bit of tidying in the ICMP processing code. Signed-off-by: David Howells diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 6ba798d..31c440a 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -245,15 +245,13 @@ void rxrpc_UDP_error_handler(struct work_struct *work) break; case SO_EE_ORIGIN_LOCAL: - _proto("Rx Received local error { error=%d }", - ee->ee_errno); + _proto("Rx Received local error { error=%d }", err); break; case SO_EE_ORIGIN_NONE: case SO_EE_ORIGIN_ICMP6: default: - _proto("Rx Received error report { orig=%u }", - ee->ee_origin); + _proto("Rx Received error report { orig=%u }", ee->ee_origin); break; } -- cgit v0.10.2 From f66d7490196055cb9fb058f8936d19111a6231b9 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:34 +0100 Subject: rxrpc: Use the peer record to distribute network errors Use the peer record to distribute network errors rather than the transport object (which I want to get rid of). An error from a particular peer terminates all calls on that peer. For future consideration: (1) For ICMP-induced errors it might be worth trying to extract the RxRPC header from the offending packet, if one is returned attached to the ICMP packet, to better direct the error. This may be overkill, though, since an ICMP packet would be expected to be relating to the destination port, machine or network. RxRPC ABORT and BUSY packets give notice at RxRPC level. (2) To also abort connection-level communications (such as CHALLENGE packets) where indicted by an error - but that requires some revamping of the connection event handling first. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 1e5c156..a63bb75 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -189,7 +189,6 @@ struct rxrpc_local { rwlock_t services_lock; /* lock for services list */ atomic_t usage; int debug_id; /* debug ID for printks */ - volatile char error_rcvd; /* T if received ICMP error outstanding */ struct sockaddr_rxrpc srx; /* local address */ }; @@ -203,14 +202,16 @@ struct rxrpc_peer { unsigned long hash_key; struct hlist_node hash_link; struct rxrpc_local *local; - struct list_head error_targets; /* targets for net error distribution */ + struct hlist_head error_targets; /* targets for net error distribution */ + struct work_struct error_distributor; spinlock_t lock; /* access lock */ unsigned int if_mtu; /* interface MTU for this peer */ unsigned int mtu; /* network MTU for this peer */ unsigned int maxdata; /* data size (MTU - hdrsize) */ unsigned short hdrsize; /* header size (IP + UDP + RxRPC) */ int debug_id; /* debug ID for printks */ - int net_error; /* network error distributed */ + int error_report; /* Net (+0) or local (+1000000) to distribute */ +#define RXRPC_LOCAL_ERROR_OFFSET 1000000 struct sockaddr_rxrpc srx; /* remote address */ /* calculated RTT cache */ @@ -229,12 +230,10 @@ struct rxrpc_peer { struct rxrpc_transport { struct rxrpc_local *local; /* local transport endpoint */ struct rxrpc_peer *peer; /* remote transport endpoint */ - struct work_struct error_handler; /* network error distributor */ struct rb_root bundles; /* client connection bundles on this transport */ struct rb_root client_conns; /* client connections on this transport */ struct rb_root server_conns; /* server connections on this transport */ struct list_head link; /* link in master session list */ - struct sk_buff_head error_queue; /* error packets awaiting processing */ unsigned long put_time; /* time at which to reap */ spinlock_t client_lock; /* client connection allocation lock */ rwlock_t conn_lock; /* lock for active/dead connections */ @@ -393,7 +392,7 @@ struct rxrpc_call { struct work_struct destroyer; /* call destroyer */ struct work_struct processor; /* packet processor and ACK generator */ struct list_head link; /* link in master call list */ - struct list_head error_link; /* link in error distribution list */ + struct hlist_node error_link; /* link in error distribution list */ struct list_head accept_link; /* calls awaiting acceptance */ struct rb_node sock_node; /* node in socket call tree */ struct rb_node conn_node; /* node in connection call tree */ @@ -411,7 +410,8 @@ struct rxrpc_call { atomic_t sequence; /* Tx data packet sequence counter */ u32 local_abort; /* local abort code */ u32 remote_abort; /* remote abort code */ - int error; /* local error incurred */ + int error_report; /* Network error (ICMP/local transport) */ + int error; /* Local error incurred */ enum rxrpc_call_state state : 8; /* current state of call */ int debug_id; /* debug ID for printks */ u8 channel; /* connection channel occupied by this call */ @@ -609,7 +609,7 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *, struct msghdr *, size_t); * peer_event.c */ void rxrpc_error_report(struct sock *); -void rxrpc_UDP_error_handler(struct work_struct *); +void rxrpc_peer_error_distributor(struct work_struct *); /* * peer_object.c diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c index 1838178..e610b10 100644 --- a/net/rxrpc/call_event.c +++ b/net/rxrpc/call_event.c @@ -864,17 +864,24 @@ void rxrpc_process_call(struct work_struct *work) } if (test_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events)) { + enum rxrpc_skb_mark mark; int error; clear_bit(RXRPC_CALL_EV_CONN_ABORT, &call->events); clear_bit(RXRPC_CALL_EV_REJECT_BUSY, &call->events); clear_bit(RXRPC_CALL_EV_ABORT, &call->events); - error = call->conn->trans->peer->net_error; - _debug("post net error %d", error); + error = call->error_report; + if (error < RXRPC_LOCAL_ERROR_OFFSET) { + mark = RXRPC_SKB_MARK_NET_ERROR; + _debug("post net error %d", error); + } else { + mark = RXRPC_SKB_MARK_LOCAL_ERROR; + error -= RXRPC_LOCAL_ERROR_OFFSET; + _debug("post net local error %d", error); + } - if (rxrpc_post_message(call, RXRPC_SKB_MARK_NET_ERROR, - error, true) < 0) + if (rxrpc_post_message(call, mark, error, true) < 0) goto no_mem; clear_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events); goto kill_ACKs; diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 68125dc..8b4d47b 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -334,7 +334,7 @@ static struct rxrpc_call *rxrpc_alloc_client_call( rxrpc_call_hash_add(call); spin_lock(&call->conn->trans->peer->lock); - list_add(&call->error_link, &call->conn->trans->peer->error_targets); + hlist_add_head(&call->error_link, &call->conn->trans->peer->error_targets); spin_unlock(&call->conn->trans->peer->lock); call->lifetimer.expires = jiffies + rxrpc_max_call_lifetime; @@ -516,7 +516,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, write_unlock_bh(&conn->lock); spin_lock(&conn->trans->peer->lock); - list_add(&call->error_link, &conn->trans->peer->error_targets); + hlist_add_head(&call->error_link, &conn->trans->peer->error_targets); spin_unlock(&conn->trans->peer->lock); write_lock_bh(&rxrpc_call_lock); @@ -812,7 +812,7 @@ static void rxrpc_cleanup_call(struct rxrpc_call *call) if (call->conn) { spin_lock(&call->conn->trans->peer->lock); - list_del(&call->error_link); + hlist_del_init(&call->error_link); spin_unlock(&call->conn->trans->peer->lock); write_lock_bh(&call->conn->lock); diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index 2e3c406..e6fb386 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -707,7 +707,9 @@ out: call_aborted: rxrpc_free_skb(skb); if (call->state == RXRPC_CALL_NETWORK_ERROR) - ret = call->conn->trans->peer->net_error; + ret = call->error_report < RXRPC_LOCAL_ERROR_OFFSET ? + call->error_report : + call->error_report - RXRPC_LOCAL_ERROR_OFFSET; else ret = -ECONNABORTED; _leave(" = %d", ret); diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 31c440a..8940674 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -1,4 +1,4 @@ -/* Error message handling (ICMP) +/* Peer event handling, typically ICMP messages. * * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) @@ -22,6 +22,8 @@ #include #include "ar-internal.h" +static void rxrpc_store_error(struct rxrpc_peer *, struct sock_exterr_skb *); + /* * Find the peer associated with an ICMP packet. */ @@ -111,12 +113,11 @@ static void rxrpc_adjust_mtu(struct rxrpc_peer *peer, struct sock_exterr_skb *se } /* - * handle an error received on the local endpoint + * Handle an error received on the local endpoint. */ void rxrpc_error_report(struct sock *sk) { struct sock_exterr_skb *serr; - struct rxrpc_transport *trans; struct rxrpc_local *local = sk->sk_user_data; struct rxrpc_peer *peer; struct sk_buff *skb; @@ -148,57 +149,37 @@ void rxrpc_error_report(struct sock *sk) return; } - trans = rxrpc_find_transport(local, peer); - if (!trans) { - rcu_read_unlock(); - rxrpc_put_peer(peer); - rxrpc_free_skb(skb); - _leave(" [no trans]"); - return; - } - if ((serr->ee.ee_origin == SO_EE_ORIGIN_ICMP && serr->ee.ee_type == ICMP_DEST_UNREACH && serr->ee.ee_code == ICMP_FRAG_NEEDED)) { rxrpc_adjust_mtu(peer, serr); + rcu_read_unlock(); rxrpc_free_skb(skb); - skb = NULL; - goto out; + rxrpc_put_peer(peer); + _leave(" [MTU update]"); + return; } -out: + rxrpc_store_error(peer, serr); rcu_read_unlock(); - rxrpc_put_peer(peer); + rxrpc_free_skb(skb); - if (skb) { - /* pass the transport ref to error_handler to release */ - skb_queue_tail(&trans->error_queue, skb); - rxrpc_queue_work(&trans->error_handler); - } else { - rxrpc_put_transport(trans); - } + /* The ref we obtained is passed off to the work item */ + rxrpc_queue_work(&peer->error_distributor); _leave(""); } /* - * deal with UDP error messages + * Map an error report to error codes on the peer record. */ -void rxrpc_UDP_error_handler(struct work_struct *work) +static void rxrpc_store_error(struct rxrpc_peer *peer, + struct sock_exterr_skb *serr) { struct sock_extended_err *ee; - struct sock_exterr_skb *serr; - struct rxrpc_transport *trans = - container_of(work, struct rxrpc_transport, error_handler); - struct sk_buff *skb; int err; _enter(""); - skb = skb_dequeue(&trans->error_queue); - if (!skb) - return; - - serr = SKB_EXT_ERR(skb); ee = &serr->ee; _net("Rx Error o=%d t=%d c=%d e=%d", @@ -244,47 +225,57 @@ void rxrpc_UDP_error_handler(struct work_struct *work) } break; + case SO_EE_ORIGIN_NONE: case SO_EE_ORIGIN_LOCAL: _proto("Rx Received local error { error=%d }", err); + err += RXRPC_LOCAL_ERROR_OFFSET; break; - case SO_EE_ORIGIN_NONE: case SO_EE_ORIGIN_ICMP6: default: _proto("Rx Received error report { orig=%u }", ee->ee_origin); break; } - /* terminate all the affected calls if there's an unrecoverable - * error */ - if (err) { - struct rxrpc_call *call, *_n; + peer->error_report = err; +} + +/* + * Distribute an error that occurred on a peer + */ +void rxrpc_peer_error_distributor(struct work_struct *work) +{ + struct rxrpc_peer *peer = + container_of(work, struct rxrpc_peer, error_distributor); + struct rxrpc_call *call; + int error_report; + + _enter(""); - _debug("ISSUE ERROR %d", err); + error_report = READ_ONCE(peer->error_report); - spin_lock_bh(&trans->peer->lock); - trans->peer->net_error = err; + _debug("ISSUE ERROR %d", error_report); - list_for_each_entry_safe(call, _n, &trans->peer->error_targets, - error_link) { - write_lock(&call->state_lock); - if (call->state != RXRPC_CALL_COMPLETE && - call->state < RXRPC_CALL_NETWORK_ERROR) { - call->state = RXRPC_CALL_NETWORK_ERROR; - set_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events); - rxrpc_queue_call(call); - } - write_unlock(&call->state_lock); - list_del_init(&call->error_link); - } + spin_lock_bh(&peer->lock); - spin_unlock_bh(&trans->peer->lock); + while (!hlist_empty(&peer->error_targets)) { + call = hlist_entry(peer->error_targets.first, + struct rxrpc_call, error_link); + hlist_del_init(&call->error_link); + + write_lock(&call->state_lock); + if (call->state != RXRPC_CALL_COMPLETE && + call->state < RXRPC_CALL_NETWORK_ERROR) { + call->error_report = error_report; + call->state = RXRPC_CALL_NETWORK_ERROR; + set_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events); + rxrpc_queue_call(call); + } + write_unlock(&call->state_lock); } - if (!skb_queue_empty(&trans->error_queue)) - rxrpc_queue_work(&trans->error_handler); + spin_unlock_bh(&peer->lock); - rxrpc_free_skb(skb); - rxrpc_put_transport(trans); + rxrpc_put_peer(peer); _leave(""); } diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c index 7fc50dc..faf222c 100644 --- a/net/rxrpc/peer_object.c +++ b/net/rxrpc/peer_object.c @@ -182,7 +182,9 @@ struct rxrpc_peer *rxrpc_alloc_peer(struct rxrpc_local *local, gfp_t gfp) if (peer) { atomic_set(&peer->usage, 1); peer->local = local; - INIT_LIST_HEAD(&peer->error_targets); + INIT_HLIST_HEAD(&peer->error_targets); + INIT_WORK(&peer->error_distributor, + &rxrpc_peer_error_distributor); spin_lock_init(&peer->lock); peer->debug_id = atomic_inc_return(&rxrpc_debug_id); } @@ -298,7 +300,7 @@ struct rxrpc_peer *rxrpc_lookup_peer(struct rxrpc_local *local, */ void __rxrpc_put_peer(struct rxrpc_peer *peer) { - ASSERT(list_empty(&peer->error_targets)); + ASSERT(hlist_empty(&peer->error_targets)); spin_lock(&rxrpc_peer_hash_lock); hash_del_rcu(&peer->hash_link); diff --git a/net/rxrpc/transport.c b/net/rxrpc/transport.c index d33387d..24c7121 100644 --- a/net/rxrpc/transport.c +++ b/net/rxrpc/transport.c @@ -49,26 +49,11 @@ static struct rxrpc_transport *rxrpc_alloc_transport(struct rxrpc_local *local, trans->bundles = RB_ROOT; trans->client_conns = RB_ROOT; trans->server_conns = RB_ROOT; - skb_queue_head_init(&trans->error_queue); spin_lock_init(&trans->client_lock); rwlock_init(&trans->conn_lock); atomic_set(&trans->usage, 1); trans->conn_idcounter = peer->srx.srx_service << 16; trans->debug_id = atomic_inc_return(&rxrpc_debug_id); - - if (peer->srx.transport.family == AF_INET) { - switch (peer->srx.transport_type) { - case SOCK_DGRAM: - INIT_WORK(&trans->error_handler, - rxrpc_UDP_error_handler); - break; - default: - BUG(); - break; - } - } else { - BUG(); - } } _leave(" = %p", trans); @@ -210,8 +195,6 @@ static void rxrpc_cleanup_transport(struct rxrpc_transport *trans) { _net("DESTROY TRANS %d", trans->debug_id); - rxrpc_purge_queue(&trans->error_queue); - rxrpc_put_local(trans->local); rxrpc_put_peer(trans->peer); kfree(trans); -- cgit v0.10.2 From 875636163b4e694c092625ed98b17e10d582b3ca Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:34 +0100 Subject: rxrpc: Separate local endpoint event handling out into its own file Separate local endpoint event handling out into its own file preparatory to overhauling the object management aspect (which remains in the original file). Signed-off-by: David Howells diff --git a/net/rxrpc/Makefile b/net/rxrpc/Makefile index a6f6f21..b005027 100644 --- a/net/rxrpc/Makefile +++ b/net/rxrpc/Makefile @@ -12,6 +12,7 @@ af-rxrpc-y := \ input.o \ insecure.o \ key.o \ + local_event.o \ local_object.o \ misc.o \ output.o \ diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index a63bb75..fa50b09 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -573,6 +573,11 @@ int rxrpc_get_server_data_key(struct rxrpc_connection *, const void *, time_t, u32); /* + * local_event.c + */ +extern void rxrpc_process_local_events(struct work_struct *); + +/* * local_object.c */ extern rwlock_t rxrpc_local_lock; diff --git a/net/rxrpc/local_event.c b/net/rxrpc/local_event.c new file mode 100644 index 0000000..194db2e --- /dev/null +++ b/net/rxrpc/local_event.c @@ -0,0 +1,120 @@ +/* AF_RXRPC local endpoint management + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "ar-internal.h" + +static const char rxrpc_version_string[65] = "linux-" UTS_RELEASE " AF_RXRPC"; + +/* + * Reply to a version request + */ +static void rxrpc_send_version_request(struct rxrpc_local *local, + struct rxrpc_host_header *hdr, + struct sk_buff *skb) +{ + struct rxrpc_wire_header whdr; + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + struct sockaddr_in sin; + struct msghdr msg; + struct kvec iov[2]; + size_t len; + int ret; + + _enter(""); + + sin.sin_family = AF_INET; + sin.sin_port = udp_hdr(skb)->source; + sin.sin_addr.s_addr = ip_hdr(skb)->saddr; + + msg.msg_name = &sin; + msg.msg_namelen = sizeof(sin); + msg.msg_control = NULL; + msg.msg_controllen = 0; + msg.msg_flags = 0; + + whdr.epoch = htonl(sp->hdr.epoch); + whdr.cid = htonl(sp->hdr.cid); + whdr.callNumber = htonl(sp->hdr.callNumber); + whdr.seq = 0; + whdr.serial = 0; + whdr.type = RXRPC_PACKET_TYPE_VERSION; + whdr.flags = RXRPC_LAST_PACKET | (~hdr->flags & RXRPC_CLIENT_INITIATED); + whdr.userStatus = 0; + whdr.securityIndex = 0; + whdr._rsvd = 0; + whdr.serviceId = htons(sp->hdr.serviceId); + + iov[0].iov_base = &whdr; + iov[0].iov_len = sizeof(whdr); + iov[1].iov_base = (char *)rxrpc_version_string; + iov[1].iov_len = sizeof(rxrpc_version_string); + + len = iov[0].iov_len + iov[1].iov_len; + + _proto("Tx VERSION (reply)"); + + ret = kernel_sendmsg(local->socket, &msg, iov, 2, len); + if (ret < 0) + _debug("sendmsg failed: %d", ret); + + _leave(""); +} + +/* + * Process event packets targetted at a local endpoint. + */ +void rxrpc_process_local_events(struct work_struct *work) +{ + struct rxrpc_local *local = container_of(work, struct rxrpc_local, event_processor); + struct sk_buff *skb; + char v; + + _enter(""); + + atomic_inc(&local->usage); + + while ((skb = skb_dequeue(&local->event_queue))) { + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + + _debug("{%d},{%u}", local->debug_id, sp->hdr.type); + + switch (sp->hdr.type) { + case RXRPC_PACKET_TYPE_VERSION: + if (skb_copy_bits(skb, 0, &v, 1) < 0) + return; + _proto("Rx VERSION { %02x }", v); + if (v == 0) + rxrpc_send_version_request(local, &sp->hdr, skb); + break; + + default: + /* Just ignore anything we don't understand */ + break; + } + + rxrpc_put_local(local); + rxrpc_free_skb(skb); + } + + rxrpc_put_local(local); + _leave(""); +} diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c index 28f9efb..c1b8d745 100644 --- a/net/rxrpc/local_object.c +++ b/net/rxrpc/local_object.c @@ -1,12 +1,12 @@ -/* AF_RXRPC local endpoint management +/* Local endpoint object management * * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License + * modify it under the terms of the GNU General Public Licence * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. + * 2 of the Licence, or (at your option) any later version. */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt @@ -19,18 +19,14 @@ #include #include #include -#include #include "ar-internal.h" -static const char rxrpc_version_string[65] = "linux-" UTS_RELEASE " AF_RXRPC"; - static LIST_HEAD(rxrpc_locals); DEFINE_RWLOCK(rxrpc_local_lock); static DECLARE_RWSEM(rxrpc_local_sem); static DECLARE_WAIT_QUEUE_HEAD(rxrpc_local_wq); static void rxrpc_destroy_local(struct work_struct *work); -static void rxrpc_process_local_events(struct work_struct *work); /* * allocate a new local @@ -320,98 +316,3 @@ void __exit rxrpc_destroy_all_locals(void) _leave(""); } - -/* - * Reply to a version request - */ -static void rxrpc_send_version_request(struct rxrpc_local *local, - struct rxrpc_host_header *hdr, - struct sk_buff *skb) -{ - struct rxrpc_wire_header whdr; - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - struct sockaddr_in sin; - struct msghdr msg; - struct kvec iov[2]; - size_t len; - int ret; - - _enter(""); - - sin.sin_family = AF_INET; - sin.sin_port = udp_hdr(skb)->source; - sin.sin_addr.s_addr = ip_hdr(skb)->saddr; - - msg.msg_name = &sin; - msg.msg_namelen = sizeof(sin); - msg.msg_control = NULL; - msg.msg_controllen = 0; - msg.msg_flags = 0; - - whdr.epoch = htonl(sp->hdr.epoch); - whdr.cid = htonl(sp->hdr.cid); - whdr.callNumber = htonl(sp->hdr.callNumber); - whdr.seq = 0; - whdr.serial = 0; - whdr.type = RXRPC_PACKET_TYPE_VERSION; - whdr.flags = RXRPC_LAST_PACKET | (~hdr->flags & RXRPC_CLIENT_INITIATED); - whdr.userStatus = 0; - whdr.securityIndex = 0; - whdr._rsvd = 0; - whdr.serviceId = htons(sp->hdr.serviceId); - - iov[0].iov_base = &whdr; - iov[0].iov_len = sizeof(whdr); - iov[1].iov_base = (char *)rxrpc_version_string; - iov[1].iov_len = sizeof(rxrpc_version_string); - - len = iov[0].iov_len + iov[1].iov_len; - - _proto("Tx VERSION (reply)"); - - ret = kernel_sendmsg(local->socket, &msg, iov, 2, len); - if (ret < 0) - _debug("sendmsg failed: %d", ret); - - _leave(""); -} - -/* - * Process event packets targetted at a local endpoint. - */ -static void rxrpc_process_local_events(struct work_struct *work) -{ - struct rxrpc_local *local = container_of(work, struct rxrpc_local, event_processor); - struct sk_buff *skb; - char v; - - _enter(""); - - atomic_inc(&local->usage); - - while ((skb = skb_dequeue(&local->event_queue))) { - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - - _debug("{%d},{%u}", local->debug_id, sp->hdr.type); - - switch (sp->hdr.type) { - case RXRPC_PACKET_TYPE_VERSION: - if (skb_copy_bits(skb, 0, &v, 1) < 0) - return; - _proto("Rx VERSION { %02x }", v); - if (v == 0) - rxrpc_send_version_request(local, &sp->hdr, skb); - break; - - default: - /* Just ignore anything we don't understand */ - break; - } - - rxrpc_put_local(local); - rxrpc_free_skb(skb); - } - - rxrpc_put_local(local); - _leave(""); -} -- cgit v0.10.2 From 4f95dd78a77edc42454de55bb32332be293fb461 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:35 +0100 Subject: rxrpc: Rework local endpoint management Rework the local RxRPC endpoint management. Local endpoint objects are maintained in a flat list as before. This should be okay as there shouldn't be more than one per open AF_RXRPC socket (there can be fewer as local endpoints can be shared if their local service ID is 0 and they share the same local transport parameters). Changes: (1) Local endpoints may now only be shared if they have local service ID 0 (ie. they're not being used for listening). This prevents a scenario where process A is listening of the Cache Manager port and process B contacts a fileserver - which may then attempt to send CM requests back to B. But if A and B are sharing a local endpoint, A will get the CM requests meant for B. (2) We use a mutex to handle lookups and don't provide RCU-only lookups since we only expect to access the list when opening a socket or destroying an endpoint. The local endpoint object is pointed to by the transport socket's sk_user_data for the life of the transport socket - allowing us to refer to it directly from the sk_data_ready and sk_error_report callbacks. (3) atomic_inc_not_zero() now exists and can be used to only share a local endpoint if the last reference hasn't yet gone. (4) We can remove rxrpc_local_lock - a spinlock that had to be taken with BH processing disabled given that we assume sk_user_data won't change under us. (5) The transport socket is shut down before we clear the sk_user_data pointer so that we can be sure that the transport socket's callbacks won't be invoked once the RCU destruction is scheduled. (6) Local endpoints have a work item that handles both destruction and event processing. The means that destruction doesn't then need to wait for event processing. The event queues can then be cleared after the transport socket is shut down. (7) Local endpoints are no longer available for resurrection beyond the life of the sockets that had them open. As soon as their last ref goes, they are scheduled for destruction and may not have their usage count moved from 0. Signed-off-by: David Howells diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index ba373ca..c83c3c7 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -102,6 +102,8 @@ static int rxrpc_validate_address(struct rxrpc_sock *rx, switch (srx->transport.family) { case AF_INET: + if (srx->transport_len < sizeof(struct sockaddr_in)) + return -EINVAL; _debug("INET: %x @ %pI4", ntohs(srx->transport.sin.sin_port), &srx->transport.sin.sin_addr); @@ -835,12 +837,27 @@ static void __exit af_rxrpc_exit(void) rxrpc_destroy_all_calls(); rxrpc_destroy_all_connections(); rxrpc_destroy_all_transports(); - rxrpc_destroy_all_locals(); ASSERTCMP(atomic_read(&rxrpc_n_skbs), ==, 0); + /* We need to flush the scheduled work twice because the local endpoint + * records involve a work item in their destruction as they can only be + * destroyed from process context. However, a connection may have a + * work item outstanding - and this will pin the local endpoint record + * until the connection goes away. + * + * Peers don't pin locals and calls pin sockets - which prevents the + * module from being unloaded - so we should only need two flushes. + */ _debug("flush scheduled work"); flush_workqueue(rxrpc_workqueue); + _debug("flush scheduled work 2"); + flush_workqueue(rxrpc_workqueue); + _debug("synchronise RCU"); + rcu_barrier(); + _debug("destroy locals"); + rxrpc_destroy_all_locals(); + remove_proc_entry("rxrpc_conns", init_net.proc_net); remove_proc_entry("rxrpc_calls", init_net.proc_net); destroy_workqueue(rxrpc_workqueue); diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index fa50b09..c168268 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -170,25 +170,26 @@ struct rxrpc_security { }; /* - * RxRPC local transport endpoint definition - * - matched by local port, address and protocol type + * RxRPC local transport endpoint description + * - owned by a single AF_RXRPC socket + * - pointed to by transport socket struct sk_user_data */ struct rxrpc_local { + struct rcu_head rcu; + atomic_t usage; + struct list_head link; struct socket *socket; /* my UDP socket */ - struct work_struct destroyer; /* endpoint destroyer */ - struct work_struct acceptor; /* incoming call processor */ - struct work_struct rejecter; /* packet reject writer */ - struct work_struct event_processor; /* endpoint event processor */ + struct work_struct processor; struct list_head services; /* services listening on this endpoint */ - struct list_head link; /* link in endpoint list */ struct rw_semaphore defrag_sem; /* control re-enablement of IP DF bit */ struct sk_buff_head accept_queue; /* incoming calls awaiting acceptance */ struct sk_buff_head reject_queue; /* packets awaiting rejection */ struct sk_buff_head event_queue; /* endpoint event packets awaiting processing */ + struct mutex conn_lock; /* Client connection creation lock */ spinlock_t lock; /* access lock */ rwlock_t services_lock; /* lock for services list */ - atomic_t usage; int debug_id; /* debug ID for printks */ + bool dead; struct sockaddr_rxrpc srx; /* local address */ }; @@ -487,7 +488,7 @@ extern struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *, /* * call_accept.c */ -void rxrpc_accept_incoming_calls(struct work_struct *); +void rxrpc_accept_incoming_calls(struct rxrpc_local *); struct rxrpc_call *rxrpc_accept_call(struct rxrpc_sock *, unsigned long); int rxrpc_reject_call(struct rxrpc_sock *); @@ -527,7 +528,7 @@ void __exit rxrpc_destroy_all_calls(void); */ void rxrpc_process_connection(struct work_struct *); void rxrpc_reject_packet(struct rxrpc_local *, struct sk_buff *); -void rxrpc_reject_packets(struct work_struct *); +void rxrpc_reject_packets(struct rxrpc_local *); /* * conn_object.c @@ -575,17 +576,32 @@ int rxrpc_get_server_data_key(struct rxrpc_connection *, const void *, time_t, /* * local_event.c */ -extern void rxrpc_process_local_events(struct work_struct *); +extern void rxrpc_process_local_events(struct rxrpc_local *); /* * local_object.c */ -extern rwlock_t rxrpc_local_lock; - -struct rxrpc_local *rxrpc_lookup_local(struct sockaddr_rxrpc *); -void rxrpc_put_local(struct rxrpc_local *); +struct rxrpc_local *rxrpc_lookup_local(const struct sockaddr_rxrpc *); +void __rxrpc_put_local(struct rxrpc_local *); void __exit rxrpc_destroy_all_locals(void); +static inline void rxrpc_get_local(struct rxrpc_local *local) +{ + atomic_inc(&local->usage); +} + +static inline +struct rxrpc_local *rxrpc_get_local_maybe(struct rxrpc_local *local) +{ + return atomic_inc_not_zero(&local->usage) ? local : NULL; +} + +static inline void rxrpc_put_local(struct rxrpc_local *local) +{ + if (atomic_dec_and_test(&local->usage)) + __rxrpc_put_local(local); +} + /* * misc.c */ @@ -874,15 +890,6 @@ static inline void rxrpc_purge_queue(struct sk_buff_head *list) rxrpc_free_skb(skb); } -static inline void __rxrpc_get_local(struct rxrpc_local *local, const char *f) -{ - CHECK_SLAB_OKAY(&local->usage); - if (atomic_inc_return(&local->usage) == 1) - printk("resurrected (%s)\n", f); -} - -#define rxrpc_get_local(LOCAL) __rxrpc_get_local((LOCAL), __func__) - #define rxrpc_get_call(CALL) \ do { \ CHECK_SLAB_OKAY(&(CALL)->usage); \ diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index e5723f4..50136c7 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -202,10 +202,8 @@ error_nofree: * accept incoming calls that need peer, transport and/or connection setting up * - the packets we get are all incoming client DATA packets that have seq == 1 */ -void rxrpc_accept_incoming_calls(struct work_struct *work) +void rxrpc_accept_incoming_calls(struct rxrpc_local *local) { - struct rxrpc_local *local = - container_of(work, struct rxrpc_local, acceptor); struct rxrpc_skb_priv *sp; struct sockaddr_rxrpc srx; struct rxrpc_sock *rx; @@ -215,21 +213,8 @@ void rxrpc_accept_incoming_calls(struct work_struct *work) _enter("%d", local->debug_id); - read_lock_bh(&rxrpc_local_lock); - if (atomic_read(&local->usage) > 0) - rxrpc_get_local(local); - else - local = NULL; - read_unlock_bh(&rxrpc_local_lock); - if (!local) { - _leave(" [local dead]"); - return; - } - -process_next_packet: skb = skb_dequeue(&local->accept_queue); if (!skb) { - rxrpc_put_local(local); _leave("\n"); return; } @@ -292,7 +277,7 @@ found_service: case -ECONNRESET: /* old calls are ignored */ case -ECONNABORTED: /* aborted calls are reaborted or ignored */ case 0: - goto process_next_packet; + return; case -ECONNREFUSED: goto invalid_service; case -EBUSY: @@ -308,18 +293,18 @@ backlog_full: busy: rxrpc_busy(local, &srx, &whdr); rxrpc_free_skb(skb); - goto process_next_packet; + return; invalid_service: skb->priority = RX_INVALID_OPERATION; rxrpc_reject_packet(local, skb); - goto process_next_packet; + return; /* can't change connection security type mid-flow */ security_mismatch: skb->priority = RX_PROTOCOL_ERROR; rxrpc_reject_packet(local, skb); - goto process_next_packet; + return; } /* diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index 8bdd692..00c92b6 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -314,19 +314,14 @@ void rxrpc_reject_packet(struct rxrpc_local *local, struct sk_buff *skb) { CHECK_SLAB_OKAY(&local->usage); - if (!atomic_inc_not_zero(&local->usage)) { - printk("resurrected on reject\n"); - BUG(); - } - skb_queue_tail(&local->reject_queue, skb); - rxrpc_queue_work(&local->rejecter); + rxrpc_queue_work(&local->processor); } /* * reject packets through the local endpoint */ -void rxrpc_reject_packets(struct work_struct *work) +void rxrpc_reject_packets(struct rxrpc_local *local) { union { struct sockaddr sa; @@ -334,16 +329,12 @@ void rxrpc_reject_packets(struct work_struct *work) } sa; struct rxrpc_skb_priv *sp; struct rxrpc_wire_header whdr; - struct rxrpc_local *local; struct sk_buff *skb; struct msghdr msg; struct kvec iov[2]; size_t size; __be32 code; - local = container_of(work, struct rxrpc_local, rejecter); - rxrpc_get_local(local); - _enter("%d", local->debug_id); iov[0].iov_base = &whdr; @@ -395,9 +386,7 @@ void rxrpc_reject_packets(struct work_struct *work) } rxrpc_free_skb(skb); - rxrpc_put_local(local); } - rxrpc_put_local(local); _leave(""); } diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 3b405db..47fb167 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -594,9 +594,8 @@ static void rxrpc_post_packet_to_local(struct rxrpc_local *local, { _enter("%p,%p", local, skb); - atomic_inc(&local->usage); skb_queue_tail(&local->event_queue, skb); - rxrpc_queue_work(&local->event_processor); + rxrpc_queue_work(&local->processor); } /* @@ -664,11 +663,15 @@ cant_find_conn: /* * handle data received on the local endpoint * - may be called in interrupt context + * + * The socket is locked by the caller and this prevents the socket from being + * shut down and the local endpoint from going away, thus sk_user_data will not + * be cleared until this function returns. */ void rxrpc_data_ready(struct sock *sk) { struct rxrpc_skb_priv *sp; - struct rxrpc_local *local; + struct rxrpc_local *local = sk->sk_user_data; struct sk_buff *skb; int ret; @@ -676,21 +679,8 @@ void rxrpc_data_ready(struct sock *sk) ASSERT(!irqs_disabled()); - read_lock_bh(&rxrpc_local_lock); - local = sk->sk_user_data; - if (local && atomic_read(&local->usage) > 0) - rxrpc_get_local(local); - else - local = NULL; - read_unlock_bh(&rxrpc_local_lock); - if (!local) { - _leave(" [local dead]"); - return; - } - skb = skb_recv_datagram(sk, 0, 1, &ret); if (!skb) { - rxrpc_put_local(local); if (ret == -EAGAIN) return; _debug("UDP socket error %d", ret); @@ -704,7 +694,6 @@ void rxrpc_data_ready(struct sock *sk) /* we'll probably need to checksum it (didn't call sock_recvmsg) */ if (skb_checksum_complete(skb)) { rxrpc_free_skb(skb); - rxrpc_put_local(local); __UDP_INC_STATS(&init_net, UDP_MIB_INERRORS, 0); _leave(" [CSUM failed]"); return; @@ -769,7 +758,6 @@ void rxrpc_data_ready(struct sock *sk) } out: - rxrpc_put_local(local); return; cant_route_call: @@ -779,8 +767,7 @@ cant_route_call: if (sp->hdr.seq == 1) { _debug("first packet"); skb_queue_tail(&local->accept_queue, skb); - rxrpc_queue_work(&local->acceptor); - rxrpc_put_local(local); + rxrpc_queue_work(&local->processor); _leave(" [incoming]"); return; } @@ -793,13 +780,11 @@ cant_route_call: _debug("reject type %d",sp->hdr.type); rxrpc_reject_packet(local, skb); } - rxrpc_put_local(local); _leave(" [no call]"); return; bad_message: skb->priority = RX_PROTOCOL_ERROR; rxrpc_reject_packet(local, skb); - rxrpc_put_local(local); _leave(" [badmsg]"); } diff --git a/net/rxrpc/local_event.c b/net/rxrpc/local_event.c index 194db2e..31a3f86 100644 --- a/net/rxrpc/local_event.c +++ b/net/rxrpc/local_event.c @@ -82,17 +82,15 @@ static void rxrpc_send_version_request(struct rxrpc_local *local, /* * Process event packets targetted at a local endpoint. */ -void rxrpc_process_local_events(struct work_struct *work) +void rxrpc_process_local_events(struct rxrpc_local *local) { - struct rxrpc_local *local = container_of(work, struct rxrpc_local, event_processor); struct sk_buff *skb; char v; _enter(""); - atomic_inc(&local->usage); - - while ((skb = skb_dequeue(&local->event_queue))) { + skb = skb_dequeue(&local->event_queue); + if (skb) { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); _debug("{%d},{%u}", local->debug_id, sp->hdr.type); @@ -111,10 +109,8 @@ void rxrpc_process_local_events(struct work_struct *work) break; } - rxrpc_put_local(local); rxrpc_free_skb(skb); } - rxrpc_put_local(local); _leave(""); } diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c index c1b8d745..009b321 100644 --- a/net/rxrpc/local_object.c +++ b/net/rxrpc/local_object.c @@ -1,6 +1,6 @@ /* Local endpoint object management * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2016 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -17,40 +17,72 @@ #include #include #include +#include #include #include #include "ar-internal.h" -static LIST_HEAD(rxrpc_locals); -DEFINE_RWLOCK(rxrpc_local_lock); -static DECLARE_RWSEM(rxrpc_local_sem); -static DECLARE_WAIT_QUEUE_HEAD(rxrpc_local_wq); +static void rxrpc_local_processor(struct work_struct *); +static void rxrpc_local_rcu(struct rcu_head *); -static void rxrpc_destroy_local(struct work_struct *work); +static DEFINE_MUTEX(rxrpc_local_mutex); +static LIST_HEAD(rxrpc_local_endpoints); /* - * allocate a new local + * Compare a local to an address. Return -ve, 0 or +ve to indicate less than, + * same or greater than. + * + * We explicitly don't compare the RxRPC service ID as we want to reject + * conflicting uses by differing services. Further, we don't want to share + * addresses with different options (IPv6), so we don't compare those bits + * either. */ -static -struct rxrpc_local *rxrpc_alloc_local(struct sockaddr_rxrpc *srx) +static long rxrpc_local_cmp_key(const struct rxrpc_local *local, + const struct sockaddr_rxrpc *srx) +{ + long diff; + + diff = ((local->srx.transport_type - srx->transport_type) ?: + (local->srx.transport_len - srx->transport_len) ?: + (local->srx.transport.family - srx->transport.family)); + if (diff != 0) + return diff; + + switch (srx->transport.family) { + case AF_INET: + /* If the choice of UDP port is left up to the transport, then + * the endpoint record doesn't match. + */ + return ((u16 __force)local->srx.transport.sin.sin_port - + (u16 __force)srx->transport.sin.sin_port) ?: + memcmp(&local->srx.transport.sin.sin_addr, + &srx->transport.sin.sin_addr, + sizeof(struct in_addr)); + default: + BUG(); + } +} + +/* + * Allocate a new local endpoint. + */ +static struct rxrpc_local *rxrpc_alloc_local(const struct sockaddr_rxrpc *srx) { struct rxrpc_local *local; local = kzalloc(sizeof(struct rxrpc_local), GFP_KERNEL); if (local) { - INIT_WORK(&local->destroyer, &rxrpc_destroy_local); - INIT_WORK(&local->acceptor, &rxrpc_accept_incoming_calls); - INIT_WORK(&local->rejecter, &rxrpc_reject_packets); - INIT_WORK(&local->event_processor, &rxrpc_process_local_events); - INIT_LIST_HEAD(&local->services); + atomic_set(&local->usage, 1); INIT_LIST_HEAD(&local->link); + INIT_WORK(&local->processor, rxrpc_local_processor); + INIT_LIST_HEAD(&local->services); init_rwsem(&local->defrag_sem); skb_queue_head_init(&local->accept_queue); skb_queue_head_init(&local->reject_queue); skb_queue_head_init(&local->event_queue); + mutex_init(&local->conn_lock); spin_lock_init(&local->lock); rwlock_init(&local->services_lock); - atomic_set(&local->usage, 1); local->debug_id = atomic_inc_return(&rxrpc_debug_id); memcpy(&local->srx, srx, sizeof(*srx)); } @@ -61,9 +93,9 @@ struct rxrpc_local *rxrpc_alloc_local(struct sockaddr_rxrpc *srx) /* * create the local socket - * - must be called with rxrpc_local_sem writelocked + * - must be called with rxrpc_local_mutex locked */ -static int rxrpc_create_local(struct rxrpc_local *local) +static int rxrpc_open_socket(struct rxrpc_local *local) { struct sock *sock; int ret, opt; @@ -82,10 +114,10 @@ static int rxrpc_create_local(struct rxrpc_local *local) if (local->srx.transport_len > sizeof(sa_family_t)) { _debug("bind"); ret = kernel_bind(local->socket, - (struct sockaddr *) &local->srx.transport, + (struct sockaddr *)&local->srx.transport, local->srx.transport_len); if (ret < 0) { - _debug("bind failed"); + _debug("bind failed %d", ret); goto error; } } @@ -108,10 +140,6 @@ static int rxrpc_create_local(struct rxrpc_local *local) goto error; } - write_lock_bh(&rxrpc_local_lock); - list_add(&local->link, &rxrpc_locals); - write_unlock_bh(&rxrpc_local_lock); - /* set the socket up */ sock = local->socket->sk; sock->sk_user_data = local; @@ -131,188 +159,227 @@ error: } /* - * create a new local endpoint using the specified UDP address + * Look up or create a new local endpoint using the specified local address. */ -struct rxrpc_local *rxrpc_lookup_local(struct sockaddr_rxrpc *srx) +struct rxrpc_local *rxrpc_lookup_local(const struct sockaddr_rxrpc *srx) { struct rxrpc_local *local; + struct list_head *cursor; + const char *age; + long diff; int ret; - _enter("{%d,%u,%pI4+%hu}", - srx->transport_type, - srx->transport.family, - &srx->transport.sin.sin_addr, - ntohs(srx->transport.sin.sin_port)); - - down_write(&rxrpc_local_sem); + if (srx->transport.family == AF_INET) { + _enter("{%d,%u,%pI4+%hu}", + srx->transport_type, + srx->transport.family, + &srx->transport.sin.sin_addr, + ntohs(srx->transport.sin.sin_port)); + } else { + _enter("{%d,%u}", + srx->transport_type, + srx->transport.family); + return ERR_PTR(-EAFNOSUPPORT); + } - /* see if we have a suitable local local endpoint already */ - read_lock_bh(&rxrpc_local_lock); + mutex_lock(&rxrpc_local_mutex); - list_for_each_entry(local, &rxrpc_locals, link) { - _debug("CMP {%d,%u,%pI4+%hu}", - local->srx.transport_type, - local->srx.transport.family, - &local->srx.transport.sin.sin_addr, - ntohs(local->srx.transport.sin.sin_port)); + for (cursor = rxrpc_local_endpoints.next; + cursor != &rxrpc_local_endpoints; + cursor = cursor->next) { + local = list_entry(cursor, struct rxrpc_local, link); - if (local->srx.transport_type != srx->transport_type || - local->srx.transport.family != srx->transport.family) + diff = rxrpc_local_cmp_key(local, srx); + if (diff < 0) continue; + if (diff > 0) + break; + + /* Services aren't allowed to share transport sockets, so + * reject that here. It is possible that the object is dying - + * but it may also still have the local transport address that + * we want bound. + */ + if (srx->srx_service) { + local = NULL; + goto addr_in_use; + } - switch (srx->transport.family) { - case AF_INET: - if (local->srx.transport.sin.sin_port != - srx->transport.sin.sin_port) - continue; - if (memcmp(&local->srx.transport.sin.sin_addr, - &srx->transport.sin.sin_addr, - sizeof(struct in_addr)) != 0) - continue; - goto found_local; - - default: - BUG(); + /* Found a match. We replace a dying object. Attempting to + * bind the transport socket may still fail if we're attempting + * to use a local address that the dying object is still using. + */ + if (!atomic_inc_not_zero(&local->usage)) { + cursor = cursor->next; + list_del_init(&local->link); + break; } - } - read_unlock_bh(&rxrpc_local_lock); + age = "old"; + goto found; + } - /* we didn't find one, so we need to create one */ local = rxrpc_alloc_local(srx); - if (!local) { - up_write(&rxrpc_local_sem); - return ERR_PTR(-ENOMEM); - } + if (!local) + goto nomem; - ret = rxrpc_create_local(local); - if (ret < 0) { - up_write(&rxrpc_local_sem); - kfree(local); - _leave(" = %d", ret); - return ERR_PTR(ret); - } + ret = rxrpc_open_socket(local); + if (ret < 0) + goto sock_error; + + list_add_tail(&local->link, cursor); + age = "new"; - up_write(&rxrpc_local_sem); +found: + mutex_unlock(&rxrpc_local_mutex); - _net("LOCAL new %d {%d,%u,%pI4+%hu}", + _net("LOCAL %s %d {%d,%u,%pI4+%hu}", + age, local->debug_id, local->srx.transport_type, local->srx.transport.family, &local->srx.transport.sin.sin_addr, ntohs(local->srx.transport.sin.sin_port)); - _leave(" = %p [new]", local); + _leave(" = %p", local); return local; -found_local: - rxrpc_get_local(local); - read_unlock_bh(&rxrpc_local_lock); - up_write(&rxrpc_local_sem); +nomem: + ret = -ENOMEM; +sock_error: + mutex_unlock(&rxrpc_local_mutex); + kfree(local); + _leave(" = %d", ret); + return ERR_PTR(ret); - _net("LOCAL old %d {%d,%u,%pI4+%hu}", - local->debug_id, - local->srx.transport_type, - local->srx.transport.family, - &local->srx.transport.sin.sin_addr, - ntohs(local->srx.transport.sin.sin_port)); +addr_in_use: + mutex_unlock(&rxrpc_local_mutex); + _leave(" = -EADDRINUSE"); + return ERR_PTR(-EADDRINUSE); +} - _leave(" = %p [reuse]", local); - return local; +/* + * A local endpoint reached its end of life. + */ +void __rxrpc_put_local(struct rxrpc_local *local) +{ + _enter("%d", local->debug_id); + rxrpc_queue_work(&local->processor); } /* - * release a local endpoint + * Destroy a local endpoint's socket and then hand the record to RCU to dispose + * of. + * + * Closing the socket cannot be done from bottom half context or RCU callback + * context because it might sleep. */ -void rxrpc_put_local(struct rxrpc_local *local) +static void rxrpc_local_destroyer(struct rxrpc_local *local) { - _enter("%p{u=%d}", local, atomic_read(&local->usage)); + struct socket *socket = local->socket; - ASSERTCMP(atomic_read(&local->usage), >, 0); + _enter("%d", local->debug_id); - /* to prevent a race, the decrement and the dequeue must be effectively - * atomic */ - write_lock_bh(&rxrpc_local_lock); - if (unlikely(atomic_dec_and_test(&local->usage))) { - _debug("destroy local"); - rxrpc_queue_work(&local->destroyer); + /* We can get a race between an incoming call packet queueing the + * processor again and the work processor starting the destruction + * process which will shut down the UDP socket. + */ + if (local->dead) { + _leave(" [already dead]"); + return; } - write_unlock_bh(&rxrpc_local_lock); - _leave(""); + local->dead = true; + + mutex_lock(&rxrpc_local_mutex); + list_del_init(&local->link); + mutex_unlock(&rxrpc_local_mutex); + + ASSERT(list_empty(&local->services)); + + if (socket) { + local->socket = NULL; + kernel_sock_shutdown(socket, SHUT_RDWR); + socket->sk->sk_user_data = NULL; + sock_release(socket); + } + + /* At this point, there should be no more packets coming in to the + * local endpoint. + */ + rxrpc_purge_queue(&local->accept_queue); + rxrpc_purge_queue(&local->reject_queue); + rxrpc_purge_queue(&local->event_queue); + + _debug("rcu local %d", local->debug_id); + call_rcu(&local->rcu, rxrpc_local_rcu); } /* - * destroy a local endpoint + * Process events on an endpoint */ -static void rxrpc_destroy_local(struct work_struct *work) +static void rxrpc_local_processor(struct work_struct *work) { struct rxrpc_local *local = - container_of(work, struct rxrpc_local, destroyer); + container_of(work, struct rxrpc_local, processor); + bool again; - _enter("%p{%d}", local, atomic_read(&local->usage)); + _enter("%d", local->debug_id); - down_write(&rxrpc_local_sem); + do { + again = false; + if (atomic_read(&local->usage) == 0) + return rxrpc_local_destroyer(local); - write_lock_bh(&rxrpc_local_lock); - if (atomic_read(&local->usage) > 0) { - write_unlock_bh(&rxrpc_local_lock); - up_read(&rxrpc_local_sem); - _leave(" [resurrected]"); - return; - } + if (!skb_queue_empty(&local->accept_queue)) { + rxrpc_accept_incoming_calls(local); + again = true; + } - list_del(&local->link); - local->socket->sk->sk_user_data = NULL; - write_unlock_bh(&rxrpc_local_lock); + if (!skb_queue_empty(&local->reject_queue)) { + rxrpc_reject_packets(local); + again = true; + } - downgrade_write(&rxrpc_local_sem); + if (!skb_queue_empty(&local->event_queue)) { + rxrpc_process_local_events(local); + again = true; + } + } while (again); +} - ASSERT(list_empty(&local->services)); - ASSERT(!work_pending(&local->acceptor)); - ASSERT(!work_pending(&local->rejecter)); - ASSERT(!work_pending(&local->event_processor)); +/* + * Destroy a local endpoint after the RCU grace period expires. + */ +static void rxrpc_local_rcu(struct rcu_head *rcu) +{ + struct rxrpc_local *local = container_of(rcu, struct rxrpc_local, rcu); - /* finish cleaning up the local descriptor */ - rxrpc_purge_queue(&local->accept_queue); - rxrpc_purge_queue(&local->reject_queue); - rxrpc_purge_queue(&local->event_queue); - kernel_sock_shutdown(local->socket, SHUT_RDWR); - sock_release(local->socket); + _enter("%d", local->debug_id); - up_read(&rxrpc_local_sem); + ASSERT(!work_pending(&local->processor)); _net("DESTROY LOCAL %d", local->debug_id); kfree(local); - - if (list_empty(&rxrpc_locals)) - wake_up_all(&rxrpc_local_wq); - _leave(""); } /* - * preemptively destroy all local local endpoint rather than waiting for - * them to be destroyed + * Verify the local endpoint list is empty by this point. */ void __exit rxrpc_destroy_all_locals(void) { - DECLARE_WAITQUEUE(myself,current); + struct rxrpc_local *local; _enter(""); - /* we simply have to wait for them to go away */ - if (!list_empty(&rxrpc_locals)) { - set_current_state(TASK_UNINTERRUPTIBLE); - add_wait_queue(&rxrpc_local_wq, &myself); - - while (!list_empty(&rxrpc_locals)) { - schedule(); - set_current_state(TASK_UNINTERRUPTIBLE); - } + if (list_empty(&rxrpc_local_endpoints)) + return; - remove_wait_queue(&rxrpc_local_wq, &myself); - set_current_state(TASK_RUNNING); + mutex_lock(&rxrpc_local_mutex); + list_for_each_entry(local, &rxrpc_local_endpoints, link) { + pr_err("AF_RXRPC: Leaked local %p {%d}\n", + local, atomic_read(&local->usage)); } - - _leave(""); + mutex_unlock(&rxrpc_local_mutex); + BUG(); } -- cgit v0.10.2 From d3cf8fd3fce1b425512f07b307855980379e0783 Mon Sep 17 00:00:00 2001 From: Alexander Shiyan Date: Mon, 13 Jun 2016 18:51:05 +0300 Subject: net: cx89x0: Add DT support Add DT support to the Cirrus Logic CS89x0 driver. Signed-off-by: Alexander Shiyan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cirrus/cs89x0.c b/drivers/net/ethernet/cirrus/cs89x0.c index 6038304..c363b58 100644 --- a/drivers/net/ethernet/cirrus/cs89x0.c +++ b/drivers/net/ethernet/cirrus/cs89x0.c @@ -53,6 +53,8 @@ #include #include #include +#include +#include #include #include #include @@ -1895,9 +1897,17 @@ static int cs89x0_platform_remove(struct platform_device *pdev) return 0; } +static const struct __maybe_unused of_device_id cs89x0_match[] = { + { .compatible = "cirrus,cs8900", }, + { .compatible = "cirrus,cs8920", }, + { }, +}; +MODULE_DEVICE_TABLE(of, cs89x0_match); + static struct platform_driver cs89x0_driver = { .driver = { - .name = DRV_NAME, + .name = DRV_NAME, + .of_match_table = of_match_ptr(cs89x0_match), }, .remove = cs89x0_platform_remove, }; -- cgit v0.10.2 From 818d49ad164cb680983568e3307a30012ab66a23 Mon Sep 17 00:00:00 2001 From: Alexander Shiyan Date: Mon, 13 Jun 2016 18:52:17 +0300 Subject: dt: bindings: Add bindings for Cirrus Logic CS89x0 ethernet chip Add device tree binding documentation details for Cirrus Logic CS8900/CS8920 ethernet chip. Signed-off-by: Alexander Shiyan Acked-by: Rob Herring Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/cirrus,cs89x0.txt b/Documentation/devicetree/bindings/net/cirrus,cs89x0.txt new file mode 100644 index 0000000..c070076 --- /dev/null +++ b/Documentation/devicetree/bindings/net/cirrus,cs89x0.txt @@ -0,0 +1,13 @@ +* Cirrus Logic CS8900/CS8920 Network Controller + +Required properties: +- compatible : Should be "cirrus,cs8900" or "cirrus,cs8920". +- reg : Address and length of the IO space. +- interrupts : Should contain the controller interrupt line. + +Examples: + eth0: eth@10000000 { + compatible = "cirrus,cs8900"; + reg = <0x10000000 0x400>; + interrupts = <10>; + }; -- cgit v0.10.2 From e53743994e21d2458f0129d07b253d66f96f5742 Mon Sep 17 00:00:00 2001 From: Eugene Crosser Date: Mon, 13 Jun 2016 18:46:14 +0200 Subject: af_iucv: use paged SKBs for big outbound messages When an outbound message is bigger than a page, allocate and fill a paged SKB, and subsequently use IUCV send primitive with IPBUFLST flag. This relaxes the pressure to allocate big contiguous kernel buffers. Signed-off-by: Eugene Crosser Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index fc3598a..38448d1 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -1033,6 +1033,7 @@ static int iucv_sock_sendmsg(struct socket *sock, struct msghdr *msg, { struct sock *sk = sock->sk; struct iucv_sock *iucv = iucv_sk(sk); + size_t headroom, linear; struct sk_buff *skb; struct iucv_message txmsg = {0}; struct cmsghdr *cmsg; @@ -1110,20 +1111,31 @@ static int iucv_sock_sendmsg(struct socket *sock, struct msghdr *msg, * this is fine for SOCK_SEQPACKET (unless we want to support * segmented records using the MSG_EOR flag), but * for SOCK_STREAM we might want to improve it in future */ - if (iucv->transport == AF_IUCV_TRANS_HIPER) - skb = sock_alloc_send_skb(sk, - len + sizeof(struct af_iucv_trans_hdr) + ETH_HLEN, - noblock, &err); - else - skb = sock_alloc_send_skb(sk, len, noblock, &err); + headroom = (iucv->transport == AF_IUCV_TRANS_HIPER) + ? sizeof(struct af_iucv_trans_hdr) + ETH_HLEN : 0; + if (headroom + len < PAGE_SIZE) { + linear = len; + } else { + /* In nonlinear "classic" iucv skb, + * reserve space for iucv_array + */ + if (iucv->transport != AF_IUCV_TRANS_HIPER) + headroom += sizeof(struct iucv_array) * + (MAX_SKB_FRAGS + 1); + linear = PAGE_SIZE - headroom; + } + skb = sock_alloc_send_pskb(sk, headroom + linear, len - linear, + noblock, &err, 0); if (!skb) goto out; - if (iucv->transport == AF_IUCV_TRANS_HIPER) - skb_reserve(skb, sizeof(struct af_iucv_trans_hdr) + ETH_HLEN); - if (memcpy_from_msg(skb_put(skb, len), msg, len)) { - err = -EFAULT; + if (headroom) + skb_reserve(skb, headroom); + skb_put(skb, linear); + skb->len = len; + skb->data_len = len - linear; + err = skb_copy_datagram_from_iter(skb, 0, &msg->msg_iter, len); + if (err) goto fail; - } /* wait if outstanding messages for iucv path has reached */ timeo = sock_sndtimeo(sk, noblock); @@ -1148,49 +1160,67 @@ static int iucv_sock_sendmsg(struct socket *sock, struct msghdr *msg, atomic_dec(&iucv->msg_sent); goto fail; } - goto release; - } - skb_queue_tail(&iucv->send_skb_q, skb); - - if (((iucv->path->flags & IUCV_IPRMDATA) & iucv->flags) - && skb->len <= 7) { - err = iucv_send_iprm(iucv->path, &txmsg, skb); + } else { /* Classic VM IUCV transport */ + skb_queue_tail(&iucv->send_skb_q, skb); + + if (((iucv->path->flags & IUCV_IPRMDATA) & iucv->flags) && + skb->len <= 7) { + err = iucv_send_iprm(iucv->path, &txmsg, skb); + + /* on success: there is no message_complete callback */ + /* for an IPRMDATA msg; remove skb from send queue */ + if (err == 0) { + skb_unlink(skb, &iucv->send_skb_q); + kfree_skb(skb); + } - /* on success: there is no message_complete callback - * for an IPRMDATA msg; remove skb from send queue */ - if (err == 0) { - skb_unlink(skb, &iucv->send_skb_q); - kfree_skb(skb); + /* this error should never happen since the */ + /* IUCV_IPRMDATA path flag is set... sever path */ + if (err == 0x15) { + pr_iucv->path_sever(iucv->path, NULL); + skb_unlink(skb, &iucv->send_skb_q); + err = -EPIPE; + goto fail; + } + } else if (skb_is_nonlinear(skb)) { + struct iucv_array *iba = (struct iucv_array *)skb->head; + int i; + + /* skip iucv_array lying in the headroom */ + iba[0].address = (u32)(addr_t)skb->data; + iba[0].length = (u32)skb_headlen(skb); + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + + iba[i + 1].address = + (u32)(addr_t)skb_frag_address(frag); + iba[i + 1].length = (u32)skb_frag_size(frag); + } + err = pr_iucv->message_send(iucv->path, &txmsg, + IUCV_IPBUFLST, 0, + (void *)iba, skb->len); + } else { /* non-IPRM Linear skb */ + err = pr_iucv->message_send(iucv->path, &txmsg, + 0, 0, (void *)skb->data, skb->len); } - - /* this error should never happen since the - * IUCV_IPRMDATA path flag is set... sever path */ - if (err == 0x15) { - pr_iucv->path_sever(iucv->path, NULL); + if (err) { + if (err == 3) { + user_id[8] = 0; + memcpy(user_id, iucv->dst_user_id, 8); + appl_id[8] = 0; + memcpy(appl_id, iucv->dst_name, 8); + pr_err( + "Application %s on z/VM guest %s exceeds message limit\n", + appl_id, user_id); + err = -EAGAIN; + } else { + err = -EPIPE; + } skb_unlink(skb, &iucv->send_skb_q); - err = -EPIPE; goto fail; } - } else - err = pr_iucv->message_send(iucv->path, &txmsg, 0, 0, - (void *) skb->data, skb->len); - if (err) { - if (err == 3) { - user_id[8] = 0; - memcpy(user_id, iucv->dst_user_id, 8); - appl_id[8] = 0; - memcpy(appl_id, iucv->dst_name, 8); - pr_err("Application %s on z/VM guest %s" - " exceeds message limit\n", - appl_id, user_id); - err = -EAGAIN; - } else - err = -EPIPE; - skb_unlink(skb, &iucv->send_skb_q); - goto fail; } -release: release_sock(sk); return len; -- cgit v0.10.2 From 291759a57532b7940b6e52c54ceebd6b8d9e113e Mon Sep 17 00:00:00 2001 From: Eugene Crosser Date: Mon, 13 Jun 2016 18:46:15 +0200 Subject: af_iucv: remove fragment_skb() to use paged SKBs Before introducing paged skbs in the receive path, get rid of the function `iucv_fragment_skb()` that replaces one large linear skb with several smaller linear skbs. Signed-off-by: Eugene Crosser Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 38448d1..9ed2adf 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -1231,44 +1231,6 @@ out: return err; } -/* iucv_fragment_skb() - Fragment a single IUCV message into multiple skb's - * - * Locking: must be called with message_q.lock held - */ -static int iucv_fragment_skb(struct sock *sk, struct sk_buff *skb, int len) -{ - int dataleft, size, copied = 0; - struct sk_buff *nskb; - - dataleft = len; - while (dataleft) { - if (dataleft >= sk->sk_rcvbuf / 4) - size = sk->sk_rcvbuf / 4; - else - size = dataleft; - - nskb = alloc_skb(size, GFP_ATOMIC | GFP_DMA); - if (!nskb) - return -ENOMEM; - - /* copy target class to control buffer of new skb */ - IUCV_SKB_CB(nskb)->class = IUCV_SKB_CB(skb)->class; - - /* copy data fragment */ - memcpy(nskb->data, skb->data + copied, size); - copied += size; - dataleft -= size; - - skb_reset_transport_header(nskb); - skb_reset_network_header(nskb); - nskb->len = size; - - skb_queue_tail(&iucv_sk(sk)->backlog_skb_q, nskb); - } - - return 0; -} - /* iucv_process_message() - Receive a single outstanding IUCV message * * Locking: must be called with message_q.lock held @@ -1300,24 +1262,9 @@ static void iucv_process_message(struct sock *sk, struct sk_buff *skb, kfree_skb(skb); return; } - /* we need to fragment iucv messages for SOCK_STREAM only; - * for SOCK_SEQPACKET, it is only relevant if we support - * record segmentation using MSG_EOR (see also recvmsg()) */ - if (sk->sk_type == SOCK_STREAM && - skb->truesize >= sk->sk_rcvbuf / 4) { - rc = iucv_fragment_skb(sk, skb, len); - kfree_skb(skb); - skb = NULL; - if (rc) { - pr_iucv->path_sever(path, NULL); - return; - } - skb = skb_dequeue(&iucv_sk(sk)->backlog_skb_q); - } else { - skb_reset_transport_header(skb); - skb_reset_network_header(skb); - skb->len = len; - } + skb_reset_transport_header(skb); + skb_reset_network_header(skb); + skb->len = len; } IUCV_SKB_CB(skb)->offset = 0; -- cgit v0.10.2 From a006353a9a8d9e28d35f94bfc97e9573d6ee28aa Mon Sep 17 00:00:00 2001 From: Eugene Crosser Date: Mon, 13 Jun 2016 18:46:16 +0200 Subject: af_iucv: use paged SKBs for big inbound messages When an inbound message is bigger than a page, allocate a paged SKB, and subsequently use IUCV receive primitive with IPBUFLST flag. This relaxes the pressure to allocate big contiguous kernel buffers. Signed-off-by: Eugene Crosser Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 9ed2adf..37d674e 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -1231,6 +1231,34 @@ out: return err; } +static struct sk_buff *alloc_iucv_recv_skb(unsigned long len) +{ + size_t headroom, linear; + struct sk_buff *skb; + int err; + + if (len < PAGE_SIZE) { + headroom = 0; + linear = len; + } else { + headroom = sizeof(struct iucv_array) * (MAX_SKB_FRAGS + 1); + linear = PAGE_SIZE - headroom; + } + skb = alloc_skb_with_frags(headroom + linear, len - linear, + 0, &err, GFP_ATOMIC | GFP_DMA); + WARN_ONCE(!skb, + "alloc of recv iucv skb len=%lu failed with errcode=%d\n", + len, err); + if (skb) { + if (headroom) + skb_reserve(skb, headroom); + skb_put(skb, linear); + skb->len = len; + skb->data_len = len - linear; + } + return skb; +} + /* iucv_process_message() - Receive a single outstanding IUCV message * * Locking: must be called with message_q.lock held @@ -1255,16 +1283,32 @@ static void iucv_process_message(struct sock *sk, struct sk_buff *skb, skb->len = 0; } } else { - rc = pr_iucv->message_receive(path, msg, + if (skb_is_nonlinear(skb)) { + struct iucv_array *iba = (struct iucv_array *)skb->head; + int i; + + iba[0].address = (u32)(addr_t)skb->data; + iba[0].length = (u32)skb_headlen(skb); + for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + + iba[i + 1].address = + (u32)(addr_t)skb_frag_address(frag); + iba[i + 1].length = (u32)skb_frag_size(frag); + } + rc = pr_iucv->message_receive(path, msg, + IUCV_IPBUFLST, + (void *)iba, len, NULL); + } else { + rc = pr_iucv->message_receive(path, msg, msg->flags & IUCV_IPRMDATA, skb->data, len, NULL); + } if (rc) { kfree_skb(skb); return; } - skb_reset_transport_header(skb); - skb_reset_network_header(skb); - skb->len = len; + WARN_ON_ONCE(skb->len != len); } IUCV_SKB_CB(skb)->offset = 0; @@ -1283,7 +1327,7 @@ static void iucv_process_message_q(struct sock *sk) struct sock_msg_q *p, *n; list_for_each_entry_safe(p, n, &iucv->message_q.list, list) { - skb = alloc_skb(iucv_msg_length(&p->msg), GFP_ATOMIC | GFP_DMA); + skb = alloc_iucv_recv_skb(iucv_msg_length(&p->msg)); if (!skb) break; iucv_process_message(sk, skb, p->path, &p->msg); @@ -1778,7 +1822,7 @@ static void iucv_callback_rx(struct iucv_path *path, struct iucv_message *msg) if (len > sk->sk_rcvbuf) goto save_message; - skb = alloc_skb(iucv_msg_length(msg), GFP_ATOMIC | GFP_DMA); + skb = alloc_iucv_recv_skb(iucv_msg_length(msg)); if (!skb) goto save_message; -- cgit v0.10.2 From cd2a9e62c8a3c5cae7691982667d79a0edc65283 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Mon, 13 Jun 2016 13:44:17 -0700 Subject: net: l3mdev: Remove const from flowi6 arg to get_rt6_dst Allow drivers to pass flow arg to functions where the arg is not const and allow the driver to make updates as needed (eg., setting oif). Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index b4d7469..d2ce76c 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -861,7 +861,7 @@ static struct sk_buff *vrf_l3_rcv(struct net_device *vrf_dev, #if IS_ENABLED(CONFIG_IPV6) static struct dst_entry *vrf_get_rt6_dst(const struct net_device *dev, - const struct flowi6 *fl6) + struct flowi6 *fl6) { struct dst_entry *dst = NULL; diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h index 34f33eb..f8a416e 100644 --- a/include/net/l3mdev.h +++ b/include/net/l3mdev.h @@ -38,7 +38,7 @@ struct l3mdev_ops { /* IPv6 ops */ struct dst_entry * (*l3mdev_get_rt6_dst)(const struct net_device *dev, - const struct flowi6 *fl6); + struct flowi6 *fl6); }; #ifdef CONFIG_NET_L3_MASTER_DEV @@ -139,7 +139,7 @@ static inline bool netif_index_is_l3_master(struct net *net, int ifindex) int l3mdev_get_saddr(struct net *net, int ifindex, struct flowi4 *fl4); -struct dst_entry *l3mdev_get_rt6_dst(struct net *net, const struct flowi6 *fl6); +struct dst_entry *l3mdev_get_rt6_dst(struct net *net, struct flowi6 *fl6); static inline struct sk_buff *l3mdev_l3_rcv(struct sk_buff *skb, u16 proto) @@ -225,7 +225,7 @@ static inline int l3mdev_get_saddr(struct net *net, int ifindex, } static inline -struct dst_entry *l3mdev_get_rt6_dst(struct net *net, const struct flowi6 *fl6) +struct dst_entry *l3mdev_get_rt6_dst(struct net *net, struct flowi6 *fl6) { return NULL; } diff --git a/net/l3mdev/l3mdev.c b/net/l3mdev/l3mdev.c index 7da9780..d90e4ef 100644 --- a/net/l3mdev/l3mdev.c +++ b/net/l3mdev/l3mdev.c @@ -108,7 +108,7 @@ EXPORT_SYMBOL_GPL(l3mdev_fib_table_by_index); */ struct dst_entry *l3mdev_get_rt6_dst(struct net *net, - const struct flowi6 *fl6) + struct flowi6 *fl6) { struct dst_entry *dst = NULL; struct net_device *dev; -- cgit v0.10.2 From ba46ee4c0ed122fa14aa2f5d6994c166a01ae2c0 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Mon, 13 Jun 2016 13:44:18 -0700 Subject: net: ipv6: Do not add multicast route for l3 master devices L3 master devices are virtual devices similar to the loopback device. Link local and multicast routes for these devices do not make sense. The ipv6 addrconf code already skips adding a linklocal address; do the same for the mcast route. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 47f837a..b125539 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2254,7 +2254,7 @@ static struct inet6_dev *addrconf_add_dev(struct net_device *dev) return ERR_PTR(-EACCES); /* Add default multicast route */ - if (!(dev->flags & IFF_LOOPBACK)) + if (!(dev->flags & IFF_LOOPBACK) && !netif_is_l3_master(dev)) addrconf_add_mroute(dev); return idev; -- cgit v0.10.2 From 9ff74384600aeecba34ebdacbbde0627489ff601 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Mon, 13 Jun 2016 13:44:19 -0700 Subject: net: vrf: Handle ipv6 multicast and link-local addresses IPv6 multicast and link-local addresses require special handling by the VRF driver: 1. Rather than using the VRF device index and full FIB lookups, packets to/from these addresses should use direct FIB lookups based on the VRF device table. 2. fail sends/receives on a VRF device to/from a multicast address (e.g, make ping6 ff02::1% fail) 3. move the setting of the flow oif to the first dst lookup and revert the change in icmpv6_echo_reply made in ca254490c8dfd ("net: Add VRF support to IPv6 stack"). Linklocal/mcast addresses require use of the skb->dev. With this change connections into and out of a VRF enslaved device work for multicast and link-local addresses work (icmp, tcp, and udp) e.g., 1. packets into VM with VRF config: ping6 -c3 fe80::e0:f9ff:fe1c:b974%br1 ping6 -c3 ff02::1%br1 ssh -6 fe80::e0:f9ff:fe1c:b974%br1 2. packets going out a VRF enslaved device: ping6 -c3 fe80::18f8:83ff:fe4b:7a2e%eth1 ping6 -c3 ff02::1%eth1 ssh -6 root@fe80::18f8:83ff:fe4b:7a2e%eth1 Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index d2ce76c..0b5b3c2 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -785,9 +785,63 @@ out: return rc; } +static struct rt6_info *vrf_ip6_route_lookup(struct net *net, + const struct net_device *dev, + struct flowi6 *fl6, + int ifindex, + int flags) +{ + struct net_vrf *vrf = netdev_priv(dev); + struct fib6_table *table = NULL; + struct rt6_info *rt6; + + rcu_read_lock(); + + /* fib6_table does not have a refcnt and can not be freed */ + rt6 = rcu_dereference(vrf->rt6); + if (likely(rt6)) + table = rt6->rt6i_table; + + rcu_read_unlock(); + + if (!table) + return NULL; + + return ip6_pol_route(net, table, ifindex, fl6, flags); +} + +static void vrf_ip6_input_dst(struct sk_buff *skb, struct net_device *vrf_dev, + int ifindex) +{ + const struct ipv6hdr *iph = ipv6_hdr(skb); + struct flowi6 fl6 = { + .daddr = iph->daddr, + .saddr = iph->saddr, + .flowlabel = ip6_flowinfo(iph), + .flowi6_mark = skb->mark, + .flowi6_proto = iph->nexthdr, + .flowi6_iif = ifindex, + }; + struct net *net = dev_net(vrf_dev); + struct rt6_info *rt6; + + rt6 = vrf_ip6_route_lookup(net, vrf_dev, &fl6, ifindex, + RT6_LOOKUP_F_HAS_SADDR | RT6_LOOKUP_F_IFACE); + if (unlikely(!rt6)) + return; + + if (unlikely(&rt6->dst == &net->ipv6.ip6_null_entry->dst)) + return; + + skb_dst_set(skb, &rt6->dst); +} + static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, struct sk_buff *skb) { + int orig_iif = skb->skb_iif; + bool need_strict; + /* loopback traffic; do not push through packet taps again. * Reset pkt_type for upper layers to process skb */ @@ -798,8 +852,11 @@ static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, goto out; } - /* if packet is NDISC keep the ingress interface */ - if (!ipv6_ndisc_frame(skb)) { + /* if packet is NDISC or addressed to multicast or link-local + * then keep the ingress interface + */ + need_strict = rt6_need_strict(&ipv6_hdr(skb)->daddr); + if (!ipv6_ndisc_frame(skb) && !need_strict) { skb->dev = vrf_dev; skb->skb_iif = vrf_dev->ifindex; @@ -810,6 +867,9 @@ static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, IP6CB(skb)->flags |= IP6SKB_L3SLAVE; } + if (need_strict) + vrf_ip6_input_dst(skb, vrf_dev, orig_iif); + out: return skb; } @@ -863,11 +923,35 @@ static struct sk_buff *vrf_l3_rcv(struct net_device *vrf_dev, static struct dst_entry *vrf_get_rt6_dst(const struct net_device *dev, struct flowi6 *fl6) { + bool need_strict = rt6_need_strict(&fl6->daddr); + struct net_vrf *vrf = netdev_priv(dev); + struct net *net = dev_net(dev); struct dst_entry *dst = NULL; + struct rt6_info *rt; - if (!(fl6->flowi6_flags & FLOWI_FLAG_L3MDEV_SRC)) { - struct net_vrf *vrf = netdev_priv(dev); - struct rt6_info *rt; + /* send to link-local or multicast address */ + if (need_strict) { + int flags = RT6_LOOKUP_F_IFACE; + + /* VRF device does not have a link-local address and + * sending packets to link-local or mcast addresses over + * a VRF device does not make sense + */ + if (fl6->flowi6_oif == dev->ifindex) { + struct dst_entry *dst = &net->ipv6.ip6_null_entry->dst; + + dst_hold(dst); + return dst; + } + + if (!ipv6_addr_any(&fl6->saddr)) + flags |= RT6_LOOKUP_F_HAS_SADDR; + + rt = vrf_ip6_route_lookup(net, dev, fl6, fl6->flowi6_oif, flags); + if (rt) + dst = &rt->dst; + + } else if (!(fl6->flowi6_flags & FLOWI_FLAG_L3MDEV_SRC)) { rcu_read_lock(); @@ -880,6 +964,10 @@ static struct dst_entry *vrf_get_rt6_dst(const struct net_device *dev, rcu_read_unlock(); } + /* make sure oif is set to VRF device for lookup */ + if (!need_strict) + fl6->flowi6_oif = dev->ifindex; + return dst; } #endif diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 54c7794..f55bf3d 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -76,6 +76,8 @@ static inline struct dst_entry *ip6_route_output(struct net *net, struct dst_entry *ip6_route_lookup(struct net *net, struct flowi6 *fl6, int flags); +struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, + int ifindex, struct flowi6 *fl6, int flags); int ip6_route_init(void); void ip6_route_cleanup(void); diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 40454bf..e32a72f 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -587,7 +587,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) fl6.daddr = ipv6_hdr(skb)->saddr; if (saddr) fl6.saddr = *saddr; - fl6.flowi6_oif = l3mdev_fib_oif(skb->dev); + fl6.flowi6_oif = skb->dev->ifindex; fl6.fl6_icmp_type = ICMPV6_ECHO_REPLY; fl6.flowi6_mark = mark; security_skb_classify_flow(skb, flowi6_to_flowi(&fl6)); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index c6ae6f9..d51a1a4 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1042,8 +1042,8 @@ static struct rt6_info *rt6_make_pcpu_route(struct rt6_info *rt) return pcpu_rt; } -static struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, int oif, - struct flowi6 *fl6, int flags) +struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, + int oif, struct flowi6 *fl6, int flags) { struct fib6_node *fn, *saved_fn; struct rt6_info *rt; @@ -1139,6 +1139,7 @@ redo_rt6_select: } } +EXPORT_SYMBOL_GPL(ip6_pol_route); static struct rt6_info *ip6_pol_route_input(struct net *net, struct fib6_table *table, struct flowi6 *fl6, int flags) -- cgit v0.10.2 From b2313077ed0db35ee186905d8076a737248edd24 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Mon, 13 Jun 2016 13:46:28 -0700 Subject: net_sched: make tcf_hash_check() boolean Cc: Jamal Hadi Salim Signed-off-by: Cong Wang Acked-by: Jamal Hadi Salim Signed-off-by: David S. Miller diff --git a/include/net/act_api.h b/include/net/act_api.h index db218a1..fb82b5b 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -155,8 +155,8 @@ int tcf_generic_walker(struct tc_action_net *tn, struct sk_buff *skb, struct tc_action *a); int tcf_hash_search(struct tc_action_net *tn, struct tc_action *a, u32 index); u32 tcf_hash_new_index(struct tc_action_net *tn); -int tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action *a, - int bind); +bool tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action *a, + int bind); int tcf_hash_create(struct tc_action_net *tn, u32 index, struct nlattr *est, struct tc_action *a, int size, int bind, bool cpustats); void tcf_hash_cleanup(struct tc_action *a, struct nlattr *est); diff --git a/net/sched/act_api.c b/net/sched/act_api.c index b6db56e..f8c61d2 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -224,8 +224,8 @@ int tcf_hash_search(struct tc_action_net *tn, struct tc_action *a, u32 index) } EXPORT_SYMBOL(tcf_hash_search); -int tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action *a, - int bind) +bool tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action *a, + int bind) { struct tcf_hashinfo *hinfo = tn->hinfo; struct tcf_common *p = NULL; @@ -235,9 +235,9 @@ int tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action *a, p->tcfc_refcnt++; a->priv = p; a->hinfo = hinfo; - return 1; + return true; } - return 0; + return false; } EXPORT_SYMBOL(tcf_hash_check); diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c index 02f5a8b..b7fa969 100644 --- a/net/sched/act_ife.c +++ b/net/sched/act_ife.c @@ -423,7 +423,8 @@ static int tcf_ife_init(struct net *net, struct nlattr *nla, u16 ife_type = 0; u8 *daddr = NULL; u8 *saddr = NULL; - int ret = 0, exists = 0; + bool exists = false; + int ret = 0; int err; err = nla_parse_nested(tb, TCA_IFE_MAX, nla, ife_policy); diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index 8998a35..6148e32 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -97,7 +97,8 @@ static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla, struct tcf_ipt *ipt; struct xt_entry_target *td, *t; char *tname; - int ret = 0, err, exists = 0; + bool exists = false; + int ret = 0, err; u32 hook = 0; u32 index = 0; diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 787751a..5b135d3 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -62,7 +62,8 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla, struct tc_mirred *parm; struct tcf_mirred *m; struct net_device *dev; - int ret, ok_push = 0, exists = 0; + int ret, ok_push = 0; + bool exists = false; if (nla == NULL) return -EINVAL; diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c index be5fbb5..318328d3 100644 --- a/net/sched/act_simple.c +++ b/net/sched/act_simple.c @@ -86,8 +86,9 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla, struct nlattr *tb[TCA_DEF_MAX + 1]; struct tc_defact *parm; struct tcf_defact *d; + bool exists = false; + int ret = 0, err; char *defdata; - int ret = 0, err, exists = 0; if (nla == NULL) return -EINVAL; diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index 7e2bc3c..53d1486 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -69,7 +69,8 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, struct tcf_skbedit *d; u32 flags = 0, *priority = NULL, *mark = NULL; u16 *queue_mapping = NULL; - int ret = 0, err, exists = 0; + bool exists = false; + int ret = 0, err; if (nla == NULL) return -EINVAL; diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index b075d50..db9b7ed 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -77,8 +77,8 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla, int action; __be16 push_vid = 0; __be16 push_proto = 0; - int ret = 0, exists = 0; - int err; + bool exists = false; + int ret = 0, err; if (!nla) return -EINVAL; -- cgit v0.10.2 From 2e0ab8ca83c122f275b21ea917d52fee506910bf Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 13 Jun 2016 23:54:31 +0300 Subject: ptr_ring: array based FIFO for pointers A simple array based FIFO of pointers. Intended for net stack which commonly has a single consumer/producer. Signed-off-by: Michael S. Tsirkin Acked-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h new file mode 100644 index 0000000..633406f --- /dev/null +++ b/include/linux/ptr_ring.h @@ -0,0 +1,264 @@ +/* + * Definitions for the 'struct ptr_ring' datastructure. + * + * Author: + * Michael S. Tsirkin + * + * Copyright (C) 2016 Red Hat, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * This is a limited-size FIFO maintaining pointers in FIFO order, with + * one CPU producing entries and another consuming entries from a FIFO. + * + * This implementation tries to minimize cache-contention when there is a + * single producer and a single consumer CPU. + */ + +#ifndef _LINUX_PTR_RING_H +#define _LINUX_PTR_RING_H 1 + +#ifdef __KERNEL__ +#include +#include +#include +#include +#include +#include +#include +#endif + +struct ptr_ring { + int producer ____cacheline_aligned_in_smp; + spinlock_t producer_lock; + int consumer ____cacheline_aligned_in_smp; + spinlock_t consumer_lock; + /* Shared consumer/producer data */ + /* Read-only by both the producer and the consumer */ + int size ____cacheline_aligned_in_smp; /* max entries in queue */ + void **queue; +}; + +/* Note: callers invoking this in a loop must use a compiler barrier, + * for example cpu_relax(). + * Callers don't need to take producer lock - if they don't + * the next call to __ptr_ring_produce may fail. + */ +static inline bool __ptr_ring_full(struct ptr_ring *r) +{ + return r->queue[r->producer]; +} + +static inline bool ptr_ring_full(struct ptr_ring *r) +{ + barrier(); + return __ptr_ring_full(r); +} + +/* Note: callers invoking this in a loop must use a compiler barrier, + * for example cpu_relax(). + */ +static inline int __ptr_ring_produce(struct ptr_ring *r, void *ptr) +{ + if (__ptr_ring_full(r)) + return -ENOSPC; + + r->queue[r->producer++] = ptr; + if (unlikely(r->producer >= r->size)) + r->producer = 0; + return 0; +} + +static inline int ptr_ring_produce(struct ptr_ring *r, void *ptr) +{ + int ret; + + spin_lock(&r->producer_lock); + ret = __ptr_ring_produce(r, ptr); + spin_unlock(&r->producer_lock); + + return ret; +} + +static inline int ptr_ring_produce_irq(struct ptr_ring *r, void *ptr) +{ + int ret; + + spin_lock_irq(&r->producer_lock); + ret = __ptr_ring_produce(r, ptr); + spin_unlock_irq(&r->producer_lock); + + return ret; +} + +static inline int ptr_ring_produce_any(struct ptr_ring *r, void *ptr) +{ + unsigned long flags; + int ret; + + spin_lock_irqsave(&r->producer_lock, flags); + ret = __ptr_ring_produce(r, ptr); + spin_unlock_irqrestore(&r->producer_lock, flags); + + return ret; +} + +static inline int ptr_ring_produce_bh(struct ptr_ring *r, void *ptr) +{ + int ret; + + spin_lock_bh(&r->producer_lock); + ret = __ptr_ring_produce(r, ptr); + spin_unlock_bh(&r->producer_lock); + + return ret; +} + +/* Note: callers invoking this in a loop must use a compiler barrier, + * for example cpu_relax(). Callers must take consumer_lock + * if they dereference the pointer - see e.g. PTR_RING_PEEK_CALL. + * There's no need for a lock if pointer is merely tested - see e.g. + * ptr_ring_empty. + */ +static inline void *__ptr_ring_peek(struct ptr_ring *r) +{ + return r->queue[r->consumer]; +} + +static inline bool ptr_ring_empty(struct ptr_ring *r) +{ + barrier(); + return !__ptr_ring_peek(r); +} + +/* Must only be called after __ptr_ring_peek returned !NULL */ +static inline void __ptr_ring_discard_one(struct ptr_ring *r) +{ + r->queue[r->consumer++] = NULL; + if (unlikely(r->consumer >= r->size)) + r->consumer = 0; +} + +static inline void *__ptr_ring_consume(struct ptr_ring *r) +{ + void *ptr; + + ptr = __ptr_ring_peek(r); + if (ptr) + __ptr_ring_discard_one(r); + + return ptr; +} + +static inline void *ptr_ring_consume(struct ptr_ring *r) +{ + void *ptr; + + spin_lock(&r->consumer_lock); + ptr = __ptr_ring_consume(r); + spin_unlock(&r->consumer_lock); + + return ptr; +} + +static inline void *ptr_ring_consume_irq(struct ptr_ring *r) +{ + void *ptr; + + spin_lock_irq(&r->consumer_lock); + ptr = __ptr_ring_consume(r); + spin_unlock_irq(&r->consumer_lock); + + return ptr; +} + +static inline void *ptr_ring_consume_any(struct ptr_ring *r) +{ + unsigned long flags; + void *ptr; + + spin_lock_irqsave(&r->consumer_lock, flags); + ptr = __ptr_ring_consume(r); + spin_unlock_irqrestore(&r->consumer_lock, flags); + + return ptr; +} + +static inline void *ptr_ring_consume_bh(struct ptr_ring *r) +{ + void *ptr; + + spin_lock_bh(&r->consumer_lock); + ptr = __ptr_ring_consume(r); + spin_unlock_bh(&r->consumer_lock); + + return ptr; +} + +/* Cast to structure type and call a function without discarding from FIFO. + * Function must return a value. + * Callers must take consumer_lock. + */ +#define __PTR_RING_PEEK_CALL(r, f) ((f)(__ptr_ring_peek(r))) + +#define PTR_RING_PEEK_CALL(r, f) ({ \ + typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \ + \ + spin_lock(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \ + spin_unlock(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v; \ +}) + +#define PTR_RING_PEEK_CALL_IRQ(r, f) ({ \ + typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \ + \ + spin_lock_irq(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \ + spin_unlock_irq(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v; \ +}) + +#define PTR_RING_PEEK_CALL_BH(r, f) ({ \ + typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \ + \ + spin_lock_bh(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \ + spin_unlock_bh(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v; \ +}) + +#define PTR_RING_PEEK_CALL_ANY(r, f) ({ \ + typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \ + unsigned long __PTR_RING_PEEK_CALL_f;\ + \ + spin_lock_irqsave(&(r)->consumer_lock, __PTR_RING_PEEK_CALL_f); \ + __PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \ + spin_unlock_irqrestore(&(r)->consumer_lock, __PTR_RING_PEEK_CALL_f); \ + __PTR_RING_PEEK_CALL_v; \ +}) + +static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) +{ + r->queue = kzalloc(ALIGN(size * sizeof *(r->queue), SMP_CACHE_BYTES), + gfp); + if (!r->queue) + return -ENOMEM; + + r->size = size; + r->producer = r->consumer = 0; + spin_lock_init(&r->producer_lock); + spin_lock_init(&r->consumer_lock); + + return 0; +} + +static inline void ptr_ring_cleanup(struct ptr_ring *r) +{ + kfree(r->queue); +} + +#endif /* _LINUX_PTR_RING_H */ -- cgit v0.10.2 From 9fb6bc5b4a78b38049aee62686c00b0b6c5ac946 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 13 Jun 2016 23:54:36 +0300 Subject: ptr_ring: ring test Add ringtest based unit test for ptr ring. Signed-off-by: Michael S. Tsirkin Signed-off-by: David S. Miller diff --git a/tools/virtio/ringtest/Makefile b/tools/virtio/ringtest/Makefile index 6ba7455..50e086c 100644 --- a/tools/virtio/ringtest/Makefile +++ b/tools/virtio/ringtest/Makefile @@ -1,6 +1,6 @@ all: -all: ring virtio_ring_0_9 virtio_ring_poll virtio_ring_inorder +all: ring virtio_ring_0_9 virtio_ring_poll virtio_ring_inorder ptr_ring CFLAGS += -Wall CFLAGS += -pthread -O2 -ggdb @@ -8,6 +8,7 @@ LDFLAGS += -pthread -O2 -ggdb main.o: main.c main.h ring.o: ring.c main.h +ptr_ring.o: ptr_ring.c main.h ../../../include/linux/ptr_ring.h virtio_ring_0_9.o: virtio_ring_0_9.c main.h virtio_ring_poll.o: virtio_ring_poll.c virtio_ring_0_9.c main.h virtio_ring_inorder.o: virtio_ring_inorder.c virtio_ring_0_9.c main.h @@ -15,11 +16,13 @@ ring: ring.o main.o virtio_ring_0_9: virtio_ring_0_9.o main.o virtio_ring_poll: virtio_ring_poll.o main.o virtio_ring_inorder: virtio_ring_inorder.o main.o +ptr_ring: ptr_ring.o main.o clean: -rm main.o -rm ring.o ring -rm virtio_ring_0_9.o virtio_ring_0_9 -rm virtio_ring_poll.o virtio_ring_poll -rm virtio_ring_inorder.o virtio_ring_inorder + -rm ptr_ring.o ptr_ring .PHONY: all clean diff --git a/tools/virtio/ringtest/ptr_ring.c b/tools/virtio/ringtest/ptr_ring.c new file mode 100644 index 0000000..74abd74 --- /dev/null +++ b/tools/virtio/ringtest/ptr_ring.c @@ -0,0 +1,192 @@ +#define _GNU_SOURCE +#include "main.h" +#include +#include +#include +#include +#include +#include +#include +#include + +#define SMP_CACHE_BYTES 64 +#define cache_line_size() SMP_CACHE_BYTES +#define ____cacheline_aligned_in_smp __attribute__ ((aligned (SMP_CACHE_BYTES))) +#define unlikely(x) (__builtin_expect(!!(x), 0)) +#define ALIGN(x, a) (((x) + (a) - 1) / (a) * (a)) +typedef pthread_spinlock_t spinlock_t; + +typedef int gfp_t; +static void *kzalloc(unsigned size, gfp_t gfp) +{ + void *p = memalign(64, size); + if (!p) + return p; + memset(p, 0, size); + + return p; +} + +static void kfree(void *p) +{ + if (p) + free(p); +} + +static void spin_lock_init(spinlock_t *lock) +{ + int r = pthread_spin_init(lock, 0); + assert(!r); +} + +static void spin_lock(spinlock_t *lock) +{ + int ret = pthread_spin_lock(lock); + assert(!ret); +} + +static void spin_unlock(spinlock_t *lock) +{ + int ret = pthread_spin_unlock(lock); + assert(!ret); +} + +static void spin_lock_bh(spinlock_t *lock) +{ + spin_lock(lock); +} + +static void spin_unlock_bh(spinlock_t *lock) +{ + spin_unlock(lock); +} + +static void spin_lock_irq(spinlock_t *lock) +{ + spin_lock(lock); +} + +static void spin_unlock_irq(spinlock_t *lock) +{ + spin_unlock(lock); +} + +static void spin_lock_irqsave(spinlock_t *lock, unsigned long f) +{ + spin_lock(lock); +} + +static void spin_unlock_irqrestore(spinlock_t *lock, unsigned long f) +{ + spin_unlock(lock); +} + +#include "../../../include/linux/ptr_ring.h" + +static unsigned long long headcnt, tailcnt; +static struct ptr_ring array ____cacheline_aligned_in_smp; + +/* implemented by ring */ +void alloc_ring(void) +{ + int ret = ptr_ring_init(&array, ring_size, 0); + assert(!ret); +} + +/* guest side */ +int add_inbuf(unsigned len, void *buf, void *datap) +{ + int ret; + + ret = __ptr_ring_produce(&array, buf); + if (ret >= 0) { + ret = 0; + headcnt++; + } + + return ret; +} + +/* + * ptr_ring API provides no way for producer to find out whether a given + * buffer was consumed. Our tests merely require that a successful get_buf + * implies that add_inbuf succeed in the past, and that add_inbuf will succeed, + * fake it accordingly. + */ +void *get_buf(unsigned *lenp, void **bufp) +{ + void *datap; + + if (tailcnt == headcnt || __ptr_ring_full(&array)) + datap = NULL; + else { + datap = "Buffer\n"; + ++tailcnt; + } + + return datap; +} + +void poll_used(void) +{ + void *b; + + do { + if (tailcnt == headcnt || __ptr_ring_full(&array)) { + b = NULL; + barrier(); + } else { + b = "Buffer\n"; + } + } while (!b); +} + +void disable_call() +{ + assert(0); +} + +bool enable_call() +{ + assert(0); +} + +void kick_available(void) +{ + assert(0); +} + +/* host side */ +void disable_kick() +{ + assert(0); +} + +bool enable_kick() +{ + assert(0); +} + +void poll_avail(void) +{ + void *b; + + do { + barrier(); + b = __ptr_ring_peek(&array); + } while (!b); +} + +bool use_buf(unsigned *lenp, void **bufp) +{ + void *ptr; + + ptr = __ptr_ring_consume(&array); + + return ptr; +} + +void call_used(void) +{ + assert(0); +} -- cgit v0.10.2 From ad69f35d1dc0a79f86627ca56e01f86512602a49 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 13 Jun 2016 23:54:41 +0300 Subject: skb_array: array based FIFO for skbs A simple array based FIFO of pointers. Intended for net stack so uses skbs for type safety. Implemented as a set of wrappers around ptr_ring. Signed-off-by: Michael S. Tsirkin Acked-by: Jesper Dangaard Brouer Tested-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h new file mode 100644 index 0000000..c4c0902 --- /dev/null +++ b/include/linux/skb_array.h @@ -0,0 +1,144 @@ +/* + * Definitions for the 'struct skb_array' datastructure. + * + * Author: + * Michael S. Tsirkin + * + * Copyright (C) 2016 Red Hat, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * Limited-size FIFO of skbs. Can be used more or less whenever + * sk_buff_head can be used, except you need to know the queue size in + * advance. + * Implemented as a type-safe wrapper around ptr_ring. + */ + +#ifndef _LINUX_SKB_ARRAY_H +#define _LINUX_SKB_ARRAY_H 1 + +#ifdef __KERNEL__ +#include +#include +#include +#endif + +struct skb_array { + struct ptr_ring ring; +}; + +/* Might be slightly faster than skb_array_full below, but callers invoking + * this in a loop must use a compiler barrier, for example cpu_relax(). + */ +static inline bool __skb_array_full(struct skb_array *a) +{ + return __ptr_ring_full(&a->ring); +} + +static inline bool skb_array_full(struct skb_array *a) +{ + return ptr_ring_full(&a->ring); +} + +static inline int skb_array_produce(struct skb_array *a, struct sk_buff *skb) +{ + return ptr_ring_produce(&a->ring, skb); +} + +static inline int skb_array_produce_irq(struct skb_array *a, struct sk_buff *skb) +{ + return ptr_ring_produce_irq(&a->ring, skb); +} + +static inline int skb_array_produce_bh(struct skb_array *a, struct sk_buff *skb) +{ + return ptr_ring_produce_bh(&a->ring, skb); +} + +static inline int skb_array_produce_any(struct skb_array *a, struct sk_buff *skb) +{ + return ptr_ring_produce_any(&a->ring, skb); +} + +/* Might be slightly faster than skb_array_empty below, but callers invoking + * this in a loop must take care to use a compiler barrier, for example + * cpu_relax(). + */ +static inline bool __skb_array_empty(struct skb_array *a) +{ + return !__ptr_ring_peek(&a->ring); +} + +static inline bool skb_array_empty(struct skb_array *a) +{ + return ptr_ring_empty(&a->ring); +} + +static inline struct sk_buff *skb_array_consume(struct skb_array *a) +{ + return ptr_ring_consume(&a->ring); +} + +static inline struct sk_buff *skb_array_consume_irq(struct skb_array *a) +{ + return ptr_ring_consume_irq(&a->ring); +} + +static inline struct sk_buff *skb_array_consume_any(struct skb_array *a) +{ + return ptr_ring_consume_any(&a->ring); +} + +static inline struct sk_buff *skb_array_consume_bh(struct skb_array *a) +{ + return ptr_ring_consume_bh(&a->ring); +} + +static inline int __skb_array_len_with_tag(struct sk_buff *skb) +{ + if (likely(skb)) { + int len = skb->len; + + if (skb_vlan_tag_present(skb)) + len += VLAN_HLEN; + + return len; + } else { + return 0; + } +} + +static inline int skb_array_peek_len(struct skb_array *a) +{ + return PTR_RING_PEEK_CALL(&a->ring, __skb_array_len_with_tag); +} + +static inline int skb_array_peek_len_irq(struct skb_array *a) +{ + return PTR_RING_PEEK_CALL_IRQ(&a->ring, __skb_array_len_with_tag); +} + +static inline int skb_array_peek_len_bh(struct skb_array *a) +{ + return PTR_RING_PEEK_CALL_BH(&a->ring, __skb_array_len_with_tag); +} + +static inline int skb_array_peek_len_any(struct skb_array *a) +{ + return PTR_RING_PEEK_CALL_ANY(&a->ring, __skb_array_len_with_tag); +} + +static inline int skb_array_init(struct skb_array *a, int size, gfp_t gfp) +{ + return ptr_ring_init(&a->ring, size, gfp); +} + +static inline void skb_array_cleanup(struct skb_array *a) +{ + ptr_ring_cleanup(&a->ring); +} + +#endif /* _LINUX_SKB_ARRAY_H */ -- cgit v0.10.2 From 5d49de532002f02755decd1758aac53063a68625 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 13 Jun 2016 23:54:45 +0300 Subject: ptr_ring: resize support This adds ring resize support. Seems to be necessary as users such as tun allow userspace control over queue size. If resize is used, this costs us ability to peek at queue without consumer lock - should not be a big deal as peek and consumer are usually run on the same CPU. If ring is made bigger, ring contents is preserved. If ring is made smaller, extra pointers are passed to an optional destructor callback. Cleanup function also gains destructor callback such that all pointers in queue can be cleaned up. This changes some APIs but we don't have any users yet, so it won't break bisect. Signed-off-by: Michael S. Tsirkin Acked-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h index 633406f..562a65e 100644 --- a/include/linux/ptr_ring.h +++ b/include/linux/ptr_ring.h @@ -43,9 +43,9 @@ struct ptr_ring { }; /* Note: callers invoking this in a loop must use a compiler barrier, - * for example cpu_relax(). - * Callers don't need to take producer lock - if they don't - * the next call to __ptr_ring_produce may fail. + * for example cpu_relax(). If ring is ever resized, callers must hold + * producer_lock - see e.g. ptr_ring_full. Otherwise, if callers don't hold + * producer_lock, the next call to __ptr_ring_produce may fail. */ static inline bool __ptr_ring_full(struct ptr_ring *r) { @@ -54,16 +54,55 @@ static inline bool __ptr_ring_full(struct ptr_ring *r) static inline bool ptr_ring_full(struct ptr_ring *r) { - barrier(); - return __ptr_ring_full(r); + bool ret; + + spin_lock(&r->producer_lock); + ret = __ptr_ring_full(r); + spin_unlock(&r->producer_lock); + + return ret; +} + +static inline bool ptr_ring_full_irq(struct ptr_ring *r) +{ + bool ret; + + spin_lock_irq(&r->producer_lock); + ret = __ptr_ring_full(r); + spin_unlock_irq(&r->producer_lock); + + return ret; +} + +static inline bool ptr_ring_full_any(struct ptr_ring *r) +{ + unsigned long flags; + bool ret; + + spin_lock_irqsave(&r->producer_lock, flags); + ret = __ptr_ring_full(r); + spin_unlock_irqrestore(&r->producer_lock, flags); + + return ret; +} + +static inline bool ptr_ring_full_bh(struct ptr_ring *r) +{ + bool ret; + + spin_lock_bh(&r->producer_lock); + ret = __ptr_ring_full(r); + spin_unlock_bh(&r->producer_lock); + + return ret; } /* Note: callers invoking this in a loop must use a compiler barrier, - * for example cpu_relax(). + * for example cpu_relax(). Callers must hold producer_lock. */ static inline int __ptr_ring_produce(struct ptr_ring *r, void *ptr) { - if (__ptr_ring_full(r)) + if (r->queue[r->producer]) return -ENOSPC; r->queue[r->producer++] = ptr; @@ -120,20 +159,68 @@ static inline int ptr_ring_produce_bh(struct ptr_ring *r, void *ptr) /* Note: callers invoking this in a loop must use a compiler barrier, * for example cpu_relax(). Callers must take consumer_lock * if they dereference the pointer - see e.g. PTR_RING_PEEK_CALL. - * There's no need for a lock if pointer is merely tested - see e.g. - * ptr_ring_empty. + * If ring is never resized, and if the pointer is merely + * tested, there's no need to take the lock - see e.g. __ptr_ring_empty. */ static inline void *__ptr_ring_peek(struct ptr_ring *r) { return r->queue[r->consumer]; } -static inline bool ptr_ring_empty(struct ptr_ring *r) +/* Note: callers invoking this in a loop must use a compiler barrier, + * for example cpu_relax(). Callers must take consumer_lock + * if the ring is ever resized - see e.g. ptr_ring_empty. + */ +static inline bool __ptr_ring_empty(struct ptr_ring *r) { - barrier(); return !__ptr_ring_peek(r); } +static inline bool ptr_ring_empty(struct ptr_ring *r) +{ + bool ret; + + spin_lock(&r->consumer_lock); + ret = __ptr_ring_empty(r); + spin_unlock(&r->consumer_lock); + + return ret; +} + +static inline bool ptr_ring_empty_irq(struct ptr_ring *r) +{ + bool ret; + + spin_lock_irq(&r->consumer_lock); + ret = __ptr_ring_empty(r); + spin_unlock_irq(&r->consumer_lock); + + return ret; +} + +static inline bool ptr_ring_empty_any(struct ptr_ring *r) +{ + unsigned long flags; + bool ret; + + spin_lock_irqsave(&r->consumer_lock, flags); + ret = __ptr_ring_empty(r); + spin_unlock_irqrestore(&r->consumer_lock, flags); + + return ret; +} + +static inline bool ptr_ring_empty_bh(struct ptr_ring *r) +{ + bool ret; + + spin_lock_bh(&r->consumer_lock); + ret = __ptr_ring_empty(r); + spin_unlock_bh(&r->consumer_lock); + + return ret; +} + /* Must only be called after __ptr_ring_peek returned !NULL */ static inline void __ptr_ring_discard_one(struct ptr_ring *r) { @@ -241,10 +328,14 @@ static inline void *ptr_ring_consume_bh(struct ptr_ring *r) __PTR_RING_PEEK_CALL_v; \ }) +static inline void **__ptr_ring_init_queue_alloc(int size, gfp_t gfp) +{ + return kzalloc(ALIGN(size * sizeof(void *), SMP_CACHE_BYTES), gfp); +} + static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) { - r->queue = kzalloc(ALIGN(size * sizeof *(r->queue), SMP_CACHE_BYTES), - gfp); + r->queue = __ptr_ring_init_queue_alloc(size, gfp); if (!r->queue) return -ENOMEM; @@ -256,8 +347,46 @@ static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) return 0; } -static inline void ptr_ring_cleanup(struct ptr_ring *r) +static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, + void (*destroy)(void *)) +{ + unsigned long flags; + int producer = 0; + void **queue = __ptr_ring_init_queue_alloc(size, gfp); + void **old; + void *ptr; + + if (!queue) + return -ENOMEM; + + spin_lock_irqsave(&(r)->producer_lock, flags); + + while ((ptr = ptr_ring_consume(r))) + if (producer < size) + queue[producer++] = ptr; + else if (destroy) + destroy(ptr); + + r->size = size; + r->producer = producer; + r->consumer = 0; + old = r->queue; + r->queue = queue; + + spin_unlock_irqrestore(&(r)->producer_lock, flags); + + kfree(old); + + return 0; +} + +static inline void ptr_ring_cleanup(struct ptr_ring *r, void (*destroy)(void *)) { + void *ptr; + + if (destroy) + while ((ptr = ptr_ring_consume(r))) + destroy(ptr); kfree(r->queue); } -- cgit v0.10.2 From 7d7072e3bad5569f636d3c54a36da40976bfd505 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 13 Jun 2016 23:54:50 +0300 Subject: skb_array: resize support Update skb_array after ptr_ring API changes. Signed-off-by: Michael S. Tsirkin Acked-by: Jesper Dangaard Brouer Tested-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index c4c0902..678bfbf 100644 --- a/include/linux/skb_array.h +++ b/include/linux/skb_array.h @@ -63,9 +63,9 @@ static inline int skb_array_produce_any(struct skb_array *a, struct sk_buff *skb return ptr_ring_produce_any(&a->ring, skb); } -/* Might be slightly faster than skb_array_empty below, but callers invoking - * this in a loop must take care to use a compiler barrier, for example - * cpu_relax(). +/* Might be slightly faster than skb_array_empty below, but only safe if the + * array is never resized. Also, callers invoking this in a loop must take care + * to use a compiler barrier, for example cpu_relax(). */ static inline bool __skb_array_empty(struct skb_array *a) { @@ -77,6 +77,21 @@ static inline bool skb_array_empty(struct skb_array *a) return ptr_ring_empty(&a->ring); } +static inline bool skb_array_empty_bh(struct skb_array *a) +{ + return ptr_ring_empty_bh(&a->ring); +} + +static inline bool skb_array_empty_irq(struct skb_array *a) +{ + return ptr_ring_empty_irq(&a->ring); +} + +static inline bool skb_array_empty_any(struct skb_array *a) +{ + return ptr_ring_empty_any(&a->ring); +} + static inline struct sk_buff *skb_array_consume(struct skb_array *a) { return ptr_ring_consume(&a->ring); @@ -136,9 +151,19 @@ static inline int skb_array_init(struct skb_array *a, int size, gfp_t gfp) return ptr_ring_init(&a->ring, size, gfp); } +void __skb_array_destroy_skb(void *ptr) +{ + kfree_skb(ptr); +} + +int skb_array_resize(struct skb_array *a, int size, gfp_t gfp) +{ + return ptr_ring_resize(&a->ring, size, gfp, __skb_array_destroy_skb); +} + static inline void skb_array_cleanup(struct skb_array *a) { - ptr_ring_cleanup(&a->ring); + ptr_ring_cleanup(&a->ring, __skb_array_destroy_skb); } #endif /* _LINUX_SKB_ARRAY_H */ -- cgit v0.10.2 From df10db98ab546e899b79d1b2b8a00a86fe8bb29d Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 14 Jun 2016 00:00:04 +0200 Subject: tun: fix csum generation for tap devices The commit 34166093639b ("tuntap: use common code for virtio_net_hdr and skb GSO conversion") replaced the tun code for header manipulation with the generic helpers. While doing so, it implictly moved the skb_partial_csum_set() invocation after eth_type_trans(), which invalidate the current gso start/offset values. Fix it by moving the helper invocation before the mac pulling. Fixes: 34166093639 ("tuntap: use common code for virtio_net_hdr and skb GSO conversion") Signed-off-by: Paolo Abeni Acked-by: Mike Rapoport Signed-off-by: David S. Miller diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 8cc6bf4..4884802 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1254,6 +1254,13 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, return -EFAULT; } + err = virtio_net_hdr_to_skb(skb, &gso, tun_is_little_endian(tun)); + if (err) { + this_cpu_inc(tun->pcpu_stats->rx_frame_errors); + kfree_skb(skb); + return -EINVAL; + } + switch (tun->flags & TUN_TYPE_MASK) { case IFF_TUN: if (tun->flags & IFF_NO_PI) { @@ -1280,13 +1287,6 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, break; } - err = virtio_net_hdr_to_skb(skb, &gso, tun_is_little_endian(tun)); - if (err) { - this_cpu_inc(tun->pcpu_stats->rx_frame_errors); - kfree_skb(skb); - return -EINVAL; - } - /* copy skb_ubuf_info for callback when skb has no error */ if (zerocopy) { skb_shinfo(skb)->destructor_arg = msg_control; -- cgit v0.10.2 From 7889681f4a6c2148e1245604bac751a1cae8f882 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Mon, 13 Jun 2016 17:14:12 -0700 Subject: net: vrf: Update flags and features settings 1. Default VRF devices to not having a qdisc (IFF_NO_QUEUE). Users can add one as desired. 2. Disable adding a VLAN to a VRF device. 3. Enable offloads and hardware features similar to other logical devices (e.g., dummy, veth) Change provides a significant boost in TCP stream Tx performance, from ~2,700 Mbps to ~18,100 Mbps and makes throughput close to the performance without a VRF (18,500 Mbps). netperf TCP_STREAM benchmark using qemu with virtio+vhost for the NICs Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 0b5b3c2..e3fc6d3 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -1099,6 +1099,20 @@ static void vrf_setup(struct net_device *dev) /* don't allow vrf devices to change network namespaces. */ dev->features |= NETIF_F_NETNS_LOCAL; + + /* does not make sense for a VLAN to be added to a vrf device */ + dev->features |= NETIF_F_VLAN_CHALLENGED; + + /* enable offload features */ + dev->features |= NETIF_F_GSO_SOFTWARE; + dev->features |= NETIF_F_RXCSUM | NETIF_F_HW_CSUM; + dev->features |= NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA; + + dev->hw_features = dev->features; + dev->hw_enc_features = dev->features; + + /* default to no qdisc; user can add if desired */ + dev->priv_flags |= IFF_NO_QUEUE; } static int vrf_validate(struct nlattr *tb[], struct nlattr *data[]) -- cgit v0.10.2 From 35c55c9877f8de0ab129fa1a309271d0ecc868b9 Mon Sep 17 00:00:00 2001 From: Jon Paul Maloy Date: Mon, 13 Jun 2016 20:46:22 -0400 Subject: tipc: add neighbor monitoring framework TIPC based clusters are by default set up with full-mesh link connectivity between all nodes. Those links are expected to provide a short failure detection time, by default set to 1500 ms. Because of this, the background load for neighbor monitoring in an N-node cluster increases with a factor N on each node, while the overall monitoring traffic through the network infrastructure increases at a ~(N * (N - 1)) rate. Experience has shown that such clusters don't scale well beyond ~100 nodes unless we significantly increase failure discovery tolerance. This commit introduces a framework and an algorithm that drastically reduces this background load, while basically maintaining the original failure detection times across the whole cluster. Using this algorithm, background load will now grow at a rate of ~(2 * sqrt(N)) per node, and at ~(2 * N * sqrt(N)) in traffic overhead. As an example, each node will now have to actively monitor 38 neighbors in a 400-node cluster, instead of as before 399. This "Overlapping Ring Supervision Algorithm" is completely distributed and employs no centralized or coordinated state. It goes as follows: - Each node makes up a linearly ascending, circular list of all its N known neighbors, based on their TIPC node identity. This algorithm must be the same on all nodes. - The node then selects the next M = sqrt(N) - 1 nodes downstream from itself in the list, and chooses to actively monitor those. This is called its "local monitoring domain". - It creates a domain record describing the monitoring domain, and piggy-backs this in the data area of all neighbor monitoring messages (LINK_PROTOCOL/STATE) leaving that node. This means that all nodes in the cluster eventually (default within 400 ms) will learn about its monitoring domain. - Whenever a node discovers a change in its local domain, e.g., a node has been added or has gone down, it creates and sends out a new version of its node record to inform all neighbors about the change. - A node receiving a domain record from anybody outside its local domain matches this against its own list (which may not look the same), and chooses to not actively monitor those members of the received domain record that are also present in its own list. Instead, it relies on indications from the direct monitoring nodes if an indirectly monitored node has gone up or down. If a node is indicated lost, the receiving node temporarily activates its own direct monitoring towards that node in order to confirm, or not, that it is actually gone. - Since each node is actively monitoring sqrt(N) downstream neighbors, each node is also actively monitored by the same number of upstream neighbors. This means that all non-direct monitoring nodes normally will receive sqrt(N) indications that a node is gone. - A major drawback with ring monitoring is how it handles failures that cause massive network partitionings. If both a lost node and all its direct monitoring neighbors are inside the lost partition, the nodes in the remaining partition will never receive indications about the loss. To overcome this, each node also chooses to actively monitor some nodes outside its local domain. Those nodes are called remote domain "heads", and are selected in such a way that no node in the cluster will be more than two direct monitoring hops away. Because of this, each node, apart from monitoring the member of its local domain, will also typically monitor sqrt(N) remote head nodes. - As an optimization, local list status, domain status and domain records are marked with a generation number. This saves senders from unnecessarily conveying unaltered domain records, and receivers from performing unneeded re-adaptations of their node monitoring list, such as re-assigning domain heads. - As a measure of caution we have added the possibility to disable the new algorithm through configuration. We do this by keeping a threshold value for the cluster size; a cluster that grows beyond this value will switch from full-mesh to ring monitoring, and vice versa when it shrinks below the value. This means that if the threshold is set to a value larger than any anticipated cluster size (default size is 32) the new algorithm is effectively disabled. A patch set for altering the threshold value and for listing the table contents will follow shortly. - This change is fully backwards compatible. Acked-by: Ying Xue Signed-off-by: Jon Maloy Signed-off-by: David S. Miller diff --git a/net/tipc/Makefile b/net/tipc/Makefile index 57e460b..31b9f9c 100644 --- a/net/tipc/Makefile +++ b/net/tipc/Makefile @@ -6,7 +6,7 @@ obj-$(CONFIG_TIPC) := tipc.o tipc-y += addr.o bcast.o bearer.o \ core.o link.o discover.o msg.o \ - name_distr.o subscr.o name_table.o net.o \ + name_distr.o subscr.o monitor.o name_table.o net.o \ netlink.o netlink_compat.o node.o socket.o eth_media.o \ server.o socket.o diff --git a/net/tipc/addr.h b/net/tipc/addr.h index 93f7c98..64f4004 100644 --- a/net/tipc/addr.h +++ b/net/tipc/addr.h @@ -73,4 +73,5 @@ int tipc_addr_node_valid(u32 addr); int tipc_in_scope(u32 domain, u32 addr); int tipc_addr_scope(u32 domain); char *tipc_addr_string_fill(char *string, u32 addr); + #endif diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 6f11c62..9a70e1d 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -1,7 +1,7 @@ /* * net/tipc/bearer.c: TIPC bearer code * - * Copyright (c) 1996-2006, 2013-2014, Ericsson AB + * Copyright (c) 1996-2006, 2013-2016, Ericsson AB * Copyright (c) 2004-2006, 2010-2013, Wind River Systems * All rights reserved. * @@ -39,6 +39,7 @@ #include "bearer.h" #include "link.h" #include "discover.h" +#include "monitor.h" #include "bcast.h" #include "netlink.h" @@ -313,6 +314,10 @@ restart: rcu_assign_pointer(tn->bearer_list[bearer_id], b); if (skb) tipc_bearer_xmit_skb(net, bearer_id, skb, &b->bcast_addr); + + if (tipc_mon_create(net, bearer_id)) + return -ENOMEM; + pr_info("Enabled bearer <%s>, discovery domain %s, priority %u\n", name, tipc_addr_string_fill(addr_string, disc_domain), priority); @@ -348,6 +353,7 @@ static void bearer_disable(struct net *net, struct tipc_bearer *b) tipc_disc_delete(b->link_req); RCU_INIT_POINTER(tn->bearer_list[bearer_id], NULL); kfree_rcu(b, rcu); + tipc_mon_delete(net, bearer_id); } int tipc_enable_l2_media(struct net *net, struct tipc_bearer *b, diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index f686e41..0d337c7 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -1,7 +1,7 @@ /* * net/tipc/bearer.h: Include file for TIPC bearer code * - * Copyright (c) 1996-2006, 2013-2014, Ericsson AB + * Copyright (c) 1996-2006, 2013-2016, Ericsson AB * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * diff --git a/net/tipc/core.c b/net/tipc/core.c index fe1b062..236b043 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -57,6 +57,7 @@ static int __net_init tipc_init_net(struct net *net) tn->net_id = 4711; tn->own_addr = 0; + tn->mon_threshold = TIPC_DEF_MON_THRESHOLD; get_random_bytes(&tn->random, sizeof(int)); INIT_LIST_HEAD(&tn->node_list); spin_lock_init(&tn->node_list_lock); diff --git a/net/tipc/core.h b/net/tipc/core.h index eff58dc..a1845fb 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -66,11 +66,13 @@ struct tipc_bc_base; struct tipc_link; struct tipc_name_table; struct tipc_server; +struct tipc_monitor; #define TIPC_MOD_VER "2.0.0" -#define NODE_HTABLE_SIZE 512 -#define MAX_BEARERS 3 +#define NODE_HTABLE_SIZE 512 +#define MAX_BEARERS 3 +#define TIPC_DEF_MON_THRESHOLD 32 extern int tipc_net_id __read_mostly; extern int sysctl_tipc_rmem[3] __read_mostly; @@ -88,6 +90,10 @@ struct tipc_net { u32 num_nodes; u32 num_links; + /* Neighbor monitoring list */ + struct tipc_monitor *monitors[MAX_BEARERS]; + int mon_threshold; + /* Bearer list */ struct tipc_bearer __rcu *bearer_list[MAX_BEARERS + 1]; @@ -126,6 +132,11 @@ static inline struct list_head *tipc_nodes(struct net *net) return &tipc_net(net)->node_list; } +static inline unsigned int tipc_hashfn(u32 addr) +{ + return addr & (NODE_HTABLE_SIZE - 1); +} + static inline u16 mod(u16 x) { return x & 0xffffu; diff --git a/net/tipc/link.c b/net/tipc/link.c index a904ccd..03f8bdf 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -42,6 +42,7 @@ #include "name_distr.h" #include "discover.h" #include "netlink.h" +#include "monitor.h" #include @@ -95,6 +96,7 @@ struct tipc_stats { * @pmsg: convenience pointer to "proto_msg" field * @priority: current link priority * @net_plane: current link network plane ('A' through 'H') + * @mon_state: cookie with information needed by link monitor * @backlog_limit: backlog queue congestion thresholds (indexed by importance) * @exp_msg_count: # of tunnelled messages expected during link changeover * @reset_rcv_checkpt: seq # of last acknowledged message at time of link reset @@ -138,6 +140,7 @@ struct tipc_link { char if_name[TIPC_MAX_IF_NAME]; u32 priority; char net_plane; + struct tipc_mon_state mon_state; u16 rst_cnt; /* Failover/synch */ @@ -708,18 +711,25 @@ int tipc_link_timeout(struct tipc_link *l, struct sk_buff_head *xmitq) bool setup = false; u16 bc_snt = l->bc_sndlink->snd_nxt - 1; u16 bc_acked = l->bc_rcvlink->acked; - - link_profile_stats(l); + struct tipc_mon_state *mstate = &l->mon_state; switch (l->state) { case LINK_ESTABLISHED: case LINK_SYNCHING: - if (l->silent_intv_cnt > l->abort_limit) - return tipc_link_fsm_evt(l, LINK_FAILURE_EVT); mtyp = STATE_MSG; + link_profile_stats(l); + tipc_mon_get_state(l->net, l->addr, mstate, l->bearer_id); + if (mstate->reset || (l->silent_intv_cnt > l->abort_limit)) + return tipc_link_fsm_evt(l, LINK_FAILURE_EVT); state = bc_acked != bc_snt; - probe = l->silent_intv_cnt; - l->silent_intv_cnt++; + state |= l->bc_rcvlink->rcv_unacked; + state |= l->rcv_unacked; + state |= !skb_queue_empty(&l->transmq); + state |= !skb_queue_empty(&l->deferdq); + probe = mstate->probing; + probe |= l->silent_intv_cnt; + if (probe || mstate->monitoring) + l->silent_intv_cnt++; break; case LINK_RESET: setup = l->rst_cnt++ <= 4; @@ -830,6 +840,7 @@ void tipc_link_reset(struct tipc_link *l) l->stats.recv_info = 0; l->stale_count = 0; l->bc_peer_is_up = false; + memset(&l->mon_state, 0, sizeof(l->mon_state)); tipc_link_reset_stats(l); } @@ -1238,6 +1249,9 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe, struct tipc_msg *hdr; struct sk_buff_head *dfq = &l->deferdq; bool node_up = link_is_up(l->bc_rcvlink); + struct tipc_mon_state *mstate = &l->mon_state; + int dlen = 0; + void *data; /* Don't send protocol message during reset or link failover */ if (tipc_link_is_blocked(l)) @@ -1250,12 +1264,13 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe, rcvgap = buf_seqno(skb_peek(dfq)) - l->rcv_nxt; skb = tipc_msg_create(LINK_PROTOCOL, mtyp, INT_H_SIZE, - TIPC_MAX_IF_NAME, l->addr, + tipc_max_domain_size, l->addr, tipc_own_addr(l->net), 0, 0, 0); if (!skb) return; hdr = buf_msg(skb); + data = msg_data(hdr); msg_set_session(hdr, l->session); msg_set_bearer_id(hdr, l->bearer_id); msg_set_net_plane(hdr, l->net_plane); @@ -1271,14 +1286,18 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe, if (mtyp == STATE_MSG) { msg_set_seq_gap(hdr, rcvgap); - msg_set_size(hdr, INT_H_SIZE); msg_set_probe(hdr, probe); + tipc_mon_prep(l->net, data, &dlen, mstate, l->bearer_id); + msg_set_size(hdr, INT_H_SIZE + dlen); + skb_trim(skb, INT_H_SIZE + dlen); l->stats.sent_states++; l->rcv_unacked = 0; } else { /* RESET_MSG or ACTIVATE_MSG */ msg_set_max_pkt(hdr, l->advertised_mtu); - strcpy(msg_data(hdr), l->if_name); + strcpy(data, l->if_name); + msg_set_size(hdr, INT_H_SIZE + TIPC_MAX_IF_NAME); + skb_trim(skb, INT_H_SIZE + TIPC_MAX_IF_NAME); } if (probe) l->stats.sent_probes++; @@ -1371,7 +1390,9 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, u16 peers_tol = msg_link_tolerance(hdr); u16 peers_prio = msg_linkprio(hdr); u16 rcv_nxt = l->rcv_nxt; + u16 dlen = msg_data_sz(hdr); int mtyp = msg_type(hdr); + void *data; char *if_name; int rc = 0; @@ -1381,6 +1402,10 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, if (tipc_own_addr(l->net) > msg_prevnode(hdr)) l->net_plane = msg_net_plane(hdr); + skb_linearize(skb); + hdr = buf_msg(skb); + data = msg_data(hdr); + switch (mtyp) { case RESET_MSG: @@ -1391,8 +1416,6 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, /* fall thru' */ case ACTIVATE_MSG: - skb_linearize(skb); - hdr = buf_msg(skb); /* Complete own link name with peer's interface name */ if_name = strrchr(l->name, ':') + 1; @@ -1400,7 +1423,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, break; if (msg_data_sz(hdr) < TIPC_MAX_IF_NAME) break; - strncpy(if_name, msg_data(hdr), TIPC_MAX_IF_NAME); + strncpy(if_name, data, TIPC_MAX_IF_NAME); /* Update own tolerance if peer indicates a non-zero value */ if (in_range(peers_tol, TIPC_MIN_LINK_TOL, TIPC_MAX_LINK_TOL)) @@ -1448,6 +1471,8 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, rc = TIPC_LINK_UP_EVT; break; } + tipc_mon_rcv(l->net, data, dlen, l->addr, + &l->mon_state, l->bearer_id); /* Send NACK if peer has sent pkts we haven't received yet */ if (more(peers_snd_nxt, rcv_nxt) && !tipc_link_is_synching(l)) diff --git a/net/tipc/monitor.c b/net/tipc/monitor.c new file mode 100644 index 0000000..87d4efe --- /dev/null +++ b/net/tipc/monitor.c @@ -0,0 +1,651 @@ +/* + * net/tipc/monitor.c + * + * Copyright (c) 2016, Ericsson AB + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "addr.h" +#include "monitor.h" + +#define MAX_MON_DOMAIN 64 +#define MON_TIMEOUT 120000 +#define MAX_PEER_DOWN_EVENTS 4 + +/* struct tipc_mon_domain: domain record to be transferred between peers + * @len: actual size of domain record + * @gen: current generation of sender's domain + * @ack_gen: most recent generation of self's domain acked by peer + * @member_cnt: number of domain member nodes described in this record + * @up_map: bit map indicating which of the members the sender considers up + * @members: identity of the domain members + */ +struct tipc_mon_domain { + u16 len; + u16 gen; + u16 ack_gen; + u16 member_cnt; + u64 up_map; + u32 members[MAX_MON_DOMAIN]; +}; + +/* struct tipc_peer: state of a peer node and its domain + * @addr: tipc node identity of peer + * @head_map: shows which other nodes currently consider peer 'up' + * @domain: most recent domain record from peer + * @hash: position in hashed lookup list + * @list: position in linked list, in circular ascending order by 'addr' + * @applied: number of reported domain members applied on this monitor list + * @is_up: peer is up as seen from this node + * @is_head: peer is assigned domain head as seen from this node + * @is_local: peer is in local domain and should be continuously monitored + * @down_cnt: - numbers of other peers which have reported this on lost + */ +struct tipc_peer { + u32 addr; + struct tipc_mon_domain *domain; + struct hlist_node hash; + struct list_head list; + u8 applied; + u8 down_cnt; + bool is_up; + bool is_head; + bool is_local; +}; + +struct tipc_monitor { + struct hlist_head peers[NODE_HTABLE_SIZE]; + int peer_cnt; + struct tipc_peer *self; + rwlock_t lock; + struct tipc_mon_domain cache; + u16 list_gen; + u16 dom_gen; + struct net *net; + struct timer_list timer; + unsigned long timer_intv; +}; + +static struct tipc_monitor *tipc_monitor(struct net *net, int bearer_id) +{ + return tipc_net(net)->monitors[bearer_id]; +} + +const int tipc_max_domain_size = sizeof(struct tipc_mon_domain); + +/* dom_rec_len(): actual length of domain record for transport + */ +static int dom_rec_len(struct tipc_mon_domain *dom, u16 mcnt) +{ + return ((void *)&dom->members - (void *)dom) + (mcnt * sizeof(u32)); +} + +/* dom_size() : calculate size of own domain based on number of peers + */ +static int dom_size(int peers) +{ + int i = 0; + + while ((i * i) < peers) + i++; + return i < MAX_MON_DOMAIN ? i : MAX_MON_DOMAIN; +} + +static void map_set(u64 *up_map, int i, unsigned int v) +{ + *up_map &= ~(1 << i); + *up_map |= (v << i); +} + +static int map_get(u64 up_map, int i) +{ + return (up_map & (1 << i)) >> i; +} + +static struct tipc_peer *peer_prev(struct tipc_peer *peer) +{ + return list_last_entry(&peer->list, struct tipc_peer, list); +} + +static struct tipc_peer *peer_nxt(struct tipc_peer *peer) +{ + return list_first_entry(&peer->list, struct tipc_peer, list); +} + +static struct tipc_peer *peer_head(struct tipc_peer *peer) +{ + while (!peer->is_head) + peer = peer_prev(peer); + return peer; +} + +static struct tipc_peer *get_peer(struct tipc_monitor *mon, u32 addr) +{ + struct tipc_peer *peer; + unsigned int thash = tipc_hashfn(addr); + + hlist_for_each_entry(peer, &mon->peers[thash], hash) { + if (peer->addr == addr) + return peer; + } + return NULL; +} + +static struct tipc_peer *get_self(struct net *net, int bearer_id) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + + return mon->self; +} + +static inline bool tipc_mon_is_active(struct net *net, struct tipc_monitor *mon) +{ + struct tipc_net *tn = tipc_net(net); + + return mon->peer_cnt > tn->mon_threshold; +} + +/* mon_identify_lost_members() : - identify amd mark potentially lost members + */ +static void mon_identify_lost_members(struct tipc_peer *peer, + struct tipc_mon_domain *dom_bef, + int applied_bef) +{ + struct tipc_peer *member = peer; + struct tipc_mon_domain *dom_aft = peer->domain; + int applied_aft = peer->applied; + int i; + + for (i = 0; i < applied_bef; i++) { + member = peer_nxt(member); + + /* Do nothing if self or peer already see member as down */ + if (!member->is_up || !map_get(dom_bef->up_map, i)) + continue; + + /* Loss of local node must be detected by active probing */ + if (member->is_local) + continue; + + /* Start probing if member was removed from applied domain */ + if (!applied_aft || (applied_aft < i)) { + member->down_cnt = 1; + continue; + } + + /* Member loss is confirmed if it is still in applied domain */ + if (!map_get(dom_aft->up_map, i)) + member->down_cnt++; + } +} + +/* mon_apply_domain() : match a peer's domain record against monitor list + */ +static void mon_apply_domain(struct tipc_monitor *mon, + struct tipc_peer *peer) +{ + struct tipc_mon_domain *dom = peer->domain; + struct tipc_peer *member; + u32 addr; + int i; + + if (!dom || !peer->is_up) + return; + + /* Scan across domain members and match against monitor list */ + peer->applied = 0; + member = peer_nxt(peer); + for (i = 0; i < dom->member_cnt; i++) { + addr = dom->members[i]; + if (addr != member->addr) + return; + peer->applied++; + member = peer_nxt(member); + } +} + +/* mon_update_local_domain() : update after peer addition/removal/up/down + */ +static void mon_update_local_domain(struct tipc_monitor *mon) +{ + struct tipc_peer *self = mon->self; + struct tipc_mon_domain *cache = &mon->cache; + struct tipc_mon_domain *dom = self->domain; + struct tipc_peer *peer = self; + u64 prev_up_map = dom->up_map; + u16 member_cnt, i; + bool diff; + + /* Update local domain size based on current size of cluster */ + member_cnt = dom_size(mon->peer_cnt) - 1; + self->applied = member_cnt; + + /* Update native and cached outgoing local domain records */ + dom->len = dom_rec_len(dom, member_cnt); + diff = dom->member_cnt != member_cnt; + dom->member_cnt = member_cnt; + for (i = 0; i < member_cnt; i++) { + peer = peer_nxt(peer); + diff |= dom->members[i] != peer->addr; + dom->members[i] = peer->addr; + map_set(&dom->up_map, i, peer->is_up); + cache->members[i] = htonl(peer->addr); + } + diff |= dom->up_map != prev_up_map; + if (!diff) + return; + dom->gen = ++mon->dom_gen; + cache->len = htons(dom->len); + cache->gen = htons(dom->gen); + cache->member_cnt = htons(member_cnt); + cache->up_map = cpu_to_be64(dom->up_map); + mon_apply_domain(mon, self); +} + +/* mon_update_neighbors() : update preceding neighbors of added/removed peer + */ +static void mon_update_neighbors(struct tipc_monitor *mon, + struct tipc_peer *peer) +{ + int dz, i; + + dz = dom_size(mon->peer_cnt); + for (i = 0; i < dz; i++) { + mon_apply_domain(mon, peer); + peer = peer_prev(peer); + } +} + +/* mon_assign_roles() : reassign peer roles after a network change + * The monitor list is consistent at this stage; i.e., each peer is monitoring + * a set of domain members as matched between domain record and the monitor list + */ +static void mon_assign_roles(struct tipc_monitor *mon, struct tipc_peer *head) +{ + struct tipc_peer *peer = peer_nxt(head); + struct tipc_peer *self = mon->self; + int i = 0; + + for (; peer != self; peer = peer_nxt(peer)) { + peer->is_local = false; + + /* Update domain member */ + if (i++ < head->applied) { + peer->is_head = false; + if (head == self) + peer->is_local = true; + continue; + } + /* Assign next domain head */ + if (!peer->is_up) + continue; + if (peer->is_head) + break; + head = peer; + head->is_head = true; + i = 0; + } + mon->list_gen++; +} + +void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + struct tipc_peer *self = get_self(net, bearer_id); + struct tipc_peer *peer, *prev, *head; + + write_lock_bh(&mon->lock); + peer = get_peer(mon, addr); + if (!peer) + goto exit; + prev = peer_prev(peer); + list_del(&peer->list); + hlist_del(&peer->hash); + kfree(peer->domain); + kfree(peer); + mon->peer_cnt--; + head = peer_head(prev); + if (head == self) + mon_update_local_domain(mon); + mon_update_neighbors(mon, prev); + + /* Revert to full-mesh monitoring if we reach threshold */ + if (!tipc_mon_is_active(net, mon)) { + list_for_each_entry(peer, &self->list, list) { + kfree(peer->domain); + peer->domain = NULL; + peer->applied = 0; + } + } + mon_assign_roles(mon, head); +exit: + write_unlock_bh(&mon->lock); +} + +static bool tipc_mon_add_peer(struct tipc_monitor *mon, u32 addr, + struct tipc_peer **peer) +{ + struct tipc_peer *self = mon->self; + struct tipc_peer *cur, *prev, *p; + + p = kzalloc(sizeof(*p), GFP_ATOMIC); + *peer = p; + if (!p) + return false; + p->addr = addr; + + /* Add new peer to lookup list */ + INIT_LIST_HEAD(&p->list); + hlist_add_head(&p->hash, &mon->peers[tipc_hashfn(addr)]); + + /* Sort new peer into iterator list, in ascending circular order */ + prev = self; + list_for_each_entry(cur, &self->list, list) { + if ((addr > prev->addr) && (addr < cur->addr)) + break; + if (((addr < cur->addr) || (addr > prev->addr)) && + (prev->addr > cur->addr)) + break; + prev = cur; + } + list_add_tail(&p->list, &cur->list); + mon->peer_cnt++; + mon_update_neighbors(mon, p); + return true; +} + +void tipc_mon_peer_up(struct net *net, u32 addr, int bearer_id) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + struct tipc_peer *self = get_self(net, bearer_id); + struct tipc_peer *peer, *head; + + write_lock_bh(&mon->lock); + peer = get_peer(mon, addr); + if (!peer && !tipc_mon_add_peer(mon, addr, &peer)) + goto exit; + peer->is_up = true; + head = peer_head(peer); + if (head == self) + mon_update_local_domain(mon); + mon_assign_roles(mon, head); +exit: + write_unlock_bh(&mon->lock); +} + +void tipc_mon_peer_down(struct net *net, u32 addr, int bearer_id) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + struct tipc_peer *self = get_self(net, bearer_id); + struct tipc_peer *peer, *head; + struct tipc_mon_domain *dom; + int applied; + + write_lock_bh(&mon->lock); + peer = get_peer(mon, addr); + if (!peer) { + pr_warn("Mon: unknown link %x/%u DOWN\n", addr, bearer_id); + goto exit; + } + applied = peer->applied; + peer->applied = 0; + dom = peer->domain; + peer->domain = NULL; + if (peer->is_head) + mon_identify_lost_members(peer, dom, applied); + kfree(dom); + peer->is_up = false; + peer->is_head = false; + peer->is_local = false; + peer->down_cnt = 0; + head = peer_head(peer); + if (head == self) + mon_update_local_domain(mon); + mon_assign_roles(mon, head); +exit: + write_unlock_bh(&mon->lock); +} + +/* tipc_mon_rcv - process monitor domain event message + */ +void tipc_mon_rcv(struct net *net, void *data, u16 dlen, u32 addr, + struct tipc_mon_state *state, int bearer_id) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + struct tipc_mon_domain *arrv_dom = data; + struct tipc_mon_domain dom_bef; + struct tipc_mon_domain *dom; + struct tipc_peer *peer; + u16 new_member_cnt = ntohs(arrv_dom->member_cnt); + int new_dlen = dom_rec_len(arrv_dom, new_member_cnt); + u16 new_gen = ntohs(arrv_dom->gen); + u16 acked_gen = ntohs(arrv_dom->ack_gen); + bool probing = state->probing; + int i, applied_bef; + + state->probing = false; + if (!dlen) + return; + + /* Sanity check received domain record */ + if ((dlen < new_dlen) || ntohs(arrv_dom->len) != new_dlen) { + pr_warn_ratelimited("Received illegal domain record\n"); + return; + } + + /* Synch generation numbers with peer if link just came up */ + if (!state->synched) { + state->peer_gen = new_gen - 1; + state->acked_gen = acked_gen; + state->synched = true; + } + + if (more(acked_gen, state->acked_gen)) + state->acked_gen = acked_gen; + + /* Drop duplicate unless we are waiting for a probe response */ + if (!more(new_gen, state->peer_gen) && !probing) + return; + + write_lock_bh(&mon->lock); + peer = get_peer(mon, addr); + if (!peer || !peer->is_up) + goto exit; + + /* Peer is confirmed, stop any ongoing probing */ + peer->down_cnt = 0; + + /* Task is done for duplicate record */ + if (!more(new_gen, state->peer_gen)) + goto exit; + + state->peer_gen = new_gen; + + /* Cache current domain record for later use */ + dom_bef.member_cnt = 0; + dom = peer->domain; + if (dom) + memcpy(&dom_bef, dom, dom->len); + + /* Transform and store received domain record */ + if (!dom || (dom->len < new_dlen)) { + kfree(dom); + dom = kmalloc(new_dlen, GFP_ATOMIC); + peer->domain = dom; + if (!dom) + goto exit; + } + dom->len = new_dlen; + dom->gen = new_gen; + dom->member_cnt = new_member_cnt; + dom->up_map = be64_to_cpu(arrv_dom->up_map); + for (i = 0; i < new_member_cnt; i++) + dom->members[i] = ntohl(arrv_dom->members[i]); + + /* Update peers affected by this domain record */ + applied_bef = peer->applied; + mon_apply_domain(mon, peer); + mon_identify_lost_members(peer, &dom_bef, applied_bef); + mon_assign_roles(mon, peer_head(peer)); +exit: + write_unlock_bh(&mon->lock); +} + +void tipc_mon_prep(struct net *net, void *data, int *dlen, + struct tipc_mon_state *state, int bearer_id) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + struct tipc_mon_domain *dom = data; + u16 gen = mon->dom_gen; + u16 len; + + if (!tipc_mon_is_active(net, mon)) + return; + + /* Send only a dummy record with ack if peer has acked our last sent */ + if (likely(state->acked_gen == gen)) { + len = dom_rec_len(dom, 0); + *dlen = len; + dom->len = htons(len); + dom->gen = htons(gen); + dom->ack_gen = htons(state->peer_gen); + dom->member_cnt = 0; + return; + } + /* Send the full record */ + read_lock_bh(&mon->lock); + len = ntohs(mon->cache.len); + *dlen = len; + memcpy(data, &mon->cache, len); + read_unlock_bh(&mon->lock); + dom->ack_gen = htons(state->peer_gen); +} + +void tipc_mon_get_state(struct net *net, u32 addr, + struct tipc_mon_state *state, + int bearer_id) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + struct tipc_peer *peer; + + /* Used cached state if table has not changed */ + if (!state->probing && + (state->list_gen == mon->list_gen) && + (state->acked_gen == mon->dom_gen)) + return; + + read_lock_bh(&mon->lock); + peer = get_peer(mon, addr); + if (peer) { + state->probing = state->acked_gen != mon->dom_gen; + state->probing |= peer->down_cnt; + state->reset |= peer->down_cnt >= MAX_PEER_DOWN_EVENTS; + state->monitoring = peer->is_local; + state->monitoring |= peer->is_head; + state->list_gen = mon->list_gen; + } + read_unlock_bh(&mon->lock); +} + +static void mon_timeout(unsigned long m) +{ + struct tipc_monitor *mon = (void *)m; + struct tipc_peer *self; + int best_member_cnt = dom_size(mon->peer_cnt) - 1; + + write_lock_bh(&mon->lock); + self = mon->self; + if (self && (best_member_cnt != self->applied)) { + mon_update_local_domain(mon); + mon_assign_roles(mon, self); + } + write_unlock_bh(&mon->lock); + mod_timer(&mon->timer, jiffies + mon->timer_intv); +} + +int tipc_mon_create(struct net *net, int bearer_id) +{ + struct tipc_net *tn = tipc_net(net); + struct tipc_monitor *mon; + struct tipc_peer *self; + struct tipc_mon_domain *dom; + + if (tn->monitors[bearer_id]) + return 0; + + mon = kzalloc(sizeof(*mon), GFP_ATOMIC); + self = kzalloc(sizeof(*self), GFP_ATOMIC); + dom = kzalloc(sizeof(*dom), GFP_ATOMIC); + if (!mon || !self || !dom) { + kfree(mon); + kfree(self); + kfree(dom); + return -ENOMEM; + } + tn->monitors[bearer_id] = mon; + rwlock_init(&mon->lock); + mon->net = net; + mon->peer_cnt = 1; + mon->self = self; + self->domain = dom; + self->addr = tipc_own_addr(net); + self->is_up = true; + self->is_head = true; + INIT_LIST_HEAD(&self->list); + setup_timer(&mon->timer, mon_timeout, (unsigned long)mon); + mon->timer_intv = msecs_to_jiffies(MON_TIMEOUT + (tn->random & 0xffff)); + mod_timer(&mon->timer, jiffies + mon->timer_intv); + return 0; +} + +void tipc_mon_delete(struct net *net, int bearer_id) +{ + struct tipc_net *tn = tipc_net(net); + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + struct tipc_peer *self = get_self(net, bearer_id); + struct tipc_peer *peer, *tmp; + + write_lock_bh(&mon->lock); + tn->monitors[bearer_id] = NULL; + list_for_each_entry_safe(peer, tmp, &self->list, list) { + list_del(&peer->list); + hlist_del(&peer->hash); + kfree(peer->domain); + kfree(peer); + } + mon->self = NULL; + write_unlock_bh(&mon->lock); + del_timer_sync(&mon->timer); + kfree(self->domain); + kfree(self); + kfree(mon); +} diff --git a/net/tipc/monitor.h b/net/tipc/monitor.h new file mode 100644 index 0000000..598459c --- /dev/null +++ b/net/tipc/monitor.h @@ -0,0 +1,73 @@ +/* + * net/tipc/monitor.h + * + * Copyright (c) 2015, Ericsson AB + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_MONITOR_H +#define _TIPC_MONITOR_H + +/* struct tipc_mon_state: link instance's cache of monitor list and domain state + * @list_gen: current generation of this node's monitor list + * @gen: current generation of this node's local domain + * @peer_gen: most recent domain generation received from peer + * @acked_gen: most recent generation of self's domain acked by peer + * @monitoring: this peer endpoint should continuously monitored + * @probing: peer endpoint should be temporarily probed for potential loss + * @synched: domain record's generation has been synched with peer after reset + */ +struct tipc_mon_state { + u16 list_gen; + u16 peer_gen; + u16 acked_gen; + bool monitoring :1; + bool probing :1; + bool reset :1; + bool synched :1; +}; + +int tipc_mon_create(struct net *net, int bearer_id); +void tipc_mon_delete(struct net *net, int bearer_id); + +void tipc_mon_peer_up(struct net *net, u32 addr, int bearer_id); +void tipc_mon_peer_down(struct net *net, u32 addr, int bearer_id); +void tipc_mon_prep(struct net *net, void *data, int *dlen, + struct tipc_mon_state *state, int bearer_id); +void tipc_mon_rcv(struct net *net, void *data, u16 dlen, u32 addr, + struct tipc_mon_state *state, int bearer_id); +void tipc_mon_get_state(struct net *net, u32 addr, + struct tipc_mon_state *state, + int bearer_id); +void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id); + +extern const int tipc_max_domain_size; +#endif diff --git a/net/tipc/node.c b/net/tipc/node.c index d6a490f..a3fc0a3 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -40,6 +40,7 @@ #include "name_distr.h" #include "socket.h" #include "bcast.h" +#include "monitor.h" #include "discover.h" #include "netlink.h" @@ -205,17 +206,6 @@ u16 tipc_node_get_capabilities(struct net *net, u32 addr) return caps; } -/* - * A trivial power-of-two bitmask technique is used for speed, since this - * operation is done for every incoming TIPC packet. The number of hash table - * entries has been chosen so that no hash chain exceeds 8 nodes and will - * usually be much smaller (typically only a single node). - */ -static unsigned int tipc_hashfn(u32 addr) -{ - return addr & (NODE_HTABLE_SIZE - 1); -} - static void tipc_node_kref_release(struct kref *kref) { struct tipc_node *n = container_of(kref, struct tipc_node, kref); @@ -279,6 +269,7 @@ static void tipc_node_write_unlock(struct tipc_node *n) u32 addr = 0; u32 flags = n->action_flags; u32 link_id = 0; + u32 bearer_id; struct list_head *publ_list; if (likely(!flags)) { @@ -288,6 +279,7 @@ static void tipc_node_write_unlock(struct tipc_node *n) addr = n->addr; link_id = n->link_id; + bearer_id = link_id & 0xffff; publ_list = &n->publ_list; n->action_flags &= ~(TIPC_NOTIFY_NODE_DOWN | TIPC_NOTIFY_NODE_UP | @@ -301,13 +293,16 @@ static void tipc_node_write_unlock(struct tipc_node *n) if (flags & TIPC_NOTIFY_NODE_UP) tipc_named_node_up(net, addr); - if (flags & TIPC_NOTIFY_LINK_UP) + if (flags & TIPC_NOTIFY_LINK_UP) { + tipc_mon_peer_up(net, addr, bearer_id); tipc_nametbl_publish(net, TIPC_LINK_STATE, addr, addr, TIPC_NODE_SCOPE, link_id, addr); - - if (flags & TIPC_NOTIFY_LINK_DOWN) + } + if (flags & TIPC_NOTIFY_LINK_DOWN) { + tipc_mon_peer_down(net, addr, bearer_id); tipc_nametbl_withdraw(net, TIPC_LINK_STATE, addr, link_id, addr); + } } struct tipc_node *tipc_node_create(struct net *net, u32 addr, u16 capabilities) @@ -691,6 +686,7 @@ static void tipc_node_link_down(struct tipc_node *n, int bearer_id, bool delete) struct tipc_link *l = le->link; struct tipc_media_addr *maddr; struct sk_buff_head xmitq; + int old_bearer_id = bearer_id; if (!l) return; @@ -710,6 +706,8 @@ static void tipc_node_link_down(struct tipc_node *n, int bearer_id, bool delete) tipc_link_fsm_evt(l, LINK_RESET_EVT); } tipc_node_write_unlock(n); + if (delete) + tipc_mon_remove_peer(n->net, n->addr, old_bearer_id); tipc_bearer_xmit(n->net, bearer_id, &xmitq, maddr); tipc_sk_rcv(n->net, &le->inputq); } -- cgit v0.10.2 From 1b5c5493e3e68181be344cb51bf9df192d05ffc2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:50 -0700 Subject: net_sched: add the ability to defer skb freeing qdisc are changed under RTNL protection and often while blocking BH and root qdisc spinlock. When lots of skbs need to be dropped, we free them under these locks causing TX/RX freezes, and more generally latency spikes. This commit adds rtnl_kfree_skbs(), used to queue skbs for deferred freeing. Actual freeing happens right after RTNL is released, with appropriate scheduling points. rtnl_qdisc_drop() can also be used in place of disc_drop() when RTNL is held. qdisc_reset_queue() and __qdisc_reset_queue() get the new behavior, so standard qdiscs like pfifo, pfifo_fast... have their ->reset() method automatically handled. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index c006cc9..2daece8 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -89,8 +89,9 @@ void net_inc_egress_queue(void); void net_dec_egress_queue(void); #endif -extern void rtnetlink_init(void); -extern void __rtnl_unlock(void); +void rtnetlink_init(void); +void __rtnl_unlock(void); +void rtnl_kfree_skbs(struct sk_buff *head, struct sk_buff *tail); #define ASSERT_RTNL() do { \ if (unlikely(!rtnl_is_locked())) { \ diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 9a0d177..4f7cee8 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -683,19 +683,21 @@ static inline struct sk_buff *qdisc_dequeue_peeked(struct Qdisc *sch) return skb; } -static inline void __qdisc_reset_queue(struct Qdisc *sch, - struct sk_buff_head *list) +static inline void __qdisc_reset_queue(struct sk_buff_head *list) { /* * We do not know the backlog in bytes of this list, it * is up to the caller to correct it */ - __skb_queue_purge(list); + if (!skb_queue_empty(list)) { + rtnl_kfree_skbs(list->next, list->prev); + __skb_queue_head_init(list); + } } static inline void qdisc_reset_queue(struct Qdisc *sch) { - __qdisc_reset_queue(sch, &sch->q); + __qdisc_reset_queue(&sch->q); sch->qstats.backlog = 0; } @@ -716,6 +718,12 @@ static inline struct Qdisc *qdisc_replace(struct Qdisc *sch, struct Qdisc *new, return old; } +static inline void rtnl_qdisc_drop(struct sk_buff *skb, struct Qdisc *sch) +{ + rtnl_kfree_skbs(skb, skb); + qdisc_qstats_drop(sch); +} + static inline int qdisc_drop(struct sk_buff *skb, struct Qdisc *sch) { kfree_skb(skb); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index d69c464..eb49ca2 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -71,9 +71,31 @@ void rtnl_lock(void) } EXPORT_SYMBOL(rtnl_lock); +static struct sk_buff *defer_kfree_skb_list; +void rtnl_kfree_skbs(struct sk_buff *head, struct sk_buff *tail) +{ + if (head && tail) { + tail->next = defer_kfree_skb_list; + defer_kfree_skb_list = head; + } +} +EXPORT_SYMBOL(rtnl_kfree_skbs); + void __rtnl_unlock(void) { + struct sk_buff *head = defer_kfree_skb_list; + + defer_kfree_skb_list = NULL; + mutex_unlock(&rtnl_mutex); + + while (head) { + struct sk_buff *next = head->next; + + kfree_skb(head); + cond_resched(); + head = next; + } } void rtnl_unlock(void) diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 0c9cb51..773b632 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -493,7 +493,7 @@ static void pfifo_fast_reset(struct Qdisc *qdisc) struct pfifo_fast_priv *priv = qdisc_priv(qdisc); for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) - __qdisc_reset_queue(qdisc, band2list(priv, prio)); + __qdisc_reset_queue(band2list(priv, prio)); priv->bitmap = 0; qdisc->qstats.backlog = 0; -- cgit v0.10.2 From f9aed311b682ca933ca78821cce3e4f1d69d4f56 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:51 -0700 Subject: net_sched: sch_choke: defer skb freeing choke_reset() and choke_change() can use rtnl_qdisc_drop() to defer expensive skb freeing after locks are released. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c index 04e0b05..789b69e 100644 --- a/net/sched/sch_choke.c +++ b/net/sched/sch_choke.c @@ -375,11 +375,11 @@ static void choke_reset(struct Qdisc *sch) q->head = (q->head + 1) & q->tab_mask; if (!skb) continue; - qdisc_qstats_backlog_dec(sch, skb); - --sch->q.qlen; - qdisc_drop(skb, sch); + rtnl_qdisc_drop(skb, sch); } + sch->q.qlen = 0; + sch->qstats.backlog = 0; memset(q->tab, 0, (q->tab_mask + 1) * sizeof(struct sk_buff *)); q->head = q->tail = 0; red_restart(&q->vars); @@ -455,7 +455,7 @@ static int choke_change(struct Qdisc *sch, struct nlattr *opt) dropped += qdisc_pkt_len(skb); qdisc_qstats_backlog_dec(sch, skb); --sch->q.qlen; - qdisc_drop(skb, sch); + rtnl_qdisc_drop(skb, sch); } qdisc_tree_reduce_backlog(sch, oqlen - sch->q.qlen, dropped); q->head = 0; -- cgit v0.10.2 From b3d7e2b29b226c986cbd4efcaf43ab3ff90e6fdb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:52 -0700 Subject: net_sched: sch_codel: defer skb freeing in codel_change() codel_change() can use rtnl_qdisc_drop() to defer expensive skb freeing after locks are released. codel_reset() already has support for deferred skb freeing because it uses qdisc_reset_queue() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c index dddf3bb..c5bc424 100644 --- a/net/sched/sch_codel.c +++ b/net/sched/sch_codel.c @@ -174,7 +174,7 @@ static int codel_change(struct Qdisc *sch, struct nlattr *opt) dropped += qdisc_pkt_len(skb); qdisc_qstats_backlog_dec(sch, skb); - qdisc_drop(skb, sch); + rtnl_qdisc_drop(skb, sch); } qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, dropped); -- cgit v0.10.2 From e14ffdfdd67a9eba36bd76af0df1f2ac363ae906 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:53 -0700 Subject: net_sched: sch_fq: defer skb freeing Both fq_change() and fq_reset() can use rtnl_kfree_skbs() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index f49c81e..6eb0667 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -514,17 +514,25 @@ out: return skb; } +static void fq_flow_purge(struct fq_flow *flow) +{ + rtnl_kfree_skbs(flow->head, flow->tail); + flow->head = NULL; + flow->qlen = 0; +} + static void fq_reset(struct Qdisc *sch) { struct fq_sched_data *q = qdisc_priv(sch); struct rb_root *root; - struct sk_buff *skb; struct rb_node *p; struct fq_flow *f; unsigned int idx; - while ((skb = fq_dequeue_head(sch, &q->internal)) != NULL) - kfree_skb(skb); + sch->q.qlen = 0; + sch->qstats.backlog = 0; + + fq_flow_purge(&q->internal); if (!q->fq_root) return; @@ -535,8 +543,7 @@ static void fq_reset(struct Qdisc *sch) f = container_of(p, struct fq_flow, fq_node); rb_erase(p, root); - while ((skb = fq_dequeue_head(sch, f)) != NULL) - kfree_skb(skb); + fq_flow_purge(f); kmem_cache_free(fq_flow_cachep, f); } @@ -737,7 +744,7 @@ static int fq_change(struct Qdisc *sch, struct nlattr *opt) if (!skb) break; drop_len += qdisc_pkt_len(skb); - kfree_skb(skb); + rtnl_kfree_skbs(skb, skb); drop_count++; } qdisc_tree_reduce_backlog(sch, drop_count, drop_len); -- cgit v0.10.2 From ece5d4c723b69a4aa0cd1f545ebbdc3a37c80dc6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:54 -0700 Subject: net_sched: fq_codel: defer skb freeing Both fq_codel_change() and fq_codel_reset() can use rtnl_kfree_skbs() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index a302e8e..2dc0a84 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -336,6 +336,12 @@ begin: return skb; } +static void fq_codel_flow_purge(struct fq_codel_flow *flow) +{ + rtnl_kfree_skbs(flow->head, flow->tail); + flow->head = NULL; +} + static void fq_codel_reset(struct Qdisc *sch) { struct fq_codel_sched_data *q = qdisc_priv(sch); @@ -346,18 +352,13 @@ static void fq_codel_reset(struct Qdisc *sch) for (i = 0; i < q->flows_cnt; i++) { struct fq_codel_flow *flow = q->flows + i; - while (flow->head) { - struct sk_buff *skb = dequeue_head(flow); - - qdisc_qstats_backlog_dec(sch, skb); - kfree_skb(skb); - } - + fq_codel_flow_purge(flow); INIT_LIST_HEAD(&flow->flowchain); codel_vars_init(&flow->cvars); } memset(q->backlogs, 0, q->flows_cnt * sizeof(u32)); sch->q.qlen = 0; + sch->qstats.backlog = 0; q->memory_usage = 0; } @@ -433,7 +434,7 @@ static int fq_codel_change(struct Qdisc *sch, struct nlattr *opt) struct sk_buff *skb = fq_codel_dequeue(sch); q->cstats.drop_len += qdisc_pkt_len(skb); - kfree_skb(skb); + rtnl_kfree_skbs(skb, skb); q->cstats.drop_count++; } qdisc_tree_reduce_backlog(sch, q->cstats.drop_count, q->cstats.drop_len); -- cgit v0.10.2 From e7e424cdc4b2fcd7507b71d3a931708d11d5a61e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:55 -0700 Subject: net_sched: sch_hhf: defer skb freeing Both hhf_reset() and hhf_change() can use rtnl_kfree_skbs() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c index c517918..c44593b 100644 --- a/net/sched/sch_hhf.c +++ b/net/sched/sch_hhf.c @@ -464,7 +464,7 @@ static void hhf_reset(struct Qdisc *sch) struct sk_buff *skb; while ((skb = hhf_dequeue(sch)) != NULL) - kfree_skb(skb); + rtnl_kfree_skbs(skb, skb); } static void *hhf_zalloc(size_t sz) @@ -574,7 +574,7 @@ static int hhf_change(struct Qdisc *sch, struct nlattr *opt) while (sch->q.qlen > sch->limit) { struct sk_buff *skb = hhf_dequeue(sch); - kfree_skb(skb); + rtnl_kfree_skbs(skb, skb); } qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, prev_backlog - sch->qstats.backlog); -- cgit v0.10.2 From a5a9f5346fb95c00d47e8b4648f6d5eada147cd4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:56 -0700 Subject: net_sched: sch_htb: defer skb freeing Both htb_reset() and htb_destroy() can use __qdisc_reset_queue() instead of __skb_queue_purge() to defer skb freeing of internal queues. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index 07dcd29..a454605 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -957,7 +957,7 @@ static void htb_reset(struct Qdisc *sch) } } qdisc_watchdog_cancel(&q->watchdog); - __skb_queue_purge(&q->direct_queue); + __qdisc_reset_queue(&q->direct_queue); sch->q.qlen = 0; sch->qstats.backlog = 0; memset(q->hlevel, 0, sizeof(q->hlevel)); @@ -1231,7 +1231,7 @@ static void htb_destroy(struct Qdisc *sch) htb_destroy_class(sch, cl); } qdisc_class_hash_destroy(&q->clhash); - __skb_queue_purge(&q->direct_queue); + __qdisc_reset_queue(&q->direct_queue); } static int htb_delete(struct Qdisc *sch, unsigned long arg) -- cgit v0.10.2 From 2f08a9a16288b60df3ddfe97c965427ce0163297 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:57 -0700 Subject: net_sched: sch_netem: defer skb freeing rtnl_kfree_skbs() can be used in tfifo_reset() It would be nice if we could iterate through rb tree instead of removing one skb at a time, and build a single skb chain. But this is left for a future patch. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 876df13..e271967 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -368,9 +368,7 @@ static void tfifo_reset(struct Qdisc *sch) struct sk_buff *skb = netem_rb_to_skb(p); rb_erase(p, &q->t_root); - skb->next = NULL; - skb->prev = NULL; - kfree_skb(skb); + rtnl_kfree_skbs(skb, skb); } } -- cgit v0.10.2 From db4879d93c351cb978db1eb4c963f44d267d63a2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:58 -0700 Subject: net_sched: sch_pie: defer skb freeing pie_change() can use rtnl_qdisc_drop() to benefit from deferred freeing. pie_reset() is already using qdisc_reset_queue() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c index 71ae3b9..912a46a 100644 --- a/net/sched/sch_pie.c +++ b/net/sched/sch_pie.c @@ -234,7 +234,7 @@ static int pie_change(struct Qdisc *sch, struct nlattr *opt) dropped += qdisc_pkt_len(skb); qdisc_qstats_backlog_dec(sch, skb); - qdisc_drop(skb, sch); + rtnl_qdisc_drop(skb, sch); } qdisc_tree_reduce_backlog(sch, qlen - sch->q.qlen, dropped); -- cgit v0.10.2 From fea024784f588a1c50e7718d6053697ebdcc033e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:59 -0700 Subject: net_sched: sch_fq: defer skb freeing sfq_reset() can use rtnl_kfree_skbs() instead of kfree_skb() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index a2e0b85..57d118b 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -520,7 +520,7 @@ sfq_reset(struct Qdisc *sch) struct sk_buff *skb; while ((skb = sfq_dequeue(sch)) != NULL) - kfree_skb(skb); + rtnl_kfree_skbs(skb, skb); } /* -- cgit v0.10.2 From c5a8c0f3aa726db65b27481d284ca6c427c7a78f Mon Sep 17 00:00:00 2001 From: Hariprasad Shenai Date: Tue, 14 Jun 2016 14:39:30 +0530 Subject: cxgb4: Force cxgb4 driver as MASTER in kdump kernel When is_kdump_kernel() is true, Forcing cxgb4 driver as Master so we can reinitialize the Firmware/Chip. Also reduce memory usage by disabling offload. Signed-off-by: Hariprasad Shenai Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 477db47..c045f65 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -64,6 +64,7 @@ #include #include #include +#include #include "cxgb4.h" #include "t4_regs.h" @@ -3735,7 +3736,8 @@ static int adap_init0(struct adapter *adap) return ret; /* Contact FW, advertising Master capability */ - ret = t4_fw_hello(adap, adap->mbox, adap->mbox, MASTER_MAY, &state); + ret = t4_fw_hello(adap, adap->mbox, adap->mbox, + is_kdump_kernel() ? MASTER_MUST : MASTER_MAY, &state); if (ret < 0) { dev_err(adap->pdev_dev, "could not connect to FW, error %d\n", ret); @@ -4366,6 +4368,11 @@ static void cfg_queues(struct adapter *adap) if (q10g > netif_get_num_default_rss_queues()) q10g = netif_get_num_default_rss_queues(); + /* Reduce memory usage in kdump environment, disable all offload. + */ + if (is_kdump_kernel()) + adap->params.offload = 0; + for_each_port(adap, i) { struct port_info *pi = adap2pinfo(adap, i); -- cgit v0.10.2 From b6244201f4197860fa589d775285fe2b5e632545 Mon Sep 17 00:00:00 2001 From: Hariprasad Shenai Date: Tue, 14 Jun 2016 14:39:31 +0530 Subject: cxgb4: Enable SR-IOV configuration via PCI sysfs interface Implement callback in the driver for the new PCI bus driver interface that allows the user to enable/disable SR-IOV virtual functions in a device via the sysfs interface. Deprecate module parameter used to configure SRIOV Signed-off-by: Hariprasad Shenai Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index c045f65..6ce4344 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -207,7 +207,7 @@ static int rx_dma_offset = 2; static unsigned int num_vf[NUM_OF_PF_WITH_SRIOV]; module_param_array(num_vf, uint, NULL, 0644); -MODULE_PARM_DESC(num_vf, "number of VFs for each of PFs 0-3"); +MODULE_PARM_DESC(num_vf, "number of VFs for each of PFs 0-3, deprecated parameter - please use the pci sysfs interface."); #endif /* TX Queue select used to determine what algorithm to use for selecting TX @@ -4836,6 +4836,60 @@ static int get_chip_type(struct pci_dev *pdev, u32 pl_rev) return -EINVAL; } +#ifdef CONFIG_PCI_IOV +static int cxgb4_iov_configure(struct pci_dev *pdev, int num_vfs) +{ + int err = 0; + int current_vfs = pci_num_vf(pdev); + u32 pcie_fw; + void __iomem *regs; + + regs = pci_ioremap_bar(pdev, 0); + if (!regs) { + dev_err(&pdev->dev, "cannot map device registers\n"); + return -ENOMEM; + } + + pcie_fw = readl(regs + PCIE_FW_A); + iounmap(regs); + /* Check if cxgb4 is the MASTER and fw is initialized */ + if (!(pcie_fw & PCIE_FW_INIT_F) || + !(pcie_fw & PCIE_FW_MASTER_VLD_F) || + PCIE_FW_MASTER_G(pcie_fw) != 4) { + dev_warn(&pdev->dev, + "cxgb4 driver needs to be MASTER to support SRIOV\n"); + return -EOPNOTSUPP; + } + + /* If any of the VF's is already assigned to Guest OS, then + * SRIOV for the same cannot be modified + */ + if (current_vfs && pci_vfs_assigned(pdev)) { + dev_err(&pdev->dev, + "Cannot modify SR-IOV while VFs are assigned\n"); + num_vfs = current_vfs; + return num_vfs; + } + + /* Disable SRIOV when zero is passed. + * One needs to disable SRIOV before modifying it, else + * stack throws the below warning: + * " 'n' VFs already enabled. Disable before enabling 'm' VFs." + */ + if (!num_vfs) { + pci_disable_sriov(pdev); + return num_vfs; + } + + if (num_vfs != current_vfs) { + err = pci_enable_sriov(pdev, num_vfs); + if (err) + return err; + } + return num_vfs; +} +#endif + static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { int func, i, err, s_qpp, qpp, num_seg; @@ -5169,11 +5223,16 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) sriov: #ifdef CONFIG_PCI_IOV - if (func < ARRAY_SIZE(num_vf) && num_vf[func] > 0) + if (func < ARRAY_SIZE(num_vf) && num_vf[func] > 0) { + dev_warn(&pdev->dev, + "Enabling SR-IOV VFs using the num_vf module " + "parameter is deprecated - please use the pci sysfs " + "interface instead.\n"); if (pci_enable_sriov(pdev, num_vf[func]) == 0) dev_info(&pdev->dev, "instantiated %u virtual functions\n", num_vf[func]); + } #endif return 0; @@ -5266,6 +5325,9 @@ static struct pci_driver cxgb4_driver = { .probe = init_one, .remove = remove_one, .shutdown = remove_one, +#ifdef CONFIG_PCI_IOV + .sriov_configure = cxgb4_iov_configure, +#endif .err_handler = &cxgb4_eeh, }; -- cgit v0.10.2 From d01f7abc91cad02bb4ff771d074c598e01ffdb58 Mon Sep 17 00:00:00 2001 From: Hariprasad Shenai Date: Tue, 14 Jun 2016 14:39:32 +0530 Subject: cxgb4/cxgb4vf: Synchronize all MAC addresses Even if interface is in Promiscuous mode/Allmulti mode synchronize MAC addresses. Signed-off-by: Hariprasad Shenai Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 6ce4344..c45de49 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -461,11 +461,8 @@ static int set_rxmode(struct net_device *dev, int mtu, bool sleep_ok) struct port_info *pi = netdev_priv(dev); struct adapter *adapter = pi->adapter; - if (!(dev->flags & IFF_PROMISC)) { - __dev_uc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync); - if (!(dev->flags & IFF_ALLMULTI)) - __dev_mc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync); - } + __dev_uc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync); + __dev_mc_sync(dev, cxgb4_mac_sync, cxgb4_mac_unsync); return t4_set_rxmode(adapter, adapter->mbox, pi->viid, mtu, (dev->flags & IFF_PROMISC) ? 1 : 0, diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index 04fc6f6..8d9b2cb 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -937,12 +937,8 @@ static int set_rxmode(struct net_device *dev, int mtu, bool sleep_ok) { struct port_info *pi = netdev_priv(dev); - if (!(dev->flags & IFF_PROMISC)) { - __dev_uc_sync(dev, cxgb4vf_mac_sync, cxgb4vf_mac_unsync); - if (!(dev->flags & IFF_ALLMULTI)) - __dev_mc_sync(dev, cxgb4vf_mac_sync, - cxgb4vf_mac_unsync); - } + __dev_uc_sync(dev, cxgb4vf_mac_sync, cxgb4vf_mac_unsync); + __dev_mc_sync(dev, cxgb4vf_mac_sync, cxgb4vf_mac_unsync); return t4vf_set_rxmode(pi->adapter, pi->viid, -1, (dev->flags & IFF_PROMISC) != 0, (dev->flags & IFF_ALLMULTI) != 0, -- cgit v0.10.2 From 8626a0c83b0d471d859bcd908d016874df951fc3 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:16 +0200 Subject: 6lowpan: add private neighbour data This patch will introduce a 6lowpan neighbour private data. Like the interface private data we handle private data for generic 6lowpan and for link-layer specific 6lowpan. The current first use case if to save the short address for a 802.15.4 6lowpan neighbour. Cc: David S. Miller Reviewed-by: Stefan Schmidt Acked-by: YOSHIFUJI Hideaki Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d101e4d..36e43bd 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1483,8 +1483,7 @@ enum netdev_priv_flags { * @perm_addr: Permanent hw address * @addr_assign_type: Hw address assignment type * @addr_len: Hardware address length - * @neigh_priv_len; Used in neigh_alloc(), - * initialized only in atm/clip.c + * @neigh_priv_len: Used in neigh_alloc() * @dev_id: Used to differentiate devices that share * the same link layer address * @dev_port: Used to differentiate devices that share diff --git a/include/net/6lowpan.h b/include/net/6lowpan.h index da84cf9..2d9b9d3 100644 --- a/include/net/6lowpan.h +++ b/include/net/6lowpan.h @@ -141,6 +141,16 @@ struct lowpan_dev { u8 priv[0] __aligned(sizeof(void *)); }; +struct lowpan_802154_neigh { + __le16 short_addr; +}; + +static inline +struct lowpan_802154_neigh *lowpan_802154_neigh(void *neigh_priv) +{ + return neigh_priv; +} + static inline struct lowpan_dev *lowpan_dev(const struct net_device *dev) { diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index 4e2b308..8c004a0 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -81,11 +81,21 @@ static int lowpan_stop(struct net_device *dev) return 0; } +static int lowpan_neigh_construct(struct neighbour *n) +{ + struct lowpan_802154_neigh *neigh = lowpan_802154_neigh(neighbour_priv(n)); + + /* default no short_addr is available for a neighbour */ + neigh->short_addr = cpu_to_le16(IEEE802154_ADDR_SHORT_UNSPEC); + return 0; +} + static const struct net_device_ops lowpan_netdev_ops = { .ndo_init = lowpan_dev_init, .ndo_start_xmit = lowpan_xmit, .ndo_open = lowpan_open, .ndo_stop = lowpan_stop, + .ndo_neigh_construct = lowpan_neigh_construct, }; static void lowpan_setup(struct net_device *ldev) @@ -150,6 +160,8 @@ static int lowpan_newlink(struct net *src_net, struct net_device *ldev, wdev->needed_headroom; ldev->needed_tailroom = wdev->needed_tailroom; + ldev->neigh_priv_len = sizeof(struct lowpan_802154_neigh); + ret = lowpan_register_netdevice(ldev, LOWPAN_LLTYPE_IEEE802154); if (ret < 0) { dev_put(wdev); -- cgit v0.10.2 From 2ad3ed59198c5404c34515cfcfd9a2b3c54d964f Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:17 +0200 Subject: 6lowpan: add 802.15.4 short addr slaac This patch adds the autoconfiguration if a valid 802.15.4 short address is available for 802.15.4 6LoWPAN interfaces. Cc: David S. Miller Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Acked-by: Hannes Frederic Sowa Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/include/net/6lowpan.h b/include/net/6lowpan.h index 2d9b9d3..5ab4c99 100644 --- a/include/net/6lowpan.h +++ b/include/net/6lowpan.h @@ -254,6 +254,12 @@ static inline bool lowpan_fetch_skb(struct sk_buff *skb, void *data, return false; } +static inline bool lowpan_802154_is_valid_src_short_addr(__le16 addr) +{ + /* First bit of addr is multicast, reserved or 802.15.4 specific */ + return !(addr & cpu_to_le16(0x8000)); +} + static inline void lowpan_push_hc_data(u8 **hc_ptr, const void *data, const size_t len) { diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 730d856..b1774eb 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -94,6 +94,9 @@ int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2, void addrconf_join_solict(struct net_device *dev, const struct in6_addr *addr); void addrconf_leave_solict(struct inet6_dev *idev, const struct in6_addr *addr); +void addrconf_add_linklocal(struct inet6_dev *idev, + const struct in6_addr *addr, u32 flags); + static inline int addrconf_ifid_eui48(u8 *eui, struct net_device *dev) { if (dev->addr_len != ETH_ALEN) diff --git a/net/6lowpan/core.c b/net/6lowpan/core.c index 7a240b3..801404c 100644 --- a/net/6lowpan/core.c +++ b/net/6lowpan/core.c @@ -14,6 +14,7 @@ #include #include +#include #include "6lowpan_i.h" @@ -72,16 +73,61 @@ void lowpan_unregister_netdev(struct net_device *dev) } EXPORT_SYMBOL(lowpan_unregister_netdev); +static int addrconf_ifid_802154_6lowpan(u8 *eui, struct net_device *dev) +{ + struct wpan_dev *wpan_dev = lowpan_802154_dev(dev)->wdev->ieee802154_ptr; + + /* Set short_addr autoconfiguration if short_addr is present only */ + if (!lowpan_802154_is_valid_src_short_addr(wpan_dev->short_addr)) + return -1; + + /* For either address format, all zero addresses MUST NOT be used */ + if (wpan_dev->pan_id == cpu_to_le16(0x0000) && + wpan_dev->short_addr == cpu_to_le16(0x0000)) + return -1; + + /* Alternatively, if no PAN ID is known, 16 zero bits may be used */ + if (wpan_dev->pan_id == cpu_to_le16(IEEE802154_PAN_ID_BROADCAST)) + memset(eui, 0, 2); + else + ieee802154_le16_to_be16(eui, &wpan_dev->pan_id); + + /* The "Universal/Local" (U/L) bit shall be set to zero */ + eui[0] &= ~2; + eui[2] = 0; + eui[3] = 0xFF; + eui[4] = 0xFE; + eui[5] = 0; + ieee802154_le16_to_be16(&eui[6], &wpan_dev->short_addr); + return 0; +} + static int lowpan_event(struct notifier_block *unused, unsigned long event, void *ptr) { struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct inet6_dev *idev; + struct in6_addr addr; int i; if (dev->type != ARPHRD_6LOWPAN) return NOTIFY_DONE; + idev = __in6_dev_get(dev); + if (!idev) + return NOTIFY_DONE; + switch (event) { + case NETDEV_UP: + case NETDEV_CHANGE: + /* (802.15.4 6LoWPAN short address slaac handling */ + if (lowpan_is_ll(dev, LOWPAN_LLTYPE_IEEE802154) && + addrconf_ifid_802154_6lowpan(addr.s6_addr + 8, dev) == 0) { + __ipv6_addr_set_half(&addr.s6_addr32[0], + htonl(0xFE800000), 0); + addrconf_add_linklocal(idev, &addr, 0); + } + break; case NETDEV_DOWN: for (i = 0; i < LOWPAN_IPHC_CTX_TABLE_SIZE; i++) clear_bit(LOWPAN_IPHC_CTX_FLAG_ACTIVE, diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index b125539..1ce4048 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2947,8 +2947,8 @@ static void init_loopback(struct net_device *dev) } } -static void addrconf_add_linklocal(struct inet6_dev *idev, - const struct in6_addr *addr, u32 flags) +void addrconf_add_linklocal(struct inet6_dev *idev, + const struct in6_addr *addr, u32 flags) { struct inet6_ifaddr *ifp; u32 addr_flags = flags | IFA_F_PERMANENT; @@ -2967,6 +2967,7 @@ static void addrconf_add_linklocal(struct inet6_dev *idev, in6_ifa_put(ifp); } } +EXPORT_SYMBOL_GPL(addrconf_add_linklocal); static bool ipv6_reserved_interfaceid(struct in6_addr address) { -- cgit v0.10.2 From 848484c93128eae5b0f879ad1c2d7204c10a8b6a Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:18 +0200 Subject: 6lowpan: remove ipv6 module request Since we use exported function from ipv6 kernel module we don't need to request the module anymore to have ipv6 functionality. Acked-by: Hannes Frederic Sowa Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/net/6lowpan/core.c b/net/6lowpan/core.c index 801404c..1c7a42b 100644 --- a/net/6lowpan/core.c +++ b/net/6lowpan/core.c @@ -158,8 +158,6 @@ static int __init lowpan_module_init(void) return ret; } - request_module_nowait("ipv6"); - request_module_nowait("nhc_dest"); request_module_nowait("nhc_fragment"); request_module_nowait("nhc_hop"); -- cgit v0.10.2 From 1e82f961ac8e94c50a933e89ee08071fc81a4bbd Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:19 +0200 Subject: ndisc: add __ndisc_opt_addr_space function This patch adds __ndisc_opt_addr_space as low-level function for ndisc_opt_addr_space which doesn't depend on net_device parameter. Cc: David S. Miller Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Acked-by: YOSHIFUJI Hideaki Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/include/net/ndisc.h b/include/net/ndisc.h index 2d8edaa..4cee826 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -127,10 +127,15 @@ static inline int ndisc_addr_option_pad(unsigned short type) } } +static inline int __ndisc_opt_addr_space(unsigned char addr_len, int pad) +{ + return NDISC_OPT_SPACE(addr_len + pad); +} + static inline int ndisc_opt_addr_space(struct net_device *dev) { - return NDISC_OPT_SPACE(dev->addr_len + - ndisc_addr_option_pad(dev->type)); + return __ndisc_opt_addr_space(dev->addr_len, + ndisc_addr_option_pad(dev->type)); } static inline u8 *ndisc_opt_addr_data(struct nd_opt_hdr *p, -- cgit v0.10.2 From 4f36ce84c54c971cd67c882035d9bde7b1a2ee53 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:20 +0200 Subject: ndisc: add __ndisc_opt_addr_data function This patch adds __ndisc_opt_addr_data as low-level function for ndisc_opt_addr_data which doesn't depend on net_device parameter. Cc: David S. Miller Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Acked-by: YOSHIFUJI Hideaki Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/include/net/ndisc.h b/include/net/ndisc.h index 4cee826..c8962ad 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -138,17 +138,23 @@ static inline int ndisc_opt_addr_space(struct net_device *dev) ndisc_addr_option_pad(dev->type)); } -static inline u8 *ndisc_opt_addr_data(struct nd_opt_hdr *p, - struct net_device *dev) +static inline u8 *__ndisc_opt_addr_data(struct nd_opt_hdr *p, + unsigned char addr_len, int prepad) { u8 *lladdr = (u8 *)(p + 1); int lladdrlen = p->nd_opt_len << 3; - int prepad = ndisc_addr_option_pad(dev->type); - if (lladdrlen != ndisc_opt_addr_space(dev)) + if (lladdrlen != __ndisc_opt_addr_space(addr_len, prepad)) return NULL; return lladdr + prepad; } +static inline u8 *ndisc_opt_addr_data(struct nd_opt_hdr *p, + struct net_device *dev) +{ + return __ndisc_opt_addr_data(p, dev->addr_len, + ndisc_addr_option_pad(dev->type)); +} + static inline u32 ndisc_hashfn(const void *pkey, const struct net_device *dev, __u32 *hash_rnd) { const u32 *p32 = pkey; -- cgit v0.10.2 From 8ec5da41502843947af0b09e795b19fc7a83edd8 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:21 +0200 Subject: ndisc: add __ndisc_fill_addr_option function This patch adds __ndisc_fill_addr_option as low-level function for ndisc_fill_addr_option which doesn't depend on net_device parameter. Cc: David S. Miller Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Acked-by: YOSHIFUJI Hideaki Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index c245895..a7b9468 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -150,11 +150,10 @@ struct neigh_table nd_tbl = { }; EXPORT_SYMBOL_GPL(nd_tbl); -static void ndisc_fill_addr_option(struct sk_buff *skb, int type, void *data) +static void __ndisc_fill_addr_option(struct sk_buff *skb, int type, void *data, + int data_len, int pad) { - int pad = ndisc_addr_option_pad(skb->dev->type); - int data_len = skb->dev->addr_len; - int space = ndisc_opt_addr_space(skb->dev); + int space = __ndisc_opt_addr_space(data_len, pad); u8 *opt = skb_put(skb, space); opt[0] = type; @@ -172,6 +171,13 @@ static void ndisc_fill_addr_option(struct sk_buff *skb, int type, void *data) memset(opt, 0, space); } +static inline void ndisc_fill_addr_option(struct sk_buff *skb, int type, + void *data) +{ + __ndisc_fill_addr_option(skb, type, data, skb->dev->addr_len, + ndisc_addr_option_pad(skb->dev->type)); +} + static struct nd_opt_hdr *ndisc_next_option(struct nd_opt_hdr *cur, struct nd_opt_hdr *end) { -- cgit v0.10.2 From 4f672235cb11c49d4be7ac7d505c65e3bd367322 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:22 +0200 Subject: addrconf: put prefix address add in an own function This patch moves the functionality to add a RA PIO prefix generated address in an own function. This move prepares to add a hook for adding a second address for a second link-layer address. E.g. short address for 802.15.4 6LoWPAN. Cc: David S. Miller Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Reviewed-by: Stefan Schmidt Acked-by: YOSHIFUJI Hideaki Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 1ce4048..44c1cef 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2333,12 +2333,110 @@ static bool is_addr_mode_generate_stable(struct inet6_dev *idev) idev->addr_gen_mode == IN6_ADDR_GEN_MODE_RANDOM; } +static int addrconf_prefix_rcv_add_addr(struct net *net, + struct net_device *dev, + const struct prefix_info *pinfo, + struct inet6_dev *in6_dev, + const struct in6_addr *addr, + int addr_type, u32 addr_flags, + bool sllao, bool tokenized, + __u32 valid_lft, u32 prefered_lft) +{ + struct inet6_ifaddr *ifp = ipv6_get_ifaddr(net, addr, dev, 1); + int create = 0, update_lft = 0; + + if (!ifp && valid_lft) { + int max_addresses = in6_dev->cnf.max_addresses; + +#ifdef CONFIG_IPV6_OPTIMISTIC_DAD + if (in6_dev->cnf.optimistic_dad && + !net->ipv6.devconf_all->forwarding && sllao) + addr_flags |= IFA_F_OPTIMISTIC; +#endif + + /* Do not allow to create too much of autoconfigured + * addresses; this would be too easy way to crash kernel. + */ + if (!max_addresses || + ipv6_count_addresses(in6_dev) < max_addresses) + ifp = ipv6_add_addr(in6_dev, addr, NULL, + pinfo->prefix_len, + addr_type&IPV6_ADDR_SCOPE_MASK, + addr_flags, valid_lft, + prefered_lft); + + if (IS_ERR_OR_NULL(ifp)) + return -1; + + update_lft = 0; + create = 1; + spin_lock_bh(&ifp->lock); + ifp->flags |= IFA_F_MANAGETEMPADDR; + ifp->cstamp = jiffies; + ifp->tokenized = tokenized; + spin_unlock_bh(&ifp->lock); + addrconf_dad_start(ifp); + } + + if (ifp) { + u32 flags; + unsigned long now; + u32 stored_lft; + + /* update lifetime (RFC2462 5.5.3 e) */ + spin_lock_bh(&ifp->lock); + now = jiffies; + if (ifp->valid_lft > (now - ifp->tstamp) / HZ) + stored_lft = ifp->valid_lft - (now - ifp->tstamp) / HZ; + else + stored_lft = 0; + if (!update_lft && !create && stored_lft) { + const u32 minimum_lft = min_t(u32, + stored_lft, MIN_VALID_LIFETIME); + valid_lft = max(valid_lft, minimum_lft); + + /* RFC4862 Section 5.5.3e: + * "Note that the preferred lifetime of the + * corresponding address is always reset to + * the Preferred Lifetime in the received + * Prefix Information option, regardless of + * whether the valid lifetime is also reset or + * ignored." + * + * So we should always update prefered_lft here. + */ + update_lft = 1; + } + + if (update_lft) { + ifp->valid_lft = valid_lft; + ifp->prefered_lft = prefered_lft; + ifp->tstamp = now; + flags = ifp->flags; + ifp->flags &= ~IFA_F_DEPRECATED; + spin_unlock_bh(&ifp->lock); + + if (!(flags&IFA_F_TENTATIVE)) + ipv6_ifa_notify(0, ifp); + } else + spin_unlock_bh(&ifp->lock); + + manage_tempaddrs(in6_dev, ifp, valid_lft, prefered_lft, + create, now); + + in6_ifa_put(ifp); + addrconf_verify(); + } + + return 0; +} + void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) { struct prefix_info *pinfo; __u32 valid_lft; __u32 prefered_lft; - int addr_type; + int addr_type, err; u32 addr_flags = 0; struct inet6_dev *in6_dev; struct net *net = dev_net(dev); @@ -2432,9 +2530,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) /* Try to figure out our local address for this prefix */ if (pinfo->autoconf && in6_dev->cnf.autoconf) { - struct inet6_ifaddr *ifp; struct in6_addr addr; - int create = 0, update_lft = 0; bool tokenized = false; if (pinfo->prefix_len == 64) { @@ -2453,106 +2549,25 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) goto ok; } else if (ipv6_generate_eui64(addr.s6_addr + 8, dev) && ipv6_inherit_eui64(addr.s6_addr + 8, in6_dev)) { - in6_dev_put(in6_dev); - return; + goto put; } goto ok; } net_dbg_ratelimited("IPv6 addrconf: prefix with wrong length %d\n", pinfo->prefix_len); - in6_dev_put(in6_dev); - return; + goto put; ok: - - ifp = ipv6_get_ifaddr(net, &addr, dev, 1); - - if (!ifp && valid_lft) { - int max_addresses = in6_dev->cnf.max_addresses; - -#ifdef CONFIG_IPV6_OPTIMISTIC_DAD - if (in6_dev->cnf.optimistic_dad && - !net->ipv6.devconf_all->forwarding && sllao) - addr_flags |= IFA_F_OPTIMISTIC; -#endif - - /* Do not allow to create too much of autoconfigured - * addresses; this would be too easy way to crash kernel. - */ - if (!max_addresses || - ipv6_count_addresses(in6_dev) < max_addresses) - ifp = ipv6_add_addr(in6_dev, &addr, NULL, - pinfo->prefix_len, - addr_type&IPV6_ADDR_SCOPE_MASK, - addr_flags, valid_lft, - prefered_lft); - - if (IS_ERR_OR_NULL(ifp)) { - in6_dev_put(in6_dev); - return; - } - - update_lft = 0; - create = 1; - spin_lock_bh(&ifp->lock); - ifp->flags |= IFA_F_MANAGETEMPADDR; - ifp->cstamp = jiffies; - ifp->tokenized = tokenized; - spin_unlock_bh(&ifp->lock); - addrconf_dad_start(ifp); - } - - if (ifp) { - u32 flags; - unsigned long now; - u32 stored_lft; - - /* update lifetime (RFC2462 5.5.3 e) */ - spin_lock_bh(&ifp->lock); - now = jiffies; - if (ifp->valid_lft > (now - ifp->tstamp) / HZ) - stored_lft = ifp->valid_lft - (now - ifp->tstamp) / HZ; - else - stored_lft = 0; - if (!update_lft && !create && stored_lft) { - const u32 minimum_lft = min_t(u32, - stored_lft, MIN_VALID_LIFETIME); - valid_lft = max(valid_lft, minimum_lft); - - /* RFC4862 Section 5.5.3e: - * "Note that the preferred lifetime of the - * corresponding address is always reset to - * the Preferred Lifetime in the received - * Prefix Information option, regardless of - * whether the valid lifetime is also reset or - * ignored." - * - * So we should always update prefered_lft here. - */ - update_lft = 1; - } - - if (update_lft) { - ifp->valid_lft = valid_lft; - ifp->prefered_lft = prefered_lft; - ifp->tstamp = now; - flags = ifp->flags; - ifp->flags &= ~IFA_F_DEPRECATED; - spin_unlock_bh(&ifp->lock); - - if (!(flags&IFA_F_TENTATIVE)) - ipv6_ifa_notify(0, ifp); - } else - spin_unlock_bh(&ifp->lock); - - manage_tempaddrs(in6_dev, ifp, valid_lft, prefered_lft, - create, now); - - in6_ifa_put(ifp); - addrconf_verify(); - } + err = addrconf_prefix_rcv_add_addr(net, dev, pinfo, in6_dev, + &addr, addr_type, + addr_flags, sllao, + tokenized, valid_lft, + prefered_lft); + if (err) + goto put; } inet6_prefix_notify(RTM_NEWPREFIX, in6_dev, pinfo); +put: in6_dev_put(in6_dev); } -- cgit v0.10.2 From f997c55c1dc8841b3ee4df0493d0ac7966d42165 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:23 +0200 Subject: ipv6: introduce neighbour discovery ops This patch introduces neighbour discovery ops callback structure. The idea is to separate the handling for 6LoWPAN into the 6lowpan module. These callback offers 6lowpan different handling, such as 802.15.4 short address handling or RFC6775 (Neighbor Discovery Optimization for IPv6 over 6LoWPANs). Cc: David S. Miller Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Acked-by: YOSHIFUJI Hideaki Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 36e43bd..890158e 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1456,6 +1456,8 @@ enum netdev_priv_flags { * @netdev_ops: Includes several pointers to callbacks, * if one wants to override the ndo_*() functions * @ethtool_ops: Management operations + * @ndisc_ops: Includes callbacks for different IPv6 neighbour + * discovery handling. Necessary for e.g. 6LoWPAN. * @header_ops: Includes callbacks for creating,parsing,caching,etc * of Layer 2 headers. * @@ -1672,6 +1674,9 @@ struct net_device { #ifdef CONFIG_NET_L3_MASTER_DEV const struct l3mdev_ops *l3mdev_ops; #endif +#if IS_ENABLED(CONFIG_IPV6) + const struct ndisc_ops *ndisc_ops; +#endif const struct header_ops *header_ops; diff --git a/include/net/ndisc.h b/include/net/ndisc.h index c8962ad..a5e2767 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -58,6 +58,7 @@ struct inet6_dev; struct net_device; struct net_proto_family; struct sk_buff; +struct prefix_info; extern struct neigh_table nd_tbl; @@ -110,9 +111,182 @@ struct ndisc_options { #define NDISC_OPT_SPACE(len) (((len)+2+7)&~7) -struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, +struct ndisc_options *ndisc_parse_options(const struct net_device *dev, + u8 *opt, int opt_len, struct ndisc_options *ndopts); +#define NDISC_OPS_REDIRECT_DATA_SPACE 2 + +/* + * This structure defines the hooks for IPv6 neighbour discovery. + * The following hooks can be defined; unless noted otherwise, they are + * optional and can be filled with a null pointer. + * + * int (*is_useropt)(u8 nd_opt_type): + * This function is called when IPv6 decide RA userspace options. if + * this function returns 1 then the option given by nd_opt_type will + * be handled as userspace option additional to the IPv6 options. + * + * int (*parse_options)(const struct net_device *dev, + * struct nd_opt_hdr *nd_opt, + * struct ndisc_options *ndopts): + * This function is called while parsing ndisc ops and put each position + * as pointer into ndopts. If this function return unequal 0, then this + * function took care about the ndisc option, if 0 then the IPv6 ndisc + * option parser will take care about that option. + * + * void (*update)(const struct net_device *dev, struct neighbour *n, + * u32 flags, u8 icmp6_type, + * const struct ndisc_options *ndopts): + * This function is called when IPv6 ndisc updates the neighbour cache + * entry. Additional options which can be updated may be previously + * parsed by parse_opts callback and accessible over ndopts parameter. + * + * int (*opt_addr_space)(const struct net_device *dev, u8 icmp6_type, + * struct neighbour *neigh, u8 *ha_buf, + * u8 **ha): + * This function is called when the necessary option space will be + * calculated before allocating a skb. The parameters neigh, ha_buf + * abd ha are available on NDISC_REDIRECT messages only. + * + * void (*fill_addr_option)(const struct net_device *dev, + * struct sk_buff *skb, u8 icmp6_type, + * const u8 *ha): + * This function is called when the skb will finally fill the option + * fields inside skb. NOTE: this callback should fill the option + * fields to the skb which are previously indicated by opt_space + * parameter. That means the decision to add such option should + * not lost between these two callbacks, e.g. protected by interface + * up state. + * + * void (*prefix_rcv_add_addr)(struct net *net, struct net_device *dev, + * const struct prefix_info *pinfo, + * struct inet6_dev *in6_dev, + * struct in6_addr *addr, + * int addr_type, u32 addr_flags, + * bool sllao, bool tokenized, + * __u32 valid_lft, u32 prefered_lft, + * bool dev_addr_generated): + * This function is called when a RA messages is received with valid + * PIO option fields and an IPv6 address will be added to the interface + * for autoconfiguration. The parameter dev_addr_generated reports about + * if the address was based on dev->dev_addr or not. This can be used + * to add a second address if link-layer operates with two link layer + * addresses. E.g. 802.15.4 6LoWPAN. + */ +struct ndisc_ops { + int (*is_useropt)(u8 nd_opt_type); + int (*parse_options)(const struct net_device *dev, + struct nd_opt_hdr *nd_opt, + struct ndisc_options *ndopts); + void (*update)(const struct net_device *dev, struct neighbour *n, + u32 flags, u8 icmp6_type, + const struct ndisc_options *ndopts); + int (*opt_addr_space)(const struct net_device *dev, u8 icmp6_type, + struct neighbour *neigh, u8 *ha_buf, + u8 **ha); + void (*fill_addr_option)(const struct net_device *dev, + struct sk_buff *skb, u8 icmp6_type, + const u8 *ha); + void (*prefix_rcv_add_addr)(struct net *net, struct net_device *dev, + const struct prefix_info *pinfo, + struct inet6_dev *in6_dev, + struct in6_addr *addr, + int addr_type, u32 addr_flags, + bool sllao, bool tokenized, + __u32 valid_lft, u32 prefered_lft, + bool dev_addr_generated); +}; + +#if IS_ENABLED(CONFIG_IPV6) +static inline int ndisc_ops_is_useropt(const struct net_device *dev, + u8 nd_opt_type) +{ + if (dev->ndisc_ops && dev->ndisc_ops->is_useropt) + return dev->ndisc_ops->is_useropt(nd_opt_type); + else + return 0; +} + +static inline int ndisc_ops_parse_options(const struct net_device *dev, + struct nd_opt_hdr *nd_opt, + struct ndisc_options *ndopts) +{ + if (dev->ndisc_ops && dev->ndisc_ops->parse_options) + return dev->ndisc_ops->parse_options(dev, nd_opt, ndopts); + else + return 0; +} + +static inline void ndisc_ops_update(const struct net_device *dev, + struct neighbour *n, u32 flags, + u8 icmp6_type, + const struct ndisc_options *ndopts) +{ + if (dev->ndisc_ops && dev->ndisc_ops->update) + dev->ndisc_ops->update(dev, n, flags, icmp6_type, ndopts); +} + +static inline int ndisc_ops_opt_addr_space(const struct net_device *dev, + u8 icmp6_type) +{ + if (dev->ndisc_ops && dev->ndisc_ops->opt_addr_space && + icmp6_type != NDISC_REDIRECT) + return dev->ndisc_ops->opt_addr_space(dev, icmp6_type, NULL, + NULL, NULL); + else + return 0; +} + +static inline int ndisc_ops_redirect_opt_addr_space(const struct net_device *dev, + struct neighbour *neigh, + u8 *ha_buf, u8 **ha) +{ + if (dev->ndisc_ops && dev->ndisc_ops->opt_addr_space) + return dev->ndisc_ops->opt_addr_space(dev, NDISC_REDIRECT, + neigh, ha_buf, ha); + else + return 0; +} + +static inline void ndisc_ops_fill_addr_option(const struct net_device *dev, + struct sk_buff *skb, + u8 icmp6_type) +{ + if (dev->ndisc_ops && dev->ndisc_ops->fill_addr_option && + icmp6_type != NDISC_REDIRECT) + dev->ndisc_ops->fill_addr_option(dev, skb, icmp6_type, NULL); +} + +static inline void ndisc_ops_fill_redirect_addr_option(const struct net_device *dev, + struct sk_buff *skb, + const u8 *ha) +{ + if (dev->ndisc_ops && dev->ndisc_ops->fill_addr_option) + dev->ndisc_ops->fill_addr_option(dev, skb, NDISC_REDIRECT, ha); +} + +static inline void ndisc_ops_prefix_rcv_add_addr(struct net *net, + struct net_device *dev, + const struct prefix_info *pinfo, + struct inet6_dev *in6_dev, + struct in6_addr *addr, + int addr_type, u32 addr_flags, + bool sllao, bool tokenized, + __u32 valid_lft, + u32 prefered_lft, + bool dev_addr_generated) +{ + if (dev->ndisc_ops && dev->ndisc_ops->prefix_rcv_add_addr) + dev->ndisc_ops->prefix_rcv_add_addr(net, dev, pinfo, in6_dev, + addr, addr_type, + addr_flags, sllao, + tokenized, valid_lft, + prefered_lft, + dev_addr_generated); +} +#endif + /* * Return the padding between the option length and the start of the * link addr. Currently only IP-over-InfiniBand needs this, although @@ -132,11 +306,25 @@ static inline int __ndisc_opt_addr_space(unsigned char addr_len, int pad) return NDISC_OPT_SPACE(addr_len + pad); } -static inline int ndisc_opt_addr_space(struct net_device *dev) +#if IS_ENABLED(CONFIG_IPV6) +static inline int ndisc_opt_addr_space(struct net_device *dev, u8 icmp6_type) +{ + return __ndisc_opt_addr_space(dev->addr_len, + ndisc_addr_option_pad(dev->type)) + + ndisc_ops_opt_addr_space(dev, icmp6_type); +} + +static inline int ndisc_redirect_opt_addr_space(struct net_device *dev, + struct neighbour *neigh, + u8 *ops_data_buf, + u8 **ops_data) { return __ndisc_opt_addr_space(dev->addr_len, - ndisc_addr_option_pad(dev->type)); + ndisc_addr_option_pad(dev->type)) + + ndisc_ops_redirect_opt_addr_space(dev, neigh, ops_data_buf, + ops_data); } +#endif static inline u8 *__ndisc_opt_addr_data(struct nd_opt_hdr *p, unsigned char addr_len, int prepad) @@ -205,6 +393,9 @@ void ndisc_send_redirect(struct sk_buff *skb, const struct in6_addr *target); int ndisc_mc_map(const struct in6_addr *addr, char *buf, struct net_device *dev, int dir); +void ndisc_update(const struct net_device *dev, struct neighbour *neigh, + const u8 *lladdr, u8 new, u32 flags, u8 icmp6_type, + struct ndisc_options *ndopts); /* * IGMP diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 44c1cef..b6e9bdc 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2531,7 +2531,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) if (pinfo->autoconf && in6_dev->cnf.autoconf) { struct in6_addr addr; - bool tokenized = false; + bool tokenized = false, dev_addr_generated = false; if (pinfo->prefix_len == 64) { memcpy(&addr, &pinfo->prefix, 8); @@ -2550,6 +2550,8 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) } else if (ipv6_generate_eui64(addr.s6_addr + 8, dev) && ipv6_inherit_eui64(addr.s6_addr + 8, in6_dev)) { goto put; + } else { + dev_addr_generated = true; } goto ok; } @@ -2565,6 +2567,15 @@ ok: prefered_lft); if (err) goto put; + + /* Ignore error case here because previous prefix add addr was + * successful which will be notified. + */ + ndisc_ops_prefix_rcv_add_addr(net, dev, pinfo, in6_dev, &addr, + addr_type, addr_flags, sllao, + tokenized, valid_lft, + prefered_lft, + dev_addr_generated); } inet6_prefix_notify(RTM_NEWPREFIX, in6_dev, pinfo); put: diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index a7b9468..2f4afd1 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -172,10 +172,19 @@ static void __ndisc_fill_addr_option(struct sk_buff *skb, int type, void *data, } static inline void ndisc_fill_addr_option(struct sk_buff *skb, int type, - void *data) + void *data, u8 icmp6_type) { __ndisc_fill_addr_option(skb, type, data, skb->dev->addr_len, ndisc_addr_option_pad(skb->dev->type)); + ndisc_ops_fill_addr_option(skb->dev, skb, icmp6_type); +} + +static inline void ndisc_fill_redirect_addr_option(struct sk_buff *skb, + void *ha, + const u8 *ops_data) +{ + ndisc_fill_addr_option(skb, ND_OPT_TARGET_LL_ADDR, ha, NDISC_REDIRECT); + ndisc_ops_fill_redirect_addr_option(skb->dev, skb, ops_data); } static struct nd_opt_hdr *ndisc_next_option(struct nd_opt_hdr *cur, @@ -191,24 +200,28 @@ static struct nd_opt_hdr *ndisc_next_option(struct nd_opt_hdr *cur, return cur <= end && cur->nd_opt_type == type ? cur : NULL; } -static inline int ndisc_is_useropt(struct nd_opt_hdr *opt) +static inline int ndisc_is_useropt(const struct net_device *dev, + struct nd_opt_hdr *opt) { return opt->nd_opt_type == ND_OPT_RDNSS || - opt->nd_opt_type == ND_OPT_DNSSL; + opt->nd_opt_type == ND_OPT_DNSSL || + ndisc_ops_is_useropt(dev, opt->nd_opt_type); } -static struct nd_opt_hdr *ndisc_next_useropt(struct nd_opt_hdr *cur, +static struct nd_opt_hdr *ndisc_next_useropt(const struct net_device *dev, + struct nd_opt_hdr *cur, struct nd_opt_hdr *end) { if (!cur || !end || cur >= end) return NULL; do { cur = ((void *)cur) + (cur->nd_opt_len << 3); - } while (cur < end && !ndisc_is_useropt(cur)); - return cur <= end && ndisc_is_useropt(cur) ? cur : NULL; + } while (cur < end && !ndisc_is_useropt(dev, cur)); + return cur <= end && ndisc_is_useropt(dev, cur) ? cur : NULL; } -struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, +struct ndisc_options *ndisc_parse_options(const struct net_device *dev, + u8 *opt, int opt_len, struct ndisc_options *ndopts) { struct nd_opt_hdr *nd_opt = (struct nd_opt_hdr *)opt; @@ -223,6 +236,8 @@ struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, l = nd_opt->nd_opt_len << 3; if (opt_len < l || l == 0) return NULL; + if (ndisc_ops_parse_options(dev, nd_opt, ndopts)) + goto next_opt; switch (nd_opt->nd_opt_type) { case ND_OPT_SOURCE_LL_ADDR: case ND_OPT_TARGET_LL_ADDR: @@ -249,7 +264,7 @@ struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, break; #endif default: - if (ndisc_is_useropt(nd_opt)) { + if (ndisc_is_useropt(dev, nd_opt)) { ndopts->nd_useropts_end = nd_opt; if (!ndopts->nd_useropts) ndopts->nd_useropts = nd_opt; @@ -266,6 +281,7 @@ struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, nd_opt->nd_opt_len); } } +next_opt: opt_len -= l; nd_opt = ((void *)nd_opt) + l; } @@ -515,7 +531,8 @@ void ndisc_send_na(struct net_device *dev, const struct in6_addr *daddr, if (!dev->addr_len) inc_opt = 0; if (inc_opt) - optlen += ndisc_opt_addr_space(dev); + optlen += ndisc_opt_addr_space(dev, + NDISC_NEIGHBOUR_ADVERTISEMENT); skb = ndisc_alloc_skb(dev, sizeof(*msg) + optlen); if (!skb) @@ -534,8 +551,8 @@ void ndisc_send_na(struct net_device *dev, const struct in6_addr *daddr, if (inc_opt) ndisc_fill_addr_option(skb, ND_OPT_TARGET_LL_ADDR, - dev->dev_addr); - + dev->dev_addr, + NDISC_NEIGHBOUR_ADVERTISEMENT); ndisc_send_skb(skb, daddr, src_addr); } @@ -580,7 +597,8 @@ void ndisc_send_ns(struct net_device *dev, const struct in6_addr *solicit, if (ipv6_addr_any(saddr)) inc_opt = false; if (inc_opt) - optlen += ndisc_opt_addr_space(dev); + optlen += ndisc_opt_addr_space(dev, + NDISC_NEIGHBOUR_SOLICITATION); skb = ndisc_alloc_skb(dev, sizeof(*msg) + optlen); if (!skb) @@ -596,7 +614,8 @@ void ndisc_send_ns(struct net_device *dev, const struct in6_addr *solicit, if (inc_opt) ndisc_fill_addr_option(skb, ND_OPT_SOURCE_LL_ADDR, - dev->dev_addr); + dev->dev_addr, + NDISC_NEIGHBOUR_SOLICITATION); ndisc_send_skb(skb, daddr, saddr); } @@ -632,7 +651,7 @@ void ndisc_send_rs(struct net_device *dev, const struct in6_addr *saddr, } #endif if (send_sllao) - optlen += ndisc_opt_addr_space(dev); + optlen += ndisc_opt_addr_space(dev, NDISC_ROUTER_SOLICITATION); skb = ndisc_alloc_skb(dev, sizeof(*msg) + optlen); if (!skb) @@ -647,7 +666,8 @@ void ndisc_send_rs(struct net_device *dev, const struct in6_addr *saddr, if (send_sllao) ndisc_fill_addr_option(skb, ND_OPT_SOURCE_LL_ADDR, - dev->dev_addr); + dev->dev_addr, + NDISC_ROUTER_SOLICITATION); ndisc_send_skb(skb, daddr, saddr); } @@ -708,6 +728,15 @@ static int pndisc_is_router(const void *pkey, return ret; } +void ndisc_update(const struct net_device *dev, struct neighbour *neigh, + const u8 *lladdr, u8 new, u32 flags, u8 icmp6_type, + struct ndisc_options *ndopts) +{ + neigh_update(neigh, lladdr, new, flags); + /* report ndisc ops about neighbour update */ + ndisc_ops_update(dev, neigh, flags, icmp6_type, ndopts); +} + static void ndisc_recv_ns(struct sk_buff *skb) { struct nd_msg *msg = (struct nd_msg *)skb_transport_header(skb); @@ -744,7 +773,7 @@ static void ndisc_recv_ns(struct sk_buff *skb) return; } - if (!ndisc_parse_options(msg->opt, ndoptlen, &ndopts)) { + if (!ndisc_parse_options(dev, msg->opt, ndoptlen, &ndopts)) { ND_PRINTK(2, warn, "NS: invalid ND options\n"); return; } @@ -862,9 +891,10 @@ have_ifp: neigh = __neigh_lookup(&nd_tbl, saddr, dev, !inc || lladdr || !dev->addr_len); if (neigh) - neigh_update(neigh, lladdr, NUD_STALE, + ndisc_update(dev, neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_WEAK_OVERRIDE| - NEIGH_UPDATE_F_OVERRIDE); + NEIGH_UPDATE_F_OVERRIDE, + NDISC_NEIGHBOUR_SOLICITATION, &ndopts); if (neigh || !dev->header_ops) { ndisc_send_na(dev, saddr, &msg->target, !!is_router, true, (ifp != NULL && inc), inc); @@ -917,7 +947,7 @@ static void ndisc_recv_na(struct sk_buff *skb) idev->cnf.drop_unsolicited_na) return; - if (!ndisc_parse_options(msg->opt, ndoptlen, &ndopts)) { + if (!ndisc_parse_options(dev, msg->opt, ndoptlen, &ndopts)) { ND_PRINTK(2, warn, "NS: invalid ND option\n"); return; } @@ -973,12 +1003,13 @@ static void ndisc_recv_na(struct sk_buff *skb) goto out; } - neigh_update(neigh, lladdr, + ndisc_update(dev, neigh, lladdr, msg->icmph.icmp6_solicited ? NUD_REACHABLE : NUD_STALE, NEIGH_UPDATE_F_WEAK_OVERRIDE| (msg->icmph.icmp6_override ? NEIGH_UPDATE_F_OVERRIDE : 0)| NEIGH_UPDATE_F_OVERRIDE_ISROUTER| - (msg->icmph.icmp6_router ? NEIGH_UPDATE_F_ISROUTER : 0)); + (msg->icmph.icmp6_router ? NEIGH_UPDATE_F_ISROUTER : 0), + NDISC_NEIGHBOUR_ADVERTISEMENT, &ndopts); if ((old_flags & ~neigh->flags) & NTF_ROUTER) { /* @@ -1023,7 +1054,7 @@ static void ndisc_recv_rs(struct sk_buff *skb) goto out; /* Parse ND options */ - if (!ndisc_parse_options(rs_msg->opt, ndoptlen, &ndopts)) { + if (!ndisc_parse_options(skb->dev, rs_msg->opt, ndoptlen, &ndopts)) { ND_PRINTK(2, notice, "NS: invalid ND option, ignored\n"); goto out; } @@ -1037,10 +1068,11 @@ static void ndisc_recv_rs(struct sk_buff *skb) neigh = __neigh_lookup(&nd_tbl, saddr, skb->dev, 1); if (neigh) { - neigh_update(neigh, lladdr, NUD_STALE, + ndisc_update(skb->dev, neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_WEAK_OVERRIDE| NEIGH_UPDATE_F_OVERRIDE| - NEIGH_UPDATE_F_OVERRIDE_ISROUTER); + NEIGH_UPDATE_F_OVERRIDE_ISROUTER, + NDISC_ROUTER_SOLICITATION, &ndopts); neigh_release(neigh); } out: @@ -1141,7 +1173,7 @@ static void ndisc_router_discovery(struct sk_buff *skb) return; } - if (!ndisc_parse_options(opt, optlen, &ndopts)) { + if (!ndisc_parse_options(skb->dev, opt, optlen, &ndopts)) { ND_PRINTK(2, warn, "RA: invalid ND options\n"); return; } @@ -1335,11 +1367,12 @@ skip_linkparms: goto out; } } - neigh_update(neigh, lladdr, NUD_STALE, + ndisc_update(skb->dev, neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_WEAK_OVERRIDE| NEIGH_UPDATE_F_OVERRIDE| NEIGH_UPDATE_F_OVERRIDE_ISROUTER| - NEIGH_UPDATE_F_ISROUTER); + NEIGH_UPDATE_F_ISROUTER, + NDISC_ROUTER_ADVERTISEMENT, &ndopts); } if (!ipv6_accept_ra(in6_dev)) { @@ -1427,7 +1460,8 @@ skip_routeinfo: struct nd_opt_hdr *p; for (p = ndopts.nd_useropts; p; - p = ndisc_next_useropt(p, ndopts.nd_useropts_end)) { + p = ndisc_next_useropt(skb->dev, p, + ndopts.nd_useropts_end)) { ndisc_ra_useropt(skb, p); } } @@ -1465,7 +1499,7 @@ static void ndisc_redirect_rcv(struct sk_buff *skb) return; } - if (!ndisc_parse_options(msg->opt, ndoptlen, &ndopts)) + if (!ndisc_parse_options(skb->dev, msg->opt, ndoptlen, &ndopts)) return; if (!ndopts.nd_opts_rh) { @@ -1510,7 +1544,8 @@ void ndisc_send_redirect(struct sk_buff *skb, const struct in6_addr *target) struct dst_entry *dst; struct flowi6 fl6; int rd_len; - u8 ha_buf[MAX_ADDR_LEN], *ha = NULL; + u8 ha_buf[MAX_ADDR_LEN], *ha = NULL, + ops_data_buf[NDISC_OPS_REDIRECT_DATA_SPACE], *ops_data = NULL; int oif = l3mdev_fib_oif(dev); bool ret; @@ -1569,7 +1604,9 @@ void ndisc_send_redirect(struct sk_buff *skb, const struct in6_addr *target) memcpy(ha_buf, neigh->ha, dev->addr_len); read_unlock_bh(&neigh->lock); ha = ha_buf; - optlen += ndisc_opt_addr_space(dev); + optlen += ndisc_redirect_opt_addr_space(dev, neigh, + ops_data_buf, + &ops_data); } else read_unlock_bh(&neigh->lock); @@ -1600,7 +1637,7 @@ void ndisc_send_redirect(struct sk_buff *skb, const struct in6_addr *target) */ if (ha) - ndisc_fill_addr_option(buff, ND_OPT_TARGET_LL_ADDR, ha); + ndisc_fill_redirect_addr_option(buff, ha, ops_data); /* * build redirect option and copy skb over to the new packet. diff --git a/net/ipv6/route.c b/net/ipv6/route.c index d51a1a4..9e15167 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2201,7 +2201,7 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu * first-hop router for the specified ICMP Destination Address. */ - if (!ndisc_parse_options(msg->opt, optlen, &ndopts)) { + if (!ndisc_parse_options(skb->dev, msg->opt, optlen, &ndopts)) { net_dbg_ratelimited("rt6_redirect: invalid ND options\n"); return; } @@ -2236,12 +2236,12 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu * We have finally decided to accept it. */ - neigh_update(neigh, lladdr, NUD_STALE, + ndisc_update(skb->dev, neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_WEAK_OVERRIDE| NEIGH_UPDATE_F_OVERRIDE| (on_link ? 0 : (NEIGH_UPDATE_F_OVERRIDE_ISROUTER| - NEIGH_UPDATE_F_ISROUTER)) - ); + NEIGH_UPDATE_F_ISROUTER)), + NDISC_REDIRECT, &ndopts); nrt = ip6_rt_cache_alloc(rt, &msg->dest, NULL); if (!nrt) -- cgit v0.10.2 From cc84b3c6b48ae81748c5e25d3558872385196162 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:24 +0200 Subject: ipv6: export several functions This patch exports some neighbour discovery functions which can be used by 6lowpan neighbour discovery ops functionality then. Cc: David S. Miller Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Acked-by: YOSHIFUJI Hideaki Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/include/net/addrconf.h b/include/net/addrconf.h index b1774eb..9826d3a 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -97,6 +97,13 @@ void addrconf_leave_solict(struct inet6_dev *idev, const struct in6_addr *addr); void addrconf_add_linklocal(struct inet6_dev *idev, const struct in6_addr *addr, u32 flags); +int addrconf_prefix_rcv_add_addr(struct net *net, struct net_device *dev, + const struct prefix_info *pinfo, + struct inet6_dev *in6_dev, + const struct in6_addr *addr, int addr_type, + u32 addr_flags, bool sllao, bool tokenized, + __u32 valid_lft, u32 prefered_lft); + static inline int addrconf_ifid_eui48(u8 *eui, struct net_device *dev) { if (dev->addr_len != ETH_ALEN) diff --git a/include/net/ndisc.h b/include/net/ndisc.h index a5e2767..3f0f41d 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -53,6 +53,15 @@ enum { #include +/* Set to 3 to get tracing... */ +#define ND_DEBUG 1 + +#define ND_PRINTK(val, level, fmt, ...) \ +do { \ + if (val <= ND_DEBUG) \ + net_##level##_ratelimited(fmt, ##__VA_ARGS__); \ +} while (0) + struct ctl_table; struct inet6_dev; struct net_device; @@ -115,6 +124,9 @@ struct ndisc_options *ndisc_parse_options(const struct net_device *dev, u8 *opt, int opt_len, struct ndisc_options *ndopts); +void __ndisc_fill_addr_option(struct sk_buff *skb, int type, void *data, + int data_len, int pad); + #define NDISC_OPS_REDIRECT_DATA_SPACE 2 /* diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index b6e9bdc..6c8fc3f 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2333,14 +2333,12 @@ static bool is_addr_mode_generate_stable(struct inet6_dev *idev) idev->addr_gen_mode == IN6_ADDR_GEN_MODE_RANDOM; } -static int addrconf_prefix_rcv_add_addr(struct net *net, - struct net_device *dev, - const struct prefix_info *pinfo, - struct inet6_dev *in6_dev, - const struct in6_addr *addr, - int addr_type, u32 addr_flags, - bool sllao, bool tokenized, - __u32 valid_lft, u32 prefered_lft) +int addrconf_prefix_rcv_add_addr(struct net *net, struct net_device *dev, + const struct prefix_info *pinfo, + struct inet6_dev *in6_dev, + const struct in6_addr *addr, int addr_type, + u32 addr_flags, bool sllao, bool tokenized, + __u32 valid_lft, u32 prefered_lft) { struct inet6_ifaddr *ifp = ipv6_get_ifaddr(net, addr, dev, 1); int create = 0, update_lft = 0; @@ -2430,6 +2428,7 @@ static int addrconf_prefix_rcv_add_addr(struct net *net, return 0; } +EXPORT_SYMBOL_GPL(addrconf_prefix_rcv_add_addr); void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) { diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 2f4afd1..fe65cdc 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -73,15 +73,6 @@ #include #include -/* Set to 3 to get tracing... */ -#define ND_DEBUG 1 - -#define ND_PRINTK(val, level, fmt, ...) \ -do { \ - if (val <= ND_DEBUG) \ - net_##level##_ratelimited(fmt, ##__VA_ARGS__); \ -} while (0) - static u32 ndisc_hash(const void *pkey, const struct net_device *dev, __u32 *hash_rnd); @@ -150,8 +141,8 @@ struct neigh_table nd_tbl = { }; EXPORT_SYMBOL_GPL(nd_tbl); -static void __ndisc_fill_addr_option(struct sk_buff *skb, int type, void *data, - int data_len, int pad) +void __ndisc_fill_addr_option(struct sk_buff *skb, int type, void *data, + int data_len, int pad) { int space = __ndisc_opt_addr_space(data_len, pad); u8 *opt = skb_put(skb, space); @@ -170,6 +161,7 @@ static void __ndisc_fill_addr_option(struct sk_buff *skb, int type, void *data, if (space > 0) memset(opt, 0, space); } +EXPORT_SYMBOL_GPL(__ndisc_fill_addr_option); static inline void ndisc_fill_addr_option(struct sk_buff *skb, int type, void *data, u8 icmp6_type) -- cgit v0.10.2 From bbe5f5cefe2818eda0392c178de141ffc5734d90 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:25 +0200 Subject: 6lowpan: introduce 6lowpan-nd This patch introduce different 6lowpan handling for receive and transmit NS/NA messages for the ipv6 neighbour discovery. The first use-case is for supporting 802.15.4 short addresses inside the option fields and handling for RFC6775 6CO option field as userspace option. Cc: David S. Miller Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Reviewed-by: Stefan Schmidt Acked-by: YOSHIFUJI Hideaki Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/include/net/ndisc.h b/include/net/ndisc.h index 3f0f41d..be1fe228 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -35,6 +35,7 @@ enum { ND_OPT_ROUTE_INFO = 24, /* RFC4191 */ ND_OPT_RDNSS = 25, /* RFC5006 */ ND_OPT_DNSSL = 31, /* RFC6106 */ + ND_OPT_6CO = 34, /* RFC6775 */ __ND_OPT_MAX }; @@ -109,14 +110,19 @@ struct ndisc_options { #endif struct nd_opt_hdr *nd_useropts; struct nd_opt_hdr *nd_useropts_end; +#if IS_ENABLED(CONFIG_IEEE802154_6LOWPAN) + struct nd_opt_hdr *nd_802154_opt_array[ND_OPT_TARGET_LL_ADDR + 1]; +#endif }; -#define nd_opts_src_lladdr nd_opt_array[ND_OPT_SOURCE_LL_ADDR] -#define nd_opts_tgt_lladdr nd_opt_array[ND_OPT_TARGET_LL_ADDR] -#define nd_opts_pi nd_opt_array[ND_OPT_PREFIX_INFO] -#define nd_opts_pi_end nd_opt_array[__ND_OPT_PREFIX_INFO_END] -#define nd_opts_rh nd_opt_array[ND_OPT_REDIRECT_HDR] -#define nd_opts_mtu nd_opt_array[ND_OPT_MTU] +#define nd_opts_src_lladdr nd_opt_array[ND_OPT_SOURCE_LL_ADDR] +#define nd_opts_tgt_lladdr nd_opt_array[ND_OPT_TARGET_LL_ADDR] +#define nd_opts_pi nd_opt_array[ND_OPT_PREFIX_INFO] +#define nd_opts_pi_end nd_opt_array[__ND_OPT_PREFIX_INFO_END] +#define nd_opts_rh nd_opt_array[ND_OPT_REDIRECT_HDR] +#define nd_opts_mtu nd_opt_array[ND_OPT_MTU] +#define nd_802154_opts_src_lladdr nd_802154_opt_array[ND_OPT_SOURCE_LL_ADDR] +#define nd_802154_opts_tgt_lladdr nd_802154_opt_array[ND_OPT_TARGET_LL_ADDR] #define NDISC_OPT_SPACE(len) (((len)+2+7)&~7) diff --git a/net/6lowpan/6lowpan_i.h b/net/6lowpan/6lowpan_i.h index 97ecc27..a67caee 100644 --- a/net/6lowpan/6lowpan_i.h +++ b/net/6lowpan/6lowpan_i.h @@ -12,6 +12,10 @@ static inline bool lowpan_is_ll(const struct net_device *dev, return lowpan_dev(dev)->lltype == lltype; } +extern const struct ndisc_ops lowpan_ndisc_ops; + +int addrconf_ifid_802154_6lowpan(u8 *eui, struct net_device *dev); + #ifdef CONFIG_6LOWPAN_DEBUGFS int lowpan_dev_debugfs_init(struct net_device *dev); void lowpan_dev_debugfs_exit(struct net_device *dev); diff --git a/net/6lowpan/Makefile b/net/6lowpan/Makefile index e44f3bf..12d131a 100644 --- a/net/6lowpan/Makefile +++ b/net/6lowpan/Makefile @@ -1,6 +1,6 @@ obj-$(CONFIG_6LOWPAN) += 6lowpan.o -6lowpan-y := core.o iphc.o nhc.o +6lowpan-y := core.o iphc.o nhc.o ndisc.o 6lowpan-$(CONFIG_6LOWPAN_DEBUGFS) += debugfs.o #rfc6282 nhcs diff --git a/net/6lowpan/core.c b/net/6lowpan/core.c index 1c7a42b..5945f7e 100644 --- a/net/6lowpan/core.c +++ b/net/6lowpan/core.c @@ -34,6 +34,8 @@ int lowpan_register_netdevice(struct net_device *dev, for (i = 0; i < LOWPAN_IPHC_CTX_TABLE_SIZE; i++) lowpan_dev(dev)->ctx.table[i].id = i; + dev->ndisc_ops = &lowpan_ndisc_ops; + ret = register_netdevice(dev); if (ret < 0) return ret; @@ -73,7 +75,7 @@ void lowpan_unregister_netdev(struct net_device *dev) } EXPORT_SYMBOL(lowpan_unregister_netdev); -static int addrconf_ifid_802154_6lowpan(u8 *eui, struct net_device *dev) +int addrconf_ifid_802154_6lowpan(u8 *eui, struct net_device *dev) { struct wpan_dev *wpan_dev = lowpan_802154_dev(dev)->wdev->ieee802154_ptr; diff --git a/net/6lowpan/ndisc.c b/net/6lowpan/ndisc.c new file mode 100644 index 0000000..ae1d419 --- /dev/null +++ b/net/6lowpan/ndisc.c @@ -0,0 +1,234 @@ +/* This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * Authors: + * (C) 2016 Pengutronix, Alexander Aring + */ + +#include +#include +#include + +#include "6lowpan_i.h" + +static int lowpan_ndisc_is_useropt(u8 nd_opt_type) +{ + return nd_opt_type == ND_OPT_6CO; +} + +#if IS_ENABLED(CONFIG_IEEE802154_6LOWPAN) +#define NDISC_802154_SHORT_ADDR_LENGTH 1 +static int lowpan_ndisc_parse_802154_options(const struct net_device *dev, + struct nd_opt_hdr *nd_opt, + struct ndisc_options *ndopts) +{ + switch (nd_opt->nd_opt_len) { + case NDISC_802154_SHORT_ADDR_LENGTH: + if (ndopts->nd_802154_opt_array[nd_opt->nd_opt_type]) + ND_PRINTK(2, warn, + "%s: duplicated short addr ND6 option found: type=%d\n", + __func__, nd_opt->nd_opt_type); + else + ndopts->nd_802154_opt_array[nd_opt->nd_opt_type] = nd_opt; + return 1; + default: + /* all others will be handled by ndisc IPv6 option parsing */ + return 0; + } +} + +static int lowpan_ndisc_parse_options(const struct net_device *dev, + struct nd_opt_hdr *nd_opt, + struct ndisc_options *ndopts) +{ + switch (nd_opt->nd_opt_type) { + case ND_OPT_SOURCE_LL_ADDR: + case ND_OPT_TARGET_LL_ADDR: + return lowpan_ndisc_parse_802154_options(dev, nd_opt, ndopts); + default: + return 0; + } +} + +static void lowpan_ndisc_802154_update(struct neighbour *n, u32 flags, + u8 icmp6_type, + const struct ndisc_options *ndopts) +{ + struct lowpan_802154_neigh *neigh = lowpan_802154_neigh(neighbour_priv(n)); + u8 *lladdr_short = NULL; + + switch (icmp6_type) { + case NDISC_ROUTER_SOLICITATION: + case NDISC_ROUTER_ADVERTISEMENT: + case NDISC_NEIGHBOUR_SOLICITATION: + if (ndopts->nd_802154_opts_src_lladdr) { + lladdr_short = __ndisc_opt_addr_data(ndopts->nd_802154_opts_src_lladdr, + IEEE802154_SHORT_ADDR_LEN, 0); + if (!lladdr_short) { + ND_PRINTK(2, warn, + "NA: invalid short link-layer address length\n"); + return; + } + } + break; + case NDISC_REDIRECT: + case NDISC_NEIGHBOUR_ADVERTISEMENT: + if (ndopts->nd_802154_opts_tgt_lladdr) { + lladdr_short = __ndisc_opt_addr_data(ndopts->nd_802154_opts_tgt_lladdr, + IEEE802154_SHORT_ADDR_LEN, 0); + if (!lladdr_short) { + ND_PRINTK(2, warn, + "NA: invalid short link-layer address length\n"); + return; + } + } + break; + default: + break; + } + + write_lock_bh(&n->lock); + if (lladdr_short) + ieee802154_be16_to_le16(&neigh->short_addr, lladdr_short); + else + neigh->short_addr = cpu_to_le16(IEEE802154_ADDR_SHORT_UNSPEC); + write_unlock_bh(&n->lock); +} + +static void lowpan_ndisc_update(const struct net_device *dev, + struct neighbour *n, u32 flags, u8 icmp6_type, + const struct ndisc_options *ndopts) +{ + if (!lowpan_is_ll(dev, LOWPAN_LLTYPE_IEEE802154)) + return; + + /* react on overrides only. TODO check if this is really right. */ + if (flags & NEIGH_UPDATE_F_OVERRIDE) + lowpan_ndisc_802154_update(n, flags, icmp6_type, ndopts); +} + +static int lowpan_ndisc_opt_addr_space(const struct net_device *dev, + u8 icmp6_type, struct neighbour *neigh, + u8 *ha_buf, u8 **ha) +{ + struct lowpan_802154_neigh *n; + struct wpan_dev *wpan_dev; + int addr_space = 0; + + if (!lowpan_is_ll(dev, LOWPAN_LLTYPE_IEEE802154)) + return 0; + + switch (icmp6_type) { + case NDISC_REDIRECT: + n = lowpan_802154_neigh(neighbour_priv(neigh)); + + read_lock_bh(&neigh->lock); + if (lowpan_802154_is_valid_src_short_addr(n->short_addr)) { + memcpy(ha_buf, &n->short_addr, + IEEE802154_SHORT_ADDR_LEN); + read_unlock_bh(&neigh->lock); + addr_space += __ndisc_opt_addr_space(IEEE802154_SHORT_ADDR_LEN, 0); + *ha = ha_buf; + } + read_unlock_bh(&neigh->lock); + break; + case NDISC_NEIGHBOUR_ADVERTISEMENT: + case NDISC_NEIGHBOUR_SOLICITATION: + case NDISC_ROUTER_SOLICITATION: + wpan_dev = lowpan_802154_dev(dev)->wdev->ieee802154_ptr; + + if (lowpan_802154_is_valid_src_short_addr(wpan_dev->short_addr)) + addr_space = __ndisc_opt_addr_space(IEEE802154_SHORT_ADDR_LEN, 0); + break; + default: + break; + } + + return addr_space; +} + +static void lowpan_ndisc_fill_addr_option(const struct net_device *dev, + struct sk_buff *skb, u8 icmp6_type, + const u8 *ha) +{ + struct wpan_dev *wpan_dev; + __be16 short_addr; + u8 opt_type; + + if (!lowpan_is_ll(dev, LOWPAN_LLTYPE_IEEE802154)) + return; + + switch (icmp6_type) { + case NDISC_REDIRECT: + if (ha) { + ieee802154_le16_to_be16(&short_addr, ha); + __ndisc_fill_addr_option(skb, ND_OPT_TARGET_LL_ADDR, + &short_addr, + IEEE802154_SHORT_ADDR_LEN, 0); + } + return; + case NDISC_NEIGHBOUR_ADVERTISEMENT: + opt_type = ND_OPT_TARGET_LL_ADDR; + break; + case NDISC_ROUTER_SOLICITATION: + case NDISC_NEIGHBOUR_SOLICITATION: + opt_type = ND_OPT_SOURCE_LL_ADDR; + break; + default: + return; + } + + wpan_dev = lowpan_802154_dev(dev)->wdev->ieee802154_ptr; + + if (lowpan_802154_is_valid_src_short_addr(wpan_dev->short_addr)) { + ieee802154_le16_to_be16(&short_addr, + &wpan_dev->short_addr); + __ndisc_fill_addr_option(skb, opt_type, &short_addr, + IEEE802154_SHORT_ADDR_LEN, 0); + } +} + +static void lowpan_ndisc_prefix_rcv_add_addr(struct net *net, + struct net_device *dev, + const struct prefix_info *pinfo, + struct inet6_dev *in6_dev, + struct in6_addr *addr, + int addr_type, u32 addr_flags, + bool sllao, bool tokenized, + __u32 valid_lft, + u32 prefered_lft, + bool dev_addr_generated) +{ + int err; + + /* generates short based address for RA PIO's */ + if (lowpan_is_ll(dev, LOWPAN_LLTYPE_IEEE802154) && dev_addr_generated && + !addrconf_ifid_802154_6lowpan(addr->s6_addr + 8, dev)) { + err = addrconf_prefix_rcv_add_addr(net, dev, pinfo, in6_dev, + addr, addr_type, addr_flags, + sllao, tokenized, valid_lft, + prefered_lft); + if (err) + ND_PRINTK(2, warn, + "RA: could not add a short address based address for prefix: %pI6c\n", + &pinfo->prefix); + } +} +#endif + +const struct ndisc_ops lowpan_ndisc_ops = { + .is_useropt = lowpan_ndisc_is_useropt, +#if IS_ENABLED(CONFIG_IEEE802154_6LOWPAN) + .parse_options = lowpan_ndisc_parse_options, + .update = lowpan_ndisc_update, + .opt_addr_space = lowpan_ndisc_opt_addr_space, + .fill_addr_option = lowpan_ndisc_fill_addr_option, + .prefix_rcv_add_addr = lowpan_ndisc_prefix_rcv_add_addr, +#endif +}; -- cgit v0.10.2 From cfce94653dad2d0661e1926c028ce63052eb20cd Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:26 +0200 Subject: 6lowpan: add support for getting short address In case of sending RA messages we need some way to get the short address from an 802.15.4 6LoWPAN interface. This patch will add a temporary debugfs entry for experimental userspace api. Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/net/6lowpan/debugfs.c b/net/6lowpan/debugfs.c index acbaa3d..24915e0 100644 --- a/net/6lowpan/debugfs.c +++ b/net/6lowpan/debugfs.c @@ -245,6 +245,41 @@ static const struct file_operations lowpan_context_fops = { .release = single_release, }; +static int lowpan_short_addr_get(void *data, u64 *val) +{ + struct wpan_dev *wdev = data; + + rtnl_lock(); + *val = le16_to_cpu(wdev->short_addr); + rtnl_unlock(); + + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(lowpan_short_addr_fops, lowpan_short_addr_get, + NULL, "0x%04llx\n"); + +static int lowpan_dev_debugfs_802154_init(const struct net_device *dev, + struct lowpan_dev *ldev) +{ + struct dentry *dentry, *root; + + if (!lowpan_is_ll(dev, LOWPAN_LLTYPE_IEEE802154)) + return 0; + + root = debugfs_create_dir("ieee802154", ldev->iface_debugfs); + if (!root) + return -EINVAL; + + dentry = debugfs_create_file("short_addr", 0444, root, + lowpan_802154_dev(dev)->wdev->ieee802154_ptr, + &lowpan_short_addr_fops); + if (!dentry) + return -EINVAL; + + return 0; +} + int lowpan_dev_debugfs_init(struct net_device *dev) { struct lowpan_dev *ldev = lowpan_dev(dev); @@ -272,6 +307,10 @@ int lowpan_dev_debugfs_init(struct net_device *dev) goto remove_root; } + ret = lowpan_dev_debugfs_802154_init(dev, ldev); + if (ret < 0) + goto remove_root; + return 0; remove_root: -- cgit v0.10.2 From eab560e58208730ec47e1e0461b8db1049d5d176 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:27 +0200 Subject: 6lowpan: add support for 802.15.4 short addr handling This patch adds necessary handling for use the short address for 802.15.4 6lowpan. It contains support for IPHC address compression and new matching algorithmn to decide which link layer address will be used for 802.15.4 frame. Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: David S. Miller diff --git a/net/6lowpan/iphc.c b/net/6lowpan/iphc.c index 8501dd5..79f1fa2 100644 --- a/net/6lowpan/iphc.c +++ b/net/6lowpan/iphc.c @@ -761,22 +761,75 @@ static const u8 lowpan_iphc_dam_to_sam_value[] = { [LOWPAN_IPHC_DAM_11] = LOWPAN_IPHC_SAM_11, }; -static u8 lowpan_compress_ctx_addr(u8 **hc_ptr, const struct in6_addr *ipaddr, +static inline bool +lowpan_iphc_compress_ctx_802154_lladdr(const struct in6_addr *ipaddr, + const struct lowpan_iphc_ctx *ctx, + const void *lladdr) +{ + const struct ieee802154_addr *addr = lladdr; + unsigned char extended_addr[EUI64_ADDR_LEN]; + bool lladdr_compress = false; + struct in6_addr tmp = {}; + + switch (addr->mode) { + case IEEE802154_ADDR_LONG: + ieee802154_le64_to_be64(&extended_addr, &addr->extended_addr); + /* check for SAM/DAM = 11 */ + memcpy(&tmp.s6_addr[8], &extended_addr, EUI64_ADDR_LEN); + /* second bit-flip (Universe/Local) is done according RFC2464 */ + tmp.s6_addr[8] ^= 0x02; + /* context information are always used */ + ipv6_addr_prefix_copy(&tmp, &ctx->pfx, ctx->plen); + if (ipv6_addr_equal(&tmp, ipaddr)) + lladdr_compress = true; + break; + case IEEE802154_ADDR_SHORT: + tmp.s6_addr[11] = 0xFF; + tmp.s6_addr[12] = 0xFE; + ieee802154_le16_to_be16(&tmp.s6_addr16[7], + &addr->short_addr); + /* context information are always used */ + ipv6_addr_prefix_copy(&tmp, &ctx->pfx, ctx->plen); + if (ipv6_addr_equal(&tmp, ipaddr)) + lladdr_compress = true; + break; + default: + /* should never handled and filtered by 802154 6lowpan */ + WARN_ON_ONCE(1); + break; + } + + return lladdr_compress; +} + +static u8 lowpan_compress_ctx_addr(u8 **hc_ptr, const struct net_device *dev, + const struct in6_addr *ipaddr, const struct lowpan_iphc_ctx *ctx, const unsigned char *lladdr, bool sam) { struct in6_addr tmp = {}; u8 dam; - /* check for SAM/DAM = 11 */ - memcpy(&tmp.s6_addr[8], lladdr, 8); - /* second bit-flip (Universe/Local) is done according RFC2464 */ - tmp.s6_addr[8] ^= 0x02; - /* context information are always used */ - ipv6_addr_prefix_copy(&tmp, &ctx->pfx, ctx->plen); - if (ipv6_addr_equal(&tmp, ipaddr)) { - dam = LOWPAN_IPHC_DAM_11; - goto out; + switch (lowpan_dev(dev)->lltype) { + case LOWPAN_LLTYPE_IEEE802154: + if (lowpan_iphc_compress_ctx_802154_lladdr(ipaddr, ctx, + lladdr)) { + dam = LOWPAN_IPHC_DAM_11; + goto out; + } + break; + default: + /* check for SAM/DAM = 11 */ + memcpy(&tmp.s6_addr[8], lladdr, EUI64_ADDR_LEN); + /* second bit-flip (Universe/Local) is done according RFC2464 */ + tmp.s6_addr[8] ^= 0x02; + /* context information are always used */ + ipv6_addr_prefix_copy(&tmp, &ctx->pfx, ctx->plen); + if (ipv6_addr_equal(&tmp, ipaddr)) { + dam = LOWPAN_IPHC_DAM_11; + goto out; + } + break; } memset(&tmp, 0, sizeof(tmp)); @@ -813,28 +866,85 @@ out: return dam; } -static u8 lowpan_compress_addr_64(u8 **hc_ptr, const struct in6_addr *ipaddr, +static inline bool +lowpan_iphc_compress_802154_lladdr(const struct in6_addr *ipaddr, + const void *lladdr) +{ + const struct ieee802154_addr *addr = lladdr; + unsigned char extended_addr[EUI64_ADDR_LEN]; + bool lladdr_compress = false; + struct in6_addr tmp = {}; + + switch (addr->mode) { + case IEEE802154_ADDR_LONG: + ieee802154_le64_to_be64(&extended_addr, &addr->extended_addr); + if (is_addr_mac_addr_based(ipaddr, extended_addr)) + lladdr_compress = true; + break; + case IEEE802154_ADDR_SHORT: + /* fe:80::ff:fe00:XXXX + * \__/ + * short_addr + * + * Universe/Local bit is zero. + */ + tmp.s6_addr[0] = 0xFE; + tmp.s6_addr[1] = 0x80; + tmp.s6_addr[11] = 0xFF; + tmp.s6_addr[12] = 0xFE; + ieee802154_le16_to_be16(&tmp.s6_addr16[7], + &addr->short_addr); + if (ipv6_addr_equal(&tmp, ipaddr)) + lladdr_compress = true; + break; + default: + /* should never handled and filtered by 802154 6lowpan */ + WARN_ON_ONCE(1); + break; + } + + return lladdr_compress; +} + +static u8 lowpan_compress_addr_64(u8 **hc_ptr, const struct net_device *dev, + const struct in6_addr *ipaddr, const unsigned char *lladdr, bool sam) { - u8 dam = LOWPAN_IPHC_DAM_00; + u8 dam = LOWPAN_IPHC_DAM_01; - if (is_addr_mac_addr_based(ipaddr, lladdr)) { - dam = LOWPAN_IPHC_DAM_11; /* 0-bits */ - pr_debug("address compression 0 bits\n"); - } else if (lowpan_is_iid_16_bit_compressable(ipaddr)) { + switch (lowpan_dev(dev)->lltype) { + case LOWPAN_LLTYPE_IEEE802154: + if (lowpan_iphc_compress_802154_lladdr(ipaddr, lladdr)) { + dam = LOWPAN_IPHC_DAM_11; /* 0-bits */ + pr_debug("address compression 0 bits\n"); + goto out; + } + break; + default: + if (is_addr_mac_addr_based(ipaddr, lladdr)) { + dam = LOWPAN_IPHC_DAM_11; /* 0-bits */ + pr_debug("address compression 0 bits\n"); + goto out; + } + break; + } + + if (lowpan_is_iid_16_bit_compressable(ipaddr)) { /* compress IID to 16 bits xxxx::XXXX */ lowpan_push_hc_data(hc_ptr, &ipaddr->s6_addr16[7], 2); dam = LOWPAN_IPHC_DAM_10; /* 16-bits */ raw_dump_inline(NULL, "Compressed ipv6 addr is (16 bits)", *hc_ptr - 2, 2); - } else { - /* do not compress IID => xxxx::IID */ - lowpan_push_hc_data(hc_ptr, &ipaddr->s6_addr16[4], 8); - dam = LOWPAN_IPHC_DAM_01; /* 64-bits */ - raw_dump_inline(NULL, "Compressed ipv6 addr is (64 bits)", - *hc_ptr - 8, 8); + goto out; } + /* do not compress IID => xxxx::IID */ + lowpan_push_hc_data(hc_ptr, &ipaddr->s6_addr16[4], 8); + raw_dump_inline(NULL, "Compressed ipv6 addr is (64 bits)", + *hc_ptr - 8, 8); + +out: + if (sam) return lowpan_iphc_dam_to_sam_value[dam]; else @@ -1013,9 +1123,6 @@ int lowpan_header_compress(struct sk_buff *skb, const struct net_device *dev, iphc0 = LOWPAN_DISPATCH_IPHC; iphc1 = 0; - raw_dump_inline(__func__, "saddr", saddr, EUI64_ADDR_LEN); - raw_dump_inline(__func__, "daddr", daddr, EUI64_ADDR_LEN); - raw_dump_table(__func__, "sending raw skb network uncompressed packet", skb->data, skb->len); @@ -1088,14 +1195,15 @@ int lowpan_header_compress(struct sk_buff *skb, const struct net_device *dev, iphc1 |= LOWPAN_IPHC_SAC; } else { if (sci) { - iphc1 |= lowpan_compress_ctx_addr(&hc_ptr, &hdr->saddr, + iphc1 |= lowpan_compress_ctx_addr(&hc_ptr, dev, + &hdr->saddr, &sci_entry, saddr, true); iphc1 |= LOWPAN_IPHC_SAC; } else { if (ipv6_saddr_type & IPV6_ADDR_LINKLOCAL && lowpan_is_linklocal_zero_padded(hdr->saddr)) { - iphc1 |= lowpan_compress_addr_64(&hc_ptr, + iphc1 |= lowpan_compress_addr_64(&hc_ptr, dev, &hdr->saddr, saddr, true); pr_debug("source address unicast link-local %pI6c iphc1 0x%02x\n", @@ -1123,14 +1231,15 @@ int lowpan_header_compress(struct sk_buff *skb, const struct net_device *dev, } } else { if (dci) { - iphc1 |= lowpan_compress_ctx_addr(&hc_ptr, &hdr->daddr, + iphc1 |= lowpan_compress_ctx_addr(&hc_ptr, dev, + &hdr->daddr, &dci_entry, daddr, false); iphc1 |= LOWPAN_IPHC_DAC; } else { if (ipv6_daddr_type & IPV6_ADDR_LINKLOCAL && lowpan_is_linklocal_zero_padded(hdr->daddr)) { - iphc1 |= lowpan_compress_addr_64(&hc_ptr, + iphc1 |= lowpan_compress_addr_64(&hc_ptr, dev, &hdr->daddr, daddr, false); pr_debug("dest address unicast link-local %pI6c iphc1 0x%02x\n", diff --git a/net/ieee802154/6lowpan/tx.c b/net/ieee802154/6lowpan/tx.c index e459afd..dbb476d 100644 --- a/net/ieee802154/6lowpan/tx.c +++ b/net/ieee802154/6lowpan/tx.c @@ -9,6 +9,7 @@ */ #include +#include #include #include @@ -17,19 +18,9 @@ #define LOWPAN_FRAG1_HEAD_SIZE 0x4 #define LOWPAN_FRAGN_HEAD_SIZE 0x5 -/* don't save pan id, it's intra pan */ -struct lowpan_addr { - u8 mode; - union { - /* IPv6 needs big endian here */ - __be64 extended_addr; - __be16 short_addr; - } u; -}; - struct lowpan_addr_info { - struct lowpan_addr daddr; - struct lowpan_addr saddr; + struct ieee802154_addr daddr; + struct ieee802154_addr saddr; }; static inline struct @@ -48,12 +39,14 @@ lowpan_addr_info *lowpan_skb_priv(const struct sk_buff *skb) * RAW/DGRAM sockets. */ int lowpan_header_create(struct sk_buff *skb, struct net_device *ldev, - unsigned short type, const void *_daddr, - const void *_saddr, unsigned int len) + unsigned short type, const void *daddr, + const void *saddr, unsigned int len) { - const u8 *saddr = _saddr; - const u8 *daddr = _daddr; - struct lowpan_addr_info *info; + struct wpan_dev *wpan_dev = lowpan_802154_dev(ldev)->wdev->ieee802154_ptr; + struct lowpan_addr_info *info = lowpan_skb_priv(skb); + struct lowpan_802154_neigh *llneigh = NULL; + const struct ipv6hdr *hdr = ipv6_hdr(skb); + struct neighbour *n; /* TODO: * if this package isn't ipv6 one, where should it be routed? @@ -61,21 +54,50 @@ int lowpan_header_create(struct sk_buff *skb, struct net_device *ldev, if (type != ETH_P_IPV6) return 0; - if (!saddr) - saddr = ldev->dev_addr; + /* intra-pan communication */ + info->saddr.pan_id = wpan_dev->pan_id; + info->daddr.pan_id = info->saddr.pan_id; - raw_dump_inline(__func__, "saddr", (unsigned char *)saddr, 8); - raw_dump_inline(__func__, "daddr", (unsigned char *)daddr, 8); + if (!memcmp(daddr, ldev->broadcast, EUI64_ADDR_LEN)) { + info->daddr.short_addr = cpu_to_le16(IEEE802154_ADDR_BROADCAST); + info->daddr.mode = IEEE802154_ADDR_SHORT; + } else { + __le16 short_addr = cpu_to_le16(IEEE802154_ADDR_SHORT_UNSPEC); + + n = neigh_lookup(&nd_tbl, &hdr->daddr, ldev); + if (n) { + llneigh = lowpan_802154_neigh(neighbour_priv(n)); + read_lock_bh(&n->lock); + short_addr = llneigh->short_addr; + read_unlock_bh(&n->lock); + } - info = lowpan_skb_priv(skb); + if (llneigh && + lowpan_802154_is_valid_src_short_addr(short_addr)) { + info->daddr.short_addr = short_addr; + info->daddr.mode = IEEE802154_ADDR_SHORT; + } else { + info->daddr.mode = IEEE802154_ADDR_LONG; + ieee802154_be64_to_le64(&info->daddr.extended_addr, + daddr); + } - /* TODO: Currently we only support extended_addr */ - info->daddr.mode = IEEE802154_ADDR_LONG; - memcpy(&info->daddr.u.extended_addr, daddr, - sizeof(info->daddr.u.extended_addr)); - info->saddr.mode = IEEE802154_ADDR_LONG; - memcpy(&info->saddr.u.extended_addr, saddr, - sizeof(info->daddr.u.extended_addr)); + if (n) + neigh_release(n); + } + + if (!saddr) { + if (lowpan_802154_is_valid_src_short_addr(wpan_dev->short_addr)) { + info->saddr.mode = IEEE802154_ADDR_SHORT; + info->saddr.short_addr = wpan_dev->short_addr; + } else { + info->saddr.mode = IEEE802154_ADDR_LONG; + info->saddr.extended_addr = wpan_dev->extended_addr; + } + } else { + info->saddr.mode = IEEE802154_ADDR_LONG; + ieee802154_be64_to_le64(&info->saddr.extended_addr, saddr); + } return 0; } @@ -209,47 +231,26 @@ static int lowpan_header(struct sk_buff *skb, struct net_device *ldev, u16 *dgram_size, u16 *dgram_offset) { struct wpan_dev *wpan_dev = lowpan_802154_dev(ldev)->wdev->ieee802154_ptr; - struct ieee802154_addr sa, da; struct ieee802154_mac_cb *cb = mac_cb_init(skb); struct lowpan_addr_info info; - void *daddr, *saddr; memcpy(&info, lowpan_skb_priv(skb), sizeof(info)); - /* TODO: Currently we only support extended_addr */ - daddr = &info.daddr.u.extended_addr; - saddr = &info.saddr.u.extended_addr; - *dgram_size = skb->len; - lowpan_header_compress(skb, ldev, daddr, saddr); + lowpan_header_compress(skb, ldev, &info.daddr, &info.saddr); /* dgram_offset = (saved bytes after compression) + lowpan header len */ *dgram_offset = (*dgram_size - skb->len) + skb_network_header_len(skb); cb->type = IEEE802154_FC_TYPE_DATA; - /* prepare wpan address data */ - sa.mode = IEEE802154_ADDR_LONG; - sa.pan_id = wpan_dev->pan_id; - sa.extended_addr = ieee802154_devaddr_from_raw(saddr); - - /* intra-PAN communications */ - da.pan_id = sa.pan_id; - - /* if the destination address is the broadcast address, use the - * corresponding short address - */ - if (!memcmp(daddr, ldev->broadcast, EUI64_ADDR_LEN)) { - da.mode = IEEE802154_ADDR_SHORT; - da.short_addr = cpu_to_le16(IEEE802154_ADDR_BROADCAST); + if (info.daddr.mode == IEEE802154_ADDR_SHORT && + ieee802154_is_broadcast_short_addr(info.daddr.short_addr)) cb->ackreq = false; - } else { - da.mode = IEEE802154_ADDR_LONG; - da.extended_addr = ieee802154_devaddr_from_raw(daddr); + else cb->ackreq = wpan_dev->ackreq; - } - return wpan_dev_hard_header(skb, lowpan_802154_dev(ldev)->wdev, &da, - &sa, 0); + return wpan_dev_hard_header(skb, lowpan_802154_dev(ldev)->wdev, + &info.daddr, &info.saddr, 0); } netdev_tx_t lowpan_xmit(struct sk_buff *skb, struct net_device *ldev) -- cgit v0.10.2 From a78c16e1b9ea74a64b356d667261a326fdc49513 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 14 Jun 2016 12:03:17 +0200 Subject: mdio: mux: avoid 'maybe-uninitialized' warning The latest changes to the MDIO code introduced a false-positive warning with gcc-6 (possibly others): drivers/net/phy/mdio-mux.c: In function 'mdio_mux_init': drivers/net/phy/mdio-mux.c:188:3: error: 'parent_bus_node' may be used uninitialized in this function [-Werror=maybe-uninitialized] It's easy to avoid the warning by making sure the parent_bus_node is initialized in both cases at the start of the function, since the later 'of_node_put()' call is also valid for a NULL pointer argument. Signed-off-by: Arnd Bergmann Fixes: f20e6657a875 ("mdio: mux: Enhanced MDIO mux framework for integrated multiplexers") Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/phy/mdio-mux.c b/drivers/net/phy/mdio-mux.c index dbd4ecc..963838d 100644 --- a/drivers/net/phy/mdio-mux.c +++ b/drivers/net/phy/mdio-mux.c @@ -115,6 +115,7 @@ int mdio_mux_init(struct device *dev, goto err_parent_bus; } } else { + parent_bus_node = NULL; parent_bus = mux_bus; } @@ -184,8 +185,7 @@ int mdio_mux_init(struct device *dev, put_device(&pb->mii_bus->dev); err_parent_bus: - if (!mux_bus) - of_node_put(parent_bus_node); + of_node_put(parent_bus_node); return ret_val; } EXPORT_SYMBOL_GPL(mdio_mux_init); -- cgit v0.10.2 From 810e530bfa1116079bf94b8c93b99b0208959261 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Tue, 14 Jun 2016 11:37:21 -0700 Subject: net: vrf: Switch dst dev to loopback on device delete Attempting to delete a VRF device with a socket bound to it can stall: unregister_netdevice: waiting for red to become free. Usage count = 1 The unregister is waiting for the dst to be released and with it references to the vrf device. Similar to dst_ifdown switch the dst dev to loopback on delete for all of the dst's for the vrf device and release the references to the vrf device. Fixes: 193125dbd8eb2 ("net: Introduce VRF device driver") Fixes: 35402e3136634 ("net: Add IPv6 support to VRF device") Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index e3fc6d3..32173aa 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -378,23 +378,37 @@ static int vrf_output6(struct net *net, struct sock *sk, struct sk_buff *skb) } /* holding rtnl */ -static void vrf_rt6_release(struct net_vrf *vrf) +static void vrf_rt6_release(struct net_device *dev, struct net_vrf *vrf) { struct rt6_info *rt6 = rtnl_dereference(vrf->rt6); struct rt6_info *rt6_local = rtnl_dereference(vrf->rt6_local); + struct net *net = dev_net(dev); + struct dst_entry *dst; RCU_INIT_POINTER(vrf->rt6, NULL); RCU_INIT_POINTER(vrf->rt6_local, NULL); synchronize_rcu(); - if (rt6) - dst_release(&rt6->dst); + /* move dev in dst's to loopback so this VRF device can be deleted + * - based on dst_ifdown + */ + if (rt6) { + dst = &rt6->dst; + dev_put(dst->dev); + dst->dev = net->loopback_dev; + dev_hold(dst->dev); + dst_release(dst); + } if (rt6_local) { if (rt6_local->rt6i_idev) in6_dev_put(rt6_local->rt6i_idev); - dst_release(&rt6_local->dst); + dst = &rt6_local->dst; + dev_put(dst->dev); + dst->dev = net->loopback_dev; + dev_hold(dst->dev); + dst_release(dst); } } @@ -449,7 +463,7 @@ out: return rc; } #else -static void vrf_rt6_release(struct net_vrf *vrf) +static void vrf_rt6_release(struct net_device *dev, struct net_vrf *vrf) { } @@ -518,20 +532,35 @@ static int vrf_output(struct net *net, struct sock *sk, struct sk_buff *skb) } /* holding rtnl */ -static void vrf_rtable_release(struct net_vrf *vrf) +static void vrf_rtable_release(struct net_device *dev, struct net_vrf *vrf) { struct rtable *rth = rtnl_dereference(vrf->rth); struct rtable *rth_local = rtnl_dereference(vrf->rth_local); + struct net *net = dev_net(dev); + struct dst_entry *dst; RCU_INIT_POINTER(vrf->rth, NULL); RCU_INIT_POINTER(vrf->rth_local, NULL); synchronize_rcu(); - if (rth) - dst_release(&rth->dst); + /* move dev in dst's to loopback so this VRF device can be deleted + * - based on dst_ifdown + */ + if (rth) { + dst = &rth->dst; + dev_put(dst->dev); + dst->dev = net->loopback_dev; + dev_hold(dst->dev); + dst_release(dst); + } - if (rth_local) - dst_release(&rth_local->dst); + if (rth_local) { + dst = &rth_local->dst; + dev_put(dst->dev); + dst->dev = net->loopback_dev; + dev_hold(dst->dev); + dst_release(dst); + } } static int vrf_rtable_create(struct net_device *dev) @@ -633,8 +662,8 @@ static void vrf_dev_uninit(struct net_device *dev) struct net_device *port_dev; struct list_head *iter; - vrf_rtable_release(vrf); - vrf_rt6_release(vrf); + vrf_rtable_release(dev, vrf); + vrf_rt6_release(dev, vrf); netdev_for_each_lower_dev(dev, port_dev, iter) vrf_del_slave(dev, port_dev); @@ -669,7 +698,7 @@ static int vrf_dev_init(struct net_device *dev) return 0; out_rth: - vrf_rtable_release(vrf); + vrf_rtable_release(dev, vrf); out_stats: free_percpu(dev->dstats); dev->dstats = NULL; -- cgit v0.10.2 From 22a59be8b7693eb2d0897a9638f5991f2f8e4ddd Mon Sep 17 00:00:00 2001 From: Philip Prindeville Date: Tue, 14 Jun 2016 15:53:02 -0600 Subject: net: ipv4: Add ability to have GRE ignore DF bit in IPv4 payloads In the presence of firewalls which improperly block ICMP Unreachable (including Fragmentation Required) messages, Path MTU Discovery is prevented from working. A workaround is to handle IPv4 payloads opaquely, ignoring the DF bit--as is done for other payloads like AppleTalk--and doing transparent fragmentation and reassembly. Redux includes the enforcement of mutual exclusion between this feature and Path MTU Discovery as suggested by Alexander Duyck. Cc: Alexander Duyck Reviewed-by: Stephen Hemminger Signed-off-by: Philip Prindeville Signed-off-by: David S. Miller diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index dbf4444..9222678 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -132,6 +132,7 @@ struct ip_tunnel { int ip_tnl_net_id; struct gro_cells gro_cells; bool collect_md; + bool ignore_df; }; #define TUNNEL_CSUM __cpu_to_be16(0x01) diff --git a/include/uapi/linux/if_tunnel.h b/include/uapi/linux/if_tunnel.h index af4de90..1046f55 100644 --- a/include/uapi/linux/if_tunnel.h +++ b/include/uapi/linux/if_tunnel.h @@ -113,6 +113,7 @@ enum { IFLA_GRE_ENCAP_SPORT, IFLA_GRE_ENCAP_DPORT, IFLA_GRE_COLLECT_METADATA, + IFLA_GRE_IGNORE_DF, __IFLA_GRE_MAX, }; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 4d2025f..0f8ca3f 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -841,17 +841,19 @@ out: return ipgre_tunnel_validate(tb, data); } -static void ipgre_netlink_parms(struct net_device *dev, +static int ipgre_netlink_parms(struct net_device *dev, struct nlattr *data[], struct nlattr *tb[], struct ip_tunnel_parm *parms) { + struct ip_tunnel *t = netdev_priv(dev); + memset(parms, 0, sizeof(*parms)); parms->iph.protocol = IPPROTO_GRE; if (!data) - return; + return 0; if (data[IFLA_GRE_LINK]) parms->link = nla_get_u32(data[IFLA_GRE_LINK]); @@ -880,16 +882,26 @@ static void ipgre_netlink_parms(struct net_device *dev, if (data[IFLA_GRE_TOS]) parms->iph.tos = nla_get_u8(data[IFLA_GRE_TOS]); - if (!data[IFLA_GRE_PMTUDISC] || nla_get_u8(data[IFLA_GRE_PMTUDISC])) + if (!data[IFLA_GRE_PMTUDISC] || nla_get_u8(data[IFLA_GRE_PMTUDISC])) { + if (t->ignore_df) + return -EINVAL; parms->iph.frag_off = htons(IP_DF); + } if (data[IFLA_GRE_COLLECT_METADATA]) { - struct ip_tunnel *t = netdev_priv(dev); - t->collect_md = true; if (dev->type == ARPHRD_IPGRE) dev->type = ARPHRD_NONE; } + + if (data[IFLA_GRE_IGNORE_DF]) { + if (nla_get_u8(data[IFLA_GRE_IGNORE_DF]) + && (parms->iph.frag_off & htons(IP_DF))) + return -EINVAL; + t->ignore_df = !!nla_get_u8(data[IFLA_GRE_IGNORE_DF]); + } + + return 0; } /* This function returns true when ENCAP attributes are present in the nl msg */ @@ -960,16 +972,19 @@ static int ipgre_newlink(struct net *src_net, struct net_device *dev, { struct ip_tunnel_parm p; struct ip_tunnel_encap ipencap; + int err; if (ipgre_netlink_encap_parms(data, &ipencap)) { struct ip_tunnel *t = netdev_priv(dev); - int err = ip_tunnel_encap_setup(t, &ipencap); + err = ip_tunnel_encap_setup(t, &ipencap); if (err < 0) return err; } - ipgre_netlink_parms(dev, data, tb, &p); + err = ipgre_netlink_parms(dev, data, tb, &p); + if (err < 0) + return err; return ip_tunnel_newlink(dev, tb, &p); } @@ -978,16 +993,19 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[], { struct ip_tunnel_parm p; struct ip_tunnel_encap ipencap; + int err; if (ipgre_netlink_encap_parms(data, &ipencap)) { struct ip_tunnel *t = netdev_priv(dev); - int err = ip_tunnel_encap_setup(t, &ipencap); + err = ip_tunnel_encap_setup(t, &ipencap); if (err < 0) return err; } - ipgre_netlink_parms(dev, data, tb, &p); + err = ipgre_netlink_parms(dev, data, tb, &p); + if (err < 0) + return err; return ip_tunnel_changelink(dev, tb, &p); } @@ -1024,6 +1042,8 @@ static size_t ipgre_get_size(const struct net_device *dev) nla_total_size(2) + /* IFLA_GRE_COLLECT_METADATA */ nla_total_size(0) + + /* IFLA_GRE_IGNORE_DF */ + nla_total_size(1) + 0; } @@ -1057,6 +1077,9 @@ static int ipgre_fill_info(struct sk_buff *skb, const struct net_device *dev) t->encap.flags)) goto nla_put_failure; + if (nla_put_u8(skb, IFLA_GRE_IGNORE_DF, t->ignore_df)) + goto nla_put_failure; + if (t->collect_md) { if (nla_put_flag(skb, IFLA_GRE_COLLECT_METADATA)) goto nla_put_failure; @@ -1084,6 +1107,7 @@ static const struct nla_policy ipgre_policy[IFLA_GRE_MAX + 1] = { [IFLA_GRE_ENCAP_SPORT] = { .type = NLA_U16 }, [IFLA_GRE_ENCAP_DPORT] = { .type = NLA_U16 }, [IFLA_GRE_COLLECT_METADATA] = { .type = NLA_FLAG }, + [IFLA_GRE_IGNORE_DF] = { .type = NLA_U8 }, }; static struct rtnl_link_ops ipgre_link_ops __read_mostly = { diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index d8f5e0a..95649eb 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -682,7 +682,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, } df = tnl_params->frag_off; - if (skb->protocol == htons(ETH_P_IP)) + if (skb->protocol == htons(ETH_P_IP) && !tunnel->ignore_df) df |= (inner_iph->frag_off&htons(IP_DF)); max_headroom = LL_RESERVED_SPACE(rt->dst.dev) + sizeof(struct iphdr) -- cgit v0.10.2 From 0b797c85894f9de091de2da16bc2ce80842898c0 Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Tue, 14 Jun 2016 16:29:15 -0700 Subject: ila: Fix checksum neutral mapping The algorithm for checksum neutral mapping is incorrect. This problem was being hidden since we were previously always performing checksum offload on the translated addresses and only with IPv6 HW csum. Enabling an ILA router shows the issue. Corrected algorithm: old_loc is the original locator in the packet, new_loc is the value to overwrite with and is found in the lookup table. old_flag is the old flag value (zero of CSUM_NEUTRAL_FLAG) and new_flag is then (old_flag ^ CSUM_NEUTRAL_FLAG) & CSUM_NEUTRAL_FLAG. Need SUM(new_id + new_flag + diff) == SUM(old_id + old_flag) for checksum neutral translation. Solving for diff gives: diff = (old_id - new_id) + (old_flag - new_flag) compute_csum_diff8(new_id, old_id) gives old_id - new_id If old_flag is set old_flag - new_flag = old_flag = CSUM_NEUTRAL_FLAG Else old_flag - new_flag = -new_flag = ~CSUM_NEUTRAL_FLAG Tested: - Implemented a user space program that creates random addresses and random locators to overwrite. Compares the checksum over the address before and after translation (must always be equal) - Enabled ILA router and showed proper operation. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller diff --git a/net/ipv6/ila/ila_common.c b/net/ipv6/ila/ila_common.c index b3d00be..ec9efbc 100644 --- a/net/ipv6/ila/ila_common.c +++ b/net/ipv6/ila/ila_common.c @@ -34,12 +34,12 @@ static void ila_csum_do_neutral(struct ila_addr *iaddr, if (p->locator_match.v64) { diff = p->csum_diff; } else { - diff = compute_csum_diff8((__be32 *)iaddr, - (__be32 *)&p->locator); + diff = compute_csum_diff8((__be32 *)&p->locator, + (__be32 *)iaddr); } fval = (__force __wsum)(ila_csum_neutral_set(iaddr->ident) ? - ~CSUM_NEUTRAL_FLAG : CSUM_NEUTRAL_FLAG); + CSUM_NEUTRAL_FLAG : ~CSUM_NEUTRAL_FLAG); diff = csum_add(diff, fval); @@ -140,8 +140,8 @@ void ila_init_saved_csum(struct ila_params *p) return; p->csum_diff = compute_csum_diff8( - (__be32 *)&p->locator_match, - (__be32 *)&p->locator); + (__be32 *)&p->locator, + (__be32 *)&p->locator_match); } static int __init ila_init(void) -- cgit v0.10.2 From ddc173a688a1ffef8b2a6547faaf68940d4dae0d Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 14 Jun 2016 16:54:43 -0700 Subject: liquidio: Avoid double free during soft command This patch is to resolve the double free issue by checking proper return values from soft command. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 245c063..1096cdb 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -317,7 +317,7 @@ octnet_mdio45_access(struct lio *lio, int op, int loc, int *value) retval = octeon_send_soft_command(oct_dev, sc); - if (retval) { + if (retval == IQ_SEND_FAILED) { dev_err(&oct_dev->pci_dev->dev, "octnet_mdio45_access instruction failed status: %x\n", retval); @@ -722,7 +722,7 @@ static int octnet_set_intrmod_cfg(void *oct, struct oct_intrmod_cfg *intr_cfg) sc->wait_time = 1000; retval = octeon_send_soft_command(oct_dev, sc); - if (retval) { + if (retval == IQ_SEND_FAILED) { octeon_free_soft_command(oct_dev, sc); return -EINVAL; } diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 655d89e..47fba0e 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -2583,7 +2583,7 @@ static inline int send_nic_timestamp_pkt(struct octeon_device *oct, retval = octeon_send_command(oct, sc->iq_no, ring_doorbell, &sc->cmd, sc, ih->dlengsz, ndata->reqtype); - if (retval) { + if (retval == IQ_SEND_FAILED) { dev_err(&oct->pci_dev->dev, "timestamp data packet failed status: %x\n", retval); octeon_free_soft_command(oct, sc); @@ -3192,7 +3192,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) sc->wait_time = 1000; retval = octeon_send_soft_command(octeon_dev, sc); - if (retval) { + if (retval == IQ_SEND_FAILED) { dev_err(&octeon_dev->pci_dev->dev, "iq/oq config failed status: %x\n", retval); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c index 1a01915..aacabe4 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c @@ -178,7 +178,7 @@ octnet_send_nic_ctrl_pkt(struct octeon_device *oct, } retval = octeon_send_soft_command(oct, sc); - if (retval) { + if (retval == IQ_SEND_FAILED) { octeon_free_soft_command(oct, sc); dev_err(&oct->pci_dev->dev, "%s soft command send failed status: %x\n", __func__, retval); -- cgit v0.10.2 From 26236fa9a13d8f37e7ecf3b2b69c74e57ad6e9d0 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 14 Jun 2016 16:54:44 -0700 Subject: liquidio: Host queue mapping changes This patch is to allocate the input queues based on Numa node in tx path and queue mapping changes based on the mapping info provided by firmware. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 1096cdb..2937c802 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -653,7 +653,7 @@ static int lio_get_intr_coalesce(struct net_device *netdev, intrmod_cfg->intrmod_mincnt_trigger; } - iq = oct->instr_queue[lio->linfo.txpciq[0]]; + iq = oct->instr_queue[lio->linfo.txpciq[0].s.q_no]; intr_coal->tx_max_coalesced_frames = iq->fill_threshold; break; @@ -859,7 +859,7 @@ static int lio_set_intr_coalesce(struct net_device *netdev, if ((intr_coal->tx_max_coalesced_frames >= CN6XXX_DB_MIN) && (intr_coal->tx_max_coalesced_frames <= CN6XXX_DB_MAX)) { for (j = 0; j < lio->linfo.num_txpciq; j++) { - q_no = lio->linfo.txpciq[j]; + q_no = lio->linfo.txpciq[j].s.q_no; oct->instr_queue[q_no]->fill_threshold = intr_coal->tx_max_coalesced_frames; } diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 47fba0e..3477a3c 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -682,7 +682,8 @@ static inline void txqs_wake(struct net_device *netdev) int i; for (i = 0; i < netdev->num_tx_queues; i++) - netif_wake_subqueue(netdev, i); + if (__netif_subqueue_stopped(netdev, i)) + netif_wake_subqueue(netdev, i); } else { netif_wake_queue(netdev); } @@ -752,11 +753,14 @@ static inline int check_txq_status(struct lio *lio) /* check each sub-queue state */ for (q = 0; q < numqs; q++) { - iq = lio->linfo.txpciq[q & (lio->linfo.num_txpciq - 1)]; + iq = lio->linfo.txpciq[q % + (lio->linfo.num_txpciq)].s.q_no; if (octnet_iq_is_full(lio->oct_dev, iq)) continue; - wake_q(lio->netdev, q); - ret_val++; + if (__netif_subqueue_stopped(lio->netdev, q)) { + wake_q(lio->netdev, q); + ret_val++; + } } } else { if (octnet_iq_is_full(lio->oct_dev, lio->txq)) @@ -1230,7 +1234,8 @@ static int liquidio_stop_nic_module(struct octeon_device *oct) for (i = 0; i < oct->ifcount; i++) { lio = GET_LIO(oct->props[i].netdev); for (j = 0; j < lio->linfo.num_rxpciq; j++) - octeon_unregister_droq_ops(oct, lio->linfo.rxpciq[j]); + octeon_unregister_droq_ops(oct, + lio->linfo.rxpciq[j].s.q_no); } for (i = 0; i < oct->ifcount; i++) @@ -1337,14 +1342,17 @@ static inline int check_txq_state(struct lio *lio, struct sk_buff *skb) if (netif_is_multiqueue(lio->netdev)) { q = skb->queue_mapping; - iq = lio->linfo.txpciq[(q & (lio->linfo.num_txpciq - 1))]; + iq = lio->linfo.txpciq[(q % (lio->linfo.num_txpciq))].s.q_no; } else { iq = lio->txq; + q = iq; } if (octnet_iq_is_full(lio->oct_dev, iq)) return 0; - wake_q(lio->netdev, q); + + if (__netif_subqueue_stopped(lio->netdev, q)) + wake_q(lio->netdev, q); return 1; } @@ -1743,14 +1751,13 @@ static void if_cfg_callback(struct octeon_device *oct, static u16 select_q(struct net_device *dev, struct sk_buff *skb, void *accel_priv, select_queue_fallback_t fallback) { - int qindex; + u32 qindex = 0; struct lio *lio; lio = GET_LIO(dev); - /* select queue on chosen queue_mapping or core */ - qindex = skb_rx_queue_recorded(skb) ? - skb_get_rx_queue(skb) : smp_processor_id(); - return (u16)(qindex & (lio->linfo.num_txpciq - 1)); + qindex = skb_tx_hash(dev, skb); + + return (u16)(qindex % (lio->linfo.num_txpciq)); } /** Routine to push packets arriving on Octeon interface upto network layer. @@ -1789,6 +1796,8 @@ liquidio_push_packet(u32 octeon_id, skb->dev = netdev; + skb_record_rx_queue(skb, droq->q_no); + if (rh->r_dh.has_hwtstamp) { /* timestamp is included from the hardware at the * beginning of the packet. @@ -1962,8 +1971,10 @@ static inline int setup_io_queues(struct octeon_device *octeon_dev, /* set up DROQs. */ for (q = 0; q < lio->linfo.num_rxpciq; q++) { - q_no = lio->linfo.rxpciq[q]; - + q_no = lio->linfo.rxpciq[q].s.q_no; + dev_dbg(&octeon_dev->pci_dev->dev, + "setup_io_queues index:%d linfo.rxpciq.s.q_no:%d\n", + q, q_no); retval = octeon_setup_droq(octeon_dev, q_no, CFG_GET_NUM_RX_DESCS_NIC_IF (octeon_get_conf(octeon_dev), @@ -2341,7 +2352,7 @@ static struct net_device_stats *liquidio_get_stats(struct net_device *netdev) oct = lio->oct_dev; for (i = 0; i < lio->linfo.num_txpciq; i++) { - iq_no = lio->linfo.txpciq[i]; + iq_no = lio->linfo.txpciq[i].s.q_no; iq_stats = &oct->instr_queue[iq_no]->stats; pkts += iq_stats->tx_done; drop += iq_stats->tx_dropped; @@ -2357,7 +2368,7 @@ static struct net_device_stats *liquidio_get_stats(struct net_device *netdev) bytes = 0; for (i = 0; i < lio->linfo.num_rxpciq; i++) { - oq_no = lio->linfo.rxpciq[i]; + oq_no = lio->linfo.rxpciq[i].s.q_no; oq_stats = &oct->droq[oq_no]->stats; pkts += oq_stats->rx_pkts_received; drop += (oq_stats->rx_dropped + @@ -2670,7 +2681,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) struct octnic_data_pkt ndata; struct octeon_device *oct; struct oct_iq_stats *stats; - int cpu = 0, status = 0; + int status = 0; int q_idx = 0, iq_no = 0; int xmit_more; u32 tag = 0; @@ -2679,9 +2690,10 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) oct = lio->oct_dev; if (netif_is_multiqueue(netdev)) { - cpu = skb->queue_mapping; - q_idx = (cpu & (lio->linfo.num_txpciq - 1)); - iq_no = lio->linfo.txpciq[q_idx]; + q_idx = skb->queue_mapping; + q_idx = (q_idx % (lio->linfo.num_txpciq)); + tag = q_idx; + iq_no = lio->linfo.txpciq[q_idx].s.q_no; } else { iq_no = lio->txq; } @@ -3125,7 +3137,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) struct liquidio_if_cfg_context *ctx; struct liquidio_if_cfg_resp *resp; struct octdev_props *props; - int retval, num_iqueues, num_oqueues, q_no; + int retval, num_iqueues, num_oqueues; u64 q_mask; int num_cpus = num_online_cpus(); union oct_nic_if_cfg if_cfg; @@ -3257,15 +3269,13 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) q_mask = resp->cfg_info.oqmask; /* q_mask is 0-based and already verified mask is nonzero */ for (j = 0; j < num_oqueues; j++) { - q_no = __ffs64(q_mask); - q_mask &= (~(1UL << q_no)); - lio->linfo.rxpciq[j] = q_no; + lio->linfo.rxpciq[j].u64 = + resp->cfg_info.linfo.rxpciq[j].u64; } q_mask = resp->cfg_info.iqmask; for (j = 0; j < num_iqueues; j++) { - q_no = __ffs64(q_mask); - q_mask &= (~(1UL << q_no)); - lio->linfo.txpciq[j] = q_no; + lio->linfo.txpciq[j].u64 = + resp->cfg_info.linfo.txpciq[j].u64; } lio->linfo.hw_addr = resp->cfg_info.linfo.hw_addr; lio->linfo.gmxport = resp->cfg_info.linfo.gmxport; @@ -3306,6 +3316,11 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) ether_addr_copy(netdev->dev_addr, mac); + /* By default all interfaces on a single Octeon uses the same + * tx and rx queues + */ + lio->txq = lio->linfo.txpciq[0].s.q_no; + lio->rxq = lio->linfo.rxpciq[0].s.q_no; if (setup_io_queues(octeon_dev, netdev)) { dev_err(&octeon_dev->pci_dev->dev, "I/O queues creation failed\n"); goto setup_nic_dev_fail; @@ -3313,12 +3328,6 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) ifstate_set(lio, LIO_IFSTATE_DROQ_OPS); - /* By default all interfaces on a single Octeon uses the same - * tx and rx queues - */ - lio->txq = lio->linfo.txpciq[0]; - lio->rxq = lio->linfo.rxpciq[0]; - lio->tx_qsize = octeon_get_tx_qsize(octeon_dev, lio->txq); lio->rx_qsize = octeon_get_rx_qsize(octeon_dev, lio->rxq); diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 0ac347c..00b3ef5 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -516,6 +516,46 @@ union oct_link_status { } s; }; +/** The txpciq info passed to host from the firmware */ + +union oct_txpciq { + u64 u64; + + struct { +#ifdef __BIG_ENDIAN_BITFIELD + u64 q_no:8; + u64 port:8; + u64 pkind:6; + u64 use_qpg:1; + u64 qpg:11; + u64 reserved:30; +#else + u64 reserved:30; + u64 qpg:11; + u64 use_qpg:1; + u64 pkind:6; + u64 port:8; + u64 q_no:8; +#endif + } s; +}; + +/** The rxpciq info passed to host from the firmware */ + +union oct_rxpciq { + u64 u64; + + struct { +#ifdef __BIG_ENDIAN_BITFIELD + u64 q_no:8; + u64 reserved:56; +#else + u64 reserved:56; + u64 q_no:8; +#endif + } s; +}; + /** Information for a OCTEON ethernet interface shared between core & host. */ struct oct_link_info { union oct_link_status link; @@ -535,8 +575,8 @@ struct oct_link_info { u16 gmxport; #endif - u8 txpciq[MAX_IOQS_PER_NICIF]; - u8 rxpciq[MAX_IOQS_PER_NICIF]; + union oct_txpciq txpciq[MAX_IOQS_PER_NICIF]; + union oct_rxpciq rxpciq[MAX_IOQS_PER_NICIF]; }; #define OCT_LINK_INFO_SIZE (sizeof(struct oct_link_info)) diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index 8e23e3f..967fe4d 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -741,36 +741,43 @@ struct octeon_device *octeon_allocate_device(u32 pci_id, return oct; } +/* this function is only for setting up the first queue */ int octeon_setup_instr_queues(struct octeon_device *oct) { - u32 i, num_iqs = 0; + u32 num_iqs = 0; u32 num_descs = 0; + u32 iq_no = 0; + union oct_txpciq txpciq; + int numa_node = cpu_to_node(iq_no % num_online_cpus()); + num_iqs = 1; /* this causes queue 0 to be default queue */ - if (OCTEON_CN6XXX(oct)) { - num_iqs = 1; + if (OCTEON_CN6XXX(oct)) num_descs = CFG_GET_NUM_DEF_TX_DESCS(CHIP_FIELD(oct, cn6xxx, conf)); - } oct->num_iqs = 0; - for (i = 0; i < num_iqs; i++) { - oct->instr_queue[i] = + oct->instr_queue[0] = vmalloc_node(sizeof(*oct->instr_queue[0]), + numa_node); + if (!oct->instr_queue[0]) + oct->instr_queue[0] = vmalloc(sizeof(struct octeon_instr_queue)); - if (!oct->instr_queue[i]) - return 1; - - memset(oct->instr_queue[i], 0, - sizeof(struct octeon_instr_queue)); - - oct->instr_queue[i]->app_ctx = (void *)(size_t)i; - if (octeon_init_instr_queue(oct, i, num_descs)) - return 1; - - oct->num_iqs++; + if (!oct->instr_queue[0]) + return 1; + memset(oct->instr_queue[0], 0, sizeof(struct octeon_instr_queue)); + oct->instr_queue[0]->app_ctx = (void *)(size_t)0; + txpciq.u64 = 0; + txpciq.s.q_no = iq_no; + txpciq.s.use_qpg = 0; + txpciq.s.qpg = 0; + if (octeon_init_instr_queue(oct, txpciq, num_descs)) { + /* prevent memory leak */ + vfree(oct->instr_queue[0]); + return 1; } + oct->num_iqs++; return 0; } diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h index 592fe49..658f1d0 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h @@ -81,8 +81,8 @@ struct octeon_instr_queue { /** Flag that indicates if the queue uses 64 byte commands. */ u32 iqcmd_64B:1; - /** Queue Number. */ - u32 iq_no:5; + /** Queue info. */ + union oct_txpciq txpciq; u32 rsvd:17; @@ -268,14 +268,15 @@ void octeon_free_soft_command(struct octeon_device *oct, /** * octeon_init_instr_queue() * @param octeon_dev - pointer to the octeon device structure. - * @param iq_no - queue to be initialized (0 <= q_no <= 3). + * @param txpciq - queue to be initialized (0 <= q_no <= 3). * * Called at driver init time for each input queue. iq_conf has the * configuration parameters for the queue. * * @return Success: 0 Failure: 1 */ -int octeon_init_instr_queue(struct octeon_device *octeon_dev, u32 iq_no, +int octeon_init_instr_queue(struct octeon_device *octeon_dev, + union oct_txpciq txpciq, u32 num_descs); /** @@ -313,7 +314,7 @@ void octeon_prepare_soft_command(struct octeon_device *oct, int octeon_send_soft_command(struct octeon_device *oct, struct octeon_soft_command *sc); -int octeon_setup_iq(struct octeon_device *oct, u32 iq_no, +int octeon_setup_iq(struct octeon_device *oct, union oct_txpciq, u32 num_descs, void *app_ctx); #endif /* __OCTEON_IQ_H__ */ diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index 9313915..1240461 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -69,12 +69,16 @@ static inline int IQ_INSTR_MODE_64B(struct octeon_device *oct, int iq_no) /* Return 0 on success, 1 on failure */ int octeon_init_instr_queue(struct octeon_device *oct, - u32 iq_no, u32 num_descs) + union oct_txpciq txpciq, + u32 num_descs) { struct octeon_instr_queue *iq; struct octeon_iq_config *conf = NULL; + u32 iq_no = (u32)txpciq.s.q_no; u32 q_size; struct cavium_wq *db_wq; + int orig_node = dev_to_node(&oct->pci_dev->dev); + int numa_node = cpu_to_node(iq_no % num_online_cpus()); if (OCTEON_CN6XXX(oct)) conf = &(CFG_GET_IQ_CFG(CHIP_FIELD(oct, cn6xxx, conf))); @@ -96,8 +100,13 @@ int octeon_init_instr_queue(struct octeon_device *oct, iq = oct->instr_queue[iq_no]; + set_dev_node(&oct->pci_dev->dev, numa_node); iq->base_addr = lio_dma_alloc(oct, q_size, (dma_addr_t *)&iq->base_addr_dma); + set_dev_node(&oct->pci_dev->dev, orig_node); + if (!iq->base_addr) + iq->base_addr = lio_dma_alloc(oct, q_size, + (dma_addr_t *)&iq->base_addr_dma); if (!iq->base_addr) { dev_err(&oct->pci_dev->dev, "Cannot allocate memory for instr queue %d\n", iq_no); @@ -109,7 +118,11 @@ int octeon_init_instr_queue(struct octeon_device *oct, /* Initialize a list to holds requests that have been posted to Octeon * but has yet to be fetched by octeon */ - iq->request_list = vmalloc(sizeof(*iq->request_list) * num_descs); + iq->request_list = vmalloc_node((sizeof(*iq->request_list) * num_descs), + numa_node); + if (!iq->request_list) + iq->request_list = vmalloc(sizeof(*iq->request_list) * + num_descs); if (!iq->request_list) { lio_dma_free(oct, q_size, iq->base_addr, iq->base_addr_dma); dev_err(&oct->pci_dev->dev, "Alloc failed for IQ[%d] nr free list\n", @@ -122,7 +135,7 @@ int octeon_init_instr_queue(struct octeon_device *oct, dev_dbg(&oct->pci_dev->dev, "IQ[%d]: base: %p basedma: %llx count: %d\n", iq_no, iq->base_addr, iq->base_addr_dma, iq->max_count); - iq->iq_no = iq_no; + iq->txpciq.u64 = txpciq.u64; iq->fill_threshold = (u32)conf->db_min; iq->fill_cnt = 0; iq->host_write_index = 0; @@ -189,18 +202,25 @@ int octeon_delete_instr_queue(struct octeon_device *oct, u32 iq_no) /* Return 0 on success, 1 on failure */ int octeon_setup_iq(struct octeon_device *oct, - u32 iq_no, + union oct_txpciq txpciq, u32 num_descs, void *app_ctx) { + u32 iq_no = (u32)txpciq.s.q_no; + int numa_node = cpu_to_node(iq_no % num_online_cpus()); + if (oct->instr_queue[iq_no]) { dev_dbg(&oct->pci_dev->dev, "IQ is in use. Cannot create the IQ: %d again\n", iq_no); + oct->instr_queue[iq_no]->txpciq.u64 = txpciq.u64; oct->instr_queue[iq_no]->app_ctx = app_ctx; return 0; } oct->instr_queue[iq_no] = - vmalloc(sizeof(struct octeon_instr_queue)); + vmalloc_node(sizeof(struct octeon_instr_queue), numa_node); + if (!oct->instr_queue[iq_no]) + oct->instr_queue[iq_no] = + vmalloc(sizeof(struct octeon_instr_queue)); if (!oct->instr_queue[iq_no]) return 1; @@ -208,7 +228,7 @@ int octeon_setup_iq(struct octeon_device *oct, sizeof(struct octeon_instr_queue)); oct->instr_queue[iq_no]->app_ctx = app_ctx; - if (octeon_init_instr_queue(oct, iq_no, num_descs)) { + if (octeon_init_instr_queue(oct, txpciq, num_descs)) { vfree(oct->instr_queue[iq_no]); oct->instr_queue[iq_no] = NULL; return 1; -- cgit v0.10.2 From fcd2b5e36ca91eae329d33a50977ee976122a3b9 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 14 Jun 2016 16:54:45 -0700 Subject: liquidio:Scatter gather list per IQ This patch is to allocate and manage scatter gather lists per input queue(iq's) and remove queue's interdependence. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 3477a3c..0daa89a 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -166,6 +166,8 @@ struct octnic_gather { * received from the IP layer. */ struct octeon_sg_entry *sg; + + u64 sg_dma_ptr; }; /** This structure is used by NIC driver to store information required @@ -791,64 +793,116 @@ static inline struct list_head *list_delete_head(struct list_head *root) } /** - * \brief Delete gather list + * \brief Delete gather lists * @param lio per-network private data */ -static void delete_glist(struct lio *lio) +static void delete_glists(struct lio *lio) { struct octnic_gather *g; + int i; - do { - g = (struct octnic_gather *) - list_delete_head(&lio->glist); - if (g) { - if (g->sg) - kfree((void *)((unsigned long)g->sg - - g->adjust)); - kfree(g); - } - } while (g); + if (!lio->glist) + return; + + for (i = 0; i < lio->linfo.num_txpciq; i++) { + do { + g = (struct octnic_gather *) + list_delete_head(&lio->glist[i]); + if (g) { + if (g->sg) { + dma_unmap_single(&lio->oct_dev-> + pci_dev->dev, + g->sg_dma_ptr, + g->sg_size, + DMA_TO_DEVICE); + kfree((void *)((unsigned long)g->sg - + g->adjust)); + } + kfree(g); + } + } while (g); + } + + kfree((void *)lio->glist); } /** - * \brief Setup gather list + * \brief Setup gather lists * @param lio per-network private data */ -static int setup_glist(struct lio *lio) +static int setup_glists(struct octeon_device *oct, struct lio *lio, int num_iqs) { - int i; + int i, j; struct octnic_gather *g; - INIT_LIST_HEAD(&lio->glist); + lio->glist_lock = kcalloc(num_iqs, sizeof(*lio->glist_lock), + GFP_KERNEL); + if (!lio->glist_lock) + return 1; - for (i = 0; i < lio->tx_qsize; i++) { - g = kzalloc(sizeof(*g), GFP_KERNEL); - if (!g) - break; + lio->glist = kcalloc(num_iqs, sizeof(*lio->glist), + GFP_KERNEL); + if (!lio->glist) { + kfree((void *)lio->glist_lock); + return 1; + } - g->sg_size = - ((ROUNDUP4(OCTNIC_MAX_SG) >> 2) * OCT_SG_ENTRY_SIZE); + for (i = 0; i < num_iqs; i++) { + int numa_node = cpu_to_node(i % num_online_cpus()); - g->sg = kmalloc(g->sg_size + 8, GFP_KERNEL); - if (!g->sg) { - kfree(g); - break; + spin_lock_init(&lio->glist_lock[i]); + + INIT_LIST_HEAD(&lio->glist[i]); + + for (j = 0; j < lio->tx_qsize; j++) { + g = kzalloc_node(sizeof(*g), GFP_KERNEL, + numa_node); + if (!g) + g = kzalloc(sizeof(*g), GFP_KERNEL); + if (!g) + break; + + g->sg_size = ((ROUNDUP4(OCTNIC_MAX_SG) >> 2) * + OCT_SG_ENTRY_SIZE); + + g->sg = kmalloc_node(g->sg_size + 8, + GFP_KERNEL, numa_node); + if (!g->sg) + g->sg = kmalloc(g->sg_size + 8, GFP_KERNEL); + if (!g->sg) { + kfree(g); + break; + } + + /* The gather component should be aligned on 64-bit + * boundary + */ + if (((unsigned long)g->sg) & 7) { + g->adjust = 8 - (((unsigned long)g->sg) & 7); + g->sg = (struct octeon_sg_entry *) + ((unsigned long)g->sg + g->adjust); + } + g->sg_dma_ptr = dma_map_single(&oct->pci_dev->dev, + g->sg, g->sg_size, + DMA_TO_DEVICE); + if (dma_mapping_error(&oct->pci_dev->dev, + g->sg_dma_ptr)) { + kfree((void *)((unsigned long)g->sg - + g->adjust)); + kfree(g); + break; + } + + list_add_tail(&g->list, &lio->glist[i]); } - /* The gather component should be aligned on 64-bit boundary */ - if (((unsigned long)g->sg) & 7) { - g->adjust = 8 - (((unsigned long)g->sg) & 7); - g->sg = (struct octeon_sg_entry *) - ((unsigned long)g->sg + g->adjust); + if (j != lio->tx_qsize) { + delete_glists(lio); + return 1; } - list_add_tail(&g->list, &lio->glist); } - if (i == lio->tx_qsize) - return 0; - - delete_glist(lio); - return 1; + return 0; } /** @@ -1209,7 +1263,7 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx) if (atomic_read(&lio->ifstate) & LIO_IFSTATE_REGISTERED) unregister_netdev(netdev); - delete_glist(lio); + delete_glists(lio); free_netdev(netdev); @@ -1331,6 +1385,16 @@ static int octeon_pci_os_setup(struct octeon_device *oct) return 0; } +static inline int skb_iq(struct lio *lio, struct sk_buff *skb) +{ + int q = 0; + + if (netif_is_multiqueue(lio->netdev)) + q = skb->queue_mapping % lio->linfo.num_txpciq; + + return q; +} + /** * \brief Check Tx queue state for a given network buffer * @param lio per-network private data @@ -1388,7 +1452,7 @@ static void free_netsgbuf(void *buf) struct sk_buff *skb; struct lio *lio; struct octnic_gather *g; - int i, frags; + int i, frags, iq; finfo = (struct octnet_buf_free_info *)buf; skb = finfo->skb; @@ -1410,13 +1474,13 @@ static void free_netsgbuf(void *buf) i++; } - dma_unmap_single(&lio->oct_dev->pci_dev->dev, - finfo->dptr, g->sg_size, - DMA_TO_DEVICE); + dma_sync_single_for_cpu(&lio->oct_dev->pci_dev->dev, + g->sg_dma_ptr, g->sg_size, DMA_TO_DEVICE); - spin_lock(&lio->lock); - list_add_tail(&g->list, &lio->glist); - spin_unlock(&lio->lock); + iq = skb_iq(lio, skb); + spin_lock(&lio->glist_lock[iq]); + list_add_tail(&g->list, &lio->glist[iq]); + spin_unlock(&lio->glist_lock[iq]); check_txq_state(lio, skb); /* mq support: sub-queue state check */ @@ -1434,7 +1498,7 @@ static void free_netsgbuf_with_resp(void *buf) struct sk_buff *skb; struct lio *lio; struct octnic_gather *g; - int i, frags; + int i, frags, iq; sc = (struct octeon_soft_command *)buf; skb = (struct sk_buff *)sc->callback_arg; @@ -1458,13 +1522,14 @@ static void free_netsgbuf_with_resp(void *buf) i++; } - dma_unmap_single(&lio->oct_dev->pci_dev->dev, - finfo->dptr, g->sg_size, - DMA_TO_DEVICE); + dma_sync_single_for_cpu(&lio->oct_dev->pci_dev->dev, + g->sg_dma_ptr, g->sg_size, DMA_TO_DEVICE); - spin_lock(&lio->lock); - list_add_tail(&g->list, &lio->glist); - spin_unlock(&lio->lock); + iq = skb_iq(lio, skb); + + spin_lock(&lio->glist_lock[iq]); + list_add_tail(&g->list, &lio->glist[iq]); + spin_unlock(&lio->glist_lock[iq]); /* Don't free the skb yet */ @@ -2683,7 +2748,8 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) struct oct_iq_stats *stats; int status = 0; int q_idx = 0, iq_no = 0; - int xmit_more; + int xmit_more, j; + u64 dptr = 0; u32 tag = 0; lio = GET_LIO(netdev); @@ -2826,9 +2892,10 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) struct skb_frag_struct *frag; struct octnic_gather *g; - spin_lock(&lio->lock); - g = (struct octnic_gather *)list_delete_head(&lio->glist); - spin_unlock(&lio->lock); + spin_lock(&lio->glist_lock[q_idx]); + g = (struct octnic_gather *) + list_delete_head(&lio->glist[q_idx]); + spin_unlock(&lio->glist_lock[q_idx]); if (!g) { netif_info(lio, tx_err, lio->netdev, @@ -2865,21 +2932,31 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) frag->size, DMA_TO_DEVICE); + if (dma_mapping_error(&oct->pci_dev->dev, + g->sg[i >> 2].ptr[i & 3])) { + dma_unmap_single(&oct->pci_dev->dev, + g->sg[0].ptr[0], + skb->len - skb->data_len, + DMA_TO_DEVICE); + for (j = 1; j < i; j++) { + frag = &skb_shinfo(skb)->frags[j - 1]; + dma_unmap_page(&oct->pci_dev->dev, + g->sg[j >> 2].ptr[j & 3], + frag->size, + DMA_TO_DEVICE); + } + dev_err(&oct->pci_dev->dev, "%s DMA mapping error 3\n", + __func__); + return NETDEV_TX_BUSY; + } + add_sg_size(&g->sg[(i >> 2)], frag->size, (i & 3)); i++; } - ndata.cmd.dptr = dma_map_single(&oct->pci_dev->dev, - g->sg, g->sg_size, - DMA_TO_DEVICE); - if (dma_mapping_error(&oct->pci_dev->dev, ndata.cmd.dptr)) { - dev_err(&oct->pci_dev->dev, "%s DMA mapping error 3\n", - __func__); - dma_unmap_single(&oct->pci_dev->dev, g->sg[0].ptr[0], - skb->len - skb->data_len, - DMA_TO_DEVICE); - return NETDEV_TX_BUSY; - } + dma_sync_single_for_device(&oct->pci_dev->dev, g->sg_dma_ptr, + g->sg_size, DMA_TO_DEVICE); + dptr = g->sg_dma_ptr; finfo->dptr = ndata.cmd.dptr; finfo->g = g; @@ -3301,7 +3378,6 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) lio->oct_dev = octeon_dev; lio->octprops = props; lio->netdev = netdev; - spin_lock_init(&lio->lock); dev_dbg(&octeon_dev->pci_dev->dev, "if%d gmx: %d hw_addr: 0x%llx\n", i, @@ -3331,7 +3407,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) lio->tx_qsize = octeon_get_tx_qsize(octeon_dev, lio->txq); lio->rx_qsize = octeon_get_rx_qsize(octeon_dev, lio->rxq); - if (setup_glist(lio)) { + if (setup_glists(octeon_dev, lio, num_iqueues)) { dev_err(&octeon_dev->pci_dev->dev, "Gather list allocation failed\n"); goto setup_nic_dev_fail; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h index b3abe58..0a50bac 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h @@ -48,11 +48,11 @@ struct lio { */ int rxq; - /** Guards the glist */ - spinlock_t lock; + /** Guards each glist */ + spinlock_t *glist_lock; - /** Linked list of gather components */ - struct list_head glist; + /** Array of gather component linked lists */ + struct list_head *glist; /** Pointer to the NIC properties for the Octeon device this network * interface is associated with. -- cgit v0.10.2 From 96ae48b7faa72260c9d7203f5ad6250e149fb085 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 14 Jun 2016 16:54:46 -0700 Subject: liquidio:RX queue alloc changes This patch is to allocate rx queue's memory based on numa node and also use page based buffers for rx traffic improvements. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index 967fe4d..c06807d 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -783,14 +783,15 @@ int octeon_setup_instr_queues(struct octeon_device *oct) int octeon_setup_output_queues(struct octeon_device *oct) { - u32 i, num_oqs = 0; + u32 num_oqs = 0; u32 num_descs = 0; u32 desc_size = 0; + u32 oq_no = 0; + int numa_node = cpu_to_node(oq_no % num_online_cpus()); + num_oqs = 1; /* this causes queue 0 to be default queue */ if (OCTEON_CN6XXX(oct)) { - /* CFG_GET_OQ_MAX_BASE_Q(CHIP_FIELD(oct, cn6xxx, conf)); */ - num_oqs = 1; num_descs = CFG_GET_NUM_DEF_RX_DESCS(CHIP_FIELD(oct, cn6xxx, conf)); desc_size = @@ -798,19 +799,15 @@ int octeon_setup_output_queues(struct octeon_device *oct) } oct->num_oqs = 0; + oct->droq[0] = vmalloc_node(sizeof(*oct->droq[0]), numa_node); + if (!oct->droq[0]) + oct->droq[0] = vmalloc(sizeof(*oct->droq[0])); + if (!oct->droq[0]) + return 1; - for (i = 0; i < num_oqs; i++) { - oct->droq[i] = vmalloc(sizeof(*oct->droq[i])); - if (!oct->droq[i]) - return 1; - - memset(oct->droq[i], 0, sizeof(struct octeon_droq)); - - if (octeon_init_droq(oct, i, num_descs, desc_size, NULL)) - return 1; - - oct->num_oqs++; - } + if (octeon_init_droq(oct, oq_no, num_descs, desc_size, NULL)) + return 1; + oct->num_oqs++; return 0; } diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c index 174072b..1f648dc 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c @@ -242,6 +242,8 @@ int octeon_init_droq(struct octeon_device *oct, struct octeon_droq *droq; u32 desc_ring_size = 0, c_num_descs = 0, c_buf_size = 0; u32 c_pkts_per_intr = 0, c_refill_threshold = 0; + int orig_node = dev_to_node(&oct->pci_dev->dev); + int numa_node = cpu_to_node(q_no % num_online_cpus()); dev_dbg(&oct->pci_dev->dev, "%s[%d]\n", __func__, q_no); @@ -261,15 +263,23 @@ int octeon_init_droq(struct octeon_device *oct, struct octeon_config *conf6x = CHIP_FIELD(oct, cn6xxx, conf); c_pkts_per_intr = (u32)CFG_GET_OQ_PKTS_PER_INTR(conf6x); - c_refill_threshold = (u32)CFG_GET_OQ_REFILL_THRESHOLD(conf6x); + c_refill_threshold = + (u32)CFG_GET_OQ_REFILL_THRESHOLD(conf6x); + } else { + return 1; } droq->max_count = c_num_descs; droq->buffer_size = c_buf_size; desc_ring_size = droq->max_count * OCT_DROQ_DESC_SIZE; + set_dev_node(&oct->pci_dev->dev, numa_node); droq->desc_ring = lio_dma_alloc(oct, desc_ring_size, (dma_addr_t *)&droq->desc_ring_dma); + set_dev_node(&oct->pci_dev->dev, orig_node); + if (!droq->desc_ring) + droq->desc_ring = lio_dma_alloc(oct, desc_ring_size, + (dma_addr_t *)&droq->desc_ring_dma); if (!droq->desc_ring) { dev_err(&oct->pci_dev->dev, @@ -283,12 +293,11 @@ int octeon_init_droq(struct octeon_device *oct, droq->max_count); droq->info_list = - cnnic_alloc_aligned_dma(oct->pci_dev, - (droq->max_count * OCT_DROQ_INFO_SIZE), - &droq->info_alloc_size, - &droq->info_base_addr, - &droq->info_list_dma); - + cnnic_numa_alloc_aligned_dma((droq->max_count * + OCT_DROQ_INFO_SIZE), + &droq->info_alloc_size, + &droq->info_base_addr, + numa_node); if (!droq->info_list) { dev_err(&oct->pci_dev->dev, "Cannot allocate memory for info list.\n"); lio_dma_free(oct, (droq->max_count * OCT_DROQ_DESC_SIZE), @@ -297,7 +306,12 @@ int octeon_init_droq(struct octeon_device *oct, } droq->recv_buf_list = (struct octeon_recv_buffer *) - vmalloc(droq->max_count * + vmalloc_node(droq->max_count * + OCT_DROQ_RECVBUF_SIZE, + numa_node); + if (!droq->recv_buf_list) + droq->recv_buf_list = (struct octeon_recv_buffer *) + vmalloc(droq->max_count * OCT_DROQ_RECVBUF_SIZE); if (!droq->recv_buf_list) { dev_err(&oct->pci_dev->dev, "Output queue recv buf list alloc failed\n"); @@ -949,6 +963,7 @@ int octeon_create_droq(struct octeon_device *oct, u32 desc_size, void *app_ctx) { struct octeon_droq *droq; + int numa_node = cpu_to_node(q_no % num_online_cpus()); if (oct->droq[q_no]) { dev_dbg(&oct->pci_dev->dev, "Droq already in use. Cannot create droq %d again\n", @@ -957,7 +972,9 @@ int octeon_create_droq(struct octeon_device *oct, } /* Allocate the DS for the new droq. */ - droq = vmalloc(sizeof(*droq)); + droq = vmalloc_node(sizeof(*droq), numa_node); + if (!droq) + droq = vmalloc(sizeof(*droq)); if (!droq) goto create_droq_fail; memset(droq, 0, sizeof(struct octeon_droq)); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_main.h b/drivers/net/ethernet/cavium/liquidio/octeon_main.h index cbd0819..0ff3efc 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_main.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_main.h @@ -126,22 +126,27 @@ static inline int octeon_map_pci_barx(struct octeon_device *oct, } static inline void * -cnnic_alloc_aligned_dma(struct pci_dev *pci_dev, - u32 size, - u32 *alloc_size, - size_t *orig_ptr, - size_t *dma_addr __attribute__((unused))) +cnnic_numa_alloc_aligned_dma(u32 size, + u32 *alloc_size, + size_t *orig_ptr, + int numa_node) { int retries = 0; void *ptr = NULL; #define OCTEON_MAX_ALLOC_RETRIES 1 do { - ptr = - (void *)__get_free_pages(GFP_KERNEL, - get_order(size)); + struct page *page = NULL; + + page = alloc_pages_node(numa_node, + GFP_KERNEL, + get_order(size)); + if (!page) + page = alloc_pages(GFP_KERNEL, + get_order(size)); + ptr = (void *)page_address(page); if ((unsigned long)ptr & 0x07) { - free_pages((unsigned long)ptr, get_order(size)); + __free_pages(page, get_order(size)); ptr = NULL; /* Increment the size required if the first * attempt failed. -- cgit v0.10.2 From cabeb13be9827991915a985d4b5683e59cba2f58 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 14 Jun 2016 16:54:47 -0700 Subject: liquidio: RX desc alloc changes This patch is to add page based buffers for receive side descriptors of the driver and separate free routines for rx and tx buffers. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 0daa89a..3a4f31f 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -1439,7 +1439,7 @@ static void free_netbuf(void *buf) check_txq_state(lio, skb); - recv_buffer_free((struct sk_buff *)skb); + tx_buffer_free(skb); } /** @@ -1484,7 +1484,7 @@ static void free_netsgbuf(void *buf) check_txq_state(lio, skb); /* mq support: sub-queue state check */ - recv_buffer_free((struct sk_buff *)skb); + tx_buffer_free(skb); } /** @@ -1862,6 +1862,32 @@ liquidio_push_packet(u32 octeon_id, skb->dev = netdev; skb_record_rx_queue(skb, droq->q_no); + if (likely(len > MIN_SKB_SIZE)) { + struct octeon_skb_page_info *pg_info; + unsigned char *va; + + pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + if (pg_info->page) { + /* For Paged allocation use the frags */ + va = page_address(pg_info->page) + + pg_info->page_offset; + memcpy(skb->data, va, MIN_SKB_SIZE); + skb_put(skb, MIN_SKB_SIZE); + skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, + pg_info->page, + pg_info->page_offset + + MIN_SKB_SIZE, + len - MIN_SKB_SIZE, + LIO_RXBUFFER_SZ); + } + } else { + struct octeon_skb_page_info *pg_info = + ((struct octeon_skb_page_info *)(skb->cb)); + skb_copy_to_linear_data(skb, page_address(pg_info->page) + + pg_info->page_offset, len); + skb_put(skb, len); + put_page(pg_info->page); + } if (rh->r_dh.has_hwtstamp) { /* timestamp is included from the hardware at the @@ -2612,7 +2638,7 @@ static void handle_timestamp(struct octeon_device *oct, } octeon_free_soft_command(oct, sc); - recv_buffer_free(skb); + tx_buffer_free(skb); } /* \brief Send a data packet that will be timestamped @@ -3001,7 +3027,7 @@ lio_xmit_failed: iq_no, stats->tx_dropped); dma_unmap_single(&oct->pci_dev->dev, ndata.cmd.dptr, ndata.datasize, DMA_TO_DEVICE); - recv_buffer_free(skb); + tx_buffer_free(skb); return NETDEV_TX_OK; } diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c index 1f648dc..a12beaa 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c @@ -151,22 +151,26 @@ octeon_droq_destroy_ring_buffers(struct octeon_device *oct, struct octeon_droq *droq) { u32 i; + struct octeon_skb_page_info *pg_info; for (i = 0; i < droq->max_count; i++) { - if (droq->recv_buf_list[i].buffer) { - if (droq->desc_ring) { - lio_unmap_ring_info(oct->pci_dev, - (u64)droq-> - desc_ring[i].info_ptr, - OCT_DROQ_INFO_SIZE); - lio_unmap_ring(oct->pci_dev, - (u64)droq->desc_ring[i]. - buffer_ptr, - droq->buffer_size); - } - recv_buffer_free(droq->recv_buf_list[i].buffer); - droq->recv_buf_list[i].buffer = NULL; - } + pg_info = &droq->recv_buf_list[i].pg_info; + + if (pg_info->dma) + lio_unmap_ring(oct->pci_dev, + (u64)pg_info->dma); + pg_info->dma = 0; + + if (pg_info->page) + recv_buffer_destroy(droq->recv_buf_list[i].buffer, + pg_info); + + if (droq->desc_ring && droq->desc_ring[i].info_ptr) + lio_unmap_ring_info(oct->pci_dev, + (u64)droq-> + desc_ring[i].info_ptr, + OCT_DROQ_INFO_SIZE); + droq->recv_buf_list[i].buffer = NULL; } octeon_droq_reset_indices(droq); @@ -181,11 +185,12 @@ octeon_droq_setup_ring_buffers(struct octeon_device *oct, struct octeon_droq_desc *desc_ring = droq->desc_ring; for (i = 0; i < droq->max_count; i++) { - buf = recv_buffer_alloc(oct, droq->q_no, droq->buffer_size); + buf = recv_buffer_alloc(oct, &droq->recv_buf_list[i].pg_info); if (!buf) { dev_err(&oct->pci_dev->dev, "%s buffer alloc failed\n", __func__); + droq->stats.rx_alloc_failure++; return -ENOMEM; } @@ -197,9 +202,7 @@ octeon_droq_setup_ring_buffers(struct octeon_device *oct, /* map ring buffers into memory */ desc_ring[i].info_ptr = lio_map_ring_info(droq, i); desc_ring[i].buffer_ptr = - lio_map_ring(oct->pci_dev, - droq->recv_buf_list[i].buffer, - droq->buffer_size); + lio_map_ring(droq->recv_buf_list[i].buffer); } octeon_droq_reset_indices(droq); @@ -372,6 +375,7 @@ static inline struct octeon_recv_info *octeon_create_recv_info( struct octeon_recv_pkt *recv_pkt; struct octeon_recv_info *recv_info; u32 i, bytes_left; + struct octeon_skb_page_info *pg_info; info = &droq->info_list[idx]; @@ -389,9 +393,14 @@ static inline struct octeon_recv_info *octeon_create_recv_info( bytes_left = (u32)info->length; while (buf_cnt) { - lio_unmap_ring(octeon_dev->pci_dev, - (u64)droq->desc_ring[idx].buffer_ptr, - droq->buffer_size); + { + pg_info = &droq->recv_buf_list[idx].pg_info; + + lio_unmap_ring(octeon_dev->pci_dev, + (u64)pg_info->dma); + pg_info->page = NULL; + pg_info->dma = 0; + } recv_pkt->buffer_size[i] = (bytes_left >= @@ -463,6 +472,7 @@ octeon_droq_refill(struct octeon_device *octeon_dev, struct octeon_droq *droq) void *buf = NULL; u8 *data; u32 desc_refilled = 0; + struct octeon_skb_page_info *pg_info; desc_ring = droq->desc_ring; @@ -472,13 +482,22 @@ octeon_droq_refill(struct octeon_device *octeon_dev, struct octeon_droq *droq) * the buffer, else allocate. */ if (!droq->recv_buf_list[droq->refill_idx].buffer) { - buf = recv_buffer_alloc(octeon_dev, droq->q_no, - droq->buffer_size); + pg_info = + &droq->recv_buf_list[droq->refill_idx].pg_info; + /* Either recycle the existing pages or go for + * new page alloc + */ + if (pg_info->page) + buf = recv_buffer_reuse(octeon_dev, pg_info); + else + buf = recv_buffer_alloc(octeon_dev, pg_info); /* If a buffer could not be allocated, no point in * continuing */ - if (!buf) + if (!buf) { + droq->stats.rx_alloc_failure++; break; + } droq->recv_buf_list[droq->refill_idx].buffer = buf; data = get_rbd(buf); @@ -490,11 +509,8 @@ octeon_droq_refill(struct octeon_device *octeon_dev, struct octeon_droq *droq) droq->recv_buf_list[droq->refill_idx].data = data; desc_ring[droq->refill_idx].buffer_ptr = - lio_map_ring(octeon_dev->pci_dev, - droq->recv_buf_list[droq-> - refill_idx].buffer, - droq->buffer_size); - + lio_map_ring(droq->recv_buf_list[droq-> + refill_idx].buffer); /* Reset any previous values in the length field. */ droq->info_list[droq->refill_idx].length = 0; @@ -600,6 +616,8 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, for (pkt = 0; pkt < pkt_count; pkt++) { u32 pkt_len = 0; struct sk_buff *nicbuf = NULL; + struct octeon_skb_page_info *pg_info; + void *buf; info = &droq->info_list[droq->read_idx]; octeon_swap_8B_data((u64 *)info, 2); @@ -619,7 +637,6 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, rh = &info->rh; total_len += (u32)info->length; - if (OPCODE_SLOW_PATH(rh)) { u32 buf_cnt; @@ -628,50 +645,44 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, droq->refill_count += buf_cnt; } else { if (info->length <= droq->buffer_size) { - lio_unmap_ring(oct->pci_dev, - (u64)droq->desc_ring[ - droq->read_idx].buffer_ptr, - droq->buffer_size); pkt_len = (u32)info->length; nicbuf = droq->recv_buf_list[ droq->read_idx].buffer; + pg_info = &droq->recv_buf_list[ + droq->read_idx].pg_info; + if (recv_buffer_recycle(oct, pg_info)) + pg_info->page = NULL; droq->recv_buf_list[droq->read_idx].buffer = NULL; INCR_INDEX_BY1(droq->read_idx, droq->max_count); - skb_put(nicbuf, pkt_len); droq->refill_count++; } else { - nicbuf = octeon_fast_packet_alloc(oct, droq, - droq->q_no, - (u32) + nicbuf = octeon_fast_packet_alloc((u32) info->length); pkt_len = 0; /* nicbuf allocation can fail. We'll handle it * inside the loop. */ while (pkt_len < info->length) { - int cpy_len; + int cpy_len, idx = droq->read_idx; - cpy_len = ((pkt_len + - droq->buffer_size) > - info->length) ? + cpy_len = ((pkt_len + droq->buffer_size) + > info->length) ? ((u32)info->length - pkt_len) : droq->buffer_size; if (nicbuf) { - lio_unmap_ring(oct->pci_dev, - (u64) - droq->desc_ring - [droq->read_idx]. - buffer_ptr, - droq-> - buffer_size); octeon_fast_packet_next(droq, nicbuf, cpy_len, - droq-> - read_idx - ); + idx); + buf = droq->recv_buf_list[idx]. + buffer; + recv_buffer_fast_free(buf); + droq->recv_buf_list[idx].buffer + = NULL; + } else { + droq->stats.rx_alloc_failure++; } pkt_len += cpy_len; @@ -682,12 +693,13 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, } if (nicbuf) { - if (droq->ops.fptr) + if (droq->ops.fptr) { droq->ops.fptr(oct->octeon_id, - nicbuf, pkt_len, - rh, &droq->napi); - else + nicbuf, pkt_len, + rh, &droq->napi); + } else { recv_buffer_free(nicbuf); + } } } @@ -695,16 +707,16 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, int desc_refilled = octeon_droq_refill(oct, droq); /* Flush the droq descriptor data to memory to be sure - * that when we update the credits the data in memory - * is accurate. - */ + * that when we update the credits the data in memory + * is accurate. + */ wmb(); writel((desc_refilled), droq->pkts_credit_reg); /* make sure mmio write completes */ mmiowb(); } - } /* for ( each packet )... */ + } /* for (each packet)... */ /* Increment refill_count by the number of buffers processed. */ droq->stats.pkts_received += pkt; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h index 7940cce..91c365c 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h @@ -65,6 +65,17 @@ struct octeon_droq_info { #define OCT_DROQ_INFO_SIZE (sizeof(struct octeon_droq_info)) +struct octeon_skb_page_info { + /* DMA address for the page */ + dma_addr_t dma; + + /* Page for the rx dma **/ + struct page *page; + + /** which offset into page */ + unsigned int page_offset; +}; + /** Pointer to data buffer. * Driver keeps a pointer to the data buffer that it made available to * the Octeon device. Since the descriptor ring keeps physical (bus) @@ -77,6 +88,9 @@ struct octeon_recv_buffer { /** Data in the packet buffer. */ u8 *data; + + /** pg_info **/ + struct octeon_skb_page_info pg_info; }; #define OCT_DROQ_RECVBUF_SIZE (sizeof(struct octeon_recv_buffer)) @@ -106,6 +120,10 @@ struct oct_droq_stats { /** Num of Packets dropped due to receive path failures. */ u64 rx_dropped; + + /** Num of failures of recv_buffer_alloc() */ + u64 rx_alloc_failure; + }; #define POLL_EVENT_INTR_ARRIVED 1 diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h index 0a50bac..0267fff 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h @@ -131,14 +131,30 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr); */ void liquidio_set_ethtool_ops(struct net_device *netdev); -static inline void -*recv_buffer_alloc(struct octeon_device *oct __attribute__((unused)), - u32 q_no __attribute__((unused)), u32 size) -{ #define SKB_ADJ_MASK 0x3F #define SKB_ADJ (SKB_ADJ_MASK + 1) - struct sk_buff *skb = dev_alloc_skb(size + SKB_ADJ); +#define MIN_SKB_SIZE 256 /* 8 bytes and more - 8 bytes for PTP */ +#define LIO_RXBUFFER_SZ 2048 + +static inline void +*recv_buffer_alloc(struct octeon_device *oct, + struct octeon_skb_page_info *pg_info) +{ + struct page *page; + struct sk_buff *skb; + struct octeon_skb_page_info *skb_pg_info; + + page = alloc_page(GFP_ATOMIC | __GFP_COLD); + if (unlikely(!page)) + return NULL; + + skb = dev_alloc_skb(MIN_SKB_SIZE + SKB_ADJ); + if (unlikely(!skb)) { + __free_page(page); + pg_info->page = NULL; + return NULL; + } if ((unsigned long)skb->data & SKB_ADJ_MASK) { u32 r = SKB_ADJ - ((unsigned long)skb->data & SKB_ADJ_MASK); @@ -146,11 +162,151 @@ static inline void skb_reserve(skb, r); } + skb_pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + /* Get DMA info */ + pg_info->dma = dma_map_page(&oct->pci_dev->dev, page, 0, + PAGE_SIZE, DMA_FROM_DEVICE); + + /* Mapping failed!! */ + if (dma_mapping_error(&oct->pci_dev->dev, pg_info->dma)) { + __free_page(page); + dev_kfree_skb_any((struct sk_buff *)skb); + pg_info->page = NULL; + return NULL; + } + + pg_info->page = page; + pg_info->page_offset = 0; + skb_pg_info->page = page; + skb_pg_info->page_offset = 0; + skb_pg_info->dma = pg_info->dma; + return (void *)skb; } +static inline void +*recv_buffer_fast_alloc(u32 size) +{ + struct sk_buff *skb; + struct octeon_skb_page_info *skb_pg_info; + + skb = dev_alloc_skb(size + SKB_ADJ); + if (unlikely(!skb)) + return NULL; + + if ((unsigned long)skb->data & SKB_ADJ_MASK) { + u32 r = SKB_ADJ - ((unsigned long)skb->data & SKB_ADJ_MASK); + + skb_reserve(skb, r); + } + + skb_pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + skb_pg_info->page = NULL; + skb_pg_info->page_offset = 0; + skb_pg_info->dma = 0; + + return skb; +} + +static inline int +recv_buffer_recycle(struct octeon_device *oct, void *buf) +{ + struct octeon_skb_page_info *pg_info = buf; + + if (!pg_info->page) { + dev_err(&oct->pci_dev->dev, "%s: pg_info->page NULL\n", + __func__); + return -ENOMEM; + } + + if (unlikely(page_count(pg_info->page) != 1) || + unlikely(page_to_nid(pg_info->page) != numa_node_id())) { + dma_unmap_page(&oct->pci_dev->dev, + pg_info->dma, (PAGE_SIZE << 0), + DMA_FROM_DEVICE); + pg_info->dma = 0; + pg_info->page = NULL; + pg_info->page_offset = 0; + return -ENOMEM; + } + + /* Flip to other half of the buffer */ + if (pg_info->page_offset == 0) + pg_info->page_offset = LIO_RXBUFFER_SZ; + else + pg_info->page_offset = 0; + page_ref_inc(pg_info->page); + + return 0; +} + +static inline void +*recv_buffer_reuse(struct octeon_device *oct, void *buf) +{ + struct octeon_skb_page_info *pg_info = buf, *skb_pg_info; + struct sk_buff *skb; + + skb = dev_alloc_skb(MIN_SKB_SIZE + SKB_ADJ); + if (unlikely(!skb)) { + dma_unmap_page(&oct->pci_dev->dev, + pg_info->dma, (PAGE_SIZE << 0), + DMA_FROM_DEVICE); + return NULL; + } + + if ((unsigned long)skb->data & SKB_ADJ_MASK) { + u32 r = SKB_ADJ - ((unsigned long)skb->data & SKB_ADJ_MASK); + + skb_reserve(skb, r); + } + + skb_pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + skb_pg_info->page = pg_info->page; + skb_pg_info->page_offset = pg_info->page_offset; + skb_pg_info->dma = pg_info->dma; + + return skb; +} + +static inline void +recv_buffer_destroy(void *buffer, struct octeon_skb_page_info *pg_info) +{ + struct sk_buff *skb = (struct sk_buff *)buffer; + + put_page(pg_info->page); + pg_info->dma = 0; + pg_info->page = NULL; + pg_info->page_offset = 0; + + if (skb) + dev_kfree_skb_any(skb); +} + static inline void recv_buffer_free(void *buffer) { + struct sk_buff *skb = (struct sk_buff *)buffer; + struct octeon_skb_page_info *pg_info; + + pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + + if (pg_info->page) { + put_page(pg_info->page); + pg_info->dma = 0; + pg_info->page = NULL; + pg_info->page_offset = 0; + } + + dev_kfree_skb_any((struct sk_buff *)buffer); +} + +static inline void +recv_buffer_fast_free(void *buffer) +{ + dev_kfree_skb_any((struct sk_buff *)buffer); +} + +static inline void tx_buffer_free(void *buffer) +{ dev_kfree_skb_any((struct sk_buff *)buffer); } @@ -159,7 +315,17 @@ static inline void recv_buffer_free(void *buffer) #define lio_dma_free(oct, size, virt_addr, dma_addr) \ dma_free_coherent(&oct->pci_dev->dev, size, virt_addr, dma_addr) -#define get_rbd(ptr) (((struct sk_buff *)(ptr))->data) +static inline +void *get_rbd(struct sk_buff *skb) +{ + struct octeon_skb_page_info *pg_info; + unsigned char *va; + + pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + va = page_address(pg_info->page) + pg_info->page_offset; + + return va; +} static inline u64 lio_map_ring_info(struct octeon_droq *droq, u32 i) @@ -183,33 +349,44 @@ lio_unmap_ring_info(struct pci_dev *pci_dev, } static inline u64 -lio_map_ring(struct pci_dev *pci_dev, - void *buf, u32 size) +lio_map_ring(void *buf) { dma_addr_t dma_addr; - dma_addr = dma_map_single(&pci_dev->dev, get_rbd(buf), size, - DMA_FROM_DEVICE); + struct sk_buff *skb = (struct sk_buff *)buf; + struct octeon_skb_page_info *pg_info; - BUG_ON(dma_mapping_error(&pci_dev->dev, dma_addr)); + pg_info = ((struct octeon_skb_page_info *)(skb->cb)); + if (!pg_info->page) { + pr_err("%s: pg_info->page NULL\n", __func__); + WARN_ON(1); + } + + /* Get DMA info */ + dma_addr = pg_info->dma; + if (!pg_info->dma) { + pr_err("%s: ERROR it should be already available\n", + __func__); + WARN_ON(1); + } + dma_addr += pg_info->page_offset; return (u64)dma_addr; } static inline void lio_unmap_ring(struct pci_dev *pci_dev, - u64 buf_ptr, u32 size) + u64 buf_ptr) + { - dma_unmap_single(&pci_dev->dev, - buf_ptr, size, - DMA_FROM_DEVICE); + dma_unmap_page(&pci_dev->dev, + buf_ptr, (PAGE_SIZE << 0), + DMA_FROM_DEVICE); } -static inline void *octeon_fast_packet_alloc(struct octeon_device *oct, - struct octeon_droq *droq, - u32 q_no, u32 size) +static inline void *octeon_fast_packet_alloc(u32 size) { - return recv_buffer_alloc(oct, q_no, size); + return recv_buffer_fast_alloc(size); } static inline void octeon_fast_packet_next(struct octeon_droq *droq, -- cgit v0.10.2 From a5b3788881289148ac34e6aeeb678fadee644bc9 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 14 Jun 2016 16:54:48 -0700 Subject: liquidio: Consider PTP for packet size calculations This patch is to refactor packet size calculations to support PTP enabled for 66xx and 68xx cards and also other cards that do not support PTP. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 3a4f31f..aa28790 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -84,6 +84,8 @@ static int conf_type; module_param(conf_type, int, 0); MODULE_PARM_DESC(conf_type, "select octeon configuration 0 default 1 ovs"); +static int ptp_enable = 1; + /* Bit mask values for lio->ifstate */ #define LIO_IFSTATE_DROQ_OPS 0x01 #define LIO_IFSTATE_REGISTERED 0x02 @@ -1851,6 +1853,7 @@ liquidio_push_packet(u32 octeon_id, if (netdev) { int packet_was_received; struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; /* Do not proceed if the interface is not in RUNNING state. */ if (!ifstate_check(lio, LIO_IFSTATE_RUNNING)) { @@ -1889,21 +1892,26 @@ liquidio_push_packet(u32 octeon_id, put_page(pg_info->page); } - if (rh->r_dh.has_hwtstamp) { - /* timestamp is included from the hardware at the - * beginning of the packet. - */ - if (ifstate_check(lio, - LIO_IFSTATE_RX_TIMESTAMP_ENABLED)) { - /* Nanoseconds are in the first 64-bits - * of the packet. + if (((oct->chip_id == OCTEON_CN66XX) || + (oct->chip_id == OCTEON_CN68XX)) && + ptp_enable) { + if (rh->r_dh.has_hwtstamp) { + /* timestamp is included from the hardware at + * the beginning of the packet. */ - memcpy(&ns, (skb->data), sizeof(ns)); - shhwtstamps = skb_hwtstamps(skb); - shhwtstamps->hwtstamp = - ns_to_ktime(ns + lio->ptp_adjust); + if (ifstate_check + (lio, LIO_IFSTATE_RX_TIMESTAMP_ENABLED)) { + /* Nanoseconds are in the first 64-bits + * of the packet. + */ + memcpy(&ns, (skb->data), sizeof(ns)); + shhwtstamps = skb_hwtstamps(skb); + shhwtstamps->hwtstamp = + ns_to_ktime(ns + + lio->ptp_adjust); + } + skb_pull(skb, sizeof(ns)); } - skb_pull(skb, sizeof(ns)); } skb->protocol = eth_type_trans(skb, skb->dev); diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 00b3ef5..84ffcae 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -174,9 +174,11 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry, /*------------------------- End Scatter/Gather ---------------------------*/ #define OCTNET_FRM_PTP_HEADER_SIZE 8 -#define OCTNET_FRM_HEADER_SIZE 30 /* PTP timestamp + VLAN + Ethernet */ -#define OCTNET_MIN_FRM_SIZE (64 + OCTNET_FRM_PTP_HEADER_SIZE) +#define OCTNET_FRM_HEADER_SIZE 22 /* VLAN + Ethernet */ + +#define OCTNET_MIN_FRM_SIZE 64 + #define OCTNET_MAX_FRM_SIZE (16000 + OCTNET_FRM_HEADER_SIZE) #define OCTNET_DEFAULT_FRM_SIZE (1500 + OCTNET_FRM_HEADER_SIZE) -- cgit v0.10.2 From 7275ebfc504c068a1250e0fa51f896e493b0edfa Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 14 Jun 2016 16:54:49 -0700 Subject: liquidio: New driver FW command structure This patch is for new driver/firmware control command structure (octnic_packet_params and octnic_cmd_setup ) and resultant code changes. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index aa28790..1f1a28d 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -2704,68 +2704,6 @@ static inline int send_nic_timestamp_pkt(struct octeon_device *oct, return retval; } -static inline int is_ipv4(struct sk_buff *skb) -{ - return (skb->protocol == htons(ETH_P_IP)) && - (ip_hdr(skb)->version == 4); -} - -static inline int is_vlan(struct sk_buff *skb) -{ - return skb->protocol == htons(ETH_P_8021Q); -} - -static inline int is_ip_fragmented(struct sk_buff *skb) -{ - /* The Don't fragment and Reserved flag fields are ignored. - * IP is fragmented if - * - the More fragments bit is set (indicating this IP is a fragment - * with more to follow; the current offset could be 0 ). - * - ths offset field is non-zero. - */ - return (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) ? 1 : 0; -} - -static inline int is_ipv6(struct sk_buff *skb) -{ - return (skb->protocol == htons(ETH_P_IPV6)) && - (ipv6_hdr(skb)->version == 6); -} - -static inline int is_with_extn_hdr(struct sk_buff *skb) -{ - return (ipv6_hdr(skb)->nexthdr != IPPROTO_TCP) && - (ipv6_hdr(skb)->nexthdr != IPPROTO_UDP); -} - -static inline int is_tcpudp(struct sk_buff *skb) -{ - return (ip_hdr(skb)->protocol == IPPROTO_TCP) || - (ip_hdr(skb)->protocol == IPPROTO_UDP); -} - -static inline u32 get_ipv4_5tuple_tag(struct sk_buff *skb) -{ - u32 tag; - struct iphdr *iphdr = ip_hdr(skb); - - tag = crc32(0, &iphdr->protocol, 1); - tag = crc32(tag, (u8 *)&iphdr->saddr, 8); - tag = crc32(tag, skb_transport_header(skb), 4); - return tag; -} - -static inline u32 get_ipv6_5tuple_tag(struct sk_buff *skb) -{ - u32 tag; - struct ipv6hdr *ipv6hdr = ipv6_hdr(skb); - - tag = crc32(0, &ipv6hdr->nexthdr, 1); - tag = crc32(tag, (u8 *)&ipv6hdr->saddr, 32); - tag = crc32(tag, skb_transport_header(skb), 4); - return tag; -} - /** \brief Transmit networks packets to the Octeon interface * @param skbuff skbuff struct to be passed to network layer. * @param netdev pointer to network device @@ -2852,52 +2790,11 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) cmdsetup.u64 = 0; cmdsetup.s.ifidx = lio->linfo.ifidx; + cmdsetup.s.iq_no = iq_no; - if (skb->ip_summed == CHECKSUM_PARTIAL) { - if (is_ipv4(skb) && !is_ip_fragmented(skb) && is_tcpudp(skb)) { - tag = get_ipv4_5tuple_tag(skb); - - cmdsetup.s.cksum_offset = sizeof(struct ethhdr) + 1; - - if (ip_hdr(skb)->ihl > 5) - cmdsetup.s.ipv4opts_ipv6exthdr = - OCT_PKT_PARAM_IPV4OPTS; - - } else if (is_ipv6(skb)) { - tag = get_ipv6_5tuple_tag(skb); + if (skb->ip_summed == CHECKSUM_PARTIAL) + cmdsetup.s.transport_csum = 1; - cmdsetup.s.cksum_offset = sizeof(struct ethhdr) + 1; - - if (is_with_extn_hdr(skb)) - cmdsetup.s.ipv4opts_ipv6exthdr = - OCT_PKT_PARAM_IPV6EXTHDR; - - } else if (is_vlan(skb)) { - if (vlan_eth_hdr(skb)->h_vlan_encapsulated_proto - == htons(ETH_P_IP) && - !is_ip_fragmented(skb) && is_tcpudp(skb)) { - tag = get_ipv4_5tuple_tag(skb); - - cmdsetup.s.cksum_offset = - sizeof(struct vlan_ethhdr) + 1; - - if (ip_hdr(skb)->ihl > 5) - cmdsetup.s.ipv4opts_ipv6exthdr = - OCT_PKT_PARAM_IPV4OPTS; - - } else if (vlan_eth_hdr(skb)->h_vlan_encapsulated_proto - == htons(ETH_P_IPV6)) { - tag = get_ipv6_5tuple_tag(skb); - - cmdsetup.s.cksum_offset = - sizeof(struct vlan_ethhdr) + 1; - - if (is_with_extn_hdr(skb)) - cmdsetup.s.ipv4opts_ipv6exthdr = - OCT_PKT_PARAM_IPV6EXTHDR; - } - } - } if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) { skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; cmdsetup.s.timestamp = 1; diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 84ffcae..ebdb802 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -463,30 +463,27 @@ union octeon_rh { #define OCT_RH_SIZE (sizeof(union octeon_rh)) -#define OCT_PKT_PARAM_IPV4OPTS 1 -#define OCT_PKT_PARAM_IPV6EXTHDR 2 - union octnic_packet_params { u32 u32; struct { #ifdef __BIG_ENDIAN_BITFIELD - u32 reserved:6; + u32 reserved:16; + u32 ip_csum:1; /* Perform IP header checksum(s) */ + /* Perform Outer transport header checksum */ + u32 transport_csum:1; + /* Find tunnel, and perform transport csum. */ u32 tnl_csum:1; - u32 ip_csum:1; - u32 ipv4opts_ipv6exthdr:2; - u32 ipsec_ops:4; - u32 tsflag:1; - u32 csoffset:9; + u32 tsflag:1; /* Timestamp this packet */ + u32 ipsec_ops:4; /* IPsec operation */ u32 ifidx:8; #else u32 ifidx:8; - u32 csoffset:9; - u32 tsflag:1; u32 ipsec_ops:4; - u32 ipv4opts_ipv6exthdr:2; - u32 ip_csum:1; + u32 tsflag:1; u32 tnl_csum:1; - u32 reserved:6; + u32 transport_csum:1; + u32 ip_csum:1; + u32 reserved:16; #endif } s; }; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h index 0238857..a61dea3 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h @@ -94,15 +94,15 @@ struct octnic_data_pkt { */ union octnic_cmd_setup { struct { - u32 ifidx:8; - u32 cksum_offset:7; + u32 iq_no:8; u32 gather:1; u32 timestamp:1; - u32 ipv4opts_ipv6exthdr:2; u32 ip_csum:1; + u32 transport_csum:1; u32 tnl_csum:1; - + u32 ifidx:8; u32 rsvd:11; + union { u32 datasize; u32 gatherptrs; @@ -172,13 +172,8 @@ octnet_prepare_pci_cmd(struct octeon_instr_64B *cmd, packet_params.u32 = 0; - if (setup->s.cksum_offset) { - packet_params.s.csoffset = setup->s.cksum_offset; - packet_params.s.ipv4opts_ipv6exthdr = - setup->s.ipv4opts_ipv6exthdr; - } - packet_params.s.ip_csum = setup->s.ip_csum; + packet_params.s.transport_csum = setup->s.transport_csum; packet_params.s.tnl_csum = setup->s.tnl_csum; packet_params.s.ifidx = setup->s.ifidx; packet_params.s.tsflag = setup->s.timestamp; -- cgit v0.10.2 From 0cece6c5832b7617c90adbdc14f231f2db23dca6 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 14 Jun 2016 16:54:50 -0700 Subject: liquidio: Replace ifidx for FW commands This patch decoupled the firmware side ifidx and host side interface number. It also has some minor name change for linkinfo sturct field. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 2937c802..4523c86 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -127,7 +127,7 @@ static int lio_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) dev_err(&oct->pci_dev->dev, "Unknown link interface reported\n"); } - if (linfo->link.s.status) { + if (linfo->link.s.link_up) { ethtool_cmd_speed_set(ecmd, linfo->link.s.speed); ecmd->duplex = linfo->link.s.duplex; } else { @@ -222,23 +222,20 @@ static int octnet_gpio_access(struct net_device *netdev, int addr, int val) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; int ret = 0; memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = OCTNET_CMD_GPIO_ACCESS; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = addr; - nctrl.ncmd.s.param3 = val; + nctrl.ncmd.s.param1 = addr; + nctrl.ncmd.s.param2 = val; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.wait_time = 100; nctrl.netpndev = (u64)netdev; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; - nparams.resp_order = OCTEON_RESP_ORDERED; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "Failed to configure gpio value\n"); return -EINVAL; @@ -303,9 +300,10 @@ octnet_mdio45_access(struct lio *lio, int op, int loc, int *value) mdio_cmd->mdio_addr = loc; if (op) mdio_cmd->value1 = *value; - mdio_cmd->value2 = lio->linfo.ifidx; octeon_swap_8B_data((u64 *)mdio_cmd, sizeof(struct oct_mdio_cmd) / 8); + sc->iq_no = lio->linfo.txpciq[0].s.q_no; + octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC, OPCODE_NIC_MDIO45, 0, 0, 0); @@ -503,10 +501,10 @@ static void lio_set_msglevel(struct net_device *netdev, u32 msglvl) if ((msglvl ^ lio->msg_enable) & NETIF_MSG_HW) { if (msglvl & NETIF_MSG_HW) liquidio_set_feature(netdev, - OCTNET_CMD_VERBOSE_ENABLE); + OCTNET_CMD_VERBOSE_ENABLE, 0); else liquidio_set_feature(netdev, - OCTNET_CMD_VERBOSE_DISABLE); + OCTNET_CMD_VERBOSE_DISABLE, 0); } lio->msg_enable = msglvl; @@ -950,7 +948,6 @@ static int lio_set_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) struct octeon_device *oct = lio->oct_dev; struct oct_link_info *linfo; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; int ret = 0; /* get the link info */ @@ -978,9 +975,9 @@ static int lio_set_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = OCTNET_CMD_SET_SETTINGS; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.wait_time = 1000; nctrl.netpndev = (u64)netdev; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; /* Passing the parameters sent by ethtool like Speed, Autoneg & Duplex @@ -990,19 +987,17 @@ static int lio_set_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) /* Autoneg ON */ nctrl.ncmd.s.more = OCTNIC_NCMD_PHY_ON | OCTNIC_NCMD_AUTONEG_ON; - nctrl.ncmd.s.param2 = ecmd->advertising; + nctrl.ncmd.s.param1 = ecmd->advertising; } else { /* Autoneg OFF */ nctrl.ncmd.s.more = OCTNIC_NCMD_PHY_ON; - nctrl.ncmd.s.param3 = ecmd->duplex; + nctrl.ncmd.s.param2 = ecmd->duplex; - nctrl.ncmd.s.param2 = ecmd->speed; + nctrl.ncmd.s.param1 = ecmd->speed; } - nparams.resp_order = OCTEON_RESP_ORDERED; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "Failed to set settings\n"); return -1; diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 1f1a28d..4119e70 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -710,7 +710,7 @@ static void start_txq(struct net_device *netdev) { struct lio *lio = GET_LIO(netdev); - if (lio->linfo.link.s.status) { + if (lio->linfo.link.s.link_up) { txqs_start(netdev); return; } @@ -918,7 +918,7 @@ static void print_link_info(struct net_device *netdev) if (atomic_read(&lio->ifstate) & LIO_IFSTATE_REGISTERED) { struct oct_link_info *linfo = &lio->linfo; - if (linfo->link.s.status) { + if (linfo->link.s.link_up) { netif_info(lio, link, lio->netdev, "%d Mbps %s Duplex UP\n", linfo->link.s.speed, (linfo->link.s.duplex) ? "Full" : "Half"); @@ -940,13 +940,15 @@ static inline void update_link_status(struct net_device *netdev, union oct_link_status *ls) { struct lio *lio = GET_LIO(netdev); + int changed = (lio->linfo.link.u64 != ls->u64); - if ((lio->intf_open) && (lio->linfo.link.u64 != ls->u64)) { - lio->linfo.link.u64 = ls->u64; + lio->linfo.link.u64 = ls->u64; + if ((lio->intf_open) && (changed)) { print_link_info(netdev); + lio->link_changes++; - if (lio->linfo.link.s.status) { + if (lio->linfo.link.s.link_up) { netif_carrier_on(netdev); /* start_txq(netdev); */ txqs_wake(netdev); @@ -1219,18 +1221,15 @@ static void octeon_destroy_resources(struct octeon_device *oct) static void send_rx_ctrl_cmd(struct lio *lio, int start_stop) { struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); nctrl.ncmd.s.cmd = OCTNET_CMD_RX_CTL; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = start_stop; + nctrl.ncmd.s.param1 = start_stop; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.netpndev = (u64)lio->netdev; - nparams.resp_order = OCTEON_RESP_NORESPONSE; - - if (octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams) < 0) + if (octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl) < 0) netif_info(lio, rx_err, lio->netdev, "Failed to send RX Control message\n"); } @@ -1269,6 +1268,8 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx) free_netdev(netdev); + oct->props[ifidx].gmxport = -1; + oct->props[ifidx].netdev = NULL; } @@ -1833,21 +1834,21 @@ static u16 select_q(struct net_device *dev, struct sk_buff *skb, * @param len - size of total data received. * @param rh - Control header associated with the packet * @param param - additional control data with the packet + * @param arg - farg registered in droq_ops */ static void liquidio_push_packet(u32 octeon_id, void *skbuff, u32 len, union octeon_rh *rh, - void *param) + void *param, + void *arg) { struct napi_struct *napi = param; - struct octeon_device *oct = lio_get_device(octeon_id); struct sk_buff *skb = (struct sk_buff *)skbuff; struct skb_shared_hwtstamps *shhwtstamps; u64 ns; - struct net_device *netdev = - (struct net_device *)oct->props[rh->r_dh.link].netdev; + struct net_device *netdev = (struct net_device *)arg; struct octeon_droq *droq = container_of(param, struct octeon_droq, napi); if (netdev) { @@ -2043,10 +2044,10 @@ static int liquidio_napi_poll(struct napi_struct *napi, int budget) * are for ingress packets. */ static inline int setup_io_queues(struct octeon_device *octeon_dev, - struct net_device *net_device) + int ifidx) { - static int first_time = 1; - static struct octeon_droq_ops droq_ops; + struct octeon_droq_ops droq_ops; + struct net_device *netdev; static int cpu_id; static int cpu_id_modulus; struct octeon_droq *droq; @@ -2055,18 +2056,19 @@ static inline int setup_io_queues(struct octeon_device *octeon_dev, struct lio *lio; int num_tx_descs; - lio = GET_LIO(net_device); - if (first_time) { - first_time = 0; - memset(&droq_ops, 0, sizeof(struct octeon_droq_ops)); + netdev = octeon_dev->props[ifidx].netdev; + + lio = GET_LIO(netdev); - droq_ops.fptr = liquidio_push_packet; + memset(&droq_ops, 0, sizeof(struct octeon_droq_ops)); - droq_ops.poll_mode = 1; - droq_ops.napi_fn = liquidio_napi_drv_callback; - cpu_id = 0; - cpu_id_modulus = num_present_cpus(); - } + droq_ops.fptr = liquidio_push_packet; + droq_ops.farg = (void *)netdev; + + droq_ops.poll_mode = 1; + droq_ops.napi_fn = liquidio_napi_drv_callback; + cpu_id = 0; + cpu_id_modulus = num_present_cpus(); /* set up DROQs. */ for (q = 0; q < lio->linfo.num_rxpciq; q++) { @@ -2090,7 +2092,11 @@ static inline int setup_io_queues(struct octeon_device *octeon_dev, droq = octeon_dev->droq[q_no]; napi = &droq->napi; - netif_napi_add(net_device, napi, liquidio_napi_poll, 64); + dev_dbg(&octeon_dev->pci_dev->dev, + "netif_napi_add netdev:%llx oct:%llx\n", + (u64)netdev, + (u64)octeon_dev); + netif_napi_add(netdev, napi, liquidio_napi_poll, 64); /* designate a CPU for this droq */ droq->cpu_id = cpu_id; @@ -2106,9 +2112,9 @@ static inline int setup_io_queues(struct octeon_device *octeon_dev, num_tx_descs = CFG_GET_NUM_TX_DESCS_NIC_IF(octeon_get_conf (octeon_dev), lio->ifidx); - retval = octeon_setup_iq(octeon_dev, lio->linfo.txpciq[q], - num_tx_descs, - netdev_get_tx_queue(net_device, q)); + retval = octeon_setup_iq(octeon_dev, ifidx, q, + lio->linfo.txpciq[q], num_tx_descs, + netdev_get_tx_queue(netdev, q)); if (retval) { dev_err(&octeon_dev->pci_dev->dev, " %s : Runtime IQ(TxQ) creation failed.\n", @@ -2206,7 +2212,8 @@ static int liquidio_stop(struct net_device *netdev) netif_info(lio, ifdown, lio->netdev, "Stopping interface!\n"); /* Inform that netif carrier is down */ lio->intf_open = 0; - lio->linfo.link.s.status = 0; + lio->linfo.link.s.link_up = 0; + lio->link_changes++; netif_carrier_off(netdev); @@ -2345,7 +2352,6 @@ static void liquidio_set_mcast_list(struct net_device *netdev) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; struct netdev_hw_addr *ha; u64 *mc; int ret, i; @@ -2356,10 +2362,10 @@ static void liquidio_set_mcast_list(struct net_device *netdev) /* Create a ctrl pkt command to be sent to core app. */ nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = OCTNET_CMD_SET_MULTI_LIST; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = get_new_flags(netdev); - nctrl.ncmd.s.param3 = mc_count; + nctrl.ncmd.s.param1 = get_new_flags(netdev); + nctrl.ncmd.s.param2 = mc_count; nctrl.ncmd.s.more = mc_count; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.netpndev = (u64)netdev; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; @@ -2380,9 +2386,7 @@ static void liquidio_set_mcast_list(struct net_device *netdev) */ nctrl.wait_time = 0; - nparams.resp_order = OCTEON_RESP_NORESPONSE; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "DEVFLAGS change failed in core (ret: 0x%x)\n", ret); @@ -2400,19 +2404,17 @@ static int liquidio_set_mac(struct net_device *netdev, void *p) struct octeon_device *oct = lio->oct_dev; struct sockaddr *addr = (struct sockaddr *)p; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; - if ((!is_valid_ether_addr(addr->sa_data)) || - (ifstate_check(lio, LIO_IFSTATE_RUNNING))) + if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = OCTNET_CMD_CHANGE_MACADDR; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = 0; + nctrl.ncmd.s.param1 = 0; nctrl.ncmd.s.more = 1; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.netpndev = (u64)netdev; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; nctrl.wait_time = 100; @@ -2421,9 +2423,7 @@ static int liquidio_set_mac(struct net_device *netdev, void *p) /* The MAC Address is presented in network byte order. */ memcpy((u8 *)&nctrl.udd[0] + 2, addr->sa_data, ETH_ALEN); - nparams.resp_order = OCTEON_RESP_ORDERED; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "MAC Address change failed\n"); return -ENOMEM; @@ -2493,7 +2493,6 @@ static int liquidio_change_mtu(struct net_device *netdev, int new_mtu) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; int max_frm_size = new_mtu + OCTNET_FRM_HEADER_SIZE; int ret = 0; @@ -2513,15 +2512,13 @@ static int liquidio_change_mtu(struct net_device *netdev, int new_mtu) nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = OCTNET_CMD_CHANGE_MTU; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = new_mtu; + nctrl.ncmd.s.param1 = new_mtu; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.wait_time = 100; nctrl.netpndev = (u64)netdev; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; - nparams.resp_order = OCTEON_RESP_ORDERED; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "Failed to set MTU\n"); return -1; @@ -2742,11 +2739,11 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) * transmitted. */ if (!(atomic_read(&lio->ifstate) & LIO_IFSTATE_RUNNING) || - (!lio->linfo.link.s.status) || + (!lio->linfo.link.s.link_up) || (skb->len <= 0)) { netif_info(lio, tx_err, lio->netdev, "Transmit failed link_status : %d\n", - lio->linfo.link.s.status); + lio->linfo.link.s.link_up); goto lio_xmit_failed; } @@ -2789,7 +2786,6 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) ndata.datasize = skb->len; cmdsetup.u64 = 0; - cmdsetup.s.ifidx = lio->linfo.ifidx; cmdsetup.s.iq_no = iq_no; if (skb->ip_summed == CHECKSUM_PARTIAL) @@ -2802,7 +2798,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) if (skb_shinfo(skb)->nr_frags == 0) { cmdsetup.s.u.datasize = skb->len; - octnet_prepare_pci_cmd(&ndata.cmd, &cmdsetup, tag); + octnet_prepare_pci_cmd(oct, &ndata.cmd, &cmdsetup, tag); /* Offload checksum calculation for TCP/UDP packets */ ndata.cmd.dptr = dma_map_single(&oct->pci_dev->dev, skb->data, @@ -2836,7 +2832,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) cmdsetup.s.gather = 1; cmdsetup.s.u.gatherptrs = (skb_shinfo(skb)->nr_frags + 1); - octnet_prepare_pci_cmd(&ndata.cmd, &cmdsetup, tag); + octnet_prepare_pci_cmd(oct, &ndata.cmd, &cmdsetup, tag); memset(g->sg, 0, g->sg_size); @@ -2952,27 +2948,24 @@ static void liquidio_tx_timeout(struct net_device *netdev) txqs_wake(netdev); } -int liquidio_set_feature(struct net_device *netdev, int cmd) +int liquidio_set_feature(struct net_device *netdev, int cmd, u16 param1) { struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; struct octnic_ctrl_pkt nctrl; - struct octnic_ctrl_params nparams; int ret = 0; memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); nctrl.ncmd.u64 = 0; nctrl.ncmd.s.cmd = cmd; - nctrl.ncmd.s.param1 = lio->linfo.ifidx; - nctrl.ncmd.s.param2 = OCTNIC_LROIPV4 | OCTNIC_LROIPV6; + nctrl.ncmd.s.param1 = param1; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; nctrl.wait_time = 100; nctrl.netpndev = (u64)netdev; nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; - nparams.resp_order = OCTEON_RESP_NORESPONSE; - - ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl, nparams); + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); if (ret < 0) { dev_err(&oct->pci_dev->dev, "Feature change failed in core (ret: 0x%x)\n", ret); @@ -3028,10 +3021,12 @@ static int liquidio_set_features(struct net_device *netdev, return 0; if ((features & NETIF_F_LRO) && (lio->dev_capability & NETIF_F_LRO)) - liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE); + liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE, + OCTNIC_LROIPV4 | OCTNIC_LROIPV6); else if (!(features & NETIF_F_LRO) && (lio->dev_capability & NETIF_F_LRO)) - liquidio_set_feature(netdev, OCTNET_CMD_LRO_DISABLE); + liquidio_set_feature(netdev, OCTNET_CMD_LRO_DISABLE, + OCTNIC_LROIPV4 | OCTNIC_LROIPV6); return 0; } @@ -3102,24 +3097,27 @@ static int lio_nic_info(struct octeon_recv_info *recv_info, void *buf) { struct octeon_device *oct = (struct octeon_device *)buf; struct octeon_recv_pkt *recv_pkt = recv_info->recv_pkt; - int ifidx = 0; + int gmxport = 0; union oct_link_status *ls; int i; - if ((recv_pkt->buffer_size[0] != sizeof(*ls)) || - (recv_pkt->rh.r_nic_info.ifidx > oct->ifcount)) { + if (recv_pkt->buffer_size[0] != sizeof(*ls)) { dev_err(&oct->pci_dev->dev, "Malformed NIC_INFO, len=%d, ifidx=%d\n", recv_pkt->buffer_size[0], - recv_pkt->rh.r_nic_info.ifidx); + recv_pkt->rh.r_nic_info.gmxport); goto nic_info_err; } - ifidx = recv_pkt->rh.r_nic_info.ifidx; + gmxport = recv_pkt->rh.r_nic_info.gmxport; ls = (union oct_link_status *)get_rbd(recv_pkt->buffer_ptr[0]); octeon_swap_8B_data((u64 *)ls, (sizeof(union oct_link_status)) >> 3); - - update_link_status(oct->props[ifidx].netdev, ls); + for (i = 0; i < oct->ifcount; i++) { + if (oct->props[i].gmxport == gmxport) { + update_link_status(oct->props[i].netdev, ls); + break; + } + } nic_info_err: for (i = 0; i < recv_pkt->buffer_count; i++) @@ -3146,12 +3144,12 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) struct liquidio_if_cfg_resp *resp; struct octdev_props *props; int retval, num_iqueues, num_oqueues; - u64 q_mask; int num_cpus = num_online_cpus(); union oct_nic_if_cfg if_cfg; unsigned int base_queue; unsigned int gmx_port_id; u32 resp_size, ctx_size; + u32 ifidx_or_pfnum; /* This is to handle link status changes */ octeon_register_dispatch_fn(octeon_dev, OPCODE_NIC, @@ -3187,13 +3185,14 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) CFG_GET_BASE_QUE_NIC_IF(octeon_get_conf(octeon_dev), i); gmx_port_id = CFG_GET_GMXID_NIC_IF(octeon_get_conf(octeon_dev), i); + ifidx_or_pfnum = i; if (num_iqueues > num_cpus) num_iqueues = num_cpus; if (num_oqueues > num_cpus) num_oqueues = num_cpus; dev_dbg(&octeon_dev->pci_dev->dev, "requesting config for interface %d, iqs %d, oqs %d\n", - i, num_iqueues, num_oqueues); + ifidx_or_pfnum, num_iqueues, num_oqueues); ACCESS_ONCE(ctx->cond) = 0; ctx->octeon_id = lio_get_device_id(octeon_dev); init_waitqueue_head(&ctx->wc); @@ -3203,8 +3202,11 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) if_cfg.s.num_oqueues = num_oqueues; if_cfg.s.base_queue = base_queue; if_cfg.s.gmx_port_id = gmx_port_id; + + sc->iq_no = 0; + octeon_prepare_soft_command(octeon_dev, sc, OPCODE_NIC, - OPCODE_NIC_IF_CFG, i, + OPCODE_NIC_IF_CFG, 0, if_cfg.u64, 0); sc->callback = if_cfg_callback; @@ -3254,8 +3256,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) goto setup_nic_dev_fail; } - props = &octeon_dev->props[i]; - props->netdev = netdev; + SET_NETDEV_DEV(netdev, &octeon_dev->pci_dev->dev); if (num_iqueues > 1) lionetdevops.ndo_select_queue = select_q; @@ -3269,18 +3270,18 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) memset(lio, 0, sizeof(struct lio)); - lio->linfo.ifidx = resp->cfg_info.ifidx; - lio->ifidx = resp->cfg_info.ifidx; + lio->ifidx = ifidx_or_pfnum; + + props = &octeon_dev->props[i]; + props->gmxport = resp->cfg_info.linfo.gmxport; + props->netdev = netdev; lio->linfo.num_rxpciq = num_oqueues; lio->linfo.num_txpciq = num_iqueues; - q_mask = resp->cfg_info.oqmask; - /* q_mask is 0-based and already verified mask is nonzero */ for (j = 0; j < num_oqueues; j++) { lio->linfo.rxpciq[j].u64 = resp->cfg_info.linfo.rxpciq[j].u64; } - q_mask = resp->cfg_info.iqmask; for (j = 0; j < num_iqueues; j++) { lio->linfo.txpciq[j].u64 = resp->cfg_info.linfo.txpciq[j].u64; @@ -3292,13 +3293,15 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) lio->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE); lio->dev_capability = NETIF_F_HIGHDMA - | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM - | NETIF_F_SG | NETIF_F_RXCSUM - | NETIF_F_TSO | NETIF_F_TSO6 - | NETIF_F_LRO; + | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM + | NETIF_F_SG | NETIF_F_RXCSUM + | NETIF_F_GRO + | NETIF_F_TSO | NETIF_F_TSO6 + | NETIF_F_LRO; netif_set_gso_max_size(netdev, OCTNIC_GSO_MAX_SIZE); - netdev->features = lio->dev_capability; + netdev->features = (lio->dev_capability & ~NETIF_F_LRO); + netdev->vlan_features = lio->dev_capability; netdev->hw_features = lio->dev_capability; @@ -3328,7 +3331,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) */ lio->txq = lio->linfo.txpciq[0].s.q_no; lio->rxq = lio->linfo.rxpciq[0].s.q_no; - if (setup_io_queues(octeon_dev, netdev)) { + if (setup_io_queues(octeon_dev, i)) { dev_err(&octeon_dev->pci_dev->dev, "I/O queues creation failed\n"); goto setup_nic_dev_fail; } @@ -3347,10 +3350,13 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) /* Register ethtool support */ liquidio_set_ethtool_ops(netdev); - liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE); + if (netdev->features & NETIF_F_LRO) + liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE, + OCTNIC_LROIPV4 | OCTNIC_LROIPV6); if ((debug != -1) && (debug & NETIF_MSG_HW)) - liquidio_set_feature(netdev, OCTNET_CMD_VERBOSE_ENABLE); + liquidio_set_feature(netdev, OCTNET_CMD_VERBOSE_ENABLE, + 0); /* Register the network device with the OS */ if (register_netdev(netdev)) { @@ -3362,13 +3368,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) "Setup NIC ifidx:%d mac:%02x%02x%02x%02x%02x%02x\n", i, mac[0], mac[1], mac[2], mac[3], mac[4], mac[5]); netif_carrier_off(netdev); - - if (lio->linfo.link.s.status) { - netif_carrier_on(netdev); - start_txq(netdev); - } else { - netif_carrier_off(netdev); - } + lio->link_changes++; ifstate_set(lio, LIO_IFSTATE_REGISTERED); @@ -3402,7 +3402,7 @@ setup_nic_dev_fail: static int liquidio_init_nic_module(struct octeon_device *oct) { struct oct_intrmod_cfg *intrmod_cfg; - int retval = 0; + int i, retval = 0; int num_nic_ports = CFG_GET_NUM_NIC_PORTS(octeon_get_conf(oct)); dev_dbg(&oct->pci_dev->dev, "Initializing network interfaces\n"); @@ -3416,6 +3416,9 @@ static int liquidio_init_nic_module(struct octeon_device *oct) memset(oct->props, 0, sizeof(struct octdev_props) * num_nic_ports); + for (i = 0; i < MAX_OCTEON_LINKS; i++) + oct->props[i].gmxport = -1; + retval = setup_nic_devices(oct); if (retval) { dev_err(&oct->pci_dev->dev, "Setup NIC devices failed\n"); diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index ebdb802..9917576 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -260,19 +260,19 @@ union octnet_cmd { u64 more:6; /* How many udd words follow the command */ - u64 param1:29; + u64 reserved:29; - u64 param2:16; + u64 param1:16; - u64 param3:8; + u64 param2:8; #else - u64 param3:8; + u64 param2:8; - u64 param2:16; + u64 param1:16; - u64 param1:29; + u64 reserved:29; u64 more:6; @@ -414,10 +414,9 @@ union octeon_rh { u64 opcode:4; u64 subcode:8; u64 len:3; /** additional 64-bit words */ - u64 rid:13; - u64 reserved:4; + u64 reserved:8; u64 extra:25; - u64 ifidx:7; + u64 gmxport:16; } r_nic_info; #else u64 u64; @@ -450,10 +449,9 @@ union octeon_rh { u64 opcode:4; } r_core_drv_init; struct { - u64 ifidx:7; + u64 gmxport:16; u64 extra:25; - u64 reserved:4; - u64 rid:13; + u64 reserved:8; u64 len:3; /** additional 64-bit words */ u64 subcode:8; u64 opcode:4; @@ -467,7 +465,7 @@ union octnic_packet_params { u32 u32; struct { #ifdef __BIG_ENDIAN_BITFIELD - u32 reserved:16; + u32 reserved:24; u32 ip_csum:1; /* Perform IP header checksum(s) */ /* Perform Outer transport header checksum */ u32 transport_csum:1; @@ -475,15 +473,13 @@ union octnic_packet_params { u32 tnl_csum:1; u32 tsflag:1; /* Timestamp this packet */ u32 ipsec_ops:4; /* IPsec operation */ - u32 ifidx:8; #else - u32 ifidx:8; u32 ipsec_ops:4; u32 tsflag:1; u32 tnl_csum:1; u32 transport_csum:1; u32 ip_csum:1; - u32 reserved:16; + u32 reserved:24; #endif } s; }; @@ -495,21 +491,21 @@ union oct_link_status { struct { #ifdef __BIG_ENDIAN_BITFIELD u64 duplex:8; - u64 status:8; u64 mtu:16; u64 speed:16; + u64 link_up:1; u64 autoneg:1; u64 interface:4; u64 pause:1; - u64 reserved:10; + u64 reserved:17; #else - u64 reserved:10; + u64 reserved:17; u64 pause:1; u64 interface:4; u64 autoneg:1; + u64 link_up:1; u64 speed:16; u64 mtu:16; - u64 status:8; u64 duplex:8; #endif } s; @@ -561,17 +557,15 @@ struct oct_link_info { u64 hw_addr; #ifdef __BIG_ENDIAN_BITFIELD - u16 gmxport; - u8 rsvd[3]; - u8 num_txpciq; - u8 num_rxpciq; - u8 ifidx; + u64 gmxport:16; + u64 rsvd:32; + u64 num_txpciq:8; + u64 num_rxpciq:8; #else - u8 ifidx; - u8 num_rxpciq; - u8 num_txpciq; - u8 rsvd[3]; - u16 gmxport; + u64 num_rxpciq:8; + u64 num_txpciq:8; + u64 rsvd:32; + u64 gmxport:16; #endif union oct_txpciq txpciq[MAX_IOQS_PER_NICIF]; @@ -581,7 +575,6 @@ struct oct_link_info { #define OCT_LINK_INFO_SIZE (sizeof(struct oct_link_info)) struct liquidio_if_cfg_info { - u64 ifidx; u64 iqmask; /** mask for IQs enabled for the port */ u64 oqmask; /** mask for OQs enabled for the port */ struct oct_link_info linfo; /** initial link information */ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index c06807d..3290009 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -766,7 +766,9 @@ int octeon_setup_instr_queues(struct octeon_device *oct) if (!oct->instr_queue[0]) return 1; memset(oct->instr_queue[0], 0, sizeof(struct octeon_instr_queue)); + oct->instr_queue[0]->q_index = 0; oct->instr_queue[0]->app_ctx = (void *)(size_t)0; + oct->instr_queue[0]->ifidx = 0; txpciq.u64 = 0; txpciq.s.q_no = iq_no; txpciq.s.use_qpg = 0; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h index 36e1f85..0950b94 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h @@ -267,6 +267,7 @@ struct octdev_props { /* Each interface in the Octeon device has a network * device pointer (used for OS specific calls). */ + int gmxport; struct net_device *netdev; }; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c index a12beaa..59a5293 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c @@ -696,7 +696,8 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, if (droq->ops.fptr) { droq->ops.fptr(oct->octeon_id, nicbuf, pkt_len, - rh, &droq->napi); + rh, &droq->napi, + droq->ops.farg); } else { recv_buffer_free(nicbuf); } @@ -963,6 +964,7 @@ int octeon_unregister_droq_ops(struct octeon_device *oct, u32 q_no) spin_lock_irqsave(&droq->lock, flags); droq->ops.fptr = NULL; + droq->ops.farg = NULL; droq->ops.drop_on_max = 0; spin_unlock_irqrestore(&droq->lock, flags); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h index 91c365c..1ca9c4f 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h @@ -231,7 +231,8 @@ struct octeon_droq_ops { * data in the buffer. The receive header gives the port * number to the caller. Function pointer is set by caller. */ - void (*fptr)(u32, void *, u32, union octeon_rh *, void *); + void (*fptr)(u32, void *, u32, union octeon_rh *, void *, void *); + void *farg; /* This function will be called by the driver for all NAPI related * events. The first param is the octeon id. The second param is the diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h index 658f1d0..14c5d70 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h @@ -147,6 +147,13 @@ struct octeon_instr_queue { /** Application context */ void *app_ctx; + + /* network stack queue index */ + int q_index; + + /*os ifidx associated with this queue */ + int ifidx; + }; /*---------------------- INSTRUCTION FORMAT ----------------------------*/ @@ -314,7 +321,8 @@ void octeon_prepare_soft_command(struct octeon_device *oct, int octeon_send_soft_command(struct octeon_device *oct, struct octeon_soft_command *sc); -int octeon_setup_iq(struct octeon_device *oct, union oct_txpciq, - u32 num_descs, void *app_ctx); +int octeon_setup_iq(struct octeon_device *oct, int ifidx, + int q_index, union oct_txpciq iq_no, u32 num_descs, + void *app_ctx); #endif /* __OCTEON_IQ_H__ */ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h index 0267fff..9c14484 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h @@ -67,6 +67,9 @@ struct lio { /** Link information sent by the core application for this interface. */ struct oct_link_info linfo; + /** counter of link changes */ + u64 link_changes; + /** Size of Tx queue for this octeon device. */ u32 tx_qsize; @@ -111,8 +114,9 @@ struct lio { * \brief Enable or disable feature * @param netdev pointer to network device * @param cmd Command that just requires acknowledgment + * @param param1 Parameter to command */ -int liquidio_set_feature(struct net_device *netdev, int cmd); +int liquidio_set_feature(struct net_device *netdev, int cmd, u16 param1); /** * \brief Link control command completion callback diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c index aacabe4..20e0122 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c @@ -119,8 +119,7 @@ static void octnet_link_ctrl_callback(struct octeon_device *oct, static inline struct octeon_soft_command *octnic_alloc_ctrl_pkt_sc(struct octeon_device *oct, - struct octnic_ctrl_pkt *nctrl, - struct octnic_ctrl_params nparams) + struct octnic_ctrl_pkt *nctrl) { struct octeon_soft_command *sc = NULL; u8 *data; @@ -143,7 +142,7 @@ static inline struct octeon_soft_command data = (u8 *)sc->virtdptr; - memcpy(data, &nctrl->ncmd, OCTNET_CMD_SIZE); + memcpy(data, &nctrl->ncmd, OCTNET_CMD_SIZE); octeon_swap_8B_data((u64 *)data, (OCTNET_CMD_SIZE >> 3)); @@ -152,6 +151,8 @@ static inline struct octeon_soft_command memcpy(data + OCTNET_CMD_SIZE, nctrl->udd, uddsize); } + sc->iq_no = (u32)nctrl->iq_no; + octeon_prepare_soft_command(oct, sc, OPCODE_NIC, OPCODE_NIC_CMD, 0, 0, 0); @@ -164,13 +165,12 @@ static inline struct octeon_soft_command int octnet_send_nic_ctrl_pkt(struct octeon_device *oct, - struct octnic_ctrl_pkt *nctrl, - struct octnic_ctrl_params nparams) + struct octnic_ctrl_pkt *nctrl) { int retval; struct octeon_soft_command *sc = NULL; - sc = octnic_alloc_ctrl_pkt_sc(oct, nctrl, nparams); + sc = octnic_alloc_ctrl_pkt_sc(oct, nctrl); if (!sc) { dev_err(&oct->pci_dev->dev, "%s soft command alloc failed\n", __func__); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h index a61dea3..5e53cc9 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h @@ -52,6 +52,9 @@ struct octnic_ctrl_pkt { /** Additional data that may be needed by some commands. */ u64 udd[MAX_NCTRL_UDD]; + /** Input queue to use to send this command. */ + u64 iq_no; + /** Time to wait for Octeon software to respond to this control command. * If wait_time is 0, OSI assumes no response is expected. */ @@ -100,8 +103,7 @@ union octnic_cmd_setup { u32 ip_csum:1; u32 transport_csum:1; u32 tnl_csum:1; - u32 ifidx:8; - u32 rsvd:11; + u32 rsvd:19; union { u32 datasize; @@ -113,10 +115,6 @@ union octnic_cmd_setup { }; -struct octnic_ctrl_params { - u32 resp_order; -}; - static inline int octnet_iq_is_full(struct octeon_device *oct, u32 q_no) { return ((u32)atomic_read(&oct->instr_queue[q_no]->instr_pending) @@ -131,12 +129,13 @@ static inline int octnet_iq_is_full(struct octeon_device *oct, u32 q_no) * Assumes the cmd instruction is pre-allocated, but no fields are filled in. */ static inline void -octnet_prepare_pci_cmd(struct octeon_instr_64B *cmd, +octnet_prepare_pci_cmd(struct octeon_device *oct, struct octeon_instr_64B *cmd, union octnic_cmd_setup *setup, u32 tag) { struct octeon_instr_ih *ih; struct octeon_instr_irh *irh; union octnic_packet_params packet_params; + int port; memset(cmd, 0, sizeof(struct octeon_instr_64B)); @@ -150,13 +149,15 @@ octnet_prepare_pci_cmd(struct octeon_instr_64B *cmd, ih->tagtype = ORDERED_TAG; ih->grp = DEFAULT_POW_GRP; + port = (int)oct->instr_queue[setup->s.iq_no]->txpciq.s.port; + if (tag) ih->tag = tag; else - ih->tag = LIO_DATA(setup->s.ifidx); + ih->tag = LIO_DATA(port); ih->raw = 1; - ih->qos = (setup->s.ifidx & 3) + 4; /* map qos based on interface */ + ih->qos = (port & 3) + 4; /* map qos based on interface */ if (!setup->s.gather) { ih->dlengsz = setup->s.u.datasize; @@ -175,7 +176,6 @@ octnet_prepare_pci_cmd(struct octeon_instr_64B *cmd, packet_params.s.ip_csum = setup->s.ip_csum; packet_params.s.transport_csum = setup->s.transport_csum; packet_params.s.tnl_csum = setup->s.tnl_csum; - packet_params.s.ifidx = setup->s.ifidx; packet_params.s.tsflag = setup->s.timestamp; irh->ossp = packet_params.u32; @@ -216,7 +216,6 @@ int octnet_send_nic_data_pkt(struct octeon_device *oct, */ int octnet_send_nic_ctrl_pkt(struct octeon_device *oct, - struct octnic_ctrl_pkt *nctrl, - struct octnic_ctrl_params nparams); + struct octnic_ctrl_pkt *nctrl); #endif diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index 1240461..b0a8d4d 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -202,6 +202,8 @@ int octeon_delete_instr_queue(struct octeon_device *oct, u32 iq_no) /* Return 0 on success, 1 on failure */ int octeon_setup_iq(struct octeon_device *oct, + int ifidx, + int q_index, union oct_txpciq txpciq, u32 num_descs, void *app_ctx) @@ -227,7 +229,10 @@ int octeon_setup_iq(struct octeon_device *oct, memset(oct->instr_queue[iq_no], 0, sizeof(struct octeon_instr_queue)); + oct->instr_queue[iq_no]->q_index = q_index; oct->instr_queue[iq_no]->app_ctx = app_ctx; + oct->instr_queue[iq_no]->ifidx = ifidx; + if (octeon_init_instr_queue(oct, txpciq, num_descs)) { vfree(oct->instr_queue[iq_no]); oct->instr_queue[iq_no] = NULL; -- cgit v0.10.2 From 6a885b60dad25bb687fe779fafe90a24886022f8 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 14 Jun 2016 16:54:51 -0700 Subject: liquidio: Introduce new octeon2/3 header Added support for new instruction header for octeon2/octeon3(ih) and corresponding changes. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 4119e70..d0ab97c 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -2658,10 +2658,9 @@ static inline int send_nic_timestamp_pkt(struct octeon_device *oct, { int retval; struct octeon_soft_command *sc; - struct octeon_instr_ih *ih; - struct octeon_instr_rdp *rdp; struct lio *lio; int ring_doorbell; + u32 len; lio = finfo->lio; @@ -2683,12 +2682,11 @@ static inline int send_nic_timestamp_pkt(struct octeon_device *oct, sc->callback_arg = finfo->skb; sc->iq_no = ndata->q_no; - ih = (struct octeon_instr_ih *)&sc->cmd.ih; - rdp = (struct octeon_instr_rdp *)&sc->cmd.rdp; + len = (u32)((struct octeon_instr_ih2 *)(&sc->cmd.cmd2.ih2))->dlengsz; ring_doorbell = !xmit_more; retval = octeon_send_command(oct, sc->iq_no, ring_doorbell, &sc->cmd, - sc, ih->dlengsz, ndata->reqtype); + sc, len, ndata->reqtype); if (retval == IQ_SEND_FAILED) { dev_err(&oct->pci_dev->dev, "timestamp data packet failed status: %x\n", @@ -2715,6 +2713,8 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) struct octnic_data_pkt ndata; struct octeon_device *oct; struct oct_iq_stats *stats; + struct octeon_instr_irh *irh; + union tx_info *tx_info; int status = 0; int q_idx = 0, iq_no = 0; int xmit_more, j; @@ -2800,18 +2800,18 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) cmdsetup.s.u.datasize = skb->len; octnet_prepare_pci_cmd(oct, &ndata.cmd, &cmdsetup, tag); /* Offload checksum calculation for TCP/UDP packets */ - ndata.cmd.dptr = dma_map_single(&oct->pci_dev->dev, - skb->data, - skb->len, - DMA_TO_DEVICE); - if (dma_mapping_error(&oct->pci_dev->dev, ndata.cmd.dptr)) { + dptr = dma_map_single(&oct->pci_dev->dev, + skb->data, + skb->len, + DMA_TO_DEVICE); + if (dma_mapping_error(&oct->pci_dev->dev, dptr)) { dev_err(&oct->pci_dev->dev, "%s DMA mapping error 1\n", __func__); return NETDEV_TX_BUSY; } - finfo->dptr = ndata.cmd.dptr; - + ndata.cmd.cmd2.dptr = dptr; + finfo->dptr = dptr; ndata.reqtype = REQTYPE_NORESP_NET; } else { @@ -2885,18 +2885,17 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) g->sg_size, DMA_TO_DEVICE); dptr = g->sg_dma_ptr; - finfo->dptr = ndata.cmd.dptr; + ndata.cmd.cmd2.dptr = dptr; + finfo->dptr = dptr; finfo->g = g; ndata.reqtype = REQTYPE_NORESP_NET_SG; } - if (skb_shinfo(skb)->gso_size) { - struct octeon_instr_irh *irh = - (struct octeon_instr_irh *)&ndata.cmd.irh; - union tx_info *tx_info = (union tx_info *)&ndata.cmd.ossp[0]; + irh = (struct octeon_instr_irh *)&ndata.cmd.cmd2.irh; + tx_info = (union tx_info *)&ndata.cmd.cmd2.ossp[0]; - irh->len = 1; /* to indicate that ossp[0] contains tx_info */ + if (skb_shinfo(skb)->gso_size) { tx_info->s.gso_size = skb_shinfo(skb)->gso_size; tx_info->s.gso_segs = skb_shinfo(skb)->gso_segs; } @@ -2926,8 +2925,9 @@ lio_xmit_failed: stats->tx_dropped++; netif_info(lio, tx_err, lio->netdev, "IQ%d Transmit dropped:%llu\n", iq_no, stats->tx_dropped); - dma_unmap_single(&oct->pci_dev->dev, ndata.cmd.dptr, - ndata.datasize, DMA_TO_DEVICE); + if (dptr) + dma_unmap_single(&oct->pci_dev->dev, dptr, + ndata.datasize, DMA_TO_DEVICE); tx_buffer_free(skb); return NETDEV_TX_OK; } diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 9917576..2179691 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -285,8 +285,140 @@ union octnet_cmd { #define OCTNET_CMD_SIZE (sizeof(union octnet_cmd)) +/* Instruction Header (DPI - CN23xx) - for OCTEON-III models */ +struct octeon_instr_ih3 { +#ifdef __BIG_ENDIAN_BITFIELD + + /** Reserved3 */ + u64 reserved3:1; + + /** Gather indicator 1=gather*/ + u64 gather:1; + + /** Data length OR no. of entries in gather list */ + u64 dlengsz:14; + + /** Front Data size */ + u64 fsz:6; + + /** Reserved2 */ + u64 reserved2:4; + + /** PKI port kind - PKIND */ + u64 pkind:6; + + /** Reserved1 */ + u64 reserved1:32; + +#else + /** Reserved1 */ + u64 reserved1:32; + + /** PKI port kind - PKIND */ + u64 pkind:6; + + /** Reserved2 */ + u64 reserved2:4; + + /** Front Data size */ + u64 fsz:6; + + /** Data length OR no. of entries in gather list */ + u64 dlengsz:14; + + /** Gather indicator 1=gather*/ + u64 gather:1; + + /** Reserved3 */ + u64 reserved3:1; + +#endif +}; + +/* Optional PKI Instruction Header(PKI IH) - for OCTEON CN23XX models */ +/** BIG ENDIAN format. */ +struct octeon_instr_pki_ih3 { +#ifdef __BIG_ENDIAN_BITFIELD + + /** Wider bit */ + u64 w:1; + + /** Raw mode indicator 1 = RAW */ + u64 raw:1; + + /** Use Tag */ + u64 utag:1; + + /** Use QPG */ + u64 uqpg:1; + + /** Reserved2 */ + u64 reserved2:1; + + /** Parse Mode */ + u64 pm:3; + + /** Skip Length */ + u64 sl:8; + + /** Use Tag Type */ + u64 utt:1; + + /** Tag type */ + u64 tagtype:2; + + /** Reserved1 */ + u64 reserved1:2; + + /** QPG Value */ + u64 qpg:11; + + /** Tag Value */ + u64 tag:32; + +#else + + /** Tag Value */ + u64 tag:32; + + /** QPG Value */ + u64 qpg:11; + + /** Reserved1 */ + u64 reserved1:2; + + /** Tag type */ + u64 tagtype:2; + + /** Use Tag Type */ + u64 utt:1; + + /** Skip Length */ + u64 sl:8; + + /** Parse Mode */ + u64 pm:3; + + /** Reserved2 */ + u64 reserved2:1; + + /** Use QPG */ + u64 uqpg:1; + + /** Use Tag */ + u64 utag:1; + + /** Raw mode indicator 1 = RAW */ + u64 raw:1; + + /** Wider bit */ + u64 w:1; +#endif + +}; + /** Instruction Header */ -struct octeon_instr_ih { +struct octeon_instr_ih2 { #ifdef __BIG_ENDIAN_BITFIELD /** Raw mode indicator 1 = RAW */ u64 raw:1; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h index 14c5d70..513f8a0 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h @@ -75,6 +75,8 @@ struct oct_iq_stats { * a Octeon device has one such structure to represent it. */ struct octeon_instr_queue { + struct octeon_device *oct_dev; + /** A spinlock to protect access to the input ring. */ spinlock_t lock; @@ -183,12 +185,12 @@ struct octeon_instr_32B { /** 64-byte instruction format. * Format of instruction for a 64-byte mode input queue. */ -struct octeon_instr_64B { +struct octeon_instr2_64B { /** Pointer where the input data is available. */ u64 dptr; /** Instruction Header. */ - u64 ih; + u64 ih2; /** Input Request Header. */ u64 irh; @@ -205,10 +207,40 @@ struct octeon_instr_64B { u64 rptr; u64 reserved; +}; + +struct octeon_instr3_64B { + /** Pointer where the input data is available. */ + u64 dptr; + + /** Instruction Header. */ + u64 ih3; + + /** Instruction Header. */ + u64 pki_ih3; + + /** Input Request Header. */ + u64 irh; + + /** opcode/subcode specific parameters */ + u64 ossp[2]; + + /** Return Data Parameters */ + u64 rdp; + + /** Pointer where the response for a RAW mode packet will be written + * by Octeon. + */ + u64 rptr; }; -#define OCT_64B_INSTR_SIZE (sizeof(struct octeon_instr_64B)) +union octeon_instr_64B { + struct octeon_instr2_64B cmd2; + struct octeon_instr3_64B cmd3; +}; + +#define OCT_64B_INSTR_SIZE (sizeof(union octeon_instr_64B)) /** The size of each buffer in soft command buffer pool */ @@ -221,7 +253,8 @@ struct octeon_soft_command { u32 size; /** Command and return status */ - struct octeon_instr_64B cmd; + union octeon_instr_64B cmd; + #define COMPLETION_WORD_INIT 0xffffffffffffffffULL u64 *status_word; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c index 20e0122..7843b8a 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c @@ -44,11 +44,11 @@ void * octeon_alloc_soft_command_resp(struct octeon_device *oct, - struct octeon_instr_64B *cmd, - size_t rdatasize) + union octeon_instr_64B *cmd, + u32 rdatasize) { struct octeon_soft_command *sc; - struct octeon_instr_ih *ih; + struct octeon_instr_ih2 *ih2; struct octeon_instr_irh *irh; struct octeon_instr_rdp *rdp; @@ -59,24 +59,25 @@ octeon_alloc_soft_command_resp(struct octeon_device *oct, return NULL; /* Copy existing command structure into the soft command */ - memcpy(&sc->cmd, cmd, sizeof(struct octeon_instr_64B)); + memcpy(&sc->cmd, cmd, sizeof(union octeon_instr_64B)); /* Add in the response related fields. Opcode and Param are already * there. */ - ih = (struct octeon_instr_ih *)&sc->cmd.ih; - ih->fsz = 40; /* irh + ossp[0] + ossp[1] + rdp + rptr = 40 bytes */ + ih2 = (struct octeon_instr_ih2 *)&sc->cmd.cmd2.ih2; + rdp = (struct octeon_instr_rdp *)&sc->cmd.cmd2.rdp; + irh = (struct octeon_instr_irh *)&sc->cmd.cmd2.irh; + ih2->fsz = 40; /* irh + ossp[0] + ossp[1] + rdp + rptr = 40 bytes */ - irh = (struct octeon_instr_irh *)&sc->cmd.irh; irh->rflag = 1; /* a response is required */ - irh->len = 4; /* means four 64-bit words immediately follow irh */ - rdp = (struct octeon_instr_rdp *)&sc->cmd.rdp; rdp->pcie_port = oct->pcie_port; rdp->rlen = rdatasize; *sc->status_word = COMPLETION_WORD_INIT; + sc->cmd.cmd2.rptr = sc->dmarptr; + sc->wait_time = 1000; sc->timeout = jiffies + sc->wait_time; @@ -123,7 +124,7 @@ static inline struct octeon_soft_command { struct octeon_soft_command *sc = NULL; u8 *data; - size_t rdatasize; + u32 rdatasize; u32 uddsize = 0, datasize = 0; uddsize = (u32)(nctrl->ncmd.s.more * 8); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h index 5e53cc9..b71a2bb 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h @@ -85,7 +85,7 @@ struct octnic_data_pkt { u32 datasize; /** Command to be passed to the Octeon device software. */ - struct octeon_instr_64B cmd; + union octeon_instr_64B cmd; /** Input queue to use to send this command. */ u32 q_no; @@ -121,52 +121,109 @@ static inline int octnet_iq_is_full(struct octeon_device *oct, u32 q_no) >= (oct->instr_queue[q_no]->max_count - 2)); } -/** Utility function to prepare a 64B NIC instruction based on a setup command - * @param cmd - pointer to instruction to be filled in. - * @param setup - pointer to the setup structure - * @param q_no - which queue for back pressure - * - * Assumes the cmd instruction is pre-allocated, but no fields are filled in. - */ static inline void -octnet_prepare_pci_cmd(struct octeon_device *oct, struct octeon_instr_64B *cmd, - union octnic_cmd_setup *setup, u32 tag) +octnet_prepare_pci_cmd_o2(struct octeon_device *oct, + union octeon_instr_64B *cmd, + union octnic_cmd_setup *setup, u32 tag) { - struct octeon_instr_ih *ih; + struct octeon_instr_ih2 *ih2; struct octeon_instr_irh *irh; union octnic_packet_params packet_params; int port; - memset(cmd, 0, sizeof(struct octeon_instr_64B)); + memset(cmd, 0, sizeof(union octeon_instr_64B)); - ih = (struct octeon_instr_ih *)&cmd->ih; + ih2 = (struct octeon_instr_ih2 *)&cmd->cmd2.ih2; /* assume that rflag is cleared so therefore front data will only have - * irh and ossp[1] and ossp[2] for a total of 24 bytes + * irh and ossp[0], ossp[1] for a total of 32 bytes */ - ih->fsz = 24; + ih2->fsz = 24; - ih->tagtype = ORDERED_TAG; - ih->grp = DEFAULT_POW_GRP; + ih2->tagtype = ORDERED_TAG; + ih2->grp = DEFAULT_POW_GRP; port = (int)oct->instr_queue[setup->s.iq_no]->txpciq.s.port; if (tag) - ih->tag = tag; + ih2->tag = tag; else - ih->tag = LIO_DATA(port); + ih2->tag = LIO_DATA(port); + + ih2->raw = 1; + ih2->qos = (port & 3) + 4; /* map qos based on interface */ + + if (!setup->s.gather) { + ih2->dlengsz = setup->s.u.datasize; + } else { + ih2->gather = 1; + ih2->dlengsz = setup->s.u.gatherptrs; + } + + irh = (struct octeon_instr_irh *)&cmd->cmd2.irh; + + irh->opcode = OPCODE_NIC; + irh->subcode = OPCODE_NIC_NW_DATA; + + packet_params.u32 = 0; + + packet_params.s.ip_csum = setup->s.ip_csum; + packet_params.s.transport_csum = setup->s.transport_csum; + packet_params.s.tnl_csum = setup->s.tnl_csum; + packet_params.s.tsflag = setup->s.timestamp; + + irh->ossp = packet_params.u32; +} + +static inline void +octnet_prepare_pci_cmd_o3(struct octeon_device *oct, + union octeon_instr_64B *cmd, + union octnic_cmd_setup *setup, u32 tag) +{ + struct octeon_instr_irh *irh; + struct octeon_instr_ih3 *ih3; + struct octeon_instr_pki_ih3 *pki_ih3; + union octnic_packet_params packet_params; + int port; - ih->raw = 1; - ih->qos = (port & 3) + 4; /* map qos based on interface */ + memset(cmd, 0, sizeof(union octeon_instr_64B)); + + ih3 = (struct octeon_instr_ih3 *)&cmd->cmd3.ih3; + pki_ih3 = (struct octeon_instr_pki_ih3 *)&cmd->cmd3.pki_ih3; + + /* assume that rflag is cleared so therefore front data will only have + * irh and ossp[1] and ossp[2] for a total of 24 bytes + */ + ih3->pkind = oct->instr_queue[setup->s.iq_no]->txpciq.s.pkind; + /*PKI IH*/ + ih3->fsz = 24 + 8; if (!setup->s.gather) { - ih->dlengsz = setup->s.u.datasize; + ih3->dlengsz = setup->s.u.datasize; } else { - ih->gather = 1; - ih->dlengsz = setup->s.u.gatherptrs; + ih3->gather = 1; + ih3->dlengsz = setup->s.u.gatherptrs; } - irh = (struct octeon_instr_irh *)&cmd->irh; + pki_ih3->w = 1; + pki_ih3->raw = 1; + pki_ih3->utag = 1; + pki_ih3->utt = 1; + pki_ih3->uqpg = oct->instr_queue[setup->s.iq_no]->txpciq.s.use_qpg; + + port = (int)oct->instr_queue[setup->s.iq_no]->txpciq.s.port; + + if (tag) + pki_ih3->tag = tag; + else + pki_ih3->tag = LIO_DATA(port); + + pki_ih3->tagtype = ORDERED_TAG; + pki_ih3->qpg = oct->instr_queue[setup->s.iq_no]->txpciq.s.qpg; + pki_ih3->pm = 0x7; /*0x7 - meant for Parse nothing, uninterpreted*/ + pki_ih3->sl = 8; /* sl will be sizeof(pki_ih3)*/ + + irh = (struct octeon_instr_irh *)&cmd->cmd3.irh; irh->opcode = OPCODE_NIC; irh->subcode = OPCODE_NIC_NW_DATA; @@ -181,6 +238,23 @@ octnet_prepare_pci_cmd(struct octeon_device *oct, struct octeon_instr_64B *cmd, irh->ossp = packet_params.u32; } +/** Utility function to prepare a 64B NIC instruction based on a setup command + * @param cmd - pointer to instruction to be filled in. + * @param setup - pointer to the setup structure + * @param q_no - which queue for back pressure + * + * Assumes the cmd instruction is pre-allocated, but no fields are filled in. + */ +static inline void +octnet_prepare_pci_cmd(struct octeon_device *oct, union octeon_instr_64B *cmd, + union octnic_cmd_setup *setup, u32 tag) +{ + if (OCTEON_CN6XXX(oct)) + octnet_prepare_pci_cmd_o2(oct, cmd, setup, tag); + else + octnet_prepare_pci_cmd_o3(oct, cmd, setup, tag); +} + /** Allocate and a soft command with space for a response immediately following * the commnad. * @param oct - octeon device pointer @@ -193,8 +267,8 @@ octnet_prepare_pci_cmd(struct octeon_device *oct, struct octeon_instr_64B *cmd, */ void * octeon_alloc_soft_command_resp(struct octeon_device *oct, - struct octeon_instr_64B *cmd, - size_t rdatasize); + union octeon_instr_64B *cmd, + u32 rdatasize); /** Send a NIC data packet to the device * @param oct - octeon device pointer @@ -209,8 +283,6 @@ int octnet_send_nic_data_pkt(struct octeon_device *oct, /** Send a NIC control packet to the device * @param oct - octeon device pointer * @param nctrl - control structure with command, timout, and callback info - * @param nparams - response control structure - * * @returns IQ_FAILED if it failed to add to the input queue. IQ_STOP if it the * queue should be stopped, and IQ_SEND_OK if it sent okay. */ diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index b0a8d4d..8649677 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -99,6 +99,7 @@ int octeon_init_instr_queue(struct octeon_device *oct, q_size = (u32)conf->instr_type * num_descs; iq = oct->instr_queue[iq_no]; + iq->oct_dev = oct; set_dev_node(&oct->pci_dev->dev, numa_node); iq->base_addr = lio_dma_alloc(oct, q_size, @@ -420,7 +421,7 @@ lio_process_iq_request_list(struct octeon_device *oct, case REQTYPE_SOFT_COMMAND: sc = buf; - irh = (struct octeon_instr_irh *)&sc->cmd.irh; + irh = (struct octeon_instr_irh *)&sc->cmd.cmd2.irh; if (irh->rflag) { /* We're expecting a response from Octeon. * It's up to lio_process_ordered_list() to @@ -583,7 +584,7 @@ octeon_prepare_soft_command(struct octeon_device *oct, u64 ossp1) { struct octeon_config *oct_cfg; - struct octeon_instr_ih *ih; + struct octeon_instr_ih2 *ih2; struct octeon_instr_irh *irh; struct octeon_instr_rdp *rdp; @@ -592,73 +593,69 @@ octeon_prepare_soft_command(struct octeon_device *oct, oct_cfg = octeon_get_conf(oct); - ih = (struct octeon_instr_ih *)&sc->cmd.ih; - ih->tagtype = ATOMIC_TAG; - ih->tag = LIO_CONTROL; - ih->raw = 1; - ih->grp = CFG_GET_CTRL_Q_GRP(oct_cfg); + ih2 = (struct octeon_instr_ih2 *)&sc->cmd.cmd2.ih2; + ih2->tagtype = ATOMIC_TAG; + ih2->tag = LIO_CONTROL; + ih2->raw = 1; + ih2->grp = CFG_GET_CTRL_Q_GRP(oct_cfg); if (sc->datasize) { - ih->dlengsz = sc->datasize; - ih->rs = 1; + ih2->dlengsz = sc->datasize; + ih2->rs = 1; } - irh = (struct octeon_instr_irh *)&sc->cmd.irh; + irh = (struct octeon_instr_irh *)&sc->cmd.cmd2.irh; irh->opcode = opcode; irh->subcode = subcode; /* opcode/subcode specific parameters (ossp) */ irh->ossp = irh_ossp; - sc->cmd.ossp[0] = ossp0; - sc->cmd.ossp[1] = ossp1; + sc->cmd.cmd2.ossp[0] = ossp0; + sc->cmd.cmd2.ossp[1] = ossp1; if (sc->rdatasize) { - rdp = (struct octeon_instr_rdp *)&sc->cmd.rdp; + rdp = (struct octeon_instr_rdp *)&sc->cmd.cmd2.rdp; rdp->pcie_port = oct->pcie_port; rdp->rlen = sc->rdatasize; irh->rflag = 1; - irh->len = 4; - ih->fsz = 40; /* irh+ossp[0]+ossp[1]+rdp+rptr = 40 bytes */ + ih2->fsz = 40; /* irh+ossp[0]+ossp[1]+rdp+rptr = 40 bytes */ } else { irh->rflag = 0; - irh->len = 2; - ih->fsz = 24; /* irh + ossp[0] + ossp[1] = 24 bytes */ + ih2->fsz = 24; /* irh + ossp[0] + ossp[1] = 24 bytes */ } - - while (!(oct->io_qmask.iq & (1 << sc->iq_no))) - sc->iq_no++; } int octeon_send_soft_command(struct octeon_device *oct, struct octeon_soft_command *sc) { - struct octeon_instr_ih *ih; + struct octeon_instr_ih2 *ih2; struct octeon_instr_irh *irh; struct octeon_instr_rdp *rdp; + u32 len; - ih = (struct octeon_instr_ih *)&sc->cmd.ih; - if (ih->dlengsz) { - BUG_ON(!sc->dmadptr); - sc->cmd.dptr = sc->dmadptr; + ih2 = (struct octeon_instr_ih2 *)&sc->cmd.cmd2.ih2; + if (ih2->dlengsz) { + WARN_ON(!sc->dmadptr); + sc->cmd.cmd2.dptr = sc->dmadptr; } - - irh = (struct octeon_instr_irh *)&sc->cmd.irh; + irh = (struct octeon_instr_irh *)&sc->cmd.cmd2.irh; if (irh->rflag) { BUG_ON(!sc->dmarptr); BUG_ON(!sc->status_word); *sc->status_word = COMPLETION_WORD_INIT; - rdp = (struct octeon_instr_rdp *)&sc->cmd.rdp; + rdp = (struct octeon_instr_rdp *)&sc->cmd.cmd2.rdp; - sc->cmd.rptr = sc->dmarptr; + sc->cmd.cmd2.rptr = sc->dmarptr; } + len = (u32)ih2->dlengsz; if (sc->wait_time) sc->timeout = jiffies + sc->wait_time; - return octeon_send_command(oct, sc->iq_no, 1, &sc->cmd, sc, - (u32)ih->dlengsz, REQTYPE_SOFT_COMMAND); + return (octeon_send_command(oct, sc->iq_no, 1, &sc->cmd, sc, + len, REQTYPE_SOFT_COMMAND)); } int octeon_setup_sc_buffer_pool(struct octeon_device *oct) diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.c b/drivers/net/ethernet/cavium/liquidio/response_manager.c index 6287a7c..e2e9103 100644 --- a/drivers/net/ethernet/cavium/liquidio/response_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/response_manager.c @@ -85,6 +85,7 @@ int lio_process_ordered_list(struct octeon_device *octeon_dev, u32 status; u64 status64; struct octeon_instr_rdp *rdp; + u64 rptr; ordered_sc_list = &octeon_dev->response_list[OCTEON_ORDERED_SC_LIST]; @@ -102,7 +103,8 @@ int lio_process_ordered_list(struct octeon_device *octeon_dev, sc = (struct octeon_soft_command *)ordered_sc_list-> head.next; - rdp = (struct octeon_instr_rdp *)&sc->cmd.rdp; + rdp = (struct octeon_instr_rdp *)&sc->cmd.cmd2.rdp; + rptr = sc->cmd.cmd2.rptr; status = OCTEON_REQUEST_PENDING; @@ -110,7 +112,7 @@ int lio_process_ordered_list(struct octeon_device *octeon_dev, * to where rptr is pointing to */ dma_sync_single_for_cpu(&octeon_dev->pci_dev->dev, - sc->cmd.rptr, rdp->rlen, + rptr, rdp->rlen, DMA_FROM_DEVICE); status64 = *sc->status_word; -- cgit v0.10.2 From 100a9db52f706aa3db01e5538f07a4130610a07b Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:42:26 +0100 Subject: sfc: Define macro with EF10 offload feature It is useful to simplify features addition. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 1f30912..89c88ca 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -4703,6 +4703,12 @@ static int efx_ef10_ptp_set_ts_config(struct efx_nic *efx, } } +#define EF10_OFFLOAD_FEATURES \ + (NETIF_F_IP_CSUM | \ + NETIF_F_IPV6_CSUM | \ + NETIF_F_RXHASH | \ + NETIF_F_NTUPLE) + const struct efx_nic_type efx_hunt_a0_vf_nic_type = { .is_vf = true, .mem_bar = EFX_MEM_VF_BAR, @@ -4798,8 +4804,7 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = { .always_rx_scatter = true, .max_interrupt_mode = EFX_INT_MODE_MSIX, .timer_period_max = 1 << ERF_DD_EVQ_IND_TIMER_VAL_WIDTH, - .offload_features = (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | - NETIF_F_RXHASH | NETIF_F_NTUPLE), + .offload_features = EF10_OFFLOAD_FEATURES, .mcdi_max_ver = 2, .max_rx_ip_filters = HUNT_FILTER_TBL_ROWS, .hwtstamp_filters = 1 << HWTSTAMP_FILTER_NONE | @@ -4919,8 +4924,7 @@ const struct efx_nic_type efx_hunt_a0_nic_type = { .always_rx_scatter = true, .max_interrupt_mode = EFX_INT_MODE_MSIX, .timer_period_max = 1 << ERF_DD_EVQ_IND_TIMER_VAL_WIDTH, - .offload_features = (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | - NETIF_F_RXHASH | NETIF_F_NTUPLE), + .offload_features = EF10_OFFLOAD_FEATURES, .mcdi_max_ver = 2, .max_rx_ip_filters = HUNT_FILTER_TBL_ROWS, .hwtstamp_filters = 1 << HWTSTAMP_FILTER_NONE | -- cgit v0.10.2 From b071c3a222db465b32bb6f04ba07264a1556a17c Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:43:00 +0100 Subject: sfc: Move last mc_promisc flag to EF10 filter table state It is used for EF10 only and logically belongs to EF10 filter table state. It is OK that it is reset to false on filter table recreation since all filters are removed on destruction. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 89c88ca..ebe3549 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -83,6 +83,8 @@ struct efx_ef10_filter_table { u16 ucdef_id; u16 bcast_id; u16 mcdef_id; +/* Whether in multicast promiscuous mode when last changed */ + bool mc_promisc_last; }; /* An arbitrary search limit for the software hash table */ @@ -3778,6 +3780,7 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) table->ucdef_id = EFX_EF10_FILTER_ID_INVALID; table->bcast_id = EFX_EF10_FILTER_ID_INVALID; table->mcdef_id = EFX_EF10_FILTER_ID_INVALID; + table->mc_promisc_last = false; efx->filter_state = table; init_waitqueue_head(&table->waitq); @@ -4243,7 +4246,7 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) /* If changing promiscuous state with cascaded multicast filters, remove * old filters first, so that packets are dropped rather than duplicated */ - if (nic_data->workaround_26807 && efx->mc_promisc != mc_promisc) + if (nic_data->workaround_26807 && table->mc_promisc_last != mc_promisc) efx_ef10_filter_remove_old(efx); if (mc_promisc) { if (nic_data->workaround_26807) { @@ -4278,7 +4281,7 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) } efx_ef10_filter_remove_old(efx); - efx->mc_promisc = mc_promisc; + table->mc_promisc_last = mc_promisc; } static int efx_ef10_set_mac_address(struct efx_nic *efx) diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index d13ddf9..77cd23c 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -916,7 +916,6 @@ struct vfdi_status; * @stats_lock: Statistics update lock. Must be held when calling * efx_nic_type::{update,start,stop}_stats. * @n_rx_noskb_drops: Count of RX packets dropped due to failure to allocate an skb - * @mc_promisc: Whether in multicast promiscuous mode when last changed * * This is stored in the private area of the &struct net_device. */ @@ -1065,7 +1064,6 @@ struct efx_nic { int last_irq_cpu; spinlock_t stats_lock; atomic_t n_rx_noskb_drops; - bool mc_promisc; }; static inline int efx_dev_registered(struct efx_nic *efx) -- cgit v0.10.2 From ebfcd0fd90b88088e4d3841c4be9a8c5753d3993 Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:43:20 +0100 Subject: sfc: Add efx_nic member with fixed netdev features It allows to change set of fixed features on datapath reset. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index 097f363..2eecd28 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -600,6 +600,7 @@ fail: */ static void efx_start_datapath(struct efx_nic *efx) { + netdev_features_t old_features = efx->net_dev->features; bool old_rx_scatter = efx->rx_scatter; struct efx_tx_queue *tx_queue; struct efx_rx_queue *rx_queue; @@ -644,6 +645,15 @@ static void efx_start_datapath(struct efx_nic *efx) efx->rx_dma_len, efx->rx_page_buf_step, efx->rx_bufs_per_page, efx->rx_pages_per_batch); + /* Restore previously fixed features in hw_features and remove + * features which are fixed now + */ + efx->net_dev->hw_features |= efx->net_dev->features; + efx->net_dev->hw_features &= ~efx->fixed_features; + efx->net_dev->features |= efx->fixed_features; + if (efx->net_dev->features != old_features) + netdev_features_change(efx->net_dev); + /* RX filters may also have scatter-enabled flags */ if (efx->rx_scatter != old_rx_scatter) efx->type->filter_update_rx_scatter(efx); @@ -3147,17 +3157,17 @@ static int efx_pci_probe(struct pci_dev *pci_dev, return -ENOMEM; efx = netdev_priv(net_dev); efx->type = (const struct efx_nic_type *) entry->driver_data; + efx->fixed_features |= NETIF_F_HIGHDMA; net_dev->features |= (efx->type->offload_features | NETIF_F_SG | - NETIF_F_HIGHDMA | NETIF_F_TSO | - NETIF_F_RXCSUM); + NETIF_F_TSO | NETIF_F_RXCSUM); if (efx->type->offload_features & (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM)) net_dev->features |= NETIF_F_TSO6; /* Mask for features that also apply to VLAN devices */ net_dev->vlan_features |= (NETIF_F_HW_CSUM | NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_ALL_TSO | NETIF_F_RXCSUM); - /* All offloads can be toggled */ - net_dev->hw_features = net_dev->features & ~NETIF_F_HIGHDMA; + net_dev->features |= efx->fixed_features; + net_dev->hw_features = net_dev->features & ~efx->fixed_features; pci_set_drvdata(pci_dev, efx); SET_NETDEV_DEV(net_dev, &pci_dev->dev); rc = efx_init_struct(efx, pci_dev, net_dev); diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index 77cd23c..26abb5c 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -868,6 +868,7 @@ struct vfdi_status; * be held to modify it. * @port_initialized: Port initialized? * @net_dev: Operating system network device. Consider holding the rtnl lock + * @fixed_features: Features which cannot be turned off * @stats_buffer: DMA buffer for statistics * @phy_type: PHY type * @phy_op: PHY interface @@ -1007,6 +1008,8 @@ struct efx_nic { bool port_initialized; struct net_device *net_dev; + netdev_features_t fixed_features; + struct efx_buffer stats_buffer; u64 rx_nodesc_drops_total; u64 rx_nodesc_drops_while_down; -- cgit v0.10.2 From dd98708cf6a7981ad5bc23b1e10c548689482ef7 Mon Sep 17 00:00:00 2001 From: Edward Cree Date: Wed, 15 Jun 2016 17:43:43 +0100 Subject: sfc: Assert filter_sem write locked when required Based on a patch by Andrew Rybchenko Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index ebe3549..66cc963 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -3735,6 +3735,12 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) size_t outlen; int rc; + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return -EINVAL; + + if (efx->filter_state) /* already probed */ + return 0; + table = kzalloc(sizeof(*table), GFP_KERNEL); if (!table) return -ENOMEM; @@ -3846,7 +3852,6 @@ static void efx_ef10_filter_table_restore(struct efx_nic *efx) nic_data->must_restore_filters = false; } -/* Caller must hold efx->filter_sem for write */ static void efx_ef10_filter_table_remove(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; @@ -3856,6 +3861,15 @@ static void efx_ef10_filter_table_remove(struct efx_nic *efx) int rc; efx->filter_state = NULL; + /* If we were called without locking, then it's not safe to free + * the table as others might be using it. So we just WARN, leak + * the memory, and potentially get an inconsistent filter table + * state. + * This should never actually happen. + */ + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return; + if (!table) return; diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h index 5e3f93f..c3ae739 100644 --- a/drivers/net/ethernet/sfc/efx.h +++ b/drivers/net/ethernet/sfc/efx.h @@ -274,4 +274,13 @@ static inline void efx_device_detach_sync(struct efx_nic *efx) netif_tx_unlock_bh(dev); } +static inline bool efx_rwsem_assert_write_locked(struct rw_semaphore *sem) +{ + if (WARN_ON(down_read_trylock(sem))) { + up_read(sem); + return false; + } + return true; +} + #endif /* EFX_EFX_H */ -- cgit v0.10.2 From 6a37958b8aa57cf556935baf67b248e0d8baddbe Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:44:20 +0100 Subject: sfc: Forget filter ID when the filter is marked old It is required to remove setting of filter IDs to invalid from multicast and unicast addresses caching functions. Add initialization to invalid when filter table is created. Add paranoid checks to track consistency. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 66cc963..120a4b7 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -3732,6 +3732,7 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PARSER_DISP_INFO_OUT_LENMAX); unsigned int pd_match_pri, pd_match_count; struct efx_ef10_filter_table *table; + unsigned int i; size_t outlen; int rc; @@ -3783,6 +3784,10 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) goto fail; } + for (i = 0; i < ARRAY_SIZE(table->dev_uc_list); i++) + table->dev_uc_list[i].id = EFX_EF10_FILTER_ID_INVALID; + for (i = 0; i < ARRAY_SIZE(table->dev_mc_list); i++) + table->dev_mc_list[i].id = EFX_EF10_FILTER_ID_INVALID; table->ucdef_id = EFX_EF10_FILTER_ID_INVALID; table->bcast_id = EFX_EF10_FILTER_ID_INVALID; table->mcdef_id = EFX_EF10_FILTER_ID_INVALID; @@ -3897,19 +3902,26 @@ static void efx_ef10_filter_table_remove(struct efx_nic *efx) kfree(table); } -#define EFX_EF10_FILTER_DO_MARK_OLD(id) \ - if (id != EFX_EF10_FILTER_ID_INVALID) { \ - filter_idx = efx_ef10_filter_get_unsafe_id(efx, id); \ - if (!table->entry[filter_idx].spec) \ - netif_dbg(efx, drv, efx->net_dev, \ - "%s: marked null spec old %04x:%04x\n", \ - __func__, id, filter_idx); \ - table->entry[filter_idx].spec |= EFX_EF10_FILTER_FLAG_AUTO_OLD;\ +static void efx_ef10_filter_mark_one_old(struct efx_nic *efx, uint16_t *id) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + unsigned int filter_idx; + + if (*id != EFX_EF10_FILTER_ID_INVALID) { + filter_idx = efx_ef10_filter_get_unsafe_id(efx, *id); + if (!table->entry[filter_idx].spec) + netif_dbg(efx, drv, efx->net_dev, + "marked null spec old %04x:%04x\n", *id, + filter_idx); + table->entry[filter_idx].spec |= EFX_EF10_FILTER_FLAG_AUTO_OLD; + *id = EFX_EF10_FILTER_ID_INVALID; } +} + static void efx_ef10_filter_mark_old(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; - unsigned int filter_idx, i; + unsigned int i; if (!table) return; @@ -3917,15 +3929,14 @@ static void efx_ef10_filter_mark_old(struct efx_nic *efx) /* Mark old filters that may need to be removed */ spin_lock_bh(&efx->filter_lock); for (i = 0; i < table->dev_uc_count; i++) - EFX_EF10_FILTER_DO_MARK_OLD(table->dev_uc_list[i].id); + efx_ef10_filter_mark_one_old(efx, &table->dev_uc_list[i].id); for (i = 0; i < table->dev_mc_count; i++) - EFX_EF10_FILTER_DO_MARK_OLD(table->dev_mc_list[i].id); - EFX_EF10_FILTER_DO_MARK_OLD(table->ucdef_id); - EFX_EF10_FILTER_DO_MARK_OLD(table->bcast_id); - EFX_EF10_FILTER_DO_MARK_OLD(table->mcdef_id); + efx_ef10_filter_mark_one_old(efx, &table->dev_mc_list[i].id); + efx_ef10_filter_mark_one_old(efx, &table->ucdef_id); + efx_ef10_filter_mark_one_old(efx, &table->bcast_id); + efx_ef10_filter_mark_one_old(efx, &table->mcdef_id); spin_unlock_bh(&efx->filter_lock); } -#undef EFX_EF10_FILTER_DO_MARK_OLD static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx, bool *promisc) { @@ -3935,7 +3946,6 @@ static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx, bool *promisc) int addr_count; unsigned int i; - table->ucdef_id = EFX_EF10_FILTER_ID_INVALID; addr_count = netdev_uc_count(net_dev); if (net_dev->flags & IFF_PROMISC) *promisc = true; @@ -3948,7 +3958,6 @@ static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx, bool *promisc) break; } ether_addr_copy(table->dev_uc_list[i].addr, uc->addr); - table->dev_uc_list[i].id = EFX_EF10_FILTER_ID_INVALID; i++; } } @@ -3960,8 +3969,6 @@ static void efx_ef10_filter_mc_addr_list(struct efx_nic *efx, bool *promisc) struct netdev_hw_addr *mc; unsigned int i, addr_count; - table->mcdef_id = EFX_EF10_FILTER_ID_INVALID; - table->bcast_id = EFX_EF10_FILTER_ID_INVALID; if (net_dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) *promisc = true; @@ -3973,7 +3980,6 @@ static void efx_ef10_filter_mc_addr_list(struct efx_nic *efx, bool *promisc) break; } ether_addr_copy(table->dev_mc_list[i].addr, mc->addr); - table->dev_mc_list[i].id = EFX_EF10_FILTER_ID_INVALID; i++; } @@ -4051,6 +4057,8 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, } return rc; } else { + EFX_WARN_ON_PARANOID(table->bcast_id != + EFX_EF10_FILTER_ID_INVALID); table->bcast_id = efx_ef10_filter_get_unsafe_id(efx, rc); } } @@ -4084,6 +4092,8 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, "%scast mismatch filter insert failed rc=%d\n", multicast ? "Multi" : "Uni", rc); } else if (multicast) { + EFX_WARN_ON_PARANOID(table->mcdef_id != + EFX_EF10_FILTER_ID_INVALID); table->mcdef_id = efx_ef10_filter_get_unsafe_id(efx, rc); if (!nic_data->workaround_26807) { /* Also need an Ethernet broadcast filter */ @@ -4106,11 +4116,14 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, return rc; } } else { + EFX_WARN_ON_PARANOID(table->bcast_id != + EFX_EF10_FILTER_ID_INVALID); table->bcast_id = efx_ef10_filter_get_unsafe_id(efx, rc); } } rc = 0; } else { + EFX_WARN_ON_PARANOID(table->ucdef_id != EFX_EF10_FILTER_ID_INVALID); table->ucdef_id = rc; rc = 0; } -- cgit v0.10.2 From dc3273e0c39b72f9f7f01aaf2d52c29badd4a2ba Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:45:36 +0100 Subject: sfc: Move filter IDs to per-VLAN data structure It is a step to support VLAN filtering in HW. Until then, there is only one struct efx_ef10_filter_vlan per struct efx_ef10_filter_table, with no VLAN information yet. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 120a4b7..7abb8a7 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -50,9 +50,21 @@ enum { #define HUNT_FILTER_TBL_ROWS 8192 #define EFX_EF10_FILTER_ID_INVALID 0xffff + +#define EFX_EF10_FILTER_DEV_UC_MAX 32 +#define EFX_EF10_FILTER_DEV_MC_MAX 256 + +/* Per-VLAN filters information */ +struct efx_ef10_filter_vlan { + u16 uc[EFX_EF10_FILTER_DEV_UC_MAX]; + u16 mc[EFX_EF10_FILTER_DEV_MC_MAX]; + u16 ucdef; + u16 bcast; + u16 mcdef; +}; + struct efx_ef10_dev_addr { u8 addr[ETH_ALEN]; - u16 id; }; struct efx_ef10_filter_table { @@ -73,18 +85,13 @@ struct efx_ef10_filter_table { } *entry; wait_queue_head_t waitq; /* Shadow of net_device address lists, guarded by mac_lock */ -#define EFX_EF10_FILTER_DEV_UC_MAX 32 -#define EFX_EF10_FILTER_DEV_MC_MAX 256 struct efx_ef10_dev_addr dev_uc_list[EFX_EF10_FILTER_DEV_UC_MAX]; struct efx_ef10_dev_addr dev_mc_list[EFX_EF10_FILTER_DEV_MC_MAX]; int dev_uc_count; int dev_mc_count; -/* Indices (like efx_ef10_dev_addr.id) for promisc/allmulti filters */ - u16 ucdef_id; - u16 bcast_id; - u16 mcdef_id; /* Whether in multicast promiscuous mode when last changed */ bool mc_promisc_last; + struct efx_ef10_filter_vlan vlan; }; /* An arbitrary search limit for the software hash table */ @@ -3732,6 +3739,7 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PARSER_DISP_INFO_OUT_LENMAX); unsigned int pd_match_pri, pd_match_count; struct efx_ef10_filter_table *table; + struct efx_ef10_filter_vlan *vlan; unsigned int i; size_t outlen; int rc; @@ -3784,13 +3792,14 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) goto fail; } - for (i = 0; i < ARRAY_SIZE(table->dev_uc_list); i++) - table->dev_uc_list[i].id = EFX_EF10_FILTER_ID_INVALID; - for (i = 0; i < ARRAY_SIZE(table->dev_mc_list); i++) - table->dev_mc_list[i].id = EFX_EF10_FILTER_ID_INVALID; - table->ucdef_id = EFX_EF10_FILTER_ID_INVALID; - table->bcast_id = EFX_EF10_FILTER_ID_INVALID; - table->mcdef_id = EFX_EF10_FILTER_ID_INVALID; + vlan = &table->vlan; + for (i = 0; i < ARRAY_SIZE(vlan->uc); i++) + vlan->uc[i] = EFX_EF10_FILTER_ID_INVALID; + for (i = 0; i < ARRAY_SIZE(vlan->mc); i++) + vlan->mc[i] = EFX_EF10_FILTER_ID_INVALID; + vlan->ucdef = EFX_EF10_FILTER_ID_INVALID; + vlan->bcast = EFX_EF10_FILTER_ID_INVALID; + vlan->mcdef = EFX_EF10_FILTER_ID_INVALID; table->mc_promisc_last = false; efx->filter_state = table; @@ -3921,6 +3930,7 @@ static void efx_ef10_filter_mark_one_old(struct efx_nic *efx, uint16_t *id) static void efx_ef10_filter_mark_old(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; + struct efx_ef10_filter_vlan *vlan = &table->vlan; unsigned int i; if (!table) @@ -3929,12 +3939,12 @@ static void efx_ef10_filter_mark_old(struct efx_nic *efx) /* Mark old filters that may need to be removed */ spin_lock_bh(&efx->filter_lock); for (i = 0; i < table->dev_uc_count; i++) - efx_ef10_filter_mark_one_old(efx, &table->dev_uc_list[i].id); + efx_ef10_filter_mark_one_old(efx, &vlan->uc[i]); for (i = 0; i < table->dev_mc_count; i++) - efx_ef10_filter_mark_one_old(efx, &table->dev_mc_list[i].id); - efx_ef10_filter_mark_one_old(efx, &table->ucdef_id); - efx_ef10_filter_mark_one_old(efx, &table->bcast_id); - efx_ef10_filter_mark_one_old(efx, &table->mcdef_id); + efx_ef10_filter_mark_one_old(efx, &vlan->mc[i]); + efx_ef10_filter_mark_one_old(efx, &vlan->ucdef); + efx_ef10_filter_mark_one_old(efx, &vlan->bcast); + efx_ef10_filter_mark_one_old(efx, &vlan->mcdef); spin_unlock_bh(&efx->filter_lock); } @@ -3990,20 +4000,24 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, bool multicast, bool rollback) { struct efx_ef10_filter_table *table = efx->filter_state; + struct efx_ef10_filter_vlan *vlan = &table->vlan; struct efx_ef10_dev_addr *addr_list; enum efx_filter_flags filter_flags; struct efx_filter_spec spec; u8 baddr[ETH_ALEN]; unsigned int i, j; int addr_count; + u16 *ids; int rc; if (multicast) { addr_list = table->dev_mc_list; addr_count = table->dev_mc_count; + ids = vlan->mc; } else { addr_list = table->dev_uc_list; addr_count = table->dev_uc_count; + ids = vlan->uc; } filter_flags = efx_rss_enabled(efx) ? EFX_FILTER_FLAG_RX_RSS : 0; @@ -4021,12 +4035,12 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, rc); /* Fall back to promiscuous */ for (j = 0; j < i; j++) { - if (addr_list[j].id == EFX_EF10_FILTER_ID_INVALID) + if (ids[j] == EFX_EF10_FILTER_ID_INVALID) continue; efx_ef10_filter_remove_unsafe( efx, EFX_FILTER_PRI_AUTO, - addr_list[j].id); - addr_list[j].id = EFX_EF10_FILTER_ID_INVALID; + ids[j]); + ids[j] = EFX_EF10_FILTER_ID_INVALID; } return rc; } else { @@ -4034,7 +4048,7 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, rc = EFX_EF10_FILTER_ID_INVALID; } } - addr_list[i].id = efx_ef10_filter_get_unsafe_id(efx, rc); + ids[i] = efx_ef10_filter_get_unsafe_id(efx, rc); } if (multicast && rollback) { @@ -4048,18 +4062,18 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, "Broadcast filter insert failed rc=%d\n", rc); /* Fall back to promiscuous */ for (j = 0; j < i; j++) { - if (addr_list[j].id == EFX_EF10_FILTER_ID_INVALID) + if (ids[j] == EFX_EF10_FILTER_ID_INVALID) continue; efx_ef10_filter_remove_unsafe( efx, EFX_FILTER_PRI_AUTO, - addr_list[j].id); - addr_list[j].id = EFX_EF10_FILTER_ID_INVALID; + ids[j]); + ids[j] = EFX_EF10_FILTER_ID_INVALID; } return rc; } else { - EFX_WARN_ON_PARANOID(table->bcast_id != + EFX_WARN_ON_PARANOID(vlan->bcast != EFX_EF10_FILTER_ID_INVALID); - table->bcast_id = efx_ef10_filter_get_unsafe_id(efx, rc); + vlan->bcast = efx_ef10_filter_get_unsafe_id(efx, rc); } } @@ -4071,6 +4085,7 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, { struct efx_ef10_filter_table *table = efx->filter_state; struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_ef10_filter_vlan *vlan = &table->vlan; enum efx_filter_flags filter_flags; struct efx_filter_spec spec; u8 baddr[ETH_ALEN]; @@ -4092,9 +4107,8 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, "%scast mismatch filter insert failed rc=%d\n", multicast ? "Multi" : "Uni", rc); } else if (multicast) { - EFX_WARN_ON_PARANOID(table->mcdef_id != - EFX_EF10_FILTER_ID_INVALID); - table->mcdef_id = efx_ef10_filter_get_unsafe_id(efx, rc); + EFX_WARN_ON_PARANOID(vlan->mcdef != EFX_EF10_FILTER_ID_INVALID); + vlan->mcdef = efx_ef10_filter_get_unsafe_id(efx, rc); if (!nic_data->workaround_26807) { /* Also need an Ethernet broadcast filter */ efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, @@ -4111,20 +4125,20 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, /* Roll back the mc_def filter */ efx_ef10_filter_remove_unsafe( efx, EFX_FILTER_PRI_AUTO, - table->mcdef_id); - table->mcdef_id = EFX_EF10_FILTER_ID_INVALID; + vlan->mcdef); + vlan->mcdef = EFX_EF10_FILTER_ID_INVALID; return rc; } } else { - EFX_WARN_ON_PARANOID(table->bcast_id != + EFX_WARN_ON_PARANOID(vlan->bcast != EFX_EF10_FILTER_ID_INVALID); - table->bcast_id = efx_ef10_filter_get_unsafe_id(efx, rc); + vlan->bcast = efx_ef10_filter_get_unsafe_id(efx, rc); } } rc = 0; } else { - EFX_WARN_ON_PARANOID(table->ucdef_id != EFX_EF10_FILTER_ID_INVALID); - table->ucdef_id = rc; + EFX_WARN_ON_PARANOID(vlan->ucdef != EFX_EF10_FILTER_ID_INVALID); + vlan->ucdef = rc; rc = 0; } return rc; -- cgit v0.10.2 From afa4ce125516cc29a8897952d7824a742bd8b4b9 Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:45:56 +0100 Subject: sfc: Store unicast and multicast promisc flag with address cache These flags are built when address cache is updated. The information will be required when VLAN filtering is added and address cache is used without re-sync. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 7abb8a7..a2a10dc 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -89,6 +89,8 @@ struct efx_ef10_filter_table { struct efx_ef10_dev_addr dev_mc_list[EFX_EF10_FILTER_DEV_MC_MAX]; int dev_uc_count; int dev_mc_count; + bool uc_promisc; + bool mc_promisc; /* Whether in multicast promiscuous mode when last changed */ bool mc_promisc_last; struct efx_ef10_filter_vlan vlan; @@ -3948,7 +3950,7 @@ static void efx_ef10_filter_mark_old(struct efx_nic *efx) spin_unlock_bh(&efx->filter_lock); } -static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx, bool *promisc) +static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; struct net_device *net_dev = efx->net_dev; @@ -3957,14 +3959,13 @@ static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx, bool *promisc) unsigned int i; addr_count = netdev_uc_count(net_dev); - if (net_dev->flags & IFF_PROMISC) - *promisc = true; + table->uc_promisc = !!(net_dev->flags & IFF_PROMISC); table->dev_uc_count = 1 + addr_count; ether_addr_copy(table->dev_uc_list[0].addr, net_dev->dev_addr); i = 1; netdev_for_each_uc_addr(uc, net_dev) { if (i >= EFX_EF10_FILTER_DEV_UC_MAX) { - *promisc = true; + table->uc_promisc = true; break; } ether_addr_copy(table->dev_uc_list[i].addr, uc->addr); @@ -3972,21 +3973,20 @@ static void efx_ef10_filter_uc_addr_list(struct efx_nic *efx, bool *promisc) } } -static void efx_ef10_filter_mc_addr_list(struct efx_nic *efx, bool *promisc) +static void efx_ef10_filter_mc_addr_list(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; struct net_device *net_dev = efx->net_dev; struct netdev_hw_addr *mc; unsigned int i, addr_count; - if (net_dev->flags & (IFF_PROMISC | IFF_ALLMULTI)) - *promisc = true; + table->mc_promisc = !!(net_dev->flags & (IFF_PROMISC | IFF_ALLMULTI)); addr_count = netdev_mc_count(net_dev); i = 0; netdev_for_each_mc_addr(mc, net_dev) { if (i >= EFX_EF10_FILTER_DEV_MC_MAX) { - *promisc = true; + table->mc_promisc = true; break; } ether_addr_copy(table->dev_mc_list[i].addr, mc->addr); @@ -4252,7 +4252,6 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) struct efx_ef10_filter_table *table = efx->filter_state; struct efx_ef10_nic_data *nic_data = efx->nic_data; struct net_device *net_dev = efx->net_dev; - bool uc_promisc = false, mc_promisc = false; if (!efx_dev_registered(efx)) return; @@ -4266,12 +4265,12 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) * address and broadcast address */ netif_addr_lock_bh(net_dev); - efx_ef10_filter_uc_addr_list(efx, &uc_promisc); - efx_ef10_filter_mc_addr_list(efx, &mc_promisc); + efx_ef10_filter_uc_addr_list(efx); + efx_ef10_filter_mc_addr_list(efx); netif_addr_unlock_bh(net_dev); /* Insert/renew unicast filters */ - if (uc_promisc) { + if (table->uc_promisc) { efx_ef10_filter_insert_def(efx, false, false); efx_ef10_filter_insert_addr_list(efx, false, false); } else { @@ -4287,9 +4286,10 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) /* If changing promiscuous state with cascaded multicast filters, remove * old filters first, so that packets are dropped rather than duplicated */ - if (nic_data->workaround_26807 && table->mc_promisc_last != mc_promisc) + if (nic_data->workaround_26807 && + table->mc_promisc_last != table->mc_promisc) efx_ef10_filter_remove_old(efx); - if (mc_promisc) { + if (table->mc_promisc) { if (nic_data->workaround_26807) { /* If we failed to insert promiscuous filters, rollback * and fall back to individual multicast filters @@ -4322,7 +4322,7 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) } efx_ef10_filter_remove_old(efx); - table->mc_promisc_last = mc_promisc; + table->mc_promisc_last = table->mc_promisc; } static int efx_ef10_set_mac_address(struct efx_nic *efx) -- cgit v0.10.2 From b3a3c03c38666a9d6e0d5ff52e81d7e4135f689e Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:47:36 +0100 Subject: sfc: Make EF10 filter management helper functions VLAN-aware It is a step to support VLAN filtering in HW. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index a2a10dc..d1bc0dc 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -56,6 +56,7 @@ enum { /* Per-VLAN filters information */ struct efx_ef10_filter_vlan { + u16 vid; u16 uc[EFX_EF10_FILTER_DEV_UC_MAX]; u16 mc[EFX_EF10_FILTER_DEV_MC_MAX]; u16 ucdef; @@ -3795,6 +3796,7 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) } vlan = &table->vlan; + vlan->vid = EFX_FILTER_VID_UNSPEC; for (i = 0; i < ARRAY_SIZE(vlan->uc); i++) vlan->uc[i] = EFX_EF10_FILTER_ID_INVALID; for (i = 0; i < ARRAY_SIZE(vlan->mc); i++) @@ -3929,17 +3931,13 @@ static void efx_ef10_filter_mark_one_old(struct efx_nic *efx, uint16_t *id) } } -static void efx_ef10_filter_mark_old(struct efx_nic *efx) +/* Mark old per-VLAN filters that may need to be removed */ +static void _efx_ef10_filter_vlan_mark_old(struct efx_nic *efx, + struct efx_ef10_filter_vlan *vlan) { struct efx_ef10_filter_table *table = efx->filter_state; - struct efx_ef10_filter_vlan *vlan = &table->vlan; unsigned int i; - if (!table) - return; - - /* Mark old filters that may need to be removed */ - spin_lock_bh(&efx->filter_lock); for (i = 0; i < table->dev_uc_count; i++) efx_ef10_filter_mark_one_old(efx, &vlan->uc[i]); for (i = 0; i < table->dev_mc_count; i++) @@ -3947,6 +3945,15 @@ static void efx_ef10_filter_mark_old(struct efx_nic *efx) efx_ef10_filter_mark_one_old(efx, &vlan->ucdef); efx_ef10_filter_mark_one_old(efx, &vlan->bcast); efx_ef10_filter_mark_one_old(efx, &vlan->mcdef); +} + +/* Mark old filters that may need to be removed */ +static void efx_ef10_filter_mark_old(struct efx_nic *efx) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + + spin_lock_bh(&efx->filter_lock); + _efx_ef10_filter_vlan_mark_old(efx, &table->vlan); spin_unlock_bh(&efx->filter_lock); } @@ -3997,10 +4004,10 @@ static void efx_ef10_filter_mc_addr_list(struct efx_nic *efx) } static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, - bool multicast, bool rollback) + struct efx_ef10_filter_vlan *vlan, + bool multicast, bool rollback) { struct efx_ef10_filter_table *table = efx->filter_state; - struct efx_ef10_filter_vlan *vlan = &table->vlan; struct efx_ef10_dev_addr *addr_list; enum efx_filter_flags filter_flags; struct efx_filter_spec spec; @@ -4025,8 +4032,7 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, /* Insert/renew filters */ for (i = 0; i < addr_count; i++) { efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0); - efx_filter_set_eth_local(&spec, EFX_FILTER_VID_UNSPEC, - addr_list[i].addr); + efx_filter_set_eth_local(&spec, vlan->vid, addr_list[i].addr); rc = efx_ef10_filter_insert(efx, &spec, true); if (rc < 0) { if (rollback) { @@ -4055,7 +4061,7 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, /* Also need an Ethernet broadcast filter */ efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0); eth_broadcast_addr(baddr); - efx_filter_set_eth_local(&spec, EFX_FILTER_VID_UNSPEC, baddr); + efx_filter_set_eth_local(&spec, vlan->vid, baddr); rc = efx_ef10_filter_insert(efx, &spec, true); if (rc < 0) { netif_warn(efx, drv, efx->net_dev, @@ -4080,12 +4086,11 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, return 0; } -static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, - bool rollback) +static int efx_ef10_filter_insert_def(struct efx_nic *efx, + struct efx_ef10_filter_vlan *vlan, + bool multicast, bool rollback) { - struct efx_ef10_filter_table *table = efx->filter_state; struct efx_ef10_nic_data *nic_data = efx->nic_data; - struct efx_ef10_filter_vlan *vlan = &table->vlan; enum efx_filter_flags filter_flags; struct efx_filter_spec spec; u8 baddr[ETH_ALEN]; @@ -4100,6 +4105,9 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, else efx_filter_set_uc_def(&spec); + if (vlan->vid != EFX_FILTER_VID_UNSPEC) + efx_filter_set_eth_local(&spec, vlan->vid, NULL); + rc = efx_ef10_filter_insert(efx, &spec, true); if (rc < 0) { netif_printk(efx, drv, rc == -EPERM ? KERN_DEBUG : KERN_WARNING, @@ -4114,8 +4122,7 @@ static int efx_ef10_filter_insert_def(struct efx_nic *efx, bool multicast, efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0); eth_broadcast_addr(baddr); - efx_filter_set_eth_local(&spec, EFX_FILTER_VID_UNSPEC, - baddr); + efx_filter_set_eth_local(&spec, vlan->vid, baddr); rc = efx_ef10_filter_insert(efx, &spec, true); if (rc < 0) { netif_warn(efx, drv, efx->net_dev, @@ -4252,6 +4259,7 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) struct efx_ef10_filter_table *table = efx->filter_state; struct efx_ef10_nic_data *nic_data = efx->nic_data; struct net_device *net_dev = efx->net_dev; + struct efx_ef10_filter_vlan *vlan; if (!efx_dev_registered(efx)) return; @@ -4269,17 +4277,19 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) efx_ef10_filter_mc_addr_list(efx); netif_addr_unlock_bh(net_dev); + vlan = &table->vlan; + /* Insert/renew unicast filters */ if (table->uc_promisc) { - efx_ef10_filter_insert_def(efx, false, false); - efx_ef10_filter_insert_addr_list(efx, false, false); + efx_ef10_filter_insert_def(efx, vlan, false, false); + efx_ef10_filter_insert_addr_list(efx, vlan, false, false); } else { /* If any of the filters failed to insert, fall back to * promiscuous mode - add in the uc_def filter. But keep * our individual unicast filters. */ - if (efx_ef10_filter_insert_addr_list(efx, false, false)) - efx_ef10_filter_insert_def(efx, false, false); + if (efx_ef10_filter_insert_addr_list(efx, vlan, false, false)) + efx_ef10_filter_insert_def(efx, vlan, false, false); } /* Insert/renew multicast filters */ @@ -4294,17 +4304,18 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) /* If we failed to insert promiscuous filters, rollback * and fall back to individual multicast filters */ - if (efx_ef10_filter_insert_def(efx, true, true)) { + if (efx_ef10_filter_insert_def(efx, vlan, true, true)) { /* Changing promisc state, so remove old filters */ efx_ef10_filter_remove_old(efx); - efx_ef10_filter_insert_addr_list(efx, true, false); + efx_ef10_filter_insert_addr_list(efx, vlan, + true, false); } } else { /* If we failed to insert promiscuous filters, don't * rollback. Regardless, also insert the mc_list */ - efx_ef10_filter_insert_def(efx, true, false); - efx_ef10_filter_insert_addr_list(efx, true, false); + efx_ef10_filter_insert_def(efx, vlan, true, false); + efx_ef10_filter_insert_addr_list(efx, vlan, true, false); } } else { /* If any filters failed to insert, rollback and fall back to @@ -4312,12 +4323,13 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) * that fails, roll back again and insert as many of our * individual multicast filters as we can. */ - if (efx_ef10_filter_insert_addr_list(efx, true, true)) { + if (efx_ef10_filter_insert_addr_list(efx, vlan, true, true)) { /* Changing promisc state, so remove old filters */ if (nic_data->workaround_26807) efx_ef10_filter_remove_old(efx); - if (efx_ef10_filter_insert_def(efx, true, true)) - efx_ef10_filter_insert_addr_list(efx, true, false); + if (efx_ef10_filter_insert_def(efx, vlan, true, true)) + efx_ef10_filter_insert_addr_list(efx, vlan, + true, false); } } -- cgit v0.10.2 From 34813fe26e173098d70655bc268aef54d3a9e488 Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:48:14 +0100 Subject: sfc: Implement list of VLANs added over interface Right now it contains dummy VLAN entry with unspecified VID only. The entry is used for the case when HW VLAN filtering is not used. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index d1bc0dc..6dd0a79 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -54,8 +54,15 @@ enum { #define EFX_EF10_FILTER_DEV_UC_MAX 32 #define EFX_EF10_FILTER_DEV_MC_MAX 256 +/* VLAN list entry */ +struct efx_ef10_vlan { + struct list_head list; + u16 vid; +}; + /* Per-VLAN filters information */ struct efx_ef10_filter_vlan { + struct list_head list; u16 vid; u16 uc[EFX_EF10_FILTER_DEV_UC_MAX]; u16 mc[EFX_EF10_FILTER_DEV_MC_MAX]; @@ -94,7 +101,7 @@ struct efx_ef10_filter_table { bool mc_promisc; /* Whether in multicast promiscuous mode when last changed */ bool mc_promisc_last; - struct efx_ef10_filter_vlan vlan; + struct list_head vlan_list; }; /* An arbitrary search limit for the software hash table */ @@ -102,6 +109,10 @@ struct efx_ef10_filter_table { static void efx_ef10_rx_free_indir_table(struct efx_nic *efx); static void efx_ef10_filter_table_remove(struct efx_nic *efx); +static int efx_ef10_filter_add_vlan(struct efx_nic *efx, u16 vid); +static void efx_ef10_filter_del_vlan_internal(struct efx_nic *efx, + struct efx_ef10_filter_vlan *vlan); +static void efx_ef10_filter_del_vlan(struct efx_nic *efx, u16 vid); static int efx_ef10_get_warm_boot_count(struct efx_nic *efx) { @@ -287,6 +298,96 @@ static ssize_t efx_ef10_show_primary_flag(struct device *dev, ? 1 : 0); } +static struct efx_ef10_vlan *efx_ef10_find_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_ef10_vlan *vlan; + + WARN_ON(!mutex_is_locked(&nic_data->vlan_lock)); + + list_for_each_entry(vlan, &nic_data->vlan_list, list) { + if (vlan->vid == vid) + return vlan; + } + + return NULL; +} + +static int efx_ef10_add_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_ef10_vlan *vlan; + int rc; + + mutex_lock(&nic_data->vlan_lock); + + vlan = efx_ef10_find_vlan(efx, vid); + if (vlan) { + netif_warn(efx, drv, efx->net_dev, + "VLAN %u already added\n", vid); + rc = -EALREADY; + goto fail_exist; + } + + rc = -ENOMEM; + vlan = kzalloc(sizeof(*vlan), GFP_KERNEL); + if (!vlan) + goto fail_alloc; + + vlan->vid = vid; + + list_add_tail(&vlan->list, &nic_data->vlan_list); + + if (efx->filter_state) { + mutex_lock(&efx->mac_lock); + down_write(&efx->filter_sem); + rc = efx_ef10_filter_add_vlan(efx, vlan->vid); + up_write(&efx->filter_sem); + mutex_unlock(&efx->mac_lock); + if (rc) + goto fail_filter_add_vlan; + } + + mutex_unlock(&nic_data->vlan_lock); + return 0; + +fail_filter_add_vlan: + list_del(&vlan->list); + kfree(vlan); +fail_alloc: +fail_exist: + mutex_unlock(&nic_data->vlan_lock); + return rc; +} + +static void efx_ef10_del_vlan_internal(struct efx_nic *efx, + struct efx_ef10_vlan *vlan) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + WARN_ON(!mutex_is_locked(&nic_data->vlan_lock)); + + if (efx->filter_state) { + down_write(&efx->filter_sem); + efx_ef10_filter_del_vlan(efx, vlan->vid); + up_write(&efx->filter_sem); + } + + list_del(&vlan->list); + kfree(vlan); +} + +static void efx_ef10_cleanup_vlans(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_ef10_vlan *vlan, *next_vlan; + + mutex_lock(&nic_data->vlan_lock); + list_for_each_entry_safe(vlan, next_vlan, &nic_data->vlan_list, list) + efx_ef10_del_vlan_internal(efx, vlan); + mutex_unlock(&nic_data->vlan_lock); +} + static DEVICE_ATTR(link_control_flag, 0444, efx_ef10_show_link_control_flag, NULL); static DEVICE_ATTR(primary_flag, 0444, efx_ef10_show_primary_flag, NULL); @@ -433,8 +534,20 @@ static int efx_ef10_probe(struct efx_nic *efx) #endif ether_addr_copy(nic_data->port_id, efx->net_dev->perm_addr); + INIT_LIST_HEAD(&nic_data->vlan_list); + mutex_init(&nic_data->vlan_lock); + + /* Add unspecified VID to support VLAN filtering being disabled */ + rc = efx_ef10_add_vlan(efx, EFX_FILTER_VID_UNSPEC); + if (rc) + goto fail_add_vid_unspec; + return 0; +fail_add_vid_unspec: + mutex_destroy(&nic_data->vlan_lock); + efx_ptp_remove(efx); + efx_mcdi_mon_remove(efx); fail5: device_remove_file(&efx->pci_dev->dev, &dev_attr_primary_flag); fail4: @@ -688,6 +801,9 @@ static void efx_ef10_remove(struct efx_nic *efx) } #endif + efx_ef10_cleanup_vlans(efx); + mutex_destroy(&nic_data->vlan_lock); + efx_ptp_remove(efx); efx_mcdi_mon_remove(efx); @@ -3736,14 +3852,30 @@ static int efx_ef10_filter_match_flags_from_mcdi(u32 mcdi_flags) return match_flags; } +static void efx_ef10_filter_cleanup_vlans(struct efx_nic *efx) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + struct efx_ef10_filter_vlan *vlan, *next_vlan; + + /* See comment in efx_ef10_filter_table_remove() */ + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return; + + if (!table) + return; + + list_for_each_entry_safe(vlan, next_vlan, &table->vlan_list, list) + efx_ef10_filter_del_vlan_internal(efx, vlan); +} + static int efx_ef10_filter_table_probe(struct efx_nic *efx) { MCDI_DECLARE_BUF(inbuf, MC_CMD_GET_PARSER_DISP_INFO_IN_LEN); MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PARSER_DISP_INFO_OUT_LENMAX); + struct efx_ef10_nic_data *nic_data = efx->nic_data; unsigned int pd_match_pri, pd_match_count; struct efx_ef10_filter_table *table; - struct efx_ef10_filter_vlan *vlan; - unsigned int i; + struct efx_ef10_vlan *vlan; size_t outlen; int rc; @@ -3795,21 +3927,23 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) goto fail; } - vlan = &table->vlan; - vlan->vid = EFX_FILTER_VID_UNSPEC; - for (i = 0; i < ARRAY_SIZE(vlan->uc); i++) - vlan->uc[i] = EFX_EF10_FILTER_ID_INVALID; - for (i = 0; i < ARRAY_SIZE(vlan->mc); i++) - vlan->mc[i] = EFX_EF10_FILTER_ID_INVALID; - vlan->ucdef = EFX_EF10_FILTER_ID_INVALID; - vlan->bcast = EFX_EF10_FILTER_ID_INVALID; - vlan->mcdef = EFX_EF10_FILTER_ID_INVALID; table->mc_promisc_last = false; + INIT_LIST_HEAD(&table->vlan_list); efx->filter_state = table; init_waitqueue_head(&table->waitq); + + list_for_each_entry(vlan, &nic_data->vlan_list, list) { + rc = efx_ef10_filter_add_vlan(efx, vlan->vid); + if (rc) + goto fail_add_vlan; + } + return 0; +fail_add_vlan: + efx_ef10_filter_cleanup_vlans(efx); + efx->filter_state = NULL; fail: kfree(table); return rc; @@ -3878,6 +4012,7 @@ static void efx_ef10_filter_table_remove(struct efx_nic *efx) unsigned int filter_idx; int rc; + efx_ef10_filter_cleanup_vlans(efx); efx->filter_state = NULL; /* If we were called without locking, then it's not safe to free * the table as others might be using it. So we just WARN, leak @@ -3947,13 +4082,18 @@ static void _efx_ef10_filter_vlan_mark_old(struct efx_nic *efx, efx_ef10_filter_mark_one_old(efx, &vlan->mcdef); } -/* Mark old filters that may need to be removed */ +/* Mark old filters that may need to be removed. + * Caller must hold efx->filter_sem for read if race against + * efx_ef10_filter_table_remove() is possible + */ static void efx_ef10_filter_mark_old(struct efx_nic *efx) { struct efx_ef10_filter_table *table = efx->filter_state; + struct efx_ef10_filter_vlan *vlan; spin_lock_bh(&efx->filter_lock); - _efx_ef10_filter_vlan_mark_old(efx, &table->vlan); + list_for_each_entry(vlan, &table->vlan_list, list) + _efx_ef10_filter_vlan_mark_old(efx, vlan); spin_unlock_bh(&efx->filter_lock); } @@ -4254,30 +4394,11 @@ reset_nic: /* Caller must hold efx->filter_sem for read if race against * efx_ef10_filter_table_remove() is possible */ -static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) +static void efx_ef10_filter_vlan_sync_rx_mode(struct efx_nic *efx, + struct efx_ef10_filter_vlan *vlan) { struct efx_ef10_filter_table *table = efx->filter_state; struct efx_ef10_nic_data *nic_data = efx->nic_data; - struct net_device *net_dev = efx->net_dev; - struct efx_ef10_filter_vlan *vlan; - - if (!efx_dev_registered(efx)) - return; - - if (!table) - return; - - efx_ef10_filter_mark_old(efx); - - /* Copy/convert the address lists; add the primary station - * address and broadcast address - */ - netif_addr_lock_bh(net_dev); - efx_ef10_filter_uc_addr_list(efx); - efx_ef10_filter_mc_addr_list(efx); - netif_addr_unlock_bh(net_dev); - - vlan = &table->vlan; /* Insert/renew unicast filters */ if (table->uc_promisc) { @@ -4332,11 +4453,145 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) true, false); } } +} + +/* Caller must hold efx->filter_sem for read if race against + * efx_ef10_filter_table_remove() is possible + */ +static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + struct net_device *net_dev = efx->net_dev; + struct efx_ef10_filter_vlan *vlan; + + if (!efx_dev_registered(efx)) + return; + + if (!table) + return; + + efx_ef10_filter_mark_old(efx); + + /* Copy/convert the address lists; add the primary station + * address and broadcast address + */ + netif_addr_lock_bh(net_dev); + efx_ef10_filter_uc_addr_list(efx); + efx_ef10_filter_mc_addr_list(efx); + netif_addr_unlock_bh(net_dev); + + list_for_each_entry(vlan, &table->vlan_list, list) + efx_ef10_filter_vlan_sync_rx_mode(efx, vlan); efx_ef10_filter_remove_old(efx); table->mc_promisc_last = table->mc_promisc; } +static struct efx_ef10_filter_vlan *efx_ef10_filter_find_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + struct efx_ef10_filter_vlan *vlan; + + WARN_ON(!rwsem_is_locked(&efx->filter_sem)); + + list_for_each_entry(vlan, &table->vlan_list, list) { + if (vlan->vid == vid) + return vlan; + } + + return NULL; +} + +static int efx_ef10_filter_add_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_filter_table *table = efx->filter_state; + struct efx_ef10_filter_vlan *vlan; + unsigned int i; + + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return -EINVAL; + + vlan = efx_ef10_filter_find_vlan(efx, vid); + if (WARN_ON(vlan)) { + netif_err(efx, drv, efx->net_dev, + "VLAN %u already added\n", vid); + return -EALREADY; + } + + vlan = kzalloc(sizeof(*vlan), GFP_KERNEL); + if (!vlan) + return -ENOMEM; + + vlan->vid = vid; + + for (i = 0; i < ARRAY_SIZE(vlan->uc); i++) + vlan->uc[i] = EFX_EF10_FILTER_ID_INVALID; + for (i = 0; i < ARRAY_SIZE(vlan->mc); i++) + vlan->mc[i] = EFX_EF10_FILTER_ID_INVALID; + vlan->ucdef = EFX_EF10_FILTER_ID_INVALID; + vlan->bcast = EFX_EF10_FILTER_ID_INVALID; + vlan->mcdef = EFX_EF10_FILTER_ID_INVALID; + + list_add_tail(&vlan->list, &table->vlan_list); + + if (efx_dev_registered(efx)) + efx_ef10_filter_vlan_sync_rx_mode(efx, vlan); + + return 0; +} + +static void efx_ef10_filter_del_vlan_internal(struct efx_nic *efx, + struct efx_ef10_filter_vlan *vlan) +{ + unsigned int i; + + /* See comment in efx_ef10_filter_table_remove() */ + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return; + + list_del(&vlan->list); + + for (i = 0; i < ARRAY_SIZE(vlan->uc); i++) { + if (vlan->uc[i] != EFX_EF10_FILTER_ID_INVALID) + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, + vlan->uc[i]); + } + for (i = 0; i < ARRAY_SIZE(vlan->mc); i++) { + if (vlan->mc[i] != EFX_EF10_FILTER_ID_INVALID) + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, + vlan->mc[i]); + } + if (vlan->ucdef != EFX_EF10_FILTER_ID_INVALID) + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, + vlan->ucdef); + if (vlan->bcast != EFX_EF10_FILTER_ID_INVALID) + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, + vlan->bcast); + if (vlan->mcdef != EFX_EF10_FILTER_ID_INVALID) + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, + vlan->mcdef); + + kfree(vlan); +} + +static void efx_ef10_filter_del_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_filter_vlan *vlan; + + /* See comment in efx_ef10_filter_table_remove() */ + if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) + return; + + vlan = efx_ef10_filter_find_vlan(efx, vid); + if (!vlan) { + netif_err(efx, drv, efx->net_dev, + "VLAN %u not found in filter state\n", vid); + return; + } + + efx_ef10_filter_del_vlan_internal(efx, vlan); +} + static int efx_ef10_set_mac_address(struct efx_nic *efx) { MCDI_DECLARE_BUF(inbuf, MC_CMD_VADAPTOR_SET_MAC_IN_LEN); diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h index 0b536e2..96944c3 100644 --- a/drivers/net/ethernet/sfc/nic.h +++ b/drivers/net/ethernet/sfc/nic.h @@ -519,6 +519,9 @@ enum { #ifdef CONFIG_SFC_SRIOV * @vf: Pointer to VF data structure #endif + * @vport_mac: The MAC address on the vport, only for PFs; VFs will be zero + * @vlan_list: List of VLANs added over the interface. Serialised by vlan_lock. + * @vlan_lock: Lock to serialize access to vlan_list. */ struct efx_ef10_nic_data { struct efx_buffer mcdi_buf; @@ -550,6 +553,8 @@ struct efx_ef10_nic_data { struct ef10_vf *vf; #endif u8 vport_mac[ETH_ALEN]; + struct list_head vlan_list; + struct mutex vlan_lock; }; int efx_init_sriov(void); -- cgit v0.10.2 From 4a53ea8a74248adfb3179c4ede3d741a5dd9ef5a Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:48:32 +0100 Subject: sfc: Implement ndo_vlan_rx_{add, kill}_vid() callbacks Supports HW VLAN filtering, en/disabled using ethtool. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 6dd0a79..a3c00ff 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -101,6 +101,7 @@ struct efx_ef10_filter_table { bool mc_promisc; /* Whether in multicast promiscuous mode when last changed */ bool mc_promisc_last; + bool vlan_filter; struct list_head vlan_list; }; @@ -323,6 +324,11 @@ static int efx_ef10_add_vlan(struct efx_nic *efx, u16 vid) vlan = efx_ef10_find_vlan(efx, vid); if (vlan) { + /* We add VID 0 on init. 8021q adds it on module init + * for all interfaces with VLAN filtring feature. + */ + if (vid == 0) + goto done_unlock; netif_warn(efx, drv, efx->net_dev, "VLAN %u already added\n", vid); rc = -EALREADY; @@ -348,6 +354,7 @@ static int efx_ef10_add_vlan(struct efx_nic *efx, u16 vid) goto fail_filter_add_vlan; } +done_unlock: mutex_unlock(&nic_data->vlan_lock); return 0; @@ -377,6 +384,35 @@ static void efx_ef10_del_vlan_internal(struct efx_nic *efx, kfree(vlan); } +static int efx_ef10_del_vlan(struct efx_nic *efx, u16 vid) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_ef10_vlan *vlan; + int rc = 0; + + /* 8021q removes VID 0 on module unload for all interfaces + * with VLAN filtering feature. We need to keep it to receive + * untagged traffic. + */ + if (vid == 0) + return 0; + + mutex_lock(&nic_data->vlan_lock); + + vlan = efx_ef10_find_vlan(efx, vid); + if (!vlan) { + netif_err(efx, drv, efx->net_dev, + "VLAN %u to be deleted not found\n", vid); + rc = -ENOENT; + } else { + efx_ef10_del_vlan_internal(efx, vlan); + } + + mutex_unlock(&nic_data->vlan_lock); + + return rc; +} + static void efx_ef10_cleanup_vlans(struct efx_nic *efx) { struct efx_ef10_nic_data *nic_data = efx->nic_data; @@ -542,8 +578,18 @@ static int efx_ef10_probe(struct efx_nic *efx) if (rc) goto fail_add_vid_unspec; + /* If VLAN filtering is enabled, we need VID 0 to get untagged + * traffic. It is added automatically if 8021q module is loaded, + * but we can't rely on it since module may be not loaded. + */ + rc = efx_ef10_add_vlan(efx, 0); + if (rc) + goto fail_add_vid_0; + return 0; +fail_add_vid_0: + efx_ef10_cleanup_vlans(efx); fail_add_vid_unspec: mutex_destroy(&nic_data->vlan_lock); efx_ptp_remove(efx); @@ -3928,6 +3974,8 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) } table->mc_promisc_last = false; + table->vlan_filter = + !!(efx->net_dev->features & NETIF_F_HW_VLAN_CTAG_FILTER); INIT_LIST_HEAD(&table->vlan_list); efx->filter_state = table; @@ -4400,6 +4448,12 @@ static void efx_ef10_filter_vlan_sync_rx_mode(struct efx_nic *efx, struct efx_ef10_filter_table *table = efx->filter_state; struct efx_ef10_nic_data *nic_data = efx->nic_data; + /* Do not install unspecified VID if VLAN filtering is enabled. + * Do not install all specified VIDs if VLAN filtering is disabled. + */ + if ((vlan->vid == EFX_FILTER_VID_UNSPEC) == table->vlan_filter) + return; + /* Insert/renew unicast filters */ if (table->uc_promisc) { efx_ef10_filter_insert_def(efx, vlan, false, false); @@ -4463,6 +4517,7 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) struct efx_ef10_filter_table *table = efx->filter_state; struct net_device *net_dev = efx->net_dev; struct efx_ef10_filter_vlan *vlan; + bool vlan_filter; if (!efx_dev_registered(efx)) return; @@ -4480,6 +4535,16 @@ static void efx_ef10_filter_sync_rx_mode(struct efx_nic *efx) efx_ef10_filter_mc_addr_list(efx); netif_addr_unlock_bh(net_dev); + /* If VLAN filtering changes, all old filters are finally removed. + * Do it in advance to avoid conflicts for unicast untagged and + * VLAN 0 tagged filters. + */ + vlan_filter = !!(net_dev->features & NETIF_F_HW_VLAN_CTAG_FILTER); + if (table->vlan_filter != vlan_filter) { + table->vlan_filter = vlan_filter; + efx_ef10_filter_remove_old(efx); + } + list_for_each_entry(vlan, &table->vlan_list, list) efx_ef10_filter_vlan_sync_rx_mode(efx, vlan); @@ -5014,8 +5079,25 @@ static int efx_ef10_ptp_set_ts_config(struct efx_nic *efx, } } +static int efx_ef10_vlan_rx_add_vid(struct efx_nic *efx, __be16 proto, u16 vid) +{ + if (proto != htons(ETH_P_8021Q)) + return -EINVAL; + + return efx_ef10_add_vlan(efx, vid); +} + +static int efx_ef10_vlan_rx_kill_vid(struct efx_nic *efx, __be16 proto, u16 vid) +{ + if (proto != htons(ETH_P_8021Q)) + return -EINVAL; + + return efx_ef10_del_vlan(efx, vid); +} + #define EF10_OFFLOAD_FEATURES \ (NETIF_F_IP_CSUM | \ + NETIF_F_HW_VLAN_CTAG_FILTER | \ NETIF_F_IPV6_CSUM | \ NETIF_F_RXHASH | \ NETIF_F_NTUPLE) @@ -5097,6 +5179,8 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = { #endif .ptp_write_host_time = efx_ef10_ptp_write_host_time_vf, .ptp_set_ts_config = efx_ef10_ptp_set_ts_config_vf, + .vlan_rx_add_vid = efx_ef10_vlan_rx_add_vid, + .vlan_rx_kill_vid = efx_ef10_vlan_rx_kill_vid, #ifdef CONFIG_SFC_SRIOV .vswitching_probe = efx_ef10_vswitching_probe_vf, .vswitching_restore = efx_ef10_vswitching_restore_vf, @@ -5207,6 +5291,8 @@ const struct efx_nic_type efx_hunt_a0_nic_type = { .ptp_write_host_time = efx_ef10_ptp_write_host_time, .ptp_set_ts_sync_events = efx_ef10_ptp_set_ts_sync_events, .ptp_set_ts_config = efx_ef10_ptp_set_ts_config, + .vlan_rx_add_vid = efx_ef10_vlan_rx_add_vid, + .vlan_rx_kill_vid = efx_ef10_vlan_rx_kill_vid, #ifdef CONFIG_SFC_SRIOV .sriov_configure = efx_ef10_sriov_configure, .sriov_init = efx_ef10_sriov_init, diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index 2eecd28..902bcf2 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -2322,14 +2322,46 @@ static void efx_set_rx_mode(struct net_device *net_dev) static int efx_set_features(struct net_device *net_dev, netdev_features_t data) { struct efx_nic *efx = netdev_priv(net_dev); + int rc; /* If disabling RX n-tuple filtering, clear existing filters */ - if (net_dev->features & ~data & NETIF_F_NTUPLE) - return efx->type->filter_clear_rx(efx, EFX_FILTER_PRI_MANUAL); + if (net_dev->features & ~data & NETIF_F_NTUPLE) { + rc = efx->type->filter_clear_rx(efx, EFX_FILTER_PRI_MANUAL); + if (rc) + return rc; + } + + /* If Rx VLAN filter is changed, update filters via mac_reconfigure */ + if ((net_dev->features ^ data) & NETIF_F_HW_VLAN_CTAG_FILTER) { + /* efx_set_rx_mode() will schedule MAC work to update filters + * when a new features are finally set in net_dev. + */ + efx_set_rx_mode(net_dev); + } return 0; } +static int efx_vlan_rx_add_vid(struct net_device *net_dev, __be16 proto, u16 vid) +{ + struct efx_nic *efx = netdev_priv(net_dev); + + if (efx->type->vlan_rx_add_vid) + return efx->type->vlan_rx_add_vid(efx, proto, vid); + else + return -EOPNOTSUPP; +} + +static int efx_vlan_rx_kill_vid(struct net_device *net_dev, __be16 proto, u16 vid) +{ + struct efx_nic *efx = netdev_priv(net_dev); + + if (efx->type->vlan_rx_kill_vid) + return efx->type->vlan_rx_kill_vid(efx, proto, vid); + else + return -EOPNOTSUPP; +} + static const struct net_device_ops efx_netdev_ops = { .ndo_open = efx_net_open, .ndo_stop = efx_net_stop, @@ -2342,6 +2374,8 @@ static const struct net_device_ops efx_netdev_ops = { .ndo_set_mac_address = efx_set_mac_address, .ndo_set_rx_mode = efx_set_rx_mode, .ndo_set_features = efx_set_features, + .ndo_vlan_rx_add_vid = efx_vlan_rx_add_vid, + .ndo_vlan_rx_kill_vid = efx_vlan_rx_kill_vid, #ifdef CONFIG_SFC_SRIOV .ndo_set_vf_mac = efx_sriov_set_vf_mac, .ndo_set_vf_vlan = efx_sriov_set_vf_vlan, diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index 26abb5c..7613f79 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -1334,6 +1334,8 @@ struct efx_nic_type { int (*ptp_set_ts_config)(struct efx_nic *efx, struct hwtstamp_config *init); int (*sriov_configure)(struct efx_nic *efx, int num_vfs); + int (*vlan_rx_add_vid)(struct efx_nic *efx, __be16 proto, u16 vid); + int (*vlan_rx_kill_vid)(struct efx_nic *efx, __be16 proto, u16 vid); int (*sriov_init)(struct efx_nic *efx); void (*sriov_fini)(struct efx_nic *efx); bool (*sriov_wanted)(struct efx_nic *efx); -- cgit v0.10.2 From d248953a3ccd218893b5119a0e63058f0dabca67 Mon Sep 17 00:00:00 2001 From: Martin Habets Date: Wed, 15 Jun 2016 17:48:49 +0100 Subject: sfc: Take mac_lock before calling efx_ef10_filter_table_probe When trying to enslave an SFC interface to a bond the following BUG_ON was hit: kernel BUG [in ef10.c]! CPU: 0 PID: 4383 Comm: ifenslave Tainted: G ... Call Trace: efx_ef10_filter_add_vlan+0x121/0x180 [sfc] efx_ef10_filter_table_probe+0x2a2/0x4f0 [sfc] efx_ef10_set_mac_address+0x370/0x6d0 [sfc] efx_set_mac_address+0x7d/0x120 [sfc] dev_set_mac_address+0x43/0xa0 bond_enslave+0x337/0xea0 [bonding] This comes from function efx_ef10_filter_vlan_sync_rx_mode. To solve the bug we ensure the mac_lock is taken before calling efx_ef10_filter_add_vlan. But to avoid a priority inversion mac_lock must be taken before filter_sem. To satisfy these requirements we end up taking mac_lock in efx_ef10_vport_set_mac_address, efx_ef10_set_mac_address, efx_ef10_sriov_set_vf_vlan and efx_probe_filters. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index a3c00ff..add9f46 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -4666,6 +4666,8 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx) efx_device_detach_sync(efx); efx_net_stop(efx->net_dev); + + mutex_lock(&efx->mac_lock); down_write(&efx->filter_sem); efx_ef10_filter_table_remove(efx); @@ -4678,6 +4680,8 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx) efx_ef10_filter_table_probe(efx); up_write(&efx->filter_sem); + mutex_unlock(&efx->mac_lock); + if (was_enabled) efx_net_open(efx->net_dev); netif_device_attach(efx->net_dev); diff --git a/drivers/net/ethernet/sfc/ef10_sriov.c b/drivers/net/ethernet/sfc/ef10_sriov.c index 3c17f27..a76610a 100644 --- a/drivers/net/ethernet/sfc/ef10_sriov.c +++ b/drivers/net/ethernet/sfc/ef10_sriov.c @@ -554,6 +554,7 @@ int efx_ef10_sriov_set_vf_vlan(struct efx_nic *efx, int vf_i, u16 vlan, efx_device_detach_sync(vf->efx); efx_net_stop(vf->efx->net_dev); + mutex_lock(&vf->efx->mac_lock); down_write(&vf->efx->filter_sem); vf->efx->type->filter_table_remove(vf->efx); @@ -630,6 +631,7 @@ restore_filters: goto reset_nic_up_write; up_write(&vf->efx->filter_sem); + mutex_unlock(&vf->efx->mac_lock); up_write(&vf->efx->filter_sem); @@ -642,9 +644,10 @@ restore_filters: return rc; reset_nic_up_write: - if (vf->efx) + if (vf->efx) { up_write(&vf->efx->filter_sem); - + mutex_unlock(&vf->efx->mac_lock); + } reset_nic: if (vf->efx) { netif_err(efx, drv, efx->net_dev, diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index 902bcf2..130ee17 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -1729,6 +1729,7 @@ static int efx_probe_filters(struct efx_nic *efx) spin_lock_init(&efx->filter_lock); init_rwsem(&efx->filter_sem); + mutex_lock(&efx->mac_lock); down_write(&efx->filter_sem); rc = efx->type->filter_table_probe(efx); if (rc) @@ -1767,6 +1768,7 @@ static int efx_probe_filters(struct efx_nic *efx) #endif out_unlock: up_write(&efx->filter_sem); + mutex_unlock(&efx->mac_lock); return rc; } -- cgit v0.10.2 From 8c91562075e8621f6d9a11cfcf31a71a7203cbbe Mon Sep 17 00:00:00 2001 From: Edward Cree Date: Wed, 15 Jun 2016 17:49:05 +0100 Subject: sfc: Refactor checks for invalid filter ID Nearly every time we call efx_ef10_filter_remove_unsafe, we first check for EFX_EF10_FILTER_ID_INVALID, in which case we do nothing. So move that check into the function, simplifying all the call sites. Also, change the return type to void, since none of the callers check it. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index add9f46..d4f9e94 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -3568,12 +3568,13 @@ static u32 efx_ef10_filter_get_unsafe_id(struct efx_nic *efx, u32 filter_id) return filter_id % HUNT_FILTER_TBL_ROWS; } -static int efx_ef10_filter_remove_unsafe(struct efx_nic *efx, - enum efx_filter_priority priority, - u32 filter_id) +static void efx_ef10_filter_remove_unsafe(struct efx_nic *efx, + enum efx_filter_priority priority, + u32 filter_id) { - return efx_ef10_filter_remove_internal(efx, 1U << priority, - filter_id, true); + if (filter_id == EFX_EF10_FILTER_ID_INVALID) + return; + efx_ef10_filter_remove_internal(efx, 1U << priority, filter_id, true); } static int efx_ef10_filter_get_safe(struct efx_nic *efx, @@ -4229,8 +4230,6 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, rc); /* Fall back to promiscuous */ for (j = 0; j < i; j++) { - if (ids[j] == EFX_EF10_FILTER_ID_INVALID) - continue; efx_ef10_filter_remove_unsafe( efx, EFX_FILTER_PRI_AUTO, ids[j]); @@ -4256,8 +4255,6 @@ static int efx_ef10_filter_insert_addr_list(struct efx_nic *efx, "Broadcast filter insert failed rc=%d\n", rc); /* Fall back to promiscuous */ for (j = 0; j < i; j++) { - if (ids[j] == EFX_EF10_FILTER_ID_INVALID) - continue; efx_ef10_filter_remove_unsafe( efx, EFX_FILTER_PRI_AUTO, ids[j]); @@ -4616,25 +4613,15 @@ static void efx_ef10_filter_del_vlan_internal(struct efx_nic *efx, list_del(&vlan->list); - for (i = 0; i < ARRAY_SIZE(vlan->uc); i++) { - if (vlan->uc[i] != EFX_EF10_FILTER_ID_INVALID) - efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, - vlan->uc[i]); - } - for (i = 0; i < ARRAY_SIZE(vlan->mc); i++) { - if (vlan->mc[i] != EFX_EF10_FILTER_ID_INVALID) - efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, - vlan->mc[i]); - } - if (vlan->ucdef != EFX_EF10_FILTER_ID_INVALID) - efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, - vlan->ucdef); - if (vlan->bcast != EFX_EF10_FILTER_ID_INVALID) + for (i = 0; i < ARRAY_SIZE(vlan->uc); i++) efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, - vlan->bcast); - if (vlan->mcdef != EFX_EF10_FILTER_ID_INVALID) + vlan->uc[i]); + for (i = 0; i < ARRAY_SIZE(vlan->mc); i++) efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, - vlan->mcdef); + vlan->mc[i]); + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, vlan->ucdef); + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, vlan->bcast); + efx_ef10_filter_remove_unsafe(efx, EFX_FILTER_PRI_AUTO, vlan->mcdef); kfree(vlan); } -- cgit v0.10.2 From 7ac0dd9de6eda438dfeafdebc30a2d67e1e5ad60 Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:49:30 +0100 Subject: sfc: Fix dup unknown multicast/unicast filters after datapath reset Filter match flags are not unique criteria to be mapped to priority because of both unknown unicast and unknown multicast are mapped to LOC_MAC_IG. So, local MAC is required to map filter to priority. MCDI filter flags is unique criteria to find filter priority. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index d4f9e94..626054d 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -76,8 +76,8 @@ struct efx_ef10_dev_addr { }; struct efx_ef10_filter_table { -/* The RX match field masks supported by this fw & hw, in order of priority */ - enum efx_filter_match_flags rx_match_flags[ +/* The MCDI match masks supported by this fw & hw, in order of priority */ + u32 rx_match_mcdi_flags[ MC_CMD_GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES_MAXNUM]; unsigned int rx_match_count; @@ -3214,15 +3214,55 @@ static int efx_ef10_filter_push(struct efx_nic *efx, return rc; } -static int efx_ef10_filter_rx_match_pri(struct efx_ef10_filter_table *table, - enum efx_filter_match_flags match_flags) +static u32 efx_ef10_filter_mcdi_flags_from_spec(const struct efx_filter_spec *spec) { + unsigned int match_flags = spec->match_flags; + u32 mcdi_flags = 0; + + if (match_flags & EFX_FILTER_MATCH_LOC_MAC_IG) { + match_flags &= ~EFX_FILTER_MATCH_LOC_MAC_IG; + mcdi_flags |= + is_multicast_ether_addr(spec->loc_mac) ? + (1 << MC_CMD_FILTER_OP_IN_MATCH_UNKNOWN_MCAST_DST_LBN) : + (1 << MC_CMD_FILTER_OP_IN_MATCH_UNKNOWN_UCAST_DST_LBN); + } + +#define MAP_FILTER_TO_MCDI_FLAG(gen_flag, mcdi_field) { \ + unsigned int old_match_flags = match_flags; \ + match_flags &= ~EFX_FILTER_MATCH_ ## gen_flag; \ + if (match_flags != old_match_flags) \ + mcdi_flags |= \ + (1 << MC_CMD_FILTER_OP_IN_MATCH_ ## \ + mcdi_field ## _LBN); \ + } + MAP_FILTER_TO_MCDI_FLAG(REM_HOST, SRC_IP); + MAP_FILTER_TO_MCDI_FLAG(LOC_HOST, DST_IP); + MAP_FILTER_TO_MCDI_FLAG(REM_MAC, SRC_MAC); + MAP_FILTER_TO_MCDI_FLAG(REM_PORT, SRC_PORT); + MAP_FILTER_TO_MCDI_FLAG(LOC_MAC, DST_MAC); + MAP_FILTER_TO_MCDI_FLAG(LOC_PORT, DST_PORT); + MAP_FILTER_TO_MCDI_FLAG(ETHER_TYPE, ETHER_TYPE); + MAP_FILTER_TO_MCDI_FLAG(INNER_VID, INNER_VLAN); + MAP_FILTER_TO_MCDI_FLAG(OUTER_VID, OUTER_VLAN); + MAP_FILTER_TO_MCDI_FLAG(IP_PROTO, IP_PROTO); +#undef MAP_FILTER_TO_MCDI_FLAG + + /* Did we map them all? */ + WARN_ON_ONCE(match_flags); + + return mcdi_flags; +} + +static int efx_ef10_filter_pri(struct efx_ef10_filter_table *table, + const struct efx_filter_spec *spec) +{ + u32 mcdi_flags = efx_ef10_filter_mcdi_flags_from_spec(spec); unsigned int match_pri; for (match_pri = 0; match_pri < table->rx_match_count; match_pri++) - if (table->rx_match_flags[match_pri] == match_flags) + if (table->rx_match_mcdi_flags[match_pri] == mcdi_flags) return match_pri; return -EPROTONOSUPPORT; @@ -3248,7 +3288,7 @@ static s32 efx_ef10_filter_insert(struct efx_nic *efx, EFX_FILTER_FLAG_RX) return -EINVAL; - rc = efx_ef10_filter_rx_match_pri(table, spec->match_flags); + rc = efx_ef10_filter_pri(table, spec); if (rc < 0) return rc; match_pri = rc; @@ -3487,7 +3527,7 @@ static int efx_ef10_filter_remove_internal(struct efx_nic *efx, spec = efx_ef10_filter_entry_spec(table, filter_idx); if (!spec || (!by_index && - efx_ef10_filter_rx_match_pri(table, spec->match_flags) != + efx_ef10_filter_pri(table, spec) != filter_id / HUNT_FILTER_TBL_ROWS)) { rc = -ENOENT; goto out_unlock; @@ -3589,7 +3629,7 @@ static int efx_ef10_filter_get_safe(struct efx_nic *efx, spin_lock_bh(&efx->filter_lock); saved_spec = efx_ef10_filter_entry_spec(table, filter_idx); if (saved_spec && saved_spec->priority == priority && - efx_ef10_filter_rx_match_pri(table, saved_spec->match_flags) == + efx_ef10_filter_pri(table, saved_spec) == filter_id / HUNT_FILTER_TBL_ROWS) { *spec = *saved_spec; rc = 0; @@ -3662,8 +3702,7 @@ static s32 efx_ef10_filter_get_rx_ids(struct efx_nic *efx, count = -EMSGSIZE; break; } - buf[count++] = (efx_ef10_filter_rx_match_pri( - table, spec->match_flags) * + buf[count++] = (efx_ef10_filter_pri(table, spec) * HUNT_FILTER_TBL_ROWS + filter_idx); } @@ -3915,6 +3954,24 @@ static void efx_ef10_filter_cleanup_vlans(struct efx_nic *efx) efx_ef10_filter_del_vlan_internal(efx, vlan); } +static bool efx_ef10_filter_match_supported(struct efx_ef10_filter_table *table, + enum efx_filter_match_flags match_flags) +{ + unsigned int match_pri; + int mf; + + for (match_pri = 0; + match_pri < table->rx_match_count; + match_pri++) { + mf = efx_ef10_filter_match_flags_from_mcdi( + table->rx_match_mcdi_flags[match_pri]); + if (mf == match_flags) + return true; + } + + return false; +} + static int efx_ef10_filter_table_probe(struct efx_nic *efx) { MCDI_DECLARE_BUF(inbuf, MC_CMD_GET_PARSER_DISP_INFO_IN_LEN); @@ -3964,7 +4021,8 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) "%s: fw flags %#x pri %u supported as driver flags %#x pri %u\n", __func__, mcdi_flags, pd_match_pri, rc, table->rx_match_count); - table->rx_match_flags[table->rx_match_count++] = rc; + table->rx_match_mcdi_flags[table->rx_match_count] = mcdi_flags; + table->rx_match_count++; } } -- cgit v0.10.2 From e4478ad14f442103d7f03406933aeb28cbdfc1bd Mon Sep 17 00:00:00 2001 From: Martin Habets Date: Wed, 15 Jun 2016 17:51:07 +0100 Subject: sfc: VLAN filters must only be created if the firmware supports this. If it is not supported we simply disable the feature. For the feature to work we need firmware filter support for OUTER_VID + LOC_MAC and for OUTER_VID + LOC_MAC_IG. The low-latency firmware can match on OUTER_VID + LOC_MAC but not on OUTER_VID + LOC_MAC_IG. For the capture packet firmware it is the other way around. Only the full-feature variant can match on both combinations. Incorporates a fix by Andrew Rybchenko in the net_dev->[hw_]features handling. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 626054d..353ceef 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -3977,6 +3977,7 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) MCDI_DECLARE_BUF(inbuf, MC_CMD_GET_PARSER_DISP_INFO_IN_LEN); MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_PARSER_DISP_INFO_OUT_LENMAX); struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct net_device *net_dev = efx->net_dev; unsigned int pd_match_pri, pd_match_count; struct efx_ef10_filter_table *table; struct efx_ef10_vlan *vlan; @@ -4026,6 +4027,18 @@ static int efx_ef10_filter_table_probe(struct efx_nic *efx) } } + if ((efx_supported_features(efx) & NETIF_F_HW_VLAN_CTAG_FILTER) && + !(efx_ef10_filter_match_supported(table, + (EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC)) && + efx_ef10_filter_match_supported(table, + (EFX_FILTER_MATCH_OUTER_VID | EFX_FILTER_MATCH_LOC_MAC_IG)))) { + netif_info(efx, probe, net_dev, + "VLAN filters are not supported in this firmware variant\n"); + net_dev->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER; + efx->fixed_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER; + net_dev->hw_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER; + } + table->entry = vzalloc(HUNT_FILTER_TBL_ROWS * sizeof(*table->entry)); if (!table->entry) { rc = -ENOMEM; diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index 7613f79..9ff062a 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -1524,4 +1524,16 @@ static inline void efx_xmit_hwtstamp_pending(struct sk_buff *skb) skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; } +/* Get all supported features. + * If a feature is not fixed, it is present in hw_features. + * If a feature is fixed, it does not present in hw_features, but + * always in features. + */ +static inline netdev_features_t efx_supported_features(const struct efx_nic *efx) +{ + const struct net_device *net_dev = efx->net_dev; + + return net_dev->features | net_dev->hw_features; +} + #endif /* EFX_NET_DRIVER_H */ -- cgit v0.10.2 From eb7cfd8c9b6e28edf339b89bf19322c11514d286 Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:51:36 +0100 Subject: sfc: Disable VLAN filtering by default if not strictly required If should be done after net_dev->hw_features initialization, to keep the feature there to be able to enable it later using ethtool. VLAN filtering is enforced and fixed if vPort requires usage of VLAN filters to receive tagged traffic. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c index 130ee17..14b821b 100644 --- a/drivers/net/ethernet/sfc/efx.c +++ b/drivers/net/ethernet/sfc/efx.c @@ -3202,8 +3202,16 @@ static int efx_pci_probe(struct pci_dev *pci_dev, net_dev->vlan_features |= (NETIF_F_HW_CSUM | NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_ALL_TSO | NETIF_F_RXCSUM); - net_dev->features |= efx->fixed_features; + net_dev->hw_features = net_dev->features & ~efx->fixed_features; + + /* Disable VLAN filtering by default. It may be enforced if + * the feature is fixed (i.e. VLAN filters are required to + * receive VLAN tagged packets due to vPort restrictions). + */ + net_dev->features &= ~NETIF_F_HW_VLAN_CTAG_FILTER; + net_dev->features |= efx->fixed_features; + pci_set_drvdata(pci_dev, efx); SET_NETDEV_DEV(net_dev, &pci_dev->dev); rc = efx_init_struct(efx, pci_dev, net_dev); -- cgit v0.10.2 From 23e202b41978d485540c89b7dc78d8915bf207e3 Mon Sep 17 00:00:00 2001 From: Edward Cree Date: Wed, 15 Jun 2016 17:51:48 +0100 Subject: sfc: Update MCDI protocol definitions Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/mcdi_pcol.h b/drivers/net/ethernet/sfc/mcdi_pcol.h index 4cc7721..c9a5b00 100644 --- a/drivers/net/ethernet/sfc/mcdi_pcol.h +++ b/drivers/net/ethernet/sfc/mcdi_pcol.h @@ -273,6 +273,9 @@ * have already installed filters. See the comment at * MC_CMD_WORKAROUND_BUG26807. */ #define MC_CMD_ERR_FILTERS_PRESENT 0x1014 +/* The clock whose frequency you've attempted to set set + * doesn't exist on this NIC */ +#define MC_CMD_ERR_NO_CLOCK 0x1015 #define MC_CMD_ERR_CODE_OFST 0 @@ -292,9 +295,11 @@ /* Point to the copycode entry point. */ #define SIENA_MC_BOOTROM_COPYCODE_VEC (0x800 - 3 * 0x4) #define HUNT_MC_BOOTROM_COPYCODE_VEC (0x8000 - 3 * 0x4) +#define MEDFORD_MC_BOOTROM_COPYCODE_VEC (0x10000 - 3 * 0x4) /* Points to the recovery mode entry point. */ #define SIENA_MC_BOOTROM_NOFLASH_VEC (0x800 - 2 * 0x4) #define HUNT_MC_BOOTROM_NOFLASH_VEC (0x8000 - 2 * 0x4) +#define MEDFORD_MC_BOOTROM_NOFLASH_VEC (0x10000 - 2 * 0x4) /* The command set exported by the boot ROM (MCDI v0) */ #define MC_CMD_GET_VERSION_V0_SUPPORTED_FUNCS { \ @@ -686,6 +691,12 @@ #define FCDI_EVENT_CODE_PTP_STATUS 0x9 /* enum: Port id config to map MC-FC port idx */ #define FCDI_EVENT_CODE_PORT_CONFIG 0xa +/* enum: Boot result or error code */ +#define FCDI_EVENT_CODE_BOOT_RESULT 0xb +#define FCDI_EVENT_REBOOT_SRC_LBN 36 +#define FCDI_EVENT_REBOOT_SRC_WIDTH 8 +#define FCDI_EVENT_REBOOT_FC_FW 0x0 /* enum */ +#define FCDI_EVENT_REBOOT_FC_BOOTLOADER 0x1 /* enum */ #define FCDI_EVENT_ASSERT_INSTR_ADDRESS_OFST 0 #define FCDI_EVENT_ASSERT_INSTR_ADDRESS_LBN 0 #define FCDI_EVENT_ASSERT_INSTR_ADDRESS_WIDTH 32 @@ -717,6 +728,11 @@ #define FCDI_EVENT_PORT_CONFIG_DATA_OFST 0 #define FCDI_EVENT_PORT_CONFIG_DATA_LBN 0 #define FCDI_EVENT_PORT_CONFIG_DATA_WIDTH 32 +#define FCDI_EVENT_BOOT_RESULT_OFST 0 +/* Enum values, see field(s): */ +/* MC_CMD_AOE/MC_CMD_AOE_OUT_INFO/FC_BOOT_RESULT */ +#define FCDI_EVENT_BOOT_RESULT_LBN 0 +#define FCDI_EVENT_BOOT_RESULT_WIDTH 32 /* FCDI_EXTENDED_EVENT_PPS structuredef: Extended FCDI event to send PPS events * to the MC. Note that this structure | is overlayed over a normal FCDI event @@ -1649,15 +1665,30 @@ /* MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS msgresponse */ #define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_LEN 16 -/* Uncorrected error on transmit timestamps in NIC clock format */ +/* Uncorrected error on PTP transmit timestamps in NIC clock format */ #define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_TRANSMIT_OFST 0 -/* Uncorrected error on receive timestamps in NIC clock format */ +/* Uncorrected error on PTP receive timestamps in NIC clock format */ #define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_RECEIVE_OFST 4 /* Uncorrected error on PPS output in NIC clock format */ #define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_PPS_OUT_OFST 8 /* Uncorrected error on PPS input in NIC clock format */ #define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_PPS_IN_OFST 12 +/* MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2 msgresponse */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_LEN 24 +/* Uncorrected error on PTP transmit timestamps in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_PTP_TX_OFST 0 +/* Uncorrected error on PTP receive timestamps in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_PTP_RX_OFST 4 +/* Uncorrected error on PPS output in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_PPS_OUT_OFST 8 +/* Uncorrected error on PPS input in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_PPS_IN_OFST 12 +/* Uncorrected error on non-PTP transmit timestamps in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_GENERAL_TX_OFST 16 +/* Uncorrected error on non-PTP receive timestamps in NIC clock format */ +#define MC_CMD_PTP_OUT_GET_TIMESTAMP_CORRECTIONS_V2_GENERAL_RX_OFST 20 + /* MC_CMD_PTP_OUT_MANFTEST_PPS msgresponse */ #define MC_CMD_PTP_OUT_MANFTEST_PPS_LEN 4 /* Results of testing */ @@ -2158,8 +2189,12 @@ /* MC_CMD_DRV_ATTACH_IN msgrequest */ #define MC_CMD_DRV_ATTACH_IN_LEN 12 -/* new state (0=detached, 1=attached) to set if UPDATE=1 */ +/* new state to set if UPDATE=1 */ #define MC_CMD_DRV_ATTACH_IN_NEW_STATE_OFST 0 +#define MC_CMD_DRV_ATTACH_LBN 0 +#define MC_CMD_DRV_ATTACH_WIDTH 1 +#define MC_CMD_DRV_PREBOOT_LBN 1 +#define MC_CMD_DRV_PREBOOT_WIDTH 1 /* 1 to set new state, or 0 to just report the existing state */ #define MC_CMD_DRV_ATTACH_IN_UPDATE_OFST 4 /* preferred datapath firmware (for Huntington; ignored for Siena) */ @@ -2181,12 +2216,12 @@ /* MC_CMD_DRV_ATTACH_OUT msgresponse */ #define MC_CMD_DRV_ATTACH_OUT_LEN 4 -/* previous or existing state (0=detached, 1=attached) */ +/* previous or existing state, see the bitmask at NEW_STATE */ #define MC_CMD_DRV_ATTACH_OUT_OLD_STATE_OFST 0 /* MC_CMD_DRV_ATTACH_EXT_OUT msgresponse */ #define MC_CMD_DRV_ATTACH_EXT_OUT_LEN 8 -/* previous or existing state (0=detached, 1=attached) */ +/* previous or existing state, see the bitmask at NEW_STATE */ #define MC_CMD_DRV_ATTACH_EXT_OUT_OLD_STATE_OFST 0 /* Flags associated with this function */ #define MC_CMD_DRV_ATTACH_EXT_OUT_FUNC_FLAGS_OFST 4 @@ -2198,6 +2233,10 @@ #define MC_CMD_DRV_ATTACH_EXT_OUT_FLAG_LINKCTRL 0x1 /* enum: The function can perform privileged operations */ #define MC_CMD_DRV_ATTACH_EXT_OUT_FLAG_TRUSTED 0x2 +/* enum: The function does not have an active port associated with it. The port + * refers to the Sorrento external FPGA port. + */ +#define MC_CMD_DRV_ATTACH_EXT_OUT_FLAG_NO_ACTIVE_PORT 0x3 /***********************************/ @@ -2892,7 +2931,7 @@ */ #define MC_CMD_SET_MAC 0x2c -#define MC_CMD_0x2c_PRIVILEGE_CTG SRIOV_CTG_LINK +#define MC_CMD_0x2c_PRIVILEGE_CTG SRIOV_CTG_GENERAL /* MC_CMD_SET_MAC_IN msgrequest */ #define MC_CMD_SET_MAC_IN_LEN 28 @@ -2927,9 +2966,66 @@ #define MC_CMD_SET_MAC_IN_FLAG_INCLUDE_FCS_LBN 0 #define MC_CMD_SET_MAC_IN_FLAG_INCLUDE_FCS_WIDTH 1 +/* MC_CMD_SET_MAC_EXT_IN msgrequest */ +#define MC_CMD_SET_MAC_EXT_IN_LEN 32 +/* The MTU is the MTU programmed directly into the XMAC/GMAC (inclusive of + * EtherII, VLAN, bug16011 padding). + */ +#define MC_CMD_SET_MAC_EXT_IN_MTU_OFST 0 +#define MC_CMD_SET_MAC_EXT_IN_DRAIN_OFST 4 +#define MC_CMD_SET_MAC_EXT_IN_ADDR_OFST 8 +#define MC_CMD_SET_MAC_EXT_IN_ADDR_LEN 8 +#define MC_CMD_SET_MAC_EXT_IN_ADDR_LO_OFST 8 +#define MC_CMD_SET_MAC_EXT_IN_ADDR_HI_OFST 12 +#define MC_CMD_SET_MAC_EXT_IN_REJECT_OFST 16 +#define MC_CMD_SET_MAC_EXT_IN_REJECT_UNCST_LBN 0 +#define MC_CMD_SET_MAC_EXT_IN_REJECT_UNCST_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_REJECT_BRDCST_LBN 1 +#define MC_CMD_SET_MAC_EXT_IN_REJECT_BRDCST_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_FCNTL_OFST 20 +/* enum: Flow control is off. */ +/* MC_CMD_FCNTL_OFF 0x0 */ +/* enum: Respond to flow control. */ +/* MC_CMD_FCNTL_RESPOND 0x1 */ +/* enum: Respond to and Issue flow control. */ +/* MC_CMD_FCNTL_BIDIR 0x2 */ +/* enum: Auto neg flow control. */ +/* MC_CMD_FCNTL_AUTO 0x3 */ +/* enum: Priority flow control (eftest builds only). */ +/* MC_CMD_FCNTL_QBB 0x4 */ +/* enum: Issue flow control. */ +/* MC_CMD_FCNTL_GENERATE 0x5 */ +#define MC_CMD_SET_MAC_EXT_IN_FLAGS_OFST 24 +#define MC_CMD_SET_MAC_EXT_IN_FLAG_INCLUDE_FCS_LBN 0 +#define MC_CMD_SET_MAC_EXT_IN_FLAG_INCLUDE_FCS_WIDTH 1 +/* Select which parameters to configure. A parameter will only be modified if + * the corresponding control flag is set. If SET_MAC_ENHANCED is not set in + * capabilities then this field is ignored (and all flags are assumed to be + * set). + */ +#define MC_CMD_SET_MAC_EXT_IN_CONTROL_OFST 28 +#define MC_CMD_SET_MAC_EXT_IN_CFG_MTU_LBN 0 +#define MC_CMD_SET_MAC_EXT_IN_CFG_MTU_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_CFG_DRAIN_LBN 1 +#define MC_CMD_SET_MAC_EXT_IN_CFG_DRAIN_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_CFG_REJECT_LBN 2 +#define MC_CMD_SET_MAC_EXT_IN_CFG_REJECT_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_CFG_FCNTL_LBN 3 +#define MC_CMD_SET_MAC_EXT_IN_CFG_FCNTL_WIDTH 1 +#define MC_CMD_SET_MAC_EXT_IN_CFG_FCS_LBN 4 +#define MC_CMD_SET_MAC_EXT_IN_CFG_FCS_WIDTH 1 + /* MC_CMD_SET_MAC_OUT msgresponse */ #define MC_CMD_SET_MAC_OUT_LEN 0 +/* MC_CMD_SET_MAC_V2_OUT msgresponse */ +#define MC_CMD_SET_MAC_V2_OUT_LEN 4 +/* MTU as configured after processing the request. See comment at + * MC_CMD_SET_MAC_IN/MTU. To query MTU without doing any changes, set CONTROL + * to 0. + */ +#define MC_CMD_SET_MAC_V2_OUT_MTU_OFST 0 + /***********************************/ /* MC_CMD_PHY_STATS @@ -3521,6 +3617,26 @@ #define MC_CMD_NVRAM_INFO_OUT_PHYSDEV_OFST 16 #define MC_CMD_NVRAM_INFO_OUT_PHYSADDR_OFST 20 +/* MC_CMD_NVRAM_INFO_V2_OUT msgresponse */ +#define MC_CMD_NVRAM_INFO_V2_OUT_LEN 28 +#define MC_CMD_NVRAM_INFO_V2_OUT_TYPE_OFST 0 +/* Enum values, see field(s): */ +/* MC_CMD_NVRAM_TYPES/MC_CMD_NVRAM_TYPES_OUT/TYPES */ +#define MC_CMD_NVRAM_INFO_V2_OUT_SIZE_OFST 4 +#define MC_CMD_NVRAM_INFO_V2_OUT_ERASESIZE_OFST 8 +#define MC_CMD_NVRAM_INFO_V2_OUT_FLAGS_OFST 12 +#define MC_CMD_NVRAM_INFO_V2_OUT_PROTECTED_LBN 0 +#define MC_CMD_NVRAM_INFO_V2_OUT_PROTECTED_WIDTH 1 +#define MC_CMD_NVRAM_INFO_V2_OUT_TLV_LBN 1 +#define MC_CMD_NVRAM_INFO_V2_OUT_TLV_WIDTH 1 +#define MC_CMD_NVRAM_INFO_V2_OUT_A_B_LBN 7 +#define MC_CMD_NVRAM_INFO_V2_OUT_A_B_WIDTH 1 +#define MC_CMD_NVRAM_INFO_V2_OUT_PHYSDEV_OFST 16 +#define MC_CMD_NVRAM_INFO_V2_OUT_PHYSADDR_OFST 20 +/* Writes must be multiples of this size. Added to support the MUM on Sorrento. + */ +#define MC_CMD_NVRAM_INFO_V2_OUT_WRITESIZE_OFST 24 + /***********************************/ /* MC_CMD_NVRAM_UPDATE_START @@ -3561,6 +3677,37 @@ /* amount to read in bytes */ #define MC_CMD_NVRAM_READ_IN_LENGTH_OFST 8 +/* MC_CMD_NVRAM_READ_IN_V2 msgrequest */ +#define MC_CMD_NVRAM_READ_IN_V2_LEN 16 +#define MC_CMD_NVRAM_READ_IN_V2_TYPE_OFST 0 +/* Enum values, see field(s): */ +/* MC_CMD_NVRAM_TYPES/MC_CMD_NVRAM_TYPES_OUT/TYPES */ +#define MC_CMD_NVRAM_READ_IN_V2_OFFSET_OFST 4 +/* amount to read in bytes */ +#define MC_CMD_NVRAM_READ_IN_V2_LENGTH_OFST 8 +/* Optional control info. If a partition is stored with an A/B versioning + * scheme (i.e. in more than one physical partition in NVRAM) the host can set + * this to control which underlying physical partition is used to read data + * from. This allows it to perform a read-modify-write-verify with the write + * lock continuously held by calling NVRAM_UPDATE_START, reading the old + * contents using MODE=TARGET_CURRENT, overwriting the old partition and then + * verifying by reading with MODE=TARGET_BACKUP. + */ +#define MC_CMD_NVRAM_READ_IN_V2_MODE_OFST 12 +/* enum: Same as omitting MODE: caller sees data in current partition unless it + * holds the write lock in which case it sees data in the partition it is + * updating. + */ +#define MC_CMD_NVRAM_READ_IN_V2_DEFAULT 0x0 +/* enum: Read from the current partition of an A/B pair, even if holding the + * write lock. + */ +#define MC_CMD_NVRAM_READ_IN_V2_TARGET_CURRENT 0x1 +/* enum: Read from the non-current (i.e. to be updated) partition of an A/B + * pair + */ +#define MC_CMD_NVRAM_READ_IN_V2_TARGET_BACKUP 0x2 + /* MC_CMD_NVRAM_READ_OUT msgresponse */ #define MC_CMD_NVRAM_READ_OUT_LENMIN 1 #define MC_CMD_NVRAM_READ_OUT_LENMAX 252 @@ -3895,6 +4042,8 @@ #define MC_CMD_SENSOR_CCOM_AVREG_1V8_SUPPLY 0x39 /* enum: CCOM AVREG 1v8 supply (external ADC): mV */ #define MC_CMD_SENSOR_CCOM_AVREG_1V8_SUPPLY_EXTADC 0x3a +/* enum: CCOM RTS temperature: degC */ +#define MC_CMD_SENSOR_CONTROLLER_RTS 0x3b /* enum: Not a sensor: reserved for the next page flag */ #define MC_CMD_SENSOR_PAGE1_NEXT 0x3f /* enum: controller internal temperature sensor voltage on master core @@ -3931,6 +4080,12 @@ #define MC_CMD_SENSOR_PHY0_VCC 0x4c /* enum: Voltage supplied to the QSFP #1 from their power supply: mV */ #define MC_CMD_SENSOR_PHY1_VCC 0x4d +/* enum: Controller die temperature (TDIODE): degC */ +#define MC_CMD_SENSOR_CONTROLLER_TDIODE_TEMP 0x4e +/* enum: Board temperature (front): degC */ +#define MC_CMD_SENSOR_BOARD_FRONT_TEMP 0x4f +/* enum: Board temperature (back): degC */ +#define MC_CMD_SENSOR_BOARD_BACK_TEMP 0x50 /* MC_CMD_SENSOR_INFO_ENTRY_TYPEDEF */ #define MC_CMD_SENSOR_ENTRY_OFST 4 #define MC_CMD_SENSOR_ENTRY_LEN 8 @@ -4007,7 +4162,7 @@ /* MC_CMD_READ_SENSORS_EXT_IN msgrequest */ #define MC_CMD_READ_SENSORS_EXT_IN_LEN 12 -/* DMA address of host buffer for sensor readings */ +/* DMA address of host buffer for sensor readings (must be 4Kbyte aligned). */ #define MC_CMD_READ_SENSORS_EXT_IN_DMA_ADDR_OFST 0 #define MC_CMD_READ_SENSORS_EXT_IN_DMA_ADDR_LEN 8 #define MC_CMD_READ_SENSORS_EXT_IN_DMA_ADDR_LO_OFST 0 @@ -4608,6 +4763,10 @@ * operations */ #define MC_CMD_MUM_OP_QSFP 0xc +/* enum: Request discrete and SODIMM DDR info (type, size, speed grade, voltage + * level) from MUM + */ +#define MC_CMD_MUM_OP_READ_DDR_INFO 0xd /* MC_CMD_MUM_IN_NULL msgrequest */ #define MC_CMD_MUM_IN_NULL_LEN 4 @@ -4793,6 +4952,10 @@ #define MC_CMD_MUM_IN_PROGRAM_CLOCKS_FLAGS_OFST 8 #define MC_CMD_MUM_IN_PROGRAM_CLOCKS_OVERCLOCK_110_LBN 0 #define MC_CMD_MUM_IN_PROGRAM_CLOCKS_OVERCLOCK_110_WIDTH 1 +#define MC_CMD_MUM_IN_PROGRAM_CLOCKS_CLOCK_NIC_FROM_FPGA_LBN 1 +#define MC_CMD_MUM_IN_PROGRAM_CLOCKS_CLOCK_NIC_FROM_FPGA_WIDTH 1 +#define MC_CMD_MUM_IN_PROGRAM_CLOCKS_CLOCK_REF_FROM_XO_LBN 2 +#define MC_CMD_MUM_IN_PROGRAM_CLOCKS_CLOCK_REF_FROM_XO_WIDTH 1 /* MC_CMD_MUM_IN_FPGA_LOAD msgrequest */ #define MC_CMD_MUM_IN_FPGA_LOAD_LEN 8 @@ -4862,6 +5025,11 @@ #define MC_CMD_MUM_IN_QSFP_POLL_BIST_HDR_OFST 4 #define MC_CMD_MUM_IN_QSFP_POLL_BIST_IDX_OFST 8 +/* MC_CMD_MUM_IN_READ_DDR_INFO msgrequest */ +#define MC_CMD_MUM_IN_READ_DDR_INFO_LEN 4 +/* MUM cmd header */ +/* MC_CMD_MUM_IN_CMD_OFST 0 */ + /* MC_CMD_MUM_OUT msgresponse */ #define MC_CMD_MUM_OUT_LEN 0 @@ -5004,6 +5172,69 @@ #define MC_CMD_MUM_OUT_QSFP_POLL_BIST_LEN 4 #define MC_CMD_MUM_OUT_QSFP_POLL_BIST_TEST_OFST 0 +/* MC_CMD_MUM_OUT_READ_DDR_INFO msgresponse */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_LENMIN 24 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_LENMAX 248 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_LEN(num) (8+8*(num)) +/* Discrete (soldered) DDR resistor strap info */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_DISCRETE_DDR_INFO_OFST 0 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_VRATIO_LBN 0 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_VRATIO_WIDTH 16 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RESERVED1_LBN 16 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RESERVED1_WIDTH 16 +/* Number of SODIMM info records */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_NUM_RECORDS_OFST 4 +/* Array of SODIMM info records */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_OFST 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_LEN 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_LO_OFST 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_HI_OFST 12 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_MINNUM 2 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SODIMM_INFO_RECORD_MAXNUM 30 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_BANK_ID_LBN 0 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_BANK_ID_WIDTH 8 +/* enum: SODIMM bank 1 (Top SODIMM for Sorrento) */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_BANK1 0x0 +/* enum: SODIMM bank 2 (Bottom SODDIMM for Sorrento) */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_BANK2 0x1 +/* enum: Total number of SODIMM banks */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_NUM_BANKS 0x2 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_TYPE_LBN 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_TYPE_WIDTH 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RANK_LBN 16 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RANK_WIDTH 4 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_VOLTAGE_LBN 20 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_VOLTAGE_WIDTH 4 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_NOT_POWERED 0x0 /* enum */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_1V25 0x1 /* enum */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_1V35 0x2 /* enum */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_1V5 0x3 /* enum */ +/* enum: Values 5-15 are reserved for future usage */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_1V8 0x4 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SIZE_LBN 24 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SIZE_WIDTH 8 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SPEED_LBN 32 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_SPEED_WIDTH 16 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_STATE_LBN 48 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_STATE_WIDTH 4 +/* enum: No module present */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_ABSENT 0x0 +/* enum: Module present supported and powered on */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_PRESENT_POWERED 0x1 +/* enum: Module present but bad type */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_PRESENT_BAD_TYPE 0x2 +/* enum: Module present but incompatible voltage */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_PRESENT_BAD_VOLTAGE 0x3 +/* enum: Module present but unknown SPD */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_PRESENT_BAD_SPD 0x4 +/* enum: Module present but slot cannot support it */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_PRESENT_BAD_SLOT 0x5 +/* enum: Modules may or may not be present, but cannot establish contact by I2C + */ +#define MC_CMD_MUM_OUT_READ_DDR_INFO_NOT_REACHABLE 0x6 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RESERVED2_LBN 52 +#define MC_CMD_MUM_OUT_READ_DDR_INFO_RESERVED2_WIDTH 12 + /* MC_CMD_RESOURCE_SPECIFIER enum */ /* enum: Any */ #define MC_CMD_RESOURCE_INSTANCE_ANY 0xffffffff @@ -5076,6 +5307,8 @@ #define NVRAM_PARTITION_TYPE_DYNAMIC_CONFIG 0x500 /* enum: Expansion ROM configuration data for port 0 */ #define NVRAM_PARTITION_TYPE_EXPROM_CONFIG_PORT0 0x600 +/* enum: Synonym for EXPROM_CONFIG_PORT0 as used in pmap files */ +#define NVRAM_PARTITION_TYPE_EXPROM_CONFIG 0x600 /* enum: Expansion ROM configuration data for port 1 */ #define NVRAM_PARTITION_TYPE_EXPROM_CONFIG_PORT1 0x601 /* enum: Expansion ROM configuration data for port 2 */ @@ -5084,6 +5317,8 @@ #define NVRAM_PARTITION_TYPE_EXPROM_CONFIG_PORT3 0x603 /* enum: Non-volatile log output partition */ #define NVRAM_PARTITION_TYPE_LOG 0x700 +/* enum: Non-volatile log output of second core on dual-core device */ +#define NVRAM_PARTITION_TYPE_LOG_SLAVE 0x701 /* enum: Device state dump output partition */ #define NVRAM_PARTITION_TYPE_DUMP 0x800 /* enum: Application license key storage partition */ @@ -5116,6 +5351,20 @@ #define NVRAM_PARTITION_TYPE_MUM_USER_ROM 0xc05 /* enum: MUM fuses and lockbits partition. */ #define NVRAM_PARTITION_TYPE_MUM_FUSELOCK 0xc06 +/* enum: UEFI expansion ROM if separate from PXE */ +#define NVRAM_PARTITION_TYPE_EXPANSION_UEFI 0xd00 +/* enum: Spare partition 0 */ +#define NVRAM_PARTITION_TYPE_SPARE_0 0x1000 +/* enum: Spare partition 1 */ +#define NVRAM_PARTITION_TYPE_SPARE_1 0x1100 +/* enum: Spare partition 2 */ +#define NVRAM_PARTITION_TYPE_SPARE_2 0x1200 +/* enum: Spare partition 3 */ +#define NVRAM_PARTITION_TYPE_SPARE_3 0x1300 +/* enum: Spare partition 4 */ +#define NVRAM_PARTITION_TYPE_SPARE_4 0x1400 +/* enum: Spare partition 5 */ +#define NVRAM_PARTITION_TYPE_SPARE_5 0x1500 /* enum: Start of reserved value range (firmware may use for any purpose) */ #define NVRAM_PARTITION_TYPE_RESERVED_VALUES_MIN 0xff00 /* enum: End of reserved value range (firmware may use for any purpose) */ @@ -5149,6 +5398,90 @@ #define LICENSED_APP_ID_ID_LBN 0 #define LICENSED_APP_ID_ID_WIDTH 32 +/* LICENSED_FEATURES structuredef */ +#define LICENSED_FEATURES_LEN 8 +/* Bitmask of licensed firmware features */ +#define LICENSED_FEATURES_MASK_OFST 0 +#define LICENSED_FEATURES_MASK_LEN 8 +#define LICENSED_FEATURES_MASK_LO_OFST 0 +#define LICENSED_FEATURES_MASK_HI_OFST 4 +#define LICENSED_FEATURES_RX_CUT_THROUGH_LBN 0 +#define LICENSED_FEATURES_RX_CUT_THROUGH_WIDTH 1 +#define LICENSED_FEATURES_PIO_LBN 1 +#define LICENSED_FEATURES_PIO_WIDTH 1 +#define LICENSED_FEATURES_EVQ_TIMER_LBN 2 +#define LICENSED_FEATURES_EVQ_TIMER_WIDTH 1 +#define LICENSED_FEATURES_CLOCK_LBN 3 +#define LICENSED_FEATURES_CLOCK_WIDTH 1 +#define LICENSED_FEATURES_RX_TIMESTAMPS_LBN 4 +#define LICENSED_FEATURES_RX_TIMESTAMPS_WIDTH 1 +#define LICENSED_FEATURES_TX_TIMESTAMPS_LBN 5 +#define LICENSED_FEATURES_TX_TIMESTAMPS_WIDTH 1 +#define LICENSED_FEATURES_RX_SNIFF_LBN 6 +#define LICENSED_FEATURES_RX_SNIFF_WIDTH 1 +#define LICENSED_FEATURES_TX_SNIFF_LBN 7 +#define LICENSED_FEATURES_TX_SNIFF_WIDTH 1 +#define LICENSED_FEATURES_PROXY_FILTER_OPS_LBN 8 +#define LICENSED_FEATURES_PROXY_FILTER_OPS_WIDTH 1 +#define LICENSED_FEATURES_EVENT_CUT_THROUGH_LBN 9 +#define LICENSED_FEATURES_EVENT_CUT_THROUGH_WIDTH 1 +#define LICENSED_FEATURES_MASK_LBN 0 +#define LICENSED_FEATURES_MASK_WIDTH 64 + +/* LICENSED_V3_APPS structuredef */ +#define LICENSED_V3_APPS_LEN 8 +/* Bitmask of licensed applications */ +#define LICENSED_V3_APPS_MASK_OFST 0 +#define LICENSED_V3_APPS_MASK_LEN 8 +#define LICENSED_V3_APPS_MASK_LO_OFST 0 +#define LICENSED_V3_APPS_MASK_HI_OFST 4 +#define LICENSED_V3_APPS_ONLOAD_LBN 0 +#define LICENSED_V3_APPS_ONLOAD_WIDTH 1 +#define LICENSED_V3_APPS_PTP_LBN 1 +#define LICENSED_V3_APPS_PTP_WIDTH 1 +#define LICENSED_V3_APPS_SOLARCAPTURE_PRO_LBN 2 +#define LICENSED_V3_APPS_SOLARCAPTURE_PRO_WIDTH 1 +#define LICENSED_V3_APPS_SOLARSECURE_LBN 3 +#define LICENSED_V3_APPS_SOLARSECURE_WIDTH 1 +#define LICENSED_V3_APPS_PERF_MONITOR_LBN 4 +#define LICENSED_V3_APPS_PERF_MONITOR_WIDTH 1 +#define LICENSED_V3_APPS_SOLARCAPTURE_LIVE_LBN 5 +#define LICENSED_V3_APPS_SOLARCAPTURE_LIVE_WIDTH 1 +#define LICENSED_V3_APPS_CAPTURE_SOLARSYSTEM_LBN 6 +#define LICENSED_V3_APPS_CAPTURE_SOLARSYSTEM_WIDTH 1 +#define LICENSED_V3_APPS_NETWORK_ACCESS_CONTROL_LBN 7 +#define LICENSED_V3_APPS_NETWORK_ACCESS_CONTROL_WIDTH 1 +#define LICENSED_V3_APPS_MASK_LBN 0 +#define LICENSED_V3_APPS_MASK_WIDTH 64 + +/* LICENSED_V3_FEATURES structuredef */ +#define LICENSED_V3_FEATURES_LEN 8 +/* Bitmask of licensed firmware features */ +#define LICENSED_V3_FEATURES_MASK_OFST 0 +#define LICENSED_V3_FEATURES_MASK_LEN 8 +#define LICENSED_V3_FEATURES_MASK_LO_OFST 0 +#define LICENSED_V3_FEATURES_MASK_HI_OFST 4 +#define LICENSED_V3_FEATURES_RX_CUT_THROUGH_LBN 0 +#define LICENSED_V3_FEATURES_RX_CUT_THROUGH_WIDTH 1 +#define LICENSED_V3_FEATURES_PIO_LBN 1 +#define LICENSED_V3_FEATURES_PIO_WIDTH 1 +#define LICENSED_V3_FEATURES_EVQ_TIMER_LBN 2 +#define LICENSED_V3_FEATURES_EVQ_TIMER_WIDTH 1 +#define LICENSED_V3_FEATURES_CLOCK_LBN 3 +#define LICENSED_V3_FEATURES_CLOCK_WIDTH 1 +#define LICENSED_V3_FEATURES_RX_TIMESTAMPS_LBN 4 +#define LICENSED_V3_FEATURES_RX_TIMESTAMPS_WIDTH 1 +#define LICENSED_V3_FEATURES_TX_TIMESTAMPS_LBN 5 +#define LICENSED_V3_FEATURES_TX_TIMESTAMPS_WIDTH 1 +#define LICENSED_V3_FEATURES_RX_SNIFF_LBN 6 +#define LICENSED_V3_FEATURES_RX_SNIFF_WIDTH 1 +#define LICENSED_V3_FEATURES_TX_SNIFF_LBN 7 +#define LICENSED_V3_FEATURES_TX_SNIFF_WIDTH 1 +#define LICENSED_V3_FEATURES_PROXY_FILTER_OPS_LBN 8 +#define LICENSED_V3_FEATURES_PROXY_FILTER_OPS_WIDTH 1 +#define LICENSED_V3_FEATURES_MASK_LBN 0 +#define LICENSED_V3_FEATURES_MASK_WIDTH 64 + /* TX_TIMESTAMP_EVENT structuredef */ #define TX_TIMESTAMP_EVENT_LEN 6 /* lower 16 bits of timestamp data */ @@ -5258,6 +5591,8 @@ #define MC_CMD_INIT_EVQ_IN_FLAG_RX_MERGE_WIDTH 1 #define MC_CMD_INIT_EVQ_IN_FLAG_TX_MERGE_LBN 5 #define MC_CMD_INIT_EVQ_IN_FLAG_TX_MERGE_WIDTH 1 +#define MC_CMD_INIT_EVQ_IN_FLAG_USE_TIMER_LBN 6 +#define MC_CMD_INIT_EVQ_IN_FLAG_USE_TIMER_WIDTH 1 #define MC_CMD_INIT_EVQ_IN_TMR_MODE_OFST 20 /* enum: Disabled */ #define MC_CMD_INIT_EVQ_IN_TMR_MODE_DIS 0x0 @@ -5362,6 +5697,8 @@ #define MC_CMD_INIT_RXQ_IN_FLAG_PREFIX_WIDTH 1 #define MC_CMD_INIT_RXQ_IN_FLAG_DISABLE_SCATTER_LBN 9 #define MC_CMD_INIT_RXQ_IN_FLAG_DISABLE_SCATTER_WIDTH 1 +#define MC_CMD_INIT_RXQ_IN_FLAG_FORCE_EV_MERGING_LBN 10 +#define MC_CMD_INIT_RXQ_IN_FLAG_FORCE_EV_MERGING_WIDTH 1 /* Owner ID to use if in buffer mode (zero if physical) */ #define MC_CMD_INIT_RXQ_IN_OWNER_ID_OFST 20 /* The port ID associated with the v-adaptor which should contain this DMAQ. */ @@ -5422,6 +5759,8 @@ #define MC_CMD_INIT_RXQ_EXT_IN_PS_BUFF_64K 0x4 /* enum */ #define MC_CMD_INIT_RXQ_EXT_IN_FLAG_WANT_OUTER_CLASSES_LBN 18 #define MC_CMD_INIT_RXQ_EXT_IN_FLAG_WANT_OUTER_CLASSES_WIDTH 1 +#define MC_CMD_INIT_RXQ_EXT_IN_FLAG_FORCE_EV_MERGING_LBN 19 +#define MC_CMD_INIT_RXQ_EXT_IN_FLAG_FORCE_EV_MERGING_WIDTH 1 /* Owner ID to use if in buffer mode (zero if physical) */ #define MC_CMD_INIT_RXQ_EXT_IN_OWNER_ID_OFST 20 /* The port ID associated with the v-adaptor which should contain this DMAQ. */ @@ -5535,6 +5874,8 @@ #define MC_CMD_INIT_TXQ_EXT_IN_FLAG_INNER_IP_CSUM_EN_WIDTH 1 #define MC_CMD_INIT_TXQ_EXT_IN_FLAG_INNER_TCP_CSUM_EN_LBN 11 #define MC_CMD_INIT_TXQ_EXT_IN_FLAG_INNER_TCP_CSUM_EN_WIDTH 1 +#define MC_CMD_INIT_TXQ_EXT_IN_FLAG_TSOV2_EN_LBN 12 +#define MC_CMD_INIT_TXQ_EXT_IN_FLAG_TSOV2_EN_WIDTH 1 /* Owner ID to use if in buffer mode (zero if physical) */ #define MC_CMD_INIT_TXQ_EXT_IN_OWNER_ID_OFST 20 /* The port ID associated with the v-adaptor which should contain this DMAQ. */ @@ -5747,6 +6088,46 @@ #define MC_CMD_PROXY_CONFIGURE_IN_ALLOWED_MCDI_MASK_OFST 44 #define MC_CMD_PROXY_CONFIGURE_IN_ALLOWED_MCDI_MASK_LEN 64 +/* MC_CMD_PROXY_CONFIGURE_EXT_IN msgrequest */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_LEN 112 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_FLAGS_OFST 0 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_ENABLE_LBN 0 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_ENABLE_WIDTH 1 +/* Host provides a contiguous memory buffer that contains at least NUM_BLOCKS + * of blocks, each of the size REQUEST_BLOCK_SIZE. + */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_STATUS_BUFF_ADDR_OFST 4 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_STATUS_BUFF_ADDR_LEN 8 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_STATUS_BUFF_ADDR_LO_OFST 4 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_STATUS_BUFF_ADDR_HI_OFST 8 +/* Must be a power of 2 */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_STATUS_BLOCK_SIZE_OFST 12 +/* Host provides a contiguous memory buffer that contains at least NUM_BLOCKS + * of blocks, each of the size REPLY_BLOCK_SIZE. + */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REQUEST_BUFF_ADDR_OFST 16 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REQUEST_BUFF_ADDR_LEN 8 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REQUEST_BUFF_ADDR_LO_OFST 16 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REQUEST_BUFF_ADDR_HI_OFST 20 +/* Must be a power of 2 */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REQUEST_BLOCK_SIZE_OFST 24 +/* Host provides a contiguous memory buffer that contains at least NUM_BLOCKS + * of blocks, each of the size STATUS_BLOCK_SIZE. This buffer is only needed if + * host intends to complete proxied operations by using MC_CMD_PROXY_CMD. + */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REPLY_BUFF_ADDR_OFST 28 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REPLY_BUFF_ADDR_LEN 8 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REPLY_BUFF_ADDR_LO_OFST 28 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REPLY_BUFF_ADDR_HI_OFST 32 +/* Must be a power of 2, or zero if this buffer is not provided */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_REPLY_BLOCK_SIZE_OFST 36 +/* Applies to all three buffers */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_NUM_BLOCKS_OFST 40 +/* A bit mask defining which MCDI operations may be proxied */ +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_ALLOWED_MCDI_MASK_OFST 44 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_ALLOWED_MCDI_MASK_LEN 64 +#define MC_CMD_PROXY_CONFIGURE_EXT_IN_RESERVED_OFST 108 + /* MC_CMD_PROXY_CONFIGURE_OUT msgresponse */ #define MC_CMD_PROXY_CONFIGURE_OUT_LEN 0 @@ -6323,6 +6704,15 @@ * client */ #define MC_CMD_GET_PARSER_DISP_INFO_IN_OP_GET_RESTRICTIONS 0x2 +/* enum: read properties relating to security rules (Medford-only; for use by + * SolarSecure apps, not directly by drivers. See SF-114946-SW.) + */ +#define MC_CMD_GET_PARSER_DISP_INFO_IN_OP_GET_SECURITY_RULE_INFO 0x3 +/* enum: read the list of supported RX filter matches for VXLAN/NVGRE + * encapsulated frames, which follow a different match sequence to normal + * frames (Medford only) + */ +#define MC_CMD_GET_PARSER_DISP_INFO_IN_OP_GET_SUPPORTED_ENCAP_RX_MATCHES 0x4 /* MC_CMD_GET_PARSER_DISP_INFO_OUT msgresponse */ #define MC_CMD_GET_PARSER_DISP_INFO_OUT_LENMIN 8 @@ -6356,7 +6746,10 @@ /***********************************/ /* MC_CMD_PARSER_DISP_RW - * Direct read/write of parser-dispatcher state (DICPUs and LUE) for debugging + * Direct read/write of parser-dispatcher state (DICPUs and LUE) for debugging. + * Please note that this interface is only of use to debug tools which have + * knowledge of firmware and hardware data structures; nothing here is intended + * for use by normal driver code. */ #define MC_CMD_PARSER_DISP_RW 0xe5 @@ -6374,6 +6767,12 @@ #define MC_CMD_PARSER_DISP_RW_IN_LUE 0x2 /* enum: Lookup engine (with requested metadata format) */ #define MC_CMD_PARSER_DISP_RW_IN_LUE_VERSIONED_METADATA 0x3 +/* enum: RX0 dispatcher CPU (alias for RX_DICPU; Medford has 2 RX DICPUs) */ +#define MC_CMD_PARSER_DISP_RW_IN_RX0_DICPU 0x0 +/* enum: RX1 dispatcher CPU (only valid for Medford) */ +#define MC_CMD_PARSER_DISP_RW_IN_RX1_DICPU 0x4 +/* enum: Miscellaneous other state (only valid for Medford) */ +#define MC_CMD_PARSER_DISP_RW_IN_MISC_STATE 0x5 /* identifies the type of operation requested */ #define MC_CMD_PARSER_DISP_RW_IN_OP_OFST 4 /* enum: read a word of DICPU DMEM or a LUE entry */ @@ -6382,8 +6781,12 @@ #define MC_CMD_PARSER_DISP_RW_IN_WRITE 0x1 /* enum: read-modify-write a word of DICPU DMEM (not valid for LUE) */ #define MC_CMD_PARSER_DISP_RW_IN_RMW 0x2 -/* data memory address or LUE index */ +/* data memory address (DICPU targets) or LUE index (LUE targets) */ #define MC_CMD_PARSER_DISP_RW_IN_ADDRESS_OFST 8 +/* selector (for MISC_STATE target) */ +#define MC_CMD_PARSER_DISP_RW_IN_SELECTOR_OFST 8 +/* enum: Port to datapath mapping */ +#define MC_CMD_PARSER_DISP_RW_IN_PORT_DP_MAPPING 0x1 /* value to write (for DMEM writes) */ #define MC_CMD_PARSER_DISP_RW_IN_DMEM_WRITE_VALUE_OFST 12 /* XOR value (for DMEM read-modify-writes: new = (old & mask) ^ value) */ @@ -6408,6 +6811,12 @@ */ #define MC_CMD_PARSER_DISP_RW_OUT_LUE_MGR_STATE_OFST 20 #define MC_CMD_PARSER_DISP_RW_OUT_LUE_MGR_STATE_LEN 32 +/* datapath(s) used for each port (for MISC_STATE PORT_DP_MAPPING selector) */ +#define MC_CMD_PARSER_DISP_RW_OUT_PORT_DP_MAPPING_OFST 0 +#define MC_CMD_PARSER_DISP_RW_OUT_PORT_DP_MAPPING_LEN 4 +#define MC_CMD_PARSER_DISP_RW_OUT_PORT_DP_MAPPING_NUM 4 +#define MC_CMD_PARSER_DISP_RW_OUT_DP0 0x1 /* enum */ +#define MC_CMD_PARSER_DISP_RW_OUT_DP1 0x2 /* enum */ /***********************************/ @@ -7071,6 +7480,24 @@ #define MC_CMD_GET_CAPABILITIES_OUT_LEN 20 /* First word of flags. */ #define MC_CMD_GET_CAPABILITIES_OUT_FLAGS1_OFST 0 +#define MC_CMD_GET_CAPABILITIES_OUT_VPORT_RECONFIGURE_LBN 3 +#define MC_CMD_GET_CAPABILITIES_OUT_VPORT_RECONFIGURE_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_TX_STRIPING_LBN 4 +#define MC_CMD_GET_CAPABILITIES_OUT_TX_STRIPING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_VADAPTOR_QUERY_LBN 5 +#define MC_CMD_GET_CAPABILITIES_OUT_VADAPTOR_QUERY_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_EVB_PORT_VLAN_RESTRICT_LBN 6 +#define MC_CMD_GET_CAPABILITIES_OUT_EVB_PORT_VLAN_RESTRICT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_DRV_ATTACH_PREBOOT_LBN 7 +#define MC_CMD_GET_CAPABILITIES_OUT_DRV_ATTACH_PREBOOT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_RX_FORCE_EVENT_MERGING_LBN 8 +#define MC_CMD_GET_CAPABILITIES_OUT_RX_FORCE_EVENT_MERGING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_SET_MAC_ENHANCED_LBN 9 +#define MC_CMD_GET_CAPABILITIES_OUT_SET_MAC_ENHANCED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_LBN 10 +#define MC_CMD_GET_CAPABILITIES_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_LBN 11 +#define MC_CMD_GET_CAPABILITIES_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_WIDTH 1 #define MC_CMD_GET_CAPABILITIES_OUT_TX_MAC_SECURITY_FILTERING_LBN 12 #define MC_CMD_GET_CAPABILITIES_OUT_TX_MAC_SECURITY_FILTERING_WIDTH 1 #define MC_CMD_GET_CAPABILITIES_OUT_ADDITIONAL_RSS_MODES_LBN 13 @@ -7138,6 +7565,8 @@ #define MC_CMD_GET_CAPABILITIES_OUT_RXDP_TEST_FW_RX_HDR_SPLIT 0x107 /* enum: RXDP Test firmware image 8 */ #define MC_CMD_GET_CAPABILITIES_OUT_RXDP_TEST_FW_DISABLE_DL 0x108 +/* enum: RXDP Test firmware image 9 */ +#define MC_CMD_GET_CAPABILITIES_OUT_RXDP_TEST_FW_DOORBELL_DELAY 0x10b /* TxDPCPU firmware id. */ #define MC_CMD_GET_CAPABILITIES_OUT_TX_DPCPU_FW_ID_OFST 6 #define MC_CMD_GET_CAPABILITIES_OUT_TX_DPCPU_FW_ID_LEN 2 @@ -7153,6 +7582,8 @@ #define MC_CMD_GET_CAPABILITIES_OUT_TXDP_TEST_FW_TSO_EDIT 0x101 /* enum: TXDP Test firmware image 2 */ #define MC_CMD_GET_CAPABILITIES_OUT_TXDP_TEST_FW_PACKET_EDITS 0x102 +/* enum: TXDP CSR bus test firmware */ +#define MC_CMD_GET_CAPABILITIES_OUT_TXDP_TEST_FW_CSR 0x103 #define MC_CMD_GET_CAPABILITIES_OUT_RXPD_FW_VERSION_OFST 8 #define MC_CMD_GET_CAPABILITIES_OUT_RXPD_FW_VERSION_LEN 2 #define MC_CMD_GET_CAPABILITIES_OUT_RXPD_FW_VERSION_REV_LBN 0 @@ -7227,6 +7658,258 @@ /* Licensed capabilities */ #define MC_CMD_GET_CAPABILITIES_OUT_LICENSE_CAPABILITIES_OFST 16 +/* MC_CMD_GET_CAPABILITIES_V2_IN msgrequest */ +#define MC_CMD_GET_CAPABILITIES_V2_IN_LEN 0 + +/* MC_CMD_GET_CAPABILITIES_V2_OUT msgresponse */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_LEN 72 +/* First word of flags. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_FLAGS1_OFST 0 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VPORT_RECONFIGURE_LBN 3 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VPORT_RECONFIGURE_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_STRIPING_LBN 4 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_STRIPING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VADAPTOR_QUERY_LBN 5 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VADAPTOR_QUERY_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVB_PORT_VLAN_RESTRICT_LBN 6 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVB_PORT_VLAN_RESTRICT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_DRV_ATTACH_PREBOOT_LBN 7 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_DRV_ATTACH_PREBOOT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_FORCE_EVENT_MERGING_LBN 8 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_FORCE_EVENT_MERGING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_SET_MAC_ENHANCED_LBN 9 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_SET_MAC_ENHANCED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_LBN 10 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_LBN 11 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MAC_SECURITY_FILTERING_LBN 12 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MAC_SECURITY_FILTERING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_ADDITIONAL_RSS_MODES_LBN 13 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_ADDITIONAL_RSS_MODES_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_QBB_LBN 14 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_QBB_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PACKED_STREAM_VAR_BUFFERS_LBN 15 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PACKED_STREAM_VAR_BUFFERS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_RSS_LIMITED_LBN 16 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_RSS_LIMITED_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PACKED_STREAM_LBN 17 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PACKED_STREAM_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_INCLUDE_FCS_LBN 18 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_INCLUDE_FCS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_VLAN_INSERTION_LBN 19 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_VLAN_INSERTION_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_VLAN_STRIPPING_LBN 20 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_VLAN_STRIPPING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_LBN 21 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PREFIX_LEN_0_LBN 22 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PREFIX_LEN_0_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PREFIX_LEN_14_LBN 23 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_PREFIX_LEN_14_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_TIMESTAMP_LBN 24 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_TIMESTAMP_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_BATCHING_LBN 25 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_BATCHING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_MCAST_FILTER_CHAINING_LBN 26 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_MCAST_FILTER_CHAINING_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PM_AND_RXDP_COUNTERS_LBN 27 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PM_AND_RXDP_COUNTERS_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DISABLE_SCATTER_LBN 28 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DISABLE_SCATTER_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MCAST_UDP_LOOPBACK_LBN 29 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MCAST_UDP_LOOPBACK_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVB_LBN 30 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVB_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VXLAN_NVGRE_LBN 31 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_VXLAN_NVGRE_WIDTH 1 +/* RxDPCPU firmware id. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DPCPU_FW_ID_OFST 4 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DPCPU_FW_ID_LEN 2 +/* enum: Standard RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP 0x0 +/* enum: Low latency RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_LOW_LATENCY 0x1 +/* enum: Packed stream RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_PACKED_STREAM 0x2 +/* enum: BIST RXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_BIST 0x10a +/* enum: RXDP Test firmware image 1 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_TO_MC_CUT_THROUGH 0x101 +/* enum: RXDP Test firmware image 2 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_TO_MC_STORE_FORWARD 0x102 +/* enum: RXDP Test firmware image 3 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_TO_MC_STORE_FORWARD_FIRST 0x103 +/* enum: RXDP Test firmware image 4 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_EVERY_EVENT_BATCHABLE 0x104 +/* enum: RXDP Test firmware image 5 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_BACKPRESSURE 0x105 +/* enum: RXDP Test firmware image 6 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_PACKET_EDITS 0x106 +/* enum: RXDP Test firmware image 7 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_RX_HDR_SPLIT 0x107 +/* enum: RXDP Test firmware image 8 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_DISABLE_DL 0x108 +/* enum: RXDP Test firmware image 9 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXDP_TEST_FW_DOORBELL_DELAY 0x10b +/* TxDPCPU firmware id. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_DPCPU_FW_ID_OFST 6 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_DPCPU_FW_ID_LEN 2 +/* enum: Standard TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP 0x0 +/* enum: Low latency TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_LOW_LATENCY 0x1 +/* enum: High packet rate TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_HIGH_PACKET_RATE 0x3 +/* enum: BIST TXDP firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_BIST 0x12d +/* enum: TXDP Test firmware image 1 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_TEST_FW_TSO_EDIT 0x101 +/* enum: TXDP Test firmware image 2 */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_TEST_FW_PACKET_EDITS 0x102 +/* enum: TXDP CSR bus test firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXDP_TEST_FW_CSR 0x103 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_OFST 8 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_LEN 2 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_REV_LBN 0 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_REV_WIDTH 12 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_TYPE_LBN 12 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_VERSION_TYPE_WIDTH 4 +/* enum: reserved value - do not use (may indicate alternative interpretation + * of REV field in future) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_RESERVED 0x0 +/* enum: Trivial RX PD firmware for early Huntington development (Huntington + * development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_FIRST_PKT 0x1 +/* enum: RX PD firmware with approximately Siena-compatible behaviour + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_SIENA_COMPAT 0x2 +/* enum: Virtual switching (full feature) RX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_VSWITCH 0x3 +/* enum: siena_compat variant RX PD firmware using PM rather than MAC + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_SIENA_COMPAT_PM 0x4 +/* enum: Low latency RX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_LOW_LATENCY 0x5 +/* enum: Packed stream RX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_PACKED_STREAM 0x6 +/* enum: RX PD firmware handling layer 2 only for high packet rate performance + * tests (Medford development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_LAYER2_PERF 0x7 +/* enum: RX PD firmware for GUE parsing prototype (Medford development only) */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_TESTFW_GUE_PROTOTYPE 0xe +/* enum: RX PD firmware parsing but not filtering network overlay tunnel + * encapsulations (Medford development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RXPD_FW_TYPE_TESTFW_ENCAP_PARSING_ONLY 0xf +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_OFST 10 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_LEN 2 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_REV_LBN 0 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_REV_WIDTH 12 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_TYPE_LBN 12 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_VERSION_TYPE_WIDTH 4 +/* enum: reserved value - do not use (may indicate alternative interpretation + * of REV field in future) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_RESERVED 0x0 +/* enum: Trivial TX PD firmware for early Huntington development (Huntington + * development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_FIRST_PKT 0x1 +/* enum: TX PD firmware with approximately Siena-compatible behaviour + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_SIENA_COMPAT 0x2 +/* enum: Virtual switching (full feature) TX PD production firmware */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_VSWITCH 0x3 +/* enum: siena_compat variant TX PD firmware using PM rather than MAC + * (Huntington development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_SIENA_COMPAT_PM 0x4 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_LOW_LATENCY 0x5 /* enum */ +/* enum: TX PD firmware handling layer 2 only for high packet rate performance + * tests (Medford development only) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_LAYER2_PERF 0x7 +/* enum: RX PD firmware for GUE parsing prototype (Medford development only) */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TXPD_FW_TYPE_TESTFW_GUE_PROTOTYPE 0xe +/* Hardware capabilities of NIC */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_HW_CAPABILITIES_OFST 12 +/* Licensed capabilities */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_LICENSE_CAPABILITIES_OFST 16 +/* Second word of flags. Not present on older firmware (check the length). */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_FLAGS2_OFST 20 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_LBN 0 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_ENCAP_LBN 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_ENCAP_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVQ_TIMER_CTRL_LBN 2 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVQ_TIMER_CTRL_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVENT_CUT_THROUGH_LBN 3 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVENT_CUT_THROUGH_WIDTH 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_CUT_THROUGH_LBN 4 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_CUT_THROUGH_WIDTH 1 +/* Number of FATSOv2 contexts per datapath supported by this NIC. Not present + * on older firmware (check the length). + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_N_CONTEXTS_OFST 24 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TSO_V2_N_CONTEXTS_LEN 2 +/* One byte per PF containing the number of the external port assigned to this + * PF, indexed by PF number. Special values indicate that a PF is either not + * present or not assigned. + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PFS_TO_PORTS_ASSIGNMENT_OFST 26 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PFS_TO_PORTS_ASSIGNMENT_LEN 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PFS_TO_PORTS_ASSIGNMENT_NUM 16 +/* enum: The caller is not permitted to access information on this PF. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_ACCESS_NOT_PERMITTED 0xff +/* enum: PF does not exist. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PF_NOT_PRESENT 0xfe +/* enum: PF does exist but is not assigned to any external port. */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_PF_NOT_ASSIGNED 0xfd +/* enum: This value indicates that PF is assigned, but it cannot be expressed + * in this field. It is intended for a possible future situation where a more + * complex scheme of PFs to ports mapping is being used. The future driver + * should look for a new field supporting the new scheme. The current/old + * driver should treat this value as PF_NOT_ASSIGNED. + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_INCOMPATIBLE_ASSIGNMENT 0xfc +/* One byte per PF containing the number of its VFs, indexed by PF number. A + * special value indicates that a PF is not present. + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VFS_PER_PF_OFST 42 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VFS_PER_PF_LEN 1 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VFS_PER_PF_NUM 16 +/* enum: The caller is not permitted to access information on this PF. */ +/* MC_CMD_GET_CAPABILITIES_V2_OUT_ACCESS_NOT_PERMITTED 0xff */ +/* enum: PF does not exist. */ +/* MC_CMD_GET_CAPABILITIES_V2_OUT_PF_NOT_PRESENT 0xfe */ +/* Number of VIs available for each external port */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VIS_PER_PORT_OFST 58 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VIS_PER_PORT_LEN 2 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_VIS_PER_PORT_NUM 4 +/* Size of RX descriptor cache expressed as binary logarithm The actual size + * equals (2 ^ RX_DESC_CACHE_SIZE) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DESC_CACHE_SIZE_OFST 66 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_DESC_CACHE_SIZE_LEN 1 +/* Size of TX descriptor cache expressed as binary logarithm The actual size + * equals (2 ^ TX_DESC_CACHE_SIZE) + */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_DESC_CACHE_SIZE_OFST 67 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_DESC_CACHE_SIZE_LEN 1 +/* Total number of available PIO buffers */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_PIO_BUFFS_OFST 68 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_NUM_PIO_BUFFS_LEN 2 +/* Size of a single PIO buffer */ +#define MC_CMD_GET_CAPABILITIES_V2_OUT_SIZE_PIO_BUFF_OFST 70 +#define MC_CMD_GET_CAPABILITIES_V2_OUT_SIZE_PIO_BUFF_LEN 2 + /***********************************/ /* MC_CMD_V2_EXTN @@ -7475,6 +8158,25 @@ /***********************************/ +/* MC_CMD_VSWITCH_QUERY + * read some config of v-switch. For now this command is an empty placeholder. + * It may be used to check if a v-switch is connected to a given EVB port (if + * not, then the command returns ENOENT). + */ +#define MC_CMD_VSWITCH_QUERY 0x63 + +#define MC_CMD_0x63_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_VSWITCH_QUERY_IN msgrequest */ +#define MC_CMD_VSWITCH_QUERY_IN_LEN 4 +/* The port to which the v-switch is connected. */ +#define MC_CMD_VSWITCH_QUERY_IN_UPSTREAM_PORT_ID_OFST 0 + +/* MC_CMD_VSWITCH_QUERY_OUT msgresponse */ +#define MC_CMD_VSWITCH_QUERY_OUT_LEN 0 + + +/***********************************/ /* MC_CMD_VPORT_ALLOC * allocate a v-port. */ @@ -7510,6 +8212,8 @@ #define MC_CMD_VPORT_ALLOC_IN_FLAGS_OFST 8 #define MC_CMD_VPORT_ALLOC_IN_FLAG_AUTO_PORT_LBN 0 #define MC_CMD_VPORT_ALLOC_IN_FLAG_AUTO_PORT_WIDTH 1 +#define MC_CMD_VPORT_ALLOC_IN_FLAG_VLAN_RESTRICT_LBN 1 +#define MC_CMD_VPORT_ALLOC_IN_FLAG_VLAN_RESTRICT_WIDTH 1 /* The number of VLAN tags to insert/remove. An error will be returned if * incompatible with the number of VLAN tags specified for the upstream * v-switch. @@ -7561,6 +8265,8 @@ #define MC_CMD_VADAPTOR_ALLOC_IN_FLAGS_OFST 8 #define MC_CMD_VADAPTOR_ALLOC_IN_FLAG_AUTO_VADAPTOR_LBN 0 #define MC_CMD_VADAPTOR_ALLOC_IN_FLAG_AUTO_VADAPTOR_WIDTH 1 +#define MC_CMD_VADAPTOR_ALLOC_IN_FLAG_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_LBN 1 +#define MC_CMD_VADAPTOR_ALLOC_IN_FLAG_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_WIDTH 1 /* The number of VLAN tags to strip on receive */ #define MC_CMD_VADAPTOR_ALLOC_IN_NUM_VLANS_OFST 12 /* The number of VLAN tags to transparently insert/remove. */ @@ -7639,6 +8345,29 @@ /***********************************/ +/* MC_CMD_VADAPTOR_QUERY + * read some config of v-adaptor. + */ +#define MC_CMD_VADAPTOR_QUERY 0x61 + +#define MC_CMD_0x61_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_VADAPTOR_QUERY_IN msgrequest */ +#define MC_CMD_VADAPTOR_QUERY_IN_LEN 4 +/* The port to which the v-adaptor is connected. */ +#define MC_CMD_VADAPTOR_QUERY_IN_UPSTREAM_PORT_ID_OFST 0 + +/* MC_CMD_VADAPTOR_QUERY_OUT msgresponse */ +#define MC_CMD_VADAPTOR_QUERY_OUT_LEN 12 +/* The EVB port flags as defined at MC_CMD_VPORT_ALLOC. */ +#define MC_CMD_VADAPTOR_QUERY_OUT_PORT_FLAGS_OFST 0 +/* The v-adaptor flags as defined at MC_CMD_VADAPTOR_ALLOC. */ +#define MC_CMD_VADAPTOR_QUERY_OUT_VADAPTOR_FLAGS_OFST 4 +/* The number of VLAN tags that may still be added */ +#define MC_CMD_VADAPTOR_QUERY_OUT_NUM_AVAILABLE_VLAN_TAGS_OFST 8 + + +/***********************************/ /* MC_CMD_EVB_PORT_ASSIGN * assign a port to a PCI function. */ @@ -7875,10 +8604,17 @@ #define MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_LEN 8 /* The handle of the RSS context */ #define MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_RSS_CONTEXT_ID_OFST 0 -/* Hash control flags. The _EN bits are always supported. The _MODE bits only - * work when the firmware reports ADDITIONAL_RSS_MODES in - * MC_CMD_GET_CAPABILITIES and override the _EN bits if any of them are not 0. - * See the RSS_MODE structure for the meaning of the mode bits. +/* Hash control flags. The _EN bits are always supported, but new modes are + * available when ADDITIONAL_RSS_MODES is reported by MC_CMD_GET_CAPABILITIES: + * in this case, the MODE fields may be set to non-zero values, and will take + * effect regardless of the settings of the _EN flags. See the RSS_MODE + * structure for the meaning of the mode bits. Drivers must check the + * capability before trying to set any _MODE fields, as older firmware will + * reject any attempt to set the FLAGS field to a value > 0xff with EINVAL. In + * the case where all the _MODE flags are zero, the _EN flags take effect, + * providing backward compatibility for existing drivers. (Setting all _MODE + * *and* all _EN flags to zero is valid, to disable RSS spreading for that + * particular packet type.) */ #define MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_FLAGS_OFST 4 #define MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_TOEPLITZ_IPV4_EN_LBN 0 @@ -7923,11 +8659,18 @@ /* MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT msgresponse */ #define MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN 8 -/* Hash control flags. If any _MODE bits are non-zero (which will only be true - * when the firmware reports ADDITIONAL_RSS_MODES) then the _EN bits should be - * disregarded (but are guaranteed to be consistent with the _MODE bits if - * RSS_CONTEXT_SET_FLAGS has never been called for this context since it was - * allocated). +/* Hash control flags. If all _MODE bits are zero (which will always be true + * for older firmware which does not report the ADDITIONAL_RSS_MODES + * capability), the _EN bits report the state. If any _MODE bits are non-zero + * (which will only be true when the firmware reports ADDITIONAL_RSS_MODES) + * then the _EN bits should be disregarded, although the _MODE flags are + * guaranteed to be consistent with the _EN flags for a freshly-allocated RSS + * context and in the case where the _EN flags were used in the SET. This + * provides backward compatibility: old drivers will not be attempting to + * derive any meaning from the _MODE bits (and can never set them to any value + * not representable by the _EN bits); new drivers can always determine the + * mode by looking only at the _MODE bits; the value returned by a GET can + * always be used for a SET regardless of old/new driver vs. old/new firmware. */ #define MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_FLAGS_OFST 4 #define MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_IPV4_EN_LBN 0 @@ -8155,6 +8898,74 @@ /***********************************/ +/* MC_CMD_VPORT_RECONFIGURE + * Replace VLAN tags and/or MAC addresses of an existing v-port. If the v-port + * has already been passed to another function (v-port's user), then that + * function will be reset before applying the changes. + */ +#define MC_CMD_VPORT_RECONFIGURE 0xeb + +#define MC_CMD_0xeb_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_VPORT_RECONFIGURE_IN msgrequest */ +#define MC_CMD_VPORT_RECONFIGURE_IN_LEN 44 +/* The handle of the v-port */ +#define MC_CMD_VPORT_RECONFIGURE_IN_VPORT_ID_OFST 0 +/* Flags requesting what should be changed. */ +#define MC_CMD_VPORT_RECONFIGURE_IN_FLAGS_OFST 4 +#define MC_CMD_VPORT_RECONFIGURE_IN_REPLACE_VLAN_TAGS_LBN 0 +#define MC_CMD_VPORT_RECONFIGURE_IN_REPLACE_VLAN_TAGS_WIDTH 1 +#define MC_CMD_VPORT_RECONFIGURE_IN_REPLACE_MACADDRS_LBN 1 +#define MC_CMD_VPORT_RECONFIGURE_IN_REPLACE_MACADDRS_WIDTH 1 +/* The number of VLAN tags to insert/remove. An error will be returned if + * incompatible with the number of VLAN tags specified for the upstream + * v-switch. + */ +#define MC_CMD_VPORT_RECONFIGURE_IN_NUM_VLAN_TAGS_OFST 8 +/* The actual VLAN tags to insert/remove */ +#define MC_CMD_VPORT_RECONFIGURE_IN_VLAN_TAGS_OFST 12 +#define MC_CMD_VPORT_RECONFIGURE_IN_VLAN_TAG_0_LBN 0 +#define MC_CMD_VPORT_RECONFIGURE_IN_VLAN_TAG_0_WIDTH 16 +#define MC_CMD_VPORT_RECONFIGURE_IN_VLAN_TAG_1_LBN 16 +#define MC_CMD_VPORT_RECONFIGURE_IN_VLAN_TAG_1_WIDTH 16 +/* The number of MAC addresses to add */ +#define MC_CMD_VPORT_RECONFIGURE_IN_NUM_MACADDRS_OFST 16 +/* MAC addresses to add */ +#define MC_CMD_VPORT_RECONFIGURE_IN_MACADDRS_OFST 20 +#define MC_CMD_VPORT_RECONFIGURE_IN_MACADDRS_LEN 6 +#define MC_CMD_VPORT_RECONFIGURE_IN_MACADDRS_NUM 4 + +/* MC_CMD_VPORT_RECONFIGURE_OUT msgresponse */ +#define MC_CMD_VPORT_RECONFIGURE_OUT_LEN 4 +#define MC_CMD_VPORT_RECONFIGURE_OUT_FLAGS_OFST 0 +#define MC_CMD_VPORT_RECONFIGURE_OUT_RESET_DONE_LBN 0 +#define MC_CMD_VPORT_RECONFIGURE_OUT_RESET_DONE_WIDTH 1 + + +/***********************************/ +/* MC_CMD_EVB_PORT_QUERY + * read some config of v-port. + */ +#define MC_CMD_EVB_PORT_QUERY 0x62 + +#define MC_CMD_0x62_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_EVB_PORT_QUERY_IN msgrequest */ +#define MC_CMD_EVB_PORT_QUERY_IN_LEN 4 +/* The handle of the v-port */ +#define MC_CMD_EVB_PORT_QUERY_IN_PORT_ID_OFST 0 + +/* MC_CMD_EVB_PORT_QUERY_OUT msgresponse */ +#define MC_CMD_EVB_PORT_QUERY_OUT_LEN 8 +/* The EVB port flags as defined at MC_CMD_VPORT_ALLOC. */ +#define MC_CMD_EVB_PORT_QUERY_OUT_PORT_FLAGS_OFST 0 +/* The number of VLAN tags that may be used on a v-adaptor connected to this + * EVB port. + */ +#define MC_CMD_EVB_PORT_QUERY_OUT_NUM_AVAILABLE_VLAN_TAGS_OFST 4 + + +/***********************************/ /* MC_CMD_DUMP_BUFTBL_ENTRIES * Dump buffer table entries, mainly for command client debug use. Dumps * absolute entries, and does not use chunk handles. All entries must be in @@ -8196,6 +9007,14 @@ #define MC_CMD_SET_RXDP_CONFIG_IN_DATA_OFST 0 #define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_DMA_LBN 0 #define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_DMA_WIDTH 1 +#define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_LEN_LBN 1 +#define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_LEN_WIDTH 2 +/* enum: pad to 64 bytes */ +#define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_64 0x0 +/* enum: pad to 128 bytes (Medford only) */ +#define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_128 0x1 +/* enum: pad to 256 bytes (Medford only) */ +#define MC_CMD_SET_RXDP_CONFIG_IN_PAD_HOST_256 0x2 /* MC_CMD_SET_RXDP_CONFIG_OUT msgresponse */ #define MC_CMD_SET_RXDP_CONFIG_OUT_LEN 0 @@ -8217,6 +9036,10 @@ #define MC_CMD_GET_RXDP_CONFIG_OUT_DATA_OFST 0 #define MC_CMD_GET_RXDP_CONFIG_OUT_PAD_HOST_DMA_LBN 0 #define MC_CMD_GET_RXDP_CONFIG_OUT_PAD_HOST_DMA_WIDTH 1 +#define MC_CMD_GET_RXDP_CONFIG_OUT_PAD_HOST_LEN_LBN 1 +#define MC_CMD_GET_RXDP_CONFIG_OUT_PAD_HOST_LEN_WIDTH 2 +/* Enum values, see field(s): */ +/* MC_CMD_SET_RXDP_CONFIG/MC_CMD_SET_RXDP_CONFIG_IN/PAD_HOST_LEN */ /***********************************/ @@ -8788,32 +9611,38 @@ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_PARAM_MAXNUM 63 #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_PARAM_ID_LBN 0 #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_PARAM_ID_WIDTH 8 -/* enum: Attenuation (0-15, TBD for Medford) */ +/* enum: Attenuation (0-15, Huntington) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_ATT 0x0 -/* enum: CTLE Boost (0-15, TBD for Medford) */ +/* enum: CTLE Boost (0-15, Huntington) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_BOOST 0x1 -/* enum: Edge DFE Tap1 (0 - max negative, 64 - zero, 127 - max positive, TBD - * for Medford) +/* enum: Edge DFE Tap1 (Huntington - 0 - max negative, 64 - zero, 127 - max + * positive, Medford - 0-31) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_TAP1 0x2 -/* enum: Edge DFE Tap2 (0 - max negative, 32 - zero, 63 - max positive, TBD for - * Medford) +/* enum: Edge DFE Tap2 (Huntington - 0 - max negative, 32 - zero, 63 - max + * positive, Medford - 0-31) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_TAP2 0x3 -/* enum: Edge DFE Tap3 (0 - max negative, 32 - zero, 63 - max positive, TBD for - * Medford) +/* enum: Edge DFE Tap3 (Huntington - 0 - max negative, 32 - zero, 63 - max + * positive, Medford - 0-16) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_TAP3 0x4 -/* enum: Edge DFE Tap4 (0 - max negative, 32 - zero, 63 - max positive, TBD for - * Medford) +/* enum: Edge DFE Tap4 (Huntington - 0 - max negative, 32 - zero, 63 - max + * positive, Medford - 0-16) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_TAP4 0x5 -/* enum: Edge DFE Tap5 (0 - max negative, 32 - zero, 63 - max positive, TBD for - * Medford) +/* enum: Edge DFE Tap5 (Huntington - 0 - max negative, 32 - zero, 63 - max + * positive, Medford - 0-16) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_TAP5 0x6 -/* enum: Edge DFE DLEV (TBD for Medford) */ +/* enum: Edge DFE DLEV (0-128 for Medford) */ #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_EDFE_DLEV 0x7 +/* enum: Variable Gain Amplifier (0-15, Medford) */ +#define MC_CMD_KR_TUNE_RXEQ_GET_OUT_VGA 0x8 +/* enum: CTLE EQ Capacitor (0-15, Medford) */ +#define MC_CMD_KR_TUNE_RXEQ_GET_OUT_CTLE_EQC 0x9 +/* enum: CTLE EQ Resistor (0-7, Medford) */ +#define MC_CMD_KR_TUNE_RXEQ_GET_OUT_CTLE_EQRES 0xa #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_PARAM_LANE_LBN 8 #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_PARAM_LANE_WIDTH 3 #define MC_CMD_KR_TUNE_RXEQ_GET_OUT_LANE_0 0x0 /* enum */ @@ -8885,26 +9714,32 @@ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_PARAM_MAXNUM 63 #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_PARAM_ID_LBN 0 #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_PARAM_ID_WIDTH 8 -/* enum: TX Amplitude */ +/* enum: TX Amplitude (Huntington, Medford) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_LEV 0x0 -/* enum: De-Emphasis Tap1 Magnitude (0-7) */ +/* enum: De-Emphasis Tap1 Magnitude (0-7) (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_MODE 0x1 /* enum: De-Emphasis Tap1 Fine */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_DTLEV 0x2 -/* enum: De-Emphasis Tap2 Magnitude (0-6) */ +/* enum: De-Emphasis Tap2 Magnitude (0-6) (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_D2 0x3 -/* enum: De-Emphasis Tap2 Fine */ +/* enum: De-Emphasis Tap2 Fine (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_D2TLEV 0x4 -/* enum: Pre-Emphasis Magnitude */ +/* enum: Pre-Emphasis Magnitude (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_E 0x5 -/* enum: Pre-Emphasis Fine */ +/* enum: Pre-Emphasis Fine (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_ETLEV 0x6 -/* enum: TX Slew Rate Coarse control */ +/* enum: TX Slew Rate Coarse control (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_PREDRV_DLY 0x7 -/* enum: TX Slew Rate Fine control */ +/* enum: TX Slew Rate Fine control (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_SR_SET 0x8 -/* enum: TX Termination Impedance control */ +/* enum: TX Termination Impedance control (Huntington) */ #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_RT_SET 0x9 +/* enum: TX Amplitude Fine control (Medford) */ +#define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TX_LEV_FINE 0xa +/* enum: Pre-shoot Tap (Medford) */ +#define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TAP_ADV 0xb +/* enum: De-emphasis Tap (Medford) */ +#define MC_CMD_KR_TUNE_TXEQ_GET_OUT_TAP_DLY 0xc #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_PARAM_LANE_LBN 8 #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_PARAM_LANE_WIDTH 3 #define MC_CMD_KR_TUNE_TXEQ_GET_OUT_LANE_0 0x0 /* enum */ @@ -9086,8 +9921,16 @@ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_DFE_TAP4 0x5 /* enum: DFE Tap5 (0 - max negative, 32 - zero, 63 - max positive) */ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_DFE_TAP5 0x6 +/* enum: DFE DLev */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_DFE_DLEV 0x7 +/* enum: Figure of Merit */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_FOM 0x8 +/* enum: CTLE EQ Capacitor (HF Gain) */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_CTLE_EQC 0x9 +/* enum: CTLE EQ Resistor (DC Gain) */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_CTLE_EQRES 0xa #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_LANE_LBN 8 -#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_LANE_WIDTH 4 +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_LANE_WIDTH 5 #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_0 0x0 /* enum */ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_1 0x1 /* enum */ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_2 0x2 /* enum */ @@ -9096,12 +9939,57 @@ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_5 0x5 /* enum */ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_6 0x6 /* enum */ #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_7 0x7 /* enum */ -#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_ALL 0x8 /* enum */ -#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_RESERVED_LBN 12 -#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_RESERVED_WIDTH 12 +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_8 0x8 /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_9 0x9 /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_10 0xa /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_11 0xb /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_12 0xc /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_13 0xd /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_14 0xe /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_15 0xf /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_LANE_ALL 0x10 /* enum */ +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_AUTOCAL_LBN 13 +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_AUTOCAL_WIDTH 1 +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_RESERVED_LBN 14 +#define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_RESERVED_WIDTH 10 #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_CURRENT_LBN 24 #define MC_CMD_PCIE_TUNE_RXEQ_GET_OUT_PARAM_CURRENT_WIDTH 8 +/* MC_CMD_PCIE_TUNE_RXEQ_SET_IN msgrequest */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_LENMIN 8 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_LENMAX 252 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_LEN(num) (4+4*(num)) +/* Requested operation */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PCIE_TUNE_OP_OFST 0 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PCIE_TUNE_OP_LEN 1 +/* Align the arguments to 32 bits */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PCIE_TUNE_RSVD_OFST 1 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PCIE_TUNE_RSVD_LEN 3 +/* RXEQ Parameter */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_OFST 4 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_LEN 4 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_MINNUM 1 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_MAXNUM 62 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_ID_LBN 0 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_ID_WIDTH 8 +/* Enum values, see field(s): */ +/* MC_CMD_PCIE_TUNE_RXEQ_GET_OUT/PARAM_ID */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_LANE_LBN 8 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_LANE_WIDTH 5 +/* Enum values, see field(s): */ +/* MC_CMD_PCIE_TUNE_RXEQ_GET_OUT/PARAM_LANE */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_AUTOCAL_LBN 13 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_AUTOCAL_WIDTH 1 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_RESERVED_LBN 14 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_RESERVED_WIDTH 2 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_INITIAL_LBN 16 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_PARAM_INITIAL_WIDTH 8 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_RESERVED2_LBN 24 +#define MC_CMD_PCIE_TUNE_RXEQ_SET_IN_RESERVED2_WIDTH 8 + +/* MC_CMD_PCIE_TUNE_RXEQ_SET_OUT msgresponse */ +#define MC_CMD_PCIE_TUNE_RXEQ_SET_OUT_LEN 0 + /* MC_CMD_PCIE_TUNE_TXEQ_GET_IN msgrequest */ #define MC_CMD_PCIE_TUNE_TXEQ_GET_IN_LEN 4 /* Requested operation */ @@ -9176,6 +10064,7 @@ /***********************************/ /* MC_CMD_LICENSING * Operations on the NVRAM_PARTITION_TYPE_LICENSE application license partition + * - not used for V3 licensing */ #define MC_CMD_LICENSING 0xf3 @@ -9220,6 +10109,93 @@ /***********************************/ +/* MC_CMD_LICENSING_V3 + * Operations on the NVRAM_PARTITION_TYPE_LICENSE application license partition + * - V3 licensing (Medford) + */ +#define MC_CMD_LICENSING_V3 0xd0 + +#define MC_CMD_0xd0_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_LICENSING_V3_IN msgrequest */ +#define MC_CMD_LICENSING_V3_IN_LEN 4 +/* identifies the type of operation requested */ +#define MC_CMD_LICENSING_V3_IN_OP_OFST 0 +/* enum: re-read and apply licenses after a license key partition update; note + * that this operation returns a zero-length response + */ +#define MC_CMD_LICENSING_V3_IN_OP_UPDATE_LICENSE 0x0 +/* enum: report counts of installed licenses */ +#define MC_CMD_LICENSING_V3_IN_OP_REPORT_LICENSE 0x1 + +/* MC_CMD_LICENSING_V3_OUT msgresponse */ +#define MC_CMD_LICENSING_V3_OUT_LEN 88 +/* count of keys which are valid */ +#define MC_CMD_LICENSING_V3_OUT_VALID_KEYS_OFST 0 +/* sum of UNVERIFIABLE_KEYS + WRONG_NODE_KEYS (for compatibility with + * MC_CMD_FC_OP_LICENSE) + */ +#define MC_CMD_LICENSING_V3_OUT_INVALID_KEYS_OFST 4 +/* count of keys which are invalid due to being unverifiable */ +#define MC_CMD_LICENSING_V3_OUT_UNVERIFIABLE_KEYS_OFST 8 +/* count of keys which are invalid due to being for the wrong node */ +#define MC_CMD_LICENSING_V3_OUT_WRONG_NODE_KEYS_OFST 12 +/* licensing state (for diagnostics; the exact meaning of the bits in this + * field are private to the firmware) + */ +#define MC_CMD_LICENSING_V3_OUT_LICENSING_STATE_OFST 16 +/* licensing subsystem self-test report (for manftest) */ +#define MC_CMD_LICENSING_V3_OUT_LICENSING_SELF_TEST_OFST 20 +/* enum: licensing subsystem self-test failed */ +#define MC_CMD_LICENSING_V3_OUT_SELF_TEST_FAIL 0x0 +/* enum: licensing subsystem self-test passed */ +#define MC_CMD_LICENSING_V3_OUT_SELF_TEST_PASS 0x1 +/* bitmask of licensed applications */ +#define MC_CMD_LICENSING_V3_OUT_LICENSED_APPS_OFST 24 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_APPS_LEN 8 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_APPS_LO_OFST 24 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_APPS_HI_OFST 28 +/* reserved for future use */ +#define MC_CMD_LICENSING_V3_OUT_RESERVED_0_OFST 32 +#define MC_CMD_LICENSING_V3_OUT_RESERVED_0_LEN 24 +/* bitmask of licensed features */ +#define MC_CMD_LICENSING_V3_OUT_LICENSED_FEATURES_OFST 56 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_FEATURES_LEN 8 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_FEATURES_LO_OFST 56 +#define MC_CMD_LICENSING_V3_OUT_LICENSED_FEATURES_HI_OFST 60 +/* reserved for future use */ +#define MC_CMD_LICENSING_V3_OUT_RESERVED_1_OFST 64 +#define MC_CMD_LICENSING_V3_OUT_RESERVED_1_LEN 24 + + +/***********************************/ +/* MC_CMD_LICENSING_GET_ID_V3 + * Get ID and type from the NVRAM_PARTITION_TYPE_LICENSE application license + * partition - V3 licensing (Medford) + */ +#define MC_CMD_LICENSING_GET_ID_V3 0xd1 + +#define MC_CMD_0xd1_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_LICENSING_GET_ID_V3_IN msgrequest */ +#define MC_CMD_LICENSING_GET_ID_V3_IN_LEN 0 + +/* MC_CMD_LICENSING_GET_ID_V3_OUT msgresponse */ +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LENMIN 8 +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LENMAX 252 +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LEN(num) (8+1*(num)) +/* type of license (eg 3) */ +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_TYPE_OFST 0 +/* length of the license ID (in bytes) */ +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_LENGTH_OFST 4 +/* the unique license ID of the adapter */ +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_OFST 8 +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_LEN 1 +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_MINNUM 0 +#define MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_MAXNUM 244 + + +/***********************************/ /* MC_CMD_MC2MC_PROXY * Execute an arbitrary MCDI command on the slave MC of a dual-core device. * This will fail on a single-core system. @@ -9239,7 +10215,7 @@ /* MC_CMD_GET_LICENSED_APP_STATE * Query the state of an individual licensed application. (Note that the actual * state may be invalidated by the MC_CMD_LICENSING OP_UPDATE_LICENSE operation - * or a reboot of the MC.) + * or a reboot of the MC.) Not used for V3 licensing */ #define MC_CMD_GET_LICENSED_APP_STATE 0xf5 @@ -9261,8 +10237,68 @@ /***********************************/ +/* MC_CMD_GET_LICENSED_V3_APP_STATE + * Query the state of an individual licensed application. (Note that the actual + * state may be invalidated by the MC_CMD_LICENSING_V3 OP_UPDATE_LICENSE + * operation or a reboot of the MC.) Used for V3 licensing (Medford) + */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE 0xd2 + +#define MC_CMD_0xd2_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_GET_LICENSED_V3_APP_STATE_IN msgrequest */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_IN_LEN 8 +/* application ID to query (LICENSED_V3_APPS_xxx) expressed as a single bit + * mask + */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_IN_APP_ID_OFST 0 +#define MC_CMD_GET_LICENSED_V3_APP_STATE_IN_APP_ID_LEN 8 +#define MC_CMD_GET_LICENSED_V3_APP_STATE_IN_APP_ID_LO_OFST 0 +#define MC_CMD_GET_LICENSED_V3_APP_STATE_IN_APP_ID_HI_OFST 4 + +/* MC_CMD_GET_LICENSED_V3_APP_STATE_OUT msgresponse */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_OUT_LEN 4 +/* state of this application */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_OUT_STATE_OFST 0 +/* enum: no (or invalid) license is present for the application */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_OUT_NOT_LICENSED 0x0 +/* enum: a valid license is present for the application */ +#define MC_CMD_GET_LICENSED_V3_APP_STATE_OUT_LICENSED 0x1 + + +/***********************************/ +/* MC_CMD_GET_LICENSED_V3_FEATURE_STATES + * Query the state of an one or more licensed features. (Note that the actual + * state may be invalidated by the MC_CMD_LICENSING_V3 OP_UPDATE_LICENSE + * operation or a reboot of the MC.) Used for V3 licensing (Medford) + */ +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES 0xd3 + +#define MC_CMD_0xd3_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN msgrequest */ +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN_LEN 8 +/* features to query (LICENSED_V3_FEATURES_xxx) expressed as a mask with one or + * more bits set + */ +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN_FEATURES_OFST 0 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN_FEATURES_LEN 8 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN_FEATURES_LO_OFST 0 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_IN_FEATURES_HI_OFST 4 + +/* MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT msgresponse */ +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT_LEN 8 +/* states of these features - bit set for licensed, clear for not licensed */ +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT_STATES_OFST 0 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT_STATES_LEN 8 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT_STATES_LO_OFST 0 +#define MC_CMD_GET_LICENSED_V3_FEATURE_STATES_OUT_STATES_HI_OFST 4 + + +/***********************************/ /* MC_CMD_LICENSED_APP_OP - * Perform an action for an individual licensed application. + * Perform an action for an individual licensed application - not used for V3 + * licensing. */ #define MC_CMD_LICENSED_APP_OP 0xf6 @@ -9328,6 +10364,67 @@ /***********************************/ +/* MC_CMD_LICENSED_V3_VALIDATE_APP + * Perform validation for an individual licensed application - V3 licensing + * (Medford) + */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP 0xd4 + +#define MC_CMD_0xd4_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_LICENSED_V3_VALIDATE_APP_IN msgrequest */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_LEN 72 +/* application ID expressed as a single bit mask */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_APP_ID_OFST 0 +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_APP_ID_LEN 8 +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_APP_ID_LO_OFST 0 +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_APP_ID_HI_OFST 4 +/* challenge for validation */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_CHALLENGE_OFST 8 +#define MC_CMD_LICENSED_V3_VALIDATE_APP_IN_CHALLENGE_LEN 64 + +/* MC_CMD_LICENSED_V3_VALIDATE_APP_OUT msgresponse */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_LEN 72 +/* application expiry time */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_EXPIRY_TIME_OFST 0 +/* application expiry units */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_EXPIRY_UNITS_OFST 4 +/* enum: expiry units are accounting units */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_EXPIRY_UNIT_ACC 0x0 +/* enum: expiry units are calendar days */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_EXPIRY_UNIT_DAYS 0x1 +/* validation response to challenge */ +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_RESPONSE_OFST 8 +#define MC_CMD_LICENSED_V3_VALIDATE_APP_OUT_RESPONSE_LEN 64 + + +/***********************************/ +/* MC_CMD_LICENSED_V3_MASK_FEATURES + * Mask features - V3 licensing (Medford) + */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES 0xd5 + +#define MC_CMD_0xd5_PRIVILEGE_CTG SRIOV_CTG_GENERAL + +/* MC_CMD_LICENSED_V3_MASK_FEATURES_IN msgrequest */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_LEN 12 +/* mask to be applied to features to be changed */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_MASK_OFST 0 +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_MASK_LEN 8 +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_MASK_LO_OFST 0 +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_MASK_HI_OFST 4 +/* whether to turn on or turn off the masked features */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_FLAG_OFST 8 +/* enum: turn the features off */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_OFF 0x0 +/* enum: turn the features back on */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_IN_ON 0x1 + +/* MC_CMD_LICENSED_V3_MASK_FEATURES_OUT msgresponse */ +#define MC_CMD_LICENSED_V3_MASK_FEATURES_OUT_LEN 0 + + +/***********************************/ /* MC_CMD_SET_PORT_SNIFF_CONFIG * Configure RX port sniffing for the physical port associated with the calling * function. Only a privileged function may change the port sniffing @@ -9696,12 +10793,27 @@ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_ONLOAD 0x4 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_PTP 0x8 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_INSECURE_FILTERS 0x10 /* enum */ -#define MC_CMD_PRIVILEGE_MASK_IN_GRP_MAC_SPOOFING 0x20 /* enum */ +/* enum: Deprecated. Equivalent to MAC_SPOOFING_TX combined with CHANGE_MAC. */ +#define MC_CMD_PRIVILEGE_MASK_IN_GRP_MAC_SPOOFING 0x20 #define MC_CMD_PRIVILEGE_MASK_IN_GRP_UNICAST 0x40 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_MULTICAST 0x80 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_BROADCAST 0x100 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_ALL_MULTICAST 0x200 /* enum */ #define MC_CMD_PRIVILEGE_MASK_IN_GRP_PROMISCUOUS 0x400 /* enum */ +/* enum: Allows to set the TX packets' source MAC address to any arbitrary MAC + * adress. + */ +#define MC_CMD_PRIVILEGE_MASK_IN_GRP_MAC_SPOOFING_TX 0x800 +/* enum: Privilege that allows a Function to change the MAC address configured + * in its associated vAdapter/vPort. + */ +#define MC_CMD_PRIVILEGE_MASK_IN_GRP_CHANGE_MAC 0x1000 +/* enum: Privilege that allows a Function to install filters that specify VLANs + * that are not in the permit list for the associated vPort. This privilege is + * primarily to support ESX where vPorts are created that restrict traffic to + * only a set of permitted VLANs. See the vPort flag FLAG_VLAN_RESTRICT. + */ +#define MC_CMD_PRIVILEGE_MASK_IN_GRP_UNRESTRICTED_VLAN 0x2000 /* enum: Set this bit to indicate that a new privilege mask is to be set, * otherwise the command will only read the existing mask. */ @@ -9951,7 +11063,7 @@ /* Sector type */ #define MC_CMD_XPM_WRITE_SECTOR_IN_TYPE_OFST 4 /* Enum values, see field(s): */ -/* MC_CMD_XPM_READ_SECTOR_OUT/TYPE */ +/* MC_CMD_XPM_READ_SECTOR/MC_CMD_XPM_READ_SECTOR_OUT/TYPE */ /* Sector size */ #define MC_CMD_XPM_WRITE_SECTOR_IN_SIZE_OFST 8 /* Sector data */ @@ -10067,4 +11179,123 @@ #define MC_CMD_XPM_WRITE_TEST_OUT_LEN 0 +/***********************************/ +/* MC_CMD_EXEC_SIGNED + * Check the CMAC of the contents of IMEM and DMEM against the value supplied + * and if correct begin execution from the start of IMEM. The caller supplies a + * key ID, the length of IMEM and DMEM to validate and the expected CMAC. CMAC + * computation runs from the start of IMEM, and from the start of DMEM + 16k, + * to match flash booting. The command will respond with EINVAL if the CMAC + * does match, otherwise it will respond with success before it jumps to IMEM. + */ +#define MC_CMD_EXEC_SIGNED 0x10c + +#define MC_CMD_0x10c_PRIVILEGE_CTG SRIOV_CTG_ADMIN + +/* MC_CMD_EXEC_SIGNED_IN msgrequest */ +#define MC_CMD_EXEC_SIGNED_IN_LEN 28 +/* the length of code to include in the CMAC */ +#define MC_CMD_EXEC_SIGNED_IN_CODELEN_OFST 0 +/* the length of date to include in the CMAC */ +#define MC_CMD_EXEC_SIGNED_IN_DATALEN_OFST 4 +/* the XPM sector containing the key to use */ +#define MC_CMD_EXEC_SIGNED_IN_KEYSECTOR_OFST 8 +/* the expected CMAC value */ +#define MC_CMD_EXEC_SIGNED_IN_CMAC_OFST 12 +#define MC_CMD_EXEC_SIGNED_IN_CMAC_LEN 16 + +/* MC_CMD_EXEC_SIGNED_OUT msgresponse */ +#define MC_CMD_EXEC_SIGNED_OUT_LEN 0 + + +/***********************************/ +/* MC_CMD_PREPARE_SIGNED + * Prepare to upload a signed image. This will scrub the specified length of + * the data region, which must be at least as large as the DATALEN supplied to + * MC_CMD_EXEC_SIGNED. + */ +#define MC_CMD_PREPARE_SIGNED 0x10d + +#define MC_CMD_0x10d_PRIVILEGE_CTG SRIOV_CTG_ADMIN + +/* MC_CMD_PREPARE_SIGNED_IN msgrequest */ +#define MC_CMD_PREPARE_SIGNED_IN_LEN 4 +/* the length of data area to clear */ +#define MC_CMD_PREPARE_SIGNED_IN_DATALEN_OFST 0 + +/* MC_CMD_PREPARE_SIGNED_OUT msgresponse */ +#define MC_CMD_PREPARE_SIGNED_OUT_LEN 0 + + +/***********************************/ +/* MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS + * Configure UDP ports for tunnel encapsulation hardware acceleration. The + * parser-dispatcher will attempt to parse traffic on these ports as tunnel + * encapsulation PDUs and filter them using the tunnel encapsulation filter + * chain rather than the standard filter chain. Note that this command can + * cause all functions to see a reset. (Available on Medford only.) + */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS 0x117 + +#define MC_CMD_0x117_PRIVILEGE_CTG SRIOV_CTG_ADMIN + +/* MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN msgrequest */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_LENMIN 4 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_LENMAX 68 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_LEN(num) (4+4*(num)) +/* Flags */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_FLAGS_OFST 0 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_FLAGS_LEN 2 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_UNLOADING_LBN 0 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_UNLOADING_WIDTH 1 +/* The number of entries in the ENTRIES array */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_NUM_ENTRIES_OFST 2 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_NUM_ENTRIES_LEN 2 +/* Entries defining the UDP port to protocol mapping, each laid out as a + * TUNNEL_ENCAP_UDP_PORT_ENTRY + */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_ENTRIES_OFST 4 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_ENTRIES_LEN 4 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_ENTRIES_MINNUM 0 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_IN_ENTRIES_MAXNUM 16 + +/* MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT msgresponse */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT_LEN 2 +/* Flags */ +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT_FLAGS_OFST 0 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT_FLAGS_LEN 2 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT_RESETTING_LBN 0 +#define MC_CMD_SET_TUNNEL_ENCAP_UDP_PORTS_OUT_RESETTING_WIDTH 1 + + +/***********************************/ +/* MC_CMD_RX_BALANCING + * Configure a port upconverter to distribute the packets on both RX engines. + * Packets are distributed based on a table with the destination vFIFO. The + * index of the table is a hash of source and destination of IPV4 and VLAN + * priority. + */ +#define MC_CMD_RX_BALANCING 0x118 + +#define MC_CMD_0x118_PRIVILEGE_CTG SRIOV_CTG_ADMIN + +/* MC_CMD_RX_BALANCING_IN msgrequest */ +#define MC_CMD_RX_BALANCING_IN_LEN 4 +/* The RX port whose upconverter table will be modified */ +#define MC_CMD_RX_BALANCING_IN_PORT_OFST 0 +#define MC_CMD_RX_BALANCING_IN_PORT_LEN 1 +/* The VLAN priority associated to the table index and vFIFO */ +#define MC_CMD_RX_BALANCING_IN_PRIORITY_OFST 1 +#define MC_CMD_RX_BALANCING_IN_PRIORITY_LEN 1 +/* The resulting bit of SRC^DST for indexing the table */ +#define MC_CMD_RX_BALANCING_IN_SRC_DST_OFST 2 +#define MC_CMD_RX_BALANCING_IN_SRC_DST_LEN 1 +/* The RX engine to which the vFIFO in the table entry will point to */ +#define MC_CMD_RX_BALANCING_IN_ENG_OFST 3 +#define MC_CMD_RX_BALANCING_IN_ENG_LEN 1 + +/* MC_CMD_RX_BALANCING_OUT msgresponse */ +#define MC_CMD_RX_BALANCING_OUT_LEN 0 + + #endif /* MCDI_PCOL_H */ -- cgit v0.10.2 From 38d27f389cf5f1973ca97ddf0fc0b4c6886001ec Mon Sep 17 00:00:00 2001 From: Andrew Rybchenko Date: Wed, 15 Jun 2016 17:52:08 +0100 Subject: sfc: Fix VLAN filtering feature if vPort has VLAN_RESTRICT flag If vPort has VLAN_RESTRICT flag, VLAN tagged traffic will not be delivered without corresponding Rx filters which may be proxied to and moderated by hypervisor. Signed-off-by: Edward Cree Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 353ceef..f658fee 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -878,6 +878,45 @@ static int efx_ef10_probe_pf(struct efx_nic *efx) return efx_ef10_probe(efx); } +int efx_ef10_vadaptor_query(struct efx_nic *efx, unsigned int port_id, + u32 *port_flags, u32 *vadaptor_flags, + unsigned int *vlan_tags) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + MCDI_DECLARE_BUF(inbuf, MC_CMD_VADAPTOR_QUERY_IN_LEN); + MCDI_DECLARE_BUF(outbuf, MC_CMD_VADAPTOR_QUERY_OUT_LEN); + size_t outlen; + int rc; + + if (nic_data->datapath_caps & + (1 << MC_CMD_GET_CAPABILITIES_OUT_VADAPTOR_QUERY_LBN)) { + MCDI_SET_DWORD(inbuf, VADAPTOR_QUERY_IN_UPSTREAM_PORT_ID, + port_id); + + rc = efx_mcdi_rpc(efx, MC_CMD_VADAPTOR_QUERY, inbuf, sizeof(inbuf), + outbuf, sizeof(outbuf), &outlen); + if (rc) + return rc; + + if (outlen < sizeof(outbuf)) { + rc = -EIO; + return rc; + } + } + + if (port_flags) + *port_flags = MCDI_DWORD(outbuf, VADAPTOR_QUERY_OUT_PORT_FLAGS); + if (vadaptor_flags) + *vadaptor_flags = + MCDI_DWORD(outbuf, VADAPTOR_QUERY_OUT_VADAPTOR_FLAGS); + if (vlan_tags) + *vlan_tags = + MCDI_DWORD(outbuf, + VADAPTOR_QUERY_OUT_NUM_AVAILABLE_VLAN_TAGS); + + return 0; +} + int efx_ef10_vadaptor_alloc(struct efx_nic *efx, unsigned int port_id) { MCDI_DECLARE_BUF(inbuf, MC_CMD_VADAPTOR_ALLOC_IN_LEN); diff --git a/drivers/net/ethernet/sfc/ef10_sriov.c b/drivers/net/ethernet/sfc/ef10_sriov.c index a76610a..a949b9d 100644 --- a/drivers/net/ethernet/sfc/ef10_sriov.c +++ b/drivers/net/ethernet/sfc/ef10_sriov.c @@ -232,6 +232,35 @@ fail: return rc; } +static int efx_ef10_vadaptor_alloc_set_features(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + u32 port_flags; + int rc; + + rc = efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + if (rc) + goto fail_vadaptor_alloc; + + rc = efx_ef10_vadaptor_query(efx, nic_data->vport_id, + &port_flags, NULL, NULL); + if (rc) + goto fail_vadaptor_query; + + if (port_flags & + (1 << MC_CMD_VPORT_ALLOC_IN_FLAG_VLAN_RESTRICT_LBN)) + efx->fixed_features |= NETIF_F_HW_VLAN_CTAG_FILTER; + else + efx->fixed_features &= ~NETIF_F_HW_VLAN_CTAG_FILTER; + + return 0; + +fail_vadaptor_query: + efx_ef10_vadaptor_free(efx, EVB_PORT_ID_ASSIGNED); +fail_vadaptor_alloc: + return rc; +} + /* On top of the default firmware vswitch setup, create a VEB vswitch and * expansion vport for use by this function. */ @@ -243,7 +272,7 @@ int efx_ef10_vswitching_probe_pf(struct efx_nic *efx) if (pci_sriov_get_totalvfs(efx->pci_dev) <= 0) { /* vswitch not needed as we have no VFs */ - efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + efx_ef10_vadaptor_alloc_set_features(efx); return 0; } @@ -263,7 +292,7 @@ int efx_ef10_vswitching_probe_pf(struct efx_nic *efx) goto fail3; ether_addr_copy(nic_data->vport_mac, net_dev->dev_addr); - rc = efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + rc = efx_ef10_vadaptor_alloc_set_features(efx); if (rc) goto fail4; @@ -282,9 +311,7 @@ fail1: int efx_ef10_vswitching_probe_vf(struct efx_nic *efx) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; - - return efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + return efx_ef10_vadaptor_alloc_set_features(efx); } int efx_ef10_vswitching_restore_pf(struct efx_nic *efx) diff --git a/drivers/net/ethernet/sfc/ef10_sriov.h b/drivers/net/ethernet/sfc/ef10_sriov.h index 6d25b92..9ceb7ef 100644 --- a/drivers/net/ethernet/sfc/ef10_sriov.h +++ b/drivers/net/ethernet/sfc/ef10_sriov.h @@ -70,6 +70,9 @@ int efx_ef10_vport_add_mac(struct efx_nic *efx, int efx_ef10_vport_del_mac(struct efx_nic *efx, unsigned int port_id, u8 *mac); int efx_ef10_vadaptor_alloc(struct efx_nic *efx, unsigned int port_id); +int efx_ef10_vadaptor_query(struct efx_nic *efx, unsigned int port_id, + u32 *port_flags, u32 *vadaptor_flags, + unsigned int *vlan_tags); int efx_ef10_vadaptor_free(struct efx_nic *efx, unsigned int port_id); #endif /* EF10_SRIOV_H */ -- cgit v0.10.2 From 61d1b6a42fec61c5065f54cc62cef02b483c69fb Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 15 Jun 2016 22:47:12 +0200 Subject: bpf, maps: add release callback Add a release callback for maps that is invoked when the last reference to its struct file is gone and the struct file about to be released by vfs. The handler will be used by fd array maps. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 1bcae82..29b5a1a 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -19,7 +19,8 @@ struct bpf_map; struct bpf_map_ops { /* funcs callable from userspace (via syscall) */ struct bpf_map *(*map_alloc)(union bpf_attr *attr); - void (*map_free)(struct bpf_map *); + void (*map_release)(struct bpf_map *map, struct file *map_file); + void (*map_free)(struct bpf_map *map); int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key); /* funcs callable from userspace and from eBPF programs */ diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 46ecce4..fc3adcd 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -124,7 +124,12 @@ void bpf_map_put_with_uref(struct bpf_map *map) static int bpf_map_release(struct inode *inode, struct file *filp) { - bpf_map_put_with_uref(filp->private_data); + struct bpf_map *map = filp->private_data; + + if (map->ops->map_release) + map->ops->map_release(map, filp); + + bpf_map_put_with_uref(map); return 0; } -- cgit v0.10.2 From d056a788765e67773124f520159185bc89f5d1ad Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 15 Jun 2016 22:47:13 +0200 Subject: bpf, maps: extend map_fd_get_ptr arguments This patch extends map_fd_get_ptr() callback that is used by fd array maps, so that struct file pointer from the related map can be passed in. It's safe to remove map_update_elem() callback for the two maps since this is only allowed from syscall side, but not from eBPF programs for these two map types. Like in per-cpu map case, bpf_fd_array_map_update_elem() needs to be called directly here due to the extra argument. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 29b5a1a..d7b43e7 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -29,8 +29,9 @@ struct bpf_map_ops { int (*map_delete_elem)(struct bpf_map *map, void *key); /* funcs called by prog_array and perf_event_array map */ - void *(*map_fd_get_ptr) (struct bpf_map *map, int fd); - void (*map_fd_put_ptr) (void *ptr); + void *(*map_fd_get_ptr)(struct bpf_map *map, struct file *map_file, + int fd); + void (*map_fd_put_ptr)(void *ptr); }; struct bpf_map { @@ -169,7 +170,7 @@ struct bpf_array { u64 bpf_tail_call(u64 ctx, u64 r2, u64 index, u64 r4, u64 r5); u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); -void bpf_fd_array_map_clear(struct bpf_map *map); + bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp); const struct bpf_func_proto *bpf_get_trace_printk_proto(void); @@ -207,8 +208,13 @@ int bpf_percpu_hash_update(struct bpf_map *map, void *key, void *value, u64 flags); int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value, u64 flags); + int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value); +int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file, + void *key, void *value, u64 map_flags); +void bpf_fd_array_map_clear(struct bpf_map *map); + /* memcpy that is used with 8-byte aligned pointers, power-of-8 size and * forced to use 'long' read/writes to try to atomically copy long counters. * Best-effort only. No barriers here, since it _will_ race with concurrent diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 76d5a79..bfedcbd 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -328,8 +328,8 @@ static void *fd_array_map_lookup_elem(struct bpf_map *map, void *key) } /* only called from syscall */ -static int fd_array_map_update_elem(struct bpf_map *map, void *key, - void *value, u64 map_flags) +int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file, + void *key, void *value, u64 map_flags) { struct bpf_array *array = container_of(map, struct bpf_array, map); void *new_ptr, *old_ptr; @@ -342,7 +342,7 @@ static int fd_array_map_update_elem(struct bpf_map *map, void *key, return -E2BIG; ufd = *(u32 *)value; - new_ptr = map->ops->map_fd_get_ptr(map, ufd); + new_ptr = map->ops->map_fd_get_ptr(map, map_file, ufd); if (IS_ERR(new_ptr)) return PTR_ERR(new_ptr); @@ -371,10 +371,12 @@ static int fd_array_map_delete_elem(struct bpf_map *map, void *key) } } -static void *prog_fd_array_get_ptr(struct bpf_map *map, int fd) +static void *prog_fd_array_get_ptr(struct bpf_map *map, + struct file *map_file, int fd) { struct bpf_array *array = container_of(map, struct bpf_array, map); struct bpf_prog *prog = bpf_prog_get(fd); + if (IS_ERR(prog)) return prog; @@ -382,6 +384,7 @@ static void *prog_fd_array_get_ptr(struct bpf_map *map, int fd) bpf_prog_put(prog); return ERR_PTR(-EINVAL); } + return prog; } @@ -407,7 +410,6 @@ static const struct bpf_map_ops prog_array_ops = { .map_free = fd_array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = fd_array_map_lookup_elem, - .map_update_elem = fd_array_map_update_elem, .map_delete_elem = fd_array_map_delete_elem, .map_fd_get_ptr = prog_fd_array_get_ptr, .map_fd_put_ptr = prog_fd_array_put_ptr, @@ -431,7 +433,8 @@ static void perf_event_array_map_free(struct bpf_map *map) fd_array_map_free(map); } -static void *perf_event_fd_array_get_ptr(struct bpf_map *map, int fd) +static void *perf_event_fd_array_get_ptr(struct bpf_map *map, + struct file *map_file, int fd) { struct perf_event *event; const struct perf_event_attr *attr; @@ -474,7 +477,6 @@ static const struct bpf_map_ops perf_event_array_ops = { .map_free = perf_event_array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = fd_array_map_lookup_elem, - .map_update_elem = fd_array_map_update_elem, .map_delete_elem = fd_array_map_delete_elem, .map_fd_get_ptr = perf_event_fd_array_get_ptr, .map_fd_put_ptr = perf_event_fd_array_put_ptr, diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index fc3adcd..c23a4e93 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -392,6 +392,12 @@ static int map_update_elem(union bpf_attr *attr) err = bpf_percpu_hash_update(map, key, value, attr->flags); } else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) { err = bpf_percpu_array_update(map, key, value, attr->flags); + } else if (map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || + map->map_type == BPF_MAP_TYPE_PROG_ARRAY) { + rcu_read_lock(); + err = bpf_fd_array_map_update_elem(map, f.file, key, value, + attr->flags); + rcu_read_unlock(); } else { rcu_read_lock(); err = map->ops->map_update_elem(map, key, value, attr->flags); -- cgit v0.10.2 From 3b1efb196eee45b2f0c4994e0c43edb5e367f620 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 15 Jun 2016 22:47:14 +0200 Subject: bpf, maps: flush own entries on perf map release The behavior of perf event arrays are quite different from all others as they are tightly coupled to perf event fds, f.e. shown recently by commit e03e7ee34fdd ("perf/bpf: Convert perf_event_array to use struct file") to make refcounting on perf event more robust. A remaining issue that the current code still has is that since additions to the perf event array take a reference on the struct file via perf_event_get() and are only released via fput() (that cleans up the perf event eventually via perf_event_release_kernel()) when the element is either manually removed from the map from user space or automatically when the last reference on the perf event map is dropped. However, this leads us to dangling struct file's when the map gets pinned after the application owning the perf event descriptor exits, and since the struct file reference will in such case only be manually dropped or via pinned file removal, it leads to the perf event living longer than necessary, consuming needlessly resources for that time. Relations between perf event fds and bpf perf event map fds can be rather complex. F.e. maps can act as demuxers among different perf event fds that can possibly be owned by different threads and based on the index selection from the program, events get dispatched to one of the per-cpu fd endpoints. One perf event fd (or, rather a per-cpu set of them) can also live in multiple perf event maps at the same time, listening for events. Also, another requirement is that perf event fds can get closed from application side after they have been attached to the perf event map, so that on exit perf event map will take care of dropping their references eventually. Likewise, when such maps are pinned, the intended behavior is that a user application does bpf_obj_get(), puts its fds in there and on exit when fd is released, they are dropped from the map again, so the map acts rather as connector endpoint. This also makes perf event maps inherently different from program arrays as described in more detail in commit c9da161c6517 ("bpf: fix clearing on persistent program array maps"). To tackle this, map entries are marked by the map struct file that added the element to the map. And when the last reference to that map struct file is released from user space, then the tracked entries are purged from the map. This is okay, because new map struct files instances resp. frontends to the anon inode are provided via bpf_map_new_fd() that is called when we invoke bpf_obj_get_user() for retrieving a pinned map, but also when an initial instance is created via map_create(). The rest is resolved by the vfs layer automatically for us by keeping reference count on the map's struct file. Any concurrent updates on the map slot are fine as well, it just means that perf_event_fd_array_release() needs to delete less of its own entires. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/bpf.h b/include/linux/bpf.h index d7b43e7..9adfef6 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -13,6 +13,7 @@ #include #include +struct perf_event; struct bpf_map; /* map is generic key/value storage optionally accesible by eBPF programs */ @@ -166,8 +167,16 @@ struct bpf_array { void __percpu *pptrs[0] __aligned(8); }; }; + #define MAX_TAIL_CALL_CNT 32 +struct bpf_event_entry { + struct perf_event *event; + struct file *perf_file; + struct file *map_file; + struct rcu_head rcu; +}; + u64 bpf_tail_call(u64 ctx, u64 r2, u64 index, u64 r4, u64 r5); u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index bfedcbd..5af3073 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -427,59 +427,105 @@ static int __init register_prog_array_map(void) } late_initcall(register_prog_array_map); -static void perf_event_array_map_free(struct bpf_map *map) +static struct bpf_event_entry *bpf_event_entry_gen(struct file *perf_file, + struct file *map_file) { - bpf_fd_array_map_clear(map); - fd_array_map_free(map); + struct bpf_event_entry *ee; + + ee = kzalloc(sizeof(*ee), GFP_KERNEL); + if (ee) { + ee->event = perf_file->private_data; + ee->perf_file = perf_file; + ee->map_file = map_file; + } + + return ee; +} + +static void __bpf_event_entry_free(struct rcu_head *rcu) +{ + struct bpf_event_entry *ee; + + ee = container_of(rcu, struct bpf_event_entry, rcu); + fput(ee->perf_file); + kfree(ee); +} + +static void bpf_event_entry_free_rcu(struct bpf_event_entry *ee) +{ + call_rcu(&ee->rcu, __bpf_event_entry_free); } static void *perf_event_fd_array_get_ptr(struct bpf_map *map, struct file *map_file, int fd) { - struct perf_event *event; const struct perf_event_attr *attr; - struct file *file; + struct bpf_event_entry *ee; + struct perf_event *event; + struct file *perf_file; - file = perf_event_get(fd); - if (IS_ERR(file)) - return file; + perf_file = perf_event_get(fd); + if (IS_ERR(perf_file)) + return perf_file; - event = file->private_data; + event = perf_file->private_data; + ee = ERR_PTR(-EINVAL); attr = perf_event_attrs(event); - if (IS_ERR(attr)) - goto err; - - if (attr->inherit) - goto err; - - if (attr->type == PERF_TYPE_RAW) - return file; - - if (attr->type == PERF_TYPE_HARDWARE) - return file; + if (IS_ERR(attr) || attr->inherit) + goto err_out; + + switch (attr->type) { + case PERF_TYPE_SOFTWARE: + if (attr->config != PERF_COUNT_SW_BPF_OUTPUT) + goto err_out; + /* fall-through */ + case PERF_TYPE_RAW: + case PERF_TYPE_HARDWARE: + ee = bpf_event_entry_gen(perf_file, map_file); + if (ee) + return ee; + ee = ERR_PTR(-ENOMEM); + /* fall-through */ + default: + break; + } - if (attr->type == PERF_TYPE_SOFTWARE && - attr->config == PERF_COUNT_SW_BPF_OUTPUT) - return file; -err: - fput(file); - return ERR_PTR(-EINVAL); +err_out: + fput(perf_file); + return ee; } static void perf_event_fd_array_put_ptr(void *ptr) { - fput((struct file *)ptr); + bpf_event_entry_free_rcu(ptr); +} + +static void perf_event_fd_array_release(struct bpf_map *map, + struct file *map_file) +{ + struct bpf_array *array = container_of(map, struct bpf_array, map); + struct bpf_event_entry *ee; + int i; + + rcu_read_lock(); + for (i = 0; i < array->map.max_entries; i++) { + ee = READ_ONCE(array->ptrs[i]); + if (ee && ee->map_file == map_file) + fd_array_map_delete_elem(map, &i); + } + rcu_read_unlock(); } static const struct bpf_map_ops perf_event_array_ops = { .map_alloc = fd_array_map_alloc, - .map_free = perf_event_array_map_free, + .map_free = fd_array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = fd_array_map_lookup_elem, .map_delete_elem = fd_array_map_delete_elem, .map_fd_get_ptr = perf_event_fd_array_get_ptr, .map_fd_put_ptr = perf_event_fd_array_put_ptr, + .map_release = perf_event_fd_array_release, }; static struct bpf_map_type_list perf_event_array_type __read_mostly = { diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 720b7bb..037ea6e 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -192,18 +192,17 @@ static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5) { struct bpf_map *map = (struct bpf_map *) (unsigned long) r1; struct bpf_array *array = container_of(map, struct bpf_array, map); + struct bpf_event_entry *ee; struct perf_event *event; - struct file *file; if (unlikely(index >= array->map.max_entries)) return -E2BIG; - file = READ_ONCE(array->ptrs[index]); - if (unlikely(!file)) + ee = READ_ONCE(array->ptrs[index]); + if (unlikely(!ee)) return -ENOENT; - event = file->private_data; - + event = ee->event; /* make sure event is local and doesn't have pmu::count */ if (event->oncpu != smp_processor_id() || event->pmu->count) @@ -233,8 +232,8 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) u64 index = flags & BPF_F_INDEX_MASK; void *data = (void *) (long) r4; struct perf_sample_data sample_data; + struct bpf_event_entry *ee; struct perf_event *event; - struct file *file; struct perf_raw_record raw = { .size = size, .data = data, @@ -247,12 +246,11 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) if (unlikely(index >= array->map.max_entries)) return -E2BIG; - file = READ_ONCE(array->ptrs[index]); - if (unlikely(!file)) + ee = READ_ONCE(array->ptrs[index]); + if (unlikely(!ee)) return -ENOENT; - event = file->private_data; - + event = ee->event; if (unlikely(event->attr.type != PERF_TYPE_SOFTWARE || event->attr.config != PERF_COUNT_SW_BPF_OUTPUT)) return -EINVAL; -- cgit v0.10.2 From 0cd33c204b98c6bff25817d69e4706b9fb610359 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Fri, 27 May 2016 10:54:28 +0200 Subject: brcmfmac: print errors if creating interface fails This is helpful for debugging. Without this all I was getting from "iw" command on failed creating of P2P interface was: > command failed: Too many open files in system (-23) Signed-off-by: Rafal Milecki [arend@broadcom.com: reduce error prints upon iface creation] Signed-off-by: Arend van Spriel Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index ba65a93..30ac4365 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -684,20 +684,24 @@ static struct wireless_dev *brcmf_cfg80211_add_iface(struct wiphy *wiphy, return ERR_PTR(-EOPNOTSUPP); case NL80211_IFTYPE_AP: wdev = brcmf_ap_add_vif(wiphy, name, flags, params); - if (!IS_ERR(wdev)) - brcmf_cfg80211_update_proto_addr_mode(wdev); - return wdev; + break; case NL80211_IFTYPE_P2P_CLIENT: case NL80211_IFTYPE_P2P_GO: case NL80211_IFTYPE_P2P_DEVICE: wdev = brcmf_p2p_add_vif(wiphy, name, name_assign_type, type, flags, params); - if (!IS_ERR(wdev)) - brcmf_cfg80211_update_proto_addr_mode(wdev); - return wdev; + break; case NL80211_IFTYPE_UNSPECIFIED: default: return ERR_PTR(-EINVAL); } + + if (IS_ERR(wdev)) + brcmf_err("add iface %s type %d failed: err=%d\n", + name, type, (int)PTR_ERR(wdev)); + else + brcmf_cfg80211_update_proto_addr_mode(wdev); + + return wdev; } static void brcmf_scan_config_mpc(struct brcmf_if *ifp, int mpc) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c index 1652a48..b7d54e9 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c @@ -2030,8 +2030,6 @@ static int brcmf_p2p_request_p2p_if(struct brcmf_p2p_info *p2p, err = brcmf_fil_iovar_data_set(ifp, "p2p_ifadd", &if_request, sizeof(if_request)); - if (err) - return err; return err; } -- cgit v0.10.2 From 6f49208fec850639f56ed850ab79dbe7f6979221 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 27 May 2016 10:18:15 -0400 Subject: mwifiex: only call mwifiex_sdio_probe_of() if dev has an OF node SDIO is an auto enumerable bus so the SDIO devices are matched using the sdio_device_id table and not using compatible strings from a OF id table. However, commit ce4f6f0c353b ("mwifiex: add platform specific wakeup interrupt support") allowed to match nodes defined as child of the SDIO host controller in the probe function using a compatible string to setup platform specific parameters in the DT. The problem is that the OF parse function is always called regardless if the SDIO dev has an OF node associated or not, and prints an error if it is not found. So, on a platform that doesn't have a node for a SDIO dev, the following misleading error message will be printed: [ 12.480042] mwifiex_sdio mmc2:0001:1: sdio platform data not available Signed-off-by: Javier Martinez Canillas Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c index 674465e..3b30f27 100644 --- a/drivers/net/wireless/marvell/mwifiex/sdio.c +++ b/drivers/net/wireless/marvell/mwifiex/sdio.c @@ -102,8 +102,7 @@ static int mwifiex_sdio_probe_of(struct device *dev, struct sdio_mmc_card *card) struct mwifiex_plt_wake_cfg *cfg; int ret; - if (!dev->of_node || - !of_match_node(mwifiex_sdio_of_match_table, dev->of_node)) { + if (!of_match_node(mwifiex_sdio_of_match_table, dev->of_node)) { dev_err(dev, "sdio platform data not available\n"); return -1; } @@ -189,7 +188,8 @@ mwifiex_sdio_probe(struct sdio_func *func, const struct sdio_device_id *id) } /* device tree node parsing and platform specific configuration*/ - mwifiex_sdio_probe_of(&func->dev, card); + if (func->dev.of_node) + mwifiex_sdio_probe_of(&func->dev, card); if (mwifiex_add_card(card, &add_remove_card_sem, &sdio_ops, MWIFIEX_SDIO)) { -- cgit v0.10.2 From cc524d1706b775466bf16a0a1130105ae5c70f43 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 27 May 2016 10:18:16 -0400 Subject: mwifiex: propagate sdio_enable_func() errno code in mwifiex_sdio_probe() If the sdio_enable_func() function fails on .probe, the -EIO errno code is always returned but that could make more difficult to debug and find the cause of why the function actually failed. Since the driver/device core prints the value returned by .probe in its error message propagate what was returned by sdio_enable_func() at fail. Signed-off-by: Javier Martinez Canillas Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c index 3b30f27..138dc1f 100644 --- a/drivers/net/wireless/marvell/mwifiex/sdio.c +++ b/drivers/net/wireless/marvell/mwifiex/sdio.c @@ -184,7 +184,7 @@ mwifiex_sdio_probe(struct sdio_func *func, const struct sdio_device_id *id) if (ret) { pr_err("%s: failed to enable function\n", __func__); kfree(card); - return -EIO; + return ret; } /* device tree node parsing and platform specific configuration*/ -- cgit v0.10.2 From 032e0f546c7e36217ebcae33f82d390c272d00ea Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 27 May 2016 10:18:17 -0400 Subject: mwifiex: propagate mwifiex_add_card() errno code in mwifiex_sdio_probe() There's only a check if mwifiex_add_card() returned a nonzero value, but the actual error code is neither stored nor propagated to the caller. So instead of always returning -1 (which is -EPERM and not a suitable errno code in this case), propagate the value returned by mwifiex_add_card(). Patch also removes the assignment of sdio_disable_func() returned value since it was overwritten anyways and what matters is to know the error value returned by the first function that failed. Signed-off-by: Javier Martinez Canillas Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c index 138dc1f..b248d10 100644 --- a/drivers/net/wireless/marvell/mwifiex/sdio.c +++ b/drivers/net/wireless/marvell/mwifiex/sdio.c @@ -191,14 +191,14 @@ mwifiex_sdio_probe(struct sdio_func *func, const struct sdio_device_id *id) if (func->dev.of_node) mwifiex_sdio_probe_of(&func->dev, card); - if (mwifiex_add_card(card, &add_remove_card_sem, &sdio_ops, - MWIFIEX_SDIO)) { + ret = mwifiex_add_card(card, &add_remove_card_sem, &sdio_ops, + MWIFIEX_SDIO); + if (ret) { pr_err("%s: add card failed\n", __func__); kfree(card); sdio_claim_host(func); - ret = sdio_disable_func(func); + sdio_disable_func(func); sdio_release_host(func); - ret = -1; } return ret; -- cgit v0.10.2 From a82f65aae143f298e7b795ffd8f1cbbe76653a90 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 27 May 2016 10:18:18 -0400 Subject: mwifiex: consolidate mwifiex_sdio_probe() error paths Instead of duplicating part of the cleanups needed in case of an error in .probe callback, have a single error path and use goto labels as is common practice in the kernel. This also has the nice side effect that the cleanup operations are made in the inverse order of their counterparts, which was not the case for the mwifiex_add_card() error path. Signed-off-by: Javier Martinez Canillas Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c index b248d10..a3fd0a1 100644 --- a/drivers/net/wireless/marvell/mwifiex/sdio.c +++ b/drivers/net/wireless/marvell/mwifiex/sdio.c @@ -183,8 +183,7 @@ mwifiex_sdio_probe(struct sdio_func *func, const struct sdio_device_id *id) if (ret) { pr_err("%s: failed to enable function\n", __func__); - kfree(card); - return ret; + goto err_free; } /* device tree node parsing and platform specific configuration*/ @@ -195,12 +194,18 @@ mwifiex_sdio_probe(struct sdio_func *func, const struct sdio_device_id *id) MWIFIEX_SDIO); if (ret) { pr_err("%s: add card failed\n", __func__); - kfree(card); - sdio_claim_host(func); - sdio_disable_func(func); - sdio_release_host(func); + goto err_disable; } + return 0; + +err_disable: + sdio_claim_host(func); + sdio_disable_func(func); + sdio_release_host(func); +err_free: + kfree(card); + return ret; } -- cgit v0.10.2 From d3f04ece53a40f6d3c83821ce0cf82d3d16880c9 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 27 May 2016 10:18:19 -0400 Subject: mwifiex: use dev_err() instead of pr_err() in mwifiex_sdio_probe() It's better to have the device name prefixed in the error message. Signed-off-by: Javier Martinez Canillas Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c index a3fd0a1..e9d90766 100644 --- a/drivers/net/wireless/marvell/mwifiex/sdio.c +++ b/drivers/net/wireless/marvell/mwifiex/sdio.c @@ -182,7 +182,7 @@ mwifiex_sdio_probe(struct sdio_func *func, const struct sdio_device_id *id) sdio_release_host(func); if (ret) { - pr_err("%s: failed to enable function\n", __func__); + dev_err(&func->dev, "failed to enable function\n"); goto err_free; } @@ -193,7 +193,7 @@ mwifiex_sdio_probe(struct sdio_func *func, const struct sdio_device_id *id) ret = mwifiex_add_card(card, &add_remove_card_sem, &sdio_ops, MWIFIEX_SDIO); if (ret) { - pr_err("%s: add card failed\n", __func__); + dev_err(&func->dev, "add card failed\n"); goto err_disable; } -- cgit v0.10.2 From 213d9421c165973f6cc9d79e91c8be2de25d0ea0 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 27 May 2016 10:18:20 -0400 Subject: mwifiex: check if mwifiex_sdio_probe_of() fails and return error The function can fail so the returned value should be checked and the error propagated to the caller in case of a failure. Signed-off-by: Javier Martinez Canillas Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c index e9d90766..c38d14f 100644 --- a/drivers/net/wireless/marvell/mwifiex/sdio.c +++ b/drivers/net/wireless/marvell/mwifiex/sdio.c @@ -187,8 +187,13 @@ mwifiex_sdio_probe(struct sdio_func *func, const struct sdio_device_id *id) } /* device tree node parsing and platform specific configuration*/ - if (func->dev.of_node) - mwifiex_sdio_probe_of(&func->dev, card); + if (func->dev.of_node) { + ret = mwifiex_sdio_probe_of(&func->dev, card); + if (ret) { + dev_err(&func->dev, "SDIO dt node parse failed\n"); + goto err_disable; + } + } ret = mwifiex_add_card(card, &add_remove_card_sem, &sdio_ops, MWIFIEX_SDIO); -- cgit v0.10.2 From 806dd220340d404857c2f1b8f4bd9f9f1f052d80 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 27 May 2016 10:18:21 -0400 Subject: mwifiex: don't print an error if an optional DT property is missing The Documentation/devicetree/bindings/net/wireless/marvell-sd8xxx.txt DT binding document say that the "interrupts" property in the child node is optional. So the property being missed shouldn't be treated as an error. Signed-off-by: Javier Martinez Canillas Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c index c38d14f..ba7a493 100644 --- a/drivers/net/wireless/marvell/mwifiex/sdio.c +++ b/drivers/net/wireless/marvell/mwifiex/sdio.c @@ -114,7 +114,7 @@ static int mwifiex_sdio_probe_of(struct device *dev, struct sdio_mmc_card *card) if (cfg && card->plt_of_node) { cfg->irq_wifi = irq_of_parse_and_map(card->plt_of_node, 0); if (!cfg->irq_wifi) { - dev_err(dev, + dev_dbg(dev, "fail to parse irq_wifi from device tree\n"); } else { ret = devm_request_irq(dev, cfg->irq_wifi, -- cgit v0.10.2 From 5e94913f676af0dddeb6e0f3de241de5bd92f3f1 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 27 May 2016 10:18:22 -0400 Subject: mwifiex: use better message and error code when OF node doesn't match The Documentation/devicetree/bindings/net/wireless/marvell-sd8xxx.txt DT binding document lists the possible compatible strings that a SDIO child node can have, so the driver checks if the defined in the node matches. But the error message when that's not the case is misleading, so change for one that makes clear what the error really is. Also, returning a -1 as errno code is not correct since that's -EPERM. A -EINVAL seems to be a more appropriate one. Signed-off-by: Javier Martinez Canillas Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c index ba7a493..3a2267a 100644 --- a/drivers/net/wireless/marvell/mwifiex/sdio.c +++ b/drivers/net/wireless/marvell/mwifiex/sdio.c @@ -103,8 +103,8 @@ static int mwifiex_sdio_probe_of(struct device *dev, struct sdio_mmc_card *card) int ret; if (!of_match_node(mwifiex_sdio_of_match_table, dev->of_node)) { - dev_err(dev, "sdio platform data not available\n"); - return -1; + dev_err(dev, "required compatible string missing\n"); + return -EINVAL; } card->plt_of_node = dev->of_node; -- cgit v0.10.2 From 2683f7dd9aaddf2ab8d41fe597bfffda2d6f4ab6 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 27 May 2016 17:08:33 +0200 Subject: wl3501_cs: avoid bogus gcc-6 warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit gcc-6 on x86 started warning about wl3501_get_encode when building with -O2: drivers/net/wireless/wl3501_cs.c: In function ‘wl3501_get_encode’: drivers/net/wireless/wl3501_cs.c:1769:5: warning: ‘implemented’ may be used uninitialized in this function drivers/net/wireless/wl3501_cs.c:1686:19: warning: ‘threshold’ may be used uninitialized in this function drivers/net/wireless/wl3501_cs.c:1702:20: warning: ‘threshold’ may be used uninitialized in this function drivers/net/wireless/wl3501_cs.c:1719:23: warning: ‘txpow’ may be used uninitialized in this function drivers/net/wireless/wl3501_cs.c:1752:20: warning: ‘retry’ may be used uninitialized in this function drivers/net/wireless/wl3501_cs.c:1806:25: warning: ‘pwr_state’ may be used uninitialized in this function drivers/net/wireless/wl3501_cs.c:1383:24: warning: ‘value’ may be used uninitialized in this function I could not figure out what exactly confuses gcc here, but splitting the wl3501_get_mib_value function into two helps the compiler to figure out that the variables are not actually used uninitialized, and makes it slightly clearer to a human reader what the function actually does and which parts of it are under the spinlock. Signed-off-by: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/wl3501_cs.c b/drivers/net/wireless/wl3501_cs.c index 13fd734..82d94f8 100644 --- a/drivers/net/wireless/wl3501_cs.c +++ b/drivers/net/wireless/wl3501_cs.c @@ -378,8 +378,7 @@ static int wl3501_esbq_exec(struct wl3501_card *this, void *sig, int sig_size) return rc; } -static int wl3501_get_mib_value(struct wl3501_card *this, u8 index, - void *bf, int size) +static int wl3501_request_mib(struct wl3501_card *this, u8 index, void *bf) { struct wl3501_get_req sig = { .sig_id = WL3501_SIG_GET_REQ, @@ -395,20 +394,32 @@ static int wl3501_get_mib_value(struct wl3501_card *this, u8 index, wl3501_set_to_wla(this, ptr, &sig, sizeof(sig)); wl3501_esbq_req(this, &ptr); this->sig_get_confirm.mib_status = 255; - spin_unlock_irqrestore(&this->lock, flags); - rc = wait_event_interruptible(this->wait, - this->sig_get_confirm.mib_status != 255); - if (!rc) - memcpy(bf, this->sig_get_confirm.mib_value, - size); - goto out; + rc = 0; } } spin_unlock_irqrestore(&this->lock, flags); -out: + return rc; } +static int wl3501_get_mib_value(struct wl3501_card *this, u8 index, + void *bf, int size) +{ + int rc; + + rc = wl3501_request_mib(this, index, bf); + if (rc) + return rc; + + rc = wait_event_interruptible(this->wait, + this->sig_get_confirm.mib_status != 255); + if (rc) + return rc; + + memcpy(bf, this->sig_get_confirm.mib_value, size); + return 0; +} + static int wl3501_pwr_mgmt(struct wl3501_card *this, int suspend) { struct wl3501_pwr_mgmt_req sig = { -- cgit v0.10.2 From 8707e08dbc1f1f3a60630c94a41793390ad22c99 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Fri, 27 May 2016 21:07:19 +0200 Subject: brcmfmac: fix setting AP channel with new firmwares MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Firmware for new chipsets is based on a new major version of code internally maintained at Broadcom. E.g. brcmfmac4366b-pcie.bin (used for BCM4366B1) is based on 10.10.69.3309 while brcmfmac43602-pcie.ap.bin was based on 7.35.177.56. Currently setting AP 5 GHz channel doesn't work reliably with BCM4366B1. When setting e.g. 36 control channel with VHT80 (center channel 42) firmware may randomly pick one of: 1) 52 control channel with 58 as center one 2) 100 control channel with 106 as center one 3) 116 control channel with 122 as center one 4) 149 control channel with 155 as center one It seems new firmwares require setting AP mode (BRCMF_C_SET_AP) before specifying a channel. Changing an order of firmware calls fixes the problem. This requirement resulted in two separated "chanspec" calls, one in AP code path and one in P2P path. This fix was verified with BCM4366B1 and tested for regressions on BCM43602. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index 30ac4365..f533dc2 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -4442,7 +4442,7 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev, struct brcmf_join_params join_params; enum nl80211_iftype dev_role; struct brcmf_fil_bss_enable_le bss_enable; - u16 chanspec; + u16 chanspec = chandef_to_chanspec(&cfg->d11inf, &settings->chandef); bool mbss; int is_11d; @@ -4518,16 +4518,8 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev, brcmf_config_ap_mgmt_ie(ifp->vif, &settings->beacon); + /* Parameters shared by all radio interfaces */ if (!mbss) { - chanspec = chandef_to_chanspec(&cfg->d11inf, - &settings->chandef); - err = brcmf_fil_iovar_int_set(ifp, "chanspec", chanspec); - if (err < 0) { - brcmf_err("Set Channel failed: chspec=%d, %d\n", - chanspec, err); - goto exit; - } - if (is_11d != ifp->vif->is_11d) { err = brcmf_fil_cmd_int_set(ifp, BRCMF_C_SET_REGULATORY, is_11d); @@ -4575,6 +4567,8 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev, err = -EINVAL; goto exit; } + + /* Interface specific setup */ if (dev_role == NL80211_IFTYPE_AP) { if ((brcmf_feat_is_enabled(ifp, BRCMF_FEAT_MBSS)) && (!mbss)) brcmf_fil_iovar_int_set(ifp, "mbss", 1); @@ -4584,6 +4578,17 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev, brcmf_err("setting AP mode failed %d\n", err); goto exit; } + if (!mbss) { + /* Firmware 10.x requires setting channel after enabling + * AP and before bringing interface up. + */ + err = brcmf_fil_iovar_int_set(ifp, "chanspec", chanspec); + if (err < 0) { + brcmf_err("Set Channel failed: chspec=%d, %d\n", + chanspec, err); + goto exit; + } + } err = brcmf_fil_cmd_int_set(ifp, BRCMF_C_UP, 1); if (err < 0) { brcmf_err("BRCMF_C_UP error (%d)\n", err); @@ -4605,7 +4610,13 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev, goto exit; } brcmf_dbg(TRACE, "AP mode configuration complete\n"); - } else { + } else if (dev_role == NL80211_IFTYPE_P2P_GO) { + err = brcmf_fil_iovar_int_set(ifp, "chanspec", chanspec); + if (err < 0) { + brcmf_err("Set Channel failed: chspec=%d, %d\n", + chanspec, err); + goto exit; + } err = brcmf_fil_bsscfg_data_set(ifp, "ssid", &ssid_le, sizeof(ssid_le)); if (err < 0) { @@ -4622,7 +4633,10 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev, } brcmf_dbg(TRACE, "GO mode configuration complete\n"); + } else { + WARN_ON(1); } + set_bit(BRCMF_VIF_STATUS_AP_CREATED, &ifp->vif->sme_state); brcmf_net_setcarrier(ifp, true); -- cgit v0.10.2 From fd3667a8d1cb17f0a91771805be4efaa5c35cda1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Mon, 30 May 2016 06:40:54 +0200 Subject: brcmfmac: don't remove interface on link down firmware event MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There are two firmware events we handle similarly in brcmfmac: BRCMF_E_LINK and BRCMF_E_IF. The difference from firmware point of view is that the first one means BSS remains present in the firmware. Trying to (re)create it (e.g. when adding new virtual interface) will result in an error. Current code treats both events in a similar way. It removes Linux interface for each of them. It works OK with e.g. BCM43602. Its firmware generates both events for each interface. It means we get BRCMF_E_LINK and remove interface. That is soon followed by BRCMF_E_IF which means BSS was also removed in a firmware. The only downside of this is a harmless error like: [ 208.643180] brcmfmac: brcmf_fweh_call_event_handler: no interface object Unfortunately BCM4366 firmware doesn't automatically remove BSS and so it doesn't generate BRCMF_E_IF. In such case we incorrectly remove Linux interface on BRCMF_E_LINK as BSS is still present in the firmware. It results in an error when trying to re-create virtual interface, e.g.: > iw phy phy1 interface add wlan1-1 type __ap [ 3602.929199] brcmfmac: brcmf_ap_add_vif: timeout occurred command failed: I/O error (-5) With this patch we don't remove Linux interface while firmware keeps BSS. Thanks to this we keep a consistent states of host driver and device firmware. Further improvement should be to mark BSS as disabled and remove interface on BRCMF_E_LINK. Then we should add support for reusing BSS-es. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index f533dc2..9651a7c 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -5432,7 +5432,6 @@ brcmf_notify_connect_status_ap(struct brcmf_cfg80211_info *cfg, struct net_device *ndev, const struct brcmf_event_msg *e, void *data) { - struct brcmf_if *ifp = netdev_priv(ndev); static int generation; u32 event = e->event_code; u32 reason = e->reason; @@ -5443,8 +5442,6 @@ brcmf_notify_connect_status_ap(struct brcmf_cfg80211_info *cfg, ndev != cfg_to_ndev(cfg)) { brcmf_dbg(CONN, "AP mode link down\n"); complete(&cfg->vif_disabled); - if (ifp->vif->mbss) - brcmf_remove_interface(ifp); return 0; } -- cgit v0.10.2 From 5345ea6a4bfb956b1d021fbd9c62daa921942d85 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 30 May 2016 17:26:16 +0200 Subject: rtlwifi: fix error handling in *_read_adapter_info() There are nine copies of the _rtl88ee_read_adapter_info() function, and most but not all of them cause a build warning in some configurations: rtl8192de/hw.c: In function '_rtl92de_read_adapter_info': rtl8192de/hw.c:1767:12: error: 'hwinfo' may be used uninitialized in this function [-Werror=maybe-uninitialized] rtl8723ae/hw.c: In function '_rtl8723e_read_adapter_info.constprop': rtlwifi/rtl8723ae/hw.c:1654:12: error: 'hwinfo' may be used uninitialized in this function [-Werror=maybe-uninitialized] The problem is that when rtlefuse->epromtype is something other than EEPROM_BOOT_EFUSE, the rest of the function uses undefined data, resulting in random behavior later. Apparently, in some drivers, the problem was already found and fixed but the fix did not make it into the others. This picks one approach to deal with the problem and applies identical code to all 9 files, to simplify the later consolidation of those. Signed-off-by: Arnd Bergmann Acked-by: Larry Finger Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c index 8ee83b0..e26a233 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c @@ -1839,20 +1839,22 @@ static void _rtl88ee_read_adapter_info(struct ieee80211_hw *hw) u8 hwinfo[HWSET_MAX_SIZE]; u16 eeprom_id; - if (rtlefuse->epromtype == EEPROM_BOOT_EFUSE) { + switch (rtlefuse->epromtype) { + case EEPROM_BOOT_EFUSE: rtl_efuse_shadow_map_update(hw); + break; - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], - HWSET_MAX_SIZE); - } else if (rtlefuse->epromtype == EEPROM_93C46) { + case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "RTL819X Not boot from eeprom, check it !!"); return; - } else { + + default: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "boot from neither eeprom nor efuse, check it !!"); return; } + memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP\n", hwinfo, HWSET_MAX_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c index 04eb5c3..58b7ac6 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c @@ -1680,21 +1680,28 @@ static void _rtl92ce_read_adapter_info(struct ieee80211_hw *hw) struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); + struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; u16 i, usvalue; u8 hwinfo[HWSET_MAX_SIZE]; u16 eeprom_id; - if (rtlefuse->epromtype == EEPROM_BOOT_EFUSE) { + switch (rtlefuse->epromtype) { + case EEPROM_BOOT_EFUSE: rtl_efuse_shadow_map_update(hw); + break; - memcpy((void *)hwinfo, - (void *)&rtlefuse->efuse_map[EFUSE_INIT_MAP][0], - HWSET_MAX_SIZE); - } else if (rtlefuse->epromtype == EEPROM_93C46) { + case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "RTL819X Not boot from eeprom, check it !!"); + return; + + default: + dev_warn(dev, "no efuse data\n"); + return; } + memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); + RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP", hwinfo, HWSET_MAX_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c index 34ce064..ae1129f 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c @@ -351,15 +351,21 @@ static void _rtl92cu_read_adapter_info(struct ieee80211_hw *hw) u8 hwinfo[HWSET_MAX_SIZE] = {0}; u16 eeprom_id; - if (rtlefuse->epromtype == EEPROM_BOOT_EFUSE) { + switch (rtlefuse->epromtype) { + case EEPROM_BOOT_EFUSE: rtl_efuse_shadow_map_update(hw); - memcpy((void *)hwinfo, - (void *)&rtlefuse->efuse_map[EFUSE_INIT_MAP][0], - HWSET_MAX_SIZE); - } else if (rtlefuse->epromtype == EEPROM_93C46) { + break; + + case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "RTL819X Not boot from eeprom, check it !!\n"); + return; + + default: + pr_warn("rtl92cu: no efuse data\n\n"); + return; } + memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_LOUD, "MAP", hwinfo, HWSET_MAX_SIZE); eeprom_id = le16_to_cpu(*((__le16 *)&hwinfo[0])); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c index f49b60d..8618c32 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c @@ -1744,23 +1744,29 @@ static void _rtl92de_read_adapter_info(struct ieee80211_hw *hw) struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); + struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; u16 i, usvalue; u8 hwinfo[HWSET_MAX_SIZE]; u16 eeprom_id; unsigned long flags; - if (rtlefuse->epromtype == EEPROM_BOOT_EFUSE) { + switch (rtlefuse->epromtype) { + case EEPROM_BOOT_EFUSE: spin_lock_irqsave(&globalmutex_for_power_and_efuse, flags); rtl_efuse_shadow_map_update(hw); _rtl92de_efuse_update_chip_version(hw); spin_unlock_irqrestore(&globalmutex_for_power_and_efuse, flags); - memcpy((void *)hwinfo, (void *)&rtlefuse->efuse_map - [EFUSE_INIT_MAP][0], - HWSET_MAX_SIZE); - } else if (rtlefuse->epromtype == EEPROM_93C46) { + break; + case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "RTL819X Not boot from eeprom, check it !!\n"); + return; + default: + dev_warn(dev, "no efuse data\n"); + return; } + + memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP", hwinfo, HWSET_MAX_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c index 9fd3f1b..28c260d 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c @@ -2102,20 +2102,22 @@ static void _rtl92ee_read_adapter_info(struct ieee80211_hw *hw) u8 hwinfo[HWSET_MAX_SIZE]; u16 eeprom_id; - if (rtlefuse->epromtype == EEPROM_BOOT_EFUSE) { + switch (rtlefuse->epromtype) { + case EEPROM_BOOT_EFUSE: rtl_efuse_shadow_map_update(hw); + break; - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], - HWSET_MAX_SIZE); - } else if (rtlefuse->epromtype == EEPROM_93C46) { + case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "RTL819X Not boot from eeprom, check it !!"); return; - } else { + + default: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "boot from neither eeprom nor efuse, check it !!"); return; } + memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP\n", hwinfo, HWSET_MAX_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c index 12b0978..442f2b6 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c @@ -1673,23 +1673,31 @@ static void _rtl92se_read_adapter_info(struct ieee80211_hw *hw) struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_phy *rtlphy = &(rtlpriv->phy); + struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; u16 i, usvalue; u16 eeprom_id; u8 tempval; u8 hwinfo[HWSET_MAX_SIZE_92S]; u8 rf_path, index; - if (rtlefuse->epromtype == EEPROM_93C46) { + switch (rtlefuse->epromtype) { + case EEPROM_BOOT_EFUSE: + rtl_efuse_shadow_map_update(hw); + break; + + case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "RTL819X Not boot from eeprom, check it !!\n"); - } else if (rtlefuse->epromtype == EEPROM_BOOT_EFUSE) { - rtl_efuse_shadow_map_update(hw); + return; - memcpy((void *)hwinfo, (void *) - &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], - HWSET_MAX_SIZE_92S); + default: + dev_warn(dev, "no efuse data\n"); + return; } + memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], + HWSET_MAX_SIZE_92S); + RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP", hwinfo, HWSET_MAX_SIZE_92S); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c index a4b7eac..57a1ba8 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c @@ -1630,6 +1630,7 @@ static void _rtl8723e_read_adapter_info(struct ieee80211_hw *hw, struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); + struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; u16 i, usvalue; u8 hwinfo[HWSET_MAX_SIZE]; u16 eeprom_id; @@ -1638,15 +1639,19 @@ static void _rtl8723e_read_adapter_info(struct ieee80211_hw *hw, /* need add */ return; } - if (rtlefuse->epromtype == EEPROM_BOOT_EFUSE) { + switch (rtlefuse->epromtype) { + case EEPROM_BOOT_EFUSE: rtl_efuse_shadow_map_update(hw); - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], - HWSET_MAX_SIZE); - } else if (rtlefuse->epromtype == EEPROM_93C46) { + case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "RTL819X Not boot from eeprom, check it !!"); + return; + + default: + dev_warn(dev, "no efuse data\n"); } + memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP\n", hwinfo, HWSET_MAX_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c index 5a3df91..08288ac 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c @@ -2026,6 +2026,7 @@ static void _rtl8723be_read_adapter_info(struct ieee80211_hw *hw, struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); + struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; u16 i, usvalue; u8 hwinfo[HWSET_MAX_SIZE]; u16 eeprom_id; @@ -2055,15 +2056,22 @@ static void _rtl8723be_read_adapter_info(struct ieee80211_hw *hw, /* needs to be added */ return; } - if (rtlefuse->epromtype == EEPROM_BOOT_EFUSE) { + + switch (rtlefuse->epromtype) { + case EEPROM_BOOT_EFUSE: rtl_efuse_shadow_map_update(hw); + break; - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], - HWSET_MAX_SIZE); - } else if (rtlefuse->epromtype == EEPROM_93C46) { + case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "RTL819X Not boot from eeprom, check it !!"); + return; + + default: + dev_warn(dev, "no efuse data\n"); + return; } + memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, ("MAP\n"), hwinfo, HWSET_MAX_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c index 71e4dd9..b9436df 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c @@ -3101,6 +3101,7 @@ static void _rtl8821ae_read_adapter_info(struct ieee80211_hw *hw, bool b_pseudo_ struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); struct rtl_pci_priv *pcipriv = rtl_pcipriv(hw); + struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; u16 i, usvalue; u8 hwinfo[HWSET_MAX_SIZE]; u16 eeprom_id; @@ -3109,14 +3110,20 @@ static void _rtl8821ae_read_adapter_info(struct ieee80211_hw *hw, bool b_pseudo_ ;/* need add */ } - if (rtlefuse->epromtype == EEPROM_BOOT_EFUSE) { + switch (rtlefuse->epromtype) { + case EEPROM_BOOT_EFUSE: rtl_efuse_shadow_map_update(hw); - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], - HWSET_MAX_SIZE); - } else if (rtlefuse->epromtype == EEPROM_93C46) { + break; + + case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "RTL819X Not boot from eeprom, check it !!"); + return; + + default: + dev_warn(dev, "no efuse data\n"); } + memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP\n", hwinfo, HWSET_MAX_SIZE); -- cgit v0.10.2 From 035ddbc59f542453491356dbcb4b8c2d548acf43 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Fri, 3 Jun 2016 13:48:55 -0700 Subject: libertas_tf: Drop unused variable and define gcc-6 reports: drivers/net/wireless/marvell/libertas_tf/main.c:30:19: error: 'lbtf_driver_version' defined but not used with -Werror=unused-const-variable=. Reported-by: Fengguang Wu [0-day test robot] Signed-off-by: Guenter Roeck Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/libertas_tf/main.c b/drivers/net/wireless/marvell/libertas_tf/main.c index 0bf8916..75bf0c8 100644 --- a/drivers/net/wireless/marvell/libertas_tf/main.c +++ b/drivers/net/wireless/marvell/libertas_tf/main.c @@ -16,7 +16,6 @@ #include #include "libertas_tf.h" -#define DRIVER_RELEASE_VERSION "004.p0" /* thinfirm version: 5.132.X.pX */ #define LBTF_FW_VER_MIN 0x05840300 #define LBTF_FW_VER_MAX 0x0584ffff @@ -27,12 +26,6 @@ unsigned int lbtf_debug; EXPORT_SYMBOL_GPL(lbtf_debug); module_param_named(libertas_tf_debug, lbtf_debug, int, 0644); -static const char lbtf_driver_version[] = "THINFIRM-USB8388-" DRIVER_RELEASE_VERSION -#ifdef DEBUG - "-dbg" -#endif - ""; - struct workqueue_struct *lbtf_wq; static const struct ieee80211_channel lbtf_channels[] = { -- cgit v0.10.2 From 508f1222ba4e14e0737a3a1b39aed9e3e8a21fc7 Mon Sep 17 00:00:00 2001 From: Lucas Stach Date: Fri, 3 Jun 2016 23:04:03 +0200 Subject: b43: only hardcode LED behavior if SPROM doesn't encode any Only hardcode the LED behavior if the SROM doesn't provide any for all LEDs of the card. This avoids instantiating LED triggers for unconnected LEDs, while (hopefully) keeping things working for old cards with a blank SROM. Signed-off-by: Lucas Stach Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/b43/leds.c b/drivers/net/wireless/broadcom/b43/leds.c index d79ab2a..cb987c2 100644 --- a/drivers/net/wireless/broadcom/b43/leds.c +++ b/drivers/net/wireless/broadcom/b43/leds.c @@ -222,7 +222,7 @@ static void b43_led_get_sprominfo(struct b43_wldev *dev, sprom[2] = dev->dev->bus_sprom->gpio2; sprom[3] = dev->dev->bus_sprom->gpio3; - if (sprom[led_index] == 0xFF) { + if ((sprom[0] & sprom[1] & sprom[2] & sprom[3]) == 0xff) { /* There is no LED information in the SPROM * for this LED. Hardcode it here. */ *activelow = false; @@ -250,7 +250,11 @@ static void b43_led_get_sprominfo(struct b43_wldev *dev, return; } } else { - *behaviour = sprom[led_index] & B43_LED_BEHAVIOUR; + /* keep LED disabled if no mapping is defined */ + if (sprom[led_index] == 0xff) + *behaviour = B43_LED_OFF; + else + *behaviour = sprom[led_index] & B43_LED_BEHAVIOUR; *activelow = !!(sprom[led_index] & B43_LED_ACTIVELOW); } } -- cgit v0.10.2 From 10d096f708e870af7707847bfba3b39fab00c621 Mon Sep 17 00:00:00 2001 From: Hante Meuleman Date: Fri, 3 Jun 2016 23:31:07 +0200 Subject: brcmfmac: fix skb priority handling SKBs can come with a prioriy. Currently a priority of 0..7 is assumed. But this assumption is incorrect. To fix this any priority of 0 or higher then 7 will be adjusted by calling cfg80211_classify8021d Reviewed-by: Arend Van Spriel Reviewed-by: Franky Lin Reviewed-by: Pieter-Paul Giesberts Signed-off-by: Hante Meuleman Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c index 5b30922..cd221ab 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c @@ -2101,7 +2101,7 @@ int brcmf_fws_process_skb(struct brcmf_if *ifp, struct sk_buff *skb) brcmf_dbg(DATA, "tx proto=0x%X\n", ntohs(eh->h_proto)); /* determine the priority */ - if (!skb->priority) + if ((skb->priority == 0) || (skb->priority > 7)) skb->priority = cfg80211_classify8021d(skb, NULL); drvr->tx_multicast += !!multicast; -- cgit v0.10.2 From d922dfa372ed94e8f2354f83204159e8fd6c1807 Mon Sep 17 00:00:00 2001 From: Wright Feng Date: Fri, 3 Jun 2016 23:31:08 +0200 Subject: brcmfmac: revise SDIO error message in brcmf_sdio_drivestrengthinit The error message is given for something that is not an error here as the drive strength configuration may not be applicable for specific devices. Therefor the error message is rephrased and changed to a debug message. Signed-off-by: Wright Feng [arend@broadcom.com: rephrase commit message] Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c index 67e69bf..22b7dc0 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c @@ -3666,7 +3666,7 @@ brcmf_sdio_drivestrengthinit(struct brcmf_sdio_dev *sdiodev, str_shift = 11; break; default: - brcmf_err("No SDIO Drive strength init done for chip %s rev %d pmurev %d\n", + brcmf_dbg(INFO, "No SDIO driver strength init needed for chip %s rev %d pmurev %d\n", ci->name, ci->chiprev, ci->pmurev); break; } -- cgit v0.10.2 From cb39288fd6bb8231a6eb41e8c28ada093d274e28 Mon Sep 17 00:00:00 2001 From: Wright Feng Date: Fri, 3 Jun 2016 23:31:09 +0200 Subject: brcmfmac: use ndev->needed_headroom to reserve additional header space When using nmap tool with FMAC, the nmap packets were be dropped by kernel because the size was too short. The kernel message showed like "nmap: packet size is too short (42 <= 50)". It is caused by the packet length is shorter than ndev->hard_header_len. According to definition of LL_RESERVED_SPACE() and hard_header_len, we should use hard_header_len to reserve for L2 header, like ethernet header(ETH_HLEN) in our case and use needed_headroom for the additional headroom needed by hardware. Reviewed-by: Arend Van Spriel Signed-off-by: Wright Feng Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c index 6a76480..faf4e46 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c @@ -516,7 +516,7 @@ int brcmf_net_attach(struct brcmf_if *ifp, bool rtnl_locked) /* set appropriate operations */ ndev->netdev_ops = &brcmf_netdev_ops_pri; - ndev->hard_header_len += drvr->hdrlen; + ndev->needed_headroom += drvr->hdrlen; ndev->ethtool_ops = &brcmf_ethtool_ops; drvr->rxsz = ndev->mtu + ndev->hard_header_len + -- cgit v0.10.2 From 43819926eda54c38f31ce6175f3edf6c2987ceac Mon Sep 17 00:00:00 2001 From: Hante Meuleman Date: Fri, 3 Jun 2016 23:31:10 +0200 Subject: brcmfmac: add support for the PCIE devices 43525 and 43465 This patch adds support for the new PCIE devices 43525 and 43465. Reviewed-by: Arend Van Spriel Reviewed-by: Franky (Zhenhui) Lin Reviewed-by: Pieter-Paul Giesberts Signed-off-by: Hante Meuleman Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c index d3fd6b1..05f22ff 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/chip.c @@ -685,6 +685,8 @@ static u32 brcmf_chip_tcm_rambase(struct brcmf_chip_priv *ci) case BRCM_CC_43602_CHIP_ID: case BRCM_CC_4371_CHIP_ID: return 0x180000; + case BRCM_CC_43465_CHIP_ID: + case BRCM_CC_43525_CHIP_ID: case BRCM_CC_4365_CHIP_ID: case BRCM_CC_4366_CHIP_ID: return 0x200000; diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c index 0af8db8..3deba90 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c @@ -54,21 +54,25 @@ BRCMF_FW_NVRAM_DEF(43570, "brcmfmac43570-pcie.bin", "brcmfmac43570-pcie.txt"); BRCMF_FW_NVRAM_DEF(4358, "brcmfmac4358-pcie.bin", "brcmfmac4358-pcie.txt"); BRCMF_FW_NVRAM_DEF(4359, "brcmfmac4359-pcie.bin", "brcmfmac4359-pcie.txt"); BRCMF_FW_NVRAM_DEF(4365B, "brcmfmac4365b-pcie.bin", "brcmfmac4365b-pcie.txt"); +BRCMF_FW_NVRAM_DEF(4365C, "brcmfmac4365c-pcie.bin", "brcmfmac4365c-pcie.txt"); BRCMF_FW_NVRAM_DEF(4366B, "brcmfmac4366b-pcie.bin", "brcmfmac4366b-pcie.txt"); BRCMF_FW_NVRAM_DEF(4366C, "brcmfmac4366c-pcie.bin", "brcmfmac4366c-pcie.txt"); BRCMF_FW_NVRAM_DEF(4371, "brcmfmac4371-pcie.bin", "brcmfmac4371-pcie.txt"); static struct brcmf_firmware_mapping brcmf_pcie_fwnames[] = { BRCMF_FW_NVRAM_ENTRY(BRCM_CC_43602_CHIP_ID, 0xFFFFFFFF, 43602), + BRCMF_FW_NVRAM_ENTRY(BRCM_CC_43465_CHIP_ID, 0xFFFFFFF0, 4366C), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4350_CHIP_ID, 0x000000FF, 4350C), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4350_CHIP_ID, 0xFFFFFF00, 4350), + BRCMF_FW_NVRAM_ENTRY(BRCM_CC_43525_CHIP_ID, 0xFFFFFFF0, 4365C), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4356_CHIP_ID, 0xFFFFFFFF, 4356), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_43567_CHIP_ID, 0xFFFFFFFF, 43570), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_43569_CHIP_ID, 0xFFFFFFFF, 43570), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_43570_CHIP_ID, 0xFFFFFFFF, 43570), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4358_CHIP_ID, 0xFFFFFFFF, 4358), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4359_CHIP_ID, 0xFFFFFFFF, 4359), - BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4365_CHIP_ID, 0xFFFFFFFF, 4365B), + BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4365_CHIP_ID, 0x0000000F, 4365B), + BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4365_CHIP_ID, 0xFFFFFFF0, 4365C), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4366_CHIP_ID, 0x0000000F, 4366B), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4366_CHIP_ID, 0xFFFFFFF0, 4366C), BRCMF_FW_NVRAM_ENTRY(BRCM_CC_4371_CHIP_ID, 0xFFFFFFFF, 4371), diff --git a/drivers/net/wireless/broadcom/brcm80211/include/brcm_hw_ids.h b/drivers/net/wireless/broadcom/brcm80211/include/brcm_hw_ids.h index 699f2c2..3cc42be 100644 --- a/drivers/net/wireless/broadcom/brcm80211/include/brcm_hw_ids.h +++ b/drivers/net/wireless/broadcom/brcm80211/include/brcm_hw_ids.h @@ -40,7 +40,9 @@ #define BRCM_CC_4339_CHIP_ID 0x4339 #define BRCM_CC_43430_CHIP_ID 43430 #define BRCM_CC_4345_CHIP_ID 0x4345 +#define BRCM_CC_43465_CHIP_ID 43465 #define BRCM_CC_4350_CHIP_ID 0x4350 +#define BRCM_CC_43525_CHIP_ID 43525 #define BRCM_CC_4354_CHIP_ID 0x4354 #define BRCM_CC_4356_CHIP_ID 0x4356 #define BRCM_CC_43566_CHIP_ID 43566 -- cgit v0.10.2 From 98aff6c005e322f2bd4a91ae2f3b3be3b8193d9b Mon Sep 17 00:00:00 2001 From: Wright Feng Date: Fri, 3 Jun 2016 23:31:11 +0200 Subject: brcmfmac: change rx_seq check log from error print to debug print The bus rx sequence is not in order because that control and event frames always cause immediate send, but data frames may be held for glomming in firmware side. It is not actually an error as the packets are still processed even if the RX sequence is not in order. Therefor the error message is rephrased and changed to a debug message. Reviewed-by: Arend Van Spriel Signed-off-by: Wright Feng Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c index 22b7dc0..5fb8b91 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c @@ -1384,8 +1384,7 @@ static int brcmf_sdio_hdparse(struct brcmf_sdio *bus, u8 *header, return -ENXIO; } if (rd->seq_num != rx_seq) { - brcmf_err("seq %d: sequence number error, expect %d\n", - rx_seq, rd->seq_num); + brcmf_dbg(SDIO, "seq %d, expected %d\n", rx_seq, rd->seq_num); bus->sdcnt.rx_badseq++; rd->seq_num = rx_seq; } -- cgit v0.10.2 From 2a734451028670cf8c14e233b79d95fd3df99bca Mon Sep 17 00:00:00 2001 From: Arend van Spriel Date: Fri, 3 Jun 2016 23:31:12 +0200 Subject: brcm80211: update maintainers email addresses Update MAINTAINERS file because of organizational changes. Reviewed-by: Hante Meuleman Reviewed-by: Pieter-Paul Giesberts Reviewed-by: Franky Lin Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/MAINTAINERS b/MAINTAINERS index f8d5a37..56934b5 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2570,12 +2570,11 @@ S: Supported F: drivers/net/ethernet/broadcom/tg3.* BROADCOM BRCM80211 IEEE802.11n WIRELESS DRIVER -M: Brett Rudley -M: Arend van Spriel -M: Franky (Zhenhui) Lin -M: Hante Meuleman +M: Arend van Spriel +M: Franky Lin +M: Hante Meuleman L: linux-wireless@vger.kernel.org -L: brcm80211-dev-list@broadcom.com +L: brcm80211-dev-list.pdl@broadcom.com S: Supported F: drivers/net/wireless/broadcom/brcm80211/ -- cgit v0.10.2 From f9f905b00b878243e1e3cbe32b920f86a8aaa68e Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Sat, 4 Jun 2016 19:29:01 +0530 Subject: libertas: Remove create_workqueue alloc_workqueue replaces deprecated create_workqueue(). In if_sdio.c, the workqueue card->workqueue has workitem &card->packet_worker, which is mapped to if_sdio_host_to_card_worker. The workitem is involved in sending packets to firmware. Forward progress under memory pressure is a requirement here. In if_spi.c, the workqueue card->workqueue has workitem &card->packet_worker, which is mapped to if_spi_host_to_card_worker. The workitem is involved in sending command packets from the host. Forward progress under memory pressure is a requirement here. Dedicated workqueues have been used in both cases since the workitems on the workqueues are involved in normal device operation with WQ_MEM_RECLAIM set to gurantee forward progress under memory pressure. Since there are only a fixed number of work items, explicit concurrency limit is unnecessary. flush_workqueue is unnecessary since destroy_workqueue() itself calls drain_workqueue() which flushes repeatedly till the workqueue becomes empty. Hence the calls to flush_workqueue() before destroy_workqueue() have been dropped. Signed-off-by: Bhaktipriya Shridhar Acked-by: Tejun Heo Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/libertas/if_sdio.c b/drivers/net/wireless/marvell/libertas/if_sdio.c index 13eae9f..47f4a14 100644 --- a/drivers/net/wireless/marvell/libertas/if_sdio.c +++ b/drivers/net/wireless/marvell/libertas/if_sdio.c @@ -1228,7 +1228,7 @@ static int if_sdio_probe(struct sdio_func *func, } spin_lock_init(&card->lock); - card->workqueue = create_workqueue("libertas_sdio"); + card->workqueue = alloc_workqueue("libertas_sdio", WQ_MEM_RECLAIM, 0); INIT_WORK(&card->packet_worker, if_sdio_host_to_card_worker); init_waitqueue_head(&card->pwron_waitq); @@ -1326,7 +1326,6 @@ static void if_sdio_remove(struct sdio_func *func) lbs_stop_card(card->priv); lbs_remove_card(card->priv); - flush_workqueue(card->workqueue); destroy_workqueue(card->workqueue); while (card->packets) { diff --git a/drivers/net/wireless/marvell/libertas/if_spi.c b/drivers/net/wireless/marvell/libertas/if_spi.c index 82c0796..c3a53cd 100644 --- a/drivers/net/wireless/marvell/libertas/if_spi.c +++ b/drivers/net/wireless/marvell/libertas/if_spi.c @@ -1180,7 +1180,7 @@ static int if_spi_probe(struct spi_device *spi) priv->fw_ready = 1; /* Initialize interrupt handling stuff. */ - card->workqueue = create_workqueue("libertas_spi"); + card->workqueue = alloc_workqueue("libertas_spi", WQ_MEM_RECLAIM, 0); INIT_WORK(&card->packet_work, if_spi_host_to_card_worker); INIT_WORK(&card->resume_work, if_spi_resume_worker); @@ -1208,7 +1208,6 @@ static int if_spi_probe(struct spi_device *spi) release_irq: free_irq(spi->irq, card); terminate_workqueue: - flush_workqueue(card->workqueue); destroy_workqueue(card->workqueue); lbs_remove_card(priv); /* will call free_netdev */ free_card: @@ -1235,7 +1234,6 @@ static int libertas_spi_remove(struct spi_device *spi) lbs_remove_card(priv); /* will call free_netdev */ free_irq(spi->irq, card); - flush_workqueue(card->workqueue); destroy_workqueue(card->workqueue); if (card->pdata->teardown) card->pdata->teardown(spi); -- cgit v0.10.2 From 9791333a840f292ab32746264803f957de3aa3a9 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sat, 4 Jun 2016 07:54:12 -0700 Subject: b43: Remove unused phy_a code MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit gcc-6 reports the following error with -Werror=unused-const-variable. drivers/net/wireless/broadcom/b43/phy_a.c:576:40: error: 'b43_phyops_a' defined but not used Per Michael Büsch: "All a-phy code is usused", so remove it all, and move the remaining Type-G initialization code into phy_g.c. Reported-by: Fengguang Wu [0-day test robot] Cc: Michael Büsch Signed-off-by: Guenter Roeck Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/b43/Makefile b/drivers/net/wireless/broadcom/b43/Makefile index ddc4df4..27fab958 100644 --- a/drivers/net/wireless/broadcom/b43/Makefile +++ b/drivers/net/wireless/broadcom/b43/Makefile @@ -1,6 +1,6 @@ b43-y += main.o b43-y += bus.o -b43-$(CONFIG_B43_PHY_G) += phy_a.o phy_g.o tables.o lo.o wa.o +b43-$(CONFIG_B43_PHY_G) += phy_g.o tables.o lo.o wa.o b43-$(CONFIG_B43_PHY_N) += tables_nphy.o b43-$(CONFIG_B43_PHY_N) += radio_2055.o b43-$(CONFIG_B43_PHY_N) += radio_2056.o diff --git a/drivers/net/wireless/broadcom/b43/phy_a.c b/drivers/net/wireless/broadcom/b43/phy_a.c deleted file mode 100644 index 99c036f..0000000 --- a/drivers/net/wireless/broadcom/b43/phy_a.c +++ /dev/null @@ -1,595 +0,0 @@ -/* - - Broadcom B43 wireless driver - IEEE 802.11a PHY driver - - Copyright (c) 2005 Martin Langer , - Copyright (c) 2005-2007 Stefano Brivio - Copyright (c) 2005-2008 Michael Buesch - Copyright (c) 2005, 2006 Danny van Dyk - Copyright (c) 2005, 2006 Andreas Jaggi - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2 of the License, or - (at your option) any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, Inc., 51 Franklin Steet, Fifth Floor, - Boston, MA 02110-1301, USA. - -*/ - -#include - -#include "b43.h" -#include "phy_a.h" -#include "phy_common.h" -#include "wa.h" -#include "tables.h" -#include "main.h" - - -/* Get the freq, as it has to be written to the device. */ -static inline u16 channel2freq_a(u8 channel) -{ - B43_WARN_ON(channel > 200); - - return (5000 + 5 * channel); -} - -static inline u16 freq_r3A_value(u16 frequency) -{ - u16 value; - - if (frequency < 5091) - value = 0x0040; - else if (frequency < 5321) - value = 0x0000; - else if (frequency < 5806) - value = 0x0080; - else - value = 0x0040; - - return value; -} - -#if 0 -/* This function converts a TSSI value to dBm in Q5.2 */ -static s8 b43_aphy_estimate_power_out(struct b43_wldev *dev, s8 tssi) -{ - struct b43_phy *phy = &dev->phy; - struct b43_phy_a *aphy = phy->a; - s8 dbm = 0; - s32 tmp; - - tmp = (aphy->tgt_idle_tssi - aphy->cur_idle_tssi + tssi); - tmp += 0x80; - tmp = clamp_val(tmp, 0x00, 0xFF); - dbm = aphy->tssi2dbm[tmp]; - //TODO: There's a FIXME on the specs - - return dbm; -} -#endif - -static void b43_radio_set_tx_iq(struct b43_wldev *dev) -{ - static const u8 data_high[5] = { 0x00, 0x40, 0x80, 0x90, 0xD0 }; - static const u8 data_low[5] = { 0x00, 0x01, 0x05, 0x06, 0x0A }; - u16 tmp = b43_radio_read16(dev, 0x001E); - int i, j; - - for (i = 0; i < 5; i++) { - for (j = 0; j < 5; j++) { - if (tmp == (data_high[i] << 4 | data_low[j])) { - b43_phy_write(dev, 0x0069, - (i - j) << 8 | 0x00C0); - return; - } - } - } -} - -static void aphy_channel_switch(struct b43_wldev *dev, unsigned int channel) -{ - u16 freq, r8, tmp; - - freq = channel2freq_a(channel); - - r8 = b43_radio_read16(dev, 0x0008); - b43_write16(dev, 0x03F0, freq); - b43_radio_write16(dev, 0x0008, r8); - - //TODO: write max channel TX power? to Radio 0x2D - tmp = b43_radio_read16(dev, 0x002E); - tmp &= 0x0080; - //TODO: OR tmp with the Power out estimation for this channel? - b43_radio_write16(dev, 0x002E, tmp); - - if (freq >= 4920 && freq <= 5500) { - /* - * r8 = (((freq * 15 * 0xE1FC780F) >> 32) / 29) & 0x0F; - * = (freq * 0.025862069 - */ - r8 = 3 * freq / 116; /* is equal to r8 = freq * 0.025862 */ - } - b43_radio_write16(dev, 0x0007, (r8 << 4) | r8); - b43_radio_write16(dev, 0x0020, (r8 << 4) | r8); - b43_radio_write16(dev, 0x0021, (r8 << 4) | r8); - b43_radio_maskset(dev, 0x0022, 0x000F, (r8 << 4)); - b43_radio_write16(dev, 0x002A, (r8 << 4)); - b43_radio_write16(dev, 0x002B, (r8 << 4)); - b43_radio_maskset(dev, 0x0008, 0x00F0, (r8 << 4)); - b43_radio_maskset(dev, 0x0029, 0xFF0F, 0x00B0); - b43_radio_write16(dev, 0x0035, 0x00AA); - b43_radio_write16(dev, 0x0036, 0x0085); - b43_radio_maskset(dev, 0x003A, 0xFF20, freq_r3A_value(freq)); - b43_radio_mask(dev, 0x003D, 0x00FF); - b43_radio_maskset(dev, 0x0081, 0xFF7F, 0x0080); - b43_radio_mask(dev, 0x0035, 0xFFEF); - b43_radio_maskset(dev, 0x0035, 0xFFEF, 0x0010); - b43_radio_set_tx_iq(dev); - //TODO: TSSI2dbm workaround -//FIXME b43_phy_xmitpower(dev); -} - -static void b43_radio_init2060(struct b43_wldev *dev) -{ - b43_radio_write16(dev, 0x0004, 0x00C0); - b43_radio_write16(dev, 0x0005, 0x0008); - b43_radio_write16(dev, 0x0009, 0x0040); - b43_radio_write16(dev, 0x0005, 0x00AA); - b43_radio_write16(dev, 0x0032, 0x008F); - b43_radio_write16(dev, 0x0006, 0x008F); - b43_radio_write16(dev, 0x0034, 0x008F); - b43_radio_write16(dev, 0x002C, 0x0007); - b43_radio_write16(dev, 0x0082, 0x0080); - b43_radio_write16(dev, 0x0080, 0x0000); - b43_radio_write16(dev, 0x003F, 0x00DA); - b43_radio_mask(dev, 0x0005, ~0x0008); - b43_radio_mask(dev, 0x0081, ~0x0010); - b43_radio_mask(dev, 0x0081, ~0x0020); - b43_radio_mask(dev, 0x0081, ~0x0020); - msleep(1); /* delay 400usec */ - - b43_radio_maskset(dev, 0x0081, ~0x0020, 0x0010); - msleep(1); /* delay 400usec */ - - b43_radio_maskset(dev, 0x0005, ~0x0008, 0x0008); - b43_radio_mask(dev, 0x0085, ~0x0010); - b43_radio_mask(dev, 0x0005, ~0x0008); - b43_radio_mask(dev, 0x0081, ~0x0040); - b43_radio_maskset(dev, 0x0081, ~0x0040, 0x0040); - b43_radio_write16(dev, 0x0005, - (b43_radio_read16(dev, 0x0081) & ~0x0008) | 0x0008); - b43_phy_write(dev, 0x0063, 0xDDC6); - b43_phy_write(dev, 0x0069, 0x07BE); - b43_phy_write(dev, 0x006A, 0x0000); - - aphy_channel_switch(dev, dev->phy.ops->get_default_chan(dev)); - - msleep(1); -} - -static void b43_phy_rssiagc(struct b43_wldev *dev, u8 enable) -{ - int i; - - if (dev->phy.rev < 3) { - if (enable) - for (i = 0; i < B43_TAB_RSSIAGC1_SIZE; i++) { - b43_ofdmtab_write16(dev, - B43_OFDMTAB_LNAHPFGAIN1, i, 0xFFF8); - b43_ofdmtab_write16(dev, - B43_OFDMTAB_WRSSI, i, 0xFFF8); - } - else - for (i = 0; i < B43_TAB_RSSIAGC1_SIZE; i++) { - b43_ofdmtab_write16(dev, - B43_OFDMTAB_LNAHPFGAIN1, i, b43_tab_rssiagc1[i]); - b43_ofdmtab_write16(dev, - B43_OFDMTAB_WRSSI, i, b43_tab_rssiagc1[i]); - } - } else { - if (enable) - for (i = 0; i < B43_TAB_RSSIAGC1_SIZE; i++) - b43_ofdmtab_write16(dev, - B43_OFDMTAB_WRSSI, i, 0x0820); - else - for (i = 0; i < B43_TAB_RSSIAGC2_SIZE; i++) - b43_ofdmtab_write16(dev, - B43_OFDMTAB_WRSSI, i, b43_tab_rssiagc2[i]); - } -} - -static void b43_phy_ww(struct b43_wldev *dev) -{ - u16 b, curr_s, best_s = 0xFFFF; - int i; - - b43_phy_mask(dev, B43_PHY_CRS0, ~B43_PHY_CRS0_EN); - b43_phy_set(dev, B43_PHY_OFDM(0x1B), 0x1000); - b43_phy_maskset(dev, B43_PHY_OFDM(0x82), 0xF0FF, 0x0300); - b43_radio_set(dev, 0x0009, 0x0080); - b43_radio_maskset(dev, 0x0012, 0xFFFC, 0x0002); - b43_wa_initgains(dev); - b43_phy_write(dev, B43_PHY_OFDM(0xBA), 0x3ED5); - b = b43_phy_read(dev, B43_PHY_PWRDOWN); - b43_phy_write(dev, B43_PHY_PWRDOWN, (b & 0xFFF8) | 0x0005); - b43_radio_set(dev, 0x0004, 0x0004); - for (i = 0x10; i <= 0x20; i++) { - b43_radio_write16(dev, 0x0013, i); - curr_s = b43_phy_read(dev, B43_PHY_OTABLEQ) & 0x00FF; - if (!curr_s) { - best_s = 0x0000; - break; - } else if (curr_s >= 0x0080) - curr_s = 0x0100 - curr_s; - if (curr_s < best_s) - best_s = curr_s; - } - b43_phy_write(dev, B43_PHY_PWRDOWN, b); - b43_radio_mask(dev, 0x0004, 0xFFFB); - b43_radio_write16(dev, 0x0013, best_s); - b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC1_R1, 0, 0xFFEC); - b43_phy_write(dev, B43_PHY_OFDM(0xB7), 0x1E80); - b43_phy_write(dev, B43_PHY_OFDM(0xB6), 0x1C00); - b43_phy_write(dev, B43_PHY_OFDM(0xB5), 0x0EC0); - b43_phy_write(dev, B43_PHY_OFDM(0xB2), 0x00C0); - b43_phy_write(dev, B43_PHY_OFDM(0xB9), 0x1FFF); - b43_phy_maskset(dev, B43_PHY_OFDM(0xBB), 0xF000, 0x0053); - b43_phy_maskset(dev, B43_PHY_OFDM61, 0xFE1F, 0x0120); - b43_phy_maskset(dev, B43_PHY_OFDM(0x13), 0x0FFF, 0x3000); - b43_phy_maskset(dev, B43_PHY_OFDM(0x14), 0x0FFF, 0x3000); - b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC1, 6, 0x0017); - for (i = 0; i < 6; i++) - b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC1, i, 0x000F); - b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC1, 0x0D, 0x000E); - b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC1, 0x0E, 0x0011); - b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC1, 0x0F, 0x0013); - b43_phy_write(dev, B43_PHY_OFDM(0x33), 0x5030); - b43_phy_set(dev, B43_PHY_CRS0, B43_PHY_CRS0_EN); -} - -static void hardware_pctl_init_aphy(struct b43_wldev *dev) -{ - //TODO -} - -void b43_phy_inita(struct b43_wldev *dev) -{ - struct b43_phy *phy = &dev->phy; - - /* This lowlevel A-PHY init is also called from G-PHY init. - * So we must not access phy->a, if called from G-PHY code. - */ - B43_WARN_ON((phy->type != B43_PHYTYPE_A) && - (phy->type != B43_PHYTYPE_G)); - - might_sleep(); - - if (phy->rev >= 6) { - if (phy->type == B43_PHYTYPE_A) - b43_phy_mask(dev, B43_PHY_OFDM(0x1B), ~0x1000); - if (b43_phy_read(dev, B43_PHY_ENCORE) & B43_PHY_ENCORE_EN) - b43_phy_set(dev, B43_PHY_ENCORE, 0x0010); - else - b43_phy_mask(dev, B43_PHY_ENCORE, ~0x1010); - } - - b43_wa_all(dev); - - if (phy->type == B43_PHYTYPE_A) { - if (phy->gmode && (phy->rev < 3)) - b43_phy_set(dev, 0x0034, 0x0001); - b43_phy_rssiagc(dev, 0); - - b43_phy_set(dev, B43_PHY_CRS0, B43_PHY_CRS0_EN); - - b43_radio_init2060(dev); - - if ((dev->dev->board_vendor == SSB_BOARDVENDOR_BCM) && - ((dev->dev->board_type == SSB_BOARD_BU4306) || - (dev->dev->board_type == SSB_BOARD_BU4309))) { - ; //TODO: A PHY LO - } - - if (phy->rev >= 3) - b43_phy_ww(dev); - - hardware_pctl_init_aphy(dev); - - //TODO: radar detection - } - - if ((phy->type == B43_PHYTYPE_G) && - (dev->dev->bus_sprom->boardflags_lo & B43_BFL_PACTRL)) { - b43_phy_maskset(dev, B43_PHY_OFDM(0x6E), 0xE000, 0x3CF); - } -} - -/* Initialise the TSSI->dBm lookup table */ -static int b43_aphy_init_tssi2dbm_table(struct b43_wldev *dev) -{ - struct b43_phy *phy = &dev->phy; - struct b43_phy_a *aphy = phy->a; - s16 pab0, pab1, pab2; - - pab0 = (s16) (dev->dev->bus_sprom->pa1b0); - pab1 = (s16) (dev->dev->bus_sprom->pa1b1); - pab2 = (s16) (dev->dev->bus_sprom->pa1b2); - - if (pab0 != 0 && pab1 != 0 && pab2 != 0 && - pab0 != -1 && pab1 != -1 && pab2 != -1) { - /* The pabX values are set in SPROM. Use them. */ - if ((s8) dev->dev->bus_sprom->itssi_a != 0 && - (s8) dev->dev->bus_sprom->itssi_a != -1) - aphy->tgt_idle_tssi = - (s8) (dev->dev->bus_sprom->itssi_a); - else - aphy->tgt_idle_tssi = 62; - aphy->tssi2dbm = b43_generate_dyn_tssi2dbm_tab(dev, pab0, - pab1, pab2); - if (!aphy->tssi2dbm) - return -ENOMEM; - } else { - /* pabX values not set in SPROM, - * but APHY needs a generated table. */ - aphy->tssi2dbm = NULL; - b43err(dev->wl, "Could not generate tssi2dBm " - "table (wrong SPROM info)!\n"); - return -ENODEV; - } - - return 0; -} - -static int b43_aphy_op_allocate(struct b43_wldev *dev) -{ - struct b43_phy_a *aphy; - int err; - - aphy = kzalloc(sizeof(*aphy), GFP_KERNEL); - if (!aphy) - return -ENOMEM; - dev->phy.a = aphy; - - err = b43_aphy_init_tssi2dbm_table(dev); - if (err) - goto err_free_aphy; - - return 0; - -err_free_aphy: - kfree(aphy); - dev->phy.a = NULL; - - return err; -} - -static void b43_aphy_op_prepare_structs(struct b43_wldev *dev) -{ - struct b43_phy *phy = &dev->phy; - struct b43_phy_a *aphy = phy->a; - const void *tssi2dbm; - int tgt_idle_tssi; - - /* tssi2dbm table is constant, so it is initialized at alloc time. - * Save a copy of the pointer. */ - tssi2dbm = aphy->tssi2dbm; - tgt_idle_tssi = aphy->tgt_idle_tssi; - - /* Zero out the whole PHY structure. */ - memset(aphy, 0, sizeof(*aphy)); - - aphy->tssi2dbm = tssi2dbm; - aphy->tgt_idle_tssi = tgt_idle_tssi; - - //TODO init struct b43_phy_a - -} - -static void b43_aphy_op_free(struct b43_wldev *dev) -{ - struct b43_phy *phy = &dev->phy; - struct b43_phy_a *aphy = phy->a; - - kfree(aphy->tssi2dbm); - aphy->tssi2dbm = NULL; - - kfree(aphy); - dev->phy.a = NULL; -} - -static int b43_aphy_op_init(struct b43_wldev *dev) -{ - b43_phy_inita(dev); - - return 0; -} - -static inline u16 adjust_phyreg(struct b43_wldev *dev, u16 offset) -{ - /* OFDM registers are base-registers for the A-PHY. */ - if ((offset & B43_PHYROUTE) == B43_PHYROUTE_OFDM_GPHY) { - offset &= ~B43_PHYROUTE; - offset |= B43_PHYROUTE_BASE; - } - -#if B43_DEBUG - if ((offset & B43_PHYROUTE) == B43_PHYROUTE_EXT_GPHY) { - /* Ext-G registers are only available on G-PHYs */ - b43err(dev->wl, "Invalid EXT-G PHY access at " - "0x%04X on A-PHY\n", offset); - dump_stack(); - } - if ((offset & B43_PHYROUTE) == B43_PHYROUTE_N_BMODE) { - /* N-BMODE registers are only available on N-PHYs */ - b43err(dev->wl, "Invalid N-BMODE PHY access at " - "0x%04X on A-PHY\n", offset); - dump_stack(); - } -#endif /* B43_DEBUG */ - - return offset; -} - -static u16 b43_aphy_op_read(struct b43_wldev *dev, u16 reg) -{ - reg = adjust_phyreg(dev, reg); - b43_write16f(dev, B43_MMIO_PHY_CONTROL, reg); - return b43_read16(dev, B43_MMIO_PHY_DATA); -} - -static void b43_aphy_op_write(struct b43_wldev *dev, u16 reg, u16 value) -{ - reg = adjust_phyreg(dev, reg); - b43_write16f(dev, B43_MMIO_PHY_CONTROL, reg); - b43_write16(dev, B43_MMIO_PHY_DATA, value); -} - -static u16 b43_aphy_op_radio_read(struct b43_wldev *dev, u16 reg) -{ - /* Register 1 is a 32-bit register. */ - B43_WARN_ON(reg == 1); - /* A-PHY needs 0x40 for read access */ - reg |= 0x40; - - b43_write16(dev, B43_MMIO_RADIO_CONTROL, reg); - return b43_read16(dev, B43_MMIO_RADIO_DATA_LOW); -} - -static void b43_aphy_op_radio_write(struct b43_wldev *dev, u16 reg, u16 value) -{ - /* Register 1 is a 32-bit register. */ - B43_WARN_ON(reg == 1); - - b43_write16(dev, B43_MMIO_RADIO_CONTROL, reg); - b43_write16(dev, B43_MMIO_RADIO_DATA_LOW, value); -} - -static bool b43_aphy_op_supports_hwpctl(struct b43_wldev *dev) -{ - return (dev->phy.rev >= 5); -} - -static void b43_aphy_op_software_rfkill(struct b43_wldev *dev, - bool blocked) -{ - struct b43_phy *phy = &dev->phy; - - if (!blocked) { - if (phy->radio_on) - return; - b43_radio_write16(dev, 0x0004, 0x00C0); - b43_radio_write16(dev, 0x0005, 0x0008); - b43_phy_mask(dev, 0x0010, 0xFFF7); - b43_phy_mask(dev, 0x0011, 0xFFF7); - b43_radio_init2060(dev); - } else { - b43_radio_write16(dev, 0x0004, 0x00FF); - b43_radio_write16(dev, 0x0005, 0x00FB); - b43_phy_set(dev, 0x0010, 0x0008); - b43_phy_set(dev, 0x0011, 0x0008); - } -} - -static int b43_aphy_op_switch_channel(struct b43_wldev *dev, - unsigned int new_channel) -{ - if (new_channel > 200) - return -EINVAL; - aphy_channel_switch(dev, new_channel); - - return 0; -} - -static unsigned int b43_aphy_op_get_default_chan(struct b43_wldev *dev) -{ - return 36; /* Default to channel 36 */ -} - -static void b43_aphy_op_set_rx_antenna(struct b43_wldev *dev, int antenna) -{//TODO - struct b43_phy *phy = &dev->phy; - u16 tmp; - int autodiv = 0; - - if (antenna == B43_ANTENNA_AUTO0 || antenna == B43_ANTENNA_AUTO1) - autodiv = 1; - - b43_hf_write(dev, b43_hf_read(dev) & ~B43_HF_ANTDIVHELP); - - b43_phy_maskset(dev, B43_PHY_BBANDCFG, ~B43_PHY_BBANDCFG_RXANT, - (autodiv ? B43_ANTENNA_AUTO1 : antenna) << - B43_PHY_BBANDCFG_RXANT_SHIFT); - - if (autodiv) { - tmp = b43_phy_read(dev, B43_PHY_ANTDWELL); - if (antenna == B43_ANTENNA_AUTO1) - tmp &= ~B43_PHY_ANTDWELL_AUTODIV1; - else - tmp |= B43_PHY_ANTDWELL_AUTODIV1; - b43_phy_write(dev, B43_PHY_ANTDWELL, tmp); - } - if (phy->rev < 3) - b43_phy_maskset(dev, B43_PHY_ANTDWELL, 0xFF00, 0x24); - else { - b43_phy_set(dev, B43_PHY_OFDM61, 0x10); - if (phy->rev == 3) { - b43_phy_write(dev, B43_PHY_CLIPPWRDOWNT, 0x1D); - b43_phy_write(dev, B43_PHY_ADIVRELATED, 8); - } else { - b43_phy_write(dev, B43_PHY_CLIPPWRDOWNT, 0x3A); - b43_phy_maskset(dev, B43_PHY_ADIVRELATED, 0xFF00, 8); - } - } - - b43_hf_write(dev, b43_hf_read(dev) | B43_HF_ANTDIVHELP); -} - -static void b43_aphy_op_adjust_txpower(struct b43_wldev *dev) -{//TODO -} - -static enum b43_txpwr_result b43_aphy_op_recalc_txpower(struct b43_wldev *dev, - bool ignore_tssi) -{//TODO - return B43_TXPWR_RES_DONE; -} - -static void b43_aphy_op_pwork_15sec(struct b43_wldev *dev) -{//TODO -} - -static void b43_aphy_op_pwork_60sec(struct b43_wldev *dev) -{//TODO -} - -static const struct b43_phy_operations b43_phyops_a = { - .allocate = b43_aphy_op_allocate, - .free = b43_aphy_op_free, - .prepare_structs = b43_aphy_op_prepare_structs, - .init = b43_aphy_op_init, - .phy_read = b43_aphy_op_read, - .phy_write = b43_aphy_op_write, - .radio_read = b43_aphy_op_radio_read, - .radio_write = b43_aphy_op_radio_write, - .supports_hwpctl = b43_aphy_op_supports_hwpctl, - .software_rfkill = b43_aphy_op_software_rfkill, - .switch_analog = b43_phyop_switch_analog_generic, - .switch_channel = b43_aphy_op_switch_channel, - .get_default_chan = b43_aphy_op_get_default_chan, - .set_rx_antenna = b43_aphy_op_set_rx_antenna, - .recalc_txpower = b43_aphy_op_recalc_txpower, - .adjust_txpower = b43_aphy_op_adjust_txpower, - .pwork_15sec = b43_aphy_op_pwork_15sec, - .pwork_60sec = b43_aphy_op_pwork_60sec, -}; diff --git a/drivers/net/wireless/broadcom/b43/phy_a.h b/drivers/net/wireless/broadcom/b43/phy_a.h index f7d0d92..0a92d01 100644 --- a/drivers/net/wireless/broadcom/b43/phy_a.h +++ b/drivers/net/wireless/broadcom/b43/phy_a.h @@ -101,26 +101,4 @@ u32 b43_ofdmtab_read32(struct b43_wldev *dev, u16 table, u16 offset); void b43_ofdmtab_write32(struct b43_wldev *dev, u16 table, u16 offset, u32 value); - -struct b43_phy_a { - /* Pointer to the table used to convert a - * TSSI value to dBm-Q5.2 */ - const s8 *tssi2dbm; - /* Target idle TSSI */ - int tgt_idle_tssi; - /* Current idle TSSI */ - int cur_idle_tssi;//FIXME value currently not set - - /* A-PHY TX Power control value. */ - u16 txpwr_offset; - - //TODO lots of missing stuff -}; - -/** - * b43_phy_inita - Lowlevel A-PHY init routine. - * This is _only_ used by the G-PHY code. - */ -void b43_phy_inita(struct b43_wldev *dev); - #endif /* LINUX_B43_PHY_A_H_ */ diff --git a/drivers/net/wireless/broadcom/b43/phy_common.h b/drivers/net/wireless/broadcom/b43/phy_common.h index 78d8652..ced054a 100644 --- a/drivers/net/wireless/broadcom/b43/phy_common.h +++ b/drivers/net/wireless/broadcom/b43/phy_common.h @@ -190,7 +190,6 @@ struct b43_phy_operations { void (*pwork_60sec)(struct b43_wldev *dev); }; -struct b43_phy_a; struct b43_phy_g; struct b43_phy_n; struct b43_phy_lp; @@ -210,8 +209,6 @@ struct b43_phy { #else union { #endif - /* A-PHY specific information */ - struct b43_phy_a *a; /* G-PHY specific information */ struct b43_phy_g *g; /* N-PHY specific information */ diff --git a/drivers/net/wireless/broadcom/b43/phy_g.c b/drivers/net/wireless/broadcom/b43/phy_g.c index 462310e..822dcaa 100644 --- a/drivers/net/wireless/broadcom/b43/phy_g.c +++ b/drivers/net/wireless/broadcom/b43/phy_g.c @@ -31,6 +31,7 @@ #include "phy_common.h" #include "lo.h" #include "main.h" +#include "wa.h" #include #include @@ -1987,6 +1988,25 @@ static void b43_phy_init_pctl(struct b43_wldev *dev) b43_shm_clear_tssi(dev); } +static void b43_phy_inita(struct b43_wldev *dev) +{ + struct b43_phy *phy = &dev->phy; + + might_sleep(); + + if (phy->rev >= 6) { + if (b43_phy_read(dev, B43_PHY_ENCORE) & B43_PHY_ENCORE_EN) + b43_phy_set(dev, B43_PHY_ENCORE, 0x0010); + else + b43_phy_mask(dev, B43_PHY_ENCORE, ~0x1010); + } + + b43_wa_all(dev); + + if (dev->dev->bus_sprom->boardflags_lo & B43_BFL_PACTRL) + b43_phy_maskset(dev, B43_PHY_OFDM(0x6E), 0xE000, 0x3CF); +} + static void b43_phy_initg(struct b43_wldev *dev) { struct b43_phy *phy = &dev->phy; @@ -2150,11 +2170,6 @@ static void default_radio_attenuation(struct b43_wldev *dev, } } - if (phy->type == B43_PHYTYPE_A) { - rf->att = 0x60; - return; - } - switch (phy->radio_ver) { case 0x2053: switch (phy->radio_rev) { -- cgit v0.10.2 From afdfdc481ea96ef1fac95ade8ec29c5e14333d2f Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sat, 4 Jun 2016 07:54:13 -0700 Subject: b43: Completely remove support for phy_a MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Per Michael Büsch: "All a-phy code is usused", so remove it all. Cc: Michael Büsch Signed-off-by: Guenter Roeck Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/b43/main.c b/drivers/net/wireless/broadcom/b43/main.c index 4ee5c58..6e5d909 100644 --- a/drivers/net/wireless/broadcom/b43/main.c +++ b/drivers/net/wireless/broadcom/b43/main.c @@ -3180,7 +3180,6 @@ static void b43_rate_memory_write(struct b43_wldev *dev, u16 rate, int is_ofdm) static void b43_rate_memory_init(struct b43_wldev *dev) { switch (dev->phy.type) { - case B43_PHYTYPE_A: case B43_PHYTYPE_G: case B43_PHYTYPE_N: case B43_PHYTYPE_LP: @@ -3194,8 +3193,6 @@ static void b43_rate_memory_init(struct b43_wldev *dev) b43_rate_memory_write(dev, B43_OFDM_RATE_36MB, 1); b43_rate_memory_write(dev, B43_OFDM_RATE_48MB, 1); b43_rate_memory_write(dev, B43_OFDM_RATE_54MB, 1); - if (dev->phy.type == B43_PHYTYPE_A) - break; /* fallthrough */ case B43_PHYTYPE_B: b43_rate_memory_write(dev, B43_CCK_RATE_1MB, 0); @@ -4604,14 +4601,6 @@ static int b43_phy_versioning(struct b43_wldev *dev) if (radio_manuf != 0x17F /* Broadcom */) unsupported = 1; switch (phy_type) { - case B43_PHYTYPE_A: - if (radio_id != 0x2060) - unsupported = 1; - if (radio_rev != 1) - unsupported = 1; - if (radio_manuf != 0x17F) - unsupported = 1; - break; case B43_PHYTYPE_B: if ((radio_id & 0xFFF0) != 0x2050) unsupported = 1; @@ -4766,10 +4755,7 @@ static void b43_set_synth_pu_delay(struct b43_wldev *dev, bool idle) u16 pu_delay; /* The time value is in microseconds. */ - if (dev->phy.type == B43_PHYTYPE_A) - pu_delay = 3700; - else - pu_delay = 1050; + pu_delay = 1050; if (b43_is_mode(dev->wl, NL80211_IFTYPE_ADHOC) || idle) pu_delay = 500; if ((dev->phy.radio_ver == 0x2050) && (dev->phy.radio_rev == 8)) @@ -4784,14 +4770,10 @@ static void b43_set_pretbtt(struct b43_wldev *dev) u16 pretbtt; /* The time value is in microseconds. */ - if (b43_is_mode(dev->wl, NL80211_IFTYPE_ADHOC)) { + if (b43_is_mode(dev->wl, NL80211_IFTYPE_ADHOC)) pretbtt = 2; - } else { - if (dev->phy.type == B43_PHYTYPE_A) - pretbtt = 120; - else - pretbtt = 250; - } + else + pretbtt = 250; b43_shm_write16(dev, B43_SHM_SHARED, B43_SHM_SH_PRETBTT, pretbtt); b43_write16(dev, B43_MMIO_TSF_CFP_PRETBTT, pretbtt); } @@ -5380,10 +5362,6 @@ static void b43_supported_bands(struct b43_wldev *dev, bool *have_2ghz_phy, /* As a fallback, try to guess using PHY type */ switch (dev->phy.type) { - case B43_PHYTYPE_A: - *have_2ghz_phy = false; - *have_5ghz_phy = true; - return; case B43_PHYTYPE_G: case B43_PHYTYPE_N: case B43_PHYTYPE_LP: @@ -5455,7 +5433,6 @@ static int b43_wireless_core_attach(struct b43_wldev *dev) /* We don't support 5 GHz on some PHYs yet */ if (have_5ghz_phy) { switch (dev->phy.type) { - case B43_PHYTYPE_A: case B43_PHYTYPE_G: case B43_PHYTYPE_LP: case B43_PHYTYPE_HT: diff --git a/drivers/net/wireless/broadcom/b43/wa.c b/drivers/net/wireless/broadcom/b43/wa.c index c218c08..0e96c08 100644 --- a/drivers/net/wireless/broadcom/b43/wa.c +++ b/drivers/net/wireless/broadcom/b43/wa.c @@ -30,33 +30,6 @@ #include "phy_common.h" #include "wa.h" -static void b43_wa_papd(struct b43_wldev *dev) -{ - u16 backup; - - backup = b43_ofdmtab_read16(dev, B43_OFDMTAB_PWRDYN2, 0); - b43_ofdmtab_write16(dev, B43_OFDMTAB_PWRDYN2, 0, 7); - b43_ofdmtab_write16(dev, B43_OFDMTAB_UNKNOWN_APHY, 0, 0); - b43_dummy_transmission(dev, true, true); - b43_ofdmtab_write16(dev, B43_OFDMTAB_PWRDYN2, 0, backup); -} - -static void b43_wa_auxclipthr(struct b43_wldev *dev) -{ - b43_phy_write(dev, B43_PHY_OFDM(0x8E), 0x3800); -} - -static void b43_wa_afcdac(struct b43_wldev *dev) -{ - b43_phy_write(dev, 0x0035, 0x03FF); - b43_phy_write(dev, 0x0036, 0x0400); -} - -static void b43_wa_txdc_offset(struct b43_wldev *dev) -{ - b43_ofdmtab_write16(dev, B43_OFDMTAB_DC, 0, 0x0051); -} - void b43_wa_initgains(struct b43_wldev *dev) { struct b43_phy *phy = &dev->phy; @@ -81,41 +54,6 @@ void b43_wa_initgains(struct b43_wldev *dev) b43_phy_write(dev, 0x00BA, 0x3ED5); } -static void b43_wa_divider(struct b43_wldev *dev) -{ - b43_phy_mask(dev, 0x002B, ~0x0100); - b43_phy_write(dev, 0x008E, 0x58C1); -} - -static void b43_wa_gt(struct b43_wldev *dev) /* Gain table. */ -{ - if (dev->phy.rev <= 2) { - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN2, 0, 15); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN2, 1, 31); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN2, 2, 42); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN2, 3, 48); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN2, 4, 58); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 0, 19); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 1, 19); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 2, 19); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 3, 19); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 4, 21); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 5, 21); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 6, 25); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN1, 0, 3); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN1, 1, 3); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN1, 2, 7); - } else { - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 0, 19); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 1, 19); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 2, 19); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 3, 19); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 4, 21); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 5, 21); - b43_ofdmtab_write16(dev, B43_OFDMTAB_GAIN0, 6, 25); - } -} - static void b43_wa_rssi_lt(struct b43_wldev *dev) /* RSSI lookup table */ { int i; @@ -133,15 +71,11 @@ static void b43_wa_rssi_lt(struct b43_wldev *dev) /* RSSI lookup table */ static void b43_wa_analog(struct b43_wldev *dev) { - struct b43_phy *phy = &dev->phy; u16 ofdmrev; ofdmrev = b43_phy_read(dev, B43_PHY_VERSION_OFDM) & B43_PHYVER_VERSION; if (ofdmrev > 2) { - if (phy->type == B43_PHYTYPE_A) - b43_phy_write(dev, B43_PHY_PWRDOWN, 0x1808); - else - b43_phy_write(dev, B43_PHY_PWRDOWN, 0x1000); + b43_phy_write(dev, B43_PHY_PWRDOWN, 0x1000); } else { b43_ofdmtab_write16(dev, B43_OFDMTAB_DAC, 3, 0x1044); b43_ofdmtab_write16(dev, B43_OFDMTAB_DAC, 4, 0x7201); @@ -149,26 +83,13 @@ static void b43_wa_analog(struct b43_wldev *dev) } } -static void b43_wa_dac(struct b43_wldev *dev) -{ - if (dev->phy.analog == 1) - b43_ofdmtab_write16(dev, B43_OFDMTAB_DAC, 1, - (b43_ofdmtab_read16(dev, B43_OFDMTAB_DAC, 1) & ~0x0034) | 0x0008); - else - b43_ofdmtab_write16(dev, B43_OFDMTAB_DAC, 1, - (b43_ofdmtab_read16(dev, B43_OFDMTAB_DAC, 1) & ~0x0078) | 0x0010); -} - static void b43_wa_fft(struct b43_wldev *dev) /* Fine frequency table */ { int i; - if (dev->phy.type == B43_PHYTYPE_A) - for (i = 0; i < B43_TAB_FINEFREQA_SIZE; i++) - b43_ofdmtab_write16(dev, B43_OFDMTAB_DACRFPABB, i, b43_tab_finefreqa[i]); - else - for (i = 0; i < B43_TAB_FINEFREQG_SIZE; i++) - b43_ofdmtab_write16(dev, B43_OFDMTAB_DACRFPABB, i, b43_tab_finefreqg[i]); + for (i = 0; i < B43_TAB_FINEFREQG_SIZE; i++) + b43_ofdmtab_write16(dev, B43_OFDMTAB_DACRFPABB, i, + b43_tab_finefreqg[i]); } static void b43_wa_nft(struct b43_wldev *dev) /* Noise figure table */ @@ -176,21 +97,14 @@ static void b43_wa_nft(struct b43_wldev *dev) /* Noise figure table */ struct b43_phy *phy = &dev->phy; int i; - if (phy->type == B43_PHYTYPE_A) { - if (phy->rev == 2) - for (i = 0; i < B43_TAB_NOISEA2_SIZE; i++) - b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC2, i, b43_tab_noisea2[i]); - else - for (i = 0; i < B43_TAB_NOISEA3_SIZE; i++) - b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC2, i, b43_tab_noisea3[i]); - } else { - if (phy->rev == 1) - for (i = 0; i < B43_TAB_NOISEG1_SIZE; i++) - b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC2, i, b43_tab_noiseg1[i]); - else - for (i = 0; i < B43_TAB_NOISEG2_SIZE; i++) - b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC2, i, b43_tab_noiseg2[i]); - } + if (phy->rev == 1) + for (i = 0; i < B43_TAB_NOISEG1_SIZE; i++) + b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC2, i, + b43_tab_noiseg1[i]); + else + for (i = 0; i < B43_TAB_NOISEG2_SIZE; i++) + b43_ofdmtab_write16(dev, B43_OFDMTAB_AGC2, i, + b43_tab_noiseg2[i]); } static void b43_wa_rt(struct b43_wldev *dev) /* Rotor table */ @@ -201,14 +115,6 @@ static void b43_wa_rt(struct b43_wldev *dev) /* Rotor table */ b43_ofdmtab_write32(dev, B43_OFDMTAB_ROTOR, i, b43_tab_rotor[i]); } -static void b43_write_null_nst(struct b43_wldev *dev) -{ - int i; - - for (i = 0; i < B43_TAB_NOISESCALE_SIZE; i++) - b43_ofdmtab_write16(dev, B43_OFDMTAB_NOISESCALE, i, 0); -} - static void b43_write_nst(struct b43_wldev *dev, const u16 *nst) { int i; @@ -221,24 +127,13 @@ static void b43_wa_nst(struct b43_wldev *dev) /* Noise scale table */ { struct b43_phy *phy = &dev->phy; - if (phy->type == B43_PHYTYPE_A) { - if (phy->rev <= 1) - b43_write_null_nst(dev); - else if (phy->rev == 2) - b43_write_nst(dev, b43_tab_noisescalea2); - else if (phy->rev == 3) - b43_write_nst(dev, b43_tab_noisescalea3); - else + if (phy->rev >= 6) { + if (b43_phy_read(dev, B43_PHY_ENCORE) & B43_PHY_ENCORE_EN) b43_write_nst(dev, b43_tab_noisescaleg3); + else + b43_write_nst(dev, b43_tab_noisescaleg2); } else { - if (phy->rev >= 6) { - if (b43_phy_read(dev, B43_PHY_ENCORE) & B43_PHY_ENCORE_EN) - b43_write_nst(dev, b43_tab_noisescaleg3); - else - b43_write_nst(dev, b43_tab_noisescaleg2); - } else { - b43_write_nst(dev, b43_tab_noisescaleg1); - } + b43_write_nst(dev, b43_tab_noisescaleg1); } } @@ -251,41 +146,13 @@ static void b43_wa_art(struct b43_wldev *dev) /* ADV retard table */ i, b43_tab_retard[i]); } -static void b43_wa_txlna_gain(struct b43_wldev *dev) -{ - b43_ofdmtab_write16(dev, B43_OFDMTAB_DC, 13, 0x0000); -} - -static void b43_wa_crs_reset(struct b43_wldev *dev) -{ - b43_phy_write(dev, 0x002C, 0x0064); -} - -static void b43_wa_2060txlna_gain(struct b43_wldev *dev) -{ - b43_hf_write(dev, b43_hf_read(dev) | - B43_HF_2060W); -} - -static void b43_wa_lms(struct b43_wldev *dev) -{ - b43_phy_maskset(dev, 0x0055, 0xFFC0, 0x0004); -} - -static void b43_wa_mixedsignal(struct b43_wldev *dev) -{ - b43_ofdmtab_write16(dev, B43_OFDMTAB_DAC, 1, 3); -} - static void b43_wa_msst(struct b43_wldev *dev) /* Min sigma square table */ { struct b43_phy *phy = &dev->phy; int i; const u16 *tab; - if (phy->type == B43_PHYTYPE_A) { - tab = b43_tab_sigmasqr1; - } else if (phy->type == B43_PHYTYPE_G) { + if (phy->type == B43_PHYTYPE_G) { tab = b43_tab_sigmasqr2; } else { B43_WARN_ON(1); @@ -298,13 +165,6 @@ static void b43_wa_msst(struct b43_wldev *dev) /* Min sigma square table */ } } -static void b43_wa_iqadc(struct b43_wldev *dev) -{ - if (dev->phy.analog == 4) - b43_ofdmtab_write16(dev, B43_OFDMTAB_DAC, 0, - b43_ofdmtab_read16(dev, B43_OFDMTAB_DAC, 0) & ~0xF000); -} - static void b43_wa_crs_ed(struct b43_wldev *dev) { struct b43_phy *phy = &dev->phy; @@ -450,38 +310,6 @@ static void b43_wa_cpll_nonpilot(struct b43_wldev *dev) b43_ofdmtab_write16(dev, B43_OFDMTAB_UNKNOWN_11, 1, 0); } -static void b43_wa_rssi_adc(struct b43_wldev *dev) -{ - if (dev->phy.analog == 4) - b43_phy_write(dev, 0x00DC, 0x7454); -} - -static void b43_wa_boards_a(struct b43_wldev *dev) -{ - if (dev->dev->board_vendor == SSB_BOARDVENDOR_BCM && - dev->dev->board_type == SSB_BOARD_BU4306 && - dev->dev->board_rev < 0x30) { - b43_phy_write(dev, 0x0010, 0xE000); - b43_phy_write(dev, 0x0013, 0x0140); - b43_phy_write(dev, 0x0014, 0x0280); - } else { - if (dev->dev->board_type == SSB_BOARD_MP4318 && - dev->dev->board_rev < 0x20) { - b43_phy_write(dev, 0x0013, 0x0210); - b43_phy_write(dev, 0x0014, 0x0840); - } else { - b43_phy_write(dev, 0x0013, 0x0140); - b43_phy_write(dev, 0x0014, 0x0280); - } - if (dev->phy.rev <= 4) - b43_phy_write(dev, 0x0010, 0xE000); - else - b43_phy_write(dev, 0x0010, 0x2000); - b43_ofdmtab_write16(dev, B43_OFDMTAB_DC, 1, 0x0039); - b43_ofdmtab_write16(dev, B43_OFDMTAB_UNKNOWN_APHY, 7, 0x0040); - } -} - static void b43_wa_boards_g(struct b43_wldev *dev) { struct ssb_sprom *sprom = dev->dev->bus_sprom; @@ -518,80 +346,7 @@ void b43_wa_all(struct b43_wldev *dev) { struct b43_phy *phy = &dev->phy; - if (phy->type == B43_PHYTYPE_A) { - switch (phy->rev) { - case 2: - b43_wa_papd(dev); - b43_wa_auxclipthr(dev); - b43_wa_afcdac(dev); - b43_wa_txdc_offset(dev); - b43_wa_initgains(dev); - b43_wa_divider(dev); - b43_wa_gt(dev); - b43_wa_rssi_lt(dev); - b43_wa_analog(dev); - b43_wa_dac(dev); - b43_wa_fft(dev); - b43_wa_nft(dev); - b43_wa_rt(dev); - b43_wa_nst(dev); - b43_wa_art(dev); - b43_wa_txlna_gain(dev); - b43_wa_crs_reset(dev); - b43_wa_2060txlna_gain(dev); - b43_wa_lms(dev); - break; - case 3: - b43_wa_papd(dev); - b43_wa_mixedsignal(dev); - b43_wa_rssi_lt(dev); - b43_wa_txdc_offset(dev); - b43_wa_initgains(dev); - b43_wa_dac(dev); - b43_wa_nft(dev); - b43_wa_nst(dev); - b43_wa_msst(dev); - b43_wa_analog(dev); - b43_wa_gt(dev); - b43_wa_txpuoff_rxpuon(dev); - b43_wa_txlna_gain(dev); - break; - case 5: - b43_wa_iqadc(dev); - case 6: - b43_wa_papd(dev); - b43_wa_rssi_lt(dev); - b43_wa_txdc_offset(dev); - b43_wa_initgains(dev); - b43_wa_dac(dev); - b43_wa_nft(dev); - b43_wa_nst(dev); - b43_wa_msst(dev); - b43_wa_analog(dev); - b43_wa_gt(dev); - b43_wa_txpuoff_rxpuon(dev); - b43_wa_txlna_gain(dev); - break; - case 7: - b43_wa_iqadc(dev); - b43_wa_papd(dev); - b43_wa_rssi_lt(dev); - b43_wa_txdc_offset(dev); - b43_wa_initgains(dev); - b43_wa_dac(dev); - b43_wa_nft(dev); - b43_wa_nst(dev); - b43_wa_msst(dev); - b43_wa_analog(dev); - b43_wa_gt(dev); - b43_wa_txpuoff_rxpuon(dev); - b43_wa_txlna_gain(dev); - b43_wa_rssi_adc(dev); - default: - B43_WARN_ON(1); - } - b43_wa_boards_a(dev); - } else if (phy->type == B43_PHYTYPE_G) { + if (phy->type == B43_PHYTYPE_G) { switch (phy->rev) { case 1://XXX review rev1 b43_wa_crs_ed(dev); diff --git a/drivers/net/wireless/broadcom/b43/xmit.c b/drivers/net/wireless/broadcom/b43/xmit.c index 7edbcdb..b068d5a 100644 --- a/drivers/net/wireless/broadcom/b43/xmit.c +++ b/drivers/net/wireless/broadcom/b43/xmit.c @@ -649,11 +649,7 @@ static s8 b43_rssinoise_postprocess(struct b43_wldev *dev, u8 in_rssi) struct b43_phy *phy = &dev->phy; s8 ret; - if (phy->type == B43_PHYTYPE_A) { - //TODO: Incomplete specs. - ret = 0; - } else - ret = b43_rssi_postprocess(dev, in_rssi, 0, 1, 1); + ret = b43_rssi_postprocess(dev, in_rssi, 0, 1, 1); return ret; } @@ -670,7 +666,6 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr) u16 uninitialized_var(chanstat), uninitialized_var(mactime); u32 uninitialized_var(macstat); u16 chanid; - u16 phytype; int padding, rate_idx; memset(&status, 0, sizeof(status)); @@ -691,7 +686,6 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr) chanstat = le16_to_cpu(rxhdr->format_351.channel); break; } - phytype = chanstat & B43_RX_CHAN_PHYTYPE; if (unlikely(macstat & B43_RX_MAC_FCSERR)) { dev->wl->ieee_stats.dot11FCSErrorCount++; @@ -762,7 +756,6 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr) else status.signal = max(rxhdr->power0, rxhdr->power1); break; - case B43_PHYTYPE_A: case B43_PHYTYPE_B: case B43_PHYTYPE_G: case B43_PHYTYPE_LP: @@ -809,14 +802,6 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr) chanid = (chanstat & B43_RX_CHAN_ID) >> B43_RX_CHAN_ID_SHIFT; switch (chanstat & B43_RX_CHAN_PHYTYPE) { - case B43_PHYTYPE_A: - status.band = NL80211_BAND_5GHZ; - B43_WARN_ON(1); - /* FIXME: We don't really know which value the "chanid" contains. - * So the following assignment might be wrong. */ - status.freq = - ieee80211_channel_to_frequency(chanid, status.band); - break; case B43_PHYTYPE_G: status.band = NL80211_BAND_2GHZ; /* Somewhere between 478.104 and 508.1084 firmware for G-PHY -- cgit v0.10.2 From 26072330df2ef37676ae396ca48668567336cdb8 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Mon, 6 Jun 2016 23:03:55 +0200 Subject: brcmfmac: drop unused pm_block vif attribute MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This attribute was added 3 years ago by commit 3eacf866559c ("brcmfmac: introduce brcmf_cfg80211_vif structure") but it remains unused since then. It seems we can safely drop it. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index 9651a7c..502b0a0 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -601,7 +601,7 @@ struct wireless_dev *brcmf_ap_add_vif(struct wiphy *wiphy, const char *name, brcmf_dbg(INFO, "Adding vif \"%s\"\n", name); - vif = brcmf_alloc_vif(cfg, NL80211_IFTYPE_AP, false); + vif = brcmf_alloc_vif(cfg, NL80211_IFTYPE_AP); if (IS_ERR(vif)) return (struct wireless_dev *)vif; @@ -5158,8 +5158,7 @@ static struct cfg80211_ops brcmf_cfg80211_ops = { }; struct brcmf_cfg80211_vif *brcmf_alloc_vif(struct brcmf_cfg80211_info *cfg, - enum nl80211_iftype type, - bool pm_block) + enum nl80211_iftype type) { struct brcmf_cfg80211_vif *vif_walk; struct brcmf_cfg80211_vif *vif; @@ -5174,8 +5173,6 @@ struct brcmf_cfg80211_vif *brcmf_alloc_vif(struct brcmf_cfg80211_info *cfg, vif->wdev.wiphy = cfg->wiphy; vif->wdev.iftype = type; - vif->pm_block = pm_block; - brcmf_init_prof(&vif->profile); if (type == NL80211_IFTYPE_AP) { @@ -6817,7 +6814,7 @@ struct brcmf_cfg80211_info *brcmf_cfg80211_attach(struct brcmf_pub *drvr, init_vif_event(&cfg->vif_event); INIT_LIST_HEAD(&cfg->vif_list); - vif = brcmf_alloc_vif(cfg, NL80211_IFTYPE_STATION, false); + vif = brcmf_alloc_vif(cfg, NL80211_IFTYPE_STATION); if (IS_ERR(vif)) goto wiphy_out; diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h index 95e35bc..c6f9986 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h @@ -167,7 +167,6 @@ struct vif_saved_ie { * @wdev: wireless device. * @profile: profile information. * @sme_state: SME state using enum brcmf_vif_status bits. - * @pm_block: power-management blocked. * @list: linked list. * @mgmt_rx_reg: registered rx mgmt frame types. * @mbss: Multiple BSS type, set if not first AP (not relevant for P2P). @@ -177,7 +176,6 @@ struct brcmf_cfg80211_vif { struct wireless_dev wdev; struct brcmf_cfg80211_profile profile; unsigned long sme_state; - bool pm_block; struct vif_saved_ie saved_ie; struct list_head list; u16 mgmt_rx_reg; @@ -388,8 +386,7 @@ s32 brcmf_cfg80211_down(struct net_device *ndev); enum nl80211_iftype brcmf_cfg80211_get_iftype(struct brcmf_if *ifp); struct brcmf_cfg80211_vif *brcmf_alloc_vif(struct brcmf_cfg80211_info *cfg, - enum nl80211_iftype type, - bool pm_block); + enum nl80211_iftype type); void brcmf_free_vif(struct brcmf_cfg80211_vif *vif); s32 brcmf_vif_set_mgmt_ie(struct brcmf_cfg80211_vif *vif, s32 pktflag, diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c index b7d54e9..f38a821 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c @@ -2074,8 +2074,7 @@ static struct wireless_dev *brcmf_p2p_create_p2pdev(struct brcmf_p2p_info *p2p, if (p2p->bss_idx[P2PAPI_BSSCFG_DEVICE].vif) return ERR_PTR(-ENOSPC); - p2p_vif = brcmf_alloc_vif(p2p->cfg, NL80211_IFTYPE_P2P_DEVICE, - false); + p2p_vif = brcmf_alloc_vif(p2p->cfg, NL80211_IFTYPE_P2P_DEVICE); if (IS_ERR(p2p_vif)) { brcmf_err("could not create discovery vif\n"); return (struct wireless_dev *)p2p_vif; @@ -2175,7 +2174,7 @@ struct wireless_dev *brcmf_p2p_add_vif(struct wiphy *wiphy, const char *name, return ERR_PTR(-EOPNOTSUPP); } - vif = brcmf_alloc_vif(cfg, type, false); + vif = brcmf_alloc_vif(cfg, type); if (IS_ERR(vif)) return (struct wireless_dev *)vif; brcmf_cfg80211_arm_vif_event(cfg, vif); -- cgit v0.10.2 From 29477269a27daf8859217ced08f853be04e1bac7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Tue, 7 Jun 2016 08:20:21 +0200 Subject: brcmfmac: include required headers in cfg80211.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Without this including cfg80211.h in a wrong order could result in: drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h:122:24: error: array type has incomplete element type struct brcmf_wsec_key key[BRCMF_MAX_DEFAULT_KEYS]; ^ drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h:291:24: error: field ‘p2p’ has incomplete type struct brcmf_p2p_info p2p; ^ drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h:297:27: error: field ‘pmk_list’ has incomplete type struct brcmf_pmk_list_le pmk_list; ^ drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h:317:28: error: field ‘assoclist’ has incomplete type struct brcmf_assoclist_le assoclist; Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h index c6f9986..04bfc7e 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h @@ -20,6 +20,9 @@ /* for brcmu_d11inf */ #include +#include "fwil_types.h" +#include "p2p.h" + #define WL_NUM_SCAN_MAX 10 #define WL_TLV_INFO_MAX 1024 #define WL_BSS_INFO_MAX 2048 -- cgit v0.10.2 From 5fb384b066d7c320832c4541658e5c655c590ac5 Mon Sep 17 00:00:00 2001 From: Fabien Siron Date: Wed, 15 Jun 2016 15:37:49 +0000 Subject: netlink: Add comment to warn about deprecated netlink rings attribute request Signed-off-by: Fabien Siron Signed-off-by: David S. Miller diff --git a/include/uapi/linux/netlink_diag.h b/include/uapi/linux/netlink_diag.h index d793993..76b4d87 100644 --- a/include/uapi/linux/netlink_diag.h +++ b/include/uapi/linux/netlink_diag.h @@ -49,6 +49,7 @@ enum { #define NDIAG_SHOW_MEMINFO 0x00000001 /* show memory info of a socket */ #define NDIAG_SHOW_GROUPS 0x00000002 /* show groups of a netlink socket */ #ifndef __KERNEL__ +/* deprecated since 4.6 */ #define NDIAG_SHOW_RING_CFG 0x00000004 /* show ring configuration */ #endif -- cgit v0.10.2 From 141ddefce7c807c5e34b67be50b4a789a51f4a56 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Thu, 16 Jun 2016 01:15:06 +0800 Subject: sctp: change sk state to CLOSED instead of CLOSING in sctp_sock_migrate Commit d46e416c11c8 ("sctp: sctp should change socket state when shutdown is received") may set sk_state CLOSING in sctp_sock_migrate, but inet_accept doesn't allow the sk_state other than ESTABLISHED/ CLOSED for sctp. So we will change sk_state to CLOSED, instead of CLOSING, as actually sk is closed already there. Fixes: d46e416c11c8 ("sctp: sctp should change socket state when shutdown is received") Reported-by: Ye Xiaolong Signed-off-by: Xin Long Signed-off-by: David S. Miller diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 6cae4c6..cdabbd8 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -7568,7 +7568,7 @@ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, * is called, set RCV_SHUTDOWN flag. */ if (sctp_state(assoc, CLOSED) && sctp_style(newsk, TCP)) { - newsk->sk_state = SCTP_SS_CLOSING; + newsk->sk_state = SCTP_SS_CLOSED; newsk->sk_shutdown |= RCV_SHUTDOWN; } else { newsk->sk_state = SCTP_SS_ESTABLISHED; -- cgit v0.10.2 From cecbc5563a02289164fa6379130243cbe08b2dd6 Mon Sep 17 00:00:00 2001 From: Vincent Palatin Date: Wed, 15 Jun 2016 11:32:21 -0700 Subject: net: stmmac: allow to split suspend/resume from init/exit callbacks Let the stmmac platform drivers provide dedicated suspend and resume callbacks rather than always re-using the init and exits callbacks. If the driver does not provide the suspend or resume callback, we fall back to the old behavior trying to use exit or init. This allows a specific platform to perform only a partial power-down on suspend if Wake-on-Lan is enabled but always perform the full shutdown sequence if the module is unloaded. Signed-off-by: Vincent Palatin Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 409db91..a96714d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -411,7 +411,9 @@ static int stmmac_pltfr_suspend(struct device *dev) struct platform_device *pdev = to_platform_device(dev); ret = stmmac_suspend(dev); - if (priv->plat->exit) + if (priv->plat->suspend) + priv->plat->suspend(pdev, priv->plat->bsp_priv); + else if (priv->plat->exit) priv->plat->exit(pdev, priv->plat->bsp_priv); return ret; @@ -430,7 +432,9 @@ static int stmmac_pltfr_resume(struct device *dev) struct stmmac_priv *priv = netdev_priv(ndev); struct platform_device *pdev = to_platform_device(dev); - if (priv->plat->init) + if (priv->plat->resume) + priv->plat->resume(pdev, priv->plat->bsp_priv); + else if (priv->plat->init) priv->plat->init(pdev, priv->plat->bsp_priv); return stmmac_resume(dev); diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index ffdaca9..0507dbf 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -135,6 +135,8 @@ struct plat_stmmacenet_data { void (*bus_setup)(void __iomem *ioaddr); int (*init)(struct platform_device *pdev, void *priv); void (*exit)(struct platform_device *pdev, void *priv); + void (*suspend)(struct platform_device *pdev, void *priv); + void (*resume)(struct platform_device *pdev, void *priv); void *bsp_priv; struct stmmac_axi *axi; int has_gmac4; -- cgit v0.10.2 From 229666c14c75ae6c5016c49d8806438e843c6807 Mon Sep 17 00:00:00 2001 From: Vincent Palatin Date: Wed, 15 Jun 2016 11:32:22 -0700 Subject: net: stmmac: dwmac-rk: keep the PHY up for WoL When suspending the machine, do not shutdown the external PHY by cutting its regulator in the mac platform driver suspend code if Wake-on-Lan is enabled, else it cannot wake us up. In order to do this, split the suspend/resume callbacks from the init/exit callbacks, so we can condition the power-down on the lack of need to wake-up from the LAN but do it unconditionally when unloading the module. Signed-off-by: Vincent Palatin Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 0cd3ecf..63c2e4f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -46,6 +46,7 @@ struct rk_priv_data { struct platform_device *pdev; int phy_iface; struct regulator *regulator; + bool suspended; const struct rk_gmac_ops *ops; bool clk_enabled; @@ -529,9 +530,8 @@ static struct rk_priv_data *rk_gmac_setup(struct platform_device *pdev, return bsp_priv; } -static int rk_gmac_init(struct platform_device *pdev, void *priv) +static int rk_gmac_powerup(struct rk_priv_data *bsp_priv) { - struct rk_priv_data *bsp_priv = priv; int ret; ret = phy_power_on(bsp_priv, true); @@ -545,14 +545,50 @@ static int rk_gmac_init(struct platform_device *pdev, void *priv) return 0; } -static void rk_gmac_exit(struct platform_device *pdev, void *priv) +static void rk_gmac_powerdown(struct rk_priv_data *gmac) { - struct rk_priv_data *gmac = priv; - phy_power_on(gmac, false); gmac_clk_enable(gmac, false); } +static int rk_gmac_init(struct platform_device *pdev, void *priv) +{ + struct rk_priv_data *bsp_priv = priv; + + return rk_gmac_powerup(bsp_priv); +} + +static void rk_gmac_exit(struct platform_device *pdev, void *priv) +{ + struct rk_priv_data *bsp_priv = priv; + + rk_gmac_powerdown(bsp_priv); +} + +static void rk_gmac_suspend(struct platform_device *pdev, void *priv) +{ + struct rk_priv_data *bsp_priv = priv; + + /* Keep the PHY up if we use Wake-on-Lan. */ + if (device_may_wakeup(&pdev->dev)) + return; + + rk_gmac_powerdown(bsp_priv); + bsp_priv->suspended = true; +} + +static void rk_gmac_resume(struct platform_device *pdev, void *priv) +{ + struct rk_priv_data *bsp_priv = priv; + + /* The PHY was up for Wake-on-Lan. */ + if (!bsp_priv->suspended) + return; + + rk_gmac_powerup(bsp_priv); + bsp_priv->suspended = false; +} + static void rk_fix_speed(void *priv, unsigned int speed) { struct rk_priv_data *bsp_priv = priv; @@ -591,6 +627,8 @@ static int rk_gmac_probe(struct platform_device *pdev) plat_dat->init = rk_gmac_init; plat_dat->exit = rk_gmac_exit; plat_dat->fix_mac_speed = rk_fix_speed; + plat_dat->suspend = rk_gmac_suspend; + plat_dat->resume = rk_gmac_resume; plat_dat->bsp_priv = rk_gmac_setup(pdev, data); if (IS_ERR(plat_dat->bsp_priv)) -- cgit v0.10.2 From d5bfbeb80973010dc87c82dc56f66e99c3e8c4da Mon Sep 17 00:00:00 2001 From: Vincent Palatin Date: Wed, 15 Jun 2016 11:32:23 -0700 Subject: ARM: dts: rockchip: add interrupt for Wake-on-Lan on RK3288 In order to use Wake-on-Lan on RK3288 integrated MAC, we need to wake-up the CPU on the PMT interrupt when the MAC and the PHY are in low power mode. Adding the interrupt declaration. Signed-off-by: Vincent Palatin Signed-off-by: David S. Miller diff --git a/arch/arm/boot/dts/rk3288.dtsi b/arch/arm/boot/dts/rk3288.dtsi index 3b44ef3..3ebee53 100644 --- a/arch/arm/boot/dts/rk3288.dtsi +++ b/arch/arm/boot/dts/rk3288.dtsi @@ -539,8 +539,9 @@ gmac: ethernet@ff290000 { compatible = "rockchip,rk3288-gmac"; reg = <0xff290000 0x10000>; - interrupts = ; - interrupt-names = "macirq"; + interrupts = , + ; + interrupt-names = "macirq", "eth_wake_irq"; rockchip,grf = <&grf>; clocks = <&cru SCLK_MAC>, <&cru SCLK_MAC_RX>, <&cru SCLK_MAC_TX>, -- cgit v0.10.2 From de0eabf86d005ab6b4fd5a5c254dcfe230d88a20 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 16 Jun 2016 00:12:48 +0200 Subject: net: ethernet: ax88796: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/8390/ax88796.c b/drivers/net/ethernet/8390/ax88796.c index c89b9ae..b446977 100644 --- a/drivers/net/ethernet/8390/ax88796.c +++ b/drivers/net/ethernet/8390/ax88796.c @@ -84,7 +84,6 @@ static u32 ax_msg_enable; struct ax_device { struct mii_bus *mii_bus; struct mdiobb_ctrl bb_ctrl; - struct phy_device *phy_dev; void __iomem *addr_memr; u8 reg_memr; int link; @@ -320,7 +319,7 @@ static void ax_block_output(struct net_device *dev, int count, static void ax_handle_link_change(struct net_device *dev) { struct ax_device *ax = to_ax_dev(dev); - struct phy_device *phy_dev = ax->phy_dev; + struct phy_device *phy_dev = dev->phydev; int status_change = 0; if (phy_dev->link && ((ax->speed != phy_dev->speed) || @@ -369,8 +368,6 @@ static int ax_mii_probe(struct net_device *dev) phy_dev->supported &= PHY_BASIC_FEATURES; phy_dev->advertising = phy_dev->supported; - ax->phy_dev = phy_dev; - netdev_info(dev, "PHY driver [%s] (mii_bus:phy_addr=%s, irq=%d)\n", phy_dev->drv->name, phydev_name(phy_dev), phy_dev->irq); @@ -410,7 +407,7 @@ static int ax_open(struct net_device *dev) ret = ax_mii_probe(dev); if (ret) goto failed_mii_probe; - phy_start(ax->phy_dev); + phy_start(dev->phydev); ret = ax_ei_open(dev); if (ret) @@ -421,7 +418,7 @@ static int ax_open(struct net_device *dev) return 0; failed_ax_ei_open: - phy_disconnect(ax->phy_dev); + phy_disconnect(dev->phydev); failed_mii_probe: ax_phy_switch(dev, 0); free_irq(dev->irq, dev); @@ -442,7 +439,7 @@ static int ax_close(struct net_device *dev) /* turn the phy off */ ax_phy_switch(dev, 0); - phy_disconnect(ax->phy_dev); + phy_disconnect(dev->phydev); free_irq(dev->irq, dev); return 0; @@ -450,8 +447,7 @@ static int ax_close(struct net_device *dev) static int ax_ioctl(struct net_device *dev, struct ifreq *req, int cmd) { - struct ax_device *ax = to_ax_dev(dev); - struct phy_device *phy_dev = ax->phy_dev; + struct phy_device *phy_dev = dev->phydev; if (!netif_running(dev)) return -EINVAL; @@ -476,8 +472,7 @@ static void ax_get_drvinfo(struct net_device *dev, static int ax_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct ax_device *ax = to_ax_dev(dev); - struct phy_device *phy_dev = ax->phy_dev; + struct phy_device *phy_dev = dev->phydev; if (!phy_dev) return -ENODEV; @@ -487,8 +482,7 @@ static int ax_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) static int ax_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct ax_device *ax = to_ax_dev(dev); - struct phy_device *phy_dev = ax->phy_dev; + struct phy_device *phy_dev = dev->phydev; if (!phy_dev) return -ENODEV; -- cgit v0.10.2 From d21cfb375eb07ce294ac29a0a51e45413fc8506d Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 16 Jun 2016 00:12:49 +0200 Subject: net: ethernet: ax88796: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/8390/ax88796.c b/drivers/net/ethernet/8390/ax88796.c index b446977..5698f53 100644 --- a/drivers/net/ethernet/8390/ax88796.c +++ b/drivers/net/ethernet/8390/ax88796.c @@ -470,26 +470,6 @@ static void ax_get_drvinfo(struct net_device *dev, strlcpy(info->bus_info, pdev->name, sizeof(info->bus_info)); } -static int ax_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phy_dev = dev->phydev; - - if (!phy_dev) - return -ENODEV; - - return phy_ethtool_gset(phy_dev, cmd); -} - -static int ax_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phy_dev = dev->phydev; - - if (!phy_dev) - return -ENODEV; - - return phy_ethtool_sset(phy_dev, cmd); -} - static u32 ax_get_msglevel(struct net_device *dev) { struct ei_device *ei_local = netdev_priv(dev); @@ -506,12 +486,12 @@ static void ax_set_msglevel(struct net_device *dev, u32 v) static const struct ethtool_ops ax_ethtool_ops = { .get_drvinfo = ax_get_drvinfo, - .get_settings = ax_get_settings, - .set_settings = ax_set_settings, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, .get_msglevel = ax_get_msglevel, .set_msglevel = ax_set_msglevel, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; #ifdef CONFIG_AX88796_93CX6 -- cgit v0.10.2 From fae5617877f2e1040106385d94529f30780525e1 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Thu, 16 Jun 2016 14:08:29 +0800 Subject: r8152: modify the check of the flag of PHY_RESET in set_speed function In set_speed(), BMCR_RESET would be set when the flag of PHY_RESET is set. Use BMCR_RESET to replace testing the flag of PHY_RESET. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index 69d1bbfd..11178f9 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -2872,7 +2872,7 @@ static int rtl8152_set_speed(struct r8152 *tp, u8 autoneg, u16 speed, u8 duplex) bmcr = BMCR_ANENABLE | BMCR_ANRESTART; } - if (test_bit(PHY_RESET, &tp->flags)) + if (test_and_clear_bit(PHY_RESET, &tp->flags)) bmcr |= BMCR_RESET; if (tp->mii.supports_gmii) @@ -2881,7 +2881,7 @@ static int rtl8152_set_speed(struct r8152 *tp, u8 autoneg, u16 speed, u8 duplex) r8152_mdio_write(tp, MII_ADVERTISE, anar); r8152_mdio_write(tp, MII_BMCR, bmcr); - if (test_and_clear_bit(PHY_RESET, &tp->flags)) { + if (bmcr & BMCR_RESET) { int i; for (i = 0; i < 50; i++) { -- cgit v0.10.2 From 0d227a8672c83f2153a0eeeb5439e3b7185c3d9c Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 16 Jun 2016 17:09:09 +0900 Subject: mpls: allow routes on ipgre devices This appears to be necessary and sufficient to provide MPLS in GRE (RFC4023) support. This can be used by establishing an ipgre tunnel device and then routing MPLS over it. The following example will forward MPLS frames received with an outermost MPLS label 100 over tun1, a GRE tunnel. The forwarded packet will have the outermost MPLS LSE removed and two new LSEs added with labels 200 (outermost) and 300 (next). ip link add name tun1 type gre remote 10.0.99.193 local 10.0.99.192 ttl 225 ip link set up dev tun1 ip addr add 10.0.98.192/24 dev tun1 ip route sh echo 1 > /proc/sys/net/mpls/conf/eth0/input echo 101 > /proc/sys/net/mpls/platform_labels ip -f mpls route add 100 as 200/300 via inet 10.0.98.193 ip -f mpls route sh Also remove unnecessary braces. Reviewed-by: Dinan Gunawardena Signed-off-by: Simon Horman Acked-by: Robert Shearman Signed-off-by: David S. Miller diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 7a4aa34..e9beaa5 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -1009,9 +1009,10 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, unsigned int flags; if (event == NETDEV_REGISTER) { - /* For now just support ethernet devices */ - if ((dev->type == ARPHRD_ETHER) || - (dev->type == ARPHRD_LOOPBACK)) { + /* For now just support Ethernet and IPGRE devices */ + if (dev->type == ARPHRD_ETHER || + dev->type == ARPHRD_LOOPBACK || + dev->type == ARPHRD_IPGRE) { mdev = mpls_add_dev(dev); if (IS_ERR(mdev)) return notifier_from_errno(PTR_ERR(mdev)); -- cgit v0.10.2 From 55e7f6abe131529a786e13464515015064f19e7d Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 11:00:05 +0200 Subject: dsa: b53: fix big-endian register access The b53 dsa register access confusingly uses __raw register accessors when both the CPU and the device are big-endian, but it uses little- endian accessors when the same device is used from a little-endian CPU, which makes no sense. This uses normal accessors in device-endianess all the time, which will work in all four combinations of register and CPU endianess, and it will have the same barrier semantics in all cases. This also seems to take care of a (false positive) warning I'm getting: drivers/net/dsa/b53/b53_mmap.c: In function 'b53_mmap_read64': drivers/net/dsa/b53/b53_mmap.c:109:10: error: 'hi' may be used uninitialized in this function [-Werror=maybe-uninitialized] *val = ((u64)hi << 32) | lo; I originally planned to submit another patch for that warning and did this one as a preparation cleanup, but it does seem to be sufficient by itself. Signed-off-by: Arnd Bergmann Acked-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/b53/b53_mmap.c b/drivers/net/dsa/b53/b53_mmap.c index f115ee2..b70daf9 100644 --- a/drivers/net/dsa/b53/b53_mmap.c +++ b/drivers/net/dsa/b53/b53_mmap.c @@ -45,9 +45,8 @@ static int b53_mmap_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) if (WARN_ON(reg % 2)) return -EINVAL; - if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && - dev->pdata->big_endian) - *val = __raw_readw(regs + (page << 8) + reg); + if (dev->pdata && dev->pdata->big_endian) + *val = ioread16be(regs + (page << 8) + reg); else *val = readw(regs + (page << 8) + reg); @@ -61,9 +60,8 @@ static int b53_mmap_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) if (WARN_ON(reg % 4)) return -EINVAL; - if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && - dev->pdata->big_endian) - *val = __raw_readl(regs + (page << 8) + reg); + if (dev->pdata && dev->pdata->big_endian) + *val = ioread32be(regs + (page << 8) + reg); else *val = readl(regs + (page << 8) + reg); @@ -128,9 +126,8 @@ static int b53_mmap_write16(struct b53_device *dev, u8 page, u8 reg, if (WARN_ON(reg % 2)) return -EINVAL; - if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && - dev->pdata->big_endian) - __raw_writew(value, regs + (page << 8) + reg); + if (dev->pdata && dev->pdata->big_endian) + iowrite16be(value, regs + (page << 8) + reg); else writew(value, regs + (page << 8) + reg); @@ -145,9 +142,8 @@ static int b53_mmap_write32(struct b53_device *dev, u8 page, u8 reg, if (WARN_ON(reg % 4)) return -EINVAL; - if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && - dev->pdata->big_endian) - __raw_writel(value, regs + (page << 8) + reg); + if (dev->pdata && dev->pdata->big_endian) + iowrite32be(value, regs + (page << 8) + reg); else writel(value, regs + (page << 8) + reg); -- cgit v0.10.2 From 287debd6aac5b9f60113d45931a53da0cd4adc94 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 13:38:22 +0200 Subject: net: qlcnic: don't set unused function argument We get a warning for qlcnic_83xx_get_mac_address when building with "make W=1": drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c: In function 'qlcnic_83xx_get_mac_address': drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c:2156:8: error: parameter 'function' set but not used [-Werror=unused-but-set-parameter] Clearly this is harmless, but there is also no point for setting the variable, so we can simply remove the assignment. Signed-off-by: Arnd Bergmann Acked-by: Rajesh Borundia Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c index f9640d5ce..bdbcd2b 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c @@ -2159,7 +2159,6 @@ int qlcnic_83xx_get_mac_address(struct qlcnic_adapter *adapter, u8 *mac, struct qlcnic_cmd_args cmd; u32 mac_low, mac_high; - function = 0; err = qlcnic_alloc_mbx_args(&cmd, adapter, QLCNIC_CMD_MAC_ADDRESS); if (err) return err; -- cgit v0.10.2 From 40309d26549ec29cfc91d9ee5fc6f06083b97fe5 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 13:38:23 +0200 Subject: net: tlan: don't set unused function argument We get a warning for tlan_handle_tx_eoc when building with "make W=1" drivers/net/ethernet/ti/tlan.c: In function 'tlan_handle_tx_eoc': drivers/net/ethernet/ti/tlan.c:1647:59: error: parameter 'host_int' set but not used [-Werror=unused-but-set-parameter] static u32 tlan_handle_tx_eoc(struct net_device *dev, u16 host_int) This is harmless, but removing the unused assignment lets us avoid the warning with no downside. Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/tlan.c b/drivers/net/ethernet/ti/tlan.c index 5617033..ece0ea0 100644 --- a/drivers/net/ethernet/ti/tlan.c +++ b/drivers/net/ethernet/ti/tlan.c @@ -1651,7 +1651,6 @@ static u32 tlan_handle_tx_eoc(struct net_device *dev, u16 host_int) dma_addr_t head_list_phys; u32 ack = 1; - host_int = 0; if (priv->tlan_rev < 0x30) { TLAN_DBG(TLAN_DEBUG_TX, "TRANSMIT: handling TX EOC (Head=%d Tail=%d) -- IRQ\n", -- cgit v0.10.2 From 84d15ae57d9478efc755306fee5ee562e0fa40e5 Mon Sep 17 00:00:00 2001 From: Wei Tang Date: Thu, 16 Jun 2016 21:17:49 +0800 Subject: net: do not initialise statics to 0 This patch fixes the checkpatch.pl error to dev.c: ERROR: do not initialise statics to 0 Signed-off-by: Wei Tang Signed-off-by: David S. Miller diff --git a/net/core/dev.c b/net/core/dev.c index b148357..441657f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2422,7 +2422,7 @@ EXPORT_SYMBOL(__skb_tx_hash); static void skb_warn_bad_offload(const struct sk_buff *skb) { - static const netdev_features_t null_features = 0; + static const netdev_features_t null_features; struct net_device *dev = skb->dev; const char *name = ""; -- cgit v0.10.2 From be4da0e340ed2a17b1a55cbe81d6bc251710ff72 Mon Sep 17 00:00:00 2001 From: Wei Tang Date: Thu, 16 Jun 2016 21:30:12 +0800 Subject: net: the space is required after ',' The space is missing after ',', and this will introduce much more noise when checking patch around. Signed-off-by: Wei Tang Signed-off-by: David S. Miller diff --git a/net/core/dev.c b/net/core/dev.c index 441657f..d40593b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5911,7 +5911,7 @@ static void netdev_adjacent_add_links(struct net_device *dev) struct net *net = dev_net(dev); list_for_each_entry(iter, &dev->adj_list.upper, list) { - if (!net_eq(net,dev_net(iter->dev))) + if (!net_eq(net, dev_net(iter->dev))) continue; netdev_adjacent_sysfs_add(iter->dev, dev, &iter->dev->adj_list.lower); @@ -5920,7 +5920,7 @@ static void netdev_adjacent_add_links(struct net_device *dev) } list_for_each_entry(iter, &dev->adj_list.lower, list) { - if (!net_eq(net,dev_net(iter->dev))) + if (!net_eq(net, dev_net(iter->dev))) continue; netdev_adjacent_sysfs_add(iter->dev, dev, &iter->dev->adj_list.upper); @@ -5936,7 +5936,7 @@ static void netdev_adjacent_del_links(struct net_device *dev) struct net *net = dev_net(dev); list_for_each_entry(iter, &dev->adj_list.upper, list) { - if (!net_eq(net,dev_net(iter->dev))) + if (!net_eq(net, dev_net(iter->dev))) continue; netdev_adjacent_sysfs_del(iter->dev, dev->name, &iter->dev->adj_list.lower); @@ -5945,7 +5945,7 @@ static void netdev_adjacent_del_links(struct net_device *dev) } list_for_each_entry(iter, &dev->adj_list.lower, list) { - if (!net_eq(net,dev_net(iter->dev))) + if (!net_eq(net, dev_net(iter->dev))) continue; netdev_adjacent_sysfs_del(iter->dev, dev->name, &iter->dev->adj_list.upper); @@ -5961,7 +5961,7 @@ void netdev_adjacent_rename_links(struct net_device *dev, char *oldname) struct net *net = dev_net(dev); list_for_each_entry(iter, &dev->adj_list.upper, list) { - if (!net_eq(net,dev_net(iter->dev))) + if (!net_eq(net, dev_net(iter->dev))) continue; netdev_adjacent_sysfs_del(iter->dev, oldname, &iter->dev->adj_list.lower); @@ -5970,7 +5970,7 @@ void netdev_adjacent_rename_links(struct net_device *dev, char *oldname) } list_for_each_entry(iter, &dev->adj_list.lower, list) { - if (!net_eq(net,dev_net(iter->dev))) + if (!net_eq(net, dev_net(iter->dev))) continue; netdev_adjacent_sysfs_del(iter->dev, oldname, &iter->dev->adj_list.upper); -- cgit v0.10.2 From 60a25d00db94f71d905952b11c5825f2d0e6ec07 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 15:52:11 +0200 Subject: hamradio: baycom: fix old-style declaration Modern C standards expect the '__inline__' keyword to come before the return type in a declaration, and we get a warning for this with "make W=1": drivers/net/hamradio/baycom_par.c:159:1: error: '__inline__' is not at beginning of declaration [-Werror=old-style-declaration] For consistency with other drivers, I'm changing '__inline__' to 'inline' at the same time. Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller diff --git a/drivers/net/hamradio/baycom_par.c b/drivers/net/hamradio/baycom_par.c index acb6369..072cddc 100644 --- a/drivers/net/hamradio/baycom_par.c +++ b/drivers/net/hamradio/baycom_par.c @@ -156,7 +156,7 @@ struct baycom_state { /* --------------------------------------------------------------------- */ -static void __inline__ baycom_int_freq(struct baycom_state *bc) +static inline void baycom_int_freq(struct baycom_state *bc) { #ifdef BAYCOM_DEBUG unsigned long cur_jiffies = jiffies; @@ -192,7 +192,7 @@ static void __inline__ baycom_int_freq(struct baycom_state *bc) /* --------------------------------------------------------------------- */ -static __inline__ void par96_tx(struct net_device *dev, struct baycom_state *bc) +static inline void par96_tx(struct net_device *dev, struct baycom_state *bc) { int i; unsigned int data = hdlcdrv_getbits(&bc->hdrv); @@ -216,7 +216,7 @@ static __inline__ void par96_tx(struct net_device *dev, struct baycom_state *bc) /* --------------------------------------------------------------------- */ -static __inline__ void par96_rx(struct net_device *dev, struct baycom_state *bc) +static inline void par96_rx(struct net_device *dev, struct baycom_state *bc) { int i; unsigned int data, mask, mask2, descx; -- cgit v0.10.2 From ac565065a60a78fe67aa1d3bb0ddc5361d2587d4 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 15:52:12 +0200 Subject: isdn: eicon: fix old-style declarations Modern C standards expect the '__inline__' keyword to come before the return type in a declaration, and we get many warnings for this with "make W=1" because the eicon driver has this in a header file: eicon/divasmain.c:448:1: error: '__inline__' is not at beginning of declaration [-Werror=old-style-declaration] eicon/divasmain.c:453:1: error: '__inline__' is not at beginning of declaration [-Werror=old-style-declaration] eicon/divasmain.c:458:1: error: '__inline__' is not at beginning of declaration [-Werror=old-style-declaration] eicon/divasmain.c:463:1: error: '__inline__' is not at beginning of declaration [-Werror=old-style-declaration] eicon/divasmain.c:468:1: error: '__inline__' is not at beginning of declaration [-Werror=old-style-declaration] eicon/divasmain.c:473:1: error: '__inline__' is not at beginning of declaration [-Werror=old-style-declaration] eicon/platform.h:274:1: error: '__inline__' is not at beginning of declaration [-Werror=old-style-declaration] eicon/platform.h:280:1: error: '__inline__' is not at beginning of declaration [-Werror=old-style-declaration] A similar warning gets printed for the diva_os_register_io_port() declaration, because 'register' is interpreted as a keyword instead of a variable name: In file included from eicon/diva_didd.c:21:0: eicon/platform.h:206:1: error: 'register' is not at beginning of declaration [-Werror=old-style-declaration] Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller diff --git a/drivers/isdn/hardware/eicon/divasmain.c b/drivers/isdn/hardware/eicon/divasmain.c index a2e0ed6..32f3451 100644 --- a/drivers/isdn/hardware/eicon/divasmain.c +++ b/drivers/isdn/hardware/eicon/divasmain.c @@ -445,32 +445,32 @@ void divasa_unmap_pci_bar(void __iomem *bar) /********************************************************* ** I/O port access *********************************************************/ -byte __inline__ inpp(void __iomem *addr) +inline byte inpp(void __iomem *addr) { return (inb((unsigned long) addr)); } -word __inline__ inppw(void __iomem *addr) +inline word inppw(void __iomem *addr) { return (inw((unsigned long) addr)); } -void __inline__ inppw_buffer(void __iomem *addr, void *P, int length) +inline void inppw_buffer(void __iomem *addr, void *P, int length) { insw((unsigned long) addr, (word *) P, length >> 1); } -void __inline__ outppw_buffer(void __iomem *addr, void *P, int length) +inline void outppw_buffer(void __iomem *addr, void *P, int length) { outsw((unsigned long) addr, (word *) P, length >> 1); } -void __inline__ outppw(void __iomem *addr, word w) +inline void outppw(void __iomem *addr, word w) { outw(w, (unsigned long) addr); } -void __inline__ outpp(void __iomem *addr, word p) +inline void outpp(void __iomem *addr, word p) { outb(p, (unsigned long) addr); } diff --git a/drivers/isdn/hardware/eicon/platform.h b/drivers/isdn/hardware/eicon/platform.h index b2edb75..62e2073 100644 --- a/drivers/isdn/hardware/eicon/platform.h +++ b/drivers/isdn/hardware/eicon/platform.h @@ -203,7 +203,7 @@ void PCIread(byte bus, byte func, int offset, void *data, int length, void *pci_ /* ** I/O Port utilities */ -int diva_os_register_io_port(void *adapter, int register, unsigned long port, +int diva_os_register_io_port(void *adapter, int reg, unsigned long port, unsigned long length, const char *name, int id); /* ** I/O port access abstraction @@ -271,13 +271,13 @@ void diva_os_get_time(dword *sec, dword *usec); ** atomic operation, fake because we use threads */ typedef int diva_os_atomic_t; -static diva_os_atomic_t __inline__ +static inline diva_os_atomic_t diva_os_atomic_increment(diva_os_atomic_t *pv) { *pv += 1; return (*pv); } -static diva_os_atomic_t __inline__ +static inline diva_os_atomic_t diva_os_atomic_decrement(diva_os_atomic_t *pv) { *pv -= 1; -- cgit v0.10.2 From 278af574dbcb3165a3b9d38a460db7ad4766a7ca Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 15:52:13 +0200 Subject: net: gianfar: fix old-style declaration Modern C standards expect the '__inline__' keyword to come before the return type in a declaration, and we get a warning for this with "make W=1": drivers/net/ethernet/freescale/gianfar.c:2278:1: error: 'inline' is not at beginning of declaration [-Werror=old-style-declaration] Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index 2e6785b..d20935d 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -2275,7 +2275,7 @@ static inline void gfar_tx_checksum(struct sk_buff *skb, struct txfcb *fcb, fcb->flags = flags; } -void inline gfar_tx_vlan(struct sk_buff *skb, struct txfcb *fcb) +static inline void gfar_tx_vlan(struct sk_buff *skb, struct txfcb *fcb) { fcb->flags |= TXFCB_VLN; fcb->vlctl = cpu_to_be16(skb_vlan_tag_get(skb)); -- cgit v0.10.2 From 318d3cc04e8e42b3b138f7dea2297290636fad7d Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 15:59:25 +0200 Subject: net: xfrm: fix old-style declaration Modern C standards expect the '__inline__' keyword to come before the return type in a declaration, and we get a couple of warnings for this with "make W=1" in the xfrm{4,6}_policy.c files: net/ipv6/xfrm6_policy.c:369:1: error: 'inline' is not at beginning of declaration [-Werror=old-style-declaration] static int inline xfrm6_net_sysctl_init(struct net *net) net/ipv6/xfrm6_policy.c:374:1: error: 'inline' is not at beginning of declaration [-Werror=old-style-declaration] static void inline xfrm6_net_sysctl_exit(struct net *net) net/ipv4/xfrm4_policy.c:339:1: error: 'inline' is not at beginning of declaration [-Werror=old-style-declaration] static int inline xfrm4_net_sysctl_init(struct net *net) net/ipv4/xfrm4_policy.c:344:1: error: 'inline' is not at beginning of declaration [-Werror=old-style-declaration] static void inline xfrm4_net_sysctl_exit(struct net *net) Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 7b0edb3..b644a23 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -295,7 +295,7 @@ static struct ctl_table xfrm4_policy_table[] = { { } }; -static int __net_init xfrm4_net_sysctl_init(struct net *net) +static __net_init int xfrm4_net_sysctl_init(struct net *net) { struct ctl_table *table; struct ctl_table_header *hdr; @@ -323,7 +323,7 @@ err_alloc: return -ENOMEM; } -static void __net_exit xfrm4_net_sysctl_exit(struct net *net) +static __net_exit void xfrm4_net_sysctl_exit(struct net *net) { struct ctl_table *table; @@ -336,12 +336,12 @@ static void __net_exit xfrm4_net_sysctl_exit(struct net *net) kfree(table); } #else /* CONFIG_SYSCTL */ -static int inline xfrm4_net_sysctl_init(struct net *net) +static inline int xfrm4_net_sysctl_init(struct net *net) { return 0; } -static void inline xfrm4_net_sysctl_exit(struct net *net) +static inline void xfrm4_net_sysctl_exit(struct net *net) { } #endif diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index c074771..6cc9700 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -366,12 +366,12 @@ static void __net_exit xfrm6_net_sysctl_exit(struct net *net) kfree(table); } #else /* CONFIG_SYSCTL */ -static int inline xfrm6_net_sysctl_init(struct net *net) +static inline int xfrm6_net_sysctl_init(struct net *net) { return 0; } -static void inline xfrm6_net_sysctl_exit(struct net *net) +static inline void xfrm6_net_sysctl_exit(struct net *net) { } #endif -- cgit v0.10.2 From 1b05cf6285c108c8c2acfec47dd02ed50d973823 Mon Sep 17 00:00:00 2001 From: Eugene Crosser Date: Thu, 16 Jun 2016 16:18:51 +0200 Subject: qeth: Include error message for "OS Mismatch" Having understood the semantics of BRIDGEPORT error code 0x0010, we can introduce a meaningful error message. Signed-off-by: Eugene Crosser Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index 80b1979..faab0b6 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -1800,6 +1800,12 @@ static int qeth_bridgeport_makerc(struct qeth_card *card, dev_err(&card->gdev->dev, "The device is not configured as a Bridge Port\n"); break; + case 0x2B10: + case 0x0010: /* OS mismatch */ + rc = -EPERM; + dev_err(&card->gdev->dev, + "A Bridge Port is already configured by a different operating system\n"); + break; case 0x2B14: case 0x0014: /* Another device is Primary */ switch (setcmd) { -- cgit v0.10.2 From 2863c61334aa9fd82000da500075b7c959361919 Mon Sep 17 00:00:00 2001 From: Eugene Crosser Date: Thu, 16 Jun 2016 16:18:52 +0200 Subject: qeth: refactor calculation of SBALE count Rewrite the functions that calculate the required number of buffer elements needed to represent SKB data, to make them hopefully more comprehensible. Plus a few cleanups. Signed-off-by: Eugene Crosser Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index ec2e014..eb8f434 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -844,6 +844,19 @@ struct qeth_trap_id { /*some helper functions*/ #define QETH_CARD_IFNAME(card) (((card)->dev)? (card)->dev->name : "") +/** + * qeth_get_elements_for_range() - find number of SBALEs to cover range. + * @start: Start of the address range. + * @end: Address after the end of the range. + * + * Returns the number of pages, and thus QDIO buffer elements, needed to cover + * the specified address range. + */ +static inline int qeth_get_elements_for_range(addr_t start, addr_t end) +{ + return PFN_UP(end - 1) - PFN_DOWN(start); +} + static inline int qeth_get_micros(void) { return (int) (get_tod_clock() >> 12); diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index b7b7477..a91a31d 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -3810,41 +3810,54 @@ int qeth_get_priority_queue(struct qeth_card *card, struct sk_buff *skb, } EXPORT_SYMBOL_GPL(qeth_get_priority_queue); +/** + * qeth_get_elements_for_frags() - find number of SBALEs for skb frags. + * @skb: SKB address + * + * Returns the number of pages, and thus QDIO buffer elements, needed to cover + * fragmented part of the SKB. Returns zero for linear SKB. + */ int qeth_get_elements_for_frags(struct sk_buff *skb) { - int cnt, length, e, elements = 0; - struct skb_frag_struct *frag; - char *data; + int cnt, elements = 0; for (cnt = 0; cnt < skb_shinfo(skb)->nr_frags; cnt++) { - frag = &skb_shinfo(skb)->frags[cnt]; - data = (char *)page_to_phys(skb_frag_page(frag)) + - frag->page_offset; - length = frag->size; - e = PFN_UP((unsigned long)data + length - 1) - - PFN_DOWN((unsigned long)data); - elements += e; + struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[cnt]; + + elements += qeth_get_elements_for_range( + (addr_t)skb_frag_address(frag), + (addr_t)skb_frag_address(frag) + skb_frag_size(frag)); } return elements; } EXPORT_SYMBOL_GPL(qeth_get_elements_for_frags); +/** + * qeth_get_elements_no() - find number of SBALEs for skb data, inc. frags. + * @card: qeth card structure, to check max. elems. + * @skb: SKB address + * @extra_elems: extra elems needed, to check against max. + * + * Returns the number of pages, and thus QDIO buffer elements, needed to cover + * skb data, including linear part and fragments. Checks if the result plus + * extra_elems fits under the limit for the card. Returns 0 if it does not. + * Note: extra_elems is not included in the returned result. + */ int qeth_get_elements_no(struct qeth_card *card, - struct sk_buff *skb, int elems) + struct sk_buff *skb, int extra_elems) { - int dlen = skb->len - skb->data_len; - int elements_needed = PFN_UP((unsigned long)skb->data + dlen - 1) - - PFN_DOWN((unsigned long)skb->data); - - elements_needed += qeth_get_elements_for_frags(skb); + int elements = qeth_get_elements_for_range( + (addr_t)skb->data, + (addr_t)skb->data + skb_headlen(skb)) + + qeth_get_elements_for_frags(skb); - if ((elements_needed + elems) > QETH_MAX_BUFFER_ELEMENTS(card)) { + if ((elements + extra_elems) > QETH_MAX_BUFFER_ELEMENTS(card)) { QETH_DBF_MESSAGE(2, "Invalid size of IP packet " "(Number=%d / Length=%d). Discarded.\n", - (elements_needed+elems), skb->len); + elements + extra_elems, skb->len); return 0; } - return elements_needed; + return elements; } EXPORT_SYMBOL_GPL(qeth_get_elements_no); @@ -3859,7 +3872,7 @@ int qeth_hdr_chk_and_bounce(struct sk_buff *skb, struct qeth_hdr **hdr, int len) rest = len - inpage; if (rest > hroom) return 1; - memmove(skb->data - rest, skb->data, skb->len - skb->data_len); + memmove(skb->data - rest, skb->data, skb_headlen(skb)); skb->data -= rest; skb->tail -= rest; *hdr = (struct qeth_hdr *)skb->data; @@ -3873,7 +3886,7 @@ static inline void __qeth_fill_buffer(struct sk_buff *skb, struct qdio_buffer *buffer, int is_tso, int *next_element_to_fill, int offset) { - int length = skb->len - skb->data_len; + int length = skb_headlen(skb); int length_here; int element; char *data; diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index ac54433..7c9968a 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -2793,15 +2793,34 @@ static void qeth_tso_fill_header(struct qeth_card *card, } } -static inline int qeth_l3_tso_elements(struct sk_buff *skb) +/** + * qeth_get_elements_no_tso() - find number of SBALEs for skb data, inc. frags. + * @card: qeth card structure, to check max. elems. + * @skb: SKB address + * @extra_elems: extra elems needed, to check against max. + * + * Returns the number of pages, and thus QDIO buffer elements, needed to cover + * skb data, including linear part and fragments, but excluding TCP header. + * (Exclusion of TCP header distinguishes it from qeth_get_elements_no().) + * Checks if the result plus extra_elems fits under the limit for the card. + * Returns 0 if it does not. + * Note: extra_elems is not included in the returned result. + */ +static int qeth_get_elements_no_tso(struct qeth_card *card, + struct sk_buff *skb, int extra_elems) { - unsigned long tcpd = (unsigned long)tcp_hdr(skb) + - tcp_hdr(skb)->doff * 4; - int tcpd_len = skb_headlen(skb) - (tcpd - (unsigned long)skb->data); - int elements = PFN_UP(tcpd + tcpd_len - 1) - PFN_DOWN(tcpd); - - elements += qeth_get_elements_for_frags(skb); + addr_t tcpdptr = (addr_t)tcp_hdr(skb) + tcp_hdrlen(skb); + int elements = qeth_get_elements_for_range( + tcpdptr, + (addr_t)skb->data + skb_headlen(skb)) + + qeth_get_elements_for_frags(skb); + if ((elements + extra_elems) > QETH_MAX_BUFFER_ELEMENTS(card)) { + QETH_DBF_MESSAGE(2, + "Invalid size of TSO IP packet (Number=%d / Length=%d). Discarded.\n", + elements + extra_elems, skb->len); + return 0; + } return elements; } @@ -2810,8 +2829,8 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) int rc; u16 *tag; struct qeth_hdr *hdr = NULL; - int elements_needed = 0; - int elems; + int hdr_elements = 0; + int elements; struct qeth_card *card = dev->ml_priv; struct sk_buff *new_skb = NULL; int ipv = qeth_get_ip_version(skb); @@ -2859,7 +2878,7 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) hdr = kmem_cache_alloc(qeth_core_header_cache, GFP_ATOMIC); if (!hdr) goto tx_drop; - elements_needed++; + hdr_elements++; } else { /* create a clone with writeable headroom */ new_skb = skb_realloc_headroom(skb, sizeof(struct qeth_hdr_tso) @@ -2895,7 +2914,7 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) * chaining we can not send long frag lists */ if (large_send) { - if (qeth_l3_tso_elements(new_skb) + 1 > 16) { + if (!qeth_get_elements_no_tso(card, new_skb, 1)) { if (skb_linearize(new_skb)) goto tx_drop; if (card->options.performance_stats) @@ -2909,7 +2928,7 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) memset(hdr, 0, sizeof(struct qeth_hdr_tso)); qeth_l3_fill_header(card, hdr, new_skb, ipv, cast_type); qeth_tso_fill_header(card, hdr, new_skb); - elements_needed++; + hdr_elements++; } else { if (data_offset < 0) { hdr = (struct qeth_hdr *)skb_push(new_skb, @@ -2930,31 +2949,29 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) qeth_l3_hdr_csum(card, hdr, new_skb); } - elems = qeth_get_elements_no(card, new_skb, elements_needed); - if (!elems) { + elements = qeth_get_elements_no(card, new_skb, hdr_elements); + if (!elements) { if (data_offset >= 0) kmem_cache_free(qeth_core_header_cache, hdr); goto tx_drop; } - elements_needed += elems; - nr_frags = skb_shinfo(new_skb)->nr_frags; + elements += hdr_elements; if (card->info.type != QETH_CARD_TYPE_IQD) { int len; if (large_send) len = ((unsigned long)tcp_hdr(new_skb) + - tcp_hdr(new_skb)->doff * 4) - + tcp_hdrlen(new_skb)) - (unsigned long)new_skb->data; else len = sizeof(struct qeth_hdr_layer3); if (qeth_hdr_chk_and_bounce(new_skb, &hdr, len)) goto tx_drop; - rc = qeth_do_send_packet(card, queue, new_skb, hdr, - elements_needed); + rc = qeth_do_send_packet(card, queue, new_skb, hdr, elements); } else rc = qeth_do_send_packet_fast(card, queue, new_skb, hdr, - elements_needed, data_offset, 0); + elements, data_offset, 0); if (!rc) { card->stats.tx_packets++; @@ -2962,6 +2979,7 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) if (new_skb != skb) dev_kfree_skb_any(skb); if (card->options.performance_stats) { + nr_frags = skb_shinfo(new_skb)->nr_frags; if (large_send) { card->perf_stats.large_send_bytes += tx_bytes; card->perf_stats.large_send_cnt++; -- cgit v0.10.2 From 41aeed58af5d1b5d35988c1c82531306fa719a66 Mon Sep 17 00:00:00 2001 From: Eugene Crosser Date: Thu, 16 Jun 2016 16:18:53 +0200 Subject: qeth: clean up condition when tso is used Make conditions under which TSO is activated more stringent. Make calculation of SBALEs required for the skb more accurate. Signed-off-by: Eugene Crosser Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index eb8f434..c3ab956 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -878,6 +879,11 @@ static inline int qeth_get_ip_version(struct sk_buff *skb) } } +static inline int qeth_get_ip_protocol(struct sk_buff *skb) +{ + return ip_hdr(skb)->protocol; +} + static inline void qeth_put_buffer_pool_entry(struct qeth_card *card, struct qeth_buffer_pool_entry *entry) { diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index 7c9968a..86b3bc2 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -2794,10 +2794,10 @@ static void qeth_tso_fill_header(struct qeth_card *card, } /** - * qeth_get_elements_no_tso() - find number of SBALEs for skb data, inc. frags. - * @card: qeth card structure, to check max. elems. - * @skb: SKB address - * @extra_elems: extra elems needed, to check against max. + * qeth_l3_get_elements_no_tso() - find number of SBALEs for skb data for tso + * @card: qeth card structure, to check max. elems. + * @skb: SKB address + * @extra_elems: extra elems needed, to check against max. * * Returns the number of pages, and thus QDIO buffer elements, needed to cover * skb data, including linear part and fragments, but excluding TCP header. @@ -2806,7 +2806,7 @@ static void qeth_tso_fill_header(struct qeth_card *card, * Returns 0 if it does not. * Note: extra_elems is not included in the returned result. */ -static int qeth_get_elements_no_tso(struct qeth_card *card, +static int qeth_l3_get_elements_no_tso(struct qeth_card *card, struct sk_buff *skb, int extra_elems) { addr_t tcpdptr = (addr_t)tcp_hdr(skb) + tcp_hdrlen(skb); @@ -2841,7 +2841,7 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) qeth_get_priority_queue(card, skb, ipv, cast_type) : card->qdio.default_out_queue]; int tx_bytes = skb->len; - bool large_send; + bool use_tso; int data_offset = -1; int nr_frags; @@ -2866,10 +2866,12 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) card->perf_stats.outbound_start_time = qeth_get_micros(); } - large_send = skb_is_gso(skb); + /* Ignore segment size from skb_is_gso(), 1 page is always used. */ + use_tso = skb_is_gso(skb) && + (qeth_get_ip_protocol(skb) == IPPROTO_TCP) && (ipv == 4); - if ((card->info.type == QETH_CARD_TYPE_IQD) && (!large_send) && - (skb_shinfo(skb)->nr_frags == 0)) { + if ((card->info.type == QETH_CARD_TYPE_IQD) && + !skb_is_nonlinear(skb)) { new_skb = skb; if (new_skb->protocol == ETH_P_AF_IUCV) data_offset = 0; @@ -2913,16 +2915,16 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) /* fix hardware limitation: as long as we do not have sbal * chaining we can not send long frag lists */ - if (large_send) { - if (!qeth_get_elements_no_tso(card, new_skb, 1)) { - if (skb_linearize(new_skb)) - goto tx_drop; - if (card->options.performance_stats) - card->perf_stats.tx_lin++; - } + if ((card->info.type != QETH_CARD_TYPE_IQD) && + ((use_tso && !qeth_l3_get_elements_no_tso(card, new_skb, 1)) || + (!use_tso && !qeth_get_elements_no(card, new_skb, 0)))) { + if (skb_linearize(new_skb)) + goto tx_drop; + if (card->options.performance_stats) + card->perf_stats.tx_lin++; } - if (large_send && (cast_type == RTN_UNSPEC)) { + if (use_tso) { hdr = (struct qeth_hdr *)skb_push(new_skb, sizeof(struct qeth_hdr_tso)); memset(hdr, 0, sizeof(struct qeth_hdr_tso)); @@ -2949,7 +2951,9 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) qeth_l3_hdr_csum(card, hdr, new_skb); } - elements = qeth_get_elements_no(card, new_skb, hdr_elements); + elements = use_tso ? + qeth_l3_get_elements_no_tso(card, new_skb, hdr_elements) : + qeth_get_elements_no(card, new_skb, hdr_elements); if (!elements) { if (data_offset >= 0) kmem_cache_free(qeth_core_header_cache, hdr); @@ -2959,7 +2963,7 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) if (card->info.type != QETH_CARD_TYPE_IQD) { int len; - if (large_send) + if (use_tso) len = ((unsigned long)tcp_hdr(new_skb) + tcp_hdrlen(new_skb)) - (unsigned long)new_skb->data; @@ -2980,7 +2984,7 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) dev_kfree_skb_any(skb); if (card->options.performance_stats) { nr_frags = skb_shinfo(new_skb)->nr_frags; - if (large_send) { + if (use_tso) { card->perf_stats.large_send_bytes += tx_bytes; card->perf_stats.large_send_cnt++; } -- cgit v0.10.2 From 8bad55f8cedb429c13a3b43cad7473443b69b397 Mon Sep 17 00:00:00 2001 From: Eugene Crosser Date: Thu, 16 Jun 2016 16:18:54 +0200 Subject: qeth: fill netdevice->gso_* attributes accurately Use QETH_MAX_BUFFER_ELEMENTS(card) instead of constant 16. Also fill gso_max_segs and gso_min_segs. Signed-off-by: Eugene Crosser Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index 86b3bc2..76ddd18 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -3217,7 +3217,9 @@ static int qeth_l3_setup_netdev(struct qeth_card *card) NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER; netif_keep_dst(card->dev); - card->dev->gso_max_size = 15 * PAGE_SIZE; + card->dev->gso_max_size = (QETH_MAX_BUFFER_ELEMENTS(card) - 1) * + PAGE_SIZE; + card->dev->gso_max_segs = (QETH_MAX_BUFFER_ELEMENTS(card) - 1); SET_NETDEV_DEV(card->dev, &card->gdev->dev); netif_napi_add(card->dev, &card->napi, qeth_l3_poll, QETH_NAPI_WEIGHT); -- cgit v0.10.2 From d52aec97e5bc40b84fbe348f9d0ab56a6f3face4 Mon Sep 17 00:00:00 2001 From: Eugene Crosser Date: Thu, 16 Jun 2016 16:18:55 +0200 Subject: qeth: enable scatter/gather in layer 2 mode The patch enables NETIF_F_SG flag for OSA in layer 2 mode. It also adds performance accounting for fragmented sends, adds a conditional skb_linearize() attempt if the skb had too many fragments for QDIO SBAL, and fills netdevice->gso_* attributes. Signed-off-by: Eugene Crosser Signed-off-by: Ursula Braun Reviewed-by: Lakhvich Dmitriy Reviewed-by: Thomas Richter Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index faab0b6..ec163e4 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -869,6 +869,7 @@ static int qeth_l2_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) int data_offset = -1; int elements_needed = 0; int hd_len = 0; + int nr_frags; if (card->qdio.do_prio_queueing || (cast_type && card->info.is_multicast_different)) @@ -892,6 +893,17 @@ static int qeth_l2_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) } netif_stop_queue(dev); + /* fix hardware limitation: as long as we do not have sbal + * chaining we can not send long frag lists + */ + if ((card->info.type != QETH_CARD_TYPE_IQD) && + !qeth_get_elements_no(card, new_skb, 0)) { + if (skb_linearize(new_skb)) + goto tx_drop; + if (card->options.performance_stats) + card->perf_stats.tx_lin++; + } + if (card->info.type == QETH_CARD_TYPE_OSN) hdr = (struct qeth_hdr *)skb->data; else { @@ -943,6 +955,14 @@ static int qeth_l2_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) if (!rc) { card->stats.tx_packets++; card->stats.tx_bytes += tx_bytes; + if (card->options.performance_stats) { + nr_frags = skb_shinfo(new_skb)->nr_frags; + if (nr_frags) { + card->perf_stats.sg_skbs_sent++; + /* nr_frags + skb->data */ + card->perf_stats.sg_frags_sent += nr_frags + 1; + } + } if (new_skb != skb) dev_kfree_skb_any(skb); rc = NETDEV_TX_OK; @@ -1118,12 +1138,16 @@ static int qeth_l2_setup_netdev(struct qeth_card *card) &qeth_l2_ethtool_ops : &qeth_l2_osn_ops; card->dev->features |= NETIF_F_HW_VLAN_CTAG_FILTER; if (card->info.type == QETH_CARD_TYPE_OSD && !card->info.guestlan) { - card->dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM; + card->dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM | + NETIF_F_SG; /* Turn on RX offloading per default */ card->dev->features |= NETIF_F_RXCSUM; } card->info.broadcast_capable = 1; qeth_l2_request_initial_mac(card); + card->dev->gso_max_size = (QETH_MAX_BUFFER_ELEMENTS(card) - 1) * + PAGE_SIZE; + card->dev->gso_max_segs = (QETH_MAX_BUFFER_ELEMENTS(card) - 1); SET_NETDEV_DEV(card->dev, &card->gdev->dev); netif_napi_add(card->dev, &card->napi, qeth_l2_poll, QETH_NAPI_WEIGHT); netif_carrier_off(card->dev); -- cgit v0.10.2 From 2601e4ed3f643da2cd4e55f174355e02c0636a04 Mon Sep 17 00:00:00 2001 From: Eugene Crosser Date: Thu, 16 Jun 2016 16:18:56 +0200 Subject: qeth: enable scatter/gather by default Set scatter/gather ON by default on OSA, for both layer 2 and layer 3 modes. We always use fragmentation over QDIO anyway, so let the upper layers of the stack take advantage of that. Signed-off-by: Eugene Crosser Signed-off-by: Ursula Braun Reviewed-by: Lakhvich Dmitriy Reviewed-by: Thomas Richter Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index ec163e4..e24627f 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -1140,8 +1140,8 @@ static int qeth_l2_setup_netdev(struct qeth_card *card) if (card->info.type == QETH_CARD_TYPE_OSD && !card->info.guestlan) { card->dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM | NETIF_F_SG; - /* Turn on RX offloading per default */ - card->dev->features |= NETIF_F_RXCSUM; + /* Turn on RX offloading and SG per default */ + card->dev->features |= NETIF_F_RXCSUM | NETIF_F_SG; } card->info.broadcast_capable = 1; qeth_l2_request_initial_mac(card); diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index 76ddd18..0afdb14 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -3191,7 +3191,8 @@ static int qeth_l3_setup_netdev(struct qeth_card *card) card->dev->hw_features = NETIF_F_SG | NETIF_F_RXCSUM | NETIF_F_IP_CSUM | NETIF_F_TSO; - card->dev->features = NETIF_F_RXCSUM; + card->dev->features = NETIF_F_RXCSUM | + NETIF_F_SG; } } } else if (card->info.type == QETH_CARD_TYPE_IQD) { -- cgit v0.10.2 From 6059c90537868bfd365163ba3dff95cf775939ff Mon Sep 17 00:00:00 2001 From: Eugene Crosser Date: Thu, 16 Jun 2016 16:18:57 +0200 Subject: qeth: introduce linearization fail count to stats When skb data touches too many pages, skb_linearize() is called opportunistically in the hope that less pages will be required for a big linear buffer than for multiple fragments. This patch intoduces a separate counter in ethtool statistics structure representing _failed_ linearization attempts. Signed-off-by: Eugene Crosser Signed-off-by: Ursula Braun Reviewed-by: Lakhvich Dmitriy Reviewed-by: Thomas Richter Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index c3ab956..b4f14e4 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -145,6 +145,7 @@ struct qeth_perf_stats { unsigned int sg_alloc_page_rx; unsigned int tx_csum; unsigned int tx_lin; + unsigned int tx_linfail; }; /* Routing stuff */ diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index a91a31d..9806ee0 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -5801,6 +5801,7 @@ static struct { {"tx do_QDIO count"}, {"tx csum"}, {"tx lin"}, + {"tx linfail"}, {"cq handler count"}, {"cq handler time"} }; @@ -5861,8 +5862,9 @@ void qeth_core_get_ethtool_stats(struct net_device *dev, data[32] = card->perf_stats.outbound_do_qdio_cnt; data[33] = card->perf_stats.tx_csum; data[34] = card->perf_stats.tx_lin; - data[35] = card->perf_stats.cq_cnt; - data[36] = card->perf_stats.cq_time; + data[35] = card->perf_stats.tx_linfail; + data[36] = card->perf_stats.cq_cnt; + data[37] = card->perf_stats.cq_time; } EXPORT_SYMBOL_GPL(qeth_core_get_ethtool_stats); diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index e24627f..08b5fa9 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -898,10 +898,16 @@ static int qeth_l2_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) */ if ((card->info.type != QETH_CARD_TYPE_IQD) && !qeth_get_elements_no(card, new_skb, 0)) { - if (skb_linearize(new_skb)) + int lin_rc = skb_linearize(new_skb); + + if (card->options.performance_stats) { + if (lin_rc) + card->perf_stats.tx_linfail++; + else + card->perf_stats.tx_lin++; + } + if (lin_rc) goto tx_drop; - if (card->options.performance_stats) - card->perf_stats.tx_lin++; } if (card->info.type == QETH_CARD_TYPE_OSN) diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index 0afdb14..51077fb 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -2918,10 +2918,16 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) if ((card->info.type != QETH_CARD_TYPE_IQD) && ((use_tso && !qeth_l3_get_elements_no_tso(card, new_skb, 1)) || (!use_tso && !qeth_get_elements_no(card, new_skb, 0)))) { - if (skb_linearize(new_skb)) + int lin_rc = skb_linearize(new_skb); + + if (card->options.performance_stats) { + if (lin_rc) + card->perf_stats.tx_linfail++; + else + card->perf_stats.tx_lin++; + } + if (lin_rc) goto tx_drop; - if (card->options.performance_stats) - card->perf_stats.tx_lin++; } if (use_tso) { -- cgit v0.10.2 From 5f78e29ceebf03a80a5141515bd5b48ca83f0495 Mon Sep 17 00:00:00 2001 From: Lakhvich Dmitriy Date: Thu, 16 Jun 2016 16:18:58 +0200 Subject: qeth: optimize IP handling in rx_mode callback In layer3 mode of the qeth driver, multicast IP addresses from struct net_device and other type of IP addresses from other sources require mapping to the OSA-card. This patch simplifies the IP address mapping logic, and changes imple- mentation of ndo_set_rx_mode callback and ip notifier events. Addresses are stored in private hashtables instead of lists now. It allows hardware registration/removal for new/deleted multicast addresses only. Signed-off-by: Lakhvich Dmitriy Signed-off-by: Ursula Braun Reviewed-by: Evgeny Cherkashin Reviewed-by: Thomas Richter Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index b4f14e4..ab0a171 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -561,7 +561,6 @@ enum qeth_ip_types { QETH_IP_TYPE_NORMAL, QETH_IP_TYPE_VIPA, QETH_IP_TYPE_RXIP, - QETH_IP_TYPE_DEL_ALL_MC, }; enum qeth_cmd_buffer_state { @@ -742,17 +741,10 @@ struct qeth_vlan_vid { unsigned short vid; }; -enum qeth_mac_disposition { - QETH_DISP_MAC_DELETE = 0, - QETH_DISP_MAC_DO_NOTHING = 1, - QETH_DISP_MAC_ADD = 2, -}; - -struct qeth_mac { - u8 mac_addr[OSA_ADDR_LEN]; - u8 is_uc:1; - u8 disp_flag:2; - struct hlist_node hnode; +enum qeth_addr_disposition { + QETH_DISP_ADDR_DELETE = 0, + QETH_DISP_ADDR_DO_NOTHING = 1, + QETH_DISP_ADDR_ADD = 2, }; struct qeth_rx { @@ -800,6 +792,8 @@ struct qeth_card { unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)]; struct list_head vid_list; DECLARE_HASHTABLE(mac_htable, 4); + DECLARE_HASHTABLE(ip_htable, 4); + DECLARE_HASHTABLE(ip_mc_htable, 4); struct work_struct kernel_thread_starter; spinlock_t thread_mask_lock; unsigned long thread_start_mask; @@ -807,8 +801,6 @@ struct qeth_card { unsigned long thread_running_mask; struct task_struct *recovery_task; spinlock_t ip_lock; - struct list_head ip_list; - struct list_head *ip_tbd_list; struct qeth_ipato ipato; struct list_head cmd_waiter_list; /* QDIO buffer handling */ diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 9806ee0..ede9ed8 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -1464,8 +1464,6 @@ static int qeth_setup_card(struct qeth_card *card) card->thread_allowed_mask = 0; card->thread_running_mask = 0; INIT_WORK(&card->kernel_thread_starter, qeth_start_kernel_thread); - INIT_LIST_HEAD(&card->ip_list); - INIT_LIST_HEAD(card->ip_tbd_list); INIT_LIST_HEAD(&card->cmd_waiter_list); init_waitqueue_head(&card->wait_q); /* initial options */ @@ -1500,11 +1498,6 @@ static struct qeth_card *qeth_alloc_card(void) if (!card) goto out; QETH_DBF_HEX(SETUP, 2, &card, sizeof(void *)); - card->ip_tbd_list = kzalloc(sizeof(struct list_head), GFP_KERNEL); - if (!card->ip_tbd_list) { - QETH_DBF_TEXT(SETUP, 0, "iptbdnom"); - goto out_card; - } if (qeth_setup_channel(&card->read)) goto out_ip; if (qeth_setup_channel(&card->write)) @@ -1517,8 +1510,6 @@ static struct qeth_card *qeth_alloc_card(void) out_channel: qeth_clean_channel(&card->read); out_ip: - kfree(card->ip_tbd_list); -out_card: kfree(card); out: return NULL; @@ -4980,7 +4971,6 @@ static void qeth_core_free_card(struct qeth_card *card) qeth_clean_channel(&card->write); if (card->dev) free_netdev(card->dev); - kfree(card->ip_tbd_list); qeth_free_qdio_buffers(card); unregister_service_level(&card->qeth_service_level); kfree(card); diff --git a/drivers/s390/net/qeth_l2.h b/drivers/s390/net/qeth_l2.h index 0767556..29d9fb3 100644 --- a/drivers/s390/net/qeth_l2.h +++ b/drivers/s390/net/qeth_l2.h @@ -12,4 +12,11 @@ int qeth_l2_create_device_attributes(struct device *); void qeth_l2_remove_device_attributes(struct device *); void qeth_l2_setup_bridgeport_attrs(struct qeth_card *card); +struct qeth_mac { + u8 mac_addr[OSA_ADDR_LEN]; + u8 is_uc:1; + u8 disp_flag:2; + struct hlist_node hnode; +}; + #endif /* __QETH_L2_H__ */ diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index 08b5fa9..2a331d1 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -780,7 +780,7 @@ qeth_l2_add_mac(struct qeth_card *card, struct netdev_hw_addr *ha, u8 is_uc) qeth_l2_mac_hash(ha->addr)) { if (is_uc == mac->is_uc && !memcmp(ha->addr, mac->mac_addr, OSA_ADDR_LEN)) { - mac->disp_flag = QETH_DISP_MAC_DO_NOTHING; + mac->disp_flag = QETH_DISP_ADDR_DO_NOTHING; return; } } @@ -792,7 +792,7 @@ qeth_l2_add_mac(struct qeth_card *card, struct netdev_hw_addr *ha, u8 is_uc) memcpy(mac->mac_addr, ha->addr, OSA_ADDR_LEN); mac->is_uc = is_uc; - mac->disp_flag = QETH_DISP_MAC_ADD; + mac->disp_flag = QETH_DISP_ADDR_ADD; hash_add(card->mac_htable, &mac->hnode, qeth_l2_mac_hash(mac->mac_addr)); @@ -825,7 +825,7 @@ static void qeth_l2_set_rx_mode(struct net_device *dev) qeth_l2_add_mac(card, ha, 1); hash_for_each_safe(card->mac_htable, i, tmp, mac, hnode) { - if (mac->disp_flag == QETH_DISP_MAC_DELETE) { + if (mac->disp_flag == QETH_DISP_ADDR_DELETE) { if (!mac->is_uc) rc = qeth_l2_send_delgroupmac(card, mac->mac_addr); @@ -837,15 +837,15 @@ static void qeth_l2_set_rx_mode(struct net_device *dev) hash_del(&mac->hnode); kfree(mac); - } else if (mac->disp_flag == QETH_DISP_MAC_ADD) { + } else if (mac->disp_flag == QETH_DISP_ADDR_ADD) { rc = qeth_l2_write_mac(card, mac); if (rc) { hash_del(&mac->hnode); kfree(mac); } else - mac->disp_flag = QETH_DISP_MAC_DELETE; + mac->disp_flag = QETH_DISP_ADDR_DELETE; } else - mac->disp_flag = QETH_DISP_MAC_DELETE; + mac->disp_flag = QETH_DISP_ADDR_DELETE; } spin_unlock_bh(&card->mclock); diff --git a/drivers/s390/net/qeth_l3.h b/drivers/s390/net/qeth_l3.h index 551a4b4..26f7953 100644 --- a/drivers/s390/net/qeth_l3.h +++ b/drivers/s390/net/qeth_l3.h @@ -10,16 +10,23 @@ #define __QETH_L3_H__ #include "qeth_core.h" +#include #define QETH_SNIFF_AVAIL 0x0008 struct qeth_ipaddr { - struct list_head entry; + struct hlist_node hnode; enum qeth_ip_types type; enum qeth_ipa_setdelip_flags set_flags; enum qeth_ipa_setdelip_flags del_flags; - int is_multicast; - int users; + u8 is_multicast:1; + u8 in_progress:1; + u8 disp_flag:2; + + /* is changed only for normal ip addresses + * for non-normal addresses it always is 1 + */ + int ref_counter; enum qeth_prot_versions proto; unsigned char mac[OSA_ADDR_LEN]; union { @@ -32,7 +39,24 @@ struct qeth_ipaddr { unsigned int pfxlen; } a6; } u; + }; +static inline u64 qeth_l3_ipaddr_hash(struct qeth_ipaddr *addr) +{ + u64 ret = 0; + u8 *point; + + if (addr->proto == QETH_PROT_IPV6) { + point = (u8 *) &addr->u.a6.addr; + ret = get_unaligned((u64 *)point) ^ + get_unaligned((u64 *) (point + 8)); + } + if (addr->proto == QETH_PROT_IPV4) { + point = (u8 *) &addr->u.a4.addr; + ret = get_unaligned((u32 *) point); + } + return ret; +} struct qeth_ipato_entry { struct list_head entry; @@ -60,6 +84,5 @@ int qeth_l3_is_addr_covered_by_ipato(struct qeth_card *, struct qeth_ipaddr *); struct qeth_ipaddr *qeth_l3_get_addr_buffer(enum qeth_prot_versions); int qeth_l3_add_ip(struct qeth_card *, struct qeth_ipaddr *); int qeth_l3_delete_ip(struct qeth_card *, struct qeth_ipaddr *); -void qeth_l3_set_ip_addr_list(struct qeth_card *); #endif /* __QETH_L3_H__ */ diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index 51077fb..6e7d06c 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -30,6 +30,7 @@ #include #include #include +#include #include "qeth_l3.h" @@ -57,7 +58,7 @@ static int qeth_l3_isxdigit(char *buf) static void qeth_l3_ipaddr4_to_string(const __u8 *addr, char *buf) { - sprintf(buf, "%i.%i.%i.%i", addr[0], addr[1], addr[2], addr[3]); + sprintf(buf, "%pI4", addr); } static int qeth_l3_string_to_ipaddr4(const char *buf, __u8 *addr) @@ -204,104 +205,129 @@ int qeth_l3_is_addr_covered_by_ipato(struct qeth_card *card, return rc; } -/* - * Add IP to be added to todo list. If there is already an "add todo" - * in this list we just incremenent the reference count. - * Returns 0 if we just incremented reference count. - */ -static int __qeth_l3_insert_ip_todo(struct qeth_card *card, - struct qeth_ipaddr *addr, int add) +inline int +qeth_l3_ipaddrs_is_equal(struct qeth_ipaddr *addr1, struct qeth_ipaddr *addr2) { - struct qeth_ipaddr *tmp, *t; - int found = 0; + return addr1->proto == addr2->proto && + !memcmp(&addr1->u, &addr2->u, sizeof(addr1->u)) && + !memcmp(&addr1->mac, &addr2->mac, sizeof(addr1->mac)); +} - if (card->options.sniffer) - return 0; - list_for_each_entry_safe(tmp, t, card->ip_tbd_list, entry) { - if ((addr->type == QETH_IP_TYPE_DEL_ALL_MC) && - (tmp->type == QETH_IP_TYPE_DEL_ALL_MC)) - return 0; - if ((tmp->proto == QETH_PROT_IPV4) && - (addr->proto == QETH_PROT_IPV4) && - (tmp->type == addr->type) && - (tmp->is_multicast == addr->is_multicast) && - (tmp->u.a4.addr == addr->u.a4.addr) && - (tmp->u.a4.mask == addr->u.a4.mask)) { - found = 1; - break; - } - if ((tmp->proto == QETH_PROT_IPV6) && - (addr->proto == QETH_PROT_IPV6) && - (tmp->type == addr->type) && - (tmp->is_multicast == addr->is_multicast) && - (tmp->u.a6.pfxlen == addr->u.a6.pfxlen) && - (memcmp(&tmp->u.a6.addr, &addr->u.a6.addr, - sizeof(struct in6_addr)) == 0)) { - found = 1; - break; - } - } - if (found) { - if (addr->users != 0) - tmp->users += addr->users; - else - tmp->users += add ? 1 : -1; - if (tmp->users == 0) { - list_del(&tmp->entry); - kfree(tmp); - } - return 0; +static struct qeth_ipaddr * +qeth_l3_ip_from_hash(struct qeth_card *card, struct qeth_ipaddr *tmp_addr) +{ + struct qeth_ipaddr *addr; + + if (tmp_addr->is_multicast) { + hash_for_each_possible(card->ip_mc_htable, addr, + hnode, qeth_l3_ipaddr_hash(tmp_addr)) + if (qeth_l3_ipaddrs_is_equal(tmp_addr, addr)) + return addr; } else { - if (addr->type == QETH_IP_TYPE_DEL_ALL_MC) - list_add(&addr->entry, card->ip_tbd_list); - else { - if (addr->users == 0) - addr->users += add ? 1 : -1; - if (add && (addr->type == QETH_IP_TYPE_NORMAL) && - qeth_l3_is_addr_covered_by_ipato(card, addr)) { - QETH_CARD_TEXT(card, 2, "tkovaddr"); - addr->set_flags |= QETH_IPA_SETIP_TAKEOVER_FLAG; - } - list_add_tail(&addr->entry, card->ip_tbd_list); - } - return 1; + hash_for_each_possible(card->ip_htable, addr, + hnode, qeth_l3_ipaddr_hash(tmp_addr)) + if (qeth_l3_ipaddrs_is_equal(tmp_addr, addr)) + return addr; } + + return NULL; } -int qeth_l3_delete_ip(struct qeth_card *card, struct qeth_ipaddr *addr) +int qeth_l3_delete_ip(struct qeth_card *card, struct qeth_ipaddr *tmp_addr) { - unsigned long flags; int rc = 0; + struct qeth_ipaddr *addr; QETH_CARD_TEXT(card, 4, "delip"); - if (addr->proto == QETH_PROT_IPV4) - QETH_CARD_HEX(card, 4, &addr->u.a4.addr, 4); + if (tmp_addr->proto == QETH_PROT_IPV4) + QETH_CARD_HEX(card, 4, &tmp_addr->u.a4.addr, 4); else { - QETH_CARD_HEX(card, 4, &addr->u.a6.addr, 8); - QETH_CARD_HEX(card, 4, ((char *)&addr->u.a6.addr) + 8, 8); + QETH_CARD_HEX(card, 4, &tmp_addr->u.a6.addr, 8); + QETH_CARD_HEX(card, 4, ((char *)&tmp_addr->u.a6.addr) + 8, 8); } - spin_lock_irqsave(&card->ip_lock, flags); - rc = __qeth_l3_insert_ip_todo(card, addr, 0); - spin_unlock_irqrestore(&card->ip_lock, flags); + + addr = qeth_l3_ip_from_hash(card, tmp_addr); + if (!addr) + return -ENOENT; + + addr->ref_counter--; + if (addr->type == QETH_IP_TYPE_NORMAL && addr->ref_counter > 0) + return rc; + if (addr->in_progress) + return -EINPROGRESS; + + rc = qeth_l3_deregister_addr_entry(card, addr); + + hash_del(&addr->hnode); + kfree(addr); + return rc; } -int qeth_l3_add_ip(struct qeth_card *card, struct qeth_ipaddr *addr) +int qeth_l3_add_ip(struct qeth_card *card, struct qeth_ipaddr *tmp_addr) { - unsigned long flags; int rc = 0; + struct qeth_ipaddr *addr; QETH_CARD_TEXT(card, 4, "addip"); - if (addr->proto == QETH_PROT_IPV4) - QETH_CARD_HEX(card, 4, &addr->u.a4.addr, 4); + + if (tmp_addr->proto == QETH_PROT_IPV4) + QETH_CARD_HEX(card, 4, &tmp_addr->u.a4.addr, 4); else { - QETH_CARD_HEX(card, 4, &addr->u.a6.addr, 8); - QETH_CARD_HEX(card, 4, ((char *)&addr->u.a6.addr) + 8, 8); + QETH_CARD_HEX(card, 4, &tmp_addr->u.a6.addr, 8); + QETH_CARD_HEX(card, 4, ((char *)&tmp_addr->u.a6.addr) + 8, 8); + } + + addr = qeth_l3_ip_from_hash(card, tmp_addr); + if (!addr) { + addr = qeth_l3_get_addr_buffer(tmp_addr->proto); + if (!addr) + return -ENOMEM; + + memcpy(addr, tmp_addr, sizeof(struct qeth_ipaddr)); + addr->ref_counter = 1; + + if (addr->type == QETH_IP_TYPE_NORMAL && + qeth_l3_is_addr_covered_by_ipato(card, addr)) { + QETH_CARD_TEXT(card, 2, "tkovaddr"); + addr->set_flags |= QETH_IPA_SETIP_TAKEOVER_FLAG; + } + hash_add(card->ip_htable, &addr->hnode, + qeth_l3_ipaddr_hash(addr)); + + /* qeth_l3_register_addr_entry can go to sleep + * if we add a IPV4 addr. It is caused by the reason + * that SETIP ipa cmd starts ARP staff for IPV4 addr. + * Thus we should unlock spinlock, and make a protection + * using in_progress variable to indicate that there is + * an hardware operation with this IPV4 address + */ + if (addr->proto == QETH_PROT_IPV4) { + addr->in_progress = 1; + spin_unlock_bh(&card->ip_lock); + rc = qeth_l3_register_addr_entry(card, addr); + spin_lock_bh(&card->ip_lock); + addr->in_progress = 0; + } else + rc = qeth_l3_register_addr_entry(card, addr); + + if (!rc || (rc == IPA_RC_DUPLICATE_IP_ADDRESS) || + (rc == IPA_RC_LAN_OFFLINE)) { + addr->disp_flag = QETH_DISP_ADDR_DO_NOTHING; + if (addr->ref_counter < 1) { + qeth_l3_delete_ip(card, addr); + kfree(addr); + } + } else { + hash_del(&addr->hnode); + kfree(addr); + } + } else { + if (addr->type == QETH_IP_TYPE_NORMAL) + addr->ref_counter++; } - spin_lock_irqsave(&card->ip_lock, flags); - rc = __qeth_l3_insert_ip_todo(card, addr, 1); - spin_unlock_irqrestore(&card->ip_lock, flags); + return rc; } @@ -312,229 +338,88 @@ struct qeth_ipaddr *qeth_l3_get_addr_buffer( struct qeth_ipaddr *addr; addr = kzalloc(sizeof(struct qeth_ipaddr), GFP_ATOMIC); - if (addr == NULL) { + if (!addr) return NULL; - } + addr->type = QETH_IP_TYPE_NORMAL; + addr->disp_flag = QETH_DISP_ADDR_DO_NOTHING; addr->proto = prot; - return addr; -} -static void qeth_l3_delete_mc_addresses(struct qeth_card *card) -{ - struct qeth_ipaddr *iptodo; - unsigned long flags; - - QETH_CARD_TEXT(card, 4, "delmc"); - iptodo = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); - if (!iptodo) { - QETH_CARD_TEXT(card, 2, "dmcnomem"); - return; - } - iptodo->type = QETH_IP_TYPE_DEL_ALL_MC; - spin_lock_irqsave(&card->ip_lock, flags); - if (!__qeth_l3_insert_ip_todo(card, iptodo, 0)) - kfree(iptodo); - spin_unlock_irqrestore(&card->ip_lock, flags); + return addr; } -/* - * Add/remove address to/from card's ip list, i.e. try to add or remove - * reference to/from an IP address that is already registered on the card. - * Returns: - * 0 address was on card and its reference count has been adjusted, - * but is still > 0, so nothing has to be done - * also returns 0 if card was not on card and the todo was to delete - * the address -> there is also nothing to be done - * 1 address was not on card and the todo is to add it to the card's ip - * list - * -1 address was on card and its reference count has been decremented - * to <= 0 by the todo -> address must be removed from card - */ -static int __qeth_l3_ref_ip_on_card(struct qeth_card *card, - struct qeth_ipaddr *todo, struct qeth_ipaddr **__addr) +static void qeth_l3_clear_ip_htable(struct qeth_card *card, int recover) { struct qeth_ipaddr *addr; - int found = 0; - - list_for_each_entry(addr, &card->ip_list, entry) { - if ((addr->proto == QETH_PROT_IPV4) && - (todo->proto == QETH_PROT_IPV4) && - (addr->type == todo->type) && - (addr->u.a4.addr == todo->u.a4.addr) && - (addr->u.a4.mask == todo->u.a4.mask)) { - found = 1; - break; - } - if ((addr->proto == QETH_PROT_IPV6) && - (todo->proto == QETH_PROT_IPV6) && - (addr->type == todo->type) && - (addr->u.a6.pfxlen == todo->u.a6.pfxlen) && - (memcmp(&addr->u.a6.addr, &todo->u.a6.addr, - sizeof(struct in6_addr)) == 0)) { - found = 1; - break; - } - } - if (found) { - addr->users += todo->users; - if (addr->users <= 0) { - *__addr = addr; - return -1; - } else { - /* for VIPA and RXIP limit refcount to 1 */ - if (addr->type != QETH_IP_TYPE_NORMAL) - addr->users = 1; - return 0; - } - } - if (todo->users > 0) { - /* for VIPA and RXIP limit refcount to 1 */ - if (todo->type != QETH_IP_TYPE_NORMAL) - todo->users = 1; - return 1; - } else - return 0; -} - -static void __qeth_l3_delete_all_mc(struct qeth_card *card, - unsigned long *flags) -{ - struct list_head fail_list; - struct qeth_ipaddr *addr, *tmp; - int rc; - - INIT_LIST_HEAD(&fail_list); -again: - list_for_each_entry_safe(addr, tmp, &card->ip_list, entry) { - if (addr->is_multicast) { - list_del(&addr->entry); - spin_unlock_irqrestore(&card->ip_lock, *flags); - rc = qeth_l3_deregister_addr_entry(card, addr); - spin_lock_irqsave(&card->ip_lock, *flags); - if (!rc || (rc == IPA_RC_MC_ADDR_NOT_FOUND)) - kfree(addr); - else - list_add_tail(&addr->entry, &fail_list); - goto again; - } - } - list_splice(&fail_list, &card->ip_list); -} - -void qeth_l3_set_ip_addr_list(struct qeth_card *card) -{ - struct list_head *tbd_list; - struct qeth_ipaddr *todo, *addr; - unsigned long flags; - int rc; + struct hlist_node *tmp; + int i; - QETH_CARD_TEXT(card, 2, "sdiplist"); - QETH_CARD_HEX(card, 2, &card, sizeof(void *)); + QETH_CARD_TEXT(card, 4, "clearip"); - if (!qeth_card_hw_is_reachable(card) || card->options.sniffer) + if (recover && card->options.sniffer) return; - spin_lock_irqsave(&card->ip_lock, flags); - tbd_list = card->ip_tbd_list; - card->ip_tbd_list = kzalloc(sizeof(struct list_head), GFP_ATOMIC); - if (!card->ip_tbd_list) { - QETH_CARD_TEXT(card, 0, "silnomem"); - card->ip_tbd_list = tbd_list; - spin_unlock_irqrestore(&card->ip_lock, flags); - return; - } else - INIT_LIST_HEAD(card->ip_tbd_list); - - while (!list_empty(tbd_list)) { - todo = list_entry(tbd_list->next, struct qeth_ipaddr, entry); - list_del(&todo->entry); - if (todo->type == QETH_IP_TYPE_DEL_ALL_MC) { - __qeth_l3_delete_all_mc(card, &flags); - kfree(todo); + spin_lock_bh(&card->ip_lock); + + hash_for_each_safe(card->ip_htable, i, tmp, addr, hnode) { + if (!recover) { + hash_del(&addr->hnode); + kfree(addr); continue; } - rc = __qeth_l3_ref_ip_on_card(card, todo, &addr); - if (rc == 0) { - /* nothing to be done; only adjusted refcount */ - kfree(todo); - } else if (rc == 1) { - /* new entry to be added to on-card list */ - spin_unlock_irqrestore(&card->ip_lock, flags); - rc = qeth_l3_register_addr_entry(card, todo); - spin_lock_irqsave(&card->ip_lock, flags); - if (!rc || (rc == IPA_RC_LAN_OFFLINE)) - list_add_tail(&todo->entry, &card->ip_list); - else - kfree(todo); - } else if (rc == -1) { - /* on-card entry to be removed */ - list_del_init(&addr->entry); - spin_unlock_irqrestore(&card->ip_lock, flags); - rc = qeth_l3_deregister_addr_entry(card, addr); - spin_lock_irqsave(&card->ip_lock, flags); - if (!rc || (rc == IPA_RC_IP_ADDRESS_NOT_DEFINED)) - kfree(addr); - else - list_add_tail(&addr->entry, &card->ip_list); - kfree(todo); - } + addr->disp_flag = QETH_DISP_ADDR_ADD; } - spin_unlock_irqrestore(&card->ip_lock, flags); - kfree(tbd_list); -} -static void qeth_l3_clear_ip_list(struct qeth_card *card, int recover) -{ - struct qeth_ipaddr *addr, *tmp; - unsigned long flags; + spin_unlock_bh(&card->ip_lock); - QETH_CARD_TEXT(card, 4, "clearip"); - if (recover && card->options.sniffer) - return; - spin_lock_irqsave(&card->ip_lock, flags); - /* clear todo list */ - list_for_each_entry_safe(addr, tmp, card->ip_tbd_list, entry) { - list_del(&addr->entry); + spin_lock_bh(&card->mclock); + + hash_for_each_safe(card->ip_mc_htable, i, tmp, addr, hnode) { + hash_del(&addr->hnode); kfree(addr); } - while (!list_empty(&card->ip_list)) { - addr = list_entry(card->ip_list.next, - struct qeth_ipaddr, entry); - list_del_init(&addr->entry); - if (!recover || addr->is_multicast) { - kfree(addr); - continue; - } - list_add_tail(&addr->entry, card->ip_tbd_list); - } - spin_unlock_irqrestore(&card->ip_lock, flags); -} + spin_unlock_bh(&card->mclock); + -static int qeth_l3_address_exists_in_list(struct list_head *list, - struct qeth_ipaddr *addr, int same_type) +} +static void qeth_l3_recover_ip(struct qeth_card *card) { - struct qeth_ipaddr *tmp; + struct qeth_ipaddr *addr; + struct hlist_node *tmp; + int i; + int rc; - list_for_each_entry(tmp, list, entry) { - if ((tmp->proto == QETH_PROT_IPV4) && - (addr->proto == QETH_PROT_IPV4) && - ((same_type && (tmp->type == addr->type)) || - (!same_type && (tmp->type != addr->type))) && - (tmp->u.a4.addr == addr->u.a4.addr)) - return 1; + QETH_CARD_TEXT(card, 4, "recoverip"); + + spin_lock_bh(&card->ip_lock); + + hash_for_each_safe(card->ip_htable, i, tmp, addr, hnode) { + if (addr->disp_flag == QETH_DISP_ADDR_ADD) { + if (addr->proto == QETH_PROT_IPV4) { + addr->in_progress = 1; + spin_unlock_bh(&card->ip_lock); + rc = qeth_l3_register_addr_entry(card, addr); + spin_lock_bh(&card->ip_lock); + addr->in_progress = 0; + } else + rc = qeth_l3_register_addr_entry(card, addr); + + if (!rc) { + addr->disp_flag = QETH_DISP_ADDR_DO_NOTHING; + if (addr->ref_counter < 1) { + qeth_l3_delete_ip(card, addr); + kfree(addr); + } + } else { + hash_del(&addr->hnode); + kfree(addr); + } + } + } - if ((tmp->proto == QETH_PROT_IPV6) && - (addr->proto == QETH_PROT_IPV6) && - ((same_type && (tmp->type == addr->type)) || - (!same_type && (tmp->type != addr->type))) && - (memcmp(&tmp->u.a6.addr, &addr->u.a6.addr, - sizeof(struct in6_addr)) == 0)) - return 1; + spin_unlock_bh(&card->ip_lock); - } - return 0; } static int qeth_l3_send_setdelmc(struct qeth_card *card, @@ -712,27 +597,28 @@ int qeth_l3_setrouting_v6(struct qeth_card *card) */ static void qeth_l3_clear_ipato_list(struct qeth_card *card) { - struct qeth_ipato_entry *ipatoe, *tmp; - unsigned long flags; - spin_lock_irqsave(&card->ip_lock, flags); + spin_lock_bh(&card->ip_lock); + list_for_each_entry_safe(ipatoe, tmp, &card->ipato.entries, entry) { list_del(&ipatoe->entry); kfree(ipatoe); } - spin_unlock_irqrestore(&card->ip_lock, flags); + + spin_unlock_bh(&card->ip_lock); } int qeth_l3_add_ipato_entry(struct qeth_card *card, struct qeth_ipato_entry *new) { struct qeth_ipato_entry *ipatoe; - unsigned long flags; int rc = 0; QETH_CARD_TEXT(card, 2, "addipato"); - spin_lock_irqsave(&card->ip_lock, flags); + + spin_lock_bh(&card->ip_lock); + list_for_each_entry(ipatoe, &card->ipato.entries, entry) { if (ipatoe->proto != new->proto) continue; @@ -743,10 +629,12 @@ int qeth_l3_add_ipato_entry(struct qeth_card *card, break; } } + if (!rc) list_add_tail(&new->entry, &card->ipato.entries); - spin_unlock_irqrestore(&card->ip_lock, flags); + spin_unlock_bh(&card->ip_lock); + return rc; } @@ -754,10 +642,11 @@ void qeth_l3_del_ipato_entry(struct qeth_card *card, enum qeth_prot_versions proto, u8 *addr, int mask_bits) { struct qeth_ipato_entry *ipatoe, *tmp; - unsigned long flags; QETH_CARD_TEXT(card, 2, "delipato"); - spin_lock_irqsave(&card->ip_lock, flags); + + spin_lock_bh(&card->ip_lock); + list_for_each_entry_safe(ipatoe, tmp, &card->ipato.entries, entry) { if (ipatoe->proto != proto) continue; @@ -768,7 +657,8 @@ void qeth_l3_del_ipato_entry(struct qeth_card *card, kfree(ipatoe); } } - spin_unlock_irqrestore(&card->ip_lock, flags); + + spin_unlock_bh(&card->ip_lock); } /* @@ -778,7 +668,6 @@ int qeth_l3_add_vipa(struct qeth_card *card, enum qeth_prot_versions proto, const u8 *addr) { struct qeth_ipaddr *ipaddr; - unsigned long flags; int rc = 0; ipaddr = qeth_l3_get_addr_buffer(proto); @@ -797,18 +686,18 @@ int qeth_l3_add_vipa(struct qeth_card *card, enum qeth_prot_versions proto, ipaddr->del_flags = QETH_IPA_DELIP_VIPA_FLAG; } else return -ENOMEM; - spin_lock_irqsave(&card->ip_lock, flags); - if (qeth_l3_address_exists_in_list(&card->ip_list, ipaddr, 0) || - qeth_l3_address_exists_in_list(card->ip_tbd_list, ipaddr, 0)) + + spin_lock_bh(&card->ip_lock); + + if (!qeth_l3_ip_from_hash(card, ipaddr)) rc = -EEXIST; - spin_unlock_irqrestore(&card->ip_lock, flags); - if (rc) { - kfree(ipaddr); - return rc; - } - if (!qeth_l3_add_ip(card, ipaddr)) - kfree(ipaddr); - qeth_l3_set_ip_addr_list(card); + else + qeth_l3_add_ip(card, ipaddr); + + spin_unlock_bh(&card->ip_lock); + + kfree(ipaddr); + return rc; } @@ -831,9 +720,12 @@ void qeth_l3_del_vipa(struct qeth_card *card, enum qeth_prot_versions proto, ipaddr->type = QETH_IP_TYPE_VIPA; } else return; - if (!qeth_l3_delete_ip(card, ipaddr)) - kfree(ipaddr); - qeth_l3_set_ip_addr_list(card); + + spin_lock_bh(&card->ip_lock); + qeth_l3_delete_ip(card, ipaddr); + spin_unlock_bh(&card->ip_lock); + + kfree(ipaddr); } /* @@ -843,7 +735,6 @@ int qeth_l3_add_rxip(struct qeth_card *card, enum qeth_prot_versions proto, const u8 *addr) { struct qeth_ipaddr *ipaddr; - unsigned long flags; int rc = 0; ipaddr = qeth_l3_get_addr_buffer(proto); @@ -857,24 +748,25 @@ int qeth_l3_add_rxip(struct qeth_card *card, enum qeth_prot_versions proto, memcpy(&ipaddr->u.a6.addr, addr, 16); ipaddr->u.a6.pfxlen = 0; } + ipaddr->type = QETH_IP_TYPE_RXIP; ipaddr->set_flags = QETH_IPA_SETIP_TAKEOVER_FLAG; ipaddr->del_flags = 0; } else return -ENOMEM; - spin_lock_irqsave(&card->ip_lock, flags); - if (qeth_l3_address_exists_in_list(&card->ip_list, ipaddr, 0) || - qeth_l3_address_exists_in_list(card->ip_tbd_list, ipaddr, 0)) + + spin_lock_bh(&card->ip_lock); + + if (!qeth_l3_ip_from_hash(card, ipaddr)) rc = -EEXIST; - spin_unlock_irqrestore(&card->ip_lock, flags); - if (rc) { - kfree(ipaddr); - return rc; - } - if (!qeth_l3_add_ip(card, ipaddr)) - kfree(ipaddr); - qeth_l3_set_ip_addr_list(card); - return 0; + else + qeth_l3_add_ip(card, ipaddr); + + spin_unlock_bh(&card->ip_lock); + + kfree(ipaddr); + + return rc; } void qeth_l3_del_rxip(struct qeth_card *card, enum qeth_prot_versions proto, @@ -896,9 +788,12 @@ void qeth_l3_del_rxip(struct qeth_card *card, enum qeth_prot_versions proto, ipaddr->type = QETH_IP_TYPE_RXIP; } else return; - if (!qeth_l3_delete_ip(card, ipaddr)) - kfree(ipaddr); - qeth_l3_set_ip_addr_list(card); + + spin_lock_bh(&card->ip_lock); + qeth_l3_delete_ip(card, ipaddr); + spin_unlock_bh(&card->ip_lock); + + kfree(ipaddr); } static int qeth_l3_register_addr_entry(struct qeth_card *card, @@ -908,6 +803,7 @@ static int qeth_l3_register_addr_entry(struct qeth_card *card, int rc = 0; int cnt = 3; + if (addr->proto == QETH_PROT_IPV4) { QETH_CARD_TEXT(card, 2, "setaddr4"); QETH_CARD_HEX(card, 3, &addr->u.a4.addr, sizeof(int)); @@ -1507,31 +1403,99 @@ qeth_diags_trace(struct qeth_card *card, enum qeth_diags_trace_cmds diags_cmd) return qeth_send_ipa_cmd(card, iob, qeth_diags_trace_cb, NULL); } -static void qeth_l3_get_mac_for_ipm(__u32 ipm, char *mac, - struct net_device *dev) +static void qeth_l3_get_mac_for_ipm(__u32 ipm, char *mac) { ip_eth_mc_map(ipm, mac); } -static void qeth_l3_add_mc(struct qeth_card *card, struct in_device *in4_dev) +static void qeth_l3_mark_all_mc_to_be_deleted(struct qeth_card *card) +{ + struct qeth_ipaddr *addr; + int i; + + hash_for_each(card->ip_mc_htable, i, addr, hnode) + addr->disp_flag = QETH_DISP_ADDR_DELETE; + +} + +static void qeth_l3_add_all_new_mc(struct qeth_card *card) +{ + struct qeth_ipaddr *addr; + struct hlist_node *tmp; + int i; + int rc; + + hash_for_each_safe(card->ip_mc_htable, i, tmp, addr, hnode) { + if (addr->disp_flag == QETH_DISP_ADDR_ADD) { + rc = qeth_l3_register_addr_entry(card, addr); + if (!rc || (rc == IPA_RC_LAN_OFFLINE)) + addr->ref_counter = 1; + else { + hash_del(&addr->hnode); + kfree(addr); + } + } + } + +} + +static void qeth_l3_delete_nonused_mc(struct qeth_card *card) +{ + struct qeth_ipaddr *addr; + struct hlist_node *tmp; + int i; + int rc; + + hash_for_each_safe(card->ip_mc_htable, i, tmp, addr, hnode) { + if (addr->disp_flag == QETH_DISP_ADDR_DELETE) { + rc = qeth_l3_deregister_addr_entry(card, addr); + if (!rc || (rc == IPA_RC_MC_ADDR_NOT_FOUND)) { + hash_del(&addr->hnode); + kfree(addr); + } + } + } + +} + + +static void +qeth_l3_add_mc_to_hash(struct qeth_card *card, struct in_device *in4_dev) { - struct qeth_ipaddr *ipm; struct ip_mc_list *im4; + struct qeth_ipaddr *tmp, *ipm; char buf[MAX_ADDR_LEN]; QETH_CARD_TEXT(card, 4, "addmc"); + + tmp = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); + if (!tmp) + return; + for (im4 = rcu_dereference(in4_dev->mc_list); im4 != NULL; im4 = rcu_dereference(im4->next_rcu)) { - qeth_l3_get_mac_for_ipm(im4->multiaddr, buf, in4_dev->dev); - ipm = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); - if (!ipm) - continue; - ipm->u.a4.addr = im4->multiaddr; - memcpy(ipm->mac, buf, OSA_ADDR_LEN); - ipm->is_multicast = 1; - if (!qeth_l3_add_ip(card, ipm)) - kfree(ipm); + qeth_l3_get_mac_for_ipm(im4->multiaddr, buf); + + tmp->u.a4.addr = im4->multiaddr; + memcpy(tmp->mac, buf, sizeof(tmp->mac)); + + ipm = qeth_l3_ip_from_hash(card, tmp); + if (ipm) { + ipm->disp_flag = QETH_DISP_ADDR_DO_NOTHING; + } else { + ipm = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); + if (!ipm) + continue; + memcpy(ipm->mac, buf, sizeof(tmp->mac)); + ipm->u.a4.addr = im4->multiaddr; + ipm->is_multicast = 1; + ipm->disp_flag = QETH_DISP_ADDR_ADD; + hash_add(card->ip_mc_htable, + &ipm->hnode, qeth_l3_ipaddr_hash(ipm)); + } } + + kfree(tmp); } /* called with rcu_read_lock */ @@ -1541,6 +1505,7 @@ static void qeth_l3_add_vlan_mc(struct qeth_card *card) u16 vid; QETH_CARD_TEXT(card, 4, "addmcvl"); + if (!qeth_is_supported(card, IPA_FULL_VLAN)) return; @@ -1555,7 +1520,7 @@ static void qeth_l3_add_vlan_mc(struct qeth_card *card) in_dev = __in_dev_get_rcu(netdev); if (!in_dev) continue; - qeth_l3_add_mc(card, in_dev); + qeth_l3_add_mc_to_hash(card, in_dev); } } @@ -1564,36 +1529,60 @@ static void qeth_l3_add_multicast_ipv4(struct qeth_card *card) struct in_device *in4_dev; QETH_CARD_TEXT(card, 4, "chkmcv4"); + rcu_read_lock(); in4_dev = __in_dev_get_rcu(card->dev); if (in4_dev == NULL) goto unlock; - qeth_l3_add_mc(card, in4_dev); + qeth_l3_add_mc_to_hash(card, in4_dev); qeth_l3_add_vlan_mc(card); unlock: rcu_read_unlock(); } #ifdef CONFIG_QETH_IPV6 -static void qeth_l3_add_mc6(struct qeth_card *card, struct inet6_dev *in6_dev) +static void +qeth_l3_add_mc6_to_hash(struct qeth_card *card, struct inet6_dev *in6_dev) { struct qeth_ipaddr *ipm; struct ifmcaddr6 *im6; + struct qeth_ipaddr *tmp; char buf[MAX_ADDR_LEN]; QETH_CARD_TEXT(card, 4, "addmc6"); + + tmp = qeth_l3_get_addr_buffer(QETH_PROT_IPV6); + if (!tmp) + return; + for (im6 = in6_dev->mc_list; im6 != NULL; im6 = im6->next) { ndisc_mc_map(&im6->mca_addr, buf, in6_dev->dev, 0); + + memcpy(tmp->mac, buf, sizeof(tmp->mac)); + memcpy(&tmp->u.a6.addr, &im6->mca_addr.s6_addr, + sizeof(struct in6_addr)); + tmp->is_multicast = 1; + + ipm = qeth_l3_ip_from_hash(card, tmp); + if (ipm) { + ipm->disp_flag = QETH_DISP_ADDR_DO_NOTHING; + continue; + } + ipm = qeth_l3_get_addr_buffer(QETH_PROT_IPV6); if (!ipm) continue; - ipm->is_multicast = 1; + memcpy(ipm->mac, buf, OSA_ADDR_LEN); memcpy(&ipm->u.a6.addr, &im6->mca_addr.s6_addr, sizeof(struct in6_addr)); - if (!qeth_l3_add_ip(card, ipm)) - kfree(ipm); + ipm->is_multicast = 1; + ipm->disp_flag = QETH_DISP_ADDR_ADD; + hash_add(card->ip_mc_htable, + &ipm->hnode, qeth_l3_ipaddr_hash(ipm)); + } + kfree(tmp); } /* called with rcu_read_lock */ @@ -1603,6 +1592,7 @@ static void qeth_l3_add_vlan_mc6(struct qeth_card *card) u16 vid; QETH_CARD_TEXT(card, 4, "admc6vl"); + if (!qeth_is_supported(card, IPA_FULL_VLAN)) return; @@ -1618,7 +1608,7 @@ static void qeth_l3_add_vlan_mc6(struct qeth_card *card) if (!in_dev) continue; read_lock_bh(&in_dev->lock); - qeth_l3_add_mc6(card, in_dev); + qeth_l3_add_mc6_to_hash(card, in_dev); read_unlock_bh(&in_dev->lock); in6_dev_put(in_dev); } @@ -1629,14 +1619,16 @@ static void qeth_l3_add_multicast_ipv6(struct qeth_card *card) struct inet6_dev *in6_dev; QETH_CARD_TEXT(card, 4, "chkmcv6"); + if (!qeth_is_supported(card, IPA_IPV6)) return ; in6_dev = in6_dev_get(card->dev); - if (in6_dev == NULL) + if (!in6_dev) return; + rcu_read_lock(); read_lock_bh(&in6_dev->lock); - qeth_l3_add_mc6(card, in6_dev); + qeth_l3_add_mc6_to_hash(card, in6_dev); qeth_l3_add_vlan_mc6(card); read_unlock_bh(&in6_dev->lock); rcu_read_unlock(); @@ -1660,16 +1652,23 @@ static void qeth_l3_free_vlan_addresses4(struct qeth_card *card, in_dev = in_dev_get(netdev); if (!in_dev) return; + + addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); + if (!addr) + return; + + spin_lock_bh(&card->ip_lock); + for (ifa = in_dev->ifa_list; ifa; ifa = ifa->ifa_next) { - addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); - if (addr) { - addr->u.a4.addr = ifa->ifa_address; - addr->u.a4.mask = ifa->ifa_mask; - addr->type = QETH_IP_TYPE_NORMAL; - if (!qeth_l3_delete_ip(card, addr)) - kfree(addr); - } + addr->u.a4.addr = ifa->ifa_address; + addr->u.a4.mask = ifa->ifa_mask; + addr->type = QETH_IP_TYPE_NORMAL; + qeth_l3_delete_ip(card, addr); } + + spin_unlock_bh(&card->ip_lock); + + kfree(addr); in_dev_put(in_dev); } @@ -1687,20 +1686,28 @@ static void qeth_l3_free_vlan_addresses6(struct qeth_card *card, netdev = __vlan_find_dev_deep_rcu(card->dev, htons(ETH_P_8021Q), vid); if (!netdev) return; + in6_dev = in6_dev_get(netdev); if (!in6_dev) return; + + addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV6); + if (!addr) + return; + + spin_lock_bh(&card->ip_lock); + list_for_each_entry(ifa, &in6_dev->addr_list, if_list) { - addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV6); - if (addr) { - memcpy(&addr->u.a6.addr, &ifa->addr, - sizeof(struct in6_addr)); - addr->u.a6.pfxlen = ifa->prefix_len; - addr->type = QETH_IP_TYPE_NORMAL; - if (!qeth_l3_delete_ip(card, addr)) - kfree(addr); - } + memcpy(&addr->u.a6.addr, &ifa->addr, + sizeof(struct in6_addr)); + addr->u.a6.pfxlen = ifa->prefix_len; + addr->type = QETH_IP_TYPE_NORMAL; + qeth_l3_delete_ip(card, addr); } + + spin_unlock_bh(&card->ip_lock); + + kfree(addr); in6_dev_put(in6_dev); #endif /* CONFIG_QETH_IPV6 */ } @@ -1727,18 +1734,16 @@ static int qeth_l3_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, u16 vid) { struct qeth_card *card = dev->ml_priv; - unsigned long flags; QETH_CARD_TEXT_(card, 4, "kid:%d", vid); + if (qeth_wait_for_threads(card, QETH_RECOVER_THREAD)) { QETH_CARD_TEXT(card, 3, "kidREC"); return 0; } - spin_lock_irqsave(&card->vlanlock, flags); /* unregister IP addresses of vlan device */ qeth_l3_free_vlan_addresses(card, vid); clear_bit(vid, card->active_vlans); - spin_unlock_irqrestore(&card->vlanlock, flags); qeth_l3_set_multicast_list(card->dev); return 0; } @@ -1994,8 +1999,8 @@ static int qeth_l3_verify_vlan_dev(struct net_device *dev, static int qeth_l3_verify_dev(struct net_device *dev) { struct qeth_card *card; - unsigned long flags; int rc = 0; + unsigned long flags; read_lock_irqsave(&qeth_core_card_list.rwlock, flags); list_for_each_entry(card, &qeth_core_card_list.list, list) { @@ -2051,7 +2056,7 @@ static void qeth_l3_stop_card(struct qeth_card *card, int recovery_mode) card->state = CARD_STATE_SOFTSETUP; } if (card->state == CARD_STATE_SOFTSETUP) { - qeth_l3_clear_ip_list(card, 1); + qeth_l3_clear_ip_htable(card, 1); qeth_clear_ipacmd_list(card); card->state = CARD_STATE_HARDSETUP; } @@ -2106,12 +2111,20 @@ static void qeth_l3_set_multicast_list(struct net_device *dev) (card->state != CARD_STATE_UP)) return; if (!card->options.sniffer) { - qeth_l3_delete_mc_addresses(card); + + spin_lock_bh(&card->mclock); + + qeth_l3_mark_all_mc_to_be_deleted(card); + qeth_l3_add_multicast_ipv4(card); #ifdef CONFIG_QETH_IPV6 qeth_l3_add_multicast_ipv6(card); #endif - qeth_l3_set_ip_addr_list(card); + qeth_l3_delete_nonused_mc(card); + qeth_l3_add_all_new_mc(card); + + spin_unlock_bh(&card->mclock); + if (!qeth_adp_supported(card, IPA_SETADP_SET_PROMISC_MODE)) return; } @@ -3261,7 +3274,7 @@ static void qeth_l3_remove_device(struct ccwgroup_device *cgdev) card->dev = NULL; } - qeth_l3_clear_ip_list(card, 0); + qeth_l3_clear_ip_htable(card, 0); qeth_l3_clear_ipato_list(card); return; } @@ -3346,7 +3359,7 @@ contin: card->state = CARD_STATE_SOFTSETUP; qeth_set_allowed_threads(card, 0xffffffff, 0); - qeth_l3_set_ip_addr_list(card); + qeth_l3_recover_ip(card); if (card->lan_online) netif_carrier_on(card->dev); else @@ -3547,6 +3560,7 @@ EXPORT_SYMBOL_GPL(qeth_l3_discipline); static int qeth_l3_ip_event(struct notifier_block *this, unsigned long event, void *ptr) { + struct in_ifaddr *ifa = (struct in_ifaddr *)ptr; struct net_device *dev = (struct net_device *)ifa->ifa_dev->dev; struct qeth_ipaddr *addr; @@ -3561,27 +3575,27 @@ static int qeth_l3_ip_event(struct notifier_block *this, QETH_CARD_TEXT(card, 3, "ipevent"); addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV4); - if (addr != NULL) { + if (addr) { addr->u.a4.addr = ifa->ifa_address; addr->u.a4.mask = ifa->ifa_mask; addr->type = QETH_IP_TYPE_NORMAL; } else - goto out; + return NOTIFY_DONE; switch (event) { case NETDEV_UP: - if (!qeth_l3_add_ip(card, addr)) - kfree(addr); + spin_lock_bh(&card->ip_lock); + qeth_l3_add_ip(card, addr); + spin_unlock_bh(&card->ip_lock); break; case NETDEV_DOWN: - if (!qeth_l3_delete_ip(card, addr)) - kfree(addr); - break; - default: + spin_lock_bh(&card->ip_lock); + qeth_l3_delete_ip(card, addr); + spin_unlock_bh(&card->ip_lock); break; } - qeth_l3_set_ip_addr_list(card); -out: + + kfree(addr); return NOTIFY_DONE; } @@ -3610,27 +3624,27 @@ static int qeth_l3_ip6_event(struct notifier_block *this, return NOTIFY_DONE; addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV6); - if (addr != NULL) { + if (addr) { memcpy(&addr->u.a6.addr, &ifa->addr, sizeof(struct in6_addr)); addr->u.a6.pfxlen = ifa->prefix_len; addr->type = QETH_IP_TYPE_NORMAL; } else - goto out; + return NOTIFY_DONE; switch (event) { case NETDEV_UP: - if (!qeth_l3_add_ip(card, addr)) - kfree(addr); + spin_lock_bh(&card->ip_lock); + qeth_l3_add_ip(card, addr); + spin_unlock_bh(&card->ip_lock); break; case NETDEV_DOWN: - if (!qeth_l3_delete_ip(card, addr)) - kfree(addr); - break; - default: + spin_lock_bh(&card->ip_lock); + qeth_l3_delete_ip(card, addr); + spin_unlock_bh(&card->ip_lock); break; } - qeth_l3_set_ip_addr_list(card); -out: + + kfree(addr); return NOTIFY_DONE; } diff --git a/drivers/s390/net/qeth_l3_sys.c b/drivers/s390/net/qeth_l3_sys.c index 386eb7b..65645b1 100644 --- a/drivers/s390/net/qeth_l3_sys.c +++ b/drivers/s390/net/qeth_l3_sys.c @@ -8,6 +8,7 @@ #include #include +#include #include "qeth_l3.h" #define QETH_DEVICE_ATTR(_id, _name, _mode, _show, _store) \ @@ -285,19 +286,19 @@ static ssize_t qeth_l3_dev_hsuid_store(struct device *dev, if (card->options.hsuid[0]) { /* delete old ip address */ addr = qeth_l3_get_addr_buffer(QETH_PROT_IPV6); - if (addr != NULL) { - addr->u.a6.addr.s6_addr32[0] = 0xfe800000; - addr->u.a6.addr.s6_addr32[1] = 0x00000000; - for (i = 8; i < 16; i++) - addr->u.a6.addr.s6_addr[i] = - card->options.hsuid[i - 8]; - addr->u.a6.pfxlen = 0; - addr->type = QETH_IP_TYPE_NORMAL; - } else + if (!addr) return -ENOMEM; - if (!qeth_l3_delete_ip(card, addr)) - kfree(addr); - qeth_l3_set_ip_addr_list(card); + + addr->u.a6.addr.s6_addr32[0] = 0xfe800000; + addr->u.a6.addr.s6_addr32[1] = 0x00000000; + for (i = 8; i < 16; i++) + addr->u.a6.addr.s6_addr[i] = + card->options.hsuid[i - 8]; + addr->u.a6.pfxlen = 0; + addr->type = QETH_IP_TYPE_NORMAL; + + qeth_l3_delete_ip(card, addr); + kfree(addr); } if (strlen(tmp) == 0) { @@ -328,9 +329,8 @@ static ssize_t qeth_l3_dev_hsuid_store(struct device *dev, addr->type = QETH_IP_TYPE_NORMAL; } else return -ENOMEM; - if (!qeth_l3_add_ip(card, addr)) - kfree(addr); - qeth_l3_set_ip_addr_list(card); + qeth_l3_add_ip(card, addr); + kfree(addr); return count; } @@ -367,8 +367,8 @@ static ssize_t qeth_l3_dev_ipato_enable_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { struct qeth_card *card = dev_get_drvdata(dev); - struct qeth_ipaddr *tmpipa, *t; - int rc = 0; + struct qeth_ipaddr *addr; + int i, rc = 0; if (!card) return -EINVAL; @@ -384,21 +384,20 @@ static ssize_t qeth_l3_dev_ipato_enable_store(struct device *dev, card->ipato.enabled = (card->ipato.enabled)? 0 : 1; } else if (sysfs_streq(buf, "1")) { card->ipato.enabled = 1; - list_for_each_entry_safe(tmpipa, t, card->ip_tbd_list, entry) { - if ((tmpipa->type == QETH_IP_TYPE_NORMAL) && - qeth_l3_is_addr_covered_by_ipato(card, tmpipa)) - tmpipa->set_flags |= + hash_for_each(card->ip_htable, i, addr, hnode) { + if ((addr->type == QETH_IP_TYPE_NORMAL) && + qeth_l3_is_addr_covered_by_ipato(card, addr)) + addr->set_flags |= QETH_IPA_SETIP_TAKEOVER_FLAG; - } - + } } else if (sysfs_streq(buf, "0")) { card->ipato.enabled = 0; - list_for_each_entry_safe(tmpipa, t, card->ip_tbd_list, entry) { - if (tmpipa->set_flags & - QETH_IPA_SETIP_TAKEOVER_FLAG) - tmpipa->set_flags &= - ~QETH_IPA_SETIP_TAKEOVER_FLAG; - } + hash_for_each(card->ip_htable, i, addr, hnode) { + if (addr->set_flags & + QETH_IPA_SETIP_TAKEOVER_FLAG) + addr->set_flags &= + ~QETH_IPA_SETIP_TAKEOVER_FLAG; + } } else rc = -EINVAL; out: @@ -452,7 +451,6 @@ static ssize_t qeth_l3_dev_ipato_add_show(char *buf, struct qeth_card *card, enum qeth_prot_versions proto) { struct qeth_ipato_entry *ipatoe; - unsigned long flags; char addr_str[40]; int entry_len; /* length of 1 entry string, differs between v4 and v6 */ int i = 0; @@ -460,7 +458,7 @@ static ssize_t qeth_l3_dev_ipato_add_show(char *buf, struct qeth_card *card, entry_len = (proto == QETH_PROT_IPV4)? 12 : 40; /* add strlen for "/\n" */ entry_len += (proto == QETH_PROT_IPV4)? 5 : 6; - spin_lock_irqsave(&card->ip_lock, flags); + spin_lock_bh(&card->ip_lock); list_for_each_entry(ipatoe, &card->ipato.entries, entry) { if (ipatoe->proto != proto) continue; @@ -473,7 +471,7 @@ static ssize_t qeth_l3_dev_ipato_add_show(char *buf, struct qeth_card *card, i += snprintf(buf + i, PAGE_SIZE - i, "%s/%i\n", addr_str, ipatoe->mask_bits); } - spin_unlock_irqrestore(&card->ip_lock, flags); + spin_unlock_bh(&card->ip_lock); i += snprintf(buf + i, PAGE_SIZE - i, "\n"); return i; @@ -689,15 +687,15 @@ static ssize_t qeth_l3_dev_vipa_add_show(char *buf, struct qeth_card *card, enum qeth_prot_versions proto) { struct qeth_ipaddr *ipaddr; + struct hlist_node *tmp; char addr_str[40]; int entry_len; /* length of 1 entry string, differs between v4 and v6 */ - unsigned long flags; int i = 0; entry_len = (proto == QETH_PROT_IPV4)? 12 : 40; entry_len += 2; /* \n + terminator */ - spin_lock_irqsave(&card->ip_lock, flags); - list_for_each_entry(ipaddr, &card->ip_list, entry) { + spin_lock_bh(&card->ip_lock); + hash_for_each_safe(card->ip_htable, i, tmp, ipaddr, hnode) { if (ipaddr->proto != proto) continue; if (ipaddr->type != QETH_IP_TYPE_VIPA) @@ -711,7 +709,7 @@ static ssize_t qeth_l3_dev_vipa_add_show(char *buf, struct qeth_card *card, addr_str); i += snprintf(buf + i, PAGE_SIZE - i, "%s\n", addr_str); } - spin_unlock_irqrestore(&card->ip_lock, flags); + spin_unlock_bh(&card->ip_lock); i += snprintf(buf + i, PAGE_SIZE - i, "\n"); return i; @@ -851,15 +849,15 @@ static ssize_t qeth_l3_dev_rxip_add_show(char *buf, struct qeth_card *card, enum qeth_prot_versions proto) { struct qeth_ipaddr *ipaddr; + struct hlist_node *tmp; char addr_str[40]; int entry_len; /* length of 1 entry string, differs between v4 and v6 */ - unsigned long flags; int i = 0; entry_len = (proto == QETH_PROT_IPV4)? 12 : 40; entry_len += 2; /* \n + terminator */ - spin_lock_irqsave(&card->ip_lock, flags); - list_for_each_entry(ipaddr, &card->ip_list, entry) { + spin_lock_bh(&card->ip_lock); + hash_for_each_safe(card->ip_htable, i, tmp, ipaddr, hnode) { if (ipaddr->proto != proto) continue; if (ipaddr->type != QETH_IP_TYPE_RXIP) @@ -873,7 +871,7 @@ static ssize_t qeth_l3_dev_rxip_add_show(char *buf, struct qeth_card *card, addr_str); i += snprintf(buf + i, PAGE_SIZE - i, "%s\n", addr_str); } - spin_unlock_irqrestore(&card->ip_lock, flags); + spin_unlock_bh(&card->ip_lock); i += snprintf(buf + i, PAGE_SIZE - i, "\n"); return i; -- cgit v0.10.2 From 8f43fb00a154712daeaa23e75704c7677294cffc Mon Sep 17 00:00:00 2001 From: Thomas Richter Date: Thu, 16 Jun 2016 16:18:59 +0200 Subject: qeth layer 2 and layer 3 common feature handling This patch introduces a common set of fix_features and set_features functions for layer 2 and layer 3. The RX, TX and TSO offload functionality on the OSA card is enabled using ethtool at user's request and not at device initialization as done before. For layer 3 the RX checksum offloading is disabled at device initialization time. Signed-off-by: Thomas Richter Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h index ab0a171..bf40063 100644 --- a/drivers/s390/net/qeth_core.h +++ b/drivers/s390/net/qeth_core.h @@ -993,12 +993,13 @@ int qeth_send_setassparms(struct qeth_card *, struct qeth_cmd_buffer *, __u16, int (*reply_cb)(struct qeth_card *, struct qeth_reply *, unsigned long), void *); +int qeth_setassparms_cb(struct qeth_card *, struct qeth_reply *, unsigned long); struct qeth_cmd_buffer *qeth_get_setassparms_cmd(struct qeth_card *, enum qeth_ipa_funcs, __u16, __u16, enum qeth_prot_versions); -int qeth_start_ipa_tx_checksum(struct qeth_card *); -int qeth_set_rx_csum(struct qeth_card *, int); +int qeth_set_features(struct net_device *, netdev_features_t); +netdev_features_t qeth_fix_features(struct net_device *, netdev_features_t); /* exports for OSN */ int qeth_osn_assist(struct net_device *, void *, int); diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index ede9ed8..19a6ee0 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -5268,8 +5268,8 @@ no_mem: } EXPORT_SYMBOL_GPL(qeth_core_get_next_skb); -static int qeth_setassparms_cb(struct qeth_card *card, - struct qeth_reply *reply, unsigned long data) +int qeth_setassparms_cb(struct qeth_card *card, + struct qeth_reply *reply, unsigned long data) { struct qeth_ipa_cmd *cmd; @@ -5297,6 +5297,7 @@ static int qeth_setassparms_cb(struct qeth_card *card, return 0; } +EXPORT_SYMBOL_GPL(qeth_setassparms_cb); struct qeth_cmd_buffer *qeth_get_setassparms_cmd(struct qeth_card *card, enum qeth_ipa_funcs ipa_func, @@ -6053,74 +6054,120 @@ int qeth_core_ethtool_get_settings(struct net_device *netdev, } EXPORT_SYMBOL_GPL(qeth_core_ethtool_get_settings); -static int qeth_send_checksum_command(struct qeth_card *card) +static int qeth_send_checksum_on(struct qeth_card *card, int cstype) { + long rxtx_arg; int rc; - rc = qeth_send_simple_setassparms(card, IPA_INBOUND_CHECKSUM, - IPA_CMD_ASS_START, 0); + rc = qeth_send_simple_setassparms(card, cstype, IPA_CMD_ASS_START, 0); if (rc) { - dev_warn(&card->gdev->dev, "Starting HW checksumming for %s " - "failed, using SW checksumming\n", - QETH_CARD_IFNAME(card)); + dev_warn(&card->gdev->dev, + "Starting HW checksumming for %s failed, using SW checksumming\n", + QETH_CARD_IFNAME(card)); return rc; } - rc = qeth_send_simple_setassparms(card, IPA_INBOUND_CHECKSUM, - IPA_CMD_ASS_ENABLE, - card->info.csum_mask); + rxtx_arg = (cstype == IPA_OUTBOUND_CHECKSUM) ? card->info.tx_csum_mask + : card->info.csum_mask; + rc = qeth_send_simple_setassparms(card, cstype, IPA_CMD_ASS_ENABLE, + rxtx_arg); if (rc) { - dev_warn(&card->gdev->dev, "Enabling HW checksumming for %s " - "failed, using SW checksumming\n", - QETH_CARD_IFNAME(card)); + dev_warn(&card->gdev->dev, + "Enabling HW checksumming for %s failed, using SW checksumming\n", + QETH_CARD_IFNAME(card)); return rc; } + + dev_info(&card->gdev->dev, "HW Checksumming (%sbound) enabled\n", + cstype == IPA_INBOUND_CHECKSUM ? "in" : "out"); return 0; } -int qeth_set_rx_csum(struct qeth_card *card, int on) +static int qeth_set_ipa_csum(struct qeth_card *card, int on, int cstype) { int rc; if (on) { - rc = qeth_send_checksum_command(card); + rc = qeth_send_checksum_on(card, cstype); if (rc) return -EIO; - dev_info(&card->gdev->dev, - "HW Checksumming (inbound) enabled\n"); } else { - rc = qeth_send_simple_setassparms(card, - IPA_INBOUND_CHECKSUM, IPA_CMD_ASS_STOP, 0); + rc = qeth_send_simple_setassparms(card, cstype, + IPA_CMD_ASS_STOP, 0); if (rc) return -EIO; } return 0; } -EXPORT_SYMBOL_GPL(qeth_set_rx_csum); -int qeth_start_ipa_tx_checksum(struct qeth_card *card) +static int qeth_set_ipa_tso(struct qeth_card *card, int on) { - int rc = 0; + int rc; - if (!qeth_is_supported(card, IPA_OUTBOUND_CHECKSUM)) - return rc; - rc = qeth_send_simple_setassparms(card, IPA_OUTBOUND_CHECKSUM, - IPA_CMD_ASS_START, 0); - if (rc) - goto err_out; - rc = qeth_send_simple_setassparms(card, IPA_OUTBOUND_CHECKSUM, - IPA_CMD_ASS_ENABLE, - card->info.tx_csum_mask); - if (rc) - goto err_out; + QETH_CARD_TEXT(card, 3, "sttso"); - dev_info(&card->gdev->dev, "HW TX Checksumming enabled\n"); - return rc; -err_out: - dev_warn(&card->gdev->dev, "Enabling HW TX checksumming for %s " - "failed, using SW TX checksumming\n", QETH_CARD_IFNAME(card)); + if (on) { + rc = qeth_send_simple_setassparms(card, IPA_OUTBOUND_TSO, + IPA_CMD_ASS_START, 0); + if (rc) { + dev_warn(&card->gdev->dev, + "Starting outbound TCP segmentation offload for %s failed\n", + QETH_CARD_IFNAME(card)); + return -EIO; + } + dev_info(&card->gdev->dev, "Outbound TSO enabled\n"); + } else { + rc = qeth_send_simple_setassparms(card, IPA_OUTBOUND_TSO, + IPA_CMD_ASS_STOP, 0); + } return rc; } -EXPORT_SYMBOL_GPL(qeth_start_ipa_tx_checksum); + +int qeth_set_features(struct net_device *dev, netdev_features_t features) +{ + struct qeth_card *card = dev->ml_priv; + netdev_features_t changed = card->dev->features ^ features; + int rc = 0; + + QETH_DBF_TEXT(SETUP, 2, "setfeat"); + QETH_DBF_HEX(SETUP, 2, &features, sizeof(features)); + + if (card->state == CARD_STATE_DOWN || + card->state == CARD_STATE_RECOVER) + return 0; + + if ((changed & NETIF_F_IP_CSUM)) + rc = qeth_set_ipa_csum(card, + features & NETIF_F_IP_CSUM ? 1 : 0, + IPA_OUTBOUND_CHECKSUM); + if ((changed & NETIF_F_RXCSUM)) + rc |= qeth_set_ipa_csum(card, + features & NETIF_F_RXCSUM ? 1 : 0, + IPA_INBOUND_CHECKSUM); + if ((changed & NETIF_F_TSO)) + rc |= qeth_set_ipa_tso(card, features & NETIF_F_TSO ? 1 : 0); + return rc ? -EIO : 0; +} +EXPORT_SYMBOL_GPL(qeth_set_features); + +netdev_features_t qeth_fix_features(struct net_device *dev, + netdev_features_t features) +{ + struct qeth_card *card = dev->ml_priv; + + QETH_DBF_TEXT(SETUP, 2, "fixfeat"); + if (!qeth_is_supported(card, IPA_OUTBOUND_CHECKSUM)) + features &= ~NETIF_F_IP_CSUM; + if (!qeth_is_supported(card, IPA_INBOUND_CHECKSUM)) + features &= ~NETIF_F_RXCSUM; + if (!qeth_is_supported(card, IPA_OUTBOUND_TSO)) { + features &= ~NETIF_F_TSO; + dev_info(&card->gdev->dev, "Outbound TSO not supported on %s\n", + QETH_CARD_IFNAME(card)); + } + QETH_DBF_HEX(SETUP, 2, &features, sizeof(features)); + return features; +} +EXPORT_SYMBOL_GPL(qeth_fix_features); static int __init qeth_core_init(void) { diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index 2a331d1..928a4ad 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -404,38 +404,6 @@ static int qeth_l2_vlan_rx_kill_vid(struct net_device *dev, return rc; } -static netdev_features_t qeth_l2_fix_features(struct net_device *dev, - netdev_features_t features) -{ - struct qeth_card *card = dev->ml_priv; - - QETH_DBF_TEXT(SETUP, 2, "fixfeat"); - if (!qeth_is_supported(card, IPA_OUTBOUND_CHECKSUM)) - features &= ~NETIF_F_IP_CSUM; - if (!qeth_is_supported(card, IPA_INBOUND_CHECKSUM)) - features &= ~NETIF_F_RXCSUM; - QETH_DBF_HEX(SETUP, 2, &features, sizeof(features)); - return features; -} - -static int qeth_l2_set_features(struct net_device *dev, - netdev_features_t features) -{ - struct qeth_card *card = dev->ml_priv; - netdev_features_t changed = dev->features ^ features; - - QETH_DBF_TEXT(SETUP, 2, "setfeat"); - QETH_DBF_HEX(SETUP, 2, &features, sizeof(features)); - - if (card->state == CARD_STATE_DOWN || - card->state == CARD_STATE_RECOVER) - return 0; - - if (!(changed & NETIF_F_RXCSUM)) - return 0; - return qeth_set_rx_csum(card, features & NETIF_F_RXCSUM ? 1 : 0); -} - static void qeth_l2_stop_card(struct qeth_card *card, int recovery_mode) { QETH_DBF_TEXT(SETUP , 2, "stopcard"); @@ -1112,8 +1080,8 @@ static const struct net_device_ops qeth_l2_netdev_ops = { .ndo_vlan_rx_add_vid = qeth_l2_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = qeth_l2_vlan_rx_kill_vid, .ndo_tx_timeout = qeth_tx_timeout, - .ndo_fix_features = qeth_l2_fix_features, - .ndo_set_features = qeth_l2_set_features + .ndo_fix_features = qeth_fix_features, + .ndo_set_features = qeth_set_features }; static int qeth_l2_setup_netdev(struct qeth_card *card) @@ -1144,10 +1112,14 @@ static int qeth_l2_setup_netdev(struct qeth_card *card) &qeth_l2_ethtool_ops : &qeth_l2_osn_ops; card->dev->features |= NETIF_F_HW_VLAN_CTAG_FILTER; if (card->info.type == QETH_CARD_TYPE_OSD && !card->info.guestlan) { - card->dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM | - NETIF_F_SG; - /* Turn on RX offloading and SG per default */ - card->dev->features |= NETIF_F_RXCSUM | NETIF_F_SG; + card->dev->hw_features = NETIF_F_SG; + /* OSA 3S and earlier has no RX/TX support */ + if (qeth_is_supported(card, IPA_OUTBOUND_CHECKSUM)) + card->dev->hw_features |= NETIF_F_IP_CSUM; + if (qeth_is_supported(card, IPA_INBOUND_CHECKSUM)) + card->dev->hw_features |= NETIF_F_RXCSUM; + /* Turn on SG per default */ + card->dev->features |= NETIF_F_SG; } card->info.broadcast_capable = 1; qeth_l2_request_initial_mac(card); @@ -1165,9 +1137,6 @@ static int qeth_l2_start_ipassists(struct qeth_card *card) /* configure isolation level */ if (qeth_set_access_ctrl_online(card, 0)) return -ENODEV; - if (qeth_is_supported(card, IPA_INBOUND_CHECKSUM)) - qeth_set_rx_csum(card, 1); - qeth_start_ipa_tx_checksum(card); return 0; } @@ -1236,7 +1205,8 @@ static int __qeth_l2_set_online(struct ccwgroup_device *gdev, int recovery_mode) contin: if ((card->info.type == QETH_CARD_TYPE_OSD) || (card->info.type == QETH_CARD_TYPE_OSX)) { - if (qeth_l2_start_ipassists(card)) + rc = qeth_l2_start_ipassists(card); + if (rc) goto out_remove; } diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index 6e7d06c..fc81776 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -909,36 +909,6 @@ static int qeth_l3_setadapter_parms(struct qeth_card *card) return rc; } -static int qeth_l3_default_setassparms_cb(struct qeth_card *card, - struct qeth_reply *reply, unsigned long data) -{ - struct qeth_ipa_cmd *cmd; - - QETH_CARD_TEXT(card, 4, "defadpcb"); - - cmd = (struct qeth_ipa_cmd *) data; - if (cmd->hdr.return_code == 0) { - cmd->hdr.return_code = cmd->data.setassparms.hdr.return_code; - if (cmd->hdr.prot_version == QETH_PROT_IPV4) - card->options.ipa4.enabled_funcs = cmd->hdr.ipa_enabled; - if (cmd->hdr.prot_version == QETH_PROT_IPV6) - card->options.ipa6.enabled_funcs = cmd->hdr.ipa_enabled; - } - if (cmd->data.setassparms.hdr.assist_no == IPA_INBOUND_CHECKSUM && - cmd->data.setassparms.hdr.command_code == IPA_CMD_ASS_START) { - card->info.csum_mask = cmd->data.setassparms.data.flags_32bit; - QETH_CARD_TEXT_(card, 3, "csum:%d", card->info.csum_mask); - } - if (cmd->data.setassparms.hdr.assist_no == IPA_OUTBOUND_CHECKSUM && - cmd->data.setassparms.hdr.command_code == IPA_CMD_ASS_START) { - card->info.tx_csum_mask = - cmd->data.setassparms.data.flags_32bit; - QETH_CARD_TEXT_(card, 3, "tcsu:%d", card->info.tx_csum_mask); - } - - return 0; -} - #ifdef CONFIG_QETH_IPV6 static int qeth_l3_send_simple_setassparms_ipv6(struct qeth_card *card, enum qeth_ipa_funcs ipa_func, __u16 cmd_code) @@ -952,7 +922,7 @@ static int qeth_l3_send_simple_setassparms_ipv6(struct qeth_card *card, if (!iob) return -ENOMEM; rc = qeth_send_setassparms(card, iob, 0, 0, - qeth_l3_default_setassparms_cb, NULL); + qeth_setassparms_cb, NULL); return rc; } #endif @@ -1187,47 +1157,6 @@ out: return rc; } -static void qeth_l3_start_ipa_checksum(struct qeth_card *card) -{ - QETH_CARD_TEXT(card, 3, "strtcsum"); - if (qeth_is_supported(card, IPA_INBOUND_CHECKSUM) - && (card->dev->features & NETIF_F_RXCSUM)) - qeth_set_rx_csum(card, 1); -} - -static void qeth_l3_start_ipa_tx_checksum(struct qeth_card *card) -{ - QETH_CARD_TEXT(card, 3, "strttxcs"); - qeth_start_ipa_tx_checksum(card); -} - -static int qeth_l3_start_ipa_tso(struct qeth_card *card) -{ - int rc; - - QETH_CARD_TEXT(card, 3, "sttso"); - - if (!qeth_is_supported(card, IPA_OUTBOUND_TSO)) { - dev_info(&card->gdev->dev, - "Outbound TSO not supported on %s\n", - QETH_CARD_IFNAME(card)); - rc = -EOPNOTSUPP; - } else { - rc = qeth_send_simple_setassparms(card, IPA_OUTBOUND_TSO, - IPA_CMD_ASS_START, 0); - if (rc) - dev_warn(&card->gdev->dev, "Starting outbound TCP " - "segmentation offload for %s failed\n", - QETH_CARD_IFNAME(card)); - else - dev_info(&card->gdev->dev, - "Outbound TSO enabled\n"); - } - if (rc) - card->dev->features &= ~NETIF_F_TSO; - return rc; -} - static int qeth_l3_start_ipassists(struct qeth_card *card) { QETH_CARD_TEXT(card, 3, "strtipas"); @@ -1241,9 +1170,6 @@ static int qeth_l3_start_ipassists(struct qeth_card *card) qeth_l3_start_ipa_multicast(card); /* go on*/ qeth_l3_start_ipa_ipv6(card); /* go on*/ qeth_l3_start_ipa_broadcast(card); /* go on*/ - qeth_l3_start_ipa_checksum(card); /* go on*/ - qeth_l3_start_ipa_tx_checksum(card); - qeth_l3_start_ipa_tso(card); /* go on*/ return 0; } @@ -2440,7 +2366,7 @@ static int qeth_l3_arp_add_entry(struct qeth_card *card, rc = qeth_send_setassparms(card, iob, sizeof(struct qeth_arp_cache_entry), (unsigned long) entry, - qeth_l3_default_setassparms_cb, NULL); + qeth_setassparms_cb, NULL); if (rc) { tmp = rc; qeth_l3_ipaddr4_to_string((u8 *)entry->ipaddr, buf); @@ -2480,7 +2406,7 @@ static int qeth_l3_arp_remove_entry(struct qeth_card *card, return -ENOMEM; rc = qeth_send_setassparms(card, iob, 12, (unsigned long)buf, - qeth_l3_default_setassparms_cb, NULL); + qeth_setassparms_cb, NULL); if (rc) { tmp = rc; memset(buf, 0, 16); @@ -2826,7 +2752,7 @@ static int qeth_l3_get_elements_no_tso(struct qeth_card *card, int elements = qeth_get_elements_for_range( tcpdptr, (addr_t)skb->data + skb_headlen(skb)) + - qeth_get_elements_for_frags(skb); + qeth_get_elements_for_frags(skb); if ((elements + extra_elems) > QETH_MAX_BUFFER_ELEMENTS(card)) { QETH_DBF_MESSAGE(2, @@ -3089,36 +3015,6 @@ static int qeth_l3_stop(struct net_device *dev) return 0; } -static netdev_features_t qeth_l3_fix_features(struct net_device *dev, - netdev_features_t features) -{ - struct qeth_card *card = dev->ml_priv; - - if (!qeth_is_supported(card, IPA_OUTBOUND_CHECKSUM)) - features &= ~NETIF_F_IP_CSUM; - if (!qeth_is_supported(card, IPA_OUTBOUND_TSO)) - features &= ~NETIF_F_TSO; - if (!qeth_is_supported(card, IPA_INBOUND_CHECKSUM)) - features &= ~NETIF_F_RXCSUM; - return features; -} - -static int qeth_l3_set_features(struct net_device *dev, - netdev_features_t features) -{ - struct qeth_card *card = dev->ml_priv; - netdev_features_t changed = dev->features ^ features; - - if (!(changed & NETIF_F_RXCSUM)) - return 0; - - if (card->state == CARD_STATE_DOWN || - card->state == CARD_STATE_RECOVER) - return 0; - - return qeth_set_rx_csum(card, features & NETIF_F_RXCSUM ? 1 : 0); -} - static const struct ethtool_ops qeth_l3_ethtool_ops = { .get_link = ethtool_op_get_link, .get_strings = qeth_core_get_strings, @@ -3161,8 +3057,8 @@ static const struct net_device_ops qeth_l3_netdev_ops = { .ndo_set_rx_mode = qeth_l3_set_multicast_list, .ndo_do_ioctl = qeth_l3_do_ioctl, .ndo_change_mtu = qeth_change_mtu, - .ndo_fix_features = qeth_l3_fix_features, - .ndo_set_features = qeth_l3_set_features, + .ndo_fix_features = qeth_fix_features, + .ndo_set_features = qeth_set_features, .ndo_vlan_rx_add_vid = qeth_l3_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = qeth_l3_vlan_rx_kill_vid, .ndo_tx_timeout = qeth_tx_timeout, @@ -3177,8 +3073,8 @@ static const struct net_device_ops qeth_l3_osa_netdev_ops = { .ndo_set_rx_mode = qeth_l3_set_multicast_list, .ndo_do_ioctl = qeth_l3_do_ioctl, .ndo_change_mtu = qeth_change_mtu, - .ndo_fix_features = qeth_l3_fix_features, - .ndo_set_features = qeth_l3_set_features, + .ndo_fix_features = qeth_fix_features, + .ndo_set_features = qeth_set_features, .ndo_vlan_rx_add_vid = qeth_l3_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = qeth_l3_vlan_rx_kill_vid, .ndo_tx_timeout = qeth_tx_timeout, @@ -3210,8 +3106,7 @@ static int qeth_l3_setup_netdev(struct qeth_card *card) card->dev->hw_features = NETIF_F_SG | NETIF_F_RXCSUM | NETIF_F_IP_CSUM | NETIF_F_TSO; - card->dev->features = NETIF_F_RXCSUM | - NETIF_F_SG; + card->dev->features = NETIF_F_SG; } } } else if (card->info.type == QETH_CARD_TYPE_IQD) { -- cgit v0.10.2 From 9bdc441102f012b70f51e1ca73b603312fff8b5d Mon Sep 17 00:00:00 2001 From: Hans Wippel Date: Thu, 16 Jun 2016 16:19:00 +0200 Subject: qeth: add network device features for VLAN devices Network device features indicate the capabilities of network devices (e.g., TX checksum offloading and TSO) and their configuration state. Additional network device features (vlan_features) indicate for each network device, which capabilities can be used on VLAN devices, that are configured on the respective base network device. In the current qeth implementation, network device features are only set for the base network devices and not for the VLAN devices. Thus, features like TX checksum offloading cannot be used on VLAN devices. This patch adds network device features to vlan_features, so they can be used by VLAN devices. Signed-off-by: Hans Wippel Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c index 928a4ad..9fd48de 100644 --- a/drivers/s390/net/qeth_l2_main.c +++ b/drivers/s390/net/qeth_l2_main.c @@ -1113,11 +1113,16 @@ static int qeth_l2_setup_netdev(struct qeth_card *card) card->dev->features |= NETIF_F_HW_VLAN_CTAG_FILTER; if (card->info.type == QETH_CARD_TYPE_OSD && !card->info.guestlan) { card->dev->hw_features = NETIF_F_SG; + card->dev->vlan_features = NETIF_F_SG; /* OSA 3S and earlier has no RX/TX support */ - if (qeth_is_supported(card, IPA_OUTBOUND_CHECKSUM)) + if (qeth_is_supported(card, IPA_OUTBOUND_CHECKSUM)) { card->dev->hw_features |= NETIF_F_IP_CSUM; - if (qeth_is_supported(card, IPA_INBOUND_CHECKSUM)) + card->dev->vlan_features |= NETIF_F_IP_CSUM; + } + if (qeth_is_supported(card, IPA_INBOUND_CHECKSUM)) { card->dev->hw_features |= NETIF_F_RXCSUM; + card->dev->vlan_features |= NETIF_F_RXCSUM; + } /* Turn on SG per default */ card->dev->features |= NETIF_F_SG; } diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index fc81776..05f764c 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -3106,6 +3106,9 @@ static int qeth_l3_setup_netdev(struct qeth_card *card) card->dev->hw_features = NETIF_F_SG | NETIF_F_RXCSUM | NETIF_F_IP_CSUM | NETIF_F_TSO; + card->dev->vlan_features = NETIF_F_SG | + NETIF_F_RXCSUM | NETIF_F_IP_CSUM | + NETIF_F_TSO; card->dev->features = NETIF_F_SG; } } -- cgit v0.10.2 From 6c7cd7124493de05d01ea53bdbfcf35918dae4c3 Mon Sep 17 00:00:00 2001 From: Hans Wippel Date: Thu, 16 Jun 2016 16:19:01 +0200 Subject: qeth: improve set_features error handling The function set_features is called to configure network device features on the hardware. If errors occur, the network device features should reflect the changed hardware state and the function should return an error in order to notify the user. In case of an error, the current implementation does not necessarily save the changed hardware state in the network device features before an error is returned. This patch improves error handling by saving features, that could be changed, to the network device features before returning an error. If the device is not running, an additional check in fix_features removes features, that require hardware changes, before they are passed to set_features. Thus, the corresponding check was removed in set_features. Signed-off-by: Hans Wippel Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 19a6ee0..44d3200 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -6125,27 +6125,38 @@ static int qeth_set_ipa_tso(struct qeth_card *card, int on) int qeth_set_features(struct net_device *dev, netdev_features_t features) { struct qeth_card *card = dev->ml_priv; - netdev_features_t changed = card->dev->features ^ features; + netdev_features_t changed = dev->features ^ features; int rc = 0; QETH_DBF_TEXT(SETUP, 2, "setfeat"); QETH_DBF_HEX(SETUP, 2, &features, sizeof(features)); - if (card->state == CARD_STATE_DOWN || - card->state == CARD_STATE_RECOVER) - return 0; - - if ((changed & NETIF_F_IP_CSUM)) + if ((changed & NETIF_F_IP_CSUM)) { rc = qeth_set_ipa_csum(card, features & NETIF_F_IP_CSUM ? 1 : 0, IPA_OUTBOUND_CHECKSUM); - if ((changed & NETIF_F_RXCSUM)) - rc |= qeth_set_ipa_csum(card, + if (rc) + changed ^= NETIF_F_IP_CSUM; + } + if ((changed & NETIF_F_RXCSUM)) { + rc = qeth_set_ipa_csum(card, features & NETIF_F_RXCSUM ? 1 : 0, IPA_INBOUND_CHECKSUM); - if ((changed & NETIF_F_TSO)) - rc |= qeth_set_ipa_tso(card, features & NETIF_F_TSO ? 1 : 0); - return rc ? -EIO : 0; + if (rc) + changed ^= NETIF_F_RXCSUM; + } + if ((changed & NETIF_F_TSO)) { + rc = qeth_set_ipa_tso(card, features & NETIF_F_TSO ? 1 : 0); + if (rc) + changed ^= NETIF_F_TSO; + } + + /* everything changed successfully? */ + if ((dev->features ^ features) == changed) + return 0; + /* something went wrong. save changed features and return error */ + dev->features ^= changed; + return -EIO; } EXPORT_SYMBOL_GPL(qeth_set_features); @@ -6164,6 +6175,11 @@ netdev_features_t qeth_fix_features(struct net_device *dev, dev_info(&card->gdev->dev, "Outbound TSO not supported on %s\n", QETH_CARD_IFNAME(card)); } + /* if the card isn't up, remove features that require hw changes */ + if (card->state == CARD_STATE_DOWN || + card->state == CARD_STATE_RECOVER) + features = features & ~(NETIF_F_IP_CSUM | NETIF_F_RXCSUM | + NETIF_F_TSO); QETH_DBF_HEX(SETUP, 2, &features, sizeof(features)); return features; } -- cgit v0.10.2 From 70deb01662b130748f738525120ef4a68b59c398 Mon Sep 17 00:00:00 2001 From: Hans Wippel Date: Thu, 16 Jun 2016 16:19:02 +0200 Subject: qeth: omit outbound queue 3 for unicast packets in Priority Queuing on HiperSockets On HiperSockets only outbound queues 0 to 2 are available for unicast packets. Current Priority Queuing implementation in the qeth driver puts outgoing packets in outbound queues 0 to 3. This puts outgoing unicast packets into outbound queue 2 instead of outbound queue 3 when using Priority Queuing on a HiperSocket. Additionally, the default outbound queue cannot be set to outbound queue 3 on HiperSockets. Signed-off-by: Hans Wippel Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c index 44d3200..7dba6c8 100644 --- a/drivers/s390/net/qeth_core_main.c +++ b/drivers/s390/net/qeth_core_main.c @@ -3748,6 +3748,14 @@ void qeth_qdio_output_handler(struct ccw_device *ccwdev, } EXPORT_SYMBOL_GPL(qeth_qdio_output_handler); +/* We cannot use outbound queue 3 for unicast packets on HiperSockets */ +static inline int qeth_cut_iqd_prio(struct qeth_card *card, int queue_num) +{ + if ((card->info.type == QETH_CARD_TYPE_IQD) && (queue_num == 3)) + return 2; + return queue_num; +} + /** * Note: Function assumes that we have 4 outbound queues. */ @@ -3775,9 +3783,9 @@ int qeth_get_priority_queue(struct qeth_card *card, struct sk_buff *skb, return card->qdio.default_out_queue; } if (card->qdio.do_prio_queueing == QETH_PRIO_Q_ING_PREC) - return ~tos >> 6 & 3; + return qeth_cut_iqd_prio(card, ~tos >> 6 & 3); if (tos & IPTOS_MINCOST) - return 3; + return qeth_cut_iqd_prio(card, 3); if (tos & IPTOS_RELIABILITY) return 2; if (tos & IPTOS_THROUGHPUT) @@ -3788,11 +3796,12 @@ int qeth_get_priority_queue(struct qeth_card *card, struct sk_buff *skb, case QETH_PRIO_Q_ING_SKB: if (skb->priority > 5) return 0; - return ~skb->priority >> 1 & 3; + return qeth_cut_iqd_prio(card, ~skb->priority >> 1 & 3); case QETH_PRIO_Q_ING_VLAN: tci = &((struct ethhdr *)skb->data)->h_proto; if (*tci == ETH_P_8021Q) - return ~*(tci + 1) >> (VLAN_PRIO_SHIFT + 1) & 3; + return qeth_cut_iqd_prio(card, ~*(tci + 1) >> + (VLAN_PRIO_SHIFT + 1) & 3); break; default: break; diff --git a/drivers/s390/net/qeth_core_sys.c b/drivers/s390/net/qeth_core_sys.c index e6e5b96..75b29fd2 100644 --- a/drivers/s390/net/qeth_core_sys.c +++ b/drivers/s390/net/qeth_core_sys.c @@ -243,6 +243,10 @@ static ssize_t qeth_dev_prioqing_store(struct device *dev, card->qdio.do_prio_queueing = QETH_NO_PRIO_QUEUEING; card->qdio.default_out_queue = 2; } else if (sysfs_streq(buf, "no_prio_queueing:3")) { + if (card->info.type == QETH_CARD_TYPE_IQD) { + rc = -EPERM; + goto out; + } card->qdio.do_prio_queueing = QETH_NO_PRIO_QUEUEING; card->qdio.default_out_queue = 3; } else if (sysfs_streq(buf, "no_prio_queueing")) { -- cgit v0.10.2 From 77a83ed10bcccf67e803d0eab7c8506c7546586a Mon Sep 17 00:00:00 2001 From: Sebastian Ott Date: Thu, 16 Jun 2016 16:19:03 +0200 Subject: s390/qeth: fix indentation in qeth_l3_arp_query gcc-6 warns about obviously wrong indentation: drivers/s390/net/qeth_l3_main.c: In function 'qeth_l3_arp_query': drivers/s390/net/qeth_l3_main.c:2315:3: warning: this 'if' clause does not guard... [-Wmisleading-indentation] if (copy_to_user(udata, qinfo.udata, 4)) ^~ drivers/s390/net/qeth_l3_main.c:2317:4: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the 'if' goto free_and_out; ^~~~ Although this particular case is harmless, fix the indentation to get rid of that warning and improve readability. Signed-off-by: Sebastian Ott Signed-off-by: Ursula Braun Signed-off-by: David S. Miller diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index 05f764c..bcd324e 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -2314,22 +2314,21 @@ static int qeth_l3_arp_query(struct qeth_card *card, char __user *udata) if (rc) { if (copy_to_user(udata, qinfo.udata, 4)) rc = -EFAULT; - goto free_and_out; - } else { + goto free_and_out; + } #ifdef CONFIG_QETH_IPV6 - if (qinfo.mask_bits & QETH_QARP_WITH_IPV6) { - /* fails in case of GuestLAN QDIO mode */ - qeth_l3_query_arp_cache_info(card, QETH_PROT_IPV6, - &qinfo); - } + if (qinfo.mask_bits & QETH_QARP_WITH_IPV6) { + /* fails in case of GuestLAN QDIO mode */ + qeth_l3_query_arp_cache_info(card, QETH_PROT_IPV6, &qinfo); + } #endif - if (copy_to_user(udata, qinfo.udata, qinfo.udata_len)) { - QETH_CARD_TEXT(card, 4, "qactf"); - rc = -EFAULT; - goto free_and_out; - } - QETH_CARD_TEXT(card, 4, "qacts"); + if (copy_to_user(udata, qinfo.udata, qinfo.udata_len)) { + QETH_CARD_TEXT(card, 4, "qactf"); + rc = -EFAULT; + goto free_and_out; } + QETH_CARD_TEXT(card, 4, "qacts"); + free_and_out: kfree(qinfo.udata); out: -- cgit v0.10.2 From 190af10f0b5a07140ec4ce8e6ef04b7cb238dde1 Mon Sep 17 00:00:00 2001 From: Shrikrishna Khare Date: Thu, 16 Jun 2016 10:51:53 -0700 Subject: vmxnet3: prepare for version 3 changes vmxnet3 is currently at version 2, but some command definitions from previous vmxnet3 versions are missing. Add those definitions before moving to version 3. Also, introduce utility macros for vmxnet3 version comparison and update Copyright information and Maintained by. Signed-off-by: Shrikrishna Khare Signed-off-by: David S. Miller diff --git a/drivers/net/vmxnet3/Makefile b/drivers/net/vmxnet3/Makefile index 880f509..8cdbb63 100644 --- a/drivers/net/vmxnet3/Makefile +++ b/drivers/net/vmxnet3/Makefile @@ -2,7 +2,7 @@ # # Linux driver for VMware's vmxnet3 ethernet NIC. # -# Copyright (C) 2007-2009, VMware, Inc. All Rights Reserved. +# Copyright (C) 2007-2016, VMware, Inc. All Rights Reserved. # # This program is free software; you can redistribute it and/or modify it # under the terms of the GNU General Public License as published by the @@ -21,7 +21,7 @@ # The full GNU General Public License is included in this distribution in # the file called "COPYING". # -# Maintained by: Shreyas Bhatewara +# Maintained by: pv-drivers@vmware.com # # ################################################################################ diff --git a/drivers/net/vmxnet3/upt1_defs.h b/drivers/net/vmxnet3/upt1_defs.h index 969c751..db9f1fd 100644 --- a/drivers/net/vmxnet3/upt1_defs.h +++ b/drivers/net/vmxnet3/upt1_defs.h @@ -1,7 +1,7 @@ /* * Linux driver for VMware's vmxnet3 ethernet NIC. * - * Copyright (C) 2008-2009, VMware, Inc. All Rights Reserved. + * Copyright (C) 2008-2016, VMware, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the @@ -20,7 +20,7 @@ * The full GNU General Public License is included in this distribution in * the file called "COPYING". * - * Maintained by: Shreyas Bhatewara + * Maintained by: pv-drivers@vmware.com * */ diff --git a/drivers/net/vmxnet3/vmxnet3_defs.h b/drivers/net/vmxnet3/vmxnet3_defs.h index 72ba8ae..8345e0c 100644 --- a/drivers/net/vmxnet3/vmxnet3_defs.h +++ b/drivers/net/vmxnet3/vmxnet3_defs.h @@ -1,7 +1,7 @@ /* * Linux driver for VMware's vmxnet3 ethernet NIC. * - * Copyright (C) 2008-2015, VMware, Inc. All Rights Reserved. + * Copyright (C) 2008-2016, VMware, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the @@ -20,7 +20,7 @@ * The full GNU General Public License is included in this distribution in * the file called "COPYING". * - * Maintained by: Shreyas Bhatewara + * Maintained by: pv-drivers@vmware.com * */ @@ -76,7 +76,9 @@ enum { VMXNET3_CMD_UPDATE_IML, VMXNET3_CMD_UPDATE_PMCFG, VMXNET3_CMD_UPDATE_FEATURE, + VMXNET3_CMD_RESERVED1, VMXNET3_CMD_LOAD_PLUGIN, + VMXNET3_CMD_RESERVED2, VMXNET3_CMD_FIRST_GET = 0xF00D0000, VMXNET3_CMD_GET_QUEUE_STATUS = VMXNET3_CMD_FIRST_GET, @@ -87,7 +89,8 @@ enum { VMXNET3_CMD_GET_DID_LO, VMXNET3_CMD_GET_DID_HI, VMXNET3_CMD_GET_DEV_EXTRA_INFO, - VMXNET3_CMD_GET_CONF_INTR + VMXNET3_CMD_GET_CONF_INTR, + VMXNET3_CMD_GET_RESERVED1, }; /* diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c index 08885bc..507c53d 100644 --- a/drivers/net/vmxnet3/vmxnet3_drv.c +++ b/drivers/net/vmxnet3/vmxnet3_drv.c @@ -1,7 +1,7 @@ /* * Linux driver for VMware's vmxnet3 ethernet NIC. * - * Copyright (C) 2008-2009, VMware, Inc. All Rights Reserved. + * Copyright (C) 2008-2016, VMware, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the @@ -20,7 +20,7 @@ * The full GNU General Public License is included in this distribution in * the file called "COPYING". * - * Maintained by: Shreyas Bhatewara + * Maintained by: pv-drivers@vmware.com * */ @@ -1363,7 +1363,7 @@ vmxnet3_rq_rx_complete(struct vmxnet3_rx_queue *rq, rbi->dma_addr = new_dma_addr; rxd->addr = cpu_to_le64(rbi->dma_addr); rxd->len = rbi->len; - if (adapter->version == 2 && + if (VMXNET3_VERSION_GE_2(adapter) && rcd->type == VMXNET3_CDTYPE_RXCOMP_LRO) { struct Vmxnet3_RxCompDescExt *rcdlro; rcdlro = (struct Vmxnet3_RxCompDescExt *)rcd; @@ -3200,12 +3200,16 @@ vmxnet3_probe_device(struct pci_dev *pdev, goto err_alloc_pci; ver = VMXNET3_READ_BAR1_REG(adapter, VMXNET3_REG_VRRS); - if (ver & 2) { - VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_VRRS, 2); - adapter->version = 2; - } else if (ver & 1) { - VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_VRRS, 1); - adapter->version = 1; + if (ver & (1 << VMXNET3_REV_2)) { + VMXNET3_WRITE_BAR1_REG(adapter, + VMXNET3_REG_VRRS, + 1 << VMXNET3_REV_2); + adapter->version = VMXNET3_REV_2 + 1; + } else if (ver & (1 << VMXNET3_REV_1)) { + VMXNET3_WRITE_BAR1_REG(adapter, + VMXNET3_REG_VRRS, + 1 << VMXNET3_REV_1); + adapter->version = VMXNET3_REV_1 + 1; } else { dev_err(&pdev->dev, "Incompatible h/w version (0x%x) for adapter\n", ver); diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c index 9ba11d7..163e99c 100644 --- a/drivers/net/vmxnet3/vmxnet3_ethtool.c +++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c @@ -1,7 +1,7 @@ /* * Linux driver for VMware's vmxnet3 ethernet NIC. * - * Copyright (C) 2008-2009, VMware, Inc. All Rights Reserved. + * Copyright (C) 2008-2016, VMware, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the @@ -20,7 +20,7 @@ * The full GNU General Public License is included in this distribution in * the file called "COPYING". * - * Maintained by: Shreyas Bhatewara + * Maintained by: pv-drivers@vmware.com * */ diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h index 3d2b64e..de068e9 100644 --- a/drivers/net/vmxnet3/vmxnet3_int.h +++ b/drivers/net/vmxnet3/vmxnet3_int.h @@ -1,7 +1,7 @@ /* * Linux driver for VMware's vmxnet3 ethernet NIC. * - * Copyright (C) 2008-2009, VMware, Inc. All Rights Reserved. + * Copyright (C) 2008-2016, VMware, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the @@ -20,7 +20,7 @@ * The full GNU General Public License is included in this distribution in * the file called "COPYING". * - * Maintained by: Shreyas Bhatewara + * Maintained by: pv-drivers@vmware.com * */ @@ -79,6 +79,10 @@ #define VMXNET3_RSS #endif +#define VMXNET3_REV_3 2 /* Vmxnet3 Rev. 3 */ +#define VMXNET3_REV_2 1 /* Vmxnet3 Rev. 2 */ +#define VMXNET3_REV_1 0 /* Vmxnet3 Rev. 1 */ + /* * Capabilities */ @@ -387,6 +391,11 @@ struct vmxnet3_adapter { #define VMXNET3_GET_ADDR_LO(dma) ((u32)(dma)) #define VMXNET3_GET_ADDR_HI(dma) ((u32)(((u64)(dma)) >> 32)) +#define VMXNET3_VERSION_GE_2(adapter) \ + (adapter->version >= VMXNET3_REV_2 + 1) +#define VMXNET3_VERSION_GE_3(adapter) \ + (adapter->version >= VMXNET3_REV_3 + 1) + /* must be a multiple of VMXNET3_RING_SIZE_ALIGN */ #define VMXNET3_DEF_TX_RING_SIZE 512 #define VMXNET3_DEF_RX_RING_SIZE 256 -- cgit v0.10.2 From f35c7480f81b70f9c3030d96a3807e8faba34cf7 Mon Sep 17 00:00:00 2001 From: Shrikrishna Khare Date: Thu, 16 Jun 2016 10:51:54 -0700 Subject: vmxnet3: introduce generalized command interface to configure the device Shared memory is used to exchange information between the vmxnet3 driver and the emulation. In order to request emulation to perform a task, the driver first populates specific fields in this shared memory and then issues corresponding command by writing to the command register(CMD). The layout of the shared memory was defined by vmxnet3 version 1 and cannot be extended for every new command without breaking backward compatibility. To address this problem, in vmxnet3 version 3, the emulation repurposed a reserved field in the shared memory to represent command information instead. For new commands, the driver first populates the command information field in the shared memory and then issues the command. The emulation interprets the data written to the command information depending on the type of the command. This patch exposes this capability to the driver. Signed-off-by: Guolin Yang Signed-off-by: Shrikrishna Khare Signed-off-by: David S. Miller diff --git a/drivers/net/vmxnet3/vmxnet3_defs.h b/drivers/net/vmxnet3/vmxnet3_defs.h index 8345e0c..a26a69d 100644 --- a/drivers/net/vmxnet3/vmxnet3_defs.h +++ b/drivers/net/vmxnet3/vmxnet3_defs.h @@ -79,6 +79,7 @@ enum { VMXNET3_CMD_RESERVED1, VMXNET3_CMD_LOAD_PLUGIN, VMXNET3_CMD_RESERVED2, + VMXNET3_CMD_RESERVED3, VMXNET3_CMD_FIRST_GET = 0xF00D0000, VMXNET3_CMD_GET_QUEUE_STATUS = VMXNET3_CMD_FIRST_GET, @@ -612,6 +613,18 @@ struct Vmxnet3_RxQueueDesc { u8 __pad[88]; /* 128 aligned */ }; +struct Vmxnet3_SetPolling { + u8 enablePolling; +}; + +/* If the command data <= 16 bytes, use the shared memory directly. + * otherwise, use variable length configuration descriptor. + */ +union Vmxnet3_CmdInfo { + struct Vmxnet3_VariableLenConfDesc varConf; + struct Vmxnet3_SetPolling setPolling; + __le64 data[2]; +}; struct Vmxnet3_DSDevRead { /* read-only region for device, read by dev in response to a SET cmd */ @@ -630,7 +643,14 @@ struct Vmxnet3_DriverShared { __le32 pad; struct Vmxnet3_DSDevRead devRead; __le32 ecr; - __le32 reserved[5]; + __le32 reserved; + union { + __le32 reserved1[4]; + union Vmxnet3_CmdInfo cmdInfo; /* only valid in the context of + * executing the relevant + * command + */ + } cu; }; -- cgit v0.10.2 From 3c8b3efc061a745d888869dc3462ac4f7dd582d9 Mon Sep 17 00:00:00 2001 From: Shrikrishna Khare Date: Thu, 16 Jun 2016 10:51:55 -0700 Subject: vmxnet3: allow variable length transmit data ring buffer vmxnet3 driver supports transmit data ring viz. a set of fixed size buffers used by the driver to copy packet headers. Small packets that fit these buffers are copied into these buffers entirely. Currently this buffer size of fixed at 128 bytes. This patch extends transmit data ring implementation to allow variable length transmit data ring buffers. The length of the buffer is read from the emulation during initialization. Signed-off-by: Sriram Rangarajan Signed-off-by: Shrikrishna Khare Signed-off-by: David S. Miller diff --git a/drivers/net/vmxnet3/vmxnet3_defs.h b/drivers/net/vmxnet3/vmxnet3_defs.h index a26a69d..701d989 100644 --- a/drivers/net/vmxnet3/vmxnet3_defs.h +++ b/drivers/net/vmxnet3/vmxnet3_defs.h @@ -92,6 +92,7 @@ enum { VMXNET3_CMD_GET_DEV_EXTRA_INFO, VMXNET3_CMD_GET_CONF_INTR, VMXNET3_CMD_GET_RESERVED1, + VMXNET3_CMD_GET_TXDATA_DESC_SIZE }; /* @@ -377,6 +378,10 @@ union Vmxnet3_GenericDesc { #define VMXNET3_RING_SIZE_ALIGN 32 #define VMXNET3_RING_SIZE_MASK (VMXNET3_RING_SIZE_ALIGN - 1) +/* Tx Data Ring buffer size must be a multiple of 64 */ +#define VMXNET3_TXDATA_DESC_SIZE_ALIGN 64 +#define VMXNET3_TXDATA_DESC_SIZE_MASK (VMXNET3_TXDATA_DESC_SIZE_ALIGN - 1) + /* Max ring size */ #define VMXNET3_TX_RING_MAX_SIZE 4096 #define VMXNET3_TC_RING_MAX_SIZE 4096 @@ -384,6 +389,9 @@ union Vmxnet3_GenericDesc { #define VMXNET3_RX_RING2_MAX_SIZE 4096 #define VMXNET3_RC_RING_MAX_SIZE 8192 +#define VMXNET3_TXDATA_DESC_MIN_SIZE 128 +#define VMXNET3_TXDATA_DESC_MAX_SIZE 2048 + /* a list of reasons for queue stop */ enum { @@ -470,7 +478,9 @@ struct Vmxnet3_TxQueueConf { __le32 compRingSize; /* # of comp desc */ __le32 ddLen; /* size of driver data */ u8 intrIdx; - u8 _pad[7]; + u8 _pad1[1]; + __le16 txDataRingDescSize; + u8 _pad2[4]; }; diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c index 507c53d..4e42eb0 100644 --- a/drivers/net/vmxnet3/vmxnet3_drv.c +++ b/drivers/net/vmxnet3/vmxnet3_drv.c @@ -435,8 +435,8 @@ vmxnet3_tq_destroy(struct vmxnet3_tx_queue *tq, tq->tx_ring.base = NULL; } if (tq->data_ring.base) { - dma_free_coherent(&adapter->pdev->dev, tq->data_ring.size * - sizeof(struct Vmxnet3_TxDataDesc), + dma_free_coherent(&adapter->pdev->dev, + tq->data_ring.size * tq->txdata_desc_size, tq->data_ring.base, tq->data_ring.basePA); tq->data_ring.base = NULL; } @@ -478,8 +478,8 @@ vmxnet3_tq_init(struct vmxnet3_tx_queue *tq, tq->tx_ring.next2fill = tq->tx_ring.next2comp = 0; tq->tx_ring.gen = VMXNET3_INIT_GEN; - memset(tq->data_ring.base, 0, tq->data_ring.size * - sizeof(struct Vmxnet3_TxDataDesc)); + memset(tq->data_ring.base, 0, + tq->data_ring.size * tq->txdata_desc_size); /* reset the tx comp ring contents to 0 and reset comp ring states */ memset(tq->comp_ring.base, 0, tq->comp_ring.size * @@ -514,10 +514,10 @@ vmxnet3_tq_create(struct vmxnet3_tx_queue *tq, } tq->data_ring.base = dma_alloc_coherent(&adapter->pdev->dev, - tq->data_ring.size * sizeof(struct Vmxnet3_TxDataDesc), + tq->data_ring.size * tq->txdata_desc_size, &tq->data_ring.basePA, GFP_KERNEL); if (!tq->data_ring.base) { - netdev_err(adapter->netdev, "failed to allocate data ring\n"); + netdev_err(adapter->netdev, "failed to allocate tx data ring\n"); goto err; } @@ -689,7 +689,7 @@ vmxnet3_map_pkt(struct sk_buff *skb, struct vmxnet3_tx_ctx *ctx, if (ctx->copy_size) { ctx->sop_txd->txd.addr = cpu_to_le64(tq->data_ring.basePA + tq->tx_ring.next2fill * - sizeof(struct Vmxnet3_TxDataDesc)); + tq->txdata_desc_size); ctx->sop_txd->dword[2] = cpu_to_le32(dw2 | ctx->copy_size); ctx->sop_txd->dword[3] = 0; @@ -873,8 +873,9 @@ vmxnet3_parse_hdr(struct sk_buff *skb, struct vmxnet3_tx_queue *tq, ctx->eth_ip_hdr_size = 0; ctx->l4_hdr_size = 0; /* copy as much as allowed */ - ctx->copy_size = min((unsigned int)VMXNET3_HDR_COPY_SIZE - , skb_headlen(skb)); + ctx->copy_size = min_t(unsigned int, + tq->txdata_desc_size, + skb_headlen(skb)); } if (skb->len <= VMXNET3_HDR_COPY_SIZE) @@ -885,7 +886,7 @@ vmxnet3_parse_hdr(struct sk_buff *skb, struct vmxnet3_tx_queue *tq, goto err; } - if (unlikely(ctx->copy_size > VMXNET3_HDR_COPY_SIZE)) { + if (unlikely(ctx->copy_size > tq->txdata_desc_size)) { tq->stats.oversized_hdr++; ctx->copy_size = 0; return 0; @@ -2336,6 +2337,7 @@ vmxnet3_setup_driver_shared(struct vmxnet3_adapter *adapter) tqc->ddPA = cpu_to_le64(tq->buf_info_pa); tqc->txRingSize = cpu_to_le32(tq->tx_ring.size); tqc->dataRingSize = cpu_to_le32(tq->data_ring.size); + tqc->txDataRingDescSize = cpu_to_le32(tq->txdata_desc_size); tqc->compRingSize = cpu_to_le32(tq->comp_ring.size); tqc->ddLen = cpu_to_le32( sizeof(struct vmxnet3_tx_buf_info) * @@ -2689,7 +2691,8 @@ vmxnet3_adjust_rx_ring_size(struct vmxnet3_adapter *adapter) int vmxnet3_create_queues(struct vmxnet3_adapter *adapter, u32 tx_ring_size, - u32 rx_ring_size, u32 rx_ring2_size) + u32 rx_ring_size, u32 rx_ring2_size, + u16 txdata_desc_size) { int err = 0, i; @@ -2698,6 +2701,7 @@ vmxnet3_create_queues(struct vmxnet3_adapter *adapter, u32 tx_ring_size, tq->tx_ring.size = tx_ring_size; tq->data_ring.size = tx_ring_size; tq->comp_ring.size = tx_ring_size; + tq->txdata_desc_size = txdata_desc_size; tq->shared = &adapter->tqd_start[i].ctrl; tq->stopped = true; tq->adapter = adapter; @@ -2754,9 +2758,34 @@ vmxnet3_open(struct net_device *netdev) for (i = 0; i < adapter->num_tx_queues; i++) spin_lock_init(&adapter->tx_queue[i].tx_lock); - err = vmxnet3_create_queues(adapter, adapter->tx_ring_size, + if (VMXNET3_VERSION_GE_3(adapter)) { + unsigned long flags; + u16 txdata_desc_size; + + spin_lock_irqsave(&adapter->cmd_lock, flags); + VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_CMD, + VMXNET3_CMD_GET_TXDATA_DESC_SIZE); + txdata_desc_size = VMXNET3_READ_BAR1_REG(adapter, + VMXNET3_REG_CMD); + spin_unlock_irqrestore(&adapter->cmd_lock, flags); + + if ((txdata_desc_size < VMXNET3_TXDATA_DESC_MIN_SIZE) || + (txdata_desc_size > VMXNET3_TXDATA_DESC_MAX_SIZE) || + (txdata_desc_size & VMXNET3_TXDATA_DESC_SIZE_MASK)) { + adapter->txdata_desc_size = + sizeof(struct Vmxnet3_TxDataDesc); + } else { + adapter->txdata_desc_size = txdata_desc_size; + } + } else { + adapter->txdata_desc_size = sizeof(struct Vmxnet3_TxDataDesc); + } + + err = vmxnet3_create_queues(adapter, + adapter->tx_ring_size, adapter->rx_ring_size, - adapter->rx_ring2_size); + adapter->rx_ring2_size, + adapter->txdata_desc_size); if (err) goto queue_err; diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c index 163e99c..3b70cfe 100644 --- a/drivers/net/vmxnet3/vmxnet3_ethtool.c +++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c @@ -396,8 +396,7 @@ vmxnet3_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *p) buf[j++] = VMXNET3_GET_ADDR_LO(tq->data_ring.basePA); buf[j++] = VMXNET3_GET_ADDR_HI(tq->data_ring.basePA); buf[j++] = tq->data_ring.size; - /* transmit data ring buffer size */ - buf[j++] = VMXNET3_HDR_COPY_SIZE; + buf[j++] = tq->txdata_desc_size; buf[j++] = VMXNET3_GET_ADDR_LO(tq->comp_ring.basePA); buf[j++] = VMXNET3_GET_ADDR_HI(tq->comp_ring.basePA); @@ -591,7 +590,8 @@ vmxnet3_set_ringparam(struct net_device *netdev, vmxnet3_rq_destroy_all(adapter); err = vmxnet3_create_queues(adapter, new_tx_ring_size, - new_rx_ring_size, new_rx_ring2_size); + new_rx_ring_size, new_rx_ring2_size, + adapter->txdata_desc_size); if (err) { /* failed, most likely because of OOM, try default @@ -604,7 +604,8 @@ vmxnet3_set_ringparam(struct net_device *netdev, err = vmxnet3_create_queues(adapter, new_tx_ring_size, new_rx_ring_size, - new_rx_ring2_size); + new_rx_ring2_size, + adapter->txdata_desc_size); if (err) { netdev_err(netdev, "failed to create queues " "with default sizes. Closing it\n"); diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h index de068e9..94010de 100644 --- a/drivers/net/vmxnet3/vmxnet3_int.h +++ b/drivers/net/vmxnet3/vmxnet3_int.h @@ -241,6 +241,7 @@ struct vmxnet3_tx_queue { int num_stop; /* # of times the queue is * stopped */ int qid; + u16 txdata_desc_size; } __attribute__((__aligned__(SMP_CACHE_BYTES))); enum vmxnet3_rx_buf_type { @@ -363,6 +364,9 @@ struct vmxnet3_adapter { u32 rx_ring_size; u32 rx_ring2_size; + /* Size of buffer in the data ring */ + u16 txdata_desc_size; + struct work_struct work; unsigned long state; /* VMXNET3_STATE_BIT_xxx */ @@ -427,7 +431,8 @@ vmxnet3_set_features(struct net_device *netdev, netdev_features_t features); int vmxnet3_create_queues(struct vmxnet3_adapter *adapter, - u32 tx_ring_size, u32 rx_ring_size, u32 rx_ring2_size); + u32 tx_ring_size, u32 rx_ring_size, u32 rx_ring2_size, + u16 txdata_desc_size); void vmxnet3_set_ethtool_ops(struct net_device *netdev); -- cgit v0.10.2 From 50a5ce3e7116a70edb7a1d1d209e3bc537752427 Mon Sep 17 00:00:00 2001 From: Shrikrishna Khare Date: Thu, 16 Jun 2016 10:51:56 -0700 Subject: vmxnet3: add receive data ring support vmxnet3 driver preallocates buffers for receiving packets and posts the buffers to the emulation. In order to deliver a received packet to the guest, the emulation must map buffer(s) and copy the packet into it. To avoid this memory mapping overhead, this patch introduces the receive data ring - a set of small sized buffers that are always mapped by the emulation. If a packet fits into the receive data ring buffer, the emulation delivers the packet via the receive data ring (which must be copied by the guest driver), or else the usual receive path is used. Receive Data Ring buffer length is configurable via ethtool -G ethX rx-mini Signed-off-by: Shrikrishna Khare Signed-off-by: David S. Miller diff --git a/drivers/net/vmxnet3/vmxnet3_defs.h b/drivers/net/vmxnet3/vmxnet3_defs.h index 701d989..f3b31c2 100644 --- a/drivers/net/vmxnet3/vmxnet3_defs.h +++ b/drivers/net/vmxnet3/vmxnet3_defs.h @@ -174,6 +174,8 @@ struct Vmxnet3_TxDataDesc { u8 data[VMXNET3_HDR_COPY_SIZE]; }; +typedef u8 Vmxnet3_RxDataDesc; + #define VMXNET3_TCD_GEN_SHIFT 31 #define VMXNET3_TCD_GEN_SIZE 1 #define VMXNET3_TCD_TXIDX_SHIFT 0 @@ -382,6 +384,10 @@ union Vmxnet3_GenericDesc { #define VMXNET3_TXDATA_DESC_SIZE_ALIGN 64 #define VMXNET3_TXDATA_DESC_SIZE_MASK (VMXNET3_TXDATA_DESC_SIZE_ALIGN - 1) +/* Rx Data Ring buffer size must be a multiple of 64 */ +#define VMXNET3_RXDATA_DESC_SIZE_ALIGN 64 +#define VMXNET3_RXDATA_DESC_SIZE_MASK (VMXNET3_RXDATA_DESC_SIZE_ALIGN - 1) + /* Max ring size */ #define VMXNET3_TX_RING_MAX_SIZE 4096 #define VMXNET3_TC_RING_MAX_SIZE 4096 @@ -392,6 +398,8 @@ union Vmxnet3_GenericDesc { #define VMXNET3_TXDATA_DESC_MIN_SIZE 128 #define VMXNET3_TXDATA_DESC_MAX_SIZE 2048 +#define VMXNET3_RXDATA_DESC_MAX_SIZE 2048 + /* a list of reasons for queue stop */ enum { @@ -488,12 +496,14 @@ struct Vmxnet3_RxQueueConf { __le64 rxRingBasePA[2]; __le64 compRingBasePA; __le64 ddPA; /* driver data */ - __le64 reserved; + __le64 rxDataRingBasePA; __le32 rxRingSize[2]; /* # of rx desc */ __le32 compRingSize; /* # of rx comp desc */ __le32 ddLen; /* size of driver data */ u8 intrIdx; - u8 _pad[7]; + u8 _pad1[1]; + __le16 rxDataRingDescSize; /* size of rx data ring buffer */ + u8 _pad2[4]; }; diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c index 4e42eb0..6449d2e 100644 --- a/drivers/net/vmxnet3/vmxnet3_drv.c +++ b/drivers/net/vmxnet3/vmxnet3_drv.c @@ -1284,9 +1284,10 @@ vmxnet3_rq_rx_complete(struct vmxnet3_rx_queue *rq, */ break; } - BUG_ON(rcd->rqID != rq->qid && rcd->rqID != rq->qid2); + BUG_ON(rcd->rqID != rq->qid && rcd->rqID != rq->qid2 && + rcd->rqID != rq->dataRingQid); idx = rcd->rxdIdx; - ring_idx = rcd->rqID < adapter->num_rx_queues ? 0 : 1; + ring_idx = VMXNET3_GET_RING_IDX(adapter, rcd->rqID); ring = rq->rx_ring + ring_idx; vmxnet3_getRxDesc(rxd, &rq->rx_ring[ring_idx].base[idx].rxd, &rxCmdDesc); @@ -1301,8 +1302,12 @@ vmxnet3_rq_rx_complete(struct vmxnet3_rx_queue *rq, } if (rcd->sop) { /* first buf of the pkt */ + bool rxDataRingUsed; + u16 len; + BUG_ON(rxd->btype != VMXNET3_RXD_BTYPE_HEAD || - rcd->rqID != rq->qid); + (rcd->rqID != rq->qid && + rcd->rqID != rq->dataRingQid)); BUG_ON(rbi->buf_type != VMXNET3_RX_BUF_SKB); BUG_ON(ctx->skb != NULL || rbi->skb == NULL); @@ -1318,8 +1323,12 @@ vmxnet3_rq_rx_complete(struct vmxnet3_rx_queue *rq, skip_page_frags = false; ctx->skb = rbi->skb; + + rxDataRingUsed = + VMXNET3_RX_DATA_RING(adapter, rcd->rqID); + len = rxDataRingUsed ? rcd->len : rbi->len; new_skb = netdev_alloc_skb_ip_align(adapter->netdev, - rbi->len); + len); if (new_skb == NULL) { /* Skb allocation failed, do not handover this * skb to stack. Reuse it. Drop the existing pkt @@ -1330,25 +1339,48 @@ vmxnet3_rq_rx_complete(struct vmxnet3_rx_queue *rq, skip_page_frags = true; goto rcd_done; } - new_dma_addr = dma_map_single(&adapter->pdev->dev, - new_skb->data, rbi->len, - PCI_DMA_FROMDEVICE); - if (dma_mapping_error(&adapter->pdev->dev, - new_dma_addr)) { - dev_kfree_skb(new_skb); - /* Skb allocation failed, do not handover this - * skb to stack. Reuse it. Drop the existing pkt - */ - rq->stats.rx_buf_alloc_failure++; - ctx->skb = NULL; - rq->stats.drop_total++; - skip_page_frags = true; - goto rcd_done; - } - dma_unmap_single(&adapter->pdev->dev, rbi->dma_addr, - rbi->len, - PCI_DMA_FROMDEVICE); + if (rxDataRingUsed) { + size_t sz; + + BUG_ON(rcd->len > rq->data_ring.desc_size); + + ctx->skb = new_skb; + sz = rcd->rxdIdx * rq->data_ring.desc_size; + memcpy(new_skb->data, + &rq->data_ring.base[sz], rcd->len); + } else { + ctx->skb = rbi->skb; + + new_dma_addr = + dma_map_single(&adapter->pdev->dev, + new_skb->data, rbi->len, + PCI_DMA_FROMDEVICE); + if (dma_mapping_error(&adapter->pdev->dev, + new_dma_addr)) { + dev_kfree_skb(new_skb); + /* Skb allocation failed, do not + * handover this skb to stack. Reuse + * it. Drop the existing pkt. + */ + rq->stats.rx_buf_alloc_failure++; + ctx->skb = NULL; + rq->stats.drop_total++; + skip_page_frags = true; + goto rcd_done; + } + + dma_unmap_single(&adapter->pdev->dev, + rbi->dma_addr, + rbi->len, + PCI_DMA_FROMDEVICE); + + /* Immediate refill */ + rbi->skb = new_skb; + rbi->dma_addr = new_dma_addr; + rxd->addr = cpu_to_le64(rbi->dma_addr); + rxd->len = rbi->len; + } #ifdef VMXNET3_RSS if (rcd->rssType != VMXNET3_RCD_RSS_TYPE_NONE && @@ -1359,11 +1391,6 @@ vmxnet3_rq_rx_complete(struct vmxnet3_rx_queue *rq, #endif skb_put(ctx->skb, rcd->len); - /* Immediate refill */ - rbi->skb = new_skb; - rbi->dma_addr = new_dma_addr; - rxd->addr = cpu_to_le64(rbi->dma_addr); - rxd->len = rbi->len; if (VMXNET3_VERSION_GE_2(adapter) && rcd->type == VMXNET3_CDTYPE_RXCOMP_LRO) { struct Vmxnet3_RxCompDescExt *rcdlro; @@ -1590,6 +1617,13 @@ static void vmxnet3_rq_destroy(struct vmxnet3_rx_queue *rq, rq->buf_info[i] = NULL; } + if (rq->data_ring.base) { + dma_free_coherent(&adapter->pdev->dev, + rq->rx_ring[0].size * rq->data_ring.desc_size, + rq->data_ring.base, rq->data_ring.basePA); + rq->data_ring.base = NULL; + } + if (rq->comp_ring.base) { dma_free_coherent(&adapter->pdev->dev, rq->comp_ring.size * sizeof(struct Vmxnet3_RxCompDesc), @@ -1605,6 +1639,25 @@ static void vmxnet3_rq_destroy(struct vmxnet3_rx_queue *rq, } } +void +vmxnet3_rq_destroy_all_rxdataring(struct vmxnet3_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_rx_queues; i++) { + struct vmxnet3_rx_queue *rq = &adapter->rx_queue[i]; + + if (rq->data_ring.base) { + dma_free_coherent(&adapter->pdev->dev, + (rq->rx_ring[0].size * + rq->data_ring.desc_size), + rq->data_ring.base, + rq->data_ring.basePA); + rq->data_ring.base = NULL; + rq->data_ring.desc_size = 0; + } + } +} static int vmxnet3_rq_init(struct vmxnet3_rx_queue *rq, @@ -1698,6 +1751,22 @@ vmxnet3_rq_create(struct vmxnet3_rx_queue *rq, struct vmxnet3_adapter *adapter) } } + if ((adapter->rxdataring_enabled) && (rq->data_ring.desc_size != 0)) { + sz = rq->rx_ring[0].size * rq->data_ring.desc_size; + rq->data_ring.base = + dma_alloc_coherent(&adapter->pdev->dev, sz, + &rq->data_ring.basePA, + GFP_KERNEL); + if (!rq->data_ring.base) { + netdev_err(adapter->netdev, + "rx data ring will be disabled\n"); + adapter->rxdataring_enabled = false; + } + } else { + rq->data_ring.base = NULL; + rq->data_ring.desc_size = 0; + } + sz = rq->comp_ring.size * sizeof(struct Vmxnet3_RxCompDesc); rq->comp_ring.base = dma_alloc_coherent(&adapter->pdev->dev, sz, &rq->comp_ring.basePA, @@ -1730,6 +1799,8 @@ vmxnet3_rq_create_all(struct vmxnet3_adapter *adapter) { int i, err = 0; + adapter->rxdataring_enabled = VMXNET3_VERSION_GE_3(adapter); + for (i = 0; i < adapter->num_rx_queues; i++) { err = vmxnet3_rq_create(&adapter->rx_queue[i], adapter); if (unlikely(err)) { @@ -1739,6 +1810,10 @@ vmxnet3_rq_create_all(struct vmxnet3_adapter *adapter) goto err_out; } } + + if (!adapter->rxdataring_enabled) + vmxnet3_rq_destroy_all_rxdataring(adapter); + return err; err_out: vmxnet3_rq_destroy_all(adapter); @@ -2046,10 +2121,9 @@ vmxnet3_request_irqs(struct vmxnet3_adapter *adapter) struct vmxnet3_rx_queue *rq = &adapter->rx_queue[i]; rq->qid = i; rq->qid2 = i + adapter->num_rx_queues; + rq->dataRingQid = i + 2 * adapter->num_rx_queues; } - - /* init our intr settings */ for (i = 0; i < intr->num_intrs; i++) intr->mod_levels[i] = UPT1_IML_ADAPTIVE; @@ -2362,6 +2436,12 @@ vmxnet3_setup_driver_shared(struct vmxnet3_adapter *adapter) (rqc->rxRingSize[0] + rqc->rxRingSize[1])); rqc->intrIdx = rq->comp_ring.intr_idx; + if (VMXNET3_VERSION_GE_3(adapter)) { + rqc->rxDataRingBasePA = + cpu_to_le64(rq->data_ring.basePA); + rqc->rxDataRingDescSize = + cpu_to_le16(rq->data_ring.desc_size); + } } #ifdef VMXNET3_RSS @@ -2692,7 +2772,7 @@ vmxnet3_adjust_rx_ring_size(struct vmxnet3_adapter *adapter) int vmxnet3_create_queues(struct vmxnet3_adapter *adapter, u32 tx_ring_size, u32 rx_ring_size, u32 rx_ring2_size, - u16 txdata_desc_size) + u16 txdata_desc_size, u16 rxdata_desc_size) { int err = 0, i; @@ -2718,12 +2798,15 @@ vmxnet3_create_queues(struct vmxnet3_adapter *adapter, u32 tx_ring_size, adapter->rx_queue[0].rx_ring[0].size = rx_ring_size; adapter->rx_queue[0].rx_ring[1].size = rx_ring2_size; vmxnet3_adjust_rx_ring_size(adapter); + + adapter->rxdataring_enabled = VMXNET3_VERSION_GE_3(adapter); for (i = 0; i < adapter->num_rx_queues; i++) { struct vmxnet3_rx_queue *rq = &adapter->rx_queue[i]; /* qid and qid2 for rx queues will be assigned later when num * of rx queues is finalized after allocating intrs */ rq->shared = &adapter->rqd_start[i].ctrl; rq->adapter = adapter; + rq->data_ring.desc_size = rxdata_desc_size; err = vmxnet3_rq_create(rq, adapter); if (err) { if (i == 0) { @@ -2741,6 +2824,10 @@ vmxnet3_create_queues(struct vmxnet3_adapter *adapter, u32 tx_ring_size, } } } + + if (!adapter->rxdataring_enabled) + vmxnet3_rq_destroy_all_rxdataring(adapter); + return err; queue_err: vmxnet3_tq_destroy_all(adapter); @@ -2785,7 +2872,8 @@ vmxnet3_open(struct net_device *netdev) adapter->tx_ring_size, adapter->rx_ring_size, adapter->rx_ring2_size, - adapter->txdata_desc_size); + adapter->txdata_desc_size, + adapter->rxdata_desc_size); if (err) goto queue_err; @@ -3260,6 +3348,9 @@ vmxnet3_probe_device(struct pci_dev *pdev, SET_NETDEV_DEV(netdev, &pdev->dev); vmxnet3_declare_features(adapter, dma64); + adapter->rxdata_desc_size = VMXNET3_VERSION_GE_3(adapter) ? + VMXNET3_DEF_RXDATA_DESC_SIZE : 0; + if (adapter->num_tx_queues == adapter->num_rx_queues) adapter->share_intr = VMXNET3_INTR_BUDDYSHARE; else diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c index 3b70cfe..38f7c79 100644 --- a/drivers/net/vmxnet3/vmxnet3_ethtool.c +++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c @@ -430,11 +430,10 @@ vmxnet3_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *p) buf[j++] = rq->rx_ring[1].next2comp; buf[j++] = rq->rx_ring[1].gen; - /* receive data ring */ - buf[j++] = 0; - buf[j++] = 0; - buf[j++] = 0; - buf[j++] = 0; + buf[j++] = VMXNET3_GET_ADDR_LO(rq->data_ring.basePA); + buf[j++] = VMXNET3_GET_ADDR_HI(rq->data_ring.basePA); + buf[j++] = rq->rx_ring[0].size; + buf[j++] = rq->data_ring.desc_size; buf[j++] = VMXNET3_GET_ADDR_LO(rq->comp_ring.basePA); buf[j++] = VMXNET3_GET_ADDR_HI(rq->comp_ring.basePA); @@ -503,12 +502,14 @@ vmxnet3_get_ringparam(struct net_device *netdev, param->rx_max_pending = VMXNET3_RX_RING_MAX_SIZE; param->tx_max_pending = VMXNET3_TX_RING_MAX_SIZE; - param->rx_mini_max_pending = 0; + param->rx_mini_max_pending = VMXNET3_VERSION_GE_3(adapter) ? + VMXNET3_RXDATA_DESC_MAX_SIZE : 0; param->rx_jumbo_max_pending = VMXNET3_RX_RING2_MAX_SIZE; param->rx_pending = adapter->rx_ring_size; param->tx_pending = adapter->tx_ring_size; - param->rx_mini_pending = 0; + param->rx_mini_pending = VMXNET3_VERSION_GE_3(adapter) ? + adapter->rxdata_desc_size : 0; param->rx_jumbo_pending = adapter->rx_ring2_size; } @@ -519,6 +520,7 @@ vmxnet3_set_ringparam(struct net_device *netdev, { struct vmxnet3_adapter *adapter = netdev_priv(netdev); u32 new_tx_ring_size, new_rx_ring_size, new_rx_ring2_size; + u16 new_rxdata_desc_size; u32 sz; int err = 0; @@ -541,6 +543,15 @@ vmxnet3_set_ringparam(struct net_device *netdev, return -EOPNOTSUPP; } + if (VMXNET3_VERSION_GE_3(adapter)) { + if (param->rx_mini_pending < 0 || + param->rx_mini_pending > VMXNET3_RXDATA_DESC_MAX_SIZE) { + return -EINVAL; + } + } else if (param->rx_mini_pending != 0) { + return -EINVAL; + } + /* round it up to a multiple of VMXNET3_RING_SIZE_ALIGN */ new_tx_ring_size = (param->tx_pending + VMXNET3_RING_SIZE_MASK) & ~VMXNET3_RING_SIZE_MASK; @@ -567,9 +578,19 @@ vmxnet3_set_ringparam(struct net_device *netdev, new_rx_ring2_size = min_t(u32, new_rx_ring2_size, VMXNET3_RX_RING2_MAX_SIZE); + /* rx data ring buffer size has to be a multiple of + * VMXNET3_RXDATA_DESC_SIZE_ALIGN + */ + new_rxdata_desc_size = + (param->rx_mini_pending + VMXNET3_RXDATA_DESC_SIZE_MASK) & + ~VMXNET3_RXDATA_DESC_SIZE_MASK; + new_rxdata_desc_size = min_t(u16, new_rxdata_desc_size, + VMXNET3_RXDATA_DESC_MAX_SIZE); + if (new_tx_ring_size == adapter->tx_ring_size && new_rx_ring_size == adapter->rx_ring_size && - new_rx_ring2_size == adapter->rx_ring2_size) { + new_rx_ring2_size == adapter->rx_ring2_size && + new_rxdata_desc_size == adapter->rxdata_desc_size) { return 0; } @@ -591,8 +612,8 @@ vmxnet3_set_ringparam(struct net_device *netdev, err = vmxnet3_create_queues(adapter, new_tx_ring_size, new_rx_ring_size, new_rx_ring2_size, - adapter->txdata_desc_size); - + adapter->txdata_desc_size, + new_rxdata_desc_size); if (err) { /* failed, most likely because of OOM, try default * size */ @@ -601,11 +622,15 @@ vmxnet3_set_ringparam(struct net_device *netdev, new_rx_ring_size = VMXNET3_DEF_RX_RING_SIZE; new_rx_ring2_size = VMXNET3_DEF_RX_RING2_SIZE; new_tx_ring_size = VMXNET3_DEF_TX_RING_SIZE; + new_rxdata_desc_size = VMXNET3_VERSION_GE_3(adapter) ? + VMXNET3_DEF_RXDATA_DESC_SIZE : 0; + err = vmxnet3_create_queues(adapter, new_tx_ring_size, new_rx_ring_size, new_rx_ring2_size, - adapter->txdata_desc_size); + adapter->txdata_desc_size, + new_rxdata_desc_size); if (err) { netdev_err(netdev, "failed to create queues " "with default sizes. Closing it\n"); @@ -621,6 +646,7 @@ vmxnet3_set_ringparam(struct net_device *netdev, adapter->tx_ring_size = new_tx_ring_size; adapter->rx_ring_size = new_rx_ring_size; adapter->rx_ring2_size = new_rx_ring2_size; + adapter->rxdata_desc_size = new_rxdata_desc_size; out: clear_bit(VMXNET3_STATE_BIT_RESETTING, &adapter->state); diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h index 94010de..c46bf09 100644 --- a/drivers/net/vmxnet3/vmxnet3_int.h +++ b/drivers/net/vmxnet3/vmxnet3_int.h @@ -272,15 +272,23 @@ struct vmxnet3_rq_driver_stats { u64 rx_buf_alloc_failure; }; +struct vmxnet3_rx_data_ring { + Vmxnet3_RxDataDesc *base; + dma_addr_t basePA; + u16 desc_size; +}; + struct vmxnet3_rx_queue { char name[IFNAMSIZ + 8]; /* To identify interrupt */ struct vmxnet3_adapter *adapter; struct napi_struct napi; struct vmxnet3_cmd_ring rx_ring[2]; + struct vmxnet3_rx_data_ring data_ring; struct vmxnet3_comp_ring comp_ring; struct vmxnet3_rx_ctx rx_ctx; u32 qid; /* rqID in RCD for buffer from 1st ring */ u32 qid2; /* rqID in RCD for buffer from 2nd ring */ + u32 dataRingQid; /* rqID in RCD for buffer from data ring */ struct vmxnet3_rx_buf_info *buf_info[2]; dma_addr_t buf_info_pa; struct Vmxnet3_RxQueueCtrl *shared; @@ -366,6 +374,9 @@ struct vmxnet3_adapter { /* Size of buffer in the data ring */ u16 txdata_desc_size; + u16 rxdata_desc_size; + + bool rxdataring_enabled; struct work_struct work; @@ -405,9 +416,19 @@ struct vmxnet3_adapter { #define VMXNET3_DEF_RX_RING_SIZE 256 #define VMXNET3_DEF_RX_RING2_SIZE 128 +#define VMXNET3_DEF_RXDATA_DESC_SIZE 128 + #define VMXNET3_MAX_ETH_HDR_SIZE 22 #define VMXNET3_MAX_SKB_BUF_SIZE (3*1024) +#define VMXNET3_GET_RING_IDX(adapter, rqID) \ + ((rqID >= adapter->num_rx_queues && \ + rqID < 2 * adapter->num_rx_queues) ? 1 : 0) \ + +#define VMXNET3_RX_DATA_RING(adapter, rqID) \ + (rqID >= 2 * adapter->num_rx_queues && \ + rqID < 3 * adapter->num_rx_queues) \ + int vmxnet3_quiesce_dev(struct vmxnet3_adapter *adapter); @@ -432,7 +453,7 @@ vmxnet3_set_features(struct net_device *netdev, netdev_features_t features); int vmxnet3_create_queues(struct vmxnet3_adapter *adapter, u32 tx_ring_size, u32 rx_ring_size, u32 rx_ring2_size, - u16 txdata_desc_size); + u16 txdata_desc_size, u16 rxdata_desc_size); void vmxnet3_set_ethtool_ops(struct net_device *netdev); -- cgit v0.10.2 From 4edef40ef5f8d09a0b1ded4d1d9b0e988cd98e97 Mon Sep 17 00:00:00 2001 From: Shrikrishna Khare Date: Thu, 16 Jun 2016 10:51:57 -0700 Subject: vmxnet3: add support for get_coalesce, set_coalesce ethtool operations The emulation supports a variety of coalescing modes viz. disabled (no coalescing), adaptive, static (number of packets to batch before raising an interrupt), rate based (number of interrupts per second). This patch implements get_coalesce and set_coalesce methods to allow querying and configuring different coalescing modes. Signed-off-by: Keyong Sun Signed-off-by: Manoj Tammali Signed-off-by: Shrikrishna Khare Signed-off-by: David S. Miller diff --git a/drivers/net/vmxnet3/vmxnet3_defs.h b/drivers/net/vmxnet3/vmxnet3_defs.h index f3b31c2..274e145 100644 --- a/drivers/net/vmxnet3/vmxnet3_defs.h +++ b/drivers/net/vmxnet3/vmxnet3_defs.h @@ -80,6 +80,7 @@ enum { VMXNET3_CMD_LOAD_PLUGIN, VMXNET3_CMD_RESERVED2, VMXNET3_CMD_RESERVED3, + VMXNET3_CMD_SET_COALESCE, VMXNET3_CMD_FIRST_GET = 0xF00D0000, VMXNET3_CMD_GET_QUEUE_STATUS = VMXNET3_CMD_FIRST_GET, @@ -92,7 +93,8 @@ enum { VMXNET3_CMD_GET_DEV_EXTRA_INFO, VMXNET3_CMD_GET_CONF_INTR, VMXNET3_CMD_GET_RESERVED1, - VMXNET3_CMD_GET_TXDATA_DESC_SIZE + VMXNET3_CMD_GET_TXDATA_DESC_SIZE, + VMXNET3_CMD_GET_COALESCE, }; /* @@ -637,6 +639,35 @@ struct Vmxnet3_SetPolling { u8 enablePolling; }; +#define VMXNET3_COAL_STATIC_MAX_DEPTH 128 +#define VMXNET3_COAL_RBC_MIN_RATE 100 +#define VMXNET3_COAL_RBC_MAX_RATE 100000 + +enum Vmxnet3_CoalesceMode { + VMXNET3_COALESCE_DISABLED = 0, + VMXNET3_COALESCE_ADAPT = 1, + VMXNET3_COALESCE_STATIC = 2, + VMXNET3_COALESCE_RBC = 3 +}; + +struct Vmxnet3_CoalesceRbc { + u32 rbc_rate; +}; + +struct Vmxnet3_CoalesceStatic { + u32 tx_depth; + u32 tx_comp_depth; + u32 rx_depth; +}; + +struct Vmxnet3_CoalesceScheme { + enum Vmxnet3_CoalesceMode coalMode; + union { + struct Vmxnet3_CoalesceRbc coalRbc; + struct Vmxnet3_CoalesceStatic coalStatic; + } coalPara; +}; + /* If the command data <= 16 bytes, use the shared memory directly. * otherwise, use variable length configuration descriptor. */ diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c index 6449d2e..d0bcc1d9 100644 --- a/drivers/net/vmxnet3/vmxnet3_drv.c +++ b/drivers/net/vmxnet3/vmxnet3_drv.c @@ -2491,6 +2491,32 @@ vmxnet3_setup_driver_shared(struct vmxnet3_adapter *adapter) /* the rest are already zeroed */ } +static void +vmxnet3_init_coalesce(struct vmxnet3_adapter *adapter) +{ + struct Vmxnet3_DriverShared *shared = adapter->shared; + union Vmxnet3_CmdInfo *cmdInfo = &shared->cu.cmdInfo; + unsigned long flags; + + if (!VMXNET3_VERSION_GE_3(adapter)) + return; + + spin_lock_irqsave(&adapter->cmd_lock, flags); + cmdInfo->varConf.confVer = 1; + cmdInfo->varConf.confLen = + cpu_to_le32(sizeof(*adapter->coal_conf)); + cmdInfo->varConf.confPA = cpu_to_le64(adapter->coal_conf_pa); + + if (adapter->default_coal_mode) { + VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_CMD, + VMXNET3_CMD_GET_COALESCE); + } else { + VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_CMD, + VMXNET3_CMD_SET_COALESCE); + } + + spin_unlock_irqrestore(&adapter->cmd_lock, flags); +} int vmxnet3_activate_dev(struct vmxnet3_adapter *adapter) @@ -2540,6 +2566,8 @@ vmxnet3_activate_dev(struct vmxnet3_adapter *adapter) goto activate_err; } + vmxnet3_init_coalesce(adapter); + for (i = 0; i < adapter->num_rx_queues; i++) { VMXNET3_WRITE_BAR0_REG(adapter, VMXNET3_REG_RXPROD + i * VMXNET3_REG_ALIGN, @@ -3345,6 +3373,22 @@ vmxnet3_probe_device(struct pci_dev *pdev, goto err_ver; } + if (VMXNET3_VERSION_GE_3(adapter)) { + adapter->coal_conf = + dma_alloc_coherent(&adapter->pdev->dev, + sizeof(struct Vmxnet3_CoalesceScheme) + , + &adapter->coal_conf_pa, + GFP_KERNEL); + if (!adapter->coal_conf) { + err = -ENOMEM; + goto err_ver; + } + memset(adapter->coal_conf, 0, sizeof(*adapter->coal_conf)); + adapter->coal_conf->coalMode = VMXNET3_COALESCE_DISABLED; + adapter->default_coal_mode = true; + } + SET_NETDEV_DEV(netdev, &pdev->dev); vmxnet3_declare_features(adapter, dma64); @@ -3407,6 +3451,11 @@ vmxnet3_probe_device(struct pci_dev *pdev, return 0; err_register: + if (VMXNET3_VERSION_GE_3(adapter)) { + dma_free_coherent(&adapter->pdev->dev, + sizeof(struct Vmxnet3_CoalesceScheme), + adapter->coal_conf, adapter->coal_conf_pa); + } vmxnet3_free_intr_resources(adapter); err_ver: vmxnet3_free_pci_resources(adapter); @@ -3457,6 +3506,11 @@ vmxnet3_remove_device(struct pci_dev *pdev) vmxnet3_free_intr_resources(adapter); vmxnet3_free_pci_resources(adapter); + if (VMXNET3_VERSION_GE_3(adapter)) { + dma_free_coherent(&adapter->pdev->dev, + sizeof(struct Vmxnet3_CoalesceScheme), + adapter->coal_conf, adapter->coal_conf_pa); + } #ifdef VMXNET3_RSS dma_free_coherent(&adapter->pdev->dev, sizeof(struct UPT1_RSSConf), adapter->rss_conf, adapter->rss_conf_pa); diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c index 38f7c79..aabc6ef 100644 --- a/drivers/net/vmxnet3/vmxnet3_ethtool.c +++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c @@ -725,6 +725,162 @@ vmxnet3_set_rss(struct net_device *netdev, const u32 *p, const u8 *key, } #endif +static int +vmxnet3_get_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec) +{ + struct vmxnet3_adapter *adapter = netdev_priv(netdev); + + if (!VMXNET3_VERSION_GE_3(adapter)) + return -EOPNOTSUPP; + + switch (adapter->coal_conf->coalMode) { + case VMXNET3_COALESCE_DISABLED: + /* struct ethtool_coalesce is already initialized to 0 */ + break; + case VMXNET3_COALESCE_ADAPT: + ec->use_adaptive_rx_coalesce = true; + break; + case VMXNET3_COALESCE_STATIC: + ec->tx_max_coalesced_frames = + adapter->coal_conf->coalPara.coalStatic.tx_comp_depth; + ec->rx_max_coalesced_frames = + adapter->coal_conf->coalPara.coalStatic.rx_depth; + break; + case VMXNET3_COALESCE_RBC: { + u32 rbc_rate; + + rbc_rate = adapter->coal_conf->coalPara.coalRbc.rbc_rate; + ec->rx_coalesce_usecs = VMXNET3_COAL_RBC_USECS(rbc_rate); + } + break; + default: + return -EOPNOTSUPP; + } + + return 0; +} + +static int +vmxnet3_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *ec) +{ + struct vmxnet3_adapter *adapter = netdev_priv(netdev); + struct Vmxnet3_DriverShared *shared = adapter->shared; + union Vmxnet3_CmdInfo *cmdInfo = &shared->cu.cmdInfo; + unsigned long flags; + + if (!VMXNET3_VERSION_GE_3(adapter)) + return -EOPNOTSUPP; + + if (ec->rx_coalesce_usecs_irq || + ec->rx_max_coalesced_frames_irq || + ec->tx_coalesce_usecs || + ec->tx_coalesce_usecs_irq || + ec->tx_max_coalesced_frames_irq || + ec->stats_block_coalesce_usecs || + ec->use_adaptive_tx_coalesce || + ec->pkt_rate_low || + ec->rx_coalesce_usecs_low || + ec->rx_max_coalesced_frames_low || + ec->tx_coalesce_usecs_low || + ec->tx_max_coalesced_frames_low || + ec->pkt_rate_high || + ec->rx_coalesce_usecs_high || + ec->rx_max_coalesced_frames_high || + ec->tx_coalesce_usecs_high || + ec->tx_max_coalesced_frames_high || + ec->rate_sample_interval) { + return -EINVAL; + } + + if ((ec->rx_coalesce_usecs == 0) && + (ec->use_adaptive_rx_coalesce == 0) && + (ec->tx_max_coalesced_frames == 0) && + (ec->rx_max_coalesced_frames == 0)) { + memset(adapter->coal_conf, 0, sizeof(*adapter->coal_conf)); + adapter->coal_conf->coalMode = VMXNET3_COALESCE_DISABLED; + goto done; + } + + if (ec->rx_coalesce_usecs != 0) { + u32 rbc_rate; + + if ((ec->use_adaptive_rx_coalesce != 0) || + (ec->tx_max_coalesced_frames != 0) || + (ec->rx_max_coalesced_frames != 0)) { + return -EINVAL; + } + + rbc_rate = VMXNET3_COAL_RBC_RATE(ec->rx_coalesce_usecs); + if (rbc_rate < VMXNET3_COAL_RBC_MIN_RATE || + rbc_rate > VMXNET3_COAL_RBC_MAX_RATE) { + return -EINVAL; + } + + memset(adapter->coal_conf, 0, sizeof(*adapter->coal_conf)); + adapter->coal_conf->coalMode = VMXNET3_COALESCE_RBC; + adapter->coal_conf->coalPara.coalRbc.rbc_rate = rbc_rate; + goto done; + } + + if (ec->use_adaptive_rx_coalesce != 0) { + if ((ec->rx_coalesce_usecs != 0) || + (ec->tx_max_coalesced_frames != 0) || + (ec->rx_max_coalesced_frames != 0)) { + return -EINVAL; + } + memset(adapter->coal_conf, 0, sizeof(*adapter->coal_conf)); + adapter->coal_conf->coalMode = VMXNET3_COALESCE_ADAPT; + goto done; + } + + if ((ec->tx_max_coalesced_frames != 0) || + (ec->rx_max_coalesced_frames != 0)) { + if ((ec->rx_coalesce_usecs != 0) || + (ec->use_adaptive_rx_coalesce != 0)) { + return -EINVAL; + } + + if ((ec->tx_max_coalesced_frames > + VMXNET3_COAL_STATIC_MAX_DEPTH) || + (ec->rx_max_coalesced_frames > + VMXNET3_COAL_STATIC_MAX_DEPTH)) { + return -EINVAL; + } + + memset(adapter->coal_conf, 0, sizeof(*adapter->coal_conf)); + adapter->coal_conf->coalMode = VMXNET3_COALESCE_STATIC; + + adapter->coal_conf->coalPara.coalStatic.tx_comp_depth = + (ec->tx_max_coalesced_frames ? + ec->tx_max_coalesced_frames : + VMXNET3_COAL_STATIC_DEFAULT_DEPTH); + + adapter->coal_conf->coalPara.coalStatic.rx_depth = + (ec->rx_max_coalesced_frames ? + ec->rx_max_coalesced_frames : + VMXNET3_COAL_STATIC_DEFAULT_DEPTH); + + adapter->coal_conf->coalPara.coalStatic.tx_depth = + VMXNET3_COAL_STATIC_DEFAULT_DEPTH; + goto done; + } + +done: + adapter->default_coal_mode = false; + if (netif_running(netdev)) { + spin_lock_irqsave(&adapter->cmd_lock, flags); + cmdInfo->varConf.confVer = 1; + cmdInfo->varConf.confLen = + cpu_to_le32(sizeof(*adapter->coal_conf)); + cmdInfo->varConf.confPA = cpu_to_le64(adapter->coal_conf_pa); + VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_CMD, + VMXNET3_CMD_SET_COALESCE); + spin_unlock_irqrestore(&adapter->cmd_lock, flags); + } + + return 0; +} + static const struct ethtool_ops vmxnet3_ethtool_ops = { .get_settings = vmxnet3_get_settings, .get_drvinfo = vmxnet3_get_drvinfo, @@ -733,6 +889,8 @@ static const struct ethtool_ops vmxnet3_ethtool_ops = { .get_wol = vmxnet3_get_wol, .set_wol = vmxnet3_set_wol, .get_link = ethtool_op_get_link, + .get_coalesce = vmxnet3_get_coalesce, + .set_coalesce = vmxnet3_set_coalesce, .get_strings = vmxnet3_get_strings, .get_sset_count = vmxnet3_get_sset_count, .get_ethtool_stats = vmxnet3_get_ethtool_stats, diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h index c46bf09..63df4f2 100644 --- a/drivers/net/vmxnet3/vmxnet3_int.h +++ b/drivers/net/vmxnet3/vmxnet3_int.h @@ -358,6 +358,7 @@ struct vmxnet3_adapter { int rx_buf_per_pkt; /* only apply to the 1st ring */ dma_addr_t shared_pa; dma_addr_t queue_desc_pa; + dma_addr_t coal_conf_pa; /* Wake-on-LAN */ u32 wol; @@ -384,6 +385,9 @@ struct vmxnet3_adapter { int share_intr; + struct Vmxnet3_CoalesceScheme *coal_conf; + bool default_coal_mode; + dma_addr_t adapter_pa; dma_addr_t pm_conf_pa; dma_addr_t rss_conf_pa; @@ -429,6 +433,11 @@ struct vmxnet3_adapter { (rqID >= 2 * adapter->num_rx_queues && \ rqID < 3 * adapter->num_rx_queues) \ +#define VMXNET3_COAL_STATIC_DEFAULT_DEPTH 64 + +#define VMXNET3_COAL_RBC_RATE(usecs) (1000000 / usecs) +#define VMXNET3_COAL_RBC_USECS(rbc_rate) (1000000 / rbc_rate) + int vmxnet3_quiesce_dev(struct vmxnet3_adapter *adapter); -- cgit v0.10.2 From 474432229f9482f0f4a2732f2e130dc48247f1d7 Mon Sep 17 00:00:00 2001 From: Shrikrishna Khare Date: Thu, 16 Jun 2016 10:51:58 -0700 Subject: vmxnet3: introduce command to register memory region In vmxnet3 version 3, the emulation added support for the vmxnet3 driver to communicate information about the memory regions the driver will use for rx/tx buffers. The driver can also indicate which rx/tx queue the memory region is applicable for. If this information is communicated to the emulation, the emulation will always keep these memory regions mapped, thereby avoiding the mapping/unmapping overhead for every packet. Currently, Linux vmxnet3 driver does not leverage this capability. The feasibility of using this approach for the Linux vmxnet3 driver will be investigated independently and if possible, will be part of a different patch. This patch only exposes the emulation capability to the driver (vmxnet3_defs.h is identical between the driver and the emulation). Signed-off-by: Guolin Yang Signed-off-by: Shrikrishna Khare Signed-off-by: David S. Miller diff --git a/drivers/net/vmxnet3/vmxnet3_defs.h b/drivers/net/vmxnet3/vmxnet3_defs.h index 274e145..c3a3164 100644 --- a/drivers/net/vmxnet3/vmxnet3_defs.h +++ b/drivers/net/vmxnet3/vmxnet3_defs.h @@ -81,6 +81,7 @@ enum { VMXNET3_CMD_RESERVED2, VMXNET3_CMD_RESERVED3, VMXNET3_CMD_SET_COALESCE, + VMXNET3_CMD_REGISTER_MEMREGS, VMXNET3_CMD_FIRST_GET = 0xF00D0000, VMXNET3_CMD_GET_QUEUE_STATUS = VMXNET3_CMD_FIRST_GET, @@ -668,6 +669,22 @@ struct Vmxnet3_CoalesceScheme { } coalPara; }; +struct Vmxnet3_MemoryRegion { + __le64 startPA; + __le32 length; + __le16 txQueueBits; + __le16 rxQueueBits; +}; + +#define MAX_MEMORY_REGION_PER_QUEUE 16 +#define MAX_MEMORY_REGION_PER_DEVICE 256 + +struct Vmxnet3_MemRegs { + __le16 numRegs; + __le16 pad[3]; + struct Vmxnet3_MemoryRegion memRegs[1]; +}; + /* If the command data <= 16 bytes, use the shared memory directly. * otherwise, use variable length configuration descriptor. */ -- cgit v0.10.2 From 6af9d787459e3ea32da446fe0aa76cff65fd2f8c Mon Sep 17 00:00:00 2001 From: Shrikrishna Khare Date: Thu, 16 Jun 2016 10:51:59 -0700 Subject: vmxnet3: update to version 3 With all vmxnet3 version 3 changes incorporated in the vmxnet3 driver, the driver can configure emulation to run at vmxnet3 version 3, provided the emulation advertises support for version 3. Signed-off-by: Shrikrishna Khare Signed-off-by: David S. Miller diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c index d0bcc1d9..c68fe49 100644 --- a/drivers/net/vmxnet3/vmxnet3_drv.c +++ b/drivers/net/vmxnet3/vmxnet3_drv.c @@ -3345,7 +3345,12 @@ vmxnet3_probe_device(struct pci_dev *pdev, goto err_alloc_pci; ver = VMXNET3_READ_BAR1_REG(adapter, VMXNET3_REG_VRRS); - if (ver & (1 << VMXNET3_REV_2)) { + if (ver & (1 << VMXNET3_REV_3)) { + VMXNET3_WRITE_BAR1_REG(adapter, + VMXNET3_REG_VRRS, + 1 << VMXNET3_REV_3); + adapter->version = VMXNET3_REV_3 + 1; + } else if (ver & (1 << VMXNET3_REV_2)) { VMXNET3_WRITE_BAR1_REG(adapter, VMXNET3_REG_VRRS, 1 << VMXNET3_REV_2); diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h index 63df4f2..74fc030 100644 --- a/drivers/net/vmxnet3/vmxnet3_int.h +++ b/drivers/net/vmxnet3/vmxnet3_int.h @@ -69,10 +69,10 @@ /* * Version numbers */ -#define VMXNET3_DRIVER_VERSION_STRING "1.4.8.0-k" +#define VMXNET3_DRIVER_VERSION_STRING "1.4.9.0-k" /* a 32-bit int, each byte encode a verion number in VMXNET3_DRIVER_VERSION */ -#define VMXNET3_DRIVER_VERSION_NUM 0x01040800 +#define VMXNET3_DRIVER_VERSION_NUM 0x01040900 #if defined(CONFIG_PCI_MSI) /* RSS only makes sense if MSI-X is supported. */ -- cgit v0.10.2 From 9f6ed032cd951d2427995578c51af1b50c054efa Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 16 Jun 2016 23:19:29 +0200 Subject: net, cls: also reject deleting all filters when TCA_KIND present When we check for RTM_DELTFILTER, we should also reject the request for deleting all filters under a given parent when TCA_KIND attribute is present. If present, it's currently just ignored but there's also no point to let it pass in the first place either since this doesn't have any meaning with wild-card removal. Fixes: ea7f8277f907 ("net, cls: allow for deleting all filters for given parent") Signed-off-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index cca1ef5..843a716 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -169,7 +169,7 @@ replay: if (prio == 0) { switch (n->nlmsg_type) { case RTM_DELTFILTER: - if (protocol || t->tcm_handle) + if (protocol || t->tcm_handle || tca[TCA_KIND]) return -ENOENT; break; case RTM_NEWTFILTER: -- cgit v0.10.2 From 3636876a5680f2816a99807cd997b142ed24fcdb Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 22 May 2016 11:05:46 +0200 Subject: MAINTAINERS: Add file patterns for can device tree bindings Submitters of device tree binding documentation may forget to CC the subsystem maintainer if this is missing. Signed-off-by: Geert Uytterhoeven Cc: Wolfgang Grandegger Cc: Marc Kleine-Budde Cc: linux-can@vger.kernel.org Signed-off-by: Marc Kleine-Budde diff --git a/MAINTAINERS b/MAINTAINERS index 0e26025..50f69ba 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2814,6 +2814,7 @@ W: https://github.com/linux-can T: git git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can.git T: git git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next.git S: Maintained +F: Documentation/devicetree/bindings/net/can/ F: drivers/net/can/ F: include/linux/can/dev.h F: include/linux/can/platform/ -- cgit v0.10.2 From a20fadf85312f7e999c7279af3e038e4f3539fbf Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Mon, 23 May 2016 15:13:00 +0200 Subject: can: build proc support only if CONFIG_PROC_FS is activated When building can subsystem with CONFIG_PROC_FS=n I detected some unused variables warning by using proc functions. In CAN the proc handling is nicely placed in one object file. This patch adds simple add a dependency on CONFIG_PROC_FS for CAN's proc.o file and corresponding static inline no-op functions. Signed-off-by: Alexander Aring Acked-by: Oliver Hartkopp [mkl: provide static inline noops instead of using #ifdefs] Signed-off-by: Marc Kleine-Budde diff --git a/net/can/Makefile b/net/can/Makefile index cef49eb..1093675 100644 --- a/net/can/Makefile +++ b/net/can/Makefile @@ -3,7 +3,8 @@ # obj-$(CONFIG_CAN) += can.o -can-y := af_can.o proc.o +can-y := af_can.o +can-$(CONFIG_PROC_FS) += proc.o obj-$(CONFIG_CAN_RAW) += can-raw.o can-raw-y := raw.o diff --git a/net/can/af_can.h b/net/can/af_can.h index fca0fe9..38a79ff 100644 --- a/net/can/af_can.h +++ b/net/can/af_can.h @@ -113,8 +113,19 @@ struct s_pstats { extern struct dev_rcv_lists can_rx_alldev_list; /* function prototypes for the CAN networklayer procfs (proc.c) */ +#ifdef CONFIG_PROC_FS void can_init_proc(void); void can_remove_proc(void); +#else +static inline void can_init_proc(void) +{ + pr_info("can: Can't create /proc/net/can. CONFIG_PROC_FS missing!\n"); +} + +static inline void can_remove_proc(void) +{ +} +#endif void can_stat_update(unsigned long data); /* structures and variables from af_can.c needed in proc.c for reading */ diff --git a/net/can/proc.c b/net/can/proc.c index 1a19b98..85ef7bb 100644 --- a/net/can/proc.c +++ b/net/can/proc.c @@ -517,8 +517,7 @@ void can_init_proc(void) can_dir = proc_mkdir("can", init_net.proc_net); if (!can_dir) { - printk(KERN_INFO "can: failed to create /proc/net/can . " - "CONFIG_PROC_FS missing?\n"); + pr_info("can: failed to create /proc/net/can.\n"); return; } -- cgit v0.10.2 From 7da29f97d6c816f5f95f964f4ced0dae14968cce Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Mon, 9 May 2016 14:13:06 +0200 Subject: can: dev: can-calc-bit-timing(): better sample point calculation This patch optimizes the calculation of the sample point. To understand what it does have a look at the original implementation. If there is a combination of timing parameters where both the bitrate and sample point error are 0 the current implementation will find it. However if the reference clock doesn't allow an optimal bitrate (this means the bitrate error is always != 0) there might be several timing parameter combinations having the same bitrate error. The original implementation will allways choose the one with the highest brp. The actual sample point error isn't taken into account. This patch changes the algorithm to minimize the sample point error, too. Now a brp/tseg combination is accepted as better if one of these condition are fulfilled: 1) the bit rate error must be smaller, or 2) the bit rate error must be equal and the sample point error must be equal or smaller If a smaller bit rate error is found the sample point error is reset. This ensures that we first optimize for small bit rate error and then for small sample point errors. Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/dev.c b/drivers/net/can/dev.c index 910c12e..7188137 100644 --- a/drivers/net/can/dev.c +++ b/drivers/net/can/dev.c @@ -69,6 +69,7 @@ EXPORT_SYMBOL_GPL(can_len2dlc); #ifdef CONFIG_CAN_CALC_BITTIMING #define CAN_CALC_MAX_ERROR 50 /* in one-tenth of a percent */ +#define CAN_CALC_SYNC_SEG 1 /* * Bit-timing calculation derived from: @@ -83,98 +84,126 @@ EXPORT_SYMBOL_GPL(can_len2dlc); * registers of the CAN controller. You can find more information * in the header file linux/can/netlink.h. */ -static int can_update_spt(const struct can_bittiming_const *btc, - int sampl_pt, int tseg, int *tseg1, int *tseg2) +static int can_update_sample_point(const struct can_bittiming_const *btc, + unsigned int sample_point_nominal, unsigned int tseg, + unsigned int *tseg1_ptr, unsigned int *tseg2_ptr, + unsigned int *sample_point_error_ptr) { - *tseg2 = tseg + 1 - (sampl_pt * (tseg + 1)) / 1000; - if (*tseg2 < btc->tseg2_min) - *tseg2 = btc->tseg2_min; - if (*tseg2 > btc->tseg2_max) - *tseg2 = btc->tseg2_max; - *tseg1 = tseg - *tseg2; - if (*tseg1 > btc->tseg1_max) { - *tseg1 = btc->tseg1_max; - *tseg2 = tseg - *tseg1; + unsigned int sample_point_error, best_sample_point_error = UINT_MAX; + unsigned int sample_point, best_sample_point = 0; + unsigned int tseg1, tseg2; + int i; + + for (i = 0; i <= 1; i++) { + tseg2 = tseg + CAN_CALC_SYNC_SEG - (sample_point_nominal * (tseg + CAN_CALC_SYNC_SEG)) / 1000 - i; + tseg2 = clamp(tseg2, btc->tseg2_min, btc->tseg2_max); + tseg1 = tseg - tseg2; + if (tseg1 > btc->tseg1_max) { + tseg1 = btc->tseg1_max; + tseg2 = tseg - tseg1; + } + + sample_point = 1000 * (tseg + CAN_CALC_SYNC_SEG - tseg2) / (tseg + CAN_CALC_SYNC_SEG); + sample_point_error = abs(sample_point_nominal - sample_point); + + if ((sample_point <= sample_point_nominal) && (sample_point_error < best_sample_point_error)) { + best_sample_point = sample_point; + best_sample_point_error = sample_point_error; + *tseg1_ptr = tseg1; + *tseg2_ptr = tseg2; + } } - return 1000 * (tseg + 1 - *tseg2) / (tseg + 1); + + if (sample_point_error_ptr) + *sample_point_error_ptr = best_sample_point_error; + + return best_sample_point; } static int can_calc_bittiming(struct net_device *dev, struct can_bittiming *bt, const struct can_bittiming_const *btc) { struct can_priv *priv = netdev_priv(dev); - long best_error = 1000000000, error = 0; - int best_tseg = 0, best_brp = 0, brp = 0; - int tsegall, tseg = 0, tseg1 = 0, tseg2 = 0; - int spt_error = 1000, spt = 0, sampl_pt; - long rate; + unsigned int bitrate; /* current bitrate */ + unsigned int bitrate_error; /* difference between current and nominal value */ + unsigned int best_bitrate_error = UINT_MAX; + unsigned int sample_point_error; /* difference between current and nominal value */ + unsigned int best_sample_point_error = UINT_MAX; + unsigned int sample_point_nominal; /* nominal sample point */ + unsigned int best_tseg = 0; /* current best value for tseg */ + unsigned int best_brp = 0; /* current best value for brp */ + unsigned int brp, tsegall, tseg, tseg1 = 0, tseg2 = 0; u64 v64; /* Use CiA recommended sample points */ if (bt->sample_point) { - sampl_pt = bt->sample_point; + sample_point_nominal = bt->sample_point; } else { if (bt->bitrate > 800000) - sampl_pt = 750; + sample_point_nominal = 750; else if (bt->bitrate > 500000) - sampl_pt = 800; + sample_point_nominal = 800; else - sampl_pt = 875; + sample_point_nominal = 875; } /* tseg even = round down, odd = round up */ for (tseg = (btc->tseg1_max + btc->tseg2_max) * 2 + 1; tseg >= (btc->tseg1_min + btc->tseg2_min) * 2; tseg--) { - tsegall = 1 + tseg / 2; + tsegall = CAN_CALC_SYNC_SEG + tseg / 2; + /* Compute all possible tseg choices (tseg=tseg1+tseg2) */ brp = priv->clock.freq / (tsegall * bt->bitrate) + tseg % 2; - /* chose brp step which is possible in system */ + + /* choose brp step which is possible in system */ brp = (brp / btc->brp_inc) * btc->brp_inc; if ((brp < btc->brp_min) || (brp > btc->brp_max)) continue; - rate = priv->clock.freq / (brp * tsegall); - error = bt->bitrate - rate; + + bitrate = priv->clock.freq / (brp * tsegall); + bitrate_error = abs(bt->bitrate - bitrate); + /* tseg brp biterror */ - if (error < 0) - error = -error; - if (error > best_error) + if (bitrate_error > best_bitrate_error) continue; - best_error = error; - if (error == 0) { - spt = can_update_spt(btc, sampl_pt, tseg / 2, - &tseg1, &tseg2); - error = sampl_pt - spt; - if (error < 0) - error = -error; - if (error > spt_error) - continue; - spt_error = error; - } + + /* reset sample point error if we have a better bitrate */ + if (bitrate_error < best_bitrate_error) + best_sample_point_error = UINT_MAX; + + can_update_sample_point(btc, sample_point_nominal, tseg / 2, &tseg1, &tseg2, &sample_point_error); + if (sample_point_error > best_sample_point_error) + continue; + + best_sample_point_error = sample_point_error; + best_bitrate_error = bitrate_error; best_tseg = tseg / 2; best_brp = brp; - if (error == 0) + + if (bitrate_error == 0 && sample_point_error == 0) break; } - if (best_error) { + if (best_bitrate_error) { /* Error in one-tenth of a percent */ - error = (best_error * 1000) / bt->bitrate; - if (error > CAN_CALC_MAX_ERROR) { + v64 = (u64)best_bitrate_error * 1000; + do_div(v64, bt->bitrate); + bitrate_error = (u32)v64; + if (bitrate_error > CAN_CALC_MAX_ERROR) { netdev_err(dev, - "bitrate error %ld.%ld%% too high\n", - error / 10, error % 10); + "bitrate error %d.%d%% too high\n", + bitrate_error / 10, bitrate_error % 10); return -EDOM; - } else { - netdev_warn(dev, "bitrate error %ld.%ld%%\n", - error / 10, error % 10); } + netdev_warn(dev, "bitrate error %d.%d%%\n", + bitrate_error / 10, bitrate_error % 10); } /* real sample point */ - bt->sample_point = can_update_spt(btc, sampl_pt, best_tseg, - &tseg1, &tseg2); + bt->sample_point = can_update_sample_point(btc, sample_point_nominal, best_tseg, + &tseg1, &tseg2, NULL); - v64 = (u64)best_brp * 1000000000UL; + v64 = (u64)best_brp * 1000 * 1000 * 1000; do_div(v64, priv->clock.freq); bt->tq = (u32)v64; bt->prop_seg = tseg1 / 2; @@ -182,9 +211,9 @@ static int can_calc_bittiming(struct net_device *dev, struct can_bittiming *bt, bt->phase_seg2 = tseg2; /* check for sjw user settings */ - if (!bt->sjw || !btc->sjw_max) + if (!bt->sjw || !btc->sjw_max) { bt->sjw = 1; - else { + } else { /* bt->sjw is at least 1 -> sanitize upper bound to sjw_max */ if (bt->sjw > btc->sjw_max) bt->sjw = btc->sjw_max; @@ -194,8 +223,9 @@ static int can_calc_bittiming(struct net_device *dev, struct can_bittiming *bt, } bt->brp = best_brp; - /* real bit-rate */ - bt->bitrate = priv->clock.freq / (bt->brp * (tseg1 + tseg2 + 1)); + + /* real bitrate */ + bt->bitrate = priv->clock.freq / (bt->brp * (CAN_CALC_SYNC_SEG + tseg1 + tseg2)); return 0; } -- cgit v0.10.2 From 21aedfdf092b0ceaf717e89495505052f8346362 Mon Sep 17 00:00:00 2001 From: Marek Vasut Date: Thu, 9 Jun 2016 21:21:00 +0200 Subject: can: slcan: Replace sizeof struct can_frame with CAN_MTU Use CAN_MTU macro instead of sizeof(struct can_frame) just like the other drivers do. Signed-off-by: Marek Vasut Cc: Oliver Hartkopp Cc: Marc Kleine-Budde Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/slcan.c b/drivers/net/can/slcan.c index 9a3f15c..eb71737 100644 --- a/drivers/net/can/slcan.c +++ b/drivers/net/can/slcan.c @@ -354,7 +354,7 @@ static netdev_tx_t slc_xmit(struct sk_buff *skb, struct net_device *dev) { struct slcan *sl = netdev_priv(dev); - if (skb->len != sizeof(struct can_frame)) + if (skb->len != CAN_MTU) goto out; spin_lock(&sl->lock); @@ -442,7 +442,7 @@ static void slc_setup(struct net_device *dev) dev->addr_len = 0; dev->tx_queue_len = 10; - dev->mtu = sizeof(struct can_frame); + dev->mtu = CAN_MTU; dev->type = ARPHRD_CAN; /* New-style flags. */ -- cgit v0.10.2 From ae5ac9b3b8a942724da578dd5a4112b6d2f7d1bb Mon Sep 17 00:00:00 2001 From: William Breathitt Gray Date: Wed, 1 Jun 2016 08:59:51 -0400 Subject: can: tscan1: Utilize the module_isa_driver macro This driver does not do anything special in module init/exit. This patch eliminates the module init/exit boilerplate code by utilizing the module_isa_driver macro. Signed-off-by: William Breathitt Gray Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/sja1000/tscan1.c b/drivers/net/can/sja1000/tscan1.c index 76513dd..7957245 100644 --- a/drivers/net/can/sja1000/tscan1.c +++ b/drivers/net/can/sja1000/tscan1.c @@ -203,14 +203,4 @@ static struct isa_driver tscan1_isa_driver = { }, }; -static int __init tscan1_init(void) -{ - return isa_register_driver(&tscan1_isa_driver, TSCAN1_MAXDEV); -} -module_init(tscan1_init); - -static void __exit tscan1_exit(void) -{ - isa_unregister_driver(&tscan1_isa_driver); -} -module_exit(tscan1_exit); +module_isa_driver(tscan1_isa_driver, TSCAN1_MAXDEV); -- cgit v0.10.2 From 9be95c91fd9b5386590e408a99db5fdfa836c5b9 Mon Sep 17 00:00:00 2001 From: Maximilian Schneider Date: Fri, 10 Jun 2016 20:39:22 +0000 Subject: can: gs_usb: codingstyle: introduce use of BIT() + fix indention This patch converts "1 << n" by BIT(n) and fixes the indention. No functional change. Signed-off-by: Hubert Denkmair Signed-off-by: Maximilian Schneider [mkl: split codingstyle changes into separate patch] Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c index 1556d42..6186a94 100644 --- a/drivers/net/can/usb/gs_usb.c +++ b/drivers/net/can/usb/gs_usb.c @@ -77,10 +77,10 @@ struct gs_device_config { } __packed; #define GS_CAN_MODE_NORMAL 0 -#define GS_CAN_MODE_LISTEN_ONLY (1<<0) -#define GS_CAN_MODE_LOOP_BACK (1<<1) -#define GS_CAN_MODE_TRIPLE_SAMPLE (1<<2) -#define GS_CAN_MODE_ONE_SHOT (1<<3) +#define GS_CAN_MODE_LISTEN_ONLY BIT(0) +#define GS_CAN_MODE_LOOP_BACK BIT(1) +#define GS_CAN_MODE_TRIPLE_SAMPLE BIT(2) +#define GS_CAN_MODE_ONE_SHOT BIT(3) struct gs_device_mode { u32 mode; @@ -101,10 +101,10 @@ struct gs_device_bittiming { u32 brp; } __packed; -#define GS_CAN_FEATURE_LISTEN_ONLY (1<<0) -#define GS_CAN_FEATURE_LOOP_BACK (1<<1) -#define GS_CAN_FEATURE_TRIPLE_SAMPLE (1<<2) -#define GS_CAN_FEATURE_ONE_SHOT (1<<3) +#define GS_CAN_FEATURE_LISTEN_ONLY BIT(0) +#define GS_CAN_FEATURE_LOOP_BACK BIT(1) +#define GS_CAN_FEATURE_TRIPLE_SAMPLE BIT(2) +#define GS_CAN_FEATURE_ONE_SHOT BIT(3) struct gs_device_bt_const { u32 feature; @@ -209,7 +209,8 @@ static void gs_free_tx_context(struct gs_tx_context *txc) /* Get a tx context by id. */ -static struct gs_tx_context *gs_get_tx_context(struct gs_can *dev, unsigned int id) +static struct gs_tx_context *gs_get_tx_context(struct gs_can *dev, + unsigned int id) { unsigned long flags; @@ -452,7 +453,8 @@ static void gs_usb_xmit_callback(struct urb *urb) netif_wake_queue(netdev); } -static netdev_tx_t gs_can_start_xmit(struct sk_buff *skb, struct net_device *netdev) +static netdev_tx_t gs_can_start_xmit(struct sk_buff *skb, + struct net_device *netdev) { struct gs_can *dev = netdev_priv(netdev); struct net_device_stats *stats = &dev->netdev->stats; @@ -658,7 +660,8 @@ static int gs_can_open(struct net_device *netdev) rc = usb_control_msg(interface_to_usbdev(dev->iface), usb_sndctrlpipe(interface_to_usbdev(dev->iface), 0), GS_USB_BREQ_MODE, - USB_DIR_OUT|USB_TYPE_VENDOR|USB_RECIP_INTERFACE, + USB_DIR_OUT | USB_TYPE_VENDOR | + USB_RECIP_INTERFACE, dev->channel, 0, dm, -- cgit v0.10.2 From 05ca5270005c18ec46decacef87992ea968f9fce Mon Sep 17 00:00:00 2001 From: Maximilian Schneider Date: Fri, 10 Jun 2016 20:39:22 +0000 Subject: can: gs_usb: add ethtool set_phys_id callback to locate physical device This patch Implements the ethtool set_phys_id callback to ease the locating of specific physical devices. Currently only supported on candleLight interfaces. Signed-off-by: Hubert Denkmair Signed-off-by: Maximilian Schneider [mkl: split codingstyle change sinto separate patch] Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c index 6186a94..3607643 100644 --- a/drivers/net/can/usb/gs_usb.c +++ b/drivers/net/can/usb/gs_usb.c @@ -39,7 +39,9 @@ enum gs_usb_breq { GS_USB_BREQ_MODE, GS_USB_BREQ_BERR, GS_USB_BREQ_BT_CONST, - GS_USB_BREQ_DEVICE_CONFIG + GS_USB_BREQ_DEVICE_CONFIG, + GS_USB_BREQ_TIMESTAMP, + GS_USB_BREQ_IDENTIFY, }; enum gs_can_mode { @@ -58,6 +60,11 @@ enum gs_can_state { GS_CAN_STATE_SLEEPING }; +enum gs_can_identify_mode { + GS_CAN_IDENTIFY_OFF = 0, + GS_CAN_IDENTIFY_ON +}; + /* data types passed between host and device */ struct gs_host_config { u32 byte_order; @@ -101,10 +108,16 @@ struct gs_device_bittiming { u32 brp; } __packed; +struct gs_identify_mode { + u32 mode; +} __packed; + #define GS_CAN_FEATURE_LISTEN_ONLY BIT(0) #define GS_CAN_FEATURE_LOOP_BACK BIT(1) #define GS_CAN_FEATURE_TRIPLE_SAMPLE BIT(2) #define GS_CAN_FEATURE_ONE_SHOT BIT(3) +#define GS_CAN_FEATURE_HW_TIMESTAMP BIT(4) +#define GS_CAN_FEATURE_IDENTIFY BIT(5) struct gs_device_bt_const { u32 feature; @@ -724,7 +737,59 @@ static const struct net_device_ops gs_usb_netdev_ops = { .ndo_change_mtu = can_change_mtu, }; -static struct gs_can *gs_make_candev(unsigned int channel, struct usb_interface *intf) +static int gs_usb_set_identify(struct net_device *netdev, bool do_identify) +{ + struct gs_can *dev = netdev_priv(netdev); + struct gs_identify_mode imode; + int rc; + + if (do_identify) + imode.mode = GS_CAN_IDENTIFY_ON; + else + imode.mode = GS_CAN_IDENTIFY_OFF; + + rc = usb_control_msg(interface_to_usbdev(dev->iface), + usb_sndctrlpipe(interface_to_usbdev(dev->iface), + 0), + GS_USB_BREQ_IDENTIFY, + USB_DIR_OUT | USB_TYPE_VENDOR | + USB_RECIP_INTERFACE, + dev->channel, + 0, + &imode, + sizeof(imode), + 100); + + return (rc > 0) ? 0 : rc; +} + +/* blink LED's for finding the this interface */ +static int gs_usb_set_phys_id(struct net_device *dev, + enum ethtool_phys_id_state state) +{ + int rc = 0; + + switch (state) { + case ETHTOOL_ID_ACTIVE: + rc = gs_usb_set_identify(dev, GS_CAN_IDENTIFY_ON); + break; + case ETHTOOL_ID_INACTIVE: + rc = gs_usb_set_identify(dev, GS_CAN_IDENTIFY_OFF); + break; + default: + break; + } + + return rc; +} + +static const struct ethtool_ops gs_usb_ethtool_ops = { + .set_phys_id = gs_usb_set_phys_id, +}; + +static struct gs_can *gs_make_candev(unsigned int channel, + struct usb_interface *intf, + struct gs_device_config *dconf) { struct gs_can *dev; struct net_device *netdev; @@ -812,10 +877,14 @@ static struct gs_can *gs_make_candev(unsigned int channel, struct usb_interface if (bt_const->feature & GS_CAN_FEATURE_ONE_SHOT) dev->can.ctrlmode_supported |= CAN_CTRLMODE_ONE_SHOT; - kfree(bt_const); - SET_NETDEV_DEV(netdev, &intf->dev); + if (dconf->sw_version > 1) + if (bt_const->feature & GS_CAN_FEATURE_IDENTIFY) + netdev->ethtool_ops = &gs_usb_ethtool_ops; + + kfree(bt_const); + rc = register_candev(dev->netdev); if (rc) { free_candev(dev->netdev); @@ -833,19 +902,16 @@ static void gs_destroy_candev(struct gs_can *dev) free_candev(dev->netdev); } -static int gs_usb_probe(struct usb_interface *intf, const struct usb_device_id *id) +static int gs_usb_probe(struct usb_interface *intf, + const struct usb_device_id *id) { struct gs_usb *dev; int rc = -ENOMEM; unsigned int icount, i; - struct gs_host_config *hconf; - struct gs_device_config *dconf; - - hconf = kmalloc(sizeof(*hconf), GFP_KERNEL); - if (!hconf) - return -ENOMEM; - - hconf->byte_order = 0x0000beef; + struct gs_host_config hconf = { + .byte_order = 0x0000beef, + }; + struct gs_device_config dconf; /* send host config */ rc = usb_control_msg(interface_to_usbdev(intf), @@ -854,22 +920,16 @@ static int gs_usb_probe(struct usb_interface *intf, const struct usb_device_id * USB_DIR_OUT|USB_TYPE_VENDOR|USB_RECIP_INTERFACE, 1, intf->altsetting[0].desc.bInterfaceNumber, - hconf, - sizeof(*hconf), + &hconf, + sizeof(hconf), 1000); - kfree(hconf); - if (rc < 0) { dev_err(&intf->dev, "Couldn't send data format (err=%d)\n", rc); return rc; } - dconf = kmalloc(sizeof(*dconf), GFP_KERNEL); - if (!dconf) - return -ENOMEM; - /* read device config */ rc = usb_control_msg(interface_to_usbdev(intf), usb_rcvctrlpipe(interface_to_usbdev(intf), 0), @@ -877,22 +937,16 @@ static int gs_usb_probe(struct usb_interface *intf, const struct usb_device_id * USB_DIR_IN|USB_TYPE_VENDOR|USB_RECIP_INTERFACE, 1, intf->altsetting[0].desc.bInterfaceNumber, - dconf, - sizeof(*dconf), + &dconf, + sizeof(dconf), 1000); if (rc < 0) { dev_err(&intf->dev, "Couldn't get device config: (err=%d)\n", rc); - - kfree(dconf); - return rc; } - icount = dconf->icount+1; - - kfree(dconf); - + icount = dconf.icount + 1; dev_info(&intf->dev, "Configuring for %d interfaces\n", icount); if (icount > GS_MAX_INTF) { @@ -913,7 +967,7 @@ static int gs_usb_probe(struct usb_interface *intf, const struct usb_device_id * dev->udev = interface_to_usbdev(intf); for (i = 0; i < icount; i++) { - dev->canch[i] = gs_make_candev(i, intf); + dev->canch[i] = gs_make_candev(i, intf, &dconf); if (IS_ERR_OR_NULL(dev->canch[i])) { /* save error code to return later */ rc = PTR_ERR(dev->canch[i]); -- cgit v0.10.2 From 95acb490ec5145015b64cf4e99f604bb5fe79250 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Fri, 17 Jun 2016 15:35:24 +0200 Subject: can: bcm: fix indention and other minor style issues Signed-off-by: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde diff --git a/net/can/bcm.c b/net/can/bcm.c index 6863310..17fb796 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -184,42 +184,40 @@ static int bcm_proc_show(struct seq_file *m, void *v) continue; seq_printf(m, "rx_op: %03X %-5s ", - op->can_id, bcm_proc_getifname(ifname, op->ifindex)); + op->can_id, bcm_proc_getifname(ifname, op->ifindex)); seq_printf(m, "[%u]%c ", op->nframes, - (op->flags & RX_CHECK_DLC)?'d':' '); + (op->flags & RX_CHECK_DLC) ? 'd' : ' '); if (op->kt_ival1.tv64) seq_printf(m, "timeo=%lld ", - (long long) - ktime_to_us(op->kt_ival1)); + (long long)ktime_to_us(op->kt_ival1)); if (op->kt_ival2.tv64) seq_printf(m, "thr=%lld ", - (long long) - ktime_to_us(op->kt_ival2)); + (long long)ktime_to_us(op->kt_ival2)); seq_printf(m, "# recv %ld (%ld) => reduction: ", - op->frames_filtered, op->frames_abs); + op->frames_filtered, op->frames_abs); reduction = 100 - (op->frames_filtered * 100) / op->frames_abs; seq_printf(m, "%s%ld%%\n", - (reduction == 100)?"near ":"", reduction); + (reduction == 100) ? "near " : "", reduction); } list_for_each_entry(op, &bo->tx_ops, list) { seq_printf(m, "tx_op: %03X %s [%u] ", - op->can_id, - bcm_proc_getifname(ifname, op->ifindex), - op->nframes); + op->can_id, + bcm_proc_getifname(ifname, op->ifindex), + op->nframes); if (op->kt_ival1.tv64) seq_printf(m, "t1=%lld ", - (long long) ktime_to_us(op->kt_ival1)); + (long long)ktime_to_us(op->kt_ival1)); if (op->kt_ival2.tv64) seq_printf(m, "t2=%lld ", - (long long) ktime_to_us(op->kt_ival2)); + (long long)ktime_to_us(op->kt_ival2)); seq_printf(m, "# sent %ld\n", op->frames_abs); } @@ -282,7 +280,7 @@ static void bcm_can_tx(struct bcm_op *op) /* reached last frame? */ if (op->currframe >= op->nframes) op->currframe = 0; - out: +out: dev_put(dev); } -- cgit v0.10.2 From 72c8a89ad2e4de18849674f30589baa5ebb4fbc1 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Fri, 17 Jun 2016 15:35:25 +0200 Subject: can: bcm: use CAN frame instead of can_frame in comments can_frame is the name of the struct can_frame which is not meant in the corrected comments. Signed-off-by: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde diff --git a/net/can/bcm.c b/net/can/bcm.c index 17fb796..83aa6cf 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -84,7 +84,7 @@ MODULE_LICENSE("Dual BSD/GPL"); MODULE_AUTHOR("Oliver Hartkopp "); MODULE_ALIAS("can-proto-2"); -/* easy access to can_frame payload */ +/* easy access to CAN frame payload */ static inline u64 GET_U64(const struct can_frame *cp) { return *(u64 *)cp->data; @@ -305,13 +305,13 @@ static void bcm_send_to_user(struct bcm_op *op, struct bcm_msg_head *head, memcpy(skb_put(skb, sizeof(*head)), head, sizeof(*head)); if (head->nframes) { - /* can_frames starting here */ + /* CAN frames starting here */ firstframe = (struct can_frame *)skb_tail_pointer(skb); memcpy(skb_put(skb, datalen), frames, datalen); /* - * the BCM uses the can_dlc-element of the can_frame + * the BCM uses the can_dlc-element of the CAN frame * structure for internal purposes. This is only * relevant for updates that are generated by the * BCM, where nframes is 1 @@ -492,7 +492,7 @@ static void bcm_rx_cmp_to_index(struct bcm_op *op, unsigned int index, return; } - /* do a real check in can_frame data section */ + /* do a real check in CAN frame data section */ if ((GET_U64(&op->frames[index]) & GET_U64(rxdata)) != (GET_U64(&op->frames[index]) & GET_U64(&op->last_frames[index]))) { @@ -501,7 +501,7 @@ static void bcm_rx_cmp_to_index(struct bcm_op *op, unsigned int index, } if (op->flags & RX_CHECK_DLC) { - /* do a real check in can_frame dlc */ + /* do a real check in CAN frame dlc */ if (rxdata->can_dlc != (op->last_frames[index].can_dlc & BCM_CAN_DLC_MASK)) { bcm_rx_update_and_send(op, &op->last_frames[index], @@ -554,7 +554,7 @@ static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) /* if user wants to be informed, when cyclic CAN-Messages come back */ if ((op->flags & RX_ANNOUNCE_RESUME) && op->last_frames) { - /* clear received can_frames to indicate 'nothing received' */ + /* clear received CAN frames to indicate 'nothing received' */ memset(op->last_frames, 0, op->nframes * CFSIZ); } @@ -840,7 +840,7 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (!ifindex) return -ENODEV; - /* check nframes boundaries - we need at least one can_frame */ + /* check nframes boundaries - we need at least one CAN frame */ if (msg_head->nframes < 1 || msg_head->nframes > MAX_NFRAMES) return -EINVAL; @@ -851,14 +851,14 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, /* update existing BCM operation */ /* - * Do we need more space for the can_frames than currently + * Do we need more space for the CAN frames than currently * allocated? -> This is a _really_ unusual use-case and * therefore (complexity / locking) it is not supported. */ if (msg_head->nframes > op->nframes) return -E2BIG; - /* update can_frames content */ + /* update CAN frames content */ for (i = 0; i < msg_head->nframes; i++) { err = memcpy_from_msg((u8 *)&op->frames[i], msg, CFSIZ); @@ -883,7 +883,7 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, op->can_id = msg_head->can_id; - /* create array for can_frames and copy the data */ + /* create array for CAN frames and copy the data */ if (msg_head->nframes > 1) { op->frames = kmalloc(msg_head->nframes * CFSIZ, GFP_KERNEL); @@ -966,7 +966,7 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (op->flags & STARTTIMER) { hrtimer_cancel(&op->timer); - /* spec: send can_frame when starting timer */ + /* spec: send CAN frame when starting timer */ op->flags |= TX_ANNOUNCE; } @@ -1015,7 +1015,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, /* update existing BCM operation */ /* - * Do we need more space for the can_frames than currently + * Do we need more space for the CAN frames than currently * allocated? -> This is a _really_ unusual use-case and * therefore (complexity / locking) it is not supported. */ @@ -1023,7 +1023,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, return -E2BIG; if (msg_head->nframes) { - /* update can_frames content */ + /* update CAN frames content */ err = memcpy_from_msg((u8 *)op->frames, msg, msg_head->nframes * CFSIZ); if (err < 0) @@ -1048,7 +1048,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, op->nframes = msg_head->nframes; if (msg_head->nframes > 1) { - /* create array for can_frames and copy the data */ + /* create array for CAN frames and copy the data */ op->frames = kmalloc(msg_head->nframes * CFSIZ, GFP_KERNEL); if (!op->frames) { @@ -1056,7 +1056,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, return -ENOMEM; } - /* create and init array for received can_frames */ + /* create and init array for received CAN frames */ op->last_frames = kzalloc(msg_head->nframes * CFSIZ, GFP_KERNEL); if (!op->last_frames) { @@ -1327,7 +1327,7 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) break; case TX_SEND: - /* we need exactly one can_frame behind the msg head */ + /* we need exactly one CAN frame behind the msg head */ if ((msg_head.nframes != 1) || (size != CFSIZ + MHSIZ)) ret = -EINVAL; else -- cgit v0.10.2 From 2b5f5f5dc114219dcd848fb0ff344acb413c11ef Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Fri, 17 Jun 2016 15:35:26 +0200 Subject: can: bcm: unify bcm_msg_head handling and prepare function parameters Signed-off-by: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde diff --git a/net/can/bcm.c b/net/can/bcm.c index 83aa6cf..f3bf387 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -693,13 +693,13 @@ rx_starttimer: /* * helpers for bcm_op handling: find & delete bcm [rx|tx] op elements */ -static struct bcm_op *bcm_find_op(struct list_head *ops, canid_t can_id, - int ifindex) +static struct bcm_op *bcm_find_op(struct list_head *ops, + struct bcm_msg_head *mh, int ifindex) { struct bcm_op *op; list_for_each_entry(op, ops, list) { - if ((op->can_id == can_id) && (op->ifindex == ifindex)) + if ((op->can_id == mh->can_id) && (op->ifindex == ifindex)) return op; } @@ -742,12 +742,13 @@ static void bcm_rx_unreg(struct net_device *dev, struct bcm_op *op) /* * bcm_delete_rx_op - find and remove a rx op (returns number of removed ops) */ -static int bcm_delete_rx_op(struct list_head *ops, canid_t can_id, int ifindex) +static int bcm_delete_rx_op(struct list_head *ops, struct bcm_msg_head *mh, + int ifindex) { struct bcm_op *op, *n; list_for_each_entry_safe(op, n, ops, list) { - if ((op->can_id == can_id) && (op->ifindex == ifindex)) { + if ((op->can_id == mh->can_id) && (op->ifindex == ifindex)) { /* * Don't care if we're bound or not (due to netdev @@ -787,12 +788,13 @@ static int bcm_delete_rx_op(struct list_head *ops, canid_t can_id, int ifindex) /* * bcm_delete_tx_op - find and remove a tx op (returns number of removed ops) */ -static int bcm_delete_tx_op(struct list_head *ops, canid_t can_id, int ifindex) +static int bcm_delete_tx_op(struct list_head *ops, struct bcm_msg_head *mh, + int ifindex) { struct bcm_op *op, *n; list_for_each_entry_safe(op, n, ops, list) { - if ((op->can_id == can_id) && (op->ifindex == ifindex)) { + if ((op->can_id == mh->can_id) && (op->ifindex == ifindex)) { list_del(&op->list); bcm_remove_op(op); return 1; /* done */ @@ -808,7 +810,7 @@ static int bcm_delete_tx_op(struct list_head *ops, canid_t can_id, int ifindex) static int bcm_read_op(struct list_head *ops, struct bcm_msg_head *msg_head, int ifindex) { - struct bcm_op *op = bcm_find_op(ops, msg_head->can_id, ifindex); + struct bcm_op *op = bcm_find_op(ops, msg_head, ifindex); if (!op) return -EINVAL; @@ -845,8 +847,7 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, return -EINVAL; /* check the given can_id */ - op = bcm_find_op(&bo->tx_ops, msg_head->can_id, ifindex); - + op = bcm_find_op(&bo->tx_ops, msg_head, ifindex); if (op) { /* update existing BCM operation */ @@ -1010,7 +1011,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, return -EINVAL; /* check the given can_id */ - op = bcm_find_op(&bo->rx_ops, msg_head->can_id, ifindex); + op = bcm_find_op(&bo->rx_ops, msg_head, ifindex); if (op) { /* update existing BCM operation */ @@ -1192,7 +1193,8 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, /* * bcm_tx_send - send a single CAN frame to the CAN interface (for bcm_sendmsg) */ -static int bcm_tx_send(struct msghdr *msg, int ifindex, struct sock *sk) +static int bcm_tx_send(struct msghdr *msg, int ifindex, struct sock *sk, + int cfsiz) { struct sk_buff *skb; struct net_device *dev; @@ -1202,13 +1204,13 @@ static int bcm_tx_send(struct msghdr *msg, int ifindex, struct sock *sk) if (!ifindex) return -ENODEV; - skb = alloc_skb(CFSIZ + sizeof(struct can_skb_priv), GFP_KERNEL); + skb = alloc_skb(cfsiz + sizeof(struct can_skb_priv), GFP_KERNEL); if (!skb) return -ENOMEM; can_skb_reserve(skb); - err = memcpy_from_msg(skb_put(skb, CFSIZ), msg, CFSIZ); + err = memcpy_from_msg(skb_put(skb, cfsiz), msg, cfsiz); if (err < 0) { kfree_skb(skb); return err; @@ -1230,7 +1232,7 @@ static int bcm_tx_send(struct msghdr *msg, int ifindex, struct sock *sk) if (err) return err; - return CFSIZ + MHSIZ; + return cfsiz + MHSIZ; } /* @@ -1248,7 +1250,15 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) return -ENOTCONN; /* check for valid message length from userspace */ - if (size < MHSIZ || (size - MHSIZ) % CFSIZ) + if (size < MHSIZ) + return -EINVAL; + + /* read message head information */ + ret = memcpy_from_msg((u8 *)&msg_head, msg, MHSIZ); + if (ret < 0) + return ret; + + if ((size - MHSIZ) % CFSIZ) return -EINVAL; /* check for alternative ifindex for this bcm_op */ @@ -1282,12 +1292,6 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) } } - /* read message head information */ - - ret = memcpy_from_msg((u8 *)&msg_head, msg, MHSIZ); - if (ret < 0) - return ret; - lock_sock(sk); switch (msg_head.opcode) { @@ -1301,14 +1305,14 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) break; case TX_DELETE: - if (bcm_delete_tx_op(&bo->tx_ops, msg_head.can_id, ifindex)) + if (bcm_delete_tx_op(&bo->tx_ops, &msg_head, ifindex)) ret = MHSIZ; else ret = -EINVAL; break; case RX_DELETE: - if (bcm_delete_rx_op(&bo->rx_ops, msg_head.can_id, ifindex)) + if (bcm_delete_rx_op(&bo->rx_ops, &msg_head, ifindex)) ret = MHSIZ; else ret = -EINVAL; @@ -1331,7 +1335,7 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) if ((msg_head.nframes != 1) || (size != CFSIZ + MHSIZ)) ret = -EINVAL; else - ret = bcm_tx_send(msg, ifindex, sk); + ret = bcm_tx_send(msg, ifindex, sk, CFSIZ); break; default: -- cgit v0.10.2 From 6f3b911d5f29b98752e5da86a295210c0c4f4e14 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Fri, 17 Jun 2016 15:35:27 +0200 Subject: can: bcm: add support for CAN FD frames The programming API of the CAN_BCM depends on struct can_frame which is given as array directly behind the bcm_msg_head structure. To follow this schema for the CAN FD frames a new flag 'CAN_FD_FRAME' in the bcm_msg_head flags indicates that the concatenated CAN frame structures behind the bcm_msg_head are defined as struct canfd_frame. This patch adds the support to handle CAN and CAN FD frames on a per BCM-op base. Main changes: - generally use struct canfd_frames instead if struct can_frames - use canfd_frame.flags instead of can_frame.can_dlc for private BCM flags - make all CAN frame sizes depending on the new CAN_FD_FRAME flags - separate between CAN and CAN FD when sending/receiving frames Due to the dependence of the CAN_FD_FRAME flag the former binary interface for classic CAN frames remains stable. Signed-off-by: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde diff --git a/include/uapi/linux/can/bcm.h b/include/uapi/linux/can/bcm.h index 7a291dc..cefb304 100644 --- a/include/uapi/linux/can/bcm.h +++ b/include/uapi/linux/can/bcm.h @@ -99,5 +99,6 @@ enum { #define RX_ANNOUNCE_RESUME 0x0100 #define TX_RESET_MULTI_IDX 0x0200 #define RX_RTR_FRAME 0x0400 +#define CAN_FD_FRAME 0x0800 #endif /* !_UAPI_CAN_BCM_H */ diff --git a/net/can/bcm.c b/net/can/bcm.c index f3bf387..8e999ff 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -1,7 +1,7 @@ /* * bcm.c - Broadcast Manager to filter/send (cyclic) CAN content * - * Copyright (c) 2002-2007 Volkswagen Group Electronic Research + * Copyright (c) 2002-2016 Volkswagen Group Electronic Research * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -67,27 +67,31 @@ */ #define MAX_NFRAMES 256 -/* use of last_frames[index].can_dlc */ +/* use of last_frames[index].flags */ #define RX_RECV 0x40 /* received data for this element */ #define RX_THR 0x80 /* element not been sent due to throttle feature */ -#define BCM_CAN_DLC_MASK 0x0F /* clean private flags in can_dlc by masking */ +#define BCM_CAN_FLAGS_MASK 0x3F /* to clean private flags after usage */ /* get best masking value for can_rx_register() for a given single can_id */ #define REGMASK(id) ((id & CAN_EFF_FLAG) ? \ (CAN_EFF_MASK | CAN_EFF_FLAG | CAN_RTR_FLAG) : \ (CAN_SFF_MASK | CAN_EFF_FLAG | CAN_RTR_FLAG)) -#define CAN_BCM_VERSION CAN_VERSION +#define CAN_BCM_VERSION "20160617" MODULE_DESCRIPTION("PF_CAN broadcast manager protocol"); MODULE_LICENSE("Dual BSD/GPL"); MODULE_AUTHOR("Oliver Hartkopp "); MODULE_ALIAS("can-proto-2"); -/* easy access to CAN frame payload */ -static inline u64 GET_U64(const struct can_frame *cp) +/* + * easy access to the first 64 bit of can(fd)_frame payload. cp->data is + * 64 bit aligned so the offset has to be multiples of 8 which is ensured + * by the only callers in bcm_rx_cmp_to_index() bcm_rx_handler(). + */ +static inline u64 get_u64(const struct canfd_frame *cp, int offset) { - return *(u64 *)cp->data; + return *(u64 *)(cp->data + offset); } struct bcm_op { @@ -101,13 +105,14 @@ struct bcm_op { struct tasklet_struct tsklet, thrtsklet; ktime_t rx_stamp, kt_ival1, kt_ival2, kt_lastmsg; int rx_ifindex; + int cfsiz; u32 count; u32 nframes; u32 currframe; - struct can_frame *frames; - struct can_frame *last_frames; - struct can_frame sframe; - struct can_frame last_sframe; + struct canfd_frame *frames; + struct canfd_frame *last_frames; + struct canfd_frame sframe; + struct canfd_frame last_sframe; struct sock *sk; struct net_device *rx_reg_dev; }; @@ -136,7 +141,7 @@ static inline ktime_t bcm_timeval_to_ktime(struct bcm_timeval tv) return ktime_set(tv.tv_sec, tv.tv_usec * NSEC_PER_USEC); } -#define CFSIZ sizeof(struct can_frame) +#define CFSIZ(flags) ((flags & CAN_FD_FRAME) ? CANFD_MTU : CAN_MTU) #define OPSIZ sizeof(struct bcm_op) #define MHSIZ sizeof(struct bcm_msg_head) @@ -183,10 +188,16 @@ static int bcm_proc_show(struct seq_file *m, void *v) if (!op->frames_abs) continue; - seq_printf(m, "rx_op: %03X %-5s ", - op->can_id, bcm_proc_getifname(ifname, op->ifindex)); - seq_printf(m, "[%u]%c ", op->nframes, - (op->flags & RX_CHECK_DLC) ? 'd' : ' '); + seq_printf(m, "rx_op: %03X %-5s ", op->can_id, + bcm_proc_getifname(ifname, op->ifindex)); + + if (op->flags & CAN_FD_FRAME) + seq_printf(m, "(%u)", op->nframes); + else + seq_printf(m, "[%u]", op->nframes); + + seq_printf(m, "%c ", (op->flags & RX_CHECK_DLC) ? 'd' : ' '); + if (op->kt_ival1.tv64) seq_printf(m, "timeo=%lld ", (long long)ktime_to_us(op->kt_ival1)); @@ -206,10 +217,13 @@ static int bcm_proc_show(struct seq_file *m, void *v) list_for_each_entry(op, &bo->tx_ops, list) { - seq_printf(m, "tx_op: %03X %s [%u] ", - op->can_id, - bcm_proc_getifname(ifname, op->ifindex), - op->nframes); + seq_printf(m, "tx_op: %03X %s ", op->can_id, + bcm_proc_getifname(ifname, op->ifindex)); + + if (op->flags & CAN_FD_FRAME) + seq_printf(m, "(%u) ", op->nframes); + else + seq_printf(m, "[%u] ", op->nframes); if (op->kt_ival1.tv64) seq_printf(m, "t1=%lld ", @@ -246,7 +260,7 @@ static void bcm_can_tx(struct bcm_op *op) { struct sk_buff *skb; struct net_device *dev; - struct can_frame *cf = &op->frames[op->currframe]; + struct canfd_frame *cf = op->frames + op->cfsiz * op->currframe; /* no target device? => exit */ if (!op->ifindex) @@ -258,7 +272,7 @@ static void bcm_can_tx(struct bcm_op *op) return; } - skb = alloc_skb(CFSIZ + sizeof(struct can_skb_priv), gfp_any()); + skb = alloc_skb(op->cfsiz + sizeof(struct can_skb_priv), gfp_any()); if (!skb) goto out; @@ -266,7 +280,7 @@ static void bcm_can_tx(struct bcm_op *op) can_skb_prv(skb)->ifindex = dev->ifindex; can_skb_prv(skb)->skbcnt = 0; - memcpy(skb_put(skb, CFSIZ), cf, CFSIZ); + memcpy(skb_put(skb, op->cfsiz), cf, op->cfsiz); /* send with loopback */ skb->dev = dev; @@ -289,13 +303,13 @@ out: * (consisting of bcm_msg_head + x CAN frames) */ static void bcm_send_to_user(struct bcm_op *op, struct bcm_msg_head *head, - struct can_frame *frames, int has_timestamp) + struct canfd_frame *frames, int has_timestamp) { struct sk_buff *skb; - struct can_frame *firstframe; + struct canfd_frame *firstframe; struct sockaddr_can *addr; struct sock *sk = op->sk; - unsigned int datalen = head->nframes * CFSIZ; + unsigned int datalen = head->nframes * op->cfsiz; int err; skb = alloc_skb(sizeof(*head) + datalen, gfp_any()); @@ -306,18 +320,18 @@ static void bcm_send_to_user(struct bcm_op *op, struct bcm_msg_head *head, if (head->nframes) { /* CAN frames starting here */ - firstframe = (struct can_frame *)skb_tail_pointer(skb); + firstframe = (struct canfd_frame *)skb_tail_pointer(skb); memcpy(skb_put(skb, datalen), frames, datalen); /* - * the BCM uses the can_dlc-element of the CAN frame + * the BCM uses the flags-element of the canfd_frame * structure for internal purposes. This is only * relevant for updates that are generated by the * BCM, where nframes is 1 */ if (head->nframes == 1) - firstframe->can_dlc &= BCM_CAN_DLC_MASK; + firstframe->flags &= BCM_CAN_FLAGS_MASK; } if (has_timestamp) { @@ -404,7 +418,7 @@ static enum hrtimer_restart bcm_tx_timeout_handler(struct hrtimer *hrtimer) /* * bcm_rx_changed - create a RX_CHANGED notification due to changed content */ -static void bcm_rx_changed(struct bcm_op *op, struct can_frame *data) +static void bcm_rx_changed(struct bcm_op *op, struct canfd_frame *data) { struct bcm_msg_head head; @@ -416,7 +430,7 @@ static void bcm_rx_changed(struct bcm_op *op, struct can_frame *data) op->frames_filtered = op->frames_abs = 0; /* this element is not throttled anymore */ - data->can_dlc &= (BCM_CAN_DLC_MASK|RX_RECV); + data->flags &= (BCM_CAN_FLAGS_MASK|RX_RECV); head.opcode = RX_CHANGED; head.flags = op->flags; @@ -435,13 +449,13 @@ static void bcm_rx_changed(struct bcm_op *op, struct can_frame *data) * 2. send a notification to the user (if possible) */ static void bcm_rx_update_and_send(struct bcm_op *op, - struct can_frame *lastdata, - const struct can_frame *rxdata) + struct canfd_frame *lastdata, + const struct canfd_frame *rxdata) { - memcpy(lastdata, rxdata, CFSIZ); + memcpy(lastdata, rxdata, op->cfsiz); /* mark as used and throttled by default */ - lastdata->can_dlc |= (RX_RECV|RX_THR); + lastdata->flags |= (RX_RECV|RX_THR); /* throttling mode inactive ? */ if (!op->kt_ival2.tv64) { @@ -479,33 +493,36 @@ rx_changed_settime: * received data stored in op->last_frames[] */ static void bcm_rx_cmp_to_index(struct bcm_op *op, unsigned int index, - const struct can_frame *rxdata) + const struct canfd_frame *rxdata) { + struct canfd_frame *cf = op->frames + op->cfsiz * index; + struct canfd_frame *lcf = op->last_frames + op->cfsiz * index; + int i; + /* - * no one uses the MSBs of can_dlc for comparison, + * no one uses the MSBs of flags for comparison, * so we use it here to detect the first time of reception */ - if (!(op->last_frames[index].can_dlc & RX_RECV)) { + if (!(lcf->flags & RX_RECV)) { /* received data for the first time => send update to user */ - bcm_rx_update_and_send(op, &op->last_frames[index], rxdata); + bcm_rx_update_and_send(op, lcf, rxdata); return; } /* do a real check in CAN frame data section */ - - if ((GET_U64(&op->frames[index]) & GET_U64(rxdata)) != - (GET_U64(&op->frames[index]) & GET_U64(&op->last_frames[index]))) { - bcm_rx_update_and_send(op, &op->last_frames[index], rxdata); - return; + for (i = 0; i < rxdata->len; i += 8) { + if ((get_u64(cf, i) & get_u64(rxdata, i)) != + (get_u64(cf, i) & get_u64(lcf, i))) { + bcm_rx_update_and_send(op, lcf, rxdata); + return; + } } if (op->flags & RX_CHECK_DLC) { - /* do a real check in CAN frame dlc */ - if (rxdata->can_dlc != (op->last_frames[index].can_dlc & - BCM_CAN_DLC_MASK)) { - bcm_rx_update_and_send(op, &op->last_frames[index], - rxdata); + /* do a real check in CAN frame length */ + if (rxdata->len != lcf->len) { + bcm_rx_update_and_send(op, lcf, rxdata); return; } } @@ -555,7 +572,7 @@ static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) /* if user wants to be informed, when cyclic CAN-Messages come back */ if ((op->flags & RX_ANNOUNCE_RESUME) && op->last_frames) { /* clear received CAN frames to indicate 'nothing received' */ - memset(op->last_frames, 0, op->nframes * CFSIZ); + memset(op->last_frames, 0, op->nframes * op->cfsiz); } return HRTIMER_NORESTART; @@ -567,9 +584,11 @@ static enum hrtimer_restart bcm_rx_timeout_handler(struct hrtimer *hrtimer) static inline int bcm_rx_do_flush(struct bcm_op *op, int update, unsigned int index) { - if ((op->last_frames) && (op->last_frames[index].can_dlc & RX_THR)) { + struct canfd_frame *lcf = op->last_frames + op->cfsiz * index; + + if ((op->last_frames) && (lcf->flags & RX_THR)) { if (update) - bcm_rx_changed(op, &op->last_frames[index]); + bcm_rx_changed(op, lcf); return 1; } return 0; @@ -634,15 +653,19 @@ static enum hrtimer_restart bcm_rx_thr_handler(struct hrtimer *hrtimer) static void bcm_rx_handler(struct sk_buff *skb, void *data) { struct bcm_op *op = (struct bcm_op *)data; - const struct can_frame *rxframe = (struct can_frame *)skb->data; + const struct canfd_frame *rxframe = (struct canfd_frame *)skb->data; unsigned int i; - /* disable timeout */ - hrtimer_cancel(&op->timer); - if (op->can_id != rxframe->can_id) return; + /* make sure to handle the correct frame type (CAN / CAN FD) */ + if (skb->len != op->cfsiz) + return; + + /* disable timeout */ + hrtimer_cancel(&op->timer); + /* save rx timestamp */ op->rx_stamp = skb->tstamp; /* save originator for recvfrom() */ @@ -673,13 +696,14 @@ static void bcm_rx_handler(struct sk_buff *skb, void *data) * multiplex compare * * find the first multiplex mask that fits. - * Remark: The MUX-mask is stored in index 0 + * Remark: The MUX-mask is stored in index 0 - but only the + * first 64 bits of the frame data[] are relevant (CAN FD) */ for (i = 1; i < op->nframes; i++) { - if ((GET_U64(&op->frames[0]) & GET_U64(rxframe)) == - (GET_U64(&op->frames[0]) & - GET_U64(&op->frames[i]))) { + if ((get_u64(op->frames, 0) & get_u64(rxframe, 0)) == + (get_u64(op->frames, 0) & + get_u64(op->frames + op->cfsiz * i, 0))) { bcm_rx_cmp_to_index(op, i, rxframe); break; } @@ -699,7 +723,8 @@ static struct bcm_op *bcm_find_op(struct list_head *ops, struct bcm_op *op; list_for_each_entry(op, ops, list) { - if ((op->can_id == mh->can_id) && (op->ifindex == ifindex)) + if ((op->can_id == mh->can_id) && (op->ifindex == ifindex) && + (op->flags & CAN_FD_FRAME) == (mh->flags & CAN_FD_FRAME)) return op; } @@ -748,7 +773,8 @@ static int bcm_delete_rx_op(struct list_head *ops, struct bcm_msg_head *mh, struct bcm_op *op, *n; list_for_each_entry_safe(op, n, ops, list) { - if ((op->can_id == mh->can_id) && (op->ifindex == ifindex)) { + if ((op->can_id == mh->can_id) && (op->ifindex == ifindex) && + (op->flags & CAN_FD_FRAME) == (mh->flags & CAN_FD_FRAME)) { /* * Don't care if we're bound or not (due to netdev @@ -794,7 +820,8 @@ static int bcm_delete_tx_op(struct list_head *ops, struct bcm_msg_head *mh, struct bcm_op *op, *n; list_for_each_entry_safe(op, n, ops, list) { - if ((op->can_id == mh->can_id) && (op->ifindex == ifindex)) { + if ((op->can_id == mh->can_id) && (op->ifindex == ifindex) && + (op->flags & CAN_FD_FRAME) == (mh->flags & CAN_FD_FRAME)) { list_del(&op->list); bcm_remove_op(op); return 1; /* done */ @@ -835,6 +862,7 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, { struct bcm_sock *bo = bcm_sk(sk); struct bcm_op *op; + struct canfd_frame *cf; unsigned int i; int err; @@ -861,19 +889,27 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, /* update CAN frames content */ for (i = 0; i < msg_head->nframes; i++) { - err = memcpy_from_msg((u8 *)&op->frames[i], msg, CFSIZ); - if (op->frames[i].can_dlc > 8) - err = -EINVAL; + cf = op->frames + op->cfsiz * i; + err = memcpy_from_msg((u8 *)cf, msg, op->cfsiz); + + if (op->flags & CAN_FD_FRAME) { + if (cf->len > 64) + err = -EINVAL; + } else { + if (cf->len > 8) + err = -EINVAL; + } if (err < 0) return err; if (msg_head->flags & TX_CP_CAN_ID) { /* copy can_id into frame */ - op->frames[i].can_id = msg_head->can_id; + cf->can_id = msg_head->can_id; } } + op->flags = msg_head->flags; } else { /* insert new BCM operation for the given can_id */ @@ -882,11 +918,13 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (!op) return -ENOMEM; - op->can_id = msg_head->can_id; + op->can_id = msg_head->can_id; + op->cfsiz = CFSIZ(msg_head->flags); + op->flags = msg_head->flags; /* create array for CAN frames and copy the data */ if (msg_head->nframes > 1) { - op->frames = kmalloc(msg_head->nframes * CFSIZ, + op->frames = kmalloc(msg_head->nframes * op->cfsiz, GFP_KERNEL); if (!op->frames) { kfree(op); @@ -896,10 +934,17 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, op->frames = &op->sframe; for (i = 0; i < msg_head->nframes; i++) { - err = memcpy_from_msg((u8 *)&op->frames[i], msg, CFSIZ); - if (op->frames[i].can_dlc > 8) - err = -EINVAL; + cf = op->frames + op->cfsiz * i; + err = memcpy_from_msg((u8 *)cf, msg, op->cfsiz); + + if (op->flags & CAN_FD_FRAME) { + if (cf->len > 64) + err = -EINVAL; + } else { + if (cf->len > 8) + err = -EINVAL; + } if (err < 0) { if (op->frames != &op->sframe) @@ -910,7 +955,7 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (msg_head->flags & TX_CP_CAN_ID) { /* copy can_id into frame */ - op->frames[i].can_id = msg_head->can_id; + cf->can_id = msg_head->can_id; } } @@ -945,8 +990,6 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, /* check flags */ - op->flags = msg_head->flags; - if (op->flags & TX_RESET_MULTI_IDX) { /* start multiple frame transmission with index 0 */ op->currframe = 0; @@ -980,7 +1023,7 @@ static int bcm_tx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (op->flags & STARTTIMER) bcm_tx_start_timer(op); - return msg_head->nframes * CFSIZ + MHSIZ; + return msg_head->nframes * op->cfsiz + MHSIZ; } /* @@ -1026,15 +1069,16 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (msg_head->nframes) { /* update CAN frames content */ err = memcpy_from_msg((u8 *)op->frames, msg, - msg_head->nframes * CFSIZ); + msg_head->nframes * op->cfsiz); if (err < 0) return err; /* clear last_frames to indicate 'nothing received' */ - memset(op->last_frames, 0, msg_head->nframes * CFSIZ); + memset(op->last_frames, 0, msg_head->nframes * op->cfsiz); } op->nframes = msg_head->nframes; + op->flags = msg_head->flags; /* Only an update -> do not call can_rx_register() */ do_rx_register = 0; @@ -1045,12 +1089,14 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (!op) return -ENOMEM; - op->can_id = msg_head->can_id; - op->nframes = msg_head->nframes; + op->can_id = msg_head->can_id; + op->nframes = msg_head->nframes; + op->cfsiz = CFSIZ(msg_head->flags); + op->flags = msg_head->flags; if (msg_head->nframes > 1) { /* create array for CAN frames and copy the data */ - op->frames = kmalloc(msg_head->nframes * CFSIZ, + op->frames = kmalloc(msg_head->nframes * op->cfsiz, GFP_KERNEL); if (!op->frames) { kfree(op); @@ -1058,7 +1104,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, } /* create and init array for received CAN frames */ - op->last_frames = kzalloc(msg_head->nframes * CFSIZ, + op->last_frames = kzalloc(msg_head->nframes * op->cfsiz, GFP_KERNEL); if (!op->last_frames) { kfree(op->frames); @@ -1073,7 +1119,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, if (msg_head->nframes) { err = memcpy_from_msg((u8 *)op->frames, msg, - msg_head->nframes * CFSIZ); + msg_head->nframes * op->cfsiz); if (err < 0) { if (op->frames != &op->sframe) kfree(op->frames); @@ -1115,7 +1161,6 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, } /* if ((op = bcm_find_op(&bo->rx_ops, msg_head->can_id, ifindex))) */ /* check flags */ - op->flags = msg_head->flags; if (op->flags & RX_RTR_FRAME) { @@ -1187,7 +1232,7 @@ static int bcm_rx_setup(struct bcm_msg_head *msg_head, struct msghdr *msg, } } - return msg_head->nframes * CFSIZ + MHSIZ; + return msg_head->nframes * op->cfsiz + MHSIZ; } /* @@ -1244,6 +1289,7 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) struct bcm_sock *bo = bcm_sk(sk); int ifindex = bo->ifindex; /* default ifindex for this bcm_op */ struct bcm_msg_head msg_head; + int cfsiz; int ret; /* read bytes or error codes as return value */ if (!bo->bound) @@ -1258,7 +1304,8 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) if (ret < 0) return ret; - if ((size - MHSIZ) % CFSIZ) + cfsiz = CFSIZ(msg_head.flags); + if ((size - MHSIZ) % cfsiz) return -EINVAL; /* check for alternative ifindex for this bcm_op */ @@ -1332,10 +1379,10 @@ static int bcm_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) case TX_SEND: /* we need exactly one CAN frame behind the msg head */ - if ((msg_head.nframes != 1) || (size != CFSIZ + MHSIZ)) + if ((msg_head.nframes != 1) || (size != cfsiz + MHSIZ)) ret = -EINVAL; else - ret = bcm_tx_send(msg, ifindex, sk, CFSIZ); + ret = bcm_tx_send(msg, ifindex, sk, cfsiz); break; default: -- cgit v0.10.2 From 9be05c7f372940a2308e0301c5bdddab3022a449 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Fri, 17 Jun 2016 15:35:28 +0200 Subject: can: bcm: add documentation for CAN FD support Signed-off-by: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde diff --git a/Documentation/networking/can.txt b/Documentation/networking/can.txt index d58ff84..aa15b9e 100644 --- a/Documentation/networking/can.txt +++ b/Documentation/networking/can.txt @@ -31,6 +31,7 @@ This file contains 4.2.4 Broadcast Manager message sequence transmission 4.2.5 Broadcast Manager receive filter timers 4.2.6 Broadcast Manager multiplex message receive filter + 4.2.7 Broadcast Manager CAN FD support 4.3 connected transport protocols (SOCK_SEQPACKET) 4.4 unconnected transport protocols (SOCK_DGRAM) @@ -799,7 +800,7 @@ solution for a couple of reasons: } mytxmsg; (..) - mytxmsg.nframes = 4; + mytxmsg.msg_head.nframes = 4; (..) write(s, &mytxmsg, sizeof(mytxmsg)); @@ -852,6 +853,28 @@ solution for a couple of reasons: write(s, &msg, sizeof(msg)); + 4.2.7 Broadcast Manager CAN FD support + + The programming API of the CAN_BCM depends on struct can_frame which is + given as array directly behind the bcm_msg_head structure. To follow this + schema for the CAN FD frames a new flag 'CAN_FD_FRAME' in the bcm_msg_head + flags indicates that the concatenated CAN frame structures behind the + bcm_msg_head are defined as struct canfd_frame. + + struct { + struct bcm_msg_head msg_head; + struct canfd_frame frame[5]; + } msg; + + msg.msg_head.opcode = RX_SETUP; + msg.msg_head.can_id = 0x42; + msg.msg_head.flags = CAN_FD_FRAME; + msg.msg_head.nframes = 5; + (..) + + When using CAN FD frames for multiplex filtering the MUX mask is still + expected in the first 64 bit of the struct canfd_frame data section. + 4.3 connected transport protocols (SOCK_SEQPACKET) 4.4 unconnected transport protocols (SOCK_DGRAM) -- cgit v0.10.2 From dd3bd23eb438919aeeb73a7116642731ef8f73a8 Mon Sep 17 00:00:00 2001 From: Ramesh Shanmugasundaram Date: Fri, 17 Jun 2016 09:20:55 +0100 Subject: can: rcar_canfd: Add Renesas R-Car CAN FD driver This patch adds support for the CAN FD controller found in Renesas R-Car SoCs. The controller operates in CAN FD only mode by default. CAN FD mode supports both Classical CAN & CAN FD frame formats. The controller supports ISO 11898-1:2015 CAN FD format only. This controller supports two channels and the driver can enable either or both of the channels. Driver uses Rx FIFOs (one per channel) for reception & Common FIFOs (one per channel) for transmission. Rx filter rules are configured to the minimum (one per channel) and it accepts Standard, Extended, Data & Remote Frame combinations. Note: There are few documentation errors in R-Car Gen3 Hardware User Manual v0.5E with respect to CAN FD controller. They are listed below: 1. CAN FD interrupt numbers 29 & 30 are listed as per channel interrupts. However, they are common to both channels (i.e.) they are global and channel interrupts respectively. 2. CANFD clock is derived from PLL1. This is not documented. 3. CANFD clock is further divided by (1/2) within the CAN FD controller. This is not documented. 4. The minimum value of NTSEG1 in RSCFDnCFDCmNCFG register is 2 Tq. It is specified 4 Tq in the manual. 5. The maximum number of message RAM area the controller can use is 3584 bytes. It is specified 10752 bytes in the manual. Signed-off-by: Ramesh Shanmugasundaram Acked-by: Rob Herring Reviewed-by: Ulrich Hecht Signed-off-by: Marc Kleine-Budde diff --git a/Documentation/devicetree/bindings/net/can/rcar_canfd.txt b/Documentation/devicetree/bindings/net/can/rcar_canfd.txt new file mode 100644 index 0000000..d45182b --- /dev/null +++ b/Documentation/devicetree/bindings/net/can/rcar_canfd.txt @@ -0,0 +1,89 @@ +Renesas R-Car CAN FD controller Device Tree Bindings +---------------------------------------------------- + +Required properties: +- compatible: Must contain one or more of the following: + - "renesas,rcar-gen3-canfd" for R-Car Gen3 compatible controller. + - "renesas,r8a7795-canfd" for R8A7795 (R-Car H3) compatible controller. + + When compatible with the generic version, nodes must list the + SoC-specific version corresponding to the platform first, followed by the + family-specific and/or generic versions. + +- reg: physical base address and size of the R-Car CAN FD register map. +- interrupts: interrupt specifier for the Global & Channel interrupts +- clocks: phandles and clock specifiers for 3 clock inputs. +- clock-names: 3 clock input name strings: "fck", "canfd", "can_clk". +- pinctrl-0: pin control group to be used for this controller. +- pinctrl-names: must be "default". + +Required child nodes: +The controller supports two channels and each is represented as a child node. +The name of the child nodes are "channel0" and "channel1" respectively. Each +child node supports the "status" property only, which is used to +enable/disable the respective channel. + +Required properties for "renesas,r8a7795-canfd" compatible: +In R8A7795 SoC, canfd clock is a div6 clock and can be used by both CAN +and CAN FD controller at the same time. It needs to be scaled to maximum +frequency if any of these controllers use it. This is done using the +below properties. + +- assigned-clocks: phandle of canfd clock. +- assigned-clock-rates: maximum frequency of this clock. + +Example +------- + +SoC common .dtsi file: + + canfd: can@e66c0000 { + compatible = "renesas,r8a7795-canfd", + "renesas,rcar-gen3-canfd"; + reg = <0 0xe66c0000 0 0x8000>; + interrupts = , + ; + clocks = <&cpg CPG_MOD 914>, + <&cpg CPG_CORE R8A7795_CLK_CANFD>, + <&can_clk>; + clock-names = "fck", "canfd", "can_clk"; + assigned-clocks = <&cpg CPG_CORE R8A7795_CLK_CANFD>; + assigned-clock-rates = <40000000>; + power-domains = <&cpg>; + status = "disabled"; + + channel0 { + status = "disabled"; + }; + + channel1 { + status = "disabled"; + }; + }; + +Board specific .dts file: + +E.g. below enables Channel 1 alone in the board. + +&canfd { + pinctrl-0 = <&canfd1_pins>; + pinctrl-names = "default"; + status = "okay"; + + channel1 { + status = "okay"; + }; +}; + +E.g. below enables Channel 0 alone in the board using External clock +as fCAN clock. + +&canfd { + pinctrl-0 = <&canfd0_pins &can_clk_pins>; + pinctrl-names = "default"; + status = "okay"; + + channel0 { + status = "okay"; + }; +}; diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig index 0d40aef..13003a9 100644 --- a/drivers/net/can/Kconfig +++ b/drivers/net/can/Kconfig @@ -152,6 +152,7 @@ source "drivers/net/can/cc770/Kconfig" source "drivers/net/can/ifi_canfd/Kconfig" source "drivers/net/can/m_can/Kconfig" source "drivers/net/can/mscan/Kconfig" +source "drivers/net/can/rcar/Kconfig" source "drivers/net/can/sja1000/Kconfig" source "drivers/net/can/softing/Kconfig" source "drivers/net/can/spi/Kconfig" diff --git a/drivers/net/can/Makefile b/drivers/net/can/Makefile index e3db0c8..226d5b5 100644 --- a/drivers/net/can/Makefile +++ b/drivers/net/can/Makefile @@ -10,6 +10,7 @@ can-dev-y := dev.o can-dev-$(CONFIG_CAN_LEDS) += led.o +obj-y += rcar/ obj-y += spi/ obj-y += usb/ obj-y += softing/ diff --git a/drivers/net/can/rcar/Kconfig b/drivers/net/can/rcar/Kconfig new file mode 100644 index 0000000..6ea64d1 --- /dev/null +++ b/drivers/net/can/rcar/Kconfig @@ -0,0 +1,11 @@ +config CAN_RCAR_CANFD + tristate "Renesas R-Car CAN FD controller" + depends on ARCH_RENESAS || ARM + ---help--- + Say Y here if you want to use CAN FD controller found on + Renesas R-Car SoCs. The driver puts the controller in CAN FD only + mode, which can interoperate with CAN2.0 nodes but does not support + dedicated CAN 2.0 mode. + + To compile this driver as a module, choose M here: the module will + be called rcar_canfd. diff --git a/drivers/net/can/rcar/Makefile b/drivers/net/can/rcar/Makefile new file mode 100644 index 0000000..cbaf498e --- /dev/null +++ b/drivers/net/can/rcar/Makefile @@ -0,0 +1,5 @@ +# +# Makefile for the Renesas R-Car CAN FD controller driver +# + +obj-$(CONFIG_CAN_RCAR_CANFD) += rcar_canfd.o diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c new file mode 100644 index 0000000..a39d922 --- /dev/null +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -0,0 +1,1695 @@ +/* Renesas R-Car CAN FD device driver + * + * Copyright (C) 2015 Renesas Electronics Corp. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +/* The R-Car CAN FD controller can operate in either one of the below two modes + * - CAN FD only mode + * - Classical CAN (CAN 2.0) only mode + * + * This driver puts the controller in CAN FD only mode by default. In this + * mode, the controller acts as a CAN FD node that can also interoperate with + * CAN 2.0 nodes. + * + * As of now, this driver does not support the Classical CAN (CAN 2.0) mode, + * which is handled by a different register map compared to CAN FD only mode. + * + * Note: The h/w manual register naming convention is clumsy and not acceptable + * to use as it is in the driver. However, those names are added as comments + * wherever it is modified to a readable name. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define RCANFD_DRV_NAME "rcar_canfd" + +/* Global register bits */ + +/* RSCFDnCFDGRMCFG */ +#define RCANFD_GRMCFG_RCMC BIT(0) + +/* RSCFDnCFDGCFG */ +#define RCANFD_GCFG_CMPOC BIT(5) +#define RCANFD_GCFG_DCS BIT(4) +#define RCANFD_GCFG_DCE BIT(1) +#define RCANFD_GCFG_TPRI BIT(0) + +/* RSCFDnCFDGCTR */ +#define RCANFD_GCTR_TSRST BIT(16) +#define RCANFD_GCTR_CFMPOFIE BIT(11) +#define RCANFD_GCTR_THLEIE BIT(10) +#define RCANFD_GCTR_MEIE BIT(9) +#define RCANFD_GCTR_DEIE BIT(8) +#define RCANFD_GCTR_GSLPR BIT(2) +#define RCANFD_GCTR_GMDC_MASK (0x3) +#define RCANFD_GCTR_GMDC_GOPM (0x0) +#define RCANFD_GCTR_GMDC_GRESET (0x1) +#define RCANFD_GCTR_GMDC_GTEST (0x2) + +/* RSCFDnCFDGSTS */ +#define RCANFD_GSTS_GRAMINIT BIT(3) +#define RCANFD_GSTS_GSLPSTS BIT(2) +#define RCANFD_GSTS_GHLTSTS BIT(1) +#define RCANFD_GSTS_GRSTSTS BIT(0) +/* Non-operational status */ +#define RCANFD_GSTS_GNOPM (BIT(0) | BIT(1) | BIT(2) | BIT(3)) + +/* RSCFDnCFDGERFL */ +#define RCANFD_GERFL_EEF1 BIT(17) +#define RCANFD_GERFL_EEF0 BIT(16) +#define RCANFD_GERFL_CMPOF BIT(3) +#define RCANFD_GERFL_THLES BIT(2) +#define RCANFD_GERFL_MES BIT(1) +#define RCANFD_GERFL_DEF BIT(0) + +#define RCANFD_GERFL_ERR(x) ((x) & (RCANFD_GERFL_EEF1 |\ + RCANFD_GERFL_EEF0 |\ + RCANFD_GERFL_MES |\ + RCANFD_GERFL_CMPOF)) + +/* AFL Rx rules registers */ + +/* RSCFDnCFDGAFLCFG0 */ +#define RCANFD_GAFLCFG_SETRNC(n, x) (((x) & 0xff) << (24 - n * 8)) +#define RCANFD_GAFLCFG_GETRNC(n, x) (((x) >> (24 - n * 8)) & 0xff) + +/* RSCFDnCFDGAFLECTR */ +#define RCANFD_GAFLECTR_AFLDAE BIT(8) +#define RCANFD_GAFLECTR_AFLPN(x) ((x) & 0x1f) + +/* RSCFDnCFDGAFLIDj */ +#define RCANFD_GAFLID_GAFLLB BIT(29) + +/* RSCFDnCFDGAFLP1_j */ +#define RCANFD_GAFLP1_GAFLFDP(x) (1 << (x)) + +/* Channel register bits */ + +/* RSCFDnCFDCmNCFG */ +#define RCANFD_NCFG_NTSEG2(x) (((x) & 0x1f) << 24) +#define RCANFD_NCFG_NTSEG1(x) (((x) & 0x7f) << 16) +#define RCANFD_NCFG_NSJW(x) (((x) & 0x1f) << 11) +#define RCANFD_NCFG_NBRP(x) (((x) & 0x3ff) << 0) + +/* RSCFDnCFDCmCTR */ +#define RCANFD_CCTR_CTME BIT(24) +#define RCANFD_CCTR_ERRD BIT(23) +#define RCANFD_CCTR_BOM_MASK (0x3 << 21) +#define RCANFD_CCTR_BOM_ISO (0x0 << 21) +#define RCANFD_CCTR_BOM_BENTRY (0x1 << 21) +#define RCANFD_CCTR_BOM_BEND (0x2 << 21) +#define RCANFD_CCTR_TDCVFIE BIT(19) +#define RCANFD_CCTR_SOCOIE BIT(18) +#define RCANFD_CCTR_EOCOIE BIT(17) +#define RCANFD_CCTR_TAIE BIT(16) +#define RCANFD_CCTR_ALIE BIT(15) +#define RCANFD_CCTR_BLIE BIT(14) +#define RCANFD_CCTR_OLIE BIT(13) +#define RCANFD_CCTR_BORIE BIT(12) +#define RCANFD_CCTR_BOEIE BIT(11) +#define RCANFD_CCTR_EPIE BIT(10) +#define RCANFD_CCTR_EWIE BIT(9) +#define RCANFD_CCTR_BEIE BIT(8) +#define RCANFD_CCTR_CSLPR BIT(2) +#define RCANFD_CCTR_CHMDC_MASK (0x3) +#define RCANFD_CCTR_CHDMC_COPM (0x0) +#define RCANFD_CCTR_CHDMC_CRESET (0x1) +#define RCANFD_CCTR_CHDMC_CHLT (0x2) + +/* RSCFDnCFDCmSTS */ +#define RCANFD_CSTS_COMSTS BIT(7) +#define RCANFD_CSTS_RECSTS BIT(6) +#define RCANFD_CSTS_TRMSTS BIT(5) +#define RCANFD_CSTS_BOSTS BIT(4) +#define RCANFD_CSTS_EPSTS BIT(3) +#define RCANFD_CSTS_SLPSTS BIT(2) +#define RCANFD_CSTS_HLTSTS BIT(1) +#define RCANFD_CSTS_CRSTSTS BIT(0) + +#define RCANFD_CSTS_TECCNT(x) (((x) >> 24) & 0xff) +#define RCANFD_CSTS_RECCNT(x) (((x) >> 16) & 0xff) + +/* RSCFDnCFDCmERFL */ +#define RCANFD_CERFL_ADERR BIT(14) +#define RCANFD_CERFL_B0ERR BIT(13) +#define RCANFD_CERFL_B1ERR BIT(12) +#define RCANFD_CERFL_CERR BIT(11) +#define RCANFD_CERFL_AERR BIT(10) +#define RCANFD_CERFL_FERR BIT(9) +#define RCANFD_CERFL_SERR BIT(8) +#define RCANFD_CERFL_ALF BIT(7) +#define RCANFD_CERFL_BLF BIT(6) +#define RCANFD_CERFL_OVLF BIT(5) +#define RCANFD_CERFL_BORF BIT(4) +#define RCANFD_CERFL_BOEF BIT(3) +#define RCANFD_CERFL_EPF BIT(2) +#define RCANFD_CERFL_EWF BIT(1) +#define RCANFD_CERFL_BEF BIT(0) + +#define RCANFD_CERFL_ERR(x) ((x) & (0x7fff)) /* above bits 14:0 */ + +/* RSCFDnCFDCmDCFG */ +#define RCANFD_DCFG_DSJW(x) (((x) & 0x7) << 24) +#define RCANFD_DCFG_DTSEG2(x) (((x) & 0x7) << 20) +#define RCANFD_DCFG_DTSEG1(x) (((x) & 0xf) << 16) +#define RCANFD_DCFG_DBRP(x) (((x) & 0xff) << 0) + +/* RSCFDnCFDCmFDCFG */ +#define RCANFD_FDCFG_TDCE BIT(9) +#define RCANFD_FDCFG_TDCOC BIT(8) +#define RCANFD_FDCFG_TDCO(x) (((x) & 0x7f) >> 16) + +/* RSCFDnCFDRFCCx */ +#define RCANFD_RFCC_RFIM BIT(12) +#define RCANFD_RFCC_RFDC(x) (((x) & 0x7) << 8) +#define RCANFD_RFCC_RFPLS(x) (((x) & 0x7) << 4) +#define RCANFD_RFCC_RFIE BIT(1) +#define RCANFD_RFCC_RFE BIT(0) + +/* RSCFDnCFDRFSTSx */ +#define RCANFD_RFSTS_RFIF BIT(3) +#define RCANFD_RFSTS_RFMLT BIT(2) +#define RCANFD_RFSTS_RFFLL BIT(1) +#define RCANFD_RFSTS_RFEMP BIT(0) + +/* RSCFDnCFDRFIDx */ +#define RCANFD_RFID_RFIDE BIT(31) +#define RCANFD_RFID_RFRTR BIT(30) + +/* RSCFDnCFDRFPTRx */ +#define RCANFD_RFPTR_RFDLC(x) (((x) >> 28) & 0xf) +#define RCANFD_RFPTR_RFPTR(x) (((x) >> 16) & 0xfff) +#define RCANFD_RFPTR_RFTS(x) (((x) >> 0) & 0xffff) + +/* RSCFDnCFDRFFDSTSx */ +#define RCANFD_RFFDSTS_RFFDF BIT(2) +#define RCANFD_RFFDSTS_RFBRS BIT(1) +#define RCANFD_RFFDSTS_RFESI BIT(0) + +/* Common FIFO bits */ + +/* RSCFDnCFDCFCCk */ +#define RCANFD_CFCC_CFTML(x) (((x) & 0xf) << 20) +#define RCANFD_CFCC_CFM(x) (((x) & 0x3) << 16) +#define RCANFD_CFCC_CFIM BIT(12) +#define RCANFD_CFCC_CFDC(x) (((x) & 0x7) << 8) +#define RCANFD_CFCC_CFPLS(x) (((x) & 0x7) << 4) +#define RCANFD_CFCC_CFTXIE BIT(2) +#define RCANFD_CFCC_CFE BIT(0) + +/* RSCFDnCFDCFSTSk */ +#define RCANFD_CFSTS_CFMC(x) (((x) >> 8) & 0xff) +#define RCANFD_CFSTS_CFTXIF BIT(4) +#define RCANFD_CFSTS_CFMLT BIT(2) +#define RCANFD_CFSTS_CFFLL BIT(1) +#define RCANFD_CFSTS_CFEMP BIT(0) + +/* RSCFDnCFDCFIDk */ +#define RCANFD_CFID_CFIDE BIT(31) +#define RCANFD_CFID_CFRTR BIT(30) +#define RCANFD_CFID_CFID_MASK(x) ((x) & 0x1fffffff) + +/* RSCFDnCFDCFPTRk */ +#define RCANFD_CFPTR_CFDLC(x) (((x) & 0xf) << 28) +#define RCANFD_CFPTR_CFPTR(x) (((x) & 0xfff) << 16) +#define RCANFD_CFPTR_CFTS(x) (((x) & 0xff) << 0) + +/* RSCFDnCFDCFFDCSTSk */ +#define RCANFD_CFFDCSTS_CFFDF BIT(2) +#define RCANFD_CFFDCSTS_CFBRS BIT(1) +#define RCANFD_CFFDCSTS_CFESI BIT(0) + +/* This controller supports classical CAN only mode or CAN FD only mode. These + * modes are supported in two separate set of register maps & names. However, + * some of the register offsets are common for both modes. Those offsets are + * listed below as Common registers. + * + * The CAN FD only specific registers are listed separately and their names + * starts with RCANFD_F_xxx names. When classical CAN only specific registers + * are added, those specific registers can be prefixed as RCANFD_C_xxx. + */ + +/* Common registers */ + +/* RSCFDnCFDCmNCFG / RSCFDnCmCFG */ +#define RCANFD_CCFG(m) (0x0000 + (0x10 * (m))) +/* RSCFDnCFDCmCTR / RSCFDnCmCTR */ +#define RCANFD_CCTR(m) (0x0004 + (0x10 * (m))) +/* RSCFDnCFDCmSTS / RSCFDnCmSTS */ +#define RCANFD_CSTS(m) (0x0008 + (0x10 * (m))) +/* RSCFDnCFDCmERFL / RSCFDnCmERFL */ +#define RCANFD_CERFL(m) (0x000C + (0x10 * (m))) + +/* RSCFDnCFDGCFG / RSCFDnGCFG */ +#define RCANFD_GCFG (0x0084) +/* RSCFDnCFDGCTR / RSCFDnGCTR */ +#define RCANFD_GCTR (0x0088) +/* RSCFDnCFDGCTS / RSCFDnGCTS */ +#define RCANFD_GSTS (0x008c) +/* RSCFDnCFDGERFL / RSCFDnGERFL */ +#define RCANFD_GERFL (0x0090) +/* RSCFDnCFDGTSC / RSCFDnGTSC */ +#define RCANFD_GTSC (0x0094) +/* RSCFDnCFDGAFLECTR / RSCFDnGAFLECTR */ +#define RCANFD_GAFLECTR (0x0098) +/* RSCFDnCFDGAFLCFG0 / RSCFDnGAFLCFG0 */ +#define RCANFD_GAFLCFG0 (0x009c) +/* RSCFDnCFDGAFLCFG1 / RSCFDnGAFLCFG1 */ +#define RCANFD_GAFLCFG1 (0x00a0) +/* RSCFDnCFDRMNB / RSCFDnRMNB */ +#define RCANFD_RMNB (0x00a4) +/* RSCFDnCFDRMND / RSCFDnRMND */ +#define RCANFD_RMND(y) (0x00a8 + (0x04 * (y))) + +/* RSCFDnCFDRFCCx / RSCFDnRFCCx */ +#define RCANFD_RFCC(x) (0x00b8 + (0x04 * (x))) +/* RSCFDnCFDRFSTSx / RSCFDnRFSTSx */ +#define RCANFD_RFSTS(x) (0x00d8 + (0x04 * (x))) +/* RSCFDnCFDRFPCTRx / RSCFDnRFPCTRx */ +#define RCANFD_RFPCTR(x) (0x00f8 + (0x04 * (x))) + +/* Common FIFO Control registers */ + +/* RSCFDnCFDCFCCx / RSCFDnCFCCx */ +#define RCANFD_CFCC(ch, idx) (0x0118 + (0x0c * (ch)) + \ + (0x04 * (idx))) +/* RSCFDnCFDCFSTSx / RSCFDnCFSTSx */ +#define RCANFD_CFSTS(ch, idx) (0x0178 + (0x0c * (ch)) + \ + (0x04 * (idx))) +/* RSCFDnCFDCFPCTRx / RSCFDnCFPCTRx */ +#define RCANFD_CFPCTR(ch, idx) (0x01d8 + (0x0c * (ch)) + \ + (0x04 * (idx))) + +/* RSCFDnCFDFESTS / RSCFDnFESTS */ +#define RCANFD_FESTS (0x0238) +/* RSCFDnCFDFFSTS / RSCFDnFFSTS */ +#define RCANFD_FFSTS (0x023c) +/* RSCFDnCFDFMSTS / RSCFDnFMSTS */ +#define RCANFD_FMSTS (0x0240) +/* RSCFDnCFDRFISTS / RSCFDnRFISTS */ +#define RCANFD_RFISTS (0x0244) +/* RSCFDnCFDCFRISTS / RSCFDnCFRISTS */ +#define RCANFD_CFRISTS (0x0248) +/* RSCFDnCFDCFTISTS / RSCFDnCFTISTS */ +#define RCANFD_CFTISTS (0x024c) + +/* RSCFDnCFDTMCp / RSCFDnTMCp */ +#define RCANFD_TMC(p) (0x0250 + (0x01 * (p))) +/* RSCFDnCFDTMSTSp / RSCFDnTMSTSp */ +#define RCANFD_TMSTS(p) (0x02d0 + (0x01 * (p))) + +/* RSCFDnCFDTMTRSTSp / RSCFDnTMTRSTSp */ +#define RCANFD_TMTRSTS(y) (0x0350 + (0x04 * (y))) +/* RSCFDnCFDTMTARSTSp / RSCFDnTMTARSTSp */ +#define RCANFD_TMTARSTS(y) (0x0360 + (0x04 * (y))) +/* RSCFDnCFDTMTCSTSp / RSCFDnTMTCSTSp */ +#define RCANFD_TMTCSTS(y) (0x0370 + (0x04 * (y))) +/* RSCFDnCFDTMTASTSp / RSCFDnTMTASTSp */ +#define RCANFD_TMTASTS(y) (0x0380 + (0x04 * (y))) +/* RSCFDnCFDTMIECy / RSCFDnTMIECy */ +#define RCANFD_TMIEC(y) (0x0390 + (0x04 * (y))) + +/* RSCFDnCFDTXQCCm / RSCFDnTXQCCm */ +#define RCANFD_TXQCC(m) (0x03a0 + (0x04 * (m))) +/* RSCFDnCFDTXQSTSm / RSCFDnTXQSTSm */ +#define RCANFD_TXQSTS(m) (0x03c0 + (0x04 * (m))) +/* RSCFDnCFDTXQPCTRm / RSCFDnTXQPCTRm */ +#define RCANFD_TXQPCTR(m) (0x03e0 + (0x04 * (m))) + +/* RSCFDnCFDTHLCCm / RSCFDnTHLCCm */ +#define RCANFD_THLCC(m) (0x0400 + (0x04 * (m))) +/* RSCFDnCFDTHLSTSm / RSCFDnTHLSTSm */ +#define RCANFD_THLSTS(m) (0x0420 + (0x04 * (m))) +/* RSCFDnCFDTHLPCTRm / RSCFDnTHLPCTRm */ +#define RCANFD_THLPCTR(m) (0x0440 + (0x04 * (m))) + +/* RSCFDnCFDGTINTSTS0 / RSCFDnGTINTSTS0 */ +#define RCANFD_GTINTSTS0 (0x0460) +/* RSCFDnCFDGTINTSTS1 / RSCFDnGTINTSTS1 */ +#define RCANFD_GTINTSTS1 (0x0464) +/* RSCFDnCFDGTSTCFG / RSCFDnGTSTCFG */ +#define RCANFD_GTSTCFG (0x0468) +/* RSCFDnCFDGTSTCTR / RSCFDnGTSTCTR */ +#define RCANFD_GTSTCTR (0x046c) +/* RSCFDnCFDGLOCKK / RSCFDnGLOCKK */ +#define RCANFD_GLOCKK (0x047c) +/* RSCFDnCFDGRMCFG / RSCFDnGRMCFG */ +#define RCANFD_GRMCFG (0x04fc) + +/* RSCFDnCFDGAFLIDj / RSCFDnGAFLIDj */ +#define RCANFD_GAFLID(offset, j) ((offset) + (0x10 * (j))) +/* RSCFDnCFDGAFLMj / RSCFDnGAFLMj */ +#define RCANFD_GAFLM(offset, j) ((offset) + 0x04 + (0x10 * (j))) +/* RSCFDnCFDGAFLP0j / RSCFDnGAFLP0j */ +#define RCANFD_GAFLP0(offset, j) ((offset) + 0x08 + (0x10 * (j))) +/* RSCFDnCFDGAFLP1j / RSCFDnGAFLP1j */ +#define RCANFD_GAFLP1(offset, j) ((offset) + 0x0c + (0x10 * (j))) + +/* CAN FD mode specific regsiter map */ + +/* RSCFDnCFDCmXXX -> RCANFD_F_XXX(m) */ +#define RCANFD_F_DCFG(m) (0x0500 + (0x20 * (m))) +#define RCANFD_F_CFDCFG(m) (0x0504 + (0x20 * (m))) +#define RCANFD_F_CFDCTR(m) (0x0508 + (0x20 * (m))) +#define RCANFD_F_CFDSTS(m) (0x050c + (0x20 * (m))) +#define RCANFD_F_CFDCRC(m) (0x0510 + (0x20 * (m))) + +/* RSCFDnCFDGAFLXXXj offset */ +#define RCANFD_F_GAFL_OFFSET (0x1000) + +/* RSCFDnCFDRMXXXq -> RCANFD_F_RMXXX(q) */ +#define RCANFD_F_RMID(q) (0x2000 + (0x20 * (q))) +#define RCANFD_F_RMPTR(q) (0x2004 + (0x20 * (q))) +#define RCANFD_F_RMFDSTS(q) (0x2008 + (0x20 * (q))) +#define RCANFD_F_RMDF(q, b) (0x200c + (0x04 * (b)) + (0x20 * (q))) + +/* RSCFDnCFDRFXXx -> RCANFD_F_RFXX(x) */ +#define RCANFD_F_RFOFFSET (0x3000) +#define RCANFD_F_RFID(x) (RCANFD_F_RFOFFSET + (0x80 * (x))) +#define RCANFD_F_RFPTR(x) (RCANFD_F_RFOFFSET + 0x04 + \ + (0x80 * (x))) +#define RCANFD_F_RFFDSTS(x) (RCANFD_F_RFOFFSET + 0x08 + \ + (0x80 * (x))) +#define RCANFD_F_RFDF(x, df) (RCANFD_F_RFOFFSET + 0x0c + \ + (0x80 * (x)) + (0x04 * (df))) + +/* RSCFDnCFDCFXXk -> RCANFD_F_CFXX(ch, k) */ +#define RCANFD_F_CFOFFSET (0x3400) +#define RCANFD_F_CFID(ch, idx) (RCANFD_F_CFOFFSET + (0x180 * (ch)) + \ + (0x80 * (idx))) +#define RCANFD_F_CFPTR(ch, idx) (RCANFD_F_CFOFFSET + 0x04 + \ + (0x180 * (ch)) + (0x80 * (idx))) +#define RCANFD_F_CFFDCSTS(ch, idx) (RCANFD_F_CFOFFSET + 0x08 + \ + (0x180 * (ch)) + (0x80 * (idx))) +#define RCANFD_F_CFDF(ch, idx, df) (RCANFD_F_CFOFFSET + 0x0c + \ + (0x180 * (ch)) + (0x80 * (idx)) + \ + (0x04 * (df))) + +/* RSCFDnCFDTMXXp -> RCANFD_F_TMXX(p) */ +#define RCANFD_F_TMID(p) (0x4000 + (0x20 * (p))) +#define RCANFD_F_TMPTR(p) (0x4004 + (0x20 * (p))) +#define RCANFD_F_TMFDCTR(p) (0x4008 + (0x20 * (p))) +#define RCANFD_F_TMDF(p, b) (0x400c + (0x20 * (p)) + (0x04 * (b))) + +/* RSCFDnCFDTHLACCm */ +#define RCANFD_F_THLACC(m) (0x6000 + (0x04 * (m))) +/* RSCFDnCFDRPGACCr */ +#define RCANFD_F_RPGACC(r) (0x6400 + (0x04 * (r))) + +/* Constants */ +#define RCANFD_FIFO_DEPTH 8 /* Tx FIFO depth */ +#define RCANFD_NAPI_WEIGHT 8 /* Rx poll quota */ + +#define RCANFD_NUM_CHANNELS 2 /* Two channels max */ +#define RCANFD_CHANNELS_MASK BIT((RCANFD_NUM_CHANNELS) - 1) + +#define RCANFD_GAFL_PAGENUM(entry) ((entry) / 16) +#define RCANFD_CHANNEL_NUMRULES 1 /* only one rule per channel */ + +/* Rx FIFO is a global resource of the controller. There are 8 such FIFOs + * available. Each channel gets a dedicated Rx FIFO (i.e.) the channel + * number is added to RFFIFO index. + */ +#define RCANFD_RFFIFO_IDX 0 + +/* Tx/Rx or Common FIFO is a per channel resource. Each channel has 3 Common + * FIFOs dedicated to them. Use the first (index 0) FIFO out of the 3 for Tx. + */ +#define RCANFD_CFFIFO_IDX 0 + +/* fCAN clock select register settings */ +enum rcar_canfd_fcanclk { + RCANFD_CANFDCLK = 0, /* CANFD clock */ + RCANFD_EXTCLK, /* Externally input clock */ +}; + +struct rcar_canfd_global; + +/* Channel priv data */ +struct rcar_canfd_channel { + struct can_priv can; /* Must be the first member */ + struct net_device *ndev; + struct rcar_canfd_global *gpriv; /* Controller reference */ + void __iomem *base; /* Register base address */ + struct napi_struct napi; + u8 tx_len[RCANFD_FIFO_DEPTH]; /* For net stats */ + u32 tx_head; /* Incremented on xmit */ + u32 tx_tail; /* Incremented on xmit done */ + u32 channel; /* Channel number */ + spinlock_t tx_lock; /* To protect tx path */ +}; + +/* Global priv data */ +struct rcar_canfd_global { + struct rcar_canfd_channel *ch[RCANFD_NUM_CHANNELS]; + void __iomem *base; /* Register base address */ + struct platform_device *pdev; /* Respective platform device */ + struct clk *clkp; /* Peripheral clock */ + struct clk *can_clk; /* fCAN clock */ + enum rcar_canfd_fcanclk fcan; /* CANFD or Ext clock */ + unsigned long channels_mask; /* Enabled channels mask */ +}; + +/* CAN FD mode nominal rate constants */ +static const struct can_bittiming_const rcar_canfd_nom_bittiming_const = { + .name = RCANFD_DRV_NAME, + .tseg1_min = 2, + .tseg1_max = 128, + .tseg2_min = 2, + .tseg2_max = 32, + .sjw_max = 32, + .brp_min = 1, + .brp_max = 1024, + .brp_inc = 1, +}; + +/* CAN FD mode data rate constants */ +static const struct can_bittiming_const rcar_canfd_data_bittiming_const = { + .name = RCANFD_DRV_NAME, + .tseg1_min = 2, + .tseg1_max = 16, + .tseg2_min = 2, + .tseg2_max = 8, + .sjw_max = 8, + .brp_min = 1, + .brp_max = 256, + .brp_inc = 1, +}; + +/* Helper functions */ +static inline void rcar_canfd_update(u32 mask, u32 val, u32 __iomem *reg) +{ + u32 data = readl(reg); + + data &= ~mask; + data |= (val & mask); + writel(data, reg); +} + +static inline u32 rcar_canfd_read(void __iomem *base, u32 offset) +{ + return readl(base + (offset)); +} + +static inline void rcar_canfd_write(void __iomem *base, u32 offset, u32 val) +{ + writel(val, base + (offset)); +} + +static void rcar_canfd_set_bit(void __iomem *base, u32 reg, u32 val) +{ + rcar_canfd_update(val, val, base + (reg)); +} + +static void rcar_canfd_clear_bit(void __iomem *base, u32 reg, u32 val) +{ + rcar_canfd_update(val, 0, base + (reg)); +} + +static void rcar_canfd_update_bit(void __iomem *base, u32 reg, + u32 mask, u32 val) +{ + rcar_canfd_update(mask, val, base + (reg)); +} + +static void rcar_canfd_get_data(struct rcar_canfd_channel *priv, + struct canfd_frame *cf, u32 off) +{ + u32 i, lwords; + + lwords = DIV_ROUND_UP(cf->len, sizeof(u32)); + for (i = 0; i < lwords; i++) + *((u32 *)cf->data + i) = + rcar_canfd_read(priv->base, off + (i * sizeof(u32))); +} + +static void rcar_canfd_put_data(struct rcar_canfd_channel *priv, + struct canfd_frame *cf, u32 off) +{ + u32 i, lwords; + + lwords = DIV_ROUND_UP(cf->len, sizeof(u32)); + for (i = 0; i < lwords; i++) + rcar_canfd_write(priv->base, off + (i * sizeof(u32)), + *((u32 *)cf->data + i)); +} + +static void rcar_canfd_tx_failure_cleanup(struct net_device *ndev) +{ + u32 i; + + for (i = 0; i < RCANFD_FIFO_DEPTH; i++) + can_free_echo_skb(ndev, i); +} + +static int rcar_canfd_reset_controller(struct rcar_canfd_global *gpriv) +{ + u32 sts, ch; + int err; + + /* Check RAMINIT flag as CAN RAM initialization takes place + * after the MCU reset + */ + err = readl_poll_timeout((gpriv->base + RCANFD_GSTS), sts, + !(sts & RCANFD_GSTS_GRAMINIT), 2, 500000); + if (err) { + dev_dbg(&gpriv->pdev->dev, "global raminit failed\n"); + return err; + } + + /* Transition to Global Reset mode */ + rcar_canfd_clear_bit(gpriv->base, RCANFD_GCTR, RCANFD_GCTR_GSLPR); + rcar_canfd_update_bit(gpriv->base, RCANFD_GCTR, + RCANFD_GCTR_GMDC_MASK, RCANFD_GCTR_GMDC_GRESET); + + /* Ensure Global reset mode */ + err = readl_poll_timeout((gpriv->base + RCANFD_GSTS), sts, + (sts & RCANFD_GSTS_GRSTSTS), 2, 500000); + if (err) { + dev_dbg(&gpriv->pdev->dev, "global reset failed\n"); + return err; + } + + /* Reset Global error flags */ + rcar_canfd_write(gpriv->base, RCANFD_GERFL, 0x0); + + /* Set the controller into FD mode */ + rcar_canfd_set_bit(gpriv->base, RCANFD_GRMCFG, RCANFD_GRMCFG_RCMC); + + /* Transition all Channels to reset mode */ + for_each_set_bit(ch, &gpriv->channels_mask, RCANFD_NUM_CHANNELS) { + rcar_canfd_clear_bit(gpriv->base, + RCANFD_CCTR(ch), RCANFD_CCTR_CSLPR); + + rcar_canfd_update_bit(gpriv->base, RCANFD_CCTR(ch), + RCANFD_CCTR_CHMDC_MASK, + RCANFD_CCTR_CHDMC_CRESET); + + /* Ensure Channel reset mode */ + err = readl_poll_timeout((gpriv->base + RCANFD_CSTS(ch)), sts, + (sts & RCANFD_CSTS_CRSTSTS), + 2, 500000); + if (err) { + dev_dbg(&gpriv->pdev->dev, + "channel %u reset failed\n", ch); + return err; + } + } + return 0; +} + +static void rcar_canfd_configure_controller(struct rcar_canfd_global *gpriv) +{ + u32 cfg, ch; + + /* Global configuration settings */ + + /* Truncate payload to configured message size RFPLS */ + cfg = RCANFD_GCFG_CMPOC; + + /* Set External Clock if selected */ + if (gpriv->fcan != RCANFD_CANFDCLK) + cfg |= RCANFD_GCFG_DCS; + + rcar_canfd_set_bit(gpriv->base, RCANFD_GCFG, cfg); + + /* Channel configuration settings */ + for_each_set_bit(ch, &gpriv->channels_mask, RCANFD_NUM_CHANNELS) { + rcar_canfd_set_bit(gpriv->base, RCANFD_CCTR(ch), + RCANFD_CCTR_ERRD); + rcar_canfd_update_bit(gpriv->base, RCANFD_CCTR(ch), + RCANFD_CCTR_BOM_MASK, + RCANFD_CCTR_BOM_BENTRY); + } +} + +static void rcar_canfd_configure_afl_rules(struct rcar_canfd_global *gpriv, + u32 ch) +{ + u32 cfg; + int start, page, num_rules = RCANFD_CHANNEL_NUMRULES; + u32 ridx = ch + RCANFD_RFFIFO_IDX; + + if (ch == 0) { + start = 0; /* Channel 0 always starts from 0th rule */ + } else { + /* Get number of Channel 0 rules and adjust */ + cfg = rcar_canfd_read(gpriv->base, RCANFD_GAFLCFG0); + start = RCANFD_GAFLCFG_GETRNC(0, cfg); + } + + /* Enable write access to entry */ + page = RCANFD_GAFL_PAGENUM(start); + rcar_canfd_set_bit(gpriv->base, RCANFD_GAFLECTR, + (RCANFD_GAFLECTR_AFLPN(page) | + RCANFD_GAFLECTR_AFLDAE)); + + /* Write number of rules for channel */ + rcar_canfd_set_bit(gpriv->base, RCANFD_GAFLCFG0, + RCANFD_GAFLCFG_SETRNC(ch, num_rules)); + + /* Accept all IDs */ + rcar_canfd_write(gpriv->base, + RCANFD_GAFLID(RCANFD_F_GAFL_OFFSET, start), 0); + /* IDE or RTR is not considered for matching */ + rcar_canfd_write(gpriv->base, + RCANFD_GAFLM(RCANFD_F_GAFL_OFFSET, start), 0); + /* Any data length accepted */ + rcar_canfd_write(gpriv->base, + RCANFD_GAFLP0(RCANFD_F_GAFL_OFFSET, start), 0); + /* Place the msg in corresponding Rx FIFO entry */ + rcar_canfd_write(gpriv->base, + RCANFD_GAFLP1(RCANFD_F_GAFL_OFFSET, start), + RCANFD_GAFLP1_GAFLFDP(ridx)); + + /* Disable write access to page */ + rcar_canfd_clear_bit(gpriv->base, + RCANFD_GAFLECTR, RCANFD_GAFLECTR_AFLDAE); +} + +static void rcar_canfd_configure_rx(struct rcar_canfd_global *gpriv, u32 ch) +{ + /* Rx FIFO is used for reception */ + u32 cfg; + u16 rfdc, rfpls; + + /* Select Rx FIFO based on channel */ + u32 ridx = ch + RCANFD_RFFIFO_IDX; + + rfdc = 2; /* b010 - 8 messages Rx FIFO depth */ + rfpls = 7; /* b111 - Max 64 bytes payload */ + + cfg = (RCANFD_RFCC_RFIM | RCANFD_RFCC_RFDC(rfdc) | + RCANFD_RFCC_RFPLS(rfpls) | RCANFD_RFCC_RFIE); + rcar_canfd_write(gpriv->base, RCANFD_RFCC(ridx), cfg); +} + +static void rcar_canfd_configure_tx(struct rcar_canfd_global *gpriv, u32 ch) +{ + /* Tx/Rx(Common) FIFO configured in Tx mode is + * used for transmission + * + * Each channel has 3 Common FIFO dedicated to them. + * Use the 1st (index 0) out of 3 + */ + u32 cfg; + u16 cftml, cfm, cfdc, cfpls; + + cftml = 0; /* 0th buffer */ + cfm = 1; /* b01 - Transmit mode */ + cfdc = 2; /* b010 - 8 messages Tx FIFO depth */ + cfpls = 7; /* b111 - Max 64 bytes payload */ + + cfg = (RCANFD_CFCC_CFTML(cftml) | RCANFD_CFCC_CFM(cfm) | + RCANFD_CFCC_CFIM | RCANFD_CFCC_CFDC(cfdc) | + RCANFD_CFCC_CFPLS(cfpls) | RCANFD_CFCC_CFTXIE); + rcar_canfd_write(gpriv->base, RCANFD_CFCC(ch, RCANFD_CFFIFO_IDX), cfg); + + /* Clear FD mode specific control/status register */ + rcar_canfd_write(gpriv->base, + RCANFD_F_CFFDCSTS(ch, RCANFD_CFFIFO_IDX), 0); +} + +static void rcar_canfd_enable_global_interrupts(struct rcar_canfd_global *gpriv) +{ + u32 ctr; + + /* Clear any stray error interrupt flags */ + rcar_canfd_write(gpriv->base, RCANFD_GERFL, 0); + + /* Global interrupts setup */ + ctr = RCANFD_GCTR_MEIE; + ctr |= RCANFD_GCTR_CFMPOFIE; + + rcar_canfd_set_bit(gpriv->base, RCANFD_GCTR, ctr); +} + +static void rcar_canfd_disable_global_interrupts(struct rcar_canfd_global + *gpriv) +{ + /* Disable all interrupts */ + rcar_canfd_write(gpriv->base, RCANFD_GCTR, 0); + + /* Clear any stray error interrupt flags */ + rcar_canfd_write(gpriv->base, RCANFD_GERFL, 0); +} + +static void rcar_canfd_enable_channel_interrupts(struct rcar_canfd_channel + *priv) +{ + u32 ctr, ch = priv->channel; + + /* Clear any stray error flags */ + rcar_canfd_write(priv->base, RCANFD_CERFL(ch), 0); + + /* Channel interrupts setup */ + ctr = (RCANFD_CCTR_TAIE | + RCANFD_CCTR_ALIE | RCANFD_CCTR_BLIE | + RCANFD_CCTR_OLIE | RCANFD_CCTR_BORIE | + RCANFD_CCTR_BOEIE | RCANFD_CCTR_EPIE | + RCANFD_CCTR_EWIE | RCANFD_CCTR_BEIE); + rcar_canfd_set_bit(priv->base, RCANFD_CCTR(ch), ctr); +} + +static void rcar_canfd_disable_channel_interrupts(struct rcar_canfd_channel + *priv) +{ + u32 ctr, ch = priv->channel; + + ctr = (RCANFD_CCTR_TAIE | + RCANFD_CCTR_ALIE | RCANFD_CCTR_BLIE | + RCANFD_CCTR_OLIE | RCANFD_CCTR_BORIE | + RCANFD_CCTR_BOEIE | RCANFD_CCTR_EPIE | + RCANFD_CCTR_EWIE | RCANFD_CCTR_BEIE); + rcar_canfd_clear_bit(priv->base, RCANFD_CCTR(ch), ctr); + + /* Clear any stray error flags */ + rcar_canfd_write(priv->base, RCANFD_CERFL(ch), 0); +} + +static void rcar_canfd_global_error(struct net_device *ndev) +{ + struct rcar_canfd_channel *priv = netdev_priv(ndev); + struct net_device_stats *stats = &ndev->stats; + u32 ch = priv->channel; + u32 gerfl, sts; + u32 ridx = ch + RCANFD_RFFIFO_IDX; + + gerfl = rcar_canfd_read(priv->base, RCANFD_GERFL); + if ((gerfl & RCANFD_GERFL_EEF0) && (ch == 0)) { + netdev_dbg(ndev, "Ch0: ECC Error flag\n"); + stats->tx_dropped++; + } + if ((gerfl & RCANFD_GERFL_EEF1) && (ch == 1)) { + netdev_dbg(ndev, "Ch1: ECC Error flag\n"); + stats->tx_dropped++; + } + if (gerfl & RCANFD_GERFL_MES) { + sts = rcar_canfd_read(priv->base, + RCANFD_CFSTS(ch, RCANFD_CFFIFO_IDX)); + if (sts & RCANFD_CFSTS_CFMLT) { + netdev_dbg(ndev, "Tx Message Lost flag\n"); + stats->tx_dropped++; + rcar_canfd_write(priv->base, + RCANFD_CFSTS(ch, RCANFD_CFFIFO_IDX), + sts & ~RCANFD_CFSTS_CFMLT); + } + + sts = rcar_canfd_read(priv->base, RCANFD_RFSTS(ridx)); + if (sts & RCANFD_RFSTS_RFMLT) { + netdev_dbg(ndev, "Rx Message Lost flag\n"); + stats->rx_dropped++; + rcar_canfd_write(priv->base, RCANFD_RFSTS(ridx), + sts & ~RCANFD_RFSTS_RFMLT); + } + } + if (gerfl & RCANFD_GERFL_CMPOF) { + /* Message Lost flag will be set for respective channel + * when this condition happens with counters and flags + * already updated. + */ + netdev_dbg(ndev, "global payload overflow interrupt\n"); + } + + /* Clear all global error interrupts. Only affected channels bits + * get cleared + */ + rcar_canfd_write(priv->base, RCANFD_GERFL, 0); +} + +static void rcar_canfd_error(struct net_device *ndev) +{ + struct rcar_canfd_channel *priv = netdev_priv(ndev); + struct net_device_stats *stats = &ndev->stats; + struct can_frame *cf; + struct sk_buff *skb; + u32 cerfl, csts; + u32 txerr = 0, rxerr = 0; + u32 ch = priv->channel; + + /* Propagate the error condition to the CAN stack */ + skb = alloc_can_err_skb(ndev, &cf); + if (!skb) { + stats->rx_dropped++; + return; + } + + /* Channel error interrupt */ + cerfl = rcar_canfd_read(priv->base, RCANFD_CERFL(ch)); + csts = rcar_canfd_read(priv->base, RCANFD_CSTS(ch)); + txerr = RCANFD_CSTS_TECCNT(csts); + rxerr = RCANFD_CSTS_RECCNT(csts); + + netdev_dbg(ndev, "ch erfl %x sts %x txerr %u rxerr %u\n", + cerfl, csts, txerr, rxerr); + + if (cerfl & RCANFD_CERFL_BEF) { + netdev_dbg(ndev, "Bus error\n"); + cf->can_id |= CAN_ERR_BUSERROR | CAN_ERR_PROT; + cf->data[2] = CAN_ERR_PROT_UNSPEC; + priv->can.can_stats.bus_error++; + } + if (cerfl & RCANFD_CERFL_ADERR) { + netdev_dbg(ndev, "ACK Delimiter Error\n"); + stats->tx_errors++; + cf->data[3] |= CAN_ERR_PROT_LOC_ACK_DEL; + } + if (cerfl & RCANFD_CERFL_B0ERR) { + netdev_dbg(ndev, "Bit Error (dominant)\n"); + stats->tx_errors++; + cf->data[2] |= CAN_ERR_PROT_BIT0; + } + if (cerfl & RCANFD_CERFL_B1ERR) { + netdev_dbg(ndev, "Bit Error (recessive)\n"); + stats->tx_errors++; + cf->data[2] |= CAN_ERR_PROT_BIT1; + } + if (cerfl & RCANFD_CERFL_CERR) { + netdev_dbg(ndev, "CRC Error\n"); + stats->rx_errors++; + cf->data[3] |= CAN_ERR_PROT_LOC_CRC_SEQ; + } + if (cerfl & RCANFD_CERFL_AERR) { + netdev_dbg(ndev, "ACK Error\n"); + stats->tx_errors++; + cf->can_id |= CAN_ERR_ACK; + cf->data[3] |= CAN_ERR_PROT_LOC_ACK; + } + if (cerfl & RCANFD_CERFL_FERR) { + netdev_dbg(ndev, "Form Error\n"); + stats->rx_errors++; + cf->data[2] |= CAN_ERR_PROT_FORM; + } + if (cerfl & RCANFD_CERFL_SERR) { + netdev_dbg(ndev, "Stuff Error\n"); + stats->rx_errors++; + cf->data[2] |= CAN_ERR_PROT_STUFF; + } + if (cerfl & RCANFD_CERFL_ALF) { + netdev_dbg(ndev, "Arbitration lost Error\n"); + priv->can.can_stats.arbitration_lost++; + cf->can_id |= CAN_ERR_LOSTARB; + cf->data[0] |= CAN_ERR_LOSTARB_UNSPEC; + } + if (cerfl & RCANFD_CERFL_BLF) { + netdev_dbg(ndev, "Bus Lock Error\n"); + stats->rx_errors++; + cf->can_id |= CAN_ERR_BUSERROR; + } + if (cerfl & RCANFD_CERFL_EWF) { + netdev_dbg(ndev, "Error warning interrupt\n"); + priv->can.state = CAN_STATE_ERROR_WARNING; + priv->can.can_stats.error_warning++; + cf->can_id |= CAN_ERR_CRTL; + cf->data[1] = txerr > rxerr ? CAN_ERR_CRTL_TX_WARNING : + CAN_ERR_CRTL_RX_WARNING; + cf->data[6] = txerr; + cf->data[7] = rxerr; + } + if (cerfl & RCANFD_CERFL_EPF) { + netdev_dbg(ndev, "Error passive interrupt\n"); + priv->can.state = CAN_STATE_ERROR_PASSIVE; + priv->can.can_stats.error_passive++; + cf->can_id |= CAN_ERR_CRTL; + cf->data[1] = txerr > rxerr ? CAN_ERR_CRTL_TX_PASSIVE : + CAN_ERR_CRTL_RX_PASSIVE; + cf->data[6] = txerr; + cf->data[7] = rxerr; + } + if (cerfl & RCANFD_CERFL_BOEF) { + netdev_dbg(ndev, "Bus-off entry interrupt\n"); + rcar_canfd_tx_failure_cleanup(ndev); + priv->can.state = CAN_STATE_BUS_OFF; + priv->can.can_stats.bus_off++; + can_bus_off(ndev); + cf->can_id |= CAN_ERR_BUSOFF; + } + if (cerfl & RCANFD_CERFL_OVLF) { + netdev_dbg(ndev, + "Overload Frame Transmission error interrupt\n"); + stats->tx_errors++; + cf->can_id |= CAN_ERR_PROT; + cf->data[2] |= CAN_ERR_PROT_OVERLOAD; + } + + /* Clear all channel error interrupts */ + rcar_canfd_write(priv->base, RCANFD_CERFL(ch), 0); + stats->rx_packets++; + stats->rx_bytes += cf->can_dlc; + netif_rx(skb); +} + +static void rcar_canfd_tx_done(struct net_device *ndev) +{ + struct rcar_canfd_channel *priv = netdev_priv(ndev); + struct net_device_stats *stats = &ndev->stats; + u32 sts; + unsigned long flags; + u32 ch = priv->channel; + + do { + u8 unsent, sent; + + sent = priv->tx_tail % RCANFD_FIFO_DEPTH; + stats->tx_packets++; + stats->tx_bytes += priv->tx_len[sent]; + priv->tx_len[sent] = 0; + can_get_echo_skb(ndev, sent); + + spin_lock_irqsave(&priv->tx_lock, flags); + priv->tx_tail++; + sts = rcar_canfd_read(priv->base, + RCANFD_CFSTS(ch, RCANFD_CFFIFO_IDX)); + unsent = RCANFD_CFSTS_CFMC(sts); + + /* Wake producer only when there is room */ + if (unsent != RCANFD_FIFO_DEPTH) + netif_wake_queue(ndev); + + if (priv->tx_head - priv->tx_tail <= unsent) { + spin_unlock_irqrestore(&priv->tx_lock, flags); + break; + } + spin_unlock_irqrestore(&priv->tx_lock, flags); + + } while (1); + + /* Clear interrupt */ + rcar_canfd_write(priv->base, RCANFD_CFSTS(ch, RCANFD_CFFIFO_IDX), + sts & ~RCANFD_CFSTS_CFTXIF); + can_led_event(ndev, CAN_LED_EVENT_TX); +} + +static irqreturn_t rcar_canfd_global_interrupt(int irq, void *dev_id) +{ + struct rcar_canfd_global *gpriv = dev_id; + struct net_device *ndev; + struct rcar_canfd_channel *priv; + u32 sts, gerfl; + u32 ch, ridx; + + /* Global error interrupts still indicate a condition specific + * to a channel. RxFIFO interrupt is a global interrupt. + */ + for_each_set_bit(ch, &gpriv->channels_mask, RCANFD_NUM_CHANNELS) { + priv = gpriv->ch[ch]; + ndev = priv->ndev; + ridx = ch + RCANFD_RFFIFO_IDX; + + /* Global error interrupts */ + gerfl = rcar_canfd_read(priv->base, RCANFD_GERFL); + if (RCANFD_GERFL_ERR(gerfl)) + rcar_canfd_global_error(ndev); + + /* Handle Rx interrupts */ + sts = rcar_canfd_read(priv->base, RCANFD_RFSTS(ridx)); + if (sts & RCANFD_RFSTS_RFIF) { + if (napi_schedule_prep(&priv->napi)) { + /* Disable Rx FIFO interrupts */ + rcar_canfd_clear_bit(priv->base, + RCANFD_RFCC(ridx), + RCANFD_RFCC_RFIE); + __napi_schedule(&priv->napi); + } + } + } + return IRQ_HANDLED; +} + +static irqreturn_t rcar_canfd_channel_interrupt(int irq, void *dev_id) +{ + struct rcar_canfd_global *gpriv = dev_id; + struct net_device *ndev; + struct rcar_canfd_channel *priv; + u32 sts, cerfl, ch; + + /* Common FIFO is a per channel resource */ + for_each_set_bit(ch, &gpriv->channels_mask, RCANFD_NUM_CHANNELS) { + priv = gpriv->ch[ch]; + ndev = priv->ndev; + + /* Channel error interrupts */ + cerfl = rcar_canfd_read(priv->base, RCANFD_CERFL(ch)); + if (RCANFD_CERFL_ERR(cerfl)) + rcar_canfd_error(ndev); + + /* Handle Tx interrupts */ + sts = rcar_canfd_read(priv->base, + RCANFD_CFSTS(ch, RCANFD_CFFIFO_IDX)); + if (sts & RCANFD_CFSTS_CFTXIF) + rcar_canfd_tx_done(ndev); + } + return IRQ_HANDLED; +} + +static void rcar_canfd_set_bittiming(struct net_device *dev) +{ + struct rcar_canfd_channel *priv = netdev_priv(dev); + const struct can_bittiming *bt = &priv->can.bittiming; + const struct can_bittiming *dbt = &priv->can.data_bittiming; + u16 brp, sjw, tseg1, tseg2; + u32 cfg; + u32 ch = priv->channel; + + /* Nominal bit timing settings */ + brp = bt->brp - 1; + sjw = bt->sjw - 1; + tseg1 = bt->prop_seg + bt->phase_seg1 - 1; + tseg2 = bt->phase_seg2 - 1; + + cfg = (RCANFD_NCFG_NTSEG1(tseg1) | RCANFD_NCFG_NBRP(brp) | + RCANFD_NCFG_NSJW(sjw) | RCANFD_NCFG_NTSEG2(tseg2)); + + rcar_canfd_write(priv->base, RCANFD_CCFG(ch), cfg); + netdev_dbg(priv->ndev, "nrate: brp %u, sjw %u, tseg1 %u, tseg2 %u\n", + brp, sjw, tseg1, tseg2); + + /* Data bit timing settings */ + brp = dbt->brp - 1; + sjw = dbt->sjw - 1; + tseg1 = dbt->prop_seg + dbt->phase_seg1 - 1; + tseg2 = dbt->phase_seg2 - 1; + + cfg = (RCANFD_DCFG_DTSEG1(tseg1) | RCANFD_DCFG_DBRP(brp) | + RCANFD_DCFG_DSJW(sjw) | RCANFD_DCFG_DTSEG2(tseg2)); + + rcar_canfd_write(priv->base, RCANFD_F_DCFG(ch), cfg); + netdev_dbg(priv->ndev, "drate: brp %u, sjw %u, tseg1 %u, tseg2 %u\n", + brp, sjw, tseg1, tseg2); +} + +static int rcar_canfd_start(struct net_device *ndev) +{ + struct rcar_canfd_channel *priv = netdev_priv(ndev); + int err = -EOPNOTSUPP; + u32 sts, ch = priv->channel; + u32 ridx = ch + RCANFD_RFFIFO_IDX; + + rcar_canfd_set_bittiming(ndev); + + rcar_canfd_enable_channel_interrupts(priv); + + /* Set channel to Operational mode */ + rcar_canfd_update_bit(priv->base, RCANFD_CCTR(ch), + RCANFD_CCTR_CHMDC_MASK, RCANFD_CCTR_CHDMC_COPM); + + /* Verify channel mode change */ + err = readl_poll_timeout((priv->base + RCANFD_CSTS(ch)), sts, + (sts & RCANFD_CSTS_COMSTS), 2, 500000); + if (err) { + netdev_err(ndev, "channel %u communication state failed\n", ch); + goto fail_mode_change; + } + + /* Enable Common & Rx FIFO */ + rcar_canfd_set_bit(priv->base, RCANFD_CFCC(ch, RCANFD_CFFIFO_IDX), + RCANFD_CFCC_CFE); + rcar_canfd_set_bit(priv->base, RCANFD_RFCC(ridx), RCANFD_RFCC_RFE); + + priv->can.state = CAN_STATE_ERROR_ACTIVE; + return 0; + +fail_mode_change: + rcar_canfd_disable_channel_interrupts(priv); + return err; +} + +static int rcar_canfd_open(struct net_device *ndev) +{ + struct rcar_canfd_channel *priv = netdev_priv(ndev); + struct rcar_canfd_global *gpriv = priv->gpriv; + int err; + + /* Peripheral clock is already enabled in probe */ + err = clk_prepare_enable(gpriv->can_clk); + if (err) { + netdev_err(ndev, "failed to enable CAN clock, error %d\n", err); + goto out_clock; + } + + err = open_candev(ndev); + if (err) { + netdev_err(ndev, "open_candev() failed, error %d\n", err); + goto out_can_clock; + } + + napi_enable(&priv->napi); + err = rcar_canfd_start(ndev); + if (err) + goto out_close; + netif_start_queue(ndev); + can_led_event(ndev, CAN_LED_EVENT_OPEN); + return 0; +out_close: + napi_disable(&priv->napi); + close_candev(ndev); +out_can_clock: + clk_disable_unprepare(gpriv->can_clk); +out_clock: + return err; +} + +static void rcar_canfd_stop(struct net_device *ndev) +{ + struct rcar_canfd_channel *priv = netdev_priv(ndev); + int err; + u32 sts, ch = priv->channel; + u32 ridx = ch + RCANFD_RFFIFO_IDX; + + /* Transition to channel reset mode */ + rcar_canfd_update_bit(priv->base, RCANFD_CCTR(ch), + RCANFD_CCTR_CHMDC_MASK, RCANFD_CCTR_CHDMC_CRESET); + + /* Check Channel reset mode */ + err = readl_poll_timeout((priv->base + RCANFD_CSTS(ch)), sts, + (sts & RCANFD_CSTS_CRSTSTS), 2, 500000); + if (err) + netdev_err(ndev, "channel %u reset failed\n", ch); + + rcar_canfd_disable_channel_interrupts(priv); + + /* Disable Common & Rx FIFO */ + rcar_canfd_clear_bit(priv->base, RCANFD_CFCC(ch, RCANFD_CFFIFO_IDX), + RCANFD_CFCC_CFE); + rcar_canfd_clear_bit(priv->base, RCANFD_RFCC(ridx), RCANFD_RFCC_RFE); + + /* Set the state as STOPPED */ + priv->can.state = CAN_STATE_STOPPED; +} + +static int rcar_canfd_close(struct net_device *ndev) +{ + struct rcar_canfd_channel *priv = netdev_priv(ndev); + struct rcar_canfd_global *gpriv = priv->gpriv; + + netif_stop_queue(ndev); + rcar_canfd_stop(ndev); + napi_disable(&priv->napi); + clk_disable_unprepare(gpriv->can_clk); + close_candev(ndev); + can_led_event(ndev, CAN_LED_EVENT_STOP); + return 0; +} + +static netdev_tx_t rcar_canfd_start_xmit(struct sk_buff *skb, + struct net_device *ndev) +{ + struct rcar_canfd_channel *priv = netdev_priv(ndev); + struct canfd_frame *cf = (struct canfd_frame *)skb->data; + u32 sts = 0, id, dlc; + unsigned long flags; + u32 ch = priv->channel; + + if (can_dropped_invalid_skb(ndev, skb)) + return NETDEV_TX_OK; + + if (cf->can_id & CAN_EFF_FLAG) { + id = cf->can_id & CAN_EFF_MASK; + id |= RCANFD_CFID_CFIDE; + } else { + id = cf->can_id & CAN_SFF_MASK; + } + + if (cf->can_id & CAN_RTR_FLAG) + id |= RCANFD_CFID_CFRTR; + + rcar_canfd_write(priv->base, + RCANFD_F_CFID(ch, RCANFD_CFFIFO_IDX), id); + dlc = RCANFD_CFPTR_CFDLC(can_len2dlc(cf->len)); + rcar_canfd_write(priv->base, + RCANFD_F_CFPTR(ch, RCANFD_CFFIFO_IDX), dlc); + + if (can_is_canfd_skb(skb)) { + /* CAN FD frame format */ + sts |= RCANFD_CFFDCSTS_CFFDF; + if (cf->flags & CANFD_BRS) + sts |= RCANFD_CFFDCSTS_CFBRS; + + if (priv->can.state == CAN_STATE_ERROR_PASSIVE) + sts |= RCANFD_CFFDCSTS_CFESI; + } + + rcar_canfd_write(priv->base, RCANFD_F_CFFDCSTS(ch, RCANFD_CFFIFO_IDX), + sts); + + rcar_canfd_put_data(priv, cf, + RCANFD_F_CFDF(ch, RCANFD_CFFIFO_IDX, 0)); + + priv->tx_len[priv->tx_head % RCANFD_FIFO_DEPTH] = cf->len; + can_put_echo_skb(skb, ndev, priv->tx_head % RCANFD_FIFO_DEPTH); + + spin_lock_irqsave(&priv->tx_lock, flags); + priv->tx_head++; + + /* Stop the queue if we've filled all FIFO entries */ + if (priv->tx_head - priv->tx_tail >= RCANFD_FIFO_DEPTH) + netif_stop_queue(ndev); + + /* Start Tx: Write 0xff to CFPC to increment the CPU-side + * pointer for the Common FIFO + */ + rcar_canfd_write(priv->base, + RCANFD_CFPCTR(ch, RCANFD_CFFIFO_IDX), 0xff); + + spin_unlock_irqrestore(&priv->tx_lock, flags); + return NETDEV_TX_OK; +} + +static void rcar_canfd_rx_pkt(struct rcar_canfd_channel *priv) +{ + struct net_device_stats *stats = &priv->ndev->stats; + struct canfd_frame *cf; + struct sk_buff *skb; + u32 sts = 0, id, ptr; + u32 ch = priv->channel; + u32 ridx = ch + RCANFD_RFFIFO_IDX; + + id = rcar_canfd_read(priv->base, RCANFD_F_RFID(ridx)); + ptr = rcar_canfd_read(priv->base, RCANFD_F_RFPTR(ridx)); + + sts = rcar_canfd_read(priv->base, RCANFD_F_RFFDSTS(ridx)); + if (sts & RCANFD_RFFDSTS_RFFDF) + skb = alloc_canfd_skb(priv->ndev, &cf); + else + skb = alloc_can_skb(priv->ndev, + (struct can_frame **)&cf); + + if (!skb) { + stats->rx_dropped++; + return; + } + + if (sts & RCANFD_RFFDSTS_RFFDF) + cf->len = can_dlc2len(RCANFD_RFPTR_RFDLC(ptr)); + else + cf->len = get_can_dlc(RCANFD_RFPTR_RFDLC(ptr)); + + if (sts & RCANFD_RFFDSTS_RFESI) { + cf->flags |= CANFD_ESI; + netdev_dbg(priv->ndev, "ESI Error\n"); + } + + if (id & RCANFD_RFID_RFIDE) + cf->can_id = (id & CAN_EFF_MASK) | CAN_EFF_FLAG; + else + cf->can_id = id & CAN_SFF_MASK; + + if (!(sts & RCANFD_RFFDSTS_RFFDF) && (id & RCANFD_RFID_RFRTR)) { + cf->can_id |= CAN_RTR_FLAG; + } else { + if (sts & RCANFD_RFFDSTS_RFBRS) + cf->flags |= CANFD_BRS; + + rcar_canfd_get_data(priv, cf, RCANFD_F_RFDF(ridx, 0)); + } + + /* Write 0xff to RFPC to increment the CPU-side + * pointer of the Rx FIFO + */ + rcar_canfd_write(priv->base, RCANFD_RFPCTR(ridx), 0xff); + + can_led_event(priv->ndev, CAN_LED_EVENT_RX); + + stats->rx_bytes += cf->len; + stats->rx_packets++; + netif_receive_skb(skb); +} + +static int rcar_canfd_rx_poll(struct napi_struct *napi, int quota) +{ + struct rcar_canfd_channel *priv = + container_of(napi, struct rcar_canfd_channel, napi); + int num_pkts; + u32 sts; + u32 ch = priv->channel; + u32 ridx = ch + RCANFD_RFFIFO_IDX; + + for (num_pkts = 0; num_pkts < quota; num_pkts++) { + sts = rcar_canfd_read(priv->base, RCANFD_RFSTS(ridx)); + /* Check FIFO empty condition */ + if (sts & RCANFD_RFSTS_RFEMP) + break; + + rcar_canfd_rx_pkt(priv); + + /* Clear interrupt bit */ + if (sts & RCANFD_RFSTS_RFIF) + rcar_canfd_write(priv->base, RCANFD_RFSTS(ridx), + sts & ~RCANFD_RFSTS_RFIF); + } + + /* All packets processed */ + if (num_pkts < quota) { + napi_complete(napi); + /* Enable Rx FIFO interrupts */ + rcar_canfd_set_bit(priv->base, RCANFD_RFCC(ridx), + RCANFD_RFCC_RFIE); + } + return num_pkts; +} + +static int rcar_canfd_do_set_mode(struct net_device *ndev, enum can_mode mode) +{ + int err; + + switch (mode) { + case CAN_MODE_START: + err = rcar_canfd_start(ndev); + if (err) + return err; + netif_wake_queue(ndev); + return 0; + default: + return -EOPNOTSUPP; + } +} + +static int rcar_canfd_get_berr_counter(const struct net_device *dev, + struct can_berr_counter *bec) +{ + struct rcar_canfd_channel *priv = netdev_priv(dev); + u32 val, ch = priv->channel; + + /* Peripheral clock is already enabled in probe */ + val = rcar_canfd_read(priv->base, RCANFD_CSTS(ch)); + bec->txerr = RCANFD_CSTS_TECCNT(val); + bec->rxerr = RCANFD_CSTS_RECCNT(val); + return 0; +} + +static const struct net_device_ops rcar_canfd_netdev_ops = { + .ndo_open = rcar_canfd_open, + .ndo_stop = rcar_canfd_close, + .ndo_start_xmit = rcar_canfd_start_xmit, + .ndo_change_mtu = can_change_mtu, +}; + +static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch, + u32 fcan_freq) +{ + struct platform_device *pdev = gpriv->pdev; + struct rcar_canfd_channel *priv; + struct net_device *ndev; + int err = -ENODEV; + + ndev = alloc_candev(sizeof(*priv), RCANFD_FIFO_DEPTH); + if (!ndev) { + dev_err(&pdev->dev, "alloc_candev() failed\n"); + err = -ENOMEM; + goto fail; + } + priv = netdev_priv(ndev); + + ndev->netdev_ops = &rcar_canfd_netdev_ops; + ndev->flags |= IFF_ECHO; + priv->ndev = ndev; + priv->base = gpriv->base; + priv->channel = ch; + priv->can.clock.freq = fcan_freq; + dev_info(&pdev->dev, "can_clk rate is %u\n", priv->can.clock.freq); + + priv->can.bittiming_const = &rcar_canfd_nom_bittiming_const; + priv->can.data_bittiming_const = + &rcar_canfd_data_bittiming_const; + + /* Controller starts in CAN FD only mode */ + can_set_static_ctrlmode(ndev, CAN_CTRLMODE_FD); + priv->can.ctrlmode_supported = CAN_CTRLMODE_BERR_REPORTING; + + priv->can.do_set_mode = rcar_canfd_do_set_mode; + priv->can.do_get_berr_counter = rcar_canfd_get_berr_counter; + priv->gpriv = gpriv; + SET_NETDEV_DEV(ndev, &pdev->dev); + + netif_napi_add(ndev, &priv->napi, rcar_canfd_rx_poll, + RCANFD_NAPI_WEIGHT); + err = register_candev(ndev); + if (err) { + dev_err(&pdev->dev, + "register_candev() failed, error %d\n", err); + goto fail_candev; + } + spin_lock_init(&priv->tx_lock); + devm_can_led_init(ndev); + gpriv->ch[priv->channel] = priv; + dev_info(&pdev->dev, "device registered (channel %u)\n", priv->channel); + return 0; + +fail_candev: + netif_napi_del(&priv->napi); + free_candev(ndev); +fail: + return err; +} + +static void rcar_canfd_channel_remove(struct rcar_canfd_global *gpriv, u32 ch) +{ + struct rcar_canfd_channel *priv = gpriv->ch[ch]; + + if (priv) { + unregister_candev(priv->ndev); + netif_napi_del(&priv->napi); + free_candev(priv->ndev); + } +} + +static int rcar_canfd_probe(struct platform_device *pdev) +{ + struct resource *mem; + void __iomem *addr; + u32 sts, ch, fcan_freq; + struct rcar_canfd_global *gpriv; + struct device_node *of_child; + unsigned long channels_mask = 0; + int err, ch_irq, g_irq; + + of_child = of_get_child_by_name(pdev->dev.of_node, "channel0"); + if (of_child && of_device_is_available(of_child)) + channels_mask |= BIT(0); /* Channel 0 */ + + of_child = of_get_child_by_name(pdev->dev.of_node, "channel1"); + if (of_child && of_device_is_available(of_child)) + channels_mask |= BIT(1); /* Channel 1 */ + + ch_irq = platform_get_irq(pdev, 0); + if (ch_irq < 0) { + dev_err(&pdev->dev, "no Channel IRQ resource\n"); + err = ch_irq; + goto fail_dev; + } + + g_irq = platform_get_irq(pdev, 1); + if (g_irq < 0) { + dev_err(&pdev->dev, "no Global IRQ resource\n"); + err = g_irq; + goto fail_dev; + } + + /* Global controller context */ + gpriv = devm_kzalloc(&pdev->dev, sizeof(*gpriv), GFP_KERNEL); + if (!gpriv) { + err = -ENOMEM; + goto fail_dev; + } + gpriv->pdev = pdev; + gpriv->channels_mask = channels_mask; + + /* Peripheral clock */ + gpriv->clkp = devm_clk_get(&pdev->dev, "fck"); + if (IS_ERR(gpriv->clkp)) { + err = PTR_ERR(gpriv->clkp); + dev_err(&pdev->dev, "cannot get peripheral clock, error %d\n", + err); + goto fail_dev; + } + + /* fCAN clock: Pick External clock. If not available fallback to + * CANFD clock + */ + gpriv->can_clk = devm_clk_get(&pdev->dev, "can_clk"); + if (IS_ERR(gpriv->can_clk) || (clk_get_rate(gpriv->can_clk) == 0)) { + gpriv->can_clk = devm_clk_get(&pdev->dev, "canfd"); + if (IS_ERR(gpriv->can_clk)) { + err = PTR_ERR(gpriv->can_clk); + dev_err(&pdev->dev, + "cannot get canfd clock, error %d\n", err); + goto fail_dev; + } + gpriv->fcan = RCANFD_CANFDCLK; + + } else { + gpriv->fcan = RCANFD_EXTCLK; + } + fcan_freq = clk_get_rate(gpriv->can_clk); + + if (gpriv->fcan == RCANFD_CANFDCLK) + /* CANFD clock is further divided by (1/2) within the IP */ + fcan_freq /= 2; + + mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); + addr = devm_ioremap_resource(&pdev->dev, mem); + if (IS_ERR(addr)) { + err = PTR_ERR(addr); + goto fail_dev; + } + gpriv->base = addr; + + /* Request IRQ that's common for both channels */ + err = devm_request_irq(&pdev->dev, ch_irq, + rcar_canfd_channel_interrupt, 0, + "canfd.chn", gpriv); + if (err) { + dev_err(&pdev->dev, "devm_request_irq(%d) failed, error %d\n", + ch_irq, err); + goto fail_dev; + } + err = devm_request_irq(&pdev->dev, g_irq, + rcar_canfd_global_interrupt, 0, + "canfd.gbl", gpriv); + if (err) { + dev_err(&pdev->dev, "devm_request_irq(%d) failed, error %d\n", + g_irq, err); + goto fail_dev; + } + + /* Enable peripheral clock for register access */ + err = clk_prepare_enable(gpriv->clkp); + if (err) { + dev_err(&pdev->dev, + "failed to enable peripheral clock, error %d\n", err); + goto fail_dev; + } + + err = rcar_canfd_reset_controller(gpriv); + if (err) { + dev_err(&pdev->dev, "reset controller failed\n"); + goto fail_clk; + } + + /* Controller in Global reset & Channel reset mode */ + rcar_canfd_configure_controller(gpriv); + + /* Configure per channel attributes */ + for_each_set_bit(ch, &gpriv->channels_mask, RCANFD_NUM_CHANNELS) { + /* Configure Channel's Rx fifo */ + rcar_canfd_configure_rx(gpriv, ch); + + /* Configure Channel's Tx (Common) fifo */ + rcar_canfd_configure_tx(gpriv, ch); + + /* Configure receive rules */ + rcar_canfd_configure_afl_rules(gpriv, ch); + } + + /* Configure common interrupts */ + rcar_canfd_enable_global_interrupts(gpriv); + + /* Start Global operation mode */ + rcar_canfd_update_bit(gpriv->base, RCANFD_GCTR, RCANFD_GCTR_GMDC_MASK, + RCANFD_GCTR_GMDC_GOPM); + + /* Verify mode change */ + err = readl_poll_timeout((gpriv->base + RCANFD_GSTS), sts, + !(sts & RCANFD_GSTS_GNOPM), 2, 500000); + if (err) { + dev_err(&pdev->dev, "global operational mode failed\n"); + goto fail_mode; + } + + for_each_set_bit(ch, &gpriv->channels_mask, RCANFD_NUM_CHANNELS) { + err = rcar_canfd_channel_probe(gpriv, ch, fcan_freq); + if (err) + goto fail_channel; + } + + platform_set_drvdata(pdev, gpriv); + dev_info(&pdev->dev, "global operational state (clk %d)\n", + gpriv->fcan); + return 0; + +fail_channel: + for_each_set_bit(ch, &gpriv->channels_mask, RCANFD_NUM_CHANNELS) + rcar_canfd_channel_remove(gpriv, ch); +fail_mode: + rcar_canfd_disable_global_interrupts(gpriv); +fail_clk: + clk_disable_unprepare(gpriv->clkp); +fail_dev: + return err; +} + +static int rcar_canfd_remove(struct platform_device *pdev) +{ + struct rcar_canfd_global *gpriv = platform_get_drvdata(pdev); + u32 ch; + + rcar_canfd_reset_controller(gpriv); + rcar_canfd_disable_global_interrupts(gpriv); + + for_each_set_bit(ch, &gpriv->channels_mask, RCANFD_NUM_CHANNELS) { + rcar_canfd_disable_channel_interrupts(gpriv->ch[ch]); + rcar_canfd_channel_remove(gpriv, ch); + } + + /* Enter global sleep mode */ + rcar_canfd_set_bit(gpriv->base, RCANFD_GCTR, RCANFD_GCTR_GSLPR); + clk_disable_unprepare(gpriv->clkp); + return 0; +} + +static int __maybe_unused rcar_canfd_suspend(struct device *dev) +{ + return 0; +} + +static int __maybe_unused rcar_canfd_resume(struct device *dev) +{ + return 0; +} + +static SIMPLE_DEV_PM_OPS(rcar_canfd_pm_ops, rcar_canfd_suspend, + rcar_canfd_resume); + +static const struct of_device_id rcar_canfd_of_table[] = { + { .compatible = "renesas,rcar-gen3-canfd" }, + { } +}; + +MODULE_DEVICE_TABLE(of, rcar_canfd_of_table); + +static struct platform_driver rcar_canfd_driver = { + .driver = { + .name = RCANFD_DRV_NAME, + .of_match_table = of_match_ptr(rcar_canfd_of_table), + .pm = &rcar_canfd_pm_ops, + }, + .probe = rcar_canfd_probe, + .remove = rcar_canfd_remove, +}; + +module_platform_driver(rcar_canfd_driver); + +MODULE_AUTHOR("Ramesh Shanmugasundaram "); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("CAN FD driver for Renesas R-Car SoC"); +MODULE_ALIAS("platform:" RCANFD_DRV_NAME); -- cgit v0.10.2 From a23b97e6255b83babd28345e43a29508981f931e Mon Sep 17 00:00:00 2001 From: Ramesh Shanmugasundaram Date: Fri, 17 Jun 2016 09:20:56 +0100 Subject: can: rcar_can: Move Renesas CAN driver to rcar dir This patch clubs the Renesas controller drivers in one rcar dir. Signed-off-by: Ramesh Shanmugasundaram Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig index 13003a9..22570ea 100644 --- a/drivers/net/can/Kconfig +++ b/drivers/net/can/Kconfig @@ -104,16 +104,6 @@ config CAN_JANZ_ICAN3 This driver can also be built as a module. If so, the module will be called janz-ican3.ko. -config CAN_RCAR - tristate "Renesas R-Car CAN controller" - depends on ARCH_RENESAS || ARM - ---help--- - Say Y here if you want to use CAN controller found on Renesas R-Car - SoCs. - - To compile this driver as a module, choose M here: the module will - be called rcar_can. - config CAN_SUN4I tristate "Allwinner A10 CAN controller" depends on MACH_SUN4I || MACH_SUN7I || COMPILE_TEST diff --git a/drivers/net/can/Makefile b/drivers/net/can/Makefile index 226d5b5..26ba4b7 100644 --- a/drivers/net/can/Makefile +++ b/drivers/net/can/Makefile @@ -25,7 +25,6 @@ obj-$(CONFIG_CAN_IFI_CANFD) += ifi_canfd/ obj-$(CONFIG_CAN_JANZ_ICAN3) += janz-ican3.o obj-$(CONFIG_CAN_MSCAN) += mscan/ obj-$(CONFIG_CAN_M_CAN) += m_can/ -obj-$(CONFIG_CAN_RCAR) += rcar_can.o obj-$(CONFIG_CAN_SJA1000) += sja1000/ obj-$(CONFIG_CAN_SUN4I) += sun4i_can.o obj-$(CONFIG_CAN_TI_HECC) += ti_hecc.o diff --git a/drivers/net/can/rcar/Kconfig b/drivers/net/can/rcar/Kconfig index 6ea64d1..7b03a3a 100644 --- a/drivers/net/can/rcar/Kconfig +++ b/drivers/net/can/rcar/Kconfig @@ -1,3 +1,13 @@ +config CAN_RCAR + tristate "Renesas R-Car CAN controller" + depends on ARCH_RENESAS || ARM + ---help--- + Say Y here if you want to use CAN controller found on Renesas R-Car + SoCs. + + To compile this driver as a module, choose M here: the module will + be called rcar_can. + config CAN_RCAR_CANFD tristate "Renesas R-Car CAN FD controller" depends on ARCH_RENESAS || ARM diff --git a/drivers/net/can/rcar/Makefile b/drivers/net/can/rcar/Makefile index cbaf498e..08de36a 100644 --- a/drivers/net/can/rcar/Makefile +++ b/drivers/net/can/rcar/Makefile @@ -1,5 +1,6 @@ # -# Makefile for the Renesas R-Car CAN FD controller driver +# Makefile for the Renesas R-Car CAN & CAN FD controller drivers # +obj-$(CONFIG_CAN_RCAR) += rcar_can.o obj-$(CONFIG_CAN_RCAR_CANFD) += rcar_canfd.o diff --git a/drivers/net/can/rcar/rcar_can.c b/drivers/net/can/rcar/rcar_can.c new file mode 100644 index 0000000..788459f --- /dev/null +++ b/drivers/net/can/rcar/rcar_can.c @@ -0,0 +1,929 @@ +/* Renesas R-Car CAN device driver + * + * Copyright (C) 2013 Cogent Embedded, Inc. + * Copyright (C) 2013 Renesas Solutions Corp. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define RCAR_CAN_DRV_NAME "rcar_can" + +/* Mailbox configuration: + * mailbox 60 - 63 - Rx FIFO mailboxes + * mailbox 56 - 59 - Tx FIFO mailboxes + * non-FIFO mailboxes are not used + */ +#define RCAR_CAN_N_MBX 64 /* Number of mailboxes in non-FIFO mode */ +#define RCAR_CAN_RX_FIFO_MBX 60 /* Mailbox - window to Rx FIFO */ +#define RCAR_CAN_TX_FIFO_MBX 56 /* Mailbox - window to Tx FIFO */ +#define RCAR_CAN_FIFO_DEPTH 4 + +/* Mailbox registers structure */ +struct rcar_can_mbox_regs { + u32 id; /* IDE and RTR bits, SID and EID */ + u8 stub; /* Not used */ + u8 dlc; /* Data Length Code - bits [0..3] */ + u8 data[8]; /* Data Bytes */ + u8 tsh; /* Time Stamp Higher Byte */ + u8 tsl; /* Time Stamp Lower Byte */ +}; + +struct rcar_can_regs { + struct rcar_can_mbox_regs mb[RCAR_CAN_N_MBX]; /* Mailbox registers */ + u32 mkr_2_9[8]; /* Mask Registers 2-9 */ + u32 fidcr[2]; /* FIFO Received ID Compare Register */ + u32 mkivlr1; /* Mask Invalid Register 1 */ + u32 mier1; /* Mailbox Interrupt Enable Register 1 */ + u32 mkr_0_1[2]; /* Mask Registers 0-1 */ + u32 mkivlr0; /* Mask Invalid Register 0*/ + u32 mier0; /* Mailbox Interrupt Enable Register 0 */ + u8 pad_440[0x3c0]; + u8 mctl[64]; /* Message Control Registers */ + u16 ctlr; /* Control Register */ + u16 str; /* Status register */ + u8 bcr[3]; /* Bit Configuration Register */ + u8 clkr; /* Clock Select Register */ + u8 rfcr; /* Receive FIFO Control Register */ + u8 rfpcr; /* Receive FIFO Pointer Control Register */ + u8 tfcr; /* Transmit FIFO Control Register */ + u8 tfpcr; /* Transmit FIFO Pointer Control Register */ + u8 eier; /* Error Interrupt Enable Register */ + u8 eifr; /* Error Interrupt Factor Judge Register */ + u8 recr; /* Receive Error Count Register */ + u8 tecr; /* Transmit Error Count Register */ + u8 ecsr; /* Error Code Store Register */ + u8 cssr; /* Channel Search Support Register */ + u8 mssr; /* Mailbox Search Status Register */ + u8 msmr; /* Mailbox Search Mode Register */ + u16 tsr; /* Time Stamp Register */ + u8 afsr; /* Acceptance Filter Support Register */ + u8 pad_857; + u8 tcr; /* Test Control Register */ + u8 pad_859[7]; + u8 ier; /* Interrupt Enable Register */ + u8 isr; /* Interrupt Status Register */ + u8 pad_862; + u8 mbsmr; /* Mailbox Search Mask Register */ +}; + +struct rcar_can_priv { + struct can_priv can; /* Must be the first member! */ + struct net_device *ndev; + struct napi_struct napi; + struct rcar_can_regs __iomem *regs; + struct clk *clk; + struct clk *can_clk; + u8 tx_dlc[RCAR_CAN_FIFO_DEPTH]; + u32 tx_head; + u32 tx_tail; + u8 clock_select; + u8 ier; +}; + +static const struct can_bittiming_const rcar_can_bittiming_const = { + .name = RCAR_CAN_DRV_NAME, + .tseg1_min = 4, + .tseg1_max = 16, + .tseg2_min = 2, + .tseg2_max = 8, + .sjw_max = 4, + .brp_min = 1, + .brp_max = 1024, + .brp_inc = 1, +}; + +/* Control Register bits */ +#define RCAR_CAN_CTLR_BOM (3 << 11) /* Bus-Off Recovery Mode Bits */ +#define RCAR_CAN_CTLR_BOM_ENT (1 << 11) /* Entry to halt mode */ + /* at bus-off entry */ +#define RCAR_CAN_CTLR_SLPM (1 << 10) +#define RCAR_CAN_CTLR_CANM (3 << 8) /* Operating Mode Select Bit */ +#define RCAR_CAN_CTLR_CANM_HALT (1 << 9) +#define RCAR_CAN_CTLR_CANM_RESET (1 << 8) +#define RCAR_CAN_CTLR_CANM_FORCE_RESET (3 << 8) +#define RCAR_CAN_CTLR_MLM (1 << 3) /* Message Lost Mode Select */ +#define RCAR_CAN_CTLR_IDFM (3 << 1) /* ID Format Mode Select Bits */ +#define RCAR_CAN_CTLR_IDFM_MIXED (1 << 2) /* Mixed ID mode */ +#define RCAR_CAN_CTLR_MBM (1 << 0) /* Mailbox Mode select */ + +/* Status Register bits */ +#define RCAR_CAN_STR_RSTST (1 << 8) /* Reset Status Bit */ + +/* FIFO Received ID Compare Registers 0 and 1 bits */ +#define RCAR_CAN_FIDCR_IDE (1 << 31) /* ID Extension Bit */ +#define RCAR_CAN_FIDCR_RTR (1 << 30) /* Remote Transmission Request Bit */ + +/* Receive FIFO Control Register bits */ +#define RCAR_CAN_RFCR_RFEST (1 << 7) /* Receive FIFO Empty Status Flag */ +#define RCAR_CAN_RFCR_RFE (1 << 0) /* Receive FIFO Enable */ + +/* Transmit FIFO Control Register bits */ +#define RCAR_CAN_TFCR_TFUST (7 << 1) /* Transmit FIFO Unsent Message */ + /* Number Status Bits */ +#define RCAR_CAN_TFCR_TFUST_SHIFT 1 /* Offset of Transmit FIFO Unsent */ + /* Message Number Status Bits */ +#define RCAR_CAN_TFCR_TFE (1 << 0) /* Transmit FIFO Enable */ + +#define RCAR_CAN_N_RX_MKREGS1 2 /* Number of mask registers */ + /* for Rx mailboxes 0-31 */ +#define RCAR_CAN_N_RX_MKREGS2 8 + +/* Bit Configuration Register settings */ +#define RCAR_CAN_BCR_TSEG1(x) (((x) & 0x0f) << 20) +#define RCAR_CAN_BCR_BPR(x) (((x) & 0x3ff) << 8) +#define RCAR_CAN_BCR_SJW(x) (((x) & 0x3) << 4) +#define RCAR_CAN_BCR_TSEG2(x) ((x) & 0x07) + +/* Mailbox and Mask Registers bits */ +#define RCAR_CAN_IDE (1 << 31) +#define RCAR_CAN_RTR (1 << 30) +#define RCAR_CAN_SID_SHIFT 18 + +/* Mailbox Interrupt Enable Register 1 bits */ +#define RCAR_CAN_MIER1_RXFIE (1 << 28) /* Receive FIFO Interrupt Enable */ +#define RCAR_CAN_MIER1_TXFIE (1 << 24) /* Transmit FIFO Interrupt Enable */ + +/* Interrupt Enable Register bits */ +#define RCAR_CAN_IER_ERSIE (1 << 5) /* Error (ERS) Interrupt Enable Bit */ +#define RCAR_CAN_IER_RXFIE (1 << 4) /* Reception FIFO Interrupt */ + /* Enable Bit */ +#define RCAR_CAN_IER_TXFIE (1 << 3) /* Transmission FIFO Interrupt */ + /* Enable Bit */ +/* Interrupt Status Register bits */ +#define RCAR_CAN_ISR_ERSF (1 << 5) /* Error (ERS) Interrupt Status Bit */ +#define RCAR_CAN_ISR_RXFF (1 << 4) /* Reception FIFO Interrupt */ + /* Status Bit */ +#define RCAR_CAN_ISR_TXFF (1 << 3) /* Transmission FIFO Interrupt */ + /* Status Bit */ + +/* Error Interrupt Enable Register bits */ +#define RCAR_CAN_EIER_BLIE (1 << 7) /* Bus Lock Interrupt Enable */ +#define RCAR_CAN_EIER_OLIE (1 << 6) /* Overload Frame Transmit */ + /* Interrupt Enable */ +#define RCAR_CAN_EIER_ORIE (1 << 5) /* Receive Overrun Interrupt Enable */ +#define RCAR_CAN_EIER_BORIE (1 << 4) /* Bus-Off Recovery Interrupt Enable */ +#define RCAR_CAN_EIER_BOEIE (1 << 3) /* Bus-Off Entry Interrupt Enable */ +#define RCAR_CAN_EIER_EPIE (1 << 2) /* Error Passive Interrupt Enable */ +#define RCAR_CAN_EIER_EWIE (1 << 1) /* Error Warning Interrupt Enable */ +#define RCAR_CAN_EIER_BEIE (1 << 0) /* Bus Error Interrupt Enable */ + +/* Error Interrupt Factor Judge Register bits */ +#define RCAR_CAN_EIFR_BLIF (1 << 7) /* Bus Lock Detect Flag */ +#define RCAR_CAN_EIFR_OLIF (1 << 6) /* Overload Frame Transmission */ + /* Detect Flag */ +#define RCAR_CAN_EIFR_ORIF (1 << 5) /* Receive Overrun Detect Flag */ +#define RCAR_CAN_EIFR_BORIF (1 << 4) /* Bus-Off Recovery Detect Flag */ +#define RCAR_CAN_EIFR_BOEIF (1 << 3) /* Bus-Off Entry Detect Flag */ +#define RCAR_CAN_EIFR_EPIF (1 << 2) /* Error Passive Detect Flag */ +#define RCAR_CAN_EIFR_EWIF (1 << 1) /* Error Warning Detect Flag */ +#define RCAR_CAN_EIFR_BEIF (1 << 0) /* Bus Error Detect Flag */ + +/* Error Code Store Register bits */ +#define RCAR_CAN_ECSR_EDPM (1 << 7) /* Error Display Mode Select Bit */ +#define RCAR_CAN_ECSR_ADEF (1 << 6) /* ACK Delimiter Error Flag */ +#define RCAR_CAN_ECSR_BE0F (1 << 5) /* Bit Error (dominant) Flag */ +#define RCAR_CAN_ECSR_BE1F (1 << 4) /* Bit Error (recessive) Flag */ +#define RCAR_CAN_ECSR_CEF (1 << 3) /* CRC Error Flag */ +#define RCAR_CAN_ECSR_AEF (1 << 2) /* ACK Error Flag */ +#define RCAR_CAN_ECSR_FEF (1 << 1) /* Form Error Flag */ +#define RCAR_CAN_ECSR_SEF (1 << 0) /* Stuff Error Flag */ + +#define RCAR_CAN_NAPI_WEIGHT 4 +#define MAX_STR_READS 0x100 + +static void tx_failure_cleanup(struct net_device *ndev) +{ + int i; + + for (i = 0; i < RCAR_CAN_FIFO_DEPTH; i++) + can_free_echo_skb(ndev, i); +} + +static void rcar_can_error(struct net_device *ndev) +{ + struct rcar_can_priv *priv = netdev_priv(ndev); + struct net_device_stats *stats = &ndev->stats; + struct can_frame *cf; + struct sk_buff *skb; + u8 eifr, txerr = 0, rxerr = 0; + + /* Propagate the error condition to the CAN stack */ + skb = alloc_can_err_skb(ndev, &cf); + + eifr = readb(&priv->regs->eifr); + if (eifr & (RCAR_CAN_EIFR_EWIF | RCAR_CAN_EIFR_EPIF)) { + txerr = readb(&priv->regs->tecr); + rxerr = readb(&priv->regs->recr); + if (skb) { + cf->can_id |= CAN_ERR_CRTL; + cf->data[6] = txerr; + cf->data[7] = rxerr; + } + } + if (eifr & RCAR_CAN_EIFR_BEIF) { + int rx_errors = 0, tx_errors = 0; + u8 ecsr; + + netdev_dbg(priv->ndev, "Bus error interrupt:\n"); + if (skb) + cf->can_id |= CAN_ERR_BUSERROR | CAN_ERR_PROT; + + ecsr = readb(&priv->regs->ecsr); + if (ecsr & RCAR_CAN_ECSR_ADEF) { + netdev_dbg(priv->ndev, "ACK Delimiter Error\n"); + tx_errors++; + writeb(~RCAR_CAN_ECSR_ADEF, &priv->regs->ecsr); + if (skb) + cf->data[3] = CAN_ERR_PROT_LOC_ACK_DEL; + } + if (ecsr & RCAR_CAN_ECSR_BE0F) { + netdev_dbg(priv->ndev, "Bit Error (dominant)\n"); + tx_errors++; + writeb(~RCAR_CAN_ECSR_BE0F, &priv->regs->ecsr); + if (skb) + cf->data[2] |= CAN_ERR_PROT_BIT0; + } + if (ecsr & RCAR_CAN_ECSR_BE1F) { + netdev_dbg(priv->ndev, "Bit Error (recessive)\n"); + tx_errors++; + writeb(~RCAR_CAN_ECSR_BE1F, &priv->regs->ecsr); + if (skb) + cf->data[2] |= CAN_ERR_PROT_BIT1; + } + if (ecsr & RCAR_CAN_ECSR_CEF) { + netdev_dbg(priv->ndev, "CRC Error\n"); + rx_errors++; + writeb(~RCAR_CAN_ECSR_CEF, &priv->regs->ecsr); + if (skb) + cf->data[3] = CAN_ERR_PROT_LOC_CRC_SEQ; + } + if (ecsr & RCAR_CAN_ECSR_AEF) { + netdev_dbg(priv->ndev, "ACK Error\n"); + tx_errors++; + writeb(~RCAR_CAN_ECSR_AEF, &priv->regs->ecsr); + if (skb) { + cf->can_id |= CAN_ERR_ACK; + cf->data[3] = CAN_ERR_PROT_LOC_ACK; + } + } + if (ecsr & RCAR_CAN_ECSR_FEF) { + netdev_dbg(priv->ndev, "Form Error\n"); + rx_errors++; + writeb(~RCAR_CAN_ECSR_FEF, &priv->regs->ecsr); + if (skb) + cf->data[2] |= CAN_ERR_PROT_FORM; + } + if (ecsr & RCAR_CAN_ECSR_SEF) { + netdev_dbg(priv->ndev, "Stuff Error\n"); + rx_errors++; + writeb(~RCAR_CAN_ECSR_SEF, &priv->regs->ecsr); + if (skb) + cf->data[2] |= CAN_ERR_PROT_STUFF; + } + + priv->can.can_stats.bus_error++; + ndev->stats.rx_errors += rx_errors; + ndev->stats.tx_errors += tx_errors; + writeb(~RCAR_CAN_EIFR_BEIF, &priv->regs->eifr); + } + if (eifr & RCAR_CAN_EIFR_EWIF) { + netdev_dbg(priv->ndev, "Error warning interrupt\n"); + priv->can.state = CAN_STATE_ERROR_WARNING; + priv->can.can_stats.error_warning++; + /* Clear interrupt condition */ + writeb(~RCAR_CAN_EIFR_EWIF, &priv->regs->eifr); + if (skb) + cf->data[1] = txerr > rxerr ? CAN_ERR_CRTL_TX_WARNING : + CAN_ERR_CRTL_RX_WARNING; + } + if (eifr & RCAR_CAN_EIFR_EPIF) { + netdev_dbg(priv->ndev, "Error passive interrupt\n"); + priv->can.state = CAN_STATE_ERROR_PASSIVE; + priv->can.can_stats.error_passive++; + /* Clear interrupt condition */ + writeb(~RCAR_CAN_EIFR_EPIF, &priv->regs->eifr); + if (skb) + cf->data[1] = txerr > rxerr ? CAN_ERR_CRTL_TX_PASSIVE : + CAN_ERR_CRTL_RX_PASSIVE; + } + if (eifr & RCAR_CAN_EIFR_BOEIF) { + netdev_dbg(priv->ndev, "Bus-off entry interrupt\n"); + tx_failure_cleanup(ndev); + priv->ier = RCAR_CAN_IER_ERSIE; + writeb(priv->ier, &priv->regs->ier); + priv->can.state = CAN_STATE_BUS_OFF; + /* Clear interrupt condition */ + writeb(~RCAR_CAN_EIFR_BOEIF, &priv->regs->eifr); + priv->can.can_stats.bus_off++; + can_bus_off(ndev); + if (skb) + cf->can_id |= CAN_ERR_BUSOFF; + } + if (eifr & RCAR_CAN_EIFR_ORIF) { + netdev_dbg(priv->ndev, "Receive overrun error interrupt\n"); + ndev->stats.rx_over_errors++; + ndev->stats.rx_errors++; + writeb(~RCAR_CAN_EIFR_ORIF, &priv->regs->eifr); + if (skb) { + cf->can_id |= CAN_ERR_CRTL; + cf->data[1] = CAN_ERR_CRTL_RX_OVERFLOW; + } + } + if (eifr & RCAR_CAN_EIFR_OLIF) { + netdev_dbg(priv->ndev, + "Overload Frame Transmission error interrupt\n"); + ndev->stats.rx_over_errors++; + ndev->stats.rx_errors++; + writeb(~RCAR_CAN_EIFR_OLIF, &priv->regs->eifr); + if (skb) { + cf->can_id |= CAN_ERR_PROT; + cf->data[2] |= CAN_ERR_PROT_OVERLOAD; + } + } + + if (skb) { + stats->rx_packets++; + stats->rx_bytes += cf->can_dlc; + netif_rx(skb); + } +} + +static void rcar_can_tx_done(struct net_device *ndev) +{ + struct rcar_can_priv *priv = netdev_priv(ndev); + struct net_device_stats *stats = &ndev->stats; + u8 isr; + + while (1) { + u8 unsent = readb(&priv->regs->tfcr); + + unsent = (unsent & RCAR_CAN_TFCR_TFUST) >> + RCAR_CAN_TFCR_TFUST_SHIFT; + if (priv->tx_head - priv->tx_tail <= unsent) + break; + stats->tx_packets++; + stats->tx_bytes += priv->tx_dlc[priv->tx_tail % + RCAR_CAN_FIFO_DEPTH]; + priv->tx_dlc[priv->tx_tail % RCAR_CAN_FIFO_DEPTH] = 0; + can_get_echo_skb(ndev, priv->tx_tail % RCAR_CAN_FIFO_DEPTH); + priv->tx_tail++; + netif_wake_queue(ndev); + } + /* Clear interrupt */ + isr = readb(&priv->regs->isr); + writeb(isr & ~RCAR_CAN_ISR_TXFF, &priv->regs->isr); + can_led_event(ndev, CAN_LED_EVENT_TX); +} + +static irqreturn_t rcar_can_interrupt(int irq, void *dev_id) +{ + struct net_device *ndev = dev_id; + struct rcar_can_priv *priv = netdev_priv(ndev); + u8 isr; + + isr = readb(&priv->regs->isr); + if (!(isr & priv->ier)) + return IRQ_NONE; + + if (isr & RCAR_CAN_ISR_ERSF) + rcar_can_error(ndev); + + if (isr & RCAR_CAN_ISR_TXFF) + rcar_can_tx_done(ndev); + + if (isr & RCAR_CAN_ISR_RXFF) { + if (napi_schedule_prep(&priv->napi)) { + /* Disable Rx FIFO interrupts */ + priv->ier &= ~RCAR_CAN_IER_RXFIE; + writeb(priv->ier, &priv->regs->ier); + __napi_schedule(&priv->napi); + } + } + + return IRQ_HANDLED; +} + +static void rcar_can_set_bittiming(struct net_device *dev) +{ + struct rcar_can_priv *priv = netdev_priv(dev); + struct can_bittiming *bt = &priv->can.bittiming; + u32 bcr; + + bcr = RCAR_CAN_BCR_TSEG1(bt->phase_seg1 + bt->prop_seg - 1) | + RCAR_CAN_BCR_BPR(bt->brp - 1) | RCAR_CAN_BCR_SJW(bt->sjw - 1) | + RCAR_CAN_BCR_TSEG2(bt->phase_seg2 - 1); + /* Don't overwrite CLKR with 32-bit BCR access; CLKR has 8-bit access. + * All the registers are big-endian but they get byte-swapped on 32-bit + * read/write (but not on 8-bit, contrary to the manuals)... + */ + writel((bcr << 8) | priv->clock_select, &priv->regs->bcr); +} + +static void rcar_can_start(struct net_device *ndev) +{ + struct rcar_can_priv *priv = netdev_priv(ndev); + u16 ctlr; + int i; + + /* Set controller to known mode: + * - FIFO mailbox mode + * - accept all messages + * - overrun mode + * CAN is in sleep mode after MCU hardware or software reset. + */ + ctlr = readw(&priv->regs->ctlr); + ctlr &= ~RCAR_CAN_CTLR_SLPM; + writew(ctlr, &priv->regs->ctlr); + /* Go to reset mode */ + ctlr |= RCAR_CAN_CTLR_CANM_FORCE_RESET; + writew(ctlr, &priv->regs->ctlr); + for (i = 0; i < MAX_STR_READS; i++) { + if (readw(&priv->regs->str) & RCAR_CAN_STR_RSTST) + break; + } + rcar_can_set_bittiming(ndev); + ctlr |= RCAR_CAN_CTLR_IDFM_MIXED; /* Select mixed ID mode */ + ctlr |= RCAR_CAN_CTLR_BOM_ENT; /* Entry to halt mode automatically */ + /* at bus-off */ + ctlr |= RCAR_CAN_CTLR_MBM; /* Select FIFO mailbox mode */ + ctlr |= RCAR_CAN_CTLR_MLM; /* Overrun mode */ + writew(ctlr, &priv->regs->ctlr); + + /* Accept all SID and EID */ + writel(0, &priv->regs->mkr_2_9[6]); + writel(0, &priv->regs->mkr_2_9[7]); + /* In FIFO mailbox mode, write "0" to bits 24 to 31 */ + writel(0, &priv->regs->mkivlr1); + /* Accept all frames */ + writel(0, &priv->regs->fidcr[0]); + writel(RCAR_CAN_FIDCR_IDE | RCAR_CAN_FIDCR_RTR, &priv->regs->fidcr[1]); + /* Enable and configure FIFO mailbox interrupts */ + writel(RCAR_CAN_MIER1_RXFIE | RCAR_CAN_MIER1_TXFIE, &priv->regs->mier1); + + priv->ier = RCAR_CAN_IER_ERSIE | RCAR_CAN_IER_RXFIE | + RCAR_CAN_IER_TXFIE; + writeb(priv->ier, &priv->regs->ier); + + /* Accumulate error codes */ + writeb(RCAR_CAN_ECSR_EDPM, &priv->regs->ecsr); + /* Enable error interrupts */ + writeb(RCAR_CAN_EIER_EWIE | RCAR_CAN_EIER_EPIE | RCAR_CAN_EIER_BOEIE | + (priv->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING ? + RCAR_CAN_EIER_BEIE : 0) | RCAR_CAN_EIER_ORIE | + RCAR_CAN_EIER_OLIE, &priv->regs->eier); + priv->can.state = CAN_STATE_ERROR_ACTIVE; + + /* Go to operation mode */ + writew(ctlr & ~RCAR_CAN_CTLR_CANM, &priv->regs->ctlr); + for (i = 0; i < MAX_STR_READS; i++) { + if (!(readw(&priv->regs->str) & RCAR_CAN_STR_RSTST)) + break; + } + /* Enable Rx and Tx FIFO */ + writeb(RCAR_CAN_RFCR_RFE, &priv->regs->rfcr); + writeb(RCAR_CAN_TFCR_TFE, &priv->regs->tfcr); +} + +static int rcar_can_open(struct net_device *ndev) +{ + struct rcar_can_priv *priv = netdev_priv(ndev); + int err; + + err = clk_prepare_enable(priv->clk); + if (err) { + netdev_err(ndev, + "failed to enable peripheral clock, error %d\n", + err); + goto out; + } + err = clk_prepare_enable(priv->can_clk); + if (err) { + netdev_err(ndev, "failed to enable CAN clock, error %d\n", + err); + goto out_clock; + } + err = open_candev(ndev); + if (err) { + netdev_err(ndev, "open_candev() failed, error %d\n", err); + goto out_can_clock; + } + napi_enable(&priv->napi); + err = request_irq(ndev->irq, rcar_can_interrupt, 0, ndev->name, ndev); + if (err) { + netdev_err(ndev, "request_irq(%d) failed, error %d\n", + ndev->irq, err); + goto out_close; + } + can_led_event(ndev, CAN_LED_EVENT_OPEN); + rcar_can_start(ndev); + netif_start_queue(ndev); + return 0; +out_close: + napi_disable(&priv->napi); + close_candev(ndev); +out_can_clock: + clk_disable_unprepare(priv->can_clk); +out_clock: + clk_disable_unprepare(priv->clk); +out: + return err; +} + +static void rcar_can_stop(struct net_device *ndev) +{ + struct rcar_can_priv *priv = netdev_priv(ndev); + u16 ctlr; + int i; + + /* Go to (force) reset mode */ + ctlr = readw(&priv->regs->ctlr); + ctlr |= RCAR_CAN_CTLR_CANM_FORCE_RESET; + writew(ctlr, &priv->regs->ctlr); + for (i = 0; i < MAX_STR_READS; i++) { + if (readw(&priv->regs->str) & RCAR_CAN_STR_RSTST) + break; + } + writel(0, &priv->regs->mier0); + writel(0, &priv->regs->mier1); + writeb(0, &priv->regs->ier); + writeb(0, &priv->regs->eier); + /* Go to sleep mode */ + ctlr |= RCAR_CAN_CTLR_SLPM; + writew(ctlr, &priv->regs->ctlr); + priv->can.state = CAN_STATE_STOPPED; +} + +static int rcar_can_close(struct net_device *ndev) +{ + struct rcar_can_priv *priv = netdev_priv(ndev); + + netif_stop_queue(ndev); + rcar_can_stop(ndev); + free_irq(ndev->irq, ndev); + napi_disable(&priv->napi); + clk_disable_unprepare(priv->can_clk); + clk_disable_unprepare(priv->clk); + close_candev(ndev); + can_led_event(ndev, CAN_LED_EVENT_STOP); + return 0; +} + +static netdev_tx_t rcar_can_start_xmit(struct sk_buff *skb, + struct net_device *ndev) +{ + struct rcar_can_priv *priv = netdev_priv(ndev); + struct can_frame *cf = (struct can_frame *)skb->data; + u32 data, i; + + if (can_dropped_invalid_skb(ndev, skb)) + return NETDEV_TX_OK; + + if (cf->can_id & CAN_EFF_FLAG) /* Extended frame format */ + data = (cf->can_id & CAN_EFF_MASK) | RCAR_CAN_IDE; + else /* Standard frame format */ + data = (cf->can_id & CAN_SFF_MASK) << RCAR_CAN_SID_SHIFT; + + if (cf->can_id & CAN_RTR_FLAG) { /* Remote transmission request */ + data |= RCAR_CAN_RTR; + } else { + for (i = 0; i < cf->can_dlc; i++) + writeb(cf->data[i], + &priv->regs->mb[RCAR_CAN_TX_FIFO_MBX].data[i]); + } + + writel(data, &priv->regs->mb[RCAR_CAN_TX_FIFO_MBX].id); + + writeb(cf->can_dlc, &priv->regs->mb[RCAR_CAN_TX_FIFO_MBX].dlc); + + priv->tx_dlc[priv->tx_head % RCAR_CAN_FIFO_DEPTH] = cf->can_dlc; + can_put_echo_skb(skb, ndev, priv->tx_head % RCAR_CAN_FIFO_DEPTH); + priv->tx_head++; + /* Start Tx: write 0xff to the TFPCR register to increment + * the CPU-side pointer for the transmit FIFO to the next + * mailbox location + */ + writeb(0xff, &priv->regs->tfpcr); + /* Stop the queue if we've filled all FIFO entries */ + if (priv->tx_head - priv->tx_tail >= RCAR_CAN_FIFO_DEPTH) + netif_stop_queue(ndev); + + return NETDEV_TX_OK; +} + +static const struct net_device_ops rcar_can_netdev_ops = { + .ndo_open = rcar_can_open, + .ndo_stop = rcar_can_close, + .ndo_start_xmit = rcar_can_start_xmit, + .ndo_change_mtu = can_change_mtu, +}; + +static void rcar_can_rx_pkt(struct rcar_can_priv *priv) +{ + struct net_device_stats *stats = &priv->ndev->stats; + struct can_frame *cf; + struct sk_buff *skb; + u32 data; + u8 dlc; + + skb = alloc_can_skb(priv->ndev, &cf); + if (!skb) { + stats->rx_dropped++; + return; + } + + data = readl(&priv->regs->mb[RCAR_CAN_RX_FIFO_MBX].id); + if (data & RCAR_CAN_IDE) + cf->can_id = (data & CAN_EFF_MASK) | CAN_EFF_FLAG; + else + cf->can_id = (data >> RCAR_CAN_SID_SHIFT) & CAN_SFF_MASK; + + dlc = readb(&priv->regs->mb[RCAR_CAN_RX_FIFO_MBX].dlc); + cf->can_dlc = get_can_dlc(dlc); + if (data & RCAR_CAN_RTR) { + cf->can_id |= CAN_RTR_FLAG; + } else { + for (dlc = 0; dlc < cf->can_dlc; dlc++) + cf->data[dlc] = + readb(&priv->regs->mb[RCAR_CAN_RX_FIFO_MBX].data[dlc]); + } + + can_led_event(priv->ndev, CAN_LED_EVENT_RX); + + stats->rx_bytes += cf->can_dlc; + stats->rx_packets++; + netif_receive_skb(skb); +} + +static int rcar_can_rx_poll(struct napi_struct *napi, int quota) +{ + struct rcar_can_priv *priv = container_of(napi, + struct rcar_can_priv, napi); + int num_pkts; + + for (num_pkts = 0; num_pkts < quota; num_pkts++) { + u8 rfcr, isr; + + isr = readb(&priv->regs->isr); + /* Clear interrupt bit */ + if (isr & RCAR_CAN_ISR_RXFF) + writeb(isr & ~RCAR_CAN_ISR_RXFF, &priv->regs->isr); + rfcr = readb(&priv->regs->rfcr); + if (rfcr & RCAR_CAN_RFCR_RFEST) + break; + rcar_can_rx_pkt(priv); + /* Write 0xff to the RFPCR register to increment + * the CPU-side pointer for the receive FIFO + * to the next mailbox location + */ + writeb(0xff, &priv->regs->rfpcr); + } + /* All packets processed */ + if (num_pkts < quota) { + napi_complete(napi); + priv->ier |= RCAR_CAN_IER_RXFIE; + writeb(priv->ier, &priv->regs->ier); + } + return num_pkts; +} + +static int rcar_can_do_set_mode(struct net_device *ndev, enum can_mode mode) +{ + switch (mode) { + case CAN_MODE_START: + rcar_can_start(ndev); + netif_wake_queue(ndev); + return 0; + default: + return -EOPNOTSUPP; + } +} + +static int rcar_can_get_berr_counter(const struct net_device *dev, + struct can_berr_counter *bec) +{ + struct rcar_can_priv *priv = netdev_priv(dev); + int err; + + err = clk_prepare_enable(priv->clk); + if (err) + return err; + bec->txerr = readb(&priv->regs->tecr); + bec->rxerr = readb(&priv->regs->recr); + clk_disable_unprepare(priv->clk); + return 0; +} + +static const char * const clock_names[] = { + [CLKR_CLKP1] = "clkp1", + [CLKR_CLKP2] = "clkp2", + [CLKR_CLKEXT] = "can_clk", +}; + +static int rcar_can_probe(struct platform_device *pdev) +{ + struct rcar_can_platform_data *pdata; + struct rcar_can_priv *priv; + struct net_device *ndev; + struct resource *mem; + void __iomem *addr; + u32 clock_select = CLKR_CLKP1; + int err = -ENODEV; + int irq; + + if (pdev->dev.of_node) { + of_property_read_u32(pdev->dev.of_node, + "renesas,can-clock-select", &clock_select); + } else { + pdata = dev_get_platdata(&pdev->dev); + if (!pdata) { + dev_err(&pdev->dev, "No platform data provided!\n"); + goto fail; + } + clock_select = pdata->clock_select; + } + + irq = platform_get_irq(pdev, 0); + if (irq < 0) { + dev_err(&pdev->dev, "No IRQ resource\n"); + err = irq; + goto fail; + } + + mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); + addr = devm_ioremap_resource(&pdev->dev, mem); + if (IS_ERR(addr)) { + err = PTR_ERR(addr); + goto fail; + } + + ndev = alloc_candev(sizeof(struct rcar_can_priv), RCAR_CAN_FIFO_DEPTH); + if (!ndev) { + dev_err(&pdev->dev, "alloc_candev() failed\n"); + err = -ENOMEM; + goto fail; + } + + priv = netdev_priv(ndev); + + priv->clk = devm_clk_get(&pdev->dev, "clkp1"); + if (IS_ERR(priv->clk)) { + err = PTR_ERR(priv->clk); + dev_err(&pdev->dev, "cannot get peripheral clock, error %d\n", + err); + goto fail_clk; + } + + if (clock_select >= ARRAY_SIZE(clock_names)) { + err = -EINVAL; + dev_err(&pdev->dev, "invalid CAN clock selected\n"); + goto fail_clk; + } + priv->can_clk = devm_clk_get(&pdev->dev, clock_names[clock_select]); + if (IS_ERR(priv->can_clk)) { + err = PTR_ERR(priv->can_clk); + dev_err(&pdev->dev, "cannot get CAN clock, error %d\n", err); + goto fail_clk; + } + + ndev->netdev_ops = &rcar_can_netdev_ops; + ndev->irq = irq; + ndev->flags |= IFF_ECHO; + priv->ndev = ndev; + priv->regs = addr; + priv->clock_select = clock_select; + priv->can.clock.freq = clk_get_rate(priv->can_clk); + priv->can.bittiming_const = &rcar_can_bittiming_const; + priv->can.do_set_mode = rcar_can_do_set_mode; + priv->can.do_get_berr_counter = rcar_can_get_berr_counter; + priv->can.ctrlmode_supported = CAN_CTRLMODE_BERR_REPORTING; + platform_set_drvdata(pdev, ndev); + SET_NETDEV_DEV(ndev, &pdev->dev); + + netif_napi_add(ndev, &priv->napi, rcar_can_rx_poll, + RCAR_CAN_NAPI_WEIGHT); + err = register_candev(ndev); + if (err) { + dev_err(&pdev->dev, "register_candev() failed, error %d\n", + err); + goto fail_candev; + } + + devm_can_led_init(ndev); + + dev_info(&pdev->dev, "device registered (regs @ %p, IRQ%d)\n", + priv->regs, ndev->irq); + + return 0; +fail_candev: + netif_napi_del(&priv->napi); +fail_clk: + free_candev(ndev); +fail: + return err; +} + +static int rcar_can_remove(struct platform_device *pdev) +{ + struct net_device *ndev = platform_get_drvdata(pdev); + struct rcar_can_priv *priv = netdev_priv(ndev); + + unregister_candev(ndev); + netif_napi_del(&priv->napi); + free_candev(ndev); + return 0; +} + +static int __maybe_unused rcar_can_suspend(struct device *dev) +{ + struct net_device *ndev = dev_get_drvdata(dev); + struct rcar_can_priv *priv = netdev_priv(ndev); + u16 ctlr; + + if (netif_running(ndev)) { + netif_stop_queue(ndev); + netif_device_detach(ndev); + } + ctlr = readw(&priv->regs->ctlr); + ctlr |= RCAR_CAN_CTLR_CANM_HALT; + writew(ctlr, &priv->regs->ctlr); + ctlr |= RCAR_CAN_CTLR_SLPM; + writew(ctlr, &priv->regs->ctlr); + priv->can.state = CAN_STATE_SLEEPING; + + clk_disable(priv->clk); + return 0; +} + +static int __maybe_unused rcar_can_resume(struct device *dev) +{ + struct net_device *ndev = dev_get_drvdata(dev); + struct rcar_can_priv *priv = netdev_priv(ndev); + u16 ctlr; + int err; + + err = clk_enable(priv->clk); + if (err) { + netdev_err(ndev, "clk_enable() failed, error %d\n", err); + return err; + } + + ctlr = readw(&priv->regs->ctlr); + ctlr &= ~RCAR_CAN_CTLR_SLPM; + writew(ctlr, &priv->regs->ctlr); + ctlr &= ~RCAR_CAN_CTLR_CANM; + writew(ctlr, &priv->regs->ctlr); + priv->can.state = CAN_STATE_ERROR_ACTIVE; + + if (netif_running(ndev)) { + netif_device_attach(ndev); + netif_start_queue(ndev); + } + return 0; +} + +static SIMPLE_DEV_PM_OPS(rcar_can_pm_ops, rcar_can_suspend, rcar_can_resume); + +static const struct of_device_id rcar_can_of_table[] __maybe_unused = { + { .compatible = "renesas,can-r8a7778" }, + { .compatible = "renesas,can-r8a7779" }, + { .compatible = "renesas,can-r8a7790" }, + { .compatible = "renesas,can-r8a7791" }, + { .compatible = "renesas,rcar-gen1-can" }, + { .compatible = "renesas,rcar-gen2-can" }, + { .compatible = "renesas,rcar-gen3-can" }, + { } +}; +MODULE_DEVICE_TABLE(of, rcar_can_of_table); + +static struct platform_driver rcar_can_driver = { + .driver = { + .name = RCAR_CAN_DRV_NAME, + .of_match_table = of_match_ptr(rcar_can_of_table), + .pm = &rcar_can_pm_ops, + }, + .probe = rcar_can_probe, + .remove = rcar_can_remove, +}; + +module_platform_driver(rcar_can_driver); + +MODULE_AUTHOR("Cogent Embedded, Inc."); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("CAN driver for Renesas R-Car SoC"); +MODULE_ALIAS("platform:" RCAR_CAN_DRV_NAME); diff --git a/drivers/net/can/rcar_can.c b/drivers/net/can/rcar_can.c deleted file mode 100644 index 788459f..0000000 --- a/drivers/net/can/rcar_can.c +++ /dev/null @@ -1,929 +0,0 @@ -/* Renesas R-Car CAN device driver - * - * Copyright (C) 2013 Cogent Embedded, Inc. - * Copyright (C) 2013 Renesas Solutions Corp. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License as published by the - * Free Software Foundation; either version 2 of the License, or (at your - * option) any later version. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#define RCAR_CAN_DRV_NAME "rcar_can" - -/* Mailbox configuration: - * mailbox 60 - 63 - Rx FIFO mailboxes - * mailbox 56 - 59 - Tx FIFO mailboxes - * non-FIFO mailboxes are not used - */ -#define RCAR_CAN_N_MBX 64 /* Number of mailboxes in non-FIFO mode */ -#define RCAR_CAN_RX_FIFO_MBX 60 /* Mailbox - window to Rx FIFO */ -#define RCAR_CAN_TX_FIFO_MBX 56 /* Mailbox - window to Tx FIFO */ -#define RCAR_CAN_FIFO_DEPTH 4 - -/* Mailbox registers structure */ -struct rcar_can_mbox_regs { - u32 id; /* IDE and RTR bits, SID and EID */ - u8 stub; /* Not used */ - u8 dlc; /* Data Length Code - bits [0..3] */ - u8 data[8]; /* Data Bytes */ - u8 tsh; /* Time Stamp Higher Byte */ - u8 tsl; /* Time Stamp Lower Byte */ -}; - -struct rcar_can_regs { - struct rcar_can_mbox_regs mb[RCAR_CAN_N_MBX]; /* Mailbox registers */ - u32 mkr_2_9[8]; /* Mask Registers 2-9 */ - u32 fidcr[2]; /* FIFO Received ID Compare Register */ - u32 mkivlr1; /* Mask Invalid Register 1 */ - u32 mier1; /* Mailbox Interrupt Enable Register 1 */ - u32 mkr_0_1[2]; /* Mask Registers 0-1 */ - u32 mkivlr0; /* Mask Invalid Register 0*/ - u32 mier0; /* Mailbox Interrupt Enable Register 0 */ - u8 pad_440[0x3c0]; - u8 mctl[64]; /* Message Control Registers */ - u16 ctlr; /* Control Register */ - u16 str; /* Status register */ - u8 bcr[3]; /* Bit Configuration Register */ - u8 clkr; /* Clock Select Register */ - u8 rfcr; /* Receive FIFO Control Register */ - u8 rfpcr; /* Receive FIFO Pointer Control Register */ - u8 tfcr; /* Transmit FIFO Control Register */ - u8 tfpcr; /* Transmit FIFO Pointer Control Register */ - u8 eier; /* Error Interrupt Enable Register */ - u8 eifr; /* Error Interrupt Factor Judge Register */ - u8 recr; /* Receive Error Count Register */ - u8 tecr; /* Transmit Error Count Register */ - u8 ecsr; /* Error Code Store Register */ - u8 cssr; /* Channel Search Support Register */ - u8 mssr; /* Mailbox Search Status Register */ - u8 msmr; /* Mailbox Search Mode Register */ - u16 tsr; /* Time Stamp Register */ - u8 afsr; /* Acceptance Filter Support Register */ - u8 pad_857; - u8 tcr; /* Test Control Register */ - u8 pad_859[7]; - u8 ier; /* Interrupt Enable Register */ - u8 isr; /* Interrupt Status Register */ - u8 pad_862; - u8 mbsmr; /* Mailbox Search Mask Register */ -}; - -struct rcar_can_priv { - struct can_priv can; /* Must be the first member! */ - struct net_device *ndev; - struct napi_struct napi; - struct rcar_can_regs __iomem *regs; - struct clk *clk; - struct clk *can_clk; - u8 tx_dlc[RCAR_CAN_FIFO_DEPTH]; - u32 tx_head; - u32 tx_tail; - u8 clock_select; - u8 ier; -}; - -static const struct can_bittiming_const rcar_can_bittiming_const = { - .name = RCAR_CAN_DRV_NAME, - .tseg1_min = 4, - .tseg1_max = 16, - .tseg2_min = 2, - .tseg2_max = 8, - .sjw_max = 4, - .brp_min = 1, - .brp_max = 1024, - .brp_inc = 1, -}; - -/* Control Register bits */ -#define RCAR_CAN_CTLR_BOM (3 << 11) /* Bus-Off Recovery Mode Bits */ -#define RCAR_CAN_CTLR_BOM_ENT (1 << 11) /* Entry to halt mode */ - /* at bus-off entry */ -#define RCAR_CAN_CTLR_SLPM (1 << 10) -#define RCAR_CAN_CTLR_CANM (3 << 8) /* Operating Mode Select Bit */ -#define RCAR_CAN_CTLR_CANM_HALT (1 << 9) -#define RCAR_CAN_CTLR_CANM_RESET (1 << 8) -#define RCAR_CAN_CTLR_CANM_FORCE_RESET (3 << 8) -#define RCAR_CAN_CTLR_MLM (1 << 3) /* Message Lost Mode Select */ -#define RCAR_CAN_CTLR_IDFM (3 << 1) /* ID Format Mode Select Bits */ -#define RCAR_CAN_CTLR_IDFM_MIXED (1 << 2) /* Mixed ID mode */ -#define RCAR_CAN_CTLR_MBM (1 << 0) /* Mailbox Mode select */ - -/* Status Register bits */ -#define RCAR_CAN_STR_RSTST (1 << 8) /* Reset Status Bit */ - -/* FIFO Received ID Compare Registers 0 and 1 bits */ -#define RCAR_CAN_FIDCR_IDE (1 << 31) /* ID Extension Bit */ -#define RCAR_CAN_FIDCR_RTR (1 << 30) /* Remote Transmission Request Bit */ - -/* Receive FIFO Control Register bits */ -#define RCAR_CAN_RFCR_RFEST (1 << 7) /* Receive FIFO Empty Status Flag */ -#define RCAR_CAN_RFCR_RFE (1 << 0) /* Receive FIFO Enable */ - -/* Transmit FIFO Control Register bits */ -#define RCAR_CAN_TFCR_TFUST (7 << 1) /* Transmit FIFO Unsent Message */ - /* Number Status Bits */ -#define RCAR_CAN_TFCR_TFUST_SHIFT 1 /* Offset of Transmit FIFO Unsent */ - /* Message Number Status Bits */ -#define RCAR_CAN_TFCR_TFE (1 << 0) /* Transmit FIFO Enable */ - -#define RCAR_CAN_N_RX_MKREGS1 2 /* Number of mask registers */ - /* for Rx mailboxes 0-31 */ -#define RCAR_CAN_N_RX_MKREGS2 8 - -/* Bit Configuration Register settings */ -#define RCAR_CAN_BCR_TSEG1(x) (((x) & 0x0f) << 20) -#define RCAR_CAN_BCR_BPR(x) (((x) & 0x3ff) << 8) -#define RCAR_CAN_BCR_SJW(x) (((x) & 0x3) << 4) -#define RCAR_CAN_BCR_TSEG2(x) ((x) & 0x07) - -/* Mailbox and Mask Registers bits */ -#define RCAR_CAN_IDE (1 << 31) -#define RCAR_CAN_RTR (1 << 30) -#define RCAR_CAN_SID_SHIFT 18 - -/* Mailbox Interrupt Enable Register 1 bits */ -#define RCAR_CAN_MIER1_RXFIE (1 << 28) /* Receive FIFO Interrupt Enable */ -#define RCAR_CAN_MIER1_TXFIE (1 << 24) /* Transmit FIFO Interrupt Enable */ - -/* Interrupt Enable Register bits */ -#define RCAR_CAN_IER_ERSIE (1 << 5) /* Error (ERS) Interrupt Enable Bit */ -#define RCAR_CAN_IER_RXFIE (1 << 4) /* Reception FIFO Interrupt */ - /* Enable Bit */ -#define RCAR_CAN_IER_TXFIE (1 << 3) /* Transmission FIFO Interrupt */ - /* Enable Bit */ -/* Interrupt Status Register bits */ -#define RCAR_CAN_ISR_ERSF (1 << 5) /* Error (ERS) Interrupt Status Bit */ -#define RCAR_CAN_ISR_RXFF (1 << 4) /* Reception FIFO Interrupt */ - /* Status Bit */ -#define RCAR_CAN_ISR_TXFF (1 << 3) /* Transmission FIFO Interrupt */ - /* Status Bit */ - -/* Error Interrupt Enable Register bits */ -#define RCAR_CAN_EIER_BLIE (1 << 7) /* Bus Lock Interrupt Enable */ -#define RCAR_CAN_EIER_OLIE (1 << 6) /* Overload Frame Transmit */ - /* Interrupt Enable */ -#define RCAR_CAN_EIER_ORIE (1 << 5) /* Receive Overrun Interrupt Enable */ -#define RCAR_CAN_EIER_BORIE (1 << 4) /* Bus-Off Recovery Interrupt Enable */ -#define RCAR_CAN_EIER_BOEIE (1 << 3) /* Bus-Off Entry Interrupt Enable */ -#define RCAR_CAN_EIER_EPIE (1 << 2) /* Error Passive Interrupt Enable */ -#define RCAR_CAN_EIER_EWIE (1 << 1) /* Error Warning Interrupt Enable */ -#define RCAR_CAN_EIER_BEIE (1 << 0) /* Bus Error Interrupt Enable */ - -/* Error Interrupt Factor Judge Register bits */ -#define RCAR_CAN_EIFR_BLIF (1 << 7) /* Bus Lock Detect Flag */ -#define RCAR_CAN_EIFR_OLIF (1 << 6) /* Overload Frame Transmission */ - /* Detect Flag */ -#define RCAR_CAN_EIFR_ORIF (1 << 5) /* Receive Overrun Detect Flag */ -#define RCAR_CAN_EIFR_BORIF (1 << 4) /* Bus-Off Recovery Detect Flag */ -#define RCAR_CAN_EIFR_BOEIF (1 << 3) /* Bus-Off Entry Detect Flag */ -#define RCAR_CAN_EIFR_EPIF (1 << 2) /* Error Passive Detect Flag */ -#define RCAR_CAN_EIFR_EWIF (1 << 1) /* Error Warning Detect Flag */ -#define RCAR_CAN_EIFR_BEIF (1 << 0) /* Bus Error Detect Flag */ - -/* Error Code Store Register bits */ -#define RCAR_CAN_ECSR_EDPM (1 << 7) /* Error Display Mode Select Bit */ -#define RCAR_CAN_ECSR_ADEF (1 << 6) /* ACK Delimiter Error Flag */ -#define RCAR_CAN_ECSR_BE0F (1 << 5) /* Bit Error (dominant) Flag */ -#define RCAR_CAN_ECSR_BE1F (1 << 4) /* Bit Error (recessive) Flag */ -#define RCAR_CAN_ECSR_CEF (1 << 3) /* CRC Error Flag */ -#define RCAR_CAN_ECSR_AEF (1 << 2) /* ACK Error Flag */ -#define RCAR_CAN_ECSR_FEF (1 << 1) /* Form Error Flag */ -#define RCAR_CAN_ECSR_SEF (1 << 0) /* Stuff Error Flag */ - -#define RCAR_CAN_NAPI_WEIGHT 4 -#define MAX_STR_READS 0x100 - -static void tx_failure_cleanup(struct net_device *ndev) -{ - int i; - - for (i = 0; i < RCAR_CAN_FIFO_DEPTH; i++) - can_free_echo_skb(ndev, i); -} - -static void rcar_can_error(struct net_device *ndev) -{ - struct rcar_can_priv *priv = netdev_priv(ndev); - struct net_device_stats *stats = &ndev->stats; - struct can_frame *cf; - struct sk_buff *skb; - u8 eifr, txerr = 0, rxerr = 0; - - /* Propagate the error condition to the CAN stack */ - skb = alloc_can_err_skb(ndev, &cf); - - eifr = readb(&priv->regs->eifr); - if (eifr & (RCAR_CAN_EIFR_EWIF | RCAR_CAN_EIFR_EPIF)) { - txerr = readb(&priv->regs->tecr); - rxerr = readb(&priv->regs->recr); - if (skb) { - cf->can_id |= CAN_ERR_CRTL; - cf->data[6] = txerr; - cf->data[7] = rxerr; - } - } - if (eifr & RCAR_CAN_EIFR_BEIF) { - int rx_errors = 0, tx_errors = 0; - u8 ecsr; - - netdev_dbg(priv->ndev, "Bus error interrupt:\n"); - if (skb) - cf->can_id |= CAN_ERR_BUSERROR | CAN_ERR_PROT; - - ecsr = readb(&priv->regs->ecsr); - if (ecsr & RCAR_CAN_ECSR_ADEF) { - netdev_dbg(priv->ndev, "ACK Delimiter Error\n"); - tx_errors++; - writeb(~RCAR_CAN_ECSR_ADEF, &priv->regs->ecsr); - if (skb) - cf->data[3] = CAN_ERR_PROT_LOC_ACK_DEL; - } - if (ecsr & RCAR_CAN_ECSR_BE0F) { - netdev_dbg(priv->ndev, "Bit Error (dominant)\n"); - tx_errors++; - writeb(~RCAR_CAN_ECSR_BE0F, &priv->regs->ecsr); - if (skb) - cf->data[2] |= CAN_ERR_PROT_BIT0; - } - if (ecsr & RCAR_CAN_ECSR_BE1F) { - netdev_dbg(priv->ndev, "Bit Error (recessive)\n"); - tx_errors++; - writeb(~RCAR_CAN_ECSR_BE1F, &priv->regs->ecsr); - if (skb) - cf->data[2] |= CAN_ERR_PROT_BIT1; - } - if (ecsr & RCAR_CAN_ECSR_CEF) { - netdev_dbg(priv->ndev, "CRC Error\n"); - rx_errors++; - writeb(~RCAR_CAN_ECSR_CEF, &priv->regs->ecsr); - if (skb) - cf->data[3] = CAN_ERR_PROT_LOC_CRC_SEQ; - } - if (ecsr & RCAR_CAN_ECSR_AEF) { - netdev_dbg(priv->ndev, "ACK Error\n"); - tx_errors++; - writeb(~RCAR_CAN_ECSR_AEF, &priv->regs->ecsr); - if (skb) { - cf->can_id |= CAN_ERR_ACK; - cf->data[3] = CAN_ERR_PROT_LOC_ACK; - } - } - if (ecsr & RCAR_CAN_ECSR_FEF) { - netdev_dbg(priv->ndev, "Form Error\n"); - rx_errors++; - writeb(~RCAR_CAN_ECSR_FEF, &priv->regs->ecsr); - if (skb) - cf->data[2] |= CAN_ERR_PROT_FORM; - } - if (ecsr & RCAR_CAN_ECSR_SEF) { - netdev_dbg(priv->ndev, "Stuff Error\n"); - rx_errors++; - writeb(~RCAR_CAN_ECSR_SEF, &priv->regs->ecsr); - if (skb) - cf->data[2] |= CAN_ERR_PROT_STUFF; - } - - priv->can.can_stats.bus_error++; - ndev->stats.rx_errors += rx_errors; - ndev->stats.tx_errors += tx_errors; - writeb(~RCAR_CAN_EIFR_BEIF, &priv->regs->eifr); - } - if (eifr & RCAR_CAN_EIFR_EWIF) { - netdev_dbg(priv->ndev, "Error warning interrupt\n"); - priv->can.state = CAN_STATE_ERROR_WARNING; - priv->can.can_stats.error_warning++; - /* Clear interrupt condition */ - writeb(~RCAR_CAN_EIFR_EWIF, &priv->regs->eifr); - if (skb) - cf->data[1] = txerr > rxerr ? CAN_ERR_CRTL_TX_WARNING : - CAN_ERR_CRTL_RX_WARNING; - } - if (eifr & RCAR_CAN_EIFR_EPIF) { - netdev_dbg(priv->ndev, "Error passive interrupt\n"); - priv->can.state = CAN_STATE_ERROR_PASSIVE; - priv->can.can_stats.error_passive++; - /* Clear interrupt condition */ - writeb(~RCAR_CAN_EIFR_EPIF, &priv->regs->eifr); - if (skb) - cf->data[1] = txerr > rxerr ? CAN_ERR_CRTL_TX_PASSIVE : - CAN_ERR_CRTL_RX_PASSIVE; - } - if (eifr & RCAR_CAN_EIFR_BOEIF) { - netdev_dbg(priv->ndev, "Bus-off entry interrupt\n"); - tx_failure_cleanup(ndev); - priv->ier = RCAR_CAN_IER_ERSIE; - writeb(priv->ier, &priv->regs->ier); - priv->can.state = CAN_STATE_BUS_OFF; - /* Clear interrupt condition */ - writeb(~RCAR_CAN_EIFR_BOEIF, &priv->regs->eifr); - priv->can.can_stats.bus_off++; - can_bus_off(ndev); - if (skb) - cf->can_id |= CAN_ERR_BUSOFF; - } - if (eifr & RCAR_CAN_EIFR_ORIF) { - netdev_dbg(priv->ndev, "Receive overrun error interrupt\n"); - ndev->stats.rx_over_errors++; - ndev->stats.rx_errors++; - writeb(~RCAR_CAN_EIFR_ORIF, &priv->regs->eifr); - if (skb) { - cf->can_id |= CAN_ERR_CRTL; - cf->data[1] = CAN_ERR_CRTL_RX_OVERFLOW; - } - } - if (eifr & RCAR_CAN_EIFR_OLIF) { - netdev_dbg(priv->ndev, - "Overload Frame Transmission error interrupt\n"); - ndev->stats.rx_over_errors++; - ndev->stats.rx_errors++; - writeb(~RCAR_CAN_EIFR_OLIF, &priv->regs->eifr); - if (skb) { - cf->can_id |= CAN_ERR_PROT; - cf->data[2] |= CAN_ERR_PROT_OVERLOAD; - } - } - - if (skb) { - stats->rx_packets++; - stats->rx_bytes += cf->can_dlc; - netif_rx(skb); - } -} - -static void rcar_can_tx_done(struct net_device *ndev) -{ - struct rcar_can_priv *priv = netdev_priv(ndev); - struct net_device_stats *stats = &ndev->stats; - u8 isr; - - while (1) { - u8 unsent = readb(&priv->regs->tfcr); - - unsent = (unsent & RCAR_CAN_TFCR_TFUST) >> - RCAR_CAN_TFCR_TFUST_SHIFT; - if (priv->tx_head - priv->tx_tail <= unsent) - break; - stats->tx_packets++; - stats->tx_bytes += priv->tx_dlc[priv->tx_tail % - RCAR_CAN_FIFO_DEPTH]; - priv->tx_dlc[priv->tx_tail % RCAR_CAN_FIFO_DEPTH] = 0; - can_get_echo_skb(ndev, priv->tx_tail % RCAR_CAN_FIFO_DEPTH); - priv->tx_tail++; - netif_wake_queue(ndev); - } - /* Clear interrupt */ - isr = readb(&priv->regs->isr); - writeb(isr & ~RCAR_CAN_ISR_TXFF, &priv->regs->isr); - can_led_event(ndev, CAN_LED_EVENT_TX); -} - -static irqreturn_t rcar_can_interrupt(int irq, void *dev_id) -{ - struct net_device *ndev = dev_id; - struct rcar_can_priv *priv = netdev_priv(ndev); - u8 isr; - - isr = readb(&priv->regs->isr); - if (!(isr & priv->ier)) - return IRQ_NONE; - - if (isr & RCAR_CAN_ISR_ERSF) - rcar_can_error(ndev); - - if (isr & RCAR_CAN_ISR_TXFF) - rcar_can_tx_done(ndev); - - if (isr & RCAR_CAN_ISR_RXFF) { - if (napi_schedule_prep(&priv->napi)) { - /* Disable Rx FIFO interrupts */ - priv->ier &= ~RCAR_CAN_IER_RXFIE; - writeb(priv->ier, &priv->regs->ier); - __napi_schedule(&priv->napi); - } - } - - return IRQ_HANDLED; -} - -static void rcar_can_set_bittiming(struct net_device *dev) -{ - struct rcar_can_priv *priv = netdev_priv(dev); - struct can_bittiming *bt = &priv->can.bittiming; - u32 bcr; - - bcr = RCAR_CAN_BCR_TSEG1(bt->phase_seg1 + bt->prop_seg - 1) | - RCAR_CAN_BCR_BPR(bt->brp - 1) | RCAR_CAN_BCR_SJW(bt->sjw - 1) | - RCAR_CAN_BCR_TSEG2(bt->phase_seg2 - 1); - /* Don't overwrite CLKR with 32-bit BCR access; CLKR has 8-bit access. - * All the registers are big-endian but they get byte-swapped on 32-bit - * read/write (but not on 8-bit, contrary to the manuals)... - */ - writel((bcr << 8) | priv->clock_select, &priv->regs->bcr); -} - -static void rcar_can_start(struct net_device *ndev) -{ - struct rcar_can_priv *priv = netdev_priv(ndev); - u16 ctlr; - int i; - - /* Set controller to known mode: - * - FIFO mailbox mode - * - accept all messages - * - overrun mode - * CAN is in sleep mode after MCU hardware or software reset. - */ - ctlr = readw(&priv->regs->ctlr); - ctlr &= ~RCAR_CAN_CTLR_SLPM; - writew(ctlr, &priv->regs->ctlr); - /* Go to reset mode */ - ctlr |= RCAR_CAN_CTLR_CANM_FORCE_RESET; - writew(ctlr, &priv->regs->ctlr); - for (i = 0; i < MAX_STR_READS; i++) { - if (readw(&priv->regs->str) & RCAR_CAN_STR_RSTST) - break; - } - rcar_can_set_bittiming(ndev); - ctlr |= RCAR_CAN_CTLR_IDFM_MIXED; /* Select mixed ID mode */ - ctlr |= RCAR_CAN_CTLR_BOM_ENT; /* Entry to halt mode automatically */ - /* at bus-off */ - ctlr |= RCAR_CAN_CTLR_MBM; /* Select FIFO mailbox mode */ - ctlr |= RCAR_CAN_CTLR_MLM; /* Overrun mode */ - writew(ctlr, &priv->regs->ctlr); - - /* Accept all SID and EID */ - writel(0, &priv->regs->mkr_2_9[6]); - writel(0, &priv->regs->mkr_2_9[7]); - /* In FIFO mailbox mode, write "0" to bits 24 to 31 */ - writel(0, &priv->regs->mkivlr1); - /* Accept all frames */ - writel(0, &priv->regs->fidcr[0]); - writel(RCAR_CAN_FIDCR_IDE | RCAR_CAN_FIDCR_RTR, &priv->regs->fidcr[1]); - /* Enable and configure FIFO mailbox interrupts */ - writel(RCAR_CAN_MIER1_RXFIE | RCAR_CAN_MIER1_TXFIE, &priv->regs->mier1); - - priv->ier = RCAR_CAN_IER_ERSIE | RCAR_CAN_IER_RXFIE | - RCAR_CAN_IER_TXFIE; - writeb(priv->ier, &priv->regs->ier); - - /* Accumulate error codes */ - writeb(RCAR_CAN_ECSR_EDPM, &priv->regs->ecsr); - /* Enable error interrupts */ - writeb(RCAR_CAN_EIER_EWIE | RCAR_CAN_EIER_EPIE | RCAR_CAN_EIER_BOEIE | - (priv->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING ? - RCAR_CAN_EIER_BEIE : 0) | RCAR_CAN_EIER_ORIE | - RCAR_CAN_EIER_OLIE, &priv->regs->eier); - priv->can.state = CAN_STATE_ERROR_ACTIVE; - - /* Go to operation mode */ - writew(ctlr & ~RCAR_CAN_CTLR_CANM, &priv->regs->ctlr); - for (i = 0; i < MAX_STR_READS; i++) { - if (!(readw(&priv->regs->str) & RCAR_CAN_STR_RSTST)) - break; - } - /* Enable Rx and Tx FIFO */ - writeb(RCAR_CAN_RFCR_RFE, &priv->regs->rfcr); - writeb(RCAR_CAN_TFCR_TFE, &priv->regs->tfcr); -} - -static int rcar_can_open(struct net_device *ndev) -{ - struct rcar_can_priv *priv = netdev_priv(ndev); - int err; - - err = clk_prepare_enable(priv->clk); - if (err) { - netdev_err(ndev, - "failed to enable peripheral clock, error %d\n", - err); - goto out; - } - err = clk_prepare_enable(priv->can_clk); - if (err) { - netdev_err(ndev, "failed to enable CAN clock, error %d\n", - err); - goto out_clock; - } - err = open_candev(ndev); - if (err) { - netdev_err(ndev, "open_candev() failed, error %d\n", err); - goto out_can_clock; - } - napi_enable(&priv->napi); - err = request_irq(ndev->irq, rcar_can_interrupt, 0, ndev->name, ndev); - if (err) { - netdev_err(ndev, "request_irq(%d) failed, error %d\n", - ndev->irq, err); - goto out_close; - } - can_led_event(ndev, CAN_LED_EVENT_OPEN); - rcar_can_start(ndev); - netif_start_queue(ndev); - return 0; -out_close: - napi_disable(&priv->napi); - close_candev(ndev); -out_can_clock: - clk_disable_unprepare(priv->can_clk); -out_clock: - clk_disable_unprepare(priv->clk); -out: - return err; -} - -static void rcar_can_stop(struct net_device *ndev) -{ - struct rcar_can_priv *priv = netdev_priv(ndev); - u16 ctlr; - int i; - - /* Go to (force) reset mode */ - ctlr = readw(&priv->regs->ctlr); - ctlr |= RCAR_CAN_CTLR_CANM_FORCE_RESET; - writew(ctlr, &priv->regs->ctlr); - for (i = 0; i < MAX_STR_READS; i++) { - if (readw(&priv->regs->str) & RCAR_CAN_STR_RSTST) - break; - } - writel(0, &priv->regs->mier0); - writel(0, &priv->regs->mier1); - writeb(0, &priv->regs->ier); - writeb(0, &priv->regs->eier); - /* Go to sleep mode */ - ctlr |= RCAR_CAN_CTLR_SLPM; - writew(ctlr, &priv->regs->ctlr); - priv->can.state = CAN_STATE_STOPPED; -} - -static int rcar_can_close(struct net_device *ndev) -{ - struct rcar_can_priv *priv = netdev_priv(ndev); - - netif_stop_queue(ndev); - rcar_can_stop(ndev); - free_irq(ndev->irq, ndev); - napi_disable(&priv->napi); - clk_disable_unprepare(priv->can_clk); - clk_disable_unprepare(priv->clk); - close_candev(ndev); - can_led_event(ndev, CAN_LED_EVENT_STOP); - return 0; -} - -static netdev_tx_t rcar_can_start_xmit(struct sk_buff *skb, - struct net_device *ndev) -{ - struct rcar_can_priv *priv = netdev_priv(ndev); - struct can_frame *cf = (struct can_frame *)skb->data; - u32 data, i; - - if (can_dropped_invalid_skb(ndev, skb)) - return NETDEV_TX_OK; - - if (cf->can_id & CAN_EFF_FLAG) /* Extended frame format */ - data = (cf->can_id & CAN_EFF_MASK) | RCAR_CAN_IDE; - else /* Standard frame format */ - data = (cf->can_id & CAN_SFF_MASK) << RCAR_CAN_SID_SHIFT; - - if (cf->can_id & CAN_RTR_FLAG) { /* Remote transmission request */ - data |= RCAR_CAN_RTR; - } else { - for (i = 0; i < cf->can_dlc; i++) - writeb(cf->data[i], - &priv->regs->mb[RCAR_CAN_TX_FIFO_MBX].data[i]); - } - - writel(data, &priv->regs->mb[RCAR_CAN_TX_FIFO_MBX].id); - - writeb(cf->can_dlc, &priv->regs->mb[RCAR_CAN_TX_FIFO_MBX].dlc); - - priv->tx_dlc[priv->tx_head % RCAR_CAN_FIFO_DEPTH] = cf->can_dlc; - can_put_echo_skb(skb, ndev, priv->tx_head % RCAR_CAN_FIFO_DEPTH); - priv->tx_head++; - /* Start Tx: write 0xff to the TFPCR register to increment - * the CPU-side pointer for the transmit FIFO to the next - * mailbox location - */ - writeb(0xff, &priv->regs->tfpcr); - /* Stop the queue if we've filled all FIFO entries */ - if (priv->tx_head - priv->tx_tail >= RCAR_CAN_FIFO_DEPTH) - netif_stop_queue(ndev); - - return NETDEV_TX_OK; -} - -static const struct net_device_ops rcar_can_netdev_ops = { - .ndo_open = rcar_can_open, - .ndo_stop = rcar_can_close, - .ndo_start_xmit = rcar_can_start_xmit, - .ndo_change_mtu = can_change_mtu, -}; - -static void rcar_can_rx_pkt(struct rcar_can_priv *priv) -{ - struct net_device_stats *stats = &priv->ndev->stats; - struct can_frame *cf; - struct sk_buff *skb; - u32 data; - u8 dlc; - - skb = alloc_can_skb(priv->ndev, &cf); - if (!skb) { - stats->rx_dropped++; - return; - } - - data = readl(&priv->regs->mb[RCAR_CAN_RX_FIFO_MBX].id); - if (data & RCAR_CAN_IDE) - cf->can_id = (data & CAN_EFF_MASK) | CAN_EFF_FLAG; - else - cf->can_id = (data >> RCAR_CAN_SID_SHIFT) & CAN_SFF_MASK; - - dlc = readb(&priv->regs->mb[RCAR_CAN_RX_FIFO_MBX].dlc); - cf->can_dlc = get_can_dlc(dlc); - if (data & RCAR_CAN_RTR) { - cf->can_id |= CAN_RTR_FLAG; - } else { - for (dlc = 0; dlc < cf->can_dlc; dlc++) - cf->data[dlc] = - readb(&priv->regs->mb[RCAR_CAN_RX_FIFO_MBX].data[dlc]); - } - - can_led_event(priv->ndev, CAN_LED_EVENT_RX); - - stats->rx_bytes += cf->can_dlc; - stats->rx_packets++; - netif_receive_skb(skb); -} - -static int rcar_can_rx_poll(struct napi_struct *napi, int quota) -{ - struct rcar_can_priv *priv = container_of(napi, - struct rcar_can_priv, napi); - int num_pkts; - - for (num_pkts = 0; num_pkts < quota; num_pkts++) { - u8 rfcr, isr; - - isr = readb(&priv->regs->isr); - /* Clear interrupt bit */ - if (isr & RCAR_CAN_ISR_RXFF) - writeb(isr & ~RCAR_CAN_ISR_RXFF, &priv->regs->isr); - rfcr = readb(&priv->regs->rfcr); - if (rfcr & RCAR_CAN_RFCR_RFEST) - break; - rcar_can_rx_pkt(priv); - /* Write 0xff to the RFPCR register to increment - * the CPU-side pointer for the receive FIFO - * to the next mailbox location - */ - writeb(0xff, &priv->regs->rfpcr); - } - /* All packets processed */ - if (num_pkts < quota) { - napi_complete(napi); - priv->ier |= RCAR_CAN_IER_RXFIE; - writeb(priv->ier, &priv->regs->ier); - } - return num_pkts; -} - -static int rcar_can_do_set_mode(struct net_device *ndev, enum can_mode mode) -{ - switch (mode) { - case CAN_MODE_START: - rcar_can_start(ndev); - netif_wake_queue(ndev); - return 0; - default: - return -EOPNOTSUPP; - } -} - -static int rcar_can_get_berr_counter(const struct net_device *dev, - struct can_berr_counter *bec) -{ - struct rcar_can_priv *priv = netdev_priv(dev); - int err; - - err = clk_prepare_enable(priv->clk); - if (err) - return err; - bec->txerr = readb(&priv->regs->tecr); - bec->rxerr = readb(&priv->regs->recr); - clk_disable_unprepare(priv->clk); - return 0; -} - -static const char * const clock_names[] = { - [CLKR_CLKP1] = "clkp1", - [CLKR_CLKP2] = "clkp2", - [CLKR_CLKEXT] = "can_clk", -}; - -static int rcar_can_probe(struct platform_device *pdev) -{ - struct rcar_can_platform_data *pdata; - struct rcar_can_priv *priv; - struct net_device *ndev; - struct resource *mem; - void __iomem *addr; - u32 clock_select = CLKR_CLKP1; - int err = -ENODEV; - int irq; - - if (pdev->dev.of_node) { - of_property_read_u32(pdev->dev.of_node, - "renesas,can-clock-select", &clock_select); - } else { - pdata = dev_get_platdata(&pdev->dev); - if (!pdata) { - dev_err(&pdev->dev, "No platform data provided!\n"); - goto fail; - } - clock_select = pdata->clock_select; - } - - irq = platform_get_irq(pdev, 0); - if (irq < 0) { - dev_err(&pdev->dev, "No IRQ resource\n"); - err = irq; - goto fail; - } - - mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); - addr = devm_ioremap_resource(&pdev->dev, mem); - if (IS_ERR(addr)) { - err = PTR_ERR(addr); - goto fail; - } - - ndev = alloc_candev(sizeof(struct rcar_can_priv), RCAR_CAN_FIFO_DEPTH); - if (!ndev) { - dev_err(&pdev->dev, "alloc_candev() failed\n"); - err = -ENOMEM; - goto fail; - } - - priv = netdev_priv(ndev); - - priv->clk = devm_clk_get(&pdev->dev, "clkp1"); - if (IS_ERR(priv->clk)) { - err = PTR_ERR(priv->clk); - dev_err(&pdev->dev, "cannot get peripheral clock, error %d\n", - err); - goto fail_clk; - } - - if (clock_select >= ARRAY_SIZE(clock_names)) { - err = -EINVAL; - dev_err(&pdev->dev, "invalid CAN clock selected\n"); - goto fail_clk; - } - priv->can_clk = devm_clk_get(&pdev->dev, clock_names[clock_select]); - if (IS_ERR(priv->can_clk)) { - err = PTR_ERR(priv->can_clk); - dev_err(&pdev->dev, "cannot get CAN clock, error %d\n", err); - goto fail_clk; - } - - ndev->netdev_ops = &rcar_can_netdev_ops; - ndev->irq = irq; - ndev->flags |= IFF_ECHO; - priv->ndev = ndev; - priv->regs = addr; - priv->clock_select = clock_select; - priv->can.clock.freq = clk_get_rate(priv->can_clk); - priv->can.bittiming_const = &rcar_can_bittiming_const; - priv->can.do_set_mode = rcar_can_do_set_mode; - priv->can.do_get_berr_counter = rcar_can_get_berr_counter; - priv->can.ctrlmode_supported = CAN_CTRLMODE_BERR_REPORTING; - platform_set_drvdata(pdev, ndev); - SET_NETDEV_DEV(ndev, &pdev->dev); - - netif_napi_add(ndev, &priv->napi, rcar_can_rx_poll, - RCAR_CAN_NAPI_WEIGHT); - err = register_candev(ndev); - if (err) { - dev_err(&pdev->dev, "register_candev() failed, error %d\n", - err); - goto fail_candev; - } - - devm_can_led_init(ndev); - - dev_info(&pdev->dev, "device registered (regs @ %p, IRQ%d)\n", - priv->regs, ndev->irq); - - return 0; -fail_candev: - netif_napi_del(&priv->napi); -fail_clk: - free_candev(ndev); -fail: - return err; -} - -static int rcar_can_remove(struct platform_device *pdev) -{ - struct net_device *ndev = platform_get_drvdata(pdev); - struct rcar_can_priv *priv = netdev_priv(ndev); - - unregister_candev(ndev); - netif_napi_del(&priv->napi); - free_candev(ndev); - return 0; -} - -static int __maybe_unused rcar_can_suspend(struct device *dev) -{ - struct net_device *ndev = dev_get_drvdata(dev); - struct rcar_can_priv *priv = netdev_priv(ndev); - u16 ctlr; - - if (netif_running(ndev)) { - netif_stop_queue(ndev); - netif_device_detach(ndev); - } - ctlr = readw(&priv->regs->ctlr); - ctlr |= RCAR_CAN_CTLR_CANM_HALT; - writew(ctlr, &priv->regs->ctlr); - ctlr |= RCAR_CAN_CTLR_SLPM; - writew(ctlr, &priv->regs->ctlr); - priv->can.state = CAN_STATE_SLEEPING; - - clk_disable(priv->clk); - return 0; -} - -static int __maybe_unused rcar_can_resume(struct device *dev) -{ - struct net_device *ndev = dev_get_drvdata(dev); - struct rcar_can_priv *priv = netdev_priv(ndev); - u16 ctlr; - int err; - - err = clk_enable(priv->clk); - if (err) { - netdev_err(ndev, "clk_enable() failed, error %d\n", err); - return err; - } - - ctlr = readw(&priv->regs->ctlr); - ctlr &= ~RCAR_CAN_CTLR_SLPM; - writew(ctlr, &priv->regs->ctlr); - ctlr &= ~RCAR_CAN_CTLR_CANM; - writew(ctlr, &priv->regs->ctlr); - priv->can.state = CAN_STATE_ERROR_ACTIVE; - - if (netif_running(ndev)) { - netif_device_attach(ndev); - netif_start_queue(ndev); - } - return 0; -} - -static SIMPLE_DEV_PM_OPS(rcar_can_pm_ops, rcar_can_suspend, rcar_can_resume); - -static const struct of_device_id rcar_can_of_table[] __maybe_unused = { - { .compatible = "renesas,can-r8a7778" }, - { .compatible = "renesas,can-r8a7779" }, - { .compatible = "renesas,can-r8a7790" }, - { .compatible = "renesas,can-r8a7791" }, - { .compatible = "renesas,rcar-gen1-can" }, - { .compatible = "renesas,rcar-gen2-can" }, - { .compatible = "renesas,rcar-gen3-can" }, - { } -}; -MODULE_DEVICE_TABLE(of, rcar_can_of_table); - -static struct platform_driver rcar_can_driver = { - .driver = { - .name = RCAR_CAN_DRV_NAME, - .of_match_table = of_match_ptr(rcar_can_of_table), - .pm = &rcar_can_pm_ops, - }, - .probe = rcar_can_probe, - .remove = rcar_can_remove, -}; - -module_platform_driver(rcar_can_driver); - -MODULE_AUTHOR("Cogent Embedded, Inc."); -MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION("CAN driver for Renesas R-Car SoC"); -MODULE_ALIAS("platform:" RCAR_CAN_DRV_NAME); -- cgit v0.10.2 From 86a98057256020e75e1be0f88d7617491a06e8f1 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:20:44 -0700 Subject: vxlan/geneve: Include udp_tunnel.h in vxlan/geneve.h and fixup includes This patch makes it so that we add udp_tunnel.h to vxlan.h and geneve.h header files. This is useful as I plan to move the generic handlers for the port offloads into the udp_tunnel header file and leave the vxlan and geneve headers to be a bit more protocol specific. I also went through and cleaned out a number of redundant includes that where in the .h and .c files for these drivers. Signed-off-by: Alexander Duyck Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index cadefe4..e5e33cd 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -12,7 +12,6 @@ #include #include -#include #include #include #include diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index f999db2..10ad41d 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -11,32 +11,18 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include -#include #include #include #include -#include -#include -#include -#include -#include #include #include -#include #include -#include -#include #include #include #include #include -#include #include -#include -#include #include -#include -#include #include #include #include @@ -44,12 +30,9 @@ #include #if IS_ENABLED(CONFIG_IPV6) -#include -#include #include #include #endif -#include #define VXLAN_VERSION "0.1" diff --git a/include/net/geneve.h b/include/net/geneve.h index cb544a5..f8aff18 100644 --- a/include/net/geneve.h +++ b/include/net/geneve.h @@ -1,10 +1,7 @@ #ifndef __NET_GENEVE_H #define __NET_GENEVE_H 1 -#ifdef CONFIG_INET #include -#endif - /* Geneve Header: * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h index 9d14f70..5901956 100644 --- a/include/net/udp_tunnel.h +++ b/include/net/udp_tunnel.h @@ -105,12 +105,14 @@ struct metadata_dst *udp_tun_rx_dst(struct sk_buff *skb, unsigned short family, __be16 flags, __be64 tunnel_id, int md_size); +#ifdef CONFIG_INET static inline int udp_tunnel_handle_offloads(struct sk_buff *skb, bool udp_csum) { int type = udp_csum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL; return iptunnel_handle_offloads(skb, type); } +#endif static inline void udp_tunnel_encap_enable(struct socket *sock) { diff --git a/include/net/vxlan.h b/include/net/vxlan.h index b880316..7d94494 100644 --- a/include/net/vxlan.h +++ b/include/net/vxlan.h @@ -1,12 +1,8 @@ #ifndef __NET_VXLAN_H #define __NET_VXLAN_H 1 -#include -#include #include -#include -#include -#include +#include #include /* VXLAN protocol (RFC 7348) header: -- cgit v0.10.2 From e7b3db5e60e8f471c3f5ef93b497bafe5863e56a Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:20:52 -0700 Subject: net: Combine GENEVE and VXLAN port notifiers into single functions This patch merges the GENEVE and VXLAN code so that both functions pass through a shared code path. This way we can start the effort of using a single function on the network device drivers to handle both of these tunnel types. Signed-off-by: Alexander Duyck Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index e5e33cd..af6f676 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -396,23 +396,6 @@ static struct socket *geneve_create_sock(struct net *net, bool ipv6, return sock; } -static void geneve_notify_add_rx_port(struct geneve_sock *gs) -{ - struct net_device *dev; - struct sock *sk = gs->sock->sk; - struct net *net = sock_net(sk); - sa_family_t sa_family = geneve_get_sk_family(gs); - __be16 port = inet_sk(sk)->inet_sport; - - rcu_read_lock(); - for_each_netdev_rcu(net, dev) { - if (dev->netdev_ops->ndo_add_geneve_port) - dev->netdev_ops->ndo_add_geneve_port(dev, sa_family, - port); - } - rcu_read_unlock(); -} - static int geneve_hlen(struct genevehdr *gh) { return sizeof(*gh) + gh->opt_len * 4; @@ -532,7 +515,7 @@ static struct geneve_sock *geneve_socket_create(struct net *net, __be16 port, INIT_HLIST_HEAD(&gs->vni_list[h]); /* Initialize the geneve udp offloads structure */ - geneve_notify_add_rx_port(gs); + udp_tunnel_notify_add_rx_port(gs->sock, UDP_TUNNEL_TYPE_GENEVE); /* Mark socket as an encapsulation socket */ memset(&tunnel_cfg, 0, sizeof(tunnel_cfg)); @@ -547,31 +530,13 @@ static struct geneve_sock *geneve_socket_create(struct net *net, __be16 port, return gs; } -static void geneve_notify_del_rx_port(struct geneve_sock *gs) -{ - struct net_device *dev; - struct sock *sk = gs->sock->sk; - struct net *net = sock_net(sk); - sa_family_t sa_family = geneve_get_sk_family(gs); - __be16 port = inet_sk(sk)->inet_sport; - - rcu_read_lock(); - for_each_netdev_rcu(net, dev) { - if (dev->netdev_ops->ndo_del_geneve_port) - dev->netdev_ops->ndo_del_geneve_port(dev, sa_family, - port); - } - - rcu_read_unlock(); -} - static void __geneve_sock_release(struct geneve_sock *gs) { if (!gs || --gs->refcnt) return; list_del(&gs->list); - geneve_notify_del_rx_port(gs); + udp_tunnel_notify_del_rx_port(gs->sock, UDP_TUNNEL_TYPE_GENEVE); udp_tunnel_sock_release(gs->sock); kfree_rcu(gs, rcu); } @@ -1164,29 +1129,20 @@ static struct device_type geneve_type = { .name = "geneve", }; -/* Calls the ndo_add_geneve_port of the caller in order to +/* Calls the ndo_add_udp_enc_port of the caller in order to * supply the listening GENEVE udp ports. Callers are expected - * to implement the ndo_add_geneve_port. + * to implement the ndo_add_udp_enc_port. */ static void geneve_push_rx_ports(struct net_device *dev) { struct net *net = dev_net(dev); struct geneve_net *gn = net_generic(net, geneve_net_id); struct geneve_sock *gs; - sa_family_t sa_family; - struct sock *sk; - __be16 port; - - if (!dev->netdev_ops->ndo_add_geneve_port) - return; rcu_read_lock(); - list_for_each_entry_rcu(gs, &gn->sock_list, list) { - sk = gs->sock->sk; - sa_family = sk->sk_family; - port = inet_sk(sk)->inet_sport; - dev->netdev_ops->ndo_add_geneve_port(dev, sa_family, port); - } + list_for_each_entry_rcu(gs, &gn->sock_list, list) + udp_tunnel_push_rx_port(dev, gs->sock, + UDP_TUNNEL_TYPE_GENEVE); rcu_read_unlock(); } diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 10ad41d..adbd979 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -602,42 +602,6 @@ static int vxlan_gro_complete(struct sock *sk, struct sk_buff *skb, int nhoff) return eth_gro_complete(skb, nhoff + sizeof(struct vxlanhdr)); } -/* Notify netdevs that UDP port started listening */ -static void vxlan_notify_add_rx_port(struct vxlan_sock *vs) -{ - struct net_device *dev; - struct sock *sk = vs->sock->sk; - struct net *net = sock_net(sk); - sa_family_t sa_family = vxlan_get_sk_family(vs); - __be16 port = inet_sk(sk)->inet_sport; - - rcu_read_lock(); - for_each_netdev_rcu(net, dev) { - if (dev->netdev_ops->ndo_add_vxlan_port) - dev->netdev_ops->ndo_add_vxlan_port(dev, sa_family, - port); - } - rcu_read_unlock(); -} - -/* Notify netdevs that UDP port is no more listening */ -static void vxlan_notify_del_rx_port(struct vxlan_sock *vs) -{ - struct net_device *dev; - struct sock *sk = vs->sock->sk; - struct net *net = sock_net(sk); - sa_family_t sa_family = vxlan_get_sk_family(vs); - __be16 port = inet_sk(sk)->inet_sport; - - rcu_read_lock(); - for_each_netdev_rcu(net, dev) { - if (dev->netdev_ops->ndo_del_vxlan_port) - dev->netdev_ops->ndo_del_vxlan_port(dev, sa_family, - port); - } - rcu_read_unlock(); -} - /* Add new entry to forwarding table -- assumes lock held */ static int vxlan_fdb_create(struct vxlan_dev *vxlan, const u8 *mac, union vxlan_addr *ip, @@ -1033,7 +997,8 @@ static bool __vxlan_sock_release_prep(struct vxlan_sock *vs) vn = net_generic(sock_net(vs->sock->sk), vxlan_net_id); spin_lock(&vn->sock_lock); hlist_del_rcu(&vs->hlist); - vxlan_notify_del_rx_port(vs); + udp_tunnel_notify_del_rx_port(vs->sock, + UDP_TUNNEL_TYPE_VXLAN); spin_unlock(&vn->sock_lock); return true; @@ -2508,30 +2473,22 @@ static struct device_type vxlan_type = { .name = "vxlan", }; -/* Calls the ndo_add_vxlan_port of the caller in order to +/* Calls the ndo_add_udp_enc_port of the caller in order to * supply the listening VXLAN udp ports. Callers are expected - * to implement the ndo_add_vxlan_port. + * to implement the ndo_add_udp_enc_port. */ static void vxlan_push_rx_ports(struct net_device *dev) { struct vxlan_sock *vs; struct net *net = dev_net(dev); struct vxlan_net *vn = net_generic(net, vxlan_net_id); - sa_family_t sa_family; - __be16 port; unsigned int i; - if (!dev->netdev_ops->ndo_add_vxlan_port) - return; - spin_lock(&vn->sock_lock); for (i = 0; i < PORT_HASH_SIZE; ++i) { - hlist_for_each_entry_rcu(vs, &vn->sock_list[i], hlist) { - port = inet_sk(vs->sock->sk)->inet_sport; - sa_family = vxlan_get_sk_family(vs); - dev->netdev_ops->ndo_add_vxlan_port(dev, sa_family, - port); - } + hlist_for_each_entry_rcu(vs, &vn->sock_list[i], hlist) + udp_tunnel_push_rx_port(dev, vs->sock, + UDP_TUNNEL_TYPE_VXLAN); } spin_unlock(&vn->sock_lock); } @@ -2733,7 +2690,8 @@ static struct vxlan_sock *vxlan_socket_create(struct net *net, bool ipv6, spin_lock(&vn->sock_lock); hlist_add_head_rcu(&vs->hlist, vs_head(net, port)); - vxlan_notify_add_rx_port(vs); + udp_tunnel_notify_add_rx_port(sock, + UDP_TUNNEL_TYPE_VXLAN); spin_unlock(&vn->sock_lock); /* Mark socket as an encapsulation socket. */ diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h index 5901956..71afbea 100644 --- a/include/net/udp_tunnel.h +++ b/include/net/udp_tunnel.h @@ -84,6 +84,39 @@ struct udp_tunnel_sock_cfg { void setup_udp_tunnel_sock(struct net *net, struct socket *sock, struct udp_tunnel_sock_cfg *sock_cfg); +/* -- List of parsable UDP tunnel types -- + * + * Adding to this list will result in serious debate. The main issue is + * that this list is essentially a list of workarounds for either poorly + * designed tunnels, or poorly designed device offloads. + * + * The parsing supported via these types should really be used for Rx + * traffic only as the network stack will have already inserted offsets for + * the location of the headers in the skb. In addition any ports that are + * pushed should be kept within the namespace without leaking to other + * devices such as VFs or other ports on the same device. + * + * It is strongly encouraged to use CHECKSUM_COMPLETE for Rx to avoid the + * need to use this for Rx checksum offload. It should not be necessary to + * call this function to perform Tx offloads on outgoing traffic. + */ +enum udp_parsable_tunnel_type { + UDP_TUNNEL_TYPE_VXLAN, /* RFC 7348 */ + UDP_TUNNEL_TYPE_GENEVE, /* draft-ietf-nvo3-geneve */ +}; + +struct udp_tunnel_info { + unsigned short type; + sa_family_t sa_family; + __be16 port; +}; + +/* Notify network devices of offloadable types */ +void udp_tunnel_push_rx_port(struct net_device *dev, struct socket *sock, + unsigned short type); +void udp_tunnel_notify_add_rx_port(struct socket *sock, unsigned short type); +void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type); + /* Transmit the skb using UDP encapsulation. */ void udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb, __be32 src, __be32 dst, __u8 tos, __u8 ttl, diff --git a/net/ipv4/udp_tunnel.c b/net/ipv4/udp_tunnel.c index 47f12c7..8174753 100644 --- a/net/ipv4/udp_tunnel.c +++ b/net/ipv4/udp_tunnel.c @@ -76,6 +76,108 @@ void setup_udp_tunnel_sock(struct net *net, struct socket *sock, } EXPORT_SYMBOL_GPL(setup_udp_tunnel_sock); +static void __udp_tunnel_push_rx_port(struct net_device *dev, + struct udp_tunnel_info *ti) +{ + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (!dev->netdev_ops->ndo_add_vxlan_port) + break; + + dev->netdev_ops->ndo_add_vxlan_port(dev, + ti->sa_family, + ti->port); + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (!dev->netdev_ops->ndo_add_geneve_port) + break; + + dev->netdev_ops->ndo_add_geneve_port(dev, + ti->sa_family, + ti->port); + break; + default: + break; + } +} + +void udp_tunnel_push_rx_port(struct net_device *dev, struct socket *sock, + unsigned short type) +{ + struct sock *sk = sock->sk; + struct udp_tunnel_info ti; + + ti.type = type; + ti.sa_family = sk->sk_family; + ti.port = inet_sk(sk)->inet_sport; + + __udp_tunnel_push_rx_port(dev, &ti); +} +EXPORT_SYMBOL_GPL(udp_tunnel_push_rx_port); + +/* Notify netdevs that UDP port started listening */ +void udp_tunnel_notify_add_rx_port(struct socket *sock, unsigned short type) +{ + struct sock *sk = sock->sk; + struct net *net = sock_net(sk); + struct udp_tunnel_info ti; + struct net_device *dev; + + ti.type = type; + ti.sa_family = sk->sk_family; + ti.port = inet_sk(sk)->inet_sport; + + rcu_read_lock(); + for_each_netdev_rcu(net, dev) + __udp_tunnel_push_rx_port(dev, &ti); + rcu_read_unlock(); +} +EXPORT_SYMBOL_GPL(udp_tunnel_notify_add_rx_port); + +static void __udp_tunnel_pull_rx_port(struct net_device *dev, + struct udp_tunnel_info *ti) +{ + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (!dev->netdev_ops->ndo_del_vxlan_port) + break; + + dev->netdev_ops->ndo_del_vxlan_port(dev, + ti->sa_family, + ti->port); + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (!dev->netdev_ops->ndo_del_geneve_port) + break; + + dev->netdev_ops->ndo_del_geneve_port(dev, + ti->sa_family, + ti->port); + break; + default: + break; + } +} + +/* Notify netdevs that UDP port is no more listening */ +void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type) +{ + struct sock *sk = sock->sk; + struct net *net = sock_net(sk); + struct udp_tunnel_info ti; + struct net_device *dev; + + ti.type = type; + ti.sa_family = sk->sk_family; + ti.port = inet_sk(sk)->inet_sport; + + rcu_read_lock(); + for_each_netdev_rcu(net, dev) + __udp_tunnel_pull_rx_port(dev, &ti); + rcu_read_unlock(); +} +EXPORT_SYMBOL_GPL(udp_tunnel_notify_del_rx_port); + void udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb, __be32 src, __be32 dst, __u8 tos, __u8 ttl, __be16 df, __be16 src_port, __be16 dst_port, -- cgit v0.10.2 From 7c46a640de6fcc4f35d0702710356a024eadf68f Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:21:00 -0700 Subject: net: Merge VXLAN and GENEVE push notifiers into a single notifier This patch merges the notifiers for VXLAN and GENEVE into a single UDP tunnel notifier. The idea is that we will want to only have to make one notifier call to receive the list of ports for VXLAN and GENEVE tunnels that need to be offloaded. In addition we add a new set of ndo functions named ndo_udp_tunnel_add and ndo_udp_tunnel_del that are meant to allow us to track the tunnel meta-data such as port and address family as tunnels are added and removed. The tunnel meta-data is now transported in a structure named udp_tunnel_info which for now carries the type, address family, and port number. In the future this could be updated so that we can include a tuple of values including things such as the destination IP address and other fields. I also ended up going with a naming scheme that consisted of using the prefix udp_tunnel on function names. I applied this to the notifier and ndo ops as well so that it hopefully points to the fact that these are primarily used in the udp_tunnel functions. Signed-off-by: Alexander Duyck Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index af6f676..aa61708 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -1497,7 +1497,7 @@ static int geneve_netdevice_event(struct notifier_block *unused, { struct net_device *dev = netdev_notifier_info_to_dev(ptr); - if (event == NETDEV_OFFLOAD_PUSH_GENEVE) + if (event == NETDEV_UDP_TUNNEL_PUSH_INFO) geneve_push_rx_ports(dev); return NOTIFY_DONE; diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index adbd979..31aeec9 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -3239,7 +3239,7 @@ static int vxlan_netdevice_event(struct notifier_block *unused, if (event == NETDEV_UNREGISTER) vxlan_handle_lowerdev_unregister(vn, dev); - else if (event == NETDEV_OFFLOAD_PUSH_VXLAN) + else if (event == NETDEV_UDP_TUNNEL_PUSH_INFO) vxlan_push_rx_ports(dev); return NOTIFY_DONE; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 890158e..577d2a1 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -61,6 +61,8 @@ struct wireless_dev; /* 802.15.4 specific */ struct wpan_dev; struct mpls_dev; +/* UDP Tunnel offloads */ +struct udp_tunnel_info; void netdev_set_default_ethtool_ops(struct net_device *dev, const struct ethtool_ops *ops); @@ -1050,6 +1052,19 @@ struct tc_to_netdev { * address family that vxlan is not listening to anymore. The operation * is protected by the vxlan_net->sock_lock. * + * void (*ndo_udp_tunnel_add)(struct net_device *dev, + * struct udp_tunnel_info *ti); + * Called by UDP tunnel to notify a driver about the UDP port and socket + * address family that a UDP tunnel is listnening to. It is called only + * when a new port starts listening. The operation is protected by the + * RTNL. + * + * void (*ndo_udp_tunnel_del)(struct net_device *dev, + * struct udp_tunnel_info *ti); + * Called by UDP tunnel to notify the driver about a UDP port and socket + * address family that the UDP tunnel is not listening to anymore. The + * operation is protected by the RTNL. + * * void* (*ndo_dfwd_add_station)(struct net_device *pdev, * struct net_device *dev) * Called by upper layer devices to accelerate switching or other @@ -1269,6 +1284,10 @@ struct net_device_ops { void (*ndo_del_geneve_port)(struct net_device *dev, sa_family_t sa_family, __be16 port); + void (*ndo_udp_tunnel_add)(struct net_device *dev, + struct udp_tunnel_info *ti); + void (*ndo_udp_tunnel_del)(struct net_device *dev, + struct udp_tunnel_info *ti); void* (*ndo_dfwd_add_station)(struct net_device *pdev, struct net_device *dev); void (*ndo_dfwd_del_station)(struct net_device *pdev, @@ -2255,8 +2274,7 @@ struct netdev_lag_lower_state_info { #define NETDEV_BONDING_INFO 0x0019 #define NETDEV_PRECHANGEUPPER 0x001A #define NETDEV_CHANGELOWERSTATE 0x001B -#define NETDEV_OFFLOAD_PUSH_VXLAN 0x001C -#define NETDEV_OFFLOAD_PUSH_GENEVE 0x001D +#define NETDEV_UDP_TUNNEL_PUSH_INFO 0x001C int register_netdevice_notifier(struct notifier_block *nb); int unregister_netdevice_notifier(struct notifier_block *nb); diff --git a/include/net/geneve.h b/include/net/geneve.h index f8aff18..3410c4b 100644 --- a/include/net/geneve.h +++ b/include/net/geneve.h @@ -61,8 +61,7 @@ struct genevehdr { static inline void geneve_get_rx_port(struct net_device *netdev) { - ASSERT_RTNL(); - call_netdevice_notifiers(NETDEV_OFFLOAD_PUSH_GENEVE, netdev); + udp_tunnel_get_rx_info(netdev); } #ifdef CONFIG_INET diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h index 71afbea..1c9408a 100644 --- a/include/net/udp_tunnel.h +++ b/include/net/udp_tunnel.h @@ -117,6 +117,12 @@ void udp_tunnel_push_rx_port(struct net_device *dev, struct socket *sock, void udp_tunnel_notify_add_rx_port(struct socket *sock, unsigned short type); void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type); +static inline void udp_tunnel_get_rx_info(struct net_device *dev) +{ + ASSERT_RTNL(); + call_netdevice_notifiers(NETDEV_UDP_TUNNEL_PUSH_INFO, dev); +} + /* Transmit the skb using UDP encapsulation. */ void udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb, __be32 src, __be32 dst, __u8 tos, __u8 ttl, diff --git a/include/net/vxlan.h b/include/net/vxlan.h index 7d94494..c62e2ed 100644 --- a/include/net/vxlan.h +++ b/include/net/vxlan.h @@ -4,6 +4,7 @@ #include #include #include +#include /* VXLAN protocol (RFC 7348) header: * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ @@ -390,8 +391,7 @@ static inline __be32 vxlan_compute_rco(unsigned int start, unsigned int offset) static inline void vxlan_get_rx_port(struct net_device *netdev) { - ASSERT_RTNL(); - call_netdevice_notifiers(NETDEV_OFFLOAD_PUSH_VXLAN, netdev); + udp_tunnel_get_rx_info(netdev); } static inline unsigned short vxlan_get_sk_family(struct vxlan_sock *vs) diff --git a/net/ipv4/udp_tunnel.c b/net/ipv4/udp_tunnel.c index 8174753..683e494 100644 --- a/net/ipv4/udp_tunnel.c +++ b/net/ipv4/udp_tunnel.c @@ -79,6 +79,11 @@ EXPORT_SYMBOL_GPL(setup_udp_tunnel_sock); static void __udp_tunnel_push_rx_port(struct net_device *dev, struct udp_tunnel_info *ti) { + if (dev->netdev_ops->ndo_udp_tunnel_add) { + dev->netdev_ops->ndo_udp_tunnel_add(dev, ti); + return; + } + switch (ti->type) { case UDP_TUNNEL_TYPE_VXLAN: if (!dev->netdev_ops->ndo_add_vxlan_port) @@ -137,6 +142,11 @@ EXPORT_SYMBOL_GPL(udp_tunnel_notify_add_rx_port); static void __udp_tunnel_pull_rx_port(struct net_device *dev, struct udp_tunnel_info *ti) { + if (dev->netdev_ops->ndo_udp_tunnel_del) { + dev->netdev_ops->ndo_udp_tunnel_del(dev, ti); + return; + } + switch (ti->type) { case UDP_TUNNEL_TYPE_VXLAN: if (!dev->netdev_ops->ndo_del_vxlan_port) -- cgit v0.10.2 From 6b3529123dbe2d9cf728231a6857ee8ce2fa7d83 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:21:09 -0700 Subject: bnx2x: Move all UDP port notifiers to single function This patch goes through and combines the notifiers for VXLAN and GENEVE into a single function for each action. So there is now one combined function for getting ports, one for adding the ports, and one for deleting the ports. I also went through and dropped the BNX2X VXLAN and GENEVE specific build flags. Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig index 18042c2..d92c37f 100644 --- a/drivers/net/ethernet/broadcom/Kconfig +++ b/drivers/net/ethernet/broadcom/Kconfig @@ -139,26 +139,6 @@ config BNX2X_SRIOV Virtualization support in the 578xx and 57712 products. This allows for virtual function acceleration in virtual environments. -config BNX2X_VXLAN - bool "Virtual eXtensible Local Area Network support" - default n - depends on BNX2X && VXLAN && !(BNX2X=y && VXLAN=m) - ---help--- - This enables hardward offload support for VXLAN protocol over the - NetXtremeII series adapters. - Say Y here if you want to enable hardware offload support for - Virtual eXtensible Local Area Network (VXLAN) in the driver. - -config BNX2X_GENEVE - bool "Generic Network Virtualization Encapsulation (GENEVE) support" - depends on BNX2X && GENEVE && !(BNX2X=y && GENEVE=m) - ---help--- - This allows one to create GENEVE virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. GENEVE is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to enable hardware offload support for - Generic Network Virtualization Encapsulation (GENEVE) in the driver. - config BGMAC tristate "BCMA bus GBit core support" depends on BCMA && BCMA_HOST_SOC diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index a59d55e..97e8925 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -59,9 +59,6 @@ #include #include #include -#if IS_ENABLED(CONFIG_BNX2X_GENEVE) -#include -#endif #include "bnx2x.h" #include "bnx2x_init.h" #include "bnx2x_init_ops.h" @@ -10076,7 +10073,6 @@ static void bnx2x_parity_recover(struct bnx2x *bp) } } -#if defined(CONFIG_BNX2X_VXLAN) || IS_ENABLED(CONFIG_BNX2X_GENEVE) static int bnx2x_udp_port_update(struct bnx2x *bp) { struct bnx2x_func_switch_update_params *switch_update_params; @@ -10177,47 +10173,42 @@ static void __bnx2x_del_udp_port(struct bnx2x *bp, u16 port, DP(BNX2X_MSG_SP, "Deleted UDP tunnel [%d] port %d\n", type, port); } -#endif - -#ifdef CONFIG_BNX2X_VXLAN -static void bnx2x_add_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) -{ - struct bnx2x *bp = netdev_priv(netdev); - u16 t_port = ntohs(port); - - __bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN); -} - -static void bnx2x_del_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) -{ - struct bnx2x *bp = netdev_priv(netdev); - u16 t_port = ntohs(port); - - __bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN); -} -#endif -#if IS_ENABLED(CONFIG_BNX2X_GENEVE) -static void bnx2x_add_geneve_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) +static void bnx2x_udp_tunnel_add(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct bnx2x *bp = netdev_priv(netdev); - u16 t_port = ntohs(port); + u16 t_port = ntohs(ti->port); - __bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE); + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + __bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN); + break; + case UDP_TUNNEL_TYPE_GENEVE: + __bnx2x_add_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE); + break; + default: + break; + } } -static void bnx2x_del_geneve_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) +static void bnx2x_udp_tunnel_del(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct bnx2x *bp = netdev_priv(netdev); - u16 t_port = ntohs(port); + u16 t_port = ntohs(ti->port); - __bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE); + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + __bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_VXLAN); + break; + case UDP_TUNNEL_TYPE_GENEVE: + __bnx2x_del_udp_port(bp, t_port, BNX2X_UDP_PORT_GENEVE); + break; + default: + break; + } } -#endif static int bnx2x_close(struct net_device *dev); @@ -10325,7 +10316,6 @@ sp_rtnl_not_reset: &bp->sp_rtnl_state)) bnx2x_update_mng_version(bp); -#if defined(CONFIG_BNX2X_VXLAN) || IS_ENABLED(CONFIG_BNX2X_GENEVE) if (test_and_clear_bit(BNX2X_SP_RTNL_CHANGE_UDP_PORT, &bp->sp_rtnl_state)) { if (bnx2x_udp_port_update(bp)) { @@ -10335,20 +10325,14 @@ sp_rtnl_not_reset: BNX2X_UDP_PORT_MAX); } else { /* Since we don't store additional port information, - * if no port is configured for any feature ask for + * if no ports are configured for any feature ask for * information about currently configured ports. */ -#ifdef CONFIG_BNX2X_VXLAN - if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count) - vxlan_get_rx_port(bp->dev); -#endif -#if IS_ENABLED(CONFIG_BNX2X_GENEVE) - if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count) - geneve_get_rx_port(bp->dev); -#endif + if (!bp->udp_tunnel_ports[BNX2X_UDP_PORT_VXLAN].count && + !bp->udp_tunnel_ports[BNX2X_UDP_PORT_GENEVE].count) + udp_tunnel_get_rx_info(bp->dev); } } -#endif /* work which needs rtnl lock not-taken (as it takes the lock itself and * can be called from other contexts as well) @@ -12551,14 +12535,8 @@ static int bnx2x_open(struct net_device *dev) if (rc) return rc; -#ifdef CONFIG_BNX2X_VXLAN - if (IS_PF(bp)) - vxlan_get_rx_port(dev); -#endif -#if IS_ENABLED(CONFIG_BNX2X_GENEVE) if (IS_PF(bp)) - geneve_get_rx_port(dev); -#endif + udp_tunnel_get_rx_info(dev); return 0; } @@ -13045,14 +13023,8 @@ static const struct net_device_ops bnx2x_netdev_ops = { .ndo_get_phys_port_id = bnx2x_get_phys_port_id, .ndo_set_vf_link_state = bnx2x_set_vf_link_state, .ndo_features_check = bnx2x_features_check, -#ifdef CONFIG_BNX2X_VXLAN - .ndo_add_vxlan_port = bnx2x_add_vxlan_port, - .ndo_del_vxlan_port = bnx2x_del_vxlan_port, -#endif -#if IS_ENABLED(CONFIG_BNX2X_GENEVE) - .ndo_add_geneve_port = bnx2x_add_geneve_port, - .ndo_del_geneve_port = bnx2x_del_geneve_port, -#endif + .ndo_udp_tunnel_add = bnx2x_udp_tunnel_add, + .ndo_udp_tunnel_del = bnx2x_udp_tunnel_del, }; static int bnx2x_set_coherency_mask(struct bnx2x *bp) -- cgit v0.10.2 From ad51b8e9f9f4f8172eb7a6219d3005861bfb9a57 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:21:19 -0700 Subject: bnxt: Update drivers to support unified UDP encapsulation offload functions This patch ends up doing several things. First it updates the driver to make use of the new unified UDP tunnel offload notifier functions. In addition I updated the code so that we can work around the bits that were checking for if VXLAN was enabled since we are now using a notifier based setup. Signed-off-by: Alexander Duyck Acked-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig index d92c37f..d74a92e 100644 --- a/drivers/net/ethernet/broadcom/Kconfig +++ b/drivers/net/ethernet/broadcom/Kconfig @@ -166,7 +166,6 @@ config SYSTEMPORT config BNXT tristate "Broadcom NetXtreme-C/E support" depends on PCI - depends on VXLAN || VXLAN=n select FW_LOADER select LIBCRC32C ---help--- diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 9aaa6a6..03a5d84 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -37,9 +37,7 @@ #include #include #include -#if defined(CONFIG_VXLAN) || defined(CONFIG_VXLAN_MODULE) -#include -#endif +#include #ifdef CONFIG_NET_RX_BUSY_POLL #include #endif @@ -5256,9 +5254,7 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init) } if (irq_re_init) { -#if defined(CONFIG_VXLAN) || defined(CONFIG_VXLAN_MODULE) - vxlan_get_rx_port(bp->dev); -#endif + udp_tunnel_get_rx_info(bp->dev); if (!bnxt_hwrm_tunnel_dst_port_alloc( bp, htons(0x17c1), TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE)) @@ -6250,47 +6246,63 @@ static void bnxt_cfg_ntp_filters(struct bnxt *bp) #endif /* CONFIG_RFS_ACCEL */ -static void bnxt_add_vxlan_port(struct net_device *dev, sa_family_t sa_family, - __be16 port) +static void bnxt_udp_tunnel_add(struct net_device *dev, + struct udp_tunnel_info *ti) { struct bnxt *bp = netdev_priv(dev); - if (!netif_running(dev)) + if (ti->sa_family != AF_INET6 && ti->sa_family != AF_INET) return; - if (sa_family != AF_INET6 && sa_family != AF_INET) + if (!netif_running(dev)) return; - if (bp->vxlan_port_cnt && bp->vxlan_port != port) - return; + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (bp->vxlan_port_cnt && bp->vxlan_port != ti->port) + return; - bp->vxlan_port_cnt++; - if (bp->vxlan_port_cnt == 1) { - bp->vxlan_port = port; - set_bit(BNXT_VXLAN_ADD_PORT_SP_EVENT, &bp->sp_event); - schedule_work(&bp->sp_task); + bp->vxlan_port_cnt++; + if (bp->vxlan_port_cnt == 1) { + bp->vxlan_port = ti->port; + set_bit(BNXT_VXLAN_ADD_PORT_SP_EVENT, &bp->sp_event); + schedule_work(&bp->sp_task); + } + break; + default: + return; } + + schedule_work(&bp->sp_task); } -static void bnxt_del_vxlan_port(struct net_device *dev, sa_family_t sa_family, - __be16 port) +static void bnxt_udp_tunnel_del(struct net_device *dev, + struct udp_tunnel_info *ti) { struct bnxt *bp = netdev_priv(dev); - if (!netif_running(dev)) + if (ti->sa_family != AF_INET6 && ti->sa_family != AF_INET) return; - if (sa_family != AF_INET6 && sa_family != AF_INET) + if (!netif_running(dev)) return; - if (bp->vxlan_port_cnt && bp->vxlan_port == port) { + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (!bp->vxlan_port_cnt || bp->vxlan_port != ti->port) + return; bp->vxlan_port_cnt--; - if (bp->vxlan_port_cnt == 0) { - set_bit(BNXT_VXLAN_DEL_PORT_SP_EVENT, &bp->sp_event); - schedule_work(&bp->sp_task); - } + if (bp->vxlan_port_cnt != 0) + return; + + set_bit(BNXT_VXLAN_DEL_PORT_SP_EVENT, &bp->sp_event); + break; + default: + return; } + + schedule_work(&bp->sp_task); } static const struct net_device_ops bnxt_netdev_ops = { @@ -6321,8 +6333,8 @@ static const struct net_device_ops bnxt_netdev_ops = { #ifdef CONFIG_RFS_ACCEL .ndo_rx_flow_steer = bnxt_rx_flow_steer, #endif - .ndo_add_vxlan_port = bnxt_add_vxlan_port, - .ndo_del_vxlan_port = bnxt_del_vxlan_port, + .ndo_udp_tunnel_add = bnxt_udp_tunnel_add, + .ndo_udp_tunnel_del = bnxt_udp_tunnel_del, #ifdef CONFIG_NET_RX_BUSY_POLL .ndo_busy_poll = bnxt_busy_poll, #endif -- cgit v0.10.2 From 7cdd5fc376a51cdf191895c23badd699eddbc901 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:21:36 -0700 Subject: bnxt: Move GENEVE support from hard-coded port to using port notifier The port number for GENEVE is hard coded into the bnxt driver. This is the kind of thing we want to avoid going forward. For now I will integrate this back into the port notifier so that we can change the GENEVE port number if we need to in the future. Signed-off-by: Alexander Duyck Acked-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 03a5d84..673f4d6 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -5253,13 +5253,8 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init) netdev_warn(bp->dev, "failed to update phy settings\n"); } - if (irq_re_init) { + if (irq_re_init) udp_tunnel_get_rx_info(bp->dev); - if (!bnxt_hwrm_tunnel_dst_port_alloc( - bp, htons(0x17c1), - TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE)) - bp->nge_port_cnt = 1; - } set_bit(BNXT_STATE_OPEN, &bp->state); bnxt_enable_int(bp); @@ -5877,6 +5872,15 @@ static void bnxt_sp_task(struct work_struct *work) bnxt_hwrm_tunnel_dst_port_free( bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN); } + if (test_and_clear_bit(BNXT_GENEVE_ADD_PORT_SP_EVENT, &bp->sp_event)) { + bnxt_hwrm_tunnel_dst_port_alloc( + bp, bp->nge_port, + TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE); + } + if (test_and_clear_bit(BNXT_GENEVE_DEL_PORT_SP_EVENT, &bp->sp_event)) { + bnxt_hwrm_tunnel_dst_port_free( + bp, TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE); + } if (test_and_clear_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event)) bnxt_reset(bp, false); @@ -6269,6 +6273,16 @@ static void bnxt_udp_tunnel_add(struct net_device *dev, schedule_work(&bp->sp_task); } break; + case UDP_TUNNEL_TYPE_GENEVE: + if (bp->nge_port_cnt && bp->nge_port != ti->port) + return; + + bp->nge_port_cnt++; + if (bp->nge_port_cnt == 1) { + bp->nge_port = ti->port; + set_bit(BNXT_GENEVE_ADD_PORT_SP_EVENT, &bp->sp_event); + } + break; default: return; } @@ -6298,6 +6312,16 @@ static void bnxt_udp_tunnel_del(struct net_device *dev, set_bit(BNXT_VXLAN_DEL_PORT_SP_EVENT, &bp->sp_event); break; + case UDP_TUNNEL_TYPE_GENEVE: + if (!bp->nge_port_cnt || bp->nge_port != ti->port) + return; + bp->nge_port_cnt--; + + if (bp->nge_port_cnt != 0) + return; + + set_bit(BNXT_GENEVE_DEL_PORT_SP_EVENT, &bp->sp_event); + break; default: return; } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 04cc69b..927ece9 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -1049,6 +1049,7 @@ struct bnxt { __be16 vxlan_port; u8 vxlan_port_cnt; __le16 vxlan_fw_dst_port_id; + __be16 nge_port; u8 nge_port_cnt; __le16 nge_fw_dst_port_id; u8 port_partition_type; @@ -1078,6 +1079,8 @@ struct bnxt { #define BNXT_PERIODIC_STATS_SP_EVENT 9 #define BNXT_HWRM_PORT_MODULE_SP_EVENT 10 #define BNXT_RESET_TASK_SILENT_SP_EVENT 11 +#define BNXT_GENEVE_ADD_PORT_SP_EVENT 12 +#define BNXT_GENEVE_DEL_PORT_SP_EVENT 13 struct bnxt_pf_info pf; #ifdef CONFIG_BNXT_SRIOV -- cgit v0.10.2 From bde6b7cdaed82040ba81d180e8c096c6af3515a9 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:21:43 -0700 Subject: benet: Replace ndo_add/del_vxlan_port with ndo_add/del_udp_enc_port This change replaces the network device operations for adding or removing a VXLAN port with operations that are more generically defined to be used for any UDP offload port but provide a type. As such by just adding a line to verify that the offload type if VXLAN we can maintain the same functionality. I have also gone though and removed the BE2NET_VXLAN config option since it no longer relies on the VXLAN code anyway. Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/Kconfig b/drivers/net/ethernet/emulex/benet/Kconfig index 7108563..b4853ec 100644 --- a/drivers/net/ethernet/emulex/benet/Kconfig +++ b/drivers/net/ethernet/emulex/benet/Kconfig @@ -13,11 +13,3 @@ config BE2NET_HWMON ---help--- Say Y here if you want to expose thermal sensor data on be2net network adapter. - -config BE2NET_VXLAN - bool "VXLAN offload support on be2net driver" - default y - depends on BE2NET && VXLAN && !(BE2NET=y && VXLAN=m) - ---help--- - Say Y here if you want to enable VXLAN offload support on - be2net driver. diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 2451a47..3d94789 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -3625,10 +3625,8 @@ static int be_open(struct net_device *netdev) be_link_status_update(adapter, link_status); netif_tx_start_all_queues(netdev); -#ifdef CONFIG_BE2NET_VXLAN if (skyhawk_chip(adapter)) - vxlan_get_rx_port(netdev); -#endif + udp_tunnel_get_rx_info(netdev); return 0; err: @@ -3755,7 +3753,6 @@ static void be_cancel_err_detection(struct be_adapter *adapter) } } -#ifdef CONFIG_BE2NET_VXLAN static void be_disable_vxlan_offloads(struct be_adapter *adapter) { struct net_device *netdev = adapter->netdev; @@ -3774,7 +3771,6 @@ static void be_disable_vxlan_offloads(struct be_adapter *adapter) netdev->hw_features &= ~(NETIF_F_GSO_UDP_TUNNEL); netdev->features &= ~(NETIF_F_GSO_UDP_TUNNEL); } -#endif static void be_calculate_vf_res(struct be_adapter *adapter, u16 num_vfs, struct be_resources *vft_res) @@ -3875,9 +3871,7 @@ static int be_clear(struct be_adapter *adapter) &vft_res); } -#ifdef CONFIG_BE2NET_VXLAN be_disable_vxlan_offloads(adapter); -#endif kfree(adapter->pmac_id); adapter->pmac_id = NULL; @@ -4705,7 +4699,6 @@ static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, 0, 0, nlflags, filter_mask, NULL); } -#ifdef CONFIG_BE2NET_VXLAN /* VxLAN offload Notes: * * The stack defines tunnel offload flags (hw_enc_features) for IP and doesn't @@ -4720,13 +4713,17 @@ static int be_ndo_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, * adds more than one port, disable offloads and don't re-enable them again * until after all the tunnels are removed. */ -static void be_add_vxlan_port(struct net_device *netdev, sa_family_t sa_family, - __be16 port) +static void be_add_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct be_adapter *adapter = netdev_priv(netdev); struct device *dev = &adapter->pdev->dev; + __be16 port = ti->port; int status; + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + if (lancer_chip(adapter) || BEx_chip(adapter) || be_is_mc(adapter)) return; @@ -4774,10 +4771,14 @@ err: be_disable_vxlan_offloads(adapter); } -static void be_del_vxlan_port(struct net_device *netdev, sa_family_t sa_family, - __be16 port) +static void be_del_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct be_adapter *adapter = netdev_priv(netdev); + __be16 port = ti->port; + + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; if (lancer_chip(adapter) || BEx_chip(adapter) || be_is_mc(adapter)) return; @@ -4839,7 +4840,6 @@ static netdev_features_t be_features_check(struct sk_buff *skb, return features; } -#endif static int be_get_phys_port_id(struct net_device *dev, struct netdev_phys_item_id *ppid) @@ -4887,11 +4887,9 @@ static const struct net_device_ops be_netdev_ops = { #ifdef CONFIG_NET_RX_BUSY_POLL .ndo_busy_poll = be_busy_poll, #endif -#ifdef CONFIG_BE2NET_VXLAN - .ndo_add_vxlan_port = be_add_vxlan_port, - .ndo_del_vxlan_port = be_del_vxlan_port, + .ndo_udp_tunnel_add = be_add_vxlan_port, + .ndo_udp_tunnel_del = be_del_vxlan_port, .ndo_features_check = be_features_check, -#endif .ndo_get_phys_port_id = be_get_phys_port_id, }; -- cgit v0.10.2 From f174cdbe5ba7ff4e25fe20f3801607d648aea8d3 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:21:57 -0700 Subject: fm10k: Replace ndo_add/del_vxlan_port with ndo_add/del_udp_enc_port This change replaces the network device operations for adding or removing a VXLAN port with operations that are more generically defined to be used for any UDP offload port but provide a type. As such by just adding a line to verify that the offload type if VXLAN we can maintain the same functionality. Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig index 714bd10..11fc5e8 100644 --- a/drivers/net/ethernet/intel/Kconfig +++ b/drivers/net/ethernet/intel/Kconfig @@ -307,15 +307,4 @@ config FM10K To compile this driver as a module, choose M here. The module will be called fm10k. MSI-X interrupt support is required -config FM10K_VXLAN - bool "Virtual eXtensible Local Area Network Support" - default n - depends on FM10K && VXLAN && !(FM10K=y && VXLAN=m) - ---help--- - This allows one to create VXLAN virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. VXLAN is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to use Virtual eXtensible Local Area Network - (VXLAN) in the driver. - endif # NET_VENDOR_INTEL diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c index 2a08d3f..d00cb19 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c @@ -20,9 +20,7 @@ #include "fm10k.h" #include -#ifdef CONFIG_FM10K_VXLAN -#include -#endif /* CONFIG_FM10K_VXLAN */ +#include /** * fm10k_setup_tx_resources - allocate Tx resources (Descriptors) @@ -436,6 +434,7 @@ static void fm10k_restore_vxlan_port(struct fm10k_intfc *interface) * @netdev: network interface device structure * @sa_family: Address family of new port * @port: port number used for VXLAN + * @type: Enumerated value specifying udp encapsulation type * * This function is called when a new VXLAN interface has added a new port * number to the range that is currently in use for VXLAN. The new port @@ -444,18 +443,21 @@ static void fm10k_restore_vxlan_port(struct fm10k_intfc *interface) * is always used as the VXLAN port number for offloads. **/ static void fm10k_add_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) { + struct udp_tunnel_info *ti) +{ struct fm10k_intfc *interface = netdev_priv(dev); struct fm10k_vxlan_port *vxlan_port; + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; /* only the PF supports configuring tunnels */ if (interface->hw.mac.type != fm10k_mac_pf) return; /* existing ports are pulled out so our new entry is always last */ fm10k_vxlan_port_for_each(vxlan_port, interface) { - if ((vxlan_port->port == port) && - (vxlan_port->sa_family == sa_family)) { + if ((vxlan_port->port == ti->port) && + (vxlan_port->sa_family == ti->sa_family)) { list_del(&vxlan_port->list); goto insert_tail; } @@ -465,8 +467,8 @@ static void fm10k_add_vxlan_port(struct net_device *dev, vxlan_port = kmalloc(sizeof(*vxlan_port), GFP_ATOMIC); if (!vxlan_port) return; - vxlan_port->port = port; - vxlan_port->sa_family = sa_family; + vxlan_port->port = ti->port; + vxlan_port->sa_family = ti->sa_family; insert_tail: /* add new port value to list */ @@ -480,6 +482,7 @@ insert_tail: * @netdev: network interface device structure * @sa_family: Address family of freed port * @port: port number used for VXLAN + * @type: Enumerated value specifying udp encapsulation type * * This function is called when a new VXLAN interface has freed a port * number from the range that is currently in use for VXLAN. The freed @@ -487,17 +490,20 @@ insert_tail: * the port number for offloads. **/ static void fm10k_del_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) { + struct udp_tunnel_info *ti) +{ struct fm10k_intfc *interface = netdev_priv(dev); struct fm10k_vxlan_port *vxlan_port; + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; if (interface->hw.mac.type != fm10k_mac_pf) return; /* find the port in the list and free it */ fm10k_vxlan_port_for_each(vxlan_port, interface) { - if ((vxlan_port->port == port) && - (vxlan_port->sa_family == sa_family)) { + if ((vxlan_port->port == ti->port) && + (vxlan_port->sa_family == ti->sa_family)) { list_del(&vxlan_port->list); kfree(vxlan_port); break; @@ -553,10 +559,8 @@ int fm10k_open(struct net_device *netdev) if (err) goto err_set_queues; -#ifdef CONFIG_FM10K_VXLAN /* update VXLAN port configuration */ - vxlan_get_rx_port(netdev); -#endif + udp_tunnel_get_rx_info(netdev); fm10k_up(interface); @@ -1375,8 +1379,8 @@ static const struct net_device_ops fm10k_netdev_ops = { .ndo_set_vf_vlan = fm10k_ndo_set_vf_vlan, .ndo_set_vf_rate = fm10k_ndo_set_vf_bw, .ndo_get_vf_config = fm10k_ndo_get_vf_config, - .ndo_add_vxlan_port = fm10k_add_vxlan_port, - .ndo_del_vxlan_port = fm10k_del_vxlan_port, + .ndo_udp_tunnel_add = fm10k_add_vxlan_port, + .ndo_udp_tunnel_del = fm10k_del_vxlan_port, .ndo_dfwd_add_station = fm10k_dfwd_add_station, .ndo_dfwd_del_station = fm10k_dfwd_del_station, #ifdef CONFIG_NET_POLL_CONTROLLER -- cgit v0.10.2 From 06a5f7f167c53d4116c3b13912d082f4116a620f Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:22:06 -0700 Subject: i40e: Move all UDP port notifiers to single function This patch goes through and combines the notifiers for VXLAN and GENEVE into a single function for each action. So there is now one combined function for getting ports, one for adding the ports, and one for deleting the ports. Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig index 11fc5e8..1fbf908 100644 --- a/drivers/net/ethernet/intel/Kconfig +++ b/drivers/net/ethernet/intel/Kconfig @@ -236,27 +236,6 @@ config I40E To compile this driver as a module, choose M here. The module will be called i40e. -config I40E_VXLAN - bool "Virtual eXtensible Local Area Network Support" - default n - depends on I40E && VXLAN && !(I40E=y && VXLAN=m) - ---help--- - This allows one to create VXLAN virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. VXLAN is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to use Virtual eXtensible Local Area Network - (VXLAN) in the driver. - -config I40E_GENEVE - bool "Generic Network Virtualization Encapsulation (GENEVE) Support" - depends on I40E && GENEVE && !(I40E=y && GENEVE=m) - default n - ---help--- - This allows one to create GENEVE virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. GENEVE is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to use GENEVE in the driver. - config I40E_DCB bool "Data Center Bridging (DCB) Support" default n diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 5ea2200..734cba6 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -31,12 +31,7 @@ /* Local includes */ #include "i40e.h" #include "i40e_diag.h" -#if IS_ENABLED(CONFIG_VXLAN) -#include -#endif -#if IS_ENABLED(CONFIG_GENEVE) -#include -#endif +#include const char i40e_driver_name[] = "i40e"; static const char i40e_driver_string[] = @@ -5342,14 +5337,7 @@ int i40e_open(struct net_device *netdev) TCP_FLAG_CWR) >> 16); wr32(&pf->hw, I40E_GLLAN_TSOMSK_L, be32_to_cpu(TCP_FLAG_CWR) >> 16); -#ifdef CONFIG_I40E_VXLAN - vxlan_get_rx_port(netdev); -#endif -#ifdef CONFIG_I40E_GENEVE - if (pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE) - geneve_get_rx_port(netdev); -#endif - + udp_tunnel_get_rx_info(netdev); i40e_notify_client_of_netdev_open(vsi); return 0; @@ -7057,7 +7045,6 @@ static void i40e_handle_mdd_event(struct i40e_pf *pf) **/ static void i40e_sync_udp_filters_subtask(struct i40e_pf *pf) { -#if IS_ENABLED(CONFIG_VXLAN) || IS_ENABLED(CONFIG_GENEVE) struct i40e_hw *hw = &pf->hw; i40e_status ret; __be16 port; @@ -7092,7 +7079,6 @@ static void i40e_sync_udp_filters_subtask(struct i40e_pf *pf) } } } -#endif } /** @@ -8628,7 +8614,6 @@ static int i40e_set_features(struct net_device *netdev, return 0; } -#if IS_ENABLED(CONFIG_VXLAN) || IS_ENABLED(CONFIG_GENEVE) /** * i40e_get_udp_port_idx - Lookup a possibly offloaded for Rx UDP port * @pf: board private structure @@ -8648,21 +8633,18 @@ static u8 i40e_get_udp_port_idx(struct i40e_pf *pf, __be16 port) return i; } -#endif - -#if IS_ENABLED(CONFIG_VXLAN) /** - * i40e_add_vxlan_port - Get notifications about VXLAN ports that come up + * i40e_udp_tunnel_add - Get notifications about UDP tunnel ports that come up * @netdev: This physical port's netdev - * @sa_family: Socket Family that VXLAN is notifying us about - * @port: New UDP port number that VXLAN started listening to + * @ti: Tunnel endpoint information **/ -static void i40e_add_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) +static void i40e_udp_tunnel_add(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct i40e_netdev_priv *np = netdev_priv(netdev); struct i40e_vsi *vsi = np->vsi; struct i40e_pf *pf = vsi->back; + __be16 port = ti->port; u8 next_idx; u8 idx; @@ -8670,7 +8652,7 @@ static void i40e_add_vxlan_port(struct net_device *netdev, /* Check if port already exists */ if (idx < I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - netdev_info(netdev, "vxlan port %d already offloaded\n", + netdev_info(netdev, "port %d already offloaded\n", ntohs(port)); return; } @@ -8679,131 +8661,75 @@ static void i40e_add_vxlan_port(struct net_device *netdev, next_idx = i40e_get_udp_port_idx(pf, 0); if (next_idx == I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - netdev_info(netdev, "maximum number of vxlan UDP ports reached, not adding port %d\n", - ntohs(port)); - return; - } - - /* New port: add it and mark its index in the bitmap */ - pf->udp_ports[next_idx].index = port; - pf->udp_ports[next_idx].type = I40E_AQC_TUNNEL_TYPE_VXLAN; - pf->pending_udp_bitmap |= BIT_ULL(next_idx); - pf->flags |= I40E_FLAG_UDP_FILTER_SYNC; -} - -/** - * i40e_del_vxlan_port - Get notifications about VXLAN ports that go away - * @netdev: This physical port's netdev - * @sa_family: Socket Family that VXLAN is notifying us about - * @port: UDP port number that VXLAN stopped listening to - **/ -static void i40e_del_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) -{ - struct i40e_netdev_priv *np = netdev_priv(netdev); - struct i40e_vsi *vsi = np->vsi; - struct i40e_pf *pf = vsi->back; - u8 idx; - - idx = i40e_get_udp_port_idx(pf, port); - - /* Check if port already exists */ - if (idx < I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - /* if port exists, set it to 0 (mark for deletion) - * and make it pending - */ - pf->udp_ports[idx].index = 0; - pf->pending_udp_bitmap |= BIT_ULL(idx); - pf->flags |= I40E_FLAG_UDP_FILTER_SYNC; - } else { - netdev_warn(netdev, "vxlan port %d was not found, not deleting\n", - ntohs(port)); - } -} -#endif - -#if IS_ENABLED(CONFIG_GENEVE) -/** - * i40e_add_geneve_port - Get notifications about GENEVE ports that come up - * @netdev: This physical port's netdev - * @sa_family: Socket Family that GENEVE is notifying us about - * @port: New UDP port number that GENEVE started listening to - **/ -static void i40e_add_geneve_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) -{ - struct i40e_netdev_priv *np = netdev_priv(netdev); - struct i40e_vsi *vsi = np->vsi; - struct i40e_pf *pf = vsi->back; - u8 next_idx; - u8 idx; - - if (!(pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE)) - return; - - idx = i40e_get_udp_port_idx(pf, port); - - /* Check if port already exists */ - if (idx < I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - netdev_info(netdev, "udp port %d already offloaded\n", + netdev_info(netdev, "maximum number of offloaded UDP ports reached, not adding port %d\n", ntohs(port)); return; } - /* Now check if there is space to add the new port */ - next_idx = i40e_get_udp_port_idx(pf, 0); - - if (next_idx == I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - netdev_info(netdev, "maximum number of UDP ports reached, not adding port %d\n", - ntohs(port)); + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + pf->udp_ports[next_idx].type = I40E_AQC_TUNNEL_TYPE_VXLAN; + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (!(pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE)) + return; + pf->udp_ports[next_idx].type = I40E_AQC_TUNNEL_TYPE_NGE; + break; + default: return; } /* New port: add it and mark its index in the bitmap */ pf->udp_ports[next_idx].index = port; - pf->udp_ports[next_idx].type = I40E_AQC_TUNNEL_TYPE_NGE; pf->pending_udp_bitmap |= BIT_ULL(next_idx); pf->flags |= I40E_FLAG_UDP_FILTER_SYNC; - - dev_info(&pf->pdev->dev, "adding geneve port %d\n", ntohs(port)); } /** - * i40e_del_geneve_port - Get notifications about GENEVE ports that go away + * i40e_udp_tunnel_del - Get notifications about UDP tunnel ports that go away * @netdev: This physical port's netdev - * @sa_family: Socket Family that GENEVE is notifying us about - * @port: UDP port number that GENEVE stopped listening to + * @ti: Tunnel endpoint information **/ -static void i40e_del_geneve_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) +static void i40e_udp_tunnel_del(struct net_device *netdev, + struct udp_tunnel_info *ti) { struct i40e_netdev_priv *np = netdev_priv(netdev); struct i40e_vsi *vsi = np->vsi; struct i40e_pf *pf = vsi->back; + __be16 port = ti->port; u8 idx; - if (!(pf->flags & I40E_FLAG_GENEVE_OFFLOAD_CAPABLE)) - return; - idx = i40e_get_udp_port_idx(pf, port); /* Check if port already exists */ - if (idx < I40E_MAX_PF_UDP_OFFLOAD_PORTS) { - /* if port exists, set it to 0 (mark for deletion) - * and make it pending - */ - pf->udp_ports[idx].index = 0; - pf->pending_udp_bitmap |= BIT_ULL(idx); - pf->flags |= I40E_FLAG_UDP_FILTER_SYNC; + if (idx >= I40E_MAX_PF_UDP_OFFLOAD_PORTS) + goto not_found; - dev_info(&pf->pdev->dev, "deleting geneve port %d\n", - ntohs(port)); - } else { - netdev_warn(netdev, "geneve port %d was not found, not deleting\n", - ntohs(port)); + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (pf->udp_ports[idx].type != I40E_AQC_TUNNEL_TYPE_VXLAN) + goto not_found; + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (pf->udp_ports[idx].type != I40E_AQC_TUNNEL_TYPE_NGE) + goto not_found; + break; + default: + goto not_found; } + + /* if port exists, set it to 0 (mark for deletion) + * and make it pending + */ + pf->udp_ports[idx].index = 0; + pf->pending_udp_bitmap |= BIT_ULL(idx); + pf->flags |= I40E_FLAG_UDP_FILTER_SYNC; + + return; +not_found: + netdev_warn(netdev, "UDP port %d was not found, not deleting\n", + ntohs(port)); } -#endif static int i40e_get_phys_port_id(struct net_device *netdev, struct netdev_phys_item_id *ppid) @@ -9033,14 +8959,8 @@ static const struct net_device_ops i40e_netdev_ops = { .ndo_set_vf_link_state = i40e_ndo_set_vf_link_state, .ndo_set_vf_spoofchk = i40e_ndo_set_vf_spoofchk, .ndo_set_vf_trust = i40e_ndo_set_vf_trust, -#if IS_ENABLED(CONFIG_VXLAN) - .ndo_add_vxlan_port = i40e_add_vxlan_port, - .ndo_del_vxlan_port = i40e_del_vxlan_port, -#endif -#if IS_ENABLED(CONFIG_GENEVE) - .ndo_add_geneve_port = i40e_add_geneve_port, - .ndo_del_geneve_port = i40e_del_geneve_port, -#endif + .ndo_udp_tunnel_add = i40e_udp_tunnel_add, + .ndo_udp_tunnel_del = i40e_udp_tunnel_del, .ndo_get_phys_port_id = i40e_get_phys_port_id, .ndo_fdb_add = i40e_ndo_fdb_add, .ndo_features_check = i40e_features_check, @@ -10689,12 +10609,8 @@ static void i40e_print_features(struct i40e_pf *pf) } if (pf->flags & I40E_FLAG_DCB_CAPABLE) i += snprintf(&buf[i], REMAIN(i), " DCB"); -#if IS_ENABLED(CONFIG_VXLAN) i += snprintf(&buf[i], REMAIN(i), " VxLAN"); -#endif -#if IS_ENABLED(CONFIG_GENEVE) i += snprintf(&buf[i], REMAIN(i), " Geneve"); -#endif if (pf->flags & I40E_FLAG_PTP) i += snprintf(&buf[i], REMAIN(i), " PTP"); #ifdef I40E_FCOE -- cgit v0.10.2 From b3a49557d53108eea959347e5a67b695729c2779 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:22:19 -0700 Subject: ixgbe: Replace ndo_add/del_vxlan_port with ndo_add/del_udp_enc_port This change replaces the network device operations for adding or removing a VXLAN port with operations that are more generically defined to be used for any UDP offload port but provide a type. As such by just adding a line to verify that the offload type is VXLAN we can maintain the same functionality. In addition I updated the socket address family check so that instead of excluding IPv6 we instead abort of type is not IPv4. This makes much more sense as we should only be supporting IPv4 outer addresses on this hardware. The last change is that I pulled the rtnl_lock/unlock into the conditional statement for IXGBE_FLAG2_VXLAN_REREG_NEEDED. The motivation behind this is to avoid unneeded bouncing of the mutex which will just slow down the handling of this call anyway. Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig index 1fbf908..c0e1743 100644 --- a/drivers/net/ethernet/intel/Kconfig +++ b/drivers/net/ethernet/intel/Kconfig @@ -167,17 +167,6 @@ config IXGBE To compile this driver as a module, choose M here. The module will be called ixgbe. -config IXGBE_VXLAN - bool "Virtual eXtensible Local Area Network Support" - default n - depends on IXGBE && VXLAN && !(IXGBE=y && VXLAN=m) - ---help--- - This allows one to create VXLAN virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. VXLAN is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to use Virtual eXtensible Local Area Network - (VXLAN) in the driver. - config IXGBE_HWMON bool "Intel(R) 10GbE PCI Express adapters HWMON support" default y diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 088c47c..468fa9d 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -50,7 +50,7 @@ #include #include #include -#include +#include #include #include #include @@ -5722,9 +5722,7 @@ static int ixgbe_sw_init(struct ixgbe_adapter *adapter) #ifdef CONFIG_IXGBE_DCA adapter->flags &= ~IXGBE_FLAG_DCA_CAPABLE; #endif -#ifdef CONFIG_IXGBE_VXLAN adapter->flags |= IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE; -#endif break; default: break; @@ -6158,9 +6156,7 @@ int ixgbe_open(struct net_device *netdev) ixgbe_up_complete(adapter); ixgbe_clear_vxlan_port(adapter); -#ifdef CONFIG_IXGBE_VXLAN - vxlan_get_rx_port(netdev); -#endif + udp_tunnel_get_rx_info(netdev); return 0; @@ -7262,14 +7258,12 @@ static void ixgbe_service_task(struct work_struct *work) ixgbe_service_event_complete(adapter); return; } -#ifdef CONFIG_IXGBE_VXLAN - rtnl_lock(); if (adapter->flags2 & IXGBE_FLAG2_VXLAN_REREG_NEEDED) { + rtnl_lock(); adapter->flags2 &= ~IXGBE_FLAG2_VXLAN_REREG_NEEDED; - vxlan_get_rx_port(adapter->netdev); + udp_tunnel_get_rx_info(adapter->netdev); + rtnl_unlock(); } - rtnl_unlock(); -#endif /* CONFIG_IXGBE_VXLAN */ ixgbe_reset_subtask(adapter); ixgbe_phy_interrupt_subtask(adapter); ixgbe_sfp_detection_subtask(adapter); @@ -7697,7 +7691,6 @@ static void ixgbe_atr(struct ixgbe_ring *ring, /* snag network header to get L4 type and address */ skb = first->skb; hdr.network = skb_network_header(skb); -#ifdef CONFIG_IXGBE_VXLAN if (skb->encapsulation && first->protocol == htons(ETH_P_IP) && hdr.ipv4->protocol != IPPROTO_UDP) { @@ -7708,7 +7701,6 @@ static void ixgbe_atr(struct ixgbe_ring *ring, udp_hdr(skb)->dest == adapter->vxlan_port) hdr.network = skb_inner_network_header(skb); } -#endif /* CONFIG_IXGBE_VXLAN */ /* Currently only IPv4/IPv6 with TCP is supported */ switch (hdr.ipv4->version) { @@ -8770,14 +8762,12 @@ static int ixgbe_set_features(struct net_device *netdev, netdev->features = features; -#ifdef CONFIG_IXGBE_VXLAN if ((adapter->flags & IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE)) { if (features & NETIF_F_RXCSUM) adapter->flags2 |= IXGBE_FLAG2_VXLAN_REREG_NEEDED; else ixgbe_clear_vxlan_port(adapter); } -#endif /* CONFIG_IXGBE_VXLAN */ if (need_reset) ixgbe_do_reset(netdev); @@ -8788,23 +8778,27 @@ static int ixgbe_set_features(struct net_device *netdev, return 0; } -#ifdef CONFIG_IXGBE_VXLAN /** * ixgbe_add_vxlan_port - Get notifications about VXLAN ports that come up * @dev: The port's netdev * @sa_family: Socket Family that VXLAN is notifiying us about * @port: New UDP port number that VXLAN started listening to + * @type: Enumerated type specifying UDP tunnel type **/ -static void ixgbe_add_vxlan_port(struct net_device *dev, sa_family_t sa_family, - __be16 port) +static void ixgbe_add_vxlan_port(struct net_device *dev, + struct udp_tunnel_info *ti) { struct ixgbe_adapter *adapter = netdev_priv(dev); struct ixgbe_hw *hw = &adapter->hw; + __be16 port = ti->port; - if (!(adapter->flags & IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE)) + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) return; - if (sa_family == AF_INET6) + if (ti->sa_family != AF_INET) + return; + + if (!(adapter->flags & IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE)) return; if (adapter->vxlan_port == port) @@ -8826,28 +8820,31 @@ static void ixgbe_add_vxlan_port(struct net_device *dev, sa_family_t sa_family, * @dev: The port's netdev * @sa_family: Socket Family that VXLAN is notifying us about * @port: UDP port number that VXLAN stopped listening to + * @type: Enumerated type specifying UDP tunnel type **/ -static void ixgbe_del_vxlan_port(struct net_device *dev, sa_family_t sa_family, - __be16 port) +static void ixgbe_del_vxlan_port(struct net_device *dev, + struct udp_tunnel_info *ti) { struct ixgbe_adapter *adapter = netdev_priv(dev); - if (!(adapter->flags & IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE)) + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) return; - if (sa_family == AF_INET6) + if (ti->sa_family != AF_INET) return; - if (adapter->vxlan_port != port) { + if (!(adapter->flags & IXGBE_FLAG_VXLAN_OFFLOAD_CAPABLE)) + return; + + if (adapter->vxlan_port != ti->port) { netdev_info(dev, "Port %d was not found, not deleting\n", - ntohs(port)); + ntohs(ti->port)); return; } ixgbe_clear_vxlan_port(adapter); adapter->flags2 |= IXGBE_FLAG2_VXLAN_REREG_NEEDED; } -#endif /* CONFIG_IXGBE_VXLAN */ static int ixgbe_ndo_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], struct net_device *dev, @@ -9160,10 +9157,8 @@ static const struct net_device_ops ixgbe_netdev_ops = { .ndo_bridge_getlink = ixgbe_ndo_bridge_getlink, .ndo_dfwd_add_station = ixgbe_fwd_add, .ndo_dfwd_del_station = ixgbe_fwd_del, -#ifdef CONFIG_IXGBE_VXLAN - .ndo_add_vxlan_port = ixgbe_add_vxlan_port, - .ndo_del_vxlan_port = ixgbe_del_vxlan_port, -#endif /* CONFIG_IXGBE_VXLAN */ + .ndo_udp_tunnel_add = ixgbe_add_vxlan_port, + .ndo_udp_tunnel_del = ixgbe_del_vxlan_port, .ndo_features_check = ixgbe_features_check, }; -- cgit v0.10.2 From a831274a1346913c145797ddee6f39e30e061318 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:22:30 -0700 Subject: mlx4_en: Replace ndo_add/del_vxlan_port with ndo_add/del_udp_enc_port This change replaces the network device operations for adding or removing a VXLAN port with operations that are more generically defined to be used for any UDP offload port but provide a type. As such by just adding a line to verify that the offload type is VXLAN we can maintain the same functionality. In addition I updated the socket address family check so that instead of excluding IPv6 we instead abort of type is not IPv4. This makes much more sense as we should only be supporting IPv4 outer addresses on this hardware. Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/Kconfig b/drivers/net/ethernet/mellanox/mlx4/Kconfig index 9ca3734..5098e7f 100644 --- a/drivers/net/ethernet/mellanox/mlx4/Kconfig +++ b/drivers/net/ethernet/mellanox/mlx4/Kconfig @@ -24,13 +24,6 @@ config MLX4_EN_DCB If unsure, set to Y -config MLX4_EN_VXLAN - bool "VXLAN offloads Support" - default y - depends on MLX4_EN && VXLAN && !(MLX4_EN=y && VXLAN=m) - ---help--- - Say Y here if you want to use VXLAN offloads in the driver. - config MLX4_CORE tristate depends on PCI diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 973391b..8e318d2 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -1692,10 +1692,9 @@ int mlx4_en_start_port(struct net_device *dev) /* Schedule multicast task to populate multicast list */ queue_work(mdev->workqueue, &priv->rx_mode_task); -#ifdef CONFIG_MLX4_EN_VXLAN if (priv->mdev->dev->caps.tunnel_offload_mode == MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) - vxlan_get_rx_port(dev); -#endif + udp_tunnel_get_rx_info(dev); + priv->port_up = true; netif_tx_start_all_queues(dev); netif_device_attach(dev); @@ -2342,7 +2341,6 @@ static int mlx4_en_get_phys_port_id(struct net_device *dev, return 0; } -#ifdef CONFIG_MLX4_EN_VXLAN static void mlx4_en_add_vxlan_offloads(struct work_struct *work) { int ret; @@ -2392,15 +2390,19 @@ static void mlx4_en_del_vxlan_offloads(struct work_struct *work) } static void mlx4_en_add_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct mlx4_en_priv *priv = netdev_priv(dev); + __be16 port = ti->port; __be16 current_port; - if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) return; - if (sa_family == AF_INET6) + if (ti->sa_family != AF_INET) + return; + + if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) return; current_port = priv->vxlan_port; @@ -2415,15 +2417,19 @@ static void mlx4_en_add_vxlan_port(struct net_device *dev, } static void mlx4_en_del_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct mlx4_en_priv *priv = netdev_priv(dev); + __be16 port = ti->port; __be16 current_port; - if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) return; - if (sa_family == AF_INET6) + if (ti->sa_family != AF_INET) + return; + + if (priv->mdev->dev->caps.tunnel_offload_mode != MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) return; current_port = priv->vxlan_port; @@ -2453,7 +2459,6 @@ static netdev_features_t mlx4_en_features_check(struct sk_buff *skb, return features; } -#endif static int mlx4_en_set_tx_maxrate(struct net_device *dev, int queue_index, u32 maxrate) { @@ -2506,11 +2511,9 @@ static const struct net_device_ops mlx4_netdev_ops = { .ndo_rx_flow_steer = mlx4_en_filter_rfs, #endif .ndo_get_phys_port_id = mlx4_en_get_phys_port_id, -#ifdef CONFIG_MLX4_EN_VXLAN - .ndo_add_vxlan_port = mlx4_en_add_vxlan_port, - .ndo_del_vxlan_port = mlx4_en_del_vxlan_port, + .ndo_udp_tunnel_add = mlx4_en_add_vxlan_port, + .ndo_udp_tunnel_del = mlx4_en_del_vxlan_port, .ndo_features_check = mlx4_en_features_check, -#endif .ndo_set_tx_maxrate = mlx4_en_set_tx_maxrate, }; @@ -2544,11 +2547,9 @@ static const struct net_device_ops mlx4_netdev_ops_master = { .ndo_rx_flow_steer = mlx4_en_filter_rfs, #endif .ndo_get_phys_port_id = mlx4_en_get_phys_port_id, -#ifdef CONFIG_MLX4_EN_VXLAN - .ndo_add_vxlan_port = mlx4_en_add_vxlan_port, - .ndo_del_vxlan_port = mlx4_en_del_vxlan_port, + .ndo_udp_tunnel_add = mlx4_en_add_vxlan_port, + .ndo_udp_tunnel_del = mlx4_en_del_vxlan_port, .ndo_features_check = mlx4_en_features_check, -#endif .ndo_set_tx_maxrate = mlx4_en_set_tx_maxrate, }; @@ -2839,10 +2840,8 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, INIT_WORK(&priv->linkstate_task, mlx4_en_linkstate); INIT_DELAYED_WORK(&priv->stats_task, mlx4_en_do_get_stats); INIT_DELAYED_WORK(&priv->service_task, mlx4_en_service_task); -#ifdef CONFIG_MLX4_EN_VXLAN INIT_WORK(&priv->vxlan_add_task, mlx4_en_add_vxlan_offloads); INIT_WORK(&priv->vxlan_del_task, mlx4_en_del_vxlan_offloads); -#endif #ifdef CONFIG_RFS_ACCEL INIT_LIST_HEAD(&priv->filters); spin_lock_init(&priv->filters_lock); diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 467d47e..6b3b0fe 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -545,10 +545,8 @@ struct mlx4_en_priv { struct work_struct linkstate_task; struct delayed_work stats_task; struct delayed_work service_task; -#ifdef CONFIG_MLX4_EN_VXLAN struct work_struct vxlan_add_task; struct work_struct vxlan_del_task; -#endif struct mlx4_en_perf_stats pstats; struct mlx4_en_pkt_stats pkstats; struct mlx4_en_counter_stats pf_stats; -- cgit v0.10.2 From 974c3f3000d85a3504ce294bcae6472ff9040770 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:22:38 -0700 Subject: mlx5_en: Replace ndo_add/del_vxlan_port with ndo_add/del_udp_enc_port This change replaces the network device operations for adding or removing a VXLAN port with operations that are more generically defined to be used for any UDP offload port but provide a type. As such by just adding a line to verify that the offload type is VXLAN we can maintain the same functionality. Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index f5c8d5d..8b7c6f3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -2520,25 +2520,31 @@ static int mlx5e_get_vf_stats(struct net_device *dev, } static void mlx5e_add_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct mlx5e_priv *priv = netdev_priv(netdev); + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + if (!mlx5e_vxlan_allowed(priv->mdev)) return; - mlx5e_vxlan_queue_work(priv, sa_family, be16_to_cpu(port), 1); + mlx5e_vxlan_queue_work(priv, ti->sa_family, be16_to_cpu(ti->port), 1); } static void mlx5e_del_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct mlx5e_priv *priv = netdev_priv(netdev); + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + if (!mlx5e_vxlan_allowed(priv->mdev)) return; - mlx5e_vxlan_queue_work(priv, sa_family, be16_to_cpu(port), 0); + mlx5e_vxlan_queue_work(priv, ti->sa_family, be16_to_cpu(ti->port), 0); } static netdev_features_t mlx5e_vxlan_features_check(struct mlx5e_priv *priv, @@ -2624,8 +2630,8 @@ static const struct net_device_ops mlx5e_netdev_ops_sriov = { .ndo_set_features = mlx5e_set_features, .ndo_change_mtu = mlx5e_change_mtu, .ndo_do_ioctl = mlx5e_ioctl, - .ndo_add_vxlan_port = mlx5e_add_vxlan_port, - .ndo_del_vxlan_port = mlx5e_del_vxlan_port, + .ndo_udp_tunnel_add = mlx5e_add_vxlan_port, + .ndo_udp_tunnel_del = mlx5e_del_vxlan_port, .ndo_features_check = mlx5e_features_check, #ifdef CONFIG_RFS_ACCEL .ndo_rx_flow_steer = mlx5e_rx_flow_steer, @@ -3128,7 +3134,7 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev) if (mlx5e_vxlan_allowed(mdev)) { rtnl_lock(); - vxlan_get_rx_port(netdev); + udp_tunnel_get_rx_info(netdev); rtnl_unlock(); } -- cgit v0.10.2 From 3ab68837062894529db2dc80789659547aaf0773 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:22:51 -0700 Subject: nfp: Replace ndo_add/del_vxlan_port with ndo_add/del_udp_enc_port This change replaces the network device operations for adding or removing a VXLAN port with operations that are more generically defined to be used for any UDP offload port but provide a type. As such by just adding a line to verify that the offload type is VXLAN we can maintain the same functionality. Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index fa47c14..2195ed3 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -1979,7 +1979,7 @@ static int __nfp_net_set_config_and_enable(struct nfp_net *nn) if (nn->ctrl & NFP_NET_CFG_CTRL_VXLAN) { memset(&nn->vxlan_ports, 0, sizeof(nn->vxlan_ports)); memset(&nn->vxlan_usecnt, 0, sizeof(nn->vxlan_usecnt)); - vxlan_get_rx_port(nn->netdev); + udp_tunnel_get_rx_info(nn->netdev); } return err; @@ -2551,26 +2551,32 @@ static int nfp_net_find_vxlan_idx(struct nfp_net *nn, __be16 port) } static void nfp_net_add_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct nfp_net *nn = netdev_priv(netdev); int idx; - idx = nfp_net_find_vxlan_idx(nn, port); + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + idx = nfp_net_find_vxlan_idx(nn, ti->port); if (idx == -ENOSPC) return; if (!nn->vxlan_usecnt[idx]++) - nfp_net_set_vxlan_port(nn, idx, port); + nfp_net_set_vxlan_port(nn, idx, ti->port); } static void nfp_net_del_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct nfp_net *nn = netdev_priv(netdev); int idx; - idx = nfp_net_find_vxlan_idx(nn, port); + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + idx = nfp_net_find_vxlan_idx(nn, ti->port); if (!nn->vxlan_usecnt[idx] || idx == -ENOSPC) return; @@ -2589,8 +2595,8 @@ static const struct net_device_ops nfp_net_netdev_ops = { .ndo_set_mac_address = eth_mac_addr, .ndo_set_features = nfp_net_set_features, .ndo_features_check = nfp_net_features_check, - .ndo_add_vxlan_port = nfp_net_add_vxlan_port, - .ndo_del_vxlan_port = nfp_net_del_vxlan_port, + .ndo_udp_tunnel_add = nfp_net_add_vxlan_port, + .ndo_udp_tunnel_del = nfp_net_del_vxlan_port, }; /** -- cgit v0.10.2 From f9f082a9b948e02742511f626080eb4d84886512 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:22:57 -0700 Subject: qede: Move all UDP port notifiers to single function This patch goes through and combines the notifiers for VXLAN and GENEVE into a single function for each action. So there is now one combined function for getting ports, one for adding the ports, and one for deleting the ports. Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/Kconfig b/drivers/net/ethernet/qlogic/Kconfig index 680d8c7..613dd28 100644 --- a/drivers/net/ethernet/qlogic/Kconfig +++ b/drivers/net/ethernet/qlogic/Kconfig @@ -114,24 +114,4 @@ config QEDE ---help--- This enables the support for ... -config QEDE_VXLAN - bool "Virtual eXtensible Local Area Network support" - default n - depends on QEDE && VXLAN && !(QEDE=y && VXLAN=m) - ---help--- - This enables hardware offload support for VXLAN protocol over - qede module. Say Y here if you want to enable hardware offload - support for Virtual eXtensible Local Area Network (VXLAN) - in the driver. - -config QEDE_GENEVE - bool "Generic Network Virtualization Encapsulation (GENEVE) support" - depends on QEDE && GENEVE && !(QEDE=y && GENEVE=m) - ---help--- - This allows one to create GENEVE virtual interfaces that provide - Layer 2 Networks over Layer 3 Networks. GENEVE is often used - to tunnel virtual network infrastructure in virtualized environments. - Say Y here if you want to enable hardware offload support for - Generic Network Virtualization Encapsulation (GENEVE) in the driver. - endif # NET_VENDOR_QLOGIC diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 423168b..2972742 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -24,12 +24,7 @@ #include #include #include -#ifdef CONFIG_QEDE_VXLAN -#include -#endif -#ifdef CONFIG_QEDE_GENEVE -#include -#endif +#include #include #include #include @@ -2112,75 +2107,75 @@ int qede_set_features(struct net_device *dev, netdev_features_t features) return 0; } -#ifdef CONFIG_QEDE_VXLAN -static void qede_add_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) +static void qede_udp_tunnel_add(struct net_device *dev, + struct udp_tunnel_info *ti) { struct qede_dev *edev = netdev_priv(dev); - u16 t_port = ntohs(port); + u16 t_port = ntohs(ti->port); - if (edev->vxlan_dst_port) - return; + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (edev->vxlan_dst_port) + return; - edev->vxlan_dst_port = t_port; + edev->vxlan_dst_port = t_port; - DP_VERBOSE(edev, QED_MSG_DEBUG, "Added vxlan port=%d", t_port); + DP_VERBOSE(edev, QED_MSG_DEBUG, "Added vxlan port=%d", + t_port); - set_bit(QEDE_SP_VXLAN_PORT_CONFIG, &edev->sp_flags); - schedule_delayed_work(&edev->sp_task, 0); -} + set_bit(QEDE_SP_VXLAN_PORT_CONFIG, &edev->sp_flags); + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (edev->geneve_dst_port) + return; -static void qede_del_vxlan_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) -{ - struct qede_dev *edev = netdev_priv(dev); - u16 t_port = ntohs(port); + edev->geneve_dst_port = t_port; - if (t_port != edev->vxlan_dst_port) + DP_VERBOSE(edev, QED_MSG_DEBUG, "Added geneve port=%d", + t_port); + set_bit(QEDE_SP_GENEVE_PORT_CONFIG, &edev->sp_flags); + break; + default: return; + } - edev->vxlan_dst_port = 0; - - DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted vxlan port=%d", t_port); - - set_bit(QEDE_SP_VXLAN_PORT_CONFIG, &edev->sp_flags); schedule_delayed_work(&edev->sp_task, 0); } -#endif -#ifdef CONFIG_QEDE_GENEVE -static void qede_add_geneve_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) +static void qede_udp_tunnel_del(struct net_device *dev, + struct udp_tunnel_info *ti) { struct qede_dev *edev = netdev_priv(dev); - u16 t_port = ntohs(port); + u16 t_port = ntohs(ti->port); - if (edev->geneve_dst_port) - return; + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + if (t_port != edev->vxlan_dst_port) + return; - edev->geneve_dst_port = t_port; + edev->vxlan_dst_port = 0; - DP_VERBOSE(edev, QED_MSG_DEBUG, "Added geneve port=%d", t_port); - set_bit(QEDE_SP_GENEVE_PORT_CONFIG, &edev->sp_flags); - schedule_delayed_work(&edev->sp_task, 0); -} + DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted vxlan port=%d", + t_port); -static void qede_del_geneve_port(struct net_device *dev, - sa_family_t sa_family, __be16 port) -{ - struct qede_dev *edev = netdev_priv(dev); - u16 t_port = ntohs(port); + set_bit(QEDE_SP_VXLAN_PORT_CONFIG, &edev->sp_flags); + break; + case UDP_TUNNEL_TYPE_GENEVE: + if (t_port != edev->geneve_dst_port) + return; - if (t_port != edev->geneve_dst_port) - return; + edev->geneve_dst_port = 0; - edev->geneve_dst_port = 0; + DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted geneve port=%d", + t_port); + set_bit(QEDE_SP_GENEVE_PORT_CONFIG, &edev->sp_flags); + break; + default: + return; + } - DP_VERBOSE(edev, QED_MSG_DEBUG, "Deleted geneve port=%d", t_port); - set_bit(QEDE_SP_GENEVE_PORT_CONFIG, &edev->sp_flags); schedule_delayed_work(&edev->sp_task, 0); } -#endif static const struct net_device_ops qede_netdev_ops = { .ndo_open = qede_open, @@ -2204,14 +2199,8 @@ static const struct net_device_ops qede_netdev_ops = { .ndo_get_vf_config = qede_get_vf_config, .ndo_set_vf_rate = qede_set_vf_rate, #endif -#ifdef CONFIG_QEDE_VXLAN - .ndo_add_vxlan_port = qede_add_vxlan_port, - .ndo_del_vxlan_port = qede_del_vxlan_port, -#endif -#ifdef CONFIG_QEDE_GENEVE - .ndo_add_geneve_port = qede_add_geneve_port, - .ndo_del_geneve_port = qede_del_geneve_port, -#endif + .ndo_udp_tunnel_add = qede_udp_tunnel_add, + .ndo_udp_tunnel_del = qede_udp_tunnel_del, }; /* ------------------------------------------------------------------------- @@ -3579,12 +3568,8 @@ static int qede_open(struct net_device *ndev) if (rc) return rc; -#ifdef CONFIG_QEDE_VXLAN - vxlan_get_rx_port(ndev); -#endif -#ifdef CONFIG_QEDE_GENEVE - geneve_get_rx_port(ndev); -#endif + udp_tunnel_get_rx_info(ndev); + return 0; } -- cgit v0.10.2 From 5e44f8e299f64c7b5ce8c9185d14f6f98ff30b21 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:23:04 -0700 Subject: qlcnic: Replace ndo_add/del_vxlan_port with ndo_add/del_udp_enc_port This change replaces the network device operations for adding or removing a VXLAN port with operations that are more generically defined to be used for any UDP offload port but provide a type. As such by just adding a line to verify that the offload type is VXLAN we can maintain the same functionality. Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/Kconfig b/drivers/net/ethernet/qlogic/Kconfig index 613dd28..6ba4840 100644 --- a/drivers/net/ethernet/qlogic/Kconfig +++ b/drivers/net/ethernet/qlogic/Kconfig @@ -54,16 +54,6 @@ config QLCNIC_DCB mode of DCB is supported. PG and PFC values are related only to Tx. -config QLCNIC_VXLAN - bool "Virtual eXtensible Local Area Network (VXLAN) offload support" - default n - depends on QLCNIC && VXLAN && !(QLCNIC=y && VXLAN=m) - ---help--- - This enables hardware offload support for VXLAN protocol over QLogic's - 84XX series adapters. - Say Y here if you want to enable hardware offload support for - Virtual eXtensible Local Area Network (VXLAN) in the driver. - config QLCNIC_HWMON bool "QLOGIC QLCNIC 82XX and 83XX family HWMON support" depends on QLCNIC && HWMON && !(QLCNIC=y && HWMON=m) diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h index caf6ddb..fd973f4 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h @@ -1026,10 +1026,8 @@ struct qlcnic_ipaddr { #define QLCNIC_HAS_PHYS_PORT_ID 0x40000 #define QLCNIC_TSS_RSS 0x80000 -#ifdef CONFIG_QLCNIC_VXLAN #define QLCNIC_ADD_VXLAN_PORT 0x100000 #define QLCNIC_DEL_VXLAN_PORT 0x200000 -#endif #define QLCNIC_VLAN_FILTERING 0x800000 diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c index bf89216..a496390 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c @@ -1020,7 +1020,6 @@ static int qlcnic_83xx_idc_check_state_validity(struct qlcnic_adapter *adapter, return 0; } -#ifdef CONFIG_QLCNIC_VXLAN #define QLC_83XX_ENCAP_TYPE_VXLAN BIT_1 #define QLC_83XX_MATCH_ENCAP_ID BIT_2 #define QLC_83XX_SET_VXLAN_UDP_DPORT BIT_3 @@ -1089,14 +1088,12 @@ static int qlcnic_set_vxlan_parsing(struct qlcnic_adapter *adapter, return ret; } -#endif static void qlcnic_83xx_periodic_tasks(struct qlcnic_adapter *adapter) { if (adapter->fhash.fnum) qlcnic_prune_lb_filters(adapter); -#ifdef CONFIG_QLCNIC_VXLAN if (adapter->flags & QLCNIC_ADD_VXLAN_PORT) { if (qlcnic_set_vxlan_port(adapter)) return; @@ -1112,7 +1109,6 @@ static void qlcnic_83xx_periodic_tasks(struct qlcnic_adapter *adapter) adapter->ahw->vxlan_port = 0; adapter->flags &= ~QLCNIC_DEL_VXLAN_PORT; } -#endif } /** diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c index 1c29105..3ebef27 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c @@ -16,9 +16,7 @@ #include #include #include -#ifdef CONFIG_QLCNIC_VXLAN #include -#endif #include "qlcnic.h" #include "qlcnic_sriov.h" @@ -474,13 +472,15 @@ static int qlcnic_get_phys_port_id(struct net_device *netdev, return 0; } -#ifdef CONFIG_QLCNIC_VXLAN static void qlcnic_add_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct qlcnic_adapter *adapter = netdev_priv(netdev); struct qlcnic_hardware_context *ahw = adapter->ahw; + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + /* Adapter supports only one VXLAN port. Use very first port * for enabling offload */ @@ -488,23 +488,26 @@ static void qlcnic_add_vxlan_port(struct net_device *netdev, return; if (!ahw->vxlan_port_count) { ahw->vxlan_port_count = 1; - ahw->vxlan_port = ntohs(port); + ahw->vxlan_port = ntohs(ti->port); adapter->flags |= QLCNIC_ADD_VXLAN_PORT; return; } - if (ahw->vxlan_port == ntohs(port)) + if (ahw->vxlan_port == ntohs(ti->port)) ahw->vxlan_port_count++; } static void qlcnic_del_vxlan_port(struct net_device *netdev, - sa_family_t sa_family, __be16 port) + struct udp_tunnel_info *ti) { struct qlcnic_adapter *adapter = netdev_priv(netdev); struct qlcnic_hardware_context *ahw = adapter->ahw; + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + if (!qlcnic_encap_rx_offload(adapter) || !ahw->vxlan_port_count || - (ahw->vxlan_port != ntohs(port))) + (ahw->vxlan_port != ntohs(ti->port))) return; ahw->vxlan_port_count--; @@ -519,7 +522,6 @@ static netdev_features_t qlcnic_features_check(struct sk_buff *skb, features = vlan_features_check(skb, features); return vxlan_features_check(skb, features); } -#endif static const struct net_device_ops qlcnic_netdev_ops = { .ndo_open = qlcnic_open, @@ -539,11 +541,9 @@ static const struct net_device_ops qlcnic_netdev_ops = { .ndo_fdb_del = qlcnic_fdb_del, .ndo_fdb_dump = qlcnic_fdb_dump, .ndo_get_phys_port_id = qlcnic_get_phys_port_id, -#ifdef CONFIG_QLCNIC_VXLAN - .ndo_add_vxlan_port = qlcnic_add_vxlan_port, - .ndo_del_vxlan_port = qlcnic_del_vxlan_port, + .ndo_udp_tunnel_add = qlcnic_add_vxlan_port, + .ndo_udp_tunnel_del = qlcnic_del_vxlan_port, .ndo_features_check = qlcnic_features_check, -#endif #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = qlcnic_poll_controller, #endif @@ -2015,10 +2015,8 @@ qlcnic_attach(struct qlcnic_adapter *adapter) qlcnic_create_sysfs_entries(adapter); -#ifdef CONFIG_QLCNIC_VXLAN if (qlcnic_encap_rx_offload(adapter)) - vxlan_get_rx_port(netdev); -#endif + udp_tunnel_get_rx_info(netdev); adapter->is_up = QLCNIC_ADAPTER_UP_MAGIC; return 0; -- cgit v0.10.2 From 1938ee1fd3de74d761a60806b048df652666afec Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:23:12 -0700 Subject: net: Remove deprecated tunnel specific UDP offload functions Now that we have all the drivers using udp_tunnel_get_rx_ports, ndo_add_udp_enc_rx_port, and ndo_del_udp_enc_rx_port we can drop the function calls that were specific to VXLAN and GENEVE. Signed-off-by: Alexander Duyck Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 577d2a1..e84d9d2 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1026,32 +1026,6 @@ struct tc_to_netdev { * not implement this, it is assumed that the hw is not able to have * multiple net devices on single physical port. * - * void (*ndo_add_vxlan_port)(struct net_device *dev, - * sa_family_t sa_family, __be16 port); - * Called by vxlan to notify a driver about the UDP port and socket - * address family that vxlan is listening to. It is called only when - * a new port starts listening. The operation is protected by the - * vxlan_net->sock_lock. - * - * void (*ndo_add_geneve_port)(struct net_device *dev, - * sa_family_t sa_family, __be16 port); - * Called by geneve to notify a driver about the UDP port and socket - * address family that geneve is listnening to. It is called only when - * a new port starts listening. The operation is protected by the - * geneve_net->sock_lock. - * - * void (*ndo_del_geneve_port)(struct net_device *dev, - * sa_family_t sa_family, __be16 port); - * Called by geneve to notify the driver about a UDP port and socket - * address family that geneve is not listening to anymore. The operation - * is protected by the geneve_net->sock_lock. - * - * void (*ndo_del_vxlan_port)(struct net_device *dev, - * sa_family_t sa_family, __be16 port); - * Called by vxlan to notify the driver about a UDP port and socket - * address family that vxlan is not listening to anymore. The operation - * is protected by the vxlan_net->sock_lock. - * * void (*ndo_udp_tunnel_add)(struct net_device *dev, * struct udp_tunnel_info *ti); * Called by UDP tunnel to notify a driver about the UDP port and socket @@ -1272,18 +1246,6 @@ struct net_device_ops { struct netdev_phys_item_id *ppid); int (*ndo_get_phys_port_name)(struct net_device *dev, char *name, size_t len); - void (*ndo_add_vxlan_port)(struct net_device *dev, - sa_family_t sa_family, - __be16 port); - void (*ndo_del_vxlan_port)(struct net_device *dev, - sa_family_t sa_family, - __be16 port); - void (*ndo_add_geneve_port)(struct net_device *dev, - sa_family_t sa_family, - __be16 port); - void (*ndo_del_geneve_port)(struct net_device *dev, - sa_family_t sa_family, - __be16 port); void (*ndo_udp_tunnel_add)(struct net_device *dev, struct udp_tunnel_info *ti); void (*ndo_udp_tunnel_del)(struct net_device *dev, diff --git a/include/net/geneve.h b/include/net/geneve.h index 3410c4b..ec0327d 100644 --- a/include/net/geneve.h +++ b/include/net/geneve.h @@ -59,11 +59,6 @@ struct genevehdr { struct geneve_opt options[]; }; -static inline void geneve_get_rx_port(struct net_device *netdev) -{ - udp_tunnel_get_rx_info(netdev); -} - #ifdef CONFIG_INET struct net_device *geneve_dev_create_fb(struct net *net, const char *name, u8 name_assign_type, u16 dst_port); diff --git a/include/net/vxlan.h b/include/net/vxlan.h index c62e2ed..b96d036 100644 --- a/include/net/vxlan.h +++ b/include/net/vxlan.h @@ -389,11 +389,6 @@ static inline __be32 vxlan_compute_rco(unsigned int start, unsigned int offset) return vni_field; } -static inline void vxlan_get_rx_port(struct net_device *netdev) -{ - udp_tunnel_get_rx_info(netdev); -} - static inline unsigned short vxlan_get_sk_family(struct vxlan_sock *vs) { return vs->sock->sk->sk_family; diff --git a/net/ipv4/udp_tunnel.c b/net/ipv4/udp_tunnel.c index 683e494..58bd39f 100644 --- a/net/ipv4/udp_tunnel.c +++ b/net/ipv4/udp_tunnel.c @@ -76,47 +76,20 @@ void setup_udp_tunnel_sock(struct net *net, struct socket *sock, } EXPORT_SYMBOL_GPL(setup_udp_tunnel_sock); -static void __udp_tunnel_push_rx_port(struct net_device *dev, - struct udp_tunnel_info *ti) -{ - if (dev->netdev_ops->ndo_udp_tunnel_add) { - dev->netdev_ops->ndo_udp_tunnel_add(dev, ti); - return; - } - - switch (ti->type) { - case UDP_TUNNEL_TYPE_VXLAN: - if (!dev->netdev_ops->ndo_add_vxlan_port) - break; - - dev->netdev_ops->ndo_add_vxlan_port(dev, - ti->sa_family, - ti->port); - break; - case UDP_TUNNEL_TYPE_GENEVE: - if (!dev->netdev_ops->ndo_add_geneve_port) - break; - - dev->netdev_ops->ndo_add_geneve_port(dev, - ti->sa_family, - ti->port); - break; - default: - break; - } -} - void udp_tunnel_push_rx_port(struct net_device *dev, struct socket *sock, unsigned short type) { struct sock *sk = sock->sk; struct udp_tunnel_info ti; + if (!dev->netdev_ops->ndo_udp_tunnel_add) + return; + ti.type = type; ti.sa_family = sk->sk_family; ti.port = inet_sk(sk)->inet_sport; - __udp_tunnel_push_rx_port(dev, &ti); + dev->netdev_ops->ndo_udp_tunnel_add(dev, &ti); } EXPORT_SYMBOL_GPL(udp_tunnel_push_rx_port); @@ -133,42 +106,15 @@ void udp_tunnel_notify_add_rx_port(struct socket *sock, unsigned short type) ti.port = inet_sk(sk)->inet_sport; rcu_read_lock(); - for_each_netdev_rcu(net, dev) - __udp_tunnel_push_rx_port(dev, &ti); + for_each_netdev_rcu(net, dev) { + if (!dev->netdev_ops->ndo_udp_tunnel_add) + continue; + dev->netdev_ops->ndo_udp_tunnel_add(dev, &ti); + } rcu_read_unlock(); } EXPORT_SYMBOL_GPL(udp_tunnel_notify_add_rx_port); -static void __udp_tunnel_pull_rx_port(struct net_device *dev, - struct udp_tunnel_info *ti) -{ - if (dev->netdev_ops->ndo_udp_tunnel_del) { - dev->netdev_ops->ndo_udp_tunnel_del(dev, ti); - return; - } - - switch (ti->type) { - case UDP_TUNNEL_TYPE_VXLAN: - if (!dev->netdev_ops->ndo_del_vxlan_port) - break; - - dev->netdev_ops->ndo_del_vxlan_port(dev, - ti->sa_family, - ti->port); - break; - case UDP_TUNNEL_TYPE_GENEVE: - if (!dev->netdev_ops->ndo_del_geneve_port) - break; - - dev->netdev_ops->ndo_del_geneve_port(dev, - ti->sa_family, - ti->port); - break; - default: - break; - } -} - /* Notify netdevs that UDP port is no more listening */ void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type) { @@ -182,8 +128,11 @@ void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type) ti.port = inet_sk(sk)->inet_sport; rcu_read_lock(); - for_each_netdev_rcu(net, dev) - __udp_tunnel_pull_rx_port(dev, &ti); + for_each_netdev_rcu(net, dev) { + if (!dev->netdev_ops->ndo_udp_tunnel_del) + continue; + dev->netdev_ops->ndo_udp_tunnel_del(dev, &ti); + } rcu_read_unlock(); } EXPORT_SYMBOL_GPL(udp_tunnel_notify_del_rx_port); -- cgit v0.10.2 From b9adcd69bd7b41625201686b4cfec7ff13357afc Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:23:19 -0700 Subject: vxlan: Add new UDP encapsulation offload type for VXLAN-GPE The fact is VXLAN with Generic Protocol Extensions cannot be supported by the same hardware parsers that support VXLAN. The protocol extensions allow for things like a Next Protocol field which in turn allows for things other than Ethernet to be passed over the tunnel. Most existing parsers will not know how to interpret this. To resolve this I am giving VXLAN-GPE its own UDP encapsulation offload type. This way hardware that does support GPE can simply add this type to the switch statement for VXLAN, and if they don't support it then this will fix any issues where headers might be interpreted incorrectly. Signed-off-by: Alexander Duyck Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index 31aeec9..abb9cd2 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -998,6 +998,8 @@ static bool __vxlan_sock_release_prep(struct vxlan_sock *vs) spin_lock(&vn->sock_lock); hlist_del_rcu(&vs->hlist); udp_tunnel_notify_del_rx_port(vs->sock, + (vs->flags & VXLAN_F_GPE) ? + UDP_TUNNEL_TYPE_VXLAN_GPE : UDP_TUNNEL_TYPE_VXLAN); spin_unlock(&vn->sock_lock); @@ -2488,6 +2490,8 @@ static void vxlan_push_rx_ports(struct net_device *dev) for (i = 0; i < PORT_HASH_SIZE; ++i) { hlist_for_each_entry_rcu(vs, &vn->sock_list[i], hlist) udp_tunnel_push_rx_port(dev, vs->sock, + (vs->flags & VXLAN_F_GPE) ? + UDP_TUNNEL_TYPE_VXLAN_GPE : UDP_TUNNEL_TYPE_VXLAN); } spin_unlock(&vn->sock_lock); @@ -2691,6 +2695,8 @@ static struct vxlan_sock *vxlan_socket_create(struct net *net, bool ipv6, spin_lock(&vn->sock_lock); hlist_add_head_rcu(&vs->hlist, vs_head(net, port)); udp_tunnel_notify_add_rx_port(sock, + (vs->flags & VXLAN_F_GPE) ? + UDP_TUNNEL_TYPE_VXLAN_GPE : UDP_TUNNEL_TYPE_VXLAN); spin_unlock(&vn->sock_lock); diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h index 1c9408a..02c5be0 100644 --- a/include/net/udp_tunnel.h +++ b/include/net/udp_tunnel.h @@ -103,6 +103,7 @@ void setup_udp_tunnel_sock(struct net *net, struct socket *sock, enum udp_parsable_tunnel_type { UDP_TUNNEL_TYPE_VXLAN, /* RFC 7348 */ UDP_TUNNEL_TYPE_GENEVE, /* draft-ietf-nvo3-geneve */ + UDP_TUNNEL_TYPE_VXLAN_GPE, /* draft-ietf-nvo3-vxlan-gpe */ }; struct udp_tunnel_info { -- cgit v0.10.2 From 0023a061d75a37f50fb6952015b826d86de8d2c4 Mon Sep 17 00:00:00 2001 From: Sudip Mukherjee Date: Thu, 16 Jun 2016 22:19:31 +0100 Subject: net: lantiq_etop: remove unused variable The variable i was declared but was never used and we were getting a build warning for that. Signed-off-by: Sudip Mukherjee Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c index dc82b1b..0d2f8e9 100644 --- a/drivers/net/ethernet/lantiq_etop.c +++ b/drivers/net/ethernet/lantiq_etop.c @@ -411,7 +411,6 @@ static int ltq_etop_mdio_init(struct net_device *dev) { struct ltq_etop_priv *priv = netdev_priv(dev); - int i; int err; priv->mii_bus = mdiobus_alloc(); -- cgit v0.10.2 From a2e2ff560f5113e8ca31432fbd985f5f1889efdc Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 16 Jun 2016 16:24:24 -0700 Subject: net: ipv6: Move ip6_route_get_saddr to inline VRF driver needs access to ip6_route_get_saddr code. Since it does little beyond ipv6_dev_get_saddr and ipv6_dev_get_saddr is already exported for modules move ip6_route_get_saddr to the header as an inline. Code move only; no functional change. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index f55bf3d..d97305d 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -18,6 +18,7 @@ struct route_info { __u8 prefix[0]; /* 0,8 or 16 */ }; +#include #include #include #include @@ -88,9 +89,23 @@ int ip6_route_add(struct fib6_config *cfg); int ip6_ins_rt(struct rt6_info *); int ip6_del_rt(struct rt6_info *); -int ip6_route_get_saddr(struct net *net, struct rt6_info *rt, - const struct in6_addr *daddr, unsigned int prefs, - struct in6_addr *saddr); +static inline int ip6_route_get_saddr(struct net *net, struct rt6_info *rt, + const struct in6_addr *daddr, + unsigned int prefs, + struct in6_addr *saddr) +{ + struct inet6_dev *idev = + rt ? ip6_dst_idev((struct dst_entry *)rt) : NULL; + int err = 0; + + if (rt && rt->rt6i_prefsrc.plen) + *saddr = rt->rt6i_prefsrc.addr; + else + err = ipv6_dev_get_saddr(net, idev ? idev->dev : NULL, + daddr, prefs, saddr); + + return err; +} struct rt6_info *rt6_lookup(struct net *net, const struct in6_addr *daddr, const struct in6_addr *saddr, int oif, int flags); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 9e15167..08b77f4 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2586,23 +2586,6 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, return rt; } -int ip6_route_get_saddr(struct net *net, - struct rt6_info *rt, - const struct in6_addr *daddr, - unsigned int prefs, - struct in6_addr *saddr) -{ - struct inet6_dev *idev = - rt ? ip6_dst_idev((struct dst_entry *)rt) : NULL; - int err = 0; - if (rt && rt->rt6i_prefsrc.plen) - *saddr = rt->rt6i_prefsrc.addr; - else - err = ipv6_dev_get_saddr(net, idev ? idev->dev : NULL, - daddr, prefs, saddr); - return err; -} - /* remove deleted ip from prefsrc entries */ struct arg_dev_net_ip { struct net_device *dev; -- cgit v0.10.2 From 0d240e7811c4ec1965760ee4643b5bbc9cfacbb3 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 16 Jun 2016 16:24:25 -0700 Subject: net: vrf: Implement get_saddr for IPv6 IPv6 source address selection needs to consider the real egress route. Similar to IPv4 implement a get_saddr6 method which is called if source address has not been set. The get_saddr6 method does a full lookup which means pulling a route from the VRF FIB table and properly considering linklocal/multicast destination addresses. Lookup failures (eg., unreachable) then cause the source address selection to fail which gets propagated back to the caller. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 32173aa..b376282 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -999,6 +999,46 @@ static struct dst_entry *vrf_get_rt6_dst(const struct net_device *dev, return dst; } + +/* called under rcu_read_lock */ +static int vrf_get_saddr6(struct net_device *dev, const struct sock *sk, + struct flowi6 *fl6) +{ + struct net *net = dev_net(dev); + struct dst_entry *dst; + struct rt6_info *rt; + int err; + + if (rt6_need_strict(&fl6->daddr)) { + rt = vrf_ip6_route_lookup(net, dev, fl6, fl6->flowi6_oif, + RT6_LOOKUP_F_IFACE); + if (unlikely(!rt)) + return 0; + + dst = &rt->dst; + } else { + __u8 flags = fl6->flowi6_flags; + + fl6->flowi6_flags |= FLOWI_FLAG_L3MDEV_SRC; + fl6->flowi6_flags |= FLOWI_FLAG_SKIP_NH_OIF; + + dst = ip6_route_output(net, sk, fl6); + rt = (struct rt6_info *)dst; + + fl6->flowi6_flags = flags; + } + + err = dst->error; + if (!err) { + err = ip6_route_get_saddr(net, rt, &fl6->daddr, + sk ? inet6_sk(sk)->srcprefs : 0, + &fl6->saddr); + } + + dst_release(dst); + + return err; +} #endif static const struct l3mdev_ops vrf_l3mdev_ops = { @@ -1008,6 +1048,7 @@ static const struct l3mdev_ops vrf_l3mdev_ops = { .l3mdev_l3_rcv = vrf_l3_rcv, #if IS_ENABLED(CONFIG_IPV6) .l3mdev_get_rt6_dst = vrf_get_rt6_dst, + .l3mdev_get_saddr6 = vrf_get_saddr6, #endif }; diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h index f8a416e..818fd4f 100644 --- a/include/net/l3mdev.h +++ b/include/net/l3mdev.h @@ -39,6 +39,9 @@ struct l3mdev_ops { /* IPv6 ops */ struct dst_entry * (*l3mdev_get_rt6_dst)(const struct net_device *dev, struct flowi6 *fl6); + int (*l3mdev_get_saddr6)(struct net_device *dev, + const struct sock *sk, + struct flowi6 *fl6); }; #ifdef CONFIG_NET_L3_MASTER_DEV @@ -140,6 +143,8 @@ static inline bool netif_index_is_l3_master(struct net *net, int ifindex) int l3mdev_get_saddr(struct net *net, int ifindex, struct flowi4 *fl4); struct dst_entry *l3mdev_get_rt6_dst(struct net *net, struct flowi6 *fl6); +int l3mdev_get_saddr6(struct net *net, const struct sock *sk, + struct flowi6 *fl6); static inline struct sk_buff *l3mdev_l3_rcv(struct sk_buff *skb, u16 proto) @@ -230,6 +235,12 @@ struct dst_entry *l3mdev_get_rt6_dst(struct net *net, struct flowi6 *fl6) return NULL; } +static inline int l3mdev_get_saddr6(struct net *net, const struct sock *sk, + struct flowi6 *fl6) +{ + return 0; +} + static inline struct sk_buff *l3mdev_ip_rcv(struct sk_buff *skb) { diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index fd32175..1dfc402 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -910,6 +910,13 @@ static int ip6_dst_lookup_tail(struct net *net, const struct sock *sk, int err; int flags = 0; + if (ipv6_addr_any(&fl6->saddr) && fl6->flowi6_oif && + (!*dst || !(*dst)->error)) { + err = l3mdev_get_saddr6(net, sk, fl6); + if (err) + goto out_err; + } + /* The correct way to handle this would be to do * ip6_route_get_saddr, and then ip6_route_output; however, * the route-specific preferred source forces the @@ -999,10 +1006,11 @@ static int ip6_dst_lookup_tail(struct net *net, const struct sock *sk, return 0; out_err_release: - if (err == -ENETUNREACH) - IP6_INC_STATS(net, NULL, IPSTATS_MIB_OUTNOROUTES); dst_release(*dst); *dst = NULL; +out_err: + if (err == -ENETUNREACH) + IP6_INC_STATS(net, NULL, IPSTATS_MIB_OUTNOROUTES); return err; } diff --git a/net/l3mdev/l3mdev.c b/net/l3mdev/l3mdev.c index d90e4ef..c4a1c3e 100644 --- a/net/l3mdev/l3mdev.c +++ b/net/l3mdev/l3mdev.c @@ -162,6 +162,30 @@ int l3mdev_get_saddr(struct net *net, int ifindex, struct flowi4 *fl4) } EXPORT_SYMBOL_GPL(l3mdev_get_saddr); +int l3mdev_get_saddr6(struct net *net, const struct sock *sk, + struct flowi6 *fl6) +{ + struct net_device *dev; + int rc = 0; + + if (fl6->flowi6_oif) { + rcu_read_lock(); + + dev = dev_get_by_index_rcu(net, fl6->flowi6_oif); + if (dev && netif_is_l3_slave(dev)) + dev = netdev_master_upper_dev_get_rcu(dev); + + if (dev && netif_is_l3_master(dev) && + dev->l3mdev_ops->l3mdev_get_saddr6) + rc = dev->l3mdev_ops->l3mdev_get_saddr6(dev, sk, fl6); + + rcu_read_unlock(); + } + + return rc; +} +EXPORT_SYMBOL_GPL(l3mdev_get_saddr6); + /** * l3mdev_fib_rule_match - Determine if flowi references an * L3 master device -- cgit v0.10.2 From afbac6010aec514998214fb19a1f37732b7a1d77 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 16 Jun 2016 16:24:26 -0700 Subject: net: ipv6: Address selection needs to consider L3 domains IPv6 version of 3f2fb9a834cb ("net: l3mdev: address selection should only consider devices in L3 domain") and the follow up commit, a17b693cdd876 ("net: l3mdev: prefer VRF master for source address selection"). That is, if outbound device is given then the address preference order is an address from that device, an address from the master device if it is enslaved, and then an address from a device in the same L3 domain. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h index 818fd4f..e900950 100644 --- a/include/net/l3mdev.h +++ b/include/net/l3mdev.h @@ -79,6 +79,31 @@ static inline int l3mdev_master_ifindex_by_index(struct net *net, int ifindex) return rc; } +static inline +const struct net_device *l3mdev_master_dev_rcu(const struct net_device *_dev) +{ + /* netdev_master_upper_dev_get_rcu calls + * list_first_or_null_rcu to walk the upper dev list. + * list_first_or_null_rcu does not handle a const arg. We aren't + * making changes, just want the master device from that list so + * typecast to remove the const + */ + struct net_device *dev = (struct net_device *)_dev; + const struct net_device *master; + + if (!dev) + return NULL; + + if (netif_is_l3_master(dev)) + master = dev; + else if (netif_is_l3_slave(dev)) + master = netdev_master_upper_dev_get_rcu(dev); + else + master = NULL; + + return master; +} + /* get index of an interface to use for FIB lookups. For devices * enslaved to an L3 master device FIB lookups are based on the * master index @@ -190,6 +215,12 @@ static inline int l3mdev_master_ifindex_by_index(struct net *net, int ifindex) return 0; } +static inline +const struct net_device *l3mdev_master_dev_rcu(const struct net_device *dev) +{ + return NULL; +} + static inline int l3mdev_fib_oif_rcu(struct net_device *dev) { return dev ? dev->ifindex : 0; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 6c8fc3f..a1f6b7b 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1524,6 +1524,28 @@ out: return hiscore_idx; } +static int ipv6_get_saddr_master(struct net *net, + const struct net_device *dst_dev, + const struct net_device *master, + struct ipv6_saddr_dst *dst, + struct ipv6_saddr_score *scores, + int hiscore_idx) +{ + struct inet6_dev *idev; + + idev = __in6_dev_get(dst_dev); + if (idev) + hiscore_idx = __ipv6_dev_get_saddr(net, dst, idev, + scores, hiscore_idx); + + idev = __in6_dev_get(master); + if (idev) + hiscore_idx = __ipv6_dev_get_saddr(net, dst, idev, + scores, hiscore_idx); + + return hiscore_idx; +} + int ipv6_dev_get_saddr(struct net *net, const struct net_device *dst_dev, const struct in6_addr *daddr, unsigned int prefs, struct in6_addr *saddr) @@ -1577,13 +1599,39 @@ int ipv6_dev_get_saddr(struct net *net, const struct net_device *dst_dev, if (idev) hiscore_idx = __ipv6_dev_get_saddr(net, &dst, idev, scores, hiscore_idx); } else { + const struct net_device *master; + int master_idx = 0; + + /* if dst_dev exists and is enslaved to an L3 device, then + * prefer addresses from dst_dev and then the master over + * any other enslaved devices in the L3 domain. + */ + master = l3mdev_master_dev_rcu(dst_dev); + if (master) { + master_idx = master->ifindex; + + hiscore_idx = ipv6_get_saddr_master(net, dst_dev, + master, &dst, + scores, hiscore_idx); + + if (scores[hiscore_idx].ifa) + goto out; + } + for_each_netdev_rcu(net, dev) { + /* only consider addresses on devices in the + * same L3 domain + */ + if (l3mdev_master_ifindex_rcu(dev) != master_idx) + continue; idev = __in6_dev_get(dev); if (!idev) continue; hiscore_idx = __ipv6_dev_get_saddr(net, &dst, idev, scores, hiscore_idx); } } + +out: rcu_read_unlock(); hiscore = &scores[hiscore_idx]; -- cgit v0.10.2 From 0350cb48fb94e168d8b4d3ff65adbdbc73759cbf Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 17 Jun 2016 12:22:26 +0300 Subject: tipc: potential shift wrapping bug in map_set() "up_map" is a u64 type but we're not using the high 32 bits. Fixes: 35c55c9877f8 ('tipc: add neighbor monitoring framework') Signed-off-by: Dan Carpenter Signed-off-by: David S. Miller diff --git a/net/tipc/monitor.c b/net/tipc/monitor.c index 87d4efe..0d489e8 100644 --- a/net/tipc/monitor.c +++ b/net/tipc/monitor.c @@ -122,8 +122,8 @@ static int dom_size(int peers) static void map_set(u64 *up_map, int i, unsigned int v) { - *up_map &= ~(1 << i); - *up_map |= (v << i); + *up_map &= ~(1ULL << i); + *up_map |= ((u64)v << i); } static int map_get(u64 up_map, int i) -- cgit v0.10.2 From 1793331e0943f1ddd8649dd1ccea11f3f267d371 Mon Sep 17 00:00:00 2001 From: Ivan Khoronzhuk Date: Fri, 17 Jun 2016 13:25:39 +0300 Subject: net: ethernet: ti: cpsw: remove rx_descs property There is no reason in rx_descs property because davinici_cpdma driver splits pool of descriptors equally between tx and rx channels. That is, if number of descriptors 256, 128 of them are for rx channels. While receiving, the descriptor is freed to the pool and then allocated with new skb. And if in DT the "rx_descs" is set to 64, then 128 - 64 = 64 descriptors are always in the pool and cannot be used, for tx, for instance. It's not correct resource usage, better to set it to half of pool, then the rx pool can be used in full. It will not have any impact on performance, as anyway, the "redundant" descriptors were unused. Signed-off-by: Ivan Khoronzhuk Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index f2a4cd6..8327328 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1277,6 +1277,7 @@ static int cpsw_ndo_open(struct net_device *ndev) if (!cpsw_common_res_usage_state(priv)) { struct cpsw_priv *priv_sl0 = cpsw_get_slave_priv(priv, 0); + int buf_num; /* setup tx dma to fixed prio and zero offset */ cpdma_control_set(priv->dma, CPDMA_TX_PRIO_FIXED, 1); @@ -1304,10 +1305,8 @@ static int cpsw_ndo_open(struct net_device *ndev) enable_irq(priv->irqs_table[0]); } - if (WARN_ON(!priv->data.rx_descs)) - priv->data.rx_descs = 128; - - for (i = 0; i < priv->data.rx_descs; i++) { + buf_num = cpdma_chan_get_rx_buf_num(priv->dma); + for (i = 0; i < buf_num; i++) { struct sk_buff *skb; ret = -ENOMEM; @@ -1998,12 +1997,6 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data, } data->bd_ram_size = prop; - if (of_property_read_u32(node, "rx_descs", &prop)) { - dev_err(&pdev->dev, "Missing rx_descs property in the DT.\n"); - return -EINVAL; - } - data->rx_descs = prop; - if (of_property_read_u32(node, "mac_control", &prop)) { dev_err(&pdev->dev, "Missing mac_control property in the DT.\n"); return -EINVAL; diff --git a/drivers/net/ethernet/ti/cpsw.h b/drivers/net/ethernet/ti/cpsw.h index e50afd1..16b54c6 100644 --- a/drivers/net/ethernet/ti/cpsw.h +++ b/drivers/net/ethernet/ti/cpsw.h @@ -35,7 +35,6 @@ struct cpsw_platform_data { u32 cpts_clock_shift; /* convert input clock ticks to nanoseconds */ u32 ale_entries; /* ale table size */ u32 bd_ram_size; /*buffer descriptor ram size */ - u32 rx_descs; /* Number of Rx Descriptios */ u32 mac_control; /* Mac control register */ u16 default_vlan; /* Def VLAN for ALE lookup in VLAN aware mode*/ bool dual_emac; /* Enable Dual EMAC mode */ diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c index 18bf3a8..bcd9e45 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.c +++ b/drivers/net/ethernet/ti/davinci_cpdma.c @@ -543,6 +543,12 @@ struct cpdma_chan *cpdma_chan_create(struct cpdma_ctlr *ctlr, int chan_num, } EXPORT_SYMBOL_GPL(cpdma_chan_create); +int cpdma_chan_get_rx_buf_num(struct cpdma_ctlr *ctlr) +{ + return ctlr->pool->num_desc / 2; +} +EXPORT_SYMBOL_GPL(cpdma_chan_get_rx_buf_num); + int cpdma_chan_destroy(struct cpdma_chan *chan) { struct cpdma_ctlr *ctlr; diff --git a/drivers/net/ethernet/ti/davinci_cpdma.h b/drivers/net/ethernet/ti/davinci_cpdma.h index 86dee48..80c015c 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.h +++ b/drivers/net/ethernet/ti/davinci_cpdma.h @@ -81,6 +81,7 @@ int cpdma_ctlr_dump(struct cpdma_ctlr *ctlr); struct cpdma_chan *cpdma_chan_create(struct cpdma_ctlr *ctlr, int chan_num, cpdma_handler_fn handler); +int cpdma_chan_get_rx_buf_num(struct cpdma_ctlr *ctlr); int cpdma_chan_destroy(struct cpdma_chan *chan); int cpdma_chan_start(struct cpdma_chan *chan); int cpdma_chan_stop(struct cpdma_chan *chan); -- cgit v0.10.2 From 6774b68b2435a129850542f9b7a3b9ce375291c6 Mon Sep 17 00:00:00 2001 From: Ivan Khoronzhuk Date: Fri, 17 Jun 2016 13:25:40 +0300 Subject: Documentation: DT: cpsw: remove rx_descs property There is no reason to hold s/w dependent parameter in device tree. Even more, there is no reason in this parameter because davinici_cpdma driver splits pool of descriptors equally between tx and rx channels anyway. Acked-by: Rob Herring Signed-off-by: Ivan Khoronzhuk Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/cpsw.txt b/Documentation/devicetree/bindings/net/cpsw.txt index 0ae0649..5ad439f 100644 --- a/Documentation/devicetree/bindings/net/cpsw.txt +++ b/Documentation/devicetree/bindings/net/cpsw.txt @@ -15,7 +15,6 @@ Required properties: - cpdma_channels : Specifies number of channels in CPDMA - ale_entries : Specifies No of entries ALE can hold - bd_ram_size : Specifies internal descriptor RAM size -- rx_descs : Specifies number of Rx descriptors - mac_control : Specifies Default MAC control register content for the specific platform - slaves : Specifies number for slaves diff --git a/arch/arm/boot/dts/am33xx.dtsi b/arch/arm/boot/dts/am33xx.dtsi index 52be48b..702126f 100644 --- a/arch/arm/boot/dts/am33xx.dtsi +++ b/arch/arm/boot/dts/am33xx.dtsi @@ -766,7 +766,6 @@ ale_entries = <1024>; bd_ram_size = <0x2000>; no_bd_ram = <0>; - rx_descs = <64>; mac_control = <0x20>; slaves = <2>; active_slave = <0>; diff --git a/arch/arm/boot/dts/am4372.dtsi b/arch/arm/boot/dts/am4372.dtsi index 12fcde4..a10fa7f 100644 --- a/arch/arm/boot/dts/am4372.dtsi +++ b/arch/arm/boot/dts/am4372.dtsi @@ -626,7 +626,6 @@ ale_entries = <1024>; bd_ram_size = <0x2000>; no_bd_ram = <0>; - rx_descs = <64>; mac_control = <0x20>; slaves = <2>; active_slave = <0>; diff --git a/arch/arm/boot/dts/dm814x.dtsi b/arch/arm/boot/dts/dm814x.dtsi index d4537dc..f23cae0c 100644 --- a/arch/arm/boot/dts/dm814x.dtsi +++ b/arch/arm/boot/dts/dm814x.dtsi @@ -509,7 +509,6 @@ ale_entries = <1024>; bd_ram_size = <0x2000>; no_bd_ram = <0>; - rx_descs = <64>; mac_control = <0x20>; slaves = <2>; active_slave = <0>; diff --git a/arch/arm/boot/dts/dra7.dtsi b/arch/arm/boot/dts/dra7.dtsi index e007401..b7ddc64 100644 --- a/arch/arm/boot/dts/dra7.dtsi +++ b/arch/arm/boot/dts/dra7.dtsi @@ -1626,7 +1626,6 @@ ale_entries = <1024>; bd_ram_size = <0x2000>; no_bd_ram = <0>; - rx_descs = <64>; mac_control = <0x20>; slaves = <2>; active_slave = <0>; -- cgit v0.10.2 From c3f3cbfbf339213a2cf1d64907571b668342e25a Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Fri, 17 Jun 2016 17:33:30 +0000 Subject: gtp: remove unused including Remove including that don't need it. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c index 4e976a0..97e0cbc 100644 --- a/drivers/net/gtp.c +++ b/drivers/net/gtp.c @@ -16,7 +16,6 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include -#include #include #include #include -- cgit v0.10.2 From 2dce5fbfc0fb8941f1df6d08cfd3e749fd7116eb Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Fri, 17 Jun 2016 17:49:33 +0000 Subject: net:liquidio: remove unused including Remove including that don't need it. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index 3290009..7b44b5c 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -19,7 +19,6 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include #include #include #include -- cgit v0.10.2 From af73e72dccaf33d4ae2e9ccf48eeb527c5f24e1a Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Fri, 17 Jun 2016 18:12:46 +0000 Subject: RDS: TCP: Fix non static symbol warnings Fixes the following sparse warnings: net/rds/tcp.c:59:5: warning: symbol 'rds_tcp_min_sndbuf' was not declared. Should it be static? net/rds/tcp.c:60:5: warning: symbol 'rds_tcp_min_rcvbuf' was not declared. Should it be static? Signed-off-by: Wei Yongjun Acked-by: Sowmini Varadhan Acked-by: Santosh Shilimkar Signed-off-by: David S. Miller diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 0e757a0..5217d49 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -57,8 +57,8 @@ static int rds_tcp_skbuf_handler(struct ctl_table *ctl, int write, void __user *buffer, size_t *lenp, loff_t *fpos); -int rds_tcp_min_sndbuf = SOCK_MIN_SNDBUF; -int rds_tcp_min_rcvbuf = SOCK_MIN_RCVBUF; +static int rds_tcp_min_sndbuf = SOCK_MIN_SNDBUF; +static int rds_tcp_min_rcvbuf = SOCK_MIN_RCVBUF; static struct ctl_table rds_tcp_sysctl_table[] = { #define RDS_TCP_SNDBUF 0 -- cgit v0.10.2 From a0bbb9fe2e26dba9524247f5d81e46f18ba69c9d Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 17 Jun 2016 23:32:14 +0200 Subject: net: ethernet: et131x: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c index 30defe6..d194992 100644 --- a/drivers/net/ethernet/agere/et131x.c +++ b/drivers/net/ethernet/agere/et131x.c @@ -440,7 +440,6 @@ struct et131x_adapter { struct net_device *netdev; struct pci_dev *pdev; struct mii_bus *mii_bus; - struct phy_device *phydev; struct napi_struct napi; /* Flags that indicate current state of the adapter */ @@ -864,7 +863,7 @@ static void et1310_config_mac_regs2(struct et131x_adapter *adapter) { int32_t delay = 0; struct mac_regs __iomem *mac = &adapter->regs->mac; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; u32 cfg1; u32 cfg2; u32 ifctrl; @@ -1035,7 +1034,7 @@ static void et1310_setup_device_for_unicast(struct et131x_adapter *adapter) static void et1310_config_rxmac_regs(struct et131x_adapter *adapter) { struct rxmac_regs __iomem *rxmac = &adapter->regs->rxmac; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; u32 sa_lo; u32 sa_hi = 0; u32 pf_ctrl = 0; @@ -1230,7 +1229,7 @@ out: static int et131x_mii_read(struct et131x_adapter *adapter, u8 reg, u16 *value) { - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; if (!phydev) return -EIO; @@ -1311,7 +1310,7 @@ static void et1310_phy_read_mii_bit(struct et131x_adapter *adapter, static void et1310_config_flow_control(struct et131x_adapter *adapter) { - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; if (phydev->duplex == DUPLEX_HALF) { adapter->flow = FLOW_NONE; @@ -1456,7 +1455,7 @@ static int et131x_mdio_write(struct mii_bus *bus, int phy_addr, static void et1310_phy_power_switch(struct et131x_adapter *adapter, bool down) { u16 data; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; et131x_mii_read(adapter, MII_BMCR, &data); data &= ~BMCR_PDOWN; @@ -1469,7 +1468,7 @@ static void et1310_phy_power_switch(struct et131x_adapter *adapter, bool down) static void et131x_xcvr_init(struct et131x_adapter *adapter) { u16 lcr2; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; /* Set the LED behavior such that LED 1 indicates speed (off = * 10Mbits, blink = 100Mbits, on = 1000Mbits) and LED 2 indicates @@ -2111,7 +2110,7 @@ static int et131x_init_recv(struct et131x_adapter *adapter) /* et131x_set_rx_dma_timer - Set the heartbeat timer according to line rate */ static void et131x_set_rx_dma_timer(struct et131x_adapter *adapter) { - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; /* For version B silicon, we do not use the RxDMA timer for 10 and 100 * Mbits/s line rates. We do not enable and RxDMA interrupt coalescing. @@ -2426,7 +2425,7 @@ static int nic_send_packet(struct et131x_adapter *adapter, struct tcb *tcb) struct sk_buff *skb = tcb->skb; u32 nr_frags = skb_shinfo(skb)->nr_frags + 1; struct skb_frag_struct *frags = &skb_shinfo(skb)->frags[0]; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; dma_addr_t dma_addr; struct tx_ring *tx_ring = &adapter->tx_ring; @@ -2794,17 +2793,13 @@ static void et131x_handle_send_pkts(struct et131x_adapter *adapter) static int et131x_get_settings(struct net_device *netdev, struct ethtool_cmd *cmd) { - struct et131x_adapter *adapter = netdev_priv(netdev); - - return phy_ethtool_gset(adapter->phydev, cmd); + return phy_ethtool_gset(netdev->phydev, cmd); } static int et131x_set_settings(struct net_device *netdev, struct ethtool_cmd *cmd) { - struct et131x_adapter *adapter = netdev_priv(netdev); - - return phy_ethtool_sset(adapter->phydev, cmd); + return phy_ethtool_sset(netdev->phydev, cmd); } static int et131x_get_regs_len(struct net_device *netdev) @@ -3098,7 +3093,7 @@ err_out: static void et131x_error_timer_handler(unsigned long data) { struct et131x_adapter *adapter = (struct et131x_adapter *)data; - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = adapter->netdev->phydev; if (et1310_in_phy_coma(adapter)) { /* Bring the device immediately out of coma, to @@ -3168,7 +3163,7 @@ static int et131x_adapter_memory_alloc(struct et131x_adapter *adapter) static void et131x_adjust_link(struct net_device *netdev) { struct et131x_adapter *adapter = netdev_priv(netdev); - struct phy_device *phydev = adapter->phydev; + struct phy_device *phydev = netdev->phydev; if (!phydev) return; @@ -3287,7 +3282,6 @@ static int et131x_mii_probe(struct net_device *netdev) phydev->advertising = phydev->supported; phydev->autoneg = AUTONEG_ENABLE; - adapter->phydev = phydev; phy_attached_info(phydev); @@ -3323,7 +3317,7 @@ static void et131x_pci_remove(struct pci_dev *pdev) unregister_netdev(netdev); netif_napi_del(&adapter->napi); - phy_disconnect(adapter->phydev); + phy_disconnect(netdev->phydev); mdiobus_unregister(adapter->mii_bus); mdiobus_free(adapter->mii_bus); @@ -3338,20 +3332,16 @@ static void et131x_pci_remove(struct pci_dev *pdev) static void et131x_up(struct net_device *netdev) { - struct et131x_adapter *adapter = netdev_priv(netdev); - et131x_enable_txrx(netdev); - phy_start(adapter->phydev); + phy_start(netdev->phydev); } static void et131x_down(struct net_device *netdev) { - struct et131x_adapter *adapter = netdev_priv(netdev); - /* Save the timestamp for the TX watchdog, prevent a timeout */ netif_trans_update(netdev); - phy_stop(adapter->phydev); + phy_stop(netdev->phydev); et131x_disable_txrx(netdev); } @@ -3684,12 +3674,10 @@ static int et131x_close(struct net_device *netdev) static int et131x_ioctl(struct net_device *netdev, struct ifreq *reqbuf, int cmd) { - struct et131x_adapter *adapter = netdev_priv(netdev); - - if (!adapter->phydev) + if (!netdev->phydev) return -EINVAL; - return phy_mii_ioctl(adapter->phydev, reqbuf, cmd); + return phy_mii_ioctl(netdev->phydev, reqbuf, cmd); } /* et131x_set_packet_filter - Configures the Rx Packet filtering */ @@ -4073,7 +4061,7 @@ out: return rc; err_phy_disconnect: - phy_disconnect(adapter->phydev); + phy_disconnect(netdev->phydev); err_mdio_unregister: mdiobus_unregister(adapter->mii_bus); err_mdio_free: -- cgit v0.10.2 From adc01582e36fc228f6f1c82188f7501eb4c6154f Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 17 Jun 2016 23:32:15 +0200 Subject: net: ethernet: et131x: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c index d194992..cd7e2e5 100644 --- a/drivers/net/ethernet/agere/et131x.c +++ b/drivers/net/ethernet/agere/et131x.c @@ -2790,18 +2790,6 @@ static void et131x_handle_send_pkts(struct et131x_adapter *adapter) spin_unlock_irqrestore(&adapter->tcb_send_qlock, flags); } -static int et131x_get_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) -{ - return phy_ethtool_gset(netdev->phydev, cmd); -} - -static int et131x_set_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) -{ - return phy_ethtool_sset(netdev->phydev, cmd); -} - static int et131x_get_regs_len(struct net_device *netdev) { #define ET131X_REGS_LEN 256 @@ -2974,12 +2962,12 @@ static void et131x_get_drvinfo(struct net_device *netdev, } static struct ethtool_ops et131x_ethtool_ops = { - .get_settings = et131x_get_settings, - .set_settings = et131x_set_settings, .get_drvinfo = et131x_get_drvinfo, .get_regs_len = et131x_get_regs_len, .get_regs = et131x_get_regs, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /* et131x_hwaddr_init - set up the MAC Address */ -- cgit v0.10.2 From 2095b1426c9c8088f7be28d70c6d7eb432640baa Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 11:01:10 +0200 Subject: mwifiex: fix link error against sdio Calling sdio_claim_host() from the interface independent part of the mwifiex driver is not only a layering violation, but also causes a link error if MMC support is disabled, or if CONFIG_MMC=m and CONFIG_MWIFIEX=y: drivers/net/built-in.o: In function `mwifiex_fw_dpc': :(.text+0xff138): undefined reference to `sdio_claim_host' :(.text+0xff158): undefined reference to `sdio_release_host' The right way to do this is to have the sdio specific code in the sdio driver front-end, and we already have a callback pointer that we can use for this after exporting the generic fw download function from the core driver. Signed-off-by: Arnd Bergmann Fixes: 65c71efe1c59 ("mwifiex: fix racing condition when downloading firmware") Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/init.c b/drivers/net/wireless/marvell/mwifiex/init.c index 78c532f..a6d86d4 100644 --- a/drivers/net/wireless/marvell/mwifiex/init.c +++ b/drivers/net/wireless/marvell/mwifiex/init.c @@ -788,3 +788,4 @@ poll_fw: return ret; } +EXPORT_SYMBOL_GPL(mwifiex_dnld_fw); diff --git a/drivers/net/wireless/marvell/mwifiex/main.c b/drivers/net/wireless/marvell/mwifiex/main.c index 2b65334..0e280f8 100644 --- a/drivers/net/wireless/marvell/mwifiex/main.c +++ b/drivers/net/wireless/marvell/mwifiex/main.c @@ -21,7 +21,6 @@ #include "wmm.h" #include "cfg80211.h" #include "11n.h" -#include "sdio.h" #define VERSION "1.0" @@ -515,7 +514,6 @@ static void mwifiex_fw_dpc(const struct firmware *firmware, void *context) struct semaphore *sem = adapter->card_sem; bool init_failed = false; struct wireless_dev *wdev; - struct sdio_mmc_card *card = adapter->card; if (!firmware) { mwifiex_dbg(adapter, ERROR, @@ -531,11 +529,7 @@ static void mwifiex_fw_dpc(const struct firmware *firmware, void *context) if (adapter->if_ops.dnld_fw) { ret = adapter->if_ops.dnld_fw(adapter, &fw); } else { - if (adapter->iface_type == MWIFIEX_SDIO) - sdio_claim_host(card->func); ret = mwifiex_dnld_fw(adapter, &fw); - if (adapter->iface_type == MWIFIEX_SDIO) - sdio_release_host(card->func); } if (ret == -1) diff --git a/drivers/net/wireless/marvell/mwifiex/sdio.c b/drivers/net/wireless/marvell/mwifiex/sdio.c index 3a2267a..d3e1561 100644 --- a/drivers/net/wireless/marvell/mwifiex/sdio.c +++ b/drivers/net/wireless/marvell/mwifiex/sdio.c @@ -554,6 +554,19 @@ static int mwifiex_pm_wakeup_card_complete(struct mwifiex_adapter *adapter) return mwifiex_write_reg(adapter, CONFIGURATION_REG, 0); } +static int mwifiex_sdio_dnld_fw(struct mwifiex_adapter *adapter, + struct mwifiex_fw_image *fw) +{ + struct sdio_mmc_card *card = adapter->card; + int ret; + + sdio_claim_host(card->func); + ret = mwifiex_dnld_fw(adapter, fw); + sdio_release_host(card->func); + + return ret; +} + /* * This function is used to initialize IO ports for the * chipsets supporting SDIO new mode eg SD8897. @@ -2742,6 +2755,7 @@ static struct mwifiex_if_ops sdio_ops = { .cleanup_mpa_buf = mwifiex_cleanup_mpa_buf, .cmdrsp_complete = mwifiex_sdio_cmdrsp_complete, .event_complete = mwifiex_sdio_event_complete, + .dnld_fw = mwifiex_sdio_dnld_fw, .card_reset = mwifiex_sdio_card_reset, .reg_dump = mwifiex_sdio_reg_dump, .device_dump = mwifiex_sdio_device_dump, -- cgit v0.10.2 From b1cadc1a0949c82ff7fcb15603e3caf2d32ff9f6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Jun 2016 21:52:02 -0700 Subject: ipv6: icmp: add a force_saddr param to icmp6_send() SIT or GRE tunnels might want to translate an IPV4 address into a v4mapped one when translating ICMP to ICMPv6. This patch adds the parameter to icmp6_send() but does not change icmpv6_send() signature. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 630f453..432611a 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -14,7 +14,8 @@ static inline struct icmp6hdr *icmp6_hdr(const struct sk_buff *skb) #if IS_ENABLED(CONFIG_IPV6) extern void icmpv6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info); -typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info); +typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info, + const struct in6_addr *force_saddr); extern int inet6_register_icmp_sender(ip6_icmp_send_t *fn); extern int inet6_unregister_icmp_sender(ip6_icmp_send_t *fn); diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index e32a72f..6c57e6e 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -388,7 +388,8 @@ relookup_failed: /* * Send an ICMP message in response to a packet in error */ -static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) +static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, + const struct in6_addr *force_saddr) { struct net *net = dev_net(skb->dev); struct inet6_dev *idev = NULL; @@ -475,6 +476,8 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) memset(&fl6, 0, sizeof(fl6)); fl6.flowi6_proto = IPPROTO_ICMPV6; fl6.daddr = hdr->saddr; + if (force_saddr) + saddr = force_saddr; if (saddr) fl6.saddr = *saddr; fl6.flowi6_mark = mark; @@ -551,7 +554,7 @@ out: */ void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) { - icmp6_send(skb, ICMPV6_PARAMPROB, code, pos); + icmp6_send(skb, ICMPV6_PARAMPROB, code, pos, NULL); kfree_skb(skb); } diff --git a/net/ipv6/ip6_icmp.c b/net/ipv6/ip6_icmp.c index 14dacc5..713676f 100644 --- a/net/ipv6/ip6_icmp.c +++ b/net/ipv6/ip6_icmp.c @@ -39,7 +39,7 @@ void icmpv6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) if (!send) goto out; - send(skb, type, code, info); + send(skb, type, code, info, NULL); out: rcu_read_unlock(); } -- cgit v0.10.2 From 5fbba8ac9358f1e796c8aedcccc3487364643723 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Jun 2016 21:52:03 -0700 Subject: ip6: move ipip6_err_gen_icmpv6_unreach() We want to use this helper from GRE as well, so this is the time to move it in net/ipv6/icmp.c Also add a @nhs parameter, since SIT and GRE have different values for the header(s) to skip. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 432611a..9796481 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -18,6 +18,7 @@ typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info, const struct in6_addr *force_saddr); extern int inet6_register_icmp_sender(ip6_icmp_send_t *fn); extern int inet6_unregister_icmp_sender(ip6_icmp_send_t *fn); +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs); #else diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 6c57e6e..07bc63c 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -558,6 +558,45 @@ void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) kfree_skb(skb); } +/* Generate icmpv6 with type/code ICMPV6_DEST_UNREACH/ICMPV6_ADDR_UNREACH + * if sufficient data bytes are available + * @nhs is the size of the tunnel header(s) : + * Either an IPv4 header for SIT encap + * an IPv4 header + GRE header for GRE encap + */ +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs) +{ + struct rt6_info *rt; + struct sk_buff *skb2; + + if (!pskb_may_pull(skb, nhs + sizeof(struct ipv6hdr) + 8)) + return 1; + + skb2 = skb_clone(skb, GFP_ATOMIC); + + if (!skb2) + return 1; + + skb_dst_drop(skb2); + skb_pull(skb2, nhs); + skb_reset_network_header(skb2); + + rt = rt6_lookup(dev_net(skb->dev), &ipv6_hdr(skb2)->saddr, NULL, 0, 0); + + if (rt && rt->dst.dev) + skb2->dev = rt->dst.dev; + + icmpv6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0); + + if (rt) + ip6_rt_put(rt); + + kfree_skb(skb2); + + return 0; +} +EXPORT_SYMBOL(ip6_err_gen_icmpv6_unreach); + static void icmpv6_echo_reply(struct sk_buff *skb) { struct net *net = dev_net(skb->dev); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index d9f2bd6..78e84d6 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -479,42 +479,6 @@ static void ipip6_tunnel_uninit(struct net_device *dev) dev_put(dev); } -/* Generate icmpv6 with type/code ICMPV6_DEST_UNREACH/ICMPV6_ADDR_UNREACH - * if sufficient data bytes are available - */ -static int ipip6_err_gen_icmpv6_unreach(struct sk_buff *skb) -{ - int ihl = ((const struct iphdr *)skb->data)->ihl*4; - struct rt6_info *rt; - struct sk_buff *skb2; - - if (!pskb_may_pull(skb, ihl + sizeof(struct ipv6hdr) + 8)) - return 1; - - skb2 = skb_clone(skb, GFP_ATOMIC); - - if (!skb2) - return 1; - - skb_dst_drop(skb2); - skb_pull(skb2, ihl); - skb_reset_network_header(skb2); - - rt = rt6_lookup(dev_net(skb->dev), &ipv6_hdr(skb2)->saddr, NULL, 0, 0); - - if (rt && rt->dst.dev) - skb2->dev = rt->dst.dev; - - icmpv6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0); - - if (rt) - ip6_rt_put(rt); - - kfree_skb(skb2); - - return 0; -} - static int ipip6_err(struct sk_buff *skb, u32 info) { const struct iphdr *iph = (const struct iphdr *)skb->data; @@ -575,7 +539,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) goto out; err = 0; - if (!ipip6_err_gen_icmpv6_unreach(skb)) + if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4)) goto out; if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED) -- cgit v0.10.2 From 2d7a3b276be2d032a6c1a48ced87a474327ee3d3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Jun 2016 21:52:04 -0700 Subject: ipv6: translate ICMP_TIME_EXCEEDED to ICMPV6_TIME_EXCEED For better traceroute/mtr support for SIT and GRE tunnels, we translate IPV4 ICMP ICMP_TIME_EXCEEDED to ICMPV6_TIME_EXCEED We also have to translate the IPv4 source IP address of ICMP message to IPv6 v4mapped. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 9796481..97ae980 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -18,7 +18,7 @@ typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info, const struct in6_addr *force_saddr); extern int inet6_register_icmp_sender(ip6_icmp_send_t *fn); extern int inet6_unregister_icmp_sender(ip6_icmp_send_t *fn); -int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs); +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type); #else diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 07bc63c..867aebc 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -564,8 +564,9 @@ void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) * Either an IPv4 header for SIT encap * an IPv4 header + GRE header for GRE encap */ -int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs) +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type) { + struct in6_addr temp_saddr; struct rt6_info *rt; struct sk_buff *skb2; @@ -586,8 +587,13 @@ int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs) if (rt && rt->dst.dev) skb2->dev = rt->dst.dev; - icmpv6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0); - + ipv6_addr_set_v4mapped(ip_hdr(skb)->saddr, &temp_saddr); + if (type == ICMP_TIME_EXCEEDED) + icmp6_send(skb2, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, + 0, &temp_saddr); + else + icmp6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, + 0, &temp_saddr); if (rt) ip6_rt_put(rt); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 78e84d6..d7a3611 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -535,11 +535,11 @@ static int ipip6_err(struct sk_buff *skb, u32 info) goto out; } - if (t->parms.iph.daddr == 0) + err = 0; + if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4, type)) goto out; - err = 0; - if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4)) + if (t->parms.iph.daddr == 0) goto out; if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED) -- cgit v0.10.2 From 9b8c6d7bf2e08a7d3eb6660a2bfaf29b8b49c329 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Jun 2016 21:52:05 -0700 Subject: gre: better support for ICMP messages for gre+ipv6 ipgre_err() can call ip6_err_gen_icmpv6_unreach() for proper support of ipv4+gre+icmp+ipv6+... frames, used for example by traceroute/mtr. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index 9222678..a5e7035 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -157,6 +157,7 @@ struct tnl_ptk_info { __be16 proto; __be32 key; __be32 seq; + int hdr_len; }; #define PACKET_RCVD 0 diff --git a/net/ipv4/gre_demux.c b/net/ipv4/gre_demux.c index 4c39f4f..c4c3e43 100644 --- a/net/ipv4/gre_demux.c +++ b/net/ipv4/gre_demux.c @@ -117,6 +117,7 @@ int gre_parse_header(struct sk_buff *skb, struct tnl_ptk_info *tpi, if ((*(u8 *)options & 0xF0) != 0x40) hdr_len += 4; } + tpi->hdr_len = hdr_len; return hdr_len; } EXPORT_SYMBOL(gre_parse_header); diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 0f8ca3f..ab4cff8 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -187,6 +187,12 @@ static void ipgre_err(struct sk_buff *skb, u32 info, if (!t) return; +#if IS_ENABLED(CONFIG_IPV6) + if (tpi->proto == htons(ETH_P_IPV6) && + !ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4 + tpi->hdr_len, type)) + return; +#endif + if (t->parms.iph.daddr == 0 || ipv4_is_multicast(t->parms.iph.daddr)) return; -- cgit v0.10.2 From 20e1954fe238dbe5f8d3a979e593fe352bd703cf Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Jun 2016 21:52:06 -0700 Subject: ipv6: RFC 4884 partial support for SIT/GRE tunnels When receiving an ICMPv4 message containing extensions as defined in RFC 4884, and translating it to ICMPv6 at SIT or GRE tunnel, we need some extra manipulation in order to properly forward the extensions. This patch only takes care of Time Exceeded messages as they are the ones that typically carry information from various routers in a fabric during a traceroute session. It also avoids complex skb logic if the data_len is not a multiple of 8. RFC states : The "original datagram" field MUST contain at least 128 octets. If the original datagram did not contain 128 octets, the "original datagram" field MUST be zero padded to 128 octets. In practice routers use 128 bytes of original datagram, not more. Initial translation was added in commit ca15a078bd90 ("sit: generate icmpv6 error when receiving icmpv4 error") Signed-off-by: Eric Dumazet Cc: Oussama Ghorbel Signed-off-by: David S. Miller diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 97ae980..57086e9 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -18,7 +18,8 @@ typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info, const struct in6_addr *force_saddr); extern int inet6_register_icmp_sender(ip6_icmp_send_t *fn); extern int inet6_unregister_icmp_sender(ip6_icmp_send_t *fn); -int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type); +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type, + unsigned int data_len); #else diff --git a/include/uapi/linux/icmp.h b/include/uapi/linux/icmp.h index 16fff05..fddd9d7 100644 --- a/include/uapi/linux/icmp.h +++ b/include/uapi/linux/icmp.h @@ -79,6 +79,7 @@ struct icmphdr { __be16 __unused; __be16 mtu; } frag; + __u8 reserved[4]; } un; }; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index ab4cff8..8eec78f 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -144,6 +144,7 @@ static void ipgre_err(struct sk_buff *skb, u32 info, const struct iphdr *iph; const int type = icmp_hdr(skb)->type; const int code = icmp_hdr(skb)->code; + unsigned int data_len = 0; struct ip_tunnel *t; switch (type) { @@ -169,6 +170,7 @@ static void ipgre_err(struct sk_buff *skb, u32 info, case ICMP_TIME_EXCEEDED: if (code != ICMP_EXC_TTL) return; + data_len = icmp_hdr(skb)->un.reserved[1] * 4; /* RFC 4884 4.1 */ break; case ICMP_REDIRECT: @@ -189,7 +191,8 @@ static void ipgre_err(struct sk_buff *skb, u32 info, #if IS_ENABLED(CONFIG_IPV6) if (tpi->proto == htons(ETH_P_IPV6) && - !ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4 + tpi->hdr_len, type)) + !ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4 + tpi->hdr_len, + type, data_len)) return; #endif diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 867aebc..fd11f58 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -564,16 +564,22 @@ void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) * Either an IPv4 header for SIT encap * an IPv4 header + GRE header for GRE encap */ -int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type) +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type, + unsigned int data_len) { struct in6_addr temp_saddr; struct rt6_info *rt; struct sk_buff *skb2; + u32 info = 0; if (!pskb_may_pull(skb, nhs + sizeof(struct ipv6hdr) + 8)) return 1; - skb2 = skb_clone(skb, GFP_ATOMIC); + /* RFC 4884 (partial) support for ICMP extensions */ + if (data_len < 128 || (data_len & 7) || skb->len < data_len) + data_len = 0; + + skb2 = data_len ? skb_copy(skb, GFP_ATOMIC) : skb_clone(skb, GFP_ATOMIC); if (!skb2) return 1; @@ -588,12 +594,26 @@ int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type) skb2->dev = rt->dst.dev; ipv6_addr_set_v4mapped(ip_hdr(skb)->saddr, &temp_saddr); + + if (data_len) { + /* RFC 4884 (partial) support : + * insert 0 padding at the end, before the extensions + */ + __skb_push(skb2, nhs); + skb_reset_network_header(skb2); + memmove(skb2->data, skb2->data + nhs, data_len - nhs); + memset(skb2->data + data_len - nhs, 0, nhs); + /* RFC 4884 4.5 : Length is measured in 64-bit words, + * and stored in reserved[0] + */ + info = (data_len/8) << 24; + } if (type == ICMP_TIME_EXCEEDED) icmp6_send(skb2, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, - 0, &temp_saddr); + info, &temp_saddr); else icmp6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, - 0, &temp_saddr); + info, &temp_saddr); if (rt) ip6_rt_put(rt); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index d7a3611..cdd7146 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -484,6 +484,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) const struct iphdr *iph = (const struct iphdr *)skb->data; const int type = icmp_hdr(skb)->type; const int code = icmp_hdr(skb)->code; + unsigned int data_len = 0; struct ip_tunnel *t; int err; @@ -508,6 +509,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) case ICMP_TIME_EXCEEDED: if (code != ICMP_EXC_TTL) return 0; + data_len = icmp_hdr(skb)->un.reserved[1] * 4; /* RFC 4884 4.1 */ break; case ICMP_REDIRECT: break; @@ -536,7 +538,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) } err = 0; - if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4, type)) + if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4, type, data_len)) goto out; if (t->parms.iph.daddr == 0) -- cgit v0.10.2 From 46bb0bb094661ef6d251c1f63f6452123f6b273b Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 18 Jun 2016 23:16:54 +0200 Subject: net: ethernet: nb8800: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Acked-by: Mans Rullgard Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/aurora/nb8800.c b/drivers/net/ethernet/aurora/nb8800.c index 08a23e6..4989d31 100644 --- a/drivers/net/ethernet/aurora/nb8800.c +++ b/drivers/net/ethernet/aurora/nb8800.c @@ -631,7 +631,7 @@ static void nb8800_mac_config(struct net_device *dev) static void nb8800_pause_config(struct net_device *dev) { struct nb8800_priv *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; u32 rxcr; if (priv->pause_aneg) { @@ -664,7 +664,7 @@ static void nb8800_pause_config(struct net_device *dev) static void nb8800_link_reconfigure(struct net_device *dev) { struct nb8800_priv *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; int change = 0; if (phydev->link) { @@ -690,7 +690,7 @@ static void nb8800_link_reconfigure(struct net_device *dev) } if (change) - phy_print_status(priv->phydev); + phy_print_status(phydev); } static void nb8800_update_mac_addr(struct net_device *dev) @@ -935,9 +935,10 @@ static int nb8800_dma_stop(struct net_device *dev) static void nb8800_pause_adv(struct net_device *dev) { struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev = dev->phydev; u32 adv = 0; - if (!priv->phydev) + if (!phydev) return; if (priv->pause_rx) @@ -945,13 +946,14 @@ static void nb8800_pause_adv(struct net_device *dev) if (priv->pause_tx) adv ^= ADVERTISED_Asym_Pause; - priv->phydev->supported |= adv; - priv->phydev->advertising |= adv; + phydev->supported |= adv; + phydev->advertising |= adv; } static int nb8800_open(struct net_device *dev) { struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev; int err; /* clear any pending interrupts */ @@ -969,10 +971,10 @@ static int nb8800_open(struct net_device *dev) nb8800_mac_rx(dev, true); nb8800_mac_tx(dev, true); - priv->phydev = of_phy_connect(dev, priv->phy_node, - nb8800_link_reconfigure, 0, - priv->phy_mode); - if (!priv->phydev) + phydev = of_phy_connect(dev, priv->phy_node, + nb8800_link_reconfigure, 0, + priv->phy_mode); + if (!phydev) goto err_free_irq; nb8800_pause_adv(dev); @@ -982,7 +984,7 @@ static int nb8800_open(struct net_device *dev) netif_start_queue(dev); nb8800_start_rx(dev); - phy_start(priv->phydev); + phy_start(phydev); return 0; @@ -997,8 +999,9 @@ err_free_dma: static int nb8800_stop(struct net_device *dev) { struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev = dev->phydev; - phy_stop(priv->phydev); + phy_stop(phydev); netif_stop_queue(dev); napi_disable(&priv->napi); @@ -1007,8 +1010,7 @@ static int nb8800_stop(struct net_device *dev) nb8800_mac_rx(dev, false); nb8800_mac_tx(dev, false); - phy_disconnect(priv->phydev); - priv->phydev = NULL; + phy_disconnect(phydev); free_irq(dev->irq, dev); @@ -1019,9 +1021,7 @@ static int nb8800_stop(struct net_device *dev) static int nb8800_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct nb8800_priv *priv = netdev_priv(dev); - - return phy_mii_ioctl(priv->phydev, rq, cmd); + return phy_mii_ioctl(dev->phydev, rq, cmd); } static const struct net_device_ops nb8800_netdev_ops = { @@ -1037,32 +1037,32 @@ static const struct net_device_ops nb8800_netdev_ops = { static int nb8800_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev = dev->phydev; - if (!priv->phydev) + if (!phydev) return -ENODEV; - return phy_ethtool_gset(priv->phydev, cmd); + return phy_ethtool_gset(phydev, cmd); } static int nb8800_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev = dev->phydev; - if (!priv->phydev) + if (!phydev) return -ENODEV; - return phy_ethtool_sset(priv->phydev, cmd); + return phy_ethtool_sset(phydev, cmd); } static int nb8800_nway_reset(struct net_device *dev) { - struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev = dev->phydev; - if (!priv->phydev) + if (!phydev) return -ENODEV; - return genphy_restart_aneg(priv->phydev); + return genphy_restart_aneg(phydev); } static void nb8800_get_pauseparam(struct net_device *dev, @@ -1079,6 +1079,7 @@ static int nb8800_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *pp) { struct nb8800_priv *priv = netdev_priv(dev); + struct phy_device *phydev = dev->phydev; priv->pause_aneg = pp->autoneg; priv->pause_rx = pp->rx_pause; @@ -1088,8 +1089,8 @@ static int nb8800_set_pauseparam(struct net_device *dev, if (!priv->pause_aneg) nb8800_pause_config(dev); - else if (priv->phydev) - phy_start_aneg(priv->phydev); + else if (phydev) + phy_start_aneg(phydev); return 0; } diff --git a/drivers/net/ethernet/aurora/nb8800.h b/drivers/net/ethernet/aurora/nb8800.h index e5adbc2..6ec4a95 100644 --- a/drivers/net/ethernet/aurora/nb8800.h +++ b/drivers/net/ethernet/aurora/nb8800.h @@ -284,7 +284,6 @@ struct nb8800_priv { struct mii_bus *mii_bus; struct device_node *phy_node; - struct phy_device *phydev; /* PHY connection type from DT */ int phy_mode; -- cgit v0.10.2 From 90789322c3bb9ece244f1fc6bf62667eae9814f2 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 18 Jun 2016 23:16:55 +0200 Subject: net: ethernet: nb8800: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Acked-by: Mans Rullgard Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/aurora/nb8800.c b/drivers/net/ethernet/aurora/nb8800.c index 4989d31..dc2c35d 100644 --- a/drivers/net/ethernet/aurora/nb8800.c +++ b/drivers/net/ethernet/aurora/nb8800.c @@ -1035,26 +1035,6 @@ static const struct net_device_ops nb8800_netdev_ops = { .ndo_validate_addr = eth_validate_addr, }; -static int nb8800_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = dev->phydev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_gset(phydev, cmd); -} - -static int nb8800_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = dev->phydev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_sset(phydev, cmd); -} - static int nb8800_nway_reset(struct net_device *dev) { struct phy_device *phydev = dev->phydev; @@ -1183,8 +1163,6 @@ static void nb8800_get_ethtool_stats(struct net_device *dev, } static const struct ethtool_ops nb8800_ethtool_ops = { - .get_settings = nb8800_get_settings, - .set_settings = nb8800_set_settings, .nway_reset = nb8800_nway_reset, .get_link = ethtool_op_get_link, .get_pauseparam = nb8800_get_pauseparam, @@ -1192,6 +1170,8 @@ static const struct ethtool_ops nb8800_ethtool_ops = { .get_sset_count = nb8800_get_sset_count, .get_strings = nb8800_get_strings, .get_ethtool_stats = nb8800_get_ethtool_stats, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int nb8800_hw_init(struct net_device *dev) -- cgit v0.10.2 From 715a022703e2d2e2e8e607fcd82f483bf8494202 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 19 Jun 2016 20:39:08 +0200 Subject: net: ethernet: bcmsysport: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index 543bf38..9775c78 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -99,23 +99,19 @@ static inline void tdma_port_write_desc_addr(struct bcm_sysport_priv *priv, static int bcm_sysport_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct bcm_sysport_priv *priv = netdev_priv(dev); - if (!netif_running(dev)) return -EINVAL; - return phy_ethtool_sset(priv->phydev, cmd); + return phy_ethtool_sset(dev->phydev, cmd); } static int bcm_sysport_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct bcm_sysport_priv *priv = netdev_priv(dev); - if (!netif_running(dev)) return -EINVAL; - return phy_ethtool_gset(priv->phydev, cmd); + return phy_ethtool_gset(dev->phydev, cmd); } static int bcm_sysport_set_rx_csum(struct net_device *dev, @@ -1127,7 +1123,7 @@ static void bcm_sysport_tx_timeout(struct net_device *dev) static void bcm_sysport_adj_link(struct net_device *dev) { struct bcm_sysport_priv *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; unsigned int changed = 0; u32 cmd_bits = 0, reg; @@ -1182,7 +1178,7 @@ static void bcm_sysport_adj_link(struct net_device *dev) umac_writel(priv, reg, UMAC_CMD); } - phy_print_status(priv->phydev); + phy_print_status(phydev); } static int bcm_sysport_init_tx_ring(struct bcm_sysport_priv *priv, @@ -1525,7 +1521,7 @@ static void bcm_sysport_netif_start(struct net_device *dev) /* Enable RX interrupt and TX ring full interrupt */ intrl2_0_mask_clear(priv, INTRL2_0_RDMA_MBDONE | INTRL2_0_TX_RING_FULL); - phy_start(priv->phydev); + phy_start(dev->phydev); /* Enable TX interrupts for the 32 TXQs */ intrl2_1_mask_clear(priv, 0xffffffff); @@ -1546,6 +1542,7 @@ static void rbuf_init(struct bcm_sysport_priv *priv) static int bcm_sysport_open(struct net_device *dev) { struct bcm_sysport_priv *priv = netdev_priv(dev); + struct phy_device *phydev; unsigned int i; int ret; @@ -1570,9 +1567,9 @@ static int bcm_sysport_open(struct net_device *dev) /* Read CRC forward */ priv->crc_fwd = !!(umac_readl(priv, UMAC_CMD) & CMD_CRC_FWD); - priv->phydev = of_phy_connect(dev, priv->phy_dn, bcm_sysport_adj_link, - 0, priv->phy_interface); - if (!priv->phydev) { + phydev = of_phy_connect(dev, priv->phy_dn, bcm_sysport_adj_link, + 0, priv->phy_interface); + if (!phydev) { netdev_err(dev, "could not attach to PHY\n"); return -ENODEV; } @@ -1650,7 +1647,7 @@ out_free_tx_ring: out_free_irq0: free_irq(priv->irq0, dev); out_phy_disconnect: - phy_disconnect(priv->phydev); + phy_disconnect(phydev); return ret; } @@ -1661,7 +1658,7 @@ static void bcm_sysport_netif_stop(struct net_device *dev) /* stop all software from updating hardware */ netif_tx_stop_all_queues(dev); napi_disable(&priv->napi); - phy_stop(priv->phydev); + phy_stop(dev->phydev); /* mask all interrupts */ intrl2_0_mask_set(priv, 0xffffffff); @@ -1708,7 +1705,7 @@ static int bcm_sysport_stop(struct net_device *dev) free_irq(priv->irq1, dev); /* Disconnect from PHY */ - phy_disconnect(priv->phydev); + phy_disconnect(dev->phydev); return 0; } @@ -1929,7 +1926,7 @@ static int bcm_sysport_suspend(struct device *d) bcm_sysport_netif_stop(dev); - phy_suspend(priv->phydev); + phy_suspend(dev->phydev); netif_device_detach(dev); @@ -2055,7 +2052,7 @@ static int bcm_sysport_resume(struct device *d) goto out_free_rx_ring; } - phy_resume(priv->phydev); + phy_resume(dev->phydev); bcm_sysport_netif_start(dev); diff --git a/drivers/net/ethernet/broadcom/bcmsysport.h b/drivers/net/ethernet/broadcom/bcmsysport.h index f28bf54..1c82e3d 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.h +++ b/drivers/net/ethernet/broadcom/bcmsysport.h @@ -670,7 +670,6 @@ struct bcm_sysport_priv { /* PHY device */ struct device_node *phy_dn; - struct phy_device *phydev; phy_interface_t phy_interface; int old_pause; int old_link; -- cgit v0.10.2 From 697666eac664dbea7c2c1fa7518fd5dfe098776f Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 19 Jun 2016 20:39:09 +0200 Subject: net: ethernet: bcmsysport: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index 9775c78..834afbb 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -96,24 +96,6 @@ static inline void tdma_port_write_desc_addr(struct bcm_sysport_priv *priv, } /* Ethtool operations */ -static int bcm_sysport_set_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - if (!netif_running(dev)) - return -EINVAL; - - return phy_ethtool_sset(dev->phydev, cmd); -} - -static int bcm_sysport_get_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - if (!netif_running(dev)) - return -EINVAL; - - return phy_ethtool_gset(dev->phydev, cmd); -} - static int bcm_sysport_set_rx_csum(struct net_device *dev, netdev_features_t wanted) { @@ -1711,8 +1693,6 @@ static int bcm_sysport_stop(struct net_device *dev) } static struct ethtool_ops bcm_sysport_ethtool_ops = { - .get_settings = bcm_sysport_get_settings, - .set_settings = bcm_sysport_set_settings, .get_drvinfo = bcm_sysport_get_drvinfo, .get_msglevel = bcm_sysport_get_msglvl, .set_msglevel = bcm_sysport_set_msglvl, @@ -1724,6 +1704,8 @@ static struct ethtool_ops bcm_sysport_ethtool_ops = { .set_wol = bcm_sysport_set_wol, .get_coalesce = bcm_sysport_get_coalesce, .set_coalesce = bcm_sysport_set_coalesce, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops bcm_sysport_netdev_ops = { -- cgit v0.10.2 From 57d3231057e9e16930cef07b4281c6c902c42670 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:13:58 -0400 Subject: net: dsa: mv88e6xxx: fix style issues This patch fixes 5 style problems reported by checkpatch: WARNING: suspect code indent for conditional statements (8, 24) #492: FILE: drivers/net/dsa/mv88e6xxx.c:492: + if (phydev->link) + reg |= PORT_PCS_CTRL_LINK_UP; CHECK: Logical continuations should be on the previous line #1318: FILE: drivers/net/dsa/mv88e6xxx.c:1318: + oldstate == PORT_CONTROL_STATE_FORWARDING) + && (state == PORT_CONTROL_STATE_DISABLED || CHECK: multiple assignments should be avoided #1662: FILE: drivers/net/dsa/mv88e6xxx.c:1662: + vlan->vid_begin = vlan->vid_end = next.vid; WARNING: line over 80 characters #2097: FILE: drivers/net/dsa/mv88e6xxx.c:2097: + const struct switchdev_obj_port_vlan *vlan, WARNING: suspect code indent for conditional statements (16, 32) #2734: FILE: drivers/net/dsa/mv88e6xxx.c:2734: + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || [...] + reg |= PORT_CONTROL_EGRESS_ADD_TAG; total: 0 errors, 3 warnings, 2 checks, 3805 lines checked It also rebases and integrates changes sent by Ben Dooks [1]: The driver has a number of functions that are not exported or declared elsewhere, so make them static to avoid the following warnings from sparse: drivers/net/dsa/mv88e6xxx.c:113:5: warning: symbol 'mv88e6xxx_reg_read' was not declared. Should it be static? drivers/net/dsa/mv88e6xxx.c:167:5: warning: symbol 'mv88e6xxx_reg_write' was not declared. Should it be static? drivers/net/dsa/mv88e6xxx.c:231:5: warning: symbol 'mv88e6xxx_set_addr' was not declared. Should it be static? drivers/net/dsa/mv88e6xxx.c:367:6: warning: symbol 'mv88e6xxx_ppu_state_init' was not declared. Should it be static? drivers/net/dsa/mv88e6xxx.c:3157:5: warning: symbol 'mv88e6xxx_phy_page_read' was not declared. Should it be static? drivers/net/dsa/mv88e6xxx.c:3169:5: warning: symbol 'mv88e6xxx_phy_page_write' was not declared. Should it be static? drivers/net/dsa/mv88e6xxx.c:3583:26: warning: symbol 'mv88e6xxx_switch_driver' was not declared. Should it be static? drivers/net/dsa/mv88e6xxx.c:3621:5: warning: symbol 'mv88e6xxx_probe' was not declared. Should it be static? [1] http://patchwork.ozlabs.org/patch/632708/ Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index ee06055..1972ec5 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -111,7 +111,8 @@ static int _mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, return ret; } -int mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, int addr, int reg) +static int mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, int addr, + int reg) { int ret; @@ -165,8 +166,8 @@ static int _mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, return __mv88e6xxx_reg_write(ps->bus, ps->sw_addr, addr, reg, val); } -int mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, - int reg, u16 val) +static int mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, + int reg, u16 val) { int ret; @@ -229,7 +230,7 @@ static int mv88e6xxx_set_addr_indirect(struct dsa_switch *ds, u8 *addr) return 0; } -int mv88e6xxx_set_addr(struct dsa_switch *ds, u8 *addr) +static int mv88e6xxx_set_addr(struct dsa_switch *ds, u8 *addr) { struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); @@ -370,7 +371,7 @@ static void mv88e6xxx_ppu_access_put(struct mv88e6xxx_priv_state *ps) mutex_unlock(&ps->ppu_mutex); } -void mv88e6xxx_ppu_state_init(struct mv88e6xxx_priv_state *ps) +static void mv88e6xxx_ppu_state_init(struct mv88e6xxx_priv_state *ps) { mutex_init(&ps->ppu_mutex); INIT_WORK(&ps->ppu_work, mv88e6xxx_ppu_reenable_work); @@ -490,7 +491,7 @@ static void mv88e6xxx_adjust_link(struct dsa_switch *ds, int port, reg |= PORT_PCS_CTRL_FORCE_LINK; if (phydev->link) - reg |= PORT_PCS_CTRL_LINK_UP; + reg |= PORT_PCS_CTRL_LINK_UP; if (mv88e6xxx_6065_family(ps) && phydev->speed > SPEED_100) goto out; @@ -1314,9 +1315,9 @@ static int _mv88e6xxx_port_state(struct mv88e6xxx_priv_state *ps, int port, * Blocking or Listening state. */ if ((oldstate == PORT_CONTROL_STATE_LEARNING || - oldstate == PORT_CONTROL_STATE_FORWARDING) - && (state == PORT_CONTROL_STATE_DISABLED || - state == PORT_CONTROL_STATE_BLOCKING)) { + oldstate == PORT_CONTROL_STATE_FORWARDING) && + (state == PORT_CONTROL_STATE_DISABLED || + state == PORT_CONTROL_STATE_BLOCKING)) { ret = _mv88e6xxx_atu_remove(ps, 0, port, false); if (ret) return ret; @@ -1659,7 +1660,8 @@ static int mv88e6xxx_port_vlan_dump(struct dsa_switch *ds, int port, continue; /* reinit and dump this VLAN obj */ - vlan->vid_begin = vlan->vid_end = next.vid; + vlan->vid_begin = next.vid; + vlan->vid_end = next.vid; vlan->flags = 0; if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED) @@ -2093,9 +2095,10 @@ unlock: return ret; } -static int mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_vlan *vlan, - struct switchdev_trans *trans) +static int +mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans) { struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); int err; @@ -2735,7 +2738,7 @@ static int mv88e6xxx_setup_port(struct mv88e6xxx_priv_state *ps, int port) mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || mv88e6xxx_6095_family(ps) || mv88e6xxx_6065_family(ps) || mv88e6xxx_6185_family(ps) || mv88e6xxx_6320_family(ps)) { - reg |= PORT_CONTROL_EGRESS_ADD_TAG; + reg |= PORT_CONTROL_EGRESS_ADD_TAG; } } if (dsa_is_dsa_port(ds, port)) { @@ -3158,7 +3161,8 @@ unlock: return err; } -int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, int reg) +static int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, + int reg) { struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); int ret; @@ -3170,8 +3174,8 @@ int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, int reg) return ret; } -int mv88e6xxx_mdio_page_write(struct dsa_switch *ds, int port, int page, - int reg, int val) +static int mv88e6xxx_mdio_page_write(struct dsa_switch *ds, int port, int page, + int reg, int val) { struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); int ret; @@ -3650,7 +3654,7 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, return name; } -struct dsa_switch_driver mv88e6xxx_switch_driver = { +static struct dsa_switch_driver mv88e6xxx_switch_driver = { .tag_protocol = DSA_TAG_PROTO_EDSA, .probe = mv88e6xxx_drv_probe, .setup = mv88e6xxx_setup, @@ -3686,7 +3690,7 @@ struct dsa_switch_driver mv88e6xxx_switch_driver = { .port_fdb_dump = mv88e6xxx_port_fdb_dump, }; -int mv88e6xxx_probe(struct mdio_device *mdiodev) +static int mv88e6xxx_probe(struct mdio_device *mdiodev) { struct device *dev = &mdiodev->dev; struct device_node *np = dev->of_node; -- cgit v0.10.2 From fbae5a4895b8694126388ee033b6dd0b33fadf2b Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:13:59 -0400 Subject: net: dsa: mv88e6xxx: remove redundant assignments The chip->ds and ds->slave_mii_bus assignments are common to both legacy and new MDIO probing and are already done in the later setup code. Remove the duplicated assignments from the MDIO probing code. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 1972ec5..673283a 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3708,7 +3708,6 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) ds->priv = ps; ds->dev = dev; ps->dev = dev; - ps->ds = ds; ps->bus = mdiodev->bus; ps->sw_addr = mdiodev->addr; mutex_init(&ps->smi_mutex); @@ -3748,8 +3747,6 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) if (err) return err; - ds->slave_mii_bus = ps->mdio_bus; - dev_set_drvdata(dev, ds); err = dsa_register_switch(ds, mdiodev->dev.of_node); -- cgit v0.10.2 From aa8ac3967e1f8d21b5f89d2a17b1281e1eb52522 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:00 -0400 Subject: net: dsa: mv88e6xxx: use already declared variables In the MDIO probing function, dev is already assigned to &mdiodev->dev and np is already assigned to mdiodev->dev.of_node, so use them. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 673283a..b3170ea 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3728,7 +3728,7 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) if (!ps->info) return -ENODEV; - ps->reset = devm_gpiod_get(&mdiodev->dev, "reset", GPIOD_ASIS); + ps->reset = devm_gpiod_get(dev, "reset", GPIOD_ASIS); if (IS_ERR(ps->reset)) { err = PTR_ERR(ps->reset); if (err == -ENOENT) { @@ -3743,13 +3743,13 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) !of_property_read_u32(np, "eeprom-length", &eeprom_len)) ps->eeprom_len = eeprom_len; - err = mv88e6xxx_mdio_register(ps, mdiodev->dev.of_node); + err = mv88e6xxx_mdio_register(ps, np); if (err) return err; dev_set_drvdata(dev, ds); - err = dsa_register_switch(ds, mdiodev->dev.of_node); + err = dsa_register_switch(ds, np); if (err) { mv88e6xxx_mdio_unregister(ps); return err; -- cgit v0.10.2 From 1d35f0b2c3fcc8a0c1fd93aba02520c940c0b855 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:01 -0400 Subject: net: dsa: mv88e6xxx: do not increment bus refcount The MDIO device probe and remove functions are respectively incrementing and decrementing the bus refcount themselves. Since these bus level actions are out of the device scope, remove them. Signed-off-by: Vivien Didelot Acked-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index b3170ea..4b4bffc 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3712,8 +3712,6 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) ps->sw_addr = mdiodev->addr; mutex_init(&ps->smi_mutex); - get_device(&ps->bus->dev); - ds->drv = &mv88e6xxx_switch_driver; id = mv88e6xxx_reg_read(ps, REG_PORT(0), PORT_SWITCH_ID); @@ -3767,7 +3765,6 @@ static void mv88e6xxx_remove(struct mdio_device *mdiodev) struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); dsa_unregister_switch(ds); - put_device(&ps->bus->dev); mv88e6xxx_mdio_unregister(ps); } -- cgit v0.10.2 From b7e66a5fad73cb0f118471e6ca8aeed3447915dc Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:02 -0400 Subject: net: dsa: mv88e6xxx: add switch register helpers The mixed assignments, allocations and registrations in the probe code make it hard to follow the logic and figure out what is DSA or chip specific. Extract the struct dsa_switch related code in a simple mv88e6xxx_register_switch helper function. For symmetry in the code, add a mv88e6xxx_unregister_switch function. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 4b4bffc..ad7735d 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3690,30 +3690,48 @@ static struct dsa_switch_driver mv88e6xxx_switch_driver = { .port_fdb_dump = mv88e6xxx_port_fdb_dump, }; +static int mv88e6xxx_register_switch(struct mv88e6xxx_priv_state *ps, + struct device_node *np) +{ + struct device *dev = ps->dev; + struct dsa_switch *ds; + + ds = devm_kzalloc(dev, sizeof(*ds), GFP_KERNEL); + if (!ds) + return -ENOMEM; + + ds->dev = dev; + ds->priv = ps; + ds->drv = &mv88e6xxx_switch_driver; + + dev_set_drvdata(dev, ds); + + return dsa_register_switch(ds, np); +} + +static void mv88e6xxx_unregister_switch(struct mv88e6xxx_priv_state *ps) +{ + dsa_unregister_switch(ps->ds); +} + static int mv88e6xxx_probe(struct mdio_device *mdiodev) { struct device *dev = &mdiodev->dev; struct device_node *np = dev->of_node; struct mv88e6xxx_priv_state *ps; int id, prod_num, rev; - struct dsa_switch *ds; u32 eeprom_len; int err; - ds = devm_kzalloc(dev, sizeof(*ds) + sizeof(*ps), GFP_KERNEL); - if (!ds) + ps = devm_kzalloc(dev, sizeof(*ps), GFP_KERNEL); + if (!ps) return -ENOMEM; - ps = (struct mv88e6xxx_priv_state *)(ds + 1); - ds->priv = ps; - ds->dev = dev; ps->dev = dev; ps->bus = mdiodev->bus; ps->sw_addr = mdiodev->addr; mutex_init(&ps->smi_mutex); - ds->drv = &mv88e6xxx_switch_driver; - id = mv88e6xxx_reg_read(ps, REG_PORT(0), PORT_SWITCH_ID); if (id < 0) return id; @@ -3745,9 +3763,7 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) if (err) return err; - dev_set_drvdata(dev, ds); - - err = dsa_register_switch(ds, np); + err = mv88e6xxx_register_switch(ps, np); if (err) { mv88e6xxx_mdio_unregister(ps); return err; @@ -3764,8 +3780,7 @@ static void mv88e6xxx_remove(struct mdio_device *mdiodev) struct dsa_switch *ds = dev_get_drvdata(&mdiodev->dev); struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - dsa_unregister_switch(ds); - + mv88e6xxx_unregister_switch(ps); mv88e6xxx_mdio_unregister(ps); } -- cgit v0.10.2 From c6d19ab609d5a7f0eb9385d8dda6b2b5b3d9122f Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:03 -0400 Subject: net: dsa: mv88e6xxx: use gpio get optional variant Use the optional variant to get the reset GPIO line, instead of checking for the -ENOENT error. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index ad7735d..ec28465 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3744,16 +3744,9 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) if (!ps->info) return -ENODEV; - ps->reset = devm_gpiod_get(dev, "reset", GPIOD_ASIS); - if (IS_ERR(ps->reset)) { - err = PTR_ERR(ps->reset); - if (err == -ENOENT) { - /* Optional, so not an error */ - ps->reset = NULL; - } else { - return err; - } - } + ps->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_ASIS); + if (IS_ERR(ps->reset)) + return PTR_ERR(ps->reset); if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM) && !of_property_read_u32(np, "eeprom-length", &eeprom_len)) -- cgit v0.10.2 From 5f7c036719536362f293c03c951f36a0509e7aec Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:04 -0400 Subject: net: dsa: mv88e6xxx: remove table args in info lookup The mv88e6xxx_table array and the mv88e6xxx_lookup_info function are static, so remove the table and size arguments from the lookup function. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index ec28465..75e5408 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3590,15 +3590,13 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { }, }; -static const struct mv88e6xxx_info * -mv88e6xxx_lookup_info(unsigned int prod_num, const struct mv88e6xxx_info *table, - unsigned int num) +static const struct mv88e6xxx_info *mv88e6xxx_lookup_info(unsigned int prod_num) { int i; - for (i = 0; i < num; ++i) - if (table[i].prod_num == prod_num) - return &table[i]; + for (i = 0; i < ARRAY_SIZE(mv88e6xxx_table); ++i) + if (mv88e6xxx_table[i].prod_num == prod_num) + return &mv88e6xxx_table[i]; return NULL; } @@ -3625,8 +3623,7 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, prod_num = (id & 0xfff0) >> 4; rev = id & 0x000f; - info = mv88e6xxx_lookup_info(prod_num, mv88e6xxx_table, - ARRAY_SIZE(mv88e6xxx_table)); + info = mv88e6xxx_lookup_info(prod_num); if (!info) return NULL; @@ -3739,8 +3736,7 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) prod_num = (id & 0xfff0) >> 4; rev = id & 0x000f; - ps->info = mv88e6xxx_lookup_info(prod_num, mv88e6xxx_table, - ARRAY_SIZE(mv88e6xxx_table)); + ps->info = mv88e6xxx_lookup_info(prod_num); if (!ps->info) return -ENODEV; -- cgit v0.10.2 From 9f8b3ee1b5be49f5757fd60c22c921ad3ef58585 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:05 -0400 Subject: net: dsa: mv88e6xxx: rename smi_mutex to reg_lock The chip smi_mutex mutex is used to protect the access to the internal switch registers, not only the Multi-chip Addressing Mode, as commented. Since we will isolate SMI-specific pieces of code, avoid the confusion now by renaming smi_mutex to reg_lock. No functional changes here. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 75e5408..2c86172 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -29,10 +29,10 @@ #include #include "mv88e6xxx.h" -static void assert_smi_lock(struct mv88e6xxx_priv_state *ps) +static void assert_reg_lock(struct mv88e6xxx_priv_state *ps) { - if (unlikely(!mutex_is_locked(&ps->smi_mutex))) { - dev_err(ps->dev, "SMI lock not held!\n"); + if (unlikely(!mutex_is_locked(&ps->reg_lock))) { + dev_err(ps->dev, "Switch registers lock not held!\n"); dump_stack(); } } @@ -99,7 +99,7 @@ static int _mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, { int ret; - assert_smi_lock(ps); + assert_reg_lock(ps); ret = __mv88e6xxx_reg_read(ps->bus, ps->sw_addr, addr, reg); if (ret < 0) @@ -116,9 +116,9 @@ static int mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, int addr, { int ret; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = _mv88e6xxx_reg_read(ps, addr, reg); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -158,7 +158,7 @@ static int __mv88e6xxx_reg_write(struct mii_bus *bus, int sw_addr, int addr, static int _mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, int reg, u16 val) { - assert_smi_lock(ps); + assert_reg_lock(ps); dev_dbg(ps->dev, "-> addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", addr, reg, val); @@ -171,9 +171,9 @@ static int mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, { int ret; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = _mv88e6xxx_reg_write(ps, addr, reg, val); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -320,7 +320,7 @@ static void mv88e6xxx_ppu_reenable_work(struct work_struct *ugly) ps = container_of(ugly, struct mv88e6xxx_priv_state, ppu_work); - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); if (mutex_trylock(&ps->ppu_mutex)) { if (mv88e6xxx_ppu_enable(ps) == 0) @@ -328,7 +328,7 @@ static void mv88e6xxx_ppu_reenable_work(struct work_struct *ugly) mutex_unlock(&ps->ppu_mutex); } - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); } static void mv88e6xxx_ppu_reenable_timer(unsigned long _ps) @@ -477,7 +477,7 @@ static void mv88e6xxx_adjust_link(struct dsa_switch *ds, int port, if (!phy_is_pseudo_fixed_link(phydev)) return; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_PCS_CTRL); if (ret < 0) @@ -528,7 +528,7 @@ static void mv88e6xxx_adjust_link(struct dsa_switch *ds, int port, _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_PCS_CTRL, reg); out: - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); } static int _mv88e6xxx_stats_wait(struct mv88e6xxx_priv_state *ps) @@ -753,11 +753,11 @@ static void mv88e6xxx_get_ethtool_stats(struct dsa_switch *ds, int port, int ret; int i, j; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = _mv88e6xxx_stats_snapshot(ps, port); if (ret < 0) { - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return; } for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { @@ -768,7 +768,7 @@ static void mv88e6xxx_get_ethtool_stats(struct dsa_switch *ds, int port, } } - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); } static int mv88e6xxx_get_regs_len(struct dsa_switch *ds, int port) @@ -787,7 +787,7 @@ static void mv88e6xxx_get_regs(struct dsa_switch *ds, int port, memset(p, 0xff, 32 * sizeof(u16)); - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); for (i = 0; i < 32; i++) { int ret; @@ -797,7 +797,7 @@ static void mv88e6xxx_get_regs(struct dsa_switch *ds, int port, p[i] = ret; } - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); } static int _mv88e6xxx_wait(struct mv88e6xxx_priv_state *ps, int reg, int offset, @@ -824,9 +824,9 @@ static int mv88e6xxx_wait(struct mv88e6xxx_priv_state *ps, int reg, { int ret; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = _mv88e6xxx_wait(ps, reg, offset, mask); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -1123,7 +1123,7 @@ static int mv88e6xxx_get_eee(struct dsa_switch *ds, int port, if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEE)) return -EOPNOTSUPP; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); reg = mv88e6xxx_mdio_read_indirect(ps, port, 16); if (reg < 0) @@ -1140,7 +1140,7 @@ static int mv88e6xxx_get_eee(struct dsa_switch *ds, int port, reg = 0; out: - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return reg; } @@ -1154,7 +1154,7 @@ static int mv88e6xxx_set_eee(struct dsa_switch *ds, int port, if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEE)) return -EOPNOTSUPP; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = mv88e6xxx_mdio_read_indirect(ps, port, 16); if (ret < 0) @@ -1168,7 +1168,7 @@ static int mv88e6xxx_set_eee(struct dsa_switch *ds, int port, ret = mv88e6xxx_mdio_write_indirect(ps, port, 16, reg); out: - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -1402,9 +1402,9 @@ static void mv88e6xxx_port_stp_state_set(struct dsa_switch *ds, int port, break; } - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); err = _mv88e6xxx_port_state(ps, port, stp_state); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); if (err) netdev_err(ds->ports[port].netdev, @@ -1638,7 +1638,7 @@ static int mv88e6xxx_port_vlan_dump(struct dsa_switch *ds, int port, if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) return -EOPNOTSUPP; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); err = _mv88e6xxx_port_pvid_get(ps, port, &pvid); if (err) @@ -1676,7 +1676,7 @@ static int mv88e6xxx_port_vlan_dump(struct dsa_switch *ds, int port, } while (next.vid < GLOBAL_VTU_VID_MASK); unlock: - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return err; } @@ -2004,7 +2004,7 @@ static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port, if (!vid_begin) return -EOPNOTSUPP; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); err = _mv88e6xxx_vtu_vid_write(ps, vid_begin - 1); if (err) @@ -2043,7 +2043,7 @@ static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port, } while (vlan.vid < vid_end); unlock: - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return err; } @@ -2066,7 +2066,7 @@ static int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port, if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) return -EOPNOTSUPP; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_CONTROL_2); if (ret < 0) @@ -2090,7 +2090,7 @@ static int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port, ret = 0; unlock: - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -2149,7 +2149,7 @@ static void mv88e6xxx_port_vlan_add(struct dsa_switch *ds, int port, if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) return; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) if (_mv88e6xxx_port_vlan_add(ps, port, vid, untagged)) @@ -2161,7 +2161,7 @@ static void mv88e6xxx_port_vlan_add(struct dsa_switch *ds, int port, netdev_err(ds->ports[port].netdev, "failed to set PVID %d\n", vlan->vid_end); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); } static int _mv88e6xxx_port_vlan_del(struct mv88e6xxx_priv_state *ps, @@ -2210,7 +2210,7 @@ static int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port, if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) return -EOPNOTSUPP; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); err = _mv88e6xxx_port_pvid_get(ps, port, &pvid); if (err) @@ -2229,7 +2229,7 @@ static int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port, } unlock: - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return err; } @@ -2341,11 +2341,11 @@ static void mv88e6xxx_port_fdb_add(struct dsa_switch *ds, int port, if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) return; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); if (_mv88e6xxx_port_fdb_load(ps, port, fdb->addr, fdb->vid, state)) netdev_err(ds->ports[port].netdev, "failed to load MAC address\n"); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); } static int mv88e6xxx_port_fdb_del(struct dsa_switch *ds, int port, @@ -2357,10 +2357,10 @@ static int mv88e6xxx_port_fdb_del(struct dsa_switch *ds, int port, if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) return -EOPNOTSUPP; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = _mv88e6xxx_port_fdb_load(ps, port, fdb->addr, fdb->vid, GLOBAL_ATU_DATA_STATE_UNUSED); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -2465,7 +2465,7 @@ static int mv88e6xxx_port_fdb_dump(struct dsa_switch *ds, int port, if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) return -EOPNOTSUPP; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); /* Dump port's default Filtering Information Database (VLAN ID 0) */ err = _mv88e6xxx_port_fid_get(ps, port, &fid); @@ -2496,7 +2496,7 @@ static int mv88e6xxx_port_fdb_dump(struct dsa_switch *ds, int port, } while (vlan.vid < GLOBAL_VTU_VID_MASK); unlock: - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return err; } @@ -2510,7 +2510,7 @@ static int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port, if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VLANTABLE)) return -EOPNOTSUPP; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); /* Assign the bridge and remap each port's VLANTable */ ps->ports[port].bridge_dev = bridge; @@ -2523,7 +2523,7 @@ static int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port, } } - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return err; } @@ -2537,7 +2537,7 @@ static void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port) if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VLANTABLE)) return; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); /* Unassign the bridge and remap each port's VLANTable */ ps->ports[port].bridge_dev = NULL; @@ -2548,7 +2548,7 @@ static void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port) netdev_warn(ds->ports[i].netdev, "failed to remap\n"); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); } static int _mv88e6xxx_mdio_page_write(struct mv88e6xxx_priv_state *ps, @@ -3139,7 +3139,7 @@ static int mv88e6xxx_setup(struct dsa_switch *ds) if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) mutex_init(&ps->eeprom_mutex); - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); err = mv88e6xxx_switch_reset(ps); if (err) @@ -3156,7 +3156,7 @@ static int mv88e6xxx_setup(struct dsa_switch *ds) } unlock: - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return err; } @@ -3167,9 +3167,9 @@ static int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); int ret; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = _mv88e6xxx_mdio_page_read(ps, port, page, reg); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -3180,9 +3180,9 @@ static int mv88e6xxx_mdio_page_write(struct dsa_switch *ds, int port, int page, struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); int ret; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = _mv88e6xxx_mdio_page_write(ps, port, page, reg, val); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -3204,7 +3204,7 @@ static int mv88e6xxx_mdio_read(struct mii_bus *bus, int port, int regnum) if (addr < 0) return 0xffff; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) ret = mv88e6xxx_mdio_read_ppu(ps, addr, regnum); @@ -3213,7 +3213,7 @@ static int mv88e6xxx_mdio_read(struct mii_bus *bus, int port, int regnum) else ret = mv88e6xxx_mdio_read_direct(ps, addr, regnum); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -3227,7 +3227,7 @@ static int mv88e6xxx_mdio_write(struct mii_bus *bus, int port, int regnum, if (addr < 0) return 0xffff; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) ret = mv88e6xxx_mdio_write_ppu(ps, addr, regnum, val); @@ -3236,7 +3236,7 @@ static int mv88e6xxx_mdio_write(struct mii_bus *bus, int port, int regnum, else ret = mv88e6xxx_mdio_write_direct(ps, addr, regnum, val); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -3310,7 +3310,7 @@ static int mv88e61xx_get_temp(struct dsa_switch *ds, int *temp) *temp = 0; - mutex_lock(&ps->smi_mutex); + mutex_lock(&ps->reg_lock); ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x16, 0x6); if (ret < 0) @@ -3343,7 +3343,7 @@ static int mv88e61xx_get_temp(struct dsa_switch *ds, int *temp) error: mv88e6xxx_mdio_write_direct(ps, 0x0, 0x16, 0x0); - mutex_unlock(&ps->smi_mutex); + mutex_unlock(&ps->reg_lock); return ret; } @@ -3637,7 +3637,7 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, ps->sw_addr = sw_addr; ps->info = info; ps->dev = dsa_dev; - mutex_init(&ps->smi_mutex); + mutex_init(&ps->reg_lock); err = mv88e6xxx_mdio_register(ps, NULL); if (err) @@ -3727,7 +3727,7 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) ps->dev = dev; ps->bus = mdiodev->bus; ps->sw_addr = mdiodev->addr; - mutex_init(&ps->smi_mutex); + mutex_init(&ps->reg_lock); id = mv88e6xxx_reg_read(ps, REG_PORT(0), PORT_SWITCH_ID); if (id < 0) diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h index 8221c3c..b279f8c 100644 --- a/drivers/net/dsa/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx.h @@ -554,11 +554,8 @@ struct mv88e6xxx_priv_state { /* The device this structure is associated to */ struct device *dev; - /* When using multi-chip addressing, this mutex protects - * access to the indirect access registers. (In single-chip - * mode, this mutex is effectively useless.) - */ - struct mutex smi_mutex; + /* This mutex protects the access to the switch registers */ + struct mutex reg_lock; /* The MII bus and the address on the bus that is used to * communication with the switch -- cgit v0.10.2 From 469d729f2a701e78d12b936bf0df63b8acae73c9 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:06 -0400 Subject: net: dsa: mv88e6xxx: add chip allocation helper Add an helper function to allocate the chip structure at the beginning of the probe functions. It will be used to initialize the SMI access. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 2c86172..113092a 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3601,6 +3601,21 @@ static const struct mv88e6xxx_info *mv88e6xxx_lookup_info(unsigned int prod_num) return NULL; } +static struct mv88e6xxx_priv_state *mv88e6xxx_alloc_chip(struct device *dev) +{ + struct mv88e6xxx_priv_state *ps; + + ps = devm_kzalloc(dev, sizeof(*ps), GFP_KERNEL); + if (!ps) + return NULL; + + ps->dev = dev; + + mutex_init(&ps->reg_lock); + + return ps; +} + static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, struct device *host_dev, int sw_addr, void **priv) @@ -3616,32 +3631,30 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, if (!bus) return NULL; + ps = mv88e6xxx_alloc_chip(dsa_dev); + if (!ps) + return NULL; + id = __mv88e6xxx_reg_read(bus, sw_addr, REG_PORT(0), PORT_SWITCH_ID); if (id < 0) - return NULL; + goto free; prod_num = (id & 0xfff0) >> 4; rev = id & 0x000f; info = mv88e6xxx_lookup_info(prod_num); if (!info) - return NULL; + goto free; name = info->name; - ps = devm_kzalloc(dsa_dev, sizeof(*ps), GFP_KERNEL); - if (!ps) - return NULL; - ps->bus = bus; ps->sw_addr = sw_addr; ps->info = info; - ps->dev = dsa_dev; - mutex_init(&ps->reg_lock); err = mv88e6xxx_mdio_register(ps, NULL); if (err) - return NULL; + goto free; *priv = ps; @@ -3649,6 +3662,10 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, prod_num, name, rev); return name; +free: + devm_kfree(dsa_dev, ps); + + return NULL; } static struct dsa_switch_driver mv88e6xxx_switch_driver = { @@ -3720,14 +3737,12 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) u32 eeprom_len; int err; - ps = devm_kzalloc(dev, sizeof(*ps), GFP_KERNEL); + ps = mv88e6xxx_alloc_chip(dev); if (!ps) return -ENOMEM; - ps->dev = dev; ps->bus = mdiodev->bus; ps->sw_addr = mdiodev->addr; - mutex_init(&ps->reg_lock); id = mv88e6xxx_reg_read(ps, REG_PORT(0), PORT_SWITCH_ID); if (id < 0) -- cgit v0.10.2 From 4a70c4ab4fa49c07cbec519dfb9f9f90d2fec8d5 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:07 -0400 Subject: net: dsa: mv88e6xxx: add SMI init helper Add an helper function to isolate SMI specific assignments and checks. This function will later help choosing the different SMI accesses based of the compatible info. Since the chip structure is already allocated in the legacy probe, use the mv88e6xxx_reg_read access routine instead of __mv88e6xxx_reg_read. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 113092a..4e24ac5 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3616,6 +3616,19 @@ static struct mv88e6xxx_priv_state *mv88e6xxx_alloc_chip(struct device *dev) return ps; } +static int mv88e6xxx_smi_init(struct mv88e6xxx_priv_state *ps, + struct mii_bus *bus, int sw_addr) +{ + /* ADDR[0] pin is unavailable externally and considered zero */ + if (sw_addr & 0x1) + return -EINVAL; + + ps->bus = bus; + ps->sw_addr = sw_addr; + + return 0; +} + static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, struct device *host_dev, int sw_addr, void **priv) @@ -3635,7 +3648,11 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, if (!ps) return NULL; - id = __mv88e6xxx_reg_read(bus, sw_addr, REG_PORT(0), PORT_SWITCH_ID); + err = mv88e6xxx_smi_init(ps, bus, sw_addr); + if (err) + goto free; + + id = mv88e6xxx_reg_read(ps, REG_PORT(0), PORT_SWITCH_ID); if (id < 0) goto free; @@ -3648,8 +3665,6 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, name = info->name; - ps->bus = bus; - ps->sw_addr = sw_addr; ps->info = info; err = mv88e6xxx_mdio_register(ps, NULL); @@ -3741,8 +3756,9 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) if (!ps) return -ENOMEM; - ps->bus = mdiodev->bus; - ps->sw_addr = mdiodev->addr; + err = mv88e6xxx_smi_init(ps, mdiodev->bus, mdiodev->addr); + if (err) + return err; id = mv88e6xxx_reg_read(ps, REG_PORT(0), PORT_SWITCH_ID); if (id < 0) -- cgit v0.10.2 From bc46a3d57cf7715d76eef8ea822c763cd271aacf Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:08 -0400 Subject: net: dsa: mv88e6xxx: add detection helper Extract the common detection code which assigns the info structure to the chip given the read switch ID. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 4e24ac5..de92add 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3601,6 +3601,30 @@ static const struct mv88e6xxx_info *mv88e6xxx_lookup_info(unsigned int prod_num) return NULL; } +static int mv88e6xxx_detect(struct mv88e6xxx_priv_state *ps) +{ + const struct mv88e6xxx_info *info; + int id, prod_num, rev; + + id = mv88e6xxx_reg_read(ps, REG_PORT(0), PORT_SWITCH_ID); + if (id < 0) + return id; + + prod_num = (id & 0xfff0) >> 4; + rev = id & 0x000f; + + info = mv88e6xxx_lookup_info(prod_num); + if (!info) + return -ENODEV; + + ps->info = info; + + dev_info(ps->dev, "switch 0x%x detected: %s, revision %u\n", + ps->info->prod_num, ps->info->name, rev); + + return 0; +} + static struct mv88e6xxx_priv_state *mv88e6xxx_alloc_chip(struct device *dev) { struct mv88e6xxx_priv_state *ps; @@ -3633,11 +3657,8 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, struct device *host_dev, int sw_addr, void **priv) { - const struct mv88e6xxx_info *info; struct mv88e6xxx_priv_state *ps; struct mii_bus *bus; - const char *name; - int id, prod_num, rev; int err; bus = dsa_host_dev_to_mii_bus(host_dev); @@ -3652,31 +3673,17 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, if (err) goto free; - id = mv88e6xxx_reg_read(ps, REG_PORT(0), PORT_SWITCH_ID); - if (id < 0) - goto free; - - prod_num = (id & 0xfff0) >> 4; - rev = id & 0x000f; - - info = mv88e6xxx_lookup_info(prod_num); - if (!info) + err = mv88e6xxx_detect(ps); + if (err) goto free; - name = info->name; - - ps->info = info; - err = mv88e6xxx_mdio_register(ps, NULL); if (err) goto free; *priv = ps; - dev_info(&ps->bus->dev, "switch 0x%x probed: %s, revision %u\n", - prod_num, name, rev); - - return name; + return ps->info->name; free: devm_kfree(dsa_dev, ps); @@ -3748,7 +3755,6 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) struct device *dev = &mdiodev->dev; struct device_node *np = dev->of_node; struct mv88e6xxx_priv_state *ps; - int id, prod_num, rev; u32 eeprom_len; int err; @@ -3760,16 +3766,9 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) if (err) return err; - id = mv88e6xxx_reg_read(ps, REG_PORT(0), PORT_SWITCH_ID); - if (id < 0) - return id; - - prod_num = (id & 0xfff0) >> 4; - rev = id & 0x000f; - - ps->info = mv88e6xxx_lookup_info(prod_num); - if (!ps->info) - return -ENODEV; + err = mv88e6xxx_detect(ps); + if (err) + return err; ps->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_ASIS); if (IS_ERR(ps->reset)) @@ -3789,9 +3788,6 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) return err; } - dev_info(dev, "switch 0x%x probed: %s, revision %u\n", - prod_num, ps->info->name, rev); - return 0; } -- cgit v0.10.2 From caac8545c861f2e78ae2565de31b573739b4034b Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:09 -0400 Subject: net: dsa: mv88e6xxx: pass compatible info After allocating the chip structure, pass it a compatible info pointer. The compatible info structure will be used later to describe how to access the switch registers and where to read the switch ID. For the standard MDIO probe, get it from the device node data. For the legacy DSA driver probing, pass it the 88E6085 info. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index de92add..cadd1e3 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include @@ -3617,6 +3618,7 @@ static int mv88e6xxx_detect(struct mv88e6xxx_priv_state *ps) if (!info) return -ENODEV; + /* Update the compatible info with the probed one */ ps->info = info; dev_info(ps->dev, "switch 0x%x detected: %s, revision %u\n", @@ -3669,6 +3671,9 @@ static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, if (!ps) return NULL; + /* Legacy SMI probing will only support chips similar to 88E6085 */ + ps->info = &mv88e6xxx_table[MV88E6085]; + err = mv88e6xxx_smi_init(ps, bus, sw_addr); if (err) goto free; @@ -3754,14 +3759,21 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) { struct device *dev = &mdiodev->dev; struct device_node *np = dev->of_node; + const struct mv88e6xxx_info *compat_info; struct mv88e6xxx_priv_state *ps; u32 eeprom_len; int err; + compat_info = of_device_get_match_data(dev); + if (!compat_info) + return -EINVAL; + ps = mv88e6xxx_alloc_chip(dev); if (!ps) return -ENOMEM; + ps->info = compat_info; + err = mv88e6xxx_smi_init(ps, mdiodev->bus, mdiodev->addr); if (err) return err; @@ -3801,7 +3813,10 @@ static void mv88e6xxx_remove(struct mdio_device *mdiodev) } static const struct of_device_id mv88e6xxx_of_match[] = { - { .compatible = "marvell,mv88e6085" }, + { + .compatible = "marvell,mv88e6085", + .data = &mv88e6xxx_table[MV88E6085], + }, { /* sentinel */ }, }; -- cgit v0.10.2 From 9dddd478d4883deb1f5b6d3fcea681c9c9e90708 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:10 -0400 Subject: net: dsa: mv88e6xxx: add port base address to info The switch ID is located at address 0x3 of every Port Registers bank. But not all Marvell switches have their Port Registers SMI Addresses starting at 0x10. 88E6060 starts at 0x8 and 88E6390 starts at 0x0. Add this data in the info structure and use it in the detection code. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index cadd1e3..789f938a 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -3443,6 +3443,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6085", .num_databases = 4096, .num_ports = 10, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6097, }, @@ -3452,6 +3453,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6095/88E6095F", .num_databases = 256, .num_ports = 11, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6095, }, @@ -3461,6 +3463,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6123", .num_databases = 4096, .num_ports = 3, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6165, }, @@ -3470,6 +3473,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6131", .num_databases = 256, .num_ports = 8, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6185, }, @@ -3479,6 +3483,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6161", .num_databases = 4096, .num_ports = 6, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6165, }, @@ -3488,6 +3493,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6165", .num_databases = 4096, .num_ports = 6, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6165, }, @@ -3497,6 +3503,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6171", .num_databases = 4096, .num_ports = 7, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6351, }, @@ -3506,6 +3513,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6172", .num_databases = 4096, .num_ports = 7, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6352, }, @@ -3515,6 +3523,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6175", .num_databases = 4096, .num_ports = 7, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6351, }, @@ -3524,6 +3533,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6176", .num_databases = 4096, .num_ports = 7, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6352, }, @@ -3533,6 +3543,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6185", .num_databases = 256, .num_ports = 10, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6185, }, @@ -3542,6 +3553,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6240", .num_databases = 4096, .num_ports = 7, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6352, }, @@ -3551,6 +3563,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6320", .num_databases = 4096, .num_ports = 7, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6320, }, @@ -3560,6 +3573,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6321", .num_databases = 4096, .num_ports = 7, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6320, }, @@ -3569,6 +3583,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6350", .num_databases = 4096, .num_ports = 7, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6351, }, @@ -3578,6 +3593,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6351", .num_databases = 4096, .num_ports = 7, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6351, }, @@ -3587,6 +3603,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .name = "Marvell 88E6352", .num_databases = 4096, .num_ports = 7, + .port_base_addr = 0x10, .flags = MV88E6XXX_FLAGS_FAMILY_6352, }, }; @@ -3607,7 +3624,7 @@ static int mv88e6xxx_detect(struct mv88e6xxx_priv_state *ps) const struct mv88e6xxx_info *info; int id, prod_num, rev; - id = mv88e6xxx_reg_read(ps, REG_PORT(0), PORT_SWITCH_ID); + id = mv88e6xxx_reg_read(ps, ps->info->port_base_addr, PORT_SWITCH_ID); if (id < 0) return id; diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h index b279f8c..8e6fe6b 100644 --- a/drivers/net/dsa/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx.h @@ -519,6 +519,7 @@ struct mv88e6xxx_info { const char *name; unsigned int num_databases; unsigned int num_ports; + unsigned int port_base_addr; unsigned long flags; }; -- cgit v0.10.2 From 914b32f65afaaab005151335e183b4194d48162d Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 20 Jun 2016 13:14:11 -0400 Subject: net: dsa: mv88e6xxx: abstract switch registers accesses When the SMI address of the switch chip is zero, the chip assumes to be the only one on the SMI master bus and thus responds to all its known SMI devices addresses (port registers, Global2, etc.) When its SMI address is not zero, some chips (e.g. 88E6352) use an indirect access through two SMI Command and Data registers. Other models (e.g. 88E6060) using less than 16 internal SMI addresses always use a direct access. Add a capability flag to describe chips supporting the (indirect) Multi-chip Addressing Mode, and a low-level API to access the registers via SMI. Other accesses (like Ethernet management frames) may be added later. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index 789f938a..9b116d8 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -38,21 +38,74 @@ static void assert_reg_lock(struct mv88e6xxx_priv_state *ps) } } -/* If the switch's ADDR[4:0] strap pins are strapped to zero, it will - * use all 32 SMI bus addresses on its SMI bus, and all switch registers - * will be directly accessible on some {device address,register address} - * pair. If the ADDR[4:0] pins are not strapped to zero, the switch - * will only respond to SMI transactions to that specific address, and - * an indirect addressing mechanism needs to be used to access its - * registers. +/* The switch ADDR[4:1] configuration pins define the chip SMI device address + * (ADDR[0] is always zero, thus only even SMI addresses can be strapped). + * + * When ADDR is all zero, the chip uses Single-chip Addressing Mode, assuming it + * is the only device connected to the SMI master. In this mode it responds to + * all 32 possible SMI addresses, and thus maps directly the internal devices. + * + * When ADDR is non-zero, the chip uses Multi-chip Addressing Mode, allowing + * multiple devices to share the SMI interface. In this mode it responds to only + * 2 registers, used to indirectly access the internal SMI devices. */ -static int mv88e6xxx_reg_wait_ready(struct mii_bus *bus, int sw_addr) + +static int mv88e6xxx_smi_read(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 *val) +{ + if (!ps->smi_ops) + return -EOPNOTSUPP; + + return ps->smi_ops->read(ps, addr, reg, val); +} + +static int mv88e6xxx_smi_write(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 val) +{ + if (!ps->smi_ops) + return -EOPNOTSUPP; + + return ps->smi_ops->write(ps, addr, reg, val); +} + +static int mv88e6xxx_smi_single_chip_read(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 *val) +{ + int ret; + + ret = mdiobus_read_nested(ps->bus, addr, reg); + if (ret < 0) + return ret; + + *val = ret & 0xffff; + + return 0; +} + +static int mv88e6xxx_smi_single_chip_write(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 val) +{ + int ret; + + ret = mdiobus_write_nested(ps->bus, addr, reg, val); + if (ret < 0) + return ret; + + return 0; +} + +static const struct mv88e6xxx_ops mv88e6xxx_smi_single_chip_ops = { + .read = mv88e6xxx_smi_single_chip_read, + .write = mv88e6xxx_smi_single_chip_write, +}; + +static int mv88e6xxx_smi_multi_chip_wait(struct mv88e6xxx_priv_state *ps) { int ret; int i; for (i = 0; i < 16; i++) { - ret = mdiobus_read_nested(bus, sw_addr, SMI_CMD); + ret = mdiobus_read_nested(ps->bus, ps->sw_addr, SMI_CMD); if (ret < 0) return ret; @@ -63,108 +116,134 @@ static int mv88e6xxx_reg_wait_ready(struct mii_bus *bus, int sw_addr) return -ETIMEDOUT; } -static int __mv88e6xxx_reg_read(struct mii_bus *bus, int sw_addr, int addr, - int reg) +static int mv88e6xxx_smi_multi_chip_read(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 *val) { int ret; - if (sw_addr == 0) - return mdiobus_read_nested(bus, addr, reg); - /* Wait for the bus to become free. */ - ret = mv88e6xxx_reg_wait_ready(bus, sw_addr); + ret = mv88e6xxx_smi_multi_chip_wait(ps); if (ret < 0) return ret; /* Transmit the read command. */ - ret = mdiobus_write_nested(bus, sw_addr, SMI_CMD, + ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_CMD, SMI_CMD_OP_22_READ | (addr << 5) | reg); if (ret < 0) return ret; /* Wait for the read command to complete. */ - ret = mv88e6xxx_reg_wait_ready(bus, sw_addr); + ret = mv88e6xxx_smi_multi_chip_wait(ps); if (ret < 0) return ret; /* Read the data. */ - ret = mdiobus_read_nested(bus, sw_addr, SMI_DATA); - if (ret < 0) - return ret; - - return ret & 0xffff; -} - -static int _mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg) -{ - int ret; - - assert_reg_lock(ps); - - ret = __mv88e6xxx_reg_read(ps->bus, ps->sw_addr, addr, reg); + ret = mdiobus_read_nested(ps->bus, ps->sw_addr, SMI_DATA); if (ret < 0) return ret; - dev_dbg(ps->dev, "<- addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", - addr, reg, ret); - - return ret; -} + *val = ret & 0xffff; -static int mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, int addr, - int reg) -{ - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_reg_read(ps, addr, reg); - mutex_unlock(&ps->reg_lock); - - return ret; + return 0; } -static int __mv88e6xxx_reg_write(struct mii_bus *bus, int sw_addr, int addr, - int reg, u16 val) +static int mv88e6xxx_smi_multi_chip_write(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 val) { int ret; - if (sw_addr == 0) - return mdiobus_write_nested(bus, addr, reg, val); - /* Wait for the bus to become free. */ - ret = mv88e6xxx_reg_wait_ready(bus, sw_addr); + ret = mv88e6xxx_smi_multi_chip_wait(ps); if (ret < 0) return ret; /* Transmit the data to write. */ - ret = mdiobus_write_nested(bus, sw_addr, SMI_DATA, val); + ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_DATA, val); if (ret < 0) return ret; /* Transmit the write command. */ - ret = mdiobus_write_nested(bus, sw_addr, SMI_CMD, + ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_CMD, SMI_CMD_OP_22_WRITE | (addr << 5) | reg); if (ret < 0) return ret; /* Wait for the write command to complete. */ - ret = mv88e6xxx_reg_wait_ready(bus, sw_addr); + ret = mv88e6xxx_smi_multi_chip_wait(ps); if (ret < 0) return ret; return 0; } -static int _mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, - int reg, u16 val) +static const struct mv88e6xxx_ops mv88e6xxx_smi_multi_chip_ops = { + .read = mv88e6xxx_smi_multi_chip_read, + .write = mv88e6xxx_smi_multi_chip_write, +}; + +static int mv88e6xxx_read(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 *val) +{ + int err; + + assert_reg_lock(ps); + + err = mv88e6xxx_smi_read(ps, addr, reg, val); + if (err) + return err; + + dev_dbg(ps->dev, "<- addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", + addr, reg, *val); + + return 0; +} + +static int mv88e6xxx_write(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 val) { + int err; + assert_reg_lock(ps); + err = mv88e6xxx_smi_write(ps, addr, reg, val); + if (err) + return err; + dev_dbg(ps->dev, "-> addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", addr, reg, val); - return __mv88e6xxx_reg_write(ps->bus, ps->sw_addr, addr, reg, val); + return 0; +} + +static int _mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, + int addr, int reg) +{ + u16 val; + int err; + + err = mv88e6xxx_read(ps, addr, reg, &val); + if (err) + return err; + + return val; +} + +static int mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, int addr, + int reg) +{ + int ret; + + mutex_lock(&ps->reg_lock); + ret = _mv88e6xxx_reg_read(ps, addr, reg); + mutex_unlock(&ps->reg_lock); + + return ret; +} + +static int _mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, + int reg, u16 val) +{ + return mv88e6xxx_write(ps, addr, reg, val); } static int mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, @@ -3666,6 +3745,13 @@ static int mv88e6xxx_smi_init(struct mv88e6xxx_priv_state *ps, if (sw_addr & 0x1) return -EINVAL; + if (sw_addr == 0) + ps->smi_ops = &mv88e6xxx_smi_single_chip_ops; + else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_MULTI_CHIP)) + ps->smi_ops = &mv88e6xxx_smi_multi_chip_ops; + else + return -EINVAL; + ps->bus = bus; ps->sw_addr = sw_addr; diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h index 8e6fe6b..a94acd8 100644 --- a/drivers/net/dsa/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx.h @@ -387,6 +387,12 @@ enum mv88e6xxx_cap { */ MV88E6XXX_CAP_EEPROM, + /* Multi-chip Addressing Mode. + * Some chips require an indirect SMI access when their SMI device + * address is not zero. See SMI_CMD and SMI_DATA. + */ + MV88E6XXX_CAP_MULTI_CHIP, + /* Port State Filtering for 802.1D Spanning Tree. * See PORT_CONTROL_STATE_* values in the PORT_CONTROL register. */ @@ -439,6 +445,7 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_ATU BIT(MV88E6XXX_CAP_ATU) #define MV88E6XXX_FLAG_EEE BIT(MV88E6XXX_CAP_EEE) #define MV88E6XXX_FLAG_EEPROM BIT(MV88E6XXX_CAP_EEPROM) +#define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) #define MV88E6XXX_FLAG_PORTSTATE BIT(MV88E6XXX_CAP_PORTSTATE) #define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) #define MV88E6XXX_FLAG_PPU_ACTIVE BIT(MV88E6XXX_CAP_PPU_ACTIVE) @@ -452,25 +459,29 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAGS_FAMILY_6095 \ (MV88E6XXX_FLAG_ATU | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_VLANTABLE | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6097 \ (MV88E6XXX_FLAG_ATU | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_VLANTABLE | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6165 \ - (MV88E6XXX_FLAG_STU | \ + (MV88E6XXX_FLAG_MULTI_CHIP | \ + MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_SWITCH_MAC | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6185 \ (MV88E6XXX_FLAG_ATU | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_VLANTABLE | \ MV88E6XXX_FLAG_VTU) @@ -479,6 +490,7 @@ enum mv88e6xxx_cap { (MV88E6XXX_FLAG_ATU | \ MV88E6XXX_FLAG_EEE | \ MV88E6XXX_FLAG_EEPROM | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PORTSTATE | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ @@ -490,6 +502,7 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAGS_FAMILY_6351 \ (MV88E6XXX_FLAG_ATU | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PORTSTATE | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ @@ -503,6 +516,7 @@ enum mv88e6xxx_cap { (MV88E6XXX_FLAG_ATU | \ MV88E6XXX_FLAG_EEE | \ MV88E6XXX_FLAG_EEPROM | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PORTSTATE | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ @@ -542,6 +556,8 @@ struct mv88e6xxx_vtu_stu_entry { u8 data[DSA_MAX_PORTS]; }; +struct mv88e6xxx_ops; + struct mv88e6xxx_priv_port { struct net_device *bridge_dev; }; @@ -561,6 +577,7 @@ struct mv88e6xxx_priv_state { /* The MII bus and the address on the bus that is used to * communication with the switch */ + const struct mv88e6xxx_ops *smi_ops; struct mii_bus *bus; int sw_addr; @@ -606,6 +623,13 @@ struct mv88e6xxx_priv_state { struct mii_bus *mdio_bus; }; +struct mv88e6xxx_ops { + int (*read)(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 *val); + int (*write)(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 val); +}; + enum stat_type { BANK0, BANK1, -- cgit v0.10.2 From 0b03fd8528f7bf00b5c058212d69f92cf123bf30 Mon Sep 17 00:00:00 2001 From: Qianqian Xie Date: Tue, 21 Jun 2016 11:56:21 +0800 Subject: net: hns: bug fix of ge reset sequence The bit fileds of PPE reset register are different between HNS v1 and HNS v2, but the current procedure just only match HNS v1. Here is a patch to fix it. Signed-off-by: Kejian Yan Signed-off-by: Qianqian Xie Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c index 96cb628..09e60d6 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c @@ -271,7 +271,11 @@ static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, } } else { reg_val_1 = 0x15540 << dsaf_dev->reset_offset; - reg_val_2 = 0x100 << dsaf_dev->reset_offset; + + if (AE_IS_VER1(dsaf_dev->dsaf_ver)) + reg_val_2 = 0x100 << dsaf_dev->reset_offset; + else + reg_val_2 = 0x40 << dsaf_dev->reset_offset; if (!dereset) { dsaf_write_sub(dsaf_dev, DSAF_SUB_SC_GE_RESET_REQ1_REG, -- cgit v0.10.2 From 14ae335a8599d735d19f600c29c6249dd2c3f7a8 Mon Sep 17 00:00:00 2001 From: Qianqian Xie Date: Tue, 21 Jun 2016 11:56:22 +0800 Subject: net: hns: fix hns dsaf v1 dont support tx_pause close For service port, hns dsaf v1 support to close tx_pause. However, the port will be invalid when it run command ethtool to close tx_pause. This patch will fix it. Signed-off-by: Qianqian Xie Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index ac03c4a..422f97d 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -1114,10 +1114,10 @@ int hns_dsaf_set_rx_mac_pause_en(struct dsaf_device *dsaf_dev, int mac_id, u32 en) { if (AE_IS_VER1(dsaf_dev->dsaf_ver)) { - if (!en) + if (!en) { dev_err(dsaf_dev->dev, "dsafv1 can't close rx_pause!\n"); - - return -EINVAL; + return -EINVAL; + } } dsaf_set_dev_bit(dsaf_dev, DSAF_PAUSE_CFG_REG + mac_id * 4, -- cgit v0.10.2 From 8379f0a8ea84f1a904e116b65751681a48a763ea Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Tue, 21 Jun 2016 11:56:23 +0800 Subject: net: hns: add skb_reset_mac_header() after skb being alloc HNS receives a packet without doing anything, but it should call skb_reset_mac_header() to initialize the header before using eth_hdr(). Fixes: 0d6b425a3773c3445b0f51b2f333821beaacb619 Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index ad742a6..15200e4 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -600,6 +600,7 @@ static int hns_nic_poll_rx_skb(struct hns_nic_ring_data *ring_data, ring->stats.sw_err_cnt++; return -ENOMEM; } + skb_reset_mac_header(skb); prefetchw(skb->data); length = le16_to_cpu(desc->rx.pkt_len); -- cgit v0.10.2 From f7211729bd74f26d4b09839a11ab1892d9cd33ea Mon Sep 17 00:00:00 2001 From: Qianqian Xie Date: Tue, 21 Jun 2016 11:56:24 +0800 Subject: net: hns: typo fix of annotation info for hns_nic_reset_subtask() The annotation info for hns_nic_reset_subtask() should be "for resetting subtask" instead of "for resetting suntask". Signed-off-by: Qianqian Xie Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index 15200e4..b978db4 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -1633,7 +1633,7 @@ static void hns_nic_dump(struct hns_nic_priv *priv) } } -/* for resetting suntask*/ +/* for resetting subtask */ static void hns_nic_reset_subtask(struct hns_nic_priv *priv) { enum hnae_port_type type = priv->ae_handle->port_type; -- cgit v0.10.2 From f6c2df1e5b913f9dec44ee013f4552840864c7f0 Mon Sep 17 00:00:00 2001 From: Qianqian Xie Date: Tue, 21 Jun 2016 11:56:25 +0800 Subject: net: hns: Remove unnecessary device resource free The driver uses devm_ioremap_resource, it will unmap the map automatically, remove the unnecessary the resource free. Signed-off-by: Qianqian Xie Reported-by: Kefeng Wang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index 422f97d..b8b2ff9 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -176,7 +176,7 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) desc_num > HNS_DSAF_MAX_DESC_CNT) { dev_err(dsaf_dev->dev, "get desc-num(%d) fail, ret=%d!\n", desc_num, ret); - goto unmap_base_addr; + return -EINVAL; } dsaf_dev->desc_num = desc_num; @@ -192,7 +192,7 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) if (ret < 0) { dev_err(dsaf_dev->dev, "get buf-size fail, ret=%d!\r\n", ret); - goto unmap_base_addr; + return ret; } dsaf_dev->buf_size = buf_size; @@ -200,7 +200,7 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) if (dsaf_dev->buf_size_type < 0) { dev_err(dsaf_dev->dev, "buf_size(%d) is wrong!\n", buf_size); - goto unmap_base_addr; + return -EINVAL; } dsaf_dev->misc_op = hns_misc_op_get(dsaf_dev); @@ -213,32 +213,6 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) dev_err(dsaf_dev->dev, "set mask to 64bit fail!\n"); return 0; - -unmap_base_addr: - if (dsaf_dev->io_base) - iounmap(dsaf_dev->io_base); - if (dsaf_dev->ppe_base) - iounmap(dsaf_dev->ppe_base); - if (dsaf_dev->sds_base) - iounmap(dsaf_dev->sds_base); - if (dsaf_dev->sc_base) - iounmap(dsaf_dev->sc_base); - return ret; -} - -static void hns_dsaf_free_cfg(struct dsaf_device *dsaf_dev) -{ - if (dsaf_dev->io_base) - iounmap(dsaf_dev->io_base); - - if (dsaf_dev->ppe_base) - iounmap(dsaf_dev->ppe_base); - - if (dsaf_dev->sds_base) - iounmap(dsaf_dev->sds_base); - - if (dsaf_dev->sc_base) - iounmap(dsaf_dev->sc_base); } /** @@ -2645,7 +2619,7 @@ static int hns_dsaf_probe(struct platform_device *pdev) ret = hns_dsaf_init(dsaf_dev); if (ret) - goto free_cfg; + goto free_dev; ret = hns_mac_init(dsaf_dev); if (ret) @@ -2670,9 +2644,6 @@ uninit_mac: uninit_dsaf: hns_dsaf_free(dsaf_dev); -free_cfg: - hns_dsaf_free_cfg(dsaf_dev); - free_dev: hns_dsaf_free_dev(dsaf_dev); @@ -2695,8 +2666,6 @@ static int hns_dsaf_remove(struct platform_device *pdev) hns_dsaf_free(dsaf_dev); - hns_dsaf_free_cfg(dsaf_dev); - hns_dsaf_free_dev(dsaf_dev); return 0; -- cgit v0.10.2 From 39c944179483a1f81e9b6c382edc667bbd24c4a6 Mon Sep 17 00:00:00 2001 From: Qianqian Xie Date: Tue, 21 Jun 2016 11:56:26 +0800 Subject: net: hns: fix the error info when dma_set_mask_and_coherent fail The error info should be printed as "set mask to 64bit fail!" instead of "set mask to 32bit fail!" in dma_set_mask_and_coherent(). Signed-off-by: Qianqian Xie Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index b978db4..00e529f 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -1972,7 +1972,7 @@ static int hns_nic_dev_probe(struct platform_device *pdev) if (!dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64))) dev_dbg(dev, "set mask to 64bit\n"); else - dev_err(dev, "set mask to 32bit fail!\n"); + dev_err(dev, "set mask to 64bit fail!\n"); /* carrier off reporting is important to ethtool even BEFORE open */ netif_carrier_off(ndev); -- cgit v0.10.2 From 89a6b1aae844a89f6724b22c63803bb6fbe5677b Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Tue, 21 Jun 2016 11:56:27 +0800 Subject: net: hns: select Hilink before serdes loopback for HNS V2 As Hilink3 and Hilink4 use the same xge training and xge u adaptor for HNSv2, it needs to select which Hilink to be set before relative serdes being configed. The hilink_access_sel is the register to do that. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c index 09e60d6..aff9d77 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c @@ -435,11 +435,6 @@ int hns_mac_get_sfp_prsnt(struct hns_mac_cb *mac_cb, int *sfp_prsnt) */ static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en) { - /* port 0-3 hilink4 base is serdes_vaddr + 0x00280000 - * port 4-7 hilink3 base is serdes_vaddr + 0x00200000 - */ - u8 *base_addr = (u8 *)mac_cb->serdes_vaddr + - (mac_cb->mac_id <= 3 ? 0x00280000 : 0x00200000); const u8 lane_id[] = { 0, /* mac 0 -> lane 0 */ 1, /* mac 1 -> lane 1 */ @@ -465,11 +460,30 @@ static int hns_mac_config_sds_loopback(struct hns_mac_cb *mac_cb, bool en) } if (mac_cb->serdes_ctrl) { - u32 origin = dsaf_read_syscon(mac_cb->serdes_ctrl, reg_offset); + u32 origin; + + if (!AE_IS_VER1(mac_cb->dsaf_dev->dsaf_ver)) { +#define HILINK_ACCESS_SEL_CFG 0x40008 + /* hilink4 & hilink3 use the same xge training and + * xge u adaptor. There is a hilink access sel cfg + * register to select which one to be configed + */ + if ((!HNS_DSAF_IS_DEBUG(mac_cb->dsaf_dev)) && + (mac_cb->mac_id <= 3)) + dsaf_write_syscon(mac_cb->serdes_ctrl, + HILINK_ACCESS_SEL_CFG, 0); + else + dsaf_write_syscon(mac_cb->serdes_ctrl, + HILINK_ACCESS_SEL_CFG, 3); + } + + origin = dsaf_read_syscon(mac_cb->serdes_ctrl, reg_offset); dsaf_set_field(origin, 1ull << 10, 10, en); dsaf_write_syscon(mac_cb->serdes_ctrl, reg_offset, origin); } else { + u8 *base_addr = (u8 *)mac_cb->serdes_vaddr + + (mac_cb->mac_id <= 3 ? 0x00280000 : 0x00200000); dsaf_set_reg_field(base_addr, reg_offset, 1ull << 10, 10, en); } -- cgit v0.10.2 From cba80bdea72cb24a93bd00286673991ffb4a6b31 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Tue, 21 Jun 2016 11:56:28 +0800 Subject: net: hns: fix ethtool loopback fail bug When run ethtool cmd(ethtool -t ethx) again and again for a long time, it will be probabilistically fail. The PHYs' registers may be on different pages, so it must be switch to the right page before setting PHYs' registers. And __lb_up() calls phy_start() to startup the PHYs device, but this function may change Copper Control Register(Page 0, Register 0) to an other value. It would cause phy loopback test fail. if we remove phy_start(), we have to remove the relative phy_stop(), phy_disconnect() when doing phy loopback to keep the phy stay in right status. Reported-by: hejun Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c index a809f52..5b3dccb 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c @@ -242,6 +242,7 @@ static const char hns_nic_test_strs[][ETH_GSTRING_LEN] = { static int hns_nic_config_phy_loopback(struct phy_device *phy_dev, u8 en) { #define COPPER_CONTROL_REG 0 +#define PHY_POWER_DOWN BIT(11) #define PHY_LOOP_BACK BIT(14) u16 val = 0; @@ -252,33 +253,40 @@ static int hns_nic_config_phy_loopback(struct phy_device *phy_dev, u8 en) /* speed : 1000M */ phy_write(phy_dev, HNS_PHY_PAGE_REG, 2); phy_write(phy_dev, 21, 0x1046); + + phy_write(phy_dev, HNS_PHY_PAGE_REG, 0); /* Force Master */ phy_write(phy_dev, 9, 0x1F00); + /* Soft-reset */ phy_write(phy_dev, 0, 0x9140); /* If autoneg disabled,two soft-reset operations */ phy_write(phy_dev, 0, 0x9140); - phy_write(phy_dev, 22, 0xFA); + + phy_write(phy_dev, HNS_PHY_PAGE_REG, 0xFA); /* Default is 0x0400 */ phy_write(phy_dev, 1, 0x418); /* Force 1000M Link, Default is 0x0200 */ phy_write(phy_dev, 7, 0x20C); - phy_write(phy_dev, 22, 0); + phy_write(phy_dev, HNS_PHY_PAGE_REG, 0); - /* Enable MAC loop-back */ + /* Enable PHY loop-back */ val = phy_read(phy_dev, COPPER_CONTROL_REG); val |= PHY_LOOP_BACK; + val &= ~PHY_POWER_DOWN; phy_write(phy_dev, COPPER_CONTROL_REG, val); } else { - phy_write(phy_dev, 22, 0xFA); + phy_write(phy_dev, HNS_PHY_PAGE_REG, 0xFA); phy_write(phy_dev, 1, 0x400); phy_write(phy_dev, 7, 0x200); - phy_write(phy_dev, 22, 0); + phy_write(phy_dev, HNS_PHY_PAGE_REG, 0); + phy_write(phy_dev, 9, 0xF00); val = phy_read(phy_dev, COPPER_CONTROL_REG); val &= ~PHY_LOOP_BACK; + val |= PHY_POWER_DOWN; phy_write(phy_dev, COPPER_CONTROL_REG, val); } return 0; @@ -339,28 +347,16 @@ static int __lb_up(struct net_device *ndev, hns_nic_net_reset(ndev); - if (priv->phy) { - phy_disconnect(priv->phy); - msleep(100); - - ret = hns_nic_init_phy(ndev, h); - if (ret) - return ret; - } - ret = __lb_setup(ndev, loop_mode); if (ret) return ret; - msleep(100); + msleep(200); ret = h->dev->ops->start ? h->dev->ops->start(h) : 0; if (ret) return ret; - if (priv->phy) - phy_start(priv->phy); - /* link adjust duplex*/ if (priv->ae_handle->phy_if != PHY_INTERFACE_MODE_XGMII) speed = 1000; @@ -561,9 +557,6 @@ static int __lb_down(struct net_device *ndev) __func__, ret); - if (priv->phy) - phy_stop(priv->phy); - if (h->dev->ops->stop) h->dev->ops->stop(h); -- cgit v0.10.2 From bb7189dc78d643b2707483763664a02ac017ac59 Mon Sep 17 00:00:00 2001 From: Qianqian Xie Date: Tue, 21 Jun 2016 11:56:29 +0800 Subject: net: hns: fix the wrong speed for bond For debug-ports,there are two non-synchronized processes: Speed-Auto-Negotiation and Link-Update-Status. The two processes are towed by two different state machines. Bond reads the speed when link up, but the speed maybe not update the right value at that time.That make for bond's wrong speed. Thus only one state machine should be used and if phy_state_machine is used, it does not need to do hns_nic_update_link_status(). Signed-off-by: Qianqian Xie Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index 00e529f..cef9d12 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -991,8 +991,26 @@ static void hns_nic_adjust_link(struct net_device *ndev) { struct hns_nic_priv *priv = netdev_priv(ndev); struct hnae_handle *h = priv->ae_handle; + int state = 1; + + if (priv->phy) { + h->dev->ops->adjust_link(h, ndev->phydev->speed, + ndev->phydev->duplex); + state = priv->phy->link; + } + state = state && h->dev->ops->get_status(h); - h->dev->ops->adjust_link(h, ndev->phydev->speed, ndev->phydev->duplex); + if (state != priv->link) { + if (state) { + netif_carrier_on(ndev); + netif_tx_wake_all_queues(ndev); + netdev_info(ndev, "link up\n"); + } else { + netif_carrier_off(ndev); + netdev_info(ndev, "link down\n"); + } + priv->link = state; + } } /** @@ -1577,27 +1595,14 @@ static void hns_nic_update_link_status(struct net_device *netdev) struct hns_nic_priv *priv = netdev_priv(netdev); struct hnae_handle *h = priv->ae_handle; - int state = 1; - if (priv->phy) { - if (!genphy_update_link(priv->phy)) - state = priv->phy->link; - else - state = 0; - } - state = state && h->dev->ops->get_status(h); + if (h->phy_dev) { + if (h->phy_if != PHY_INTERFACE_MODE_XGMII) + return; - if (state != priv->link) { - if (state) { - netif_carrier_on(netdev); - netif_tx_wake_all_queues(netdev); - netdev_info(netdev, "link up\n"); - } else { - netif_carrier_off(netdev); - netdev_info(netdev, "link down\n"); - } - priv->link = state; + (void)genphy_read_status(h->phy_dev); } + hns_nic_adjust_link(netdev); } /* for dumping key regs*/ diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c index 5b3dccb..564ae1e 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c @@ -49,7 +49,7 @@ static u32 hns_nic_get_link(struct net_device *net_dev) h = priv->ae_handle; if (priv->phy) { - if (!genphy_update_link(priv->phy)) + if (!genphy_read_status(priv->phy)) link_stat = priv->phy->link; else link_stat = 0; -- cgit v0.10.2 From 379d3954923537df4d895cb9d1ce36d8a7888d17 Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Tue, 21 Jun 2016 11:56:30 +0800 Subject: net: hns: bugfix about pfc pause frame statistics For SoC hip06, PFC pause handled in dsaf, while hip05 in XGMAC, so change the statistics of pfc pause in dsaf and remove the old pfc pause frame statistics. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c index d37b778..b97cc75 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c @@ -587,6 +587,7 @@ void hns_ae_get_strings(struct hnae_handle *handle, int idx; struct hns_mac_cb *mac_cb; struct hns_ppe_cb *ppe_cb; + struct dsaf_device *dsaf_dev = hns_ae_get_dsaf_dev(handle->dev); u8 *p = data; struct hnae_vf_cb *vf_cb; @@ -609,13 +610,14 @@ void hns_ae_get_strings(struct hnae_handle *handle, p += ETH_GSTRING_LEN * hns_mac_get_sset_count(mac_cb, stringset); if (mac_cb->mac_type == HNAE_PORT_SERVICE) - hns_dsaf_get_strings(stringset, p, port); + hns_dsaf_get_strings(stringset, p, port, dsaf_dev); } int hns_ae_get_sset_count(struct hnae_handle *handle, int stringset) { u32 sset_count = 0; struct hns_mac_cb *mac_cb; + struct dsaf_device *dsaf_dev = hns_ae_get_dsaf_dev(handle->dev); assert(handle); @@ -626,7 +628,7 @@ int hns_ae_get_sset_count(struct hnae_handle *handle, int stringset) sset_count += hns_mac_get_sset_count(mac_cb, stringset); if (mac_cb->mac_type == HNAE_PORT_SERVICE) - sset_count += hns_dsaf_get_sset_count(stringset); + sset_count += hns_dsaf_get_sset_count(dsaf_dev, stringset); return sset_count; } diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index b8b2ff9..0edea9c 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -2096,11 +2096,24 @@ void hns_dsaf_fix_mac_mode(struct hns_mac_cb *mac_cb) hns_dsaf_port_work_rate_cfg(dsaf_dev, mac_id, mode); } +static u32 hns_dsaf_get_inode_prio_reg(int index) +{ + int base_index, offset; + u32 base_addr = DSAF_INODE_IN_PRIO_PAUSE_BASE_REG; + + base_index = (index + 1) / DSAF_REG_PER_ZONE; + offset = (index + 1) % DSAF_REG_PER_ZONE; + + return base_addr + DSAF_INODE_IN_PRIO_PAUSE_BASE_OFFSET * base_index + + DSAF_INODE_IN_PRIO_PAUSE_OFFSET * offset; +} + void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 node_num) { struct dsaf_hw_stats *hw_stats = &dsaf_dev->hw_stats[node_num]; bool is_ver1 = AE_IS_VER1(dsaf_dev->dsaf_ver); + int i; u32 reg_tmp; hw_stats->pad_drop += dsaf_read_dev(dsaf_dev, @@ -2135,6 +2148,18 @@ void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 node_num) hw_stats->stp_drop += dsaf_read_dev(dsaf_dev, DSAF_INODE_IN_DATA_STP_DISC_0_REG + 0x80 * (u64)node_num); + /* pfc pause frame statistics stored in dsaf inode*/ + if ((node_num < DSAF_SERVICE_NW_NUM) && !is_ver1) { + for (i = 0; i < DSAF_PRIO_NR; i++) { + reg_tmp = hns_dsaf_get_inode_prio_reg(i); + hw_stats->rx_pfc[i] += dsaf_read_dev(dsaf_dev, + reg_tmp + 0x4 * (u64)node_num); + hw_stats->tx_pfc[i] += dsaf_read_dev(dsaf_dev, + DSAF_XOD_XGE_PFC_PRIO_CNT_BASE_REG + + DSAF_XOD_XGE_PFC_PRIO_CNT_OFFSET * i + + 0xF0 * (u64)node_num); + } + } hw_stats->tx_pkts += dsaf_read_dev(dsaf_dev, DSAF_XOD_RCVPKT_CNT_0_REG + 0x90 * (u64)node_num); } @@ -2472,9 +2497,12 @@ void hns_dsaf_get_regs(struct dsaf_device *ddev, u32 port, void *data) p[i] = 0xdddddddd; } -static char *hns_dsaf_get_node_stats_strings(char *data, int node) +static char *hns_dsaf_get_node_stats_strings(char *data, int node, + struct dsaf_device *dsaf_dev) { char *buff = data; + int i; + bool is_ver1 = AE_IS_VER1(dsaf_dev->dsaf_ver); snprintf(buff, ETH_GSTRING_LEN, "innod%d_pad_drop_pkts", node); buff = buff + ETH_GSTRING_LEN; @@ -2502,6 +2530,18 @@ static char *hns_dsaf_get_node_stats_strings(char *data, int node) buff = buff + ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_stp_drop_pkts", node); buff = buff + ETH_GSTRING_LEN; + if ((node < DSAF_SERVICE_NW_NUM) && (!is_ver1)) { + for (i = 0; i < DSAF_PRIO_NR; i++) { + snprintf(buff, ETH_GSTRING_LEN, + "inod%d_pfc_prio%d_pkts", node, i); + buff = buff + ETH_GSTRING_LEN; + } + for (i = 0; i < DSAF_PRIO_NR; i++) { + snprintf(buff, ETH_GSTRING_LEN, + "onod%d_pfc_prio%d_pkts", node, i); + buff = buff + ETH_GSTRING_LEN; + } + } snprintf(buff, ETH_GSTRING_LEN, "onnod%d_tx_pkts", node); buff = buff + ETH_GSTRING_LEN; @@ -2512,7 +2552,9 @@ static u64 *hns_dsaf_get_node_stats(struct dsaf_device *ddev, u64 *data, int node_num) { u64 *p = data; + int i; struct dsaf_hw_stats *hw_stats = &ddev->hw_stats[node_num]; + bool is_ver1 = AE_IS_VER1(ddev->dsaf_ver); p[0] = hw_stats->pad_drop; p[1] = hw_stats->man_pkts; @@ -2527,8 +2569,16 @@ static u64 *hns_dsaf_get_node_stats(struct dsaf_device *ddev, u64 *data, p[10] = hw_stats->local_addr_false; p[11] = hw_stats->vlan_drop; p[12] = hw_stats->stp_drop; - p[13] = hw_stats->tx_pkts; + if ((node_num < DSAF_SERVICE_NW_NUM) && (!is_ver1)) { + for (i = 0; i < DSAF_PRIO_NR; i++) { + p[13 + i] = hw_stats->rx_pfc[i]; + p[13 + i + DSAF_PRIO_NR] = hw_stats->tx_pfc[i]; + } + p[29] = hw_stats->tx_pkts; + return &p[30]; + } + p[13] = hw_stats->tx_pkts; return &p[14]; } @@ -2556,11 +2606,16 @@ void hns_dsaf_get_stats(struct dsaf_device *ddev, u64 *data, int port) *@stringset: type of values in data *return dsaf string name count */ -int hns_dsaf_get_sset_count(int stringset) +int hns_dsaf_get_sset_count(struct dsaf_device *dsaf_dev, int stringset) { - if (stringset == ETH_SS_STATS) - return DSAF_STATIC_NUM; + bool is_ver1 = AE_IS_VER1(dsaf_dev->dsaf_ver); + if (stringset == ETH_SS_STATS) { + if (is_ver1) + return DSAF_STATIC_NUM; + else + return DSAF_V2_STATIC_NUM; + } return 0; } @@ -2570,7 +2625,8 @@ int hns_dsaf_get_sset_count(int stringset) *@data:strings name value *@port:port index */ -void hns_dsaf_get_strings(int stringset, u8 *data, int port) +void hns_dsaf_get_strings(int stringset, u8 *data, int port, + struct dsaf_device *dsaf_dev) { char *buff = (char *)data; int node = port; @@ -2579,11 +2635,11 @@ void hns_dsaf_get_strings(int stringset, u8 *data, int port) return; /* for ge/xge node info */ - buff = hns_dsaf_get_node_stats_strings(buff, node); + buff = hns_dsaf_get_node_stats_strings(buff, node, dsaf_dev); /* for ppe node info */ node = port + DSAF_PPE_INODE_BASE; - (void)hns_dsaf_get_node_stats_strings(buff, node); + (void)hns_dsaf_get_node_stats_strings(buff, node, dsaf_dev); } /** diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h index 2e55b3c..00a13de 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h @@ -39,6 +39,9 @@ struct hns_mac_cb; #define DSAF_DUMP_REGS_NUM 504 #define DSAF_STATIC_NUM 28 +#define DSAF_V2_STATIC_NUM 44 +#define DSAF_PRIO_NR 8 +#define DSAF_REG_PER_ZONE 3 #define DSAF_STATS_READ(p, offset) (*((u64 *)((u8 *)(p) + (offset)))) #define HNS_DSAF_IS_DEBUG(dev) (dev->dsaf_mode == DSAF_MODE_DISABLE_SP) @@ -176,6 +179,8 @@ struct dsaf_hw_stats { u64 local_addr_false; u64 vlan_drop; u64 stp_drop; + u64 rx_pfc[DSAF_PRIO_NR]; + u64 tx_pfc[DSAF_PRIO_NR]; u64 tx_pkts; }; @@ -417,9 +422,10 @@ void hns_dsaf_ae_uninit(struct dsaf_device *dsaf_dev); void hns_dsaf_update_stats(struct dsaf_device *dsaf_dev, u32 inode_num); -int hns_dsaf_get_sset_count(int stringset); +int hns_dsaf_get_sset_count(struct dsaf_device *dsaf_dev, int stringset); void hns_dsaf_get_stats(struct dsaf_device *ddev, u64 *data, int port); -void hns_dsaf_get_strings(int stringset, u8 *data, int port); +void hns_dsaf_get_strings(int stringset, u8 *data, int port, + struct dsaf_device *dsaf_dev); void hns_dsaf_get_regs(struct dsaf_device *ddev, u32 port, void *data); int hns_dsaf_get_regs_count(void); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h index 7c3b510..e35d0cb 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h @@ -166,6 +166,9 @@ #define DSAF_INODE_GE_FC_EN_0_REG 0x1B00 #define DSAF_INODE_VC0_IN_PKT_NUM_0_REG 0x1B50 #define DSAF_INODE_VC1_IN_PKT_NUM_0_REG 0x1C00 +#define DSAF_INODE_IN_PRIO_PAUSE_BASE_REG 0x1C00 +#define DSAF_INODE_IN_PRIO_PAUSE_BASE_OFFSET 0x100 +#define DSAF_INODE_IN_PRIO_PAUSE_OFFSET 0x50 #define DSAF_SBM_CFG_REG_0_REG 0x2000 #define DSAF_SBM_BP_CFG_0_XGE_REG_0_REG 0x2004 @@ -232,6 +235,8 @@ #define DSAF_XOD_ROCEE_RCVIN0_CNT_0_REG 0x3074 #define DSAF_XOD_ROCEE_RCVIN1_CNT_0_REG 0x3078 #define DSAF_XOD_FIFO_STATUS_0_REG 0x307C +#define DSAF_XOD_XGE_PFC_PRIO_CNT_BASE_REG 0x3A00 +#define DSAF_XOD_XGE_PFC_PRIO_CNT_OFFSET 0x4 #define DSAF_VOQ_ECC_INVERT_EN_0_REG 0x4004 #define DSAF_VOQ_SRAM_PKT_NUM_0_REG 0x4008 -- cgit v0.10.2 From b76238168991f8b000c90ba053f6f125e4cf1e8c Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Tue, 21 Jun 2016 11:56:31 +0800 Subject: net: hns: add spin lock for tcam table operation This patch adds spin lock for tcam table operation, there maybe a race condition happens when more than one thread try to change the tcam talbe entries. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index 0edea9c..0958ceb 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -860,6 +860,8 @@ static void hns_dsaf_single_line_tbl_cfg( struct dsaf_device *dsaf_dev, u32 address, struct dsaf_tbl_line_cfg *ptbl_line) { + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_line_addr_cfg(dsaf_dev, address); @@ -868,6 +870,8 @@ static void hns_dsaf_single_line_tbl_cfg( /*Write Plus*/ hns_dsaf_tbl_line_pul(dsaf_dev); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -881,6 +885,8 @@ static void hns_dsaf_tcam_uc_cfg( struct dsaf_tbl_tcam_data *ptbl_tcam_data, struct dsaf_tbl_tcam_ucast_cfg *ptbl_tcam_ucast) { + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address); /*Write Tcam Data*/ @@ -889,6 +895,8 @@ static void hns_dsaf_tcam_uc_cfg( hns_dsaf_tbl_tcam_ucast_cfg(dsaf_dev, ptbl_tcam_ucast); /*Write Plus*/ hns_dsaf_tbl_tcam_data_ucast_pul(dsaf_dev); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -903,6 +911,8 @@ static void hns_dsaf_tcam_mc_cfg( struct dsaf_tbl_tcam_data *ptbl_tcam_data, struct dsaf_tbl_tcam_mcast_cfg *ptbl_tcam_mcast) { + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address); /*Write Tcam Data*/ @@ -911,6 +921,8 @@ static void hns_dsaf_tcam_mc_cfg( hns_dsaf_tbl_tcam_mcast_cfg(dsaf_dev, ptbl_tcam_mcast); /*Write Plus*/ hns_dsaf_tbl_tcam_data_mcast_pul(dsaf_dev); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -920,6 +932,8 @@ static void hns_dsaf_tcam_mc_cfg( */ static void hns_dsaf_tcam_mc_invld(struct dsaf_device *dsaf_dev, u32 address) { + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address); @@ -932,6 +946,8 @@ static void hns_dsaf_tcam_mc_invld(struct dsaf_device *dsaf_dev, u32 address) /*Write Plus*/ hns_dsaf_tbl_tcam_mcast_pul(dsaf_dev); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -949,6 +965,8 @@ static void hns_dsaf_tcam_uc_get( u32 tcam_read_data0; u32 tcam_read_data4; + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address); @@ -981,6 +999,8 @@ static void hns_dsaf_tcam_uc_get( DSAF_TBL_UCAST_CFG1_OUT_PORT_S); ptbl_tcam_ucast->tbl_ucast_dvc = dsaf_get_bit(tcam_read_data0, DSAF_TBL_UCAST_CFG1_DVC_S); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -997,6 +1017,8 @@ static void hns_dsaf_tcam_mc_get( { u32 data_tmp; + spin_lock_bh(&dsaf_dev->tcam_lock); + /*Write Addr*/ hns_dsaf_tbl_tcam_addr_cfg(dsaf_dev, address); @@ -1027,6 +1049,8 @@ static void hns_dsaf_tcam_mc_get( ptbl_tcam_mcast->tbl_mcast_port_msk[4] = dsaf_get_field(data_tmp, DSAF_TBL_MCAST_CFG4_VM128_112_M, DSAF_TBL_MCAST_CFG4_VM128_112_S); + + spin_unlock_bh(&dsaf_dev->tcam_lock); } /** @@ -1351,6 +1375,7 @@ static int hns_dsaf_init(struct dsaf_device *dsaf_dev) if (HNS_DSAF_IS_DEBUG(dsaf_dev)) return 0; + spin_lock_init(&dsaf_dev->tcam_lock); ret = hns_dsaf_init_hw(dsaf_dev); if (ret) return ret; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h index 00a13de..1daf018 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h @@ -322,6 +322,8 @@ struct dsaf_device { struct dsaf_hw_stats hw_stats[DSAF_NODE_NUM]; struct dsaf_int_stat int_stat; + /* make sure tcam table config spinlock */ + spinlock_t tcam_lock; }; static inline void *hns_dsaf_dev_priv(const struct dsaf_device *dsaf_dev) -- cgit v0.10.2 From f56c1b3de7a2c7435e991a01350b0b32664ba5fb Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Tue, 21 Jun 2016 11:56:32 +0800 Subject: net: hns: fix bug of getting the wrong tcam data The current driver stores the high bit value of tcam data register to the tcam data low element, stores the low bit value of tcam data register to tcam data high element, this patch fixes this bug. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index 0958ceb..7f5c248 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -975,9 +975,9 @@ static void hns_dsaf_tcam_uc_get( /*read tcam data*/ ptbl_tcam_data->tbl_tcam_data_high - = dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_LOW_0_REG); - ptbl_tcam_data->tbl_tcam_data_low = dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_HIGH_0_REG); + ptbl_tcam_data->tbl_tcam_data_low + = dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_LOW_0_REG); /*read tcam mcast*/ tcam_read_data0 = dsaf_read_dev(dsaf_dev, @@ -1027,9 +1027,9 @@ static void hns_dsaf_tcam_mc_get( /*read tcam data*/ ptbl_tcam_data->tbl_tcam_data_high = - dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_LOW_0_REG); - ptbl_tcam_data->tbl_tcam_data_low = dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_HIGH_0_REG); + ptbl_tcam_data->tbl_tcam_data_low = + dsaf_read_dev(dsaf_dev, DSAF_TBL_TCAM_RDATA_LOW_0_REG); /*read tcam mcast*/ ptbl_tcam_mcast->tbl_mcast_port_msk[0] = -- cgit v0.10.2 From ad59a17f0a8ea31e6235e77fb4dda3cd22978914 Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Tue, 21 Jun 2016 11:56:33 +0800 Subject: net: hns: add get_coalesce_range api for hns This patch adds get_coalesce_range api for hns, it shows range of coalesce usecs and frames that can be set on this interface. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.h b/drivers/net/ethernet/hisilicon/hns/hnae.h index 529cb13..962e445 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.h +++ b/drivers/net/ethernet/hisilicon/hns/hnae.h @@ -473,6 +473,11 @@ struct hnae_ae_ops { int (*set_coalesce_usecs)(struct hnae_handle *handle, u32 timeout); int (*set_coalesce_frames)(struct hnae_handle *handle, u32 coalesce_frames); + void (*get_coalesce_range)(struct hnae_handle *handle, + u32 *tx_frames_low, u32 *rx_frames_low, + u32 *tx_frames_high, u32 *rx_frames_high, + u32 *tx_usecs_low, u32 *rx_usecs_low, + u32 *tx_usecs_high, u32 *rx_usecs_high); void (*set_promisc_mode)(struct hnae_handle *handle, u32 en); int (*get_mac_addr)(struct hnae_handle *handle, void **p); int (*set_mac_addr)(struct hnae_handle *handle, void *p); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c index b97cc75..6b3796f 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c @@ -465,6 +465,30 @@ static int hns_ae_set_coalesce_frames(struct hnae_handle *handle, ring_pair->port_id_in_comm, coalesce_frames); } +static void hns_ae_get_coalesce_range(struct hnae_handle *handle, + u32 *tx_frames_low, u32 *rx_frames_low, + u32 *tx_frames_high, u32 *rx_frames_high, + u32 *tx_usecs_low, u32 *rx_usecs_low, + u32 *tx_usecs_high, u32 *rx_usecs_high) +{ + struct dsaf_device *dsaf_dev; + + dsaf_dev = hns_ae_get_dsaf_dev(handle->dev); + + *tx_frames_low = HNS_RCB_MIN_COALESCED_FRAMES; + *rx_frames_low = HNS_RCB_MIN_COALESCED_FRAMES; + *tx_frames_high = + (dsaf_dev->desc_num - 1 > HNS_RCB_MAX_COALESCED_FRAMES) ? + HNS_RCB_MAX_COALESCED_FRAMES : dsaf_dev->desc_num - 1; + *rx_frames_high = + (dsaf_dev->desc_num - 1 > HNS_RCB_MAX_COALESCED_FRAMES) ? + HNS_RCB_MAX_COALESCED_FRAMES : dsaf_dev->desc_num - 1; + *tx_usecs_low = 0; + *rx_usecs_low = 0; + *tx_usecs_high = HNS_RCB_MAX_COALESCED_USECS; + *rx_usecs_high = HNS_RCB_MAX_COALESCED_USECS; +} + void hns_ae_update_stats(struct hnae_handle *handle, struct net_device_stats *net_stats) { @@ -798,6 +822,7 @@ static struct hnae_ae_ops hns_dsaf_ops = { .get_rx_max_coalesced_frames = hns_ae_get_rx_max_coalesced_frames, .set_coalesce_usecs = hns_ae_set_coalesce_usecs, .set_coalesce_frames = hns_ae_set_coalesce_frames, + .get_coalesce_range = hns_ae_get_coalesce_range, .set_promisc_mode = hns_ae_set_promisc_mode, .set_mac_addr = hns_ae_set_mac_address, .set_mc_addr = hns_ae_set_multicast_one, diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c index 564ae1e..a395ca1 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c @@ -751,6 +751,16 @@ static int hns_get_coalesce(struct net_device *net_dev, &ec->tx_max_coalesced_frames, &ec->rx_max_coalesced_frames); + ops->get_coalesce_range(priv->ae_handle, + &ec->tx_max_coalesced_frames_low, + &ec->rx_max_coalesced_frames_low, + &ec->tx_max_coalesced_frames_high, + &ec->rx_max_coalesced_frames_high, + &ec->tx_coalesce_usecs_low, + &ec->rx_coalesce_usecs_low, + &ec->tx_coalesce_usecs_high, + &ec->rx_coalesce_usecs_high); + return 0; } -- cgit v0.10.2 From 454784d85de372991ad2e473d562d4ad31f4502a Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Tue, 21 Jun 2016 11:56:34 +0800 Subject: net: hns: delete redundancy ring enable operations When network interface is enabled, the ring enable operation is conducted twice. This patch deletes the redundancy code of ring enable, and integrates hnae_ae_ops.toggle_queue_status other functions to hns_ae_start. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.c b/drivers/net/ethernet/hisilicon/hns/hnae.c index 5d3047c..c54c6fa 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.c +++ b/drivers/net/ethernet/hisilicon/hns/hnae.c @@ -400,7 +400,6 @@ int hnae_ae_register(struct hnae_ae_dev *hdev, struct module *owner) if (!hdev->ops || !hdev->ops->get_handle || !hdev->ops->toggle_ring_irq || - !hdev->ops->toggle_queue_status || !hdev->ops->get_status || !hdev->ops->adjust_link) return -EINVAL; diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.h b/drivers/net/ethernet/hisilicon/hns/hnae.h index 962e445..3869322 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.h +++ b/drivers/net/ethernet/hisilicon/hns/hnae.h @@ -454,7 +454,6 @@ struct hnae_ae_ops { int (*get_info)(struct hnae_handle *handle, u8 *auto_neg, u16 *speed, u8 *duplex); void (*toggle_ring_irq)(struct hnae_ring *ring, u32 val); - void (*toggle_queue_status)(struct hnae_queue *queue, u32 val); void (*adjust_link)(struct hnae_handle *handle, int speed, int duplex); int (*set_loopback)(struct hnae_handle *handle, enum hnae_loop loop_mode, int en); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c index 6b3796f..835521b 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c @@ -247,12 +247,21 @@ static void hns_ae_set_tso_stats(struct hnae_handle *handle, int enable) static int hns_ae_start(struct hnae_handle *handle) { int ret; + int k; struct hns_mac_cb *mac_cb = hns_get_mac_cb(handle); ret = hns_mac_vm_config_bc_en(mac_cb, 0, true); if (ret) return ret; + for (k = 0; k < handle->q_num; k++) { + if (AE_IS_VER1(mac_cb->dsaf_dev->dsaf_ver)) + hns_rcb_int_clr_hw(handle->qs[k], + RCB_INT_FLAG_TX | RCB_INT_FLAG_RX); + else + hns_rcbv2_int_clr_hw(handle->qs[k], + RCB_INT_FLAG_TX | RCB_INT_FLAG_RX); + } hns_ae_ring_enable_all(handle, 1); msleep(100); @@ -313,18 +322,6 @@ static void hns_aev2_toggle_ring_irq(struct hnae_ring *ring, u32 mask) hns_rcbv2_int_ctrl_hw(ring->q, flag, mask); } -static void hns_ae_toggle_queue_status(struct hnae_queue *queue, u32 val) -{ - struct dsaf_device *dsaf_dev = hns_ae_get_dsaf_dev(queue->dev); - - if (AE_IS_VER1(dsaf_dev->dsaf_ver)) - hns_rcb_int_clr_hw(queue, RCB_INT_FLAG_TX | RCB_INT_FLAG_RX); - else - hns_rcbv2_int_clr_hw(queue, RCB_INT_FLAG_TX | RCB_INT_FLAG_RX); - - hns_rcb_start(queue, val); -} - static int hns_ae_get_link_status(struct hnae_handle *handle) { u32 link_status; @@ -808,7 +805,6 @@ static struct hnae_ae_ops hns_dsaf_ops = { .stop = hns_ae_stop, .reset = hns_ae_reset, .toggle_ring_irq = hns_ae_toggle_ring_irq, - .toggle_queue_status = hns_ae_toggle_queue_status, .get_status = hns_ae_get_link_status, .get_info = hns_ae_get_mac_info, .adjust_link = hns_ae_adjust_link, diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index cef9d12..f49246d 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -1200,7 +1200,7 @@ static int hns_nic_net_up(struct net_device *ndev) { struct hns_nic_priv *priv = netdev_priv(ndev); struct hnae_handle *h = priv->ae_handle; - int i, j, k; + int i, j; int ret; ret = hns_nic_init_irq(priv); @@ -1215,9 +1215,6 @@ static int hns_nic_net_up(struct net_device *ndev) goto out_has_some_queues; } - for (k = 0; k < h->q_num; k++) - h->dev->ops->toggle_queue_status(h->qs[k], 1); - ret = h->dev->ops->set_mac_addr(h, ndev->dev_addr); if (ret) goto out_set_mac_addr_err; @@ -1237,8 +1234,6 @@ static int hns_nic_net_up(struct net_device *ndev) out_start_err: netif_stop_queue(ndev); out_set_mac_addr_err: - for (k = 0; k < h->q_num; k++) - h->dev->ops->toggle_queue_status(h->qs[k], 0); out_has_some_queues: for (j = i - 1; j >= 0; j--) hns_nic_ring_close(ndev, j); -- cgit v0.10.2 From d8a8371e8ba24e49d8e42d691a0c47e9cfc42cdd Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Tue, 21 Jun 2016 11:56:35 +0800 Subject: net: hns: bug fix about led control logic when link down The default driver sets anchor led bit to 0 when link down, actually, the anchor bit should be set to 1, so fixes it when link status is down. Secondly, change the return value of cpld_set_led_id to 0, which means leave the cpld to control led blink frequece other than the driver itself. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c index aff9d77..8473287 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c @@ -86,9 +86,10 @@ static void hns_cpld_set_led(struct hns_mac_cb *mac_cb, int link_status, mac_cb->cpld_led_value = value; } } else { - dsaf_write_syscon(mac_cb->cpld_ctrl, mac_cb->cpld_ctrl_reg, - CPLD_LED_DEFAULT_VALUE); - mac_cb->cpld_led_value = CPLD_LED_DEFAULT_VALUE; + value = (mac_cb->cpld_led_value) & (0x1 << DSAF_LED_ANCHOR_B); + dsaf_write_syscon(mac_cb->cpld_ctrl, + mac_cb->cpld_ctrl_reg, value); + mac_cb->cpld_led_value = value; } } @@ -114,7 +115,7 @@ static int cpld_set_led_id(struct hns_mac_cb *mac_cb, CPLD_LED_ON_VALUE); dsaf_write_syscon(mac_cb->cpld_ctrl, mac_cb->cpld_ctrl_reg, mac_cb->cpld_led_value); - return 2; + break; case HNAE_LED_INACTIVE: dsaf_set_bit(mac_cb->cpld_led_value, DSAF_LED_ANCHOR_B, CPLD_LED_DEFAULT_VALUE); @@ -122,7 +123,8 @@ static int cpld_set_led_id(struct hns_mac_cb *mac_cb, mac_cb->cpld_led_value); break; default: - break; + dev_err(mac_cb->dev, "invalid led state: %d!", status); + return -EINVAL; } return 0; -- cgit v0.10.2 From 3a31b64e3e4f37452709d73f2c0b5c433faad3f3 Mon Sep 17 00:00:00 2001 From: Jun He Date: Tue, 21 Jun 2016 11:56:36 +0800 Subject: net: hns: fix bug that alloc skb fail lead to port unavailable When hns_nic_poll_rx_skb alloc skb fail, it will break receive cycle and read new fbd_num to start new receive cycle. It recomputes cycle num is fbd_num minus clean_count, actually this cycle num is too big because it drop out receive cycle. It brings about the port unavailable. So we will goto out when alloc skb fail to fix this bug. Signed-off-by: Jun He Signed-off-by: Ding Tianhong Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index f49246d..c0ce37b 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -768,10 +768,10 @@ recv: clean_count = 0; } - /* poll one pkg*/ + /* poll one pkt*/ err = hns_nic_poll_rx_skb(ring_data, &skb, &bnum); if (unlikely(!skb)) /* this fault cannot be repaired */ - break; + goto out; recv_bds += bnum; clean_count += bnum; @@ -797,6 +797,7 @@ recv: } } +out: /* make all data has been write before submit */ if (clean_count > 0) hns_nic_alloc_rx_buffers(ring_data, clean_count); -- cgit v0.10.2 From 8ae7b8a599383b8d7022df52732e68c0888c0d5e Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Tue, 21 Jun 2016 11:56:37 +0800 Subject: net: hns: fix sbm default parameters config error The default sbm config parameter leaves little buffer when there is heavy traffic, which will cause packets drop. This patch changes them to make enough buffers for handling packets. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index 7f5c248..67e8e13 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -516,10 +516,10 @@ static void hns_dsafv2_sbm_bp_wl_cfg(struct dsaf_device *dsaf_dev) o_sbm_bp_cfg = dsaf_read_dev(dsaf_dev, reg); dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG3_SET_BUF_NUM_NO_PFC_M, - DSAFV2_SBM_CFG3_SET_BUF_NUM_NO_PFC_S, 110); + DSAFV2_SBM_CFG3_SET_BUF_NUM_NO_PFC_S, 48); dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG3_RESET_BUF_NUM_NO_PFC_M, - DSAFV2_SBM_CFG3_RESET_BUF_NUM_NO_PFC_S, 160); + DSAFV2_SBM_CFG3_RESET_BUF_NUM_NO_PFC_S, 80); dsaf_write_dev(dsaf_dev, reg, o_sbm_bp_cfg); /* for no enable pfc mode */ @@ -527,29 +527,39 @@ static void hns_dsafv2_sbm_bp_wl_cfg(struct dsaf_device *dsaf_dev) o_sbm_bp_cfg = dsaf_read_dev(dsaf_dev, reg); dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG4_SET_BUF_NUM_NO_PFC_M, - DSAFV2_SBM_CFG4_SET_BUF_NUM_NO_PFC_S, 128); + DSAFV2_SBM_CFG4_SET_BUF_NUM_NO_PFC_S, 192); dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG4_RESET_BUF_NUM_NO_PFC_M, - DSAFV2_SBM_CFG4_RESET_BUF_NUM_NO_PFC_S, 192); + DSAFV2_SBM_CFG4_RESET_BUF_NUM_NO_PFC_S, 240); dsaf_write_dev(dsaf_dev, reg, o_sbm_bp_cfg); } /* PPE */ - reg = DSAF_SBM_BP_CFG_2_PPE_REG_0_REG + 0x80 * i; - o_sbm_bp_cfg = dsaf_read_dev(dsaf_dev, reg); - dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG2_SET_BUF_NUM_M, - DSAFV2_SBM_CFG2_SET_BUF_NUM_S, 10); - dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG2_RESET_BUF_NUM_M, - DSAFV2_SBM_CFG2_RESET_BUF_NUM_S, 12); - dsaf_write_dev(dsaf_dev, reg, o_sbm_bp_cfg); + for (i = 0; i < DSAFV2_SBM_PPE_CHN; i++) { + reg = DSAF_SBM_BP_CFG_2_PPE_REG_0_REG + 0x80 * i; + o_sbm_bp_cfg = dsaf_read_dev(dsaf_dev, reg); + dsaf_set_field(o_sbm_bp_cfg, + DSAFV2_SBM_CFG2_PPE_SET_BUF_NUM_M, + DSAFV2_SBM_CFG2_PPE_SET_BUF_NUM_S, 2); + dsaf_set_field(o_sbm_bp_cfg, + DSAFV2_SBM_CFG2_PPE_RESET_BUF_NUM_M, + DSAFV2_SBM_CFG2_PPE_RESET_BUF_NUM_S, 3); + dsaf_set_field(o_sbm_bp_cfg, + DSAFV2_SBM_CFG2_PPE_CFG_USEFUL_NUM_M, + DSAFV2_SBM_CFG2_PPE_CFG_USEFUL_NUM_S, 52); + dsaf_write_dev(dsaf_dev, reg, o_sbm_bp_cfg); + } + /* RoCEE */ for (i = 0; i < DASFV2_ROCEE_CRD_NUM; i++) { reg = DSAFV2_SBM_BP_CFG_2_ROCEE_REG_0_REG + 0x80 * i; o_sbm_bp_cfg = dsaf_read_dev(dsaf_dev, reg); - dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG2_SET_BUF_NUM_M, - DSAFV2_SBM_CFG2_SET_BUF_NUM_S, 2); - dsaf_set_field(o_sbm_bp_cfg, DSAFV2_SBM_CFG2_RESET_BUF_NUM_M, - DSAFV2_SBM_CFG2_RESET_BUF_NUM_S, 4); + dsaf_set_field(o_sbm_bp_cfg, + DSAFV2_SBM_CFG2_ROCEE_SET_BUF_NUM_M, + DSAFV2_SBM_CFG2_ROCEE_SET_BUF_NUM_S, 2); + dsaf_set_field(o_sbm_bp_cfg, + DSAFV2_SBM_CFG2_ROCEE_RESET_BUF_NUM_M, + DSAFV2_SBM_CFG2_ROCEE_RESET_BUF_NUM_S, 4); dsaf_write_dev(dsaf_dev, reg, o_sbm_bp_cfg); } } diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h index e35d0cb..235f744 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h @@ -32,7 +32,7 @@ #define DSAFV2_SBM_NUM 8 #define DSAFV2_SBM_XGE_CHN 6 #define DSAFV2_SBM_PPE_CHN 1 -#define DASFV2_ROCEE_CRD_NUM 8 +#define DASFV2_ROCEE_CRD_NUM 1 #define DSAF_VOQ_NUM DSAF_NODE_NUM #define DSAF_INODE_NUM DSAF_NODE_NUM @@ -178,7 +178,7 @@ #define DSAF_SBM_BP_CFG_2_XGE_REG_0_REG 0x200C #define DSAF_SBM_BP_CFG_2_PPE_REG_0_REG 0x230C #define DSAF_SBM_BP_CFG_2_ROCEE_REG_0_REG 0x260C -#define DSAFV2_SBM_BP_CFG_2_ROCEE_REG_0_REG 0x238C +#define DSAFV2_SBM_BP_CFG_2_ROCEE_REG_0_REG 0x238C #define DSAF_SBM_FREE_CNT_0_0_REG 0x2010 #define DSAF_SBM_FREE_CNT_1_0_REG 0x2014 #define DSAF_SBM_BP_CNT_0_0_REG 0x2018 @@ -796,6 +796,18 @@ #define DSAFV2_SBM_CFG4_RESET_BUF_NUM_NO_PFC_S 9 #define DSAFV2_SBM_CFG4_RESET_BUF_NUM_NO_PFC_M (((1ULL << 9) - 1) << 9) +#define DSAFV2_SBM_CFG2_ROCEE_SET_BUF_NUM_S 0 +#define DSAFV2_SBM_CFG2_ROCEE_SET_BUF_NUM_M (((1ULL << 8) - 1) << 0) +#define DSAFV2_SBM_CFG2_ROCEE_RESET_BUF_NUM_S 8 +#define DSAFV2_SBM_CFG2_ROCEE_RESET_BUF_NUM_M (((1ULL << 8) - 1) << 8) + +#define DSAFV2_SBM_CFG2_PPE_SET_BUF_NUM_S (0) +#define DSAFV2_SBM_CFG2_PPE_SET_BUF_NUM_M (((1ULL << 6) - 1) << 0) +#define DSAFV2_SBM_CFG2_PPE_RESET_BUF_NUM_S (6) +#define DSAFV2_SBM_CFG2_PPE_RESET_BUF_NUM_M (((1ULL << 6) - 1) << 6) +#define DSAFV2_SBM_CFG2_PPE_CFG_USEFUL_NUM_S (12) +#define DSAFV2_SBM_CFG2_PPE_CFG_USEFUL_NUM_M (((1ULL << 6) - 1) << 12) + #define DSAF_TBL_TCAM_ADDR_S 0 #define DSAF_TBL_TCAM_ADDR_M ((1ULL << 9) - 1) -- cgit v0.10.2 From f28f34cdefe5323468eab79f23d792b014b7d758 Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Tue, 21 Jun 2016 11:56:38 +0800 Subject: net: hns: change the default coalesce usecs The default coalesce timeout is 3us, which is will cause CPU usage is too high. This patch change it to 50us in order to reduce CPU usage and the value makes sure network latency also meets requirement. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c index 3ce2409..ef11077 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.c @@ -540,7 +540,7 @@ int hns_rcb_set_coalesce_usecs( } if (timeout > HNS_RCB_MAX_COALESCED_USECS) { dev_err(rcb_common->dsaf_dev->dev, - "error: not support coalesce %dus!\n", timeout); + "error: coalesce_usecs setting supports 0~1023us\n"); return -EINVAL; } hns_rcb_set_port_timeout(rcb_common, port_idx, timeout); diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h index bd54dac..99b4e1b 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_rcb.h @@ -40,7 +40,7 @@ struct rcb_common_cb; #define HNS_RCB_DEF_COALESCED_FRAMES 50 #define HNS_RCB_CLK_FREQ_MHZ 350 #define HNS_RCB_MAX_COALESCED_USECS 0x3ff -#define HNS_RCB_DEF_COALESCED_USECS 3 +#define HNS_RCB_DEF_COALESCED_USECS 50 #define HNS_RCB_COMMON_ENDIAN 1 -- cgit v0.10.2 From 6fe27464d8e1e738c0a4779f6f5423017be052a2 Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Tue, 21 Jun 2016 11:56:39 +0800 Subject: net: hns: bug fix about TSO on|off when there is traffic When enable/disable tso, the driver tries to access the hardware register, but this operation will cause the port unavalible when there is traffic. This patch tries to enable TSO when initialize, then control tso through TSE bit in transmit descriptor. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index c0ce37b..d5297ec 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -1441,7 +1441,6 @@ static int hns_nic_set_features(struct net_device *netdev, netdev_features_t features) { struct hns_nic_priv *priv = netdev_priv(netdev); - struct hnae_handle *h = priv->ae_handle; switch (priv->enet_ver) { case AE_VERSION_1: @@ -1454,11 +1453,9 @@ static int hns_nic_set_features(struct net_device *netdev, priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tso; /* The chip only support 7*4096 */ netif_set_gso_max_size(netdev, 7 * 4096); - h->dev->ops->set_tso_stats(h, 1); } else { priv->ops.fill_desc = fill_v2_desc; priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tx; - h->dev->ops->set_tso_stats(h, 0); } break; } @@ -1804,11 +1801,14 @@ static void hns_nic_set_priv_ops(struct net_device *netdev) priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tso; /* This chip only support 7*4096 */ netif_set_gso_max_size(netdev, 7 * 4096); - h->dev->ops->set_tso_stats(h, 1); } else { priv->ops.fill_desc = fill_v2_desc; priv->ops.maybe_stop_tx = hns_nic_maybe_stop_tx; } + /* enable tso when init + * control tso on/off through TSE bit in bd + */ + h->dev->ops->set_tso_stats(h, 1); } } -- cgit v0.10.2 From 80bedf1a62d65fb79eaaf030e75174886f1a794c Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:03:59 +0200 Subject: mlxsw: spectrum: Use notifier_from_errno() in notifier block Instead of checking the error value and returning NOTIFY_BAD, just use notifier_from_errno() and simplify the code. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 6f9e3dd..3d57008 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -49,6 +49,7 @@ #include #include #include +#include #include #include #include @@ -3024,7 +3025,7 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, struct mlxsw_sp_port *mlxsw_sp_port; struct net_device *upper_dev; struct mlxsw_sp *mlxsw_sp; - int err; + int err = 0; mlxsw_sp_port = netdev_priv(dev); mlxsw_sp = mlxsw_sp_port->mlxsw_sp; @@ -3038,68 +3039,42 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, /* HW limitation forbids to put ports to multiple bridges. */ if (netif_is_bridge_master(upper_dev) && !mlxsw_sp_master_bridge_check(mlxsw_sp, upper_dev)) - return NOTIFY_BAD; + return -EINVAL; if (netif_is_lag_master(upper_dev) && !mlxsw_sp_master_lag_check(mlxsw_sp, upper_dev, info->upper_info)) - return NOTIFY_BAD; + return -EINVAL; break; case NETDEV_CHANGEUPPER: upper_dev = info->upper_dev; if (is_vlan_dev(upper_dev)) { - if (info->linking) { + if (info->linking) err = mlxsw_sp_port_vlan_link(mlxsw_sp_port, upper_dev); - if (err) { - netdev_err(dev, "Failed to link VLAN device\n"); - return NOTIFY_BAD; - } - } else { + else err = mlxsw_sp_port_vlan_unlink(mlxsw_sp_port, upper_dev); - if (err) { - netdev_err(dev, "Failed to unlink VLAN device\n"); - return NOTIFY_BAD; - } - } } else if (netif_is_bridge_master(upper_dev)) { if (info->linking) { err = mlxsw_sp_port_bridge_join(mlxsw_sp_port); - if (err) { - netdev_err(dev, "Failed to join bridge\n"); - return NOTIFY_BAD; - } mlxsw_sp_master_bridge_inc(mlxsw_sp, upper_dev); } else { err = mlxsw_sp_port_bridge_leave(mlxsw_sp_port, true); mlxsw_sp_master_bridge_dec(mlxsw_sp, upper_dev); - if (err) { - netdev_err(dev, "Failed to leave bridge\n"); - return NOTIFY_BAD; - } } } else if (netif_is_lag_master(upper_dev)) { - if (info->linking) { + if (info->linking) err = mlxsw_sp_port_lag_join(mlxsw_sp_port, upper_dev); - if (err) { - netdev_err(dev, "Failed to join link aggregation\n"); - return NOTIFY_BAD; - } - } else { + else err = mlxsw_sp_port_lag_leave(mlxsw_sp_port, upper_dev); - if (err) { - netdev_err(dev, "Failed to leave link aggregation\n"); - return NOTIFY_BAD; - } - } } break; } - return NOTIFY_DONE; + return err; } static int mlxsw_sp_netdevice_port_lower_event(struct net_device *dev, @@ -3123,7 +3098,7 @@ static int mlxsw_sp_netdevice_port_lower_event(struct net_device *dev, break; } - return NOTIFY_DONE; + return 0; } static int mlxsw_sp_netdevice_port_event(struct net_device *dev, @@ -3137,7 +3112,7 @@ static int mlxsw_sp_netdevice_port_event(struct net_device *dev, return mlxsw_sp_netdevice_port_lower_event(dev, event, ptr); } - return NOTIFY_DONE; + return 0; } static int mlxsw_sp_netdevice_lag_event(struct net_device *lag_dev, @@ -3150,12 +3125,12 @@ static int mlxsw_sp_netdevice_lag_event(struct net_device *lag_dev, netdev_for_each_lower_dev(lag_dev, dev, iter) { if (mlxsw_sp_port_dev_check(dev)) { ret = mlxsw_sp_netdevice_port_event(dev, event, ptr); - if (ret == NOTIFY_BAD) + if (ret) return ret; } } - return NOTIFY_DONE; + return 0; } static struct mlxsw_sp_vfid * @@ -3446,7 +3421,7 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, struct netdev_notifier_changeupper_info *info = ptr; struct mlxsw_sp_port *mlxsw_sp_vport; struct net_device *upper_dev; - int err; + int err = 0; mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid); @@ -3456,13 +3431,13 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, if (!info->master || !info->linking) break; if (!netif_is_bridge_master(upper_dev)) - return NOTIFY_BAD; + return -EINVAL; /* We can't have multiple VLAN interfaces configured on * the same port and being members in the same bridge. */ if (!mlxsw_sp_port_master_bridge_check(mlxsw_sp_port, upper_dev)) - return NOTIFY_BAD; + return -EINVAL; break; case NETDEV_CHANGEUPPER: upper_dev = info->upper_dev; @@ -3471,31 +3446,23 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, if (info->linking) { if (!mlxsw_sp_vport) { WARN_ON(!mlxsw_sp_vport); - return NOTIFY_BAD; + return -EINVAL; } err = mlxsw_sp_vport_bridge_join(mlxsw_sp_vport, upper_dev); - if (err) { - netdev_err(dev, "Failed to join bridge\n"); - return NOTIFY_BAD; - } } else { /* We ignore bridge's unlinking notifications if vPort * is gone, since we already left the bridge when the * VLAN device was unlinked from the real device. */ if (!mlxsw_sp_vport) - return NOTIFY_DONE; + return 0; err = mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, upper_dev, true); - if (err) { - netdev_err(dev, "Failed to leave bridge\n"); - return NOTIFY_BAD; - } } } - return NOTIFY_DONE; + return err; } static int mlxsw_sp_netdevice_lag_vport_event(struct net_device *lag_dev, @@ -3510,12 +3477,12 @@ static int mlxsw_sp_netdevice_lag_vport_event(struct net_device *lag_dev, if (mlxsw_sp_port_dev_check(dev)) { ret = mlxsw_sp_netdevice_vport_event(dev, event, ptr, vid); - if (ret == NOTIFY_BAD) + if (ret) return ret; } } - return NOTIFY_DONE; + return 0; } static int mlxsw_sp_netdevice_vlan_event(struct net_device *vlan_dev, @@ -3531,24 +3498,23 @@ static int mlxsw_sp_netdevice_vlan_event(struct net_device *vlan_dev, return mlxsw_sp_netdevice_lag_vport_event(real_dev, event, ptr, vid); - return NOTIFY_DONE; + return 0; } static int mlxsw_sp_netdevice_event(struct notifier_block *unused, unsigned long event, void *ptr) { struct net_device *dev = netdev_notifier_info_to_dev(ptr); + int err = 0; if (mlxsw_sp_port_dev_check(dev)) - return mlxsw_sp_netdevice_port_event(dev, event, ptr); - - if (netif_is_lag_master(dev)) - return mlxsw_sp_netdevice_lag_event(dev, event, ptr); - - if (is_vlan_dev(dev)) - return mlxsw_sp_netdevice_vlan_event(dev, event, ptr); + err = mlxsw_sp_netdevice_port_event(dev, event, ptr); + else if (netif_is_lag_master(dev)) + err = mlxsw_sp_netdevice_lag_event(dev, event, ptr); + else if (is_vlan_dev(dev)) + err = mlxsw_sp_netdevice_vlan_event(dev, event, ptr); - return NOTIFY_DONE; + return notifier_from_errno(err); } static struct notifier_block mlxsw_sp_netdevice_nb __read_mostly = { -- cgit v0.10.2 From 59fe9b3f842751688d9ee180f673314190813be1 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:00 +0200 Subject: mlxsw: spectrum: Sanitize port netdev upper devices We currently only support the following upper devices for port netdevs: 1) Bridge 2) LAG (bond / team) 3) VLAN Any other device is forbidden, so return an error. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 3d57008..b47e3fb 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -3034,6 +3034,10 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, switch (event) { case NETDEV_PRECHANGEUPPER: upper_dev = info->upper_dev; + if (!is_vlan_dev(upper_dev) && + !netif_is_lag_master(upper_dev) && + !netif_is_bridge_master(upper_dev)) + return -EINVAL; if (!info->master || !info->linking) break; /* HW limitation forbids to put ports to multiple bridges. */ @@ -3070,6 +3074,9 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, else err = mlxsw_sp_port_lag_leave(mlxsw_sp_port, upper_dev); + } else { + err = -EINVAL; + WARN_ON(1); } break; } -- cgit v0.10.2 From 6ec439043b0dcbd7845b9e268a12daac92ebd190 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:01 +0200 Subject: mlxsw: spectrum: Forbid LAG slave from having VLAN uppers When a port netdev is put under LAG it cannot have VLAN upper devices, so forbid that. The LAG device itself can have VLAN upper devices. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index b47e3fb..46c867f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -3038,7 +3038,7 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, !netif_is_lag_master(upper_dev) && !netif_is_bridge_master(upper_dev)) return -EINVAL; - if (!info->master || !info->linking) + if (!info->linking) break; /* HW limitation forbids to put ports to multiple bridges. */ if (netif_is_bridge_master(upper_dev) && @@ -3048,6 +3048,11 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, !mlxsw_sp_master_lag_check(mlxsw_sp, upper_dev, info->upper_info)) return -EINVAL; + if (netif_is_lag_master(upper_dev) && vlan_uses_dev(dev)) + return -EINVAL; + if (netif_is_lag_port(dev) && is_vlan_dev(upper_dev) && + !netif_is_lag_master(vlan_dev_real_dev(upper_dev))) + return -EINVAL; break; case NETDEV_CHANGEUPPER: upper_dev = info->upper_dev; -- cgit v0.10.2 From ddbe993dbe4d832a6ea31e4359150fbc5a4b93d4 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:02 +0200 Subject: mlxsw: spectrum: Remove unnecessary checks from event processing When upper device of a VLAN device changes we already made sure it's a bridge device in PRECHANGEUPPER, so no need to check it's a master device in CHANGEUPPER. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 46c867f..e55c685 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -3440,10 +3440,10 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, switch (event) { case NETDEV_PRECHANGEUPPER: upper_dev = info->upper_dev; - if (!info->master || !info->linking) - break; if (!netif_is_bridge_master(upper_dev)) return -EINVAL; + if (!info->linking) + break; /* We can't have multiple VLAN interfaces configured on * the same port and being members in the same bridge. */ @@ -3453,8 +3453,6 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, break; case NETDEV_CHANGEUPPER: upper_dev = info->upper_dev; - if (!info->master) - break; if (info->linking) { if (!mlxsw_sp_vport) { WARN_ON(!mlxsw_sp_vport); -- cgit v0.10.2 From 423b937e7dde1291fc0e87bb7a382aafde9d712a Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:03 +0200 Subject: mlxsw: spectrum: Use WARN_ON() return value Instead of checking for a condition and then issue the warning, just do it in one go and simplify the code. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index e55c685..f297b10 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2980,10 +2980,8 @@ static int mlxsw_sp_port_vlan_link(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid = vlan_dev_vlan_id(vlan_dev); mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid); - if (!mlxsw_sp_vport) { - WARN_ON(!mlxsw_sp_vport); + if (WARN_ON(!mlxsw_sp_vport)) return -EINVAL; - } mlxsw_sp_vport->dev = vlan_dev; @@ -2997,10 +2995,8 @@ static int mlxsw_sp_port_vlan_unlink(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid = vlan_dev_vlan_id(vlan_dev); mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid); - if (!mlxsw_sp_vport) { - WARN_ON(!mlxsw_sp_vport); + if (WARN_ON(!mlxsw_sp_vport)) return -EINVAL; - } /* When removing a VLAN device while still bridged we should first * remove it from the bridge, as we receive the bridge's notification @@ -3236,10 +3232,8 @@ static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, int err; vfid = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); - if (!vfid) { - WARN_ON(!vfid); + if (WARN_ON(!vfid)) return -EINVAL; - } /* We need a vFID to go back to after leaving the bridge's vFID. */ new_vfid = mlxsw_sp_vfid_find(mlxsw_sp, vid); @@ -3454,10 +3448,8 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, case NETDEV_CHANGEUPPER: upper_dev = info->upper_dev; if (info->linking) { - if (!mlxsw_sp_vport) { - WARN_ON(!mlxsw_sp_vport); + if (WARN_ON(!mlxsw_sp_vport)) return -EINVAL; - } err = mlxsw_sp_vport_bridge_join(mlxsw_sp_vport, upper_dev); } else { -- cgit v0.10.2 From 82e6db034b1fca1370c3930e7c4b61f0d8bb2c4b Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:04 +0200 Subject: mlxsw: spectrum: Make unlinking functions return void When responding to unlinking CHANGEUPPER notifications we shouldn't return any value, as it's not checked by upper layers. In addition, there's nothing the driver can do in case of failure, so it should simply continue and try to free as much resources as possible and not stop on first error. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index f297b10..3500a72 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2673,8 +2673,8 @@ static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port) return 0; } -static int mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port, - bool flush_fdb) +static void mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port, + bool flush_fdb) { struct net_device *dev = mlxsw_sp_port->dev; @@ -2691,7 +2691,7 @@ static int mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port, /* Add implicit VLAN interface in the device, so that untagged * packets will be classified to the default vFID. */ - return mlxsw_sp_port_add_vid(dev, 0, 1); + mlxsw_sp_port_add_vid(dev, 0, 1); } static bool mlxsw_sp_master_bridge_check(struct mlxsw_sp *mlxsw_sp, @@ -2873,30 +2873,25 @@ err_col_port_add: return err; } -static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, - struct net_device *br_dev, - bool flush_fdb); +static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *br_dev, + bool flush_fdb); -static int mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, - struct net_device *lag_dev) +static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, + struct net_device *lag_dev) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct mlxsw_sp_port *mlxsw_sp_vport; struct mlxsw_sp_upper *lag; u16 lag_id = mlxsw_sp_port->lag_id; - int err; if (!mlxsw_sp_port->lagged) - return 0; + return; lag = mlxsw_sp_lag_get(mlxsw_sp, lag_id); WARN_ON(lag->ref_count == 0); - err = mlxsw_sp_lag_col_port_disable(mlxsw_sp_port, lag_id); - if (err) - return err; - err = mlxsw_sp_lag_col_port_remove(mlxsw_sp_port, lag_id); - if (err) - return err; + mlxsw_sp_lag_col_port_disable(mlxsw_sp_port, lag_id); + mlxsw_sp_lag_col_port_remove(mlxsw_sp_port, lag_id); /* In case we leave a LAG device that has bridges built on top, * then their teardown sequence is never issued and we need to @@ -2922,16 +2917,13 @@ static int mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, if (lag->ref_count == 1) { if (mlxsw_sp_port_fdb_flush_by_lag_id(mlxsw_sp_port)) netdev_err(mlxsw_sp_port->dev, "Failed to flush FDB\n"); - err = mlxsw_sp_lag_destroy(mlxsw_sp, lag_id); - if (err) - return err; + mlxsw_sp_lag_destroy(mlxsw_sp, lag_id); } mlxsw_core_lag_mapping_clear(mlxsw_sp->core, lag_id, mlxsw_sp_port->local_port); mlxsw_sp_port->lagged = 0; lag->ref_count--; - return 0; } static int mlxsw_sp_lag_dist_port_add(struct mlxsw_sp_port *mlxsw_sp_port, @@ -2988,15 +2980,15 @@ static int mlxsw_sp_port_vlan_link(struct mlxsw_sp_port *mlxsw_sp_port, return 0; } -static int mlxsw_sp_port_vlan_unlink(struct mlxsw_sp_port *mlxsw_sp_port, - struct net_device *vlan_dev) +static void mlxsw_sp_port_vlan_unlink(struct mlxsw_sp_port *mlxsw_sp_port, + struct net_device *vlan_dev) { struct mlxsw_sp_port *mlxsw_sp_vport; u16 vid = vlan_dev_vlan_id(vlan_dev); mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid); if (WARN_ON(!mlxsw_sp_vport)) - return -EINVAL; + return; /* When removing a VLAN device while still bridged we should first * remove it from the bridge, as we receive the bridge's notification @@ -3010,8 +3002,6 @@ static int mlxsw_sp_port_vlan_unlink(struct mlxsw_sp_port *mlxsw_sp_port, } mlxsw_sp_vport->dev = mlxsw_sp_port->dev; - - return 0; } static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, @@ -3057,15 +3047,14 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, err = mlxsw_sp_port_vlan_link(mlxsw_sp_port, upper_dev); else - err = mlxsw_sp_port_vlan_unlink(mlxsw_sp_port, - upper_dev); + mlxsw_sp_port_vlan_unlink(mlxsw_sp_port, + upper_dev); } else if (netif_is_bridge_master(upper_dev)) { if (info->linking) { err = mlxsw_sp_port_bridge_join(mlxsw_sp_port); mlxsw_sp_master_bridge_inc(mlxsw_sp, upper_dev); } else { - err = mlxsw_sp_port_bridge_leave(mlxsw_sp_port, - true); + mlxsw_sp_port_bridge_leave(mlxsw_sp_port, true); mlxsw_sp_master_bridge_dec(mlxsw_sp, upper_dev); } } else if (netif_is_lag_master(upper_dev)) { @@ -3073,8 +3062,8 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, err = mlxsw_sp_port_lag_join(mlxsw_sp_port, upper_dev); else - err = mlxsw_sp_port_lag_leave(mlxsw_sp_port, - upper_dev); + mlxsw_sp_port_lag_leave(mlxsw_sp_port, + upper_dev); } else { err = -EINVAL; WARN_ON(1); @@ -3221,9 +3210,9 @@ static void mlxsw_sp_br_vfid_destroy(struct mlxsw_sp *mlxsw_sp, kfree(vfid); } -static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, - struct net_device *br_dev, - bool flush_fdb) +static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *br_dev, + bool flush_fdb) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); @@ -3233,7 +3222,7 @@ static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, vfid = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); if (WARN_ON(!vfid)) - return -EINVAL; + return; /* We need a vFID to go back to after leaving the bridge's vFID. */ new_vfid = mlxsw_sp_vfid_find(mlxsw_sp, vid); @@ -3242,7 +3231,7 @@ static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, if (IS_ERR(new_vfid)) { netdev_err(dev, "Failed to create vFID for VID=%d\n", vid); - return PTR_ERR(new_vfid); + return; } } @@ -3306,7 +3295,7 @@ static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, mlxsw_sp_vport->uc_flood = 0; mlxsw_sp_vport->bridged = 0; - return 0; + return; err_port_stp_state_set: err_vport_flood_set: @@ -3316,7 +3305,6 @@ err_port_vid_to_fid_invalidate: /* Rollback vFID only if new. */ if (!new_vfid->nr_vports) mlxsw_sp_vfid_destroy(mlxsw_sp, new_vfid); - return err; } static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, @@ -3459,8 +3447,8 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, */ if (!mlxsw_sp_vport) return 0; - err = mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, - upper_dev, true); + mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, upper_dev, + true); } } -- cgit v0.10.2 From 279438952b7f5423f33ce4ddf639a7e796377b59 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:05 +0200 Subject: mlxsw: spectrum: Remove unnecessary function argument The argument 'br_dev' is never used, so remove it. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 3500a72..cd85316 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2708,8 +2708,7 @@ static void mlxsw_sp_master_bridge_inc(struct mlxsw_sp *mlxsw_sp, mlxsw_sp->master_bridge.ref_count++; } -static void mlxsw_sp_master_bridge_dec(struct mlxsw_sp *mlxsw_sp, - struct net_device *br_dev) +static void mlxsw_sp_master_bridge_dec(struct mlxsw_sp *mlxsw_sp) { if (--mlxsw_sp->master_bridge.ref_count == 0) mlxsw_sp->master_bridge.dev = NULL; @@ -2911,7 +2910,7 @@ static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, if (mlxsw_sp_port->bridged) { mlxsw_sp_port_active_vlans_del(mlxsw_sp_port); mlxsw_sp_port_bridge_leave(mlxsw_sp_port, false); - mlxsw_sp_master_bridge_dec(mlxsw_sp, NULL); + mlxsw_sp_master_bridge_dec(mlxsw_sp); } if (lag->ref_count == 1) { @@ -3055,7 +3054,7 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, mlxsw_sp_master_bridge_inc(mlxsw_sp, upper_dev); } else { mlxsw_sp_port_bridge_leave(mlxsw_sp_port, true); - mlxsw_sp_master_bridge_dec(mlxsw_sp, upper_dev); + mlxsw_sp_master_bridge_dec(mlxsw_sp); } } else if (netif_is_lag_master(upper_dev)) { if (info->linking) -- cgit v0.10.2 From 7117a570b93b8c1f0e20d4efd2695abe929c1fc4 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:06 +0200 Subject: mlxsw: spectrum: Centralize VLAN-aware bridge ref counting We hold a reference count on the number of ports member in the VLAN-aware bridge, as we only support one. Instead of always incrementing / decrementing the reference count after joining / leaving the bridge, simply do this accounting in the join / leave functions. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index cd85316..b999802 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2651,7 +2651,28 @@ static bool mlxsw_sp_port_dev_check(const struct net_device *dev) return dev->netdev_ops == &mlxsw_sp_port_netdev_ops; } -static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port) +static bool mlxsw_sp_master_bridge_check(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev) +{ + return !mlxsw_sp->master_bridge.dev || + mlxsw_sp->master_bridge.dev == br_dev; +} + +static void mlxsw_sp_master_bridge_inc(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev) +{ + mlxsw_sp->master_bridge.dev = br_dev; + mlxsw_sp->master_bridge.ref_count++; +} + +static void mlxsw_sp_master_bridge_dec(struct mlxsw_sp *mlxsw_sp) +{ + if (--mlxsw_sp->master_bridge.ref_count == 0) + mlxsw_sp->master_bridge.dev = NULL; +} + +static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port, + struct net_device *br_dev) { struct net_device *dev = mlxsw_sp_port->dev; int err; @@ -2665,6 +2686,8 @@ static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port) if (err) return err; + mlxsw_sp_master_bridge_inc(mlxsw_sp_port->mlxsw_sp, br_dev); + mlxsw_sp_port->learning = 1; mlxsw_sp_port->learning_sync = 1; mlxsw_sp_port->uc_flood = 1; @@ -2683,6 +2706,8 @@ static void mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_port_pvid_set(mlxsw_sp_port, 1); + mlxsw_sp_master_bridge_dec(mlxsw_sp_port->mlxsw_sp); + mlxsw_sp_port->learning = 0; mlxsw_sp_port->learning_sync = 0; mlxsw_sp_port->uc_flood = 0; @@ -2694,26 +2719,6 @@ static void mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_port_add_vid(dev, 0, 1); } -static bool mlxsw_sp_master_bridge_check(struct mlxsw_sp *mlxsw_sp, - struct net_device *br_dev) -{ - return !mlxsw_sp->master_bridge.dev || - mlxsw_sp->master_bridge.dev == br_dev; -} - -static void mlxsw_sp_master_bridge_inc(struct mlxsw_sp *mlxsw_sp, - struct net_device *br_dev) -{ - mlxsw_sp->master_bridge.dev = br_dev; - mlxsw_sp->master_bridge.ref_count++; -} - -static void mlxsw_sp_master_bridge_dec(struct mlxsw_sp *mlxsw_sp) -{ - if (--mlxsw_sp->master_bridge.ref_count == 0) - mlxsw_sp->master_bridge.dev = NULL; -} - static int mlxsw_sp_lag_create(struct mlxsw_sp *mlxsw_sp, u16 lag_id) { char sldr_pl[MLXSW_REG_SLDR_LEN]; @@ -2910,7 +2915,6 @@ static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, if (mlxsw_sp_port->bridged) { mlxsw_sp_port_active_vlans_del(mlxsw_sp_port); mlxsw_sp_port_bridge_leave(mlxsw_sp_port, false); - mlxsw_sp_master_bridge_dec(mlxsw_sp); } if (lag->ref_count == 1) { @@ -3049,13 +3053,11 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, mlxsw_sp_port_vlan_unlink(mlxsw_sp_port, upper_dev); } else if (netif_is_bridge_master(upper_dev)) { - if (info->linking) { - err = mlxsw_sp_port_bridge_join(mlxsw_sp_port); - mlxsw_sp_master_bridge_inc(mlxsw_sp, upper_dev); - } else { + if (info->linking) + err = mlxsw_sp_port_bridge_join(mlxsw_sp_port, + upper_dev); + else mlxsw_sp_port_bridge_leave(mlxsw_sp_port, true); - mlxsw_sp_master_bridge_dec(mlxsw_sp); - } } else if (netif_is_lag_master(upper_dev)) { if (info->linking) err = mlxsw_sp_port_lag_join(mlxsw_sp_port, -- cgit v0.10.2 From d8651fd8868e340eddc41e1907a973a3d21de993 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:07 +0200 Subject: mlxsw: spectrum: Use DECLARE_BITMAP() macro There is a macro to do this kind of declarations, so use it. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 13b30ea..5da98b4 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -155,17 +155,17 @@ struct mlxsw_sp_sb { struct mlxsw_sp { struct { struct list_head list; - unsigned long mapped[BITS_TO_LONGS(MLXSW_SP_VFID_PORT_MAX)]; + DECLARE_BITMAP(mapped, MLXSW_SP_VFID_PORT_MAX); } port_vfids; struct { struct list_head list; - unsigned long mapped[BITS_TO_LONGS(MLXSW_SP_VFID_BR_MAX)]; + DECLARE_BITMAP(mapped, MLXSW_SP_VFID_BR_MAX); } br_vfids; struct { struct list_head list; - unsigned long mapped[BITS_TO_LONGS(MLXSW_SP_MID_MAX)]; + DECLARE_BITMAP(mapped, MLXSW_SP_MID_MAX); } br_mids; - unsigned long active_fids[BITS_TO_LONGS(VLAN_N_VID)]; + DECLARE_BITMAP(active_fids, VLAN_N_VID); struct mlxsw_sp_port **ports; struct mlxsw_core *core; const struct mlxsw_bus_info *bus_info; -- cgit v0.10.2 From 47a0a9e6c38325028d8fa4eb764687fbeaee8ef2 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:08 +0200 Subject: mlxsw: spectrum: Remove redundant function argument In all call sites 'only_uc' is set to false, so strip it. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index b999802..90cf6c5 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -791,7 +791,7 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, if (!vfid->nr_vports) { err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, - true, false); + true); if (err) { netdev_err(dev, "Failed to setup flooding for vFID=%d\n", vfid->vfid); @@ -859,8 +859,7 @@ err_port_vid_to_fid_set: mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port); err_port_vp_mode_trans: if (!vfid->nr_vports) - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false, - false); + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false); err_vport_flood_set: mlxsw_sp_port_vport_destroy(mlxsw_sp_vport); err_port_vport_create: @@ -3267,8 +3266,7 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, goto err_port_vid_learning_set; } - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false, - false); + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false); if (err) { netdev_err(dev, "Failed clear to clear flooding\n"); goto err_vport_flood_set; @@ -3327,7 +3325,7 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, } } - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, true, false); + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, true); if (err) { netdev_err(dev, "Failed to setup flooding for vFID=%d\n", vfid->vfid); @@ -3386,7 +3384,7 @@ err_port_vid_to_fid_validate: err_port_vid_to_fid_invalidate: mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); err_port_vid_learning_set: - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false, false); + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false); err_port_flood_set: if (!vfid->nr_vports) mlxsw_sp_br_vfid_destroy(mlxsw_sp, vfid); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 5da98b4..58a2a5d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -367,7 +367,7 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, int mlxsw_sp_port_kill_vid(struct net_device *dev, __be16 __always_unused proto, u16 vid); int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vfid, - bool set, bool only_uc); + bool set); void mlxsw_sp_port_active_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port); int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid); int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 3710f19..02c126c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -261,13 +261,13 @@ err_port_flood_set: } int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vfid, - bool set, bool only_uc) + bool set) { /* In case of vFIDs, index into the flooding table is relative to * the start of the vFIDs range. */ return __mlxsw_sp_port_flood_set(mlxsw_sp_vport, vfid, vfid, set, - only_uc); + false); } static int mlxsw_sp_port_attr_br_flags_set(struct mlxsw_sp_port *mlxsw_sp_port, -- cgit v0.10.2 From c7e920b5be6b2deaef51ae65a9ce18da3c8fa7da Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:09 +0200 Subject: mlxsw: spectrum: Use only one function to create vFIDs Simplify the code and use only one function for vFID creation / destruction. Unlike before, the function receives a FID index as its argument and not a vFID index. Instead of passing 0, now one would need to pass 4K, which is the first vFID. This is the first step in creating a generic FID struct that will be used for all three types of FIDs. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 90cf6c5..4f53dd2 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -652,68 +652,61 @@ static u16 mlxsw_sp_avail_vfid_get(const struct mlxsw_sp *mlxsw_sp) MLXSW_SP_VFID_PORT_MAX); } -static int __mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, u16 vfid) +static int mlxsw_sp_vfid_op(struct mlxsw_sp *mlxsw_sp, u16 fid, bool create) { - u16 fid = mlxsw_sp_vfid_to_fid(vfid); char sfmr_pl[MLXSW_REG_SFMR_LEN]; - mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_CREATE_FID, fid, 0); + mlxsw_reg_sfmr_pack(sfmr_pl, !create, fid, 0); return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); } -static void __mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, u16 vfid) -{ - u16 fid = mlxsw_sp_vfid_to_fid(vfid); - char sfmr_pl[MLXSW_REG_SFMR_LEN]; - - mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_DESTROY_FID, fid, 0); - mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); -} - static struct mlxsw_sp_vfid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, u16 vid) { struct device *dev = mlxsw_sp->bus_info->dev; - struct mlxsw_sp_vfid *vfid; - u16 n_vfid; + struct mlxsw_sp_vfid *f; + u16 vfid, fid; int err; - n_vfid = mlxsw_sp_avail_vfid_get(mlxsw_sp); - if (n_vfid == MLXSW_SP_VFID_PORT_MAX) { + vfid = mlxsw_sp_avail_vfid_get(mlxsw_sp); + if (vfid == MLXSW_SP_VFID_PORT_MAX) { dev_err(dev, "No available vFIDs\n"); return ERR_PTR(-ERANGE); } - err = __mlxsw_sp_vfid_create(mlxsw_sp, n_vfid); + fid = mlxsw_sp_vfid_to_fid(vfid); + err = mlxsw_sp_vfid_op(mlxsw_sp, fid, true); if (err) { - dev_err(dev, "Failed to create vFID=%d\n", n_vfid); + dev_err(dev, "Failed to create FID=%d\n", fid); return ERR_PTR(err); } - vfid = kzalloc(sizeof(*vfid), GFP_KERNEL); - if (!vfid) + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (!f) goto err_allocate_vfid; - vfid->vfid = n_vfid; - vfid->vid = vid; + f->vfid = vfid; + f->vid = vid; - list_add(&vfid->list, &mlxsw_sp->port_vfids.list); - set_bit(n_vfid, mlxsw_sp->port_vfids.mapped); + list_add(&f->list, &mlxsw_sp->port_vfids.list); + set_bit(vfid, mlxsw_sp->port_vfids.mapped); - return vfid; + return f; err_allocate_vfid: - __mlxsw_sp_vfid_destroy(mlxsw_sp, n_vfid); + mlxsw_sp_vfid_op(mlxsw_sp, fid, false); return ERR_PTR(-ENOMEM); } static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_vfid *vfid) { + u16 fid = mlxsw_sp_vfid_to_fid(vfid->vfid); + clear_bit(vfid->vfid, mlxsw_sp->port_vfids.mapped); list_del(&vfid->list); - __mlxsw_sp_vfid_destroy(mlxsw_sp, vfid->vfid); + mlxsw_sp_vfid_op(mlxsw_sp, fid, false); kfree(vfid); } @@ -3164,36 +3157,37 @@ static struct mlxsw_sp_vfid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, struct net_device *br_dev) { struct device *dev = mlxsw_sp->bus_info->dev; - struct mlxsw_sp_vfid *vfid; - u16 n_vfid; + struct mlxsw_sp_vfid *f; + u16 vfid, fid; int err; - n_vfid = mlxsw_sp_br_vfid_to_vfid(mlxsw_sp_avail_br_vfid_get(mlxsw_sp)); - if (n_vfid == MLXSW_SP_VFID_MAX) { + vfid = mlxsw_sp_br_vfid_to_vfid(mlxsw_sp_avail_br_vfid_get(mlxsw_sp)); + if (vfid == MLXSW_SP_VFID_MAX) { dev_err(dev, "No available vFIDs\n"); return ERR_PTR(-ERANGE); } - err = __mlxsw_sp_vfid_create(mlxsw_sp, n_vfid); + fid = mlxsw_sp_vfid_to_fid(vfid); + err = mlxsw_sp_vfid_op(mlxsw_sp, fid, true); if (err) { - dev_err(dev, "Failed to create vFID=%d\n", n_vfid); + dev_err(dev, "Failed to create FID=%d\n", fid); return ERR_PTR(err); } - vfid = kzalloc(sizeof(*vfid), GFP_KERNEL); - if (!vfid) + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (!f) goto err_allocate_vfid; - vfid->vfid = n_vfid; - vfid->br_dev = br_dev; + f->vfid = vfid; + f->br_dev = br_dev; - list_add(&vfid->list, &mlxsw_sp->br_vfids.list); - set_bit(mlxsw_sp_vfid_to_br_vfid(n_vfid), mlxsw_sp->br_vfids.mapped); + list_add(&f->list, &mlxsw_sp->br_vfids.list); + set_bit(mlxsw_sp_vfid_to_br_vfid(vfid), mlxsw_sp->br_vfids.mapped); - return vfid; + return f; err_allocate_vfid: - __mlxsw_sp_vfid_destroy(mlxsw_sp, n_vfid); + mlxsw_sp_vfid_op(mlxsw_sp, fid, false); return ERR_PTR(-ENOMEM); } @@ -3201,11 +3195,12 @@ static void mlxsw_sp_br_vfid_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_vfid *vfid) { u16 br_vfid = mlxsw_sp_vfid_to_br_vfid(vfid->vfid); + u16 fid = mlxsw_sp_vfid_to_fid(vfid->vfid); clear_bit(br_vfid, mlxsw_sp->br_vfids.mapped); list_del(&vfid->list); - __mlxsw_sp_vfid_destroy(mlxsw_sp, vfid->vfid); + mlxsw_sp_vfid_op(mlxsw_sp, fid, false); kfree(vfid); } -- cgit v0.10.2 From 9c4d442314e0dcc0ba82dfa3850ef938af123a02 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:10 +0200 Subject: mlxsw: spectrum: Create a function to map vPort's FID A FID used by a vPort (vFID, but also rFID later in the series) is always mapped using {Port, VID} and not only VID as with the 4K FIDs of the VLAN-aware bridge. Instead of specifying all the arguments each time, just wrap this operation using a dedicated function and simplify the code. As before, the function takes FID as its argument in preparation for a generic FID struct. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 4f53dd2..6df944b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -745,6 +745,16 @@ static void mlxsw_sp_port_vport_destroy(struct mlxsw_sp_port *mlxsw_sp_vport) kfree(mlxsw_sp_vport); } +static int mlxsw_sp_vport_fid_map(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, + bool valid) +{ + enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; + u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); + + return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, mt, valid, fid, + vid); +} + int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, u16 vid) { @@ -752,6 +762,7 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct mlxsw_sp_port *mlxsw_sp_vport; struct mlxsw_sp_vfid *vfid; + u16 fid; int err; /* VLAN 0 is added to HW filter when device goes up, but it is @@ -804,15 +815,12 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, } } - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - true, - mlxsw_sp_vfid_to_fid(vfid->vfid), - vid); + fid = mlxsw_sp_vfid_to_fid(vfid->vfid); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, true); if (err) { netdev_err(dev, "Failed to map {Port, VID=%d} to vFID=%d\n", vid, vfid->vfid); - goto err_port_vid_to_fid_set; + goto err_vport_fid_map; } err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); @@ -844,10 +852,8 @@ err_port_stp_state_set: err_port_add_vid: mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true); err_port_vid_learning_set: - mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, false, - mlxsw_sp_vfid_to_fid(vfid->vfid), vid); -err_port_vid_to_fid_set: + mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, false); +err_vport_fid_map: if (list_is_singular(&mlxsw_sp_port->vports_list)) mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port); err_port_vp_mode_trans: @@ -867,6 +873,7 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); struct mlxsw_sp_port *mlxsw_sp_vport; struct mlxsw_sp_vfid *vfid; + u16 fid; int err; /* VLAN 0 is removed from HW filter when device goes down, but @@ -903,11 +910,8 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, return err; } - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - false, - mlxsw_sp_vfid_to_fid(vfid->vfid), - vid); + fid = mlxsw_sp_vfid_to_fid(vfid->vfid); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, false); if (err) { netdev_err(dev, "Failed to invalidate {Port, VID=%d} to vFID=%d mapping\n", vid, vfid->vfid); @@ -3213,6 +3217,7 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); struct net_device *dev = mlxsw_sp_vport->dev; struct mlxsw_sp_vfid *vfid, *new_vfid; + u16 fid, new_fid; int err; vfid = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); @@ -3233,26 +3238,20 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, /* Invalidate existing {Port, VID} to vFID mapping and create a new * one for the new vFID. */ - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - false, - mlxsw_sp_vfid_to_fid(vfid->vfid), - vid); + fid = mlxsw_sp_vfid_to_fid(vfid->vfid); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, false); if (err) { netdev_err(dev, "Failed to invalidate {Port, VID} to vFID=%d mapping\n", vfid->vfid); - goto err_port_vid_to_fid_invalidate; + goto err_vport_fid_unmap; } - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - true, - mlxsw_sp_vfid_to_fid(new_vfid->vfid), - vid); + new_fid = mlxsw_sp_vfid_to_fid(new_vfid->vfid); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, new_fid, true); if (err) { netdev_err(dev, "Failed to map {Port, VID} to vFID=%d\n", new_vfid->vfid); - goto err_port_vid_to_fid_validate; + goto err_vport_fid_map; } err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); @@ -3294,8 +3293,8 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, err_port_stp_state_set: err_vport_flood_set: err_port_vid_learning_set: -err_port_vid_to_fid_validate: -err_port_vid_to_fid_invalidate: +err_vport_fid_map: +err_vport_fid_unmap: /* Rollback vFID only if new. */ if (!new_vfid->nr_vports) mlxsw_sp_vfid_destroy(mlxsw_sp, new_vfid); @@ -3309,6 +3308,7 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); struct net_device *dev = mlxsw_sp_vport->dev; struct mlxsw_sp_vfid *vfid; + u16 fid, old_fid; int err; vfid = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); @@ -3336,26 +3336,20 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, /* We need to invalidate existing {Port, VID} to vFID mapping and * create a new one for the bridge's vFID. */ - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - false, - mlxsw_sp_vfid_to_fid(old_vfid->vfid), - vid); + old_fid = mlxsw_sp_vfid_to_fid(old_vfid->vfid); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, old_fid, false); if (err) { netdev_err(dev, "Failed to invalidate {Port, VID} to vFID=%d mapping\n", old_vfid->vfid); - goto err_port_vid_to_fid_invalidate; + goto err_vport_fid_unmap; } - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, - true, - mlxsw_sp_vfid_to_fid(vfid->vfid), - vid); + fid = mlxsw_sp_vfid_to_fid(vfid->vfid); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, true); if (err) { netdev_err(dev, "Failed to map {Port, VID} to vFID=%d\n", vfid->vfid); - goto err_port_vid_to_fid_validate; + goto err_vport_fid_map; } /* Switch between the vFIDs and destroy the old one if needed. */ @@ -3372,11 +3366,9 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, return 0; -err_port_vid_to_fid_validate: - mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, - MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, false, - mlxsw_sp_vfid_to_fid(old_vfid->vfid), vid); -err_port_vid_to_fid_invalidate: +err_vport_fid_map: + mlxsw_sp_vport_fid_map(mlxsw_sp_vport, old_fid, true); +err_vport_fid_unmap: mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); err_port_vid_learning_set: mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false); -- cgit v0.10.2 From e6060027215d7e21f240f378b96c223ab438995c Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:11 +0200 Subject: mlxsw: spectrum: Use FID instead of vFID to setup flooding Use a FID index instead of vFID and ease the transition towards a generic FID struct. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 6df944b..7a75a45 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -793,9 +793,9 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, goto err_port_vport_create; } + fid = mlxsw_sp_vfid_to_fid(vfid->vfid); if (!vfid->nr_vports) { - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, - true); + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, fid, true); if (err) { netdev_err(dev, "Failed to setup flooding for vFID=%d\n", vfid->vfid); @@ -815,7 +815,6 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, } } - fid = mlxsw_sp_vfid_to_fid(vfid->vfid); err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, true); if (err) { netdev_err(dev, "Failed to map {Port, VID=%d} to vFID=%d\n", @@ -858,7 +857,7 @@ err_vport_fid_map: mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port); err_port_vp_mode_trans: if (!vfid->nr_vports) - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false); + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, fid, false); err_vport_flood_set: mlxsw_sp_port_vport_destroy(mlxsw_sp_vport); err_port_vport_create: @@ -3260,7 +3259,7 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, goto err_port_vid_learning_set; } - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false); + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, fid, false); if (err) { netdev_err(dev, "Failed clear to clear flooding\n"); goto err_vport_flood_set; @@ -3320,7 +3319,8 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, } } - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, true); + fid = mlxsw_sp_vfid_to_fid(vfid->vfid); + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, fid, true); if (err) { netdev_err(dev, "Failed to setup flooding for vFID=%d\n", vfid->vfid); @@ -3344,7 +3344,6 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, goto err_vport_fid_unmap; } - fid = mlxsw_sp_vfid_to_fid(vfid->vfid); err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, true); if (err) { netdev_err(dev, "Failed to map {Port, VID} to vFID=%d\n", @@ -3371,7 +3370,7 @@ err_vport_fid_map: err_vport_fid_unmap: mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); err_port_vid_learning_set: - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, vfid->vfid, false); + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, fid, false); err_port_flood_set: if (!vfid->nr_vports) mlxsw_sp_br_vfid_destroy(mlxsw_sp, vfid); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 58a2a5d..0ae929e 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -366,7 +366,7 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, u16 vid); int mlxsw_sp_port_kill_vid(struct net_device *dev, __be16 __always_unused proto, u16 vid); -int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vfid, +int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, bool set); void mlxsw_sp_port_active_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port); int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 02c126c..76f53c2 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -260,12 +260,15 @@ err_port_flood_set: return err; } -int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vfid, +int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, bool set) { + u16 vfid; + /* In case of vFIDs, index into the flooding table is relative to * the start of the vFIDs range. */ + vfid = mlxsw_sp_fid_to_vfid(fid); return __mlxsw_sp_port_flood_set(mlxsw_sp_vport, vfid, vfid, set, false); } -- cgit v0.10.2 From d0ec875a2f2eaacef59ad182dc2a061a7387013e Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:12 +0200 Subject: mlxsw: spectrum: Make vFID struct generic Up until now we had a dedicated struct only for vFIDs, but before introducing support for L3 interfaces we need to make it generic and use it for all three types of FIDs: 1) FIDs - 0..4K-1, used for the VLAN-aware bridge 2) vFIDs - 4K..15K-1, used for VLAN-unaware bridges 3) rFIDs - 15K..16K-1, used to direct traffic to / from the router in the device. Will be introduced later in the series. The three types of L3 interfaces - Router InterFaces, RIFs - that will be introduced correspond to the three types of FIDs and are configured using them. Therefore, we'll need to store the links between them as well as a reference count on the underlying FID, so that the corresponding RIF will be destroyed when it reaches zero. Note that the lower 0.5K vFIDs are currently used for for non-bridged netdevs, so that traffic could be flooded to the CPU port. However, when rFIDs will be introduced we'll no longer need these and they too will be used for VLAN-unaware bridges. Make the vFID struct generic by renaming it and some of its fields. FIDs will be converted to use it later in the series. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 7a75a45..8ef8d0b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -633,14 +633,14 @@ static int mlxsw_sp_port_vlan_mode_trans(struct mlxsw_sp_port *mlxsw_sp_port) return 0; } -static struct mlxsw_sp_vfid * +static struct mlxsw_sp_fid * mlxsw_sp_vfid_find(const struct mlxsw_sp *mlxsw_sp, u16 vid) { - struct mlxsw_sp_vfid *vfid; + struct mlxsw_sp_fid *f; - list_for_each_entry(vfid, &mlxsw_sp->port_vfids.list, list) { - if (vfid->vid == vid) - return vfid; + list_for_each_entry(f, &mlxsw_sp->port_vfids.list, list) { + if (f->vid == vid) + return f; } return NULL; @@ -660,11 +660,11 @@ static int mlxsw_sp_vfid_op(struct mlxsw_sp *mlxsw_sp, u16 fid, bool create) return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); } -static struct mlxsw_sp_vfid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, - u16 vid) +static struct mlxsw_sp_fid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, + u16 vid) { struct device *dev = mlxsw_sp->bus_info->dev; - struct mlxsw_sp_vfid *f; + struct mlxsw_sp_fid *f; u16 vfid, fid; int err; @@ -685,7 +685,7 @@ static struct mlxsw_sp_vfid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, if (!f) goto err_allocate_vfid; - f->vfid = vfid; + f->fid = fid; f->vid = vid; list_add(&f->list, &mlxsw_sp->port_vfids.list); @@ -699,21 +699,21 @@ err_allocate_vfid: } static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_vfid *vfid) + struct mlxsw_sp_fid *f) { - u16 fid = mlxsw_sp_vfid_to_fid(vfid->vfid); + u16 vfid = mlxsw_sp_fid_to_vfid(f->fid); - clear_bit(vfid->vfid, mlxsw_sp->port_vfids.mapped); - list_del(&vfid->list); + clear_bit(vfid, mlxsw_sp->port_vfids.mapped); + list_del(&f->list); - mlxsw_sp_vfid_op(mlxsw_sp, fid, false); + mlxsw_sp_vfid_op(mlxsw_sp, f->fid, false); - kfree(vfid); + kfree(f); } static struct mlxsw_sp_port * mlxsw_sp_port_vport_create(struct mlxsw_sp_port *mlxsw_sp_port, - struct mlxsw_sp_vfid *vfid) + struct mlxsw_sp_fid *f) { struct mlxsw_sp_port *mlxsw_sp_vport; @@ -731,8 +731,8 @@ mlxsw_sp_port_vport_create(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_vport->stp_state = BR_STATE_FORWARDING; mlxsw_sp_vport->lagged = mlxsw_sp_port->lagged; mlxsw_sp_vport->lag_id = mlxsw_sp_port->lag_id; - mlxsw_sp_vport->vport.vfid = vfid; - mlxsw_sp_vport->vport.vid = vfid->vid; + mlxsw_sp_vport->vport.f = f; + mlxsw_sp_vport->vport.vid = f->vid; list_add(&mlxsw_sp_vport->vport.list, &mlxsw_sp_port->vports_list); @@ -761,8 +761,7 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct mlxsw_sp_port *mlxsw_sp_vport; - struct mlxsw_sp_vfid *vfid; - u16 fid; + struct mlxsw_sp_fid *f; int err; /* VLAN 0 is added to HW filter when device goes up, but it is @@ -776,29 +775,28 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, return 0; } - vfid = mlxsw_sp_vfid_find(mlxsw_sp, vid); - if (!vfid) { - vfid = mlxsw_sp_vfid_create(mlxsw_sp, vid); - if (IS_ERR(vfid)) { + f = mlxsw_sp_vfid_find(mlxsw_sp, vid); + if (!f) { + f = mlxsw_sp_vfid_create(mlxsw_sp, vid); + if (IS_ERR(f)) { netdev_err(dev, "Failed to create vFID for VID=%d\n", vid); - return PTR_ERR(vfid); + return PTR_ERR(f); } } - mlxsw_sp_vport = mlxsw_sp_port_vport_create(mlxsw_sp_port, vfid); + mlxsw_sp_vport = mlxsw_sp_port_vport_create(mlxsw_sp_port, f); if (!mlxsw_sp_vport) { netdev_err(dev, "Failed to create vPort for VID=%d\n", vid); err = -ENOMEM; goto err_port_vport_create; } - fid = mlxsw_sp_vfid_to_fid(vfid->vfid); - if (!vfid->nr_vports) { - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, fid, true); + if (!f->ref_count) { + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, true); if (err) { - netdev_err(dev, "Failed to setup flooding for vFID=%d\n", - vfid->vfid); + netdev_err(dev, "Failed to setup flooding for FID=%d\n", + f->fid); goto err_vport_flood_set; } } @@ -815,10 +813,10 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, } } - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, true); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, true); if (err) { - netdev_err(dev, "Failed to map {Port, VID=%d} to vFID=%d\n", - vid, vfid->vfid); + netdev_err(dev, "Failed to map {Port, VID=%d} to FID=%d\n", + vid, f->fid); goto err_vport_fid_map; } @@ -842,7 +840,7 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, goto err_port_stp_state_set; } - vfid->nr_vports++; + f->ref_count++; return 0; @@ -851,18 +849,18 @@ err_port_stp_state_set: err_port_add_vid: mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true); err_port_vid_learning_set: - mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, false); + mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); err_vport_fid_map: if (list_is_singular(&mlxsw_sp_port->vports_list)) mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port); err_port_vp_mode_trans: - if (!vfid->nr_vports) - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, fid, false); + if (!f->ref_count) + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); err_vport_flood_set: mlxsw_sp_port_vport_destroy(mlxsw_sp_vport); err_port_vport_create: - if (!vfid->nr_vports) - mlxsw_sp_vfid_destroy(mlxsw_sp, vfid); + if (!f->ref_count) + mlxsw_sp_vfid_destroy(mlxsw_sp, f); return err; } @@ -871,8 +869,7 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); struct mlxsw_sp_port *mlxsw_sp_vport; - struct mlxsw_sp_vfid *vfid; - u16 fid; + struct mlxsw_sp_fid *f; int err; /* VLAN 0 is removed from HW filter when device goes down, but @@ -887,7 +884,7 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, return 0; } - vfid = mlxsw_sp_vport->vport.vfid; + f = mlxsw_sp_vport->vport.f; err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, MLXSW_REG_SPMS_STATE_DISCARDING); @@ -909,11 +906,10 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, return err; } - fid = mlxsw_sp_vfid_to_fid(vfid->vfid); - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, false); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); if (err) { - netdev_err(dev, "Failed to invalidate {Port, VID=%d} to vFID=%d mapping\n", - vid, vfid->vfid); + netdev_err(dev, "Failed to invalidate {Port, VID=%d} to FID=%d mapping\n", + vid, f->fid); return err; } @@ -929,12 +925,12 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, } } - vfid->nr_vports--; + f->ref_count--; mlxsw_sp_port_vport_destroy(mlxsw_sp_vport); /* Destroy the vFID if no vPorts are assigned to it anymore. */ - if (!vfid->nr_vports) - mlxsw_sp_vfid_destroy(mlxsw_sp_port->mlxsw_sp, vfid); + if (!f->ref_count) + mlxsw_sp_vfid_destroy(mlxsw_sp_port->mlxsw_sp, f); return 0; } @@ -2631,8 +2627,7 @@ static int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port) static int mlxsw_sp_vport_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_vport) { - u16 vfid = mlxsw_sp_vport_vfid_get(mlxsw_sp_vport); - u16 fid = mlxsw_sp_vfid_to_fid(vfid); + u16 fid = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); if (mlxsw_sp_vport->lagged) return mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_vport, @@ -3126,15 +3121,15 @@ static int mlxsw_sp_netdevice_lag_event(struct net_device *lag_dev, return 0; } -static struct mlxsw_sp_vfid * +static struct mlxsw_sp_fid * mlxsw_sp_br_vfid_find(const struct mlxsw_sp *mlxsw_sp, const struct net_device *br_dev) { - struct mlxsw_sp_vfid *vfid; + struct mlxsw_sp_fid *f; - list_for_each_entry(vfid, &mlxsw_sp->br_vfids.list, list) { - if (vfid->br_dev == br_dev) - return vfid; + list_for_each_entry(f, &mlxsw_sp->br_vfids.list, list) { + if (f->dev == br_dev) + return f; } return NULL; @@ -3156,11 +3151,11 @@ static u16 mlxsw_sp_avail_br_vfid_get(const struct mlxsw_sp *mlxsw_sp) MLXSW_SP_VFID_BR_MAX); } -static struct mlxsw_sp_vfid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, - struct net_device *br_dev) +static struct mlxsw_sp_fid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev) { struct device *dev = mlxsw_sp->bus_info->dev; - struct mlxsw_sp_vfid *f; + struct mlxsw_sp_fid *f; u16 vfid, fid; int err; @@ -3181,8 +3176,8 @@ static struct mlxsw_sp_vfid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, if (!f) goto err_allocate_vfid; - f->vfid = vfid; - f->br_dev = br_dev; + f->fid = fid; + f->dev = br_dev; list_add(&f->list, &mlxsw_sp->br_vfids.list); set_bit(mlxsw_sp_vfid_to_br_vfid(vfid), mlxsw_sp->br_vfids.mapped); @@ -3195,17 +3190,17 @@ err_allocate_vfid: } static void mlxsw_sp_br_vfid_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_vfid *vfid) + struct mlxsw_sp_fid *f) { - u16 br_vfid = mlxsw_sp_vfid_to_br_vfid(vfid->vfid); - u16 fid = mlxsw_sp_vfid_to_fid(vfid->vfid); + u16 vfid = mlxsw_sp_fid_to_vfid(f->fid); + u16 br_vfid = mlxsw_sp_vfid_to_br_vfid(vfid); clear_bit(br_vfid, mlxsw_sp->br_vfids.mapped); - list_del(&vfid->list); + list_del(&f->list); - mlxsw_sp_vfid_op(mlxsw_sp, fid, false); + mlxsw_sp_vfid_op(mlxsw_sp, f->fid, false); - kfree(vfid); + kfree(f); } static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, @@ -3215,19 +3210,18 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); struct net_device *dev = mlxsw_sp_vport->dev; - struct mlxsw_sp_vfid *vfid, *new_vfid; - u16 fid, new_fid; + struct mlxsw_sp_fid *f, *new_f; int err; - vfid = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); - if (WARN_ON(!vfid)) + f = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); + if (WARN_ON(!f)) return; /* We need a vFID to go back to after leaving the bridge's vFID. */ - new_vfid = mlxsw_sp_vfid_find(mlxsw_sp, vid); - if (!new_vfid) { - new_vfid = mlxsw_sp_vfid_create(mlxsw_sp, vid); - if (IS_ERR(new_vfid)) { + new_f = mlxsw_sp_vfid_find(mlxsw_sp, vid); + if (!new_f) { + new_f = mlxsw_sp_vfid_create(mlxsw_sp, vid); + if (IS_ERR(new_f)) { netdev_err(dev, "Failed to create vFID for VID=%d\n", vid); return; @@ -3237,19 +3231,17 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, /* Invalidate existing {Port, VID} to vFID mapping and create a new * one for the new vFID. */ - fid = mlxsw_sp_vfid_to_fid(vfid->vfid); - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, false); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); if (err) { - netdev_err(dev, "Failed to invalidate {Port, VID} to vFID=%d mapping\n", - vfid->vfid); + netdev_err(dev, "Failed to invalidate {Port, VID} to FID=%d mapping\n", + f->fid); goto err_vport_fid_unmap; } - new_fid = mlxsw_sp_vfid_to_fid(new_vfid->vfid); - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, new_fid, true); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, new_f->fid, true); if (err) { - netdev_err(dev, "Failed to map {Port, VID} to vFID=%d\n", - new_vfid->vfid); + netdev_err(dev, "Failed to map {Port, VID} to FID=%d\n", + new_f->fid); goto err_vport_fid_map; } @@ -3259,7 +3251,7 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, goto err_port_vid_learning_set; } - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, fid, false); + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); if (err) { netdev_err(dev, "Failed clear to clear flooding\n"); goto err_vport_flood_set; @@ -3276,11 +3268,11 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, netdev_err(dev, "Failed to flush FDB\n"); /* Switch between the vFIDs and destroy the old one if needed. */ - new_vfid->nr_vports++; - mlxsw_sp_vport->vport.vfid = new_vfid; - vfid->nr_vports--; - if (!vfid->nr_vports) - mlxsw_sp_br_vfid_destroy(mlxsw_sp, vfid); + new_f->ref_count++; + mlxsw_sp_vport->vport.f = new_f; + f->ref_count--; + if (!f->ref_count) + mlxsw_sp_br_vfid_destroy(mlxsw_sp, f); mlxsw_sp_vport->learning = 0; mlxsw_sp_vport->learning_sync = 0; @@ -3295,35 +3287,33 @@ err_port_vid_learning_set: err_vport_fid_map: err_vport_fid_unmap: /* Rollback vFID only if new. */ - if (!new_vfid->nr_vports) - mlxsw_sp_vfid_destroy(mlxsw_sp, new_vfid); + if (!new_f->ref_count) + mlxsw_sp_vfid_destroy(mlxsw_sp, new_f); } static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, struct net_device *br_dev) { - struct mlxsw_sp_vfid *old_vfid = mlxsw_sp_vport->vport.vfid; + struct mlxsw_sp_fid *old_f = mlxsw_sp_vport->vport.f; struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); struct net_device *dev = mlxsw_sp_vport->dev; - struct mlxsw_sp_vfid *vfid; - u16 fid, old_fid; + struct mlxsw_sp_fid *f; int err; - vfid = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); - if (!vfid) { - vfid = mlxsw_sp_br_vfid_create(mlxsw_sp, br_dev); - if (IS_ERR(vfid)) { + f = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); + if (!f) { + f = mlxsw_sp_br_vfid_create(mlxsw_sp, br_dev); + if (IS_ERR(f)) { netdev_err(dev, "Failed to create bridge vFID\n"); - return PTR_ERR(vfid); + return PTR_ERR(f); } } - fid = mlxsw_sp_vfid_to_fid(vfid->vfid); - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, fid, true); + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, true); if (err) { - netdev_err(dev, "Failed to setup flooding for vFID=%d\n", - vfid->vfid); + netdev_err(dev, "Failed to setup flooding for FID=%d\n", + f->fid); goto err_port_flood_set; } @@ -3336,27 +3326,26 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, /* We need to invalidate existing {Port, VID} to vFID mapping and * create a new one for the bridge's vFID. */ - old_fid = mlxsw_sp_vfid_to_fid(old_vfid->vfid); - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, old_fid, false); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, old_f->fid, false); if (err) { - netdev_err(dev, "Failed to invalidate {Port, VID} to vFID=%d mapping\n", - old_vfid->vfid); + netdev_err(dev, "Failed to invalidate {Port, VID} to FID=%d mapping\n", + old_f->fid); goto err_vport_fid_unmap; } - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, fid, true); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, true); if (err) { - netdev_err(dev, "Failed to map {Port, VID} to vFID=%d\n", - vfid->vfid); + netdev_err(dev, "Failed to map {Port, VID} to FID=%d\n", + f->fid); goto err_vport_fid_map; } /* Switch between the vFIDs and destroy the old one if needed. */ - vfid->nr_vports++; - mlxsw_sp_vport->vport.vfid = vfid; - old_vfid->nr_vports--; - if (!old_vfid->nr_vports) - mlxsw_sp_vfid_destroy(mlxsw_sp, old_vfid); + f->ref_count++; + mlxsw_sp_vport->vport.f = f; + old_f->ref_count--; + if (!old_f->ref_count) + mlxsw_sp_vfid_destroy(mlxsw_sp, old_f); mlxsw_sp_vport->learning = 1; mlxsw_sp_vport->learning_sync = 1; @@ -3366,14 +3355,14 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, return 0; err_vport_fid_map: - mlxsw_sp_vport_fid_map(mlxsw_sp_vport, old_fid, true); + mlxsw_sp_vport_fid_map(mlxsw_sp_vport, old_f->fid, true); err_vport_fid_unmap: mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); err_port_vid_learning_set: - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, fid, false); + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); err_port_flood_set: - if (!vfid->nr_vports) - mlxsw_sp_br_vfid_destroy(mlxsw_sp, vfid); + if (!f->ref_count) + mlxsw_sp_br_vfid_destroy(mlxsw_sp, f); return err; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 0ae929e..c973ab5 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -87,11 +87,11 @@ struct mlxsw_sp_upper { unsigned int ref_count; }; -struct mlxsw_sp_vfid { +struct mlxsw_sp_fid { struct list_head list; - u16 nr_vports; - u16 vfid; /* Starting at 0 */ - struct net_device *br_dev; + unsigned int ref_count; + struct net_device *dev; + u16 fid; u16 vid; }; @@ -217,7 +217,7 @@ struct mlxsw_sp_port { u16 lag_id; struct { struct list_head list; - struct mlxsw_sp_vfid *vfid; + struct mlxsw_sp_fid *f; u16 vid; } vport; struct { @@ -262,13 +262,13 @@ mlxsw_sp_port_lagged_get(struct mlxsw_sp *mlxsw_sp, u16 lag_id, u8 port_index) static inline bool mlxsw_sp_port_is_vport(const struct mlxsw_sp_port *mlxsw_sp_port) { - return mlxsw_sp_port->vport.vfid; + return mlxsw_sp_port->vport.f; } static inline struct net_device * mlxsw_sp_vport_br_get(const struct mlxsw_sp_port *mlxsw_sp_vport) { - return mlxsw_sp_vport->vport.vfid->br_dev; + return mlxsw_sp_vport->vport.f->dev; } static inline u16 @@ -278,9 +278,9 @@ mlxsw_sp_vport_vid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) } static inline u16 -mlxsw_sp_vport_vfid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) +mlxsw_sp_vport_fid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) { - return mlxsw_sp_vport->vport.vfid->vfid; + return mlxsw_sp_vport->vport.f->fid; } static inline struct mlxsw_sp_port * @@ -298,14 +298,14 @@ mlxsw_sp_port_vport_find(const struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) } static inline struct mlxsw_sp_port * -mlxsw_sp_port_vport_find_by_vfid(const struct mlxsw_sp_port *mlxsw_sp_port, - u16 vfid) +mlxsw_sp_port_vport_find_by_fid(const struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid) { struct mlxsw_sp_port *mlxsw_sp_vport; list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list, vport.list) { - if (mlxsw_sp_vport_vfid_get(mlxsw_sp_vport) == vfid) + if (mlxsw_sp_vport_fid_get(mlxsw_sp_vport) == fid) return mlxsw_sp_vport; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 76f53c2..3e9ba58 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -57,11 +57,8 @@ static u16 mlxsw_sp_port_vid_to_fid_get(struct mlxsw_sp_port *mlxsw_sp_port, { u16 fid = vid; - if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) { - u16 vfid = mlxsw_sp_vport_vfid_get(mlxsw_sp_port); - - fid = mlxsw_sp_vfid_to_fid(vfid); - } + if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) + fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port); if (!fid) fid = mlxsw_sp_port->pvid; @@ -236,8 +233,9 @@ static int mlxsw_sp_port_uc_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, int err; if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) { - u16 vfid = mlxsw_sp_vport_vfid_get(mlxsw_sp_port); + u16 vfid, fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port); + vfid = mlxsw_sp_fid_to_vfid(fid); return __mlxsw_sp_port_flood_set(mlxsw_sp_port, vfid, vfid, set, true); } @@ -1136,12 +1134,8 @@ static int mlxsw_sp_port_fdb_dump(struct mlxsw_sp_port *mlxsw_sp_port, if (!sfd_pl) return -ENOMEM; - if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) { - u16 tmp; - - tmp = mlxsw_sp_vport_vfid_get(mlxsw_sp_port); - vport_fid = mlxsw_sp_vfid_to_fid(tmp); - } + if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) + vport_fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port); mlxsw_reg_sfd_pack(sfd_pl, MLXSW_REG_SFD_OP_QUERY_DUMP, 0); do { @@ -1313,11 +1307,10 @@ static void mlxsw_sp_fdb_notify_mac_process(struct mlxsw_sp *mlxsw_sp, } if (mlxsw_sp_fid_is_vfid(fid)) { - u16 vfid = mlxsw_sp_fid_to_vfid(fid); struct mlxsw_sp_port *mlxsw_sp_vport; - mlxsw_sp_vport = mlxsw_sp_port_vport_find_by_vfid(mlxsw_sp_port, - vfid); + mlxsw_sp_vport = mlxsw_sp_port_vport_find_by_fid(mlxsw_sp_port, + fid); if (!mlxsw_sp_vport) { netdev_err(mlxsw_sp_port->dev, "Failed to find a matching vPort following FDB notification\n"); goto just_remove; @@ -1373,11 +1366,10 @@ static void mlxsw_sp_fdb_notify_mac_lag_process(struct mlxsw_sp *mlxsw_sp, } if (mlxsw_sp_fid_is_vfid(fid)) { - u16 vfid = mlxsw_sp_fid_to_vfid(fid); struct mlxsw_sp_port *mlxsw_sp_vport; - mlxsw_sp_vport = mlxsw_sp_port_vport_find_by_vfid(mlxsw_sp_port, - vfid); + mlxsw_sp_vport = mlxsw_sp_port_vport_find_by_fid(mlxsw_sp_port, + fid); if (!mlxsw_sp_vport) { netdev_err(mlxsw_sp_port->dev, "Failed to find a matching vPort following FDB notification\n"); goto just_remove; -- cgit v0.10.2 From 0355b59fbb523c86f020ba2e970cbdd420060dfe Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:13 +0200 Subject: mlxsw: spectrum: Use join / leave functions for vFID operations When a vPort is created or when it joins a bridge we always do the same set of operations: 1) Create the vFID, if not already created 2) Setup flooding for the vFID 3) Map the {Port, VID} to the vFID When a vPort is destroyed or when it leaves a bridge the reverse is performed. Encapsulate the above in join / leave functions and simplify the code. FIDs and rFIDs will use a similar set of functions. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 8ef8d0b..ad8e447 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -712,8 +712,7 @@ static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, } static struct mlxsw_sp_port * -mlxsw_sp_port_vport_create(struct mlxsw_sp_port *mlxsw_sp_port, - struct mlxsw_sp_fid *f) +mlxsw_sp_port_vport_create(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) { struct mlxsw_sp_port *mlxsw_sp_vport; @@ -731,8 +730,7 @@ mlxsw_sp_port_vport_create(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_vport->stp_state = BR_STATE_FORWARDING; mlxsw_sp_vport->lagged = mlxsw_sp_port->lagged; mlxsw_sp_vport->lag_id = mlxsw_sp_port->lag_id; - mlxsw_sp_vport->vport.f = f; - mlxsw_sp_vport->vport.vid = f->vid; + mlxsw_sp_vport->vport.vid = vid; list_add(&mlxsw_sp_vport->vport.list, &mlxsw_sp_port->vports_list); @@ -755,13 +753,62 @@ static int mlxsw_sp_vport_fid_map(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, vid); } +static int mlxsw_sp_vport_vfid_join(struct mlxsw_sp_port *mlxsw_sp_vport) +{ + u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); + struct mlxsw_sp_fid *f; + int err; + + f = mlxsw_sp_vfid_find(mlxsw_sp_vport->mlxsw_sp, vid); + if (!f) { + f = mlxsw_sp_vfid_create(mlxsw_sp_vport->mlxsw_sp, vid); + if (IS_ERR(f)) + return PTR_ERR(f); + } + + if (!f->ref_count) { + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, true); + if (err) + goto err_vport_flood_set; + } + + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, true); + if (err) + goto err_vport_fid_map; + + mlxsw_sp_vport->vport.f = f; + f->ref_count++; + + return 0; + +err_vport_fid_map: + if (!f->ref_count) + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); +err_vport_flood_set: + if (!f->ref_count) + mlxsw_sp_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); + return err; +} + +static void mlxsw_sp_vport_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) +{ + struct mlxsw_sp_fid *f = mlxsw_sp_vport->vport.f; + + mlxsw_sp_vport->vport.f = NULL; + + mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); + + if (--f->ref_count == 0) { + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); + mlxsw_sp_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); + } +} + int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, u16 vid) { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct mlxsw_sp_port *mlxsw_sp_vport; - struct mlxsw_sp_fid *f; int err; /* VLAN 0 is added to HW filter when device goes up, but it is @@ -775,30 +822,10 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, return 0; } - f = mlxsw_sp_vfid_find(mlxsw_sp, vid); - if (!f) { - f = mlxsw_sp_vfid_create(mlxsw_sp, vid); - if (IS_ERR(f)) { - netdev_err(dev, "Failed to create vFID for VID=%d\n", - vid); - return PTR_ERR(f); - } - } - - mlxsw_sp_vport = mlxsw_sp_port_vport_create(mlxsw_sp_port, f); + mlxsw_sp_vport = mlxsw_sp_port_vport_create(mlxsw_sp_port, vid); if (!mlxsw_sp_vport) { netdev_err(dev, "Failed to create vPort for VID=%d\n", vid); - err = -ENOMEM; - goto err_port_vport_create; - } - - if (!f->ref_count) { - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, true); - if (err) { - netdev_err(dev, "Failed to setup flooding for FID=%d\n", - f->fid); - goto err_vport_flood_set; - } + return -ENOMEM; } /* When adding the first VLAN interface on a bridged port we need to @@ -813,11 +840,10 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, } } - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, true); + err = mlxsw_sp_vport_vfid_join(mlxsw_sp_vport); if (err) { - netdev_err(dev, "Failed to map {Port, VID=%d} to FID=%d\n", - vid, f->fid); - goto err_vport_fid_map; + netdev_err(dev, "Failed to join vFID\n"); + goto err_vport_vfid_join; } err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); @@ -840,8 +866,6 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, goto err_port_stp_state_set; } - f->ref_count++; - return 0; err_port_stp_state_set: @@ -849,18 +873,12 @@ err_port_stp_state_set: err_port_add_vid: mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true); err_port_vid_learning_set: - mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); -err_vport_fid_map: + mlxsw_sp_vport_vfid_leave(mlxsw_sp_vport); +err_vport_vfid_join: if (list_is_singular(&mlxsw_sp_port->vports_list)) mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port); err_port_vp_mode_trans: - if (!f->ref_count) - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); -err_vport_flood_set: mlxsw_sp_port_vport_destroy(mlxsw_sp_vport); -err_port_vport_create: - if (!f->ref_count) - mlxsw_sp_vfid_destroy(mlxsw_sp, f); return err; } @@ -869,7 +887,6 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); struct mlxsw_sp_port *mlxsw_sp_vport; - struct mlxsw_sp_fid *f; int err; /* VLAN 0 is removed from HW filter when device goes down, but @@ -884,8 +901,6 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, return 0; } - f = mlxsw_sp_vport->vport.f; - err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, MLXSW_REG_SPMS_STATE_DISCARDING); if (err) { @@ -906,12 +921,7 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, return err; } - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); - if (err) { - netdev_err(dev, "Failed to invalidate {Port, VID=%d} to FID=%d mapping\n", - vid, f->fid); - return err; - } + mlxsw_sp_vport_vfid_leave(mlxsw_sp_vport); /* When removing the last VLAN interface on a bridged port we need to * transition all active 802.1Q bridge VLANs to use VID to FID @@ -925,13 +935,8 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, } } - f->ref_count--; mlxsw_sp_port_vport_destroy(mlxsw_sp_vport); - /* Destroy the vFID if no vPorts are assigned to it anymore. */ - if (!f->ref_count) - mlxsw_sp_vfid_destroy(mlxsw_sp_port->mlxsw_sp, f); - return 0; } @@ -2625,10 +2630,9 @@ static int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port) return mlxsw_sp_port_fdb_flush_by_port(mlxsw_sp_port); } -static int mlxsw_sp_vport_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_vport) +static int mlxsw_sp_vport_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_vport, + u16 fid) { - u16 fid = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); - if (mlxsw_sp_vport->lagged) return mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_vport, fid); @@ -3203,118 +3207,66 @@ static void mlxsw_sp_br_vfid_destroy(struct mlxsw_sp *mlxsw_sp, kfree(f); } -static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, - struct net_device *br_dev, - bool flush_fdb) +static int mlxsw_sp_vport_br_vfid_join(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *br_dev) { - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; - u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); - struct net_device *dev = mlxsw_sp_vport->dev; - struct mlxsw_sp_fid *f, *new_f; + struct mlxsw_sp_fid *f; int err; - f = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); - if (WARN_ON(!f)) - return; - - /* We need a vFID to go back to after leaving the bridge's vFID. */ - new_f = mlxsw_sp_vfid_find(mlxsw_sp, vid); - if (!new_f) { - new_f = mlxsw_sp_vfid_create(mlxsw_sp, vid); - if (IS_ERR(new_f)) { - netdev_err(dev, "Failed to create vFID for VID=%d\n", - vid); - return; - } + f = mlxsw_sp_br_vfid_find(mlxsw_sp_vport->mlxsw_sp, br_dev); + if (!f) { + f = mlxsw_sp_br_vfid_create(mlxsw_sp_vport->mlxsw_sp, br_dev); + if (IS_ERR(f)) + return PTR_ERR(f); } - /* Invalidate existing {Port, VID} to vFID mapping and create a new - * one for the new vFID. - */ - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); - if (err) { - netdev_err(dev, "Failed to invalidate {Port, VID} to FID=%d mapping\n", - f->fid); - goto err_vport_fid_unmap; - } + err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, true); + if (err) + goto err_vport_flood_set; - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, new_f->fid, true); - if (err) { - netdev_err(dev, "Failed to map {Port, VID} to FID=%d\n", - new_f->fid); + err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, true); + if (err) goto err_vport_fid_map; - } - - err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); - if (err) { - netdev_err(dev, "Failed to disable learning\n"); - goto err_port_vid_learning_set; - } - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); - if (err) { - netdev_err(dev, "Failed clear to clear flooding\n"); - goto err_vport_flood_set; - } - - err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, - MLXSW_REG_SPMS_STATE_FORWARDING); - if (err) { - netdev_err(dev, "Failed to set STP state\n"); - goto err_port_stp_state_set; - } + mlxsw_sp_vport->vport.f = f; + f->ref_count++; - if (flush_fdb && mlxsw_sp_vport_fdb_flush(mlxsw_sp_vport)) - netdev_err(dev, "Failed to flush FDB\n"); + return 0; - /* Switch between the vFIDs and destroy the old one if needed. */ - new_f->ref_count++; - mlxsw_sp_vport->vport.f = new_f; - f->ref_count--; +err_vport_fid_map: + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); +err_vport_flood_set: if (!f->ref_count) - mlxsw_sp_br_vfid_destroy(mlxsw_sp, f); + mlxsw_sp_br_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); + return err; +} - mlxsw_sp_vport->learning = 0; - mlxsw_sp_vport->learning_sync = 0; - mlxsw_sp_vport->uc_flood = 0; - mlxsw_sp_vport->bridged = 0; +static void mlxsw_sp_vport_br_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) +{ + struct mlxsw_sp_fid *f = mlxsw_sp_vport->vport.f; - return; + mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); -err_port_stp_state_set: -err_vport_flood_set: -err_port_vid_learning_set: -err_vport_fid_map: -err_vport_fid_unmap: - /* Rollback vFID only if new. */ - if (!new_f->ref_count) - mlxsw_sp_vfid_destroy(mlxsw_sp, new_f); + mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); + + mlxsw_sp_vport->vport.f = NULL; + if (--f->ref_count == 0) + mlxsw_sp_br_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); } static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, struct net_device *br_dev) { - struct mlxsw_sp_fid *old_f = mlxsw_sp_vport->vport.f; - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); struct net_device *dev = mlxsw_sp_vport->dev; - struct mlxsw_sp_fid *f; int err; - f = mlxsw_sp_br_vfid_find(mlxsw_sp, br_dev); - if (!f) { - f = mlxsw_sp_br_vfid_create(mlxsw_sp, br_dev); - if (IS_ERR(f)) { - netdev_err(dev, "Failed to create bridge vFID\n"); - return PTR_ERR(f); - } - } + mlxsw_sp_vport_vfid_leave(mlxsw_sp_vport); - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, true); + err = mlxsw_sp_vport_br_vfid_join(mlxsw_sp_vport, br_dev); if (err) { - netdev_err(dev, "Failed to setup flooding for FID=%d\n", - f->fid); - goto err_port_flood_set; + netdev_err(dev, "Failed to join vFID\n"); + goto err_vport_br_vfid_join; } err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true); @@ -3323,30 +3275,6 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, goto err_port_vid_learning_set; } - /* We need to invalidate existing {Port, VID} to vFID mapping and - * create a new one for the bridge's vFID. - */ - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, old_f->fid, false); - if (err) { - netdev_err(dev, "Failed to invalidate {Port, VID} to FID=%d mapping\n", - old_f->fid); - goto err_vport_fid_unmap; - } - - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, true); - if (err) { - netdev_err(dev, "Failed to map {Port, VID} to FID=%d\n", - f->fid); - goto err_vport_fid_map; - } - - /* Switch between the vFIDs and destroy the old one if needed. */ - f->ref_count++; - mlxsw_sp_vport->vport.f = f; - old_f->ref_count--; - if (!old_f->ref_count) - mlxsw_sp_vfid_destroy(mlxsw_sp, old_f); - mlxsw_sp_vport->learning = 1; mlxsw_sp_vport->learning_sync = 1; mlxsw_sp_vport->uc_flood = 1; @@ -3354,18 +3282,38 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, return 0; -err_vport_fid_map: - mlxsw_sp_vport_fid_map(mlxsw_sp_vport, old_f->fid, true); -err_vport_fid_unmap: - mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); err_port_vid_learning_set: - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); -err_port_flood_set: - if (!f->ref_count) - mlxsw_sp_br_vfid_destroy(mlxsw_sp, f); + mlxsw_sp_vport_br_vfid_leave(mlxsw_sp_vport); +err_vport_br_vfid_join: + mlxsw_sp_vport_vfid_join(mlxsw_sp_vport); return err; } +static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *br_dev, + bool flush_fdb) +{ + u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); + u16 fid = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + + mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); + + mlxsw_sp_vport_br_vfid_leave(mlxsw_sp_vport); + + mlxsw_sp_vport_vfid_join(mlxsw_sp_vport); + + mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, + MLXSW_REG_SPMS_STATE_FORWARDING); + + if (flush_fdb) + mlxsw_sp_vport_fdb_flush(mlxsw_sp_vport, fid); + + mlxsw_sp_vport->learning = 0; + mlxsw_sp_vport->learning_sync = 0; + mlxsw_sp_vport->uc_flood = 0; + mlxsw_sp_vport->bridged = 0; +} + static bool mlxsw_sp_port_master_bridge_check(const struct mlxsw_sp_port *mlxsw_sp_port, const struct net_device *br_dev) -- cgit v0.10.2 From 37286d2571ccc8eda60970074ea0cd641e39fa05 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:14 +0200 Subject: mlxsw: spectrum: Remove unused function argument Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index ad8e447..4869392 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2872,7 +2872,6 @@ err_col_port_add: } static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, - struct net_device *br_dev, bool flush_fdb); static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, @@ -2903,7 +2902,7 @@ static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, continue; br_dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport); - mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, br_dev, false); + mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, false); } if (mlxsw_sp_port->bridged) { @@ -2995,7 +2994,7 @@ static void mlxsw_sp_port_vlan_unlink(struct mlxsw_sp_port *mlxsw_sp_port, struct net_device *br_dev; br_dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport); - mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, br_dev, true); + mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, true); } mlxsw_sp_vport->dev = mlxsw_sp_port->dev; @@ -3290,7 +3289,6 @@ err_vport_br_vfid_join: } static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, - struct net_device *br_dev, bool flush_fdb) { u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); @@ -3369,8 +3367,7 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, */ if (!mlxsw_sp_vport) return 0; - mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, upper_dev, - true); + mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, true); } } -- cgit v0.10.2 From 14d39461b3f4f0f827db64a2f1fc7bc44c452684 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:15 +0200 Subject: mlxsw: spectrum: Use per-FID struct for the VLAN-aware bridge In a very similar way to the vFIDs, make the first 4K FIDs - used in the VLAN-aware bridge - use the new FID struct. Upon first use of the FID by any of the ports do the following: 1) Create the FID 2) Setup a matching flooding entry 3) Create a mapping for the FID Unlike vFIDs, upon creation of a FID we always create a global VID-to-FID mapping, so that ports without upper vPorts can use it instead of creating an explicit {Port, VID} to FID mapping. When a port leaves a FID the reverse is performed. Whenever the FID's reference count reaches zero the FID is deleted along with the global mapping. The per-FID struct will later allow us to configure L3 interfaces on top of the VLAN-aware bridge. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 4869392..08f3f51 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2396,6 +2396,7 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->core = mlxsw_core; mlxsw_sp->bus_info = mlxsw_bus_info; + INIT_LIST_HEAD(&mlxsw_sp->fids); INIT_LIST_HEAD(&mlxsw_sp->port_vfids.list); INIT_LIST_HEAD(&mlxsw_sp->br_vfids.list); INIT_LIST_HEAD(&mlxsw_sp->br_mids.list); @@ -2472,6 +2473,7 @@ static void mlxsw_sp_fini(struct mlxsw_core *mlxsw_core) mlxsw_sp_traps_fini(mlxsw_sp); mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE); mlxsw_sp_ports_remove(mlxsw_sp); + WARN_ON(!list_empty(&mlxsw_sp->fids)); } static struct mlxsw_config_profile mlxsw_sp_config_profile = { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index c973ab5..d6cf6de 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -165,7 +165,7 @@ struct mlxsw_sp { struct list_head list; DECLARE_BITMAP(mapped, MLXSW_SP_MID_MAX); } br_mids; - DECLARE_BITMAP(active_fids, VLAN_N_VID); + struct list_head fids; /* VLAN-aware bridge FIDs */ struct mlxsw_sp_port **ports; struct mlxsw_core *core; const struct mlxsw_bus_info *bus_info; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 3e9ba58..6713769 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -384,6 +384,192 @@ static int mlxsw_sp_port_attr_set(struct net_device *dev, return err; } +static struct mlxsw_sp_fid *mlxsw_sp_fid_find(struct mlxsw_sp *mlxsw_sp, + u16 fid) +{ + struct mlxsw_sp_fid *f; + + list_for_each_entry(f, &mlxsw_sp->fids, list) + if (f->fid == fid) + return f; + + return NULL; +} + +static int mlxsw_sp_fid_op(struct mlxsw_sp *mlxsw_sp, u16 fid, bool create) +{ + char sfmr_pl[MLXSW_REG_SFMR_LEN]; + + mlxsw_reg_sfmr_pack(sfmr_pl, !create, fid, fid); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); +} + +static int mlxsw_sp_fid_map(struct mlxsw_sp *mlxsw_sp, u16 fid, bool valid) +{ + enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_VID_TO_FID; + char svfa_pl[MLXSW_REG_SVFA_LEN]; + + mlxsw_reg_svfa_pack(svfa_pl, 0, mt, valid, fid, fid); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(svfa), svfa_pl); +} + +static struct mlxsw_sp_fid *mlxsw_sp_fid_alloc(u16 fid) +{ + struct mlxsw_sp_fid *f; + + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (!f) + return NULL; + + f->fid = fid; + + return f; +} + +static struct mlxsw_sp_fid *mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, + u16 fid) +{ + struct mlxsw_sp_fid *f; + int err; + + err = mlxsw_sp_fid_op(mlxsw_sp, fid, true); + if (err) + return ERR_PTR(err); + + /* Although all the ports member in the FID might be using a + * {Port, VID} to FID mapping, we create a global VID-to-FID + * mapping. This allows a port to transition to VLAN mode, + * knowing the global mapping exists. + */ + err = mlxsw_sp_fid_map(mlxsw_sp, fid, true); + if (err) + goto err_fid_map; + + f = mlxsw_sp_fid_alloc(fid); + if (!f) { + err = -ENOMEM; + goto err_allocate_fid; + } + + list_add(&f->list, &mlxsw_sp->fids); + + return f; + +err_allocate_fid: + mlxsw_sp_fid_map(mlxsw_sp, fid, false); +err_fid_map: + mlxsw_sp_fid_op(mlxsw_sp, fid, false); + return ERR_PTR(err); +} + +static void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fid *f) +{ + u16 fid = f->fid; + + list_del(&f->list); + + kfree(f); + + mlxsw_sp_fid_op(mlxsw_sp, fid, false); +} + +static int __mlxsw_sp_port_fid_join(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid) +{ + struct mlxsw_sp_fid *f; + + f = mlxsw_sp_fid_find(mlxsw_sp_port->mlxsw_sp, fid); + if (!f) { + f = mlxsw_sp_fid_create(mlxsw_sp_port->mlxsw_sp, fid); + if (IS_ERR(f)) + return PTR_ERR(f); + } + + f->ref_count++; + + return 0; +} + +static void __mlxsw_sp_port_fid_leave(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid) +{ + struct mlxsw_sp_fid *f; + + f = mlxsw_sp_fid_find(mlxsw_sp_port->mlxsw_sp, fid); + if (WARN_ON(!f)) + return; + + if (--f->ref_count == 0) + mlxsw_sp_fid_destroy(mlxsw_sp_port->mlxsw_sp, f); +} + +static int mlxsw_sp_port_fid_map(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid, + bool valid) +{ + enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; + + /* If port doesn't have vPorts, then it can use the global + * VID-to-FID mapping. + */ + if (list_empty(&mlxsw_sp_port->vports_list)) + return 0; + + return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, valid, fid, fid); +} + +static int mlxsw_sp_port_fid_join(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid_begin, u16 fid_end) +{ + int fid, err; + + for (fid = fid_begin; fid <= fid_end; fid++) { + err = __mlxsw_sp_port_fid_join(mlxsw_sp_port, fid); + if (err) + goto err_port_fid_join; + } + + err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, fid_begin, fid_end, + true, false); + if (err) + goto err_port_flood_set; + + for (fid = fid_begin; fid <= fid_end; fid++) { + err = mlxsw_sp_port_fid_map(mlxsw_sp_port, fid, true); + if (err) + goto err_port_fid_map; + } + + return 0; + +err_port_fid_map: + for (fid--; fid >= fid_begin; fid--) + mlxsw_sp_port_fid_map(mlxsw_sp_port, fid, false); + __mlxsw_sp_port_flood_set(mlxsw_sp_port, fid_begin, fid_end, false, + false); +err_port_flood_set: + fid = fid_end; +err_port_fid_join: + for (fid--; fid >= fid_begin; fid--) + __mlxsw_sp_port_fid_leave(mlxsw_sp_port, fid); + return err; +} + +static void mlxsw_sp_port_fid_leave(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid_begin, u16 fid_end) +{ + int fid; + + for (fid = fid_begin; fid <= fid_end; fid++) + mlxsw_sp_port_fid_map(mlxsw_sp_port, fid, false); + + __mlxsw_sp_port_flood_set(mlxsw_sp_port, fid_begin, fid_end, false, + false); + + for (fid = fid_begin; fid <= fid_end; fid++) + __mlxsw_sp_port_fid_leave(mlxsw_sp_port, fid); +} + static int __mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) { @@ -441,55 +627,6 @@ err_port_allow_untagged_set: return err; } -static int mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, u16 fid) -{ - char sfmr_pl[MLXSW_REG_SFMR_LEN]; - int err; - - mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_CREATE_FID, fid, fid); - err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); - - if (err) - return err; - - set_bit(fid, mlxsw_sp->active_fids); - return 0; -} - -static void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, u16 fid) -{ - char sfmr_pl[MLXSW_REG_SFMR_LEN]; - - clear_bit(fid, mlxsw_sp->active_fids); - - mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_DESTROY_FID, - fid, fid); - mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); -} - -static int mlxsw_sp_port_fid_map(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) -{ - enum mlxsw_reg_svfa_mt mt; - - if (!list_empty(&mlxsw_sp_port->vports_list)) - mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; - else - mt = MLXSW_REG_SVFA_MT_VID_TO_FID; - - return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, true, fid, fid); -} - -static int mlxsw_sp_port_fid_unmap(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) -{ - enum mlxsw_reg_svfa_mt mt; - - if (list_empty(&mlxsw_sp_port->vports_list)) - return 0; - - mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; - return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, false, fid, fid); -} - static int mlxsw_sp_port_add_vids(struct net_device *dev, u16 vid_begin, u16 vid_end) { @@ -534,10 +671,8 @@ static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin, u16 vid_end, bool flag_untagged, bool flag_pvid) { - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct net_device *dev = mlxsw_sp_port->dev; - u16 vid, last_visited_vid, old_pvid; - enum mlxsw_reg_svfa_mt mt; + u16 vid, old_pvid; int err; /* In case this is invoked with BRIDGE_FLAGS_SELF and port is @@ -547,44 +682,10 @@ static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port, if (!mlxsw_sp_port->bridged) return mlxsw_sp_port_add_vids(dev, vid_begin, vid_end); - for (vid = vid_begin; vid <= vid_end; vid++) { - if (!test_bit(vid, mlxsw_sp->active_fids)) { - err = mlxsw_sp_fid_create(mlxsw_sp, vid); - if (err) { - netdev_err(dev, "Failed to create FID=%d\n", - vid); - return err; - } - - /* When creating a FID, we set a VID to FID mapping - * regardless of the port's mode. - */ - mt = MLXSW_REG_SVFA_MT_VID_TO_FID; - err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, - true, vid, vid); - if (err) { - netdev_err(dev, "Failed to create FID=VID=%d mapping\n", - vid); - goto err_port_vid_to_fid_set; - } - } - } - - /* Set FID mapping according to port's mode */ - for (vid = vid_begin; vid <= vid_end; vid++) { - err = mlxsw_sp_port_fid_map(mlxsw_sp_port, vid); - if (err) { - netdev_err(dev, "Failed to map FID=%d", vid); - last_visited_vid = --vid; - goto err_port_fid_map; - } - } - - err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end, - true, false); + err = mlxsw_sp_port_fid_join(mlxsw_sp_port, vid_begin, vid_end); if (err) { - netdev_err(dev, "Failed to configure flooding\n"); - goto err_port_flood_set; + netdev_err(dev, "Failed to join FIDs\n"); + return err; } err = __mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end, @@ -629,10 +730,6 @@ static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port, return 0; -err_port_vid_to_fid_set: - mlxsw_sp_fid_destroy(mlxsw_sp, vid); - return err; - err_port_stp_state_set: for (vid = vid_begin; vid <= vid_end; vid++) clear_bit(vid, mlxsw_sp_port->active_vlans); @@ -642,13 +739,7 @@ err_port_pvid_set: __mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end, false, false); err_port_vlans_set: - __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end, false, - false); -err_port_flood_set: - last_visited_vid = vid_end; -err_port_fid_map: - for (vid = last_visited_vid; vid >= vid_begin; vid--) - mlxsw_sp_port_fid_unmap(mlxsw_sp_port, vid); + mlxsw_sp_port_fid_leave(mlxsw_sp_port, vid_begin, vid_end); return err; } @@ -971,21 +1062,7 @@ static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port, } } - err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end, - false, false); - if (err) { - netdev_err(dev, "Failed to clear flooding\n"); - return err; - } - - for (vid = vid_begin; vid <= vid_end; vid++) { - /* Remove FID mapping in case of Virtual mode */ - err = mlxsw_sp_port_fid_unmap(mlxsw_sp_port, vid); - if (err) { - netdev_err(dev, "Failed to unmap FID=%d", vid); - return err; - } - } + mlxsw_sp_port_fid_leave(mlxsw_sp_port, vid_begin, vid_end); out: /* Changing activity bits only if HW operation succeded */ @@ -1490,14 +1567,6 @@ static void mlxsw_sp_fdb_fini(struct mlxsw_sp *mlxsw_sp) cancel_delayed_work_sync(&mlxsw_sp->fdb_notify.dw); } -static void mlxsw_sp_fids_fini(struct mlxsw_sp *mlxsw_sp) -{ - u16 fid; - - for_each_set_bit(fid, mlxsw_sp->active_fids, VLAN_N_VID) - mlxsw_sp_fid_destroy(mlxsw_sp, fid); -} - int mlxsw_sp_switchdev_init(struct mlxsw_sp *mlxsw_sp) { return mlxsw_sp_fdb_init(mlxsw_sp); @@ -1506,7 +1575,6 @@ int mlxsw_sp_switchdev_init(struct mlxsw_sp *mlxsw_sp) void mlxsw_sp_switchdev_fini(struct mlxsw_sp *mlxsw_sp) { mlxsw_sp_fdb_fini(mlxsw_sp); - mlxsw_sp_fids_fini(mlxsw_sp); } int mlxsw_sp_port_vlan_init(struct mlxsw_sp_port *mlxsw_sp_port) -- cgit v0.10.2 From 6381b3a85feba85bba671a2f1299550e43a7abb1 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:16 +0200 Subject: mlxsw: spectrum: Check if port is vPort using its VID When L3 interfaces will be introduced a vPort won't necessarily have a FID assigned to it. This can happen if it's not member in a bridge (in which case it's assigned a vFID) or doesn't have an IP address (in which case it's assigned an rFID). Therefore, instead check the VID parameter to test whether a port is a vPort or not. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index d6cf6de..1d34419 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -259,12 +259,6 @@ mlxsw_sp_port_lagged_get(struct mlxsw_sp *mlxsw_sp, u16 lag_id, u8 port_index) return mlxsw_sp_port && mlxsw_sp_port->lagged ? mlxsw_sp_port : NULL; } -static inline bool -mlxsw_sp_port_is_vport(const struct mlxsw_sp_port *mlxsw_sp_port) -{ - return mlxsw_sp_port->vport.f; -} - static inline struct net_device * mlxsw_sp_vport_br_get(const struct mlxsw_sp_port *mlxsw_sp_vport) { @@ -277,6 +271,14 @@ mlxsw_sp_vport_vid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) return mlxsw_sp_vport->vport.vid; } +static inline bool +mlxsw_sp_port_is_vport(const struct mlxsw_sp_port *mlxsw_sp_port) +{ + u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_port); + + return vid != 0; +} + static inline u16 mlxsw_sp_vport_fid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) { -- cgit v0.10.2 From 41b996cc94c7e7f8567585d05ea1d52a3a181146 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:17 +0200 Subject: mlxsw: spectrum: Add FID get / set functions As previously explained, not all vPorts will be assigned FIDs, so instead of returning the FID index of a vPort, return a pointer to its FID struct. This will allow us to know whether it's legal to access the vPort's FID parameters such as index and device. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 08f3f51..ea007f0 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -776,7 +776,7 @@ static int mlxsw_sp_vport_vfid_join(struct mlxsw_sp_port *mlxsw_sp_vport) if (err) goto err_vport_fid_map; - mlxsw_sp_vport->vport.f = f; + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, f); f->ref_count++; return 0; @@ -792,9 +792,9 @@ err_vport_flood_set: static void mlxsw_sp_vport_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) { - struct mlxsw_sp_fid *f = mlxsw_sp_vport->vport.f; + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); - mlxsw_sp_vport->vport.f = NULL; + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, NULL); mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); @@ -2639,7 +2639,8 @@ static int mlxsw_sp_vport_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_vport, return mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_vport, fid); else - return mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_vport, fid); + return mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_vport, + fid); } static bool mlxsw_sp_port_dev_check(const struct net_device *dev) @@ -3229,7 +3230,7 @@ static int mlxsw_sp_vport_br_vfid_join(struct mlxsw_sp_port *mlxsw_sp_vport, if (err) goto err_vport_fid_map; - mlxsw_sp_vport->vport.f = f; + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, f); f->ref_count++; return 0; @@ -3244,13 +3245,13 @@ err_vport_flood_set: static void mlxsw_sp_vport_br_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) { - struct mlxsw_sp_fid *f = mlxsw_sp_vport->vport.f; + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); - mlxsw_sp_vport->vport.f = NULL; + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, NULL); if (--f->ref_count == 0) mlxsw_sp_br_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); } @@ -3293,8 +3294,8 @@ err_vport_br_vfid_join: static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, bool flush_fdb) { + u16 fid = mlxsw_sp_vport_fid_get(mlxsw_sp_vport)->fid; u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); - u16 fid = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 1d34419..a6dd295 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -259,12 +259,6 @@ mlxsw_sp_port_lagged_get(struct mlxsw_sp *mlxsw_sp, u16 lag_id, u8 port_index) return mlxsw_sp_port && mlxsw_sp_port->lagged ? mlxsw_sp_port : NULL; } -static inline struct net_device * -mlxsw_sp_vport_br_get(const struct mlxsw_sp_port *mlxsw_sp_vport) -{ - return mlxsw_sp_vport->vport.f->dev; -} - static inline u16 mlxsw_sp_vport_vid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) { @@ -279,10 +273,24 @@ mlxsw_sp_port_is_vport(const struct mlxsw_sp_port *mlxsw_sp_port) return vid != 0; } -static inline u16 +static inline void mlxsw_sp_vport_fid_set(struct mlxsw_sp_port *mlxsw_sp_vport, + struct mlxsw_sp_fid *f) +{ + mlxsw_sp_vport->vport.f = f; +} + +static inline struct mlxsw_sp_fid * mlxsw_sp_vport_fid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) { - return mlxsw_sp_vport->vport.f->fid; + return mlxsw_sp_vport->vport.f; +} + +static inline struct net_device * +mlxsw_sp_vport_br_get(const struct mlxsw_sp_port *mlxsw_sp_vport) +{ + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + + return f->dev; } static inline struct mlxsw_sp_port * @@ -307,7 +315,9 @@ mlxsw_sp_port_vport_find_by_fid(const struct mlxsw_sp_port *mlxsw_sp_port, list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list, vport.list) { - if (mlxsw_sp_vport_fid_get(mlxsw_sp_vport) == fid) + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + + if (f->fid == fid) return mlxsw_sp_vport; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 6713769..b31b2ce 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -58,7 +58,7 @@ static u16 mlxsw_sp_port_vid_to_fid_get(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid = vid; if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) - fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port); + fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port)->fid; if (!fid) fid = mlxsw_sp_port->pvid; @@ -233,9 +233,9 @@ static int mlxsw_sp_port_uc_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, int err; if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) { - u16 vfid, fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port); + u16 fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port)->fid; + u16 vfid = mlxsw_sp_fid_to_vfid(fid); - vfid = mlxsw_sp_fid_to_vfid(fid); return __mlxsw_sp_port_flood_set(mlxsw_sp_port, vfid, vfid, set, true); } @@ -1212,7 +1212,7 @@ static int mlxsw_sp_port_fdb_dump(struct mlxsw_sp_port *mlxsw_sp_port, return -ENOMEM; if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) - vport_fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port); + vport_fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port)->fid; mlxsw_reg_sfd_pack(sfd_pl, MLXSW_REG_SFD_OP_QUERY_DUMP, 0); do { -- cgit v0.10.2 From 56918b6b0a66e449b3e11adb3e1c93849964d429 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:18 +0200 Subject: mlxsw: spectrum: Don't count on FID being present Not all vPorts will have FIDs assigned to them, so make sure functions first test for FID presence. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index ea007f0..0c42515 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -3323,7 +3323,9 @@ mlxsw_sp_port_master_bridge_check(const struct mlxsw_sp_port *mlxsw_sp_port, list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list, vport.list) { - if (mlxsw_sp_vport_br_get(mlxsw_sp_vport) == br_dev) + struct net_device *dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport); + + if (dev && dev == br_dev) return false; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index a6dd295..f419535 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -290,7 +290,7 @@ mlxsw_sp_vport_br_get(const struct mlxsw_sp_port *mlxsw_sp_vport) { struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); - return f->dev; + return f ? f->dev : NULL; } static inline struct mlxsw_sp_port * @@ -317,7 +317,7 @@ mlxsw_sp_port_vport_find_by_fid(const struct mlxsw_sp_port *mlxsw_sp_port, vport.list) { struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); - if (f->fid == fid) + if (f && f->fid == fid) return mlxsw_sp_vport; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index b31b2ce..fc34c46 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -55,10 +55,10 @@ static u16 mlxsw_sp_port_vid_to_fid_get(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) { + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_port); u16 fid = vid; - if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) - fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port)->fid; + fid = f ? f->fid : fid; if (!fid) fid = mlxsw_sp_port->pvid; @@ -1196,7 +1196,8 @@ static int mlxsw_sp_port_fdb_dump(struct mlxsw_sp_port *mlxsw_sp_port, { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; struct mlxsw_sp_port *tmp; - u16 vport_fid = 0; + struct mlxsw_sp_fid *f; + u16 vport_fid; char *sfd_pl; char mac[ETH_ALEN]; u16 fid; @@ -1211,8 +1212,8 @@ static int mlxsw_sp_port_fdb_dump(struct mlxsw_sp_port *mlxsw_sp_port, if (!sfd_pl) return -ENOMEM; - if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) - vport_fid = mlxsw_sp_vport_fid_get(mlxsw_sp_port)->fid; + f = mlxsw_sp_vport_fid_get(mlxsw_sp_port); + vport_fid = f ? f->fid : 0; mlxsw_reg_sfd_pack(sfd_pl, MLXSW_REG_SFD_OP_QUERY_DUMP, 0); do { -- cgit v0.10.2 From fe3f6d144a843c9845d2eccb78fe72d1bebae0e9 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:19 +0200 Subject: mlxsw: spectrum: Refactor FDB flushing logic FDB entries are learned using {Port / LAG ID, FID} and therefore should be flushed whenever a port (vPort) leaves its FID (vFID). However, when the bridge port is a LAG device (or a VLAN device on top), then FDB flushing is conditional. Ports removed from such LAG configurations must not trigger flushing, as other ports might still be members in the LAG and therefore the bridge port is still active. The decision whether to flush or not was previously computed in the netdevice notification block, but in order to flush the entries when a port leaves its FID this decision should be computed there. Strip the notification block from this logic and instead move it to one FDB flushing function that is invoked from both the FID / vFID leave functions. When port isn't member in LAG, FDB flushing should always occur. Otherwise, it should occur only when the last port (vPort) member in the LAG leaves the FID (vFID). This will allow us - in the next patch - to simplify the cleanup code paths that are hit whenever the topology above the port netdevs changes. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 0c42515..2b1748d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2535,16 +2535,37 @@ static struct mlxsw_driver mlxsw_sp_driver = { .profile = &mlxsw_sp_config_profile, }; -static int -mlxsw_sp_port_fdb_flush_by_port(const struct mlxsw_sp_port *mlxsw_sp_port) +static bool mlxsw_sp_lag_port_fid_member(struct mlxsw_sp_port *lag_port, + u16 fid) +{ + if (mlxsw_sp_fid_is_vfid(fid)) + return mlxsw_sp_port_vport_find_by_fid(lag_port, fid); + else + return test_bit(fid, lag_port->active_vlans); +} + +static bool mlxsw_sp_port_fdb_should_flush(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - char sfdf_pl[MLXSW_REG_SFDF_LEN]; + u8 local_port = mlxsw_sp_port->local_port; + u16 lag_id = mlxsw_sp_port->lag_id; + int i, count = 0; - mlxsw_reg_sfdf_pack(sfdf_pl, MLXSW_REG_SFDF_FLUSH_PER_PORT); - mlxsw_reg_sfdf_system_port_set(sfdf_pl, mlxsw_sp_port->local_port); + if (!mlxsw_sp_port->lagged) + return true; - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl); + for (i = 0; i < MLXSW_SP_PORT_PER_LAG_MAX; i++) { + struct mlxsw_sp_port *lag_port; + + lag_port = mlxsw_sp_port_lagged_get(mlxsw_sp, lag_id, i); + if (!lag_port || lag_port->local_port == local_port) + continue; + if (mlxsw_sp_lag_port_fid_member(lag_port, fid)) + count++; + } + + return !count; } static int @@ -2563,18 +2584,6 @@ mlxsw_sp_port_fdb_flush_by_port_fid(const struct mlxsw_sp_port *mlxsw_sp_port, } static int -mlxsw_sp_port_fdb_flush_by_lag_id(const struct mlxsw_sp_port *mlxsw_sp_port) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - char sfdf_pl[MLXSW_REG_SFDF_LEN]; - - mlxsw_reg_sfdf_pack(sfdf_pl, MLXSW_REG_SFDF_FLUSH_PER_LAG); - mlxsw_reg_sfdf_lag_id_set(sfdf_pl, mlxsw_sp_port->lag_id); - - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl); -} - -static int mlxsw_sp_port_fdb_flush_by_lag_id_fid(const struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) { @@ -2588,59 +2597,16 @@ mlxsw_sp_port_fdb_flush_by_lag_id_fid(const struct mlxsw_sp_port *mlxsw_sp_port, return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl); } -static int -__mlxsw_sp_port_fdb_flush(const struct mlxsw_sp_port *mlxsw_sp_port) -{ - int err, last_err = 0; - u16 vid; - - for (vid = 1; vid < VLAN_N_VID - 1; vid++) { - err = mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_port, vid); - if (err) - last_err = err; - } - - return last_err; -} - -static int -__mlxsw_sp_port_fdb_flush_lagged(const struct mlxsw_sp_port *mlxsw_sp_port) +int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) { - int err, last_err = 0; - u16 vid; - - for (vid = 1; vid < VLAN_N_VID - 1; vid++) { - err = mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_port, vid); - if (err) - last_err = err; - } - - return last_err; -} - -static int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port) -{ - if (!list_empty(&mlxsw_sp_port->vports_list)) - if (mlxsw_sp_port->lagged) - return __mlxsw_sp_port_fdb_flush_lagged(mlxsw_sp_port); - else - return __mlxsw_sp_port_fdb_flush(mlxsw_sp_port); - else - if (mlxsw_sp_port->lagged) - return mlxsw_sp_port_fdb_flush_by_lag_id(mlxsw_sp_port); - else - return mlxsw_sp_port_fdb_flush_by_port(mlxsw_sp_port); -} + if (!mlxsw_sp_port_fdb_should_flush(mlxsw_sp_port, fid)) + return 0; -static int mlxsw_sp_vport_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_vport, - u16 fid) -{ - if (mlxsw_sp_vport->lagged) - return mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_vport, + if (mlxsw_sp_port->lagged) + return mlxsw_sp_port_fdb_flush_by_lag_id_fid(mlxsw_sp_port, fid); else - return mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_vport, - fid); + return mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_port, fid); } static bool mlxsw_sp_port_dev_check(const struct net_device *dev) @@ -2693,14 +2659,10 @@ static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port, return 0; } -static void mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port, - bool flush_fdb) +static void mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port) { struct net_device *dev = mlxsw_sp_port->dev; - if (flush_fdb && mlxsw_sp_port_fdb_flush(mlxsw_sp_port)) - netdev_err(mlxsw_sp_port->dev, "Failed to flush FDB\n"); - mlxsw_sp_port_pvid_set(mlxsw_sp_port, 1); mlxsw_sp_master_bridge_dec(mlxsw_sp_port->mlxsw_sp); @@ -2874,8 +2836,7 @@ err_col_port_add: return err; } -static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, - bool flush_fdb); +static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport); static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, struct net_device *lag_dev) @@ -2905,19 +2866,16 @@ static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, continue; br_dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport); - mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, false); + mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport); } if (mlxsw_sp_port->bridged) { mlxsw_sp_port_active_vlans_del(mlxsw_sp_port); - mlxsw_sp_port_bridge_leave(mlxsw_sp_port, false); + mlxsw_sp_port_bridge_leave(mlxsw_sp_port); } - if (lag->ref_count == 1) { - if (mlxsw_sp_port_fdb_flush_by_lag_id(mlxsw_sp_port)) - netdev_err(mlxsw_sp_port->dev, "Failed to flush FDB\n"); + if (lag->ref_count == 1) mlxsw_sp_lag_destroy(mlxsw_sp, lag_id); - } mlxsw_core_lag_mapping_clear(mlxsw_sp->core, lag_id, mlxsw_sp_port->local_port); @@ -2997,7 +2955,7 @@ static void mlxsw_sp_port_vlan_unlink(struct mlxsw_sp_port *mlxsw_sp_port, struct net_device *br_dev; br_dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport); - mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, true); + mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport); } mlxsw_sp_vport->dev = mlxsw_sp_port->dev; @@ -3053,7 +3011,7 @@ static int mlxsw_sp_netdevice_port_upper_event(struct net_device *dev, err = mlxsw_sp_port_bridge_join(mlxsw_sp_port, upper_dev); else - mlxsw_sp_port_bridge_leave(mlxsw_sp_port, true); + mlxsw_sp_port_bridge_leave(mlxsw_sp_port); } else if (netif_is_lag_master(upper_dev)) { if (info->linking) err = mlxsw_sp_port_lag_join(mlxsw_sp_port, @@ -3251,6 +3209,8 @@ static void mlxsw_sp_vport_br_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); + mlxsw_sp_port_fdb_flush(mlxsw_sp_vport, f->fid); + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, NULL); if (--f->ref_count == 0) mlxsw_sp_br_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); @@ -3291,10 +3251,8 @@ err_vport_br_vfid_join: return err; } -static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, - bool flush_fdb) +static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport) { - u16 fid = mlxsw_sp_vport_fid_get(mlxsw_sp_vport)->fid; u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); @@ -3306,9 +3264,6 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport, mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, MLXSW_REG_SPMS_STATE_FORWARDING); - if (flush_fdb) - mlxsw_sp_vport_fdb_flush(mlxsw_sp_vport, fid); - mlxsw_sp_vport->learning = 0; mlxsw_sp_vport->learning_sync = 0; mlxsw_sp_vport->uc_flood = 0; @@ -3372,7 +3327,7 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, */ if (!mlxsw_sp_vport) return 0; - mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport, true); + mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport); } } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index f419535..9122abb 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -382,6 +382,7 @@ int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, bool set); void mlxsw_sp_port_active_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port); int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid); +int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid); int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index, bool dwrr, u8 dwrr_weight); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index fc34c46..7c2b0f8 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -500,6 +500,8 @@ static void __mlxsw_sp_port_fid_leave(struct mlxsw_sp_port *mlxsw_sp_port, if (WARN_ON(!f)) return; + mlxsw_sp_port_fdb_flush(mlxsw_sp_port, fid); + if (--f->ref_count == 0) mlxsw_sp_fid_destroy(mlxsw_sp_port->mlxsw_sp, f); } -- cgit v0.10.2 From 1c800759078d400a02ed4ed1655672ea5fdcfd66 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:20 +0200 Subject: mlxsw: spectrum: Free resources upon vPort destruction There are situations in which a vPort is destroyed while still holding references to device's resources such as FIDs and FDB records. This can happen, for example, when a VLAN device is deleted while still being bridged. Instead of trying to make sure vPort destruction is invoked when it no longer uses device's resources, just free them upon destruction. This simplifies the code, as we no longer need to take different situations into account when events are received - cleanup is taken care of in one place. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 2b1748d..9f48ec5 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -660,6 +660,8 @@ static int mlxsw_sp_vfid_op(struct mlxsw_sp *mlxsw_sp, u16 fid, bool create) return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); } +static void mlxsw_sp_vport_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport); + static struct mlxsw_sp_fid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, u16 vid) { @@ -685,6 +687,7 @@ static struct mlxsw_sp_fid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, if (!f) goto err_allocate_vfid; + f->leave = mlxsw_sp_vport_vfid_leave; f->fid = fid; f->vid = vid; @@ -887,6 +890,7 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); struct mlxsw_sp_port *mlxsw_sp_vport; + struct mlxsw_sp_fid *f; int err; /* VLAN 0 is removed from HW filter when device goes down, but @@ -921,7 +925,12 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev, return err; } - mlxsw_sp_vport_vfid_leave(mlxsw_sp_vport); + /* Drop FID reference. If this was the last reference the + * resources will be freed. + */ + f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + if (f && !WARN_ON(!f->leave)) + f->leave(mlxsw_sp_vport); /* When removing the last VLAN interface on a bridged port we need to * transition all active 802.1Q bridge VLANs to use VID to FID @@ -2836,15 +2845,12 @@ err_col_port_add: return err; } -static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport); - static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, struct net_device *lag_dev) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - struct mlxsw_sp_port *mlxsw_sp_vport; - struct mlxsw_sp_upper *lag; u16 lag_id = mlxsw_sp_port->lag_id; + struct mlxsw_sp_upper *lag; if (!mlxsw_sp_port->lagged) return; @@ -2854,21 +2860,6 @@ static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_lag_col_port_disable(mlxsw_sp_port, lag_id); mlxsw_sp_lag_col_port_remove(mlxsw_sp_port, lag_id); - /* In case we leave a LAG device that has bridges built on top, - * then their teardown sequence is never issued and we need to - * invoke the necessary cleanup routines ourselves. - */ - list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list, - vport.list) { - struct net_device *br_dev; - - if (!mlxsw_sp_vport->bridged) - continue; - - br_dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport); - mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport); - } - if (mlxsw_sp_port->bridged) { mlxsw_sp_port_active_vlans_del(mlxsw_sp_port); mlxsw_sp_port_bridge_leave(mlxsw_sp_port); @@ -2947,17 +2938,6 @@ static void mlxsw_sp_port_vlan_unlink(struct mlxsw_sp_port *mlxsw_sp_port, if (WARN_ON(!mlxsw_sp_vport)) return; - /* When removing a VLAN device while still bridged we should first - * remove it from the bridge, as we receive the bridge's notification - * when the vPort is already gone. - */ - if (mlxsw_sp_vport->bridged) { - struct net_device *br_dev; - - br_dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport); - mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport); - } - mlxsw_sp_vport->dev = mlxsw_sp_port->dev; } @@ -3115,6 +3095,8 @@ static u16 mlxsw_sp_avail_br_vfid_get(const struct mlxsw_sp *mlxsw_sp) MLXSW_SP_VFID_BR_MAX); } +static void mlxsw_sp_vport_br_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport); + static struct mlxsw_sp_fid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, struct net_device *br_dev) { @@ -3140,6 +3122,7 @@ static struct mlxsw_sp_fid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, if (!f) goto err_allocate_vfid; + f->leave = mlxsw_sp_vport_br_vfid_leave; f->fid = fid; f->dev = br_dev; @@ -3321,10 +3304,6 @@ static int mlxsw_sp_netdevice_vport_event(struct net_device *dev, err = mlxsw_sp_vport_bridge_join(mlxsw_sp_vport, upper_dev); } else { - /* We ignore bridge's unlinking notifications if vPort - * is gone, since we already left the bridge when the - * VLAN device was unlinked from the real device. - */ if (!mlxsw_sp_vport) return 0; mlxsw_sp_vport_bridge_leave(mlxsw_sp_vport); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 9122abb..36c9835 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -88,6 +88,7 @@ struct mlxsw_sp_upper { }; struct mlxsw_sp_fid { + void (*leave)(struct mlxsw_sp_port *mlxsw_sp_vport); struct list_head list; unsigned int ref_count; struct net_device *dev; -- cgit v0.10.2 From 223053783b2374fa45f22600a127c9dcb3bf018c Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 20 Jun 2016 23:04:21 +0200 Subject: mlxsw: spectrum: Add debug prints For debug purposes, it's useful to know the order in which the driver responds to changes in the topology of its upper devices. Add debug prints to signal these events. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 9f48ec5..d23948b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2589,6 +2589,9 @@ mlxsw_sp_port_fdb_flush_by_port_fid(const struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_reg_sfdf_port_fid_system_port_set(sfdf_pl, mlxsw_sp_port->local_port); + netdev_dbg(mlxsw_sp_port->dev, "FDB flushed using Port=%d, FID=%d\n", + mlxsw_sp_port->local_port, fid); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl); } @@ -2603,6 +2606,9 @@ mlxsw_sp_port_fdb_flush_by_lag_id_fid(const struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_reg_sfdf_fid_set(sfdf_pl, fid); mlxsw_reg_sfdf_lag_fid_lag_id_set(sfdf_pl, mlxsw_sp_port->lag_id); + netdev_dbg(mlxsw_sp_port->dev, "FDB flushed using LAG ID=%d, FID=%d\n", + mlxsw_sp_port->lag_id, fid); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdf), sfdf_pl); } @@ -3174,6 +3180,8 @@ static int mlxsw_sp_vport_br_vfid_join(struct mlxsw_sp_port *mlxsw_sp_vport, mlxsw_sp_vport_fid_set(mlxsw_sp_vport, f); f->ref_count++; + netdev_dbg(mlxsw_sp_vport->dev, "Joined FID=%d\n", f->fid); + return 0; err_vport_fid_map: @@ -3188,6 +3196,8 @@ static void mlxsw_sp_vport_br_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) { struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + netdev_dbg(mlxsw_sp_vport->dev, "Left FID=%d\n", f->fid); + mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 7c2b0f8..a0c7376 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -488,6 +488,8 @@ static int __mlxsw_sp_port_fid_join(struct mlxsw_sp_port *mlxsw_sp_port, f->ref_count++; + netdev_dbg(mlxsw_sp_port->dev, "Joined FID=%d\n", fid); + return 0; } @@ -500,6 +502,8 @@ static void __mlxsw_sp_port_fid_leave(struct mlxsw_sp_port *mlxsw_sp_port, if (WARN_ON(!f)) return; + netdev_dbg(mlxsw_sp_port->dev, "Left FID=%d\n", fid); + mlxsw_sp_port_fdb_flush(mlxsw_sp_port, fid); if (--f->ref_count == 0) -- cgit v0.10.2 From 0e4699e4a37be4cfa07f0340cef6a3fa6a46f5f8 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Sat, 18 Jun 2016 11:44:03 +0300 Subject: rxrpc: checking for IS_ERR() instead of NULL rxrpc_lookup_peer_rcu() and rxrpc_lookup_peer() return NULL on error, never error pointers, so IS_ERR() can't be used. Fix three callers of those functions. Fixes: be6e6707f6ee ('rxrpc: Rework peer object handling to use hash table and RCU') Signed-off-by: Dan Carpenter Signed-off-by: David Howells diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index c83c3c7..9dd160b 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -247,8 +247,8 @@ struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *rx, /* find a remote transport endpoint from the local one */ peer = rxrpc_lookup_peer(rx->local, srx, gfp); - if (IS_ERR(peer)) - return ERR_CAST(peer); + if (!peer) + return ERR_PTR(-ENOMEM); /* find a transport */ trans = rxrpc_get_transport(rx->local, peer, gfp); diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index 50136c7..553b67c1 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -96,7 +96,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, notification->mark = RXRPC_SKB_MARK_NEW_CALL; peer = rxrpc_lookup_peer(local, srx, GFP_NOIO); - if (IS_ERR(peer)) { + if (!peer) { _debug("no peer"); ret = -EBUSY; goto error; diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 47fb167..e11e4d7 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -639,7 +639,7 @@ static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local, rxrpc_get_addr_from_skb(local, skb, &srx); rcu_read_lock(); peer = rxrpc_lookup_peer_rcu(local, &srx); - if (IS_ERR(peer)) + if (!peer) goto cant_find_peer; trans = rxrpc_find_transport(local, peer); -- cgit v0.10.2 From 2f9f9f5210887b1019fbd0327ffdf7c3aff271fd Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 17 Jun 2016 11:55:22 +0200 Subject: rxrpc: fix uninitialized variable use Hashing the peer key was introduced for AF_INET, but gcc warns about the rxrpc_peer_hash_key function returning uninitialized data for any other value of srx->transport.family: net/rxrpc/peer_object.c: In function 'rxrpc_peer_hash_key': net/rxrpc/peer_object.c:57:15: error: 'p' may be used uninitialized in this function [-Werror=maybe-uninitialized] Assuming that nothing else can be set here, this changes the function to just return zero in case of an unknown address family. Fixes: be6e6707f6ee ("rxrpc: Rework peer object handling to use hash table and RCU") Signed-off-by: Arnd Bergmann Signed-off-by: David Howells diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c index faf222c..6baad70 100644 --- a/net/rxrpc/peer_object.c +++ b/net/rxrpc/peer_object.c @@ -50,6 +50,9 @@ static unsigned long rxrpc_peer_hash_key(struct rxrpc_local *local, size = sizeof(srx->transport.sin.sin_addr); p = (u16 *)&srx->transport.sin.sin_addr; break; + default: + WARN(1, "AF_RXRPC: Unsupported transport address family\n"); + return 0; } /* Step through the peer address in 16-bit portions for speed */ -- cgit v0.10.2 From 19ffa01c9c45861ad6b181323e0d36904298e326 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:36 +0100 Subject: rxrpc: Use structs to hold connection params and protocol info Define and use a structure to hold connection parameters. This makes it easier to pass multiple connection parameters around. Define and use a structure to hold protocol information used to hash a connection for lookup on incoming packet. Most of these fields will be disposed of eventually, including the duplicate local pointer. Whilst we're at it rename "proto" to "family" when referring to a protocol family. Signed-off-by: David Howells diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 9dd160b..48b45a0 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -97,7 +97,7 @@ static int rxrpc_validate_address(struct rxrpc_sock *rx, srx->transport_len > len) return -EINVAL; - if (srx->transport.family != rx->proto) + if (srx->transport.family != rx->family) return -EAFNOSUPPORT; switch (srx->transport.family) { @@ -227,32 +227,30 @@ static int rxrpc_listen(struct socket *sock, int backlog) /* * find a transport by address */ -struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *rx, - struct sockaddr *addr, - int addr_len, int flags, - gfp_t gfp) +struct rxrpc_transport * +rxrpc_name_to_transport(struct rxrpc_conn_parameters *cp, + struct sockaddr *addr, + int addr_len, + gfp_t gfp) { struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *) addr; struct rxrpc_transport *trans; - struct rxrpc_peer *peer; - _enter("%p,%p,%d,%d", rx, addr, addr_len, flags); - - ASSERT(rx->local != NULL); + _enter("%p,%d", addr, addr_len); - if (rx->srx.transport_type != srx->transport_type) + if (cp->local->srx.transport_type != srx->transport_type) return ERR_PTR(-ESOCKTNOSUPPORT); - if (rx->srx.transport.family != srx->transport.family) + if (cp->local->srx.transport.family != srx->transport.family) return ERR_PTR(-EAFNOSUPPORT); /* find a remote transport endpoint from the local one */ - peer = rxrpc_lookup_peer(rx->local, srx, gfp); - if (!peer) + cp->peer = rxrpc_lookup_peer(cp->local, srx, gfp); + if (!cp->peer) return ERR_PTR(-ENOMEM); /* find a transport */ - trans = rxrpc_get_transport(rx->local, peer, gfp); - rxrpc_put_peer(peer); + trans = rxrpc_get_transport(cp->local, cp->peer, gfp); + rxrpc_put_peer(cp->peer); _leave(" = %p", trans); return trans; } @@ -277,6 +275,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, unsigned long user_call_ID, gfp_t gfp) { + struct rxrpc_conn_parameters cp; struct rxrpc_conn_bundle *bundle; struct rxrpc_transport *trans; struct rxrpc_call *call; @@ -286,18 +285,26 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, lock_sock(&rx->sk); - trans = rxrpc_name_to_transport(rx, (struct sockaddr *)srx, - sizeof(*srx), 0, gfp); + if (!key) + key = rx->key; + if (key && !key->payload.data[0]) + key = NULL; /* a no-security key */ + + memset(&cp, 0, sizeof(cp)); + cp.local = rx->local; + cp.key = key; + cp.security_level = 0; + cp.exclusive = false; + cp.service_id = srx->srx_service; + + trans = rxrpc_name_to_transport(&cp, (struct sockaddr *)srx, + sizeof(*srx), gfp); if (IS_ERR(trans)) { call = ERR_CAST(trans); trans = NULL; goto out_notrans; } - - if (!key) - key = rx->key; - if (key && !key->payload.data[0]) - key = NULL; /* a no-security key */ + cp.peer = trans->peer; bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, gfp); if (IS_ERR(bundle)) { @@ -305,7 +312,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, goto out; } - call = rxrpc_new_client_call(rx, trans, bundle, user_call_ID, gfp); + call = rxrpc_new_client_call(rx, &cp, trans, bundle, user_call_ID, gfp); rxrpc_put_bundle(trans, bundle); out: rxrpc_put_transport(trans); @@ -600,7 +607,7 @@ static int rxrpc_create(struct net *net, struct socket *sock, int protocol, sk->sk_destruct = rxrpc_sock_destructor; rx = rxrpc_sk(sk); - rx->proto = protocol; + rx->family = protocol; rx->calls = RB_ROOT; INIT_LIST_HEAD(&rx->listen_link); diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index c168268..efe6673 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -72,7 +72,7 @@ struct rxrpc_sock { #define RXRPC_SECURITY_MAX RXRPC_SECURITY_ENCRYPT struct sockaddr_rxrpc srx; /* local address */ struct sockaddr_rxrpc connect_srx; /* Default client address from connect() */ - sa_family_t proto; /* protocol created with */ + sa_family_t family; /* protocol family created with */ }; #define rxrpc_sk(__sk) container_of((__sk), struct rxrpc_sock, sk) @@ -262,6 +262,34 @@ struct rxrpc_conn_bundle { }; /* + * Keys for matching a connection. + */ +struct rxrpc_conn_proto { + unsigned long hash_key; + struct rxrpc_local *local; /* Representation of local endpoint */ + u32 epoch; /* epoch of this connection */ + u32 cid; /* connection ID */ + u8 in_clientflag; /* RXRPC_CLIENT_INITIATED if we are server */ + u8 addr_size; /* Size of the address */ + sa_family_t family; /* Transport protocol */ + __be16 port; /* Peer UDP/UDP6 port */ + union { /* Peer address */ + struct in_addr ipv4_addr; + struct in6_addr ipv6_addr; + u32 raw_addr[0]; + }; +}; + +struct rxrpc_conn_parameters { + struct rxrpc_local *local; /* Representation of local endpoint */ + struct rxrpc_peer *peer; /* Remote endpoint */ + struct key *key; /* Security details */ + bool exclusive; /* T if conn is exclusive */ + u16 service_id; /* Service ID for this connection */ + u32 security_level; /* Security level selected */ +}; + +/* * RxRPC connection definition * - matched by { transport, service_id, conn_id, direction, key } * - each connection can only handle four simultaneous calls @@ -269,6 +297,9 @@ struct rxrpc_conn_bundle { struct rxrpc_connection { struct rxrpc_transport *trans; /* transport session */ struct rxrpc_conn_bundle *bundle; /* connection bundle (client) */ + struct rxrpc_conn_proto proto; + struct rxrpc_conn_parameters params; + struct work_struct processor; /* connection event processor */ struct rb_node node; /* node in transport's lookup tree */ struct list_head link; /* link in master connection list */ @@ -277,7 +308,6 @@ struct rxrpc_connection { struct sk_buff_head rx_queue; /* received conn-level packets */ struct rxrpc_call *channels[RXRPC_MAXCALLS]; /* channels (active calls) */ const struct rxrpc_security *security; /* applied security module */ - struct key *key; /* security for this connection (client) */ struct key *server_key; /* security for this service */ struct crypto_skcipher *cipher; /* encryption handle */ struct rxrpc_crypt csum_iv; /* packet checksum base */ @@ -308,13 +338,8 @@ struct rxrpc_connection { u8 size_align; /* data size alignment (for security) */ u8 header_size; /* rxrpc + security header size */ u8 security_size; /* security header size */ - u32 security_level; /* security level negotiated */ u32 security_nonce; /* response re-use preventer */ - u32 epoch; /* epoch of this connection */ - u32 cid; /* connection ID */ - u16 service_id; /* service ID for this connection */ u8 security_ix; /* security type */ - u8 in_clientflag; /* RXRPC_CLIENT_INITIATED if we are server */ u8 out_clientflag; /* RXRPC_CLIENT_INITIATED if we are client */ }; @@ -448,7 +473,7 @@ struct rxrpc_call { unsigned long hash_key; /* Full hash key */ u8 in_clientflag; /* Copy of conn->in_clientflag for hashing */ struct rxrpc_local *local; /* Local endpoint. Used for hashing. */ - sa_family_t proto; /* Frame protocol */ + sa_family_t family; /* Frame protocol */ u32 call_id; /* call ID on connection */ u32 cid; /* connection ID plus channel index */ u32 epoch; /* epoch of this connection */ @@ -481,9 +506,9 @@ extern u32 rxrpc_epoch; extern atomic_t rxrpc_debug_id; extern struct workqueue_struct *rxrpc_workqueue; -extern struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *, +extern struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_conn_parameters *, struct sockaddr *, - int, int, gfp_t); + int, gfp_t); /* * call_accept.c @@ -512,6 +537,7 @@ struct rxrpc_call *rxrpc_find_call_hash(struct rxrpc_host_header *, void *, sa_family_t, const void *); struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *, unsigned long); struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *, + struct rxrpc_conn_parameters *, struct rxrpc_transport *, struct rxrpc_conn_bundle *, unsigned long, gfp_t); @@ -541,8 +567,9 @@ struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *, struct rxrpc_transport *, struct key *, u16, gfp_t); void rxrpc_put_bundle(struct rxrpc_transport *, struct rxrpc_conn_bundle *); -int rxrpc_connect_call(struct rxrpc_sock *, struct rxrpc_transport *, - struct rxrpc_conn_bundle *, struct rxrpc_call *, gfp_t); +int rxrpc_connect_call(struct rxrpc_sock *, struct rxrpc_conn_parameters *, + struct rxrpc_transport *, struct rxrpc_conn_bundle *, + struct rxrpc_call *, gfp_t); void rxrpc_put_connection(struct rxrpc_connection *); void __exit rxrpc_destroy_all_connections(void); struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *, @@ -550,6 +577,16 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *, extern struct rxrpc_connection * rxrpc_incoming_connection(struct rxrpc_transport *, struct rxrpc_host_header *); +static inline bool rxrpc_conn_is_client(const struct rxrpc_connection *conn) +{ + return conn->out_clientflag; +} + +static inline bool rxrpc_conn_is_service(const struct rxrpc_connection *conn) +{ + return conn->proto.in_clientflag; +} + /* * input.c */ diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c index e610b10..1571dfb 100644 --- a/net/rxrpc/call_event.c +++ b/net/rxrpc/call_event.c @@ -842,7 +842,7 @@ void rxrpc_process_call(struct work_struct *work) msg.msg_controllen = 0; msg.msg_flags = 0; - whdr.epoch = htonl(call->conn->epoch); + whdr.epoch = htonl(call->conn->proto.epoch); whdr.cid = htonl(call->cid); whdr.callNumber = htonl(call->call_id); whdr.seq = 0; @@ -1264,7 +1264,7 @@ maybe_reschedule: if (call->state >= RXRPC_CALL_COMPLETE && !list_empty(&call->accept_link)) { _debug("X unlinking once-pending call %p { e=%lx f=%lx c=%x }", - call, call->events, call->flags, call->conn->cid); + call, call->events, call->flags, call->conn->proto.cid); read_lock_bh(&call->state_lock); if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && @@ -1282,7 +1282,7 @@ error: * this means there's a race between clearing the flag and setting the * work pending bit and the work item being processed again */ if (call->events && !work_pending(&call->processor)) { - _debug("jumpstart %x", call->conn->cid); + _debug("jumpstart %x", call->conn->proto.cid); rxrpc_queue_call(call); } diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 8b4d47b..b7c6011 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -71,7 +71,7 @@ static unsigned long rxrpc_call_hashfunc( u32 call_id, u32 epoch, u16 service_id, - sa_family_t proto, + sa_family_t family, void *localptr, unsigned int addr_size, const u8 *peer_addr) @@ -92,7 +92,7 @@ static unsigned long rxrpc_call_hashfunc( key += (cid & RXRPC_CIDMASK) >> RXRPC_CIDSHIFT; key += cid & RXRPC_CHANNELMASK; key += in_clientflag; - key += proto; + key += family; /* Step through the peer address in 16-bit portions for speed */ for (i = 0, p = (const u16 *)peer_addr; i < addr_size >> 1; i++, p++) key += *p; @@ -109,7 +109,7 @@ static void rxrpc_call_hash_add(struct rxrpc_call *call) unsigned int addr_size = 0; _enter(""); - switch (call->proto) { + switch (call->family) { case AF_INET: addr_size = sizeof(call->peer_ip.ipv4_addr); break; @@ -121,7 +121,7 @@ static void rxrpc_call_hash_add(struct rxrpc_call *call) } key = rxrpc_call_hashfunc(call->in_clientflag, call->cid, call->call_id, call->epoch, - call->service_id, call->proto, + call->service_id, call->family, call->conn->trans->local, addr_size, call->peer_ip.ipv6_addr); /* Store the full key in the call */ @@ -151,7 +151,7 @@ static void rxrpc_call_hash_del(struct rxrpc_call *call) struct rxrpc_call *rxrpc_find_call_hash( struct rxrpc_host_header *hdr, void *localptr, - sa_family_t proto, + sa_family_t family, const void *peer_addr) { unsigned long key; @@ -161,7 +161,7 @@ struct rxrpc_call *rxrpc_find_call_hash( u8 in_clientflag = hdr->flags & RXRPC_CLIENT_INITIATED; _enter(""); - switch (proto) { + switch (family) { case AF_INET: addr_size = sizeof(call->peer_ip.ipv4_addr); break; @@ -174,7 +174,7 @@ struct rxrpc_call *rxrpc_find_call_hash( key = rxrpc_call_hashfunc(in_clientflag, hdr->cid, hdr->callNumber, hdr->epoch, hdr->serviceId, - proto, localptr, addr_size, + family, localptr, addr_size, peer_addr); hash_for_each_possible_rcu(rxrpc_call_hash, call, hash_node, key) { if (call->hash_key == key && @@ -182,7 +182,7 @@ struct rxrpc_call *rxrpc_find_call_hash( call->cid == hdr->cid && call->in_clientflag == in_clientflag && call->service_id == hdr->serviceId && - call->proto == proto && + call->family == family && call->local == localptr && memcmp(call->peer_ip.ipv6_addr, peer_addr, addr_size) == 0 && @@ -286,6 +286,7 @@ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp) */ static struct rxrpc_call *rxrpc_alloc_client_call( struct rxrpc_sock *rx, + struct rxrpc_conn_parameters *cp, struct rxrpc_transport *trans, struct rxrpc_conn_bundle *bundle, gfp_t gfp) @@ -307,16 +308,16 @@ static struct rxrpc_call *rxrpc_alloc_client_call( call->socket = rx; call->rx_data_post = 1; - ret = rxrpc_connect_call(rx, trans, bundle, call, gfp); + ret = rxrpc_connect_call(rx, cp, trans, bundle, call, gfp); if (ret < 0) { kmem_cache_free(rxrpc_call_jar, call); return ERR_PTR(ret); } /* Record copies of information for hashtable lookup */ - call->proto = rx->proto; - call->local = trans->local; - switch (call->proto) { + call->family = rx->family; + call->local = call->conn->params.local; + switch (call->family) { case AF_INET: call->peer_ip.ipv4_addr = trans->peer->srx.transport.sin.sin_addr.s_addr; @@ -327,9 +328,9 @@ static struct rxrpc_call *rxrpc_alloc_client_call( sizeof(call->peer_ip.ipv6_addr)); break; } - call->epoch = call->conn->epoch; - call->service_id = call->conn->service_id; - call->in_clientflag = call->conn->in_clientflag; + call->epoch = call->conn->proto.epoch; + call->service_id = call->conn->params.service_id; + call->in_clientflag = call->conn->proto.in_clientflag; /* Add the new call to the hashtable */ rxrpc_call_hash_add(call); @@ -349,6 +350,7 @@ static struct rxrpc_call *rxrpc_alloc_client_call( * - called in process context with IRQs enabled */ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, + struct rxrpc_conn_parameters *cp, struct rxrpc_transport *trans, struct rxrpc_conn_bundle *bundle, unsigned long user_call_ID, @@ -361,7 +363,7 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, rx, trans->debug_id, bundle ? bundle->debug_id : -1, user_call_ID); - call = rxrpc_alloc_client_call(rx, trans, bundle, gfp); + call = rxrpc_alloc_client_call(rx, cp, trans, bundle, gfp); if (IS_ERR(call)) { _leave(" = %ld", PTR_ERR(call)); return call; @@ -524,9 +526,9 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, write_unlock_bh(&rxrpc_call_lock); /* Record copies of information for hashtable lookup */ - call->proto = rx->proto; + call->family = rx->family; call->local = conn->trans->local; - switch (call->proto) { + switch (call->family) { case AF_INET: call->peer_ip.ipv4_addr = conn->trans->peer->srx.transport.sin.sin_addr.s_addr; @@ -539,9 +541,9 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, default: break; } - call->epoch = conn->epoch; - call->service_id = conn->service_id; - call->in_clientflag = conn->in_clientflag; + call->epoch = conn->proto.epoch; + call->service_id = conn->params.service_id; + call->in_clientflag = conn->proto.in_clientflag; /* Add the new call to the hashtable */ rxrpc_call_hash_add(call); diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index 00c92b6..51e280c 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -94,8 +94,8 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn, msg.msg_controllen = 0; msg.msg_flags = 0; - whdr.epoch = htonl(conn->epoch); - whdr.cid = htonl(conn->cid); + whdr.epoch = htonl(conn->proto.epoch); + whdr.cid = htonl(conn->proto.cid); whdr.callNumber = 0; whdr.seq = 0; whdr.type = RXRPC_PACKET_TYPE_ABORT; @@ -103,7 +103,7 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn, whdr.userStatus = 0; whdr.securityIndex = conn->security_ix; whdr._rsvd = 0; - whdr.serviceId = htons(conn->service_id); + whdr.serviceId = htons(conn->params.service_id); word = htonl(conn->local_abort); @@ -220,7 +220,7 @@ static void rxrpc_secure_connection(struct rxrpc_connection *conn) ASSERT(conn->security_ix != 0); - if (!conn->key) { + if (!conn->params.key) { _debug("set up security"); ret = rxrpc_init_server_conn_security(conn); switch (ret) { diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 8ecde4b..c6787b6 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -220,7 +220,7 @@ static void rxrpc_assign_connection_id(struct rxrpc_connection *conn) _enter(""); - epoch = conn->epoch; + epoch = conn->proto.epoch; write_lock_bh(&conn->trans->conn_lock); @@ -237,13 +237,13 @@ attempt_insertion: parent = *p; xconn = rb_entry(parent, struct rxrpc_connection, node); - if (epoch < xconn->epoch) + if (epoch < xconn->proto.epoch) p = &(*p)->rb_left; - else if (epoch > xconn->epoch) + else if (epoch > xconn->proto.epoch) p = &(*p)->rb_right; - else if (cid < xconn->cid) + else if (cid < xconn->proto.cid) p = &(*p)->rb_left; - else if (cid > xconn->cid) + else if (cid > xconn->proto.cid) p = &(*p)->rb_right; else goto id_exists; @@ -254,7 +254,7 @@ attempt_insertion: rb_link_node(&conn->node, parent, p); rb_insert_color(&conn->node, &conn->trans->client_conns); - conn->cid = cid; + conn->proto.cid = cid; write_unlock_bh(&conn->trans->conn_lock); _leave(" [CID %x]", cid); return; @@ -275,8 +275,8 @@ id_exists: goto attempt_insertion; xconn = rb_entry(parent, struct rxrpc_connection, node); - if (epoch < xconn->epoch || - cid < xconn->cid) + if (epoch < xconn->proto.epoch || + cid < xconn->proto.cid) goto attempt_insertion; } } @@ -318,8 +318,8 @@ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn, * connect a call on an exclusive connection */ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, + struct rxrpc_conn_parameters *cp, struct rxrpc_transport *trans, - u16 service_id, struct rxrpc_call *call, gfp_t gfp) { @@ -340,19 +340,21 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, conn->trans = trans; conn->bundle = NULL; - conn->service_id = service_id; - conn->epoch = rxrpc_epoch; - conn->in_clientflag = 0; + conn->params = *cp; + conn->proto.local = cp->local; + conn->proto.epoch = rxrpc_epoch; + conn->proto.cid = 0; + conn->proto.in_clientflag = 0; + conn->proto.family = cp->peer->srx.transport.family; conn->out_clientflag = RXRPC_CLIENT_INITIATED; - conn->cid = 0; conn->state = RXRPC_CONN_CLIENT; conn->avail_calls = RXRPC_MAXCALLS - 1; - conn->security_level = rx->min_sec_level; - conn->key = key_get(rx->key); + + key_get(conn->params.key); ret = rxrpc_init_client_conn_security(conn); if (ret < 0) { - key_put(conn->key); + key_put(conn->params.key); kfree(conn); _leave(" = %d [key]", ret); return ret; @@ -389,7 +391,7 @@ found_channel: conn->channels[chan] = call; call->conn = conn; call->channel = chan; - call->cid = conn->cid | chan; + call->cid = conn->proto.cid | chan; call->call_id = ++conn->call_counter; _net("CONNECT client on conn %d chan %d as call %x", @@ -412,6 +414,7 @@ no_free_channels: * - called in process context with IRQs enabled */ int rxrpc_connect_call(struct rxrpc_sock *rx, + struct rxrpc_conn_parameters *cp, struct rxrpc_transport *trans, struct rxrpc_conn_bundle *bundle, struct rxrpc_call *call, @@ -425,8 +428,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx, _enter("%p,%lx,", rx, call->user_call_ID); if (test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags)) - return rxrpc_connect_exclusive(rx, trans, bundle->service_id, - call, gfp); + return rxrpc_connect_exclusive(rx, cp, trans, call, gfp); spin_lock(&trans->client_lock); for (;;) { @@ -517,19 +519,21 @@ int rxrpc_connect_call(struct rxrpc_sock *rx, candidate->trans = trans; candidate->bundle = bundle; - candidate->service_id = bundle->service_id; - candidate->epoch = rxrpc_epoch; - candidate->in_clientflag = 0; + candidate->params = *cp; + candidate->proto.local = cp->local; + candidate->proto.epoch = rxrpc_epoch; + candidate->proto.cid = 0; + candidate->proto.in_clientflag = 0; + candidate->proto.family = cp->peer->srx.transport.family; candidate->out_clientflag = RXRPC_CLIENT_INITIATED; - candidate->cid = 0; candidate->state = RXRPC_CONN_CLIENT; candidate->avail_calls = RXRPC_MAXCALLS; - candidate->security_level = rx->min_sec_level; - candidate->key = key_get(bundle->key); + + key_get(candidate->params.key); ret = rxrpc_init_client_conn_security(candidate); if (ret < 0) { - key_put(candidate->key); + key_put(candidate->params.key); kfree(candidate); _leave(" = %d [key]", ret); return ret; @@ -577,7 +581,7 @@ found_channel: conn->channels[chan] = call; call->conn = conn; call->channel = chan; - call->cid = conn->cid | chan; + call->cid = conn->proto.cid | chan; call->call_id = ++conn->call_counter; _net("CONNECT client on conn %d chan %d as call %x", @@ -626,15 +630,15 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, while (p) { conn = rb_entry(p, struct rxrpc_connection, node); - _debug("maybe %x", conn->cid); + _debug("maybe %x", conn->proto.cid); - if (epoch < conn->epoch) + if (epoch < conn->proto.epoch) p = p->rb_left; - else if (epoch > conn->epoch) + else if (epoch > conn->proto.epoch) p = p->rb_right; - else if (cid < conn->cid) + else if (cid < conn->proto.cid) p = p->rb_left; - else if (cid > conn->cid) + else if (cid > conn->proto.cid) p = p->rb_right; else goto found_extant_connection; @@ -650,14 +654,17 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, } candidate->trans = trans; - candidate->epoch = hdr->epoch; - candidate->cid = hdr->cid & RXRPC_CIDMASK; - candidate->service_id = hdr->serviceId; + candidate->proto.local = trans->local; + candidate->proto.epoch = hdr->epoch; + candidate->proto.cid = hdr->cid & RXRPC_CIDMASK; + candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED; + candidate->params.local = trans->local; + candidate->params.peer = trans->peer; + candidate->params.service_id = hdr->serviceId; candidate->security_ix = hdr->securityIndex; - candidate->in_clientflag = RXRPC_CLIENT_INITIATED; candidate->out_clientflag = 0; candidate->state = RXRPC_CONN_SERVER; - if (candidate->service_id) + if (candidate->params.service_id) candidate->state = RXRPC_CONN_SERVER_UNSECURED; write_lock_bh(&trans->conn_lock); @@ -668,13 +675,13 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, p = *pp; conn = rb_entry(p, struct rxrpc_connection, node); - if (epoch < conn->epoch) + if (epoch < conn->proto.epoch) pp = &(*pp)->rb_left; - else if (epoch > conn->epoch) + else if (epoch > conn->proto.epoch) pp = &(*pp)->rb_right; - else if (cid < conn->cid) + else if (cid < conn->proto.cid) pp = &(*pp)->rb_left; - else if (cid > conn->cid) + else if (cid > conn->proto.cid) pp = &(*pp)->rb_right; else goto found_extant_second; @@ -696,7 +703,7 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, new = "new"; success: - _net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->cid); + _net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->proto.cid); _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); return conn; @@ -754,15 +761,15 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans, while (p) { conn = rb_entry(p, struct rxrpc_connection, node); - _debug("maybe %x", conn->cid); + _debug("maybe %x", conn->proto.cid); - if (epoch < conn->epoch) + if (epoch < conn->proto.epoch) p = p->rb_left; - else if (epoch > conn->epoch) + else if (epoch > conn->proto.epoch) p = p->rb_right; - else if (cid < conn->cid) + else if (cid < conn->proto.cid) p = p->rb_left; - else if (cid > conn->cid) + else if (cid > conn->proto.cid) p = p->rb_right; else goto found; @@ -816,7 +823,7 @@ static void rxrpc_destroy_connection(struct rxrpc_connection *conn) rxrpc_purge_queue(&conn->rx_queue); conn->security->clear(conn); - key_put(conn->key); + key_put(conn->params.key); key_put(conn->server_key); rxrpc_put_transport(conn->trans); diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index e11e4d7..c030abd 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -360,7 +360,7 @@ void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb) case RXRPC_PACKET_TYPE_BUSY: _proto("Rx BUSY %%%u", sp->hdr.serial); - if (call->conn->out_clientflag) + if (rxrpc_conn_is_service(call->conn)) goto protocol_error; write_lock_bh(&call->state_lock); @@ -533,7 +533,7 @@ static void rxrpc_post_packet_to_call(struct rxrpc_call *call, case RXRPC_CALL_COMPLETE: case RXRPC_CALL_CLIENT_FINAL_ACK: /* complete server call */ - if (call->conn->in_clientflag) + if (rxrpc_conn_is_service(call->conn)) goto dead_call; /* resend last packet of a completed call */ _debug("final ack again"); diff --git a/net/rxrpc/key.c b/net/rxrpc/key.c index 4ad56fa..18c737a 100644 --- a/net/rxrpc/key.c +++ b/net/rxrpc/key.c @@ -987,7 +987,7 @@ int rxrpc_get_server_data_key(struct rxrpc_connection *conn, if (ret < 0) goto error; - conn->key = key; + conn->params.key = key; _leave(" = 0 [%d]", key_serial(key)); return 0; diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index e6fb386..8c51745 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -133,6 +133,7 @@ static struct rxrpc_call * rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, unsigned long user_call_ID) { + struct rxrpc_conn_parameters cp; struct rxrpc_conn_bundle *bundle; struct rxrpc_transport *trans; struct rxrpc_call *call; @@ -146,23 +147,32 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, if (!msg->msg_name) return ERR_PTR(-EDESTADDRREQ); - trans = rxrpc_name_to_transport(rx, msg->msg_name, msg->msg_namelen, 0, + key = rx->key; + if (key && !rx->key->payload.data[0]) + key = NULL; + + memset(&cp, 0, sizeof(cp)); + cp.local = rx->local; + cp.key = rx->key; + cp.security_level = rx->min_sec_level; + cp.exclusive = test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags); + cp.service_id = srx->srx_service; + trans = rxrpc_name_to_transport(&cp, msg->msg_name, msg->msg_namelen, GFP_KERNEL); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out; } + cp.peer = trans->peer; - key = rx->key; - if (key && !rx->key->payload.data[0]) - key = NULL; - bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, GFP_KERNEL); + bundle = rxrpc_get_bundle(rx, trans, cp.key, srx->srx_service, + GFP_KERNEL); if (IS_ERR(bundle)) { ret = PTR_ERR(bundle); goto out_trans; } - call = rxrpc_new_client_call(rx, trans, bundle, user_call_ID, + call = rxrpc_new_client_call(rx, &cp, trans, bundle, user_call_ID, GFP_KERNEL); rxrpc_put_bundle(trans, bundle); rxrpc_put_transport(trans); @@ -664,7 +674,7 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, seq = atomic_inc_return(&call->sequence); - sp->hdr.epoch = conn->epoch; + sp->hdr.epoch = conn->proto.epoch; sp->hdr.cid = call->cid; sp->hdr.callNumber = call->call_id; sp->hdr.seq = seq; diff --git a/net/rxrpc/proc.c b/net/rxrpc/proc.c index 225163b..bbee058 100644 --- a/net/rxrpc/proc.c +++ b/net/rxrpc/proc.c @@ -74,10 +74,10 @@ static int rxrpc_call_seq_show(struct seq_file *seq, void *v) " %-8.8s %08x %lx\n", lbuff, rbuff, - call->conn->service_id, + call->conn->params.service_id, call->cid, call->call_id, - call->conn->in_clientflag ? "Svc" : "Clt", + rxrpc_conn_is_service(call->conn) ? "Svc" : "Clt", atomic_read(&call->usage), rxrpc_call_states[call->state], call->remote_abort ?: call->local_abort, @@ -157,13 +157,13 @@ static int rxrpc_connection_seq_show(struct seq_file *seq, void *v) " %s %08x %08x %08x\n", lbuff, rbuff, - conn->service_id, - conn->cid, + conn->params.service_id, + conn->proto.cid, conn->call_counter, - conn->in_clientflag ? "Svc" : "Clt", + rxrpc_conn_is_service(conn) ? "Svc" : "Clt", atomic_read(&conn->usage), rxrpc_conn_states[conn->state], - key_serial(conn->key), + key_serial(conn->params.key), atomic_read(&conn->serial), atomic_read(&conn->hi_serial)); diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c index 59706b9..c5bac4e 100644 --- a/net/rxrpc/recvmsg.c +++ b/net/rxrpc/recvmsg.c @@ -205,7 +205,7 @@ int rxrpc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, /* we transferred the whole data packet */ if (sp->hdr.flags & RXRPC_LAST_PACKET) { _debug("last"); - if (call->conn->out_clientflag) { + if (rxrpc_conn_is_client(call->conn)) { /* last byte of reply received */ ret = copied; goto terminal_message; diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index 36a6340..134c271 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -58,9 +58,9 @@ static int rxkad_init_connection_security(struct rxrpc_connection *conn) struct rxrpc_key_token *token; int ret; - _enter("{%d},{%x}", conn->debug_id, key_serial(conn->key)); + _enter("{%d},{%x}", conn->debug_id, key_serial(conn->params.key)); - token = conn->key->payload.data[0]; + token = conn->params.key->payload.data[0]; conn->security_ix = token->security_index; ci = crypto_alloc_skcipher("pcbc(fcrypt)", 0, CRYPTO_ALG_ASYNC); @@ -74,7 +74,7 @@ static int rxkad_init_connection_security(struct rxrpc_connection *conn) sizeof(token->kad->session_key)) < 0) BUG(); - switch (conn->security_level) { + switch (conn->params.security_level) { case RXRPC_SECURITY_PLAIN: break; case RXRPC_SECURITY_AUTH: @@ -115,14 +115,14 @@ static void rxkad_prime_packet_security(struct rxrpc_connection *conn) _enter(""); - if (!conn->key) + if (!conn->params.key) return; - token = conn->key->payload.data[0]; + token = conn->params.key->payload.data[0]; memcpy(&iv, token->kad->session_key, sizeof(iv)); - tmpbuf.x[0] = htonl(conn->epoch); - tmpbuf.x[1] = htonl(conn->cid); + tmpbuf.x[0] = htonl(conn->proto.epoch); + tmpbuf.x[1] = htonl(conn->proto.cid); tmpbuf.x[2] = 0; tmpbuf.x[3] = htonl(conn->security_ix); @@ -220,7 +220,7 @@ static int rxkad_secure_packet_encrypt(const struct rxrpc_call *call, rxkhdr.checksum = 0; /* encrypt from the session key */ - token = call->conn->key->payload.data[0]; + token = call->conn->params.key->payload.data[0]; memcpy(&iv, token->kad->session_key, sizeof(iv)); sg_init_one(&sg[0], sechdr, sizeof(rxkhdr)); @@ -277,13 +277,13 @@ static int rxkad_secure_packet(const struct rxrpc_call *call, sp = rxrpc_skb(skb); _enter("{%d{%x}},{#%u},%zu,", - call->debug_id, key_serial(call->conn->key), sp->hdr.seq, - data_size); + call->debug_id, key_serial(call->conn->params.key), + sp->hdr.seq, data_size); if (!call->conn->cipher) return 0; - ret = key_validate(call->conn->key); + ret = key_validate(call->conn->params.key); if (ret < 0) return ret; @@ -312,7 +312,7 @@ static int rxkad_secure_packet(const struct rxrpc_call *call, y = 1; /* zero checksums are not permitted */ sp->hdr.cksum = y; - switch (call->conn->security_level) { + switch (call->conn->params.security_level) { case RXRPC_SECURITY_PLAIN: ret = 0; break; @@ -446,7 +446,7 @@ static int rxkad_verify_packet_encrypt(const struct rxrpc_call *call, skb_to_sgvec(skb, sg, 0, skb->len); /* decrypt from the session key */ - token = call->conn->key->payload.data[0]; + token = call->conn->params.key->payload.data[0]; memcpy(&iv, token->kad->session_key, sizeof(iv)); skcipher_request_set_tfm(req, call->conn->cipher); @@ -516,7 +516,7 @@ static int rxkad_verify_packet(const struct rxrpc_call *call, sp = rxrpc_skb(skb); _enter("{%d{%x}},{#%u}", - call->debug_id, key_serial(call->conn->key), sp->hdr.seq); + call->debug_id, key_serial(call->conn->params.key), sp->hdr.seq); if (!call->conn->cipher) return 0; @@ -557,7 +557,7 @@ static int rxkad_verify_packet(const struct rxrpc_call *call, return -EPROTO; } - switch (call->conn->security_level) { + switch (call->conn->params.security_level) { case RXRPC_SECURITY_PLAIN: ret = 0; break; @@ -589,9 +589,9 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn) u32 serial; int ret; - _enter("{%d,%x}", conn->debug_id, key_serial(conn->key)); + _enter("{%d,%x}", conn->debug_id, key_serial(conn->params.key)); - ret = key_validate(conn->key); + ret = key_validate(conn->params.key); if (ret < 0) return ret; @@ -608,8 +608,8 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn) msg.msg_controllen = 0; msg.msg_flags = 0; - whdr.epoch = htonl(conn->epoch); - whdr.cid = htonl(conn->cid); + whdr.epoch = htonl(conn->proto.epoch); + whdr.cid = htonl(conn->proto.cid); whdr.callNumber = 0; whdr.seq = 0; whdr.type = RXRPC_PACKET_TYPE_CHALLENGE; @@ -617,7 +617,7 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn) whdr.userStatus = 0; whdr.securityIndex = conn->security_ix; whdr._rsvd = 0; - whdr.serviceId = htons(conn->service_id); + whdr.serviceId = htons(conn->params.service_id); iov[0].iov_base = &whdr; iov[0].iov_len = sizeof(whdr); @@ -771,14 +771,14 @@ static int rxkad_respond_to_challenge(struct rxrpc_connection *conn, u32 version, nonce, min_level, abort_code; int ret; - _enter("{%d,%x}", conn->debug_id, key_serial(conn->key)); + _enter("{%d,%x}", conn->debug_id, key_serial(conn->params.key)); - if (!conn->key) { + if (!conn->params.key) { _leave(" = -EPROTO [no key]"); return -EPROTO; } - ret = key_validate(conn->key); + ret = key_validate(conn->params.key); if (ret < 0) { *_abort_code = RXKADEXPIRED; return ret; @@ -801,20 +801,20 @@ static int rxkad_respond_to_challenge(struct rxrpc_connection *conn, goto protocol_error; abort_code = RXKADLEVELFAIL; - if (conn->security_level < min_level) + if (conn->params.security_level < min_level) goto protocol_error; - token = conn->key->payload.data[0]; + token = conn->params.key->payload.data[0]; /* build the response packet */ memset(&resp, 0, sizeof(resp)); resp.version = htonl(RXKAD_VERSION); - resp.encrypted.epoch = htonl(conn->epoch); - resp.encrypted.cid = htonl(conn->cid); + resp.encrypted.epoch = htonl(conn->proto.epoch); + resp.encrypted.cid = htonl(conn->proto.cid); resp.encrypted.securityIndex = htonl(conn->security_ix); resp.encrypted.inc_nonce = htonl(nonce + 1); - resp.encrypted.level = htonl(conn->security_level); + resp.encrypted.level = htonl(conn->params.security_level); resp.kvno = htonl(token->kad->kvno); resp.ticket_len = htonl(token->kad->ticket_len); @@ -1096,9 +1096,9 @@ static int rxkad_verify_response(struct rxrpc_connection *conn, rxkad_decrypt_response(conn, &response, &session_key); abort_code = RXKADSEALEDINCON; - if (ntohl(response.encrypted.epoch) != conn->epoch) + if (ntohl(response.encrypted.epoch) != conn->proto.epoch) goto protocol_error_free; - if (ntohl(response.encrypted.cid) != conn->cid) + if (ntohl(response.encrypted.cid) != conn->proto.cid) goto protocol_error_free; if (ntohl(response.encrypted.securityIndex) != conn->security_ix) goto protocol_error_free; @@ -1122,7 +1122,7 @@ static int rxkad_verify_response(struct rxrpc_connection *conn, level = ntohl(response.encrypted.level); if (level > RXRPC_SECURITY_ENCRYPT) goto protocol_error_free; - conn->security_level = level; + conn->params.security_level = level; /* create a key to hold the security data and expiration time - after * this the connection security can be handled in exactly the same way diff --git a/net/rxrpc/security.c b/net/rxrpc/security.c index d223253..40955d0 100644 --- a/net/rxrpc/security.c +++ b/net/rxrpc/security.c @@ -76,7 +76,7 @@ int rxrpc_init_client_conn_security(struct rxrpc_connection *conn) { const struct rxrpc_security *sec; struct rxrpc_key_token *token; - struct key *key = conn->key; + struct key *key = conn->params.key; int ret; _enter("{%d},{%x}", conn->debug_id, key_serial(key)); @@ -121,7 +121,7 @@ int rxrpc_init_server_conn_security(struct rxrpc_connection *conn) _enter(""); - sprintf(kdesc, "%u:%u", conn->service_id, conn->security_ix); + sprintf(kdesc, "%u:%u", conn->params.service_id, conn->security_ix); sec = rxrpc_security_lookup(conn->security_ix); if (!sec) { @@ -132,7 +132,7 @@ int rxrpc_init_server_conn_security(struct rxrpc_connection *conn) /* find the service */ read_lock_bh(&local->services_lock); list_for_each_entry(rx, &local->services, listen_link) { - if (rx->srx.srx_service == conn->service_id) + if (rx->srx.srx_service == conn->params.service_id) goto found_service; } -- cgit v0.10.2 From 85f32278bd98fa89dff528b0baea4ae6eea4cc5d Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:36 +0100 Subject: rxrpc: Replace conn->trans->{local,peer} with conn->params.{local,peer} Replace accesses of conn->trans->{local,peer} with conn->params.{local,peer} thus making it easier for a future commit to remove the rxrpc_transport struct. This also reduces the number of memory accesses involved. Signed-off-by: David Howells diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c index 1571dfb..b43faf5 100644 --- a/net/rxrpc/call_event.c +++ b/net/rxrpc/call_event.c @@ -545,7 +545,7 @@ static void rxrpc_extract_ackinfo(struct rxrpc_call *call, struct sk_buff *skb, mtu = min(ntohl(ackinfo.rxMTU), ntohl(ackinfo.maxMTU)); - peer = call->conn->trans->peer; + peer = call->conn->params.peer; if (mtu < peer->maxdata) { spin_lock_bh(&peer->lock); peer->maxdata = mtu; @@ -836,8 +836,8 @@ void rxrpc_process_call(struct work_struct *work) /* there's a good chance we're going to have to send a message, so set * one up in advance */ - msg.msg_name = &call->conn->trans->peer->srx.transport; - msg.msg_namelen = call->conn->trans->peer->srx.transport_len; + msg.msg_name = &call->conn->params.peer->srx.transport; + msg.msg_namelen = call->conn->params.peer->srx.transport_len; msg.msg_control = NULL; msg.msg_controllen = 0; msg.msg_flags = 0; @@ -1151,8 +1151,8 @@ send_ACK_with_skew: ack.maxSkew = htons(atomic_read(&call->conn->hi_serial) - ntohl(ack.serial)); send_ACK: - mtu = call->conn->trans->peer->if_mtu; - mtu -= call->conn->trans->peer->hdrsize; + mtu = call->conn->params.peer->if_mtu; + mtu -= call->conn->params.peer->hdrsize; ackinfo.maxMTU = htonl(mtu); ackinfo.rwind = htonl(rxrpc_rx_window_size); @@ -1206,7 +1206,7 @@ send_message_2: len += iov[1].iov_len; } - ret = kernel_sendmsg(call->conn->trans->local->socket, + ret = kernel_sendmsg(call->conn->params.local->socket, &msg, iov, ioc, len); if (ret < 0) { _debug("sendmsg failed: %d", ret); diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index b7c6011..5c2dcea 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -122,7 +122,7 @@ static void rxrpc_call_hash_add(struct rxrpc_call *call) key = rxrpc_call_hashfunc(call->in_clientflag, call->cid, call->call_id, call->epoch, call->service_id, call->family, - call->conn->trans->local, addr_size, + call->conn->params.local, addr_size, call->peer_ip.ipv6_addr); /* Store the full key in the call */ call->hash_key = key; @@ -320,11 +320,11 @@ static struct rxrpc_call *rxrpc_alloc_client_call( switch (call->family) { case AF_INET: call->peer_ip.ipv4_addr = - trans->peer->srx.transport.sin.sin_addr.s_addr; + call->conn->params.peer->srx.transport.sin.sin_addr.s_addr; break; case AF_INET6: memcpy(call->peer_ip.ipv6_addr, - trans->peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8, + call->conn->params.peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8, sizeof(call->peer_ip.ipv6_addr)); break; } @@ -334,9 +334,9 @@ static struct rxrpc_call *rxrpc_alloc_client_call( /* Add the new call to the hashtable */ rxrpc_call_hash_add(call); - spin_lock(&call->conn->trans->peer->lock); - hlist_add_head(&call->error_link, &call->conn->trans->peer->error_targets); - spin_unlock(&call->conn->trans->peer->lock); + spin_lock(&call->conn->params.peer->lock); + hlist_add_head(&call->error_link, &call->conn->params.peer->error_targets); + spin_unlock(&call->conn->params.peer->lock); call->lifetimer.expires = jiffies + rxrpc_max_call_lifetime; add_timer(&call->lifetimer); @@ -517,9 +517,9 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, atomic_inc(&conn->usage); write_unlock_bh(&conn->lock); - spin_lock(&conn->trans->peer->lock); - hlist_add_head(&call->error_link, &conn->trans->peer->error_targets); - spin_unlock(&conn->trans->peer->lock); + spin_lock(&conn->params.peer->lock); + hlist_add_head(&call->error_link, &conn->params.peer->error_targets); + spin_unlock(&conn->params.peer->lock); write_lock_bh(&rxrpc_call_lock); list_add_tail(&call->link, &rxrpc_calls); @@ -527,15 +527,15 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, /* Record copies of information for hashtable lookup */ call->family = rx->family; - call->local = conn->trans->local; + call->local = conn->params.local; switch (call->family) { case AF_INET: call->peer_ip.ipv4_addr = - conn->trans->peer->srx.transport.sin.sin_addr.s_addr; + conn->params.peer->srx.transport.sin.sin_addr.s_addr; break; case AF_INET6: memcpy(call->peer_ip.ipv6_addr, - conn->trans->peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8, + conn->params.peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8, sizeof(call->peer_ip.ipv6_addr)); break; default: @@ -813,9 +813,9 @@ static void rxrpc_cleanup_call(struct rxrpc_call *call) } if (call->conn) { - spin_lock(&call->conn->trans->peer->lock); + spin_lock(&call->conn->params.peer->lock); hlist_del_init(&call->error_link); - spin_unlock(&call->conn->trans->peer->lock); + spin_unlock(&call->conn->params.peer->lock); write_lock_bh(&call->conn->lock); rb_erase(&call->conn_node, &call->conn->calls); diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index 51e280c..a022439 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -88,8 +88,8 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn, rxrpc_abort_calls(conn, RXRPC_CALL_LOCALLY_ABORTED, abort_code); - msg.msg_name = &conn->trans->peer->srx.transport; - msg.msg_namelen = conn->trans->peer->srx.transport_len; + msg.msg_name = &conn->params.peer->srx.transport; + msg.msg_namelen = conn->params.peer->srx.transport_len; msg.msg_control = NULL; msg.msg_controllen = 0; msg.msg_flags = 0; @@ -118,7 +118,7 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn, whdr.serial = htonl(serial); _proto("Tx CONN ABORT %%%u { %d }", serial, conn->local_abort); - ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 2, len); + ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 2, len); if (ret < 0) { _debug("sendmsg failed: %d", ret); return -EAGAIN; diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index c030abd..6af7f40 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -560,7 +560,7 @@ static void rxrpc_post_packet_to_call(struct rxrpc_call *call, dead_call: if (sp->hdr.type != RXRPC_PACKET_TYPE_ABORT) { skb->priority = RX_CALL_DEAD; - rxrpc_reject_packet(call->conn->trans->local, skb); + rxrpc_reject_packet(call->conn->params.local, skb); goto unlock; } free_unlock: diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index 8c51745..becbaa7 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -583,7 +583,7 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, goto maybe_error; } - max = call->conn->trans->peer->maxdata; + max = call->conn->params.peer->maxdata; max -= call->conn->security_size; max &= ~(call->conn->size_align - 1UL); diff --git a/net/rxrpc/proc.c b/net/rxrpc/proc.c index bbee058..9863270 100644 --- a/net/rxrpc/proc.c +++ b/net/rxrpc/proc.c @@ -46,7 +46,7 @@ static void rxrpc_call_seq_stop(struct seq_file *seq, void *v) static int rxrpc_call_seq_show(struct seq_file *seq, void *v) { - struct rxrpc_transport *trans; + struct rxrpc_connection *conn; struct rxrpc_call *call; char lbuff[4 + 4 + 4 + 4 + 5 + 1], rbuff[4 + 4 + 4 + 4 + 5 + 1]; @@ -59,15 +59,15 @@ static int rxrpc_call_seq_show(struct seq_file *seq, void *v) } call = list_entry(v, struct rxrpc_call, link); - trans = call->conn->trans; + conn = call->conn; sprintf(lbuff, "%pI4:%u", - &trans->local->srx.transport.sin.sin_addr, - ntohs(trans->local->srx.transport.sin.sin_port)); + &conn->params.local->srx.transport.sin.sin_addr, + ntohs(conn->params.local->srx.transport.sin.sin_port)); sprintf(rbuff, "%pI4:%u", - &trans->peer->srx.transport.sin.sin_addr, - ntohs(trans->peer->srx.transport.sin.sin_port)); + &conn->params.peer->srx.transport.sin.sin_addr, + ntohs(conn->params.peer->srx.transport.sin.sin_port)); seq_printf(seq, "UDP %-22.22s %-22.22s %4x %08x %08x %s %3u" @@ -129,7 +129,6 @@ static void rxrpc_connection_seq_stop(struct seq_file *seq, void *v) static int rxrpc_connection_seq_show(struct seq_file *seq, void *v) { struct rxrpc_connection *conn; - struct rxrpc_transport *trans; char lbuff[4 + 4 + 4 + 4 + 5 + 1], rbuff[4 + 4 + 4 + 4 + 5 + 1]; if (v == &rxrpc_connections) { @@ -142,15 +141,14 @@ static int rxrpc_connection_seq_show(struct seq_file *seq, void *v) } conn = list_entry(v, struct rxrpc_connection, link); - trans = conn->trans; sprintf(lbuff, "%pI4:%u", - &trans->local->srx.transport.sin.sin_addr, - ntohs(trans->local->srx.transport.sin.sin_port)); + &conn->params.local->srx.transport.sin.sin_addr, + ntohs(conn->params.local->srx.transport.sin.sin_port)); sprintf(rbuff, "%pI4:%u", - &trans->peer->srx.transport.sin.sin_addr, - ntohs(trans->peer->srx.transport.sin.sin_port)); + &conn->params.peer->srx.transport.sin.sin_addr, + ntohs(conn->params.peer->srx.transport.sin.sin_port)); seq_printf(seq, "UDP %-22.22s %-22.22s %4x %08x %08x %s %3u" diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c index c5bac4e..a3fa2ed 100644 --- a/net/rxrpc/recvmsg.c +++ b/net/rxrpc/recvmsg.c @@ -147,9 +147,9 @@ int rxrpc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, if (!continue_call) { if (msg->msg_name) { size_t len = - sizeof(call->conn->trans->peer->srx); + sizeof(call->conn->params.peer->srx); memcpy(msg->msg_name, - &call->conn->trans->peer->srx, len); + &call->conn->params.peer->srx, len); msg->msg_namelen = len; } sock_recv_timestamp(msg, &rx->sk, skb); diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index 134c271..23c05ec 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -602,8 +602,8 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn) challenge.min_level = htonl(0); challenge.__padding = 0; - msg.msg_name = &conn->trans->peer->srx.transport.sin; - msg.msg_namelen = sizeof(conn->trans->peer->srx.transport.sin); + msg.msg_name = &conn->params.peer->srx.transport.sin; + msg.msg_namelen = sizeof(conn->params.peer->srx.transport.sin); msg.msg_control = NULL; msg.msg_controllen = 0; msg.msg_flags = 0; @@ -630,7 +630,7 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn) whdr.serial = htonl(serial); _proto("Tx CHALLENGE %%%u", serial); - ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 2, len); + ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 2, len); if (ret < 0) { _debug("sendmsg failed: %d", ret); return -EAGAIN; @@ -657,8 +657,8 @@ static int rxkad_send_response(struct rxrpc_connection *conn, _enter(""); - msg.msg_name = &conn->trans->peer->srx.transport.sin; - msg.msg_namelen = sizeof(conn->trans->peer->srx.transport.sin); + msg.msg_name = &conn->params.peer->srx.transport.sin; + msg.msg_namelen = sizeof(conn->params.peer->srx.transport.sin); msg.msg_control = NULL; msg.msg_controllen = 0; msg.msg_flags = 0; @@ -684,7 +684,7 @@ static int rxkad_send_response(struct rxrpc_connection *conn, whdr.serial = htonl(serial); _proto("Tx RESPONSE %%%u", serial); - ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 3, len); + ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 3, len); if (ret < 0) { _debug("sendmsg failed: %d", ret); return -EAGAIN; diff --git a/net/rxrpc/security.c b/net/rxrpc/security.c index 40955d0..814d285 100644 --- a/net/rxrpc/security.c +++ b/net/rxrpc/security.c @@ -113,7 +113,7 @@ int rxrpc_init_client_conn_security(struct rxrpc_connection *conn) int rxrpc_init_server_conn_security(struct rxrpc_connection *conn) { const struct rxrpc_security *sec; - struct rxrpc_local *local = conn->trans->local; + struct rxrpc_local *local = conn->params.local; struct rxrpc_sock *rx; struct key *key; key_ref_t kref; -- cgit v0.10.2 From cc8feb8edd92d854be552fe4f5e0eeabca40b9ee Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:37 +0100 Subject: rxrpc: Fix exclusive connection handling "Exclusive connections" are meant to be used for a single client call and then scrapped. The idea is to limit the use of the negotiated security context. The current code, however, isn't doing this: it is instead restricting the socket to a single virtual connection and doing all the calls over that. This is changed such that the socket no longer maintains a special virtual connection over which it will do all the calls, but rather gets a new one each time a new exclusive call is made. Further, using a socket option for this is a poor choice. It should be done on sendmsg with a control message marker instead so that calls can be marked exclusive individually. To that end, add RXRPC_EXCLUSIVE_CALL which, if passed to sendmsg() as a control message element, will cause the call to be done on an single-use connection. The socket option (RXRPC_EXCLUSIVE_CONNECTION) still exists and, if set, will override any lack of RXRPC_EXCLUSIVE_CALL being specified so that programs using the setsockopt() will appear to work the same. Signed-off-by: David Howells diff --git a/include/linux/rxrpc.h b/include/linux/rxrpc.h index 1e8f216..c68307b 100644 --- a/include/linux/rxrpc.h +++ b/include/linux/rxrpc.h @@ -35,7 +35,7 @@ struct sockaddr_rxrpc { */ #define RXRPC_SECURITY_KEY 1 /* [clnt] set client security key */ #define RXRPC_SECURITY_KEYRING 2 /* [srvr] set ring of server security keys */ -#define RXRPC_EXCLUSIVE_CONNECTION 3 /* [clnt] use exclusive RxRPC connection */ +#define RXRPC_EXCLUSIVE_CONNECTION 3 /* Deprecated; use RXRPC_EXCLUSIVE_CALL instead */ #define RXRPC_MIN_SECURITY_LEVEL 4 /* minimum security level */ /* @@ -52,6 +52,7 @@ struct sockaddr_rxrpc { #define RXRPC_LOCAL_ERROR 7 /* -r: local error generated [terminal] */ #define RXRPC_NEW_CALL 8 /* -r: [Service] new incoming call notification */ #define RXRPC_ACCEPT 9 /* s-: [Service] accept request */ +#define RXRPC_EXCLUSIVE_CALL 10 /* s-: Call should be on exclusive connection */ /* * RxRPC security levels diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 48b45a0..73f5c55 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -494,7 +494,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname, ret = -EISCONN; if (rx->sk.sk_state != RXRPC_UNBOUND) goto error; - set_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags); + rx->exclusive = true; goto success; case RXRPC_SECURITY_KEY: @@ -669,11 +669,6 @@ static int rxrpc_release_sock(struct sock *sk) flush_workqueue(rxrpc_workqueue); rxrpc_purge_queue(&sk->sk_receive_queue); - if (rx->conn) { - rxrpc_put_connection(rx->conn); - rx->conn = NULL; - } - if (rx->local) { rxrpc_put_local(rx->local); rx->local = NULL; diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index efe6673..4ca9944 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -37,6 +37,8 @@ struct rxrpc_crypt { #define rxrpc_queue_call(CALL) rxrpc_queue_work(&(CALL)->processor) #define rxrpc_queue_conn(CONN) rxrpc_queue_work(&(CONN)->processor) +struct rxrpc_connection; + /* * sk_state for RxRPC sockets */ @@ -57,7 +59,6 @@ struct rxrpc_sock { struct sock sk; rxrpc_interceptor_t interceptor; /* kernel service Rx interceptor function */ struct rxrpc_local *local; /* local endpoint */ - struct rxrpc_connection *conn; /* exclusive virtual connection */ struct list_head listen_link; /* link in the local endpoint's listen list */ struct list_head secureq; /* calls awaiting connection security clearance */ struct list_head acceptq; /* calls awaiting acceptance */ @@ -66,13 +67,13 @@ struct rxrpc_sock { struct rb_root calls; /* outstanding calls on this socket */ unsigned long flags; #define RXRPC_SOCK_CONNECTED 0 /* connect_srx is set */ -#define RXRPC_SOCK_EXCLUSIVE_CONN 1 /* exclusive connection for a client socket */ rwlock_t call_lock; /* lock for calls */ u32 min_sec_level; /* minimum security level */ #define RXRPC_SECURITY_MAX RXRPC_SECURITY_ENCRYPT + bool exclusive; /* Exclusive connection for a client socket */ + sa_family_t family; /* Protocol family created with */ struct sockaddr_rxrpc srx; /* local address */ struct sockaddr_rxrpc connect_srx; /* Default client address from connect() */ - sa_family_t family; /* protocol family created with */ }; #define rxrpc_sk(__sk) container_of((__sk), struct rxrpc_sock, sk) diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index c6787b6..6164373 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -328,71 +328,57 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, _enter(""); - conn = rx->conn; + conn = rxrpc_alloc_connection(gfp); if (!conn) { - /* not yet present - create a candidate for a new connection - * and then redo the check */ - conn = rxrpc_alloc_connection(gfp); - if (!conn) { - _leave(" = -ENOMEM"); - return -ENOMEM; - } + _leave(" = -ENOMEM"); + return -ENOMEM; + } - conn->trans = trans; - conn->bundle = NULL; - conn->params = *cp; - conn->proto.local = cp->local; - conn->proto.epoch = rxrpc_epoch; - conn->proto.cid = 0; - conn->proto.in_clientflag = 0; - conn->proto.family = cp->peer->srx.transport.family; - conn->out_clientflag = RXRPC_CLIENT_INITIATED; - conn->state = RXRPC_CONN_CLIENT; - conn->avail_calls = RXRPC_MAXCALLS - 1; - - key_get(conn->params.key); - - ret = rxrpc_init_client_conn_security(conn); - if (ret < 0) { - key_put(conn->params.key); - kfree(conn); - _leave(" = %d [key]", ret); - return ret; - } + conn->trans = trans; + conn->bundle = NULL; + conn->params = *cp; + conn->proto.local = cp->local; + conn->proto.epoch = rxrpc_epoch; + conn->proto.cid = 0; + conn->proto.in_clientflag = 0; + conn->proto.family = cp->peer->srx.transport.family; + conn->out_clientflag = RXRPC_CLIENT_INITIATED; + conn->state = RXRPC_CONN_CLIENT; + conn->avail_calls = RXRPC_MAXCALLS - 1; + + key_get(conn->params.key); + + ret = rxrpc_init_client_conn_security(conn); + if (ret < 0) { + key_put(conn->params.key); + kfree(conn); + _leave(" = %d [key]", ret); + return ret; + } - write_lock_bh(&rxrpc_connection_lock); - list_add_tail(&conn->link, &rxrpc_connections); - write_unlock_bh(&rxrpc_connection_lock); + write_lock_bh(&rxrpc_connection_lock); + list_add_tail(&conn->link, &rxrpc_connections); + write_unlock_bh(&rxrpc_connection_lock); - spin_lock(&trans->client_lock); - atomic_inc(&trans->usage); + spin_lock(&trans->client_lock); + atomic_inc(&trans->usage); - _net("CONNECT EXCL new %d on TRANS %d", - conn->debug_id, conn->trans->debug_id); + _net("CONNECT EXCL new %d on TRANS %d", + conn->debug_id, conn->trans->debug_id); - rxrpc_assign_connection_id(conn); - rx->conn = conn; - } else { - spin_lock(&trans->client_lock); - } + rxrpc_assign_connection_id(conn); - /* we've got a connection with a free channel and we can now attach the - * call to it - * - we're holding the transport's client lock - * - we're holding a reference on the connection + /* Since no one else can use the connection, we just use the first + * channel. */ - for (chan = 0; chan < RXRPC_MAXCALLS; chan++) - if (!conn->channels[chan]) - goto found_channel; - goto no_free_channels; - -found_channel: + chan = 0; atomic_inc(&conn->usage); conn->channels[chan] = call; + conn->call_counter = 1; call->conn = conn; call->channel = chan; call->cid = conn->proto.cid | chan; - call->call_id = ++conn->call_counter; + call->call_id = 1; _net("CONNECT client on conn %d chan %d as call %x", conn->debug_id, chan, call->call_id); @@ -402,11 +388,6 @@ found_channel: rxrpc_add_call_ID_to_conn(conn, call); _leave(" = 0"); return 0; - -no_free_channels: - spin_unlock(&trans->client_lock); - _leave(" = -ENOSR"); - return -ENOSR; } /* @@ -427,7 +408,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx, _enter("%p,%lx,", rx, call->user_call_ID); - if (test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags)) + if (cp->exclusive) return rxrpc_connect_exclusive(rx, cp, trans, call, gfp); spin_lock(&trans->client_lock); diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index becbaa7..6f8ab0e 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -35,7 +35,8 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, static int rxrpc_sendmsg_cmsg(struct msghdr *msg, unsigned long *user_call_ID, enum rxrpc_command *command, - u32 *abort_code) + u32 *abort_code, + bool *_exclusive) { struct cmsghdr *cmsg; bool got_user_ID = false; @@ -93,6 +94,11 @@ static int rxrpc_sendmsg_cmsg(struct msghdr *msg, return -EINVAL; break; + case RXRPC_EXCLUSIVE_CALL: + *_exclusive = true; + if (len != 0) + return -EINVAL; + break; default: return -EINVAL; } @@ -131,7 +137,7 @@ static void rxrpc_send_abort(struct rxrpc_call *call, u32 abort_code) */ static struct rxrpc_call * rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, - unsigned long user_call_ID) + unsigned long user_call_ID, bool exclusive) { struct rxrpc_conn_parameters cp; struct rxrpc_conn_bundle *bundle; @@ -155,7 +161,7 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, cp.local = rx->local; cp.key = rx->key; cp.security_level = rx->min_sec_level; - cp.exclusive = test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags); + cp.exclusive = rx->exclusive | exclusive; cp.service_id = srx->srx_service; trans = rxrpc_name_to_transport(&cp, msg->msg_name, msg->msg_namelen, GFP_KERNEL); @@ -201,12 +207,14 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) enum rxrpc_command cmd; struct rxrpc_call *call; unsigned long user_call_ID = 0; + bool exclusive = false; u32 abort_code = 0; int ret; _enter(""); - ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code); + ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code, + &exclusive); if (ret < 0) return ret; @@ -224,7 +232,8 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) if (!call) { if (cmd != RXRPC_CMD_SEND_DATA) return -EBADSLT; - call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID); + call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID, + exclusive); if (IS_ERR(call)) return PTR_ERR(call); } -- cgit v0.10.2 From 42886ffe77f142c36ecf585d60fff2edd06b5be8 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 16 Jun 2016 13:31:07 +0100 Subject: rxrpc: Pass sk_buff * rather than rxrpc_host_header * to functions Pass a pointer to struct sk_buff rather than struct rxrpc_host_header to functions so that they can in the future get at transport protocol parameters rather than just RxRPC parameters. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 4ca9944..60ba22f 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -544,7 +544,7 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *, unsigned long, gfp_t); struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *, struct rxrpc_connection *, - struct rxrpc_host_header *); + struct sk_buff *); void rxrpc_release_call(struct rxrpc_call *); void rxrpc_release_calls_on_socket(struct rxrpc_sock *); void __rxrpc_put_call(struct rxrpc_call *); @@ -574,9 +574,9 @@ int rxrpc_connect_call(struct rxrpc_sock *, struct rxrpc_conn_parameters *, void rxrpc_put_connection(struct rxrpc_connection *); void __exit rxrpc_destroy_all_connections(void); struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *, - struct rxrpc_host_header *); + struct sk_buff *); extern struct rxrpc_connection * -rxrpc_incoming_connection(struct rxrpc_transport *, struct rxrpc_host_header *); +rxrpc_incoming_connection(struct rxrpc_transport *, struct sk_buff *); static inline bool rxrpc_conn_is_client(const struct rxrpc_connection *conn) { diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index 553b67c1..5a70dc4 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -110,7 +110,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, goto error; } - conn = rxrpc_incoming_connection(trans, &sp->hdr); + conn = rxrpc_incoming_connection(trans, skb); rxrpc_put_transport(trans); if (IS_ERR(conn)) { _debug("no conn"); @@ -118,7 +118,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, goto error; } - call = rxrpc_incoming_call(rx, conn, &sp->hdr); + call = rxrpc_incoming_call(rx, conn, skb); rxrpc_put_connection(conn); if (IS_ERR(call)) { _debug("no call"); diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 5c2dcea..d83f2cb 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -421,8 +421,9 @@ found_user_ID_now_present: */ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, struct rxrpc_connection *conn, - struct rxrpc_host_header *hdr) + struct sk_buff *skb) { + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rxrpc_call *call, *candidate; struct rb_node **p, *parent; u32 call_id; @@ -435,13 +436,13 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, if (!candidate) return ERR_PTR(-EBUSY); - candidate->socket = rx; - candidate->conn = conn; - candidate->cid = hdr->cid; - candidate->call_id = hdr->callNumber; - candidate->channel = hdr->cid & RXRPC_CHANNELMASK; - candidate->rx_data_post = 0; - candidate->state = RXRPC_CALL_SERVER_ACCEPTING; + candidate->socket = rx; + candidate->conn = conn; + candidate->cid = sp->hdr.cid; + candidate->call_id = sp->hdr.callNumber; + candidate->channel = sp->hdr.cid & RXRPC_CHANNELMASK; + candidate->rx_data_post = 0; + candidate->state = RXRPC_CALL_SERVER_ACCEPTING; if (conn->security_ix > 0) candidate->state = RXRPC_CALL_SERVER_SECURING; @@ -450,7 +451,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, /* set the channel for this call */ call = conn->channels[candidate->channel]; _debug("channel[%u] is %p", candidate->channel, call); - if (call && call->call_id == hdr->callNumber) { + if (call && call->call_id == sp->hdr.callNumber) { /* already set; must've been a duplicate packet */ _debug("extant call [%d]", call->state); ASSERTCMP(call->conn, ==, conn); @@ -488,7 +489,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, /* check the call number isn't duplicate */ _debug("check dup"); - call_id = hdr->callNumber; + call_id = sp->hdr.callNumber; p = &conn->calls.rb_node; parent = NULL; while (*p) { diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 6164373..3b42fc4 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -588,10 +588,10 @@ interrupted: * get a record of an incoming connection */ struct rxrpc_connection * -rxrpc_incoming_connection(struct rxrpc_transport *trans, - struct rxrpc_host_header *hdr) +rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb) { struct rxrpc_connection *conn, *candidate = NULL; + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rb_node *p, **pp; const char *new = "old"; __be32 epoch; @@ -599,10 +599,10 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, _enter(""); - ASSERT(hdr->flags & RXRPC_CLIENT_INITIATED); + ASSERT(sp->hdr.flags & RXRPC_CLIENT_INITIATED); - epoch = hdr->epoch; - cid = hdr->cid & RXRPC_CIDMASK; + epoch = sp->hdr.epoch; + cid = sp->hdr.cid & RXRPC_CIDMASK; /* search the connection list first */ read_lock_bh(&trans->conn_lock); @@ -634,19 +634,19 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, return ERR_PTR(-ENOMEM); } - candidate->trans = trans; - candidate->proto.local = trans->local; - candidate->proto.epoch = hdr->epoch; - candidate->proto.cid = hdr->cid & RXRPC_CIDMASK; - candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED; - candidate->params.local = trans->local; - candidate->params.peer = trans->peer; - candidate->params.service_id = hdr->serviceId; - candidate->security_ix = hdr->securityIndex; - candidate->out_clientflag = 0; - candidate->state = RXRPC_CONN_SERVER; + candidate->trans = trans; + candidate->proto.local = trans->local; + candidate->proto.epoch = sp->hdr.epoch; + candidate->proto.cid = sp->hdr.cid & RXRPC_CIDMASK; + candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED; + candidate->params.local = trans->local; + candidate->params.peer = trans->peer; + candidate->params.service_id = sp->hdr.serviceId; + candidate->security_ix = sp->hdr.securityIndex; + candidate->out_clientflag = 0; + candidate->state = RXRPC_CONN_SERVER; if (candidate->params.service_id) - candidate->state = RXRPC_CONN_SERVER_UNSECURED; + candidate->state = RXRPC_CONN_SERVER_UNSECURED; write_lock_bh(&trans->conn_lock); @@ -691,7 +691,7 @@ success: /* we found the connection in the list immediately */ found_extant_connection: - if (hdr->securityIndex != conn->security_ix) { + if (sp->hdr.securityIndex != conn->security_ix) { read_unlock_bh(&trans->conn_lock); goto security_mismatch; } @@ -701,7 +701,7 @@ found_extant_connection: /* we found the connection on the second time through the list */ found_extant_second: - if (hdr->securityIndex != conn->security_ix) { + if (sp->hdr.securityIndex != conn->security_ix) { write_unlock_bh(&trans->conn_lock); goto security_mismatch; } @@ -721,20 +721,21 @@ security_mismatch: * packet */ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans, - struct rxrpc_host_header *hdr) + struct sk_buff *skb) { struct rxrpc_connection *conn; + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rb_node *p; u32 epoch, cid; - _enter(",{%x,%x}", hdr->cid, hdr->flags); + _enter(",{%x,%x}", sp->hdr.cid, sp->hdr.flags); read_lock_bh(&trans->conn_lock); - cid = hdr->cid & RXRPC_CIDMASK; - epoch = hdr->epoch; + cid = sp->hdr.cid & RXRPC_CIDMASK; + epoch = sp->hdr.epoch; - if (hdr->flags & RXRPC_CLIENT_INITIATED) + if (sp->hdr.flags & RXRPC_CLIENT_INITIATED) p = trans->server_conns.rb_node; else p = trans->client_conns.rb_node; diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 6af7f40..cf540ef 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -628,8 +628,7 @@ int rxrpc_extract_header(struct rxrpc_skb_priv *sp, struct sk_buff *skb) } static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local, - struct sk_buff *skb, - struct rxrpc_skb_priv *sp) + struct sk_buff *skb) { struct rxrpc_peer *peer; struct rxrpc_transport *trans; @@ -647,7 +646,7 @@ static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local, if (!trans) goto cant_find_conn; - conn = rxrpc_find_connection(trans, &sp->hdr); + conn = rxrpc_find_connection(trans, skb); rxrpc_put_transport(trans); if (!conn) goto cant_find_conn; @@ -739,7 +738,7 @@ void rxrpc_data_ready(struct sock *sk) * old-fashioned way doesn't really hurt */ struct rxrpc_connection *conn; - conn = rxrpc_conn_from_local(local, skb, sp); + conn = rxrpc_conn_from_local(local, skb); if (!conn) goto cant_route_call; -- cgit v0.10.2 From b3f575043fcd2926616a794db3f22280740fea6d Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 21 Jun 2016 16:10:03 +0100 Subject: rxrpc: rxrpc_connection_lock shouldn't be a BH lock, but conn_lock is rxrpc_connection_lock shouldn't be accessed as a BH-excluding lock. It's only accessed in a few places and none of those are in BH-context. rxrpc_transport::conn_lock, however, *is* a BH-excluding lock and should be accessed so consistently. Signed-off-by: David Howells diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 3b42fc4..cab2f6d 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -356,9 +356,9 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, return ret; } - write_lock_bh(&rxrpc_connection_lock); + write_lock(&rxrpc_connection_lock); list_add_tail(&conn->link, &rxrpc_connections); - write_unlock_bh(&rxrpc_connection_lock); + write_unlock(&rxrpc_connection_lock); spin_lock(&trans->client_lock); atomic_inc(&trans->usage); @@ -677,9 +677,9 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb) write_unlock_bh(&trans->conn_lock); - write_lock_bh(&rxrpc_connection_lock); + write_lock(&rxrpc_connection_lock); list_add_tail(&conn->link, &rxrpc_connections); - write_unlock_bh(&rxrpc_connection_lock); + write_unlock(&rxrpc_connection_lock); new = "new"; @@ -828,7 +828,7 @@ static void rxrpc_connection_reaper(struct work_struct *work) now = ktime_get_seconds(); earliest = ULONG_MAX; - write_lock_bh(&rxrpc_connection_lock); + write_lock(&rxrpc_connection_lock); list_for_each_entry_safe(conn, _p, &rxrpc_connections, link) { _debug("reap CONN %d { u=%d,t=%ld }", conn->debug_id, atomic_read(&conn->usage), @@ -838,7 +838,7 @@ static void rxrpc_connection_reaper(struct work_struct *work) continue; spin_lock(&conn->trans->client_lock); - write_lock(&conn->trans->conn_lock); + write_lock_bh(&conn->trans->conn_lock); reap_time = conn->put_time + rxrpc_connection_expiry; if (atomic_read(&conn->usage) > 0) { @@ -860,10 +860,10 @@ static void rxrpc_connection_reaper(struct work_struct *work) earliest = reap_time; } - write_unlock(&conn->trans->conn_lock); + write_unlock_bh(&conn->trans->conn_lock); spin_unlock(&conn->trans->client_lock); } - write_unlock_bh(&rxrpc_connection_lock); + write_unlock(&rxrpc_connection_lock); if (earliest != ULONG_MAX) { _debug("reschedule reaper %ld", (long) earliest - now); -- cgit v0.10.2 From 4a3388c8033e4ea00f06a341d5ed4a20a7da89de Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:37 +0100 Subject: rxrpc: Use IDR to allocate client conn IDs on a machine-wide basis Use the IDR facility to allocate client connection IDs on a machine-wide basis so that each client connection has a unique identifier. When the connection ID space wraps, we advance the epoch by 1, thereby effectively having a 62-bit ID space. The IDR facility is then used to look up client connections during incoming packet routing instead of using an rbtree rooted on the transport. This change allows for the removal of the transport in the future and also means that client connections can be looked up directly in the data-ready handler by connection ID. The ID management code is placed in a new file, conn-client.c, to which all the client connection-specific code will eventually move. Note that the IDR tree gets very expensive on memory if the connection IDs are widely scattered throughout the number space, so we shall need to retire connections that have, say, an ID more than four times the maximum number of client conns away from the current allocation point to try and keep the IDs concentrated. We will also need to retire connections from an old epoch. Also note that, for the moment, a pointer to the transport has to be passed through into the ID allocation function so that we can take a BH lock to prevent a locking issue against in-BH lookup of client connections. This will go away later when RCU is used for server connections also. Signed-off-by: David Howells diff --git a/net/rxrpc/Makefile b/net/rxrpc/Makefile index b005027..cfa2215 100644 --- a/net/rxrpc/Makefile +++ b/net/rxrpc/Makefile @@ -7,6 +7,7 @@ af-rxrpc-y := \ call_accept.o \ call_event.o \ call_object.o \ + conn_client.o \ conn_event.o \ conn_object.o \ input.o \ diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 73f5c55..408bd02 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -858,6 +858,8 @@ static void __exit af_rxrpc_exit(void) _debug("synchronise RCU"); rcu_barrier(); _debug("destroy locals"); + ASSERT(idr_is_empty(&rxrpc_client_conn_ids)); + idr_destroy(&rxrpc_client_conn_ids); rxrpc_destroy_all_locals(); remove_proc_entry("rxrpc_conns", init_net.proc_net); diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 60ba22f..8996650 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -233,7 +233,6 @@ struct rxrpc_transport { struct rxrpc_local *local; /* local transport endpoint */ struct rxrpc_peer *peer; /* remote transport endpoint */ struct rb_root bundles; /* client connection bundles on this transport */ - struct rb_root client_conns; /* client connections on this transport */ struct rb_root server_conns; /* server connections on this transport */ struct list_head link; /* link in master session list */ unsigned long put_time; /* time at which to reap */ @@ -241,7 +240,6 @@ struct rxrpc_transport { rwlock_t conn_lock; /* lock for active/dead connections */ atomic_t usage; int debug_id; /* debug ID for printks */ - unsigned int conn_idcounter; /* connection ID counter (client) */ }; /* @@ -312,6 +310,8 @@ struct rxrpc_connection { struct key *server_key; /* security for this service */ struct crypto_skcipher *cipher; /* encryption handle */ struct rxrpc_crypt csum_iv; /* packet checksum base */ + unsigned long flags; +#define RXRPC_CONN_HAS_IDR 0 /* - Has a client conn ID assigned */ unsigned long events; #define RXRPC_CONN_CHALLENGE 0 /* send challenge packet */ unsigned long put_time; /* time at which to reap */ @@ -551,6 +551,15 @@ void __rxrpc_put_call(struct rxrpc_call *); void __exit rxrpc_destroy_all_calls(void); /* + * conn_client.c + */ +extern struct idr rxrpc_client_conn_ids; + +int rxrpc_get_client_connection_id(struct rxrpc_connection *, + struct rxrpc_transport *, gfp_t); +void rxrpc_put_client_connection_id(struct rxrpc_connection *); + +/* * conn_event.c */ void rxrpc_process_connection(struct work_struct *); diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c new file mode 100644 index 0000000..2cccb4b --- /dev/null +++ b/net/rxrpc/conn_client.c @@ -0,0 +1,99 @@ +/* Client connection-specific management code. + * + * Copyright (C) 2016 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include "ar-internal.h" + +/* + * We use machine-unique IDs for our client connections. + */ +DEFINE_IDR(rxrpc_client_conn_ids); +static DEFINE_SPINLOCK(rxrpc_conn_id_lock); + +/* + * Get a connection ID and epoch for a client connection from the global pool. + * The connection struct pointer is then recorded in the idr radix tree. The + * epoch is changed if this wraps. + * + * TODO: The IDR tree gets very expensive on memory if the connection IDs are + * widely scattered throughout the number space, so we shall need to retire + * connections that have, say, an ID more than four times the maximum number of + * client conns away from the current allocation point to try and keep the IDs + * concentrated. We will also need to retire connections from an old epoch. + */ +int rxrpc_get_client_connection_id(struct rxrpc_connection *conn, + struct rxrpc_transport *trans, + gfp_t gfp) +{ + u32 epoch; + int id; + + _enter(""); + + idr_preload(gfp); + write_lock_bh(&trans->conn_lock); + spin_lock(&rxrpc_conn_id_lock); + + epoch = rxrpc_epoch; + + /* We could use idr_alloc_cyclic() here, but we really need to know + * when the thing wraps so that we can advance the epoch. + */ + if (rxrpc_client_conn_ids.cur == 0) + rxrpc_client_conn_ids.cur = 1; + id = idr_alloc(&rxrpc_client_conn_ids, conn, + rxrpc_client_conn_ids.cur, 0x40000000, GFP_NOWAIT); + if (id < 0) { + if (id != -ENOSPC) + goto error; + id = idr_alloc(&rxrpc_client_conn_ids, conn, + 1, 0x40000000, GFP_NOWAIT); + if (id < 0) + goto error; + epoch++; + rxrpc_epoch = epoch; + } + rxrpc_client_conn_ids.cur = id + 1; + + spin_unlock(&rxrpc_conn_id_lock); + write_unlock_bh(&trans->conn_lock); + idr_preload_end(); + + conn->proto.epoch = epoch; + conn->proto.cid = id << RXRPC_CIDSHIFT; + set_bit(RXRPC_CONN_HAS_IDR, &conn->flags); + _leave(" [CID %x:%x]", epoch, conn->proto.cid); + return 0; + +error: + spin_unlock(&rxrpc_conn_id_lock); + write_unlock_bh(&trans->conn_lock); + idr_preload_end(); + _leave(" = %d", id); + return id; +} + +/* + * Release a connection ID for a client connection from the global pool. + */ +void rxrpc_put_client_connection_id(struct rxrpc_connection *conn) +{ + if (test_bit(RXRPC_CONN_HAS_IDR, &conn->flags)) { + spin_lock(&rxrpc_conn_id_lock); + idr_remove(&rxrpc_client_conn_ids, + conn->proto.cid >> RXRPC_CIDSHIFT); + spin_unlock(&rxrpc_conn_id_lock); + } +} diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index cab2f6d..312b750 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -207,81 +207,6 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) } /* - * assign a connection ID to a connection and add it to the transport's - * connection lookup tree - * - called with transport client lock held - */ -static void rxrpc_assign_connection_id(struct rxrpc_connection *conn) -{ - struct rxrpc_connection *xconn; - struct rb_node *parent, **p; - __be32 epoch; - u32 cid; - - _enter(""); - - epoch = conn->proto.epoch; - - write_lock_bh(&conn->trans->conn_lock); - - conn->trans->conn_idcounter += RXRPC_CID_INC; - if (conn->trans->conn_idcounter < RXRPC_CID_INC) - conn->trans->conn_idcounter = RXRPC_CID_INC; - cid = conn->trans->conn_idcounter; - -attempt_insertion: - parent = NULL; - p = &conn->trans->client_conns.rb_node; - - while (*p) { - parent = *p; - xconn = rb_entry(parent, struct rxrpc_connection, node); - - if (epoch < xconn->proto.epoch) - p = &(*p)->rb_left; - else if (epoch > xconn->proto.epoch) - p = &(*p)->rb_right; - else if (cid < xconn->proto.cid) - p = &(*p)->rb_left; - else if (cid > xconn->proto.cid) - p = &(*p)->rb_right; - else - goto id_exists; - } - - /* we've found a suitable hole - arrange for this connection to occupy - * it */ - rb_link_node(&conn->node, parent, p); - rb_insert_color(&conn->node, &conn->trans->client_conns); - - conn->proto.cid = cid; - write_unlock_bh(&conn->trans->conn_lock); - _leave(" [CID %x]", cid); - return; - - /* we found a connection with the proposed ID - walk the tree from that - * point looking for the next unused ID */ -id_exists: - for (;;) { - cid += RXRPC_CID_INC; - if (cid < RXRPC_CID_INC) { - cid = RXRPC_CID_INC; - conn->trans->conn_idcounter = cid; - goto attempt_insertion; - } - - parent = rb_next(parent); - if (!parent) - goto attempt_insertion; - - xconn = rb_entry(parent, struct rxrpc_connection, node); - if (epoch < xconn->proto.epoch || - cid < xconn->proto.cid) - goto attempt_insertion; - } -} - -/* * add a call to a connection's call-by-ID tree */ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn, @@ -315,27 +240,24 @@ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn, } /* - * connect a call on an exclusive connection + * Allocate a client connection. */ -static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, - struct rxrpc_conn_parameters *cp, - struct rxrpc_transport *trans, - struct rxrpc_call *call, - gfp_t gfp) +static struct rxrpc_connection * +rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, + struct rxrpc_transport *trans, + gfp_t gfp) { struct rxrpc_connection *conn; - int chan, ret; + int ret; _enter(""); conn = rxrpc_alloc_connection(gfp); if (!conn) { _leave(" = -ENOMEM"); - return -ENOMEM; + return ERR_PTR(-ENOMEM); } - conn->trans = trans; - conn->bundle = NULL; conn->params = *cp; conn->proto.local = cp->local; conn->proto.epoch = rxrpc_epoch; @@ -344,35 +266,75 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, conn->proto.family = cp->peer->srx.transport.family; conn->out_clientflag = RXRPC_CLIENT_INITIATED; conn->state = RXRPC_CONN_CLIENT; - conn->avail_calls = RXRPC_MAXCALLS - 1; - key_get(conn->params.key); + switch (conn->proto.family) { + case AF_INET: + conn->proto.addr_size = sizeof(conn->proto.ipv4_addr); + conn->proto.ipv4_addr = cp->peer->srx.transport.sin.sin_addr; + conn->proto.port = cp->peer->srx.transport.sin.sin_port; + break; + } + + ret = rxrpc_get_client_connection_id(conn, trans, gfp); + if (ret < 0) + goto error_0; ret = rxrpc_init_client_conn_security(conn); - if (ret < 0) { - key_put(conn->params.key); - kfree(conn); - _leave(" = %d [key]", ret); - return ret; - } + if (ret < 0) + goto error_1; + + conn->security->prime_packet_security(conn); write_lock(&rxrpc_connection_lock); list_add_tail(&conn->link, &rxrpc_connections); write_unlock(&rxrpc_connection_lock); - spin_lock(&trans->client_lock); + key_get(conn->params.key); + + _leave(" = %p", conn); + return conn; + +error_1: + rxrpc_put_client_connection_id(conn); +error_0: + kfree(conn); + _leave(" = %d", ret); + return ERR_PTR(ret); +} + +/* + * connect a call on an exclusive connection + */ +static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, + struct rxrpc_conn_parameters *cp, + struct rxrpc_transport *trans, + struct rxrpc_call *call, + gfp_t gfp) +{ + struct rxrpc_connection *conn; + int chan; + + _enter(""); + + conn = rxrpc_alloc_client_connection(cp, trans, gfp); + if (IS_ERR(conn)) { + _leave(" = %ld", PTR_ERR(conn)); + return PTR_ERR(conn); + } + atomic_inc(&trans->usage); + conn->trans = trans; + conn->bundle = NULL; _net("CONNECT EXCL new %d on TRANS %d", conn->debug_id, conn->trans->debug_id); - rxrpc_assign_connection_id(conn); - /* Since no one else can use the connection, we just use the first * channel. */ chan = 0; atomic_inc(&conn->usage); + conn->avail_calls = RXRPC_MAXCALLS - 1; conn->channels[chan] = call; conn->call_counter = 1; call->conn = conn; @@ -383,8 +345,6 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, _net("CONNECT client on conn %d chan %d as call %x", conn->debug_id, chan, call->call_id); - spin_unlock(&trans->client_lock); - rxrpc_add_call_ID_to_conn(conn, call); _leave(" = 0"); return 0; @@ -402,7 +362,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx, gfp_t gfp) { struct rxrpc_connection *conn, *candidate; - int chan, ret; + int chan; DECLARE_WAITQUEUE(myself, current); @@ -492,51 +452,25 @@ int rxrpc_connect_call(struct rxrpc_sock *rx, /* not yet present - create a candidate for a new connection and then * redo the check */ - candidate = rxrpc_alloc_connection(gfp); + candidate = rxrpc_alloc_client_connection(cp, trans, gfp); if (!candidate) { _leave(" = -ENOMEM"); return -ENOMEM; } + atomic_inc(&bundle->usage); + atomic_inc(&trans->usage); candidate->trans = trans; candidate->bundle = bundle; - candidate->params = *cp; - candidate->proto.local = cp->local; - candidate->proto.epoch = rxrpc_epoch; - candidate->proto.cid = 0; - candidate->proto.in_clientflag = 0; - candidate->proto.family = cp->peer->srx.transport.family; - candidate->out_clientflag = RXRPC_CLIENT_INITIATED; - candidate->state = RXRPC_CONN_CLIENT; - candidate->avail_calls = RXRPC_MAXCALLS; - - key_get(candidate->params.key); - - ret = rxrpc_init_client_conn_security(candidate); - if (ret < 0) { - key_put(candidate->params.key); - kfree(candidate); - _leave(" = %d [key]", ret); - return ret; - } - - write_lock_bh(&rxrpc_connection_lock); - list_add_tail(&candidate->link, &rxrpc_connections); - write_unlock_bh(&rxrpc_connection_lock); spin_lock(&trans->client_lock); list_add(&candidate->bundle_link, &bundle->unused_conns); bundle->num_conns++; - atomic_inc(&bundle->usage); - atomic_inc(&trans->usage); _net("CONNECT new %d on TRANS %d", candidate->debug_id, candidate->trans->debug_id); - rxrpc_assign_connection_id(candidate); - candidate->security->prime_packet_security(candidate); - /* leave the candidate lurking in zombie mode attached to the * bundle until we're ready for it */ rxrpc_put_connection(candidate); @@ -735,25 +669,27 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans, cid = sp->hdr.cid & RXRPC_CIDMASK; epoch = sp->hdr.epoch; - if (sp->hdr.flags & RXRPC_CLIENT_INITIATED) + if (sp->hdr.flags & RXRPC_CLIENT_INITIATED) { p = trans->server_conns.rb_node; - else - p = trans->client_conns.rb_node; - - while (p) { - conn = rb_entry(p, struct rxrpc_connection, node); - - _debug("maybe %x", conn->proto.cid); - - if (epoch < conn->proto.epoch) - p = p->rb_left; - else if (epoch > conn->proto.epoch) - p = p->rb_right; - else if (cid < conn->proto.cid) - p = p->rb_left; - else if (cid > conn->proto.cid) - p = p->rb_right; - else + while (p) { + conn = rb_entry(p, struct rxrpc_connection, node); + + _debug("maybe %x", conn->proto.cid); + + if (epoch < conn->proto.epoch) + p = p->rb_left; + else if (epoch > conn->proto.epoch) + p = p->rb_right; + else if (cid < conn->proto.cid) + p = p->rb_left; + else if (cid > conn->proto.cid) + p = p->rb_right; + else + goto found; + } + } else { + conn = idr_find(&rxrpc_client_conn_ids, cid >> RXRPC_CIDSHIFT); + if (conn && conn->proto.epoch == epoch) goto found; } @@ -846,8 +782,7 @@ static void rxrpc_connection_reaper(struct work_struct *work) } else if (reap_time <= now) { list_move_tail(&conn->link, &graveyard); if (conn->out_clientflag) - rb_erase(&conn->node, - &conn->trans->client_conns); + rxrpc_put_client_connection_id(conn); else rb_erase(&conn->node, &conn->trans->server_conns); diff --git a/net/rxrpc/transport.c b/net/rxrpc/transport.c index 24c7121..140628d 100644 --- a/net/rxrpc/transport.c +++ b/net/rxrpc/transport.c @@ -47,12 +47,10 @@ static struct rxrpc_transport *rxrpc_alloc_transport(struct rxrpc_local *local, trans->peer = peer; INIT_LIST_HEAD(&trans->link); trans->bundles = RB_ROOT; - trans->client_conns = RB_ROOT; trans->server_conns = RB_ROOT; spin_lock_init(&trans->client_lock); rwlock_init(&trans->conn_lock); atomic_set(&trans->usage, 1); - trans->conn_idcounter = peer->srx.srx_service << 16; trans->debug_id = atomic_inc_return(&rxrpc_debug_id); } -- cgit v0.10.2 From f4552c2d248e9d9f6f728ea32eb25f600d3d6cd6 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 17 Jun 2016 11:00:48 +0100 Subject: rxrpc: Validate the net address given to rxrpc_kernel_begin_call() Validate the net address given to rxrpc_kernel_begin_call() before using it. Whilst this should be mostly unnecessary for in-kernel users, it does clear the tail of the address struct in case we want to hash or compare the whole thing. Signed-off-by: David Howells diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 408bd02..b29bb50 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -280,9 +280,14 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, struct rxrpc_transport *trans; struct rxrpc_call *call; struct rxrpc_sock *rx = rxrpc_sk(sock->sk); + int ret; _enter(",,%x,%lx", key_serial(key), user_call_ID); + ret = rxrpc_validate_address(rx, srx, sizeof(*srx)); + if (ret < 0) + return ERR_PTR(ret); + lock_sock(&rx->sk); if (!key) -- cgit v0.10.2 From f4e7da8cde87d0f7e9fb806918f7ec283912b694 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 17 Jun 2016 11:07:55 +0100 Subject: rxrpc: Calls displayed in /proc may in future lack a connection Allocated rxrpc calls displayed in /proc/net/rxrpc_calls may in future be on the proc list before they're connected or after they've been disconnected - in which case they may not have a pointer to a connection struct that can be used to get data from there. Deal with this by using stuff from the call struct in preference where possible and printing "no_connection" rather than a peer address if no connection is assigned. This change also has the added bonus that the service ID is now taken from the call rather the connection which will allow per-call service upgrades to be shown - something required for AuriStor server compatibility. Signed-off-by: David Howells diff --git a/net/rxrpc/proc.c b/net/rxrpc/proc.c index 9863270..500cdcd 100644 --- a/net/rxrpc/proc.c +++ b/net/rxrpc/proc.c @@ -59,25 +59,28 @@ static int rxrpc_call_seq_show(struct seq_file *seq, void *v) } call = list_entry(v, struct rxrpc_call, link); - conn = call->conn; sprintf(lbuff, "%pI4:%u", - &conn->params.local->srx.transport.sin.sin_addr, - ntohs(conn->params.local->srx.transport.sin.sin_port)); + &call->local->srx.transport.sin.sin_addr, + ntohs(call->local->srx.transport.sin.sin_port)); - sprintf(rbuff, "%pI4:%u", - &conn->params.peer->srx.transport.sin.sin_addr, - ntohs(conn->params.peer->srx.transport.sin.sin_port)); + conn = call->conn; + if (conn) + sprintf(rbuff, "%pI4:%u", + &conn->params.peer->srx.transport.sin.sin_addr, + ntohs(conn->params.peer->srx.transport.sin.sin_port)); + else + strcpy(rbuff, "no_connection"); seq_printf(seq, "UDP %-22.22s %-22.22s %4x %08x %08x %s %3u" " %-8.8s %08x %lx\n", lbuff, rbuff, - call->conn->params.service_id, + call->service_id, call->cid, call->call_id, - rxrpc_conn_is_service(call->conn) ? "Svc" : "Clt", + call->in_clientflag ? "Svc" : "Clt", atomic_read(&call->usage), rxrpc_call_states[call->state], call->remote_abort ?: call->local_abort, -- cgit v0.10.2 From 985a5c824a52e9f7cae59c850e2db98954f21c7c Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 17 Jun 2016 11:53:37 +0100 Subject: rxrpc: Make rxrpc_send_packet() take a connection not a transport Make rxrpc_send_packet() take a connection not a transport as part of the phasing out of the rxrpc_transport struct. Whilst we're at it, rename the function to rxrpc_send_data_packet() to differentiate it from the other packet sending functions. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 8996650..cfbd028 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -670,7 +670,7 @@ extern const char *rxrpc_acks(u8 reason); */ extern unsigned int rxrpc_resend_timeout; -int rxrpc_send_packet(struct rxrpc_transport *, struct sk_buff *); +int rxrpc_send_data_packet(struct rxrpc_connection *, struct sk_buff *); int rxrpc_do_sendmsg(struct rxrpc_sock *, struct msghdr *, size_t); /* diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c index b43faf5..0ba8429 100644 --- a/net/rxrpc/call_event.c +++ b/net/rxrpc/call_event.c @@ -187,7 +187,7 @@ static void rxrpc_resend(struct rxrpc_call *call) _proto("Tx DATA %%%u { #%d }", sp->hdr.serial, sp->hdr.seq); - if (rxrpc_send_packet(call->conn->trans, txb) < 0) { + if (rxrpc_send_data_packet(call->conn, txb) < 0) { stop = true; sp->resend_at = jiffies + 3; } else { diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index 6f8ab0e..db3933c 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -338,7 +338,7 @@ EXPORT_SYMBOL(rxrpc_kernel_abort_call); /* * send a packet through the transport endpoint */ -int rxrpc_send_packet(struct rxrpc_transport *trans, struct sk_buff *skb) +int rxrpc_send_data_packet(struct rxrpc_connection *conn, struct sk_buff *skb) { struct kvec iov[1]; struct msghdr msg; @@ -349,30 +349,30 @@ int rxrpc_send_packet(struct rxrpc_transport *trans, struct sk_buff *skb) iov[0].iov_base = skb->head; iov[0].iov_len = skb->len; - msg.msg_name = &trans->peer->srx.transport.sin; - msg.msg_namelen = sizeof(trans->peer->srx.transport.sin); + msg.msg_name = &conn->params.peer->srx.transport; + msg.msg_namelen = conn->params.peer->srx.transport_len; msg.msg_control = NULL; msg.msg_controllen = 0; msg.msg_flags = 0; /* send the packet with the don't fragment bit set if we currently * think it's small enough */ - if (skb->len - sizeof(struct rxrpc_wire_header) < trans->peer->maxdata) { - down_read(&trans->local->defrag_sem); + if (skb->len - sizeof(struct rxrpc_wire_header) < conn->params.peer->maxdata) { + down_read(&conn->params.local->defrag_sem); /* send the packet by UDP * - returns -EMSGSIZE if UDP would have to fragment the packet * to go out of the interface * - in which case, we'll have processed the ICMP error * message and update the peer record */ - ret = kernel_sendmsg(trans->local->socket, &msg, iov, 1, + ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 1, iov[0].iov_len); - up_read(&trans->local->defrag_sem); + up_read(&conn->params.local->defrag_sem); if (ret == -EMSGSIZE) goto send_fragmentable; - _leave(" = %d [%u]", ret, trans->peer->maxdata); + _leave(" = %d [%u]", ret, conn->params.peer->maxdata); return ret; } @@ -380,21 +380,28 @@ send_fragmentable: /* attempt to send this message with fragmentation enabled */ _debug("send fragment"); - down_write(&trans->local->defrag_sem); - opt = IP_PMTUDISC_DONT; - ret = kernel_setsockopt(trans->local->socket, SOL_IP, IP_MTU_DISCOVER, - (char *) &opt, sizeof(opt)); - if (ret == 0) { - ret = kernel_sendmsg(trans->local->socket, &msg, iov, 1, - iov[0].iov_len); - - opt = IP_PMTUDISC_DO; - kernel_setsockopt(trans->local->socket, SOL_IP, - IP_MTU_DISCOVER, (char *) &opt, sizeof(opt)); + down_write(&conn->params.local->defrag_sem); + + switch (conn->params.local->srx.transport.family) { + case AF_INET: + opt = IP_PMTUDISC_DONT; + ret = kernel_setsockopt(conn->params.local->socket, + SOL_IP, IP_MTU_DISCOVER, + (char *)&opt, sizeof(opt)); + if (ret == 0) { + ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 1, + iov[0].iov_len); + + opt = IP_PMTUDISC_DO; + kernel_setsockopt(conn->params.local->socket, SOL_IP, + IP_MTU_DISCOVER, + (char *)&opt, sizeof(opt)); + } + break; } - up_write(&trans->local->defrag_sem); - _leave(" = %d [frag %u]", ret, trans->peer->maxdata); + up_write(&conn->params.local->defrag_sem); + _leave(" = %d [frag %u]", ret, conn->params.peer->maxdata); return ret; } @@ -506,7 +513,7 @@ static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb, if (try_to_del_timer_sync(&call->ack_timer) >= 0) { /* the packet may be freed by rxrpc_process_call() before this * returns */ - ret = rxrpc_send_packet(call->conn->trans, skb); + ret = rxrpc_send_data_packet(call->conn, skb); _net("sent skb %p", skb); } else { _debug("failed to delete ACK timer"); -- cgit v0.10.2 From 5627cc8b961e4b07d5d649d9bd01ac929dcc1a95 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:38 +0100 Subject: rxrpc: Provide more refcount helper functions Provide refcount helper functions for connections so that the code doesn't touch local or connection usage counts directly. Also make it such that local and peer put functions can take a NULL pointer. Signed-off-by: David Howells diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index b29bb50..57dcbfc 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -674,11 +674,8 @@ static int rxrpc_release_sock(struct sock *sk) flush_workqueue(rxrpc_workqueue); rxrpc_purge_queue(&sk->sk_receive_queue); - if (rx->local) { - rxrpc_put_local(rx->local); - rx->local = NULL; - } - + rxrpc_put_local(rx->local); + rx->local = NULL; key_put(rx->key); rx->key = NULL; key_put(rx->securities); diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index cfbd028..c0ed5e7 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -597,6 +597,17 @@ static inline bool rxrpc_conn_is_service(const struct rxrpc_connection *conn) return conn->proto.in_clientflag; } +static inline void rxrpc_get_connection(struct rxrpc_connection *conn) +{ + atomic_inc(&conn->usage); +} + +static inline +struct rxrpc_connection *rxrpc_get_connection_maybe(struct rxrpc_connection *conn) +{ + return atomic_inc_not_zero(&conn->usage) ? conn : NULL; +} + /* * input.c */ @@ -645,7 +656,7 @@ struct rxrpc_local *rxrpc_get_local_maybe(struct rxrpc_local *local) static inline void rxrpc_put_local(struct rxrpc_local *local) { - if (atomic_dec_and_test(&local->usage)) + if (local && atomic_dec_and_test(&local->usage)) __rxrpc_put_local(local); } @@ -702,7 +713,7 @@ struct rxrpc_peer *rxrpc_get_peer_maybe(struct rxrpc_peer *peer) extern void __rxrpc_put_peer(struct rxrpc_peer *peer); static inline void rxrpc_put_peer(struct rxrpc_peer *peer) { - if (atomic_dec_and_test(&peer->usage)) + if (peer && atomic_dec_and_test(&peer->usage)) __rxrpc_put_peer(peer); } diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index 5a70dc4..833ad06 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -141,7 +141,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, _debug("await conn sec"); list_add_tail(&call->accept_link, &rx->secureq); call->conn->state = RXRPC_CONN_SERVER_CHALLENGING; - atomic_inc(&call->conn->usage); + rxrpc_get_connection(call->conn); set_bit(RXRPC_CONN_CHALLENGE, &call->conn->events); rxrpc_queue_conn(call->conn); } else { diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index d83f2cb..45849a6 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -515,7 +515,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, rb_insert_color(&call->conn_node, &conn->calls); conn->channels[call->channel] = call; sock_hold(&rx->sk); - atomic_inc(&conn->usage); + rxrpc_get_connection(conn); write_unlock_bh(&conn->lock); spin_lock(&conn->params.peer->lock); diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index a022439..bf69715 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -263,7 +263,7 @@ void rxrpc_process_connection(struct work_struct *work) _enter("{%d}", conn->debug_id); - atomic_inc(&conn->usage); + rxrpc_get_connection(conn); if (test_and_clear_bit(RXRPC_CONN_CHALLENGE, &conn->events)) { rxrpc_secure_connection(conn); diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 312b750..1754f2e 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -333,7 +333,7 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, * channel. */ chan = 0; - atomic_inc(&conn->usage); + rxrpc_get_connection(conn); conn->avail_calls = RXRPC_MAXCALLS - 1; conn->channels[chan] = call; conn->call_counter = 1; @@ -392,7 +392,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx, conn->channels[1] == NULL || conn->channels[2] == NULL || conn->channels[3] == NULL); - atomic_inc(&conn->usage); + rxrpc_get_connection(conn); break; } @@ -412,7 +412,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx, conn->channels[1] == NULL && conn->channels[2] == NULL && conn->channels[3] == NULL); - atomic_inc(&conn->usage); + rxrpc_get_connection(conn); list_move(&conn->bundle_link, &bundle->avail_conns); break; } @@ -629,7 +629,7 @@ found_extant_connection: read_unlock_bh(&trans->conn_lock); goto security_mismatch; } - atomic_inc(&conn->usage); + rxrpc_get_connection(conn); read_unlock_bh(&trans->conn_lock); goto success; @@ -639,7 +639,7 @@ found_extant_second: write_unlock_bh(&trans->conn_lock); goto security_mismatch; } - atomic_inc(&conn->usage); + rxrpc_get_connection(conn); write_unlock_bh(&trans->conn_lock); kfree(candidate); goto success; @@ -698,7 +698,7 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans, return NULL; found: - atomic_inc(&conn->usage); + rxrpc_get_connection(conn); read_unlock_bh(&trans->conn_lock); _leave(" = %p", conn); return conn; diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index cf540ef..799aec1 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -580,7 +580,7 @@ static void rxrpc_post_packet_to_conn(struct rxrpc_connection *conn, { _enter("%p,%p", conn, skb); - atomic_inc(&conn->usage); + rxrpc_get_connection(conn); skb_queue_tail(&conn->rx_queue, skb); rxrpc_queue_conn(conn); } diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c index 009b321..5703b0d 100644 --- a/net/rxrpc/local_object.c +++ b/net/rxrpc/local_object.c @@ -209,7 +209,7 @@ struct rxrpc_local *rxrpc_lookup_local(const struct sockaddr_rxrpc *srx) * bind the transport socket may still fail if we're attempting * to use a local address that the dying object is still using. */ - if (!atomic_inc_not_zero(&local->usage)) { + if (!rxrpc_get_local_maybe(local)) { cursor = cursor->next; list_del_init(&local->link); break; -- cgit v0.10.2 From 999b69f89241c9384c104b84329c13350fd696ef Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 17 Jun 2016 15:42:35 +0100 Subject: rxrpc: Kill the client connection bundle concept Kill off the concept of maintaining a bundle of connections to a particular target service to increase the number of call slots available for any beyond four for that service (there are four call slots per connection). This will make cleaning up the connection handling code easier and facilitate removal of the rxrpc_transport struct. Bundling can be reintroduced later if necessary. Signed-off-by: David Howells diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 57dcbfc..f3b6ed8 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -276,7 +276,6 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, gfp_t gfp) { struct rxrpc_conn_parameters cp; - struct rxrpc_conn_bundle *bundle; struct rxrpc_transport *trans; struct rxrpc_call *call; struct rxrpc_sock *rx = rxrpc_sk(sock->sk); @@ -311,15 +310,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, } cp.peer = trans->peer; - bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, gfp); - if (IS_ERR(bundle)) { - call = ERR_CAST(bundle); - goto out; - } - - call = rxrpc_new_client_call(rx, &cp, trans, bundle, user_call_ID, gfp); - rxrpc_put_bundle(trans, bundle); -out: + call = rxrpc_new_client_call(rx, &cp, trans, srx, user_call_ID, gfp); rxrpc_put_transport(trans); out_notrans: release_sock(&rx->sk); diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index c0ed5e7..26fe137 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -186,7 +186,8 @@ struct rxrpc_local { struct sk_buff_head accept_queue; /* incoming calls awaiting acceptance */ struct sk_buff_head reject_queue; /* packets awaiting rejection */ struct sk_buff_head event_queue; /* endpoint event packets awaiting processing */ - struct mutex conn_lock; /* Client connection creation lock */ + struct rb_root client_conns; /* Client connections by socket params */ + spinlock_t client_conns_lock; /* Lock for client_conns */ spinlock_t lock; /* access lock */ rwlock_t services_lock; /* lock for services list */ int debug_id; /* debug ID for printks */ @@ -232,35 +233,15 @@ struct rxrpc_peer { struct rxrpc_transport { struct rxrpc_local *local; /* local transport endpoint */ struct rxrpc_peer *peer; /* remote transport endpoint */ - struct rb_root bundles; /* client connection bundles on this transport */ struct rb_root server_conns; /* server connections on this transport */ struct list_head link; /* link in master session list */ unsigned long put_time; /* time at which to reap */ - spinlock_t client_lock; /* client connection allocation lock */ rwlock_t conn_lock; /* lock for active/dead connections */ atomic_t usage; int debug_id; /* debug ID for printks */ }; /* - * RxRPC client connection bundle - * - matched by { transport, service_id, key } - */ -struct rxrpc_conn_bundle { - struct rb_node node; /* node in transport's lookup tree */ - struct list_head unused_conns; /* unused connections in this bundle */ - struct list_head avail_conns; /* available connections in this bundle */ - struct list_head busy_conns; /* busy connections in this bundle */ - struct key *key; /* security for this bundle */ - wait_queue_head_t chanwait; /* wait for channel to become available */ - atomic_t usage; - int debug_id; /* debug ID for printks */ - unsigned short num_conns; /* number of connections in this bundle */ - u16 service_id; /* Service ID for this bundle */ - u8 security_ix; /* security type */ -}; - -/* * Keys for matching a connection. */ struct rxrpc_conn_proto { @@ -295,17 +276,21 @@ struct rxrpc_conn_parameters { */ struct rxrpc_connection { struct rxrpc_transport *trans; /* transport session */ - struct rxrpc_conn_bundle *bundle; /* connection bundle (client) */ struct rxrpc_conn_proto proto; struct rxrpc_conn_parameters params; + spinlock_t channel_lock; + struct rxrpc_call *channels[RXRPC_MAXCALLS]; /* active calls */ + wait_queue_head_t channel_wq; /* queue to wait for channel to become available */ + struct work_struct processor; /* connection event processor */ - struct rb_node node; /* node in transport's lookup tree */ + union { + struct rb_node client_node; /* Node in local->client_conns */ + struct rb_node service_node; /* Node in trans->server_conns */ + }; struct list_head link; /* link in master connection list */ - struct list_head bundle_link; /* link in bundle */ struct rb_root calls; /* calls on this connection */ struct sk_buff_head rx_queue; /* received conn-level packets */ - struct rxrpc_call *channels[RXRPC_MAXCALLS]; /* channels (active calls) */ const struct rxrpc_security *security; /* applied security module */ struct key *server_key; /* security for this service */ struct crypto_skcipher *cipher; /* encryption handle */ @@ -314,7 +299,7 @@ struct rxrpc_connection { #define RXRPC_CONN_HAS_IDR 0 /* - Has a client conn ID assigned */ unsigned long events; #define RXRPC_CONN_CHALLENGE 0 /* send challenge packet */ - unsigned long put_time; /* time at which to reap */ + unsigned long put_time; /* Time at which last put */ rwlock_t lock; /* access lock */ spinlock_t state_lock; /* state-change lock */ atomic_t usage; @@ -335,7 +320,7 @@ struct rxrpc_connection { unsigned int call_counter; /* call ID counter */ atomic_t serial; /* packet serial number counter */ atomic_t hi_serial; /* highest serial number received */ - u8 avail_calls; /* number of calls available */ + atomic_t avail_chans; /* number of channels available */ u8 size_align; /* data size alignment (for security) */ u8 header_size; /* rxrpc + security header size */ u8 security_size; /* security header size */ @@ -386,6 +371,8 @@ enum rxrpc_call_event { * The states that a call can be in. */ enum rxrpc_call_state { + RXRPC_CALL_UNINITIALISED, + RXRPC_CALL_CLIENT_AWAIT_CONN, /* - client waiting for connection to become available */ RXRPC_CALL_CLIENT_SEND_REQUEST, /* - client sending request phase */ RXRPC_CALL_CLIENT_AWAIT_REPLY, /* - client awaiting reply */ RXRPC_CALL_CLIENT_RECV_REPLY, /* - client receiving reply phase */ @@ -540,7 +527,7 @@ struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *, unsigned long struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *, struct rxrpc_conn_parameters *, struct rxrpc_transport *, - struct rxrpc_conn_bundle *, + struct sockaddr_rxrpc *, unsigned long, gfp_t); struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *, struct rxrpc_connection *, @@ -555,8 +542,7 @@ void __exit rxrpc_destroy_all_calls(void); */ extern struct idr rxrpc_client_conn_ids; -int rxrpc_get_client_connection_id(struct rxrpc_connection *, - struct rxrpc_transport *, gfp_t); +int rxrpc_get_client_connection_id(struct rxrpc_connection *, gfp_t); void rxrpc_put_client_connection_id(struct rxrpc_connection *); /* @@ -573,13 +559,10 @@ extern unsigned int rxrpc_connection_expiry; extern struct list_head rxrpc_connections; extern rwlock_t rxrpc_connection_lock; -struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *, - struct rxrpc_transport *, - struct key *, u16, gfp_t); -void rxrpc_put_bundle(struct rxrpc_transport *, struct rxrpc_conn_bundle *); -int rxrpc_connect_call(struct rxrpc_sock *, struct rxrpc_conn_parameters *, - struct rxrpc_transport *, struct rxrpc_conn_bundle *, - struct rxrpc_call *, gfp_t); +int rxrpc_connect_call(struct rxrpc_call *, struct rxrpc_conn_parameters *, + struct rxrpc_transport *, + struct sockaddr_rxrpc *, gfp_t); +void rxrpc_disconnect_call(struct rxrpc_call *); void rxrpc_put_connection(struct rxrpc_connection *); void __exit rxrpc_destroy_all_connections(void); struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *, diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 45849a6..9b3b48a 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -31,6 +31,8 @@ unsigned int rxrpc_max_call_lifetime = 60 * HZ; unsigned int rxrpc_dead_call_expiry = 2 * HZ; const char *const rxrpc_call_states[NR__RXRPC_CALL_STATES] = { + [RXRPC_CALL_UNINITIALISED] = "Uninit", + [RXRPC_CALL_CLIENT_AWAIT_CONN] = "ClWtConn", [RXRPC_CALL_CLIENT_SEND_REQUEST] = "ClSndReq", [RXRPC_CALL_CLIENT_AWAIT_REPLY] = "ClAwtRpl", [RXRPC_CALL_CLIENT_RECV_REPLY] = "ClRcvRpl", @@ -261,6 +263,7 @@ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp) (unsigned long) call); INIT_WORK(&call->destroyer, &rxrpc_destroy_call); INIT_WORK(&call->processor, &rxrpc_process_call); + INIT_LIST_HEAD(&call->link); INIT_LIST_HEAD(&call->accept_link); skb_queue_head_init(&call->rx_queue); skb_queue_head_init(&call->rx_oos_queue); @@ -269,7 +272,6 @@ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp) rwlock_init(&call->state_lock); atomic_set(&call->usage, 1); call->debug_id = atomic_inc_return(&rxrpc_debug_id); - call->state = RXRPC_CALL_CLIENT_SEND_REQUEST; memset(&call->sock_node, 0xed, sizeof(call->sock_node)); @@ -282,55 +284,70 @@ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp) } /* - * allocate a new client call and attempt to get a connection slot for it + * Allocate a new client call. */ static struct rxrpc_call *rxrpc_alloc_client_call( struct rxrpc_sock *rx, struct rxrpc_conn_parameters *cp, - struct rxrpc_transport *trans, - struct rxrpc_conn_bundle *bundle, + struct sockaddr_rxrpc *srx, gfp_t gfp) { struct rxrpc_call *call; - int ret; _enter(""); - ASSERT(rx != NULL); - ASSERT(trans != NULL); - ASSERT(bundle != NULL); + ASSERT(rx->local != NULL); call = rxrpc_alloc_call(gfp); if (!call) return ERR_PTR(-ENOMEM); + call->state = RXRPC_CALL_CLIENT_AWAIT_CONN; sock_hold(&rx->sk); call->socket = rx; call->rx_data_post = 1; - ret = rxrpc_connect_call(rx, cp, trans, bundle, call, gfp); - if (ret < 0) { - kmem_cache_free(rxrpc_call_jar, call); - return ERR_PTR(ret); - } - /* Record copies of information for hashtable lookup */ call->family = rx->family; - call->local = call->conn->params.local; + call->local = rx->local; switch (call->family) { case AF_INET: - call->peer_ip.ipv4_addr = - call->conn->params.peer->srx.transport.sin.sin_addr.s_addr; + call->peer_ip.ipv4_addr = srx->transport.sin.sin_addr.s_addr; break; case AF_INET6: memcpy(call->peer_ip.ipv6_addr, - call->conn->params.peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8, + srx->transport.sin6.sin6_addr.in6_u.u6_addr8, sizeof(call->peer_ip.ipv6_addr)); break; } - call->epoch = call->conn->proto.epoch; - call->service_id = call->conn->params.service_id; - call->in_clientflag = call->conn->proto.in_clientflag; + + call->service_id = srx->srx_service; + call->in_clientflag = 0; + + _leave(" = %p", call); + return call; +} + +/* + * Begin client call. + */ +static int rxrpc_begin_client_call(struct rxrpc_call *call, + struct rxrpc_conn_parameters *cp, + struct rxrpc_transport *trans, + struct sockaddr_rxrpc *srx, + gfp_t gfp) +{ + int ret; + + /* Set up or get a connection record and set the protocol parameters, + * including channel number and call ID. + */ + ret = rxrpc_connect_call(call, cp, trans, srx, gfp); + if (ret < 0) + return ret; + + call->state = RXRPC_CALL_CLIENT_SEND_REQUEST; + /* Add the new call to the hashtable */ rxrpc_call_hash_add(call); @@ -340,9 +357,7 @@ static struct rxrpc_call *rxrpc_alloc_client_call( call->lifetimer.expires = jiffies + rxrpc_max_call_lifetime; add_timer(&call->lifetimer); - - _leave(" = %p", call); - return call; + return 0; } /* @@ -352,23 +367,23 @@ static struct rxrpc_call *rxrpc_alloc_client_call( struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, struct rxrpc_conn_parameters *cp, struct rxrpc_transport *trans, - struct rxrpc_conn_bundle *bundle, + struct sockaddr_rxrpc *srx, unsigned long user_call_ID, gfp_t gfp) { struct rxrpc_call *call, *xcall; struct rb_node *parent, **pp; + int ret; - _enter("%p,%d,%d,%lx", - rx, trans->debug_id, bundle ? bundle->debug_id : -1, - user_call_ID); + _enter("%p,%lx", rx, user_call_ID); - call = rxrpc_alloc_client_call(rx, cp, trans, bundle, gfp); + call = rxrpc_alloc_client_call(rx, cp, srx, gfp); if (IS_ERR(call)) { _leave(" = %ld", PTR_ERR(call)); return call; } + /* Publish the call, even though it is incompletely set up as yet */ call->user_call_ID = user_call_ID; __set_bit(RXRPC_CALL_HAS_USERID, &call->flags); @@ -398,11 +413,29 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, list_add_tail(&call->link, &rxrpc_calls); write_unlock_bh(&rxrpc_call_lock); + ret = rxrpc_begin_client_call(call, cp, trans, srx, gfp); + if (ret < 0) + goto error; + _net("CALL new %d on CONN %d", call->debug_id, call->conn->debug_id); _leave(" = %p [new]", call); return call; +error: + write_lock(&rx->call_lock); + rb_erase(&call->sock_node, &rx->calls); + write_unlock(&rx->call_lock); + rxrpc_put_call(call); + + write_lock_bh(&rxrpc_call_lock); + list_del(&call->link); + write_unlock_bh(&rxrpc_call_lock); + + rxrpc_put_call(call); + _leave(" = %d", ret); + return ERR_PTR(ret); + /* We unexpectedly found the user ID in the list after taking * the call_lock. This shouldn't happen unless the user races * with itself and tries to add the same user ID twice at the @@ -612,40 +645,13 @@ void rxrpc_release_call(struct rxrpc_call *call) write_unlock_bh(&rx->call_lock); /* free up the channel for reuse */ - spin_lock(&conn->trans->client_lock); + spin_lock(&conn->channel_lock); write_lock_bh(&conn->lock); write_lock(&call->state_lock); - if (conn->channels[call->channel] == call) - conn->channels[call->channel] = NULL; - - if (conn->out_clientflag && conn->bundle) { - conn->avail_calls++; - switch (conn->avail_calls) { - case 1: - list_move_tail(&conn->bundle_link, - &conn->bundle->avail_conns); - case 2 ... RXRPC_MAXCALLS - 1: - ASSERT(conn->channels[0] == NULL || - conn->channels[1] == NULL || - conn->channels[2] == NULL || - conn->channels[3] == NULL); - break; - case RXRPC_MAXCALLS: - list_move_tail(&conn->bundle_link, - &conn->bundle->unused_conns); - ASSERT(conn->channels[0] == NULL && - conn->channels[1] == NULL && - conn->channels[2] == NULL && - conn->channels[3] == NULL); - break; - default: - pr_err("conn->avail_calls=%d\n", conn->avail_calls); - BUG(); - } - } + rxrpc_disconnect_call(call); - spin_unlock(&conn->trans->client_lock); + spin_unlock(&conn->channel_lock); if (call->state < RXRPC_CALL_COMPLETE && call->state != RXRPC_CALL_CLIENT_FINAL_ACK) { diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c index 2cccb4b..82488d6 100644 --- a/net/rxrpc/conn_client.c +++ b/net/rxrpc/conn_client.c @@ -33,9 +33,7 @@ static DEFINE_SPINLOCK(rxrpc_conn_id_lock); * client conns away from the current allocation point to try and keep the IDs * concentrated. We will also need to retire connections from an old epoch. */ -int rxrpc_get_client_connection_id(struct rxrpc_connection *conn, - struct rxrpc_transport *trans, - gfp_t gfp) +int rxrpc_get_client_connection_id(struct rxrpc_connection *conn, gfp_t gfp) { u32 epoch; int id; @@ -43,7 +41,6 @@ int rxrpc_get_client_connection_id(struct rxrpc_connection *conn, _enter(""); idr_preload(gfp); - write_lock_bh(&trans->conn_lock); spin_lock(&rxrpc_conn_id_lock); epoch = rxrpc_epoch; @@ -68,7 +65,6 @@ int rxrpc_get_client_connection_id(struct rxrpc_connection *conn, rxrpc_client_conn_ids.cur = id + 1; spin_unlock(&rxrpc_conn_id_lock); - write_unlock_bh(&trans->conn_lock); idr_preload_end(); conn->proto.epoch = epoch; @@ -79,7 +75,6 @@ int rxrpc_get_client_connection_id(struct rxrpc_connection *conn, error: spin_unlock(&rxrpc_conn_id_lock); - write_unlock_bh(&trans->conn_lock); idr_preload_end(); _leave(" = %d", id); return id; diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 1754f2e..276ff50 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -32,152 +32,6 @@ DEFINE_RWLOCK(rxrpc_connection_lock); static DECLARE_DELAYED_WORK(rxrpc_connection_reap, rxrpc_connection_reaper); /* - * allocate a new client connection bundle - */ -static struct rxrpc_conn_bundle *rxrpc_alloc_bundle(gfp_t gfp) -{ - struct rxrpc_conn_bundle *bundle; - - _enter(""); - - bundle = kzalloc(sizeof(struct rxrpc_conn_bundle), gfp); - if (bundle) { - INIT_LIST_HEAD(&bundle->unused_conns); - INIT_LIST_HEAD(&bundle->avail_conns); - INIT_LIST_HEAD(&bundle->busy_conns); - init_waitqueue_head(&bundle->chanwait); - atomic_set(&bundle->usage, 1); - } - - _leave(" = %p", bundle); - return bundle; -} - -/* - * compare bundle parameters with what we're looking for - * - return -ve, 0 or +ve - */ -static inline -int rxrpc_cmp_bundle(const struct rxrpc_conn_bundle *bundle, - struct key *key, u16 service_id) -{ - return (bundle->service_id - service_id) ?: - ((unsigned long)bundle->key - (unsigned long)key); -} - -/* - * get bundle of client connections that a client socket can make use of - */ -struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *rx, - struct rxrpc_transport *trans, - struct key *key, - u16 service_id, - gfp_t gfp) -{ - struct rxrpc_conn_bundle *bundle, *candidate; - struct rb_node *p, *parent, **pp; - - _enter("%p{%x},%x,%hx,", - rx, key_serial(key), trans->debug_id, service_id); - - /* search the extant bundles first for one that matches the specified - * user ID */ - spin_lock(&trans->client_lock); - - p = trans->bundles.rb_node; - while (p) { - bundle = rb_entry(p, struct rxrpc_conn_bundle, node); - - if (rxrpc_cmp_bundle(bundle, key, service_id) < 0) - p = p->rb_left; - else if (rxrpc_cmp_bundle(bundle, key, service_id) > 0) - p = p->rb_right; - else - goto found_extant_bundle; - } - - spin_unlock(&trans->client_lock); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_bundle(gfp); - if (!candidate) { - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } - - candidate->key = key_get(key); - candidate->service_id = service_id; - - spin_lock(&trans->client_lock); - - pp = &trans->bundles.rb_node; - parent = NULL; - while (*pp) { - parent = *pp; - bundle = rb_entry(parent, struct rxrpc_conn_bundle, node); - - if (rxrpc_cmp_bundle(bundle, key, service_id) < 0) - pp = &(*pp)->rb_left; - else if (rxrpc_cmp_bundle(bundle, key, service_id) > 0) - pp = &(*pp)->rb_right; - else - goto found_extant_second; - } - - /* second search also failed; add the new bundle */ - bundle = candidate; - candidate = NULL; - - rb_link_node(&bundle->node, parent, pp); - rb_insert_color(&bundle->node, &trans->bundles); - spin_unlock(&trans->client_lock); - _net("BUNDLE new on trans %d", trans->debug_id); - _leave(" = %p [new]", bundle); - return bundle; - - /* we found the bundle in the list immediately */ -found_extant_bundle: - atomic_inc(&bundle->usage); - spin_unlock(&trans->client_lock); - _net("BUNDLE old on trans %d", trans->debug_id); - _leave(" = %p [extant %d]", bundle, atomic_read(&bundle->usage)); - return bundle; - - /* we found the bundle on the second time through the list */ -found_extant_second: - atomic_inc(&bundle->usage); - spin_unlock(&trans->client_lock); - kfree(candidate); - _net("BUNDLE old2 on trans %d", trans->debug_id); - _leave(" = %p [second %d]", bundle, atomic_read(&bundle->usage)); - return bundle; -} - -/* - * release a bundle - */ -void rxrpc_put_bundle(struct rxrpc_transport *trans, - struct rxrpc_conn_bundle *bundle) -{ - _enter("%p,%p{%d}",trans, bundle, atomic_read(&bundle->usage)); - - if (atomic_dec_and_lock(&bundle->usage, &trans->client_lock)) { - _debug("Destroy bundle"); - rb_erase(&bundle->node, &trans->bundles); - spin_unlock(&trans->client_lock); - ASSERT(list_empty(&bundle->unused_conns)); - ASSERT(list_empty(&bundle->avail_conns)); - ASSERT(list_empty(&bundle->busy_conns)); - ASSERTCMP(bundle->num_conns, ==, 0); - key_put(bundle->key); - kfree(bundle); - } - - _leave(""); -} - -/* * allocate a new connection */ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) @@ -188,8 +42,10 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) conn = kzalloc(sizeof(struct rxrpc_connection), gfp); if (conn) { + spin_lock_init(&conn->channel_lock); + init_waitqueue_head(&conn->channel_wq); INIT_WORK(&conn->processor, &rxrpc_process_connection); - INIT_LIST_HEAD(&conn->bundle_link); + INIT_LIST_HEAD(&conn->link); conn->calls = RB_ROOT; skb_queue_head_init(&conn->rx_queue); conn->security = &rxrpc_no_security; @@ -197,7 +53,7 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) spin_lock_init(&conn->state_lock); atomic_set(&conn->usage, 1); conn->debug_id = atomic_inc_return(&rxrpc_debug_id); - conn->avail_calls = RXRPC_MAXCALLS; + atomic_set(&conn->avail_chans, RXRPC_MAXCALLS); conn->size_align = 4; conn->header_size = sizeof(struct rxrpc_wire_header); } @@ -240,7 +96,8 @@ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn, } /* - * Allocate a client connection. + * Allocate a client connection. The caller must take care to clear any + * padding bytes in *cp. */ static struct rxrpc_connection * rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, @@ -275,7 +132,7 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, break; } - ret = rxrpc_get_client_connection_id(conn, trans, gfp); + ret = rxrpc_get_client_connection_id(conn, gfp); if (ret < 0) goto error_0; @@ -290,6 +147,8 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, write_unlock(&rxrpc_connection_lock); key_get(conn->params.key); + conn->trans = trans; + atomic_inc(&trans->usage); _leave(" = %p", conn); return conn; @@ -303,217 +162,173 @@ error_0: } /* - * connect a call on an exclusive connection - */ -static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, - struct rxrpc_conn_parameters *cp, - struct rxrpc_transport *trans, - struct rxrpc_call *call, - gfp_t gfp) -{ - struct rxrpc_connection *conn; - int chan; - - _enter(""); - - conn = rxrpc_alloc_client_connection(cp, trans, gfp); - if (IS_ERR(conn)) { - _leave(" = %ld", PTR_ERR(conn)); - return PTR_ERR(conn); - } - - atomic_inc(&trans->usage); - conn->trans = trans; - conn->bundle = NULL; - - _net("CONNECT EXCL new %d on TRANS %d", - conn->debug_id, conn->trans->debug_id); - - /* Since no one else can use the connection, we just use the first - * channel. - */ - chan = 0; - rxrpc_get_connection(conn); - conn->avail_calls = RXRPC_MAXCALLS - 1; - conn->channels[chan] = call; - conn->call_counter = 1; - call->conn = conn; - call->channel = chan; - call->cid = conn->proto.cid | chan; - call->call_id = 1; - - _net("CONNECT client on conn %d chan %d as call %x", - conn->debug_id, chan, call->call_id); - - rxrpc_add_call_ID_to_conn(conn, call); - _leave(" = 0"); - return 0; -} - -/* * find a connection for a call * - called in process context with IRQs enabled */ -int rxrpc_connect_call(struct rxrpc_sock *rx, +int rxrpc_connect_call(struct rxrpc_call *call, struct rxrpc_conn_parameters *cp, struct rxrpc_transport *trans, - struct rxrpc_conn_bundle *bundle, - struct rxrpc_call *call, + struct sockaddr_rxrpc *srx, gfp_t gfp) { - struct rxrpc_connection *conn, *candidate; + struct rxrpc_connection *conn, *candidate = NULL; + struct rxrpc_local *local = cp->local; + struct rb_node *p, **pp, *parent; + long diff; int chan; DECLARE_WAITQUEUE(myself, current); - _enter("%p,%lx,", rx, call->user_call_ID); - - if (cp->exclusive) - return rxrpc_connect_exclusive(rx, cp, trans, call, gfp); - - spin_lock(&trans->client_lock); - for (;;) { - /* see if the bundle has a call slot available */ - if (!list_empty(&bundle->avail_conns)) { - _debug("avail"); - conn = list_entry(bundle->avail_conns.next, - struct rxrpc_connection, - bundle_link); - if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) { - list_del_init(&conn->bundle_link); - bundle->num_conns--; - continue; - } - if (--conn->avail_calls == 0) - list_move(&conn->bundle_link, - &bundle->busy_conns); - ASSERTCMP(conn->avail_calls, <, RXRPC_MAXCALLS); - ASSERT(conn->channels[0] == NULL || - conn->channels[1] == NULL || - conn->channels[2] == NULL || - conn->channels[3] == NULL); - rxrpc_get_connection(conn); - break; - } + _enter("{%d,%lx},", call->debug_id, call->user_call_ID); - if (!list_empty(&bundle->unused_conns)) { - _debug("unused"); - conn = list_entry(bundle->unused_conns.next, - struct rxrpc_connection, - bundle_link); - if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) { - list_del_init(&conn->bundle_link); - bundle->num_conns--; - continue; - } - ASSERTCMP(conn->avail_calls, ==, RXRPC_MAXCALLS); - conn->avail_calls = RXRPC_MAXCALLS - 1; - ASSERT(conn->channels[0] == NULL && - conn->channels[1] == NULL && - conn->channels[2] == NULL && - conn->channels[3] == NULL); - rxrpc_get_connection(conn); - list_move(&conn->bundle_link, &bundle->avail_conns); - break; + cp->peer = trans->peer; + rxrpc_get_peer(cp->peer); + + if (!cp->exclusive) { + /* Search for a existing client connection unless this is going + * to be a connection that's used exclusively for a single call. + */ + _debug("search 1"); + spin_lock(&local->client_conns_lock); + p = local->client_conns.rb_node; + while (p) { + conn = rb_entry(p, struct rxrpc_connection, client_node); + +#define cmp(X) ((long)conn->params.X - (long)cp->X) + diff = (cmp(peer) ?: + cmp(key) ?: + cmp(security_level)); + if (diff < 0) + p = p->rb_left; + else if (diff > 0) + p = p->rb_right; + else + goto found_extant_conn; } + spin_unlock(&local->client_conns_lock); + } - /* need to allocate a new connection */ - _debug("get new conn [%d]", bundle->num_conns); + /* We didn't find a connection or we want an exclusive one. */ + _debug("get new conn"); + candidate = rxrpc_alloc_client_connection(cp, trans, gfp); + if (!candidate) { + _leave(" = -ENOMEM"); + return -ENOMEM; + } - spin_unlock(&trans->client_lock); + if (cp->exclusive) { + /* Assign the call on an exclusive connection to channel 0 and + * don't add the connection to the endpoint's shareable conn + * lookup tree. + */ + _debug("exclusive chan 0"); + conn = candidate; + atomic_set(&conn->avail_chans, RXRPC_MAXCALLS - 1); + spin_lock(&conn->channel_lock); + chan = 0; + goto found_channel; + } - if (signal_pending(current)) - goto interrupted; + /* We need to redo the search before attempting to add a new connection + * lest we race with someone else adding a conflicting instance. + */ + _debug("search 2"); + spin_lock(&local->client_conns_lock); - if (bundle->num_conns >= 20) { - _debug("too many conns"); + pp = &local->client_conns.rb_node; + parent = NULL; + while (*pp) { + parent = *pp; + conn = rb_entry(parent, struct rxrpc_connection, client_node); - if (!gfpflags_allow_blocking(gfp)) { - _leave(" = -EAGAIN"); - return -EAGAIN; - } + diff = (cmp(peer) ?: + cmp(key) ?: + cmp(security_level)); + if (diff < 0) + pp = &(*pp)->rb_left; + else if (diff > 0) + pp = &(*pp)->rb_right; + else + goto found_extant_conn; + } - add_wait_queue(&bundle->chanwait, &myself); - for (;;) { - set_current_state(TASK_INTERRUPTIBLE); - if (bundle->num_conns < 20 || - !list_empty(&bundle->unused_conns) || - !list_empty(&bundle->avail_conns)) - break; - if (signal_pending(current)) - goto interrupted_dequeue; - schedule(); - } - remove_wait_queue(&bundle->chanwait, &myself); - __set_current_state(TASK_RUNNING); - spin_lock(&trans->client_lock); - continue; - } + /* The second search also failed; simply add the new connection with + * the new call in channel 0. Note that we need to take the channel + * lock before dropping the client conn lock. + */ + _debug("new conn"); + conn = candidate; + candidate = NULL; - /* not yet present - create a candidate for a new connection and then - * redo the check */ - candidate = rxrpc_alloc_client_connection(cp, trans, gfp); - if (!candidate) { - _leave(" = -ENOMEM"); - return -ENOMEM; - } + rb_link_node(&conn->client_node, parent, pp); + rb_insert_color(&conn->client_node, &local->client_conns); + + atomic_set(&conn->avail_chans, RXRPC_MAXCALLS - 1); + spin_lock(&conn->channel_lock); + spin_unlock(&local->client_conns_lock); + chan = 0; + +found_channel: + _debug("found chan"); + call->conn = conn; + call->channel = chan; + call->epoch = conn->proto.epoch; + call->cid = conn->proto.cid | chan; + call->call_id = ++conn->call_counter; + rcu_assign_pointer(conn->channels[chan], call); - atomic_inc(&bundle->usage); - atomic_inc(&trans->usage); - candidate->trans = trans; - candidate->bundle = bundle; + _net("CONNECT call %d on conn %d", call->debug_id, conn->debug_id); - spin_lock(&trans->client_lock); + rxrpc_add_call_ID_to_conn(conn, call); + spin_unlock(&conn->channel_lock); + _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); + return 0; + + /* We found a suitable connection already in existence. Discard any + * candidate we may have allocated, and try to get a channel on this + * one. + */ +found_extant_conn: + _debug("found conn"); + rxrpc_get_connection(conn); + spin_unlock(&local->client_conns_lock); - list_add(&candidate->bundle_link, &bundle->unused_conns); - bundle->num_conns++; + rxrpc_put_connection(candidate); - _net("CONNECT new %d on TRANS %d", - candidate->debug_id, candidate->trans->debug_id); + if (!atomic_add_unless(&conn->avail_chans, -1, 0)) { + if (!gfpflags_allow_blocking(gfp)) { + rxrpc_put_connection(conn); + _leave(" = -EAGAIN"); + return -EAGAIN; + } - /* leave the candidate lurking in zombie mode attached to the - * bundle until we're ready for it */ - rxrpc_put_connection(candidate); - candidate = NULL; + add_wait_queue(&conn->channel_wq, &myself); + for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + if (atomic_add_unless(&conn->avail_chans, -1, 0)) + break; + if (signal_pending(current)) + goto interrupted; + schedule(); + } + remove_wait_queue(&conn->channel_wq, &myself); + __set_current_state(TASK_RUNNING); } - /* we've got a connection with a free channel and we can now attach the - * call to it - * - we're holding the transport's client lock - * - we're holding a reference on the connection - * - we're holding a reference on the bundle + /* The connection allegedly now has a free channel and we can now + * attach the call to it. */ + spin_lock(&conn->channel_lock); + for (chan = 0; chan < RXRPC_MAXCALLS; chan++) if (!conn->channels[chan]) goto found_channel; - ASSERT(conn->channels[0] == NULL || - conn->channels[1] == NULL || - conn->channels[2] == NULL || - conn->channels[3] == NULL); BUG(); -found_channel: - conn->channels[chan] = call; - call->conn = conn; - call->channel = chan; - call->cid = conn->proto.cid | chan; - call->call_id = ++conn->call_counter; - - _net("CONNECT client on conn %d chan %d as call %x", - conn->debug_id, chan, call->call_id); - - ASSERTCMP(conn->avail_calls, <, RXRPC_MAXCALLS); - spin_unlock(&trans->client_lock); - - rxrpc_add_call_ID_to_conn(conn, call); - - _leave(" = 0"); - return 0; - -interrupted_dequeue: - remove_wait_queue(&bundle->chanwait, &myself); - __set_current_state(TASK_RUNNING); interrupted: + remove_wait_queue(&conn->channel_wq, &myself); + __set_current_state(TASK_RUNNING); + rxrpc_put_connection(conn); _leave(" = -ERESTARTSYS"); return -ERESTARTSYS; } @@ -521,8 +336,8 @@ interrupted: /* * get a record of an incoming connection */ -struct rxrpc_connection * -rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb) +struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans, + struct sk_buff *skb) { struct rxrpc_connection *conn, *candidate = NULL; struct rxrpc_skb_priv *sp = rxrpc_skb(skb); @@ -543,7 +358,7 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb) p = trans->server_conns.rb_node; while (p) { - conn = rb_entry(p, struct rxrpc_connection, node); + conn = rb_entry(p, struct rxrpc_connection, service_node); _debug("maybe %x", conn->proto.cid); @@ -588,7 +403,7 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb) p = NULL; while (*pp) { p = *pp; - conn = rb_entry(p, struct rxrpc_connection, node); + conn = rb_entry(p, struct rxrpc_connection, service_node); if (epoch < conn->proto.epoch) pp = &(*pp)->rb_left; @@ -605,8 +420,8 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb) /* we can now add the new candidate to the list */ conn = candidate; candidate = NULL; - rb_link_node(&conn->node, p, pp); - rb_insert_color(&conn->node, &trans->server_conns); + rb_link_node(&conn->service_node, p, pp); + rb_insert_color(&conn->service_node, &trans->server_conns); atomic_inc(&conn->trans->usage); write_unlock_bh(&trans->conn_lock); @@ -672,7 +487,7 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans, if (sp->hdr.flags & RXRPC_CLIENT_INITIATED) { p = trans->server_conns.rb_node; while (p) { - conn = rb_entry(p, struct rxrpc_connection, node); + conn = rb_entry(p, struct rxrpc_connection, service_node); _debug("maybe %x", conn->proto.cid); @@ -705,10 +520,31 @@ found: } /* + * Disconnect a call and clear any channel it occupies when that call + * terminates. + */ +void rxrpc_disconnect_call(struct rxrpc_call *call) +{ + struct rxrpc_connection *conn = call->conn; + unsigned chan = call->channel; + + _enter("%d,%d", conn->debug_id, call->channel); + + if (conn->channels[chan] == call) { + rcu_assign_pointer(conn->channels[chan], NULL); + atomic_inc(&conn->avail_chans); + wake_up(&conn->channel_wq); + } +} + +/* * release a virtual connection */ void rxrpc_put_connection(struct rxrpc_connection *conn) { + if (!conn) + return; + _enter("%p{u=%d,d=%d}", conn, atomic_read(&conn->usage), conn->debug_id); @@ -734,9 +570,6 @@ static void rxrpc_destroy_connection(struct rxrpc_connection *conn) _net("DESTROY CONN %d", conn->debug_id); - if (conn->bundle) - rxrpc_put_bundle(conn->trans, conn->bundle); - ASSERT(RB_EMPTY_ROOT(&conn->calls)); rxrpc_purge_queue(&conn->rx_queue); @@ -773,30 +606,39 @@ static void rxrpc_connection_reaper(struct work_struct *work) if (likely(atomic_read(&conn->usage) > 0)) continue; - spin_lock(&conn->trans->client_lock); - write_lock_bh(&conn->trans->conn_lock); - reap_time = conn->put_time + rxrpc_connection_expiry; + if (rxrpc_conn_is_client(conn)) { + struct rxrpc_local *local = conn->params.local; + spin_lock(&local->client_conns_lock); + reap_time = conn->put_time + rxrpc_connection_expiry; - if (atomic_read(&conn->usage) > 0) { - ; - } else if (reap_time <= now) { - list_move_tail(&conn->link, &graveyard); - if (conn->out_clientflag) + if (atomic_read(&conn->usage) > 0) { + ; + } else if (reap_time <= now) { + list_move_tail(&conn->link, &graveyard); rxrpc_put_client_connection_id(conn); - else - rb_erase(&conn->node, + rb_erase(&conn->client_node, + &local->client_conns); + } else if (reap_time < earliest) { + earliest = reap_time; + } + + spin_unlock(&local->client_conns_lock); + } else { + write_lock_bh(&conn->trans->conn_lock); + reap_time = conn->put_time + rxrpc_connection_expiry; + + if (atomic_read(&conn->usage) > 0) { + ; + } else if (reap_time <= now) { + list_move_tail(&conn->link, &graveyard); + rb_erase(&conn->service_node, &conn->trans->server_conns); - if (conn->bundle) { - list_del_init(&conn->bundle_link); - conn->bundle->num_conns--; + } else if (reap_time < earliest) { + earliest = reap_time; } - } else if (reap_time < earliest) { - earliest = reap_time; + write_unlock_bh(&conn->trans->conn_lock); } - - write_unlock_bh(&conn->trans->conn_lock); - spin_unlock(&conn->trans->client_lock); } write_unlock(&rxrpc_connection_lock); diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c index 5703b0d..3ab7764 100644 --- a/net/rxrpc/local_object.c +++ b/net/rxrpc/local_object.c @@ -80,7 +80,8 @@ static struct rxrpc_local *rxrpc_alloc_local(const struct sockaddr_rxrpc *srx) skb_queue_head_init(&local->accept_queue); skb_queue_head_init(&local->reject_queue); skb_queue_head_init(&local->event_queue); - mutex_init(&local->conn_lock); + local->client_conns = RB_ROOT; + spin_lock_init(&local->client_conns_lock); spin_lock_init(&local->lock); rwlock_init(&local->services_lock); local->debug_id = atomic_inc_return(&rxrpc_debug_id); @@ -294,6 +295,7 @@ static void rxrpc_local_destroyer(struct rxrpc_local *local) list_del_init(&local->link); mutex_unlock(&rxrpc_local_mutex); + ASSERT(RB_EMPTY_ROOT(&local->client_conns)); ASSERT(list_empty(&local->services)); if (socket) { diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index db3933c..8e24939 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -140,7 +140,6 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, unsigned long user_call_ID, bool exclusive) { struct rxrpc_conn_parameters cp; - struct rxrpc_conn_bundle *bundle; struct rxrpc_transport *trans; struct rxrpc_call *call; struct key *key; @@ -171,16 +170,8 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, } cp.peer = trans->peer; - bundle = rxrpc_get_bundle(rx, trans, cp.key, srx->srx_service, - GFP_KERNEL); - if (IS_ERR(bundle)) { - ret = PTR_ERR(bundle); - goto out_trans; - } - - call = rxrpc_new_client_call(rx, &cp, trans, bundle, user_call_ID, + call = rxrpc_new_client_call(rx, &cp, trans, srx, user_call_ID, GFP_KERNEL); - rxrpc_put_bundle(trans, bundle); rxrpc_put_transport(trans); if (IS_ERR(call)) { ret = PTR_ERR(call); diff --git a/net/rxrpc/transport.c b/net/rxrpc/transport.c index 140628d..7194740 100644 --- a/net/rxrpc/transport.c +++ b/net/rxrpc/transport.c @@ -46,9 +46,7 @@ static struct rxrpc_transport *rxrpc_alloc_transport(struct rxrpc_local *local, trans->local = local; trans->peer = peer; INIT_LIST_HEAD(&trans->link); - trans->bundles = RB_ROOT; trans->server_conns = RB_ROOT; - spin_lock_init(&trans->client_lock); rwlock_init(&trans->conn_lock); atomic_set(&trans->usage, 1); trans->debug_id = atomic_inc_return(&rxrpc_debug_id); -- cgit v0.10.2 From aa390bbe2113dd0de99cf35c39d7701d4412b744 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 17 Jun 2016 10:06:56 +0100 Subject: rxrpc: Kill off the rxrpc_transport struct The rxrpc_transport struct is now redundant, given that the rxrpc_peer struct is now per peer port rather than per peer host, so get rid of it. Service connection lists are transferred to the rxrpc_peer struct, as is the conn_lock. Previous patches moved the client connection handling out of the rxrpc_transport struct and discarded the connection bundling code. Signed-off-by: David Howells diff --git a/net/rxrpc/Makefile b/net/rxrpc/Makefile index cfa2215..6522e50 100644 --- a/net/rxrpc/Makefile +++ b/net/rxrpc/Makefile @@ -22,7 +22,6 @@ af-rxrpc-y := \ recvmsg.o \ security.o \ skbuff.o \ - transport.o \ utils.o af-rxrpc-$(CONFIG_PROC_FS) += proc.o diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index f3b6ed8..5d3e795 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -224,37 +224,6 @@ static int rxrpc_listen(struct socket *sock, int backlog) return ret; } -/* - * find a transport by address - */ -struct rxrpc_transport * -rxrpc_name_to_transport(struct rxrpc_conn_parameters *cp, - struct sockaddr *addr, - int addr_len, - gfp_t gfp) -{ - struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *) addr; - struct rxrpc_transport *trans; - - _enter("%p,%d", addr, addr_len); - - if (cp->local->srx.transport_type != srx->transport_type) - return ERR_PTR(-ESOCKTNOSUPPORT); - if (cp->local->srx.transport.family != srx->transport.family) - return ERR_PTR(-EAFNOSUPPORT); - - /* find a remote transport endpoint from the local one */ - cp->peer = rxrpc_lookup_peer(cp->local, srx, gfp); - if (!cp->peer) - return ERR_PTR(-ENOMEM); - - /* find a transport */ - trans = rxrpc_get_transport(cp->local, cp->peer, gfp); - rxrpc_put_peer(cp->peer); - _leave(" = %p", trans); - return trans; -} - /** * rxrpc_kernel_begin_call - Allow a kernel service to begin a call * @sock: The socket on which to make the call @@ -276,7 +245,6 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, gfp_t gfp) { struct rxrpc_conn_parameters cp; - struct rxrpc_transport *trans; struct rxrpc_call *call; struct rxrpc_sock *rx = rxrpc_sk(sock->sk); int ret; @@ -300,19 +268,8 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, cp.security_level = 0; cp.exclusive = false; cp.service_id = srx->srx_service; + call = rxrpc_new_client_call(rx, &cp, srx, user_call_ID, gfp); - trans = rxrpc_name_to_transport(&cp, (struct sockaddr *)srx, - sizeof(*srx), gfp); - if (IS_ERR(trans)) { - call = ERR_CAST(trans); - trans = NULL; - goto out_notrans; - } - cp.peer = trans->peer; - - call = rxrpc_new_client_call(rx, &cp, trans, srx, user_call_ID, gfp); - rxrpc_put_transport(trans); -out_notrans: release_sock(&rx->sk); _leave(" = %p", call); return call; @@ -831,7 +788,6 @@ static void __exit af_rxrpc_exit(void) proto_unregister(&rxrpc_proto); rxrpc_destroy_all_calls(); rxrpc_destroy_all_connections(); - rxrpc_destroy_all_transports(); ASSERTCMP(atomic_read(&rxrpc_n_skbs), ==, 0); diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 26fe137..702db72 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -207,6 +207,8 @@ struct rxrpc_peer { struct rxrpc_local *local; struct hlist_head error_targets; /* targets for net error distribution */ struct work_struct error_distributor; + struct rb_root service_conns; /* Service connections */ + rwlock_t conn_lock; spinlock_t lock; /* access lock */ unsigned int if_mtu; /* interface MTU for this peer */ unsigned int mtu; /* network MTU for this peer */ @@ -226,22 +228,6 @@ struct rxrpc_peer { }; /* - * RxRPC point-to-point transport / connection manager definition - * - handles a bundle of connections between two endpoints - * - matched by { local, peer } - */ -struct rxrpc_transport { - struct rxrpc_local *local; /* local transport endpoint */ - struct rxrpc_peer *peer; /* remote transport endpoint */ - struct rb_root server_conns; /* server connections on this transport */ - struct list_head link; /* link in master session list */ - unsigned long put_time; /* time at which to reap */ - rwlock_t conn_lock; /* lock for active/dead connections */ - atomic_t usage; - int debug_id; /* debug ID for printks */ -}; - -/* * Keys for matching a connection. */ struct rxrpc_conn_proto { @@ -271,11 +257,10 @@ struct rxrpc_conn_parameters { /* * RxRPC connection definition - * - matched by { transport, service_id, conn_id, direction, key } + * - matched by { local, peer, epoch, conn_id, direction } * - each connection can only handle four simultaneous calls */ struct rxrpc_connection { - struct rxrpc_transport *trans; /* transport session */ struct rxrpc_conn_proto proto; struct rxrpc_conn_parameters params; @@ -286,7 +271,7 @@ struct rxrpc_connection { struct work_struct processor; /* connection event processor */ union { struct rb_node client_node; /* Node in local->client_conns */ - struct rb_node service_node; /* Node in trans->server_conns */ + struct rb_node service_node; /* Node in peer->service_conns */ }; struct list_head link; /* link in master connection list */ struct rb_root calls; /* calls on this connection */ @@ -494,10 +479,6 @@ extern u32 rxrpc_epoch; extern atomic_t rxrpc_debug_id; extern struct workqueue_struct *rxrpc_workqueue; -extern struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_conn_parameters *, - struct sockaddr *, - int, gfp_t); - /* * call_accept.c */ @@ -526,7 +507,6 @@ struct rxrpc_call *rxrpc_find_call_hash(struct rxrpc_host_header *, struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *, unsigned long); struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *, struct rxrpc_conn_parameters *, - struct rxrpc_transport *, struct sockaddr_rxrpc *, unsigned long, gfp_t); struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *, @@ -560,15 +540,16 @@ extern struct list_head rxrpc_connections; extern rwlock_t rxrpc_connection_lock; int rxrpc_connect_call(struct rxrpc_call *, struct rxrpc_conn_parameters *, - struct rxrpc_transport *, struct sockaddr_rxrpc *, gfp_t); +struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *, + struct rxrpc_peer *, + struct sk_buff *); void rxrpc_disconnect_call(struct rxrpc_call *); void rxrpc_put_connection(struct rxrpc_connection *); void __exit rxrpc_destroy_all_connections(void); -struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *, - struct sk_buff *); -extern struct rxrpc_connection * -rxrpc_incoming_connection(struct rxrpc_transport *, struct sk_buff *); +struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *, + struct rxrpc_peer *, + struct sk_buff *); static inline bool rxrpc_conn_is_client(const struct rxrpc_connection *conn) { @@ -585,12 +566,6 @@ static inline void rxrpc_get_connection(struct rxrpc_connection *conn) atomic_inc(&conn->usage); } -static inline -struct rxrpc_connection *rxrpc_get_connection_maybe(struct rxrpc_connection *conn) -{ - return atomic_inc_not_zero(&conn->usage) ? conn : NULL; -} - /* * input.c */ @@ -745,18 +720,6 @@ static inline void rxrpc_sysctl_exit(void) {} #endif /* - * transport.c - */ -extern unsigned int rxrpc_transport_expiry; - -struct rxrpc_transport *rxrpc_get_transport(struct rxrpc_local *, - struct rxrpc_peer *, gfp_t); -void rxrpc_put_transport(struct rxrpc_transport *); -void __exit rxrpc_destroy_all_transports(void); -struct rxrpc_transport *rxrpc_find_transport(struct rxrpc_local *, - struct rxrpc_peer *); - -/* * utils.c */ void rxrpc_get_addr_from_skb(struct rxrpc_local *, const struct sk_buff *, diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index 833ad06..202e053 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -74,7 +74,6 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, struct sockaddr_rxrpc *srx) { struct rxrpc_connection *conn; - struct rxrpc_transport *trans; struct rxrpc_skb_priv *sp, *nsp; struct rxrpc_peer *peer; struct rxrpc_call *call; @@ -102,16 +101,8 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, goto error; } - trans = rxrpc_get_transport(local, peer, GFP_NOIO); + conn = rxrpc_incoming_connection(local, peer, skb); rxrpc_put_peer(peer); - if (IS_ERR(trans)) { - _debug("no trans"); - ret = -EBUSY; - goto error; - } - - conn = rxrpc_incoming_connection(trans, skb); - rxrpc_put_transport(trans); if (IS_ERR(conn)) { _debug("no conn"); ret = PTR_ERR(conn); diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 9b3b48a..ad933da 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -286,11 +286,9 @@ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp) /* * Allocate a new client call. */ -static struct rxrpc_call *rxrpc_alloc_client_call( - struct rxrpc_sock *rx, - struct rxrpc_conn_parameters *cp, - struct sockaddr_rxrpc *srx, - gfp_t gfp) +static struct rxrpc_call *rxrpc_alloc_client_call(struct rxrpc_sock *rx, + struct sockaddr_rxrpc *srx, + gfp_t gfp) { struct rxrpc_call *call; @@ -333,7 +331,6 @@ static struct rxrpc_call *rxrpc_alloc_client_call( */ static int rxrpc_begin_client_call(struct rxrpc_call *call, struct rxrpc_conn_parameters *cp, - struct rxrpc_transport *trans, struct sockaddr_rxrpc *srx, gfp_t gfp) { @@ -342,7 +339,7 @@ static int rxrpc_begin_client_call(struct rxrpc_call *call, /* Set up or get a connection record and set the protocol parameters, * including channel number and call ID. */ - ret = rxrpc_connect_call(call, cp, trans, srx, gfp); + ret = rxrpc_connect_call(call, cp, srx, gfp); if (ret < 0) return ret; @@ -366,7 +363,6 @@ static int rxrpc_begin_client_call(struct rxrpc_call *call, */ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, struct rxrpc_conn_parameters *cp, - struct rxrpc_transport *trans, struct sockaddr_rxrpc *srx, unsigned long user_call_ID, gfp_t gfp) @@ -377,7 +373,7 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, _enter("%p,%lx", rx, user_call_ID); - call = rxrpc_alloc_client_call(rx, cp, srx, gfp); + call = rxrpc_alloc_client_call(rx, srx, gfp); if (IS_ERR(call)) { _leave(" = %ld", PTR_ERR(call)); return call; @@ -413,7 +409,7 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, list_add_tail(&call->link, &rxrpc_calls); write_unlock_bh(&rxrpc_call_lock); - ret = rxrpc_begin_client_call(call, cp, trans, srx, gfp); + ret = rxrpc_begin_client_call(call, cp, srx, gfp); if (ret < 0) goto error; diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 276ff50..4bfad7c 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -100,9 +100,7 @@ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn, * padding bytes in *cp. */ static struct rxrpc_connection * -rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, - struct rxrpc_transport *trans, - gfp_t gfp) +rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, gfp_t gfp) { struct rxrpc_connection *conn; int ret; @@ -146,9 +144,10 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, list_add_tail(&conn->link, &rxrpc_connections); write_unlock(&rxrpc_connection_lock); + /* We steal the caller's peer ref. */ + cp->peer = NULL; + rxrpc_get_local(conn->params.local); key_get(conn->params.key); - conn->trans = trans; - atomic_inc(&trans->usage); _leave(" = %p", conn); return conn; @@ -167,7 +166,6 @@ error_0: */ int rxrpc_connect_call(struct rxrpc_call *call, struct rxrpc_conn_parameters *cp, - struct rxrpc_transport *trans, struct sockaddr_rxrpc *srx, gfp_t gfp) { @@ -181,8 +179,9 @@ int rxrpc_connect_call(struct rxrpc_call *call, _enter("{%d,%lx},", call->debug_id, call->user_call_ID); - cp->peer = trans->peer; - rxrpc_get_peer(cp->peer); + cp->peer = rxrpc_lookup_peer(cp->local, srx, gfp); + if (!cp->peer) + return -ENOMEM; if (!cp->exclusive) { /* Search for a existing client connection unless this is going @@ -210,7 +209,7 @@ int rxrpc_connect_call(struct rxrpc_call *call, /* We didn't find a connection or we want an exclusive one. */ _debug("get new conn"); - candidate = rxrpc_alloc_client_connection(cp, trans, gfp); + candidate = rxrpc_alloc_client_connection(cp, gfp); if (!candidate) { _leave(" = -ENOMEM"); return -ENOMEM; @@ -281,6 +280,8 @@ found_channel: rxrpc_add_call_ID_to_conn(conn, call); spin_unlock(&conn->channel_lock); + rxrpc_put_peer(cp->peer); + cp->peer = NULL; _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); return 0; @@ -329,6 +330,8 @@ interrupted: remove_wait_queue(&conn->channel_wq, &myself); __set_current_state(TASK_RUNNING); rxrpc_put_connection(conn); + rxrpc_put_peer(cp->peer); + cp->peer = NULL; _leave(" = -ERESTARTSYS"); return -ERESTARTSYS; } @@ -336,7 +339,8 @@ interrupted: /* * get a record of an incoming connection */ -struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans, +struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, + struct rxrpc_peer *peer, struct sk_buff *skb) { struct rxrpc_connection *conn, *candidate = NULL; @@ -354,9 +358,9 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans cid = sp->hdr.cid & RXRPC_CIDMASK; /* search the connection list first */ - read_lock_bh(&trans->conn_lock); + read_lock_bh(&peer->conn_lock); - p = trans->server_conns.rb_node; + p = peer->service_conns.rb_node; while (p) { conn = rb_entry(p, struct rxrpc_connection, service_node); @@ -373,7 +377,7 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans else goto found_extant_connection; } - read_unlock_bh(&trans->conn_lock); + read_unlock_bh(&peer->conn_lock); /* not yet present - create a candidate for a new record and then * redo the search */ @@ -383,13 +387,12 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans return ERR_PTR(-ENOMEM); } - candidate->trans = trans; - candidate->proto.local = trans->local; + candidate->proto.local = local; candidate->proto.epoch = sp->hdr.epoch; candidate->proto.cid = sp->hdr.cid & RXRPC_CIDMASK; candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED; - candidate->params.local = trans->local; - candidate->params.peer = trans->peer; + candidate->params.local = local; + candidate->params.peer = peer; candidate->params.service_id = sp->hdr.serviceId; candidate->security_ix = sp->hdr.securityIndex; candidate->out_clientflag = 0; @@ -397,9 +400,9 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans if (candidate->params.service_id) candidate->state = RXRPC_CONN_SERVER_UNSECURED; - write_lock_bh(&trans->conn_lock); + write_lock_bh(&peer->conn_lock); - pp = &trans->server_conns.rb_node; + pp = &peer->service_conns.rb_node; p = NULL; while (*pp) { p = *pp; @@ -421,10 +424,11 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans conn = candidate; candidate = NULL; rb_link_node(&conn->service_node, p, pp); - rb_insert_color(&conn->service_node, &trans->server_conns); - atomic_inc(&conn->trans->usage); + rb_insert_color(&conn->service_node, &peer->service_conns); + rxrpc_get_peer(peer); + rxrpc_get_local(local); - write_unlock_bh(&trans->conn_lock); + write_unlock_bh(&peer->conn_lock); write_lock(&rxrpc_connection_lock); list_add_tail(&conn->link, &rxrpc_connections); @@ -441,21 +445,21 @@ success: /* we found the connection in the list immediately */ found_extant_connection: if (sp->hdr.securityIndex != conn->security_ix) { - read_unlock_bh(&trans->conn_lock); + read_unlock_bh(&peer->conn_lock); goto security_mismatch; } rxrpc_get_connection(conn); - read_unlock_bh(&trans->conn_lock); + read_unlock_bh(&peer->conn_lock); goto success; /* we found the connection on the second time through the list */ found_extant_second: if (sp->hdr.securityIndex != conn->security_ix) { - write_unlock_bh(&trans->conn_lock); + write_unlock_bh(&peer->conn_lock); goto security_mismatch; } rxrpc_get_connection(conn); - write_unlock_bh(&trans->conn_lock); + write_unlock_bh(&peer->conn_lock); kfree(candidate); goto success; @@ -469,7 +473,8 @@ security_mismatch: * find a connection based on transport and RxRPC connection ID for an incoming * packet */ -struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans, +struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *local, + struct rxrpc_peer *peer, struct sk_buff *skb) { struct rxrpc_connection *conn; @@ -479,13 +484,13 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans, _enter(",{%x,%x}", sp->hdr.cid, sp->hdr.flags); - read_lock_bh(&trans->conn_lock); + read_lock_bh(&peer->conn_lock); cid = sp->hdr.cid & RXRPC_CIDMASK; epoch = sp->hdr.epoch; if (sp->hdr.flags & RXRPC_CLIENT_INITIATED) { - p = trans->server_conns.rb_node; + p = peer->service_conns.rb_node; while (p) { conn = rb_entry(p, struct rxrpc_connection, service_node); @@ -508,13 +513,13 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans, goto found; } - read_unlock_bh(&trans->conn_lock); + read_unlock_bh(&peer->conn_lock); _leave(" = NULL"); return NULL; found: rxrpc_get_connection(conn); - read_unlock_bh(&trans->conn_lock); + read_unlock_bh(&peer->conn_lock); _leave(" = %p", conn); return conn; } @@ -576,8 +581,9 @@ static void rxrpc_destroy_connection(struct rxrpc_connection *conn) conn->security->clear(conn); key_put(conn->params.key); key_put(conn->server_key); + rxrpc_put_peer(conn->params.peer); + rxrpc_put_local(conn->params.local); - rxrpc_put_transport(conn->trans); kfree(conn); _leave(""); } @@ -588,6 +594,7 @@ static void rxrpc_destroy_connection(struct rxrpc_connection *conn) static void rxrpc_connection_reaper(struct work_struct *work) { struct rxrpc_connection *conn, *_p; + struct rxrpc_peer *peer; unsigned long now, earliest, reap_time; LIST_HEAD(graveyard); @@ -624,7 +631,8 @@ static void rxrpc_connection_reaper(struct work_struct *work) spin_unlock(&local->client_conns_lock); } else { - write_lock_bh(&conn->trans->conn_lock); + peer = conn->params.peer; + write_lock_bh(&peer->conn_lock); reap_time = conn->put_time + rxrpc_connection_expiry; if (atomic_read(&conn->usage) > 0) { @@ -632,12 +640,12 @@ static void rxrpc_connection_reaper(struct work_struct *work) } else if (reap_time <= now) { list_move_tail(&conn->link, &graveyard); rb_erase(&conn->service_node, - &conn->trans->server_conns); + &peer->service_conns); } else if (reap_time < earliest) { earliest = reap_time; } - write_unlock_bh(&conn->trans->conn_lock); + write_unlock_bh(&peer->conn_lock); } } write_unlock(&rxrpc_connection_lock); diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 799aec1..f4bd57b 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -631,7 +631,6 @@ static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local, struct sk_buff *skb) { struct rxrpc_peer *peer; - struct rxrpc_transport *trans; struct rxrpc_connection *conn; struct sockaddr_rxrpc srx; @@ -641,13 +640,8 @@ static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local, if (!peer) goto cant_find_peer; - trans = rxrpc_find_transport(local, peer); + conn = rxrpc_find_connection(local, peer, skb); rcu_read_unlock(); - if (!trans) - goto cant_find_conn; - - conn = rxrpc_find_connection(trans, skb); - rxrpc_put_transport(trans); if (!conn) goto cant_find_conn; diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index 8e24939..f4bda06 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -140,10 +140,8 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, unsigned long user_call_ID, bool exclusive) { struct rxrpc_conn_parameters cp; - struct rxrpc_transport *trans; struct rxrpc_call *call; struct key *key; - long ret; DECLARE_SOCKADDR(struct sockaddr_rxrpc *, srx, msg->msg_name); @@ -162,30 +160,10 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, cp.security_level = rx->min_sec_level; cp.exclusive = rx->exclusive | exclusive; cp.service_id = srx->srx_service; - trans = rxrpc_name_to_transport(&cp, msg->msg_name, msg->msg_namelen, - GFP_KERNEL); - if (IS_ERR(trans)) { - ret = PTR_ERR(trans); - goto out; - } - cp.peer = trans->peer; - - call = rxrpc_new_client_call(rx, &cp, trans, srx, user_call_ID, - GFP_KERNEL); - rxrpc_put_transport(trans); - if (IS_ERR(call)) { - ret = PTR_ERR(call); - goto out_trans; - } + call = rxrpc_new_client_call(rx, &cp, srx, user_call_ID, GFP_KERNEL); _leave(" = %p\n", call); return call; - -out_trans: - rxrpc_put_transport(trans); -out: - _leave(" = %ld", ret); - return ERR_PTR(ret); } /* diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c index 6baad70..01d4930 100644 --- a/net/rxrpc/peer_object.c +++ b/net/rxrpc/peer_object.c @@ -188,6 +188,8 @@ struct rxrpc_peer *rxrpc_alloc_peer(struct rxrpc_local *local, gfp_t gfp) INIT_HLIST_HEAD(&peer->error_targets); INIT_WORK(&peer->error_distributor, &rxrpc_peer_error_distributor); + peer->service_conns = RB_ROOT; + rwlock_init(&peer->conn_lock); spin_lock_init(&peer->lock); peer->debug_id = atomic_inc_return(&rxrpc_debug_id); } diff --git a/net/rxrpc/sysctl.c b/net/rxrpc/sysctl.c index a99690a..03ad087 100644 --- a/net/rxrpc/sysctl.c +++ b/net/rxrpc/sysctl.c @@ -90,14 +90,6 @@ static struct ctl_table rxrpc_sysctl_table[] = { .proc_handler = proc_dointvec_minmax, .extra1 = (void *)&one, }, - { - .procname = "transport_expiry", - .data = &rxrpc_transport_expiry, - .maxlen = sizeof(unsigned int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = (void *)&one, - }, /* Non-time values */ { diff --git a/net/rxrpc/transport.c b/net/rxrpc/transport.c deleted file mode 100644 index 7194740..0000000 --- a/net/rxrpc/transport.c +++ /dev/null @@ -1,265 +0,0 @@ -/* RxRPC point-to-point transport session management - * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include -#include "ar-internal.h" - -/* - * Time after last use at which transport record is cleaned up. - */ -unsigned int rxrpc_transport_expiry = 3600 * 24; - -static void rxrpc_transport_reaper(struct work_struct *work); - -static LIST_HEAD(rxrpc_transports); -static DEFINE_RWLOCK(rxrpc_transport_lock); -static DECLARE_DELAYED_WORK(rxrpc_transport_reap, rxrpc_transport_reaper); - -/* - * allocate a new transport session manager - */ -static struct rxrpc_transport *rxrpc_alloc_transport(struct rxrpc_local *local, - struct rxrpc_peer *peer, - gfp_t gfp) -{ - struct rxrpc_transport *trans; - - _enter(""); - - trans = kzalloc(sizeof(struct rxrpc_transport), gfp); - if (trans) { - trans->local = local; - trans->peer = peer; - INIT_LIST_HEAD(&trans->link); - trans->server_conns = RB_ROOT; - rwlock_init(&trans->conn_lock); - atomic_set(&trans->usage, 1); - trans->debug_id = atomic_inc_return(&rxrpc_debug_id); - } - - _leave(" = %p", trans); - return trans; -} - -/* - * obtain a transport session for the nominated endpoints - */ -struct rxrpc_transport *rxrpc_get_transport(struct rxrpc_local *local, - struct rxrpc_peer *peer, - gfp_t gfp) -{ - struct rxrpc_transport *trans, *candidate; - const char *new = "old"; - int usage; - - _enter("{%pI4+%hu},{%pI4+%hu},", - &local->srx.transport.sin.sin_addr, - ntohs(local->srx.transport.sin.sin_port), - &peer->srx.transport.sin.sin_addr, - ntohs(peer->srx.transport.sin.sin_port)); - - /* search the transport list first */ - read_lock_bh(&rxrpc_transport_lock); - list_for_each_entry(trans, &rxrpc_transports, link) { - if (trans->local == local && trans->peer == peer) - goto found_extant_transport; - } - read_unlock_bh(&rxrpc_transport_lock); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_transport(local, peer, gfp); - if (!candidate) { - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } - - write_lock_bh(&rxrpc_transport_lock); - - list_for_each_entry(trans, &rxrpc_transports, link) { - if (trans->local == local && trans->peer == peer) - goto found_extant_second; - } - - /* we can now add the new candidate to the list */ - trans = candidate; - candidate = NULL; - usage = atomic_read(&trans->usage); - - rxrpc_get_local(trans->local); - rxrpc_get_peer(trans->peer); - list_add_tail(&trans->link, &rxrpc_transports); - write_unlock_bh(&rxrpc_transport_lock); - new = "new"; - -success: - _net("TRANSPORT %s %d local %d -> peer %d", - new, - trans->debug_id, - trans->local->debug_id, - trans->peer->debug_id); - - _leave(" = %p {u=%d}", trans, usage); - return trans; - - /* we found the transport in the list immediately */ -found_extant_transport: - usage = atomic_inc_return(&trans->usage); - read_unlock_bh(&rxrpc_transport_lock); - goto success; - - /* we found the transport on the second time through the list */ -found_extant_second: - usage = atomic_inc_return(&trans->usage); - write_unlock_bh(&rxrpc_transport_lock); - kfree(candidate); - goto success; -} - -/* - * find the transport connecting two endpoints - */ -struct rxrpc_transport *rxrpc_find_transport(struct rxrpc_local *local, - struct rxrpc_peer *peer) -{ - struct rxrpc_transport *trans; - - _enter("{%pI4+%hu},{%pI4+%hu},", - &local->srx.transport.sin.sin_addr, - ntohs(local->srx.transport.sin.sin_port), - &peer->srx.transport.sin.sin_addr, - ntohs(peer->srx.transport.sin.sin_port)); - - /* search the transport list */ - read_lock_bh(&rxrpc_transport_lock); - - list_for_each_entry(trans, &rxrpc_transports, link) { - if (trans->local == local && trans->peer == peer) - goto found_extant_transport; - } - - read_unlock_bh(&rxrpc_transport_lock); - _leave(" = NULL"); - return NULL; - -found_extant_transport: - atomic_inc(&trans->usage); - read_unlock_bh(&rxrpc_transport_lock); - _leave(" = %p", trans); - return trans; -} - -/* - * release a transport session - */ -void rxrpc_put_transport(struct rxrpc_transport *trans) -{ - _enter("%p{u=%d}", trans, atomic_read(&trans->usage)); - - ASSERTCMP(atomic_read(&trans->usage), >, 0); - - trans->put_time = ktime_get_seconds(); - if (unlikely(atomic_dec_and_test(&trans->usage))) { - _debug("zombie"); - /* let the reaper determine the timeout to avoid a race with - * overextending the timeout if the reaper is running at the - * same time */ - rxrpc_queue_delayed_work(&rxrpc_transport_reap, 0); - } - _leave(""); -} - -/* - * clean up a transport session - */ -static void rxrpc_cleanup_transport(struct rxrpc_transport *trans) -{ - _net("DESTROY TRANS %d", trans->debug_id); - - rxrpc_put_local(trans->local); - rxrpc_put_peer(trans->peer); - kfree(trans); -} - -/* - * reap dead transports that have passed their expiry date - */ -static void rxrpc_transport_reaper(struct work_struct *work) -{ - struct rxrpc_transport *trans, *_p; - unsigned long now, earliest, reap_time; - - LIST_HEAD(graveyard); - - _enter(""); - - now = ktime_get_seconds(); - earliest = ULONG_MAX; - - /* extract all the transports that have been dead too long */ - write_lock_bh(&rxrpc_transport_lock); - list_for_each_entry_safe(trans, _p, &rxrpc_transports, link) { - _debug("reap TRANS %d { u=%d t=%ld }", - trans->debug_id, atomic_read(&trans->usage), - (long) now - (long) trans->put_time); - - if (likely(atomic_read(&trans->usage) > 0)) - continue; - - reap_time = trans->put_time + rxrpc_transport_expiry; - if (reap_time <= now) - list_move_tail(&trans->link, &graveyard); - else if (reap_time < earliest) - earliest = reap_time; - } - write_unlock_bh(&rxrpc_transport_lock); - - if (earliest != ULONG_MAX) { - _debug("reschedule reaper %ld", (long) earliest - now); - ASSERTCMP(earliest, >, now); - rxrpc_queue_delayed_work(&rxrpc_transport_reap, - (earliest - now) * HZ); - } - - /* then destroy all those pulled out */ - while (!list_empty(&graveyard)) { - trans = list_entry(graveyard.next, struct rxrpc_transport, - link); - list_del_init(&trans->link); - - ASSERTCMP(atomic_read(&trans->usage), ==, 0); - rxrpc_cleanup_transport(trans); - } - - _leave(""); -} - -/* - * preemptively destroy all the transport session records rather than waiting - * for them to time out - */ -void __exit rxrpc_destroy_all_transports(void) -{ - _enter(""); - - rxrpc_transport_expiry = 0; - cancel_delayed_work(&rxrpc_transport_reap); - rxrpc_queue_delayed_work(&rxrpc_transport_reap, 0); - - _leave(""); -} -- cgit v0.10.2 From 5427ccafe61a95b945730a23c3f7645a96a24fd7 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 18 Jun 2016 15:15:39 +0200 Subject: net: ethernet: sun4i-emac: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c index de2c4bf..6e31cb6 100644 --- a/drivers/net/ethernet/allwinner/sun4i-emac.c +++ b/drivers/net/ethernet/allwinner/sun4i-emac.c @@ -77,7 +77,6 @@ struct emac_board_info { int emacrx_completed_flag; - struct phy_device *phy_dev; struct device_node *phy_node; unsigned int link; unsigned int speed; @@ -115,7 +114,7 @@ static void emac_update_duplex(struct net_device *dev) static void emac_handle_link_change(struct net_device *dev) { struct emac_board_info *db = netdev_priv(dev); - struct phy_device *phydev = db->phy_dev; + struct phy_device *phydev = dev->phydev; unsigned long flags; int status_change = 0; @@ -154,21 +153,22 @@ static void emac_handle_link_change(struct net_device *dev) static int emac_mdio_probe(struct net_device *dev) { struct emac_board_info *db = netdev_priv(dev); + struct phy_device *phydev; /* to-do: PHY interrupts are currently not supported */ /* attach the mac to the phy */ - db->phy_dev = of_phy_connect(db->ndev, db->phy_node, - &emac_handle_link_change, 0, - db->phy_interface); - if (!db->phy_dev) { + phydev = of_phy_connect(db->ndev, db->phy_node, + &emac_handle_link_change, 0, + db->phy_interface); + if (!phydev) { netdev_err(db->ndev, "could not find the PHY\n"); return -ENODEV; } /* mask with MAC supported features */ - db->phy_dev->supported &= PHY_BASIC_FEATURES; - db->phy_dev->advertising = db->phy_dev->supported; + phydev->supported &= PHY_BASIC_FEATURES; + phydev->advertising = phydev->supported; db->link = 0; db->speed = 0; @@ -179,10 +179,7 @@ static int emac_mdio_probe(struct net_device *dev) static void emac_mdio_remove(struct net_device *dev) { - struct emac_board_info *db = netdev_priv(dev); - - phy_disconnect(db->phy_dev); - db->phy_dev = NULL; + phy_disconnect(dev->phydev); } static void emac_reset(struct emac_board_info *db) @@ -208,8 +205,7 @@ static void emac_inblk_32bit(void __iomem *reg, void *data, int count) static int emac_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct emac_board_info *dm = netdev_priv(dev); - struct phy_device *phydev = dm->phy_dev; + struct phy_device *phydev = dev->phydev; if (!netif_running(dev)) return -EINVAL; @@ -231,8 +227,7 @@ static void emac_get_drvinfo(struct net_device *dev, static int emac_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct emac_board_info *dm = netdev_priv(dev); - struct phy_device *phydev = dm->phy_dev; + struct phy_device *phydev = dev->phydev; if (!phydev) return -ENODEV; @@ -242,8 +237,7 @@ static int emac_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) static int emac_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct emac_board_info *dm = netdev_priv(dev); - struct phy_device *phydev = dm->phy_dev; + struct phy_device *phydev = dev->phydev; if (!phydev) return -ENODEV; @@ -744,7 +738,7 @@ static int emac_open(struct net_device *dev) return ret; } - phy_start(db->phy_dev); + phy_start(dev->phydev); netif_start_queue(dev); return 0; @@ -781,7 +775,7 @@ static int emac_stop(struct net_device *ndev) netif_stop_queue(ndev); netif_carrier_off(ndev); - phy_stop(db->phy_dev); + phy_stop(ndev->phydev); emac_mdio_remove(ndev); -- cgit v0.10.2 From 91cd3b4499bfbf6d5a412b4f9e3da9d260a1412f Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 18 Jun 2016 15:15:40 +0200 Subject: net: ethernet: sun4i-emac: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c index 6e31cb6..6ffdff6 100644 --- a/drivers/net/ethernet/allwinner/sun4i-emac.c +++ b/drivers/net/ethernet/allwinner/sun4i-emac.c @@ -225,31 +225,11 @@ static void emac_get_drvinfo(struct net_device *dev, strlcpy(info->bus_info, dev_name(&dev->dev), sizeof(info->bus_info)); } -static int emac_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = dev->phydev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_gset(phydev, cmd); -} - -static int emac_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = dev->phydev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_sset(phydev, cmd); -} - static const struct ethtool_ops emac_ethtool_ops = { .get_drvinfo = emac_get_drvinfo, - .get_settings = emac_get_settings, - .set_settings = emac_set_settings, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static unsigned int emac_setup(struct net_device *ndev) -- cgit v0.10.2 From 941ea69e176bc5cc8a359c6b608250fc950be0b1 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 18 Jun 2016 16:37:20 +0200 Subject: net: ethernet: altera_tse: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/altera/altera_tse.h b/drivers/net/ethernet/altera/altera_tse.h index 103c30d..e005200 100644 --- a/drivers/net/ethernet/altera/altera_tse.h +++ b/drivers/net/ethernet/altera/altera_tse.h @@ -473,7 +473,6 @@ struct altera_tse_private { int phy_addr; /* PHY's MDIO address, -1 for autodetection */ phy_interface_t phy_iface; struct mii_bus *mdio; - struct phy_device *phydev; int oldspeed; int oldduplex; int oldlink; diff --git a/drivers/net/ethernet/altera/altera_tse_ethtool.c b/drivers/net/ethernet/altera/altera_tse_ethtool.c index be72e1e..c1b5608 100644 --- a/drivers/net/ethernet/altera/altera_tse_ethtool.c +++ b/drivers/net/ethernet/altera/altera_tse_ethtool.c @@ -235,8 +235,7 @@ static void tse_get_regs(struct net_device *dev, struct ethtool_regs *regs, static int tse_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct altera_tse_private *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; if (phydev == NULL) return -ENODEV; @@ -246,8 +245,7 @@ static int tse_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) static int tse_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct altera_tse_private *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; if (phydev == NULL) return -ENODEV; diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index f749e4d..49025e9 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -625,7 +625,7 @@ out: static void altera_tse_adjust_link(struct net_device *dev) { struct altera_tse_private *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; int new_state = 0; /* only change config if there is a link */ @@ -845,7 +845,6 @@ static int init_phy(struct net_device *dev) netdev_dbg(dev, "attached to PHY %d UID 0x%08x Link = %d\n", phydev->mdio.addr, phydev->phy_id, phydev->link); - priv->phydev = phydev; return 0; } @@ -1172,8 +1171,8 @@ static int tse_open(struct net_device *dev) spin_unlock_irqrestore(&priv->rxdma_irq_lock, flags); - if (priv->phydev) - phy_start(priv->phydev); + if (dev->phydev) + phy_start(dev->phydev); napi_enable(&priv->napi); netif_start_queue(dev); @@ -1205,8 +1204,8 @@ static int tse_shutdown(struct net_device *dev) unsigned long int flags; /* Stop the PHY */ - if (priv->phydev) - phy_stop(priv->phydev); + if (dev->phydev) + phy_stop(dev->phydev); netif_stop_queue(dev); napi_disable(&priv->napi); @@ -1545,10 +1544,9 @@ err_free_netdev: static int altera_tse_remove(struct platform_device *pdev) { struct net_device *ndev = platform_get_drvdata(pdev); - struct altera_tse_private *priv = netdev_priv(ndev); - if (priv->phydev) - phy_disconnect(priv->phydev); + if (ndev->phydev) + phy_disconnect(ndev->phydev); platform_set_drvdata(pdev, NULL); altera_tse_mdio_destroy(ndev); -- cgit v0.10.2 From 11bbb171b921dc777113a2383c945d66db229eaf Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 18 Jun 2016 16:37:21 +0200 Subject: net: ethernet: altera_tse: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/altera/altera_tse_ethtool.c b/drivers/net/ethernet/altera/altera_tse_ethtool.c index c1b5608..7c36771 100644 --- a/drivers/net/ethernet/altera/altera_tse_ethtool.c +++ b/drivers/net/ethernet/altera/altera_tse_ethtool.c @@ -233,38 +233,18 @@ static void tse_get_regs(struct net_device *dev, struct ethtool_regs *regs, buf[i] = csrrd32(priv->mac_dev, i * 4); } -static int tse_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = dev->phydev; - - if (phydev == NULL) - return -ENODEV; - - return phy_ethtool_gset(phydev, cmd); -} - -static int tse_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = dev->phydev; - - if (phydev == NULL) - return -ENODEV; - - return phy_ethtool_sset(phydev, cmd); -} - static const struct ethtool_ops tse_ethtool_ops = { .get_drvinfo = tse_get_drvinfo, .get_regs_len = tse_reglen, .get_regs = tse_get_regs, .get_link = ethtool_op_get_link, - .get_settings = tse_get_settings, - .set_settings = tse_set_settings, .get_strings = tse_gstrings, .get_sset_count = tse_sset_count, .get_ethtool_stats = tse_fill_stats, .get_msglevel = tse_get_msglevel, .set_msglevel = tse_set_msglevel, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; void altera_tse_set_ethtool_ops(struct net_device *netdev) -- cgit v0.10.2 From b21fcb259313bcf7d4f73ecd5e44948995c8957c Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 19 Jun 2016 22:37:05 +0200 Subject: net: ethernet: bgmac: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index 0ee34cc..90ed244 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -1320,7 +1320,7 @@ static int bgmac_open(struct net_device *net_dev) } napi_enable(&bgmac->napi); - phy_start(bgmac->phy_dev); + phy_start(net_dev->phydev); netif_carrier_on(net_dev); return 0; @@ -1332,7 +1332,7 @@ static int bgmac_stop(struct net_device *net_dev) netif_carrier_off(net_dev); - phy_stop(bgmac->phy_dev); + phy_stop(net_dev->phydev); napi_disable(&bgmac->napi); bgmac_chip_intrs_off(bgmac); @@ -1370,12 +1370,10 @@ static int bgmac_set_mac_address(struct net_device *net_dev, void *addr) static int bgmac_ioctl(struct net_device *net_dev, struct ifreq *ifr, int cmd) { - struct bgmac *bgmac = netdev_priv(net_dev); - if (!netif_running(net_dev)) return -EINVAL; - return phy_mii_ioctl(bgmac->phy_dev, ifr, cmd); + return phy_mii_ioctl(net_dev->phydev, ifr, cmd); } static const struct net_device_ops bgmac_netdev_ops = { @@ -1518,7 +1516,7 @@ static int bgmac_get_settings(struct net_device *net_dev, { struct bgmac *bgmac = netdev_priv(net_dev); - return phy_ethtool_gset(bgmac->phy_dev, cmd); + return phy_ethtool_gset(net_dev->phydev, cmd); } static int bgmac_set_settings(struct net_device *net_dev, @@ -1526,7 +1524,7 @@ static int bgmac_set_settings(struct net_device *net_dev, { struct bgmac *bgmac = netdev_priv(net_dev); - return phy_ethtool_sset(bgmac->phy_dev, cmd); + return phy_ethtool_sset(net_dev->phydev, cmd); } static void bgmac_get_drvinfo(struct net_device *net_dev, @@ -1563,7 +1561,7 @@ static int bgmac_mii_write(struct mii_bus *bus, int mii_id, int regnum, static void bgmac_adjust_link(struct net_device *net_dev) { struct bgmac *bgmac = netdev_priv(net_dev); - struct phy_device *phy_dev = bgmac->phy_dev; + struct phy_device *phy_dev = net_dev->phydev; bool update = false; if (phy_dev->link) { @@ -1607,8 +1605,6 @@ static int bgmac_fixed_phy_register(struct bgmac *bgmac) return err; } - bgmac->phy_dev = phy_dev; - return err; } @@ -1653,7 +1649,6 @@ static int bgmac_mii_register(struct bgmac *bgmac) err = PTR_ERR(phy_dev); goto err_unregister_bus; } - bgmac->phy_dev = phy_dev; return err; diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h index 853d72b..99beb18 100644 --- a/drivers/net/ethernet/broadcom/bgmac.h +++ b/drivers/net/ethernet/broadcom/bgmac.h @@ -441,7 +441,6 @@ struct bgmac { struct net_device *net_dev; struct napi_struct napi; struct mii_bus *mii_bus; - struct phy_device *phy_dev; /* DMA */ struct bgmac_dma_ring tx_ring[BGMAC_MAX_TX_RINGS]; -- cgit v0.10.2 From 904632a224b2ea39920e526cdc93565aaf1fb43e Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 19 Jun 2016 22:37:06 +0200 Subject: net: ethernet: bgmac: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index 90ed244..e6e74ca 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -1511,22 +1511,6 @@ static void bgmac_get_ethtool_stats(struct net_device *dev, } } -static int bgmac_get_settings(struct net_device *net_dev, - struct ethtool_cmd *cmd) -{ - struct bgmac *bgmac = netdev_priv(net_dev); - - return phy_ethtool_gset(net_dev->phydev, cmd); -} - -static int bgmac_set_settings(struct net_device *net_dev, - struct ethtool_cmd *cmd) -{ - struct bgmac *bgmac = netdev_priv(net_dev); - - return phy_ethtool_sset(net_dev->phydev, cmd); -} - static void bgmac_get_drvinfo(struct net_device *net_dev, struct ethtool_drvinfo *info) { @@ -1538,9 +1522,9 @@ static const struct ethtool_ops bgmac_ethtool_ops = { .get_strings = bgmac_get_strings, .get_sset_count = bgmac_get_sset_count, .get_ethtool_stats = bgmac_get_ethtool_stats, - .get_settings = bgmac_get_settings, - .set_settings = bgmac_set_settings, .get_drvinfo = bgmac_get_drvinfo, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /************************************************** -- cgit v0.10.2 From b95e5928fcc76d156352570858abdea7b2628efd Mon Sep 17 00:00:00 2001 From: William Tu Date: Mon, 20 Jun 2016 07:26:17 -0700 Subject: openvswitch: Add packet len info to upcall. The commit f2a4d086ed4c ("openvswitch: Add packet truncation support.") introduces packet truncation before sending to userspace upcall receiver. This patch passes up the skb->len before truncation so that the upcall receiver knows the original packet size. Potentially this will be used by sFlow, where OVS translates sFlow config header=N to a sample action, truncating packet to N byte in kernel datapath. Thus, only N bytes instead of full-packet size is copied from kernel to userspace, saving the kernel-to-userspace bandwidth. Signed-off-by: William Tu Cc: Pravin Shelar Acked-by: Pravin B Shelar Signed-off-by: David S. Miller diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h index 8274675..d95a301 100644 --- a/include/uapi/linux/openvswitch.h +++ b/include/uapi/linux/openvswitch.h @@ -166,6 +166,7 @@ enum ovs_packet_cmd { * output port is actually a tunnel port. Contains the output tunnel key * extracted from the packet as nested %OVS_TUNNEL_KEY_ATTR_* attributes. * @OVS_PACKET_ATTR_MRU: Present for an %OVS_PACKET_CMD_ACTION and + * @OVS_PACKET_ATTR_LEN: Packet size before truncation. * %OVS_PACKET_ATTR_USERSPACE action specify the Maximum received fragment * size. * @@ -185,6 +186,7 @@ enum ovs_packet_attr { OVS_PACKET_ATTR_PROBE, /* Packet operation is a feature probe, error logging should be suppressed. */ OVS_PACKET_ATTR_MRU, /* Maximum received IP fragment size. */ + OVS_PACKET_ATTR_LEN, /* Packet size before truncation. */ __OVS_PACKET_ATTR_MAX }; diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 6739342..524c0fd 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -387,7 +387,8 @@ static size_t upcall_msg_size(const struct dp_upcall_info *upcall_info, { size_t size = NLMSG_ALIGN(sizeof(struct ovs_header)) + nla_total_size(hdrlen) /* OVS_PACKET_ATTR_PACKET */ - + nla_total_size(ovs_key_attr_size()); /* OVS_PACKET_ATTR_KEY */ + + nla_total_size(ovs_key_attr_size()) /* OVS_PACKET_ATTR_KEY */ + + nla_total_size(sizeof(unsigned int)); /* OVS_PACKET_ATTR_LEN */ /* OVS_PACKET_ATTR_USERDATA */ if (upcall_info->userdata) @@ -514,6 +515,16 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, pad_packet(dp, user_skb); } + /* Add OVS_PACKET_ATTR_LEN when packet is truncated */ + if (cutlen > 0) { + if (nla_put_u32(user_skb, OVS_PACKET_ATTR_LEN, + skb->len)) { + err = -ENOBUFS; + goto out; + } + pad_packet(dp, user_skb); + } + /* Only reserve room for attribute header, packet data is added * in skb_zerocopy() */ if (!(nla = nla_reserve(user_skb, OVS_PACKET_ATTR_PACKET, 0))) { -- cgit v0.10.2 From 2781ff5c8fc7722e97503f96686bf6d7093069a9 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 20 Jun 2016 17:51:52 +0200 Subject: can: only call can_stat_update with procfs The change to leave out procfs support in CAN when CONFIG_PROC_FS is not set was incomplete and leads to a build error: net/built-in.o: In function `can_init': :(.init.text+0x9858): undefined reference to `can_stat_update' ERROR: "can_stat_update" [net/can/can.ko] undefined! This tries a better approach, encapsulating all of the calls within IS_ENABLED(), so we also leave out the timer function from the object file. Signed-off-by: Arnd Bergmann Fixes: a20fadf85312 ("can: build proc support only if CONFIG_PROC_FS is activated") Signed-off-by: Marc Kleine-Budde diff --git a/net/can/af_can.c b/net/can/af_can.c index 166d436..1108079 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -911,14 +911,14 @@ static __init int can_init(void) if (!rcv_cache) return -ENOMEM; - if (stats_timer) { + if (IS_ENABLED(CONFIG_PROC_FS)) { + if (stats_timer) { /* the statistics are updated every second (timer triggered) */ - setup_timer(&can_stattimer, can_stat_update, 0); - mod_timer(&can_stattimer, round_jiffies(jiffies + HZ)); - } else - can_stattimer.function = NULL; - - can_init_proc(); + setup_timer(&can_stattimer, can_stat_update, 0); + mod_timer(&can_stattimer, round_jiffies(jiffies + HZ)); + } + can_init_proc(); + } /* protocol register */ sock_register(&can_family_ops); @@ -933,10 +933,12 @@ static __exit void can_exit(void) { struct net_device *dev; - if (stats_timer) - del_timer_sync(&can_stattimer); + if (IS_ENABLED(CONFIG_PROC_FS)) { + if (stats_timer) + del_timer_sync(&can_stattimer); - can_remove_proc(); + can_remove_proc(); + } /* protocol unregister */ dev_remove_pack(&canfd_packet); diff --git a/net/can/af_can.h b/net/can/af_can.h index 38a79ff..fca0fe9 100644 --- a/net/can/af_can.h +++ b/net/can/af_can.h @@ -113,19 +113,8 @@ struct s_pstats { extern struct dev_rcv_lists can_rx_alldev_list; /* function prototypes for the CAN networklayer procfs (proc.c) */ -#ifdef CONFIG_PROC_FS void can_init_proc(void); void can_remove_proc(void); -#else -static inline void can_init_proc(void) -{ - pr_info("can: Can't create /proc/net/can. CONFIG_PROC_FS missing!\n"); -} - -static inline void can_remove_proc(void) -{ -} -#endif void can_stat_update(unsigned long data); /* structures and variables from af_can.c needed in proc.c for reading */ -- cgit v0.10.2 From 6f4c2eea353809fb85386d5ce17a30e37042847d Mon Sep 17 00:00:00 2001 From: Ramesh Shanmugasundaram Date: Wed, 22 Jun 2016 13:31:46 +0100 Subject: can: rcar_canfd: Add Classical CAN only mode support The controller can operate in one of the two global modes - CAN FD only mode (default) - Classical CAN (CAN2.0) only mode This patch adds support for Classical CAN only mode. It can be enabled by defining the optional device tree property "renesas,no-can-fd" of this node. Note: R-Car Gen3 h/w manual v0.51E shows bit6 of RSCFDnCFDGCFG as reserved, which is incorrect. This bit is same as RSCFDnGCFG. Signed-off-by: Ramesh Shanmugasundaram Signed-off-by: Marc Kleine-Budde diff --git a/Documentation/devicetree/bindings/net/can/rcar_canfd.txt b/Documentation/devicetree/bindings/net/can/rcar_canfd.txt index d45182b..22a6f10 100644 --- a/Documentation/devicetree/bindings/net/can/rcar_canfd.txt +++ b/Documentation/devicetree/bindings/net/can/rcar_canfd.txt @@ -32,6 +32,12 @@ below properties. - assigned-clocks: phandle of canfd clock. - assigned-clock-rates: maximum frequency of this clock. +Optional property: +The controller can operate in either CAN FD only mode (default) or +Classical CAN only mode. The mode is global to both the channels. In order to +enable the later, define the following optional property. + - renesas,no-can-fd: puts the controller in Classical CAN only mode. + Example ------- @@ -63,12 +69,13 @@ SoC common .dtsi file: Board specific .dts file: -E.g. below enables Channel 1 alone in the board. +E.g. below enables Channel 1 alone in the board in Classical CAN only mode. &canfd { - pinctrl-0 = <&canfd1_pins>; - pinctrl-names = "default"; - status = "okay"; + pinctrl-0 = <&canfd1_pins>; + pinctrl-names = "default"; + renesas,no-can-fd; + status = "okay"; channel1 { status = "okay"; @@ -79,9 +86,9 @@ E.g. below enables Channel 0 alone in the board using External clock as fCAN clock. &canfd { - pinctrl-0 = <&canfd0_pins &can_clk_pins>; - pinctrl-names = "default"; - status = "okay"; + pinctrl-0 = <&canfd0_pins &can_clk_pins>; + pinctrl-names = "default"; + status = "okay"; channel0 { status = "okay"; diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index a39d922..6bcc474 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -16,8 +16,9 @@ * mode, the controller acts as a CAN FD node that can also interoperate with * CAN 2.0 nodes. * - * As of now, this driver does not support the Classical CAN (CAN 2.0) mode, - * which is handled by a different register map compared to CAN FD only mode. + * To switch the controller to Classical CAN (CAN 2.0) only mode, add + * "renesas,no-can-fd" optional property to the device tree node. A h/w reset is + * also required to switch modes. * * Note: The h/w manual register naming convention is clumsy and not acceptable * to use as it is in the driver. However, those names are added as comments @@ -48,15 +49,16 @@ /* RSCFDnCFDGRMCFG */ #define RCANFD_GRMCFG_RCMC BIT(0) -/* RSCFDnCFDGCFG */ -#define RCANFD_GCFG_CMPOC BIT(5) +/* RSCFDnCFDGCFG / RSCFDnGCFG */ +#define RCANFD_GCFG_EEFE BIT(6) +#define RCANFD_GCFG_CMPOC BIT(5) /* CAN FD only */ #define RCANFD_GCFG_DCS BIT(4) #define RCANFD_GCFG_DCE BIT(1) #define RCANFD_GCFG_TPRI BIT(0) -/* RSCFDnCFDGCTR */ +/* RSCFDnCFDGCTR / RSCFDnGCTR */ #define RCANFD_GCTR_TSRST BIT(16) -#define RCANFD_GCTR_CFMPOFIE BIT(11) +#define RCANFD_GCTR_CFMPOFIE BIT(11) /* CAN FD only */ #define RCANFD_GCTR_THLEIE BIT(10) #define RCANFD_GCTR_MEIE BIT(9) #define RCANFD_GCTR_DEIE BIT(8) @@ -66,7 +68,7 @@ #define RCANFD_GCTR_GMDC_GRESET (0x1) #define RCANFD_GCTR_GMDC_GTEST (0x2) -/* RSCFDnCFDGSTS */ +/* RSCFDnCFDGSTS / RSCFDnGSTS */ #define RCANFD_GSTS_GRAMINIT BIT(3) #define RCANFD_GSTS_GSLPSTS BIT(2) #define RCANFD_GSTS_GHLTSTS BIT(1) @@ -74,44 +76,50 @@ /* Non-operational status */ #define RCANFD_GSTS_GNOPM (BIT(0) | BIT(1) | BIT(2) | BIT(3)) -/* RSCFDnCFDGERFL */ +/* RSCFDnCFDGERFL / RSCFDnGERFL */ #define RCANFD_GERFL_EEF1 BIT(17) #define RCANFD_GERFL_EEF0 BIT(16) -#define RCANFD_GERFL_CMPOF BIT(3) +#define RCANFD_GERFL_CMPOF BIT(3) /* CAN FD only */ #define RCANFD_GERFL_THLES BIT(2) #define RCANFD_GERFL_MES BIT(1) #define RCANFD_GERFL_DEF BIT(0) -#define RCANFD_GERFL_ERR(x) ((x) & (RCANFD_GERFL_EEF1 |\ - RCANFD_GERFL_EEF0 |\ - RCANFD_GERFL_MES |\ - RCANFD_GERFL_CMPOF)) +#define RCANFD_GERFL_ERR(gpriv, x) ((x) & (RCANFD_GERFL_EEF1 |\ + RCANFD_GERFL_EEF0 | RCANFD_GERFL_MES |\ + (gpriv->fdmode ?\ + RCANFD_GERFL_CMPOF : 0))) /* AFL Rx rules registers */ -/* RSCFDnCFDGAFLCFG0 */ +/* RSCFDnCFDGAFLCFG0 / RSCFDnGAFLCFG0 */ #define RCANFD_GAFLCFG_SETRNC(n, x) (((x) & 0xff) << (24 - n * 8)) #define RCANFD_GAFLCFG_GETRNC(n, x) (((x) >> (24 - n * 8)) & 0xff) -/* RSCFDnCFDGAFLECTR */ +/* RSCFDnCFDGAFLECTR / RSCFDnGAFLECTR */ #define RCANFD_GAFLECTR_AFLDAE BIT(8) #define RCANFD_GAFLECTR_AFLPN(x) ((x) & 0x1f) -/* RSCFDnCFDGAFLIDj */ +/* RSCFDnCFDGAFLIDj / RSCFDnGAFLIDj */ #define RCANFD_GAFLID_GAFLLB BIT(29) -/* RSCFDnCFDGAFLP1_j */ +/* RSCFDnCFDGAFLP1_j / RSCFDnGAFLP1_j */ #define RCANFD_GAFLP1_GAFLFDP(x) (1 << (x)) /* Channel register bits */ -/* RSCFDnCFDCmNCFG */ +/* RSCFDnCmCFG - Classical CAN only */ +#define RCANFD_CFG_SJW(x) (((x) & 0x3) << 24) +#define RCANFD_CFG_TSEG2(x) (((x) & 0x7) << 20) +#define RCANFD_CFG_TSEG1(x) (((x) & 0xf) << 16) +#define RCANFD_CFG_BRP(x) (((x) & 0x3ff) << 0) + +/* RSCFDnCFDCmNCFG - CAN FD only */ #define RCANFD_NCFG_NTSEG2(x) (((x) & 0x1f) << 24) #define RCANFD_NCFG_NTSEG1(x) (((x) & 0x7f) << 16) #define RCANFD_NCFG_NSJW(x) (((x) & 0x1f) << 11) #define RCANFD_NCFG_NBRP(x) (((x) & 0x3ff) << 0) -/* RSCFDnCFDCmCTR */ +/* RSCFDnCFDCmCTR / RSCFDnCmCTR */ #define RCANFD_CCTR_CTME BIT(24) #define RCANFD_CCTR_ERRD BIT(23) #define RCANFD_CCTR_BOM_MASK (0x3 << 21) @@ -136,7 +144,7 @@ #define RCANFD_CCTR_CHDMC_CRESET (0x1) #define RCANFD_CCTR_CHDMC_CHLT (0x2) -/* RSCFDnCFDCmSTS */ +/* RSCFDnCFDCmSTS / RSCFDnCmSTS */ #define RCANFD_CSTS_COMSTS BIT(7) #define RCANFD_CSTS_RECSTS BIT(6) #define RCANFD_CSTS_TRMSTS BIT(5) @@ -149,7 +157,7 @@ #define RCANFD_CSTS_TECCNT(x) (((x) >> 24) & 0xff) #define RCANFD_CSTS_RECCNT(x) (((x) >> 16) & 0xff) -/* RSCFDnCFDCmERFL */ +/* RSCFDnCFDCmERFL / RSCFDnCmERFL */ #define RCANFD_CERFL_ADERR BIT(14) #define RCANFD_CERFL_B0ERR BIT(13) #define RCANFD_CERFL_B1ERR BIT(12) @@ -239,14 +247,14 @@ #define RCANFD_CFFDCSTS_CFBRS BIT(1) #define RCANFD_CFFDCSTS_CFESI BIT(0) -/* This controller supports classical CAN only mode or CAN FD only mode. These - * modes are supported in two separate set of register maps & names. However, - * some of the register offsets are common for both modes. Those offsets are - * listed below as Common registers. +/* This controller supports either Classical CAN only mode or CAN FD only mode. + * These modes are supported in two separate set of register maps & names. + * However, some of the register offsets are common for both modes. Those + * offsets are listed below as Common registers. * - * The CAN FD only specific registers are listed separately and their names - * starts with RCANFD_F_xxx names. When classical CAN only specific registers - * are added, those specific registers can be prefixed as RCANFD_C_xxx. + * The CAN FD only mode specific registers & Classical CAN only mode specific + * registers are listed separately. Their register names starts with + * RCANFD_F_xxx & RCANFD_C_xxx respectively. */ /* Common registers */ @@ -353,7 +361,7 @@ #define RCANFD_GTSTCTR (0x046c) /* RSCFDnCFDGLOCKK / RSCFDnGLOCKK */ #define RCANFD_GLOCKK (0x047c) -/* RSCFDnCFDGRMCFG / RSCFDnGRMCFG */ +/* RSCFDnCFDGRMCFG */ #define RCANFD_GRMCFG (0x04fc) /* RSCFDnCFDGAFLIDj / RSCFDnGAFLIDj */ @@ -365,6 +373,46 @@ /* RSCFDnCFDGAFLP1j / RSCFDnGAFLP1j */ #define RCANFD_GAFLP1(offset, j) ((offset) + 0x0c + (0x10 * (j))) +/* Classical CAN only mode register map */ + +/* RSCFDnGAFLXXXj offset */ +#define RCANFD_C_GAFL_OFFSET (0x0500) + +/* RSCFDnRMXXXq -> RCANFD_C_RMXXX(q) */ +#define RCANFD_C_RMID(q) (0x0600 + (0x10 * (q))) +#define RCANFD_C_RMPTR(q) (0x0604 + (0x10 * (q))) +#define RCANFD_C_RMDF0(q) (0x0608 + (0x10 * (q))) +#define RCANFD_C_RMDF1(q) (0x060c + (0x10 * (q))) + +/* RSCFDnRFXXx -> RCANFD_C_RFXX(x) */ +#define RCANFD_C_RFOFFSET (0x0e00) +#define RCANFD_C_RFID(x) (RCANFD_C_RFOFFSET + (0x10 * (x))) +#define RCANFD_C_RFPTR(x) (RCANFD_C_RFOFFSET + 0x04 + \ + (0x10 * (x))) +#define RCANFD_C_RFDF(x, df) (RCANFD_C_RFOFFSET + 0x08 + \ + (0x10 * (x)) + (0x04 * (df))) + +/* RSCFDnCFXXk -> RCANFD_C_CFXX(ch, k) */ +#define RCANFD_C_CFOFFSET (0x0e80) +#define RCANFD_C_CFID(ch, idx) (RCANFD_C_CFOFFSET + (0x30 * (ch)) + \ + (0x10 * (idx))) +#define RCANFD_C_CFPTR(ch, idx) (RCANFD_C_CFOFFSET + 0x04 + \ + (0x30 * (ch)) + (0x10 * (idx))) +#define RCANFD_C_CFDF(ch, idx, df) (RCANFD_C_CFOFFSET + 0x08 + \ + (0x30 * (ch)) + (0x10 * (idx)) + \ + (0x04 * (df))) + +/* RSCFDnTMXXp -> RCANFD_C_TMXX(p) */ +#define RCANFD_C_TMID(p) (0x1000 + (0x10 * (p))) +#define RCANFD_C_TMPTR(p) (0x1004 + (0x10 * (p))) +#define RCANFD_C_TMDF0(p) (0x1008 + (0x10 * (p))) +#define RCANFD_C_TMDF1(p) (0x100c + (0x10 * (p))) + +/* RSCFDnTHLACCm */ +#define RCANFD_C_THLACC(m) (0x1800 + (0x04 * (m))) +/* RSCFDnRPGACCr */ +#define RCANFD_C_RPGACC(r) (0x1900 + (0x04 * (r))) + /* CAN FD mode specific regsiter map */ /* RSCFDnCFDCmXXX -> RCANFD_F_XXX(m) */ @@ -468,6 +516,7 @@ struct rcar_canfd_global { struct clk *can_clk; /* fCAN clock */ enum rcar_canfd_fcanclk fcan; /* CANFD or Ext clock */ unsigned long channels_mask; /* Enabled channels mask */ + bool fdmode; /* CAN FD or Classical CAN only mode */ }; /* CAN FD mode nominal rate constants */ @@ -496,6 +545,19 @@ static const struct can_bittiming_const rcar_canfd_data_bittiming_const = { .brp_inc = 1, }; +/* Classical CAN mode bitrate constants */ +static const struct can_bittiming_const rcar_canfd_bittiming_const = { + .name = RCANFD_DRV_NAME, + .tseg1_min = 4, + .tseg1_max = 16, + .tseg2_min = 2, + .tseg2_max = 8, + .sjw_max = 4, + .brp_min = 1, + .brp_max = 1024, + .brp_inc = 1, +}; + /* Helper functions */ static inline void rcar_canfd_update(u32 mask, u32 val, u32 __iomem *reg) { @@ -593,8 +655,13 @@ static int rcar_canfd_reset_controller(struct rcar_canfd_global *gpriv) /* Reset Global error flags */ rcar_canfd_write(gpriv->base, RCANFD_GERFL, 0x0); - /* Set the controller into FD mode */ - rcar_canfd_set_bit(gpriv->base, RCANFD_GRMCFG, RCANFD_GRMCFG_RCMC); + /* Set the controller into appropriate mode */ + if (gpriv->fdmode) + rcar_canfd_set_bit(gpriv->base, RCANFD_GRMCFG, + RCANFD_GRMCFG_RCMC); + else + rcar_canfd_clear_bit(gpriv->base, RCANFD_GRMCFG, + RCANFD_GRMCFG_RCMC); /* Transition all Channels to reset mode */ for_each_set_bit(ch, &gpriv->channels_mask, RCANFD_NUM_CHANNELS) { @@ -624,8 +691,12 @@ static void rcar_canfd_configure_controller(struct rcar_canfd_global *gpriv) /* Global configuration settings */ - /* Truncate payload to configured message size RFPLS */ - cfg = RCANFD_GCFG_CMPOC; + /* ECC Error flag Enable */ + cfg = RCANFD_GCFG_EEFE; + + if (gpriv->fdmode) + /* Truncate payload to configured message size RFPLS */ + cfg |= RCANFD_GCFG_CMPOC; /* Set External Clock if selected */ if (gpriv->fcan != RCANFD_CANFDCLK) @@ -647,7 +718,7 @@ static void rcar_canfd_configure_afl_rules(struct rcar_canfd_global *gpriv, u32 ch) { u32 cfg; - int start, page, num_rules = RCANFD_CHANNEL_NUMRULES; + int offset, start, page, num_rules = RCANFD_CHANNEL_NUMRULES; u32 ridx = ch + RCANFD_RFFIFO_IDX; if (ch == 0) { @@ -667,19 +738,19 @@ static void rcar_canfd_configure_afl_rules(struct rcar_canfd_global *gpriv, /* Write number of rules for channel */ rcar_canfd_set_bit(gpriv->base, RCANFD_GAFLCFG0, RCANFD_GAFLCFG_SETRNC(ch, num_rules)); + if (gpriv->fdmode) + offset = RCANFD_F_GAFL_OFFSET; + else + offset = RCANFD_C_GAFL_OFFSET; /* Accept all IDs */ - rcar_canfd_write(gpriv->base, - RCANFD_GAFLID(RCANFD_F_GAFL_OFFSET, start), 0); + rcar_canfd_write(gpriv->base, RCANFD_GAFLID(offset, start), 0); /* IDE or RTR is not considered for matching */ - rcar_canfd_write(gpriv->base, - RCANFD_GAFLM(RCANFD_F_GAFL_OFFSET, start), 0); + rcar_canfd_write(gpriv->base, RCANFD_GAFLM(offset, start), 0); /* Any data length accepted */ - rcar_canfd_write(gpriv->base, - RCANFD_GAFLP0(RCANFD_F_GAFL_OFFSET, start), 0); + rcar_canfd_write(gpriv->base, RCANFD_GAFLP0(offset, start), 0); /* Place the msg in corresponding Rx FIFO entry */ - rcar_canfd_write(gpriv->base, - RCANFD_GAFLP1(RCANFD_F_GAFL_OFFSET, start), + rcar_canfd_write(gpriv->base, RCANFD_GAFLP1(offset, start), RCANFD_GAFLP1_GAFLFDP(ridx)); /* Disable write access to page */ @@ -697,7 +768,10 @@ static void rcar_canfd_configure_rx(struct rcar_canfd_global *gpriv, u32 ch) u32 ridx = ch + RCANFD_RFFIFO_IDX; rfdc = 2; /* b010 - 8 messages Rx FIFO depth */ - rfpls = 7; /* b111 - Max 64 bytes payload */ + if (gpriv->fdmode) + rfpls = 7; /* b111 - Max 64 bytes payload */ + else + rfpls = 0; /* b000 - Max 8 bytes payload */ cfg = (RCANFD_RFCC_RFIM | RCANFD_RFCC_RFDC(rfdc) | RCANFD_RFCC_RFPLS(rfpls) | RCANFD_RFCC_RFIE); @@ -718,16 +792,20 @@ static void rcar_canfd_configure_tx(struct rcar_canfd_global *gpriv, u32 ch) cftml = 0; /* 0th buffer */ cfm = 1; /* b01 - Transmit mode */ cfdc = 2; /* b010 - 8 messages Tx FIFO depth */ - cfpls = 7; /* b111 - Max 64 bytes payload */ + if (gpriv->fdmode) + cfpls = 7; /* b111 - Max 64 bytes payload */ + else + cfpls = 0; /* b000 - Max 8 bytes payload */ cfg = (RCANFD_CFCC_CFTML(cftml) | RCANFD_CFCC_CFM(cfm) | RCANFD_CFCC_CFIM | RCANFD_CFCC_CFDC(cfdc) | RCANFD_CFCC_CFPLS(cfpls) | RCANFD_CFCC_CFTXIE); rcar_canfd_write(gpriv->base, RCANFD_CFCC(ch, RCANFD_CFFIFO_IDX), cfg); - /* Clear FD mode specific control/status register */ - rcar_canfd_write(gpriv->base, - RCANFD_F_CFFDCSTS(ch, RCANFD_CFFIFO_IDX), 0); + if (gpriv->fdmode) + /* Clear FD mode specific control/status register */ + rcar_canfd_write(gpriv->base, + RCANFD_F_CFFDCSTS(ch, RCANFD_CFFIFO_IDX), 0); } static void rcar_canfd_enable_global_interrupts(struct rcar_canfd_global *gpriv) @@ -739,7 +817,8 @@ static void rcar_canfd_enable_global_interrupts(struct rcar_canfd_global *gpriv) /* Global interrupts setup */ ctr = RCANFD_GCTR_MEIE; - ctr |= RCANFD_GCTR_CFMPOFIE; + if (gpriv->fdmode) + ctr |= RCANFD_GCTR_CFMPOFIE; rcar_canfd_set_bit(gpriv->base, RCANFD_GCTR, ctr); } @@ -790,6 +869,7 @@ static void rcar_canfd_disable_channel_interrupts(struct rcar_canfd_channel static void rcar_canfd_global_error(struct net_device *ndev) { struct rcar_canfd_channel *priv = netdev_priv(ndev); + struct rcar_canfd_global *gpriv = priv->gpriv; struct net_device_stats *stats = &ndev->stats; u32 ch = priv->channel; u32 gerfl, sts; @@ -823,7 +903,7 @@ static void rcar_canfd_global_error(struct net_device *ndev) sts & ~RCANFD_RFSTS_RFMLT); } } - if (gerfl & RCANFD_GERFL_CMPOF) { + if (gpriv->fdmode && gerfl & RCANFD_GERFL_CMPOF) { /* Message Lost flag will be set for respective channel * when this condition happens with counters and flags * already updated. @@ -1018,7 +1098,7 @@ static irqreturn_t rcar_canfd_global_interrupt(int irq, void *dev_id) /* Global error interrupts */ gerfl = rcar_canfd_read(priv->base, RCANFD_GERFL); - if (RCANFD_GERFL_ERR(gerfl)) + if (RCANFD_GERFL_ERR(gpriv, gerfl)) rcar_canfd_global_error(ndev); /* Handle Rx interrupts */ @@ -1077,25 +1157,37 @@ static void rcar_canfd_set_bittiming(struct net_device *dev) tseg1 = bt->prop_seg + bt->phase_seg1 - 1; tseg2 = bt->phase_seg2 - 1; - cfg = (RCANFD_NCFG_NTSEG1(tseg1) | RCANFD_NCFG_NBRP(brp) | - RCANFD_NCFG_NSJW(sjw) | RCANFD_NCFG_NTSEG2(tseg2)); + if (priv->can.ctrlmode & CAN_CTRLMODE_FD) { + /* CAN FD only mode */ + cfg = (RCANFD_NCFG_NTSEG1(tseg1) | RCANFD_NCFG_NBRP(brp) | + RCANFD_NCFG_NSJW(sjw) | RCANFD_NCFG_NTSEG2(tseg2)); - rcar_canfd_write(priv->base, RCANFD_CCFG(ch), cfg); - netdev_dbg(priv->ndev, "nrate: brp %u, sjw %u, tseg1 %u, tseg2 %u\n", - brp, sjw, tseg1, tseg2); + rcar_canfd_write(priv->base, RCANFD_CCFG(ch), cfg); + netdev_dbg(priv->ndev, "nrate: brp %u, sjw %u, tseg1 %u, tseg2 %u\n", + brp, sjw, tseg1, tseg2); - /* Data bit timing settings */ - brp = dbt->brp - 1; - sjw = dbt->sjw - 1; - tseg1 = dbt->prop_seg + dbt->phase_seg1 - 1; - tseg2 = dbt->phase_seg2 - 1; + /* Data bit timing settings */ + brp = dbt->brp - 1; + sjw = dbt->sjw - 1; + tseg1 = dbt->prop_seg + dbt->phase_seg1 - 1; + tseg2 = dbt->phase_seg2 - 1; - cfg = (RCANFD_DCFG_DTSEG1(tseg1) | RCANFD_DCFG_DBRP(brp) | - RCANFD_DCFG_DSJW(sjw) | RCANFD_DCFG_DTSEG2(tseg2)); + cfg = (RCANFD_DCFG_DTSEG1(tseg1) | RCANFD_DCFG_DBRP(brp) | + RCANFD_DCFG_DSJW(sjw) | RCANFD_DCFG_DTSEG2(tseg2)); - rcar_canfd_write(priv->base, RCANFD_F_DCFG(ch), cfg); - netdev_dbg(priv->ndev, "drate: brp %u, sjw %u, tseg1 %u, tseg2 %u\n", - brp, sjw, tseg1, tseg2); + rcar_canfd_write(priv->base, RCANFD_F_DCFG(ch), cfg); + netdev_dbg(priv->ndev, "drate: brp %u, sjw %u, tseg1 %u, tseg2 %u\n", + brp, sjw, tseg1, tseg2); + } else { + /* Classical CAN only mode */ + cfg = (RCANFD_CFG_TSEG1(tseg1) | RCANFD_CFG_BRP(brp) | + RCANFD_CFG_SJW(sjw) | RCANFD_CFG_TSEG2(tseg2)); + + rcar_canfd_write(priv->base, RCANFD_CCFG(ch), cfg); + netdev_dbg(priv->ndev, + "rate: brp %u, sjw %u, tseg1 %u, tseg2 %u\n", + brp, sjw, tseg1, tseg2); + } } static int rcar_canfd_start(struct net_device *ndev) @@ -1233,27 +1325,37 @@ static netdev_tx_t rcar_canfd_start_xmit(struct sk_buff *skb, if (cf->can_id & CAN_RTR_FLAG) id |= RCANFD_CFID_CFRTR; - rcar_canfd_write(priv->base, - RCANFD_F_CFID(ch, RCANFD_CFFIFO_IDX), id); dlc = RCANFD_CFPTR_CFDLC(can_len2dlc(cf->len)); - rcar_canfd_write(priv->base, - RCANFD_F_CFPTR(ch, RCANFD_CFFIFO_IDX), dlc); - if (can_is_canfd_skb(skb)) { - /* CAN FD frame format */ - sts |= RCANFD_CFFDCSTS_CFFDF; - if (cf->flags & CANFD_BRS) - sts |= RCANFD_CFFDCSTS_CFBRS; + if (priv->can.ctrlmode & CAN_CTRLMODE_FD) { + rcar_canfd_write(priv->base, + RCANFD_F_CFID(ch, RCANFD_CFFIFO_IDX), id); + rcar_canfd_write(priv->base, + RCANFD_F_CFPTR(ch, RCANFD_CFFIFO_IDX), dlc); - if (priv->can.state == CAN_STATE_ERROR_PASSIVE) - sts |= RCANFD_CFFDCSTS_CFESI; - } + if (can_is_canfd_skb(skb)) { + /* CAN FD frame format */ + sts |= RCANFD_CFFDCSTS_CFFDF; + if (cf->flags & CANFD_BRS) + sts |= RCANFD_CFFDCSTS_CFBRS; - rcar_canfd_write(priv->base, RCANFD_F_CFFDCSTS(ch, RCANFD_CFFIFO_IDX), - sts); + if (priv->can.state == CAN_STATE_ERROR_PASSIVE) + sts |= RCANFD_CFFDCSTS_CFESI; + } + + rcar_canfd_write(priv->base, + RCANFD_F_CFFDCSTS(ch, RCANFD_CFFIFO_IDX), sts); - rcar_canfd_put_data(priv, cf, - RCANFD_F_CFDF(ch, RCANFD_CFFIFO_IDX, 0)); + rcar_canfd_put_data(priv, cf, + RCANFD_F_CFDF(ch, RCANFD_CFFIFO_IDX, 0)); + } else { + rcar_canfd_write(priv->base, + RCANFD_C_CFID(ch, RCANFD_CFFIFO_IDX), id); + rcar_canfd_write(priv->base, + RCANFD_C_CFPTR(ch, RCANFD_CFFIFO_IDX), dlc); + rcar_canfd_put_data(priv, cf, + RCANFD_C_CFDF(ch, RCANFD_CFFIFO_IDX, 0)); + } priv->tx_len[priv->tx_head % RCANFD_FIFO_DEPTH] = cf->len; can_put_echo_skb(skb, ndev, priv->tx_head % RCANFD_FIFO_DEPTH); @@ -1280,47 +1382,61 @@ static void rcar_canfd_rx_pkt(struct rcar_canfd_channel *priv) struct net_device_stats *stats = &priv->ndev->stats; struct canfd_frame *cf; struct sk_buff *skb; - u32 sts = 0, id, ptr; + u32 sts = 0, id, dlc; u32 ch = priv->channel; u32 ridx = ch + RCANFD_RFFIFO_IDX; - id = rcar_canfd_read(priv->base, RCANFD_F_RFID(ridx)); - ptr = rcar_canfd_read(priv->base, RCANFD_F_RFPTR(ridx)); + if (priv->can.ctrlmode & CAN_CTRLMODE_FD) { + id = rcar_canfd_read(priv->base, RCANFD_F_RFID(ridx)); + dlc = rcar_canfd_read(priv->base, RCANFD_F_RFPTR(ridx)); - sts = rcar_canfd_read(priv->base, RCANFD_F_RFFDSTS(ridx)); - if (sts & RCANFD_RFFDSTS_RFFDF) - skb = alloc_canfd_skb(priv->ndev, &cf); - else - skb = alloc_can_skb(priv->ndev, - (struct can_frame **)&cf); + sts = rcar_canfd_read(priv->base, RCANFD_F_RFFDSTS(ridx)); + if (sts & RCANFD_RFFDSTS_RFFDF) + skb = alloc_canfd_skb(priv->ndev, &cf); + else + skb = alloc_can_skb(priv->ndev, + (struct can_frame **)&cf); + } else { + id = rcar_canfd_read(priv->base, RCANFD_C_RFID(ridx)); + dlc = rcar_canfd_read(priv->base, RCANFD_C_RFPTR(ridx)); + skb = alloc_can_skb(priv->ndev, (struct can_frame **)&cf); + } if (!skb) { stats->rx_dropped++; return; } - if (sts & RCANFD_RFFDSTS_RFFDF) - cf->len = can_dlc2len(RCANFD_RFPTR_RFDLC(ptr)); - else - cf->len = get_can_dlc(RCANFD_RFPTR_RFDLC(ptr)); - - if (sts & RCANFD_RFFDSTS_RFESI) { - cf->flags |= CANFD_ESI; - netdev_dbg(priv->ndev, "ESI Error\n"); - } - if (id & RCANFD_RFID_RFIDE) cf->can_id = (id & CAN_EFF_MASK) | CAN_EFF_FLAG; else cf->can_id = id & CAN_SFF_MASK; - if (!(sts & RCANFD_RFFDSTS_RFFDF) && (id & RCANFD_RFID_RFRTR)) { - cf->can_id |= CAN_RTR_FLAG; - } else { - if (sts & RCANFD_RFFDSTS_RFBRS) - cf->flags |= CANFD_BRS; + if (priv->can.ctrlmode & CAN_CTRLMODE_FD) { + if (sts & RCANFD_RFFDSTS_RFFDF) + cf->len = can_dlc2len(RCANFD_RFPTR_RFDLC(dlc)); + else + cf->len = get_can_dlc(RCANFD_RFPTR_RFDLC(dlc)); + + if (sts & RCANFD_RFFDSTS_RFESI) { + cf->flags |= CANFD_ESI; + netdev_dbg(priv->ndev, "ESI Error\n"); + } + + if (!(sts & RCANFD_RFFDSTS_RFFDF) && (id & RCANFD_RFID_RFRTR)) { + cf->can_id |= CAN_RTR_FLAG; + } else { + if (sts & RCANFD_RFFDSTS_RFBRS) + cf->flags |= CANFD_BRS; - rcar_canfd_get_data(priv, cf, RCANFD_F_RFDF(ridx, 0)); + rcar_canfd_get_data(priv, cf, RCANFD_F_RFDF(ridx, 0)); + } + } else { + cf->len = get_can_dlc(RCANFD_RFPTR_RFDLC(dlc)); + if (id & RCANFD_RFID_RFRTR) + cf->can_id |= CAN_RTR_FLAG; + else + rcar_canfd_get_data(priv, cf, RCANFD_C_RFDF(ridx, 0)); } /* Write 0xff to RFPC to increment the CPU-side @@ -1428,13 +1544,19 @@ static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch, priv->can.clock.freq = fcan_freq; dev_info(&pdev->dev, "can_clk rate is %u\n", priv->can.clock.freq); - priv->can.bittiming_const = &rcar_canfd_nom_bittiming_const; - priv->can.data_bittiming_const = - &rcar_canfd_data_bittiming_const; + if (gpriv->fdmode) { + priv->can.bittiming_const = &rcar_canfd_nom_bittiming_const; + priv->can.data_bittiming_const = + &rcar_canfd_data_bittiming_const; - /* Controller starts in CAN FD only mode */ - can_set_static_ctrlmode(ndev, CAN_CTRLMODE_FD); - priv->can.ctrlmode_supported = CAN_CTRLMODE_BERR_REPORTING; + /* Controller starts in CAN FD only mode */ + can_set_static_ctrlmode(ndev, CAN_CTRLMODE_FD); + priv->can.ctrlmode_supported = CAN_CTRLMODE_BERR_REPORTING; + } else { + /* Controller starts in Classical CAN only mode */ + priv->can.bittiming_const = &rcar_canfd_bittiming_const; + priv->can.ctrlmode_supported = CAN_CTRLMODE_BERR_REPORTING; + } priv->can.do_set_mode = rcar_canfd_do_set_mode; priv->can.do_get_berr_counter = rcar_canfd_get_berr_counter; @@ -1482,6 +1604,10 @@ static int rcar_canfd_probe(struct platform_device *pdev) struct device_node *of_child; unsigned long channels_mask = 0; int err, ch_irq, g_irq; + bool fdmode = true; /* CAN FD only mode - default */ + + if (of_property_read_bool(pdev->dev.of_node, "renesas,no-can-fd")) + fdmode = false; /* Classical CAN only mode */ of_child = of_get_child_by_name(pdev->dev.of_node, "channel0"); if (of_child && of_device_is_available(of_child)) @@ -1513,6 +1639,7 @@ static int rcar_canfd_probe(struct platform_device *pdev) } gpriv->pdev = pdev; gpriv->channels_mask = channels_mask; + gpriv->fdmode = fdmode; /* Peripheral clock */ gpriv->clkp = devm_clk_get(&pdev->dev, "fck"); @@ -1623,8 +1750,8 @@ static int rcar_canfd_probe(struct platform_device *pdev) } platform_set_drvdata(pdev, gpriv); - dev_info(&pdev->dev, "global operational state (clk %d)\n", - gpriv->fcan); + dev_info(&pdev->dev, "global operational state (clk %d, fdmode %d)\n", + gpriv->fcan, gpriv->fdmode); return 0; fail_channel: -- cgit v0.10.2 From 926f10359a69fa82f0ec022f3b54ca8a05ea8440 Mon Sep 17 00:00:00 2001 From: Ramesh Shanmugasundaram Date: Wed, 22 Jun 2016 13:31:47 +0100 Subject: can: rcar_canfd: Add back-to-error-active support As per Wolfgang G, all new drivers should support decreasing state transition(back-to-error-active). This patch adds this support. This driver configures the controller to halt on bus-off entry. Hence, when in error states less than bus off state, the TEC/REC counters are checked for lower state transition eligibility and action. Signed-off-by: Ramesh Shanmugasundaram Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index 6bcc474..43cdd55 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -917,16 +917,17 @@ static void rcar_canfd_global_error(struct net_device *ndev) rcar_canfd_write(priv->base, RCANFD_GERFL, 0); } -static void rcar_canfd_error(struct net_device *ndev) +static void rcar_canfd_error(struct net_device *ndev, u32 cerfl, + u16 txerr, u16 rxerr) { struct rcar_canfd_channel *priv = netdev_priv(ndev); struct net_device_stats *stats = &ndev->stats; struct can_frame *cf; struct sk_buff *skb; - u32 cerfl, csts; - u32 txerr = 0, rxerr = 0; u32 ch = priv->channel; + netdev_dbg(ndev, "ch erfl %x txerr %u rxerr %u\n", cerfl, txerr, rxerr); + /* Propagate the error condition to the CAN stack */ skb = alloc_can_err_skb(ndev, &cf); if (!skb) { @@ -934,15 +935,7 @@ static void rcar_canfd_error(struct net_device *ndev) return; } - /* Channel error interrupt */ - cerfl = rcar_canfd_read(priv->base, RCANFD_CERFL(ch)); - csts = rcar_canfd_read(priv->base, RCANFD_CSTS(ch)); - txerr = RCANFD_CSTS_TECCNT(csts); - rxerr = RCANFD_CSTS_RECCNT(csts); - - netdev_dbg(ndev, "ch erfl %x sts %x txerr %u rxerr %u\n", - cerfl, csts, txerr, rxerr); - + /* Channel error interrupts */ if (cerfl & RCANFD_CERFL_BEF) { netdev_dbg(ndev, "Bus error\n"); cf->can_id |= CAN_ERR_BUSERROR | CAN_ERR_PROT; @@ -1032,8 +1025,9 @@ static void rcar_canfd_error(struct net_device *ndev) cf->data[2] |= CAN_ERR_PROT_OVERLOAD; } - /* Clear all channel error interrupts */ - rcar_canfd_write(priv->base, RCANFD_CERFL(ch), 0); + /* Clear channel error interrupts that are handled */ + rcar_canfd_write(priv->base, RCANFD_CERFL(ch), + RCANFD_CERFL_ERR(~cerfl)); stats->rx_packets++; stats->rx_bytes += cf->can_dlc; netif_rx(skb); @@ -1098,12 +1092,12 @@ static irqreturn_t rcar_canfd_global_interrupt(int irq, void *dev_id) /* Global error interrupts */ gerfl = rcar_canfd_read(priv->base, RCANFD_GERFL); - if (RCANFD_GERFL_ERR(gpriv, gerfl)) + if (unlikely(RCANFD_GERFL_ERR(gpriv, gerfl))) rcar_canfd_global_error(ndev); /* Handle Rx interrupts */ sts = rcar_canfd_read(priv->base, RCANFD_RFSTS(ridx)); - if (sts & RCANFD_RFSTS_RFIF) { + if (likely(sts & RCANFD_RFSTS_RFIF)) { if (napi_schedule_prep(&priv->napi)) { /* Disable Rx FIFO interrupts */ rcar_canfd_clear_bit(priv->base, @@ -1116,12 +1110,46 @@ static irqreturn_t rcar_canfd_global_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } +static void rcar_canfd_state_change(struct net_device *ndev, + u16 txerr, u16 rxerr) +{ + struct rcar_canfd_channel *priv = netdev_priv(ndev); + struct net_device_stats *stats = &ndev->stats; + enum can_state rx_state, tx_state, state = priv->can.state; + struct can_frame *cf; + struct sk_buff *skb; + + /* Handle transition from error to normal states */ + if (txerr < 96 && rxerr < 96) + state = CAN_STATE_ERROR_ACTIVE; + else if (txerr < 128 && rxerr < 128) + state = CAN_STATE_ERROR_WARNING; + + if (state != priv->can.state) { + netdev_dbg(ndev, "state: new %d, old %d: txerr %u, rxerr %u\n", + state, priv->can.state, txerr, rxerr); + skb = alloc_can_err_skb(ndev, &cf); + if (!skb) { + stats->rx_dropped++; + return; + } + tx_state = txerr >= rxerr ? state : 0; + rx_state = txerr <= rxerr ? state : 0; + + can_change_state(ndev, cf, tx_state, rx_state); + stats->rx_packets++; + stats->rx_bytes += cf->can_dlc; + netif_rx(skb); + } +} + static irqreturn_t rcar_canfd_channel_interrupt(int irq, void *dev_id) { struct rcar_canfd_global *gpriv = dev_id; struct net_device *ndev; struct rcar_canfd_channel *priv; - u32 sts, cerfl, ch; + u32 sts, ch, cerfl; + u16 txerr, rxerr; /* Common FIFO is a per channel resource */ for_each_set_bit(ch, &gpriv->channels_mask, RCANFD_NUM_CHANNELS) { @@ -1130,13 +1158,21 @@ static irqreturn_t rcar_canfd_channel_interrupt(int irq, void *dev_id) /* Channel error interrupts */ cerfl = rcar_canfd_read(priv->base, RCANFD_CERFL(ch)); - if (RCANFD_CERFL_ERR(cerfl)) - rcar_canfd_error(ndev); + sts = rcar_canfd_read(priv->base, RCANFD_CSTS(ch)); + txerr = RCANFD_CSTS_TECCNT(sts); + rxerr = RCANFD_CSTS_RECCNT(sts); + if (unlikely(RCANFD_CERFL_ERR(cerfl))) + rcar_canfd_error(ndev, cerfl, txerr, rxerr); + + /* Handle state change to lower states */ + if (unlikely((priv->can.state != CAN_STATE_ERROR_ACTIVE) && + (priv->can.state != CAN_STATE_BUS_OFF))) + rcar_canfd_state_change(ndev, txerr, rxerr); /* Handle Tx interrupts */ sts = rcar_canfd_read(priv->base, RCANFD_CFSTS(ch, RCANFD_CFFIFO_IDX)); - if (sts & RCANFD_CFSTS_CFTXIF) + if (likely(sts & RCANFD_CFSTS_CFTXIF)) rcar_canfd_tx_done(ndev); } return IRQ_HANDLED; -- cgit v0.10.2 From b63f69d0fc1fa1e25842a2266633862d523c380f Mon Sep 17 00:00:00 2001 From: Ed Spiridonov Date: Mon, 20 Jun 2016 21:40:15 +0300 Subject: can: mcp251x: add message about sucessful/unsuccessful probe Silent ignorance of errors during probe procedure is a bad thing, this patch fixes it. Extra message added for hardware initialization failure. Such common issues are mostly caused by wrong wiring. Message about success added as well, it should be useful to debug new hardware configuration, especially in case of several CAN buses. Signed-off-by: Ed Spiridonov Signed-off-by: Marc Kleine-Budde diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c index cf36d26..f3f05fe 100644 --- a/drivers/net/can/spi/mcp251x.c +++ b/drivers/net/can/spi/mcp251x.c @@ -1145,8 +1145,11 @@ static int mcp251x_can_probe(struct spi_device *spi) /* Here is OK to not lock the MCP, no one knows about it yet */ ret = mcp251x_hw_probe(spi); - if (ret) + if (ret) { + if (ret == -ENODEV) + dev_err(&spi->dev, "Cannot initialize MCP%x. Wrong wiring?\n", priv->model); goto error_probe; + } mcp251x_hw_sleep(spi); @@ -1156,6 +1159,7 @@ static int mcp251x_can_probe(struct spi_device *spi) devm_can_led_init(net); + netdev_info(net, "MCP%x successfully initialized.\n", priv->model); return 0; error_probe: @@ -1168,6 +1172,7 @@ out_clk: out_free: free_candev(net); + dev_err(&spi->dev, "Probe failed, err=%d\n", -ret); return ret; } -- cgit v0.10.2 From a6d0bae14858a43ab9d76d6332d7c3f2a618a6a2 Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Thu, 2 Jun 2016 10:59:56 +0800 Subject: netfilter: x_tables: fix possible ZERO_SIZE_PTR pointer dereferencing error. Since we cannot make sure that the 'hook_mask' will always be none zero here. If it equals to zero, the num_hooks will be zero too, and then kmalloc() will return ZERO_SIZE_PTR, which is (void *)16. Then the following error check will fails: ops = kmalloc(sizeof(*ops) * num_hooks, GFP_KERNEL); if (ops == NULL) return ERR_PTR(-ENOMEM); So this patch will fix this with just doing the zero check before kmalloc() is called. Maybe the case above will never happen here, but in theory. Signed-off-by: Xiubo Li Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index c69c892..8aff34e 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -1460,6 +1460,9 @@ xt_hook_ops_alloc(const struct xt_table *table, nf_hookfn *fn) uint8_t hooknum; struct nf_hook_ops *ops; + if (!num_hooks) + return ERR_PTR(-EINVAL); + ops = kmalloc(sizeof(*ops) * num_hooks, GFP_KERNEL); if (ops == NULL) return ERR_PTR(-ENOMEM); -- cgit v0.10.2 From f3bb53338e0965c3084c185020e821ac49015832 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Wed, 8 Jun 2016 20:43:17 +0800 Subject: netfilter: nf_log: handle NFPROTO_INET properly in nf_logger_[find_get|put] When we request NFPROTO_INET, it means both NFPROTO_IPV4 and NFPROTO_IPV6. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index a5d41df..73b845d 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -159,6 +159,20 @@ int nf_logger_find_get(int pf, enum nf_log_type type) struct nf_logger *logger; int ret = -ENOENT; + if (pf == NFPROTO_INET) { + ret = nf_logger_find_get(NFPROTO_IPV4, type); + if (ret < 0) + return ret; + + ret = nf_logger_find_get(NFPROTO_IPV6, type); + if (ret < 0) { + nf_logger_put(NFPROTO_IPV4, type); + return ret; + } + + return 0; + } + if (rcu_access_pointer(loggers[pf][type]) == NULL) request_module("nf-logger-%u-%u", pf, type); @@ -179,6 +193,12 @@ void nf_logger_put(int pf, enum nf_log_type type) { struct nf_logger *logger; + if (pf == NFPROTO_INET) { + nf_logger_put(NFPROTO_IPV4, type); + nf_logger_put(NFPROTO_IPV6, type); + return; + } + BUG_ON(loggers[pf][type] == NULL); rcu_read_lock(); diff --git a/net/netfilter/nft_log.c b/net/netfilter/nft_log.c index 319c22b..713d668 100644 --- a/net/netfilter/nft_log.c +++ b/net/netfilter/nft_log.c @@ -52,7 +52,6 @@ static int nft_log_init(const struct nft_ctx *ctx, struct nft_log *priv = nft_expr_priv(expr); struct nf_loginfo *li = &priv->loginfo; const struct nlattr *nla; - int ret; nla = tb[NFTA_LOG_PREFIX]; if (nla != NULL) { @@ -97,19 +96,6 @@ static int nft_log_init(const struct nft_ctx *ctx, break; } - if (ctx->afi->family == NFPROTO_INET) { - ret = nf_logger_find_get(NFPROTO_IPV4, li->type); - if (ret < 0) - return ret; - - ret = nf_logger_find_get(NFPROTO_IPV6, li->type); - if (ret < 0) { - nf_logger_put(NFPROTO_IPV4, li->type); - return ret; - } - return 0; - } - return nf_logger_find_get(ctx->afi->family, li->type); } @@ -122,12 +108,7 @@ static void nft_log_destroy(const struct nft_ctx *ctx, if (priv->prefix != nft_log_null_prefix) kfree(priv->prefix); - if (ctx->afi->family == NFPROTO_INET) { - nf_logger_put(NFPROTO_IPV4, li->type); - nf_logger_put(NFPROTO_IPV6, li->type); - } else { - nf_logger_put(ctx->afi->family, li->type); - } + nf_logger_put(ctx->afi->family, li->type); } static int nft_log_dump(struct sk_buff *skb, const struct nft_expr *expr) -- cgit v0.10.2 From 36f959c491abc7e0acf94b631a6d7a3e2e3699b0 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Wed, 8 Jun 2016 20:43:19 +0800 Subject: netfilter: xt_TRACE: add explicitly nf_logger_find_get call Consider such situation, if nf_log_ipv4 kernel module is not installed, and the user add a following iptables rule: # iptables -t raw -I PREROUTING -j TRACE There will be no trace log generated until the user install nf_log_ipv4 module manully. So we should add request related nf_log module appropriately here. Signed-off-by: Liping Zhang Acked-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/xt_TRACE.c b/net/netfilter/xt_TRACE.c index df48967..858d189 100644 --- a/net/netfilter/xt_TRACE.c +++ b/net/netfilter/xt_TRACE.c @@ -4,12 +4,23 @@ #include #include +#include MODULE_DESCRIPTION("Xtables: packet flow tracing"); MODULE_LICENSE("GPL"); MODULE_ALIAS("ipt_TRACE"); MODULE_ALIAS("ip6t_TRACE"); +static int trace_tg_check(const struct xt_tgchk_param *par) +{ + return nf_logger_find_get(par->family, NF_LOG_TYPE_LOG); +} + +static void trace_tg_destroy(const struct xt_tgdtor_param *par) +{ + nf_logger_put(par->family, NF_LOG_TYPE_LOG); +} + static unsigned int trace_tg(struct sk_buff *skb, const struct xt_action_param *par) { @@ -18,12 +29,14 @@ trace_tg(struct sk_buff *skb, const struct xt_action_param *par) } static struct xt_target trace_tg_reg __read_mostly = { - .name = "TRACE", - .revision = 0, - .family = NFPROTO_UNSPEC, - .table = "raw", - .target = trace_tg, - .me = THIS_MODULE, + .name = "TRACE", + .revision = 0, + .family = NFPROTO_UNSPEC, + .table = "raw", + .target = trace_tg, + .checkentry = trace_tg_check, + .destroy = trace_tg_destroy, + .me = THIS_MODULE, }; static int __init trace_tg_init(void) -- cgit v0.10.2 From 5a75cdebabc4576ca31f497a9272ac558421b119 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 10 Jun 2016 22:25:22 +0200 Subject: netfilter: conntrack: align nf_conn on cacheline boundary increases struct size by 32 bytes (288 -> 320), but it is the right thing, else any attempt to (re-)arrange nf_conn members by cacheline won't work. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index db2312e..2903bb4 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1731,7 +1731,7 @@ int nf_conntrack_init_start(void) nf_conntrack_cachep = kmem_cache_create("nf_conntrack", sizeof(struct nf_conn), 0, - SLAB_DESTROY_BY_RCU, NULL); + SLAB_DESTROY_BY_RCU | SLAB_HWCACHE_ALIGN, NULL); if (!nf_conntrack_cachep) goto err_cachep; -- cgit v0.10.2 From 506e65df52f2bf250aa9b4264efd180d1646bdec Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 10 Jun 2016 23:09:01 +0200 Subject: netfilter: make comparision helpers stub functions in ZONES=n case Those comparisions are useless in case of ZONES=n; all conntracks will reside in the same zone by definition. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_conntrack_zones.h b/include/net/netfilter/nf_conntrack_zones.h index 4e32512..bd46926 100644 --- a/include/net/netfilter/nf_conntrack_zones.h +++ b/include/net/netfilter/nf_conntrack_zones.h @@ -68,22 +68,34 @@ static inline bool nf_ct_zone_matches_dir(const struct nf_conntrack_zone *zone, static inline u16 nf_ct_zone_id(const struct nf_conntrack_zone *zone, enum ip_conntrack_dir dir) { +#ifdef CONFIG_NF_CONNTRACK_ZONES return nf_ct_zone_matches_dir(zone, dir) ? zone->id : NF_CT_DEFAULT_ZONE_ID; +#else + return NF_CT_DEFAULT_ZONE_ID; +#endif } static inline bool nf_ct_zone_equal(const struct nf_conn *a, const struct nf_conntrack_zone *b, enum ip_conntrack_dir dir) { +#ifdef CONFIG_NF_CONNTRACK_ZONES return nf_ct_zone_id(nf_ct_zone(a), dir) == nf_ct_zone_id(b, dir); +#else + return true; +#endif } static inline bool nf_ct_zone_equal_any(const struct nf_conn *a, const struct nf_conntrack_zone *b) { +#ifdef CONFIG_NF_CONNTRACK_ZONES return nf_ct_zone(a)->id == b->id; +#else + return true; +#endif } #endif /* IS_ENABLED(CONFIG_NF_CONNTRACK) */ #endif /* _NF_CONNTRACK_ZONES_H */ -- cgit v0.10.2 From 7e53e7f8ca24e01292d114373f35b2999301d879 Mon Sep 17 00:00:00 2001 From: Shivani Bhardwaj Date: Sun, 12 Jun 2016 00:26:10 +0530 Subject: netfilter: nf_log: Remove NULL check If 'logger' was NULL, there would be a direct jump to the label 'out', since it has already been checked for NULL, remove this unnecessary check. Signed-off-by: Shivani Bhardwaj Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index 73b845d..18e325c 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -181,7 +181,7 @@ int nf_logger_find_get(int pf, enum nf_log_type type) if (logger == NULL) goto out; - if (logger && try_module_get(logger->me)) + if (try_module_get(logger->me)) ret = 0; out: rcu_read_unlock(); -- cgit v0.10.2 From 6c8dee9842461e6ee6eb46081478999b3d5cb297 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sat, 11 Jun 2016 21:57:35 +0200 Subject: netfilter: move zone info into struct nf_conn Curently we store zone information as a conntrack extension. This has one drawback: for every lookup we need to fetch the zone data from the extension area. This change place the zone data directly into the main conntrack object structure and then removes the zone conntrack extension. The zone data is just 4 bytes, it fits into a padding hole before the tuplehash info, so we do not even increase the nf_conn structure size. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index dd78bea..9c0ed3d 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -85,6 +85,9 @@ struct nf_conn { spinlock_t lock; u16 cpu; +#ifdef CONFIG_NF_CONNTRACK_ZONES + struct nf_conntrack_zone zone; +#endif /* XXX should I move this to the tail ? - Y.K */ /* These are my tuples; original and reply */ struct nf_conntrack_tuple_hash tuplehash[IP_CT_DIR_MAX]; diff --git a/include/net/netfilter/nf_conntrack_extend.h b/include/net/netfilter/nf_conntrack_extend.h index 55d1504..b925395 100644 --- a/include/net/netfilter/nf_conntrack_extend.h +++ b/include/net/netfilter/nf_conntrack_extend.h @@ -15,9 +15,6 @@ enum nf_ct_ext_id { #ifdef CONFIG_NF_CONNTRACK_EVENTS NF_CT_EXT_ECACHE, #endif -#ifdef CONFIG_NF_CONNTRACK_ZONES - NF_CT_EXT_ZONE, -#endif #ifdef CONFIG_NF_CONNTRACK_TIMESTAMP NF_CT_EXT_TSTAMP, #endif @@ -38,7 +35,6 @@ enum nf_ct_ext_id { #define NF_CT_EXT_SEQADJ_TYPE struct nf_conn_seqadj #define NF_CT_EXT_ACCT_TYPE struct nf_conn_acct #define NF_CT_EXT_ECACHE_TYPE struct nf_conntrack_ecache -#define NF_CT_EXT_ZONE_TYPE struct nf_conntrack_zone #define NF_CT_EXT_TSTAMP_TYPE struct nf_conn_tstamp #define NF_CT_EXT_TIMEOUT_TYPE struct nf_conn_timeout #define NF_CT_EXT_LABELS_TYPE struct nf_conn_labels diff --git a/include/net/netfilter/nf_conntrack_zones.h b/include/net/netfilter/nf_conntrack_zones.h index bd46926..64a718b 100644 --- a/include/net/netfilter/nf_conntrack_zones.h +++ b/include/net/netfilter/nf_conntrack_zones.h @@ -9,12 +9,11 @@ static inline const struct nf_conntrack_zone * nf_ct_zone(const struct nf_conn *ct) { - const struct nf_conntrack_zone *nf_ct_zone = NULL; - #ifdef CONFIG_NF_CONNTRACK_ZONES - nf_ct_zone = nf_ct_ext_find(ct, NF_CT_EXT_ZONE); + return &ct->zone; +#else + return &nf_ct_zone_dflt; #endif - return nf_ct_zone ? nf_ct_zone : &nf_ct_zone_dflt; } static inline const struct nf_conntrack_zone * @@ -31,32 +30,22 @@ static inline const struct nf_conntrack_zone * nf_ct_zone_tmpl(const struct nf_conn *tmpl, const struct sk_buff *skb, struct nf_conntrack_zone *tmp) { - const struct nf_conntrack_zone *zone; - +#ifdef CONFIG_NF_CONNTRACK_ZONES if (!tmpl) return &nf_ct_zone_dflt; - zone = nf_ct_zone(tmpl); - if (zone->flags & NF_CT_FLAG_MARK) - zone = nf_ct_zone_init(tmp, skb->mark, zone->dir, 0); - - return zone; + if (tmpl->zone.flags & NF_CT_FLAG_MARK) + return nf_ct_zone_init(tmp, skb->mark, tmpl->zone.dir, 0); +#endif + return nf_ct_zone(tmpl); } -static inline int nf_ct_zone_add(struct nf_conn *ct, gfp_t flags, - const struct nf_conntrack_zone *info) +static inline void nf_ct_zone_add(struct nf_conn *ct, + const struct nf_conntrack_zone *zone) { #ifdef CONFIG_NF_CONNTRACK_ZONES - struct nf_conntrack_zone *nf_ct_zone; - - nf_ct_zone = nf_ct_ext_add(ct, NF_CT_EXT_ZONE, flags); - if (!nf_ct_zone) - return -ENOMEM; - - nf_ct_zone_init(nf_ct_zone, info->id, info->dir, - info->flags); + ct->zone = *zone; #endif - return 0; } static inline bool nf_ct_zone_matches_dir(const struct nf_conntrack_zone *zone, diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 2903bb4..a459176 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -327,16 +327,10 @@ struct nf_conn *nf_ct_tmpl_alloc(struct net *net, tmpl->status = IPS_TEMPLATE; write_pnet(&tmpl->ct_net, net); - - if (nf_ct_zone_add(tmpl, flags, zone) < 0) - goto out_free; - + nf_ct_zone_add(tmpl, zone); atomic_set(&tmpl->ct_general.use, 0); return tmpl; -out_free: - kfree(tmpl); - return NULL; } EXPORT_SYMBOL_GPL(nf_ct_tmpl_alloc); @@ -929,16 +923,13 @@ __nf_conntrack_alloc(struct net *net, offsetof(struct nf_conn, proto) - offsetof(struct nf_conn, __nfct_init_offset[0])); - if (zone && nf_ct_zone_add(ct, GFP_ATOMIC, zone) < 0) - goto out_free; + nf_ct_zone_add(ct, zone); /* Because we use RCU lookups, we set ct_general.use to zero before * this is inserted in any list. */ atomic_set(&ct->ct_general.use, 0); return ct; -out_free: - kmem_cache_free(nf_conntrack_cachep, ct); out: atomic_dec(&net->ct.count); return ERR_PTR(-ENOMEM); @@ -1342,14 +1333,6 @@ bool __nf_ct_kill_acct(struct nf_conn *ct, } EXPORT_SYMBOL_GPL(__nf_ct_kill_acct); -#ifdef CONFIG_NF_CONNTRACK_ZONES -static struct nf_ct_ext_type nf_ct_zone_extend __read_mostly = { - .len = sizeof(struct nf_conntrack_zone), - .align = __alignof__(struct nf_conntrack_zone), - .id = NF_CT_EXT_ZONE, -}; -#endif - #if IS_ENABLED(CONFIG_NF_CT_NETLINK) #include @@ -1532,9 +1515,6 @@ void nf_conntrack_cleanup_end(void) nf_ct_free_hashtable(nf_conntrack_hash, nf_conntrack_htable_size); -#ifdef CONFIG_NF_CONNTRACK_ZONES - nf_ct_extend_unregister(&nf_ct_zone_extend); -#endif nf_conntrack_proto_fini(); nf_conntrack_seqadj_fini(); nf_conntrack_labels_fini(); @@ -1771,11 +1751,6 @@ int nf_conntrack_init_start(void) if (ret < 0) goto err_seqadj; -#ifdef CONFIG_NF_CONNTRACK_ZONES - ret = nf_ct_extend_register(&nf_ct_zone_extend); - if (ret < 0) - goto err_extend; -#endif ret = nf_conntrack_proto_init(); if (ret < 0) goto err_proto; @@ -1791,10 +1766,6 @@ int nf_conntrack_init_start(void) return 0; err_proto: -#ifdef CONFIG_NF_CONNTRACK_ZONES - nf_ct_extend_unregister(&nf_ct_zone_extend); -err_extend: -#endif nf_conntrack_seqadj_fini(); err_seqadj: nf_conntrack_labels_fini(); -- cgit v0.10.2 From 9847371a84b0be330f4bc4aaa98904101ee8573d Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Tue, 14 Jun 2016 15:14:12 -0700 Subject: netfilter: Allow xt_owner in any user namespace Making this work is a little tricky as it really isn't kosher to change the xt_owner_match_info in a check function. Without changing xt_owner_match_info we need to know the user namespace the uids and gids are specified in. In the common case net->user_ns == current_user_ns(). Verify net->user_ns == current_user_ns() in owner_check so we can later assume it in owner_mt. In owner_check also verify that all of the uids and gids specified are in net->user_ns and that the expected min/max relationship exists between the uids and gids in xt_owner_match_info. In owner_mt get the network namespace from the outgoing socket, as this must be the same network namespace as the netfilter rules, and use that network namespace to find the user namespace the uids and gids in xt_match_owner_info are encoded in. Then convert from their encoded from into the kernel internal format for uids and gids and perform the owner match. Similar to ping_group_range, this code does not try to detect noncontiguous UID/GID ranges. Signed-off-by: "Eric W. Biederman" Signed-off-by: Kevin Cernekee Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/xt_owner.c b/net/netfilter/xt_owner.c index 1302b47..a20e731 100644 --- a/net/netfilter/xt_owner.c +++ b/net/netfilter/xt_owner.c @@ -21,11 +21,39 @@ static int owner_check(const struct xt_mtchk_param *par) { struct xt_owner_match_info *info = par->matchinfo; + struct net *net = par->net; - /* For now only allow adding matches from the initial user namespace */ + /* Only allow the common case where the userns of the writer + * matches the userns of the network namespace. + */ if ((info->match & (XT_OWNER_UID|XT_OWNER_GID)) && - (current_user_ns() != &init_user_ns)) + (current_user_ns() != net->user_ns)) return -EINVAL; + + /* Ensure the uids are valid */ + if (info->match & XT_OWNER_UID) { + kuid_t uid_min = make_kuid(net->user_ns, info->uid_min); + kuid_t uid_max = make_kuid(net->user_ns, info->uid_max); + + if (!uid_valid(uid_min) || !uid_valid(uid_max) || + (info->uid_max < info->uid_min) || + uid_lt(uid_max, uid_min)) { + return -EINVAL; + } + } + + /* Ensure the gids are valid */ + if (info->match & XT_OWNER_GID) { + kgid_t gid_min = make_kgid(net->user_ns, info->gid_min); + kgid_t gid_max = make_kgid(net->user_ns, info->gid_max); + + if (!gid_valid(gid_min) || !gid_valid(gid_max) || + (info->gid_max < info->gid_min) || + gid_lt(gid_max, gid_min)) { + return -EINVAL; + } + } + return 0; } @@ -35,6 +63,7 @@ owner_mt(const struct sk_buff *skb, struct xt_action_param *par) const struct xt_owner_match_info *info = par->matchinfo; const struct file *filp; struct sock *sk = skb_to_full_sk(skb); + struct net *net = par->net; if (sk == NULL || sk->sk_socket == NULL) return (info->match ^ info->invert) == 0; @@ -51,8 +80,8 @@ owner_mt(const struct sk_buff *skb, struct xt_action_param *par) (XT_OWNER_UID | XT_OWNER_GID)) == 0; if (info->match & XT_OWNER_UID) { - kuid_t uid_min = make_kuid(&init_user_ns, info->uid_min); - kuid_t uid_max = make_kuid(&init_user_ns, info->uid_max); + kuid_t uid_min = make_kuid(net->user_ns, info->uid_min); + kuid_t uid_max = make_kuid(net->user_ns, info->uid_max); if ((uid_gte(filp->f_cred->fsuid, uid_min) && uid_lte(filp->f_cred->fsuid, uid_max)) ^ !(info->invert & XT_OWNER_UID)) @@ -60,8 +89,8 @@ owner_mt(const struct sk_buff *skb, struct xt_action_param *par) } if (info->match & XT_OWNER_GID) { - kgid_t gid_min = make_kgid(&init_user_ns, info->gid_min); - kgid_t gid_max = make_kgid(&init_user_ns, info->gid_max); + kgid_t gid_min = make_kgid(net->user_ns, info->gid_min); + kgid_t gid_max = make_kgid(net->user_ns, info->gid_max); if ((gid_gte(filp->f_cred->fsgid, gid_min) && gid_lte(filp->f_cred->fsgid, gid_max)) ^ !(info->invert & XT_OWNER_GID)) -- cgit v0.10.2 From 8658aaf2539795e0b58fe101ae61d989f81700ff Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Tue, 21 Jun 2016 01:16:31 +0100 Subject: ti_cpsw: Check for disabled child nodes Dual MAC devices don't necessarily have both MACs wired up, so ignore those that are disabled. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 8327328..9c924f1 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -2014,7 +2014,7 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data, if (ret) dev_warn(&pdev->dev, "Doesn't have any child node\n"); - for_each_child_of_node(node, slave_node) { + for_each_available_child_of_node(node, slave_node) { struct cpsw_slave_data *slave_data = data->slave_data + i; const void *mac_addr = NULL; int lenp; -- cgit v0.10.2 From d1bd330a229fc8a69f0e7532138dfd42b4542fd4 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Tue, 21 Jun 2016 01:17:17 +0100 Subject: of_mdio: Enable fixed PHY support if driver is a module The fixed_phy driver doesn't have to be built-in, and it's important that of_mdio supports it even if it's a module. Signed-off-by: Ben Hutchings Acked-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index e051e1b..de68707 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -361,7 +361,7 @@ struct phy_device *of_phy_attach(struct net_device *dev, } EXPORT_SYMBOL(of_phy_attach); -#if defined(CONFIG_FIXED_PHY) +#if IS_ENABLED(CONFIG_FIXED_PHY) /* * of_phy_is_fixed_link() and of_phy_register_fixed_link() must * support two DT bindings: diff --git a/include/linux/of_mdio.h b/include/linux/of_mdio.h index 8f2237e..6c8cb9a 100644 --- a/include/linux/of_mdio.h +++ b/include/linux/of_mdio.h @@ -69,7 +69,7 @@ static inline int of_mdio_parse_addr(struct device *dev, } #endif /* CONFIG_OF */ -#if defined(CONFIG_OF) && defined(CONFIG_FIXED_PHY) +#if defined(CONFIG_OF) && IS_ENABLED(CONFIG_FIXED_PHY) extern int of_phy_register_fixed_link(struct device_node *np); extern bool of_phy_is_fixed_link(struct device_node *np); #else -- cgit v0.10.2 From 51dca8a1cfd917e1b53b118ddd3fbe5eb1b6a9a8 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 20 Jun 2016 18:26:53 -0700 Subject: net: dsa: b53: Fix statistics readings Due to a typo we would always be using the MIB counter width of the first element of the counter array instead of the current element, and we would always be accessing the register statistics with a 64-bits read, while some could be 32-bits. This got unnoticed in testing with MDIO and SRAB which tolerate doing this, but testing with the SPI bus revealed bogus values being returned. Fix this by using the proper iterator here. Fixes: 967dd82ffc52 ("net: dsa: b53: Add support for Broadcom RoboSwitch") Reported-by: Jonas Gorski Signed-off-by: Florian Fainelli Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 5321083..444de66 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -679,7 +679,7 @@ static void b53_get_ethtool_stats(struct dsa_switch *ds, int port, for (i = 0; i < mib_size; i++) { s = &mibs[i]; - if (mibs->size == 8) { + if (s->size == 8) { b53_read64(dev, B53_MIB_PAGE(port), s->offset, &val); } else { u32 val32; -- cgit v0.10.2 From af7d5185263133f859dd4f35d45594deef9db854 Mon Sep 17 00:00:00 2001 From: Rana Shahout Date: Tue, 21 Jun 2016 12:43:59 +0300 Subject: net/mlx4_en: Add DCB PFC support through CEE netlink commands This patch adds support for reading and updating priority flow control (PFC) attributes in the driver via netlink. Signed-off-by: Rana Shahout Signed-off-by: Eran Ben Elisha Signed-off-by: Eugenia Emantayev Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c index f01918c..99c6bbd 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c @@ -37,6 +37,11 @@ #include "mlx4_en.h" #include "fw_qos.h" +enum { + MLX4_CEE_STATE_DOWN = 0, + MLX4_CEE_STATE_UP = 1, +}; + /* Definitions for QCN */ @@ -80,13 +85,202 @@ struct mlx4_congestion_control_mb_prio_802_1_qau_statistics { __be32 reserved3[4]; }; +static u8 mlx4_en_dcbnl_getcap(struct net_device *dev, int capid, u8 *cap) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + + switch (capid) { + case DCB_CAP_ATTR_PFC: + *cap = true; + break; + case DCB_CAP_ATTR_DCBX: + *cap = priv->cee_params.dcbx_cap; + break; + case DCB_CAP_ATTR_PFC_TCS: + *cap = 1 << mlx4_max_tc(priv->mdev->dev); + break; + default: + *cap = false; + break; + } + + return 0; +} + +static u8 mlx4_en_dcbnl_getpfcstate(struct net_device *netdev) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + return priv->cee_params.dcb_cfg.pfc_state; +} + +static void mlx4_en_dcbnl_setpfcstate(struct net_device *netdev, u8 state) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + priv->cee_params.dcb_cfg.pfc_state = state; +} + +static void mlx4_en_dcbnl_get_pfc_cfg(struct net_device *netdev, int priority, + u8 *setting) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + *setting = priv->cee_params.dcb_cfg.tc_config[priority].dcb_pfc; +} + +static void mlx4_en_dcbnl_set_pfc_cfg(struct net_device *netdev, int priority, + u8 setting) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + priv->cee_params.dcb_cfg.tc_config[priority].dcb_pfc = setting; + priv->cee_params.dcb_cfg.pfc_state = true; +} + +static int mlx4_en_dcbnl_getnumtcs(struct net_device *netdev, int tcid, u8 *num) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + if (!(priv->flags & MLX4_EN_FLAG_DCB_ENABLED)) + return -EINVAL; + + if (tcid == DCB_NUMTCS_ATTR_PFC) + *num = mlx4_max_tc(priv->mdev->dev); + else + *num = 0; + + return 0; +} + +static u8 mlx4_en_dcbnl_set_all(struct net_device *netdev) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + struct mlx4_en_dev *mdev = priv->mdev; + struct mlx4_en_cee_config *dcb_cfg = &priv->cee_params.dcb_cfg; + int err = 0; + + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return -EINVAL; + + if (dcb_cfg->pfc_state) { + int tc; + + priv->prof->rx_pause = 0; + priv->prof->tx_pause = 0; + for (tc = 0; tc < CEE_DCBX_MAX_PRIO; tc++) { + u8 tc_mask = 1 << tc; + + switch (dcb_cfg->tc_config[tc].dcb_pfc) { + case pfc_disabled: + priv->prof->tx_ppp &= ~tc_mask; + priv->prof->rx_ppp &= ~tc_mask; + break; + case pfc_enabled_full: + priv->prof->tx_ppp |= tc_mask; + priv->prof->rx_ppp |= tc_mask; + break; + case pfc_enabled_tx: + priv->prof->tx_ppp |= tc_mask; + priv->prof->rx_ppp &= ~tc_mask; + break; + case pfc_enabled_rx: + priv->prof->tx_ppp &= ~tc_mask; + priv->prof->rx_ppp |= tc_mask; + break; + default: + break; + } + } + en_dbg(DRV, priv, "Set pfc on\n"); + } else { + priv->prof->rx_pause = 1; + priv->prof->tx_pause = 1; + en_dbg(DRV, priv, "Set pfc off\n"); + } + + err = mlx4_SET_PORT_general(mdev->dev, priv->port, + priv->rx_skb_size + ETH_FCS_LEN, + priv->prof->tx_pause, + priv->prof->tx_ppp, + priv->prof->rx_pause, + priv->prof->rx_ppp); + if (err) + en_err(priv, "Failed setting pause params\n"); + return err; +} + +static u8 mlx4_en_dcbnl_get_state(struct net_device *dev) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + + if (priv->flags & MLX4_EN_FLAG_DCB_ENABLED) + return MLX4_CEE_STATE_UP; + + return MLX4_CEE_STATE_DOWN; +} + +static u8 mlx4_en_dcbnl_set_state(struct net_device *dev, u8 state) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + int num_tcs = 0; + + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return 1; + + if (!!(state) == !!(priv->flags & MLX4_EN_FLAG_DCB_ENABLED)) + return 0; + + if (state) { + priv->flags |= MLX4_EN_FLAG_DCB_ENABLED; + num_tcs = IEEE_8021QAZ_MAX_TCS; + } else { + priv->flags &= ~MLX4_EN_FLAG_DCB_ENABLED; + } + + return mlx4_en_setup_tc(dev, num_tcs); +} + +/* On success returns a non-zero 802.1p user priority bitmap + * otherwise returns 0 as the invalid user priority bitmap to + * indicate an error. + */ +static int mlx4_en_dcbnl_getapp(struct net_device *netdev, u8 idtype, u16 id) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + struct dcb_app app = { + .selector = idtype, + .protocol = id, + }; + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return 0; + + return dcb_getapp(netdev, &app); +} + +static int mlx4_en_dcbnl_setapp(struct net_device *netdev, u8 idtype, + u16 id, u8 up) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + struct dcb_app app; + + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return -EINVAL; + + memset(&app, 0, sizeof(struct dcb_app)); + app.selector = idtype; + app.protocol = id; + app.priority = up; + + return dcb_setapp(netdev, &app); +} + static int mlx4_en_dcbnl_ieee_getets(struct net_device *dev, struct ieee_ets *ets) { struct mlx4_en_priv *priv = netdev_priv(dev); struct ieee_ets *my_ets = &priv->ets; - /* No IEEE PFC settings available */ if (!my_ets) return -EINVAL; @@ -237,18 +431,51 @@ static int mlx4_en_dcbnl_ieee_setpfc(struct net_device *dev, static u8 mlx4_en_dcbnl_getdcbx(struct net_device *dev) { - return DCB_CAP_DCBX_HOST | DCB_CAP_DCBX_VER_IEEE; + struct mlx4_en_priv *priv = netdev_priv(dev); + + return priv->cee_params.dcbx_cap; } static u8 mlx4_en_dcbnl_setdcbx(struct net_device *dev, u8 mode) { + struct mlx4_en_priv *priv = netdev_priv(dev); + struct ieee_ets ets = {0}; + struct ieee_pfc pfc = {0}; + + if (mode == priv->cee_params.dcbx_cap) + return 0; + if ((mode & DCB_CAP_DCBX_LLD_MANAGED) || - (mode & DCB_CAP_DCBX_VER_CEE) || - !(mode & DCB_CAP_DCBX_VER_IEEE) || + ((mode & DCB_CAP_DCBX_VER_IEEE) && + (mode & DCB_CAP_DCBX_VER_CEE)) || !(mode & DCB_CAP_DCBX_HOST)) - return 1; + goto err; + + priv->cee_params.dcbx_cap = mode; + + ets.ets_cap = IEEE_8021QAZ_MAX_TCS; + pfc.pfc_cap = IEEE_8021QAZ_MAX_TCS; + + if (mode & DCB_CAP_DCBX_VER_IEEE) { + if (mlx4_en_dcbnl_ieee_setets(dev, &ets)) + goto err; + if (mlx4_en_dcbnl_ieee_setpfc(dev, &pfc)) + goto err; + } else if (mode & DCB_CAP_DCBX_VER_CEE) { + if (mlx4_en_dcbnl_set_all(dev)) + goto err; + } else { + if (mlx4_en_dcbnl_ieee_setets(dev, &ets)) + goto err; + if (mlx4_en_dcbnl_ieee_setpfc(dev, &pfc)) + goto err; + if (mlx4_en_setup_tc(dev, 0)) + goto err; + } return 0; +err: + return 1; } #define MLX4_RATELIMIT_UNITS_IN_KB 100000 /* rate-limit HW unit in Kbps */ @@ -463,24 +690,46 @@ static int mlx4_en_dcbnl_ieee_getqcnstats(struct net_device *dev, } const struct dcbnl_rtnl_ops mlx4_en_dcbnl_ops = { - .ieee_getets = mlx4_en_dcbnl_ieee_getets, - .ieee_setets = mlx4_en_dcbnl_ieee_setets, - .ieee_getmaxrate = mlx4_en_dcbnl_ieee_getmaxrate, - .ieee_setmaxrate = mlx4_en_dcbnl_ieee_setmaxrate, - .ieee_getpfc = mlx4_en_dcbnl_ieee_getpfc, - .ieee_setpfc = mlx4_en_dcbnl_ieee_setpfc, + .ieee_getets = mlx4_en_dcbnl_ieee_getets, + .ieee_setets = mlx4_en_dcbnl_ieee_setets, + .ieee_getmaxrate = mlx4_en_dcbnl_ieee_getmaxrate, + .ieee_setmaxrate = mlx4_en_dcbnl_ieee_setmaxrate, + .ieee_getqcn = mlx4_en_dcbnl_ieee_getqcn, + .ieee_setqcn = mlx4_en_dcbnl_ieee_setqcn, + .ieee_getqcnstats = mlx4_en_dcbnl_ieee_getqcnstats, + .ieee_getpfc = mlx4_en_dcbnl_ieee_getpfc, + .ieee_setpfc = mlx4_en_dcbnl_ieee_setpfc, + + .getstate = mlx4_en_dcbnl_get_state, + .setstate = mlx4_en_dcbnl_set_state, + .getpfccfg = mlx4_en_dcbnl_get_pfc_cfg, + .setpfccfg = mlx4_en_dcbnl_set_pfc_cfg, + .setall = mlx4_en_dcbnl_set_all, + .getcap = mlx4_en_dcbnl_getcap, + .getnumtcs = mlx4_en_dcbnl_getnumtcs, + .getpfcstate = mlx4_en_dcbnl_getpfcstate, + .setpfcstate = mlx4_en_dcbnl_setpfcstate, + .getapp = mlx4_en_dcbnl_getapp, + .setapp = mlx4_en_dcbnl_setapp, .getdcbx = mlx4_en_dcbnl_getdcbx, .setdcbx = mlx4_en_dcbnl_setdcbx, - .ieee_getqcn = mlx4_en_dcbnl_ieee_getqcn, - .ieee_setqcn = mlx4_en_dcbnl_ieee_setqcn, - .ieee_getqcnstats = mlx4_en_dcbnl_ieee_getqcnstats, }; const struct dcbnl_rtnl_ops mlx4_en_dcbnl_pfc_ops = { .ieee_getpfc = mlx4_en_dcbnl_ieee_getpfc, .ieee_setpfc = mlx4_en_dcbnl_ieee_setpfc, + .setstate = mlx4_en_dcbnl_set_state, + .getpfccfg = mlx4_en_dcbnl_get_pfc_cfg, + .setpfccfg = mlx4_en_dcbnl_set_pfc_cfg, + .setall = mlx4_en_dcbnl_set_all, + .getnumtcs = mlx4_en_dcbnl_getnumtcs, + .getpfcstate = mlx4_en_dcbnl_getpfcstate, + .setpfcstate = mlx4_en_dcbnl_setpfcstate, + .getapp = mlx4_en_dcbnl_getapp, + .setapp = mlx4_en_dcbnl_setapp, + .getdcbx = mlx4_en_dcbnl_getdcbx, .setdcbx = mlx4_en_dcbnl_setdcbx, }; diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 8e318d2..d42083a 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -67,6 +67,17 @@ int mlx4_en_setup_tc(struct net_device *dev, u8 up) offset += priv->num_tx_rings_p_up; } +#ifdef CONFIG_MLX4_EN_DCB + if (!mlx4_is_slave(priv->mdev->dev)) { + if (up) { + priv->flags |= MLX4_EN_FLAG_DCB_ENABLED; + } else { + priv->flags &= ~MLX4_EN_FLAG_DCB_ENABLED; + priv->cee_params.dcb_cfg.pfc_state = false; + } + } +#endif /* CONFIG_MLX4_EN_DCB */ + return 0; } @@ -2815,6 +2826,9 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, struct mlx4_en_priv *priv; int i; int err; +#ifdef CONFIG_MLX4_EN_DCB + struct tc_configuration *tc; +#endif dev = alloc_etherdev_mqs(sizeof(struct mlx4_en_priv), MAX_TX_RINGS, MAX_RX_RINGS); @@ -2881,6 +2895,17 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, priv->msg_enable = MLX4_EN_MSG_LEVEL; #ifdef CONFIG_MLX4_EN_DCB if (!mlx4_is_slave(priv->mdev->dev)) { + priv->cee_params.dcbx_cap = DCB_CAP_DCBX_VER_CEE | + DCB_CAP_DCBX_HOST | + DCB_CAP_DCBX_VER_IEEE; + priv->flags |= MLX4_EN_DCB_ENABLED; + priv->cee_params.dcb_cfg.pfc_state = false; + + for (i = 0; i < MLX4_EN_NUM_UP; i++) { + tc = &priv->cee_params.dcb_cfg.tc_config[i]; + tc->dcb_pfc = pfc_disabled; + } + if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ETS_CFG) { dev->dcbnl_ops = &mlx4_en_dcbnl_ops; } else { diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index e970945..f4497cf 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -1128,6 +1128,7 @@ int mlx4_QUERY_PORT(struct mlx4_dev *dev, int port, struct mlx4_port_cap *port_c port_cap->max_pkeys = 1 << (field & 0xf); MLX4_GET(field, outbox, QUERY_PORT_MAX_VL_OFFSET); port_cap->max_vl = field & 0xf; + port_cap->max_tc_eth = field >> 4; MLX4_GET(field, outbox, QUERY_PORT_MAX_MACVLAN_OFFSET); port_cap->log_max_macs = field & 0xf; port_cap->log_max_vlans = field >> 4; diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.h b/drivers/net/ethernet/mellanox/mlx4/fw.h index 7ea258a..cdbd76f 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.h +++ b/drivers/net/ethernet/mellanox/mlx4/fw.h @@ -53,6 +53,7 @@ struct mlx4_port_cap { int ib_mtu; int max_port_width; int max_vl; + int max_tc_eth; int max_gids; int max_pkeys; u64 def_mac; diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index 12c77a7..3564aad 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -292,6 +292,7 @@ static int _mlx4_dev_port(struct mlx4_dev *dev, int port, dev->caps.pkey_table_len[port] = port_cap->max_pkeys; dev->caps.port_width_cap[port] = port_cap->max_port_width; dev->caps.eth_mtu_cap[port] = port_cap->eth_mtu; + dev->caps.max_tc_eth = port_cap->max_tc_eth; dev->caps.def_mac[port] = port_cap->def_mac; dev->caps.supported_type[port] = port_cap->supported_port_types; dev->caps.suggested_type[port] = port_cap->suggested_type; diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 6b3b0fe..d39bf59 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -448,6 +448,27 @@ struct mlx4_en_frag_info { #define MLX4_EN_TC_ETS 7 +enum dcb_pfc_type { + pfc_disabled = 0, + pfc_enabled_full, + pfc_enabled_tx, + pfc_enabled_rx +}; + +struct tc_configuration { + enum dcb_pfc_type dcb_pfc; +}; + +struct mlx4_en_cee_config { + bool pfc_state; + struct tc_configuration tc_config[MLX4_EN_NUM_UP]; +}; + +struct mlx4_en_cee_params { + u8 dcbx_cap; + struct mlx4_en_cee_config dcb_cfg; +}; + #endif struct ethtool_flow_id { @@ -467,6 +488,9 @@ enum { MLX4_EN_FLAG_RX_FILTER_NEEDED = (1 << 3), MLX4_EN_FLAG_FORCE_PROMISC = (1 << 4), MLX4_EN_FLAG_RX_CSUM_NON_TCP_UDP = (1 << 5), +#ifdef CONFIG_MLX4_EN_DCB + MLX4_EN_FLAG_DCB_ENABLED = (1 << 6), +#endif }; #define PORT_BEACON_MAX_LIMIT (65535) @@ -568,9 +592,11 @@ struct mlx4_en_priv { u32 counter_index; #ifdef CONFIG_MLX4_EN_DCB +#define MLX4_EN_DCB_ENABLED 0x3 struct ieee_ets ets; u16 maxrate[IEEE_8021QAZ_MAX_TCS]; enum dcbnl_cndd_states cndd_state[IEEE_8021QAZ_MAX_TCS]; + struct mlx4_en_cee_params cee_params; #endif #ifdef CONFIG_RFS_ACCEL spinlock_t filters_lock; diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c index 087b23b..3d2095e 100644 --- a/drivers/net/ethernet/mellanox/mlx4/port.c +++ b/drivers/net/ethernet/mellanox/mlx4/port.c @@ -52,6 +52,7 @@ #define MLX4_FLAG_V_IGNORE_FCS_MASK 0x2 #define MLX4_IGNORE_FCS_MASK 0x1 +#define MLNX4_TX_MAX_NUMBER 8 void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table) { @@ -2015,3 +2016,14 @@ out: return ret; } EXPORT_SYMBOL(mlx4_get_module_info); + +int mlx4_max_tc(struct mlx4_dev *dev) +{ + u8 num_tc = dev->caps.max_tc_eth; + + if (!num_tc) + num_tc = MLNX4_TX_MAX_NUMBER; + + return num_tc; +} +EXPORT_SYMBOL(mlx4_max_tc); diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index 80dec87..4dbc145 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -535,6 +535,7 @@ struct mlx4_caps { int max_rq_desc_sz; int max_qp_init_rdma; int max_qp_dest_rdma; + int max_tc_eth; u32 *qp0_qkey; u32 *qp0_proxy; u32 *qp1_proxy; @@ -1494,6 +1495,7 @@ int mlx4_mr_rereg_mem_write(struct mlx4_dev *dev, struct mlx4_mr *mr, int mlx4_get_module_info(struct mlx4_dev *dev, u8 port, u16 offset, u16 size, u8 *data); +int mlx4_max_tc(struct mlx4_dev *dev); /* Returns true if running in low memory profile (kdump kernel) */ static inline bool mlx4_low_memory_profile(void) -- cgit v0.10.2 From 722003ac40c2c397bd5bc2b714125bc82ab27043 Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Tue, 21 Jun 2016 09:36:21 -0400 Subject: qed: Add support for coalescing config read/update. This patch adds support for configuring the device tx/rx coalescing timeout values in the order of micro seconds. It also adds APIs for upper layer drivers for reading/updating the coalescing values. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index e45cff4..b26fe26 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -2222,6 +2222,110 @@ int qed_fw_rss_eng(struct qed_hwfn *p_hwfn, return 0; } +static int qed_set_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u32 hw_addr, void *p_eth_qzone, + size_t eth_qzone_size, u8 timeset) +{ + struct coalescing_timeset *p_coal_timeset; + + if (p_hwfn->cdev->int_coalescing_mode != QED_COAL_MODE_ENABLE) { + DP_NOTICE(p_hwfn, "Coalescing configuration not enabled\n"); + return -EINVAL; + } + + p_coal_timeset = p_eth_qzone; + memset(p_coal_timeset, 0, eth_qzone_size); + SET_FIELD(p_coal_timeset->value, COALESCING_TIMESET_TIMESET, timeset); + SET_FIELD(p_coal_timeset->value, COALESCING_TIMESET_VALID, 1); + qed_memcpy_to(p_hwfn, p_ptt, hw_addr, p_eth_qzone, eth_qzone_size); + + return 0; +} + +int qed_set_rxq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id) +{ + struct ustorm_eth_queue_zone eth_qzone; + u8 timeset, timer_res; + u16 fw_qid = 0; + u32 address; + int rc; + + /* Coalesce = (timeset << timer-resolution), timeset is 7bit wide */ + if (coalesce <= 0x7F) { + timer_res = 0; + } else if (coalesce <= 0xFF) { + timer_res = 1; + } else if (coalesce <= 0x1FF) { + timer_res = 2; + } else { + DP_ERR(p_hwfn, "Invalid coalesce value - %d\n", coalesce); + return -EINVAL; + } + timeset = (u8)(coalesce >> timer_res); + + rc = qed_fw_l2_queue(p_hwfn, (u16)qid, &fw_qid); + if (rc) + return rc; + + rc = qed_int_set_timer_res(p_hwfn, p_ptt, timer_res, sb_id, false); + if (rc) + goto out; + + address = BAR0_MAP_REG_USDM_RAM + USTORM_ETH_QUEUE_ZONE_OFFSET(fw_qid); + + rc = qed_set_coalesce(p_hwfn, p_ptt, address, ð_qzone, + sizeof(struct ustorm_eth_queue_zone), timeset); + if (rc) + goto out; + + p_hwfn->cdev->rx_coalesce_usecs = coalesce; +out: + return rc; +} + +int qed_set_txq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id) +{ + struct xstorm_eth_queue_zone eth_qzone; + u8 timeset, timer_res; + u16 fw_qid = 0; + u32 address; + int rc; + + /* Coalesce = (timeset << timer-resolution), timeset is 7bit wide */ + if (coalesce <= 0x7F) { + timer_res = 0; + } else if (coalesce <= 0xFF) { + timer_res = 1; + } else if (coalesce <= 0x1FF) { + timer_res = 2; + } else { + DP_ERR(p_hwfn, "Invalid coalesce value - %d\n", coalesce); + return -EINVAL; + } + timeset = (u8)(coalesce >> timer_res); + + rc = qed_fw_l2_queue(p_hwfn, (u16)qid, &fw_qid); + if (rc) + return rc; + + rc = qed_int_set_timer_res(p_hwfn, p_ptt, timer_res, sb_id, true); + if (rc) + goto out; + + address = BAR0_MAP_REG_XSDM_RAM + XSTORM_ETH_QUEUE_ZONE_OFFSET(fw_qid); + + rc = qed_set_coalesce(p_hwfn, p_ptt, address, ð_qzone, + sizeof(struct xstorm_eth_queue_zone), timeset); + if (rc) + goto out; + + p_hwfn->cdev->tx_coalesce_usecs = coalesce; +out: + return rc; +} + /* Calculate final WFQ values for all vports and configure them. * After this configuration each vport will have * approx min rate = min_pf_rate * (vport_wfq / QED_WFQ_UNIT) diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h index f810ce4..343bb03 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h +++ b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h @@ -212,6 +212,20 @@ qed_dmae_host2grc(struct qed_hwfn *p_hwfn, u32 size_in_dwords, u32 flags); + /** + * @brief qed_dmae_grc2host - Read data from dmae data offset + * to source address using the given ptt + * + * @param p_ptt + * @param grc_addr (dmae_data_offset) + * @param dest_addr + * @param size_in_dwords + * @param flags - one of the flags defined above + */ +int qed_dmae_grc2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u32 grc_addr, dma_addr_t dest_addr, u32 size_in_dwords, + u32 flags); + /** * @brief qed_dmae_host2host - copy data from to source address * to a destination adress (for SRIOV) using the given ptt @@ -308,4 +322,37 @@ int qed_fw_rss_eng(struct qed_hwfn *p_hwfn, int qed_final_cleanup(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u16 id, bool is_vf); +/** + * @brief qed_set_rxq_coalesce - Configure coalesce parameters for an Rx queue + * The fact that we can configure coalescing to up to 511, but on varying + * accuracy [the bigger the value the less accurate] up to a mistake of 3usec + * for the highest values. + * + * @param p_hwfn + * @param p_ptt + * @param coalesce - Coalesce value in micro seconds. + * @param qid - Queue index. + * @param qid - SB Id + * + * @return int + */ +int qed_set_rxq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id); + +/** + * @brief qed_set_txq_coalesce - Configure coalesce parameters for a Tx queue + * While the API allows setting coalescing per-qid, all tx queues sharing a + * SB should be in same range [i.e., either 0-0x7f, 0x80-0xff or 0x100-0x1ff] + * otherwise configuration would break. + * + * @param p_hwfn + * @param p_ptt + * @param coalesce - Coalesce value in micro seconds. + * @param qid - Queue index. + * @param qid - SB Id + * + * @return int + */ +int qed_set_txq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id); #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c index 2693c30..e178853 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.c +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c @@ -769,6 +769,29 @@ int qed_dmae_host2grc(struct qed_hwfn *p_hwfn, } int +qed_dmae_grc2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u32 grc_addr, + dma_addr_t dest_addr, u32 size_in_dwords, u32 flags) +{ + u32 grc_addr_in_dw = grc_addr / sizeof(u32); + struct qed_dmae_params params; + int rc; + + memset(¶ms, 0, sizeof(struct qed_dmae_params)); + params.flags = flags; + + mutex_lock(&p_hwfn->dmae_info.mutex); + + rc = qed_dmae_execute_command(p_hwfn, p_ptt, grc_addr_in_dw, + dest_addr, QED_DMAE_ADDRESS_GRC, + QED_DMAE_ADDRESS_HOST_VIRT, + size_in_dwords, ¶ms); + + mutex_unlock(&p_hwfn->dmae_info.mutex); + + return rc; +} + +int qed_dmae_host2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, dma_addr_t source_addr, diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c index 09a6ad3..8fa50fa 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.c +++ b/drivers/net/ethernet/qlogic/qed/qed_int.c @@ -2418,6 +2418,7 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, { struct qed_dev *cdev = p_hwfn->cdev; u32 cau_state; + u8 timer_res; memset(p_sb_entry, 0, sizeof(*p_sb_entry)); @@ -2443,6 +2444,23 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, cdev->tx_coalesce_usecs = QED_CAU_DEF_TX_USECS; } + /* Coalesce = (timeset << timer-res), timeset is 7bit wide */ + if (cdev->rx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (cdev->rx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + SET_FIELD(p_sb_entry->params, CAU_SB_ENTRY_TIMER_RES0, timer_res); + + if (cdev->tx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (cdev->tx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + SET_FIELD(p_sb_entry->params, CAU_SB_ENTRY_TIMER_RES1, timer_res); + SET_FIELD(p_sb_entry->data, CAU_SB_ENTRY_STATE0, cau_state); SET_FIELD(p_sb_entry->data, CAU_SB_ENTRY_STATE1, cau_state); } @@ -2484,17 +2502,28 @@ void qed_int_cau_conf_sb(struct qed_hwfn *p_hwfn, /* Configure pi coalescing if set */ if (p_hwfn->cdev->int_coalescing_mode == QED_COAL_MODE_ENABLE) { - u8 timeset = p_hwfn->cdev->rx_coalesce_usecs >> - (QED_CAU_DEF_RX_TIMER_RES + 1); + u8 timeset, timer_res; u8 num_tc = 1, i; + /* timeset = (coalesce >> timer-res), timeset is 7bit wide */ + if (p_hwfn->cdev->rx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (p_hwfn->cdev->rx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + timeset = (u8)(p_hwfn->cdev->rx_coalesce_usecs >> timer_res); qed_int_cau_conf_pi(p_hwfn, p_ptt, igu_sb_id, RX_PI, QED_COAL_RX_STATE_MACHINE, timeset); - timeset = p_hwfn->cdev->tx_coalesce_usecs >> - (QED_CAU_DEF_TX_TIMER_RES + 1); - + if (p_hwfn->cdev->tx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (p_hwfn->cdev->tx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + timeset = (u8)(p_hwfn->cdev->tx_coalesce_usecs >> timer_res); for (i = 0; i < num_tc; i++) { qed_int_cau_conf_pi(p_hwfn, p_ptt, igu_sb_id, TX_PI(i), @@ -3199,3 +3228,39 @@ void qed_int_disable_post_isr_release(struct qed_dev *cdev) for_each_hwfn(cdev, i) cdev->hwfns[i].b_int_requested = false; } + +int qed_int_set_timer_res(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u8 timer_res, u16 sb_id, bool tx) +{ + struct cau_sb_entry sb_entry; + int rc; + + if (!p_hwfn->hw_init_done) { + DP_ERR(p_hwfn, "hardware not initialized yet\n"); + return -EINVAL; + } + + rc = qed_dmae_grc2host(p_hwfn, p_ptt, CAU_REG_SB_VAR_MEMORY + + sb_id * sizeof(u64), + (u64)(uintptr_t)&sb_entry, 2, 0); + if (rc) { + DP_ERR(p_hwfn, "dmae_grc2host failed %d\n", rc); + return rc; + } + + if (tx) + SET_FIELD(sb_entry.params, CAU_SB_ENTRY_TIMER_RES1, timer_res); + else + SET_FIELD(sb_entry.params, CAU_SB_ENTRY_TIMER_RES0, timer_res); + + rc = qed_dmae_host2grc(p_hwfn, p_ptt, + (u64)(uintptr_t)&sb_entry, + CAU_REG_SB_VAR_MEMORY + + sb_id * sizeof(u64), 2, 0); + if (rc) { + DP_ERR(p_hwfn, "dmae_host2grc failed %d\n", rc); + return rc; + } + + return rc; +} diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.h b/drivers/net/ethernet/qlogic/qed/qed_int.h index 20b4686..0948be6 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.h +++ b/drivers/net/ethernet/qlogic/qed/qed_int.h @@ -389,6 +389,9 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, u16 vf_number, u8 vf_valid); +int qed_int_set_timer_res(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u8 timer_res, u16 sb_id, bool tx); + #define QED_MAPPING_MEMORY_SIZE(dev) (NUM_OF_SBS(dev)) #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 6c4606b..e32ee57 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -1303,6 +1303,38 @@ static int qed_drain(struct qed_dev *cdev) return 0; } +static void qed_get_coalesce(struct qed_dev *cdev, u16 *rx_coal, u16 *tx_coal) +{ + *rx_coal = cdev->rx_coalesce_usecs; + *tx_coal = cdev->tx_coalesce_usecs; +} + +static int qed_set_coalesce(struct qed_dev *cdev, u16 rx_coal, u16 tx_coal, + u8 qid, u16 sb_id) +{ + struct qed_hwfn *hwfn; + struct qed_ptt *ptt; + int hwfn_index; + int status = 0; + + hwfn_index = qid % cdev->num_hwfns; + hwfn = &cdev->hwfns[hwfn_index]; + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EAGAIN; + + status = qed_set_rxq_coalesce(hwfn, ptt, rx_coal, + qid / cdev->num_hwfns, sb_id); + if (status) + goto out; + status = qed_set_txq_coalesce(hwfn, ptt, tx_coal, + qid / cdev->num_hwfns, sb_id); +out: + qed_ptt_release(hwfn, ptt); + + return status; +} + static int qed_set_led(struct qed_dev *cdev, enum qed_led_mode mode) { struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); @@ -1349,5 +1381,7 @@ const struct qed_common_ops qed_common_ops_pass = { .update_msglvl = &qed_init_dp, .chain_alloc = &qed_chain_alloc, .chain_free = &qed_chain_free, + .get_coalesce = &qed_get_coalesce, + .set_coalesce = &qed_set_coalesce, .set_led = &qed_set_led, }; diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h index aa08ddb..f6b86ca 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h +++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h @@ -80,6 +80,8 @@ 0x1f00000UL #define BAR0_MAP_REG_TSDM_RAM \ 0x1c80000UL +#define BAR0_MAP_REG_XSDM_RAM \ + 0x1e00000UL #define NIG_REG_RX_LLH_BRB_GATE_DNTFWD_PERPF \ 0x5011f4UL #define PRS_REG_SEARCH_TCP \ diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index e1d5122..b1e3c57 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -489,6 +489,30 @@ struct qed_common_ops { struct qed_chain *p_chain); /** + * @brief get_coalesce - Get coalesce parameters in usec + * + * @param cdev + * @param rx_coal - Rx coalesce value in usec + * @param tx_coal - Tx coalesce value in usec + * + */ + void (*get_coalesce)(struct qed_dev *cdev, u16 *rx_coal, u16 *tx_coal); + +/** + * @brief set_coalesce - Configure Rx coalesce value in usec + * + * @param cdev + * @param rx_coal - Rx coalesce value in usec + * @param tx_coal - Tx coalesce value in usec + * @param qid - Queue index + * @param sb_id - Status Block Id + * + * @return 0 on success, error otherwise. + */ + int (*set_coalesce)(struct qed_dev *cdev, u16 rx_coal, u16 tx_coal, + u8 qid, u16 sb_id); + +/** * @brief set_led - Configure LED mode * * @param cdev -- cgit v0.10.2 From d552fa84cb3573eb86b6722329f5e72c3ed9029e Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Tue, 21 Jun 2016 09:36:22 -0400 Subject: qede: Add support for coalescing config read/update. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index 6836d44..6228482 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -426,6 +426,57 @@ static u32 qede_get_link(struct net_device *dev) return current_link.link_up; } +static int qede_get_coalesce(struct net_device *dev, + struct ethtool_coalesce *coal) +{ + struct qede_dev *edev = netdev_priv(dev); + + memset(coal, 0, sizeof(struct ethtool_coalesce)); + edev->ops->common->get_coalesce(edev->cdev, + (u16 *)&coal->rx_coalesce_usecs, + (u16 *)&coal->tx_coalesce_usecs); + + return 0; +} + +static int qede_set_coalesce(struct net_device *dev, + struct ethtool_coalesce *coal) +{ + struct qede_dev *edev = netdev_priv(dev); + int i, rc = 0; + u16 rxc, txc; + u8 sb_id; + + if (!netif_running(dev)) { + DP_INFO(edev, "Interface is down\n"); + return -EINVAL; + } + + if (coal->rx_coalesce_usecs > QED_COALESCE_MAX || + coal->tx_coalesce_usecs > QED_COALESCE_MAX) { + DP_INFO(edev, + "Can't support requested %s coalesce value [max supported value %d]\n", + coal->rx_coalesce_usecs > QED_COALESCE_MAX ? "rx" + : "tx", + QED_COALESCE_MAX); + return -EINVAL; + } + + rxc = (u16)coal->rx_coalesce_usecs; + txc = (u16)coal->tx_coalesce_usecs; + for_each_rss(i) { + sb_id = edev->fp_array[i].sb_info->igu_sb_id; + rc = edev->ops->common->set_coalesce(edev->cdev, rxc, txc, + (u8)i, sb_id); + if (rc) { + DP_INFO(edev, "Set coalesce error, rc = %d\n", rc); + return rc; + } + } + + return rc; +} + static void qede_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ering) { @@ -1139,6 +1190,8 @@ static const struct ethtool_ops qede_ethtool_ops = { .set_msglevel = qede_set_msglevel, .nway_reset = qede_nway_reset, .get_link = qede_get_link, + .get_coalesce = qede_get_coalesce, + .set_coalesce = qede_set_coalesce, .get_ringparam = qede_get_ringparam, .set_ringparam = qede_set_ringparam, .get_pauseparam = qede_get_pauseparam, -- cgit v0.10.2 From e1dbbc5907b53d8d53c009b3cb3dd2a0366ce45c Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 20 Jun 2016 21:26:28 +0800 Subject: netfilter: nf_reject_ipv4: don't send tcp RST if the packet is non-TCP In iptables, if the user add a rule to send tcp RST and specify the non-TCP protocol, such as UDP, kernel will reject this request. But in nftables, this validity check only occurs in nft tool, i.e. only in userspace. This means that user can add such a rule like follows via nfnetlink: "nft add rule filter forward ip protocol udp reject with tcp reset" This will generate some confusing tcp RST packets. So we should send tcp RST only when it is TCP packet. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/ipv4/netfilter/nf_reject_ipv4.c b/net/ipv4/netfilter/nf_reject_ipv4.c index b6ea57e..fd82202 100644 --- a/net/ipv4/netfilter/nf_reject_ipv4.c +++ b/net/ipv4/netfilter/nf_reject_ipv4.c @@ -24,6 +24,9 @@ const struct tcphdr *nf_reject_ip_tcphdr_get(struct sk_buff *oldskb, if (ip_hdr(oldskb)->frag_off & htons(IP_OFFSET)) return NULL; + if (ip_hdr(oldskb)->protocol != IPPROTO_TCP) + return NULL; + oth = skb_header_pointer(oldskb, ip_hdrlen(oldskb), sizeof(struct tcphdr), _oth); if (oth == NULL) -- cgit v0.10.2 From 7643507fe8b5bd8ab7522f6a81058cc1209d2585 Mon Sep 17 00:00:00 2001 From: Vishwanath Pai Date: Tue, 21 Jun 2016 14:58:46 -0400 Subject: netfilter: xt_NFLOG: nflog-range does not truncate packets li->u.ulog.copy_len is currently ignored by the kernel, we should truncate the packet to either li->u.ulog.copy_len (if set) or copy_range before sending it to userspace. 0 is a valid input for copy_len, so add a new flag to indicate whether this was option was specified by the user or not. Add two flags to indicate whether nflog-size/copy_len was set or not. XT_NFLOG_F_COPY_LEN is for XT_NFLOG and NFLOG_F_COPY_LEN for nfnetlink_log On the userspace side, this was initially represented by the option nflog-range, this will be replaced by --nflog-size now. --nflog-range would still exist but does not do anything. Reported-by: Joe Dollard Reviewed-by: Josh Hunt Signed-off-by: Vishwanath Pai Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_log.h b/include/net/netfilter/nf_log.h index 57639fc..83d855b 100644 --- a/include/net/netfilter/nf_log.h +++ b/include/net/netfilter/nf_log.h @@ -12,6 +12,9 @@ #define NF_LOG_UID 0x08 /* Log UID owning local socket */ #define NF_LOG_MASK 0x0f +/* This flag indicates that copy_len field in nf_loginfo is set */ +#define NF_LOG_F_COPY_LEN 0x1 + enum nf_log_type { NF_LOG_TYPE_LOG = 0, NF_LOG_TYPE_ULOG, @@ -22,9 +25,13 @@ struct nf_loginfo { u_int8_t type; union { struct { + /* copy_len will be used iff you set + * NF_LOG_F_COPY_LEN in flags + */ u_int32_t copy_len; u_int16_t group; u_int16_t qthreshold; + u_int16_t flags; } ulog; struct { u_int8_t level; diff --git a/include/uapi/linux/netfilter/xt_NFLOG.h b/include/uapi/linux/netfilter/xt_NFLOG.h index 87b5831..f330707 100644 --- a/include/uapi/linux/netfilter/xt_NFLOG.h +++ b/include/uapi/linux/netfilter/xt_NFLOG.h @@ -6,9 +6,13 @@ #define XT_NFLOG_DEFAULT_GROUP 0x1 #define XT_NFLOG_DEFAULT_THRESHOLD 0 -#define XT_NFLOG_MASK 0x0 +#define XT_NFLOG_MASK 0x1 + +/* This flag indicates that 'len' field in xt_nflog_info is set*/ +#define XT_NFLOG_F_COPY_LEN 0x1 struct xt_nflog_info { + /* 'len' will be used iff you set XT_NFLOG_F_COPY_LEN in flags */ __u32 len; __u16 group; __u16 threshold; diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 11f81c8..cbcfdfb 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -700,10 +700,13 @@ nfulnl_log_packet(struct net *net, break; case NFULNL_COPY_PACKET: - if (inst->copy_range > skb->len) + data_len = inst->copy_range; + if ((li->u.ulog.flags & NF_LOG_F_COPY_LEN) && + (li->u.ulog.copy_len < data_len)) + data_len = li->u.ulog.copy_len; + + if (data_len > skb->len) data_len = skb->len; - else - data_len = inst->copy_range; size += nla_total_size(data_len); break; diff --git a/net/netfilter/xt_NFLOG.c b/net/netfilter/xt_NFLOG.c index a1fa2c8..018eed7 100644 --- a/net/netfilter/xt_NFLOG.c +++ b/net/netfilter/xt_NFLOG.c @@ -33,6 +33,9 @@ nflog_tg(struct sk_buff *skb, const struct xt_action_param *par) li.u.ulog.group = info->group; li.u.ulog.qthreshold = info->threshold; + if (info->flags & XT_NFLOG_F_COPY_LEN) + li.u.ulog.flags |= NF_LOG_F_COPY_LEN; + nfulnl_log_packet(net, par->family, par->hooknum, skb, par->in, par->out, &li, info->prefix); return XT_CONTINUE; -- cgit v0.10.2 From 889f7ee7c6e84251215d43cbc856ea116c72d3f2 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Sun, 12 Jun 2016 18:07:07 +0200 Subject: netfilter: nf_tables: add generic macros to check for generation mask Thus, we can reuse these to check the genmask of any object type, not only rules. This is required now that tables, chain and sets will get a generation mask field too in follow up patches. Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 0922354..d0778cb 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -970,6 +970,30 @@ static inline u8 nft_genmask_cur(const struct net *net) #define NFT_GENMASK_ANY ((1 << 0) | (1 << 1)) /* + * Generic transaction helpers + */ + +/* Check if this object is currently active. */ +#define nft_is_active(__net, __obj) \ + (((__obj)->genmask & nft_genmask_cur(__net)) == 0) + +/* Check if this object is active in the next generation. */ +#define nft_is_active_next(__net, __obj) \ + (((__obj)->genmask & nft_genmask_next(__net)) == 0) + +/* This object becomes active in the next generation. */ +#define nft_activate_next(__net, __obj) \ + (__obj)->genmask = nft_genmask_cur(__net) + +/* This object becomes inactive in the next generation. */ +#define nft_deactivate_next(__net, __obj) \ + (__obj)->genmask = nft_genmask_next(__net) + +/* After committing the ruleset, clear the stale generation bit. */ +#define nft_clear(__net, __obj) \ + (__obj)->genmask &= ~nft_genmask_next(__net) + +/* * Set element transaction helpers */ diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 4d292b9..d9f0f07 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -234,42 +234,12 @@ static int nft_delchain(struct nft_ctx *ctx) return err; } -static inline bool -nft_rule_is_active(struct net *net, const struct nft_rule *rule) -{ - return (rule->genmask & nft_genmask_cur(net)) == 0; -} - -static inline int -nft_rule_is_active_next(struct net *net, const struct nft_rule *rule) -{ - return (rule->genmask & nft_genmask_next(net)) == 0; -} - -static inline void -nft_rule_activate_next(struct net *net, struct nft_rule *rule) -{ - /* Now inactive, will be active in the future */ - rule->genmask = nft_genmask_cur(net); -} - -static inline void -nft_rule_deactivate_next(struct net *net, struct nft_rule *rule) -{ - rule->genmask = nft_genmask_next(net); -} - -static inline void nft_rule_clear(struct net *net, struct nft_rule *rule) -{ - rule->genmask &= ~nft_genmask_next(net); -} - static int nf_tables_delrule_deactivate(struct nft_ctx *ctx, struct nft_rule *rule) { /* You cannot delete the same rule twice */ - if (nft_rule_is_active_next(ctx->net, rule)) { - nft_rule_deactivate_next(ctx->net, rule); + if (nft_is_active_next(ctx->net, rule)) { + nft_deactivate_next(ctx->net, rule); ctx->chain->use--; return 0; } @@ -1898,7 +1868,7 @@ static int nf_tables_dump_rules(struct sk_buff *skb, list_for_each_entry_rcu(table, &afi->tables, list) { list_for_each_entry_rcu(chain, &table->chains, list) { list_for_each_entry_rcu(rule, &chain->rules, list) { - if (!nft_rule_is_active(net, rule)) + if (!nft_is_active(net, rule)) goto cont; if (idx < s_idx) goto cont; @@ -2102,7 +2072,7 @@ static int nf_tables_newrule(struct net *net, struct sock *nlsk, if (rule == NULL) goto err1; - nft_rule_activate_next(net, rule); + nft_activate_next(net, rule); rule->handle = handle; rule->dlen = size; @@ -2124,14 +2094,14 @@ static int nf_tables_newrule(struct net *net, struct sock *nlsk, } if (nlh->nlmsg_flags & NLM_F_REPLACE) { - if (nft_rule_is_active_next(net, old_rule)) { + if (nft_is_active_next(net, old_rule)) { trans = nft_trans_rule_add(&ctx, NFT_MSG_DELRULE, old_rule); if (trans == NULL) { err = -ENOMEM; goto err2; } - nft_rule_deactivate_next(net, old_rule); + nft_deactivate_next(net, old_rule); chain->use--; list_add_tail_rcu(&rule->list, &old_rule->list); } else { @@ -3980,7 +3950,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) trans->ctx.afi->nops); break; case NFT_MSG_NEWRULE: - nft_rule_clear(trans->ctx.net, nft_trans_rule(trans)); + nft_clear(trans->ctx.net, nft_trans_rule(trans)); nf_tables_rule_notify(&trans->ctx, nft_trans_rule(trans), NFT_MSG_NEWRULE); @@ -4116,7 +4086,7 @@ static int nf_tables_abort(struct net *net, struct sk_buff *skb) break; case NFT_MSG_DELRULE: trans->ctx.chain->use++; - nft_rule_clear(trans->ctx.net, nft_trans_rule(trans)); + nft_clear(trans->ctx.net, nft_trans_rule(trans)); nft_trans_destroy(trans); break; case NFT_MSG_NEWSET: -- cgit v0.10.2 From f2a6d766765d2794e26e25655d4ffcfe29c3ec2f Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 14 Jun 2016 17:29:18 +0200 Subject: netfilter: nf_tables: add generation mask to tables This patch addresses two problems: 1) The netlink dump is inconsistent when interfering with an ongoing transaction update for several reasons: 1.a) We don't honor the internal NFT_TABLE_INACTIVE flag, and we should be skipping these inactive objects in the dump. 1.b) We perform speculative deletion during the preparation phase, that may result in skipping active objects. 1.c) The listing order changes, which generates noise when tracking incremental ruleset update via tools like git or our own testsuite. 2) We don't allow to add and to update the object in the same batch, eg. add table x; add table x { flags dormant\; }. In order to resolve these problems: 1) If the user requests a deletion, the object becomes inactive in the next generation. Then, ignore objects that scheduled to be deleted from the lookup path, as they will be effectively removed in the next generation. 2) From the get/dump path, if the object is not currently active, we skip it. 3) Support 'add X -> update X' sequence from a transaction. After this update, we obtain a consistent list as long as we stay in the same generation. The userspace side can detect interferences through the generation counter so it can restart the dumping. Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index d0778cb..05c9a64b 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -838,6 +838,7 @@ unsigned int nft_do_chain(struct nft_pktinfo *pkt, void *priv); * @hgenerator: handle generator state * @use: number of chain references to this table * @flags: table flag (see enum nft_table_flags) + * @genmask: generation mask * @name: name of the table */ struct nft_table { @@ -846,7 +847,8 @@ struct nft_table { struct list_head sets; u64 hgenerator; u32 use; - u16 flags; + u16 flags:14, + genmask:2; char name[NFT_TABLE_MAXNAMELEN]; }; @@ -992,6 +994,8 @@ static inline u8 nft_genmask_cur(const struct net *net) /* After committing the ruleset, clear the stale generation bit. */ #define nft_clear(__net, __obj) \ (__obj)->genmask &= ~nft_genmask_next(__net) +#define nft_active_genmask(__obj, __genmask) \ + !((__obj)->genmask & __genmask) /* * Set element transaction helpers diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index d9f0f07..a4a77d6 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -175,9 +175,6 @@ static void nf_tables_unregister_hooks(const struct nft_table *table, nft_unregister_basechain(nft_base_chain(chain), hook_nops); } -/* Internal table flags */ -#define NFT_TABLE_INACTIVE (1 << 15) - static int nft_trans_table_add(struct nft_ctx *ctx, int msg_type) { struct nft_trans *trans; @@ -187,7 +184,7 @@ static int nft_trans_table_add(struct nft_ctx *ctx, int msg_type) return -ENOMEM; if (msg_type == NFT_MSG_NEWTABLE) - ctx->table->flags |= NFT_TABLE_INACTIVE; + nft_activate_next(ctx->net, ctx->table); list_add_tail(&trans->list, &ctx->net->nft.commit_list); return 0; @@ -201,7 +198,7 @@ static int nft_deltable(struct nft_ctx *ctx) if (err < 0) return err; - list_del_rcu(&ctx->table->list); + nft_deactivate_next(ctx->net, ctx->table); return err; } @@ -334,26 +331,29 @@ static int nft_delset(struct nft_ctx *ctx, struct nft_set *set) */ static struct nft_table *nft_table_lookup(const struct nft_af_info *afi, - const struct nlattr *nla) + const struct nlattr *nla, + u8 genmask) { struct nft_table *table; list_for_each_entry(table, &afi->tables, list) { - if (!nla_strcmp(nla, table->name)) + if (!nla_strcmp(nla, table->name) && + nft_active_genmask(table, genmask)) return table; } return NULL; } static struct nft_table *nf_tables_table_lookup(const struct nft_af_info *afi, - const struct nlattr *nla) + const struct nlattr *nla, + u8 genmask) { struct nft_table *table; if (nla == NULL) return ERR_PTR(-EINVAL); - table = nft_table_lookup(afi, nla); + table = nft_table_lookup(afi, nla, genmask); if (table != NULL) return table; @@ -494,6 +494,8 @@ static int nf_tables_dump_tables(struct sk_buff *skb, if (idx > s_idx) memset(&cb->args[1], 0, sizeof(cb->args) - sizeof(cb->args[0])); + if (!nft_is_active(net, table)) + continue; if (nf_tables_fill_table_info(skb, net, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, @@ -518,6 +520,7 @@ static int nf_tables_gettable(struct net *net, struct sock *nlsk, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + u8 genmask = nft_genmask_cur(net); const struct nft_af_info *afi; const struct nft_table *table; struct sk_buff *skb2; @@ -535,11 +538,9 @@ static int nf_tables_gettable(struct net *net, struct sock *nlsk, if (IS_ERR(afi)) return PTR_ERR(afi); - table = nf_tables_table_lookup(afi, nla[NFTA_TABLE_NAME]); + table = nf_tables_table_lookup(afi, nla[NFTA_TABLE_NAME], genmask); if (IS_ERR(table)) return PTR_ERR(table); - if (table->flags & NFT_TABLE_INACTIVE) - return -ENOENT; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) @@ -648,6 +649,7 @@ static int nf_tables_newtable(struct net *net, struct sock *nlsk, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + u8 genmask = nft_genmask_next(net); const struct nlattr *name; struct nft_af_info *afi; struct nft_table *table; @@ -661,7 +663,7 @@ static int nf_tables_newtable(struct net *net, struct sock *nlsk, return PTR_ERR(afi); name = nla[NFTA_TABLE_NAME]; - table = nf_tables_table_lookup(afi, name); + table = nf_tables_table_lookup(afi, name, genmask); if (IS_ERR(table)) { if (PTR_ERR(table) != -ENOENT) return PTR_ERR(table); @@ -669,8 +671,6 @@ static int nf_tables_newtable(struct net *net, struct sock *nlsk, } if (table != NULL) { - if (table->flags & NFT_TABLE_INACTIVE) - return -ENOENT; if (nlh->nlmsg_flags & NLM_F_EXCL) return -EEXIST; if (nlh->nlmsg_flags & NLM_F_REPLACE) @@ -765,6 +765,9 @@ static int nft_flush(struct nft_ctx *ctx, int family) ctx->afi = afi; list_for_each_entry_safe(table, nt, &afi->tables, list) { + if (!nft_is_active_next(ctx->net, table)) + continue; + if (nla[NFTA_TABLE_NAME] && nla_strcmp(nla[NFTA_TABLE_NAME], table->name) != 0) continue; @@ -785,6 +788,7 @@ static int nf_tables_deltable(struct net *net, struct sock *nlsk, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + u8 genmask = nft_genmask_next(net); struct nft_af_info *afi; struct nft_table *table; int family = nfmsg->nfgen_family; @@ -798,7 +802,7 @@ static int nf_tables_deltable(struct net *net, struct sock *nlsk, if (IS_ERR(afi)) return PTR_ERR(afi); - table = nf_tables_table_lookup(afi, nla[NFTA_TABLE_NAME]); + table = nf_tables_table_lookup(afi, nla[NFTA_TABLE_NAME], genmask); if (IS_ERR(table)) return PTR_ERR(table); @@ -1074,6 +1078,7 @@ static int nf_tables_getchain(struct net *net, struct sock *nlsk, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + u8 genmask = nft_genmask_cur(net); const struct nft_af_info *afi; const struct nft_table *table; const struct nft_chain *chain; @@ -1092,11 +1097,9 @@ static int nf_tables_getchain(struct net *net, struct sock *nlsk, if (IS_ERR(afi)) return PTR_ERR(afi); - table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE]); + table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE], genmask); if (IS_ERR(table)) return PTR_ERR(table); - if (table->flags & NFT_TABLE_INACTIVE) - return -ENOENT; chain = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME]); if (IS_ERR(chain)) @@ -1201,6 +1204,7 @@ static int nf_tables_newchain(struct net *net, struct sock *nlsk, struct nft_chain *chain; struct nft_base_chain *basechain = NULL; struct nlattr *ha[NFTA_HOOK_MAX + 1]; + u8 genmask = nft_genmask_next(net); int family = nfmsg->nfgen_family; struct net_device *dev = NULL; u8 policy = NF_ACCEPT; @@ -1217,7 +1221,7 @@ static int nf_tables_newchain(struct net *net, struct sock *nlsk, if (IS_ERR(afi)) return PTR_ERR(afi); - table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE]); + table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE], genmask); if (IS_ERR(table)) return PTR_ERR(table); @@ -1449,6 +1453,7 @@ static int nf_tables_delchain(struct net *net, struct sock *nlsk, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + u8 genmask = nft_genmask_next(net); struct nft_af_info *afi; struct nft_table *table; struct nft_chain *chain; @@ -1459,7 +1464,7 @@ static int nf_tables_delchain(struct net *net, struct sock *nlsk, if (IS_ERR(afi)) return PTR_ERR(afi); - table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE]); + table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE], genmask); if (IS_ERR(table)) return PTR_ERR(table); @@ -1901,6 +1906,7 @@ static int nf_tables_getrule(struct net *net, struct sock *nlsk, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + u8 genmask = nft_genmask_cur(net); const struct nft_af_info *afi; const struct nft_table *table; const struct nft_chain *chain; @@ -1920,11 +1926,9 @@ static int nf_tables_getrule(struct net *net, struct sock *nlsk, if (IS_ERR(afi)) return PTR_ERR(afi); - table = nf_tables_table_lookup(afi, nla[NFTA_RULE_TABLE]); + table = nf_tables_table_lookup(afi, nla[NFTA_RULE_TABLE], genmask); if (IS_ERR(table)) return PTR_ERR(table); - if (table->flags & NFT_TABLE_INACTIVE) - return -ENOENT; chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN]); if (IS_ERR(chain)) @@ -1979,6 +1983,7 @@ static int nf_tables_newrule(struct net *net, struct sock *nlsk, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + u8 genmask = nft_genmask_next(net); struct nft_af_info *afi; struct nft_table *table; struct nft_chain *chain; @@ -1999,7 +2004,7 @@ static int nf_tables_newrule(struct net *net, struct sock *nlsk, if (IS_ERR(afi)) return PTR_ERR(afi); - table = nf_tables_table_lookup(afi, nla[NFTA_RULE_TABLE]); + table = nf_tables_table_lookup(afi, nla[NFTA_RULE_TABLE], genmask); if (IS_ERR(table)) return PTR_ERR(table); @@ -2144,6 +2149,7 @@ static int nf_tables_delrule(struct net *net, struct sock *nlsk, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + u8 genmask = nft_genmask_next(net); struct nft_af_info *afi; struct nft_table *table; struct nft_chain *chain = NULL; @@ -2155,7 +2161,7 @@ static int nf_tables_delrule(struct net *net, struct sock *nlsk, if (IS_ERR(afi)) return PTR_ERR(afi); - table = nf_tables_table_lookup(afi, nla[NFTA_RULE_TABLE]); + table = nf_tables_table_lookup(afi, nla[NFTA_RULE_TABLE], genmask); if (IS_ERR(table)) return PTR_ERR(table); @@ -2309,7 +2315,8 @@ static const struct nla_policy nft_set_desc_policy[NFTA_SET_DESC_MAX + 1] = { static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, struct net *net, const struct sk_buff *skb, const struct nlmsghdr *nlh, - const struct nlattr * const nla[]) + const struct nlattr * const nla[], + u8 genmask) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); struct nft_af_info *afi = NULL; @@ -2325,7 +2332,8 @@ static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, struct net *net, if (afi == NULL) return -EAFNOSUPPORT; - table = nf_tables_table_lookup(afi, nla[NFTA_SET_TABLE]); + table = nf_tables_table_lookup(afi, nla[NFTA_SET_TABLE], + genmask); if (IS_ERR(table)) return PTR_ERR(table); } @@ -2586,6 +2594,7 @@ static int nf_tables_getset(struct net *net, struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { + u8 genmask = nft_genmask_cur(net); const struct nft_set *set; struct nft_ctx ctx; struct sk_buff *skb2; @@ -2593,7 +2602,7 @@ static int nf_tables_getset(struct net *net, struct sock *nlsk, int err; /* Verify existence before starting dump */ - err = nft_ctx_init_from_setattr(&ctx, net, skb, nlh, nla); + err = nft_ctx_init_from_setattr(&ctx, net, skb, nlh, nla, genmask); if (err < 0) return err; @@ -2661,6 +2670,7 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + u8 genmask = nft_genmask_next(net); const struct nft_set_ops *ops; struct nft_af_info *afi; struct nft_table *table; @@ -2758,7 +2768,7 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk, if (IS_ERR(afi)) return PTR_ERR(afi); - table = nf_tables_table_lookup(afi, nla[NFTA_SET_TABLE]); + table = nf_tables_table_lookup(afi, nla[NFTA_SET_TABLE], genmask); if (IS_ERR(table)) return PTR_ERR(table); @@ -2863,6 +2873,7 @@ static int nf_tables_delset(struct net *net, struct sock *nlsk, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); + u8 genmask = nft_genmask_next(net); struct nft_set *set; struct nft_ctx ctx; int err; @@ -2872,7 +2883,7 @@ static int nf_tables_delset(struct net *net, struct sock *nlsk, if (nla[NFTA_SET_TABLE] == NULL) return -EINVAL; - err = nft_ctx_init_from_setattr(&ctx, net, skb, nlh, nla); + err = nft_ctx_init_from_setattr(&ctx, net, skb, nlh, nla, genmask); if (err < 0) return err; @@ -3001,7 +3012,8 @@ static const struct nla_policy nft_set_elem_list_policy[NFTA_SET_ELEM_LIST_MAX + static int nft_ctx_init_from_elemattr(struct nft_ctx *ctx, struct net *net, const struct sk_buff *skb, const struct nlmsghdr *nlh, - const struct nlattr * const nla[]) + const struct nlattr * const nla[], + u8 genmask) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); struct nft_af_info *afi; @@ -3011,7 +3023,8 @@ static int nft_ctx_init_from_elemattr(struct nft_ctx *ctx, struct net *net, if (IS_ERR(afi)) return PTR_ERR(afi); - table = nf_tables_table_lookup(afi, nla[NFTA_SET_ELEM_LIST_TABLE]); + table = nf_tables_table_lookup(afi, nla[NFTA_SET_ELEM_LIST_TABLE], + genmask); if (IS_ERR(table)) return PTR_ERR(table); @@ -3108,6 +3121,7 @@ static int nf_tables_dump_setelem(const struct nft_ctx *ctx, static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb) { struct net *net = sock_net(skb->sk); + u8 genmask = nft_genmask_cur(net); const struct nft_set *set; struct nft_set_dump_args args; struct nft_ctx ctx; @@ -3124,11 +3138,9 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb) return err; err = nft_ctx_init_from_elemattr(&ctx, net, cb->skb, cb->nlh, - (void *)nla); + (void *)nla, genmask); if (err < 0) return err; - if (ctx.table->flags & NFT_TABLE_INACTIVE) - return -ENOENT; set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); if (IS_ERR(set)) @@ -3187,15 +3199,14 @@ static int nf_tables_getsetelem(struct net *net, struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { + u8 genmask = nft_genmask_cur(net); const struct nft_set *set; struct nft_ctx ctx; int err; - err = nft_ctx_init_from_elemattr(&ctx, net, skb, nlh, nla); + err = nft_ctx_init_from_elemattr(&ctx, net, skb, nlh, nla, genmask); if (err < 0) return err; - if (ctx.table->flags & NFT_TABLE_INACTIVE) - return -ENOENT; set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); if (IS_ERR(set)) @@ -3519,6 +3530,7 @@ static int nf_tables_newsetelem(struct net *net, struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { + u8 genmask = nft_genmask_next(net); const struct nlattr *attr; struct nft_set *set; struct nft_ctx ctx; @@ -3527,7 +3539,7 @@ static int nf_tables_newsetelem(struct net *net, struct sock *nlsk, if (nla[NFTA_SET_ELEM_LIST_ELEMENTS] == NULL) return -EINVAL; - err = nft_ctx_init_from_elemattr(&ctx, net, skb, nlh, nla); + err = nft_ctx_init_from_elemattr(&ctx, net, skb, nlh, nla, genmask); if (err < 0) return err; @@ -3641,6 +3653,7 @@ static int nf_tables_delsetelem(struct net *net, struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { + u8 genmask = nft_genmask_next(net); const struct nlattr *attr; struct nft_set *set; struct nft_ctx ctx; @@ -3649,7 +3662,7 @@ static int nf_tables_delsetelem(struct net *net, struct sock *nlsk, if (nla[NFTA_SET_ELEM_LIST_ELEMENTS] == NULL) return -EINVAL; - err = nft_ctx_init_from_elemattr(&ctx, net, skb, nlh, nla); + err = nft_ctx_init_from_elemattr(&ctx, net, skb, nlh, nla, genmask); if (err < 0) return err; @@ -3926,12 +3939,13 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) trans->ctx.table->flags |= NFT_TABLE_F_DORMANT; } } else { - trans->ctx.table->flags &= ~NFT_TABLE_INACTIVE; + nft_clear(net, trans->ctx.table); } nf_tables_table_notify(&trans->ctx, NFT_MSG_NEWTABLE); nft_trans_destroy(trans); break; case NFT_MSG_DELTABLE: + list_del_rcu(&trans->ctx.table->list); nf_tables_table_notify(&trans->ctx, NFT_MSG_DELTABLE); break; case NFT_MSG_NEWCHAIN: @@ -4057,8 +4071,7 @@ static int nf_tables_abort(struct net *net, struct sk_buff *skb) } break; case NFT_MSG_DELTABLE: - list_add_tail_rcu(&trans->ctx.table->list, - &trans->ctx.afi->tables); + nft_clear(trans->ctx.net, trans->ctx.table); nft_trans_destroy(trans); break; case NFT_MSG_NEWCHAIN: -- cgit v0.10.2 From 664b0f8cd8c66d02d14168ee7ac6a957cc88177f Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Sun, 12 Jun 2016 19:21:31 +0200 Subject: netfilter: nf_tables: add generation mask to chains Similar to ("netfilter: nf_tables: add generation mask to tables"). Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 05c9a64b..b023e28 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -732,7 +732,6 @@ static inline struct nft_userdata *nft_userdata(const struct nft_rule *rule) enum nft_chain_flags { NFT_BASE_CHAIN = 0x1, - NFT_CHAIN_INACTIVE = 0x2, }; /** @@ -754,7 +753,8 @@ struct nft_chain { u64 handle; u32 use; u16 level; - u8 flags; + u8 flags:6, + genmask:2; char name[NFT_CHAIN_MAXNAMELEN]; }; diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index a4a77d6..cae88f8 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -211,7 +211,7 @@ static int nft_trans_chain_add(struct nft_ctx *ctx, int msg_type) return -ENOMEM; if (msg_type == NFT_MSG_NEWCHAIN) - ctx->chain->flags |= NFT_CHAIN_INACTIVE; + nft_activate_next(ctx->net, ctx->chain); list_add_tail(&trans->list, &ctx->net->nft.commit_list); return 0; @@ -226,7 +226,7 @@ static int nft_delchain(struct nft_ctx *ctx) return err; ctx->table->use--; - list_del_rcu(&ctx->chain->list); + nft_deactivate_next(ctx->net, ctx->chain); return err; } @@ -559,13 +559,16 @@ err: return err; } -static int nf_tables_table_enable(const struct nft_af_info *afi, +static int nf_tables_table_enable(struct net *net, + const struct nft_af_info *afi, struct nft_table *table) { struct nft_chain *chain; int err, i = 0; list_for_each_entry(chain, &table->chains, list) { + if (!nft_is_active_next(net, chain)) + continue; if (!(chain->flags & NFT_BASE_CHAIN)) continue; @@ -578,6 +581,8 @@ static int nf_tables_table_enable(const struct nft_af_info *afi, return 0; err: list_for_each_entry(chain, &table->chains, list) { + if (!nft_is_active_next(net, chain)) + continue; if (!(chain->flags & NFT_BASE_CHAIN)) continue; @@ -589,12 +594,15 @@ err: return err; } -static void nf_tables_table_disable(const struct nft_af_info *afi, +static void nf_tables_table_disable(struct net *net, + const struct nft_af_info *afi, struct nft_table *table) { struct nft_chain *chain; list_for_each_entry(chain, &table->chains, list) { + if (!nft_is_active_next(net, chain)) + continue; if (chain->flags & NFT_BASE_CHAIN) nft_unregister_basechain(nft_base_chain(chain), afi->nops); @@ -627,7 +635,7 @@ static int nf_tables_updtable(struct nft_ctx *ctx) nft_trans_table_enable(trans) = false; } else if (!(flags & NFT_TABLE_F_DORMANT) && ctx->table->flags & NFT_TABLE_F_DORMANT) { - ret = nf_tables_table_enable(ctx->afi, ctx->table); + ret = nf_tables_table_enable(ctx->net, ctx->afi, ctx->table); if (ret >= 0) { ctx->table->flags &= ~NFT_TABLE_F_DORMANT; nft_trans_table_enable(trans) = true; @@ -722,6 +730,9 @@ static int nft_flush_table(struct nft_ctx *ctx) struct nft_set *set, *ns; list_for_each_entry(chain, &ctx->table->chains, list) { + if (!nft_is_active_next(ctx->net, chain)) + continue; + ctx->chain = chain; err = nft_delrule_by_chain(ctx); @@ -740,6 +751,9 @@ static int nft_flush_table(struct nft_ctx *ctx) } list_for_each_entry_safe(chain, nc, &ctx->table->chains, list) { + if (!nft_is_active_next(ctx->net, chain)) + continue; + ctx->chain = chain; err = nft_delchain(ctx); @@ -849,12 +863,14 @@ EXPORT_SYMBOL_GPL(nft_unregister_chain_type); */ static struct nft_chain * -nf_tables_chain_lookup_byhandle(const struct nft_table *table, u64 handle) +nf_tables_chain_lookup_byhandle(const struct nft_table *table, u64 handle, + u8 genmask) { struct nft_chain *chain; list_for_each_entry(chain, &table->chains, list) { - if (chain->handle == handle) + if (chain->handle == handle && + nft_active_genmask(chain, genmask)) return chain; } @@ -862,7 +878,8 @@ nf_tables_chain_lookup_byhandle(const struct nft_table *table, u64 handle) } static struct nft_chain *nf_tables_chain_lookup(const struct nft_table *table, - const struct nlattr *nla) + const struct nlattr *nla, + u8 genmask) { struct nft_chain *chain; @@ -870,7 +887,8 @@ static struct nft_chain *nf_tables_chain_lookup(const struct nft_table *table, return ERR_PTR(-EINVAL); list_for_each_entry(chain, &table->chains, list) { - if (!nla_strcmp(nla, chain->name)) + if (!nla_strcmp(nla, chain->name) && + nft_active_genmask(chain, genmask)) return chain; } @@ -1053,6 +1071,8 @@ static int nf_tables_dump_chains(struct sk_buff *skb, if (idx > s_idx) memset(&cb->args[1], 0, sizeof(cb->args) - sizeof(cb->args[0])); + if (!nft_is_active(net, chain)) + continue; if (nf_tables_fill_chain_info(skb, net, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, @@ -1101,11 +1121,9 @@ static int nf_tables_getchain(struct net *net, struct sock *nlsk, if (IS_ERR(table)) return PTR_ERR(table); - chain = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME]); + chain = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME], genmask); if (IS_ERR(chain)) return PTR_ERR(chain); - if (chain->flags & NFT_CHAIN_INACTIVE) - return -ENOENT; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) @@ -1230,11 +1248,11 @@ static int nf_tables_newchain(struct net *net, struct sock *nlsk, if (nla[NFTA_CHAIN_HANDLE]) { handle = be64_to_cpu(nla_get_be64(nla[NFTA_CHAIN_HANDLE])); - chain = nf_tables_chain_lookup_byhandle(table, handle); + chain = nf_tables_chain_lookup_byhandle(table, handle, genmask); if (IS_ERR(chain)) return PTR_ERR(chain); } else { - chain = nf_tables_chain_lookup(table, name); + chain = nf_tables_chain_lookup(table, name, genmask); if (IS_ERR(chain)) { if (PTR_ERR(chain) != -ENOENT) return PTR_ERR(chain); @@ -1265,16 +1283,20 @@ static int nf_tables_newchain(struct net *net, struct sock *nlsk, struct nft_stats *stats = NULL; struct nft_trans *trans; - if (chain->flags & NFT_CHAIN_INACTIVE) - return -ENOENT; if (nlh->nlmsg_flags & NLM_F_EXCL) return -EEXIST; if (nlh->nlmsg_flags & NLM_F_REPLACE) return -EOPNOTSUPP; - if (nla[NFTA_CHAIN_HANDLE] && name && - !IS_ERR(nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME]))) - return -EEXIST; + if (nla[NFTA_CHAIN_HANDLE] && name) { + struct nft_chain *chain2; + + chain2 = nf_tables_chain_lookup(table, + nla[NFTA_CHAIN_NAME], + genmask); + if (IS_ERR(chain2)) + return PTR_ERR(chain2); + } if (nla[NFTA_CHAIN_COUNTERS]) { if (!(chain->flags & NFT_BASE_CHAIN)) @@ -1468,7 +1490,7 @@ static int nf_tables_delchain(struct net *net, struct sock *nlsk, if (IS_ERR(table)) return PTR_ERR(table); - chain = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME]); + chain = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME], genmask); if (IS_ERR(chain)) return PTR_ERR(chain); if (chain->use > 0) @@ -1930,11 +1952,9 @@ static int nf_tables_getrule(struct net *net, struct sock *nlsk, if (IS_ERR(table)) return PTR_ERR(table); - chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN]); + chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN], genmask); if (IS_ERR(chain)) return PTR_ERR(chain); - if (chain->flags & NFT_CHAIN_INACTIVE) - return -ENOENT; rule = nf_tables_rule_lookup(chain, nla[NFTA_RULE_HANDLE]); if (IS_ERR(rule)) @@ -2008,7 +2028,7 @@ static int nf_tables_newrule(struct net *net, struct sock *nlsk, if (IS_ERR(table)) return PTR_ERR(table); - chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN]); + chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN], genmask); if (IS_ERR(chain)) return PTR_ERR(chain); @@ -2166,7 +2186,8 @@ static int nf_tables_delrule(struct net *net, struct sock *nlsk, return PTR_ERR(table); if (nla[NFTA_RULE_CHAIN]) { - chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN]); + chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN], + genmask); if (IS_ERR(chain)) return PTR_ERR(chain); } @@ -2186,6 +2207,9 @@ static int nf_tables_delrule(struct net *net, struct sock *nlsk, } } else { list_for_each_entry(chain, &table->chains, list) { + if (!nft_is_active_next(net, chain)) + continue; + ctx.chain = chain; err = nft_delrule_by_chain(&ctx); if (err < 0) @@ -3934,7 +3958,8 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) case NFT_MSG_NEWTABLE: if (nft_trans_table_update(trans)) { if (!nft_trans_table_enable(trans)) { - nf_tables_table_disable(trans->ctx.afi, + nf_tables_table_disable(net, + trans->ctx.afi, trans->ctx.table); trans->ctx.table->flags |= NFT_TABLE_F_DORMANT; } @@ -3952,12 +3977,13 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) if (nft_trans_chain_update(trans)) nft_chain_commit_update(trans); else - trans->ctx.chain->flags &= ~NFT_CHAIN_INACTIVE; + nft_clear(net, trans->ctx.chain); nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN); nft_trans_destroy(trans); break; case NFT_MSG_DELCHAIN: + list_del_rcu(&trans->ctx.chain->list); nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN); nf_tables_unregister_hooks(trans->ctx.table, trans->ctx.chain, @@ -4061,7 +4087,8 @@ static int nf_tables_abort(struct net *net, struct sk_buff *skb) case NFT_MSG_NEWTABLE: if (nft_trans_table_update(trans)) { if (nft_trans_table_enable(trans)) { - nf_tables_table_disable(trans->ctx.afi, + nf_tables_table_disable(net, + trans->ctx.afi, trans->ctx.table); trans->ctx.table->flags |= NFT_TABLE_F_DORMANT; } @@ -4089,8 +4116,7 @@ static int nf_tables_abort(struct net *net, struct sk_buff *skb) break; case NFT_MSG_DELCHAIN: trans->ctx.table->use++; - list_add_tail_rcu(&trans->ctx.chain->list, - &trans->ctx.table->chains); + nft_clear(trans->ctx.net, trans->ctx.chain); nft_trans_destroy(trans); break; case NFT_MSG_NEWRULE: @@ -4413,6 +4439,7 @@ static const struct nla_policy nft_verdict_policy[NFTA_VERDICT_MAX + 1] = { static int nft_verdict_init(const struct nft_ctx *ctx, struct nft_data *data, struct nft_data_desc *desc, const struct nlattr *nla) { + u8 genmask = nft_genmask_next(ctx->net); struct nlattr *tb[NFTA_VERDICT_MAX + 1]; struct nft_chain *chain; int err; @@ -4445,7 +4472,7 @@ static int nft_verdict_init(const struct nft_ctx *ctx, struct nft_data *data, if (!tb[NFTA_VERDICT_CHAIN]) return -EINVAL; chain = nf_tables_chain_lookup(ctx->table, - tb[NFTA_VERDICT_CHAIN]); + tb[NFTA_VERDICT_CHAIN], genmask); if (IS_ERR(chain)) return PTR_ERR(chain); if (chain->flags & NFT_BASE_CHAIN) -- cgit v0.10.2 From 37a9cc52552579f22e18cca401cfc4351b6cbc72 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Sun, 12 Jun 2016 22:52:45 +0200 Subject: netfilter: nf_tables: add generation mask to sets Similar to ("netfilter: nf_tables: add generation mask to tables"). Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index b023e28..07a5ba4 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -296,6 +296,7 @@ void nft_unregister_set(struct nft_set_ops *ops); * @ops: set ops * @pnet: network namespace * @flags: set flags + * @genmask: generation mask * @klen: key length * @dlen: data length * @data: private set data @@ -317,7 +318,8 @@ struct nft_set { /* runtime data below here */ const struct nft_set_ops *ops ____cacheline_aligned; possible_net_t pnet; - u16 flags; + u16 flags:14, + genmask:2; u8 klen; u8 dlen; unsigned char data[] @@ -335,9 +337,9 @@ static inline struct nft_set *nft_set_container_of(const void *priv) } struct nft_set *nf_tables_set_lookup(const struct nft_table *table, - const struct nlattr *nla); + const struct nlattr *nla, u8 genmask); struct nft_set *nf_tables_set_lookup_byid(const struct net *net, - const struct nlattr *nla); + const struct nlattr *nla, u8 genmask); static inline unsigned long nft_set_gc_interval(const struct nft_set *set) { diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index cae88f8..3316bce 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -289,9 +289,6 @@ static int nft_delrule_by_chain(struct nft_ctx *ctx) return 0; } -/* Internal set flag */ -#define NFT_SET_INACTIVE (1 << 15) - static int nft_trans_set_add(struct nft_ctx *ctx, int msg_type, struct nft_set *set) { @@ -304,7 +301,7 @@ static int nft_trans_set_add(struct nft_ctx *ctx, int msg_type, if (msg_type == NFT_MSG_NEWSET && ctx->nla[NFTA_SET_ID] != NULL) { nft_trans_set_id(trans) = ntohl(nla_get_be32(ctx->nla[NFTA_SET_ID])); - set->flags |= NFT_SET_INACTIVE; + nft_activate_next(ctx->net, set); } nft_trans_set(trans) = set; list_add_tail(&trans->list, &ctx->net->nft.commit_list); @@ -320,7 +317,7 @@ static int nft_delset(struct nft_ctx *ctx, struct nft_set *set) if (err < 0) return err; - list_del_rcu(&set->list); + nft_deactivate_next(ctx->net, set); ctx->table->use--; return err; @@ -741,6 +738,9 @@ static int nft_flush_table(struct nft_ctx *ctx) } list_for_each_entry_safe(set, ns, &ctx->table->sets, list) { + if (!nft_is_active_next(ctx->net, set)) + continue; + if (set->flags & NFT_SET_ANONYMOUS && !list_empty(&set->bindings)) continue; @@ -2367,7 +2367,7 @@ static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, struct net *net, } struct nft_set *nf_tables_set_lookup(const struct nft_table *table, - const struct nlattr *nla) + const struct nlattr *nla, u8 genmask) { struct nft_set *set; @@ -2375,22 +2375,27 @@ struct nft_set *nf_tables_set_lookup(const struct nft_table *table, return ERR_PTR(-EINVAL); list_for_each_entry(set, &table->sets, list) { - if (!nla_strcmp(nla, set->name)) + if (!nla_strcmp(nla, set->name) && + nft_active_genmask(set, genmask)) return set; } return ERR_PTR(-ENOENT); } struct nft_set *nf_tables_set_lookup_byid(const struct net *net, - const struct nlattr *nla) + const struct nlattr *nla, + u8 genmask) { struct nft_trans *trans; u32 id = ntohl(nla_get_be32(nla)); list_for_each_entry(trans, &net->nft.commit_list, list) { + struct nft_set *set = nft_trans_set(trans); + if (trans->msg_type == NFT_MSG_NEWSET && - id == nft_trans_set_id(trans)) - return nft_trans_set(trans); + id == nft_trans_set_id(trans) && + nft_active_genmask(set, genmask)) + return set; } return ERR_PTR(-ENOENT); } @@ -2415,6 +2420,8 @@ cont: list_for_each_entry(i, &ctx->table->sets, list) { int tmp; + if (!nft_is_active_next(ctx->net, set)) + continue; if (!sscanf(i->name, name, &tmp)) continue; if (tmp < min || tmp >= min + BITS_PER_BYTE * PAGE_SIZE) @@ -2434,6 +2441,8 @@ cont: snprintf(set->name, sizeof(set->name), name, min + n); list_for_each_entry(i, &ctx->table->sets, list) { + if (!nft_is_active_next(ctx->net, i)) + continue; if (!strcmp(set->name, i->name)) return -ENFILE; } @@ -2582,6 +2591,8 @@ static int nf_tables_dump_sets(struct sk_buff *skb, struct netlink_callback *cb) list_for_each_entry_rcu(set, &table->sets, list) { if (idx < s_idx) goto cont; + if (!nft_is_active(net, set)) + goto cont; ctx_set = *ctx; ctx_set.table = table; @@ -2651,11 +2662,9 @@ static int nf_tables_getset(struct net *net, struct sock *nlsk, if (nfmsg->nfgen_family == NFPROTO_UNSPEC) return -EAFNOSUPPORT; - set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); + set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME], genmask); if (IS_ERR(set)) return PTR_ERR(set); - if (set->flags & NFT_SET_INACTIVE) - return -ENOENT; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (skb2 == NULL) @@ -2798,7 +2807,7 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk, nft_ctx_init(&ctx, net, skb, nlh, afi, table, NULL, nla); - set = nf_tables_set_lookup(table, nla[NFTA_SET_NAME]); + set = nf_tables_set_lookup(table, nla[NFTA_SET_NAME], genmask); if (IS_ERR(set)) { if (PTR_ERR(set) != -ENOENT) return PTR_ERR(set); @@ -2911,7 +2920,7 @@ static int nf_tables_delset(struct net *net, struct sock *nlsk, if (err < 0) return err; - set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); + set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME], genmask); if (IS_ERR(set)) return PTR_ERR(set); if (!list_empty(&set->bindings)) @@ -2980,7 +2989,7 @@ void nf_tables_unbind_set(const struct nft_ctx *ctx, struct nft_set *set, list_del_rcu(&binding->list); if (list_empty(&set->bindings) && set->flags & NFT_SET_ANONYMOUS && - !(set->flags & NFT_SET_INACTIVE)) + nft_is_active(ctx->net, set)) nf_tables_set_destroy(ctx, set); } @@ -3166,11 +3175,10 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb) if (err < 0) return err; - set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); + set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET], + genmask); if (IS_ERR(set)) return PTR_ERR(set); - if (set->flags & NFT_SET_INACTIVE) - return -ENOENT; event = NFT_MSG_NEWSETELEM; event |= NFNL_SUBSYS_NFTABLES << 8; @@ -3232,11 +3240,10 @@ static int nf_tables_getsetelem(struct net *net, struct sock *nlsk, if (err < 0) return err; - set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); + set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET], + genmask); if (IS_ERR(set)) return PTR_ERR(set); - if (set->flags & NFT_SET_INACTIVE) - return -ENOENT; if (nlh->nlmsg_flags & NLM_F_DUMP) { struct netlink_dump_control c = { @@ -3567,11 +3574,13 @@ static int nf_tables_newsetelem(struct net *net, struct sock *nlsk, if (err < 0) return err; - set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); + set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET], + genmask); if (IS_ERR(set)) { if (nla[NFTA_SET_ELEM_LIST_SET_ID]) { set = nf_tables_set_lookup_byid(net, - nla[NFTA_SET_ELEM_LIST_SET_ID]); + nla[NFTA_SET_ELEM_LIST_SET_ID], + genmask); } if (IS_ERR(set)) return PTR_ERR(set); @@ -3690,7 +3699,8 @@ static int nf_tables_delsetelem(struct net *net, struct sock *nlsk, if (err < 0) return err; - set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); + set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET], + genmask); if (IS_ERR(set)) return PTR_ERR(set); if (!list_empty(&set->bindings) && set->flags & NFT_SET_CONSTANT) @@ -4003,7 +4013,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) NFT_MSG_DELRULE); break; case NFT_MSG_NEWSET: - nft_trans_set(trans)->flags &= ~NFT_SET_INACTIVE; + nft_clear(net, nft_trans_set(trans)); /* This avoids hitting -EBUSY when deleting the table * from the transaction. */ @@ -4016,6 +4026,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) nft_trans_destroy(trans); break; case NFT_MSG_DELSET: + list_del_rcu(&nft_trans_set(trans)->list); nf_tables_set_notify(&trans->ctx, nft_trans_set(trans), NFT_MSG_DELSET, GFP_KERNEL); break; @@ -4134,8 +4145,7 @@ static int nf_tables_abort(struct net *net, struct sk_buff *skb) break; case NFT_MSG_DELSET: trans->ctx.table->use++; - list_add_tail_rcu(&nft_trans_set(trans)->list, - &trans->ctx.table->sets); + nft_clear(trans->ctx.net, nft_trans_set(trans)); nft_trans_destroy(trans); break; case NFT_MSG_NEWSETELEM: @@ -4282,6 +4292,8 @@ static int nf_tables_check_loops(const struct nft_ctx *ctx, } list_for_each_entry(set, &ctx->table->sets, list) { + if (!nft_is_active_next(ctx->net, set)) + continue; if (!(set->flags & NFT_SET_MAP) || set->dtype != NFT_DATA_VERDICT) continue; diff --git a/net/netfilter/nft_dynset.c b/net/netfilter/nft_dynset.c index 78d4914..0af2669 100644 --- a/net/netfilter/nft_dynset.c +++ b/net/netfilter/nft_dynset.c @@ -103,6 +103,7 @@ static int nft_dynset_init(const struct nft_ctx *ctx, const struct nlattr * const tb[]) { struct nft_dynset *priv = nft_expr_priv(expr); + u8 genmask = nft_genmask_next(ctx->net); struct nft_set *set; u64 timeout; int err; @@ -112,11 +113,13 @@ static int nft_dynset_init(const struct nft_ctx *ctx, tb[NFTA_DYNSET_SREG_KEY] == NULL) return -EINVAL; - set = nf_tables_set_lookup(ctx->table, tb[NFTA_DYNSET_SET_NAME]); + set = nf_tables_set_lookup(ctx->table, tb[NFTA_DYNSET_SET_NAME], + genmask); if (IS_ERR(set)) { if (tb[NFTA_DYNSET_SET_ID]) set = nf_tables_set_lookup_byid(ctx->net, - tb[NFTA_DYNSET_SET_ID]); + tb[NFTA_DYNSET_SET_ID], + genmask); if (IS_ERR(set)) return PTR_ERR(set); } diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index b3c31ef..8a102cf 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -54,6 +54,7 @@ static int nft_lookup_init(const struct nft_ctx *ctx, const struct nlattr * const tb[]) { struct nft_lookup *priv = nft_expr_priv(expr); + u8 genmask = nft_genmask_next(ctx->net); struct nft_set *set; int err; @@ -61,11 +62,12 @@ static int nft_lookup_init(const struct nft_ctx *ctx, tb[NFTA_LOOKUP_SREG] == NULL) return -EINVAL; - set = nf_tables_set_lookup(ctx->table, tb[NFTA_LOOKUP_SET]); + set = nf_tables_set_lookup(ctx->table, tb[NFTA_LOOKUP_SET], genmask); if (IS_ERR(set)) { if (tb[NFTA_LOOKUP_SET_ID]) { set = nf_tables_set_lookup_byid(ctx->net, - tb[NFTA_LOOKUP_SET_ID]); + tb[NFTA_LOOKUP_SET_ID], + genmask); } if (IS_ERR(set)) return PTR_ERR(set); -- cgit v0.10.2 From 4e5001651f5e488eac378ebabc5bde2a8f1ea861 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 21 Jun 2016 00:12:15 +0200 Subject: netfilter: nft_rbtree: check for next generation when deactivating elements set->ops->deactivate() is invoked from nft_del_setelem() that happens from the transaction path, so we have to check if the object is active in the next generation, not the current. Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nft_rbtree.c b/net/netfilter/nft_rbtree.c index f762094..86fbe5e 100644 --- a/net/netfilter/nft_rbtree.c +++ b/net/netfilter/nft_rbtree.c @@ -170,7 +170,7 @@ static void *nft_rbtree_deactivate(const struct nft_set *set, const struct nft_rbtree *priv = nft_set_priv(set); const struct rb_node *parent = priv->root.rb_node; struct nft_rbtree_elem *rbe, *this = elem->priv; - u8 genmask = nft_genmask_cur(read_pnet(&set->pnet)); + u8 genmask = nft_genmask_next(read_pnet(&set->pnet)); int d; while (parent != NULL) { -- cgit v0.10.2 From 8eee54be73f4b938dbf48e95c0dbecb5f19b08ee Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 21 Jun 2016 00:12:26 +0200 Subject: netfilter: nft_hash: support deletion of inactive elements New elements are inactive in the preparation phase, and its NFT_SET_ELEM_BUSY_MASK flag is set on. This busy flag doesn't allow us to delete it from the same transaction, following a sequence like: begin transaction add element X delete element X end transaction This sequence is valid and may be triggered by robots. To resolve this problem, allow deactivating elements that are active in the current generation (ie. those that has been just added in this batch). Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nft_hash.c b/net/netfilter/nft_hash.c index 6fa0165..d3a507d 100644 --- a/net/netfilter/nft_hash.c +++ b/net/netfilter/nft_hash.c @@ -153,9 +153,10 @@ static void *nft_hash_deactivate(const struct nft_set *set, const struct nft_set_elem *elem) { struct nft_hash *priv = nft_set_priv(set); + struct net *net = read_pnet(&set->pnet); struct nft_hash_elem *he; struct nft_hash_cmp_arg arg = { - .genmask = nft_genmask_next(read_pnet(&set->pnet)), + .genmask = nft_genmask_next(net), .set = set, .key = elem->key.val.data, }; @@ -163,7 +164,8 @@ static void *nft_hash_deactivate(const struct nft_set *set, rcu_read_lock(); he = rhashtable_lookup_fast(&priv->ht, &arg, nft_hash_params); if (he != NULL) { - if (!nft_set_elem_mark_busy(&he->ext)) + if (!nft_set_elem_mark_busy(&he->ext) || + !nft_is_active(net, &he->ext)) nft_set_elem_change_active(set, &he->ext); else he = NULL; -- cgit v0.10.2 From 3183ab8997a477c8d9ad175a1cef70dff77c6dbc Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 22 Jun 2016 13:26:10 +0200 Subject: netfilter: conntrack: allow increasing bucket size via sysctl too No need to restrict this to module parameter. We export a copy of the real hash size -- when user alters the value we allocate the new table, copy entries etc before we update the real size to the requested one. This is also needed because the real size is used by concurrent readers and cannot be changed without synchronizing the conntrack generation seqcnt. We only allow changing this value from the initial net namespace. Tested using http-client-benchmark vs. httpterm with concurrent while true;do echo $RANDOM > /proc/sys/net/netfilter/nf_conntrack_buckets done Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/Documentation/networking/nf_conntrack-sysctl.txt b/Documentation/networking/nf_conntrack-sysctl.txt index f55599c..4fb51d3 100644 --- a/Documentation/networking/nf_conntrack-sysctl.txt +++ b/Documentation/networking/nf_conntrack-sysctl.txt @@ -7,12 +7,13 @@ nf_conntrack_acct - BOOLEAN Enable connection tracking flow accounting. 64-bit byte and packet counters per flow are added. -nf_conntrack_buckets - INTEGER (read-only) +nf_conntrack_buckets - INTEGER Size of hash table. If not specified as parameter during module loading, the default size is calculated by dividing total memory by 16384 to determine the number of buckets but the hash table will never have fewer than 32 and limited to 16384 buckets. For systems with more than 4GB of memory it will be 65536 buckets. + This sysctl is only writeable in the initial net namespace. nf_conntrack_checksum - BOOLEAN 0 - disabled diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 9c0ed3d..5d3397f 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -290,6 +290,7 @@ static inline bool nf_is_loopback_packet(const struct sk_buff *skb) struct kernel_param; int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp); +int nf_conntrack_hash_resize(unsigned int hashsize); extern unsigned int nf_conntrack_htable_size; extern unsigned int nf_conntrack_max; diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index a459176..e17d5c7 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1595,24 +1595,14 @@ void *nf_ct_alloc_hashtable(unsigned int *sizep, int nulls) } EXPORT_SYMBOL_GPL(nf_ct_alloc_hashtable); -int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) +int nf_conntrack_hash_resize(unsigned int hashsize) { - int i, bucket, rc; - unsigned int hashsize, old_size; + int i, bucket; + unsigned int old_size; struct hlist_nulls_head *hash, *old_hash; struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; - if (current->nsproxy->net_ns != &init_net) - return -EOPNOTSUPP; - - /* On boot, we can set this without any fancy locking. */ - if (!nf_conntrack_htable_size) - return param_set_uint(val, kp); - - rc = kstrtouint(val, 0, &hashsize); - if (rc) - return rc; if (!hashsize) return -EINVAL; @@ -1620,6 +1610,12 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) if (!hash) return -ENOMEM; + old_size = nf_conntrack_htable_size; + if (old_size == hashsize) { + nf_ct_free_hashtable(hash, hashsize); + return 0; + } + local_bh_disable(); nf_conntrack_all_lock(); write_seqcount_begin(&nf_conntrack_generation); @@ -1655,6 +1651,25 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) nf_ct_free_hashtable(old_hash, old_size); return 0; } + +int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) +{ + unsigned int hashsize; + int rc; + + if (current->nsproxy->net_ns != &init_net) + return -EOPNOTSUPP; + + /* On boot, we can set this without any fancy locking. */ + if (!nf_conntrack_htable_size) + return param_set_uint(val, kp); + + rc = kstrtouint(val, 0, &hashsize); + if (rc) + return rc; + + return nf_conntrack_hash_resize(hashsize); +} EXPORT_SYMBOL_GPL(nf_conntrack_set_hashsize); module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint, diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index f87e84e..a0cc191 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -434,8 +434,29 @@ static void nf_conntrack_standalone_fini_proc(struct net *net) #ifdef CONFIG_SYSCTL /* Log invalid packets of a given protocol */ -static int log_invalid_proto_min = 0; -static int log_invalid_proto_max = 255; +static int log_invalid_proto_min __read_mostly; +static int log_invalid_proto_max __read_mostly = 255; + +/* size the user *wants to set */ +static unsigned int nf_conntrack_htable_size_user __read_mostly; + +static int +nf_conntrack_hash_sysctl(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int ret; + + ret = proc_dointvec(table, write, buffer, lenp, ppos); + if (ret < 0 || !write) + return ret; + + /* update ret, we might not be able to satisfy request */ + ret = nf_conntrack_hash_resize(nf_conntrack_htable_size_user); + + /* update it to the actual value used by conntrack */ + nf_conntrack_htable_size_user = nf_conntrack_htable_size; + return ret; +} static struct ctl_table_header *nf_ct_netfilter_header; @@ -456,10 +477,10 @@ static struct ctl_table nf_ct_sysctl_table[] = { }, { .procname = "nf_conntrack_buckets", - .data = &nf_conntrack_htable_size, + .data = &nf_conntrack_htable_size_user, .maxlen = sizeof(unsigned int), - .mode = 0444, - .proc_handler = proc_dointvec, + .mode = 0644, + .proc_handler = nf_conntrack_hash_sysctl, }, { .procname = "nf_conntrack_checksum", @@ -517,6 +538,9 @@ static int nf_conntrack_standalone_init_sysctl(struct net *net) if (net->user_ns != &init_user_ns) table[0].procname = NULL; + if (!net_eq(&init_net, net)) + table[2].mode = 0444; + net->ct.sysctl_header = register_net_sysctl(net, "net/netfilter", table); if (!net->ct.sysctl_header) goto out_unregister_netfilter; @@ -606,6 +630,8 @@ static int __init nf_conntrack_standalone_init(void) ret = -ENOMEM; goto out_sysctl; } + + nf_conntrack_htable_size_user = nf_conntrack_htable_size; #endif ret = register_pernet_subsys(&nf_conntrack_net_ops); -- cgit v0.10.2 From 82bec71d46b83f39860e2838ff8394e4fcd6efab Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 22 Jun 2016 14:26:33 +0200 Subject: netfilter: nf_tables: get rid of NFT_BASECHAIN_DISABLED This flag was introduced to restore rulesets from the new netdev family, but since 5ebe0b0eec9d6f7 ("netfilter: nf_tables: destroy basechain and rules on netdevice removal") the ruleset is released once the netdev is gone. This also removes nft_register_basechain() and nft_unregister_basechain() since they have no clients anymore after this rework. Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 07a5ba4..1ea19a6 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -798,7 +798,6 @@ struct nft_stats { }; #define NFT_HOOK_OPS_MAX 2 -#define NFT_BASECHAIN_DISABLED (1 << 0) /** * struct nft_base_chain - nf_tables base chain diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 3316bce..92c9fae 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -131,29 +131,8 @@ static void nft_trans_destroy(struct nft_trans *trans) kfree(trans); } -static int nft_register_basechain(struct nft_base_chain *basechain, - unsigned int hook_nops) -{ - struct net *net = read_pnet(&basechain->pnet); - - if (basechain->flags & NFT_BASECHAIN_DISABLED) - return 0; - - return nf_register_net_hooks(net, basechain->ops, hook_nops); -} - -static void nft_unregister_basechain(struct nft_base_chain *basechain, - unsigned int hook_nops) -{ - struct net *net = read_pnet(&basechain->pnet); - - if (basechain->flags & NFT_BASECHAIN_DISABLED) - return; - - nf_unregister_net_hooks(net, basechain->ops, hook_nops); -} - -static int nf_tables_register_hooks(const struct nft_table *table, +static int nf_tables_register_hooks(struct net *net, + const struct nft_table *table, struct nft_chain *chain, unsigned int hook_nops) { @@ -161,10 +140,12 @@ static int nf_tables_register_hooks(const struct nft_table *table, !(chain->flags & NFT_BASE_CHAIN)) return 0; - return nft_register_basechain(nft_base_chain(chain), hook_nops); + return nf_register_net_hooks(net, nft_base_chain(chain)->ops, + hook_nops); } -static void nf_tables_unregister_hooks(const struct nft_table *table, +static void nf_tables_unregister_hooks(struct net *net, + const struct nft_table *table, struct nft_chain *chain, unsigned int hook_nops) { @@ -172,7 +153,7 @@ static void nf_tables_unregister_hooks(const struct nft_table *table, !(chain->flags & NFT_BASE_CHAIN)) return; - nft_unregister_basechain(nft_base_chain(chain), hook_nops); + nf_unregister_net_hooks(net, nft_base_chain(chain)->ops, hook_nops); } static int nft_trans_table_add(struct nft_ctx *ctx, int msg_type) @@ -569,7 +550,8 @@ static int nf_tables_table_enable(struct net *net, if (!(chain->flags & NFT_BASE_CHAIN)) continue; - err = nft_register_basechain(nft_base_chain(chain), afi->nops); + err = nf_register_net_hooks(net, nft_base_chain(chain)->ops, + afi->nops); if (err < 0) goto err; @@ -586,7 +568,8 @@ err: if (i-- <= 0) break; - nft_unregister_basechain(nft_base_chain(chain), afi->nops); + nf_unregister_net_hooks(net, nft_base_chain(chain)->ops, + afi->nops); } return err; } @@ -600,9 +583,11 @@ static void nf_tables_table_disable(struct net *net, list_for_each_entry(chain, &table->chains, list) { if (!nft_is_active_next(net, chain)) continue; - if (chain->flags & NFT_BASE_CHAIN) - nft_unregister_basechain(nft_base_chain(chain), - afi->nops); + if (!(chain->flags & NFT_BASE_CHAIN)) + continue; + + nf_unregister_net_hooks(net, nft_base_chain(chain)->ops, + afi->nops); } } @@ -1451,7 +1436,7 @@ static int nf_tables_newchain(struct net *net, struct sock *nlsk, chain->table = table; nla_strlcpy(chain->name, name, NFT_CHAIN_MAXNAMELEN); - err = nf_tables_register_hooks(table, chain, afi->nops); + err = nf_tables_register_hooks(net, table, chain, afi->nops); if (err < 0) goto err1; @@ -1464,7 +1449,7 @@ static int nf_tables_newchain(struct net *net, struct sock *nlsk, list_add_tail_rcu(&chain->list, &table->chains); return 0; err2: - nf_tables_unregister_hooks(table, chain, afi->nops); + nf_tables_unregister_hooks(net, table, chain, afi->nops); err1: nf_tables_chain_destroy(chain); return err; @@ -3995,7 +3980,8 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) case NFT_MSG_DELCHAIN: list_del_rcu(&trans->ctx.chain->list); nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN); - nf_tables_unregister_hooks(trans->ctx.table, + nf_tables_unregister_hooks(trans->ctx.net, + trans->ctx.table, trans->ctx.chain, trans->ctx.afi->nops); break; @@ -4120,7 +4106,8 @@ static int nf_tables_abort(struct net *net, struct sk_buff *skb) } else { trans->ctx.table->use--; list_del_rcu(&trans->ctx.chain->list); - nf_tables_unregister_hooks(trans->ctx.table, + nf_tables_unregister_hooks(trans->ctx.net, + trans->ctx.table, trans->ctx.chain, trans->ctx.afi->nops); } @@ -4662,7 +4649,7 @@ int __nft_release_basechain(struct nft_ctx *ctx) BUG_ON(!(ctx->chain->flags & NFT_BASE_CHAIN)); - nf_tables_unregister_hooks(ctx->chain->table, ctx->chain, + nf_tables_unregister_hooks(ctx->net, ctx->chain->table, ctx->chain, ctx->afi->nops); list_for_each_entry_safe(rule, nr, &ctx->chain->rules, list) { list_del(&rule->list); @@ -4691,7 +4678,8 @@ static void __nft_release_afinfo(struct net *net, struct nft_af_info *afi) list_for_each_entry_safe(table, nt, &afi->tables, list) { list_for_each_entry(chain, &table->chains, list) - nf_tables_unregister_hooks(table, chain, afi->nops); + nf_tables_unregister_hooks(net, table, chain, + afi->nops); /* No packets are walking on these chains anymore. */ ctx.table = table; list_for_each_entry(chain, &table->chains, list) { -- cgit v0.10.2 From 0071e184a535e40ce487528cb04f4690cb0da881 Mon Sep 17 00:00:00 2001 From: Arturo Borrero Date: Thu, 23 Jun 2016 12:24:08 +0200 Subject: netfilter: nf_tables: add support for inverted logic in nft_lookup Introduce a new configuration option for this expression, which allows users to invert the logic of set lookups. In _init() we will now return EINVAL if NFT_LOOKUP_F_INV is in anyway related to a map lookup. The code in the _eval() function has been untangled and updated to sopport the XOR of options, as we should consider 4 cases: * lookup false, invert false -> NFT_BREAK * lookup false, invert true -> return w/o NFT_BREAK * lookup true, invert false -> return w/o NFT_BREAK * lookup true, invert true -> NFT_BREAK Signed-off-by: Arturo Borrero Gonzalez Signed-off-by: Pablo Neira Ayuso diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 6a4dbe0..01751fa 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -546,6 +546,10 @@ enum nft_cmp_attributes { }; #define NFTA_CMP_MAX (__NFTA_CMP_MAX - 1) +enum nft_lookup_flags { + NFT_LOOKUP_F_INV = (1 << 0), +}; + /** * enum nft_lookup_attributes - nf_tables set lookup expression netlink attributes * @@ -553,6 +557,7 @@ enum nft_cmp_attributes { * @NFTA_LOOKUP_SREG: source register of the data to look for (NLA_U32: nft_registers) * @NFTA_LOOKUP_DREG: destination register (NLA_U32: nft_registers) * @NFTA_LOOKUP_SET_ID: uniquely identifies a set in a transaction (NLA_U32) + * @NFTA_LOOKUP_FLAGS: flags (NLA_U32: enum nft_lookup_flags) */ enum nft_lookup_attributes { NFTA_LOOKUP_UNSPEC, @@ -560,6 +565,7 @@ enum nft_lookup_attributes { NFTA_LOOKUP_SREG, NFTA_LOOKUP_DREG, NFTA_LOOKUP_SET_ID, + NFTA_LOOKUP_FLAGS, __NFTA_LOOKUP_MAX }; #define NFTA_LOOKUP_MAX (__NFTA_LOOKUP_MAX - 1) diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index 8a102cf..b8d18f5 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -22,6 +22,7 @@ struct nft_lookup { struct nft_set *set; enum nft_registers sreg:8; enum nft_registers dreg:8; + bool invert; struct nft_set_binding binding; }; @@ -32,14 +33,20 @@ static void nft_lookup_eval(const struct nft_expr *expr, const struct nft_lookup *priv = nft_expr_priv(expr); const struct nft_set *set = priv->set; const struct nft_set_ext *ext; + bool found; - if (set->ops->lookup(set, ®s->data[priv->sreg], &ext)) { - if (set->flags & NFT_SET_MAP) - nft_data_copy(®s->data[priv->dreg], - nft_set_ext_data(ext), set->dlen); + found = set->ops->lookup(set, ®s->data[priv->sreg], &ext) ^ + priv->invert; + + if (!found) { + regs->verdict.code = NFT_BREAK; return; } - regs->verdict.code = NFT_BREAK; + + if (found && set->flags & NFT_SET_MAP) + nft_data_copy(®s->data[priv->dreg], + nft_set_ext_data(ext), set->dlen); + } static const struct nla_policy nft_lookup_policy[NFTA_LOOKUP_MAX + 1] = { @@ -47,6 +54,7 @@ static const struct nla_policy nft_lookup_policy[NFTA_LOOKUP_MAX + 1] = { [NFTA_LOOKUP_SET_ID] = { .type = NLA_U32 }, [NFTA_LOOKUP_SREG] = { .type = NLA_U32 }, [NFTA_LOOKUP_DREG] = { .type = NLA_U32 }, + [NFTA_LOOKUP_FLAGS] = { .type = NLA_U32 }, }; static int nft_lookup_init(const struct nft_ctx *ctx, @@ -56,6 +64,7 @@ static int nft_lookup_init(const struct nft_ctx *ctx, struct nft_lookup *priv = nft_expr_priv(expr); u8 genmask = nft_genmask_next(ctx->net); struct nft_set *set; + u32 flags; int err; if (tb[NFTA_LOOKUP_SET] == NULL || @@ -81,7 +90,22 @@ static int nft_lookup_init(const struct nft_ctx *ctx, if (err < 0) return err; + if (tb[NFTA_LOOKUP_FLAGS]) { + flags = ntohl(nla_get_be32(tb[NFTA_LOOKUP_FLAGS])); + + if (flags & ~NFT_LOOKUP_F_INV) + return -EINVAL; + + if (flags & NFT_LOOKUP_F_INV) { + if (set->flags & NFT_SET_MAP) + return -EINVAL; + priv->invert = true; + } + } + if (tb[NFTA_LOOKUP_DREG] != NULL) { + if (priv->invert) + return -EINVAL; if (!(set->flags & NFT_SET_MAP)) return -EINVAL; @@ -114,6 +138,7 @@ static void nft_lookup_destroy(const struct nft_ctx *ctx, static int nft_lookup_dump(struct sk_buff *skb, const struct nft_expr *expr) { const struct nft_lookup *priv = nft_expr_priv(expr); + u32 flags = priv->invert ? NFT_LOOKUP_F_INV : 0; if (nla_put_string(skb, NFTA_LOOKUP_SET, priv->set->name)) goto nla_put_failure; @@ -122,6 +147,8 @@ static int nft_lookup_dump(struct sk_buff *skb, const struct nft_expr *expr) if (priv->set->flags & NFT_SET_MAP) if (nft_dump_register(skb, NFTA_LOOKUP_DREG, priv->dreg)) goto nla_put_failure; + if (nla_put_be32(skb, NFTA_LOOKUP_FLAGS, htonl(flags))) + goto nla_put_failure; return 0; nla_put_failure: -- cgit v0.10.2 From 0d3cd4b6b49865e83ae648b66cf815d466085914 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Tue, 21 Jun 2016 12:28:19 -0400 Subject: net: dsa: mv88e6xxx: move driver in its own folder With the upcoming support for cross-chip operations and other mv88e6xxx enhancements, new files will be added. Similarly to mlxsw or b53, move mv88e6xxx files into their own folder. In the meantime, update the MAINTAINERS entry to please checkpatch.pl, by replacing the invalid 88E6352 entry with 88E6XXX, maintained by Andrew and myself. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/MAINTAINERS b/MAINTAINERS index c9544f5..3a171a9 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7171,6 +7171,12 @@ W: http://www.kernel.org/doc/man-pages L: linux-man@vger.kernel.org S: Maintained +MARVELL 88E6XXX ETHERNET SWITCH FABRIC DRIVER +M: Andrew Lunn +M: Vivien Didelot +S: Maintained +F: drivers/net/dsa/mv88e6xxx/ + MARVELL ARMADA DRM SUPPORT M: Russell King S: Maintained @@ -7178,11 +7184,6 @@ F: drivers/gpu/drm/armada/ F: include/uapi/drm/armada_drm.h F: Documentation/devicetree/bindings/display/armada/ -MARVELL 88E6352 DSA support -M: Guenter Roeck -S: Maintained -F: drivers/net/dsa/mv88e6352.c - MARVELL CRYPTO DRIVER M: Boris Brezillon M: Arnaud Ebalard diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig index be481e1..8f45443 100644 --- a/drivers/net/dsa/Kconfig +++ b/drivers/net/dsa/Kconfig @@ -9,14 +9,6 @@ config NET_DSA_MV88E6060 This enables support for the Marvell 88E6060 ethernet switch chip. -config NET_DSA_MV88E6XXX - tristate "Marvell 88E6xxx Ethernet switch chip support" - depends on NET_DSA - select NET_DSA_TAG_EDSA - ---help--- - This enables support for most of the Marvell 88E6xxx models of - Ethernet switch chips, except 88E6060. - config NET_DSA_BCM_SF2 tristate "Broadcom Starfighter 2 Ethernet switch support" depends on HAS_IOMEM && NET_DSA @@ -30,4 +22,6 @@ config NET_DSA_BCM_SF2 source "drivers/net/dsa/b53/Kconfig" +source "drivers/net/dsa/mv88e6xxx/Kconfig" + endmenu diff --git a/drivers/net/dsa/Makefile b/drivers/net/dsa/Makefile index 97bc70a..ca1e71b 100644 --- a/drivers/net/dsa/Makefile +++ b/drivers/net/dsa/Makefile @@ -1,5 +1,5 @@ obj-$(CONFIG_NET_DSA_MV88E6060) += mv88e6060.o -obj-$(CONFIG_NET_DSA_MV88E6XXX) += mv88e6xxx.o obj-$(CONFIG_NET_DSA_BCM_SF2) += bcm_sf2.o obj-y += b53/ +obj-y += mv88e6xxx/ diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c deleted file mode 100644 index 9b116d8..0000000 --- a/drivers/net/dsa/mv88e6xxx.c +++ /dev/null @@ -1,3953 +0,0 @@ -/* - * net/dsa/mv88e6xxx.c - Marvell 88e6xxx switch chip support - * Copyright (c) 2008 Marvell Semiconductor - * - * Copyright (c) 2015 CMC Electronics, Inc. - * Added support for VLAN Table Unit operations - * - * Copyright (c) 2016 Andrew Lunn - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "mv88e6xxx.h" - -static void assert_reg_lock(struct mv88e6xxx_priv_state *ps) -{ - if (unlikely(!mutex_is_locked(&ps->reg_lock))) { - dev_err(ps->dev, "Switch registers lock not held!\n"); - dump_stack(); - } -} - -/* The switch ADDR[4:1] configuration pins define the chip SMI device address - * (ADDR[0] is always zero, thus only even SMI addresses can be strapped). - * - * When ADDR is all zero, the chip uses Single-chip Addressing Mode, assuming it - * is the only device connected to the SMI master. In this mode it responds to - * all 32 possible SMI addresses, and thus maps directly the internal devices. - * - * When ADDR is non-zero, the chip uses Multi-chip Addressing Mode, allowing - * multiple devices to share the SMI interface. In this mode it responds to only - * 2 registers, used to indirectly access the internal SMI devices. - */ - -static int mv88e6xxx_smi_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 *val) -{ - if (!ps->smi_ops) - return -EOPNOTSUPP; - - return ps->smi_ops->read(ps, addr, reg, val); -} - -static int mv88e6xxx_smi_write(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 val) -{ - if (!ps->smi_ops) - return -EOPNOTSUPP; - - return ps->smi_ops->write(ps, addr, reg, val); -} - -static int mv88e6xxx_smi_single_chip_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 *val) -{ - int ret; - - ret = mdiobus_read_nested(ps->bus, addr, reg); - if (ret < 0) - return ret; - - *val = ret & 0xffff; - - return 0; -} - -static int mv88e6xxx_smi_single_chip_write(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 val) -{ - int ret; - - ret = mdiobus_write_nested(ps->bus, addr, reg, val); - if (ret < 0) - return ret; - - return 0; -} - -static const struct mv88e6xxx_ops mv88e6xxx_smi_single_chip_ops = { - .read = mv88e6xxx_smi_single_chip_read, - .write = mv88e6xxx_smi_single_chip_write, -}; - -static int mv88e6xxx_smi_multi_chip_wait(struct mv88e6xxx_priv_state *ps) -{ - int ret; - int i; - - for (i = 0; i < 16; i++) { - ret = mdiobus_read_nested(ps->bus, ps->sw_addr, SMI_CMD); - if (ret < 0) - return ret; - - if ((ret & SMI_CMD_BUSY) == 0) - return 0; - } - - return -ETIMEDOUT; -} - -static int mv88e6xxx_smi_multi_chip_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 *val) -{ - int ret; - - /* Wait for the bus to become free. */ - ret = mv88e6xxx_smi_multi_chip_wait(ps); - if (ret < 0) - return ret; - - /* Transmit the read command. */ - ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_CMD, - SMI_CMD_OP_22_READ | (addr << 5) | reg); - if (ret < 0) - return ret; - - /* Wait for the read command to complete. */ - ret = mv88e6xxx_smi_multi_chip_wait(ps); - if (ret < 0) - return ret; - - /* Read the data. */ - ret = mdiobus_read_nested(ps->bus, ps->sw_addr, SMI_DATA); - if (ret < 0) - return ret; - - *val = ret & 0xffff; - - return 0; -} - -static int mv88e6xxx_smi_multi_chip_write(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 val) -{ - int ret; - - /* Wait for the bus to become free. */ - ret = mv88e6xxx_smi_multi_chip_wait(ps); - if (ret < 0) - return ret; - - /* Transmit the data to write. */ - ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_DATA, val); - if (ret < 0) - return ret; - - /* Transmit the write command. */ - ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_CMD, - SMI_CMD_OP_22_WRITE | (addr << 5) | reg); - if (ret < 0) - return ret; - - /* Wait for the write command to complete. */ - ret = mv88e6xxx_smi_multi_chip_wait(ps); - if (ret < 0) - return ret; - - return 0; -} - -static const struct mv88e6xxx_ops mv88e6xxx_smi_multi_chip_ops = { - .read = mv88e6xxx_smi_multi_chip_read, - .write = mv88e6xxx_smi_multi_chip_write, -}; - -static int mv88e6xxx_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 *val) -{ - int err; - - assert_reg_lock(ps); - - err = mv88e6xxx_smi_read(ps, addr, reg, val); - if (err) - return err; - - dev_dbg(ps->dev, "<- addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", - addr, reg, *val); - - return 0; -} - -static int mv88e6xxx_write(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 val) -{ - int err; - - assert_reg_lock(ps); - - err = mv88e6xxx_smi_write(ps, addr, reg, val); - if (err) - return err; - - dev_dbg(ps->dev, "-> addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", - addr, reg, val); - - return 0; -} - -static int _mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg) -{ - u16 val; - int err; - - err = mv88e6xxx_read(ps, addr, reg, &val); - if (err) - return err; - - return val; -} - -static int mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, int addr, - int reg) -{ - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_reg_read(ps, addr, reg); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int _mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, - int reg, u16 val) -{ - return mv88e6xxx_write(ps, addr, reg, val); -} - -static int mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, - int reg, u16 val) -{ - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_reg_write(ps, addr, reg, val); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int mv88e6xxx_set_addr_direct(struct dsa_switch *ds, u8 *addr) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int err; - - err = mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MAC_01, - (addr[0] << 8) | addr[1]); - if (err) - return err; - - err = mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MAC_23, - (addr[2] << 8) | addr[3]); - if (err) - return err; - - return mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MAC_45, - (addr[4] << 8) | addr[5]); -} - -static int mv88e6xxx_set_addr_indirect(struct dsa_switch *ds, u8 *addr) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - int i; - - for (i = 0; i < 6; i++) { - int j; - - /* Write the MAC address byte. */ - ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SWITCH_MAC, - GLOBAL2_SWITCH_MAC_BUSY | - (i << 8) | addr[i]); - if (ret) - return ret; - - /* Wait for the write to complete. */ - for (j = 0; j < 16; j++) { - ret = mv88e6xxx_reg_read(ps, REG_GLOBAL2, - GLOBAL2_SWITCH_MAC); - if (ret < 0) - return ret; - - if ((ret & GLOBAL2_SWITCH_MAC_BUSY) == 0) - break; - } - if (j == 16) - return -ETIMEDOUT; - } - - return 0; -} - -static int mv88e6xxx_set_addr(struct dsa_switch *ds, u8 *addr) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SWITCH_MAC)) - return mv88e6xxx_set_addr_indirect(ds, addr); - else - return mv88e6xxx_set_addr_direct(ds, addr); -} - -static int mv88e6xxx_mdio_read_direct(struct mv88e6xxx_priv_state *ps, - int addr, int regnum) -{ - if (addr >= 0) - return _mv88e6xxx_reg_read(ps, addr, regnum); - return 0xffff; -} - -static int mv88e6xxx_mdio_write_direct(struct mv88e6xxx_priv_state *ps, - int addr, int regnum, u16 val) -{ - if (addr >= 0) - return _mv88e6xxx_reg_write(ps, addr, regnum, val); - return 0; -} - -static int mv88e6xxx_ppu_disable(struct mv88e6xxx_priv_state *ps) -{ - int ret; - unsigned long timeout; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_CONTROL); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, - ret & ~GLOBAL_CONTROL_PPU_ENABLE); - if (ret) - return ret; - - timeout = jiffies + 1 * HZ; - while (time_before(jiffies, timeout)) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATUS); - if (ret < 0) - return ret; - - usleep_range(1000, 2000); - if ((ret & GLOBAL_STATUS_PPU_MASK) != - GLOBAL_STATUS_PPU_POLLING) - return 0; - } - - return -ETIMEDOUT; -} - -static int mv88e6xxx_ppu_enable(struct mv88e6xxx_priv_state *ps) -{ - int ret, err; - unsigned long timeout; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_CONTROL); - if (ret < 0) - return ret; - - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, - ret | GLOBAL_CONTROL_PPU_ENABLE); - if (err) - return err; - - timeout = jiffies + 1 * HZ; - while (time_before(jiffies, timeout)) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATUS); - if (ret < 0) - return ret; - - usleep_range(1000, 2000); - if ((ret & GLOBAL_STATUS_PPU_MASK) == - GLOBAL_STATUS_PPU_POLLING) - return 0; - } - - return -ETIMEDOUT; -} - -static void mv88e6xxx_ppu_reenable_work(struct work_struct *ugly) -{ - struct mv88e6xxx_priv_state *ps; - - ps = container_of(ugly, struct mv88e6xxx_priv_state, ppu_work); - - mutex_lock(&ps->reg_lock); - - if (mutex_trylock(&ps->ppu_mutex)) { - if (mv88e6xxx_ppu_enable(ps) == 0) - ps->ppu_disabled = 0; - mutex_unlock(&ps->ppu_mutex); - } - - mutex_unlock(&ps->reg_lock); -} - -static void mv88e6xxx_ppu_reenable_timer(unsigned long _ps) -{ - struct mv88e6xxx_priv_state *ps = (void *)_ps; - - schedule_work(&ps->ppu_work); -} - -static int mv88e6xxx_ppu_access_get(struct mv88e6xxx_priv_state *ps) -{ - int ret; - - mutex_lock(&ps->ppu_mutex); - - /* If the PHY polling unit is enabled, disable it so that - * we can access the PHY registers. If it was already - * disabled, cancel the timer that is going to re-enable - * it. - */ - if (!ps->ppu_disabled) { - ret = mv88e6xxx_ppu_disable(ps); - if (ret < 0) { - mutex_unlock(&ps->ppu_mutex); - return ret; - } - ps->ppu_disabled = 1; - } else { - del_timer(&ps->ppu_timer); - ret = 0; - } - - return ret; -} - -static void mv88e6xxx_ppu_access_put(struct mv88e6xxx_priv_state *ps) -{ - /* Schedule a timer to re-enable the PHY polling unit. */ - mod_timer(&ps->ppu_timer, jiffies + msecs_to_jiffies(10)); - mutex_unlock(&ps->ppu_mutex); -} - -static void mv88e6xxx_ppu_state_init(struct mv88e6xxx_priv_state *ps) -{ - mutex_init(&ps->ppu_mutex); - INIT_WORK(&ps->ppu_work, mv88e6xxx_ppu_reenable_work); - init_timer(&ps->ppu_timer); - ps->ppu_timer.data = (unsigned long)ps; - ps->ppu_timer.function = mv88e6xxx_ppu_reenable_timer; -} - -static int mv88e6xxx_mdio_read_ppu(struct mv88e6xxx_priv_state *ps, int addr, - int regnum) -{ - int ret; - - ret = mv88e6xxx_ppu_access_get(ps); - if (ret >= 0) { - ret = _mv88e6xxx_reg_read(ps, addr, regnum); - mv88e6xxx_ppu_access_put(ps); - } - - return ret; -} - -static int mv88e6xxx_mdio_write_ppu(struct mv88e6xxx_priv_state *ps, int addr, - int regnum, u16 val) -{ - int ret; - - ret = mv88e6xxx_ppu_access_get(ps); - if (ret >= 0) { - ret = _mv88e6xxx_reg_write(ps, addr, regnum, val); - mv88e6xxx_ppu_access_put(ps); - } - - return ret; -} - -static bool mv88e6xxx_6065_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6065; -} - -static bool mv88e6xxx_6095_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6095; -} - -static bool mv88e6xxx_6097_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6097; -} - -static bool mv88e6xxx_6165_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6165; -} - -static bool mv88e6xxx_6185_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6185; -} - -static bool mv88e6xxx_6320_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6320; -} - -static bool mv88e6xxx_6351_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6351; -} - -static bool mv88e6xxx_6352_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6352; -} - -static unsigned int mv88e6xxx_num_databases(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->num_databases; -} - -static bool mv88e6xxx_has_fid_reg(struct mv88e6xxx_priv_state *ps) -{ - /* Does the device have dedicated FID registers for ATU and VTU ops? */ - if (mv88e6xxx_6097_family(ps) || mv88e6xxx_6165_family(ps) || - mv88e6xxx_6351_family(ps) || mv88e6xxx_6352_family(ps)) - return true; - - return false; -} - -/* We expect the switch to perform auto negotiation if there is a real - * phy. However, in the case of a fixed link phy, we force the port - * settings from the fixed link settings. - */ -static void mv88e6xxx_adjust_link(struct dsa_switch *ds, int port, - struct phy_device *phydev) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - u32 reg; - int ret; - - if (!phy_is_pseudo_fixed_link(phydev)) - return; - - mutex_lock(&ps->reg_lock); - - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_PCS_CTRL); - if (ret < 0) - goto out; - - reg = ret & ~(PORT_PCS_CTRL_LINK_UP | - PORT_PCS_CTRL_FORCE_LINK | - PORT_PCS_CTRL_DUPLEX_FULL | - PORT_PCS_CTRL_FORCE_DUPLEX | - PORT_PCS_CTRL_UNFORCED); - - reg |= PORT_PCS_CTRL_FORCE_LINK; - if (phydev->link) - reg |= PORT_PCS_CTRL_LINK_UP; - - if (mv88e6xxx_6065_family(ps) && phydev->speed > SPEED_100) - goto out; - - switch (phydev->speed) { - case SPEED_1000: - reg |= PORT_PCS_CTRL_1000; - break; - case SPEED_100: - reg |= PORT_PCS_CTRL_100; - break; - case SPEED_10: - reg |= PORT_PCS_CTRL_10; - break; - default: - pr_info("Unknown speed"); - goto out; - } - - reg |= PORT_PCS_CTRL_FORCE_DUPLEX; - if (phydev->duplex == DUPLEX_FULL) - reg |= PORT_PCS_CTRL_DUPLEX_FULL; - - if ((mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps)) && - (port >= ps->info->num_ports - 2)) { - if (phydev->interface == PHY_INTERFACE_MODE_RGMII_RXID) - reg |= PORT_PCS_CTRL_RGMII_DELAY_RXCLK; - if (phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID) - reg |= PORT_PCS_CTRL_RGMII_DELAY_TXCLK; - if (phydev->interface == PHY_INTERFACE_MODE_RGMII_ID) - reg |= (PORT_PCS_CTRL_RGMII_DELAY_RXCLK | - PORT_PCS_CTRL_RGMII_DELAY_TXCLK); - } - _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_PCS_CTRL, reg); - -out: - mutex_unlock(&ps->reg_lock); -} - -static int _mv88e6xxx_stats_wait(struct mv88e6xxx_priv_state *ps) -{ - int ret; - int i; - - for (i = 0; i < 10; i++) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATS_OP); - if ((ret & GLOBAL_STATS_OP_BUSY) == 0) - return 0; - } - - return -ETIMEDOUT; -} - -static int _mv88e6xxx_stats_snapshot(struct mv88e6xxx_priv_state *ps, - int port) -{ - int ret; - - if (mv88e6xxx_6320_family(ps) || mv88e6xxx_6352_family(ps)) - port = (port + 1) << 5; - - /* Snapshot the hardware statistics counters for this port. */ - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_STATS_OP, - GLOBAL_STATS_OP_CAPTURE_PORT | - GLOBAL_STATS_OP_HIST_RX_TX | port); - if (ret < 0) - return ret; - - /* Wait for the snapshotting to complete. */ - ret = _mv88e6xxx_stats_wait(ps); - if (ret < 0) - return ret; - - return 0; -} - -static void _mv88e6xxx_stats_read(struct mv88e6xxx_priv_state *ps, - int stat, u32 *val) -{ - u32 _val; - int ret; - - *val = 0; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_STATS_OP, - GLOBAL_STATS_OP_READ_CAPTURED | - GLOBAL_STATS_OP_HIST_RX_TX | stat); - if (ret < 0) - return; - - ret = _mv88e6xxx_stats_wait(ps); - if (ret < 0) - return; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATS_COUNTER_32); - if (ret < 0) - return; - - _val = ret << 16; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATS_COUNTER_01); - if (ret < 0) - return; - - *val = _val | ret; -} - -static struct mv88e6xxx_hw_stat mv88e6xxx_hw_stats[] = { - { "in_good_octets", 8, 0x00, BANK0, }, - { "in_bad_octets", 4, 0x02, BANK0, }, - { "in_unicast", 4, 0x04, BANK0, }, - { "in_broadcasts", 4, 0x06, BANK0, }, - { "in_multicasts", 4, 0x07, BANK0, }, - { "in_pause", 4, 0x16, BANK0, }, - { "in_undersize", 4, 0x18, BANK0, }, - { "in_fragments", 4, 0x19, BANK0, }, - { "in_oversize", 4, 0x1a, BANK0, }, - { "in_jabber", 4, 0x1b, BANK0, }, - { "in_rx_error", 4, 0x1c, BANK0, }, - { "in_fcs_error", 4, 0x1d, BANK0, }, - { "out_octets", 8, 0x0e, BANK0, }, - { "out_unicast", 4, 0x10, BANK0, }, - { "out_broadcasts", 4, 0x13, BANK0, }, - { "out_multicasts", 4, 0x12, BANK0, }, - { "out_pause", 4, 0x15, BANK0, }, - { "excessive", 4, 0x11, BANK0, }, - { "collisions", 4, 0x1e, BANK0, }, - { "deferred", 4, 0x05, BANK0, }, - { "single", 4, 0x14, BANK0, }, - { "multiple", 4, 0x17, BANK0, }, - { "out_fcs_error", 4, 0x03, BANK0, }, - { "late", 4, 0x1f, BANK0, }, - { "hist_64bytes", 4, 0x08, BANK0, }, - { "hist_65_127bytes", 4, 0x09, BANK0, }, - { "hist_128_255bytes", 4, 0x0a, BANK0, }, - { "hist_256_511bytes", 4, 0x0b, BANK0, }, - { "hist_512_1023bytes", 4, 0x0c, BANK0, }, - { "hist_1024_max_bytes", 4, 0x0d, BANK0, }, - { "sw_in_discards", 4, 0x10, PORT, }, - { "sw_in_filtered", 2, 0x12, PORT, }, - { "sw_out_filtered", 2, 0x13, PORT, }, - { "in_discards", 4, 0x00 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_filtered", 4, 0x01 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_accepted", 4, 0x02 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_bad_accepted", 4, 0x03 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_good_avb_class_a", 4, 0x04 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_good_avb_class_b", 4, 0x05 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_bad_avb_class_a", 4, 0x06 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_bad_avb_class_b", 4, 0x07 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "tcam_counter_0", 4, 0x08 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "tcam_counter_1", 4, 0x09 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "tcam_counter_2", 4, 0x0a | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "tcam_counter_3", 4, 0x0b | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_da_unknown", 4, 0x0e | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_management", 4, 0x0f | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_0", 4, 0x10 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_1", 4, 0x11 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_2", 4, 0x12 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_3", 4, 0x13 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_4", 4, 0x14 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_5", 4, 0x15 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_6", 4, 0x16 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_7", 4, 0x17 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_cut_through", 4, 0x18 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_octets_a", 4, 0x1a | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_octets_b", 4, 0x1b | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_management", 4, 0x1f | GLOBAL_STATS_OP_BANK_1, BANK1, }, -}; - -static bool mv88e6xxx_has_stat(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_hw_stat *stat) -{ - switch (stat->type) { - case BANK0: - return true; - case BANK1: - return mv88e6xxx_6320_family(ps); - case PORT: - return mv88e6xxx_6095_family(ps) || - mv88e6xxx_6185_family(ps) || - mv88e6xxx_6097_family(ps) || - mv88e6xxx_6165_family(ps) || - mv88e6xxx_6351_family(ps) || - mv88e6xxx_6352_family(ps); - } - return false; -} - -static uint64_t _mv88e6xxx_get_ethtool_stat(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_hw_stat *s, - int port) -{ - u32 low; - u32 high = 0; - int ret; - u64 value; - - switch (s->type) { - case PORT: - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), s->reg); - if (ret < 0) - return UINT64_MAX; - - low = ret; - if (s->sizeof_stat == 4) { - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), - s->reg + 1); - if (ret < 0) - return UINT64_MAX; - high = ret; - } - break; - case BANK0: - case BANK1: - _mv88e6xxx_stats_read(ps, s->reg, &low); - if (s->sizeof_stat == 8) - _mv88e6xxx_stats_read(ps, s->reg + 1, &high); - } - value = (((u64)high) << 16) | low; - return value; -} - -static void mv88e6xxx_get_strings(struct dsa_switch *ds, int port, - uint8_t *data) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_hw_stat *stat; - int i, j; - - for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { - stat = &mv88e6xxx_hw_stats[i]; - if (mv88e6xxx_has_stat(ps, stat)) { - memcpy(data + j * ETH_GSTRING_LEN, stat->string, - ETH_GSTRING_LEN); - j++; - } - } -} - -static int mv88e6xxx_get_sset_count(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_hw_stat *stat; - int i, j; - - for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { - stat = &mv88e6xxx_hw_stats[i]; - if (mv88e6xxx_has_stat(ps, stat)) - j++; - } - return j; -} - -static void mv88e6xxx_get_ethtool_stats(struct dsa_switch *ds, int port, - uint64_t *data) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_hw_stat *stat; - int ret; - int i, j; - - mutex_lock(&ps->reg_lock); - - ret = _mv88e6xxx_stats_snapshot(ps, port); - if (ret < 0) { - mutex_unlock(&ps->reg_lock); - return; - } - for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { - stat = &mv88e6xxx_hw_stats[i]; - if (mv88e6xxx_has_stat(ps, stat)) { - data[j] = _mv88e6xxx_get_ethtool_stat(ps, stat, port); - j++; - } - } - - mutex_unlock(&ps->reg_lock); -} - -static int mv88e6xxx_get_regs_len(struct dsa_switch *ds, int port) -{ - return 32 * sizeof(u16); -} - -static void mv88e6xxx_get_regs(struct dsa_switch *ds, int port, - struct ethtool_regs *regs, void *_p) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - u16 *p = _p; - int i; - - regs->version = 0; - - memset(p, 0xff, 32 * sizeof(u16)); - - mutex_lock(&ps->reg_lock); - - for (i = 0; i < 32; i++) { - int ret; - - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), i); - if (ret >= 0) - p[i] = ret; - } - - mutex_unlock(&ps->reg_lock); -} - -static int _mv88e6xxx_wait(struct mv88e6xxx_priv_state *ps, int reg, int offset, - u16 mask) -{ - unsigned long timeout = jiffies + HZ / 10; - - while (time_before(jiffies, timeout)) { - int ret; - - ret = _mv88e6xxx_reg_read(ps, reg, offset); - if (ret < 0) - return ret; - if (!(ret & mask)) - return 0; - - usleep_range(1000, 2000); - } - return -ETIMEDOUT; -} - -static int mv88e6xxx_wait(struct mv88e6xxx_priv_state *ps, int reg, - int offset, u16 mask) -{ - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_wait(ps, reg, offset, mask); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int mv88e6xxx_mdio_wait(struct mv88e6xxx_priv_state *ps) -{ - return _mv88e6xxx_wait(ps, REG_GLOBAL2, GLOBAL2_SMI_OP, - GLOBAL2_SMI_OP_BUSY); -} - -static int mv88e6xxx_eeprom_load_wait(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - return mv88e6xxx_wait(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_LOAD); -} - -static int mv88e6xxx_eeprom_busy_wait(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - return mv88e6xxx_wait(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_BUSY); -} - -static int mv88e6xxx_read_eeprom_word(struct dsa_switch *ds, int addr) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - mutex_lock(&ps->eeprom_mutex); - - ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_READ | - (addr & GLOBAL2_EEPROM_OP_ADDR_MASK)); - if (ret < 0) - goto error; - - ret = mv88e6xxx_eeprom_busy_wait(ds); - if (ret < 0) - goto error; - - ret = mv88e6xxx_reg_read(ps, REG_GLOBAL2, GLOBAL2_EEPROM_DATA); -error: - mutex_unlock(&ps->eeprom_mutex); - return ret; -} - -static int mv88e6xxx_get_eeprom_len(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) - return ps->eeprom_len; - - return 0; -} - -static int mv88e6xxx_get_eeprom(struct dsa_switch *ds, - struct ethtool_eeprom *eeprom, u8 *data) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int offset; - int len; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) - return -EOPNOTSUPP; - - offset = eeprom->offset; - len = eeprom->len; - eeprom->len = 0; - - eeprom->magic = 0xc3ec4951; - - ret = mv88e6xxx_eeprom_load_wait(ds); - if (ret < 0) - return ret; - - if (offset & 1) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - *data++ = (word >> 8) & 0xff; - - offset++; - len--; - eeprom->len++; - } - - while (len >= 2) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - *data++ = word & 0xff; - *data++ = (word >> 8) & 0xff; - - offset += 2; - len -= 2; - eeprom->len += 2; - } - - if (len) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - *data++ = word & 0xff; - - offset++; - len--; - eeprom->len++; - } - - return 0; -} - -static int mv88e6xxx_eeprom_is_readonly(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - ret = mv88e6xxx_reg_read(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP); - if (ret < 0) - return ret; - - if (!(ret & GLOBAL2_EEPROM_OP_WRITE_EN)) - return -EROFS; - - return 0; -} - -static int mv88e6xxx_write_eeprom_word(struct dsa_switch *ds, int addr, - u16 data) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - mutex_lock(&ps->eeprom_mutex); - - ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_EEPROM_DATA, data); - if (ret < 0) - goto error; - - ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_WRITE | - (addr & GLOBAL2_EEPROM_OP_ADDR_MASK)); - if (ret < 0) - goto error; - - ret = mv88e6xxx_eeprom_busy_wait(ds); -error: - mutex_unlock(&ps->eeprom_mutex); - return ret; -} - -static int mv88e6xxx_set_eeprom(struct dsa_switch *ds, - struct ethtool_eeprom *eeprom, u8 *data) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int offset; - int ret; - int len; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) - return -EOPNOTSUPP; - - if (eeprom->magic != 0xc3ec4951) - return -EINVAL; - - ret = mv88e6xxx_eeprom_is_readonly(ds); - if (ret) - return ret; - - offset = eeprom->offset; - len = eeprom->len; - eeprom->len = 0; - - ret = mv88e6xxx_eeprom_load_wait(ds); - if (ret < 0) - return ret; - - if (offset & 1) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - word = (*data++ << 8) | (word & 0xff); - - ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); - if (ret < 0) - return ret; - - offset++; - len--; - eeprom->len++; - } - - while (len >= 2) { - int word; - - word = *data++; - word |= *data++ << 8; - - ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); - if (ret < 0) - return ret; - - offset += 2; - len -= 2; - eeprom->len += 2; - } - - if (len) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - word = (word & 0xff00) | *data++; - - ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); - if (ret < 0) - return ret; - - offset++; - len--; - eeprom->len++; - } - - return 0; -} - -static int _mv88e6xxx_atu_wait(struct mv88e6xxx_priv_state *ps) -{ - return _mv88e6xxx_wait(ps, REG_GLOBAL, GLOBAL_ATU_OP, - GLOBAL_ATU_OP_BUSY); -} - -static int mv88e6xxx_mdio_read_indirect(struct mv88e6xxx_priv_state *ps, - int addr, int regnum) -{ - int ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SMI_OP, - GLOBAL2_SMI_OP_22_READ | (addr << 5) | - regnum); - if (ret < 0) - return ret; - - ret = mv88e6xxx_mdio_wait(ps); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL2, GLOBAL2_SMI_DATA); - - return ret; -} - -static int mv88e6xxx_mdio_write_indirect(struct mv88e6xxx_priv_state *ps, - int addr, int regnum, u16 val) -{ - int ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SMI_DATA, val); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SMI_OP, - GLOBAL2_SMI_OP_22_WRITE | (addr << 5) | - regnum); - - return mv88e6xxx_mdio_wait(ps); -} - -static int mv88e6xxx_get_eee(struct dsa_switch *ds, int port, - struct ethtool_eee *e) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int reg; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEE)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - reg = mv88e6xxx_mdio_read_indirect(ps, port, 16); - if (reg < 0) - goto out; - - e->eee_enabled = !!(reg & 0x0200); - e->tx_lpi_enabled = !!(reg & 0x0100); - - reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_STATUS); - if (reg < 0) - goto out; - - e->eee_active = !!(reg & PORT_STATUS_EEE); - reg = 0; - -out: - mutex_unlock(&ps->reg_lock); - return reg; -} - -static int mv88e6xxx_set_eee(struct dsa_switch *ds, int port, - struct phy_device *phydev, struct ethtool_eee *e) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int reg; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEE)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - ret = mv88e6xxx_mdio_read_indirect(ps, port, 16); - if (ret < 0) - goto out; - - reg = ret & ~0x0300; - if (e->eee_enabled) - reg |= 0x0200; - if (e->tx_lpi_enabled) - reg |= 0x0100; - - ret = mv88e6xxx_mdio_write_indirect(ps, port, 16, reg); -out: - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int _mv88e6xxx_atu_cmd(struct mv88e6xxx_priv_state *ps, u16 fid, u16 cmd) -{ - int ret; - - if (mv88e6xxx_has_fid_reg(ps)) { - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_FID, fid); - if (ret < 0) - return ret; - } else if (mv88e6xxx_num_databases(ps) == 256) { - /* ATU DBNum[7:4] are located in ATU Control 15:12 */ - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_ATU_CONTROL); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_CONTROL, - (ret & 0xfff) | - ((fid << 8) & 0xf000)); - if (ret < 0) - return ret; - - /* ATU DBNum[3:0] are located in ATU Operation 3:0 */ - cmd |= fid & 0xf; - } - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_OP, cmd); - if (ret < 0) - return ret; - - return _mv88e6xxx_atu_wait(ps); -} - -static int _mv88e6xxx_atu_data_write(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_atu_entry *entry) -{ - u16 data = entry->state & GLOBAL_ATU_DATA_STATE_MASK; - - if (entry->state != GLOBAL_ATU_DATA_STATE_UNUSED) { - unsigned int mask, shift; - - if (entry->trunk) { - data |= GLOBAL_ATU_DATA_TRUNK; - mask = GLOBAL_ATU_DATA_TRUNK_ID_MASK; - shift = GLOBAL_ATU_DATA_TRUNK_ID_SHIFT; - } else { - mask = GLOBAL_ATU_DATA_PORT_VECTOR_MASK; - shift = GLOBAL_ATU_DATA_PORT_VECTOR_SHIFT; - } - - data |= (entry->portv_trunkid << shift) & mask; - } - - return _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_DATA, data); -} - -static int _mv88e6xxx_atu_flush_move(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_atu_entry *entry, - bool static_too) -{ - int op; - int err; - - err = _mv88e6xxx_atu_wait(ps); - if (err) - return err; - - err = _mv88e6xxx_atu_data_write(ps, entry); - if (err) - return err; - - if (entry->fid) { - op = static_too ? GLOBAL_ATU_OP_FLUSH_MOVE_ALL_DB : - GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC_DB; - } else { - op = static_too ? GLOBAL_ATU_OP_FLUSH_MOVE_ALL : - GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC; - } - - return _mv88e6xxx_atu_cmd(ps, entry->fid, op); -} - -static int _mv88e6xxx_atu_flush(struct mv88e6xxx_priv_state *ps, - u16 fid, bool static_too) -{ - struct mv88e6xxx_atu_entry entry = { - .fid = fid, - .state = 0, /* EntryState bits must be 0 */ - }; - - return _mv88e6xxx_atu_flush_move(ps, &entry, static_too); -} - -static int _mv88e6xxx_atu_move(struct mv88e6xxx_priv_state *ps, u16 fid, - int from_port, int to_port, bool static_too) -{ - struct mv88e6xxx_atu_entry entry = { - .trunk = false, - .fid = fid, - }; - - /* EntryState bits must be 0xF */ - entry.state = GLOBAL_ATU_DATA_STATE_MASK; - - /* ToPort and FromPort are respectively in PortVec bits 7:4 and 3:0 */ - entry.portv_trunkid = (to_port & 0x0f) << 4; - entry.portv_trunkid |= from_port & 0x0f; - - return _mv88e6xxx_atu_flush_move(ps, &entry, static_too); -} - -static int _mv88e6xxx_atu_remove(struct mv88e6xxx_priv_state *ps, u16 fid, - int port, bool static_too) -{ - /* Destination port 0xF means remove the entries */ - return _mv88e6xxx_atu_move(ps, fid, port, 0x0f, static_too); -} - -static const char * const mv88e6xxx_port_state_names[] = { - [PORT_CONTROL_STATE_DISABLED] = "Disabled", - [PORT_CONTROL_STATE_BLOCKING] = "Blocking/Listening", - [PORT_CONTROL_STATE_LEARNING] = "Learning", - [PORT_CONTROL_STATE_FORWARDING] = "Forwarding", -}; - -static int _mv88e6xxx_port_state(struct mv88e6xxx_priv_state *ps, int port, - u8 state) -{ - struct dsa_switch *ds = ps->ds; - int reg, ret = 0; - u8 oldstate; - - reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_CONTROL); - if (reg < 0) - return reg; - - oldstate = reg & PORT_CONTROL_STATE_MASK; - - if (oldstate != state) { - /* Flush forwarding database if we're moving a port - * from Learning or Forwarding state to Disabled or - * Blocking or Listening state. - */ - if ((oldstate == PORT_CONTROL_STATE_LEARNING || - oldstate == PORT_CONTROL_STATE_FORWARDING) && - (state == PORT_CONTROL_STATE_DISABLED || - state == PORT_CONTROL_STATE_BLOCKING)) { - ret = _mv88e6xxx_atu_remove(ps, 0, port, false); - if (ret) - return ret; - } - - reg = (reg & ~PORT_CONTROL_STATE_MASK) | state; - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL, - reg); - if (ret) - return ret; - - netdev_dbg(ds->ports[port].netdev, "PortState %s (was %s)\n", - mv88e6xxx_port_state_names[state], - mv88e6xxx_port_state_names[oldstate]); - } - - return ret; -} - -static int _mv88e6xxx_port_based_vlan_map(struct mv88e6xxx_priv_state *ps, - int port) -{ - struct net_device *bridge = ps->ports[port].bridge_dev; - const u16 mask = (1 << ps->info->num_ports) - 1; - struct dsa_switch *ds = ps->ds; - u16 output_ports = 0; - int reg; - int i; - - /* allow CPU port or DSA link(s) to send frames to every port */ - if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) { - output_ports = mask; - } else { - for (i = 0; i < ps->info->num_ports; ++i) { - /* allow sending frames to every group member */ - if (bridge && ps->ports[i].bridge_dev == bridge) - output_ports |= BIT(i); - - /* allow sending frames to CPU port and DSA link(s) */ - if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i)) - output_ports |= BIT(i); - } - } - - /* prevent frames from going back out of the port they came in on */ - output_ports &= ~BIT(port); - - reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_BASE_VLAN); - if (reg < 0) - return reg; - - reg &= ~mask; - reg |= output_ports & mask; - - return _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_BASE_VLAN, reg); -} - -static void mv88e6xxx_port_stp_state_set(struct dsa_switch *ds, int port, - u8 state) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int stp_state; - int err; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_PORTSTATE)) - return; - - switch (state) { - case BR_STATE_DISABLED: - stp_state = PORT_CONTROL_STATE_DISABLED; - break; - case BR_STATE_BLOCKING: - case BR_STATE_LISTENING: - stp_state = PORT_CONTROL_STATE_BLOCKING; - break; - case BR_STATE_LEARNING: - stp_state = PORT_CONTROL_STATE_LEARNING; - break; - case BR_STATE_FORWARDING: - default: - stp_state = PORT_CONTROL_STATE_FORWARDING; - break; - } - - mutex_lock(&ps->reg_lock); - err = _mv88e6xxx_port_state(ps, port, stp_state); - mutex_unlock(&ps->reg_lock); - - if (err) - netdev_err(ds->ports[port].netdev, - "failed to update state to %s\n", - mv88e6xxx_port_state_names[stp_state]); -} - -static int _mv88e6xxx_port_pvid(struct mv88e6xxx_priv_state *ps, int port, - u16 *new, u16 *old) -{ - struct dsa_switch *ds = ps->ds; - u16 pvid; - int ret; - - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_DEFAULT_VLAN); - if (ret < 0) - return ret; - - pvid = ret & PORT_DEFAULT_VLAN_MASK; - - if (new) { - ret &= ~PORT_DEFAULT_VLAN_MASK; - ret |= *new & PORT_DEFAULT_VLAN_MASK; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_DEFAULT_VLAN, ret); - if (ret < 0) - return ret; - - netdev_dbg(ds->ports[port].netdev, - "DefaultVID %d (was %d)\n", *new, pvid); - } - - if (old) - *old = pvid; - - return 0; -} - -static int _mv88e6xxx_port_pvid_get(struct mv88e6xxx_priv_state *ps, - int port, u16 *pvid) -{ - return _mv88e6xxx_port_pvid(ps, port, NULL, pvid); -} - -static int _mv88e6xxx_port_pvid_set(struct mv88e6xxx_priv_state *ps, - int port, u16 pvid) -{ - return _mv88e6xxx_port_pvid(ps, port, &pvid, NULL); -} - -static int _mv88e6xxx_vtu_wait(struct mv88e6xxx_priv_state *ps) -{ - return _mv88e6xxx_wait(ps, REG_GLOBAL, GLOBAL_VTU_OP, - GLOBAL_VTU_OP_BUSY); -} - -static int _mv88e6xxx_vtu_cmd(struct mv88e6xxx_priv_state *ps, u16 op) -{ - int ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_OP, op); - if (ret < 0) - return ret; - - return _mv88e6xxx_vtu_wait(ps); -} - -static int _mv88e6xxx_vtu_stu_flush(struct mv88e6xxx_priv_state *ps) -{ - int ret; - - ret = _mv88e6xxx_vtu_wait(ps); - if (ret < 0) - return ret; - - return _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_FLUSH_ALL); -} - -static int _mv88e6xxx_vtu_stu_data_read(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry, - unsigned int nibble_offset) -{ - u16 regs[3]; - int i; - int ret; - - for (i = 0; i < 3; ++i) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, - GLOBAL_VTU_DATA_0_3 + i); - if (ret < 0) - return ret; - - regs[i] = ret; - } - - for (i = 0; i < ps->info->num_ports; ++i) { - unsigned int shift = (i % 4) * 4 + nibble_offset; - u16 reg = regs[i / 4]; - - entry->data[i] = (reg >> shift) & GLOBAL_VTU_STU_DATA_MASK; - } - - return 0; -} - -static int mv88e6xxx_vtu_data_read(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - return _mv88e6xxx_vtu_stu_data_read(ps, entry, 0); -} - -static int mv88e6xxx_stu_data_read(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - return _mv88e6xxx_vtu_stu_data_read(ps, entry, 2); -} - -static int _mv88e6xxx_vtu_stu_data_write(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry, - unsigned int nibble_offset) -{ - u16 regs[3] = { 0 }; - int i; - int ret; - - for (i = 0; i < ps->info->num_ports; ++i) { - unsigned int shift = (i % 4) * 4 + nibble_offset; - u8 data = entry->data[i]; - - regs[i / 4] |= (data & GLOBAL_VTU_STU_DATA_MASK) << shift; - } - - for (i = 0; i < 3; ++i) { - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, - GLOBAL_VTU_DATA_0_3 + i, regs[i]); - if (ret < 0) - return ret; - } - - return 0; -} - -static int mv88e6xxx_vtu_data_write(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - return _mv88e6xxx_vtu_stu_data_write(ps, entry, 0); -} - -static int mv88e6xxx_stu_data_write(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - return _mv88e6xxx_vtu_stu_data_write(ps, entry, 2); -} - -static int _mv88e6xxx_vtu_vid_write(struct mv88e6xxx_priv_state *ps, u16 vid) -{ - return _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_VID, - vid & GLOBAL_VTU_VID_MASK); -} - -static int _mv88e6xxx_vtu_getnext(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - struct mv88e6xxx_vtu_stu_entry next = { 0 }; - int ret; - - ret = _mv88e6xxx_vtu_wait(ps); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_VTU_GET_NEXT); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_VTU_VID); - if (ret < 0) - return ret; - - next.vid = ret & GLOBAL_VTU_VID_MASK; - next.valid = !!(ret & GLOBAL_VTU_VID_VALID); - - if (next.valid) { - ret = mv88e6xxx_vtu_data_read(ps, &next); - if (ret < 0) - return ret; - - if (mv88e6xxx_has_fid_reg(ps)) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, - GLOBAL_VTU_FID); - if (ret < 0) - return ret; - - next.fid = ret & GLOBAL_VTU_FID_MASK; - } else if (mv88e6xxx_num_databases(ps) == 256) { - /* VTU DBNum[7:4] are located in VTU Operation 11:8, and - * VTU DBNum[3:0] are located in VTU Operation 3:0 - */ - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, - GLOBAL_VTU_OP); - if (ret < 0) - return ret; - - next.fid = (ret & 0xf00) >> 4; - next.fid |= ret & 0xf; - } - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_STU)) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, - GLOBAL_VTU_SID); - if (ret < 0) - return ret; - - next.sid = ret & GLOBAL_VTU_SID_MASK; - } - } - - *entry = next; - return 0; -} - -static int mv88e6xxx_port_vlan_dump(struct dsa_switch *ds, int port, - struct switchdev_obj_port_vlan *vlan, - int (*cb)(struct switchdev_obj *obj)) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_vtu_stu_entry next; - u16 pvid; - int err; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - err = _mv88e6xxx_port_pvid_get(ps, port, &pvid); - if (err) - goto unlock; - - err = _mv88e6xxx_vtu_vid_write(ps, GLOBAL_VTU_VID_MASK); - if (err) - goto unlock; - - do { - err = _mv88e6xxx_vtu_getnext(ps, &next); - if (err) - break; - - if (!next.valid) - break; - - if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) - continue; - - /* reinit and dump this VLAN obj */ - vlan->vid_begin = next.vid; - vlan->vid_end = next.vid; - vlan->flags = 0; - - if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED) - vlan->flags |= BRIDGE_VLAN_INFO_UNTAGGED; - - if (next.vid == pvid) - vlan->flags |= BRIDGE_VLAN_INFO_PVID; - - err = cb(&vlan->obj); - if (err) - break; - } while (next.vid < GLOBAL_VTU_VID_MASK); - -unlock: - mutex_unlock(&ps->reg_lock); - - return err; -} - -static int _mv88e6xxx_vtu_loadpurge(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - u16 op = GLOBAL_VTU_OP_VTU_LOAD_PURGE; - u16 reg = 0; - int ret; - - ret = _mv88e6xxx_vtu_wait(ps); - if (ret < 0) - return ret; - - if (!entry->valid) - goto loadpurge; - - /* Write port member tags */ - ret = mv88e6xxx_vtu_data_write(ps, entry); - if (ret < 0) - return ret; - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_STU)) { - reg = entry->sid & GLOBAL_VTU_SID_MASK; - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_SID, reg); - if (ret < 0) - return ret; - } - - if (mv88e6xxx_has_fid_reg(ps)) { - reg = entry->fid & GLOBAL_VTU_FID_MASK; - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_FID, reg); - if (ret < 0) - return ret; - } else if (mv88e6xxx_num_databases(ps) == 256) { - /* VTU DBNum[7:4] are located in VTU Operation 11:8, and - * VTU DBNum[3:0] are located in VTU Operation 3:0 - */ - op |= (entry->fid & 0xf0) << 8; - op |= entry->fid & 0xf; - } - - reg = GLOBAL_VTU_VID_VALID; -loadpurge: - reg |= entry->vid & GLOBAL_VTU_VID_MASK; - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_VID, reg); - if (ret < 0) - return ret; - - return _mv88e6xxx_vtu_cmd(ps, op); -} - -static int _mv88e6xxx_stu_getnext(struct mv88e6xxx_priv_state *ps, u8 sid, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - struct mv88e6xxx_vtu_stu_entry next = { 0 }; - int ret; - - ret = _mv88e6xxx_vtu_wait(ps); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_SID, - sid & GLOBAL_VTU_SID_MASK); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_STU_GET_NEXT); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_VTU_SID); - if (ret < 0) - return ret; - - next.sid = ret & GLOBAL_VTU_SID_MASK; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_VTU_VID); - if (ret < 0) - return ret; - - next.valid = !!(ret & GLOBAL_VTU_VID_VALID); - - if (next.valid) { - ret = mv88e6xxx_stu_data_read(ps, &next); - if (ret < 0) - return ret; - } - - *entry = next; - return 0; -} - -static int _mv88e6xxx_stu_loadpurge(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - u16 reg = 0; - int ret; - - ret = _mv88e6xxx_vtu_wait(ps); - if (ret < 0) - return ret; - - if (!entry->valid) - goto loadpurge; - - /* Write port states */ - ret = mv88e6xxx_stu_data_write(ps, entry); - if (ret < 0) - return ret; - - reg = GLOBAL_VTU_VID_VALID; -loadpurge: - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_VID, reg); - if (ret < 0) - return ret; - - reg = entry->sid & GLOBAL_VTU_SID_MASK; - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_SID, reg); - if (ret < 0) - return ret; - - return _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_STU_LOAD_PURGE); -} - -static int _mv88e6xxx_port_fid(struct mv88e6xxx_priv_state *ps, int port, - u16 *new, u16 *old) -{ - struct dsa_switch *ds = ps->ds; - u16 upper_mask; - u16 fid; - int ret; - - if (mv88e6xxx_num_databases(ps) == 4096) - upper_mask = 0xff; - else if (mv88e6xxx_num_databases(ps) == 256) - upper_mask = 0xf; - else - return -EOPNOTSUPP; - - /* Port's default FID bits 3:0 are located in reg 0x06, offset 12 */ - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_BASE_VLAN); - if (ret < 0) - return ret; - - fid = (ret & PORT_BASE_VLAN_FID_3_0_MASK) >> 12; - - if (new) { - ret &= ~PORT_BASE_VLAN_FID_3_0_MASK; - ret |= (*new << 12) & PORT_BASE_VLAN_FID_3_0_MASK; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_BASE_VLAN, - ret); - if (ret < 0) - return ret; - } - - /* Port's default FID bits 11:4 are located in reg 0x05, offset 0 */ - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_CONTROL_1); - if (ret < 0) - return ret; - - fid |= (ret & upper_mask) << 4; - - if (new) { - ret &= ~upper_mask; - ret |= (*new >> 4) & upper_mask; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL_1, - ret); - if (ret < 0) - return ret; - - netdev_dbg(ds->ports[port].netdev, - "FID %d (was %d)\n", *new, fid); - } - - if (old) - *old = fid; - - return 0; -} - -static int _mv88e6xxx_port_fid_get(struct mv88e6xxx_priv_state *ps, - int port, u16 *fid) -{ - return _mv88e6xxx_port_fid(ps, port, NULL, fid); -} - -static int _mv88e6xxx_port_fid_set(struct mv88e6xxx_priv_state *ps, - int port, u16 fid) -{ - return _mv88e6xxx_port_fid(ps, port, &fid, NULL); -} - -static int _mv88e6xxx_fid_new(struct mv88e6xxx_priv_state *ps, u16 *fid) -{ - DECLARE_BITMAP(fid_bitmap, MV88E6XXX_N_FID); - struct mv88e6xxx_vtu_stu_entry vlan; - int i, err; - - bitmap_zero(fid_bitmap, MV88E6XXX_N_FID); - - /* Set every FID bit used by the (un)bridged ports */ - for (i = 0; i < ps->info->num_ports; ++i) { - err = _mv88e6xxx_port_fid_get(ps, i, fid); - if (err) - return err; - - set_bit(*fid, fid_bitmap); - } - - /* Set every FID bit used by the VLAN entries */ - err = _mv88e6xxx_vtu_vid_write(ps, GLOBAL_VTU_VID_MASK); - if (err) - return err; - - do { - err = _mv88e6xxx_vtu_getnext(ps, &vlan); - if (err) - return err; - - if (!vlan.valid) - break; - - set_bit(vlan.fid, fid_bitmap); - } while (vlan.vid < GLOBAL_VTU_VID_MASK); - - /* The reset value 0x000 is used to indicate that multiple address - * databases are not needed. Return the next positive available. - */ - *fid = find_next_zero_bit(fid_bitmap, MV88E6XXX_N_FID, 1); - if (unlikely(*fid >= mv88e6xxx_num_databases(ps))) - return -ENOSPC; - - /* Clear the database */ - return _mv88e6xxx_atu_flush(ps, *fid, true); -} - -static int _mv88e6xxx_vtu_new(struct mv88e6xxx_priv_state *ps, u16 vid, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - struct dsa_switch *ds = ps->ds; - struct mv88e6xxx_vtu_stu_entry vlan = { - .valid = true, - .vid = vid, - }; - int i, err; - - err = _mv88e6xxx_fid_new(ps, &vlan.fid); - if (err) - return err; - - /* exclude all ports except the CPU and DSA ports */ - for (i = 0; i < ps->info->num_ports; ++i) - vlan.data[i] = dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i) - ? GLOBAL_VTU_DATA_MEMBER_TAG_UNMODIFIED - : GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER; - - if (mv88e6xxx_6097_family(ps) || mv88e6xxx_6165_family(ps) || - mv88e6xxx_6351_family(ps) || mv88e6xxx_6352_family(ps)) { - struct mv88e6xxx_vtu_stu_entry vstp; - - /* Adding a VTU entry requires a valid STU entry. As VSTP is not - * implemented, only one STU entry is needed to cover all VTU - * entries. Thus, validate the SID 0. - */ - vlan.sid = 0; - err = _mv88e6xxx_stu_getnext(ps, GLOBAL_VTU_SID_MASK, &vstp); - if (err) - return err; - - if (vstp.sid != vlan.sid || !vstp.valid) { - memset(&vstp, 0, sizeof(vstp)); - vstp.valid = true; - vstp.sid = vlan.sid; - - err = _mv88e6xxx_stu_loadpurge(ps, &vstp); - if (err) - return err; - } - } - - *entry = vlan; - return 0; -} - -static int _mv88e6xxx_vtu_get(struct mv88e6xxx_priv_state *ps, u16 vid, - struct mv88e6xxx_vtu_stu_entry *entry, bool creat) -{ - int err; - - if (!vid) - return -EINVAL; - - err = _mv88e6xxx_vtu_vid_write(ps, vid - 1); - if (err) - return err; - - err = _mv88e6xxx_vtu_getnext(ps, entry); - if (err) - return err; - - if (entry->vid != vid || !entry->valid) { - if (!creat) - return -EOPNOTSUPP; - /* -ENOENT would've been more appropriate, but switchdev expects - * -EOPNOTSUPP to inform bridge about an eventual software VLAN. - */ - - err = _mv88e6xxx_vtu_new(ps, vid, entry); - } - - return err; -} - -static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port, - u16 vid_begin, u16 vid_end) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_vtu_stu_entry vlan; - int i, err; - - if (!vid_begin) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - err = _mv88e6xxx_vtu_vid_write(ps, vid_begin - 1); - if (err) - goto unlock; - - do { - err = _mv88e6xxx_vtu_getnext(ps, &vlan); - if (err) - goto unlock; - - if (!vlan.valid) - break; - - if (vlan.vid > vid_end) - break; - - for (i = 0; i < ps->info->num_ports; ++i) { - if (dsa_is_dsa_port(ds, i) || dsa_is_cpu_port(ds, i)) - continue; - - if (vlan.data[i] == - GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) - continue; - - if (ps->ports[i].bridge_dev == - ps->ports[port].bridge_dev) - break; /* same bridge, check next VLAN */ - - netdev_warn(ds->ports[port].netdev, - "hardware VLAN %d already used by %s\n", - vlan.vid, - netdev_name(ps->ports[i].bridge_dev)); - err = -EOPNOTSUPP; - goto unlock; - } - } while (vlan.vid < vid_end); - -unlock: - mutex_unlock(&ps->reg_lock); - - return err; -} - -static const char * const mv88e6xxx_port_8021q_mode_names[] = { - [PORT_CONTROL_2_8021Q_DISABLED] = "Disabled", - [PORT_CONTROL_2_8021Q_FALLBACK] = "Fallback", - [PORT_CONTROL_2_8021Q_CHECK] = "Check", - [PORT_CONTROL_2_8021Q_SECURE] = "Secure", -}; - -static int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port, - bool vlan_filtering) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - u16 old, new = vlan_filtering ? PORT_CONTROL_2_8021Q_SECURE : - PORT_CONTROL_2_8021Q_DISABLED; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_CONTROL_2); - if (ret < 0) - goto unlock; - - old = ret & PORT_CONTROL_2_8021Q_MASK; - - if (new != old) { - ret &= ~PORT_CONTROL_2_8021Q_MASK; - ret |= new & PORT_CONTROL_2_8021Q_MASK; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL_2, - ret); - if (ret < 0) - goto unlock; - - netdev_dbg(ds->ports[port].netdev, "802.1Q Mode %s (was %s)\n", - mv88e6xxx_port_8021q_mode_names[new], - mv88e6xxx_port_8021q_mode_names[old]); - } - - ret = 0; -unlock: - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int -mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_vlan *vlan, - struct switchdev_trans *trans) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int err; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) - return -EOPNOTSUPP; - - /* If the requested port doesn't belong to the same bridge as the VLAN - * members, do not support it (yet) and fallback to software VLAN. - */ - err = mv88e6xxx_port_check_hw_vlan(ds, port, vlan->vid_begin, - vlan->vid_end); - if (err) - return err; - - /* We don't need any dynamic resource from the kernel (yet), - * so skip the prepare phase. - */ - return 0; -} - -static int _mv88e6xxx_port_vlan_add(struct mv88e6xxx_priv_state *ps, int port, - u16 vid, bool untagged) -{ - struct mv88e6xxx_vtu_stu_entry vlan; - int err; - - err = _mv88e6xxx_vtu_get(ps, vid, &vlan, true); - if (err) - return err; - - vlan.data[port] = untagged ? - GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED : - GLOBAL_VTU_DATA_MEMBER_TAG_TAGGED; - - return _mv88e6xxx_vtu_loadpurge(ps, &vlan); -} - -static void mv88e6xxx_port_vlan_add(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_vlan *vlan, - struct switchdev_trans *trans) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; - bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; - u16 vid; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) - return; - - mutex_lock(&ps->reg_lock); - - for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) - if (_mv88e6xxx_port_vlan_add(ps, port, vid, untagged)) - netdev_err(ds->ports[port].netdev, - "failed to add VLAN %d%c\n", - vid, untagged ? 'u' : 't'); - - if (pvid && _mv88e6xxx_port_pvid_set(ps, port, vlan->vid_end)) - netdev_err(ds->ports[port].netdev, "failed to set PVID %d\n", - vlan->vid_end); - - mutex_unlock(&ps->reg_lock); -} - -static int _mv88e6xxx_port_vlan_del(struct mv88e6xxx_priv_state *ps, - int port, u16 vid) -{ - struct dsa_switch *ds = ps->ds; - struct mv88e6xxx_vtu_stu_entry vlan; - int i, err; - - err = _mv88e6xxx_vtu_get(ps, vid, &vlan, false); - if (err) - return err; - - /* Tell switchdev if this VLAN is handled in software */ - if (vlan.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) - return -EOPNOTSUPP; - - vlan.data[port] = GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER; - - /* keep the VLAN unless all ports are excluded */ - vlan.valid = false; - for (i = 0; i < ps->info->num_ports; ++i) { - if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i)) - continue; - - if (vlan.data[i] != GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) { - vlan.valid = true; - break; - } - } - - err = _mv88e6xxx_vtu_loadpurge(ps, &vlan); - if (err) - return err; - - return _mv88e6xxx_atu_remove(ps, vlan.fid, port, false); -} - -static int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_vlan *vlan) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - u16 pvid, vid; - int err = 0; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - err = _mv88e6xxx_port_pvid_get(ps, port, &pvid); - if (err) - goto unlock; - - for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) { - err = _mv88e6xxx_port_vlan_del(ps, port, vid); - if (err) - goto unlock; - - if (vid == pvid) { - err = _mv88e6xxx_port_pvid_set(ps, port, 0); - if (err) - goto unlock; - } - } - -unlock: - mutex_unlock(&ps->reg_lock); - - return err; -} - -static int _mv88e6xxx_atu_mac_write(struct mv88e6xxx_priv_state *ps, - const unsigned char *addr) -{ - int i, ret; - - for (i = 0; i < 3; i++) { - ret = _mv88e6xxx_reg_write( - ps, REG_GLOBAL, GLOBAL_ATU_MAC_01 + i, - (addr[i * 2] << 8) | addr[i * 2 + 1]); - if (ret < 0) - return ret; - } - - return 0; -} - -static int _mv88e6xxx_atu_mac_read(struct mv88e6xxx_priv_state *ps, - unsigned char *addr) -{ - int i, ret; - - for (i = 0; i < 3; i++) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, - GLOBAL_ATU_MAC_01 + i); - if (ret < 0) - return ret; - addr[i * 2] = ret >> 8; - addr[i * 2 + 1] = ret & 0xff; - } - - return 0; -} - -static int _mv88e6xxx_atu_load(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_atu_entry *entry) -{ - int ret; - - ret = _mv88e6xxx_atu_wait(ps); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_atu_mac_write(ps, entry->mac); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_atu_data_write(ps, entry); - if (ret < 0) - return ret; - - return _mv88e6xxx_atu_cmd(ps, entry->fid, GLOBAL_ATU_OP_LOAD_DB); -} - -static int _mv88e6xxx_port_fdb_load(struct mv88e6xxx_priv_state *ps, int port, - const unsigned char *addr, u16 vid, - u8 state) -{ - struct mv88e6xxx_atu_entry entry = { 0 }; - struct mv88e6xxx_vtu_stu_entry vlan; - int err; - - /* Null VLAN ID corresponds to the port private database */ - if (vid == 0) - err = _mv88e6xxx_port_fid_get(ps, port, &vlan.fid); - else - err = _mv88e6xxx_vtu_get(ps, vid, &vlan, false); - if (err) - return err; - - entry.fid = vlan.fid; - entry.state = state; - ether_addr_copy(entry.mac, addr); - if (state != GLOBAL_ATU_DATA_STATE_UNUSED) { - entry.trunk = false; - entry.portv_trunkid = BIT(port); - } - - return _mv88e6xxx_atu_load(ps, &entry); -} - -static int mv88e6xxx_port_fdb_prepare(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_fdb *fdb, - struct switchdev_trans *trans) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) - return -EOPNOTSUPP; - - /* We don't need any dynamic resource from the kernel (yet), - * so skip the prepare phase. - */ - return 0; -} - -static void mv88e6xxx_port_fdb_add(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_fdb *fdb, - struct switchdev_trans *trans) -{ - int state = is_multicast_ether_addr(fdb->addr) ? - GLOBAL_ATU_DATA_STATE_MC_STATIC : - GLOBAL_ATU_DATA_STATE_UC_STATIC; - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) - return; - - mutex_lock(&ps->reg_lock); - if (_mv88e6xxx_port_fdb_load(ps, port, fdb->addr, fdb->vid, state)) - netdev_err(ds->ports[port].netdev, - "failed to load MAC address\n"); - mutex_unlock(&ps->reg_lock); -} - -static int mv88e6xxx_port_fdb_del(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_fdb *fdb) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_port_fdb_load(ps, port, fdb->addr, fdb->vid, - GLOBAL_ATU_DATA_STATE_UNUSED); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int _mv88e6xxx_atu_getnext(struct mv88e6xxx_priv_state *ps, u16 fid, - struct mv88e6xxx_atu_entry *entry) -{ - struct mv88e6xxx_atu_entry next = { 0 }; - int ret; - - next.fid = fid; - - ret = _mv88e6xxx_atu_wait(ps); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_atu_cmd(ps, fid, GLOBAL_ATU_OP_GET_NEXT_DB); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_atu_mac_read(ps, next.mac); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_ATU_DATA); - if (ret < 0) - return ret; - - next.state = ret & GLOBAL_ATU_DATA_STATE_MASK; - if (next.state != GLOBAL_ATU_DATA_STATE_UNUSED) { - unsigned int mask, shift; - - if (ret & GLOBAL_ATU_DATA_TRUNK) { - next.trunk = true; - mask = GLOBAL_ATU_DATA_TRUNK_ID_MASK; - shift = GLOBAL_ATU_DATA_TRUNK_ID_SHIFT; - } else { - next.trunk = false; - mask = GLOBAL_ATU_DATA_PORT_VECTOR_MASK; - shift = GLOBAL_ATU_DATA_PORT_VECTOR_SHIFT; - } - - next.portv_trunkid = (ret & mask) >> shift; - } - - *entry = next; - return 0; -} - -static int _mv88e6xxx_port_fdb_dump_one(struct mv88e6xxx_priv_state *ps, - u16 fid, u16 vid, int port, - struct switchdev_obj_port_fdb *fdb, - int (*cb)(struct switchdev_obj *obj)) -{ - struct mv88e6xxx_atu_entry addr = { - .mac = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }, - }; - int err; - - err = _mv88e6xxx_atu_mac_write(ps, addr.mac); - if (err) - return err; - - do { - err = _mv88e6xxx_atu_getnext(ps, fid, &addr); - if (err) - break; - - if (addr.state == GLOBAL_ATU_DATA_STATE_UNUSED) - break; - - if (!addr.trunk && addr.portv_trunkid & BIT(port)) { - bool is_static = addr.state == - (is_multicast_ether_addr(addr.mac) ? - GLOBAL_ATU_DATA_STATE_MC_STATIC : - GLOBAL_ATU_DATA_STATE_UC_STATIC); - - fdb->vid = vid; - ether_addr_copy(fdb->addr, addr.mac); - fdb->ndm_state = is_static ? NUD_NOARP : NUD_REACHABLE; - - err = cb(&fdb->obj); - if (err) - break; - } - } while (!is_broadcast_ether_addr(addr.mac)); - - return err; -} - -static int mv88e6xxx_port_fdb_dump(struct dsa_switch *ds, int port, - struct switchdev_obj_port_fdb *fdb, - int (*cb)(struct switchdev_obj *obj)) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_vtu_stu_entry vlan = { - .vid = GLOBAL_VTU_VID_MASK, /* all ones */ - }; - u16 fid; - int err; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - /* Dump port's default Filtering Information Database (VLAN ID 0) */ - err = _mv88e6xxx_port_fid_get(ps, port, &fid); - if (err) - goto unlock; - - err = _mv88e6xxx_port_fdb_dump_one(ps, fid, 0, port, fdb, cb); - if (err) - goto unlock; - - /* Dump VLANs' Filtering Information Databases */ - err = _mv88e6xxx_vtu_vid_write(ps, vlan.vid); - if (err) - goto unlock; - - do { - err = _mv88e6xxx_vtu_getnext(ps, &vlan); - if (err) - break; - - if (!vlan.valid) - break; - - err = _mv88e6xxx_port_fdb_dump_one(ps, vlan.fid, vlan.vid, port, - fdb, cb); - if (err) - break; - } while (vlan.vid < GLOBAL_VTU_VID_MASK); - -unlock: - mutex_unlock(&ps->reg_lock); - - return err; -} - -static int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port, - struct net_device *bridge) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int i, err = 0; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VLANTABLE)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - /* Assign the bridge and remap each port's VLANTable */ - ps->ports[port].bridge_dev = bridge; - - for (i = 0; i < ps->info->num_ports; ++i) { - if (ps->ports[i].bridge_dev == bridge) { - err = _mv88e6xxx_port_based_vlan_map(ps, i); - if (err) - break; - } - } - - mutex_unlock(&ps->reg_lock); - - return err; -} - -static void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct net_device *bridge = ps->ports[port].bridge_dev; - int i; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VLANTABLE)) - return; - - mutex_lock(&ps->reg_lock); - - /* Unassign the bridge and remap each port's VLANTable */ - ps->ports[port].bridge_dev = NULL; - - for (i = 0; i < ps->info->num_ports; ++i) - if (i == port || ps->ports[i].bridge_dev == bridge) - if (_mv88e6xxx_port_based_vlan_map(ps, i)) - netdev_warn(ds->ports[i].netdev, - "failed to remap\n"); - - mutex_unlock(&ps->reg_lock); -} - -static int _mv88e6xxx_mdio_page_write(struct mv88e6xxx_priv_state *ps, - int port, int page, int reg, int val) -{ - int ret; - - ret = mv88e6xxx_mdio_write_indirect(ps, port, 0x16, page); - if (ret < 0) - goto restore_page_0; - - ret = mv88e6xxx_mdio_write_indirect(ps, port, reg, val); -restore_page_0: - mv88e6xxx_mdio_write_indirect(ps, port, 0x16, 0x0); - - return ret; -} - -static int _mv88e6xxx_mdio_page_read(struct mv88e6xxx_priv_state *ps, - int port, int page, int reg) -{ - int ret; - - ret = mv88e6xxx_mdio_write_indirect(ps, port, 0x16, page); - if (ret < 0) - goto restore_page_0; - - ret = mv88e6xxx_mdio_read_indirect(ps, port, reg); -restore_page_0: - mv88e6xxx_mdio_write_indirect(ps, port, 0x16, 0x0); - - return ret; -} - -static int mv88e6xxx_switch_reset(struct mv88e6xxx_priv_state *ps) -{ - bool ppu_active = mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU_ACTIVE); - u16 is_reset = (ppu_active ? 0x8800 : 0xc800); - struct gpio_desc *gpiod = ps->reset; - unsigned long timeout; - int ret; - int i; - - /* Set all ports to the disabled state. */ - for (i = 0; i < ps->info->num_ports; i++) { - ret = _mv88e6xxx_reg_read(ps, REG_PORT(i), PORT_CONTROL); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(i), PORT_CONTROL, - ret & 0xfffc); - if (ret) - return ret; - } - - /* Wait for transmit queues to drain. */ - usleep_range(2000, 4000); - - /* If there is a gpio connected to the reset pin, toggle it */ - if (gpiod) { - gpiod_set_value_cansleep(gpiod, 1); - usleep_range(10000, 20000); - gpiod_set_value_cansleep(gpiod, 0); - usleep_range(10000, 20000); - } - - /* Reset the switch. Keep the PPU active if requested. The PPU - * needs to be active to support indirect phy register access - * through global registers 0x18 and 0x19. - */ - if (ppu_active) - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, 0x04, 0xc000); - else - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, 0x04, 0xc400); - if (ret) - return ret; - - /* Wait up to one second for reset to complete. */ - timeout = jiffies + 1 * HZ; - while (time_before(jiffies, timeout)) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, 0x00); - if (ret < 0) - return ret; - - if ((ret & is_reset) == is_reset) - break; - usleep_range(1000, 2000); - } - if (time_after(jiffies, timeout)) - ret = -ETIMEDOUT; - else - ret = 0; - - return ret; -} - -static int mv88e6xxx_power_on_serdes(struct mv88e6xxx_priv_state *ps) -{ - int ret; - - ret = _mv88e6xxx_mdio_page_read(ps, REG_FIBER_SERDES, - PAGE_FIBER_SERDES, MII_BMCR); - if (ret < 0) - return ret; - - if (ret & BMCR_PDOWN) { - ret &= ~BMCR_PDOWN; - ret = _mv88e6xxx_mdio_page_write(ps, REG_FIBER_SERDES, - PAGE_FIBER_SERDES, MII_BMCR, - ret); - } - - return ret; -} - -static int mv88e6xxx_setup_port(struct mv88e6xxx_priv_state *ps, int port) -{ - struct dsa_switch *ds = ps->ds; - int ret; - u16 reg; - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6185_family(ps) || mv88e6xxx_6095_family(ps) || - mv88e6xxx_6065_family(ps) || mv88e6xxx_6320_family(ps)) { - /* MAC Forcing register: don't force link, speed, - * duplex or flow control state to any particular - * values on physical ports, but force the CPU port - * and all DSA ports to their maximum bandwidth and - * full duplex. - */ - reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_PCS_CTRL); - if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) { - reg &= ~PORT_PCS_CTRL_UNFORCED; - reg |= PORT_PCS_CTRL_FORCE_LINK | - PORT_PCS_CTRL_LINK_UP | - PORT_PCS_CTRL_DUPLEX_FULL | - PORT_PCS_CTRL_FORCE_DUPLEX; - if (mv88e6xxx_6065_family(ps)) - reg |= PORT_PCS_CTRL_100; - else - reg |= PORT_PCS_CTRL_1000; - } else { - reg |= PORT_PCS_CTRL_UNFORCED; - } - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_PCS_CTRL, reg); - if (ret) - return ret; - } - - /* Port Control: disable Drop-on-Unlock, disable Drop-on-Lock, - * disable Header mode, enable IGMP/MLD snooping, disable VLAN - * tunneling, determine priority by looking at 802.1p and IP - * priority fields (IP prio has precedence), and set STP state - * to Forwarding. - * - * If this is the CPU link, use DSA or EDSA tagging depending - * on which tagging mode was configured. - * - * If this is a link to another switch, use DSA tagging mode. - * - * If this is the upstream port for this switch, enable - * forwarding of unknown unicasts and multicasts. - */ - reg = 0; - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6095_family(ps) || mv88e6xxx_6065_family(ps) || - mv88e6xxx_6185_family(ps) || mv88e6xxx_6320_family(ps)) - reg = PORT_CONTROL_IGMP_MLD_SNOOP | - PORT_CONTROL_USE_TAG | PORT_CONTROL_USE_IP | - PORT_CONTROL_STATE_FORWARDING; - if (dsa_is_cpu_port(ds, port)) { - if (mv88e6xxx_6095_family(ps) || mv88e6xxx_6185_family(ps)) - reg |= PORT_CONTROL_DSA_TAG; - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6320_family(ps)) { - reg |= PORT_CONTROL_FRAME_ETHER_TYPE_DSA | - PORT_CONTROL_FORWARD_UNKNOWN | - PORT_CONTROL_FORWARD_UNKNOWN_MC; - } - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6095_family(ps) || mv88e6xxx_6065_family(ps) || - mv88e6xxx_6185_family(ps) || mv88e6xxx_6320_family(ps)) { - reg |= PORT_CONTROL_EGRESS_ADD_TAG; - } - } - if (dsa_is_dsa_port(ds, port)) { - if (mv88e6xxx_6095_family(ps) || mv88e6xxx_6185_family(ps)) - reg |= PORT_CONTROL_DSA_TAG; - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6320_family(ps)) { - reg |= PORT_CONTROL_FRAME_MODE_DSA; - } - - if (port == dsa_upstream_port(ds)) - reg |= PORT_CONTROL_FORWARD_UNKNOWN | - PORT_CONTROL_FORWARD_UNKNOWN_MC; - } - if (reg) { - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_CONTROL, reg); - if (ret) - return ret; - } - - /* If this port is connected to a SerDes, make sure the SerDes is not - * powered down. - */ - if (mv88e6xxx_6352_family(ps)) { - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_STATUS); - if (ret < 0) - return ret; - ret &= PORT_STATUS_CMODE_MASK; - if ((ret == PORT_STATUS_CMODE_100BASE_X) || - (ret == PORT_STATUS_CMODE_1000BASE_X) || - (ret == PORT_STATUS_CMODE_SGMII)) { - ret = mv88e6xxx_power_on_serdes(ps); - if (ret < 0) - return ret; - } - } - - /* Port Control 2: don't force a good FCS, set the maximum frame size to - * 10240 bytes, disable 802.1q tags checking, don't discard tagged or - * untagged frames on this port, do a destination address lookup on all - * received packets as usual, disable ARP mirroring and don't send a - * copy of all transmitted/received frames on this port to the CPU. - */ - reg = 0; - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6095_family(ps) || mv88e6xxx_6320_family(ps) || - mv88e6xxx_6185_family(ps)) - reg = PORT_CONTROL_2_MAP_DA; - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6320_family(ps)) - reg |= PORT_CONTROL_2_JUMBO_10240; - - if (mv88e6xxx_6095_family(ps) || mv88e6xxx_6185_family(ps)) { - /* Set the upstream port this port should use */ - reg |= dsa_upstream_port(ds); - /* enable forwarding of unknown multicast addresses to - * the upstream port - */ - if (port == dsa_upstream_port(ds)) - reg |= PORT_CONTROL_2_FORWARD_UNKNOWN; - } - - reg |= PORT_CONTROL_2_8021Q_DISABLED; - - if (reg) { - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_CONTROL_2, reg); - if (ret) - return ret; - } - - /* Port Association Vector: when learning source addresses - * of packets, add the address to the address database using - * a port bitmap that has only the bit for this port set and - * the other bits clear. - */ - reg = 1 << port; - /* Disable learning for CPU port */ - if (dsa_is_cpu_port(ds, port)) - reg = 0; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_ASSOC_VECTOR, reg); - if (ret) - return ret; - - /* Egress rate control 2: disable egress rate control. */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_RATE_CONTROL_2, - 0x0000); - if (ret) - return ret; - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6320_family(ps)) { - /* Do not limit the period of time that this port can - * be paused for by the remote end or the period of - * time that this port can pause the remote end. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_PAUSE_CTRL, 0x0000); - if (ret) - return ret; - - /* Port ATU control: disable limiting the number of - * address database entries that this port is allowed - * to use. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_ATU_CONTROL, 0x0000); - /* Priority Override: disable DA, SA and VTU priority - * override. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_PRI_OVERRIDE, 0x0000); - if (ret) - return ret; - - /* Port Ethertype: use the Ethertype DSA Ethertype - * value. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_ETH_TYPE, ETH_P_EDSA); - if (ret) - return ret; - /* Tag Remap: use an identity 802.1p prio -> switch - * prio mapping. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_TAG_REGMAP_0123, 0x3210); - if (ret) - return ret; - - /* Tag Remap 2: use an identity 802.1p prio -> switch - * prio mapping. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_TAG_REGMAP_4567, 0x7654); - if (ret) - return ret; - } - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6185_family(ps) || mv88e6xxx_6095_family(ps) || - mv88e6xxx_6320_family(ps)) { - /* Rate Control: disable ingress rate limiting. */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_RATE_CONTROL, 0x0001); - if (ret) - return ret; - } - - /* Port Control 1: disable trunking, disable sending - * learning messages to this port. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL_1, 0x0000); - if (ret) - return ret; - - /* Port based VLAN map: give each port the same default address - * database, and allow bidirectional communication between the - * CPU and DSA port(s), and the other ports. - */ - ret = _mv88e6xxx_port_fid_set(ps, port, 0); - if (ret) - return ret; - - ret = _mv88e6xxx_port_based_vlan_map(ps, port); - if (ret) - return ret; - - /* Default VLAN ID and priority: don't set a default VLAN - * ID, and set the default packet priority to zero. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_DEFAULT_VLAN, - 0x0000); - if (ret) - return ret; - - return 0; -} - -static int mv88e6xxx_setup_global(struct mv88e6xxx_priv_state *ps) -{ - struct dsa_switch *ds = ps->ds; - u32 upstream_port = dsa_upstream_port(ds); - u16 reg; - int err; - int i; - - /* Enable the PHY Polling Unit if present, don't discard any packets, - * and mask all interrupt sources. - */ - reg = 0; - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU) || - mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU_ACTIVE)) - reg |= GLOBAL_CONTROL_PPU_ENABLE; - - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, reg); - if (err) - return err; - - /* Configure the upstream port, and configure it as the port to which - * ingress and egress and ARP monitor frames are to be sent. - */ - reg = upstream_port << GLOBAL_MONITOR_CONTROL_INGRESS_SHIFT | - upstream_port << GLOBAL_MONITOR_CONTROL_EGRESS_SHIFT | - upstream_port << GLOBAL_MONITOR_CONTROL_ARP_SHIFT; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MONITOR_CONTROL, reg); - if (err) - return err; - - /* Disable remote management, and set the switch's DSA device number. */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL_2, - GLOBAL_CONTROL_2_MULTIPLE_CASCADE | - (ds->index & 0x1f)); - if (err) - return err; - - /* Set the default address aging time to 5 minutes, and - * enable address learn messages to be sent to all message - * ports. - */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_CONTROL, - 0x0140 | GLOBAL_ATU_CONTROL_LEARN2ALL); - if (err) - return err; - - /* Configure the IP ToS mapping registers. */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_0, 0x0000); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_1, 0x0000); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_2, 0x5555); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_3, 0x5555); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_4, 0xaaaa); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_5, 0xaaaa); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_6, 0xffff); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_7, 0xffff); - if (err) - return err; - - /* Configure the IEEE 802.1p priority mapping register. */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IEEE_PRI, 0xfa41); - if (err) - return err; - - /* Send all frames with destination addresses matching - * 01:80:c2:00:00:0x to the CPU port. - */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_MGMT_EN_0X, 0xffff); - if (err) - return err; - - /* Ignore removed tag data on doubly tagged packets, disable - * flow control messages, force flow control priority to the - * highest, and send all special multicast frames to the CPU - * port at the highest priority. - */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SWITCH_MGMT, - 0x7 | GLOBAL2_SWITCH_MGMT_RSVD2CPU | 0x70 | - GLOBAL2_SWITCH_MGMT_FORCE_FLOW_CTRL_PRI); - if (err) - return err; - - /* Program the DSA routing table. */ - for (i = 0; i < 32; i++) { - int nexthop = 0x1f; - - if (i != ds->index && i < DSA_MAX_SWITCHES) - nexthop = ds->rtable[i] & 0x1f; - - err = _mv88e6xxx_reg_write( - ps, REG_GLOBAL2, - GLOBAL2_DEVICE_MAPPING, - GLOBAL2_DEVICE_MAPPING_UPDATE | - (i << GLOBAL2_DEVICE_MAPPING_TARGET_SHIFT) | nexthop); - if (err) - return err; - } - - /* Clear all trunk masks. */ - for (i = 0; i < 8; i++) { - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_TRUNK_MASK, - 0x8000 | - (i << GLOBAL2_TRUNK_MASK_NUM_SHIFT) | - ((1 << ps->info->num_ports) - 1)); - if (err) - return err; - } - - /* Clear all trunk mappings. */ - for (i = 0; i < 16; i++) { - err = _mv88e6xxx_reg_write( - ps, REG_GLOBAL2, - GLOBAL2_TRUNK_MAPPING, - GLOBAL2_TRUNK_MAPPING_UPDATE | - (i << GLOBAL2_TRUNK_MAPPING_ID_SHIFT)); - if (err) - return err; - } - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6320_family(ps)) { - /* Send all frames with destination addresses matching - * 01:80:c2:00:00:2x to the CPU port. - */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, - GLOBAL2_MGMT_EN_2X, 0xffff); - if (err) - return err; - - /* Initialise cross-chip port VLAN table to reset - * defaults. - */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, - GLOBAL2_PVT_ADDR, 0x9000); - if (err) - return err; - - /* Clear the priority override table. */ - for (i = 0; i < 16; i++) { - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, - GLOBAL2_PRIO_OVERRIDE, - 0x8000 | (i << 8)); - if (err) - return err; - } - } - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6185_family(ps) || mv88e6xxx_6095_family(ps) || - mv88e6xxx_6320_family(ps)) { - /* Disable ingress rate limiting by resetting all - * ingress rate limit registers to their initial - * state. - */ - for (i = 0; i < ps->info->num_ports; i++) { - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, - GLOBAL2_INGRESS_OP, - 0x9000 | (i << 8)); - if (err) - return err; - } - } - - /* Clear the statistics counters for all ports */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_STATS_OP, - GLOBAL_STATS_OP_FLUSH_ALL); - if (err) - return err; - - /* Wait for the flush to complete. */ - err = _mv88e6xxx_stats_wait(ps); - if (err) - return err; - - /* Clear all ATU entries */ - err = _mv88e6xxx_atu_flush(ps, 0, true); - if (err) - return err; - - /* Clear all the VTU and STU entries */ - err = _mv88e6xxx_vtu_stu_flush(ps); - if (err < 0) - return err; - - return err; -} - -static int mv88e6xxx_setup(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int err; - int i; - - ps->ds = ds; - ds->slave_mii_bus = ps->mdio_bus; - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) - mutex_init(&ps->eeprom_mutex); - - mutex_lock(&ps->reg_lock); - - err = mv88e6xxx_switch_reset(ps); - if (err) - goto unlock; - - err = mv88e6xxx_setup_global(ps); - if (err) - goto unlock; - - for (i = 0; i < ps->info->num_ports; i++) { - err = mv88e6xxx_setup_port(ps, i); - if (err) - goto unlock; - } - -unlock: - mutex_unlock(&ps->reg_lock); - - return err; -} - -static int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, - int reg) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_mdio_page_read(ps, port, page, reg); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int mv88e6xxx_mdio_page_write(struct dsa_switch *ds, int port, int page, - int reg, int val) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_mdio_page_write(ps, port, page, reg, val); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int mv88e6xxx_port_to_mdio_addr(struct mv88e6xxx_priv_state *ps, - int port) -{ - if (port >= 0 && port < ps->info->num_ports) - return port; - return -EINVAL; -} - -static int mv88e6xxx_mdio_read(struct mii_bus *bus, int port, int regnum) -{ - struct mv88e6xxx_priv_state *ps = bus->priv; - int addr = mv88e6xxx_port_to_mdio_addr(ps, port); - int ret; - - if (addr < 0) - return 0xffff; - - mutex_lock(&ps->reg_lock); - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) - ret = mv88e6xxx_mdio_read_ppu(ps, addr, regnum); - else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SMI_PHY)) - ret = mv88e6xxx_mdio_read_indirect(ps, addr, regnum); - else - ret = mv88e6xxx_mdio_read_direct(ps, addr, regnum); - - mutex_unlock(&ps->reg_lock); - return ret; -} - -static int mv88e6xxx_mdio_write(struct mii_bus *bus, int port, int regnum, - u16 val) -{ - struct mv88e6xxx_priv_state *ps = bus->priv; - int addr = mv88e6xxx_port_to_mdio_addr(ps, port); - int ret; - - if (addr < 0) - return 0xffff; - - mutex_lock(&ps->reg_lock); - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) - ret = mv88e6xxx_mdio_write_ppu(ps, addr, regnum, val); - else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SMI_PHY)) - ret = mv88e6xxx_mdio_write_indirect(ps, addr, regnum, val); - else - ret = mv88e6xxx_mdio_write_direct(ps, addr, regnum, val); - - mutex_unlock(&ps->reg_lock); - return ret; -} - -static int mv88e6xxx_mdio_register(struct mv88e6xxx_priv_state *ps, - struct device_node *np) -{ - static int index; - struct mii_bus *bus; - int err; - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) - mv88e6xxx_ppu_state_init(ps); - - if (np) - ps->mdio_np = of_get_child_by_name(np, "mdio"); - - bus = devm_mdiobus_alloc(ps->dev); - if (!bus) - return -ENOMEM; - - bus->priv = (void *)ps; - if (np) { - bus->name = np->full_name; - snprintf(bus->id, MII_BUS_ID_SIZE, "%s", np->full_name); - } else { - bus->name = "mv88e6xxx SMI"; - snprintf(bus->id, MII_BUS_ID_SIZE, "mv88e6xxx-%d", index++); - } - - bus->read = mv88e6xxx_mdio_read; - bus->write = mv88e6xxx_mdio_write; - bus->parent = ps->dev; - - if (ps->mdio_np) - err = of_mdiobus_register(bus, ps->mdio_np); - else - err = mdiobus_register(bus); - if (err) { - dev_err(ps->dev, "Cannot register MDIO bus (%d)\n", err); - goto out; - } - ps->mdio_bus = bus; - - return 0; - -out: - if (ps->mdio_np) - of_node_put(ps->mdio_np); - - return err; -} - -static void mv88e6xxx_mdio_unregister(struct mv88e6xxx_priv_state *ps) - -{ - struct mii_bus *bus = ps->mdio_bus; - - mdiobus_unregister(bus); - - if (ps->mdio_np) - of_node_put(ps->mdio_np); -} - -#ifdef CONFIG_NET_DSA_HWMON - -static int mv88e61xx_get_temp(struct dsa_switch *ds, int *temp) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - int val; - - *temp = 0; - - mutex_lock(&ps->reg_lock); - - ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x16, 0x6); - if (ret < 0) - goto error; - - /* Enable temperature sensor */ - ret = mv88e6xxx_mdio_read_direct(ps, 0x0, 0x1a); - if (ret < 0) - goto error; - - ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x1a, ret | (1 << 5)); - if (ret < 0) - goto error; - - /* Wait for temperature to stabilize */ - usleep_range(10000, 12000); - - val = mv88e6xxx_mdio_read_direct(ps, 0x0, 0x1a); - if (val < 0) { - ret = val; - goto error; - } - - /* Disable temperature sensor */ - ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x1a, ret & ~(1 << 5)); - if (ret < 0) - goto error; - - *temp = ((val & 0x1f) - 5) * 5; - -error: - mv88e6xxx_mdio_write_direct(ps, 0x0, 0x16, 0x0); - mutex_unlock(&ps->reg_lock); - return ret; -} - -static int mv88e63xx_get_temp(struct dsa_switch *ds, int *temp) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; - int ret; - - *temp = 0; - - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 27); - if (ret < 0) - return ret; - - *temp = (ret & 0xff) - 25; - - return 0; -} - -static int mv88e6xxx_get_temp(struct dsa_switch *ds, int *temp) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP)) - return -EOPNOTSUPP; - - if (mv88e6xxx_6320_family(ps) || mv88e6xxx_6352_family(ps)) - return mv88e63xx_get_temp(ds, temp); - - return mv88e61xx_get_temp(ds, temp); -} - -static int mv88e6xxx_get_temp_limit(struct dsa_switch *ds, int *temp) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP_LIMIT)) - return -EOPNOTSUPP; - - *temp = 0; - - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); - if (ret < 0) - return ret; - - *temp = (((ret >> 8) & 0x1f) * 5) - 25; - - return 0; -} - -static int mv88e6xxx_set_temp_limit(struct dsa_switch *ds, int temp) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP_LIMIT)) - return -EOPNOTSUPP; - - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); - if (ret < 0) - return ret; - temp = clamp_val(DIV_ROUND_CLOSEST(temp, 5) + 5, 0, 0x1f); - return mv88e6xxx_mdio_page_write(ds, phy, 6, 26, - (ret & 0xe0ff) | (temp << 8)); -} - -static int mv88e6xxx_get_temp_alarm(struct dsa_switch *ds, bool *alarm) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP_LIMIT)) - return -EOPNOTSUPP; - - *alarm = false; - - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); - if (ret < 0) - return ret; - - *alarm = !!(ret & 0x40); - - return 0; -} -#endif /* CONFIG_NET_DSA_HWMON */ - -static const struct mv88e6xxx_info mv88e6xxx_table[] = { - [MV88E6085] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6085, - .family = MV88E6XXX_FAMILY_6097, - .name = "Marvell 88E6085", - .num_databases = 4096, - .num_ports = 10, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6097, - }, - - [MV88E6095] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6095, - .family = MV88E6XXX_FAMILY_6095, - .name = "Marvell 88E6095/88E6095F", - .num_databases = 256, - .num_ports = 11, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6095, - }, - - [MV88E6123] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6123, - .family = MV88E6XXX_FAMILY_6165, - .name = "Marvell 88E6123", - .num_databases = 4096, - .num_ports = 3, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6165, - }, - - [MV88E6131] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6131, - .family = MV88E6XXX_FAMILY_6185, - .name = "Marvell 88E6131", - .num_databases = 256, - .num_ports = 8, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6185, - }, - - [MV88E6161] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6161, - .family = MV88E6XXX_FAMILY_6165, - .name = "Marvell 88E6161", - .num_databases = 4096, - .num_ports = 6, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6165, - }, - - [MV88E6165] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6165, - .family = MV88E6XXX_FAMILY_6165, - .name = "Marvell 88E6165", - .num_databases = 4096, - .num_ports = 6, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6165, - }, - - [MV88E6171] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6171, - .family = MV88E6XXX_FAMILY_6351, - .name = "Marvell 88E6171", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6351, - }, - - [MV88E6172] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6172, - .family = MV88E6XXX_FAMILY_6352, - .name = "Marvell 88E6172", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6352, - }, - - [MV88E6175] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6175, - .family = MV88E6XXX_FAMILY_6351, - .name = "Marvell 88E6175", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6351, - }, - - [MV88E6176] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6176, - .family = MV88E6XXX_FAMILY_6352, - .name = "Marvell 88E6176", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6352, - }, - - [MV88E6185] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6185, - .family = MV88E6XXX_FAMILY_6185, - .name = "Marvell 88E6185", - .num_databases = 256, - .num_ports = 10, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6185, - }, - - [MV88E6240] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6240, - .family = MV88E6XXX_FAMILY_6352, - .name = "Marvell 88E6240", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6352, - }, - - [MV88E6320] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6320, - .family = MV88E6XXX_FAMILY_6320, - .name = "Marvell 88E6320", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6320, - }, - - [MV88E6321] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6321, - .family = MV88E6XXX_FAMILY_6320, - .name = "Marvell 88E6321", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6320, - }, - - [MV88E6350] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6350, - .family = MV88E6XXX_FAMILY_6351, - .name = "Marvell 88E6350", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6351, - }, - - [MV88E6351] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6351, - .family = MV88E6XXX_FAMILY_6351, - .name = "Marvell 88E6351", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6351, - }, - - [MV88E6352] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6352, - .family = MV88E6XXX_FAMILY_6352, - .name = "Marvell 88E6352", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6352, - }, -}; - -static const struct mv88e6xxx_info *mv88e6xxx_lookup_info(unsigned int prod_num) -{ - int i; - - for (i = 0; i < ARRAY_SIZE(mv88e6xxx_table); ++i) - if (mv88e6xxx_table[i].prod_num == prod_num) - return &mv88e6xxx_table[i]; - - return NULL; -} - -static int mv88e6xxx_detect(struct mv88e6xxx_priv_state *ps) -{ - const struct mv88e6xxx_info *info; - int id, prod_num, rev; - - id = mv88e6xxx_reg_read(ps, ps->info->port_base_addr, PORT_SWITCH_ID); - if (id < 0) - return id; - - prod_num = (id & 0xfff0) >> 4; - rev = id & 0x000f; - - info = mv88e6xxx_lookup_info(prod_num); - if (!info) - return -ENODEV; - - /* Update the compatible info with the probed one */ - ps->info = info; - - dev_info(ps->dev, "switch 0x%x detected: %s, revision %u\n", - ps->info->prod_num, ps->info->name, rev); - - return 0; -} - -static struct mv88e6xxx_priv_state *mv88e6xxx_alloc_chip(struct device *dev) -{ - struct mv88e6xxx_priv_state *ps; - - ps = devm_kzalloc(dev, sizeof(*ps), GFP_KERNEL); - if (!ps) - return NULL; - - ps->dev = dev; - - mutex_init(&ps->reg_lock); - - return ps; -} - -static int mv88e6xxx_smi_init(struct mv88e6xxx_priv_state *ps, - struct mii_bus *bus, int sw_addr) -{ - /* ADDR[0] pin is unavailable externally and considered zero */ - if (sw_addr & 0x1) - return -EINVAL; - - if (sw_addr == 0) - ps->smi_ops = &mv88e6xxx_smi_single_chip_ops; - else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_MULTI_CHIP)) - ps->smi_ops = &mv88e6xxx_smi_multi_chip_ops; - else - return -EINVAL; - - ps->bus = bus; - ps->sw_addr = sw_addr; - - return 0; -} - -static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, - struct device *host_dev, int sw_addr, - void **priv) -{ - struct mv88e6xxx_priv_state *ps; - struct mii_bus *bus; - int err; - - bus = dsa_host_dev_to_mii_bus(host_dev); - if (!bus) - return NULL; - - ps = mv88e6xxx_alloc_chip(dsa_dev); - if (!ps) - return NULL; - - /* Legacy SMI probing will only support chips similar to 88E6085 */ - ps->info = &mv88e6xxx_table[MV88E6085]; - - err = mv88e6xxx_smi_init(ps, bus, sw_addr); - if (err) - goto free; - - err = mv88e6xxx_detect(ps); - if (err) - goto free; - - err = mv88e6xxx_mdio_register(ps, NULL); - if (err) - goto free; - - *priv = ps; - - return ps->info->name; -free: - devm_kfree(dsa_dev, ps); - - return NULL; -} - -static struct dsa_switch_driver mv88e6xxx_switch_driver = { - .tag_protocol = DSA_TAG_PROTO_EDSA, - .probe = mv88e6xxx_drv_probe, - .setup = mv88e6xxx_setup, - .set_addr = mv88e6xxx_set_addr, - .adjust_link = mv88e6xxx_adjust_link, - .get_strings = mv88e6xxx_get_strings, - .get_ethtool_stats = mv88e6xxx_get_ethtool_stats, - .get_sset_count = mv88e6xxx_get_sset_count, - .set_eee = mv88e6xxx_set_eee, - .get_eee = mv88e6xxx_get_eee, -#ifdef CONFIG_NET_DSA_HWMON - .get_temp = mv88e6xxx_get_temp, - .get_temp_limit = mv88e6xxx_get_temp_limit, - .set_temp_limit = mv88e6xxx_set_temp_limit, - .get_temp_alarm = mv88e6xxx_get_temp_alarm, -#endif - .get_eeprom_len = mv88e6xxx_get_eeprom_len, - .get_eeprom = mv88e6xxx_get_eeprom, - .set_eeprom = mv88e6xxx_set_eeprom, - .get_regs_len = mv88e6xxx_get_regs_len, - .get_regs = mv88e6xxx_get_regs, - .port_bridge_join = mv88e6xxx_port_bridge_join, - .port_bridge_leave = mv88e6xxx_port_bridge_leave, - .port_stp_state_set = mv88e6xxx_port_stp_state_set, - .port_vlan_filtering = mv88e6xxx_port_vlan_filtering, - .port_vlan_prepare = mv88e6xxx_port_vlan_prepare, - .port_vlan_add = mv88e6xxx_port_vlan_add, - .port_vlan_del = mv88e6xxx_port_vlan_del, - .port_vlan_dump = mv88e6xxx_port_vlan_dump, - .port_fdb_prepare = mv88e6xxx_port_fdb_prepare, - .port_fdb_add = mv88e6xxx_port_fdb_add, - .port_fdb_del = mv88e6xxx_port_fdb_del, - .port_fdb_dump = mv88e6xxx_port_fdb_dump, -}; - -static int mv88e6xxx_register_switch(struct mv88e6xxx_priv_state *ps, - struct device_node *np) -{ - struct device *dev = ps->dev; - struct dsa_switch *ds; - - ds = devm_kzalloc(dev, sizeof(*ds), GFP_KERNEL); - if (!ds) - return -ENOMEM; - - ds->dev = dev; - ds->priv = ps; - ds->drv = &mv88e6xxx_switch_driver; - - dev_set_drvdata(dev, ds); - - return dsa_register_switch(ds, np); -} - -static void mv88e6xxx_unregister_switch(struct mv88e6xxx_priv_state *ps) -{ - dsa_unregister_switch(ps->ds); -} - -static int mv88e6xxx_probe(struct mdio_device *mdiodev) -{ - struct device *dev = &mdiodev->dev; - struct device_node *np = dev->of_node; - const struct mv88e6xxx_info *compat_info; - struct mv88e6xxx_priv_state *ps; - u32 eeprom_len; - int err; - - compat_info = of_device_get_match_data(dev); - if (!compat_info) - return -EINVAL; - - ps = mv88e6xxx_alloc_chip(dev); - if (!ps) - return -ENOMEM; - - ps->info = compat_info; - - err = mv88e6xxx_smi_init(ps, mdiodev->bus, mdiodev->addr); - if (err) - return err; - - err = mv88e6xxx_detect(ps); - if (err) - return err; - - ps->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_ASIS); - if (IS_ERR(ps->reset)) - return PTR_ERR(ps->reset); - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM) && - !of_property_read_u32(np, "eeprom-length", &eeprom_len)) - ps->eeprom_len = eeprom_len; - - err = mv88e6xxx_mdio_register(ps, np); - if (err) - return err; - - err = mv88e6xxx_register_switch(ps, np); - if (err) { - mv88e6xxx_mdio_unregister(ps); - return err; - } - - return 0; -} - -static void mv88e6xxx_remove(struct mdio_device *mdiodev) -{ - struct dsa_switch *ds = dev_get_drvdata(&mdiodev->dev); - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - mv88e6xxx_unregister_switch(ps); - mv88e6xxx_mdio_unregister(ps); -} - -static const struct of_device_id mv88e6xxx_of_match[] = { - { - .compatible = "marvell,mv88e6085", - .data = &mv88e6xxx_table[MV88E6085], - }, - { /* sentinel */ }, -}; - -MODULE_DEVICE_TABLE(of, mv88e6xxx_of_match); - -static struct mdio_driver mv88e6xxx_driver = { - .probe = mv88e6xxx_probe, - .remove = mv88e6xxx_remove, - .mdiodrv.driver = { - .name = "mv88e6085", - .of_match_table = mv88e6xxx_of_match, - }, -}; - -static int __init mv88e6xxx_init(void) -{ - register_switch_driver(&mv88e6xxx_switch_driver); - return mdio_driver_register(&mv88e6xxx_driver); -} -module_init(mv88e6xxx_init); - -static void __exit mv88e6xxx_cleanup(void) -{ - mdio_driver_unregister(&mv88e6xxx_driver); - unregister_switch_driver(&mv88e6xxx_switch_driver); -} -module_exit(mv88e6xxx_cleanup); - -MODULE_AUTHOR("Lennert Buytenhek "); -MODULE_DESCRIPTION("Driver for Marvell 88E6XXX ethernet switch chips"); -MODULE_LICENSE("GPL"); diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h deleted file mode 100644 index a94acd8..0000000 --- a/drivers/net/dsa/mv88e6xxx.h +++ /dev/null @@ -1,652 +0,0 @@ -/* - * net/dsa/mv88e6xxx.h - Marvell 88e6xxx switch chip support - * Copyright (c) 2008 Marvell Semiconductor - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - */ - -#ifndef __MV88E6XXX_H -#define __MV88E6XXX_H - -#include -#include - -#ifndef UINT64_MAX -#define UINT64_MAX (u64)(~((u64)0)) -#endif - -#define SMI_CMD 0x00 -#define SMI_CMD_BUSY BIT(15) -#define SMI_CMD_CLAUSE_22 BIT(12) -#define SMI_CMD_OP_22_WRITE ((1 << 10) | SMI_CMD_BUSY | SMI_CMD_CLAUSE_22) -#define SMI_CMD_OP_22_READ ((2 << 10) | SMI_CMD_BUSY | SMI_CMD_CLAUSE_22) -#define SMI_CMD_OP_45_WRITE_ADDR ((0 << 10) | SMI_CMD_BUSY) -#define SMI_CMD_OP_45_WRITE_DATA ((1 << 10) | SMI_CMD_BUSY) -#define SMI_CMD_OP_45_READ_DATA ((2 << 10) | SMI_CMD_BUSY) -#define SMI_CMD_OP_45_READ_DATA_INC ((3 << 10) | SMI_CMD_BUSY) -#define SMI_DATA 0x01 - -/* Fiber/SERDES Registers are located at SMI address F, page 1 */ -#define REG_FIBER_SERDES 0x0f -#define PAGE_FIBER_SERDES 0x01 - -#define REG_PORT(p) (0x10 + (p)) -#define PORT_STATUS 0x00 -#define PORT_STATUS_PAUSE_EN BIT(15) -#define PORT_STATUS_MY_PAUSE BIT(14) -#define PORT_STATUS_HD_FLOW BIT(13) -#define PORT_STATUS_PHY_DETECT BIT(12) -#define PORT_STATUS_LINK BIT(11) -#define PORT_STATUS_DUPLEX BIT(10) -#define PORT_STATUS_SPEED_MASK 0x0300 -#define PORT_STATUS_SPEED_10 0x0000 -#define PORT_STATUS_SPEED_100 0x0100 -#define PORT_STATUS_SPEED_1000 0x0200 -#define PORT_STATUS_EEE BIT(6) /* 6352 */ -#define PORT_STATUS_AM_DIS BIT(6) /* 6165 */ -#define PORT_STATUS_MGMII BIT(6) /* 6185 */ -#define PORT_STATUS_TX_PAUSED BIT(5) -#define PORT_STATUS_FLOW_CTRL BIT(4) -#define PORT_STATUS_CMODE_MASK 0x0f -#define PORT_STATUS_CMODE_100BASE_X 0x8 -#define PORT_STATUS_CMODE_1000BASE_X 0x9 -#define PORT_STATUS_CMODE_SGMII 0xa -#define PORT_PCS_CTRL 0x01 -#define PORT_PCS_CTRL_RGMII_DELAY_RXCLK BIT(15) -#define PORT_PCS_CTRL_RGMII_DELAY_TXCLK BIT(14) -#define PORT_PCS_CTRL_FC BIT(7) -#define PORT_PCS_CTRL_FORCE_FC BIT(6) -#define PORT_PCS_CTRL_LINK_UP BIT(5) -#define PORT_PCS_CTRL_FORCE_LINK BIT(4) -#define PORT_PCS_CTRL_DUPLEX_FULL BIT(3) -#define PORT_PCS_CTRL_FORCE_DUPLEX BIT(2) -#define PORT_PCS_CTRL_10 0x00 -#define PORT_PCS_CTRL_100 0x01 -#define PORT_PCS_CTRL_1000 0x02 -#define PORT_PCS_CTRL_UNFORCED 0x03 -#define PORT_PAUSE_CTRL 0x02 -#define PORT_SWITCH_ID 0x03 -#define PORT_SWITCH_ID_PROD_NUM_6085 0x04a -#define PORT_SWITCH_ID_PROD_NUM_6095 0x095 -#define PORT_SWITCH_ID_PROD_NUM_6131 0x106 -#define PORT_SWITCH_ID_PROD_NUM_6320 0x115 -#define PORT_SWITCH_ID_PROD_NUM_6123 0x121 -#define PORT_SWITCH_ID_PROD_NUM_6161 0x161 -#define PORT_SWITCH_ID_PROD_NUM_6165 0x165 -#define PORT_SWITCH_ID_PROD_NUM_6171 0x171 -#define PORT_SWITCH_ID_PROD_NUM_6172 0x172 -#define PORT_SWITCH_ID_PROD_NUM_6175 0x175 -#define PORT_SWITCH_ID_PROD_NUM_6176 0x176 -#define PORT_SWITCH_ID_PROD_NUM_6185 0x1a7 -#define PORT_SWITCH_ID_PROD_NUM_6240 0x240 -#define PORT_SWITCH_ID_PROD_NUM_6321 0x310 -#define PORT_SWITCH_ID_PROD_NUM_6352 0x352 -#define PORT_SWITCH_ID_PROD_NUM_6350 0x371 -#define PORT_SWITCH_ID_PROD_NUM_6351 0x375 -#define PORT_CONTROL 0x04 -#define PORT_CONTROL_USE_CORE_TAG BIT(15) -#define PORT_CONTROL_DROP_ON_LOCK BIT(14) -#define PORT_CONTROL_EGRESS_UNMODIFIED (0x0 << 12) -#define PORT_CONTROL_EGRESS_UNTAGGED (0x1 << 12) -#define PORT_CONTROL_EGRESS_TAGGED (0x2 << 12) -#define PORT_CONTROL_EGRESS_ADD_TAG (0x3 << 12) -#define PORT_CONTROL_HEADER BIT(11) -#define PORT_CONTROL_IGMP_MLD_SNOOP BIT(10) -#define PORT_CONTROL_DOUBLE_TAG BIT(9) -#define PORT_CONTROL_FRAME_MODE_NORMAL (0x0 << 8) -#define PORT_CONTROL_FRAME_MODE_DSA (0x1 << 8) -#define PORT_CONTROL_FRAME_MODE_PROVIDER (0x2 << 8) -#define PORT_CONTROL_FRAME_ETHER_TYPE_DSA (0x3 << 8) -#define PORT_CONTROL_DSA_TAG BIT(8) -#define PORT_CONTROL_VLAN_TUNNEL BIT(7) -#define PORT_CONTROL_TAG_IF_BOTH BIT(6) -#define PORT_CONTROL_USE_IP BIT(5) -#define PORT_CONTROL_USE_TAG BIT(4) -#define PORT_CONTROL_FORWARD_UNKNOWN_MC BIT(3) -#define PORT_CONTROL_FORWARD_UNKNOWN BIT(2) -#define PORT_CONTROL_STATE_MASK 0x03 -#define PORT_CONTROL_STATE_DISABLED 0x00 -#define PORT_CONTROL_STATE_BLOCKING 0x01 -#define PORT_CONTROL_STATE_LEARNING 0x02 -#define PORT_CONTROL_STATE_FORWARDING 0x03 -#define PORT_CONTROL_1 0x05 -#define PORT_CONTROL_1_FID_11_4_MASK (0xff << 0) -#define PORT_BASE_VLAN 0x06 -#define PORT_BASE_VLAN_FID_3_0_MASK (0xf << 12) -#define PORT_DEFAULT_VLAN 0x07 -#define PORT_DEFAULT_VLAN_MASK 0xfff -#define PORT_CONTROL_2 0x08 -#define PORT_CONTROL_2_IGNORE_FCS BIT(15) -#define PORT_CONTROL_2_VTU_PRI_OVERRIDE BIT(14) -#define PORT_CONTROL_2_SA_PRIO_OVERRIDE BIT(13) -#define PORT_CONTROL_2_DA_PRIO_OVERRIDE BIT(12) -#define PORT_CONTROL_2_JUMBO_1522 (0x00 << 12) -#define PORT_CONTROL_2_JUMBO_2048 (0x01 << 12) -#define PORT_CONTROL_2_JUMBO_10240 (0x02 << 12) -#define PORT_CONTROL_2_8021Q_MASK (0x03 << 10) -#define PORT_CONTROL_2_8021Q_DISABLED (0x00 << 10) -#define PORT_CONTROL_2_8021Q_FALLBACK (0x01 << 10) -#define PORT_CONTROL_2_8021Q_CHECK (0x02 << 10) -#define PORT_CONTROL_2_8021Q_SECURE (0x03 << 10) -#define PORT_CONTROL_2_DISCARD_TAGGED BIT(9) -#define PORT_CONTROL_2_DISCARD_UNTAGGED BIT(8) -#define PORT_CONTROL_2_MAP_DA BIT(7) -#define PORT_CONTROL_2_DEFAULT_FORWARD BIT(6) -#define PORT_CONTROL_2_FORWARD_UNKNOWN BIT(6) -#define PORT_CONTROL_2_EGRESS_MONITOR BIT(5) -#define PORT_CONTROL_2_INGRESS_MONITOR BIT(4) -#define PORT_RATE_CONTROL 0x09 -#define PORT_RATE_CONTROL_2 0x0a -#define PORT_ASSOC_VECTOR 0x0b -#define PORT_ASSOC_VECTOR_HOLD_AT_1 BIT(15) -#define PORT_ASSOC_VECTOR_INT_AGE_OUT BIT(14) -#define PORT_ASSOC_VECTOR_LOCKED_PORT BIT(13) -#define PORT_ASSOC_VECTOR_IGNORE_WRONG BIT(12) -#define PORT_ASSOC_VECTOR_REFRESH_LOCKED BIT(11) -#define PORT_ATU_CONTROL 0x0c -#define PORT_PRI_OVERRIDE 0x0d -#define PORT_ETH_TYPE 0x0f -#define PORT_IN_DISCARD_LO 0x10 -#define PORT_IN_DISCARD_HI 0x11 -#define PORT_IN_FILTERED 0x12 -#define PORT_OUT_FILTERED 0x13 -#define PORT_TAG_REGMAP_0123 0x18 -#define PORT_TAG_REGMAP_4567 0x19 - -#define REG_GLOBAL 0x1b -#define GLOBAL_STATUS 0x00 -#define GLOBAL_STATUS_PPU_STATE BIT(15) /* 6351 and 6171 */ -/* Two bits for 6165, 6185 etc */ -#define GLOBAL_STATUS_PPU_MASK (0x3 << 14) -#define GLOBAL_STATUS_PPU_DISABLED_RST (0x0 << 14) -#define GLOBAL_STATUS_PPU_INITIALIZING (0x1 << 14) -#define GLOBAL_STATUS_PPU_DISABLED (0x2 << 14) -#define GLOBAL_STATUS_PPU_POLLING (0x3 << 14) -#define GLOBAL_MAC_01 0x01 -#define GLOBAL_MAC_23 0x02 -#define GLOBAL_MAC_45 0x03 -#define GLOBAL_ATU_FID 0x01 /* 6097 6165 6351 6352 */ -#define GLOBAL_VTU_FID 0x02 /* 6097 6165 6351 6352 */ -#define GLOBAL_VTU_FID_MASK 0xfff -#define GLOBAL_VTU_SID 0x03 /* 6097 6165 6351 6352 */ -#define GLOBAL_VTU_SID_MASK 0x3f -#define GLOBAL_CONTROL 0x04 -#define GLOBAL_CONTROL_SW_RESET BIT(15) -#define GLOBAL_CONTROL_PPU_ENABLE BIT(14) -#define GLOBAL_CONTROL_DISCARD_EXCESS BIT(13) /* 6352 */ -#define GLOBAL_CONTROL_SCHED_PRIO BIT(11) /* 6152 */ -#define GLOBAL_CONTROL_MAX_FRAME_1632 BIT(10) /* 6152 */ -#define GLOBAL_CONTROL_RELOAD_EEPROM BIT(9) /* 6152 */ -#define GLOBAL_CONTROL_DEVICE_EN BIT(7) -#define GLOBAL_CONTROL_STATS_DONE_EN BIT(6) -#define GLOBAL_CONTROL_VTU_PROBLEM_EN BIT(5) -#define GLOBAL_CONTROL_VTU_DONE_EN BIT(4) -#define GLOBAL_CONTROL_ATU_PROBLEM_EN BIT(3) -#define GLOBAL_CONTROL_ATU_DONE_EN BIT(2) -#define GLOBAL_CONTROL_TCAM_EN BIT(1) -#define GLOBAL_CONTROL_EEPROM_DONE_EN BIT(0) -#define GLOBAL_VTU_OP 0x05 -#define GLOBAL_VTU_OP_BUSY BIT(15) -#define GLOBAL_VTU_OP_FLUSH_ALL ((0x01 << 12) | GLOBAL_VTU_OP_BUSY) -#define GLOBAL_VTU_OP_VTU_LOAD_PURGE ((0x03 << 12) | GLOBAL_VTU_OP_BUSY) -#define GLOBAL_VTU_OP_VTU_GET_NEXT ((0x04 << 12) | GLOBAL_VTU_OP_BUSY) -#define GLOBAL_VTU_OP_STU_LOAD_PURGE ((0x05 << 12) | GLOBAL_VTU_OP_BUSY) -#define GLOBAL_VTU_OP_STU_GET_NEXT ((0x06 << 12) | GLOBAL_VTU_OP_BUSY) -#define GLOBAL_VTU_VID 0x06 -#define GLOBAL_VTU_VID_MASK 0xfff -#define GLOBAL_VTU_VID_VALID BIT(12) -#define GLOBAL_VTU_DATA_0_3 0x07 -#define GLOBAL_VTU_DATA_4_7 0x08 -#define GLOBAL_VTU_DATA_8_11 0x09 -#define GLOBAL_VTU_STU_DATA_MASK 0x03 -#define GLOBAL_VTU_DATA_MEMBER_TAG_UNMODIFIED 0x00 -#define GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED 0x01 -#define GLOBAL_VTU_DATA_MEMBER_TAG_TAGGED 0x02 -#define GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER 0x03 -#define GLOBAL_STU_DATA_PORT_STATE_DISABLED 0x00 -#define GLOBAL_STU_DATA_PORT_STATE_BLOCKING 0x01 -#define GLOBAL_STU_DATA_PORT_STATE_LEARNING 0x02 -#define GLOBAL_STU_DATA_PORT_STATE_FORWARDING 0x03 -#define GLOBAL_ATU_CONTROL 0x0a -#define GLOBAL_ATU_CONTROL_LEARN2ALL BIT(3) -#define GLOBAL_ATU_OP 0x0b -#define GLOBAL_ATU_OP_BUSY BIT(15) -#define GLOBAL_ATU_OP_NOP (0 << 12) -#define GLOBAL_ATU_OP_FLUSH_MOVE_ALL ((1 << 12) | GLOBAL_ATU_OP_BUSY) -#define GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC ((2 << 12) | GLOBAL_ATU_OP_BUSY) -#define GLOBAL_ATU_OP_LOAD_DB ((3 << 12) | GLOBAL_ATU_OP_BUSY) -#define GLOBAL_ATU_OP_GET_NEXT_DB ((4 << 12) | GLOBAL_ATU_OP_BUSY) -#define GLOBAL_ATU_OP_FLUSH_MOVE_ALL_DB ((5 << 12) | GLOBAL_ATU_OP_BUSY) -#define GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC_DB ((6 << 12) | GLOBAL_ATU_OP_BUSY) -#define GLOBAL_ATU_OP_GET_CLR_VIOLATION ((7 << 12) | GLOBAL_ATU_OP_BUSY) -#define GLOBAL_ATU_DATA 0x0c -#define GLOBAL_ATU_DATA_TRUNK BIT(15) -#define GLOBAL_ATU_DATA_TRUNK_ID_MASK 0x00f0 -#define GLOBAL_ATU_DATA_TRUNK_ID_SHIFT 4 -#define GLOBAL_ATU_DATA_PORT_VECTOR_MASK 0x3ff0 -#define GLOBAL_ATU_DATA_PORT_VECTOR_SHIFT 4 -#define GLOBAL_ATU_DATA_STATE_MASK 0x0f -#define GLOBAL_ATU_DATA_STATE_UNUSED 0x00 -#define GLOBAL_ATU_DATA_STATE_UC_MGMT 0x0d -#define GLOBAL_ATU_DATA_STATE_UC_STATIC 0x0e -#define GLOBAL_ATU_DATA_STATE_UC_PRIO_OVER 0x0f -#define GLOBAL_ATU_DATA_STATE_MC_NONE_RATE 0x05 -#define GLOBAL_ATU_DATA_STATE_MC_STATIC 0x07 -#define GLOBAL_ATU_DATA_STATE_MC_MGMT 0x0e -#define GLOBAL_ATU_DATA_STATE_MC_PRIO_OVER 0x0f -#define GLOBAL_ATU_MAC_01 0x0d -#define GLOBAL_ATU_MAC_23 0x0e -#define GLOBAL_ATU_MAC_45 0x0f -#define GLOBAL_IP_PRI_0 0x10 -#define GLOBAL_IP_PRI_1 0x11 -#define GLOBAL_IP_PRI_2 0x12 -#define GLOBAL_IP_PRI_3 0x13 -#define GLOBAL_IP_PRI_4 0x14 -#define GLOBAL_IP_PRI_5 0x15 -#define GLOBAL_IP_PRI_6 0x16 -#define GLOBAL_IP_PRI_7 0x17 -#define GLOBAL_IEEE_PRI 0x18 -#define GLOBAL_CORE_TAG_TYPE 0x19 -#define GLOBAL_MONITOR_CONTROL 0x1a -#define GLOBAL_MONITOR_CONTROL_INGRESS_SHIFT 12 -#define GLOBAL_MONITOR_CONTROL_EGRESS_SHIFT 8 -#define GLOBAL_MONITOR_CONTROL_ARP_SHIFT 4 -#define GLOBAL_MONITOR_CONTROL_MIRROR_SHIFT 0 -#define GLOBAL_MONITOR_CONTROL_ARP_DISABLED (0xf0) -#define GLOBAL_CONTROL_2 0x1c -#define GLOBAL_CONTROL_2_NO_CASCADE 0xe000 -#define GLOBAL_CONTROL_2_MULTIPLE_CASCADE 0xf000 - -#define GLOBAL_STATS_OP 0x1d -#define GLOBAL_STATS_OP_BUSY BIT(15) -#define GLOBAL_STATS_OP_NOP (0 << 12) -#define GLOBAL_STATS_OP_FLUSH_ALL ((1 << 12) | GLOBAL_STATS_OP_BUSY) -#define GLOBAL_STATS_OP_FLUSH_PORT ((2 << 12) | GLOBAL_STATS_OP_BUSY) -#define GLOBAL_STATS_OP_READ_CAPTURED ((4 << 12) | GLOBAL_STATS_OP_BUSY) -#define GLOBAL_STATS_OP_CAPTURE_PORT ((5 << 12) | GLOBAL_STATS_OP_BUSY) -#define GLOBAL_STATS_OP_HIST_RX ((1 << 10) | GLOBAL_STATS_OP_BUSY) -#define GLOBAL_STATS_OP_HIST_TX ((2 << 10) | GLOBAL_STATS_OP_BUSY) -#define GLOBAL_STATS_OP_HIST_RX_TX ((3 << 10) | GLOBAL_STATS_OP_BUSY) -#define GLOBAL_STATS_OP_BANK_1 BIT(9) -#define GLOBAL_STATS_COUNTER_32 0x1e -#define GLOBAL_STATS_COUNTER_01 0x1f - -#define REG_GLOBAL2 0x1c -#define GLOBAL2_INT_SOURCE 0x00 -#define GLOBAL2_INT_MASK 0x01 -#define GLOBAL2_MGMT_EN_2X 0x02 -#define GLOBAL2_MGMT_EN_0X 0x03 -#define GLOBAL2_FLOW_CONTROL 0x04 -#define GLOBAL2_SWITCH_MGMT 0x05 -#define GLOBAL2_SWITCH_MGMT_USE_DOUBLE_TAG_DATA BIT(15) -#define GLOBAL2_SWITCH_MGMT_PREVENT_LOOPS BIT(14) -#define GLOBAL2_SWITCH_MGMT_FLOW_CONTROL_MSG BIT(13) -#define GLOBAL2_SWITCH_MGMT_FORCE_FLOW_CTRL_PRI BIT(7) -#define GLOBAL2_SWITCH_MGMT_RSVD2CPU BIT(3) -#define GLOBAL2_DEVICE_MAPPING 0x06 -#define GLOBAL2_DEVICE_MAPPING_UPDATE BIT(15) -#define GLOBAL2_DEVICE_MAPPING_TARGET_SHIFT 8 -#define GLOBAL2_DEVICE_MAPPING_PORT_MASK 0x0f -#define GLOBAL2_TRUNK_MASK 0x07 -#define GLOBAL2_TRUNK_MASK_UPDATE BIT(15) -#define GLOBAL2_TRUNK_MASK_NUM_SHIFT 12 -#define GLOBAL2_TRUNK_MAPPING 0x08 -#define GLOBAL2_TRUNK_MAPPING_UPDATE BIT(15) -#define GLOBAL2_TRUNK_MAPPING_ID_SHIFT 11 -#define GLOBAL2_INGRESS_OP 0x09 -#define GLOBAL2_INGRESS_DATA 0x0a -#define GLOBAL2_PVT_ADDR 0x0b -#define GLOBAL2_PVT_DATA 0x0c -#define GLOBAL2_SWITCH_MAC 0x0d -#define GLOBAL2_SWITCH_MAC_BUSY BIT(15) -#define GLOBAL2_ATU_STATS 0x0e -#define GLOBAL2_PRIO_OVERRIDE 0x0f -#define GLOBAL2_PRIO_OVERRIDE_FORCE_SNOOP BIT(7) -#define GLOBAL2_PRIO_OVERRIDE_SNOOP_SHIFT 4 -#define GLOBAL2_PRIO_OVERRIDE_FORCE_ARP BIT(3) -#define GLOBAL2_PRIO_OVERRIDE_ARP_SHIFT 0 -#define GLOBAL2_EEPROM_OP 0x14 -#define GLOBAL2_EEPROM_OP_BUSY BIT(15) -#define GLOBAL2_EEPROM_OP_WRITE ((3 << 12) | GLOBAL2_EEPROM_OP_BUSY) -#define GLOBAL2_EEPROM_OP_READ ((4 << 12) | GLOBAL2_EEPROM_OP_BUSY) -#define GLOBAL2_EEPROM_OP_LOAD BIT(11) -#define GLOBAL2_EEPROM_OP_WRITE_EN BIT(10) -#define GLOBAL2_EEPROM_OP_ADDR_MASK 0xff -#define GLOBAL2_EEPROM_DATA 0x15 -#define GLOBAL2_PTP_AVB_OP 0x16 -#define GLOBAL2_PTP_AVB_DATA 0x17 -#define GLOBAL2_SMI_OP 0x18 -#define GLOBAL2_SMI_OP_BUSY BIT(15) -#define GLOBAL2_SMI_OP_CLAUSE_22 BIT(12) -#define GLOBAL2_SMI_OP_22_WRITE ((1 << 10) | GLOBAL2_SMI_OP_BUSY | \ - GLOBAL2_SMI_OP_CLAUSE_22) -#define GLOBAL2_SMI_OP_22_READ ((2 << 10) | GLOBAL2_SMI_OP_BUSY | \ - GLOBAL2_SMI_OP_CLAUSE_22) -#define GLOBAL2_SMI_OP_45_WRITE_ADDR ((0 << 10) | GLOBAL2_SMI_OP_BUSY) -#define GLOBAL2_SMI_OP_45_WRITE_DATA ((1 << 10) | GLOBAL2_SMI_OP_BUSY) -#define GLOBAL2_SMI_OP_45_READ_DATA ((2 << 10) | GLOBAL2_SMI_OP_BUSY) -#define GLOBAL2_SMI_DATA 0x19 -#define GLOBAL2_SCRATCH_MISC 0x1a -#define GLOBAL2_SCRATCH_BUSY BIT(15) -#define GLOBAL2_SCRATCH_REGISTER_SHIFT 8 -#define GLOBAL2_SCRATCH_VALUE_MASK 0xff -#define GLOBAL2_WDOG_CONTROL 0x1b -#define GLOBAL2_QOS_WEIGHT 0x1c -#define GLOBAL2_MISC 0x1d - -#define MV88E6XXX_N_FID 4096 - -/* List of supported models */ -enum mv88e6xxx_model { - MV88E6085, - MV88E6095, - MV88E6123, - MV88E6131, - MV88E6161, - MV88E6165, - MV88E6171, - MV88E6172, - MV88E6175, - MV88E6176, - MV88E6185, - MV88E6240, - MV88E6320, - MV88E6321, - MV88E6350, - MV88E6351, - MV88E6352, -}; - -enum mv88e6xxx_family { - MV88E6XXX_FAMILY_NONE, - MV88E6XXX_FAMILY_6065, /* 6031 6035 6061 6065 */ - MV88E6XXX_FAMILY_6095, /* 6092 6095 */ - MV88E6XXX_FAMILY_6097, /* 6046 6085 6096 6097 */ - MV88E6XXX_FAMILY_6165, /* 6123 6161 6165 */ - MV88E6XXX_FAMILY_6185, /* 6108 6121 6122 6131 6152 6155 6182 6185 */ - MV88E6XXX_FAMILY_6320, /* 6320 6321 */ - MV88E6XXX_FAMILY_6351, /* 6171 6175 6350 6351 */ - MV88E6XXX_FAMILY_6352, /* 6172 6176 6240 6352 */ -}; - -enum mv88e6xxx_cap { - /* Address Translation Unit. - * The ATU is used to lookup and learn MAC addresses. See GLOBAL_ATU_OP. - */ - MV88E6XXX_CAP_ATU, - - /* Energy Efficient Ethernet. - */ - MV88E6XXX_CAP_EEE, - - /* EEPROM Command and Data registers. - * See GLOBAL2_EEPROM_OP and GLOBAL2_EEPROM_DATA. - */ - MV88E6XXX_CAP_EEPROM, - - /* Multi-chip Addressing Mode. - * Some chips require an indirect SMI access when their SMI device - * address is not zero. See SMI_CMD and SMI_DATA. - */ - MV88E6XXX_CAP_MULTI_CHIP, - - /* Port State Filtering for 802.1D Spanning Tree. - * See PORT_CONTROL_STATE_* values in the PORT_CONTROL register. - */ - MV88E6XXX_CAP_PORTSTATE, - - /* PHY Polling Unit. - * See GLOBAL_CONTROL_PPU_ENABLE and GLOBAL_STATUS_PPU_POLLING. - */ - MV88E6XXX_CAP_PPU, - MV88E6XXX_CAP_PPU_ACTIVE, - - /* SMI PHY Command and Data registers. - * This requires an indirect access to PHY registers through - * GLOBAL2_SMI_OP, otherwise direct access to PHY registers is done. - */ - MV88E6XXX_CAP_SMI_PHY, - - /* Per VLAN Spanning Tree Unit (STU). - * The Port State database, if present, is accessed through VTU - * operations and dedicated SID registers. See GLOBAL_VTU_SID. - */ - MV88E6XXX_CAP_STU, - - /* Switch MAC/WoL/WoF register. - * This requires an indirect access to set the switch MAC address - * through GLOBAL2_SWITCH_MAC, otherwise GLOBAL_MAC_01, GLOBAL_MAC_23, - * and GLOBAL_MAC_45 are used with a direct access. - */ - MV88E6XXX_CAP_SWITCH_MAC_WOL_WOF, - - /* Internal temperature sensor. - * Available from any enabled port's PHY register 26, page 6. - */ - MV88E6XXX_CAP_TEMP, - MV88E6XXX_CAP_TEMP_LIMIT, - - /* In-chip Port Based VLANs. - * Each port VLANTable register (see PORT_BASE_VLAN) is used to restrict - * the output (or egress) ports to which it is allowed to send frames. - */ - MV88E6XXX_CAP_VLANTABLE, - - /* VLAN Table Unit. - * The VTU is used to program 802.1Q VLANs. See GLOBAL_VTU_OP. - */ - MV88E6XXX_CAP_VTU, -}; - -/* Bitmask of capabilities */ -#define MV88E6XXX_FLAG_ATU BIT(MV88E6XXX_CAP_ATU) -#define MV88E6XXX_FLAG_EEE BIT(MV88E6XXX_CAP_EEE) -#define MV88E6XXX_FLAG_EEPROM BIT(MV88E6XXX_CAP_EEPROM) -#define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) -#define MV88E6XXX_FLAG_PORTSTATE BIT(MV88E6XXX_CAP_PORTSTATE) -#define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) -#define MV88E6XXX_FLAG_PPU_ACTIVE BIT(MV88E6XXX_CAP_PPU_ACTIVE) -#define MV88E6XXX_FLAG_SMI_PHY BIT(MV88E6XXX_CAP_SMI_PHY) -#define MV88E6XXX_FLAG_STU BIT(MV88E6XXX_CAP_STU) -#define MV88E6XXX_FLAG_SWITCH_MAC BIT(MV88E6XXX_CAP_SWITCH_MAC_WOL_WOF) -#define MV88E6XXX_FLAG_TEMP BIT(MV88E6XXX_CAP_TEMP) -#define MV88E6XXX_FLAG_TEMP_LIMIT BIT(MV88E6XXX_CAP_TEMP_LIMIT) -#define MV88E6XXX_FLAG_VLANTABLE BIT(MV88E6XXX_CAP_VLANTABLE) -#define MV88E6XXX_FLAG_VTU BIT(MV88E6XXX_CAP_VTU) - -#define MV88E6XXX_FLAGS_FAMILY_6095 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ - MV88E6XXX_FLAG_PPU | \ - MV88E6XXX_FLAG_VLANTABLE | \ - MV88E6XXX_FLAG_VTU) - -#define MV88E6XXX_FLAGS_FAMILY_6097 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ - MV88E6XXX_FLAG_PPU | \ - MV88E6XXX_FLAG_STU | \ - MV88E6XXX_FLAG_VLANTABLE | \ - MV88E6XXX_FLAG_VTU) - -#define MV88E6XXX_FLAGS_FAMILY_6165 \ - (MV88E6XXX_FLAG_MULTI_CHIP | \ - MV88E6XXX_FLAG_STU | \ - MV88E6XXX_FLAG_SWITCH_MAC | \ - MV88E6XXX_FLAG_TEMP | \ - MV88E6XXX_FLAG_VTU) - -#define MV88E6XXX_FLAGS_FAMILY_6185 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ - MV88E6XXX_FLAG_PPU | \ - MV88E6XXX_FLAG_VLANTABLE | \ - MV88E6XXX_FLAG_VTU) - -#define MV88E6XXX_FLAGS_FAMILY_6320 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_EEE | \ - MV88E6XXX_FLAG_EEPROM | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ - MV88E6XXX_FLAG_PORTSTATE | \ - MV88E6XXX_FLAG_PPU_ACTIVE | \ - MV88E6XXX_FLAG_SMI_PHY | \ - MV88E6XXX_FLAG_SWITCH_MAC | \ - MV88E6XXX_FLAG_TEMP | \ - MV88E6XXX_FLAG_TEMP_LIMIT | \ - MV88E6XXX_FLAG_VLANTABLE | \ - MV88E6XXX_FLAG_VTU) - -#define MV88E6XXX_FLAGS_FAMILY_6351 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ - MV88E6XXX_FLAG_PORTSTATE | \ - MV88E6XXX_FLAG_PPU_ACTIVE | \ - MV88E6XXX_FLAG_SMI_PHY | \ - MV88E6XXX_FLAG_STU | \ - MV88E6XXX_FLAG_SWITCH_MAC | \ - MV88E6XXX_FLAG_TEMP | \ - MV88E6XXX_FLAG_VLANTABLE | \ - MV88E6XXX_FLAG_VTU) - -#define MV88E6XXX_FLAGS_FAMILY_6352 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_EEE | \ - MV88E6XXX_FLAG_EEPROM | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ - MV88E6XXX_FLAG_PORTSTATE | \ - MV88E6XXX_FLAG_PPU_ACTIVE | \ - MV88E6XXX_FLAG_SMI_PHY | \ - MV88E6XXX_FLAG_STU | \ - MV88E6XXX_FLAG_SWITCH_MAC | \ - MV88E6XXX_FLAG_TEMP | \ - MV88E6XXX_FLAG_TEMP_LIMIT | \ - MV88E6XXX_FLAG_VLANTABLE | \ - MV88E6XXX_FLAG_VTU) - -struct mv88e6xxx_info { - enum mv88e6xxx_family family; - u16 prod_num; - const char *name; - unsigned int num_databases; - unsigned int num_ports; - unsigned int port_base_addr; - unsigned long flags; -}; - -struct mv88e6xxx_atu_entry { - u16 fid; - u8 state; - bool trunk; - u16 portv_trunkid; - u8 mac[ETH_ALEN]; -}; - -struct mv88e6xxx_vtu_stu_entry { - /* VTU only */ - u16 vid; - u16 fid; - - /* VTU and STU */ - u8 sid; - bool valid; - u8 data[DSA_MAX_PORTS]; -}; - -struct mv88e6xxx_ops; - -struct mv88e6xxx_priv_port { - struct net_device *bridge_dev; -}; - -struct mv88e6xxx_priv_state { - const struct mv88e6xxx_info *info; - - /* The dsa_switch this private structure is related to */ - struct dsa_switch *ds; - - /* The device this structure is associated to */ - struct device *dev; - - /* This mutex protects the access to the switch registers */ - struct mutex reg_lock; - - /* The MII bus and the address on the bus that is used to - * communication with the switch - */ - const struct mv88e6xxx_ops *smi_ops; - struct mii_bus *bus; - int sw_addr; - - /* Handles automatic disabling and re-enabling of the PHY - * polling unit. - */ - struct mutex ppu_mutex; - int ppu_disabled; - struct work_struct ppu_work; - struct timer_list ppu_timer; - - /* This mutex serialises access to the statistics unit. - * Hold this mutex over snapshot + dump sequences. - */ - struct mutex stats_mutex; - - /* This mutex serializes phy access for chips with - * indirect phy addressing. It is unused for chips - * with direct phy access. - */ - struct mutex phy_mutex; - - /* This mutex serializes eeprom access for chips with - * eeprom support. - */ - struct mutex eeprom_mutex; - - struct mv88e6xxx_priv_port ports[DSA_MAX_PORTS]; - - /* A switch may have a GPIO line tied to its reset pin. Parse - * this from the device tree, and use it before performing - * switch soft reset. - */ - struct gpio_desc *reset; - - /* set to size of eeprom if supported by the switch */ - int eeprom_len; - - /* Device node for the MDIO bus */ - struct device_node *mdio_np; - - /* And the MDIO bus itself */ - struct mii_bus *mdio_bus; -}; - -struct mv88e6xxx_ops { - int (*read)(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 *val); - int (*write)(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 val); -}; - -enum stat_type { - BANK0, - BANK1, - PORT, -}; - -struct mv88e6xxx_hw_stat { - char string[ETH_GSTRING_LEN]; - int sizeof_stat; - int reg; - enum stat_type type; -}; - -static inline bool mv88e6xxx_has(struct mv88e6xxx_priv_state *ps, - unsigned long flags) -{ - return (ps->info->flags & flags) == flags; -} - -#endif diff --git a/drivers/net/dsa/mv88e6xxx/Kconfig b/drivers/net/dsa/mv88e6xxx/Kconfig new file mode 100644 index 0000000..490bc06 --- /dev/null +++ b/drivers/net/dsa/mv88e6xxx/Kconfig @@ -0,0 +1,7 @@ +config NET_DSA_MV88E6XXX + tristate "Marvell 88E6xxx Ethernet switch fabric support" + depends on NET_DSA + select NET_DSA_TAG_EDSA + help + This driver adds support for most of the Marvell 88E6xxx models of + Ethernet switch chips, except 88E6060. diff --git a/drivers/net/dsa/mv88e6xxx/Makefile b/drivers/net/dsa/mv88e6xxx/Makefile new file mode 100644 index 0000000..1128fc7 --- /dev/null +++ b/drivers/net/dsa/mv88e6xxx/Makefile @@ -0,0 +1 @@ +obj-$(CONFIG_NET_DSA_MV88E6XXX) += mv88e6xxx.o diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.c new file mode 100644 index 0000000..2073f7b --- /dev/null +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.c @@ -0,0 +1,3954 @@ +/* + * Marvell 88e6xxx Ethernet switch single-chip support + * + * Copyright (c) 2008 Marvell Semiconductor + * + * Copyright (c) 2015 CMC Electronics, Inc. + * Added support for VLAN Table Unit operations + * + * Copyright (c) 2016 Andrew Lunn + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "mv88e6xxx.h" + +static void assert_reg_lock(struct mv88e6xxx_priv_state *ps) +{ + if (unlikely(!mutex_is_locked(&ps->reg_lock))) { + dev_err(ps->dev, "Switch registers lock not held!\n"); + dump_stack(); + } +} + +/* The switch ADDR[4:1] configuration pins define the chip SMI device address + * (ADDR[0] is always zero, thus only even SMI addresses can be strapped). + * + * When ADDR is all zero, the chip uses Single-chip Addressing Mode, assuming it + * is the only device connected to the SMI master. In this mode it responds to + * all 32 possible SMI addresses, and thus maps directly the internal devices. + * + * When ADDR is non-zero, the chip uses Multi-chip Addressing Mode, allowing + * multiple devices to share the SMI interface. In this mode it responds to only + * 2 registers, used to indirectly access the internal SMI devices. + */ + +static int mv88e6xxx_smi_read(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 *val) +{ + if (!ps->smi_ops) + return -EOPNOTSUPP; + + return ps->smi_ops->read(ps, addr, reg, val); +} + +static int mv88e6xxx_smi_write(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 val) +{ + if (!ps->smi_ops) + return -EOPNOTSUPP; + + return ps->smi_ops->write(ps, addr, reg, val); +} + +static int mv88e6xxx_smi_single_chip_read(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 *val) +{ + int ret; + + ret = mdiobus_read_nested(ps->bus, addr, reg); + if (ret < 0) + return ret; + + *val = ret & 0xffff; + + return 0; +} + +static int mv88e6xxx_smi_single_chip_write(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 val) +{ + int ret; + + ret = mdiobus_write_nested(ps->bus, addr, reg, val); + if (ret < 0) + return ret; + + return 0; +} + +static const struct mv88e6xxx_ops mv88e6xxx_smi_single_chip_ops = { + .read = mv88e6xxx_smi_single_chip_read, + .write = mv88e6xxx_smi_single_chip_write, +}; + +static int mv88e6xxx_smi_multi_chip_wait(struct mv88e6xxx_priv_state *ps) +{ + int ret; + int i; + + for (i = 0; i < 16; i++) { + ret = mdiobus_read_nested(ps->bus, ps->sw_addr, SMI_CMD); + if (ret < 0) + return ret; + + if ((ret & SMI_CMD_BUSY) == 0) + return 0; + } + + return -ETIMEDOUT; +} + +static int mv88e6xxx_smi_multi_chip_read(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 *val) +{ + int ret; + + /* Wait for the bus to become free. */ + ret = mv88e6xxx_smi_multi_chip_wait(ps); + if (ret < 0) + return ret; + + /* Transmit the read command. */ + ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_CMD, + SMI_CMD_OP_22_READ | (addr << 5) | reg); + if (ret < 0) + return ret; + + /* Wait for the read command to complete. */ + ret = mv88e6xxx_smi_multi_chip_wait(ps); + if (ret < 0) + return ret; + + /* Read the data. */ + ret = mdiobus_read_nested(ps->bus, ps->sw_addr, SMI_DATA); + if (ret < 0) + return ret; + + *val = ret & 0xffff; + + return 0; +} + +static int mv88e6xxx_smi_multi_chip_write(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 val) +{ + int ret; + + /* Wait for the bus to become free. */ + ret = mv88e6xxx_smi_multi_chip_wait(ps); + if (ret < 0) + return ret; + + /* Transmit the data to write. */ + ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_DATA, val); + if (ret < 0) + return ret; + + /* Transmit the write command. */ + ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_CMD, + SMI_CMD_OP_22_WRITE | (addr << 5) | reg); + if (ret < 0) + return ret; + + /* Wait for the write command to complete. */ + ret = mv88e6xxx_smi_multi_chip_wait(ps); + if (ret < 0) + return ret; + + return 0; +} + +static const struct mv88e6xxx_ops mv88e6xxx_smi_multi_chip_ops = { + .read = mv88e6xxx_smi_multi_chip_read, + .write = mv88e6xxx_smi_multi_chip_write, +}; + +static int mv88e6xxx_read(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 *val) +{ + int err; + + assert_reg_lock(ps); + + err = mv88e6xxx_smi_read(ps, addr, reg, val); + if (err) + return err; + + dev_dbg(ps->dev, "<- addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", + addr, reg, *val); + + return 0; +} + +static int mv88e6xxx_write(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 val) +{ + int err; + + assert_reg_lock(ps); + + err = mv88e6xxx_smi_write(ps, addr, reg, val); + if (err) + return err; + + dev_dbg(ps->dev, "-> addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", + addr, reg, val); + + return 0; +} + +static int _mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, + int addr, int reg) +{ + u16 val; + int err; + + err = mv88e6xxx_read(ps, addr, reg, &val); + if (err) + return err; + + return val; +} + +static int mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, int addr, + int reg) +{ + int ret; + + mutex_lock(&ps->reg_lock); + ret = _mv88e6xxx_reg_read(ps, addr, reg); + mutex_unlock(&ps->reg_lock); + + return ret; +} + +static int _mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, + int reg, u16 val) +{ + return mv88e6xxx_write(ps, addr, reg, val); +} + +static int mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, + int reg, u16 val) +{ + int ret; + + mutex_lock(&ps->reg_lock); + ret = _mv88e6xxx_reg_write(ps, addr, reg, val); + mutex_unlock(&ps->reg_lock); + + return ret; +} + +static int mv88e6xxx_set_addr_direct(struct dsa_switch *ds, u8 *addr) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int err; + + err = mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MAC_01, + (addr[0] << 8) | addr[1]); + if (err) + return err; + + err = mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MAC_23, + (addr[2] << 8) | addr[3]); + if (err) + return err; + + return mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MAC_45, + (addr[4] << 8) | addr[5]); +} + +static int mv88e6xxx_set_addr_indirect(struct dsa_switch *ds, u8 *addr) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int ret; + int i; + + for (i = 0; i < 6; i++) { + int j; + + /* Write the MAC address byte. */ + ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SWITCH_MAC, + GLOBAL2_SWITCH_MAC_BUSY | + (i << 8) | addr[i]); + if (ret) + return ret; + + /* Wait for the write to complete. */ + for (j = 0; j < 16; j++) { + ret = mv88e6xxx_reg_read(ps, REG_GLOBAL2, + GLOBAL2_SWITCH_MAC); + if (ret < 0) + return ret; + + if ((ret & GLOBAL2_SWITCH_MAC_BUSY) == 0) + break; + } + if (j == 16) + return -ETIMEDOUT; + } + + return 0; +} + +static int mv88e6xxx_set_addr(struct dsa_switch *ds, u8 *addr) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SWITCH_MAC)) + return mv88e6xxx_set_addr_indirect(ds, addr); + else + return mv88e6xxx_set_addr_direct(ds, addr); +} + +static int mv88e6xxx_mdio_read_direct(struct mv88e6xxx_priv_state *ps, + int addr, int regnum) +{ + if (addr >= 0) + return _mv88e6xxx_reg_read(ps, addr, regnum); + return 0xffff; +} + +static int mv88e6xxx_mdio_write_direct(struct mv88e6xxx_priv_state *ps, + int addr, int regnum, u16 val) +{ + if (addr >= 0) + return _mv88e6xxx_reg_write(ps, addr, regnum, val); + return 0; +} + +static int mv88e6xxx_ppu_disable(struct mv88e6xxx_priv_state *ps) +{ + int ret; + unsigned long timeout; + + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_CONTROL); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, + ret & ~GLOBAL_CONTROL_PPU_ENABLE); + if (ret) + return ret; + + timeout = jiffies + 1 * HZ; + while (time_before(jiffies, timeout)) { + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATUS); + if (ret < 0) + return ret; + + usleep_range(1000, 2000); + if ((ret & GLOBAL_STATUS_PPU_MASK) != + GLOBAL_STATUS_PPU_POLLING) + return 0; + } + + return -ETIMEDOUT; +} + +static int mv88e6xxx_ppu_enable(struct mv88e6xxx_priv_state *ps) +{ + int ret, err; + unsigned long timeout; + + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_CONTROL); + if (ret < 0) + return ret; + + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, + ret | GLOBAL_CONTROL_PPU_ENABLE); + if (err) + return err; + + timeout = jiffies + 1 * HZ; + while (time_before(jiffies, timeout)) { + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATUS); + if (ret < 0) + return ret; + + usleep_range(1000, 2000); + if ((ret & GLOBAL_STATUS_PPU_MASK) == + GLOBAL_STATUS_PPU_POLLING) + return 0; + } + + return -ETIMEDOUT; +} + +static void mv88e6xxx_ppu_reenable_work(struct work_struct *ugly) +{ + struct mv88e6xxx_priv_state *ps; + + ps = container_of(ugly, struct mv88e6xxx_priv_state, ppu_work); + + mutex_lock(&ps->reg_lock); + + if (mutex_trylock(&ps->ppu_mutex)) { + if (mv88e6xxx_ppu_enable(ps) == 0) + ps->ppu_disabled = 0; + mutex_unlock(&ps->ppu_mutex); + } + + mutex_unlock(&ps->reg_lock); +} + +static void mv88e6xxx_ppu_reenable_timer(unsigned long _ps) +{ + struct mv88e6xxx_priv_state *ps = (void *)_ps; + + schedule_work(&ps->ppu_work); +} + +static int mv88e6xxx_ppu_access_get(struct mv88e6xxx_priv_state *ps) +{ + int ret; + + mutex_lock(&ps->ppu_mutex); + + /* If the PHY polling unit is enabled, disable it so that + * we can access the PHY registers. If it was already + * disabled, cancel the timer that is going to re-enable + * it. + */ + if (!ps->ppu_disabled) { + ret = mv88e6xxx_ppu_disable(ps); + if (ret < 0) { + mutex_unlock(&ps->ppu_mutex); + return ret; + } + ps->ppu_disabled = 1; + } else { + del_timer(&ps->ppu_timer); + ret = 0; + } + + return ret; +} + +static void mv88e6xxx_ppu_access_put(struct mv88e6xxx_priv_state *ps) +{ + /* Schedule a timer to re-enable the PHY polling unit. */ + mod_timer(&ps->ppu_timer, jiffies + msecs_to_jiffies(10)); + mutex_unlock(&ps->ppu_mutex); +} + +static void mv88e6xxx_ppu_state_init(struct mv88e6xxx_priv_state *ps) +{ + mutex_init(&ps->ppu_mutex); + INIT_WORK(&ps->ppu_work, mv88e6xxx_ppu_reenable_work); + init_timer(&ps->ppu_timer); + ps->ppu_timer.data = (unsigned long)ps; + ps->ppu_timer.function = mv88e6xxx_ppu_reenable_timer; +} + +static int mv88e6xxx_mdio_read_ppu(struct mv88e6xxx_priv_state *ps, int addr, + int regnum) +{ + int ret; + + ret = mv88e6xxx_ppu_access_get(ps); + if (ret >= 0) { + ret = _mv88e6xxx_reg_read(ps, addr, regnum); + mv88e6xxx_ppu_access_put(ps); + } + + return ret; +} + +static int mv88e6xxx_mdio_write_ppu(struct mv88e6xxx_priv_state *ps, int addr, + int regnum, u16 val) +{ + int ret; + + ret = mv88e6xxx_ppu_access_get(ps); + if (ret >= 0) { + ret = _mv88e6xxx_reg_write(ps, addr, regnum, val); + mv88e6xxx_ppu_access_put(ps); + } + + return ret; +} + +static bool mv88e6xxx_6065_family(struct mv88e6xxx_priv_state *ps) +{ + return ps->info->family == MV88E6XXX_FAMILY_6065; +} + +static bool mv88e6xxx_6095_family(struct mv88e6xxx_priv_state *ps) +{ + return ps->info->family == MV88E6XXX_FAMILY_6095; +} + +static bool mv88e6xxx_6097_family(struct mv88e6xxx_priv_state *ps) +{ + return ps->info->family == MV88E6XXX_FAMILY_6097; +} + +static bool mv88e6xxx_6165_family(struct mv88e6xxx_priv_state *ps) +{ + return ps->info->family == MV88E6XXX_FAMILY_6165; +} + +static bool mv88e6xxx_6185_family(struct mv88e6xxx_priv_state *ps) +{ + return ps->info->family == MV88E6XXX_FAMILY_6185; +} + +static bool mv88e6xxx_6320_family(struct mv88e6xxx_priv_state *ps) +{ + return ps->info->family == MV88E6XXX_FAMILY_6320; +} + +static bool mv88e6xxx_6351_family(struct mv88e6xxx_priv_state *ps) +{ + return ps->info->family == MV88E6XXX_FAMILY_6351; +} + +static bool mv88e6xxx_6352_family(struct mv88e6xxx_priv_state *ps) +{ + return ps->info->family == MV88E6XXX_FAMILY_6352; +} + +static unsigned int mv88e6xxx_num_databases(struct mv88e6xxx_priv_state *ps) +{ + return ps->info->num_databases; +} + +static bool mv88e6xxx_has_fid_reg(struct mv88e6xxx_priv_state *ps) +{ + /* Does the device have dedicated FID registers for ATU and VTU ops? */ + if (mv88e6xxx_6097_family(ps) || mv88e6xxx_6165_family(ps) || + mv88e6xxx_6351_family(ps) || mv88e6xxx_6352_family(ps)) + return true; + + return false; +} + +/* We expect the switch to perform auto negotiation if there is a real + * phy. However, in the case of a fixed link phy, we force the port + * settings from the fixed link settings. + */ +static void mv88e6xxx_adjust_link(struct dsa_switch *ds, int port, + struct phy_device *phydev) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + u32 reg; + int ret; + + if (!phy_is_pseudo_fixed_link(phydev)) + return; + + mutex_lock(&ps->reg_lock); + + ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_PCS_CTRL); + if (ret < 0) + goto out; + + reg = ret & ~(PORT_PCS_CTRL_LINK_UP | + PORT_PCS_CTRL_FORCE_LINK | + PORT_PCS_CTRL_DUPLEX_FULL | + PORT_PCS_CTRL_FORCE_DUPLEX | + PORT_PCS_CTRL_UNFORCED); + + reg |= PORT_PCS_CTRL_FORCE_LINK; + if (phydev->link) + reg |= PORT_PCS_CTRL_LINK_UP; + + if (mv88e6xxx_6065_family(ps) && phydev->speed > SPEED_100) + goto out; + + switch (phydev->speed) { + case SPEED_1000: + reg |= PORT_PCS_CTRL_1000; + break; + case SPEED_100: + reg |= PORT_PCS_CTRL_100; + break; + case SPEED_10: + reg |= PORT_PCS_CTRL_10; + break; + default: + pr_info("Unknown speed"); + goto out; + } + + reg |= PORT_PCS_CTRL_FORCE_DUPLEX; + if (phydev->duplex == DUPLEX_FULL) + reg |= PORT_PCS_CTRL_DUPLEX_FULL; + + if ((mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps)) && + (port >= ps->info->num_ports - 2)) { + if (phydev->interface == PHY_INTERFACE_MODE_RGMII_RXID) + reg |= PORT_PCS_CTRL_RGMII_DELAY_RXCLK; + if (phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID) + reg |= PORT_PCS_CTRL_RGMII_DELAY_TXCLK; + if (phydev->interface == PHY_INTERFACE_MODE_RGMII_ID) + reg |= (PORT_PCS_CTRL_RGMII_DELAY_RXCLK | + PORT_PCS_CTRL_RGMII_DELAY_TXCLK); + } + _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_PCS_CTRL, reg); + +out: + mutex_unlock(&ps->reg_lock); +} + +static int _mv88e6xxx_stats_wait(struct mv88e6xxx_priv_state *ps) +{ + int ret; + int i; + + for (i = 0; i < 10; i++) { + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATS_OP); + if ((ret & GLOBAL_STATS_OP_BUSY) == 0) + return 0; + } + + return -ETIMEDOUT; +} + +static int _mv88e6xxx_stats_snapshot(struct mv88e6xxx_priv_state *ps, + int port) +{ + int ret; + + if (mv88e6xxx_6320_family(ps) || mv88e6xxx_6352_family(ps)) + port = (port + 1) << 5; + + /* Snapshot the hardware statistics counters for this port. */ + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_STATS_OP, + GLOBAL_STATS_OP_CAPTURE_PORT | + GLOBAL_STATS_OP_HIST_RX_TX | port); + if (ret < 0) + return ret; + + /* Wait for the snapshotting to complete. */ + ret = _mv88e6xxx_stats_wait(ps); + if (ret < 0) + return ret; + + return 0; +} + +static void _mv88e6xxx_stats_read(struct mv88e6xxx_priv_state *ps, + int stat, u32 *val) +{ + u32 _val; + int ret; + + *val = 0; + + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_STATS_OP, + GLOBAL_STATS_OP_READ_CAPTURED | + GLOBAL_STATS_OP_HIST_RX_TX | stat); + if (ret < 0) + return; + + ret = _mv88e6xxx_stats_wait(ps); + if (ret < 0) + return; + + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATS_COUNTER_32); + if (ret < 0) + return; + + _val = ret << 16; + + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATS_COUNTER_01); + if (ret < 0) + return; + + *val = _val | ret; +} + +static struct mv88e6xxx_hw_stat mv88e6xxx_hw_stats[] = { + { "in_good_octets", 8, 0x00, BANK0, }, + { "in_bad_octets", 4, 0x02, BANK0, }, + { "in_unicast", 4, 0x04, BANK0, }, + { "in_broadcasts", 4, 0x06, BANK0, }, + { "in_multicasts", 4, 0x07, BANK0, }, + { "in_pause", 4, 0x16, BANK0, }, + { "in_undersize", 4, 0x18, BANK0, }, + { "in_fragments", 4, 0x19, BANK0, }, + { "in_oversize", 4, 0x1a, BANK0, }, + { "in_jabber", 4, 0x1b, BANK0, }, + { "in_rx_error", 4, 0x1c, BANK0, }, + { "in_fcs_error", 4, 0x1d, BANK0, }, + { "out_octets", 8, 0x0e, BANK0, }, + { "out_unicast", 4, 0x10, BANK0, }, + { "out_broadcasts", 4, 0x13, BANK0, }, + { "out_multicasts", 4, 0x12, BANK0, }, + { "out_pause", 4, 0x15, BANK0, }, + { "excessive", 4, 0x11, BANK0, }, + { "collisions", 4, 0x1e, BANK0, }, + { "deferred", 4, 0x05, BANK0, }, + { "single", 4, 0x14, BANK0, }, + { "multiple", 4, 0x17, BANK0, }, + { "out_fcs_error", 4, 0x03, BANK0, }, + { "late", 4, 0x1f, BANK0, }, + { "hist_64bytes", 4, 0x08, BANK0, }, + { "hist_65_127bytes", 4, 0x09, BANK0, }, + { "hist_128_255bytes", 4, 0x0a, BANK0, }, + { "hist_256_511bytes", 4, 0x0b, BANK0, }, + { "hist_512_1023bytes", 4, 0x0c, BANK0, }, + { "hist_1024_max_bytes", 4, 0x0d, BANK0, }, + { "sw_in_discards", 4, 0x10, PORT, }, + { "sw_in_filtered", 2, 0x12, PORT, }, + { "sw_out_filtered", 2, 0x13, PORT, }, + { "in_discards", 4, 0x00 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_filtered", 4, 0x01 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_accepted", 4, 0x02 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_bad_accepted", 4, 0x03 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_good_avb_class_a", 4, 0x04 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_good_avb_class_b", 4, 0x05 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_bad_avb_class_a", 4, 0x06 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_bad_avb_class_b", 4, 0x07 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "tcam_counter_0", 4, 0x08 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "tcam_counter_1", 4, 0x09 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "tcam_counter_2", 4, 0x0a | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "tcam_counter_3", 4, 0x0b | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_da_unknown", 4, 0x0e | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_management", 4, 0x0f | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_0", 4, 0x10 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_1", 4, 0x11 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_2", 4, 0x12 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_3", 4, 0x13 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_4", 4, 0x14 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_5", 4, 0x15 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_6", 4, 0x16 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_7", 4, 0x17 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_cut_through", 4, 0x18 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_octets_a", 4, 0x1a | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_octets_b", 4, 0x1b | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_management", 4, 0x1f | GLOBAL_STATS_OP_BANK_1, BANK1, }, +}; + +static bool mv88e6xxx_has_stat(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_hw_stat *stat) +{ + switch (stat->type) { + case BANK0: + return true; + case BANK1: + return mv88e6xxx_6320_family(ps); + case PORT: + return mv88e6xxx_6095_family(ps) || + mv88e6xxx_6185_family(ps) || + mv88e6xxx_6097_family(ps) || + mv88e6xxx_6165_family(ps) || + mv88e6xxx_6351_family(ps) || + mv88e6xxx_6352_family(ps); + } + return false; +} + +static uint64_t _mv88e6xxx_get_ethtool_stat(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_hw_stat *s, + int port) +{ + u32 low; + u32 high = 0; + int ret; + u64 value; + + switch (s->type) { + case PORT: + ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), s->reg); + if (ret < 0) + return UINT64_MAX; + + low = ret; + if (s->sizeof_stat == 4) { + ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), + s->reg + 1); + if (ret < 0) + return UINT64_MAX; + high = ret; + } + break; + case BANK0: + case BANK1: + _mv88e6xxx_stats_read(ps, s->reg, &low); + if (s->sizeof_stat == 8) + _mv88e6xxx_stats_read(ps, s->reg + 1, &high); + } + value = (((u64)high) << 16) | low; + return value; +} + +static void mv88e6xxx_get_strings(struct dsa_switch *ds, int port, + uint8_t *data) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + struct mv88e6xxx_hw_stat *stat; + int i, j; + + for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { + stat = &mv88e6xxx_hw_stats[i]; + if (mv88e6xxx_has_stat(ps, stat)) { + memcpy(data + j * ETH_GSTRING_LEN, stat->string, + ETH_GSTRING_LEN); + j++; + } + } +} + +static int mv88e6xxx_get_sset_count(struct dsa_switch *ds) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + struct mv88e6xxx_hw_stat *stat; + int i, j; + + for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { + stat = &mv88e6xxx_hw_stats[i]; + if (mv88e6xxx_has_stat(ps, stat)) + j++; + } + return j; +} + +static void mv88e6xxx_get_ethtool_stats(struct dsa_switch *ds, int port, + uint64_t *data) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + struct mv88e6xxx_hw_stat *stat; + int ret; + int i, j; + + mutex_lock(&ps->reg_lock); + + ret = _mv88e6xxx_stats_snapshot(ps, port); + if (ret < 0) { + mutex_unlock(&ps->reg_lock); + return; + } + for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { + stat = &mv88e6xxx_hw_stats[i]; + if (mv88e6xxx_has_stat(ps, stat)) { + data[j] = _mv88e6xxx_get_ethtool_stat(ps, stat, port); + j++; + } + } + + mutex_unlock(&ps->reg_lock); +} + +static int mv88e6xxx_get_regs_len(struct dsa_switch *ds, int port) +{ + return 32 * sizeof(u16); +} + +static void mv88e6xxx_get_regs(struct dsa_switch *ds, int port, + struct ethtool_regs *regs, void *_p) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + u16 *p = _p; + int i; + + regs->version = 0; + + memset(p, 0xff, 32 * sizeof(u16)); + + mutex_lock(&ps->reg_lock); + + for (i = 0; i < 32; i++) { + int ret; + + ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), i); + if (ret >= 0) + p[i] = ret; + } + + mutex_unlock(&ps->reg_lock); +} + +static int _mv88e6xxx_wait(struct mv88e6xxx_priv_state *ps, int reg, int offset, + u16 mask) +{ + unsigned long timeout = jiffies + HZ / 10; + + while (time_before(jiffies, timeout)) { + int ret; + + ret = _mv88e6xxx_reg_read(ps, reg, offset); + if (ret < 0) + return ret; + if (!(ret & mask)) + return 0; + + usleep_range(1000, 2000); + } + return -ETIMEDOUT; +} + +static int mv88e6xxx_wait(struct mv88e6xxx_priv_state *ps, int reg, + int offset, u16 mask) +{ + int ret; + + mutex_lock(&ps->reg_lock); + ret = _mv88e6xxx_wait(ps, reg, offset, mask); + mutex_unlock(&ps->reg_lock); + + return ret; +} + +static int mv88e6xxx_mdio_wait(struct mv88e6xxx_priv_state *ps) +{ + return _mv88e6xxx_wait(ps, REG_GLOBAL2, GLOBAL2_SMI_OP, + GLOBAL2_SMI_OP_BUSY); +} + +static int mv88e6xxx_eeprom_load_wait(struct dsa_switch *ds) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + + return mv88e6xxx_wait(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, + GLOBAL2_EEPROM_OP_LOAD); +} + +static int mv88e6xxx_eeprom_busy_wait(struct dsa_switch *ds) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + + return mv88e6xxx_wait(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, + GLOBAL2_EEPROM_OP_BUSY); +} + +static int mv88e6xxx_read_eeprom_word(struct dsa_switch *ds, int addr) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int ret; + + mutex_lock(&ps->eeprom_mutex); + + ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, + GLOBAL2_EEPROM_OP_READ | + (addr & GLOBAL2_EEPROM_OP_ADDR_MASK)); + if (ret < 0) + goto error; + + ret = mv88e6xxx_eeprom_busy_wait(ds); + if (ret < 0) + goto error; + + ret = mv88e6xxx_reg_read(ps, REG_GLOBAL2, GLOBAL2_EEPROM_DATA); +error: + mutex_unlock(&ps->eeprom_mutex); + return ret; +} + +static int mv88e6xxx_get_eeprom_len(struct dsa_switch *ds) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) + return ps->eeprom_len; + + return 0; +} + +static int mv88e6xxx_get_eeprom(struct dsa_switch *ds, + struct ethtool_eeprom *eeprom, u8 *data) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int offset; + int len; + int ret; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) + return -EOPNOTSUPP; + + offset = eeprom->offset; + len = eeprom->len; + eeprom->len = 0; + + eeprom->magic = 0xc3ec4951; + + ret = mv88e6xxx_eeprom_load_wait(ds); + if (ret < 0) + return ret; + + if (offset & 1) { + int word; + + word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); + if (word < 0) + return word; + + *data++ = (word >> 8) & 0xff; + + offset++; + len--; + eeprom->len++; + } + + while (len >= 2) { + int word; + + word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); + if (word < 0) + return word; + + *data++ = word & 0xff; + *data++ = (word >> 8) & 0xff; + + offset += 2; + len -= 2; + eeprom->len += 2; + } + + if (len) { + int word; + + word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); + if (word < 0) + return word; + + *data++ = word & 0xff; + + offset++; + len--; + eeprom->len++; + } + + return 0; +} + +static int mv88e6xxx_eeprom_is_readonly(struct dsa_switch *ds) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int ret; + + ret = mv88e6xxx_reg_read(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP); + if (ret < 0) + return ret; + + if (!(ret & GLOBAL2_EEPROM_OP_WRITE_EN)) + return -EROFS; + + return 0; +} + +static int mv88e6xxx_write_eeprom_word(struct dsa_switch *ds, int addr, + u16 data) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int ret; + + mutex_lock(&ps->eeprom_mutex); + + ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_EEPROM_DATA, data); + if (ret < 0) + goto error; + + ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, + GLOBAL2_EEPROM_OP_WRITE | + (addr & GLOBAL2_EEPROM_OP_ADDR_MASK)); + if (ret < 0) + goto error; + + ret = mv88e6xxx_eeprom_busy_wait(ds); +error: + mutex_unlock(&ps->eeprom_mutex); + return ret; +} + +static int mv88e6xxx_set_eeprom(struct dsa_switch *ds, + struct ethtool_eeprom *eeprom, u8 *data) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int offset; + int ret; + int len; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) + return -EOPNOTSUPP; + + if (eeprom->magic != 0xc3ec4951) + return -EINVAL; + + ret = mv88e6xxx_eeprom_is_readonly(ds); + if (ret) + return ret; + + offset = eeprom->offset; + len = eeprom->len; + eeprom->len = 0; + + ret = mv88e6xxx_eeprom_load_wait(ds); + if (ret < 0) + return ret; + + if (offset & 1) { + int word; + + word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); + if (word < 0) + return word; + + word = (*data++ << 8) | (word & 0xff); + + ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); + if (ret < 0) + return ret; + + offset++; + len--; + eeprom->len++; + } + + while (len >= 2) { + int word; + + word = *data++; + word |= *data++ << 8; + + ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); + if (ret < 0) + return ret; + + offset += 2; + len -= 2; + eeprom->len += 2; + } + + if (len) { + int word; + + word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); + if (word < 0) + return word; + + word = (word & 0xff00) | *data++; + + ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); + if (ret < 0) + return ret; + + offset++; + len--; + eeprom->len++; + } + + return 0; +} + +static int _mv88e6xxx_atu_wait(struct mv88e6xxx_priv_state *ps) +{ + return _mv88e6xxx_wait(ps, REG_GLOBAL, GLOBAL_ATU_OP, + GLOBAL_ATU_OP_BUSY); +} + +static int mv88e6xxx_mdio_read_indirect(struct mv88e6xxx_priv_state *ps, + int addr, int regnum) +{ + int ret; + + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SMI_OP, + GLOBAL2_SMI_OP_22_READ | (addr << 5) | + regnum); + if (ret < 0) + return ret; + + ret = mv88e6xxx_mdio_wait(ps); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL2, GLOBAL2_SMI_DATA); + + return ret; +} + +static int mv88e6xxx_mdio_write_indirect(struct mv88e6xxx_priv_state *ps, + int addr, int regnum, u16 val) +{ + int ret; + + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SMI_DATA, val); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SMI_OP, + GLOBAL2_SMI_OP_22_WRITE | (addr << 5) | + regnum); + + return mv88e6xxx_mdio_wait(ps); +} + +static int mv88e6xxx_get_eee(struct dsa_switch *ds, int port, + struct ethtool_eee *e) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int reg; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEE)) + return -EOPNOTSUPP; + + mutex_lock(&ps->reg_lock); + + reg = mv88e6xxx_mdio_read_indirect(ps, port, 16); + if (reg < 0) + goto out; + + e->eee_enabled = !!(reg & 0x0200); + e->tx_lpi_enabled = !!(reg & 0x0100); + + reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_STATUS); + if (reg < 0) + goto out; + + e->eee_active = !!(reg & PORT_STATUS_EEE); + reg = 0; + +out: + mutex_unlock(&ps->reg_lock); + return reg; +} + +static int mv88e6xxx_set_eee(struct dsa_switch *ds, int port, + struct phy_device *phydev, struct ethtool_eee *e) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int reg; + int ret; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEE)) + return -EOPNOTSUPP; + + mutex_lock(&ps->reg_lock); + + ret = mv88e6xxx_mdio_read_indirect(ps, port, 16); + if (ret < 0) + goto out; + + reg = ret & ~0x0300; + if (e->eee_enabled) + reg |= 0x0200; + if (e->tx_lpi_enabled) + reg |= 0x0100; + + ret = mv88e6xxx_mdio_write_indirect(ps, port, 16, reg); +out: + mutex_unlock(&ps->reg_lock); + + return ret; +} + +static int _mv88e6xxx_atu_cmd(struct mv88e6xxx_priv_state *ps, u16 fid, u16 cmd) +{ + int ret; + + if (mv88e6xxx_has_fid_reg(ps)) { + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_FID, fid); + if (ret < 0) + return ret; + } else if (mv88e6xxx_num_databases(ps) == 256) { + /* ATU DBNum[7:4] are located in ATU Control 15:12 */ + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_ATU_CONTROL); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_CONTROL, + (ret & 0xfff) | + ((fid << 8) & 0xf000)); + if (ret < 0) + return ret; + + /* ATU DBNum[3:0] are located in ATU Operation 3:0 */ + cmd |= fid & 0xf; + } + + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_OP, cmd); + if (ret < 0) + return ret; + + return _mv88e6xxx_atu_wait(ps); +} + +static int _mv88e6xxx_atu_data_write(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_atu_entry *entry) +{ + u16 data = entry->state & GLOBAL_ATU_DATA_STATE_MASK; + + if (entry->state != GLOBAL_ATU_DATA_STATE_UNUSED) { + unsigned int mask, shift; + + if (entry->trunk) { + data |= GLOBAL_ATU_DATA_TRUNK; + mask = GLOBAL_ATU_DATA_TRUNK_ID_MASK; + shift = GLOBAL_ATU_DATA_TRUNK_ID_SHIFT; + } else { + mask = GLOBAL_ATU_DATA_PORT_VECTOR_MASK; + shift = GLOBAL_ATU_DATA_PORT_VECTOR_SHIFT; + } + + data |= (entry->portv_trunkid << shift) & mask; + } + + return _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_DATA, data); +} + +static int _mv88e6xxx_atu_flush_move(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_atu_entry *entry, + bool static_too) +{ + int op; + int err; + + err = _mv88e6xxx_atu_wait(ps); + if (err) + return err; + + err = _mv88e6xxx_atu_data_write(ps, entry); + if (err) + return err; + + if (entry->fid) { + op = static_too ? GLOBAL_ATU_OP_FLUSH_MOVE_ALL_DB : + GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC_DB; + } else { + op = static_too ? GLOBAL_ATU_OP_FLUSH_MOVE_ALL : + GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC; + } + + return _mv88e6xxx_atu_cmd(ps, entry->fid, op); +} + +static int _mv88e6xxx_atu_flush(struct mv88e6xxx_priv_state *ps, + u16 fid, bool static_too) +{ + struct mv88e6xxx_atu_entry entry = { + .fid = fid, + .state = 0, /* EntryState bits must be 0 */ + }; + + return _mv88e6xxx_atu_flush_move(ps, &entry, static_too); +} + +static int _mv88e6xxx_atu_move(struct mv88e6xxx_priv_state *ps, u16 fid, + int from_port, int to_port, bool static_too) +{ + struct mv88e6xxx_atu_entry entry = { + .trunk = false, + .fid = fid, + }; + + /* EntryState bits must be 0xF */ + entry.state = GLOBAL_ATU_DATA_STATE_MASK; + + /* ToPort and FromPort are respectively in PortVec bits 7:4 and 3:0 */ + entry.portv_trunkid = (to_port & 0x0f) << 4; + entry.portv_trunkid |= from_port & 0x0f; + + return _mv88e6xxx_atu_flush_move(ps, &entry, static_too); +} + +static int _mv88e6xxx_atu_remove(struct mv88e6xxx_priv_state *ps, u16 fid, + int port, bool static_too) +{ + /* Destination port 0xF means remove the entries */ + return _mv88e6xxx_atu_move(ps, fid, port, 0x0f, static_too); +} + +static const char * const mv88e6xxx_port_state_names[] = { + [PORT_CONTROL_STATE_DISABLED] = "Disabled", + [PORT_CONTROL_STATE_BLOCKING] = "Blocking/Listening", + [PORT_CONTROL_STATE_LEARNING] = "Learning", + [PORT_CONTROL_STATE_FORWARDING] = "Forwarding", +}; + +static int _mv88e6xxx_port_state(struct mv88e6xxx_priv_state *ps, int port, + u8 state) +{ + struct dsa_switch *ds = ps->ds; + int reg, ret = 0; + u8 oldstate; + + reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_CONTROL); + if (reg < 0) + return reg; + + oldstate = reg & PORT_CONTROL_STATE_MASK; + + if (oldstate != state) { + /* Flush forwarding database if we're moving a port + * from Learning or Forwarding state to Disabled or + * Blocking or Listening state. + */ + if ((oldstate == PORT_CONTROL_STATE_LEARNING || + oldstate == PORT_CONTROL_STATE_FORWARDING) && + (state == PORT_CONTROL_STATE_DISABLED || + state == PORT_CONTROL_STATE_BLOCKING)) { + ret = _mv88e6xxx_atu_remove(ps, 0, port, false); + if (ret) + return ret; + } + + reg = (reg & ~PORT_CONTROL_STATE_MASK) | state; + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL, + reg); + if (ret) + return ret; + + netdev_dbg(ds->ports[port].netdev, "PortState %s (was %s)\n", + mv88e6xxx_port_state_names[state], + mv88e6xxx_port_state_names[oldstate]); + } + + return ret; +} + +static int _mv88e6xxx_port_based_vlan_map(struct mv88e6xxx_priv_state *ps, + int port) +{ + struct net_device *bridge = ps->ports[port].bridge_dev; + const u16 mask = (1 << ps->info->num_ports) - 1; + struct dsa_switch *ds = ps->ds; + u16 output_ports = 0; + int reg; + int i; + + /* allow CPU port or DSA link(s) to send frames to every port */ + if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) { + output_ports = mask; + } else { + for (i = 0; i < ps->info->num_ports; ++i) { + /* allow sending frames to every group member */ + if (bridge && ps->ports[i].bridge_dev == bridge) + output_ports |= BIT(i); + + /* allow sending frames to CPU port and DSA link(s) */ + if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i)) + output_ports |= BIT(i); + } + } + + /* prevent frames from going back out of the port they came in on */ + output_ports &= ~BIT(port); + + reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_BASE_VLAN); + if (reg < 0) + return reg; + + reg &= ~mask; + reg |= output_ports & mask; + + return _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_BASE_VLAN, reg); +} + +static void mv88e6xxx_port_stp_state_set(struct dsa_switch *ds, int port, + u8 state) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int stp_state; + int err; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_PORTSTATE)) + return; + + switch (state) { + case BR_STATE_DISABLED: + stp_state = PORT_CONTROL_STATE_DISABLED; + break; + case BR_STATE_BLOCKING: + case BR_STATE_LISTENING: + stp_state = PORT_CONTROL_STATE_BLOCKING; + break; + case BR_STATE_LEARNING: + stp_state = PORT_CONTROL_STATE_LEARNING; + break; + case BR_STATE_FORWARDING: + default: + stp_state = PORT_CONTROL_STATE_FORWARDING; + break; + } + + mutex_lock(&ps->reg_lock); + err = _mv88e6xxx_port_state(ps, port, stp_state); + mutex_unlock(&ps->reg_lock); + + if (err) + netdev_err(ds->ports[port].netdev, + "failed to update state to %s\n", + mv88e6xxx_port_state_names[stp_state]); +} + +static int _mv88e6xxx_port_pvid(struct mv88e6xxx_priv_state *ps, int port, + u16 *new, u16 *old) +{ + struct dsa_switch *ds = ps->ds; + u16 pvid; + int ret; + + ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_DEFAULT_VLAN); + if (ret < 0) + return ret; + + pvid = ret & PORT_DEFAULT_VLAN_MASK; + + if (new) { + ret &= ~PORT_DEFAULT_VLAN_MASK; + ret |= *new & PORT_DEFAULT_VLAN_MASK; + + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_DEFAULT_VLAN, ret); + if (ret < 0) + return ret; + + netdev_dbg(ds->ports[port].netdev, + "DefaultVID %d (was %d)\n", *new, pvid); + } + + if (old) + *old = pvid; + + return 0; +} + +static int _mv88e6xxx_port_pvid_get(struct mv88e6xxx_priv_state *ps, + int port, u16 *pvid) +{ + return _mv88e6xxx_port_pvid(ps, port, NULL, pvid); +} + +static int _mv88e6xxx_port_pvid_set(struct mv88e6xxx_priv_state *ps, + int port, u16 pvid) +{ + return _mv88e6xxx_port_pvid(ps, port, &pvid, NULL); +} + +static int _mv88e6xxx_vtu_wait(struct mv88e6xxx_priv_state *ps) +{ + return _mv88e6xxx_wait(ps, REG_GLOBAL, GLOBAL_VTU_OP, + GLOBAL_VTU_OP_BUSY); +} + +static int _mv88e6xxx_vtu_cmd(struct mv88e6xxx_priv_state *ps, u16 op) +{ + int ret; + + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_OP, op); + if (ret < 0) + return ret; + + return _mv88e6xxx_vtu_wait(ps); +} + +static int _mv88e6xxx_vtu_stu_flush(struct mv88e6xxx_priv_state *ps) +{ + int ret; + + ret = _mv88e6xxx_vtu_wait(ps); + if (ret < 0) + return ret; + + return _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_FLUSH_ALL); +} + +static int _mv88e6xxx_vtu_stu_data_read(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_vtu_stu_entry *entry, + unsigned int nibble_offset) +{ + u16 regs[3]; + int i; + int ret; + + for (i = 0; i < 3; ++i) { + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, + GLOBAL_VTU_DATA_0_3 + i); + if (ret < 0) + return ret; + + regs[i] = ret; + } + + for (i = 0; i < ps->info->num_ports; ++i) { + unsigned int shift = (i % 4) * 4 + nibble_offset; + u16 reg = regs[i / 4]; + + entry->data[i] = (reg >> shift) & GLOBAL_VTU_STU_DATA_MASK; + } + + return 0; +} + +static int mv88e6xxx_vtu_data_read(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + return _mv88e6xxx_vtu_stu_data_read(ps, entry, 0); +} + +static int mv88e6xxx_stu_data_read(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + return _mv88e6xxx_vtu_stu_data_read(ps, entry, 2); +} + +static int _mv88e6xxx_vtu_stu_data_write(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_vtu_stu_entry *entry, + unsigned int nibble_offset) +{ + u16 regs[3] = { 0 }; + int i; + int ret; + + for (i = 0; i < ps->info->num_ports; ++i) { + unsigned int shift = (i % 4) * 4 + nibble_offset; + u8 data = entry->data[i]; + + regs[i / 4] |= (data & GLOBAL_VTU_STU_DATA_MASK) << shift; + } + + for (i = 0; i < 3; ++i) { + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, + GLOBAL_VTU_DATA_0_3 + i, regs[i]); + if (ret < 0) + return ret; + } + + return 0; +} + +static int mv88e6xxx_vtu_data_write(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + return _mv88e6xxx_vtu_stu_data_write(ps, entry, 0); +} + +static int mv88e6xxx_stu_data_write(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + return _mv88e6xxx_vtu_stu_data_write(ps, entry, 2); +} + +static int _mv88e6xxx_vtu_vid_write(struct mv88e6xxx_priv_state *ps, u16 vid) +{ + return _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_VID, + vid & GLOBAL_VTU_VID_MASK); +} + +static int _mv88e6xxx_vtu_getnext(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + struct mv88e6xxx_vtu_stu_entry next = { 0 }; + int ret; + + ret = _mv88e6xxx_vtu_wait(ps); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_VTU_GET_NEXT); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_VTU_VID); + if (ret < 0) + return ret; + + next.vid = ret & GLOBAL_VTU_VID_MASK; + next.valid = !!(ret & GLOBAL_VTU_VID_VALID); + + if (next.valid) { + ret = mv88e6xxx_vtu_data_read(ps, &next); + if (ret < 0) + return ret; + + if (mv88e6xxx_has_fid_reg(ps)) { + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, + GLOBAL_VTU_FID); + if (ret < 0) + return ret; + + next.fid = ret & GLOBAL_VTU_FID_MASK; + } else if (mv88e6xxx_num_databases(ps) == 256) { + /* VTU DBNum[7:4] are located in VTU Operation 11:8, and + * VTU DBNum[3:0] are located in VTU Operation 3:0 + */ + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, + GLOBAL_VTU_OP); + if (ret < 0) + return ret; + + next.fid = (ret & 0xf00) >> 4; + next.fid |= ret & 0xf; + } + + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_STU)) { + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, + GLOBAL_VTU_SID); + if (ret < 0) + return ret; + + next.sid = ret & GLOBAL_VTU_SID_MASK; + } + } + + *entry = next; + return 0; +} + +static int mv88e6xxx_port_vlan_dump(struct dsa_switch *ds, int port, + struct switchdev_obj_port_vlan *vlan, + int (*cb)(struct switchdev_obj *obj)) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + struct mv88e6xxx_vtu_stu_entry next; + u16 pvid; + int err; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) + return -EOPNOTSUPP; + + mutex_lock(&ps->reg_lock); + + err = _mv88e6xxx_port_pvid_get(ps, port, &pvid); + if (err) + goto unlock; + + err = _mv88e6xxx_vtu_vid_write(ps, GLOBAL_VTU_VID_MASK); + if (err) + goto unlock; + + do { + err = _mv88e6xxx_vtu_getnext(ps, &next); + if (err) + break; + + if (!next.valid) + break; + + if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) + continue; + + /* reinit and dump this VLAN obj */ + vlan->vid_begin = next.vid; + vlan->vid_end = next.vid; + vlan->flags = 0; + + if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED) + vlan->flags |= BRIDGE_VLAN_INFO_UNTAGGED; + + if (next.vid == pvid) + vlan->flags |= BRIDGE_VLAN_INFO_PVID; + + err = cb(&vlan->obj); + if (err) + break; + } while (next.vid < GLOBAL_VTU_VID_MASK); + +unlock: + mutex_unlock(&ps->reg_lock); + + return err; +} + +static int _mv88e6xxx_vtu_loadpurge(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + u16 op = GLOBAL_VTU_OP_VTU_LOAD_PURGE; + u16 reg = 0; + int ret; + + ret = _mv88e6xxx_vtu_wait(ps); + if (ret < 0) + return ret; + + if (!entry->valid) + goto loadpurge; + + /* Write port member tags */ + ret = mv88e6xxx_vtu_data_write(ps, entry); + if (ret < 0) + return ret; + + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_STU)) { + reg = entry->sid & GLOBAL_VTU_SID_MASK; + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_SID, reg); + if (ret < 0) + return ret; + } + + if (mv88e6xxx_has_fid_reg(ps)) { + reg = entry->fid & GLOBAL_VTU_FID_MASK; + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_FID, reg); + if (ret < 0) + return ret; + } else if (mv88e6xxx_num_databases(ps) == 256) { + /* VTU DBNum[7:4] are located in VTU Operation 11:8, and + * VTU DBNum[3:0] are located in VTU Operation 3:0 + */ + op |= (entry->fid & 0xf0) << 8; + op |= entry->fid & 0xf; + } + + reg = GLOBAL_VTU_VID_VALID; +loadpurge: + reg |= entry->vid & GLOBAL_VTU_VID_MASK; + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_VID, reg); + if (ret < 0) + return ret; + + return _mv88e6xxx_vtu_cmd(ps, op); +} + +static int _mv88e6xxx_stu_getnext(struct mv88e6xxx_priv_state *ps, u8 sid, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + struct mv88e6xxx_vtu_stu_entry next = { 0 }; + int ret; + + ret = _mv88e6xxx_vtu_wait(ps); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_SID, + sid & GLOBAL_VTU_SID_MASK); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_STU_GET_NEXT); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_VTU_SID); + if (ret < 0) + return ret; + + next.sid = ret & GLOBAL_VTU_SID_MASK; + + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_VTU_VID); + if (ret < 0) + return ret; + + next.valid = !!(ret & GLOBAL_VTU_VID_VALID); + + if (next.valid) { + ret = mv88e6xxx_stu_data_read(ps, &next); + if (ret < 0) + return ret; + } + + *entry = next; + return 0; +} + +static int _mv88e6xxx_stu_loadpurge(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + u16 reg = 0; + int ret; + + ret = _mv88e6xxx_vtu_wait(ps); + if (ret < 0) + return ret; + + if (!entry->valid) + goto loadpurge; + + /* Write port states */ + ret = mv88e6xxx_stu_data_write(ps, entry); + if (ret < 0) + return ret; + + reg = GLOBAL_VTU_VID_VALID; +loadpurge: + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_VID, reg); + if (ret < 0) + return ret; + + reg = entry->sid & GLOBAL_VTU_SID_MASK; + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_SID, reg); + if (ret < 0) + return ret; + + return _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_STU_LOAD_PURGE); +} + +static int _mv88e6xxx_port_fid(struct mv88e6xxx_priv_state *ps, int port, + u16 *new, u16 *old) +{ + struct dsa_switch *ds = ps->ds; + u16 upper_mask; + u16 fid; + int ret; + + if (mv88e6xxx_num_databases(ps) == 4096) + upper_mask = 0xff; + else if (mv88e6xxx_num_databases(ps) == 256) + upper_mask = 0xf; + else + return -EOPNOTSUPP; + + /* Port's default FID bits 3:0 are located in reg 0x06, offset 12 */ + ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_BASE_VLAN); + if (ret < 0) + return ret; + + fid = (ret & PORT_BASE_VLAN_FID_3_0_MASK) >> 12; + + if (new) { + ret &= ~PORT_BASE_VLAN_FID_3_0_MASK; + ret |= (*new << 12) & PORT_BASE_VLAN_FID_3_0_MASK; + + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_BASE_VLAN, + ret); + if (ret < 0) + return ret; + } + + /* Port's default FID bits 11:4 are located in reg 0x05, offset 0 */ + ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_CONTROL_1); + if (ret < 0) + return ret; + + fid |= (ret & upper_mask) << 4; + + if (new) { + ret &= ~upper_mask; + ret |= (*new >> 4) & upper_mask; + + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL_1, + ret); + if (ret < 0) + return ret; + + netdev_dbg(ds->ports[port].netdev, + "FID %d (was %d)\n", *new, fid); + } + + if (old) + *old = fid; + + return 0; +} + +static int _mv88e6xxx_port_fid_get(struct mv88e6xxx_priv_state *ps, + int port, u16 *fid) +{ + return _mv88e6xxx_port_fid(ps, port, NULL, fid); +} + +static int _mv88e6xxx_port_fid_set(struct mv88e6xxx_priv_state *ps, + int port, u16 fid) +{ + return _mv88e6xxx_port_fid(ps, port, &fid, NULL); +} + +static int _mv88e6xxx_fid_new(struct mv88e6xxx_priv_state *ps, u16 *fid) +{ + DECLARE_BITMAP(fid_bitmap, MV88E6XXX_N_FID); + struct mv88e6xxx_vtu_stu_entry vlan; + int i, err; + + bitmap_zero(fid_bitmap, MV88E6XXX_N_FID); + + /* Set every FID bit used by the (un)bridged ports */ + for (i = 0; i < ps->info->num_ports; ++i) { + err = _mv88e6xxx_port_fid_get(ps, i, fid); + if (err) + return err; + + set_bit(*fid, fid_bitmap); + } + + /* Set every FID bit used by the VLAN entries */ + err = _mv88e6xxx_vtu_vid_write(ps, GLOBAL_VTU_VID_MASK); + if (err) + return err; + + do { + err = _mv88e6xxx_vtu_getnext(ps, &vlan); + if (err) + return err; + + if (!vlan.valid) + break; + + set_bit(vlan.fid, fid_bitmap); + } while (vlan.vid < GLOBAL_VTU_VID_MASK); + + /* The reset value 0x000 is used to indicate that multiple address + * databases are not needed. Return the next positive available. + */ + *fid = find_next_zero_bit(fid_bitmap, MV88E6XXX_N_FID, 1); + if (unlikely(*fid >= mv88e6xxx_num_databases(ps))) + return -ENOSPC; + + /* Clear the database */ + return _mv88e6xxx_atu_flush(ps, *fid, true); +} + +static int _mv88e6xxx_vtu_new(struct mv88e6xxx_priv_state *ps, u16 vid, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + struct dsa_switch *ds = ps->ds; + struct mv88e6xxx_vtu_stu_entry vlan = { + .valid = true, + .vid = vid, + }; + int i, err; + + err = _mv88e6xxx_fid_new(ps, &vlan.fid); + if (err) + return err; + + /* exclude all ports except the CPU and DSA ports */ + for (i = 0; i < ps->info->num_ports; ++i) + vlan.data[i] = dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i) + ? GLOBAL_VTU_DATA_MEMBER_TAG_UNMODIFIED + : GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER; + + if (mv88e6xxx_6097_family(ps) || mv88e6xxx_6165_family(ps) || + mv88e6xxx_6351_family(ps) || mv88e6xxx_6352_family(ps)) { + struct mv88e6xxx_vtu_stu_entry vstp; + + /* Adding a VTU entry requires a valid STU entry. As VSTP is not + * implemented, only one STU entry is needed to cover all VTU + * entries. Thus, validate the SID 0. + */ + vlan.sid = 0; + err = _mv88e6xxx_stu_getnext(ps, GLOBAL_VTU_SID_MASK, &vstp); + if (err) + return err; + + if (vstp.sid != vlan.sid || !vstp.valid) { + memset(&vstp, 0, sizeof(vstp)); + vstp.valid = true; + vstp.sid = vlan.sid; + + err = _mv88e6xxx_stu_loadpurge(ps, &vstp); + if (err) + return err; + } + } + + *entry = vlan; + return 0; +} + +static int _mv88e6xxx_vtu_get(struct mv88e6xxx_priv_state *ps, u16 vid, + struct mv88e6xxx_vtu_stu_entry *entry, bool creat) +{ + int err; + + if (!vid) + return -EINVAL; + + err = _mv88e6xxx_vtu_vid_write(ps, vid - 1); + if (err) + return err; + + err = _mv88e6xxx_vtu_getnext(ps, entry); + if (err) + return err; + + if (entry->vid != vid || !entry->valid) { + if (!creat) + return -EOPNOTSUPP; + /* -ENOENT would've been more appropriate, but switchdev expects + * -EOPNOTSUPP to inform bridge about an eventual software VLAN. + */ + + err = _mv88e6xxx_vtu_new(ps, vid, entry); + } + + return err; +} + +static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port, + u16 vid_begin, u16 vid_end) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + struct mv88e6xxx_vtu_stu_entry vlan; + int i, err; + + if (!vid_begin) + return -EOPNOTSUPP; + + mutex_lock(&ps->reg_lock); + + err = _mv88e6xxx_vtu_vid_write(ps, vid_begin - 1); + if (err) + goto unlock; + + do { + err = _mv88e6xxx_vtu_getnext(ps, &vlan); + if (err) + goto unlock; + + if (!vlan.valid) + break; + + if (vlan.vid > vid_end) + break; + + for (i = 0; i < ps->info->num_ports; ++i) { + if (dsa_is_dsa_port(ds, i) || dsa_is_cpu_port(ds, i)) + continue; + + if (vlan.data[i] == + GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) + continue; + + if (ps->ports[i].bridge_dev == + ps->ports[port].bridge_dev) + break; /* same bridge, check next VLAN */ + + netdev_warn(ds->ports[port].netdev, + "hardware VLAN %d already used by %s\n", + vlan.vid, + netdev_name(ps->ports[i].bridge_dev)); + err = -EOPNOTSUPP; + goto unlock; + } + } while (vlan.vid < vid_end); + +unlock: + mutex_unlock(&ps->reg_lock); + + return err; +} + +static const char * const mv88e6xxx_port_8021q_mode_names[] = { + [PORT_CONTROL_2_8021Q_DISABLED] = "Disabled", + [PORT_CONTROL_2_8021Q_FALLBACK] = "Fallback", + [PORT_CONTROL_2_8021Q_CHECK] = "Check", + [PORT_CONTROL_2_8021Q_SECURE] = "Secure", +}; + +static int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port, + bool vlan_filtering) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + u16 old, new = vlan_filtering ? PORT_CONTROL_2_8021Q_SECURE : + PORT_CONTROL_2_8021Q_DISABLED; + int ret; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) + return -EOPNOTSUPP; + + mutex_lock(&ps->reg_lock); + + ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_CONTROL_2); + if (ret < 0) + goto unlock; + + old = ret & PORT_CONTROL_2_8021Q_MASK; + + if (new != old) { + ret &= ~PORT_CONTROL_2_8021Q_MASK; + ret |= new & PORT_CONTROL_2_8021Q_MASK; + + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL_2, + ret); + if (ret < 0) + goto unlock; + + netdev_dbg(ds->ports[port].netdev, "802.1Q Mode %s (was %s)\n", + mv88e6xxx_port_8021q_mode_names[new], + mv88e6xxx_port_8021q_mode_names[old]); + } + + ret = 0; +unlock: + mutex_unlock(&ps->reg_lock); + + return ret; +} + +static int +mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int err; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) + return -EOPNOTSUPP; + + /* If the requested port doesn't belong to the same bridge as the VLAN + * members, do not support it (yet) and fallback to software VLAN. + */ + err = mv88e6xxx_port_check_hw_vlan(ds, port, vlan->vid_begin, + vlan->vid_end); + if (err) + return err; + + /* We don't need any dynamic resource from the kernel (yet), + * so skip the prepare phase. + */ + return 0; +} + +static int _mv88e6xxx_port_vlan_add(struct mv88e6xxx_priv_state *ps, int port, + u16 vid, bool untagged) +{ + struct mv88e6xxx_vtu_stu_entry vlan; + int err; + + err = _mv88e6xxx_vtu_get(ps, vid, &vlan, true); + if (err) + return err; + + vlan.data[port] = untagged ? + GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED : + GLOBAL_VTU_DATA_MEMBER_TAG_TAGGED; + + return _mv88e6xxx_vtu_loadpurge(ps, &vlan); +} + +static void mv88e6xxx_port_vlan_add(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; + bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; + u16 vid; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) + return; + + mutex_lock(&ps->reg_lock); + + for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) + if (_mv88e6xxx_port_vlan_add(ps, port, vid, untagged)) + netdev_err(ds->ports[port].netdev, + "failed to add VLAN %d%c\n", + vid, untagged ? 'u' : 't'); + + if (pvid && _mv88e6xxx_port_pvid_set(ps, port, vlan->vid_end)) + netdev_err(ds->ports[port].netdev, "failed to set PVID %d\n", + vlan->vid_end); + + mutex_unlock(&ps->reg_lock); +} + +static int _mv88e6xxx_port_vlan_del(struct mv88e6xxx_priv_state *ps, + int port, u16 vid) +{ + struct dsa_switch *ds = ps->ds; + struct mv88e6xxx_vtu_stu_entry vlan; + int i, err; + + err = _mv88e6xxx_vtu_get(ps, vid, &vlan, false); + if (err) + return err; + + /* Tell switchdev if this VLAN is handled in software */ + if (vlan.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) + return -EOPNOTSUPP; + + vlan.data[port] = GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER; + + /* keep the VLAN unless all ports are excluded */ + vlan.valid = false; + for (i = 0; i < ps->info->num_ports; ++i) { + if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i)) + continue; + + if (vlan.data[i] != GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) { + vlan.valid = true; + break; + } + } + + err = _mv88e6xxx_vtu_loadpurge(ps, &vlan); + if (err) + return err; + + return _mv88e6xxx_atu_remove(ps, vlan.fid, port, false); +} + +static int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + u16 pvid, vid; + int err = 0; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) + return -EOPNOTSUPP; + + mutex_lock(&ps->reg_lock); + + err = _mv88e6xxx_port_pvid_get(ps, port, &pvid); + if (err) + goto unlock; + + for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) { + err = _mv88e6xxx_port_vlan_del(ps, port, vid); + if (err) + goto unlock; + + if (vid == pvid) { + err = _mv88e6xxx_port_pvid_set(ps, port, 0); + if (err) + goto unlock; + } + } + +unlock: + mutex_unlock(&ps->reg_lock); + + return err; +} + +static int _mv88e6xxx_atu_mac_write(struct mv88e6xxx_priv_state *ps, + const unsigned char *addr) +{ + int i, ret; + + for (i = 0; i < 3; i++) { + ret = _mv88e6xxx_reg_write( + ps, REG_GLOBAL, GLOBAL_ATU_MAC_01 + i, + (addr[i * 2] << 8) | addr[i * 2 + 1]); + if (ret < 0) + return ret; + } + + return 0; +} + +static int _mv88e6xxx_atu_mac_read(struct mv88e6xxx_priv_state *ps, + unsigned char *addr) +{ + int i, ret; + + for (i = 0; i < 3; i++) { + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, + GLOBAL_ATU_MAC_01 + i); + if (ret < 0) + return ret; + addr[i * 2] = ret >> 8; + addr[i * 2 + 1] = ret & 0xff; + } + + return 0; +} + +static int _mv88e6xxx_atu_load(struct mv88e6xxx_priv_state *ps, + struct mv88e6xxx_atu_entry *entry) +{ + int ret; + + ret = _mv88e6xxx_atu_wait(ps); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_atu_mac_write(ps, entry->mac); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_atu_data_write(ps, entry); + if (ret < 0) + return ret; + + return _mv88e6xxx_atu_cmd(ps, entry->fid, GLOBAL_ATU_OP_LOAD_DB); +} + +static int _mv88e6xxx_port_fdb_load(struct mv88e6xxx_priv_state *ps, int port, + const unsigned char *addr, u16 vid, + u8 state) +{ + struct mv88e6xxx_atu_entry entry = { 0 }; + struct mv88e6xxx_vtu_stu_entry vlan; + int err; + + /* Null VLAN ID corresponds to the port private database */ + if (vid == 0) + err = _mv88e6xxx_port_fid_get(ps, port, &vlan.fid); + else + err = _mv88e6xxx_vtu_get(ps, vid, &vlan, false); + if (err) + return err; + + entry.fid = vlan.fid; + entry.state = state; + ether_addr_copy(entry.mac, addr); + if (state != GLOBAL_ATU_DATA_STATE_UNUSED) { + entry.trunk = false; + entry.portv_trunkid = BIT(port); + } + + return _mv88e6xxx_atu_load(ps, &entry); +} + +static int mv88e6xxx_port_fdb_prepare(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_fdb *fdb, + struct switchdev_trans *trans) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) + return -EOPNOTSUPP; + + /* We don't need any dynamic resource from the kernel (yet), + * so skip the prepare phase. + */ + return 0; +} + +static void mv88e6xxx_port_fdb_add(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_fdb *fdb, + struct switchdev_trans *trans) +{ + int state = is_multicast_ether_addr(fdb->addr) ? + GLOBAL_ATU_DATA_STATE_MC_STATIC : + GLOBAL_ATU_DATA_STATE_UC_STATIC; + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) + return; + + mutex_lock(&ps->reg_lock); + if (_mv88e6xxx_port_fdb_load(ps, port, fdb->addr, fdb->vid, state)) + netdev_err(ds->ports[port].netdev, + "failed to load MAC address\n"); + mutex_unlock(&ps->reg_lock); +} + +static int mv88e6xxx_port_fdb_del(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_fdb *fdb) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int ret; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) + return -EOPNOTSUPP; + + mutex_lock(&ps->reg_lock); + ret = _mv88e6xxx_port_fdb_load(ps, port, fdb->addr, fdb->vid, + GLOBAL_ATU_DATA_STATE_UNUSED); + mutex_unlock(&ps->reg_lock); + + return ret; +} + +static int _mv88e6xxx_atu_getnext(struct mv88e6xxx_priv_state *ps, u16 fid, + struct mv88e6xxx_atu_entry *entry) +{ + struct mv88e6xxx_atu_entry next = { 0 }; + int ret; + + next.fid = fid; + + ret = _mv88e6xxx_atu_wait(ps); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_atu_cmd(ps, fid, GLOBAL_ATU_OP_GET_NEXT_DB); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_atu_mac_read(ps, next.mac); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_ATU_DATA); + if (ret < 0) + return ret; + + next.state = ret & GLOBAL_ATU_DATA_STATE_MASK; + if (next.state != GLOBAL_ATU_DATA_STATE_UNUSED) { + unsigned int mask, shift; + + if (ret & GLOBAL_ATU_DATA_TRUNK) { + next.trunk = true; + mask = GLOBAL_ATU_DATA_TRUNK_ID_MASK; + shift = GLOBAL_ATU_DATA_TRUNK_ID_SHIFT; + } else { + next.trunk = false; + mask = GLOBAL_ATU_DATA_PORT_VECTOR_MASK; + shift = GLOBAL_ATU_DATA_PORT_VECTOR_SHIFT; + } + + next.portv_trunkid = (ret & mask) >> shift; + } + + *entry = next; + return 0; +} + +static int _mv88e6xxx_port_fdb_dump_one(struct mv88e6xxx_priv_state *ps, + u16 fid, u16 vid, int port, + struct switchdev_obj_port_fdb *fdb, + int (*cb)(struct switchdev_obj *obj)) +{ + struct mv88e6xxx_atu_entry addr = { + .mac = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }, + }; + int err; + + err = _mv88e6xxx_atu_mac_write(ps, addr.mac); + if (err) + return err; + + do { + err = _mv88e6xxx_atu_getnext(ps, fid, &addr); + if (err) + break; + + if (addr.state == GLOBAL_ATU_DATA_STATE_UNUSED) + break; + + if (!addr.trunk && addr.portv_trunkid & BIT(port)) { + bool is_static = addr.state == + (is_multicast_ether_addr(addr.mac) ? + GLOBAL_ATU_DATA_STATE_MC_STATIC : + GLOBAL_ATU_DATA_STATE_UC_STATIC); + + fdb->vid = vid; + ether_addr_copy(fdb->addr, addr.mac); + fdb->ndm_state = is_static ? NUD_NOARP : NUD_REACHABLE; + + err = cb(&fdb->obj); + if (err) + break; + } + } while (!is_broadcast_ether_addr(addr.mac)); + + return err; +} + +static int mv88e6xxx_port_fdb_dump(struct dsa_switch *ds, int port, + struct switchdev_obj_port_fdb *fdb, + int (*cb)(struct switchdev_obj *obj)) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + struct mv88e6xxx_vtu_stu_entry vlan = { + .vid = GLOBAL_VTU_VID_MASK, /* all ones */ + }; + u16 fid; + int err; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) + return -EOPNOTSUPP; + + mutex_lock(&ps->reg_lock); + + /* Dump port's default Filtering Information Database (VLAN ID 0) */ + err = _mv88e6xxx_port_fid_get(ps, port, &fid); + if (err) + goto unlock; + + err = _mv88e6xxx_port_fdb_dump_one(ps, fid, 0, port, fdb, cb); + if (err) + goto unlock; + + /* Dump VLANs' Filtering Information Databases */ + err = _mv88e6xxx_vtu_vid_write(ps, vlan.vid); + if (err) + goto unlock; + + do { + err = _mv88e6xxx_vtu_getnext(ps, &vlan); + if (err) + break; + + if (!vlan.valid) + break; + + err = _mv88e6xxx_port_fdb_dump_one(ps, vlan.fid, vlan.vid, port, + fdb, cb); + if (err) + break; + } while (vlan.vid < GLOBAL_VTU_VID_MASK); + +unlock: + mutex_unlock(&ps->reg_lock); + + return err; +} + +static int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port, + struct net_device *bridge) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int i, err = 0; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VLANTABLE)) + return -EOPNOTSUPP; + + mutex_lock(&ps->reg_lock); + + /* Assign the bridge and remap each port's VLANTable */ + ps->ports[port].bridge_dev = bridge; + + for (i = 0; i < ps->info->num_ports; ++i) { + if (ps->ports[i].bridge_dev == bridge) { + err = _mv88e6xxx_port_based_vlan_map(ps, i); + if (err) + break; + } + } + + mutex_unlock(&ps->reg_lock); + + return err; +} + +static void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + struct net_device *bridge = ps->ports[port].bridge_dev; + int i; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VLANTABLE)) + return; + + mutex_lock(&ps->reg_lock); + + /* Unassign the bridge and remap each port's VLANTable */ + ps->ports[port].bridge_dev = NULL; + + for (i = 0; i < ps->info->num_ports; ++i) + if (i == port || ps->ports[i].bridge_dev == bridge) + if (_mv88e6xxx_port_based_vlan_map(ps, i)) + netdev_warn(ds->ports[i].netdev, + "failed to remap\n"); + + mutex_unlock(&ps->reg_lock); +} + +static int _mv88e6xxx_mdio_page_write(struct mv88e6xxx_priv_state *ps, + int port, int page, int reg, int val) +{ + int ret; + + ret = mv88e6xxx_mdio_write_indirect(ps, port, 0x16, page); + if (ret < 0) + goto restore_page_0; + + ret = mv88e6xxx_mdio_write_indirect(ps, port, reg, val); +restore_page_0: + mv88e6xxx_mdio_write_indirect(ps, port, 0x16, 0x0); + + return ret; +} + +static int _mv88e6xxx_mdio_page_read(struct mv88e6xxx_priv_state *ps, + int port, int page, int reg) +{ + int ret; + + ret = mv88e6xxx_mdio_write_indirect(ps, port, 0x16, page); + if (ret < 0) + goto restore_page_0; + + ret = mv88e6xxx_mdio_read_indirect(ps, port, reg); +restore_page_0: + mv88e6xxx_mdio_write_indirect(ps, port, 0x16, 0x0); + + return ret; +} + +static int mv88e6xxx_switch_reset(struct mv88e6xxx_priv_state *ps) +{ + bool ppu_active = mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU_ACTIVE); + u16 is_reset = (ppu_active ? 0x8800 : 0xc800); + struct gpio_desc *gpiod = ps->reset; + unsigned long timeout; + int ret; + int i; + + /* Set all ports to the disabled state. */ + for (i = 0; i < ps->info->num_ports; i++) { + ret = _mv88e6xxx_reg_read(ps, REG_PORT(i), PORT_CONTROL); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_write(ps, REG_PORT(i), PORT_CONTROL, + ret & 0xfffc); + if (ret) + return ret; + } + + /* Wait for transmit queues to drain. */ + usleep_range(2000, 4000); + + /* If there is a gpio connected to the reset pin, toggle it */ + if (gpiod) { + gpiod_set_value_cansleep(gpiod, 1); + usleep_range(10000, 20000); + gpiod_set_value_cansleep(gpiod, 0); + usleep_range(10000, 20000); + } + + /* Reset the switch. Keep the PPU active if requested. The PPU + * needs to be active to support indirect phy register access + * through global registers 0x18 and 0x19. + */ + if (ppu_active) + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, 0x04, 0xc000); + else + ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, 0x04, 0xc400); + if (ret) + return ret; + + /* Wait up to one second for reset to complete. */ + timeout = jiffies + 1 * HZ; + while (time_before(jiffies, timeout)) { + ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, 0x00); + if (ret < 0) + return ret; + + if ((ret & is_reset) == is_reset) + break; + usleep_range(1000, 2000); + } + if (time_after(jiffies, timeout)) + ret = -ETIMEDOUT; + else + ret = 0; + + return ret; +} + +static int mv88e6xxx_power_on_serdes(struct mv88e6xxx_priv_state *ps) +{ + int ret; + + ret = _mv88e6xxx_mdio_page_read(ps, REG_FIBER_SERDES, + PAGE_FIBER_SERDES, MII_BMCR); + if (ret < 0) + return ret; + + if (ret & BMCR_PDOWN) { + ret &= ~BMCR_PDOWN; + ret = _mv88e6xxx_mdio_page_write(ps, REG_FIBER_SERDES, + PAGE_FIBER_SERDES, MII_BMCR, + ret); + } + + return ret; +} + +static int mv88e6xxx_setup_port(struct mv88e6xxx_priv_state *ps, int port) +{ + struct dsa_switch *ds = ps->ds; + int ret; + u16 reg; + + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || + mv88e6xxx_6185_family(ps) || mv88e6xxx_6095_family(ps) || + mv88e6xxx_6065_family(ps) || mv88e6xxx_6320_family(ps)) { + /* MAC Forcing register: don't force link, speed, + * duplex or flow control state to any particular + * values on physical ports, but force the CPU port + * and all DSA ports to their maximum bandwidth and + * full duplex. + */ + reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_PCS_CTRL); + if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) { + reg &= ~PORT_PCS_CTRL_UNFORCED; + reg |= PORT_PCS_CTRL_FORCE_LINK | + PORT_PCS_CTRL_LINK_UP | + PORT_PCS_CTRL_DUPLEX_FULL | + PORT_PCS_CTRL_FORCE_DUPLEX; + if (mv88e6xxx_6065_family(ps)) + reg |= PORT_PCS_CTRL_100; + else + reg |= PORT_PCS_CTRL_1000; + } else { + reg |= PORT_PCS_CTRL_UNFORCED; + } + + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_PCS_CTRL, reg); + if (ret) + return ret; + } + + /* Port Control: disable Drop-on-Unlock, disable Drop-on-Lock, + * disable Header mode, enable IGMP/MLD snooping, disable VLAN + * tunneling, determine priority by looking at 802.1p and IP + * priority fields (IP prio has precedence), and set STP state + * to Forwarding. + * + * If this is the CPU link, use DSA or EDSA tagging depending + * on which tagging mode was configured. + * + * If this is a link to another switch, use DSA tagging mode. + * + * If this is the upstream port for this switch, enable + * forwarding of unknown unicasts and multicasts. + */ + reg = 0; + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || + mv88e6xxx_6095_family(ps) || mv88e6xxx_6065_family(ps) || + mv88e6xxx_6185_family(ps) || mv88e6xxx_6320_family(ps)) + reg = PORT_CONTROL_IGMP_MLD_SNOOP | + PORT_CONTROL_USE_TAG | PORT_CONTROL_USE_IP | + PORT_CONTROL_STATE_FORWARDING; + if (dsa_is_cpu_port(ds, port)) { + if (mv88e6xxx_6095_family(ps) || mv88e6xxx_6185_family(ps)) + reg |= PORT_CONTROL_DSA_TAG; + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || + mv88e6xxx_6320_family(ps)) { + reg |= PORT_CONTROL_FRAME_ETHER_TYPE_DSA | + PORT_CONTROL_FORWARD_UNKNOWN | + PORT_CONTROL_FORWARD_UNKNOWN_MC; + } + + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || + mv88e6xxx_6095_family(ps) || mv88e6xxx_6065_family(ps) || + mv88e6xxx_6185_family(ps) || mv88e6xxx_6320_family(ps)) { + reg |= PORT_CONTROL_EGRESS_ADD_TAG; + } + } + if (dsa_is_dsa_port(ds, port)) { + if (mv88e6xxx_6095_family(ps) || mv88e6xxx_6185_family(ps)) + reg |= PORT_CONTROL_DSA_TAG; + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || + mv88e6xxx_6320_family(ps)) { + reg |= PORT_CONTROL_FRAME_MODE_DSA; + } + + if (port == dsa_upstream_port(ds)) + reg |= PORT_CONTROL_FORWARD_UNKNOWN | + PORT_CONTROL_FORWARD_UNKNOWN_MC; + } + if (reg) { + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_CONTROL, reg); + if (ret) + return ret; + } + + /* If this port is connected to a SerDes, make sure the SerDes is not + * powered down. + */ + if (mv88e6xxx_6352_family(ps)) { + ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_STATUS); + if (ret < 0) + return ret; + ret &= PORT_STATUS_CMODE_MASK; + if ((ret == PORT_STATUS_CMODE_100BASE_X) || + (ret == PORT_STATUS_CMODE_1000BASE_X) || + (ret == PORT_STATUS_CMODE_SGMII)) { + ret = mv88e6xxx_power_on_serdes(ps); + if (ret < 0) + return ret; + } + } + + /* Port Control 2: don't force a good FCS, set the maximum frame size to + * 10240 bytes, disable 802.1q tags checking, don't discard tagged or + * untagged frames on this port, do a destination address lookup on all + * received packets as usual, disable ARP mirroring and don't send a + * copy of all transmitted/received frames on this port to the CPU. + */ + reg = 0; + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || + mv88e6xxx_6095_family(ps) || mv88e6xxx_6320_family(ps) || + mv88e6xxx_6185_family(ps)) + reg = PORT_CONTROL_2_MAP_DA; + + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6320_family(ps)) + reg |= PORT_CONTROL_2_JUMBO_10240; + + if (mv88e6xxx_6095_family(ps) || mv88e6xxx_6185_family(ps)) { + /* Set the upstream port this port should use */ + reg |= dsa_upstream_port(ds); + /* enable forwarding of unknown multicast addresses to + * the upstream port + */ + if (port == dsa_upstream_port(ds)) + reg |= PORT_CONTROL_2_FORWARD_UNKNOWN; + } + + reg |= PORT_CONTROL_2_8021Q_DISABLED; + + if (reg) { + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_CONTROL_2, reg); + if (ret) + return ret; + } + + /* Port Association Vector: when learning source addresses + * of packets, add the address to the address database using + * a port bitmap that has only the bit for this port set and + * the other bits clear. + */ + reg = 1 << port; + /* Disable learning for CPU port */ + if (dsa_is_cpu_port(ds, port)) + reg = 0; + + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_ASSOC_VECTOR, reg); + if (ret) + return ret; + + /* Egress rate control 2: disable egress rate control. */ + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_RATE_CONTROL_2, + 0x0000); + if (ret) + return ret; + + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || + mv88e6xxx_6320_family(ps)) { + /* Do not limit the period of time that this port can + * be paused for by the remote end or the period of + * time that this port can pause the remote end. + */ + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_PAUSE_CTRL, 0x0000); + if (ret) + return ret; + + /* Port ATU control: disable limiting the number of + * address database entries that this port is allowed + * to use. + */ + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_ATU_CONTROL, 0x0000); + /* Priority Override: disable DA, SA and VTU priority + * override. + */ + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_PRI_OVERRIDE, 0x0000); + if (ret) + return ret; + + /* Port Ethertype: use the Ethertype DSA Ethertype + * value. + */ + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_ETH_TYPE, ETH_P_EDSA); + if (ret) + return ret; + /* Tag Remap: use an identity 802.1p prio -> switch + * prio mapping. + */ + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_TAG_REGMAP_0123, 0x3210); + if (ret) + return ret; + + /* Tag Remap 2: use an identity 802.1p prio -> switch + * prio mapping. + */ + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_TAG_REGMAP_4567, 0x7654); + if (ret) + return ret; + } + + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || + mv88e6xxx_6185_family(ps) || mv88e6xxx_6095_family(ps) || + mv88e6xxx_6320_family(ps)) { + /* Rate Control: disable ingress rate limiting. */ + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), + PORT_RATE_CONTROL, 0x0001); + if (ret) + return ret; + } + + /* Port Control 1: disable trunking, disable sending + * learning messages to this port. + */ + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL_1, 0x0000); + if (ret) + return ret; + + /* Port based VLAN map: give each port the same default address + * database, and allow bidirectional communication between the + * CPU and DSA port(s), and the other ports. + */ + ret = _mv88e6xxx_port_fid_set(ps, port, 0); + if (ret) + return ret; + + ret = _mv88e6xxx_port_based_vlan_map(ps, port); + if (ret) + return ret; + + /* Default VLAN ID and priority: don't set a default VLAN + * ID, and set the default packet priority to zero. + */ + ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_DEFAULT_VLAN, + 0x0000); + if (ret) + return ret; + + return 0; +} + +static int mv88e6xxx_setup_global(struct mv88e6xxx_priv_state *ps) +{ + struct dsa_switch *ds = ps->ds; + u32 upstream_port = dsa_upstream_port(ds); + u16 reg; + int err; + int i; + + /* Enable the PHY Polling Unit if present, don't discard any packets, + * and mask all interrupt sources. + */ + reg = 0; + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU) || + mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU_ACTIVE)) + reg |= GLOBAL_CONTROL_PPU_ENABLE; + + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, reg); + if (err) + return err; + + /* Configure the upstream port, and configure it as the port to which + * ingress and egress and ARP monitor frames are to be sent. + */ + reg = upstream_port << GLOBAL_MONITOR_CONTROL_INGRESS_SHIFT | + upstream_port << GLOBAL_MONITOR_CONTROL_EGRESS_SHIFT | + upstream_port << GLOBAL_MONITOR_CONTROL_ARP_SHIFT; + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MONITOR_CONTROL, reg); + if (err) + return err; + + /* Disable remote management, and set the switch's DSA device number. */ + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL_2, + GLOBAL_CONTROL_2_MULTIPLE_CASCADE | + (ds->index & 0x1f)); + if (err) + return err; + + /* Set the default address aging time to 5 minutes, and + * enable address learn messages to be sent to all message + * ports. + */ + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_CONTROL, + 0x0140 | GLOBAL_ATU_CONTROL_LEARN2ALL); + if (err) + return err; + + /* Configure the IP ToS mapping registers. */ + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_0, 0x0000); + if (err) + return err; + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_1, 0x0000); + if (err) + return err; + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_2, 0x5555); + if (err) + return err; + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_3, 0x5555); + if (err) + return err; + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_4, 0xaaaa); + if (err) + return err; + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_5, 0xaaaa); + if (err) + return err; + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_6, 0xffff); + if (err) + return err; + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_7, 0xffff); + if (err) + return err; + + /* Configure the IEEE 802.1p priority mapping register. */ + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IEEE_PRI, 0xfa41); + if (err) + return err; + + /* Send all frames with destination addresses matching + * 01:80:c2:00:00:0x to the CPU port. + */ + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_MGMT_EN_0X, 0xffff); + if (err) + return err; + + /* Ignore removed tag data on doubly tagged packets, disable + * flow control messages, force flow control priority to the + * highest, and send all special multicast frames to the CPU + * port at the highest priority. + */ + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SWITCH_MGMT, + 0x7 | GLOBAL2_SWITCH_MGMT_RSVD2CPU | 0x70 | + GLOBAL2_SWITCH_MGMT_FORCE_FLOW_CTRL_PRI); + if (err) + return err; + + /* Program the DSA routing table. */ + for (i = 0; i < 32; i++) { + int nexthop = 0x1f; + + if (i != ds->index && i < DSA_MAX_SWITCHES) + nexthop = ds->rtable[i] & 0x1f; + + err = _mv88e6xxx_reg_write( + ps, REG_GLOBAL2, + GLOBAL2_DEVICE_MAPPING, + GLOBAL2_DEVICE_MAPPING_UPDATE | + (i << GLOBAL2_DEVICE_MAPPING_TARGET_SHIFT) | nexthop); + if (err) + return err; + } + + /* Clear all trunk masks. */ + for (i = 0; i < 8; i++) { + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_TRUNK_MASK, + 0x8000 | + (i << GLOBAL2_TRUNK_MASK_NUM_SHIFT) | + ((1 << ps->info->num_ports) - 1)); + if (err) + return err; + } + + /* Clear all trunk mappings. */ + for (i = 0; i < 16; i++) { + err = _mv88e6xxx_reg_write( + ps, REG_GLOBAL2, + GLOBAL2_TRUNK_MAPPING, + GLOBAL2_TRUNK_MAPPING_UPDATE | + (i << GLOBAL2_TRUNK_MAPPING_ID_SHIFT)); + if (err) + return err; + } + + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || + mv88e6xxx_6320_family(ps)) { + /* Send all frames with destination addresses matching + * 01:80:c2:00:00:2x to the CPU port. + */ + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, + GLOBAL2_MGMT_EN_2X, 0xffff); + if (err) + return err; + + /* Initialise cross-chip port VLAN table to reset + * defaults. + */ + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, + GLOBAL2_PVT_ADDR, 0x9000); + if (err) + return err; + + /* Clear the priority override table. */ + for (i = 0; i < 16; i++) { + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, + GLOBAL2_PRIO_OVERRIDE, + 0x8000 | (i << 8)); + if (err) + return err; + } + } + + if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || + mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || + mv88e6xxx_6185_family(ps) || mv88e6xxx_6095_family(ps) || + mv88e6xxx_6320_family(ps)) { + /* Disable ingress rate limiting by resetting all + * ingress rate limit registers to their initial + * state. + */ + for (i = 0; i < ps->info->num_ports; i++) { + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, + GLOBAL2_INGRESS_OP, + 0x9000 | (i << 8)); + if (err) + return err; + } + } + + /* Clear the statistics counters for all ports */ + err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_STATS_OP, + GLOBAL_STATS_OP_FLUSH_ALL); + if (err) + return err; + + /* Wait for the flush to complete. */ + err = _mv88e6xxx_stats_wait(ps); + if (err) + return err; + + /* Clear all ATU entries */ + err = _mv88e6xxx_atu_flush(ps, 0, true); + if (err) + return err; + + /* Clear all the VTU and STU entries */ + err = _mv88e6xxx_vtu_stu_flush(ps); + if (err < 0) + return err; + + return err; +} + +static int mv88e6xxx_setup(struct dsa_switch *ds) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int err; + int i; + + ps->ds = ds; + ds->slave_mii_bus = ps->mdio_bus; + + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) + mutex_init(&ps->eeprom_mutex); + + mutex_lock(&ps->reg_lock); + + err = mv88e6xxx_switch_reset(ps); + if (err) + goto unlock; + + err = mv88e6xxx_setup_global(ps); + if (err) + goto unlock; + + for (i = 0; i < ps->info->num_ports; i++) { + err = mv88e6xxx_setup_port(ps, i); + if (err) + goto unlock; + } + +unlock: + mutex_unlock(&ps->reg_lock); + + return err; +} + +static int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, + int reg) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int ret; + + mutex_lock(&ps->reg_lock); + ret = _mv88e6xxx_mdio_page_read(ps, port, page, reg); + mutex_unlock(&ps->reg_lock); + + return ret; +} + +static int mv88e6xxx_mdio_page_write(struct dsa_switch *ds, int port, int page, + int reg, int val) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int ret; + + mutex_lock(&ps->reg_lock); + ret = _mv88e6xxx_mdio_page_write(ps, port, page, reg, val); + mutex_unlock(&ps->reg_lock); + + return ret; +} + +static int mv88e6xxx_port_to_mdio_addr(struct mv88e6xxx_priv_state *ps, + int port) +{ + if (port >= 0 && port < ps->info->num_ports) + return port; + return -EINVAL; +} + +static int mv88e6xxx_mdio_read(struct mii_bus *bus, int port, int regnum) +{ + struct mv88e6xxx_priv_state *ps = bus->priv; + int addr = mv88e6xxx_port_to_mdio_addr(ps, port); + int ret; + + if (addr < 0) + return 0xffff; + + mutex_lock(&ps->reg_lock); + + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) + ret = mv88e6xxx_mdio_read_ppu(ps, addr, regnum); + else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SMI_PHY)) + ret = mv88e6xxx_mdio_read_indirect(ps, addr, regnum); + else + ret = mv88e6xxx_mdio_read_direct(ps, addr, regnum); + + mutex_unlock(&ps->reg_lock); + return ret; +} + +static int mv88e6xxx_mdio_write(struct mii_bus *bus, int port, int regnum, + u16 val) +{ + struct mv88e6xxx_priv_state *ps = bus->priv; + int addr = mv88e6xxx_port_to_mdio_addr(ps, port); + int ret; + + if (addr < 0) + return 0xffff; + + mutex_lock(&ps->reg_lock); + + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) + ret = mv88e6xxx_mdio_write_ppu(ps, addr, regnum, val); + else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SMI_PHY)) + ret = mv88e6xxx_mdio_write_indirect(ps, addr, regnum, val); + else + ret = mv88e6xxx_mdio_write_direct(ps, addr, regnum, val); + + mutex_unlock(&ps->reg_lock); + return ret; +} + +static int mv88e6xxx_mdio_register(struct mv88e6xxx_priv_state *ps, + struct device_node *np) +{ + static int index; + struct mii_bus *bus; + int err; + + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) + mv88e6xxx_ppu_state_init(ps); + + if (np) + ps->mdio_np = of_get_child_by_name(np, "mdio"); + + bus = devm_mdiobus_alloc(ps->dev); + if (!bus) + return -ENOMEM; + + bus->priv = (void *)ps; + if (np) { + bus->name = np->full_name; + snprintf(bus->id, MII_BUS_ID_SIZE, "%s", np->full_name); + } else { + bus->name = "mv88e6xxx SMI"; + snprintf(bus->id, MII_BUS_ID_SIZE, "mv88e6xxx-%d", index++); + } + + bus->read = mv88e6xxx_mdio_read; + bus->write = mv88e6xxx_mdio_write; + bus->parent = ps->dev; + + if (ps->mdio_np) + err = of_mdiobus_register(bus, ps->mdio_np); + else + err = mdiobus_register(bus); + if (err) { + dev_err(ps->dev, "Cannot register MDIO bus (%d)\n", err); + goto out; + } + ps->mdio_bus = bus; + + return 0; + +out: + if (ps->mdio_np) + of_node_put(ps->mdio_np); + + return err; +} + +static void mv88e6xxx_mdio_unregister(struct mv88e6xxx_priv_state *ps) + +{ + struct mii_bus *bus = ps->mdio_bus; + + mdiobus_unregister(bus); + + if (ps->mdio_np) + of_node_put(ps->mdio_np); +} + +#ifdef CONFIG_NET_DSA_HWMON + +static int mv88e61xx_get_temp(struct dsa_switch *ds, int *temp) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int ret; + int val; + + *temp = 0; + + mutex_lock(&ps->reg_lock); + + ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x16, 0x6); + if (ret < 0) + goto error; + + /* Enable temperature sensor */ + ret = mv88e6xxx_mdio_read_direct(ps, 0x0, 0x1a); + if (ret < 0) + goto error; + + ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x1a, ret | (1 << 5)); + if (ret < 0) + goto error; + + /* Wait for temperature to stabilize */ + usleep_range(10000, 12000); + + val = mv88e6xxx_mdio_read_direct(ps, 0x0, 0x1a); + if (val < 0) { + ret = val; + goto error; + } + + /* Disable temperature sensor */ + ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x1a, ret & ~(1 << 5)); + if (ret < 0) + goto error; + + *temp = ((val & 0x1f) - 5) * 5; + +error: + mv88e6xxx_mdio_write_direct(ps, 0x0, 0x16, 0x0); + mutex_unlock(&ps->reg_lock); + return ret; +} + +static int mv88e63xx_get_temp(struct dsa_switch *ds, int *temp) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; + int ret; + + *temp = 0; + + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 27); + if (ret < 0) + return ret; + + *temp = (ret & 0xff) - 25; + + return 0; +} + +static int mv88e6xxx_get_temp(struct dsa_switch *ds, int *temp) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP)) + return -EOPNOTSUPP; + + if (mv88e6xxx_6320_family(ps) || mv88e6xxx_6352_family(ps)) + return mv88e63xx_get_temp(ds, temp); + + return mv88e61xx_get_temp(ds, temp); +} + +static int mv88e6xxx_get_temp_limit(struct dsa_switch *ds, int *temp) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; + int ret; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP_LIMIT)) + return -EOPNOTSUPP; + + *temp = 0; + + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); + if (ret < 0) + return ret; + + *temp = (((ret >> 8) & 0x1f) * 5) - 25; + + return 0; +} + +static int mv88e6xxx_set_temp_limit(struct dsa_switch *ds, int temp) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; + int ret; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP_LIMIT)) + return -EOPNOTSUPP; + + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); + if (ret < 0) + return ret; + temp = clamp_val(DIV_ROUND_CLOSEST(temp, 5) + 5, 0, 0x1f); + return mv88e6xxx_mdio_page_write(ds, phy, 6, 26, + (ret & 0xe0ff) | (temp << 8)); +} + +static int mv88e6xxx_get_temp_alarm(struct dsa_switch *ds, bool *alarm) +{ + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; + int ret; + + if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP_LIMIT)) + return -EOPNOTSUPP; + + *alarm = false; + + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); + if (ret < 0) + return ret; + + *alarm = !!(ret & 0x40); + + return 0; +} +#endif /* CONFIG_NET_DSA_HWMON */ + +static const struct mv88e6xxx_info mv88e6xxx_table[] = { + [MV88E6085] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6085, + .family = MV88E6XXX_FAMILY_6097, + .name = "Marvell 88E6085", + .num_databases = 4096, + .num_ports = 10, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6097, + }, + + [MV88E6095] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6095, + .family = MV88E6XXX_FAMILY_6095, + .name = "Marvell 88E6095/88E6095F", + .num_databases = 256, + .num_ports = 11, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6095, + }, + + [MV88E6123] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6123, + .family = MV88E6XXX_FAMILY_6165, + .name = "Marvell 88E6123", + .num_databases = 4096, + .num_ports = 3, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6165, + }, + + [MV88E6131] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6131, + .family = MV88E6XXX_FAMILY_6185, + .name = "Marvell 88E6131", + .num_databases = 256, + .num_ports = 8, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6185, + }, + + [MV88E6161] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6161, + .family = MV88E6XXX_FAMILY_6165, + .name = "Marvell 88E6161", + .num_databases = 4096, + .num_ports = 6, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6165, + }, + + [MV88E6165] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6165, + .family = MV88E6XXX_FAMILY_6165, + .name = "Marvell 88E6165", + .num_databases = 4096, + .num_ports = 6, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6165, + }, + + [MV88E6171] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6171, + .family = MV88E6XXX_FAMILY_6351, + .name = "Marvell 88E6171", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6351, + }, + + [MV88E6172] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6172, + .family = MV88E6XXX_FAMILY_6352, + .name = "Marvell 88E6172", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6352, + }, + + [MV88E6175] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6175, + .family = MV88E6XXX_FAMILY_6351, + .name = "Marvell 88E6175", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6351, + }, + + [MV88E6176] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6176, + .family = MV88E6XXX_FAMILY_6352, + .name = "Marvell 88E6176", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6352, + }, + + [MV88E6185] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6185, + .family = MV88E6XXX_FAMILY_6185, + .name = "Marvell 88E6185", + .num_databases = 256, + .num_ports = 10, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6185, + }, + + [MV88E6240] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6240, + .family = MV88E6XXX_FAMILY_6352, + .name = "Marvell 88E6240", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6352, + }, + + [MV88E6320] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6320, + .family = MV88E6XXX_FAMILY_6320, + .name = "Marvell 88E6320", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6320, + }, + + [MV88E6321] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6321, + .family = MV88E6XXX_FAMILY_6320, + .name = "Marvell 88E6321", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6320, + }, + + [MV88E6350] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6350, + .family = MV88E6XXX_FAMILY_6351, + .name = "Marvell 88E6350", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6351, + }, + + [MV88E6351] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6351, + .family = MV88E6XXX_FAMILY_6351, + .name = "Marvell 88E6351", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6351, + }, + + [MV88E6352] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6352, + .family = MV88E6XXX_FAMILY_6352, + .name = "Marvell 88E6352", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6352, + }, +}; + +static const struct mv88e6xxx_info *mv88e6xxx_lookup_info(unsigned int prod_num) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(mv88e6xxx_table); ++i) + if (mv88e6xxx_table[i].prod_num == prod_num) + return &mv88e6xxx_table[i]; + + return NULL; +} + +static int mv88e6xxx_detect(struct mv88e6xxx_priv_state *ps) +{ + const struct mv88e6xxx_info *info; + int id, prod_num, rev; + + id = mv88e6xxx_reg_read(ps, ps->info->port_base_addr, PORT_SWITCH_ID); + if (id < 0) + return id; + + prod_num = (id & 0xfff0) >> 4; + rev = id & 0x000f; + + info = mv88e6xxx_lookup_info(prod_num); + if (!info) + return -ENODEV; + + /* Update the compatible info with the probed one */ + ps->info = info; + + dev_info(ps->dev, "switch 0x%x detected: %s, revision %u\n", + ps->info->prod_num, ps->info->name, rev); + + return 0; +} + +static struct mv88e6xxx_priv_state *mv88e6xxx_alloc_chip(struct device *dev) +{ + struct mv88e6xxx_priv_state *ps; + + ps = devm_kzalloc(dev, sizeof(*ps), GFP_KERNEL); + if (!ps) + return NULL; + + ps->dev = dev; + + mutex_init(&ps->reg_lock); + + return ps; +} + +static int mv88e6xxx_smi_init(struct mv88e6xxx_priv_state *ps, + struct mii_bus *bus, int sw_addr) +{ + /* ADDR[0] pin is unavailable externally and considered zero */ + if (sw_addr & 0x1) + return -EINVAL; + + if (sw_addr == 0) + ps->smi_ops = &mv88e6xxx_smi_single_chip_ops; + else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_MULTI_CHIP)) + ps->smi_ops = &mv88e6xxx_smi_multi_chip_ops; + else + return -EINVAL; + + ps->bus = bus; + ps->sw_addr = sw_addr; + + return 0; +} + +static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, + struct device *host_dev, int sw_addr, + void **priv) +{ + struct mv88e6xxx_priv_state *ps; + struct mii_bus *bus; + int err; + + bus = dsa_host_dev_to_mii_bus(host_dev); + if (!bus) + return NULL; + + ps = mv88e6xxx_alloc_chip(dsa_dev); + if (!ps) + return NULL; + + /* Legacy SMI probing will only support chips similar to 88E6085 */ + ps->info = &mv88e6xxx_table[MV88E6085]; + + err = mv88e6xxx_smi_init(ps, bus, sw_addr); + if (err) + goto free; + + err = mv88e6xxx_detect(ps); + if (err) + goto free; + + err = mv88e6xxx_mdio_register(ps, NULL); + if (err) + goto free; + + *priv = ps; + + return ps->info->name; +free: + devm_kfree(dsa_dev, ps); + + return NULL; +} + +static struct dsa_switch_driver mv88e6xxx_switch_driver = { + .tag_protocol = DSA_TAG_PROTO_EDSA, + .probe = mv88e6xxx_drv_probe, + .setup = mv88e6xxx_setup, + .set_addr = mv88e6xxx_set_addr, + .adjust_link = mv88e6xxx_adjust_link, + .get_strings = mv88e6xxx_get_strings, + .get_ethtool_stats = mv88e6xxx_get_ethtool_stats, + .get_sset_count = mv88e6xxx_get_sset_count, + .set_eee = mv88e6xxx_set_eee, + .get_eee = mv88e6xxx_get_eee, +#ifdef CONFIG_NET_DSA_HWMON + .get_temp = mv88e6xxx_get_temp, + .get_temp_limit = mv88e6xxx_get_temp_limit, + .set_temp_limit = mv88e6xxx_set_temp_limit, + .get_temp_alarm = mv88e6xxx_get_temp_alarm, +#endif + .get_eeprom_len = mv88e6xxx_get_eeprom_len, + .get_eeprom = mv88e6xxx_get_eeprom, + .set_eeprom = mv88e6xxx_set_eeprom, + .get_regs_len = mv88e6xxx_get_regs_len, + .get_regs = mv88e6xxx_get_regs, + .port_bridge_join = mv88e6xxx_port_bridge_join, + .port_bridge_leave = mv88e6xxx_port_bridge_leave, + .port_stp_state_set = mv88e6xxx_port_stp_state_set, + .port_vlan_filtering = mv88e6xxx_port_vlan_filtering, + .port_vlan_prepare = mv88e6xxx_port_vlan_prepare, + .port_vlan_add = mv88e6xxx_port_vlan_add, + .port_vlan_del = mv88e6xxx_port_vlan_del, + .port_vlan_dump = mv88e6xxx_port_vlan_dump, + .port_fdb_prepare = mv88e6xxx_port_fdb_prepare, + .port_fdb_add = mv88e6xxx_port_fdb_add, + .port_fdb_del = mv88e6xxx_port_fdb_del, + .port_fdb_dump = mv88e6xxx_port_fdb_dump, +}; + +static int mv88e6xxx_register_switch(struct mv88e6xxx_priv_state *ps, + struct device_node *np) +{ + struct device *dev = ps->dev; + struct dsa_switch *ds; + + ds = devm_kzalloc(dev, sizeof(*ds), GFP_KERNEL); + if (!ds) + return -ENOMEM; + + ds->dev = dev; + ds->priv = ps; + ds->drv = &mv88e6xxx_switch_driver; + + dev_set_drvdata(dev, ds); + + return dsa_register_switch(ds, np); +} + +static void mv88e6xxx_unregister_switch(struct mv88e6xxx_priv_state *ps) +{ + dsa_unregister_switch(ps->ds); +} + +static int mv88e6xxx_probe(struct mdio_device *mdiodev) +{ + struct device *dev = &mdiodev->dev; + struct device_node *np = dev->of_node; + const struct mv88e6xxx_info *compat_info; + struct mv88e6xxx_priv_state *ps; + u32 eeprom_len; + int err; + + compat_info = of_device_get_match_data(dev); + if (!compat_info) + return -EINVAL; + + ps = mv88e6xxx_alloc_chip(dev); + if (!ps) + return -ENOMEM; + + ps->info = compat_info; + + err = mv88e6xxx_smi_init(ps, mdiodev->bus, mdiodev->addr); + if (err) + return err; + + err = mv88e6xxx_detect(ps); + if (err) + return err; + + ps->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_ASIS); + if (IS_ERR(ps->reset)) + return PTR_ERR(ps->reset); + + if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM) && + !of_property_read_u32(np, "eeprom-length", &eeprom_len)) + ps->eeprom_len = eeprom_len; + + err = mv88e6xxx_mdio_register(ps, np); + if (err) + return err; + + err = mv88e6xxx_register_switch(ps, np); + if (err) { + mv88e6xxx_mdio_unregister(ps); + return err; + } + + return 0; +} + +static void mv88e6xxx_remove(struct mdio_device *mdiodev) +{ + struct dsa_switch *ds = dev_get_drvdata(&mdiodev->dev); + struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); + + mv88e6xxx_unregister_switch(ps); + mv88e6xxx_mdio_unregister(ps); +} + +static const struct of_device_id mv88e6xxx_of_match[] = { + { + .compatible = "marvell,mv88e6085", + .data = &mv88e6xxx_table[MV88E6085], + }, + { /* sentinel */ }, +}; + +MODULE_DEVICE_TABLE(of, mv88e6xxx_of_match); + +static struct mdio_driver mv88e6xxx_driver = { + .probe = mv88e6xxx_probe, + .remove = mv88e6xxx_remove, + .mdiodrv.driver = { + .name = "mv88e6085", + .of_match_table = mv88e6xxx_of_match, + }, +}; + +static int __init mv88e6xxx_init(void) +{ + register_switch_driver(&mv88e6xxx_switch_driver); + return mdio_driver_register(&mv88e6xxx_driver); +} +module_init(mv88e6xxx_init); + +static void __exit mv88e6xxx_cleanup(void) +{ + mdio_driver_unregister(&mv88e6xxx_driver); + unregister_switch_driver(&mv88e6xxx_switch_driver); +} +module_exit(mv88e6xxx_cleanup); + +MODULE_AUTHOR("Lennert Buytenhek "); +MODULE_DESCRIPTION("Driver for Marvell 88E6XXX ethernet switch chips"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h new file mode 100644 index 0000000..856c6e5 --- /dev/null +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -0,0 +1,653 @@ +/* + * Marvell 88e6xxx common definitions + * + * Copyright (c) 2008 Marvell Semiconductor + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __MV88E6XXX_H +#define __MV88E6XXX_H + +#include +#include + +#ifndef UINT64_MAX +#define UINT64_MAX (u64)(~((u64)0)) +#endif + +#define SMI_CMD 0x00 +#define SMI_CMD_BUSY BIT(15) +#define SMI_CMD_CLAUSE_22 BIT(12) +#define SMI_CMD_OP_22_WRITE ((1 << 10) | SMI_CMD_BUSY | SMI_CMD_CLAUSE_22) +#define SMI_CMD_OP_22_READ ((2 << 10) | SMI_CMD_BUSY | SMI_CMD_CLAUSE_22) +#define SMI_CMD_OP_45_WRITE_ADDR ((0 << 10) | SMI_CMD_BUSY) +#define SMI_CMD_OP_45_WRITE_DATA ((1 << 10) | SMI_CMD_BUSY) +#define SMI_CMD_OP_45_READ_DATA ((2 << 10) | SMI_CMD_BUSY) +#define SMI_CMD_OP_45_READ_DATA_INC ((3 << 10) | SMI_CMD_BUSY) +#define SMI_DATA 0x01 + +/* Fiber/SERDES Registers are located at SMI address F, page 1 */ +#define REG_FIBER_SERDES 0x0f +#define PAGE_FIBER_SERDES 0x01 + +#define REG_PORT(p) (0x10 + (p)) +#define PORT_STATUS 0x00 +#define PORT_STATUS_PAUSE_EN BIT(15) +#define PORT_STATUS_MY_PAUSE BIT(14) +#define PORT_STATUS_HD_FLOW BIT(13) +#define PORT_STATUS_PHY_DETECT BIT(12) +#define PORT_STATUS_LINK BIT(11) +#define PORT_STATUS_DUPLEX BIT(10) +#define PORT_STATUS_SPEED_MASK 0x0300 +#define PORT_STATUS_SPEED_10 0x0000 +#define PORT_STATUS_SPEED_100 0x0100 +#define PORT_STATUS_SPEED_1000 0x0200 +#define PORT_STATUS_EEE BIT(6) /* 6352 */ +#define PORT_STATUS_AM_DIS BIT(6) /* 6165 */ +#define PORT_STATUS_MGMII BIT(6) /* 6185 */ +#define PORT_STATUS_TX_PAUSED BIT(5) +#define PORT_STATUS_FLOW_CTRL BIT(4) +#define PORT_STATUS_CMODE_MASK 0x0f +#define PORT_STATUS_CMODE_100BASE_X 0x8 +#define PORT_STATUS_CMODE_1000BASE_X 0x9 +#define PORT_STATUS_CMODE_SGMII 0xa +#define PORT_PCS_CTRL 0x01 +#define PORT_PCS_CTRL_RGMII_DELAY_RXCLK BIT(15) +#define PORT_PCS_CTRL_RGMII_DELAY_TXCLK BIT(14) +#define PORT_PCS_CTRL_FC BIT(7) +#define PORT_PCS_CTRL_FORCE_FC BIT(6) +#define PORT_PCS_CTRL_LINK_UP BIT(5) +#define PORT_PCS_CTRL_FORCE_LINK BIT(4) +#define PORT_PCS_CTRL_DUPLEX_FULL BIT(3) +#define PORT_PCS_CTRL_FORCE_DUPLEX BIT(2) +#define PORT_PCS_CTRL_10 0x00 +#define PORT_PCS_CTRL_100 0x01 +#define PORT_PCS_CTRL_1000 0x02 +#define PORT_PCS_CTRL_UNFORCED 0x03 +#define PORT_PAUSE_CTRL 0x02 +#define PORT_SWITCH_ID 0x03 +#define PORT_SWITCH_ID_PROD_NUM_6085 0x04a +#define PORT_SWITCH_ID_PROD_NUM_6095 0x095 +#define PORT_SWITCH_ID_PROD_NUM_6131 0x106 +#define PORT_SWITCH_ID_PROD_NUM_6320 0x115 +#define PORT_SWITCH_ID_PROD_NUM_6123 0x121 +#define PORT_SWITCH_ID_PROD_NUM_6161 0x161 +#define PORT_SWITCH_ID_PROD_NUM_6165 0x165 +#define PORT_SWITCH_ID_PROD_NUM_6171 0x171 +#define PORT_SWITCH_ID_PROD_NUM_6172 0x172 +#define PORT_SWITCH_ID_PROD_NUM_6175 0x175 +#define PORT_SWITCH_ID_PROD_NUM_6176 0x176 +#define PORT_SWITCH_ID_PROD_NUM_6185 0x1a7 +#define PORT_SWITCH_ID_PROD_NUM_6240 0x240 +#define PORT_SWITCH_ID_PROD_NUM_6321 0x310 +#define PORT_SWITCH_ID_PROD_NUM_6352 0x352 +#define PORT_SWITCH_ID_PROD_NUM_6350 0x371 +#define PORT_SWITCH_ID_PROD_NUM_6351 0x375 +#define PORT_CONTROL 0x04 +#define PORT_CONTROL_USE_CORE_TAG BIT(15) +#define PORT_CONTROL_DROP_ON_LOCK BIT(14) +#define PORT_CONTROL_EGRESS_UNMODIFIED (0x0 << 12) +#define PORT_CONTROL_EGRESS_UNTAGGED (0x1 << 12) +#define PORT_CONTROL_EGRESS_TAGGED (0x2 << 12) +#define PORT_CONTROL_EGRESS_ADD_TAG (0x3 << 12) +#define PORT_CONTROL_HEADER BIT(11) +#define PORT_CONTROL_IGMP_MLD_SNOOP BIT(10) +#define PORT_CONTROL_DOUBLE_TAG BIT(9) +#define PORT_CONTROL_FRAME_MODE_NORMAL (0x0 << 8) +#define PORT_CONTROL_FRAME_MODE_DSA (0x1 << 8) +#define PORT_CONTROL_FRAME_MODE_PROVIDER (0x2 << 8) +#define PORT_CONTROL_FRAME_ETHER_TYPE_DSA (0x3 << 8) +#define PORT_CONTROL_DSA_TAG BIT(8) +#define PORT_CONTROL_VLAN_TUNNEL BIT(7) +#define PORT_CONTROL_TAG_IF_BOTH BIT(6) +#define PORT_CONTROL_USE_IP BIT(5) +#define PORT_CONTROL_USE_TAG BIT(4) +#define PORT_CONTROL_FORWARD_UNKNOWN_MC BIT(3) +#define PORT_CONTROL_FORWARD_UNKNOWN BIT(2) +#define PORT_CONTROL_STATE_MASK 0x03 +#define PORT_CONTROL_STATE_DISABLED 0x00 +#define PORT_CONTROL_STATE_BLOCKING 0x01 +#define PORT_CONTROL_STATE_LEARNING 0x02 +#define PORT_CONTROL_STATE_FORWARDING 0x03 +#define PORT_CONTROL_1 0x05 +#define PORT_CONTROL_1_FID_11_4_MASK (0xff << 0) +#define PORT_BASE_VLAN 0x06 +#define PORT_BASE_VLAN_FID_3_0_MASK (0xf << 12) +#define PORT_DEFAULT_VLAN 0x07 +#define PORT_DEFAULT_VLAN_MASK 0xfff +#define PORT_CONTROL_2 0x08 +#define PORT_CONTROL_2_IGNORE_FCS BIT(15) +#define PORT_CONTROL_2_VTU_PRI_OVERRIDE BIT(14) +#define PORT_CONTROL_2_SA_PRIO_OVERRIDE BIT(13) +#define PORT_CONTROL_2_DA_PRIO_OVERRIDE BIT(12) +#define PORT_CONTROL_2_JUMBO_1522 (0x00 << 12) +#define PORT_CONTROL_2_JUMBO_2048 (0x01 << 12) +#define PORT_CONTROL_2_JUMBO_10240 (0x02 << 12) +#define PORT_CONTROL_2_8021Q_MASK (0x03 << 10) +#define PORT_CONTROL_2_8021Q_DISABLED (0x00 << 10) +#define PORT_CONTROL_2_8021Q_FALLBACK (0x01 << 10) +#define PORT_CONTROL_2_8021Q_CHECK (0x02 << 10) +#define PORT_CONTROL_2_8021Q_SECURE (0x03 << 10) +#define PORT_CONTROL_2_DISCARD_TAGGED BIT(9) +#define PORT_CONTROL_2_DISCARD_UNTAGGED BIT(8) +#define PORT_CONTROL_2_MAP_DA BIT(7) +#define PORT_CONTROL_2_DEFAULT_FORWARD BIT(6) +#define PORT_CONTROL_2_FORWARD_UNKNOWN BIT(6) +#define PORT_CONTROL_2_EGRESS_MONITOR BIT(5) +#define PORT_CONTROL_2_INGRESS_MONITOR BIT(4) +#define PORT_RATE_CONTROL 0x09 +#define PORT_RATE_CONTROL_2 0x0a +#define PORT_ASSOC_VECTOR 0x0b +#define PORT_ASSOC_VECTOR_HOLD_AT_1 BIT(15) +#define PORT_ASSOC_VECTOR_INT_AGE_OUT BIT(14) +#define PORT_ASSOC_VECTOR_LOCKED_PORT BIT(13) +#define PORT_ASSOC_VECTOR_IGNORE_WRONG BIT(12) +#define PORT_ASSOC_VECTOR_REFRESH_LOCKED BIT(11) +#define PORT_ATU_CONTROL 0x0c +#define PORT_PRI_OVERRIDE 0x0d +#define PORT_ETH_TYPE 0x0f +#define PORT_IN_DISCARD_LO 0x10 +#define PORT_IN_DISCARD_HI 0x11 +#define PORT_IN_FILTERED 0x12 +#define PORT_OUT_FILTERED 0x13 +#define PORT_TAG_REGMAP_0123 0x18 +#define PORT_TAG_REGMAP_4567 0x19 + +#define REG_GLOBAL 0x1b +#define GLOBAL_STATUS 0x00 +#define GLOBAL_STATUS_PPU_STATE BIT(15) /* 6351 and 6171 */ +/* Two bits for 6165, 6185 etc */ +#define GLOBAL_STATUS_PPU_MASK (0x3 << 14) +#define GLOBAL_STATUS_PPU_DISABLED_RST (0x0 << 14) +#define GLOBAL_STATUS_PPU_INITIALIZING (0x1 << 14) +#define GLOBAL_STATUS_PPU_DISABLED (0x2 << 14) +#define GLOBAL_STATUS_PPU_POLLING (0x3 << 14) +#define GLOBAL_MAC_01 0x01 +#define GLOBAL_MAC_23 0x02 +#define GLOBAL_MAC_45 0x03 +#define GLOBAL_ATU_FID 0x01 /* 6097 6165 6351 6352 */ +#define GLOBAL_VTU_FID 0x02 /* 6097 6165 6351 6352 */ +#define GLOBAL_VTU_FID_MASK 0xfff +#define GLOBAL_VTU_SID 0x03 /* 6097 6165 6351 6352 */ +#define GLOBAL_VTU_SID_MASK 0x3f +#define GLOBAL_CONTROL 0x04 +#define GLOBAL_CONTROL_SW_RESET BIT(15) +#define GLOBAL_CONTROL_PPU_ENABLE BIT(14) +#define GLOBAL_CONTROL_DISCARD_EXCESS BIT(13) /* 6352 */ +#define GLOBAL_CONTROL_SCHED_PRIO BIT(11) /* 6152 */ +#define GLOBAL_CONTROL_MAX_FRAME_1632 BIT(10) /* 6152 */ +#define GLOBAL_CONTROL_RELOAD_EEPROM BIT(9) /* 6152 */ +#define GLOBAL_CONTROL_DEVICE_EN BIT(7) +#define GLOBAL_CONTROL_STATS_DONE_EN BIT(6) +#define GLOBAL_CONTROL_VTU_PROBLEM_EN BIT(5) +#define GLOBAL_CONTROL_VTU_DONE_EN BIT(4) +#define GLOBAL_CONTROL_ATU_PROBLEM_EN BIT(3) +#define GLOBAL_CONTROL_ATU_DONE_EN BIT(2) +#define GLOBAL_CONTROL_TCAM_EN BIT(1) +#define GLOBAL_CONTROL_EEPROM_DONE_EN BIT(0) +#define GLOBAL_VTU_OP 0x05 +#define GLOBAL_VTU_OP_BUSY BIT(15) +#define GLOBAL_VTU_OP_FLUSH_ALL ((0x01 << 12) | GLOBAL_VTU_OP_BUSY) +#define GLOBAL_VTU_OP_VTU_LOAD_PURGE ((0x03 << 12) | GLOBAL_VTU_OP_BUSY) +#define GLOBAL_VTU_OP_VTU_GET_NEXT ((0x04 << 12) | GLOBAL_VTU_OP_BUSY) +#define GLOBAL_VTU_OP_STU_LOAD_PURGE ((0x05 << 12) | GLOBAL_VTU_OP_BUSY) +#define GLOBAL_VTU_OP_STU_GET_NEXT ((0x06 << 12) | GLOBAL_VTU_OP_BUSY) +#define GLOBAL_VTU_VID 0x06 +#define GLOBAL_VTU_VID_MASK 0xfff +#define GLOBAL_VTU_VID_VALID BIT(12) +#define GLOBAL_VTU_DATA_0_3 0x07 +#define GLOBAL_VTU_DATA_4_7 0x08 +#define GLOBAL_VTU_DATA_8_11 0x09 +#define GLOBAL_VTU_STU_DATA_MASK 0x03 +#define GLOBAL_VTU_DATA_MEMBER_TAG_UNMODIFIED 0x00 +#define GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED 0x01 +#define GLOBAL_VTU_DATA_MEMBER_TAG_TAGGED 0x02 +#define GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER 0x03 +#define GLOBAL_STU_DATA_PORT_STATE_DISABLED 0x00 +#define GLOBAL_STU_DATA_PORT_STATE_BLOCKING 0x01 +#define GLOBAL_STU_DATA_PORT_STATE_LEARNING 0x02 +#define GLOBAL_STU_DATA_PORT_STATE_FORWARDING 0x03 +#define GLOBAL_ATU_CONTROL 0x0a +#define GLOBAL_ATU_CONTROL_LEARN2ALL BIT(3) +#define GLOBAL_ATU_OP 0x0b +#define GLOBAL_ATU_OP_BUSY BIT(15) +#define GLOBAL_ATU_OP_NOP (0 << 12) +#define GLOBAL_ATU_OP_FLUSH_MOVE_ALL ((1 << 12) | GLOBAL_ATU_OP_BUSY) +#define GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC ((2 << 12) | GLOBAL_ATU_OP_BUSY) +#define GLOBAL_ATU_OP_LOAD_DB ((3 << 12) | GLOBAL_ATU_OP_BUSY) +#define GLOBAL_ATU_OP_GET_NEXT_DB ((4 << 12) | GLOBAL_ATU_OP_BUSY) +#define GLOBAL_ATU_OP_FLUSH_MOVE_ALL_DB ((5 << 12) | GLOBAL_ATU_OP_BUSY) +#define GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC_DB ((6 << 12) | GLOBAL_ATU_OP_BUSY) +#define GLOBAL_ATU_OP_GET_CLR_VIOLATION ((7 << 12) | GLOBAL_ATU_OP_BUSY) +#define GLOBAL_ATU_DATA 0x0c +#define GLOBAL_ATU_DATA_TRUNK BIT(15) +#define GLOBAL_ATU_DATA_TRUNK_ID_MASK 0x00f0 +#define GLOBAL_ATU_DATA_TRUNK_ID_SHIFT 4 +#define GLOBAL_ATU_DATA_PORT_VECTOR_MASK 0x3ff0 +#define GLOBAL_ATU_DATA_PORT_VECTOR_SHIFT 4 +#define GLOBAL_ATU_DATA_STATE_MASK 0x0f +#define GLOBAL_ATU_DATA_STATE_UNUSED 0x00 +#define GLOBAL_ATU_DATA_STATE_UC_MGMT 0x0d +#define GLOBAL_ATU_DATA_STATE_UC_STATIC 0x0e +#define GLOBAL_ATU_DATA_STATE_UC_PRIO_OVER 0x0f +#define GLOBAL_ATU_DATA_STATE_MC_NONE_RATE 0x05 +#define GLOBAL_ATU_DATA_STATE_MC_STATIC 0x07 +#define GLOBAL_ATU_DATA_STATE_MC_MGMT 0x0e +#define GLOBAL_ATU_DATA_STATE_MC_PRIO_OVER 0x0f +#define GLOBAL_ATU_MAC_01 0x0d +#define GLOBAL_ATU_MAC_23 0x0e +#define GLOBAL_ATU_MAC_45 0x0f +#define GLOBAL_IP_PRI_0 0x10 +#define GLOBAL_IP_PRI_1 0x11 +#define GLOBAL_IP_PRI_2 0x12 +#define GLOBAL_IP_PRI_3 0x13 +#define GLOBAL_IP_PRI_4 0x14 +#define GLOBAL_IP_PRI_5 0x15 +#define GLOBAL_IP_PRI_6 0x16 +#define GLOBAL_IP_PRI_7 0x17 +#define GLOBAL_IEEE_PRI 0x18 +#define GLOBAL_CORE_TAG_TYPE 0x19 +#define GLOBAL_MONITOR_CONTROL 0x1a +#define GLOBAL_MONITOR_CONTROL_INGRESS_SHIFT 12 +#define GLOBAL_MONITOR_CONTROL_EGRESS_SHIFT 8 +#define GLOBAL_MONITOR_CONTROL_ARP_SHIFT 4 +#define GLOBAL_MONITOR_CONTROL_MIRROR_SHIFT 0 +#define GLOBAL_MONITOR_CONTROL_ARP_DISABLED (0xf0) +#define GLOBAL_CONTROL_2 0x1c +#define GLOBAL_CONTROL_2_NO_CASCADE 0xe000 +#define GLOBAL_CONTROL_2_MULTIPLE_CASCADE 0xf000 + +#define GLOBAL_STATS_OP 0x1d +#define GLOBAL_STATS_OP_BUSY BIT(15) +#define GLOBAL_STATS_OP_NOP (0 << 12) +#define GLOBAL_STATS_OP_FLUSH_ALL ((1 << 12) | GLOBAL_STATS_OP_BUSY) +#define GLOBAL_STATS_OP_FLUSH_PORT ((2 << 12) | GLOBAL_STATS_OP_BUSY) +#define GLOBAL_STATS_OP_READ_CAPTURED ((4 << 12) | GLOBAL_STATS_OP_BUSY) +#define GLOBAL_STATS_OP_CAPTURE_PORT ((5 << 12) | GLOBAL_STATS_OP_BUSY) +#define GLOBAL_STATS_OP_HIST_RX ((1 << 10) | GLOBAL_STATS_OP_BUSY) +#define GLOBAL_STATS_OP_HIST_TX ((2 << 10) | GLOBAL_STATS_OP_BUSY) +#define GLOBAL_STATS_OP_HIST_RX_TX ((3 << 10) | GLOBAL_STATS_OP_BUSY) +#define GLOBAL_STATS_OP_BANK_1 BIT(9) +#define GLOBAL_STATS_COUNTER_32 0x1e +#define GLOBAL_STATS_COUNTER_01 0x1f + +#define REG_GLOBAL2 0x1c +#define GLOBAL2_INT_SOURCE 0x00 +#define GLOBAL2_INT_MASK 0x01 +#define GLOBAL2_MGMT_EN_2X 0x02 +#define GLOBAL2_MGMT_EN_0X 0x03 +#define GLOBAL2_FLOW_CONTROL 0x04 +#define GLOBAL2_SWITCH_MGMT 0x05 +#define GLOBAL2_SWITCH_MGMT_USE_DOUBLE_TAG_DATA BIT(15) +#define GLOBAL2_SWITCH_MGMT_PREVENT_LOOPS BIT(14) +#define GLOBAL2_SWITCH_MGMT_FLOW_CONTROL_MSG BIT(13) +#define GLOBAL2_SWITCH_MGMT_FORCE_FLOW_CTRL_PRI BIT(7) +#define GLOBAL2_SWITCH_MGMT_RSVD2CPU BIT(3) +#define GLOBAL2_DEVICE_MAPPING 0x06 +#define GLOBAL2_DEVICE_MAPPING_UPDATE BIT(15) +#define GLOBAL2_DEVICE_MAPPING_TARGET_SHIFT 8 +#define GLOBAL2_DEVICE_MAPPING_PORT_MASK 0x0f +#define GLOBAL2_TRUNK_MASK 0x07 +#define GLOBAL2_TRUNK_MASK_UPDATE BIT(15) +#define GLOBAL2_TRUNK_MASK_NUM_SHIFT 12 +#define GLOBAL2_TRUNK_MAPPING 0x08 +#define GLOBAL2_TRUNK_MAPPING_UPDATE BIT(15) +#define GLOBAL2_TRUNK_MAPPING_ID_SHIFT 11 +#define GLOBAL2_INGRESS_OP 0x09 +#define GLOBAL2_INGRESS_DATA 0x0a +#define GLOBAL2_PVT_ADDR 0x0b +#define GLOBAL2_PVT_DATA 0x0c +#define GLOBAL2_SWITCH_MAC 0x0d +#define GLOBAL2_SWITCH_MAC_BUSY BIT(15) +#define GLOBAL2_ATU_STATS 0x0e +#define GLOBAL2_PRIO_OVERRIDE 0x0f +#define GLOBAL2_PRIO_OVERRIDE_FORCE_SNOOP BIT(7) +#define GLOBAL2_PRIO_OVERRIDE_SNOOP_SHIFT 4 +#define GLOBAL2_PRIO_OVERRIDE_FORCE_ARP BIT(3) +#define GLOBAL2_PRIO_OVERRIDE_ARP_SHIFT 0 +#define GLOBAL2_EEPROM_OP 0x14 +#define GLOBAL2_EEPROM_OP_BUSY BIT(15) +#define GLOBAL2_EEPROM_OP_WRITE ((3 << 12) | GLOBAL2_EEPROM_OP_BUSY) +#define GLOBAL2_EEPROM_OP_READ ((4 << 12) | GLOBAL2_EEPROM_OP_BUSY) +#define GLOBAL2_EEPROM_OP_LOAD BIT(11) +#define GLOBAL2_EEPROM_OP_WRITE_EN BIT(10) +#define GLOBAL2_EEPROM_OP_ADDR_MASK 0xff +#define GLOBAL2_EEPROM_DATA 0x15 +#define GLOBAL2_PTP_AVB_OP 0x16 +#define GLOBAL2_PTP_AVB_DATA 0x17 +#define GLOBAL2_SMI_OP 0x18 +#define GLOBAL2_SMI_OP_BUSY BIT(15) +#define GLOBAL2_SMI_OP_CLAUSE_22 BIT(12) +#define GLOBAL2_SMI_OP_22_WRITE ((1 << 10) | GLOBAL2_SMI_OP_BUSY | \ + GLOBAL2_SMI_OP_CLAUSE_22) +#define GLOBAL2_SMI_OP_22_READ ((2 << 10) | GLOBAL2_SMI_OP_BUSY | \ + GLOBAL2_SMI_OP_CLAUSE_22) +#define GLOBAL2_SMI_OP_45_WRITE_ADDR ((0 << 10) | GLOBAL2_SMI_OP_BUSY) +#define GLOBAL2_SMI_OP_45_WRITE_DATA ((1 << 10) | GLOBAL2_SMI_OP_BUSY) +#define GLOBAL2_SMI_OP_45_READ_DATA ((2 << 10) | GLOBAL2_SMI_OP_BUSY) +#define GLOBAL2_SMI_DATA 0x19 +#define GLOBAL2_SCRATCH_MISC 0x1a +#define GLOBAL2_SCRATCH_BUSY BIT(15) +#define GLOBAL2_SCRATCH_REGISTER_SHIFT 8 +#define GLOBAL2_SCRATCH_VALUE_MASK 0xff +#define GLOBAL2_WDOG_CONTROL 0x1b +#define GLOBAL2_QOS_WEIGHT 0x1c +#define GLOBAL2_MISC 0x1d + +#define MV88E6XXX_N_FID 4096 + +/* List of supported models */ +enum mv88e6xxx_model { + MV88E6085, + MV88E6095, + MV88E6123, + MV88E6131, + MV88E6161, + MV88E6165, + MV88E6171, + MV88E6172, + MV88E6175, + MV88E6176, + MV88E6185, + MV88E6240, + MV88E6320, + MV88E6321, + MV88E6350, + MV88E6351, + MV88E6352, +}; + +enum mv88e6xxx_family { + MV88E6XXX_FAMILY_NONE, + MV88E6XXX_FAMILY_6065, /* 6031 6035 6061 6065 */ + MV88E6XXX_FAMILY_6095, /* 6092 6095 */ + MV88E6XXX_FAMILY_6097, /* 6046 6085 6096 6097 */ + MV88E6XXX_FAMILY_6165, /* 6123 6161 6165 */ + MV88E6XXX_FAMILY_6185, /* 6108 6121 6122 6131 6152 6155 6182 6185 */ + MV88E6XXX_FAMILY_6320, /* 6320 6321 */ + MV88E6XXX_FAMILY_6351, /* 6171 6175 6350 6351 */ + MV88E6XXX_FAMILY_6352, /* 6172 6176 6240 6352 */ +}; + +enum mv88e6xxx_cap { + /* Address Translation Unit. + * The ATU is used to lookup and learn MAC addresses. See GLOBAL_ATU_OP. + */ + MV88E6XXX_CAP_ATU, + + /* Energy Efficient Ethernet. + */ + MV88E6XXX_CAP_EEE, + + /* EEPROM Command and Data registers. + * See GLOBAL2_EEPROM_OP and GLOBAL2_EEPROM_DATA. + */ + MV88E6XXX_CAP_EEPROM, + + /* Multi-chip Addressing Mode. + * Some chips require an indirect SMI access when their SMI device + * address is not zero. See SMI_CMD and SMI_DATA. + */ + MV88E6XXX_CAP_MULTI_CHIP, + + /* Port State Filtering for 802.1D Spanning Tree. + * See PORT_CONTROL_STATE_* values in the PORT_CONTROL register. + */ + MV88E6XXX_CAP_PORTSTATE, + + /* PHY Polling Unit. + * See GLOBAL_CONTROL_PPU_ENABLE and GLOBAL_STATUS_PPU_POLLING. + */ + MV88E6XXX_CAP_PPU, + MV88E6XXX_CAP_PPU_ACTIVE, + + /* SMI PHY Command and Data registers. + * This requires an indirect access to PHY registers through + * GLOBAL2_SMI_OP, otherwise direct access to PHY registers is done. + */ + MV88E6XXX_CAP_SMI_PHY, + + /* Per VLAN Spanning Tree Unit (STU). + * The Port State database, if present, is accessed through VTU + * operations and dedicated SID registers. See GLOBAL_VTU_SID. + */ + MV88E6XXX_CAP_STU, + + /* Switch MAC/WoL/WoF register. + * This requires an indirect access to set the switch MAC address + * through GLOBAL2_SWITCH_MAC, otherwise GLOBAL_MAC_01, GLOBAL_MAC_23, + * and GLOBAL_MAC_45 are used with a direct access. + */ + MV88E6XXX_CAP_SWITCH_MAC_WOL_WOF, + + /* Internal temperature sensor. + * Available from any enabled port's PHY register 26, page 6. + */ + MV88E6XXX_CAP_TEMP, + MV88E6XXX_CAP_TEMP_LIMIT, + + /* In-chip Port Based VLANs. + * Each port VLANTable register (see PORT_BASE_VLAN) is used to restrict + * the output (or egress) ports to which it is allowed to send frames. + */ + MV88E6XXX_CAP_VLANTABLE, + + /* VLAN Table Unit. + * The VTU is used to program 802.1Q VLANs. See GLOBAL_VTU_OP. + */ + MV88E6XXX_CAP_VTU, +}; + +/* Bitmask of capabilities */ +#define MV88E6XXX_FLAG_ATU BIT(MV88E6XXX_CAP_ATU) +#define MV88E6XXX_FLAG_EEE BIT(MV88E6XXX_CAP_EEE) +#define MV88E6XXX_FLAG_EEPROM BIT(MV88E6XXX_CAP_EEPROM) +#define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) +#define MV88E6XXX_FLAG_PORTSTATE BIT(MV88E6XXX_CAP_PORTSTATE) +#define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) +#define MV88E6XXX_FLAG_PPU_ACTIVE BIT(MV88E6XXX_CAP_PPU_ACTIVE) +#define MV88E6XXX_FLAG_SMI_PHY BIT(MV88E6XXX_CAP_SMI_PHY) +#define MV88E6XXX_FLAG_STU BIT(MV88E6XXX_CAP_STU) +#define MV88E6XXX_FLAG_SWITCH_MAC BIT(MV88E6XXX_CAP_SWITCH_MAC_WOL_WOF) +#define MV88E6XXX_FLAG_TEMP BIT(MV88E6XXX_CAP_TEMP) +#define MV88E6XXX_FLAG_TEMP_LIMIT BIT(MV88E6XXX_CAP_TEMP_LIMIT) +#define MV88E6XXX_FLAG_VLANTABLE BIT(MV88E6XXX_CAP_VLANTABLE) +#define MV88E6XXX_FLAG_VTU BIT(MV88E6XXX_CAP_VTU) + +#define MV88E6XXX_FLAGS_FAMILY_6095 \ + (MV88E6XXX_FLAG_ATU | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ + MV88E6XXX_FLAG_PPU | \ + MV88E6XXX_FLAG_VLANTABLE | \ + MV88E6XXX_FLAG_VTU) + +#define MV88E6XXX_FLAGS_FAMILY_6097 \ + (MV88E6XXX_FLAG_ATU | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ + MV88E6XXX_FLAG_PPU | \ + MV88E6XXX_FLAG_STU | \ + MV88E6XXX_FLAG_VLANTABLE | \ + MV88E6XXX_FLAG_VTU) + +#define MV88E6XXX_FLAGS_FAMILY_6165 \ + (MV88E6XXX_FLAG_MULTI_CHIP | \ + MV88E6XXX_FLAG_STU | \ + MV88E6XXX_FLAG_SWITCH_MAC | \ + MV88E6XXX_FLAG_TEMP | \ + MV88E6XXX_FLAG_VTU) + +#define MV88E6XXX_FLAGS_FAMILY_6185 \ + (MV88E6XXX_FLAG_ATU | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ + MV88E6XXX_FLAG_PPU | \ + MV88E6XXX_FLAG_VLANTABLE | \ + MV88E6XXX_FLAG_VTU) + +#define MV88E6XXX_FLAGS_FAMILY_6320 \ + (MV88E6XXX_FLAG_ATU | \ + MV88E6XXX_FLAG_EEE | \ + MV88E6XXX_FLAG_EEPROM | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ + MV88E6XXX_FLAG_PORTSTATE | \ + MV88E6XXX_FLAG_PPU_ACTIVE | \ + MV88E6XXX_FLAG_SMI_PHY | \ + MV88E6XXX_FLAG_SWITCH_MAC | \ + MV88E6XXX_FLAG_TEMP | \ + MV88E6XXX_FLAG_TEMP_LIMIT | \ + MV88E6XXX_FLAG_VLANTABLE | \ + MV88E6XXX_FLAG_VTU) + +#define MV88E6XXX_FLAGS_FAMILY_6351 \ + (MV88E6XXX_FLAG_ATU | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ + MV88E6XXX_FLAG_PORTSTATE | \ + MV88E6XXX_FLAG_PPU_ACTIVE | \ + MV88E6XXX_FLAG_SMI_PHY | \ + MV88E6XXX_FLAG_STU | \ + MV88E6XXX_FLAG_SWITCH_MAC | \ + MV88E6XXX_FLAG_TEMP | \ + MV88E6XXX_FLAG_VLANTABLE | \ + MV88E6XXX_FLAG_VTU) + +#define MV88E6XXX_FLAGS_FAMILY_6352 \ + (MV88E6XXX_FLAG_ATU | \ + MV88E6XXX_FLAG_EEE | \ + MV88E6XXX_FLAG_EEPROM | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ + MV88E6XXX_FLAG_PORTSTATE | \ + MV88E6XXX_FLAG_PPU_ACTIVE | \ + MV88E6XXX_FLAG_SMI_PHY | \ + MV88E6XXX_FLAG_STU | \ + MV88E6XXX_FLAG_SWITCH_MAC | \ + MV88E6XXX_FLAG_TEMP | \ + MV88E6XXX_FLAG_TEMP_LIMIT | \ + MV88E6XXX_FLAG_VLANTABLE | \ + MV88E6XXX_FLAG_VTU) + +struct mv88e6xxx_info { + enum mv88e6xxx_family family; + u16 prod_num; + const char *name; + unsigned int num_databases; + unsigned int num_ports; + unsigned int port_base_addr; + unsigned long flags; +}; + +struct mv88e6xxx_atu_entry { + u16 fid; + u8 state; + bool trunk; + u16 portv_trunkid; + u8 mac[ETH_ALEN]; +}; + +struct mv88e6xxx_vtu_stu_entry { + /* VTU only */ + u16 vid; + u16 fid; + + /* VTU and STU */ + u8 sid; + bool valid; + u8 data[DSA_MAX_PORTS]; +}; + +struct mv88e6xxx_ops; + +struct mv88e6xxx_priv_port { + struct net_device *bridge_dev; +}; + +struct mv88e6xxx_priv_state { + const struct mv88e6xxx_info *info; + + /* The dsa_switch this private structure is related to */ + struct dsa_switch *ds; + + /* The device this structure is associated to */ + struct device *dev; + + /* This mutex protects the access to the switch registers */ + struct mutex reg_lock; + + /* The MII bus and the address on the bus that is used to + * communication with the switch + */ + const struct mv88e6xxx_ops *smi_ops; + struct mii_bus *bus; + int sw_addr; + + /* Handles automatic disabling and re-enabling of the PHY + * polling unit. + */ + struct mutex ppu_mutex; + int ppu_disabled; + struct work_struct ppu_work; + struct timer_list ppu_timer; + + /* This mutex serialises access to the statistics unit. + * Hold this mutex over snapshot + dump sequences. + */ + struct mutex stats_mutex; + + /* This mutex serializes phy access for chips with + * indirect phy addressing. It is unused for chips + * with direct phy access. + */ + struct mutex phy_mutex; + + /* This mutex serializes eeprom access for chips with + * eeprom support. + */ + struct mutex eeprom_mutex; + + struct mv88e6xxx_priv_port ports[DSA_MAX_PORTS]; + + /* A switch may have a GPIO line tied to its reset pin. Parse + * this from the device tree, and use it before performing + * switch soft reset. + */ + struct gpio_desc *reset; + + /* set to size of eeprom if supported by the switch */ + int eeprom_len; + + /* Device node for the MDIO bus */ + struct device_node *mdio_np; + + /* And the MDIO bus itself */ + struct mii_bus *mdio_bus; +}; + +struct mv88e6xxx_ops { + int (*read)(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 *val); + int (*write)(struct mv88e6xxx_priv_state *ps, + int addr, int reg, u16 val); +}; + +enum stat_type { + BANK0, + BANK1, + PORT, +}; + +struct mv88e6xxx_hw_stat { + char string[ETH_GSTRING_LEN]; + int sizeof_stat; + int reg; + enum stat_type type; +}; + +static inline bool mv88e6xxx_has(struct mv88e6xxx_priv_state *ps, + unsigned long flags) +{ + return (ps->info->flags & flags) == flags; +} + +#endif -- cgit v0.10.2 From fad09c73c27020001cd472343efdacf60a93f8ea Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Tue, 21 Jun 2016 12:28:20 -0400 Subject: net: dsa: mv88e6xxx: rename single-chip support With the upcoming support for cross-chip operations, it will be hard to distinguish portions of code supporting a single-chip or a switch fabric of interconnected chips. Make the code clearer now, by renaming the mv88e6xxx_priv_state chip structure to mv88e6xxx_chip. This patch brings no functional changes. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/Makefile b/drivers/net/dsa/mv88e6xxx/Makefile index 1128fc7..6e29a75 100644 --- a/drivers/net/dsa/mv88e6xxx/Makefile +++ b/drivers/net/dsa/mv88e6xxx/Makefile @@ -1 +1 @@ -obj-$(CONFIG_NET_DSA_MV88E6XXX) += mv88e6xxx.o +obj-$(CONFIG_NET_DSA_MV88E6XXX) += chip.o diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c new file mode 100644 index 0000000..5cb06f7 --- /dev/null +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -0,0 +1,3967 @@ +/* + * Marvell 88e6xxx Ethernet switch single-chip support + * + * Copyright (c) 2008 Marvell Semiconductor + * + * Copyright (c) 2015 CMC Electronics, Inc. + * Added support for VLAN Table Unit operations + * + * Copyright (c) 2016 Andrew Lunn + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "mv88e6xxx.h" + +static void assert_reg_lock(struct mv88e6xxx_chip *chip) +{ + if (unlikely(!mutex_is_locked(&chip->reg_lock))) { + dev_err(chip->dev, "Switch registers lock not held!\n"); + dump_stack(); + } +} + +/* The switch ADDR[4:1] configuration pins define the chip SMI device address + * (ADDR[0] is always zero, thus only even SMI addresses can be strapped). + * + * When ADDR is all zero, the chip uses Single-chip Addressing Mode, assuming it + * is the only device connected to the SMI master. In this mode it responds to + * all 32 possible SMI addresses, and thus maps directly the internal devices. + * + * When ADDR is non-zero, the chip uses Multi-chip Addressing Mode, allowing + * multiple devices to share the SMI interface. In this mode it responds to only + * 2 registers, used to indirectly access the internal SMI devices. + */ + +static int mv88e6xxx_smi_read(struct mv88e6xxx_chip *chip, + int addr, int reg, u16 *val) +{ + if (!chip->smi_ops) + return -EOPNOTSUPP; + + return chip->smi_ops->read(chip, addr, reg, val); +} + +static int mv88e6xxx_smi_write(struct mv88e6xxx_chip *chip, + int addr, int reg, u16 val) +{ + if (!chip->smi_ops) + return -EOPNOTSUPP; + + return chip->smi_ops->write(chip, addr, reg, val); +} + +static int mv88e6xxx_smi_single_chip_read(struct mv88e6xxx_chip *chip, + int addr, int reg, u16 *val) +{ + int ret; + + ret = mdiobus_read_nested(chip->bus, addr, reg); + if (ret < 0) + return ret; + + *val = ret & 0xffff; + + return 0; +} + +static int mv88e6xxx_smi_single_chip_write(struct mv88e6xxx_chip *chip, + int addr, int reg, u16 val) +{ + int ret; + + ret = mdiobus_write_nested(chip->bus, addr, reg, val); + if (ret < 0) + return ret; + + return 0; +} + +static const struct mv88e6xxx_ops mv88e6xxx_smi_single_chip_ops = { + .read = mv88e6xxx_smi_single_chip_read, + .write = mv88e6xxx_smi_single_chip_write, +}; + +static int mv88e6xxx_smi_multi_chip_wait(struct mv88e6xxx_chip *chip) +{ + int ret; + int i; + + for (i = 0; i < 16; i++) { + ret = mdiobus_read_nested(chip->bus, chip->sw_addr, SMI_CMD); + if (ret < 0) + return ret; + + if ((ret & SMI_CMD_BUSY) == 0) + return 0; + } + + return -ETIMEDOUT; +} + +static int mv88e6xxx_smi_multi_chip_read(struct mv88e6xxx_chip *chip, + int addr, int reg, u16 *val) +{ + int ret; + + /* Wait for the bus to become free. */ + ret = mv88e6xxx_smi_multi_chip_wait(chip); + if (ret < 0) + return ret; + + /* Transmit the read command. */ + ret = mdiobus_write_nested(chip->bus, chip->sw_addr, SMI_CMD, + SMI_CMD_OP_22_READ | (addr << 5) | reg); + if (ret < 0) + return ret; + + /* Wait for the read command to complete. */ + ret = mv88e6xxx_smi_multi_chip_wait(chip); + if (ret < 0) + return ret; + + /* Read the data. */ + ret = mdiobus_read_nested(chip->bus, chip->sw_addr, SMI_DATA); + if (ret < 0) + return ret; + + *val = ret & 0xffff; + + return 0; +} + +static int mv88e6xxx_smi_multi_chip_write(struct mv88e6xxx_chip *chip, + int addr, int reg, u16 val) +{ + int ret; + + /* Wait for the bus to become free. */ + ret = mv88e6xxx_smi_multi_chip_wait(chip); + if (ret < 0) + return ret; + + /* Transmit the data to write. */ + ret = mdiobus_write_nested(chip->bus, chip->sw_addr, SMI_DATA, val); + if (ret < 0) + return ret; + + /* Transmit the write command. */ + ret = mdiobus_write_nested(chip->bus, chip->sw_addr, SMI_CMD, + SMI_CMD_OP_22_WRITE | (addr << 5) | reg); + if (ret < 0) + return ret; + + /* Wait for the write command to complete. */ + ret = mv88e6xxx_smi_multi_chip_wait(chip); + if (ret < 0) + return ret; + + return 0; +} + +static const struct mv88e6xxx_ops mv88e6xxx_smi_multi_chip_ops = { + .read = mv88e6xxx_smi_multi_chip_read, + .write = mv88e6xxx_smi_multi_chip_write, +}; + +static int mv88e6xxx_read(struct mv88e6xxx_chip *chip, + int addr, int reg, u16 *val) +{ + int err; + + assert_reg_lock(chip); + + err = mv88e6xxx_smi_read(chip, addr, reg, val); + if (err) + return err; + + dev_dbg(chip->dev, "<- addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", + addr, reg, *val); + + return 0; +} + +static int mv88e6xxx_write(struct mv88e6xxx_chip *chip, + int addr, int reg, u16 val) +{ + int err; + + assert_reg_lock(chip); + + err = mv88e6xxx_smi_write(chip, addr, reg, val); + if (err) + return err; + + dev_dbg(chip->dev, "-> addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", + addr, reg, val); + + return 0; +} + +static int _mv88e6xxx_reg_read(struct mv88e6xxx_chip *chip, int addr, int reg) +{ + u16 val; + int err; + + err = mv88e6xxx_read(chip, addr, reg, &val); + if (err) + return err; + + return val; +} + +static int mv88e6xxx_reg_read(struct mv88e6xxx_chip *chip, int addr, int reg) +{ + int ret; + + mutex_lock(&chip->reg_lock); + ret = _mv88e6xxx_reg_read(chip, addr, reg); + mutex_unlock(&chip->reg_lock); + + return ret; +} + +static int _mv88e6xxx_reg_write(struct mv88e6xxx_chip *chip, int addr, + int reg, u16 val) +{ + return mv88e6xxx_write(chip, addr, reg, val); +} + +static int mv88e6xxx_reg_write(struct mv88e6xxx_chip *chip, int addr, + int reg, u16 val) +{ + int ret; + + mutex_lock(&chip->reg_lock); + ret = _mv88e6xxx_reg_write(chip, addr, reg, val); + mutex_unlock(&chip->reg_lock); + + return ret; +} + +static int mv88e6xxx_set_addr_direct(struct dsa_switch *ds, u8 *addr) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int err; + + err = mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_MAC_01, + (addr[0] << 8) | addr[1]); + if (err) + return err; + + err = mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_MAC_23, + (addr[2] << 8) | addr[3]); + if (err) + return err; + + return mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_MAC_45, + (addr[4] << 8) | addr[5]); +} + +static int mv88e6xxx_set_addr_indirect(struct dsa_switch *ds, u8 *addr) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int ret; + int i; + + for (i = 0; i < 6; i++) { + int j; + + /* Write the MAC address byte. */ + ret = mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_SWITCH_MAC, + GLOBAL2_SWITCH_MAC_BUSY | + (i << 8) | addr[i]); + if (ret) + return ret; + + /* Wait for the write to complete. */ + for (j = 0; j < 16; j++) { + ret = mv88e6xxx_reg_read(chip, REG_GLOBAL2, + GLOBAL2_SWITCH_MAC); + if (ret < 0) + return ret; + + if ((ret & GLOBAL2_SWITCH_MAC_BUSY) == 0) + break; + } + if (j == 16) + return -ETIMEDOUT; + } + + return 0; +} + +static int mv88e6xxx_set_addr(struct dsa_switch *ds, u8 *addr) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_SWITCH_MAC)) + return mv88e6xxx_set_addr_indirect(ds, addr); + else + return mv88e6xxx_set_addr_direct(ds, addr); +} + +static int mv88e6xxx_mdio_read_direct(struct mv88e6xxx_chip *chip, + int addr, int regnum) +{ + if (addr >= 0) + return _mv88e6xxx_reg_read(chip, addr, regnum); + return 0xffff; +} + +static int mv88e6xxx_mdio_write_direct(struct mv88e6xxx_chip *chip, + int addr, int regnum, u16 val) +{ + if (addr >= 0) + return _mv88e6xxx_reg_write(chip, addr, regnum, val); + return 0; +} + +static int mv88e6xxx_ppu_disable(struct mv88e6xxx_chip *chip) +{ + int ret; + unsigned long timeout; + + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_CONTROL); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_CONTROL, + ret & ~GLOBAL_CONTROL_PPU_ENABLE); + if (ret) + return ret; + + timeout = jiffies + 1 * HZ; + while (time_before(jiffies, timeout)) { + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_STATUS); + if (ret < 0) + return ret; + + usleep_range(1000, 2000); + if ((ret & GLOBAL_STATUS_PPU_MASK) != + GLOBAL_STATUS_PPU_POLLING) + return 0; + } + + return -ETIMEDOUT; +} + +static int mv88e6xxx_ppu_enable(struct mv88e6xxx_chip *chip) +{ + int ret, err; + unsigned long timeout; + + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_CONTROL); + if (ret < 0) + return ret; + + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_CONTROL, + ret | GLOBAL_CONTROL_PPU_ENABLE); + if (err) + return err; + + timeout = jiffies + 1 * HZ; + while (time_before(jiffies, timeout)) { + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_STATUS); + if (ret < 0) + return ret; + + usleep_range(1000, 2000); + if ((ret & GLOBAL_STATUS_PPU_MASK) == + GLOBAL_STATUS_PPU_POLLING) + return 0; + } + + return -ETIMEDOUT; +} + +static void mv88e6xxx_ppu_reenable_work(struct work_struct *ugly) +{ + struct mv88e6xxx_chip *chip; + + chip = container_of(ugly, struct mv88e6xxx_chip, ppu_work); + + mutex_lock(&chip->reg_lock); + + if (mutex_trylock(&chip->ppu_mutex)) { + if (mv88e6xxx_ppu_enable(chip) == 0) + chip->ppu_disabled = 0; + mutex_unlock(&chip->ppu_mutex); + } + + mutex_unlock(&chip->reg_lock); +} + +static void mv88e6xxx_ppu_reenable_timer(unsigned long _ps) +{ + struct mv88e6xxx_chip *chip = (void *)_ps; + + schedule_work(&chip->ppu_work); +} + +static int mv88e6xxx_ppu_access_get(struct mv88e6xxx_chip *chip) +{ + int ret; + + mutex_lock(&chip->ppu_mutex); + + /* If the PHY polling unit is enabled, disable it so that + * we can access the PHY registers. If it was already + * disabled, cancel the timer that is going to re-enable + * it. + */ + if (!chip->ppu_disabled) { + ret = mv88e6xxx_ppu_disable(chip); + if (ret < 0) { + mutex_unlock(&chip->ppu_mutex); + return ret; + } + chip->ppu_disabled = 1; + } else { + del_timer(&chip->ppu_timer); + ret = 0; + } + + return ret; +} + +static void mv88e6xxx_ppu_access_put(struct mv88e6xxx_chip *chip) +{ + /* Schedule a timer to re-enable the PHY polling unit. */ + mod_timer(&chip->ppu_timer, jiffies + msecs_to_jiffies(10)); + mutex_unlock(&chip->ppu_mutex); +} + +static void mv88e6xxx_ppu_state_init(struct mv88e6xxx_chip *chip) +{ + mutex_init(&chip->ppu_mutex); + INIT_WORK(&chip->ppu_work, mv88e6xxx_ppu_reenable_work); + init_timer(&chip->ppu_timer); + chip->ppu_timer.data = (unsigned long)chip; + chip->ppu_timer.function = mv88e6xxx_ppu_reenable_timer; +} + +static int mv88e6xxx_mdio_read_ppu(struct mv88e6xxx_chip *chip, int addr, + int regnum) +{ + int ret; + + ret = mv88e6xxx_ppu_access_get(chip); + if (ret >= 0) { + ret = _mv88e6xxx_reg_read(chip, addr, regnum); + mv88e6xxx_ppu_access_put(chip); + } + + return ret; +} + +static int mv88e6xxx_mdio_write_ppu(struct mv88e6xxx_chip *chip, int addr, + int regnum, u16 val) +{ + int ret; + + ret = mv88e6xxx_ppu_access_get(chip); + if (ret >= 0) { + ret = _mv88e6xxx_reg_write(chip, addr, regnum, val); + mv88e6xxx_ppu_access_put(chip); + } + + return ret; +} + +static bool mv88e6xxx_6065_family(struct mv88e6xxx_chip *chip) +{ + return chip->info->family == MV88E6XXX_FAMILY_6065; +} + +static bool mv88e6xxx_6095_family(struct mv88e6xxx_chip *chip) +{ + return chip->info->family == MV88E6XXX_FAMILY_6095; +} + +static bool mv88e6xxx_6097_family(struct mv88e6xxx_chip *chip) +{ + return chip->info->family == MV88E6XXX_FAMILY_6097; +} + +static bool mv88e6xxx_6165_family(struct mv88e6xxx_chip *chip) +{ + return chip->info->family == MV88E6XXX_FAMILY_6165; +} + +static bool mv88e6xxx_6185_family(struct mv88e6xxx_chip *chip) +{ + return chip->info->family == MV88E6XXX_FAMILY_6185; +} + +static bool mv88e6xxx_6320_family(struct mv88e6xxx_chip *chip) +{ + return chip->info->family == MV88E6XXX_FAMILY_6320; +} + +static bool mv88e6xxx_6351_family(struct mv88e6xxx_chip *chip) +{ + return chip->info->family == MV88E6XXX_FAMILY_6351; +} + +static bool mv88e6xxx_6352_family(struct mv88e6xxx_chip *chip) +{ + return chip->info->family == MV88E6XXX_FAMILY_6352; +} + +static unsigned int mv88e6xxx_num_databases(struct mv88e6xxx_chip *chip) +{ + return chip->info->num_databases; +} + +static bool mv88e6xxx_has_fid_reg(struct mv88e6xxx_chip *chip) +{ + /* Does the device have dedicated FID registers for ATU and VTU ops? */ + if (mv88e6xxx_6097_family(chip) || mv88e6xxx_6165_family(chip) || + mv88e6xxx_6351_family(chip) || mv88e6xxx_6352_family(chip)) + return true; + + return false; +} + +/* We expect the switch to perform auto negotiation if there is a real + * phy. However, in the case of a fixed link phy, we force the port + * settings from the fixed link settings. + */ +static void mv88e6xxx_adjust_link(struct dsa_switch *ds, int port, + struct phy_device *phydev) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + u32 reg; + int ret; + + if (!phy_is_pseudo_fixed_link(phydev)) + return; + + mutex_lock(&chip->reg_lock); + + ret = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_PCS_CTRL); + if (ret < 0) + goto out; + + reg = ret & ~(PORT_PCS_CTRL_LINK_UP | + PORT_PCS_CTRL_FORCE_LINK | + PORT_PCS_CTRL_DUPLEX_FULL | + PORT_PCS_CTRL_FORCE_DUPLEX | + PORT_PCS_CTRL_UNFORCED); + + reg |= PORT_PCS_CTRL_FORCE_LINK; + if (phydev->link) + reg |= PORT_PCS_CTRL_LINK_UP; + + if (mv88e6xxx_6065_family(chip) && phydev->speed > SPEED_100) + goto out; + + switch (phydev->speed) { + case SPEED_1000: + reg |= PORT_PCS_CTRL_1000; + break; + case SPEED_100: + reg |= PORT_PCS_CTRL_100; + break; + case SPEED_10: + reg |= PORT_PCS_CTRL_10; + break; + default: + pr_info("Unknown speed"); + goto out; + } + + reg |= PORT_PCS_CTRL_FORCE_DUPLEX; + if (phydev->duplex == DUPLEX_FULL) + reg |= PORT_PCS_CTRL_DUPLEX_FULL; + + if ((mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip)) && + (port >= chip->info->num_ports - 2)) { + if (phydev->interface == PHY_INTERFACE_MODE_RGMII_RXID) + reg |= PORT_PCS_CTRL_RGMII_DELAY_RXCLK; + if (phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID) + reg |= PORT_PCS_CTRL_RGMII_DELAY_TXCLK; + if (phydev->interface == PHY_INTERFACE_MODE_RGMII_ID) + reg |= (PORT_PCS_CTRL_RGMII_DELAY_RXCLK | + PORT_PCS_CTRL_RGMII_DELAY_TXCLK); + } + _mv88e6xxx_reg_write(chip, REG_PORT(port), PORT_PCS_CTRL, reg); + +out: + mutex_unlock(&chip->reg_lock); +} + +static int _mv88e6xxx_stats_wait(struct mv88e6xxx_chip *chip) +{ + int ret; + int i; + + for (i = 0; i < 10; i++) { + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_STATS_OP); + if ((ret & GLOBAL_STATS_OP_BUSY) == 0) + return 0; + } + + return -ETIMEDOUT; +} + +static int _mv88e6xxx_stats_snapshot(struct mv88e6xxx_chip *chip, int port) +{ + int ret; + + if (mv88e6xxx_6320_family(chip) || mv88e6xxx_6352_family(chip)) + port = (port + 1) << 5; + + /* Snapshot the hardware statistics counters for this port. */ + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_STATS_OP, + GLOBAL_STATS_OP_CAPTURE_PORT | + GLOBAL_STATS_OP_HIST_RX_TX | port); + if (ret < 0) + return ret; + + /* Wait for the snapshotting to complete. */ + ret = _mv88e6xxx_stats_wait(chip); + if (ret < 0) + return ret; + + return 0; +} + +static void _mv88e6xxx_stats_read(struct mv88e6xxx_chip *chip, + int stat, u32 *val) +{ + u32 _val; + int ret; + + *val = 0; + + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_STATS_OP, + GLOBAL_STATS_OP_READ_CAPTURED | + GLOBAL_STATS_OP_HIST_RX_TX | stat); + if (ret < 0) + return; + + ret = _mv88e6xxx_stats_wait(chip); + if (ret < 0) + return; + + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_STATS_COUNTER_32); + if (ret < 0) + return; + + _val = ret << 16; + + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_STATS_COUNTER_01); + if (ret < 0) + return; + + *val = _val | ret; +} + +static struct mv88e6xxx_hw_stat mv88e6xxx_hw_stats[] = { + { "in_good_octets", 8, 0x00, BANK0, }, + { "in_bad_octets", 4, 0x02, BANK0, }, + { "in_unicast", 4, 0x04, BANK0, }, + { "in_broadcasts", 4, 0x06, BANK0, }, + { "in_multicasts", 4, 0x07, BANK0, }, + { "in_pause", 4, 0x16, BANK0, }, + { "in_undersize", 4, 0x18, BANK0, }, + { "in_fragments", 4, 0x19, BANK0, }, + { "in_oversize", 4, 0x1a, BANK0, }, + { "in_jabber", 4, 0x1b, BANK0, }, + { "in_rx_error", 4, 0x1c, BANK0, }, + { "in_fcs_error", 4, 0x1d, BANK0, }, + { "out_octets", 8, 0x0e, BANK0, }, + { "out_unicast", 4, 0x10, BANK0, }, + { "out_broadcasts", 4, 0x13, BANK0, }, + { "out_multicasts", 4, 0x12, BANK0, }, + { "out_pause", 4, 0x15, BANK0, }, + { "excessive", 4, 0x11, BANK0, }, + { "collisions", 4, 0x1e, BANK0, }, + { "deferred", 4, 0x05, BANK0, }, + { "single", 4, 0x14, BANK0, }, + { "multiple", 4, 0x17, BANK0, }, + { "out_fcs_error", 4, 0x03, BANK0, }, + { "late", 4, 0x1f, BANK0, }, + { "hist_64bytes", 4, 0x08, BANK0, }, + { "hist_65_127bytes", 4, 0x09, BANK0, }, + { "hist_128_255bytes", 4, 0x0a, BANK0, }, + { "hist_256_511bytes", 4, 0x0b, BANK0, }, + { "hist_512_1023bytes", 4, 0x0c, BANK0, }, + { "hist_1024_max_bytes", 4, 0x0d, BANK0, }, + { "sw_in_discards", 4, 0x10, PORT, }, + { "sw_in_filtered", 2, 0x12, PORT, }, + { "sw_out_filtered", 2, 0x13, PORT, }, + { "in_discards", 4, 0x00 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_filtered", 4, 0x01 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_accepted", 4, 0x02 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_bad_accepted", 4, 0x03 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_good_avb_class_a", 4, 0x04 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_good_avb_class_b", 4, 0x05 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_bad_avb_class_a", 4, 0x06 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_bad_avb_class_b", 4, 0x07 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "tcam_counter_0", 4, 0x08 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "tcam_counter_1", 4, 0x09 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "tcam_counter_2", 4, 0x0a | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "tcam_counter_3", 4, 0x0b | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_da_unknown", 4, 0x0e | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "in_management", 4, 0x0f | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_0", 4, 0x10 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_1", 4, 0x11 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_2", 4, 0x12 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_3", 4, 0x13 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_4", 4, 0x14 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_5", 4, 0x15 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_6", 4, 0x16 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_queue_7", 4, 0x17 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_cut_through", 4, 0x18 | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_octets_a", 4, 0x1a | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_octets_b", 4, 0x1b | GLOBAL_STATS_OP_BANK_1, BANK1, }, + { "out_management", 4, 0x1f | GLOBAL_STATS_OP_BANK_1, BANK1, }, +}; + +static bool mv88e6xxx_has_stat(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_hw_stat *stat) +{ + switch (stat->type) { + case BANK0: + return true; + case BANK1: + return mv88e6xxx_6320_family(chip); + case PORT: + return mv88e6xxx_6095_family(chip) || + mv88e6xxx_6185_family(chip) || + mv88e6xxx_6097_family(chip) || + mv88e6xxx_6165_family(chip) || + mv88e6xxx_6351_family(chip) || + mv88e6xxx_6352_family(chip); + } + return false; +} + +static uint64_t _mv88e6xxx_get_ethtool_stat(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_hw_stat *s, + int port) +{ + u32 low; + u32 high = 0; + int ret; + u64 value; + + switch (s->type) { + case PORT: + ret = _mv88e6xxx_reg_read(chip, REG_PORT(port), s->reg); + if (ret < 0) + return UINT64_MAX; + + low = ret; + if (s->sizeof_stat == 4) { + ret = _mv88e6xxx_reg_read(chip, REG_PORT(port), + s->reg + 1); + if (ret < 0) + return UINT64_MAX; + high = ret; + } + break; + case BANK0: + case BANK1: + _mv88e6xxx_stats_read(chip, s->reg, &low); + if (s->sizeof_stat == 8) + _mv88e6xxx_stats_read(chip, s->reg + 1, &high); + } + value = (((u64)high) << 16) | low; + return value; +} + +static void mv88e6xxx_get_strings(struct dsa_switch *ds, int port, + uint8_t *data) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + struct mv88e6xxx_hw_stat *stat; + int i, j; + + for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { + stat = &mv88e6xxx_hw_stats[i]; + if (mv88e6xxx_has_stat(chip, stat)) { + memcpy(data + j * ETH_GSTRING_LEN, stat->string, + ETH_GSTRING_LEN); + j++; + } + } +} + +static int mv88e6xxx_get_sset_count(struct dsa_switch *ds) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + struct mv88e6xxx_hw_stat *stat; + int i, j; + + for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { + stat = &mv88e6xxx_hw_stats[i]; + if (mv88e6xxx_has_stat(chip, stat)) + j++; + } + return j; +} + +static void mv88e6xxx_get_ethtool_stats(struct dsa_switch *ds, int port, + uint64_t *data) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + struct mv88e6xxx_hw_stat *stat; + int ret; + int i, j; + + mutex_lock(&chip->reg_lock); + + ret = _mv88e6xxx_stats_snapshot(chip, port); + if (ret < 0) { + mutex_unlock(&chip->reg_lock); + return; + } + for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { + stat = &mv88e6xxx_hw_stats[i]; + if (mv88e6xxx_has_stat(chip, stat)) { + data[j] = _mv88e6xxx_get_ethtool_stat(chip, stat, port); + j++; + } + } + + mutex_unlock(&chip->reg_lock); +} + +static int mv88e6xxx_get_regs_len(struct dsa_switch *ds, int port) +{ + return 32 * sizeof(u16); +} + +static void mv88e6xxx_get_regs(struct dsa_switch *ds, int port, + struct ethtool_regs *regs, void *_p) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + u16 *p = _p; + int i; + + regs->version = 0; + + memset(p, 0xff, 32 * sizeof(u16)); + + mutex_lock(&chip->reg_lock); + + for (i = 0; i < 32; i++) { + int ret; + + ret = _mv88e6xxx_reg_read(chip, REG_PORT(port), i); + if (ret >= 0) + p[i] = ret; + } + + mutex_unlock(&chip->reg_lock); +} + +static int _mv88e6xxx_wait(struct mv88e6xxx_chip *chip, int reg, int offset, + u16 mask) +{ + unsigned long timeout = jiffies + HZ / 10; + + while (time_before(jiffies, timeout)) { + int ret; + + ret = _mv88e6xxx_reg_read(chip, reg, offset); + if (ret < 0) + return ret; + if (!(ret & mask)) + return 0; + + usleep_range(1000, 2000); + } + return -ETIMEDOUT; +} + +static int mv88e6xxx_wait(struct mv88e6xxx_chip *chip, int reg, + int offset, u16 mask) +{ + int ret; + + mutex_lock(&chip->reg_lock); + ret = _mv88e6xxx_wait(chip, reg, offset, mask); + mutex_unlock(&chip->reg_lock); + + return ret; +} + +static int mv88e6xxx_mdio_wait(struct mv88e6xxx_chip *chip) +{ + return _mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_SMI_OP, + GLOBAL2_SMI_OP_BUSY); +} + +static int mv88e6xxx_eeprom_load_wait(struct dsa_switch *ds) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + + return mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_EEPROM_OP, + GLOBAL2_EEPROM_OP_LOAD); +} + +static int mv88e6xxx_eeprom_busy_wait(struct dsa_switch *ds) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + + return mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_EEPROM_OP, + GLOBAL2_EEPROM_OP_BUSY); +} + +static int mv88e6xxx_read_eeprom_word(struct dsa_switch *ds, int addr) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int ret; + + mutex_lock(&chip->eeprom_mutex); + + ret = mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_EEPROM_OP, + GLOBAL2_EEPROM_OP_READ | + (addr & GLOBAL2_EEPROM_OP_ADDR_MASK)); + if (ret < 0) + goto error; + + ret = mv88e6xxx_eeprom_busy_wait(ds); + if (ret < 0) + goto error; + + ret = mv88e6xxx_reg_read(chip, REG_GLOBAL2, GLOBAL2_EEPROM_DATA); +error: + mutex_unlock(&chip->eeprom_mutex); + return ret; +} + +static int mv88e6xxx_get_eeprom_len(struct dsa_switch *ds) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEPROM)) + return chip->eeprom_len; + + return 0; +} + +static int mv88e6xxx_get_eeprom(struct dsa_switch *ds, + struct ethtool_eeprom *eeprom, u8 *data) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int offset; + int len; + int ret; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEPROM)) + return -EOPNOTSUPP; + + offset = eeprom->offset; + len = eeprom->len; + eeprom->len = 0; + + eeprom->magic = 0xc3ec4951; + + ret = mv88e6xxx_eeprom_load_wait(ds); + if (ret < 0) + return ret; + + if (offset & 1) { + int word; + + word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); + if (word < 0) + return word; + + *data++ = (word >> 8) & 0xff; + + offset++; + len--; + eeprom->len++; + } + + while (len >= 2) { + int word; + + word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); + if (word < 0) + return word; + + *data++ = word & 0xff; + *data++ = (word >> 8) & 0xff; + + offset += 2; + len -= 2; + eeprom->len += 2; + } + + if (len) { + int word; + + word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); + if (word < 0) + return word; + + *data++ = word & 0xff; + + offset++; + len--; + eeprom->len++; + } + + return 0; +} + +static int mv88e6xxx_eeprom_is_readonly(struct dsa_switch *ds) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int ret; + + ret = mv88e6xxx_reg_read(chip, REG_GLOBAL2, GLOBAL2_EEPROM_OP); + if (ret < 0) + return ret; + + if (!(ret & GLOBAL2_EEPROM_OP_WRITE_EN)) + return -EROFS; + + return 0; +} + +static int mv88e6xxx_write_eeprom_word(struct dsa_switch *ds, int addr, + u16 data) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int ret; + + mutex_lock(&chip->eeprom_mutex); + + ret = mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_EEPROM_DATA, data); + if (ret < 0) + goto error; + + ret = mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_EEPROM_OP, + GLOBAL2_EEPROM_OP_WRITE | + (addr & GLOBAL2_EEPROM_OP_ADDR_MASK)); + if (ret < 0) + goto error; + + ret = mv88e6xxx_eeprom_busy_wait(ds); +error: + mutex_unlock(&chip->eeprom_mutex); + return ret; +} + +static int mv88e6xxx_set_eeprom(struct dsa_switch *ds, + struct ethtool_eeprom *eeprom, u8 *data) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int offset; + int ret; + int len; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEPROM)) + return -EOPNOTSUPP; + + if (eeprom->magic != 0xc3ec4951) + return -EINVAL; + + ret = mv88e6xxx_eeprom_is_readonly(ds); + if (ret) + return ret; + + offset = eeprom->offset; + len = eeprom->len; + eeprom->len = 0; + + ret = mv88e6xxx_eeprom_load_wait(ds); + if (ret < 0) + return ret; + + if (offset & 1) { + int word; + + word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); + if (word < 0) + return word; + + word = (*data++ << 8) | (word & 0xff); + + ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); + if (ret < 0) + return ret; + + offset++; + len--; + eeprom->len++; + } + + while (len >= 2) { + int word; + + word = *data++; + word |= *data++ << 8; + + ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); + if (ret < 0) + return ret; + + offset += 2; + len -= 2; + eeprom->len += 2; + } + + if (len) { + int word; + + word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); + if (word < 0) + return word; + + word = (word & 0xff00) | *data++; + + ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); + if (ret < 0) + return ret; + + offset++; + len--; + eeprom->len++; + } + + return 0; +} + +static int _mv88e6xxx_atu_wait(struct mv88e6xxx_chip *chip) +{ + return _mv88e6xxx_wait(chip, REG_GLOBAL, GLOBAL_ATU_OP, + GLOBAL_ATU_OP_BUSY); +} + +static int mv88e6xxx_mdio_read_indirect(struct mv88e6xxx_chip *chip, + int addr, int regnum) +{ + int ret; + + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_SMI_OP, + GLOBAL2_SMI_OP_22_READ | (addr << 5) | + regnum); + if (ret < 0) + return ret; + + ret = mv88e6xxx_mdio_wait(chip); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL2, GLOBAL2_SMI_DATA); + + return ret; +} + +static int mv88e6xxx_mdio_write_indirect(struct mv88e6xxx_chip *chip, + int addr, int regnum, u16 val) +{ + int ret; + + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_SMI_DATA, val); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_SMI_OP, + GLOBAL2_SMI_OP_22_WRITE | (addr << 5) | + regnum); + + return mv88e6xxx_mdio_wait(chip); +} + +static int mv88e6xxx_get_eee(struct dsa_switch *ds, int port, + struct ethtool_eee *e) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int reg; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEE)) + return -EOPNOTSUPP; + + mutex_lock(&chip->reg_lock); + + reg = mv88e6xxx_mdio_read_indirect(chip, port, 16); + if (reg < 0) + goto out; + + e->eee_enabled = !!(reg & 0x0200); + e->tx_lpi_enabled = !!(reg & 0x0100); + + reg = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_STATUS); + if (reg < 0) + goto out; + + e->eee_active = !!(reg & PORT_STATUS_EEE); + reg = 0; + +out: + mutex_unlock(&chip->reg_lock); + return reg; +} + +static int mv88e6xxx_set_eee(struct dsa_switch *ds, int port, + struct phy_device *phydev, struct ethtool_eee *e) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int reg; + int ret; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEE)) + return -EOPNOTSUPP; + + mutex_lock(&chip->reg_lock); + + ret = mv88e6xxx_mdio_read_indirect(chip, port, 16); + if (ret < 0) + goto out; + + reg = ret & ~0x0300; + if (e->eee_enabled) + reg |= 0x0200; + if (e->tx_lpi_enabled) + reg |= 0x0100; + + ret = mv88e6xxx_mdio_write_indirect(chip, port, 16, reg); +out: + mutex_unlock(&chip->reg_lock); + + return ret; +} + +static int _mv88e6xxx_atu_cmd(struct mv88e6xxx_chip *chip, u16 fid, u16 cmd) +{ + int ret; + + if (mv88e6xxx_has_fid_reg(chip)) { + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_ATU_FID, + fid); + if (ret < 0) + return ret; + } else if (mv88e6xxx_num_databases(chip) == 256) { + /* ATU DBNum[7:4] are located in ATU Control 15:12 */ + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_ATU_CONTROL); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_ATU_CONTROL, + (ret & 0xfff) | + ((fid << 8) & 0xf000)); + if (ret < 0) + return ret; + + /* ATU DBNum[3:0] are located in ATU Operation 3:0 */ + cmd |= fid & 0xf; + } + + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_ATU_OP, cmd); + if (ret < 0) + return ret; + + return _mv88e6xxx_atu_wait(chip); +} + +static int _mv88e6xxx_atu_data_write(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_atu_entry *entry) +{ + u16 data = entry->state & GLOBAL_ATU_DATA_STATE_MASK; + + if (entry->state != GLOBAL_ATU_DATA_STATE_UNUSED) { + unsigned int mask, shift; + + if (entry->trunk) { + data |= GLOBAL_ATU_DATA_TRUNK; + mask = GLOBAL_ATU_DATA_TRUNK_ID_MASK; + shift = GLOBAL_ATU_DATA_TRUNK_ID_SHIFT; + } else { + mask = GLOBAL_ATU_DATA_PORT_VECTOR_MASK; + shift = GLOBAL_ATU_DATA_PORT_VECTOR_SHIFT; + } + + data |= (entry->portv_trunkid << shift) & mask; + } + + return _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_ATU_DATA, data); +} + +static int _mv88e6xxx_atu_flush_move(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_atu_entry *entry, + bool static_too) +{ + int op; + int err; + + err = _mv88e6xxx_atu_wait(chip); + if (err) + return err; + + err = _mv88e6xxx_atu_data_write(chip, entry); + if (err) + return err; + + if (entry->fid) { + op = static_too ? GLOBAL_ATU_OP_FLUSH_MOVE_ALL_DB : + GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC_DB; + } else { + op = static_too ? GLOBAL_ATU_OP_FLUSH_MOVE_ALL : + GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC; + } + + return _mv88e6xxx_atu_cmd(chip, entry->fid, op); +} + +static int _mv88e6xxx_atu_flush(struct mv88e6xxx_chip *chip, + u16 fid, bool static_too) +{ + struct mv88e6xxx_atu_entry entry = { + .fid = fid, + .state = 0, /* EntryState bits must be 0 */ + }; + + return _mv88e6xxx_atu_flush_move(chip, &entry, static_too); +} + +static int _mv88e6xxx_atu_move(struct mv88e6xxx_chip *chip, u16 fid, + int from_port, int to_port, bool static_too) +{ + struct mv88e6xxx_atu_entry entry = { + .trunk = false, + .fid = fid, + }; + + /* EntryState bits must be 0xF */ + entry.state = GLOBAL_ATU_DATA_STATE_MASK; + + /* ToPort and FromPort are respectively in PortVec bits 7:4 and 3:0 */ + entry.portv_trunkid = (to_port & 0x0f) << 4; + entry.portv_trunkid |= from_port & 0x0f; + + return _mv88e6xxx_atu_flush_move(chip, &entry, static_too); +} + +static int _mv88e6xxx_atu_remove(struct mv88e6xxx_chip *chip, u16 fid, + int port, bool static_too) +{ + /* Destination port 0xF means remove the entries */ + return _mv88e6xxx_atu_move(chip, fid, port, 0x0f, static_too); +} + +static const char * const mv88e6xxx_port_state_names[] = { + [PORT_CONTROL_STATE_DISABLED] = "Disabled", + [PORT_CONTROL_STATE_BLOCKING] = "Blocking/Listening", + [PORT_CONTROL_STATE_LEARNING] = "Learning", + [PORT_CONTROL_STATE_FORWARDING] = "Forwarding", +}; + +static int _mv88e6xxx_port_state(struct mv88e6xxx_chip *chip, int port, + u8 state) +{ + struct dsa_switch *ds = chip->ds; + int reg, ret = 0; + u8 oldstate; + + reg = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_CONTROL); + if (reg < 0) + return reg; + + oldstate = reg & PORT_CONTROL_STATE_MASK; + + if (oldstate != state) { + /* Flush forwarding database if we're moving a port + * from Learning or Forwarding state to Disabled or + * Blocking or Listening state. + */ + if ((oldstate == PORT_CONTROL_STATE_LEARNING || + oldstate == PORT_CONTROL_STATE_FORWARDING) && + (state == PORT_CONTROL_STATE_DISABLED || + state == PORT_CONTROL_STATE_BLOCKING)) { + ret = _mv88e6xxx_atu_remove(chip, 0, port, false); + if (ret) + return ret; + } + + reg = (reg & ~PORT_CONTROL_STATE_MASK) | state; + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), PORT_CONTROL, + reg); + if (ret) + return ret; + + netdev_dbg(ds->ports[port].netdev, "PortState %s (was %s)\n", + mv88e6xxx_port_state_names[state], + mv88e6xxx_port_state_names[oldstate]); + } + + return ret; +} + +static int _mv88e6xxx_port_based_vlan_map(struct mv88e6xxx_chip *chip, int port) +{ + struct net_device *bridge = chip->ports[port].bridge_dev; + const u16 mask = (1 << chip->info->num_ports) - 1; + struct dsa_switch *ds = chip->ds; + u16 output_ports = 0; + int reg; + int i; + + /* allow CPU port or DSA link(s) to send frames to every port */ + if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) { + output_ports = mask; + } else { + for (i = 0; i < chip->info->num_ports; ++i) { + /* allow sending frames to every group member */ + if (bridge && chip->ports[i].bridge_dev == bridge) + output_ports |= BIT(i); + + /* allow sending frames to CPU port and DSA link(s) */ + if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i)) + output_ports |= BIT(i); + } + } + + /* prevent frames from going back out of the port they came in on */ + output_ports &= ~BIT(port); + + reg = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_BASE_VLAN); + if (reg < 0) + return reg; + + reg &= ~mask; + reg |= output_ports & mask; + + return _mv88e6xxx_reg_write(chip, REG_PORT(port), PORT_BASE_VLAN, reg); +} + +static void mv88e6xxx_port_stp_state_set(struct dsa_switch *ds, int port, + u8 state) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int stp_state; + int err; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_PORTSTATE)) + return; + + switch (state) { + case BR_STATE_DISABLED: + stp_state = PORT_CONTROL_STATE_DISABLED; + break; + case BR_STATE_BLOCKING: + case BR_STATE_LISTENING: + stp_state = PORT_CONTROL_STATE_BLOCKING; + break; + case BR_STATE_LEARNING: + stp_state = PORT_CONTROL_STATE_LEARNING; + break; + case BR_STATE_FORWARDING: + default: + stp_state = PORT_CONTROL_STATE_FORWARDING; + break; + } + + mutex_lock(&chip->reg_lock); + err = _mv88e6xxx_port_state(chip, port, stp_state); + mutex_unlock(&chip->reg_lock); + + if (err) + netdev_err(ds->ports[port].netdev, + "failed to update state to %s\n", + mv88e6xxx_port_state_names[stp_state]); +} + +static int _mv88e6xxx_port_pvid(struct mv88e6xxx_chip *chip, int port, + u16 *new, u16 *old) +{ + struct dsa_switch *ds = chip->ds; + u16 pvid; + int ret; + + ret = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_DEFAULT_VLAN); + if (ret < 0) + return ret; + + pvid = ret & PORT_DEFAULT_VLAN_MASK; + + if (new) { + ret &= ~PORT_DEFAULT_VLAN_MASK; + ret |= *new & PORT_DEFAULT_VLAN_MASK; + + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_DEFAULT_VLAN, ret); + if (ret < 0) + return ret; + + netdev_dbg(ds->ports[port].netdev, + "DefaultVID %d (was %d)\n", *new, pvid); + } + + if (old) + *old = pvid; + + return 0; +} + +static int _mv88e6xxx_port_pvid_get(struct mv88e6xxx_chip *chip, + int port, u16 *pvid) +{ + return _mv88e6xxx_port_pvid(chip, port, NULL, pvid); +} + +static int _mv88e6xxx_port_pvid_set(struct mv88e6xxx_chip *chip, + int port, u16 pvid) +{ + return _mv88e6xxx_port_pvid(chip, port, &pvid, NULL); +} + +static int _mv88e6xxx_vtu_wait(struct mv88e6xxx_chip *chip) +{ + return _mv88e6xxx_wait(chip, REG_GLOBAL, GLOBAL_VTU_OP, + GLOBAL_VTU_OP_BUSY); +} + +static int _mv88e6xxx_vtu_cmd(struct mv88e6xxx_chip *chip, u16 op) +{ + int ret; + + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_VTU_OP, op); + if (ret < 0) + return ret; + + return _mv88e6xxx_vtu_wait(chip); +} + +static int _mv88e6xxx_vtu_stu_flush(struct mv88e6xxx_chip *chip) +{ + int ret; + + ret = _mv88e6xxx_vtu_wait(chip); + if (ret < 0) + return ret; + + return _mv88e6xxx_vtu_cmd(chip, GLOBAL_VTU_OP_FLUSH_ALL); +} + +static int _mv88e6xxx_vtu_stu_data_read(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_vtu_stu_entry *entry, + unsigned int nibble_offset) +{ + u16 regs[3]; + int i; + int ret; + + for (i = 0; i < 3; ++i) { + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, + GLOBAL_VTU_DATA_0_3 + i); + if (ret < 0) + return ret; + + regs[i] = ret; + } + + for (i = 0; i < chip->info->num_ports; ++i) { + unsigned int shift = (i % 4) * 4 + nibble_offset; + u16 reg = regs[i / 4]; + + entry->data[i] = (reg >> shift) & GLOBAL_VTU_STU_DATA_MASK; + } + + return 0; +} + +static int mv88e6xxx_vtu_data_read(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + return _mv88e6xxx_vtu_stu_data_read(chip, entry, 0); +} + +static int mv88e6xxx_stu_data_read(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + return _mv88e6xxx_vtu_stu_data_read(chip, entry, 2); +} + +static int _mv88e6xxx_vtu_stu_data_write(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_vtu_stu_entry *entry, + unsigned int nibble_offset) +{ + u16 regs[3] = { 0 }; + int i; + int ret; + + for (i = 0; i < chip->info->num_ports; ++i) { + unsigned int shift = (i % 4) * 4 + nibble_offset; + u8 data = entry->data[i]; + + regs[i / 4] |= (data & GLOBAL_VTU_STU_DATA_MASK) << shift; + } + + for (i = 0; i < 3; ++i) { + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, + GLOBAL_VTU_DATA_0_3 + i, regs[i]); + if (ret < 0) + return ret; + } + + return 0; +} + +static int mv88e6xxx_vtu_data_write(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + return _mv88e6xxx_vtu_stu_data_write(chip, entry, 0); +} + +static int mv88e6xxx_stu_data_write(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + return _mv88e6xxx_vtu_stu_data_write(chip, entry, 2); +} + +static int _mv88e6xxx_vtu_vid_write(struct mv88e6xxx_chip *chip, u16 vid) +{ + return _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_VTU_VID, + vid & GLOBAL_VTU_VID_MASK); +} + +static int _mv88e6xxx_vtu_getnext(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + struct mv88e6xxx_vtu_stu_entry next = { 0 }; + int ret; + + ret = _mv88e6xxx_vtu_wait(chip); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_vtu_cmd(chip, GLOBAL_VTU_OP_VTU_GET_NEXT); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_VTU_VID); + if (ret < 0) + return ret; + + next.vid = ret & GLOBAL_VTU_VID_MASK; + next.valid = !!(ret & GLOBAL_VTU_VID_VALID); + + if (next.valid) { + ret = mv88e6xxx_vtu_data_read(chip, &next); + if (ret < 0) + return ret; + + if (mv88e6xxx_has_fid_reg(chip)) { + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, + GLOBAL_VTU_FID); + if (ret < 0) + return ret; + + next.fid = ret & GLOBAL_VTU_FID_MASK; + } else if (mv88e6xxx_num_databases(chip) == 256) { + /* VTU DBNum[7:4] are located in VTU Operation 11:8, and + * VTU DBNum[3:0] are located in VTU Operation 3:0 + */ + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, + GLOBAL_VTU_OP); + if (ret < 0) + return ret; + + next.fid = (ret & 0xf00) >> 4; + next.fid |= ret & 0xf; + } + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_STU)) { + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, + GLOBAL_VTU_SID); + if (ret < 0) + return ret; + + next.sid = ret & GLOBAL_VTU_SID_MASK; + } + } + + *entry = next; + return 0; +} + +static int mv88e6xxx_port_vlan_dump(struct dsa_switch *ds, int port, + struct switchdev_obj_port_vlan *vlan, + int (*cb)(struct switchdev_obj *obj)) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + struct mv88e6xxx_vtu_stu_entry next; + u16 pvid; + int err; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_VTU)) + return -EOPNOTSUPP; + + mutex_lock(&chip->reg_lock); + + err = _mv88e6xxx_port_pvid_get(chip, port, &pvid); + if (err) + goto unlock; + + err = _mv88e6xxx_vtu_vid_write(chip, GLOBAL_VTU_VID_MASK); + if (err) + goto unlock; + + do { + err = _mv88e6xxx_vtu_getnext(chip, &next); + if (err) + break; + + if (!next.valid) + break; + + if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) + continue; + + /* reinit and dump this VLAN obj */ + vlan->vid_begin = next.vid; + vlan->vid_end = next.vid; + vlan->flags = 0; + + if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED) + vlan->flags |= BRIDGE_VLAN_INFO_UNTAGGED; + + if (next.vid == pvid) + vlan->flags |= BRIDGE_VLAN_INFO_PVID; + + err = cb(&vlan->obj); + if (err) + break; + } while (next.vid < GLOBAL_VTU_VID_MASK); + +unlock: + mutex_unlock(&chip->reg_lock); + + return err; +} + +static int _mv88e6xxx_vtu_loadpurge(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + u16 op = GLOBAL_VTU_OP_VTU_LOAD_PURGE; + u16 reg = 0; + int ret; + + ret = _mv88e6xxx_vtu_wait(chip); + if (ret < 0) + return ret; + + if (!entry->valid) + goto loadpurge; + + /* Write port member tags */ + ret = mv88e6xxx_vtu_data_write(chip, entry); + if (ret < 0) + return ret; + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_STU)) { + reg = entry->sid & GLOBAL_VTU_SID_MASK; + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_VTU_SID, + reg); + if (ret < 0) + return ret; + } + + if (mv88e6xxx_has_fid_reg(chip)) { + reg = entry->fid & GLOBAL_VTU_FID_MASK; + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_VTU_FID, + reg); + if (ret < 0) + return ret; + } else if (mv88e6xxx_num_databases(chip) == 256) { + /* VTU DBNum[7:4] are located in VTU Operation 11:8, and + * VTU DBNum[3:0] are located in VTU Operation 3:0 + */ + op |= (entry->fid & 0xf0) << 8; + op |= entry->fid & 0xf; + } + + reg = GLOBAL_VTU_VID_VALID; +loadpurge: + reg |= entry->vid & GLOBAL_VTU_VID_MASK; + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_VTU_VID, reg); + if (ret < 0) + return ret; + + return _mv88e6xxx_vtu_cmd(chip, op); +} + +static int _mv88e6xxx_stu_getnext(struct mv88e6xxx_chip *chip, u8 sid, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + struct mv88e6xxx_vtu_stu_entry next = { 0 }; + int ret; + + ret = _mv88e6xxx_vtu_wait(chip); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_VTU_SID, + sid & GLOBAL_VTU_SID_MASK); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_vtu_cmd(chip, GLOBAL_VTU_OP_STU_GET_NEXT); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_VTU_SID); + if (ret < 0) + return ret; + + next.sid = ret & GLOBAL_VTU_SID_MASK; + + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_VTU_VID); + if (ret < 0) + return ret; + + next.valid = !!(ret & GLOBAL_VTU_VID_VALID); + + if (next.valid) { + ret = mv88e6xxx_stu_data_read(chip, &next); + if (ret < 0) + return ret; + } + + *entry = next; + return 0; +} + +static int _mv88e6xxx_stu_loadpurge(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + u16 reg = 0; + int ret; + + ret = _mv88e6xxx_vtu_wait(chip); + if (ret < 0) + return ret; + + if (!entry->valid) + goto loadpurge; + + /* Write port states */ + ret = mv88e6xxx_stu_data_write(chip, entry); + if (ret < 0) + return ret; + + reg = GLOBAL_VTU_VID_VALID; +loadpurge: + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_VTU_VID, reg); + if (ret < 0) + return ret; + + reg = entry->sid & GLOBAL_VTU_SID_MASK; + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_VTU_SID, reg); + if (ret < 0) + return ret; + + return _mv88e6xxx_vtu_cmd(chip, GLOBAL_VTU_OP_STU_LOAD_PURGE); +} + +static int _mv88e6xxx_port_fid(struct mv88e6xxx_chip *chip, int port, + u16 *new, u16 *old) +{ + struct dsa_switch *ds = chip->ds; + u16 upper_mask; + u16 fid; + int ret; + + if (mv88e6xxx_num_databases(chip) == 4096) + upper_mask = 0xff; + else if (mv88e6xxx_num_databases(chip) == 256) + upper_mask = 0xf; + else + return -EOPNOTSUPP; + + /* Port's default FID bits 3:0 are located in reg 0x06, offset 12 */ + ret = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_BASE_VLAN); + if (ret < 0) + return ret; + + fid = (ret & PORT_BASE_VLAN_FID_3_0_MASK) >> 12; + + if (new) { + ret &= ~PORT_BASE_VLAN_FID_3_0_MASK; + ret |= (*new << 12) & PORT_BASE_VLAN_FID_3_0_MASK; + + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), PORT_BASE_VLAN, + ret); + if (ret < 0) + return ret; + } + + /* Port's default FID bits 11:4 are located in reg 0x05, offset 0 */ + ret = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_CONTROL_1); + if (ret < 0) + return ret; + + fid |= (ret & upper_mask) << 4; + + if (new) { + ret &= ~upper_mask; + ret |= (*new >> 4) & upper_mask; + + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), PORT_CONTROL_1, + ret); + if (ret < 0) + return ret; + + netdev_dbg(ds->ports[port].netdev, + "FID %d (was %d)\n", *new, fid); + } + + if (old) + *old = fid; + + return 0; +} + +static int _mv88e6xxx_port_fid_get(struct mv88e6xxx_chip *chip, + int port, u16 *fid) +{ + return _mv88e6xxx_port_fid(chip, port, NULL, fid); +} + +static int _mv88e6xxx_port_fid_set(struct mv88e6xxx_chip *chip, + int port, u16 fid) +{ + return _mv88e6xxx_port_fid(chip, port, &fid, NULL); +} + +static int _mv88e6xxx_fid_new(struct mv88e6xxx_chip *chip, u16 *fid) +{ + DECLARE_BITMAP(fid_bitmap, MV88E6XXX_N_FID); + struct mv88e6xxx_vtu_stu_entry vlan; + int i, err; + + bitmap_zero(fid_bitmap, MV88E6XXX_N_FID); + + /* Set every FID bit used by the (un)bridged ports */ + for (i = 0; i < chip->info->num_ports; ++i) { + err = _mv88e6xxx_port_fid_get(chip, i, fid); + if (err) + return err; + + set_bit(*fid, fid_bitmap); + } + + /* Set every FID bit used by the VLAN entries */ + err = _mv88e6xxx_vtu_vid_write(chip, GLOBAL_VTU_VID_MASK); + if (err) + return err; + + do { + err = _mv88e6xxx_vtu_getnext(chip, &vlan); + if (err) + return err; + + if (!vlan.valid) + break; + + set_bit(vlan.fid, fid_bitmap); + } while (vlan.vid < GLOBAL_VTU_VID_MASK); + + /* The reset value 0x000 is used to indicate that multiple address + * databases are not needed. Return the next positive available. + */ + *fid = find_next_zero_bit(fid_bitmap, MV88E6XXX_N_FID, 1); + if (unlikely(*fid >= mv88e6xxx_num_databases(chip))) + return -ENOSPC; + + /* Clear the database */ + return _mv88e6xxx_atu_flush(chip, *fid, true); +} + +static int _mv88e6xxx_vtu_new(struct mv88e6xxx_chip *chip, u16 vid, + struct mv88e6xxx_vtu_stu_entry *entry) +{ + struct dsa_switch *ds = chip->ds; + struct mv88e6xxx_vtu_stu_entry vlan = { + .valid = true, + .vid = vid, + }; + int i, err; + + err = _mv88e6xxx_fid_new(chip, &vlan.fid); + if (err) + return err; + + /* exclude all ports except the CPU and DSA ports */ + for (i = 0; i < chip->info->num_ports; ++i) + vlan.data[i] = dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i) + ? GLOBAL_VTU_DATA_MEMBER_TAG_UNMODIFIED + : GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER; + + if (mv88e6xxx_6097_family(chip) || mv88e6xxx_6165_family(chip) || + mv88e6xxx_6351_family(chip) || mv88e6xxx_6352_family(chip)) { + struct mv88e6xxx_vtu_stu_entry vstp; + + /* Adding a VTU entry requires a valid STU entry. As VSTP is not + * implemented, only one STU entry is needed to cover all VTU + * entries. Thus, validate the SID 0. + */ + vlan.sid = 0; + err = _mv88e6xxx_stu_getnext(chip, GLOBAL_VTU_SID_MASK, &vstp); + if (err) + return err; + + if (vstp.sid != vlan.sid || !vstp.valid) { + memset(&vstp, 0, sizeof(vstp)); + vstp.valid = true; + vstp.sid = vlan.sid; + + err = _mv88e6xxx_stu_loadpurge(chip, &vstp); + if (err) + return err; + } + } + + *entry = vlan; + return 0; +} + +static int _mv88e6xxx_vtu_get(struct mv88e6xxx_chip *chip, u16 vid, + struct mv88e6xxx_vtu_stu_entry *entry, bool creat) +{ + int err; + + if (!vid) + return -EINVAL; + + err = _mv88e6xxx_vtu_vid_write(chip, vid - 1); + if (err) + return err; + + err = _mv88e6xxx_vtu_getnext(chip, entry); + if (err) + return err; + + if (entry->vid != vid || !entry->valid) { + if (!creat) + return -EOPNOTSUPP; + /* -ENOENT would've been more appropriate, but switchdev expects + * -EOPNOTSUPP to inform bridge about an eventual software VLAN. + */ + + err = _mv88e6xxx_vtu_new(chip, vid, entry); + } + + return err; +} + +static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port, + u16 vid_begin, u16 vid_end) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + struct mv88e6xxx_vtu_stu_entry vlan; + int i, err; + + if (!vid_begin) + return -EOPNOTSUPP; + + mutex_lock(&chip->reg_lock); + + err = _mv88e6xxx_vtu_vid_write(chip, vid_begin - 1); + if (err) + goto unlock; + + do { + err = _mv88e6xxx_vtu_getnext(chip, &vlan); + if (err) + goto unlock; + + if (!vlan.valid) + break; + + if (vlan.vid > vid_end) + break; + + for (i = 0; i < chip->info->num_ports; ++i) { + if (dsa_is_dsa_port(ds, i) || dsa_is_cpu_port(ds, i)) + continue; + + if (vlan.data[i] == + GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) + continue; + + if (chip->ports[i].bridge_dev == + chip->ports[port].bridge_dev) + break; /* same bridge, check next VLAN */ + + netdev_warn(ds->ports[port].netdev, + "hardware VLAN %d already used by %s\n", + vlan.vid, + netdev_name(chip->ports[i].bridge_dev)); + err = -EOPNOTSUPP; + goto unlock; + } + } while (vlan.vid < vid_end); + +unlock: + mutex_unlock(&chip->reg_lock); + + return err; +} + +static const char * const mv88e6xxx_port_8021q_mode_names[] = { + [PORT_CONTROL_2_8021Q_DISABLED] = "Disabled", + [PORT_CONTROL_2_8021Q_FALLBACK] = "Fallback", + [PORT_CONTROL_2_8021Q_CHECK] = "Check", + [PORT_CONTROL_2_8021Q_SECURE] = "Secure", +}; + +static int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port, + bool vlan_filtering) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + u16 old, new = vlan_filtering ? PORT_CONTROL_2_8021Q_SECURE : + PORT_CONTROL_2_8021Q_DISABLED; + int ret; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_VTU)) + return -EOPNOTSUPP; + + mutex_lock(&chip->reg_lock); + + ret = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_CONTROL_2); + if (ret < 0) + goto unlock; + + old = ret & PORT_CONTROL_2_8021Q_MASK; + + if (new != old) { + ret &= ~PORT_CONTROL_2_8021Q_MASK; + ret |= new & PORT_CONTROL_2_8021Q_MASK; + + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), PORT_CONTROL_2, + ret); + if (ret < 0) + goto unlock; + + netdev_dbg(ds->ports[port].netdev, "802.1Q Mode %s (was %s)\n", + mv88e6xxx_port_8021q_mode_names[new], + mv88e6xxx_port_8021q_mode_names[old]); + } + + ret = 0; +unlock: + mutex_unlock(&chip->reg_lock); + + return ret; +} + +static int +mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int err; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_VTU)) + return -EOPNOTSUPP; + + /* If the requested port doesn't belong to the same bridge as the VLAN + * members, do not support it (yet) and fallback to software VLAN. + */ + err = mv88e6xxx_port_check_hw_vlan(ds, port, vlan->vid_begin, + vlan->vid_end); + if (err) + return err; + + /* We don't need any dynamic resource from the kernel (yet), + * so skip the prepare phase. + */ + return 0; +} + +static int _mv88e6xxx_port_vlan_add(struct mv88e6xxx_chip *chip, int port, + u16 vid, bool untagged) +{ + struct mv88e6xxx_vtu_stu_entry vlan; + int err; + + err = _mv88e6xxx_vtu_get(chip, vid, &vlan, true); + if (err) + return err; + + vlan.data[port] = untagged ? + GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED : + GLOBAL_VTU_DATA_MEMBER_TAG_TAGGED; + + return _mv88e6xxx_vtu_loadpurge(chip, &vlan); +} + +static void mv88e6xxx_port_vlan_add(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; + bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; + u16 vid; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_VTU)) + return; + + mutex_lock(&chip->reg_lock); + + for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) + if (_mv88e6xxx_port_vlan_add(chip, port, vid, untagged)) + netdev_err(ds->ports[port].netdev, + "failed to add VLAN %d%c\n", + vid, untagged ? 'u' : 't'); + + if (pvid && _mv88e6xxx_port_pvid_set(chip, port, vlan->vid_end)) + netdev_err(ds->ports[port].netdev, "failed to set PVID %d\n", + vlan->vid_end); + + mutex_unlock(&chip->reg_lock); +} + +static int _mv88e6xxx_port_vlan_del(struct mv88e6xxx_chip *chip, + int port, u16 vid) +{ + struct dsa_switch *ds = chip->ds; + struct mv88e6xxx_vtu_stu_entry vlan; + int i, err; + + err = _mv88e6xxx_vtu_get(chip, vid, &vlan, false); + if (err) + return err; + + /* Tell switchdev if this VLAN is handled in software */ + if (vlan.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) + return -EOPNOTSUPP; + + vlan.data[port] = GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER; + + /* keep the VLAN unless all ports are excluded */ + vlan.valid = false; + for (i = 0; i < chip->info->num_ports; ++i) { + if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i)) + continue; + + if (vlan.data[i] != GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) { + vlan.valid = true; + break; + } + } + + err = _mv88e6xxx_vtu_loadpurge(chip, &vlan); + if (err) + return err; + + return _mv88e6xxx_atu_remove(chip, vlan.fid, port, false); +} + +static int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_vlan *vlan) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + u16 pvid, vid; + int err = 0; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_VTU)) + return -EOPNOTSUPP; + + mutex_lock(&chip->reg_lock); + + err = _mv88e6xxx_port_pvid_get(chip, port, &pvid); + if (err) + goto unlock; + + for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) { + err = _mv88e6xxx_port_vlan_del(chip, port, vid); + if (err) + goto unlock; + + if (vid == pvid) { + err = _mv88e6xxx_port_pvid_set(chip, port, 0); + if (err) + goto unlock; + } + } + +unlock: + mutex_unlock(&chip->reg_lock); + + return err; +} + +static int _mv88e6xxx_atu_mac_write(struct mv88e6xxx_chip *chip, + const unsigned char *addr) +{ + int i, ret; + + for (i = 0; i < 3; i++) { + ret = _mv88e6xxx_reg_write( + chip, REG_GLOBAL, GLOBAL_ATU_MAC_01 + i, + (addr[i * 2] << 8) | addr[i * 2 + 1]); + if (ret < 0) + return ret; + } + + return 0; +} + +static int _mv88e6xxx_atu_mac_read(struct mv88e6xxx_chip *chip, + unsigned char *addr) +{ + int i, ret; + + for (i = 0; i < 3; i++) { + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, + GLOBAL_ATU_MAC_01 + i); + if (ret < 0) + return ret; + addr[i * 2] = ret >> 8; + addr[i * 2 + 1] = ret & 0xff; + } + + return 0; +} + +static int _mv88e6xxx_atu_load(struct mv88e6xxx_chip *chip, + struct mv88e6xxx_atu_entry *entry) +{ + int ret; + + ret = _mv88e6xxx_atu_wait(chip); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_atu_mac_write(chip, entry->mac); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_atu_data_write(chip, entry); + if (ret < 0) + return ret; + + return _mv88e6xxx_atu_cmd(chip, entry->fid, GLOBAL_ATU_OP_LOAD_DB); +} + +static int _mv88e6xxx_port_fdb_load(struct mv88e6xxx_chip *chip, int port, + const unsigned char *addr, u16 vid, + u8 state) +{ + struct mv88e6xxx_atu_entry entry = { 0 }; + struct mv88e6xxx_vtu_stu_entry vlan; + int err; + + /* Null VLAN ID corresponds to the port private database */ + if (vid == 0) + err = _mv88e6xxx_port_fid_get(chip, port, &vlan.fid); + else + err = _mv88e6xxx_vtu_get(chip, vid, &vlan, false); + if (err) + return err; + + entry.fid = vlan.fid; + entry.state = state; + ether_addr_copy(entry.mac, addr); + if (state != GLOBAL_ATU_DATA_STATE_UNUSED) { + entry.trunk = false; + entry.portv_trunkid = BIT(port); + } + + return _mv88e6xxx_atu_load(chip, &entry); +} + +static int mv88e6xxx_port_fdb_prepare(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_fdb *fdb, + struct switchdev_trans *trans) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_ATU)) + return -EOPNOTSUPP; + + /* We don't need any dynamic resource from the kernel (yet), + * so skip the prepare phase. + */ + return 0; +} + +static void mv88e6xxx_port_fdb_add(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_fdb *fdb, + struct switchdev_trans *trans) +{ + int state = is_multicast_ether_addr(fdb->addr) ? + GLOBAL_ATU_DATA_STATE_MC_STATIC : + GLOBAL_ATU_DATA_STATE_UC_STATIC; + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_ATU)) + return; + + mutex_lock(&chip->reg_lock); + if (_mv88e6xxx_port_fdb_load(chip, port, fdb->addr, fdb->vid, state)) + netdev_err(ds->ports[port].netdev, + "failed to load MAC address\n"); + mutex_unlock(&chip->reg_lock); +} + +static int mv88e6xxx_port_fdb_del(struct dsa_switch *ds, int port, + const struct switchdev_obj_port_fdb *fdb) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int ret; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_ATU)) + return -EOPNOTSUPP; + + mutex_lock(&chip->reg_lock); + ret = _mv88e6xxx_port_fdb_load(chip, port, fdb->addr, fdb->vid, + GLOBAL_ATU_DATA_STATE_UNUSED); + mutex_unlock(&chip->reg_lock); + + return ret; +} + +static int _mv88e6xxx_atu_getnext(struct mv88e6xxx_chip *chip, u16 fid, + struct mv88e6xxx_atu_entry *entry) +{ + struct mv88e6xxx_atu_entry next = { 0 }; + int ret; + + next.fid = fid; + + ret = _mv88e6xxx_atu_wait(chip); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_atu_cmd(chip, fid, GLOBAL_ATU_OP_GET_NEXT_DB); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_atu_mac_read(chip, next.mac); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, GLOBAL_ATU_DATA); + if (ret < 0) + return ret; + + next.state = ret & GLOBAL_ATU_DATA_STATE_MASK; + if (next.state != GLOBAL_ATU_DATA_STATE_UNUSED) { + unsigned int mask, shift; + + if (ret & GLOBAL_ATU_DATA_TRUNK) { + next.trunk = true; + mask = GLOBAL_ATU_DATA_TRUNK_ID_MASK; + shift = GLOBAL_ATU_DATA_TRUNK_ID_SHIFT; + } else { + next.trunk = false; + mask = GLOBAL_ATU_DATA_PORT_VECTOR_MASK; + shift = GLOBAL_ATU_DATA_PORT_VECTOR_SHIFT; + } + + next.portv_trunkid = (ret & mask) >> shift; + } + + *entry = next; + return 0; +} + +static int _mv88e6xxx_port_fdb_dump_one(struct mv88e6xxx_chip *chip, + u16 fid, u16 vid, int port, + struct switchdev_obj_port_fdb *fdb, + int (*cb)(struct switchdev_obj *obj)) +{ + struct mv88e6xxx_atu_entry addr = { + .mac = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }, + }; + int err; + + err = _mv88e6xxx_atu_mac_write(chip, addr.mac); + if (err) + return err; + + do { + err = _mv88e6xxx_atu_getnext(chip, fid, &addr); + if (err) + break; + + if (addr.state == GLOBAL_ATU_DATA_STATE_UNUSED) + break; + + if (!addr.trunk && addr.portv_trunkid & BIT(port)) { + bool is_static = addr.state == + (is_multicast_ether_addr(addr.mac) ? + GLOBAL_ATU_DATA_STATE_MC_STATIC : + GLOBAL_ATU_DATA_STATE_UC_STATIC); + + fdb->vid = vid; + ether_addr_copy(fdb->addr, addr.mac); + fdb->ndm_state = is_static ? NUD_NOARP : NUD_REACHABLE; + + err = cb(&fdb->obj); + if (err) + break; + } + } while (!is_broadcast_ether_addr(addr.mac)); + + return err; +} + +static int mv88e6xxx_port_fdb_dump(struct dsa_switch *ds, int port, + struct switchdev_obj_port_fdb *fdb, + int (*cb)(struct switchdev_obj *obj)) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + struct mv88e6xxx_vtu_stu_entry vlan = { + .vid = GLOBAL_VTU_VID_MASK, /* all ones */ + }; + u16 fid; + int err; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_ATU)) + return -EOPNOTSUPP; + + mutex_lock(&chip->reg_lock); + + /* Dump port's default Filtering Information Database (VLAN ID 0) */ + err = _mv88e6xxx_port_fid_get(chip, port, &fid); + if (err) + goto unlock; + + err = _mv88e6xxx_port_fdb_dump_one(chip, fid, 0, port, fdb, cb); + if (err) + goto unlock; + + /* Dump VLANs' Filtering Information Databases */ + err = _mv88e6xxx_vtu_vid_write(chip, vlan.vid); + if (err) + goto unlock; + + do { + err = _mv88e6xxx_vtu_getnext(chip, &vlan); + if (err) + break; + + if (!vlan.valid) + break; + + err = _mv88e6xxx_port_fdb_dump_one(chip, vlan.fid, vlan.vid, + port, fdb, cb); + if (err) + break; + } while (vlan.vid < GLOBAL_VTU_VID_MASK); + +unlock: + mutex_unlock(&chip->reg_lock); + + return err; +} + +static int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port, + struct net_device *bridge) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int i, err = 0; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_VLANTABLE)) + return -EOPNOTSUPP; + + mutex_lock(&chip->reg_lock); + + /* Assign the bridge and remap each port's VLANTable */ + chip->ports[port].bridge_dev = bridge; + + for (i = 0; i < chip->info->num_ports; ++i) { + if (chip->ports[i].bridge_dev == bridge) { + err = _mv88e6xxx_port_based_vlan_map(chip, i); + if (err) + break; + } + } + + mutex_unlock(&chip->reg_lock); + + return err; +} + +static void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + struct net_device *bridge = chip->ports[port].bridge_dev; + int i; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_VLANTABLE)) + return; + + mutex_lock(&chip->reg_lock); + + /* Unassign the bridge and remap each port's VLANTable */ + chip->ports[port].bridge_dev = NULL; + + for (i = 0; i < chip->info->num_ports; ++i) + if (i == port || chip->ports[i].bridge_dev == bridge) + if (_mv88e6xxx_port_based_vlan_map(chip, i)) + netdev_warn(ds->ports[i].netdev, + "failed to remap\n"); + + mutex_unlock(&chip->reg_lock); +} + +static int _mv88e6xxx_mdio_page_write(struct mv88e6xxx_chip *chip, + int port, int page, int reg, int val) +{ + int ret; + + ret = mv88e6xxx_mdio_write_indirect(chip, port, 0x16, page); + if (ret < 0) + goto restore_page_0; + + ret = mv88e6xxx_mdio_write_indirect(chip, port, reg, val); +restore_page_0: + mv88e6xxx_mdio_write_indirect(chip, port, 0x16, 0x0); + + return ret; +} + +static int _mv88e6xxx_mdio_page_read(struct mv88e6xxx_chip *chip, + int port, int page, int reg) +{ + int ret; + + ret = mv88e6xxx_mdio_write_indirect(chip, port, 0x16, page); + if (ret < 0) + goto restore_page_0; + + ret = mv88e6xxx_mdio_read_indirect(chip, port, reg); +restore_page_0: + mv88e6xxx_mdio_write_indirect(chip, port, 0x16, 0x0); + + return ret; +} + +static int mv88e6xxx_switch_reset(struct mv88e6xxx_chip *chip) +{ + bool ppu_active = mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU_ACTIVE); + u16 is_reset = (ppu_active ? 0x8800 : 0xc800); + struct gpio_desc *gpiod = chip->reset; + unsigned long timeout; + int ret; + int i; + + /* Set all ports to the disabled state. */ + for (i = 0; i < chip->info->num_ports; i++) { + ret = _mv88e6xxx_reg_read(chip, REG_PORT(i), PORT_CONTROL); + if (ret < 0) + return ret; + + ret = _mv88e6xxx_reg_write(chip, REG_PORT(i), PORT_CONTROL, + ret & 0xfffc); + if (ret) + return ret; + } + + /* Wait for transmit queues to drain. */ + usleep_range(2000, 4000); + + /* If there is a gpio connected to the reset pin, toggle it */ + if (gpiod) { + gpiod_set_value_cansleep(gpiod, 1); + usleep_range(10000, 20000); + gpiod_set_value_cansleep(gpiod, 0); + usleep_range(10000, 20000); + } + + /* Reset the switch. Keep the PPU active if requested. The PPU + * needs to be active to support indirect phy register access + * through global registers 0x18 and 0x19. + */ + if (ppu_active) + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, 0x04, 0xc000); + else + ret = _mv88e6xxx_reg_write(chip, REG_GLOBAL, 0x04, 0xc400); + if (ret) + return ret; + + /* Wait up to one second for reset to complete. */ + timeout = jiffies + 1 * HZ; + while (time_before(jiffies, timeout)) { + ret = _mv88e6xxx_reg_read(chip, REG_GLOBAL, 0x00); + if (ret < 0) + return ret; + + if ((ret & is_reset) == is_reset) + break; + usleep_range(1000, 2000); + } + if (time_after(jiffies, timeout)) + ret = -ETIMEDOUT; + else + ret = 0; + + return ret; +} + +static int mv88e6xxx_power_on_serdes(struct mv88e6xxx_chip *chip) +{ + int ret; + + ret = _mv88e6xxx_mdio_page_read(chip, REG_FIBER_SERDES, + PAGE_FIBER_SERDES, MII_BMCR); + if (ret < 0) + return ret; + + if (ret & BMCR_PDOWN) { + ret &= ~BMCR_PDOWN; + ret = _mv88e6xxx_mdio_page_write(chip, REG_FIBER_SERDES, + PAGE_FIBER_SERDES, MII_BMCR, + ret); + } + + return ret; +} + +static int mv88e6xxx_setup_port(struct mv88e6xxx_chip *chip, int port) +{ + struct dsa_switch *ds = chip->ds; + int ret; + u16 reg; + + if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || + mv88e6xxx_6185_family(chip) || mv88e6xxx_6095_family(chip) || + mv88e6xxx_6065_family(chip) || mv88e6xxx_6320_family(chip)) { + /* MAC Forcing register: don't force link, speed, + * duplex or flow control state to any particular + * values on physical ports, but force the CPU port + * and all DSA ports to their maximum bandwidth and + * full duplex. + */ + reg = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_PCS_CTRL); + if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) { + reg &= ~PORT_PCS_CTRL_UNFORCED; + reg |= PORT_PCS_CTRL_FORCE_LINK | + PORT_PCS_CTRL_LINK_UP | + PORT_PCS_CTRL_DUPLEX_FULL | + PORT_PCS_CTRL_FORCE_DUPLEX; + if (mv88e6xxx_6065_family(chip)) + reg |= PORT_PCS_CTRL_100; + else + reg |= PORT_PCS_CTRL_1000; + } else { + reg |= PORT_PCS_CTRL_UNFORCED; + } + + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_PCS_CTRL, reg); + if (ret) + return ret; + } + + /* Port Control: disable Drop-on-Unlock, disable Drop-on-Lock, + * disable Header mode, enable IGMP/MLD snooping, disable VLAN + * tunneling, determine priority by looking at 802.1p and IP + * priority fields (IP prio has precedence), and set STP state + * to Forwarding. + * + * If this is the CPU link, use DSA or EDSA tagging depending + * on which tagging mode was configured. + * + * If this is a link to another switch, use DSA tagging mode. + * + * If this is the upstream port for this switch, enable + * forwarding of unknown unicasts and multicasts. + */ + reg = 0; + if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || + mv88e6xxx_6095_family(chip) || mv88e6xxx_6065_family(chip) || + mv88e6xxx_6185_family(chip) || mv88e6xxx_6320_family(chip)) + reg = PORT_CONTROL_IGMP_MLD_SNOOP | + PORT_CONTROL_USE_TAG | PORT_CONTROL_USE_IP | + PORT_CONTROL_STATE_FORWARDING; + if (dsa_is_cpu_port(ds, port)) { + if (mv88e6xxx_6095_family(chip) || mv88e6xxx_6185_family(chip)) + reg |= PORT_CONTROL_DSA_TAG; + if (mv88e6xxx_6352_family(chip) || + mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || + mv88e6xxx_6097_family(chip) || + mv88e6xxx_6320_family(chip)) { + reg |= PORT_CONTROL_FRAME_ETHER_TYPE_DSA | + PORT_CONTROL_FORWARD_UNKNOWN | + PORT_CONTROL_FORWARD_UNKNOWN_MC; + } + + if (mv88e6xxx_6352_family(chip) || + mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || + mv88e6xxx_6097_family(chip) || + mv88e6xxx_6095_family(chip) || + mv88e6xxx_6065_family(chip) || + mv88e6xxx_6185_family(chip) || + mv88e6xxx_6320_family(chip)) { + reg |= PORT_CONTROL_EGRESS_ADD_TAG; + } + } + if (dsa_is_dsa_port(ds, port)) { + if (mv88e6xxx_6095_family(chip) || + mv88e6xxx_6185_family(chip)) + reg |= PORT_CONTROL_DSA_TAG; + if (mv88e6xxx_6352_family(chip) || + mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || + mv88e6xxx_6097_family(chip) || + mv88e6xxx_6320_family(chip)) { + reg |= PORT_CONTROL_FRAME_MODE_DSA; + } + + if (port == dsa_upstream_port(ds)) + reg |= PORT_CONTROL_FORWARD_UNKNOWN | + PORT_CONTROL_FORWARD_UNKNOWN_MC; + } + if (reg) { + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_CONTROL, reg); + if (ret) + return ret; + } + + /* If this port is connected to a SerDes, make sure the SerDes is not + * powered down. + */ + if (mv88e6xxx_6352_family(chip)) { + ret = _mv88e6xxx_reg_read(chip, REG_PORT(port), PORT_STATUS); + if (ret < 0) + return ret; + ret &= PORT_STATUS_CMODE_MASK; + if ((ret == PORT_STATUS_CMODE_100BASE_X) || + (ret == PORT_STATUS_CMODE_1000BASE_X) || + (ret == PORT_STATUS_CMODE_SGMII)) { + ret = mv88e6xxx_power_on_serdes(chip); + if (ret < 0) + return ret; + } + } + + /* Port Control 2: don't force a good FCS, set the maximum frame size to + * 10240 bytes, disable 802.1q tags checking, don't discard tagged or + * untagged frames on this port, do a destination address lookup on all + * received packets as usual, disable ARP mirroring and don't send a + * copy of all transmitted/received frames on this port to the CPU. + */ + reg = 0; + if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || + mv88e6xxx_6095_family(chip) || mv88e6xxx_6320_family(chip) || + mv88e6xxx_6185_family(chip)) + reg = PORT_CONTROL_2_MAP_DA; + + if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || mv88e6xxx_6320_family(chip)) + reg |= PORT_CONTROL_2_JUMBO_10240; + + if (mv88e6xxx_6095_family(chip) || mv88e6xxx_6185_family(chip)) { + /* Set the upstream port this port should use */ + reg |= dsa_upstream_port(ds); + /* enable forwarding of unknown multicast addresses to + * the upstream port + */ + if (port == dsa_upstream_port(ds)) + reg |= PORT_CONTROL_2_FORWARD_UNKNOWN; + } + + reg |= PORT_CONTROL_2_8021Q_DISABLED; + + if (reg) { + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_CONTROL_2, reg); + if (ret) + return ret; + } + + /* Port Association Vector: when learning source addresses + * of packets, add the address to the address database using + * a port bitmap that has only the bit for this port set and + * the other bits clear. + */ + reg = 1 << port; + /* Disable learning for CPU port */ + if (dsa_is_cpu_port(ds, port)) + reg = 0; + + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), PORT_ASSOC_VECTOR, + reg); + if (ret) + return ret; + + /* Egress rate control 2: disable egress rate control. */ + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), PORT_RATE_CONTROL_2, + 0x0000); + if (ret) + return ret; + + if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || + mv88e6xxx_6320_family(chip)) { + /* Do not limit the period of time that this port can + * be paused for by the remote end or the period of + * time that this port can pause the remote end. + */ + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_PAUSE_CTRL, 0x0000); + if (ret) + return ret; + + /* Port ATU control: disable limiting the number of + * address database entries that this port is allowed + * to use. + */ + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_ATU_CONTROL, 0x0000); + /* Priority Override: disable DA, SA and VTU priority + * override. + */ + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_PRI_OVERRIDE, 0x0000); + if (ret) + return ret; + + /* Port Ethertype: use the Ethertype DSA Ethertype + * value. + */ + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_ETH_TYPE, ETH_P_EDSA); + if (ret) + return ret; + /* Tag Remap: use an identity 802.1p prio -> switch + * prio mapping. + */ + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_TAG_REGMAP_0123, 0x3210); + if (ret) + return ret; + + /* Tag Remap 2: use an identity 802.1p prio -> switch + * prio mapping. + */ + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_TAG_REGMAP_4567, 0x7654); + if (ret) + return ret; + } + + if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || + mv88e6xxx_6185_family(chip) || mv88e6xxx_6095_family(chip) || + mv88e6xxx_6320_family(chip)) { + /* Rate Control: disable ingress rate limiting. */ + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), + PORT_RATE_CONTROL, 0x0001); + if (ret) + return ret; + } + + /* Port Control 1: disable trunking, disable sending + * learning messages to this port. + */ + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), PORT_CONTROL_1, + 0x0000); + if (ret) + return ret; + + /* Port based VLAN map: give each port the same default address + * database, and allow bidirectional communication between the + * CPU and DSA port(s), and the other ports. + */ + ret = _mv88e6xxx_port_fid_set(chip, port, 0); + if (ret) + return ret; + + ret = _mv88e6xxx_port_based_vlan_map(chip, port); + if (ret) + return ret; + + /* Default VLAN ID and priority: don't set a default VLAN + * ID, and set the default packet priority to zero. + */ + ret = _mv88e6xxx_reg_write(chip, REG_PORT(port), PORT_DEFAULT_VLAN, + 0x0000); + if (ret) + return ret; + + return 0; +} + +static int mv88e6xxx_setup_global(struct mv88e6xxx_chip *chip) +{ + struct dsa_switch *ds = chip->ds; + u32 upstream_port = dsa_upstream_port(ds); + u16 reg; + int err; + int i; + + /* Enable the PHY Polling Unit if present, don't discard any packets, + * and mask all interrupt sources. + */ + reg = 0; + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU) || + mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU_ACTIVE)) + reg |= GLOBAL_CONTROL_PPU_ENABLE; + + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_CONTROL, reg); + if (err) + return err; + + /* Configure the upstream port, and configure it as the port to which + * ingress and egress and ARP monitor frames are to be sent. + */ + reg = upstream_port << GLOBAL_MONITOR_CONTROL_INGRESS_SHIFT | + upstream_port << GLOBAL_MONITOR_CONTROL_EGRESS_SHIFT | + upstream_port << GLOBAL_MONITOR_CONTROL_ARP_SHIFT; + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_MONITOR_CONTROL, + reg); + if (err) + return err; + + /* Disable remote management, and set the switch's DSA device number. */ + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_CONTROL_2, + GLOBAL_CONTROL_2_MULTIPLE_CASCADE | + (ds->index & 0x1f)); + if (err) + return err; + + /* Set the default address aging time to 5 minutes, and + * enable address learn messages to be sent to all message + * ports. + */ + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_ATU_CONTROL, + 0x0140 | GLOBAL_ATU_CONTROL_LEARN2ALL); + if (err) + return err; + + /* Configure the IP ToS mapping registers. */ + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_IP_PRI_0, 0x0000); + if (err) + return err; + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_IP_PRI_1, 0x0000); + if (err) + return err; + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_IP_PRI_2, 0x5555); + if (err) + return err; + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_IP_PRI_3, 0x5555); + if (err) + return err; + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_IP_PRI_4, 0xaaaa); + if (err) + return err; + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_IP_PRI_5, 0xaaaa); + if (err) + return err; + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_IP_PRI_6, 0xffff); + if (err) + return err; + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_IP_PRI_7, 0xffff); + if (err) + return err; + + /* Configure the IEEE 802.1p priority mapping register. */ + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_IEEE_PRI, 0xfa41); + if (err) + return err; + + /* Send all frames with destination addresses matching + * 01:80:c2:00:00:0x to the CPU port. + */ + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_MGMT_EN_0X, + 0xffff); + if (err) + return err; + + /* Ignore removed tag data on doubly tagged packets, disable + * flow control messages, force flow control priority to the + * highest, and send all special multicast frames to the CPU + * port at the highest priority. + */ + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_SWITCH_MGMT, + 0x7 | GLOBAL2_SWITCH_MGMT_RSVD2CPU | 0x70 | + GLOBAL2_SWITCH_MGMT_FORCE_FLOW_CTRL_PRI); + if (err) + return err; + + /* Program the DSA routing table. */ + for (i = 0; i < 32; i++) { + int nexthop = 0x1f; + + if (i != ds->index && i < DSA_MAX_SWITCHES) + nexthop = ds->rtable[i] & 0x1f; + + err = _mv88e6xxx_reg_write( + chip, REG_GLOBAL2, + GLOBAL2_DEVICE_MAPPING, + GLOBAL2_DEVICE_MAPPING_UPDATE | + (i << GLOBAL2_DEVICE_MAPPING_TARGET_SHIFT) | nexthop); + if (err) + return err; + } + + /* Clear all trunk masks. */ + for (i = 0; i < 8; i++) { + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, + GLOBAL2_TRUNK_MASK, + 0x8000 | + (i << GLOBAL2_TRUNK_MASK_NUM_SHIFT) | + ((1 << chip->info->num_ports) - 1)); + if (err) + return err; + } + + /* Clear all trunk mappings. */ + for (i = 0; i < 16; i++) { + err = _mv88e6xxx_reg_write( + chip, REG_GLOBAL2, + GLOBAL2_TRUNK_MAPPING, + GLOBAL2_TRUNK_MAPPING_UPDATE | + (i << GLOBAL2_TRUNK_MAPPING_ID_SHIFT)); + if (err) + return err; + } + + if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || + mv88e6xxx_6320_family(chip)) { + /* Send all frames with destination addresses matching + * 01:80:c2:00:00:2x to the CPU port. + */ + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, + GLOBAL2_MGMT_EN_2X, 0xffff); + if (err) + return err; + + /* Initialise cross-chip port VLAN table to reset + * defaults. + */ + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, + GLOBAL2_PVT_ADDR, 0x9000); + if (err) + return err; + + /* Clear the priority override table. */ + for (i = 0; i < 16; i++) { + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, + GLOBAL2_PRIO_OVERRIDE, + 0x8000 | (i << 8)); + if (err) + return err; + } + } + + if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || + mv88e6xxx_6185_family(chip) || mv88e6xxx_6095_family(chip) || + mv88e6xxx_6320_family(chip)) { + /* Disable ingress rate limiting by resetting all + * ingress rate limit registers to their initial + * state. + */ + for (i = 0; i < chip->info->num_ports; i++) { + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, + GLOBAL2_INGRESS_OP, + 0x9000 | (i << 8)); + if (err) + return err; + } + } + + /* Clear the statistics counters for all ports */ + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_STATS_OP, + GLOBAL_STATS_OP_FLUSH_ALL); + if (err) + return err; + + /* Wait for the flush to complete. */ + err = _mv88e6xxx_stats_wait(chip); + if (err) + return err; + + /* Clear all ATU entries */ + err = _mv88e6xxx_atu_flush(chip, 0, true); + if (err) + return err; + + /* Clear all the VTU and STU entries */ + err = _mv88e6xxx_vtu_stu_flush(chip); + if (err < 0) + return err; + + return err; +} + +static int mv88e6xxx_setup(struct dsa_switch *ds) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int err; + int i; + + chip->ds = ds; + ds->slave_mii_bus = chip->mdio_bus; + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEPROM)) + mutex_init(&chip->eeprom_mutex); + + mutex_lock(&chip->reg_lock); + + err = mv88e6xxx_switch_reset(chip); + if (err) + goto unlock; + + err = mv88e6xxx_setup_global(chip); + if (err) + goto unlock; + + for (i = 0; i < chip->info->num_ports; i++) { + err = mv88e6xxx_setup_port(chip, i); + if (err) + goto unlock; + } + +unlock: + mutex_unlock(&chip->reg_lock); + + return err; +} + +static int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, + int reg) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int ret; + + mutex_lock(&chip->reg_lock); + ret = _mv88e6xxx_mdio_page_read(chip, port, page, reg); + mutex_unlock(&chip->reg_lock); + + return ret; +} + +static int mv88e6xxx_mdio_page_write(struct dsa_switch *ds, int port, int page, + int reg, int val) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int ret; + + mutex_lock(&chip->reg_lock); + ret = _mv88e6xxx_mdio_page_write(chip, port, page, reg, val); + mutex_unlock(&chip->reg_lock); + + return ret; +} + +static int mv88e6xxx_port_to_mdio_addr(struct mv88e6xxx_chip *chip, int port) +{ + if (port >= 0 && port < chip->info->num_ports) + return port; + return -EINVAL; +} + +static int mv88e6xxx_mdio_read(struct mii_bus *bus, int port, int regnum) +{ + struct mv88e6xxx_chip *chip = bus->priv; + int addr = mv88e6xxx_port_to_mdio_addr(chip, port); + int ret; + + if (addr < 0) + return 0xffff; + + mutex_lock(&chip->reg_lock); + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU)) + ret = mv88e6xxx_mdio_read_ppu(chip, addr, regnum); + else if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_SMI_PHY)) + ret = mv88e6xxx_mdio_read_indirect(chip, addr, regnum); + else + ret = mv88e6xxx_mdio_read_direct(chip, addr, regnum); + + mutex_unlock(&chip->reg_lock); + return ret; +} + +static int mv88e6xxx_mdio_write(struct mii_bus *bus, int port, int regnum, + u16 val) +{ + struct mv88e6xxx_chip *chip = bus->priv; + int addr = mv88e6xxx_port_to_mdio_addr(chip, port); + int ret; + + if (addr < 0) + return 0xffff; + + mutex_lock(&chip->reg_lock); + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU)) + ret = mv88e6xxx_mdio_write_ppu(chip, addr, regnum, val); + else if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_SMI_PHY)) + ret = mv88e6xxx_mdio_write_indirect(chip, addr, regnum, val); + else + ret = mv88e6xxx_mdio_write_direct(chip, addr, regnum, val); + + mutex_unlock(&chip->reg_lock); + return ret; +} + +static int mv88e6xxx_mdio_register(struct mv88e6xxx_chip *chip, + struct device_node *np) +{ + static int index; + struct mii_bus *bus; + int err; + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_PPU)) + mv88e6xxx_ppu_state_init(chip); + + if (np) + chip->mdio_np = of_get_child_by_name(np, "mdio"); + + bus = devm_mdiobus_alloc(chip->dev); + if (!bus) + return -ENOMEM; + + bus->priv = (void *)chip; + if (np) { + bus->name = np->full_name; + snprintf(bus->id, MII_BUS_ID_SIZE, "%s", np->full_name); + } else { + bus->name = "mv88e6xxx SMI"; + snprintf(bus->id, MII_BUS_ID_SIZE, "mv88e6xxx-%d", index++); + } + + bus->read = mv88e6xxx_mdio_read; + bus->write = mv88e6xxx_mdio_write; + bus->parent = chip->dev; + + if (chip->mdio_np) + err = of_mdiobus_register(bus, chip->mdio_np); + else + err = mdiobus_register(bus); + if (err) { + dev_err(chip->dev, "Cannot register MDIO bus (%d)\n", err); + goto out; + } + chip->mdio_bus = bus; + + return 0; + +out: + if (chip->mdio_np) + of_node_put(chip->mdio_np); + + return err; +} + +static void mv88e6xxx_mdio_unregister(struct mv88e6xxx_chip *chip) + +{ + struct mii_bus *bus = chip->mdio_bus; + + mdiobus_unregister(bus); + + if (chip->mdio_np) + of_node_put(chip->mdio_np); +} + +#ifdef CONFIG_NET_DSA_HWMON + +static int mv88e61xx_get_temp(struct dsa_switch *ds, int *temp) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int ret; + int val; + + *temp = 0; + + mutex_lock(&chip->reg_lock); + + ret = mv88e6xxx_mdio_write_direct(chip, 0x0, 0x16, 0x6); + if (ret < 0) + goto error; + + /* Enable temperature sensor */ + ret = mv88e6xxx_mdio_read_direct(chip, 0x0, 0x1a); + if (ret < 0) + goto error; + + ret = mv88e6xxx_mdio_write_direct(chip, 0x0, 0x1a, ret | (1 << 5)); + if (ret < 0) + goto error; + + /* Wait for temperature to stabilize */ + usleep_range(10000, 12000); + + val = mv88e6xxx_mdio_read_direct(chip, 0x0, 0x1a); + if (val < 0) { + ret = val; + goto error; + } + + /* Disable temperature sensor */ + ret = mv88e6xxx_mdio_write_direct(chip, 0x0, 0x1a, ret & ~(1 << 5)); + if (ret < 0) + goto error; + + *temp = ((val & 0x1f) - 5) * 5; + +error: + mv88e6xxx_mdio_write_direct(chip, 0x0, 0x16, 0x0); + mutex_unlock(&chip->reg_lock); + return ret; +} + +static int mv88e63xx_get_temp(struct dsa_switch *ds, int *temp) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int phy = mv88e6xxx_6320_family(chip) ? 3 : 0; + int ret; + + *temp = 0; + + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 27); + if (ret < 0) + return ret; + + *temp = (ret & 0xff) - 25; + + return 0; +} + +static int mv88e6xxx_get_temp(struct dsa_switch *ds, int *temp) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_TEMP)) + return -EOPNOTSUPP; + + if (mv88e6xxx_6320_family(chip) || mv88e6xxx_6352_family(chip)) + return mv88e63xx_get_temp(ds, temp); + + return mv88e61xx_get_temp(ds, temp); +} + +static int mv88e6xxx_get_temp_limit(struct dsa_switch *ds, int *temp) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int phy = mv88e6xxx_6320_family(chip) ? 3 : 0; + int ret; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_TEMP_LIMIT)) + return -EOPNOTSUPP; + + *temp = 0; + + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); + if (ret < 0) + return ret; + + *temp = (((ret >> 8) & 0x1f) * 5) - 25; + + return 0; +} + +static int mv88e6xxx_set_temp_limit(struct dsa_switch *ds, int temp) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int phy = mv88e6xxx_6320_family(chip) ? 3 : 0; + int ret; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_TEMP_LIMIT)) + return -EOPNOTSUPP; + + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); + if (ret < 0) + return ret; + temp = clamp_val(DIV_ROUND_CLOSEST(temp, 5) + 5, 0, 0x1f); + return mv88e6xxx_mdio_page_write(ds, phy, 6, 26, + (ret & 0xe0ff) | (temp << 8)); +} + +static int mv88e6xxx_get_temp_alarm(struct dsa_switch *ds, bool *alarm) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int phy = mv88e6xxx_6320_family(chip) ? 3 : 0; + int ret; + + if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_TEMP_LIMIT)) + return -EOPNOTSUPP; + + *alarm = false; + + ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); + if (ret < 0) + return ret; + + *alarm = !!(ret & 0x40); + + return 0; +} +#endif /* CONFIG_NET_DSA_HWMON */ + +static const struct mv88e6xxx_info mv88e6xxx_table[] = { + [MV88E6085] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6085, + .family = MV88E6XXX_FAMILY_6097, + .name = "Marvell 88E6085", + .num_databases = 4096, + .num_ports = 10, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6097, + }, + + [MV88E6095] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6095, + .family = MV88E6XXX_FAMILY_6095, + .name = "Marvell 88E6095/88E6095F", + .num_databases = 256, + .num_ports = 11, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6095, + }, + + [MV88E6123] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6123, + .family = MV88E6XXX_FAMILY_6165, + .name = "Marvell 88E6123", + .num_databases = 4096, + .num_ports = 3, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6165, + }, + + [MV88E6131] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6131, + .family = MV88E6XXX_FAMILY_6185, + .name = "Marvell 88E6131", + .num_databases = 256, + .num_ports = 8, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6185, + }, + + [MV88E6161] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6161, + .family = MV88E6XXX_FAMILY_6165, + .name = "Marvell 88E6161", + .num_databases = 4096, + .num_ports = 6, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6165, + }, + + [MV88E6165] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6165, + .family = MV88E6XXX_FAMILY_6165, + .name = "Marvell 88E6165", + .num_databases = 4096, + .num_ports = 6, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6165, + }, + + [MV88E6171] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6171, + .family = MV88E6XXX_FAMILY_6351, + .name = "Marvell 88E6171", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6351, + }, + + [MV88E6172] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6172, + .family = MV88E6XXX_FAMILY_6352, + .name = "Marvell 88E6172", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6352, + }, + + [MV88E6175] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6175, + .family = MV88E6XXX_FAMILY_6351, + .name = "Marvell 88E6175", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6351, + }, + + [MV88E6176] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6176, + .family = MV88E6XXX_FAMILY_6352, + .name = "Marvell 88E6176", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6352, + }, + + [MV88E6185] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6185, + .family = MV88E6XXX_FAMILY_6185, + .name = "Marvell 88E6185", + .num_databases = 256, + .num_ports = 10, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6185, + }, + + [MV88E6240] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6240, + .family = MV88E6XXX_FAMILY_6352, + .name = "Marvell 88E6240", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6352, + }, + + [MV88E6320] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6320, + .family = MV88E6XXX_FAMILY_6320, + .name = "Marvell 88E6320", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6320, + }, + + [MV88E6321] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6321, + .family = MV88E6XXX_FAMILY_6320, + .name = "Marvell 88E6321", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6320, + }, + + [MV88E6350] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6350, + .family = MV88E6XXX_FAMILY_6351, + .name = "Marvell 88E6350", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6351, + }, + + [MV88E6351] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6351, + .family = MV88E6XXX_FAMILY_6351, + .name = "Marvell 88E6351", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6351, + }, + + [MV88E6352] = { + .prod_num = PORT_SWITCH_ID_PROD_NUM_6352, + .family = MV88E6XXX_FAMILY_6352, + .name = "Marvell 88E6352", + .num_databases = 4096, + .num_ports = 7, + .port_base_addr = 0x10, + .flags = MV88E6XXX_FLAGS_FAMILY_6352, + }, +}; + +static const struct mv88e6xxx_info *mv88e6xxx_lookup_info(unsigned int prod_num) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(mv88e6xxx_table); ++i) + if (mv88e6xxx_table[i].prod_num == prod_num) + return &mv88e6xxx_table[i]; + + return NULL; +} + +static int mv88e6xxx_detect(struct mv88e6xxx_chip *chip) +{ + const struct mv88e6xxx_info *info; + int id, prod_num, rev; + + id = mv88e6xxx_reg_read(chip, chip->info->port_base_addr, + PORT_SWITCH_ID); + if (id < 0) + return id; + + prod_num = (id & 0xfff0) >> 4; + rev = id & 0x000f; + + info = mv88e6xxx_lookup_info(prod_num); + if (!info) + return -ENODEV; + + /* Update the compatible info with the probed one */ + chip->info = info; + + dev_info(chip->dev, "switch 0x%x detected: %s, revision %u\n", + chip->info->prod_num, chip->info->name, rev); + + return 0; +} + +static struct mv88e6xxx_chip *mv88e6xxx_alloc_chip(struct device *dev) +{ + struct mv88e6xxx_chip *chip; + + chip = devm_kzalloc(dev, sizeof(*chip), GFP_KERNEL); + if (!chip) + return NULL; + + chip->dev = dev; + + mutex_init(&chip->reg_lock); + + return chip; +} + +static int mv88e6xxx_smi_init(struct mv88e6xxx_chip *chip, + struct mii_bus *bus, int sw_addr) +{ + /* ADDR[0] pin is unavailable externally and considered zero */ + if (sw_addr & 0x1) + return -EINVAL; + + if (sw_addr == 0) + chip->smi_ops = &mv88e6xxx_smi_single_chip_ops; + else if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_MULTI_CHIP)) + chip->smi_ops = &mv88e6xxx_smi_multi_chip_ops; + else + return -EINVAL; + + chip->bus = bus; + chip->sw_addr = sw_addr; + + return 0; +} + +static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, + struct device *host_dev, int sw_addr, + void **priv) +{ + struct mv88e6xxx_chip *chip; + struct mii_bus *bus; + int err; + + bus = dsa_host_dev_to_mii_bus(host_dev); + if (!bus) + return NULL; + + chip = mv88e6xxx_alloc_chip(dsa_dev); + if (!chip) + return NULL; + + /* Legacy SMI probing will only support chips similar to 88E6085 */ + chip->info = &mv88e6xxx_table[MV88E6085]; + + err = mv88e6xxx_smi_init(chip, bus, sw_addr); + if (err) + goto free; + + err = mv88e6xxx_detect(chip); + if (err) + goto free; + + err = mv88e6xxx_mdio_register(chip, NULL); + if (err) + goto free; + + *priv = chip; + + return chip->info->name; +free: + devm_kfree(dsa_dev, chip); + + return NULL; +} + +static struct dsa_switch_driver mv88e6xxx_switch_driver = { + .tag_protocol = DSA_TAG_PROTO_EDSA, + .probe = mv88e6xxx_drv_probe, + .setup = mv88e6xxx_setup, + .set_addr = mv88e6xxx_set_addr, + .adjust_link = mv88e6xxx_adjust_link, + .get_strings = mv88e6xxx_get_strings, + .get_ethtool_stats = mv88e6xxx_get_ethtool_stats, + .get_sset_count = mv88e6xxx_get_sset_count, + .set_eee = mv88e6xxx_set_eee, + .get_eee = mv88e6xxx_get_eee, +#ifdef CONFIG_NET_DSA_HWMON + .get_temp = mv88e6xxx_get_temp, + .get_temp_limit = mv88e6xxx_get_temp_limit, + .set_temp_limit = mv88e6xxx_set_temp_limit, + .get_temp_alarm = mv88e6xxx_get_temp_alarm, +#endif + .get_eeprom_len = mv88e6xxx_get_eeprom_len, + .get_eeprom = mv88e6xxx_get_eeprom, + .set_eeprom = mv88e6xxx_set_eeprom, + .get_regs_len = mv88e6xxx_get_regs_len, + .get_regs = mv88e6xxx_get_regs, + .port_bridge_join = mv88e6xxx_port_bridge_join, + .port_bridge_leave = mv88e6xxx_port_bridge_leave, + .port_stp_state_set = mv88e6xxx_port_stp_state_set, + .port_vlan_filtering = mv88e6xxx_port_vlan_filtering, + .port_vlan_prepare = mv88e6xxx_port_vlan_prepare, + .port_vlan_add = mv88e6xxx_port_vlan_add, + .port_vlan_del = mv88e6xxx_port_vlan_del, + .port_vlan_dump = mv88e6xxx_port_vlan_dump, + .port_fdb_prepare = mv88e6xxx_port_fdb_prepare, + .port_fdb_add = mv88e6xxx_port_fdb_add, + .port_fdb_del = mv88e6xxx_port_fdb_del, + .port_fdb_dump = mv88e6xxx_port_fdb_dump, +}; + +static int mv88e6xxx_register_switch(struct mv88e6xxx_chip *chip, + struct device_node *np) +{ + struct device *dev = chip->dev; + struct dsa_switch *ds; + + ds = devm_kzalloc(dev, sizeof(*ds), GFP_KERNEL); + if (!ds) + return -ENOMEM; + + ds->dev = dev; + ds->priv = chip; + ds->drv = &mv88e6xxx_switch_driver; + + dev_set_drvdata(dev, ds); + + return dsa_register_switch(ds, np); +} + +static void mv88e6xxx_unregister_switch(struct mv88e6xxx_chip *chip) +{ + dsa_unregister_switch(chip->ds); +} + +static int mv88e6xxx_probe(struct mdio_device *mdiodev) +{ + struct device *dev = &mdiodev->dev; + struct device_node *np = dev->of_node; + const struct mv88e6xxx_info *compat_info; + struct mv88e6xxx_chip *chip; + u32 eeprom_len; + int err; + + compat_info = of_device_get_match_data(dev); + if (!compat_info) + return -EINVAL; + + chip = mv88e6xxx_alloc_chip(dev); + if (!chip) + return -ENOMEM; + + chip->info = compat_info; + + err = mv88e6xxx_smi_init(chip, mdiodev->bus, mdiodev->addr); + if (err) + return err; + + err = mv88e6xxx_detect(chip); + if (err) + return err; + + chip->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_ASIS); + if (IS_ERR(chip->reset)) + return PTR_ERR(chip->reset); + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEPROM) && + !of_property_read_u32(np, "eeprom-length", &eeprom_len)) + chip->eeprom_len = eeprom_len; + + err = mv88e6xxx_mdio_register(chip, np); + if (err) + return err; + + err = mv88e6xxx_register_switch(chip, np); + if (err) { + mv88e6xxx_mdio_unregister(chip); + return err; + } + + return 0; +} + +static void mv88e6xxx_remove(struct mdio_device *mdiodev) +{ + struct dsa_switch *ds = dev_get_drvdata(&mdiodev->dev); + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + + mv88e6xxx_unregister_switch(chip); + mv88e6xxx_mdio_unregister(chip); +} + +static const struct of_device_id mv88e6xxx_of_match[] = { + { + .compatible = "marvell,mv88e6085", + .data = &mv88e6xxx_table[MV88E6085], + }, + { /* sentinel */ }, +}; + +MODULE_DEVICE_TABLE(of, mv88e6xxx_of_match); + +static struct mdio_driver mv88e6xxx_driver = { + .probe = mv88e6xxx_probe, + .remove = mv88e6xxx_remove, + .mdiodrv.driver = { + .name = "mv88e6085", + .of_match_table = mv88e6xxx_of_match, + }, +}; + +static int __init mv88e6xxx_init(void) +{ + register_switch_driver(&mv88e6xxx_switch_driver); + return mdio_driver_register(&mv88e6xxx_driver); +} +module_init(mv88e6xxx_init); + +static void __exit mv88e6xxx_cleanup(void) +{ + mdio_driver_unregister(&mv88e6xxx_driver); + unregister_switch_driver(&mv88e6xxx_switch_driver); +} +module_exit(mv88e6xxx_cleanup); + +MODULE_AUTHOR("Lennert Buytenhek "); +MODULE_DESCRIPTION("Driver for Marvell 88E6XXX ethernet switch chips"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.c deleted file mode 100644 index 2073f7b..0000000 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.c +++ /dev/null @@ -1,3954 +0,0 @@ -/* - * Marvell 88e6xxx Ethernet switch single-chip support - * - * Copyright (c) 2008 Marvell Semiconductor - * - * Copyright (c) 2015 CMC Electronics, Inc. - * Added support for VLAN Table Unit operations - * - * Copyright (c) 2016 Andrew Lunn - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "mv88e6xxx.h" - -static void assert_reg_lock(struct mv88e6xxx_priv_state *ps) -{ - if (unlikely(!mutex_is_locked(&ps->reg_lock))) { - dev_err(ps->dev, "Switch registers lock not held!\n"); - dump_stack(); - } -} - -/* The switch ADDR[4:1] configuration pins define the chip SMI device address - * (ADDR[0] is always zero, thus only even SMI addresses can be strapped). - * - * When ADDR is all zero, the chip uses Single-chip Addressing Mode, assuming it - * is the only device connected to the SMI master. In this mode it responds to - * all 32 possible SMI addresses, and thus maps directly the internal devices. - * - * When ADDR is non-zero, the chip uses Multi-chip Addressing Mode, allowing - * multiple devices to share the SMI interface. In this mode it responds to only - * 2 registers, used to indirectly access the internal SMI devices. - */ - -static int mv88e6xxx_smi_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 *val) -{ - if (!ps->smi_ops) - return -EOPNOTSUPP; - - return ps->smi_ops->read(ps, addr, reg, val); -} - -static int mv88e6xxx_smi_write(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 val) -{ - if (!ps->smi_ops) - return -EOPNOTSUPP; - - return ps->smi_ops->write(ps, addr, reg, val); -} - -static int mv88e6xxx_smi_single_chip_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 *val) -{ - int ret; - - ret = mdiobus_read_nested(ps->bus, addr, reg); - if (ret < 0) - return ret; - - *val = ret & 0xffff; - - return 0; -} - -static int mv88e6xxx_smi_single_chip_write(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 val) -{ - int ret; - - ret = mdiobus_write_nested(ps->bus, addr, reg, val); - if (ret < 0) - return ret; - - return 0; -} - -static const struct mv88e6xxx_ops mv88e6xxx_smi_single_chip_ops = { - .read = mv88e6xxx_smi_single_chip_read, - .write = mv88e6xxx_smi_single_chip_write, -}; - -static int mv88e6xxx_smi_multi_chip_wait(struct mv88e6xxx_priv_state *ps) -{ - int ret; - int i; - - for (i = 0; i < 16; i++) { - ret = mdiobus_read_nested(ps->bus, ps->sw_addr, SMI_CMD); - if (ret < 0) - return ret; - - if ((ret & SMI_CMD_BUSY) == 0) - return 0; - } - - return -ETIMEDOUT; -} - -static int mv88e6xxx_smi_multi_chip_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 *val) -{ - int ret; - - /* Wait for the bus to become free. */ - ret = mv88e6xxx_smi_multi_chip_wait(ps); - if (ret < 0) - return ret; - - /* Transmit the read command. */ - ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_CMD, - SMI_CMD_OP_22_READ | (addr << 5) | reg); - if (ret < 0) - return ret; - - /* Wait for the read command to complete. */ - ret = mv88e6xxx_smi_multi_chip_wait(ps); - if (ret < 0) - return ret; - - /* Read the data. */ - ret = mdiobus_read_nested(ps->bus, ps->sw_addr, SMI_DATA); - if (ret < 0) - return ret; - - *val = ret & 0xffff; - - return 0; -} - -static int mv88e6xxx_smi_multi_chip_write(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 val) -{ - int ret; - - /* Wait for the bus to become free. */ - ret = mv88e6xxx_smi_multi_chip_wait(ps); - if (ret < 0) - return ret; - - /* Transmit the data to write. */ - ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_DATA, val); - if (ret < 0) - return ret; - - /* Transmit the write command. */ - ret = mdiobus_write_nested(ps->bus, ps->sw_addr, SMI_CMD, - SMI_CMD_OP_22_WRITE | (addr << 5) | reg); - if (ret < 0) - return ret; - - /* Wait for the write command to complete. */ - ret = mv88e6xxx_smi_multi_chip_wait(ps); - if (ret < 0) - return ret; - - return 0; -} - -static const struct mv88e6xxx_ops mv88e6xxx_smi_multi_chip_ops = { - .read = mv88e6xxx_smi_multi_chip_read, - .write = mv88e6xxx_smi_multi_chip_write, -}; - -static int mv88e6xxx_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 *val) -{ - int err; - - assert_reg_lock(ps); - - err = mv88e6xxx_smi_read(ps, addr, reg, val); - if (err) - return err; - - dev_dbg(ps->dev, "<- addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", - addr, reg, *val); - - return 0; -} - -static int mv88e6xxx_write(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 val) -{ - int err; - - assert_reg_lock(ps); - - err = mv88e6xxx_smi_write(ps, addr, reg, val); - if (err) - return err; - - dev_dbg(ps->dev, "-> addr: 0x%.2x reg: 0x%.2x val: 0x%.4x\n", - addr, reg, val); - - return 0; -} - -static int _mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, - int addr, int reg) -{ - u16 val; - int err; - - err = mv88e6xxx_read(ps, addr, reg, &val); - if (err) - return err; - - return val; -} - -static int mv88e6xxx_reg_read(struct mv88e6xxx_priv_state *ps, int addr, - int reg) -{ - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_reg_read(ps, addr, reg); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int _mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, - int reg, u16 val) -{ - return mv88e6xxx_write(ps, addr, reg, val); -} - -static int mv88e6xxx_reg_write(struct mv88e6xxx_priv_state *ps, int addr, - int reg, u16 val) -{ - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_reg_write(ps, addr, reg, val); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int mv88e6xxx_set_addr_direct(struct dsa_switch *ds, u8 *addr) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int err; - - err = mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MAC_01, - (addr[0] << 8) | addr[1]); - if (err) - return err; - - err = mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MAC_23, - (addr[2] << 8) | addr[3]); - if (err) - return err; - - return mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MAC_45, - (addr[4] << 8) | addr[5]); -} - -static int mv88e6xxx_set_addr_indirect(struct dsa_switch *ds, u8 *addr) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - int i; - - for (i = 0; i < 6; i++) { - int j; - - /* Write the MAC address byte. */ - ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SWITCH_MAC, - GLOBAL2_SWITCH_MAC_BUSY | - (i << 8) | addr[i]); - if (ret) - return ret; - - /* Wait for the write to complete. */ - for (j = 0; j < 16; j++) { - ret = mv88e6xxx_reg_read(ps, REG_GLOBAL2, - GLOBAL2_SWITCH_MAC); - if (ret < 0) - return ret; - - if ((ret & GLOBAL2_SWITCH_MAC_BUSY) == 0) - break; - } - if (j == 16) - return -ETIMEDOUT; - } - - return 0; -} - -static int mv88e6xxx_set_addr(struct dsa_switch *ds, u8 *addr) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SWITCH_MAC)) - return mv88e6xxx_set_addr_indirect(ds, addr); - else - return mv88e6xxx_set_addr_direct(ds, addr); -} - -static int mv88e6xxx_mdio_read_direct(struct mv88e6xxx_priv_state *ps, - int addr, int regnum) -{ - if (addr >= 0) - return _mv88e6xxx_reg_read(ps, addr, regnum); - return 0xffff; -} - -static int mv88e6xxx_mdio_write_direct(struct mv88e6xxx_priv_state *ps, - int addr, int regnum, u16 val) -{ - if (addr >= 0) - return _mv88e6xxx_reg_write(ps, addr, regnum, val); - return 0; -} - -static int mv88e6xxx_ppu_disable(struct mv88e6xxx_priv_state *ps) -{ - int ret; - unsigned long timeout; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_CONTROL); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, - ret & ~GLOBAL_CONTROL_PPU_ENABLE); - if (ret) - return ret; - - timeout = jiffies + 1 * HZ; - while (time_before(jiffies, timeout)) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATUS); - if (ret < 0) - return ret; - - usleep_range(1000, 2000); - if ((ret & GLOBAL_STATUS_PPU_MASK) != - GLOBAL_STATUS_PPU_POLLING) - return 0; - } - - return -ETIMEDOUT; -} - -static int mv88e6xxx_ppu_enable(struct mv88e6xxx_priv_state *ps) -{ - int ret, err; - unsigned long timeout; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_CONTROL); - if (ret < 0) - return ret; - - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, - ret | GLOBAL_CONTROL_PPU_ENABLE); - if (err) - return err; - - timeout = jiffies + 1 * HZ; - while (time_before(jiffies, timeout)) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATUS); - if (ret < 0) - return ret; - - usleep_range(1000, 2000); - if ((ret & GLOBAL_STATUS_PPU_MASK) == - GLOBAL_STATUS_PPU_POLLING) - return 0; - } - - return -ETIMEDOUT; -} - -static void mv88e6xxx_ppu_reenable_work(struct work_struct *ugly) -{ - struct mv88e6xxx_priv_state *ps; - - ps = container_of(ugly, struct mv88e6xxx_priv_state, ppu_work); - - mutex_lock(&ps->reg_lock); - - if (mutex_trylock(&ps->ppu_mutex)) { - if (mv88e6xxx_ppu_enable(ps) == 0) - ps->ppu_disabled = 0; - mutex_unlock(&ps->ppu_mutex); - } - - mutex_unlock(&ps->reg_lock); -} - -static void mv88e6xxx_ppu_reenable_timer(unsigned long _ps) -{ - struct mv88e6xxx_priv_state *ps = (void *)_ps; - - schedule_work(&ps->ppu_work); -} - -static int mv88e6xxx_ppu_access_get(struct mv88e6xxx_priv_state *ps) -{ - int ret; - - mutex_lock(&ps->ppu_mutex); - - /* If the PHY polling unit is enabled, disable it so that - * we can access the PHY registers. If it was already - * disabled, cancel the timer that is going to re-enable - * it. - */ - if (!ps->ppu_disabled) { - ret = mv88e6xxx_ppu_disable(ps); - if (ret < 0) { - mutex_unlock(&ps->ppu_mutex); - return ret; - } - ps->ppu_disabled = 1; - } else { - del_timer(&ps->ppu_timer); - ret = 0; - } - - return ret; -} - -static void mv88e6xxx_ppu_access_put(struct mv88e6xxx_priv_state *ps) -{ - /* Schedule a timer to re-enable the PHY polling unit. */ - mod_timer(&ps->ppu_timer, jiffies + msecs_to_jiffies(10)); - mutex_unlock(&ps->ppu_mutex); -} - -static void mv88e6xxx_ppu_state_init(struct mv88e6xxx_priv_state *ps) -{ - mutex_init(&ps->ppu_mutex); - INIT_WORK(&ps->ppu_work, mv88e6xxx_ppu_reenable_work); - init_timer(&ps->ppu_timer); - ps->ppu_timer.data = (unsigned long)ps; - ps->ppu_timer.function = mv88e6xxx_ppu_reenable_timer; -} - -static int mv88e6xxx_mdio_read_ppu(struct mv88e6xxx_priv_state *ps, int addr, - int regnum) -{ - int ret; - - ret = mv88e6xxx_ppu_access_get(ps); - if (ret >= 0) { - ret = _mv88e6xxx_reg_read(ps, addr, regnum); - mv88e6xxx_ppu_access_put(ps); - } - - return ret; -} - -static int mv88e6xxx_mdio_write_ppu(struct mv88e6xxx_priv_state *ps, int addr, - int regnum, u16 val) -{ - int ret; - - ret = mv88e6xxx_ppu_access_get(ps); - if (ret >= 0) { - ret = _mv88e6xxx_reg_write(ps, addr, regnum, val); - mv88e6xxx_ppu_access_put(ps); - } - - return ret; -} - -static bool mv88e6xxx_6065_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6065; -} - -static bool mv88e6xxx_6095_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6095; -} - -static bool mv88e6xxx_6097_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6097; -} - -static bool mv88e6xxx_6165_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6165; -} - -static bool mv88e6xxx_6185_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6185; -} - -static bool mv88e6xxx_6320_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6320; -} - -static bool mv88e6xxx_6351_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6351; -} - -static bool mv88e6xxx_6352_family(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->family == MV88E6XXX_FAMILY_6352; -} - -static unsigned int mv88e6xxx_num_databases(struct mv88e6xxx_priv_state *ps) -{ - return ps->info->num_databases; -} - -static bool mv88e6xxx_has_fid_reg(struct mv88e6xxx_priv_state *ps) -{ - /* Does the device have dedicated FID registers for ATU and VTU ops? */ - if (mv88e6xxx_6097_family(ps) || mv88e6xxx_6165_family(ps) || - mv88e6xxx_6351_family(ps) || mv88e6xxx_6352_family(ps)) - return true; - - return false; -} - -/* We expect the switch to perform auto negotiation if there is a real - * phy. However, in the case of a fixed link phy, we force the port - * settings from the fixed link settings. - */ -static void mv88e6xxx_adjust_link(struct dsa_switch *ds, int port, - struct phy_device *phydev) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - u32 reg; - int ret; - - if (!phy_is_pseudo_fixed_link(phydev)) - return; - - mutex_lock(&ps->reg_lock); - - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_PCS_CTRL); - if (ret < 0) - goto out; - - reg = ret & ~(PORT_PCS_CTRL_LINK_UP | - PORT_PCS_CTRL_FORCE_LINK | - PORT_PCS_CTRL_DUPLEX_FULL | - PORT_PCS_CTRL_FORCE_DUPLEX | - PORT_PCS_CTRL_UNFORCED); - - reg |= PORT_PCS_CTRL_FORCE_LINK; - if (phydev->link) - reg |= PORT_PCS_CTRL_LINK_UP; - - if (mv88e6xxx_6065_family(ps) && phydev->speed > SPEED_100) - goto out; - - switch (phydev->speed) { - case SPEED_1000: - reg |= PORT_PCS_CTRL_1000; - break; - case SPEED_100: - reg |= PORT_PCS_CTRL_100; - break; - case SPEED_10: - reg |= PORT_PCS_CTRL_10; - break; - default: - pr_info("Unknown speed"); - goto out; - } - - reg |= PORT_PCS_CTRL_FORCE_DUPLEX; - if (phydev->duplex == DUPLEX_FULL) - reg |= PORT_PCS_CTRL_DUPLEX_FULL; - - if ((mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps)) && - (port >= ps->info->num_ports - 2)) { - if (phydev->interface == PHY_INTERFACE_MODE_RGMII_RXID) - reg |= PORT_PCS_CTRL_RGMII_DELAY_RXCLK; - if (phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID) - reg |= PORT_PCS_CTRL_RGMII_DELAY_TXCLK; - if (phydev->interface == PHY_INTERFACE_MODE_RGMII_ID) - reg |= (PORT_PCS_CTRL_RGMII_DELAY_RXCLK | - PORT_PCS_CTRL_RGMII_DELAY_TXCLK); - } - _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_PCS_CTRL, reg); - -out: - mutex_unlock(&ps->reg_lock); -} - -static int _mv88e6xxx_stats_wait(struct mv88e6xxx_priv_state *ps) -{ - int ret; - int i; - - for (i = 0; i < 10; i++) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATS_OP); - if ((ret & GLOBAL_STATS_OP_BUSY) == 0) - return 0; - } - - return -ETIMEDOUT; -} - -static int _mv88e6xxx_stats_snapshot(struct mv88e6xxx_priv_state *ps, - int port) -{ - int ret; - - if (mv88e6xxx_6320_family(ps) || mv88e6xxx_6352_family(ps)) - port = (port + 1) << 5; - - /* Snapshot the hardware statistics counters for this port. */ - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_STATS_OP, - GLOBAL_STATS_OP_CAPTURE_PORT | - GLOBAL_STATS_OP_HIST_RX_TX | port); - if (ret < 0) - return ret; - - /* Wait for the snapshotting to complete. */ - ret = _mv88e6xxx_stats_wait(ps); - if (ret < 0) - return ret; - - return 0; -} - -static void _mv88e6xxx_stats_read(struct mv88e6xxx_priv_state *ps, - int stat, u32 *val) -{ - u32 _val; - int ret; - - *val = 0; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_STATS_OP, - GLOBAL_STATS_OP_READ_CAPTURED | - GLOBAL_STATS_OP_HIST_RX_TX | stat); - if (ret < 0) - return; - - ret = _mv88e6xxx_stats_wait(ps); - if (ret < 0) - return; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATS_COUNTER_32); - if (ret < 0) - return; - - _val = ret << 16; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_STATS_COUNTER_01); - if (ret < 0) - return; - - *val = _val | ret; -} - -static struct mv88e6xxx_hw_stat mv88e6xxx_hw_stats[] = { - { "in_good_octets", 8, 0x00, BANK0, }, - { "in_bad_octets", 4, 0x02, BANK0, }, - { "in_unicast", 4, 0x04, BANK0, }, - { "in_broadcasts", 4, 0x06, BANK0, }, - { "in_multicasts", 4, 0x07, BANK0, }, - { "in_pause", 4, 0x16, BANK0, }, - { "in_undersize", 4, 0x18, BANK0, }, - { "in_fragments", 4, 0x19, BANK0, }, - { "in_oversize", 4, 0x1a, BANK0, }, - { "in_jabber", 4, 0x1b, BANK0, }, - { "in_rx_error", 4, 0x1c, BANK0, }, - { "in_fcs_error", 4, 0x1d, BANK0, }, - { "out_octets", 8, 0x0e, BANK0, }, - { "out_unicast", 4, 0x10, BANK0, }, - { "out_broadcasts", 4, 0x13, BANK0, }, - { "out_multicasts", 4, 0x12, BANK0, }, - { "out_pause", 4, 0x15, BANK0, }, - { "excessive", 4, 0x11, BANK0, }, - { "collisions", 4, 0x1e, BANK0, }, - { "deferred", 4, 0x05, BANK0, }, - { "single", 4, 0x14, BANK0, }, - { "multiple", 4, 0x17, BANK0, }, - { "out_fcs_error", 4, 0x03, BANK0, }, - { "late", 4, 0x1f, BANK0, }, - { "hist_64bytes", 4, 0x08, BANK0, }, - { "hist_65_127bytes", 4, 0x09, BANK0, }, - { "hist_128_255bytes", 4, 0x0a, BANK0, }, - { "hist_256_511bytes", 4, 0x0b, BANK0, }, - { "hist_512_1023bytes", 4, 0x0c, BANK0, }, - { "hist_1024_max_bytes", 4, 0x0d, BANK0, }, - { "sw_in_discards", 4, 0x10, PORT, }, - { "sw_in_filtered", 2, 0x12, PORT, }, - { "sw_out_filtered", 2, 0x13, PORT, }, - { "in_discards", 4, 0x00 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_filtered", 4, 0x01 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_accepted", 4, 0x02 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_bad_accepted", 4, 0x03 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_good_avb_class_a", 4, 0x04 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_good_avb_class_b", 4, 0x05 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_bad_avb_class_a", 4, 0x06 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_bad_avb_class_b", 4, 0x07 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "tcam_counter_0", 4, 0x08 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "tcam_counter_1", 4, 0x09 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "tcam_counter_2", 4, 0x0a | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "tcam_counter_3", 4, 0x0b | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_da_unknown", 4, 0x0e | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "in_management", 4, 0x0f | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_0", 4, 0x10 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_1", 4, 0x11 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_2", 4, 0x12 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_3", 4, 0x13 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_4", 4, 0x14 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_5", 4, 0x15 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_6", 4, 0x16 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_queue_7", 4, 0x17 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_cut_through", 4, 0x18 | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_octets_a", 4, 0x1a | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_octets_b", 4, 0x1b | GLOBAL_STATS_OP_BANK_1, BANK1, }, - { "out_management", 4, 0x1f | GLOBAL_STATS_OP_BANK_1, BANK1, }, -}; - -static bool mv88e6xxx_has_stat(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_hw_stat *stat) -{ - switch (stat->type) { - case BANK0: - return true; - case BANK1: - return mv88e6xxx_6320_family(ps); - case PORT: - return mv88e6xxx_6095_family(ps) || - mv88e6xxx_6185_family(ps) || - mv88e6xxx_6097_family(ps) || - mv88e6xxx_6165_family(ps) || - mv88e6xxx_6351_family(ps) || - mv88e6xxx_6352_family(ps); - } - return false; -} - -static uint64_t _mv88e6xxx_get_ethtool_stat(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_hw_stat *s, - int port) -{ - u32 low; - u32 high = 0; - int ret; - u64 value; - - switch (s->type) { - case PORT: - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), s->reg); - if (ret < 0) - return UINT64_MAX; - - low = ret; - if (s->sizeof_stat == 4) { - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), - s->reg + 1); - if (ret < 0) - return UINT64_MAX; - high = ret; - } - break; - case BANK0: - case BANK1: - _mv88e6xxx_stats_read(ps, s->reg, &low); - if (s->sizeof_stat == 8) - _mv88e6xxx_stats_read(ps, s->reg + 1, &high); - } - value = (((u64)high) << 16) | low; - return value; -} - -static void mv88e6xxx_get_strings(struct dsa_switch *ds, int port, - uint8_t *data) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_hw_stat *stat; - int i, j; - - for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { - stat = &mv88e6xxx_hw_stats[i]; - if (mv88e6xxx_has_stat(ps, stat)) { - memcpy(data + j * ETH_GSTRING_LEN, stat->string, - ETH_GSTRING_LEN); - j++; - } - } -} - -static int mv88e6xxx_get_sset_count(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_hw_stat *stat; - int i, j; - - for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { - stat = &mv88e6xxx_hw_stats[i]; - if (mv88e6xxx_has_stat(ps, stat)) - j++; - } - return j; -} - -static void mv88e6xxx_get_ethtool_stats(struct dsa_switch *ds, int port, - uint64_t *data) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_hw_stat *stat; - int ret; - int i, j; - - mutex_lock(&ps->reg_lock); - - ret = _mv88e6xxx_stats_snapshot(ps, port); - if (ret < 0) { - mutex_unlock(&ps->reg_lock); - return; - } - for (i = 0, j = 0; i < ARRAY_SIZE(mv88e6xxx_hw_stats); i++) { - stat = &mv88e6xxx_hw_stats[i]; - if (mv88e6xxx_has_stat(ps, stat)) { - data[j] = _mv88e6xxx_get_ethtool_stat(ps, stat, port); - j++; - } - } - - mutex_unlock(&ps->reg_lock); -} - -static int mv88e6xxx_get_regs_len(struct dsa_switch *ds, int port) -{ - return 32 * sizeof(u16); -} - -static void mv88e6xxx_get_regs(struct dsa_switch *ds, int port, - struct ethtool_regs *regs, void *_p) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - u16 *p = _p; - int i; - - regs->version = 0; - - memset(p, 0xff, 32 * sizeof(u16)); - - mutex_lock(&ps->reg_lock); - - for (i = 0; i < 32; i++) { - int ret; - - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), i); - if (ret >= 0) - p[i] = ret; - } - - mutex_unlock(&ps->reg_lock); -} - -static int _mv88e6xxx_wait(struct mv88e6xxx_priv_state *ps, int reg, int offset, - u16 mask) -{ - unsigned long timeout = jiffies + HZ / 10; - - while (time_before(jiffies, timeout)) { - int ret; - - ret = _mv88e6xxx_reg_read(ps, reg, offset); - if (ret < 0) - return ret; - if (!(ret & mask)) - return 0; - - usleep_range(1000, 2000); - } - return -ETIMEDOUT; -} - -static int mv88e6xxx_wait(struct mv88e6xxx_priv_state *ps, int reg, - int offset, u16 mask) -{ - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_wait(ps, reg, offset, mask); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int mv88e6xxx_mdio_wait(struct mv88e6xxx_priv_state *ps) -{ - return _mv88e6xxx_wait(ps, REG_GLOBAL2, GLOBAL2_SMI_OP, - GLOBAL2_SMI_OP_BUSY); -} - -static int mv88e6xxx_eeprom_load_wait(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - return mv88e6xxx_wait(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_LOAD); -} - -static int mv88e6xxx_eeprom_busy_wait(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - return mv88e6xxx_wait(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_BUSY); -} - -static int mv88e6xxx_read_eeprom_word(struct dsa_switch *ds, int addr) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - mutex_lock(&ps->eeprom_mutex); - - ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_READ | - (addr & GLOBAL2_EEPROM_OP_ADDR_MASK)); - if (ret < 0) - goto error; - - ret = mv88e6xxx_eeprom_busy_wait(ds); - if (ret < 0) - goto error; - - ret = mv88e6xxx_reg_read(ps, REG_GLOBAL2, GLOBAL2_EEPROM_DATA); -error: - mutex_unlock(&ps->eeprom_mutex); - return ret; -} - -static int mv88e6xxx_get_eeprom_len(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) - return ps->eeprom_len; - - return 0; -} - -static int mv88e6xxx_get_eeprom(struct dsa_switch *ds, - struct ethtool_eeprom *eeprom, u8 *data) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int offset; - int len; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) - return -EOPNOTSUPP; - - offset = eeprom->offset; - len = eeprom->len; - eeprom->len = 0; - - eeprom->magic = 0xc3ec4951; - - ret = mv88e6xxx_eeprom_load_wait(ds); - if (ret < 0) - return ret; - - if (offset & 1) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - *data++ = (word >> 8) & 0xff; - - offset++; - len--; - eeprom->len++; - } - - while (len >= 2) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - *data++ = word & 0xff; - *data++ = (word >> 8) & 0xff; - - offset += 2; - len -= 2; - eeprom->len += 2; - } - - if (len) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - *data++ = word & 0xff; - - offset++; - len--; - eeprom->len++; - } - - return 0; -} - -static int mv88e6xxx_eeprom_is_readonly(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - ret = mv88e6xxx_reg_read(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP); - if (ret < 0) - return ret; - - if (!(ret & GLOBAL2_EEPROM_OP_WRITE_EN)) - return -EROFS; - - return 0; -} - -static int mv88e6xxx_write_eeprom_word(struct dsa_switch *ds, int addr, - u16 data) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - mutex_lock(&ps->eeprom_mutex); - - ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_EEPROM_DATA, data); - if (ret < 0) - goto error; - - ret = mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_WRITE | - (addr & GLOBAL2_EEPROM_OP_ADDR_MASK)); - if (ret < 0) - goto error; - - ret = mv88e6xxx_eeprom_busy_wait(ds); -error: - mutex_unlock(&ps->eeprom_mutex); - return ret; -} - -static int mv88e6xxx_set_eeprom(struct dsa_switch *ds, - struct ethtool_eeprom *eeprom, u8 *data) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int offset; - int ret; - int len; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) - return -EOPNOTSUPP; - - if (eeprom->magic != 0xc3ec4951) - return -EINVAL; - - ret = mv88e6xxx_eeprom_is_readonly(ds); - if (ret) - return ret; - - offset = eeprom->offset; - len = eeprom->len; - eeprom->len = 0; - - ret = mv88e6xxx_eeprom_load_wait(ds); - if (ret < 0) - return ret; - - if (offset & 1) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - word = (*data++ << 8) | (word & 0xff); - - ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); - if (ret < 0) - return ret; - - offset++; - len--; - eeprom->len++; - } - - while (len >= 2) { - int word; - - word = *data++; - word |= *data++ << 8; - - ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); - if (ret < 0) - return ret; - - offset += 2; - len -= 2; - eeprom->len += 2; - } - - if (len) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - word = (word & 0xff00) | *data++; - - ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); - if (ret < 0) - return ret; - - offset++; - len--; - eeprom->len++; - } - - return 0; -} - -static int _mv88e6xxx_atu_wait(struct mv88e6xxx_priv_state *ps) -{ - return _mv88e6xxx_wait(ps, REG_GLOBAL, GLOBAL_ATU_OP, - GLOBAL_ATU_OP_BUSY); -} - -static int mv88e6xxx_mdio_read_indirect(struct mv88e6xxx_priv_state *ps, - int addr, int regnum) -{ - int ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SMI_OP, - GLOBAL2_SMI_OP_22_READ | (addr << 5) | - regnum); - if (ret < 0) - return ret; - - ret = mv88e6xxx_mdio_wait(ps); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL2, GLOBAL2_SMI_DATA); - - return ret; -} - -static int mv88e6xxx_mdio_write_indirect(struct mv88e6xxx_priv_state *ps, - int addr, int regnum, u16 val) -{ - int ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SMI_DATA, val); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SMI_OP, - GLOBAL2_SMI_OP_22_WRITE | (addr << 5) | - regnum); - - return mv88e6xxx_mdio_wait(ps); -} - -static int mv88e6xxx_get_eee(struct dsa_switch *ds, int port, - struct ethtool_eee *e) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int reg; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEE)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - reg = mv88e6xxx_mdio_read_indirect(ps, port, 16); - if (reg < 0) - goto out; - - e->eee_enabled = !!(reg & 0x0200); - e->tx_lpi_enabled = !!(reg & 0x0100); - - reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_STATUS); - if (reg < 0) - goto out; - - e->eee_active = !!(reg & PORT_STATUS_EEE); - reg = 0; - -out: - mutex_unlock(&ps->reg_lock); - return reg; -} - -static int mv88e6xxx_set_eee(struct dsa_switch *ds, int port, - struct phy_device *phydev, struct ethtool_eee *e) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int reg; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEE)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - ret = mv88e6xxx_mdio_read_indirect(ps, port, 16); - if (ret < 0) - goto out; - - reg = ret & ~0x0300; - if (e->eee_enabled) - reg |= 0x0200; - if (e->tx_lpi_enabled) - reg |= 0x0100; - - ret = mv88e6xxx_mdio_write_indirect(ps, port, 16, reg); -out: - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int _mv88e6xxx_atu_cmd(struct mv88e6xxx_priv_state *ps, u16 fid, u16 cmd) -{ - int ret; - - if (mv88e6xxx_has_fid_reg(ps)) { - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_FID, fid); - if (ret < 0) - return ret; - } else if (mv88e6xxx_num_databases(ps) == 256) { - /* ATU DBNum[7:4] are located in ATU Control 15:12 */ - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_ATU_CONTROL); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_CONTROL, - (ret & 0xfff) | - ((fid << 8) & 0xf000)); - if (ret < 0) - return ret; - - /* ATU DBNum[3:0] are located in ATU Operation 3:0 */ - cmd |= fid & 0xf; - } - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_OP, cmd); - if (ret < 0) - return ret; - - return _mv88e6xxx_atu_wait(ps); -} - -static int _mv88e6xxx_atu_data_write(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_atu_entry *entry) -{ - u16 data = entry->state & GLOBAL_ATU_DATA_STATE_MASK; - - if (entry->state != GLOBAL_ATU_DATA_STATE_UNUSED) { - unsigned int mask, shift; - - if (entry->trunk) { - data |= GLOBAL_ATU_DATA_TRUNK; - mask = GLOBAL_ATU_DATA_TRUNK_ID_MASK; - shift = GLOBAL_ATU_DATA_TRUNK_ID_SHIFT; - } else { - mask = GLOBAL_ATU_DATA_PORT_VECTOR_MASK; - shift = GLOBAL_ATU_DATA_PORT_VECTOR_SHIFT; - } - - data |= (entry->portv_trunkid << shift) & mask; - } - - return _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_DATA, data); -} - -static int _mv88e6xxx_atu_flush_move(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_atu_entry *entry, - bool static_too) -{ - int op; - int err; - - err = _mv88e6xxx_atu_wait(ps); - if (err) - return err; - - err = _mv88e6xxx_atu_data_write(ps, entry); - if (err) - return err; - - if (entry->fid) { - op = static_too ? GLOBAL_ATU_OP_FLUSH_MOVE_ALL_DB : - GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC_DB; - } else { - op = static_too ? GLOBAL_ATU_OP_FLUSH_MOVE_ALL : - GLOBAL_ATU_OP_FLUSH_MOVE_NON_STATIC; - } - - return _mv88e6xxx_atu_cmd(ps, entry->fid, op); -} - -static int _mv88e6xxx_atu_flush(struct mv88e6xxx_priv_state *ps, - u16 fid, bool static_too) -{ - struct mv88e6xxx_atu_entry entry = { - .fid = fid, - .state = 0, /* EntryState bits must be 0 */ - }; - - return _mv88e6xxx_atu_flush_move(ps, &entry, static_too); -} - -static int _mv88e6xxx_atu_move(struct mv88e6xxx_priv_state *ps, u16 fid, - int from_port, int to_port, bool static_too) -{ - struct mv88e6xxx_atu_entry entry = { - .trunk = false, - .fid = fid, - }; - - /* EntryState bits must be 0xF */ - entry.state = GLOBAL_ATU_DATA_STATE_MASK; - - /* ToPort and FromPort are respectively in PortVec bits 7:4 and 3:0 */ - entry.portv_trunkid = (to_port & 0x0f) << 4; - entry.portv_trunkid |= from_port & 0x0f; - - return _mv88e6xxx_atu_flush_move(ps, &entry, static_too); -} - -static int _mv88e6xxx_atu_remove(struct mv88e6xxx_priv_state *ps, u16 fid, - int port, bool static_too) -{ - /* Destination port 0xF means remove the entries */ - return _mv88e6xxx_atu_move(ps, fid, port, 0x0f, static_too); -} - -static const char * const mv88e6xxx_port_state_names[] = { - [PORT_CONTROL_STATE_DISABLED] = "Disabled", - [PORT_CONTROL_STATE_BLOCKING] = "Blocking/Listening", - [PORT_CONTROL_STATE_LEARNING] = "Learning", - [PORT_CONTROL_STATE_FORWARDING] = "Forwarding", -}; - -static int _mv88e6xxx_port_state(struct mv88e6xxx_priv_state *ps, int port, - u8 state) -{ - struct dsa_switch *ds = ps->ds; - int reg, ret = 0; - u8 oldstate; - - reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_CONTROL); - if (reg < 0) - return reg; - - oldstate = reg & PORT_CONTROL_STATE_MASK; - - if (oldstate != state) { - /* Flush forwarding database if we're moving a port - * from Learning or Forwarding state to Disabled or - * Blocking or Listening state. - */ - if ((oldstate == PORT_CONTROL_STATE_LEARNING || - oldstate == PORT_CONTROL_STATE_FORWARDING) && - (state == PORT_CONTROL_STATE_DISABLED || - state == PORT_CONTROL_STATE_BLOCKING)) { - ret = _mv88e6xxx_atu_remove(ps, 0, port, false); - if (ret) - return ret; - } - - reg = (reg & ~PORT_CONTROL_STATE_MASK) | state; - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL, - reg); - if (ret) - return ret; - - netdev_dbg(ds->ports[port].netdev, "PortState %s (was %s)\n", - mv88e6xxx_port_state_names[state], - mv88e6xxx_port_state_names[oldstate]); - } - - return ret; -} - -static int _mv88e6xxx_port_based_vlan_map(struct mv88e6xxx_priv_state *ps, - int port) -{ - struct net_device *bridge = ps->ports[port].bridge_dev; - const u16 mask = (1 << ps->info->num_ports) - 1; - struct dsa_switch *ds = ps->ds; - u16 output_ports = 0; - int reg; - int i; - - /* allow CPU port or DSA link(s) to send frames to every port */ - if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) { - output_ports = mask; - } else { - for (i = 0; i < ps->info->num_ports; ++i) { - /* allow sending frames to every group member */ - if (bridge && ps->ports[i].bridge_dev == bridge) - output_ports |= BIT(i); - - /* allow sending frames to CPU port and DSA link(s) */ - if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i)) - output_ports |= BIT(i); - } - } - - /* prevent frames from going back out of the port they came in on */ - output_ports &= ~BIT(port); - - reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_BASE_VLAN); - if (reg < 0) - return reg; - - reg &= ~mask; - reg |= output_ports & mask; - - return _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_BASE_VLAN, reg); -} - -static void mv88e6xxx_port_stp_state_set(struct dsa_switch *ds, int port, - u8 state) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int stp_state; - int err; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_PORTSTATE)) - return; - - switch (state) { - case BR_STATE_DISABLED: - stp_state = PORT_CONTROL_STATE_DISABLED; - break; - case BR_STATE_BLOCKING: - case BR_STATE_LISTENING: - stp_state = PORT_CONTROL_STATE_BLOCKING; - break; - case BR_STATE_LEARNING: - stp_state = PORT_CONTROL_STATE_LEARNING; - break; - case BR_STATE_FORWARDING: - default: - stp_state = PORT_CONTROL_STATE_FORWARDING; - break; - } - - mutex_lock(&ps->reg_lock); - err = _mv88e6xxx_port_state(ps, port, stp_state); - mutex_unlock(&ps->reg_lock); - - if (err) - netdev_err(ds->ports[port].netdev, - "failed to update state to %s\n", - mv88e6xxx_port_state_names[stp_state]); -} - -static int _mv88e6xxx_port_pvid(struct mv88e6xxx_priv_state *ps, int port, - u16 *new, u16 *old) -{ - struct dsa_switch *ds = ps->ds; - u16 pvid; - int ret; - - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_DEFAULT_VLAN); - if (ret < 0) - return ret; - - pvid = ret & PORT_DEFAULT_VLAN_MASK; - - if (new) { - ret &= ~PORT_DEFAULT_VLAN_MASK; - ret |= *new & PORT_DEFAULT_VLAN_MASK; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_DEFAULT_VLAN, ret); - if (ret < 0) - return ret; - - netdev_dbg(ds->ports[port].netdev, - "DefaultVID %d (was %d)\n", *new, pvid); - } - - if (old) - *old = pvid; - - return 0; -} - -static int _mv88e6xxx_port_pvid_get(struct mv88e6xxx_priv_state *ps, - int port, u16 *pvid) -{ - return _mv88e6xxx_port_pvid(ps, port, NULL, pvid); -} - -static int _mv88e6xxx_port_pvid_set(struct mv88e6xxx_priv_state *ps, - int port, u16 pvid) -{ - return _mv88e6xxx_port_pvid(ps, port, &pvid, NULL); -} - -static int _mv88e6xxx_vtu_wait(struct mv88e6xxx_priv_state *ps) -{ - return _mv88e6xxx_wait(ps, REG_GLOBAL, GLOBAL_VTU_OP, - GLOBAL_VTU_OP_BUSY); -} - -static int _mv88e6xxx_vtu_cmd(struct mv88e6xxx_priv_state *ps, u16 op) -{ - int ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_OP, op); - if (ret < 0) - return ret; - - return _mv88e6xxx_vtu_wait(ps); -} - -static int _mv88e6xxx_vtu_stu_flush(struct mv88e6xxx_priv_state *ps) -{ - int ret; - - ret = _mv88e6xxx_vtu_wait(ps); - if (ret < 0) - return ret; - - return _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_FLUSH_ALL); -} - -static int _mv88e6xxx_vtu_stu_data_read(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry, - unsigned int nibble_offset) -{ - u16 regs[3]; - int i; - int ret; - - for (i = 0; i < 3; ++i) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, - GLOBAL_VTU_DATA_0_3 + i); - if (ret < 0) - return ret; - - regs[i] = ret; - } - - for (i = 0; i < ps->info->num_ports; ++i) { - unsigned int shift = (i % 4) * 4 + nibble_offset; - u16 reg = regs[i / 4]; - - entry->data[i] = (reg >> shift) & GLOBAL_VTU_STU_DATA_MASK; - } - - return 0; -} - -static int mv88e6xxx_vtu_data_read(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - return _mv88e6xxx_vtu_stu_data_read(ps, entry, 0); -} - -static int mv88e6xxx_stu_data_read(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - return _mv88e6xxx_vtu_stu_data_read(ps, entry, 2); -} - -static int _mv88e6xxx_vtu_stu_data_write(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry, - unsigned int nibble_offset) -{ - u16 regs[3] = { 0 }; - int i; - int ret; - - for (i = 0; i < ps->info->num_ports; ++i) { - unsigned int shift = (i % 4) * 4 + nibble_offset; - u8 data = entry->data[i]; - - regs[i / 4] |= (data & GLOBAL_VTU_STU_DATA_MASK) << shift; - } - - for (i = 0; i < 3; ++i) { - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, - GLOBAL_VTU_DATA_0_3 + i, regs[i]); - if (ret < 0) - return ret; - } - - return 0; -} - -static int mv88e6xxx_vtu_data_write(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - return _mv88e6xxx_vtu_stu_data_write(ps, entry, 0); -} - -static int mv88e6xxx_stu_data_write(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - return _mv88e6xxx_vtu_stu_data_write(ps, entry, 2); -} - -static int _mv88e6xxx_vtu_vid_write(struct mv88e6xxx_priv_state *ps, u16 vid) -{ - return _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_VID, - vid & GLOBAL_VTU_VID_MASK); -} - -static int _mv88e6xxx_vtu_getnext(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - struct mv88e6xxx_vtu_stu_entry next = { 0 }; - int ret; - - ret = _mv88e6xxx_vtu_wait(ps); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_VTU_GET_NEXT); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_VTU_VID); - if (ret < 0) - return ret; - - next.vid = ret & GLOBAL_VTU_VID_MASK; - next.valid = !!(ret & GLOBAL_VTU_VID_VALID); - - if (next.valid) { - ret = mv88e6xxx_vtu_data_read(ps, &next); - if (ret < 0) - return ret; - - if (mv88e6xxx_has_fid_reg(ps)) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, - GLOBAL_VTU_FID); - if (ret < 0) - return ret; - - next.fid = ret & GLOBAL_VTU_FID_MASK; - } else if (mv88e6xxx_num_databases(ps) == 256) { - /* VTU DBNum[7:4] are located in VTU Operation 11:8, and - * VTU DBNum[3:0] are located in VTU Operation 3:0 - */ - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, - GLOBAL_VTU_OP); - if (ret < 0) - return ret; - - next.fid = (ret & 0xf00) >> 4; - next.fid |= ret & 0xf; - } - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_STU)) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, - GLOBAL_VTU_SID); - if (ret < 0) - return ret; - - next.sid = ret & GLOBAL_VTU_SID_MASK; - } - } - - *entry = next; - return 0; -} - -static int mv88e6xxx_port_vlan_dump(struct dsa_switch *ds, int port, - struct switchdev_obj_port_vlan *vlan, - int (*cb)(struct switchdev_obj *obj)) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_vtu_stu_entry next; - u16 pvid; - int err; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - err = _mv88e6xxx_port_pvid_get(ps, port, &pvid); - if (err) - goto unlock; - - err = _mv88e6xxx_vtu_vid_write(ps, GLOBAL_VTU_VID_MASK); - if (err) - goto unlock; - - do { - err = _mv88e6xxx_vtu_getnext(ps, &next); - if (err) - break; - - if (!next.valid) - break; - - if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) - continue; - - /* reinit and dump this VLAN obj */ - vlan->vid_begin = next.vid; - vlan->vid_end = next.vid; - vlan->flags = 0; - - if (next.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED) - vlan->flags |= BRIDGE_VLAN_INFO_UNTAGGED; - - if (next.vid == pvid) - vlan->flags |= BRIDGE_VLAN_INFO_PVID; - - err = cb(&vlan->obj); - if (err) - break; - } while (next.vid < GLOBAL_VTU_VID_MASK); - -unlock: - mutex_unlock(&ps->reg_lock); - - return err; -} - -static int _mv88e6xxx_vtu_loadpurge(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - u16 op = GLOBAL_VTU_OP_VTU_LOAD_PURGE; - u16 reg = 0; - int ret; - - ret = _mv88e6xxx_vtu_wait(ps); - if (ret < 0) - return ret; - - if (!entry->valid) - goto loadpurge; - - /* Write port member tags */ - ret = mv88e6xxx_vtu_data_write(ps, entry); - if (ret < 0) - return ret; - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_STU)) { - reg = entry->sid & GLOBAL_VTU_SID_MASK; - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_SID, reg); - if (ret < 0) - return ret; - } - - if (mv88e6xxx_has_fid_reg(ps)) { - reg = entry->fid & GLOBAL_VTU_FID_MASK; - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_FID, reg); - if (ret < 0) - return ret; - } else if (mv88e6xxx_num_databases(ps) == 256) { - /* VTU DBNum[7:4] are located in VTU Operation 11:8, and - * VTU DBNum[3:0] are located in VTU Operation 3:0 - */ - op |= (entry->fid & 0xf0) << 8; - op |= entry->fid & 0xf; - } - - reg = GLOBAL_VTU_VID_VALID; -loadpurge: - reg |= entry->vid & GLOBAL_VTU_VID_MASK; - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_VID, reg); - if (ret < 0) - return ret; - - return _mv88e6xxx_vtu_cmd(ps, op); -} - -static int _mv88e6xxx_stu_getnext(struct mv88e6xxx_priv_state *ps, u8 sid, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - struct mv88e6xxx_vtu_stu_entry next = { 0 }; - int ret; - - ret = _mv88e6xxx_vtu_wait(ps); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_SID, - sid & GLOBAL_VTU_SID_MASK); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_STU_GET_NEXT); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_VTU_SID); - if (ret < 0) - return ret; - - next.sid = ret & GLOBAL_VTU_SID_MASK; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_VTU_VID); - if (ret < 0) - return ret; - - next.valid = !!(ret & GLOBAL_VTU_VID_VALID); - - if (next.valid) { - ret = mv88e6xxx_stu_data_read(ps, &next); - if (ret < 0) - return ret; - } - - *entry = next; - return 0; -} - -static int _mv88e6xxx_stu_loadpurge(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - u16 reg = 0; - int ret; - - ret = _mv88e6xxx_vtu_wait(ps); - if (ret < 0) - return ret; - - if (!entry->valid) - goto loadpurge; - - /* Write port states */ - ret = mv88e6xxx_stu_data_write(ps, entry); - if (ret < 0) - return ret; - - reg = GLOBAL_VTU_VID_VALID; -loadpurge: - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_VID, reg); - if (ret < 0) - return ret; - - reg = entry->sid & GLOBAL_VTU_SID_MASK; - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_VTU_SID, reg); - if (ret < 0) - return ret; - - return _mv88e6xxx_vtu_cmd(ps, GLOBAL_VTU_OP_STU_LOAD_PURGE); -} - -static int _mv88e6xxx_port_fid(struct mv88e6xxx_priv_state *ps, int port, - u16 *new, u16 *old) -{ - struct dsa_switch *ds = ps->ds; - u16 upper_mask; - u16 fid; - int ret; - - if (mv88e6xxx_num_databases(ps) == 4096) - upper_mask = 0xff; - else if (mv88e6xxx_num_databases(ps) == 256) - upper_mask = 0xf; - else - return -EOPNOTSUPP; - - /* Port's default FID bits 3:0 are located in reg 0x06, offset 12 */ - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_BASE_VLAN); - if (ret < 0) - return ret; - - fid = (ret & PORT_BASE_VLAN_FID_3_0_MASK) >> 12; - - if (new) { - ret &= ~PORT_BASE_VLAN_FID_3_0_MASK; - ret |= (*new << 12) & PORT_BASE_VLAN_FID_3_0_MASK; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_BASE_VLAN, - ret); - if (ret < 0) - return ret; - } - - /* Port's default FID bits 11:4 are located in reg 0x05, offset 0 */ - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_CONTROL_1); - if (ret < 0) - return ret; - - fid |= (ret & upper_mask) << 4; - - if (new) { - ret &= ~upper_mask; - ret |= (*new >> 4) & upper_mask; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL_1, - ret); - if (ret < 0) - return ret; - - netdev_dbg(ds->ports[port].netdev, - "FID %d (was %d)\n", *new, fid); - } - - if (old) - *old = fid; - - return 0; -} - -static int _mv88e6xxx_port_fid_get(struct mv88e6xxx_priv_state *ps, - int port, u16 *fid) -{ - return _mv88e6xxx_port_fid(ps, port, NULL, fid); -} - -static int _mv88e6xxx_port_fid_set(struct mv88e6xxx_priv_state *ps, - int port, u16 fid) -{ - return _mv88e6xxx_port_fid(ps, port, &fid, NULL); -} - -static int _mv88e6xxx_fid_new(struct mv88e6xxx_priv_state *ps, u16 *fid) -{ - DECLARE_BITMAP(fid_bitmap, MV88E6XXX_N_FID); - struct mv88e6xxx_vtu_stu_entry vlan; - int i, err; - - bitmap_zero(fid_bitmap, MV88E6XXX_N_FID); - - /* Set every FID bit used by the (un)bridged ports */ - for (i = 0; i < ps->info->num_ports; ++i) { - err = _mv88e6xxx_port_fid_get(ps, i, fid); - if (err) - return err; - - set_bit(*fid, fid_bitmap); - } - - /* Set every FID bit used by the VLAN entries */ - err = _mv88e6xxx_vtu_vid_write(ps, GLOBAL_VTU_VID_MASK); - if (err) - return err; - - do { - err = _mv88e6xxx_vtu_getnext(ps, &vlan); - if (err) - return err; - - if (!vlan.valid) - break; - - set_bit(vlan.fid, fid_bitmap); - } while (vlan.vid < GLOBAL_VTU_VID_MASK); - - /* The reset value 0x000 is used to indicate that multiple address - * databases are not needed. Return the next positive available. - */ - *fid = find_next_zero_bit(fid_bitmap, MV88E6XXX_N_FID, 1); - if (unlikely(*fid >= mv88e6xxx_num_databases(ps))) - return -ENOSPC; - - /* Clear the database */ - return _mv88e6xxx_atu_flush(ps, *fid, true); -} - -static int _mv88e6xxx_vtu_new(struct mv88e6xxx_priv_state *ps, u16 vid, - struct mv88e6xxx_vtu_stu_entry *entry) -{ - struct dsa_switch *ds = ps->ds; - struct mv88e6xxx_vtu_stu_entry vlan = { - .valid = true, - .vid = vid, - }; - int i, err; - - err = _mv88e6xxx_fid_new(ps, &vlan.fid); - if (err) - return err; - - /* exclude all ports except the CPU and DSA ports */ - for (i = 0; i < ps->info->num_ports; ++i) - vlan.data[i] = dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i) - ? GLOBAL_VTU_DATA_MEMBER_TAG_UNMODIFIED - : GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER; - - if (mv88e6xxx_6097_family(ps) || mv88e6xxx_6165_family(ps) || - mv88e6xxx_6351_family(ps) || mv88e6xxx_6352_family(ps)) { - struct mv88e6xxx_vtu_stu_entry vstp; - - /* Adding a VTU entry requires a valid STU entry. As VSTP is not - * implemented, only one STU entry is needed to cover all VTU - * entries. Thus, validate the SID 0. - */ - vlan.sid = 0; - err = _mv88e6xxx_stu_getnext(ps, GLOBAL_VTU_SID_MASK, &vstp); - if (err) - return err; - - if (vstp.sid != vlan.sid || !vstp.valid) { - memset(&vstp, 0, sizeof(vstp)); - vstp.valid = true; - vstp.sid = vlan.sid; - - err = _mv88e6xxx_stu_loadpurge(ps, &vstp); - if (err) - return err; - } - } - - *entry = vlan; - return 0; -} - -static int _mv88e6xxx_vtu_get(struct mv88e6xxx_priv_state *ps, u16 vid, - struct mv88e6xxx_vtu_stu_entry *entry, bool creat) -{ - int err; - - if (!vid) - return -EINVAL; - - err = _mv88e6xxx_vtu_vid_write(ps, vid - 1); - if (err) - return err; - - err = _mv88e6xxx_vtu_getnext(ps, entry); - if (err) - return err; - - if (entry->vid != vid || !entry->valid) { - if (!creat) - return -EOPNOTSUPP; - /* -ENOENT would've been more appropriate, but switchdev expects - * -EOPNOTSUPP to inform bridge about an eventual software VLAN. - */ - - err = _mv88e6xxx_vtu_new(ps, vid, entry); - } - - return err; -} - -static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port, - u16 vid_begin, u16 vid_end) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_vtu_stu_entry vlan; - int i, err; - - if (!vid_begin) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - err = _mv88e6xxx_vtu_vid_write(ps, vid_begin - 1); - if (err) - goto unlock; - - do { - err = _mv88e6xxx_vtu_getnext(ps, &vlan); - if (err) - goto unlock; - - if (!vlan.valid) - break; - - if (vlan.vid > vid_end) - break; - - for (i = 0; i < ps->info->num_ports; ++i) { - if (dsa_is_dsa_port(ds, i) || dsa_is_cpu_port(ds, i)) - continue; - - if (vlan.data[i] == - GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) - continue; - - if (ps->ports[i].bridge_dev == - ps->ports[port].bridge_dev) - break; /* same bridge, check next VLAN */ - - netdev_warn(ds->ports[port].netdev, - "hardware VLAN %d already used by %s\n", - vlan.vid, - netdev_name(ps->ports[i].bridge_dev)); - err = -EOPNOTSUPP; - goto unlock; - } - } while (vlan.vid < vid_end); - -unlock: - mutex_unlock(&ps->reg_lock); - - return err; -} - -static const char * const mv88e6xxx_port_8021q_mode_names[] = { - [PORT_CONTROL_2_8021Q_DISABLED] = "Disabled", - [PORT_CONTROL_2_8021Q_FALLBACK] = "Fallback", - [PORT_CONTROL_2_8021Q_CHECK] = "Check", - [PORT_CONTROL_2_8021Q_SECURE] = "Secure", -}; - -static int mv88e6xxx_port_vlan_filtering(struct dsa_switch *ds, int port, - bool vlan_filtering) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - u16 old, new = vlan_filtering ? PORT_CONTROL_2_8021Q_SECURE : - PORT_CONTROL_2_8021Q_DISABLED; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_CONTROL_2); - if (ret < 0) - goto unlock; - - old = ret & PORT_CONTROL_2_8021Q_MASK; - - if (new != old) { - ret &= ~PORT_CONTROL_2_8021Q_MASK; - ret |= new & PORT_CONTROL_2_8021Q_MASK; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL_2, - ret); - if (ret < 0) - goto unlock; - - netdev_dbg(ds->ports[port].netdev, "802.1Q Mode %s (was %s)\n", - mv88e6xxx_port_8021q_mode_names[new], - mv88e6xxx_port_8021q_mode_names[old]); - } - - ret = 0; -unlock: - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int -mv88e6xxx_port_vlan_prepare(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_vlan *vlan, - struct switchdev_trans *trans) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int err; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) - return -EOPNOTSUPP; - - /* If the requested port doesn't belong to the same bridge as the VLAN - * members, do not support it (yet) and fallback to software VLAN. - */ - err = mv88e6xxx_port_check_hw_vlan(ds, port, vlan->vid_begin, - vlan->vid_end); - if (err) - return err; - - /* We don't need any dynamic resource from the kernel (yet), - * so skip the prepare phase. - */ - return 0; -} - -static int _mv88e6xxx_port_vlan_add(struct mv88e6xxx_priv_state *ps, int port, - u16 vid, bool untagged) -{ - struct mv88e6xxx_vtu_stu_entry vlan; - int err; - - err = _mv88e6xxx_vtu_get(ps, vid, &vlan, true); - if (err) - return err; - - vlan.data[port] = untagged ? - GLOBAL_VTU_DATA_MEMBER_TAG_UNTAGGED : - GLOBAL_VTU_DATA_MEMBER_TAG_TAGGED; - - return _mv88e6xxx_vtu_loadpurge(ps, &vlan); -} - -static void mv88e6xxx_port_vlan_add(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_vlan *vlan, - struct switchdev_trans *trans) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - bool untagged = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; - bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID; - u16 vid; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) - return; - - mutex_lock(&ps->reg_lock); - - for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) - if (_mv88e6xxx_port_vlan_add(ps, port, vid, untagged)) - netdev_err(ds->ports[port].netdev, - "failed to add VLAN %d%c\n", - vid, untagged ? 'u' : 't'); - - if (pvid && _mv88e6xxx_port_pvid_set(ps, port, vlan->vid_end)) - netdev_err(ds->ports[port].netdev, "failed to set PVID %d\n", - vlan->vid_end); - - mutex_unlock(&ps->reg_lock); -} - -static int _mv88e6xxx_port_vlan_del(struct mv88e6xxx_priv_state *ps, - int port, u16 vid) -{ - struct dsa_switch *ds = ps->ds; - struct mv88e6xxx_vtu_stu_entry vlan; - int i, err; - - err = _mv88e6xxx_vtu_get(ps, vid, &vlan, false); - if (err) - return err; - - /* Tell switchdev if this VLAN is handled in software */ - if (vlan.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) - return -EOPNOTSUPP; - - vlan.data[port] = GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER; - - /* keep the VLAN unless all ports are excluded */ - vlan.valid = false; - for (i = 0; i < ps->info->num_ports; ++i) { - if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i)) - continue; - - if (vlan.data[i] != GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER) { - vlan.valid = true; - break; - } - } - - err = _mv88e6xxx_vtu_loadpurge(ps, &vlan); - if (err) - return err; - - return _mv88e6xxx_atu_remove(ps, vlan.fid, port, false); -} - -static int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_vlan *vlan) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - u16 pvid, vid; - int err = 0; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VTU)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - err = _mv88e6xxx_port_pvid_get(ps, port, &pvid); - if (err) - goto unlock; - - for (vid = vlan->vid_begin; vid <= vlan->vid_end; ++vid) { - err = _mv88e6xxx_port_vlan_del(ps, port, vid); - if (err) - goto unlock; - - if (vid == pvid) { - err = _mv88e6xxx_port_pvid_set(ps, port, 0); - if (err) - goto unlock; - } - } - -unlock: - mutex_unlock(&ps->reg_lock); - - return err; -} - -static int _mv88e6xxx_atu_mac_write(struct mv88e6xxx_priv_state *ps, - const unsigned char *addr) -{ - int i, ret; - - for (i = 0; i < 3; i++) { - ret = _mv88e6xxx_reg_write( - ps, REG_GLOBAL, GLOBAL_ATU_MAC_01 + i, - (addr[i * 2] << 8) | addr[i * 2 + 1]); - if (ret < 0) - return ret; - } - - return 0; -} - -static int _mv88e6xxx_atu_mac_read(struct mv88e6xxx_priv_state *ps, - unsigned char *addr) -{ - int i, ret; - - for (i = 0; i < 3; i++) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, - GLOBAL_ATU_MAC_01 + i); - if (ret < 0) - return ret; - addr[i * 2] = ret >> 8; - addr[i * 2 + 1] = ret & 0xff; - } - - return 0; -} - -static int _mv88e6xxx_atu_load(struct mv88e6xxx_priv_state *ps, - struct mv88e6xxx_atu_entry *entry) -{ - int ret; - - ret = _mv88e6xxx_atu_wait(ps); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_atu_mac_write(ps, entry->mac); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_atu_data_write(ps, entry); - if (ret < 0) - return ret; - - return _mv88e6xxx_atu_cmd(ps, entry->fid, GLOBAL_ATU_OP_LOAD_DB); -} - -static int _mv88e6xxx_port_fdb_load(struct mv88e6xxx_priv_state *ps, int port, - const unsigned char *addr, u16 vid, - u8 state) -{ - struct mv88e6xxx_atu_entry entry = { 0 }; - struct mv88e6xxx_vtu_stu_entry vlan; - int err; - - /* Null VLAN ID corresponds to the port private database */ - if (vid == 0) - err = _mv88e6xxx_port_fid_get(ps, port, &vlan.fid); - else - err = _mv88e6xxx_vtu_get(ps, vid, &vlan, false); - if (err) - return err; - - entry.fid = vlan.fid; - entry.state = state; - ether_addr_copy(entry.mac, addr); - if (state != GLOBAL_ATU_DATA_STATE_UNUSED) { - entry.trunk = false; - entry.portv_trunkid = BIT(port); - } - - return _mv88e6xxx_atu_load(ps, &entry); -} - -static int mv88e6xxx_port_fdb_prepare(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_fdb *fdb, - struct switchdev_trans *trans) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) - return -EOPNOTSUPP; - - /* We don't need any dynamic resource from the kernel (yet), - * so skip the prepare phase. - */ - return 0; -} - -static void mv88e6xxx_port_fdb_add(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_fdb *fdb, - struct switchdev_trans *trans) -{ - int state = is_multicast_ether_addr(fdb->addr) ? - GLOBAL_ATU_DATA_STATE_MC_STATIC : - GLOBAL_ATU_DATA_STATE_UC_STATIC; - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) - return; - - mutex_lock(&ps->reg_lock); - if (_mv88e6xxx_port_fdb_load(ps, port, fdb->addr, fdb->vid, state)) - netdev_err(ds->ports[port].netdev, - "failed to load MAC address\n"); - mutex_unlock(&ps->reg_lock); -} - -static int mv88e6xxx_port_fdb_del(struct dsa_switch *ds, int port, - const struct switchdev_obj_port_fdb *fdb) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_port_fdb_load(ps, port, fdb->addr, fdb->vid, - GLOBAL_ATU_DATA_STATE_UNUSED); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int _mv88e6xxx_atu_getnext(struct mv88e6xxx_priv_state *ps, u16 fid, - struct mv88e6xxx_atu_entry *entry) -{ - struct mv88e6xxx_atu_entry next = { 0 }; - int ret; - - next.fid = fid; - - ret = _mv88e6xxx_atu_wait(ps); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_atu_cmd(ps, fid, GLOBAL_ATU_OP_GET_NEXT_DB); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_atu_mac_read(ps, next.mac); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, GLOBAL_ATU_DATA); - if (ret < 0) - return ret; - - next.state = ret & GLOBAL_ATU_DATA_STATE_MASK; - if (next.state != GLOBAL_ATU_DATA_STATE_UNUSED) { - unsigned int mask, shift; - - if (ret & GLOBAL_ATU_DATA_TRUNK) { - next.trunk = true; - mask = GLOBAL_ATU_DATA_TRUNK_ID_MASK; - shift = GLOBAL_ATU_DATA_TRUNK_ID_SHIFT; - } else { - next.trunk = false; - mask = GLOBAL_ATU_DATA_PORT_VECTOR_MASK; - shift = GLOBAL_ATU_DATA_PORT_VECTOR_SHIFT; - } - - next.portv_trunkid = (ret & mask) >> shift; - } - - *entry = next; - return 0; -} - -static int _mv88e6xxx_port_fdb_dump_one(struct mv88e6xxx_priv_state *ps, - u16 fid, u16 vid, int port, - struct switchdev_obj_port_fdb *fdb, - int (*cb)(struct switchdev_obj *obj)) -{ - struct mv88e6xxx_atu_entry addr = { - .mac = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }, - }; - int err; - - err = _mv88e6xxx_atu_mac_write(ps, addr.mac); - if (err) - return err; - - do { - err = _mv88e6xxx_atu_getnext(ps, fid, &addr); - if (err) - break; - - if (addr.state == GLOBAL_ATU_DATA_STATE_UNUSED) - break; - - if (!addr.trunk && addr.portv_trunkid & BIT(port)) { - bool is_static = addr.state == - (is_multicast_ether_addr(addr.mac) ? - GLOBAL_ATU_DATA_STATE_MC_STATIC : - GLOBAL_ATU_DATA_STATE_UC_STATIC); - - fdb->vid = vid; - ether_addr_copy(fdb->addr, addr.mac); - fdb->ndm_state = is_static ? NUD_NOARP : NUD_REACHABLE; - - err = cb(&fdb->obj); - if (err) - break; - } - } while (!is_broadcast_ether_addr(addr.mac)); - - return err; -} - -static int mv88e6xxx_port_fdb_dump(struct dsa_switch *ds, int port, - struct switchdev_obj_port_fdb *fdb, - int (*cb)(struct switchdev_obj *obj)) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct mv88e6xxx_vtu_stu_entry vlan = { - .vid = GLOBAL_VTU_VID_MASK, /* all ones */ - }; - u16 fid; - int err; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_ATU)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - /* Dump port's default Filtering Information Database (VLAN ID 0) */ - err = _mv88e6xxx_port_fid_get(ps, port, &fid); - if (err) - goto unlock; - - err = _mv88e6xxx_port_fdb_dump_one(ps, fid, 0, port, fdb, cb); - if (err) - goto unlock; - - /* Dump VLANs' Filtering Information Databases */ - err = _mv88e6xxx_vtu_vid_write(ps, vlan.vid); - if (err) - goto unlock; - - do { - err = _mv88e6xxx_vtu_getnext(ps, &vlan); - if (err) - break; - - if (!vlan.valid) - break; - - err = _mv88e6xxx_port_fdb_dump_one(ps, vlan.fid, vlan.vid, port, - fdb, cb); - if (err) - break; - } while (vlan.vid < GLOBAL_VTU_VID_MASK); - -unlock: - mutex_unlock(&ps->reg_lock); - - return err; -} - -static int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port, - struct net_device *bridge) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int i, err = 0; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VLANTABLE)) - return -EOPNOTSUPP; - - mutex_lock(&ps->reg_lock); - - /* Assign the bridge and remap each port's VLANTable */ - ps->ports[port].bridge_dev = bridge; - - for (i = 0; i < ps->info->num_ports; ++i) { - if (ps->ports[i].bridge_dev == bridge) { - err = _mv88e6xxx_port_based_vlan_map(ps, i); - if (err) - break; - } - } - - mutex_unlock(&ps->reg_lock); - - return err; -} - -static void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - struct net_device *bridge = ps->ports[port].bridge_dev; - int i; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_VLANTABLE)) - return; - - mutex_lock(&ps->reg_lock); - - /* Unassign the bridge and remap each port's VLANTable */ - ps->ports[port].bridge_dev = NULL; - - for (i = 0; i < ps->info->num_ports; ++i) - if (i == port || ps->ports[i].bridge_dev == bridge) - if (_mv88e6xxx_port_based_vlan_map(ps, i)) - netdev_warn(ds->ports[i].netdev, - "failed to remap\n"); - - mutex_unlock(&ps->reg_lock); -} - -static int _mv88e6xxx_mdio_page_write(struct mv88e6xxx_priv_state *ps, - int port, int page, int reg, int val) -{ - int ret; - - ret = mv88e6xxx_mdio_write_indirect(ps, port, 0x16, page); - if (ret < 0) - goto restore_page_0; - - ret = mv88e6xxx_mdio_write_indirect(ps, port, reg, val); -restore_page_0: - mv88e6xxx_mdio_write_indirect(ps, port, 0x16, 0x0); - - return ret; -} - -static int _mv88e6xxx_mdio_page_read(struct mv88e6xxx_priv_state *ps, - int port, int page, int reg) -{ - int ret; - - ret = mv88e6xxx_mdio_write_indirect(ps, port, 0x16, page); - if (ret < 0) - goto restore_page_0; - - ret = mv88e6xxx_mdio_read_indirect(ps, port, reg); -restore_page_0: - mv88e6xxx_mdio_write_indirect(ps, port, 0x16, 0x0); - - return ret; -} - -static int mv88e6xxx_switch_reset(struct mv88e6xxx_priv_state *ps) -{ - bool ppu_active = mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU_ACTIVE); - u16 is_reset = (ppu_active ? 0x8800 : 0xc800); - struct gpio_desc *gpiod = ps->reset; - unsigned long timeout; - int ret; - int i; - - /* Set all ports to the disabled state. */ - for (i = 0; i < ps->info->num_ports; i++) { - ret = _mv88e6xxx_reg_read(ps, REG_PORT(i), PORT_CONTROL); - if (ret < 0) - return ret; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(i), PORT_CONTROL, - ret & 0xfffc); - if (ret) - return ret; - } - - /* Wait for transmit queues to drain. */ - usleep_range(2000, 4000); - - /* If there is a gpio connected to the reset pin, toggle it */ - if (gpiod) { - gpiod_set_value_cansleep(gpiod, 1); - usleep_range(10000, 20000); - gpiod_set_value_cansleep(gpiod, 0); - usleep_range(10000, 20000); - } - - /* Reset the switch. Keep the PPU active if requested. The PPU - * needs to be active to support indirect phy register access - * through global registers 0x18 and 0x19. - */ - if (ppu_active) - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, 0x04, 0xc000); - else - ret = _mv88e6xxx_reg_write(ps, REG_GLOBAL, 0x04, 0xc400); - if (ret) - return ret; - - /* Wait up to one second for reset to complete. */ - timeout = jiffies + 1 * HZ; - while (time_before(jiffies, timeout)) { - ret = _mv88e6xxx_reg_read(ps, REG_GLOBAL, 0x00); - if (ret < 0) - return ret; - - if ((ret & is_reset) == is_reset) - break; - usleep_range(1000, 2000); - } - if (time_after(jiffies, timeout)) - ret = -ETIMEDOUT; - else - ret = 0; - - return ret; -} - -static int mv88e6xxx_power_on_serdes(struct mv88e6xxx_priv_state *ps) -{ - int ret; - - ret = _mv88e6xxx_mdio_page_read(ps, REG_FIBER_SERDES, - PAGE_FIBER_SERDES, MII_BMCR); - if (ret < 0) - return ret; - - if (ret & BMCR_PDOWN) { - ret &= ~BMCR_PDOWN; - ret = _mv88e6xxx_mdio_page_write(ps, REG_FIBER_SERDES, - PAGE_FIBER_SERDES, MII_BMCR, - ret); - } - - return ret; -} - -static int mv88e6xxx_setup_port(struct mv88e6xxx_priv_state *ps, int port) -{ - struct dsa_switch *ds = ps->ds; - int ret; - u16 reg; - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6185_family(ps) || mv88e6xxx_6095_family(ps) || - mv88e6xxx_6065_family(ps) || mv88e6xxx_6320_family(ps)) { - /* MAC Forcing register: don't force link, speed, - * duplex or flow control state to any particular - * values on physical ports, but force the CPU port - * and all DSA ports to their maximum bandwidth and - * full duplex. - */ - reg = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_PCS_CTRL); - if (dsa_is_cpu_port(ds, port) || dsa_is_dsa_port(ds, port)) { - reg &= ~PORT_PCS_CTRL_UNFORCED; - reg |= PORT_PCS_CTRL_FORCE_LINK | - PORT_PCS_CTRL_LINK_UP | - PORT_PCS_CTRL_DUPLEX_FULL | - PORT_PCS_CTRL_FORCE_DUPLEX; - if (mv88e6xxx_6065_family(ps)) - reg |= PORT_PCS_CTRL_100; - else - reg |= PORT_PCS_CTRL_1000; - } else { - reg |= PORT_PCS_CTRL_UNFORCED; - } - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_PCS_CTRL, reg); - if (ret) - return ret; - } - - /* Port Control: disable Drop-on-Unlock, disable Drop-on-Lock, - * disable Header mode, enable IGMP/MLD snooping, disable VLAN - * tunneling, determine priority by looking at 802.1p and IP - * priority fields (IP prio has precedence), and set STP state - * to Forwarding. - * - * If this is the CPU link, use DSA or EDSA tagging depending - * on which tagging mode was configured. - * - * If this is a link to another switch, use DSA tagging mode. - * - * If this is the upstream port for this switch, enable - * forwarding of unknown unicasts and multicasts. - */ - reg = 0; - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6095_family(ps) || mv88e6xxx_6065_family(ps) || - mv88e6xxx_6185_family(ps) || mv88e6xxx_6320_family(ps)) - reg = PORT_CONTROL_IGMP_MLD_SNOOP | - PORT_CONTROL_USE_TAG | PORT_CONTROL_USE_IP | - PORT_CONTROL_STATE_FORWARDING; - if (dsa_is_cpu_port(ds, port)) { - if (mv88e6xxx_6095_family(ps) || mv88e6xxx_6185_family(ps)) - reg |= PORT_CONTROL_DSA_TAG; - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6320_family(ps)) { - reg |= PORT_CONTROL_FRAME_ETHER_TYPE_DSA | - PORT_CONTROL_FORWARD_UNKNOWN | - PORT_CONTROL_FORWARD_UNKNOWN_MC; - } - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6095_family(ps) || mv88e6xxx_6065_family(ps) || - mv88e6xxx_6185_family(ps) || mv88e6xxx_6320_family(ps)) { - reg |= PORT_CONTROL_EGRESS_ADD_TAG; - } - } - if (dsa_is_dsa_port(ds, port)) { - if (mv88e6xxx_6095_family(ps) || mv88e6xxx_6185_family(ps)) - reg |= PORT_CONTROL_DSA_TAG; - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6320_family(ps)) { - reg |= PORT_CONTROL_FRAME_MODE_DSA; - } - - if (port == dsa_upstream_port(ds)) - reg |= PORT_CONTROL_FORWARD_UNKNOWN | - PORT_CONTROL_FORWARD_UNKNOWN_MC; - } - if (reg) { - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_CONTROL, reg); - if (ret) - return ret; - } - - /* If this port is connected to a SerDes, make sure the SerDes is not - * powered down. - */ - if (mv88e6xxx_6352_family(ps)) { - ret = _mv88e6xxx_reg_read(ps, REG_PORT(port), PORT_STATUS); - if (ret < 0) - return ret; - ret &= PORT_STATUS_CMODE_MASK; - if ((ret == PORT_STATUS_CMODE_100BASE_X) || - (ret == PORT_STATUS_CMODE_1000BASE_X) || - (ret == PORT_STATUS_CMODE_SGMII)) { - ret = mv88e6xxx_power_on_serdes(ps); - if (ret < 0) - return ret; - } - } - - /* Port Control 2: don't force a good FCS, set the maximum frame size to - * 10240 bytes, disable 802.1q tags checking, don't discard tagged or - * untagged frames on this port, do a destination address lookup on all - * received packets as usual, disable ARP mirroring and don't send a - * copy of all transmitted/received frames on this port to the CPU. - */ - reg = 0; - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6095_family(ps) || mv88e6xxx_6320_family(ps) || - mv88e6xxx_6185_family(ps)) - reg = PORT_CONTROL_2_MAP_DA; - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6320_family(ps)) - reg |= PORT_CONTROL_2_JUMBO_10240; - - if (mv88e6xxx_6095_family(ps) || mv88e6xxx_6185_family(ps)) { - /* Set the upstream port this port should use */ - reg |= dsa_upstream_port(ds); - /* enable forwarding of unknown multicast addresses to - * the upstream port - */ - if (port == dsa_upstream_port(ds)) - reg |= PORT_CONTROL_2_FORWARD_UNKNOWN; - } - - reg |= PORT_CONTROL_2_8021Q_DISABLED; - - if (reg) { - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_CONTROL_2, reg); - if (ret) - return ret; - } - - /* Port Association Vector: when learning source addresses - * of packets, add the address to the address database using - * a port bitmap that has only the bit for this port set and - * the other bits clear. - */ - reg = 1 << port; - /* Disable learning for CPU port */ - if (dsa_is_cpu_port(ds, port)) - reg = 0; - - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_ASSOC_VECTOR, reg); - if (ret) - return ret; - - /* Egress rate control 2: disable egress rate control. */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_RATE_CONTROL_2, - 0x0000); - if (ret) - return ret; - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6320_family(ps)) { - /* Do not limit the period of time that this port can - * be paused for by the remote end or the period of - * time that this port can pause the remote end. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_PAUSE_CTRL, 0x0000); - if (ret) - return ret; - - /* Port ATU control: disable limiting the number of - * address database entries that this port is allowed - * to use. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_ATU_CONTROL, 0x0000); - /* Priority Override: disable DA, SA and VTU priority - * override. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_PRI_OVERRIDE, 0x0000); - if (ret) - return ret; - - /* Port Ethertype: use the Ethertype DSA Ethertype - * value. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_ETH_TYPE, ETH_P_EDSA); - if (ret) - return ret; - /* Tag Remap: use an identity 802.1p prio -> switch - * prio mapping. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_TAG_REGMAP_0123, 0x3210); - if (ret) - return ret; - - /* Tag Remap 2: use an identity 802.1p prio -> switch - * prio mapping. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_TAG_REGMAP_4567, 0x7654); - if (ret) - return ret; - } - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6185_family(ps) || mv88e6xxx_6095_family(ps) || - mv88e6xxx_6320_family(ps)) { - /* Rate Control: disable ingress rate limiting. */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), - PORT_RATE_CONTROL, 0x0001); - if (ret) - return ret; - } - - /* Port Control 1: disable trunking, disable sending - * learning messages to this port. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_CONTROL_1, 0x0000); - if (ret) - return ret; - - /* Port based VLAN map: give each port the same default address - * database, and allow bidirectional communication between the - * CPU and DSA port(s), and the other ports. - */ - ret = _mv88e6xxx_port_fid_set(ps, port, 0); - if (ret) - return ret; - - ret = _mv88e6xxx_port_based_vlan_map(ps, port); - if (ret) - return ret; - - /* Default VLAN ID and priority: don't set a default VLAN - * ID, and set the default packet priority to zero. - */ - ret = _mv88e6xxx_reg_write(ps, REG_PORT(port), PORT_DEFAULT_VLAN, - 0x0000); - if (ret) - return ret; - - return 0; -} - -static int mv88e6xxx_setup_global(struct mv88e6xxx_priv_state *ps) -{ - struct dsa_switch *ds = ps->ds; - u32 upstream_port = dsa_upstream_port(ds); - u16 reg; - int err; - int i; - - /* Enable the PHY Polling Unit if present, don't discard any packets, - * and mask all interrupt sources. - */ - reg = 0; - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU) || - mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU_ACTIVE)) - reg |= GLOBAL_CONTROL_PPU_ENABLE; - - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL, reg); - if (err) - return err; - - /* Configure the upstream port, and configure it as the port to which - * ingress and egress and ARP monitor frames are to be sent. - */ - reg = upstream_port << GLOBAL_MONITOR_CONTROL_INGRESS_SHIFT | - upstream_port << GLOBAL_MONITOR_CONTROL_EGRESS_SHIFT | - upstream_port << GLOBAL_MONITOR_CONTROL_ARP_SHIFT; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_MONITOR_CONTROL, reg); - if (err) - return err; - - /* Disable remote management, and set the switch's DSA device number. */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_CONTROL_2, - GLOBAL_CONTROL_2_MULTIPLE_CASCADE | - (ds->index & 0x1f)); - if (err) - return err; - - /* Set the default address aging time to 5 minutes, and - * enable address learn messages to be sent to all message - * ports. - */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_ATU_CONTROL, - 0x0140 | GLOBAL_ATU_CONTROL_LEARN2ALL); - if (err) - return err; - - /* Configure the IP ToS mapping registers. */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_0, 0x0000); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_1, 0x0000); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_2, 0x5555); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_3, 0x5555); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_4, 0xaaaa); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_5, 0xaaaa); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_6, 0xffff); - if (err) - return err; - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IP_PRI_7, 0xffff); - if (err) - return err; - - /* Configure the IEEE 802.1p priority mapping register. */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_IEEE_PRI, 0xfa41); - if (err) - return err; - - /* Send all frames with destination addresses matching - * 01:80:c2:00:00:0x to the CPU port. - */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_MGMT_EN_0X, 0xffff); - if (err) - return err; - - /* Ignore removed tag data on doubly tagged packets, disable - * flow control messages, force flow control priority to the - * highest, and send all special multicast frames to the CPU - * port at the highest priority. - */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_SWITCH_MGMT, - 0x7 | GLOBAL2_SWITCH_MGMT_RSVD2CPU | 0x70 | - GLOBAL2_SWITCH_MGMT_FORCE_FLOW_CTRL_PRI); - if (err) - return err; - - /* Program the DSA routing table. */ - for (i = 0; i < 32; i++) { - int nexthop = 0x1f; - - if (i != ds->index && i < DSA_MAX_SWITCHES) - nexthop = ds->rtable[i] & 0x1f; - - err = _mv88e6xxx_reg_write( - ps, REG_GLOBAL2, - GLOBAL2_DEVICE_MAPPING, - GLOBAL2_DEVICE_MAPPING_UPDATE | - (i << GLOBAL2_DEVICE_MAPPING_TARGET_SHIFT) | nexthop); - if (err) - return err; - } - - /* Clear all trunk masks. */ - for (i = 0; i < 8; i++) { - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, GLOBAL2_TRUNK_MASK, - 0x8000 | - (i << GLOBAL2_TRUNK_MASK_NUM_SHIFT) | - ((1 << ps->info->num_ports) - 1)); - if (err) - return err; - } - - /* Clear all trunk mappings. */ - for (i = 0; i < 16; i++) { - err = _mv88e6xxx_reg_write( - ps, REG_GLOBAL2, - GLOBAL2_TRUNK_MAPPING, - GLOBAL2_TRUNK_MAPPING_UPDATE | - (i << GLOBAL2_TRUNK_MAPPING_ID_SHIFT)); - if (err) - return err; - } - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6320_family(ps)) { - /* Send all frames with destination addresses matching - * 01:80:c2:00:00:2x to the CPU port. - */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, - GLOBAL2_MGMT_EN_2X, 0xffff); - if (err) - return err; - - /* Initialise cross-chip port VLAN table to reset - * defaults. - */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, - GLOBAL2_PVT_ADDR, 0x9000); - if (err) - return err; - - /* Clear the priority override table. */ - for (i = 0; i < 16; i++) { - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, - GLOBAL2_PRIO_OVERRIDE, - 0x8000 | (i << 8)); - if (err) - return err; - } - } - - if (mv88e6xxx_6352_family(ps) || mv88e6xxx_6351_family(ps) || - mv88e6xxx_6165_family(ps) || mv88e6xxx_6097_family(ps) || - mv88e6xxx_6185_family(ps) || mv88e6xxx_6095_family(ps) || - mv88e6xxx_6320_family(ps)) { - /* Disable ingress rate limiting by resetting all - * ingress rate limit registers to their initial - * state. - */ - for (i = 0; i < ps->info->num_ports; i++) { - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL2, - GLOBAL2_INGRESS_OP, - 0x9000 | (i << 8)); - if (err) - return err; - } - } - - /* Clear the statistics counters for all ports */ - err = _mv88e6xxx_reg_write(ps, REG_GLOBAL, GLOBAL_STATS_OP, - GLOBAL_STATS_OP_FLUSH_ALL); - if (err) - return err; - - /* Wait for the flush to complete. */ - err = _mv88e6xxx_stats_wait(ps); - if (err) - return err; - - /* Clear all ATU entries */ - err = _mv88e6xxx_atu_flush(ps, 0, true); - if (err) - return err; - - /* Clear all the VTU and STU entries */ - err = _mv88e6xxx_vtu_stu_flush(ps); - if (err < 0) - return err; - - return err; -} - -static int mv88e6xxx_setup(struct dsa_switch *ds) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int err; - int i; - - ps->ds = ds; - ds->slave_mii_bus = ps->mdio_bus; - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM)) - mutex_init(&ps->eeprom_mutex); - - mutex_lock(&ps->reg_lock); - - err = mv88e6xxx_switch_reset(ps); - if (err) - goto unlock; - - err = mv88e6xxx_setup_global(ps); - if (err) - goto unlock; - - for (i = 0; i < ps->info->num_ports; i++) { - err = mv88e6xxx_setup_port(ps, i); - if (err) - goto unlock; - } - -unlock: - mutex_unlock(&ps->reg_lock); - - return err; -} - -static int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, - int reg) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_mdio_page_read(ps, port, page, reg); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int mv88e6xxx_mdio_page_write(struct dsa_switch *ds, int port, int page, - int reg, int val) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - - mutex_lock(&ps->reg_lock); - ret = _mv88e6xxx_mdio_page_write(ps, port, page, reg, val); - mutex_unlock(&ps->reg_lock); - - return ret; -} - -static int mv88e6xxx_port_to_mdio_addr(struct mv88e6xxx_priv_state *ps, - int port) -{ - if (port >= 0 && port < ps->info->num_ports) - return port; - return -EINVAL; -} - -static int mv88e6xxx_mdio_read(struct mii_bus *bus, int port, int regnum) -{ - struct mv88e6xxx_priv_state *ps = bus->priv; - int addr = mv88e6xxx_port_to_mdio_addr(ps, port); - int ret; - - if (addr < 0) - return 0xffff; - - mutex_lock(&ps->reg_lock); - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) - ret = mv88e6xxx_mdio_read_ppu(ps, addr, regnum); - else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SMI_PHY)) - ret = mv88e6xxx_mdio_read_indirect(ps, addr, regnum); - else - ret = mv88e6xxx_mdio_read_direct(ps, addr, regnum); - - mutex_unlock(&ps->reg_lock); - return ret; -} - -static int mv88e6xxx_mdio_write(struct mii_bus *bus, int port, int regnum, - u16 val) -{ - struct mv88e6xxx_priv_state *ps = bus->priv; - int addr = mv88e6xxx_port_to_mdio_addr(ps, port); - int ret; - - if (addr < 0) - return 0xffff; - - mutex_lock(&ps->reg_lock); - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) - ret = mv88e6xxx_mdio_write_ppu(ps, addr, regnum, val); - else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_SMI_PHY)) - ret = mv88e6xxx_mdio_write_indirect(ps, addr, regnum, val); - else - ret = mv88e6xxx_mdio_write_direct(ps, addr, regnum, val); - - mutex_unlock(&ps->reg_lock); - return ret; -} - -static int mv88e6xxx_mdio_register(struct mv88e6xxx_priv_state *ps, - struct device_node *np) -{ - static int index; - struct mii_bus *bus; - int err; - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_PPU)) - mv88e6xxx_ppu_state_init(ps); - - if (np) - ps->mdio_np = of_get_child_by_name(np, "mdio"); - - bus = devm_mdiobus_alloc(ps->dev); - if (!bus) - return -ENOMEM; - - bus->priv = (void *)ps; - if (np) { - bus->name = np->full_name; - snprintf(bus->id, MII_BUS_ID_SIZE, "%s", np->full_name); - } else { - bus->name = "mv88e6xxx SMI"; - snprintf(bus->id, MII_BUS_ID_SIZE, "mv88e6xxx-%d", index++); - } - - bus->read = mv88e6xxx_mdio_read; - bus->write = mv88e6xxx_mdio_write; - bus->parent = ps->dev; - - if (ps->mdio_np) - err = of_mdiobus_register(bus, ps->mdio_np); - else - err = mdiobus_register(bus); - if (err) { - dev_err(ps->dev, "Cannot register MDIO bus (%d)\n", err); - goto out; - } - ps->mdio_bus = bus; - - return 0; - -out: - if (ps->mdio_np) - of_node_put(ps->mdio_np); - - return err; -} - -static void mv88e6xxx_mdio_unregister(struct mv88e6xxx_priv_state *ps) - -{ - struct mii_bus *bus = ps->mdio_bus; - - mdiobus_unregister(bus); - - if (ps->mdio_np) - of_node_put(ps->mdio_np); -} - -#ifdef CONFIG_NET_DSA_HWMON - -static int mv88e61xx_get_temp(struct dsa_switch *ds, int *temp) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int ret; - int val; - - *temp = 0; - - mutex_lock(&ps->reg_lock); - - ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x16, 0x6); - if (ret < 0) - goto error; - - /* Enable temperature sensor */ - ret = mv88e6xxx_mdio_read_direct(ps, 0x0, 0x1a); - if (ret < 0) - goto error; - - ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x1a, ret | (1 << 5)); - if (ret < 0) - goto error; - - /* Wait for temperature to stabilize */ - usleep_range(10000, 12000); - - val = mv88e6xxx_mdio_read_direct(ps, 0x0, 0x1a); - if (val < 0) { - ret = val; - goto error; - } - - /* Disable temperature sensor */ - ret = mv88e6xxx_mdio_write_direct(ps, 0x0, 0x1a, ret & ~(1 << 5)); - if (ret < 0) - goto error; - - *temp = ((val & 0x1f) - 5) * 5; - -error: - mv88e6xxx_mdio_write_direct(ps, 0x0, 0x16, 0x0); - mutex_unlock(&ps->reg_lock); - return ret; -} - -static int mv88e63xx_get_temp(struct dsa_switch *ds, int *temp) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; - int ret; - - *temp = 0; - - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 27); - if (ret < 0) - return ret; - - *temp = (ret & 0xff) - 25; - - return 0; -} - -static int mv88e6xxx_get_temp(struct dsa_switch *ds, int *temp) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP)) - return -EOPNOTSUPP; - - if (mv88e6xxx_6320_family(ps) || mv88e6xxx_6352_family(ps)) - return mv88e63xx_get_temp(ds, temp); - - return mv88e61xx_get_temp(ds, temp); -} - -static int mv88e6xxx_get_temp_limit(struct dsa_switch *ds, int *temp) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP_LIMIT)) - return -EOPNOTSUPP; - - *temp = 0; - - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); - if (ret < 0) - return ret; - - *temp = (((ret >> 8) & 0x1f) * 5) - 25; - - return 0; -} - -static int mv88e6xxx_set_temp_limit(struct dsa_switch *ds, int temp) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP_LIMIT)) - return -EOPNOTSUPP; - - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); - if (ret < 0) - return ret; - temp = clamp_val(DIV_ROUND_CLOSEST(temp, 5) + 5, 0, 0x1f); - return mv88e6xxx_mdio_page_write(ds, phy, 6, 26, - (ret & 0xe0ff) | (temp << 8)); -} - -static int mv88e6xxx_get_temp_alarm(struct dsa_switch *ds, bool *alarm) -{ - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - int phy = mv88e6xxx_6320_family(ps) ? 3 : 0; - int ret; - - if (!mv88e6xxx_has(ps, MV88E6XXX_FLAG_TEMP_LIMIT)) - return -EOPNOTSUPP; - - *alarm = false; - - ret = mv88e6xxx_mdio_page_read(ds, phy, 6, 26); - if (ret < 0) - return ret; - - *alarm = !!(ret & 0x40); - - return 0; -} -#endif /* CONFIG_NET_DSA_HWMON */ - -static const struct mv88e6xxx_info mv88e6xxx_table[] = { - [MV88E6085] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6085, - .family = MV88E6XXX_FAMILY_6097, - .name = "Marvell 88E6085", - .num_databases = 4096, - .num_ports = 10, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6097, - }, - - [MV88E6095] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6095, - .family = MV88E6XXX_FAMILY_6095, - .name = "Marvell 88E6095/88E6095F", - .num_databases = 256, - .num_ports = 11, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6095, - }, - - [MV88E6123] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6123, - .family = MV88E6XXX_FAMILY_6165, - .name = "Marvell 88E6123", - .num_databases = 4096, - .num_ports = 3, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6165, - }, - - [MV88E6131] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6131, - .family = MV88E6XXX_FAMILY_6185, - .name = "Marvell 88E6131", - .num_databases = 256, - .num_ports = 8, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6185, - }, - - [MV88E6161] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6161, - .family = MV88E6XXX_FAMILY_6165, - .name = "Marvell 88E6161", - .num_databases = 4096, - .num_ports = 6, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6165, - }, - - [MV88E6165] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6165, - .family = MV88E6XXX_FAMILY_6165, - .name = "Marvell 88E6165", - .num_databases = 4096, - .num_ports = 6, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6165, - }, - - [MV88E6171] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6171, - .family = MV88E6XXX_FAMILY_6351, - .name = "Marvell 88E6171", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6351, - }, - - [MV88E6172] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6172, - .family = MV88E6XXX_FAMILY_6352, - .name = "Marvell 88E6172", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6352, - }, - - [MV88E6175] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6175, - .family = MV88E6XXX_FAMILY_6351, - .name = "Marvell 88E6175", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6351, - }, - - [MV88E6176] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6176, - .family = MV88E6XXX_FAMILY_6352, - .name = "Marvell 88E6176", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6352, - }, - - [MV88E6185] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6185, - .family = MV88E6XXX_FAMILY_6185, - .name = "Marvell 88E6185", - .num_databases = 256, - .num_ports = 10, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6185, - }, - - [MV88E6240] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6240, - .family = MV88E6XXX_FAMILY_6352, - .name = "Marvell 88E6240", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6352, - }, - - [MV88E6320] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6320, - .family = MV88E6XXX_FAMILY_6320, - .name = "Marvell 88E6320", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6320, - }, - - [MV88E6321] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6321, - .family = MV88E6XXX_FAMILY_6320, - .name = "Marvell 88E6321", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6320, - }, - - [MV88E6350] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6350, - .family = MV88E6XXX_FAMILY_6351, - .name = "Marvell 88E6350", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6351, - }, - - [MV88E6351] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6351, - .family = MV88E6XXX_FAMILY_6351, - .name = "Marvell 88E6351", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6351, - }, - - [MV88E6352] = { - .prod_num = PORT_SWITCH_ID_PROD_NUM_6352, - .family = MV88E6XXX_FAMILY_6352, - .name = "Marvell 88E6352", - .num_databases = 4096, - .num_ports = 7, - .port_base_addr = 0x10, - .flags = MV88E6XXX_FLAGS_FAMILY_6352, - }, -}; - -static const struct mv88e6xxx_info *mv88e6xxx_lookup_info(unsigned int prod_num) -{ - int i; - - for (i = 0; i < ARRAY_SIZE(mv88e6xxx_table); ++i) - if (mv88e6xxx_table[i].prod_num == prod_num) - return &mv88e6xxx_table[i]; - - return NULL; -} - -static int mv88e6xxx_detect(struct mv88e6xxx_priv_state *ps) -{ - const struct mv88e6xxx_info *info; - int id, prod_num, rev; - - id = mv88e6xxx_reg_read(ps, ps->info->port_base_addr, PORT_SWITCH_ID); - if (id < 0) - return id; - - prod_num = (id & 0xfff0) >> 4; - rev = id & 0x000f; - - info = mv88e6xxx_lookup_info(prod_num); - if (!info) - return -ENODEV; - - /* Update the compatible info with the probed one */ - ps->info = info; - - dev_info(ps->dev, "switch 0x%x detected: %s, revision %u\n", - ps->info->prod_num, ps->info->name, rev); - - return 0; -} - -static struct mv88e6xxx_priv_state *mv88e6xxx_alloc_chip(struct device *dev) -{ - struct mv88e6xxx_priv_state *ps; - - ps = devm_kzalloc(dev, sizeof(*ps), GFP_KERNEL); - if (!ps) - return NULL; - - ps->dev = dev; - - mutex_init(&ps->reg_lock); - - return ps; -} - -static int mv88e6xxx_smi_init(struct mv88e6xxx_priv_state *ps, - struct mii_bus *bus, int sw_addr) -{ - /* ADDR[0] pin is unavailable externally and considered zero */ - if (sw_addr & 0x1) - return -EINVAL; - - if (sw_addr == 0) - ps->smi_ops = &mv88e6xxx_smi_single_chip_ops; - else if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_MULTI_CHIP)) - ps->smi_ops = &mv88e6xxx_smi_multi_chip_ops; - else - return -EINVAL; - - ps->bus = bus; - ps->sw_addr = sw_addr; - - return 0; -} - -static const char *mv88e6xxx_drv_probe(struct device *dsa_dev, - struct device *host_dev, int sw_addr, - void **priv) -{ - struct mv88e6xxx_priv_state *ps; - struct mii_bus *bus; - int err; - - bus = dsa_host_dev_to_mii_bus(host_dev); - if (!bus) - return NULL; - - ps = mv88e6xxx_alloc_chip(dsa_dev); - if (!ps) - return NULL; - - /* Legacy SMI probing will only support chips similar to 88E6085 */ - ps->info = &mv88e6xxx_table[MV88E6085]; - - err = mv88e6xxx_smi_init(ps, bus, sw_addr); - if (err) - goto free; - - err = mv88e6xxx_detect(ps); - if (err) - goto free; - - err = mv88e6xxx_mdio_register(ps, NULL); - if (err) - goto free; - - *priv = ps; - - return ps->info->name; -free: - devm_kfree(dsa_dev, ps); - - return NULL; -} - -static struct dsa_switch_driver mv88e6xxx_switch_driver = { - .tag_protocol = DSA_TAG_PROTO_EDSA, - .probe = mv88e6xxx_drv_probe, - .setup = mv88e6xxx_setup, - .set_addr = mv88e6xxx_set_addr, - .adjust_link = mv88e6xxx_adjust_link, - .get_strings = mv88e6xxx_get_strings, - .get_ethtool_stats = mv88e6xxx_get_ethtool_stats, - .get_sset_count = mv88e6xxx_get_sset_count, - .set_eee = mv88e6xxx_set_eee, - .get_eee = mv88e6xxx_get_eee, -#ifdef CONFIG_NET_DSA_HWMON - .get_temp = mv88e6xxx_get_temp, - .get_temp_limit = mv88e6xxx_get_temp_limit, - .set_temp_limit = mv88e6xxx_set_temp_limit, - .get_temp_alarm = mv88e6xxx_get_temp_alarm, -#endif - .get_eeprom_len = mv88e6xxx_get_eeprom_len, - .get_eeprom = mv88e6xxx_get_eeprom, - .set_eeprom = mv88e6xxx_set_eeprom, - .get_regs_len = mv88e6xxx_get_regs_len, - .get_regs = mv88e6xxx_get_regs, - .port_bridge_join = mv88e6xxx_port_bridge_join, - .port_bridge_leave = mv88e6xxx_port_bridge_leave, - .port_stp_state_set = mv88e6xxx_port_stp_state_set, - .port_vlan_filtering = mv88e6xxx_port_vlan_filtering, - .port_vlan_prepare = mv88e6xxx_port_vlan_prepare, - .port_vlan_add = mv88e6xxx_port_vlan_add, - .port_vlan_del = mv88e6xxx_port_vlan_del, - .port_vlan_dump = mv88e6xxx_port_vlan_dump, - .port_fdb_prepare = mv88e6xxx_port_fdb_prepare, - .port_fdb_add = mv88e6xxx_port_fdb_add, - .port_fdb_del = mv88e6xxx_port_fdb_del, - .port_fdb_dump = mv88e6xxx_port_fdb_dump, -}; - -static int mv88e6xxx_register_switch(struct mv88e6xxx_priv_state *ps, - struct device_node *np) -{ - struct device *dev = ps->dev; - struct dsa_switch *ds; - - ds = devm_kzalloc(dev, sizeof(*ds), GFP_KERNEL); - if (!ds) - return -ENOMEM; - - ds->dev = dev; - ds->priv = ps; - ds->drv = &mv88e6xxx_switch_driver; - - dev_set_drvdata(dev, ds); - - return dsa_register_switch(ds, np); -} - -static void mv88e6xxx_unregister_switch(struct mv88e6xxx_priv_state *ps) -{ - dsa_unregister_switch(ps->ds); -} - -static int mv88e6xxx_probe(struct mdio_device *mdiodev) -{ - struct device *dev = &mdiodev->dev; - struct device_node *np = dev->of_node; - const struct mv88e6xxx_info *compat_info; - struct mv88e6xxx_priv_state *ps; - u32 eeprom_len; - int err; - - compat_info = of_device_get_match_data(dev); - if (!compat_info) - return -EINVAL; - - ps = mv88e6xxx_alloc_chip(dev); - if (!ps) - return -ENOMEM; - - ps->info = compat_info; - - err = mv88e6xxx_smi_init(ps, mdiodev->bus, mdiodev->addr); - if (err) - return err; - - err = mv88e6xxx_detect(ps); - if (err) - return err; - - ps->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_ASIS); - if (IS_ERR(ps->reset)) - return PTR_ERR(ps->reset); - - if (mv88e6xxx_has(ps, MV88E6XXX_FLAG_EEPROM) && - !of_property_read_u32(np, "eeprom-length", &eeprom_len)) - ps->eeprom_len = eeprom_len; - - err = mv88e6xxx_mdio_register(ps, np); - if (err) - return err; - - err = mv88e6xxx_register_switch(ps, np); - if (err) { - mv88e6xxx_mdio_unregister(ps); - return err; - } - - return 0; -} - -static void mv88e6xxx_remove(struct mdio_device *mdiodev) -{ - struct dsa_switch *ds = dev_get_drvdata(&mdiodev->dev); - struct mv88e6xxx_priv_state *ps = ds_to_priv(ds); - - mv88e6xxx_unregister_switch(ps); - mv88e6xxx_mdio_unregister(ps); -} - -static const struct of_device_id mv88e6xxx_of_match[] = { - { - .compatible = "marvell,mv88e6085", - .data = &mv88e6xxx_table[MV88E6085], - }, - { /* sentinel */ }, -}; - -MODULE_DEVICE_TABLE(of, mv88e6xxx_of_match); - -static struct mdio_driver mv88e6xxx_driver = { - .probe = mv88e6xxx_probe, - .remove = mv88e6xxx_remove, - .mdiodrv.driver = { - .name = "mv88e6085", - .of_match_table = mv88e6xxx_of_match, - }, -}; - -static int __init mv88e6xxx_init(void) -{ - register_switch_driver(&mv88e6xxx_switch_driver); - return mdio_driver_register(&mv88e6xxx_driver); -} -module_init(mv88e6xxx_init); - -static void __exit mv88e6xxx_cleanup(void) -{ - mdio_driver_unregister(&mv88e6xxx_driver); - unregister_switch_driver(&mv88e6xxx_switch_driver); -} -module_exit(mv88e6xxx_cleanup); - -MODULE_AUTHOR("Lennert Buytenhek "); -MODULE_DESCRIPTION("Driver for Marvell 88E6XXX ethernet switch chips"); -MODULE_LICENSE("GPL"); diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 856c6e5..83f0662 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -563,7 +563,7 @@ struct mv88e6xxx_priv_port { struct net_device *bridge_dev; }; -struct mv88e6xxx_priv_state { +struct mv88e6xxx_chip { const struct mv88e6xxx_info *info; /* The dsa_switch this private structure is related to */ @@ -625,10 +625,8 @@ struct mv88e6xxx_priv_state { }; struct mv88e6xxx_ops { - int (*read)(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 *val); - int (*write)(struct mv88e6xxx_priv_state *ps, - int addr, int reg, u16 val); + int (*read)(struct mv88e6xxx_chip *chip, int addr, int reg, u16 *val); + int (*write)(struct mv88e6xxx_chip *chip, int addr, int reg, u16 val); }; enum stat_type { @@ -644,10 +642,10 @@ struct mv88e6xxx_hw_stat { enum stat_type type; }; -static inline bool mv88e6xxx_has(struct mv88e6xxx_priv_state *ps, +static inline bool mv88e6xxx_has(struct mv88e6xxx_chip *chip, unsigned long flags) { - return (ps->info->flags & flags) == flags; + return (chip->info->flags & flags) == flags; } #endif -- cgit v0.10.2 From 1c1779fa54b2a9d4e1de990095d790d64b9e00a1 Mon Sep 17 00:00:00 2001 From: Jarno Rajahalme Date: Tue, 21 Jun 2016 14:59:37 -0700 Subject: openvswitch: Set mark and labels before confirming. Set conntrack mark and labels right before committing so that the initial conntrack NEW event has the mark and labels. Signed-off-by: Jarno Rajahalme Acked-by: Joe Stringer Signed-off-by: David S. Miller diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index 3d5feed..23fd4fb 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -824,23 +824,6 @@ static int ovs_ct_lookup(struct net *net, struct sw_flow_key *key, return 0; } -/* Lookup connection and confirm if unconfirmed. */ -static int ovs_ct_commit(struct net *net, struct sw_flow_key *key, - const struct ovs_conntrack_info *info, - struct sk_buff *skb) -{ - int err; - - err = __ovs_ct_lookup(net, key, info, skb); - if (err) - return err; - /* This is a no-op if the connection has already been confirmed. */ - if (nf_conntrack_confirm(skb) != NF_ACCEPT) - return -EINVAL; - - return 0; -} - static bool labels_nonzero(const struct ovs_key_ct_labels *labels) { size_t i; @@ -873,21 +856,33 @@ int ovs_ct_execute(struct net *net, struct sk_buff *skb, } if (info->commit) - err = ovs_ct_commit(net, key, info, skb); + err = __ovs_ct_lookup(net, key, info, skb); else err = ovs_ct_lookup(net, key, info, skb); if (err) goto err; + /* Apply changes before confirming the connection so that the initial + * conntrack NEW netlink event carries the values given in the CT + * action. + */ if (info->mark.mask) { err = ovs_ct_set_mark(skb, key, info->mark.value, info->mark.mask); if (err) goto err; } - if (labels_nonzero(&info->labels.mask)) + if (labels_nonzero(&info->labels.mask)) { err = ovs_ct_set_labels(skb, key, &info->labels.value, &info->labels.mask); + if (err) + goto err; + } + /* This will take care of sending queued events even if the connection + * is already confirmed. + */ + if (info->commit && nf_conntrack_confirm(skb) != NF_ACCEPT) + err = -EINVAL; err: skb_push(skb, nh_ofs); if (err) -- cgit v0.10.2 From 7d904c7bcd51f72579c0c3134a50896c5a3efb9f Mon Sep 17 00:00:00 2001 From: Jarno Rajahalme Date: Tue, 21 Jun 2016 14:59:38 -0700 Subject: openvswitch: Only set mark and labels with a commit flag. Only set conntrack mark or labels when the commit flag is specified. This makes sure we can not set them before the connection has been persisted, as in that case the mark and labels would be lost in an event of an userspace upcall. OVS userspace already requires the commit flag to accept setting ct_mark and/or ct_labels. Validate for this in the kernel API. Signed-off-by: Jarno Rajahalme Signed-off-by: David S. Miller diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index 23fd4fb..52f3b9b 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -835,6 +835,42 @@ static bool labels_nonzero(const struct ovs_key_ct_labels *labels) return false; } +/* Lookup connection and confirm if unconfirmed. */ +static int ovs_ct_commit(struct net *net, struct sw_flow_key *key, + const struct ovs_conntrack_info *info, + struct sk_buff *skb) +{ + int err; + + err = __ovs_ct_lookup(net, key, info, skb); + if (err) + return err; + + /* Apply changes before confirming the connection so that the initial + * conntrack NEW netlink event carries the values given in the CT + * action. + */ + if (info->mark.mask) { + err = ovs_ct_set_mark(skb, key, info->mark.value, + info->mark.mask); + if (err) + return err; + } + if (labels_nonzero(&info->labels.mask)) { + err = ovs_ct_set_labels(skb, key, &info->labels.value, + &info->labels.mask); + if (err) + return err; + } + /* This will take care of sending queued events even if the connection + * is already confirmed. + */ + if (nf_conntrack_confirm(skb) != NF_ACCEPT) + return -EINVAL; + + return 0; +} + /* Returns 0 on success, -EINPROGRESS if 'skb' is stolen, or other nonzero * value if 'skb' is freed. */ @@ -856,34 +892,10 @@ int ovs_ct_execute(struct net *net, struct sk_buff *skb, } if (info->commit) - err = __ovs_ct_lookup(net, key, info, skb); + err = ovs_ct_commit(net, key, info, skb); else err = ovs_ct_lookup(net, key, info, skb); - if (err) - goto err; - /* Apply changes before confirming the connection so that the initial - * conntrack NEW netlink event carries the values given in the CT - * action. - */ - if (info->mark.mask) { - err = ovs_ct_set_mark(skb, key, info->mark.value, - info->mark.mask); - if (err) - goto err; - } - if (labels_nonzero(&info->labels.mask)) { - err = ovs_ct_set_labels(skb, key, &info->labels.value, - &info->labels.mask); - if (err) - goto err; - } - /* This will take care of sending queued events even if the connection - * is already confirmed. - */ - if (info->commit && nf_conntrack_confirm(skb) != NF_ACCEPT) - err = -EINVAL; -err: skb_push(skb, nh_ofs); if (err) kfree_skb(skb); @@ -1140,6 +1152,20 @@ static int parse_ct(const struct nlattr *attr, struct ovs_conntrack_info *info, } } +#ifdef CONFIG_NF_CONNTRACK_MARK + if (!info->commit && info->mark.mask) { + OVS_NLERR(log, + "Setting conntrack mark requires 'commit' flag."); + return -EINVAL; + } +#endif +#ifdef CONFIG_NF_CONNTRACK_LABELS + if (!info->commit && labels_nonzero(&info->labels.mask)) { + OVS_NLERR(log, + "Setting conntrack labels requires 'commit' flag."); + return -EINVAL; + } +#endif if (rem > 0) { OVS_NLERR(log, "Conntrack attr has %d unknown bytes", rem); return -EINVAL; -- cgit v0.10.2 From 0a91281e5190aa6bc42fa86ba88758add20087fa Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Wed, 22 Jun 2016 00:32:35 +0200 Subject: net: ethernet: macb: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index cb07d95..090463f 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -304,7 +304,7 @@ static void macb_set_tx_clk(struct clk *clk, int speed, struct net_device *dev) static void macb_handle_link_change(struct net_device *dev) { struct macb *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; unsigned long flags; int status_change = 0; @@ -414,7 +414,6 @@ static int macb_mii_probe(struct net_device *dev) bp->link = 0; bp->speed = 0; bp->duplex = -1; - bp->phy_dev = phydev; return 0; } @@ -1886,7 +1885,7 @@ static int macb_open(struct net_device *dev) netif_carrier_off(dev); /* if the phy is not yet register, retry later*/ - if (!bp->phy_dev) + if (!dev->phydev) return -EAGAIN; /* RX buffers initialization */ @@ -1905,7 +1904,7 @@ static int macb_open(struct net_device *dev) macb_init_hw(bp); /* schedule a link state check */ - phy_start(bp->phy_dev); + phy_start(dev->phydev); netif_tx_start_all_queues(dev); @@ -1920,8 +1919,8 @@ static int macb_close(struct net_device *dev) netif_tx_stop_all_queues(dev); napi_disable(&bp->napi); - if (bp->phy_dev) - phy_stop(bp->phy_dev); + if (dev->phydev) + phy_stop(dev->phydev); spin_lock_irqsave(&bp->lock, flags); macb_reset_hw(bp); @@ -2095,7 +2094,7 @@ static struct net_device_stats *macb_get_stats(struct net_device *dev) static int macb_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { struct macb *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; if (!phydev) return -ENODEV; @@ -2106,7 +2105,7 @@ static int macb_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) static int macb_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { struct macb *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; if (!phydev) return -ENODEV; @@ -2210,8 +2209,7 @@ static const struct ethtool_ops gem_ethtool_ops = { static int macb_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct macb *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; if (!netif_running(dev)) return -EINVAL; @@ -2570,7 +2568,7 @@ static int at91ether_open(struct net_device *dev) MACB_BIT(HRESP)); /* schedule a link state check */ - phy_start(lp->phy_dev); + phy_start(dev->phydev); netif_start_queue(dev); @@ -3010,7 +3008,7 @@ static int macb_probe(struct platform_device *pdev) if (err) goto err_out_free_netdev; - phydev = bp->phy_dev; + phydev = dev->phydev; netif_carrier_off(dev); @@ -3029,7 +3027,7 @@ static int macb_probe(struct platform_device *pdev) return 0; err_out_unregister_mdio: - phy_disconnect(bp->phy_dev); + phy_disconnect(dev->phydev); mdiobus_unregister(bp->mii_bus); mdiobus_free(bp->mii_bus); @@ -3057,8 +3055,8 @@ static int macb_remove(struct platform_device *pdev) if (dev) { bp = netdev_priv(dev); - if (bp->phy_dev) - phy_disconnect(bp->phy_dev); + if (dev->phydev) + phy_disconnect(dev->phydev); mdiobus_unregister(bp->mii_bus); mdiobus_free(bp->mii_bus); diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h index 8a13824..36893d8 100644 --- a/drivers/net/ethernet/cadence/macb.h +++ b/drivers/net/ethernet/cadence/macb.h @@ -823,7 +823,6 @@ struct macb { struct macb_or_gem_ops macbgem_ops; struct mii_bus *mii_bus; - struct phy_device *phy_dev; int link; int speed; int duplex; -- cgit v0.10.2 From 176275a261fbffa4e1c009b97265fabac428c891 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Wed, 22 Jun 2016 00:32:36 +0200 Subject: net: ethernet: macb: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cadence/macb.c b/drivers/net/ethernet/cadence/macb.c index 090463f..89c0cfa 100644 --- a/drivers/net/ethernet/cadence/macb.c +++ b/drivers/net/ethernet/cadence/macb.c @@ -2091,28 +2091,6 @@ static struct net_device_stats *macb_get_stats(struct net_device *dev) return nstat; } -static int macb_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct macb *bp = netdev_priv(dev); - struct phy_device *phydev = dev->phydev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_gset(phydev, cmd); -} - -static int macb_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct macb *bp = netdev_priv(dev); - struct phy_device *phydev = dev->phydev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_sset(phydev, cmd); -} - static int macb_get_regs_len(struct net_device *netdev) { return MACB_GREGS_NBR * sizeof(u32); @@ -2185,19 +2163,17 @@ static int macb_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) } static const struct ethtool_ops macb_ethtool_ops = { - .get_settings = macb_get_settings, - .set_settings = macb_set_settings, .get_regs_len = macb_get_regs_len, .get_regs = macb_get_regs, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, .get_wol = macb_get_wol, .set_wol = macb_set_wol, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct ethtool_ops gem_ethtool_ops = { - .get_settings = macb_get_settings, - .set_settings = macb_set_settings, .get_regs_len = macb_get_regs_len, .get_regs = macb_get_regs, .get_link = ethtool_op_get_link, @@ -2205,6 +2181,8 @@ static const struct ethtool_ops gem_ethtool_ops = { .get_ethtool_stats = gem_get_ethtool_stats, .get_strings = gem_get_ethtool_strings, .get_sset_count = gem_get_sset_count, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int macb_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) -- cgit v0.10.2 From eb88d58559b756065667f97ef5891b5c23c57c76 Mon Sep 17 00:00:00 2001 From: William Tu Date: Tue, 21 Jun 2016 21:05:58 -0700 Subject: samples/bpf: set max locked memory to ulimited Signed-off-by: William Tu Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/samples/bpf/sockex2_user.c b/samples/bpf/sockex2_user.c index 29a276d..8a4085c 100644 --- a/samples/bpf/sockex2_user.c +++ b/samples/bpf/sockex2_user.c @@ -5,6 +5,7 @@ #include "bpf_load.h" #include #include +#include struct pair { __u64 packets; @@ -13,11 +14,13 @@ struct pair { int main(int ac, char **argv) { + struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY}; char filename[256]; FILE *f; int i, sock; snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]); + setrlimit(RLIMIT_MEMLOCK, &r); if (load_bpf_file(filename)) { printf("%s", bpf_log_buf); diff --git a/samples/bpf/sockex3_user.c b/samples/bpf/sockex3_user.c index 2617772..d4184ab 100644 --- a/samples/bpf/sockex3_user.c +++ b/samples/bpf/sockex3_user.c @@ -5,6 +5,7 @@ #include "bpf_load.h" #include #include +#include struct flow_keys { __be32 src; @@ -23,11 +24,13 @@ struct pair { int main(int argc, char **argv) { + struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY}; char filename[256]; FILE *f; int i, sock; snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]); + setrlimit(RLIMIT_MEMLOCK, &r); if (load_bpf_file(filename)) { printf("%s", bpf_log_buf); -- cgit v0.10.2 From b38066daaa81fd2f9ca2c9f103d4d56aad8906ee Mon Sep 17 00:00:00 2001 From: Hariprasad Shenai Date: Wed, 22 Jun 2016 09:39:29 +0530 Subject: cxgb4vf: Synchronize access to mailbox The issue comes when there are multiple threads attempting to use the mailbox facility at the same time. The issue is the for the Virtual Function Driver, the only way to get the Virtual Interface statistics is to issue mailbox commands to ask the firmware for the VI Stats. And, because the VI Stats command can only retrieve a smallish number of stats per mailbox command, we have to issue three mailbox commands in quick succession. When ethtool or netstat command to get interface stats and interface up/down is run in a loop for every 0.1 sec, we observed mailbox collisions. And out of the two commands one would fail with the present code, since we don't queue the second command. To overcome the above issue, added a queue to access the mailbox. Whenever a mailbox command is issued add it to the queue. If its at the head issue the mailbox command, else wait for the existing command to complete. Usually command takes less than a milli-second to complete. Also timeout from the loop, if the command under execution takes long time to run. In reality, the number of mailbox access collisions is going to be very rare since no one runs such abusive script. Signed-off-by: Hariprasad Shenai Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h b/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h index 734dd77..109bc63 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h +++ b/drivers/net/ethernet/chelsio/cxgb4vf/adapter.h @@ -353,6 +353,10 @@ struct hash_mac_addr { u8 addr[ETH_ALEN]; }; +struct mbox_list { + struct list_head list; +}; + /* * Per-"adapter" (Virtual Function) information. */ @@ -387,6 +391,10 @@ struct adapter { /* various locks */ spinlock_t stats_lock; + /* lock for mailbox cmd list */ + spinlock_t mbox_lock; + struct mbox_list mlist; + /* support for mailbox command/reply logging */ #define T4VF_OS_LOG_MBOX_CMDS 256 struct mbox_cmd_log *mbox_log; diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index 8d9b2cb..9f55264 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -2774,6 +2774,8 @@ static int cxgb4vf_pci_probe(struct pci_dev *pdev, * Initialize SMP data synchronization resources. */ spin_lock_init(&adapter->stats_lock); + spin_lock_init(&adapter->mbox_lock); + INIT_LIST_HEAD(&adapter->mlist.list); /* * Map our I/O registers in BAR0. diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c index 955ff7c..61bfe86 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c @@ -139,6 +139,7 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, u32 mbox_ctl = T4VF_CIM_BASE_ADDR + CIM_VF_EXT_MAILBOX_CTRL; u32 cmd_op = FW_CMD_OP_G(be32_to_cpu(((struct fw_cmd_hdr *)cmd)->hi)); __be64 cmd_rpl[MBOX_LEN / 8]; + struct mbox_list entry; /* In T6, mailbox size is changed to 128 bytes to avoid * invalidating the entire prefetch buffer. @@ -156,6 +157,51 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, size > NUM_CIM_VF_MAILBOX_DATA_INSTANCES * 4) return -EINVAL; + /* Queue ourselves onto the mailbox access list. When our entry is at + * the front of the list, we have rights to access the mailbox. So we + * wait [for a while] till we're at the front [or bail out with an + * EBUSY] ... + */ + spin_lock(&adapter->mbox_lock); + list_add_tail(&entry.list, &adapter->mlist.list); + spin_unlock(&adapter->mbox_lock); + + delay_idx = 0; + ms = delay[0]; + + for (i = 0; ; i += ms) { + /* If we've waited too long, return a busy indication. This + * really ought to be based on our initial position in the + * mailbox access list but this is a start. We very rearely + * contend on access to the mailbox ... + */ + if (i > FW_CMD_MAX_TIMEOUT) { + spin_lock(&adapter->mbox_lock); + list_del(&entry.list); + spin_unlock(&adapter->mbox_lock); + ret = -EBUSY; + t4vf_record_mbox(adapter, cmd, size, access, ret); + return ret; + } + + /* If we're at the head, break out and start the mailbox + * protocol. + */ + if (list_first_entry(&adapter->mlist.list, struct mbox_list, + list) == &entry) + break; + + /* Delay for a bit before checking again ... */ + if (sleep_ok) { + ms = delay[delay_idx]; /* last element may repeat */ + if (delay_idx < ARRAY_SIZE(delay) - 1) + delay_idx++; + msleep(ms); + } else { + mdelay(ms); + } + } + /* * Loop trying to get ownership of the mailbox. Return an error * if we can't gain ownership. @@ -164,6 +210,9 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, for (i = 0; v == MBOX_OWNER_NONE && i < 3; i++) v = MBOWNER_G(t4_read_reg(adapter, mbox_ctl)); if (v != MBOX_OWNER_DRV) { + spin_lock(&adapter->mbox_lock); + list_del(&entry.list); + spin_unlock(&adapter->mbox_lock); ret = (v == MBOX_OWNER_FW) ? -EBUSY : -ETIMEDOUT; t4vf_record_mbox(adapter, cmd, size, access, ret); return ret; @@ -248,6 +297,9 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, if (cmd_op != FW_VI_STATS_CMD) t4vf_record_mbox(adapter, cmd_rpl, size, access, execute); + spin_lock(&adapter->mbox_lock); + list_del(&entry.list); + spin_unlock(&adapter->mbox_lock); return -FW_CMD_RETVAL_G(v); } } @@ -255,6 +307,9 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, /* We timed out. Return the error ... */ ret = -ETIMEDOUT; t4vf_record_mbox(adapter, cmd, size, access, ret); + spin_lock(&adapter->mbox_lock); + list_del(&entry.list); + spin_unlock(&adapter->mbox_lock); return ret; } -- cgit v0.10.2 From 63da84049bac7658286f1c774594fde3e77b8603 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:03 -0700 Subject: liquidio: soft command buffer limits This patch increases the limits of soft command buffer size and num command buffers. This patch also has changes for queue macros and limit related changes for new chips. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c index 8ad7425..c577559 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c +++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c @@ -367,7 +367,8 @@ void lio_cn6xxx_enable_io_queues(struct octeon_device *oct) void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) { - u32 mask, i, loop = HZ; + int i; + u32 mask, loop = HZ; u32 d32; /* Reset the Enable bits for Input Queues. */ @@ -376,7 +377,7 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) octeon_write_csr(oct, CN6XXX_SLI_PKT_INSTR_ENB, mask); /* Wait until hardware indicates that the queues are out of reset. */ - mask = oct->io_qmask.iq; + mask = (u32)oct->io_qmask.iq; d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_IQ); while (((d32 & mask) != mask) && loop--) { d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_IQ); @@ -384,8 +385,8 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) } /* Reset the doorbell register for each Input queue. */ - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - if (!(oct->io_qmask.iq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { + if (!(oct->io_qmask.iq & (1ULL << i))) continue; octeon_write_csr(oct, CN6XXX_SLI_IQ_DOORBELL(i), 0xFFFFFFFF); d32 = octeon_read_csr(oct, CN6XXX_SLI_IQ_DOORBELL(i)); @@ -398,7 +399,7 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) /* Wait until hardware indicates that the queues are out of reset. */ loop = HZ; - mask = oct->io_qmask.oq; + mask = (u32)oct->io_qmask.oq; d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_OQ); while (((d32 & mask) != mask) && loop--) { d32 = octeon_read_csr(oct, CN6XXX_SLI_PORT_IN_RST_OQ); @@ -408,8 +409,8 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) /* Reset the doorbell register for each Output queue. */ /* for (i = 0; i < oct->num_oqs; i++) { */ - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct->io_qmask.oq & (1ULL << i))) continue; octeon_write_csr(oct, CN6XXX_SLI_OQ_PKTS_CREDIT(i), 0xFFFFFFFF); d32 = octeon_read_csr(oct, CN6XXX_SLI_OQ_PKTS_CREDIT(i)); @@ -429,16 +430,16 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct) void lio_cn6xxx_reinit_regs(struct octeon_device *oct) { - u32 i; + int i; - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - if (!(oct->io_qmask.iq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { + if (!(oct->io_qmask.iq & (1ULL << i))) continue; oct->fn_list.setup_iq_regs(oct, i); } - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct->io_qmask.oq & (1ULL << i))) continue; oct->fn_list.setup_oq_regs(oct, i); } @@ -450,8 +451,8 @@ void lio_cn6xxx_reinit_regs(struct octeon_device *oct) oct->fn_list.enable_io_queues(oct); /* for (i = 0; i < oct->num_oqs; i++) { */ - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct->io_qmask.oq & (1ULL << i))) continue; writel(oct->droq[i]->max_count, oct->droq[i]->pkts_credit_reg); } @@ -557,7 +558,8 @@ lio_cn6xxx_process_pcie_error_intr(struct octeon_device *oct, u64 intr64) int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct) { struct octeon_droq *droq; - u32 oq_no, pkt_count, droq_time_mask, droq_mask, droq_int_enb; + int oq_no; + u32 pkt_count, droq_time_mask, droq_mask, droq_int_enb; u32 droq_cnt_enb, droq_cnt_mask; droq_cnt_enb = octeon_read_csr(oct, CN6XXX_SLI_PKT_CNT_INT_ENB); @@ -573,8 +575,8 @@ int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct) oct->droq_intr = 0; /* for (oq_no = 0; oq_no < oct->num_oqs; oq_no++) { */ - for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES; oq_no++) { - if (!(droq_mask & (1 << oq_no))) + for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES(oct); oq_no++) { + if (!(droq_mask & (1ULL << oq_no))) continue; droq = oct->droq[oq_no]; diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 4523c86..56f465b 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -528,8 +528,8 @@ lio_get_ethtool_stats(struct net_device *netdev, struct octeon_device *oct_dev = lio->oct_dev; int i = 0, j; - for (j = 0; j < MAX_OCTEON_INSTR_QUEUES; j++) { - if (!(oct_dev->io_qmask.iq & (1UL << j))) + for (j = 0; j < MAX_OCTEON_INSTR_QUEUES(oct); j++) { + if (!(oct_dev->io_qmask.iq & (1ULL << j))) continue; data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.instr_posted); @@ -556,8 +556,8 @@ lio_get_ethtool_stats(struct net_device *netdev, } /* for (j = 0; j < oct_dev->num_oqs; j++){ */ - for (j = 0; j < MAX_OCTEON_OUTPUT_QUEUES; j++) { - if (!(oct_dev->io_qmask.oq & (1UL << j))) + for (j = 0; j < MAX_OCTEON_OUTPUT_QUEUES(oct); j++) { + if (!(oct_dev->io_qmask.oq & (1ULL << j))) continue; data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.pkts_received); data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.bytes_received); @@ -581,8 +581,8 @@ static void lio_get_strings(struct net_device *netdev, u32 stringset, u8 *data) int num_iq_stats, num_oq_stats, i, j; num_iq_stats = ARRAY_SIZE(oct_iq_stats_strings); - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - if (!(oct_dev->io_qmask.iq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { + if (!(oct_dev->io_qmask.iq & (1ULL << i))) continue; for (j = 0; j < num_iq_stats; j++) { sprintf(data, "IQ%d %s", i, oct_iq_stats_strings[j]); @@ -592,8 +592,8 @@ static void lio_get_strings(struct net_device *netdev, u32 stringset, u8 *data) num_oq_stats = ARRAY_SIZE(oct_droq_stats_strings); /* for (i = 0; i < oct_dev->num_oqs; i++) { */ - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct_dev->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct_dev->io_qmask.oq & (1ULL << i))) continue; for (j = 0; j < num_oq_stats; j++) { sprintf(data, "OQ%d %s", i, oct_droq_stats_strings[j]); diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index d0ab97c..5a0977f 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -224,8 +224,8 @@ static void octeon_droq_bh(unsigned long pdev) (struct octeon_device_priv *)oct->priv; /* for (q_no = 0; q_no < oct->num_oqs; q_no++) { */ - for (q_no = 0; q_no < MAX_OCTEON_OUTPUT_QUEUES; q_no++) { - if (!(oct->io_qmask.oq & (1UL << q_no))) + for (q_no = 0; q_no < MAX_OCTEON_OUTPUT_QUEUES(oct); q_no++) { + if (!(oct->io_qmask.oq & (1ULL << q_no))) continue; reschedule |= octeon_droq_process_packets(oct, oct->droq[q_no], MAX_PACKET_BUDGET); @@ -245,8 +245,8 @@ static int lio_wait_for_oq_pkts(struct octeon_device *oct) do { pending_pkts = 0; - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct->io_qmask.oq & (1ULL << i))) continue; pkt_cnt += octeon_droq_check_hw_for_pkts(oct, oct->droq[i]); @@ -396,10 +396,10 @@ static inline void pcierror_quiesce_device(struct octeon_device *oct) dev_err(&oct->pci_dev->dev, "There were pending requests\n"); /* Force all requests waiting to be fetched by OCTEON to complete. */ - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { struct octeon_instr_queue *iq; - if (!(oct->io_qmask.iq & (1UL << i))) + if (!(oct->io_qmask.iq & (1ULL << i))) continue; iq = oct->instr_queue[i]; @@ -972,8 +972,9 @@ void liquidio_schedule_droq_pkt_handlers(struct octeon_device *oct) struct octeon_droq *droq; if (oct->int_status & OCT_DEV_INTR_PKT_DATA) { - for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES; oq_no++) { - if (!(oct->droq_intr & (1 << oq_no))) + for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES(oct); + oq_no++) { + if (!(oct->droq_intr & (1ULL << oq_no))) continue; droq = oct->droq[oq_no]; @@ -1160,8 +1161,8 @@ static void octeon_destroy_resources(struct octeon_device *oct) case OCT_DEV_DROQ_INIT_DONE: /*atomic_set(&oct->status, OCT_DEV_DROQ_INIT_DONE);*/ mdelay(100); - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { - if (!(oct->io_qmask.oq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { + if (!(oct->io_qmask.oq & (1ULL << i))) continue; octeon_delete_droq(oct, i); } @@ -1188,8 +1189,8 @@ static void octeon_destroy_resources(struct octeon_device *oct) /* fallthrough */ case OCT_DEV_INSTR_QUEUE_INIT_DONE: - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - if (!(oct->io_qmask.iq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { + if (!(oct->io_qmask.iq & (1ULL << i))) continue; octeon_delete_instr_queue(oct, i); } diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_config.h b/drivers/net/ethernet/cavium/liquidio/octeon_config.h index 62a8dd5..4b8c948 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_config.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_config.h @@ -37,7 +37,7 @@ /* Maximum octeon devices defined as MAX_OCTEON_NICIF to support * multiple(<= MAX_OCTEON_NICIF) Miniports */ -#define MAX_OCTEON_NICIF 32 +#define MAX_OCTEON_NICIF 128 #define MAX_OCTEON_DEVICES MAX_OCTEON_NICIF #define MAX_OCTEON_LINKS MAX_OCTEON_NICIF #define MAX_OCTEON_MULTICAST_ADDR 32 @@ -135,7 +135,7 @@ #define CFG_GET_IS_SLI_BP_ON(cfg) ((cfg)->misc.enable_sli_oq_bp) /* Max IOQs per OCTEON Link */ -#define MAX_IOQS_PER_NICIF 32 +#define MAX_IOQS_PER_NICIF 64 enum lio_card_type { LIO_210SV = 0, /* Two port, 66xx */ @@ -416,9 +416,11 @@ struct octeon_config { #define DISPATCH_LIST_SIZE BIT(OPCODE_MASK_BITS) /* Maximum number of Octeon Instruction (command) queues */ -#define MAX_OCTEON_INSTR_QUEUES CN6XXX_MAX_INPUT_QUEUES +#define MAX_OCTEON_INSTR_QUEUES(oct) CN6XXX_MAX_INPUT_QUEUES +/* Maximum number of Octeon Output queues */ +#define MAX_OCTEON_OUTPUT_QUEUES(oct) CN6XXX_MAX_OUTPUT_QUEUES -/* Maximum number of Octeon Instruction (command) queues */ -#define MAX_OCTEON_OUTPUT_QUEUES CN6XXX_MAX_OUTPUT_QUEUES +#define MAX_POSSIBLE_OCTEON_INSTR_QUEUES CN6XXX_MAX_INPUT_QUEUES +#define MAX_POSSIBLE_OCTEON_OUTPUT_QUEUES CN6XXX_MAX_OUTPUT_QUEUES #endif /* __OCTEON_CONFIG_H__ */ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index 7b44b5c..e1ca617 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -645,12 +645,12 @@ void octeon_free_device_mem(struct octeon_device *oct) { u32 i; - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES; i++) { + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { /* could check mask as well */ vfree(oct->droq[i]); } - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { /* could check mask as well */ vfree(oct->instr_queue[i]); } @@ -734,7 +734,7 @@ struct octeon_device *octeon_allocate_device(u32 pci_id, octeon_device[oct_idx] = oct; oct->octeon_id = oct_idx; - snprintf((oct->device_name), sizeof(oct->device_name), + snprintf(oct->device_name, sizeof(oct->device_name), "LiquidIO%d", (oct->octeon_id)); return oct; @@ -1157,8 +1157,8 @@ core_drv_init_err: int octeon_get_tx_qsize(struct octeon_device *oct, u32 q_no) { - if (oct && (q_no < MAX_OCTEON_INSTR_QUEUES) && - (oct->io_qmask.iq & (1UL << q_no))) + if (oct && (q_no < MAX_OCTEON_INSTR_QUEUES(oct)) && + (oct->io_qmask.iq & (1ULL << q_no))) return oct->instr_queue[q_no]->max_count; return -1; @@ -1166,8 +1166,8 @@ int octeon_get_tx_qsize(struct octeon_device *oct, u32 q_no) int octeon_get_rx_qsize(struct octeon_device *oct, u32 q_no) { - if (oct && (q_no < MAX_OCTEON_OUTPUT_QUEUES) && - (oct->io_qmask.oq & (1UL << q_no))) + if (oct && (q_no < MAX_OCTEON_OUTPUT_QUEUES(oct)) && + (oct->io_qmask.oq & (1ULL << q_no))) return oct->droq[q_no]->max_count; return -1; } @@ -1258,10 +1258,10 @@ void lio_pci_writeq(struct octeon_device *oct, int octeon_mem_access_ok(struct octeon_device *oct) { u64 access_okay = 0; + u64 lmc0_reset_ctl; /* Check to make sure a DDR interface is enabled */ - u64 lmc0_reset_ctl = lio_pci_readq(oct, CN6XXX_LMC0_RESET_CTL); - + lmc0_reset_ctl = lio_pci_readq(oct, CN6XXX_LMC0_RESET_CTL); access_okay = (lmc0_reset_ctl & CN6XXX_LMC0_RESET_CTL_DDR3RST_MASK); return access_okay ? 0 : 1; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h index 0950b94..dd88df7 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h @@ -152,9 +152,9 @@ struct octeon_mmio { #define MAX_OCTEON_MAPS 32 struct octeon_io_enable { - u32 iq; - u32 oq; - u32 iq64B; + u64 iq; + u64 oq; + u64 iq64B; }; struct octeon_reg_list { @@ -325,7 +325,8 @@ struct octeon_device { struct octeon_sc_buffer_pool sc_buf_pool; /** The input instruction queues */ - struct octeon_instr_queue *instr_queue[MAX_OCTEON_INSTR_QUEUES]; + struct octeon_instr_queue *instr_queue + [MAX_POSSIBLE_OCTEON_INSTR_QUEUES]; /** The doubly-linked list of instruction response */ struct octeon_response_list response_list[MAX_RESPONSE_LISTS]; @@ -333,7 +334,7 @@ struct octeon_device { u32 num_oqs; /** The DROQ output queues */ - struct octeon_droq *droq[MAX_OCTEON_OUTPUT_QUEUES]; + struct octeon_droq *droq[MAX_POSSIBLE_OCTEON_OUTPUT_QUEUES]; struct octeon_io_enable io_qmask; @@ -382,7 +383,7 @@ struct octeon_device { struct cavium_wq dma_comp_wq; - struct cavium_wq check_db_wq[MAX_OCTEON_INSTR_QUEUES]; + struct cavium_wq check_db_wq[MAX_POSSIBLE_OCTEON_INSTR_QUEUES]; struct cavium_wk nic_poll_work; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c index 59a5293..d9bb2f7 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c @@ -337,7 +337,7 @@ int octeon_init_droq(struct octeon_device *oct, /* For 56xx Pass1, this function won't be called, so no checks. */ oct->fn_list.setup_oq_regs(oct, q_no); - oct->io_qmask.oq |= (1 << q_no); + oct->io_qmask.oq |= (1ULL << q_no); return 0; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h index 513f8a0..7c275ef 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h @@ -244,7 +244,7 @@ union octeon_instr_64B { /** The size of each buffer in soft command buffer pool */ -#define SOFT_COMMAND_BUFFER_SIZE 1024 +#define SOFT_COMMAND_BUFFER_SIZE 1536 struct octeon_soft_command { /** Soft command buffer info. */ @@ -282,7 +282,7 @@ struct octeon_soft_command { /** Maximum number of buffers to allocate into soft command buffer pool */ -#define MAX_SOFT_COMMAND_BUFFERS 16 +#define MAX_SOFT_COMMAND_BUFFERS 256 /** Head of a soft command buffer pool. */ diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index 8649677..51031010 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -150,7 +150,7 @@ int octeon_init_instr_queue(struct octeon_device *oct, /* Initialize the spinlock for this instruction queue */ spin_lock_init(&iq->lock); - oct->io_qmask.iq |= (1 << iq_no); + oct->io_qmask.iq |= (1ULL << iq_no); /* Set the 32B/64B mode for each input queue */ oct->io_qmask.iq64B |= ((conf->instr_type == 64) << iq_no); @@ -253,8 +253,8 @@ int lio_wait_for_instr_fetch(struct octeon_device *oct) instr_cnt = 0; /*for (i = 0; i < oct->num_iqs; i++) {*/ - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES; i++) { - if (!(oct->io_qmask.iq & (1UL << i))) + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { + if (!(oct->io_qmask.iq & (1ULL << i))) continue; pending = atomic_read(&oct-> -- cgit v0.10.2 From 0da0b77cde2a4d55968457ddee2e987d0783f29b Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:04 -0700 Subject: liquidio: Vlan offloads changes This patch adds support for vlan offloads for the driver and receive header structures are also modified appropriately. Also requestID will not be used in reveive header any more. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 5a0977f..4b95dbf 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -1849,6 +1849,7 @@ liquidio_push_packet(u32 octeon_id, struct sk_buff *skb = (struct sk_buff *)skbuff; struct skb_shared_hwtstamps *shhwtstamps; u64 ns; + u16 vtag = 0; struct net_device *netdev = (struct net_device *)arg; struct octeon_droq *droq = container_of(param, struct octeon_droq, napi); @@ -1925,6 +1926,16 @@ liquidio_push_packet(u32 octeon_id, else skb->ip_summed = CHECKSUM_NONE; + /* inbound VLAN tag */ + if ((netdev->features & NETIF_F_HW_VLAN_CTAG_RX) && + (rh->r_dh.vlan != 0)) { + u16 vid = rh->r_dh.vlan; + u16 priority = rh->r_dh.priority; + + vtag = priority << 13 | vid; + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vtag); + } + packet_was_received = napi_gro_receive(napi, skb) != GRO_DROP; if (packet_was_received) { @@ -2900,6 +2911,11 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) tx_info->s.gso_size = skb_shinfo(skb)->gso_size; tx_info->s.gso_segs = skb_shinfo(skb)->gso_segs; } + /* HW insert VLAN tag */ + if (skb_vlan_tag_present(skb)) { + irh->priority = skb_vlan_tag_get(skb) >> 13; + irh->vlan = skb_vlan_tag_get(skb) & 0xfff; + } xmit_more = skb->xmit_more; @@ -3301,11 +3317,17 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) | NETIF_F_LRO; netif_set_gso_max_size(netdev, OCTNIC_GSO_MAX_SIZE); - netdev->features = (lio->dev_capability & ~NETIF_F_LRO); - netdev->vlan_features = lio->dev_capability; + /* Add any unchangeable hw features */ + lio->dev_capability |= NETIF_F_HW_VLAN_CTAG_RX | + NETIF_F_HW_VLAN_CTAG_TX; + + netdev->features = (lio->dev_capability & ~NETIF_F_LRO); netdev->hw_features = lio->dev_capability; + /*HW_VLAN_RX and HW_VLAN_FILTER is always on*/ + netdev->hw_features = netdev->hw_features & + ~NETIF_F_HW_VLAN_CTAG_RX; /* Point to the properties for octeon device to which this * interface belongs. diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 2179691..c86421f 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -482,15 +482,15 @@ struct octeon_instr_irh { u64 opcode:4; u64 rflag:1; u64 subcode:7; - u64 len:3; - u64 rid:13; - u64 reserved:4; + u64 vlan:12; + u64 priority:3; + u64 reserved:5; u64 ossp:32; /* opcode/subcode specific parameters */ #else u64 ossp:32; /* opcode/subcode specific parameters */ - u64 reserved:4; - u64 rid:13; - u64 len:3; + u64 reserved:5; + u64 priority:3; + u64 vlan:12; u64 subcode:7; u64 rflag:1; u64 opcode:4; @@ -517,28 +517,27 @@ union octeon_rh { struct { u64 opcode:4; u64 subcode:8; - u64 len:3; /** additional 64-bit words */ - u64 rid:13; /** request id in response to pkt sent by host */ - u64 reserved:4; - u64 ossp:32; /** opcode/subcode specific parameters */ + u64 len:3; /** additional 64-bit words */ + u64 reserved:17; + u64 ossp:32; /** opcode/subcode specific parameters */ } r; struct { u64 opcode:4; u64 subcode:8; - u64 len:3; /** additional 64-bit words */ - u64 rid:13; /** request id in response to pkt sent by host */ - u64 extra:24; - u64 link:8; + u64 len:3; /** additional 64-bit words */ + u64 extra:28; + u64 vlan:12; + u64 priority:3; u64 csum_verified:3; /** checksum verified. */ u64 has_hwtstamp:1; /** Has hardware timestamp. 1 = yes. */ } r_dh; struct { u64 opcode:4; u64 subcode:8; - u64 len:3; /** additional 64-bit words */ - u64 rid:13; /** request id in response to pkt sent by host */ + u64 len:3; /** additional 64-bit words */ + u64 reserved:11; u64 num_gmx_ports:8; - u64 max_nic_ports:8; + u64 max_nic_ports:10; u64 app_cap_flags:4; u64 app_mode:16; } r_core_drv_init; @@ -554,8 +553,7 @@ union octeon_rh { u64 u64; struct { u64 ossp:32; /** opcode/subcode specific parameters */ - u64 reserved:4; - u64 rid:13; /** req id in response to pkt sent by host */ + u64 reserved:17; u64 len:3; /** additional 64-bit words */ u64 subcode:8; u64 opcode:4; @@ -563,9 +561,9 @@ union octeon_rh { struct { u64 has_hwtstamp:1; /** 1 = has hwtstamp */ u64 csum_verified:3; /** checksum verified. */ - u64 link:8; - u64 extra:24; - u64 rid:13; /** req id in response to pkt sent by host */ + u64 priority:3; + u64 vlan:12; + u64 extra:28; u64 len:3; /** additional 64-bit words */ u64 subcode:8; u64 opcode:4; @@ -573,9 +571,9 @@ union octeon_rh { struct { u64 app_mode:16; u64 app_cap_flags:4; - u64 max_nic_ports:8; + u64 max_nic_ports:10; u64 num_gmx_ports:8; - u64 rid:13; + u64 reserved:11; u64 len:3; /** additional 64-bit words */ u64 subcode:8; u64 opcode:4; -- cgit v0.10.2 From 63245f25715c5cff19bfdf15bf28511a5ff90c8f Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:05 -0700 Subject: liquidio: Vlan filtering This patch adds supports for Vlan filtering for liquidio driver. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 4b95dbf..9f3a93b 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -2310,6 +2310,21 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr) netdev->name); break; + case OCTNET_CMD_ENABLE_VLAN_FILTER: + dev_info(&oct->pci_dev->dev, "%s VLAN filter enabled\n", + netdev->name); + break; + + case OCTNET_CMD_ADD_VLAN_FILTER: + dev_info(&oct->pci_dev->dev, "%s VLAN filter %d added\n", + netdev->name, nctrl->ncmd.s.param1); + break; + + case OCTNET_CMD_DEL_VLAN_FILTER: + dev_info(&oct->pci_dev->dev, "%s VLAN filter %d removed\n", + netdev->name, nctrl->ncmd.s.param1); + break; + case OCTNET_CMD_SET_SETTINGS: dev_info(&oct->pci_dev->dev, "%s settings changed\n", netdev->name); @@ -2965,6 +2980,61 @@ static void liquidio_tx_timeout(struct net_device *netdev) txqs_wake(netdev); } +static int liquidio_vlan_rx_add_vid(struct net_device *netdev, + __be16 proto __attribute__((unused)), + u16 vid) +{ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; + struct octnic_ctrl_pkt nctrl; + int ret = 0; + + memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); + + nctrl.ncmd.u64 = 0; + nctrl.ncmd.s.cmd = OCTNET_CMD_ADD_VLAN_FILTER; + nctrl.ncmd.s.param1 = vid; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; + nctrl.wait_time = 100; + nctrl.netpndev = (u64)netdev; + nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; + + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); + if (ret < 0) { + dev_err(&oct->pci_dev->dev, "Add VLAN filter failed in core (ret: 0x%x)\n", + ret); + } + + return ret; +} + +static int liquidio_vlan_rx_kill_vid(struct net_device *netdev, + __be16 proto __attribute__((unused)), + u16 vid) +{ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; + struct octnic_ctrl_pkt nctrl; + int ret = 0; + + memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); + + nctrl.ncmd.u64 = 0; + nctrl.ncmd.s.cmd = OCTNET_CMD_DEL_VLAN_FILTER; + nctrl.ncmd.s.param1 = vid; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; + nctrl.wait_time = 100; + nctrl.netpndev = (u64)netdev; + nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; + + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); + if (ret < 0) { + dev_err(&oct->pci_dev->dev, "Add VLAN filter failed in core (ret: 0x%x)\n", + ret); + } + return ret; +} + int liquidio_set_feature(struct net_device *netdev, int cmd, u16 param1) { struct lio *lio = GET_LIO(netdev); @@ -3056,6 +3126,9 @@ static struct net_device_ops lionetdevops = { .ndo_set_mac_address = liquidio_set_mac, .ndo_set_rx_mode = liquidio_set_mcast_list, .ndo_tx_timeout = liquidio_tx_timeout, + + .ndo_vlan_rx_add_vid = liquidio_vlan_rx_add_vid, + .ndo_vlan_rx_kill_vid = liquidio_vlan_rx_kill_vid, .ndo_change_mtu = liquidio_change_mtu, .ndo_do_ioctl = liquidio_ioctl, .ndo_fix_features = liquidio_fix_features, @@ -3319,7 +3392,8 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) netdev->vlan_features = lio->dev_capability; /* Add any unchangeable hw features */ - lio->dev_capability |= NETIF_F_HW_VLAN_CTAG_RX | + lio->dev_capability |= NETIF_F_HW_VLAN_CTAG_FILTER | + NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_TX; netdev->features = (lio->dev_capability & ~NETIF_F_LRO); @@ -3377,9 +3451,11 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE, OCTNIC_LROIPV4 | OCTNIC_LROIPV6); + liquidio_set_feature(netdev, OCTNET_CMD_ENABLE_VLAN_FILTER, 0); + if ((debug != -1) && (debug & NETIF_MSG_HW)) - liquidio_set_feature(netdev, OCTNET_CMD_VERBOSE_ENABLE, - 0); + liquidio_set_feature(netdev, + OCTNET_CMD_VERBOSE_ENABLE, 0); /* Register the network device with the OS */ if (register_netdev(netdev)) { diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index c86421f..3738877 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -214,6 +214,10 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry, #define OCTNET_CMD_VERBOSE_ENABLE 0x14 #define OCTNET_CMD_VERBOSE_DISABLE 0x15 +#define OCTNET_CMD_ENABLE_VLAN_FILTER 0x16 +#define OCTNET_CMD_ADD_VLAN_FILTER 0x17 +#define OCTNET_CMD_DEL_VLAN_FILTER 0x18 + /* RX(packets coming from wire) Checksum verification flags */ /* TCP/UDP csum */ #define CNNIC_L4SUM_VERIFIED 0x1 -- cgit v0.10.2 From 9a96bde4e1b61705aaa8e769349f5577b94c1fc4 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:06 -0700 Subject: liquidio: Napi rx/tx traffic This Patch adds tx buffer handling to Napi along with RX traffic. Also separate spinlocks are introduced for handling iq posting and buffer reclaim so that tx path and tx interrupt do not compete against each other. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c index c577559..d35864a 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c +++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c @@ -496,8 +496,7 @@ u32 lio_cn6xxx_bar1_idx_read(struct octeon_device *oct, u32 idx) } u32 -lio_cn6xxx_update_read_index(struct octeon_device *oct __attribute__((unused)), - struct octeon_instr_queue *iq) +lio_cn6xxx_update_read_index(struct octeon_instr_queue *iq) { u32 new_idx = readl(iq->inst_cnt_reg); diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h index f779187..fe2932c 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h +++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h @@ -91,8 +91,7 @@ void lio_cn6xxx_bar1_idx_setup(struct octeon_device *oct, u64 core_addr, void lio_cn6xxx_bar1_idx_write(struct octeon_device *oct, u32 idx, u32 mask); u32 lio_cn6xxx_bar1_idx_read(struct octeon_device *oct, u32 idx); u32 -lio_cn6xxx_update_read_index(struct octeon_device *oct __attribute__((unused)), - struct octeon_instr_queue *iq); +lio_cn6xxx_update_read_index(struct octeon_instr_queue *iq); void lio_cn6xxx_enable_interrupt(void *chip); void lio_cn6xxx_disable_interrupt(void *chip); void cn6xxx_get_pcie_qlmport(struct octeon_device *oct); diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 9f3a93b..8310eb8 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -365,7 +365,7 @@ static int wait_for_pending_requests(struct octeon_device *oct) [OCTEON_ORDERED_SC_LIST].pending_req_count); if (pcount) schedule_timeout_uninterruptible(HZ / 10); - else + else break; } @@ -409,7 +409,7 @@ static inline void pcierror_quiesce_device(struct octeon_device *oct) iq->octeon_read_index = iq->host_write_index; iq->stats.instr_processed += atomic_read(&iq->instr_pending); - lio_process_iq_request_list(oct, iq); + lio_process_iq_request_list(oct, iq, 0); spin_unlock_bh(&iq->lock); } } @@ -959,6 +959,36 @@ static inline void update_link_status(struct net_device *netdev, } } +/* Runs in interrupt context. */ +static void update_txq_status(struct octeon_device *oct, int iq_num) +{ + struct net_device *netdev; + struct lio *lio; + struct octeon_instr_queue *iq = oct->instr_queue[iq_num]; + + /*octeon_update_iq_read_idx(oct, iq);*/ + + netdev = oct->props[iq->ifidx].netdev; + + /* This is needed because the first IQ does not have + * a netdev associated with it. + */ + if (!netdev) + return; + + lio = GET_LIO(netdev); + if (netif_is_multiqueue(netdev)) { + if (__netif_subqueue_stopped(netdev, iq->q_index) && + lio->linfo.link.s.link_up && + (!octnet_iq_is_full(oct, iq_num))) { + netif_wake_subqueue(netdev, iq->q_index); + } else { + if (!octnet_iq_is_full(oct, lio->txq)) + wake_q(netdev, lio->txq); + } + } +} + /** * \brief Droq packet processor sceduler * @param oct octeon device @@ -1246,6 +1276,7 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx) { struct net_device *netdev = oct->props[ifidx].netdev; struct lio *lio; + struct napi_struct *napi, *n; if (!netdev) { dev_err(&oct->pci_dev->dev, "%s No netdevice ptr for index %d\n", @@ -1262,6 +1293,13 @@ static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx) if (atomic_read(&lio->ifstate) & LIO_IFSTATE_RUNNING) txqs_stop(netdev); + if (oct->props[lio->ifidx].napi_enabled == 1) { + list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list) + napi_disable(napi); + + oct->props[lio->ifidx].napi_enabled = 0; + } + if (atomic_read(&lio->ifstate) & LIO_IFSTATE_REGISTERED) unregister_netdev(netdev); @@ -1990,39 +2028,6 @@ static void liquidio_napi_drv_callback(void *arg) } /** - * \brief Main NAPI poll function - * @param droq octeon output queue - * @param budget maximum number of items to process - */ -static int liquidio_napi_do_rx(struct octeon_droq *droq, int budget) -{ - int work_done; - struct lio *lio = GET_LIO(droq->napi.dev); - struct octeon_device *oct = lio->oct_dev; - - work_done = octeon_process_droq_poll_cmd(oct, droq->q_no, - POLL_EVENT_PROCESS_PKTS, - budget); - if (work_done < 0) { - netif_info(lio, rx_err, lio->netdev, - "Receive work_done < 0, rxq:%d\n", droq->q_no); - goto octnet_napi_finish; - } - - if (work_done > budget) - dev_err(&oct->pci_dev->dev, ">>>> %s work_done: %d budget: %d\n", - __func__, work_done, budget); - - return work_done; - -octnet_napi_finish: - napi_complete(&droq->napi); - octeon_process_droq_poll_cmd(oct, droq->q_no, POLL_EVENT_ENABLE_INTR, - 0); - return 0; -} - -/** * \brief Entry point for NAPI polling * @param napi NAPI structure * @param budget maximum number of items to process @@ -2031,19 +2036,41 @@ static int liquidio_napi_poll(struct napi_struct *napi, int budget) { struct octeon_droq *droq; int work_done; + int tx_done = 0, iq_no; + struct octeon_instr_queue *iq; + struct octeon_device *oct; droq = container_of(napi, struct octeon_droq, napi); + oct = droq->oct_dev; + iq_no = droq->q_no; + /* Handle Droq descriptors */ + work_done = octeon_process_droq_poll_cmd(oct, droq->q_no, + POLL_EVENT_PROCESS_PKTS, + budget); - work_done = liquidio_napi_do_rx(droq, budget); + /* Flush the instruction queue */ + iq = oct->instr_queue[iq_no]; + if (iq) { + /* Process iq buffers with in the budget limits */ + tx_done = octeon_flush_iq(oct, iq, 1, budget); + /* Update iq read-index rather than waiting for next interrupt. + * Return back if tx_done is false. + */ + update_txq_status(oct, iq_no); + /*tx_done = (iq->flush_index == iq->octeon_read_index);*/ + } else { + dev_err(&oct->pci_dev->dev, "%s: iq (%d) num invalid\n", + __func__, iq_no); + } - if (work_done < budget) { + if ((work_done < budget) && (tx_done)) { napi_complete(napi); octeon_process_droq_poll_cmd(droq->oct_dev, droq->q_no, POLL_EVENT_ENABLE_INTR, 0); return 0; } - return work_done; + return (!tx_done) ? (budget) : (work_done); } /** @@ -2177,6 +2204,14 @@ static inline void setup_tx_poll_fn(struct net_device *netdev) &lio->txq_status_wq.wk.work, msecs_to_jiffies(1)); } +static inline void cleanup_tx_poll_fn(struct net_device *netdev) +{ + struct lio *lio = GET_LIO(netdev); + + cancel_delayed_work_sync(&lio->txq_status_wq.wk.work); + destroy_workqueue(lio->txq_status_wq.wq); +} + /** * \brief Net device open for LiquidIO * @param netdev network device @@ -2187,17 +2222,22 @@ static int liquidio_open(struct net_device *netdev) struct octeon_device *oct = lio->oct_dev; struct napi_struct *napi, *n; - list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list) - napi_enable(napi); + if (oct->props[lio->ifidx].napi_enabled == 0) { + list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list) + napi_enable(napi); + + oct->props[lio->ifidx].napi_enabled = 1; + } oct_ptp_open(netdev); ifstate_set(lio, LIO_IFSTATE_RUNNING); + setup_tx_poll_fn(netdev); + start_txq(netdev); netif_info(lio, ifup, lio->netdev, "Interface Open, ready for traffic\n"); - try_module_get(THIS_MODULE); /* tell Octeon to start forwarding packets to host */ send_rx_ctrl_cmd(lio, 1); @@ -2217,39 +2257,35 @@ static int liquidio_open(struct net_device *netdev) */ static int liquidio_stop(struct net_device *netdev) { - struct napi_struct *napi, *n; struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; - netif_info(lio, ifdown, lio->netdev, "Stopping interface!\n"); + ifstate_reset(lio, LIO_IFSTATE_RUNNING); + + netif_tx_disable(netdev); + /* Inform that netif carrier is down */ + netif_carrier_off(netdev); lio->intf_open = 0; lio->linfo.link.s.link_up = 0; lio->link_changes++; - netif_carrier_off(netdev); + /* Pause for a moment and wait for Octeon to flush out (to the wire) any + * egress packets that are in-flight. + */ + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(msecs_to_jiffies(100)); - /* tell Octeon to stop forwarding packets to host */ + /* Now it should be safe to tell Octeon that nic interface is down. */ send_rx_ctrl_cmd(lio, 0); - cancel_delayed_work_sync(&lio->txq_status_wq.wk.work); - destroy_workqueue(lio->txq_status_wq.wq); + cleanup_tx_poll_fn(netdev); if (lio->ptp_clock) { ptp_clock_unregister(lio->ptp_clock); lio->ptp_clock = NULL; } - ifstate_reset(lio, LIO_IFSTATE_RUNNING); - - /* This is a hack that allows DHCP to continue working. */ - set_bit(__LINK_STATE_START, &lio->netdev->state); - - list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list) - napi_disable(napi); - - txqs_stop(netdev); - dev_info(&oct->pci_dev->dev, "%s interface is stopped\n", netdev->name); module_put(THIS_MODULE); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h index dd88df7..abfc0d6 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h @@ -204,8 +204,7 @@ struct octeon_fn_list { void (*bar1_idx_setup)(struct octeon_device *, u64, u32, int); void (*bar1_idx_write)(struct octeon_device *, u32, u32); u32 (*bar1_idx_read)(struct octeon_device *, u32); - u32 (*update_iq_read_idx)(struct octeon_device *, - struct octeon_instr_queue *); + u32 (*update_iq_read_idx)(struct octeon_instr_queue *); void (*enable_oq_pkt_time_intr)(struct octeon_device *, u32); void (*disable_oq_pkt_time_intr)(struct octeon_device *, u32); @@ -267,6 +266,7 @@ struct octdev_props { /* Each interface in the Octeon device has a network * device pointer (used for OS specific calls). */ + int napi_enabled; int gmxport; struct net_device *netdev; }; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h index 7c275ef..69d5b91 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h @@ -80,6 +80,12 @@ struct octeon_instr_queue { /** A spinlock to protect access to the input ring. */ spinlock_t lock; + /** A spinlock to protect while posting on the ring. */ + spinlock_t post_lock; + + /** A spinlock to protect access to the input ring.*/ + spinlock_t iq_flush_running_lock; + /** Flag that indicates if the queue uses 64 byte commands. */ u32 iqcmd_64B:1; @@ -339,7 +345,7 @@ octeon_register_reqtype_free_fn(struct octeon_device *oct, int reqtype, int lio_process_iq_request_list(struct octeon_device *oct, - struct octeon_instr_queue *iq); + struct octeon_instr_queue *iq, u32 napi_budget); int octeon_send_command(struct octeon_device *oct, u32 iq_no, u32 force_db, void *cmd, void *buf, @@ -357,5 +363,7 @@ int octeon_send_soft_command(struct octeon_device *oct, int octeon_setup_iq(struct octeon_device *oct, int ifidx, int q_index, union oct_txpciq iq_no, u32 num_descs, void *app_ctx); - +int +octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq, + u32 pending_thresh, u32 napi_budget); #endif /* __OCTEON_IQ_H__ */ diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index 51031010..7eafa75 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -51,7 +51,7 @@ struct iq_post_status { }; static void check_db_timeout(struct work_struct *work); -static void __check_db_timeout(struct octeon_device *oct, unsigned long iq_no); +static void __check_db_timeout(struct octeon_device *oct, u64 iq_no); static void (*reqtype_free_fn[MAX_OCTEON_DEVICES][REQTYPE_LAST + 1]) (void *); @@ -149,6 +149,9 @@ int octeon_init_instr_queue(struct octeon_device *oct, /* Initialize the spinlock for this instruction queue */ spin_lock_init(&iq->lock); + spin_lock_init(&iq->post_lock); + + spin_lock_init(&iq->iq_flush_running_lock); oct->io_qmask.iq |= (1ULL << iq_no); @@ -391,13 +394,13 @@ __add_to_request_list(struct octeon_instr_queue *iq, int lio_process_iq_request_list(struct octeon_device *oct, - struct octeon_instr_queue *iq) + struct octeon_instr_queue *iq, u32 napi_budget) { int reqtype; void *buf; u32 old = iq->flush_index; u32 inst_count = 0; - unsigned pkts_compl = 0, bytes_compl = 0; + unsigned int pkts_compl = 0, bytes_compl = 0; struct octeon_soft_command *sc; struct octeon_instr_irh *irh; @@ -457,6 +460,9 @@ lio_process_iq_request_list(struct octeon_device *oct, skip_this: inst_count++; INCR_INDEX_BY1(old, iq->max_count); + + if ((napi_budget) && (inst_count >= napi_budget)) + break; } if (bytes_compl) octeon_report_tx_completion_to_bql(iq->app_ctx, pkts_compl, @@ -466,38 +472,63 @@ lio_process_iq_request_list(struct octeon_device *oct, return inst_count; } -static inline void -update_iq_indices(struct octeon_device *oct, struct octeon_instr_queue *iq) +/* Can only be called from process context */ +int +octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq, + u32 pending_thresh, u32 napi_budget) { u32 inst_processed = 0; + u32 tot_inst_processed = 0; + int tx_done = 1; - /* Calculate how many commands Octeon has read and move the read index - * accordingly. - */ - iq->octeon_read_index = oct->fn_list.update_iq_read_idx(oct, iq); + if (!spin_trylock(&iq->iq_flush_running_lock)) + return tx_done; - /* Move the NORESPONSE requests to the per-device completion list. */ - if (iq->flush_index != iq->octeon_read_index) - inst_processed = lio_process_iq_request_list(oct, iq); + spin_lock_bh(&iq->lock); - if (inst_processed) { - atomic_sub(inst_processed, &iq->instr_pending); - iq->stats.instr_processed += inst_processed; - } -} + iq->octeon_read_index = oct->fn_list.update_iq_read_idx(iq); -static void -octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq, - u32 pending_thresh) -{ if (atomic_read(&iq->instr_pending) >= (s32)pending_thresh) { - spin_lock_bh(&iq->lock); - update_iq_indices(oct, iq); - spin_unlock_bh(&iq->lock); + do { + /* Process any outstanding IQ packets. */ + if (iq->flush_index == iq->octeon_read_index) + break; + + if (napi_budget) + inst_processed = lio_process_iq_request_list + (oct, iq, + napi_budget - tot_inst_processed); + else + inst_processed = + lio_process_iq_request_list(oct, iq, 0); + + if (inst_processed) { + atomic_sub(inst_processed, &iq->instr_pending); + iq->stats.instr_processed += inst_processed; + } + + tot_inst_processed += inst_processed; + inst_processed = 0; + + } while (tot_inst_processed < napi_budget); + + if (napi_budget && (tot_inst_processed >= napi_budget)) + tx_done = 0; } + + iq->last_db_time = jiffies; + + spin_unlock_bh(&iq->lock); + + spin_unlock(&iq->iq_flush_running_lock); + + return tx_done; } -static void __check_db_timeout(struct octeon_device *oct, unsigned long iq_no) +/* Process instruction queue after timeout. + * This routine gets called from a workqueue or when removing the module. + */ +static void __check_db_timeout(struct octeon_device *oct, u64 iq_no) { struct octeon_instr_queue *iq; u64 next_time; @@ -508,24 +539,17 @@ static void __check_db_timeout(struct octeon_device *oct, unsigned long iq_no) if (!iq) return; + /* return immediately, if no work pending */ + if (!atomic_read(&iq->instr_pending)) + return; /* If jiffies - last_db_time < db_timeout do nothing */ next_time = iq->last_db_time + iq->db_timeout; if (!time_after(jiffies, (unsigned long)next_time)) return; iq->last_db_time = jiffies; - /* Get the lock and prevent tasklets. This routine gets called from - * the poll thread. Instructions can now be posted in tasklet context - */ - spin_lock_bh(&iq->lock); - if (iq->fill_cnt != 0) - ring_doorbell(oct, iq); - - spin_unlock_bh(&iq->lock); - /* Flush the instruction queue */ - if (iq->do_auto_flush) - octeon_flush_iq(oct, iq, 1); + octeon_flush_iq(oct, iq, 1, 0); } /* Called by the Poll thread at regular intervals to check the instruction @@ -550,7 +574,10 @@ octeon_send_command(struct octeon_device *oct, u32 iq_no, struct iq_post_status st; struct octeon_instr_queue *iq = oct->instr_queue[iq_no]; - spin_lock_bh(&iq->lock); + /* Get the lock and prevent other tasks and tx interrupt handler from + * running. + */ + spin_lock_bh(&iq->post_lock); st = __post_command2(oct, iq, force_db, cmd); @@ -566,10 +593,13 @@ octeon_send_command(struct octeon_device *oct, u32 iq_no, INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, instr_dropped, 1); } - spin_unlock_bh(&iq->lock); + spin_unlock_bh(&iq->post_lock); - if (iq->do_auto_flush) - octeon_flush_iq(oct, iq, 2); + /* This is only done here to expedite packets being flushed + * for cases where there are no IQ completion interrupts. + */ + /*if (iq->do_auto_flush)*/ + /* octeon_flush_iq(oct, iq, 2, 0);*/ return st.status; } -- cgit v0.10.2 From d3d7e6c65f75de18ced10a98595a847f9f95f0ce Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:07 -0700 Subject: liquidio: Firmware image download This patch has firmware image related changes for: firmware release upon failure, support latest firmware version and firmware download in 4MB chunks. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 8310eb8..5fb1b79 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -1375,6 +1375,7 @@ static int octeon_chip_specific_setup(struct octeon_device *oct) { u32 dev_id, rev_id; int ret = 1; + char *s; pci_read_config_dword(oct->pci_dev, 0, &dev_id); pci_read_config_dword(oct->pci_dev, 8, &rev_id); @@ -1384,22 +1385,27 @@ static int octeon_chip_specific_setup(struct octeon_device *oct) case OCTEON_CN68XX_PCIID: oct->chip_id = OCTEON_CN68XX; ret = lio_setup_cn68xx_octeon_device(oct); + s = "CN68XX"; break; case OCTEON_CN66XX_PCIID: oct->chip_id = OCTEON_CN66XX; ret = lio_setup_cn66xx_octeon_device(oct); + s = "CN66XX"; break; + default: + s = "?"; dev_err(&oct->pci_dev->dev, "Unknown device found (dev_id: %x)\n", dev_id); } if (!ret) - dev_info(&oct->pci_dev->dev, "CN68XX PASS%d.%d %s\n", + dev_info(&oct->pci_dev->dev, "%s PASS%d.%d %s Version: %s\n", s, OCTEON_MAJOR_REV(oct), OCTEON_MINOR_REV(oct), - octeon_get_conf(oct)->card_name); + octeon_get_conf(oct)->card_name, + LIQUIDIO_VERSION); return ret; } @@ -1772,6 +1778,7 @@ static int load_firmware(struct octeon_device *oct) if (ret) { dev_err(&oct->pci_dev->dev, "Request firmware failed. Could not find file %s.\n.", fw_name); + release_firmware(fw); return ret; } @@ -1841,6 +1848,9 @@ static void if_cfg_callback(struct octeon_device *oct, CVM_CAST64(resp->status)); ACCESS_ONCE(ctx->cond) = 1; + snprintf(oct->fw_info.liquidio_firmware_version, 32, "%s", + resp->cfg_info.liquidio_firmware_version); + /* This barrier is required to be sure that the response has been * written fully before waking up the handler */ @@ -3635,6 +3645,7 @@ static void nic_starter(struct work_struct *work) static int octeon_device_init(struct octeon_device *octeon_dev) { int j, ret; + char bootcmd[] = "\n"; struct octeon_device_priv *oct_priv = (struct octeon_device_priv *)octeon_dev->priv; atomic_set(&octeon_dev->status, OCT_DEV_BEGIN_STATE); @@ -3767,6 +3778,9 @@ static int octeon_device_init(struct octeon_device *octeon_dev) return 1; } + /* Divert uboot to take commands from host instead. */ + ret = octeon_console_send_cmd(octeon_dev, bootcmd, 50); + dev_dbg(&octeon_dev->pci_dev->dev, "Initializing consoles\n"); ret = octeon_init_consoles(octeon_dev); if (ret) { diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 3738877..1ef9001 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -30,11 +30,10 @@ #include "octeon_config.h" -#define LIQUIDIO_VERSION "1.1.9" -#define LIQUIDIO_MAJOR_VERSION 1 -#define LIQUIDIO_MINOR_VERSION 1 -#define LIQUIDIO_MICRO_VERSION 9 - +#define LIQUIDIO_BASE_VERSION "1.4" +#define LIQUIDIO_MICRO_VERSION ".1" +#define LIQUIDIO_PACKAGE "" +#define LIQUIDIO_VERSION "1.4.1" #define CONTROL_IQ 0 /** Tag types used by Octeon cores in its work. */ enum octeon_tag_type { @@ -712,6 +711,7 @@ struct liquidio_if_cfg_info { u64 iqmask; /** mask for IQs enabled for the port */ u64 oqmask; /** mask for OQs enabled for the port */ struct oct_link_info linfo; /** initial link information */ + char liquidio_firmware_version[32]; }; /** Stats for each NIC port in RX direction. */ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index e1ca617..bc4d6af 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -549,17 +549,19 @@ static char *get_oct_app_string(u32 app_mode) return oct_dev_app_str[CVM_DRV_INVALID_APP - CVM_DRV_APP_START]; } +u8 fbuf[4 * 1024 * 1024]; + int octeon_download_firmware(struct octeon_device *oct, const u8 *data, size_t size) { int ret = 0; - u8 *p; - u8 *buffer; + u8 *p = fbuf; u32 crc32_result; u64 load_addr; u32 image_len; struct octeon_firmware_file_header *h; - u32 i; + u32 i, rem, base_len = strlen(LIQUIDIO_BASE_VERSION); + char *base; if (size < sizeof(struct octeon_firmware_file_header)) { dev_err(&oct->pci_dev->dev, "Firmware file too small (%d < %d).\n", @@ -575,19 +577,26 @@ int octeon_download_firmware(struct octeon_device *oct, const u8 *data, return -EINVAL; } - crc32_result = - crc32(~0, data, - sizeof(struct octeon_firmware_file_header) - - sizeof(u32)) ^ ~0U; + crc32_result = crc32((unsigned int)~0, data, + sizeof(struct octeon_firmware_file_header) - + sizeof(u32)) ^ ~0U; if (crc32_result != be32_to_cpu(h->crc32)) { dev_err(&oct->pci_dev->dev, "Firmware CRC mismatch (0x%08x != 0x%08x).\n", crc32_result, be32_to_cpu(h->crc32)); return -EINVAL; } - if (memcmp(LIQUIDIO_VERSION, h->version, strlen(LIQUIDIO_VERSION))) { - dev_err(&oct->pci_dev->dev, "Unmatched firmware version. Expected %s, got %s.\n", - LIQUIDIO_VERSION, h->version); + if (strncmp(LIQUIDIO_PACKAGE, h->version, strlen(LIQUIDIO_PACKAGE))) { + dev_err(&oct->pci_dev->dev, "Unmatched firmware package type. Expected %s, got %s.\n", + LIQUIDIO_PACKAGE, h->version); + return -EINVAL; + } + + base = h->version + strlen(LIQUIDIO_PACKAGE); + ret = memcmp(LIQUIDIO_BASE_VERSION, base, base_len); + if (ret) { + dev_err(&oct->pci_dev->dev, "Unmatched firmware version. Expected %s.x, got %s.\n", + LIQUIDIO_BASE_VERSION, base); return -EINVAL; } @@ -601,44 +610,44 @@ int octeon_download_firmware(struct octeon_device *oct, const u8 *data, snprintf(oct->fw_info.liquidio_firmware_version, 32, "LIQUIDIO: %s", h->version); - buffer = kmemdup(data, size, GFP_KERNEL); - if (!buffer) - return -ENOMEM; - - p = buffer + sizeof(struct octeon_firmware_file_header); + data += sizeof(struct octeon_firmware_file_header); + dev_info(&oct->pci_dev->dev, "%s: Loading %d images\n", __func__, + be32_to_cpu(h->num_images)); /* load all images */ for (i = 0; i < be32_to_cpu(h->num_images); i++) { load_addr = be64_to_cpu(h->desc[i].addr); image_len = be32_to_cpu(h->desc[i].len); - /* validate the image */ - crc32_result = crc32(~0, p, image_len) ^ ~0U; - if (crc32_result != be32_to_cpu(h->desc[i].crc32)) { - dev_err(&oct->pci_dev->dev, - "Firmware CRC mismatch in image %d (0x%08x != 0x%08x).\n", - i, crc32_result, - be32_to_cpu(h->desc[i].crc32)); - ret = -EINVAL; - goto done_downloading; - } + dev_info(&oct->pci_dev->dev, "Loading firmware %d at %llx\n", + image_len, load_addr); - /* download the image */ - octeon_pci_write_core_mem(oct, load_addr, p, image_len); + /* Write in 4MB chunks*/ + rem = image_len; - p += image_len; - dev_dbg(&oct->pci_dev->dev, - "Downloaded image %d (%d bytes) to address 0x%016llx\n", - i, image_len, load_addr); + while (rem) { + if (rem < (4 * 1024 * 1024)) + size = rem; + else + size = 4 * 1024 * 1024; + + memcpy(p, data, size); + + /* download the image */ + octeon_pci_write_core_mem(oct, load_addr, p, (u32)size); + + data += size; + rem -= (u32)size; + load_addr += size; + } } + dev_info(&oct->pci_dev->dev, "Writing boot command: %s\n", + h->bootcmd); /* Invoke the bootcmd */ ret = octeon_console_send_cmd(oct, h->bootcmd, 50); -done_downloading: - kfree(buffer); - - return ret; + return 0; } void octeon_free_device_mem(struct octeon_device *oct) -- cgit v0.10.2 From 60441888ec9cf5cdf58e43ecf93f403210cf3383 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:08 -0700 Subject: liquidio: New unload state This patch adds new state so that the ctrl packets are not sent to firmware during unload time and only rx packets are allowed. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 5fb1b79..4440086 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -1327,6 +1327,10 @@ static int liquidio_stop_nic_module(struct octeon_device *oct) return 1; } + spin_lock_bh(&oct->cmd_resp_wqlock); + oct->cmd_resp_state = OCT_DRV_OFFLINE; + spin_unlock_bh(&oct->cmd_resp_wqlock); + for (i = 0; i < oct->ifcount; i++) { lio = GET_LIO(oct->props[i].netdev); for (j = 0; j < lio->linfo.num_rxpciq; j++) diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h index abfc0d6..ceb905d 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h @@ -383,6 +383,10 @@ struct octeon_device { struct cavium_wq dma_comp_wq; + /** Lock for dma response list */ + spinlock_t cmd_resp_wqlock; + u32 cmd_resp_state; + struct cavium_wq check_db_wq[MAX_POSSIBLE_OCTEON_INSTR_QUEUES]; struct cavium_wk nic_poll_work; @@ -392,6 +396,8 @@ struct octeon_device { void *priv; }; +#define OCT_DRV_ONLINE 1 +#define OCT_DRV_OFFLINE 2 #define OCTEON_CN6XXX(oct) ((oct->chip_id == OCTEON_CN66XX) || \ (oct->chip_id == OCTEON_CN68XX)) #define CHIP_FIELD(oct, TYPE, field) \ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c index 7843b8a..36f1970 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c @@ -171,20 +171,36 @@ octnet_send_nic_ctrl_pkt(struct octeon_device *oct, int retval; struct octeon_soft_command *sc = NULL; + spin_lock_bh(&oct->cmd_resp_wqlock); + /* Allow only rx ctrl command to stop traffic on the chip + * during offline operations + */ + if ((oct->cmd_resp_state == OCT_DRV_OFFLINE) && + (nctrl->ncmd.s.cmd != OCTNET_CMD_RX_CTL)) { + spin_unlock_bh(&oct->cmd_resp_wqlock); + dev_err(&oct->pci_dev->dev, + "%s cmd:%d not processed since driver offline\n", + __func__, nctrl->ncmd.s.cmd); + return -1; + } + sc = octnic_alloc_ctrl_pkt_sc(oct, nctrl); if (!sc) { dev_err(&oct->pci_dev->dev, "%s soft command alloc failed\n", __func__); + spin_unlock_bh(&oct->cmd_resp_wqlock); return -1; } retval = octeon_send_soft_command(oct, sc); if (retval == IQ_SEND_FAILED) { octeon_free_soft_command(oct, sc); - dev_err(&oct->pci_dev->dev, "%s soft command send failed status: %x\n", - __func__, retval); + dev_err(&oct->pci_dev->dev, "%s soft command:%d send failed status: %x\n", + __func__, nctrl->ncmd.s.cmd, retval); + spin_unlock_bh(&oct->cmd_resp_wqlock); return -1; } + spin_unlock_bh(&oct->cmd_resp_wqlock); return retval; } diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.c b/drivers/net/ethernet/cavium/liquidio/response_manager.c index e2e9103..c93210f 100644 --- a/drivers/net/ethernet/cavium/liquidio/response_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/response_manager.c @@ -54,6 +54,7 @@ int octeon_setup_response_list(struct octeon_device *oct) spin_lock_init(&oct->response_list[i].lock); atomic_set(&oct->response_list[i].pending_req_count, 0); } + spin_lock_init(&oct->cmd_resp_wqlock); oct->dma_comp_wq.wq = alloc_workqueue("dma-comp", WQ_MEM_RECLAIM, 0); if (!oct->dma_comp_wq.wq) { @@ -64,6 +65,7 @@ int octeon_setup_response_list(struct octeon_device *oct) cwq = &oct->dma_comp_wq; INIT_DELAYED_WORK(&cwq->wk.work, oct_poll_req_completion); cwq->wk.ctxptr = oct; + oct->cmd_resp_state = OCT_DRV_ONLINE; queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(100)); return ret; -- cgit v0.10.2 From 60b48c5a83eb1a72e514641b8c3939f1795198f9 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:09 -0700 Subject: liquidio: chip reset changes This patch resolves the order of chip reset while destroying the resources by postoponing soft reset in destroy resources function until all queues are removed properly. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 4440086..56b1d67 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -1180,12 +1180,6 @@ static void octeon_destroy_resources(struct octeon_device *oct) if (oct->flags & LIO_FLAG_MSI_ENABLED) pci_disable_msi(oct->pci_dev); - /* Soft reset the octeon device before exiting */ - oct->fn_list.soft_reset(oct); - - /* Disable the device, releasing the PCI INT */ - pci_disable_device(oct->pci_dev); - /* fallthrough */ case OCT_DEV_IN_RESET: case OCT_DEV_DROQ_INIT_DONE: @@ -1232,11 +1226,18 @@ static void octeon_destroy_resources(struct octeon_device *oct) /* fallthrough */ case OCT_DEV_PCI_MAP_DONE: + + /* Soft reset the octeon device before exiting */ + oct->fn_list.soft_reset(oct); + octeon_unmap_pci_barx(oct, 0); octeon_unmap_pci_barx(oct, 1); /* fallthrough */ case OCT_DEV_BEGIN_STATE: + /* Disable the device, releasing the PCI INT */ + pci_disable_device(oct->pci_dev); + /* Nothing to be done here either */ break; } /* end switch(oct->status) */ -- cgit v0.10.2 From 78e6a9b4a43b2e45c5d0f26d487b2890ffb0f0b5 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:10 -0700 Subject: liquidio: tx rx interrupt moderation This patch has new tx/rx interrupt moderation defaults of count/timer for better throughput and utilisation. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 56f465b..9c6b58a 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -616,50 +616,50 @@ static int lio_get_intr_coalesce(struct net_device *netdev, { struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; - struct octeon_cn6xxx *cn6xxx = (struct octeon_cn6xxx *)oct->chip; struct octeon_instr_queue *iq; struct oct_intrmod_cfg *intrmod_cfg; intrmod_cfg = &oct->intrmod; switch (oct->chip_id) { - /* case OCTEON_CN73XX: Todo */ - /* break; */ case OCTEON_CN68XX: - case OCTEON_CN66XX: - if (!intrmod_cfg->intrmod_enable) { + case OCTEON_CN66XX: { + struct octeon_cn6xxx *cn6xxx = + (struct octeon_cn6xxx *)oct->chip; + + if (!intrmod_cfg->rx_enable) { intr_coal->rx_coalesce_usecs = CFG_GET_OQ_INTR_TIME(cn6xxx->conf); intr_coal->rx_max_coalesced_frames = CFG_GET_OQ_INTR_PKT(cn6xxx->conf); - } else { - intr_coal->use_adaptive_rx_coalesce = - intrmod_cfg->intrmod_enable; - intr_coal->rate_sample_interval = - intrmod_cfg->intrmod_check_intrvl; - intr_coal->pkt_rate_high = - intrmod_cfg->intrmod_maxpkt_ratethr; - intr_coal->pkt_rate_low = - intrmod_cfg->intrmod_minpkt_ratethr; - intr_coal->rx_max_coalesced_frames_high = - intrmod_cfg->intrmod_maxcnt_trigger; - intr_coal->rx_coalesce_usecs_high = - intrmod_cfg->intrmod_maxtmr_trigger; - intr_coal->rx_coalesce_usecs_low = - intrmod_cfg->intrmod_mintmr_trigger; - intr_coal->rx_max_coalesced_frames_low = - intrmod_cfg->intrmod_mincnt_trigger; } iq = oct->instr_queue[lio->linfo.txpciq[0].s.q_no]; intr_coal->tx_max_coalesced_frames = iq->fill_threshold; break; - + } default: netif_info(lio, drv, lio->netdev, "Unknown Chip !!\n"); return -EINVAL; } - + if (intrmod_cfg->rx_enable) { + intr_coal->use_adaptive_rx_coalesce = + intrmod_cfg->rx_enable; + intr_coal->rate_sample_interval = + intrmod_cfg->check_intrvl; + intr_coal->pkt_rate_high = + intrmod_cfg->maxpkt_ratethr; + intr_coal->pkt_rate_low = + intrmod_cfg->minpkt_ratethr; + intr_coal->rx_max_coalesced_frames_high = + intrmod_cfg->rx_maxcnt_trigger; + intr_coal->rx_coalesce_usecs_high = + intrmod_cfg->rx_maxtmr_trigger; + intr_coal->rx_coalesce_usecs_low = + intrmod_cfg->rx_mintmr_trigger; + intr_coal->rx_max_coalesced_frames_low = + intrmod_cfg->rx_mincnt_trigger; + } return 0; } @@ -679,19 +679,20 @@ static void octnet_intrmod_callback(struct octeon_device *oct_dev, else dev_info(&oct_dev->pci_dev->dev, "Rx-Adaptive Interrupt moderation enabled:%llx\n", - oct_dev->intrmod.intrmod_enable); + oct_dev->intrmod.rx_enable); octeon_free_soft_command(oct_dev, sc); } /* Configure interrupt moderation parameters */ -static int octnet_set_intrmod_cfg(void *oct, struct oct_intrmod_cfg *intr_cfg) +static int octnet_set_intrmod_cfg(struct lio *lio, + struct oct_intrmod_cfg *intr_cfg) { struct octeon_soft_command *sc; struct oct_intrmod_cmd *cmd; struct oct_intrmod_cfg *cfg; int retval; - struct octeon_device *oct_dev = (struct octeon_device *)oct; + struct octeon_device *oct_dev = lio->oct_dev; /* Alloc soft command */ sc = (struct octeon_soft_command *) @@ -712,6 +713,8 @@ static int octnet_set_intrmod_cfg(void *oct, struct oct_intrmod_cfg *intr_cfg) cmd->cfg = cfg; cmd->oct_dev = oct_dev; + sc->iq_no = lio->linfo.txpciq[0].s.q_no; + octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC, OPCODE_NIC_INTRMOD_CFG, 0, 0, 0); @@ -730,7 +733,7 @@ static int octnet_set_intrmod_cfg(void *oct, struct oct_intrmod_cfg *intr_cfg) /* Enable/Disable auto interrupt Moderation */ static int oct_cfg_adaptive_intr(struct lio *lio, struct ethtool_coalesce - *intr_coal, int adaptive) + *intr_coal) { int ret = 0; struct octeon_device *oct = lio->oct_dev; @@ -738,59 +741,73 @@ static int oct_cfg_adaptive_intr(struct lio *lio, struct ethtool_coalesce intrmod_cfg = &oct->intrmod; - if (adaptive) { + if (oct->intrmod.rx_enable || oct->intrmod.tx_enable) { if (intr_coal->rate_sample_interval) - intrmod_cfg->intrmod_check_intrvl = + intrmod_cfg->check_intrvl = intr_coal->rate_sample_interval; else - intrmod_cfg->intrmod_check_intrvl = + intrmod_cfg->check_intrvl = LIO_INTRMOD_CHECK_INTERVAL; if (intr_coal->pkt_rate_high) - intrmod_cfg->intrmod_maxpkt_ratethr = + intrmod_cfg->maxpkt_ratethr = intr_coal->pkt_rate_high; else - intrmod_cfg->intrmod_maxpkt_ratethr = + intrmod_cfg->maxpkt_ratethr = LIO_INTRMOD_MAXPKT_RATETHR; if (intr_coal->pkt_rate_low) - intrmod_cfg->intrmod_minpkt_ratethr = + intrmod_cfg->minpkt_ratethr = intr_coal->pkt_rate_low; else - intrmod_cfg->intrmod_minpkt_ratethr = + intrmod_cfg->minpkt_ratethr = LIO_INTRMOD_MINPKT_RATETHR; - + } + if (oct->intrmod.rx_enable) { if (intr_coal->rx_max_coalesced_frames_high) - intrmod_cfg->intrmod_maxcnt_trigger = + intrmod_cfg->rx_maxcnt_trigger = intr_coal->rx_max_coalesced_frames_high; else - intrmod_cfg->intrmod_maxcnt_trigger = - LIO_INTRMOD_MAXCNT_TRIGGER; + intrmod_cfg->rx_maxcnt_trigger = + LIO_INTRMOD_RXMAXCNT_TRIGGER; if (intr_coal->rx_coalesce_usecs_high) - intrmod_cfg->intrmod_maxtmr_trigger = + intrmod_cfg->rx_maxtmr_trigger = intr_coal->rx_coalesce_usecs_high; else - intrmod_cfg->intrmod_maxtmr_trigger = - LIO_INTRMOD_MAXTMR_TRIGGER; + intrmod_cfg->rx_maxtmr_trigger = + LIO_INTRMOD_RXMAXTMR_TRIGGER; if (intr_coal->rx_coalesce_usecs_low) - intrmod_cfg->intrmod_mintmr_trigger = + intrmod_cfg->rx_mintmr_trigger = intr_coal->rx_coalesce_usecs_low; else - intrmod_cfg->intrmod_mintmr_trigger = - LIO_INTRMOD_MINTMR_TRIGGER; + intrmod_cfg->rx_mintmr_trigger = + LIO_INTRMOD_RXMINTMR_TRIGGER; if (intr_coal->rx_max_coalesced_frames_low) - intrmod_cfg->intrmod_mincnt_trigger = + intrmod_cfg->rx_mincnt_trigger = intr_coal->rx_max_coalesced_frames_low; else - intrmod_cfg->intrmod_mincnt_trigger = - LIO_INTRMOD_MINCNT_TRIGGER; + intrmod_cfg->rx_mincnt_trigger = + LIO_INTRMOD_RXMINCNT_TRIGGER; + } + if (oct->intrmod.tx_enable) { + if (intr_coal->tx_max_coalesced_frames_high) + intrmod_cfg->tx_maxcnt_trigger = + intr_coal->tx_max_coalesced_frames_high; + else + intrmod_cfg->tx_maxcnt_trigger = + LIO_INTRMOD_TXMAXCNT_TRIGGER; + if (intr_coal->tx_max_coalesced_frames_low) + intrmod_cfg->tx_mincnt_trigger = + intr_coal->tx_max_coalesced_frames_low; + else + intrmod_cfg->tx_mincnt_trigger = + LIO_INTRMOD_TXMINCNT_TRIGGER; } - intrmod_cfg->intrmod_enable = adaptive; - ret = octnet_set_intrmod_cfg(oct, intrmod_cfg); + ret = octnet_set_intrmod_cfg(lio, intrmod_cfg); return ret; } @@ -798,54 +815,82 @@ static int oct_cfg_adaptive_intr(struct lio *lio, struct ethtool_coalesce static int oct_cfg_rx_intrcnt(struct lio *lio, struct ethtool_coalesce *intr_coal) { - int ret; struct octeon_device *oct = lio->oct_dev; - struct octeon_cn6xxx *cn6xxx = (struct octeon_cn6xxx *)oct->chip; u32 rx_max_coalesced_frames; - if (!intr_coal->rx_max_coalesced_frames) - rx_max_coalesced_frames = CN6XXX_OQ_INTR_PKT; - else - rx_max_coalesced_frames = intr_coal->rx_max_coalesced_frames; - - /* Disable adaptive interrupt modulation */ - ret = oct_cfg_adaptive_intr(lio, intr_coal, 0); - if (ret) - return ret; - /* Config Cnt based interrupt values */ - octeon_write_csr(oct, CN6XXX_SLI_OQ_INT_LEVEL_PKTS, - rx_max_coalesced_frames); - CFG_SET_OQ_INTR_PKT(cn6xxx->conf, rx_max_coalesced_frames); + switch (oct->chip_id) { + case OCTEON_CN68XX: + case OCTEON_CN66XX: { + struct octeon_cn6xxx *cn6xxx = + (struct octeon_cn6xxx *)oct->chip; + + if (!intr_coal->rx_max_coalesced_frames) + rx_max_coalesced_frames = CN6XXX_OQ_INTR_PKT; + else + rx_max_coalesced_frames = + intr_coal->rx_max_coalesced_frames; + octeon_write_csr(oct, CN6XXX_SLI_OQ_INT_LEVEL_PKTS, + rx_max_coalesced_frames); + CFG_SET_OQ_INTR_PKT(cn6xxx->conf, rx_max_coalesced_frames); + break; + } + default: + return -EINVAL; + } return 0; } static int oct_cfg_rx_intrtime(struct lio *lio, struct ethtool_coalesce *intr_coal) { - int ret; struct octeon_device *oct = lio->oct_dev; - struct octeon_cn6xxx *cn6xxx = (struct octeon_cn6xxx *)oct->chip; u32 time_threshold, rx_coalesce_usecs; - if (!intr_coal->rx_coalesce_usecs) - rx_coalesce_usecs = CN6XXX_OQ_INTR_TIME; - else - rx_coalesce_usecs = intr_coal->rx_coalesce_usecs; + /* Config Time based interrupt values */ + switch (oct->chip_id) { + case OCTEON_CN68XX: + case OCTEON_CN66XX: { + struct octeon_cn6xxx *cn6xxx = + (struct octeon_cn6xxx *)oct->chip; + if (!intr_coal->rx_coalesce_usecs) + rx_coalesce_usecs = CN6XXX_OQ_INTR_TIME; + else + rx_coalesce_usecs = intr_coal->rx_coalesce_usecs; - /* Disable adaptive interrupt modulation */ - ret = oct_cfg_adaptive_intr(lio, intr_coal, 0); - if (ret) - return ret; + time_threshold = lio_cn6xxx_get_oq_ticks(oct, + rx_coalesce_usecs); + octeon_write_csr(oct, + CN6XXX_SLI_OQ_INT_LEVEL_TIME, + time_threshold); - /* Config Time based interrupt values */ - time_threshold = lio_cn6xxx_get_oq_ticks(oct, rx_coalesce_usecs); - octeon_write_csr(oct, CN6XXX_SLI_OQ_INT_LEVEL_TIME, time_threshold); - CFG_SET_OQ_INTR_TIME(cn6xxx->conf, rx_coalesce_usecs); + CFG_SET_OQ_INTR_TIME(cn6xxx->conf, rx_coalesce_usecs); + break; + } + default: + return -EINVAL; + } return 0; } +static int +oct_cfg_tx_intrcnt(struct lio *lio, struct ethtool_coalesce *intr_coal + __attribute__((unused))) +{ + struct octeon_device *oct = lio->oct_dev; + + /* Config Cnt based interrupt values */ + switch (oct->chip_id) { + case OCTEON_CN68XX: + case OCTEON_CN66XX: + break; + default: + return -EINVAL; + } + return 0; +} + static int lio_set_intr_coalesce(struct net_device *netdev, struct ethtool_coalesce *intr_coal) { @@ -853,59 +898,48 @@ static int lio_set_intr_coalesce(struct net_device *netdev, int ret; struct octeon_device *oct = lio->oct_dev; u32 j, q_no; + int db_max, db_min; - if ((intr_coal->tx_max_coalesced_frames >= CN6XXX_DB_MIN) && - (intr_coal->tx_max_coalesced_frames <= CN6XXX_DB_MAX)) { - for (j = 0; j < lio->linfo.num_txpciq; j++) { - q_no = lio->linfo.txpciq[j].s.q_no; - oct->instr_queue[q_no]->fill_threshold = - intr_coal->tx_max_coalesced_frames; + switch (oct->chip_id) { + case OCTEON_CN68XX: + case OCTEON_CN66XX: + db_min = CN6XXX_DB_MIN; + db_max = CN6XXX_DB_MAX; + if ((intr_coal->tx_max_coalesced_frames >= db_min) && + (intr_coal->tx_max_coalesced_frames <= db_max)) { + for (j = 0; j < lio->linfo.num_txpciq; j++) { + q_no = lio->linfo.txpciq[j].s.q_no; + oct->instr_queue[q_no]->fill_threshold = + intr_coal->tx_max_coalesced_frames; + } + } else { + dev_err(&oct->pci_dev->dev, + "LIQUIDIO: Invalid tx-frames:%d. Range is min:%d max:%d\n", + intr_coal->tx_max_coalesced_frames, db_min, + db_max); + return -EINVAL; } - } else { - dev_err(&oct->pci_dev->dev, - "LIQUIDIO: Invalid tx-frames:%d. Range is min:%d max:%d\n", - intr_coal->tx_max_coalesced_frames, CN6XXX_DB_MIN, - CN6XXX_DB_MAX); + break; + default: return -EINVAL; } - /* User requested adaptive-rx on */ - if (intr_coal->use_adaptive_rx_coalesce) { - ret = oct_cfg_adaptive_intr(lio, intr_coal, 1); - if (ret) - goto ret_intrmod; - } + oct->intrmod.rx_enable = intr_coal->use_adaptive_rx_coalesce ? 1 : 0; + oct->intrmod.tx_enable = intr_coal->use_adaptive_tx_coalesce ? 1 : 0; + + ret = oct_cfg_adaptive_intr(lio, intr_coal); - /* User requested adaptive-rx off and rx coalesce */ - if ((intr_coal->rx_coalesce_usecs) && - (!intr_coal->use_adaptive_rx_coalesce)) { + if (!intr_coal->use_adaptive_rx_coalesce) { ret = oct_cfg_rx_intrtime(lio, intr_coal); if (ret) goto ret_intrmod; - } - /* User requested adaptive-rx off and rx coalesce */ - if ((intr_coal->rx_max_coalesced_frames) && - (!intr_coal->use_adaptive_rx_coalesce)) { ret = oct_cfg_rx_intrcnt(lio, intr_coal); if (ret) goto ret_intrmod; } - - /* User requested adaptive-rx off, so use default coalesce params */ - if ((!intr_coal->rx_max_coalesced_frames) && - (!intr_coal->use_adaptive_rx_coalesce) && - (!intr_coal->rx_coalesce_usecs)) { - dev_info(&oct->pci_dev->dev, - "Turning off adaptive-rx interrupt moderation\n"); - dev_info(&oct->pci_dev->dev, - "Using RX Coalesce Default values rx_coalesce_usecs:%d rx_max_coalesced_frames:%d\n", - CN6XXX_OQ_INTR_TIME, CN6XXX_OQ_INTR_PKT); - ret = oct_cfg_rx_intrtime(lio, intr_coal); - if (ret) - goto ret_intrmod; - - ret = oct_cfg_rx_intrcnt(lio, intr_coal); + if (!intr_coal->use_adaptive_tx_coalesce) { + ret = oct_cfg_tx_intrcnt(lio, intr_coal); if (ret) goto ret_intrmod; } diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 56b1d67..a1e31ac 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -3579,15 +3579,19 @@ static int liquidio_init_nic_module(struct octeon_device *oct) /* Initialize interrupt moderation params */ intrmod_cfg = &((struct octeon_device *)oct)->intrmod; - intrmod_cfg->intrmod_enable = 1; - intrmod_cfg->intrmod_check_intrvl = LIO_INTRMOD_CHECK_INTERVAL; - intrmod_cfg->intrmod_maxpkt_ratethr = LIO_INTRMOD_MAXPKT_RATETHR; - intrmod_cfg->intrmod_minpkt_ratethr = LIO_INTRMOD_MINPKT_RATETHR; - intrmod_cfg->intrmod_maxcnt_trigger = LIO_INTRMOD_MAXCNT_TRIGGER; - intrmod_cfg->intrmod_maxtmr_trigger = LIO_INTRMOD_MAXTMR_TRIGGER; - intrmod_cfg->intrmod_mintmr_trigger = LIO_INTRMOD_MINTMR_TRIGGER; - intrmod_cfg->intrmod_mincnt_trigger = LIO_INTRMOD_MINCNT_TRIGGER; - + intrmod_cfg->rx_enable = 1; + intrmod_cfg->check_intrvl = LIO_INTRMOD_CHECK_INTERVAL; + intrmod_cfg->maxpkt_ratethr = LIO_INTRMOD_MAXPKT_RATETHR; + intrmod_cfg->minpkt_ratethr = LIO_INTRMOD_MINPKT_RATETHR; + intrmod_cfg->rx_maxcnt_trigger = LIO_INTRMOD_RXMAXCNT_TRIGGER; + intrmod_cfg->rx_maxtmr_trigger = LIO_INTRMOD_RXMAXTMR_TRIGGER; + intrmod_cfg->rx_mintmr_trigger = LIO_INTRMOD_RXMINTMR_TRIGGER; + intrmod_cfg->rx_mincnt_trigger = LIO_INTRMOD_RXMINCNT_TRIGGER; + intrmod_cfg->tx_enable = 1; + intrmod_cfg->tx_maxcnt_trigger = LIO_INTRMOD_TXMAXCNT_TRIGGER; + intrmod_cfg->tx_mincnt_trigger = LIO_INTRMOD_TXMINCNT_TRIGGER; + intrmod_cfg->rx_frames = CFG_GET_OQ_INTR_PKT(octeon_get_conf(oct)); + intrmod_cfg->rx_usecs = CFG_GET_OQ_INTR_TIME(octeon_get_conf(oct)); dev_dbg(&oct->pci_dev->dev, "Network interfaces ready\n"); return retval; diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 1ef9001..b5ab704 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -796,23 +796,44 @@ struct oct_mdio_cmd { #define OCT_LINK_STATS_SIZE (sizeof(struct oct_link_stats)) +/* intrmod: max. packet rate threshold */ +#define LIO_INTRMOD_MAXPKT_RATETHR 196608 +/* intrmod: min. packet rate threshold */ +#define LIO_INTRMOD_MINPKT_RATETHR 9216 +/* intrmod: max. packets to trigger interrupt */ +#define LIO_INTRMOD_RXMAXCNT_TRIGGER 384 +/* intrmod: min. packets to trigger interrupt */ +#define LIO_INTRMOD_RXMINCNT_TRIGGER 1 +/* intrmod: max. time to trigger interrupt */ +#define LIO_INTRMOD_RXMAXTMR_TRIGGER 128 +/* 66xx:intrmod: min. time to trigger interrupt + * (value of 1 is optimum for TCP_RR) + */ +#define LIO_INTRMOD_RXMINTMR_TRIGGER 1 + +/* intrmod: max. packets to trigger interrupt */ +#define LIO_INTRMOD_TXMAXCNT_TRIGGER 64 +/* intrmod: min. packets to trigger interrupt */ +#define LIO_INTRMOD_TXMINCNT_TRIGGER 0 + +/* intrmod: poll interval in seconds */ #define LIO_INTRMOD_CHECK_INTERVAL 1 -#define LIO_INTRMOD_MAXPKT_RATETHR 196608 /* max pkt rate threshold */ -#define LIO_INTRMOD_MINPKT_RATETHR 9216 /* min pkt rate threshold */ -#define LIO_INTRMOD_MAXCNT_TRIGGER 384 /* max pkts to trigger interrupt */ -#define LIO_INTRMOD_MINCNT_TRIGGER 1 /* min pkts to trigger interrupt */ -#define LIO_INTRMOD_MAXTMR_TRIGGER 128 /* max time to trigger interrupt */ -#define LIO_INTRMOD_MINTMR_TRIGGER 32 /* min time to trigger interrupt */ struct oct_intrmod_cfg { - u64 intrmod_enable; - u64 intrmod_check_intrvl; - u64 intrmod_maxpkt_ratethr; - u64 intrmod_minpkt_ratethr; - u64 intrmod_maxcnt_trigger; - u64 intrmod_maxtmr_trigger; - u64 intrmod_mincnt_trigger; - u64 intrmod_mintmr_trigger; + u64 rx_enable; + u64 tx_enable; + u64 check_intrvl; + u64 maxpkt_ratethr; + u64 minpkt_ratethr; + u64 rx_maxcnt_trigger; + u64 rx_mincnt_trigger; + u64 rx_maxtmr_trigger; + u64 rx_mintmr_trigger; + u64 tx_mincnt_trigger; + u64 tx_maxcnt_trigger; + u64 rx_frames; + u64 tx_frames; + u64 rx_usecs; }; #define BASE_QUEUE_NOT_REQUESTED 65535 -- cgit v0.10.2 From 1f164717cb1ee722dc58dc32bbcb01ac506d254f Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:11 -0700 Subject: liquidio: New statistics support This patch adds extensive support of statistics for data path, control path and firmware. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 9c6b58a..2b03095 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -40,6 +40,8 @@ #include "cn68xx_device.h" #include "liquidio_image.h" +static int octnet_get_link_stats(struct net_device *netdev); + struct oct_mdio_cmd_context { int octeon_id; wait_queue_head_t wc; @@ -77,28 +79,109 @@ enum { #define OCT_ETHTOOL_REGDUMP_LEN 4096 #define OCT_ETHTOOL_REGSVER 1 +/* statistics of PF */ +static const char oct_stats_strings[][ETH_GSTRING_LEN] = { + "rx_packets", + "tx_packets", + "rx_bytes", + "tx_bytes", + "rx_errors", /*jabber_err+l2_err+frame_err */ + "tx_errors", /*fw_err_pko+fw_err_link+fw_err_drop */ + "rx_dropped", /*st->fromwire.total_rcvd - st->fromwire.fw_total_rcvd + *+st->fromwire.dmac_drop + st->fromwire.fw_err_drop + */ + "tx_dropped", + + "tx_total_sent", + "tx_total_fwd", + "tx_err_pko", + "tx_err_link", + "tx_err_drop", + + "tx_tso", + "tx_tso_packets", + "tx_tso_err", + + "mac_tx_total_pkts", + "mac_tx_total_bytes", + "mac_tx_mcast_pkts", + "mac_tx_bcast_pkts", + "mac_tx_ctl_packets", /*oct->link_stats.fromhost.ctl_sent */ + "mac_tx_total_collisions", + "mac_tx_one_collision", + "mac_tx_multi_collison", + "mac_tx_max_collision_fail", + "mac_tx_max_deferal_fail", + "mac_tx_fifo_err", + "mac_tx_runts", + + "rx_total_rcvd", + "rx_total_fwd", + "rx_jabber_err", + "rx_l2_err", + "rx_frame_err", + "rx_err_pko", + "rx_err_link", + "rx_err_drop", + + "rx_lro_pkts", + "rx_lro_bytes", + "rx_total_lro", + + "rx_lro_aborts", + "rx_lro_aborts_port", + "rx_lro_aborts_seq", + "rx_lro_aborts_tsval", + "rx_lro_aborts_timer", + "rx_fwd_rate", + + "mac_rx_total_rcvd", + "mac_rx_bytes", + "mac_rx_total_bcst", + "mac_rx_total_mcst", + "mac_rx_runts", + "mac_rx_ctl_packets", + "mac_rx_fifo_err", + "mac_rx_dma_drop", + "mac_rx_fcs_err", + + "link_state_changes", +}; + +/* statistics of host tx queue */ static const char oct_iq_stats_strings[][ETH_GSTRING_LEN] = { - "Instr posted", - "Instr processed", - "Instr dropped", - "Bytes Sent", - "Sgentry_sent", - "Inst cntreg", - "Tx done", - "Tx Iq busy", - "Tx dropped", - "Tx bytes", + "packets", /*oct->instr_queue[iq_no]->stats.tx_done*/ + "bytes", /*oct->instr_queue[iq_no]->stats.tx_tot_bytes*/ + "dropped", + "iq_busy", + "sgentry_sent", + + "fw_instr_posted", + "fw_instr_processed", + "fw_instr_dropped", + "fw_bytes_sent", + + "tso", + "txq_restart", }; +/* statistics of host rx queue */ static const char oct_droq_stats_strings[][ETH_GSTRING_LEN] = { - "OQ Pkts Received", - "OQ Bytes Received", - "Dropped no dispatch", - "Dropped nomem", - "Dropped toomany", - "Stack RX cnt", - "Stack RX Bytes", - "RX dropped", + "packets", /*oct->droq[oq_no]->stats.rx_pkts_received */ + "bytes", /*oct->droq[oq_no]->stats.rx_bytes_received */ + "dropped", /*oct->droq[oq_no]->stats.rx_dropped+ + *oct->droq[oq_no]->stats.dropped_nodispatch+ + *oct->droq[oq_no]->stats.dropped_toomany+ + *oct->droq[oq_no]->stats.dropped_nomem + */ + "dropped_nomem", + "dropped_toomany", + "fw_dropped", + "fw_pkts_received", + "fw_bytes_received", + "fw_dropped_nodispatch", + + "buffer_alloc_failure", }; #define OCTNIC_NCMD_AUTONEG_ON 0x1 @@ -516,8 +599,13 @@ lio_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) /* Notes: Not supporting any auto negotiation in these * drivers. Just report pause frame support. */ - pause->tx_pause = 1; - pause->rx_pause = 1; /* TODO: Need to support RX pause frame!!. */ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; + + pause->autoneg = 0; + + pause->tx_pause = oct->tx_pause; + pause->rx_pause = oct->rx_pause; } static void @@ -526,51 +614,245 @@ lio_get_ethtool_stats(struct net_device *netdev, { struct lio *lio = GET_LIO(netdev); struct octeon_device *oct_dev = lio->oct_dev; + struct net_device_stats *netstats = &netdev->stats; int i = 0, j; - for (j = 0; j < MAX_OCTEON_INSTR_QUEUES(oct); j++) { + netdev->netdev_ops->ndo_get_stats(netdev); + octnet_get_link_stats(netdev); + + /*sum of oct->droq[oq_no]->stats->rx_pkts_received */ + data[i++] = CVM_CAST64(netstats->rx_packets); + /*sum of oct->instr_queue[iq_no]->stats.tx_done */ + data[i++] = CVM_CAST64(netstats->tx_packets); + /*sum of oct->droq[oq_no]->stats->rx_bytes_received */ + data[i++] = CVM_CAST64(netstats->rx_bytes); + /*sum of oct->instr_queue[iq_no]->stats.tx_tot_bytes */ + data[i++] = CVM_CAST64(netstats->tx_bytes); + data[i++] = CVM_CAST64(netstats->rx_errors); + data[i++] = CVM_CAST64(netstats->tx_errors); + /*sum of oct->droq[oq_no]->stats->rx_dropped + + *oct->droq[oq_no]->stats->dropped_nodispatch + + *oct->droq[oq_no]->stats->dropped_toomany + + *oct->droq[oq_no]->stats->dropped_nomem + */ + data[i++] = CVM_CAST64(netstats->rx_dropped); + /*sum of oct->instr_queue[iq_no]->stats.tx_dropped */ + data[i++] = CVM_CAST64(netstats->tx_dropped); + + /*data[i++] = CVM_CAST64(stats->multicast); */ + /*data[i++] = CVM_CAST64(stats->collisions); */ + + /* firmware tx stats */ + /*per_core_stats[cvmx_get_core_num()].link_stats[mdata->from_ifidx]. + *fromhost.fw_total_sent + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_total_sent); + /*per_core_stats[i].link_stats[port].fromwire.fw_total_fwd */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_total_fwd); + /*per_core_stats[j].link_stats[i].fromhost.fw_err_pko */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_pko); + /*per_core_stats[j].link_stats[i].fromhost.fw_err_link */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_link); + /*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost. + *fw_err_drop + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_drop); + + /*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost.fw_tso */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_tso); + /*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost. + *fw_tso_fwd + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_tso_fwd); + /*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost. + *fw_err_tso + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_tso); + + /* mac tx statistics */ + /*CVMX_BGXX_CMRX_TX_STAT5 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.total_pkts_sent); + /*CVMX_BGXX_CMRX_TX_STAT4 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.total_bytes_sent); + /*CVMX_BGXX_CMRX_TX_STAT15 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.mcast_pkts_sent); + /*CVMX_BGXX_CMRX_TX_STAT14 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.bcast_pkts_sent); + /*CVMX_BGXX_CMRX_TX_STAT17 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.ctl_sent); + /*CVMX_BGXX_CMRX_TX_STAT0 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.total_collisions); + /*CVMX_BGXX_CMRX_TX_STAT3 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.one_collision_sent); + /*CVMX_BGXX_CMRX_TX_STAT2 */ + data[i++] = + CVM_CAST64(oct_dev->link_stats.fromhost.multi_collision_sent); + /*CVMX_BGXX_CMRX_TX_STAT0 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.max_collision_fail); + /*CVMX_BGXX_CMRX_TX_STAT1 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.max_deferral_fail); + /*CVMX_BGXX_CMRX_TX_STAT16 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fifo_err); + /*CVMX_BGXX_CMRX_TX_STAT6 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.runts); + + /* RX firmware stats */ + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_total_rcvd + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_total_rcvd); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_total_fwd + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_total_fwd); + /*per_core_stats[core_id].link_stats[ifidx].fromwire.jabber_err */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.jabber_err); + /*per_core_stats[core_id].link_stats[ifidx].fromwire.l2_err */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.l2_err); + /*per_core_stats[core_id].link_stats[ifidx].fromwire.frame_err */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.frame_err); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_err_pko + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_err_pko); + /*per_core_stats[j].link_stats[i].fromwire.fw_err_link */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_err_link); + /*per_core_stats[cvmx_get_core_num()].link_stats[lro_ctx->ifidx]. + *fromwire.fw_err_drop + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_err_drop); + + /* LRO */ + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_pkts + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_pkts); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_octs + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_octs); + /*per_core_stats[j].link_stats[i].fromwire.fw_total_lro */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_total_lro); + /*per_core_stats[j].link_stats[i].fromwire.fw_lro_aborts */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_aborts); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_aborts_port + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_aborts_port); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_aborts_seq + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_aborts_seq); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_aborts_tsval + */ + data[i++] = + CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_aborts_tsval); + /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. + *fw_lro_aborts_timer + */ + /* intrmod: packet forward rate */ + data[i++] = + CVM_CAST64(oct_dev->link_stats.fromwire.fw_lro_aborts_timer); + /*per_core_stats[j].link_stats[i].fromwire.fw_lro_aborts */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fwd_rate); + + /* mac: link-level stats */ + /*CVMX_BGXX_CMRX_RX_STAT0 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.total_rcvd); + /*CVMX_BGXX_CMRX_RX_STAT1 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.bytes_rcvd); + /*CVMX_PKI_STATX_STAT5 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.total_bcst); + /*CVMX_PKI_STATX_STAT5 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.total_mcst); + /*wqe->word2.err_code or wqe->word2.err_level */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.runts); + /*CVMX_BGXX_CMRX_RX_STAT2 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.ctl_rcvd); + /*CVMX_BGXX_CMRX_RX_STAT6 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fifo_err); + /*CVMX_BGXX_CMRX_RX_STAT4 */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.dmac_drop); + /*wqe->word2.err_code or wqe->word2.err_level */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fcs_err); + /*lio->link_changes*/ + data[i++] = CVM_CAST64(lio->link_changes); + + /* TX -- lio_update_stats(lio); */ + for (j = 0; j < MAX_OCTEON_INSTR_QUEUES(oct_dev); j++) { if (!(oct_dev->io_qmask.iq & (1ULL << j))) continue; + /*packets to network port*/ + /*# of packets tx to network */ + data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_done); + /*# of bytes tx to network */ data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.instr_posted); - data[i++] = - CVM_CAST64( - oct_dev->instr_queue[j]->stats.instr_processed); + CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_tot_bytes); + /*# of packets dropped */ data[i++] = - CVM_CAST64( - oct_dev->instr_queue[j]->stats.instr_dropped); + CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_dropped); + /*# of tx fails due to queue full */ data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.bytes_sent); + CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_iq_busy); + /*XXX gather entries sent */ data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.sgentry_sent); + + /*instruction to firmware: data and control */ + /*# of instructions to the queue */ data[i++] = - readl(oct_dev->instr_queue[j]->inst_cnt_reg); - data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_done); - data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_iq_busy); + CVM_CAST64(oct_dev->instr_queue[j]->stats.instr_posted); + /*# of instructions processed */ + data[i++] = CVM_CAST64(oct_dev->instr_queue[j]-> + stats.instr_processed); + /*# of instructions could not be processed */ + data[i++] = CVM_CAST64(oct_dev->instr_queue[j]-> + stats.instr_dropped); + /*bytes sent through the queue */ data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_dropped); + CVM_CAST64(oct_dev->instr_queue[j]->stats.bytes_sent); + + /*tso request*/ + data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_gso); + /*txq restart*/ data[i++] = - CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_tot_bytes); + CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_restart); } - /* for (j = 0; j < oct_dev->num_oqs; j++){ */ - for (j = 0; j < MAX_OCTEON_OUTPUT_QUEUES(oct); j++) { + /* RX */ + /* for (j = 0; j < oct_dev->num_oqs; j++) { */ + for (j = 0; j < MAX_OCTEON_OUTPUT_QUEUES(oct_dev); j++) { if (!(oct_dev->io_qmask.oq & (1ULL << j))) continue; - data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.pkts_received); - data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.bytes_received); - data[i++] = - CVM_CAST64(oct_dev->droq[j]->stats.dropped_nodispatch); - data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.dropped_nomem); - data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.dropped_toomany); + + /*packets send to TCP/IP network stack */ + /*# of packets to network stack */ data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.rx_pkts_received); + /*# of bytes to network stack */ data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.rx_bytes_received); + /*# of packets dropped */ + data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.dropped_nomem + + oct_dev->droq[j]->stats.dropped_toomany + + oct_dev->droq[j]->stats.rx_dropped); + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.dropped_nomem); + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.dropped_toomany); data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.rx_dropped); + + /*control and data path*/ + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.pkts_received); + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.bytes_received); + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.dropped_nodispatch); + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.rx_alloc_failure); } } @@ -579,26 +861,43 @@ static void lio_get_strings(struct net_device *netdev, u32 stringset, u8 *data) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct_dev = lio->oct_dev; int num_iq_stats, num_oq_stats, i, j; + int num_stats; - num_iq_stats = ARRAY_SIZE(oct_iq_stats_strings); - for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { - if (!(oct_dev->io_qmask.iq & (1ULL << i))) - continue; - for (j = 0; j < num_iq_stats; j++) { - sprintf(data, "IQ%d %s", i, oct_iq_stats_strings[j]); + switch (stringset) { + case ETH_SS_STATS: + num_stats = ARRAY_SIZE(oct_stats_strings); + for (j = 0; j < num_stats; j++) { + sprintf(data, "%s", oct_stats_strings[j]); data += ETH_GSTRING_LEN; } - } - num_oq_stats = ARRAY_SIZE(oct_droq_stats_strings); - /* for (i = 0; i < oct_dev->num_oqs; i++) { */ - for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { - if (!(oct_dev->io_qmask.oq & (1ULL << i))) - continue; - for (j = 0; j < num_oq_stats; j++) { - sprintf(data, "OQ%d %s", i, oct_droq_stats_strings[j]); - data += ETH_GSTRING_LEN; + num_iq_stats = ARRAY_SIZE(oct_iq_stats_strings); + for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct_dev); i++) { + if (!(oct_dev->io_qmask.iq & (1ULL << i))) + continue; + for (j = 0; j < num_iq_stats; j++) { + sprintf(data, "tx-%d-%s", i, + oct_iq_stats_strings[j]); + data += ETH_GSTRING_LEN; + } } + + num_oq_stats = ARRAY_SIZE(oct_droq_stats_strings); + /* for (i = 0; i < oct_dev->num_oqs; i++) { */ + for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct_dev); i++) { + if (!(oct_dev->io_qmask.oq & (1ULL << i))) + continue; + for (j = 0; j < num_oq_stats; j++) { + sprintf(data, "rx-%d-%s", i, + oct_droq_stats_strings[j]); + data += ETH_GSTRING_LEN; + } + } + break; + + default: + netif_info(lio, drv, lio->netdev, "Unknown Stringset !!\n"); + break; } } @@ -607,8 +906,14 @@ static int lio_get_sset_count(struct net_device *netdev, int sset) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct_dev = lio->oct_dev; - return (ARRAY_SIZE(oct_iq_stats_strings) * oct_dev->num_iqs) + - (ARRAY_SIZE(oct_droq_stats_strings) * oct_dev->num_oqs); + switch (sset) { + case ETH_SS_STATS: + return (ARRAY_SIZE(oct_stats_strings) + + ARRAY_SIZE(oct_iq_stats_strings) * oct_dev->num_iqs + + ARRAY_SIZE(oct_droq_stats_strings) * oct_dev->num_oqs); + default: + return -EOPNOTSUPP; + } } static int lio_get_intr_coalesce(struct net_device *netdev, @@ -731,6 +1036,155 @@ static int octnet_set_intrmod_cfg(struct lio *lio, return 0; } +void +octnet_nic_stats_callback(struct octeon_device *oct_dev, + u32 status, void *ptr) +{ + struct octeon_soft_command *sc = (struct octeon_soft_command *)ptr; + struct oct_nic_stats_resp *resp = (struct oct_nic_stats_resp *) + sc->virtrptr; + struct oct_nic_stats_ctrl *ctrl = (struct oct_nic_stats_ctrl *) + sc->ctxptr; + struct nic_rx_stats *rsp_rstats = &resp->stats.fromwire; + struct nic_tx_stats *rsp_tstats = &resp->stats.fromhost; + + struct nic_rx_stats *rstats = &oct_dev->link_stats.fromwire; + struct nic_tx_stats *tstats = &oct_dev->link_stats.fromhost; + + if ((status != OCTEON_REQUEST_TIMEOUT) && !resp->status) { + octeon_swap_8B_data((u64 *)&resp->stats, + (sizeof(struct oct_link_stats)) >> 3); + + /* RX link-level stats */ + rstats->total_rcvd = rsp_rstats->total_rcvd; + rstats->bytes_rcvd = rsp_rstats->bytes_rcvd; + rstats->total_bcst = rsp_rstats->total_bcst; + rstats->total_mcst = rsp_rstats->total_mcst; + rstats->runts = rsp_rstats->runts; + rstats->ctl_rcvd = rsp_rstats->ctl_rcvd; + /* Accounts for over/under-run of buffers */ + rstats->fifo_err = rsp_rstats->fifo_err; + rstats->dmac_drop = rsp_rstats->dmac_drop; + rstats->fcs_err = rsp_rstats->fcs_err; + rstats->jabber_err = rsp_rstats->jabber_err; + rstats->l2_err = rsp_rstats->l2_err; + rstats->frame_err = rsp_rstats->frame_err; + + /* RX firmware stats */ + rstats->fw_total_rcvd = rsp_rstats->fw_total_rcvd; + rstats->fw_total_fwd = rsp_rstats->fw_total_fwd; + rstats->fw_err_pko = rsp_rstats->fw_err_pko; + rstats->fw_err_link = rsp_rstats->fw_err_link; + rstats->fw_err_drop = rsp_rstats->fw_err_drop; + /* Number of packets that are LROed */ + rstats->fw_lro_pkts = rsp_rstats->fw_lro_pkts; + /* Number of octets that are LROed */ + rstats->fw_lro_octs = rsp_rstats->fw_lro_octs; + /* Number of LRO packets formed */ + rstats->fw_total_lro = rsp_rstats->fw_total_lro; + /* Number of times lRO of packet aborted */ + rstats->fw_lro_aborts = rsp_rstats->fw_lro_aborts; + rstats->fw_lro_aborts_port = rsp_rstats->fw_lro_aborts_port; + rstats->fw_lro_aborts_seq = rsp_rstats->fw_lro_aborts_seq; + rstats->fw_lro_aborts_tsval = rsp_rstats->fw_lro_aborts_tsval; + rstats->fw_lro_aborts_timer = rsp_rstats->fw_lro_aborts_timer; + /* intrmod: packet forward rate */ + rstats->fwd_rate = rsp_rstats->fwd_rate; + + /* TX link-level stats */ + tstats->total_pkts_sent = rsp_tstats->total_pkts_sent; + tstats->total_bytes_sent = rsp_tstats->total_bytes_sent; + tstats->mcast_pkts_sent = rsp_tstats->mcast_pkts_sent; + tstats->bcast_pkts_sent = rsp_tstats->bcast_pkts_sent; + tstats->ctl_sent = rsp_tstats->ctl_sent; + /* Packets sent after one collision*/ + tstats->one_collision_sent = rsp_tstats->one_collision_sent; + /* Packets sent after multiple collision*/ + tstats->multi_collision_sent = rsp_tstats->multi_collision_sent; + /* Packets not sent due to max collisions */ + tstats->max_collision_fail = rsp_tstats->max_collision_fail; + /* Packets not sent due to max deferrals */ + tstats->max_deferral_fail = rsp_tstats->max_deferral_fail; + /* Accounts for over/under-run of buffers */ + tstats->fifo_err = rsp_tstats->fifo_err; + tstats->runts = rsp_tstats->runts; + /* Total number of collisions detected */ + tstats->total_collisions = rsp_tstats->total_collisions; + + /* firmware stats */ + tstats->fw_total_sent = rsp_tstats->fw_total_sent; + tstats->fw_total_fwd = rsp_tstats->fw_total_fwd; + tstats->fw_err_pko = rsp_tstats->fw_err_pko; + tstats->fw_err_link = rsp_tstats->fw_err_link; + tstats->fw_err_drop = rsp_tstats->fw_err_drop; + tstats->fw_tso = rsp_tstats->fw_tso; + tstats->fw_tso_fwd = rsp_tstats->fw_tso_fwd; + tstats->fw_err_tso = rsp_tstats->fw_err_tso; + resp->status = 1; + } else { + resp->status = -1; + } + complete(&ctrl->complete); +} + +/* Configure interrupt moderation parameters */ +static int octnet_get_link_stats(struct net_device *netdev) +{ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct_dev = lio->oct_dev; + + struct octeon_soft_command *sc; + struct oct_nic_stats_ctrl *ctrl; + struct oct_nic_stats_resp *resp; + + int retval; + + /* Alloc soft command */ + sc = (struct octeon_soft_command *) + octeon_alloc_soft_command(oct_dev, + 0, + sizeof(struct oct_nic_stats_resp), + sizeof(struct octnic_ctrl_pkt)); + + if (!sc) + return -ENOMEM; + + resp = (struct oct_nic_stats_resp *)sc->virtrptr; + memset(resp, 0, sizeof(struct oct_nic_stats_resp)); + + ctrl = (struct oct_nic_stats_ctrl *)sc->ctxptr; + memset(ctrl, 0, sizeof(struct oct_nic_stats_ctrl)); + ctrl->netdev = netdev; + init_completion(&ctrl->complete); + + sc->iq_no = lio->linfo.txpciq[0].s.q_no; + + octeon_prepare_soft_command(oct_dev, sc, OPCODE_NIC, + OPCODE_NIC_PORT_STATS, 0, 0, 0); + + sc->callback = octnet_nic_stats_callback; + sc->callback_arg = sc; + sc->wait_time = 500; /*in milli seconds*/ + + retval = octeon_send_soft_command(oct_dev, sc); + if (retval == IQ_SEND_FAILED) { + octeon_free_soft_command(oct_dev, sc); + return -EINVAL; + } + + wait_for_completion_timeout(&ctrl->complete, msecs_to_jiffies(1000)); + + if (resp->status != 1) { + octeon_free_soft_command(oct_dev, sc); + + return -EINVAL; + } + + octeon_free_soft_command(oct_dev, sc); + + return 0; +} + /* Enable/Disable auto interrupt Moderation */ static int oct_cfg_adaptive_intr(struct lio *lio, struct ethtool_coalesce *intr_coal) diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index a1e31ac..9f97d11 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -72,6 +72,9 @@ MODULE_PARM_DESC(console_bitmask, #define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK) +#define INCR_INSTRQUEUE_PKT_COUNT(octeon_dev_ptr, iq_no, field, count) \ + (octeon_dev_ptr->instr_queue[iq_no]->stats.field += count) + static int debug = -1; module_param(debug, int, 0644); MODULE_PARM_DESC(debug, "NETIF_MSG debug bits"); @@ -682,13 +685,24 @@ static inline void txqs_start(struct net_device *netdev) */ static inline void txqs_wake(struct net_device *netdev) { + struct lio *lio = GET_LIO(netdev); + if (netif_is_multiqueue(netdev)) { int i; - for (i = 0; i < netdev->num_tx_queues; i++) - if (__netif_subqueue_stopped(netdev, i)) + for (i = 0; i < netdev->num_tx_queues; i++) { + int qno = lio->linfo.txpciq[i % + (lio->linfo.num_txpciq)].s.q_no; + + if (__netif_subqueue_stopped(netdev, i)) { + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, qno, + tx_restart, 1); netif_wake_subqueue(netdev, i); + } + } } else { + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, lio->txq, + tx_restart, 1); netif_wake_queue(netdev); } } @@ -763,6 +777,8 @@ static inline int check_txq_status(struct lio *lio) continue; if (__netif_subqueue_stopped(lio->netdev, q)) { wake_q(lio->netdev, q); + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq, + tx_restart, 1); ret_val++; } } @@ -770,6 +786,8 @@ static inline int check_txq_status(struct lio *lio) if (octnet_iq_is_full(lio->oct_dev, lio->txq)) return 0; wake_q(lio->netdev, lio->txq); + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, lio->txq, + tx_restart, 1); ret_val = 1; } return ret_val; @@ -981,10 +999,16 @@ static void update_txq_status(struct octeon_device *oct, int iq_num) if (__netif_subqueue_stopped(netdev, iq->q_index) && lio->linfo.link.s.link_up && (!octnet_iq_is_full(oct, iq_num))) { + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq_num, + tx_restart, 1); netif_wake_subqueue(netdev, iq->q_index); } else { - if (!octnet_iq_is_full(oct, lio->txq)) + if (!octnet_iq_is_full(oct, lio->txq)) { + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, + lio->txq, + tx_restart, 1); wake_q(netdev, lio->txq); + } } } } @@ -1115,6 +1139,9 @@ static int liquidio_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return -ENOMEM; } + oct_dev->rx_pause = 1; + oct_dev->tx_pause = 1; + dev_dbg(&oct_dev->pci_dev->dev, "Device is ready\n"); return 0; @@ -1468,8 +1495,10 @@ static inline int check_txq_state(struct lio *lio, struct sk_buff *skb) if (octnet_iq_is_full(lio->oct_dev, iq)) return 0; - if (__netif_subqueue_stopped(lio->netdev, q)) + if (__netif_subqueue_stopped(lio->netdev, q)) { + INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq, tx_restart, 1); wake_q(lio->netdev, q); + } return 1; } @@ -2382,6 +2411,10 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr) break; + case OCTNET_CMD_SET_FLOW_CTL: + netif_info(lio, probe, lio->netdev, "Set RX/TX flow control parameters\n"); + break; + default: dev_err(&oct->pci_dev->dev, "%s Unknown cmd %d\n", __func__, nctrl->ncmd.s.cmd); @@ -2976,7 +3009,9 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) if (skb_shinfo(skb)->gso_size) { tx_info->s.gso_size = skb_shinfo(skb)->gso_size; tx_info->s.gso_segs = skb_shinfo(skb)->gso_segs; + stats->tx_gso++; } + /* HW insert VLAN tag */ if (skb_vlan_tag_present(skb)) { irh->priority = skb_vlan_tag_get(skb) >> 13; @@ -2999,7 +3034,10 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) netif_trans_update(netdev); - stats->tx_done++; + if (skb_shinfo(skb)->gso_size) + stats->tx_done += skb_shinfo(skb)->gso_segs; + else + stats->tx_done++; stats->tx_tot_bytes += skb->len; return NETDEV_TX_OK; diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index b5ab704..4dc8e54 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -736,10 +736,16 @@ struct nic_rx_stats { u64 fw_err_pko; u64 fw_err_link; u64 fw_err_drop; + + /* LRO */ u64 fw_lro_pkts; /* Number of packets that are LROed */ u64 fw_lro_octs; /* Number of octets that are LROed */ u64 fw_total_lro; /* Number of LRO packets formed */ u64 fw_lro_aborts; /* Number of times lRO of packet aborted */ + u64 fw_lro_aborts_port; + u64 fw_lro_aborts_seq; + u64 fw_lro_aborts_tsval; + u64 fw_lro_aborts_timer; /* intrmod: packet forward rate */ u64 fwd_rate; }; @@ -763,9 +769,13 @@ struct nic_tx_stats { /* firmware stats */ u64 fw_total_sent; u64 fw_total_fwd; + u64 fw_total_fwd_bytes; u64 fw_err_pko; u64 fw_err_link; u64 fw_err_drop; + u64 fw_err_tso; + u64 fw_tso; /* number of tso requests */ + u64 fw_tso_fwd; /* number of packets segmented in tso */ }; struct oct_link_stats { diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h index ceb905d..95b4eb7 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h @@ -394,6 +394,12 @@ struct octeon_device { struct cavium_wk console_poll_work[MAX_OCTEON_MAPS]; void *priv; + + int rx_pause; + int tx_pause; + + struct oct_link_stats link_stats; /*stastics from firmware*/ + }; #define OCT_DRV_ONLINE 1 diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h index 69d5b91..caa2b4f 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h @@ -65,6 +65,10 @@ struct oct_iq_stats { u64 tx_iq_busy;/**< Numof times this iq was found to be full. */ u64 tx_dropped;/**< Numof pkts dropped dueto xmitpath errors. */ u64 tx_tot_bytes;/**< Total count of bytes sento to network. */ + u64 tx_gso; /* count of tso */ + u64 tx_dmamap_fail; + u64 tx_restart; + /*u64 tx_timeout_count;*/ }; #define OCT_IQ_STATS_SIZE (sizeof(struct oct_iq_stats)) diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h index 9c14484..b481edc 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h @@ -30,6 +30,17 @@ #include #include +struct oct_nic_stats_resp { + u64 rh; + struct oct_link_stats stats; + u64 status; +}; + +struct oct_nic_stats_ctrl { + struct completion complete; + struct net_device *netdev; +}; + /** LiquidIO per-interface network private data */ struct lio { /** State of the interface. Rx/Tx happens only in the RUNNING state. */ -- cgit v0.10.2 From 9eb60844c8c54c4fb4e26c3e7621c0f85127d9f7 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:12 -0700 Subject: liquidio: New xaui info This patch adds support for host driver support for new Xaui interfaces. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 2b03095..88994c5 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -73,6 +73,11 @@ enum { INTERFACE_MODE_RXAUI, INTERFACE_MODE_QSGMII, INTERFACE_MODE_AGL, + INTERFACE_MODE_XLAUI, + INTERFACE_MODE_XFI, + INTERFACE_MODE_10G_KR, + INTERFACE_MODE_40G_KR4, + INTERFACE_MODE_MIXED, }; #define ARRAY_LENGTH(a) (sizeof(a) / sizeof((a)[0])) @@ -195,8 +200,9 @@ static int lio_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) linfo = &lio->linfo; - if (linfo->link.s.interface == INTERFACE_MODE_XAUI || - linfo->link.s.interface == INTERFACE_MODE_RXAUI) { + if (linfo->link.s.if_mode == INTERFACE_MODE_XAUI || + linfo->link.s.if_mode == INTERFACE_MODE_RXAUI || + linfo->link.s.if_mode == INTERFACE_MODE_XFI) { ecmd->port = PORT_FIBRE; ecmd->supported = (SUPPORTED_10000baseT_Full | SUPPORTED_FIBRE | @@ -207,7 +213,8 @@ static int lio_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) ecmd->autoneg = AUTONEG_DISABLE; } else { - dev_err(&oct->pci_dev->dev, "Unknown link interface reported\n"); + dev_err(&oct->pci_dev->dev, "Unknown link interface reported %d\n", + linfo->link.s.if_mode); } if (linfo->link.s.link_up) { @@ -1450,12 +1457,14 @@ static int lio_set_settings(struct net_device *netdev, struct ethtool_cmd *ecmd) ecmd->duplex != DUPLEX_FULL))) return -EINVAL; - /* Ethtool Support is not provided for XAUI and RXAUI Interfaces + /* Ethtool Support is not provided for XAUI, RXAUI, and XFI Interfaces * as they operate at fixed Speed and Duplex settings */ - if (linfo->link.s.interface == INTERFACE_MODE_XAUI || - linfo->link.s.interface == INTERFACE_MODE_RXAUI) { - dev_info(&oct->pci_dev->dev, "XAUI IFs settings cannot be modified.\n"); + if (linfo->link.s.if_mode == INTERFACE_MODE_XAUI || + linfo->link.s.if_mode == INTERFACE_MODE_RXAUI || + linfo->link.s.if_mode == INTERFACE_MODE_XFI) { + dev_info(&oct->pci_dev->dev, + "Autonegotiation, duplex and speed settings cannot be modified.\n"); return -EINVAL; } diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 4dc8e54..5aa01f4 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -628,13 +628,13 @@ union oct_link_status { u64 speed:16; u64 link_up:1; u64 autoneg:1; - u64 interface:4; + u64 if_mode:5; u64 pause:1; - u64 reserved:17; + u64 reserved:16; #else - u64 reserved:17; + u64 reserved:16; u64 pause:1; - u64 interface:4; + u64 if_mode:5; u64 autoneg:1; u64 link_up:1; u64 speed:16; -- cgit v0.10.2 From 178cc10e3d9b577f6fb3c87f1cf215091c339014 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:13 -0700 Subject: liquidio: ptp info This patch has minor changes for proper ptp info retreival. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 88994c5..85f5d17 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -1416,23 +1416,28 @@ static int lio_get_ts_info(struct net_device *netdev, struct lio *lio = GET_LIO(netdev); info->so_timestamping = +#ifdef PTP_HARDWARE_TIMESTAMPING SOF_TIMESTAMPING_TX_HARDWARE | - SOF_TIMESTAMPING_TX_SOFTWARE | SOF_TIMESTAMPING_RX_HARDWARE | + SOF_TIMESTAMPING_RAW_HARDWARE | + SOF_TIMESTAMPING_TX_SOFTWARE | +#endif SOF_TIMESTAMPING_RX_SOFTWARE | - SOF_TIMESTAMPING_SOFTWARE | SOF_TIMESTAMPING_RAW_HARDWARE; + SOF_TIMESTAMPING_SOFTWARE; if (lio->ptp_clock) info->phc_index = ptp_clock_index(lio->ptp_clock); else info->phc_index = -1; +#ifdef PTP_HARDWARE_TIMESTAMPING info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON); info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) | (1 << HWTSTAMP_FILTER_PTP_V1_L4_EVENT) | (1 << HWTSTAMP_FILTER_PTP_V2_L2_EVENT) | (1 << HWTSTAMP_FILTER_PTP_V2_L4_EVENT); +#endif return 0; } -- cgit v0.10.2 From f5a20472e28ffe8810306ec99d4f57b4b62bb09c Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:14 -0700 Subject: liquidio: Support priv flag This patch adds support for private flags for the driver. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 85f5d17..03bfa97 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -1683,6 +1683,23 @@ static void lio_get_regs(struct net_device *dev, } } +static u32 lio_get_priv_flags(struct net_device *netdev) +{ + struct lio *lio = GET_LIO(netdev); + + return lio->oct_dev->priv_flags; +} + +static int lio_set_priv_flags(struct net_device *netdev, u32 flags) +{ + struct lio *lio = GET_LIO(netdev); + bool intr_by_tx_bytes = !!(flags & (0x1 << OCT_PRIV_FLAG_TX_BYTES)); + + lio_set_priv_flag(lio->oct_dev, OCT_PRIV_FLAG_TX_BYTES, + intr_by_tx_bytes); + return 0; +} + static const struct ethtool_ops lio_ethtool_ops = { .get_settings = lio_get_settings, .get_link = ethtool_op_get_link, @@ -1704,6 +1721,8 @@ static const struct ethtool_ops lio_ethtool_ops = { .set_settings = lio_set_settings, .get_coalesce = lio_get_intr_coalesce, .set_coalesce = lio_set_intr_coalesce, + .get_priv_flags = lio_get_priv_flags, + .set_priv_flags = lio_set_priv_flags, .get_ts_info = lio_get_ts_info, }; diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 9f97d11..a8328f2 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -3535,6 +3535,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) /* Register ethtool support */ liquidio_set_ethtool_ops(netdev); + octeon_dev->priv_flags = 0x0; if (netdev->features & NETIF_F_LRO) liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE, diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h index 95b4eb7..b4e566d 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h @@ -400,6 +400,8 @@ struct octeon_device { struct oct_link_stats link_stats; /*stastics from firmware*/ + /* private flags to control driver-specific features through ethtool */ + u32 priv_flags; }; #define OCT_DRV_ONLINE 1 @@ -660,4 +662,17 @@ void *oct_get_config_info(struct octeon_device *oct, u16 card_type); */ struct octeon_config *octeon_get_conf(struct octeon_device *oct); +/* LiquidIO driver pivate flags */ +enum { + OCT_PRIV_FLAG_TX_BYTES = 0, /* Tx interrupts by pending byte count */ +}; + +static inline void lio_set_priv_flag(struct octeon_device *octdev, u32 flag, + u32 val) +{ + if (val) + octdev->priv_flags |= (0x1 << flag); + else + octdev->priv_flags &= ~(0x1 << flag); +} #endif -- cgit v0.10.2 From 4b129ae3b692653de1794ff6a5746691ad77c433 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Tue, 21 Jun 2016 22:53:15 -0700 Subject: liquidio: ddr timeout Adds support for ddr_timeout during device init. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index a8328f2..1a584eb 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -3805,14 +3805,19 @@ static int octeon_device_init(struct octeon_device *octeon_dev) dev_dbg(&octeon_dev->pci_dev->dev, "Waiting for DDR initialization...\n"); - if (ddr_timeout == 0) { - dev_info(&octeon_dev->pci_dev->dev, - "WAITING. Set ddr_timeout to non-zero value to proceed with initialization.\n"); - } + if (ddr_timeout == 0) + dev_info(&octeon_dev->pci_dev->dev, "WAITING. Set ddr_timeout to non-zero value to proceed with initialization.\n"); schedule_timeout_uninterruptible(HZ * LIO_RESET_SECS); /* Wait for the octeon to initialize DDR after the soft-reset. */ + while (ddr_timeout == 0) { + set_current_state(TASK_INTERRUPTIBLE); + if (schedule_timeout(HZ / 10)) { + /* user probably pressed Control-C */ + return 1; + } + } ret = octeon_wait_for_ddr_init(octeon_dev, &ddr_timeout); if (ret) { dev_err(&octeon_dev->pci_dev->dev, diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index bc4d6af..3372207 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -1284,9 +1284,6 @@ int octeon_wait_for_ddr_init(struct octeon_device *oct, u32 *timeout) if (!timeout) return ret; - while (*timeout == 0) - schedule_timeout_uninterruptible(HZ / 10); - for (ms = 0; (ret != 0) && ((*timeout == 0) || (ms <= *timeout)); ms += HZ / 10) { ret = octeon_mem_access_ok(oct); -- cgit v0.10.2 From 520ac30f45519b0a82dd92117c181d1d6144677b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 21 Jun 2016 23:16:49 -0700 Subject: net_sched: drop packets after root qdisc lock is released Qdisc performance suffers when packets are dropped at enqueue() time because drops (kfree_skb()) are done while qdisc lock is held, delaying a dequeue() draining the queue. Nominal throughput can be reduced by 50 % when this happens, at a time we would like the dequeue() to proceed as fast as possible. Even FQ is vulnerable to this problem, while one of FQ goals was to provide some flow isolation. This patch adds a 'struct sk_buff **to_free' parameter to all qdisc->enqueue(), and in qdisc_drop() helper. I measured a performance increase of up to 12 %, but this patch is a prereq so that future batches in enqueue() can fly. Signed-off-by: Eric Dumazet Acked-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 4f7cee8..04e84c0 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -37,8 +37,10 @@ struct qdisc_size_table { }; struct Qdisc { - int (*enqueue)(struct sk_buff *skb, struct Qdisc *dev); - struct sk_buff * (*dequeue)(struct Qdisc *dev); + int (*enqueue)(struct sk_buff *skb, + struct Qdisc *sch, + struct sk_buff **to_free); + struct sk_buff * (*dequeue)(struct Qdisc *sch); unsigned int flags; #define TCQ_F_BUILTIN 1 #define TCQ_F_INGRESS 2 @@ -160,7 +162,9 @@ struct Qdisc_ops { char id[IFNAMSIZ]; int priv_size; - int (*enqueue)(struct sk_buff *, struct Qdisc *); + int (*enqueue)(struct sk_buff *skb, + struct Qdisc *sch, + struct sk_buff **to_free); struct sk_buff * (*dequeue)(struct Qdisc *); struct sk_buff * (*peek)(struct Qdisc *); @@ -498,10 +502,11 @@ static inline void qdisc_calculate_pkt_len(struct sk_buff *skb, #endif } -static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { qdisc_calculate_pkt_len(skb, sch); - return sch->enqueue(skb, sch); + return sch->enqueue(skb, sch, to_free); } static inline bool qdisc_is_percpu_stats(const struct Qdisc *q) @@ -626,24 +631,36 @@ static inline struct sk_buff *qdisc_dequeue_head(struct Qdisc *sch) return __qdisc_dequeue_head(sch, &sch->q); } +/* Instead of calling kfree_skb() while root qdisc lock is held, + * queue the skb for future freeing at end of __dev_xmit_skb() + */ +static inline void __qdisc_drop(struct sk_buff *skb, struct sk_buff **to_free) +{ + skb->next = *to_free; + *to_free = skb; +} + static inline unsigned int __qdisc_queue_drop_head(struct Qdisc *sch, - struct sk_buff_head *list) + struct sk_buff_head *list, + struct sk_buff **to_free) { struct sk_buff *skb = __skb_dequeue(list); if (likely(skb != NULL)) { unsigned int len = qdisc_pkt_len(skb); + qdisc_qstats_backlog_dec(sch, skb); - kfree_skb(skb); + __qdisc_drop(skb, to_free); return len; } return 0; } -static inline unsigned int qdisc_queue_drop_head(struct Qdisc *sch) +static inline unsigned int qdisc_queue_drop_head(struct Qdisc *sch, + struct sk_buff **to_free) { - return __qdisc_queue_drop_head(sch, &sch->q); + return __qdisc_queue_drop_head(sch, &sch->q, to_free); } static inline struct sk_buff *qdisc_peek_head(struct Qdisc *sch) @@ -724,9 +741,11 @@ static inline void rtnl_qdisc_drop(struct sk_buff *skb, struct Qdisc *sch) qdisc_qstats_drop(sch); } -static inline int qdisc_drop(struct sk_buff *skb, struct Qdisc *sch) + +static inline int qdisc_drop(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { - kfree_skb(skb); + __qdisc_drop(skb, to_free); qdisc_qstats_drop(sch); return NET_XMIT_DROP; diff --git a/net/core/dev.c b/net/core/dev.c index d40593b..aba10d2 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3070,6 +3070,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, struct netdev_queue *txq) { spinlock_t *root_lock = qdisc_lock(q); + struct sk_buff *to_free = NULL; bool contended; int rc; @@ -3086,7 +3087,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, spin_lock(root_lock); if (unlikely(test_bit(__QDISC_STATE_DEACTIVATED, &q->state))) { - kfree_skb(skb); + __qdisc_drop(skb, &to_free); rc = NET_XMIT_DROP; } else if ((q->flags & TCQ_F_CAN_BYPASS) && !qdisc_qlen(q) && qdisc_run_begin(q)) { @@ -3109,7 +3110,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, rc = NET_XMIT_SUCCESS; } else { - rc = q->enqueue(skb, q) & NET_XMIT_MASK; + rc = q->enqueue(skb, q, &to_free) & NET_XMIT_MASK; if (qdisc_run_begin(q)) { if (unlikely(contended)) { spin_unlock(&q->busylock); @@ -3119,6 +3120,8 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, } } spin_unlock(root_lock); + if (unlikely(to_free)) + kfree_skb_list(to_free); if (unlikely(contended)) spin_unlock(&q->busylock); return rc; diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index e04ea69..481e4f1 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -357,7 +357,8 @@ static struct tcf_proto __rcu **atm_tc_find_tcf(struct Qdisc *sch, /* --------------------------- Qdisc operations ---------------------------- */ -static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct atm_qdisc_data *p = qdisc_priv(sch); struct atm_flow_data *flow; @@ -398,10 +399,10 @@ done: switch (result) { case TC_ACT_QUEUED: case TC_ACT_STOLEN: - kfree_skb(skb); + __qdisc_drop(skb, to_free); return NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; case TC_ACT_SHOT: - kfree_skb(skb); + __qdisc_drop(skb, to_free); goto drop; case TC_ACT_RECLASSIFY: if (flow->excess) @@ -413,7 +414,7 @@ done: #endif } - ret = qdisc_enqueue(skb, flow->q); + ret = qdisc_enqueue(skb, flow->q, to_free); if (ret != NET_XMIT_SUCCESS) { drop: __maybe_unused if (net_xmit_drop_count(ret)) { diff --git a/net/sched/sch_blackhole.c b/net/sched/sch_blackhole.c index 3fee70d..c98a61e 100644 --- a/net/sched/sch_blackhole.c +++ b/net/sched/sch_blackhole.c @@ -17,9 +17,10 @@ #include #include -static int blackhole_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int blackhole_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { - qdisc_drop(skb, sch); + qdisc_drop(skb, sch, to_free); return NET_XMIT_SUCCESS; } diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index a29fd81..beb554a 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -358,7 +358,8 @@ cbq_mark_toplevel(struct cbq_sched_data *q, struct cbq_class *cl) } static int -cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch) +cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct cbq_sched_data *q = qdisc_priv(sch); int uninitialized_var(ret); @@ -370,11 +371,11 @@ cbq_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (cl == NULL) { if (ret & __NET_XMIT_BYPASS) qdisc_qstats_drop(sch); - kfree_skb(skb); + __qdisc_drop(skb, to_free); return ret; } - ret = qdisc_enqueue(skb, cl->q); + ret = qdisc_enqueue(skb, cl->q, to_free); if (ret == NET_XMIT_SUCCESS) { sch->q.qlen++; cbq_mark_toplevel(q, cl); diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c index 789b69e..3b6d5bd 100644 --- a/net/sched/sch_choke.c +++ b/net/sched/sch_choke.c @@ -115,7 +115,8 @@ static void choke_zap_tail_holes(struct choke_sched_data *q) } /* Drop packet from queue array by creating a "hole" */ -static void choke_drop_by_idx(struct Qdisc *sch, unsigned int idx) +static void choke_drop_by_idx(struct Qdisc *sch, unsigned int idx, + struct sk_buff **to_free) { struct choke_sched_data *q = qdisc_priv(sch); struct sk_buff *skb = q->tab[idx]; @@ -129,7 +130,7 @@ static void choke_drop_by_idx(struct Qdisc *sch, unsigned int idx) qdisc_qstats_backlog_dec(sch, skb); qdisc_tree_reduce_backlog(sch, 1, qdisc_pkt_len(skb)); - qdisc_drop(skb, sch); + qdisc_drop(skb, sch, to_free); --sch->q.qlen; } @@ -261,7 +262,8 @@ static bool choke_match_random(const struct choke_sched_data *q, return choke_match_flow(oskb, nskb); } -static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { int ret = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; struct choke_sched_data *q = qdisc_priv(sch); @@ -288,7 +290,7 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch) /* Draw a packet at random from queue and compare flow */ if (choke_match_random(q, skb, &idx)) { q->stats.matched++; - choke_drop_by_idx(sch, idx); + choke_drop_by_idx(sch, idx, to_free); goto congestion_drop; } @@ -331,16 +333,16 @@ static int choke_enqueue(struct sk_buff *skb, struct Qdisc *sch) } q->stats.pdrop++; - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); congestion_drop: - qdisc_drop(skb, sch); + qdisc_drop(skb, sch, to_free); return NET_XMIT_CN; other_drop: if (ret & __NET_XMIT_BYPASS) qdisc_qstats_drop(sch); - kfree_skb(skb); + __qdisc_drop(skb, to_free); return ret; } diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c index c5bc424..4002df3 100644 --- a/net/sched/sch_codel.c +++ b/net/sched/sch_codel.c @@ -82,7 +82,8 @@ static void drop_func(struct sk_buff *skb, void *ctx) { struct Qdisc *sch = ctx; - qdisc_drop(skb, sch); + kfree_skb(skb); + qdisc_qstats_drop(sch); } static struct sk_buff *codel_qdisc_dequeue(struct Qdisc *sch) @@ -107,7 +108,8 @@ static struct sk_buff *codel_qdisc_dequeue(struct Qdisc *sch) return skb; } -static int codel_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int codel_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct codel_sched_data *q; @@ -117,7 +119,7 @@ static int codel_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch) } q = qdisc_priv(sch); q->drop_overlimit++; - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); } static const struct nla_policy codel_policy[TCA_CODEL_MAX + 1] = { diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c index 22609e4..8af5c59 100644 --- a/net/sched/sch_drr.c +++ b/net/sched/sch_drr.c @@ -350,7 +350,8 @@ static struct drr_class *drr_classify(struct sk_buff *skb, struct Qdisc *sch, return NULL; } -static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct drr_sched *q = qdisc_priv(sch); struct drr_class *cl; @@ -360,11 +361,11 @@ static int drr_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (cl == NULL) { if (err & __NET_XMIT_BYPASS) qdisc_qstats_drop(sch); - kfree_skb(skb); + __qdisc_drop(skb, to_free); return err; } - err = qdisc_enqueue(skb, cl->qdisc); + err = qdisc_enqueue(skb, cl->qdisc, to_free); if (unlikely(err != NET_XMIT_SUCCESS)) { if (net_xmit_drop_count(err)) { cl->qstats.drops++; diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c index b9ba5f6..1308bbf 100644 --- a/net/sched/sch_dsmark.c +++ b/net/sched/sch_dsmark.c @@ -191,7 +191,8 @@ static inline struct tcf_proto __rcu **dsmark_find_tcf(struct Qdisc *sch, /* --------------------------- Qdisc operations ---------------------------- */ -static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct dsmark_qdisc_data *p = qdisc_priv(sch); int err; @@ -234,7 +235,7 @@ static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch) #ifdef CONFIG_NET_CLS_ACT case TC_ACT_QUEUED: case TC_ACT_STOLEN: - kfree_skb(skb); + __qdisc_drop(skb, to_free); return NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; case TC_ACT_SHOT: @@ -251,7 +252,7 @@ static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch) } } - err = qdisc_enqueue(skb, p->q); + err = qdisc_enqueue(skb, p->q, to_free); if (err != NET_XMIT_SUCCESS) { if (net_xmit_drop_count(err)) qdisc_qstats_drop(sch); @@ -264,7 +265,7 @@ static int dsmark_enqueue(struct sk_buff *skb, struct Qdisc *sch) return NET_XMIT_SUCCESS; drop: - qdisc_drop(skb, sch); + qdisc_drop(skb, sch, to_free); return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; } diff --git a/net/sched/sch_fifo.c b/net/sched/sch_fifo.c index dea70e3..6ea0db4 100644 --- a/net/sched/sch_fifo.c +++ b/net/sched/sch_fifo.c @@ -19,29 +19,32 @@ /* 1 band FIFO pseudo-"scheduler" */ -static int bfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int bfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { if (likely(sch->qstats.backlog + qdisc_pkt_len(skb) <= sch->limit)) return qdisc_enqueue_tail(skb, sch); - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); } -static int pfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int pfifo_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { if (likely(skb_queue_len(&sch->q) < sch->limit)) return qdisc_enqueue_tail(skb, sch); - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); } -static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { if (likely(skb_queue_len(&sch->q) < sch->limit)) return qdisc_enqueue_tail(skb, sch); /* queue full, remove one skb to fulfill the limit */ - __qdisc_queue_drop_head(sch, &sch->q); + __qdisc_queue_drop_head(sch, &sch->q, to_free); qdisc_qstats_drop(sch); qdisc_enqueue_tail(skb, sch); diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index 6eb0667..e5458b9 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -368,18 +368,19 @@ static void flow_queue_add(struct fq_flow *flow, struct sk_buff *skb) } } -static int fq_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int fq_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct fq_sched_data *q = qdisc_priv(sch); struct fq_flow *f; if (unlikely(sch->q.qlen >= sch->limit)) - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); f = fq_classify(skb, q); if (unlikely(f->qlen >= q->flow_plimit && f != &q->internal)) { q->stat_flows_plimit++; - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); } f->qlen++; diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index 2dc0a84..f715195 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -139,7 +139,8 @@ static inline void flow_queue_add(struct fq_codel_flow *flow, skb->next = NULL; } -static unsigned int fq_codel_drop(struct Qdisc *sch, unsigned int max_packets) +static unsigned int fq_codel_drop(struct Qdisc *sch, unsigned int max_packets, + struct sk_buff **to_free) { struct fq_codel_sched_data *q = qdisc_priv(sch); struct sk_buff *skb; @@ -172,7 +173,7 @@ static unsigned int fq_codel_drop(struct Qdisc *sch, unsigned int max_packets) skb = dequeue_head(flow); len += qdisc_pkt_len(skb); mem += skb->truesize; - kfree_skb(skb); + __qdisc_drop(skb, to_free); } while (++i < max_packets && len < threshold); flow->dropped += i; @@ -184,7 +185,8 @@ static unsigned int fq_codel_drop(struct Qdisc *sch, unsigned int max_packets) return idx; } -static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct fq_codel_sched_data *q = qdisc_priv(sch); unsigned int idx, prev_backlog, prev_qlen; @@ -197,7 +199,7 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (idx == 0) { if (ret & __NET_XMIT_BYPASS) qdisc_qstats_drop(sch); - kfree_skb(skb); + __qdisc_drop(skb, to_free); return ret; } idx--; @@ -229,7 +231,7 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch) * So instead of dropping a single packet, drop half of its backlog * with a 64 packets limit to not add a too big cpu spike here. */ - ret = fq_codel_drop(sch, q->drop_batch_size); + ret = fq_codel_drop(sch, q->drop_batch_size, to_free); prev_qlen -= sch->q.qlen; prev_backlog -= sch->qstats.backlog; @@ -276,7 +278,8 @@ static void drop_func(struct sk_buff *skb, void *ctx) { struct Qdisc *sch = ctx; - qdisc_drop(skb, sch); + kfree_skb(skb); + qdisc_qstats_drop(sch); } static struct sk_buff *fq_codel_dequeue(struct Qdisc *sch) diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 773b632..ff86606 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -348,9 +348,10 @@ EXPORT_SYMBOL(netif_carrier_off); cheaper. */ -static int noop_enqueue(struct sk_buff *skb, struct Qdisc *qdisc) +static int noop_enqueue(struct sk_buff *skb, struct Qdisc *qdisc, + struct sk_buff **to_free) { - kfree_skb(skb); + __qdisc_drop(skb, to_free); return NET_XMIT_CN; } @@ -439,7 +440,8 @@ static inline struct sk_buff_head *band2list(struct pfifo_fast_priv *priv, return priv->q + band; } -static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc *qdisc) +static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc *qdisc, + struct sk_buff **to_free) { if (skb_queue_len(&qdisc->q) < qdisc_dev(qdisc)->tx_queue_len) { int band = prio2band[skb->priority & TC_PRIO_MAX]; @@ -451,7 +453,7 @@ static int pfifo_fast_enqueue(struct sk_buff *skb, struct Qdisc *qdisc) return __qdisc_enqueue_tail(skb, qdisc, list); } - return qdisc_drop(skb, qdisc); + return qdisc_drop(skb, qdisc, to_free); } static struct sk_buff *pfifo_fast_dequeue(struct Qdisc *qdisc) diff --git a/net/sched/sch_gred.c b/net/sched/sch_gred.c index b5fb63c7..c78a093 100644 --- a/net/sched/sch_gred.c +++ b/net/sched/sch_gred.c @@ -149,7 +149,8 @@ static inline int gred_use_harddrop(struct gred_sched *t) return t->red_flags & TC_RED_HARDDROP; } -static int gred_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int gred_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct gred_sched_data *q = NULL; struct gred_sched *t = qdisc_priv(sch); @@ -237,10 +238,10 @@ static int gred_enqueue(struct sk_buff *skb, struct Qdisc *sch) q->stats.pdrop++; drop: - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); congestion_drop: - qdisc_drop(skb, sch); + qdisc_drop(skb, sch, to_free); return NET_XMIT_CN; } diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index bd08c36..8cb5eff 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1572,7 +1572,7 @@ hfsc_dump_qdisc(struct Qdisc *sch, struct sk_buff *skb) } static int -hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch) +hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) { struct hfsc_class *cl; int uninitialized_var(err); @@ -1581,11 +1581,11 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (cl == NULL) { if (err & __NET_XMIT_BYPASS) qdisc_qstats_drop(sch); - kfree_skb(skb); + __qdisc_drop(skb, to_free); return err; } - err = qdisc_enqueue(skb, cl->qdisc); + err = qdisc_enqueue(skb, cl->qdisc, to_free); if (unlikely(err != NET_XMIT_SUCCESS)) { if (net_xmit_drop_count(err)) { cl->qstats.drops++; diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c index c44593b..e3d0458 100644 --- a/net/sched/sch_hhf.c +++ b/net/sched/sch_hhf.c @@ -345,7 +345,7 @@ static void bucket_add(struct wdrr_bucket *bucket, struct sk_buff *skb) skb->next = NULL; } -static unsigned int hhf_drop(struct Qdisc *sch) +static unsigned int hhf_drop(struct Qdisc *sch, struct sk_buff **to_free) { struct hhf_sched_data *q = qdisc_priv(sch); struct wdrr_bucket *bucket; @@ -359,16 +359,16 @@ static unsigned int hhf_drop(struct Qdisc *sch) struct sk_buff *skb = dequeue_head(bucket); sch->q.qlen--; - qdisc_qstats_drop(sch); qdisc_qstats_backlog_dec(sch, skb); - kfree_skb(skb); + qdisc_drop(skb, sch, to_free); } /* Return id of the bucket from which the packet was dropped. */ return bucket - q->buckets; } -static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct hhf_sched_data *q = qdisc_priv(sch); enum wdrr_bucket_idx idx; @@ -406,7 +406,7 @@ static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch) /* Return Congestion Notification only if we dropped a packet from this * bucket. */ - if (hhf_drop(sch) == idx) + if (hhf_drop(sch, to_free) == idx) return NET_XMIT_CN; /* As we dropped a packet, better let upper stack know this. */ diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index a454605..f388225 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -569,7 +569,8 @@ static inline void htb_deactivate(struct htb_sched *q, struct htb_class *cl) list_del_init(&cl->un.leaf.drop_list); } -static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { int uninitialized_var(ret); struct htb_sched *q = qdisc_priv(sch); @@ -581,16 +582,17 @@ static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch) __skb_queue_tail(&q->direct_queue, skb); q->direct_pkts++; } else { - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); } #ifdef CONFIG_NET_CLS_ACT } else if (!cl) { if (ret & __NET_XMIT_BYPASS) qdisc_qstats_drop(sch); - kfree_skb(skb); + __qdisc_drop(skb, to_free); return ret; #endif - } else if ((ret = qdisc_enqueue(skb, cl->un.leaf.q)) != NET_XMIT_SUCCESS) { + } else if ((ret = qdisc_enqueue(skb, cl->un.leaf.q, + to_free)) != NET_XMIT_SUCCESS) { if (net_xmit_drop_count(ret)) { qdisc_qstats_drop(sch); cl->qstats.drops++; diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c index 5ea9330..9ffbb02 100644 --- a/net/sched/sch_multiq.c +++ b/net/sched/sch_multiq.c @@ -65,7 +65,8 @@ multiq_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) } static int -multiq_enqueue(struct sk_buff *skb, struct Qdisc *sch) +multiq_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct Qdisc *qdisc; int ret; @@ -76,12 +77,12 @@ multiq_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (ret & __NET_XMIT_BYPASS) qdisc_qstats_drop(sch); - kfree_skb(skb); + __qdisc_drop(skb, to_free); return ret; } #endif - ret = qdisc_enqueue(skb, qdisc); + ret = qdisc_enqueue(skb, qdisc, to_free); if (ret == NET_XMIT_SUCCESS) { sch->q.qlen++; return NET_XMIT_SUCCESS; diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index e271967..ccca8ca 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -397,7 +397,8 @@ static void tfifo_enqueue(struct sk_buff *nskb, struct Qdisc *sch) * when we statistically choose to corrupt one, we instead segment it, returning * the first packet to be corrupted, and re-enqueue the remaining frames */ -static struct sk_buff *netem_segment(struct sk_buff *skb, struct Qdisc *sch) +static struct sk_buff *netem_segment(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct sk_buff *segs; netdev_features_t features = netif_skb_features(skb); @@ -405,7 +406,7 @@ static struct sk_buff *netem_segment(struct sk_buff *skb, struct Qdisc *sch) segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); if (IS_ERR_OR_NULL(segs)) { - qdisc_drop(skb, sch); + qdisc_drop(skb, sch, to_free); return NULL; } consume_skb(skb); @@ -418,7 +419,8 @@ static struct sk_buff *netem_segment(struct sk_buff *skb, struct Qdisc *sch) * NET_XMIT_DROP: queue length didn't change. * NET_XMIT_SUCCESS: one skb was queued. */ -static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct netem_sched_data *q = qdisc_priv(sch); /* We don't fill cb now as skb_unshare() may invalidate it */ @@ -443,7 +445,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) } if (count == 0) { qdisc_qstats_drop(sch); - kfree_skb(skb); + __qdisc_drop(skb, to_free); return NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; } @@ -463,7 +465,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) u32 dupsave = q->duplicate; /* prevent duplicating a dup... */ q->duplicate = 0; - rootq->enqueue(skb2, rootq); + rootq->enqueue(skb2, rootq, to_free); q->duplicate = dupsave; } @@ -475,7 +477,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) */ if (q->corrupt && q->corrupt >= get_crandom(&q->corrupt_cor)) { if (skb_is_gso(skb)) { - segs = netem_segment(skb, sch); + segs = netem_segment(skb, sch, to_free); if (!segs) return NET_XMIT_DROP; } else { @@ -488,7 +490,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (!(skb = skb_unshare(skb, GFP_ATOMIC)) || (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_help(skb))) { - rc = qdisc_drop(skb, sch); + rc = qdisc_drop(skb, sch, to_free); goto finish_segs; } @@ -497,7 +499,7 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch) } if (unlikely(skb_queue_len(&sch->q) >= sch->limit)) - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); qdisc_qstats_backlog_inc(sch, skb); @@ -557,7 +559,7 @@ finish_segs: segs->next = NULL; qdisc_skb_cb(segs)->pkt_len = segs->len; last_len = segs->len; - rc = qdisc_enqueue(segs, sch); + rc = qdisc_enqueue(segs, sch, to_free); if (rc != NET_XMIT_SUCCESS) { if (net_xmit_drop_count(rc)) qdisc_qstats_drop(sch); @@ -615,8 +617,11 @@ deliver: #endif if (q->qdisc) { - int err = qdisc_enqueue(skb, q->qdisc); + struct sk_buff *to_free = NULL; + int err; + err = qdisc_enqueue(skb, q->qdisc, &to_free); + kfree_skb_list(to_free); if (unlikely(err != NET_XMIT_SUCCESS)) { if (net_xmit_drop_count(err)) { qdisc_qstats_drop(sch); diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c index 912a46a..a570b0b 100644 --- a/net/sched/sch_pie.c +++ b/net/sched/sch_pie.c @@ -134,7 +134,8 @@ static bool drop_early(struct Qdisc *sch, u32 packet_size) return false; } -static int pie_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int pie_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct pie_sched_data *q = qdisc_priv(sch); bool enqueue = false; @@ -166,7 +167,7 @@ static int pie_qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch) out: q->stats.dropped++; - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); } static const struct nla_policy pie_policy[TCA_PIE_MAX + 1] = { diff --git a/net/sched/sch_plug.c b/net/sched/sch_plug.c index a12cd37..1c6cbab 100644 --- a/net/sched/sch_plug.c +++ b/net/sched/sch_plug.c @@ -88,7 +88,8 @@ struct plug_sched_data { u32 pkts_to_release; }; -static int plug_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int plug_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct plug_sched_data *q = qdisc_priv(sch); @@ -98,7 +99,7 @@ static int plug_enqueue(struct sk_buff *skb, struct Qdisc *sch) return qdisc_enqueue_tail(skb, sch); } - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); } static struct sk_buff *plug_dequeue(struct Qdisc *sch) diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index de49268..f4d443a 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -67,7 +67,7 @@ prio_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) } static int -prio_enqueue(struct sk_buff *skb, struct Qdisc *sch) +prio_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) { struct Qdisc *qdisc; int ret; @@ -83,7 +83,7 @@ prio_enqueue(struct sk_buff *skb, struct Qdisc *sch) } #endif - ret = qdisc_enqueue(skb, qdisc); + ret = qdisc_enqueue(skb, qdisc, to_free); if (ret == NET_XMIT_SUCCESS) { qdisc_qstats_backlog_inc(sch, skb); sch->q.qlen++; diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 0427fa8..f27ffee 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -1217,7 +1217,8 @@ static struct qfq_aggregate *qfq_choose_next_agg(struct qfq_sched *q) return agg; } -static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct qfq_sched *q = qdisc_priv(sch); struct qfq_class *cl; @@ -1240,11 +1241,11 @@ static int qfq_enqueue(struct sk_buff *skb, struct Qdisc *sch) qdisc_pkt_len(skb)); if (err) { cl->qstats.drops++; - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); } } - err = qdisc_enqueue(skb, cl->qdisc); + err = qdisc_enqueue(skb, cl->qdisc, to_free); if (unlikely(err != NET_XMIT_SUCCESS)) { pr_debug("qfq_enqueue: enqueue failed %d\n", err); if (net_xmit_drop_count(err)) { diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c index a0d5753..249b2a1 100644 --- a/net/sched/sch_red.c +++ b/net/sched/sch_red.c @@ -56,7 +56,8 @@ static inline int red_use_harddrop(struct red_sched_data *q) return q->flags & TC_RED_HARDDROP; } -static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct red_sched_data *q = qdisc_priv(sch); struct Qdisc *child = q->qdisc; @@ -95,7 +96,7 @@ static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch) break; } - ret = qdisc_enqueue(skb, child); + ret = qdisc_enqueue(skb, child, to_free); if (likely(ret == NET_XMIT_SUCCESS)) { qdisc_qstats_backlog_inc(sch, skb); sch->q.qlen++; @@ -106,7 +107,7 @@ static int red_enqueue(struct sk_buff *skb, struct Qdisc *sch) return ret; congestion_drop: - qdisc_drop(skb, sch); + qdisc_drop(skb, sch, to_free); return NET_XMIT_CN; } diff --git a/net/sched/sch_sfb.c b/net/sched/sch_sfb.c index c696116..add3cc7 100644 --- a/net/sched/sch_sfb.c +++ b/net/sched/sch_sfb.c @@ -275,7 +275,8 @@ static bool sfb_classify(struct sk_buff *skb, struct tcf_proto *fl, return false; } -static int sfb_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int sfb_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct sfb_sched_data *q = qdisc_priv(sch); @@ -397,7 +398,7 @@ static int sfb_enqueue(struct sk_buff *skb, struct Qdisc *sch) } enqueue: - ret = qdisc_enqueue(skb, child); + ret = qdisc_enqueue(skb, child, to_free); if (likely(ret == NET_XMIT_SUCCESS)) { sch->q.qlen++; increment_qlen(skb, q); @@ -408,7 +409,7 @@ enqueue: return ret; drop: - qdisc_drop(skb, sch); + qdisc_drop(skb, sch, to_free); return NET_XMIT_CN; other_drop: if (ret & __NET_XMIT_BYPASS) diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 57d118b..7f195ed 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -343,7 +343,7 @@ static int sfq_headdrop(const struct sfq_sched_data *q) } static int -sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch) +sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) { struct sfq_sched_data *q = qdisc_priv(sch); unsigned int hash, dropped; @@ -367,7 +367,7 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (x == SFQ_EMPTY_SLOT) { x = q->dep[0].next; /* get a free slot */ if (x >= SFQ_MAX_FLOWS) - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); q->ht[hash] = x; slot = &q->slots[x]; slot->hash = hash; @@ -424,14 +424,14 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch) if (slot->qlen >= q->maxdepth) { congestion_drop: if (!sfq_headdrop(q)) - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); /* We know we have at least one packet in queue */ head = slot_dequeue_head(slot); delta = qdisc_pkt_len(head) - qdisc_pkt_len(skb); sch->qstats.backlog -= delta; slot->backlog -= delta; - qdisc_drop(head, sch); + qdisc_drop(head, sch, to_free); slot_queue_add(slot, skb); return NET_XMIT_CN; diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c index c12df84..303355c 100644 --- a/net/sched/sch_tbf.c +++ b/net/sched/sch_tbf.c @@ -155,7 +155,8 @@ static unsigned int skb_gso_mac_seglen(const struct sk_buff *skb) /* GSO packet is too big, segment it so that tbf can transmit * each segment in time */ -static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch) +static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct tbf_sched_data *q = qdisc_priv(sch); struct sk_buff *segs, *nskb; @@ -166,7 +167,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch) segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); if (IS_ERR_OR_NULL(segs)) - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); nb = 0; while (segs) { @@ -174,7 +175,7 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch) segs->next = NULL; qdisc_skb_cb(segs)->pkt_len = segs->len; len += segs->len; - ret = qdisc_enqueue(segs, q->qdisc); + ret = qdisc_enqueue(segs, q->qdisc, to_free); if (ret != NET_XMIT_SUCCESS) { if (net_xmit_drop_count(ret)) qdisc_qstats_drop(sch); @@ -190,17 +191,18 @@ static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch) return nb > 0 ? NET_XMIT_SUCCESS : NET_XMIT_DROP; } -static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch) +static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff **to_free) { struct tbf_sched_data *q = qdisc_priv(sch); int ret; if (qdisc_pkt_len(skb) > q->max_size) { if (skb_is_gso(skb) && skb_gso_mac_seglen(skb) <= q->max_size) - return tbf_segment(skb, sch); - return qdisc_drop(skb, sch); + return tbf_segment(skb, sch, to_free); + return qdisc_drop(skb, sch, to_free); } - ret = qdisc_enqueue(skb, q->qdisc); + ret = qdisc_enqueue(skb, q->qdisc, to_free); if (ret != NET_XMIT_SUCCESS) { if (net_xmit_drop_count(ret)) qdisc_qstats_drop(sch); diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index e026871..2cd9b44 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -77,7 +77,7 @@ struct teql_sched_data { /* "teql*" qdisc routines */ static int -teql_enqueue(struct sk_buff *skb, struct Qdisc *sch) +teql_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) { struct net_device *dev = qdisc_dev(sch); struct teql_sched_data *q = qdisc_priv(sch); @@ -87,7 +87,7 @@ teql_enqueue(struct sk_buff *skb, struct Qdisc *sch) return NET_XMIT_SUCCESS; } - return qdisc_drop(skb, sch); + return qdisc_drop(skb, sch, to_free); } static struct sk_buff * -- cgit v0.10.2 From 008830bc321c0fc22c0db8d5b0b56f854ed90a5c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 21 Jun 2016 23:16:50 -0700 Subject: net_sched: fq_codel: cache skb->truesize into skb->cb Now we defer skb drops, it makes sense to keep a copy of skb->truesize in struct codel_skb_cb to avoid one cache line miss per dropped skb in fq_codel_drop(), to reduce latencies a bit further. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/include/net/codel_qdisc.h b/include/net/codel_qdisc.h index 8144d9c..098630f 100644 --- a/include/net/codel_qdisc.h +++ b/include/net/codel_qdisc.h @@ -52,6 +52,7 @@ /* Qdiscs using codel plugin must use codel_skb_cb in their own cb[] */ struct codel_skb_cb { codel_time_t enqueue_time; + unsigned int mem_usage; }; static struct codel_skb_cb *get_codel_cb(const struct sk_buff *skb) diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index f715195..a5ea0e9 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -172,7 +172,7 @@ static unsigned int fq_codel_drop(struct Qdisc *sch, unsigned int max_packets, do { skb = dequeue_head(flow); len += qdisc_pkt_len(skb); - mem += skb->truesize; + mem += get_codel_cb(skb)->mem_usage; __qdisc_drop(skb, to_free); } while (++i < max_packets && len < threshold); @@ -216,7 +216,8 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch, flow->deficit = q->quantum; flow->dropped = 0; } - q->memory_usage += skb->truesize; + get_codel_cb(skb)->mem_usage = skb->truesize; + q->memory_usage += get_codel_cb(skb)->mem_usage; memory_limited = q->memory_usage > q->memory_limit; if (++sch->q.qlen <= sch->limit && !memory_limited) return NET_XMIT_SUCCESS; @@ -267,7 +268,7 @@ static struct sk_buff *dequeue_func(struct codel_vars *vars, void *ctx) if (flow->head) { skb = dequeue_head(flow); q->backlogs[flow - q->flows] -= qdisc_pkt_len(skb); - q->memory_usage -= skb->truesize; + q->memory_usage -= get_codel_cb(skb)->mem_usage; sch->q.qlen--; sch->qstats.backlog -= qdisc_pkt_len(skb); } -- cgit v0.10.2 From 338ed9b4de57c4b7965b96364593e27de0d89582 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 21 Jun 2016 23:16:51 -0700 Subject: net_sched: sch_htb: export class backlog in dumps We already get child qdisc qlen, we also can get its backlog so that class dumps can report it. Also replace qstats by a single drop counter, but move it in a separate cache line so that drops do not dirty useful cache lines. Tested: $ tc -s cl sh dev eth0 class htb 1:1 root leaf 3: prio 0 rate 1Gbit ceil 1Gbit burst 500000b cburst 500000b Sent 2183346912 bytes 9021815 pkt (dropped 2340774, overlimits 0 requeues 0) rate 1001Mbit 517543pps backlog 120758b 499p requeues 0 lended: 9021770 borrowed: 0 giants: 0 tokens: 9 ctokens: 9 Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index f388225..ba098f2 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -117,7 +117,6 @@ struct htb_class { * Written often fields */ struct gnet_stats_basic_packed bstats; - struct gnet_stats_queue qstats; struct tc_htb_xstats xstats; /* our special stats */ /* token bucket parameters */ @@ -140,6 +139,8 @@ struct htb_class { enum htb_cmode cmode; /* current mode of the class */ struct rb_node pq_node; /* node for event queue */ struct rb_node node[TC_HTB_NUMPRIO]; /* node for self or feed tree */ + + unsigned int drops ____cacheline_aligned_in_smp; }; struct htb_level { @@ -595,7 +596,7 @@ static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch, to_free)) != NET_XMIT_SUCCESS) { if (net_xmit_drop_count(ret)) { qdisc_qstats_drop(sch); - cl->qstats.drops++; + cl->drops++; } return ret; } else { @@ -1110,17 +1111,22 @@ static int htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d) { struct htb_class *cl = (struct htb_class *)arg; + struct gnet_stats_queue qs = { + .drops = cl->drops, + }; __u32 qlen = 0; - if (!cl->level && cl->un.leaf.q) + if (!cl->level && cl->un.leaf.q) { qlen = cl->un.leaf.q->q.qlen; + qs.backlog = cl->un.leaf.q->qstats.backlog; + } cl->xstats.tokens = PSCHED_NS2TICKS(cl->tokens); cl->xstats.ctokens = PSCHED_NS2TICKS(cl->ctokens); if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), d, NULL, &cl->bstats) < 0 || gnet_stats_copy_rate_est(d, NULL, &cl->rate_est) < 0 || - gnet_stats_copy_queue(d, NULL, &cl->qstats, qlen) < 0) + gnet_stats_copy_queue(d, NULL, &qs, qlen) < 0) return -1; return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats)); -- cgit v0.10.2 From 4d202a0d31b96ab3324b21e7500d9a2da9ef57dd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 21 Jun 2016 23:16:52 -0700 Subject: net_sched: generalize bulk dequeue When qdisc bulk dequeue was added in linux-3.18 (commit 5772e9a3463b "qdisc: bulk dequeue support for qdiscs with TCQ_F_ONETXQUEUE"), it was constrained to some specific qdiscs. With some extra care, we can extend this to all qdiscs, so that typical traffic shaping solutions can benefit from small batches (8 packets in this patch). For example, HTB is often used on some multi queue device. And bonding/team are multi queue devices... Idea is to bulk-dequeue packets mapping to the same transmit queue. This brings between 35 and 80 % performance increase in HTB setup under pressure on a bonding setup : 1) NUMA node contention : 610,000 pps -> 1,110,000 pps 2) No node contention : 1,380,000 pps -> 1,930,000 pps Now we should work to add batches on the enqueue() side ;) Signed-off-by: Eric Dumazet Cc: John Fastabend Cc: Jesper Dangaard Brouer Cc: Hannes Frederic Sowa Cc: Florian Westphal Cc: Daniel Borkmann Acked-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 04e84c0..909aff2 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -75,13 +75,14 @@ struct Qdisc { /* * For performance sake on SMP, we put highly modified fields at the end */ - struct Qdisc *next_sched ____cacheline_aligned_in_smp; - struct sk_buff *gso_skb; - unsigned long state; + struct sk_buff *gso_skb ____cacheline_aligned_in_smp; struct sk_buff_head q; struct gnet_stats_basic_packed bstats; seqcount_t running; struct gnet_stats_queue qstats; + unsigned long state; + struct Qdisc *next_sched; + struct sk_buff *skb_bad_txq; struct rcu_head rcu_head; int padded; atomic_t refcnt; diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index ff86606..e95b67c 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -77,6 +77,34 @@ static void try_bulk_dequeue_skb(struct Qdisc *q, skb->next = NULL; } +/* This variant of try_bulk_dequeue_skb() makes sure + * all skbs in the chain are for the same txq + */ +static void try_bulk_dequeue_skb_slow(struct Qdisc *q, + struct sk_buff *skb, + int *packets) +{ + int mapping = skb_get_queue_mapping(skb); + struct sk_buff *nskb; + int cnt = 0; + + do { + nskb = q->dequeue(q); + if (!nskb) + break; + if (unlikely(skb_get_queue_mapping(nskb) != mapping)) { + q->skb_bad_txq = nskb; + qdisc_qstats_backlog_inc(q, nskb); + q->q.qlen++; + break; + } + skb->next = nskb; + skb = nskb; + } while (++cnt < 8); + (*packets) += cnt; + skb->next = NULL; +} + /* Note that dequeue_skb can possibly return a SKB list (via skb->next). * A requeued skb (via q->gso_skb) can also be a SKB list. */ @@ -87,8 +115,9 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate, const struct netdev_queue *txq = q->dev_queue; *packets = 1; - *validate = true; if (unlikely(skb)) { + /* skb in gso_skb were already validated */ + *validate = false; /* check the reason of requeuing without tx lock first */ txq = skb_get_tx_queue(txq->dev, skb); if (!netif_xmit_frozen_or_stopped(txq)) { @@ -97,15 +126,30 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate, q->q.qlen--; } else skb = NULL; - /* skb in gso_skb were already validated */ - *validate = false; - } else { - if (!(q->flags & TCQ_F_ONETXQUEUE) || - !netif_xmit_frozen_or_stopped(txq)) { - skb = q->dequeue(q); - if (skb && qdisc_may_bulk(q)) - try_bulk_dequeue_skb(q, skb, txq, packets); + return skb; + } + *validate = true; + skb = q->skb_bad_txq; + if (unlikely(skb)) { + /* check the reason of requeuing without tx lock first */ + txq = skb_get_tx_queue(txq->dev, skb); + if (!netif_xmit_frozen_or_stopped(txq)) { + q->skb_bad_txq = NULL; + qdisc_qstats_backlog_dec(q, skb); + q->q.qlen--; + goto bulk; } + return NULL; + } + if (!(q->flags & TCQ_F_ONETXQUEUE) || + !netif_xmit_frozen_or_stopped(txq)) + skb = q->dequeue(q); + if (skb) { +bulk: + if (qdisc_may_bulk(q)) + try_bulk_dequeue_skb(q, skb, txq, packets); + else + try_bulk_dequeue_skb_slow(q, skb, packets); } return skb; } @@ -624,11 +668,14 @@ void qdisc_reset(struct Qdisc *qdisc) if (ops->reset) ops->reset(qdisc); + kfree_skb(qdisc->skb_bad_txq); + qdisc->skb_bad_txq = NULL; + if (qdisc->gso_skb) { kfree_skb_list(qdisc->gso_skb); qdisc->gso_skb = NULL; - qdisc->q.qlen = 0; } + qdisc->q.qlen = 0; } EXPORT_SYMBOL(qdisc_reset); @@ -667,6 +714,7 @@ void qdisc_destroy(struct Qdisc *qdisc) dev_put(qdisc_dev(qdisc)); kfree_skb_list(qdisc->gso_skb); + kfree_skb(qdisc->skb_bad_txq); /* * gen_estimator est_timer() might access qdisc->q.lock, * wait a RCU grace period before freeing qdisc. -- cgit v0.10.2 From e7ffd81233334b7755050523cb7e0456ae3d2e53 Mon Sep 17 00:00:00 2001 From: Xing Zheng Date: Tue, 21 Jun 2016 20:33:28 +0800 Subject: net: stmmac: dwmac-rk: add rk3228-specific data Add constants and callback functions for the dwmac on rk3228/rk3229 socs. As can be seen, the base structure is the same, only registers and the bits in them moved slightly. Signed-off-by: Xing Zheng Reviewed-by: Heiko Stuebner Acked-by: Rob Herring Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/rockchip-dwmac.txt b/Documentation/devicetree/bindings/net/rockchip-dwmac.txt index 93eac7c..cccd945 100644 --- a/Documentation/devicetree/bindings/net/rockchip-dwmac.txt +++ b/Documentation/devicetree/bindings/net/rockchip-dwmac.txt @@ -3,7 +3,8 @@ Rockchip SoC RK3288 10/100/1000 Ethernet driver(GMAC) The device node has following properties. Required properties: - - compatible: Can be one of "rockchip,rk3288-gmac", "rockchip,rk3368-gmac" + - compatible: Can be one of "rockchip,rk3228-gmac", "rockchip,rk3288-gmac", + "rockchip,rk3368-gmac" - reg: addresses and length of the register sets for the device. - interrupts: Should contain the GMAC interrupts. - interrupt-names: Should contain the interrupt names "macirq". diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c index 63c2e4f..9210591 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-rk.c @@ -73,6 +73,122 @@ struct rk_priv_data { #define GRF_BIT(nr) (BIT(nr) | BIT(nr+16)) #define GRF_CLR_BIT(nr) (BIT(nr+16)) +#define RK3228_GRF_MAC_CON0 0x0900 +#define RK3228_GRF_MAC_CON1 0x0904 + +/* RK3228_GRF_MAC_CON0 */ +#define RK3228_GMAC_CLK_RX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 7) +#define RK3228_GMAC_CLK_TX_DL_CFG(val) HIWORD_UPDATE(val, 0x7F, 0) + +/* RK3228_GRF_MAC_CON1 */ +#define RK3228_GMAC_PHY_INTF_SEL_RGMII \ + (GRF_BIT(4) | GRF_CLR_BIT(5) | GRF_CLR_BIT(6)) +#define RK3228_GMAC_PHY_INTF_SEL_RMII \ + (GRF_CLR_BIT(4) | GRF_CLR_BIT(5) | GRF_BIT(6)) +#define RK3228_GMAC_FLOW_CTRL GRF_BIT(3) +#define RK3228_GMAC_FLOW_CTRL_CLR GRF_CLR_BIT(3) +#define RK3228_GMAC_SPEED_10M GRF_CLR_BIT(2) +#define RK3228_GMAC_SPEED_100M GRF_BIT(2) +#define RK3228_GMAC_RMII_CLK_25M GRF_BIT(7) +#define RK3228_GMAC_RMII_CLK_2_5M GRF_CLR_BIT(7) +#define RK3228_GMAC_CLK_125M (GRF_CLR_BIT(8) | GRF_CLR_BIT(9)) +#define RK3228_GMAC_CLK_25M (GRF_BIT(8) | GRF_BIT(9)) +#define RK3228_GMAC_CLK_2_5M (GRF_CLR_BIT(8) | GRF_BIT(9)) +#define RK3228_GMAC_RMII_MODE GRF_BIT(10) +#define RK3228_GMAC_RMII_MODE_CLR GRF_CLR_BIT(10) +#define RK3228_GMAC_TXCLK_DLY_ENABLE GRF_BIT(0) +#define RK3228_GMAC_TXCLK_DLY_DISABLE GRF_CLR_BIT(0) +#define RK3228_GMAC_RXCLK_DLY_ENABLE GRF_BIT(1) +#define RK3228_GMAC_RXCLK_DLY_DISABLE GRF_CLR_BIT(1) + +static void rk3228_set_to_rgmii(struct rk_priv_data *bsp_priv, + int tx_delay, int rx_delay) +{ + struct device *dev = &bsp_priv->pdev->dev; + + if (IS_ERR(bsp_priv->grf)) { + dev_err(dev, "Missing rockchip,grf property\n"); + return; + } + + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_PHY_INTF_SEL_RGMII | + RK3228_GMAC_RMII_MODE_CLR | + RK3228_GMAC_RXCLK_DLY_ENABLE | + RK3228_GMAC_TXCLK_DLY_ENABLE); + + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON0, + RK3228_GMAC_CLK_RX_DL_CFG(rx_delay) | + RK3228_GMAC_CLK_TX_DL_CFG(tx_delay)); +} + +static void rk3228_set_to_rmii(struct rk_priv_data *bsp_priv) +{ + struct device *dev = &bsp_priv->pdev->dev; + + if (IS_ERR(bsp_priv->grf)) { + dev_err(dev, "Missing rockchip,grf property\n"); + return; + } + + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_PHY_INTF_SEL_RMII | + RK3228_GMAC_RMII_MODE); + + /* set MAC to RMII mode */ + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, GRF_BIT(11)); +} + +static void rk3228_set_rgmii_speed(struct rk_priv_data *bsp_priv, int speed) +{ + struct device *dev = &bsp_priv->pdev->dev; + + if (IS_ERR(bsp_priv->grf)) { + dev_err(dev, "Missing rockchip,grf property\n"); + return; + } + + if (speed == 10) + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_CLK_2_5M); + else if (speed == 100) + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_CLK_25M); + else if (speed == 1000) + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_CLK_125M); + else + dev_err(dev, "unknown speed value for RGMII! speed=%d", speed); +} + +static void rk3228_set_rmii_speed(struct rk_priv_data *bsp_priv, int speed) +{ + struct device *dev = &bsp_priv->pdev->dev; + + if (IS_ERR(bsp_priv->grf)) { + dev_err(dev, "Missing rockchip,grf property\n"); + return; + } + + if (speed == 10) + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_RMII_CLK_2_5M | + RK3228_GMAC_SPEED_10M); + else if (speed == 100) + regmap_write(bsp_priv->grf, RK3228_GRF_MAC_CON1, + RK3228_GMAC_RMII_CLK_25M | + RK3228_GMAC_SPEED_100M); + else + dev_err(dev, "unknown speed value for RMII! speed=%d", speed); +} + +static const struct rk_gmac_ops rk3228_ops = { + .set_to_rgmii = rk3228_set_to_rgmii, + .set_to_rmii = rk3228_set_to_rmii, + .set_rgmii_speed = rk3228_set_rgmii_speed, + .set_rmii_speed = rk3228_set_rmii_speed, +}; + #define RK3288_GRF_SOC_CON1 0x0248 #define RK3288_GRF_SOC_CON3 0x0250 @@ -642,6 +758,7 @@ static int rk_gmac_probe(struct platform_device *pdev) } static const struct of_device_id rk_gmac_dwmac_match[] = { + { .compatible = "rockchip,rk3228-gmac", .data = &rk3228_ops }, { .compatible = "rockchip,rk3288-gmac", .data = &rk3288_ops }, { .compatible = "rockchip,rk3368-gmac", .data = &rk3368_ops }, { } -- cgit v0.10.2 From ff7566b8d71fd28123a152d05d0726b81e86ef06 Mon Sep 17 00:00:00 2001 From: Fugang Duan Date: Wed, 22 Jun 2016 18:52:35 +0800 Subject: net: fec: add interrupt coalesc quirk flag Different i.MX SOC FEC support different features like : - i.MX6Q/DL FEC does not support AVB and interrupt coalesc - i.MX6SX/i.MX7D supports AVB and interrupt coalesc - i.MX6UL/ULL does not support AVB, but support interrupt coalesc So, add new quirk flag to judge the supported features. Signed-off-by: Fugang Duan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h index f58f9ea..92fd5c0 100644 --- a/drivers/net/ethernet/freescale/fec.h +++ b/drivers/net/ethernet/freescale/fec.h @@ -442,6 +442,8 @@ struct bufdesc_ex { #define FEC_QUIRK_SINGLE_MDIO (1 << 11) /* Controller supports RACC register */ #define FEC_QUIRK_HAS_RACC (1 << 12) +/* Controller supports interrupt coalesc */ +#define FEC_QUIRK_HAS_COALESCE (1 << 13) struct bufdesc_prop { int qid; diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index c36401e..8afef20 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -111,7 +111,7 @@ static struct platform_device_id fec_devtype[] = { FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM | FEC_QUIRK_HAS_VLAN | FEC_QUIRK_HAS_AVB | FEC_QUIRK_ERR007885 | FEC_QUIRK_BUG_CAPTURE | - FEC_QUIRK_HAS_RACC, + FEC_QUIRK_HAS_RACC | FEC_QUIRK_HAS_COALESCE, }, { /* sentinel */ } @@ -2358,9 +2358,6 @@ static void fec_enet_itr_coal_set(struct net_device *ndev) struct fec_enet_private *fep = netdev_priv(ndev); int rx_itr, tx_itr; - if (!(fep->quirks & FEC_QUIRK_HAS_AVB)) - return; - /* Must be greater than zero to avoid unpredictable behavior */ if (!fep->rx_time_itr || !fep->rx_pkts_itr || !fep->tx_time_itr || !fep->tx_pkts_itr) @@ -2383,10 +2380,12 @@ static void fec_enet_itr_coal_set(struct net_device *ndev) writel(tx_itr, fep->hwp + FEC_TXIC0); writel(rx_itr, fep->hwp + FEC_RXIC0); - writel(tx_itr, fep->hwp + FEC_TXIC1); - writel(rx_itr, fep->hwp + FEC_RXIC1); - writel(tx_itr, fep->hwp + FEC_TXIC2); - writel(rx_itr, fep->hwp + FEC_RXIC2); + if (fep->quirks & FEC_QUIRK_HAS_AVB) { + writel(tx_itr, fep->hwp + FEC_TXIC1); + writel(rx_itr, fep->hwp + FEC_RXIC1); + writel(tx_itr, fep->hwp + FEC_TXIC2); + writel(rx_itr, fep->hwp + FEC_RXIC2); + } } static int @@ -2394,7 +2393,7 @@ fec_enet_get_coalesce(struct net_device *ndev, struct ethtool_coalesce *ec) { struct fec_enet_private *fep = netdev_priv(ndev); - if (!(fep->quirks & FEC_QUIRK_HAS_AVB)) + if (!(fep->quirks & FEC_QUIRK_HAS_COALESCE)) return -EOPNOTSUPP; ec->rx_coalesce_usecs = fep->rx_time_itr; @@ -2412,7 +2411,7 @@ fec_enet_set_coalesce(struct net_device *ndev, struct ethtool_coalesce *ec) struct fec_enet_private *fep = netdev_priv(ndev); unsigned int cycle; - if (!(fep->quirks & FEC_QUIRK_HAS_AVB)) + if (!(fep->quirks & FEC_QUIRK_HAS_COALESCE)) return -EOPNOTSUPP; if (ec->rx_max_coalesced_frames > 255) { -- cgit v0.10.2 From a51d3ab50702a8d846207151d33b2439a3661bd3 Mon Sep 17 00:00:00 2001 From: Fugang Duan Date: Wed, 22 Jun 2016 18:52:36 +0800 Subject: net: fec: use a more proper compatible string for i.MX6UL type device i.MX6UL is a member in i.MX series family, the SOC FEC inherits from i.MX6SX but removes some IP features, lets define a new type for fec device. Signed-off-by: Fugang Duan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 8afef20..4040003 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -113,6 +113,12 @@ static struct platform_device_id fec_devtype[] = { FEC_QUIRK_ERR007885 | FEC_QUIRK_BUG_CAPTURE | FEC_QUIRK_HAS_RACC | FEC_QUIRK_HAS_COALESCE, }, { + .name = "imx6ul-fec", + .driver_data = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT | + FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM | + FEC_QUIRK_HAS_VLAN | FEC_QUIRK_BUG_CAPTURE | + FEC_QUIRK_HAS_RACC | FEC_QUIRK_HAS_COALESCE, + }, { /* sentinel */ } }; @@ -125,6 +131,7 @@ enum imx_fec_type { IMX6Q_FEC, MVF600_FEC, IMX6SX_FEC, + IMX6UL_FEC, }; static const struct of_device_id fec_dt_ids[] = { @@ -134,6 +141,7 @@ static const struct of_device_id fec_dt_ids[] = { { .compatible = "fsl,imx6q-fec", .data = &fec_devtype[IMX6Q_FEC], }, { .compatible = "fsl,mvf600-fec", .data = &fec_devtype[MVF600_FEC], }, { .compatible = "fsl,imx6sx-fec", .data = &fec_devtype[IMX6SX_FEC], }, + { .compatible = "fsl,imx6ul-fec", .data = &fec_devtype[IMX6UL_FEC], }, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, fec_dt_ids); -- cgit v0.10.2 From ce7faf0a071b6d05f55d8b7b36fd796cab527427 Mon Sep 17 00:00:00 2001 From: Sathya Perla Date: Wed, 22 Jun 2016 08:54:53 -0400 Subject: be2net: fix definition of be_max_eqs() The EQs available on a function are shared between NIC and RoCE. The be_max_eqs() macro was so far being used to refer to the max number of EQs available for NIC. This has caused some confusion in the code. To fix this confusion this patch introduces a new macro called be_max_nic_eqs() to refer to the max number of EQs avialable for NIC only and renames be_max_eqs() to be_max_func_eqs(). Signed-off-by: Sathya Perla Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index e1e6c40..f8040ef 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -443,6 +443,7 @@ struct be_resources { u16 max_iface_count; u16 max_mcc_count; u16 max_evt_qs; + u16 max_nic_evt_qs; /* NIC's share of evt qs */ u32 if_cap_flags; u32 vf_if_cap_flags; /* VF if capability flags */ u32 flags; @@ -644,7 +645,10 @@ struct be_adapter { #define be_max_txqs(adapter) (adapter->res.max_tx_qs) #define be_max_prio_txqs(adapter) (adapter->res.max_prio_tx_qs) #define be_max_rxqs(adapter) (adapter->res.max_rx_qs) -#define be_max_eqs(adapter) (adapter->res.max_evt_qs) +/* Max number of EQs available for the function (NIC + RoCE (if enabled)) */ +#define be_max_func_eqs(adapter) (adapter->res.max_evt_qs) +/* Max number of EQs available avaialble only for NIC */ +#define be_max_nic_eqs(adapter) (adapter->res.max_nic_evt_qs) #define be_if_cap_flags(adapter) (adapter->res.if_cap_flags) #define be_max_pf_pool_rss_tables(adapter) \ (adapter->pool_res.max_rss_tables) @@ -654,7 +658,7 @@ static inline u16 be_max_qs(struct be_adapter *adapter) /* If no RSS, need atleast the one def RXQ */ u16 num = max_t(u16, be_max_rss(adapter), 1); - num = min(num, be_max_eqs(adapter)); + num = min(num, be_max_nic_eqs(adapter)); return min_t(u16, num, num_online_cpus()); } diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 3d94789..c67830f 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -3252,12 +3252,12 @@ static int be_msix_enable(struct be_adapter *adapter) int i, num_vec; struct device *dev = &adapter->pdev->dev; - /* If RoCE is supported, program the max number of NIC vectors that - * may be configured via set-channels, along with vectors needed for - * RoCe. Else, just program the number we'll use initially. + /* If RoCE is supported, program the max number of vectors that + * could be used for NIC and RoCE, else, just program the number + * we'll use initially. */ if (be_roce_supported(adapter)) - num_vec = min_t(int, 2 * be_max_eqs(adapter), + num_vec = min_t(int, be_max_func_eqs(adapter), 2 * num_online_cpus()); else num_vec = adapter->cfg_num_qs; @@ -4219,16 +4219,13 @@ static int be_get_resources(struct be_adapter *adapter) struct be_resources res = {0}; int status; - if (BEx_chip(adapter)) { - BEx_get_resources(adapter, &res); - adapter->res = res; - } - /* For Lancer, SH etc read per-function resource limits from FW. * GET_FUNC_CONFIG returns per function guaranteed limits. * GET_PROFILE_CONFIG returns PCI-E related limits PF-pool limits */ - if (!BEx_chip(adapter)) { + if (BEx_chip(adapter)) { + BEx_get_resources(adapter, &res); + } else { status = be_cmd_get_func_config(adapter, &res); if (status) return status; @@ -4237,13 +4234,13 @@ static int be_get_resources(struct be_adapter *adapter) if (res.max_rss_qs && res.max_rss_qs == res.max_rx_qs && !(res.if_cap_flags & BE_IF_FLAGS_DEFQ_RSS)) res.max_rss_qs -= 1; - - /* If RoCE may be enabled stash away half the EQs for RoCE */ - if (be_roce_supported(adapter)) - res.max_evt_qs /= 2; - adapter->res = res; } + /* If RoCE is supported stash away half the EQs for RoCE */ + res.max_nic_evt_qs = be_roce_supported(adapter) ? + res.max_evt_qs / 2 : res.max_evt_qs; + adapter->res = res; + /* If FW supports RSS default queue, then skip creating non-RSS * queue for non-IP traffic. */ @@ -4252,7 +4249,7 @@ static int be_get_resources(struct be_adapter *adapter) dev_info(dev, "Max: txqs %d, rxqs %d, rss %d, eqs %d, vfs %d\n", be_max_txqs(adapter), be_max_rxqs(adapter), - be_max_rss(adapter), be_max_eqs(adapter), + be_max_rss(adapter), be_max_nic_eqs(adapter), be_max_vfs(adapter)); dev_info(dev, "Max: uc-macs %d, mc-macs %d, vlans %d\n", be_max_uc(adapter), be_max_mc(adapter), -- cgit v0.10.2 From e261768e9e395b3bd71946104afd5550f77d049b Mon Sep 17 00:00:00 2001 From: Sathya Perla Date: Wed, 22 Jun 2016 08:54:54 -0400 Subject: be2net: support asymmetric rx/tx queue counts be2net so far supported creation of RX/TX queues only in pairs. On configs where rx and tx queue counts are different, creation of only the lesser number of queues has been supported. This patch now allows a combination of RX/TX-only channels along with combined channels. N TX-queues and M RX-queues can be created with the following cmds: ethtool -L ethX combined N rx M-N (when N < M) ethtool -L ethX combined M tx N-M (when M < N) Setting both RX-only and TX-only channels is still not supported. It is mandatory to create atleast one combined channel. Signed-off-by: Sathya Perla Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index f8040ef..096cb1f5 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -526,7 +526,8 @@ struct be_adapter { spinlock_t mcc_lock; /* For serializing mcc cmds to BE card */ spinlock_t mcc_cq_lock; - u16 cfg_num_qs; /* configured via set-channels */ + u16 cfg_num_rx_irqs; /* configured via set-channels */ + u16 cfg_num_tx_irqs; /* configured via set-channels */ u16 num_evt_qs; u16 num_msix_vec; struct be_eq_obj eq_obj[MAX_EVT_QS]; @@ -652,14 +653,35 @@ struct be_adapter { #define be_if_cap_flags(adapter) (adapter->res.if_cap_flags) #define be_max_pf_pool_rss_tables(adapter) \ (adapter->pool_res.max_rss_tables) +/* Max irqs avaialble for NIC */ +#define be_max_irqs(adapter) \ + (min_t(u16, be_max_nic_eqs(adapter), num_online_cpus())) -static inline u16 be_max_qs(struct be_adapter *adapter) +/* Max irqs *needed* for RX queues */ +static inline u16 be_max_rx_irqs(struct be_adapter *adapter) { - /* If no RSS, need atleast the one def RXQ */ + /* If no RSS, need atleast one irq for def-RXQ */ u16 num = max_t(u16, be_max_rss(adapter), 1); - num = min(num, be_max_nic_eqs(adapter)); - return min_t(u16, num, num_online_cpus()); + return min_t(u16, num, be_max_irqs(adapter)); +} + +/* Max irqs *needed* for TX queues */ +static inline u16 be_max_tx_irqs(struct be_adapter *adapter) +{ + return min_t(u16, be_max_txqs(adapter), be_max_irqs(adapter)); +} + +/* Max irqs *needed* for combined queues */ +static inline u16 be_max_qp_irqs(struct be_adapter *adapter) +{ + return min(be_max_tx_irqs(adapter), be_max_rx_irqs(adapter)); +} + +/* Max irqs *needed* for RX and TX queues together */ +static inline u16 be_max_any_irqs(struct be_adapter *adapter) +{ + return max(be_max_tx_irqs(adapter), be_max_rx_irqs(adapter)); } /* Is BE in pvid_tagging mode */ diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c index c569cd7..71940b9 100644 --- a/drivers/net/ethernet/emulex/benet/be_ethtool.c +++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c @@ -1196,9 +1196,17 @@ static void be_get_channels(struct net_device *netdev, struct ethtool_channels *ch) { struct be_adapter *adapter = netdev_priv(netdev); + u16 num_rx_irqs = max_t(u16, adapter->num_rss_qs, 1); - ch->combined_count = adapter->num_evt_qs; - ch->max_combined = be_max_qs(adapter); + /* num_tx_qs is always same as the number of irqs used for TX */ + ch->combined_count = min(adapter->num_tx_qs, num_rx_irqs); + ch->rx_count = num_rx_irqs - ch->combined_count; + ch->tx_count = adapter->num_tx_qs - ch->combined_count; + + ch->max_combined = be_max_qp_irqs(adapter); + /* The user must create atleast one combined channel */ + ch->max_rx = be_max_rx_irqs(adapter) - 1; + ch->max_tx = be_max_tx_irqs(adapter) - 1; } static int be_set_channels(struct net_device *netdev, @@ -1207,11 +1215,22 @@ static int be_set_channels(struct net_device *netdev, struct be_adapter *adapter = netdev_priv(netdev); int status; - if (ch->rx_count || ch->tx_count || ch->other_count || - !ch->combined_count || ch->combined_count > be_max_qs(adapter)) + /* we support either only combined channels or a combination of + * combined and either RX-only or TX-only channels. + */ + if (ch->other_count || !ch->combined_count || + (ch->rx_count && ch->tx_count)) + return -EINVAL; + + if (ch->combined_count > be_max_qp_irqs(adapter) || + (ch->rx_count && + (ch->rx_count + ch->combined_count) > be_max_rx_irqs(adapter)) || + (ch->tx_count && + (ch->tx_count + ch->combined_count) > be_max_tx_irqs(adapter))) return -EINVAL; - adapter->cfg_num_qs = ch->combined_count; + adapter->cfg_num_rx_irqs = ch->combined_count + ch->rx_count; + adapter->cfg_num_tx_irqs = ch->combined_count + ch->tx_count; status = be_update_queues(adapter); return be_cmd_status(status); diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index c67830f..a654c12 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -2620,8 +2620,10 @@ static int be_evt_queues_create(struct be_adapter *adapter) struct be_aic_obj *aic; int i, rc; + /* need enough EQs to service both RX and TX queues */ adapter->num_evt_qs = min_t(u16, num_irqs(adapter), - adapter->cfg_num_qs); + max(adapter->cfg_num_rx_irqs, + adapter->cfg_num_tx_irqs)); for_all_evt_queues(adapter, eqo, i) { int numa_node = dev_to_node(&adapter->pdev->dev); @@ -2726,7 +2728,7 @@ static int be_tx_qs_create(struct be_adapter *adapter) struct be_eq_obj *eqo; int status, i; - adapter->num_tx_qs = min(adapter->num_evt_qs, be_max_txqs(adapter)); + adapter->num_tx_qs = min(adapter->num_evt_qs, adapter->cfg_num_tx_irqs); for_all_tx_queues(adapter, txo, i) { cq = &txo->cq; @@ -2784,11 +2786,11 @@ static int be_rx_cqs_create(struct be_adapter *adapter) struct be_rx_obj *rxo; int rc, i; - /* We can create as many RSS rings as there are EQs. */ - adapter->num_rss_qs = adapter->num_evt_qs; + adapter->num_rss_qs = + min(adapter->num_evt_qs, adapter->cfg_num_rx_irqs); /* We'll use RSS only if atleast 2 RSS rings are supported. */ - if (adapter->num_rss_qs <= 1) + if (adapter->num_rss_qs < 2) adapter->num_rss_qs = 0; adapter->num_rx_qs = adapter->num_rss_qs + adapter->need_def_rxq; @@ -3249,18 +3251,22 @@ static void be_msix_disable(struct be_adapter *adapter) static int be_msix_enable(struct be_adapter *adapter) { - int i, num_vec; + unsigned int i, num_vec, max_roce_eqs; struct device *dev = &adapter->pdev->dev; /* If RoCE is supported, program the max number of vectors that * could be used for NIC and RoCE, else, just program the number * we'll use initially. */ - if (be_roce_supported(adapter)) - num_vec = min_t(int, be_max_func_eqs(adapter), - 2 * num_online_cpus()); - else - num_vec = adapter->cfg_num_qs; + if (be_roce_supported(adapter)) { + max_roce_eqs = + be_max_func_eqs(adapter) - be_max_nic_eqs(adapter); + max_roce_eqs = min(max_roce_eqs, num_online_cpus()); + num_vec = be_max_any_irqs(adapter) + max_roce_eqs; + } else { + num_vec = max(adapter->cfg_num_rx_irqs, + adapter->cfg_num_tx_irqs); + } for (i = 0; i < num_vec; i++) adapter->msix_entries[i].entry = i; @@ -4255,9 +4261,11 @@ static int be_get_resources(struct be_adapter *adapter) be_max_uc(adapter), be_max_mc(adapter), be_max_vlans(adapter)); - /* Sanitize cfg_num_qs based on HW and platform limits */ - adapter->cfg_num_qs = min_t(u16, netif_get_num_default_rss_queues(), - be_max_qs(adapter)); + /* Ensure RX and TX queues are created in pairs at init time */ + adapter->cfg_num_rx_irqs = + min_t(u16, netif_get_num_default_rss_queues(), + be_max_qp_irqs(adapter)); + adapter->cfg_num_tx_irqs = adapter->cfg_num_rx_irqs; return 0; } @@ -4370,7 +4378,7 @@ static int be_if_create(struct be_adapter *adapter) u32 cap_flags = be_if_cap_flags(adapter); int status; - if (adapter->cfg_num_qs == 1) + if (adapter->cfg_num_rx_irqs == 1) cap_flags &= ~(BE_IF_FLAGS_DEFQ_RSS | BE_IF_FLAGS_RSS); en_flags &= cap_flags; -- cgit v0.10.2 From 884476be065e23bb8e5abda3aad9ba04c17341c3 Mon Sep 17 00:00:00 2001 From: Somnath Kotur Date: Wed, 22 Jun 2016 08:54:55 -0400 Subject: be2net: Fix broadcast echoes from EVB in BE3 On SR-IOV profiles, when the user connects a Linux Bridge or OVS to a BE3 vport, they suffer the "broadcast/multicast echo" problem. BE3 EVB echoes broadcast and multicast packets back to PF's vport confusing the Linux bridge. BE3 relies on the src-mac addr being programmed on the interface to avoid sending back an echo of a broadcast or multicast packet on a vPort. When a Linux bridge is connected to a BE3, the mac-addr of the VM behind the bridge doesn't get configured on the vPort and so echo cancellation doesn't work. This patch worksaround this problem by disabling the EVB initially and re-enabling it *only* when SR-IOV is enabled by the user. For the driver fix to work, the BE3 FW version must be >= 11.1.84.0. Signed-off-by: Somnath Kotur Signed-off-by: Sathya Perla Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index 29aeb91..a98b6ab 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -87,6 +87,11 @@ static struct be_cmd_priv_map cmd_priv_map[] = { CMD_SUBSYSTEM_LOWLEVEL, BE_PRIV_DEVCFG | BE_PRIV_DEVSEC }, + { + OPCODE_COMMON_SET_HSW_CONFIG, + CMD_SUBSYSTEM_COMMON, + BE_PRIV_DEVCFG | BE_PRIV_VHADM + }, }; static bool be_cmd_allowed(struct be_adapter *adapter, u8 opcode, u8 subsystem) @@ -3850,6 +3855,10 @@ int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid, void *ctxt; int status; + if (!be_cmd_allowed(adapter, OPCODE_COMMON_SET_HSW_CONFIG, + CMD_SUBSYSTEM_COMMON)) + return -EPERM; + spin_lock_bh(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); @@ -3871,7 +3880,7 @@ int be_cmd_set_hsw_config(struct be_adapter *adapter, u16 pvid, AMAP_SET_BITS(struct amap_set_hsw_context, pvid_valid, ctxt, 1); AMAP_SET_BITS(struct amap_set_hsw_context, pvid, ctxt, pvid); } - if (!BEx_chip(adapter) && hsw_mode) { + if (hsw_mode) { AMAP_SET_BITS(struct amap_set_hsw_context, interface_id, ctxt, adapter->hba_port_num); AMAP_SET_BITS(struct amap_set_hsw_context, pport, ctxt, 1); diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index a654c12..f11a19b 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -3729,6 +3729,11 @@ static void be_vf_clear(struct be_adapter *adapter) be_cmd_if_destroy(adapter, vf_cfg->if_handle, vf + 1); } + + if (BE3_chip(adapter)) + be_cmd_set_hsw_config(adapter, 0, 0, + adapter->if_handle, + PORT_FWD_TYPE_PASSTHRU, 0); done: kfree(adapter->vf_cfg); adapter->num_vfs = 0; @@ -4019,6 +4024,15 @@ static int be_vf_setup(struct be_adapter *adapter) } } + if (BE3_chip(adapter)) { + /* On BE3, enable VEB only when SRIOV is enabled */ + status = be_cmd_set_hsw_config(adapter, 0, 0, + adapter->if_handle, + PORT_FWD_TYPE_VEB, 0); + if (status) + goto err; + } + adapter->flags |= BE_FLAGS_SRIOV_ENABLED; return 0; err: @@ -4564,6 +4578,15 @@ static int be_setup(struct be_adapter *adapter) be_cmd_set_logical_link_config(adapter, IFLA_VF_LINK_STATE_AUTO, 0); + /* BE3 EVB echoes broadcast/multicast packets back to PF's vport + * confusing a linux bridge or OVS that it might be connected to. + * Set the EVB to PASSTHRU mode which effectively disables the EVB + * when SRIOV is not enabled. + */ + if (BE3_chip(adapter)) + be_cmd_set_hsw_config(adapter, 0, 0, adapter->if_handle, + PORT_FWD_TYPE_PASSTHRU, 0); + if (adapter->num_vfs) be_vf_setup(adapter); -- cgit v0.10.2 From 7dfbe7d799ffd5cafd02c79434f3bf93bbe4fe52 Mon Sep 17 00:00:00 2001 From: Somnath Kotur Date: Wed, 22 Jun 2016 08:54:56 -0400 Subject: be2net: Change copyright markings in source files This patch updates year and company name in the copyright markings in the be2net source files. Signed-off-by: Somnath Kotur Signed-off-by: Sathya Perla Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index 096cb1f5..4555e04 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index a98b6ab..2cc1175 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h index cb96ddd..0d6be22 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.h +++ b/drivers/net/ethernet/emulex/benet/be_cmds.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c index 71940b9..50e7be5 100644 --- a/drivers/net/ethernet/emulex/benet/be_ethtool.c +++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index f11a19b..1873c74 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or diff --git a/drivers/net/ethernet/emulex/benet/be_roce.c b/drivers/net/ethernet/emulex/benet/be_roce.c index 4089156..2b62841 100644 --- a/drivers/net/ethernet/emulex/benet/be_roce.c +++ b/drivers/net/ethernet/emulex/benet/be_roce.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or diff --git a/drivers/net/ethernet/emulex/benet/be_roce.h b/drivers/net/ethernet/emulex/benet/be_roce.h index fde6097..e51719a 100644 --- a/drivers/net/ethernet/emulex/benet/be_roce.h +++ b/drivers/net/ethernet/emulex/benet/be_roce.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2005 - 2015 Emulex + * Copyright (C) 2005 - 2016 Broadcom * All rights reserved. * * This program is free software; you can redistribute it and/or -- cgit v0.10.2 From d2ee76fa0a999e1c174b219e91c748f65f42ac16 Mon Sep 17 00:00:00 2001 From: Sathya Perla Date: Wed, 22 Jun 2016 08:54:57 -0400 Subject: be2net: update be2net maintainers list This patch removes Padmanabh's name from the maintainers list as he's no longer with the company. It also adds the driver name on the headline to make it easy to lookup the maintainers list by the driver name. Signed-off-by: Sathya Perla Signed-off-by: David S. Miller diff --git a/MAINTAINERS b/MAINTAINERS index 3a171a9..f5ddaa9 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -10280,10 +10280,9 @@ W: http://www.avagotech.com S: Supported F: drivers/scsi/be2iscsi/ -Emulex 10Gbps NIC BE2, BE3-R, Lancer, Skyhawk-R DRIVER +Emulex 10Gbps NIC BE2, BE3-R, Lancer, Skyhawk-R DRIVER (be2net) M: Sathya Perla M: Ajit Khaparde -M: Padmanabh Ratnakar M: Sriharsha Basavapatna M: Somnath Kotur L: netdev@vger.kernel.org -- cgit v0.10.2 From 1466cc5b23d18e7b6b8f1a45443d595393dbcae7 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Thu, 23 Jun 2016 17:02:37 +0300 Subject: net/mlx5: Rate limit tables support Configuring and managing HW rate limit tables. The HW holds a table of rate limits, each rate is associated with an index in that table. Later a Send Queue uses this index to set the rate limit. Multiple Send Queues can have the same rate limit, which is represented by a single entry in this table. Even though a rate can be shared, each queue is being rate limited independently of others. The SW shadow of this table holds the rate itself, the index in the HW table and the refcount (number of queues) working with this rate. The exported functions are mlx5_rl_add_rate and mlx5_rl_remove_rate. Number of different rates and their values are derived from HW capabilities. Signed-off-by: Yevgeny Petrilin Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 9ea7b58..0c8a7dc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -1,8 +1,9 @@ obj-$(CONFIG_MLX5_CORE) += mlx5_core.o mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ - health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ - mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o fs_counters.o + health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ + mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o \ + fs_counters.o rl.o mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \ en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index 75c7ae6..77fc1aa 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -151,6 +151,12 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev) return err; } + if (MLX5_CAP_GEN(dev, qos)) { + err = mlx5_core_get_caps(dev, MLX5_CAP_QOS); + if (err) + return err; + } + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index a19b593..08cae34 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1144,6 +1144,13 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) dev_err(&pdev->dev, "Failed to init flow steering\n"); goto err_fs; } + + err = mlx5_init_rl_table(dev); + if (err) { + dev_err(&pdev->dev, "Failed to init rate limiting\n"); + goto err_rl; + } + #ifdef CONFIG_MLX5_CORE_EN err = mlx5_eswitch_init(dev); if (err) { @@ -1183,6 +1190,8 @@ err_sriov: mlx5_eswitch_cleanup(dev->priv.eswitch); #endif err_reg_dev: + mlx5_cleanup_rl_table(dev); +err_rl: mlx5_cleanup_fs(dev); err_fs: mlx5_cleanup_mkey_table(dev); @@ -1253,6 +1262,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) mlx5_eswitch_cleanup(dev->priv.eswitch); #endif + mlx5_cleanup_rl_table(dev); mlx5_cleanup_fs(dev); mlx5_cleanup_mkey_table(dev); mlx5_cleanup_srq_table(dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/rl.c b/drivers/net/ethernet/mellanox/mlx5/core/rl.c new file mode 100644 index 0000000..c07c28b --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/rl.c @@ -0,0 +1,209 @@ +/* + * Copyright (c) 2013-2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include "mlx5_core.h" + +/* Finds an entry where we can register the given rate + * If the rate already exists, return the entry where it is registered, + * otherwise return the first available entry. + * If the table is full, return NULL + */ +static struct mlx5_rl_entry *find_rl_entry(struct mlx5_rl_table *table, + u32 rate) +{ + struct mlx5_rl_entry *ret_entry = NULL; + bool empty_found = false; + int i; + + for (i = 0; i < table->max_size; i++) { + if (table->rl_entry[i].rate == rate) + return &table->rl_entry[i]; + if (!empty_found && !table->rl_entry[i].rate) { + empty_found = true; + ret_entry = &table->rl_entry[i]; + } + } + + return ret_entry; +} + +static int mlx5_set_rate_limit_cmd(struct mlx5_core_dev *dev, + u32 rate, u16 index) +{ + u32 in[MLX5_ST_SZ_DW(set_rate_limit_in)]; + u32 out[MLX5_ST_SZ_DW(set_rate_limit_out)]; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + + MLX5_SET(set_rate_limit_in, in, opcode, + MLX5_CMD_OP_SET_RATE_LIMIT); + MLX5_SET(set_rate_limit_in, in, rate_limit_index, index); + MLX5_SET(set_rate_limit_in, in, rate_limit, rate); + + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), + out, sizeof(out)); +} + +bool mlx5_rl_is_in_range(struct mlx5_core_dev *dev, u32 rate) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + + return (rate <= table->max_rate && rate >= table->min_rate); +} +EXPORT_SYMBOL(mlx5_rl_is_in_range); + +int mlx5_rl_add_rate(struct mlx5_core_dev *dev, u32 rate, u16 *index) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + struct mlx5_rl_entry *entry; + int err = 0; + + mutex_lock(&table->rl_lock); + + if (!rate || !mlx5_rl_is_in_range(dev, rate)) { + mlx5_core_err(dev, "Invalid rate: %u, should be %u to %u\n", + rate, table->min_rate, table->max_rate); + err = -EINVAL; + goto out; + } + + entry = find_rl_entry(table, rate); + if (!entry) { + mlx5_core_err(dev, "Max number of %u rates reached\n", + table->max_size); + err = -ENOSPC; + goto out; + } + if (entry->refcount) { + /* rate already configured */ + entry->refcount++; + } else { + /* new rate limit */ + err = mlx5_set_rate_limit_cmd(dev, rate, entry->index); + if (err) { + mlx5_core_err(dev, "Failed configuring rate: %u (%d)\n", + rate, err); + goto out; + } + entry->rate = rate; + entry->refcount = 1; + } + *index = entry->index; + +out: + mutex_unlock(&table->rl_lock); + return err; +} +EXPORT_SYMBOL(mlx5_rl_add_rate); + +void mlx5_rl_remove_rate(struct mlx5_core_dev *dev, u32 rate) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + struct mlx5_rl_entry *entry = NULL; + + /* 0 is a reserved value for unlimited rate */ + if (rate == 0) + return; + + mutex_lock(&table->rl_lock); + entry = find_rl_entry(table, rate); + if (!entry || !entry->refcount) { + mlx5_core_warn(dev, "Rate %u is not configured\n", rate); + goto out; + } + + entry->refcount--; + if (!entry->refcount) { + /* need to remove rate */ + mlx5_set_rate_limit_cmd(dev, 0, entry->index); + entry->rate = 0; + } + +out: + mutex_unlock(&table->rl_lock); +} +EXPORT_SYMBOL(mlx5_rl_remove_rate); + +int mlx5_init_rl_table(struct mlx5_core_dev *dev) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + int i; + + mutex_init(&table->rl_lock); + if (!MLX5_CAP_GEN(dev, qos) || !MLX5_CAP_QOS(dev, packet_pacing)) { + table->max_size = 0; + return 0; + } + + /* First entry is reserved for unlimited rate */ + table->max_size = MLX5_CAP_QOS(dev, packet_pacing_rate_table_size) - 1; + table->max_rate = MLX5_CAP_QOS(dev, packet_pacing_max_rate); + table->min_rate = MLX5_CAP_QOS(dev, packet_pacing_min_rate); + + table->rl_entry = kcalloc(table->max_size, sizeof(struct mlx5_rl_entry), + GFP_KERNEL); + if (!table->rl_entry) + return -ENOMEM; + + /* The index represents the index in HW rate limit table + * Index 0 is reserved for unlimited rate + */ + for (i = 0; i < table->max_size; i++) + table->rl_entry[i].index = i + 1; + + /* Index 0 is reserved */ + mlx5_core_info(dev, "Rate limit: %u rates are supported, range: %uMbps to %uMbps\n", + table->max_size, + table->min_rate >> 10, + table->max_rate >> 10); + + return 0; +} + +void mlx5_cleanup_rl_table(struct mlx5_core_dev *dev) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + int i; + + /* Clear all configured rates */ + for (i = 0; i < table->max_size; i++) + if (table->rl_entry[i].rate) + mlx5_set_rate_limit_cmd(dev, 0, + table->rl_entry[i].index); + + kfree(dev->priv.rl_table.rl_entry); +} diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 73a4847..e0a3ed7 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -1330,6 +1330,7 @@ enum mlx5_cap_type { MLX5_CAP_ESWITCH, MLX5_CAP_RESERVED, MLX5_CAP_VECTOR_CALC, + MLX5_CAP_QOS, /* NUM OF CAP Types */ MLX5_CAP_NUM }; @@ -1414,6 +1415,9 @@ enum mlx5_cap_type { MLX5_GET(vector_calc_cap, \ mdev->hca_caps_cur[MLX5_CAP_VECTOR_CALC], cap) +#define MLX5_CAP_QOS(mdev, cap)\ + MLX5_GET(qos_cap, mdev->hca_caps_cur[MLX5_CAP_QOS], cap) + enum { MLX5_CMD_STAT_OK = 0x0, MLX5_CMD_STAT_INT_ERR = 0x1, diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 80776d0..46260fd 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -481,6 +481,21 @@ struct mlx5_fc_stats { struct mlx5_eswitch; +struct mlx5_rl_entry { + u32 rate; + u16 index; + u16 refcount; +}; + +struct mlx5_rl_table { + /* protect rate limit table */ + struct mutex rl_lock; + u16 max_size; + u32 max_rate; + u32 min_rate; + struct mlx5_rl_entry *rl_entry; +}; + struct mlx5_priv { char name[MLX5_MAX_NAME_LEN]; struct mlx5_eq_table eq_table; @@ -544,6 +559,7 @@ struct mlx5_priv { struct mlx5_flow_root_namespace *esw_ingress_root_ns; struct mlx5_fc_stats fc_stats; + struct mlx5_rl_table rl_table; }; enum mlx5_device_state { @@ -861,6 +877,12 @@ int mlx5_query_odp_caps(struct mlx5_core_dev *dev, int mlx5_core_query_ib_ppcnt(struct mlx5_core_dev *dev, u8 port_num, void *out, size_t sz); +int mlx5_init_rl_table(struct mlx5_core_dev *dev); +void mlx5_cleanup_rl_table(struct mlx5_core_dev *dev); +int mlx5_rl_add_rate(struct mlx5_core_dev *dev, u32 rate, u16 *index); +void mlx5_rl_remove_rate(struct mlx5_core_dev *dev, u32 rate); +bool mlx5_rl_is_in_range(struct mlx5_core_dev *dev, u32 rate); + static inline int fw_initializing(struct mlx5_core_dev *dev) { return ioread32be(&dev->iseg->initializing) >> 31; @@ -938,6 +960,11 @@ static inline int mlx5_get_gid_table_len(u16 param) return 8 * (1 << param); } +static inline bool mlx5_rl_is_supported(struct mlx5_core_dev *dev) +{ + return !!(dev->priv.rl_table.max_size); +} + enum { MLX5_TRIGGERED_CMD_COMP = (u64)1 << 32, }; -- cgit v0.10.2 From 507f0c817f7a28bbf4facb3a8dca72a68bc25248 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Thu, 23 Jun 2016 17:02:38 +0300 Subject: net/mlx5e: Add TXQ set max rate support Implement set_maxrate ndo. Use the rate index from the hardware table to attach to channel SQ/TXQ. In case of failure to configure new rate, the queue remains with unlimited rate. We save the configuration on priv structure and apply it each time Send Queues are being reinitialized (after open/close) operations. Signed-off-by: Yevgeny Petrilin Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index e8a6c33..017e047 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -88,6 +88,7 @@ #define MLX5E_LOG_INDIR_RQT_SIZE 0x7 #define MLX5E_INDIR_RQT_SIZE BIT(MLX5E_LOG_INDIR_RQT_SIZE) #define MLX5E_MAX_NUM_CHANNELS (MLX5E_INDIR_RQT_SIZE >> 1) +#define MLX5E_MAX_NUM_SQS (MLX5E_MAX_NUM_CHANNELS * MLX5E_MAX_NUM_TC) #define MLX5E_TX_CQ_POLL_BUDGET 128 #define MLX5E_UPDATE_STATS_INTERVAL 200 /* msecs */ #define MLX5E_SQ_BF_BUDGET 16 @@ -354,6 +355,7 @@ struct mlx5e_sq { struct mlx5e_channel *channel; int tc; struct mlx5e_ico_wqe_info *ico_wqe_info; + u32 rate_limit; } ____cacheline_aligned_in_smp; static inline bool mlx5e_sq_has_room_for(struct mlx5e_sq *sq, u16 n) @@ -530,6 +532,7 @@ struct mlx5e_priv { u32 indir_rqtn; u32 indir_tirn[MLX5E_NUM_INDIR_TIRS]; struct mlx5e_direct_tir direct_tir[MLX5E_MAX_NUM_CHANNELS]; + u32 tx_rates[MLX5E_MAX_NUM_SQS]; struct mlx5e_flow_steering fs; struct mlx5e_vxlan_db vxlan; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 8b7c6f3..e5a2cef 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -702,7 +702,8 @@ static int mlx5e_enable_sq(struct mlx5e_sq *sq, struct mlx5e_sq_param *param) return err; } -static int mlx5e_modify_sq(struct mlx5e_sq *sq, int curr_state, int next_state) +static int mlx5e_modify_sq(struct mlx5e_sq *sq, int curr_state, + int next_state, bool update_rl, int rl_index) { struct mlx5e_channel *c = sq->channel; struct mlx5e_priv *priv = c->priv; @@ -722,6 +723,10 @@ static int mlx5e_modify_sq(struct mlx5e_sq *sq, int curr_state, int next_state) MLX5_SET(modify_sq_in, in, sq_state, curr_state); MLX5_SET(sqc, sqc, state, next_state); + if (update_rl && next_state == MLX5_SQC_STATE_RDY) { + MLX5_SET64(modify_sq_in, in, modify_bitmask, 1); + MLX5_SET(sqc, sqc, packet_pacing_rate_limit_index, rl_index); + } err = mlx5_core_modify_sq(mdev, sq->sqn, in, inlen); @@ -737,6 +742,8 @@ static void mlx5e_disable_sq(struct mlx5e_sq *sq) struct mlx5_core_dev *mdev = priv->mdev; mlx5_core_destroy_sq(mdev, sq->sqn); + if (sq->rate_limit) + mlx5_rl_remove_rate(mdev, sq->rate_limit); } static int mlx5e_open_sq(struct mlx5e_channel *c, @@ -754,7 +761,8 @@ static int mlx5e_open_sq(struct mlx5e_channel *c, if (err) goto err_destroy_sq; - err = mlx5e_modify_sq(sq, MLX5_SQC_STATE_RST, MLX5_SQC_STATE_RDY); + err = mlx5e_modify_sq(sq, MLX5_SQC_STATE_RST, MLX5_SQC_STATE_RDY, + false, 0); if (err) goto err_disable_sq; @@ -793,7 +801,8 @@ static void mlx5e_close_sq(struct mlx5e_sq *sq) if (mlx5e_sq_has_room_for(sq, 1)) mlx5e_send_nop(sq, true); - mlx5e_modify_sq(sq, MLX5_SQC_STATE_RDY, MLX5_SQC_STATE_ERR); + mlx5e_modify_sq(sq, MLX5_SQC_STATE_RDY, MLX5_SQC_STATE_ERR, + false, 0); } while (sq->cc != sq->pc) /* wait till sq is empty */ @@ -1024,6 +1033,79 @@ static void mlx5e_build_channeltc_to_txq_map(struct mlx5e_priv *priv, int ix) ix + i * priv->params.num_channels; } +static int mlx5e_set_sq_maxrate(struct net_device *dev, + struct mlx5e_sq *sq, u32 rate) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5_core_dev *mdev = priv->mdev; + u16 rl_index = 0; + int err; + + if (rate == sq->rate_limit) + /* nothing to do */ + return 0; + + if (sq->rate_limit) + /* remove current rl index to free space to next ones */ + mlx5_rl_remove_rate(mdev, sq->rate_limit); + + sq->rate_limit = 0; + + if (rate) { + err = mlx5_rl_add_rate(mdev, rate, &rl_index); + if (err) { + netdev_err(dev, "Failed configuring rate %u: %d\n", + rate, err); + return err; + } + } + + err = mlx5e_modify_sq(sq, MLX5_SQC_STATE_RDY, + MLX5_SQC_STATE_RDY, true, rl_index); + if (err) { + netdev_err(dev, "Failed configuring rate %u: %d\n", + rate, err); + /* remove the rate from the table */ + if (rate) + mlx5_rl_remove_rate(mdev, rate); + return err; + } + + sq->rate_limit = rate; + return 0; +} + +static int mlx5e_set_tx_maxrate(struct net_device *dev, int index, u32 rate) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5e_sq *sq = priv->txq_to_sq_map[index]; + int err = 0; + + if (!mlx5_rl_is_supported(mdev)) { + netdev_err(dev, "Rate limiting is not supported on this device\n"); + return -EINVAL; + } + + /* rate is given in Mb/sec, HW config is in Kb/sec */ + rate = rate << 10; + + /* Check whether rate in valid range, 0 is always valid */ + if (rate && !mlx5_rl_is_in_range(mdev, rate)) { + netdev_err(dev, "TX rate %u, is not in range\n", rate); + return -ERANGE; + } + + mutex_lock(&priv->state_lock); + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + err = mlx5e_set_sq_maxrate(dev, sq, rate); + if (!err) + priv->tx_rates[index] = rate; + mutex_unlock(&priv->state_lock); + + return err; +} + static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, struct mlx5e_channel_param *cparam, struct mlx5e_channel **cp) @@ -1031,7 +1113,9 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, struct net_device *netdev = priv->netdev; int cpu = mlx5e_get_cpu(priv, ix); struct mlx5e_channel *c; + struct mlx5e_sq *sq; int err; + int i; c = kzalloc_node(sizeof(*c), GFP_KERNEL, cpu_to_node(cpu)); if (!c) @@ -1073,6 +1157,16 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, if (err) goto err_close_icosq; + for (i = 0; i < priv->params.num_tc; i++) { + u32 txq_ix = priv->channeltc_to_txq_map[ix][i]; + + if (priv->tx_rates[txq_ix]) { + sq = priv->txq_to_sq_map[txq_ix]; + mlx5e_set_sq_maxrate(priv->netdev, sq, + priv->tx_rates[txq_ix]); + } + } + err = mlx5e_open_rq(c, &cparam->rq, &c->rq); if (err) goto err_close_sqs; @@ -2611,6 +2705,7 @@ static const struct net_device_ops mlx5e_netdev_ops_basic = { .ndo_set_features = mlx5e_set_features, .ndo_change_mtu = mlx5e_change_mtu, .ndo_do_ioctl = mlx5e_ioctl, + .ndo_set_tx_maxrate = mlx5e_set_tx_maxrate, #ifdef CONFIG_RFS_ACCEL .ndo_rx_flow_steer = mlx5e_rx_flow_steer, #endif @@ -2632,6 +2727,7 @@ static const struct net_device_ops mlx5e_netdev_ops_sriov = { .ndo_do_ioctl = mlx5e_ioctl, .ndo_udp_tunnel_add = mlx5e_add_vxlan_port, .ndo_udp_tunnel_del = mlx5e_del_vxlan_port, + .ndo_set_tx_maxrate = mlx5e_set_tx_maxrate, .ndo_features_check = mlx5e_features_check, #ifdef CONFIG_RFS_ACCEL .ndo_rx_flow_steer = mlx5e_rx_flow_steer, -- cgit v0.10.2 From 4e59e288813901815b39c82fc00d4e9fe78ce16b Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 23 Jun 2016 17:02:39 +0300 Subject: net/mlx5e: Introduce net device priv flags infrastructure Introduce an infrastructure for getting/setting private net device flags. Currently a 'nop' priv flag is added, following patches will override the flag will actual feature specific flags. Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 017e047..02fa4da 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -144,6 +144,22 @@ struct mlx5e_umr_wqe { struct mlx5_wqe_data_seg data; }; +static const char mlx5e_priv_flags[][ETH_GSTRING_LEN] = { + "nop", +}; + +enum mlx5e_priv_flag { + MLX5E_PFLAG_NOP = (1 << 0), +}; + +#define MLX5E_SET_PRIV_FLAG(priv, pflag, enable) \ + do { \ + if (enable) \ + priv->pflags |= pflag; \ + else \ + priv->pflags &= ~pflag; \ + } while (0) + #ifdef CONFIG_MLX5_CORE_EN_DCB #define MLX5E_MAX_BW_ALLOC 100 /* Max percentage of BW allocation */ #define MLX5E_MIN_BW_ALLOC 1 /* Min percentage of BW allocation */ @@ -543,6 +559,7 @@ struct mlx5e_priv { struct work_struct set_rx_mode_work; struct delayed_work update_stats_work; + u32 pflags; struct mlx5_core_dev *mdev; struct net_device *netdev; struct mlx5e_stats stats; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index fc7dcc0..f8bbc2b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -198,6 +198,8 @@ static int mlx5e_get_sset_count(struct net_device *dev, int sset) MLX5E_NUM_RQ_STATS(priv) + MLX5E_NUM_SQ_STATS(priv) + MLX5E_NUM_PFC_COUNTERS(priv); + case ETH_SS_PRIV_FLAGS: + return ARRAY_SIZE(mlx5e_priv_flags); /* fallthrough */ default: return -EOPNOTSUPP; @@ -272,9 +274,12 @@ static void mlx5e_get_strings(struct net_device *dev, uint32_t stringset, uint8_t *data) { struct mlx5e_priv *priv = netdev_priv(dev); + int i; switch (stringset) { case ETH_SS_PRIV_FLAGS: + for (i = 0; i < ARRAY_SIZE(mlx5e_priv_flags); i++) + strcpy(data + i * ETH_GSTRING_LEN, mlx5e_priv_flags[i]); break; case ETH_SS_TEST: @@ -1272,6 +1277,58 @@ static int mlx5e_get_module_eeprom(struct net_device *netdev, return 0; } +typedef int (*mlx5e_pflag_handler)(struct net_device *netdev, bool enable); + +static int set_pflag_nop(struct net_device *netdev, bool enable) +{ + return 0; +} + +static int mlx5e_handle_pflag(struct net_device *netdev, + u32 wanted_flags, + enum mlx5e_priv_flag flag, + mlx5e_pflag_handler pflag_handler) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + bool enable = !!(wanted_flags & flag); + u32 changes = wanted_flags ^ priv->pflags; + int err; + + if (!(changes & flag)) + return 0; + + err = pflag_handler(netdev, enable); + if (err) { + netdev_err(netdev, "%s private flag 0x%x failed err %d\n", + enable ? "Enable" : "Disable", flag, err); + return err; + } + + MLX5E_SET_PRIV_FLAG(priv, flag, enable); + return 0; +} + +static int mlx5e_set_priv_flags(struct net_device *netdev, u32 pflags) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + int err; + + mutex_lock(&priv->state_lock); + + err = mlx5e_handle_pflag(netdev, pflags, MLX5E_PFLAG_NOP, + set_pflag_nop); + + mutex_unlock(&priv->state_lock); + return err ? -EINVAL : 0; +} + +static u32 mlx5e_get_priv_flags(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + return priv->pflags; +} + const struct ethtool_ops mlx5e_ethtool_ops = { .get_drvinfo = mlx5e_get_drvinfo, .get_link = ethtool_op_get_link, @@ -1301,4 +1358,6 @@ const struct ethtool_ops mlx5e_ethtool_ops = { .set_wol = mlx5e_set_wol, .get_module_info = mlx5e_get_module_info, .get_module_eeprom = mlx5e_get_module_eeprom, + .get_priv_flags = mlx5e_get_priv_flags, + .set_priv_flags = mlx5e_set_priv_flags }; -- cgit v0.10.2 From 9908aa292971ee3320ea13a71d75f90a52929892 Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Thu, 23 Jun 2016 17:02:40 +0300 Subject: net/mlx5e: CQE based moderation In this mode the moderation timer will restart upon new completion (CQE) generation rather than upon interrupt generation. The outcome is that for bursty traffic the period timer will never expire and thus only the moderation frames counter will dictate interrupt generation, thus the interrupt rate will be relative to the incoming packets size. If the burst seizes for "moderation period" time then an interrupt will be issued immediately. CQE based moderation is off by default and can be controlled via ethtool set_priv_flags. Performance tested on ConnectX4-Lx 50G. Less packet loss in netperf UDP and TCP tests, with no bw degradation, for both single and multi streams, with message sizes of 64, 1024, 1472 and 32768 byte. Signed-off-by: Tariq Toukan Signed-off-by: Achiad Shochat Signed-off-by: Saeed Mahameed Signed-off-by: Gal Pressman Signed-off-by: Gil Rockah Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 02fa4da..36f625d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -79,6 +79,7 @@ #define MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ (64 * 1024) #define MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC 0x10 +#define MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC_FROM_CQE 0x3 #define MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_PKTS 0x20 #define MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_USEC 0x10 #define MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_PKTS 0x20 @@ -145,11 +146,11 @@ struct mlx5e_umr_wqe { }; static const char mlx5e_priv_flags[][ETH_GSTRING_LEN] = { - "nop", + "rx_cqe_moder", }; enum mlx5e_priv_flag { - MLX5E_PFLAG_NOP = (1 << 0), + MLX5E_PFLAG_RX_CQE_BASED_MODER = (1 << 0), }; #define MLX5E_SET_PRIV_FLAG(priv, pflag, enable) \ @@ -165,6 +166,11 @@ enum mlx5e_priv_flag { #define MLX5E_MIN_BW_ALLOC 1 /* Min percentage of BW allocation */ #endif +struct mlx5e_cq_moder { + u16 usec; + u16 pkts; +}; + struct mlx5e_params { u8 log_sq_size; u8 rq_wq_type; @@ -173,12 +179,11 @@ struct mlx5e_params { u8 log_rq_size; u16 num_channels; u8 num_tc; + u8 rx_cq_period_mode; bool rx_cqe_compress_admin; bool rx_cqe_compress; - u16 rx_cq_moderation_usec; - u16 rx_cq_moderation_pkts; - u16 tx_cq_moderation_usec; - u16 tx_cq_moderation_pkts; + struct mlx5e_cq_moder rx_cq_moderation; + struct mlx5e_cq_moder tx_cq_moderation; u16 min_rx_wqes; bool lro_en; u32 lro_wqe_sz; @@ -667,6 +672,9 @@ void mlx5e_build_default_indir_rqt(struct mlx5_core_dev *mdev, int num_channels); int mlx5e_get_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed); +void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, + u8 cq_period_mode); + static inline void mlx5e_tx_notify_hw(struct mlx5e_sq *sq, struct mlx5_wqe_ctrl_seg *ctrl, int bf_sz) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index f8bbc2b..4f433d3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -524,10 +524,10 @@ static int mlx5e_get_coalesce(struct net_device *netdev, if (!MLX5_CAP_GEN(priv->mdev, cq_moderation)) return -ENOTSUPP; - coal->rx_coalesce_usecs = priv->params.rx_cq_moderation_usec; - coal->rx_max_coalesced_frames = priv->params.rx_cq_moderation_pkts; - coal->tx_coalesce_usecs = priv->params.tx_cq_moderation_usec; - coal->tx_max_coalesced_frames = priv->params.tx_cq_moderation_pkts; + coal->rx_coalesce_usecs = priv->params.rx_cq_moderation.usec; + coal->rx_max_coalesced_frames = priv->params.rx_cq_moderation.pkts; + coal->tx_coalesce_usecs = priv->params.tx_cq_moderation.usec; + coal->tx_max_coalesced_frames = priv->params.tx_cq_moderation.pkts; return 0; } @@ -545,10 +545,11 @@ static int mlx5e_set_coalesce(struct net_device *netdev, return -ENOTSUPP; mutex_lock(&priv->state_lock); - priv->params.tx_cq_moderation_usec = coal->tx_coalesce_usecs; - priv->params.tx_cq_moderation_pkts = coal->tx_max_coalesced_frames; - priv->params.rx_cq_moderation_usec = coal->rx_coalesce_usecs; - priv->params.rx_cq_moderation_pkts = coal->rx_max_coalesced_frames; + + priv->params.tx_cq_moderation.usec = coal->tx_coalesce_usecs; + priv->params.tx_cq_moderation.pkts = coal->tx_max_coalesced_frames; + priv->params.rx_cq_moderation.usec = coal->rx_coalesce_usecs; + priv->params.rx_cq_moderation.pkts = coal->rx_max_coalesced_frames; if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) goto out; @@ -1279,9 +1280,37 @@ static int mlx5e_get_module_eeprom(struct net_device *netdev, typedef int (*mlx5e_pflag_handler)(struct net_device *netdev, bool enable); -static int set_pflag_nop(struct net_device *netdev, bool enable) +static int set_pflag_rx_cqe_based_moder(struct net_device *netdev, bool enable) { - return 0; + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + bool rx_mode_changed; + u8 rx_cq_period_mode; + int err = 0; + bool reset; + + rx_cq_period_mode = enable ? + MLX5_CQ_PERIOD_MODE_START_FROM_CQE : + MLX5_CQ_PERIOD_MODE_START_FROM_EQE; + rx_mode_changed = rx_cq_period_mode != priv->params.rx_cq_period_mode; + + if (rx_cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE && + !MLX5_CAP_GEN(mdev, cq_period_start_from_cqe)) + return -ENOTSUPP; + + if (!rx_mode_changed) + return 0; + + reset = test_bit(MLX5E_STATE_OPENED, &priv->state); + if (reset) + mlx5e_close_locked(netdev); + + mlx5e_set_rx_cq_mode_params(&priv->params, rx_cq_period_mode); + + if (reset) + err = mlx5e_open_locked(netdev); + + return err; } static int mlx5e_handle_pflag(struct net_device *netdev, @@ -1315,8 +1344,9 @@ static int mlx5e_set_priv_flags(struct net_device *netdev, u32 pflags) mutex_lock(&priv->state_lock); - err = mlx5e_handle_pflag(netdev, pflags, MLX5E_PFLAG_NOP, - set_pflag_nop); + err = mlx5e_handle_pflag(netdev, pflags, + MLX5E_PFLAG_RX_CQE_BASED_MODER, + set_pflag_rx_cqe_based_moder); mutex_unlock(&priv->state_lock); return err ? -EINVAL : 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index e5a2cef..13e7a45 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -55,6 +55,7 @@ struct mlx5e_cq_param { u32 cqc[MLX5_ST_SZ_DW(cqc)]; struct mlx5_wq_param wq; u16 eq_ix; + u8 cq_period_mode; }; struct mlx5e_channel_param { @@ -896,6 +897,7 @@ static int mlx5e_enable_cq(struct mlx5e_cq *cq, struct mlx5e_cq_param *param) mlx5_vector2eqn(mdev, param->eq_ix, &eqn, &irqn_not_used); + MLX5_SET(cqc, cqc, cq_period_mode, param->cq_period_mode); MLX5_SET(cqc, cqc, c_eqn, eqn); MLX5_SET(cqc, cqc, uar_page, mcq->uar->index); MLX5_SET(cqc, cqc, log_page_size, cq->wq_ctrl.buf.page_shift - @@ -925,8 +927,7 @@ static void mlx5e_disable_cq(struct mlx5e_cq *cq) static int mlx5e_open_cq(struct mlx5e_channel *c, struct mlx5e_cq_param *param, struct mlx5e_cq *cq, - u16 moderation_usecs, - u16 moderation_frames) + struct mlx5e_cq_moder moderation) { int err; struct mlx5e_priv *priv = c->priv; @@ -942,8 +943,8 @@ static int mlx5e_open_cq(struct mlx5e_channel *c, if (MLX5_CAP_GEN(mdev, cq_moderation)) mlx5_core_modify_cq_moderation(mdev, &cq->mcq, - moderation_usecs, - moderation_frames); + moderation.usec, + moderation.pkts); return 0; err_destroy_cq: @@ -972,8 +973,7 @@ static int mlx5e_open_tx_cqs(struct mlx5e_channel *c, for (tc = 0; tc < c->num_tc; tc++) { err = mlx5e_open_cq(c, &cparam->tx_cq, &c->sq[tc].cq, - priv->params.tx_cq_moderation_usec, - priv->params.tx_cq_moderation_pkts); + priv->params.tx_cq_moderation); if (err) goto err_close_tx_cqs; } @@ -1110,6 +1110,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, struct mlx5e_channel_param *cparam, struct mlx5e_channel **cp) { + struct mlx5e_cq_moder icosq_cq_moder = {0, 0}; struct net_device *netdev = priv->netdev; int cpu = mlx5e_get_cpu(priv, ix); struct mlx5e_channel *c; @@ -1133,7 +1134,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, netif_napi_add(netdev, &c->napi, mlx5e_napi_poll, 64); - err = mlx5e_open_cq(c, &cparam->icosq_cq, &c->icosq.cq, 0, 0); + err = mlx5e_open_cq(c, &cparam->icosq_cq, &c->icosq.cq, icosq_cq_moder); if (err) goto err_napi_del; @@ -1142,8 +1143,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, goto err_close_icosq_cq; err = mlx5e_open_cq(c, &cparam->rx_cq, &c->rq.cq, - priv->params.rx_cq_moderation_usec, - priv->params.rx_cq_moderation_pkts); + priv->params.rx_cq_moderation); if (err) goto err_close_tx_cqs; @@ -1308,6 +1308,8 @@ static void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, } mlx5e_build_common_cq_param(priv, param); + + param->cq_period_mode = priv->params.rx_cq_period_mode; } static void mlx5e_build_tx_cq_param(struct mlx5e_priv *priv, @@ -1318,6 +1320,8 @@ static void mlx5e_build_tx_cq_param(struct mlx5e_priv *priv, MLX5_SET(cqc, cqc, log_cq_size, priv->params.log_sq_size); mlx5e_build_common_cq_param(priv, param); + + param->cq_period_mode = MLX5_CQ_PERIOD_MODE_START_FROM_EQE; } static void mlx5e_build_ico_cq_param(struct mlx5e_priv *priv, @@ -1329,6 +1333,8 @@ static void mlx5e_build_ico_cq_param(struct mlx5e_priv *priv, MLX5_SET(cqc, cqc, log_cq_size, log_wq_size); mlx5e_build_common_cq_param(priv, param); + + param->cq_period_mode = MLX5_CQ_PERIOD_MODE_START_FROM_EQE; } static void mlx5e_build_icosq_param(struct mlx5e_priv *priv, @@ -2856,6 +2862,20 @@ static bool cqe_compress_heuristic(u32 link_speed, u32 pci_bw) (pci_bw < 40000) && (pci_bw < link_speed)); } +void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) +{ + params->rx_cq_period_mode = cq_period_mode; + + params->rx_cq_moderation.pkts = + MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_PKTS; + params->rx_cq_moderation.usec = + MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC; + + if (cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE) + params->rx_cq_moderation.usec = + MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC_FROM_CQE; +} + static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, struct net_device *netdev, int num_channels) @@ -2908,13 +2928,13 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, priv->params.min_rx_wqes = mlx5_min_rx_wqes(priv->params.rq_wq_type, BIT(priv->params.log_rq_size)); - priv->params.rx_cq_moderation_usec = - MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC; - priv->params.rx_cq_moderation_pkts = - MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_PKTS; - priv->params.tx_cq_moderation_usec = + + mlx5e_set_rx_cq_mode_params(&priv->params, + MLX5_CQ_PERIOD_MODE_START_FROM_EQE); + + priv->params.tx_cq_moderation.usec = MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_USEC; - priv->params.tx_cq_moderation_pkts = + priv->params.tx_cq_moderation.pkts = MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_PKTS; priv->params.tx_max_inline = mlx5e_get_max_inline_cap(mdev); priv->params.num_tc = 1; @@ -2929,6 +2949,10 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, priv->params.lro_wqe_sz = MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ; + /* Initialize pflags */ + MLX5E_SET_PRIV_FLAG(priv, MLX5E_PFLAG_RX_CQE_BASED_MODER, + priv->params.rx_cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE); + priv->mdev = mdev; priv->netdev = netdev; priv->params.num_channels = num_channels; -- cgit v0.10.2 From cb3c7fd4f8393e0c42cbb13367b60454ae4e05f7 Mon Sep 17 00:00:00 2001 From: Gil Rockah Date: Thu, 23 Jun 2016 17:02:41 +0300 Subject: net/mlx5e: Support adaptive RX coalescing Striving for high message rate and low interrupt rate. Usage: ethtool -C adaptive-rx on/off Signed-off-by: Gil Rockah Signed-off-by: Achiad Shochat Signed-off-by: Saeed Mahameed CC: Arnd Bergmann Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 0c8a7dc..c4f450f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -7,6 +7,7 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \ en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \ - en_txrx.o en_clock.o vxlan.o en_tc.o en_arfs.o + en_rx_am.o en_txrx.o en_clock.o vxlan.o en_tc.o \ + en_arfs.o mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 36f625d..aa36a3a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -195,6 +195,7 @@ struct mlx5e_params { #ifdef CONFIG_MLX5_CORE_EN_DCB struct ieee_ets ets; #endif + bool rx_am_enabled; }; struct mlx5e_tstamp { @@ -213,6 +214,7 @@ struct mlx5e_tstamp { enum { MLX5E_RQ_STATE_POST_WQES_ENABLE, MLX5E_RQ_STATE_UMR_WQE_IN_PROGRESS, + MLX5E_RQ_STATE_AM, }; struct mlx5e_cq { @@ -220,6 +222,7 @@ struct mlx5e_cq { struct mlx5_cqwq wq; /* data path - accessed per napi poll */ + u16 event_ctr; struct napi_struct *napi; struct mlx5_core_cq mcq; struct mlx5e_channel *channel; @@ -247,6 +250,30 @@ struct mlx5e_dma_info { dma_addr_t addr; }; +struct mlx5e_rx_am_stats { + int ppms; /* packets per msec */ + int epms; /* events per msec */ +}; + +struct mlx5e_rx_am_sample { + ktime_t time; + unsigned int pkt_ctr; + u16 event_ctr; +}; + +struct mlx5e_rx_am { /* Adaptive Moderation */ + u8 state; + struct mlx5e_rx_am_stats prev_stats; + struct mlx5e_rx_am_sample start_sample; + struct work_struct work; + u8 profile_ix; + u8 mode; + u8 tune_state; + u8 steps_right; + u8 steps_left; + u8 tired; +}; + struct mlx5e_rq { /* data path */ struct mlx5_wq_ll wq; @@ -267,6 +294,8 @@ struct mlx5e_rq { unsigned long state; int ix; + struct mlx5e_rx_am am; /* Adaptive Moderation */ + /* control */ struct mlx5_wq_ctrl wq_ctrl; u8 wq_type; @@ -637,6 +666,10 @@ void mlx5e_free_rx_fragmented_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi); struct mlx5_cqe64 *mlx5e_get_cqe(struct mlx5e_cq *cq); +void mlx5e_rx_am(struct mlx5e_rq *rq); +void mlx5e_rx_am_work(struct work_struct *work); +struct mlx5e_cq_moder mlx5e_am_get_def_profile(u8 rx_cq_period_mode); + void mlx5e_update_stats(struct mlx5e_priv *priv); int mlx5e_create_flow_steering(struct mlx5e_priv *priv); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 4f433d3..c4be394 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -528,6 +528,7 @@ static int mlx5e_get_coalesce(struct net_device *netdev, coal->rx_max_coalesced_frames = priv->params.rx_cq_moderation.pkts; coal->tx_coalesce_usecs = priv->params.tx_cq_moderation.usec; coal->tx_max_coalesced_frames = priv->params.tx_cq_moderation.pkts; + coal->use_adaptive_rx_coalesce = priv->params.rx_am_enabled; return 0; } @@ -538,6 +539,10 @@ static int mlx5e_set_coalesce(struct net_device *netdev, struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; struct mlx5e_channel *c; + bool restart = + !!coal->use_adaptive_rx_coalesce != priv->params.rx_am_enabled; + bool was_opened; + int err = 0; int tc; int i; @@ -546,12 +551,18 @@ static int mlx5e_set_coalesce(struct net_device *netdev, mutex_lock(&priv->state_lock); + was_opened = test_bit(MLX5E_STATE_OPENED, &priv->state); + if (was_opened && restart) { + mlx5e_close_locked(netdev); + priv->params.rx_am_enabled = !!coal->use_adaptive_rx_coalesce; + } + priv->params.tx_cq_moderation.usec = coal->tx_coalesce_usecs; priv->params.tx_cq_moderation.pkts = coal->tx_max_coalesced_frames; priv->params.rx_cq_moderation.usec = coal->rx_coalesce_usecs; priv->params.rx_cq_moderation.pkts = coal->rx_max_coalesced_frames; - if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) + if (!was_opened || restart) goto out; for (i = 0; i < priv->params.num_channels; ++i) { @@ -570,8 +581,11 @@ static int mlx5e_set_coalesce(struct net_device *netdev, } out: + if (was_opened && restart) + err = mlx5e_open_locked(netdev); + mutex_unlock(&priv->state_lock); - return 0; + return err; } static u32 ptys2ethtool_supported_link(u32 eth_proto_cap) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 13e7a45..39c0686 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -40,8 +40,9 @@ #include "vxlan.h" struct mlx5e_rq_param { - u32 rqc[MLX5_ST_SZ_DW(rqc)]; - struct mlx5_wq_param wq; + u32 rqc[MLX5_ST_SZ_DW(rqc)]; + struct mlx5_wq_param wq; + bool am_enabled; }; struct mlx5e_sq_param { @@ -337,6 +338,9 @@ static int mlx5e_create_rq(struct mlx5e_channel *c, wqe->data.byte_count = cpu_to_be32(byte_count); } + INIT_WORK(&rq->am.work, mlx5e_rx_am_work); + rq->am.mode = priv->params.rx_cq_period_mode; + rq->wq_type = priv->params.rq_wq_type; rq->pdev = c->pdev; rq->netdev = c->netdev; @@ -509,6 +513,9 @@ static int mlx5e_open_rq(struct mlx5e_channel *c, if (err) goto err_disable_rq; + if (param->am_enabled) + set_bit(MLX5E_RQ_STATE_AM, &c->rq.state); + set_bit(MLX5E_RQ_STATE_POST_WQES_ENABLE, &rq->state); sq->ico_wqe_info[pi].opcode = MLX5_OPCODE_NOP; @@ -537,6 +544,8 @@ static void mlx5e_close_rq(struct mlx5e_rq *rq) /* avoid destroying rq before mlx5e_poll_rx_cq() is done with it */ napi_synchronize(&rq->channel->napi); + cancel_work_sync(&rq->am.work); + mlx5e_disable_rq(rq); mlx5e_destroy_rq(rq); } @@ -1112,6 +1121,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, { struct mlx5e_cq_moder icosq_cq_moder = {0, 0}; struct net_device *netdev = priv->netdev; + struct mlx5e_cq_moder rx_cq_profile; int cpu = mlx5e_get_cpu(priv, ix); struct mlx5e_channel *c; struct mlx5e_sq *sq; @@ -1130,6 +1140,11 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->mkey_be = cpu_to_be32(priv->mkey.key); c->num_tc = priv->params.num_tc; + if (priv->params.rx_am_enabled) + rx_cq_profile = mlx5e_am_get_def_profile(priv->params.rx_cq_period_mode); + else + rx_cq_profile = priv->params.rx_cq_moderation; + mlx5e_build_channeltc_to_txq_map(priv, ix); netif_napi_add(netdev, &c->napi, mlx5e_napi_poll, 64); @@ -1143,7 +1158,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, goto err_close_icosq_cq; err = mlx5e_open_cq(c, &cparam->rx_cq, &c->rq.cq, - priv->params.rx_cq_moderation); + rx_cq_profile); if (err) goto err_close_tx_cqs; @@ -1243,6 +1258,8 @@ static void mlx5e_build_rq_param(struct mlx5e_priv *priv, param->wq.buf_numa_node = dev_to_node(&priv->mdev->pdev->dev); param->wq.linear = 1; + + param->am_enabled = priv->params.rx_am_enabled; } static void mlx5e_build_drop_rq_param(struct mlx5e_rq_param *param) @@ -2883,6 +2900,9 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, struct mlx5e_priv *priv = netdev_priv(netdev); u32 link_speed = 0; u32 pci_bw = 0; + u8 cq_period_mode = MLX5_CAP_GEN(mdev, cq_period_start_from_cqe) ? + MLX5_CQ_PERIOD_MODE_START_FROM_CQE : + MLX5_CQ_PERIOD_MODE_START_FROM_EQE; priv->params.log_sq_size = MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE; @@ -2929,8 +2949,8 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, priv->params.min_rx_wqes = mlx5_min_rx_wqes(priv->params.rq_wq_type, BIT(priv->params.log_rq_size)); - mlx5e_set_rx_cq_mode_params(&priv->params, - MLX5_CQ_PERIOD_MODE_START_FROM_EQE); + priv->params.rx_am_enabled = MLX5_CAP_GEN(mdev, cq_moderation); + mlx5e_set_rx_cq_mode_params(&priv->params, cq_period_mode); priv->params.tx_cq_moderation.usec = MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_USEC; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c new file mode 100644 index 0000000..1fffe48 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c @@ -0,0 +1,335 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "en.h" + +/* Adaptive moderation profiles */ +#define MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE 256 +#define MLX5E_RX_AM_DEF_PROFILE_CQE 1 +#define MLX5E_RX_AM_DEF_PROFILE_EQE 1 +#define MLX5E_PARAMS_AM_NUM_PROFILES 5 + +/* All profiles sizes must be MLX5E_PARAMS_AM_NUM_PROFILES */ +#define MLX5_AM_EQE_PROFILES { \ + {1, MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE}, \ + {8, MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE}, \ + {64, MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE}, \ + {128, MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE}, \ + {256, MLX5E_AM_DEFAULT_RX_CQ_MODERATION_PKTS_FROM_EQE}, \ +} + +#define MLX5_AM_CQE_PROFILES { \ + {2, 256}, \ + {8, 128}, \ + {16, 64}, \ + {32, 64}, \ + {64, 64} \ +} + +static const struct mlx5e_cq_moder +profile[MLX5_CQ_PERIOD_NUM_MODES][MLX5E_PARAMS_AM_NUM_PROFILES] = { + MLX5_AM_EQE_PROFILES, + MLX5_AM_CQE_PROFILES, +}; + +static inline struct mlx5e_cq_moder mlx5e_am_get_profile(u8 cq_period_mode, int ix) +{ + return profile[cq_period_mode][ix]; +} + +struct mlx5e_cq_moder mlx5e_am_get_def_profile(u8 rx_cq_period_mode) +{ + int default_profile_ix; + + if (rx_cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE) + default_profile_ix = MLX5E_RX_AM_DEF_PROFILE_CQE; + else /* MLX5_CQ_PERIOD_MODE_START_FROM_EQE */ + default_profile_ix = MLX5E_RX_AM_DEF_PROFILE_EQE; + + return profile[rx_cq_period_mode][default_profile_ix]; +} + +/* Adaptive moderation logic */ +enum { + MLX5E_AM_START_MEASURE, + MLX5E_AM_MEASURE_IN_PROGRESS, + MLX5E_AM_APPLY_NEW_PROFILE, +}; + +enum { + MLX5E_AM_PARKING_ON_TOP, + MLX5E_AM_PARKING_TIRED, + MLX5E_AM_GOING_RIGHT, + MLX5E_AM_GOING_LEFT, +}; + +enum { + MLX5E_AM_STATS_WORSE, + MLX5E_AM_STATS_SAME, + MLX5E_AM_STATS_BETTER, +}; + +enum { + MLX5E_AM_STEPPED, + MLX5E_AM_TOO_TIRED, + MLX5E_AM_ON_EDGE, +}; + +static bool mlx5e_am_on_top(struct mlx5e_rx_am *am) +{ + switch (am->tune_state) { + case MLX5E_AM_PARKING_ON_TOP: + case MLX5E_AM_PARKING_TIRED: + WARN_ONCE(true, "mlx5e_am_on_top: PARKING\n"); + return true; + case MLX5E_AM_GOING_RIGHT: + return (am->steps_left > 1) && (am->steps_right == 1); + default: /* MLX5E_AM_GOING_LEFT */ + return (am->steps_right > 1) && (am->steps_left == 1); + } +} + +static void mlx5e_am_turn(struct mlx5e_rx_am *am) +{ + switch (am->tune_state) { + case MLX5E_AM_PARKING_ON_TOP: + case MLX5E_AM_PARKING_TIRED: + WARN_ONCE(true, "mlx5e_am_turn: PARKING\n"); + break; + case MLX5E_AM_GOING_RIGHT: + am->tune_state = MLX5E_AM_GOING_LEFT; + am->steps_left = 0; + break; + case MLX5E_AM_GOING_LEFT: + am->tune_state = MLX5E_AM_GOING_RIGHT; + am->steps_right = 0; + break; + } +} + +static int mlx5e_am_step(struct mlx5e_rx_am *am) +{ + if (am->tired == (MLX5E_PARAMS_AM_NUM_PROFILES * 2)) + return MLX5E_AM_TOO_TIRED; + + switch (am->tune_state) { + case MLX5E_AM_PARKING_ON_TOP: + case MLX5E_AM_PARKING_TIRED: + WARN_ONCE(true, "mlx5e_am_step: PARKING\n"); + break; + case MLX5E_AM_GOING_RIGHT: + if (am->profile_ix == (MLX5E_PARAMS_AM_NUM_PROFILES - 1)) + return MLX5E_AM_ON_EDGE; + am->profile_ix++; + am->steps_right++; + break; + case MLX5E_AM_GOING_LEFT: + if (am->profile_ix == 0) + return MLX5E_AM_ON_EDGE; + am->profile_ix--; + am->steps_left++; + break; + } + + am->tired++; + return MLX5E_AM_STEPPED; +} + +static void mlx5e_am_park_on_top(struct mlx5e_rx_am *am) +{ + am->steps_right = 0; + am->steps_left = 0; + am->tired = 0; + am->tune_state = MLX5E_AM_PARKING_ON_TOP; +} + +static void mlx5e_am_park_tired(struct mlx5e_rx_am *am) +{ + am->steps_right = 0; + am->steps_left = 0; + am->tune_state = MLX5E_AM_PARKING_TIRED; +} + +static void mlx5e_am_exit_parking(struct mlx5e_rx_am *am) +{ + am->tune_state = am->profile_ix ? MLX5E_AM_GOING_LEFT : + MLX5E_AM_GOING_RIGHT; + mlx5e_am_step(am); +} + +static int mlx5e_am_stats_compare(struct mlx5e_rx_am_stats *curr, + struct mlx5e_rx_am_stats *prev) +{ + int diff; + + if (!prev->ppms) + return curr->ppms ? MLX5E_AM_STATS_BETTER : + MLX5E_AM_STATS_SAME; + + diff = curr->ppms - prev->ppms; + if (((100 * abs(diff)) / prev->ppms) > 10) /* more than 10% diff */ + return (diff > 0) ? MLX5E_AM_STATS_BETTER : + MLX5E_AM_STATS_WORSE; + + if (!prev->epms) + return curr->epms ? MLX5E_AM_STATS_WORSE : + MLX5E_AM_STATS_SAME; + + diff = curr->epms - prev->epms; + if (((100 * abs(diff)) / prev->epms) > 10) /* more than 10% diff */ + return (diff < 0) ? MLX5E_AM_STATS_BETTER : + MLX5E_AM_STATS_WORSE; + + return MLX5E_AM_STATS_SAME; +} + +static bool mlx5e_am_decision(struct mlx5e_rx_am_stats *curr_stats, + struct mlx5e_rx_am *am) +{ + int prev_state = am->tune_state; + int prev_ix = am->profile_ix; + int stats_res; + int step_res; + + switch (am->tune_state) { + case MLX5E_AM_PARKING_ON_TOP: + stats_res = mlx5e_am_stats_compare(curr_stats, &am->prev_stats); + if (stats_res != MLX5E_AM_STATS_SAME) + mlx5e_am_exit_parking(am); + break; + + case MLX5E_AM_PARKING_TIRED: + am->tired--; + if (!am->tired) + mlx5e_am_exit_parking(am); + break; + + case MLX5E_AM_GOING_RIGHT: + case MLX5E_AM_GOING_LEFT: + stats_res = mlx5e_am_stats_compare(curr_stats, &am->prev_stats); + if (stats_res != MLX5E_AM_STATS_BETTER) + mlx5e_am_turn(am); + + if (mlx5e_am_on_top(am)) { + mlx5e_am_park_on_top(am); + break; + } + + step_res = mlx5e_am_step(am); + switch (step_res) { + case MLX5E_AM_ON_EDGE: + mlx5e_am_park_on_top(am); + break; + case MLX5E_AM_TOO_TIRED: + mlx5e_am_park_tired(am); + break; + } + + break; + } + + if ((prev_state != MLX5E_AM_PARKING_ON_TOP) || + (am->tune_state != MLX5E_AM_PARKING_ON_TOP)) + am->prev_stats = *curr_stats; + + return am->profile_ix != prev_ix; +} + +static void mlx5e_am_sample(struct mlx5e_rq *rq, + struct mlx5e_rx_am_sample *s) +{ + s->time = ktime_get(); + s->pkt_ctr = rq->stats.packets; + s->event_ctr = rq->cq.event_ctr; +} + +#define MLX5E_AM_NEVENTS 64 + +static void mlx5e_am_calc_stats(struct mlx5e_rx_am_sample *start, + struct mlx5e_rx_am_sample *end, + struct mlx5e_rx_am_stats *curr_stats) +{ + /* u32 holds up to 71 minutes, should be enough */ + u32 delta_us = ktime_us_delta(end->time, start->time); + unsigned int npkts = end->pkt_ctr - start->pkt_ctr; + + if (!delta_us) { + WARN_ONCE(true, "mlx5e_am_calc_stats: delta_us=0\n"); + return; + } + + curr_stats->ppms = (npkts * USEC_PER_MSEC) / delta_us; + curr_stats->epms = (MLX5E_AM_NEVENTS * USEC_PER_MSEC) / delta_us; +} + +void mlx5e_rx_am_work(struct work_struct *work) +{ + struct mlx5e_rx_am *am = container_of(work, struct mlx5e_rx_am, + work); + struct mlx5e_rq *rq = container_of(am, struct mlx5e_rq, am); + struct mlx5e_cq_moder cur_profile = profile[am->mode][am->profile_ix]; + + mlx5_core_modify_cq_moderation(rq->priv->mdev, &rq->cq.mcq, + cur_profile.usec, cur_profile.pkts); + + am->state = MLX5E_AM_START_MEASURE; +} + +void mlx5e_rx_am(struct mlx5e_rq *rq) +{ + struct mlx5e_rx_am *am = &rq->am; + struct mlx5e_rx_am_sample end_sample; + struct mlx5e_rx_am_stats curr_stats; + u16 nevents; + + switch (am->state) { + case MLX5E_AM_MEASURE_IN_PROGRESS: + nevents = rq->cq.event_ctr - am->start_sample.event_ctr; + if (nevents < MLX5E_AM_NEVENTS) + break; + mlx5e_am_sample(rq, &end_sample); + mlx5e_am_calc_stats(&am->start_sample, &end_sample, + &curr_stats); + if (mlx5e_am_decision(&curr_stats, am)) { + am->state = MLX5E_AM_APPLY_NEW_PROFILE; + schedule_work(&am->work); + break; + } + /* fall through */ + case MLX5E_AM_START_MEASURE: + mlx5e_am_sample(rq, &am->start_sample); + am->state = MLX5E_AM_MEASURE_IN_PROGRESS; + break; + case MLX5E_AM_APPLY_NEW_PROFILE: + break; + } +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c index c38781f..64ae2e8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c @@ -136,6 +136,10 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget) for (i = 0; i < c->num_tc; i++) mlx5e_cq_arm(&c->sq[i].cq); + + if (test_bit(MLX5E_RQ_STATE_AM, &c->rq.state)) + mlx5e_rx_am(&c->rq); + mlx5e_cq_arm(&c->rq.cq); mlx5e_cq_arm(&c->icosq.cq); @@ -146,6 +150,7 @@ void mlx5e_completion_event(struct mlx5_core_cq *mcq) { struct mlx5e_cq *cq = container_of(mcq, struct mlx5e_cq, mcq); + cq->event_ctr++; set_bit(MLX5E_CHANNEL_NAPI_SCHED, &cq->channel->flags); napi_schedule(cq->napi); } -- cgit v0.10.2 From 667daedaecd15b89d0ded7af49519f28d6ea2cf4 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 23 Jun 2016 17:02:42 +0300 Subject: net/mlx5e: Toggle link only after modifying port parameters Add a dedicated function to toggle port link. It should be called only after setting a port register. Toggle will set port link to down and bring it back up in case that it's admin status was up. Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index b2db180..e688313 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -191,7 +191,6 @@ static int mlx5e_dcbnl_ieee_setpfc(struct net_device *dev, { struct mlx5e_priv *priv = netdev_priv(dev); struct mlx5_core_dev *mdev = priv->mdev; - enum mlx5_port_status ps; u8 curr_pfc_en; int ret; @@ -200,14 +199,8 @@ static int mlx5e_dcbnl_ieee_setpfc(struct net_device *dev, if (pfc->pfc_en == curr_pfc_en) return 0; - mlx5_query_port_admin_status(mdev, &ps); - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_DOWN); - ret = mlx5_set_port_pfc(mdev, pfc->pfc_en, pfc->pfc_en); - - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_UP); + mlx5_toggle_port_link(mdev); return ret; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index c4be394..d0d3dcf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -795,7 +795,6 @@ static int mlx5e_set_settings(struct net_device *netdev, u32 link_modes; u32 speed; u32 eth_proto_cap, eth_proto_admin; - enum mlx5_port_status ps; int err; speed = ethtool_cmd_speed(cmd); @@ -829,12 +828,8 @@ static int mlx5e_set_settings(struct net_device *netdev, if (link_modes == eth_proto_admin) goto out; - mlx5_query_port_admin_status(mdev, &ps); - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_DOWN); mlx5_set_port_proto(mdev, link_modes, MLX5_PTYS_EN); - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_UP); + mlx5_toggle_port_link(mdev); out: return err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index 3e35611..1562e73 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -222,6 +222,18 @@ int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, } EXPORT_SYMBOL_GPL(mlx5_set_port_proto); +/* This function should be used after setting a port register only */ +void mlx5_toggle_port_link(struct mlx5_core_dev *dev) +{ + enum mlx5_port_status ps; + + mlx5_query_port_admin_status(dev, &ps); + mlx5_set_port_admin_status(dev, MLX5_PORT_DOWN); + if (ps == MLX5_PORT_UP) + mlx5_set_port_admin_status(dev, MLX5_PORT_UP); +} +EXPORT_SYMBOL_GPL(mlx5_toggle_port_link); + int mlx5_set_port_admin_status(struct mlx5_core_dev *dev, enum mlx5_port_status status) { diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h index 9851862..4adfac1 100644 --- a/include/linux/mlx5/port.h +++ b/include/linux/mlx5/port.h @@ -67,6 +67,7 @@ int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev, u8 local_port); int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, int proto_mask); +void mlx5_toggle_port_link(struct mlx5_core_dev *dev); int mlx5_set_port_admin_status(struct mlx5_core_dev *dev, enum mlx5_port_status status); int mlx5_query_port_admin_status(struct mlx5_core_dev *dev, -- cgit v0.10.2 From 89da45b8b5b2187734a11038b8593714f964ffd1 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 23 Jun 2016 17:02:43 +0300 Subject: ethtool: Add 50G baseSR2 link mode Add ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT bit. Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Cc: Ben Hutchings Cc: David Decotigny Acked-By: David Decotigny Signed-off-by: David S. Miller diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h index 5f030b4..b8f38e8 100644 --- a/include/uapi/linux/ethtool.h +++ b/include/uapi/linux/ethtool.h @@ -1362,6 +1362,7 @@ enum ethtool_link_mode_bit_indices { ETHTOOL_LINK_MODE_100000baseSR4_Full_BIT = 37, ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT = 38, ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT = 39, + ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT = 40, /* Last allowed bit for __ETHTOOL_LINK_MODE_LEGACY_MASK is bit * 31. Please do NOT define any SUPPORTED_* or ADVERTISED_* @@ -1370,7 +1371,7 @@ enum ethtool_link_mode_bit_indices { */ __ETHTOOL_LINK_MODE_LAST - = ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT, + = ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT, }; #define __ETHTOOL_LINK_MODE_LEGACY_MASK(base_name) \ -- cgit v0.10.2 From 4a50e35b04c9008558a73ed4e349b3b483ef6739 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 23 Jun 2016 17:02:44 +0300 Subject: net/mlx5e: Add missing 50G baseSR2 link mode Add MLX5E_50GBASE_SR2 as ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT. Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Cc: Ben Hutchings Cc: David Decotigny Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index aa36a3a..b8732e6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -616,6 +616,7 @@ enum mlx5e_link_mode { MLX5E_10GBASE_ER = 14, MLX5E_40GBASE_SR4 = 15, MLX5E_40GBASE_LR4 = 16, + MLX5E_50GBASE_SR2 = 18, MLX5E_100GBASE_CR4 = 20, MLX5E_100GBASE_SR4 = 21, MLX5E_100GBASE_KR4 = 22, -- cgit v0.10.2 From 665bc53969d79af9cfd080e25b91b0415a2b5eec Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 23 Jun 2016 17:02:45 +0300 Subject: net/mlx5e: Use new ethtool get/set link ksettings API Use new get/set link ksettings and remove get/set settings legacy callbacks. This allows us to use bitmasks longer than 32 bit for supported and advertised link modes and use modes that were previously not supported. Signed-off-by: Gal Pressman CC: Ben Hutchings CC: David Decotigny Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index b8732e6..da885c0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -634,6 +634,9 @@ enum mlx5e_link_mode { #define MLX5E_PROT_MASK(link_mode) (1 << link_mode) + +void mlx5e_build_ptys2ethtool_map(void); + void mlx5e_send_nop(struct mlx5e_sq *sq, bool notify_hw); u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, void *accel_priv, select_queue_fallback_t fallback); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index d0d3dcf..4c560e0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -48,123 +48,85 @@ static void mlx5e_get_drvinfo(struct net_device *dev, sizeof(drvinfo->bus_info)); } -static const struct { - u32 supported; - u32 advertised; +struct ptys2ethtool_config { + __ETHTOOL_DECLARE_LINK_MODE_MASK(supported); + __ETHTOOL_DECLARE_LINK_MODE_MASK(advertised); u32 speed; -} ptys2ethtool_table[MLX5E_LINK_MODES_NUMBER] = { - [MLX5E_1000BASE_CX_SGMII] = { - .supported = SUPPORTED_1000baseKX_Full, - .advertised = ADVERTISED_1000baseKX_Full, - .speed = 1000, - }, - [MLX5E_1000BASE_KX] = { - .supported = SUPPORTED_1000baseKX_Full, - .advertised = ADVERTISED_1000baseKX_Full, - .speed = 1000, - }, - [MLX5E_10GBASE_CX4] = { - .supported = SUPPORTED_10000baseKX4_Full, - .advertised = ADVERTISED_10000baseKX4_Full, - .speed = 10000, - }, - [MLX5E_10GBASE_KX4] = { - .supported = SUPPORTED_10000baseKX4_Full, - .advertised = ADVERTISED_10000baseKX4_Full, - .speed = 10000, - }, - [MLX5E_10GBASE_KR] = { - .supported = SUPPORTED_10000baseKR_Full, - .advertised = ADVERTISED_10000baseKR_Full, - .speed = 10000, - }, - [MLX5E_20GBASE_KR2] = { - .supported = SUPPORTED_20000baseKR2_Full, - .advertised = ADVERTISED_20000baseKR2_Full, - .speed = 20000, - }, - [MLX5E_40GBASE_CR4] = { - .supported = SUPPORTED_40000baseCR4_Full, - .advertised = ADVERTISED_40000baseCR4_Full, - .speed = 40000, - }, - [MLX5E_40GBASE_KR4] = { - .supported = SUPPORTED_40000baseKR4_Full, - .advertised = ADVERTISED_40000baseKR4_Full, - .speed = 40000, - }, - [MLX5E_56GBASE_R4] = { - .supported = SUPPORTED_56000baseKR4_Full, - .advertised = ADVERTISED_56000baseKR4_Full, - .speed = 56000, - }, - [MLX5E_10GBASE_CR] = { - .supported = SUPPORTED_10000baseKR_Full, - .advertised = ADVERTISED_10000baseKR_Full, - .speed = 10000, - }, - [MLX5E_10GBASE_SR] = { - .supported = SUPPORTED_10000baseKR_Full, - .advertised = ADVERTISED_10000baseKR_Full, - .speed = 10000, - }, - [MLX5E_10GBASE_ER] = { - .supported = SUPPORTED_10000baseKR_Full, - .advertised = ADVERTISED_10000baseKR_Full, - .speed = 10000, - }, - [MLX5E_40GBASE_SR4] = { - .supported = SUPPORTED_40000baseSR4_Full, - .advertised = ADVERTISED_40000baseSR4_Full, - .speed = 40000, - }, - [MLX5E_40GBASE_LR4] = { - .supported = SUPPORTED_40000baseLR4_Full, - .advertised = ADVERTISED_40000baseLR4_Full, - .speed = 40000, - }, - [MLX5E_100GBASE_CR4] = { - .speed = 100000, - }, - [MLX5E_100GBASE_SR4] = { - .speed = 100000, - }, - [MLX5E_100GBASE_KR4] = { - .speed = 100000, - }, - [MLX5E_100GBASE_LR4] = { - .speed = 100000, - }, - [MLX5E_100BASE_TX] = { - .speed = 100, - }, - [MLX5E_1000BASE_T] = { - .supported = SUPPORTED_1000baseT_Full, - .advertised = ADVERTISED_1000baseT_Full, - .speed = 1000, - }, - [MLX5E_10GBASE_T] = { - .supported = SUPPORTED_10000baseT_Full, - .advertised = ADVERTISED_10000baseT_Full, - .speed = 1000, - }, - [MLX5E_25GBASE_CR] = { - .speed = 25000, - }, - [MLX5E_25GBASE_KR] = { - .speed = 25000, - }, - [MLX5E_25GBASE_SR] = { - .speed = 25000, - }, - [MLX5E_50GBASE_CR2] = { - .speed = 50000, - }, - [MLX5E_50GBASE_KR2] = { - .speed = 50000, - }, }; +static struct ptys2ethtool_config ptys2ethtool_table[MLX5E_LINK_MODES_NUMBER]; + +#define MLX5_BUILD_PTYS2ETHTOOL_CONFIG(reg_, speed_, ...) \ + ({ \ + struct ptys2ethtool_config *cfg; \ + const unsigned int modes[] = { __VA_ARGS__ }; \ + unsigned int i; \ + cfg = &ptys2ethtool_table[reg_]; \ + cfg->speed = speed_; \ + bitmap_zero(cfg->supported, \ + __ETHTOOL_LINK_MODE_MASK_NBITS); \ + bitmap_zero(cfg->advertised, \ + __ETHTOOL_LINK_MODE_MASK_NBITS); \ + for (i = 0 ; i < ARRAY_SIZE(modes) ; ++i) { \ + __set_bit(modes[i], cfg->supported); \ + __set_bit(modes[i], cfg->advertised); \ + } \ + }) + +void mlx5e_build_ptys2ethtool_map(void) +{ + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_1000BASE_CX_SGMII, SPEED_1000, + ETHTOOL_LINK_MODE_1000baseKX_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_1000BASE_KX, SPEED_1000, + ETHTOOL_LINK_MODE_1000baseKX_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_CX4, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_KX4, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_KR, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_20GBASE_KR2, SPEED_20000, + ETHTOOL_LINK_MODE_20000baseKR2_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_40GBASE_CR4, SPEED_40000, + ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_40GBASE_KR4, SPEED_40000, + ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_56GBASE_R4, SPEED_56000, + ETHTOOL_LINK_MODE_56000baseKR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_CR, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_SR, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_ER, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseKR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_40GBASE_SR4, SPEED_40000, + ETHTOOL_LINK_MODE_40000baseSR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_40GBASE_LR4, SPEED_40000, + ETHTOOL_LINK_MODE_40000baseLR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_50GBASE_SR2, SPEED_50000, + ETHTOOL_LINK_MODE_50000baseSR2_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_100GBASE_CR4, SPEED_100000, + ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_100GBASE_SR4, SPEED_100000, + ETHTOOL_LINK_MODE_100000baseSR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_100GBASE_KR4, SPEED_100000, + ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_100GBASE_LR4, SPEED_100000, + ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_10GBASE_T, SPEED_10000, + ETHTOOL_LINK_MODE_10000baseT_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_25GBASE_CR, SPEED_25000, + ETHTOOL_LINK_MODE_25000baseCR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_25GBASE_KR, SPEED_25000, + ETHTOOL_LINK_MODE_25000baseKR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_25GBASE_SR, SPEED_25000, + ETHTOOL_LINK_MODE_25000baseSR_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_50GBASE_CR2, SPEED_50000, + ETHTOOL_LINK_MODE_50000baseCR2_Full_BIT); + MLX5_BUILD_PTYS2ETHTOOL_CONFIG(MLX5E_50GBASE_KR2, SPEED_50000, + ETHTOOL_LINK_MODE_50000baseKR2_Full_BIT); +} + static unsigned long mlx5e_query_pfc_combined(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; @@ -588,31 +550,30 @@ out: return err; } -static u32 ptys2ethtool_supported_link(u32 eth_proto_cap) +static void ptys2ethtool_supported_link(unsigned long *supported_modes, + u32 eth_proto_cap) { - int i; - u32 supported_modes = 0; + int proto; - for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { - if (eth_proto_cap & MLX5E_PROT_MASK(i)) - supported_modes |= ptys2ethtool_table[i].supported; - } - return supported_modes; + for_each_set_bit(proto, (unsigned long *)ð_proto_cap, MLX5E_LINK_MODES_NUMBER) + bitmap_or(supported_modes, supported_modes, + ptys2ethtool_table[proto].supported, + __ETHTOOL_LINK_MODE_MASK_NBITS); } -static u32 ptys2ethtool_adver_link(u32 eth_proto_cap) +static void ptys2ethtool_adver_link(unsigned long *advertising_modes, + u32 eth_proto_cap) { - int i; - u32 advertising_modes = 0; + int proto; - for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { - if (eth_proto_cap & MLX5E_PROT_MASK(i)) - advertising_modes |= ptys2ethtool_table[i].advertised; - } - return advertising_modes; + for_each_set_bit(proto, (unsigned long *)ð_proto_cap, MLX5E_LINK_MODES_NUMBER) + bitmap_or(advertising_modes, advertising_modes, + ptys2ethtool_table[proto].advertised, + __ETHTOOL_LINK_MODE_MASK_NBITS); } -static u32 ptys2ethtool_supported_port(u32 eth_proto_cap) +static void ptys2ethtool_supported_port(struct ethtool_link_ksettings *link_ksettings, + u32 eth_proto_cap) { if (eth_proto_cap & (MLX5E_PROT_MASK(MLX5E_10GBASE_CR) | MLX5E_PROT_MASK(MLX5E_10GBASE_SR) @@ -620,7 +581,7 @@ static u32 ptys2ethtool_supported_port(u32 eth_proto_cap) | MLX5E_PROT_MASK(MLX5E_40GBASE_SR4) | MLX5E_PROT_MASK(MLX5E_100GBASE_SR4) | MLX5E_PROT_MASK(MLX5E_1000BASE_CX_SGMII))) { - return SUPPORTED_FIBRE; + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, FIBRE); } if (eth_proto_cap & (MLX5E_PROT_MASK(MLX5E_100GBASE_KR4) @@ -628,9 +589,8 @@ static u32 ptys2ethtool_supported_port(u32 eth_proto_cap) | MLX5E_PROT_MASK(MLX5E_10GBASE_KR) | MLX5E_PROT_MASK(MLX5E_10GBASE_KX4) | MLX5E_PROT_MASK(MLX5E_1000BASE_KX))) { - return SUPPORTED_Backplane; + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, Backplane); } - return 0; } int mlx5e_get_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed) @@ -654,7 +614,7 @@ int mlx5e_get_max_linkspeed(struct mlx5_core_dev *mdev, u32 *speed) static void get_speed_duplex(struct net_device *netdev, u32 eth_proto_oper, - struct ethtool_cmd *cmd) + struct ethtool_link_ksettings *link_ksettings) { int i; u32 speed = SPEED_UNKNOWN; @@ -671,23 +631,32 @@ static void get_speed_duplex(struct net_device *netdev, } } out: - ethtool_cmd_speed_set(cmd, speed); - cmd->duplex = duplex; + link_ksettings->base.speed = speed; + link_ksettings->base.duplex = duplex; } -static void get_supported(u32 eth_proto_cap, u32 *supported) +static void get_supported(u32 eth_proto_cap, + struct ethtool_link_ksettings *link_ksettings) { - *supported |= ptys2ethtool_supported_port(eth_proto_cap); - *supported |= ptys2ethtool_supported_link(eth_proto_cap); - *supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause; + unsigned long *supported = link_ksettings->link_modes.supported; + + ptys2ethtool_supported_port(link_ksettings, eth_proto_cap); + ptys2ethtool_supported_link(supported, eth_proto_cap); + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, Pause); + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, Asym_Pause); } static void get_advertising(u32 eth_proto_cap, u8 tx_pause, - u8 rx_pause, u32 *advertising) + u8 rx_pause, + struct ethtool_link_ksettings *link_ksettings) { - *advertising |= ptys2ethtool_adver_link(eth_proto_cap); - *advertising |= tx_pause ? ADVERTISED_Pause : 0; - *advertising |= (tx_pause ^ rx_pause) ? ADVERTISED_Asym_Pause : 0; + unsigned long *advertising = link_ksettings->link_modes.advertising; + + ptys2ethtool_adver_link(advertising, eth_proto_cap); + if (tx_pause) + ethtool_link_ksettings_add_link_mode(link_ksettings, advertising, Pause); + if (tx_pause ^ rx_pause) + ethtool_link_ksettings_add_link_mode(link_ksettings, advertising, Asym_Pause); } static u8 get_connector_port(u32 eth_proto) @@ -715,13 +684,16 @@ static u8 get_connector_port(u32 eth_proto) return PORT_OTHER; } -static void get_lp_advertising(u32 eth_proto_lp, u32 *lp_advertising) +static void get_lp_advertising(u32 eth_proto_lp, + struct ethtool_link_ksettings *link_ksettings) { - *lp_advertising = ptys2ethtool_adver_link(eth_proto_lp); + unsigned long *lp_advertising = link_ksettings->link_modes.lp_advertising; + + ptys2ethtool_adver_link(lp_advertising, eth_proto_lp); } -static int mlx5e_get_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) +static int mlx5e_get_link_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *link_ksettings) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; @@ -745,30 +717,30 @@ static int mlx5e_get_settings(struct net_device *netdev, eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper); eth_proto_lp = MLX5_GET(ptys_reg, out, eth_proto_lp_advertise); - cmd->supported = 0; - cmd->advertising = 0; + ethtool_link_ksettings_zero_link_mode(link_ksettings, supported); + ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising); - get_supported(eth_proto_cap, &cmd->supported); - get_advertising(eth_proto_admin, 0, 0, &cmd->advertising); - get_speed_duplex(netdev, eth_proto_oper, cmd); + get_supported(eth_proto_cap, link_ksettings); + get_advertising(eth_proto_admin, 0, 0, link_ksettings); + get_speed_duplex(netdev, eth_proto_oper, link_ksettings); eth_proto_oper = eth_proto_oper ? eth_proto_oper : eth_proto_cap; - cmd->port = get_connector_port(eth_proto_oper); - get_lp_advertising(eth_proto_lp, &cmd->lp_advertising); - - cmd->transceiver = XCVR_INTERNAL; + link_ksettings->base.port = get_connector_port(eth_proto_oper); + get_lp_advertising(eth_proto_lp, link_ksettings); err_query_ptys: return err; } -static u32 mlx5e_ethtool2ptys_adver_link(u32 link_modes) +static u32 mlx5e_ethtool2ptys_adver_link(const unsigned long *link_modes) { u32 i, ptys_modes = 0; for (i = 0; i < MLX5E_LINK_MODES_NUMBER; ++i) { - if (ptys2ethtool_table[i].advertised & link_modes) + if (bitmap_intersects(ptys2ethtool_table[i].advertised, + link_modes, + __ETHTOOL_LINK_MODE_MASK_NBITS)) ptys_modes |= MLX5E_PROT_MASK(i); } @@ -787,8 +759,8 @@ static u32 mlx5e_ethtool2ptys_speed_link(u32 speed) return speed_links; } -static int mlx5e_set_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) +static int mlx5e_set_link_ksettings(struct net_device *netdev, + const struct ethtool_link_ksettings *link_ksettings) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; @@ -797,10 +769,10 @@ static int mlx5e_set_settings(struct net_device *netdev, u32 eth_proto_cap, eth_proto_admin; int err; - speed = ethtool_cmd_speed(cmd); + speed = link_ksettings->base.speed; - link_modes = cmd->autoneg == AUTONEG_ENABLE ? - mlx5e_ethtool2ptys_adver_link(cmd->advertising) : + link_modes = link_ksettings->base.autoneg == AUTONEG_ENABLE ? + mlx5e_ethtool2ptys_adver_link(link_ksettings->link_modes.advertising) : mlx5e_ethtool2ptys_speed_link(speed); err = mlx5_query_port_proto_cap(mdev, ð_proto_cap, MLX5_PTYS_EN); @@ -1380,8 +1352,8 @@ const struct ethtool_ops mlx5e_ethtool_ops = { .set_channels = mlx5e_set_channels, .get_coalesce = mlx5e_get_coalesce, .set_coalesce = mlx5e_set_coalesce, - .get_settings = mlx5e_get_settings, - .set_settings = mlx5e_set_settings, + .get_link_ksettings = mlx5e_get_link_ksettings, + .set_link_ksettings = mlx5e_set_link_ksettings, .get_rxfh_key_size = mlx5e_get_rxfh_key_size, .get_rxfh_indir_size = mlx5e_get_rxfh_indir_size, .get_rxfh = mlx5e_get_rxfh, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 39c0686..02a0f17 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -3380,6 +3380,7 @@ static struct mlx5_interface mlx5e_interface = { void mlx5e_init(void) { + mlx5e_build_ptys2ethtool_map(); mlx5_register_interface(&mlx5e_interface); } -- cgit v0.10.2 From 52244d960755936fa9c8ce54d583d0ed46f24fb6 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 23 Jun 2016 17:02:46 +0300 Subject: net/mlx5e: Report correct auto negotiation and allow toggling Previous to this patch auto negotiation was reported off although it was on by default in hardware. This patch reports the correct information to ethtool and allows the user to toggle it on/off. Added another parameter to set port proto function in order to pass the auto negotiation field to the hardware. Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 4c560e0..39a4d96 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -702,6 +702,8 @@ static int mlx5e_get_link_ksettings(struct net_device *netdev, u32 eth_proto_admin; u32 eth_proto_lp; u32 eth_proto_oper; + u8 an_disable_admin; + u8 an_status; int err; err = mlx5_query_port_ptys(mdev, out, sizeof(out), MLX5_PTYS_EN, 1); @@ -712,10 +714,12 @@ static int mlx5e_get_link_ksettings(struct net_device *netdev, goto err_query_ptys; } - eth_proto_cap = MLX5_GET(ptys_reg, out, eth_proto_capability); - eth_proto_admin = MLX5_GET(ptys_reg, out, eth_proto_admin); - eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper); - eth_proto_lp = MLX5_GET(ptys_reg, out, eth_proto_lp_advertise); + eth_proto_cap = MLX5_GET(ptys_reg, out, eth_proto_capability); + eth_proto_admin = MLX5_GET(ptys_reg, out, eth_proto_admin); + eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper); + eth_proto_lp = MLX5_GET(ptys_reg, out, eth_proto_lp_advertise); + an_disable_admin = MLX5_GET(ptys_reg, out, an_disable_admin); + an_status = MLX5_GET(ptys_reg, out, an_status); ethtool_link_ksettings_zero_link_mode(link_ksettings, supported); ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising); @@ -729,6 +733,18 @@ static int mlx5e_get_link_ksettings(struct net_device *netdev, link_ksettings->base.port = get_connector_port(eth_proto_oper); get_lp_advertising(eth_proto_lp, link_ksettings); + if (an_status == MLX5_AN_COMPLETE) + ethtool_link_ksettings_add_link_mode(link_ksettings, + lp_advertising, Autoneg); + + link_ksettings->base.autoneg = an_disable_admin ? AUTONEG_DISABLE : + AUTONEG_ENABLE; + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, + Autoneg); + if (!an_disable_admin) + ethtool_link_ksettings_add_link_mode(link_ksettings, + advertising, Autoneg); + err_query_ptys: return err; } @@ -764,9 +780,14 @@ static int mlx5e_set_link_ksettings(struct net_device *netdev, { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; + u32 eth_proto_cap, eth_proto_admin; + bool an_changes = false; + u8 an_disable_admin; + u8 an_disable_cap; + bool an_disable; u32 link_modes; + u8 an_status; u32 speed; - u32 eth_proto_cap, eth_proto_admin; int err; speed = link_ksettings->base.speed; @@ -797,10 +818,17 @@ static int mlx5e_set_link_ksettings(struct net_device *netdev, goto out; } - if (link_modes == eth_proto_admin) + mlx5_query_port_autoneg(mdev, MLX5_PTYS_EN, &an_status, + &an_disable_cap, &an_disable_admin); + + an_disable = link_ksettings->base.autoneg == AUTONEG_DISABLE; + an_changes = ((!an_disable && an_disable_admin) || + (an_disable && !an_disable_admin)); + + if (!an_changes && link_modes == eth_proto_admin) goto out; - mlx5_set_port_proto(mdev, link_modes, MLX5_PTYS_EN); + mlx5_set_port_ptys(mdev, an_disable, link_modes, MLX5_PTYS_EN); mlx5_toggle_port_link(mdev); out: diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index 1562e73..752c081 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -202,15 +202,24 @@ int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev, } EXPORT_SYMBOL_GPL(mlx5_query_port_proto_oper); -int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, - int proto_mask) +int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable, + u32 proto_admin, int proto_mask) { - u32 in[MLX5_ST_SZ_DW(ptys_reg)]; u32 out[MLX5_ST_SZ_DW(ptys_reg)]; + u32 in[MLX5_ST_SZ_DW(ptys_reg)]; + u8 an_disable_admin; + u8 an_disable_cap; + u8 an_status; + + mlx5_query_port_autoneg(dev, proto_mask, &an_status, + &an_disable_cap, &an_disable_admin); + if (!an_disable_cap && an_disable) + return -EPERM; memset(in, 0, sizeof(in)); MLX5_SET(ptys_reg, in, local_port, 1); + MLX5_SET(ptys_reg, in, an_disable_admin, an_disable); MLX5_SET(ptys_reg, in, proto_mask, proto_mask); if (proto_mask == MLX5_PTYS_EN) MLX5_SET(ptys_reg, in, eth_proto_admin, proto_admin); @@ -220,7 +229,7 @@ int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, return mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out), MLX5_REG_PTYS, 0, 1); } -EXPORT_SYMBOL_GPL(mlx5_set_port_proto); +EXPORT_SYMBOL_GPL(mlx5_set_port_ptys); /* This function should be used after setting a port register only */ void mlx5_toggle_port_link(struct mlx5_core_dev *dev) @@ -530,6 +539,25 @@ int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx, u8 *pfc_en_rx) } EXPORT_SYMBOL_GPL(mlx5_query_port_pfc); +void mlx5_query_port_autoneg(struct mlx5_core_dev *dev, int proto_mask, + u8 *an_status, + u8 *an_disable_cap, u8 *an_disable_admin) +{ + u32 out[MLX5_ST_SZ_DW(ptys_reg)]; + + *an_status = 0; + *an_disable_cap = 0; + *an_disable_admin = 0; + + if (mlx5_query_port_ptys(dev, out, sizeof(out), proto_mask, 1)) + return; + + *an_status = MLX5_GET(ptys_reg, out, an_status); + *an_disable_cap = MLX5_GET(ptys_reg, out, an_disable_cap); + *an_disable_admin = MLX5_GET(ptys_reg, out, an_disable_admin); +} +EXPORT_SYMBOL_GPL(mlx5_query_port_autoneg); + int mlx5_max_tc(struct mlx5_core_dev *mdev) { u8 num_tc = MLX5_CAP_GEN(mdev, max_tc) ? : 8; diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h index 4adfac1..e3012cc 100644 --- a/include/linux/mlx5/port.h +++ b/include/linux/mlx5/port.h @@ -47,6 +47,14 @@ enum mlx5_module_id { MLX5_MODULE_ID_QSFP28 = 0x11, }; +enum mlx5_an_status { + MLX5_AN_UNAVAILABLE = 0, + MLX5_AN_COMPLETE = 1, + MLX5_AN_FAILED = 2, + MLX5_AN_LINK_UP = 3, + MLX5_AN_LINK_DOWN = 4, +}; + #define MLX5_EEPROM_MAX_BYTES 32 #define MLX5_EEPROM_IDENTIFIER_BYTE_MASK 0x000000ff #define MLX5_I2C_ADDR_LOW 0x50 @@ -65,14 +73,17 @@ int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev, int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev, u8 *proto_oper, int proto_mask, u8 local_port); -int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, - int proto_mask); +int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable, + u32 proto_admin, int proto_mask); void mlx5_toggle_port_link(struct mlx5_core_dev *dev); int mlx5_set_port_admin_status(struct mlx5_core_dev *dev, enum mlx5_port_status status); int mlx5_query_port_admin_status(struct mlx5_core_dev *dev, enum mlx5_port_status *status); int mlx5_set_port_beacon(struct mlx5_core_dev *dev, u16 beacon_duration); +void mlx5_query_port_autoneg(struct mlx5_core_dev *dev, int proto_mask, + u8 *an_status, + u8 *an_disable_cap, u8 *an_disable_admin); int mlx5_set_port_mtu(struct mlx5_core_dev *dev, u16 mtu, u8 port); void mlx5_query_port_max_mtu(struct mlx5_core_dev *dev, u16 *max_mtu, u8 port); -- cgit v0.10.2 From 1b283247e1dfb22f60242a1c7af6d5cbb8f797fe Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Wed, 22 Jun 2016 17:42:21 +0100 Subject: net: tc35815: fix spelling mistake on "descriptors" trivial fixes to spelling mistakes of the word "descriptors" Signed-off-by: Colin Ian King Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c index 5487478..74e6719 100644 --- a/drivers/net/ethernet/toshiba/tc35815.c +++ b/drivers/net/ethernet/toshiba/tc35815.c @@ -280,7 +280,7 @@ struct tc35815_regs { * Descriptors */ -/* Frame descripter */ +/* Frame descriptor */ struct FDesc { volatile __u32 FDNext; volatile __u32 FDSystem; @@ -288,7 +288,7 @@ struct FDesc { volatile __u32 FDCtl; }; -/* Buffer descripter */ +/* Buffer descriptor */ struct BDesc { volatile __u32 BuffData; volatile __u32 BDCtl; @@ -296,7 +296,7 @@ struct BDesc { #define FD_ALIGN 16 -/* Frame Descripter bit assign ---------------------------------------------- */ +/* Frame Descriptor bit assign ---------------------------------------------- */ #define FD_FDLength_MASK 0x0000FFFF /* Length MASK */ #define FD_BDCnt_MASK 0x001F0000 /* BD count MASK in FD */ #define FD_FrmOpt_MASK 0x7C000000 /* Frame option MASK */ @@ -309,7 +309,7 @@ struct BDesc { #define FD_Next_EOL 0x00000001 /* FD EOL indicator */ #define FD_BDCnt_SHIFT 16 -/* Buffer Descripter bit assign --------------------------------------------- */ +/* Buffer Descriptor bit assign --------------------------------------------- */ #define BD_BuffLength_MASK 0x0000FFFF /* Receive Data Size */ #define BD_RxBDID_MASK 0x00FF0000 /* BD ID Number MASK */ #define BD_RxBDSeqN_MASK 0x7F000000 /* Rx BD Sequence Number */ @@ -1387,7 +1387,7 @@ static int tc35815_do_interrupt(struct net_device *dev, u32 status, int limit) if (status & Int_IntExBD) { if (netif_msg_rx_err(lp)) dev_warn(&dev->dev, - "Excessive Buffer Descriptiors (%#x).\n", + "Excessive Buffer Descriptors (%#x).\n", status); dev->stats.rx_length_errors++; ret = 0; -- cgit v0.10.2 From ac5fd4f4b97f5ac9a01303c024d26f34a99633c8 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Wed, 22 Jun 2016 18:35:05 +0100 Subject: ethernet: xircom: fix spelling mistakes on "excessive collisions" trivial fixes to spelling mistakes of the words "excessive collisions" Signed-off-by: Colin Ian King Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/xircom/xirc2ps_cs.c b/drivers/net/ethernet/xircom/xirc2ps_cs.c index 7b44968..ddced28 100644 --- a/drivers/net/ethernet/xircom/xirc2ps_cs.c +++ b/drivers/net/ethernet/xircom/xirc2ps_cs.c @@ -1144,8 +1144,8 @@ xirc2ps_interrupt(int irq, void *dev_id) dev->stats.tx_packets += lp->last_ptr_value - n; netif_wake_queue(dev); } - if (tx_status & 0x0002) { /* Execessive collissions */ - pr_debug("tx restarted due to execssive collissions\n"); + if (tx_status & 0x0002) { /* Excessive collisions */ + pr_debug("tx restarted due to excessive collisions\n"); PutByte(XIRCREG_CR, RestartTx); /* restart transmitter process */ } if (tx_status & 0x0040) -- cgit v0.10.2 From 810bf11033637a2069952afb9c37f3afd3bbfe41 Mon Sep 17 00:00:00 2001 From: Amitoj Kaur Chawla Date: Thu, 23 Jun 2016 10:19:37 +0530 Subject: tipc: Use kmemdup instead of kmalloc and memcpy Replace calls to kmalloc followed by a memcpy with a direct call to kmemdup. The Coccinelle semantic patch used to make this change is as follows: @@ expression from,to,size,flag; statement S; @@ - to = \(kmalloc\|kzalloc\)(size,flag); + to = kmemdup(from,size,flag); if (to==NULL || ...) S - memcpy(to, from, size); Signed-off-by: Amitoj Kaur Chawla Acked-by: Ying Xue Signed-off-by: David S. Miller diff --git a/net/tipc/server.c b/net/tipc/server.c index 272d20a..215849c 100644 --- a/net/tipc/server.c +++ b/net/tipc/server.c @@ -418,13 +418,12 @@ static struct outqueue_entry *tipc_alloc_entry(void *data, int len) if (!entry) return NULL; - buf = kmalloc(len, GFP_ATOMIC); + buf = kmemdup(data, len, GFP_ATOMIC); if (!buf) { kfree(entry); return NULL; } - memcpy(buf, data, len); entry->iov.iov_base = buf; entry->iov.iov_len = len; -- cgit v0.10.2 From 5ae68b0ce134f9cadae2668da82d5f9a77523314 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 23 Jun 2016 14:50:05 +0100 Subject: phy: move fixed_phy MII register generation to a library Move the fixed_phy MII register generation to a library to allow other software phy implementations to use this code. Reviewed-by: Florian Fainelli Signed-off-by: Russell King Signed-off-by: David S. Miller diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig index 8dac88a..f968294 100644 --- a/drivers/net/phy/Kconfig +++ b/drivers/net/phy/Kconfig @@ -12,6 +12,9 @@ menuconfig PHYLIB if PHYLIB +config SWPHY + bool + comment "MII PHY device drivers" config AQUANTIA_PHY @@ -159,6 +162,7 @@ config MICROCHIP_PHY config FIXED_PHY tristate "Driver for MDIO Bus/PHY emulation with fixed speed/link PHYs" depends on PHYLIB + select SWPHY ---help--- Adds the platform "fixed" MDIO Bus to cover the boards that use PHYs that are not connected to the real MDIO bus. diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile index 4170642..7158274 100644 --- a/drivers/net/phy/Makefile +++ b/drivers/net/phy/Makefile @@ -1,6 +1,7 @@ # Makefile for Linux PHY drivers -libphy-objs := phy.o phy_device.o mdio_bus.o mdio_device.o +libphy-y := phy.o phy_device.o mdio_bus.o mdio_device.o +libphy-$(CONFIG_SWPHY) += swphy.o obj-$(CONFIG_PHYLIB) += libphy.o obj-$(CONFIG_AQUANTIA_PHY) += aquantia.o diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index 2d2e433..d98a0d9 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -24,6 +24,8 @@ #include #include +#include "swphy.h" + #define MII_REGS_NUM 29 struct fixed_mdio_bus { @@ -48,101 +50,10 @@ static struct fixed_mdio_bus platform_fmb = { static int fixed_phy_update_regs(struct fixed_phy *fp) { - u16 bmsr = BMSR_ANEGCAPABLE; - u16 bmcr = 0; - u16 lpagb = 0; - u16 lpa = 0; - if (gpio_is_valid(fp->link_gpio)) fp->status.link = !!gpio_get_value_cansleep(fp->link_gpio); - if (fp->status.duplex) { - switch (fp->status.speed) { - case 1000: - bmsr |= BMSR_ESTATEN; - break; - case 100: - bmsr |= BMSR_100FULL; - break; - case 10: - bmsr |= BMSR_10FULL; - break; - default: - break; - } - } else { - switch (fp->status.speed) { - case 1000: - bmsr |= BMSR_ESTATEN; - break; - case 100: - bmsr |= BMSR_100HALF; - break; - case 10: - bmsr |= BMSR_10HALF; - break; - default: - break; - } - } - - if (fp->status.link) { - bmsr |= BMSR_LSTATUS | BMSR_ANEGCOMPLETE; - - if (fp->status.duplex) { - bmcr |= BMCR_FULLDPLX; - - switch (fp->status.speed) { - case 1000: - bmcr |= BMCR_SPEED1000; - lpagb |= LPA_1000FULL; - break; - case 100: - bmcr |= BMCR_SPEED100; - lpa |= LPA_100FULL; - break; - case 10: - lpa |= LPA_10FULL; - break; - default: - pr_warn("fixed phy: unknown speed\n"); - return -EINVAL; - } - } else { - switch (fp->status.speed) { - case 1000: - bmcr |= BMCR_SPEED1000; - lpagb |= LPA_1000HALF; - break; - case 100: - bmcr |= BMCR_SPEED100; - lpa |= LPA_100HALF; - break; - case 10: - lpa |= LPA_10HALF; - break; - default: - pr_warn("fixed phy: unknown speed\n"); - return -EINVAL; - } - } - - if (fp->status.pause) - lpa |= LPA_PAUSE_CAP; - - if (fp->status.asym_pause) - lpa |= LPA_PAUSE_ASYM; - } - - fp->regs[MII_PHYSID1] = 0; - fp->regs[MII_PHYSID2] = 0; - - fp->regs[MII_BMSR] = bmsr; - fp->regs[MII_BMCR] = bmcr; - fp->regs[MII_LPA] = lpa; - fp->regs[MII_STAT1000] = lpagb; - - return 0; + return swphy_update_regs(fp->regs, &fp->status); } static int fixed_mdio_read(struct mii_bus *bus, int phy_addr, int reg_num) diff --git a/drivers/net/phy/swphy.c b/drivers/net/phy/swphy.c new file mode 100644 index 0000000..0551a79 --- /dev/null +++ b/drivers/net/phy/swphy.c @@ -0,0 +1,126 @@ +/* + * Software PHY emulation + * + * Code taken from fixed_phy.c by Russell King + * + * Author: Vitaly Bordug + * Anton Vorontsov + * + * Copyright (c) 2006-2007 MontaVista Software, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ +#include +#include +#include +#include + +#include "swphy.h" + +/** + * swphy_update_regs - update MII register array with fixed phy state + * @regs: array of 32 registers to update + * @state: fixed phy status + * + * Update the array of MII registers with the fixed phy link, speed, + * duplex and pause mode settings. + */ +int swphy_update_regs(u16 *regs, const struct fixed_phy_status *state) +{ + u16 bmsr = BMSR_ANEGCAPABLE; + u16 bmcr = 0; + u16 lpagb = 0; + u16 lpa = 0; + + if (state->duplex) { + switch (state->speed) { + case 1000: + bmsr |= BMSR_ESTATEN; + break; + case 100: + bmsr |= BMSR_100FULL; + break; + case 10: + bmsr |= BMSR_10FULL; + break; + default: + break; + } + } else { + switch (state->speed) { + case 1000: + bmsr |= BMSR_ESTATEN; + break; + case 100: + bmsr |= BMSR_100HALF; + break; + case 10: + bmsr |= BMSR_10HALF; + break; + default: + break; + } + } + + if (state->link) { + bmsr |= BMSR_LSTATUS | BMSR_ANEGCOMPLETE; + + if (state->duplex) { + bmcr |= BMCR_FULLDPLX; + + switch (state->speed) { + case 1000: + bmcr |= BMCR_SPEED1000; + lpagb |= LPA_1000FULL; + break; + case 100: + bmcr |= BMCR_SPEED100; + lpa |= LPA_100FULL; + break; + case 10: + lpa |= LPA_10FULL; + break; + default: + pr_warn("swphy: unknown speed\n"); + return -EINVAL; + } + } else { + switch (state->speed) { + case 1000: + bmcr |= BMCR_SPEED1000; + lpagb |= LPA_1000HALF; + break; + case 100: + bmcr |= BMCR_SPEED100; + lpa |= LPA_100HALF; + break; + case 10: + lpa |= LPA_10HALF; + break; + default: + pr_warn("swphy: unknown speed\n"); + return -EINVAL; + } + } + + if (state->pause) + lpa |= LPA_PAUSE_CAP; + + if (state->asym_pause) + lpa |= LPA_PAUSE_ASYM; + } + + regs[MII_PHYSID1] = 0; + regs[MII_PHYSID2] = 0; + + regs[MII_BMSR] = bmsr; + regs[MII_BMCR] = bmcr; + regs[MII_LPA] = lpa; + regs[MII_STAT1000] = lpagb; + + return 0; +} +EXPORT_SYMBOL_GPL(swphy_update_regs); diff --git a/drivers/net/phy/swphy.h b/drivers/net/phy/swphy.h new file mode 100644 index 0000000..feaa38f --- /dev/null +++ b/drivers/net/phy/swphy.h @@ -0,0 +1,8 @@ +#ifndef SWPHY_H +#define SWPHY_H + +struct fixed_phy_status; + +int swphy_update_regs(u16 *regs, const struct fixed_phy_status *state); + +#endif -- cgit v0.10.2 From 0629bf17eaab9330bef427bdf4a3985c2d8972af Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 23 Jun 2016 14:50:10 +0100 Subject: phy: convert swphy register generation to tabular form Convert the swphy register generation to tabular form which allows us to eliminate multiple switch() statements. This results in a smaller object code size, more efficient, and easier to add support for faster speeds. Before: Idx Name Size VMA LMA File off Algn 0 .text 00000164 00000000 00000000 00000034 2**2 text data bss dec hex filename 388 0 0 388 184 swphy.o After: Idx Name Size VMA LMA File off Algn 0 .text 000000fc 00000000 00000000 00000034 2**2 5 .rodata 00000028 00000000 00000000 00000138 2**2 text data bss dec hex filename 324 0 0 324 144 swphy.o Reviewed-by: Florian Fainelli Signed-off-by: Russell King Signed-off-by: David S. Miller diff --git a/drivers/net/phy/swphy.c b/drivers/net/phy/swphy.c index 0551a79..c88a194 100644 --- a/drivers/net/phy/swphy.c +++ b/drivers/net/phy/swphy.c @@ -20,6 +20,72 @@ #include "swphy.h" +struct swmii_regs { + u16 bmcr; + u16 bmsr; + u16 lpa; + u16 lpagb; +}; + +enum { + SWMII_SPEED_10 = 0, + SWMII_SPEED_100, + SWMII_SPEED_1000, + SWMII_DUPLEX_HALF = 0, + SWMII_DUPLEX_FULL, +}; + +/* + * These two tables get bitwise-anded together to produce the final result. + * This means the speed table must contain both duplex settings, and the + * duplex table must contain all speed settings. + */ +static const struct swmii_regs speed[] = { + [SWMII_SPEED_10] = { + .bmcr = BMCR_FULLDPLX, + .lpa = LPA_10FULL | LPA_10HALF, + }, + [SWMII_SPEED_100] = { + .bmcr = BMCR_FULLDPLX | BMCR_SPEED100, + .bmsr = BMSR_100FULL | BMSR_100HALF, + .lpa = LPA_100FULL | LPA_100HALF, + }, + [SWMII_SPEED_1000] = { + .bmcr = BMCR_FULLDPLX | BMCR_SPEED1000, + .bmsr = BMSR_ESTATEN, + .lpagb = LPA_1000FULL | LPA_1000HALF, + }, +}; + +static const struct swmii_regs duplex[] = { + [SWMII_DUPLEX_HALF] = { + .bmcr = ~BMCR_FULLDPLX, + .bmsr = BMSR_ESTATEN | BMSR_100HALF, + .lpa = LPA_10HALF | LPA_100HALF, + .lpagb = LPA_1000HALF, + }, + [SWMII_DUPLEX_FULL] = { + .bmcr = ~0, + .bmsr = BMSR_ESTATEN | BMSR_100FULL, + .lpa = LPA_10FULL | LPA_100FULL, + .lpagb = LPA_1000FULL, + }, +}; + +static int swphy_decode_speed(int speed) +{ + switch (speed) { + case 1000: + return SWMII_SPEED_1000; + case 100: + return SWMII_SPEED_100; + case 10: + return SWMII_SPEED_10; + default: + return -EINVAL; + } +} + /** * swphy_update_regs - update MII register array with fixed phy state * @regs: array of 32 registers to update @@ -30,81 +96,28 @@ */ int swphy_update_regs(u16 *regs, const struct fixed_phy_status *state) { + int speed_index, duplex_index; u16 bmsr = BMSR_ANEGCAPABLE; u16 bmcr = 0; u16 lpagb = 0; u16 lpa = 0; - if (state->duplex) { - switch (state->speed) { - case 1000: - bmsr |= BMSR_ESTATEN; - break; - case 100: - bmsr |= BMSR_100FULL; - break; - case 10: - bmsr |= BMSR_10FULL; - break; - default: - break; - } - } else { - switch (state->speed) { - case 1000: - bmsr |= BMSR_ESTATEN; - break; - case 100: - bmsr |= BMSR_100HALF; - break; - case 10: - bmsr |= BMSR_10HALF; - break; - default: - break; - } + speed_index = swphy_decode_speed(state->speed); + if (speed_index < 0) { + pr_warn("swphy: unknown speed\n"); + return -EINVAL; } + duplex_index = state->duplex ? SWMII_DUPLEX_FULL : SWMII_DUPLEX_HALF; + + bmsr |= speed[speed_index].bmsr & duplex[duplex_index].bmsr; + if (state->link) { bmsr |= BMSR_LSTATUS | BMSR_ANEGCOMPLETE; - if (state->duplex) { - bmcr |= BMCR_FULLDPLX; - - switch (state->speed) { - case 1000: - bmcr |= BMCR_SPEED1000; - lpagb |= LPA_1000FULL; - break; - case 100: - bmcr |= BMCR_SPEED100; - lpa |= LPA_100FULL; - break; - case 10: - lpa |= LPA_10FULL; - break; - default: - pr_warn("swphy: unknown speed\n"); - return -EINVAL; - } - } else { - switch (state->speed) { - case 1000: - bmcr |= BMCR_SPEED1000; - lpagb |= LPA_1000HALF; - break; - case 100: - bmcr |= BMCR_SPEED100; - lpa |= LPA_100HALF; - break; - case 10: - lpa |= LPA_10HALF; - break; - default: - pr_warn("swphy: unknown speed\n"); - return -EINVAL; - } - } + bmcr |= speed[speed_index].bmcr & duplex[duplex_index].bmcr; + lpa |= speed[speed_index].lpa & duplex[duplex_index].lpa; + lpagb |= speed[speed_index].lpagb & duplex[duplex_index].lpagb; if (state->pause) lpa |= LPA_PAUSE_CAP; -- cgit v0.10.2 From 68888ce075a5e77f0df659b128a7cb7c597a73cb Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 23 Jun 2016 14:50:15 +0100 Subject: phy: separate swphy state validation from register generation Separate out the generation of MII registers from the state validation. This allows us to simplify the error handing in fixed_phy() by allowing earlier error detection. Reviewed-by: Florian Fainelli Signed-off-by: Russell King Signed-off-by: David S. Miller diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index d98a0d9..d84e30c 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -48,12 +48,12 @@ static struct fixed_mdio_bus platform_fmb = { .phys = LIST_HEAD_INIT(platform_fmb.phys), }; -static int fixed_phy_update_regs(struct fixed_phy *fp) +static void fixed_phy_update_regs(struct fixed_phy *fp) { if (gpio_is_valid(fp->link_gpio)) fp->status.link = !!gpio_get_value_cansleep(fp->link_gpio); - return swphy_update_regs(fp->regs, &fp->status); + swphy_update_regs(fp->regs, &fp->status); } static int fixed_mdio_read(struct mii_bus *bus, int phy_addr, int reg_num) @@ -160,6 +160,10 @@ int fixed_phy_add(unsigned int irq, int phy_addr, struct fixed_mdio_bus *fmb = &platform_fmb; struct fixed_phy *fp; + ret = swphy_validate_state(status); + if (ret < 0) + return ret; + fp = kzalloc(sizeof(*fp), GFP_KERNEL); if (!fp) return -ENOMEM; @@ -180,17 +184,12 @@ int fixed_phy_add(unsigned int irq, int phy_addr, goto err_regs; } - ret = fixed_phy_update_regs(fp); - if (ret) - goto err_gpio; + fixed_phy_update_regs(fp); list_add_tail(&fp->node, &fmb->phys); return 0; -err_gpio: - if (gpio_is_valid(fp->link_gpio)) - gpio_free(fp->link_gpio); err_regs: kfree(fp); return ret; diff --git a/drivers/net/phy/swphy.c b/drivers/net/phy/swphy.c index c88a194..21a9bd8 100644 --- a/drivers/net/phy/swphy.c +++ b/drivers/net/phy/swphy.c @@ -87,6 +87,29 @@ static int swphy_decode_speed(int speed) } /** + * swphy_validate_state - validate the software phy status + * @state: software phy status + * + * This checks that we can represent the state stored in @state can be + * represented in the emulated MII registers. Returns 0 if it can, + * otherwise returns -EINVAL. + */ +int swphy_validate_state(const struct fixed_phy_status *state) +{ + int err; + + if (state->link) { + err = swphy_decode_speed(state->speed); + if (err < 0) { + pr_warn("swphy: unknown speed\n"); + return -EINVAL; + } + } + return 0; +} +EXPORT_SYMBOL_GPL(swphy_validate_state); + +/** * swphy_update_regs - update MII register array with fixed phy state * @regs: array of 32 registers to update * @state: fixed phy status @@ -94,7 +117,7 @@ static int swphy_decode_speed(int speed) * Update the array of MII registers with the fixed phy link, speed, * duplex and pause mode settings. */ -int swphy_update_regs(u16 *regs, const struct fixed_phy_status *state) +void swphy_update_regs(u16 *regs, const struct fixed_phy_status *state) { int speed_index, duplex_index; u16 bmsr = BMSR_ANEGCAPABLE; @@ -103,10 +126,8 @@ int swphy_update_regs(u16 *regs, const struct fixed_phy_status *state) u16 lpa = 0; speed_index = swphy_decode_speed(state->speed); - if (speed_index < 0) { - pr_warn("swphy: unknown speed\n"); - return -EINVAL; - } + if (WARN_ON(speed_index < 0)) + return; duplex_index = state->duplex ? SWMII_DUPLEX_FULL : SWMII_DUPLEX_HALF; @@ -133,7 +154,5 @@ int swphy_update_regs(u16 *regs, const struct fixed_phy_status *state) regs[MII_BMCR] = bmcr; regs[MII_LPA] = lpa; regs[MII_STAT1000] = lpagb; - - return 0; } EXPORT_SYMBOL_GPL(swphy_update_regs); diff --git a/drivers/net/phy/swphy.h b/drivers/net/phy/swphy.h index feaa38f..33d2e061 100644 --- a/drivers/net/phy/swphy.h +++ b/drivers/net/phy/swphy.h @@ -3,6 +3,7 @@ struct fixed_phy_status; -int swphy_update_regs(u16 *regs, const struct fixed_phy_status *state); +int swphy_validate_state(const struct fixed_phy_status *state); +void swphy_update_regs(u16 *regs, const struct fixed_phy_status *state); #endif -- cgit v0.10.2 From 37688e3f53c327523caeccdb1ffb3830b4aea9a7 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 23 Jun 2016 14:50:20 +0100 Subject: phy: generate swphy registers on the fly Generate software phy registers as and when requested, rather than duplicating the state in fixed_phy. This allows us to eliminate the duplicate storage of of the same data, which is only different in format. As fixed_phy_update_regs() no longer updates register state, rename it to fixed_phy_update(). Reviewed-by: Florian Fainelli Signed-off-by: Russell King Signed-off-by: David S. Miller diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index d84e30c..0dfed86 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -26,8 +26,6 @@ #include "swphy.h" -#define MII_REGS_NUM 29 - struct fixed_mdio_bus { struct mii_bus *mii_bus; struct list_head phys; @@ -35,7 +33,6 @@ struct fixed_mdio_bus { struct fixed_phy { int addr; - u16 regs[MII_REGS_NUM]; struct phy_device *phydev; struct fixed_phy_status status; int (*link_update)(struct net_device *, struct fixed_phy_status *); @@ -48,12 +45,10 @@ static struct fixed_mdio_bus platform_fmb = { .phys = LIST_HEAD_INIT(platform_fmb.phys), }; -static void fixed_phy_update_regs(struct fixed_phy *fp) +static void fixed_phy_update(struct fixed_phy *fp) { if (gpio_is_valid(fp->link_gpio)) fp->status.link = !!gpio_get_value_cansleep(fp->link_gpio); - - swphy_update_regs(fp->regs, &fp->status); } static int fixed_mdio_read(struct mii_bus *bus, int phy_addr, int reg_num) @@ -61,29 +56,15 @@ static int fixed_mdio_read(struct mii_bus *bus, int phy_addr, int reg_num) struct fixed_mdio_bus *fmb = bus->priv; struct fixed_phy *fp; - if (reg_num >= MII_REGS_NUM) - return -1; - - /* We do not support emulating Clause 45 over Clause 22 register reads - * return an error instead of bogus data. - */ - switch (reg_num) { - case MII_MMD_CTRL: - case MII_MMD_DATA: - return -1; - default: - break; - } - list_for_each_entry(fp, &fmb->phys, node) { if (fp->addr == phy_addr) { /* Issue callback if user registered it. */ if (fp->link_update) { fp->link_update(fp->phydev->attached_dev, &fp->status); - fixed_phy_update_regs(fp); + fixed_phy_update(fp); } - return fp->regs[reg_num]; + return swphy_read_reg(reg_num, &fp->status); } } @@ -143,7 +124,7 @@ int fixed_phy_update_state(struct phy_device *phydev, _UPD(pause); _UPD(asym_pause); #undef _UPD - fixed_phy_update_regs(fp); + fixed_phy_update(fp); return 0; } } @@ -168,8 +149,6 @@ int fixed_phy_add(unsigned int irq, int phy_addr, if (!fp) return -ENOMEM; - memset(fp->regs, 0xFF, sizeof(fp->regs[0]) * MII_REGS_NUM); - if (irq != PHY_POLL) fmb->mii_bus->irq[phy_addr] = irq; @@ -184,7 +163,7 @@ int fixed_phy_add(unsigned int irq, int phy_addr, goto err_regs; } - fixed_phy_update_regs(fp); + fixed_phy_update(fp); list_add_tail(&fp->node, &fmb->phys); diff --git a/drivers/net/phy/swphy.c b/drivers/net/phy/swphy.c index 21a9bd8..34f58f2 100644 --- a/drivers/net/phy/swphy.c +++ b/drivers/net/phy/swphy.c @@ -20,6 +20,8 @@ #include "swphy.h" +#define MII_REGS_NUM 29 + struct swmii_regs { u16 bmcr; u16 bmsr; @@ -110,14 +112,13 @@ int swphy_validate_state(const struct fixed_phy_status *state) EXPORT_SYMBOL_GPL(swphy_validate_state); /** - * swphy_update_regs - update MII register array with fixed phy state - * @regs: array of 32 registers to update + * swphy_read_reg - return a MII register from the fixed phy state + * @reg: MII register * @state: fixed phy status * - * Update the array of MII registers with the fixed phy link, speed, - * duplex and pause mode settings. + * Return the MII @reg register generated from the fixed phy state @state. */ -void swphy_update_regs(u16 *regs, const struct fixed_phy_status *state) +int swphy_read_reg(int reg, const struct fixed_phy_status *state) { int speed_index, duplex_index; u16 bmsr = BMSR_ANEGCAPABLE; @@ -125,9 +126,12 @@ void swphy_update_regs(u16 *regs, const struct fixed_phy_status *state) u16 lpagb = 0; u16 lpa = 0; + if (reg > MII_REGS_NUM) + return -1; + speed_index = swphy_decode_speed(state->speed); if (WARN_ON(speed_index < 0)) - return; + return 0; duplex_index = state->duplex ? SWMII_DUPLEX_FULL : SWMII_DUPLEX_HALF; @@ -147,12 +151,29 @@ void swphy_update_regs(u16 *regs, const struct fixed_phy_status *state) lpa |= LPA_PAUSE_ASYM; } - regs[MII_PHYSID1] = 0; - regs[MII_PHYSID2] = 0; + switch (reg) { + case MII_BMCR: + return bmcr; + case MII_BMSR: + return bmsr; + case MII_PHYSID1: + case MII_PHYSID2: + return 0; + case MII_LPA: + return lpa; + case MII_STAT1000: + return lpagb; + + /* + * We do not support emulating Clause 45 over Clause 22 register + * reads. Return an error instead of bogus data. + */ + case MII_MMD_CTRL: + case MII_MMD_DATA: + return -1; - regs[MII_BMSR] = bmsr; - regs[MII_BMCR] = bmcr; - regs[MII_LPA] = lpa; - regs[MII_STAT1000] = lpagb; + default: + return 0xffff; + } } -EXPORT_SYMBOL_GPL(swphy_update_regs); +EXPORT_SYMBOL_GPL(swphy_read_reg); diff --git a/drivers/net/phy/swphy.h b/drivers/net/phy/swphy.h index 33d2e061..2f09ac3 100644 --- a/drivers/net/phy/swphy.h +++ b/drivers/net/phy/swphy.h @@ -4,6 +4,6 @@ struct fixed_phy_status; int swphy_validate_state(const struct fixed_phy_status *state); -void swphy_update_regs(u16 *regs, const struct fixed_phy_status *state); +int swphy_read_reg(int reg, const struct fixed_phy_status *state); #endif -- cgit v0.10.2 From bf7afb29d545a6875fa44e17ddd23398e3dc30de Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 23 Jun 2016 14:50:25 +0100 Subject: phy: improve safety of fixed-phy MII register reading There is no prevention of a concurrent call to both fixed_mdio_read() and fixed_phy_update_state(), which can result in the state being modified while it's being inspected. Fix this by using a seqcount to detect modifications, and memcpy()ing the state. We remain slightly naughty here, calling link_update() and updating the link status within the read-side loop - which would need rework of the design to change. Reviewed-by: Florian Fainelli Signed-off-by: Russell King Signed-off-by: David S. Miller diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c index 0dfed86..b376ada 100644 --- a/drivers/net/phy/fixed_phy.c +++ b/drivers/net/phy/fixed_phy.c @@ -23,6 +23,7 @@ #include #include #include +#include #include "swphy.h" @@ -34,6 +35,7 @@ struct fixed_mdio_bus { struct fixed_phy { int addr; struct phy_device *phydev; + seqcount_t seqcount; struct fixed_phy_status status; int (*link_update)(struct net_device *, struct fixed_phy_status *); struct list_head node; @@ -58,13 +60,21 @@ static int fixed_mdio_read(struct mii_bus *bus, int phy_addr, int reg_num) list_for_each_entry(fp, &fmb->phys, node) { if (fp->addr == phy_addr) { - /* Issue callback if user registered it. */ - if (fp->link_update) { - fp->link_update(fp->phydev->attached_dev, - &fp->status); - fixed_phy_update(fp); - } - return swphy_read_reg(reg_num, &fp->status); + struct fixed_phy_status state; + int s; + + do { + s = read_seqcount_begin(&fp->seqcount); + /* Issue callback if user registered it. */ + if (fp->link_update) { + fp->link_update(fp->phydev->attached_dev, + &fp->status); + fixed_phy_update(fp); + } + state = fp->status; + } while (read_seqcount_retry(&fp->seqcount, s)); + + return swphy_read_reg(reg_num, &state); } } @@ -116,6 +126,7 @@ int fixed_phy_update_state(struct phy_device *phydev, list_for_each_entry(fp, &fmb->phys, node) { if (fp->addr == phydev->mdio.addr) { + write_seqcount_begin(&fp->seqcount); #define _UPD(x) if (changed->x) \ fp->status.x = status->x _UPD(link); @@ -125,6 +136,7 @@ int fixed_phy_update_state(struct phy_device *phydev, _UPD(asym_pause); #undef _UPD fixed_phy_update(fp); + write_seqcount_end(&fp->seqcount); return 0; } } @@ -149,6 +161,8 @@ int fixed_phy_add(unsigned int irq, int phy_addr, if (!fp) return -ENOMEM; + seqcount_init(&fp->seqcount); + if (irq != PHY_POLL) fmb->mii_bus->irq[phy_addr] = irq; -- cgit v0.10.2 From fb70fabad86d79330c9d70449b09ee88f7a7c22a Mon Sep 17 00:00:00 2001 From: Mitch Williams Date: Mon, 16 May 2016 10:26:31 -0700 Subject: i40e: add functions to control default VSI Add functions to enable and disable default VSI on a VEB. This allows for configuration of limited promiscuous mode specifically for bridging purposes. Change-ID: I0cc5bd68b31c500fdff4d47e1f15d50d2739faf4 Signed-off-by: Mitch Williams Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c index 422b41d..e447dc4 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_common.c +++ b/drivers/net/ethernet/intel/i40e/i40e_common.c @@ -1967,6 +1967,62 @@ aq_add_vsi_exit: } /** + * i40e_aq_set_default_vsi + * @hw: pointer to the hw struct + * @seid: vsi number + * @cmd_details: pointer to command details structure or NULL + **/ +i40e_status i40e_aq_set_default_vsi(struct i40e_hw *hw, + u16 seid, + struct i40e_asq_cmd_details *cmd_details) +{ + struct i40e_aq_desc desc; + struct i40e_aqc_set_vsi_promiscuous_modes *cmd = + (struct i40e_aqc_set_vsi_promiscuous_modes *) + &desc.params.raw; + i40e_status status; + + i40e_fill_default_direct_cmd_desc(&desc, + i40e_aqc_opc_set_vsi_promiscuous_modes); + + cmd->promiscuous_flags = cpu_to_le16(I40E_AQC_SET_VSI_DEFAULT); + cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_DEFAULT); + cmd->seid = cpu_to_le16(seid); + + status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details); + + return status; +} + +/** + * i40e_aq_clear_default_vsi + * @hw: pointer to the hw struct + * @seid: vsi number + * @cmd_details: pointer to command details structure or NULL + **/ +i40e_status i40e_aq_clear_default_vsi(struct i40e_hw *hw, + u16 seid, + struct i40e_asq_cmd_details *cmd_details) +{ + struct i40e_aq_desc desc; + struct i40e_aqc_set_vsi_promiscuous_modes *cmd = + (struct i40e_aqc_set_vsi_promiscuous_modes *) + &desc.params.raw; + i40e_status status; + + i40e_fill_default_direct_cmd_desc(&desc, + i40e_aqc_opc_set_vsi_promiscuous_modes); + + cmd->promiscuous_flags = cpu_to_le16(0); + cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_DEFAULT); + cmd->seid = cpu_to_le16(seid); + + status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details); + + return status; +} + +/** * i40e_aq_set_vsi_unicast_promiscuous * @hw: pointer to the hw struct * @seid: vsi number diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h index 80403c6..4660c5a 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h +++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h @@ -98,6 +98,8 @@ i40e_status i40e_aq_set_phy_debug(struct i40e_hw *hw, u8 cmd_flags, struct i40e_asq_cmd_details *cmd_details); i40e_status i40e_aq_set_default_vsi(struct i40e_hw *hw, u16 vsi_id, struct i40e_asq_cmd_details *cmd_details); +i40e_status i40e_aq_clear_default_vsi(struct i40e_hw *hw, u16 vsi_id, + struct i40e_asq_cmd_details *cmd_details); enum i40e_status_code i40e_aq_get_phy_capabilities(struct i40e_hw *hw, bool qualified_modules, bool report_init, struct i40e_aq_get_phy_abilities_resp *abilities, -- cgit v0.10.2 From 3e25a8f31af1c740c6ba2c7ad74d91830fd630c8 Mon Sep 17 00:00:00 2001 From: Mitch Williams Date: Mon, 16 May 2016 10:26:32 -0700 Subject: i40e: add hw struct local variable This function uses the i40e_hw struct all over the place, so why doesn't it keep a pointer to the struct? Add this pointer as a local variable and use it consistently throughout the function. Change-ID: I10eb688fe40909433fcb8ac7ac891cef67445d72 Signed-off-by: Mitch Williams Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 734cba6..9a26ecc 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -1840,6 +1840,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) { struct list_head tmp_del_list, tmp_add_list; struct i40e_mac_filter *f, *ftmp, *fclone; + struct i40e_hw *hw = &vsi->back->hw; bool promisc_forced_on = false; bool add_happened = false; int filter_list_len = 0; @@ -1920,7 +1921,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) if (!list_empty(&tmp_del_list)) { int del_list_size; - filter_list_len = pf->hw.aq.asq_buf_size / + filter_list_len = hw->aq.asq_buf_size / sizeof(struct i40e_aqc_remove_macvlan_element_data); del_list_size = filter_list_len * sizeof(struct i40e_aqc_remove_macvlan_element_data); @@ -1952,12 +1953,11 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) /* flush a full buffer */ if (num_del == filter_list_len) { - aq_ret = i40e_aq_remove_macvlan(&pf->hw, - vsi->seid, - del_list, - num_del, - NULL); - aq_err = pf->hw.aq.asq_last_status; + aq_ret = + i40e_aq_remove_macvlan(hw, vsi->seid, + del_list, + num_del, NULL); + aq_err = hw->aq.asq_last_status; num_del = 0; memset(del_list, 0, del_list_size); @@ -1965,8 +1965,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) retval = -EIO; dev_err(&pf->pdev->dev, "ignoring delete macvlan error, err %s, aq_err %s while flushing a full buffer\n", - i40e_stat_str(&pf->hw, aq_ret), - i40e_aq_str(&pf->hw, aq_err)); + + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, aq_err)); } } /* Release memory for MAC filter entries which were @@ -1977,17 +1978,16 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) } if (num_del) { - aq_ret = i40e_aq_remove_macvlan(&pf->hw, vsi->seid, - del_list, num_del, - NULL); - aq_err = pf->hw.aq.asq_last_status; + aq_ret = i40e_aq_remove_macvlan(hw, vsi->seid, del_list, + num_del, NULL); + aq_err = hw->aq.asq_last_status; num_del = 0; if (aq_ret && aq_err != I40E_AQ_RC_ENOENT) dev_info(&pf->pdev->dev, "ignoring delete macvlan error, err %s aq_err %s\n", - i40e_stat_str(&pf->hw, aq_ret), - i40e_aq_str(&pf->hw, aq_err)); + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, aq_err)); } kfree(del_list); @@ -1998,7 +1998,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) int add_list_size; /* do all the adds now */ - filter_list_len = pf->hw.aq.asq_buf_size / + filter_list_len = hw->aq.asq_buf_size / sizeof(struct i40e_aqc_add_macvlan_element_data), add_list_size = filter_list_len * sizeof(struct i40e_aqc_add_macvlan_element_data); @@ -2033,10 +2033,10 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) /* flush a full buffer */ if (num_add == filter_list_len) { - aq_ret = i40e_aq_add_macvlan(&pf->hw, vsi->seid, + aq_ret = i40e_aq_add_macvlan(hw, vsi->seid, add_list, num_add, NULL); - aq_err = pf->hw.aq.asq_last_status; + aq_err = hw->aq.asq_last_status; num_add = 0; if (aq_ret) @@ -2051,9 +2051,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) } if (num_add) { - aq_ret = i40e_aq_add_macvlan(&pf->hw, vsi->seid, + aq_ret = i40e_aq_add_macvlan(hw, vsi->seid, add_list, num_add, NULL); - aq_err = pf->hw.aq.asq_last_status; + aq_err = hw->aq.asq_last_status; num_add = 0; } kfree(add_list); @@ -2063,9 +2063,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) retval = i40e_aq_rc_to_posix(aq_ret, aq_err); dev_info(&pf->pdev->dev, "add filter failed, err %s aq_err %s\n", - i40e_stat_str(&pf->hw, aq_ret), - i40e_aq_str(&pf->hw, aq_err)); - if ((pf->hw.aq.asq_last_status == I40E_AQ_RC_ENOSPC) && + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, aq_err)); + if ((hw->aq.asq_last_status == I40E_AQ_RC_ENOSPC) && !test_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state)) { promisc_forced_on = true; @@ -2093,12 +2093,11 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) NULL); if (aq_ret) { retval = i40e_aq_rc_to_posix(aq_ret, - pf->hw.aq.asq_last_status); + hw->aq.asq_last_status); dev_info(&pf->pdev->dev, "set multi promisc failed, err %s aq_err %s\n", - i40e_stat_str(&pf->hw, aq_ret), - i40e_aq_str(&pf->hw, - pf->hw.aq.asq_last_status)); + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, hw->aq.asq_last_status)); } } if ((changed_flags & IFF_PROMISC) || promisc_forced_on) { @@ -2121,29 +2120,33 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) } } else { aq_ret = i40e_aq_set_vsi_unicast_promiscuous( - &vsi->back->hw, + hw, vsi->seid, cur_promisc, NULL, true); if (aq_ret) { retval = i40e_aq_rc_to_posix(aq_ret, - pf->hw.aq.asq_last_status); + hw->aq.asq_last_status); dev_info(&pf->pdev->dev, - "set unicast promisc failed, err %d, aq_err %d\n", - aq_ret, pf->hw.aq.asq_last_status); + "set unicast promisc failed, err %s, aq_err %s\n", + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, + hw->aq.asq_last_status)); } aq_ret = i40e_aq_set_vsi_multicast_promiscuous( - &vsi->back->hw, + hw, vsi->seid, cur_promisc, NULL); if (aq_ret) { retval = i40e_aq_rc_to_posix(aq_ret, - pf->hw.aq.asq_last_status); + hw->aq.asq_last_status); dev_info(&pf->pdev->dev, - "set multicast promisc failed, err %d, aq_err %d\n", - aq_ret, pf->hw.aq.asq_last_status); + "set multicast promisc failed, err %s, aq_err %s\n", + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, + hw->aq.asq_last_status)); } } aq_ret = i40e_aq_set_vsi_broadcast(&vsi->back->hw, @@ -2154,9 +2157,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) pf->hw.aq.asq_last_status); dev_info(&pf->pdev->dev, "set brdcast promisc failed, err %s, aq_err %s\n", - i40e_stat_str(&pf->hw, aq_ret), - i40e_aq_str(&pf->hw, - pf->hw.aq.asq_last_status)); + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, + hw->aq.asq_last_status)); } } out: -- cgit v0.10.2 From bb36071721699c531e19ea7c3e7eebd605e8b61d Mon Sep 17 00:00:00 2001 From: Mitch Williams Date: Mon, 16 May 2016 10:26:33 -0700 Subject: i40e: write HENA for VFs Now that VF RSS is configured by the PF driver, it needs to set the RSS Hash Enable registers by default. Without this, no packets will be hashed and they'll all end up on queue 0. Change-ID: I38e425f40ddb81e3b19a951cfbb939fa5b1123f1 Signed-off-by: Mitch Williams Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 1fcafcf..6fcbf76 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -665,6 +665,8 @@ static int i40e_alloc_vsi_res(struct i40e_vf *vf, enum i40e_vsi_type type) goto error_alloc_vsi_res; } if (type == I40E_VSI_SRIOV) { + u64 hena = i40e_pf_get_default_rss_hena(pf); + vf->lan_vsi_idx = vsi->idx; vf->lan_vsi_id = vsi->id; /* If the port VLAN has been configured and then the @@ -687,6 +689,10 @@ static int i40e_alloc_vsi_res(struct i40e_vf *vf, enum i40e_vsi_type type) vf->default_lan_addr.addr, vf->vf_id); } spin_unlock_bh(&vsi->mac_filter_list_lock); + i40e_write_rx_ctl(&pf->hw, I40E_VFQF_HENA1(0, vf->vf_id), + (u32)hena); + i40e_write_rx_ctl(&pf->hw, I40E_VFQF_HENA1(1, vf->vf_id), + (u32)(hena >> 32)); } /* program mac filter */ -- cgit v0.10.2 From f980d445e5ae71c64e1a8e07e9cbf7bb285d1106 Mon Sep 17 00:00:00 2001 From: Catherine Sullivan Date: Mon, 16 May 2016 10:26:34 -0700 Subject: i40e: Add a call to set the client interface down We were failing to set the client interface down when we put the VSI down. Add this call so that the client doesn't get an open called with no close. Also remove an un-needed delay. The VF should not be affected at all by i40e_down. Change-ID: I1135dffef534bf84e6fed57cf51bcf590e6cfaf7 Signed-off-by: Catherine Sullivan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 9a26ecc..a2b4012 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -5181,12 +5181,6 @@ static void i40e_vsi_reinit_locked(struct i40e_vsi *vsi) usleep_range(1000, 2000); i40e_down(vsi); - /* Give a VF some time to respond to the reset. The - * two second wait is based upon the watchdog cycle in - * the VF driver. - */ - if (vsi->type == I40E_VSI_SRIOV) - msleep(2000); i40e_up(vsi); clear_bit(__I40E_CONFIG_BUSY, &pf->state); } @@ -5229,6 +5223,9 @@ void i40e_down(struct i40e_vsi *vsi) i40e_clean_tx_ring(vsi->tx_rings[i]); i40e_clean_rx_ring(vsi->rx_rings[i]); } + + i40e_notify_client_of_netdev_close(vsi, false); + } /** @@ -5931,7 +5928,6 @@ static void i40e_fdir_flush_and_replay(struct i40e_pf *pf) if (I40E_DEBUG_FD & pf->hw.debug_mask) dev_info(&pf->pdev->dev, "FD Filter table flushed and FD-SB replayed.\n"); } - } /** -- cgit v0.10.2 From 0e8d95f8965dc2e6f22e5b321b73de7b0396dc4a Mon Sep 17 00:00:00 2001 From: Mitch Williams Date: Mon, 16 May 2016 10:26:36 -0700 Subject: i40evf: don't overflow buffer If the user adds an obscene amount of MAC addresses, the driver will run into the situation where it has too many address requests to fit into a single PF message. The driver checks for this case, and calculates the maximum number of messages that it can send. Then it completely ignores this count and overflows the buffer. Fix this by checking the address count and bailing out of the loop at the appropriate time. Change-ID: If8dcbb04602c75941dc0cd8309065e1de9ca791c Signed-off-by: Mitch Williams Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c index f134456..d76c221 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c @@ -434,6 +434,8 @@ void i40evf_add_ether_addrs(struct i40evf_adapter *adapter) ether_addr_copy(veal->list[i].addr, f->macaddr); i++; f->add = false; + if (i == count) + break; } } if (!more) @@ -497,6 +499,8 @@ void i40evf_del_ether_addrs(struct i40evf_adapter *adapter) i++; list_del(&f->list); kfree(f); + if (i == count) + break; } } if (!more) @@ -560,6 +564,8 @@ void i40evf_add_vlans(struct i40evf_adapter *adapter) vvfl->vlan_id[i] = f->vlan; i++; f->add = false; + if (i == count) + break; } } if (!more) @@ -623,6 +629,8 @@ void i40evf_del_vlans(struct i40evf_adapter *adapter) i++; list_del(&f->list); kfree(f); + if (i == count) + break; } } if (!more) -- cgit v0.10.2 From b33d3b7321bb5ae291851fce31688d2d3c432e6b Mon Sep 17 00:00:00 2001 From: Greg Rose Date: Mon, 16 May 2016 10:26:37 -0700 Subject: i40e: Clean up MSIX IRQs before suspend The i40e_suspend() function calls another function that preps the device for the power save and resume by freeing all the Tx/Rx resources and interrupts but that function does not free the "other" causes interrupt vector and IRQ. It also fails to call synchronize_irq() before freeing the IRQ vectors. This sometimes may result in some AER errors on those systems with that PCIe error reporting feature enabled. Call synchronize_irq() before freeing IRQ vectors and explicitly free the other causes interrupt resources and shut down that MSIX interrupt. Change-ID: Ib88e4536756518a352446da0232189716618ad81 Signed-off-by: Greg Rose Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index a2b4012..e071c22 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -3950,6 +3950,7 @@ static void i40e_vsi_free_irq(struct i40e_vsi *vsi) /* clear the affinity_mask in the IRQ descriptor */ irq_set_affinity_hint(pf->msix_entries[vector].vector, NULL); + synchronize_irq(pf->msix_entries[vector].vector); free_irq(pf->msix_entries[vector].vector, vsi->q_vectors[i]); @@ -11451,6 +11452,8 @@ static int i40e_suspend(struct pci_dev *pdev, pm_message_t state) wr32(hw, I40E_PFPM_APM, (pf->wol_en ? I40E_PFPM_APM_APME_MASK : 0)); wr32(hw, I40E_PFPM_WUFC, (pf->wol_en ? I40E_PFPM_WUFC_MAG_MASK : 0)); + i40e_stop_misc_vector(pf); + pci_wake_from_d3(pdev, pf->wol_en); pci_set_power_state(pdev, PCI_D3hot); -- cgit v0.10.2 From 059ff69b5fa136a23a2d71df6d9814e86485e8b9 Mon Sep 17 00:00:00 2001 From: Greg Rose Date: Mon, 16 May 2016 10:26:38 -0700 Subject: i40e: Save PCI state before suspend The i40e_suspend() function was failing to save PCI state and this would result in a kernel stack trace from a WARN_ONCE in the pci_legacy_suspend() function. Add a call to pci_save_state() to fix that problem. Change-ID: I4736e62bb660966bd208cc8af617a14cb07fc4bd Signed-off-by: Greg Rose Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index e071c22..52d9d28 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -11441,6 +11441,7 @@ static int i40e_suspend(struct pci_dev *pdev, pm_message_t state) { struct i40e_pf *pf = pci_get_drvdata(pdev); struct i40e_hw *hw = &pf->hw; + int retval = 0; set_bit(__I40E_SUSPENDED, &pf->state); set_bit(__I40E_DOWN, &pf->state); @@ -11454,10 +11455,14 @@ static int i40e_suspend(struct pci_dev *pdev, pm_message_t state) i40e_stop_misc_vector(pf); + retval = pci_save_state(pdev); + if (retval) + return retval; + pci_wake_from_d3(pdev, pf->wol_en); pci_set_power_state(pdev, PCI_D3hot); - return 0; + return retval; } /** -- cgit v0.10.2 From 6536227d1dd60dcd4a4a4a32825842c1456fa78c Mon Sep 17 00:00:00 2001 From: Serey Kong Date: Mon, 16 May 2016 10:26:39 -0700 Subject: i40e: fix missing DA cable check When a Direct Attach (DA) cable is used, if the i40e_set_settings function is called it would return an error. Add the DA type so the function won't fail. Change-ID: I2b802f27a5d91cfefa72fd1f852acb4d74647a8e Signed-off-by: Serey Kong Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 5e8d84f..8381dab 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -663,6 +663,7 @@ static int i40e_set_settings(struct net_device *netdev, if (hw->phy.media_type != I40E_MEDIA_TYPE_BASET && hw->phy.media_type != I40E_MEDIA_TYPE_FIBER && hw->phy.media_type != I40E_MEDIA_TYPE_BACKPLANE && + hw->phy.media_type != I40E_MEDIA_TYPE_DA && hw->phy.link_info.link_info & I40E_AQ_LINK_UP) return -EOPNOTSUPP; -- cgit v0.10.2 From 01a7a9fef44e5b2f994174c2c1fda80178214d8d Mon Sep 17 00:00:00 2001 From: Avinash Dayanand Date: Mon, 16 May 2016 10:26:40 -0700 Subject: i40e: Removing unnecessary code which caused supported link mode bug Removing this code which wasn't allowing 100BaseT to show up in the supported link modes for 10GBaseT PHYs. Change-ID: Iada2eafa7ef6b4bac9a2a1380ff533ae5de51e1d Signed-off-by: Avinash Dayanand Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 8381dab..4962e85 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -313,8 +313,7 @@ static void i40e_phy_type_to_ethtool(struct i40e_pf *pf, u32 *supported, *advertising |= ADVERTISED_Autoneg | ADVERTISED_40000baseCR4_Full; } - if ((phy_types & I40E_CAP_PHY_TYPE_100BASE_TX) && - !(phy_types & I40E_CAP_PHY_TYPE_1000BASE_T)) { + if (phy_types & I40E_CAP_PHY_TYPE_100BASE_TX) { *supported |= SUPPORTED_Autoneg | SUPPORTED_100baseT_Full; *advertising |= ADVERTISED_Autoneg | -- cgit v0.10.2 From 7d64402f5ae5acb8258860ace394a21b4584fe8f Mon Sep 17 00:00:00 2001 From: Catherine Sullivan Date: Mon, 16 May 2016 10:26:41 -0700 Subject: i40e: Fix RSS to not be limited by the number of CPUs Limiting qcount to pf->num_lan_msix, effectively limits the RSS queues to only use the number of CPUs, and ignore all other queues. We don't want to do this. If the user has changed the RSS settings to use more queues then CPUS, we want to trust they know what they are doing and let them. More importantly, if we tell them that is what we did, we want to actually do it and allow traffic into all of the queues we have allocated. This does not change the default setting to initially allocate only the number of CPUS of queue pairs. Change-ID: Ie941a96e806e4bcd016addb4e17affb46770ada5 Signed-off-by: Catherine Sullivan Tested-by: Andrew Bowers diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 52d9d28..a32be7c 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -1579,14 +1579,8 @@ static void i40e_vsi_setup_queue_map(struct i40e_vsi *vsi, vsi->tc_config.numtc = numtc; vsi->tc_config.enabled_tc = enabled_tc ? enabled_tc : 1; /* Number of queues per enabled TC */ - /* In MFP case we can have a much lower count of MSIx - * vectors available and so we need to lower the used - * q count. - */ - if (pf->flags & I40E_FLAG_MSIX_ENABLED) - qcount = min_t(int, vsi->alloc_queue_pairs, pf->num_lan_msix); - else - qcount = vsi->alloc_queue_pairs; + qcount = vsi->alloc_queue_pairs; + num_tc_qps = qcount / numtc; num_tc_qps = min_t(int, num_tc_qps, i40e_pf_get_max_q_per_tc(pf)); -- cgit v0.10.2 From 63590b6129aa4a991c4d162cad5caab1632a1b9a Mon Sep 17 00:00:00 2001 From: Mitch Williams Date: Mon, 16 May 2016 10:26:42 -0700 Subject: i40evf: always activate correct MAC address filter Always add MAC address at the tail of the MAC filter list. Since the device's "real" MAC address is added first, it will always be at the beginning of the list. This prevents an issue where the "real" MAC filter might not get added if too many other filters are added before bringing the interface up. Change-ID: I34a8aeebeb0cb87a44b24118adc4176c7b943c1c Signed-off-by: Mitch Williams Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c index 16c5529..da9221b 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c @@ -825,7 +825,7 @@ i40evf_mac_filter *i40evf_add_filter(struct i40evf_adapter *adapter, ether_addr_copy(f->macaddr, macaddr); - list_add(&f->list, &adapter->mac_filter_list); + list_add_tail(&f->list, &adapter->mac_filter_list); f->add = true; adapter->aq_required |= I40EVF_FLAG_AQ_ADD_MAC_FILTER; } -- cgit v0.10.2 From 5bc160319f8a1e7ea23d7136e725f9e6a4a7628a Mon Sep 17 00:00:00 2001 From: Mitch Williams Date: Mon, 16 May 2016 10:26:43 -0700 Subject: i40e: set default VSI without a reset Remove the need for a reset when the device enters limited promiscuous mode. This was causing heartburn for people who were using VFs and bridging, since this would require all of the VFs to undergo a reset each time the PF changed its promiscuity. Change-ID: I0a83495c5e4d68112bbc7a7a076d20fa8dd3b61c Signed-off-by: Mitch Williams Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index a32be7c..d6a5106 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -2110,7 +2110,25 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) */ if (pf->cur_promisc != cur_promisc) { pf->cur_promisc = cur_promisc; - set_bit(__I40E_PF_RESET_REQUESTED, &pf->state); + if (cur_promisc) + aq_ret = + i40e_aq_set_default_vsi(hw, + vsi->seid, + NULL); + else + aq_ret = + i40e_aq_clear_default_vsi(hw, + vsi->seid, + NULL); + if (aq_ret) { + retval = i40e_aq_rc_to_posix(aq_ret, + hw->aq.asq_last_status); + dev_info(&pf->pdev->dev, + "Set default VSI failed, err %s, aq_err %s\n", + i40e_stat_str(hw, aq_ret), + i40e_aq_str(hw, + hw->aq.asq_last_status)); + } } } else { aq_ret = i40e_aq_set_vsi_unicast_promiscuous( @@ -10047,14 +10065,14 @@ void i40e_veb_release(struct i40e_veb *veb) static int i40e_add_veb(struct i40e_veb *veb, struct i40e_vsi *vsi) { struct i40e_pf *pf = veb->pf; - bool is_default = veb->pf->cur_promisc; bool enable_stats = !!(pf->flags & I40E_FLAG_VEB_STATS_ENABLED); int ret; - /* get a VEB from the hardware */ ret = i40e_aq_add_veb(&pf->hw, veb->uplink_seid, vsi->seid, - veb->enabled_tc, is_default, + veb->enabled_tc, false, &veb->seid, enable_stats, NULL); + + /* get a VEB from the hardware */ if (ret) { dev_info(&pf->pdev->dev, "couldn't add VEB, err %s aq_err %s\n", -- cgit v0.10.2 From 2d1de8283f371467c58e16ed0b27372e369f2568 Mon Sep 17 00:00:00 2001 From: Shannon Nelson Date: Mon, 16 May 2016 10:26:44 -0700 Subject: i40e: add VSI info to macaddr messages Since the macaddr add and delete happens asynchronously, error messages don't easily get associated to the actual request. Here we add a bit of information to the error messages to help determine the source of the error. Change-ID: Id2d6df5287141c3579677d72d8bd21122823d79f Signed-off-by: Shannon Nelson Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index d6a5106..880602f 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -1837,6 +1837,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) struct i40e_hw *hw = &vsi->back->hw; bool promisc_forced_on = false; bool add_happened = false; + char vsi_name[16] = "PF"; int filter_list_len = 0; u32 changed_flags = 0; i40e_status aq_ret = 0; @@ -1864,6 +1865,11 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) INIT_LIST_HEAD(&tmp_del_list); INIT_LIST_HEAD(&tmp_add_list); + if (vsi->type == I40E_VSI_SRIOV) + snprintf(vsi_name, sizeof(vsi_name) - 1, "VF %d", vsi->vf_id); + else if (vsi->type != I40E_VSI_MAIN) + snprintf(vsi_name, sizeof(vsi_name) - 1, "vsi %d", vsi->seid); + if (vsi->flags & I40E_VSI_FLAG_FILTER_CHANGED) { vsi->flags &= ~I40E_VSI_FLAG_FILTER_CHANGED; @@ -1958,8 +1964,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) if (aq_ret && aq_err != I40E_AQ_RC_ENOENT) { retval = -EIO; dev_err(&pf->pdev->dev, - "ignoring delete macvlan error, err %s, aq_err %s while flushing a full buffer\n", - + "ignoring delete macvlan error on %s, err %s, aq_err %s while flushing a full buffer\n", + vsi_name, i40e_stat_str(hw, aq_ret), i40e_aq_str(hw, aq_err)); } @@ -1979,7 +1985,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) if (aq_ret && aq_err != I40E_AQ_RC_ENOENT) dev_info(&pf->pdev->dev, - "ignoring delete macvlan error, err %s aq_err %s\n", + "ignoring delete macvlan error on %s, err %s aq_err %s\n", + vsi_name, i40e_stat_str(hw, aq_ret), i40e_aq_str(hw, aq_err)); } @@ -2056,7 +2063,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) if (add_happened && aq_ret && aq_err != I40E_AQ_RC_EINVAL) { retval = i40e_aq_rc_to_posix(aq_ret, aq_err); dev_info(&pf->pdev->dev, - "add filter failed, err %s aq_err %s\n", + "add filter failed on %s, err %s aq_err %s\n", + vsi_name, i40e_stat_str(hw, aq_ret), i40e_aq_str(hw, aq_err)); if ((hw->aq.asq_last_status == I40E_AQ_RC_ENOSPC) && @@ -2065,7 +2073,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) promisc_forced_on = true; set_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state); - dev_info(&pf->pdev->dev, "promiscuous mode forced on\n"); + dev_info(&pf->pdev->dev, "promiscuous mode forced on %s\n", + vsi_name); } } } @@ -2089,7 +2098,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) retval = i40e_aq_rc_to_posix(aq_ret, hw->aq.asq_last_status); dev_info(&pf->pdev->dev, - "set multi promisc failed, err %s aq_err %s\n", + "set multi promisc failed on %s, err %s aq_err %s\n", + vsi_name, i40e_stat_str(hw, aq_ret), i40e_aq_str(hw, hw->aq.asq_last_status)); } @@ -2124,7 +2134,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) retval = i40e_aq_rc_to_posix(aq_ret, hw->aq.asq_last_status); dev_info(&pf->pdev->dev, - "Set default VSI failed, err %s, aq_err %s\n", + "Set default VSI failed on %s, err %s, aq_err %s\n", + vsi_name, i40e_stat_str(hw, aq_ret), i40e_aq_str(hw, hw->aq.asq_last_status)); @@ -2141,7 +2152,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) i40e_aq_rc_to_posix(aq_ret, hw->aq.asq_last_status); dev_info(&pf->pdev->dev, - "set unicast promisc failed, err %s, aq_err %s\n", + "set unicast promisc failed on %s, err %s, aq_err %s\n", + vsi_name, i40e_stat_str(hw, aq_ret), i40e_aq_str(hw, hw->aq.asq_last_status)); @@ -2155,7 +2167,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) i40e_aq_rc_to_posix(aq_ret, hw->aq.asq_last_status); dev_info(&pf->pdev->dev, - "set multicast promisc failed, err %s, aq_err %s\n", + "set multicast promisc failed on %s, err %s, aq_err %s\n", + vsi_name, i40e_stat_str(hw, aq_ret), i40e_aq_str(hw, hw->aq.asq_last_status)); -- cgit v0.10.2 From 070619580217277dc081a86299974848dd16f451 Mon Sep 17 00:00:00 2001 From: Bimmy Pujari Date: Mon, 16 May 2016 10:26:45 -0700 Subject: i40e/i40evf: Bump version from 1.5.16 to 1.6.4 Signed-off-by: Bimmy Pujari Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 880602f..3e394bf 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -40,8 +40,8 @@ static const char i40e_driver_string[] = #define DRV_KERN "-k" #define DRV_VERSION_MAJOR 1 -#define DRV_VERSION_MINOR 5 -#define DRV_VERSION_BUILD 16 +#define DRV_VERSION_MINOR 6 +#define DRV_VERSION_BUILD 4 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ __stringify(DRV_VERSION_MINOR) "." \ __stringify(DRV_VERSION_BUILD) DRV_KERN diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c index da9221b..eac057b 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c @@ -37,8 +37,8 @@ static const char i40evf_driver_string[] = #define DRV_KERN "-k" #define DRV_VERSION_MAJOR 1 -#define DRV_VERSION_MINOR 5 -#define DRV_VERSION_BUILD 10 +#define DRV_VERSION_MINOR 6 +#define DRV_VERSION_BUILD 4 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ __stringify(DRV_VERSION_MINOR) "." \ __stringify(DRV_VERSION_BUILD) \ -- cgit v0.10.2 From a70e407f6d0b0f63b17d468f78b666d33f264ba1 Mon Sep 17 00:00:00 2001 From: Tushar Dave Date: Mon, 16 May 2016 12:40:53 -0700 Subject: i40e: Fix errors resulted while turning off TSO On systems with 128 CPUs, turning off TSO results in errors, i40e 0000:03:00.0: failed to get tracking for 1 vectors for VSI 400, err=-12 i40e 0000:03:00.0: Couldn't create FDir VSI i40e 0000:03:00.0: i40e_ptp_init: PTP not supported on eth0 i40e 0000:03:00.0: couldn't add VEB, err I40E_ERR_ADMIN_QUEUE_ERROR aq_err I40E_AQ_RC_ENOENT i40e 0000:03:00.0: rebuild of switch failed: -1, will try to set up simple PF connection i40e 0000:03:00.0 eth0: adding 00:10:e0:8a:24:b6 vid=0 Enabling FD_SB without checking availability of MSI-X vector is the root cause. This change adds necessary check. Signed-off-by: Tushar Dave Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h index 9c44739..e83fc8a 100644 --- a/drivers/net/ethernet/intel/i40e/i40e.h +++ b/drivers/net/ethernet/intel/i40e/i40e.h @@ -283,6 +283,7 @@ struct i40e_pf { #endif /* I40E_FCOE */ u16 num_lan_qps; /* num lan queues this PF has set up */ u16 num_lan_msix; /* num queue vectors for the base PF vsi */ + u16 num_fdsb_msix; /* num queue vectors for sideband Fdir */ u16 num_iwarp_msix; /* num of iwarp vectors for this PF */ int iwarp_base_vector; int queues_left; /* queues left unclaimed */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 3e394bf..a313194 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -7185,7 +7185,7 @@ static int i40e_set_num_rings_in_vsi(struct i40e_vsi *vsi) vsi->alloc_queue_pairs = 1; vsi->num_desc = ALIGN(I40E_FDIR_RING_COUNT, I40E_REQ_DESCRIPTOR_MULTIPLE); - vsi->num_q_vectors = 1; + vsi->num_q_vectors = pf->num_fdsb_msix; break; case I40E_VSI_VMDQ2: @@ -7569,9 +7569,11 @@ static int i40e_init_msix(struct i40e_pf *pf) /* reserve one vector for sideband flow director */ if (pf->flags & I40E_FLAG_FD_SB_ENABLED) { if (vectors_left) { + pf->num_fdsb_msix = 1; v_budget++; vectors_left--; } else { + pf->num_fdsb_msix = 0; pf->flags &= ~I40E_FLAG_FD_SB_ENABLED; } } @@ -8590,7 +8592,9 @@ bool i40e_set_ntuple(struct i40e_pf *pf, netdev_features_t features) /* Enable filters and mark for reset */ if (!(pf->flags & I40E_FLAG_FD_SB_ENABLED)) need_reset = true; - pf->flags |= I40E_FLAG_FD_SB_ENABLED; + /* enable FD_SB only if there is MSI-X vector */ + if (pf->num_fdsb_msix > 0) + pf->flags |= I40E_FLAG_FD_SB_ENABLED; } else { /* turn off filters, mark for reset and clear SW filter list */ if (pf->flags & I40E_FLAG_FD_SB_ENABLED) { -- cgit v0.10.2 From 85a1aab79c54c7e44cb0f98e5aa797fbb0457866 Mon Sep 17 00:00:00 2001 From: Neerav Parikh Date: Tue, 7 Jun 2016 09:14:55 -0700 Subject: i40e: Don't notify client(s) for DCB changes on all VSIs When LLDP/DCBX change happens the i40e driver code flow tried to notify the client(s) for each of the PF VSIs. This resulted into kernel panic on the first VSI that didn't have any netdev associated to it. The DCB change notification to the client(s) should be done only once for the PF/LAN VSI where the client(s) instances have been added to. Also, move the notification call after the PF driver has made changes related to the updated DCB configuration. Signed-off-by: Neerav Parikh Signed-off-by: Usha Ketineni Tested-by: Ronald J Bynoe Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index a313194..2b11405 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -4982,7 +4982,6 @@ static void i40e_dcb_reconfigure(struct i40e_pf *pf) if (pf->vsi[v]->netdev) i40e_dcbnl_set_all(pf->vsi[v]); } - i40e_notify_client_of_l2_param_changes(pf->vsi[v]); } } @@ -5730,6 +5729,8 @@ static int i40e_handle_lldp_event(struct i40e_pf *pf, i40e_service_event_schedule(pf); } else { i40e_pf_unquiesce_all_vsi(pf); + /* Notify the client for the DCB changes */ + i40e_notify_client_of_l2_param_changes(pf->vsi[pf->lan_vsi]); } exit: -- cgit v0.10.2 From 958974fdaf7fcd26fcfcf409e3e61fdc3ab4020c Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 23 Jun 2016 23:48:58 +0200 Subject: net: ethernet: dnet: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c index b69a9ea..9c6955f 100644 --- a/drivers/net/ethernet/dnet.c +++ b/drivers/net/ethernet/dnet.c @@ -173,7 +173,7 @@ static int dnet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, static void dnet_handle_link_change(struct net_device *dev) { struct dnet *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; unsigned long flags; u32 mode_reg, ctl_reg; @@ -295,7 +295,6 @@ static int dnet_mii_probe(struct net_device *dev) bp->link = 0; bp->speed = 0; bp->duplex = -1; - bp->phy_dev = phydev; return 0; } @@ -629,16 +628,16 @@ static int dnet_open(struct net_device *dev) struct dnet *bp = netdev_priv(dev); /* if the phy is not yet register, retry later */ - if (!bp->phy_dev) + if (!dev->phydev) return -EAGAIN; napi_enable(&bp->napi); dnet_init_hw(bp); - phy_start_aneg(bp->phy_dev); + phy_start_aneg(dev->phydev); /* schedule a link state check */ - phy_start(bp->phy_dev); + phy_start(dev->phydev); netif_start_queue(dev); @@ -652,8 +651,8 @@ static int dnet_close(struct net_device *dev) netif_stop_queue(dev); napi_disable(&bp->napi); - if (bp->phy_dev) - phy_stop(bp->phy_dev); + if (dev->phydev) + phy_stop(dev->phydev); dnet_reset_hw(bp); netif_carrier_off(dev); @@ -733,8 +732,7 @@ static struct net_device_stats *dnet_get_stats(struct net_device *dev) static int dnet_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct dnet *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; if (!phydev) return -ENODEV; @@ -744,8 +742,7 @@ static int dnet_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) static int dnet_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct dnet *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; if (!phydev) return -ENODEV; @@ -755,8 +752,7 @@ static int dnet_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) static int dnet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct dnet *bp = netdev_priv(dev); - struct phy_device *phydev = bp->phy_dev; + struct phy_device *phydev = dev->phydev; if (!netif_running(dev)) return -EINVAL; @@ -875,7 +871,7 @@ static int dnet_probe(struct platform_device *pdev) (bp->capabilities & DNET_HAS_IRQ) ? "" : "no ", (bp->capabilities & DNET_HAS_GIGABIT) ? "" : "no ", (bp->capabilities & DNET_HAS_DMA) ? "" : "no "); - phydev = bp->phy_dev; + phydev = dev->phydev; phy_attached_info(phydev); return 0; @@ -899,8 +895,8 @@ static int dnet_remove(struct platform_device *pdev) if (dev) { bp = netdev_priv(dev); - if (bp->phy_dev) - phy_disconnect(bp->phy_dev); + if (dev->phydev) + phy_disconnect(dev->phydev); mdiobus_unregister(bp->mii_bus); mdiobus_free(bp->mii_bus); unregister_netdev(dev); diff --git a/drivers/net/ethernet/dnet.h b/drivers/net/ethernet/dnet.h index 37f5b30..d985080 100644 --- a/drivers/net/ethernet/dnet.h +++ b/drivers/net/ethernet/dnet.h @@ -216,7 +216,6 @@ struct dnet { /* PHY stuff */ struct mii_bus *mii_bus; - struct phy_device *phy_dev; unsigned int link; unsigned int speed; unsigned int duplex; -- cgit v0.10.2 From 1ba44a1f4d8160387bdaab0f414bdbcc74f2d59b Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 23 Jun 2016 23:48:59 +0200 Subject: net: ethernet: dnet: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c index 9c6955f..c3b64cd 100644 --- a/drivers/net/ethernet/dnet.c +++ b/drivers/net/ethernet/dnet.c @@ -730,26 +730,6 @@ static struct net_device_stats *dnet_get_stats(struct net_device *dev) return nstat; } -static int dnet_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = dev->phydev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_gset(phydev, cmd); -} - -static int dnet_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = dev->phydev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_sset(phydev, cmd); -} - static int dnet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { struct phy_device *phydev = dev->phydev; @@ -772,11 +752,11 @@ static void dnet_get_drvinfo(struct net_device *dev, } static const struct ethtool_ops dnet_ethtool_ops = { - .get_settings = dnet_get_settings, - .set_settings = dnet_set_settings, .get_drvinfo = dnet_get_drvinfo, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops dnet_netdev_ops = { -- cgit v0.10.2 From 637c841dd7a5f9bd97b75cbe90b526fa1a52e530 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 23 Jun 2016 18:42:51 -0700 Subject: net: diag: Add support to filter on device index Add support to inet_diag facility to filter sockets based on device index. If an interface index is in the filter only sockets bound to that index (sk_bound_dev_if) are returned. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/include/uapi/linux/inet_diag.h b/include/uapi/linux/inet_diag.h index a166437..abbd1dc 100644 --- a/include/uapi/linux/inet_diag.h +++ b/include/uapi/linux/inet_diag.h @@ -72,6 +72,7 @@ enum { INET_DIAG_BC_AUTO, INET_DIAG_BC_S_COND, INET_DIAG_BC_D_COND, + INET_DIAG_BC_DEV_COND, /* u32 ifindex */ }; struct inet_diag_hostcond { diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index 25af124..38c2c47 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -44,6 +44,7 @@ struct inet_diag_entry { u16 dport; u16 family; u16 userlocks; + u32 ifindex; }; static DEFINE_MUTEX(inet_diag_table_mutex); @@ -571,6 +572,14 @@ static int inet_diag_bc_run(const struct nlattr *_bc, yes = 0; break; } + case INET_DIAG_BC_DEV_COND: { + u32 ifindex; + + ifindex = *((const u32 *)(op + 1)); + if (ifindex != entry->ifindex) + yes = 0; + break; + } } if (yes) { @@ -613,6 +622,7 @@ int inet_diag_bc_sk(const struct nlattr *bc, struct sock *sk) entry_fill_addrs(&entry, sk); entry.sport = inet->inet_num; entry.dport = ntohs(inet->inet_dport); + entry.ifindex = sk->sk_bound_dev_if; entry.userlocks = sk_fullsock(sk) ? sk->sk_userlocks : 0; return inet_diag_bc_run(bc, &entry); @@ -636,6 +646,17 @@ static int valid_cc(const void *bc, int len, int cc) return 0; } +/* data is u32 ifindex */ +static bool valid_devcond(const struct inet_diag_bc_op *op, int len, + int *min_len) +{ + /* Check ifindex space. */ + *min_len += sizeof(u32); + if (len < *min_len) + return false; + + return true; +} /* Validate an inet_diag_hostcond. */ static bool valid_hostcond(const struct inet_diag_bc_op *op, int len, int *min_len) @@ -700,6 +721,10 @@ static int inet_diag_bc_audit(const void *bytecode, int bytecode_len) if (!valid_hostcond(bc, len, &min_len)) return -EINVAL; break; + case INET_DIAG_BC_DEV_COND: + if (!valid_devcond(bc, len, &min_len)) + return -EINVAL; + break; case INET_DIAG_BC_S_GE: case INET_DIAG_BC_S_LE: case INET_DIAG_BC_D_GE: -- cgit v0.10.2 From 56e2f23b7225d2e7b42826aee065cbf96834114d Mon Sep 17 00:00:00 2001 From: Amitoj Kaur Chawla Date: Fri, 24 Jun 2016 11:53:54 +0530 Subject: caif: Remove unneeded header file Drop redundant include of moduleparam.h The Coccinelle semantic patch used to make this change is as follows: @ includesmodule @ @@ #include @ depends on includesmodule @ @@ - #include Signed-off-by: Amitoj Kaur Chawla Signed-off-by: David S. Miller diff --git a/net/caif/chnl_net.c b/net/caif/chnl_net.c index 67a4a36..3408ed5 100644 --- a/net/caif/chnl_net.c +++ b/net/caif/chnl_net.c @@ -13,7 +13,6 @@ #include #include #include -#include #include #include #include -- cgit v0.10.2 From a5e4bd991362223346e1d3561e61d7a25797fe25 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 24 Jun 2016 11:24:08 +0200 Subject: of_mdio: select fixed phy support unconditionally Calling the fixed-phy functions when CONFIG_FIXED_PHY=m as a previous change tried cannot work if the caller is in built-in code: drivers/of/built-in.o: In function `of_phy_register_fixed_link': of_reserved_mem.c:(.text+0x85e0): undefined reference to `fixed_phy_register' Making of_mdio depend on 'FIXED_PHY || !FIXED_PHY' would solve this dependency by enforcing that OF_MDIO itself becomes a loadable module when FIXED_PHY=y, but that creates a different dependency as it breaks any built-in ethernet driver that uses of_mdio. Making FIXED_PHY a bool option also cannot work, since it depends on PHYLIB, which again is tristate. This version now uses 'select FIXED_PHY' to ensure that the fixed-phy portion of of_mdio is not optional. The main downside of this is a small increase in code size for cases that do not need fixed phy support, but it should avoid all of the link-time problems. Signed-off-by: Arnd Bergmann Fixes: d1bd330a229f ("of_mdio: Enable fixed PHY support if driver is a module") Acked-by: Randy Dunlap Signed-off-by: David S. Miller diff --git a/drivers/of/Kconfig b/drivers/of/Kconfig index b3bec3a..bc07ad3 100644 --- a/drivers/of/Kconfig +++ b/drivers/of/Kconfig @@ -74,6 +74,7 @@ config OF_NET config OF_MDIO def_tristate PHYLIB depends on PHYLIB + select FIXED_PHY help OpenFirmware MDIO bus (Ethernet PHY) accessors diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index de68707..e2b50bc 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -361,7 +361,6 @@ struct phy_device *of_phy_attach(struct net_device *dev, } EXPORT_SYMBOL(of_phy_attach); -#if IS_ENABLED(CONFIG_FIXED_PHY) /* * of_phy_is_fixed_link() and of_phy_register_fixed_link() must * support two DT bindings: @@ -451,4 +450,3 @@ int of_phy_register_fixed_link(struct device_node *np) return -ENODEV; } EXPORT_SYMBOL(of_phy_register_fixed_link); -#endif diff --git a/include/linux/of_mdio.h b/include/linux/of_mdio.h index 6c8cb9a..4b04587 100644 --- a/include/linux/of_mdio.h +++ b/include/linux/of_mdio.h @@ -25,6 +25,8 @@ struct phy_device *of_phy_attach(struct net_device *dev, extern struct mii_bus *of_mdio_find_bus(struct device_node *mdio_np); extern int of_mdio_parse_addr(struct device *dev, const struct device_node *np); +extern int of_phy_register_fixed_link(struct device_node *np); +extern bool of_phy_is_fixed_link(struct device_node *np); #else /* CONFIG_OF */ static inline int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np) @@ -67,12 +69,6 @@ static inline int of_mdio_parse_addr(struct device *dev, { return -ENOSYS; } -#endif /* CONFIG_OF */ - -#if defined(CONFIG_OF) && IS_ENABLED(CONFIG_FIXED_PHY) -extern int of_phy_register_fixed_link(struct device_node *np); -extern bool of_phy_is_fixed_link(struct device_node *np); -#else static inline int of_phy_register_fixed_link(struct device_node *np) { return -ENOSYS; -- cgit v0.10.2 From 70523e639bf8ca09b3357371c3546cee55c06351 Mon Sep 17 00:00:00 2001 From: Giuseppe CAVALLARO Date: Fri, 24 Jun 2016 15:16:24 +0200 Subject: drivers: net: stmmac: reworking the PCS code. The 3.xx and 4.xx synopsys gmacs have a very similar PCS embedded module and they share almost the same registers: for example: AN_Control, AN_Status, AN_Advertisement, AN_Link_Partner_Ability, AN_Expansion, TBI_Extended_Status. Just the RGMII/SMII Control/Status register differs. So This patch aims to reorganize and enhance the PCS support. It removes the existent support from the dwmac1000/dwmac4_core.c moving basic PCS functions inside a new file called: stmmac_pcs.h. The patch also reviews the available APIs to be better shared among different hardware and easily enhanced to support new features. Signed-off-by: Giuseppe Cavallaro Signed-off-by: David S. Miller diff --git a/Documentation/networking/stmmac.txt b/Documentation/networking/stmmac.txt index 671fe3d..e226f89 100644 --- a/Documentation/networking/stmmac.txt +++ b/Documentation/networking/stmmac.txt @@ -285,6 +285,7 @@ Please see the following document: o mmc_core.c/mmc.h: Management MAC Counters; o stmmac_hwtstamp.c: HW timestamp support for PTP; o stmmac_ptp.c: PTP 1588 clock; + o stmmac_pcs.h: Physical Coding Sublayer common implementation; o dwmac-.c: these are for the platform glue-logic file; e.g. dwmac-sti.c for STMicroelectronics SoCs. diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index fc60368..86eba2a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -232,6 +232,11 @@ struct stmmac_extra_stats { #define DMA_HW_FEAT_ACTPHYIF 0x70000000 /* Active/selected PHY iface */ #define DEFAULT_DMA_PBL 8 +/* PCS status and mask defines */ +#define PCS_ANE_IRQ BIT(2) /* PCS Auto-Negotiation */ +#define PCS_LINK_IRQ BIT(1) /* PCS Link */ +#define PCS_RGSMIIIS_IRQ BIT(0) /* RGMII or SMII Interrupt */ + /* Max/Min RI Watchdog Timer count value */ #define MAX_DMA_RIWT 0xff #define MIN_DMA_RIWT 0x20 @@ -272,9 +277,6 @@ enum dma_irq_status { #define CORE_IRQ_RX_PATH_IN_LPI_MODE (1 << 2) #define CORE_IRQ_RX_PATH_EXIT_LPI_MODE (1 << 3) -#define CORE_PCS_ANE_COMPLETE (1 << 5) -#define CORE_PCS_LINK_STATUS (1 << 6) -#define CORE_RGMII_IRQ (1 << 7) #define CORE_IRQ_MTL_RX_OVERFLOW BIT(8) /* Physical Coding Sublayer */ @@ -469,9 +471,12 @@ struct stmmac_ops { void (*reset_eee_mode)(struct mac_device_info *hw); void (*set_eee_timer)(struct mac_device_info *hw, int ls, int tw); void (*set_eee_pls)(struct mac_device_info *hw, int link); - void (*ctrl_ane)(struct mac_device_info *hw, bool restart); - void (*get_adv)(struct mac_device_info *hw, struct rgmii_adv *adv); void (*debug)(void __iomem *ioaddr, struct stmmac_extra_stats *x); + /* PCS calls */ + void (*pcs_ctrl_ane)(void __iomem *ioaddr, bool ane, bool srgmi_ral, + bool loopback); + void (*pcs_rane)(void __iomem *ioaddr, bool restart); + void (*pcs_get_adv_lp)(void __iomem *ioaddr, struct rgmii_adv *adv); }; /* PTP and HW Timer helpers */ @@ -546,6 +551,7 @@ void stmmac_dwmac4_get_mac_addr(void __iomem *ioaddr, unsigned char *addr, void stmmac_dwmac4_set_mac(void __iomem *ioaddr, bool enable); void dwmac_dma_flush_tx_fifo(void __iomem *ioaddr); + extern const struct stmmac_mode_ops ring_mode_ops; extern const struct stmmac_mode_ops chain_mode_ops; extern const struct stmmac_desc_ops dwmac4_desc_ops; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h index b0593a4..e671360 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h @@ -46,9 +46,6 @@ enum dwmac1000_irq_status { mmc_rx_irq = 0x0020, mmc_irq = 0x0010, pmt_irq = 0x0008, - pcs_ane_irq = 0x0004, - pcs_link_irq = 0x0002, - rgmii_irq = 0x0001, }; #define GMAC_INT_MASK 0x0000003c /* interrupt mask register */ @@ -90,42 +87,23 @@ enum power_event { (reg * 8)) #define GMAC_MAX_PERFECT_ADDRESSES 1 -/* PCS registers (AN/TBI/SGMII/RGMII) offset */ -#define GMAC_AN_CTRL 0x000000c0 /* AN control */ -#define GMAC_AN_STATUS 0x000000c4 /* AN status */ -#define GMAC_ANE_ADV 0x000000c8 /* Auto-Neg. Advertisement */ -#define GMAC_ANE_LPA 0x000000cc /* Auto-Neg. link partener ability */ -#define GMAC_ANE_EXP 0x000000d0 /* ANE expansion */ -#define GMAC_TBI 0x000000d4 /* TBI extend status */ -#define GMAC_S_R_GMII 0x000000d8 /* SGMII RGMII status */ - -/* AN Configuration defines */ -#define GMAC_AN_CTRL_RAN 0x00000200 /* Restart Auto-Negotiation */ -#define GMAC_AN_CTRL_ANE 0x00001000 /* Auto-Negotiation Enable */ -#define GMAC_AN_CTRL_ELE 0x00004000 /* External Loopback Enable */ -#define GMAC_AN_CTRL_ECD 0x00010000 /* Enable Comma Detect */ -#define GMAC_AN_CTRL_LR 0x00020000 /* Lock to Reference */ -#define GMAC_AN_CTRL_SGMRAL 0x00040000 /* SGMII RAL Control */ - -/* AN Status defines */ -#define GMAC_AN_STATUS_LS 0x00000004 /* Link Status 0:down 1:up */ -#define GMAC_AN_STATUS_ANA 0x00000008 /* Auto-Negotiation Ability */ -#define GMAC_AN_STATUS_ANC 0x00000020 /* Auto-Negotiation Complete */ -#define GMAC_AN_STATUS_ES 0x00000100 /* Extended Status */ - -/* Register 54 (SGMII/RGMII status register) */ -#define GMAC_S_R_GMII_LINK 0x8 -#define GMAC_S_R_GMII_SPEED 0x5 -#define GMAC_S_R_GMII_SPEED_SHIFT 0x1 -#define GMAC_S_R_GMII_MODE 0x1 -#define GMAC_S_R_GMII_SPEED_125 2 -#define GMAC_S_R_GMII_SPEED_25 1 - -/* Common ADV and LPA defines */ -#define GMAC_ANE_FD (1 << 5) -#define GMAC_ANE_HD (1 << 6) -#define GMAC_ANE_PSE (3 << 7) -#define GMAC_ANE_PSE_SHIFT 7 +#define GMAC_PCS_BASE 0x000000c0 /* PCS register base */ +#define GMAC_RGSMIIIS 0x000000d8 /* RGMII/SMII status */ + +/* SGMII/RGMII status register */ +#define GMAC_RGSMIIIS_LNKMODE BIT(0) +#define GMAC_RGSMIIIS_SPEED GENMASK(2, 1) +#define GMAC_RGSMIIIS_SPEED_SHIFT 1 +#define GMAC_RGSMIIIS_LNKSTS BIT(3) +#define GMAC_RGSMIIIS_JABTO BIT(4) +#define GMAC_RGSMIIIS_FALSECARDET BIT(5) +#define GMAC_RGSMIIIS_SMIDRXS BIT(16) +/* LNKMOD */ +#define GMAC_RGSMIIIS_LNKMOD_MASK 0x1 +/* LNKSPEED */ +#define GMAC_RGSMIIIS_SPEED_125 0x2 +#define GMAC_RGSMIIIS_SPEED_25 0x1 +#define GMAC_RGSMIIIS_SPEED_2_5 0x0 /* GMAC Configuration defines */ #define GMAC_CONTROL_2K 0x08000000 /* IEEE 802.3as 2K packets */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index fb1eb57..9772a43c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -30,6 +30,7 @@ #include #include #include +#include "stmmac_pcs.h" #include "dwmac1000.h" static void dwmac1000_core_init(struct mac_device_info *hw, int mtu) @@ -241,6 +242,39 @@ static void dwmac1000_pmt(struct mac_device_info *hw, unsigned long mode) writel(pmt, ioaddr + GMAC_PMT); } +/* RGMII or SMII interface */ +static void dwmac1000_rgsmii(void __iomem *ioaddr, struct stmmac_extra_stats *x) +{ + u32 status; + + status = readl(ioaddr + GMAC_RGSMIIIS); + x->irq_rgmii_n++; + + /* Check the link status */ + if (status & GMAC_RGSMIIIS_LNKSTS) { + int speed_value; + + x->pcs_link = 1; + + speed_value = ((status & GMAC_RGSMIIIS_SPEED) >> + GMAC_RGSMIIIS_SPEED_SHIFT); + if (speed_value == GMAC_RGSMIIIS_SPEED_125) + x->pcs_speed = SPEED_1000; + else if (speed_value == GMAC_RGSMIIIS_SPEED_25) + x->pcs_speed = SPEED_100; + else + x->pcs_speed = SPEED_10; + + x->pcs_duplex = (status & GMAC_RGSMIIIS_LNKMOD_MASK); + + pr_info("Link is Up - %d/%s\n", (int)x->pcs_speed, + x->pcs_duplex ? "Full" : "Half"); + } else { + x->pcs_link = 0; + pr_info("Link is Down\n"); + } +} + static int dwmac1000_irq_status(struct mac_device_info *hw, struct stmmac_extra_stats *x) { @@ -260,6 +294,7 @@ static int dwmac1000_irq_status(struct mac_device_info *hw, readl(ioaddr + GMAC_PMT); x->irq_receive_pmt_irq_n++; } + /* MAC trx/rx EEE LPI entry/exit interrupts */ if (intr_status & lpiis_irq) { /* Clean LPI interrupt by reading the Reg 12 */ @@ -275,36 +310,10 @@ static int dwmac1000_irq_status(struct mac_device_info *hw, x->irq_rx_path_exit_lpi_mode_n++; } - if ((intr_status & pcs_ane_irq) || (intr_status & pcs_link_irq)) { - readl(ioaddr + GMAC_AN_STATUS); - x->irq_pcs_ane_n++; - } - if (intr_status & rgmii_irq) { - u32 status = readl(ioaddr + GMAC_S_R_GMII); - x->irq_rgmii_n++; - - /* Save and dump the link status. */ - if (status & GMAC_S_R_GMII_LINK) { - int speed_value = (status & GMAC_S_R_GMII_SPEED) >> - GMAC_S_R_GMII_SPEED_SHIFT; - x->pcs_duplex = (status & GMAC_S_R_GMII_MODE); - - if (speed_value == GMAC_S_R_GMII_SPEED_125) - x->pcs_speed = SPEED_1000; - else if (speed_value == GMAC_S_R_GMII_SPEED_25) - x->pcs_speed = SPEED_100; - else - x->pcs_speed = SPEED_10; - - x->pcs_link = 1; - pr_debug("%s: Link is Up - %d/%s\n", __func__, - (int)x->pcs_speed, - x->pcs_duplex ? "Full" : "Half"); - } else { - x->pcs_link = 0; - pr_debug("%s: Link is Down\n", __func__); - } - } + dwmac_pcs_isr(ioaddr, GMAC_PCS_BASE, intr_status, x); + + if (intr_status & PCS_RGSMIIIS_IRQ) + dwmac1000_rgsmii(ioaddr, x); return ret; } @@ -363,38 +372,20 @@ static void dwmac1000_set_eee_timer(struct mac_device_info *hw, int ls, int tw) writel(value, ioaddr + LPI_TIMER_CTRL); } -static void dwmac1000_ctrl_ane(struct mac_device_info *hw, bool restart) +static void dwmac1000_ctrl_ane(void __iomem *ioaddr, bool ane, bool srgmi_ral, + bool loopback) { - void __iomem *ioaddr = hw->pcsr; - /* auto negotiation enable and External Loopback enable */ - u32 value = GMAC_AN_CTRL_ANE | GMAC_AN_CTRL_ELE; - - if (restart) - value |= GMAC_AN_CTRL_RAN; - - writel(value, ioaddr + GMAC_AN_CTRL); + dwmac_ctrl_ane(ioaddr, GMAC_PCS_BASE, ane, srgmi_ral, loopback); } -static void dwmac1000_get_adv(struct mac_device_info *hw, struct rgmii_adv *adv) +static void dwmac1000_rane(void __iomem *ioaddr, bool restart) { - void __iomem *ioaddr = hw->pcsr; - u32 value = readl(ioaddr + GMAC_ANE_ADV); - - if (value & GMAC_ANE_FD) - adv->duplex = DUPLEX_FULL; - if (value & GMAC_ANE_HD) - adv->duplex |= DUPLEX_HALF; - - adv->pause = (value & GMAC_ANE_PSE) >> GMAC_ANE_PSE_SHIFT; - - value = readl(ioaddr + GMAC_ANE_LPA); - - if (value & GMAC_ANE_FD) - adv->lp_duplex = DUPLEX_FULL; - if (value & GMAC_ANE_HD) - adv->lp_duplex = DUPLEX_HALF; + dwmac_rane(ioaddr, GMAC_PCS_BASE, restart); +} - adv->lp_pause = (value & GMAC_ANE_PSE) >> GMAC_ANE_PSE_SHIFT; +static void dwmac1000_get_adv_lp(void __iomem *ioaddr, struct rgmii_adv *adv) +{ + dwmac_get_adv_lp(ioaddr, GMAC_PCS_BASE, adv); } static void dwmac1000_debug(void __iomem *ioaddr, struct stmmac_extra_stats *x) @@ -485,9 +476,10 @@ static const struct stmmac_ops dwmac1000_ops = { .reset_eee_mode = dwmac1000_reset_eee_mode, .set_eee_timer = dwmac1000_set_eee_timer, .set_eee_pls = dwmac1000_set_eee_pls, - .ctrl_ane = dwmac1000_ctrl_ane, - .get_adv = dwmac1000_get_adv, .debug = dwmac1000_debug, + .pcs_ctrl_ane = dwmac1000_ctrl_ane, + .pcs_rane = dwmac1000_rane, + .pcs_get_adv_lp = dwmac1000_get_adv_lp, }; struct mac_device_info *dwmac1000_setup(void __iomem *ioaddr, int mcbins, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h index bc50952..227fa20 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h @@ -24,10 +24,8 @@ #define GMAC_QX_TX_FLOW_CTRL(x) (0x70 + x * 4) #define GMAC_INT_STATUS 0x000000b0 #define GMAC_INT_EN 0x000000b4 -#define GMAC_AN_CTRL 0x000000e0 -#define GMAC_AN_STATUS 0x000000e4 -#define GMAC_AN_ADV 0x000000e8 -#define GMAC_AN_LPA 0x000000ec +#define GMAC_PCS_BASE 0x000000e0 +#define GMAC_PHYIF_CONTROL_STATUS 0x000000f8 #define GMAC_PMT 0x000000c0 #define GMAC_VERSION 0x00000110 #define GMAC_DEBUG 0x00000114 @@ -64,19 +62,8 @@ enum dwmac4_irq_status { mmc_rx_irq = 0x00000200, mmc_irq = 0x00000100, pmt_irq = 0x00000010, - pcs_ane_irq = 0x00000004, - pcs_link_irq = 0x00000002, }; -/* MAC Auto-Neg bitmap*/ -#define GMAC_AN_CTRL_RAN BIT(9) -#define GMAC_AN_CTRL_ANE BIT(12) -#define GMAC_AN_CTRL_ELE BIT(14) -#define GMAC_AN_FD BIT(5) -#define GMAC_AN_HD BIT(6) -#define GMAC_AN_PSE_MASK GENMASK(8, 7) -#define GMAC_AN_PSE_SHIFT 7 - /* MAC PMT bitmap */ enum power_event { pointer_reset = 0x80000000, @@ -250,6 +237,23 @@ enum power_event { #define MTL_DEBUG_RRCSTS_FLUSH 3 #define MTL_DEBUG_RWCSTS BIT(0) +/* SGMII/RGMII status register */ +#define GMAC_PHYIF_CTRLSTATUS_TC BIT(0) +#define GMAC_PHYIF_CTRLSTATUS_LUD BIT(1) +#define GMAC_PHYIF_CTRLSTATUS_SMIDRXS BIT(4) +#define GMAC_PHYIF_CTRLSTATUS_LNKMOD BIT(16) +#define GMAC_PHYIF_CTRLSTATUS_SPEED GENMASK(18, 17) +#define GMAC_PHYIF_CTRLSTATUS_SPEED_SHIFT 17 +#define GMAC_PHYIF_CTRLSTATUS_LNKSTS BIT(19) +#define GMAC_PHYIF_CTRLSTATUS_JABTO BIT(20) +#define GMAC_PHYIF_CTRLSTATUS_FALSECARDET BIT(21) +/* LNKMOD */ +#define GMAC_PHYIF_CTRLSTATUS_LNKMOD_MASK 0x1 +/* LNKSPEED */ +#define GMAC_PHYIF_CTRLSTATUS_SPEED_125 0x2 +#define GMAC_PHYIF_CTRLSTATUS_SPEED_25 0x1 +#define GMAC_PHYIF_CTRLSTATUS_SPEED_2_5 0x0 + extern const struct stmmac_dma_ops dwmac4_dma_ops; extern const struct stmmac_dma_ops dwmac410_dma_ops; #endif /* __DWMAC4_H__ */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 44da877..207d8bb 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -17,6 +17,7 @@ #include #include #include +#include "stmmac_pcs.h" #include "dwmac4.h" static void dwmac4_core_init(struct mac_device_info *hw, int mtu) @@ -190,39 +191,53 @@ static void dwmac4_flow_ctrl(struct mac_device_info *hw, unsigned int duplex, } } -static void dwmac4_ctrl_ane(struct mac_device_info *hw, bool restart) +static void dwmac4_ctrl_ane(void __iomem *ioaddr, bool ane, bool srgmi_ral, + bool loopback) { - void __iomem *ioaddr = hw->pcsr; - - /* auto negotiation enable and External Loopback enable */ - u32 value = GMAC_AN_CTRL_ANE | GMAC_AN_CTRL_ELE; + dwmac_ctrl_ane(ioaddr, GMAC_PCS_BASE, ane, srgmi_ral, loopback); +} - if (restart) - value |= GMAC_AN_CTRL_RAN; +static void dwmac4_rane(void __iomem *ioaddr, bool restart) +{ + dwmac_rane(ioaddr, GMAC_PCS_BASE, restart); +} - writel(value, ioaddr + GMAC_AN_CTRL); +static void dwmac4_get_adv_lp(void __iomem *ioaddr, struct rgmii_adv *adv) +{ + dwmac_get_adv_lp(ioaddr, GMAC_PCS_BASE, adv); } -static void dwmac4_get_adv(struct mac_device_info *hw, struct rgmii_adv *adv) +/* RGMII or SMII interface */ +static void dwmac4_phystatus(void __iomem *ioaddr, struct stmmac_extra_stats *x) { - void __iomem *ioaddr = hw->pcsr; - u32 value = readl(ioaddr + GMAC_AN_ADV); + u32 status; - if (value & GMAC_AN_FD) - adv->duplex = DUPLEX_FULL; - if (value & GMAC_AN_HD) - adv->duplex |= DUPLEX_HALF; + status = readl(ioaddr + GMAC_PHYIF_CONTROL_STATUS); + x->irq_rgmii_n++; - adv->pause = (value & GMAC_AN_PSE_MASK) >> GMAC_AN_PSE_SHIFT; + /* Check the link status */ + if (status & GMAC_PHYIF_CTRLSTATUS_LNKSTS) { + int speed_value; - value = readl(ioaddr + GMAC_AN_LPA); + x->pcs_link = 1; - if (value & GMAC_AN_FD) - adv->lp_duplex = DUPLEX_FULL; - if (value & GMAC_AN_HD) - adv->lp_duplex = DUPLEX_HALF; + speed_value = ((status & GMAC_PHYIF_CTRLSTATUS_SPEED) >> + GMAC_PHYIF_CTRLSTATUS_SPEED_SHIFT); + if (speed_value == GMAC_PHYIF_CTRLSTATUS_SPEED_125) + x->pcs_speed = SPEED_1000; + else if (speed_value == GMAC_PHYIF_CTRLSTATUS_SPEED_25) + x->pcs_speed = SPEED_100; + else + x->pcs_speed = SPEED_10; + + x->pcs_duplex = (status & GMAC_PHYIF_CTRLSTATUS_LNKMOD_MASK); - adv->lp_pause = (value & GMAC_AN_PSE_MASK) >> GMAC_AN_PSE_SHIFT; + pr_info("Link is Up - %d/%s\n", (int)x->pcs_speed, + x->pcs_duplex ? "Full" : "Half"); + } else { + x->pcs_link = 0; + pr_info("Link is Down\n"); + } } static int dwmac4_irq_status(struct mac_device_info *hw, @@ -248,11 +263,6 @@ static int dwmac4_irq_status(struct mac_device_info *hw, x->irq_receive_pmt_irq_n++; } - if ((intr_status & pcs_ane_irq) || (intr_status & pcs_link_irq)) { - readl(ioaddr + GMAC_AN_STATUS); - x->irq_pcs_ane_n++; - } - mtl_int_qx_status = readl(ioaddr + MTL_INT_STATUS); /* Check MTL Interrupt: Currently only one queue is used: Q0. */ if (mtl_int_qx_status & MTL_INT_Q0) { @@ -267,6 +277,10 @@ static int dwmac4_irq_status(struct mac_device_info *hw, } } + dwmac_pcs_isr(ioaddr, GMAC_PCS_BASE, intr_status, x); + if (intr_status & PCS_RGSMIIIS_IRQ) + dwmac4_phystatus(ioaddr, x); + return ret; } @@ -363,8 +377,9 @@ static const struct stmmac_ops dwmac4_ops = { .pmt = dwmac4_pmt, .set_umac_addr = dwmac4_set_umac_addr, .get_umac_addr = dwmac4_get_umac_addr, - .ctrl_ane = dwmac4_ctrl_ane, - .get_adv = dwmac4_get_adv, + .pcs_ctrl_ane = dwmac4_ctrl_ane, + .pcs_rane = dwmac4_rane, + .pcs_get_adv_lp = dwmac4_get_adv_lp, .debug = dwmac4_debug, .set_filter = dwmac4_set_filter, }; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index e2b98b0..a5f4f46 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -289,10 +289,10 @@ static int stmmac_ethtool_getsettings(struct net_device *dev, ethtool_cmd_speed_set(cmd, priv->xstats.pcs_speed); /* Get and convert ADV/LP_ADV from the HW AN registers */ - if (!priv->hw->mac->get_adv) + if (!priv->hw->mac->pcs_get_adv_lp) return -EOPNOTSUPP; /* should never happen indeed */ - priv->hw->mac->get_adv(priv->hw, &adv); + priv->hw->mac->pcs_get_adv_lp(priv->ioaddr, &adv); /* Encoding of PSE bits is defined in 802.3z, 37.2.1.4 */ @@ -376,8 +376,10 @@ static int stmmac_ethtool_setsettings(struct net_device *dev, ADVERTISED_10baseT_Full); spin_lock(&priv->lock); - if (priv->hw->mac->ctrl_ane) - priv->hw->mac->ctrl_ane(priv->hw, 1); + + if (priv->hw->mac->pcs_ctrl_ane) + priv->hw->mac->pcs_ctrl_ane(priv->ioaddr, 1, 0, 0); + spin_unlock(&priv->lock); return 0; @@ -452,11 +454,22 @@ stmmac_get_pauseparam(struct net_device *netdev, { struct stmmac_priv *priv = netdev_priv(netdev); - if (priv->pcs) /* FIXME */ - return; - pause->rx_pause = 0; pause->tx_pause = 0; + + if (priv->pcs && priv->hw->mac->pcs_get_adv_lp) { + struct rgmii_adv adv_lp; + + pause->autoneg = 1; + priv->hw->mac->pcs_get_adv_lp(priv->ioaddr, &adv_lp); + if (!adv_lp.pause) + return; + } else { + if (!(priv->phydev->supported & SUPPORTED_Pause) || + !(priv->phydev->supported & SUPPORTED_Asym_Pause)) + return; + } + pause->autoneg = priv->phydev->autoneg; if (priv->flow_ctrl & FLOW_RX) @@ -473,10 +486,19 @@ stmmac_set_pauseparam(struct net_device *netdev, struct stmmac_priv *priv = netdev_priv(netdev); struct phy_device *phy = priv->phydev; int new_pause = FLOW_OFF; - int ret = 0; - if (priv->pcs) /* FIXME */ - return -EOPNOTSUPP; + if (priv->pcs && priv->hw->mac->pcs_get_adv_lp) { + struct rgmii_adv adv_lp; + + pause->autoneg = 1; + priv->hw->mac->pcs_get_adv_lp(priv->ioaddr, &adv_lp); + if (!adv_lp.pause) + return -EOPNOTSUPP; + } else { + if (!(phy->supported & SUPPORTED_Pause) || + !(phy->supported & SUPPORTED_Asym_Pause)) + return -EOPNOTSUPP; + } if (pause->rx_pause) new_pause |= FLOW_RX; @@ -485,14 +507,14 @@ stmmac_set_pauseparam(struct net_device *netdev, priv->flow_ctrl = new_pause; phy->autoneg = pause->autoneg; - if (phy->autoneg) { if (netif_running(netdev)) - ret = phy_start_aneg(phy); - } else - priv->hw->mac->flow_ctrl(priv->hw, phy->duplex, - priv->flow_ctrl, priv->pause); - return ret; + return phy_start_aneg(phy); + } + + priv->hw->mac->flow_ctrl(priv->hw, phy->duplex, priv->flow_ctrl, + priv->pause); + return 0; } static void stmmac_get_ethtool_stats(struct net_device *dev, diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index a473c18..6c43d68 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1714,8 +1714,8 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) priv->hw->dma->rx_watchdog(priv->ioaddr, MAX_DMA_RIWT); } - if (priv->pcs && priv->hw->mac->ctrl_ane) - priv->hw->mac->ctrl_ane(priv->hw, 0); + if (priv->pcs && priv->hw->mac->pcs_ctrl_ane) + priv->hw->mac->pcs_ctrl_ane(priv->hw, 1, 0, 0); /* set TX ring length */ if (priv->hw->dma->set_tx_ring_len) @@ -2809,6 +2809,14 @@ static irqreturn_t stmmac_interrupt(int irq, void *dev_id) priv->rx_tail_addr, STMMAC_CHAN0); } + + /* PCS link status */ + if (priv->pcs) { + if (priv->xstats.pcs_link) + netif_carrier_on(dev); + else + netif_carrier_off(dev); + } } /* To handle DMA interrupts */ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h new file mode 100644 index 0000000..eba41c2 --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pcs.h @@ -0,0 +1,159 @@ +/* + * stmmac_pcs.h: Physical Coding Sublayer Header File + * + * Copyright (C) 2016 STMicroelectronics (R&D) Limited + * Author: Giuseppe Cavallaro + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __STMMAC_PCS_H__ +#define __STMMAC_PCS_H__ + +#include +#include +#include "common.h" + +/* PCS registers (AN/TBI/SGMII/RGMII) offsets */ +#define GMAC_AN_CTRL(x) (x) /* AN control */ +#define GMAC_AN_STATUS(x) (x + 0x4) /* AN status */ +#define GMAC_ANE_ADV(x) (x + 0x8) /* ANE Advertisement */ +#define GMAC_ANE_LPA(x) (x + 0xc) /* ANE link partener ability */ +#define GMAC_ANE_EXP(x) (x + 0x10) /* ANE expansion */ +#define GMAC_TBI(x) (x + 0x14) /* TBI extend status */ + +/* AN Configuration defines */ +#define GMAC_AN_CTRL_RAN BIT(9) /* Restart Auto-Negotiation */ +#define GMAC_AN_CTRL_ANE BIT(12) /* Auto-Negotiation Enable */ +#define GMAC_AN_CTRL_ELE BIT(14) /* External Loopback Enable */ +#define GMAC_AN_CTRL_ECD BIT(16) /* Enable Comma Detect */ +#define GMAC_AN_CTRL_LR BIT(17) /* Lock to Reference */ +#define GMAC_AN_CTRL_SGMRAL BIT(18) /* SGMII RAL Control */ + +/* AN Status defines */ +#define GMAC_AN_STATUS_LS BIT(2) /* Link Status 0:down 1:up */ +#define GMAC_AN_STATUS_ANA BIT(3) /* Auto-Negotiation Ability */ +#define GMAC_AN_STATUS_ANC BIT(5) /* Auto-Negotiation Complete */ +#define GMAC_AN_STATUS_ES BIT(8) /* Extended Status */ + +/* ADV and LPA defines */ +#define GMAC_ANE_FD BIT(5) +#define GMAC_ANE_HD BIT(6) +#define GMAC_ANE_PSE GENMASK(8, 7) +#define GMAC_ANE_PSE_SHIFT 7 +#define GMAC_ANE_RFE GENMASK(13, 12) +#define GMAC_ANE_RFE_SHIFT 12 +#define GMAC_ANE_ACK BIT(14) + +/** + * dwmac_pcs_isr - TBI, RTBI, or SGMII PHY ISR + * @ioaddr: IO registers pointer + * @reg: Base address of the AN Control Register. + * @intr_status: GMAC core interrupt status + * @x: pointer to log these events as stats + * Description: it is the ISR for PCS events: Auto-Negotiation Completed and + * Link status. + */ +static inline void dwmac_pcs_isr(void __iomem *ioaddr, u32 reg, + unsigned int intr_status, + struct stmmac_extra_stats *x) +{ + u32 val = readl(ioaddr + GMAC_AN_STATUS(reg)); + + if (intr_status & PCS_ANE_IRQ) { + x->irq_pcs_ane_n++; + if (val & GMAC_AN_STATUS_ANC) + pr_info("stmmac_pcs: ANE process completed\n"); + } + + if (intr_status & PCS_LINK_IRQ) { + x->irq_pcs_link_n++; + if (val & GMAC_AN_STATUS_LS) + pr_info("stmmac_pcs: Link Up\n"); + else + pr_info("stmmac_pcs: Link Down\n"); + } +} + +/** + * dwmac_rane - To restart ANE + * @ioaddr: IO registers pointer + * @reg: Base address of the AN Control Register. + * @restart: to restart ANE + * Description: this is to just restart the Auto-Negotiation. + */ +static inline void dwmac_rane(void __iomem *ioaddr, u32 reg, bool restart) +{ + u32 value = readl(ioaddr + GMAC_AN_CTRL(reg)); + + if (restart) + value |= GMAC_AN_CTRL_RAN; + + writel(value, ioaddr + GMAC_AN_CTRL(reg)); +} + +/** + * dwmac_ctrl_ane - To program the AN Control Register. + * @ioaddr: IO registers pointer + * @reg: Base address of the AN Control Register. + * @ane: to enable the auto-negotiation + * @srgmi_ral: to manage MAC-2-MAC SGMII connections. + * @loopback: to cause the PHY to loopback tx data into rx path. + * Description: this is the main function to configure the AN control register + * and init the ANE, select loopback (usually for debugging purpose) and + * configure SGMII RAL. + */ +static inline void dwmac_ctrl_ane(void __iomem *ioaddr, u32 reg, bool ane, + bool srgmi_ral, bool loopback) +{ + u32 value = readl(ioaddr + GMAC_AN_CTRL(reg)); + + /* Enable and restart the Auto-Negotiation */ + if (ane) + value |= GMAC_AN_CTRL_ANE | GMAC_AN_CTRL_RAN; + + /* In case of MAC-2-MAC connection, block is configured to operate + * according to MAC conf register. + */ + if (srgmi_ral) + value |= GMAC_AN_CTRL_SGMRAL; + + if (loopback) + value |= GMAC_AN_CTRL_ELE; + + writel(value, ioaddr + GMAC_AN_CTRL(reg)); +} + +/** + * dwmac_get_adv_lp - Get ADV and LP cap + * @ioaddr: IO registers pointer + * @reg: Base address of the AN Control Register. + * @adv_lp: structure to store the adv,lp status + * Description: this is to expose the ANE advertisement and Link partner ability + * status to ethtool support. + */ +static inline void dwmac_get_adv_lp(void __iomem *ioaddr, u32 reg, + struct rgmii_adv *adv_lp) +{ + u32 value = readl(ioaddr + GMAC_ANE_ADV(reg)); + + if (value & GMAC_ANE_FD) + adv_lp->duplex = DUPLEX_FULL; + if (value & GMAC_ANE_HD) + adv_lp->duplex |= DUPLEX_HALF; + + adv_lp->pause = (value & GMAC_ANE_PSE) >> GMAC_ANE_PSE_SHIFT; + + value = readl(ioaddr + GMAC_ANE_LPA(reg)); + + if (value & GMAC_ANE_FD) + adv_lp->lp_duplex = DUPLEX_FULL; + if (value & GMAC_ANE_HD) + adv_lp->lp_duplex = DUPLEX_HALF; + + adv_lp->lp_pause = (value & GMAC_ANE_PSE) >> GMAC_ANE_PSE_SHIFT; +} +#endif /* __STMMAC_PCS_H__ */ -- cgit v0.10.2 From 3fe5cadbd3494b9d0fcea41ccefeb319528c774e Mon Sep 17 00:00:00 2001 From: Giuseppe CAVALLARO Date: Fri, 24 Jun 2016 15:16:25 +0200 Subject: drivers: net: stmmac: rework core ISR to better manage PCS and PMT By default, all gmac cores disable the PCS block and always enable the PMT. Note that this is done in a different way by 3.x and 4.x cores. With this rework, PCS and PMT interrupt masks can be driven by parameters now moved inside the mac_device_info structure and the settings follow what the HW capability register reports. Signed-off-by: Giuseppe Cavallaro Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 86eba2a..51077a8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -529,6 +529,8 @@ struct mac_device_info { int unicast_filter_entries; int mcast_bits_log2; unsigned int rx_csum; + unsigned int pcs; + unsigned int pmt; }; struct mac_device_info *dwmac1000_setup(void __iomem *ioaddr, int mcbins, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h index e671360..ff3e5ab 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000.h @@ -38,16 +38,26 @@ #define GMAC_WAKEUP_FILTER 0x00000028 /* Wake-up Frame Filter */ #define GMAC_INT_STATUS 0x00000038 /* interrupt status register */ -enum dwmac1000_irq_status { - lpiis_irq = 0x400, - time_stamp_irq = 0x0200, - mmc_rx_csum_offload_irq = 0x0080, - mmc_tx_irq = 0x0040, - mmc_rx_irq = 0x0020, - mmc_irq = 0x0010, - pmt_irq = 0x0008, -}; -#define GMAC_INT_MASK 0x0000003c /* interrupt mask register */ +#define GMAC_INT_STATUS_PMT BIT(3) +#define GMAC_INT_STATUS_MMCIS BIT(4) +#define GMAC_INT_STATUS_MMCRIS BIT(5) +#define GMAC_INT_STATUS_MMCTIS BIT(6) +#define GMAC_INT_STATUS_MMCCSUM BIT(7) +#define GMAC_INT_STATUS_TSTAMP BIT(9) +#define GMAC_INT_STATUS_LPIIS BIT(10) + +/* interrupt mask register */ +#define GMAC_INT_MASK 0x0000003c +#define GMAC_INT_DISABLE_RGMII BIT(0) +#define GMAC_INT_DISABLE_PCSLINK BIT(1) +#define GMAC_INT_DISABLE_PCSAN BIT(2) +#define GMAC_INT_DISABLE_PMT BIT(3) +#define GMAC_INT_DISABLE_TIMESTAMP BIT(9) +#define GMAC_INT_DISABLE_PCS (GMAC_INT_DISABLE_RGMII | \ + GMAC_INT_DISABLE_PCSLINK | \ + GMAC_INT_DISABLE_PCSAN) +#define GMAC_INT_DEFAULT_MASK (GMAC_INT_DISABLE_TIMESTAMP | \ + GMAC_INT_DISABLE_PCS) /* PMT Control and Status */ #define GMAC_PMT 0x0000002c diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index 9772a43c..0d31f2f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -37,7 +37,10 @@ static void dwmac1000_core_init(struct mac_device_info *hw, int mtu) { void __iomem *ioaddr = hw->pcsr; u32 value = readl(ioaddr + GMAC_CONTROL); + + /* Configure GMAC core */ value |= GMAC_CORE_INIT; + if (mtu > 1500) value |= GMAC_CONTROL_2K; if (mtu > 2000) @@ -46,7 +49,14 @@ static void dwmac1000_core_init(struct mac_device_info *hw, int mtu) writel(value, ioaddr + GMAC_CONTROL); /* Mask GMAC interrupts */ - writel(0x207, ioaddr + GMAC_INT_MASK); + value = GMAC_INT_DEFAULT_MASK; + + if (hw->pmt) + value &= ~GMAC_INT_DISABLE_PMT; + if (hw->pcs) + value &= ~GMAC_INT_DISABLE_PCS; + + writel(value, ioaddr + GMAC_INT_MASK); #ifdef STMMAC_VLAN_TAG_USED /* Tag detection without filtering */ @@ -283,20 +293,20 @@ static int dwmac1000_irq_status(struct mac_device_info *hw, int ret = 0; /* Not used events (e.g. MMC interrupts) are not handled. */ - if ((intr_status & mmc_tx_irq)) + if ((intr_status & GMAC_INT_STATUS_MMCTIS)) x->mmc_tx_irq_n++; - if (unlikely(intr_status & mmc_rx_irq)) + if (unlikely(intr_status & GMAC_INT_STATUS_MMCRIS)) x->mmc_rx_irq_n++; - if (unlikely(intr_status & mmc_rx_csum_offload_irq)) + if (unlikely(intr_status & GMAC_INT_STATUS_MMCCSUM)) x->mmc_rx_csum_offload_irq_n++; - if (unlikely(intr_status & pmt_irq)) { + if (unlikely(intr_status & GMAC_INT_DISABLE_PMT)) { /* clear the PMT bits 5 and 6 by reading the PMT status reg */ readl(ioaddr + GMAC_PMT); x->irq_receive_pmt_irq_n++; } - /* MAC trx/rx EEE LPI entry/exit interrupts */ - if (intr_status & lpiis_irq) { + /* MAC tx/rx EEE LPI entry/exit interrupts */ + if (intr_status & GMAC_INT_STATUS_LPIIS) { /* Clean LPI interrupt by reading the Reg 12 */ ret = readl(ioaddr + LPI_CTRL_STATUS); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h index 227fa20..6f4f5ce 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h @@ -52,9 +52,18 @@ #define GMAC_TX_FLOW_CTRL_PT_SHIFT 16 /* MAC Interrupt bitmap*/ +#define GMAC_INT_RGSMIIS BIT(0) +#define GMAC_INT_PCS_LINK BIT(1) +#define GMAC_INT_PCS_ANE BIT(2) +#define GMAC_INT_PCS_PHYIS BIT(3) #define GMAC_INT_PMT_EN BIT(4) #define GMAC_INT_LPI_EN BIT(5) +#define GMAC_PCS_IRQ_DEFAULT (GMAC_INT_RGSMIIS | GMAC_INT_PCS_LINK | \ + GMAC_INT_PCS_ANE) + +#define GMAC_INT_DEFAULT_MASK GMAC_INT_PMT_EN + enum dwmac4_irq_status { time_stamp_irq = 0x00001000, mmc_rx_csum_offload_irq = 0x00000800, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 207d8bb..747f3cf 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -35,7 +35,13 @@ static void dwmac4_core_init(struct mac_device_info *hw, int mtu) writel(value, ioaddr + GMAC_CONFIG); /* Mask GMAC interrupts */ - writel(GMAC_INT_PMT_EN, ioaddr + GMAC_INT_EN); + value = GMAC_INT_DEFAULT_MASK; + if (hw->pmt) + value |= GMAC_INT_PMT_EN; + if (hw->pcs) + value |= GMAC_PCS_IRQ_DEFAULT; + + writel(value, ioaddr + GMAC_INT_EN); } static void dwmac4_dump_regs(struct mac_device_info *hw) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index 59ae608..8dc9056 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -117,7 +117,6 @@ struct stmmac_priv { int eee_enabled; int eee_active; int tx_lpi_timer; - int pcs; unsigned int mode; int extend_desc; struct ptp_clock *ptp_clock; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index a5f4f46..da2d9b5 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -276,7 +276,8 @@ static int stmmac_ethtool_getsettings(struct net_device *dev, struct phy_device *phy = priv->phydev; int rc; - if ((priv->pcs & STMMAC_PCS_RGMII) || (priv->pcs & STMMAC_PCS_SGMII)) { + if (priv->hw->pcs & STMMAC_PCS_RGMII || + priv->hw->pcs & STMMAC_PCS_SGMII) { struct rgmii_adv adv; if (!priv->xstats.pcs_link) { @@ -361,7 +362,8 @@ static int stmmac_ethtool_setsettings(struct net_device *dev, struct phy_device *phy = priv->phydev; int rc; - if ((priv->pcs & STMMAC_PCS_RGMII) || (priv->pcs & STMMAC_PCS_SGMII)) { + if (priv->hw->pcs & STMMAC_PCS_RGMII || + priv->hw->pcs & STMMAC_PCS_SGMII) { u32 mask = ADVERTISED_Autoneg | ADVERTISED_Pause; /* Only support ANE */ @@ -457,7 +459,7 @@ stmmac_get_pauseparam(struct net_device *netdev, pause->rx_pause = 0; pause->tx_pause = 0; - if (priv->pcs && priv->hw->mac->pcs_get_adv_lp) { + if (priv->hw->pcs && priv->hw->mac->pcs_get_adv_lp) { struct rgmii_adv adv_lp; pause->autoneg = 1; @@ -487,7 +489,7 @@ stmmac_set_pauseparam(struct net_device *netdev, struct phy_device *phy = priv->phydev; int new_pause = FLOW_OFF; - if (priv->pcs && priv->hw->mac->pcs_get_adv_lp) { + if (priv->hw->pcs && priv->hw->mac->pcs_get_adv_lp) { struct rgmii_adv adv_lp; pause->autoneg = 1; @@ -507,6 +509,7 @@ stmmac_set_pauseparam(struct net_device *netdev, priv->flow_ctrl = new_pause; phy->autoneg = pause->autoneg; + if (phy->autoneg) { if (netif_running(netdev)) return phy_start_aneg(phy); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 6c43d68..6142fce 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -285,8 +285,9 @@ bool stmmac_eee_init(struct stmmac_priv *priv) /* Using PCS we cannot dial with the phy registers at this stage * so we do not support extra feature like EEE. */ - if ((priv->pcs == STMMAC_PCS_RGMII) || (priv->pcs == STMMAC_PCS_TBI) || - (priv->pcs == STMMAC_PCS_RTBI)) + if ((priv->hw->pcs == STMMAC_PCS_RGMII) || + (priv->hw->pcs == STMMAC_PCS_TBI) || + (priv->hw->pcs == STMMAC_PCS_RTBI)) goto out; /* MAC core supports the EEE feature. */ @@ -799,10 +800,10 @@ static void stmmac_check_pcs_mode(struct stmmac_priv *priv) (interface == PHY_INTERFACE_MODE_RGMII_RXID) || (interface == PHY_INTERFACE_MODE_RGMII_TXID)) { pr_debug("STMMAC: PCS RGMII support enable\n"); - priv->pcs = STMMAC_PCS_RGMII; + priv->hw->pcs = STMMAC_PCS_RGMII; } else if (interface == PHY_INTERFACE_MODE_SGMII) { pr_debug("STMMAC: PCS SGMII support enable\n"); - priv->pcs = STMMAC_PCS_SGMII; + priv->hw->pcs = STMMAC_PCS_SGMII; } } } @@ -1714,7 +1715,7 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) priv->hw->dma->rx_watchdog(priv->ioaddr, MAX_DMA_RIWT); } - if (priv->pcs && priv->hw->mac->pcs_ctrl_ane) + if (priv->hw->pcs && priv->hw->mac->pcs_ctrl_ane) priv->hw->mac->pcs_ctrl_ane(priv->hw, 1, 0, 0); /* set TX ring length */ @@ -1748,8 +1749,9 @@ static int stmmac_open(struct net_device *dev) stmmac_check_ether_addr(priv); - if (priv->pcs != STMMAC_PCS_RGMII && priv->pcs != STMMAC_PCS_TBI && - priv->pcs != STMMAC_PCS_RTBI) { + if (priv->hw->pcs != STMMAC_PCS_RGMII && + priv->hw->pcs != STMMAC_PCS_TBI && + priv->hw->pcs != STMMAC_PCS_RTBI) { ret = stmmac_init_phy(dev); if (ret) { pr_err("%s: Cannot attach to PHY (error: %d)\n", @@ -2811,7 +2813,7 @@ static irqreturn_t stmmac_interrupt(int irq, void *dev_id) } /* PCS link status */ - if (priv->pcs) { + if (priv->hw->pcs) { if (priv->xstats.pcs_link) netif_carrier_on(dev); else @@ -3138,6 +3140,7 @@ static int stmmac_hw_init(struct stmmac_priv *priv) */ priv->plat->enh_desc = priv->dma_cap.enh_desc; priv->plat->pmt = priv->dma_cap.pmt_remote_wake_up; + priv->hw->pmt = priv->plat->pmt; /* TXCOE doesn't work in thresh DMA mode */ if (priv->plat->force_thresh_dma_mode) @@ -3333,8 +3336,9 @@ int stmmac_dvr_probe(struct device *device, stmmac_check_pcs_mode(priv); - if (priv->pcs != STMMAC_PCS_RGMII && priv->pcs != STMMAC_PCS_TBI && - priv->pcs != STMMAC_PCS_RTBI) { + if (priv->hw->pcs != STMMAC_PCS_RGMII && + priv->hw->pcs != STMMAC_PCS_TBI && + priv->hw->pcs != STMMAC_PCS_RTBI) { /* MDIO bus Registration */ ret = stmmac_mdio_register(ndev); if (ret < 0) { @@ -3384,8 +3388,9 @@ int stmmac_dvr_remove(struct device *dev) reset_control_assert(priv->stmmac_rst); clk_disable_unprepare(priv->pclk); clk_disable_unprepare(priv->stmmac_clk); - if (priv->pcs != STMMAC_PCS_RGMII && priv->pcs != STMMAC_PCS_TBI && - priv->pcs != STMMAC_PCS_RTBI) + if (priv->hw->pcs != STMMAC_PCS_RGMII && + priv->hw->pcs != STMMAC_PCS_TBI && + priv->hw->pcs != STMMAC_PCS_RTBI) stmmac_mdio_unregister(ndev); free_netdev(ndev); -- cgit v0.10.2 From 02e57b9d7c8ce9e403f15f48fb91dd6549aaf465 Mon Sep 17 00:00:00 2001 From: Giuseppe CAVALLARO Date: Fri, 24 Jun 2016 15:16:26 +0200 Subject: drivers: net: stmmac: add port selection programming In case of SGMII more, for example when a MAC2MAC connection is needed, the port selection bits (inside the MAC configuration registers) have to be programmed according to the link selected. So the patch adds a new DT parameter to pass the port selection and to programmed related PCS and CORE to use it. Signed-off-by: Giuseppe Cavallaro Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/stmmac.txt b/Documentation/devicetree/bindings/net/stmmac.txt index 95816c5..41b49e6 100644 --- a/Documentation/devicetree/bindings/net/stmmac.txt +++ b/Documentation/devicetree/bindings/net/stmmac.txt @@ -47,6 +47,9 @@ Optional properties: supported by this device instance - snps,perfect-filter-entries: Number of perfect filter entries supported by this device instance +- snps,ps-speed: port selection speed that can be passed to the core when + PCS is supported. For example, this is used in case of SGMII + and MAC2MAC connection. - AXI BUS Mode parameters: below the list of all the parameters to program the AXI register inside the DMA module: - snps,lpi_en: enable Low Power Interface diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 51077a8..2533b91 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -531,6 +531,7 @@ struct mac_device_info { unsigned int rx_csum; unsigned int pcs; unsigned int pmt; + unsigned int ps; }; struct mac_device_info *dwmac1000_setup(void __iomem *ioaddr, int mcbins, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index 0d31f2f..cbefe9e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -46,6 +46,21 @@ static void dwmac1000_core_init(struct mac_device_info *hw, int mtu) if (mtu > 2000) value |= GMAC_CONTROL_JE; + if (hw->ps) { + value |= GMAC_CONTROL_TE; + + if (hw->ps == SPEED_1000) { + value &= ~GMAC_CONTROL_PS; + } else { + value |= GMAC_CONTROL_PS; + + if (hw->ps == SPEED_10) + value &= ~GMAC_CONTROL_FES; + else + value |= GMAC_CONTROL_FES; + } + } + writel(value, ioaddr + GMAC_CONTROL); /* Mask GMAC interrupts */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 747f3cf..df5580d 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -32,6 +32,21 @@ static void dwmac4_core_init(struct mac_device_info *hw, int mtu) if (mtu > 2000) value |= GMAC_CONFIG_JE; + if (hw->ps) { + value |= GMAC_CONFIG_TE; + + if (hw->ps == SPEED_1000) { + value &= ~GMAC_CONFIG_PS; + } else { + value |= GMAC_CONFIG_PS; + + if (hw->ps == SPEED_10) + value &= ~GMAC_CONFIG_FES; + else + value |= GMAC_CONFIG_FES; + } + } + writel(value, ioaddr + GMAC_CONFIG); /* Mask GMAC interrupts */ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index da2d9b5..1e06173 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -380,7 +380,8 @@ static int stmmac_ethtool_setsettings(struct net_device *dev, spin_lock(&priv->lock); if (priv->hw->mac->pcs_ctrl_ane) - priv->hw->mac->pcs_ctrl_ane(priv->ioaddr, 1, 0, 0); + priv->hw->mac->pcs_ctrl_ane(priv->ioaddr, 1, + priv->hw->ps, 0); spin_unlock(&priv->lock); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 6142fce..aab777c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1666,6 +1666,19 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) if (priv->plat->bus_setup) priv->plat->bus_setup(priv->ioaddr); + /* PS and related bits will be programmed according to the speed */ + if (priv->hw->pcs) { + int speed = priv->plat->mac_port_sel_speed; + + if ((speed == SPEED_10) || (speed == SPEED_100) || + (speed == SPEED_1000)) { + priv->hw->ps = speed; + } else { + dev_warn(priv->device, "invalid port speed\n"); + priv->hw->ps = 0; + } + } + /* Initialize the MAC Core */ priv->hw->mac->core_init(priv->hw, dev->mtu); @@ -1716,7 +1729,7 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) } if (priv->hw->pcs && priv->hw->mac->pcs_ctrl_ane) - priv->hw->mac->pcs_ctrl_ane(priv->hw, 1, 0, 0); + priv->hw->mac->pcs_ctrl_ane(priv->hw, 1, priv->hw->ps, 0); /* set TX ring length */ if (priv->hw->dma->set_tx_ring_len) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index a96714d..f7dfc0a 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -319,6 +319,8 @@ stmmac_probe_config_dt(struct platform_device *pdev, const char **mac) pr_warn("force_sf_dma_mode is ignored if force_thresh_dma_mode is set."); } + of_property_read_u32(np, "snps,ps-speed", &plat->mac_port_sel_speed); + plat->axi = stmmac_axi_setup(pdev); return plat; diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index 0507dbf..705840e 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -141,5 +141,6 @@ struct plat_stmmacenet_data { struct stmmac_axi *axi; int has_gmac4; bool tso_en; + int mac_port_sel_speed; }; #endif -- cgit v0.10.2 From 1f95ba000c92cafccea007129d40532a4f35b1a6 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:41 +0300 Subject: drivers: net: cpsw: fix suspend when all ethX devices are down The cpsw_suspend() could trigger L3 error and CPSW will stop functioning if System enters suspend when all ethX net-devices are down - in this case CPSW could be already suspended by PM runtime, but cpsw_suspend() will try to call soft_reset_slave() unconditionally and access CPSW registers. Hence, fix it by moving soft_reset_slave() from cpsw_suspend() to cpsw_slave_stop(). This way slave ports will be reset when CPSW is active and will be in proper state during Suspend. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 9c924f1..6e0f0c6 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1243,6 +1243,7 @@ static void cpsw_slave_stop(struct cpsw_slave *slave, struct cpsw_priv *priv) slave->phy = NULL; cpsw_ale_control_set(priv->ale, slave_port, ALE_PORT_STATE, ALE_PORT_STATE_DISABLE); + soft_reset_slave(slave); } static int cpsw_ndo_open(struct net_device *ndev) @@ -2548,12 +2549,10 @@ static int cpsw_suspend(struct device *dev) for (i = 0; i < priv->data.slaves; i++) { if (netif_running(priv->slaves[i].ndev)) cpsw_ndo_stop(priv->slaves[i].ndev); - soft_reset_slave(priv->slaves + i); } } else { if (netif_running(ndev)) cpsw_ndo_stop(ndev); - for_each_slave(priv, soft_reset_slave); } pm_runtime_put_sync(&pdev->dev); -- cgit v0.10.2 From 108a653730e4c654bd0815276276e59ea0d64578 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:42 +0300 Subject: drivers: net: cpsw: check return code from pm runtime calls Add missed check of return code from PM runtime get() calls. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 6e0f0c6..4e3d519 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1252,7 +1252,11 @@ static int cpsw_ndo_open(struct net_device *ndev) int i, ret; u32 reg; - pm_runtime_get_sync(&priv->pdev->dev); + ret = pm_runtime_get_sync(&priv->pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&priv->pdev->dev); + return ret; + } if (!cpsw_common_res_usage_state(priv)) cpsw_intr_disable(priv); @@ -2312,7 +2316,11 @@ static int cpsw_probe(struct platform_device *pdev) /* Need to enable clocks with runtime PM api to access module * registers */ - pm_runtime_get_sync(&pdev->dev); + ret = pm_runtime_get_sync(&pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&pdev->dev); + goto clean_runtime_disable_ret; + } priv->version = readl(&priv->regs->id_ver); pm_runtime_put_sync(&pdev->dev); -- cgit v0.10.2 From 74556f516702331444b3c2d900a6a1d1fc9fc4cd Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:43 +0300 Subject: drivers: net: cpsw: remove pm runtime calls from suspend callbacks PM runtime is properly handled in cpsw_ndo_open/stop(), as result it isn't required to duplicate these calls in .suspend()/.resume() callbacks. Moreover, it might cause unnecessary RPM resume of CPSW during System suspend in the case it's already suspended because all ethX interfaces are down already, before System suspend started. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 4e3d519..b6c508e 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -2563,8 +2563,6 @@ static int cpsw_suspend(struct device *dev) cpsw_ndo_stop(ndev); } - pm_runtime_put_sync(&pdev->dev); - /* Select sleep pin state */ pinctrl_pm_select_sleep_state(&pdev->dev); @@ -2577,8 +2575,6 @@ static int cpsw_resume(struct device *dev) struct net_device *ndev = platform_get_drvdata(pdev); struct cpsw_priv *priv = netdev_priv(ndev); - pm_runtime_get_sync(&pdev->dev); - /* Select default pin state */ pinctrl_pm_select_default_state(&pdev->dev); -- cgit v0.10.2 From 7898b1daf055537a8f661828272defecdee49f27 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:44 +0300 Subject: drivers: net: cpsw: ethtool: fix accessing to suspended device The CPSW might be suspended by RPM if all ethX interfaces are down, but it still could be accesible through ethtool interfce. In this case ethtool operations, requiring registers access, will cause L3 errors and CPSW crash. ethtool callbcaks which need to access CPSW registers now: .set_coalesce(), .get_ethtool_stats(), .set_pauseparam(), .get_regs() Hence, fix it by adding .begin()/.complete() ethtool callbacks, which will be called before/after each ethtool operation runs, and do CPSW RPM handling in these callbacks. That way CPSW will be active while handling ethtool requests. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index b6c508e..2b21cad 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1905,10 +1905,33 @@ static int cpsw_set_pauseparam(struct net_device *ndev, priv->tx_pause = pause->tx_pause ? true : false; for_each_slave(priv, _cpsw_adjust_link, priv, &link); - return 0; } +static int cpsw_ethtool_op_begin(struct net_device *ndev) +{ + struct cpsw_priv *priv = netdev_priv(ndev); + int ret; + + ret = pm_runtime_get_sync(&priv->pdev->dev); + if (ret < 0) { + cpsw_err(priv, drv, "ethtool begin failed %d\n", ret); + pm_runtime_put_noidle(&priv->pdev->dev); + } + + return ret; +} + +static void cpsw_ethtool_op_complete(struct net_device *ndev) +{ + struct cpsw_priv *priv = netdev_priv(ndev); + int ret; + + ret = pm_runtime_put(&priv->pdev->dev); + if (ret < 0) + cpsw_err(priv, drv, "ethtool complete failed %d\n", ret); +} + static const struct ethtool_ops cpsw_ethtool_ops = { .get_drvinfo = cpsw_get_drvinfo, .get_msglevel = cpsw_get_msglevel, @@ -1928,6 +1951,8 @@ static const struct ethtool_ops cpsw_ethtool_ops = { .set_wol = cpsw_set_wol, .get_regs_len = cpsw_get_regs_len, .get_regs = cpsw_get_regs, + .begin = cpsw_ethtool_op_begin, + .complete = cpsw_ethtool_op_complete, }; static void cpsw_slave_init(struct cpsw_slave *slave, struct cpsw_priv *priv, -- cgit v0.10.2 From a6c5d14f5136e4919ace42bdbf392d2f6e68d4da Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:45 +0300 Subject: drivers: net: cpsw: ndev: fix accessing to suspended device The CPSW might be suspended by RPM if all ethX interfaces are down, but it still could be accesible through net_device_ops interfce. In this case net_device_ops operations requiring registers access will cause L3 errors and CPSW crash. Hence, fix it by adding RPM get/put calls in net_device_ops callbacks which need to access CPSW registers: .ndo_set_mac_address(), .ndo_vlan_rx_add_vid(), .ndo_vlan_rx_kill_vid(). Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 2b21cad..6d0c5a0 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -1614,10 +1614,17 @@ static int cpsw_ndo_set_mac_address(struct net_device *ndev, void *p) struct sockaddr *addr = (struct sockaddr *)p; int flags = 0; u16 vid = 0; + int ret; if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; + ret = pm_runtime_get_sync(&priv->pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&priv->pdev->dev); + return ret; + } + if (priv->data.dual_emac) { vid = priv->slaves[priv->emac_port].port_vlan; flags = ALE_VLAN; @@ -1632,6 +1639,8 @@ static int cpsw_ndo_set_mac_address(struct net_device *ndev, void *p) memcpy(ndev->dev_addr, priv->mac_addr, ETH_ALEN); for_each_slave(priv, cpsw_set_slave_mac, priv); + pm_runtime_put(&priv->pdev->dev); + return 0; } @@ -1696,10 +1705,17 @@ static int cpsw_ndo_vlan_rx_add_vid(struct net_device *ndev, __be16 proto, u16 vid) { struct cpsw_priv *priv = netdev_priv(ndev); + int ret; if (vid == priv->data.default_vlan) return 0; + ret = pm_runtime_get_sync(&priv->pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&priv->pdev->dev); + return ret; + } + if (priv->data.dual_emac) { /* In dual EMAC, reserved VLAN id should not be used for * creating VLAN interfaces as this can break the dual @@ -1714,7 +1730,10 @@ static int cpsw_ndo_vlan_rx_add_vid(struct net_device *ndev, } dev_info(priv->dev, "Adding vlanid %d to vlan filter\n", vid); - return cpsw_add_vlan_ale_entry(priv, vid); + ret = cpsw_add_vlan_ale_entry(priv, vid); + + pm_runtime_put(&priv->pdev->dev); + return ret; } static int cpsw_ndo_vlan_rx_kill_vid(struct net_device *ndev, @@ -1726,6 +1745,12 @@ static int cpsw_ndo_vlan_rx_kill_vid(struct net_device *ndev, if (vid == priv->data.default_vlan) return 0; + ret = pm_runtime_get_sync(&priv->pdev->dev); + if (ret < 0) { + pm_runtime_put_noidle(&priv->pdev->dev); + return ret; + } + if (priv->data.dual_emac) { int i; @@ -1745,8 +1770,10 @@ static int cpsw_ndo_vlan_rx_kill_vid(struct net_device *ndev, if (ret != 0) return ret; - return cpsw_ale_del_mcast(priv->ale, priv->ndev->broadcast, - 0, ALE_VLAN, vid); + ret = cpsw_ale_del_mcast(priv->ale, priv->ndev->broadcast, + 0, ALE_VLAN, vid); + pm_runtime_put(&priv->pdev->dev); + return ret; } static const struct net_device_ops cpsw_netdev_ops = { -- cgit v0.10.2 From 909892a647fe16415c743cdb779257cb1f747cb4 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:46 +0300 Subject: drivers: net: davinci_mdio: do pm runtime initialization later in probe Do PM runtime initialization later in probe - this allows to simplify error handling a bit. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index 4e7c9b9..2e19dd1 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -356,14 +356,10 @@ static int davinci_mdio_probe(struct platform_device *pdev) data->bus->parent = dev; data->bus->priv = data; - pm_runtime_enable(&pdev->dev); - pm_runtime_get_sync(&pdev->dev); data->clk = devm_clk_get(dev, "fck"); if (IS_ERR(data->clk)) { dev_err(dev, "failed to get device clock\n"); - ret = PTR_ERR(data->clk); - data->clk = NULL; - goto bail_out; + return PTR_ERR(data->clk); } dev_set_drvdata(dev, data); @@ -372,10 +368,11 @@ static int davinci_mdio_probe(struct platform_device *pdev) res = platform_get_resource(pdev, IORESOURCE_MEM, 0); data->regs = devm_ioremap_resource(dev, res); - if (IS_ERR(data->regs)) { - ret = PTR_ERR(data->regs); - goto bail_out; - } + if (IS_ERR(data->regs)) + return PTR_ERR(data->regs); + + pm_runtime_enable(&pdev->dev); + pm_runtime_get_sync(&pdev->dev); /* register the mii bus * Create PHYs from DT only in case if PHY child nodes are explicitly -- cgit v0.10.2 From 3f655909e1246077501e6ecb50dc95c795760e35 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:47 +0300 Subject: drivers: net: davinci_mdio: remove pm runtime calls from suspend callbacks PM runtime is disabled when Davinci MDIO .suspend_late() and .resume_early() callbacks are called. As result, any PM runtime calls here will be just a nop and can be removed. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index 2e19dd1..291c42e 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -436,7 +436,6 @@ static int davinci_mdio_suspend(struct device *dev) data->suspended = true; spin_unlock(&data->lock); - pm_runtime_put_sync(data->dev); /* Select sleep pin state */ pinctrl_pm_select_sleep_state(dev); @@ -451,8 +450,6 @@ static int davinci_mdio_resume(struct device *dev) /* Select default pin state */ pinctrl_pm_select_default_state(dev); - pm_runtime_get_sync(data->dev); - spin_lock(&data->lock); /* restart the scan state machine */ __davinci_mdio_reset(data); -- cgit v0.10.2 From a01d7baa1f2e291b1b3d82d57ae41d2fd9d438c3 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:48 +0300 Subject: drivers: net: davinci_mdio: drop suspended and lock fields from mdio_data It's not expected Davinci MDIO to be accessible after its suspend callbacks have been called: - all consumers of Davinci MDIO will stop/disconnect phys at Device suspend stage; - all phys are expected to be suspned already by PHY/MDIO core; - MDIO locking is done by MDIO Bus code. Hence, it's safe to drop "suspended" and "lock" fields from mdio_data. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index 291c42e..b6d0059 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -90,11 +90,9 @@ static const struct mdio_platform_data default_pdata = { struct davinci_mdio_data { struct mdio_platform_data pdata; struct davinci_mdio_regs __iomem *regs; - spinlock_t lock; struct clk *clk; struct device *dev; struct mii_bus *bus; - bool suspended; unsigned long access_time; /* jiffies */ /* Indicates that driver shouldn't modify phy_mask in case * if MDIO bus is registered from DT. @@ -225,13 +223,6 @@ static int davinci_mdio_read(struct mii_bus *bus, int phy_id, int phy_reg) if (phy_reg & ~PHY_REG_MASK || phy_id & ~PHY_ID_MASK) return -EINVAL; - spin_lock(&data->lock); - - if (data->suspended) { - spin_unlock(&data->lock); - return -ENODEV; - } - reg = (USERACCESS_GO | USERACCESS_READ | (phy_reg << 21) | (phy_id << 16)); @@ -255,8 +246,6 @@ static int davinci_mdio_read(struct mii_bus *bus, int phy_id, int phy_reg) break; } - spin_unlock(&data->lock); - return ret; } @@ -270,13 +259,6 @@ static int davinci_mdio_write(struct mii_bus *bus, int phy_id, if (phy_reg & ~PHY_REG_MASK || phy_id & ~PHY_ID_MASK) return -EINVAL; - spin_lock(&data->lock); - - if (data->suspended) { - spin_unlock(&data->lock); - return -ENODEV; - } - reg = (USERACCESS_GO | USERACCESS_WRITE | (phy_reg << 21) | (phy_id << 16) | (phy_data & USERACCESS_DATA)); @@ -295,8 +277,6 @@ static int davinci_mdio_write(struct mii_bus *bus, int phy_id, break; } - spin_unlock(&data->lock); - return 0; } @@ -364,7 +344,6 @@ static int davinci_mdio_probe(struct platform_device *pdev) dev_set_drvdata(dev, data); data->dev = dev; - spin_lock_init(&data->lock); res = platform_get_resource(pdev, IORESOURCE_MEM, 0); data->regs = devm_ioremap_resource(dev, res); @@ -426,17 +405,12 @@ static int davinci_mdio_suspend(struct device *dev) struct davinci_mdio_data *data = dev_get_drvdata(dev); u32 ctrl; - spin_lock(&data->lock); - /* shutdown the scan state machine */ ctrl = __raw_readl(&data->regs->control); ctrl &= ~CONTROL_ENABLE; __raw_writel(ctrl, &data->regs->control); wait_for_idle(data); - data->suspended = true; - spin_unlock(&data->lock); - /* Select sleep pin state */ pinctrl_pm_select_sleep_state(dev); @@ -450,13 +424,9 @@ static int davinci_mdio_resume(struct device *dev) /* Select default pin state */ pinctrl_pm_select_default_state(dev); - spin_lock(&data->lock); /* restart the scan state machine */ __davinci_mdio_reset(data); - data->suspended = false; - spin_unlock(&data->lock); - return 0; } #endif -- cgit v0.10.2 From 28f0ccb9958f54d9f4e7ebcf09b400304d8ba7cb Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:49 +0300 Subject: drivers: net: davinci_mdio: split reset function on init_clk and enable The Davinci MDIO MDIO_CONTROL.CLKDIV can be calculated only once during probe, hence split __davinci_mdio_reset() on davinci_mdio_init_clk() and davinci_mdio_enable(). Initialize and save CLKDIV in .probe(). Then just use saved value. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index b6d0059..b206fd3 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -98,9 +98,10 @@ struct davinci_mdio_data { * if MDIO bus is registered from DT. */ bool skip_scan; + u32 clk_div; }; -static void __davinci_mdio_reset(struct davinci_mdio_data *data) +static void davinci_mdio_init_clk(struct davinci_mdio_data *data) { u32 mdio_in, div, mdio_out_khz, access_time; @@ -109,9 +110,7 @@ static void __davinci_mdio_reset(struct davinci_mdio_data *data) if (div > CONTROL_MAX_DIV) div = CONTROL_MAX_DIV; - /* set enable and clock divider */ - __raw_writel(div | CONTROL_ENABLE, &data->regs->control); - + data->clk_div = div; /* * One mdio transaction consists of: * 32 bits of preamble @@ -132,12 +131,18 @@ static void __davinci_mdio_reset(struct davinci_mdio_data *data) data->access_time = 1; } +static void davinci_mdio_enable(struct davinci_mdio_data *data) +{ + /* set enable and clock divider */ + __raw_writel(data->clk_div | CONTROL_ENABLE, &data->regs->control); +} + static int davinci_mdio_reset(struct mii_bus *bus) { struct davinci_mdio_data *data = bus->priv; u32 phy_mask, ver; - __davinci_mdio_reset(data); + davinci_mdio_enable(data); /* wait for scan logic to settle */ msleep(PHY_MAX_ADDR * data->access_time); @@ -188,7 +193,7 @@ static inline int wait_for_user_access(struct davinci_mdio_data *data) * operation */ dev_warn(data->dev, "resetting idled controller\n"); - __davinci_mdio_reset(data); + davinci_mdio_enable(data); return -EAGAIN; } @@ -350,6 +355,8 @@ static int davinci_mdio_probe(struct platform_device *pdev) if (IS_ERR(data->regs)) return PTR_ERR(data->regs); + davinci_mdio_init_clk(data); + pm_runtime_enable(&pdev->dev); pm_runtime_get_sync(&pdev->dev); @@ -425,7 +432,7 @@ static int davinci_mdio_resume(struct device *dev) pinctrl_pm_select_default_state(dev); /* restart the scan state machine */ - __davinci_mdio_reset(data); + davinci_mdio_enable(data); return 0; } -- cgit v0.10.2 From 651652aace74207a41802a911963a8a4ced37da2 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:50 +0300 Subject: drivers: net: davinci_mdio: add pm runtime callbacks Add PM runtime .runtime_suspend()/.runtime_resume() callbacks and perform Davinci MDIO enabling/disabling from these callbacks. This allows to reuse pm_runtime_force_suspend/resume() APIs during System suspend and required for further implementation of PM runtime autosuspend. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index b206fd3..13f5080 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -406,8 +406,8 @@ static int davinci_mdio_remove(struct platform_device *pdev) return 0; } -#ifdef CONFIG_PM_SLEEP -static int davinci_mdio_suspend(struct device *dev) +#ifdef CONFIG_PM +static int davinci_mdio_runtime_suspend(struct device *dev) { struct davinci_mdio_data *data = dev_get_drvdata(dev); u32 ctrl; @@ -418,6 +418,28 @@ static int davinci_mdio_suspend(struct device *dev) __raw_writel(ctrl, &data->regs->control); wait_for_idle(data); + return 0; +} + +static int davinci_mdio_runtime_resume(struct device *dev) +{ + struct davinci_mdio_data *data = dev_get_drvdata(dev); + + davinci_mdio_enable(data); + return 0; +} +#endif + +#ifdef CONFIG_PM_SLEEP +static int davinci_mdio_suspend(struct device *dev) +{ + struct davinci_mdio_data *data = dev_get_drvdata(dev); + int ret = 0; + + ret = pm_runtime_force_suspend(dev); + if (ret < 0) + return ret; + /* Select sleep pin state */ pinctrl_pm_select_sleep_state(dev); @@ -431,14 +453,15 @@ static int davinci_mdio_resume(struct device *dev) /* Select default pin state */ pinctrl_pm_select_default_state(dev); - /* restart the scan state machine */ - davinci_mdio_enable(data); + pm_runtime_force_resume(dev); return 0; } #endif static const struct dev_pm_ops davinci_mdio_pm_ops = { + SET_RUNTIME_PM_OPS(davinci_mdio_runtime_suspend, + davinci_mdio_runtime_resume, NULL) SET_LATE_SYSTEM_SLEEP_PM_OPS(davinci_mdio_suspend, davinci_mdio_resume) }; -- cgit v0.10.2 From 8ea63bbaabb83adf54609c7aa6cb0ab504023a99 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:51 +0300 Subject: drivers: net: davinci_mdio: implement pm runtime auto mode Davinci MDIO is always used as slave device which services read/write requests from MDIO/PHY core. It doesn't use IRQ also. As result, It's possible to relax PM runtime constraints for Davinci MDIO and enable it on demand, instead of powering it during probe and powering off during removal. Hence, implement PM runtime autosuspend for Davinci MDIO, but keep it disabled by default, because Davinci MDIO is integrated in big set of TI devices and not all of them expected to work corectly with RPM autosuspend enabled: - expected to work on SoCs where MDIO is part of TI CPSW (cpsw.c DRA7/am57x, am437x, am335x, dm814x) - not verified on Keystone 2 and other SoCs where MDIO is used with TI EMAC IP (davinci_emac.c: dm6467-emac, am3517-emac, dm816-emac). Davinci MDIO RPM autosuspend can be enabled through sysfs: echo 100 > /sys/devices/../48484000.ethernet/48485000.mdio/power/autosuspend_delay_ms Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index 13f5080..ce3ec42 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -93,6 +93,7 @@ struct davinci_mdio_data { struct clk *clk; struct device *dev; struct mii_bus *bus; + bool active_in_suspend; unsigned long access_time; /* jiffies */ /* Indicates that driver shouldn't modify phy_mask in case * if MDIO bus is registered from DT. @@ -141,8 +142,13 @@ static int davinci_mdio_reset(struct mii_bus *bus) { struct davinci_mdio_data *data = bus->priv; u32 phy_mask, ver; + int ret; - davinci_mdio_enable(data); + ret = pm_runtime_get_sync(data->dev); + if (ret < 0) { + pm_runtime_put_noidle(data->dev); + return ret; + } /* wait for scan logic to settle */ msleep(PHY_MAX_ADDR * data->access_time); @@ -153,7 +159,7 @@ static int davinci_mdio_reset(struct mii_bus *bus) (ver >> 8) & 0xff, ver & 0xff); if (data->skip_scan) - return 0; + goto done; /* get phy mask from the alive register */ phy_mask = __raw_readl(&data->regs->alive); @@ -168,6 +174,10 @@ static int davinci_mdio_reset(struct mii_bus *bus) } data->bus->phy_mask = phy_mask; +done: + pm_runtime_mark_last_busy(data->dev); + pm_runtime_put_autosuspend(data->dev); + return 0; } @@ -228,6 +238,12 @@ static int davinci_mdio_read(struct mii_bus *bus, int phy_id, int phy_reg) if (phy_reg & ~PHY_REG_MASK || phy_id & ~PHY_ID_MASK) return -EINVAL; + ret = pm_runtime_get_sync(data->dev); + if (ret < 0) { + pm_runtime_put_noidle(data->dev); + return ret; + } + reg = (USERACCESS_GO | USERACCESS_READ | (phy_reg << 21) | (phy_id << 16)); @@ -251,6 +267,8 @@ static int davinci_mdio_read(struct mii_bus *bus, int phy_id, int phy_reg) break; } + pm_runtime_mark_last_busy(data->dev); + pm_runtime_put_autosuspend(data->dev); return ret; } @@ -264,6 +282,12 @@ static int davinci_mdio_write(struct mii_bus *bus, int phy_id, if (phy_reg & ~PHY_REG_MASK || phy_id & ~PHY_ID_MASK) return -EINVAL; + ret = pm_runtime_get_sync(data->dev); + if (ret < 0) { + pm_runtime_put_noidle(data->dev); + return ret; + } + reg = (USERACCESS_GO | USERACCESS_WRITE | (phy_reg << 21) | (phy_id << 16) | (phy_data & USERACCESS_DATA)); @@ -282,7 +306,10 @@ static int davinci_mdio_write(struct mii_bus *bus, int phy_id, break; } - return 0; + pm_runtime_mark_last_busy(data->dev); + pm_runtime_put_autosuspend(data->dev); + + return ret; } #if IS_ENABLED(CONFIG_OF) @@ -357,8 +384,9 @@ static int davinci_mdio_probe(struct platform_device *pdev) davinci_mdio_init_clk(data); + pm_runtime_set_autosuspend_delay(&pdev->dev, -1); + pm_runtime_use_autosuspend(&pdev->dev); pm_runtime_enable(&pdev->dev); - pm_runtime_get_sync(&pdev->dev); /* register the mii bus * Create PHYs from DT only in case if PHY child nodes are explicitly @@ -387,9 +415,8 @@ static int davinci_mdio_probe(struct platform_device *pdev) return 0; bail_out: - pm_runtime_put_sync(&pdev->dev); + pm_runtime_dont_use_autosuspend(&pdev->dev); pm_runtime_disable(&pdev->dev); - return ret; } @@ -400,7 +427,7 @@ static int davinci_mdio_remove(struct platform_device *pdev) if (data->bus) mdiobus_unregister(data->bus); - pm_runtime_put_sync(&pdev->dev); + pm_runtime_dont_use_autosuspend(&pdev->dev); pm_runtime_disable(&pdev->dev); return 0; @@ -436,7 +463,9 @@ static int davinci_mdio_suspend(struct device *dev) struct davinci_mdio_data *data = dev_get_drvdata(dev); int ret = 0; - ret = pm_runtime_force_suspend(dev); + data->active_in_suspend = !pm_runtime_status_suspended(dev); + if (data->active_in_suspend) + ret = pm_runtime_force_suspend(dev); if (ret < 0) return ret; @@ -453,7 +482,8 @@ static int davinci_mdio_resume(struct device *dev) /* Select default pin state */ pinctrl_pm_select_default_state(dev); - pm_runtime_force_resume(dev); + if (data->active_in_suspend) + pm_runtime_force_resume(dev); return 0; } -- cgit v0.10.2 From beb1a04619f6bcf0e0452780b4c3eaab995e0b43 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:52 +0300 Subject: net: davinci_mdio: document missed "ti, am4372-mdio" compat string Document missed "ti,am4372-mdio" compat string used for TI am437x SoC (am4372.dtsi). Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/davinci-mdio.txt b/Documentation/devicetree/bindings/net/davinci-mdio.txt index 0369e25..b6a4f48 100644 --- a/Documentation/devicetree/bindings/net/davinci-mdio.txt +++ b/Documentation/devicetree/bindings/net/davinci-mdio.txt @@ -2,7 +2,9 @@ TI SoC Davinci/Keystone2 MDIO Controller Device Tree Bindings --------------------------------------------------- Required properties: -- compatible : Should be "ti,davinci_mdio" or "ti,keystone_mdio" +- compatible : Should be "ti,davinci_mdio" + and "ti,keystone_mdio" for Keystone 2 SoCs + and "ti,am4372-mdio" for am472x SoC - reg : physical base address and size of the davinci mdio registers map - bus_freq : Mdio Bus frequency -- cgit v0.10.2 From 22899eca586b862d5761f1b4795a3951fc0e502f Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:53 +0300 Subject: net: davinci_mdio: introduce "ti,cpsw-mdio" compat string Introduce "ti,cpsw-mdio" compatible string for Davinci MDIO, because it's required to distinguish the case when MDIO is part of TI CPSW to enable features supported by TI CPSW (for example, enable PM management). Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/davinci-mdio.txt b/Documentation/devicetree/bindings/net/davinci-mdio.txt index b6a4f48..621156c 100644 --- a/Documentation/devicetree/bindings/net/davinci-mdio.txt +++ b/Documentation/devicetree/bindings/net/davinci-mdio.txt @@ -4,6 +4,7 @@ TI SoC Davinci/Keystone2 MDIO Controller Device Tree Bindings Required properties: - compatible : Should be "ti,davinci_mdio" and "ti,keystone_mdio" for Keystone 2 SoCs + and "ti,cpsw-mdio" for am335x, am472x, am57xx/dra7, dm814x SoCs and "ti,am4372-mdio" for am472x SoC - reg : physical base address and size of the davinci mdio registers map -- cgit v0.10.2 From 9eae9c7d087592fce04ea7b9b8b486b84b668208 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:54 +0300 Subject: drivers: net: davinci_mdio: enable pm runtime auto for ti cpsw-mdio Use "ti,cpsw-mdio" to enable PM runtime auto-suspend on supported platforms, where MDIO is implemented as part of TI CPSW. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index ce3ec42..33df340 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -53,6 +53,10 @@ #define DEF_OUT_FREQ 2200000 /* 2.2 MHz */ +struct davinci_mdio_of_param { + int autosuspend_delay_ms; +}; + struct davinci_mdio_regs { u32 version; u32 control; @@ -332,6 +336,19 @@ static int davinci_mdio_probe_dt(struct mdio_platform_data *data, } #endif +#if IS_ENABLED(CONFIG_OF) +static const struct davinci_mdio_of_param of_cpsw_mdio_data = { + .autosuspend_delay_ms = 100, +}; + +static const struct of_device_id davinci_mdio_of_mtable[] = { + { .compatible = "ti,davinci_mdio", }, + { .compatible = "ti,cpsw-mdio", .data = &of_cpsw_mdio_data}, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, davinci_mdio_of_mtable); +#endif + static int davinci_mdio_probe(struct platform_device *pdev) { struct mdio_platform_data *pdata = dev_get_platdata(&pdev->dev); @@ -340,6 +357,7 @@ static int davinci_mdio_probe(struct platform_device *pdev) struct resource *res; struct phy_device *phy; int ret, addr; + int autosuspend_delay_ms = -1; data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL); if (!data) @@ -352,9 +370,22 @@ static int davinci_mdio_probe(struct platform_device *pdev) } if (dev->of_node) { - if (davinci_mdio_probe_dt(&data->pdata, pdev)) - data->pdata = default_pdata; + const struct of_device_id *of_id; + + ret = davinci_mdio_probe_dt(&data->pdata, pdev); + if (ret) + return ret; snprintf(data->bus->id, MII_BUS_ID_SIZE, "%s", pdev->name); + + of_id = of_match_device(davinci_mdio_of_mtable, &pdev->dev); + if (of_id) { + const struct davinci_mdio_of_param *of_mdio_data; + + of_mdio_data = of_id->data; + if (of_mdio_data) + autosuspend_delay_ms = + of_mdio_data->autosuspend_delay_ms; + } } else { data->pdata = pdata ? (*pdata) : default_pdata; snprintf(data->bus->id, MII_BUS_ID_SIZE, "%s-%x", @@ -384,7 +415,7 @@ static int davinci_mdio_probe(struct platform_device *pdev) davinci_mdio_init_clk(data); - pm_runtime_set_autosuspend_delay(&pdev->dev, -1); + pm_runtime_set_autosuspend_delay(&pdev->dev, autosuspend_delay_ms); pm_runtime_use_autosuspend(&pdev->dev); pm_runtime_enable(&pdev->dev); @@ -495,14 +526,6 @@ static const struct dev_pm_ops davinci_mdio_pm_ops = { SET_LATE_SYSTEM_SLEEP_PM_OPS(davinci_mdio_suspend, davinci_mdio_resume) }; -#if IS_ENABLED(CONFIG_OF) -static const struct of_device_id davinci_mdio_of_mtable[] = { - { .compatible = "ti,davinci_mdio", }, - { /* sentinel */ }, -}; -MODULE_DEVICE_TABLE(of, davinci_mdio_of_mtable); -#endif - static struct platform_driver davinci_mdio_driver = { .driver = { .name = "davinci_mdio", -- cgit v0.10.2 From 9efd1a6f60b4ed7cf87a1421a5150477dc359b25 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Fri, 24 Jun 2016 21:23:55 +0300 Subject: ARM: dts: am335x/am437x/dra7: use new "ti, cpsw-mdio" compat string Add "ti,cpsw-mdio" for am335x/am437x/dra7 SoCs where MDIO is implemented as part of TI CPSW and, this way, enable PM runtime auto suspend for Davinci MDIO driver on these paltforms. Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/arch/arm/boot/dts/am33xx.dtsi b/arch/arm/boot/dts/am33xx.dtsi index 702126f..7fa2951 100644 --- a/arch/arm/boot/dts/am33xx.dtsi +++ b/arch/arm/boot/dts/am33xx.dtsi @@ -788,7 +788,7 @@ status = "disabled"; davinci_mdio: mdio@4a101000 { - compatible = "ti,davinci_mdio"; + compatible = "ti,cpsw-mdio","ti,davinci_mdio"; #address-cells = <1>; #size-cells = <0>; ti,hwmods = "davinci_mdio"; diff --git a/arch/arm/boot/dts/am4372.dtsi b/arch/arm/boot/dts/am4372.dtsi index a10fa7f..cd81ecf 100644 --- a/arch/arm/boot/dts/am4372.dtsi +++ b/arch/arm/boot/dts/am4372.dtsi @@ -635,7 +635,7 @@ syscon = <&scm_conf>; davinci_mdio: mdio@4a101000 { - compatible = "ti,am4372-mdio","ti,davinci_mdio"; + compatible = "ti,am4372-mdio","ti,cpsw-mdio","ti,davinci_mdio"; reg = <0x4a101000 0x100>; #address-cells = <1>; #size-cells = <0>; diff --git a/arch/arm/boot/dts/dra7.dtsi b/arch/arm/boot/dts/dra7.dtsi index b7ddc64..f8b39a5 100644 --- a/arch/arm/boot/dts/dra7.dtsi +++ b/arch/arm/boot/dts/dra7.dtsi @@ -1660,7 +1660,7 @@ status = "disabled"; davinci_mdio: mdio@48485000 { - compatible = "ti,davinci_mdio"; + compatible = "ti,cpsw-mdio","ti,davinci_mdio"; #address-cells = <1>; #size-cells = <0>; ti,hwmods = "davinci_mdio"; -- cgit v0.10.2 From 6f96608ec7948475c28e3f2264a38058d28c4141 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 25 Jun 2016 16:55:12 +0200 Subject: net: ethernet: hix5hd2: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c index b9f2ea5..fed1c0f 100644 --- a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c +++ b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c @@ -218,7 +218,6 @@ struct hix5hd2_priv { struct device *dev; struct net_device *netdev; - struct phy_device *phy; struct device_node *phy_node; phy_interface_t phy_mode; @@ -402,7 +401,7 @@ static int hix5hd2_net_set_mac_address(struct net_device *dev, void *p) static void hix5hd2_adjust_link(struct net_device *dev) { struct hix5hd2_priv *priv = netdev_priv(dev); - struct phy_device *phy = priv->phy; + struct phy_device *phy = dev->phydev; if ((priv->speed != phy->speed) || (priv->duplex != phy->duplex)) { hix5hd2_config_port(dev, phy->speed, phy->duplex); @@ -679,6 +678,7 @@ static void hix5hd2_free_dma_desc_rings(struct hix5hd2_priv *priv) static int hix5hd2_net_open(struct net_device *dev) { struct hix5hd2_priv *priv = netdev_priv(dev); + struct phy_device *phy; int ret; ret = clk_prepare_enable(priv->clk); @@ -687,12 +687,12 @@ static int hix5hd2_net_open(struct net_device *dev) return ret; } - priv->phy = of_phy_connect(dev, priv->phy_node, - &hix5hd2_adjust_link, 0, priv->phy_mode); - if (!priv->phy) + phy = of_phy_connect(dev, priv->phy_node, + &hix5hd2_adjust_link, 0, priv->phy_mode); + if (!phy) return -ENODEV; - phy_start(priv->phy); + phy_start(phy); hix5hd2_hw_init(priv); hix5hd2_rx_refill(priv); @@ -716,9 +716,9 @@ static int hix5hd2_net_close(struct net_device *dev) netif_stop_queue(dev); hix5hd2_free_dma_desc_rings(priv); - if (priv->phy) { - phy_stop(priv->phy); - phy_disconnect(priv->phy); + if (dev->phydev) { + phy_stop(dev->phydev); + phy_disconnect(dev->phydev); } clk_disable_unprepare(priv->clk); @@ -753,23 +753,19 @@ static const struct net_device_ops hix5hd2_netdev_ops = { static int hix5hd2_get_settings(struct net_device *net_dev, struct ethtool_cmd *cmd) { - struct hix5hd2_priv *priv = netdev_priv(net_dev); - - if (!priv->phy) + if (!net_dev->phydev) return -ENODEV; - return phy_ethtool_gset(priv->phy, cmd); + return phy_ethtool_gset(net_dev->phydev, cmd); } static int hix5hd2_set_settings(struct net_device *net_dev, struct ethtool_cmd *cmd) { - struct hix5hd2_priv *priv = netdev_priv(net_dev); - - if (!priv->phy) + if (!net_dev->phydev) return -ENODEV; - return phy_ethtool_sset(priv->phy, cmd); + return phy_ethtool_sset(net_dev->phydev, cmd); } static struct ethtool_ops hix5hd2_ethtools_ops = { -- cgit v0.10.2 From 802fe79e15ed2c5444a3da60792cc6ddb93bb2b3 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 25 Jun 2016 16:55:13 +0200 Subject: net: ethernet: hix5hd2: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c index fed1c0f..275618b 100644 --- a/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c +++ b/drivers/net/ethernet/hisilicon/hix5hd2_gmac.c @@ -750,28 +750,10 @@ static const struct net_device_ops hix5hd2_netdev_ops = { .ndo_set_mac_address = hix5hd2_net_set_mac_address, }; -static int hix5hd2_get_settings(struct net_device *net_dev, - struct ethtool_cmd *cmd) -{ - if (!net_dev->phydev) - return -ENODEV; - - return phy_ethtool_gset(net_dev->phydev, cmd); -} - -static int hix5hd2_set_settings(struct net_device *net_dev, - struct ethtool_cmd *cmd) -{ - if (!net_dev->phydev) - return -ENODEV; - - return phy_ethtool_sset(net_dev->phydev, cmd); -} - static struct ethtool_ops hix5hd2_ethtools_ops = { .get_link = ethtool_op_get_link, - .get_settings = hix5hd2_get_settings, - .set_settings = hix5hd2_set_settings, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int hix5hd2_mdio_wait_ready(struct mii_bus *bus) -- cgit v0.10.2 From 542808f5f68810b579c3b9a44d88dbdbe5e68ac2 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 25 Jun 2016 21:09:01 +0200 Subject: net: ethernet: r6040: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 6b541e5..f2bc7b0 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -200,7 +200,6 @@ struct r6040_private { struct mii_bus *mii_bus; struct napi_struct napi; void __iomem *base; - struct phy_device *phydev; int old_link; int old_duplex; }; @@ -474,7 +473,7 @@ static void r6040_down(struct net_device *dev) iowrite16(adrp[1], ioaddr + MID_0M); iowrite16(adrp[2], ioaddr + MID_0H); - phy_stop(lp->phydev); + phy_stop(dev->phydev); } static int r6040_close(struct net_device *dev) @@ -515,12 +514,10 @@ static int r6040_close(struct net_device *dev) static int r6040_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct r6040_private *lp = netdev_priv(dev); - - if (!lp->phydev) + if (!dev->phydev) return -EINVAL; - return phy_mii_ioctl(lp->phydev, rq, cmd); + return phy_mii_ioctl(dev->phydev, rq, cmd); } static int r6040_rx(struct net_device *dev, int limit) @@ -732,7 +729,7 @@ static int r6040_up(struct net_device *dev) /* Initialize all MAC registers */ r6040_init_mac_regs(dev); - phy_start(lp->phydev); + phy_start(dev->phydev); return 0; } @@ -959,16 +956,12 @@ static void netdev_get_drvinfo(struct net_device *dev, static int netdev_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct r6040_private *rp = netdev_priv(dev); - - return phy_ethtool_gset(rp->phydev, cmd); + return phy_ethtool_gset(dev->phydev, cmd); } static int netdev_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct r6040_private *rp = netdev_priv(dev); - - return phy_ethtool_sset(rp->phydev, cmd); + return phy_ethtool_sset(dev->phydev, cmd); } static const struct ethtool_ops netdev_ethtool_ops = { @@ -998,7 +991,7 @@ static const struct net_device_ops r6040_netdev_ops = { static void r6040_adjust_link(struct net_device *dev) { struct r6040_private *lp = netdev_priv(dev); - struct phy_device *phydev = lp->phydev; + struct phy_device *phydev = dev->phydev; int status_changed = 0; void __iomem *ioaddr = lp->base; @@ -1057,7 +1050,6 @@ static int r6040_mii_probe(struct net_device *dev) | SUPPORTED_TP); phydev->advertising = phydev->supported; - lp->phydev = phydev; lp->old_link = 0; lp->old_duplex = -1; -- cgit v0.10.2 From cffce3615dd582ac30a2a0d81d9a46aaa8f4e0bb Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 25 Jun 2016 21:09:02 +0200 Subject: net: ethernet: r6040: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index f2bc7b0..7a7a395 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -954,22 +954,12 @@ static void netdev_get_drvinfo(struct net_device *dev, strlcpy(info->bus_info, pci_name(rp->pdev), sizeof(info->bus_info)); } -static int netdev_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - return phy_ethtool_gset(dev->phydev, cmd); -} - -static int netdev_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - return phy_ethtool_sset(dev->phydev, cmd); -} - static const struct ethtool_ops netdev_ethtool_ops = { .get_drvinfo = netdev_get_drvinfo, - .get_settings = netdev_get_settings, - .set_settings = netdev_set_settings, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops r6040_netdev_ops = { -- cgit v0.10.2 From 2ebc440a1f2be7c572c8ea174d1b2a6a0d4e41a3 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 25 Jun 2016 22:05:26 +0200 Subject: net: ethernet: sxgbe: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_common.h b/drivers/net/ethernet/samsung/sxgbe/sxgbe_common.h index 4501964..5cb51b6 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_common.h +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_common.h @@ -475,7 +475,6 @@ struct sxgbe_priv_data { int rxcsum_insertion; spinlock_t stats_lock; /* lock for tx/rx statatics */ - struct phy_device *phydev; int oldlink; int speed; int oldduplex; diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c index c0981ae..72c9108 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c @@ -147,7 +147,7 @@ static int sxgbe_get_eee(struct net_device *dev, edata->eee_active = priv->eee_active; edata->tx_lpi_timer = priv->tx_lpi_timer; - return phy_ethtool_get_eee(priv->phydev, edata); + return phy_ethtool_get_eee(dev->phydev, edata); } static int sxgbe_set_eee(struct net_device *dev, @@ -172,7 +172,7 @@ static int sxgbe_set_eee(struct net_device *dev, priv->tx_lpi_timer = edata->tx_lpi_timer; } - return phy_ethtool_set_eee(priv->phydev, edata); + return phy_ethtool_set_eee(dev->phydev, edata); } static void sxgbe_getdrvinfo(struct net_device *dev, @@ -185,20 +185,16 @@ static void sxgbe_getdrvinfo(struct net_device *dev, static int sxgbe_getsettings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct sxgbe_priv_data *priv = netdev_priv(dev); - - if (priv->phydev) - return phy_ethtool_gset(priv->phydev, cmd); + if (dev->phydev) + return phy_ethtool_gset(dev->phydev, cmd); return -EOPNOTSUPP; } static int sxgbe_setsettings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct sxgbe_priv_data *priv = netdev_priv(dev); - - if (priv->phydev) - return phy_ethtool_sset(priv->phydev, cmd); + if (dev->phydev) + return phy_ethtool_sset(dev->phydev, cmd); return -EOPNOTSUPP; } @@ -255,7 +251,7 @@ static void sxgbe_get_ethtool_stats(struct net_device *dev, char *p; if (priv->eee_enabled) { - int val = phy_get_eee_err(priv->phydev); + int val = phy_get_eee_err(dev->phydev); if (val) priv->xstats.eee_wakeup_error_n = val; diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c index 413ea14a..ea44a24 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c @@ -124,12 +124,13 @@ static void sxgbe_eee_ctrl_timer(unsigned long arg) */ bool sxgbe_eee_init(struct sxgbe_priv_data * const priv) { + struct net_device *ndev = priv->dev; bool ret = false; /* MAC core supports the EEE feature. */ if (priv->hw_cap.eee) { /* Check if the PHY supports EEE */ - if (phy_init_eee(priv->phydev, 1)) + if (phy_init_eee(ndev->phydev, 1)) return false; priv->eee_active = 1; @@ -152,12 +153,14 @@ bool sxgbe_eee_init(struct sxgbe_priv_data * const priv) static void sxgbe_eee_adjust(const struct sxgbe_priv_data *priv) { + struct net_device *ndev = priv->dev; + /* When the EEE has been already initialised we have to * modify the PLS bit in the LPI ctrl & status reg according * to the PHY link status. For this reason. */ if (priv->eee_enabled) - priv->hw->mac->set_eee_pls(priv->ioaddr, priv->phydev->link); + priv->hw->mac->set_eee_pls(priv->ioaddr, ndev->phydev->link); } /** @@ -203,7 +206,7 @@ static inline u32 sxgbe_tx_avail(struct sxgbe_tx_queue *queue, int tx_qsize) static void sxgbe_adjust_link(struct net_device *dev) { struct sxgbe_priv_data *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; u8 new_state = 0; u8 speed = 0xff; @@ -306,9 +309,6 @@ static int sxgbe_init_phy(struct net_device *ndev) netdev_dbg(ndev, "%s: attached to PHY (UID 0x%x) Link = %d\n", __func__, phydev->phy_id, phydev->link); - /* save phy device in private structure */ - priv->phydev = phydev; - return 0; } @@ -1173,8 +1173,8 @@ static int sxgbe_open(struct net_device *dev) priv->hw->dma->start_tx(priv->ioaddr, SXGBE_TX_QUEUES); priv->hw->dma->start_rx(priv->ioaddr, SXGBE_RX_QUEUES); - if (priv->phydev) - phy_start(priv->phydev); + if (dev->phydev) + phy_start(dev->phydev); /* initialise TX coalesce parameters */ sxgbe_tx_init_coalesce(priv); @@ -1194,8 +1194,8 @@ static int sxgbe_open(struct net_device *dev) init_error: free_dma_desc_resources(priv); - if (priv->phydev) - phy_disconnect(priv->phydev); + if (dev->phydev) + phy_disconnect(dev->phydev); phy_error: clk_disable_unprepare(priv->sxgbe_clk); @@ -1216,10 +1216,9 @@ static int sxgbe_release(struct net_device *dev) del_timer_sync(&priv->eee_ctrl_timer); /* Stop and disconnect the PHY */ - if (priv->phydev) { - phy_stop(priv->phydev); - phy_disconnect(priv->phydev); - priv->phydev = NULL; + if (dev->phydev) { + phy_stop(dev->phydev); + phy_disconnect(dev->phydev); } netif_tx_stop_all_queues(dev); @@ -1969,7 +1968,6 @@ static void sxgbe_poll_controller(struct net_device *dev) */ static int sxgbe_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct sxgbe_priv_data *priv = netdev_priv(dev); int ret = -EOPNOTSUPP; if (!netif_running(dev)) @@ -1979,9 +1977,9 @@ static int sxgbe_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) case SIOCGMIIPHY: case SIOCGMIIREG: case SIOCSMIIREG: - if (!priv->phydev) + if (!dev->phydev) return -EINVAL; - ret = phy_mii_ioctl(priv->phydev, rq, cmd); + ret = phy_mii_ioctl(dev->phydev, rq, cmd); break; default: break; -- cgit v0.10.2 From 13e4c230e05c24794e4bcba647ce44a3d13beaf7 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 25 Jun 2016 22:05:27 +0200 Subject: net: ethernet: sxgbe: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c b/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c index 72c9108..542b67d 100644 --- a/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c +++ b/drivers/net/ethernet/samsung/sxgbe/sxgbe_ethtool.c @@ -182,23 +182,6 @@ static void sxgbe_getdrvinfo(struct net_device *dev, strlcpy(info->version, DRV_VERSION, sizeof(info->version)); } -static int sxgbe_getsettings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - if (dev->phydev) - return phy_ethtool_gset(dev->phydev, cmd); - - return -EOPNOTSUPP; -} - -static int sxgbe_setsettings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - if (dev->phydev) - return phy_ethtool_sset(dev->phydev, cmd); - - return -EOPNOTSUPP; -} - static u32 sxgbe_getmsglevel(struct net_device *dev) { struct sxgbe_priv_data *priv = netdev_priv(dev); @@ -495,8 +478,6 @@ static int sxgbe_get_regs_len(struct net_device *dev) static const struct ethtool_ops sxgbe_ethtool_ops = { .get_drvinfo = sxgbe_getdrvinfo, - .get_settings = sxgbe_getsettings, - .set_settings = sxgbe_setsettings, .get_msglevel = sxgbe_getmsglevel, .set_msglevel = sxgbe_setmsglevel, .get_link = ethtool_op_get_link, @@ -512,6 +493,8 @@ static const struct ethtool_ops sxgbe_ethtool_ops = { .get_regs_len = sxgbe_get_regs_len, .get_eee = sxgbe_get_eee, .set_eee = sxgbe_set_eee, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; void sxgbe_set_ethtool_ops(struct net_device *netdev) -- cgit v0.10.2 From ce554d32eb16c459221a06ee8c93f39d7d648ef4 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 25 Jun 2016 23:05:15 +0200 Subject: net: ethernet: dwc_eth_qos: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/synopsys/dwc_eth_qos.c b/drivers/net/ethernet/synopsys/dwc_eth_qos.c index 158213c..094234f 100644 --- a/drivers/net/ethernet/synopsys/dwc_eth_qos.c +++ b/drivers/net/ethernet/synopsys/dwc_eth_qos.c @@ -598,7 +598,6 @@ struct net_local { struct work_struct txtimeout_reinit; phy_interface_t phy_interface; - struct phy_device *phy_dev; struct mii_bus *mii_bus; unsigned int link; @@ -816,7 +815,7 @@ static int dwceqos_mdio_write(struct mii_bus *bus, int mii_id, int phyreg, static int dwceqos_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd) { struct net_local *lp = netdev_priv(ndev); - struct phy_device *phydev = lp->phy_dev; + struct phy_device *phydev = ndev->phydev; if (!netif_running(ndev)) return -EINVAL; @@ -850,6 +849,7 @@ static void dwceqos_link_down(struct net_local *lp) static void dwceqos_link_up(struct net_local *lp) { + struct net_device *ndev = lp->ndev; u32 regval; unsigned long flags; @@ -860,7 +860,7 @@ static void dwceqos_link_up(struct net_local *lp) dwceqos_write(lp, REG_DWCEQOS_MAC_LPI_CTRL_STATUS, regval); spin_unlock_irqrestore(&lp->hw_lock, flags); - lp->eee_active = !phy_init_eee(lp->phy_dev, 0); + lp->eee_active = !phy_init_eee(ndev->phydev, 0); /* Check for changed EEE capability */ if (!lp->eee_active && lp->eee_enabled) { @@ -876,7 +876,8 @@ static void dwceqos_link_up(struct net_local *lp) static void dwceqos_set_speed(struct net_local *lp) { - struct phy_device *phydev = lp->phy_dev; + struct net_device *ndev = lp->ndev; + struct phy_device *phydev = ndev->phydev; u32 regval; regval = dwceqos_read(lp, REG_DWCEQOS_MAC_CFG); @@ -903,7 +904,7 @@ static void dwceqos_set_speed(struct net_local *lp) static void dwceqos_adjust_link(struct net_device *ndev) { struct net_local *lp = netdev_priv(ndev); - struct phy_device *phydev = lp->phy_dev; + struct phy_device *phydev = ndev->phydev; int status_change = 0; if (lp->phy_defer) @@ -987,7 +988,6 @@ static int dwceqos_mii_probe(struct net_device *ndev) lp->link = 0; lp->speed = 0; lp->duplex = DUPLEX_UNKNOWN; - lp->phy_dev = phydev; return 0; } @@ -1531,6 +1531,7 @@ static void dwceqos_configure_bus(struct net_local *lp) static void dwceqos_init_hw(struct net_local *lp) { + struct net_device *ndev = lp->ndev; u32 regval; u32 buswidth; u32 dma_skip; @@ -1645,10 +1646,10 @@ static void dwceqos_init_hw(struct net_local *lp) regval | DWCEQOS_MAC_CFG_TE | DWCEQOS_MAC_CFG_RE); lp->phy_defer = false; - mutex_lock(&lp->phy_dev->lock); - phy_read_status(lp->phy_dev); + mutex_lock(&ndev->phydev->lock); + phy_read_status(ndev->phydev); dwceqos_adjust_link(lp->ndev); - mutex_unlock(&lp->phy_dev->lock); + mutex_unlock(&ndev->phydev->lock); } static void dwceqos_tx_reclaim(unsigned long data) @@ -1898,7 +1899,7 @@ static int dwceqos_open(struct net_device *ndev) * hence the unusual init order with phy_start first. */ lp->phy_defer = true; - phy_start(lp->phy_dev); + phy_start(ndev->phydev); dwceqos_init_hw(lp); napi_enable(&lp->napi); @@ -1943,7 +1944,7 @@ static int dwceqos_stop(struct net_device *ndev) dwceqos_drain_dma(lp); dwceqos_reset_hw(lp); - phy_stop(lp->phy_dev); + phy_stop(ndev->phydev); dwceqos_descriptor_free(lp); @@ -2526,8 +2527,7 @@ dwceqos_get_stats64(struct net_device *ndev, struct rtnl_link_stats64 *s) static int dwceqos_get_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) { - struct net_local *lp = netdev_priv(ndev); - struct phy_device *phydev = lp->phy_dev; + struct phy_device *phydev = ndev->phydev; if (!phydev) return -ENODEV; @@ -2538,8 +2538,7 @@ dwceqos_get_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) static int dwceqos_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) { - struct net_local *lp = netdev_priv(ndev); - struct phy_device *phydev = lp->phy_dev; + struct phy_device *phydev = ndev->phydev; if (!phydev) return -ENODEV; @@ -2574,17 +2573,17 @@ static int dwceqos_set_pauseparam(struct net_device *ndev, lp->flowcontrol.autoneg = pp->autoneg; if (pp->autoneg) { - lp->phy_dev->advertising |= ADVERTISED_Pause; - lp->phy_dev->advertising |= ADVERTISED_Asym_Pause; + ndev->phydev->advertising |= ADVERTISED_Pause; + ndev->phydev->advertising |= ADVERTISED_Asym_Pause; } else { - lp->phy_dev->advertising &= ~ADVERTISED_Pause; - lp->phy_dev->advertising &= ~ADVERTISED_Asym_Pause; + ndev->phydev->advertising &= ~ADVERTISED_Pause; + ndev->phydev->advertising &= ~ADVERTISED_Asym_Pause; lp->flowcontrol.rx = pp->rx_pause; lp->flowcontrol.tx = pp->tx_pause; } if (netif_running(ndev)) - ret = phy_start_aneg(lp->phy_dev); + ret = phy_start_aneg(ndev->phydev); return ret; } @@ -2705,7 +2704,7 @@ static int dwceqos_get_eee(struct net_device *ndev, struct ethtool_eee *edata) dwceqos_get_tx_lpi_state(regval)); } - return phy_ethtool_get_eee(lp->phy_dev, edata); + return phy_ethtool_get_eee(ndev->phydev, edata); } static int dwceqos_set_eee(struct net_device *ndev, struct ethtool_eee *edata) @@ -2747,7 +2746,7 @@ static int dwceqos_set_eee(struct net_device *ndev, struct ethtool_eee *edata) spin_unlock_irqrestore(&lp->hw_lock, flags); } - return phy_ethtool_set_eee(lp->phy_dev, edata); + return phy_ethtool_set_eee(ndev->phydev, edata); } static u32 dwceqos_get_msglevel(struct net_device *ndev) @@ -2981,8 +2980,8 @@ static int dwceqos_remove(struct platform_device *pdev) if (ndev) { lp = netdev_priv(ndev); - if (lp->phy_dev) - phy_disconnect(lp->phy_dev); + if (ndev->phydev) + phy_disconnect(ndev->phydev); mdiobus_unregister(lp->mii_bus); mdiobus_free(lp->mii_bus); -- cgit v0.10.2 From 8a79813c140122b9448bb8d24ec58dff3b15ea31 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 25 Jun 2016 23:05:16 +0200 Subject: net: ethernet: dwc_eth_qos: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/synopsys/dwc_eth_qos.c b/drivers/net/ethernet/synopsys/dwc_eth_qos.c index 094234f..c14fa91 100644 --- a/drivers/net/ethernet/synopsys/dwc_eth_qos.c +++ b/drivers/net/ethernet/synopsys/dwc_eth_qos.c @@ -2524,28 +2524,6 @@ dwceqos_get_stats64(struct net_device *ndev, struct rtnl_link_stats64 *s) return s; } -static int -dwceqos_get_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) -{ - struct phy_device *phydev = ndev->phydev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_gset(phydev, ecmd); -} - -static int -dwceqos_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) -{ - struct phy_device *phydev = ndev->phydev; - - if (!phydev) - return -ENODEV; - - return phy_ethtool_sset(phydev, ecmd); -} - static void dwceqos_get_drvinfo(struct net_device *ndev, struct ethtool_drvinfo *ed) { @@ -2764,8 +2742,6 @@ static void dwceqos_set_msglevel(struct net_device *ndev, u32 msglevel) } static struct ethtool_ops dwceqos_ethtool_ops = { - .get_settings = dwceqos_get_settings, - .set_settings = dwceqos_set_settings, .get_drvinfo = dwceqos_get_drvinfo, .get_link = ethtool_op_get_link, .get_pauseparam = dwceqos_get_pauseparam, @@ -2779,6 +2755,8 @@ static struct ethtool_ops dwceqos_ethtool_ops = { .set_eee = dwceqos_set_eee, .get_msglevel = dwceqos_get_msglevel, .set_msglevel = dwceqos_set_msglevel, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static struct net_device_ops netdev_ops = { -- cgit v0.10.2 From c37d4a0085c58d9e45930ead6acd13ac75a8fb67 Mon Sep 17 00:00:00 2001 From: Zhao Qiang Date: Mon, 27 Jun 2016 09:30:22 +0800 Subject: Maxim/driver: Add driver for maxim ds26522 Signed-off-by: Zhao Qiang Signed-off-by: David S. Miller diff --git a/drivers/net/wan/Kconfig b/drivers/net/wan/Kconfig index 9e314b7..33ab334 100644 --- a/drivers/net/wan/Kconfig +++ b/drivers/net/wan/Kconfig @@ -291,6 +291,17 @@ config FSL_UCC_HDLC To compile this driver as a module, choose M here: the module will be called fsl_ucc_hdlc. +config SLIC_DS26522 + tristate "Slic Maxim ds26522 card support" + depends on SPI + depends on FSL_SOC || ARCH_MXC || ARCH_LAYERSCAPE + help + This module initializes and configures the slic maxim card + in T1 or E1 mode. + + To compile this driver as a module, choose M here: the + module will be called slic_ds26522. + config DSCC4_PCISYNC bool "Etinc PCISYNC features" depends on DSCC4 diff --git a/drivers/net/wan/Makefile b/drivers/net/wan/Makefile index 25fec40..73c2326 100644 --- a/drivers/net/wan/Makefile +++ b/drivers/net/wan/Makefile @@ -33,6 +33,7 @@ obj-$(CONFIG_PCI200SYN) += pci200syn.o obj-$(CONFIG_PC300TOO) += pc300too.o obj-$(CONFIG_IXP4XX_HSS) += ixp4xx_hss.o obj-$(CONFIG_FSL_UCC_HDLC) += fsl_ucc_hdlc.o +obj-$(CONFIG_SLIC_DS26522) += slic_ds26522.o clean-files := wanxlfw.inc $(obj)/wanxl.o: $(obj)/wanxlfw.inc diff --git a/drivers/net/wan/slic_ds26522.c b/drivers/net/wan/slic_ds26522.c new file mode 100644 index 0000000..d06a887 --- /dev/null +++ b/drivers/net/wan/slic_ds26522.c @@ -0,0 +1,255 @@ +/* + * drivers/net/wan/slic_ds26522.c + * + * Copyright (C) 2016 Freescale Semiconductor, Inc. + * + * Author:Zhao Qiang + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "slic_ds26522.h" + +#define DRV_NAME "ds26522" + +#define SLIC_TRANS_LEN 1 +#define SLIC_TWO_LEN 2 +#define SLIC_THREE_LEN 3 + +static struct spi_device *g_spi; + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Zhao Qiang"); + +/* the read/write format of address is + * w/r|A13|A12|A11|A10|A9|A8|A7|A6|A5|A4|A3|A2|A1|A0|x + */ +static void slic_write(struct spi_device *spi, u16 addr, + u8 data) +{ + u8 temp[3]; + + addr = bitrev16(addr) >> 1; + data = bitrev8(data); + temp[0] = (u8)((addr >> 8) & 0x7f); + temp[1] = (u8)(addr & 0xfe); + temp[2] = data; + + /* write spi addr and value */ + spi_write(spi, &temp[0], SLIC_THREE_LEN); +} + +static u8 slic_read(struct spi_device *spi, u16 addr) +{ + u8 temp[2]; + u8 data; + + addr = bitrev16(addr) >> 1; + temp[0] = (u8)(((addr >> 8) & 0x7f) | 0x80); + temp[1] = (u8)(addr & 0xfe); + + spi_write_then_read(spi, &temp[0], SLIC_TWO_LEN, &data, + SLIC_TRANS_LEN); + + data = bitrev8(data); + return data; +} + +static bool get_slic_product_code(struct spi_device *spi) +{ + u8 device_id; + + device_id = slic_read(spi, DS26522_IDR_ADDR); + if ((device_id & 0xf8) == 0x68) + return true; + else + return false; +} + +static void ds26522_e1_spec_config(struct spi_device *spi) +{ + /* Receive E1 Mode, Framer Disabled */ + slic_write(spi, DS26522_RMMR_ADDR, DS26522_RMMR_E1); + + /* Transmit E1 Mode, Framer Disable */ + slic_write(spi, DS26522_TMMR_ADDR, DS26522_TMMR_E1); + + /* Receive E1 Mode Framer Enable */ + slic_write(spi, DS26522_RMMR_ADDR, + slic_read(spi, DS26522_RMMR_ADDR) | DS26522_RMMR_FRM_EN); + + /* Transmit E1 Mode Framer Enable */ + slic_write(spi, DS26522_TMMR_ADDR, + slic_read(spi, DS26522_TMMR_ADDR) | DS26522_TMMR_FRM_EN); + + /* RCR1, receive E1 B8zs & ESF */ + slic_write(spi, DS26522_RCR1_ADDR, + DS26522_RCR1_E1_HDB3 | DS26522_RCR1_E1_CCS); + + /* RSYSCLK=2.048MHz, RSYNC-Output */ + slic_write(spi, DS26522_RIOCR_ADDR, + DS26522_RIOCR_2048KHZ | DS26522_RIOCR_RSIO_OUT); + + /* TCR1 Transmit E1 b8zs */ + slic_write(spi, DS26522_TCR1_ADDR, DS26522_TCR1_TB8ZS); + + /* TSYSCLK=2.048MHz, TSYNC-Output */ + slic_write(spi, DS26522_TIOCR_ADDR, + DS26522_TIOCR_2048KHZ | DS26522_TIOCR_TSIO_OUT); + + /* Set E1TAF */ + slic_write(spi, DS26522_E1TAF_ADDR, DS26522_E1TAF_DEFAULT); + + /* Set E1TNAF register */ + slic_write(spi, DS26522_E1TNAF_ADDR, DS26522_E1TNAF_DEFAULT); + + /* Receive E1 Mode Framer Enable & init Done */ + slic_write(spi, DS26522_RMMR_ADDR, slic_read(spi, DS26522_RMMR_ADDR) | + DS26522_RMMR_INIT_DONE); + + /* Transmit E1 Mode Framer Enable & init Done */ + slic_write(spi, DS26522_TMMR_ADDR, slic_read(spi, DS26522_TMMR_ADDR) | + DS26522_TMMR_INIT_DONE); + + /* Configure LIU E1 mode */ + slic_write(spi, DS26522_LTRCR_ADDR, DS26522_LTRCR_E1); + + /* E1 Mode default 75 ohm w/Transmit Impedance Matlinking */ + slic_write(spi, DS26522_LTITSR_ADDR, + DS26522_LTITSR_TLIS_75OHM | DS26522_LTITSR_LBOS_75OHM); + + /* E1 Mode default 75 ohm Long Haul w/Receive Impedance Matlinking */ + slic_write(spi, DS26522_LRISMR_ADDR, + DS26522_LRISMR_75OHM | DS26522_LRISMR_MAX); + + /* Enable Transmit output */ + slic_write(spi, DS26522_LMCR_ADDR, DS26522_LMCR_TE); +} + +static int slic_ds26522_init_configure(struct spi_device *spi) +{ + u16 addr; + + /* set clock */ + slic_write(spi, DS26522_GTCCR_ADDR, DS26522_GTCCR_BPREFSEL_REFCLKIN | + DS26522_GTCCR_BFREQSEL_2048KHZ | + DS26522_GTCCR_FREQSEL_2048KHZ); + slic_write(spi, DS26522_GTCR2_ADDR, DS26522_GTCR2_TSSYNCOUT); + slic_write(spi, DS26522_GFCR_ADDR, DS26522_GFCR_BPCLK_2048KHZ); + + /* set gtcr */ + slic_write(spi, DS26522_GTCR1_ADDR, DS26522_GTCR1); + + /* Global LIU Software Reset Register */ + slic_write(spi, DS26522_GLSRR_ADDR, DS26522_GLSRR_RESET); + + /* Global Framer and BERT Software Reset Register */ + slic_write(spi, DS26522_GFSRR_ADDR, DS26522_GFSRR_RESET); + + usleep_range(100, 120); + + slic_write(spi, DS26522_GLSRR_ADDR, DS26522_GLSRR_NORMAL); + slic_write(spi, DS26522_GFSRR_ADDR, DS26522_GFSRR_NORMAL); + + /* Perform RX/TX SRESET,Reset receiver */ + slic_write(spi, DS26522_RMMR_ADDR, DS26522_RMMR_SFTRST); + + /* Reset tranceiver */ + slic_write(spi, DS26522_TMMR_ADDR, DS26522_TMMR_SFTRST); + + usleep_range(100, 120); + + /* Zero all Framer Registers */ + for (addr = DS26522_RF_ADDR_START; addr <= DS26522_RF_ADDR_END; + addr++) + slic_write(spi, addr, 0); + + for (addr = DS26522_TF_ADDR_START; addr <= DS26522_TF_ADDR_END; + addr++) + slic_write(spi, addr, 0); + + for (addr = DS26522_LIU_ADDR_START; addr <= DS26522_LIU_ADDR_END; + addr++) + slic_write(spi, addr, 0); + + for (addr = DS26522_BERT_ADDR_START; addr <= DS26522_BERT_ADDR_END; + addr++) + slic_write(spi, addr, 0); + + /* setup ds26522 for E1 specification */ + ds26522_e1_spec_config(spi); + + slic_write(spi, DS26522_GTCR1_ADDR, 0x00); + + return 0; +} + +static int slic_ds26522_remove(struct spi_device *spi) +{ + pr_info("DS26522 module uninstalled\n"); + return 0; +} + +static int slic_ds26522_probe(struct spi_device *spi) +{ + int ret = 0; + + g_spi = spi; + spi->bits_per_word = 8; + + if (!get_slic_product_code(spi)) + return ret; + + ret = slic_ds26522_init_configure(spi); + if (ret == 0) + pr_info("DS26522 cs%d configurated\n", spi->chip_select); + + return ret; +} + +static const struct of_device_id slic_ds26522_match[] = { + { + .compatible = "maxim,ds26522", + }, + {}, +}; + +static struct spi_driver slic_ds26522_driver = { + .driver = { + .name = "ds26522", + .bus = &spi_bus_type, + .owner = THIS_MODULE, + .of_match_table = slic_ds26522_match, + }, + .probe = slic_ds26522_probe, + .remove = slic_ds26522_remove, +}; + +static int __init slic_ds26522_init(void) +{ + return spi_register_driver(&slic_ds26522_driver); +} + +static void __exit slic_ds26522_exit(void) +{ + spi_unregister_driver(&slic_ds26522_driver); +} + +module_init(slic_ds26522_init); +module_exit(slic_ds26522_exit); diff --git a/drivers/net/wan/slic_ds26522.h b/drivers/net/wan/slic_ds26522.h new file mode 100644 index 0000000..22aa0ec --- /dev/null +++ b/drivers/net/wan/slic_ds26522.h @@ -0,0 +1,134 @@ +/* + * drivers/tdm/line_ctrl/slic_ds26522.h + * + * Copyright 2016 Freescale Semiconductor, Inc. + * + * Author: Zhao Qiang + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +#define DS26522_RF_ADDR_START 0x00 +#define DS26522_RF_ADDR_END 0xef +#define DS26522_GLB_ADDR_START 0xf0 +#define DS26522_GLB_ADDR_END 0xff +#define DS26522_TF_ADDR_START 0x100 +#define DS26522_TF_ADDR_END 0x1ef +#define DS26522_LIU_ADDR_START 0x1000 +#define DS26522_LIU_ADDR_END 0x101f +#define DS26522_TEST_ADDR_START 0x1008 +#define DS26522_TEST_ADDR_END 0x101f +#define DS26522_BERT_ADDR_START 0x1100 +#define DS26522_BERT_ADDR_END 0x110f + +#define DS26522_RMMR_ADDR 0x80 +#define DS26522_RCR1_ADDR 0x81 +#define DS26522_RCR3_ADDR 0x83 +#define DS26522_RIOCR_ADDR 0x84 + +#define DS26522_GTCR1_ADDR 0xf0 +#define DS26522_GFCR_ADDR 0xf1 +#define DS26522_GTCR2_ADDR 0xf2 +#define DS26522_GTCCR_ADDR 0xf3 +#define DS26522_GLSRR_ADDR 0xf5 +#define DS26522_GFSRR_ADDR 0xf6 +#define DS26522_IDR_ADDR 0xf8 + +#define DS26522_E1TAF_ADDR 0x164 +#define DS26522_E1TNAF_ADDR 0x165 +#define DS26522_TMMR_ADDR 0x180 +#define DS26522_TCR1_ADDR 0x181 +#define DS26522_TIOCR_ADDR 0x184 + +#define DS26522_LTRCR_ADDR 0x1000 +#define DS26522_LTITSR_ADDR 0x1001 +#define DS26522_LMCR_ADDR 0x1002 +#define DS26522_LRISMR_ADDR 0x1007 + +#define MAX_NUM_OF_CHANNELS 8 +#define PQ_MDS_8E1T1_BRD_REV 0x00 +#define PQ_MDS_8E1T1_PLD_REV 0x00 + +#define DS26522_GTCCR_BPREFSEL_REFCLKIN 0xa0 +#define DS26522_GTCCR_BFREQSEL_1544KHZ 0x08 +#define DS26522_GTCCR_FREQSEL_1544KHZ 0x04 +#define DS26522_GTCCR_BFREQSEL_2048KHZ 0x00 +#define DS26522_GTCCR_FREQSEL_2048KHZ 0x00 + +#define DS26522_GFCR_BPCLK_2048KHZ 0x00 + +#define DS26522_GTCR2_TSSYNCOUT 0x02 +#define DS26522_GTCR1 0x00 + +#define DS26522_GFSRR_RESET 0x01 +#define DS26522_GFSRR_NORMAL 0x00 + +#define DS26522_GLSRR_RESET 0x01 +#define DS26522_GLSRR_NORMAL 0x00 + +#define DS26522_RMMR_SFTRST 0x02 +#define DS26522_RMMR_FRM_EN 0x80 +#define DS26522_RMMR_INIT_DONE 0x40 +#define DS26522_RMMR_T1 0x00 +#define DS26522_RMMR_E1 0x01 + +#define DS26522_E1TAF_DEFAULT 0x1b +#define DS26522_E1TNAF_DEFAULT 0x40 + +#define DS26522_TMMR_SFTRST 0x02 +#define DS26522_TMMR_FRM_EN 0x80 +#define DS26522_TMMR_INIT_DONE 0x40 +#define DS26522_TMMR_T1 0x00 +#define DS26522_TMMR_E1 0x01 + +#define DS26522_RCR1_T1_SYNCT 0x80 +#define DS26522_RCR1_T1_RB8ZS 0x40 +#define DS26522_RCR1_T1_SYNCC 0x08 + +#define DS26522_RCR1_E1_HDB3 0x40 +#define DS26522_RCR1_E1_CCS 0x20 + +#define DS26522_RIOCR_1544KHZ 0x00 +#define DS26522_RIOCR_2048KHZ 0x10 +#define DS26522_RIOCR_RSIO_OUT 0x00 + +#define DS26522_RCR3_FLB 0x01 + +#define DS26522_TIOCR_1544KHZ 0x00 +#define DS26522_TIOCR_2048KHZ 0x10 +#define DS26522_TIOCR_TSIO_OUT 0x04 + +#define DS26522_TCR1_TB8ZS 0x04 + +#define DS26522_LTRCR_T1 0x02 +#define DS26522_LTRCR_E1 0x00 + +#define DS26522_LTITSR_TLIS_75OHM 0x00 +#define DS26522_LTITSR_LBOS_75OHM 0x00 +#define DS26522_LTITSR_TLIS_100OHM 0x10 +#define DS26522_LTITSR_TLIS_0DB_CSU 0x00 + +#define DS26522_LRISMR_75OHM 0x00 +#define DS26522_LRISMR_100OHM 0x10 +#define DS26522_LRISMR_MAX 0x03 + +#define DS26522_LMCR_TE 0x01 + +enum line_rate { + LINE_RATE_T1, /* T1 line rate (1.544 Mbps) */ + LINE_RATE_E1 /* E1 line rate (2.048 Mbps) */ +}; + +enum tdm_trans_mode { + NORMAL = 0, + FRAMER_LB +}; + +enum card_support_type { + LM_CARD = 0, + DS26522_CARD, + NO_CARD +}; -- cgit v0.10.2 From 742fb20fd4c75bd08733b0ea232c7e0fa67a6f87 Mon Sep 17 00:00:00 2001 From: Grygorii Strashko Date: Mon, 27 Jun 2016 12:05:11 +0300 Subject: net: ethernet: ti: cpdma: switch to use genalloc TI CPDMA currently uses a bitmap for tracking descriptors alloactions allocations, but The genalloc already handles the same and can be used as with special memory (SRAM) as with DMA cherent memory chank (dma_alloc_coherent()). Hence, switch to using genalloc and add desc_num property for each channel for limitation of max number of allowed descriptors for each CPDMA channel. This patch do not affect on net throuput. Acked-by: Mugunthan V N Tested-by: Ivan Khoronzhuk Signed-off-by: Grygorii Strashko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c index bcd9e45..1c653ca 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.c +++ b/drivers/net/ethernet/ti/davinci_cpdma.c @@ -21,7 +21,7 @@ #include #include #include - +#include #include "davinci_cpdma.h" /* DMA Registers */ @@ -87,9 +87,8 @@ struct cpdma_desc_pool { void *cpumap; /* dma_alloc map */ int desc_size, mem_size; int num_desc, used_desc; - unsigned long *bitmap; struct device *dev; - spinlock_t lock; + struct gen_pool *gen_pool; }; enum cpdma_state { @@ -117,6 +116,7 @@ struct cpdma_chan { int chan_num; spinlock_t lock; int count; + u32 desc_num; u32 mask; cpdma_handler_fn handler; enum dma_data_direction dir; @@ -145,6 +145,19 @@ struct cpdma_chan { (directed << CPDMA_TO_PORT_SHIFT)); \ } while (0) +static void cpdma_desc_pool_destroy(struct cpdma_desc_pool *pool) +{ + if (!pool) + return; + + WARN_ON(pool->used_desc); + if (pool->cpumap) + dma_free_coherent(pool->dev, pool->mem_size, pool->cpumap, + pool->phys); + else + iounmap(pool->iomap); +} + /* * Utility constructs for a cpdma descriptor pool. Some devices (e.g. davinci * emac) have dedicated on-chip memory for these descriptors. Some other @@ -155,24 +168,25 @@ static struct cpdma_desc_pool * cpdma_desc_pool_create(struct device *dev, u32 phys, dma_addr_t hw_addr, int size, int align) { - int bitmap_size; struct cpdma_desc_pool *pool; + int ret; pool = devm_kzalloc(dev, sizeof(*pool), GFP_KERNEL); if (!pool) - goto fail; - - spin_lock_init(&pool->lock); + goto gen_pool_create_fail; pool->dev = dev; pool->mem_size = size; pool->desc_size = ALIGN(sizeof(struct cpdma_desc), align); pool->num_desc = size / pool->desc_size; - bitmap_size = (pool->num_desc / BITS_PER_LONG) * sizeof(long); - pool->bitmap = devm_kzalloc(dev, bitmap_size, GFP_KERNEL); - if (!pool->bitmap) - goto fail; + pool->gen_pool = devm_gen_pool_create(dev, ilog2(pool->desc_size), -1, + "cpdma"); + if (IS_ERR(pool->gen_pool)) { + dev_err(dev, "pool create failed %ld\n", + PTR_ERR(pool->gen_pool)); + goto gen_pool_create_fail; + } if (phys) { pool->phys = phys; @@ -185,24 +199,22 @@ cpdma_desc_pool_create(struct device *dev, u32 phys, dma_addr_t hw_addr, pool->phys = pool->hw_addr; /* assumes no IOMMU, don't use this value */ } - if (pool->iomap) - return pool; -fail: - return NULL; -} + if (!pool->iomap) + goto gen_pool_create_fail; -static void cpdma_desc_pool_destroy(struct cpdma_desc_pool *pool) -{ - if (!pool) - return; - - WARN_ON(pool->used_desc); - if (pool->cpumap) { - dma_free_coherent(pool->dev, pool->mem_size, pool->cpumap, - pool->phys); - } else { - iounmap(pool->iomap); + ret = gen_pool_add_virt(pool->gen_pool, (unsigned long)pool->iomap, + pool->phys, pool->mem_size, -1); + if (ret < 0) { + dev_err(dev, "pool add failed %d\n", ret); + goto gen_pool_add_virt_fail; } + + return pool; + +gen_pool_add_virt_fail: + cpdma_desc_pool_destroy(pool); +gen_pool_create_fail: + return NULL; } static inline dma_addr_t desc_phys(struct cpdma_desc_pool *pool, @@ -220,47 +232,23 @@ desc_from_phys(struct cpdma_desc_pool *pool, dma_addr_t dma) } static struct cpdma_desc __iomem * -cpdma_desc_alloc(struct cpdma_desc_pool *pool, int num_desc, bool is_rx) +cpdma_desc_alloc(struct cpdma_desc_pool *pool) { - unsigned long flags; - int index; - int desc_start; - int desc_end; struct cpdma_desc __iomem *desc = NULL; - spin_lock_irqsave(&pool->lock, flags); - - if (is_rx) { - desc_start = 0; - desc_end = pool->num_desc/2; - } else { - desc_start = pool->num_desc/2; - desc_end = pool->num_desc; - } - - index = bitmap_find_next_zero_area(pool->bitmap, - desc_end, desc_start, num_desc, 0); - if (index < desc_end) { - bitmap_set(pool->bitmap, index, num_desc); - desc = pool->iomap + pool->desc_size * index; + desc = (struct cpdma_desc __iomem *)gen_pool_alloc(pool->gen_pool, + pool->desc_size); + if (desc) pool->used_desc++; - } - spin_unlock_irqrestore(&pool->lock, flags); return desc; } static void cpdma_desc_free(struct cpdma_desc_pool *pool, struct cpdma_desc __iomem *desc, int num_desc) { - unsigned long flags, index; - - index = ((unsigned long)desc - (unsigned long)pool->iomap) / - pool->desc_size; - spin_lock_irqsave(&pool->lock, flags); - bitmap_clear(pool->bitmap, index, num_desc); + gen_pool_free(pool->gen_pool, (unsigned long)desc, pool->desc_size); pool->used_desc--; - spin_unlock_irqrestore(&pool->lock, flags); } struct cpdma_ctlr *cpdma_ctlr_create(struct cpdma_params *params) @@ -516,6 +504,7 @@ struct cpdma_chan *cpdma_chan_create(struct cpdma_ctlr *ctlr, int chan_num, chan->state = CPDMA_STATE_IDLE; chan->chan_num = chan_num; chan->handler = handler; + chan->desc_num = ctlr->pool->num_desc / 2; if (is_rx_chan(chan)) { chan->hdp = ctlr->params.rxhdp + offset; @@ -681,7 +670,13 @@ int cpdma_chan_submit(struct cpdma_chan *chan, void *token, void *data, goto unlock_ret; } - desc = cpdma_desc_alloc(ctlr->pool, 1, is_rx_chan(chan)); + if (chan->count >= chan->desc_num) { + chan->stats.desc_alloc_fail++; + ret = -ENOMEM; + goto unlock_ret; + } + + desc = cpdma_desc_alloc(ctlr->pool); if (!desc) { chan->stats.desc_alloc_fail++; ret = -ENOMEM; @@ -727,24 +722,16 @@ EXPORT_SYMBOL_GPL(cpdma_chan_submit); bool cpdma_check_free_tx_desc(struct cpdma_chan *chan) { - unsigned long flags; - int index; - bool ret; struct cpdma_ctlr *ctlr = chan->ctlr; struct cpdma_desc_pool *pool = ctlr->pool; + bool free_tx_desc; + unsigned long flags; - spin_lock_irqsave(&pool->lock, flags); - - index = bitmap_find_next_zero_area(pool->bitmap, - pool->num_desc, pool->num_desc/2, 1, 0); - - if (index < pool->num_desc) - ret = true; - else - ret = false; - - spin_unlock_irqrestore(&pool->lock, flags); - return ret; + spin_lock_irqsave(&chan->lock, flags); + free_tx_desc = (chan->count < chan->desc_num) && + gen_pool_avail(pool->gen_pool); + spin_unlock_irqrestore(&chan->lock, flags); + return free_tx_desc; } EXPORT_SYMBOL_GPL(cpdma_check_free_tx_desc); -- cgit v0.10.2 From 5eca2914f3cf787cced92e818bff5f1ac4f0a8b2 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 27 Jun 2016 11:19:13 +0200 Subject: dsa: b53: avoid 'maybe-uninitialized' warning In some configurations, gcc produces a warning for correct code in this driver: drivers/net/dsa/b53/b53_mmap.c: In function 'b53_mmap_read64': drivers/net/dsa/b53/b53_mmap.c:107:10: error: 'hi' may be used uninitialized in this function [-Werror=maybe-uninitialized] *val = ((u64)hi << 32) | lo; ^~~~~~~ drivers/net/dsa/b53/b53_mmap.c: In function 'b53_mmap_read48': drivers/net/dsa/b53/b53_mmap.c:91:11: error: 'hi' may be used uninitialized in this function [-Werror=maybe-uninitialized] *val = ((u64)hi << 32) | lo; ^~~~~~~ drivers/net/dsa/b53/b53_mmap.c:83:11: error: 'hi' may be used uninitialized in this function [-Werror=maybe-uninitialized] *val = ((u64)hi << 16) | lo; I have seen the warning before and at the time thought I had fixed it with 55e7f6abe131 ("dsa: b53: fix big-endian register access"), however it now came back in a different randconfig build that happens to have different inlining decisions in the compiler. The mistake that gcc makes here is that it thinks the second call to readl() might fail because the address 'reg + 4' is not a multiple of four despite having knowing that 'reg' itself is a multiple of four. By open-coding the two reads without the redundant alignment check, we can avoid the warning and produce slightly better object code, but get slightly longer source code instead. Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/b53/b53_mmap.c b/drivers/net/dsa/b53/b53_mmap.c index b70daf9..21f1068 100644 --- a/drivers/net/dsa/b53/b53_mmap.c +++ b/drivers/net/dsa/b53/b53_mmap.c @@ -70,6 +70,8 @@ static int b53_mmap_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) static int b53_mmap_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) { + u8 __iomem *regs = dev->priv; + if (WARN_ON(reg % 2)) return -EINVAL; @@ -77,16 +79,26 @@ static int b53_mmap_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) u16 lo; u32 hi; - b53_mmap_read16(dev, page, reg, &lo); - b53_mmap_read32(dev, page, reg + 2, &hi); + if (dev->pdata && dev->pdata->big_endian) { + lo = ioread16be(regs + (page << 8) + reg); + hi = ioread32be(regs + (page << 8) + reg + 2); + } else { + lo = readw(regs + (page << 8) + reg); + hi = readl(regs + (page << 8) + reg + 2); + } *val = ((u64)hi << 16) | lo; } else { u32 lo; u16 hi; - b53_mmap_read32(dev, page, reg, &lo); - b53_mmap_read16(dev, page, reg + 4, &hi); + if (dev->pdata && dev->pdata->big_endian) { + lo = ioread32be(regs + (page << 8) + reg); + hi = ioread16be(regs + (page << 8) + reg + 4); + } else { + lo = readl(regs + (page << 8) + reg); + hi = readw(regs + (page << 8) + reg + 4); + } *val = ((u64)hi << 32) | lo; } @@ -96,13 +108,19 @@ static int b53_mmap_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) static int b53_mmap_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) { + u8 __iomem *regs = dev->priv; u32 hi, lo; if (WARN_ON(reg % 4)) return -EINVAL; - b53_mmap_read32(dev, page, reg, &lo); - b53_mmap_read32(dev, page, reg + 4, &hi); + if (dev->pdata && dev->pdata->big_endian) { + lo = ioread32be(regs + (page << 8) + reg); + hi = ioread32be(regs + (page << 8) + reg + 4); + } else { + lo = readl(regs + (page << 8) + reg); + hi = readl(regs + (page << 8) + reg + 4); + } *val = ((u64)hi << 32) | lo; -- cgit v0.10.2 From 8a01ed70ebe4ddf37a759e8e9b4e8e71fb26b47c Mon Sep 17 00:00:00 2001 From: Wei Tang Date: Mon, 27 Jun 2016 18:12:46 +0800 Subject: net: the space is required before the open parenthesis '(' The space is missing before the open parenthesis '(', and this will introduce much more noise when checking patch around. Signed-off-by: Wei Tang Signed-off-by: David S. Miller diff --git a/net/core/utils.c b/net/core/utils.c index 3d17ca8..cf5622b 100644 --- a/net/core/utils.c +++ b/net/core/utils.c @@ -133,7 +133,7 @@ int in4_pton(const char *src, int srclen, s = src; d = dbuf; i = 0; - while(1) { + while (1) { int c; c = xdigit2bin(srclen > 0 ? *s : '\0', delim); if (!(c & (IN6PTON_DIGIT | IN6PTON_DOT | IN6PTON_DELIM | IN6PTON_COLON_MASK))) { @@ -283,11 +283,11 @@ cont: i = 15; d--; if (dc) { - while(d >= dc) + while (d >= dc) dst[i--] = *d--; - while(i >= dc - dbuf) + while (i >= dc - dbuf) dst[i--] = 0; - while(i >= 0) + while (i >= 0) dst[i--] = *d--; } else memcpy(dst, dbuf, sizeof(dbuf)); -- cgit v0.10.2 From e99429232e3622a7e390c3b540c4971b1ccf75c8 Mon Sep 17 00:00:00 2001 From: Richard Alpe Date: Mon, 27 Jun 2016 13:34:06 +0200 Subject: tipc: honor msg2addr return value The UDP msg2addr function tipc_udp_msg2addr() can return -EINVAL which prior to this patch was unhanded in the caller. Signed-off-by: Richard Alpe Acked-by: Jon Maloy Acked-by: Ying Xue Signed-off-by: David S. Miller diff --git a/net/tipc/discover.c b/net/tipc/discover.c index ad9d477..6b109a8 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -135,9 +135,12 @@ void tipc_disc_rcv(struct net *net, struct sk_buff *skb, u16 caps = msg_node_capabilities(hdr); bool respond = false; bool dupl_addr = false; + int err; - bearer->media->msg2addr(bearer, &maddr, msg_media_addr(hdr)); + err = bearer->media->msg2addr(bearer, &maddr, msg_media_addr(hdr)); kfree_skb(skb); + if (err) + return; /* Ensure message from node is valid and communication is permitted */ if (net_id != tn->net_id) -- cgit v0.10.2 From bc3a334cc2c49779c90d7057c42c4537cd36256f Mon Sep 17 00:00:00 2001 From: Richard Alpe Date: Mon, 27 Jun 2016 13:34:07 +0200 Subject: tipc: rename udp_port in struct udp_media_addr Context implies that port in struct "udp_media_addr" is referring to a UDP port. Signed-off-by: Richard Alpe Acked-by: Jon Maloy Acked-by: Ying Xue Signed-off-by: David S. Miller diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c index c9cf2be..b016c01 100644 --- a/net/tipc/udp_media.c +++ b/net/tipc/udp_media.c @@ -63,7 +63,7 @@ */ struct udp_media_addr { __be16 proto; - __be16 udp_port; + __be16 port; union { struct in_addr ipv4; struct in6_addr ipv6; @@ -108,9 +108,9 @@ static int tipc_udp_addr2str(struct tipc_media_addr *a, char *buf, int size) struct udp_media_addr *ua = (struct udp_media_addr *)&a->value; if (ntohs(ua->proto) == ETH_P_IP) - snprintf(buf, size, "%pI4:%u", &ua->ipv4, ntohs(ua->udp_port)); + snprintf(buf, size, "%pI4:%u", &ua->ipv4, ntohs(ua->port)); else if (ntohs(ua->proto) == ETH_P_IPV6) - snprintf(buf, size, "%pI6:%u", &ua->ipv6, ntohs(ua->udp_port)); + snprintf(buf, size, "%pI6:%u", &ua->ipv6, ntohs(ua->port)); else pr_err("Invalid UDP media address\n"); return 0; @@ -178,8 +178,8 @@ static int tipc_udp_send_msg(struct net *net, struct sk_buff *skb, skb->dev = rt->dst.dev; ttl = ip4_dst_hoplimit(&rt->dst); udp_tunnel_xmit_skb(rt, ub->ubsock->sk, skb, src->ipv4.s_addr, - dst->ipv4.s_addr, 0, ttl, 0, src->udp_port, - dst->udp_port, false, true); + dst->ipv4.s_addr, 0, ttl, 0, src->port, + dst->port, false, true); #if IS_ENABLED(CONFIG_IPV6) } else { struct dst_entry *ndst; @@ -196,8 +196,8 @@ static int tipc_udp_send_msg(struct net *net, struct sk_buff *skb, ttl = ip6_dst_hoplimit(ndst); err = udp_tunnel6_xmit_skb(ndst, ub->ubsock->sk, skb, ndst->dev, &src->ipv6, - &dst->ipv6, 0, ttl, 0, src->udp_port, - dst->udp_port, false); + &dst->ipv6, 0, ttl, 0, src->port, + dst->port, false); #endif } return err; @@ -292,12 +292,12 @@ err: ip4 = (struct sockaddr_in *)&sa_local; local->proto = htons(ETH_P_IP); - local->udp_port = ip4->sin_port; + local->port = ip4->sin_port; local->ipv4.s_addr = ip4->sin_addr.s_addr; ip4 = (struct sockaddr_in *)&sa_remote; remote->proto = htons(ETH_P_IP); - remote->udp_port = ip4->sin_port; + remote->port = ip4->sin_port; remote->ipv4.s_addr = ip4->sin_addr.s_addr; return 0; @@ -312,13 +312,13 @@ err: return -EINVAL; local->proto = htons(ETH_P_IPV6); - local->udp_port = ip6->sin6_port; + local->port = ip6->sin6_port; memcpy(&local->ipv6, &ip6->sin6_addr, sizeof(struct in6_addr)); ub->ifindex = ip6->sin6_scope_id; ip6 = (struct sockaddr_in6 *)&sa_remote; remote->proto = htons(ETH_P_IPV6); - remote->udp_port = ip6->sin6_port; + remote->port = ip6->sin6_port; memcpy(&remote->ipv6, &ip6->sin6_addr, sizeof(struct in6_addr)); return 0; #endif @@ -386,7 +386,7 @@ static int tipc_udp_enable(struct net *net, struct tipc_bearer *b, err = -EAFNOSUPPORT; goto err; } - udp_conf.local_udp_port = local.udp_port; + udp_conf.local_udp_port = local.port; err = udp_sock_create(net, &udp_conf, &ub->ubsock); if (err) goto err; -- cgit v0.10.2 From 8e07269de1fd09a01a2c00aba68b0c13160a6bac Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Tue, 28 Jun 2016 00:08:11 +0200 Subject: net: ethernet: mvpp2: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c index 868a957..18477fe 100644 --- a/drivers/net/ethernet/marvell/mvpp2.c +++ b/drivers/net/ethernet/marvell/mvpp2.c @@ -699,7 +699,6 @@ struct mvpp2_port { u16 rx_ring_size; struct mvpp2_pcpu_stats __percpu *stats; - struct phy_device *phy_dev; phy_interface_t phy_interface; struct device_node *phy_node; unsigned int link; @@ -4850,7 +4849,7 @@ static irqreturn_t mvpp2_isr(int irq, void *dev_id) static void mvpp2_link_event(struct net_device *dev) { struct mvpp2_port *port = netdev_priv(dev); - struct phy_device *phydev = port->phy_dev; + struct phy_device *phydev = dev->phydev; int status_change = 0; u32 val; @@ -5416,6 +5415,8 @@ static int mvpp2_poll(struct napi_struct *napi, int budget) /* Set hw internals when starting port */ static void mvpp2_start_dev(struct mvpp2_port *port) { + struct net_device *ndev = port->dev; + mvpp2_gmac_max_rx_size_set(port); mvpp2_txp_max_tx_size_set(port); @@ -5425,13 +5426,15 @@ static void mvpp2_start_dev(struct mvpp2_port *port) mvpp2_interrupts_enable(port); mvpp2_port_enable(port); - phy_start(port->phy_dev); + phy_start(ndev->phydev); netif_tx_start_all_queues(port->dev); } /* Set hw internals when stopping port */ static void mvpp2_stop_dev(struct mvpp2_port *port) { + struct net_device *ndev = port->dev; + /* Stop new packets from arriving to RXQs */ mvpp2_ingress_disable(port); @@ -5447,7 +5450,7 @@ static void mvpp2_stop_dev(struct mvpp2_port *port) mvpp2_egress_disable(port); mvpp2_port_disable(port); - phy_stop(port->phy_dev); + phy_stop(ndev->phydev); } /* Return positive if MTU is valid */ @@ -5535,7 +5538,6 @@ static int mvpp2_phy_connect(struct mvpp2_port *port) phy_dev->supported &= PHY_GBIT_FEATURES; phy_dev->advertising = phy_dev->supported; - port->phy_dev = phy_dev; port->link = 0; port->duplex = 0; port->speed = 0; @@ -5545,8 +5547,9 @@ static int mvpp2_phy_connect(struct mvpp2_port *port) static void mvpp2_phy_disconnect(struct mvpp2_port *port) { - phy_disconnect(port->phy_dev); - port->phy_dev = NULL; + struct net_device *ndev = port->dev; + + phy_disconnect(ndev->phydev); } static int mvpp2_open(struct net_device *dev) @@ -5796,13 +5799,12 @@ mvpp2_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) static int mvpp2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { - struct mvpp2_port *port = netdev_priv(dev); int ret; - if (!port->phy_dev) + if (!dev->phydev) return -ENOTSUPP; - ret = phy_mii_ioctl(port->phy_dev, ifr, cmd); + ret = phy_mii_ioctl(dev->phydev, ifr, cmd); if (!ret) mvpp2_link_event(dev); @@ -5815,22 +5817,18 @@ static int mvpp2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) static int mvpp2_ethtool_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct mvpp2_port *port = netdev_priv(dev); - - if (!port->phy_dev) + if (!dev->phydev) return -ENODEV; - return phy_ethtool_gset(port->phy_dev, cmd); + return phy_ethtool_gset(dev->phydev, cmd); } /* Set settings (phy address, speed) for ethtools */ static int mvpp2_ethtool_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct mvpp2_port *port = netdev_priv(dev); - - if (!port->phy_dev) + if (!dev->phydev) return -ENODEV; - return phy_ethtool_sset(port->phy_dev, cmd); + return phy_ethtool_sset(dev->phydev, cmd); } /* Set interrupt coalescing for ethtools */ -- cgit v0.10.2 From fb773e975e80a04cb7c0cfd0c1aea2dca095e968 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Tue, 28 Jun 2016 00:08:12 +0200 Subject: net: ethernet: mvpp2: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c index 18477fe..0b04717 100644 --- a/drivers/net/ethernet/marvell/mvpp2.c +++ b/drivers/net/ethernet/marvell/mvpp2.c @@ -5813,24 +5813,6 @@ static int mvpp2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) /* Ethtool methods */ -/* Get settings (phy address, speed) for ethtools */ -static int mvpp2_ethtool_get_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - if (!dev->phydev) - return -ENODEV; - return phy_ethtool_gset(dev->phydev, cmd); -} - -/* Set settings (phy address, speed) for ethtools */ -static int mvpp2_ethtool_set_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - if (!dev->phydev) - return -ENODEV; - return phy_ethtool_sset(dev->phydev, cmd); -} - /* Set interrupt coalescing for ethtools */ static int mvpp2_ethtool_set_coalesce(struct net_device *dev, struct ethtool_coalesce *c) @@ -5965,13 +5947,13 @@ static const struct net_device_ops mvpp2_netdev_ops = { static const struct ethtool_ops mvpp2_eth_tool_ops = { .get_link = ethtool_op_get_link, - .get_settings = mvpp2_ethtool_get_settings, - .set_settings = mvpp2_ethtool_set_settings, .set_coalesce = mvpp2_ethtool_set_coalesce, .get_coalesce = mvpp2_ethtool_get_coalesce, .get_drvinfo = mvpp2_ethtool_get_drvinfo, .get_ringparam = mvpp2_ethtool_get_ringparam, .set_ringparam = mvpp2_ethtool_set_ringparam, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /* Driver initialization */ -- cgit v0.10.2 From 51d99880879c8911ebcab1ec77ca7da2f4cb470d Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Tue, 28 Jun 2016 02:10:58 -0400 Subject: qed: Fix static checker warnings. Static checker warnings: drivers/net/ethernet/qlogic/qed/qed_int.c:2450 qed_init_cau_sb_entry() warn: always true condition '(cdev->rx_coalesce_usecs <= 255) => (0-255 <= 255)' drivers/net/ethernet/qlogic/qed/qed_int.c:2511 qed_int_cau_conf_sb() warn: always true condition '(p_hwfn->cdev->rx_coalesce_usecs <= 255) => (0-255 <= 255)' .. The data types for rx/tx_coalesce_usecs should be u16. Fixes: commit 722003ac40c2 ("qed: Add support for coalescing config read/update.") Reported-by: Dan Carpenter Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h index 9a63df1..35e5377 100644 --- a/drivers/net/ethernet/qlogic/qed/qed.h +++ b/drivers/net/ethernet/qlogic/qed/qed.h @@ -489,8 +489,8 @@ struct qed_dev { u32 int_mode; enum qed_coalescing_mode int_coalescing_mode; - u8 rx_coalesce_usecs; - u8 tx_coalesce_usecs; + u16 rx_coalesce_usecs; + u16 tx_coalesce_usecs; /* Start Bar offset of first hwfn */ void __iomem *regview; -- cgit v0.10.2 From d2890dea2984749f59bfeffaff7add9030326542 Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Tue, 28 Jun 2016 02:10:59 -0400 Subject: qede: Fix the static checker warnings. Static checker warnings: drivers/net/ethernet/qlogic/qede/qede_ethtool.c:435 qede_get_coalesce() warn: passing casted pointer '&coal->rx_coalesce_usecs' to 'edev->ops->common->get_coalesce()' 32 vs 16. The u32 pointer is being typecasted to u16 which may fail for big-endian platforms. Fixes: d552fa84cb35 ("qede: Add support for coalescing config read/update.") Reported-by: Dan Carpenter Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index 6228482..c5c658a 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -430,11 +430,13 @@ static int qede_get_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) { struct qede_dev *edev = netdev_priv(dev); + u16 rxc, txc; memset(coal, 0, sizeof(struct ethtool_coalesce)); - edev->ops->common->get_coalesce(edev->cdev, - (u16 *)&coal->rx_coalesce_usecs, - (u16 *)&coal->tx_coalesce_usecs); + edev->ops->common->get_coalesce(edev->cdev, &rxc, &txc); + + coal->rx_coalesce_usecs = rxc; + coal->tx_coalesce_usecs = txc; return 0; } -- cgit v0.10.2 From 8a6e9c670341db1ee913e3888cb44a08f18e7489 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 28 Jun 2016 10:30:08 +0200 Subject: net_sched: netem: do not call qdisc_drop() with a NULL skb If skb_unshare() fails, we call qdisc_drop() with a NULL skb, which is no longer supported. Fixes: 520ac30f4551 ("net_sched: drop packets after root qdisc lock is released") Signed-off-by: Eric Dumazet Reported-by: Dan Carpenter Signed-off-by: David S. Miller diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index ccca8ca..6eac3d8 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -487,10 +487,14 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, skb = segs; segs = segs->next; - if (!(skb = skb_unshare(skb, GFP_ATOMIC)) || - (skb->ip_summed == CHECKSUM_PARTIAL && - skb_checksum_help(skb))) { - rc = qdisc_drop(skb, sch, to_free); + skb = skb_unshare(skb, GFP_ATOMIC); + if (unlikely(!skb)) { + qdisc_qstats_drop(sch); + goto finish_segs; + } + if (skb->ip_summed == CHECKSUM_PARTIAL && + skb_checksum_help(skb)) { + qdisc_drop(skb, sch, to_free); goto finish_segs; } -- cgit v0.10.2 From f786f3564c4f02d5026189d4e4fc5e544d125a0c Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Tue, 28 Jun 2016 23:59:44 +0200 Subject: net: ethernet: lpc_eth: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c index b1ce7aa..b003501 100644 --- a/drivers/net/ethernet/nxp/lpc_eth.c +++ b/drivers/net/ethernet/nxp/lpc_eth.c @@ -425,7 +425,6 @@ struct netdata_local { unsigned int last_tx_idx; unsigned int num_used_tx_buffs; struct mii_bus *mii_bus; - struct phy_device *phy_dev; struct clk *clk; dma_addr_t dma_buff_base_p; void *dma_buff_base_v; @@ -750,7 +749,7 @@ static int lpc_mdio_reset(struct mii_bus *bus) static void lpc_handle_link_change(struct net_device *ndev) { struct netdata_local *pldat = netdev_priv(ndev); - struct phy_device *phydev = pldat->phy_dev; + struct phy_device *phydev = ndev->phydev; unsigned long flags; bool status_change = false; @@ -814,7 +813,6 @@ static int lpc_mii_probe(struct net_device *ndev) pldat->link = 0; pldat->speed = 0; pldat->duplex = -1; - pldat->phy_dev = phydev; phy_attached_info(phydev); @@ -1048,8 +1046,8 @@ static int lpc_eth_close(struct net_device *ndev) napi_disable(&pldat->napi); netif_stop_queue(ndev); - if (pldat->phy_dev) - phy_stop(pldat->phy_dev); + if (ndev->phydev) + phy_stop(ndev->phydev); spin_lock_irqsave(&pldat->lock, flags); __lpc_eth_reset(pldat); @@ -1186,7 +1184,7 @@ static void lpc_eth_set_multicast_list(struct net_device *ndev) static int lpc_eth_ioctl(struct net_device *ndev, struct ifreq *req, int cmd) { struct netdata_local *pldat = netdev_priv(ndev); - struct phy_device *phydev = pldat->phy_dev; + struct phy_device *phydev = ndev->phydev; if (!netif_running(ndev)) return -EINVAL; @@ -1207,14 +1205,14 @@ static int lpc_eth_open(struct net_device *ndev) __lpc_eth_clock_enable(pldat, true); /* Suspended PHY makes LPC ethernet core block, so resume now */ - phy_resume(pldat->phy_dev); + phy_resume(ndev->phydev); /* Reset and initialize */ __lpc_eth_reset(pldat); __lpc_eth_init(pldat); /* schedule a link state check */ - phy_start(pldat->phy_dev); + phy_start(ndev->phydev); netif_start_queue(ndev); napi_enable(&pldat->napi); @@ -1250,8 +1248,7 @@ static void lpc_eth_ethtool_setmsglevel(struct net_device *ndev, u32 level) static int lpc_eth_ethtool_getsettings(struct net_device *ndev, struct ethtool_cmd *cmd) { - struct netdata_local *pldat = netdev_priv(ndev); - struct phy_device *phydev = pldat->phy_dev; + struct phy_device *phydev = ndev->phydev; if (!phydev) return -EOPNOTSUPP; @@ -1262,8 +1259,7 @@ static int lpc_eth_ethtool_getsettings(struct net_device *ndev, static int lpc_eth_ethtool_setsettings(struct net_device *ndev, struct ethtool_cmd *cmd) { - struct netdata_local *pldat = netdev_priv(ndev); - struct phy_device *phydev = pldat->phy_dev; + struct phy_device *phydev = ndev->phydev; if (!phydev) return -EOPNOTSUPP; @@ -1460,7 +1456,7 @@ static int lpc_eth_drv_probe(struct platform_device *pdev) netdev_info(ndev, "LPC mac at 0x%08x irq %d\n", res->start, ndev->irq); - phydev = pldat->phy_dev; + phydev = ndev->phydev; device_init_wakeup(&pdev->dev, 1); device_set_wakeup_enable(&pdev->dev, 0); -- cgit v0.10.2 From cb90d3e15de20e44ec77c92c0a1cb2ae47a5a27f Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Tue, 28 Jun 2016 23:59:45 +0200 Subject: net: ethernet: lpc_eth: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c index b003501..01b50ff 100644 --- a/drivers/net/ethernet/nxp/lpc_eth.c +++ b/drivers/net/ethernet/nxp/lpc_eth.c @@ -1245,35 +1245,13 @@ static void lpc_eth_ethtool_setmsglevel(struct net_device *ndev, u32 level) pldat->msg_enable = level; } -static int lpc_eth_ethtool_getsettings(struct net_device *ndev, - struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = ndev->phydev; - - if (!phydev) - return -EOPNOTSUPP; - - return phy_ethtool_gset(phydev, cmd); -} - -static int lpc_eth_ethtool_setsettings(struct net_device *ndev, - struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = ndev->phydev; - - if (!phydev) - return -EOPNOTSUPP; - - return phy_ethtool_sset(phydev, cmd); -} - static const struct ethtool_ops lpc_eth_ethtool_ops = { .get_drvinfo = lpc_eth_ethtool_getdrvinfo, - .get_settings = lpc_eth_ethtool_getsettings, - .set_settings = lpc_eth_ethtool_setsettings, .get_msglevel = lpc_eth_ethtool_getmsglevel, .set_msglevel = lpc_eth_ethtool_setmsglevel, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops lpc_netdev_ops = { -- cgit v0.10.2 From f568adac7d1a50a7412a902ebc831730a9b80bf9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Tue, 7 Jun 2016 21:10:18 +0200 Subject: brcmfmac: slightly simplify building interface combinations MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This change reorders some operations in brcmf_setup_ifmodes in hope to make it simpler: 1) It allocates arrays right before filling them. This way it's easier to follow requested array length as it's immediately followed by code filling it. It's easier to check e.g. why we need 4 entries for P2P. Other than that it deduplicates some checks (e.g. for P2P). 2) It reorders code to first prepare limits and then define a new combo. Previously this was mixed (e.g. we were setting num of channels before preparing limits). 3) It modifies mbss code to use i variable just like other combos do. Signed-off-by: Rafał Miłecki Acked-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index 264bd63..a5db953 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -6284,29 +6284,15 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp) if (!combo) goto err; - c0_limits = kcalloc(p2p ? 3 : 2, sizeof(*c0_limits), GFP_KERNEL); - if (!c0_limits) - goto err; - - if (p2p) { - p2p_limits = kcalloc(4, sizeof(*p2p_limits), GFP_KERNEL); - if (!p2p_limits) - goto err; - } - - if (mbss) { - mbss_limits = kcalloc(1, sizeof(*mbss_limits), GFP_KERNEL); - if (!mbss_limits) - goto err; - } - wiphy->interface_modes = BIT(NL80211_IFTYPE_STATION) | BIT(NL80211_IFTYPE_ADHOC) | BIT(NL80211_IFTYPE_AP); c = 0; i = 0; - combo[c].num_different_channels = 1; + c0_limits = kcalloc(p2p ? 3 : 2, sizeof(*c0_limits), GFP_KERNEL); + if (!c0_limits) + goto err; c0_limits[i].max = 1; c0_limits[i++].types = BIT(NL80211_IFTYPE_STATION); if (p2p) { @@ -6324,6 +6310,7 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp) c0_limits[i].max = 1; c0_limits[i++].types = BIT(NL80211_IFTYPE_AP); } + combo[c].num_different_channels = 1; combo[c].max_interfaces = i; combo[c].n_limits = i; combo[c].limits = c0_limits; @@ -6331,7 +6318,9 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp) if (p2p) { c++; i = 0; - combo[c].num_different_channels = 1; + p2p_limits = kcalloc(4, sizeof(*p2p_limits), GFP_KERNEL); + if (!p2p_limits) + goto err; p2p_limits[i].max = 1; p2p_limits[i++].types = BIT(NL80211_IFTYPE_STATION); p2p_limits[i].max = 1; @@ -6340,6 +6329,7 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp) p2p_limits[i++].types = BIT(NL80211_IFTYPE_P2P_CLIENT); p2p_limits[i].max = 1; p2p_limits[i++].types = BIT(NL80211_IFTYPE_P2P_DEVICE); + combo[c].num_different_channels = 1; combo[c].max_interfaces = i; combo[c].n_limits = i; combo[c].limits = p2p_limits; @@ -6347,14 +6337,19 @@ static int brcmf_setup_ifmodes(struct wiphy *wiphy, struct brcmf_if *ifp) if (mbss) { c++; + i = 0; + mbss_limits = kcalloc(1, sizeof(*mbss_limits), GFP_KERNEL); + if (!mbss_limits) + goto err; + mbss_limits[i].max = 4; + mbss_limits[i++].types = BIT(NL80211_IFTYPE_AP); combo[c].beacon_int_infra_match = true; combo[c].num_different_channels = 1; - mbss_limits[0].max = 4; - mbss_limits[0].types = BIT(NL80211_IFTYPE_AP); combo[c].max_interfaces = 4; - combo[c].n_limits = 1; + combo[c].n_limits = i; combo[c].limits = mbss_limits; } + wiphy->n_iface_combinations = n_combos; wiphy->iface_combinations = combo; return 0; -- cgit v0.10.2 From 452fa86e98061ebd528dd79e22befd5f87c83269 Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Wed, 8 Jun 2016 01:38:53 +0530 Subject: libertas_tf: Remove create_workqueue alloc_workqueue replaces deprecated create_workqueue(). A dedicated workqueue has been used since the workitem (viz &priv->cmd_work per priv, which maps to lbtf_cmd_work) is involved in actual command processing and may be used on a memory reclaim path. The workitems require forward progress under memory pressure and hence, WQ_MEM_RECLAIM has been set. Since there are only a fixed number of work items, explicit concurrency limit is unnecessary here. Signed-off-by: Bhaktipriya Shridhar Acked-by: Tejun Heo Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/libertas_tf/main.c b/drivers/net/wireless/marvell/libertas_tf/main.c index 75bf0c8..54e426c 100644 --- a/drivers/net/wireless/marvell/libertas_tf/main.c +++ b/drivers/net/wireless/marvell/libertas_tf/main.c @@ -735,7 +735,7 @@ EXPORT_SYMBOL_GPL(lbtf_bcn_sent); static int __init lbtf_init_module(void) { lbtf_deb_enter(LBTF_DEB_MAIN); - lbtf_wq = create_workqueue("libertastf"); + lbtf_wq = alloc_workqueue("libertastf", WQ_MEM_RECLAIM, 0); if (lbtf_wq == NULL) { printk(KERN_ERR "libertastf: couldn't create workqueue\n"); return -ENOMEM; -- cgit v0.10.2 From e4ac0a8ac8ba3cbcafcc3c6142c022245439d057 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Thu, 9 Jun 2016 14:38:47 -0400 Subject: rtl8xxxu: Add bit definitions for REG_USB_SPECIAL_OPTION Documentation for enabling USB aggregation and whether to select interrupt or bulk delivery of interrupt events. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h index b0e0c64..bff883c 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h @@ -1052,6 +1052,10 @@ #define USB_HIMR_ROK BIT(0) /* Receive DMA OK Interrupt */ #define REG_USB_SPECIAL_OPTION 0xfe55 +#define USB_SPEC_USB_AGG_ENABLE BIT(3) /* Enable USB aggregation */ +#define USB_SPEC_INT_BULK_SELECT BIT(4) /* Use interrupt endpoint to + deliver interrupt packet. + 0: Use int, 1: use bulk */ #define REG_USB_HRPWM 0xfe58 #define REG_USB_DMA_AGG_TO 0xfe5b #define REG_USB_AGG_TO 0xfe5c -- cgit v0.10.2 From 08eca32ebc891e9501802522dc4bff5136edde86 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Thu, 9 Jun 2016 14:38:48 -0400 Subject: rtl8xxxu: Add additional documentation for RX DMA registers This also renames REG_USB_AGG_{TO,TH} to REG_USB_AGG_{TIMEOUT,THRESH} Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h index bff883c..921c565 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_regs.h @@ -405,7 +405,11 @@ #define REG_DWBCN1_CTRL_8723B 0x0228 /* 0x0280 ~ 0x02FF RXDMA Configuration */ -#define REG_RXDMA_AGG_PG_TH 0x0280 +#define REG_RXDMA_AGG_PG_TH 0x0280 /* 0-7 : USB DMA size bits + 8-14: USB DMA timeout + 15 : Aggregation enable + Only seems to be used + on 8723bu/8192eu */ #define RXDMA_USB_AGG_ENABLE BIT(31) #define REG_RXPKT_NUM 0x0284 #define RXPKT_NUM_RXDMA_IDLE BIT(17) @@ -1058,8 +1062,8 @@ 0: Use int, 1: use bulk */ #define REG_USB_HRPWM 0xfe58 #define REG_USB_DMA_AGG_TO 0xfe5b -#define REG_USB_AGG_TO 0xfe5c -#define REG_USB_AGG_TH 0xfe5d +#define REG_USB_AGG_TIMEOUT 0xfe5c +#define REG_USB_AGG_THRESH 0xfe5d #define REG_NORMAL_SIE_VID 0xfe60 /* 0xfe60 - 0xfe61 */ #define REG_NORMAL_SIE_PID 0xfe62 /* 0xfe62 - 0xfe63 */ -- cgit v0.10.2 From 2b9c9f52dc03b298c845def62ea890a2d77e9f21 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Thu, 9 Jun 2016 14:38:49 -0400 Subject: rtl8xxxu: tuse %*ph to dump buffers Use %*ph specifier to dump small buffers in hex format instead of doing this byte-by-byte. Signed-off-by: Andy Shevchenko Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c index 2c86b55..eefb7ca 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c @@ -413,13 +413,8 @@ static int rtl8192cu_parse_efuse(struct rtl8xxxu_priv *priv) dev_info(&priv->udev->dev, "%s: dumping efuse (0x%02zx bytes):\n", __func__, sizeof(struct rtl8192cu_efuse)); - for (i = 0; i < sizeof(struct rtl8192cu_efuse); i += 8) { - dev_info(&priv->udev->dev, "%02x: " - "%02x %02x %02x %02x %02x %02x %02x %02x\n", i, - raw[i], raw[i + 1], raw[i + 2], - raw[i + 3], raw[i + 4], raw[i + 5], - raw[i + 6], raw[i + 7]); - } + for (i = 0; i < sizeof(struct rtl8192cu_efuse); i += 8) + dev_info(&priv->udev->dev, "%02x: %8ph\n", i, &raw[i]); } return 0; } diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c index fe19ace..65c079a 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c @@ -622,13 +622,8 @@ static int rtl8192eu_parse_efuse(struct rtl8xxxu_priv *priv) dev_info(&priv->udev->dev, "%s: dumping efuse (0x%02zx bytes):\n", __func__, sizeof(struct rtl8192eu_efuse)); - for (i = 0; i < sizeof(struct rtl8192eu_efuse); i += 8) { - dev_info(&priv->udev->dev, "%02x: " - "%02x %02x %02x %02x %02x %02x %02x %02x\n", i, - raw[i], raw[i + 1], raw[i + 2], - raw[i + 3], raw[i + 4], raw[i + 5], - raw[i + 6], raw[i + 7]); - } + for (i = 0; i < sizeof(struct rtl8192eu_efuse); i += 8) + dev_info(&priv->udev->dev, "%02x: %8ph\n", i, &raw[i]); } return 0; } diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c index 4186e7c..9d45afb 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723b.c @@ -466,13 +466,8 @@ static int rtl8723bu_parse_efuse(struct rtl8xxxu_priv *priv) dev_info(&priv->udev->dev, "%s: dumping efuse (0x%02zx bytes):\n", __func__, sizeof(struct rtl8723bu_efuse)); - for (i = 0; i < sizeof(struct rtl8723bu_efuse); i += 8) { - dev_info(&priv->udev->dev, "%02x: " - "%02x %02x %02x %02x %02x %02x %02x %02x\n", i, - raw[i], raw[i + 1], raw[i + 2], - raw[i + 3], raw[i + 4], raw[i + 5], - raw[i + 6], raw[i + 7]); - } + for (i = 0; i < sizeof(struct rtl8723bu_efuse); i += 8) + dev_info(&priv->udev->dev, "%02x: %8ph\n", i, &raw[i]); } return 0; -- cgit v0.10.2 From 6edc119ed3b5e860535d49852f8cc8e5be95538d Mon Sep 17 00:00:00 2001 From: Bruno Herrera Date: Thu, 9 Jun 2016 21:46:46 -0300 Subject: wlcore: sdio: Fix crash on wlcore_probe_of when failing to parse/map irq pdev_data pointer is being freed with kfree but the pointer is not dynamic allocated. Signed-off-by: Bruno Herrera Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ti/wlcore/sdio.c b/drivers/net/wireless/ti/wlcore/sdio.c index c172da5..5839acb 100644 --- a/drivers/net/wireless/ti/wlcore/sdio.c +++ b/drivers/net/wireless/ti/wlcore/sdio.c @@ -241,7 +241,6 @@ static int wlcore_probe_of(struct device *dev, int *irq, *irq = irq_of_parse_and_map(np, 0); if (!*irq) { dev_err(dev, "No irq in platform data\n"); - kfree(pdev_data); return -EINVAL; } -- cgit v0.10.2 From f52b041aed77592862c97726b98d78e8dccd72c9 Mon Sep 17 00:00:00 2001 From: Pavel Andrianov Date: Wed, 15 Jun 2016 15:34:03 +0400 Subject: libertas: Add spinlock to avoid race condition lbs_mac_event_disconnected may free priv->currenttxskb while lbs_hard_start_xmit accesses to it. The patch adds a spinlock for mutual exclusion. Tested on OLPC XO-1 (usb8388) and XO-1.5 (sd8686) with v4.7-rc3. Confirmed that lbs_mac_event_disconnected is being called on the station when hostapd on access point is given SIGHUP. Signed-off-by: Pavel Tested-by: James Cameron Acked-by: Vaishali Thakkar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/libertas/cmdresp.c b/drivers/net/wireless/marvell/libertas/cmdresp.c index c95bf6d..c753e36 100644 --- a/drivers/net/wireless/marvell/libertas/cmdresp.c +++ b/drivers/net/wireless/marvell/libertas/cmdresp.c @@ -27,6 +27,8 @@ void lbs_mac_event_disconnected(struct lbs_private *priv, bool locally_generated) { + unsigned long flags; + if (priv->connect_status != LBS_CONNECTED) return; @@ -46,9 +48,11 @@ void lbs_mac_event_disconnected(struct lbs_private *priv, netif_carrier_off(priv->dev); /* Free Tx and Rx packets */ + spin_lock_irqsave(&priv->driver_lock, flags); kfree_skb(priv->currenttxskb); priv->currenttxskb = NULL; priv->tx_pending_len = 0; + spin_unlock_irqrestore(&priv->driver_lock, flags); priv->connect_status = LBS_DISCONNECTED; -- cgit v0.10.2 From 08aba42fcc7eea5e24558b3c59d1b9e86c3b9e75 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 15 Jun 2016 23:30:43 +0200 Subject: rtlwifi: use s8 instead of char Compiling the rtlwifi drivers for ARM with gcc -Wextra warns about lots of incorrect code that results from 'char' being unsigned here, e.g. realtek/rtlwifi/rc.c:113:18: error: comparison is always true due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8188ee/dm.c:1070:22: error: comparison is always false due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8192ce/trx.c:54:16: error: comparison is always false due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8192cu/mac.c:601:16: error: comparison is always false due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8192de/trx.c:53:16: error: comparison is always false due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8192ee/phy.c:1268:12: error: comparison is always true due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8192se/rf.c:150:20: error: comparison is always false due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8723be/dm.c:877:29: error: comparison is always false due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8723be/phy.c:386:16: error: comparison is always true due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8821ae/dm.c:1514:38: error: comparison is always false due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8821ae/phy.c:1558:11: error: comparison is always false due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8821ae/phy.c:386:24: error: comparison is always false due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/rtl8821ae/trx.c:55:12: error: comparison is always false due to limited range of data type [-Werror=type-limits] realtek/rtlwifi/stats.c:31:16: error: comparison is always false due to limited range of data type [-Werror=type-limits] This patch changes all uses of 'char' in this driver that refer to 8-bit integers to use 's8' instead, which is signed on all architectures. Signed-off-by: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c index b660c21..91cc139 100644 --- a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c +++ b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.c @@ -901,7 +901,7 @@ void exhalbtc_stack_update_profile_info(void) { } -void exhalbtc_update_min_bt_rssi(char bt_rssi) +void exhalbtc_update_min_bt_rssi(s8 bt_rssi) { struct btc_coexist *btcoexist = &gl_bt_coexist; diff --git a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.h b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.h index 3cbe34c..3d308eb 100644 --- a/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.h +++ b/drivers/net/wireless/realtek/rtlwifi/btcoexist/halbtcoutsrc.h @@ -433,7 +433,7 @@ struct btc_stack_info { u8 num_of_hid; bool pan_exist; bool unknown_acl_exist; - char min_bt_rssi; + s8 min_bt_rssi; }; struct btc_statistics { @@ -537,7 +537,7 @@ void exhalbtc_dbg_control(struct btc_coexist *btcoexist, u8 code, u8 len, void exhalbtc_stack_update_profile_info(void); void exhalbtc_set_hci_version(u16 hci_version); void exhalbtc_set_bt_patch_version(u16 bt_hci_version, u16 bt_patch_version); -void exhalbtc_update_min_bt_rssi(char bt_rssi); +void exhalbtc_update_min_bt_rssi(s8 bt_rssi); void exhalbtc_set_bt_exist(bool bt_exist); void exhalbtc_set_chip_type(u8 chip_type); void exhalbtc_set_ant_num(struct rtl_priv *rtlpriv, u8 type, u8 ant_num); diff --git a/drivers/net/wireless/realtek/rtlwifi/rc.c b/drivers/net/wireless/realtek/rtlwifi/rc.c index 1aca777..ce8621a 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rc.c +++ b/drivers/net/wireless/realtek/rtlwifi/rc.c @@ -94,7 +94,7 @@ static void _rtl_rc_rate_set_series(struct rtl_priv *rtlpriv, struct ieee80211_sta *sta, struct ieee80211_tx_rate *rate, struct ieee80211_tx_rate_control *txrc, - u8 tries, char rix, int rtsctsenable, + u8 tries, s8 rix, int rtsctsenable, bool not_data) { struct rtl_mac *mac = rtl_mac(rtlpriv); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/dm.c index db9a782..6e7b673 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/dm.c @@ -886,7 +886,7 @@ static void dm_txpower_track_cb_therm(struct ieee80211_hw *hw) u8 thermalvalue_avg_count = 0; u32 thermalvalue_avg = 0; long ele_d, temp_cck; - char ofdm_index[2], cck_index = 0, + s8 ofdm_index[2], cck_index = 0, ofdm_index_old[2] = {0, 0}, cck_index_old = 0; int i = 0; /*bool is2t = false;*/ @@ -898,7 +898,7 @@ static void dm_txpower_track_cb_therm(struct ieee80211_hw *hw) /*0.1 the following TWO tables decide the *final index of OFDM/CCK swing table */ - char delta_swing_table_idx[2][15] = { + s8 delta_swing_table_idx[2][15] = { {0, 0, 2, 3, 4, 4, 5, 6, 7, 7, 8, 9, 10, 10, 11}, {0, 0, -1, -2, -3, -4, -4, -4, -4, -5, -7, -8, -9, -9, -10} }; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c index 1170106..1016ad4 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c @@ -59,7 +59,7 @@ static void _rtl88ee_query_rxphystatus(struct ieee80211_hw *hw, struct phy_status_rpt *phystrpt = (struct phy_status_rpt *)p_drvinfo; struct rtl_dm *rtldm = rtl_dm(rtl_priv(hw)); - char rx_pwr_all = 0, rx_pwr[4]; + s8 rx_pwr_all = 0, rx_pwr[4]; u8 rf_rx_num = 0, evm, pwdb_all; u8 i, max_spatial_stream; u32 rssi, total_rssi = 0; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h index 5a24d19..9a1c208 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.h @@ -593,8 +593,8 @@ struct rx_fwinfo_88e { u8 pwdb_all; u8 cfosho[4]; u8 cfotail[4]; - char rxevm[2]; - char rxsnr[4]; + s8 rxevm[2]; + s8 rxsnr[4]; u8 pdsnr[2]; u8 csi_current[2]; u8 csi_target[2]; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/dm_common.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/dm_common.h index 4422e31..6a72d0c 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/dm_common.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/dm_common.h @@ -135,7 +135,7 @@ void rtl92c_dm_init_edca_turbo(struct ieee80211_hw *hw); void rtl92c_dm_check_txpower_tracking(struct ieee80211_hw *hw); void rtl92c_dm_init_rate_adaptive_mask(struct ieee80211_hw *hw); void rtl92c_dm_rf_saving(struct ieee80211_hw *hw, u8 bforce_in_normal); -void rtl92c_phy_ap_calibrate(struct ieee80211_hw *hw, char delta); +void rtl92c_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta); void rtl92c_phy_lc_calibrate(struct ieee80211_hw *hw); void rtl92c_phy_iq_calibrate(struct ieee80211_hw *hw, bool recovery); void rtl92c_dm_dynamic_txpower(struct ieee80211_hw *hw); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c index 77e61b1..24162e0 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c @@ -1353,7 +1353,7 @@ static void _rtl92c_phy_iq_calibrate(struct ieee80211_hw *hw, } static void _rtl92c_phy_ap_calibrate(struct ieee80211_hw *hw, - char delta, bool is2t) + s8 delta, bool is2t) { } @@ -1518,7 +1518,7 @@ void rtl92c_phy_lc_calibrate(struct ieee80211_hw *hw) } EXPORT_SYMBOL(rtl92c_phy_lc_calibrate); -void rtl92c_phy_ap_calibrate(struct ieee80211_hw *hw, char delta) +void rtl92c_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta) { struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_phy *rtlphy = &(rtlpriv->phy); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.h index 64bc49f..2024125 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.h @@ -210,7 +210,7 @@ u8 rtl92c_phy_sw_chnl(struct ieee80211_hw *hw); void rtl92c_phy_iq_calibrate(struct ieee80211_hw *hw, bool b_recovery); void rtl92c_phy_set_beacon_hw_reg(struct ieee80211_hw *hw, u16 beaconinterval); -void rtl92c_phy_ap_calibrate(struct ieee80211_hw *hw, char delta); +void rtl92c_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta); void rtl92c_phy_lc_calibrate(struct ieee80211_hw *hw); void rtl92c_phy_set_rfpath_switch(struct ieee80211_hw *hw, bool bmain); bool rtl92c_phy_config_rf_with_headerfile(struct ieee80211_hw *hw, diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/phy.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/phy.h index e5e1353..dadc02b 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/phy.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/phy.h @@ -102,7 +102,7 @@ void rtl92c_phy_sw_chnl_callback(struct ieee80211_hw *hw); u8 rtl92c_phy_sw_chnl(struct ieee80211_hw *hw); void rtl92c_phy_iq_calibrate(struct ieee80211_hw *hw, bool b_recovery); void rtl92c_phy_set_beacon_hw_reg(struct ieee80211_hw *hw, u16 beaconinterval); -void rtl92c_phy_ap_calibrate(struct ieee80211_hw *hw, char delta); +void rtl92c_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta); void rtl92c_phy_lc_calibrate(struct ieee80211_hw *hw); void _rtl92ce_phy_lc_calibrate(struct ieee80211_hw *hw, bool is2t); void rtl92c_phy_set_rfpath_switch(struct ieee80211_hw *hw, bool bmain); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c index 84ddd4d..93d3fba 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c @@ -49,7 +49,7 @@ static u8 _rtl92ce_map_hwqueue_to_fwqueue(struct sk_buff *skb, u8 hw_queue) return skb->priority; } -static u8 _rtl92c_query_rxpwrpercentage(char antpower) +static u8 _rtl92c_query_rxpwrpercentage(s8 antpower) { if ((antpower <= -100) || (antpower >= 20)) return 0; @@ -59,9 +59,9 @@ static u8 _rtl92c_query_rxpwrpercentage(char antpower) return 100 + antpower; } -static u8 _rtl92c_evm_db_to_percentage(char value) +static u8 _rtl92c_evm_db_to_percentage(s8 value) { - char ret_val; + s8 ret_val; ret_val = value; if (ret_val >= 0) diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.h index 4bec4b0..6073045 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.h @@ -537,8 +537,8 @@ struct rx_fwinfo_92c { u8 pwdb_all; u8 cfosho[4]; u8 cfotail[4]; - char rxevm[2]; - char rxsnr[4]; + s8 rxevm[2]; + s8 rxsnr[4]; u8 pdsnr[2]; u8 csi_current[2]; u8 csi_target[2]; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c index 0357133..68ca734 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.c @@ -596,7 +596,7 @@ void rtl92c_set_min_space(struct ieee80211_hw *hw, bool is2T) /*==============================================================*/ -static u8 _rtl92c_query_rxpwrpercentage(char antpower) +static u8 _rtl92c_query_rxpwrpercentage(s8 antpower) { if ((antpower <= -100) || (antpower >= 20)) return 0; @@ -606,9 +606,9 @@ static u8 _rtl92c_query_rxpwrpercentage(char antpower) return 100 + antpower; } -static u8 _rtl92c_evm_db_to_percentage(char value) +static u8 _rtl92c_evm_db_to_percentage(s8 value) { - char ret_val; + s8 ret_val; ret_val = value; if (ret_val >= 0) diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.h index 553a4bf..20a49ec 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/mac.h @@ -79,8 +79,8 @@ struct rx_fwinfo_92c { u8 pwdb_all; u8 cfosho[4]; u8 cfotail[4]; - char rxevm[2]; - char rxsnr[4]; + s8 rxevm[2]; + s8 rxsnr[4]; u8 pdsnr[2]; u8 csi_current[2]; u8 csi_target[2]; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.c index 7810fe8..d334d2a 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.c @@ -2695,7 +2695,7 @@ void rtl92d_phy_lc_calibrate(struct ieee80211_hw *hw) RTPRINT(rtlpriv, FINIT, INIT_IQK, "LCK:Finish!!!\n"); } -void rtl92d_phy_ap_calibrate(struct ieee80211_hw *hw, char delta) +void rtl92d_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta) { return; } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.h index 48d5c68..8115bf4 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/phy.h @@ -160,7 +160,7 @@ void rtl92d_phy_config_maccoexist_rfpage(struct ieee80211_hw *hw); bool rtl92d_phy_check_poweroff(struct ieee80211_hw *hw); void rtl92d_phy_lc_calibrate(struct ieee80211_hw *hw); void rtl92d_update_bbrf_configuration(struct ieee80211_hw *hw); -void rtl92d_phy_ap_calibrate(struct ieee80211_hw *hw, char delta); +void rtl92d_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta); void rtl92d_phy_iq_calibrate(struct ieee80211_hw *hw); void rtl92d_phy_reset_iqk_result(struct ieee80211_hw *hw); void rtl92d_release_cckandrw_pagea_ctl(struct ieee80211_hw *hw, diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c index 1feaa62..274b0e4 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c @@ -48,7 +48,7 @@ static u8 _rtl92de_map_hwqueue_to_fwqueue(struct sk_buff *skb, u8 hw_queue) return skb->priority; } -static u8 _rtl92d_query_rxpwrpercentage(char antpower) +static u8 _rtl92d_query_rxpwrpercentage(s8 antpower) { if ((antpower <= -100) || (antpower >= 20)) return 0; @@ -58,9 +58,9 @@ static u8 _rtl92d_query_rxpwrpercentage(char antpower) return 100 + antpower; } -static u8 _rtl92d_evm_db_to_percentage(char value) +static u8 _rtl92d_evm_db_to_percentage(s8 value) { - char ret_val = value; + s8 ret_val = value; if (ret_val >= 0) ret_val = 0; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h index fb5cf06..194d99f 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.h @@ -554,8 +554,8 @@ struct rx_fwinfo_92d { u8 pwdb_all; u8 cfosho[4]; u8 cfotail[4]; - char rxevm[2]; - char rxsnr[4]; + s8 rxevm[2]; + s8 rxsnr[4]; u8 pdsnr[2]; u8 csi_current[2]; u8 csi_target[2]; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c index c2bf8d1..b7184f5 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c @@ -547,7 +547,7 @@ static void _rtl92ee_phy_store_txpower_by_rate_base(struct ieee80211_hw *hw) static void _phy_convert_txpower_dbm_to_relative_value(u32 *data, u8 start, u8 end, u8 base) { - char i = 0; + s8 i = 0; u8 tmp = 0; u32 temp_data = 0; @@ -1189,7 +1189,7 @@ static u8 _rtl92ee_get_txpower_by_rate(struct ieee80211_hw *hw, struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_phy *rtlphy = &rtlpriv->phy; u8 shift = 0, sec, tx_num; - char diff = 0; + s8 diff = 0; sec = _rtl92ee_phy_get_ratesection_intxpower_byrate(rf, rate); tx_num = RF_TX_NUM_NONIMPLEMENT; @@ -1265,14 +1265,14 @@ static u8 _rtl92ee_get_txpower_index(struct ieee80211_hw *hw, "Illegal channel!!\n"); } - if (IS_CCK_RATE(rate)) + if (IS_CCK_RATE((s8)rate)) tx_power = rtlefuse->txpwrlevel_cck[rfpath][index]; else if (DESC92C_RATE6M <= rate) tx_power = rtlefuse->txpwrlevel_ht40_1s[rfpath][index]; /* OFDM-1T*/ if (DESC92C_RATE6M <= rate && rate <= DESC92C_RATE54M && - !IS_CCK_RATE(rate)) + !IS_CCK_RATE((s8)rate)) tx_power += rtlefuse->txpwr_legacyhtdiff[rfpath][TX_1S]; /* BW20-1S, BW20-2S */ @@ -2969,7 +2969,7 @@ void rtl92ee_phy_lc_calibrate(struct ieee80211_hw *hw) rtlphy->lck_inprogress = false; } -void rtl92ee_phy_ap_calibrate(struct ieee80211_hw *hw, char delta) +void rtl92ee_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta) { } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.h index c6e97c8..49bd0e5 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.h @@ -141,7 +141,7 @@ void rtl92ee_phy_set_bw_mode(struct ieee80211_hw *hw, void rtl92ee_phy_sw_chnl_callback(struct ieee80211_hw *hw); u8 rtl92ee_phy_sw_chnl(struct ieee80211_hw *hw); void rtl92ee_phy_iq_calibrate(struct ieee80211_hw *hw, bool b_recovery); -void rtl92ee_phy_ap_calibrate(struct ieee80211_hw *hw, char delta); +void rtl92ee_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta); void rtl92ee_phy_lc_calibrate(struct ieee80211_hw *hw); void rtl92ee_phy_set_rfpath_switch(struct ieee80211_hw *hw, bool bmain); bool rtl92ee_phy_config_rf_with_headerfile(struct ieee80211_hw *hw, diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c index 35e6bf7..582b1fa 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c @@ -56,7 +56,7 @@ static void _rtl92ee_query_rxphystatus(struct ieee80211_hw *hw, { struct rtl_priv *rtlpriv = rtl_priv(hw); struct phy_status_rpt *p_phystrpt = (struct phy_status_rpt *)p_drvinfo; - char rx_pwr_all = 0, rx_pwr[4]; + s8 rx_pwr_all = 0, rx_pwr[4]; u8 rf_rx_num = 0, evm, pwdb_all; u8 i, max_spatial_stream; u32 rssi, total_rssi = 0; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.h index a4c3834..8053d1b 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.h @@ -650,8 +650,8 @@ struct rx_fwinfo { u8 pwdb_all; u8 cfosho[4]; u8 cfotail[4]; - char rxevm[2]; - char rxsnr[4]; + s8 rxevm[2]; + s8 rxsnr[4]; u8 pdsnr[2]; u8 csi_current[2]; u8 csi_target[2]; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/rf.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/rf.c index 9475aa2..34e88a3 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/rf.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/rf.c @@ -137,7 +137,7 @@ static void _rtl92s_set_antennadiff(struct ieee80211_hw *hw, struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_phy *rtlphy = &(rtlpriv->phy); - char ant_pwr_diff = 0; + s8 ant_pwr_diff = 0; u32 u4reg_val = 0; if (rtlphy->rf_type == RF_2T2R) { diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.h index 32970bf..43d4c79 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.h @@ -522,8 +522,8 @@ struct rx_fwinfo_8723e { u8 pwdb_all; u8 cfosho[4]; u8 cfotail[4]; - char rxevm[2]; - char rxsnr[4]; + s8 rxevm[2]; + s8 rxsnr[4]; u8 pdsnr[2]; u8 csi_current[2]; u8 csi_target[2]; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/dm.c index 3a81cdb..6f07cd6 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/dm.c @@ -758,11 +758,11 @@ static void rtl8723be_dm_txpower_tracking_callback_thermalmeter( u8 ofdm_min_index = 6; u8 index_for_channel = 0; - char delta_swing_table_idx_tup_a[TXSCALE_TABLE_SIZE] = { + s8 delta_swing_table_idx_tup_a[TXSCALE_TABLE_SIZE] = { 0, 0, 1, 2, 2, 2, 3, 3, 3, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 9, 10, 10, 11, 11, 12, 12, 13, 14, 15}; - char delta_swing_table_idx_tdown_a[TXSCALE_TABLE_SIZE] = { + s8 delta_swing_table_idx_tdown_a[TXSCALE_TABLE_SIZE] = { 0, 0, 1, 2, 2, 2, 3, 3, 3, 4, 5, 5, 6, 6, 6, 6, 7, 7, 7, 8, 8, 9, 9, 10, 10, 11, 12, 13, 14, 15}; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c index c5ca9df..aa2e670 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c @@ -379,7 +379,7 @@ static void _rtl8723be_phy_store_txpower_by_rate_base(struct ieee80211_hw *hw) static void _phy_convert_txpower_dbm_to_relative_value(u32 *data, u8 start, u8 end, u8 base_val) { - char i = 0; + s8 i = 0; u8 temp_value = 0; u32 temp_data = 0; @@ -953,7 +953,7 @@ static u8 _rtl8723be_get_txpower_by_rate(struct ieee80211_hw *hw, struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_phy *rtlphy = &rtlpriv->phy; u8 shift = 0, rate_section, tx_num; - char tx_pwr_diff = 0; + s8 tx_pwr_diff = 0; rate_section = _rtl8723be_phy_get_ratesection_intxpower_byrate(rfpath, rate); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c index 6034597..e881ef8 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c @@ -56,7 +56,7 @@ static void _rtl8723be_query_rxphystatus(struct ieee80211_hw *hw, { struct rtl_priv *rtlpriv = rtl_priv(hw); struct phy_status_rpt *p_phystrpt = (struct phy_status_rpt *)p_drvinfo; - char rx_pwr_all = 0, rx_pwr[4]; + s8 rx_pwr_all = 0, rx_pwr[4]; u8 rf_rx_num = 0, evm, pwdb_all, pwdb_all_bt = 0; u8 i, max_spatial_stream; u32 rssi, total_rssi = 0; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.h index 40c36607..8a9fe41 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.h @@ -385,9 +385,9 @@ struct phy_status_rpt { u8 cck_rpt_b_ofdm_cfosho_b; u8 rsvd_1;/* ch_corr_msb; */ u8 noise_power_db_msb; - char path_cfotail[2]; + s8 path_cfotail[2]; u8 pcts_mask[2]; - char stream_rxevm[2]; + s8 stream_rxevm[2]; u8 path_rxsnr[2]; u8 noise_power_db_lsb; u8 rsvd_2[3]; @@ -422,8 +422,8 @@ struct rx_fwinfo_8723be { u8 pwdb_all; u8 cfosho[4]; u8 cfotail[4]; - char rxevm[2]; - char rxsnr[2]; + s8 rxevm[2]; + s8 rxsnr[2]; u8 pcts_msk_rpt[2]; u8 pdsnr[2]; u8 csi_current[2]; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c index 17a6817..69de835 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c @@ -1355,7 +1355,7 @@ void rtl8812ae_dm_txpwr_track_set_pwr(struct ieee80211_hw *hw, u32 final_swing_idx[2]; u8 pwr_tracking_limit = 26; /*+1.0dB*/ u8 tx_rate = 0xFF; - char final_ofdm_swing_index = 0; + s8 final_ofdm_swing_index = 0; if (rtldm->tx_rate != 0xFF) tx_rate = @@ -2045,7 +2045,7 @@ void rtl8821ae_dm_txpwr_track_set_pwr(struct ieee80211_hw *hw, u32 final_swing_idx[1]; u8 pwr_tracking_limit = 26; /*+1.0dB*/ u8 tx_rate = 0xFF; - char final_ofdm_swing_index = 0; + s8 final_ofdm_swing_index = 0; if (rtldm->tx_rate != 0xFF) tx_rate = rtl8821ae_hw_rate_to_mrate(hw, rtldm->tx_rate); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c index 0c3b9ce..46ea4f8 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c @@ -366,12 +366,12 @@ u32 phy_get_tx_swing_8812A(struct ieee80211_hw *hw, u8 band, struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); struct rtl_dm *rtldm = rtl_dm(rtlpriv); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); - char reg_swing_2g = -1;/* 0xff; */ - char reg_swing_5g = -1;/* 0xff; */ - char swing_2g = -1 * reg_swing_2g; - char swing_5g = -1 * reg_swing_5g; + s8 reg_swing_2g = -1;/* 0xff; */ + s8 reg_swing_5g = -1;/* 0xff; */ + s8 swing_2g = -1 * reg_swing_2g; + s8 swing_5g = -1 * reg_swing_5g; u32 out = 0x200; - const char auto_temp = -1; + const s8 auto_temp = -1; RT_TRACE(rtlpriv, COMP_SCAN, DBG_LOUD, "===> PHY_GetTxBBSwing_8812A, bbSwing_2G: %d, bbSwing_5G: %d,autoload_failflag=%d.\n", @@ -524,7 +524,7 @@ void rtl8821ae_phy_switch_wirelessband(struct ieee80211_hw *hw, u8 band) struct rtl_dm *rtldm = rtl_dm(rtlpriv); u8 current_band = rtlhal->current_bandtype; u32 txpath, rxpath; - char bb_diff_between_band; + s8 bb_diff_between_band; txpath = rtl8821ae_phy_query_bb_reg(hw, RTXPATH, 0xf0); rxpath = rtl8821ae_phy_query_bb_reg(hw, RCCK_RX, 0x0f000000); @@ -986,7 +986,7 @@ static void _rtl8812ae_phy_cross_reference_ht_and_vht_txpower_limit(struct ieee8 struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_phy *rtlphy = &rtlpriv->phy; u8 regulation, bw, channel, rate_section; - char temp_pwrlmt = 0; + s8 temp_pwrlmt = 0; for (regulation = 0; regulation < MAX_REGULATION_NUM; ++regulation) { for (bw = 0; bw < MAX_5G_BANDWITH_NUM; ++bw) { @@ -1155,7 +1155,7 @@ static void _rtl8812ae_phy_convert_txpower_limit_to_power_index(struct ieee80211 u8 regulation, bw, channel, rate_section; u8 base_index2_4G = 0; u8 base_index5G = 0; - char temp_value = 0, temp_pwrlmt = 0; + s8 temp_value = 0, temp_pwrlmt = 0; u8 rf_path = 0; RT_TRACE(rtlpriv, COMP_INIT, DBG_TRACE, @@ -1467,11 +1467,11 @@ static bool _rtl8812ae_eq_n_byte(u8 *str1, u8 *str2, u32 num) return true; } -static char _rtl8812ae_phy_get_chnl_idx_of_txpwr_lmt(struct ieee80211_hw *hw, +static s8 _rtl8812ae_phy_get_chnl_idx_of_txpwr_lmt(struct ieee80211_hw *hw, u8 band, u8 channel) { struct rtl_priv *rtlpriv = rtl_priv(hw); - char channel_index = -1; + s8 channel_index = -1; u8 i = 0; if (band == BAND_ON_2_4G) @@ -1502,7 +1502,7 @@ static void _rtl8812ae_phy_set_txpower_limit(struct ieee80211_hw *hw, u8 *pregul struct rtl_phy *rtlphy = &rtlpriv->phy; u8 regulation = 0, bandwidth = 0, rate_section = 0, channel; u8 channel_index; - char power_limit = 0, prev_power_limit, ret; + s8 power_limit = 0, prev_power_limit, ret; if (!_rtl8812ae_get_integer_from_string((char *)pchannel, &channel) || !_rtl8812ae_get_integer_from_string((char *)ppower_limit, @@ -2254,9 +2254,9 @@ static bool _rtl8821ae_phy_get_chnl_index(u8 channel, u8 *chnl_index) return in_24g; } -static char _rtl8821ae_phy_get_ratesection_intxpower_byrate(u8 path, u8 rate) +static s8 _rtl8821ae_phy_get_ratesection_intxpower_byrate(u8 path, u8 rate) { - char rate_section = 0; + s8 rate_section = 0; switch (rate) { case DESC_RATE1M: case DESC_RATE2M: @@ -2338,9 +2338,9 @@ static char _rtl8821ae_phy_get_ratesection_intxpower_byrate(u8 path, u8 rate) return rate_section; } -static char _rtl8812ae_phy_get_world_wide_limit(char *limit_table) +static s8 _rtl8812ae_phy_get_world_wide_limit(s8 *limit_table) { - char min = limit_table[0]; + s8 min = limit_table[0]; u8 i = 0; for (i = 0; i < MAX_REGULATION_NUM; ++i) { @@ -2350,7 +2350,7 @@ static char _rtl8812ae_phy_get_world_wide_limit(char *limit_table) return min; } -static char _rtl8812ae_phy_get_txpower_limit(struct ieee80211_hw *hw, +static s8 _rtl8812ae_phy_get_txpower_limit(struct ieee80211_hw *hw, u8 band, enum ht_channel_width bandwidth, enum radio_path rf_path, @@ -2362,7 +2362,7 @@ static char _rtl8812ae_phy_get_txpower_limit(struct ieee80211_hw *hw, short band_temp = -1, regulation = -1, bandwidth_temp = -1, rate_section = -1, channel_temp = -1; u16 bd, regu, bdwidth, sec, chnl; - char power_limit = MAX_POWER_INDEX; + s8 power_limit = MAX_POWER_INDEX; if (rtlefuse->eeprom_regulatory == 2) return MAX_POWER_INDEX; @@ -2489,7 +2489,7 @@ static char _rtl8812ae_phy_get_txpower_limit(struct ieee80211_hw *hw, chnl = channel_temp; if (band == BAND_ON_2_4G) { - char limits[10] = {0}; + s8 limits[10] = {0}; u8 i; for (i = 0; i < 4; ++i) @@ -2501,7 +2501,7 @@ static char _rtl8812ae_phy_get_txpower_limit(struct ieee80211_hw *hw, rtlphy->txpwr_limit_2_4g[regu][bdwidth] [sec][chnl][rf_path]; } else if (band == BAND_ON_5G) { - char limits[10] = {0}; + s8 limits[10] = {0}; u8 i; for (i = 0; i < MAX_REGULATION_NUM; ++i) @@ -2519,14 +2519,14 @@ static char _rtl8812ae_phy_get_txpower_limit(struct ieee80211_hw *hw, return power_limit; } -static char _rtl8821ae_phy_get_txpower_by_rate(struct ieee80211_hw *hw, +static s8 _rtl8821ae_phy_get_txpower_by_rate(struct ieee80211_hw *hw, u8 band, u8 path, u8 rate) { struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_phy *rtlphy = &rtlpriv->phy; u8 shift = 0, rate_section, tx_num; - char tx_pwr_diff = 0; - char limit = 0; + s8 tx_pwr_diff = 0; + s8 limit = 0; rate_section = _rtl8821ae_phy_get_ratesection_intxpower_byrate(path, rate); tx_num = RF_TX_NUM_NONIMPLEMENT; @@ -2639,7 +2639,7 @@ static u8 _rtl8821ae_get_txpower_index(struct ieee80211_hw *hw, u8 path, u8 index = (channel - 1); u8 txpower = 0; bool in_24g = false; - char powerdiff_byrate = 0; + s8 powerdiff_byrate = 0; if (((rtlhal->current_bandtype == BAND_ON_2_4G) && (channel > 14 || channel < 1)) || @@ -4637,7 +4637,7 @@ void rtl8821ae_phy_lc_calibrate(struct ieee80211_hw *hw) { } -void rtl8821ae_phy_ap_calibrate(struct ieee80211_hw *hw, char delta) +void rtl8821ae_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta) { } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.h b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.h index c411f0a..1285e1a 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.h @@ -236,7 +236,7 @@ void rtl8821ae_phy_iq_calibrate(struct ieee80211_hw *hw, bool b_recovery); void rtl8812ae_phy_iq_calibrate(struct ieee80211_hw *hw, bool b_recovery); -void rtl8821ae_phy_ap_calibrate(struct ieee80211_hw *hw, char delta); +void rtl8821ae_phy_ap_calibrate(struct ieee80211_hw *hw, s8 delta); void rtl8821ae_phy_lc_calibrate(struct ieee80211_hw *hw); void rtl8821ae_phy_set_rfpath_switch(struct ieee80211_hw *hw, bool bmain); bool rtl8812ae_phy_config_rf_with_headerfile(struct ieee80211_hw *hw, diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c index 41efaa1..8c2a5e2 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c @@ -48,7 +48,7 @@ static u8 _rtl8821ae_map_hwqueue_to_fwqueue(struct sk_buff *skb, u8 hw_queue) return skb->priority; } -static u16 odm_cfo(char value) +static u16 odm_cfo(s8 value) { int ret_val; @@ -64,9 +64,9 @@ static u16 odm_cfo(char value) return ret_val; } -static u8 _rtl8821ae_evm_dbm_jaguar(char value) +static u8 _rtl8821ae_evm_dbm_jaguar(s8 value) { - char ret_val = value; + s8 ret_val = value; /* -33dB~0dB to 33dB ~ 0dB*/ if (ret_val == -128) @@ -88,7 +88,7 @@ static void query_rxphystatus(struct ieee80211_hw *hw, struct phy_status_rpt *p_phystrpt = (struct phy_status_rpt *)p_drvinfo; struct rtl_dm *rtldm = rtl_dm(rtl_priv(hw)); struct rtl_phy *rtlphy = &rtlpriv->phy; - char rx_pwr_all = 0, rx_pwr[4]; + s8 rx_pwr_all = 0, rx_pwr[4]; u8 rf_rx_num = 0, evm, evmdbm, pwdb_all; u8 i, max_spatial_stream; u32 rssi, total_rssi = 0; @@ -170,7 +170,7 @@ static void query_rxphystatus(struct ieee80211_hw *hw, pwdb_all = 100; } } else { /* 8821 */ - char pout = -6; + s8 pout = -6; switch (lan_idx) { case 5: @@ -275,7 +275,7 @@ static void query_rxphystatus(struct ieee80211_hw *hw, if (bpacket_match_bssid) { for (i = RF90_PATH_A; i <= RF90_PATH_B; i++) rtl_priv(hw)->dm.cfo_tail[i] = - (char)p_phystrpt->cfotail[i]; + (s8)p_phystrpt->cfotail[i]; rtl_priv(hw)->dm.packet_count++; } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.h b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.h index ad565be..b6f3c56 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.h +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.h @@ -390,11 +390,11 @@ struct phy_status_rpt { u8 cfosho[4]; /* DW 1 byte 1 DW 2 byte 0 */ /* DWORD 2 */ - char cfotail[4]; /* DW 2 byte 1 DW 3 byte 0 */ + s8 cfotail[4]; /* DW 2 byte 1 DW 3 byte 0 */ /* DWORD 3 */ - char rxevm[2]; /* DW 3 byte 1 DW 3 byte 2 */ - char rxsnr[2]; /* DW 3 byte 3 DW 4 byte 0 */ + s8 rxevm[2]; /* DW 3 byte 1 DW 3 byte 2 */ + s8 rxsnr[2]; /* DW 3 byte 3 DW 4 byte 0 */ /* DWORD 4 */ u8 pcts_msk_rpt[2]; @@ -418,8 +418,8 @@ struct rx_fwinfo_8821ae { u8 pwdb_all; u8 cfosho[4]; u8 cfotail[4]; - char rxevm[2]; - char rxsnr[4]; + s8 rxevm[2]; + s8 rxsnr[4]; u8 pdsnr[2]; u8 csi_current[2]; u8 csi_target[2]; diff --git a/drivers/net/wireless/realtek/rtlwifi/stats.c b/drivers/net/wireless/realtek/rtlwifi/stats.c index d8b3069..61700fa 100644 --- a/drivers/net/wireless/realtek/rtlwifi/stats.c +++ b/drivers/net/wireless/realtek/rtlwifi/stats.c @@ -26,7 +26,7 @@ #include "stats.h" #include -u8 rtl_query_rxpwrpercentage(char antpower) +u8 rtl_query_rxpwrpercentage(s8 antpower) { if ((antpower <= -100) || (antpower >= 20)) return 0; @@ -37,9 +37,9 @@ u8 rtl_query_rxpwrpercentage(char antpower) } EXPORT_SYMBOL(rtl_query_rxpwrpercentage); -u8 rtl_evm_db_to_percentage(char value) +u8 rtl_evm_db_to_percentage(s8 value) { - char ret_val = clamp(-value, 0, 33) * 3; + s8 ret_val = clamp(-value, 0, 33) * 3; if (ret_val == 99) ret_val = 100; diff --git a/drivers/net/wireless/realtek/rtlwifi/stats.h b/drivers/net/wireless/realtek/rtlwifi/stats.h index 2b57dff..bd0108f 100644 --- a/drivers/net/wireless/realtek/rtlwifi/stats.h +++ b/drivers/net/wireless/realtek/rtlwifi/stats.h @@ -33,8 +33,8 @@ /* Rx smooth factor */ #define RX_SMOOTH_FACTOR 20 -u8 rtl_query_rxpwrpercentage(char antpower); -u8 rtl_evm_db_to_percentage(char value); +u8 rtl_query_rxpwrpercentage(s8 antpower); +u8 rtl_evm_db_to_percentage(s8 value); long rtl_signal_scale_mapping(struct ieee80211_hw *hw, long currsig); void rtl_process_phyinfo(struct ieee80211_hw *hw, u8 *buffer, struct rtl_stats *pstatus); diff --git a/drivers/net/wireless/realtek/rtlwifi/wifi.h b/drivers/net/wireless/realtek/rtlwifi/wifi.h index 4e0ab4d..c5086c2 100644 --- a/drivers/net/wireless/realtek/rtlwifi/wifi.h +++ b/drivers/net/wireless/realtek/rtlwifi/wifi.h @@ -1089,7 +1089,7 @@ struct dynamic_primary_cca { }; struct rtl_regulatory { - char alpha2[2]; + s8 alpha2[2]; u16 country_code; u16 max_power_level; u32 tp_scale; @@ -1256,16 +1256,16 @@ struct rtl_phy { u8 cur_bw20_txpwridx; u8 cur_bw40_txpwridx; - char txpwr_limit_2_4g[MAX_REGULATION_NUM] - [MAX_2_4G_BANDWITH_NUM] - [MAX_RATE_SECTION_NUM] - [CHANNEL_MAX_NUMBER_2G] - [MAX_RF_PATH_NUM]; - char txpwr_limit_5g[MAX_REGULATION_NUM] - [MAX_5G_BANDWITH_NUM] + s8 txpwr_limit_2_4g[MAX_REGULATION_NUM] + [MAX_2_4G_BANDWITH_NUM] [MAX_RATE_SECTION_NUM] - [CHANNEL_MAX_NUMBER_5G] + [CHANNEL_MAX_NUMBER_2G] [MAX_RF_PATH_NUM]; + s8 txpwr_limit_5g[MAX_REGULATION_NUM] + [MAX_5G_BANDWITH_NUM] + [MAX_RATE_SECTION_NUM] + [CHANNEL_MAX_NUMBER_5G] + [MAX_RF_PATH_NUM]; u32 rfreg_chnlval[2]; bool apk_done; @@ -1639,7 +1639,7 @@ struct fast_ant_training { }; struct dm_phy_dbg_info { - char rx_snrdb[4]; + s8 rx_snrdb[4]; u64 num_qry_phy_status; u64 num_qry_phy_status_cck; u64 num_qry_phy_status_ofdm; @@ -1688,16 +1688,16 @@ struct rtl_dm { u8 txpower_track_control; bool interrupt_migration; bool disable_tx_int; - char ofdm_index[MAX_RF_PATH]; + s8 ofdm_index[MAX_RF_PATH]; u8 default_ofdm_index; u8 default_cck_index; - char cck_index; - char delta_power_index[MAX_RF_PATH]; - char delta_power_index_last[MAX_RF_PATH]; - char power_index_offset[MAX_RF_PATH]; - char absolute_ofdm_swing_idx[MAX_RF_PATH]; - char remnant_ofdm_swing_idx[MAX_RF_PATH]; - char remnant_cck_idx; + s8 cck_index; + s8 delta_power_index[MAX_RF_PATH]; + s8 delta_power_index_last[MAX_RF_PATH]; + s8 power_index_offset[MAX_RF_PATH]; + s8 absolute_ofdm_swing_idx[MAX_RF_PATH]; + s8 remnant_ofdm_swing_idx[MAX_RF_PATH]; + s8 remnant_cck_idx; bool modify_txagc_flag_path_a; bool modify_txagc_flag_path_b; @@ -1726,8 +1726,8 @@ struct rtl_dm { u8 swing_idx_cck_base; bool swing_flag_cck; - char swing_diff_2g; - char swing_diff_5g; + s8 swing_diff_2g; + s8 swing_diff_5g; u8 delta_swing_table_idx_24gccka_p[DEL_SW_IDX_SZ]; u8 delta_swing_table_idx_24gccka_n[DEL_SW_IDX_SZ]; @@ -1838,17 +1838,17 @@ struct rtl_efuse { * * Sizes of these arrays are decided by the larger ones. */ - char txpwr_cckdiff[MAX_RF_PATH][CHANNEL_MAX_NUMBER]; - char txpwr_ht20diff[MAX_RF_PATH][CHANNEL_MAX_NUMBER]; - char txpwr_ht40diff[MAX_RF_PATH][CHANNEL_MAX_NUMBER]; - char txpwr_legacyhtdiff[MAX_RF_PATH][CHANNEL_MAX_NUMBER]; + s8 txpwr_cckdiff[MAX_RF_PATH][CHANNEL_MAX_NUMBER]; + s8 txpwr_ht20diff[MAX_RF_PATH][CHANNEL_MAX_NUMBER]; + s8 txpwr_ht40diff[MAX_RF_PATH][CHANNEL_MAX_NUMBER]; + s8 txpwr_legacyhtdiff[MAX_RF_PATH][CHANNEL_MAX_NUMBER]; u8 txpwr_5g_bw40base[MAX_RF_PATH][CHANNEL_MAX_NUMBER]; u8 txpwr_5g_bw80base[MAX_RF_PATH][CHANNEL_MAX_NUMBER_5G_80M]; - char txpwr_5g_ofdmdiff[MAX_RF_PATH][MAX_TX_COUNT]; - char txpwr_5g_bw20diff[MAX_RF_PATH][MAX_TX_COUNT]; - char txpwr_5g_bw40diff[MAX_RF_PATH][MAX_TX_COUNT]; - char txpwr_5g_bw80diff[MAX_RF_PATH][MAX_TX_COUNT]; + s8 txpwr_5g_ofdmdiff[MAX_RF_PATH][MAX_TX_COUNT]; + s8 txpwr_5g_bw20diff[MAX_RF_PATH][MAX_TX_COUNT]; + s8 txpwr_5g_bw40diff[MAX_RF_PATH][MAX_TX_COUNT]; + s8 txpwr_5g_bw80diff[MAX_RF_PATH][MAX_TX_COUNT]; u8 txpwr_safetyflag; /* Band edge enable flag */ u16 eeprom_txpowerdiff; @@ -2006,7 +2006,7 @@ struct rtl_stats { bool is_ht; bool packet_toself; bool packet_beacon; /*for rssi */ - char cck_adc_pwdb[4]; /*for rx path selection */ + s8 cck_adc_pwdb[4]; /*for rx path selection */ bool is_vht; bool is_short_gi; @@ -2413,9 +2413,9 @@ struct dig_t { u8 presta_cstate; u8 curmultista_cstate; u8 stop_dig; - char back_val; - char back_range_max; - char back_range_min; + s8 back_val; + s8 back_range_max; + s8 back_range_min; u8 rx_gain_max; u8 rx_gain_min; u8 min_undec_pwdb_for_dm; @@ -2441,8 +2441,8 @@ struct dig_t { u8 cur_cs_ratiostate; u8 pre_cs_ratiostate; u8 backoff_enable_flag; - char backoffval_range_max; - char backoffval_range_min; + s8 backoffval_range_max; + s8 backoffval_range_min; u8 dig_min_0; u8 dig_min_1; u8 bt30_cur_igi; -- cgit v0.10.2 From 4699fc3f2dcbcce27d499343c7b85b172b17ee76 Mon Sep 17 00:00:00 2001 From: Ganapathi Bhat Date: Thu, 16 Jun 2016 18:52:21 +0530 Subject: mwifiex: Fix an issue spotted by KASAN When an association command is sent to firmware but the process is killed before the command response arrives, driver will try to access bss_desc which is already freed. This issue is fixed by checking return value of bss_start. Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/join.c b/drivers/net/wireless/marvell/mwifiex/join.c index a4b773d..1c7b006 100644 --- a/drivers/net/wireless/marvell/mwifiex/join.c +++ b/drivers/net/wireless/marvell/mwifiex/join.c @@ -647,6 +647,12 @@ int mwifiex_ret_802_11_associate(struct mwifiex_private *priv, const u8 *ie_ptr; struct ieee80211_ht_operation *assoc_resp_ht_oper; + if (!priv->attempted_bss_desc) { + mwifiex_dbg(priv->adapter, ERROR, + "ASSOC_RESP: failed, association terminated by host\n"); + goto done; + } + assoc_rsp = (struct ieee_types_assoc_rsp *) &resp->params; cap_info = le16_to_cpu(assoc_rsp->cap_info_bitmap); @@ -1270,6 +1276,12 @@ int mwifiex_ret_802_11_ad_hoc(struct mwifiex_private *priv, u16 cmd = le16_to_cpu(resp->command); u8 result; + if (!priv->attempted_bss_desc) { + mwifiex_dbg(priv->adapter, ERROR, + "ADHOC_RESP: failed, association terminated by host\n"); + goto done; + } + if (cmd == HostCmd_CMD_802_11_AD_HOC_START) result = start_result->result; else diff --git a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c index 8e08626..2ba5397 100644 --- a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c +++ b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c @@ -426,6 +426,10 @@ done: if (bss_desc) kfree(bss_desc->beacon_buf); kfree(bss_desc); + + if (ret < 0) + priv->attempted_bss_desc = NULL; + return ret; } -- cgit v0.10.2 From a9c790ba23eb3b3649f012f9633be2cb5e73e588 Mon Sep 17 00:00:00 2001 From: Xinming Hu Date: Thu, 16 Jun 2016 18:52:22 +0530 Subject: mwifiex: factor out mwifiex_cancel_scan This patch creates common function mwifiex_cancel_scan to remove duplication of code. Signed-off-by: Xinming Hu Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/cmdevt.c b/drivers/net/wireless/marvell/mwifiex/cmdevt.c index 6bc2011..c29f26d 100644 --- a/drivers/net/wireless/marvell/mwifiex/cmdevt.c +++ b/drivers/net/wireless/marvell/mwifiex/cmdevt.c @@ -1020,8 +1020,6 @@ mwifiex_cancel_all_pending_cmd(struct mwifiex_adapter *adapter) { struct cmd_ctrl_node *cmd_node = NULL, *tmp_node; unsigned long flags, cmd_flags; - struct mwifiex_private *priv; - int i; spin_lock_irqsave(&adapter->mwifiex_cmd_lock, cmd_flags); /* Cancel current cmd */ @@ -1046,23 +1044,7 @@ mwifiex_cancel_all_pending_cmd(struct mwifiex_adapter *adapter) spin_unlock_irqrestore(&adapter->cmd_pending_q_lock, flags); spin_unlock_irqrestore(&adapter->mwifiex_cmd_lock, cmd_flags); - mwifiex_cancel_pending_scan_cmd(adapter); - - if (adapter->scan_processing) { - spin_lock_irqsave(&adapter->mwifiex_cmd_lock, cmd_flags); - adapter->scan_processing = false; - spin_unlock_irqrestore(&adapter->mwifiex_cmd_lock, cmd_flags); - for (i = 0; i < adapter->priv_num; i++) { - priv = adapter->priv[i]; - if (!priv) - continue; - if (priv->scan_request) { - mwifiex_dbg(adapter, WARN, "info: aborting scan\n"); - cfg80211_scan_done(priv->scan_request, 1); - priv->scan_request = NULL; - } - } - } + mwifiex_cancel_scan(adapter); } /* @@ -1080,8 +1062,6 @@ mwifiex_cancel_pending_ioctl(struct mwifiex_adapter *adapter) { struct cmd_ctrl_node *cmd_node = NULL; unsigned long cmd_flags; - struct mwifiex_private *priv; - int i; if ((adapter->curr_cmd) && (adapter->curr_cmd->wait_q_enabled)) { @@ -1101,23 +1081,7 @@ mwifiex_cancel_pending_ioctl(struct mwifiex_adapter *adapter) mwifiex_recycle_cmd_node(adapter, cmd_node); } - mwifiex_cancel_pending_scan_cmd(adapter); - - if (adapter->scan_processing) { - spin_lock_irqsave(&adapter->mwifiex_cmd_lock, cmd_flags); - adapter->scan_processing = false; - spin_unlock_irqrestore(&adapter->mwifiex_cmd_lock, cmd_flags); - for (i = 0; i < adapter->priv_num; i++) { - priv = adapter->priv[i]; - if (!priv) - continue; - if (priv->scan_request) { - mwifiex_dbg(adapter, WARN, "info: aborting scan\n"); - cfg80211_scan_done(priv->scan_request, 1); - priv->scan_request = NULL; - } - } - } + mwifiex_cancel_scan(adapter); } /* diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h index f0cd055..0d38a72 100644 --- a/drivers/net/wireless/marvell/mwifiex/main.h +++ b/drivers/net/wireless/marvell/mwifiex/main.h @@ -1054,6 +1054,7 @@ int mwifiex_free_cmd_buffer(struct mwifiex_adapter *adapter); void mwifiex_cancel_all_pending_cmd(struct mwifiex_adapter *adapter); void mwifiex_cancel_pending_ioctl(struct mwifiex_adapter *adapter); void mwifiex_cancel_pending_scan_cmd(struct mwifiex_adapter *adapter); +void mwifiex_cancel_scan(struct mwifiex_adapter *adapter); void mwifiex_recycle_cmd_node(struct mwifiex_adapter *adapter, struct cmd_ctrl_node *cmd_node); diff --git a/drivers/net/wireless/marvell/mwifiex/scan.c b/drivers/net/wireless/marvell/mwifiex/scan.c index bc5e52c..e331122 100644 --- a/drivers/net/wireless/marvell/mwifiex/scan.c +++ b/drivers/net/wireless/marvell/mwifiex/scan.c @@ -2001,6 +2001,32 @@ static void mwifiex_check_next_scan_command(struct mwifiex_private *priv) return; } +void mwifiex_cancel_scan(struct mwifiex_adapter *adapter) +{ + struct mwifiex_private *priv; + unsigned long cmd_flags; + int i; + + mwifiex_cancel_pending_scan_cmd(adapter); + + if (adapter->scan_processing) { + spin_lock_irqsave(&adapter->mwifiex_cmd_lock, cmd_flags); + adapter->scan_processing = false; + spin_unlock_irqrestore(&adapter->mwifiex_cmd_lock, cmd_flags); + for (i = 0; i < adapter->priv_num; i++) { + priv = adapter->priv[i]; + if (!priv) + continue; + if (priv->scan_request) { + mwifiex_dbg(adapter, INFO, + "info: aborting scan\n"); + cfg80211_scan_done(priv->scan_request, 1); + priv->scan_request = NULL; + } + } + } +} + /* * This function handles the command response of scan. * -- cgit v0.10.2 From dec277f781cec519d96151bb740963b2a49f5f5d Mon Sep 17 00:00:00 2001 From: Xinming Hu Date: Thu, 16 Jun 2016 18:52:23 +0530 Subject: mwifiex: cancel pending scan during disconnect It is obeserved that sometimes scan operation will block the disconnect during system suspend. It's ok to cancel ongoing scan in this case. It reduces unnecessary system suspend delay. Signed-off-by: Xinming Hu Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c index ff948a9..a05ae00 100644 --- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c +++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c @@ -1672,6 +1672,9 @@ static int mwifiex_cfg80211_change_beacon(struct wiphy *wiphy, struct cfg80211_beacon_data *data) { struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev); + struct mwifiex_adapter *adapter = priv->adapter; + + mwifiex_cancel_scan(adapter); if (GET_BSS_ROLE(priv) != MWIFIEX_BSS_ROLE_UAP) { mwifiex_dbg(priv->adapter, ERROR, -- cgit v0.10.2 From 437322ea2a36d112e20aa7282c869bf924b3a836 Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Thu, 16 Jun 2016 18:52:24 +0530 Subject: mwifiex: fix system hang problem after resume On some platforms, driver is unable to wakeup firmware after system resume due to a problem at MMC subsystem. Triggering card reset in this case has a race with card removal from MMC which causes system hang. This patch resolves the problem by not triggering card reset. Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/init.c b/drivers/net/wireless/marvell/mwifiex/init.c index a6d86d4..07eac5b 100644 --- a/drivers/net/wireless/marvell/mwifiex/init.c +++ b/drivers/net/wireless/marvell/mwifiex/init.c @@ -60,7 +60,7 @@ static void wakeup_timer_fn(unsigned long data) adapter->hw_status = MWIFIEX_HW_STATUS_RESET; mwifiex_cancel_all_pending_cmd(adapter); - if (adapter->if_ops.card_reset) + if (adapter->if_ops.card_reset && !adapter->hs_activated) adapter->if_ops.card_reset(adapter); } -- cgit v0.10.2 From 568fb26ec8be5890a698d672472aeb6b6db34e57 Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Thu, 16 Jun 2016 18:52:25 +0530 Subject: mwifiex: fix AP unable to start in VHT40 problem This patch populates secondary channel offset and downloads it to firmware to fix the problem. Signed-off-by: Amitkumar Karwar Signed-off-by: Cathy Luo Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/ioctl.h b/drivers/net/wireless/marvell/mwifiex/ioctl.h index a5a48c1..f5b8fd1 100644 --- a/drivers/net/wireless/marvell/mwifiex/ioctl.h +++ b/drivers/net/wireless/marvell/mwifiex/ioctl.h @@ -83,6 +83,8 @@ struct wep_key { #define MWIFIEX_AUTH_MODE_AUTO 0xFF #define BAND_CONFIG_BG 0x00 #define BAND_CONFIG_A 0x01 +#define MWIFIEX_SEC_CHAN_BELOW 0x30 +#define MWIFIEX_SEC_CHAN_ABOVE 0x10 #define MWIFIEX_SUPPORTED_RATES 14 #define MWIFIEX_SUPPORTED_RATES_EXT 32 #define MWIFIEX_TDLS_SUPPORTED_RATES 8 diff --git a/drivers/net/wireless/marvell/mwifiex/uap_cmd.c b/drivers/net/wireless/marvell/mwifiex/uap_cmd.c index f79d00d..a7e9f54 100644 --- a/drivers/net/wireless/marvell/mwifiex/uap_cmd.c +++ b/drivers/net/wireless/marvell/mwifiex/uap_cmd.c @@ -19,6 +19,7 @@ #include "main.h" #include "11ac.h" +#include "11n.h" /* This function parses security related parameters from cfg80211_ap_settings * and sets into FW understandable bss_config structure. @@ -521,9 +522,9 @@ mwifiex_uap_bss_param_prepare(u8 *tlv, void *cmd_buf, u16 *param_size) tlv += sizeof(struct host_cmd_tlv_rates) + i; } if (bss_cfg->channel && - ((bss_cfg->band_cfg == BAND_CONFIG_BG && + (((bss_cfg->band_cfg & BIT(0)) == BAND_CONFIG_BG && bss_cfg->channel <= MAX_CHANNEL_BAND_BG) || - (bss_cfg->band_cfg == BAND_CONFIG_A && + ((bss_cfg->band_cfg & BIT(0)) == BAND_CONFIG_A && bss_cfg->channel <= MAX_CHANNEL_BAND_A))) { chan_band = (struct host_cmd_tlv_channel_band *)tlv; chan_band->header.type = cpu_to_le16(TLV_TYPE_CHANNELBANDLIST); @@ -833,6 +834,31 @@ void mwifiex_uap_set_channel(struct mwifiex_private *priv, config_bands |= BAND_AAC; } + switch (chandef.width) { + case NL80211_CHAN_WIDTH_5: + case NL80211_CHAN_WIDTH_10: + case NL80211_CHAN_WIDTH_20_NOHT: + case NL80211_CHAN_WIDTH_20: + break; + case NL80211_CHAN_WIDTH_40: + if (chandef.center_freq1 < chandef.chan->center_freq) + bss_cfg->band_cfg |= MWIFIEX_SEC_CHAN_BELOW; + else + bss_cfg->band_cfg |= MWIFIEX_SEC_CHAN_ABOVE; + break; + case NL80211_CHAN_WIDTH_80: + case NL80211_CHAN_WIDTH_80P80: + case NL80211_CHAN_WIDTH_160: + bss_cfg->band_cfg |= + mwifiex_get_sec_chan_offset(bss_cfg->channel) << 4; + break; + default: + mwifiex_dbg(priv->adapter, + WARN, "Unknown channel width: %d\n", + chandef.width); + break; + } + priv->adapter->config_bands = config_bands; if (old_bands != config_bands) { -- cgit v0.10.2 From 7311ea85007954e14cb7f5c60ce969f7a4f618c6 Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Thu, 16 Jun 2016 18:52:26 +0530 Subject: mwifiex: fix AP start problem for newly added interface It's been observed that if interface type is changed from managed to __ap, AP can be successfully started. But there is a problem if new ap interface is added. The problem got resolved after sending appropriate commands to firmware in add_interface handler. Signed-off-by: Amitkumar Karwar Signed-off-by: Cathy Luo Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c index a05ae00..99e8cf1 100644 --- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c +++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c @@ -2737,6 +2737,7 @@ struct wireless_dev *mwifiex_add_virtual_intf(struct wiphy *wiphy, struct mwifiex_private *priv; struct net_device *dev; void *mdev_priv; + int ret; if (!adapter) return ERR_PTR(-EFAULT); @@ -2862,6 +2863,13 @@ struct wireless_dev *mwifiex_add_virtual_intf(struct wiphy *wiphy, mwifiex_init_priv_params(priv, dev); priv->netdev = dev; + ret = mwifiex_send_cmd(priv, HostCmd_CMD_SET_BSS_MODE, + HostCmd_ACT_GEN_SET, 0, NULL, true); + return ERR_PTR(ret); + + ret = mwifiex_sta_init_cmd(priv, false, false); + return ERR_PTR(ret); + mwifiex_setup_ht_caps(&wiphy->bands[NL80211_BAND_2GHZ]->ht_cap, priv); if (adapter->is_hw_11ac_capable) mwifiex_setup_vht_caps( -- cgit v0.10.2 From 88e97c32068ff365970c0958771b9dce131c3590 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 15:52:08 +0200 Subject: wireless: airo: rename 'register' variable 'register' is a keyword in C and cannot be used in place of a variable name, as shown by this -Wextra warning: drivers/net/wireless/cisco/airo.c:1105:29: error: 'register' is not at beginning of declaration [-Werror=old-style-declaration] This replaces the 'register' keyword with a 'reg' identifier in the declaration, which matches the definition and has the intended meaning. Signed-off-by: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/cisco/airo.c b/drivers/net/wireless/cisco/airo.c index ca3cd21..69b826d 100644 --- a/drivers/net/wireless/cisco/airo.c +++ b/drivers/net/wireless/cisco/airo.c @@ -1102,8 +1102,8 @@ static const char version[] = "airo.c 0.6 (Ben Reed & Javier Achirica)"; struct airo_info; static int get_dec_u16( char *buffer, int *start, int limit ); -static void OUT4500( struct airo_info *, u16 register, u16 value ); -static unsigned short IN4500( struct airo_info *, u16 register ); +static void OUT4500( struct airo_info *, u16 reg, u16 value ); +static unsigned short IN4500( struct airo_info *, u16 reg ); static u16 setup_card(struct airo_info*, u8 *mac, int lock); static int enable_MAC(struct airo_info *ai, int lock); static void disable_MAC(struct airo_info *ai, int lock); -- cgit v0.10.2 From 2a063835ce7cf584cee03484af803cd98e5fc9f2 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 15:52:09 +0200 Subject: wireless: brcmsmac: fix old-style declaration Modern C standards expect the 'static' keyword to come first in a declaration, and we get a warning for this with "make W=1": drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c:3353:1: error: 'static' is not at beginning of declaration [-Werror=old-style-declaration] Signed-off-by: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c index e16ee60..c2a938b 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/main.c @@ -3349,8 +3349,8 @@ static void brcms_b_coreinit(struct brcms_c_info *wlc) dma_rxfill(wlc_hw->di[RX_FIFO]); } -void -static brcms_b_init(struct brcms_hardware *wlc_hw, u16 chanspec) { +static void brcms_b_init(struct brcms_hardware *wlc_hw, u16 chanspec) +{ u32 macintmask; bool fastclk; struct brcms_c_info *wlc = wlc_hw->wlc; -- cgit v0.10.2 From 6f07e0f12a577a5820dd6b57c3f3b0f7f4a3c52c Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 16 Jun 2016 15:52:10 +0200 Subject: wireless: ipw2200: fix old-style declaration Modern C standards expect the 'inline' keyword to come before the return type in a declaration, and we get a warning for this with "make W=1": drivers/net/wireless/intel/ipw2x00/ipw2200.c:4096:1: error: 'inline' is not at beginning of declaration [-Werror=old-style-declaration] Signed-off-by: Arnd Bergmann Acked-by: Stanislav Yakovlev Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/intel/ipw2x00/ipw2200.c b/drivers/net/wireless/intel/ipw2x00/ipw2200.c index 5adb7ce..bfd6861 100644 --- a/drivers/net/wireless/intel/ipw2x00/ipw2200.c +++ b/drivers/net/wireless/intel/ipw2x00/ipw2200.c @@ -4093,7 +4093,7 @@ static const char *ipw_get_status_code(u16 status) return "Unknown status value."; } -static void inline average_init(struct average *avg) +static inline void average_init(struct average *avg) { memset(avg, 0, sizeof(*avg)); } -- cgit v0.10.2 From b50ddfa8530e9b5f52e873fdd6ff04f327a88799 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Fri, 17 Jun 2016 12:29:21 +0200 Subject: brcmfmac: fix lockup when removing P2P interface after event timeout MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Removing P2P interface is handled by sending a proper request to the firmware. On success firmware triggers an event and driver's handler removes a matching interface. However on event timeout we remove interface directly from the cfg80211 callback. Current code doesn't handle this case correctly as it always assumes rtnl to be unlocked. Fix it by adding an extra rtnl_locked parameter to functions and calling unregister_netdevice when needed. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c index faf4e46..7b38c2b 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c @@ -548,12 +548,16 @@ fail: return -EBADE; } -static void brcmf_net_detach(struct net_device *ndev) +static void brcmf_net_detach(struct net_device *ndev, bool rtnl_locked) { - if (ndev->reg_state == NETREG_REGISTERED) - unregister_netdev(ndev); - else + if (ndev->reg_state == NETREG_REGISTERED) { + if (rtnl_locked) + unregister_netdevice(ndev); + else + unregister_netdev(ndev); + } else { brcmf_cfg80211_free_netdev(ndev); + } } void brcmf_net_setcarrier(struct brcmf_if *ifp, bool on) @@ -651,7 +655,7 @@ struct brcmf_if *brcmf_add_if(struct brcmf_pub *drvr, s32 bsscfgidx, s32 ifidx, brcmf_err("ERROR: netdev:%s already exists\n", ifp->ndev->name); netif_stop_queue(ifp->ndev); - brcmf_net_detach(ifp->ndev); + brcmf_net_detach(ifp->ndev, false); drvr->iflist[bsscfgidx] = NULL; } else { brcmf_dbg(INFO, "netdev:%s ignore IF event\n", @@ -699,7 +703,8 @@ struct brcmf_if *brcmf_add_if(struct brcmf_pub *drvr, s32 bsscfgidx, s32 ifidx, return ifp; } -static void brcmf_del_if(struct brcmf_pub *drvr, s32 bsscfgidx) +static void brcmf_del_if(struct brcmf_pub *drvr, s32 bsscfgidx, + bool rtnl_locked) { struct brcmf_if *ifp; @@ -729,7 +734,7 @@ static void brcmf_del_if(struct brcmf_pub *drvr, s32 bsscfgidx) cancel_work_sync(&ifp->multicast_work); cancel_work_sync(&ifp->ndoffload_work); } - brcmf_net_detach(ifp->ndev); + brcmf_net_detach(ifp->ndev, rtnl_locked); } else { /* Only p2p device interfaces which get dynamically created * end up here. In this case the p2p module should be informed @@ -743,14 +748,14 @@ static void brcmf_del_if(struct brcmf_pub *drvr, s32 bsscfgidx) } } -void brcmf_remove_interface(struct brcmf_if *ifp) +void brcmf_remove_interface(struct brcmf_if *ifp, bool rtnl_locked) { if (!ifp || WARN_ON(ifp->drvr->iflist[ifp->bsscfgidx] != ifp)) return; brcmf_dbg(TRACE, "Enter, bsscfgidx=%d, ifidx=%d\n", ifp->bsscfgidx, ifp->ifidx); brcmf_fws_del_interface(ifp); - brcmf_del_if(ifp->drvr, ifp->bsscfgidx); + brcmf_del_if(ifp->drvr, ifp->bsscfgidx, rtnl_locked); } #ifdef CONFIG_INET @@ -1057,9 +1062,9 @@ fail: brcmf_fws_deinit(drvr); } if (ifp) - brcmf_net_detach(ifp->ndev); + brcmf_net_detach(ifp->ndev, false); if (p2p_ifp) - brcmf_net_detach(p2p_ifp->ndev); + brcmf_net_detach(p2p_ifp->ndev, false); drvr->iflist[0] = NULL; drvr->iflist[1] = NULL; if (drvr->settings->ignore_probe_fail) @@ -1128,7 +1133,7 @@ void brcmf_detach(struct device *dev) /* make sure primary interface removed last */ for (i = BRCMF_MAX_IFS-1; i > -1; i--) - brcmf_remove_interface(drvr->iflist[i]); + brcmf_remove_interface(drvr->iflist[i], false); brcmf_cfg80211_detach(drvr->config); diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h index 2a075c5..a0a6f7f 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h @@ -216,7 +216,7 @@ struct brcmf_if *brcmf_get_ifp(struct brcmf_pub *drvr, int ifidx); int brcmf_net_attach(struct brcmf_if *ifp, bool rtnl_locked); struct brcmf_if *brcmf_add_if(struct brcmf_pub *drvr, s32 bsscfgidx, s32 ifidx, bool is_p2pdev, char *name, u8 *mac_addr); -void brcmf_remove_interface(struct brcmf_if *ifp); +void brcmf_remove_interface(struct brcmf_if *ifp, bool rtnl_locked); void brcmf_txflowblock_if(struct brcmf_if *ifp, enum brcmf_netif_stop_reason reason, bool state); void brcmf_txfinalize(struct brcmf_if *ifp, struct sk_buff *txp, bool success); diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c index b390561..9da7a4c 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c @@ -183,7 +183,7 @@ static void brcmf_fweh_handle_if_event(struct brcmf_pub *drvr, err = brcmf_fweh_call_event_handler(ifp, emsg->event_code, emsg, data); if (ifp && ifevent->action == BRCMF_E_IF_DEL) - brcmf_remove_interface(ifp); + brcmf_remove_interface(ifp, false); } /** diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c index f38a821..426ff05 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c @@ -2287,7 +2287,7 @@ int brcmf_p2p_del_vif(struct wiphy *wiphy, struct wireless_dev *wdev) err = 0; } if (err) - brcmf_remove_interface(vif->ifp); + brcmf_remove_interface(vif->ifp, true); brcmf_cfg80211_arm_vif_event(cfg, NULL); if (vif->wdev.iftype != NL80211_IFTYPE_P2P_DEVICE) @@ -2393,7 +2393,7 @@ void brcmf_p2p_detach(struct brcmf_p2p_info *p2p) if (vif != NULL) { brcmf_p2p_cancel_remain_on_channel(vif->ifp); brcmf_p2p_deinit_discovery(p2p); - brcmf_remove_interface(vif->ifp); + brcmf_remove_interface(vif->ifp, false); } /* just set it all to zero */ memset(p2p, 0, sizeof(*p2p)); -- cgit v0.10.2 From 54264e7ea09a1a1eb78262b71a16f7eb46de2f2f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Fri, 17 Jun 2016 12:48:44 +0200 Subject: brcmfmac: use const char * for interface name in brcmf_add_if MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This function can work just fine with const pointer, it only calls alloc_netdev which take const as well. Moreover it makes this function more flexible as some cfg80211 callback may provide const char * as well, e.g. add_virtual_intf. This will be needed for more advanced interface management. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c index 7b38c2b..8d16f02 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.c @@ -638,7 +638,7 @@ fail: } struct brcmf_if *brcmf_add_if(struct brcmf_pub *drvr, s32 bsscfgidx, s32 ifidx, - bool is_p2pdev, char *name, u8 *mac_addr) + bool is_p2pdev, const char *name, u8 *mac_addr) { struct brcmf_if *ifp; struct net_device *ndev; diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h index a0a6f7f..8fa34ca 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/core.h @@ -215,7 +215,7 @@ char *brcmf_ifname(struct brcmf_if *ifp); struct brcmf_if *brcmf_get_ifp(struct brcmf_pub *drvr, int ifidx); int brcmf_net_attach(struct brcmf_if *ifp, bool rtnl_locked); struct brcmf_if *brcmf_add_if(struct brcmf_pub *drvr, s32 bsscfgidx, s32 ifidx, - bool is_p2pdev, char *name, u8 *mac_addr); + bool is_p2pdev, const char *name, u8 *mac_addr); void brcmf_remove_interface(struct brcmf_if *ifp, bool rtnl_locked); void brcmf_txflowblock_if(struct brcmf_if *ifp, enum brcmf_netif_stop_reason reason, bool state); -- cgit v0.10.2 From bda9d01427f5b646e332e9e3d4b772477fd23c14 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Sat, 18 Jun 2016 18:49:38 +0200 Subject: brcmfmac: include also core.h header in cfg80211.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This header provides two inline functions using struct brcmf_if so we need core.h to avoid: drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h: In function ‘ndev_to_prof’: drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h:368:13: error: dereferencing pointer to incomplete type return &ifp->vif->profile; ^ drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h: In function ‘ndev_to_vif’: drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h:374:12: error: dereferencing pointer to incomplete type return ifp->vif; ^ Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h index 04bfc7e..7d77f86 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.h @@ -20,6 +20,7 @@ /* for brcmu_d11inf */ #include +#include "core.h" #include "fwil_types.h" #include "p2p.h" -- cgit v0.10.2 From 20856adf22800ef90760d8256b6cf88675709584 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Sun, 19 Jun 2016 01:55:57 +0200 Subject: brcmfmac: add missing break when deleting P2P_DEVICE MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We obviously don't want to fall through in that switch. With this change 1) We wait for event (triggered by p2p_disc) as expected 2) We remove interface manually on timeout 3) We return 0 on success instead of -ENOTSUPP Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c index 426ff05..f6241fd 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c @@ -2261,6 +2261,8 @@ int brcmf_p2p_del_vif(struct wiphy *wiphy, struct wireless_dev *wdev) return 0; brcmf_p2p_cancel_remain_on_channel(vif->ifp); brcmf_p2p_deinit_discovery(p2p); + break; + default: return -ENOTSUPP; } -- cgit v0.10.2 From 535633a5ba4ea2504fa6c33176633becf0e59339 Mon Sep 17 00:00:00 2001 From: Guy Mishol Date: Sun, 19 Jun 2016 17:08:58 +0300 Subject: wlcore: reconfigure sta rates on authorization Since stations can now be added before association (NL80211_FEATURE_FULL_AP_CLIENT_STATE support), no supported rates are set when the station is added to the fw, resulting in fw recovery. Fix it by first configuring the AP basic rates as the station configured rates (when the station is first added to the driver), and after the station was authorized re-configure it, now with the actual supported rates. Signed-off-by: Guy Mishol Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ti/wlcore/cmd.c b/drivers/net/wireless/ti/wlcore/cmd.c index 3315356..5f360ce 100644 --- a/drivers/net/wireless/ti/wlcore/cmd.c +++ b/drivers/net/wireless/ti/wlcore/cmd.c @@ -1566,6 +1566,13 @@ int wl12xx_cmd_add_peer(struct wl1271 *wl, struct wl12xx_vif *wlvif, cpu_to_le32(wl1271_tx_enabled_rates_get(wl, sta_rates, wlvif->band)); + if (!cmd->supported_rates) { + wl1271_debug(DEBUG_CMD, + "peer has no supported rates yet, configuring basic rates: 0x%x", + wlvif->basic_rate_set); + cmd->supported_rates = cpu_to_le32(wlvif->basic_rate_set); + } + wl1271_debug(DEBUG_CMD, "new peer rates=0x%x queues=0x%x", cmd->supported_rates, sta->uapsd_queues); diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c index 10fd24c..a53033d 100644 --- a/drivers/net/wireless/ti/wlcore/main.c +++ b/drivers/net/wireless/ti/wlcore/main.c @@ -5091,6 +5091,11 @@ static int wl12xx_update_sta_state(struct wl1271 *wl, if (ret < 0) return ret; + /* reconfigure rates */ + ret = wl12xx_cmd_add_peer(wl, wlvif, sta, wl_sta->hlid); + if (ret < 0) + return ret; + ret = wl1271_acx_set_ht_capabilities(wl, &sta->ht_cap, true, wl_sta->hlid); if (ret) -- cgit v0.10.2 From 124579de462e6c232c4c321c3ec1ed81964cf78b Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Thu, 14 Apr 2016 13:17:27 -0700 Subject: fm10k: don't use BIT() macro where the value isn't a bitmask The FM10K_MAX_DATA_PER_TXD is really just using a bitshift as a power of 2 operation in an efficient manner. We shouldn't represent this as a BIT() because that obscures the intention of the operation. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k.h b/drivers/net/ethernet/intel/fm10k/fm10k.h index fcf106e..e98b86b 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k.h +++ b/drivers/net/ethernet/intel/fm10k/fm10k.h @@ -406,7 +406,7 @@ static inline u16 fm10k_desc_unused(struct fm10k_ring *ring) (&(((union fm10k_rx_desc *)((R)->desc))[i])) #define FM10K_MAX_TXD_PWR 14 -#define FM10K_MAX_DATA_PER_TXD BIT(FM10K_MAX_TXD_PWR) +#define FM10K_MAX_DATA_PER_TXD (1u << FM10K_MAX_TXD_PWR) /* Tx Descriptors needed, worst case */ #define TXD_USE_COUNT(S) DIV_ROUND_UP((S), FM10K_MAX_DATA_PER_TXD) -- cgit v0.10.2 From fb5677aa26a2ae2c94ffb889ab6802a42bf54b0b Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Fri, 15 Apr 2016 13:00:46 -0400 Subject: fm10k: Align Rx buffers to 512B blocks While reviewing the i40e driver changes to support page based receive I realized that I had overlooked the fact that the fm10k hardware required a 512 byte alignment for Rx buffers. This patch is meant to address that by changing the alignment for Rx buffers to 512 bytes instead of allowing it to be L1 cache aligned. Signed-off-by: Alexander Duyck Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c index 0e166e9..53d02416 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c @@ -272,7 +272,7 @@ static bool fm10k_add_rx_frag(struct fm10k_rx_buffer *rx_buffer, #if (PAGE_SIZE < 8192) unsigned int truesize = FM10K_RX_BUFSZ; #else - unsigned int truesize = SKB_DATA_ALIGN(size); + unsigned int truesize = ALIGN(size, 512); #endif unsigned int pull_len; -- cgit v0.10.2 From 34875887f360d7bd0b7f0a89f7c6d65eca616ee3 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Mon, 18 Apr 2016 15:45:00 -0700 Subject: fm10k: fix incorrect index calculation in fm10k_write_reta The index calculated when looping through the indir array passed to fm10k_write_reta was incorrectly calculated as the first part i needs to be multiplied by 4. Fixes: 0cfea7a65738 ("fm10k: fix possible null pointer deref after kcalloc", 2016-04-13) Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c index 9c0d875..9b51954 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c @@ -983,9 +983,10 @@ void fm10k_write_reta(struct fm10k_intfc *interface, const u32 *indir) /* generate a new table if we weren't given one */ for (j = 0; j < 4; j++) { if (indir) - n = indir[i + j]; + n = indir[4 * i + j]; else - n = ethtool_rxfh_indir_default(i + j, rss_i); + n = ethtool_rxfh_indir_default(4 * i + j, + rss_i); table[j] = n; } -- cgit v0.10.2 From 3d05b15b03daa4e8c350a97d0d83d2c2abc8b8ef Mon Sep 17 00:00:00 2001 From: Denys Vlasenko Date: Fri, 6 May 2016 21:41:51 +0200 Subject: e1000e: prevent division by zero if TIMINCA is zero Users report that under VMWare, er32(TIMINCA) returns zero. This causes division by zero at init time as follows: ==> incvalue = er32(TIMINCA) & E1000_TIMINCA_INCVALUE_MASK; for (i = 0; i < E1000_MAX_82574_SYSTIM_REREADS; i++) { /* latch SYSTIMH on read of SYSTIML */ systim_next = (cycle_t)er32(SYSTIML); systim_next |= (cycle_t)er32(SYSTIMH) << 32; time_delta = systim_next - systim; temp = time_delta; ====> rem = do_div(temp, incvalue); This change makes kernel survive this, and users report that NIC does work after this change. Since on real hardware incvalue is never zero, this should not affect real hardware use case. Signed-off-by: Denys Vlasenko Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c index 75e6089..9d5bab8 100644 --- a/drivers/net/ethernet/intel/e1000e/netdev.c +++ b/drivers/net/ethernet/intel/e1000e/netdev.c @@ -4352,7 +4352,8 @@ static cycle_t e1000e_cyclecounter_read(const struct cyclecounter *cc) time_delta = systim_next - systim; temp = time_delta; - rem = do_div(temp, incvalue); + /* VMWare users have seen incvalue of zero, don't div / 0 */ + rem = incvalue ? do_div(temp, incvalue) : (time_delta != 0); systim = systim_next; -- cgit v0.10.2 From 1ecedc926be12a91271e41913ebeba8cf32e9a6c Mon Sep 17 00:00:00 2001 From: Amritha Nambiar Date: Fri, 6 May 2016 19:09:51 -0700 Subject: ixgbe: Fix deleting link filters for cls_u32 offloads On deleting filters which are links to a child hash table, the filters in the child hash table must be cleared from the hardware if there is no link between the parent and child hash table. Verified with the following filters: Create a child hash table: handle 1: u32 divisor 1 Link to the child hash table from parent hash table: handle 800:0:10 u32 ht 800: link 1: \ offset at 0 mask 0f00 shift 6 plus 0 eat \ match ip protocol 6 ff match ip dst 15.0.0.1/32 Add filters into child hash table: handle 1:0:2 u32 ht 1: \ match tcp src 22 ffff action drop handle 1:0:3 u32 ht 1: \ match tcp src 33 ffff action drop Delete link filter from parent hash table: handle 800:0:10 u32 Signed-off-by: Amritha Nambiar Acked-by: Sridhar Samudrala Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 468fa9d..75e6855 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -8300,14 +8300,50 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc) static int ixgbe_delete_clsu32(struct ixgbe_adapter *adapter, struct tc_cls_u32_offload *cls) { + u32 hdl = cls->knode.handle; u32 uhtid = TC_U32_USERHTID(cls->knode.handle); - u32 loc; - int err; + u32 loc = cls->knode.handle & 0xfffff; + int err = 0, i, j; + struct ixgbe_jump_table *jump = NULL; + + if (loc > IXGBE_MAX_HW_ENTRIES) + return -EINVAL; if ((uhtid != 0x800) && (uhtid >= IXGBE_MAX_LINK_HANDLE)) return -EINVAL; - loc = cls->knode.handle & 0xfffff; + /* Clear this filter in the link data it is associated with */ + if (uhtid != 0x800) { + jump = adapter->jump_tables[uhtid]; + if (jump) + clear_bit(loc - 1, jump->child_loc_map); + } + + /* Check if the filter being deleted is a link */ + for (i = 1; i < IXGBE_MAX_LINK_HANDLE; i++) { + jump = adapter->jump_tables[i]; + if (jump && jump->link_hdl == hdl) { + /* Delete filters in the hardware in the child hash + * table associated with this link + */ + for (j = 0; j < IXGBE_MAX_HW_ENTRIES; j++) { + if (!test_bit(j, jump->child_loc_map)) + continue; + spin_lock(&adapter->fdir_perfect_lock); + err = ixgbe_update_ethtool_fdir_entry(adapter, + NULL, + j + 1); + spin_unlock(&adapter->fdir_perfect_lock); + clear_bit(j, jump->child_loc_map); + } + /* Remove resources for this link */ + kfree(jump->input); + kfree(jump->mask); + kfree(jump); + adapter->jump_tables[i] = NULL; + return err; + } + } spin_lock(&adapter->fdir_perfect_lock); err = ixgbe_update_ethtool_fdir_entry(adapter, NULL, loc); @@ -8541,6 +8577,18 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, if (!test_bit(link_uhtid - 1, &adapter->tables)) return err; + /* Multiple filters as links to the same hash table are not + * supported. To add a new filter with the same next header + * but different match/jump conditions, create a new hash table + * and link to it. + */ + if (adapter->jump_tables[link_uhtid] && + (adapter->jump_tables[link_uhtid])->link_hdl) { + e_err(drv, "Link filter exists for link: %x\n", + link_uhtid); + return err; + } + for (i = 0; nexthdr[i].jump; i++) { if (nexthdr[i].o != cls->knode.sel->offoff || nexthdr[i].s != cls->knode.sel->offshift || @@ -8558,10 +8606,12 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, mask = kzalloc(sizeof(*mask), GFP_KERNEL); if (!mask) { err = -ENOMEM; - goto free_input; + goto err_out; } jump->input = input; jump->mask = mask; + jump->link_hdl = cls->knode.handle; + err = ixgbe_clsu32_build_input(input, mask, cls, field_ptr, &nexthdr[i]); if (!err) { @@ -8579,7 +8629,7 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, mask = kzalloc(sizeof(*mask), GFP_KERNEL); if (!mask) { err = -ENOMEM; - goto free_input; + goto err_out; } if ((uhtid != 0x800) && (adapter->jump_tables[uhtid])) { @@ -8620,14 +8670,25 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, ixgbe_update_ethtool_fdir_entry(adapter, input, input->sw_idx); spin_unlock(&adapter->fdir_perfect_lock); + if ((uhtid != 0x800) && (adapter->jump_tables[uhtid])) { + struct ixgbe_jump_table *link = adapter->jump_tables[uhtid]; + + if (test_bit(loc - 1, link->child_loc_map)) { + e_err(drv, "Filter: %x exists in hash table: %x\n", + loc, uhtid); + err = -EINVAL; + goto free_mask; + } + set_bit(loc - 1, link->child_loc_map); + } kfree(mask); return err; err_out_w_lock: spin_unlock(&adapter->fdir_perfect_lock); err_out: - kfree(mask); -free_input: kfree(input); +free_mask: + kfree(mask); free_jump: kfree(jump); return err; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h index a8bed3d..538a1c54 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_model.h @@ -42,8 +42,12 @@ struct ixgbe_jump_table { struct ixgbe_mat_field *mat; struct ixgbe_fdir_filter *input; union ixgbe_atr_input *mask; + u32 link_hdl; + unsigned long child_loc_map[32]; }; +#define IXGBE_MAX_HW_ENTRIES 2045 + static inline int ixgbe_mat_prgm_sip(struct ixgbe_fdir_filter *input, union ixgbe_atr_input *mask, u32 val, u32 m) -- cgit v0.10.2 From 12746fd21e9fe1bf9103a28f15abbf343a3a66d0 Mon Sep 17 00:00:00 2001 From: Amritha Nambiar Date: Mon, 16 May 2016 18:33:20 -0700 Subject: ixgbe: Error handler for duplicate filter locations in hardware for cls_u32 offloads For u32 classifier filters, avoid overwriting existing filter in a hardware location without removing it first, to clean up inconsistencies due to duplicate values for filter location. Verified with the following filters: Create child hash tables: handle 1: u32 divisor 1 handle 2: u32 divisor 1 Link to the child hash table from parent hash table: handle 800:0:11 u32 ht 800: link 1: \ offset at 0 mask 0f00 shift 6 plus 0 eat \ match ip protocol 6 ff match ip dst 15.0.0.1/32 handle 800:0:12 u32 ht 800: link 2: \ offset at 0 mask 0f00 shift 6 plus 0 eat \ match ip protocol 17 ff match ip dst 16.0.0.1/32 Add filter into child hash table: handle 1:0:3 u32 ht 1: \ match tcp src 22 ffff action drop Add another filter to the same location: handle 2:0:3 u32 ht 2: \ match tcp src 33 ffff action drop Signed-off-by: Amritha Nambiar Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 75e6855..fd5a761 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -8315,8 +8315,11 @@ static int ixgbe_delete_clsu32(struct ixgbe_adapter *adapter, /* Clear this filter in the link data it is associated with */ if (uhtid != 0x800) { jump = adapter->jump_tables[uhtid]; - if (jump) - clear_bit(loc - 1, jump->child_loc_map); + if (!jump) + return -EINVAL; + if (!test_bit(loc - 1, jump->child_loc_map)) + return -EINVAL; + clear_bit(loc - 1, jump->child_loc_map); } /* Check if the filter being deleted is a link */ @@ -8606,7 +8609,7 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, mask = kzalloc(sizeof(*mask), GFP_KERNEL); if (!mask) { err = -ENOMEM; - goto err_out; + goto free_input; } jump->input = input; jump->mask = mask; @@ -8629,7 +8632,7 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, mask = kzalloc(sizeof(*mask), GFP_KERNEL); if (!mask) { err = -ENOMEM; - goto err_out; + goto free_input; } if ((uhtid != 0x800) && (adapter->jump_tables[uhtid])) { @@ -8639,6 +8642,20 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, if ((adapter->jump_tables[uhtid])->mask) memcpy(mask, (adapter->jump_tables[uhtid])->mask, sizeof(*mask)); + + /* Lookup in all child hash tables if this location is already + * filled with a filter + */ + for (i = 1; i < IXGBE_MAX_LINK_HANDLE; i++) { + struct ixgbe_jump_table *link = adapter->jump_tables[i]; + + if (link && (test_bit(loc - 1, link->child_loc_map))) { + e_err(drv, "Filter exists in location: %x\n", + loc); + err = -EINVAL; + goto err_out; + } + } } err = ixgbe_clsu32_build_input(input, mask, cls, field_ptr, NULL); if (err) @@ -8670,25 +8687,17 @@ static int ixgbe_configure_clsu32(struct ixgbe_adapter *adapter, ixgbe_update_ethtool_fdir_entry(adapter, input, input->sw_idx); spin_unlock(&adapter->fdir_perfect_lock); - if ((uhtid != 0x800) && (adapter->jump_tables[uhtid])) { - struct ixgbe_jump_table *link = adapter->jump_tables[uhtid]; + if ((uhtid != 0x800) && (adapter->jump_tables[uhtid])) + set_bit(loc - 1, (adapter->jump_tables[uhtid])->child_loc_map); - if (test_bit(loc - 1, link->child_loc_map)) { - e_err(drv, "Filter: %x exists in hash table: %x\n", - loc, uhtid); - err = -EINVAL; - goto free_mask; - } - set_bit(loc - 1, link->child_loc_map); - } kfree(mask); return err; err_out_w_lock: spin_unlock(&adapter->fdir_perfect_lock); err_out: - kfree(input); -free_mask: kfree(mask); +free_input: + kfree(input); free_jump: kfree(jump); return err; -- cgit v0.10.2 From 462f11888207934bbecd0f154f6cf0bedac5ecd0 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 24 May 2016 13:56:27 -0700 Subject: igb: introduce ptp_flags variable and use it to replace IGB_FLAG_PTP Upcoming patches will introduce new PTP specific flags. To avoid cluttering the normal flags variable, introduce PTP specific "ptp_flags" variable for this purpose, and move IGB_FLAG_PTP to become IGB_PTP_ENABLED. Signed-off-by: Jacob Keller Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h index b9609af..1e18a9e 100644 --- a/drivers/net/ethernet/intel/igb/igb.h +++ b/drivers/net/ethernet/intel/igb/igb.h @@ -445,6 +445,7 @@ struct igb_adapter { unsigned long ptp_tx_start; unsigned long last_rx_ptp_check; unsigned long last_rx_timestamp; + unsigned int ptp_flags; spinlock_t tmreg_lock; struct cyclecounter cc; struct timecounter tc; @@ -474,12 +475,14 @@ struct igb_adapter { u16 eee_advert; }; +/* flags controlling PTP/1588 function */ +#define IGB_PTP_ENABLED BIT(0) + #define IGB_FLAG_HAS_MSI BIT(0) #define IGB_FLAG_DCA_ENABLED BIT(1) #define IGB_FLAG_QUAD_PORT_A BIT(2) #define IGB_FLAG_QUEUE_PAIRS BIT(3) #define IGB_FLAG_DMAC BIT(4) -#define IGB_FLAG_PTP BIT(5) #define IGB_FLAG_RSS_FIELD_IPV4_UDP BIT(6) #define IGB_FLAG_RSS_FIELD_IPV6_UDP BIT(7) #define IGB_FLAG_WOL_SUPPORTED BIT(8) diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c index f097c5a..504102b 100644 --- a/drivers/net/ethernet/intel/igb/igb_ptp.c +++ b/drivers/net/ethernet/intel/igb/igb_ptp.c @@ -684,6 +684,7 @@ void igb_ptp_rx_hang(struct igb_adapter *adapter) u32 tsyncrxctl = rd32(E1000_TSYNCRXCTL); unsigned long rx_event; + /* Other hardware uses per-packet timestamps */ if (hw->mac.type != e1000_82576) return; @@ -1156,7 +1157,7 @@ void igb_ptp_init(struct igb_adapter *adapter) } else { dev_info(&adapter->pdev->dev, "added PHC on %s\n", adapter->netdev->name); - adapter->flags |= IGB_FLAG_PTP; + adapter->ptp_flags |= IGB_PTP_ENABLED; } } @@ -1194,7 +1195,7 @@ void igb_ptp_stop(struct igb_adapter *adapter) ptp_clock_unregister(adapter->ptp_clock); dev_info(&adapter->pdev->dev, "removed PHC on %s\n", adapter->netdev->name); - adapter->flags &= ~IGB_FLAG_PTP; + adapter->ptp_flags &= ~IGB_PTP_ENABLED; } } @@ -1209,7 +1210,7 @@ void igb_ptp_reset(struct igb_adapter *adapter) struct e1000_hw *hw = &adapter->hw; unsigned long flags; - if (!(adapter->flags & IGB_FLAG_PTP)) + if (!(adapter->ptp_flags & IGB_PTP_ENABLED)) return; /* reset the tstamp_config */ -- cgit v0.10.2 From 63737166a056fdcc83be46175ad4f63538279d7c Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 24 May 2016 13:56:28 -0700 Subject: igb: introduce IGB_PTP_OVERFLOW_CHECK flag Don't continue to use complex MAC type checks for handling various cases where we have overflow check code. Make this code more obvious by introducing a flag which is enabled for hardware that needs these checks. Signed-off-by: Jacob Keller Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h index 1e18a9e..38daaab 100644 --- a/drivers/net/ethernet/intel/igb/igb.h +++ b/drivers/net/ethernet/intel/igb/igb.h @@ -477,6 +477,7 @@ struct igb_adapter { /* flags controlling PTP/1588 function */ #define IGB_PTP_ENABLED BIT(0) +#define IGB_PTP_OVERFLOW_CHECK BIT(1) #define IGB_FLAG_HAS_MSI BIT(0) #define IGB_FLAG_DCA_ENABLED BIT(1) diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c index 504102b..88d367d 100644 --- a/drivers/net/ethernet/intel/igb/igb_ptp.c +++ b/drivers/net/ethernet/intel/igb/igb_ptp.c @@ -1067,6 +1067,7 @@ void igb_ptp_init(struct igb_adapter *adapter) adapter->cc.shift = IGB_82576_TSYNC_SHIFT; /* Dial the nominal frequency. */ wr32(E1000_TIMINCA, INCPERIOD_82576 | INCVALUE_82576); + adapter->ptp_flags |= IGB_PTP_OVERFLOW_CHECK; break; case e1000_82580: case e1000_i354: @@ -1087,6 +1088,7 @@ void igb_ptp_init(struct igb_adapter *adapter) adapter->cc.shift = 0; /* Enable the timer functions by clearing bit 31. */ wr32(E1000_TSAUXC, 0x0); + adapter->ptp_flags |= IGB_PTP_OVERFLOW_CHECK; break; case e1000_i210: case e1000_i211: @@ -1132,7 +1134,9 @@ void igb_ptp_init(struct igb_adapter *adapter) } else { timecounter_init(&adapter->tc, &adapter->cc, ktime_to_ns(ktime_get_real())); + } + if (adapter->ptp_flags & IGB_PTP_OVERFLOW_CHECK) { INIT_DELAYED_WORK(&adapter->ptp_overflow_work, igb_ptp_overflow_check); @@ -1169,20 +1173,11 @@ void igb_ptp_init(struct igb_adapter *adapter) **/ void igb_ptp_stop(struct igb_adapter *adapter) { - switch (adapter->hw.mac.type) { - case e1000_82576: - case e1000_82580: - case e1000_i354: - case e1000_i350: - cancel_delayed_work_sync(&adapter->ptp_overflow_work); - break; - case e1000_i210: - case e1000_i211: - /* No delayed work to cancel. */ - break; - default: + if (!(adapter->ptp_flags & IGB_PTP_ENABLED)) return; - } + + if (adapter->ptp_flags & IGB_PTP_OVERFLOW_CHECK) + cancel_delayed_work_sync(&adapter->ptp_overflow_work); cancel_work_sync(&adapter->ptp_tx_work); if (adapter->ptp_tx_skb) { -- cgit v0.10.2 From 4f3ce71bb8d9c7591f25d3d8315249a250d208f0 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 24 May 2016 13:56:29 -0700 Subject: igb: re-use igb_ptp_reset in igb_ptp_init Modify igb_ptp_init to take advantage of igb_ptp_reset, and remove duplicated work that was occurring in both igb_ptp_reset and igb_ptp_init. In total, resetting the TSAUXC register, and resetting the system time both happen in igb_ptp_reset already. igb_ptp_reset now also takes care of starting the delayed work item for overflow checks, as well. Signed-off-by: Jacob Keller Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index ef3d642..750a5d8 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -2027,7 +2027,8 @@ void igb_reset(struct igb_adapter *adapter) wr32(E1000_VET, ETHERNET_IEEE_VLAN_TYPE); /* Re-enable PTP, where applicable. */ - igb_ptp_reset(adapter); + if (adapter->ptp_flags & IGB_PTP_ENABLED) + igb_ptp_reset(adapter); igb_get_phy_info(hw); } diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c index 88d367d..907c225 100644 --- a/drivers/net/ethernet/intel/igb/igb_ptp.c +++ b/drivers/net/ethernet/intel/igb/igb_ptp.c @@ -1043,6 +1043,13 @@ int igb_ptp_set_ts_config(struct net_device *netdev, struct ifreq *ifr) -EFAULT : 0; } +/** + * igb_ptp_init - Initialize PTP functionality + * @adapter: Board private structure + * + * This function is called at device probe to initialize the PTP + * functionality. + */ void igb_ptp_init(struct igb_adapter *adapter) { struct e1000_hw *hw = &adapter->hw; @@ -1065,8 +1072,6 @@ void igb_ptp_init(struct igb_adapter *adapter) adapter->cc.mask = CYCLECOUNTER_MASK(64); adapter->cc.mult = 1; adapter->cc.shift = IGB_82576_TSYNC_SHIFT; - /* Dial the nominal frequency. */ - wr32(E1000_TIMINCA, INCPERIOD_82576 | INCVALUE_82576); adapter->ptp_flags |= IGB_PTP_OVERFLOW_CHECK; break; case e1000_82580: @@ -1086,8 +1091,6 @@ void igb_ptp_init(struct igb_adapter *adapter) adapter->cc.mask = CYCLECOUNTER_MASK(IGB_NBITS_82580); adapter->cc.mult = 1; adapter->cc.shift = 0; - /* Enable the timer functions by clearing bit 31. */ - wr32(E1000_TSAUXC, 0x0); adapter->ptp_flags |= IGB_PTP_OVERFLOW_CHECK; break; case e1000_i210: @@ -1113,46 +1116,24 @@ void igb_ptp_init(struct igb_adapter *adapter) adapter->ptp_caps.settime64 = igb_ptp_settime_i210; adapter->ptp_caps.enable = igb_ptp_feature_enable_i210; adapter->ptp_caps.verify = igb_ptp_verify_pin; - /* Enable the timer functions by clearing bit 31. */ - wr32(E1000_TSAUXC, 0x0); break; default: adapter->ptp_clock = NULL; return; } - wrfl(); - spin_lock_init(&adapter->tmreg_lock); INIT_WORK(&adapter->ptp_tx_work, igb_ptp_tx_work); - /* Initialize the clock and overflow work for devices that need it. */ - if ((hw->mac.type == e1000_i210) || (hw->mac.type == e1000_i211)) { - struct timespec64 ts = ktime_to_timespec64(ktime_get_real()); - - igb_ptp_settime_i210(&adapter->ptp_caps, &ts); - } else { - timecounter_init(&adapter->tc, &adapter->cc, - ktime_to_ns(ktime_get_real())); - } - - if (adapter->ptp_flags & IGB_PTP_OVERFLOW_CHECK) { + if (adapter->ptp_flags & IGB_PTP_OVERFLOW_CHECK) INIT_DELAYED_WORK(&adapter->ptp_overflow_work, igb_ptp_overflow_check); - schedule_delayed_work(&adapter->ptp_overflow_work, - IGB_SYSTIM_OVERFLOW_PERIOD); - } - - /* Initialize the time sync interrupts for devices that support it. */ - if (hw->mac.type >= e1000_82580) { - wr32(E1000_TSIM, TSYNC_INTERRUPTS); - wr32(E1000_IMS, E1000_IMS_TS); - } - adapter->tstamp_config.rx_filter = HWTSTAMP_FILTER_NONE; adapter->tstamp_config.tx_type = HWTSTAMP_TX_OFF; + igb_ptp_reset(adapter); + adapter->ptp_clock = ptp_clock_register(&adapter->ptp_caps, &adapter->pdev->dev); if (IS_ERR(adapter->ptp_clock)) { @@ -1205,9 +1186,6 @@ void igb_ptp_reset(struct igb_adapter *adapter) struct e1000_hw *hw = &adapter->hw; unsigned long flags; - if (!(adapter->ptp_flags & IGB_PTP_ENABLED)) - return; - /* reset the tstamp_config */ igb_ptp_set_timestamp_mode(adapter, &adapter->tstamp_config); @@ -1244,4 +1222,10 @@ void igb_ptp_reset(struct igb_adapter *adapter) } out: spin_unlock_irqrestore(&adapter->tmreg_lock, flags); + + wrfl(); + + if (adapter->ptp_flags & IGB_PTP_OVERFLOW_CHECK) + schedule_delayed_work(&adapter->ptp_overflow_work, + IGB_SYSTIM_OVERFLOW_PERIOD); } -- cgit v0.10.2 From e3f2350de829eb0c3349f416feed81c0a3ac0732 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 24 May 2016 13:56:30 -0700 Subject: igb: implement igb_ptp_suspend Make igb_ptp_stop take advantage of this new function to reduce code duplication. Signed-off-by: Jacob Keller Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h index 38daaab..5387b3a 100644 --- a/drivers/net/ethernet/intel/igb/igb.h +++ b/drivers/net/ethernet/intel/igb/igb.h @@ -550,6 +550,7 @@ void igb_set_fw_version(struct igb_adapter *); void igb_ptp_init(struct igb_adapter *adapter); void igb_ptp_stop(struct igb_adapter *adapter); void igb_ptp_reset(struct igb_adapter *adapter); +void igb_ptp_suspend(struct igb_adapter *adapter); void igb_ptp_rx_hang(struct igb_adapter *adapter); void igb_ptp_rx_rgtstamp(struct igb_q_vector *q_vector, struct sk_buff *skb); void igb_ptp_rx_pktstamp(struct igb_q_vector *q_vector, unsigned char *va, diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c index 907c225..e61b647 100644 --- a/drivers/net/ethernet/intel/igb/igb_ptp.c +++ b/drivers/net/ethernet/intel/igb/igb_ptp.c @@ -1147,12 +1147,13 @@ void igb_ptp_init(struct igb_adapter *adapter) } /** - * igb_ptp_stop - Disable PTP device and stop the overflow check. - * @adapter: Board private structure. + * igb_ptp_suspend - Disable PTP work items and prepare for suspend + * @adapter: Board private structure * - * This function stops the PTP support and cancels the delayed work. - **/ -void igb_ptp_stop(struct igb_adapter *adapter) + * This function stops the overflow check work and PTP Tx timestamp work, and + * will prepare the device for OS suspend. + */ +void igb_ptp_suspend(struct igb_adapter *adapter) { if (!(adapter->ptp_flags & IGB_PTP_ENABLED)) return; @@ -1166,6 +1167,17 @@ void igb_ptp_stop(struct igb_adapter *adapter) adapter->ptp_tx_skb = NULL; clear_bit_unlock(__IGB_PTP_TX_IN_PROGRESS, &adapter->state); } +} + +/** + * igb_ptp_stop - Disable PTP device and stop the overflow check. + * @adapter: Board private structure. + * + * This function stops the PTP support and cancels the delayed work. + **/ +void igb_ptp_stop(struct igb_adapter *adapter) +{ + igb_ptp_suspend(adapter); if (adapter->ptp_clock) { ptp_clock_unregister(adapter->ptp_clock); -- cgit v0.10.2 From 8646f7b4cdf2d0557e718c4524a3e31455b92ad7 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 24 May 2016 13:56:31 -0700 Subject: igb: call igb_ptp_suspend during suspend/resume cycle Properly stop the extra workqueue items and ensure that we resume cleanly. This is better than using igb_ptp_init and igb_ptp_stop since these functions destroy the PHC device, which will cause other problems if we do so. Since igb_ptp_reset now re-schedules the work-queue item we don't need an equivalent igb_ptp_resume in the resume workflow. Signed-off-by: Jacob Keller Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index 750a5d8..a15f826 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -7528,6 +7528,8 @@ static int __igb_shutdown(struct pci_dev *pdev, bool *enable_wake, if (netif_running(netdev)) __igb_close(netdev, true); + igb_ptp_suspend(adapter); + igb_clear_interrupt_scheme(adapter); #ifdef CONFIG_PM -- cgit v0.10.2 From 0a38c17a21a0965b4853211afa1d3e85428e6170 Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Wed, 1 Jun 2016 21:10:09 +0530 Subject: fm10k: Remove create_workqueue alloc_workqueue replaces deprecated create_workqueue(). A dedicated workqueue has been used since the workitem (viz fm10k_service_task, which manages and runs other subtasks) is involved in normal device operation and requires forward progress under memory pressure. create_workqueue has been replaced with alloc_workqueue with max_active as 0 since there is no need for throttling the number of active work items. Since network devices may be used in memory reclaim path, WQ_MEM_RECLAIM has been set to guarantee forward progress. flush_workqueue is unnecessary since destroy_workqueue() itself calls drain_workqueue() which flushes repeatedly till the workqueue becomes empty. Hence the call to flush_workqueue() has been dropped. Signed-off-by: Bhaktipriya Shridhar Acked-by: Tejun Heo Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c index 53d02416..a9ccc1e 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c @@ -56,7 +56,7 @@ static int __init fm10k_init_module(void) pr_info("%s\n", fm10k_copyright); /* create driver workqueue */ - fm10k_workqueue = create_workqueue("fm10k"); + fm10k_workqueue = alloc_workqueue("fm10k", WQ_MEM_RECLAIM, 0); fm10k_dbg_init(); @@ -77,7 +77,6 @@ static void __exit fm10k_exit_module(void) fm10k_dbg_exit(); /* destroy driver workqueue */ - flush_workqueue(fm10k_workqueue); destroy_workqueue(fm10k_workqueue); } module_exit(fm10k_exit_module); -- cgit v0.10.2 From 918b89e77fa554e185a5cc09d10655397aacdfa2 Mon Sep 17 00:00:00 2001 From: Tony Nguyen Date: Wed, 1 Jun 2016 09:50:43 -0700 Subject: ixgbe: Correct reporting of timestamping for x550 Update ixgbe_ethtool_get_ts_info() to show that x550 supports hardware timestamping of all packets. Reported-by: Guy Harris Signed-off-by: Tony Nguyen Signed-off-by: Jacob Keller Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c index 59b771b..8a84507 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -2991,10 +2991,15 @@ static int ixgbe_get_ts_info(struct net_device *dev, { struct ixgbe_adapter *adapter = netdev_priv(dev); + /* we always support timestamping disabled */ + info->rx_filters = BIT(HWTSTAMP_FILTER_NONE); + switch (adapter->hw.mac.type) { case ixgbe_mac_X550: case ixgbe_mac_X550EM_x: case ixgbe_mac_x550em_a: + info->rx_filters |= BIT(HWTSTAMP_FILTER_ALL); + /* fallthrough */ case ixgbe_mac_X540: case ixgbe_mac_82599EB: info->so_timestamping = @@ -3014,8 +3019,7 @@ static int ixgbe_get_ts_info(struct net_device *dev, BIT(HWTSTAMP_TX_OFF) | BIT(HWTSTAMP_TX_ON); - info->rx_filters = - BIT(HWTSTAMP_FILTER_NONE) | + info->rx_filters |= BIT(HWTSTAMP_FILTER_PTP_V1_L4_SYNC) | BIT(HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ) | BIT(HWTSTAMP_FILTER_PTP_V2_EVENT); -- cgit v0.10.2 From 581e0c7df90b1a7f92e7ac3e69000b414319f161 Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Wed, 1 Jun 2016 18:59:44 -0700 Subject: ixgbe: fix spoofed packets with macvlans When setting spoofing, both VLAN and MAC need to be set together. This change resolves an issue where MAC-VLANs on the VF fail to pass traffic due to spoofed packets. Signed-off-by: Emil Tantilov Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c index c5caacd..8618599 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c @@ -954,6 +954,7 @@ static int ixgbe_set_vf_macvlan_msg(struct ixgbe_adapter *adapter, struct ixgbe_hw *hw = &adapter->hw; hw->mac.ops.set_mac_anti_spoofing(hw, false, vf); + hw->mac.ops.set_vlan_anti_spoofing(hw, false, vf); } } -- cgit v0.10.2 From 64f2525ca4e76b1704b867458808ed6ffc58b803 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Fri, 3 Jun 2016 23:03:25 +0200 Subject: igb: Only DMA sync frame length On some platforms, syncing a buffer for DMA is expensive. Rather than sync the whole 2K receive buffer, only synchronise the length of the frame, which will typically be the MTU, or a much smaller TCP ACK. For an IMX6Q, this gives around 6% increased TCP receive performance, which is cache operations bound and reduces CPU load for TCP transmit. Signed-off-by: Andrew Lunn Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index a15f826..9bcba42 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -6856,12 +6856,12 @@ static bool igb_can_reuse_rx_page(struct igb_rx_buffer *rx_buffer, **/ static bool igb_add_rx_frag(struct igb_ring *rx_ring, struct igb_rx_buffer *rx_buffer, + unsigned int size, union e1000_adv_rx_desc *rx_desc, struct sk_buff *skb) { struct page *page = rx_buffer->page; unsigned char *va = page_address(page) + rx_buffer->page_offset; - unsigned int size = le16_to_cpu(rx_desc->wb.upper.length); #if (PAGE_SIZE < 8192) unsigned int truesize = IGB_RX_BUFSZ; #else @@ -6913,6 +6913,7 @@ static struct sk_buff *igb_fetch_rx_buffer(struct igb_ring *rx_ring, union e1000_adv_rx_desc *rx_desc, struct sk_buff *skb) { + unsigned int size = le16_to_cpu(rx_desc->wb.upper.length); struct igb_rx_buffer *rx_buffer; struct page *page; @@ -6948,11 +6949,11 @@ static struct sk_buff *igb_fetch_rx_buffer(struct igb_ring *rx_ring, dma_sync_single_range_for_cpu(rx_ring->dev, rx_buffer->dma, rx_buffer->page_offset, - IGB_RX_BUFSZ, + size, DMA_FROM_DEVICE); /* pull page into skb */ - if (igb_add_rx_frag(rx_ring, rx_buffer, rx_desc, skb)) { + if (igb_add_rx_frag(rx_ring, rx_buffer, size, rx_desc, skb)) { /* hand second half of page back to the ring */ igb_reuse_rx_page(rx_ring, rx_buffer); } else { -- cgit v0.10.2 From 7bed2ab8c62aed4e9ba55373d2a81faf481c3041 Mon Sep 17 00:00:00 2001 From: Simon Wunderlich Date: Thu, 9 Jun 2016 15:46:49 +0200 Subject: batman-adv: Start new development cycle Signed-off-by: Simon Wunderlich Signed-off-by: Sven Eckelmann diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 76925266..c356d91 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -24,7 +24,7 @@ #define BATADV_DRIVER_DEVICE "batman-adv" #ifndef BATADV_SOURCE_VERSION -#define BATADV_SOURCE_VERSION "2016.2" +#define BATADV_SOURCE_VERSION "2016.3" #endif /* B.A.T.M.A.N. parameters */ -- cgit v0.10.2 From 92d2b1a5b35e05a5c4506cd73a6a5dc5f8394b60 Mon Sep 17 00:00:00 2001 From: Antonio Quartulli Date: Wed, 25 May 2016 23:27:33 +0800 Subject: batman-adv: statically print gateway table header To make it easier to search through the code it is better to print static strings directly instead of using format strings printing constants. This was addressed in a previous patch, but the Gateway table header was not updated accordingly. Signed-off-by: Antonio Quartulli Reviewed-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c index 5839c56..b9d72e9 100644 --- a/net/batman-adv/gateway_client.c +++ b/net/batman-adv/gateway_client.c @@ -638,8 +638,7 @@ int batadv_gw_client_seq_print_text(struct seq_file *seq, void *offset) goto out; seq_printf(seq, - " %-12s (%s/%i) %17s [%10s]: advertised uplink bandwidth ... [B.A.T.M.A.N. adv %s, MainIF/MAC: %s/%pM (%s)]\n", - "Gateway", "#", BATADV_TQ_MAX_VALUE, "Nexthop", "outgoingIF", + " Gateway (#/255) Nexthop [outgoingIF]: advertised uplink bandwidth ... [B.A.T.M.A.N. adv %s, MainIF/MAC: %s/%pM (%s)]\n", BATADV_SOURCE_VERSION, primary_if->net_dev->name, primary_if->net_dev->dev_addr, net_dev->name); -- cgit v0.10.2 From 118dc950fc2156de9c2f1d31be54aa37f2d3d116 Mon Sep 17 00:00:00 2001 From: Simon Wunderlich Date: Fri, 20 May 2016 20:42:19 +0200 Subject: batman-adv: remove unused vid local variable in tt seq print Signed-off-by: Simon Wunderlich Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index feaf492b..87bb203 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -995,7 +995,6 @@ int batadv_tt_local_seq_print_text(struct seq_file *seq, void *offset) struct batadv_tt_local_entry *tt_local; struct batadv_hard_iface *primary_if; struct hlist_head *head; - unsigned short vid; u32 i; int last_seen_secs; int last_seen_msecs; @@ -1022,7 +1021,6 @@ int batadv_tt_local_seq_print_text(struct seq_file *seq, void *offset) tt_local = container_of(tt_common_entry, struct batadv_tt_local_entry, common); - vid = tt_common_entry->vid; last_seen_jiffies = jiffies - tt_local->last_seen; last_seen_msecs = jiffies_to_msecs(last_seen_jiffies); last_seen_secs = last_seen_msecs / 1000; -- cgit v0.10.2 From 29cac32b67de6675b79f665567305fa57283f327 Mon Sep 17 00:00:00 2001 From: Marek Lindner Date: Mon, 9 May 2016 18:42:11 +0800 Subject: batman-adv: document sysfs files in alphabetical order Also update obsolete email address. Signed-off-by: Marek Lindner Acked-by: Antonio Quartulli Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/Documentation/ABI/testing/sysfs-class-net-batman-adv b/Documentation/ABI/testing/sysfs-class-net-batman-adv index 518f6a1..77d3083 100644 --- a/Documentation/ABI/testing/sysfs-class-net-batman-adv +++ b/Documentation/ABI/testing/sysfs-class-net-batman-adv @@ -1,13 +1,4 @@ -What: /sys/class/net//batman-adv/throughput_override -Date: Feb 2014 -Contact: Antonio Quartulli -description: - Defines the throughput value to be used by B.A.T.M.A.N. V - when estimating the link throughput using this interface. - If the value is set to 0 then batman-adv will try to - estimate the throughput by itself. - What: /sys/class/net//batman-adv/elp_interval Date: Feb 2014 Contact: Linus Lüssing @@ -28,3 +19,12 @@ Description: The /sys/class/net//batman-adv/mesh_iface file displays the batman mesh interface this currently is associated with. + +What: /sys/class/net//batman-adv/throughput_override +Date: Feb 2014 +Contact: Antonio Quartulli +description: + Defines the throughput value to be used by B.A.T.M.A.N. V + when estimating the link throughput using this interface. + If the value is set to 0 then batman-adv will try to + estimate the throughput by itself. -- cgit v0.10.2 From 21c02be7cfc5c6b7927290716a243674f134242b Mon Sep 17 00:00:00 2001 From: Marek Lindner Date: Mon, 9 May 2016 18:42:12 +0800 Subject: batman-adv: update elp interval documentation Signed-off-by: Marek Lindner Acked-by: Antonio Quartulli Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/Documentation/ABI/testing/sysfs-class-net-batman-adv b/Documentation/ABI/testing/sysfs-class-net-batman-adv index 77d3083..8981068 100644 --- a/Documentation/ABI/testing/sysfs-class-net-batman-adv +++ b/Documentation/ABI/testing/sysfs-class-net-batman-adv @@ -4,7 +4,7 @@ Date: Feb 2014 Contact: Linus Lüssing Description: Defines the interval in milliseconds in which batman - sends its probing packets for link quality measurements. + emits probing packets for neighbor sensing (ELP). What: /sys/class/net//batman-adv/iface_status Date: May 2010 -- cgit v0.10.2 From 6f0a6b5ee84fee29abd634e69ea67d6cc87817a5 Mon Sep 17 00:00:00 2001 From: Marek Lindner Date: Tue, 3 May 2016 01:52:08 +0800 Subject: batman-adv: refactor batadv_neigh_node_* functions to follow common style Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index ce2f203..aa11296 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -336,7 +336,8 @@ batadv_iv_ogm_neigh_new(struct batadv_hard_iface *hard_iface, { struct batadv_neigh_node *neigh_node; - neigh_node = batadv_neigh_node_new(orig_node, hard_iface, neigh_addr); + neigh_node = batadv_neigh_node_get_or_create(orig_node, + hard_iface, neigh_addr); if (!neigh_node) goto out; diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c index df42eb1..8909d1e 100644 --- a/net/batman-adv/bat_v_elp.c +++ b/net/batman-adv/bat_v_elp.c @@ -443,7 +443,8 @@ static void batadv_v_elp_neigh_update(struct batadv_priv *bat_priv, if (!orig_neigh) return; - neigh = batadv_neigh_node_new(orig_neigh, if_incoming, neigh_addr); + neigh = batadv_neigh_node_get_or_create(orig_neigh, + if_incoming, neigh_addr); if (!neigh) goto orig_free; diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c index 473ebb9..23ea9bf 100644 --- a/net/batman-adv/bat_v_ogm.c +++ b/net/batman-adv/bat_v_ogm.c @@ -683,8 +683,8 @@ static void batadv_v_ogm_process(const struct sk_buff *skb, int ogm_offset, if (!orig_node) return; - neigh_node = batadv_neigh_node_new(orig_node, if_incoming, - ethhdr->h_source); + neigh_node = batadv_neigh_node_get_or_create(orig_node, if_incoming, + ethhdr->h_source); if (!neigh_node) goto out; diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index 7f51bc2..b0bad57 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -602,19 +602,19 @@ batadv_hardif_neigh_get(const struct batadv_hard_iface *hard_iface, } /** - * batadv_neigh_node_new - create and init a new neigh_node object + * batadv_neigh_node_create - create a neigh node object * @orig_node: originator object representing the neighbour * @hard_iface: the interface where the neighbour is connected to * @neigh_addr: the mac address of the neighbour interface * * Allocates a new neigh_node object and initialises all the generic fields. * - * Return: neighbor when found. Othwerwise NULL + * Return: the neighbour node if found or created or NULL otherwise. */ -struct batadv_neigh_node * -batadv_neigh_node_new(struct batadv_orig_node *orig_node, - struct batadv_hard_iface *hard_iface, - const u8 *neigh_addr) +static struct batadv_neigh_node * +batadv_neigh_node_create(struct batadv_orig_node *orig_node, + struct batadv_hard_iface *hard_iface, + const u8 *neigh_addr) { struct batadv_neigh_node *neigh_node; struct batadv_hardif_neigh_node *hardif_neigh = NULL; @@ -667,6 +667,29 @@ out: } /** + * batadv_neigh_node_get_or_create - retrieve or create a neigh node object + * @orig_node: originator object representing the neighbour + * @hard_iface: the interface where the neighbour is connected to + * @neigh_addr: the mac address of the neighbour interface + * + * Return: the neighbour node if found or created or NULL otherwise. + */ +struct batadv_neigh_node * +batadv_neigh_node_get_or_create(struct batadv_orig_node *orig_node, + struct batadv_hard_iface *hard_iface, + const u8 *neigh_addr) +{ + struct batadv_neigh_node *neigh_node = NULL; + + /* first check without locking to avoid the overhead */ + neigh_node = batadv_neigh_node_get(orig_node, hard_iface, neigh_addr); + if (neigh_node) + return neigh_node; + + return batadv_neigh_node_create(orig_node, hard_iface, neigh_addr); +} + +/** * batadv_hardif_neigh_seq_print_text - print the single hop neighbour list * @seq: neighbour table seq_file struct * @offset: not used diff --git a/net/batman-adv/originator.h b/net/batman-adv/originator.h index 64a8951..566306b 100644 --- a/net/batman-adv/originator.h +++ b/net/batman-adv/originator.h @@ -46,9 +46,9 @@ batadv_hardif_neigh_get(const struct batadv_hard_iface *hard_iface, void batadv_hardif_neigh_put(struct batadv_hardif_neigh_node *hardif_neigh); struct batadv_neigh_node * -batadv_neigh_node_new(struct batadv_orig_node *orig_node, - struct batadv_hard_iface *hard_iface, - const u8 *neigh_addr); +batadv_neigh_node_get_or_create(struct batadv_orig_node *orig_node, + struct batadv_hard_iface *hard_iface, + const u8 *neigh_addr); void batadv_neigh_node_put(struct batadv_neigh_node *neigh_node); struct batadv_neigh_node * batadv_orig_router_get(struct batadv_orig_node *orig_node, -- cgit v0.10.2 From d9f179877e50ae2681fe7b0b83e0d9f63b6165ad Mon Sep 17 00:00:00 2001 From: Marek Lindner Date: Mon, 2 May 2016 21:58:50 +0800 Subject: batman-adv: remove unused callback from batadv_algo_ops struct Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index b0bad57..076d258 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -251,10 +251,8 @@ static void batadv_neigh_node_release(struct kref *ref) struct hlist_node *node_tmp; struct batadv_neigh_node *neigh_node; struct batadv_neigh_ifinfo *neigh_ifinfo; - struct batadv_algo_ops *bao; neigh_node = container_of(ref, struct batadv_neigh_node, refcount); - bao = neigh_node->orig_node->bat_priv->bat_algo_ops; hlist_for_each_entry_safe(neigh_ifinfo, node_tmp, &neigh_node->ifinfo_list, list) { @@ -263,9 +261,6 @@ static void batadv_neigh_node_release(struct kref *ref) batadv_hardif_neigh_put(neigh_node->hardif_neigh); - if (bao->bat_neigh_free) - bao->bat_neigh_free(neigh_node); - batadv_hardif_put(neigh_node->if_incoming); kfree_rcu(neigh_node, rcu); diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 6a577f4..114d150 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -1278,8 +1278,6 @@ struct batadv_forw_packet { * better than neigh2 for their respective outgoing interface from the metric * prospective * @bat_neigh_print: print the single hop neighbor list (optional) - * @bat_neigh_free: free the resources allocated by the routing algorithm for a - * neigh_node object * @bat_orig_print: print the originator table (optional) * @bat_orig_free: free the resources allocated by the routing algorithm for an * orig_node object @@ -1310,7 +1308,6 @@ struct batadv_algo_ops { struct batadv_neigh_node *neigh2, struct batadv_hard_iface *if_outgoing2); void (*bat_neigh_print)(struct batadv_priv *priv, struct seq_file *seq); - void (*bat_neigh_free)(struct batadv_neigh_node *neigh); /* orig_node handling API */ void (*bat_orig_print)(struct batadv_priv *priv, struct seq_file *seq, struct batadv_hard_iface *hard_iface); -- cgit v0.10.2 From f0d97253fb5fe87a7a91e7dc1ba4becf9d89d896 Mon Sep 17 00:00:00 2001 From: Antonio Quartulli Date: Tue, 3 May 2016 01:45:34 +0800 Subject: batman-adv: remove ogm_emit and ogm_schedule API calls The ogm_emit and ogm_schedule API calls were rather tight to the B.A.T.M.A.N. IV logic and therefore rather difficult to use with other algorithm implementations. Remove such calls and move the surrounding logic into the B.A.T.M.A.N. IV specific code. Signed-off-by: Antonio Quartulli Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index aa11296..4815db9 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include @@ -58,6 +59,8 @@ #include "send.h" #include "translation-table.h" +static void batadv_iv_send_outstanding_bat_ogm_packet(struct work_struct *work); + /** * enum batadv_dup_status - duplicate status * @BATADV_NO_DUP: the packet is no duplicate @@ -731,7 +734,7 @@ static void batadv_iv_ogm_aggregate_new(const unsigned char *packet_buff, /* start timer for this packet */ INIT_DELAYED_WORK(&forw_packet_aggr->delayed_work, - batadv_send_outstanding_bat_ogm_packet); + batadv_iv_send_outstanding_bat_ogm_packet); queue_delayed_work(batadv_event_workqueue, &forw_packet_aggr->delayed_work, send_time - jiffies); @@ -938,6 +941,19 @@ static void batadv_iv_ogm_schedule(struct batadv_hard_iface *hard_iface) u16 tvlv_len = 0; unsigned long send_time; + if ((hard_iface->if_status == BATADV_IF_NOT_IN_USE) || + (hard_iface->if_status == BATADV_IF_TO_BE_REMOVED)) + return; + + /* the interface gets activated here to avoid race conditions between + * the moment of activating the interface in + * hardif_activate_interface() where the originator mac is set and + * outdated packets (especially uninitialized mac addresses) in the + * packet queue + */ + if (hard_iface->if_status == BATADV_IF_TO_BE_ACTIVATED) + hard_iface->if_status = BATADV_IF_ACTIVE; + primary_if = batadv_primary_if_get_selected(bat_priv); if (hard_iface == primary_if) { @@ -1779,6 +1795,45 @@ static void batadv_iv_ogm_process(const struct sk_buff *skb, int ogm_offset, batadv_orig_node_put(orig_node); } +static void batadv_iv_send_outstanding_bat_ogm_packet(struct work_struct *work) +{ + struct delayed_work *delayed_work; + struct batadv_forw_packet *forw_packet; + struct batadv_priv *bat_priv; + + delayed_work = to_delayed_work(work); + forw_packet = container_of(delayed_work, struct batadv_forw_packet, + delayed_work); + bat_priv = netdev_priv(forw_packet->if_incoming->soft_iface); + spin_lock_bh(&bat_priv->forw_bat_list_lock); + hlist_del(&forw_packet->list); + spin_unlock_bh(&bat_priv->forw_bat_list_lock); + + if (atomic_read(&bat_priv->mesh_state) == BATADV_MESH_DEACTIVATING) + goto out; + + batadv_iv_ogm_emit(forw_packet); + + /* we have to have at least one packet in the queue to determine the + * queues wake up time unless we are shutting down. + * + * only re-schedule if this is the "original" copy, e.g. the OGM of the + * primary interface should only be rescheduled once per period, but + * this function will be called for the forw_packet instances of the + * other secondary interfaces as well. + */ + if (forw_packet->own && + forw_packet->if_incoming == forw_packet->if_outgoing) + batadv_iv_ogm_schedule(forw_packet->if_incoming); + +out: + /* don't count own packet */ + if (!forw_packet->own) + atomic_inc(&bat_priv->batman_queue_left); + + batadv_forw_packet_free(forw_packet); +} + static int batadv_iv_ogm_receive(struct sk_buff *skb, struct batadv_hard_iface *if_incoming) { @@ -1795,7 +1850,8 @@ static int batadv_iv_ogm_receive(struct sk_buff *skb, /* did we receive a B.A.T.M.A.N. IV OGM packet on an interface * that does not have B.A.T.M.A.N. IV enabled ? */ - if (bat_priv->bat_algo_ops->bat_ogm_emit != batadv_iv_ogm_emit) + if (bat_priv->bat_algo_ops->bat_iface_enable != + batadv_iv_ogm_iface_enable) return NET_RX_DROP; batadv_inc_counter(bat_priv, BATADV_CNT_MGMT_RX); @@ -2053,14 +2109,19 @@ out: return ret; } +static void batadv_iv_iface_activate(struct batadv_hard_iface *hard_iface) +{ + /* begin scheduling originator messages on that interface */ + batadv_iv_ogm_schedule(hard_iface); +} + static struct batadv_algo_ops batadv_batman_iv __read_mostly = { .name = "BATMAN_IV", + .bat_iface_activate = batadv_iv_iface_activate, .bat_iface_enable = batadv_iv_ogm_iface_enable, .bat_iface_disable = batadv_iv_ogm_iface_disable, .bat_iface_update_mac = batadv_iv_ogm_iface_update_mac, .bat_primary_iface_set = batadv_iv_ogm_primary_iface_set, - .bat_ogm_schedule = batadv_iv_ogm_schedule, - .bat_ogm_emit = batadv_iv_ogm_emit, .bat_neigh_cmp = batadv_iv_ogm_neigh_cmp, .bat_neigh_is_similar_or_better = batadv_iv_ogm_neigh_is_sob, .bat_neigh_print = batadv_iv_neigh_print, diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c index 0a12e5c..c16cd44 100644 --- a/net/batman-adv/bat_v.c +++ b/net/batman-adv/bat_v.c @@ -119,14 +119,6 @@ batadv_v_hardif_neigh_init(struct batadv_hardif_neigh_node *hardif_neigh) batadv_v_elp_throughput_metric_update); } -static void batadv_v_ogm_schedule(struct batadv_hard_iface *hard_iface) -{ -} - -static void batadv_v_ogm_emit(struct batadv_forw_packet *forw_packet) -{ -} - /** * batadv_v_orig_print_neigh - print neighbors for the originator table * @orig_node: the orig_node for which the neighbors are printed @@ -340,8 +332,6 @@ static struct batadv_algo_ops batadv_batman_v __read_mostly = { .bat_iface_update_mac = batadv_v_iface_update_mac, .bat_primary_iface_set = batadv_v_primary_iface_set, .bat_hardif_neigh_init = batadv_v_hardif_neigh_init, - .bat_ogm_emit = batadv_v_ogm_emit, - .bat_ogm_schedule = batadv_v_ogm_schedule, .bat_orig_print = batadv_v_orig_print, .bat_neigh_cmp = batadv_v_neigh_cmp, .bat_neigh_is_similar_or_better = batadv_v_neigh_is_sob, diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index 8c2f399..db2009d 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -553,9 +553,6 @@ int batadv_hardif_enable_interface(struct batadv_hard_iface *hard_iface, batadv_hardif_recalc_extra_skbroom(soft_iface); - /* begin scheduling originator messages on that interface */ - batadv_schedule_bat_ogm(hard_iface); - out: return 0; diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index 5f2974b..627d14e 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -569,8 +569,6 @@ int batadv_algo_register(struct batadv_algo_ops *bat_algo_ops) !bat_algo_ops->bat_iface_disable || !bat_algo_ops->bat_iface_update_mac || !bat_algo_ops->bat_primary_iface_set || - !bat_algo_ops->bat_ogm_schedule || - !bat_algo_ops->bat_ogm_emit || !bat_algo_ops->bat_neigh_cmp || !bat_algo_ops->bat_neigh_is_similar_or_better) { pr_info("Routing algo '%s' does not implement required ops\n", diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c index b1a4e8a..59e695b 100644 --- a/net/batman-adv/send.c +++ b/net/batman-adv/send.c @@ -428,27 +428,7 @@ int batadv_send_skb_via_gw(struct batadv_priv *bat_priv, struct sk_buff *skb, orig_node, vid); } -void batadv_schedule_bat_ogm(struct batadv_hard_iface *hard_iface) -{ - struct batadv_priv *bat_priv = netdev_priv(hard_iface->soft_iface); - - if ((hard_iface->if_status == BATADV_IF_NOT_IN_USE) || - (hard_iface->if_status == BATADV_IF_TO_BE_REMOVED)) - return; - - /* the interface gets activated here to avoid race conditions between - * the moment of activating the interface in - * hardif_activate_interface() where the originator mac is set and - * outdated packets (especially uninitialized mac addresses) in the - * packet queue - */ - if (hard_iface->if_status == BATADV_IF_TO_BE_ACTIVATED) - hard_iface->if_status = BATADV_IF_ACTIVE; - - bat_priv->bat_algo_ops->bat_ogm_schedule(hard_iface); -} - -static void batadv_forw_packet_free(struct batadv_forw_packet *forw_packet) +void batadv_forw_packet_free(struct batadv_forw_packet *forw_packet) { kfree_skb(forw_packet->skb); if (forw_packet->if_incoming) @@ -604,45 +584,6 @@ out: atomic_inc(&bat_priv->bcast_queue_left); } -void batadv_send_outstanding_bat_ogm_packet(struct work_struct *work) -{ - struct delayed_work *delayed_work; - struct batadv_forw_packet *forw_packet; - struct batadv_priv *bat_priv; - - delayed_work = to_delayed_work(work); - forw_packet = container_of(delayed_work, struct batadv_forw_packet, - delayed_work); - bat_priv = netdev_priv(forw_packet->if_incoming->soft_iface); - spin_lock_bh(&bat_priv->forw_bat_list_lock); - hlist_del(&forw_packet->list); - spin_unlock_bh(&bat_priv->forw_bat_list_lock); - - if (atomic_read(&bat_priv->mesh_state) == BATADV_MESH_DEACTIVATING) - goto out; - - bat_priv->bat_algo_ops->bat_ogm_emit(forw_packet); - - /* we have to have at least one packet in the queue to determine the - * queues wake up time unless we are shutting down. - * - * only re-schedule if this is the "original" copy, e.g. the OGM of the - * primary interface should only be rescheduled once per period, but - * this function will be called for the forw_packet instances of the - * other secondary interfaces as well. - */ - if (forw_packet->own && - forw_packet->if_incoming == forw_packet->if_outgoing) - batadv_schedule_bat_ogm(forw_packet->if_incoming); - -out: - /* don't count own packet */ - if (!forw_packet->own) - atomic_inc(&bat_priv->batman_queue_left); - - batadv_forw_packet_free(forw_packet); -} - void batadv_purge_outstanding_packets(struct batadv_priv *bat_priv, const struct batadv_hard_iface *hard_iface) diff --git a/net/batman-adv/send.h b/net/batman-adv/send.h index 6fd7270..7cecb75 100644 --- a/net/batman-adv/send.h +++ b/net/batman-adv/send.h @@ -26,8 +26,8 @@ #include "packet.h" struct sk_buff; -struct work_struct; +void batadv_forw_packet_free(struct batadv_forw_packet *forw_packet); int batadv_send_skb_to_orig(struct sk_buff *skb, struct batadv_orig_node *orig_node, struct batadv_hard_iface *recv_if); @@ -38,11 +38,9 @@ int batadv_send_broadcast_skb(struct sk_buff *skb, struct batadv_hard_iface *hard_iface); int batadv_send_unicast_skb(struct sk_buff *skb, struct batadv_neigh_node *neigh_node); -void batadv_schedule_bat_ogm(struct batadv_hard_iface *hard_iface); int batadv_add_bcast_packet_to_list(struct batadv_priv *bat_priv, const struct sk_buff *skb, unsigned long delay); -void batadv_send_outstanding_bat_ogm_packet(struct work_struct *work); void batadv_purge_outstanding_packets(struct batadv_priv *bat_priv, const struct batadv_hard_iface *hard_iface); diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 114d150..b70b6ae 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -1269,8 +1269,6 @@ struct batadv_forw_packet { * @bat_iface_update_mac: (re-)init mac addresses of the protocol information * belonging to this hard-interface * @bat_primary_iface_set: called when primary interface is selected / changed - * @bat_ogm_schedule: prepare a new outgoing OGM for the send queue - * @bat_ogm_emit: send scheduled OGM * @bat_hardif_neigh_init: called on creation of single hop entry * @bat_neigh_cmp: compare the metrics of two neighbors for their respective * outgoing interfaces @@ -1294,8 +1292,6 @@ struct batadv_algo_ops { void (*bat_iface_disable)(struct batadv_hard_iface *hard_iface); void (*bat_iface_update_mac)(struct batadv_hard_iface *hard_iface); void (*bat_primary_iface_set)(struct batadv_hard_iface *hard_iface); - void (*bat_ogm_schedule)(struct batadv_hard_iface *hard_iface); - void (*bat_ogm_emit)(struct batadv_forw_packet *forw_packet); /* neigh_node handling API */ void (*bat_hardif_neigh_init)(struct batadv_hardif_neigh_node *neigh); int (*bat_neigh_cmp)(struct batadv_neigh_node *neigh1, -- cgit v0.10.2 From c149ca72e58ac511ad8bf71df833efa9764115a4 Mon Sep 17 00:00:00 2001 From: Antonio Quartulli Date: Fri, 6 May 2016 02:46:37 +0800 Subject: batman-adv: remove useless inline attribute for sysfs helper function the compiler can optimize functions within the same C file and therefore there is no need to make it explicit. Remove the useless inline attribute for __batadv_store_uint_attr() Signed-off-by: Antonio Quartulli Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c index 414b207..ef5832f 100644 --- a/net/batman-adv/sysfs.c +++ b/net/batman-adv/sysfs.c @@ -389,12 +389,12 @@ static int batadv_store_uint_attr(const char *buff, size_t count, return count; } -static inline ssize_t -__batadv_store_uint_attr(const char *buff, size_t count, - int min, int max, - void (*post_func)(struct net_device *), - const struct attribute *attr, - atomic_t *attr_store, struct net_device *net_dev) +static ssize_t __batadv_store_uint_attr(const char *buff, size_t count, + int min, int max, + void (*post_func)(struct net_device *), + const struct attribute *attr, + atomic_t *attr_store, + struct net_device *net_dev) { int ret; -- cgit v0.10.2 From 3a24a63e74af1bffc7aeb5d83adcd63b37e38425 Mon Sep 17 00:00:00 2001 From: Antonio Quartulli Date: Fri, 6 May 2016 02:46:38 +0800 Subject: batman-adv: move GW mode and selection class to private data structure To reduce the field pollution in our main batadv_priv data structure we've already created some substructures so that we could group fields in a convenient manner. However gw_mode and gw_sel_class are still part of the main object. More both fields to the GW private substructure. Signed-off-by: Antonio Quartulli Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c index b9d72e9..18c3715 100644 --- a/net/batman-adv/gateway_client.c +++ b/net/batman-adv/gateway_client.c @@ -192,7 +192,7 @@ batadv_gw_get_best_gw_node(struct batadv_priv *bat_priv) tq_avg = router_ifinfo->bat_iv.tq_avg; - switch (atomic_read(&bat_priv->gw_sel_class)) { + switch (atomic_read(&bat_priv->gw.sel_class)) { case 1: /* fast connection */ tmp_gw_factor = tq_avg * tq_avg; tmp_gw_factor *= gw_node->bandwidth_down; @@ -255,7 +255,7 @@ void batadv_gw_check_client_stop(struct batadv_priv *bat_priv) { struct batadv_gw_node *curr_gw; - if (atomic_read(&bat_priv->gw_mode) != BATADV_GW_MODE_CLIENT) + if (atomic_read(&bat_priv->gw.mode) != BATADV_GW_MODE_CLIENT) return; curr_gw = batadv_gw_get_selected_gw_node(bat_priv); @@ -283,7 +283,7 @@ void batadv_gw_election(struct batadv_priv *bat_priv) struct batadv_neigh_ifinfo *router_ifinfo = NULL; char gw_addr[18] = { '\0' }; - if (atomic_read(&bat_priv->gw_mode) != BATADV_GW_MODE_CLIENT) + if (atomic_read(&bat_priv->gw.mode) != BATADV_GW_MODE_CLIENT) goto out; curr_gw = batadv_gw_get_selected_gw_node(bat_priv); @@ -402,8 +402,8 @@ void batadv_gw_check_election(struct batadv_priv *bat_priv, /* if the routing class is greater than 3 the value tells us how much * greater the TQ value of the new gateway must be */ - if ((atomic_read(&bat_priv->gw_sel_class) > 3) && - (orig_tq_avg - gw_tq_avg < atomic_read(&bat_priv->gw_sel_class))) + if ((atomic_read(&bat_priv->gw.sel_class) > 3) && + (orig_tq_avg - gw_tq_avg < atomic_read(&bat_priv->gw.sel_class))) goto out; batadv_dbg(BATADV_DBG_BATMAN, bat_priv, @@ -820,7 +820,7 @@ bool batadv_gw_out_of_range(struct batadv_priv *bat_priv, if (!gw_node) goto out; - switch (atomic_read(&bat_priv->gw_mode)) { + switch (atomic_read(&bat_priv->gw.mode)) { case BATADV_GW_MODE_SERVER: /* If we are a GW then we are our best GW. We can artificially * set the tq towards ourself as the maximum value diff --git a/net/batman-adv/gateway_common.c b/net/batman-adv/gateway_common.c index 4423047..3c26945 100644 --- a/net/batman-adv/gateway_common.c +++ b/net/batman-adv/gateway_common.c @@ -144,7 +144,7 @@ void batadv_gw_tvlv_container_update(struct batadv_priv *bat_priv) u32 down, up; char gw_mode; - gw_mode = atomic_read(&bat_priv->gw_mode); + gw_mode = atomic_read(&bat_priv->gw.mode); switch (gw_mode) { case BATADV_GW_MODE_OFF: @@ -241,8 +241,8 @@ static void batadv_gw_tvlv_ogm_handler_v1(struct batadv_priv *bat_priv, /* restart gateway selection if fast or late switching was enabled */ if ((gateway.bandwidth_down != 0) && - (atomic_read(&bat_priv->gw_mode) == BATADV_GW_MODE_CLIENT) && - (atomic_read(&bat_priv->gw_sel_class) > 2)) + (atomic_read(&bat_priv->gw.mode) == BATADV_GW_MODE_CLIENT) && + (atomic_read(&bat_priv->gw.sel_class) > 2)) batadv_gw_check_election(bat_priv, orig); } diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index 343d2c9..81665b1 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -255,7 +255,7 @@ static int batadv_interface_tx(struct sk_buff *skb, if (batadv_compare_eth(ethhdr->h_dest, ectp_addr)) goto dropped; - gw_mode = atomic_read(&bat_priv->gw_mode); + gw_mode = atomic_read(&bat_priv->gw.mode); if (is_multicast_ether_addr(ethhdr->h_dest)) { /* if gw mode is off, broadcast every packet */ if (gw_mode == BATADV_GW_MODE_OFF) { @@ -815,8 +815,8 @@ static int batadv_softif_init_late(struct net_device *dev) atomic_set(&bat_priv->mcast.num_want_all_ipv4, 0); atomic_set(&bat_priv->mcast.num_want_all_ipv6, 0); #endif - atomic_set(&bat_priv->gw_mode, BATADV_GW_MODE_OFF); - atomic_set(&bat_priv->gw_sel_class, 20); + atomic_set(&bat_priv->gw.mode, BATADV_GW_MODE_OFF); + atomic_set(&bat_priv->gw.sel_class, 20); atomic_set(&bat_priv->gw.bandwidth_down, 100); atomic_set(&bat_priv->gw.bandwidth_up, 20); atomic_set(&bat_priv->orig_interval, 1000); diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c index ef5832f..233abcf 100644 --- a/net/batman-adv/sysfs.c +++ b/net/batman-adv/sysfs.c @@ -427,7 +427,7 @@ static ssize_t batadv_show_gw_mode(struct kobject *kobj, struct attribute *attr, struct batadv_priv *bat_priv = batadv_kobj_to_batpriv(kobj); int bytes_written; - switch (atomic_read(&bat_priv->gw_mode)) { + switch (atomic_read(&bat_priv->gw.mode)) { case BATADV_GW_MODE_CLIENT: bytes_written = sprintf(buff, "%s\n", BATADV_GW_MODE_CLIENT_NAME); @@ -476,10 +476,10 @@ static ssize_t batadv_store_gw_mode(struct kobject *kobj, return -EINVAL; } - if (atomic_read(&bat_priv->gw_mode) == gw_mode_tmp) + if (atomic_read(&bat_priv->gw.mode) == gw_mode_tmp) return count; - switch (atomic_read(&bat_priv->gw_mode)) { + switch (atomic_read(&bat_priv->gw.mode)) { case BATADV_GW_MODE_CLIENT: curr_gw_mode_str = BATADV_GW_MODE_CLIENT_NAME; break; @@ -508,7 +508,7 @@ static ssize_t batadv_store_gw_mode(struct kobject *kobj, * state */ batadv_gw_check_client_stop(bat_priv); - atomic_set(&bat_priv->gw_mode, (unsigned int)gw_mode_tmp); + atomic_set(&bat_priv->gw.mode, (unsigned int)gw_mode_tmp); batadv_gw_tvlv_container_update(bat_priv); return count; } @@ -624,7 +624,7 @@ BATADV_ATTR_SIF_UINT(orig_interval, orig_interval, S_IRUGO | S_IWUSR, 2 * BATADV_JITTER, INT_MAX, NULL); BATADV_ATTR_SIF_UINT(hop_penalty, hop_penalty, S_IRUGO | S_IWUSR, 0, BATADV_TQ_MAX_VALUE, NULL); -BATADV_ATTR_SIF_UINT(gw_sel_class, gw_sel_class, S_IRUGO | S_IWUSR, 1, +BATADV_ATTR_SIF_UINT(gw_sel_class, gw.sel_class, S_IRUGO | S_IWUSR, 1, BATADV_TQ_MAX_VALUE, batadv_post_gw_reselect); static BATADV_ATTR(gw_bandwidth, S_IRUGO | S_IWUSR, batadv_show_gw_bwidth, batadv_store_gw_bwidth); diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index b70b6ae..32c6d0e 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -707,6 +707,8 @@ struct batadv_priv_debug_log { * @list: list of available gateway nodes * @list_lock: lock protecting gw_list & curr_gw * @curr_gw: pointer to currently selected gateway node + * @mode: gateway operation: off, client or server (see batadv_gw_modes) + * @sel_class: gateway selection class (applies if gw_mode client) * @bandwidth_down: advertised uplink download bandwidth (if gw_mode server) * @bandwidth_up: advertised uplink upload bandwidth (if gw_mode server) * @reselect: bool indicating a gateway re-selection is in progress @@ -715,6 +717,8 @@ struct batadv_priv_gw { struct hlist_head list; spinlock_t list_lock; /* protects gw_list & curr_gw */ struct batadv_gw_node __rcu *curr_gw; /* rcu protected pointer */ + atomic_t mode; + atomic_t sel_class; atomic_t bandwidth_down; atomic_t bandwidth_up; atomic_t reselect; @@ -865,8 +869,6 @@ struct batadv_priv_bat_v { * enabled * @multicast_mode: Enable or disable multicast optimizations on this node's * sender/originating side - * @gw_mode: gateway operation: off, client or server (see batadv_gw_modes) - * @gw_sel_class: gateway selection class (applies if gw_mode client) * @orig_interval: OGM broadcast interval in milliseconds * @hop_penalty: penalty which will be applied to an OGM's tq-field on every hop * @log_level: configured log level (see batadv_dbg_level) @@ -922,8 +924,6 @@ struct batadv_priv { #ifdef CONFIG_BATMAN_ADV_MCAST atomic_t multicast_mode; #endif - atomic_t gw_mode; - atomic_t gw_sel_class; atomic_t orig_interval; atomic_t hop_penalty; #ifdef CONFIG_BATMAN_ADV_DEBUG -- cgit v0.10.2 From 7db682d1c39b2198a9c9d0bee5812d9c4329123d Mon Sep 17 00:00:00 2001 From: Marek Lindner Date: Tue, 10 May 2016 22:31:59 +0800 Subject: batman-adv: init ELP tweaking options only once The ELP interval and throughput override interface settings are initialized with default settings on every time an interface is added to a mesh. This patch prevents this behavior by moving the configuration init to the interface detection routine which runs only once per interface. Signed-off-by: Marek Lindner [a@unstable.cc: move initialization to batadv_v_hardif_init] Signed-off-by: Antonio Quartulli Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/bat_algo.h b/net/batman-adv/bat_algo.h index 03dafd3..3654296 100644 --- a/net/batman-adv/bat_algo.h +++ b/net/batman-adv/bat_algo.h @@ -18,13 +18,14 @@ #ifndef _NET_BATMAN_ADV_BAT_ALGO_H_ #define _NET_BATMAN_ADV_BAT_ALGO_H_ -struct batadv_priv; +#include "main.h" int batadv_iv_init(void); #ifdef CONFIG_BATMAN_ADV_BATMAN_V int batadv_v_init(void); +void batadv_v_hardif_init(struct batadv_hard_iface *hardif); int batadv_v_mesh_init(struct batadv_priv *bat_priv); void batadv_v_mesh_free(struct batadv_priv *bat_priv); @@ -35,6 +36,10 @@ static inline int batadv_v_init(void) return 0; } +static inline void batadv_v_hardif_init(struct batadv_hard_iface *hardif) +{ +} + static inline int batadv_v_mesh_init(struct batadv_priv *bat_priv) { return 0; diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c index c16cd44..c2fea81 100644 --- a/net/batman-adv/bat_v.c +++ b/net/batman-adv/bat_v.c @@ -70,11 +70,6 @@ static int batadv_v_iface_enable(struct batadv_hard_iface *hard_iface) if (ret < 0) batadv_v_elp_iface_disable(hard_iface); - /* enable link throughput auto-detection by setting the throughput - * override to zero - */ - atomic_set(&hard_iface->bat_v.throughput_override, 0); - return ret; } @@ -339,6 +334,20 @@ static struct batadv_algo_ops batadv_batman_v __read_mostly = { }; /** + * batadv_v_hardif_init - initialize the algorithm specific fields in the + * hard-interface object + * @hard_iface: the hard-interface to initialize + */ +void batadv_v_hardif_init(struct batadv_hard_iface *hard_iface) +{ + /* enable link throughput auto-detection by setting the throughput + * override to zero + */ + atomic_set(&hard_iface->bat_v.throughput_override, 0); + atomic_set(&hard_iface->bat_v.elp_interval, 500); +} + +/** * batadv_v_mesh_init - initialize the B.A.T.M.A.N. V private resources for a * mesh * @bat_priv: the object representing the mesh interface to initialise diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c index 8909d1e..cf0262b 100644 --- a/net/batman-adv/bat_v_elp.c +++ b/net/batman-adv/bat_v_elp.c @@ -344,7 +344,6 @@ int batadv_v_elp_iface_enable(struct batadv_hard_iface *hard_iface) /* randomize initial seqno to avoid collision */ get_random_bytes(&random_seqno, sizeof(random_seqno)); atomic_set(&hard_iface->bat_v.elp_seqno, random_seqno); - atomic_set(&hard_iface->bat_v.elp_interval, 500); /* assume full-duplex by default */ hard_iface->bat_v.flags |= BATADV_FULL_DUPLEX; diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index db2009d..3696929 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -37,6 +37,7 @@ #include #include +#include "bat_algo.h" #include "bridge_loop_avoidance.h" #include "debugfs.h" #include "distributed-arp-table.h" @@ -683,6 +684,8 @@ batadv_hardif_add_interface(struct net_device *net_dev) if (batadv_is_wifi_netdev(net_dev)) hard_iface->num_bcasts = BATADV_NUM_BCASTS_WIRELESS; + batadv_v_hardif_init(hard_iface); + /* extra reference for return */ kref_init(&hard_iface->refcount); kref_get(&hard_iface->refcount); -- cgit v0.10.2 From 1914848e0d641f8d6173bb53b6a4a392468ac725 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Mon, 9 May 2016 20:03:35 +0200 Subject: batman-adv: Set skb priority in fragments BATMAN will set the skb->priority based on the IP precedence or 802.1q tag. However, if it needs to fragment the frame, it currently leaves the fragment skb with the default priority and actually overwrites the priority in the unfragmented frame. Fix this. Signed-off-by: Andrew Lunn Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c index 65536db..a119b6a 100644 --- a/net/batman-adv/fragmentation.c +++ b/net/batman-adv/fragmentation.c @@ -27,7 +27,6 @@ #include #include #include -#include #include #include #include @@ -414,7 +413,7 @@ static struct sk_buff *batadv_frag_create(struct sk_buff *skb, if (!skb_fragment) goto err; - skb->priority = TC_PRIO_CONTROL; + skb_fragment->priority = skb->priority; /* Eat the last mtu-bytes of the skb */ skb_reserve(skb_fragment, header_size + ETH_HLEN); -- cgit v0.10.2 From 67a5613ed09e1b9d5d188507bcbaeb37a6d0fe12 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sun, 15 May 2016 11:07:41 +0200 Subject: batman-adv: Include main.h in bat_v_ogm.h main.h includes statements which (re)define preprocessor variables which influence the compiled code. This makes it necessary to include it in all files. For example, it redefines pr_fmt used to the module as prefix for each pr_* message. Reported-by: Antonio Quartulli Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/bat_v_ogm.h b/net/batman-adv/bat_v_ogm.h index d849c75..4c4d45c 100644 --- a/net/batman-adv/bat_v_ogm.h +++ b/net/batman-adv/bat_v_ogm.h @@ -18,10 +18,10 @@ #ifndef _BATMAN_ADV_BATADV_V_OGM_H_ #define _BATMAN_ADV_BATADV_V_OGM_H_ +#include "main.h" + #include -struct batadv_hard_iface; -struct batadv_priv; struct sk_buff; int batadv_v_ogm_init(struct batadv_priv *bat_priv); -- cgit v0.10.2 From c0f25c802b3300b28d1e67c58c702d29555838de Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Mon, 9 May 2016 20:03:36 +0200 Subject: batman-adv: Include frame priority in fragment header Unfragmented frames which traverse a node have their skb->priority set by looking at the IP ToS byte, or the 802.1p header. However for fragments this is not possible, only one of the fragments will contain the headers. Instead, place the priority into the fragment header and on receiving a fragment, use this information to set the skb->priority for when the fragment is forwarded. Signed-off-by: Andrew Lunn Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c index a119b6a..9f41a0a 100644 --- a/net/batman-adv/fragmentation.c +++ b/net/batman-adv/fragmentation.c @@ -472,6 +472,15 @@ bool batadv_frag_send_packet(struct sk_buff *skb, frag_header.reserved = 0; frag_header.no = 0; frag_header.total_size = htons(skb->len); + + /* skb->priority values from 256->263 are magic values to + * directly indicate a specific 802.1d priority. This is used + * to allow 802.1d priority to be passed directly in from VLAN + * tags, etc. + */ + if (skb->priority >= 256 && skb->priority <= 263) + frag_header.priority = skb->priority - 256; + ether_addr_copy(frag_header.orig, primary_if->net_dev->dev_addr); ether_addr_copy(frag_header.dest, orig_node->orig); diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h index 372128d..7156779 100644 --- a/net/batman-adv/packet.h +++ b/net/batman-adv/packet.h @@ -420,6 +420,7 @@ struct batadv_unicast_4addr_packet { * @dest: final destination used when routing fragments * @orig: originator of the fragment used when merging the packet * @no: fragment number within this sequence + * @priority: priority of frame, from ToS IP precedence or 802.1p * @reserved: reserved byte for alignment * @seqno: sequence identification * @total_size: size of the merged packet @@ -430,9 +431,11 @@ struct batadv_frag_packet { u8 ttl; #if defined(__BIG_ENDIAN_BITFIELD) u8 no:4; - u8 reserved:4; + u8 priority:3; + u8 reserved:1; #elif defined(__LITTLE_ENDIAN_BITFIELD) - u8 reserved:4; + u8 reserved:1; + u8 priority:3; u8 no:4; #else #error "unknown bitfield endianness" diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index f75091c..24fc753 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -1006,6 +1006,8 @@ int batadv_recv_frag_packet(struct sk_buff *skb, if (!orig_node_src) goto out; + skb->priority = frag_packet->priority + 256; + /* Route the fragment if it is not for us and too big to be merged. */ if (!batadv_is_my_mac(bat_priv, frag_packet->dest) && batadv_frag_skb_fwd(skb, recv_if, orig_node_src)) { -- cgit v0.10.2 From fcafa5e74b42a182a5bcc5c7f94ca026d4e5f06e Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sun, 15 May 2016 11:07:42 +0200 Subject: batman-adv: Keep includes ordered by filename It is easier to detect if a include is already there for a used functionality when the includes are ordered. Using an alphabetic order together with the grouping in commit 1e2c2a4fe4a5 ("batman-adv: Add required includes to all files") makes includes better manageable. Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index 4815db9..6dc89b0 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -31,8 +31,8 @@ #include #include #include -#include #include +#include #include #include #include diff --git a/net/batman-adv/bat_v_elp.h b/net/batman-adv/bat_v_elp.h index cc130b2..be17c0b 100644 --- a/net/batman-adv/bat_v_elp.h +++ b/net/batman-adv/bat_v_elp.h @@ -15,11 +15,11 @@ * along with this program; if not, see . */ -#include "main.h" - #ifndef _NET_BATMAN_ADV_BAT_V_ELP_H_ #define _NET_BATMAN_ADV_BAT_V_ELP_H_ +#include "main.h" + struct sk_buff; struct work_struct; diff --git a/net/batman-adv/gateway_common.c b/net/batman-adv/gateway_common.c index 3c26945..7754435 100644 --- a/net/batman-adv/gateway_common.c +++ b/net/batman-adv/gateway_common.c @@ -19,8 +19,8 @@ #include "main.h" #include -#include #include +#include #include #include #include diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index 3696929..a3483f6 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -23,9 +23,9 @@ #include #include #include +#include #include #include -#include #include #include #include diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index cd83e28..38f9e55 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -181,12 +181,12 @@ enum batadv_uev_type { #include #include #include /* for packet.h */ +#include +#include #include +#include #include #include -#include -#include -#include #include "types.h" diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c index 59e695b..4e49454 100644 --- a/net/batman-adv/send.c +++ b/net/batman-adv/send.c @@ -22,8 +22,8 @@ #include #include #include -#include #include +#include #include #include #include diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c index 233abcf..6244a9a 100644 --- a/net/batman-adv/sysfs.c +++ b/net/batman-adv/sysfs.c @@ -25,8 +25,8 @@ #include #include #include -#include #include +#include #include #include #include @@ -38,10 +38,10 @@ #include #include +#include "bridge_loop_avoidance.h" #include "distributed-arp-table.h" #include "gateway_client.h" #include "gateway_common.h" -#include "bridge_loop_avoidance.h" #include "hard-interface.h" #include "network-coding.h" #include "packet.h" -- cgit v0.10.2 From bd2a979e53fd9dd64b7e27553a23001d53201005 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Linus=20L=C3=BCssing?= Date: Tue, 10 May 2016 18:41:24 +0200 Subject: batman-adv: Always flood IGMP/MLD reports MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With this patch IGMP or MLD reports are always flooded. This is necessary for the upcoming bridge integration to function without multicast packet loss. With the report handling so far bridges might miss interested multicast listeners, leading to wrongly excluding ports from multicast packet forwarding. Currently we are treating IGMP/MLD reports, the messages bridges use to learn about interested multicast listeners, just as any other multicast packet: We try to send them to nodes matching its multicast destination. Unfortunately, the destination address of reports of the older IGMPv2/MLDv1 protocol families do not strictly adhere to their own protocol: More precisely, the interested receiver, an IGMPv2 or MLDv1 querier, itself usually does not listen to the multicast destination address of any reports. Therefore with this patch we are simply excluding IGMP/MLD reports from the multicast forwarding code path and keep flooding them. By that any bridge receives them and can properly learn about listeners. To avoid compatibility issues with older nodes not yet implementing this report handling, we need to force them to flood reports: We do this by bumping the multicast TVLV version to 2, effectively disabling their multicast optimization. Tested-by: Simon Wunderlich Signed-off-by: Linus Lüssing Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig index f66930e..b7ba97d 100644 --- a/net/batman-adv/Kconfig +++ b/net/batman-adv/Kconfig @@ -66,7 +66,7 @@ config BATMAN_ADV_NC config BATMAN_ADV_MCAST bool "Multicast optimisation" - depends on BATMAN_ADV + depends on BATMAN_ADV && INET default n help This option enables the multicast optimisation which aims to diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index c32f24f..4673328 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -25,9 +25,11 @@ #include #include #include +#include #include -#include +#include #include +#include #include #include #include @@ -236,7 +238,7 @@ static bool batadv_mcast_mla_tvlv_update(struct batadv_priv *bat_priv) if (batadv_mcast_has_bridge(bat_priv)) { if (bat_priv->mcast.enabled) { batadv_tvlv_container_unregister(bat_priv, - BATADV_TVLV_MCAST, 1); + BATADV_TVLV_MCAST, 2); bat_priv->mcast.enabled = false; } @@ -245,7 +247,7 @@ static bool batadv_mcast_mla_tvlv_update(struct batadv_priv *bat_priv) if (!bat_priv->mcast.enabled || mcast_data.flags != bat_priv->mcast.flags) { - batadv_tvlv_container_register(bat_priv, BATADV_TVLV_MCAST, 1, + batadv_tvlv_container_register(bat_priv, BATADV_TVLV_MCAST, 2, &mcast_data, sizeof(mcast_data)); bat_priv->mcast.flags = mcast_data.flags; bat_priv->mcast.enabled = true; @@ -283,6 +285,31 @@ out: } /** + * batadv_mcast_is_report_ipv4 - check for IGMP reports + * @skb: the ethernet frame destined for the mesh + * + * This call might reallocate skb data. + * + * Checks whether the given frame is a valid IGMP report. + * + * Return: If so then true, otherwise false. + */ +static bool batadv_mcast_is_report_ipv4(struct sk_buff *skb) +{ + if (ip_mc_check_igmp(skb, NULL) < 0) + return false; + + switch (igmp_hdr(skb)->type) { + case IGMP_HOST_MEMBERSHIP_REPORT: + case IGMPV2_HOST_MEMBERSHIP_REPORT: + case IGMPV3_HOST_MEMBERSHIP_REPORT: + return true; + } + + return false; +} + +/** * batadv_mcast_forw_mode_check_ipv4 - check for optimized forwarding potential * @bat_priv: the bat priv with all the soft interface information * @skb: the IPv4 packet to check @@ -304,6 +331,9 @@ static int batadv_mcast_forw_mode_check_ipv4(struct batadv_priv *bat_priv, if (!pskb_may_pull(skb, sizeof(struct ethhdr) + sizeof(*iphdr))) return -ENOMEM; + if (batadv_mcast_is_report_ipv4(skb)) + return -EINVAL; + iphdr = ip_hdr(skb); /* TODO: Implement Multicast Router Discovery (RFC4286), @@ -320,6 +350,31 @@ static int batadv_mcast_forw_mode_check_ipv4(struct batadv_priv *bat_priv, return 0; } +#if IS_ENABLED(CONFIG_IPV6) +/** + * batadv_mcast_is_report_ipv6 - check for MLD reports + * @skb: the ethernet frame destined for the mesh + * + * This call might reallocate skb data. + * + * Checks whether the given frame is a valid MLD report. + * + * Return: If so then true, otherwise false. + */ +static bool batadv_mcast_is_report_ipv6(struct sk_buff *skb) +{ + if (ipv6_mc_check_mld(skb, NULL) < 0) + return false; + + switch (icmp6_hdr(skb)->icmp6_type) { + case ICMPV6_MGM_REPORT: + case ICMPV6_MLD2_REPORT: + return true; + } + + return false; +} + /** * batadv_mcast_forw_mode_check_ipv6 - check for optimized forwarding potential * @bat_priv: the bat priv with all the soft interface information @@ -341,6 +396,9 @@ static int batadv_mcast_forw_mode_check_ipv6(struct batadv_priv *bat_priv, if (!pskb_may_pull(skb, sizeof(struct ethhdr) + sizeof(*ip6hdr))) return -ENOMEM; + if (batadv_mcast_is_report_ipv6(skb)) + return -EINVAL; + ip6hdr = ipv6_hdr(skb); /* TODO: Implement Multicast Router Discovery (RFC4286), @@ -357,6 +415,7 @@ static int batadv_mcast_forw_mode_check_ipv6(struct batadv_priv *bat_priv, return 0; } +#endif /** * batadv_mcast_forw_mode_check - check for optimized forwarding potential @@ -385,9 +444,11 @@ static int batadv_mcast_forw_mode_check(struct batadv_priv *bat_priv, case ETH_P_IP: return batadv_mcast_forw_mode_check_ipv4(bat_priv, skb, is_unsnoopable); +#if IS_ENABLED(CONFIG_IPV6) case ETH_P_IPV6: return batadv_mcast_forw_mode_check_ipv6(bat_priv, skb, is_unsnoopable); +#endif default: return -EINVAL; } @@ -728,18 +789,18 @@ static void batadv_mcast_want_ipv6_update(struct batadv_priv *bat_priv, } /** - * batadv_mcast_tvlv_ogm_handler_v1 - process incoming multicast tvlv container + * batadv_mcast_tvlv_ogm_handler - process incoming multicast tvlv container * @bat_priv: the bat priv with all the soft interface information * @orig: the orig_node of the ogm * @flags: flags indicating the tvlv state (see batadv_tvlv_handler_flags) * @tvlv_value: tvlv buffer containing the multicast data * @tvlv_value_len: tvlv buffer length */ -static void batadv_mcast_tvlv_ogm_handler_v1(struct batadv_priv *bat_priv, - struct batadv_orig_node *orig, - u8 flags, - void *tvlv_value, - u16 tvlv_value_len) +static void batadv_mcast_tvlv_ogm_handler(struct batadv_priv *bat_priv, + struct batadv_orig_node *orig, + u8 flags, + void *tvlv_value, + u16 tvlv_value_len) { bool orig_mcast_enabled = !(flags & BATADV_TVLV_HANDLER_OGM_CIFNOTFND); u8 mcast_flags = BATADV_NO_FLAGS; @@ -789,8 +850,8 @@ static void batadv_mcast_tvlv_ogm_handler_v1(struct batadv_priv *bat_priv, */ void batadv_mcast_init(struct batadv_priv *bat_priv) { - batadv_tvlv_handler_register(bat_priv, batadv_mcast_tvlv_ogm_handler_v1, - NULL, BATADV_TVLV_MCAST, 1, + batadv_tvlv_handler_register(bat_priv, batadv_mcast_tvlv_ogm_handler, + NULL, BATADV_TVLV_MCAST, 2, BATADV_TVLV_HANDLER_OGM_CIFNOTFND); } @@ -800,8 +861,8 @@ void batadv_mcast_init(struct batadv_priv *bat_priv) */ void batadv_mcast_free(struct batadv_priv *bat_priv) { - batadv_tvlv_container_unregister(bat_priv, BATADV_TVLV_MCAST, 1); - batadv_tvlv_handler_unregister(bat_priv, BATADV_TVLV_MCAST, 1); + batadv_tvlv_container_unregister(bat_priv, BATADV_TVLV_MCAST, 2); + batadv_tvlv_handler_unregister(bat_priv, BATADV_TVLV_MCAST, 2); spin_lock_bh(&bat_priv->tt.commit_lock); batadv_mcast_mla_tt_retract(bat_priv, NULL); -- cgit v0.10.2 From 1f8dce4992d03fc15cfbaf67cd09f0d1648c4606 Mon Sep 17 00:00:00 2001 From: Markus Pargmann Date: Sun, 15 May 2016 11:07:43 +0200 Subject: batman-adv: split tvlv into a separate file The tvlv functionality in main.c is mostly unrelated to the rest of the content. It still takes up a large portion of this source file (~45%, 588 lines). Moving it to a separate file makes it better visible as a main component of the batman-adv implementation and hides it less in the other helper functions in main.c Signed-off-by: Markus Pargmann [sven@narfation.org: fix conflicts with current version, fix includes, rewrote commit message] Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile index 797cf2f..5c6ece0 100644 --- a/net/batman-adv/Makefile +++ b/net/batman-adv/Makefile @@ -40,3 +40,4 @@ batman-adv-y += send.o batman-adv-y += soft-interface.o batman-adv-y += sysfs.o batman-adv-y += translation-table.o +batman-adv-y += tvlv.o diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index 6dc89b0..948a5b4 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -58,6 +58,7 @@ #include "routing.h" #include "send.h" #include "translation-table.h" +#include "tvlv.h" static void batadv_iv_send_outstanding_bat_ogm_packet(struct work_struct *work); diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c index 23ea9bf..ca5a679 100644 --- a/net/batman-adv/bat_v_ogm.c +++ b/net/batman-adv/bat_v_ogm.c @@ -46,6 +46,7 @@ #include "routing.h" #include "send.h" #include "translation-table.h" +#include "tvlv.h" /** * batadv_v_ogm_orig_get - retrieve and possibly create an originator node diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c index 278800a..584b827 100644 --- a/net/batman-adv/distributed-arp-table.c +++ b/net/batman-adv/distributed-arp-table.c @@ -48,6 +48,7 @@ #include "originator.h" #include "send.h" #include "translation-table.h" +#include "tvlv.h" static void batadv_dat_purge(struct work_struct *work); diff --git a/net/batman-adv/gateway_common.c b/net/batman-adv/gateway_common.c index 7754435..6a6f2d4 100644 --- a/net/batman-adv/gateway_common.c +++ b/net/batman-adv/gateway_common.c @@ -29,6 +29,7 @@ #include "gateway_client.h" #include "packet.h" +#include "tvlv.h" /** * batadv_parse_throughput - parse supplied string buffer to extract throughput diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index 627d14e..225d63e 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -31,16 +31,13 @@ #include #include #include -#include #include #include #include -#include #include #include #include #include -#include #include #include #include @@ -642,594 +639,6 @@ __be32 batadv_skb_crc32(struct sk_buff *skb, u8 *payload_ptr) } /** - * batadv_tvlv_handler_release - release tvlv handler from lists and queue for - * free after rcu grace period - * @ref: kref pointer of the tvlv - */ -static void batadv_tvlv_handler_release(struct kref *ref) -{ - struct batadv_tvlv_handler *tvlv_handler; - - tvlv_handler = container_of(ref, struct batadv_tvlv_handler, refcount); - kfree_rcu(tvlv_handler, rcu); -} - -/** - * batadv_tvlv_handler_put - decrement the tvlv container refcounter and - * possibly release it - * @tvlv_handler: the tvlv handler to free - */ -static void batadv_tvlv_handler_put(struct batadv_tvlv_handler *tvlv_handler) -{ - kref_put(&tvlv_handler->refcount, batadv_tvlv_handler_release); -} - -/** - * batadv_tvlv_handler_get - retrieve tvlv handler from the tvlv handler list - * based on the provided type and version (both need to match) - * @bat_priv: the bat priv with all the soft interface information - * @type: tvlv handler type to look for - * @version: tvlv handler version to look for - * - * Return: tvlv handler if found or NULL otherwise. - */ -static struct batadv_tvlv_handler * -batadv_tvlv_handler_get(struct batadv_priv *bat_priv, u8 type, u8 version) -{ - struct batadv_tvlv_handler *tvlv_handler_tmp, *tvlv_handler = NULL; - - rcu_read_lock(); - hlist_for_each_entry_rcu(tvlv_handler_tmp, - &bat_priv->tvlv.handler_list, list) { - if (tvlv_handler_tmp->type != type) - continue; - - if (tvlv_handler_tmp->version != version) - continue; - - if (!kref_get_unless_zero(&tvlv_handler_tmp->refcount)) - continue; - - tvlv_handler = tvlv_handler_tmp; - break; - } - rcu_read_unlock(); - - return tvlv_handler; -} - -/** - * batadv_tvlv_container_release - release tvlv from lists and free - * @ref: kref pointer of the tvlv - */ -static void batadv_tvlv_container_release(struct kref *ref) -{ - struct batadv_tvlv_container *tvlv; - - tvlv = container_of(ref, struct batadv_tvlv_container, refcount); - kfree(tvlv); -} - -/** - * batadv_tvlv_container_put - decrement the tvlv container refcounter and - * possibly release it - * @tvlv: the tvlv container to free - */ -static void batadv_tvlv_container_put(struct batadv_tvlv_container *tvlv) -{ - kref_put(&tvlv->refcount, batadv_tvlv_container_release); -} - -/** - * batadv_tvlv_container_get - retrieve tvlv container from the tvlv container - * list based on the provided type and version (both need to match) - * @bat_priv: the bat priv with all the soft interface information - * @type: tvlv container type to look for - * @version: tvlv container version to look for - * - * Has to be called with the appropriate locks being acquired - * (tvlv.container_list_lock). - * - * Return: tvlv container if found or NULL otherwise. - */ -static struct batadv_tvlv_container * -batadv_tvlv_container_get(struct batadv_priv *bat_priv, u8 type, u8 version) -{ - struct batadv_tvlv_container *tvlv_tmp, *tvlv = NULL; - - lockdep_assert_held(&bat_priv->tvlv.container_list_lock); - - hlist_for_each_entry(tvlv_tmp, &bat_priv->tvlv.container_list, list) { - if (tvlv_tmp->tvlv_hdr.type != type) - continue; - - if (tvlv_tmp->tvlv_hdr.version != version) - continue; - - kref_get(&tvlv_tmp->refcount); - tvlv = tvlv_tmp; - break; - } - - return tvlv; -} - -/** - * batadv_tvlv_container_list_size - calculate the size of the tvlv container - * list entries - * @bat_priv: the bat priv with all the soft interface information - * - * Has to be called with the appropriate locks being acquired - * (tvlv.container_list_lock). - * - * Return: size of all currently registered tvlv containers in bytes. - */ -static u16 batadv_tvlv_container_list_size(struct batadv_priv *bat_priv) -{ - struct batadv_tvlv_container *tvlv; - u16 tvlv_len = 0; - - lockdep_assert_held(&bat_priv->tvlv.container_list_lock); - - hlist_for_each_entry(tvlv, &bat_priv->tvlv.container_list, list) { - tvlv_len += sizeof(struct batadv_tvlv_hdr); - tvlv_len += ntohs(tvlv->tvlv_hdr.len); - } - - return tvlv_len; -} - -/** - * batadv_tvlv_container_remove - remove tvlv container from the tvlv container - * list - * @bat_priv: the bat priv with all the soft interface information - * @tvlv: the to be removed tvlv container - * - * Has to be called with the appropriate locks being acquired - * (tvlv.container_list_lock). - */ -static void batadv_tvlv_container_remove(struct batadv_priv *bat_priv, - struct batadv_tvlv_container *tvlv) -{ - lockdep_assert_held(&bat_priv->tvlv.container_list_lock); - - if (!tvlv) - return; - - hlist_del(&tvlv->list); - - /* first call to decrement the counter, second call to free */ - batadv_tvlv_container_put(tvlv); - batadv_tvlv_container_put(tvlv); -} - -/** - * batadv_tvlv_container_unregister - unregister tvlv container based on the - * provided type and version (both need to match) - * @bat_priv: the bat priv with all the soft interface information - * @type: tvlv container type to unregister - * @version: tvlv container type to unregister - */ -void batadv_tvlv_container_unregister(struct batadv_priv *bat_priv, - u8 type, u8 version) -{ - struct batadv_tvlv_container *tvlv; - - spin_lock_bh(&bat_priv->tvlv.container_list_lock); - tvlv = batadv_tvlv_container_get(bat_priv, type, version); - batadv_tvlv_container_remove(bat_priv, tvlv); - spin_unlock_bh(&bat_priv->tvlv.container_list_lock); -} - -/** - * batadv_tvlv_container_register - register tvlv type, version and content - * to be propagated with each (primary interface) OGM - * @bat_priv: the bat priv with all the soft interface information - * @type: tvlv container type - * @version: tvlv container version - * @tvlv_value: tvlv container content - * @tvlv_value_len: tvlv container content length - * - * If a container of the same type and version was already registered the new - * content is going to replace the old one. - */ -void batadv_tvlv_container_register(struct batadv_priv *bat_priv, - u8 type, u8 version, - void *tvlv_value, u16 tvlv_value_len) -{ - struct batadv_tvlv_container *tvlv_old, *tvlv_new; - - if (!tvlv_value) - tvlv_value_len = 0; - - tvlv_new = kzalloc(sizeof(*tvlv_new) + tvlv_value_len, GFP_ATOMIC); - if (!tvlv_new) - return; - - tvlv_new->tvlv_hdr.version = version; - tvlv_new->tvlv_hdr.type = type; - tvlv_new->tvlv_hdr.len = htons(tvlv_value_len); - - memcpy(tvlv_new + 1, tvlv_value, ntohs(tvlv_new->tvlv_hdr.len)); - INIT_HLIST_NODE(&tvlv_new->list); - kref_init(&tvlv_new->refcount); - - spin_lock_bh(&bat_priv->tvlv.container_list_lock); - tvlv_old = batadv_tvlv_container_get(bat_priv, type, version); - batadv_tvlv_container_remove(bat_priv, tvlv_old); - hlist_add_head(&tvlv_new->list, &bat_priv->tvlv.container_list); - spin_unlock_bh(&bat_priv->tvlv.container_list_lock); -} - -/** - * batadv_tvlv_realloc_packet_buff - reallocate packet buffer to accommodate - * requested packet size - * @packet_buff: packet buffer - * @packet_buff_len: packet buffer size - * @min_packet_len: requested packet minimum size - * @additional_packet_len: requested additional packet size on top of minimum - * size - * - * Return: true of the packet buffer could be changed to the requested size, - * false otherwise. - */ -static bool batadv_tvlv_realloc_packet_buff(unsigned char **packet_buff, - int *packet_buff_len, - int min_packet_len, - int additional_packet_len) -{ - unsigned char *new_buff; - - new_buff = kmalloc(min_packet_len + additional_packet_len, GFP_ATOMIC); - - /* keep old buffer if kmalloc should fail */ - if (!new_buff) - return false; - - memcpy(new_buff, *packet_buff, min_packet_len); - kfree(*packet_buff); - *packet_buff = new_buff; - *packet_buff_len = min_packet_len + additional_packet_len; - - return true; -} - -/** - * batadv_tvlv_container_ogm_append - append tvlv container content to given - * OGM packet buffer - * @bat_priv: the bat priv with all the soft interface information - * @packet_buff: ogm packet buffer - * @packet_buff_len: ogm packet buffer size including ogm header and tvlv - * content - * @packet_min_len: ogm header size to be preserved for the OGM itself - * - * The ogm packet might be enlarged or shrunk depending on the current size - * and the size of the to-be-appended tvlv containers. - * - * Return: size of all appended tvlv containers in bytes. - */ -u16 batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, - unsigned char **packet_buff, - int *packet_buff_len, int packet_min_len) -{ - struct batadv_tvlv_container *tvlv; - struct batadv_tvlv_hdr *tvlv_hdr; - u16 tvlv_value_len; - void *tvlv_value; - bool ret; - - spin_lock_bh(&bat_priv->tvlv.container_list_lock); - tvlv_value_len = batadv_tvlv_container_list_size(bat_priv); - - ret = batadv_tvlv_realloc_packet_buff(packet_buff, packet_buff_len, - packet_min_len, tvlv_value_len); - - if (!ret) - goto end; - - if (!tvlv_value_len) - goto end; - - tvlv_value = (*packet_buff) + packet_min_len; - - hlist_for_each_entry(tvlv, &bat_priv->tvlv.container_list, list) { - tvlv_hdr = tvlv_value; - tvlv_hdr->type = tvlv->tvlv_hdr.type; - tvlv_hdr->version = tvlv->tvlv_hdr.version; - tvlv_hdr->len = tvlv->tvlv_hdr.len; - tvlv_value = tvlv_hdr + 1; - memcpy(tvlv_value, tvlv + 1, ntohs(tvlv->tvlv_hdr.len)); - tvlv_value = (u8 *)tvlv_value + ntohs(tvlv->tvlv_hdr.len); - } - -end: - spin_unlock_bh(&bat_priv->tvlv.container_list_lock); - return tvlv_value_len; -} - -/** - * batadv_tvlv_call_handler - parse the given tvlv buffer to call the - * appropriate handlers - * @bat_priv: the bat priv with all the soft interface information - * @tvlv_handler: tvlv callback function handling the tvlv content - * @ogm_source: flag indicating whether the tvlv is an ogm or a unicast packet - * @orig_node: orig node emitting the ogm packet - * @src: source mac address of the unicast packet - * @dst: destination mac address of the unicast packet - * @tvlv_value: tvlv content - * @tvlv_value_len: tvlv content length - * - * Return: success if handler was not found or the return value of the handler - * callback. - */ -static int batadv_tvlv_call_handler(struct batadv_priv *bat_priv, - struct batadv_tvlv_handler *tvlv_handler, - bool ogm_source, - struct batadv_orig_node *orig_node, - u8 *src, u8 *dst, - void *tvlv_value, u16 tvlv_value_len) -{ - if (!tvlv_handler) - return NET_RX_SUCCESS; - - if (ogm_source) { - if (!tvlv_handler->ogm_handler) - return NET_RX_SUCCESS; - - if (!orig_node) - return NET_RX_SUCCESS; - - tvlv_handler->ogm_handler(bat_priv, orig_node, - BATADV_NO_FLAGS, - tvlv_value, tvlv_value_len); - tvlv_handler->flags |= BATADV_TVLV_HANDLER_OGM_CALLED; - } else { - if (!src) - return NET_RX_SUCCESS; - - if (!dst) - return NET_RX_SUCCESS; - - if (!tvlv_handler->unicast_handler) - return NET_RX_SUCCESS; - - return tvlv_handler->unicast_handler(bat_priv, src, - dst, tvlv_value, - tvlv_value_len); - } - - return NET_RX_SUCCESS; -} - -/** - * batadv_tvlv_containers_process - parse the given tvlv buffer to call the - * appropriate handlers - * @bat_priv: the bat priv with all the soft interface information - * @ogm_source: flag indicating whether the tvlv is an ogm or a unicast packet - * @orig_node: orig node emitting the ogm packet - * @src: source mac address of the unicast packet - * @dst: destination mac address of the unicast packet - * @tvlv_value: tvlv content - * @tvlv_value_len: tvlv content length - * - * Return: success when processing an OGM or the return value of all called - * handler callbacks. - */ -int batadv_tvlv_containers_process(struct batadv_priv *bat_priv, - bool ogm_source, - struct batadv_orig_node *orig_node, - u8 *src, u8 *dst, - void *tvlv_value, u16 tvlv_value_len) -{ - struct batadv_tvlv_handler *tvlv_handler; - struct batadv_tvlv_hdr *tvlv_hdr; - u16 tvlv_value_cont_len; - u8 cifnotfound = BATADV_TVLV_HANDLER_OGM_CIFNOTFND; - int ret = NET_RX_SUCCESS; - - while (tvlv_value_len >= sizeof(*tvlv_hdr)) { - tvlv_hdr = tvlv_value; - tvlv_value_cont_len = ntohs(tvlv_hdr->len); - tvlv_value = tvlv_hdr + 1; - tvlv_value_len -= sizeof(*tvlv_hdr); - - if (tvlv_value_cont_len > tvlv_value_len) - break; - - tvlv_handler = batadv_tvlv_handler_get(bat_priv, - tvlv_hdr->type, - tvlv_hdr->version); - - ret |= batadv_tvlv_call_handler(bat_priv, tvlv_handler, - ogm_source, orig_node, - src, dst, tvlv_value, - tvlv_value_cont_len); - if (tvlv_handler) - batadv_tvlv_handler_put(tvlv_handler); - tvlv_value = (u8 *)tvlv_value + tvlv_value_cont_len; - tvlv_value_len -= tvlv_value_cont_len; - } - - if (!ogm_source) - return ret; - - rcu_read_lock(); - hlist_for_each_entry_rcu(tvlv_handler, - &bat_priv->tvlv.handler_list, list) { - if ((tvlv_handler->flags & BATADV_TVLV_HANDLER_OGM_CIFNOTFND) && - !(tvlv_handler->flags & BATADV_TVLV_HANDLER_OGM_CALLED)) - tvlv_handler->ogm_handler(bat_priv, orig_node, - cifnotfound, NULL, 0); - - tvlv_handler->flags &= ~BATADV_TVLV_HANDLER_OGM_CALLED; - } - rcu_read_unlock(); - - return NET_RX_SUCCESS; -} - -/** - * batadv_tvlv_ogm_receive - process an incoming ogm and call the appropriate - * handlers - * @bat_priv: the bat priv with all the soft interface information - * @batadv_ogm_packet: ogm packet containing the tvlv containers - * @orig_node: orig node emitting the ogm packet - */ -void batadv_tvlv_ogm_receive(struct batadv_priv *bat_priv, - struct batadv_ogm_packet *batadv_ogm_packet, - struct batadv_orig_node *orig_node) -{ - void *tvlv_value; - u16 tvlv_value_len; - - if (!batadv_ogm_packet) - return; - - tvlv_value_len = ntohs(batadv_ogm_packet->tvlv_len); - if (!tvlv_value_len) - return; - - tvlv_value = batadv_ogm_packet + 1; - - batadv_tvlv_containers_process(bat_priv, true, orig_node, NULL, NULL, - tvlv_value, tvlv_value_len); -} - -/** - * batadv_tvlv_handler_register - register tvlv handler based on the provided - * type and version (both need to match) for ogm tvlv payload and/or unicast - * payload - * @bat_priv: the bat priv with all the soft interface information - * @optr: ogm tvlv handler callback function. This function receives the orig - * node, flags and the tvlv content as argument to process. - * @uptr: unicast tvlv handler callback function. This function receives the - * source & destination of the unicast packet as well as the tvlv content - * to process. - * @type: tvlv handler type to be registered - * @version: tvlv handler version to be registered - * @flags: flags to enable or disable TVLV API behavior - */ -void batadv_tvlv_handler_register(struct batadv_priv *bat_priv, - void (*optr)(struct batadv_priv *bat_priv, - struct batadv_orig_node *orig, - u8 flags, - void *tvlv_value, - u16 tvlv_value_len), - int (*uptr)(struct batadv_priv *bat_priv, - u8 *src, u8 *dst, - void *tvlv_value, - u16 tvlv_value_len), - u8 type, u8 version, u8 flags) -{ - struct batadv_tvlv_handler *tvlv_handler; - - tvlv_handler = batadv_tvlv_handler_get(bat_priv, type, version); - if (tvlv_handler) { - batadv_tvlv_handler_put(tvlv_handler); - return; - } - - tvlv_handler = kzalloc(sizeof(*tvlv_handler), GFP_ATOMIC); - if (!tvlv_handler) - return; - - tvlv_handler->ogm_handler = optr; - tvlv_handler->unicast_handler = uptr; - tvlv_handler->type = type; - tvlv_handler->version = version; - tvlv_handler->flags = flags; - kref_init(&tvlv_handler->refcount); - INIT_HLIST_NODE(&tvlv_handler->list); - - spin_lock_bh(&bat_priv->tvlv.handler_list_lock); - hlist_add_head_rcu(&tvlv_handler->list, &bat_priv->tvlv.handler_list); - spin_unlock_bh(&bat_priv->tvlv.handler_list_lock); -} - -/** - * batadv_tvlv_handler_unregister - unregister tvlv handler based on the - * provided type and version (both need to match) - * @bat_priv: the bat priv with all the soft interface information - * @type: tvlv handler type to be unregistered - * @version: tvlv handler version to be unregistered - */ -void batadv_tvlv_handler_unregister(struct batadv_priv *bat_priv, - u8 type, u8 version) -{ - struct batadv_tvlv_handler *tvlv_handler; - - tvlv_handler = batadv_tvlv_handler_get(bat_priv, type, version); - if (!tvlv_handler) - return; - - batadv_tvlv_handler_put(tvlv_handler); - spin_lock_bh(&bat_priv->tvlv.handler_list_lock); - hlist_del_rcu(&tvlv_handler->list); - spin_unlock_bh(&bat_priv->tvlv.handler_list_lock); - batadv_tvlv_handler_put(tvlv_handler); -} - -/** - * batadv_tvlv_unicast_send - send a unicast packet with tvlv payload to the - * specified host - * @bat_priv: the bat priv with all the soft interface information - * @src: source mac address of the unicast packet - * @dst: destination mac address of the unicast packet - * @type: tvlv type - * @version: tvlv version - * @tvlv_value: tvlv content - * @tvlv_value_len: tvlv content length - */ -void batadv_tvlv_unicast_send(struct batadv_priv *bat_priv, u8 *src, - u8 *dst, u8 type, u8 version, - void *tvlv_value, u16 tvlv_value_len) -{ - struct batadv_unicast_tvlv_packet *unicast_tvlv_packet; - struct batadv_tvlv_hdr *tvlv_hdr; - struct batadv_orig_node *orig_node; - struct sk_buff *skb; - unsigned char *tvlv_buff; - unsigned int tvlv_len; - ssize_t hdr_len = sizeof(*unicast_tvlv_packet); - - orig_node = batadv_orig_hash_find(bat_priv, dst); - if (!orig_node) - return; - - tvlv_len = sizeof(*tvlv_hdr) + tvlv_value_len; - - skb = netdev_alloc_skb_ip_align(NULL, ETH_HLEN + hdr_len + tvlv_len); - if (!skb) - goto out; - - skb->priority = TC_PRIO_CONTROL; - skb_reserve(skb, ETH_HLEN); - tvlv_buff = skb_put(skb, sizeof(*unicast_tvlv_packet) + tvlv_len); - unicast_tvlv_packet = (struct batadv_unicast_tvlv_packet *)tvlv_buff; - unicast_tvlv_packet->packet_type = BATADV_UNICAST_TVLV; - unicast_tvlv_packet->version = BATADV_COMPAT_VERSION; - unicast_tvlv_packet->ttl = BATADV_TTL; - unicast_tvlv_packet->reserved = 0; - unicast_tvlv_packet->tvlv_len = htons(tvlv_len); - unicast_tvlv_packet->align = 0; - ether_addr_copy(unicast_tvlv_packet->src, src); - ether_addr_copy(unicast_tvlv_packet->dst, dst); - - tvlv_buff = (unsigned char *)(unicast_tvlv_packet + 1); - tvlv_hdr = (struct batadv_tvlv_hdr *)tvlv_buff; - tvlv_hdr->version = version; - tvlv_hdr->type = type; - tvlv_hdr->len = htons(tvlv_value_len); - tvlv_buff += sizeof(*tvlv_hdr); - memcpy(tvlv_buff, tvlv_value, tvlv_value_len); - - if (batadv_send_skb_to_orig(skb, orig_node, NULL) == NET_XMIT_DROP) - kfree_skb(skb); -out: - batadv_orig_node_put(orig_node); -} - -/** * batadv_get_vid - extract the VLAN identifier from skb if any * @skb: the buffer containing the packet * @header_len: length of the batman header preceding the ethernet header diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 38f9e55..3af6582 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -190,7 +190,6 @@ enum batadv_uev_type { #include "types.h" -struct batadv_ogm_packet; struct seq_file; struct sk_buff; @@ -372,39 +371,6 @@ static inline u64 batadv_sum_counter(struct batadv_priv *bat_priv, size_t idx) */ #define BATADV_SKB_CB(__skb) ((struct batadv_skb_cb *)&((__skb)->cb[0])) -void batadv_tvlv_container_register(struct batadv_priv *bat_priv, - u8 type, u8 version, - void *tvlv_value, u16 tvlv_value_len); -u16 batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, - unsigned char **packet_buff, - int *packet_buff_len, int packet_min_len); -void batadv_tvlv_ogm_receive(struct batadv_priv *bat_priv, - struct batadv_ogm_packet *batadv_ogm_packet, - struct batadv_orig_node *orig_node); -void batadv_tvlv_container_unregister(struct batadv_priv *bat_priv, - u8 type, u8 version); - -void batadv_tvlv_handler_register(struct batadv_priv *bat_priv, - void (*optr)(struct batadv_priv *bat_priv, - struct batadv_orig_node *orig, - u8 flags, - void *tvlv_value, - u16 tvlv_value_len), - int (*uptr)(struct batadv_priv *bat_priv, - u8 *src, u8 *dst, - void *tvlv_value, - u16 tvlv_value_len), - u8 type, u8 version, u8 flags); -void batadv_tvlv_handler_unregister(struct batadv_priv *bat_priv, - u8 type, u8 version); -int batadv_tvlv_containers_process(struct batadv_priv *bat_priv, - bool ogm_source, - struct batadv_orig_node *orig_node, - u8 *src, u8 *dst, - void *tvlv_buff, u16 tvlv_buff_len); -void batadv_tvlv_unicast_send(struct batadv_priv *bat_priv, u8 *src, - u8 *dst, u8 type, u8 version, - void *tvlv_value, u16 tvlv_value_len); unsigned short batadv_get_vid(struct sk_buff *skb, size_t header_len); bool batadv_vlan_ap_isola_get(struct batadv_priv *bat_priv, unsigned short vid); diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index d3222db..0e7d78f 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -57,6 +57,7 @@ #include "hash.h" #include "packet.h" #include "translation-table.h" +#include "tvlv.h" /** * batadv_mcast_get_bridge - get the bridge on top of the softif if it exists diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c index 678f068..d0383de 100644 --- a/net/batman-adv/network-coding.c +++ b/net/batman-adv/network-coding.c @@ -55,6 +55,7 @@ #include "packet.h" #include "routing.h" #include "send.h" +#include "tvlv.h" static struct lock_class_key batadv_nc_coding_hash_lock_class_key; static struct lock_class_key batadv_nc_decoding_hash_lock_class_key; diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index 24fc753..8cb459a 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -46,6 +46,7 @@ #include "send.h" #include "soft-interface.h" #include "translation-table.h" +#include "tvlv.h" static int batadv_route_unicast_packet(struct sk_buff *skb, struct batadv_hard_iface *recv_if); diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 87bb203..6c8d624 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -51,6 +51,7 @@ #include "originator.h" #include "packet.h" #include "soft-interface.h" +#include "tvlv.h" /* hash class keys */ static struct lock_class_key batadv_tt_local_hash_lock_class_key; diff --git a/net/batman-adv/tvlv.c b/net/batman-adv/tvlv.c new file mode 100644 index 0000000..2fd542e --- /dev/null +++ b/net/batman-adv/tvlv.c @@ -0,0 +1,630 @@ +/* Copyright (C) 2007-2016 B.A.T.M.A.N. contributors: + * + * Marek Lindner, Simon Wunderlich + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#include "main.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "originator.h" +#include "packet.h" +#include "send.h" +#include "tvlv.h" + +/** + * batadv_tvlv_handler_release - release tvlv handler from lists and queue for + * free after rcu grace period + * @ref: kref pointer of the tvlv + */ +static void batadv_tvlv_handler_release(struct kref *ref) +{ + struct batadv_tvlv_handler *tvlv_handler; + + tvlv_handler = container_of(ref, struct batadv_tvlv_handler, refcount); + kfree_rcu(tvlv_handler, rcu); +} + +/** + * batadv_tvlv_handler_put - decrement the tvlv container refcounter and + * possibly release it + * @tvlv_handler: the tvlv handler to free + */ +static void batadv_tvlv_handler_put(struct batadv_tvlv_handler *tvlv_handler) +{ + kref_put(&tvlv_handler->refcount, batadv_tvlv_handler_release); +} + +/** + * batadv_tvlv_handler_get - retrieve tvlv handler from the tvlv handler list + * based on the provided type and version (both need to match) + * @bat_priv: the bat priv with all the soft interface information + * @type: tvlv handler type to look for + * @version: tvlv handler version to look for + * + * Return: tvlv handler if found or NULL otherwise. + */ +static struct batadv_tvlv_handler * +batadv_tvlv_handler_get(struct batadv_priv *bat_priv, u8 type, u8 version) +{ + struct batadv_tvlv_handler *tvlv_handler_tmp, *tvlv_handler = NULL; + + rcu_read_lock(); + hlist_for_each_entry_rcu(tvlv_handler_tmp, + &bat_priv->tvlv.handler_list, list) { + if (tvlv_handler_tmp->type != type) + continue; + + if (tvlv_handler_tmp->version != version) + continue; + + if (!kref_get_unless_zero(&tvlv_handler_tmp->refcount)) + continue; + + tvlv_handler = tvlv_handler_tmp; + break; + } + rcu_read_unlock(); + + return tvlv_handler; +} + +/** + * batadv_tvlv_container_release - release tvlv from lists and free + * @ref: kref pointer of the tvlv + */ +static void batadv_tvlv_container_release(struct kref *ref) +{ + struct batadv_tvlv_container *tvlv; + + tvlv = container_of(ref, struct batadv_tvlv_container, refcount); + kfree(tvlv); +} + +/** + * batadv_tvlv_container_put - decrement the tvlv container refcounter and + * possibly release it + * @tvlv: the tvlv container to free + */ +static void batadv_tvlv_container_put(struct batadv_tvlv_container *tvlv) +{ + kref_put(&tvlv->refcount, batadv_tvlv_container_release); +} + +/** + * batadv_tvlv_container_get - retrieve tvlv container from the tvlv container + * list based on the provided type and version (both need to match) + * @bat_priv: the bat priv with all the soft interface information + * @type: tvlv container type to look for + * @version: tvlv container version to look for + * + * Has to be called with the appropriate locks being acquired + * (tvlv.container_list_lock). + * + * Return: tvlv container if found or NULL otherwise. + */ +static struct batadv_tvlv_container * +batadv_tvlv_container_get(struct batadv_priv *bat_priv, u8 type, u8 version) +{ + struct batadv_tvlv_container *tvlv_tmp, *tvlv = NULL; + + lockdep_assert_held(&bat_priv->tvlv.container_list_lock); + + hlist_for_each_entry(tvlv_tmp, &bat_priv->tvlv.container_list, list) { + if (tvlv_tmp->tvlv_hdr.type != type) + continue; + + if (tvlv_tmp->tvlv_hdr.version != version) + continue; + + kref_get(&tvlv_tmp->refcount); + tvlv = tvlv_tmp; + break; + } + + return tvlv; +} + +/** + * batadv_tvlv_container_list_size - calculate the size of the tvlv container + * list entries + * @bat_priv: the bat priv with all the soft interface information + * + * Has to be called with the appropriate locks being acquired + * (tvlv.container_list_lock). + * + * Return: size of all currently registered tvlv containers in bytes. + */ +static u16 batadv_tvlv_container_list_size(struct batadv_priv *bat_priv) +{ + struct batadv_tvlv_container *tvlv; + u16 tvlv_len = 0; + + lockdep_assert_held(&bat_priv->tvlv.container_list_lock); + + hlist_for_each_entry(tvlv, &bat_priv->tvlv.container_list, list) { + tvlv_len += sizeof(struct batadv_tvlv_hdr); + tvlv_len += ntohs(tvlv->tvlv_hdr.len); + } + + return tvlv_len; +} + +/** + * batadv_tvlv_container_remove - remove tvlv container from the tvlv container + * list + * @bat_priv: the bat priv with all the soft interface information + * @tvlv: the to be removed tvlv container + * + * Has to be called with the appropriate locks being acquired + * (tvlv.container_list_lock). + */ +static void batadv_tvlv_container_remove(struct batadv_priv *bat_priv, + struct batadv_tvlv_container *tvlv) +{ + lockdep_assert_held(&bat_priv->tvlv.container_list_lock); + + if (!tvlv) + return; + + hlist_del(&tvlv->list); + + /* first call to decrement the counter, second call to free */ + batadv_tvlv_container_put(tvlv); + batadv_tvlv_container_put(tvlv); +} + +/** + * batadv_tvlv_container_unregister - unregister tvlv container based on the + * provided type and version (both need to match) + * @bat_priv: the bat priv with all the soft interface information + * @type: tvlv container type to unregister + * @version: tvlv container type to unregister + */ +void batadv_tvlv_container_unregister(struct batadv_priv *bat_priv, + u8 type, u8 version) +{ + struct batadv_tvlv_container *tvlv; + + spin_lock_bh(&bat_priv->tvlv.container_list_lock); + tvlv = batadv_tvlv_container_get(bat_priv, type, version); + batadv_tvlv_container_remove(bat_priv, tvlv); + spin_unlock_bh(&bat_priv->tvlv.container_list_lock); +} + +/** + * batadv_tvlv_container_register - register tvlv type, version and content + * to be propagated with each (primary interface) OGM + * @bat_priv: the bat priv with all the soft interface information + * @type: tvlv container type + * @version: tvlv container version + * @tvlv_value: tvlv container content + * @tvlv_value_len: tvlv container content length + * + * If a container of the same type and version was already registered the new + * content is going to replace the old one. + */ +void batadv_tvlv_container_register(struct batadv_priv *bat_priv, + u8 type, u8 version, + void *tvlv_value, u16 tvlv_value_len) +{ + struct batadv_tvlv_container *tvlv_old, *tvlv_new; + + if (!tvlv_value) + tvlv_value_len = 0; + + tvlv_new = kzalloc(sizeof(*tvlv_new) + tvlv_value_len, GFP_ATOMIC); + if (!tvlv_new) + return; + + tvlv_new->tvlv_hdr.version = version; + tvlv_new->tvlv_hdr.type = type; + tvlv_new->tvlv_hdr.len = htons(tvlv_value_len); + + memcpy(tvlv_new + 1, tvlv_value, ntohs(tvlv_new->tvlv_hdr.len)); + INIT_HLIST_NODE(&tvlv_new->list); + kref_init(&tvlv_new->refcount); + + spin_lock_bh(&bat_priv->tvlv.container_list_lock); + tvlv_old = batadv_tvlv_container_get(bat_priv, type, version); + batadv_tvlv_container_remove(bat_priv, tvlv_old); + hlist_add_head(&tvlv_new->list, &bat_priv->tvlv.container_list); + spin_unlock_bh(&bat_priv->tvlv.container_list_lock); +} + +/** + * batadv_tvlv_realloc_packet_buff - reallocate packet buffer to accommodate + * requested packet size + * @packet_buff: packet buffer + * @packet_buff_len: packet buffer size + * @min_packet_len: requested packet minimum size + * @additional_packet_len: requested additional packet size on top of minimum + * size + * + * Return: true of the packet buffer could be changed to the requested size, + * false otherwise. + */ +static bool batadv_tvlv_realloc_packet_buff(unsigned char **packet_buff, + int *packet_buff_len, + int min_packet_len, + int additional_packet_len) +{ + unsigned char *new_buff; + + new_buff = kmalloc(min_packet_len + additional_packet_len, GFP_ATOMIC); + + /* keep old buffer if kmalloc should fail */ + if (!new_buff) + return false; + + memcpy(new_buff, *packet_buff, min_packet_len); + kfree(*packet_buff); + *packet_buff = new_buff; + *packet_buff_len = min_packet_len + additional_packet_len; + + return true; +} + +/** + * batadv_tvlv_container_ogm_append - append tvlv container content to given + * OGM packet buffer + * @bat_priv: the bat priv with all the soft interface information + * @packet_buff: ogm packet buffer + * @packet_buff_len: ogm packet buffer size including ogm header and tvlv + * content + * @packet_min_len: ogm header size to be preserved for the OGM itself + * + * The ogm packet might be enlarged or shrunk depending on the current size + * and the size of the to-be-appended tvlv containers. + * + * Return: size of all appended tvlv containers in bytes. + */ +u16 batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, + unsigned char **packet_buff, + int *packet_buff_len, int packet_min_len) +{ + struct batadv_tvlv_container *tvlv; + struct batadv_tvlv_hdr *tvlv_hdr; + u16 tvlv_value_len; + void *tvlv_value; + bool ret; + + spin_lock_bh(&bat_priv->tvlv.container_list_lock); + tvlv_value_len = batadv_tvlv_container_list_size(bat_priv); + + ret = batadv_tvlv_realloc_packet_buff(packet_buff, packet_buff_len, + packet_min_len, tvlv_value_len); + + if (!ret) + goto end; + + if (!tvlv_value_len) + goto end; + + tvlv_value = (*packet_buff) + packet_min_len; + + hlist_for_each_entry(tvlv, &bat_priv->tvlv.container_list, list) { + tvlv_hdr = tvlv_value; + tvlv_hdr->type = tvlv->tvlv_hdr.type; + tvlv_hdr->version = tvlv->tvlv_hdr.version; + tvlv_hdr->len = tvlv->tvlv_hdr.len; + tvlv_value = tvlv_hdr + 1; + memcpy(tvlv_value, tvlv + 1, ntohs(tvlv->tvlv_hdr.len)); + tvlv_value = (u8 *)tvlv_value + ntohs(tvlv->tvlv_hdr.len); + } + +end: + spin_unlock_bh(&bat_priv->tvlv.container_list_lock); + return tvlv_value_len; +} + +/** + * batadv_tvlv_call_handler - parse the given tvlv buffer to call the + * appropriate handlers + * @bat_priv: the bat priv with all the soft interface information + * @tvlv_handler: tvlv callback function handling the tvlv content + * @ogm_source: flag indicating whether the tvlv is an ogm or a unicast packet + * @orig_node: orig node emitting the ogm packet + * @src: source mac address of the unicast packet + * @dst: destination mac address of the unicast packet + * @tvlv_value: tvlv content + * @tvlv_value_len: tvlv content length + * + * Return: success if handler was not found or the return value of the handler + * callback. + */ +static int batadv_tvlv_call_handler(struct batadv_priv *bat_priv, + struct batadv_tvlv_handler *tvlv_handler, + bool ogm_source, + struct batadv_orig_node *orig_node, + u8 *src, u8 *dst, + void *tvlv_value, u16 tvlv_value_len) +{ + if (!tvlv_handler) + return NET_RX_SUCCESS; + + if (ogm_source) { + if (!tvlv_handler->ogm_handler) + return NET_RX_SUCCESS; + + if (!orig_node) + return NET_RX_SUCCESS; + + tvlv_handler->ogm_handler(bat_priv, orig_node, + BATADV_NO_FLAGS, + tvlv_value, tvlv_value_len); + tvlv_handler->flags |= BATADV_TVLV_HANDLER_OGM_CALLED; + } else { + if (!src) + return NET_RX_SUCCESS; + + if (!dst) + return NET_RX_SUCCESS; + + if (!tvlv_handler->unicast_handler) + return NET_RX_SUCCESS; + + return tvlv_handler->unicast_handler(bat_priv, src, + dst, tvlv_value, + tvlv_value_len); + } + + return NET_RX_SUCCESS; +} + +/** + * batadv_tvlv_containers_process - parse the given tvlv buffer to call the + * appropriate handlers + * @bat_priv: the bat priv with all the soft interface information + * @ogm_source: flag indicating whether the tvlv is an ogm or a unicast packet + * @orig_node: orig node emitting the ogm packet + * @src: source mac address of the unicast packet + * @dst: destination mac address of the unicast packet + * @tvlv_value: tvlv content + * @tvlv_value_len: tvlv content length + * + * Return: success when processing an OGM or the return value of all called + * handler callbacks. + */ +int batadv_tvlv_containers_process(struct batadv_priv *bat_priv, + bool ogm_source, + struct batadv_orig_node *orig_node, + u8 *src, u8 *dst, + void *tvlv_value, u16 tvlv_value_len) +{ + struct batadv_tvlv_handler *tvlv_handler; + struct batadv_tvlv_hdr *tvlv_hdr; + u16 tvlv_value_cont_len; + u8 cifnotfound = BATADV_TVLV_HANDLER_OGM_CIFNOTFND; + int ret = NET_RX_SUCCESS; + + while (tvlv_value_len >= sizeof(*tvlv_hdr)) { + tvlv_hdr = tvlv_value; + tvlv_value_cont_len = ntohs(tvlv_hdr->len); + tvlv_value = tvlv_hdr + 1; + tvlv_value_len -= sizeof(*tvlv_hdr); + + if (tvlv_value_cont_len > tvlv_value_len) + break; + + tvlv_handler = batadv_tvlv_handler_get(bat_priv, + tvlv_hdr->type, + tvlv_hdr->version); + + ret |= batadv_tvlv_call_handler(bat_priv, tvlv_handler, + ogm_source, orig_node, + src, dst, tvlv_value, + tvlv_value_cont_len); + if (tvlv_handler) + batadv_tvlv_handler_put(tvlv_handler); + tvlv_value = (u8 *)tvlv_value + tvlv_value_cont_len; + tvlv_value_len -= tvlv_value_cont_len; + } + + if (!ogm_source) + return ret; + + rcu_read_lock(); + hlist_for_each_entry_rcu(tvlv_handler, + &bat_priv->tvlv.handler_list, list) { + if ((tvlv_handler->flags & BATADV_TVLV_HANDLER_OGM_CIFNOTFND) && + !(tvlv_handler->flags & BATADV_TVLV_HANDLER_OGM_CALLED)) + tvlv_handler->ogm_handler(bat_priv, orig_node, + cifnotfound, NULL, 0); + + tvlv_handler->flags &= ~BATADV_TVLV_HANDLER_OGM_CALLED; + } + rcu_read_unlock(); + + return NET_RX_SUCCESS; +} + +/** + * batadv_tvlv_ogm_receive - process an incoming ogm and call the appropriate + * handlers + * @bat_priv: the bat priv with all the soft interface information + * @batadv_ogm_packet: ogm packet containing the tvlv containers + * @orig_node: orig node emitting the ogm packet + */ +void batadv_tvlv_ogm_receive(struct batadv_priv *bat_priv, + struct batadv_ogm_packet *batadv_ogm_packet, + struct batadv_orig_node *orig_node) +{ + void *tvlv_value; + u16 tvlv_value_len; + + if (!batadv_ogm_packet) + return; + + tvlv_value_len = ntohs(batadv_ogm_packet->tvlv_len); + if (!tvlv_value_len) + return; + + tvlv_value = batadv_ogm_packet + 1; + + batadv_tvlv_containers_process(bat_priv, true, orig_node, NULL, NULL, + tvlv_value, tvlv_value_len); +} + +/** + * batadv_tvlv_handler_register - register tvlv handler based on the provided + * type and version (both need to match) for ogm tvlv payload and/or unicast + * payload + * @bat_priv: the bat priv with all the soft interface information + * @optr: ogm tvlv handler callback function. This function receives the orig + * node, flags and the tvlv content as argument to process. + * @uptr: unicast tvlv handler callback function. This function receives the + * source & destination of the unicast packet as well as the tvlv content + * to process. + * @type: tvlv handler type to be registered + * @version: tvlv handler version to be registered + * @flags: flags to enable or disable TVLV API behavior + */ +void batadv_tvlv_handler_register(struct batadv_priv *bat_priv, + void (*optr)(struct batadv_priv *bat_priv, + struct batadv_orig_node *orig, + u8 flags, + void *tvlv_value, + u16 tvlv_value_len), + int (*uptr)(struct batadv_priv *bat_priv, + u8 *src, u8 *dst, + void *tvlv_value, + u16 tvlv_value_len), + u8 type, u8 version, u8 flags) +{ + struct batadv_tvlv_handler *tvlv_handler; + + tvlv_handler = batadv_tvlv_handler_get(bat_priv, type, version); + if (tvlv_handler) { + batadv_tvlv_handler_put(tvlv_handler); + return; + } + + tvlv_handler = kzalloc(sizeof(*tvlv_handler), GFP_ATOMIC); + if (!tvlv_handler) + return; + + tvlv_handler->ogm_handler = optr; + tvlv_handler->unicast_handler = uptr; + tvlv_handler->type = type; + tvlv_handler->version = version; + tvlv_handler->flags = flags; + kref_init(&tvlv_handler->refcount); + INIT_HLIST_NODE(&tvlv_handler->list); + + spin_lock_bh(&bat_priv->tvlv.handler_list_lock); + hlist_add_head_rcu(&tvlv_handler->list, &bat_priv->tvlv.handler_list); + spin_unlock_bh(&bat_priv->tvlv.handler_list_lock); +} + +/** + * batadv_tvlv_handler_unregister - unregister tvlv handler based on the + * provided type and version (both need to match) + * @bat_priv: the bat priv with all the soft interface information + * @type: tvlv handler type to be unregistered + * @version: tvlv handler version to be unregistered + */ +void batadv_tvlv_handler_unregister(struct batadv_priv *bat_priv, + u8 type, u8 version) +{ + struct batadv_tvlv_handler *tvlv_handler; + + tvlv_handler = batadv_tvlv_handler_get(bat_priv, type, version); + if (!tvlv_handler) + return; + + batadv_tvlv_handler_put(tvlv_handler); + spin_lock_bh(&bat_priv->tvlv.handler_list_lock); + hlist_del_rcu(&tvlv_handler->list); + spin_unlock_bh(&bat_priv->tvlv.handler_list_lock); + batadv_tvlv_handler_put(tvlv_handler); +} + +/** + * batadv_tvlv_unicast_send - send a unicast packet with tvlv payload to the + * specified host + * @bat_priv: the bat priv with all the soft interface information + * @src: source mac address of the unicast packet + * @dst: destination mac address of the unicast packet + * @type: tvlv type + * @version: tvlv version + * @tvlv_value: tvlv content + * @tvlv_value_len: tvlv content length + */ +void batadv_tvlv_unicast_send(struct batadv_priv *bat_priv, u8 *src, + u8 *dst, u8 type, u8 version, + void *tvlv_value, u16 tvlv_value_len) +{ + struct batadv_unicast_tvlv_packet *unicast_tvlv_packet; + struct batadv_tvlv_hdr *tvlv_hdr; + struct batadv_orig_node *orig_node; + struct sk_buff *skb; + unsigned char *tvlv_buff; + unsigned int tvlv_len; + ssize_t hdr_len = sizeof(*unicast_tvlv_packet); + + orig_node = batadv_orig_hash_find(bat_priv, dst); + if (!orig_node) + return; + + tvlv_len = sizeof(*tvlv_hdr) + tvlv_value_len; + + skb = netdev_alloc_skb_ip_align(NULL, ETH_HLEN + hdr_len + tvlv_len); + if (!skb) + goto out; + + skb->priority = TC_PRIO_CONTROL; + skb_reserve(skb, ETH_HLEN); + tvlv_buff = skb_put(skb, sizeof(*unicast_tvlv_packet) + tvlv_len); + unicast_tvlv_packet = (struct batadv_unicast_tvlv_packet *)tvlv_buff; + unicast_tvlv_packet->packet_type = BATADV_UNICAST_TVLV; + unicast_tvlv_packet->version = BATADV_COMPAT_VERSION; + unicast_tvlv_packet->ttl = BATADV_TTL; + unicast_tvlv_packet->reserved = 0; + unicast_tvlv_packet->tvlv_len = htons(tvlv_len); + unicast_tvlv_packet->align = 0; + ether_addr_copy(unicast_tvlv_packet->src, src); + ether_addr_copy(unicast_tvlv_packet->dst, dst); + + tvlv_buff = (unsigned char *)(unicast_tvlv_packet + 1); + tvlv_hdr = (struct batadv_tvlv_hdr *)tvlv_buff; + tvlv_hdr->version = version; + tvlv_hdr->type = type; + tvlv_hdr->len = htons(tvlv_value_len); + tvlv_buff += sizeof(*tvlv_hdr); + memcpy(tvlv_buff, tvlv_value, tvlv_value_len); + + if (batadv_send_skb_to_orig(skb, orig_node, NULL) == NET_XMIT_DROP) + kfree_skb(skb); +out: + batadv_orig_node_put(orig_node); +} diff --git a/net/batman-adv/tvlv.h b/net/batman-adv/tvlv.h new file mode 100644 index 0000000..e4369b5 --- /dev/null +++ b/net/batman-adv/tvlv.h @@ -0,0 +1,61 @@ +/* Copyright (C) 2007-2016 B.A.T.M.A.N. contributors: + * + * Marek Lindner, Simon Wunderlich + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#ifndef _NET_BATMAN_ADV_TVLV_H_ +#define _NET_BATMAN_ADV_TVLV_H_ + +#include "main.h" + +#include + +struct batadv_ogm_packet; + +void batadv_tvlv_container_register(struct batadv_priv *bat_priv, + u8 type, u8 version, + void *tvlv_value, u16 tvlv_value_len); +u16 batadv_tvlv_container_ogm_append(struct batadv_priv *bat_priv, + unsigned char **packet_buff, + int *packet_buff_len, int packet_min_len); +void batadv_tvlv_ogm_receive(struct batadv_priv *bat_priv, + struct batadv_ogm_packet *batadv_ogm_packet, + struct batadv_orig_node *orig_node); +void batadv_tvlv_container_unregister(struct batadv_priv *bat_priv, + u8 type, u8 version); + +void batadv_tvlv_handler_register(struct batadv_priv *bat_priv, + void (*optr)(struct batadv_priv *bat_priv, + struct batadv_orig_node *orig, + u8 flags, + void *tvlv_value, + u16 tvlv_value_len), + int (*uptr)(struct batadv_priv *bat_priv, + u8 *src, u8 *dst, + void *tvlv_value, + u16 tvlv_value_len), + u8 type, u8 version, u8 flags); +void batadv_tvlv_handler_unregister(struct batadv_priv *bat_priv, + u8 type, u8 version); +int batadv_tvlv_containers_process(struct batadv_priv *bat_priv, + bool ogm_source, + struct batadv_orig_node *orig_node, + u8 *src, u8 *dst, + void *tvlv_buff, u16 tvlv_buff_len); +void batadv_tvlv_unicast_send(struct batadv_priv *bat_priv, u8 *src, + u8 *dst, u8 type, u8 version, + void *tvlv_value, u16 tvlv_value_len); + +#endif /* _NET_BATMAN_ADV_TVLV_H_ */ -- cgit v0.10.2 From 687937ab34896d9c39b80b68d304c68ca3c2b207 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Linus=20L=C3=BCssing?= Date: Tue, 10 May 2016 18:41:25 +0200 Subject: batman-adv: Add multicast optimization support for bridged setups MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With this patch we are finally able to support multicast optimizations in bridged setups, too. So far, if a bridge was added on top of a soft-interface (e.g. bat0) the batman-adv multicast optimizations needed to be disabled to avoid packetloss. Current Linux bridge implementations and API can now provide us with the so far missing information about interested but "remote" multicast receivers behind bridge ports. The Linux bridge performs the detection of remote participants interested in multicast packets with its own and mature so called IGMP and MLD snooping code and stores that in its database. With the new API provided by the bridge batman-adv can now simply hook into this database. We then reliably announce the gathered multicast listeners to other nodes through the batman-adv translation table. Additionally, the Linux bridge provides us with the information about whether an IGMP/MLD querier exists. If there is none then we need to disable multicast optimizations as we cannot learn about multicast listeners on external, bridged-in host then. Tested-by: Simon Wunderlich Signed-off-by: Linus Lüssing Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/Kconfig b/net/batman-adv/Kconfig index b7ba97d..833bb14 100644 --- a/net/batman-adv/Kconfig +++ b/net/batman-adv/Kconfig @@ -66,7 +66,7 @@ config BATMAN_ADV_NC config BATMAN_ADV_MCAST bool "Multicast optimisation" - depends on BATMAN_ADV && INET + depends on BATMAN_ADV && INET && !(BRIDGE=m && BATMAN_ADV=y) default n help This option enables the multicast optimisation which aims to diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index 4673328..eb30316 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include @@ -36,6 +37,7 @@ #include #include #include +#include #include #include #include @@ -45,18 +47,53 @@ #include #include #include +#include +#include #include #include "packet.h" #include "translation-table.h" /** + * batadv_mcast_get_bridge - get the bridge on top of the softif if it exists + * @soft_iface: netdev struct of the mesh interface + * + * If the given soft interface has a bridge on top then the refcount + * of the according net device is increased. + * + * Return: NULL if no such bridge exists. Otherwise the net device of the + * bridge. + */ +static struct net_device *batadv_mcast_get_bridge(struct net_device *soft_iface) +{ + struct net_device *upper = soft_iface; + + rcu_read_lock(); + do { + upper = netdev_master_upper_dev_get_rcu(upper); + } while (upper && !(upper->priv_flags & IFF_EBRIDGE)); + + if (upper) + dev_hold(upper); + rcu_read_unlock(); + + return upper; +} + +/** * batadv_mcast_mla_softif_get - get softif multicast listeners * @dev: the device to collect multicast addresses from * @mcast_list: a list to put found addresses into * - * Collect multicast addresses of the local multicast listeners - * on the given soft interface, dev, in the given mcast_list. + * Collects multicast addresses of multicast listeners residing + * on this kernel on the given soft interface, dev, in + * the given mcast_list. In general, multicast listeners provided by + * your multicast receiving applications run directly on this node. + * + * If there is a bridge interface on top of dev, collects from that one + * instead. Just like with IP addresses and routes, multicast listeners + * will(/should) register to the bridge interface instead of an + * enslaved bat0. * * Return: -ENOMEM on memory allocation error or the number of * items added to the mcast_list otherwise. @@ -64,12 +101,13 @@ static int batadv_mcast_mla_softif_get(struct net_device *dev, struct hlist_head *mcast_list) { + struct net_device *bridge = batadv_mcast_get_bridge(dev); struct netdev_hw_addr *mc_list_entry; struct batadv_hw_addr *new; int ret = 0; - netif_addr_lock_bh(dev); - netdev_for_each_mc_addr(mc_list_entry, dev) { + netif_addr_lock_bh(bridge ? bridge : dev); + netdev_for_each_mc_addr(mc_list_entry, bridge ? bridge : dev) { new = kmalloc(sizeof(*new), GFP_ATOMIC); if (!new) { ret = -ENOMEM; @@ -80,7 +118,10 @@ static int batadv_mcast_mla_softif_get(struct net_device *dev, hlist_add_head(&new->list, mcast_list); ret++; } - netif_addr_unlock_bh(dev); + netif_addr_unlock_bh(bridge ? bridge : dev); + + if (bridge) + dev_put(bridge); return ret; } @@ -106,6 +147,83 @@ static bool batadv_mcast_mla_is_duplicate(u8 *mcast_addr, } /** + * batadv_mcast_mla_br_addr_cpy - copy a bridge multicast address + * @dst: destination to write to - a multicast MAC address + * @src: source to read from - a multicast IP address + * + * Converts a given multicast IPv4/IPv6 address from a bridge + * to its matching multicast MAC address and copies it into the given + * destination buffer. + * + * Caller needs to make sure the destination buffer can hold + * at least ETH_ALEN bytes. + */ +static void batadv_mcast_mla_br_addr_cpy(char *dst, const struct br_ip *src) +{ + if (src->proto == htons(ETH_P_IP)) + ip_eth_mc_map(src->u.ip4, dst); +#if IS_ENABLED(CONFIG_IPV6) + else if (src->proto == htons(ETH_P_IPV6)) + ipv6_eth_mc_map(&src->u.ip6, dst); +#endif + else + eth_zero_addr(dst); +} + +/** + * batadv_mcast_mla_bridge_get - get bridged-in multicast listeners + * @dev: a bridge slave whose bridge to collect multicast addresses from + * @mcast_list: a list to put found addresses into + * + * Collects multicast addresses of multicast listeners residing + * on foreign, non-mesh devices which we gave access to our mesh via + * a bridge on top of the given soft interface, dev, in the given + * mcast_list. + * + * Return: -ENOMEM on memory allocation error or the number of + * items added to the mcast_list otherwise. + */ +static int batadv_mcast_mla_bridge_get(struct net_device *dev, + struct hlist_head *mcast_list) +{ + struct list_head bridge_mcast_list = LIST_HEAD_INIT(bridge_mcast_list); + struct br_ip_list *br_ip_entry, *tmp; + struct batadv_hw_addr *new; + u8 mcast_addr[ETH_ALEN]; + int ret; + + /* we don't need to detect these devices/listeners, the IGMP/MLD + * snooping code of the Linux bridge already does that for us + */ + ret = br_multicast_list_adjacent(dev, &bridge_mcast_list); + if (ret < 0) + goto out; + + list_for_each_entry(br_ip_entry, &bridge_mcast_list, list) { + batadv_mcast_mla_br_addr_cpy(mcast_addr, &br_ip_entry->addr); + if (batadv_mcast_mla_is_duplicate(mcast_addr, mcast_list)) + continue; + + new = kmalloc(sizeof(*new), GFP_ATOMIC); + if (!new) { + ret = -ENOMEM; + break; + } + + ether_addr_copy(new->addr, mcast_addr); + hlist_add_head(&new->list, mcast_list); + } + +out: + list_for_each_entry_safe(br_ip_entry, tmp, &bridge_mcast_list, list) { + list_del(&br_ip_entry->list); + kfree(br_ip_entry); + } + + return ret; +} + +/** * batadv_mcast_mla_list_free - free a list of multicast addresses * @bat_priv: the bat priv with all the soft interface information * @mcast_list: the list to free @@ -222,29 +340,51 @@ static bool batadv_mcast_has_bridge(struct batadv_priv *bat_priv) * Updates the own multicast tvlv with our current multicast related settings, * capabilities and inabilities. * - * Return: true if the tvlv container is registered afterwards. Otherwise - * returns false. + * Return: false if we want all IPv4 && IPv6 multicast traffic and true + * otherwise. */ static bool batadv_mcast_mla_tvlv_update(struct batadv_priv *bat_priv) { struct batadv_tvlv_mcast_data mcast_data; + struct batadv_mcast_querier_state querier4 = {false, false}; + struct batadv_mcast_querier_state querier6 = {false, false}; + struct net_device *dev = bat_priv->soft_iface; mcast_data.flags = BATADV_NO_FLAGS; memset(mcast_data.reserved, 0, sizeof(mcast_data.reserved)); - /* Avoid attaching MLAs, if there is a bridge on top of our soft - * interface, we don't support that yet (TODO) + bat_priv->mcast.bridged = batadv_mcast_has_bridge(bat_priv); + if (!bat_priv->mcast.bridged) + goto update; + +#if !IS_ENABLED(CONFIG_BRIDGE_IGMP_SNOOPING) + pr_warn_once("No bridge IGMP snooping compiled - multicast optimizations disabled\n"); +#endif + + querier4.exists = br_multicast_has_querier_anywhere(dev, ETH_P_IP); + querier4.shadowing = br_multicast_has_querier_adjacent(dev, ETH_P_IP); + + querier6.exists = br_multicast_has_querier_anywhere(dev, ETH_P_IPV6); + querier6.shadowing = br_multicast_has_querier_adjacent(dev, ETH_P_IPV6); + + mcast_data.flags |= BATADV_MCAST_WANT_ALL_UNSNOOPABLES; + + /* 1) If no querier exists at all, then multicast listeners on + * our local TT clients behind the bridge will keep silent. + * 2) If the selected querier is on one of our local TT clients, + * behind the bridge, then this querier might shadow multicast + * listeners on our local TT clients, behind this bridge. + * + * In both cases, we will signalize other batman nodes that + * we need all multicast traffic of the according protocol. */ - if (batadv_mcast_has_bridge(bat_priv)) { - if (bat_priv->mcast.enabled) { - batadv_tvlv_container_unregister(bat_priv, - BATADV_TVLV_MCAST, 2); - bat_priv->mcast.enabled = false; - } + if (!querier4.exists || querier4.shadowing) + mcast_data.flags |= BATADV_MCAST_WANT_ALL_IPV4; - return false; - } + if (!querier6.exists || querier6.shadowing) + mcast_data.flags |= BATADV_MCAST_WANT_ALL_IPV6; +update: if (!bat_priv->mcast.enabled || mcast_data.flags != bat_priv->mcast.flags) { batadv_tvlv_container_register(bat_priv, BATADV_TVLV_MCAST, 2, @@ -253,7 +393,8 @@ static bool batadv_mcast_mla_tvlv_update(struct batadv_priv *bat_priv) bat_priv->mcast.enabled = true; } - return true; + return !(mcast_data.flags & + (BATADV_MCAST_WANT_ALL_IPV4 + BATADV_MCAST_WANT_ALL_IPV6)); } /** @@ -276,6 +417,10 @@ void batadv_mcast_mla_update(struct batadv_priv *bat_priv) if (ret < 0) goto out; + ret = batadv_mcast_mla_bridge_get(soft_iface, &mcast_list); + if (ret < 0) + goto out; + update: batadv_mcast_mla_tt_retract(bat_priv, &mcast_list); batadv_mcast_mla_tt_add(bat_priv, &mcast_list); diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 32c6d0e..83303c2 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -755,6 +755,17 @@ struct batadv_priv_dat { #ifdef CONFIG_BATMAN_ADV_MCAST /** + * struct batadv_mcast_querier_state - IGMP/MLD querier state when bridged + * @exists: whether a querier exists in the mesh + * @shadowing: if a querier exists, whether it is potentially shadowing + * multicast listeners (i.e. querier is behind our own bridge segment) + */ +struct batadv_mcast_querier_state { + bool exists; + bool shadowing; +}; + +/** * struct batadv_priv_mcast - per mesh interface mcast data * @mla_list: list of multicast addresses we are currently announcing via TT * @want_all_unsnoopables_list: a list of orig_nodes wanting all unsnoopable @@ -763,6 +774,7 @@ struct batadv_priv_dat { * @want_all_ipv6_list: a list of orig_nodes wanting all IPv6 multicast traffic * @flags: the flags we have last sent in our mcast tvlv * @enabled: whether the multicast tvlv is currently enabled + * @bridged: whether the soft interface has a bridge on top * @num_disabled: number of nodes that have no mcast tvlv * @num_want_all_unsnoopables: number of nodes wanting unsnoopable IP traffic * @num_want_all_ipv4: counter for items in want_all_ipv4_list @@ -777,6 +789,7 @@ struct batadv_priv_mcast { struct hlist_head want_all_ipv6_list; u8 flags; bool enabled; + bool bridged; atomic_t num_disabled; atomic_t num_want_all_unsnoopables; atomic_t num_want_all_ipv4; -- cgit v0.10.2 From 01d350d14712d1e8dbf2b00c82d2fc7c48d34e04 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sun, 15 May 2016 11:07:44 +0200 Subject: batman-adv: move bat_algo functions into a separate file The bat_algo functionality in main.c is mostly unrelated to the rest of the content. It still takes up a large portion of this source file (~15%, 103 lines). Moving it to a separate file makes it better visible as a main component of the batman-adv implementation and hides it less in the other helper functions in main.c. Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile index 5c6ece0..5260c17 100644 --- a/net/batman-adv/Makefile +++ b/net/batman-adv/Makefile @@ -17,6 +17,7 @@ # obj-$(CONFIG_BATMAN_ADV) += batman-adv.o +batman-adv-y += bat_algo.o batman-adv-y += bat_iv_ogm.o batman-adv-$(CONFIG_BATMAN_ADV_BATMAN_V) += bat_v.o batman-adv-$(CONFIG_BATMAN_ADV_BATMAN_V) += bat_v_elp.o diff --git a/net/batman-adv/bat_algo.c b/net/batman-adv/bat_algo.c new file mode 100644 index 0000000..610d4de --- /dev/null +++ b/net/batman-adv/bat_algo.c @@ -0,0 +1,140 @@ +/* Copyright (C) 2007-2016 B.A.T.M.A.N. contributors: + * + * Marek Lindner, Simon Wunderlich + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#include "main.h" + +#include +#include +#include +#include +#include +#include +#include + +#include "bat_algo.h" + +char batadv_routing_algo[20] = "BATMAN_IV"; +static struct hlist_head batadv_algo_list; + +/** + * batadv_algo_init - Initialize batman-adv algorithm management data structures + */ +void batadv_algo_init(void) +{ + INIT_HLIST_HEAD(&batadv_algo_list); +} + +static struct batadv_algo_ops *batadv_algo_get(char *name) +{ + struct batadv_algo_ops *bat_algo_ops = NULL, *bat_algo_ops_tmp; + + hlist_for_each_entry(bat_algo_ops_tmp, &batadv_algo_list, list) { + if (strcmp(bat_algo_ops_tmp->name, name) != 0) + continue; + + bat_algo_ops = bat_algo_ops_tmp; + break; + } + + return bat_algo_ops; +} + +int batadv_algo_register(struct batadv_algo_ops *bat_algo_ops) +{ + struct batadv_algo_ops *bat_algo_ops_tmp; + + bat_algo_ops_tmp = batadv_algo_get(bat_algo_ops->name); + if (bat_algo_ops_tmp) { + pr_info("Trying to register already registered routing algorithm: %s\n", + bat_algo_ops->name); + return -EEXIST; + } + + /* all algorithms must implement all ops (for now) */ + if (!bat_algo_ops->bat_iface_enable || + !bat_algo_ops->bat_iface_disable || + !bat_algo_ops->bat_iface_update_mac || + !bat_algo_ops->bat_primary_iface_set || + !bat_algo_ops->bat_neigh_cmp || + !bat_algo_ops->bat_neigh_is_similar_or_better) { + pr_info("Routing algo '%s' does not implement required ops\n", + bat_algo_ops->name); + return -EINVAL; + } + + INIT_HLIST_NODE(&bat_algo_ops->list); + hlist_add_head(&bat_algo_ops->list, &batadv_algo_list); + + return 0; +} + +int batadv_algo_select(struct batadv_priv *bat_priv, char *name) +{ + struct batadv_algo_ops *bat_algo_ops; + + bat_algo_ops = batadv_algo_get(name); + if (!bat_algo_ops) + return -EINVAL; + + bat_priv->bat_algo_ops = bat_algo_ops; + + return 0; +} + +int batadv_algo_seq_print_text(struct seq_file *seq, void *offset) +{ + struct batadv_algo_ops *bat_algo_ops; + + seq_puts(seq, "Available routing algorithms:\n"); + + hlist_for_each_entry(bat_algo_ops, &batadv_algo_list, list) { + seq_printf(seq, " * %s\n", bat_algo_ops->name); + } + + return 0; +} + +static int batadv_param_set_ra(const char *val, const struct kernel_param *kp) +{ + struct batadv_algo_ops *bat_algo_ops; + char *algo_name = (char *)val; + size_t name_len = strlen(algo_name); + + if (name_len > 0 && algo_name[name_len - 1] == '\n') + algo_name[name_len - 1] = '\0'; + + bat_algo_ops = batadv_algo_get(algo_name); + if (!bat_algo_ops) { + pr_err("Routing algorithm '%s' is not supported\n", algo_name); + return -EINVAL; + } + + return param_set_copystring(algo_name, kp); +} + +static const struct kernel_param_ops batadv_param_ops_ra = { + .set = batadv_param_set_ra, + .get = param_get_string, +}; + +static struct kparam_string batadv_param_string_ra = { + .maxlen = sizeof(batadv_routing_algo), + .string = batadv_routing_algo, +}; + +module_param_cb(routing_algo, &batadv_param_ops_ra, &batadv_param_string_ra, + 0644); diff --git a/net/batman-adv/bat_algo.h b/net/batman-adv/bat_algo.h index 3654296..8c7e761 100644 --- a/net/batman-adv/bat_algo.h +++ b/net/batman-adv/bat_algo.h @@ -20,8 +20,20 @@ #include "main.h" +#include + +struct seq_file; + int batadv_iv_init(void); +extern char batadv_routing_algo[]; +extern struct list_head batadv_hardif_list; + +void batadv_algo_init(void); +int batadv_algo_register(struct batadv_algo_ops *bat_algo_ops); +int batadv_algo_select(struct batadv_priv *bat_priv, char *name); +int batadv_algo_seq_print_text(struct seq_file *seq, void *offset); + #ifdef CONFIG_BATMAN_ADV_BATMAN_V int batadv_v_init(void); diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c index ca5a679..93e3d76 100644 --- a/net/batman-adv/bat_v_ogm.c +++ b/net/batman-adv/bat_v_ogm.c @@ -39,6 +39,7 @@ #include #include +#include "bat_algo.h" #include "hard-interface.h" #include "hash.h" #include "originator.h" diff --git a/net/batman-adv/debugfs.c b/net/batman-adv/debugfs.c index f187a8f..227c84b 100644 --- a/net/batman-adv/debugfs.c +++ b/net/batman-adv/debugfs.c @@ -44,6 +44,7 @@ #include #include +#include "bat_algo.h" #include "bridge_loop_avoidance.h" #include "distributed-arp-table.h" #include "gateway_client.h" diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index 225d63e..c5a7cab 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -32,7 +32,6 @@ #include #include #include -#include #include #include #include @@ -68,8 +67,6 @@ struct list_head batadv_hardif_list; static int (*batadv_rx_handler[256])(struct sk_buff *, struct batadv_hard_iface *); -char batadv_routing_algo[20] = "BATMAN_IV"; -static struct hlist_head batadv_algo_list; unsigned char batadv_broadcast_addr[] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; @@ -80,7 +77,7 @@ static void batadv_recv_handler_init(void); static int __init batadv_init(void) { INIT_LIST_HEAD(&batadv_hardif_list); - INIT_HLIST_HEAD(&batadv_algo_list); + batadv_algo_init(); batadv_recv_handler_init(); @@ -535,76 +532,6 @@ void batadv_recv_handler_unregister(u8 packet_type) batadv_rx_handler[packet_type] = batadv_recv_unhandled_packet; } -static struct batadv_algo_ops *batadv_algo_get(char *name) -{ - struct batadv_algo_ops *bat_algo_ops = NULL, *bat_algo_ops_tmp; - - hlist_for_each_entry(bat_algo_ops_tmp, &batadv_algo_list, list) { - if (strcmp(bat_algo_ops_tmp->name, name) != 0) - continue; - - bat_algo_ops = bat_algo_ops_tmp; - break; - } - - return bat_algo_ops; -} - -int batadv_algo_register(struct batadv_algo_ops *bat_algo_ops) -{ - struct batadv_algo_ops *bat_algo_ops_tmp; - - bat_algo_ops_tmp = batadv_algo_get(bat_algo_ops->name); - if (bat_algo_ops_tmp) { - pr_info("Trying to register already registered routing algorithm: %s\n", - bat_algo_ops->name); - return -EEXIST; - } - - /* all algorithms must implement all ops (for now) */ - if (!bat_algo_ops->bat_iface_enable || - !bat_algo_ops->bat_iface_disable || - !bat_algo_ops->bat_iface_update_mac || - !bat_algo_ops->bat_primary_iface_set || - !bat_algo_ops->bat_neigh_cmp || - !bat_algo_ops->bat_neigh_is_similar_or_better) { - pr_info("Routing algo '%s' does not implement required ops\n", - bat_algo_ops->name); - return -EINVAL; - } - - INIT_HLIST_NODE(&bat_algo_ops->list); - hlist_add_head(&bat_algo_ops->list, &batadv_algo_list); - - return 0; -} - -int batadv_algo_select(struct batadv_priv *bat_priv, char *name) -{ - struct batadv_algo_ops *bat_algo_ops; - - bat_algo_ops = batadv_algo_get(name); - if (!bat_algo_ops) - return -EINVAL; - - bat_priv->bat_algo_ops = bat_algo_ops; - - return 0; -} - -int batadv_algo_seq_print_text(struct seq_file *seq, void *offset) -{ - struct batadv_algo_ops *bat_algo_ops; - - seq_puts(seq, "Available routing algorithms:\n"); - - hlist_for_each_entry(bat_algo_ops, &batadv_algo_list, list) { - seq_printf(seq, " * %s\n", bat_algo_ops->name); - } - - return 0; -} - /** * batadv_skb_crc32 - calculate CRC32 of the whole packet and skip bytes in * the header @@ -691,36 +618,6 @@ bool batadv_vlan_ap_isola_get(struct batadv_priv *bat_priv, unsigned short vid) return ap_isolation_enabled; } -static int batadv_param_set_ra(const char *val, const struct kernel_param *kp) -{ - struct batadv_algo_ops *bat_algo_ops; - char *algo_name = (char *)val; - size_t name_len = strlen(algo_name); - - if (name_len > 0 && algo_name[name_len - 1] == '\n') - algo_name[name_len - 1] = '\0'; - - bat_algo_ops = batadv_algo_get(algo_name); - if (!bat_algo_ops) { - pr_err("Routing algorithm '%s' is not supported\n", algo_name); - return -EINVAL; - } - - return param_set_copystring(algo_name, kp); -} - -static const struct kernel_param_ops batadv_param_ops_ra = { - .set = batadv_param_set_ra, - .get = param_get_string, -}; - -static struct kparam_string batadv_param_string_ra = { - .maxlen = sizeof(batadv_routing_algo), - .string = batadv_routing_algo, -}; - -module_param_cb(routing_algo, &batadv_param_ops_ra, &batadv_param_string_ra, - 0644); module_init(batadv_init); module_exit(batadv_exit); diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 3af6582..3ec6285 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -196,7 +196,6 @@ struct sk_buff; #define BATADV_PRINT_VID(vid) ((vid & BATADV_VLAN_HAS_TAG) ? \ (int)(vid & VLAN_VID_MASK) : -1) -extern char batadv_routing_algo[]; extern struct list_head batadv_hardif_list; extern unsigned char batadv_broadcast_addr[]; @@ -217,9 +216,6 @@ batadv_recv_handler_register(u8 packet_type, int (*recv_handler)(struct sk_buff *, struct batadv_hard_iface *)); void batadv_recv_handler_unregister(u8 packet_type); -int batadv_algo_register(struct batadv_algo_ops *bat_algo_ops); -int batadv_algo_select(struct batadv_priv *bat_priv, char *name); -int batadv_algo_seq_print_text(struct seq_file *seq, void *offset); __be32 batadv_skb_crc32(struct sk_buff *skb, u8 *payload_ptr); /** diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index 076d258..592cbda 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -34,6 +34,7 @@ #include #include +#include "bat_algo.h" #include "distributed-arp-table.h" #include "fragmentation.h" #include "gateway_client.h" diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index 8cb459a..b9c7325 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -34,6 +34,7 @@ #include #include +#include "bat_algo.h" #include "bitarray.h" #include "bridge_loop_avoidance.h" #include "distributed-arp-table.h" diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index b60999d..f75631e 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -48,6 +48,7 @@ #include #include +#include "bat_algo.h" #include "bridge_loop_avoidance.h" #include "debugfs.h" #include "distributed-arp-table.h" diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 6c8d624..5c3cf7f 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -44,6 +44,7 @@ #include #include +#include "bat_algo.h" #include "bridge_loop_avoidance.h" #include "hard-interface.h" #include "hash.h" -- cgit v0.10.2 From 72f7b2deafde895012f93fa4827d4b1307a138e9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Linus=20L=C3=BCssing?= Date: Tue, 10 May 2016 18:41:26 +0200 Subject: batman-adv: Adding logging of mcast flag changes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With this patch changes relevant to a node's own multicast flags are printed to the 'mcast' log level. Tested-by: Simon Wunderlich Signed-off-by: Linus Lüssing Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index c356d91..cd83e28 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -231,6 +231,7 @@ __be32 batadv_skb_crc32(struct sk_buff *skb, u8 *payload_ptr); * @BATADV_DBG_BLA: bridge loop avoidance messages * @BATADV_DBG_DAT: ARP snooping and DAT related messages * @BATADV_DBG_NC: network coding related messages + * @BATADV_DBG_MCAST: multicast related messages * @BATADV_DBG_ALL: the union of all the above log levels */ enum batadv_dbg_level { @@ -240,7 +241,8 @@ enum batadv_dbg_level { BATADV_DBG_BLA = BIT(3), BATADV_DBG_DAT = BIT(4), BATADV_DBG_NC = BIT(5), - BATADV_DBG_ALL = 63, + BATADV_DBG_MCAST = BIT(6), + BATADV_DBG_ALL = 127, }; #ifdef CONFIG_BATMAN_ADV_DEBUG diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index eb30316..2d1a896 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include @@ -334,6 +335,122 @@ static bool batadv_mcast_has_bridge(struct batadv_priv *bat_priv) } /** + * batadv_mcast_querier_log - debug output regarding the querier status on link + * @bat_priv: the bat priv with all the soft interface information + * @str_proto: a string for the querier protocol (e.g. "IGMP" or "MLD") + * @old_state: the previous querier state on our link + * @new_state: the new querier state on our link + * + * Outputs debug messages to the logging facility with log level 'mcast' + * regarding changes to the querier status on the link which are relevant + * to our multicast optimizations. + * + * Usually this is about whether a querier appeared or vanished in + * our mesh or whether the querier is in the suboptimal position of being + * behind our local bridge segment: Snooping switches will directly + * forward listener reports to the querier, therefore batman-adv and + * the bridge will potentially not see these listeners - the querier is + * potentially shadowing listeners from us then. + * + * This is only interesting for nodes with a bridge on top of their + * soft interface. + */ +static void +batadv_mcast_querier_log(struct batadv_priv *bat_priv, char *str_proto, + struct batadv_mcast_querier_state *old_state, + struct batadv_mcast_querier_state *new_state) +{ + if (!old_state->exists && new_state->exists) + batadv_info(bat_priv->soft_iface, "%s Querier appeared\n", + str_proto); + else if (old_state->exists && !new_state->exists) + batadv_info(bat_priv->soft_iface, + "%s Querier disappeared - multicast optimizations disabled\n", + str_proto); + else if (!bat_priv->mcast.bridged && !new_state->exists) + batadv_info(bat_priv->soft_iface, + "No %s Querier present - multicast optimizations disabled\n", + str_proto); + + if (new_state->exists) { + if ((!old_state->shadowing && new_state->shadowing) || + (!old_state->exists && new_state->shadowing)) + batadv_dbg(BATADV_DBG_MCAST, bat_priv, + "%s Querier is behind our bridged segment: Might shadow listeners\n", + str_proto); + else if (old_state->shadowing && !new_state->shadowing) + batadv_dbg(BATADV_DBG_MCAST, bat_priv, + "%s Querier is not behind our bridged segment\n", + str_proto); + } +} + +/** + * batadv_mcast_bridge_log - debug output for topology changes in bridged setups + * @bat_priv: the bat priv with all the soft interface information + * @bridged: a flag about whether the soft interface is currently bridged or not + * @querier_ipv4: (maybe) new status of a potential, selected IGMP querier + * @querier_ipv6: (maybe) new status of a potential, selected MLD querier + * + * If no bridges are ever used on this node, then this function does nothing. + * + * Otherwise this function outputs debug information to the 'mcast' log level + * which might be relevant to our multicast optimizations. + * + * More precisely, it outputs information when a bridge interface is added or + * removed from a soft interface. And when a bridge is present, it further + * outputs information about the querier state which is relevant for the + * multicast flags this node is going to set. + */ +static void +batadv_mcast_bridge_log(struct batadv_priv *bat_priv, bool bridged, + struct batadv_mcast_querier_state *querier_ipv4, + struct batadv_mcast_querier_state *querier_ipv6) +{ + if (!bat_priv->mcast.bridged && bridged) + batadv_dbg(BATADV_DBG_MCAST, bat_priv, + "Bridge added: Setting Unsnoopables(U)-flag\n"); + else if (bat_priv->mcast.bridged && !bridged) + batadv_dbg(BATADV_DBG_MCAST, bat_priv, + "Bridge removed: Unsetting Unsnoopables(U)-flag\n"); + + if (bridged) { + batadv_mcast_querier_log(bat_priv, "IGMP", + &bat_priv->mcast.querier_ipv4, + querier_ipv4); + batadv_mcast_querier_log(bat_priv, "MLD", + &bat_priv->mcast.querier_ipv6, + querier_ipv6); + } +} + +/** + * batadv_mcast_flags_logs - output debug information about mcast flag changes + * @bat_priv: the bat priv with all the soft interface information + * @flags: flags indicating the new multicast state + * + * Whenever the multicast flags this nodes announces changes (@mcast_flags vs. + * bat_priv->mcast.flags), this notifies userspace via the 'mcast' log level. + */ +static void batadv_mcast_flags_log(struct batadv_priv *bat_priv, u8 flags) +{ + u8 old_flags = bat_priv->mcast.flags; + char str_old_flags[] = "[...]"; + + sprintf(str_old_flags, "[%c%c%c]", + (old_flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES) ? 'U' : '.', + (old_flags & BATADV_MCAST_WANT_ALL_IPV4) ? '4' : '.', + (old_flags & BATADV_MCAST_WANT_ALL_IPV6) ? '6' : '.'); + + batadv_dbg(BATADV_DBG_MCAST, bat_priv, + "Changing multicast flags from '%s' to '[%c%c%c]'\n", + bat_priv->mcast.enabled ? str_old_flags : "", + (flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES) ? 'U' : '.', + (flags & BATADV_MCAST_WANT_ALL_IPV4) ? '4' : '.', + (flags & BATADV_MCAST_WANT_ALL_IPV6) ? '6' : '.'); +} + +/** * batadv_mcast_mla_tvlv_update - update multicast tvlv * @bat_priv: the bat priv with all the soft interface information * @@ -349,12 +466,13 @@ static bool batadv_mcast_mla_tvlv_update(struct batadv_priv *bat_priv) struct batadv_mcast_querier_state querier4 = {false, false}; struct batadv_mcast_querier_state querier6 = {false, false}; struct net_device *dev = bat_priv->soft_iface; + bool bridged; mcast_data.flags = BATADV_NO_FLAGS; memset(mcast_data.reserved, 0, sizeof(mcast_data.reserved)); - bat_priv->mcast.bridged = batadv_mcast_has_bridge(bat_priv); - if (!bat_priv->mcast.bridged) + bridged = batadv_mcast_has_bridge(bat_priv); + if (!bridged) goto update; #if !IS_ENABLED(CONFIG_BRIDGE_IGMP_SNOOPING) @@ -385,8 +503,19 @@ static bool batadv_mcast_mla_tvlv_update(struct batadv_priv *bat_priv) mcast_data.flags |= BATADV_MCAST_WANT_ALL_IPV6; update: + batadv_mcast_bridge_log(bat_priv, bridged, &querier4, &querier6); + + bat_priv->mcast.querier_ipv4.exists = querier4.exists; + bat_priv->mcast.querier_ipv4.shadowing = querier4.shadowing; + + bat_priv->mcast.querier_ipv6.exists = querier6.exists; + bat_priv->mcast.querier_ipv6.shadowing = querier6.shadowing; + + bat_priv->mcast.bridged = bridged; + if (!bat_priv->mcast.enabled || mcast_data.flags != bat_priv->mcast.flags) { + batadv_mcast_flags_log(bat_priv, mcast_data.flags); batadv_tvlv_container_register(bat_priv, BATADV_TVLV_MCAST, 2, &mcast_data, sizeof(mcast_data)); bat_priv->mcast.flags = mcast_data.flags; diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index 81665b1..b60999d 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -808,6 +808,10 @@ static int batadv_softif_init_late(struct net_device *dev) atomic_set(&bat_priv->distributed_arp_table, 1); #endif #ifdef CONFIG_BATMAN_ADV_MCAST + bat_priv->mcast.querier_ipv4.exists = false; + bat_priv->mcast.querier_ipv4.shadowing = false; + bat_priv->mcast.querier_ipv6.exists = false; + bat_priv->mcast.querier_ipv6.shadowing = false; bat_priv->mcast.flags = BATADV_NO_FLAGS; atomic_set(&bat_priv->multicast_mode, 1); atomic_set(&bat_priv->mcast.num_disabled, 0); diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 83303c2..ab863a5 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -772,6 +772,8 @@ struct batadv_mcast_querier_state { * multicast traffic * @want_all_ipv4_list: a list of orig_nodes wanting all IPv4 multicast traffic * @want_all_ipv6_list: a list of orig_nodes wanting all IPv6 multicast traffic + * @querier_ipv4: the current state of an IGMP querier in the mesh + * @querier_ipv6: the current state of an MLD querier in the mesh * @flags: the flags we have last sent in our mcast tvlv * @enabled: whether the multicast tvlv is currently enabled * @bridged: whether the soft interface has a bridge on top @@ -787,6 +789,8 @@ struct batadv_priv_mcast { struct hlist_head want_all_unsnoopables_list; struct hlist_head want_all_ipv4_list; struct hlist_head want_all_ipv6_list; + struct batadv_mcast_querier_state querier_ipv4; + struct batadv_mcast_querier_state querier_ipv6; u8 flags; bool enabled; bool bridged; -- cgit v0.10.2 From ba412080fb6461b5a40dbc5e44186ed029d67b8d Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sun, 15 May 2016 23:48:31 +0200 Subject: batman-adv: Consolidate logging related functions There are several places in batman-adv which provide logging related functions. These should be grouped together in the log.* files to make them easier to find. Reported-by: Markus Pargmann Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile index 5260c17..a55f4ec 100644 --- a/net/batman-adv/Makefile +++ b/net/batman-adv/Makefile @@ -32,6 +32,7 @@ batman-adv-y += gateway_common.o batman-adv-y += hard-interface.o batman-adv-y += hash.o batman-adv-y += icmp_socket.o +batman-adv-$(CONFIG_BATMAN_ADV_DEBUG) += log.o batman-adv-y += main.o batman-adv-$(CONFIG_BATMAN_ADV_MCAST) += multicast.o batman-adv-$(CONFIG_BATMAN_ADV_NC) += network-coding.o diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index 948a5b4..805532a 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -52,6 +52,7 @@ #include "bitarray.h" #include "hard-interface.h" #include "hash.h" +#include "log.h" #include "network-coding.h" #include "originator.h" #include "packet.h" diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c index cf0262b..15cf272 100644 --- a/net/batman-adv/bat_v_elp.c +++ b/net/batman-adv/bat_v_elp.c @@ -43,6 +43,7 @@ #include "bat_algo.h" #include "bat_v_ogm.h" #include "hard-interface.h" +#include "log.h" #include "originator.h" #include "packet.h" #include "routing.h" diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c index 93e3d76..7ac9e0b 100644 --- a/net/batman-adv/bat_v_ogm.c +++ b/net/batman-adv/bat_v_ogm.c @@ -42,6 +42,7 @@ #include "bat_algo.h" #include "hard-interface.h" #include "hash.h" +#include "log.h" #include "originator.h" #include "packet.h" #include "routing.h" diff --git a/net/batman-adv/bitarray.c b/net/batman-adv/bitarray.c index a0c7913..0322714 100644 --- a/net/batman-adv/bitarray.c +++ b/net/batman-adv/bitarray.c @@ -20,6 +20,8 @@ #include +#include "log.h" + /* shift the packet array by n places. */ static void batadv_bitmap_shift_left(unsigned long *seq_bits, s32 n) { diff --git a/net/batman-adv/bridge_loop_avoidance.c b/net/batman-adv/bridge_loop_avoidance.c index 748a9ea..e4f7494 100644 --- a/net/batman-adv/bridge_loop_avoidance.c +++ b/net/batman-adv/bridge_loop_avoidance.c @@ -48,6 +48,7 @@ #include "hard-interface.h" #include "hash.h" +#include "log.h" #include "originator.h" #include "packet.h" #include "sysfs.h" diff --git a/net/batman-adv/debugfs.c b/net/batman-adv/debugfs.c index 227c84b..1d68b6e 100644 --- a/net/batman-adv/debugfs.c +++ b/net/batman-adv/debugfs.c @@ -18,37 +18,26 @@ #include "debugfs.h" #include "main.h" -#include #include #include #include #include -#include #include -#include -#include -#include #include -#include #include #include /* for linux/wait.h */ #include -#include -#include #include #include #include #include -#include -#include -#include -#include #include "bat_algo.h" #include "bridge_loop_avoidance.h" #include "distributed-arp-table.h" #include "gateway_client.h" #include "icmp_socket.h" +#include "log.h" #include "multicast.h" #include "network-coding.h" #include "originator.h" @@ -56,209 +45,6 @@ static struct dentry *batadv_debugfs; -#ifdef CONFIG_BATMAN_ADV_DEBUG -#define BATADV_LOG_BUFF_MASK (batadv_log_buff_len - 1) - -static const int batadv_log_buff_len = BATADV_LOG_BUF_LEN; - -static char *batadv_log_char_addr(struct batadv_priv_debug_log *debug_log, - size_t idx) -{ - return &debug_log->log_buff[idx & BATADV_LOG_BUFF_MASK]; -} - -static void batadv_emit_log_char(struct batadv_priv_debug_log *debug_log, - char c) -{ - char *char_addr; - - char_addr = batadv_log_char_addr(debug_log, debug_log->log_end); - *char_addr = c; - debug_log->log_end++; - - if (debug_log->log_end - debug_log->log_start > batadv_log_buff_len) - debug_log->log_start = debug_log->log_end - batadv_log_buff_len; -} - -__printf(2, 3) -static int batadv_fdebug_log(struct batadv_priv_debug_log *debug_log, - const char *fmt, ...) -{ - va_list args; - static char debug_log_buf[256]; - char *p; - - if (!debug_log) - return 0; - - spin_lock_bh(&debug_log->lock); - va_start(args, fmt); - vscnprintf(debug_log_buf, sizeof(debug_log_buf), fmt, args); - va_end(args); - - for (p = debug_log_buf; *p != 0; p++) - batadv_emit_log_char(debug_log, *p); - - spin_unlock_bh(&debug_log->lock); - - wake_up(&debug_log->queue_wait); - - return 0; -} - -int batadv_debug_log(struct batadv_priv *bat_priv, const char *fmt, ...) -{ - va_list args; - char tmp_log_buf[256]; - - va_start(args, fmt); - vscnprintf(tmp_log_buf, sizeof(tmp_log_buf), fmt, args); - batadv_fdebug_log(bat_priv->debug_log, "[%10u] %s", - jiffies_to_msecs(jiffies), tmp_log_buf); - va_end(args); - - return 0; -} - -static int batadv_log_open(struct inode *inode, struct file *file) -{ - if (!try_module_get(THIS_MODULE)) - return -EBUSY; - - nonseekable_open(inode, file); - file->private_data = inode->i_private; - return 0; -} - -static int batadv_log_release(struct inode *inode, struct file *file) -{ - module_put(THIS_MODULE); - return 0; -} - -static bool batadv_log_empty(struct batadv_priv_debug_log *debug_log) -{ - return !(debug_log->log_start - debug_log->log_end); -} - -static ssize_t batadv_log_read(struct file *file, char __user *buf, - size_t count, loff_t *ppos) -{ - struct batadv_priv *bat_priv = file->private_data; - struct batadv_priv_debug_log *debug_log = bat_priv->debug_log; - int error, i = 0; - char *char_addr; - char c; - - if ((file->f_flags & O_NONBLOCK) && batadv_log_empty(debug_log)) - return -EAGAIN; - - if (!buf) - return -EINVAL; - - if (count == 0) - return 0; - - if (!access_ok(VERIFY_WRITE, buf, count)) - return -EFAULT; - - error = wait_event_interruptible(debug_log->queue_wait, - (!batadv_log_empty(debug_log))); - - if (error) - return error; - - spin_lock_bh(&debug_log->lock); - - while ((!error) && (i < count) && - (debug_log->log_start != debug_log->log_end)) { - char_addr = batadv_log_char_addr(debug_log, - debug_log->log_start); - c = *char_addr; - - debug_log->log_start++; - - spin_unlock_bh(&debug_log->lock); - - error = __put_user(c, buf); - - spin_lock_bh(&debug_log->lock); - - buf++; - i++; - } - - spin_unlock_bh(&debug_log->lock); - - if (!error) - return i; - - return error; -} - -static unsigned int batadv_log_poll(struct file *file, poll_table *wait) -{ - struct batadv_priv *bat_priv = file->private_data; - struct batadv_priv_debug_log *debug_log = bat_priv->debug_log; - - poll_wait(file, &debug_log->queue_wait, wait); - - if (!batadv_log_empty(debug_log)) - return POLLIN | POLLRDNORM; - - return 0; -} - -static const struct file_operations batadv_log_fops = { - .open = batadv_log_open, - .release = batadv_log_release, - .read = batadv_log_read, - .poll = batadv_log_poll, - .llseek = no_llseek, -}; - -static int batadv_debug_log_setup(struct batadv_priv *bat_priv) -{ - struct dentry *d; - - if (!bat_priv->debug_dir) - goto err; - - bat_priv->debug_log = kzalloc(sizeof(*bat_priv->debug_log), GFP_ATOMIC); - if (!bat_priv->debug_log) - goto err; - - spin_lock_init(&bat_priv->debug_log->lock); - init_waitqueue_head(&bat_priv->debug_log->queue_wait); - - d = debugfs_create_file("log", S_IFREG | S_IRUSR, - bat_priv->debug_dir, bat_priv, - &batadv_log_fops); - if (!d) - goto err; - - return 0; - -err: - return -ENOMEM; -} - -static void batadv_debug_log_cleanup(struct batadv_priv *bat_priv) -{ - kfree(bat_priv->debug_log); - bat_priv->debug_log = NULL; -} -#else /* CONFIG_BATMAN_ADV_DEBUG */ -static int batadv_debug_log_setup(struct batadv_priv *bat_priv) -{ - return 0; -} - -static void batadv_debug_log_cleanup(struct batadv_priv *bat_priv) -{ -} -#endif - static int batadv_algorithms_open(struct inode *inode, struct file *file) { return single_open(file, batadv_algo_seq_print_text, NULL); diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c index 584b827..fa76465 100644 --- a/net/batman-adv/distributed-arp-table.c +++ b/net/batman-adv/distributed-arp-table.c @@ -45,6 +45,7 @@ #include "hard-interface.h" #include "hash.h" +#include "log.h" #include "originator.h" #include "send.h" #include "translation-table.h" diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c index 18c3715..63a805d 100644 --- a/net/batman-adv/gateway_client.c +++ b/net/batman-adv/gateway_client.c @@ -42,6 +42,7 @@ #include "gateway_common.h" #include "hard-interface.h" +#include "log.h" #include "originator.h" #include "packet.h" #include "routing.h" diff --git a/net/batman-adv/gateway_common.c b/net/batman-adv/gateway_common.c index 6a6f2d4..d7bc6a8 100644 --- a/net/batman-adv/gateway_common.c +++ b/net/batman-adv/gateway_common.c @@ -28,6 +28,7 @@ #include #include "gateway_client.h" +#include "log.h" #include "packet.h" #include "tvlv.h" diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index a3483f6..ad2c37c 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -42,6 +42,7 @@ #include "debugfs.h" #include "distributed-arp-table.h" #include "gateway_client.h" +#include "log.h" #include "originator.h" #include "packet.h" #include "send.h" diff --git a/net/batman-adv/icmp_socket.c b/net/batman-adv/icmp_socket.c index 777aea1..378cc11 100644 --- a/net/batman-adv/icmp_socket.c +++ b/net/batman-adv/icmp_socket.c @@ -45,6 +45,7 @@ #include #include "hard-interface.h" +#include "log.h" #include "originator.h" #include "packet.h" #include "send.h" diff --git a/net/batman-adv/log.c b/net/batman-adv/log.c new file mode 100644 index 0000000..56dc532 --- /dev/null +++ b/net/batman-adv/log.c @@ -0,0 +1,231 @@ +/* Copyright (C) 2010-2016 B.A.T.M.A.N. contributors: + * + * Marek Lindner + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#include "log.h" +#include "main.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include /* for linux/wait.h */ +#include +#include +#include +#include +#include +#include +#include +#include + +#define BATADV_LOG_BUFF_MASK (batadv_log_buff_len - 1) + +static const int batadv_log_buff_len = BATADV_LOG_BUF_LEN; + +static char *batadv_log_char_addr(struct batadv_priv_debug_log *debug_log, + size_t idx) +{ + return &debug_log->log_buff[idx & BATADV_LOG_BUFF_MASK]; +} + +static void batadv_emit_log_char(struct batadv_priv_debug_log *debug_log, + char c) +{ + char *char_addr; + + char_addr = batadv_log_char_addr(debug_log, debug_log->log_end); + *char_addr = c; + debug_log->log_end++; + + if (debug_log->log_end - debug_log->log_start > batadv_log_buff_len) + debug_log->log_start = debug_log->log_end - batadv_log_buff_len; +} + +__printf(2, 3) +static int batadv_fdebug_log(struct batadv_priv_debug_log *debug_log, + const char *fmt, ...) +{ + va_list args; + static char debug_log_buf[256]; + char *p; + + if (!debug_log) + return 0; + + spin_lock_bh(&debug_log->lock); + va_start(args, fmt); + vscnprintf(debug_log_buf, sizeof(debug_log_buf), fmt, args); + va_end(args); + + for (p = debug_log_buf; *p != 0; p++) + batadv_emit_log_char(debug_log, *p); + + spin_unlock_bh(&debug_log->lock); + + wake_up(&debug_log->queue_wait); + + return 0; +} + +int batadv_debug_log(struct batadv_priv *bat_priv, const char *fmt, ...) +{ + va_list args; + char tmp_log_buf[256]; + + va_start(args, fmt); + vscnprintf(tmp_log_buf, sizeof(tmp_log_buf), fmt, args); + batadv_fdebug_log(bat_priv->debug_log, "[%10u] %s", + jiffies_to_msecs(jiffies), tmp_log_buf); + va_end(args); + + return 0; +} + +static int batadv_log_open(struct inode *inode, struct file *file) +{ + if (!try_module_get(THIS_MODULE)) + return -EBUSY; + + nonseekable_open(inode, file); + file->private_data = inode->i_private; + return 0; +} + +static int batadv_log_release(struct inode *inode, struct file *file) +{ + module_put(THIS_MODULE); + return 0; +} + +static bool batadv_log_empty(struct batadv_priv_debug_log *debug_log) +{ + return !(debug_log->log_start - debug_log->log_end); +} + +static ssize_t batadv_log_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + struct batadv_priv *bat_priv = file->private_data; + struct batadv_priv_debug_log *debug_log = bat_priv->debug_log; + int error, i = 0; + char *char_addr; + char c; + + if ((file->f_flags & O_NONBLOCK) && batadv_log_empty(debug_log)) + return -EAGAIN; + + if (!buf) + return -EINVAL; + + if (count == 0) + return 0; + + if (!access_ok(VERIFY_WRITE, buf, count)) + return -EFAULT; + + error = wait_event_interruptible(debug_log->queue_wait, + (!batadv_log_empty(debug_log))); + + if (error) + return error; + + spin_lock_bh(&debug_log->lock); + + while ((!error) && (i < count) && + (debug_log->log_start != debug_log->log_end)) { + char_addr = batadv_log_char_addr(debug_log, + debug_log->log_start); + c = *char_addr; + + debug_log->log_start++; + + spin_unlock_bh(&debug_log->lock); + + error = __put_user(c, buf); + + spin_lock_bh(&debug_log->lock); + + buf++; + i++; + } + + spin_unlock_bh(&debug_log->lock); + + if (!error) + return i; + + return error; +} + +static unsigned int batadv_log_poll(struct file *file, poll_table *wait) +{ + struct batadv_priv *bat_priv = file->private_data; + struct batadv_priv_debug_log *debug_log = bat_priv->debug_log; + + poll_wait(file, &debug_log->queue_wait, wait); + + if (!batadv_log_empty(debug_log)) + return POLLIN | POLLRDNORM; + + return 0; +} + +static const struct file_operations batadv_log_fops = { + .open = batadv_log_open, + .release = batadv_log_release, + .read = batadv_log_read, + .poll = batadv_log_poll, + .llseek = no_llseek, +}; + +int batadv_debug_log_setup(struct batadv_priv *bat_priv) +{ + struct dentry *d; + + if (!bat_priv->debug_dir) + goto err; + + bat_priv->debug_log = kzalloc(sizeof(*bat_priv->debug_log), GFP_ATOMIC); + if (!bat_priv->debug_log) + goto err; + + spin_lock_init(&bat_priv->debug_log->lock); + init_waitqueue_head(&bat_priv->debug_log->queue_wait); + + d = debugfs_create_file("log", S_IFREG | S_IRUSR, + bat_priv->debug_dir, bat_priv, + &batadv_log_fops); + if (!d) + goto err; + + return 0; + +err: + return -ENOMEM; +} + +void batadv_debug_log_cleanup(struct batadv_priv *bat_priv) +{ + kfree(bat_priv->debug_log); + bat_priv->debug_log = NULL; +} diff --git a/net/batman-adv/log.h b/net/batman-adv/log.h new file mode 100644 index 0000000..9948e56 --- /dev/null +++ b/net/batman-adv/log.h @@ -0,0 +1,109 @@ +/* Copyright (C) 2007-2016 B.A.T.M.A.N. contributors: + * + * Marek Lindner, Simon Wunderlich + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#ifndef _NET_BATMAN_ADV_LOG_H_ +#define _NET_BATMAN_ADV_LOG_H_ + +#include "main.h" + +#include +#include +#include + +#ifdef CONFIG_BATMAN_ADV_DEBUG + +int batadv_debug_log_setup(struct batadv_priv *bat_priv); +void batadv_debug_log_cleanup(struct batadv_priv *bat_priv); + +#else + +static inline int batadv_debug_log_setup(struct batadv_priv *bat_priv) +{ + return 0; +} + +static inline void batadv_debug_log_cleanup(struct batadv_priv *bat_priv) +{ +} + +#endif + +/** + * enum batadv_dbg_level - available log levels + * @BATADV_DBG_BATMAN: OGM and TQ computations related messages + * @BATADV_DBG_ROUTES: route added / changed / deleted + * @BATADV_DBG_TT: translation table messages + * @BATADV_DBG_BLA: bridge loop avoidance messages + * @BATADV_DBG_DAT: ARP snooping and DAT related messages + * @BATADV_DBG_NC: network coding related messages + * @BATADV_DBG_MCAST: multicast related messages + * @BATADV_DBG_ALL: the union of all the above log levels + */ +enum batadv_dbg_level { + BATADV_DBG_BATMAN = BIT(0), + BATADV_DBG_ROUTES = BIT(1), + BATADV_DBG_TT = BIT(2), + BATADV_DBG_BLA = BIT(3), + BATADV_DBG_DAT = BIT(4), + BATADV_DBG_NC = BIT(5), + BATADV_DBG_MCAST = BIT(6), + BATADV_DBG_ALL = 127, +}; + +#ifdef CONFIG_BATMAN_ADV_DEBUG +int batadv_debug_log(struct batadv_priv *bat_priv, const char *fmt, ...) +__printf(2, 3); + +/* possibly ratelimited debug output */ +#define _batadv_dbg(type, bat_priv, ratelimited, fmt, arg...) \ + do { \ + if (atomic_read(&bat_priv->log_level) & type && \ + (!ratelimited || net_ratelimit())) \ + batadv_debug_log(bat_priv, fmt, ## arg);\ + } \ + while (0) +#else /* !CONFIG_BATMAN_ADV_DEBUG */ +__printf(4, 5) +static inline void _batadv_dbg(int type __always_unused, + struct batadv_priv *bat_priv __always_unused, + int ratelimited __always_unused, + const char *fmt __always_unused, ...) +{ +} +#endif + +#define batadv_dbg(type, bat_priv, arg...) \ + _batadv_dbg(type, bat_priv, 0, ## arg) +#define batadv_dbg_ratelimited(type, bat_priv, arg...) \ + _batadv_dbg(type, bat_priv, 1, ## arg) + +#define batadv_info(net_dev, fmt, arg...) \ + do { \ + struct net_device *_netdev = (net_dev); \ + struct batadv_priv *_batpriv = netdev_priv(_netdev); \ + batadv_dbg(BATADV_DBG_ALL, _batpriv, fmt, ## arg); \ + pr_info("%s: " fmt, _netdev->name, ## arg); \ + } while (0) +#define batadv_err(net_dev, fmt, arg...) \ + do { \ + struct net_device *_netdev = (net_dev); \ + struct batadv_priv *_batpriv = netdev_priv(_netdev); \ + batadv_dbg(BATADV_DBG_ALL, _batpriv, fmt, ## arg); \ + pr_err("%s: " fmt, _netdev->name, ## arg); \ + } while (0) + +#endif /* _NET_BATMAN_ADV_LOG_H_ */ diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index c5a7cab..05e559c 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include @@ -52,6 +53,7 @@ #include "gateway_common.h" #include "hard-interface.h" #include "icmp_socket.h" +#include "log.h" #include "multicast.h" #include "network-coding.h" #include "originator.h" diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 3ec6285..857fb5a 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -175,7 +175,6 @@ enum batadv_uev_type { /* Kernel headers */ -#include #include /* for packet.h */ #include #include @@ -183,13 +182,13 @@ enum batadv_uev_type { #include /* for packet.h */ #include #include -#include #include -#include #include #include "types.h" +struct net_device; +struct packet_type; struct seq_file; struct sk_buff; @@ -219,70 +218,6 @@ void batadv_recv_handler_unregister(u8 packet_type); __be32 batadv_skb_crc32(struct sk_buff *skb, u8 *payload_ptr); /** - * enum batadv_dbg_level - available log levels - * @BATADV_DBG_BATMAN: OGM and TQ computations related messages - * @BATADV_DBG_ROUTES: route added / changed / deleted - * @BATADV_DBG_TT: translation table messages - * @BATADV_DBG_BLA: bridge loop avoidance messages - * @BATADV_DBG_DAT: ARP snooping and DAT related messages - * @BATADV_DBG_NC: network coding related messages - * @BATADV_DBG_MCAST: multicast related messages - * @BATADV_DBG_ALL: the union of all the above log levels - */ -enum batadv_dbg_level { - BATADV_DBG_BATMAN = BIT(0), - BATADV_DBG_ROUTES = BIT(1), - BATADV_DBG_TT = BIT(2), - BATADV_DBG_BLA = BIT(3), - BATADV_DBG_DAT = BIT(4), - BATADV_DBG_NC = BIT(5), - BATADV_DBG_MCAST = BIT(6), - BATADV_DBG_ALL = 127, -}; - -#ifdef CONFIG_BATMAN_ADV_DEBUG -int batadv_debug_log(struct batadv_priv *bat_priv, const char *fmt, ...) -__printf(2, 3); - -/* possibly ratelimited debug output */ -#define _batadv_dbg(type, bat_priv, ratelimited, fmt, arg...) \ - do { \ - if (atomic_read(&bat_priv->log_level) & type && \ - (!ratelimited || net_ratelimit())) \ - batadv_debug_log(bat_priv, fmt, ## arg);\ - } \ - while (0) -#else /* !CONFIG_BATMAN_ADV_DEBUG */ -__printf(4, 5) -static inline void _batadv_dbg(int type __always_unused, - struct batadv_priv *bat_priv __always_unused, - int ratelimited __always_unused, - const char *fmt __always_unused, ...) -{ -} -#endif - -#define batadv_dbg(type, bat_priv, arg...) \ - _batadv_dbg(type, bat_priv, 0, ## arg) -#define batadv_dbg_ratelimited(type, bat_priv, arg...) \ - _batadv_dbg(type, bat_priv, 1, ## arg) - -#define batadv_info(net_dev, fmt, arg...) \ - do { \ - struct net_device *_netdev = (net_dev); \ - struct batadv_priv *_batpriv = netdev_priv(_netdev); \ - batadv_dbg(BATADV_DBG_ALL, _batpriv, fmt, ## arg); \ - pr_info("%s: " fmt, _netdev->name, ## arg); \ - } while (0) -#define batadv_err(net_dev, fmt, arg...) \ - do { \ - struct net_device *_netdev = (net_dev); \ - struct batadv_priv *_batpriv = netdev_priv(_netdev); \ - batadv_dbg(BATADV_DBG_ALL, _batpriv, fmt, ## arg); \ - pr_err("%s: " fmt, _netdev->name, ## arg); \ - } while (0) - -/** * batadv_compare_eth - Compare two not u16 aligned Ethernet addresses * @data1: Pointer to a six-byte array containing the Ethernet address * @data2: Pointer other six-byte array containing the Ethernet address diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index 0e7d78f..cc91507 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -55,6 +55,7 @@ #include "hard-interface.h" #include "hash.h" +#include "log.h" #include "packet.h" #include "translation-table.h" #include "tvlv.h" diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c index d0383de..293ef4f 100644 --- a/net/batman-adv/network-coding.c +++ b/net/batman-adv/network-coding.c @@ -51,6 +51,7 @@ #include "hard-interface.h" #include "hash.h" +#include "log.h" #include "originator.h" #include "packet.h" #include "routing.h" diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index 592cbda..8ad17ad 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -40,6 +40,7 @@ #include "gateway_client.h" #include "hard-interface.h" #include "hash.h" +#include "log.h" #include "multicast.h" #include "network-coding.h" #include "routing.h" diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index b9c7325..a5b53a3 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -41,6 +41,7 @@ #include "fragmentation.h" #include "hard-interface.h" #include "icmp_socket.h" +#include "log.h" #include "network-coding.h" #include "originator.h" #include "packet.h" diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c index 4e49454..3a59df2 100644 --- a/net/batman-adv/send.c +++ b/net/batman-adv/send.c @@ -42,6 +42,7 @@ #include "fragmentation.h" #include "gateway_client.h" #include "hard-interface.h" +#include "log.h" #include "network-coding.h" #include "originator.h" #include "routing.h" diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c index 6244a9a..1a7942d 100644 --- a/net/batman-adv/sysfs.c +++ b/net/batman-adv/sysfs.c @@ -43,6 +43,7 @@ #include "gateway_client.h" #include "gateway_common.h" #include "hard-interface.h" +#include "log.h" #include "network-coding.h" #include "packet.h" #include "soft-interface.h" diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 5c3cf7f..53458d6 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -48,6 +48,7 @@ #include "bridge_loop_avoidance.h" #include "hard-interface.h" #include "hash.h" +#include "log.h" #include "multicast.h" #include "originator.h" #include "packet.h" -- cgit v0.10.2 From 4e3e823b5a503235630921287f130e1d8d22d200 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Linus=20L=C3=BCssing?= Date: Tue, 10 May 2016 18:41:27 +0200 Subject: batman-adv: Add debugfs table for mcast flags MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds a debugfs table with originators and their according multicast flags to help users figure out why multicast optimizations might be enabled or disabled for them. Tested-by: Simon Wunderlich Signed-off-by: Linus Lüssing Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/debugfs.c b/net/batman-adv/debugfs.c index 9529004..f187a8f 100644 --- a/net/batman-adv/debugfs.c +++ b/net/batman-adv/debugfs.c @@ -48,6 +48,7 @@ #include "distributed-arp-table.h" #include "gateway_client.h" #include "icmp_socket.h" +#include "multicast.h" #include "network-coding.h" #include "originator.h" #include "translation-table.h" @@ -363,6 +364,22 @@ static int batadv_nc_nodes_open(struct inode *inode, struct file *file) } #endif +#ifdef CONFIG_BATMAN_ADV_MCAST +/** + * batadv_mcast_flags_open - prepare file handler for reads from mcast_flags + * @inode: inode which was opened + * @file: file handle to be initialized + * + * Return: 0 on success or negative error number in case of failure + */ +static int batadv_mcast_flags_open(struct inode *inode, struct file *file) +{ + struct net_device *net_dev = (struct net_device *)inode->i_private; + + return single_open(file, batadv_mcast_flags_seq_print_text, net_dev); +} +#endif + #define BATADV_DEBUGINFO(_name, _mode, _open) \ struct batadv_debuginfo batadv_debuginfo_##_name = { \ .attr = { \ @@ -407,6 +424,9 @@ static BATADV_DEBUGINFO(transtable_local, S_IRUGO, #ifdef CONFIG_BATMAN_ADV_NC static BATADV_DEBUGINFO(nc_nodes, S_IRUGO, batadv_nc_nodes_open); #endif +#ifdef CONFIG_BATMAN_ADV_MCAST +static BATADV_DEBUGINFO(mcast_flags, S_IRUGO, batadv_mcast_flags_open); +#endif static struct batadv_debuginfo *batadv_mesh_debuginfos[] = { &batadv_debuginfo_neighbors, @@ -424,6 +444,9 @@ static struct batadv_debuginfo *batadv_mesh_debuginfos[] = { #ifdef CONFIG_BATMAN_ADV_NC &batadv_debuginfo_nc_nodes, #endif +#ifdef CONFIG_BATMAN_ADV_MCAST + &batadv_debuginfo_mcast_flags, +#endif NULL, }; diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index 2d1a896..d3222db 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -41,6 +41,7 @@ #include #include #include +#include #include #include #include @@ -52,6 +53,8 @@ #include #include +#include "hard-interface.h" +#include "hash.h" #include "packet.h" #include "translation-table.h" @@ -1130,6 +1133,107 @@ void batadv_mcast_init(struct batadv_priv *bat_priv) } /** + * batadv_mcast_flags_print_header - print own mcast flags to debugfs table + * @bat_priv: the bat priv with all the soft interface information + * @seq: debugfs table seq_file struct + * + * Prints our own multicast flags including a more specific reason why + * they are set, that is prints the bridge and querier state too, to + * the debugfs table specified via @seq. + */ +static void batadv_mcast_flags_print_header(struct batadv_priv *bat_priv, + struct seq_file *seq) +{ + u8 flags = bat_priv->mcast.flags; + char querier4, querier6, shadowing4, shadowing6; + bool bridged = bat_priv->mcast.bridged; + + if (bridged) { + querier4 = bat_priv->mcast.querier_ipv4.exists ? '.' : '4'; + querier6 = bat_priv->mcast.querier_ipv6.exists ? '.' : '6'; + shadowing4 = bat_priv->mcast.querier_ipv4.shadowing ? '4' : '.'; + shadowing6 = bat_priv->mcast.querier_ipv6.shadowing ? '6' : '.'; + } else { + querier4 = '?'; + querier6 = '?'; + shadowing4 = '?'; + shadowing6 = '?'; + } + + seq_printf(seq, "Multicast flags (own flags: [%c%c%c])\n", + (flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES) ? 'U' : '.', + (flags & BATADV_MCAST_WANT_ALL_IPV4) ? '4' : '.', + (flags & BATADV_MCAST_WANT_ALL_IPV6) ? '6' : '.'); + seq_printf(seq, "* Bridged [U]\t\t\t\t%c\n", bridged ? 'U' : '.'); + seq_printf(seq, "* No IGMP/MLD Querier [4/6]:\t\t%c/%c\n", + querier4, querier6); + seq_printf(seq, "* Shadowing IGMP/MLD Querier [4/6]:\t%c/%c\n", + shadowing4, shadowing6); + seq_puts(seq, "-------------------------------------------\n"); + seq_printf(seq, " %-10s %s\n", "Originator", "Flags"); +} + +/** + * batadv_mcast_flags_seq_print_text - print the mcast flags of other nodes + * @seq: seq file to print on + * @offset: not used + * + * This prints a table of (primary) originators and their according + * multicast flags, including (in the header) our own. + * + * Return: always 0 + */ +int batadv_mcast_flags_seq_print_text(struct seq_file *seq, void *offset) +{ + struct net_device *net_dev = (struct net_device *)seq->private; + struct batadv_priv *bat_priv = netdev_priv(net_dev); + struct batadv_hard_iface *primary_if; + struct batadv_hashtable *hash = bat_priv->orig_hash; + struct batadv_orig_node *orig_node; + struct hlist_head *head; + u8 flags; + u32 i; + + primary_if = batadv_seq_print_text_primary_if_get(seq); + if (!primary_if) + return 0; + + batadv_mcast_flags_print_header(bat_priv, seq); + + for (i = 0; i < hash->size; i++) { + head = &hash->table[i]; + + rcu_read_lock(); + hlist_for_each_entry_rcu(orig_node, head, hash_entry) { + if (!test_bit(BATADV_ORIG_CAPA_HAS_MCAST, + &orig_node->capa_initialized)) + continue; + + if (!test_bit(BATADV_ORIG_CAPA_HAS_MCAST, + &orig_node->capabilities)) { + seq_printf(seq, "%pM -\n", orig_node->orig); + continue; + } + + flags = orig_node->mcast_flags; + + seq_printf(seq, "%pM [%c%c%c]\n", orig_node->orig, + (flags & BATADV_MCAST_WANT_ALL_UNSNOOPABLES) + ? 'U' : '.', + (flags & BATADV_MCAST_WANT_ALL_IPV4) + ? '4' : '.', + (flags & BATADV_MCAST_WANT_ALL_IPV6) + ? '6' : '.'); + } + rcu_read_unlock(); + } + + batadv_hardif_put(primary_if); + + return 0; +} + +/** * batadv_mcast_free - free the multicast optimizations structures * @bat_priv: the bat priv with all the soft interface information */ diff --git a/net/batman-adv/multicast.h b/net/batman-adv/multicast.h index 80bceec..1fb00ba 100644 --- a/net/batman-adv/multicast.h +++ b/net/batman-adv/multicast.h @@ -20,6 +20,7 @@ #include "main.h" +struct seq_file; struct sk_buff; /** @@ -46,6 +47,8 @@ batadv_mcast_forw_mode(struct batadv_priv *bat_priv, struct sk_buff *skb, void batadv_mcast_init(struct batadv_priv *bat_priv); +int batadv_mcast_flags_seq_print_text(struct seq_file *seq, void *offset); + void batadv_mcast_free(struct batadv_priv *bat_priv); void batadv_mcast_purge_orig(struct batadv_orig_node *orig_node); -- cgit v0.10.2 From a2d0816608df1ca69fcdbb9135a2b6df0c65d954 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Sun, 15 May 2016 11:07:46 +0200 Subject: batman-adv: Fix bat_(iv|v) function declaration header The bat_algo.h had some functions declared which were not part of the bat_algo.c file. These are instead stored in bat_v.c and bat_iv_ogm.c. The declaration should therefore be also in bat_v.h and bat_iv_ogm,h to make them easier to find. Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/bat_algo.h b/net/batman-adv/bat_algo.h index 8c7e761..860d773 100644 --- a/net/batman-adv/bat_algo.h +++ b/net/batman-adv/bat_algo.h @@ -24,8 +24,6 @@ struct seq_file; -int batadv_iv_init(void); - extern char batadv_routing_algo[]; extern struct list_head batadv_hardif_list; @@ -34,33 +32,4 @@ int batadv_algo_register(struct batadv_algo_ops *bat_algo_ops); int batadv_algo_select(struct batadv_priv *bat_priv, char *name); int batadv_algo_seq_print_text(struct seq_file *seq, void *offset); -#ifdef CONFIG_BATMAN_ADV_BATMAN_V - -int batadv_v_init(void); -void batadv_v_hardif_init(struct batadv_hard_iface *hardif); -int batadv_v_mesh_init(struct batadv_priv *bat_priv); -void batadv_v_mesh_free(struct batadv_priv *bat_priv); - -#else - -static inline int batadv_v_init(void) -{ - return 0; -} - -static inline void batadv_v_hardif_init(struct batadv_hard_iface *hardif) -{ -} - -static inline int batadv_v_mesh_init(struct batadv_priv *bat_priv) -{ - return 0; -} - -static inline void batadv_v_mesh_free(struct batadv_priv *bat_priv) -{ -} - -#endif /* CONFIG_BATMAN_ADV_BATMAN_V */ - #endif /* _NET_BATMAN_ADV_BAT_ALGO_H_ */ diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index 805532a..e2d8848 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -15,7 +15,7 @@ * along with this program; if not, see . */ -#include "bat_algo.h" +#include "bat_iv_ogm.h" #include "main.h" #include @@ -49,6 +49,7 @@ #include #include +#include "bat_algo.h" #include "bitarray.h" #include "hard-interface.h" #include "hash.h" diff --git a/net/batman-adv/bat_iv_ogm.h b/net/batman-adv/bat_iv_ogm.h new file mode 100644 index 0000000..b9f3550 --- /dev/null +++ b/net/batman-adv/bat_iv_ogm.h @@ -0,0 +1,25 @@ +/* Copyright (C) 2007-2016 B.A.T.M.A.N. contributors: + * + * Marek Lindner, Simon Wunderlich + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#ifndef _BATMAN_ADV_BATADV_IV_OGM_H_ +#define _BATMAN_ADV_BATADV_IV_OGM_H_ + +#include "main.h" + +int batadv_iv_init(void); + +#endif /* _BATMAN_ADV_BATADV_IV_OGM_H_ */ diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c index c2fea81..7231440 100644 --- a/net/batman-adv/bat_v.c +++ b/net/batman-adv/bat_v.c @@ -15,7 +15,7 @@ * along with this program; if not, see . */ -#include "bat_algo.h" +#include "bat_v.h" #include "main.h" #include @@ -31,6 +31,7 @@ #include #include +#include "bat_algo.h" #include "bat_v_elp.h" #include "bat_v_ogm.h" #include "hard-interface.h" diff --git a/net/batman-adv/bat_v.h b/net/batman-adv/bat_v.h new file mode 100644 index 0000000..83b7763 --- /dev/null +++ b/net/batman-adv/bat_v.h @@ -0,0 +1,52 @@ +/* Copyright (C) 2011-2016 B.A.T.M.A.N. contributors: + * + * Marek Lindner, Linus Lüssing + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#ifndef _NET_BATMAN_ADV_BAT_V_H_ +#define _NET_BATMAN_ADV_BAT_V_H_ + +#include "main.h" + +#ifdef CONFIG_BATMAN_ADV_BATMAN_V + +int batadv_v_init(void); +void batadv_v_hardif_init(struct batadv_hard_iface *hardif); +int batadv_v_mesh_init(struct batadv_priv *bat_priv); +void batadv_v_mesh_free(struct batadv_priv *bat_priv); + +#else + +static inline int batadv_v_init(void) +{ + return 0; +} + +static inline void batadv_v_hardif_init(struct batadv_hard_iface *hardif) +{ +} + +static inline int batadv_v_mesh_init(struct batadv_priv *bat_priv) +{ + return 0; +} + +static inline void batadv_v_mesh_free(struct batadv_priv *bat_priv) +{ +} + +#endif /* CONFIG_BATMAN_ADV_BATMAN_V */ + +#endif /* _NET_BATMAN_ADV_BAT_V_H_ */ diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index ad2c37c..70841c1 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -37,7 +37,7 @@ #include #include -#include "bat_algo.h" +#include "bat_v.h" #include "bridge_loop_avoidance.h" #include "debugfs.h" #include "distributed-arp-table.h" diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index 05e559c..eab9d1b 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -46,6 +46,8 @@ #include #include "bat_algo.h" +#include "bat_iv_ogm.h" +#include "bat_v.h" #include "bridge_loop_avoidance.h" #include "debugfs.h" #include "distributed-arp-table.h" diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index a5b53a3..5833ab3 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -34,7 +34,6 @@ #include #include -#include "bat_algo.h" #include "bitarray.h" #include "bridge_loop_avoidance.h" #include "distributed-arp-table.h" diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 53458d6..48ce788 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -44,7 +44,6 @@ #include #include -#include "bat_algo.h" #include "bridge_loop_avoidance.h" #include "hard-interface.h" #include "hash.h" -- cgit v0.10.2 From 1ca1cc98bf7418c680415bfce05699f67510a7fd Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 28 Jun 2016 12:18:23 +0200 Subject: bpf: minor cleanups on fd maps and helpers Some minor cleanups: i) Remove the unlikely() from fd array map lookups and let the CPU branch predictor do its job, scenarios where there is not always a map entry are very well valid. ii) Move the attribute type check in the bpf_perf_event_read() helper a bit earlier so it's consistent wrt checks with bpf_perf_event_output() helper as well. iii) remove some comments that are self-documenting in kprobe_prog_is_valid_access() and therefore make it consistent to tp_prog_is_valid_access() as well. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index b94a365..d638062 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -719,14 +719,13 @@ select_insn: if (unlikely(index >= array->map.max_entries)) goto out; - if (unlikely(tail_call_cnt > MAX_TAIL_CALL_CNT)) goto out; tail_call_cnt++; prog = READ_ONCE(array->ptrs[index]); - if (unlikely(!prog)) + if (!prog) goto out; /* ARG1 at this point is guaranteed to point to CTX from diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 3de25fb..4e61f74 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -199,19 +199,19 @@ static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5) return -E2BIG; ee = READ_ONCE(array->ptrs[index]); - if (unlikely(!ee)) + if (!ee) return -ENOENT; event = ee->event; + if (unlikely(event->attr.type != PERF_TYPE_HARDWARE && + event->attr.type != PERF_TYPE_RAW)) + return -EINVAL; + /* make sure event is local and doesn't have pmu::count */ if (event->oncpu != smp_processor_id() || event->pmu->count) return -EINVAL; - if (unlikely(event->attr.type != PERF_TYPE_HARDWARE && - event->attr.type != PERF_TYPE_RAW)) - return -EINVAL; - /* * we don't know if the function is run successfully by the * return value. It can be judged in other places, such as @@ -251,7 +251,7 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) return -E2BIG; ee = READ_ONCE(array->ptrs[index]); - if (unlikely(!ee)) + if (!ee) return -ENOENT; event = ee->event; @@ -354,18 +354,12 @@ static const struct bpf_func_proto *kprobe_prog_func_proto(enum bpf_func_id func static bool kprobe_prog_is_valid_access(int off, int size, enum bpf_access_type type, enum bpf_reg_type *reg_type) { - /* check bounds */ if (off < 0 || off >= sizeof(struct pt_regs)) return false; - - /* only read is allowed */ if (type != BPF_READ) return false; - - /* disallow misaligned access */ if (off % size != 0) return false; - return true; } -- cgit v0.10.2 From d79313303181d357d293453fb8486bdc87bfd2f4 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 28 Jun 2016 12:18:24 +0200 Subject: bpf, trace: fetch current cpu only once We currently have two invocations, which is unnecessary. Fetch it only once and use the smp_processor_id() variant, so we also get preemption checks along with it when DEBUG_PREEMPT is set. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 4e61f74..505f9e9 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -233,6 +233,7 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) struct pt_regs *regs = (struct pt_regs *) (long) r1; struct bpf_map *map = (struct bpf_map *) (long) r2; struct bpf_array *array = container_of(map, struct bpf_array, map); + unsigned int cpu = smp_processor_id(); u64 index = flags & BPF_F_INDEX_MASK; void *data = (void *) (long) r4; struct perf_sample_data sample_data; @@ -246,7 +247,7 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) if (unlikely(flags & ~(BPF_F_INDEX_MASK))) return -EINVAL; if (index == BPF_F_CURRENT_CPU) - index = raw_smp_processor_id(); + index = cpu; if (unlikely(index >= array->map.max_entries)) return -E2BIG; @@ -259,7 +260,7 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) event->attr.config != PERF_COUNT_SW_BPF_OUTPUT)) return -EINVAL; - if (unlikely(event->oncpu != smp_processor_id())) + if (unlikely(event->oncpu != cpu)) return -EOPNOTSUPP; perf_sample_data_init(&sample_data, 0, 0); -- cgit v0.10.2 From 6816a7ffce32e999601825ddfd887f36d3052932 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 28 Jun 2016 12:18:25 +0200 Subject: bpf, trace: add BPF_F_CURRENT_CPU flag for bpf_perf_event_read Follow-up commit to 1e33759c788c ("bpf, trace: add BPF_F_CURRENT_CPU flag for bpf_perf_event_output") to add the same functionality into bpf_perf_event_read() helper. The split of index into flags and index component is also safe here, since such large maps are rejected during map allocation time. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 406459b..58df2da 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -347,7 +347,7 @@ enum bpf_func_id { #define BPF_F_ZERO_CSUM_TX (1ULL << 1) #define BPF_F_DONT_FRAGMENT (1ULL << 2) -/* BPF_FUNC_perf_event_output flags. */ +/* BPF_FUNC_perf_event_output and BPF_FUNC_perf_event_read flags. */ #define BPF_F_INDEX_MASK 0xffffffffULL #define BPF_F_CURRENT_CPU BPF_F_INDEX_MASK diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 505f9e9..19c5b4a 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -188,13 +188,19 @@ const struct bpf_func_proto *bpf_get_trace_printk_proto(void) return &bpf_trace_printk_proto; } -static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5) +static u64 bpf_perf_event_read(u64 r1, u64 flags, u64 r3, u64 r4, u64 r5) { struct bpf_map *map = (struct bpf_map *) (unsigned long) r1; struct bpf_array *array = container_of(map, struct bpf_array, map); + unsigned int cpu = smp_processor_id(); + u64 index = flags & BPF_F_INDEX_MASK; struct bpf_event_entry *ee; struct perf_event *event; + if (unlikely(flags & ~(BPF_F_INDEX_MASK))) + return -EINVAL; + if (index == BPF_F_CURRENT_CPU) + index = cpu; if (unlikely(index >= array->map.max_entries)) return -E2BIG; @@ -208,8 +214,7 @@ static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5) return -EINVAL; /* make sure event is local and doesn't have pmu::count */ - if (event->oncpu != smp_processor_id() || - event->pmu->count) + if (unlikely(event->oncpu != cpu || event->pmu->count)) return -EINVAL; /* -- cgit v0.10.2 From 80b48c445797a634d869c7e5a53e182ba2688931 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 28 Jun 2016 12:18:26 +0200 Subject: bpf: don't use raw processor id in generic helper Use smp_processor_id() for the generic helper bpf_get_smp_processor_id() instead of the raw variant. This allows for preemption checks when we have DEBUG_PREEMPT, and otherwise uses the raw variant anyway. We only need to keep the raw variant for socket filters, but we can reuse the helper that is already there from cBPF side. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c index ad7a057..1ea3afb 100644 --- a/kernel/bpf/helpers.c +++ b/kernel/bpf/helpers.c @@ -101,7 +101,7 @@ const struct bpf_func_proto bpf_get_prandom_u32_proto = { static u64 bpf_get_smp_processor_id(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) { - return raw_smp_processor_id(); + return smp_processor_id(); } const struct bpf_func_proto bpf_get_smp_processor_id_proto = { diff --git a/net/core/filter.c b/net/core/filter.c index cb9fc16..46c88d9 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -150,6 +150,12 @@ static u64 __get_raw_cpu_id(u64 ctx, u64 a, u64 x, u64 r4, u64 r5) return raw_smp_processor_id(); } +static const struct bpf_func_proto bpf_get_raw_smp_processor_id_proto = { + .func = __get_raw_cpu_id, + .gpl_only = false, + .ret_type = RET_INTEGER, +}; + static u32 convert_skb_access(int skb_field, int dst_reg, int src_reg, struct bpf_insn *insn_buf) { @@ -2037,7 +2043,7 @@ sk_filter_func_proto(enum bpf_func_id func_id) case BPF_FUNC_get_prandom_u32: return &bpf_get_prandom_u32_proto; case BPF_FUNC_get_smp_processor_id: - return &bpf_get_smp_processor_id_proto; + return &bpf_get_raw_smp_processor_id_proto; case BPF_FUNC_tail_call: return &bpf_tail_call_proto; case BPF_FUNC_ktime_get_ns: @@ -2086,6 +2092,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) return &bpf_get_route_realm_proto; case BPF_FUNC_perf_event_output: return bpf_get_event_output_proto(); + case BPF_FUNC_get_smp_processor_id: + return &bpf_get_smp_processor_id_proto; default: return sk_filter_func_proto(func_id); } -- cgit v0.10.2 From 6578171a7ff0c31dc73258f93da7407510abf085 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 28 Jun 2016 12:18:27 +0200 Subject: bpf: add bpf_skb_change_proto helper This patch adds a minimal helper for doing the groundwork of changing the skb->protocol in a controlled way. Currently supported is v4 to v6 and vice versa transitions, which allows f.e. for a minimal, static nat64 implementation where applications in containers that still require IPv4 can be transparently operated in an IPv6-only environment. For example, host facing veth of the container can transparently do the transitions in a programmatic way with the help of clsact qdisc and cls_bpf. Idea is to separate concerns for keeping complexity of the helper lower, which means that the programs utilize bpf_skb_change_proto(), bpf_skb_store_bytes() and bpf_lX_csum_replace() to get the job done, instead of doing everything in a single helper (and thus partially duplicating helper functionality). Also, bpf_skb_change_proto() shouldn't need to deal with raw packet data as this is done by other helpers. bpf_skb_proto_6_to_4() and bpf_skb_proto_4_to_6() unclone the skb to operate on a private one, push or pop additionally required header space and migrate the gso/gro meta data from the shared info. We do mark the gso type as dodgy so that headers are checked and segs recalculated by the gso/gro engine. The gso_size target is adapted as well. The flags argument added is currently reserved and can be used for future extensions. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 58df2da..66cd738 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -313,6 +313,20 @@ enum bpf_func_id { */ BPF_FUNC_skb_get_tunnel_opt, BPF_FUNC_skb_set_tunnel_opt, + + /** + * bpf_skb_change_proto(skb, proto, flags) + * Change protocol of the skb. Currently supported is + * v4 -> v6, v6 -> v4 transitions. The helper will also + * resize the skb. eBPF program is expected to fill the + * new headers via skb_store_bytes and lX_csum_replace. + * @skb: pointer to skb + * @proto: new skb->protocol type + * @flags: reserved + * Return: 0 on success or negative error + */ + BPF_FUNC_skb_change_proto, + __BPF_FUNC_MAX_ID, }; diff --git a/net/core/filter.c b/net/core/filter.c index 46c88d9..d983e76 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1783,6 +1783,202 @@ const struct bpf_func_proto bpf_skb_vlan_pop_proto = { }; EXPORT_SYMBOL_GPL(bpf_skb_vlan_pop_proto); +static int bpf_skb_generic_push(struct sk_buff *skb, u32 off, u32 len) +{ + /* Caller already did skb_cow() with len as headroom, + * so no need to do it here. + */ + skb_push(skb, len); + memmove(skb->data, skb->data + len, off); + memset(skb->data + off, 0, len); + + /* No skb_postpush_rcsum(skb, skb->data + off, len) + * needed here as it does not change the skb->csum + * result for checksum complete when summing over + * zeroed blocks. + */ + return 0; +} + +static int bpf_skb_generic_pop(struct sk_buff *skb, u32 off, u32 len) +{ + /* skb_ensure_writable() is not needed here, as we're + * already working on an uncloned skb. + */ + if (unlikely(!pskb_may_pull(skb, off + len))) + return -ENOMEM; + + skb_postpull_rcsum(skb, skb->data + off, len); + memmove(skb->data + len, skb->data, off); + __skb_pull(skb, len); + + return 0; +} + +static int bpf_skb_net_hdr_push(struct sk_buff *skb, u32 off, u32 len) +{ + bool trans_same = skb->transport_header == skb->network_header; + int ret; + + /* There's no need for __skb_push()/__skb_pull() pair to + * get to the start of the mac header as we're guaranteed + * to always start from here under eBPF. + */ + ret = bpf_skb_generic_push(skb, off, len); + if (likely(!ret)) { + skb->mac_header -= len; + skb->network_header -= len; + if (trans_same) + skb->transport_header = skb->network_header; + } + + return ret; +} + +static int bpf_skb_net_hdr_pop(struct sk_buff *skb, u32 off, u32 len) +{ + bool trans_same = skb->transport_header == skb->network_header; + int ret; + + /* Same here, __skb_push()/__skb_pull() pair not needed. */ + ret = bpf_skb_generic_pop(skb, off, len); + if (likely(!ret)) { + skb->mac_header += len; + skb->network_header += len; + if (trans_same) + skb->transport_header = skb->network_header; + } + + return ret; +} + +static int bpf_skb_proto_4_to_6(struct sk_buff *skb) +{ + const u32 len_diff = sizeof(struct ipv6hdr) - sizeof(struct iphdr); + u32 off = skb->network_header - skb->mac_header; + int ret; + + ret = skb_cow(skb, len_diff); + if (unlikely(ret < 0)) + return ret; + + ret = bpf_skb_net_hdr_push(skb, off, len_diff); + if (unlikely(ret < 0)) + return ret; + + if (skb_is_gso(skb)) { + /* SKB_GSO_UDP stays as is. SKB_GSO_TCPV4 needs to + * be changed into SKB_GSO_TCPV6. + */ + if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4) { + skb_shinfo(skb)->gso_type &= ~SKB_GSO_TCPV4; + skb_shinfo(skb)->gso_type |= SKB_GSO_TCPV6; + } + + /* Due to IPv6 header, MSS needs to be downgraded. */ + skb_shinfo(skb)->gso_size -= len_diff; + /* Header must be checked, and gso_segs recomputed. */ + skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; + skb_shinfo(skb)->gso_segs = 0; + } + + skb->protocol = htons(ETH_P_IPV6); + skb_clear_hash(skb); + + return 0; +} + +static int bpf_skb_proto_6_to_4(struct sk_buff *skb) +{ + const u32 len_diff = sizeof(struct ipv6hdr) - sizeof(struct iphdr); + u32 off = skb->network_header - skb->mac_header; + int ret; + + ret = skb_unclone(skb, GFP_ATOMIC); + if (unlikely(ret < 0)) + return ret; + + ret = bpf_skb_net_hdr_pop(skb, off, len_diff); + if (unlikely(ret < 0)) + return ret; + + if (skb_is_gso(skb)) { + /* SKB_GSO_UDP stays as is. SKB_GSO_TCPV6 needs to + * be changed into SKB_GSO_TCPV4. + */ + if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) { + skb_shinfo(skb)->gso_type &= ~SKB_GSO_TCPV6; + skb_shinfo(skb)->gso_type |= SKB_GSO_TCPV4; + } + + /* Due to IPv4 header, MSS can be upgraded. */ + skb_shinfo(skb)->gso_size += len_diff; + /* Header must be checked, and gso_segs recomputed. */ + skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; + skb_shinfo(skb)->gso_segs = 0; + } + + skb->protocol = htons(ETH_P_IP); + skb_clear_hash(skb); + + return 0; +} + +static int bpf_skb_proto_xlat(struct sk_buff *skb, __be16 to_proto) +{ + __be16 from_proto = skb->protocol; + + if (from_proto == htons(ETH_P_IP) && + to_proto == htons(ETH_P_IPV6)) + return bpf_skb_proto_4_to_6(skb); + + if (from_proto == htons(ETH_P_IPV6) && + to_proto == htons(ETH_P_IP)) + return bpf_skb_proto_6_to_4(skb); + + return -ENOTSUPP; +} + +static u64 bpf_skb_change_proto(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5) +{ + struct sk_buff *skb = (struct sk_buff *) (long) r1; + __be16 proto = (__force __be16) r2; + int ret; + + if (unlikely(flags)) + return -EINVAL; + + /* General idea is that this helper does the basic groundwork + * needed for changing the protocol, and eBPF program fills the + * rest through bpf_skb_store_bytes(), bpf_lX_csum_replace() + * and other helpers, rather than passing a raw buffer here. + * + * The rationale is to keep this minimal and without a need to + * deal with raw packet data. F.e. even if we would pass buffers + * here, the program still needs to call the bpf_lX_csum_replace() + * helpers anyway. Plus, this way we keep also separation of + * concerns, since f.e. bpf_skb_store_bytes() should only take + * care of stores. + * + * Currently, additional options and extension header space are + * not supported, but flags register is reserved so we can adapt + * that. For offloads, we mark packet as dodgy, so that headers + * need to be verified first. + */ + ret = bpf_skb_proto_xlat(skb, proto); + bpf_compute_data_end(skb); + return ret; +} + +static const struct bpf_func_proto bpf_skb_change_proto_proto = { + .func = bpf_skb_change_proto, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_ANYTHING, + .arg3_type = ARG_ANYTHING, +}; + bool bpf_helper_changes_skb_data(void *func) { if (func == bpf_skb_vlan_push) @@ -1791,6 +1987,8 @@ bool bpf_helper_changes_skb_data(void *func) return true; if (func == bpf_skb_store_bytes) return true; + if (func == bpf_skb_change_proto) + return true; if (func == bpf_l3_csum_replace) return true; if (func == bpf_l4_csum_replace) @@ -2078,6 +2276,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) return &bpf_skb_vlan_push_proto; case BPF_FUNC_skb_vlan_pop: return &bpf_skb_vlan_pop_proto; + case BPF_FUNC_skb_change_proto: + return &bpf_skb_change_proto_proto; case BPF_FUNC_skb_get_tunnel_key: return &bpf_skb_get_tunnel_key_proto; case BPF_FUNC_skb_set_tunnel_key: -- cgit v0.10.2 From d2485c4242a826fdf493fd3a27b8b792965b9b9e Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 28 Jun 2016 12:18:28 +0200 Subject: bpf: add bpf_skb_change_type helper This work adds a helper for changing skb->pkt_type in a controlled way. We only allow a subset of possible values and can extend that in future should other use cases come up. Doing this as a helper has the advantage that errors can be handeled gracefully and thus helper kept extensible. It's a write counterpart to pkt_type member we can already read from struct __sk_buff context. Major use case is to change incoming skbs to PACKET_HOST in a programmatic way instead of having to recirculate via redirect(..., BPF_F_INGRESS), for example. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 66cd738..be6ac12 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -327,6 +327,15 @@ enum bpf_func_id { */ BPF_FUNC_skb_change_proto, + /** + * bpf_skb_change_type(skb, type) + * Change packet type of skb. + * @skb: pointer to skb + * @type: new skb->pkt_type type + * Return: 0 on success or negative error + */ + BPF_FUNC_skb_change_type, + __BPF_FUNC_MAX_ID, }; diff --git a/net/core/filter.c b/net/core/filter.c index d983e76..76f9a49 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1979,6 +1979,28 @@ static const struct bpf_func_proto bpf_skb_change_proto_proto = { .arg3_type = ARG_ANYTHING, }; +static u64 bpf_skb_change_type(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + struct sk_buff *skb = (struct sk_buff *) (long) r1; + u32 pkt_type = r2; + + /* We only allow a restricted subset to be changed for now. */ + if (unlikely(skb->pkt_type > PACKET_OTHERHOST || + pkt_type > PACKET_OTHERHOST)) + return -EINVAL; + + skb->pkt_type = pkt_type; + return 0; +} + +static const struct bpf_func_proto bpf_skb_change_type_proto = { + .func = bpf_skb_change_type, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_ANYTHING, +}; + bool bpf_helper_changes_skb_data(void *func) { if (func == bpf_skb_vlan_push) @@ -2278,6 +2300,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) return &bpf_skb_vlan_pop_proto; case BPF_FUNC_skb_change_proto: return &bpf_skb_change_proto_proto; + case BPF_FUNC_skb_change_type: + return &bpf_skb_change_type_proto; case BPF_FUNC_skb_get_tunnel_key: return &bpf_skb_get_tunnel_key_proto; case BPF_FUNC_skb_set_tunnel_key: -- cgit v0.10.2 From f21e4d8ed16bb9cae27c7f1a07a2e0cd4daab736 Mon Sep 17 00:00:00 2001 From: Martin Willi Date: Sat, 14 May 2016 10:34:44 +0200 Subject: mac80211_hwsim: Allow wmediumd to attach to radios created in its netns Registering wmediumd is currently limited to the initial network namespace. This patch enables wmediumd to attach from non-initial network namespaces using a user namespace having CAP_NET_ADMIN. A registered wmediumd can forward frames on radios that have been created in the same network namespace, even if they have been moved to other network namespaces. The wmediumd Netlink portid is tracked per net namespace. Additionally, the portid is stored on all radios created in that net namespace to simplify the portid lookup in the data path. Signed-off-by: Martin Willi Signed-off-by: Johannes Berg diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index a1e28a4..382109bb 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -41,8 +41,6 @@ MODULE_AUTHOR("Jouni Malinen"); MODULE_DESCRIPTION("Software simulator of 802.11 radio(s) for mac80211"); MODULE_LICENSE("GPL"); -static u32 wmediumd_portid; - static int radios = 2; module_param(radios, int, 0444); MODULE_PARM_DESC(radios, "Number of simulated radios"); @@ -258,6 +256,7 @@ static int hwsim_netgroup; struct hwsim_net { int netgroup; + u32 wmediumd; }; static inline int hwsim_net_get_netgroup(struct net *net) @@ -274,6 +273,20 @@ static inline void hwsim_net_set_netgroup(struct net *net) hwsim_net->netgroup = hwsim_netgroup++; } +static inline u32 hwsim_net_get_wmediumd(struct net *net) +{ + struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id); + + return hwsim_net->wmediumd; +} + +static inline void hwsim_net_set_wmediumd(struct net *net, u32 portid) +{ + struct hwsim_net *hwsim_net = net_generic(net, hwsim_net_id); + + hwsim_net->wmediumd = portid; +} + static struct class *hwsim_class; static struct net_device *hwsim_mon; /* global monitor netdev */ @@ -552,6 +565,8 @@ struct mac80211_hwsim_data { /* group shared by radios created in the same netns */ int netgroup; + /* wmediumd portid responsible for netgroup of this radio */ + u32 wmediumd; int power_level; @@ -983,6 +998,29 @@ static bool hwsim_ps_rx_ok(struct mac80211_hwsim_data *data, return true; } +static int hwsim_unicast_netgroup(struct mac80211_hwsim_data *data, + struct sk_buff *skb, int portid) +{ + struct net *net; + bool found = false; + int res = -ENOENT; + + rcu_read_lock(); + for_each_net_rcu(net) { + if (data->netgroup == hwsim_net_get_netgroup(net)) { + res = genlmsg_unicast(net, skb, portid); + found = true; + break; + } + } + rcu_read_unlock(); + + if (!found) + nlmsg_free(skb); + + return res; +} + static void mac80211_hwsim_tx_frame_nl(struct ieee80211_hw *hw, struct sk_buff *my_skb, int dst_portid) @@ -1062,7 +1100,7 @@ static void mac80211_hwsim_tx_frame_nl(struct ieee80211_hw *hw, goto nla_put_failure; genlmsg_end(skb, msg_head); - if (genlmsg_unicast(&init_net, skb, dst_portid)) + if (hwsim_unicast_netgroup(data, skb, dst_portid)) goto err_free_txskb; /* Enqueue the packet */ @@ -1355,7 +1393,7 @@ static void mac80211_hwsim_tx(struct ieee80211_hw *hw, mac80211_hwsim_monitor_rx(hw, skb, channel); /* wmediumd mode check */ - _portid = ACCESS_ONCE(wmediumd_portid); + _portid = ACCESS_ONCE(data->wmediumd); if (_portid) return mac80211_hwsim_tx_frame_nl(hw, skb, _portid); @@ -1451,7 +1489,8 @@ static void mac80211_hwsim_tx_frame(struct ieee80211_hw *hw, struct sk_buff *skb, struct ieee80211_channel *chan) { - u32 _pid = ACCESS_ONCE(wmediumd_portid); + struct mac80211_hwsim_data *data = hw->priv; + u32 _pid = ACCESS_ONCE(data->wmediumd); if (ieee80211_hw_check(hw, SUPPORTS_RC_TABLE)) { struct ieee80211_tx_info *txi = IEEE80211_SKB_CB(skb); @@ -2796,6 +2835,20 @@ static struct mac80211_hwsim_data *get_hwsim_data_ref_from_addr(const u8 *addr) return data; } +static void hwsim_register_wmediumd(struct net *net, u32 portid) +{ + struct mac80211_hwsim_data *data; + + hwsim_net_set_wmediumd(net, portid); + + spin_lock_bh(&hwsim_radio_lock); + list_for_each_entry(data, &hwsim_radios, list) { + if (data->netgroup == hwsim_net_get_netgroup(net)) + data->wmediumd = portid; + } + spin_unlock_bh(&hwsim_radio_lock); +} + static int hwsim_tx_info_frame_received_nl(struct sk_buff *skb_2, struct genl_info *info) { @@ -2811,9 +2864,6 @@ static int hwsim_tx_info_frame_received_nl(struct sk_buff *skb_2, int i; bool found = false; - if (info->snd_portid != wmediumd_portid) - return -EINVAL; - if (!info->attrs[HWSIM_ATTR_ADDR_TRANSMITTER] || !info->attrs[HWSIM_ATTR_FLAGS] || !info->attrs[HWSIM_ATTR_COOKIE] || @@ -2829,6 +2879,12 @@ static int hwsim_tx_info_frame_received_nl(struct sk_buff *skb_2, if (!data2) goto out; + if (hwsim_net_get_netgroup(genl_info_net(info)) != data2->netgroup) + goto out; + + if (info->snd_portid != data2->wmediumd) + goto out; + /* look for the skb matching the cookie passed back from user */ skb_queue_walk_safe(&data2->pending, skb, tmp) { u64 skb_cookie; @@ -2892,9 +2948,6 @@ static int hwsim_cloned_frame_received_nl(struct sk_buff *skb_2, void *frame_data; struct sk_buff *skb = NULL; - if (info->snd_portid != wmediumd_portid) - return -EINVAL; - if (!info->attrs[HWSIM_ATTR_ADDR_RECEIVER] || !info->attrs[HWSIM_ATTR_FRAME] || !info->attrs[HWSIM_ATTR_RX_RATE] || @@ -2920,6 +2973,12 @@ static int hwsim_cloned_frame_received_nl(struct sk_buff *skb_2, if (!data2) goto out; + if (hwsim_net_get_netgroup(genl_info_net(info)) != data2->netgroup) + goto out; + + if (info->snd_portid != data2->wmediumd) + goto out; + /* check if radio is configured properly */ if (data2->idle || !data2->started) @@ -2966,6 +3025,7 @@ out: static int hwsim_register_received_nl(struct sk_buff *skb_2, struct genl_info *info) { + struct net *net = genl_info_net(info); struct mac80211_hwsim_data *data; int chans = 1; @@ -2982,10 +3042,10 @@ static int hwsim_register_received_nl(struct sk_buff *skb_2, if (chans > 1) return -EOPNOTSUPP; - if (wmediumd_portid) + if (hwsim_net_get_wmediumd(net)) return -EBUSY; - wmediumd_portid = info->snd_portid; + hwsim_register_wmediumd(net, info->snd_portid); printk(KERN_DEBUG "mac80211_hwsim: received a REGISTER, " "switching to wmediumd mode with pid %d\n", info->snd_portid); @@ -3152,7 +3212,7 @@ static const struct genl_ops hwsim_ops[] = { .cmd = HWSIM_CMD_REGISTER, .policy = hwsim_genl_policy, .doit = hwsim_register_received_nl, - .flags = GENL_ADMIN_PERM, + .flags = GENL_UNS_ADMIN_PERM, }, { .cmd = HWSIM_CMD_FRAME, @@ -3218,10 +3278,10 @@ static int mac80211_hwsim_netlink_notify(struct notifier_block *nb, remove_user_radios(notify->portid); - if (notify->portid == wmediumd_portid) { + if (notify->portid == hwsim_net_get_wmediumd(notify->net)) { printk(KERN_INFO "mac80211_hwsim: wmediumd released netlink" " socket, switching to perfect channel medium\n"); - wmediumd_portid = 0; + hwsim_register_wmediumd(notify->net, 0); } return NOTIFY_DONE; -- cgit v0.10.2 From f151d9db4c1e7f7ac202ae75f4cbc62cfc784156 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 15 Jun 2016 22:29:41 +0200 Subject: nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann Signed-off-by: Johannes Berg diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index c503e96..244d552 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -5287,6 +5287,51 @@ static const struct nla_policy [NL80211_MESH_SETUP_USERSPACE_AMPE] = { .type = NLA_FLAG }, }; +static int nl80211_check_bool(const struct nlattr *nla, u8 min, u8 max, bool *out) +{ + u8 val = nla_get_u8(nla); + if (val < min || val > max) + return -EINVAL; + *out = val; + return 0; +} + +static int nl80211_check_u8(const struct nlattr *nla, u8 min, u8 max, u8 *out) +{ + u8 val = nla_get_u8(nla); + if (val < min || val > max) + return -EINVAL; + *out = val; + return 0; +} + +static int nl80211_check_u16(const struct nlattr *nla, u16 min, u16 max, u16 *out) +{ + u16 val = nla_get_u16(nla); + if (val < min || val > max) + return -EINVAL; + *out = val; + return 0; +} + +static int nl80211_check_u32(const struct nlattr *nla, u32 min, u32 max, u32 *out) +{ + u32 val = nla_get_u32(nla); + if (val < min || val > max) + return -EINVAL; + *out = val; + return 0; +} + +static int nl80211_check_s32(const struct nlattr *nla, s32 min, s32 max, s32 *out) +{ + s32 val = nla_get_s32(nla); + if (val < min || val > max) + return -EINVAL; + *out = val; + return 0; +} + static int nl80211_parse_mesh_config(struct genl_info *info, struct mesh_config *cfg, u32 *mask_out) @@ -5297,9 +5342,8 @@ static int nl80211_parse_mesh_config(struct genl_info *info, #define FILL_IN_MESH_PARAM_IF_SET(tb, cfg, param, min, max, mask, attr, fn) \ do { \ if (tb[attr]) { \ - if (fn(tb[attr]) < min || fn(tb[attr]) > max) \ + if (fn(tb[attr], min, max, &cfg->param)) \ return -EINVAL; \ - cfg->param = fn(tb[attr]); \ mask |= (1 << (attr - 1)); \ } \ } while (0) @@ -5318,99 +5362,99 @@ do { \ /* Fill in the params struct */ FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshRetryTimeout, 1, 255, mask, NL80211_MESHCONF_RETRY_TIMEOUT, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshConfirmTimeout, 1, 255, mask, NL80211_MESHCONF_CONFIRM_TIMEOUT, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHoldingTimeout, 1, 255, mask, NL80211_MESHCONF_HOLDING_TIMEOUT, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshMaxPeerLinks, 0, 255, mask, NL80211_MESHCONF_MAX_PEER_LINKS, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshMaxRetries, 0, 16, mask, NL80211_MESHCONF_MAX_RETRIES, - nla_get_u8); + nl80211_check_u8); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshTTL, 1, 255, - mask, NL80211_MESHCONF_TTL, nla_get_u8); + mask, NL80211_MESHCONF_TTL, nl80211_check_u8); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, element_ttl, 1, 255, mask, NL80211_MESHCONF_ELEMENT_TTL, - nla_get_u8); + nl80211_check_u8); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, auto_open_plinks, 0, 1, mask, NL80211_MESHCONF_AUTO_OPEN_PLINKS, - nla_get_u8); + nl80211_check_bool); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshNbrOffsetMaxNeighbor, 1, 255, mask, NL80211_MESHCONF_SYNC_OFFSET_MAX_NEIGHBOR, - nla_get_u32); + nl80211_check_u32); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPmaxPREQretries, 0, 255, mask, NL80211_MESHCONF_HWMP_MAX_PREQ_RETRIES, - nla_get_u8); + nl80211_check_u8); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, path_refresh_time, 1, 65535, mask, NL80211_MESHCONF_PATH_REFRESH_TIME, - nla_get_u32); + nl80211_check_u32); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, min_discovery_timeout, 1, 65535, mask, NL80211_MESHCONF_MIN_DISCOVERY_TIMEOUT, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPactivePathTimeout, 1, 65535, mask, NL80211_MESHCONF_HWMP_ACTIVE_PATH_TIMEOUT, - nla_get_u32); + nl80211_check_u32); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPpreqMinInterval, 1, 65535, mask, NL80211_MESHCONF_HWMP_PREQ_MIN_INTERVAL, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPperrMinInterval, 1, 65535, mask, NL80211_MESHCONF_HWMP_PERR_MIN_INTERVAL, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPnetDiameterTraversalTime, 1, 65535, mask, NL80211_MESHCONF_HWMP_NET_DIAM_TRVS_TIME, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPRootMode, 0, 4, mask, NL80211_MESHCONF_HWMP_ROOTMODE, - nla_get_u8); + nl80211_check_u8); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPRannInterval, 1, 65535, mask, NL80211_MESHCONF_HWMP_RANN_INTERVAL, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshGateAnnouncementProtocol, 0, 1, mask, NL80211_MESHCONF_GATE_ANNOUNCEMENTS, - nla_get_u8); + nl80211_check_bool); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshForwarding, 0, 1, mask, NL80211_MESHCONF_FORWARDING, - nla_get_u8); + nl80211_check_bool); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, rssi_threshold, -255, 0, mask, NL80211_MESHCONF_RSSI_THRESHOLD, - nla_get_s32); + nl80211_check_s32); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, ht_opmode, 0, 16, mask, NL80211_MESHCONF_HT_OPMODE, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPactivePathToRootTimeout, 1, 65535, mask, NL80211_MESHCONF_HWMP_PATH_TO_ROOT_TIMEOUT, - nla_get_u32); + nl80211_check_u32); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMProotInterval, 1, 65535, mask, NL80211_MESHCONF_HWMP_ROOT_INTERVAL, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPconfirmationInterval, 1, 65535, mask, NL80211_MESHCONF_HWMP_CONFIRMATION_INTERVAL, - nla_get_u16); + nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, power_mode, NL80211_MESH_POWER_ACTIVE, NL80211_MESH_POWER_MAX, mask, NL80211_MESHCONF_POWER_MODE, - nla_get_u32); + nl80211_check_u32); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshAwakeWindowDuration, 0, 65535, mask, - NL80211_MESHCONF_AWAKE_WINDOW, nla_get_u16); + NL80211_MESHCONF_AWAKE_WINDOW, nl80211_check_u16); FILL_IN_MESH_PARAM_IF_SET(tb, cfg, plink_timeout, 0, 0xffffffff, mask, NL80211_MESHCONF_PLINK_TIMEOUT, - nla_get_u32); + nl80211_check_u32); if (mask_out) *mask_out = mask; -- cgit v0.10.2 From e98e915e11ad1efb11147122bd4932ec6b3425da Mon Sep 17 00:00:00 2001 From: Masashi Honma Date: Wed, 22 Jun 2016 20:23:03 +0900 Subject: wireless: Use macro instead of number Use IEEE80211_MIN_ACTION_SIZE macro for robust management frame check. Signed-off-by: Masashi Honma Signed-off-by: Johannes Berg diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index b118744..1daebb3 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -2464,7 +2464,7 @@ static inline bool _ieee80211_is_robust_mgmt_frame(struct ieee80211_hdr *hdr) */ static inline bool ieee80211_is_robust_mgmt_frame(struct sk_buff *skb) { - if (skb->len < 25) + if (skb->len < IEEE80211_MIN_ACTION_SIZE) return false; return _ieee80211_is_robust_mgmt_frame((void *)skb->data); } -- cgit v0.10.2 From 49708e3772ce648be425778702a266b207e89d4e Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 27 Jun 2016 17:31:18 +0300 Subject: mac80211: silence an uninitialized variable warning We normally return an uninitialized value, but no one checks it so it doesn't matter. Anyway, let's silence the static checker warning. Signed-off-by: Dan Carpenter Signed-off-by: Johannes Berg diff --git a/net/mac80211/tdls.c b/net/mac80211/tdls.c index 1c7d45a..b5d28f1 100644 --- a/net/mac80211/tdls.c +++ b/net/mac80211/tdls.c @@ -1747,6 +1747,7 @@ ieee80211_process_tdls_channel_switch_resp(struct ieee80211_sub_if_data *sdata, goto out; } + ret = 0; call_drv: drv_tdls_recv_channel_switch(sdata->local, sdata, ¶ms); -- cgit v0.10.2 From 46f6b06050b736dab4d41494dae27b883cddc365 Mon Sep 17 00:00:00 2001 From: Masashi Honma Date: Wed, 22 Jun 2016 19:55:20 +0900 Subject: mac80211: Encrypt "Group addressed privacy" action frames Previously, the action frames to group address was not encrypted. But [1] "Table 8-38 Category values" indicates "Mesh" and "Multihop" category action frames should be encrypted (Group addressed privacy == yes). And the encyption key should be MGTK ([1] 10.13 Group addressed robust management frame procedures). So this patch modifies the code to make it suitable for spec. [1] IEEE Std 802.11-2012 Signed-off-by: Masashi Honma Signed-off-by: Johannes Berg diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 1daebb3..a80516f 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -19,6 +19,7 @@ #include #include +#include #include #include @@ -2487,6 +2488,35 @@ static inline bool ieee80211_is_public_action(struct ieee80211_hdr *hdr, } /** + * _ieee80211_is_group_privacy_action - check if frame is a group addressed + * privacy action frame + * @hdr: the frame + */ +static inline bool _ieee80211_is_group_privacy_action(struct ieee80211_hdr *hdr) +{ + struct ieee80211_mgmt *mgmt = (void *)hdr; + + if (!ieee80211_is_action(hdr->frame_control) || + !is_multicast_ether_addr(hdr->addr1)) + return false; + + return mgmt->u.action.category == WLAN_CATEGORY_MESH_ACTION || + mgmt->u.action.category == WLAN_CATEGORY_MULTIHOP_ACTION; +} + +/** + * ieee80211_is_group_privacy_action - check if frame is a group addressed + * privacy action frame + * @skb: the skb containing the frame, length will be checked + */ +static inline bool ieee80211_is_group_privacy_action(struct sk_buff *skb) +{ + if (skb->len < IEEE80211_MIN_ACTION_SIZE) + return false; + return _ieee80211_is_group_privacy_action((void *)skb->data); +} + +/** * ieee80211_tu_to_usec - convert time units (TU) to microseconds * @tu: the TUs */ diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 9a1eb70..2e8a902 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -1624,8 +1624,13 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) if (mmie_keyidx < NUM_DEFAULT_KEYS || mmie_keyidx >= NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS) return RX_DROP_MONITOR; /* unexpected BIP keyidx */ - if (rx->sta) + if (rx->sta) { + if (ieee80211_is_group_privacy_action(skb) && + test_sta_flag(rx->sta, WLAN_STA_MFP)) + return RX_DROP_MONITOR; + rx->key = rcu_dereference(rx->sta->gtk[mmie_keyidx]); + } if (!rx->key) rx->key = rcu_dereference(rx->sdata->keys[mmie_keyidx]); } else if (!ieee80211_has_protected(fc)) { diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 44ec605..fa8d38e 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -593,6 +593,9 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx) else if (tx->sta && (key = rcu_dereference(tx->sta->ptk[tx->sta->ptk_idx]))) tx->key = key; + else if (ieee80211_is_group_privacy_action(tx->skb) && + (key = rcu_dereference(tx->sdata->default_multicast_key))) + tx->key = key; else if (ieee80211_is_mgmt(hdr->frame_control) && is_multicast_ether_addr(hdr->addr1) && ieee80211_is_robust_mgmt_frame(tx->skb) && @@ -625,7 +628,8 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx) case WLAN_CIPHER_SUITE_GCMP_256: if (!ieee80211_is_data_present(hdr->frame_control) && !ieee80211_use_mfp(hdr->frame_control, tx->sta, - tx->skb)) + tx->skb) && + !ieee80211_is_group_privacy_action(tx->skb)) tx->key = NULL; else skip_hw = (tx->key->conf.flags & -- cgit v0.10.2 From efc401f49adf9c53a95f0430496c7a5433612e74 Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Sat, 25 Jun 2016 19:14:16 -0400 Subject: mac80211: use common cleanup for user/!user_mpm We've accumulated a couple of different fixes now to mesh_sta_cleanup() due to the different paths that user_mpm and !user_mpm cases take -- one fix to flush nexthop paths and one to fix the counting. The only caller of mesh_plink_deactivate() is mesh_sta_cleanup(), so we can push the user_mpm checks down into there in order to share more code. In doing so, we can remove an extra call to mesh_path_flush_by_nexthop() and the (unnecessary) call to mesh_accept_plinks_update(). This will also ensure the powersaving state code gets called in the user_mpm case. The only cleanup tasks we need to avoid when MPM is in user-space are sending the peering frames and stopping the plink timer, so wrap those in the appropriate check. Signed-off-by: Bob Copeland Signed-off-by: Johannes Berg diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c index 6a1603b..c66411d 100644 --- a/net/mac80211/mesh.c +++ b/net/mac80211/mesh.c @@ -148,25 +148,7 @@ u32 mesh_accept_plinks_update(struct ieee80211_sub_if_data *sdata) void mesh_sta_cleanup(struct sta_info *sta) { struct ieee80211_sub_if_data *sdata = sta->sdata; - u32 changed = 0; - - /* - * maybe userspace handles peer allocation and peering, but in either - * case the beacon is still generated by the kernel and we might need - * an update. - */ - if (sdata->u.mesh.user_mpm && - sta->mesh->plink_state == NL80211_PLINK_ESTAB) - changed |= mesh_plink_dec_estab_count(sdata); - changed |= mesh_accept_plinks_update(sdata); - if (!sdata->u.mesh.user_mpm) { - changed |= mesh_plink_deactivate(sta); - del_timer_sync(&sta->mesh->plink_timer); - } - - /* make sure no readers can access nexthop sta from here on */ - mesh_path_flush_by_nexthop(sta); - synchronize_net(); + u32 changed = mesh_plink_deactivate(sta); if (changed) ieee80211_mbss_info_change_notify(sdata, changed); diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c index 79f2a0a..7fcdcf6 100644 --- a/net/mac80211/mesh_plink.c +++ b/net/mac80211/mesh_plink.c @@ -370,13 +370,21 @@ u32 mesh_plink_deactivate(struct sta_info *sta) spin_lock_bh(&sta->mesh->plink_lock); changed = __mesh_plink_deactivate(sta); - sta->mesh->reason = WLAN_REASON_MESH_PEER_CANCELED; - mesh_plink_frame_tx(sdata, sta, WLAN_SP_MESH_PEERING_CLOSE, - sta->sta.addr, sta->mesh->llid, sta->mesh->plid, - sta->mesh->reason); + + if (!sdata->u.mesh.user_mpm) { + sta->mesh->reason = WLAN_REASON_MESH_PEER_CANCELED; + mesh_plink_frame_tx(sdata, sta, WLAN_SP_MESH_PEERING_CLOSE, + sta->sta.addr, sta->mesh->llid, + sta->mesh->plid, sta->mesh->reason); + } spin_unlock_bh(&sta->mesh->plink_lock); + if (!sdata->u.mesh.user_mpm) + del_timer_sync(&sta->mesh->plink_timer); mesh_path_flush_by_nexthop(sta); + /* make sure no readers can access nexthop sta from here on */ + synchronize_net(); + return changed; } -- cgit v0.10.2 From 59a7c828d7e7d5a1be224a0d68a41ca2302843ea Mon Sep 17 00:00:00 2001 From: Michal Kazior Date: Wed, 29 Jun 2016 14:00:34 +0200 Subject: mac80211: fix fq lockdep warnings Some lockdep assertions were not fulfilled and resulted in a kernel warning/call trace if driver used intermediate software queues (e.g. ath10k). Existing code sequences should've guaranteed safety but it's always good to be extra careful. The call trace could look like this: [ 237.335805] ------------[ cut here ]------------ [ 237.335852] WARNING: CPU: 3 PID: 1921 at include/net/fq_impl.h:22 fq_flow_dequeue+0xed/0x140 [mac80211] [ 237.335855] Modules linked in: ath10k_pci(E-) ath10k_core(E) ath(E) mac80211(E) cfg80211(E) [ 237.335913] CPU: 3 PID: 1921 Comm: rmmod Tainted: G W E 4.7.0-rc4-wt-ath+ #1377 [ 237.335916] Hardware name: Hewlett-Packard HP ProBook 6540b/1722, BIOS 68CDD Ver. F.04 01/27/2010 [ 237.335918] 00200286 00200286 eff85dac c14151e2 f901574e 00000000 eff85de0 c1081075 [ 237.335928] c1ab91f0 00000003 00000781 f901574e 00000016 f8fbabad f8fbabad 00000016 [ 237.335938] eb24ff60 00000000 ef3886c0 eff85df4 c10810ba 00000009 00000000 00000000 [ 237.335948] Call Trace: [ 237.335953] [] dump_stack+0x76/0xb4 [ 237.335957] [] __warn+0xe5/0x100 [ 237.336002] [] ? fq_flow_dequeue+0xed/0x140 [mac80211] [ 237.336046] [] ? fq_flow_dequeue+0xed/0x140 [mac80211] [ 237.336053] [] warn_slowpath_null+0x2a/0x30 [ 237.336095] [] fq_flow_dequeue+0xed/0x140 [mac80211] [ 237.336137] [] fq_flow_reset.constprop.56+0x2a/0x90 [mac80211] [ 237.336180] [] fq_reset.constprop.59+0x2a/0x50 [mac80211] [ 237.336222] [] ieee80211_txq_teardown_flows+0x38/0x40 [mac80211] [ 237.336258] [] ieee80211_unregister_hw+0xe4/0x120 [mac80211] [ 237.336275] [] ath10k_mac_unregister+0x16/0x50 [ath10k_core] [ 237.336292] [] ath10k_core_unregister+0x3d/0x90 [ath10k_core] [ 237.336301] [] ath10k_pci_remove+0x36/0xa0 [ath10k_pci] [ 237.336307] [] pci_device_remove+0x38/0xb0 ... Fixes: 5caa328e3811 ("mac80211: implement codel on fair queuing flows") Fixes: fa962b92120b ("mac80211: implement fair queueing per txq") Tested-by: Kalle Valo Reported-by: Kalle Valo Signed-off-by: Michal Kazior Signed-off-by: Johannes Berg diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index fa8d38e..91461c4 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1449,7 +1449,9 @@ int ieee80211_txq_setup_flows(struct ieee80211_local *local) local->cvars = kcalloc(fq->flows_cnt, sizeof(local->cvars[0]), GFP_KERNEL); if (!local->cvars) { + spin_lock_bh(&fq->lock); fq_reset(fq, fq_skb_free_func); + spin_unlock_bh(&fq->lock); return -ENOMEM; } @@ -1469,7 +1471,9 @@ void ieee80211_txq_teardown_flows(struct ieee80211_local *local) kfree(local->cvars); local->cvars = NULL; + spin_lock_bh(&fq->lock); fq_reset(fq, fq_skb_free_func); + spin_unlock_bh(&fq->lock); } struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, -- cgit v0.10.2 From 80e73cc563c4359be809a03bcb8e7e28141a813a Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 28 Jun 2016 16:57:05 +0200 Subject: net: rtnetlink: add support for the IFLA_STATS_LINK_XSTATS_SLAVE attribute This patch adds support for the IFLA_STATS_LINK_XSTATS_SLAVE attribute which allows to export per-slave statistics if the master device supports the linkxstats callback. The attribute is passed down to the linkxstats callback and it is up to the callback user to use it (an example has been added to the only current user - the bridge). This allows us to query only specific slaves of master devices like bridge ports and export only what we're interested in instead of having to dump all ports and searching only for a single one. This will be used to export per-port IGMP/MLD stats and also per-port vlan stats in the future, possibly other statistics as well. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h index 006a7b8..4113916 100644 --- a/include/net/rtnetlink.h +++ b/include/net/rtnetlink.h @@ -98,10 +98,11 @@ struct rtnl_link_ops { const struct net_device *dev, const struct net_device *slave_dev); struct net *(*get_link_net)(const struct net_device *dev); - size_t (*get_linkxstats_size)(const struct net_device *dev); + size_t (*get_linkxstats_size)(const struct net_device *dev, + int attr); int (*fill_linkxstats)(struct sk_buff *skb, const struct net_device *dev, - int *prividx); + int *prividx, int attr); }; int __rtnl_link_register(struct rtnl_link_ops *ops); diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index bb36bd5..db2458a 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -822,6 +822,7 @@ enum { IFLA_STATS_UNSPEC, /* also used as 64bit pad attribute */ IFLA_STATS_LINK_64, IFLA_STATS_LINK_XSTATS, + IFLA_STATS_LINK_XSTATS_SLAVE, __IFLA_STATS_MAX, }; diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index 85e89f6..ed75ff9 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -1234,7 +1234,7 @@ static int br_fill_info(struct sk_buff *skb, const struct net_device *brdev) return 0; } -static size_t br_get_linkxstats_size(const struct net_device *dev) +static size_t bridge_get_linkxstats_size(const struct net_device *dev) { struct net_bridge *br = netdev_priv(dev); struct net_bridge_vlan_group *vg; @@ -1254,8 +1254,30 @@ static size_t br_get_linkxstats_size(const struct net_device *dev) nla_total_size(0); } -static int br_fill_linkxstats(struct sk_buff *skb, const struct net_device *dev, - int *prividx) +static size_t brport_get_linkxstats_size(const struct net_device *dev) +{ + return nla_total_size(0); +} + +static size_t br_get_linkxstats_size(const struct net_device *dev, int attr) +{ + size_t retsize = 0; + + switch (attr) { + case IFLA_STATS_LINK_XSTATS: + retsize = bridge_get_linkxstats_size(dev); + break; + case IFLA_STATS_LINK_XSTATS_SLAVE: + retsize = brport_get_linkxstats_size(dev); + break; + } + + return retsize; +} + +static int bridge_fill_linkxstats(struct sk_buff *skb, + const struct net_device *dev, + int *prividx) { struct net_bridge *br = netdev_priv(dev); struct net_bridge_vlan_group *vg; @@ -1298,6 +1320,37 @@ nla_put_failure: return -EMSGSIZE; } +static int brport_fill_linkxstats(struct sk_buff *skb, + const struct net_device *dev, + int *prividx) +{ + struct nlattr *nest; + + nest = nla_nest_start(skb, LINK_XSTATS_TYPE_BRIDGE); + if (!nest) + return -EMSGSIZE; + nla_nest_end(skb, nest); + + return 0; +} + +static int br_fill_linkxstats(struct sk_buff *skb, const struct net_device *dev, + int *prividx, int attr) +{ + int ret = -EINVAL; + + switch (attr) { + case IFLA_STATS_LINK_XSTATS: + ret = bridge_fill_linkxstats(skb, dev, prividx); + break; + case IFLA_STATS_LINK_XSTATS_SLAVE: + ret = brport_fill_linkxstats(skb, dev, prividx); + break; + } + + return ret; +} + static struct rtnl_af_ops br_af_ops __read_mostly = { .family = AF_BRIDGE, .get_link_af_size = br_get_link_af_size_filtered, diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index eb49ca2..cfed7bc 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -3519,7 +3519,32 @@ static int rtnl_fill_statsinfo(struct sk_buff *skb, struct net_device *dev, if (!attr) goto nla_put_failure; - err = ops->fill_linkxstats(skb, dev, prividx); + err = ops->fill_linkxstats(skb, dev, prividx, *idxattr); + nla_nest_end(skb, attr); + if (err) + goto nla_put_failure; + *idxattr = 0; + } + } + + if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS_SLAVE, + *idxattr)) { + const struct rtnl_link_ops *ops = NULL; + const struct net_device *master; + + master = netdev_master_upper_dev_get(dev); + if (master) + ops = master->rtnl_link_ops; + if (ops && ops->fill_linkxstats) { + int err; + + *idxattr = IFLA_STATS_LINK_XSTATS_SLAVE; + attr = nla_nest_start(skb, + IFLA_STATS_LINK_XSTATS_SLAVE); + if (!attr) + goto nla_put_failure; + + err = ops->fill_linkxstats(skb, dev, prividx, *idxattr); nla_nest_end(skb, attr); if (err) goto nla_put_failure; @@ -3555,14 +3580,35 @@ static size_t if_nlmsg_stats_size(const struct net_device *dev, if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS, 0)) { const struct rtnl_link_ops *ops = dev->rtnl_link_ops; + int attr = IFLA_STATS_LINK_XSTATS; if (ops && ops->get_linkxstats_size) { - size += nla_total_size(ops->get_linkxstats_size(dev)); + size += nla_total_size(ops->get_linkxstats_size(dev, + attr)); /* for IFLA_STATS_LINK_XSTATS */ size += nla_total_size(0); } } + if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_XSTATS_SLAVE, 0)) { + struct net_device *_dev = (struct net_device *)dev; + const struct rtnl_link_ops *ops = NULL; + const struct net_device *master; + + /* netdev_master_upper_dev_get can't take const */ + master = netdev_master_upper_dev_get(_dev); + if (master) + ops = master->rtnl_link_ops; + if (ops && ops->get_linkxstats_size) { + int attr = IFLA_STATS_LINK_XSTATS_SLAVE; + + size += nla_total_size(ops->get_linkxstats_size(dev, + attr)); + /* for IFLA_STATS_LINK_XSTATS_SLAVE */ + size += nla_total_size(0); + } + } + return size; } -- cgit v0.10.2 From 1080ab95e3c7bdd77870e209aff83c763fdcf439 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 28 Jun 2016 16:57:06 +0200 Subject: net: bridge: add support for IGMP/MLD stats and export them via netlink This patch adds stats support for the currently used IGMP/MLD types by the bridge. The stats are per-port (plus one stat per-bridge) and per-direction (RX/TX). The stats are exported via netlink via the new linkxstats API (RTM_GETSTATS). In order to minimize the performance impact, a new option is used to enable/disable the stats - multicast_stats_enabled, similar to the recent vlan stats. Also in order to avoid multiple IGMP/MLD type lookups and checks, we make use of the current "igmp" member of the bridge private skb->cb region to record the type on Rx (both host-generated and external packets pass by multicast_rcv()). We can do that since the igmp member was used as a boolean and all the valid IGMP/MLD types are positive values. The normal bridge fast-path is not affected at all, the only affected paths are the flooding ones and since we make use of the IGMP/MLD type, we can quickly determine if the packet should be counted using cache-hot data (cb's igmp member). We add counters for: * IGMP Queries * IGMP Leaves * IGMP v1/v2/v3 reports * MLD Queries * MLD Leaves * MLD v1/v2 reports These are invaluable when monitoring or debugging complex multicast setups with bridges. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h index 397d503..8304fe6 100644 --- a/include/uapi/linux/if_bridge.h +++ b/include/uapi/linux/if_bridge.h @@ -247,8 +247,34 @@ enum { enum { BRIDGE_XSTATS_UNSPEC, BRIDGE_XSTATS_VLAN, + BRIDGE_XSTATS_MCAST, + BRIDGE_XSTATS_PAD, __BRIDGE_XSTATS_MAX }; #define BRIDGE_XSTATS_MAX (__BRIDGE_XSTATS_MAX - 1) +enum { + BR_MCAST_DIR_RX, + BR_MCAST_DIR_TX, + BR_MCAST_DIR_SIZE +}; + +/* IGMP/MLD statistics */ +struct br_mcast_stats { + __u64 igmp_queries[BR_MCAST_DIR_SIZE]; + __u64 igmp_leaves[BR_MCAST_DIR_SIZE]; + __u64 igmp_v1reports[BR_MCAST_DIR_SIZE]; + __u64 igmp_v2reports[BR_MCAST_DIR_SIZE]; + __u64 igmp_v3reports[BR_MCAST_DIR_SIZE]; + __u64 igmp_parse_errors; + + __u64 mld_queries[BR_MCAST_DIR_SIZE]; + __u64 mld_leaves[BR_MCAST_DIR_SIZE]; + __u64 mld_v1reports[BR_MCAST_DIR_SIZE]; + __u64 mld_v2reports[BR_MCAST_DIR_SIZE]; + __u64 mld_parse_errors; + + __u64 mcast_bytes[BR_MCAST_DIR_SIZE]; + __u64 mcast_packets[BR_MCAST_DIR_SIZE]; +}; #endif /* _UAPI_LINUX_IF_BRIDGE_H */ diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index db2458a..4285ac3 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -273,6 +273,7 @@ enum { IFLA_BR_VLAN_DEFAULT_PVID, IFLA_BR_PAD, IFLA_BR_VLAN_STATS_ENABLED, + IFLA_BR_MCAST_STATS_ENABLED, __IFLA_BR_MAX, }; diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 2c8095a..0c39e0f 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -104,8 +104,16 @@ static int br_dev_init(struct net_device *dev) return -ENOMEM; err = br_vlan_init(br); - if (err) + if (err) { free_percpu(br->stats); + return err; + } + + err = br_multicast_init_stats(br); + if (err) { + free_percpu(br->stats); + br_vlan_flush(br); + } br_set_lockdep_class(dev); return err; diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index f47759f..6c19603 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -198,8 +198,10 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, struct sk_buff *skb), bool unicast) { - struct net_bridge_port *p; + u8 igmp_type = br_multicast_igmp_type(skb); + __be16 proto = skb->protocol; struct net_bridge_port *prev; + struct net_bridge_port *p; prev = NULL; @@ -218,6 +220,9 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, prev = maybe_deliver(prev, p, skb, __packet_hook); if (IS_ERR(prev)) goto out; + if (prev == p) + br_multicast_count(p->br, p, proto, igmp_type, + BR_MCAST_DIR_TX); } if (!prev) @@ -257,9 +262,12 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, struct sk_buff *skb)) { struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev; + u8 igmp_type = br_multicast_igmp_type(skb); struct net_bridge *br = netdev_priv(dev); struct net_bridge_port *prev = NULL; struct net_bridge_port_group *p; + __be16 proto = skb->protocol; + struct hlist_node *rp; rp = rcu_dereference(hlist_first_rcu(&br->router_list)); @@ -277,6 +285,9 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, prev = maybe_deliver(prev, port, skb, __packet_hook); if (IS_ERR(prev)) goto out; + if (prev == port) + br_multicast_count(port->br, port, proto, igmp_type, + BR_MCAST_DIR_TX); if ((unsigned long)lport >= (unsigned long)port) p = rcu_dereference(p->next); diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 8217aec..f2fede0 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -345,8 +345,8 @@ static int find_portno(struct net_bridge *br) static struct net_bridge_port *new_nbp(struct net_bridge *br, struct net_device *dev) { - int index; struct net_bridge_port *p; + int index, err; index = find_portno(br); if (index < 0) @@ -366,7 +366,12 @@ static struct net_bridge_port *new_nbp(struct net_bridge *br, br_init_port(p); br_set_state(p, BR_STATE_DISABLED); br_stp_port_timer_init(p); - br_multicast_add_port(p); + err = br_multicast_add_port(p); + if (err) { + dev_put(dev); + kfree(p); + p = ERR_PTR(err); + } return p; } diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 43d2cd8..786602b 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -60,6 +60,9 @@ static int br_pass_frame_up(struct sk_buff *skb) skb = br_handle_vlan(br, vg, skb); if (!skb) return NET_RX_DROP; + /* update the multicast stats if the packet is IGMP/MLD */ + br_multicast_count(br, NULL, skb->protocol, br_multicast_igmp_type(skb), + BR_MCAST_DIR_TX); return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, dev_net(indev), NULL, skb, indev, NULL, diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 4384414..e405eef 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -361,7 +361,8 @@ out: } static struct sk_buff *br_ip4_multicast_alloc_query(struct net_bridge *br, - __be32 group) + __be32 group, + u8 *igmp_type) { struct sk_buff *skb; struct igmphdr *ih; @@ -411,6 +412,7 @@ static struct sk_buff *br_ip4_multicast_alloc_query(struct net_bridge *br, skb_set_transport_header(skb, skb->len); ih = igmp_hdr(skb); + *igmp_type = IGMP_HOST_MEMBERSHIP_QUERY; ih->type = IGMP_HOST_MEMBERSHIP_QUERY; ih->code = (group ? br->multicast_last_member_interval : br->multicast_query_response_interval) / @@ -428,7 +430,8 @@ out: #if IS_ENABLED(CONFIG_IPV6) static struct sk_buff *br_ip6_multicast_alloc_query(struct net_bridge *br, - const struct in6_addr *group) + const struct in6_addr *grp, + u8 *igmp_type) { struct sk_buff *skb; struct ipv6hdr *ip6h; @@ -487,16 +490,17 @@ static struct sk_buff *br_ip6_multicast_alloc_query(struct net_bridge *br, skb_set_transport_header(skb, skb->len); mldq = (struct mld_msg *) icmp6_hdr(skb); - interval = ipv6_addr_any(group) ? + interval = ipv6_addr_any(grp) ? br->multicast_query_response_interval : br->multicast_last_member_interval; + *igmp_type = ICMPV6_MGM_QUERY; mldq->mld_type = ICMPV6_MGM_QUERY; mldq->mld_code = 0; mldq->mld_cksum = 0; mldq->mld_maxdelay = htons((u16)jiffies_to_msecs(interval)); mldq->mld_reserved = 0; - mldq->mld_mca = *group; + mldq->mld_mca = *grp; /* checksum */ mldq->mld_cksum = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, @@ -513,14 +517,16 @@ out: #endif static struct sk_buff *br_multicast_alloc_query(struct net_bridge *br, - struct br_ip *addr) + struct br_ip *addr, + u8 *igmp_type) { switch (addr->proto) { case htons(ETH_P_IP): - return br_ip4_multicast_alloc_query(br, addr->u.ip4); + return br_ip4_multicast_alloc_query(br, addr->u.ip4, igmp_type); #if IS_ENABLED(CONFIG_IPV6) case htons(ETH_P_IPV6): - return br_ip6_multicast_alloc_query(br, &addr->u.ip6); + return br_ip6_multicast_alloc_query(br, &addr->u.ip6, + igmp_type); #endif } return NULL; @@ -829,18 +835,23 @@ static void __br_multicast_send_query(struct net_bridge *br, struct br_ip *ip) { struct sk_buff *skb; + u8 igmp_type; - skb = br_multicast_alloc_query(br, ip); + skb = br_multicast_alloc_query(br, ip, &igmp_type); if (!skb) return; if (port) { skb->dev = port->dev; + br_multicast_count(br, port, skb->protocol, igmp_type, + BR_MCAST_DIR_TX); NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, dev_net(port->dev), NULL, skb, NULL, skb->dev, br_dev_queue_push_xmit); } else { br_multicast_select_own_querier(br, ip, skb); + br_multicast_count(br, port, skb->protocol, igmp_type, + BR_MCAST_DIR_RX); netif_rx(skb); } } @@ -918,7 +929,7 @@ static void br_ip6_multicast_port_query_expired(unsigned long data) } #endif -void br_multicast_add_port(struct net_bridge_port *port) +int br_multicast_add_port(struct net_bridge_port *port) { port->multicast_router = MDB_RTR_TYPE_TEMP_QUERY; @@ -930,6 +941,11 @@ void br_multicast_add_port(struct net_bridge_port *port) setup_timer(&port->ip6_own_query.timer, br_ip6_multicast_port_query_expired, (unsigned long)port); #endif + port->mcast_stats = netdev_alloc_pcpu_stats(struct bridge_mcast_stats); + if (!port->mcast_stats) + return -ENOMEM; + + return 0; } void br_multicast_del_port(struct net_bridge_port *port) @@ -944,6 +960,7 @@ void br_multicast_del_port(struct net_bridge_port *port) br_multicast_del_pg(br, pg); spin_unlock_bh(&br->multicast_lock); del_timer_sync(&port->multicast_router_timer); + free_percpu(port->mcast_stats); } static void br_multicast_enable(struct bridge_mcast_own_query *query) @@ -1583,6 +1600,39 @@ static void br_ip6_multicast_leave_group(struct net_bridge *br, } #endif +static void br_multicast_err_count(const struct net_bridge *br, + const struct net_bridge_port *p, + __be16 proto) +{ + struct bridge_mcast_stats __percpu *stats; + struct bridge_mcast_stats *pstats; + + if (!br->multicast_stats_enabled) + return; + + if (p) + stats = p->mcast_stats; + else + stats = br->mcast_stats; + if (WARN_ON(!stats)) + return; + + pstats = this_cpu_ptr(stats); + + u64_stats_update_begin(&pstats->syncp); + switch (proto) { + case htons(ETH_P_IP): + pstats->mstats.igmp_parse_errors++; + break; +#if IS_ENABLED(CONFIG_IPV6) + case htons(ETH_P_IPV6): + pstats->mstats.mld_parse_errors++; + break; +#endif + } + u64_stats_update_end(&pstats->syncp); +} + static int br_multicast_ipv4_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, @@ -1599,11 +1649,12 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, BR_INPUT_SKB_CB(skb)->mrouters_only = 1; return 0; } else if (err < 0) { + br_multicast_err_count(br, port, skb->protocol); return err; } - BR_INPUT_SKB_CB(skb)->igmp = 1; ih = igmp_hdr(skb); + BR_INPUT_SKB_CB(skb)->igmp = ih->type; switch (ih->type) { case IGMP_HOST_MEMBERSHIP_REPORT: @@ -1625,6 +1676,9 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, if (skb_trimmed && skb_trimmed != skb) kfree_skb(skb_trimmed); + br_multicast_count(br, port, skb->protocol, BR_INPUT_SKB_CB(skb)->igmp, + BR_MCAST_DIR_RX); + return err; } @@ -1645,11 +1699,12 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, BR_INPUT_SKB_CB(skb)->mrouters_only = 1; return 0; } else if (err < 0) { + br_multicast_err_count(br, port, skb->protocol); return err; } - BR_INPUT_SKB_CB(skb)->igmp = 1; mld = (struct mld_msg *)skb_transport_header(skb); + BR_INPUT_SKB_CB(skb)->igmp = mld->mld_type; switch (mld->mld_type) { case ICMPV6_MGM_REPORT: @@ -1670,6 +1725,9 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, if (skb_trimmed && skb_trimmed != skb) kfree_skb(skb_trimmed); + br_multicast_count(br, port, skb->protocol, BR_INPUT_SKB_CB(skb)->igmp, + BR_MCAST_DIR_RX); + return err; } #endif @@ -1677,6 +1735,8 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, u16 vid) { + int ret = 0; + BR_INPUT_SKB_CB(skb)->igmp = 0; BR_INPUT_SKB_CB(skb)->mrouters_only = 0; @@ -1685,14 +1745,16 @@ int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, switch (skb->protocol) { case htons(ETH_P_IP): - return br_multicast_ipv4_rcv(br, port, skb, vid); + ret = br_multicast_ipv4_rcv(br, port, skb, vid); + break; #if IS_ENABLED(CONFIG_IPV6) case htons(ETH_P_IPV6): - return br_multicast_ipv6_rcv(br, port, skb, vid); + ret = br_multicast_ipv6_rcv(br, port, skb, vid); + break; #endif } - return 0; + return ret; } static void br_multicast_query_expired(struct net_bridge *br, @@ -1831,6 +1893,8 @@ void br_multicast_dev_del(struct net_bridge *br) out: spin_unlock_bh(&br->multicast_lock); + + free_percpu(br->mcast_stats); } int br_multicast_set_router(struct net_bridge *br, unsigned long val) @@ -2185,3 +2249,128 @@ unlock: return ret; } EXPORT_SYMBOL_GPL(br_multicast_has_querier_adjacent); + +static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, + __be16 proto, u8 type, u8 dir) +{ + struct bridge_mcast_stats *pstats = this_cpu_ptr(stats); + + u64_stats_update_begin(&pstats->syncp); + switch (proto) { + case htons(ETH_P_IP): + switch (type) { + case IGMP_HOST_MEMBERSHIP_REPORT: + pstats->mstats.igmp_v1reports[dir]++; + break; + case IGMPV2_HOST_MEMBERSHIP_REPORT: + pstats->mstats.igmp_v2reports[dir]++; + break; + case IGMPV3_HOST_MEMBERSHIP_REPORT: + pstats->mstats.igmp_v3reports[dir]++; + break; + case IGMP_HOST_MEMBERSHIP_QUERY: + pstats->mstats.igmp_queries[dir]++; + break; + case IGMP_HOST_LEAVE_MESSAGE: + pstats->mstats.igmp_leaves[dir]++; + break; + } + break; +#if IS_ENABLED(CONFIG_IPV6) + case htons(ETH_P_IPV6): + switch (type) { + case ICMPV6_MGM_REPORT: + pstats->mstats.mld_v1reports[dir]++; + break; + case ICMPV6_MLD2_REPORT: + pstats->mstats.mld_v2reports[dir]++; + break; + case ICMPV6_MGM_QUERY: + pstats->mstats.mld_queries[dir]++; + break; + case ICMPV6_MGM_REDUCTION: + pstats->mstats.mld_leaves[dir]++; + break; + } + break; +#endif /* CONFIG_IPV6 */ + } + u64_stats_update_end(&pstats->syncp); +} + +void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, + __be16 proto, u8 type, u8 dir) +{ + struct bridge_mcast_stats __percpu *stats; + + /* if multicast_disabled is true then igmp type can't be set */ + if (!type || !br->multicast_stats_enabled) + return; + + if (p) + stats = p->mcast_stats; + else + stats = br->mcast_stats; + if (WARN_ON(!stats)) + return; + + br_mcast_stats_add(stats, proto, type, dir); +} + +int br_multicast_init_stats(struct net_bridge *br) +{ + br->mcast_stats = netdev_alloc_pcpu_stats(struct bridge_mcast_stats); + if (!br->mcast_stats) + return -ENOMEM; + + return 0; +} + +static void mcast_stats_add_dir(u64 *dst, u64 *src) +{ + dst[BR_MCAST_DIR_RX] += src[BR_MCAST_DIR_RX]; + dst[BR_MCAST_DIR_TX] += src[BR_MCAST_DIR_TX]; +} + +void br_multicast_get_stats(const struct net_bridge *br, + const struct net_bridge_port *p, + struct br_mcast_stats *dest) +{ + struct bridge_mcast_stats __percpu *stats; + struct br_mcast_stats tdst; + int i; + + memset(dest, 0, sizeof(*dest)); + if (p) + stats = p->mcast_stats; + else + stats = br->mcast_stats; + if (WARN_ON(!stats)) + return; + + memset(&tdst, 0, sizeof(tdst)); + for_each_possible_cpu(i) { + struct bridge_mcast_stats *cpu_stats = per_cpu_ptr(stats, i); + struct br_mcast_stats temp; + unsigned int start; + + do { + start = u64_stats_fetch_begin_irq(&cpu_stats->syncp); + memcpy(&temp, &cpu_stats->mstats, sizeof(temp)); + } while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start)); + + mcast_stats_add_dir(tdst.igmp_queries, temp.igmp_queries); + mcast_stats_add_dir(tdst.igmp_leaves, temp.igmp_leaves); + mcast_stats_add_dir(tdst.igmp_v1reports, temp.igmp_v1reports); + mcast_stats_add_dir(tdst.igmp_v2reports, temp.igmp_v2reports); + mcast_stats_add_dir(tdst.igmp_v3reports, temp.igmp_v3reports); + tdst.igmp_parse_errors += temp.igmp_parse_errors; + + mcast_stats_add_dir(tdst.mld_queries, temp.mld_queries); + mcast_stats_add_dir(tdst.mld_leaves, temp.mld_leaves); + mcast_stats_add_dir(tdst.mld_v1reports, temp.mld_v1reports); + mcast_stats_add_dir(tdst.mld_v2reports, temp.mld_v2reports); + tdst.mld_parse_errors += temp.mld_parse_errors; + } + memcpy(dest, &tdst, sizeof(*dest)); +} diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index ed75ff9..f2a29e4 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -851,6 +851,7 @@ static const struct nla_policy br_policy[IFLA_BR_MAX + 1] = { [IFLA_BR_NF_CALL_ARPTABLES] = { .type = NLA_U8 }, [IFLA_BR_VLAN_DEFAULT_PVID] = { .type = NLA_U16 }, [IFLA_BR_VLAN_STATS_ENABLED] = { .type = NLA_U8 }, + [IFLA_BR_MCAST_STATS_ENABLED] = { .type = NLA_U8 }, }; static int br_changelink(struct net_device *brdev, struct nlattr *tb[], @@ -1055,6 +1056,13 @@ static int br_changelink(struct net_device *brdev, struct nlattr *tb[], br->multicast_startup_query_interval = clock_t_to_jiffies(val); } + + if (data[IFLA_BR_MCAST_STATS_ENABLED]) { + __u8 mcast_stats; + + mcast_stats = nla_get_u8(data[IFLA_BR_MCAST_STATS_ENABLED]); + br->multicast_stats_enabled = !!mcast_stats; + } #endif #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) if (data[IFLA_BR_NF_CALL_IPTABLES]) { @@ -1110,6 +1118,7 @@ static size_t br_get_size(const struct net_device *brdev) nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_SNOOPING */ nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_QUERY_USE_IFADDR */ nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_QUERIER */ + nla_total_size(sizeof(u8)) + /* IFLA_BR_MCAST_STATS_ENABLED */ nla_total_size(sizeof(u32)) + /* IFLA_BR_MCAST_HASH_ELASTICITY */ nla_total_size(sizeof(u32)) + /* IFLA_BR_MCAST_HASH_MAX */ nla_total_size(sizeof(u32)) + /* IFLA_BR_MCAST_LAST_MEMBER_CNT */ @@ -1187,6 +1196,8 @@ static int br_fill_info(struct sk_buff *skb, const struct net_device *brdev) nla_put_u8(skb, IFLA_BR_MCAST_QUERY_USE_IFADDR, br->multicast_query_use_ifaddr) || nla_put_u8(skb, IFLA_BR_MCAST_QUERIER, br->multicast_querier) || + nla_put_u8(skb, IFLA_BR_MCAST_STATS_ENABLED, + br->multicast_stats_enabled) || nla_put_u32(skb, IFLA_BR_MCAST_HASH_ELASTICITY, br->hash_elasticity) || nla_put_u32(skb, IFLA_BR_MCAST_HASH_MAX, br->hash_max) || @@ -1242,21 +1253,21 @@ static size_t bridge_get_linkxstats_size(const struct net_device *dev) int numvls = 0; vg = br_vlan_group(br); - if (!vg) - return 0; - - /* we need to count all, even placeholder entries */ - list_for_each_entry(v, &vg->vlan_list, vlist) - numvls++; + if (vg) { + /* we need to count all, even placeholder entries */ + list_for_each_entry(v, &vg->vlan_list, vlist) + numvls++; + } - /* account for the vlans and the link xstats type nest attribute */ return numvls * nla_total_size(sizeof(struct bridge_vlan_xstats)) + + nla_total_size(sizeof(struct br_mcast_stats)) + nla_total_size(0); } static size_t brport_get_linkxstats_size(const struct net_device *dev) { - return nla_total_size(0); + return nla_total_size(sizeof(struct br_mcast_stats)) + + nla_total_size(0); } static size_t br_get_linkxstats_size(const struct net_device *dev, int attr) @@ -1280,37 +1291,50 @@ static int bridge_fill_linkxstats(struct sk_buff *skb, int *prividx) { struct net_bridge *br = netdev_priv(dev); + struct nlattr *nla __maybe_unused; struct net_bridge_vlan_group *vg; struct net_bridge_vlan *v; struct nlattr *nest; int vl_idx = 0; - vg = br_vlan_group(br); - if (!vg) - goto out; nest = nla_nest_start(skb, LINK_XSTATS_TYPE_BRIDGE); if (!nest) return -EMSGSIZE; - list_for_each_entry(v, &vg->vlan_list, vlist) { - struct bridge_vlan_xstats vxi; - struct br_vlan_stats stats; - if (++vl_idx < *prividx) - continue; - memset(&vxi, 0, sizeof(vxi)); - vxi.vid = v->vid; - br_vlan_get_stats(v, &stats); - vxi.rx_bytes = stats.rx_bytes; - vxi.rx_packets = stats.rx_packets; - vxi.tx_bytes = stats.tx_bytes; - vxi.tx_packets = stats.tx_packets; - - if (nla_put(skb, BRIDGE_XSTATS_VLAN, sizeof(vxi), &vxi)) + vg = br_vlan_group(br); + if (vg) { + list_for_each_entry(v, &vg->vlan_list, vlist) { + struct bridge_vlan_xstats vxi; + struct br_vlan_stats stats; + + if (++vl_idx < *prividx) + continue; + memset(&vxi, 0, sizeof(vxi)); + vxi.vid = v->vid; + br_vlan_get_stats(v, &stats); + vxi.rx_bytes = stats.rx_bytes; + vxi.rx_packets = stats.rx_packets; + vxi.tx_bytes = stats.tx_bytes; + vxi.tx_packets = stats.tx_packets; + + if (nla_put(skb, BRIDGE_XSTATS_VLAN, sizeof(vxi), &vxi)) + goto nla_put_failure; + } + } + +#ifdef CONFIG_BRIDGE_IGMP_SNOOPING + if (++vl_idx >= *prividx) { + nla = nla_reserve_64bit(skb, BRIDGE_XSTATS_MCAST, + sizeof(struct br_mcast_stats), + BRIDGE_XSTATS_PAD); + if (!nla) goto nla_put_failure; + br_multicast_get_stats(br, NULL, nla_data(nla)); } +#endif nla_nest_end(skb, nest); *prividx = 0; -out: + return 0; nla_put_failure: @@ -1324,11 +1348,26 @@ static int brport_fill_linkxstats(struct sk_buff *skb, const struct net_device *dev, int *prividx) { + struct net_bridge_port *p = br_port_get_rtnl(dev); + struct nlattr *nla __maybe_unused; struct nlattr *nest; + if (!p) + return 0; + nest = nla_nest_start(skb, LINK_XSTATS_TYPE_BRIDGE); if (!nest) return -EMSGSIZE; +#ifdef CONFIG_BRIDGE_IGMP_SNOOPING + nla = nla_reserve_64bit(skb, BRIDGE_XSTATS_MCAST, + sizeof(struct br_mcast_stats), + BRIDGE_XSTATS_PAD); + if (!nla) { + nla_nest_end(skb, nest); + return -EMSGSIZE; + } + br_multicast_get_stats(p->br, p, nla_data(nla)); +#endif nla_nest_end(skb, nest); return 0; diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 52edecf..4dc8511 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -75,6 +75,12 @@ struct bridge_mcast_querier { struct br_ip addr; struct net_bridge_port __rcu *port; }; + +/* IGMP/MLD statistics */ +struct bridge_mcast_stats { + struct br_mcast_stats mstats; + struct u64_stats_sync syncp; +}; #endif struct br_vlan_stats { @@ -229,6 +235,7 @@ struct net_bridge_port struct bridge_mcast_own_query ip6_own_query; #endif /* IS_ENABLED(CONFIG_IPV6) */ unsigned char multicast_router; + struct bridge_mcast_stats __percpu *mcast_stats; struct timer_list multicast_router_timer; struct hlist_head mglist; struct hlist_node rlist; @@ -315,6 +322,7 @@ struct net_bridge u8 multicast_querier:1; u8 multicast_query_use_ifaddr:1; u8 has_ipv6_addr:1; + u8 multicast_stats_enabled:1; u32 hash_elasticity; u32 hash_max; @@ -337,6 +345,7 @@ struct net_bridge struct bridge_mcast_other_query ip4_other_query; struct bridge_mcast_own_query ip4_own_query; struct bridge_mcast_querier ip4_querier; + struct bridge_mcast_stats __percpu *mcast_stats; #if IS_ENABLED(CONFIG_IPV6) struct bridge_mcast_other_query ip6_other_query; struct bridge_mcast_own_query ip6_own_query; @@ -543,7 +552,7 @@ int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, u16 vid); struct net_bridge_mdb_entry *br_mdb_get(struct net_bridge *br, struct sk_buff *skb, u16 vid); -void br_multicast_add_port(struct net_bridge_port *port); +int br_multicast_add_port(struct net_bridge_port *port); void br_multicast_del_port(struct net_bridge_port *port); void br_multicast_enable_port(struct net_bridge_port *port); void br_multicast_disable_port(struct net_bridge_port *port); @@ -576,6 +585,12 @@ void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port, struct br_ip *group, int type, u8 flags); void br_rtr_notify(struct net_device *dev, struct net_bridge_port *port, int type); +void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, + __be16 proto, u8 type, u8 dir); +int br_multicast_init_stats(struct net_bridge *br); +void br_multicast_get_stats(const struct net_bridge *br, + const struct net_bridge_port *p, + struct br_mcast_stats *dest); #define mlock_dereference(X, br) \ rcu_dereference_protected(X, lockdep_is_held(&br->multicast_lock)) @@ -623,6 +638,11 @@ static inline bool br_multicast_querier_exists(struct net_bridge *br, return false; } } + +static inline int br_multicast_igmp_type(const struct sk_buff *skb) +{ + return BR_INPUT_SKB_CB(skb)->igmp; +} #else static inline int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port, @@ -638,8 +658,9 @@ static inline struct net_bridge_mdb_entry *br_mdb_get(struct net_bridge *br, return NULL; } -static inline void br_multicast_add_port(struct net_bridge_port *port) +static inline int br_multicast_add_port(struct net_bridge_port *port) { + return 0; } static inline void br_multicast_del_port(struct net_bridge_port *port) @@ -695,6 +716,22 @@ static inline void br_mdb_init(void) static inline void br_mdb_uninit(void) { } + +static inline void br_multicast_count(struct net_bridge *br, + const struct net_bridge_port *p, + __be16 proto, u8 type, u8 dir) +{ +} + +static inline int br_multicast_init_stats(struct net_bridge *br) +{ + return 0; +} + +static inline int br_multicast_igmp_type(const struct sk_buff *skb) +{ + return 0; +} #endif /* br_vlan.c */ diff --git a/net/bridge/br_sysfs_br.c b/net/bridge/br_sysfs_br.c index beb4707..e120307 100644 --- a/net/bridge/br_sysfs_br.c +++ b/net/bridge/br_sysfs_br.c @@ -618,6 +618,30 @@ static ssize_t multicast_startup_query_interval_store( return store_bridge_parm(d, buf, len, set_startup_query_interval); } static DEVICE_ATTR_RW(multicast_startup_query_interval); + +static ssize_t multicast_stats_enabled_show(struct device *d, + struct device_attribute *attr, + char *buf) +{ + struct net_bridge *br = to_bridge(d); + + return sprintf(buf, "%u\n", br->multicast_stats_enabled); +} + +static int set_stats_enabled(struct net_bridge *br, unsigned long val) +{ + br->multicast_stats_enabled = !!val; + return 0; +} + +static ssize_t multicast_stats_enabled_store(struct device *d, + struct device_attribute *attr, + const char *buf, + size_t len) +{ + return store_bridge_parm(d, buf, len, set_stats_enabled); +} +static DEVICE_ATTR_RW(multicast_stats_enabled); #endif #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) static ssize_t nf_call_iptables_show( @@ -784,6 +808,7 @@ static struct attribute *bridge_attrs[] = { &dev_attr_multicast_query_interval.attr, &dev_attr_multicast_query_response_interval.attr, &dev_attr_multicast_startup_query_interval.attr, + &dev_attr_multicast_stats_enabled.attr, #endif #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) &dev_attr_nf_call_iptables.attr, -- cgit v0.10.2 From fb7caababc024e9086342e8d0aa238565b4a87e4 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Wed, 29 Jun 2016 19:29:24 +0300 Subject: ath10k: fix crash during card removal Usually when the firmware crashes we check for the value 'FW_IND_EVENT_PENDING' in 'FW_INDICATOR_ADDRESS' and proceed with disabling the irq and dumping firmware 'crash dump'. Now when the PCI card is unplugged from the device the PCI controller seems to generate a spurious interrupt after some time which was as treated a firmware crash and resulting in the below race condition (and eventually crashing the system) ath10k_core_unregister -> ath10k_core_free_board_files ...... device unplug spurious interrupt ......... ath10k_pci_taklet -> ath10k_pci_fw_crashed_dump ...etc Clearly even after the firmware board files related data structure is freed up we are getting a spurious interrupt from PCI with 0xfffffff in the 'FW_INDICATOR_ADDRESS' resulting in scheduling of the pci tasklet and doing a crash dump, printing f/w board related info resulting in the below crash. Fix this by detecting this spurious interrupt in ath10k PCI irq handler itself and return IRQ_NONE. Thanks to Michal Kazior for helping us conclude the most appropriate fix. Call trace: EIP is at ath10k_debug_print_board_info+0x39/0xb0 [ath10k_core] EAX: 00000000 EBX: d4de15a0 ECX: 00000000 EDX: 00000064 ESI: f615ddd0 EDI: f8530000 EBP: f615de3c ESP: f615ddbc DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068 CR0: 80050033 CR2: 00000004 CR3: 01c0a000 CR4: 000006f0 Stack: f615ddd0 00000064 f8b4ecdd 00000000 00000000 00412f4e 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 Call Trace: [] ath10k_print_driver_info+0x17/0x30 [ath10k_core] [] ath10k_pci_fw_crashed_dump+0x7a/0xe0 [ath10k_pci] [] ath10k_pci_tasklet+0x70/0x90 [ath10k_pci] [] tasklet_action+0x9e/0xb0 Cc: Michal Kazior Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c index f06dd39..20128fe 100644 --- a/drivers/net/wireless/ath/ath10k/pci.c +++ b/drivers/net/wireless/ath/ath10k/pci.c @@ -2216,6 +2216,14 @@ static void ath10k_pci_fw_crashed_clear(struct ath10k *ar) ath10k_pci_write32(ar, FW_INDICATOR_ADDRESS, val); } +static bool ath10k_pci_has_device_gone(struct ath10k *ar) +{ + u32 val; + + val = ath10k_pci_read32(ar, FW_INDICATOR_ADDRESS); + return (val == 0xffffffff); +} + /* this function effectively clears target memory controller assert line */ static void ath10k_pci_warm_reset_si0(struct ath10k *ar) { @@ -2748,6 +2756,9 @@ static irqreturn_t ath10k_pci_interrupt_handler(int irq, void *arg) struct ath10k_pci *ar_pci = ath10k_pci_priv(ar); int ret; + if (ath10k_pci_has_device_gone(ar)) + return IRQ_NONE; + ret = ath10k_pci_force_wake(ar); if (ret) { ath10k_warn(ar, "failed to wake device up on irq: %d\n", ret); -- cgit v0.10.2 From 569fba2cbb6dd1f4f57231c417accb1abec32493 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Wed, 29 Jun 2016 19:29:24 +0300 Subject: ath10k: remove unneccessary WARN_ON_ONCE in rx during ACS The below warning message seems to hit occasionally with the following combination (IPQ4019 + ACS scan) where we receive packets as a self peer when hostapd does ACS when we bring up AP mode . ath10k has the below fall back mechanism to fetch current operating channel in rx (it will check for the next channel tracking variable if the current one is NULL) [scan channel] --> [rx channel] --> [peer channel] --> [vdev channel] --> [any vdev channel] --> [target oper channel] 'scan channel' and 'target operating channel' are directly fetched from firmware events. All the others should be updated by mac80211. During ACS scan we wouldn't have a valid channel context assigned from mac80211 ('ar->rx_channel'), and also relying on ('ar->scan_channel') is not helpful (it becomes NULL when it goes to BSS channel and also when the scan event is completed). In short we cannot always rely on these two channel tracking variables. 'Target Operating Channel' (ar->tgt_oper_chan) seems to keep track of the current operating even while we are doing ACS scan and etc. Hence remove this un-necessary warning message and continue with target_operating channel. At the worst case scenario when the target operating channel is invalid (NULL) we already have an ath10k warning message to notify we really don't have a proper channel configured in rx to update the rx status("no channel configured; ignoring frame(s)!") WARNING: CPU: 0 PID: 0 at ath/ath10k/htt_rx.c:803 [] (warn_slowpath_null) from [] (ath10k_htt_rx_h_channel+0xe0/0x1b8 [ath10k_core]) [] (ath10k_htt_rx_h_channel [ath10k_core]) from [] (ath10k_htt_rx_h_ppdu+0x80/0x288 [ath10k_core]) [] (ath10k_htt_rx_h_ppdu [ath10k_core]) from [] (ath10k_htt_txrx_compl_task+0x724/0x9d4 [ath10k_core]) [] (ath10k_htt_txrx_compl_task [ath10k_core]) Fixes:3b0499e9ce42 ("ath10k: reduce warning messages during rx without proper channel context") Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c index 3b35c7a..8bae023 100644 --- a/drivers/net/wireless/ath/ath10k/htt_rx.c +++ b/drivers/net/wireless/ath/ath10k/htt_rx.c @@ -748,7 +748,7 @@ ath10k_htt_rx_h_peer_channel(struct ath10k *ar, struct htt_rx_desc *rxd) if (WARN_ON_ONCE(!arvif)) return NULL; - if (WARN_ON_ONCE(ath10k_mac_vif_chan(arvif->vif, &def))) + if (ath10k_mac_vif_chan(arvif->vif, &def)) return NULL; return def.chan; -- cgit v0.10.2 From a66cd733a729167567afd35bca8c3e3a3ace04f6 Mon Sep 17 00:00:00 2001 From: Bob Copeland Date: Wed, 29 Jun 2016 19:29:25 +0300 Subject: ath10k: fix potential null dereference bugs Smatch warns about a number of cases in ath10k where a pointer is null-checked after it has already been dereferenced, in code involving ath10k private virtual interface pointers. Fix these by making the dereference happen later. Addresses the following smatch warnings: drivers/net/wireless/ath/ath10k/mac.c:3651 ath10k_mac_txq_init() warn: variable dereferenced before check 'txq' (see line 3649) drivers/net/wireless/ath/ath10k/mac.c:3664 ath10k_mac_txq_unref() warn: variable dereferenced before check 'txq' (see line 3659) drivers/net/wireless/ath/ath10k/htt_tx.c:70 __ath10k_htt_tx_txq_recalc() warn: variable dereferenced before check 'txq->sta' (see line 52) drivers/net/wireless/ath/ath10k/htt_tx.c:740 ath10k_htt_tx_get_vdev_id() warn: variable dereferenced before check 'cb->vif' (see line 736) drivers/net/wireless/ath/ath10k/txrx.c:86 ath10k_txrx_tx_unref() warn: variable dereferenced before check 'txq' (see line 84) drivers/net/wireless/ath/ath10k/wmi.c:1837 ath10k_wmi_op_gen_mgmt_tx() warn: variable dereferenced before check 'cb->vif' (see line 1825) Signed-off-by: Bob Copeland Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c index 6269c61..ae5b33f 100644 --- a/drivers/net/wireless/ath/ath10k/htt_tx.c +++ b/drivers/net/wireless/ath/ath10k/htt_tx.c @@ -49,7 +49,7 @@ static void __ath10k_htt_tx_txq_recalc(struct ieee80211_hw *hw, struct ieee80211_txq *txq) { struct ath10k *ar = hw->priv; - struct ath10k_sta *arsta = (void *)txq->sta->drv_priv; + struct ath10k_sta *arsta; struct ath10k_vif *arvif = (void *)txq->vif->drv_priv; unsigned long frame_cnt; unsigned long byte_cnt; @@ -67,10 +67,12 @@ static void __ath10k_htt_tx_txq_recalc(struct ieee80211_hw *hw, if (ar->htt.tx_q_state.mode != HTT_TX_MODE_SWITCH_PUSH_PULL) return; - if (txq->sta) + if (txq->sta) { + arsta = (void *)txq->sta->drv_priv; peer_id = arsta->peer_id; - else + } else { peer_id = arvif->peer_id; + } tid = txq->tid; bit = BIT(peer_id % 32); @@ -733,16 +735,18 @@ static u8 ath10k_htt_tx_get_vdev_id(struct ath10k *ar, struct sk_buff *skb) { struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); struct ath10k_skb_cb *cb = ATH10K_SKB_CB(skb); - struct ath10k_vif *arvif = (void *)cb->vif->drv_priv; + struct ath10k_vif *arvif; - if (info->flags & IEEE80211_TX_CTL_TX_OFFCHAN) + if (info->flags & IEEE80211_TX_CTL_TX_OFFCHAN) { return ar->scan.vdev_id; - else if (cb->vif) + } else if (cb->vif) { + arvif = (void *)cb->vif->drv_priv; return arvif->vdev_id; - else if (ar->monitor_started) + } else if (ar->monitor_started) { return ar->monitor_vdev_id; - else + } else { return 0; + } } static u8 ath10k_htt_tx_get_tid(struct sk_buff *skb, bool is_eth) diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index 3a170b1..3560ae9 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -3675,17 +3675,18 @@ void ath10k_mgmt_over_wmi_tx_work(struct work_struct *work) static void ath10k_mac_txq_init(struct ieee80211_txq *txq) { - struct ath10k_txq *artxq = (void *)txq->drv_priv; + struct ath10k_txq *artxq; if (!txq) return; + artxq = (void *)txq->drv_priv; INIT_LIST_HEAD(&artxq->list); } static void ath10k_mac_txq_unref(struct ath10k *ar, struct ieee80211_txq *txq) { - struct ath10k_txq *artxq = (void *)txq->drv_priv; + struct ath10k_txq *artxq; struct ath10k_skb_cb *cb; struct sk_buff *msdu; int msdu_id; @@ -3693,6 +3694,7 @@ static void ath10k_mac_txq_unref(struct ath10k *ar, struct ieee80211_txq *txq) if (!txq) return; + artxq = (void *)txq->drv_priv; spin_lock_bh(&ar->txqs_lock); if (!list_empty(&artxq->list)) list_del_init(&artxq->list); diff --git a/drivers/net/wireless/ath/ath10k/txrx.c b/drivers/net/wireless/ath/ath10k/txrx.c index 1966c78..f524ac0 100644 --- a/drivers/net/wireless/ath/ath10k/txrx.c +++ b/drivers/net/wireless/ath/ath10k/txrx.c @@ -81,10 +81,11 @@ int ath10k_txrx_tx_unref(struct ath10k_htt *htt, skb_cb = ATH10K_SKB_CB(msdu); txq = skb_cb->txq; - artxq = (void *)txq->drv_priv; - if (txq) + if (txq) { + artxq = (void *)txq->drv_priv; artxq->num_fw_queued--; + } ath10k_htt_tx_free_msdu_id(htt, tx_done->msdu_id); ath10k_htt_tx_dec_pending(htt); diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c index 6279ab4..4ecf3b3 100644 --- a/drivers/net/wireless/ath/ath10k/wmi.c +++ b/drivers/net/wireless/ath/ath10k/wmi.c @@ -1826,7 +1826,7 @@ static struct sk_buff * ath10k_wmi_op_gen_mgmt_tx(struct ath10k *ar, struct sk_buff *msdu) { struct ath10k_skb_cb *cb = ATH10K_SKB_CB(msdu); - struct ath10k_vif *arvif = (void *)cb->vif->drv_priv; + struct ath10k_vif *arvif; struct wmi_mgmt_tx_cmd *cmd; struct ieee80211_hdr *hdr; struct sk_buff *skb; @@ -1838,10 +1838,12 @@ ath10k_wmi_op_gen_mgmt_tx(struct ath10k *ar, struct sk_buff *msdu) hdr = (struct ieee80211_hdr *)msdu->data; fc = le16_to_cpu(hdr->frame_control); - if (cb->vif) + if (cb->vif) { + arvif = (void *)cb->vif->drv_priv; vdev_id = arvif->vdev_id; - else + } else { vdev_id = 0; + } if (WARN_ON_ONCE(!ieee80211_is_mgmt(hdr->frame_control))) return ERR_PTR(-EINVAL); -- cgit v0.10.2 From 343bf960f03c982a80dbf5ee3af58a7734a3f1d2 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Wed, 29 Jun 2016 19:29:26 +0300 Subject: ath10k: enable beacon loss detection support for 10.4 Enable beacon loss detection support for 10.4 by handling roam event. With this change QCA99X0 station is able to detect beacon loss when the AP is powered off Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c index 4ecf3b3..12a4347 100644 --- a/drivers/net/wireless/ath/ath10k/wmi.c +++ b/drivers/net/wireless/ath/ath10k/wmi.c @@ -5259,6 +5259,9 @@ static void ath10k_wmi_10_4_op_rx(struct ath10k *ar, struct sk_buff *skb) case WMI_10_4_PEER_STA_KICKOUT_EVENTID: ath10k_wmi_event_peer_sta_kickout(ar, skb); break; + case WMI_10_4_ROAM_EVENTID: + ath10k_wmi_event_roam(ar, skb); + break; case WMI_10_4_HOST_SWBA_EVENTID: ath10k_wmi_event_host_swba(ar, skb); break; @@ -7905,6 +7908,7 @@ static const struct wmi_ops wmi_10_4_ops = { .pull_phyerr = ath10k_wmi_10_4_op_pull_phyerr_ev, .pull_svc_rdy = ath10k_wmi_main_op_pull_svc_rdy_ev, .pull_rdy = ath10k_wmi_op_pull_rdy_ev, + .pull_roam_ev = ath10k_wmi_op_pull_roam_ev, .get_txbf_conf_scheme = ath10k_wmi_10_4_txbf_conf_scheme, .gen_pdev_suspend = ath10k_wmi_op_gen_pdev_suspend, -- cgit v0.10.2 From 34663241d81f9d2a8e3730f3f3973adfcb913b83 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Wed, 29 Jun 2016 19:29:27 +0300 Subject: ath10k: disable TX_STBC for tx chainmask of 1 Disable TX_STBC for both HT and VHT if the devices tx chainmask is '1' TX_STBC is required only for devices with tx_chainmask > 1. This fixes a ping failure for QCA9887 (1x1) in HT/VHT mode Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index 3560ae9..e9d464c 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -4226,6 +4226,9 @@ static struct ieee80211_sta_vht_cap ath10k_create_vht_cap(struct ath10k *ar) mcs_map |= IEEE80211_VHT_MCS_NOT_SUPPORTED << (i * 2); } + if (ar->cfg_tx_chainmask <= 1) + vht_cap.cap &= ~IEEE80211_VHT_CAP_TXSTBC; + vht_cap.vht_mcs.rx_mcs_map = cpu_to_le16(mcs_map); vht_cap.vht_mcs.tx_mcs_map = cpu_to_le16(mcs_map); @@ -4263,7 +4266,7 @@ static struct ieee80211_sta_ht_cap ath10k_get_ht_cap(struct ath10k *ar) ht_cap.cap |= smps; } - if (ar->ht_cap_info & WMI_HT_CAP_TX_STBC) + if (ar->ht_cap_info & WMI_HT_CAP_TX_STBC && (ar->cfg_tx_chainmask > 1)) ht_cap.cap |= IEEE80211_HT_CAP_TX_STBC; if (ar->ht_cap_info & WMI_HT_CAP_RX_STBC) { -- cgit v0.10.2 From 3fa35bacc16a01b4acfa4a2a34947f84b1237056 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Wed, 29 Jun 2016 19:29:29 +0300 Subject: ath10k: fix some typo in spectral code commments Found this obvious typo while going through the spectral code design in ath10k Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/spectral.c b/drivers/net/wireless/ath/ath10k/spectral.c index 4671cfb..7d9b0da 100644 --- a/drivers/net/wireless/ath/ath10k/spectral.c +++ b/drivers/net/wireless/ath/ath10k/spectral.c @@ -101,9 +101,9 @@ int ath10k_spectral_process_fft(struct ath10k *ar, break; case 80: /* TODO: As experiments with an analogue sender and various - * configuaritions (fft-sizes of 64/128/256 and 20/40/80 Mhz) + * configurations (fft-sizes of 64/128/256 and 20/40/80 Mhz) * show, the particular configuration of 80 MHz/64 bins does - * not match with the other smaples at all. Until the reason + * not match with the other samples at all. Until the reason * for that is found, don't report these samples. */ if (bin_len == 64) -- cgit v0.10.2 From 0f27ac40fb64dc0ac8254913b414dde3386196f4 Mon Sep 17 00:00:00 2001 From: Eduardo Abinader Date: Wed, 29 Jun 2016 19:29:29 +0300 Subject: ath9k: return false when reading wrong eeprom offset Just setting the proper return for reading beyond the eeprom data. Signed-off-by: Eduardo Abinader Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/pci.c b/drivers/net/wireless/ath/ath9k/pci.c index 7cdaf40..aa04b13 100644 --- a/drivers/net/wireless/ath/ath9k/pci.c +++ b/drivers/net/wireless/ath/ath9k/pci.c @@ -794,6 +794,8 @@ static bool ath_pci_eeprom_read(struct ath_common *common, u32 off, u16 *data) ath_err(common, "%s: eeprom read failed, offset %08x is out of range\n", __func__, off); + + return false; } *data = pdata->eeprom_data[off]; -- cgit v0.10.2 From aaab50fcea78ae3414c3afc25aae8d0603df34d0 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 29 Jun 2016 19:29:30 +0300 Subject: ath9k: Fix programming of minCCA power threshold The function ar9003_hw_apply_minccapwr_thresh takes as second parameter not a pointer to the channel but a boolean value describing whether the channel is 2.4GHz or not. This broke (according to the origin commit) the ETSI regulatory compliance on 5GHz channels. Fixes: 3533bf6b15a0 ("ath9k: Fix regulatory compliance") Signed-off-by: Sven Eckelmann Cc: Simon Wunderlich Cc: Sujith Manoharan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c b/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c index d0224fc..5bd2cba 100644 --- a/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c +++ b/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c @@ -4175,7 +4175,7 @@ static void ath9k_hw_ar9300_set_board_values(struct ath_hw *ah, if (!AR_SREV_9330(ah) && !AR_SREV_9340(ah) && !AR_SREV_9531(ah)) ar9003_hw_internal_regulator_apply(ah); ar9003_hw_apply_tuning_caps(ah); - ar9003_hw_apply_minccapwr_thresh(ah, chan); + ar9003_hw_apply_minccapwr_thresh(ah, is2ghz); ar9003_hw_txend_to_xpa_off_apply(ah, is2ghz); ar9003_hw_thermometer_apply(ah); ar9003_hw_thermo_cal_apply(ah); -- cgit v0.10.2 From b1ed4c4fa9a5ccf325184fd90edc50978ef6e33a Mon Sep 17 00:00:00 2001 From: Andrey Vagin Date: Mon, 27 Jun 2016 15:33:56 -0700 Subject: tcp: add an ability to dump and restore window parameters We found that sometimes a restored tcp socket doesn't work. A reason of this bug is incorrect window parameters and in this case tcp_acceptable_seq() returns tcp_wnd_end(tp) instead of tp->snd_nxt. The other side drops packets with this seq, because seq is less than tp->rcv_nxt ( tcp_sequence() ). Data from a send queue is sent only if there is enough space in a window, so when we restore unacked data, we need to expand a window to fit this data. This was in a first version of this patch: "tcp: extend window to fit all restored unacked data in a send queue" Then Alexey recommended me to restore window parameters instead of adjusted them according with data in a sent queue. This sounds resonable. rcv_wnd has to be restored, because it was reported to another side and the offered window is never shrunk. One of reasons why we need to restore snd_wnd was described above. Cc: Pavel Emelyanov Cc: "David S. Miller" Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Signed-off-by: Andrey Vagin Signed-off-by: David S. Miller diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h index 53e8e3f..482898f 100644 --- a/include/uapi/linux/tcp.h +++ b/include/uapi/linux/tcp.h @@ -115,12 +115,22 @@ enum { #define TCP_CC_INFO 26 /* Get Congestion Control (optional) info */ #define TCP_SAVE_SYN 27 /* Record SYN headers for new connections */ #define TCP_SAVED_SYN 28 /* Get SYN headers recorded for connection */ +#define TCP_REPAIR_WINDOW 29 /* Get/set window parameters */ struct tcp_repair_opt { __u32 opt_code; __u32 opt_val; }; +struct tcp_repair_window { + __u32 snd_wl1; + __u32 snd_wnd; + __u32 max_window; + + __u32 rcv_wnd; + __u32 rcv_wup; +}; + enum { TCP_NO_QUEUE, TCP_RECV_QUEUE, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 5c7ed14..108ef2a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2277,6 +2277,38 @@ static inline bool tcp_can_repair_sock(const struct sock *sk) ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_ESTABLISHED)); } +static int tcp_repair_set_window(struct tcp_sock *tp, char __user *optbuf, int len) +{ + struct tcp_repair_window opt; + + if (!tp->repair) + return -EPERM; + + if (len != sizeof(opt)) + return -EINVAL; + + if (copy_from_user(&opt, optbuf, sizeof(opt))) + return -EFAULT; + + if (opt.max_window < opt.snd_wnd) + return -EINVAL; + + if (after(opt.snd_wl1, tp->rcv_nxt + opt.rcv_wnd)) + return -EINVAL; + + if (after(opt.rcv_wup, tp->rcv_nxt)) + return -EINVAL; + + tp->snd_wl1 = opt.snd_wl1; + tp->snd_wnd = opt.snd_wnd; + tp->max_window = opt.max_window; + + tp->rcv_wnd = opt.rcv_wnd; + tp->rcv_wup = opt.rcv_wup; + + return 0; +} + static int tcp_repair_options_est(struct tcp_sock *tp, struct tcp_repair_opt __user *optbuf, unsigned int len) { @@ -2604,6 +2636,9 @@ static int do_tcp_setsockopt(struct sock *sk, int level, else tp->tsoffset = val - tcp_time_stamp; break; + case TCP_REPAIR_WINDOW: + err = tcp_repair_set_window(tp, optval, optlen); + break; case TCP_NOTSENT_LOWAT: tp->notsent_lowat = val; sk->sk_write_space(sk); @@ -2860,6 +2895,28 @@ static int do_tcp_getsockopt(struct sock *sk, int level, return -EINVAL; break; + case TCP_REPAIR_WINDOW: { + struct tcp_repair_window opt; + + if (get_user(len, optlen)) + return -EFAULT; + + if (len != sizeof(opt)) + return -EINVAL; + + if (!tp->repair) + return -EPERM; + + opt.snd_wl1 = tp->snd_wl1; + opt.snd_wnd = tp->snd_wnd; + opt.max_window = tp->max_window; + opt.rcv_wnd = tp->rcv_wnd; + opt.rcv_wup = tp->rcv_wup; + + if (copy_to_user(optval, &opt, len)) + return -EFAULT; + return 0; + } case TCP_QUEUE_SEQ: if (tp->repair_queue == TCP_SEND_QUEUE) val = tp->write_seq; -- cgit v0.10.2 From 2631b79f6cb8b634fe41b77de9c4add0ec6b3cae Mon Sep 17 00:00:00 2001 From: "Seymour, Shane M" Date: Tue, 28 Jun 2016 23:06:48 +0000 Subject: tcp: increase size at which tcp_bound_to_half_wnd bounds to > TCP_MSS_DEFAULT In previous commit 01f83d69844d307be2aa6fea88b0e8fe5cbdb2f4 the following comments were added: "When peer uses tiny windows, there is no use in packetizing to sub-MSS pieces for the sake of SWS or making sure there are enough packets in the pipe for fast recovery." The test should be > TCP_MSS_DEFAULT not >= 512. This allows low end devices that send an MSS of 536 (TCP_MSS_DEFAULT) to see better network performance by sending it 536 bytes of data at a time instead of bounding to half window size (268). Other network stacks work this way, e.g. HP-UX. Signed-off-by: Shane Seymour Signed-off-by: David S. Miller diff --git a/include/net/tcp.h b/include/net/tcp.h index a79894b..d825858 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -589,7 +589,7 @@ static inline int tcp_bound_to_half_wnd(struct tcp_sock *tp, int pktsize) * On the other hand, for extremely large MSS devices, handling * smaller than MSS windows in this way does make sense. */ - if (tp->max_window >= 512) + if (tp->max_window > TCP_MSS_DEFAULT) cutoff = (tp->max_window >> 1); else cutoff = tp->max_window; -- cgit v0.10.2 From 153380ec4b9b6802bba61ebd34da432a54994e9d Mon Sep 17 00:00:00 2001 From: Mateusz Bajorski Date: Wed, 29 Jun 2016 09:22:10 +0200 Subject: fib_rules: Added NLM_F_EXCL support to fib_nl_newrule When adding rule with NLM_F_EXCL flag then check if the same rule exist. If yes then exit with -EEXIST. This is already implemented in iproute2: if (cmd == RTM_NEWRULE) { req.n.nlmsg_flags |= NLM_F_CREATE|NLM_F_EXCL; req.r.rtm_type = RTN_UNICAST; } Tested ipv4 and ipv6 with net-next linux on qemu x86 expected behavior after patch: localhost ~ # ip rule 0: from all lookup local 32766: from all lookup main 32767: from all lookup default localhost ~ # ip rule add from 10.46.177.97 lookup 104 pref 1005 localhost ~ # ip rule add from 10.46.177.97 lookup 104 pref 1005 RTNETLINK answers: File exists localhost ~ # ip rule 0: from all lookup local 1005: from 10.46.177.97 lookup 104 32766: from all lookup main 32767: from all lookup default There was already topic regarding this but I don't see any changes merged and problem still occurs. https://lkml.kernel.org/r/1135778809.5944.7.camel+%28%29+localhost+%21+localdomain Signed-off-by: Mateusz Bajorski Acked-by: David Ahern Signed-off-by: David S. Miller diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 98298b1..be4629c 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -269,6 +269,49 @@ errout: return err; } +static int rule_exists(struct fib_rules_ops *ops, struct fib_rule_hdr *frh, + struct nlattr **tb, struct fib_rule *rule) +{ + struct fib_rule *r; + + list_for_each_entry(r, &ops->rules_list, list) { + if (r->action != rule->action) + continue; + + if (r->table != rule->table) + continue; + + if (r->pref != rule->pref) + continue; + + if (memcmp(r->iifname, rule->iifname, IFNAMSIZ)) + continue; + + if (memcmp(r->oifname, rule->oifname, IFNAMSIZ)) + continue; + + if (r->mark != rule->mark) + continue; + + if (r->mark_mask != rule->mark_mask) + continue; + + if (r->tun_id != rule->tun_id) + continue; + + if (r->fr_net != rule->fr_net) + continue; + + if (r->l3mdev != rule->l3mdev) + continue; + + if (!ops->compare(r, frh, tb)) + continue; + return 1; + } + return 0; +} + int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh) { struct net *net = sock_net(skb->sk); @@ -386,6 +429,12 @@ int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh) if (rule->l3mdev && rule->table) goto errout_free; + if ((nlh->nlmsg_flags & NLM_F_EXCL) && + rule_exists(ops, frh, tb, rule)) { + err = -EEXIST; + goto errout_free; + } + err = ops->configure(rule, skb, frh, tb); if (err < 0) goto errout_free; -- cgit v0.10.2 From 6e6edd8b96e859bcbb3fd03eafd76ad306026b21 Mon Sep 17 00:00:00 2001 From: John Crispin Date: Wed, 29 Jun 2016 13:38:08 +0200 Subject: net-next: mediatek: remove superfluous register reads The driver was originally written for MIPS based SoC. These required the IRQ mask register to be read after writing it to ensure that the content was actually applied. As this version only works on ARM based SoCs, we can safely remove the 2 reads as they are no longer required. Signed-off-by: John Crispin Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index d1cdc2d..5a3a4e9 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -332,8 +332,6 @@ static inline void mtk_irq_disable(struct mtk_eth *eth, u32 mask) val = mtk_r32(eth, MTK_QDMA_INT_MASK); mtk_w32(eth, val & ~mask, MTK_QDMA_INT_MASK); - /* flush write */ - mtk_r32(eth, MTK_QDMA_INT_MASK); } static inline void mtk_irq_enable(struct mtk_eth *eth, u32 mask) @@ -342,8 +340,6 @@ static inline void mtk_irq_enable(struct mtk_eth *eth, u32 mask) val = mtk_r32(eth, MTK_QDMA_INT_MASK); mtk_w32(eth, val | mask, MTK_QDMA_INT_MASK); - /* flush write */ - mtk_r32(eth, MTK_QDMA_INT_MASK); } static int mtk_set_mac_address(struct net_device *dev, void *p) -- cgit v0.10.2 From eece71e8fb0953e18e3ac631b4414bb21af924a7 Mon Sep 17 00:00:00 2001 From: John Crispin Date: Wed, 29 Jun 2016 13:38:09 +0200 Subject: net-next: mediatek: don't use intermediate variables to store IRQ masks The code currently uses variables to store and never modify the bit masks of interrupts. This is legacy code from an early version of the driver that supported MIPS based SoCs where the IRQ bits depended on the actual SoC. As the bits are the same for all ARM based SoCs using this driver we can remove the intermediate variables. Signed-off-by: John Crispin Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 5a3a4e9..d6c3a17 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -794,7 +794,7 @@ drop: } static int mtk_poll_rx(struct napi_struct *napi, int budget, - struct mtk_eth *eth, u32 rx_intr) + struct mtk_eth *eth) { struct mtk_rx_ring *ring = ð->rx_ring; int idx = ring->calc_idx; @@ -882,7 +882,7 @@ release_desc: } if (done < budget) - mtk_w32(eth, rx_intr, MTK_QMTK_INT_STATUS); + mtk_w32(eth, MTK_RX_DONE_INT, MTK_QMTK_INT_STATUS); return done; } @@ -967,28 +967,26 @@ static int mtk_poll_tx(struct mtk_eth *eth, int budget, bool *tx_again) static int mtk_poll(struct napi_struct *napi, int budget) { struct mtk_eth *eth = container_of(napi, struct mtk_eth, rx_napi); - u32 status, status2, mask, tx_intr, rx_intr, status_intr; + u32 status, status2, mask; int tx_done, rx_done; bool tx_again = false; status = mtk_r32(eth, MTK_QMTK_INT_STATUS); status2 = mtk_r32(eth, MTK_INT_STATUS2); - tx_intr = MTK_TX_DONE_INT; - rx_intr = MTK_RX_DONE_INT; - status_intr = (MTK_GDM1_AF | MTK_GDM2_AF); tx_done = 0; rx_done = 0; tx_again = 0; - if (status & tx_intr) + if (status & MTK_TX_DONE_INT) tx_done = mtk_poll_tx(eth, budget, &tx_again); - if (status & rx_intr) - rx_done = mtk_poll_rx(napi, budget, eth, rx_intr); + if (status & MTK_RX_DONE_INT) + rx_done = mtk_poll_rx(napi, budget, eth); - if (unlikely(status2 & status_intr)) { + if (unlikely(status2 & (MTK_GDM1_AF | MTK_GDM2_AF))) { mtk_stats_update(eth); - mtk_w32(eth, status_intr, MTK_INT_STATUS2); + mtk_w32(eth, (MTK_GDM1_AF | MTK_GDM2_AF), + MTK_INT_STATUS2); } if (unlikely(netif_msg_intr(eth))) { @@ -1006,7 +1004,7 @@ static int mtk_poll(struct napi_struct *napi, int budget) return budget; napi_complete(napi); - mtk_irq_enable(eth, tx_intr | rx_intr); + mtk_irq_enable(eth, MTK_RX_DONE_INT | MTK_RX_DONE_INT); return rx_done; } -- cgit v0.10.2 From 7bc9ccec34f5260bf698de5d5c0366b6c6ef6c9e Mon Sep 17 00:00:00 2001 From: John Crispin Date: Wed, 29 Jun 2016 13:38:10 +0200 Subject: net-next: mediatek: add IRQ locking The code that enables and disables IRQs is missing proper locking. After adding the IRQ grouping patch and routing the RX and TX IRQs to different cores we experienced IRQ stalls. Fix this by adding proper locking. We use a dedicated lock to reduce the latency if the IRQ code. Signed-off-by: John Crispin Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index d6c3a17..698caba 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -328,18 +328,24 @@ static void mtk_mdio_cleanup(struct mtk_eth *eth) static inline void mtk_irq_disable(struct mtk_eth *eth, u32 mask) { + unsigned long flags; u32 val; + spin_lock_irqsave(ð->irq_lock, flags); val = mtk_r32(eth, MTK_QDMA_INT_MASK); mtk_w32(eth, val & ~mask, MTK_QDMA_INT_MASK); + spin_unlock_irqrestore(ð->irq_lock, flags); } static inline void mtk_irq_enable(struct mtk_eth *eth, u32 mask) { + unsigned long flags; u32 val; + spin_lock_irqsave(ð->irq_lock, flags); val = mtk_r32(eth, MTK_QDMA_INT_MASK); mtk_w32(eth, val | mask, MTK_QDMA_INT_MASK); + spin_unlock_irqrestore(ð->irq_lock, flags); } static int mtk_set_mac_address(struct net_device *dev, void *p) @@ -1760,6 +1766,7 @@ static int mtk_probe(struct platform_device *pdev) return PTR_ERR(eth->base); spin_lock_init(ð->page_lock); + spin_lock_init(ð->irq_lock); eth->ethsys = syscon_regmap_lookup_by_phandle(pdev->dev.of_node, "mediatek,ethsys"); diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h index a5eb7c6..3159d2a 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h @@ -373,6 +373,7 @@ struct mtk_eth { void __iomem *base; struct reset_control *rstc; spinlock_t page_lock; + spinlock_t irq_lock; struct net_device dummy_dev; struct net_device *netdev[MTK_MAX_DEVS]; struct mtk_mac *mac[MTK_MAX_DEVS]; -- cgit v0.10.2 From 8067302973a1959605af1124e828cbb3e66f2a3c Mon Sep 17 00:00:00 2001 From: John Crispin Date: Wed, 29 Jun 2016 13:38:11 +0200 Subject: net-next: mediatek: add support for IRQ grouping The ethernet core has 3 IRQs. Using the IRQ grouping registers we are able to separate TX and RX IRQs, which allows us to service them on separate cores. This patch splits the IRQ handler into 2 separate functions, one for TX and another for RX. The TX housekeeping is split out into its own NAPI handler. Signed-off-by: John Crispin Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 698caba..7056a37 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -893,17 +893,17 @@ release_desc: return done; } -static int mtk_poll_tx(struct mtk_eth *eth, int budget, bool *tx_again) +static int mtk_poll_tx(struct mtk_eth *eth, int budget) { struct mtk_tx_ring *ring = ð->tx_ring; struct mtk_tx_dma *desc; struct sk_buff *skb; struct mtk_tx_buf *tx_buf; - int total = 0, done[MTK_MAX_DEVS]; + unsigned int done[MTK_MAX_DEVS]; unsigned int bytes[MTK_MAX_DEVS]; u32 cpu, dma; static int condition; - int i; + int total = 0, i; memset(done, 0, sizeof(done)); memset(bytes, 0, sizeof(bytes)); @@ -954,15 +954,6 @@ static int mtk_poll_tx(struct mtk_eth *eth, int budget, bool *tx_again) total += done[i]; } - /* read hw index again make sure no new tx packet */ - if (cpu != dma || cpu != mtk_r32(eth, MTK_QTX_DRX_PTR)) - *tx_again = true; - else - mtk_w32(eth, MTK_TX_DONE_INT, MTK_QMTK_INT_STATUS); - - if (!total) - return 0; - if (mtk_queue_stopped(eth) && (atomic_read(&ring->free_count) > ring->thresh)) mtk_wake_queue(eth); @@ -970,47 +961,75 @@ static int mtk_poll_tx(struct mtk_eth *eth, int budget, bool *tx_again) return total; } -static int mtk_poll(struct napi_struct *napi, int budget) +static void mtk_handle_status_irq(struct mtk_eth *eth) { - struct mtk_eth *eth = container_of(napi, struct mtk_eth, rx_napi); - u32 status, status2, mask; - int tx_done, rx_done; - bool tx_again = false; - - status = mtk_r32(eth, MTK_QMTK_INT_STATUS); - status2 = mtk_r32(eth, MTK_INT_STATUS2); - tx_done = 0; - rx_done = 0; - tx_again = 0; - - if (status & MTK_TX_DONE_INT) - tx_done = mtk_poll_tx(eth, budget, &tx_again); - - if (status & MTK_RX_DONE_INT) - rx_done = mtk_poll_rx(napi, budget, eth); + u32 status2 = mtk_r32(eth, MTK_INT_STATUS2); if (unlikely(status2 & (MTK_GDM1_AF | MTK_GDM2_AF))) { mtk_stats_update(eth); mtk_w32(eth, (MTK_GDM1_AF | MTK_GDM2_AF), MTK_INT_STATUS2); } +} + +static int mtk_napi_tx(struct napi_struct *napi, int budget) +{ + struct mtk_eth *eth = container_of(napi, struct mtk_eth, tx_napi); + u32 status, mask; + int tx_done = 0; + + mtk_handle_status_irq(eth); + mtk_w32(eth, MTK_TX_DONE_INT, MTK_QMTK_INT_STATUS); + tx_done = mtk_poll_tx(eth, budget); + + if (unlikely(netif_msg_intr(eth))) { + status = mtk_r32(eth, MTK_QMTK_INT_STATUS); + mask = mtk_r32(eth, MTK_QDMA_INT_MASK); + dev_info(eth->dev, + "done tx %d, intr 0x%08x/0x%x\n", + tx_done, status, mask); + } + + if (tx_done == budget) + return budget; + + status = mtk_r32(eth, MTK_QMTK_INT_STATUS); + if (status & MTK_TX_DONE_INT) + return budget; + + napi_complete(napi); + mtk_irq_enable(eth, MTK_TX_DONE_INT); + + return tx_done; +} + +static int mtk_napi_rx(struct napi_struct *napi, int budget) +{ + struct mtk_eth *eth = container_of(napi, struct mtk_eth, rx_napi); + u32 status, mask; + int rx_done = 0; + + mtk_handle_status_irq(eth); + mtk_w32(eth, MTK_RX_DONE_INT, MTK_QMTK_INT_STATUS); + rx_done = mtk_poll_rx(napi, budget, eth); if (unlikely(netif_msg_intr(eth))) { + status = mtk_r32(eth, MTK_QMTK_INT_STATUS); mask = mtk_r32(eth, MTK_QDMA_INT_MASK); - netdev_info(eth->netdev[0], - "done tx %d, rx %d, intr 0x%08x/0x%x\n", - tx_done, rx_done, status, mask); + dev_info(eth->dev, + "done rx %d, intr 0x%08x/0x%x\n", + rx_done, status, mask); } - if (tx_again || rx_done == budget) + if (rx_done == budget) return budget; status = mtk_r32(eth, MTK_QMTK_INT_STATUS); - if (status & (tx_intr | rx_intr)) + if (status & MTK_RX_DONE_INT) return budget; napi_complete(napi); - mtk_irq_enable(eth, MTK_RX_DONE_INT | MTK_RX_DONE_INT); + mtk_irq_enable(eth, MTK_RX_DONE_INT); return rx_done; } @@ -1246,22 +1265,26 @@ static void mtk_tx_timeout(struct net_device *dev) schedule_work(ð->pending_work); } -static irqreturn_t mtk_handle_irq(int irq, void *_eth) +static irqreturn_t mtk_handle_irq_rx(int irq, void *_eth) { struct mtk_eth *eth = _eth; - u32 status; - status = mtk_r32(eth, MTK_QMTK_INT_STATUS); - if (unlikely(!status)) - return IRQ_NONE; + if (likely(napi_schedule_prep(ð->rx_napi))) { + __napi_schedule(ð->rx_napi); + mtk_irq_disable(eth, MTK_RX_DONE_INT); + } - if (likely(status & (MTK_RX_DONE_INT | MTK_TX_DONE_INT))) { - if (likely(napi_schedule_prep(ð->rx_napi))) - __napi_schedule(ð->rx_napi); - } else { - mtk_w32(eth, status, MTK_QMTK_INT_STATUS); + return IRQ_HANDLED; +} + +static irqreturn_t mtk_handle_irq_tx(int irq, void *_eth) +{ + struct mtk_eth *eth = _eth; + + if (likely(napi_schedule_prep(ð->tx_napi))) { + __napi_schedule(ð->tx_napi); + mtk_irq_disable(eth, MTK_TX_DONE_INT); } - mtk_irq_disable(eth, (MTK_RX_DONE_INT | MTK_TX_DONE_INT)); return IRQ_HANDLED; } @@ -1274,7 +1297,7 @@ static void mtk_poll_controller(struct net_device *dev) u32 int_mask = MTK_TX_DONE_INT | MTK_RX_DONE_INT; mtk_irq_disable(eth, int_mask); - mtk_handle_irq(dev->irq, dev); + mtk_handle_irq(dev->irq[0], dev); mtk_irq_enable(eth, int_mask); } #endif @@ -1310,6 +1333,7 @@ static int mtk_open(struct net_device *dev) if (err) return err; + napi_enable(ð->tx_napi); napi_enable(ð->rx_napi); mtk_irq_enable(eth, MTK_TX_DONE_INT | MTK_RX_DONE_INT); } @@ -1358,6 +1382,7 @@ static int mtk_stop(struct net_device *dev) return 0; mtk_irq_disable(eth, MTK_TX_DONE_INT | MTK_RX_DONE_INT); + napi_disable(ð->tx_napi); napi_disable(ð->rx_napi); mtk_stop_dma(eth, MTK_QDMA_GLO_CFG); @@ -1395,7 +1420,11 @@ static int __init mtk_hw_init(struct mtk_eth *eth) /* Enable RX VLan Offloading */ mtk_w32(eth, 1, MTK_CDMP_EG_CTRL); - err = devm_request_irq(eth->dev, eth->irq, mtk_handle_irq, 0, + err = devm_request_irq(eth->dev, eth->irq[1], mtk_handle_irq_tx, 0, + dev_name(eth->dev), eth); + if (err) + return err; + err = devm_request_irq(eth->dev, eth->irq[2], mtk_handle_irq_rx, 0, dev_name(eth->dev), eth); if (err) return err; @@ -1411,7 +1440,11 @@ static int __init mtk_hw_init(struct mtk_eth *eth) mtk_w32(eth, 0, MTK_RST_GL); /* FE int grouping */ - mtk_w32(eth, 0, MTK_FE_INT_GRP); + mtk_w32(eth, MTK_TX_DONE_INT, MTK_PDMA_INT_GRP1); + mtk_w32(eth, MTK_RX_DONE_INT, MTK_PDMA_INT_GRP2); + mtk_w32(eth, MTK_TX_DONE_INT, MTK_QDMA_INT_GRP1); + mtk_w32(eth, MTK_RX_DONE_INT, MTK_QDMA_INT_GRP2); + mtk_w32(eth, 0x21021000, MTK_FE_INT_GRP); for (i = 0; i < 2; i++) { u32 val = mtk_r32(eth, MTK_GDMA_FWD_CFG(i)); @@ -1459,7 +1492,9 @@ static void mtk_uninit(struct net_device *dev) phy_disconnect(mac->phy_dev); mtk_mdio_cleanup(eth); mtk_irq_disable(eth, ~0); - free_irq(dev->irq, dev); + free_irq(eth->irq[0], dev); + free_irq(eth->irq[1], dev); + free_irq(eth->irq[2], dev); } static int mtk_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) @@ -1733,10 +1768,10 @@ static int mtk_add_mac(struct mtk_eth *eth, struct device_node *np) dev_err(eth->dev, "error bringing up device\n"); goto free_netdev; } - eth->netdev[id]->irq = eth->irq; + eth->netdev[id]->irq = eth->irq[0]; netif_info(eth, probe, eth->netdev[id], "mediatek frame engine at 0x%08lx, irq %d\n", - eth->netdev[id]->base_addr, eth->netdev[id]->irq); + eth->netdev[id]->base_addr, eth->irq[0]); return 0; @@ -1753,6 +1788,7 @@ static int mtk_probe(struct platform_device *pdev) struct mtk_soc_data *soc; struct mtk_eth *eth; int err; + int i; match = of_match_device(of_mtk_match, &pdev->dev); soc = (struct mtk_soc_data *)match->data; @@ -1788,10 +1824,12 @@ static int mtk_probe(struct platform_device *pdev) return PTR_ERR(eth->rstc); } - eth->irq = platform_get_irq(pdev, 0); - if (eth->irq < 0) { - dev_err(&pdev->dev, "no IRQ resource found\n"); - return -ENXIO; + for (i = 0; i < 3; i++) { + eth->irq[i] = platform_get_irq(pdev, i); + if (eth->irq[i] < 0) { + dev_err(&pdev->dev, "no IRQ%d resource found\n", i); + return -ENXIO; + } } eth->clk_ethif = devm_clk_get(&pdev->dev, "ethif"); @@ -1832,7 +1870,9 @@ static int mtk_probe(struct platform_device *pdev) * for NAPI to work */ init_dummy_netdev(ð->dummy_dev); - netif_napi_add(ð->dummy_dev, ð->rx_napi, mtk_poll, + netif_napi_add(ð->dummy_dev, ð->tx_napi, mtk_napi_tx, + MTK_NAPI_WEIGHT); + netif_napi_add(ð->dummy_dev, ð->rx_napi, mtk_napi_rx, MTK_NAPI_WEIGHT); platform_set_drvdata(pdev, eth); @@ -1853,6 +1893,7 @@ static int mtk_remove(struct platform_device *pdev) clk_disable_unprepare(eth->clk_gp1); clk_disable_unprepare(eth->clk_gp2); + netif_napi_del(ð->tx_napi); netif_napi_del(ð->rx_napi); mtk_cleanup(eth); platform_set_drvdata(pdev, NULL); diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h index 3159d2a..f82e3ac 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h @@ -68,6 +68,10 @@ /* Unicast Filter MAC Address Register - High */ #define MTK_GDMA_MAC_ADRH(x) (0x50C + (x * 0x1000)) +/* PDMA Interrupt grouping registers */ +#define MTK_PDMA_INT_GRP1 0xa50 +#define MTK_PDMA_INT_GRP2 0xa54 + /* QDMA TX Queue Configuration Registers */ #define MTK_QTX_CFG(x) (0x1800 + (x * 0x10)) #define QDMA_RES_THRES 4 @@ -125,6 +129,11 @@ #define MTK_TX_DONE_INT (MTK_TX_DONE_INT0 | MTK_TX_DONE_INT1 | \ MTK_TX_DONE_INT2 | MTK_TX_DONE_INT3) +/* QDMA Interrupt grouping registers */ +#define MTK_QDMA_INT_GRP1 0x1a20 +#define MTK_QDMA_INT_GRP2 0x1a24 +#define MTK_RLS_DONE_INT BIT(0) + /* QDMA Interrupt Status Register */ #define MTK_QDMA_INT_MASK 0x1A1C @@ -356,7 +365,8 @@ struct mtk_rx_ring { * @dma_refcnt: track how many netdevs are using the DMA engine * @tx_ring: Pointer to the memore holding info about the TX ring * @rx_ring: Pointer to the memore holding info about the RX ring - * @rx_napi: The NAPI struct + * @tx_napi: The TX NAPI struct + * @rx_napi: The RX NAPI struct * @scratch_ring: Newer SoCs need memory for a second HW managed TX ring * @phy_scratch_ring: physical address of scratch_ring * @scratch_head: The scratch memory that scratch_ring points to. @@ -377,7 +387,7 @@ struct mtk_eth { struct net_device dummy_dev; struct net_device *netdev[MTK_MAX_DEVS]; struct mtk_mac *mac[MTK_MAX_DEVS]; - int irq; + int irq[3]; u32 msg_enable; unsigned long sysclk; struct regmap *ethsys; @@ -385,6 +395,7 @@ struct mtk_eth { atomic_t dma_refcnt; struct mtk_tx_ring tx_ring; struct mtk_rx_ring rx_ring; + struct napi_struct tx_napi; struct napi_struct rx_napi; struct mtk_tx_dma *scratch_ring; dma_addr_t phy_scratch_ring; -- cgit v0.10.2 From 9b9a553c902e79867b6527301fe64e45be23e36b Mon Sep 17 00:00:00 2001 From: Masanari Iida Date: Wed, 29 Jun 2016 23:36:20 +0900 Subject: net: netcp: Fix a typo in keystone-netcp.txt This patch fix a spelling typo in keystone-netcp.txt Signed-off-by: Masanari Iida Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/keystone-netcp.txt b/Documentation/devicetree/bindings/net/keystone-netcp.txt index b30ab6b..04ba1dc 100644 --- a/Documentation/devicetree/bindings/net/keystone-netcp.txt +++ b/Documentation/devicetree/bindings/net/keystone-netcp.txt @@ -2,7 +2,7 @@ This document describes the device tree bindings associated with the keystone network coprocessor(NetCP) driver support. The network coprocessor (NetCP) is a hardware accelerator that processes -Ethernet packets. NetCP has a gigabit Ethernet (GbE) subsytem with a ethernet +Ethernet packets. NetCP has a gigabit Ethernet (GbE) subsystem with a ethernet switch sub-module to send and receive packets. NetCP also includes a packet accelerator (PA) module to perform packet classification operations such as header matching, and packet modification operations such as checksum -- cgit v0.10.2 From 6fde0e63eccbaf21fa278b240b8129fec14b864b Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 29 Jun 2016 17:39:43 +0300 Subject: be2net: signedness bug in be_msix_enable() "num_vec" needs to be signed for the error handling to work. Fixes: e261768e9e39 ('be2net: support asymmetric rx/tx queue counts') Signed-off-by: Dan Carpenter Acked-by: Sathya Perla Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 1873c74..1f16e73 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -3251,8 +3251,9 @@ static void be_msix_disable(struct be_adapter *adapter) static int be_msix_enable(struct be_adapter *adapter) { - unsigned int i, num_vec, max_roce_eqs; + unsigned int i, max_roce_eqs; struct device *dev = &adapter->pdev->dev; + int num_vec; /* If RoCE is supported, program the max number of vectors that * could be used for NIC and RoCE, else, just program the number -- cgit v0.10.2 From 796312cd00d308ca142bace8dd512b02f24d178a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 29 Jun 2016 21:55:53 +0100 Subject: nfp: correct name of control BAR define Spell abbreviation of control as ctrl not crtl. Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h index e744acc..6906356 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net.h +++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h @@ -63,7 +63,7 @@ #define NFP_NET_POLL_TIMEOUT 5 /* Bar allocation */ -#define NFP_NET_CRTL_BAR 0 +#define NFP_NET_CTRL_BAR 0 #define NFP_NET_Q0_BAR 2 #define NFP_NET_Q1_BAR 4 /* OBSOLETE */ diff --git a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c index e2b22b8..37abef0 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_netvf_main.c @@ -124,11 +124,11 @@ static int nfp_netvf_pci_probe(struct pci_dev *pdev, * first NFP_NET_CFG_BAR_SZ of the BAR. This keeps the code * the identical for PF and VF drivers. */ - ctrl_bar = ioremap_nocache(pci_resource_start(pdev, NFP_NET_CRTL_BAR), + ctrl_bar = ioremap_nocache(pci_resource_start(pdev, NFP_NET_CTRL_BAR), NFP_NET_CFG_BAR_SZ); if (!ctrl_bar) { dev_err(&pdev->dev, - "Failed to map resource %d\n", NFP_NET_CRTL_BAR); + "Failed to map resource %d\n", NFP_NET_CTRL_BAR); err = -EIO; goto err_pci_regions; } -- cgit v0.10.2 From f642963bff1921b7a0b3c8250520b4bde0943387 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 29 Jun 2016 21:55:54 +0100 Subject: nfp: remove unused parameter from nfp_net_write_mac_addr() nfp_net_write_mac_addr() always writes to the BAR the current device address taken from netdev struct. The address given as parameter is actually ignored. Since all callers pass netdev->dev_addr simply remove the parameter. While at it improve the function's kdoc a bit. Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index c25a8ba..1e74b91 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -1845,13 +1845,14 @@ void nfp_net_coalesce_write_cfg(struct nfp_net *nn) } /** - * nfp_net_write_mac_addr() - Write mac address to device registers + * nfp_net_write_mac_addr() - Write mac address to the device control BAR * @nn: NFP Net device to reconfigure - * @mac: Six-byte MAC address to be written * - * We do a bit of byte swapping dance because firmware is LE. + * Writes the MAC address from the netdev to the device control BAR. Does not + * perform the required reconfig. We do a bit of byte swapping dance because + * firmware is LE. */ -static void nfp_net_write_mac_addr(struct nfp_net *nn, const u8 *mac) +static void nfp_net_write_mac_addr(struct nfp_net *nn) { nn_writel(nn, NFP_NET_CFG_MACADDR + 0, get_unaligned_be32(nn->netdev->dev_addr)); @@ -1952,7 +1953,7 @@ static int __nfp_net_set_config_and_enable(struct nfp_net *nn) nn_writeq(nn, NFP_NET_CFG_RXRS_ENABLE, nn->num_rx_rings == 64 ? 0xffffffffffffffffULL : ((u64)1 << nn->num_rx_rings) - 1); - nfp_net_write_mac_addr(nn, nn->netdev->dev_addr); + nfp_net_write_mac_addr(nn); nn_writel(nn, NFP_NET_CFG_MTU, nn->netdev->mtu); nn_writel(nn, NFP_NET_CFG_FLBUFSZ, nn->fl_bufsz); @@ -2739,7 +2740,7 @@ int nfp_net_netdev_init(struct net_device *netdev) nn->cap = nn_readl(nn, NFP_NET_CFG_CAP); nn->max_mtu = nn_readl(nn, NFP_NET_CFG_MAX_MTU); - nfp_net_write_mac_addr(nn, nn->netdev->dev_addr); + nfp_net_write_mac_addr(nn); /* Set default MTU and Freelist buffer size */ if (nn->max_mtu < NFP_NET_DEFAULT_MTU) -- cgit v0.10.2 From 2370def2e42e488cfae6a0ad0ca8331d10457cbe Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 29 Jun 2016 21:55:55 +0100 Subject: nfp: implement ethtool .get_link() callback Point the ethtool .get_link() callback to the standard ethtool_op_get_link() implementation. Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c index ccfef1f..7d7933d 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c @@ -605,6 +605,7 @@ static int nfp_net_set_coalesce(struct net_device *netdev, static const struct ethtool_ops nfp_net_ethtool_ops = { .get_drvinfo = nfp_net_get_drvinfo, + .get_link = ethtool_op_get_link, .get_ringparam = nfp_net_get_ringparam, .set_ringparam = nfp_net_set_ringparam, .get_strings = nfp_net_get_strings, -- cgit v0.10.2 From ac5d26836cb6c01505d186180a79b4362ee7b4ac Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 1 Jul 2016 08:35:02 +0100 Subject: rxrpc: Fix processing of authenticated/encrypted jumbo packets When a jumbo packet is being split up and processed, the crypto checksum for each split-out packet is in the jumbo header and needs placing in the reconstructed packet header. When the code was changed to keep the stored copy of the packet header in host byte order, this reconstruction was missed. Found with sparse with CF=-D__CHECK_ENDIAN__: ../net/rxrpc/input.c:479:33: warning: incorrect type in assignment (different base types) ../net/rxrpc/input.c:479:33: expected unsigned short [unsigned] [usertype] _rsvd ../net/rxrpc/input.c:479:33: got restricted __be16 [addressable] [usertype] _rsvd Fixes: 0d12f8a4027d021c9cc942f09f38d28288020c5d ("rxrpc: Keep the skb private record of the Rx header in host byte order") Signed-off-by: David Howells diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index f4bd57b..5f26cae 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -476,7 +476,7 @@ static void rxrpc_process_jumbo_packet(struct rxrpc_call *call, sp->hdr.seq += 1; sp->hdr.serial += 1; sp->hdr.flags = jhdr.flags; - sp->hdr._rsvd = jhdr._rsvd; + sp->hdr._rsvd = ntohs(jhdr._rsvd); _proto("Rx DATA Jumbo %%%u", sp->hdr.serial - 1); -- cgit v0.10.2 From 19689e38eca5d7b32755182d4e62efd7a5376c45 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 27 Jun 2016 18:51:53 +0200 Subject: tcp: md5: use kmalloc() backed scratch areas Some arches have virtually mapped kernel stacks, or will soon have. tcp_md5_hash_header() uses an automatic variable to copy tcp header before mangling th->check and calling crypto function, which might be problematic on such arches. David says that using percpu storage is also problematic on non SMP builds. Just use kmalloc() to allocate scratch areas. Signed-off-by: Eric Dumazet Reported-by: Andy Lutomirski Signed-off-by: David S. Miller diff --git a/include/net/tcp.h b/include/net/tcp.h index d825858..c00e7d5 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1384,7 +1384,7 @@ union tcp_md5sum_block { /* - pool: digest algorithm, hash description and scratch buffer */ struct tcp_md5sig_pool { struct ahash_request *md5_req; - union tcp_md5sum_block md5_blk; + void *scratch; }; /* - functions */ @@ -1420,7 +1420,6 @@ static inline void tcp_put_md5sig_pool(void) local_bh_enable(); } -int tcp_md5_hash_header(struct tcp_md5sig_pool *, const struct tcphdr *); int tcp_md5_hash_skb_data(struct tcp_md5sig_pool *, const struct sk_buff *, unsigned int header_len); int tcp_md5_hash_key(struct tcp_md5sig_pool *hp, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 108ef2a..032a96d 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -3026,8 +3026,18 @@ static void __tcp_alloc_md5sig_pool(void) return; for_each_possible_cpu(cpu) { + void *scratch = per_cpu(tcp_md5sig_pool, cpu).scratch; struct ahash_request *req; + if (!scratch) { + scratch = kmalloc_node(sizeof(union tcp_md5sum_block) + + sizeof(struct tcphdr), + GFP_KERNEL, + cpu_to_node(cpu)); + if (!scratch) + return; + per_cpu(tcp_md5sig_pool, cpu).scratch = scratch; + } if (per_cpu(tcp_md5sig_pool, cpu).md5_req) continue; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 3708de2..32b048e 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1018,27 +1018,28 @@ static int tcp_v4_parse_md5_keys(struct sock *sk, char __user *optval, GFP_KERNEL); } -static int tcp_v4_md5_hash_pseudoheader(struct tcp_md5sig_pool *hp, - __be32 daddr, __be32 saddr, int nbytes) +static int tcp_v4_md5_hash_headers(struct tcp_md5sig_pool *hp, + __be32 daddr, __be32 saddr, + const struct tcphdr *th, int nbytes) { struct tcp4_pseudohdr *bp; struct scatterlist sg; + struct tcphdr *_th; - bp = &hp->md5_blk.ip4; - - /* - * 1. the TCP pseudo-header (in the order: source IP address, - * destination IP address, zero-padded protocol number, and - * segment length) - */ + bp = hp->scratch; bp->saddr = saddr; bp->daddr = daddr; bp->pad = 0; bp->protocol = IPPROTO_TCP; bp->len = cpu_to_be16(nbytes); - sg_init_one(&sg, bp, sizeof(*bp)); - ahash_request_set_crypt(hp->md5_req, &sg, NULL, sizeof(*bp)); + _th = (struct tcphdr *)(bp + 1); + memcpy(_th, th, sizeof(*th)); + _th->check = 0; + + sg_init_one(&sg, bp, sizeof(*bp) + sizeof(*th)); + ahash_request_set_crypt(hp->md5_req, &sg, NULL, + sizeof(*bp) + sizeof(*th)); return crypto_ahash_update(hp->md5_req); } @@ -1055,9 +1056,7 @@ static int tcp_v4_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key, if (crypto_ahash_init(req)) goto clear_hash; - if (tcp_v4_md5_hash_pseudoheader(hp, daddr, saddr, th->doff << 2)) - goto clear_hash; - if (tcp_md5_hash_header(hp, th)) + if (tcp_v4_md5_hash_headers(hp, daddr, saddr, th, th->doff << 2)) goto clear_hash; if (tcp_md5_hash_key(hp, key)) goto clear_hash; @@ -1101,9 +1100,7 @@ int tcp_v4_md5_hash_skb(char *md5_hash, const struct tcp_md5sig_key *key, if (crypto_ahash_init(req)) goto clear_hash; - if (tcp_v4_md5_hash_pseudoheader(hp, daddr, saddr, skb->len)) - goto clear_hash; - if (tcp_md5_hash_header(hp, th)) + if (tcp_v4_md5_hash_headers(hp, daddr, saddr, th, skb->len)) goto clear_hash; if (tcp_md5_hash_skb_data(hp, skb, th->doff << 2)) goto clear_hash; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 2255d2b..37cf913 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -526,26 +526,33 @@ static int tcp_v6_parse_md5_keys(struct sock *sk, char __user *optval, AF_INET6, cmd.tcpm_key, cmd.tcpm_keylen, GFP_KERNEL); } -static int tcp_v6_md5_hash_pseudoheader(struct tcp_md5sig_pool *hp, - const struct in6_addr *daddr, - const struct in6_addr *saddr, int nbytes) +static int tcp_v6_md5_hash_headers(struct tcp_md5sig_pool *hp, + const struct in6_addr *daddr, + const struct in6_addr *saddr, + const struct tcphdr *th, int nbytes) { struct tcp6_pseudohdr *bp; struct scatterlist sg; + struct tcphdr *_th; - bp = &hp->md5_blk.ip6; + bp = hp->scratch; /* 1. TCP pseudo-header (RFC2460) */ bp->saddr = *saddr; bp->daddr = *daddr; bp->protocol = cpu_to_be32(IPPROTO_TCP); bp->len = cpu_to_be32(nbytes); - sg_init_one(&sg, bp, sizeof(*bp)); - ahash_request_set_crypt(hp->md5_req, &sg, NULL, sizeof(*bp)); + _th = (struct tcphdr *)(bp + 1); + memcpy(_th, th, sizeof(*th)); + _th->check = 0; + + sg_init_one(&sg, bp, sizeof(*bp) + sizeof(*th)); + ahash_request_set_crypt(hp->md5_req, &sg, NULL, + sizeof(*bp) + sizeof(*th)); return crypto_ahash_update(hp->md5_req); } -static int tcp_v6_md5_hash_hdr(char *md5_hash, struct tcp_md5sig_key *key, +static int tcp_v6_md5_hash_hdr(char *md5_hash, const struct tcp_md5sig_key *key, const struct in6_addr *daddr, struct in6_addr *saddr, const struct tcphdr *th) { @@ -559,9 +566,7 @@ static int tcp_v6_md5_hash_hdr(char *md5_hash, struct tcp_md5sig_key *key, if (crypto_ahash_init(req)) goto clear_hash; - if (tcp_v6_md5_hash_pseudoheader(hp, daddr, saddr, th->doff << 2)) - goto clear_hash; - if (tcp_md5_hash_header(hp, th)) + if (tcp_v6_md5_hash_headers(hp, daddr, saddr, th, th->doff << 2)) goto clear_hash; if (tcp_md5_hash_key(hp, key)) goto clear_hash; @@ -606,9 +611,7 @@ static int tcp_v6_md5_hash_skb(char *md5_hash, if (crypto_ahash_init(req)) goto clear_hash; - if (tcp_v6_md5_hash_pseudoheader(hp, daddr, saddr, skb->len)) - goto clear_hash; - if (tcp_md5_hash_header(hp, th)) + if (tcp_v6_md5_hash_headers(hp, daddr, saddr, th, skb->len)) goto clear_hash; if (tcp_md5_hash_skb_data(hp, skb, th->doff << 2)) goto clear_hash; -- cgit v0.10.2 From 12d0ad3be9c3854e52ec74bb83bb6f43612827c7 Mon Sep 17 00:00:00 2001 From: Michal Soltys Date: Thu, 30 Jun 2016 02:26:44 +0200 Subject: net/sched/sch_hfsc.c: handle corner cases where head may change invalidating calculated deadline Realtime scheduling implemented in HFSC uses head of the queue to make the decision about which packet to schedule next. But in case of any head drop, the deadline calculated for the previous head is not necessarily correct for the next head (unless both packets have the same length). Thanks to peek() function used during dequeue - which internally is a dequeue operation - hfsc is almost safe from this issue, as peek() dequeues and isolates the head storing it temporarily until the real dequeue happens. But there is one exception: if after the class activation a drop happens before the first dequeue operation, there's never a chance to do the peek(). Adding peek() call in enqueue - if this is the first packet in a new backlog period AND the scheduler has realtime curve defined - fixes that one corner case. The 1st hfsc_dequeue() will use that peeked packet, similarly as every subsequent hfsc_dequeue() call uses packet peeked by the previous call. Signed-off-by: Michal Soltys Signed-off-by: David S. Miller diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index 8cb5eff..6d6df6b 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1594,8 +1594,17 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch, struct sk_buff **to_free) return err; } - if (cl->qdisc->q.qlen == 1) + if (cl->qdisc->q.qlen == 1) { set_active(cl, qdisc_pkt_len(skb)); + /* + * If this is the first packet, isolate the head so an eventual + * head drop before the first dequeue operation has no chance + * to invalidate the deadline. + */ + if (cl->cl_flags & HFSC_RSC) + cl->qdisc->ops->peek(cl->qdisc); + + } qdisc_qstats_backlog_inc(sch, skb); sch->q.qlen++; -- cgit v0.10.2 From d1d0fc5e4c6822c5dadd9389297c7c1b8eea314f Mon Sep 17 00:00:00 2001 From: Michal Soltys Date: Thu, 30 Jun 2016 02:26:45 +0200 Subject: net/sched/sch_hfsc.c: add unlikely() in qdisc_peek_len() The condition can only succeed on wrong configurations. Signed-off-by: Michal Soltys Signed-off-by: David S. Miller diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index 6d6df6b..e2244bb 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -882,7 +882,7 @@ qdisc_peek_len(struct Qdisc *sch) unsigned int len; skb = sch->ops->peek(sch); - if (skb == NULL) { + if (unlikely(skb == NULL)) { qdisc_warn_nonwc("qdisc_peek_len", sch); return 0; } -- cgit v0.10.2 From 2354f056f6847125a95b42732ab481730389c099 Mon Sep 17 00:00:00 2001 From: Michal Soltys Date: Thu, 30 Jun 2016 02:26:46 +0200 Subject: net/sched/sch_hfsc.c: remove leftover dlist and droplist This is update to: commit a09ceb0e08140a ("sched: remove qdisc->drop") That commit removed qdisc->drop, but left alone dlist and droplist that no longer serve any meaningful purpose. Signed-off-by: Michal Soltys Signed-off-by: David S. Miller diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index e2244bb..df07f06 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -130,7 +130,6 @@ struct hfsc_class { struct rb_node vt_node; /* parent's vt_tree member */ struct rb_root cf_tree; /* active children sorted by cl_f */ struct rb_node cf_node; /* parent's cf_heap member */ - struct list_head dlist; /* drop list member */ u64 cl_total; /* total work in bytes */ u64 cl_cumul; /* cumulative work in bytes done by @@ -177,8 +176,6 @@ struct hfsc_sched { struct hfsc_class root; /* root class */ struct Qdisc_class_hash clhash; /* class hash */ struct rb_root eligible; /* eligible tree */ - struct list_head droplist; /* active leaf class list (for - dropping) */ struct qdisc_watchdog watchdog; /* watchdog timer */ }; @@ -858,7 +855,6 @@ set_active(struct hfsc_class *cl, unsigned int len) if (cl->cl_flags & HFSC_FSC) init_vf(cl, len); - list_add_tail(&cl->dlist, &cl->sched->droplist); } static void @@ -867,8 +863,6 @@ set_passive(struct hfsc_class *cl) if (cl->cl_flags & HFSC_RSC) eltree_remove(cl); - list_del(&cl->dlist); - /* * vttree is now handled in update_vf() so that update_vf(cl, 0, 0) * needs to be called explicitly to remove a class from vttree. @@ -1443,7 +1437,6 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt) if (err < 0) return err; q->eligible = RB_ROOT; - INIT_LIST_HEAD(&q->droplist); q->root.cl_common.classid = sch->handle; q->root.refcnt = 1; @@ -1527,7 +1520,6 @@ hfsc_reset_qdisc(struct Qdisc *sch) hfsc_reset_class(cl); } q->eligible = RB_ROOT; - INIT_LIST_HEAD(&q->droplist); qdisc_watchdog_cancel(&q->watchdog); sch->qstats.backlog = 0; sch->q.qlen = 0; -- cgit v0.10.2 From ab12cb4742cf608cfee43c84fe07fa56bd473dcb Mon Sep 17 00:00:00 2001 From: Michal Soltys Date: Thu, 30 Jun 2016 02:26:47 +0200 Subject: net/sched/sch_hfsc.c: go passive after vt update When a class is going passive, it should update its cl_vt first to be consistent with the last dequeue operation. Otherwise its cl_vt will be one packet behind and parent's cvtmax might not be updated as well. One possible side effect is if some class goes passive and subsequently goes active /without/ its parent going passive - with cl_vt lagging one packet behind - comparison made in init_vf() will be affected (same period). Signed-off-by: Michal Soltys Signed-off-by: David S. Miller diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index df07f06..4eef857 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -778,6 +778,20 @@ update_vf(struct hfsc_class *cl, unsigned int len, u64 cur_time) else go_passive = 0; + /* update vt */ + cl->cl_vt = rtsc_y2x(&cl->cl_virtual, cl->cl_total) + - cl->cl_vtoff + cl->cl_vtadj; + + /* + * if vt of the class is smaller than cvtmin, + * the class was skipped in the past due to non-fit. + * if so, we need to adjust vtadj. + */ + if (cl->cl_vt < cl->cl_parent->cl_cvtmin) { + cl->cl_vtadj += cl->cl_parent->cl_cvtmin - cl->cl_vt; + cl->cl_vt = cl->cl_parent->cl_cvtmin; + } + if (go_passive) { /* no more active child, going passive */ @@ -794,25 +808,10 @@ update_vf(struct hfsc_class *cl, unsigned int len, u64 cur_time) continue; } - /* - * update vt and f - */ - cl->cl_vt = rtsc_y2x(&cl->cl_virtual, cl->cl_total) - - cl->cl_vtoff + cl->cl_vtadj; - - /* - * if vt of the class is smaller than cvtmin, - * the class was skipped in the past due to non-fit. - * if so, we need to adjust vtadj. - */ - if (cl->cl_vt < cl->cl_parent->cl_cvtmin) { - cl->cl_vtadj += cl->cl_parent->cl_cvtmin - cl->cl_vt; - cl->cl_vt = cl->cl_parent->cl_cvtmin; - } - /* update the vt tree */ vttree_update(cl); + /* update f */ if (cl->cl_flags & HFSC_USC) { cl->cl_myf = cl->cl_myfadj + rtsc_y2x(&cl->cl_ulimit, cl->cl_total); -- cgit v0.10.2 From 33ef84a77d7502359abd097a28dbeb67d5466a7c Mon Sep 17 00:00:00 2001 From: Michal Soltys Date: Thu, 30 Jun 2016 02:26:48 +0200 Subject: net/sched/sch_hfsc.c: anchor virtual curve at proper vt in hfsc_change_fsc() cl->cl_vt alone is relative only to the current backlog period, while the curve operates on cumulative virtual time. This patch adds missing cl->cl_vtoff. Signed-off-by: Michal Soltys Signed-off-by: David S. Miller diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index 4eef857..dff92ea 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -940,7 +940,7 @@ static void hfsc_change_fsc(struct hfsc_class *cl, struct tc_service_curve *fsc) { sc2isc(fsc, &cl->cl_fsc); - rtsc_init(&cl->cl_virtual, &cl->cl_fsc, cl->cl_vt, cl->cl_total); + rtsc_init(&cl->cl_virtual, &cl->cl_fsc, cl->cl_vtoff + cl->cl_vt, cl->cl_total); cl->cl_flags |= HFSC_FSC; } -- cgit v0.10.2 From 982fb490c298896d15e9323a882f34a57c11ff56 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 30 Jun 2016 14:45:31 +0800 Subject: ptr_ring: support zero length ring Sometimes, we need zero length ring. But current code will crash since we don't do any check before accessing the ring. This patch fixes this. Signed-off-by: Jason Wang Signed-off-by: David S. Miller diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h index 562a65e..d78b8b8 100644 --- a/include/linux/ptr_ring.h +++ b/include/linux/ptr_ring.h @@ -102,7 +102,7 @@ static inline bool ptr_ring_full_bh(struct ptr_ring *r) */ static inline int __ptr_ring_produce(struct ptr_ring *r, void *ptr) { - if (r->queue[r->producer]) + if (unlikely(!r->size) || r->queue[r->producer]) return -ENOSPC; r->queue[r->producer++] = ptr; @@ -164,7 +164,9 @@ static inline int ptr_ring_produce_bh(struct ptr_ring *r, void *ptr) */ static inline void *__ptr_ring_peek(struct ptr_ring *r) { - return r->queue[r->consumer]; + if (likely(r->size)) + return r->queue[r->consumer]; + return NULL; } /* Note: callers invoking this in a loop must use a compiler barrier, -- cgit v0.10.2 From fd68adec9de3104c236ffbcb3bd829d3e635a444 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 30 Jun 2016 14:45:32 +0800 Subject: skb_array: minor tweak Signed-off-by: Michael S. Tsirkin Signed-off-by: Jason Wang Signed-off-by: David S. Miller diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index 678bfbf..2dd0d1e 100644 --- a/include/linux/skb_array.h +++ b/include/linux/skb_array.h @@ -151,12 +151,12 @@ static inline int skb_array_init(struct skb_array *a, int size, gfp_t gfp) return ptr_ring_init(&a->ring, size, gfp); } -void __skb_array_destroy_skb(void *ptr) +static void __skb_array_destroy_skb(void *ptr) { kfree_skb(ptr); } -int skb_array_resize(struct skb_array *a, int size, gfp_t gfp) +static inline int skb_array_resize(struct skb_array *a, int size, gfp_t gfp) { return ptr_ring_resize(&a->ring, size, gfp, __skb_array_destroy_skb); } -- cgit v0.10.2 From 59e6ae53248a72d83cec77dd704b6990b2394479 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 30 Jun 2016 14:45:33 +0800 Subject: ptr_ring: support resizing multiple queues Sometimes, we need support resizing multiple queues at once. This is because it was not easy to recover to recover from a partial failure of multiple queues resizing. Signed-off-by: Michael S. Tsirkin Signed-off-by: Jason Wang Signed-off-by: David S. Miller diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h index d78b8b8..2052011 100644 --- a/include/linux/ptr_ring.h +++ b/include/linux/ptr_ring.h @@ -349,20 +349,14 @@ static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) return 0; } -static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, - void (*destroy)(void *)) +static inline void **__ptr_ring_swap_queue(struct ptr_ring *r, void **queue, + int size, gfp_t gfp, + void (*destroy)(void *)) { - unsigned long flags; int producer = 0; - void **queue = __ptr_ring_init_queue_alloc(size, gfp); void **old; void *ptr; - if (!queue) - return -ENOMEM; - - spin_lock_irqsave(&(r)->producer_lock, flags); - while ((ptr = ptr_ring_consume(r))) if (producer < size) queue[producer++] = ptr; @@ -375,6 +369,23 @@ static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, old = r->queue; r->queue = queue; + return old; +} + +static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, + void (*destroy)(void *)) +{ + unsigned long flags; + void **queue = __ptr_ring_init_queue_alloc(size, gfp); + void **old; + + if (!queue) + return -ENOMEM; + + spin_lock_irqsave(&(r)->producer_lock, flags); + + old = __ptr_ring_swap_queue(r, queue, size, gfp, destroy); + spin_unlock_irqrestore(&(r)->producer_lock, flags); kfree(old); @@ -382,6 +393,48 @@ static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, return 0; } +static inline int ptr_ring_resize_multiple(struct ptr_ring **rings, int nrings, + int size, + gfp_t gfp, void (*destroy)(void *)) +{ + unsigned long flags; + void ***queues; + int i; + + queues = kmalloc(nrings * sizeof *queues, gfp); + if (!queues) + goto noqueues; + + for (i = 0; i < nrings; ++i) { + queues[i] = __ptr_ring_init_queue_alloc(size, gfp); + if (!queues[i]) + goto nomem; + } + + for (i = 0; i < nrings; ++i) { + spin_lock_irqsave(&(rings[i])->producer_lock, flags); + queues[i] = __ptr_ring_swap_queue(rings[i], queues[i], + size, gfp, destroy); + spin_unlock_irqrestore(&(rings[i])->producer_lock, flags); + } + + for (i = 0; i < nrings; ++i) + kfree(queues[i]); + + kfree(queues); + + return 0; + +nomem: + while (--i >= 0) + kfree(queues[i]); + + kfree(queues); + +noqueues: + return -ENOMEM; +} + static inline void ptr_ring_cleanup(struct ptr_ring *r, void (*destroy)(void *)) { void *ptr; diff --git a/tools/virtio/ringtest/ptr_ring.c b/tools/virtio/ringtest/ptr_ring.c index 74abd74..68e4f9f 100644 --- a/tools/virtio/ringtest/ptr_ring.c +++ b/tools/virtio/ringtest/ptr_ring.c @@ -17,6 +17,11 @@ typedef pthread_spinlock_t spinlock_t; typedef int gfp_t; +static void *kmalloc(unsigned size, gfp_t gfp) +{ + return memalign(64, size); +} + static void *kzalloc(unsigned size, gfp_t gfp) { void *p = memalign(64, size); -- cgit v0.10.2 From bf900b3dbefee49855c17aa09fb4245346a78fb3 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 30 Jun 2016 14:45:34 +0800 Subject: skb_array: add wrappers for resizing Signed-off-by: Michael S. Tsirkin Signed-off-by: Jason Wang Signed-off-by: David S. Miller diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index 2dd0d1e..f4dfade 100644 --- a/include/linux/skb_array.h +++ b/include/linux/skb_array.h @@ -161,6 +161,15 @@ static inline int skb_array_resize(struct skb_array *a, int size, gfp_t gfp) return ptr_ring_resize(&a->ring, size, gfp, __skb_array_destroy_skb); } +static inline int skb_array_resize_multiple(struct skb_array **rings, + int nrings, int size, gfp_t gfp) +{ + BUILD_BUG_ON(offsetof(struct skb_array, ring)); + return ptr_ring_resize_multiple((struct ptr_ring **)rings, + nrings, size, gfp, + __skb_array_destroy_skb); +} + static inline void skb_array_cleanup(struct skb_array *a) { ptr_ring_cleanup(&a->ring, __skb_array_destroy_skb); -- cgit v0.10.2 From 08294a26e15d7baf1e14ee569e9f2bc82a7ae768 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 30 Jun 2016 14:45:35 +0800 Subject: net: introduce NETDEV_CHANGE_TX_QUEUE_LEN This patch introduces a new event - NETDEV_CHANGE_TX_QUEUE_LEN, this will be triggered when tx_queue_len. It could be used by net device who want to do some processing at that time. An example is tun who may want to resize tx array when tx_queue_len is changed. Cc: John Fastabend Signed-off-by: Jason Wang Acked-by: John Fastabend Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index e84d9d2..7dc2ec7 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2237,6 +2237,7 @@ struct netdev_lag_lower_state_info { #define NETDEV_PRECHANGEUPPER 0x001A #define NETDEV_CHANGELOWERSTATE 0x001B #define NETDEV_UDP_TUNNEL_PUSH_INFO 0x001C +#define NETDEV_CHANGE_TX_QUEUE_LEN 0x001E int register_netdevice_notifier(struct notifier_block *nb); int unregister_netdevice_notifier(struct notifier_block *nb); diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 7a0b616..6e4f347 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -322,7 +322,20 @@ NETDEVICE_SHOW_RW(flags, fmt_hex); static int change_tx_queue_len(struct net_device *dev, unsigned long new_len) { - dev->tx_queue_len = new_len; + int res, orig_len = dev->tx_queue_len; + + if (new_len != orig_len) { + dev->tx_queue_len = new_len; + res = call_netdevice_notifiers(NETDEV_CHANGE_TX_QUEUE_LEN, dev); + res = notifier_to_errno(res); + if (res) { + netdev_err(dev, + "refused to change device tx_queue_len\n"); + dev->tx_queue_len = orig_len; + return -EFAULT; + } + } + return 0; } diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index cfed7bc..a9e3805 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1927,11 +1927,19 @@ static int do_setlink(const struct sk_buff *skb, if (tb[IFLA_TXQLEN]) { unsigned long value = nla_get_u32(tb[IFLA_TXQLEN]); - - if (dev->tx_queue_len ^ value) + unsigned long orig_len = dev->tx_queue_len; + + if (dev->tx_queue_len ^ value) { + dev->tx_queue_len = value; + err = call_netdevice_notifiers( + NETDEV_CHANGE_TX_QUEUE_LEN, dev); + err = notifier_to_errno(err); + if (err) { + dev->tx_queue_len = orig_len; + goto errout; + } status |= DO_SETLINK_NOTIFY; - - dev->tx_queue_len = value; + } } if (tb[IFLA_OPERSTATE]) -- cgit v0.10.2 From 1576d98605998fb59d121a39581129e134217182 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 30 Jun 2016 14:45:36 +0800 Subject: tun: switch to use skb array for tx We used to queue tx packets in sk_receive_queue, this is less efficient since it requires spinlocks to synchronize between producer and consumer. This patch tries to address this by: - switch from sk_receive_queue to a skb_array, and resize it when tx_queue_len was changed. - introduce a new proto_ops peek_len which was used for peeking the skb length. - implement a tun version of peek_len for vhost_net to use and convert vhost_net to use peek_len if possible. Pktgen test shows about 15.3% improvement on guest receiving pps for small buffers: Before: ~1300000pps After : ~1500000pps Signed-off-by: Jason Wang Signed-off-by: David S. Miller diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 4884802..7475215 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -71,6 +71,7 @@ #include #include #include +#include #include @@ -167,6 +168,7 @@ struct tun_file { }; struct list_head next; struct tun_struct *detached; + struct skb_array tx_array; }; struct tun_flow_entry { @@ -515,7 +517,11 @@ static struct tun_struct *tun_enable_queue(struct tun_file *tfile) static void tun_queue_purge(struct tun_file *tfile) { - skb_queue_purge(&tfile->sk.sk_receive_queue); + struct sk_buff *skb; + + while ((skb = skb_array_consume(&tfile->tx_array)) != NULL) + kfree_skb(skb); + skb_queue_purge(&tfile->sk.sk_error_queue); } @@ -560,6 +566,8 @@ static void __tun_detach(struct tun_file *tfile, bool clean) tun->dev->reg_state == NETREG_REGISTERED) unregister_netdevice(tun->dev); } + if (tun) + skb_array_cleanup(&tfile->tx_array); sock_put(&tfile->sk); } } @@ -613,6 +621,7 @@ static void tun_detach_all(struct net_device *dev) static int tun_attach(struct tun_struct *tun, struct file *file, bool skip_filter) { struct tun_file *tfile = file->private_data; + struct net_device *dev = tun->dev; int err; err = security_tun_dev_attach(tfile->socket.sk, tun->security); @@ -642,6 +651,13 @@ static int tun_attach(struct tun_struct *tun, struct file *file, bool skip_filte if (!err) goto out; } + + if (!tfile->detached && + skb_array_init(&tfile->tx_array, dev->tx_queue_len, GFP_KERNEL)) { + err = -ENOMEM; + goto out; + } + tfile->queue_index = tun->numqueues; tfile->socket.sk->sk_shutdown &= ~RCV_SHUTDOWN; rcu_assign_pointer(tfile->tun, tun); @@ -891,8 +907,8 @@ static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev) nf_reset(skb); - /* Enqueue packet */ - skb_queue_tail(&tfile->socket.sk->sk_receive_queue, skb); + if (skb_array_produce(&tfile->tx_array, skb)) + goto drop; /* Notify and wake up reader process */ if (tfile->flags & TUN_FASYNC) @@ -1107,7 +1123,7 @@ static unsigned int tun_chr_poll(struct file *file, poll_table *wait) poll_wait(file, sk_sleep(sk), wait); - if (!skb_queue_empty(&sk->sk_receive_queue)) + if (!skb_array_empty(&tfile->tx_array)) mask |= POLLIN | POLLRDNORM; if (sock_writeable(sk) || @@ -1426,22 +1442,61 @@ done: return total; } +static struct sk_buff *tun_ring_recv(struct tun_file *tfile, int noblock, + int *err) +{ + DECLARE_WAITQUEUE(wait, current); + struct sk_buff *skb = NULL; + + skb = skb_array_consume(&tfile->tx_array); + if (skb) + goto out; + if (noblock) { + *err = -EAGAIN; + goto out; + } + + add_wait_queue(&tfile->wq.wait, &wait); + current->state = TASK_INTERRUPTIBLE; + + while (1) { + skb = skb_array_consume(&tfile->tx_array); + if (skb) + break; + if (signal_pending(current)) { + *err = -ERESTARTSYS; + break; + } + if (tfile->socket.sk->sk_shutdown & RCV_SHUTDOWN) { + *err = -EFAULT; + break; + } + + schedule(); + } + + current->state = TASK_RUNNING; + remove_wait_queue(&tfile->wq.wait, &wait); + +out: + return skb; +} + static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile, struct iov_iter *to, int noblock) { struct sk_buff *skb; ssize_t ret; - int peeked, err, off = 0; + int err; tun_debug(KERN_INFO, tun, "tun_do_read\n"); if (!iov_iter_count(to)) return 0; - /* Read frames from queue */ - skb = __skb_recv_datagram(tfile->socket.sk, noblock ? MSG_DONTWAIT : 0, - &peeked, &off, &err); + /* Read frames from ring */ + skb = tun_ring_recv(tfile, noblock, &err); if (!skb) return err; @@ -1574,8 +1629,25 @@ out: return ret; } +static int tun_peek_len(struct socket *sock) +{ + struct tun_file *tfile = container_of(sock, struct tun_file, socket); + struct tun_struct *tun; + int ret = 0; + + tun = __tun_get(tfile); + if (!tun) + return 0; + + ret = skb_array_peek_len(&tfile->tx_array); + tun_put(tun); + + return ret; +} + /* Ops structure to mimic raw sockets with tun */ static const struct proto_ops tun_socket_ops = { + .peek_len = tun_peek_len, .sendmsg = tun_sendmsg, .recvmsg = tun_recvmsg, }; @@ -2397,6 +2469,53 @@ static const struct ethtool_ops tun_ethtool_ops = { .get_ts_info = ethtool_op_get_ts_info, }; +static int tun_queue_resize(struct tun_struct *tun) +{ + struct net_device *dev = tun->dev; + struct tun_file *tfile; + struct skb_array **arrays; + int n = tun->numqueues + tun->numdisabled; + int ret, i; + + arrays = kmalloc(sizeof *arrays * n, GFP_KERNEL); + if (!arrays) + return -ENOMEM; + + for (i = 0; i < tun->numqueues; i++) { + tfile = rtnl_dereference(tun->tfiles[i]); + arrays[i] = &tfile->tx_array; + } + list_for_each_entry(tfile, &tun->disabled, next) + arrays[i++] = &tfile->tx_array; + + ret = skb_array_resize_multiple(arrays, n, + dev->tx_queue_len, GFP_KERNEL); + + kfree(arrays); + return ret; +} + +static int tun_device_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct tun_struct *tun = netdev_priv(dev); + + switch (event) { + case NETDEV_CHANGE_TX_QUEUE_LEN: + if (tun_queue_resize(tun)) + return NOTIFY_BAD; + break; + default: + break; + } + + return NOTIFY_DONE; +} + +static struct notifier_block tun_notifier_block __read_mostly = { + .notifier_call = tun_device_event, +}; static int __init tun_init(void) { @@ -2416,6 +2535,8 @@ static int __init tun_init(void) pr_err("Can't register misc device %d\n", TUN_MINOR); goto err_misc; } + + register_netdevice_notifier(&tun_notifier_block); return 0; err_misc: rtnl_link_unregister(&tun_link_ops); @@ -2427,6 +2548,7 @@ static void tun_cleanup(void) { misc_deregister(&tun_miscdev); rtnl_link_unregister(&tun_link_ops); + unregister_netdevice_notifier(&tun_notifier_block); } /* Get an underlying socket object from tun file. Returns error unless file is diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index 1d3e45f..e032ca3 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -481,10 +481,14 @@ out: static int peek_head_len(struct sock *sk) { + struct socket *sock = sk->sk_socket; struct sk_buff *head; int len = 0; unsigned long flags; + if (sock->ops->peek_len) + return sock->ops->peek_len(sock); + spin_lock_irqsave(&sk->sk_receive_queue.lock, flags); head = skb_peek(&sk->sk_receive_queue); if (likely(head)) { @@ -497,6 +501,16 @@ static int peek_head_len(struct sock *sk) return len; } +static int sk_has_rx_data(struct sock *sk) +{ + struct socket *sock = sk->sk_socket; + + if (sock->ops->peek_len) + return sock->ops->peek_len(sock); + + return skb_queue_empty(&sk->sk_receive_queue); +} + static int vhost_net_rx_peek_head_len(struct vhost_net *net, struct sock *sk) { struct vhost_net_virtqueue *nvq = &net->vqs[VHOST_NET_VQ_TX]; @@ -513,7 +527,7 @@ static int vhost_net_rx_peek_head_len(struct vhost_net *net, struct sock *sk) endtime = busy_clock() + vq->busyloop_timeout; while (vhost_can_busy_poll(&net->dev, endtime) && - skb_queue_empty(&sk->sk_receive_queue) && + !sk_has_rx_data(sk) && vhost_vq_avail_empty(&net->dev, vq)) cpu_relax_lowlatency(); diff --git a/include/linux/net.h b/include/linux/net.h index 25aa03b..b9f0ff4 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -185,6 +185,7 @@ struct proto_ops { ssize_t (*splice_read)(struct socket *sock, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags); int (*set_peek_off)(struct sock *sk, int val); + int (*peek_len)(struct socket *sock); }; #define DECLARE_SOCKADDR(type, dst, src) \ -- cgit v0.10.2 From c72a6125d00f37fbb3173dc0d70c634ecf24e544 Mon Sep 17 00:00:00 2001 From: Manish Chopra Date: Thu, 30 Jun 2016 02:35:18 -0400 Subject: qede: Add support for handling IP fragmented packets. When handling IP fragmented packets with csum in their transport header, the csum isn't changed as part of the fragmentation. As a result, the packet containing the transport headers would have the correct csum of the original packet, but one that mismatches the actual packet that passes on the wire. As a result, on receive path HW would give an indication that the packet has incorrect csum, which would cause qede to discard the incoming packet. Since HW also delivers a notification of IP fragments, change driver behavior to pass such incoming packets to stack and let it make the decision whether it needs to be dropped. Signed-off-by: Manish Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 1441c8f..677213e 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -235,6 +235,7 @@ struct qede_rx_queue { u64 rx_hw_errors; u64 rx_alloc_errors; + u64 rx_ip_frags; }; union db_prod { diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index c5c658a..85633cf 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -37,6 +37,7 @@ static const struct { } qede_rqstats_arr[] = { QEDE_RQSTAT(rx_hw_errors), QEDE_RQSTAT(rx_alloc_errors), + QEDE_RQSTAT(rx_ip_frags), }; #define QEDE_NUM_RQSTATS ARRAY_SIZE(qede_rqstats_arr) diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 19bc631..ae4c53b 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -1348,6 +1348,20 @@ static u8 qede_check_csum(u16 flag) return qede_check_tunn_csum(flag); } +static bool qede_pkt_is_ip_fragmented(struct eth_fast_path_rx_reg_cqe *cqe, + u16 flag) +{ + u8 tun_pars_flg = cqe->tunnel_pars_flags.flags; + + if ((tun_pars_flg & (ETH_TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_MASK << + ETH_TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_SHIFT)) || + (flag & (PARSING_AND_ERR_FLAGS_IPV4FRAG_MASK << + PARSING_AND_ERR_FLAGS_IPV4FRAG_SHIFT))) + return true; + + return false; +} + static int qede_rx_int(struct qede_fastpath *fp, int budget) { struct qede_dev *edev = fp->edev; @@ -1426,6 +1440,12 @@ static int qede_rx_int(struct qede_fastpath *fp, int budget) csum_flag = qede_check_csum(parse_flag); if (unlikely(csum_flag == QEDE_CSUM_ERROR)) { + if (qede_pkt_is_ip_fragmented(&cqe->fast_path_regular, + parse_flag)) { + rxq->rx_ip_frags++; + goto alloc_skb; + } + DP_NOTICE(edev, "CQE in CONS = %u has error, flags = %x, dropping incoming packet\n", sw_comp_cons, parse_flag); @@ -1434,6 +1454,7 @@ static int qede_rx_int(struct qede_fastpath *fp, int budget) goto next_cqe; } +alloc_skb: skb = netdev_alloc_skb(edev->ndev, QEDE_RX_HDR_SIZE); if (unlikely(!skb)) { DP_NOTICE(edev, -- cgit v0.10.2 From c774169d8f7e1c05a72035756000ff590db08501 Mon Sep 17 00:00:00 2001 From: Manish Chopra Date: Thu, 30 Jun 2016 02:35:19 -0400 Subject: qede: qede_poll refactoring This patch cleanups qede_poll() routine a bit and allows qede_poll() to do single iteration to handle TX completion [As under heavy TX load qede_poll() might run for indefinite time in the while(1) loop for TX completion processing and cause CPU stuck]. Signed-off-by: Manish Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index ae4c53b..9bc4c9f 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -1597,56 +1597,49 @@ next_cqe: /* don't consume bd rx buffer */ static int qede_poll(struct napi_struct *napi, int budget) { - int work_done = 0; struct qede_fastpath *fp = container_of(napi, struct qede_fastpath, - napi); + napi); struct qede_dev *edev = fp->edev; + int rx_work_done = 0; + u8 tc; - while (1) { - u8 tc; - - for (tc = 0; tc < edev->num_tc; tc++) - if (qede_txq_has_work(&fp->txqs[tc])) - qede_tx_int(edev, &fp->txqs[tc]); - - if (qede_has_rx_work(fp->rxq)) { - work_done += qede_rx_int(fp, budget - work_done); - - /* must not complete if we consumed full budget */ - if (work_done >= budget) - break; - } + for (tc = 0; tc < edev->num_tc; tc++) + if (qede_txq_has_work(&fp->txqs[tc])) + qede_tx_int(edev, &fp->txqs[tc]); + + rx_work_done = qede_has_rx_work(fp->rxq) ? + qede_rx_int(fp, budget) : 0; + if (rx_work_done < budget) { + qed_sb_update_sb_idx(fp->sb_info); + /* *_has_*_work() reads the status block, + * thus we need to ensure that status block indices + * have been actually read (qed_sb_update_sb_idx) + * prior to this check (*_has_*_work) so that + * we won't write the "newer" value of the status block + * to HW (if there was a DMA right after + * qede_has_rx_work and if there is no rmb, the memory + * reading (qed_sb_update_sb_idx) may be postponed + * to right before *_ack_sb). In this case there + * will never be another interrupt until there is + * another update of the status block, while there + * is still unhandled work. + */ + rmb(); /* Fall out from the NAPI loop if needed */ - if (!(qede_has_rx_work(fp->rxq) || qede_has_tx_work(fp))) { - qed_sb_update_sb_idx(fp->sb_info); - /* *_has_*_work() reads the status block, - * thus we need to ensure that status block indices - * have been actually read (qed_sb_update_sb_idx) - * prior to this check (*_has_*_work) so that - * we won't write the "newer" value of the status block - * to HW (if there was a DMA right after - * qede_has_rx_work and if there is no rmb, the memory - * reading (qed_sb_update_sb_idx) may be postponed - * to right before *_ack_sb). In this case there - * will never be another interrupt until there is - * another update of the status block, while there - * is still unhandled work. - */ - rmb(); - - if (!(qede_has_rx_work(fp->rxq) || - qede_has_tx_work(fp))) { - napi_complete(napi); - /* Update and reenable interrupts */ - qed_sb_ack(fp->sb_info, IGU_INT_ENABLE, - 1 /*update*/); - break; - } + if (!(qede_has_rx_work(fp->rxq) || + qede_has_tx_work(fp))) { + napi_complete(napi); + + /* Update and reenable interrupts */ + qed_sb_ack(fp->sb_info, IGU_INT_ENABLE, + 1 /*update*/); + } else { + rx_work_done = budget; } } - return work_done; + return rx_work_done; } static irqreturn_t qede_msix_fp_int(int irq, void *fp_cookie) -- cgit v0.10.2 From 312e06761c9976c81bd1e487a2c03dce7c07c102 Mon Sep 17 00:00:00 2001 From: Manish Chopra Date: Thu, 30 Jun 2016 02:35:20 -0400 Subject: qede: Utilize xmit_more This patch uses xmit_more optimization to reduce number of TX doorbells write per packet. Signed-off-by: Manish Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 9bc4c9f..ea2df14 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -485,6 +485,24 @@ static bool qede_pkt_req_lin(struct qede_dev *edev, struct sk_buff *skb, } #endif +static inline void qede_update_tx_producer(struct qede_tx_queue *txq) +{ + /* wmb makes sure that the BDs data is updated before updating the + * producer, otherwise FW may read old data from the BDs. + */ + wmb(); + barrier(); + writel(txq->tx_db.raw, txq->doorbell_addr); + + /* mmiowb is needed to synchronize doorbell writes from more than one + * processor. It guarantees that the write arrives to the device before + * the queue lock is released and another start_xmit is called (possibly + * on another CPU). Without this barrier, the next doorbell can bypass + * this doorbell. This is applicable to IA64/Altix systems. + */ + mmiowb(); +} + /* Main transmit function */ static netdev_tx_t qede_start_xmit(struct sk_buff *skb, @@ -543,6 +561,7 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, if (unlikely(dma_mapping_error(&edev->pdev->dev, mapping))) { DP_NOTICE(edev, "SKB mapping failed\n"); qede_free_failed_tx_pkt(edev, txq, first_bd, 0, false); + qede_update_tx_producer(txq); return NETDEV_TX_OK; } nbd++; @@ -657,6 +676,7 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, if (rc) { qede_free_failed_tx_pkt(edev, txq, first_bd, nbd, data_split); + qede_update_tx_producer(txq); return NETDEV_TX_OK; } @@ -681,6 +701,7 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, if (rc) { qede_free_failed_tx_pkt(edev, txq, first_bd, nbd, data_split); + qede_update_tx_producer(txq); return NETDEV_TX_OK; } } @@ -701,20 +722,8 @@ netdev_tx_t qede_start_xmit(struct sk_buff *skb, txq->tx_db.data.bd_prod = cpu_to_le16(qed_chain_get_prod_idx(&txq->tx_pbl)); - /* wmb makes sure that the BDs data is updated before updating the - * producer, otherwise FW may read old data from the BDs. - */ - wmb(); - barrier(); - writel(txq->tx_db.raw, txq->doorbell_addr); - - /* mmiowb is needed to synchronize doorbell writes from more than one - * processor. It guarantees that the write arrives to the device before - * the queue lock is released and another start_xmit is called (possibly - * on another CPU). Without this barrier, the next doorbell can bypass - * this doorbell. This is applicable to IA64/Altix systems. - */ - mmiowb(); + if (!skb->xmit_more || netif_tx_queue_stopped(netdev_txq)) + qede_update_tx_producer(txq); if (unlikely(qed_chain_get_elem_left(&txq->tx_pbl) < (MAX_SKB_FRAGS + 1))) { -- cgit v0.10.2 From 3d789994b0ae5a148e9d0c3317be21c597462d86 Mon Sep 17 00:00:00 2001 From: Manish Chopra Date: Thu, 30 Jun 2016 02:35:21 -0400 Subject: qede: Add get/set rx copy break tunable support Signed-off-by: Manish Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 677213e..4e3f1a0 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -143,6 +143,8 @@ struct qede_dev { struct mutex qede_lock; u32 state; /* Protected by qede_lock */ u16 rx_buf_size; + u32 rx_copybreak; + /* L2 header size + 2*VLANs (8 bytes) + LLC SNAP (8 bytes) */ #define ETH_OVERHEAD (ETH_HLEN + 8 + 8) /* Max supported alignment is 256 (8 shift) @@ -333,6 +335,7 @@ void qede_recycle_rx_bd_ring(struct qede_rx_queue *rxq, struct qede_dev *edev, #define NUM_TX_BDS_MIN 128 #define NUM_TX_BDS_DEF NUM_TX_BDS_MAX +#define QEDE_MIN_PKT_LEN 64 #define QEDE_RX_HDR_SIZE 256 #define for_each_rss(i) for (i = 0; i < edev->num_rss; i++) diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index 85633cf..f8492ca 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -1185,6 +1185,48 @@ static void qede_self_test(struct net_device *dev, } } +static int qede_set_tunable(struct net_device *dev, + const struct ethtool_tunable *tuna, + const void *data) +{ + struct qede_dev *edev = netdev_priv(dev); + u32 val; + + switch (tuna->id) { + case ETHTOOL_RX_COPYBREAK: + val = *(u32 *)data; + if (val < QEDE_MIN_PKT_LEN || val > QEDE_RX_HDR_SIZE) { + DP_VERBOSE(edev, QED_MSG_DEBUG, + "Invalid rx copy break value, range is [%u, %u]", + QEDE_MIN_PKT_LEN, QEDE_RX_HDR_SIZE); + return -EINVAL; + } + + edev->rx_copybreak = *(u32 *)data; + break; + default: + return -EOPNOTSUPP; + } + + return 0; +} + +static int qede_get_tunable(struct net_device *dev, + const struct ethtool_tunable *tuna, void *data) +{ + struct qede_dev *edev = netdev_priv(dev); + + switch (tuna->id) { + case ETHTOOL_RX_COPYBREAK: + *(u32 *)data = edev->rx_copybreak; + break; + default: + return -EOPNOTSUPP; + } + + return 0; +} + static const struct ethtool_ops qede_ethtool_ops = { .get_settings = qede_get_settings, .set_settings = qede_set_settings, @@ -1213,6 +1255,8 @@ static const struct ethtool_ops qede_ethtool_ops = { .get_channels = qede_get_channels, .set_channels = qede_set_channels, .self_test = qede_self_test, + .get_tunable = qede_get_tunable, + .set_tunable = qede_set_tunable, }; static const struct ethtool_ops qede_vf_ethtool_ops = { @@ -1235,6 +1279,8 @@ static const struct ethtool_ops qede_vf_ethtool_ops = { .set_rxfh = qede_set_rxfh, .get_channels = qede_get_channels, .set_channels = qede_set_channels, + .get_tunable = qede_get_tunable, + .set_tunable = qede_set_tunable, }; void qede_set_ethtool_ops(struct net_device *dev) diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index ea2df14..91e7bb0 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -1474,7 +1474,7 @@ alloc_skb: } /* Copy data into SKB */ - if (len + pad <= QEDE_RX_HDR_SIZE) { + if (len + pad <= edev->rx_copybreak) { memcpy(skb_put(skb, len), page_address(data) + pad + sw_rx_data->page_offset, len); @@ -2519,6 +2519,7 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); mutex_init(&edev->qede_lock); + edev->rx_copybreak = QEDE_RX_HDR_SIZE; DP_INFO(edev, "Ending successfully qede probe\n"); -- cgit v0.10.2 From 831a8e6c40e6dfd8742d6b638ca9fda088355ec3 Mon Sep 17 00:00:00 2001 From: Manish Chopra Date: Thu, 30 Jun 2016 02:35:22 -0400 Subject: qede: Bump up driver version to 8.10.1.20 Signed-off-by: Manish Chopra Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 4e3f1a0..02b06d4 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -24,7 +24,7 @@ #include #define QEDE_MAJOR_VERSION 8 -#define QEDE_MINOR_VERSION 7 +#define QEDE_MINOR_VERSION 10 #define QEDE_REVISION_VERSION 1 #define QEDE_ENGINEERING_VERSION 20 #define DRV_MODULE_VERSION __stringify(QEDE_MAJOR_VERSION) "." \ -- cgit v0.10.2 From 466fc793da3aea0caf365e7fadc67473e4cdaa80 Mon Sep 17 00:00:00 2001 From: Amitoj Kaur Chawla Date: Thu, 30 Jun 2016 14:14:01 +0530 Subject: atm: horizon: Use setup_timer Convert a call to init_timer and accompanying intializations of the timer's data and function fields to a call to setup_timer. The Coccinelle semantic patch that fixes this problem is as follows: @@ expression t,d,f,e1; identifier x1; statement S1; @@ ( -t.data = d; | -t.function = f; | -init_timer(&t); +setup_timer(&t,f,d); | -init_timer_on_stack(&t); +setup_timer_on_stack(&t,f,d); ) <... when != S1 t.x1 = e1; ...> Signed-off-by: Amitoj Kaur Chawla Signed-off-by: David S. Miller diff --git a/drivers/atm/horizon.c b/drivers/atm/horizon.c index 527bbd5..5fc81e2 100644 --- a/drivers/atm/horizon.c +++ b/drivers/atm/horizon.c @@ -2795,9 +2795,7 @@ static int hrz_probe(struct pci_dev *pci_dev, dev->atm_dev->ci_range.vpi_bits = vpi_bits; dev->atm_dev->ci_range.vci_bits = 10-vpi_bits; - init_timer(&dev->housekeeping); - dev->housekeeping.function = do_housekeeping; - dev->housekeeping.data = (unsigned long) dev; + setup_timer(&dev->housekeeping, do_housekeeping, (unsigned long) dev); mod_timer(&dev->housekeeping, jiffies); out: -- cgit v0.10.2 From 4bdd4dfe7a893594a75ca324057f7010b1762bd2 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Thu, 7 Apr 2016 16:44:42 +0300 Subject: iwlwifi: advertise maximal MPDU length when Rx MQ is supported The new hardware that supports multiple queue also de-aggregates A-MSDUs. This means that we can advertise the maximal size of A-MSDUs regardless of the receive buffer's size. In order to be able to forcefully use a lower A-MSDU size, add a default value for the module parameter. Pre-9000 will have a default of 4K, and 9000 will have 12K. Setting the amsdu_size module parameter to 4K will limit the A-MSDU on 9000 as well. Signed-off-by: Emmanuel Grumbach Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/main.c b/drivers/net/wireless/intel/iwlwifi/dvm/main.c index 37b32a6..5915914 100644 --- a/drivers/net/wireless/intel/iwlwifi/dvm/main.c +++ b/drivers/net/wireless/intel/iwlwifi/dvm/main.c @@ -1317,6 +1317,7 @@ static struct iwl_op_mode *iwl_op_mode_dvm_start(struct iwl_trans *trans, trans_cfg.n_no_reclaim_cmds = ARRAY_SIZE(no_reclaim_cmds); switch (iwlwifi_mod_params.amsdu_size) { + case IWL_AMSDU_DEF: case IWL_AMSDU_4K: trans_cfg.rx_buf_size = IWL_AMSDU_4K; break; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c index 60a6c36..5c5b9f2 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c @@ -1655,7 +1655,8 @@ MODULE_PARM_DESC(11n_disable, "disable 11n functionality, bitmap: 1: full, 2: disable agg TX, 4: disable agg RX, 8 enable agg TX"); module_param_named(amsdu_size, iwlwifi_mod_params.amsdu_size, int, S_IRUGO); -MODULE_PARM_DESC(amsdu_size, "amsdu size 0:4K 1:8K 2:12K (default 0)"); +MODULE_PARM_DESC(amsdu_size, + "amsdu size 0: 12K for multi Rx queue devices, 4K for other devices 1:4K 2:8K 3:12K (default 0)"); module_param_named(fw_restart, iwlwifi_mod_params.restart_fw, bool, S_IRUGO); MODULE_PARM_DESC(fw_restart, "restart firmware in case of error (default true)"); diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.c index bf1b69a..3199d34 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.c @@ -766,7 +766,9 @@ void iwl_init_ht_hw_capab(const struct iwl_cfg *cfg, if (cfg->ht_params->ldpc) ht_info->cap |= IEEE80211_HT_CAP_LDPC_CODING; - if (iwlwifi_mod_params.amsdu_size >= IWL_AMSDU_8K) + if ((cfg->mq_rx_supported && + iwlwifi_mod_params.amsdu_size != IWL_AMSDU_4K) || + iwlwifi_mod_params.amsdu_size >= IWL_AMSDU_8K) ht_info->cap |= IEEE80211_HT_CAP_MAX_AMSDU; ht_info->ampdu_factor = cfg->max_ht_ampdu_exponent; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h index 6c5c2f9..0379899 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h @@ -87,9 +87,10 @@ enum iwl_disable_11n { }; enum iwl_amsdu_size { - IWL_AMSDU_4K = 0, - IWL_AMSDU_8K = 1, - IWL_AMSDU_12K = 2, + IWL_AMSDU_DEF = 0, + IWL_AMSDU_4K = 1, + IWL_AMSDU_8K = 2, + IWL_AMSDU_12K = 3, }; enum iwl_uapsd_disable { @@ -105,7 +106,7 @@ enum iwl_uapsd_disable { * @sw_crypto: using hardware encryption, default = 0 * @disable_11n: disable 11n capabilities, default = 0, * use IWL_[DIS,EN]ABLE_HT_* constants - * @amsdu_size: enable 8K amsdu size, default = 4K. enum iwl_amsdu_size. + * @amsdu_size: See &enum iwl_amsdu_size. * @restart_fw: restart firmware, default = 1 * @bt_coex_active: enable bt coex, default = true * @led_mode: system default, default = 0 diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c index 21653fe..43f8f7d 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c @@ -397,6 +397,13 @@ static void iwl_init_vht_hw_capab(const struct iwl_cfg *cfg, vht_cap->cap |= IEEE80211_VHT_CAP_TX_ANTENNA_PATTERN; switch (iwlwifi_mod_params.amsdu_size) { + case IWL_AMSDU_DEF: + if (cfg->mq_rx_supported) + vht_cap->cap |= + IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_11454; + else + vht_cap->cap |= IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_3895; + break; case IWL_AMSDU_4K: vht_cap->cap |= IEEE80211_VHT_CAP_MAX_MPDU_LENGTH_3895; break; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c index a68054f..64d2b4f 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c @@ -603,6 +603,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg, trans_cfg.no_reclaim_cmds = no_reclaim_cmds; trans_cfg.n_no_reclaim_cmds = ARRAY_SIZE(no_reclaim_cmds); switch (iwlwifi_mod_params.amsdu_size) { + case IWL_AMSDU_DEF: case IWL_AMSDU_4K: trans_cfg.rx_buf_size = IWL_AMSDU_4K; break; @@ -617,6 +618,10 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg, iwlwifi_mod_params.amsdu_size); trans_cfg.rx_buf_size = IWL_AMSDU_4K; } + + /* the hardware splits the A-MSDU */ + if (mvm->cfg->mq_rx_supported) + trans_cfg.rx_buf_size = IWL_AMSDU_4K; trans_cfg.wide_cmd_header = fw_has_api(&mvm->fw->ucode_capa, IWL_UCODE_TLV_API_WIDE_CMD_HDR); -- cgit v0.10.2 From e7e14089e9a8e84fd8e2885ae9cacba2c30a3403 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Sun, 17 Apr 2016 14:15:17 +0300 Subject: iwlwifi: mvm: do not trust NSSN for amsdu sub-frames We cannot trust NSSN for AMSDU sub-frames that are not the last. The reason is that NSSN advances on the first sub-frame, and may cause the reorder buffer to advance before all the sub-frames arrive. Example: Reorder buffer contains SN 0 & 2. We receive AMSDU with SN 1 and NSSN for first sub frame 3. The result us that driver releases SN 0,1, 2. When sub-frame 1 arrives - reorder buffer is already ahead and it will be dropped. If the last sub-frame is not on this queue - we will get frame release notification with up to date NSSN. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c index ac2c571..0f26c70 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c @@ -587,6 +587,8 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm, struct sk_buff *tail; u32 reorder = le32_to_cpu(desc->reorder_data); bool amsdu = desc->mac_flags2 & IWL_RX_MPDU_MFLG2_AMSDU; + bool last_subframe = + desc->amsdu_info & IWL_RX_MPDU_AMSDU_LAST_SUBFRAME; u8 tid = *ieee80211_get_qos_ctl(hdr) & IEEE80211_QOS_CTL_TID_MASK; u8 sub_frame_idx = desc->amsdu_info & IWL_RX_MPDU_AMSDU_SUBFRAME_IDX_MASK; @@ -651,7 +653,8 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm, /* release immediately if allowed by nssn and no stored frames */ if (!buffer->num_stored && ieee80211_sn_less(sn, nssn)) { if (iwl_mvm_is_sn_less(buffer->head_sn, nssn, - buffer->buf_size)) + buffer->buf_size) && + (!amsdu || last_subframe)) buffer->head_sn = nssn; /* No need to update AMSDU last SN - we are moving the head */ spin_unlock_bh(&buffer->lock); @@ -685,7 +688,20 @@ static bool iwl_mvm_reorder(struct iwl_mvm *mvm, buffer->last_sub_index = sub_frame_idx; } - iwl_mvm_release_frames(mvm, sta, napi, buffer, nssn); + /* + * We cannot trust NSSN for AMSDU sub-frames that are not the last. + * The reason is that NSSN advances on the first sub-frame, and may + * cause the reorder buffer to advance before all the sub-frames arrive. + * Example: reorder buffer contains SN 0 & 2, and we receive AMSDU with + * SN 1. NSSN for first sub frame will be 3 with the result of driver + * releasing SN 0,1, 2. When sub-frame 1 arrives - reorder buffer is + * already ahead and it will be dropped. + * If the last sub-frame is not on this queue - we will get frame + * release notification with up to date NSSN. + */ + if (!amsdu || last_subframe) + iwl_mvm_release_frames(mvm, sta, napi, buffer, nssn); + spin_unlock_bh(&buffer->lock); return true; -- cgit v0.10.2 From 7bde4c683127d62a0c6342075fae8b46b7b000f5 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 13 Apr 2016 14:31:40 +0200 Subject: iwlwifi: mvm: fix comment indentation Somehow we ended up without leading spaces here, fix that. Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h index 1ca8e49..bb8ff40 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h @@ -435,26 +435,26 @@ struct iwl_rxq_sync_notification { } __packed; /* MULTI_QUEUE_DRV_SYNC_HDR_CMD_API_S_VER_1 */ /** -* Internal message identifier -* -* @IWL_MVM_RXQ_EMPTY: empty sync notification -* @IWL_MVM_RXQ_NOTIF_DEL_BA: notify RSS queues of delBA -*/ + * Internal message identifier + * + * @IWL_MVM_RXQ_EMPTY: empty sync notification + * @IWL_MVM_RXQ_NOTIF_DEL_BA: notify RSS queues of delBA + */ enum iwl_mvm_rxq_notif_type { IWL_MVM_RXQ_EMPTY, IWL_MVM_RXQ_NOTIF_DEL_BA, }; /** -* struct iwl_mvm_internal_rxq_notif - Internal representation of the data sent -* in &iwl_rxq_sync_cmd. Should be DWORD aligned. -* FW is agnostic to the payload, so there are no endianity requirements. -* -* @type: value from &iwl_mvm_rxq_notif_type -* @sync: ctrl path is waiting for all notifications to be received -* @cookie: internal cookie to identify old notifications -* @data: payload -*/ + * struct iwl_mvm_internal_rxq_notif - Internal representation of the data sent + * in &iwl_rxq_sync_cmd. Should be DWORD aligned. + * FW is agnostic to the payload, so there are no endianity requirements. + * + * @type: value from &iwl_mvm_rxq_notif_type + * @sync: ctrl path is waiting for all notifications to be received + * @cookie: internal cookie to identify old notifications + * @data: payload + */ struct iwl_mvm_internal_rxq_notif { u16 type; u16 sync; -- cgit v0.10.2 From 468b021b944922e8fe0a30b6b6e0532bb95e4edc Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 24 Jun 2016 19:48:30 +0200 Subject: netfilter: x_tables: simplify ip{6}table_mangle_hook() No need for a special case to handle NF_INET_POST_ROUTING, this is basically the same handling as for prerouting, input, forward. Signed-off-by: Pablo Neira Ayuso diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index 57fc97c..aebdb33 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -87,10 +87,6 @@ iptable_mangle_hook(void *priv, { if (state->hook == NF_INET_LOCAL_OUT) return ipt_mangle_out(skb, state); - if (state->hook == NF_INET_POST_ROUTING) - return ipt_do_table(skb, state, - state->net->ipv4.iptable_mangle); - /* PREROUTING/INPUT/FORWARD: */ return ipt_do_table(skb, state, state->net->ipv4.iptable_mangle); } diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index cb2b288..2b1a9dc 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -83,10 +83,6 @@ ip6table_mangle_hook(void *priv, struct sk_buff *skb, { if (state->hook == NF_INET_LOCAL_OUT) return ip6t_mangle_out(skb, state); - if (state->hook == NF_INET_POST_ROUTING) - return ip6t_do_table(skb, state, - state->net->ipv6.ip6table_mangle); - /* INPUT/FORWARD */ return ip6t_do_table(skb, state, state->net->ipv6.ip6table_mangle); } -- cgit v0.10.2 From 4ae89ad92477219b504a49966ee010fe8dcb85af Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Fri, 24 Jun 2016 11:32:26 -0700 Subject: etherdevice.h & bridge: netfilter: Add and use ether_addr_equal_masked There are code duplications of a masked ethernet address comparison here so make it a separate function instead. Miscellanea: o Neaten alignment of FWINV macro uses to make it clearer for the reader Signed-off-by: Joe Perches Acked-by: David S. Miller Signed-off-by: Pablo Neira Ayuso diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index 37ff4a6..6fec9e8 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -374,6 +374,29 @@ static inline bool ether_addr_equal_unaligned(const u8 *addr1, const u8 *addr2) } /** + * ether_addr_equal_masked - Compare two Ethernet addresses with a mask + * @addr1: Pointer to a six-byte array containing the 1st Ethernet address + * @addr2: Pointer to a six-byte array containing the 2nd Ethernet address + * @mask: Pointer to a six-byte array containing the Ethernet address bitmask + * + * Compare two Ethernet addresses with a mask, returns true if for every bit + * set in the bitmask the equivalent bits in the ethernet addresses are equal. + * Using a mask with all bits set is a slower ether_addr_equal. + */ +static inline bool ether_addr_equal_masked(const u8 *addr1, const u8 *addr2, + const u8 *mask) +{ + int i; + + for (i = 0; i < ETH_ALEN; i++) { + if ((addr1[i] ^ addr2[i]) & mask[i]) + return false; + } + + return true; +} + +/** * is_etherdev_addr - Tell if given Ethernet address belongs to the device. * @dev: Pointer to a device structure * @addr: Pointer to a six-byte array containing the Ethernet address diff --git a/net/bridge/netfilter/ebt_arp.c b/net/bridge/netfilter/ebt_arp.c index cd457b8..cca0a89 100644 --- a/net/bridge/netfilter/ebt_arp.c +++ b/net/bridge/netfilter/ebt_arp.c @@ -65,7 +65,6 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) if (info->bitmask & (EBT_ARP_SRC_MAC | EBT_ARP_DST_MAC)) { const unsigned char *mp; unsigned char _mac[ETH_ALEN]; - uint8_t verdict, i; if (ah->ar_hln != ETH_ALEN || ah->ar_hrd != htons(ARPHRD_ETHER)) return false; @@ -74,11 +73,9 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) sizeof(_mac), &_mac); if (mp == NULL) return false; - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (mp[i] ^ info->smaddr[i]) & - info->smmsk[i]; - if (FWINV(verdict != 0, EBT_ARP_SRC_MAC)) + if (FWINV(!ether_addr_equal_masked(mp, info->smaddr, + info->smmsk), + EBT_ARP_SRC_MAC)) return false; } @@ -88,11 +85,9 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) sizeof(_mac), &_mac); if (mp == NULL) return false; - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (mp[i] ^ info->dmaddr[i]) & - info->dmmsk[i]; - if (FWINV(verdict != 0, EBT_ARP_DST_MAC)) + if (FWINV(!ether_addr_equal_masked(mp, info->dmaddr, + info->dmmsk), + EBT_ARP_DST_MAC)) return false; } } diff --git a/net/bridge/netfilter/ebt_stp.c b/net/bridge/netfilter/ebt_stp.c index e77f90b..45f73d5 100644 --- a/net/bridge/netfilter/ebt_stp.c +++ b/net/bridge/netfilter/ebt_stp.c @@ -46,7 +46,6 @@ static bool ebt_filter_config(const struct ebt_stp_info *info, const struct ebt_stp_config_info *c; u16 v16; u32 v32; - int verdict, i; c = &info->config; if ((info->bitmask & EBT_STP_FLAGS) && @@ -54,66 +53,62 @@ static bool ebt_filter_config(const struct ebt_stp_info *info, return false; if (info->bitmask & EBT_STP_ROOTPRIO) { v16 = NR16(stpc->root); - if (FWINV(v16 < c->root_priol || - v16 > c->root_priou, EBT_STP_ROOTPRIO)) + if (FWINV(v16 < c->root_priol || v16 > c->root_priou, + EBT_STP_ROOTPRIO)) return false; } if (info->bitmask & EBT_STP_ROOTADDR) { - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (stpc->root[2+i] ^ c->root_addr[i]) & - c->root_addrmsk[i]; - if (FWINV(verdict != 0, EBT_STP_ROOTADDR)) + if (FWINV(!ether_addr_equal_masked(&stpc->root[2], c->root_addr, + c->root_addrmsk), + EBT_STP_ROOTADDR)) return false; } if (info->bitmask & EBT_STP_ROOTCOST) { v32 = NR32(stpc->root_cost); - if (FWINV(v32 < c->root_costl || - v32 > c->root_costu, EBT_STP_ROOTCOST)) + if (FWINV(v32 < c->root_costl || v32 > c->root_costu, + EBT_STP_ROOTCOST)) return false; } if (info->bitmask & EBT_STP_SENDERPRIO) { v16 = NR16(stpc->sender); - if (FWINV(v16 < c->sender_priol || - v16 > c->sender_priou, EBT_STP_SENDERPRIO)) + if (FWINV(v16 < c->sender_priol || v16 > c->sender_priou, + EBT_STP_SENDERPRIO)) return false; } if (info->bitmask & EBT_STP_SENDERADDR) { - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (stpc->sender[2+i] ^ c->sender_addr[i]) & - c->sender_addrmsk[i]; - if (FWINV(verdict != 0, EBT_STP_SENDERADDR)) + if (FWINV(!ether_addr_equal_masked(&stpc->sender[2], + c->sender_addr, + c->sender_addrmsk), + EBT_STP_SENDERADDR)) return false; } if (info->bitmask & EBT_STP_PORT) { v16 = NR16(stpc->port); - if (FWINV(v16 < c->portl || - v16 > c->portu, EBT_STP_PORT)) + if (FWINV(v16 < c->portl || v16 > c->portu, EBT_STP_PORT)) return false; } if (info->bitmask & EBT_STP_MSGAGE) { v16 = NR16(stpc->msg_age); - if (FWINV(v16 < c->msg_agel || - v16 > c->msg_ageu, EBT_STP_MSGAGE)) + if (FWINV(v16 < c->msg_agel || v16 > c->msg_ageu, + EBT_STP_MSGAGE)) return false; } if (info->bitmask & EBT_STP_MAXAGE) { v16 = NR16(stpc->max_age); - if (FWINV(v16 < c->max_agel || - v16 > c->max_ageu, EBT_STP_MAXAGE)) + if (FWINV(v16 < c->max_agel || v16 > c->max_ageu, + EBT_STP_MAXAGE)) return false; } if (info->bitmask & EBT_STP_HELLOTIME) { v16 = NR16(stpc->hello_time); - if (FWINV(v16 < c->hello_timel || - v16 > c->hello_timeu, EBT_STP_HELLOTIME)) + if (FWINV(v16 < c->hello_timel || v16 > c->hello_timeu, + EBT_STP_HELLOTIME)) return false; } if (info->bitmask & EBT_STP_FWDD) { v16 = NR16(stpc->forward_delay); - if (FWINV(v16 < c->forward_delayl || - v16 > c->forward_delayu, EBT_STP_FWDD)) + if (FWINV(v16 < c->forward_delayl || v16 > c->forward_delayu, + EBT_STP_FWDD)) return false; } return true; diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 5a61f35..5721a25 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -130,7 +130,6 @@ ebt_basic_match(const struct ebt_entry *e, const struct sk_buff *skb, const struct ethhdr *h = eth_hdr(skb); const struct net_bridge_port *p; __be16 ethproto; - int verdict, i; if (skb_vlan_tag_present(skb)) ethproto = htons(ETH_P_8021Q); @@ -157,19 +156,15 @@ ebt_basic_match(const struct ebt_entry *e, const struct sk_buff *skb, return 1; if (e->bitmask & EBT_SOURCEMAC) { - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (h->h_source[i] ^ e->sourcemac[i]) & - e->sourcemsk[i]; - if (FWINV2(verdict != 0, EBT_ISOURCE)) + if (FWINV2(!ether_addr_equal_masked(h->h_source, + e->sourcemac, e->sourcemsk), + EBT_ISOURCE)) return 1; } if (e->bitmask & EBT_DESTMAC) { - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (h->h_dest[i] ^ e->destmac[i]) & - e->destmsk[i]; - if (FWINV2(verdict != 0, EBT_IDEST)) + if (FWINV2(!ether_addr_equal_masked(h->h_dest, + e->destmac, e->destmsk), + EBT_IDEST)) return 1; } return 0; -- cgit v0.10.2 From f1504307b9ab60e73ba31eece4be8298ebc9c1b7 Mon Sep 17 00:00:00 2001 From: Moritz Sichert Date: Thu, 30 Jun 2016 11:46:28 +0200 Subject: netfilter: Remove references to obsolete CONFIG_IP_ROUTE_FWMARK This option was removed in commit 47dcf0cb1005 ("[NET]: Rethink mark field in struct flowi"). Signed-off-by: Moritz Sichert Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 95e757c..9266cee 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -609,9 +609,8 @@ config NETFILTER_XT_MARK The target allows you to create rules in the "mangle" table which alter the netfilter mark (nfmark) field associated with the packet. - Prior to routing, the nfmark can influence the routing method (see - "Use netfilter MARK value as routing key") and can also be used by - other subsystems to change their behavior. + Prior to routing, the nfmark can influence the routing method and can + also be used by other subsystems to change their behavior. config NETFILTER_XT_CONNMARK tristate 'ctmark target and match support' @@ -753,9 +752,8 @@ config NETFILTER_XT_TARGET_HMARK The target allows you to create rules in the "raw" and "mangle" tables which set the skbuff mark by means of hash calculation within a given - range. The nfmark can influence the routing method (see "Use netfilter - MARK value as routing key") and can also be used by other subsystems to - change their behaviour. + range. The nfmark can influence the routing method and can also be used + by other subsystems to change their behaviour. To compile it as a module, choose M here. If unsure, say N. -- cgit v0.10.2 From c5241b0c8c0bb2bfd69effaa81e30fa26a16adda Mon Sep 17 00:00:00 2001 From: Avraham Stern Date: Wed, 20 Apr 2016 09:29:18 +0300 Subject: iwlwifi: rename CAPA_P2P_STANDALONE_UAPSD to CAPA_P2P_SCM_UAPSD Ucode capability bit 26 indicates support for UAPSD on P2P interface even with a simultaneous BSS station interface, as long as both interfaces are in the same binding. Change the name of the capability bit to reflect that. Signed-off-by: Avraham Stern Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h index 843232b..3f5029e 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h @@ -299,7 +299,8 @@ typedef unsigned int __bitwise__ iwl_ucode_tlv_capa_t; * @IWL_UCODE_TLV_CAPA_DC2DC_SUPPORT: supports DC2DC Command * @IWL_UCODE_TLV_CAPA_CSUM_SUPPORT: supports TCP Checksum Offload * @IWL_UCODE_TLV_CAPA_RADIO_BEACON_STATS: support radio and beacon statistics - * @IWL_UCODE_TLV_CAPA_P2P_STANDALONE_UAPSD: support p2p standalone U-APSD + * @IWL_UCODE_TLV_CAPA_P2P_SCM_UAPSD: supports U-APSD on p2p interface when it + * is standalone or with a BSS station interface in the same binding. * @IWL_UCODE_TLV_CAPA_BT_COEX_PLCR: enabled BT Coex packet level co-running * @IWL_UCODE_TLV_CAPA_LAR_MULTI_MCC: ucode supports LAR updates with different * sources for the MCC. This TLV bit is a future replacement to @@ -345,7 +346,7 @@ enum iwl_ucode_tlv_capa { IWL_UCODE_TLV_CAPA_DC2DC_CONFIG_SUPPORT = (__force iwl_ucode_tlv_capa_t)19, IWL_UCODE_TLV_CAPA_CSUM_SUPPORT = (__force iwl_ucode_tlv_capa_t)21, IWL_UCODE_TLV_CAPA_RADIO_BEACON_STATS = (__force iwl_ucode_tlv_capa_t)22, - IWL_UCODE_TLV_CAPA_P2P_STANDALONE_UAPSD = (__force iwl_ucode_tlv_capa_t)26, + IWL_UCODE_TLV_CAPA_P2P_SCM_UAPSD = (__force iwl_ucode_tlv_capa_t)26, IWL_UCODE_TLV_CAPA_BT_COEX_PLCR = (__force iwl_ucode_tlv_capa_t)28, IWL_UCODE_TLV_CAPA_LAR_MULTI_MCC = (__force iwl_ucode_tlv_capa_t)29, IWL_UCODE_TLV_CAPA_BT_COEX_RRC = (__force iwl_ucode_tlv_capa_t)30, diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index e5f267b..b7014bc 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -2360,7 +2360,7 @@ static void iwl_mvm_check_uapsd(struct iwl_mvm *mvm, struct ieee80211_vif *vif, if (!(mvm->fw->ucode_capa.flags & IWL_UCODE_TLV_FLAGS_UAPSD_SUPPORT)) return; - if (vif->p2p && !iwl_mvm_is_p2p_standalone_uapsd_supported(mvm)) { + if (vif->p2p && !iwl_mvm_is_p2p_scm_uapsd_supported(mvm)) { vif->driver_flags &= ~IEEE80211_VIF_SUPPORTS_UAPSD; return; } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index ffbd41d..4843e02 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -1158,10 +1158,10 @@ static inline bool iwl_mvm_is_mplut_supported(struct iwl_mvm *mvm) } static inline -bool iwl_mvm_is_p2p_standalone_uapsd_supported(struct iwl_mvm *mvm) +bool iwl_mvm_is_p2p_scm_uapsd_supported(struct iwl_mvm *mvm) { return fw_has_capa(&mvm->fw->ucode_capa, - IWL_UCODE_TLV_CAPA_P2P_STANDALONE_UAPSD) && + IWL_UCODE_TLV_CAPA_P2P_SCM_UAPSD) && !(iwlwifi_mod_params.uapsd_disable & IWL_DISABLE_UAPSD_P2P_CLIENT); } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/power.c b/drivers/net/wireless/intel/iwlwifi/mvm/power.c index 7b1f6ad..ff85865 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/power.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/power.c @@ -308,7 +308,7 @@ static bool iwl_mvm_power_allow_uapsd(struct iwl_mvm *mvm, /* Allow U-APSD only if p2p is stand alone */ bool is_p2p_standalone = true; - if (!iwl_mvm_is_p2p_standalone_uapsd_supported(mvm)) + if (!iwl_mvm_is_p2p_scm_uapsd_supported(mvm)) return false; ieee80211_iterate_active_interfaces_atomic(mvm->hw, -- cgit v0.10.2 From 6d99c88f6003062796cade71997731caec0dd8dc Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Sun, 3 Apr 2016 15:27:55 +0300 Subject: iwlwifi: mvm: remove RX_PHY support for 9000 device In multiple RX queues architecture, the RX_PHY notification is no longer useful as it is received in the default queue even for packets that are received on RSS queue, and cannot be accessed without locking. All the needed data is in the new RX packet metadata and firmware will no longer send this notification for 9000 devices. Remove support of it. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index 4843e02..dd0b8f8 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -1321,7 +1321,6 @@ bool iwl_mvm_bcast_filter_build_cmd(struct iwl_mvm *mvm, void iwl_mvm_rx_rx_phy_cmd(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb); void iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct napi_struct *napi, struct iwl_rx_cmd_buffer *rxb); -void iwl_mvm_rx_phy_cmd_mq(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb); void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi, struct iwl_rx_cmd_buffer *rxb, int queue); void iwl_mvm_rx_frame_release(struct iwl_mvm *mvm, struct napi_struct *napi, diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c index 64d2b4f..55f114d 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c @@ -958,8 +958,6 @@ static void iwl_mvm_rx_mq(struct iwl_op_mode *op_mode, if (likely(pkt->hdr.cmd == REPLY_RX_MPDU_CMD)) iwl_mvm_rx_mpdu_mq(mvm, napi, rxb, 0); - else if (pkt->hdr.cmd == REPLY_RX_PHY_CMD) - iwl_mvm_rx_phy_cmd_mq(mvm, rxb); else if (unlikely(pkt->hdr.group_id == DATA_PATH_GROUP && pkt->hdr.cmd == RX_QUEUES_NOTIFICATION)) iwl_mvm_rx_queue_notif(mvm, rxb, 0); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c index 0f26c70..67924fc 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c @@ -65,19 +65,6 @@ #include "fw-api.h" #include "fw-dbg.h" -void iwl_mvm_rx_phy_cmd_mq(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb) -{ - mvm->ampdu_ref++; - -#ifdef CONFIG_IWLWIFI_DEBUGFS - if (mvm->last_phy_info.phy_flags & cpu_to_le16(RX_RES_PHY_FLAGS_AGG)) { - spin_lock(&mvm->drv_stats_lock); - mvm->drv_rx_stats.ampdu_count++; - spin_unlock(&mvm->drv_stats_lock); - } -#endif -} - static inline int iwl_mvm_check_pn(struct iwl_mvm *mvm, struct sk_buff *skb, int queue, struct ieee80211_sta *sta) { -- cgit v0.10.2 From fbe4112791b8fe12bb3cae0b1e11830519f9dcfa Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Mon, 4 Apr 2016 19:28:45 +0300 Subject: iwlwifi: mvm: update mpdu metadata API rx_phy notification is no longer sent in devices with multiple rx queues. All the needed data is now set in the metadata - update code accordingly to reflect all the features as in the previous RX path. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h index 8193d36..2f024c7 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h @@ -211,6 +211,7 @@ struct iwl_cmd_header_wide { #define FH_RSCSR_FRAME_SIZE_MSK 0x00003FFF /* bits 0-13 */ #define FH_RSCSR_FRAME_INVALID 0x55550000 #define FH_RSCSR_FRAME_ALIGN 0x40 +#define FH_RSCSR_RPA_EN BIT(25) struct iwl_rx_packet { /* @@ -220,7 +221,9 @@ struct iwl_rx_packet { * 31: flag flush RB request * 30: flag ignore TC (terminal counter) request * 29: flag fast IRQ request - * 28-14: Reserved + * 28-26: Reserved + * 25: Offload enabled + * 24-14: Reserved * 13-00: RX frame size */ __le32 len_n_flags; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h index bb8ff40..1994331 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h @@ -336,6 +336,18 @@ enum iwl_rx_mpdu_reorder_data { IWL_RX_MPDU_REORDER_BA_OLD_SN = 0x80000000, }; +enum iwl_rx_mpdu_phy_info { + IWL_RX_MPDU_PHY_AMPDU = BIT(5), + IWL_RX_MPDU_PHY_AMPDU_TOGGLE = BIT(6), + IWL_RX_MPDU_PHY_SHORT_PREAMBLE = BIT(7), + IWL_RX_MPDU_PHY_TSF_OVERLOAD = BIT(8), +}; + +enum iwl_rx_mpdu_mac_info { + IWL_RX_MPDU_PHY_MAC_INDEX_MASK = 0x0f, + IWL_RX_MPDU_PHY_PHY_INDEX_MASK = 0xf0, +}; + struct iwl_rx_mpdu_desc { /* DW2 */ __le16 mpdu_len; @@ -343,9 +355,9 @@ struct iwl_rx_mpdu_desc { u8 mac_flags2; /* DW3 */ u8 amsdu_info; - __le16 reserved_for_software; + __le16 phy_info; u8 mac_phy_idx; - /* DW4 */ + /* DW4 - carries csum data only when rpa_en == 1 */ __le16 raw_csum; /* alledgedly unreliable */ __le16 l3l4_flags; /* DW5 */ @@ -354,17 +366,17 @@ struct iwl_rx_mpdu_desc { u8 sta_id_flags; /* DW6 */ __le32 reorder_data; - /* DW7 */ + /* DW7 - carries rss_hash only when rpa_en == 1 */ __le32 rss_hash; - /* DW8 */ + /* DW8 - carries filter_match only when rpa_en == 1 */ __le32 filter_match; /* DW9 */ __le32 rate_n_flags; /* DW10 */ - u8 energy_a, energy_b, channel, reserved; + u8 energy_a, energy_b, channel, mac_context; /* DW11 */ __le32 gp2_on_air_rise; - /* DW12 & DW13 */ + /* DW12 & DW13 - carries TSF only TSF_OVERLOAD bit == 0 */ __le64 tsf_on_air_rise; } __packed; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index dd0b8f8..65f9a4b 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -731,6 +731,7 @@ struct iwl_mvm { struct iwl_sf_region sf_space; u32 ampdu_ref; + bool ampdu_toggle; struct iwl_notif_wait_data notif_wait; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c index 67924fc..1f4fef4 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c @@ -737,6 +737,7 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi, struct ieee80211_hdr *hdr = (void *)(pkt->data + sizeof(*desc)); u32 len = le16_to_cpu(desc->mpdu_len); u32 rate_n_flags = le32_to_cpu(desc->rate_n_flags); + u16 phy_info = le16_to_cpu(desc->phy_info); struct ieee80211_sta *sta = NULL; struct sk_buff *skb; u8 crypt_len = 0; @@ -767,16 +768,34 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi, le16_to_cpu(desc->status)); rx_status->flag |= RX_FLAG_FAILED_FCS_CRC; } - - rx_status->mactime = le64_to_cpu(desc->tsf_on_air_rise); + /* set the preamble flag if appropriate */ + if (phy_info & IWL_RX_MPDU_PHY_SHORT_PREAMBLE) + rx_status->flag |= RX_FLAG_SHORTPRE; + + if (likely(!(phy_info & IWL_RX_MPDU_PHY_TSF_OVERLOAD))) { + rx_status->mactime = le64_to_cpu(desc->tsf_on_air_rise); + /* TSF as indicated by the firmware is at INA time */ + rx_status->flag |= RX_FLAG_MACTIME_PLCP_START; + } rx_status->device_timestamp = le32_to_cpu(desc->gp2_on_air_rise); rx_status->band = desc->channel > 14 ? NL80211_BAND_5GHZ : NL80211_BAND_2GHZ; rx_status->freq = ieee80211_channel_to_frequency(desc->channel, rx_status->band); iwl_mvm_get_signal_strength(mvm, desc, rx_status); - /* TSF as indicated by the firmware is at INA time */ - rx_status->flag |= RX_FLAG_MACTIME_PLCP_START; + + /* update aggregation data for monitor sake on default queue */ + if (!queue && (phy_info & IWL_RX_MPDU_PHY_AMPDU)) { + bool toggle_bit = phy_info & IWL_RX_MPDU_PHY_AMPDU_TOGGLE; + + rx_status->flag |= RX_FLAG_AMPDU_DETAILS; + rx_status->ampdu_reference = mvm->ampdu_ref; + /* toggle is switched whenever new aggregation starts */ + if (toggle_bit != mvm->ampdu_toggle) { + mvm->ampdu_ref++; + mvm->ampdu_toggle = toggle_bit; + } + } rcu_read_lock(); @@ -831,8 +850,6 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi, iwl_mvm_fw_dbg_collect_trig(mvm, trig, NULL); } - /* TODO: multi queue TCM */ - if (ieee80211_is_data(hdr->frame_control)) iwl_mvm_rx_csum(sta, skb, desc); @@ -857,14 +874,6 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi, iwl_mvm_agg_rx_received(mvm, baid); } - /* - * TODO: PHY info. - * Verify we don't have the information in the MPDU descriptor and - * that it is not needed. - * Make sure for monitor mode that we are on default queue, update - * ampdu_ref and the rest of phy info then - */ - /* Set up the HT phy flags */ switch (rate_n_flags & RATE_MCS_CHAN_WIDTH_MSK) { case RATE_MCS_CHAN_WIDTH_20: @@ -908,8 +917,18 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi, rx_status->band); } - /* TODO: PHY info - update ampdu queue statistics (for debugfs) */ - /* TODO: PHY info - gscan */ + /* management stuff on default queue */ + if (!queue) { + if (unlikely((ieee80211_is_beacon(hdr->frame_control) || + ieee80211_is_probe_resp(hdr->frame_control)) && + mvm->sched_scan_pass_all == + SCHED_SCAN_PASS_ALL_ENABLED)) + mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_FOUND; + + if (unlikely(ieee80211_is_beacon(hdr->frame_control) || + ieee80211_is_probe_resp(hdr->frame_control))) + rx_status->boottime_ns = ktime_get_boot_ns(); + } iwl_mvm_create_skb(skb, hdr, len, crypt_len, rxb); if (!iwl_mvm_reorder(mvm, napi, queue, sta, skb, desc)) -- cgit v0.10.2 From a6f035a008366a03378ed27560794d84a237895b Mon Sep 17 00:00:00 2001 From: Liad Kaufman Date: Mon, 24 Aug 2015 15:23:14 +0300 Subject: iwlwifi: mvm: free dqa queues on STA removal also in non-bss Support queue removal in DQA mode in iwl_mvm_rm_sta() also when the device isn't a STA connected to an AP. Signed-off-by: Liad Kaufman Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index fea4d3437..64b07b1 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -819,8 +819,9 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm, if (iwl_mvm_has_new_rx_api(mvm)) kfree(mvm_sta->dup_data); - if (vif->type == NL80211_IFTYPE_STATION && - mvmvif->ap_sta_id == mvm_sta->sta_id) { + if ((vif->type == NL80211_IFTYPE_STATION && + mvmvif->ap_sta_id == mvm_sta->sta_id) || + iwl_mvm_is_dqa_supported(mvm)){ ret = iwl_mvm_drain_sta(mvm, mvm_sta, true); if (ret) return ret; -- cgit v0.10.2 From 1316d5957be3b311b1494382172d4acb2f36ea59 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Sun, 17 Apr 2016 16:28:18 +0300 Subject: iwlwifi: pcie: workaround HW shadow registers bug Integrated 9000 devices have a bug with shadow registers value retention. If driver writes RBD registers while MAC is asleep the values are stored in shadow registers to be copied whenever MAC wakes up. However, in 9000 devices a MAC wakeup is not triggered and when the bus powers down due to inactivity the shadow values and dirty bits are lost. Turn on the chicken-bits that cause MAC wakeup for RX-related values as well when the device is in D0. When the device is in low power mode turn the RX wakeup chicken bits off since driver is idle and this W/A is not needed. Remove previous W/A which was ineffective. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c index 3ac298f..5c1e71f 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c @@ -178,6 +178,7 @@ const struct iwl_cfg iwl5165_2ac_cfg = { .nvm_ver = IWL9000_NVM_VERSION, .nvm_calib_ver = IWL9000_TX_POWER_VERSION, .max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K, + .integrated = true, }; MODULE_FIRMWARE(IWL9000_MODULE_FIRMWARE(IWL9000_UCODE_API_MAX)); diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h index 4a0af7d..57b14f4 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h @@ -319,6 +319,7 @@ struct iwl_pwr_tx_backoff { * @mq_rx_supported: multi-queue rx support * @vht_mu_mimo_supported: VHT MU-MIMO support * @rf_id: need to read rf_id to determine the firmware image + * @integrated: discrete or integrated * * We enable the driver to be backward compatible wrt. hardware features. * API differences in uCode shouldn't be handled here but through TLVs @@ -362,7 +363,8 @@ struct iwl_cfg { apmg_not_supported:1, mq_rx_supported:1, vht_mu_mimo_supported:1, - rf_id:1; + rf_id:1, + integrated:1; u8 valid_tx_ant; u8 valid_rx_ant; u8 non_shared_ant; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h index b5291344..871ad02 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-csr.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-csr.h @@ -145,8 +145,10 @@ #define CSR_LED_REG (CSR_BASE+0x094) #define CSR_DRAM_INT_TBL_REG (CSR_BASE+0x0A0) -#define CSR_MAC_SHADOW_REG_CTRL (CSR_BASE+0x0A8) /* 6000 and up */ - +#define CSR_MAC_SHADOW_REG_CTRL (CSR_BASE + 0x0A8) /* 6000 and up */ +#define CSR_MAC_SHADOW_REG_CTRL_RX_WAKE BIT(20) +#define CSR_MAC_SHADOW_REG_CTL2 (CSR_BASE + 0x0AC) +#define CSR_MAC_SHADOW_REG_CTL2_RX_WAKE 0xFFFF /* GIO Chicken Bits (PCI Express bus link power management) */ #define CSR_GIO_CHICKEN_BITS (CSR_BASE+0x100) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h index de6974f..482836e 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h @@ -673,4 +673,6 @@ static inline int iwl_trans_pcie_dbgfs_register(struct iwl_trans *trans) int iwl_pci_fw_exit_d0i3(struct iwl_trans *trans); int iwl_pci_fw_enter_d0i3(struct iwl_trans *trans); +void iwl_pcie_enable_rx_wake(struct iwl_trans *trans, bool enable); + #endif /* __iwl_trans_int_pcie_h__ */ diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index 0a4a3c5..f1e309d 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -211,12 +211,8 @@ static void iwl_pcie_rxq_inc_wr_ptr(struct iwl_trans *trans, if (trans->cfg->mq_rx_supported) iwl_write32(trans, RFH_Q_FRBDCB_WIDX_TRG(rxq->id), rxq->write_actual); - /* - * write to FH_RSCSR_CHNL0_WPTR register even in MQ as a W/A to - * hardware shadow registers bug - writing to RFH_Q_FRBDCB_WIDX will - * not wake the NIC. - */ - iwl_write32(trans, FH_RSCSR_CHNL0_WPTR, rxq->write_actual); + else + iwl_write32(trans, FH_RSCSR_CHNL0_WPTR, rxq->write_actual); } static void iwl_pcie_rxq_check_wrptr(struct iwl_trans *trans) @@ -764,6 +760,23 @@ static void iwl_pcie_rx_hw_init(struct iwl_trans *trans, struct iwl_rxq *rxq) iwl_set_bit(trans, CSR_INT_COALESCING, IWL_HOST_INT_OPER_MODE); } +void iwl_pcie_enable_rx_wake(struct iwl_trans *trans, bool enable) +{ + /* + * Turn on the chicken-bits that cause MAC wakeup for RX-related + * values. + * This costs some power, but needed for W/A 9000 integrated A-step + * bug where shadow registers are not in the retention list and their + * value is lost when NIC powers down + */ + if (trans->cfg->integrated) { + iwl_set_bit(trans, CSR_MAC_SHADOW_REG_CTRL, + CSR_MAC_SHADOW_REG_CTRL_RX_WAKE); + iwl_set_bit(trans, CSR_MAC_SHADOW_REG_CTL2, + CSR_MAC_SHADOW_REG_CTL2_RX_WAKE); + } +} + static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); @@ -849,6 +862,8 @@ static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans) /* Set interrupt coalescing timer to default (2048 usecs) */ iwl_write8(trans, CSR_INT_COALESCING, IWL_HOST_INT_TIMEOUT_DEF); + + iwl_pcie_enable_rx_wake(trans, true); } static void iwl_pcie_rx_init_rxb_lists(struct iwl_rxq *rxq) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index f603d78..33fd217 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -1286,6 +1286,8 @@ static void iwl_trans_pcie_d3_suspend(struct iwl_trans *trans, bool test, iwl_clear_bit(trans, CSR_GP_CNTRL, CSR_GP_CNTRL_REG_FLAG_INIT_DONE); + iwl_pcie_enable_rx_wake(trans, false); + if (reset) { /* * reset TX queues -- some of their registers reset during S3 @@ -1311,6 +1313,8 @@ static int iwl_trans_pcie_d3_resume(struct iwl_trans *trans, return 0; } + iwl_pcie_enable_rx_wake(trans, true); + /* * Also enables interrupts - none will happen as the device doesn't * know we're waking it up, only when the opmode actually tells it -- cgit v0.10.2 From b0262f07f47f782b804963f0a7ce3e5e0eba3491 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Thu, 21 Apr 2016 16:38:43 +0300 Subject: iwlwifi: pcie: set RB chunk size per bus For 9000 devices we can have PCIe bus for discrete devices and IOSF bus for integrated devices. PCIe supports maximum transfer size of 128B while IOSF bus supports maximum transfer size of 64B. Configure RB size accordingly. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h index 270f39e..f08cdee 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h @@ -384,7 +384,9 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(unsigned int chnl) #define RFH_GEN_CFG 0xA09800 #define RFH_GEN_CFG_SERVICE_DMA_SNOOP BIT(0) #define RFH_GEN_CFG_RFH_DMA_SNOOP BIT(1) -#define RFH_GEN_CFG_RB_CHUNK_SIZE BIT(4) /* 0 - 64B, 1- 128B */ +#define RFH_GEN_CFG_RB_CHUNK_SIZE_POS 4 +#define RFH_GEN_CFG_RB_CHUNK_SIZE_128 1 +#define RFH_GEN_CFG_RB_CHUNK_SIZE_64 0 #define RFH_GEN_CFG_DEFAULT_RXQ_NUM_MASK 0xF00 #define RFH_GEN_CFG_DEFAULT_RXQ_NUM_POS 8 diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index f1e309d..27ff74d 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -848,13 +848,17 @@ static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans) /* * Activate DMA snooping. - * Set RX DMA chunk size to 64B + * Set RX DMA chunk size to 64B for IOSF and 128B for PCIe * Default queue is 0 */ iwl_write_prph_no_grab(trans, RFH_GEN_CFG, RFH_GEN_CFG_RFH_DMA_SNOOP | (DEFAULT_RXQ_NUM << RFH_GEN_CFG_DEFAULT_RXQ_NUM_POS) | - RFH_GEN_CFG_SERVICE_DMA_SNOOP); + RFH_GEN_CFG_SERVICE_DMA_SNOOP | + (trans->cfg->integrated ? + RFH_GEN_CFG_RB_CHUNK_SIZE_64 : + RFH_GEN_CFG_RB_CHUNK_SIZE_128) << + RFH_GEN_CFG_RB_CHUNK_SIZE_POS); /* Enable the relevant rx queues */ iwl_write_prph_no_grab(trans, RFH_RXF_RXQ_ACTIVE, enabled); -- cgit v0.10.2 From 630443355a31ebf2f80cb5206b68cf2ee945a368 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Thu, 21 Apr 2016 17:41:39 +0300 Subject: iwlwifi: pcie: allow more than one frame in RB for 9000 devices We now have 9000 devices that support multiple frames in a single RB. Enable it. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index 27ff74d..4fad808 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -833,15 +833,13 @@ static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans) /* * Enable Rx DMA - * Single frame mode * Rx buffer size 4 or 8k or 12k * Min RB size 4 or 8 * Drop frames that exceed RB size * 512 RBDs */ iwl_write_prph_no_grab(trans, RFH_RXF_DMA_CFG, - RFH_DMA_EN_ENABLE_VAL | - rb_size | RFH_RXF_DMA_SINGLE_FRAME_MASK | + RFH_DMA_EN_ENABLE_VAL | rb_size | RFH_RXF_DMA_MIN_RB_4_8 | RFH_RXF_DMA_DROP_TOO_LARGE_MASK | RFH_RXF_DMA_RBDCB_SIZE_512); -- cgit v0.10.2 From ab2e696bd25c11bf4baf84f285555b069ae2dd30 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Thu, 21 Apr 2016 20:15:40 +0300 Subject: iwlwifi: pcie: make sure packet arrived to destined queue Add a warning in case packet didn't end up in the HW destined queue. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h index 2f024c7..a45be1d 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h @@ -212,6 +212,8 @@ struct iwl_cmd_header_wide { #define FH_RSCSR_FRAME_INVALID 0x55550000 #define FH_RSCSR_FRAME_ALIGN 0x40 #define FH_RSCSR_RPA_EN BIT(25) +#define FH_RSCSR_RXQ_POS 16 +#define FH_RSCSR_RXQ_MASK 0x3F0000 struct iwl_rx_packet { /* @@ -223,7 +225,11 @@ struct iwl_rx_packet { * 29: flag fast IRQ request * 28-26: Reserved * 25: Offload enabled - * 24-14: Reserved + * 24: RPF enabled + * 23: RSS enabled + * 22: Checksum enabled + * 21-16: RX queue + * 15-14: Reserved * 13-00: RX frame size */ __le32 len_n_flags; diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index 4fad808..70e39e4 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -1104,6 +1104,9 @@ static void iwl_pcie_rx_handle_rb(struct iwl_trans *trans, if (pkt->len_n_flags == cpu_to_le32(FH_RSCSR_FRAME_INVALID)) break; + WARN_ON((le32_to_cpu(pkt->len_n_flags) & FH_RSCSR_RXQ_MASK) >> + FH_RSCSR_RXQ_POS != rxq->id); + IWL_DEBUG_RX(trans, "cmd at offset %d: %s (0x%.2x, seq 0x%x)\n", rxcb._offset, -- cgit v0.10.2 From d3a108a48dc670d539c58d4339d211b914a1e1b5 Mon Sep 17 00:00:00 2001 From: Andrei Otcheretianski Date: Sun, 28 Feb 2016 17:12:21 +0200 Subject: iwlwifi: mvm: Support CSA countdown offloading Add support CSA countdown offloading. When CSA starts, the driver specifies the offsets to the eCSA and CSA IEs in the beacon template command and the fw performs the countdown. The fw notifies the driver when the channel switch flow should be performed. Beacon sent notifications are not used anymore. Signed-off-by: Andrei Otcheretianski Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h index 3f5029e..eb18de8 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h @@ -311,6 +311,9 @@ typedef unsigned int __bitwise__ iwl_ucode_tlv_capa_t; * @IWL_UCODE_TLV_CAPA_EXTENDED_DTS_MEASURE: extended DTS measurement * @IWL_UCODE_TLV_CAPA_SHORT_PM_TIMEOUTS: supports short PM timeouts * @IWL_UCODE_TLV_CAPA_BT_MPLUT_SUPPORT: supports bt-coex Multi-priority LUT + * @IWL_UCODE_TLV_CAPA_CSA_AND_TBTT_OFFLOAD: the firmware supports CSA + * countdown offloading. Beacon notifications are not sent to the host. + * The fw also offloads TBTT alignment. * @IWL_UCODE_TLV_CAPA_BEACON_ANT_SELECTION: firmware will decide on what * antenna the beacon should be transmitted * @IWL_UCODE_TLV_CAPA_BEACON_STORING: firmware will store the latest beacon @@ -355,6 +358,7 @@ enum iwl_ucode_tlv_capa { IWL_UCODE_TLV_CAPA_SHORT_PM_TIMEOUTS = (__force iwl_ucode_tlv_capa_t)65, IWL_UCODE_TLV_CAPA_BT_MPLUT_SUPPORT = (__force iwl_ucode_tlv_capa_t)67, IWL_UCODE_TLV_CAPA_MULTI_QUEUE_RX_SUPPORT = (__force iwl_ucode_tlv_capa_t)68, + IWL_UCODE_TLV_CAPA_CSA_AND_TBTT_OFFLOAD = (__force iwl_ucode_tlv_capa_t)70, IWL_UCODE_TLV_CAPA_BEACON_ANT_SELECTION = (__force iwl_ucode_tlv_capa_t)71, IWL_UCODE_TLV_CAPA_BEACON_STORING = (__force iwl_ucode_tlv_capa_t)72, IWL_UCODE_TLV_CAPA_LAR_SUPPORT_V2 = (__force iwl_ucode_tlv_capa_t)73, diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-tx.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-tx.h index dadcccd..ee59511 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-tx.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-tx.h @@ -562,8 +562,8 @@ struct iwl_mvm_ba_notif { u8 reserved1; } __packed; -/* - * struct iwl_mac_beacon_cmd - beacon template command +/** + * struct iwl_mac_beacon_cmd_v6 - beacon template command * @tx: the tx commands associated with the beacon frame * @template_id: currently equal to the mac context id of the coresponding * mac. @@ -571,13 +571,34 @@ struct iwl_mvm_ba_notif { * @tim_size: the length of the tim IE * @frame: the template of the beacon frame */ +struct iwl_mac_beacon_cmd_v6 { + struct iwl_tx_cmd tx; + __le32 template_id; + __le32 tim_idx; + __le32 tim_size; + struct ieee80211_hdr frame[0]; +} __packed; /* BEACON_TEMPLATE_CMD_API_S_VER_6 */ + +/** + * struct iwl_mac_beacon_cmd - beacon template command with offloaded CSA + * @tx: the tx commands associated with the beacon frame + * @template_id: currently equal to the mac context id of the coresponding + * mac. + * @tim_idx: the offset of the tim IE in the beacon + * @tim_size: the length of the tim IE + * @ecsa_offset: offset to the ECSA IE if present + * @csa_offset: offset to the CSA IE if present + * @frame: the template of the beacon frame + */ struct iwl_mac_beacon_cmd { struct iwl_tx_cmd tx; __le32 template_id; __le32 tim_idx; __le32 tim_size; + __le32 ecsa_offset; + __le32 csa_offset; struct ieee80211_hdr frame[0]; -} __packed; +} __packed; /* BEACON_TEMPLATE_CMD_API_S_VER_7 */ struct iwl_beacon_notif { struct iwl_mvm_tx_resp beacon_notify_hdr; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h index 41b80ae..b06380d 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h @@ -314,6 +314,7 @@ enum { enum iwl_mac_conf_subcmd_ids { LINK_QUALITY_MEASUREMENT_CMD = 0x1, LINK_QUALITY_MEASUREMENT_COMPLETE_NOTIF = 0xFE, + CHANNEL_SWITCH_NOA_NOTIF = 0xFF, }; enum iwl_phy_ops_subcmd_ids { @@ -732,7 +733,7 @@ enum iwl_time_event_type { /* P2P GO Events */ TE_P2P_GO_ASSOC_PROT, - TE_P2P_GO_REPETITIVE_NOA, + TE_P2P_GO_REPETITIVET_NOA, TE_P2P_GO_CT_WINDOW, /* WiDi Sync Events */ @@ -2111,4 +2112,13 @@ struct iwl_link_qual_msrmnt_notif { __le32 reserved[3]; } __packed; /* LQM_MEASUREMENT_COMPLETE_NTF_API_S_VER1 */ +/** + * Channel switch NOA notification + * + * @id_and_color: ID and color of the MAC + */ +struct iwl_channel_switch_noa_notif { + __le32 id_and_color; +} __packed; /* CHANNEL_SWITCH_START_NTFY_API_S_VER_1 */ + #endif /* __fw_api_h__ */ diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c index 7aae068..69c42ce 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac-ctxt.c @@ -1006,7 +1006,7 @@ static int iwl_mvm_mac_ctxt_cmd_p2p_device(struct iwl_mvm *mvm, } static void iwl_mvm_mac_ctxt_set_tim(struct iwl_mvm *mvm, - struct iwl_mac_beacon_cmd *beacon_cmd, + struct iwl_mac_beacon_cmd_v6 *beacon_cmd, u8 *beacon, u32 frame_size) { u32 tim_idx; @@ -1030,6 +1030,23 @@ static void iwl_mvm_mac_ctxt_set_tim(struct iwl_mvm *mvm, } } +static u32 iwl_mvm_find_ie_offset(u8 *beacon, u8 eid, u32 frame_size) +{ + struct ieee80211_mgmt *mgmt = (void *)beacon; + const u8 *ie; + + if (WARN_ON_ONCE(frame_size <= (mgmt->u.beacon.variable - beacon))) + return 0; + + frame_size -= mgmt->u.beacon.variable - beacon; + + ie = cfg80211_find_ie(eid, mgmt->u.beacon.variable, frame_size); + if (!ie) + return 0; + + return ie - beacon; +} + static int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm, struct ieee80211_vif *vif, struct sk_buff *beacon) @@ -1039,7 +1056,10 @@ static int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm, .id = BEACON_TEMPLATE_CMD, .flags = CMD_ASYNC, }; - struct iwl_mac_beacon_cmd beacon_cmd = {}; + union { + struct iwl_mac_beacon_cmd_v6 beacon_cmd_v6; + struct iwl_mac_beacon_cmd beacon_cmd; + } u = {}; struct ieee80211_tx_info *info; u32 beacon_skb_len; u32 rate, tx_flags; @@ -1051,18 +1071,18 @@ static int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm, /* TODO: for now the beacon template id is set to be the mac context id. * Might be better to handle it as another resource ... */ - beacon_cmd.template_id = cpu_to_le32((u32)mvmvif->id); + u.beacon_cmd_v6.template_id = cpu_to_le32((u32)mvmvif->id); info = IEEE80211_SKB_CB(beacon); /* Set up TX command fields */ - beacon_cmd.tx.len = cpu_to_le16((u16)beacon_skb_len); - beacon_cmd.tx.sta_id = mvmvif->bcast_sta.sta_id; - beacon_cmd.tx.life_time = cpu_to_le32(TX_CMD_LIFE_TIME_INFINITE); + u.beacon_cmd_v6.tx.len = cpu_to_le16((u16)beacon_skb_len); + u.beacon_cmd_v6.tx.sta_id = mvmvif->bcast_sta.sta_id; + u.beacon_cmd_v6.tx.life_time = cpu_to_le32(TX_CMD_LIFE_TIME_INFINITE); tx_flags = TX_CMD_FLG_SEQ_CTL | TX_CMD_FLG_TSF; tx_flags |= iwl_mvm_bt_coex_tx_prio(mvm, (void *)beacon->data, info, 0) << TX_CMD_FLG_BT_PRIO_POS; - beacon_cmd.tx.tx_flags = cpu_to_le32(tx_flags); + u.beacon_cmd_v6.tx.tx_flags = cpu_to_le32(tx_flags); if (!fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_BEACON_ANT_SELECTION)) { @@ -1071,7 +1091,7 @@ static int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm, mvm->mgmt_last_antenna_idx); } - beacon_cmd.tx.rate_n_flags = + u.beacon_cmd_v6.tx.rate_n_flags = cpu_to_le32(BIT(mvm->mgmt_last_antenna_idx) << RATE_MCS_ANT_POS); @@ -1079,20 +1099,37 @@ static int iwl_mvm_mac_ctxt_send_beacon(struct iwl_mvm *mvm, rate = IWL_FIRST_OFDM_RATE; } else { rate = IWL_FIRST_CCK_RATE; - beacon_cmd.tx.rate_n_flags |= cpu_to_le32(RATE_MCS_CCK_MSK); + u.beacon_cmd_v6.tx.rate_n_flags |= + cpu_to_le32(RATE_MCS_CCK_MSK); } - beacon_cmd.tx.rate_n_flags |= + u.beacon_cmd_v6.tx.rate_n_flags |= cpu_to_le32(iwl_mvm_mac80211_idx_to_hwrate(rate)); /* Set up TX beacon command fields */ if (vif->type == NL80211_IFTYPE_AP) - iwl_mvm_mac_ctxt_set_tim(mvm, &beacon_cmd, + iwl_mvm_mac_ctxt_set_tim(mvm, &u.beacon_cmd_v6, beacon->data, beacon_skb_len); /* Submit command */ - cmd.len[0] = sizeof(beacon_cmd); - cmd.data[0] = &beacon_cmd; + + if (fw_has_capa(&mvm->fw->ucode_capa, + IWL_UCODE_TLV_CAPA_CSA_AND_TBTT_OFFLOAD)) { + u.beacon_cmd.csa_offset = + cpu_to_le32(iwl_mvm_find_ie_offset(beacon->data, + WLAN_EID_CHANNEL_SWITCH, + beacon_skb_len)); + u.beacon_cmd.ecsa_offset = + cpu_to_le32(iwl_mvm_find_ie_offset(beacon->data, + WLAN_EID_EXT_CHANSWITCH_ANN, + beacon_skb_len)); + + cmd.len[0] = sizeof(u.beacon_cmd); + } else { + cmd.len[0] = sizeof(u.beacon_cmd_v6); + } + + cmd.data[0] = &u; cmd.dataflags[0] = 0; cmd.len[1] = beacon_skb_len; cmd.data[1] = beacon->data; @@ -1538,3 +1575,48 @@ void iwl_mvm_rx_stored_beacon_notif(struct iwl_mvm *mvm, /* pass it as regular rx to mac80211 */ ieee80211_rx_napi(mvm->hw, NULL, skb, NULL); } + +void iwl_mvm_channel_switch_noa_notif(struct iwl_mvm *mvm, + struct iwl_rx_cmd_buffer *rxb) +{ + struct iwl_rx_packet *pkt = rxb_addr(rxb); + struct iwl_channel_switch_noa_notif *notif = (void *)pkt->data; + struct ieee80211_vif *csa_vif; + struct iwl_mvm_vif *mvmvif; + int len = iwl_rx_packet_payload_len(pkt); + u32 id_n_color; + + if (WARN_ON_ONCE(len < sizeof(*notif))) + return; + + rcu_read_lock(); + + csa_vif = rcu_dereference(mvm->csa_vif); + if (WARN_ON(!csa_vif || !csa_vif->csa_active)) + goto out_unlock; + + id_n_color = le32_to_cpu(notif->id_and_color); + + mvmvif = iwl_mvm_vif_from_mac80211(csa_vif); + if (WARN(FW_CMD_ID_AND_COLOR(mvmvif->id, mvmvif->color) != id_n_color, + "channel switch noa notification on unexpected vif (csa_vif=%d, notif=%d)", + FW_CMD_ID_AND_COLOR(mvmvif->id, mvmvif->color), id_n_color)) + goto out_unlock; + + IWL_DEBUG_INFO(mvm, "Channel Switch Started Notification\n"); + + queue_delayed_work(system_wq, &mvm->cs_tx_unblock_dwork, + msecs_to_jiffies(IWL_MVM_CS_UNBLOCK_TX_TIMEOUT * + csa_vif->bss_conf.beacon_int)); + + ieee80211_csa_finish(csa_vif); + + rcu_read_unlock(); + + RCU_INIT_POINTER(mvm->csa_vif, NULL); + + return; + +out_unlock: + rcu_read_unlock(); +} diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index b7014bc..e5cb7db 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -1199,6 +1199,7 @@ static void iwl_mvm_mac_stop(struct ieee80211_hw *hw) flush_work(&mvm->async_handlers_wk); flush_work(&mvm->add_stream_wk); cancel_delayed_work_sync(&mvm->fw_dump_wk); + cancel_delayed_work_sync(&mvm->cs_tx_unblock_dwork); iwl_mvm_free_fw_dump_desc(mvm); mutex_lock(&mvm->mutex); @@ -3687,6 +3688,13 @@ static int iwl_mvm_pre_channel_switch(struct ieee80211_hw *hw, goto out_unlock; } + /* we still didn't unblock tx. prevent new CS meanwhile */ + if (rcu_dereference_protected(mvm->csa_tx_blocked_vif, + lockdep_is_held(&mvm->mutex))) { + ret = -EBUSY; + goto out_unlock; + } + rcu_assign_pointer(mvm->csa_vif, vif); if (WARN_ONCE(mvmvif->csa_countdown, @@ -3695,6 +3703,8 @@ static int iwl_mvm_pre_channel_switch(struct ieee80211_hw *hw, goto out_unlock; } + mvmvif->csa_target_freq = chsw->chandef.chan->center_freq; + break; case NL80211_IFTYPE_STATION: if (mvmvif->lqm_active) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index 65f9a4b..4b75b92 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -452,6 +452,7 @@ struct iwl_mvm_vif { /* Indicates that CSA countdown may be started */ bool csa_countdown; bool csa_failed; + u16 csa_target_freq; /* TCP Checksum Offload */ netdev_features_t features; @@ -1007,6 +1008,8 @@ struct iwl_mvm { * clients. */ bool drop_bcn_ap_mode; + + struct delayed_work cs_tx_unblock_dwork; }; /* Extract MVM priv from op_mode and _hw */ @@ -1381,6 +1384,8 @@ void iwl_mvm_mac_ctxt_recalc_tsf_id(struct iwl_mvm *mvm, struct ieee80211_vif *vif); unsigned long iwl_mvm_get_used_hw_queues(struct iwl_mvm *mvm, struct ieee80211_vif *exclude_vif); +void iwl_mvm_channel_switch_noa_notif(struct iwl_mvm *mvm, + struct iwl_rx_cmd_buffer *rxb); /* Bindings */ int iwl_mvm_binding_add_vif(struct iwl_mvm *mvm, struct ieee80211_vif *vif); int iwl_mvm_binding_remove_vif(struct iwl_mvm *mvm, struct ieee80211_vif *vif); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c index 55f114d..ddc4004 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c @@ -431,6 +431,7 @@ static const struct iwl_hcmd_names iwl_mvm_system_names[] = { static const struct iwl_hcmd_names iwl_mvm_mac_conf_names[] = { HCMD_NAME(LINK_QUALITY_MEASUREMENT_CMD), HCMD_NAME(LINK_QUALITY_MEASUREMENT_COMPLETE_NOTIF), + HCMD_NAME(CHANNEL_SWITCH_NOA_NOTIF), }; /* Please keep this array *SORTED* by hex value. @@ -494,6 +495,29 @@ static u32 calc_min_backoff(struct iwl_trans *trans, const struct iwl_cfg *cfg) static void iwl_mvm_fw_error_dump_wk(struct work_struct *work); +static void iwl_mvm_tx_unblock_dwork(struct work_struct *work) +{ + struct iwl_mvm *mvm = + container_of(work, struct iwl_mvm, cs_tx_unblock_dwork.work); + struct ieee80211_vif *tx_blocked_vif; + struct iwl_mvm_vif *mvmvif; + + mutex_lock(&mvm->mutex); + + tx_blocked_vif = + rcu_dereference_protected(mvm->csa_tx_blocked_vif, + lockdep_is_held(&mvm->mutex)); + + if (!tx_blocked_vif) + goto unlock; + + mvmvif = iwl_mvm_vif_from_mac80211(tx_blocked_vif); + iwl_mvm_modify_all_sta_disable_tx(mvm, mvmvif, false); + RCU_INIT_POINTER(mvm->csa_tx_blocked_vif, NULL); +unlock: + mutex_unlock(&mvm->mutex); +} + static struct iwl_op_mode * iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg, const struct iwl_fw *fw, struct dentry *dbgfs_dir) @@ -595,6 +619,8 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg, SET_IEEE80211_DEV(mvm->hw, mvm->trans->dev); + INIT_DELAYED_WORK(&mvm->cs_tx_unblock_dwork, iwl_mvm_tx_unblock_dwork); + /* * Populate the state variables that the transport layer needs * to know about. diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c index ab7f7ed..6d096b6 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c @@ -354,13 +354,22 @@ void iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct napi_struct *napi, if (sta) { struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta); + struct ieee80211_vif *tx_blocked_vif = + rcu_dereference(mvm->csa_tx_blocked_vif); /* We have tx blocked stations (with CS bit). If we heard * frames from a blocked station on a new channel we can * TX to it again. */ - if (unlikely(mvm->csa_tx_block_bcn_timeout)) - iwl_mvm_sta_modify_disable_tx_ap(mvm, sta, false); + if (unlikely(tx_blocked_vif) && + mvmsta->vif == tx_blocked_vif) { + struct iwl_mvm_vif *mvmvif = + iwl_mvm_vif_from_mac80211(tx_blocked_vif); + + if (mvmvif->csa_target_freq == rx_status->freq) + iwl_mvm_sta_modify_disable_tx_ap(mvm, sta, + false); + } rs_update_last_rssi(mvm, &mvmsta->lq_sta, rx_status); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c index 1f4fef4..d13397a 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c @@ -817,6 +817,8 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi, if (sta) { struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta); + struct ieee80211_vif *tx_blocked_vif = + rcu_dereference(mvm->csa_tx_blocked_vif); u8 baid = (u8)((le32_to_cpu(desc->reorder_data) & IWL_RX_MPDU_REORDER_BAID_MASK) >> IWL_RX_MPDU_REORDER_BAID_SHIFT); @@ -826,8 +828,15 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi, * frames from a blocked station on a new channel we can * TX to it again. */ - if (unlikely(mvm->csa_tx_block_bcn_timeout)) - iwl_mvm_sta_modify_disable_tx_ap(mvm, sta, false); + if (unlikely(tx_blocked_vif) && + tx_blocked_vif == mvmsta->vif) { + struct iwl_mvm_vif *mvmvif = + iwl_mvm_vif_from_mac80211(tx_blocked_vif); + + if (mvmvif->csa_target_freq == rx_status->freq) + iwl_mvm_sta_modify_disable_tx_ap(mvm, sta, + false); + } rs_update_last_rssi(mvm, &mvmsta->lq_sta, rx_status); -- cgit v0.10.2 From 58035432d60616cc2ef6514a3d0e6d6ad01bf705 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 27 Apr 2016 13:33:26 +0200 Subject: iwlwifi: mvm: handle FRAME_RELEASE in MQ code For some reason, the FRAME_RELEASE message handling for the default queue ended up being in the only/default queue for non-RSS devices; fix that and handle FRAME_RELEASE properly on the default queue for RSS devices. Fixes: 585a6fccf5b8 ("iwlwifi: mvm: infrastructure for frame-release message") Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c index ddc4004..632b1dc 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c @@ -967,8 +967,6 @@ static void iwl_mvm_rx(struct iwl_op_mode *op_mode, if (likely(pkt->hdr.cmd == REPLY_RX_MPDU_CMD)) iwl_mvm_rx_rx_mpdu(mvm, napi, rxb); - else if (pkt->hdr.cmd == FRAME_RELEASE) - iwl_mvm_rx_frame_release(mvm, napi, rxb, 0); else if (pkt->hdr.cmd == REPLY_RX_PHY_CMD) iwl_mvm_rx_rx_phy_cmd(mvm, rxb); else @@ -987,6 +985,8 @@ static void iwl_mvm_rx_mq(struct iwl_op_mode *op_mode, else if (unlikely(pkt->hdr.group_id == DATA_PATH_GROUP && pkt->hdr.cmd == RX_QUEUES_NOTIFICATION)) iwl_mvm_rx_queue_notif(mvm, rxb, 0); + else if (pkt->hdr.cmd == FRAME_RELEASE) + iwl_mvm_rx_frame_release(mvm, napi, rxb, 0); else iwl_mvm_rx_common(mvm, rxb, pkt); } -- cgit v0.10.2 From 1aacde3d22c42281236155c1ef6d7a5aa32a826b Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 30 Jun 2016 17:24:43 +0200 Subject: bpf: generally move prog destruction to RCU deferral Jann Horn reported following analysis that could potentially result in a very hard to trigger (if not impossible) UAF race, to quote his event timeline: - Set up a process with threads T1, T2 and T3 - Let T1 set up a socket filter F1 that invokes another filter F2 through a BPF map [tail call] - Let T1 trigger the socket filter via a unix domain socket write, don't wait for completion - Let T2 call PERF_EVENT_IOC_SET_BPF with F2, don't wait for completion - Now T2 should be behind bpf_prog_get(), but before bpf_prog_put() - Let T3 close the file descriptor for F2, dropping the reference count of F2 to 2 - At this point, T1 should have looked up F2 from the map, but not finished executing it - Let T3 remove F2 from the BPF map, dropping the reference count of F2 to 1 - Now T2 should call bpf_prog_put() (wrong BPF program type), dropping the reference count of F2 to 0 and scheduling bpf_prog_free_deferred() via schedule_work() - At this point, the BPF program could be freed - BPF execution is still running in a freed BPF program While at PERF_EVENT_IOC_SET_BPF time it's only guaranteed that the perf event fd we're doing the syscall on doesn't disappear from underneath us for whole syscall time, it may not be the case for the bpf fd used as an argument only after we did the put. It needs to be a valid fd pointing to a BPF program at the time of the call to make the bpf_prog_get() and while T2 gets preempted, F2 must have dropped reference to 1 on the other CPU. The fput() from the close() in T3 should also add additionally delay to the reference drop via exit_task_work() when bpf_prog_release() gets called as well as scheduling bpf_prog_free_deferred(). That said, it makes nevertheless sense to move the BPF prog destruction generally after RCU grace period to guarantee that such scenario above, but also others as recently fixed in ceb56070359b ("bpf, perf: delay release of BPF prog after grace period") with regards to tail calls won't happen. Integrating bpf_prog_free_deferred() directly into the RCU callback is not allowed since the invocation might happen from either softirq or process context, so we're not permitted to block. Reviewing all bpf_prog_put() invocations from eBPF side (note, cBPF -> eBPF progs don't use this for their destruction) with call_rcu() look good to me. Since we don't know whether at the time of attaching the program, we're already part of a tail call map, we need to use RCU variant. However, due to this, there won't be severely more stress on the RCU callback queue: situations with above bpf_prog_get() and bpf_prog_put() combo in practice normally won't lead to releases, but even if they would, enough effort/ cycles have to be put into loading a BPF program into the kernel already. Reported-by: Jann Horn Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 8411032..7495498 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -220,7 +220,6 @@ void bpf_register_map_type(struct bpf_map_type_list *tl); struct bpf_prog *bpf_prog_get(u32 ufd); struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog); void bpf_prog_put(struct bpf_prog *prog); -void bpf_prog_put_rcu(struct bpf_prog *prog); struct bpf_map *bpf_map_get_with_uref(u32 ufd); struct bpf_map *__bpf_map_get(struct fd f); @@ -281,10 +280,6 @@ static inline struct bpf_prog *bpf_prog_get(u32 ufd) static inline void bpf_prog_put(struct bpf_prog *prog) { } - -static inline void bpf_prog_put_rcu(struct bpf_prog *prog) -{ -} #endif /* CONFIG_BPF_SYSCALL */ /* verifier prototypes for helper functions called from eBPF programs */ diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 5af3073..4ec57a6 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -390,9 +390,7 @@ static void *prog_fd_array_get_ptr(struct bpf_map *map, static void prog_fd_array_put_ptr(void *ptr) { - struct bpf_prog *prog = ptr; - - bpf_prog_put_rcu(prog); + bpf_prog_put(ptr); } /* decrement refcnt of all bpf_progs that are stored in this map */ diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index c23a4e93..f6806a1 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -623,7 +623,7 @@ static void bpf_prog_uncharge_memlock(struct bpf_prog *prog) free_uid(user); } -static void __prog_put_common(struct rcu_head *rcu) +static void __bpf_prog_put_rcu(struct rcu_head *rcu) { struct bpf_prog_aux *aux = container_of(rcu, struct bpf_prog_aux, rcu); @@ -632,17 +632,10 @@ static void __prog_put_common(struct rcu_head *rcu) bpf_prog_free(aux->prog); } -/* version of bpf_prog_put() that is called after a grace period */ -void bpf_prog_put_rcu(struct bpf_prog *prog) -{ - if (atomic_dec_and_test(&prog->aux->refcnt)) - call_rcu(&prog->aux->rcu, __prog_put_common); -} - void bpf_prog_put(struct bpf_prog *prog) { if (atomic_dec_and_test(&prog->aux->refcnt)) - __prog_put_common(&prog->aux->rcu); + call_rcu(&prog->aux->rcu, __bpf_prog_put_rcu); } EXPORT_SYMBOL_GPL(bpf_prog_put); @@ -650,7 +643,7 @@ static int bpf_prog_release(struct inode *inode, struct file *filp) { struct bpf_prog *prog = filp->private_data; - bpf_prog_put_rcu(prog); + bpf_prog_put(prog); return 0; } diff --git a/kernel/events/core.c b/kernel/events/core.c index 85cd418..9c51ec3 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -7529,7 +7529,7 @@ static void perf_event_free_bpf_prog(struct perf_event *event) prog = event->tp_event->prog; if (prog) { event->tp_event->prog = NULL; - bpf_prog_put_rcu(prog); + bpf_prog_put(prog); } } -- cgit v0.10.2 From 113214be7f6c98dd6d0435e4765aea8dea91662c Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 30 Jun 2016 17:24:44 +0200 Subject: bpf: refactor bpf_prog_get and type check into helper Since bpf_prog_get() and program type check is used in a couple of places, refactor this into a small helper function that we can make use of. Since the non RO prog->aux part is not used in performance critical paths and a program destruction via RCU is rather very unlikley when doing the put, we shouldn't have an issue just doing the bpf_prog_get() + prog->type != type check, but actually not taking the ref at all (due to being in fdget() / fdput() section of the bpf fd) is even cleaner and makes the diff smaller as well, so just go for that. Callsites are changed to make use of the new helper where possible. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 7495498..b3336b4 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -218,6 +218,7 @@ void bpf_register_prog_type(struct bpf_prog_type_list *tl); void bpf_register_map_type(struct bpf_map_type_list *tl); struct bpf_prog *bpf_prog_get(u32 ufd); +struct bpf_prog *bpf_prog_get_type(u32 ufd, enum bpf_prog_type type); struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog); void bpf_prog_put(struct bpf_prog *prog); @@ -277,6 +278,12 @@ static inline struct bpf_prog *bpf_prog_get(u32 ufd) return ERR_PTR(-EOPNOTSUPP); } +static inline struct bpf_prog *bpf_prog_get_type(u32 ufd, + enum bpf_prog_type type) +{ + return ERR_PTR(-EOPNOTSUPP); +} + static inline void bpf_prog_put(struct bpf_prog *prog) { } diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index f6806a1..22863d9 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -657,7 +657,7 @@ int bpf_prog_new_fd(struct bpf_prog *prog) O_RDWR | O_CLOEXEC); } -static struct bpf_prog *__bpf_prog_get(struct fd f) +static struct bpf_prog *____bpf_prog_get(struct fd f) { if (!f.file) return ERR_PTR(-EBADF); @@ -678,24 +678,35 @@ struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog) return prog; } -/* called by sockets/tracing/seccomp before attaching program to an event - * pairs with bpf_prog_put() - */ -struct bpf_prog *bpf_prog_get(u32 ufd) +static struct bpf_prog *__bpf_prog_get(u32 ufd, enum bpf_prog_type *type) { struct fd f = fdget(ufd); struct bpf_prog *prog; - prog = __bpf_prog_get(f); + prog = ____bpf_prog_get(f); if (IS_ERR(prog)) return prog; + if (type && prog->type != *type) { + prog = ERR_PTR(-EINVAL); + goto out; + } prog = bpf_prog_inc(prog); +out: fdput(f); - return prog; } -EXPORT_SYMBOL_GPL(bpf_prog_get); + +struct bpf_prog *bpf_prog_get(u32 ufd) +{ + return __bpf_prog_get(ufd, NULL); +} + +struct bpf_prog *bpf_prog_get_type(u32 ufd, enum bpf_prog_type type) +{ + return __bpf_prog_get(ufd, &type); +} +EXPORT_SYMBOL_GPL(bpf_prog_get_type); /* last field in 'union bpf_attr' used by this command */ #define BPF_PROG_LOAD_LAST_FIELD kern_version diff --git a/net/core/filter.c b/net/core/filter.c index 76f9a49..76fee35 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1301,21 +1301,10 @@ int sk_reuseport_attach_filter(struct sock_fprog *fprog, struct sock *sk) static struct bpf_prog *__get_bpf(u32 ufd, struct sock *sk) { - struct bpf_prog *prog; - if (sock_flag(sk, SOCK_FILTER_LOCKED)) return ERR_PTR(-EPERM); - prog = bpf_prog_get(ufd); - if (IS_ERR(prog)) - return prog; - - if (prog->type != BPF_PROG_TYPE_SOCKET_FILTER) { - bpf_prog_put(prog); - return ERR_PTR(-EINVAL); - } - - return prog; + return bpf_prog_get_type(ufd, BPF_PROG_TYPE_SOCKET_FILTER); } int sk_attach_bpf(u32 ufd, struct sock *sk) diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c index 0b68ba7..cb39e05 100644 --- a/net/kcm/kcmsock.c +++ b/net/kcm/kcmsock.c @@ -1765,18 +1765,12 @@ static int kcm_attach_ioctl(struct socket *sock, struct kcm_attach *info) if (!csock) return -ENOENT; - prog = bpf_prog_get(info->bpf_fd); + prog = bpf_prog_get_type(info->bpf_fd, BPF_PROG_TYPE_SOCKET_FILTER); if (IS_ERR(prog)) { err = PTR_ERR(prog); goto out; } - if (prog->type != BPF_PROG_TYPE_SOCKET_FILTER) { - bpf_prog_put(prog); - err = -EINVAL; - goto out; - } - err = kcm_attach(sock, csock, prog); if (err) { bpf_prog_put(prog); diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index d1f3b9e..48b5895 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1588,13 +1588,9 @@ static int fanout_set_data_ebpf(struct packet_sock *po, char __user *data, if (copy_from_user(&fd, data, len)) return -EFAULT; - new = bpf_prog_get(fd); + new = bpf_prog_get_type(fd, BPF_PROG_TYPE_SOCKET_FILTER); if (IS_ERR(new)) return PTR_ERR(new); - if (new->type != BPF_PROG_TYPE_SOCKET_FILTER) { - bpf_prog_put(new); - return -EINVAL; - } __fanout_set_data_bpf(po->fanout, new); return 0; diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index f7b6cf4..ef74bff 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -223,15 +223,10 @@ static int tcf_bpf_init_from_efd(struct nlattr **tb, struct tcf_bpf_cfg *cfg) bpf_fd = nla_get_u32(tb[TCA_ACT_BPF_FD]); - fp = bpf_prog_get(bpf_fd); + fp = bpf_prog_get_type(bpf_fd, BPF_PROG_TYPE_SCHED_ACT); if (IS_ERR(fp)) return PTR_ERR(fp); - if (fp->type != BPF_PROG_TYPE_SCHED_ACT) { - bpf_prog_put(fp); - return -EINVAL; - } - if (tb[TCA_ACT_BPF_NAME]) { name = kmemdup(nla_data(tb[TCA_ACT_BPF_NAME]), nla_len(tb[TCA_ACT_BPF_NAME]), diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index 7b342c7..c3002c2 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -272,15 +272,10 @@ static int cls_bpf_prog_from_efd(struct nlattr **tb, struct cls_bpf_prog *prog, bpf_fd = nla_get_u32(tb[TCA_BPF_FD]); - fp = bpf_prog_get(bpf_fd); + fp = bpf_prog_get_type(bpf_fd, BPF_PROG_TYPE_SCHED_CLS); if (IS_ERR(fp)) return PTR_ERR(fp); - if (fp->type != BPF_PROG_TYPE_SCHED_CLS) { - bpf_prog_put(fp); - return -EINVAL; - } - if (tb[TCA_BPF_NAME]) { name = kmemdup(nla_data(tb[TCA_BPF_NAME]), nla_len(tb[TCA_BPF_NAME]), -- cgit v0.10.2 From 1f3fe7ebf6136c341012db9f554d4caa566fcbaa Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 30 Jun 2016 10:28:42 -0700 Subject: cgroup: Add cgroup_get_from_fd Add a helper function to get a cgroup2 from a fd. It will be stored in a bpf array (BPF_MAP_TYPE_CGROUP_ARRAY) which will be introduced in the later patch. Signed-off-by: Martin KaFai Lau Cc: Alexei Starovoitov Cc: Daniel Borkmann Cc: Tejun Heo Acked-by: Tejun Heo Signed-off-by: David S. Miller diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index a20320c..984f73b 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -87,6 +87,7 @@ struct cgroup_subsys_state *css_tryget_online_from_dir(struct dentry *dentry, struct cgroup_subsys *ss); struct cgroup *cgroup_get_from_path(const char *path); +struct cgroup *cgroup_get_from_fd(int fd); int cgroup_attach_task_all(struct task_struct *from, struct task_struct *); int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from); diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 75c0ff0..50787cd 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -62,6 +62,7 @@ #include #include #include +#include #include /* @@ -6209,6 +6210,40 @@ struct cgroup *cgroup_get_from_path(const char *path) } EXPORT_SYMBOL_GPL(cgroup_get_from_path); +/** + * cgroup_get_from_fd - get a cgroup pointer from a fd + * @fd: fd obtained by open(cgroup2_dir) + * + * Find the cgroup from a fd which should be obtained + * by opening a cgroup directory. Returns a pointer to the + * cgroup on success. ERR_PTR is returned if the cgroup + * cannot be found. + */ +struct cgroup *cgroup_get_from_fd(int fd) +{ + struct cgroup_subsys_state *css; + struct cgroup *cgrp; + struct file *f; + + f = fget_raw(fd); + if (!f) + return ERR_PTR(-EBADF); + + css = css_tryget_online_from_dir(f->f_path.dentry, NULL); + fput(f); + if (IS_ERR(css)) + return ERR_CAST(css); + + cgrp = css->cgroup; + if (!cgroup_on_dfl(cgrp)) { + cgroup_put(cgrp); + return ERR_PTR(-EBADF); + } + + return cgrp; +} +EXPORT_SYMBOL_GPL(cgroup_get_from_fd); + /* * sock->sk_cgrp_data handling. For more info, see sock_cgroup_data * definition in cgroup-defs.h. -- cgit v0.10.2 From 4ed8ec521ed57c4e207ad464ca0388776de74d4b Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 30 Jun 2016 10:28:43 -0700 Subject: cgroup: bpf: Add BPF_MAP_TYPE_CGROUP_ARRAY Add a BPF_MAP_TYPE_CGROUP_ARRAY and its bpf_map_ops's implementations. To update an element, the caller is expected to obtain a cgroup2 backed fd by open(cgroup2_dir) and then update the array with that fd. Signed-off-by: Martin KaFai Lau Cc: Alexei Starovoitov Cc: Daniel Borkmann Cc: Tejun Heo Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index be6ac12..26c04be 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -84,6 +84,7 @@ enum bpf_map_type { BPF_MAP_TYPE_PERCPU_HASH, BPF_MAP_TYPE_PERCPU_ARRAY, BPF_MAP_TYPE_STACK_TRACE, + BPF_MAP_TYPE_CGROUP_ARRAY, }; enum bpf_prog_type { diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 4ec57a6..db1a743 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -537,3 +537,46 @@ static int __init register_perf_event_array_map(void) return 0; } late_initcall(register_perf_event_array_map); + +#ifdef CONFIG_SOCK_CGROUP_DATA +static void *cgroup_fd_array_get_ptr(struct bpf_map *map, + struct file *map_file /* not used */, + int fd) +{ + return cgroup_get_from_fd(fd); +} + +static void cgroup_fd_array_put_ptr(void *ptr) +{ + /* cgroup_put free cgrp after a rcu grace period */ + cgroup_put(ptr); +} + +static void cgroup_fd_array_free(struct bpf_map *map) +{ + bpf_fd_array_map_clear(map); + fd_array_map_free(map); +} + +static const struct bpf_map_ops cgroup_array_ops = { + .map_alloc = fd_array_map_alloc, + .map_free = cgroup_fd_array_free, + .map_get_next_key = array_map_get_next_key, + .map_lookup_elem = fd_array_map_lookup_elem, + .map_delete_elem = fd_array_map_delete_elem, + .map_fd_get_ptr = cgroup_fd_array_get_ptr, + .map_fd_put_ptr = cgroup_fd_array_put_ptr, +}; + +static struct bpf_map_type_list cgroup_array_type __read_mostly = { + .ops = &cgroup_array_ops, + .type = BPF_MAP_TYPE_CGROUP_ARRAY, +}; + +static int __init register_cgroup_array_map(void) +{ + bpf_register_map_type(&cgroup_array_type); + return 0; +} +late_initcall(register_cgroup_array_map); +#endif diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 22863d9..96d938a 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -393,7 +393,8 @@ static int map_update_elem(union bpf_attr *attr) } else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) { err = bpf_percpu_array_update(map, key, value, attr->flags); } else if (map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || - map->map_type == BPF_MAP_TYPE_PROG_ARRAY) { + map->map_type == BPF_MAP_TYPE_PROG_ARRAY || + map->map_type == BPF_MAP_TYPE_CGROUP_ARRAY) { rcu_read_lock(); err = bpf_fd_array_map_update_elem(map, f.file, key, value, attr->flags); diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index eec9f90..69ba225 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1035,6 +1035,8 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) if (func_id != BPF_FUNC_get_stackid) goto error; break; + case BPF_MAP_TYPE_CGROUP_ARRAY: + goto error; default: break; } -- cgit v0.10.2 From 4a482f34afcc162d8456f449b137ec2a95be60d8 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 30 Jun 2016 10:28:44 -0700 Subject: cgroup: bpf: Add bpf_skb_in_cgroup_proto Adds a bpf helper, bpf_skb_in_cgroup, to decide if a skb->sk belongs to a descendant of a cgroup2. It is similar to the feature added in netfilter: commit c38c4597e4bf ("netfilter: implement xt_cgroup cgroup2 path match") The user is expected to populate a BPF_MAP_TYPE_CGROUP_ARRAY which will be used by the bpf_skb_in_cgroup. Modifications to the bpf verifier is to ensure BPF_MAP_TYPE_CGROUP_ARRAY and bpf_skb_in_cgroup() are always used together. Signed-off-by: Martin KaFai Lau Cc: Alexei Starovoitov Cc: Daniel Borkmann Cc: Tejun Heo Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 26c04be..f44504d 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -337,6 +337,17 @@ enum bpf_func_id { */ BPF_FUNC_skb_change_type, + /** + * bpf_skb_in_cgroup(skb, map, index) - Check cgroup2 membership of skb + * @skb: pointer to skb + * @map: pointer to bpf_map in BPF_MAP_TYPE_CGROUP_ARRAY type + * @index: index of the cgroup in the bpf_map + * Return: + * == 0 skb failed the cgroup2 descendant test + * == 1 skb succeeded the cgroup2 descendant test + * < 0 error + */ + BPF_FUNC_skb_in_cgroup, __BPF_FUNC_MAX_ID, }; diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 69ba225..e206c21 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1036,7 +1036,9 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) goto error; break; case BPF_MAP_TYPE_CGROUP_ARRAY: - goto error; + if (func_id != BPF_FUNC_skb_in_cgroup) + goto error; + break; default: break; } @@ -1056,6 +1058,10 @@ static int check_map_func_compatibility(struct bpf_map *map, int func_id) if (map->map_type != BPF_MAP_TYPE_STACK_TRACE) goto error; break; + case BPF_FUNC_skb_in_cgroup: + if (map->map_type != BPF_MAP_TYPE_CGROUP_ARRAY) + goto error; + break; default: break; } diff --git a/net/core/filter.c b/net/core/filter.c index 76fee35..54071cf 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2239,6 +2239,40 @@ bpf_get_skb_set_tunnel_proto(enum bpf_func_id which) } } +#ifdef CONFIG_SOCK_CGROUP_DATA +static u64 bpf_skb_in_cgroup(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + struct sk_buff *skb = (struct sk_buff *)(long)r1; + struct bpf_map *map = (struct bpf_map *)(long)r2; + struct bpf_array *array = container_of(map, struct bpf_array, map); + struct cgroup *cgrp; + struct sock *sk; + u32 i = (u32)r3; + + sk = skb->sk; + if (!sk || !sk_fullsock(sk)) + return -ENOENT; + + if (unlikely(i >= array->map.max_entries)) + return -E2BIG; + + cgrp = READ_ONCE(array->ptrs[i]); + if (unlikely(!cgrp)) + return -EAGAIN; + + return cgroup_is_descendant(sock_cgroup_ptr(&sk->sk_cgrp_data), cgrp); +} + +static const struct bpf_func_proto bpf_skb_in_cgroup_proto = { + .func = bpf_skb_in_cgroup, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_CONST_MAP_PTR, + .arg3_type = ARG_ANYTHING, +}; +#endif + static const struct bpf_func_proto * sk_filter_func_proto(enum bpf_func_id func_id) { @@ -2307,6 +2341,10 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) return bpf_get_event_output_proto(); case BPF_FUNC_get_smp_processor_id: return &bpf_get_smp_processor_id_proto; +#ifdef CONFIG_SOCK_CGROUP_DATA + case BPF_FUNC_skb_in_cgroup: + return &bpf_skb_in_cgroup_proto; +#endif default: return sk_filter_func_proto(func_id); } -- cgit v0.10.2 From a3f74617340b598dbc7eb5b68d4ed53b4a70f5eb Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 30 Jun 2016 10:28:45 -0700 Subject: cgroup: bpf: Add an example to do cgroup checking in BPF test_cgrp2_array_pin.c: A userland program that creates a bpf_map (BPF_MAP_TYPE_GROUP_ARRAY), pouplates/updates it with a cgroup2's backed fd and pins it to a bpf-fs's file. The pinned file can be loaded by tc and then used by the bpf prog later. This program can also update an existing pinned array and it could be useful for debugging/testing purpose. test_cgrp2_tc_kern.c: A bpf prog which should be loaded by tc. It is to demonstrate the usage of bpf_skb_in_cgroup. test_cgrp2_tc.sh: A script that glues the test_cgrp2_array_pin.c and test_cgrp2_tc_kern.c together. The idea is like: 1. Load the test_cgrp2_tc_kern.o by tc 2. Use test_cgrp2_array_pin.c to populate a BPF_MAP_TYPE_CGROUP_ARRAY with a cgroup fd 3. Do a 'ping -6 ff02::1%ve' to ensure the packet has been dropped because of a match on the cgroup Most of the lines in test_cgrp2_tc.sh is the boilerplate to setup the cgroup/bpf-fs/net-devices/netns...etc. It is not bulletproof on errors but should work well enough and give enough debug info if things did not go well. Signed-off-by: Martin KaFai Lau Cc: Alexei Starovoitov Cc: Daniel Borkmann Cc: Tejun Heo Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile index 0bf2478..a98b780 100644 --- a/samples/bpf/Makefile +++ b/samples/bpf/Makefile @@ -20,6 +20,7 @@ hostprogs-y += offwaketime hostprogs-y += spintest hostprogs-y += map_perf_test hostprogs-y += test_overhead +hostprogs-y += test_cgrp2_array_pin test_verifier-objs := test_verifier.o libbpf.o test_maps-objs := test_maps.o libbpf.o @@ -40,6 +41,7 @@ offwaketime-objs := bpf_load.o libbpf.o offwaketime_user.o spintest-objs := bpf_load.o libbpf.o spintest_user.o map_perf_test-objs := bpf_load.o libbpf.o map_perf_test_user.o test_overhead-objs := bpf_load.o libbpf.o test_overhead_user.o +test_cgrp2_array_pin-objs := libbpf.o test_cgrp2_array_pin.o # Tell kbuild to always build the programs always := $(hostprogs-y) @@ -61,6 +63,7 @@ always += map_perf_test_kern.o always += test_overhead_tp_kern.o always += test_overhead_kprobe_kern.o always += parse_varlen.o parse_simple.o parse_ldabs.o +always += test_cgrp2_tc_kern.o HOSTCFLAGS += -I$(objtree)/usr/include diff --git a/samples/bpf/bpf_helpers.h b/samples/bpf/bpf_helpers.h index 7904a2a..84e3fd9 100644 --- a/samples/bpf/bpf_helpers.h +++ b/samples/bpf/bpf_helpers.h @@ -70,6 +70,8 @@ static int (*bpf_l3_csum_replace)(void *ctx, int off, int from, int to, int flag (void *) BPF_FUNC_l3_csum_replace; static int (*bpf_l4_csum_replace)(void *ctx, int off, int from, int to, int flags) = (void *) BPF_FUNC_l4_csum_replace; +static int (*bpf_skb_in_cgroup)(void *ctx, void *map, int index) = + (void *) BPF_FUNC_skb_in_cgroup; #if defined(__x86_64__) diff --git a/samples/bpf/test_cgrp2_array_pin.c b/samples/bpf/test_cgrp2_array_pin.c new file mode 100644 index 0000000..70e86f7 --- /dev/null +++ b/samples/bpf/test_cgrp2_array_pin.c @@ -0,0 +1,109 @@ +/* Copyright (c) 2016 Facebook + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ +#include +#include + +#include +#include +#include +#include +#include +#include + +#include "libbpf.h" + +static void usage(void) +{ + printf("Usage: test_cgrp2_array_pin [...]\n"); + printf(" -F File to pin an BPF cgroup array\n"); + printf(" -U Update an already pinned BPF cgroup array\n"); + printf(" -v Full path of the cgroup2\n"); + printf(" -h Display this help\n"); +} + +int main(int argc, char **argv) +{ + const char *pinned_file = NULL, *cg2 = NULL; + int create_array = 1; + int array_key = 0; + int array_fd = -1; + int cg2_fd = -1; + int ret = -1; + int opt; + + while ((opt = getopt(argc, argv, "F:U:v:")) != -1) { + switch (opt) { + /* General args */ + case 'F': + pinned_file = optarg; + break; + case 'U': + pinned_file = optarg; + create_array = 0; + break; + case 'v': + cg2 = optarg; + break; + default: + usage(); + goto out; + } + } + + if (!cg2 || !pinned_file) { + usage(); + goto out; + } + + cg2_fd = open(cg2, O_RDONLY); + if (cg2_fd < 0) { + fprintf(stderr, "open(%s,...): %s(%d)\n", + cg2, strerror(errno), errno); + goto out; + } + + if (create_array) { + array_fd = bpf_create_map(BPF_MAP_TYPE_CGROUP_ARRAY, + sizeof(uint32_t), sizeof(uint32_t), + 1, 0); + if (array_fd < 0) { + fprintf(stderr, + "bpf_create_map(BPF_MAP_TYPE_CGROUP_ARRAY,...): %s(%d)\n", + strerror(errno), errno); + goto out; + } + } else { + array_fd = bpf_obj_get(pinned_file); + if (array_fd < 0) { + fprintf(stderr, "bpf_obj_get(%s): %s(%d)\n", + pinned_file, strerror(errno), errno); + goto out; + } + } + + ret = bpf_update_elem(array_fd, &array_key, &cg2_fd, 0); + if (ret) { + perror("bpf_update_elem"); + goto out; + } + + if (create_array) { + ret = bpf_obj_pin(array_fd, pinned_file); + if (ret) { + fprintf(stderr, "bpf_obj_pin(..., %s): %s(%d)\n", + pinned_file, strerror(errno), errno); + goto out; + } + } + +out: + if (array_fd != -1) + close(array_fd); + if (cg2_fd != -1) + close(cg2_fd); + return ret; +} diff --git a/samples/bpf/test_cgrp2_tc.sh b/samples/bpf/test_cgrp2_tc.sh new file mode 100755 index 0000000..0b119ee --- /dev/null +++ b/samples/bpf/test_cgrp2_tc.sh @@ -0,0 +1,184 @@ +#!/bin/bash + +MY_DIR=$(dirname $0) +# Details on the bpf prog +BPF_CGRP2_ARRAY_NAME='test_cgrp2_array_pin' +BPF_PROG="$MY_DIR/test_cgrp2_tc_kern.o" +BPF_SECTION='filter' + +[ -z "$TC" ] && TC='tc' +[ -z "$IP" ] && IP='ip' + +# Names of the veth interface, net namespace...etc. +HOST_IFC='ve' +NS_IFC='vens' +NS='ns' + +find_mnt() { + cat /proc/mounts | \ + awk '{ if ($3 == "'$1'" && mnt == "") { mnt = $2 }} END { print mnt }' +} + +# Init cgroup2 vars +init_cgrp2_vars() { + CGRP2_ROOT=$(find_mnt cgroup2) + if [ -z "$CGRP2_ROOT" ] + then + CGRP2_ROOT='/mnt/cgroup2' + MOUNT_CGRP2="yes" + fi + CGRP2_TC="$CGRP2_ROOT/tc" + CGRP2_TC_LEAF="$CGRP2_TC/leaf" +} + +# Init bpf fs vars +init_bpf_fs_vars() { + local bpf_fs_root=$(find_mnt bpf) + [ -n "$bpf_fs_root" ] || return -1 + BPF_FS_TC_SHARE="$bpf_fs_root/tc/globals" +} + +setup_cgrp2() { + case $1 in + start) + if [ "$MOUNT_CGRP2" == 'yes' ] + then + [ -d $CGRP2_ROOT ] || mkdir -p $CGRP2_ROOT + mount -t cgroup2 none $CGRP2_ROOT || return $? + fi + mkdir -p $CGRP2_TC_LEAF + ;; + *) + rmdir $CGRP2_TC_LEAF && rmdir $CGRP2_TC + [ "$MOUNT_CGRP2" == 'yes' ] && umount $CGRP2_ROOT + ;; + esac +} + +setup_bpf_cgrp2_array() { + local bpf_cgrp2_array="$BPF_FS_TC_SHARE/$BPF_CGRP2_ARRAY_NAME" + case $1 in + start) + $MY_DIR/test_cgrp2_array_pin -U $bpf_cgrp2_array -v $CGRP2_TC + ;; + *) + [ -d "$BPF_FS_TC_SHARE" ] && rm -f $bpf_cgrp2_array + ;; + esac +} + +setup_net() { + case $1 in + start) + $IP link add $HOST_IFC type veth peer name $NS_IFC || return $? + $IP link set dev $HOST_IFC up || return $? + sysctl -q net.ipv6.conf.$HOST_IFC.accept_dad=0 + + $IP netns add ns || return $? + $IP link set dev $NS_IFC netns ns || return $? + $IP -n $NS link set dev $NS_IFC up || return $? + $IP netns exec $NS sysctl -q net.ipv6.conf.$NS_IFC.accept_dad=0 + $TC qdisc add dev $HOST_IFC clsact || return $? + $TC filter add dev $HOST_IFC egress bpf da obj $BPF_PROG sec $BPF_SECTION || return $? + ;; + *) + $IP netns del $NS + $IP link del $HOST_IFC + ;; + esac +} + +run_in_cgrp() { + # Fork another bash and move it under the specified cgroup. + # It makes the cgroup cleanup easier at the end of the test. + cmd='echo $$ > ' + cmd="$cmd $1/cgroup.procs; exec $2" + bash -c "$cmd" +} + +do_test() { + run_in_cgrp $CGRP2_TC_LEAF "ping -6 -c3 ff02::1%$HOST_IFC >& /dev/null" + local dropped=$($TC -s qdisc show dev $HOST_IFC | tail -3 | \ + awk '/drop/{print substr($7, 0, index($7, ",")-1)}') + if [[ $dropped -eq 0 ]] + then + echo "FAIL" + return 1 + else + echo "Successfully filtered $dropped packets" + return 0 + fi +} + +do_exit() { + if [ "$DEBUG" == "yes" ] && [ "$MODE" != 'cleanuponly' ] + then + echo "------ DEBUG ------" + echo "mount: "; mount | egrep '(cgroup2|bpf)'; echo + echo "$CGRP2_TC_LEAF: "; ls -l $CGRP2_TC_LEAF; echo + if [ -d "$BPF_FS_TC_SHARE" ] + then + echo "$BPF_FS_TC_SHARE: "; ls -l $BPF_FS_TC_SHARE; echo + fi + echo "Host net:" + $IP netns + $IP link show dev $HOST_IFC + $IP -6 a show dev $HOST_IFC + $TC -s qdisc show dev $HOST_IFC + echo + echo "$NS net:" + $IP -n $NS link show dev $NS_IFC + $IP -n $NS -6 link show dev $NS_IFC + echo "------ DEBUG ------" + echo + fi + + if [ "$MODE" != 'nocleanup' ] + then + setup_net stop + setup_bpf_cgrp2_array stop + setup_cgrp2 stop + fi +} + +init_cgrp2_vars +init_bpf_fs_vars + +while [[ $# -ge 1 ]] +do + a="$1" + case $a in + debug) + DEBUG='yes' + shift 1 + ;; + cleanup-only) + MODE='cleanuponly' + shift 1 + ;; + no-cleanup) + MODE='nocleanup' + shift 1 + ;; + *) + echo "test_cgrp2_tc [debug] [cleanup-only | no-cleanup]" + echo " debug: Print cgrp and network setup details at the end of the test" + echo " cleanup-only: Try to cleanup things from last test. No test will be run" + echo " no-cleanup: Run the test but don't do cleanup at the end" + echo "[Note: If no arg is given, it will run the test and do cleanup at the end]" + echo + exit -1 + ;; + esac +done + +trap do_exit 0 + +[ "$MODE" == 'cleanuponly' ] && exit + +setup_cgrp2 start || exit $? +setup_net start || exit $? +init_bpf_fs_vars || exit $? +setup_bpf_cgrp2_array start || exit $? +do_test +echo diff --git a/samples/bpf/test_cgrp2_tc_kern.c b/samples/bpf/test_cgrp2_tc_kern.c new file mode 100644 index 0000000..2732c37 --- /dev/null +++ b/samples/bpf/test_cgrp2_tc_kern.c @@ -0,0 +1,69 @@ +/* Copyright (c) 2016 Facebook + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include +#include "bpf_helpers.h" + +/* copy of 'struct ethhdr' without __packed */ +struct eth_hdr { + unsigned char h_dest[ETH_ALEN]; + unsigned char h_source[ETH_ALEN]; + unsigned short h_proto; +}; + +#define PIN_GLOBAL_NS 2 +struct bpf_elf_map { + __u32 type; + __u32 size_key; + __u32 size_value; + __u32 max_elem; + __u32 flags; + __u32 id; + __u32 pinning; +}; + +struct bpf_elf_map SEC("maps") test_cgrp2_array_pin = { + .type = BPF_MAP_TYPE_CGROUP_ARRAY, + .size_key = sizeof(uint32_t), + .size_value = sizeof(uint32_t), + .pinning = PIN_GLOBAL_NS, + .max_elem = 1, +}; + +SEC("filter") +int handle_egress(struct __sk_buff *skb) +{ + void *data = (void *)(long)skb->data; + struct eth_hdr *eth = data; + struct ipv6hdr *ip6h = data + sizeof(*eth); + void *data_end = (void *)(long)skb->data_end; + char dont_care_msg[] = "dont care %04x %d\n"; + char pass_msg[] = "pass\n"; + char reject_msg[] = "reject\n"; + + /* single length check */ + if (data + sizeof(*eth) + sizeof(*ip6h) > data_end) + return TC_ACT_OK; + + if (eth->h_proto != htons(ETH_P_IPV6) || + ip6h->nexthdr != IPPROTO_ICMPV6) { + bpf_trace_printk(dont_care_msg, sizeof(dont_care_msg), + eth->h_proto, ip6h->nexthdr); + return TC_ACT_OK; + } else if (bpf_skb_in_cgroup(skb, &test_cgrp2_array_pin, 0) != 1) { + bpf_trace_printk(pass_msg, sizeof(pass_msg)); + return TC_ACT_OK; + } else { + bpf_trace_printk(reject_msg, sizeof(reject_msg)); + return TC_ACT_SHOT; + } +} + +char _license[] SEC("license") = "GPL"; -- cgit v0.10.2 From 226f7a7d97e37220a442f52eb85ebff2cd6fc0d2 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 30 Jun 2016 16:11:10 -0700 Subject: RDS: Rework path specific indirections Refactor code to avoid separate indirections for single-path and multipath transports. All transports (both single and mp-capable) will get a pointer to the rds_conn_path, and can trivially derive the rds_connection from the ->cp_conn. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/connection.c b/net/rds/connection.c index a4b07c8..17c2f25 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -326,10 +326,7 @@ void rds_conn_shutdown(struct rds_conn_path *cp) wait_event(cp->cp_waitq, !test_bit(RDS_RECV_REFILL, &cp->cp_flags)); - if (!conn->c_trans->t_mp_capable) - conn->c_trans->conn_shutdown(conn); - else - conn->c_trans->conn_path_shutdown(cp); + conn->c_trans->conn_path_shutdown(cp); rds_conn_path_reset(cp); if (!rds_conn_path_transition(cp, RDS_CONN_DISCONNECTING, diff --git a/net/rds/ib.c b/net/rds/ib.c index 44946a6..1b29ec9 100644 --- a/net/rds/ib.c +++ b/net/rds/ib.c @@ -381,7 +381,7 @@ void rds_ib_exit(void) struct rds_transport rds_ib_transport = { .laddr_check = rds_ib_laddr_check, - .xmit_complete = rds_ib_xmit_complete, + .xmit_path_complete = rds_ib_xmit_path_complete, .xmit = rds_ib_xmit, .xmit_rdma = rds_ib_xmit_rdma, .xmit_atomic = rds_ib_xmit_atomic, @@ -389,7 +389,7 @@ struct rds_transport rds_ib_transport = { .conn_alloc = rds_ib_conn_alloc, .conn_free = rds_ib_conn_free, .conn_connect = rds_ib_conn_connect, - .conn_shutdown = rds_ib_conn_shutdown, + .conn_path_shutdown = rds_ib_conn_path_shutdown, .inc_copy_to_user = rds_ib_inc_copy_to_user, .inc_free = rds_ib_inc_free, .cm_initiate_connect = rds_ib_cm_initiate_connect, diff --git a/net/rds/ib.h b/net/rds/ib.h index 627fb79..2051f4b 100644 --- a/net/rds/ib.h +++ b/net/rds/ib.h @@ -329,7 +329,7 @@ extern struct list_head ib_nodev_conns; int rds_ib_conn_alloc(struct rds_connection *conn, gfp_t gfp); void rds_ib_conn_free(void *arg); int rds_ib_conn_connect(struct rds_connection *conn); -void rds_ib_conn_shutdown(struct rds_connection *conn); +void rds_ib_conn_path_shutdown(struct rds_conn_path *cp); void rds_ib_state_change(struct sock *sk); int rds_ib_listen_init(void); void rds_ib_listen_stop(void); @@ -384,7 +384,7 @@ u32 rds_ib_ring_completed(struct rds_ib_work_ring *ring, u32 wr_id, u32 oldest); extern wait_queue_head_t rds_ib_ring_empty_wait; /* ib_send.c */ -void rds_ib_xmit_complete(struct rds_connection *conn); +void rds_ib_xmit_path_complete(struct rds_conn_path *cp); int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm, unsigned int hdr_off, unsigned int sg, unsigned int off); void rds_ib_send_cqe_handler(struct rds_ib_connection *ic, struct ib_wc *wc); diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c index e48bb1b..e34ea0b 100644 --- a/net/rds/ib_cm.c +++ b/net/rds/ib_cm.c @@ -731,8 +731,9 @@ out: * so that it can be called at any point during startup. In fact it * can be called multiple times for a given connection. */ -void rds_ib_conn_shutdown(struct rds_connection *conn) +void rds_ib_conn_path_shutdown(struct rds_conn_path *cp) { + struct rds_connection *conn = cp->cp_conn; struct rds_ib_connection *ic = conn->c_transport_data; int err = 0; diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c index 6e4110a..84d90c9 100644 --- a/net/rds/ib_send.c +++ b/net/rds/ib_send.c @@ -980,8 +980,9 @@ out: return ret; } -void rds_ib_xmit_complete(struct rds_connection *conn) +void rds_ib_xmit_path_complete(struct rds_conn_path *cp) { + struct rds_connection *conn = cp->cp_conn; struct rds_ib_connection *ic = conn->c_transport_data; /* We may have a pending ACK or window update we were unable diff --git a/net/rds/loop.c b/net/rds/loop.c index 15f83db..318c21d 100644 --- a/net/rds/loop.c +++ b/net/rds/loop.c @@ -156,7 +156,7 @@ static int rds_loop_conn_connect(struct rds_connection *conn) return 0; } -static void rds_loop_conn_shutdown(struct rds_connection *conn) +static void rds_loop_conn_path_shutdown(struct rds_conn_path *cp) { } @@ -189,7 +189,7 @@ struct rds_transport rds_loop_transport = { .conn_alloc = rds_loop_conn_alloc, .conn_free = rds_loop_conn_free, .conn_connect = rds_loop_conn_connect, - .conn_shutdown = rds_loop_conn_shutdown, + .conn_path_shutdown = rds_loop_conn_path_shutdown, .inc_copy_to_user = rds_message_inc_copy_to_user, .inc_free = rds_loop_inc_free, .t_name = "loopback", diff --git a/net/rds/rds.h b/net/rds/rds.h index 2e35b73..5bbad08 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -455,11 +455,8 @@ struct rds_transport { int (*conn_alloc)(struct rds_connection *conn, gfp_t gfp); void (*conn_free)(void *data); int (*conn_connect)(struct rds_connection *conn); - void (*conn_shutdown)(struct rds_connection *conn); void (*conn_path_shutdown)(struct rds_conn_path *conn); - void (*xmit_prepare)(struct rds_connection *conn); void (*xmit_path_prepare)(struct rds_conn_path *cp); - void (*xmit_complete)(struct rds_connection *conn); void (*xmit_path_complete)(struct rds_conn_path *cp); int (*xmit)(struct rds_connection *conn, struct rds_message *rm, unsigned int hdr_off, unsigned int sg, unsigned int off); diff --git a/net/rds/send.c b/net/rds/send.c index ee43d6b..5a9caf1 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -183,12 +183,8 @@ restart: goto out; } - if (conn->c_trans->t_mp_capable) { - if (conn->c_trans->xmit_path_prepare) - conn->c_trans->xmit_path_prepare(cp); - } else if (conn->c_trans->xmit_prepare) { - conn->c_trans->xmit_prepare(conn); - } + if (conn->c_trans->xmit_path_prepare) + conn->c_trans->xmit_path_prepare(cp); /* * spin trying to push headers and data down the connection until @@ -403,12 +399,8 @@ restart: } over_batch: - if (conn->c_trans->t_mp_capable) { - if (conn->c_trans->xmit_path_complete) - conn->c_trans->xmit_path_complete(cp); - } else if (conn->c_trans->xmit_complete) { - conn->c_trans->xmit_complete(conn); - } + if (conn->c_trans->xmit_path_complete) + conn->c_trans->xmit_path_complete(cp); release_in_xmit(cp); /* Nuke any messages we decided not to retransmit. */ diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 5217d49..b139630 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -340,14 +340,14 @@ static void rds_tcp_exit(void); struct rds_transport rds_tcp_transport = { .laddr_check = rds_tcp_laddr_check, - .xmit_prepare = rds_tcp_xmit_prepare, - .xmit_complete = rds_tcp_xmit_complete, + .xmit_path_prepare = rds_tcp_xmit_path_prepare, + .xmit_path_complete = rds_tcp_xmit_path_complete, .xmit = rds_tcp_xmit, .recv = rds_tcp_recv, .conn_alloc = rds_tcp_conn_alloc, .conn_free = rds_tcp_conn_free, .conn_connect = rds_tcp_conn_connect, - .conn_shutdown = rds_tcp_conn_shutdown, + .conn_path_shutdown = rds_tcp_conn_path_shutdown, .inc_copy_to_user = rds_tcp_inc_copy_to_user, .inc_free = rds_tcp_inc_free, .stats_info_copy = rds_tcp_stats_info_copy, diff --git a/net/rds/tcp.h b/net/rds/tcp.h index 7940bab..728abe2 100644 --- a/net/rds/tcp.h +++ b/net/rds/tcp.h @@ -61,7 +61,7 @@ void rds_tcp_accept_work(struct sock *sk); /* tcp_connect.c */ int rds_tcp_conn_connect(struct rds_connection *conn); -void rds_tcp_conn_shutdown(struct rds_connection *conn); +void rds_tcp_conn_path_shutdown(struct rds_conn_path *conn); void rds_tcp_state_change(struct sock *sk); /* tcp_listen.c */ @@ -80,8 +80,8 @@ void rds_tcp_inc_free(struct rds_incoming *inc); int rds_tcp_inc_copy_to_user(struct rds_incoming *inc, struct iov_iter *to); /* tcp_send.c */ -void rds_tcp_xmit_prepare(struct rds_connection *conn); -void rds_tcp_xmit_complete(struct rds_connection *conn); +void rds_tcp_xmit_path_prepare(struct rds_conn_path *cp); +void rds_tcp_xmit_path_complete(struct rds_conn_path *cp); int rds_tcp_xmit(struct rds_connection *conn, struct rds_message *rm, unsigned int hdr_off, unsigned int sg, unsigned int off); void rds_tcp_write_space(struct sock *sk); diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c index 96c2c4d..aa65c16 100644 --- a/net/rds/tcp_connect.c +++ b/net/rds/tcp_connect.c @@ -144,12 +144,13 @@ out: * callbacks to those set by TCP. Our callbacks won't execute again once we * hold the sock lock. */ -void rds_tcp_conn_shutdown(struct rds_connection *conn) +void rds_tcp_conn_path_shutdown(struct rds_conn_path *cp) { - struct rds_tcp_connection *tc = conn->c_transport_data; + struct rds_tcp_connection *tc = cp->cp_transport_data; struct socket *sock = tc->t_sock; - rdsdebug("shutting down conn %p tc %p sock %p\n", conn, tc, sock); + rdsdebug("shutting down conn %p tc %p sock %p\n", + cp->cp_conn, tc, sock); if (sock) { sock->ops->shutdown(sock, RCV_SHUTDOWN | SEND_SHUTDOWN); diff --git a/net/rds/tcp_send.c b/net/rds/tcp_send.c index 710f1aa..52cda94 100644 --- a/net/rds/tcp_send.c +++ b/net/rds/tcp_send.c @@ -49,16 +49,16 @@ static void rds_tcp_cork(struct socket *sock, int val) set_fs(oldfs); } -void rds_tcp_xmit_prepare(struct rds_connection *conn) +void rds_tcp_xmit_path_prepare(struct rds_conn_path *cp) { - struct rds_tcp_connection *tc = conn->c_transport_data; + struct rds_tcp_connection *tc = cp->cp_transport_data; rds_tcp_cork(tc->t_sock, 1); } -void rds_tcp_xmit_complete(struct rds_connection *conn) +void rds_tcp_xmit_path_complete(struct rds_conn_path *cp) { - struct rds_tcp_connection *tc = conn->c_transport_data; + struct rds_tcp_connection *tc = cp->cp_transport_data; rds_tcp_cork(tc->t_sock, 0); } -- cgit v0.10.2 From 26e4e6bb683028546f339018ab4cd394300a92a4 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 30 Jun 2016 16:11:11 -0700 Subject: RDS: TCP: Remove dead logic around c_passive in rds-tcp The c_passive bit is only intended for the IB transport and will never be encountered in rds-tcp, so remove the dead logic that predicates on this bit. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/tcp.c b/net/rds/tcp.c index b139630..c56fff2 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -329,11 +329,8 @@ static void rds_tcp_destroy_conns(void) INIT_LIST_HEAD(&rds_tcp_conn_list); spin_unlock_irq(&rds_tcp_conn_lock); - list_for_each_entry_safe(tc, _tc, &tmp_list, t_tcp_node) { - if (tc->conn->c_passive) - rds_conn_destroy(tc->conn->c_passive); + list_for_each_entry_safe(tc, _tc, &tmp_list, t_tcp_node) rds_conn_destroy(tc->conn); - } } static void rds_tcp_exit(void); @@ -512,8 +509,6 @@ static void rds_tcp_kill_sock(struct net *net) sk = tc->t_sock->sk; sk->sk_prot->disconnect(sk, 0); tcp_done(sk); - if (tc->conn->c_passive) - rds_conn_destroy(tc->conn->c_passive); rds_conn_destroy(tc->conn); } } -- cgit v0.10.2 From 02105b2ccdd6344146e0296172a9e0f17ff624ef Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 30 Jun 2016 16:11:12 -0700 Subject: RDS: TCP: Make rds_tcp_connection track the rds_conn_path The struct rds_tcp_connection is the transport-specific private data structure that tracks TCP information per rds_conn_path. Modify this structure to have a back-pointer to the rds_conn_path for which it is the ->cp_transport_data. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/connection.c b/net/rds/connection.c index 17c2f25..1b0c2a78 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -253,9 +253,12 @@ static struct rds_connection *__rds_conn_create(struct net *net, for (i = 0; i < RDS_MPATH_WORKERS; i++) { cp = &conn->c_path[i]; - trans->conn_free(cp->cp_transport_data); - if (!trans->t_mp_capable) - break; + /* The ->conn_alloc invocation may have + * allocated resource for all paths, so all + * of them may have to be freed here. + */ + if (cp->cp_transport_data) + trans->conn_free(cp->cp_transport_data); } kmem_cache_free(rds_conn_slab, conn); conn = found; @@ -367,6 +370,9 @@ static void rds_conn_path_destroy(struct rds_conn_path *cp) { struct rds_message *rm, *rtmp; + if (!cp->cp_transport_data) + return; + rds_conn_path_drop(cp); flush_work(&cp->cp_down_w); @@ -398,6 +404,8 @@ static void rds_conn_path_destroy(struct rds_conn_path *cp) void rds_conn_destroy(struct rds_connection *conn) { unsigned long flags; + int i; + struct rds_conn_path *cp; rdsdebug("freeing conn %p for %pI4 -> " "%pI4\n", conn, &conn->c_laddr, @@ -410,18 +418,10 @@ void rds_conn_destroy(struct rds_connection *conn) synchronize_rcu(); /* shut the connection down */ - if (!conn->c_trans->t_mp_capable) { - rds_conn_path_destroy(&conn->c_path[0]); - BUG_ON(!list_empty(&conn->c_path[0].cp_retrans)); - } else { - int i; - struct rds_conn_path *cp; - - for (i = 0; i < RDS_MPATH_WORKERS; i++) { - cp = &conn->c_path[i]; - rds_conn_path_destroy(cp); - BUG_ON(!list_empty(&cp->cp_retrans)); - } + for (i = 0; i < RDS_MPATH_WORKERS; i++) { + cp = &conn->c_path[i]; + rds_conn_path_destroy(cp); + BUG_ON(!list_empty(&cp->cp_retrans)); } /* diff --git a/net/rds/tcp.c b/net/rds/tcp.c index c56fff2..c6b47f6 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -221,7 +221,7 @@ void rds_tcp_set_callbacks(struct socket *sock, struct rds_connection *conn) sock->sk->sk_data_ready = sock->sk->sk_user_data; tc->t_sock = sock; - tc->conn = conn; + tc->t_cpath = &conn->c_path[0]; tc->t_orig_data_ready = sock->sk->sk_data_ready; tc->t_orig_write_space = sock->sk->sk_write_space; tc->t_orig_state_change = sock->sk->sk_state_change; @@ -284,24 +284,29 @@ static int rds_tcp_laddr_check(struct net *net, __be32 addr) static int rds_tcp_conn_alloc(struct rds_connection *conn, gfp_t gfp) { struct rds_tcp_connection *tc; + int i; - tc = kmem_cache_alloc(rds_tcp_conn_slab, gfp); - if (!tc) - return -ENOMEM; + for (i = 0; i < RDS_MPATH_WORKERS; i++) { + tc = kmem_cache_alloc(rds_tcp_conn_slab, gfp); + if (!tc) + return -ENOMEM; - mutex_init(&tc->t_conn_lock); - tc->t_sock = NULL; - tc->t_tinc = NULL; - tc->t_tinc_hdr_rem = sizeof(struct rds_header); - tc->t_tinc_data_rem = 0; + mutex_init(&tc->t_conn_path_lock); + tc->t_sock = NULL; + tc->t_tinc = NULL; + tc->t_tinc_hdr_rem = sizeof(struct rds_header); + tc->t_tinc_data_rem = 0; - conn->c_transport_data = tc; + conn->c_path[i].cp_transport_data = tc; + tc->t_cpath = &conn->c_path[i]; - spin_lock_irq(&rds_tcp_conn_lock); - list_add_tail(&tc->t_tcp_node, &rds_tcp_conn_list); - spin_unlock_irq(&rds_tcp_conn_lock); + spin_lock_irq(&rds_tcp_conn_lock); + list_add_tail(&tc->t_tcp_node, &rds_tcp_conn_list); + spin_unlock_irq(&rds_tcp_conn_lock); + rdsdebug("rds_conn_path [%d] tc %p\n", i, + conn->c_path[i].cp_transport_data); + } - rdsdebug("alloced tc %p\n", conn->c_transport_data); return 0; } @@ -330,7 +335,7 @@ static void rds_tcp_destroy_conns(void) spin_unlock_irq(&rds_tcp_conn_lock); list_for_each_entry_safe(tc, _tc, &tmp_list, t_tcp_node) - rds_conn_destroy(tc->conn); + rds_conn_destroy(tc->t_cpath->cp_conn); } static void rds_tcp_exit(void); @@ -498,7 +503,7 @@ static void rds_tcp_kill_sock(struct net *net) flush_work(&rtn->rds_tcp_accept_w); spin_lock_irq(&rds_tcp_conn_lock); list_for_each_entry_safe(tc, _tc, &rds_tcp_conn_list, t_tcp_node) { - struct net *c_net = read_pnet(&tc->conn->c_net); + struct net *c_net = read_pnet(&tc->t_cpath->cp_conn->c_net); if (net != c_net || !tc->t_sock) continue; @@ -509,7 +514,7 @@ static void rds_tcp_kill_sock(struct net *net) sk = tc->t_sock->sk; sk->sk_prot->disconnect(sk, 0); tcp_done(sk); - rds_conn_destroy(tc->conn); + rds_conn_destroy(tc->t_cpath->cp_conn); } } @@ -547,12 +552,13 @@ static void rds_tcp_sysctl_reset(struct net *net) spin_lock_irq(&rds_tcp_conn_lock); list_for_each_entry_safe(tc, _tc, &rds_tcp_conn_list, t_tcp_node) { - struct net *c_net = read_pnet(&tc->conn->c_net); + struct net *c_net = read_pnet(&tc->t_cpath->cp_conn->c_net); if (net != c_net || !tc->t_sock) continue; - rds_conn_drop(tc->conn); /* reconnect with new parameters */ + /* reconnect with new parameters */ + rds_conn_path_drop(tc->t_cpath); } spin_unlock_irq(&rds_tcp_conn_lock); } diff --git a/net/rds/tcp.h b/net/rds/tcp.h index 728abe2..e1ff169 100644 --- a/net/rds/tcp.h +++ b/net/rds/tcp.h @@ -11,11 +11,11 @@ struct rds_tcp_incoming { struct rds_tcp_connection { struct list_head t_tcp_node; - struct rds_connection *conn; - /* t_conn_lock synchronizes the connection establishment between + struct rds_conn_path *t_cpath; + /* t_conn_path_lock synchronizes the connection establishment between * rds_tcp_accept_one and rds_tcp_conn_connect */ - struct mutex t_conn_lock; + struct mutex t_conn_path_lock; struct socket *t_sock; void *t_orig_write_space; void *t_orig_data_ready; diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c index aa65c16..146692c 100644 --- a/net/rds/tcp_connect.c +++ b/net/rds/tcp_connect.c @@ -82,10 +82,10 @@ int rds_tcp_conn_connect(struct rds_connection *conn) int ret; struct rds_tcp_connection *tc = conn->c_transport_data; - mutex_lock(&tc->t_conn_lock); + mutex_lock(&tc->t_conn_path_lock); if (rds_conn_up(conn)) { - mutex_unlock(&tc->t_conn_lock); + mutex_unlock(&tc->t_conn_path_lock); return 0; } ret = sock_create_kern(rds_conn_net(conn), PF_INET, @@ -129,7 +129,7 @@ int rds_tcp_conn_connect(struct rds_connection *conn) } out: - mutex_unlock(&tc->t_conn_lock); + mutex_unlock(&tc->t_conn_path_lock); if (sock) sock_release(sock); return ret; diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index f9cc945..d893346 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -121,7 +121,7 @@ int rds_tcp_accept_one(struct socket *sock) */ rs_tcp = (struct rds_tcp_connection *)conn->c_transport_data; rds_conn_transition(conn, RDS_CONN_DOWN, RDS_CONN_CONNECTING); - mutex_lock(&rs_tcp->t_conn_lock); + mutex_lock(&rs_tcp->t_conn_path_lock); conn_state = rds_conn_state(conn); if (conn_state != RDS_CONN_CONNECTING && conn_state != RDS_CONN_UP) goto rst_nsk; @@ -156,7 +156,7 @@ rst_nsk: ret = 0; out: if (rs_tcp) - mutex_unlock(&rs_tcp->t_conn_lock); + mutex_unlock(&rs_tcp->t_conn_path_lock); if (new_sock) sock_release(new_sock); return ret; -- cgit v0.10.2 From afb4164d91c7486a1d4ab098a1b88e27b5e25772 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 30 Jun 2016 16:11:13 -0700 Subject: RDS: TCP: Refactor connection destruction to handle multiple paths A single rds_connection may have multiple rds_conn_paths that have to be carefully and correctly destroyed, for both rmmod and netns-delete cases. For both cases, we extract a single rds_tcp_connection for each conn into a temporary list, and then invoke rds_conn_destroy() which iteratively dismantles every path in the rds_connection. For the netns deletion case, we additionally have to make sure that we do not leave a socket in TIME_WAIT state, as this will hold up the netns deletion. Thus we call rds_tcp_conn_paths_destroy() to reset state quickly. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/tcp.c b/net/rds/tcp.c index c6b47f6..b327727 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -323,6 +323,17 @@ static void rds_tcp_conn_free(void *arg) kmem_cache_free(rds_tcp_conn_slab, tc); } +static bool list_has_conn(struct list_head *list, struct rds_connection *conn) +{ + struct rds_tcp_connection *tc, *_tc; + + list_for_each_entry_safe(tc, _tc, list, t_tcp_node) { + if (tc->t_cpath->cp_conn == conn) + return true; + } + return false; +} + static void rds_tcp_destroy_conns(void) { struct rds_tcp_connection *tc, *_tc; @@ -330,8 +341,10 @@ static void rds_tcp_destroy_conns(void) /* avoid calling conn_destroy with irqs off */ spin_lock_irq(&rds_tcp_conn_lock); - list_splice(&rds_tcp_conn_list, &tmp_list); - INIT_LIST_HEAD(&rds_tcp_conn_list); + list_for_each_entry_safe(tc, _tc, &rds_tcp_conn_list, t_tcp_node) { + if (!list_has_conn(&tmp_list, tc->t_cpath->cp_conn)) + list_move_tail(&tc->t_tcp_node, &tmp_list); + } spin_unlock_irq(&rds_tcp_conn_lock); list_for_each_entry_safe(tc, _tc, &tmp_list, t_tcp_node) @@ -491,10 +504,30 @@ static struct pernet_operations rds_tcp_net_ops = { .size = sizeof(struct rds_tcp_net), }; +/* explicitly send a RST on each socket, thereby releasing any socket refcnts + * that may otherwise hold up netns deletion. + */ +static void rds_tcp_conn_paths_destroy(struct rds_connection *conn) +{ + struct rds_conn_path *cp; + struct rds_tcp_connection *tc; + int i; + struct sock *sk; + + for (i = 0; i < RDS_MPATH_WORKERS; i++) { + cp = &conn->c_path[i]; + tc = cp->cp_transport_data; + if (!tc->t_sock) + continue; + sk = tc->t_sock->sk; + sk->sk_prot->disconnect(sk, 0); + tcp_done(sk); + } +} + static void rds_tcp_kill_sock(struct net *net) { struct rds_tcp_connection *tc, *_tc; - struct sock *sk; LIST_HEAD(tmp_list); struct rds_tcp_net *rtn = net_generic(net, rds_tcp_netid); @@ -507,13 +540,12 @@ static void rds_tcp_kill_sock(struct net *net) if (net != c_net || !tc->t_sock) continue; - list_move_tail(&tc->t_tcp_node, &tmp_list); + if (!list_has_conn(&tmp_list, tc->t_cpath->cp_conn)) + list_move_tail(&tc->t_tcp_node, &tmp_list); } spin_unlock_irq(&rds_tcp_conn_lock); list_for_each_entry_safe(tc, _tc, &tmp_list, t_tcp_node) { - sk = tc->t_sock->sk; - sk->sk_prot->disconnect(sk, 0); - tcp_done(sk); + rds_tcp_conn_paths_destroy(tc->t_cpath->cp_conn); rds_conn_destroy(tc->t_cpath->cp_conn); } } -- cgit v0.10.2 From ea3b1ea53930879c9847044f5cb9c97411cae797 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 30 Jun 2016 16:11:14 -0700 Subject: RDS: TCP: make ->sk_user_data point to a rds_conn_path The socket callbacks should all operate on a struct rds_conn_path, in preparation for a MP capable RDS-TCP. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/tcp.c b/net/rds/tcp.c index b327727..5658f3e 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -136,9 +136,9 @@ void rds_tcp_restore_callbacks(struct socket *sock, * from being called while it isn't set. */ void rds_tcp_reset_callbacks(struct socket *sock, - struct rds_connection *conn) + struct rds_conn_path *cp) { - struct rds_tcp_connection *tc = conn->c_transport_data; + struct rds_tcp_connection *tc = cp->cp_transport_data; struct socket *osock = tc->t_sock; if (!osock) @@ -148,8 +148,8 @@ void rds_tcp_reset_callbacks(struct socket *sock, * We have an outstanding SYN to this peer, which may * potentially have transitioned to the RDS_CONN_UP state, * so we must quiesce any send threads before resetting - * c_transport_data. We quiesce these threads by setting - * c_state to something other than RDS_CONN_UP, and then + * cp_transport_data. We quiesce these threads by setting + * cp_state to something other than RDS_CONN_UP, and then * waiting for any existing threads in rds_send_xmit to * complete release_in_xmit(). (Subsequent threads entering * rds_send_xmit() will bail on !rds_conn_up(). @@ -164,8 +164,8 @@ void rds_tcp_reset_callbacks(struct socket *sock, * RDS_CONN_RESETTTING, to ensure that rds_tcp_state_change * cannot mark rds_conn_path_up() in the window before lock_sock() */ - atomic_set(&conn->c_state, RDS_CONN_RESETTING); - wait_event(conn->c_waitq, !test_bit(RDS_IN_XMIT, &conn->c_flags)); + atomic_set(&cp->cp_state, RDS_CONN_RESETTING); + wait_event(cp->cp_waitq, !test_bit(RDS_IN_XMIT, &cp->cp_flags)); lock_sock(osock->sk); /* reset receive side state for rds_tcp_data_recv() for osock */ if (tc->t_tinc) { @@ -186,11 +186,12 @@ void rds_tcp_reset_callbacks(struct socket *sock, release_sock(osock->sk); sock_release(osock); newsock: - rds_send_path_reset(&conn->c_path[0]); + rds_send_path_reset(cp); lock_sock(sock->sk); write_lock_bh(&sock->sk->sk_callback_lock); tc->t_sock = sock; - sock->sk->sk_user_data = conn; + tc->t_cpath = cp; + sock->sk->sk_user_data = cp; sock->sk->sk_data_ready = rds_tcp_data_ready; sock->sk->sk_write_space = rds_tcp_write_space; sock->sk->sk_state_change = rds_tcp_state_change; @@ -203,9 +204,9 @@ newsock: * above rds_tcp_reset_callbacks for notes about synchronization * with data path */ -void rds_tcp_set_callbacks(struct socket *sock, struct rds_connection *conn) +void rds_tcp_set_callbacks(struct socket *sock, struct rds_conn_path *cp) { - struct rds_tcp_connection *tc = conn->c_transport_data; + struct rds_tcp_connection *tc = cp->cp_transport_data; rdsdebug("setting sock %p callbacks to tc %p\n", sock, tc); write_lock_bh(&sock->sk->sk_callback_lock); @@ -221,12 +222,12 @@ void rds_tcp_set_callbacks(struct socket *sock, struct rds_connection *conn) sock->sk->sk_data_ready = sock->sk->sk_user_data; tc->t_sock = sock; - tc->t_cpath = &conn->c_path[0]; + tc->t_cpath = cp; tc->t_orig_data_ready = sock->sk->sk_data_ready; tc->t_orig_write_space = sock->sk->sk_write_space; tc->t_orig_state_change = sock->sk->sk_state_change; - sock->sk->sk_user_data = conn; + sock->sk->sk_user_data = cp; sock->sk->sk_data_ready = rds_tcp_data_ready; sock->sk->sk_write_space = rds_tcp_write_space; sock->sk->sk_state_change = rds_tcp_state_change; diff --git a/net/rds/tcp.h b/net/rds/tcp.h index e1ff169..151b09d 100644 --- a/net/rds/tcp.h +++ b/net/rds/tcp.h @@ -49,8 +49,8 @@ struct rds_tcp_statistics { /* tcp.c */ void rds_tcp_tune(struct socket *sock); void rds_tcp_nonagle(struct socket *sock); -void rds_tcp_set_callbacks(struct socket *sock, struct rds_connection *conn); -void rds_tcp_reset_callbacks(struct socket *sock, struct rds_connection *conn); +void rds_tcp_set_callbacks(struct socket *sock, struct rds_conn_path *cp); +void rds_tcp_reset_callbacks(struct socket *sock, struct rds_conn_path *cp); void rds_tcp_restore_callbacks(struct socket *sock, struct rds_tcp_connection *tc); u32 rds_tcp_snd_nxt(struct rds_tcp_connection *tc); diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c index 146692c..7eddce5 100644 --- a/net/rds/tcp_connect.c +++ b/net/rds/tcp_connect.c @@ -41,16 +41,16 @@ void rds_tcp_state_change(struct sock *sk) { void (*state_change)(struct sock *sk); - struct rds_connection *conn; + struct rds_conn_path *cp; struct rds_tcp_connection *tc; read_lock_bh(&sk->sk_callback_lock); - conn = sk->sk_user_data; - if (!conn) { + cp = sk->sk_user_data; + if (!cp) { state_change = sk->sk_state_change; goto out; } - tc = conn->c_transport_data; + tc = cp->cp_transport_data; state_change = tc->t_orig_state_change; rdsdebug("sock %p state_change to %d\n", tc->t_sock, sk->sk_state); @@ -61,12 +61,11 @@ void rds_tcp_state_change(struct sock *sk) case TCP_SYN_RECV: break; case TCP_ESTABLISHED: - rds_connect_path_complete(&conn->c_path[0], - RDS_CONN_CONNECTING); + rds_connect_path_complete(cp, RDS_CONN_CONNECTING); break; case TCP_CLOSE_WAIT: case TCP_CLOSE: - rds_conn_drop(conn); + rds_conn_path_drop(cp); default: break; } @@ -81,6 +80,7 @@ int rds_tcp_conn_connect(struct rds_connection *conn) struct sockaddr_in src, dest; int ret; struct rds_tcp_connection *tc = conn->c_transport_data; + struct rds_conn_path *cp = &conn->c_path[0]; mutex_lock(&tc->t_conn_path_lock); @@ -114,7 +114,7 @@ int rds_tcp_conn_connect(struct rds_connection *conn) * once we call connect() we can start getting callbacks and they * own the socket */ - rds_tcp_set_callbacks(sock, conn); + rds_tcp_set_callbacks(sock, cp); ret = sock->ops->connect(sock, (struct sockaddr *)&dest, sizeof(dest), O_NONBLOCK); diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index d893346..ca975a2 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -79,6 +79,7 @@ int rds_tcp_accept_one(struct socket *sock) struct inet_sock *inet; struct rds_tcp_connection *rs_tcp = NULL; int conn_state; + struct rds_conn_path *cp; if (!sock) /* module unload or netns delete in progress */ return -ENETUNREACH; @@ -120,6 +121,7 @@ int rds_tcp_accept_one(struct socket *sock) * rds_tcp_state_change() will do that cleanup */ rs_tcp = (struct rds_tcp_connection *)conn->c_transport_data; + cp = &conn->c_path[0]; rds_conn_transition(conn, RDS_CONN_DOWN, RDS_CONN_CONNECTING); mutex_lock(&rs_tcp->t_conn_path_lock); conn_state = rds_conn_state(conn); @@ -136,16 +138,14 @@ int rds_tcp_accept_one(struct socket *sock) !conn->c_path[0].cp_outgoing) { goto rst_nsk; } else { - rds_tcp_reset_callbacks(new_sock, conn); + rds_tcp_reset_callbacks(new_sock, cp); conn->c_path[0].cp_outgoing = 0; /* rds_connect_path_complete() marks RDS_CONN_UP */ - rds_connect_path_complete(&conn->c_path[0], - RDS_CONN_RESETTING); + rds_connect_path_complete(cp, RDS_CONN_RESETTING); } } else { - rds_tcp_set_callbacks(new_sock, conn); - rds_connect_path_complete(&conn->c_path[0], - RDS_CONN_CONNECTING); + rds_tcp_set_callbacks(new_sock, cp); + rds_connect_path_complete(cp, RDS_CONN_CONNECTING); } new_sock = NULL; ret = 0; diff --git a/net/rds/tcp_recv.c b/net/rds/tcp_recv.c index 4a87d9e..aa7a79a 100644 --- a/net/rds/tcp_recv.c +++ b/net/rds/tcp_recv.c @@ -297,24 +297,24 @@ int rds_tcp_recv(struct rds_connection *conn) void rds_tcp_data_ready(struct sock *sk) { void (*ready)(struct sock *sk); - struct rds_connection *conn; + struct rds_conn_path *cp; struct rds_tcp_connection *tc; rdsdebug("data ready sk %p\n", sk); read_lock_bh(&sk->sk_callback_lock); - conn = sk->sk_user_data; - if (!conn) { /* check for teardown race */ + cp = sk->sk_user_data; + if (!cp) { /* check for teardown race */ ready = sk->sk_data_ready; goto out; } - tc = conn->c_transport_data; + tc = cp->cp_transport_data; ready = tc->t_orig_data_ready; rds_tcp_stats_inc(s_tcp_data_ready_calls); - if (rds_tcp_read_sock(conn, GFP_ATOMIC) == -ENOMEM) - queue_delayed_work(rds_wq, &conn->c_recv_w, 0); + if (rds_tcp_read_sock(cp->cp_conn, GFP_ATOMIC) == -ENOMEM) + queue_delayed_work(rds_wq, &cp->cp_recv_w, 0); out: read_unlock_bh(&sk->sk_callback_lock); ready(sk); diff --git a/net/rds/tcp_send.c b/net/rds/tcp_send.c index 52cda94..57e0f58 100644 --- a/net/rds/tcp_send.c +++ b/net/rds/tcp_send.c @@ -178,27 +178,27 @@ static int rds_tcp_is_acked(struct rds_message *rm, uint64_t ack) void rds_tcp_write_space(struct sock *sk) { void (*write_space)(struct sock *sk); - struct rds_connection *conn; + struct rds_conn_path *cp; struct rds_tcp_connection *tc; read_lock_bh(&sk->sk_callback_lock); - conn = sk->sk_user_data; - if (!conn) { + cp = sk->sk_user_data; + if (!cp) { write_space = sk->sk_write_space; goto out; } - tc = conn->c_transport_data; + tc = cp->cp_transport_data; rdsdebug("write_space for tc %p\n", tc); write_space = tc->t_orig_write_space; rds_tcp_stats_inc(s_tcp_write_space_calls); rdsdebug("tcp una %u\n", rds_tcp_snd_una(tc)); tc->t_last_seen_una = rds_tcp_snd_una(tc); - rds_send_drop_acked(conn, rds_tcp_snd_una(tc), rds_tcp_is_acked); + rds_send_path_drop_acked(cp, rds_tcp_snd_una(tc), rds_tcp_is_acked); if ((atomic_read(&sk->sk_wmem_alloc) << 1) <= sk->sk_sndbuf) - queue_delayed_work(rds_wq, &conn->c_send_w, 0); + queue_delayed_work(rds_wq, &cp->cp_send_w, 0); out: read_unlock_bh(&sk->sk_callback_lock); -- cgit v0.10.2 From 2da43c4a1b517d02e71d9611a2242273e7d399ba Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 30 Jun 2016 16:11:15 -0700 Subject: RDS: TCP: make receive path use the rds_conn_path The ->sk_user_data contains a pointer to the rds_conn_path for the socket. Use this consistently in the rds_tcp_data_ready callbacks to get the rds_conn_path for rds_recv_incoming. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/ib.c b/net/rds/ib.c index 1b29ec9..e6ba856 100644 --- a/net/rds/ib.c +++ b/net/rds/ib.c @@ -385,7 +385,7 @@ struct rds_transport rds_ib_transport = { .xmit = rds_ib_xmit, .xmit_rdma = rds_ib_xmit_rdma, .xmit_atomic = rds_ib_xmit_atomic, - .recv = rds_ib_recv, + .recv_path = rds_ib_recv_path, .conn_alloc = rds_ib_conn_alloc, .conn_free = rds_ib_conn_free, .conn_connect = rds_ib_conn_connect, diff --git a/net/rds/ib.h b/net/rds/ib.h index 2051f4b..579de7e 100644 --- a/net/rds/ib.h +++ b/net/rds/ib.h @@ -354,7 +354,7 @@ void rds_ib_mr_cqe_handler(struct rds_ib_connection *ic, struct ib_wc *wc); /* ib_recv.c */ int rds_ib_recv_init(void); void rds_ib_recv_exit(void); -int rds_ib_recv(struct rds_connection *conn); +int rds_ib_recv_path(struct rds_conn_path *conn); int rds_ib_recv_alloc_caches(struct rds_ib_connection *ic); void rds_ib_recv_free_caches(struct rds_ib_connection *ic); void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp); diff --git a/net/rds/ib_recv.c b/net/rds/ib_recv.c index 4ea8cb1..606a11f 100644 --- a/net/rds/ib_recv.c +++ b/net/rds/ib_recv.c @@ -1009,8 +1009,9 @@ void rds_ib_recv_cqe_handler(struct rds_ib_connection *ic, rds_ib_recv_refill(conn, 0, GFP_NOWAIT); } -int rds_ib_recv(struct rds_connection *conn) +int rds_ib_recv_path(struct rds_conn_path *cp) { + struct rds_connection *conn = cp->cp_conn; struct rds_ib_connection *ic = conn->c_transport_data; int ret = 0; diff --git a/net/rds/loop.c b/net/rds/loop.c index 318c21d..20284a4 100644 --- a/net/rds/loop.c +++ b/net/rds/loop.c @@ -102,7 +102,7 @@ static void rds_loop_inc_free(struct rds_incoming *inc) } /* we need to at least give the thread something to succeed */ -static int rds_loop_recv(struct rds_connection *conn) +static int rds_loop_recv_path(struct rds_conn_path *cp) { return 0; } @@ -185,7 +185,7 @@ void rds_loop_exit(void) */ struct rds_transport rds_loop_transport = { .xmit = rds_loop_xmit, - .recv = rds_loop_recv, + .recv_path = rds_loop_recv_path, .conn_alloc = rds_loop_conn_alloc, .conn_free = rds_loop_conn_free, .conn_connect = rds_loop_conn_connect, diff --git a/net/rds/rds.h b/net/rds/rds.h index 5bbad08..0faca30 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -462,7 +462,7 @@ struct rds_transport { unsigned int hdr_off, unsigned int sg, unsigned int off); int (*xmit_rdma)(struct rds_connection *conn, struct rm_rdma_op *op); int (*xmit_atomic)(struct rds_connection *conn, struct rm_atomic_op *op); - int (*recv)(struct rds_connection *conn); + int (*recv_path)(struct rds_conn_path *cp); int (*inc_copy_to_user)(struct rds_incoming *inc, struct iov_iter *to); void (*inc_free)(struct rds_incoming *inc); diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 5658f3e..7bc136c 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -359,7 +359,7 @@ struct rds_transport rds_tcp_transport = { .xmit_path_prepare = rds_tcp_xmit_path_prepare, .xmit_path_complete = rds_tcp_xmit_path_complete, .xmit = rds_tcp_xmit, - .recv = rds_tcp_recv, + .recv_path = rds_tcp_recv_path, .conn_alloc = rds_tcp_conn_alloc, .conn_free = rds_tcp_conn_free, .conn_connect = rds_tcp_conn_connect, diff --git a/net/rds/tcp.h b/net/rds/tcp.h index 151b09d..5a5f91a 100644 --- a/net/rds/tcp.h +++ b/net/rds/tcp.h @@ -75,7 +75,7 @@ int rds_tcp_keepalive(struct socket *sock); int rds_tcp_recv_init(void); void rds_tcp_recv_exit(void); void rds_tcp_data_ready(struct sock *sk); -int rds_tcp_recv(struct rds_connection *conn); +int rds_tcp_recv_path(struct rds_conn_path *cp); void rds_tcp_inc_free(struct rds_incoming *inc); int rds_tcp_inc_copy_to_user(struct rds_incoming *inc, struct iov_iter *to); diff --git a/net/rds/tcp_recv.c b/net/rds/tcp_recv.c index aa7a79a..ad4892e 100644 --- a/net/rds/tcp_recv.c +++ b/net/rds/tcp_recv.c @@ -34,7 +34,6 @@ #include #include -#include "rds_single_path.h" #include "rds.h" #include "tcp.h" @@ -148,7 +147,7 @@ static void rds_tcp_cong_recv(struct rds_connection *conn, } struct rds_tcp_desc_arg { - struct rds_connection *conn; + struct rds_conn_path *conn_path; gfp_t gfp; }; @@ -156,8 +155,8 @@ static int rds_tcp_data_recv(read_descriptor_t *desc, struct sk_buff *skb, unsigned int offset, size_t len) { struct rds_tcp_desc_arg *arg = desc->arg.data; - struct rds_connection *conn = arg->conn; - struct rds_tcp_connection *tc = conn->c_transport_data; + struct rds_conn_path *cp = arg->conn_path; + struct rds_tcp_connection *tc = cp->cp_transport_data; struct rds_tcp_incoming *tinc = tc->t_tinc; struct sk_buff *clone; size_t left = len, to_copy; @@ -179,7 +178,8 @@ static int rds_tcp_data_recv(read_descriptor_t *desc, struct sk_buff *skb, } tc->t_tinc = tinc; rdsdebug("alloced tinc %p\n", tinc); - rds_inc_init(&tinc->ti_inc, conn, conn->c_faddr); + rds_inc_path_init(&tinc->ti_inc, cp, + cp->cp_conn->c_faddr); /* * XXX * we might be able to use the __ variants when * we've already serialized at a higher level. @@ -229,6 +229,8 @@ static int rds_tcp_data_recv(read_descriptor_t *desc, struct sk_buff *skb, } if (tc->t_tinc_hdr_rem == 0 && tc->t_tinc_data_rem == 0) { + struct rds_connection *conn = cp->cp_conn; + if (tinc->ti_inc.i_hdr.h_flags == RDS_FLAG_CONG_BITMAP) rds_tcp_cong_recv(conn, tinc); else @@ -251,15 +253,15 @@ out: } /* the caller has to hold the sock lock */ -static int rds_tcp_read_sock(struct rds_connection *conn, gfp_t gfp) +static int rds_tcp_read_sock(struct rds_conn_path *cp, gfp_t gfp) { - struct rds_tcp_connection *tc = conn->c_transport_data; + struct rds_tcp_connection *tc = cp->cp_transport_data; struct socket *sock = tc->t_sock; read_descriptor_t desc; struct rds_tcp_desc_arg arg; /* It's like glib in the kernel! */ - arg.conn = conn; + arg.conn_path = cp; arg.gfp = gfp; desc.arg.data = &arg; desc.error = 0; @@ -279,16 +281,17 @@ static int rds_tcp_read_sock(struct rds_connection *conn, gfp_t gfp) * if we fail to allocate we're in trouble.. blindly wait some time before * trying again to see if the VM can free up something for us. */ -int rds_tcp_recv(struct rds_connection *conn) +int rds_tcp_recv_path(struct rds_conn_path *cp) { - struct rds_tcp_connection *tc = conn->c_transport_data; + struct rds_tcp_connection *tc = cp->cp_transport_data; struct socket *sock = tc->t_sock; int ret = 0; - rdsdebug("recv worker conn %p tc %p sock %p\n", conn, tc, sock); + rdsdebug("recv worker path [%d] tc %p sock %p\n", + cp->cp_index, tc, sock); lock_sock(sock->sk); - ret = rds_tcp_read_sock(conn, GFP_KERNEL); + ret = rds_tcp_read_sock(cp, GFP_KERNEL); release_sock(sock->sk); return ret; @@ -313,7 +316,7 @@ void rds_tcp_data_ready(struct sock *sk) ready = tc->t_orig_data_ready; rds_tcp_stats_inc(s_tcp_data_ready_calls); - if (rds_tcp_read_sock(cp->cp_conn, GFP_ATOMIC) == -ENOMEM) + if (rds_tcp_read_sock(cp, GFP_ATOMIC) == -ENOMEM) queue_delayed_work(rds_wq, &cp->cp_recv_w, 0); out: read_unlock_bh(&sk->sk_callback_lock); diff --git a/net/rds/threads.c b/net/rds/threads.c index 9fbe95b..f717b69 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -203,7 +203,7 @@ void rds_recv_worker(struct work_struct *work) int ret; if (rds_conn_path_state(cp) == RDS_CONN_UP) { - ret = cp->cp_conn->c_trans->recv(cp->cp_conn); + ret = cp->cp_conn->c_trans->recv_path(cp); rdsdebug("conn %p ret %d\n", cp->cp_conn, ret); switch (ret) { case -EAGAIN: -- cgit v0.10.2 From b04e8554f7637999af8f54cca4dcfcf49f2ae7c8 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 30 Jun 2016 16:11:16 -0700 Subject: RDS: TCP: Hooks to set up a single connection path This patch adds ->conn_path_connect callbacks in the rds_transport that are used to set up a single connection path. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/ib.c b/net/rds/ib.c index e6ba856..7eaf887 100644 --- a/net/rds/ib.c +++ b/net/rds/ib.c @@ -388,7 +388,7 @@ struct rds_transport rds_ib_transport = { .recv_path = rds_ib_recv_path, .conn_alloc = rds_ib_conn_alloc, .conn_free = rds_ib_conn_free, - .conn_connect = rds_ib_conn_connect, + .conn_path_connect = rds_ib_conn_path_connect, .conn_path_shutdown = rds_ib_conn_path_shutdown, .inc_copy_to_user = rds_ib_inc_copy_to_user, .inc_free = rds_ib_inc_free, diff --git a/net/rds/ib.h b/net/rds/ib.h index 579de7e..046f750 100644 --- a/net/rds/ib.h +++ b/net/rds/ib.h @@ -328,7 +328,7 @@ extern struct list_head ib_nodev_conns; /* ib_cm.c */ int rds_ib_conn_alloc(struct rds_connection *conn, gfp_t gfp); void rds_ib_conn_free(void *arg); -int rds_ib_conn_connect(struct rds_connection *conn); +int rds_ib_conn_path_connect(struct rds_conn_path *cp); void rds_ib_conn_path_shutdown(struct rds_conn_path *cp); void rds_ib_state_change(struct sock *sk); int rds_ib_listen_init(void); diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c index e34ea0b..5b2ab95 100644 --- a/net/rds/ib_cm.c +++ b/net/rds/ib_cm.c @@ -685,8 +685,9 @@ out: return ret; } -int rds_ib_conn_connect(struct rds_connection *conn) +int rds_ib_conn_path_connect(struct rds_conn_path *cp) { + struct rds_connection *conn = cp->cp_conn; struct rds_ib_connection *ic = conn->c_transport_data; struct sockaddr_in src, dest; int ret; diff --git a/net/rds/loop.c b/net/rds/loop.c index 20284a4..f2bf78d 100644 --- a/net/rds/loop.c +++ b/net/rds/loop.c @@ -150,9 +150,9 @@ static void rds_loop_conn_free(void *arg) kfree(lc); } -static int rds_loop_conn_connect(struct rds_connection *conn) +static int rds_loop_conn_path_connect(struct rds_conn_path *cp) { - rds_connect_complete(conn); + rds_connect_complete(cp->cp_conn); return 0; } @@ -188,7 +188,7 @@ struct rds_transport rds_loop_transport = { .recv_path = rds_loop_recv_path, .conn_alloc = rds_loop_conn_alloc, .conn_free = rds_loop_conn_free, - .conn_connect = rds_loop_conn_connect, + .conn_path_connect = rds_loop_conn_path_connect, .conn_path_shutdown = rds_loop_conn_path_shutdown, .inc_copy_to_user = rds_message_inc_copy_to_user, .inc_free = rds_loop_inc_free, diff --git a/net/rds/rds.h b/net/rds/rds.h index 0faca30..6ef07bd 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -454,7 +454,7 @@ struct rds_transport { int (*laddr_check)(struct net *net, __be32 addr); int (*conn_alloc)(struct rds_connection *conn, gfp_t gfp); void (*conn_free)(void *data); - int (*conn_connect)(struct rds_connection *conn); + int (*conn_path_connect)(struct rds_conn_path *cp); void (*conn_path_shutdown)(struct rds_conn_path *conn); void (*xmit_path_prepare)(struct rds_conn_path *cp); void (*xmit_path_complete)(struct rds_conn_path *cp); diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 7bc136c..d278432 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -362,7 +362,7 @@ struct rds_transport rds_tcp_transport = { .recv_path = rds_tcp_recv_path, .conn_alloc = rds_tcp_conn_alloc, .conn_free = rds_tcp_conn_free, - .conn_connect = rds_tcp_conn_connect, + .conn_path_connect = rds_tcp_conn_path_connect, .conn_path_shutdown = rds_tcp_conn_path_shutdown, .inc_copy_to_user = rds_tcp_inc_copy_to_user, .inc_free = rds_tcp_inc_free, diff --git a/net/rds/tcp.h b/net/rds/tcp.h index 5a5f91a..1c3160f 100644 --- a/net/rds/tcp.h +++ b/net/rds/tcp.h @@ -13,7 +13,7 @@ struct rds_tcp_connection { struct list_head t_tcp_node; struct rds_conn_path *t_cpath; /* t_conn_path_lock synchronizes the connection establishment between - * rds_tcp_accept_one and rds_tcp_conn_connect + * rds_tcp_accept_one and rds_tcp_conn_path_connect */ struct mutex t_conn_path_lock; struct socket *t_sock; @@ -60,7 +60,7 @@ extern struct rds_transport rds_tcp_transport; void rds_tcp_accept_work(struct sock *sk); /* tcp_connect.c */ -int rds_tcp_conn_connect(struct rds_connection *conn); +int rds_tcp_conn_path_connect(struct rds_conn_path *cp); void rds_tcp_conn_path_shutdown(struct rds_conn_path *conn); void rds_tcp_state_change(struct sock *sk); diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c index 7eddce5..c916715 100644 --- a/net/rds/tcp_connect.c +++ b/net/rds/tcp_connect.c @@ -74,17 +74,17 @@ out: state_change(sk); } -int rds_tcp_conn_connect(struct rds_connection *conn) +int rds_tcp_conn_path_connect(struct rds_conn_path *cp) { struct socket *sock = NULL; struct sockaddr_in src, dest; int ret; - struct rds_tcp_connection *tc = conn->c_transport_data; - struct rds_conn_path *cp = &conn->c_path[0]; + struct rds_connection *conn = cp->cp_conn; + struct rds_tcp_connection *tc = cp->cp_transport_data; mutex_lock(&tc->t_conn_path_lock); - if (rds_conn_up(conn)) { + if (rds_conn_path_up(cp)) { mutex_unlock(&tc->t_conn_path_lock); return 0; } @@ -118,6 +118,7 @@ int rds_tcp_conn_connect(struct rds_connection *conn) ret = sock->ops->connect(sock, (struct sockaddr *)&dest, sizeof(dest), O_NONBLOCK); + cp->cp_outgoing = 1; rdsdebug("connect to address %pI4 returned %d\n", &conn->c_faddr, ret); if (ret == -EINPROGRESS) ret = 0; @@ -125,7 +126,7 @@ int rds_tcp_conn_connect(struct rds_connection *conn) rds_tcp_keepalive(sock); sock = NULL; } else { - rds_tcp_restore_callbacks(sock, conn->c_transport_data); + rds_tcp_restore_callbacks(sock, cp->cp_transport_data); } out: diff --git a/net/rds/threads.c b/net/rds/threads.c index f717b69..e8f0941 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -152,8 +152,9 @@ void rds_connect_worker(struct work_struct *work) int ret; clear_bit(RDS_RECONNECT_PENDING, &cp->cp_flags); - if (rds_conn_path_transition(cp, RDS_CONN_DOWN, RDS_CONN_CONNECTING)) { - ret = conn->c_trans->conn_connect(conn); + ret = rds_conn_path_transition(cp, RDS_CONN_DOWN, RDS_CONN_CONNECTING); + if (ret) { + ret = conn->c_trans->conn_path_connect(cp); rdsdebug("conn %p for %pI4 to %pI4 dispatched, ret %d\n", conn, &conn->c_laddr, &conn->c_faddr, ret); -- cgit v0.10.2 From 8315011ad67670691545ed394968435f0a0bb29e Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 30 Jun 2016 16:11:17 -0700 Subject: RDS: TCP: Simplify reconnect to avoid duelling reconnnect attempts When reconnecting, the peer with the smaller IP address will initiate the reconnect, to avoid needless duelling SYN issues. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/connection.c b/net/rds/connection.c index 1b0c2a78..19a4fee 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -355,9 +355,7 @@ void rds_conn_shutdown(struct rds_conn_path *cp) rcu_read_lock(); if (!hlist_unhashed(&conn->c_hash_node)) { rcu_read_unlock(); - if (conn->c_trans->t_type != RDS_TRANS_TCP || - cp->cp_outgoing == 1) - rds_queue_reconnect(cp); + rds_queue_reconnect(cp); } else { rcu_read_unlock(); } diff --git a/net/rds/threads.c b/net/rds/threads.c index e8f0941..bc97d67 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -125,6 +125,11 @@ void rds_queue_reconnect(struct rds_conn_path *cp) conn, &conn->c_laddr, &conn->c_faddr, cp->cp_reconnect_jiffies); + /* let peer with smaller addr initiate reconnect, to avoid duels */ + if (conn->c_trans->t_type == RDS_TRANS_TCP && + conn->c_laddr > conn->c_faddr) + return; + set_bit(RDS_RECONNECT_PENDING, &cp->cp_flags); if (cp->cp_reconnect_jiffies == 0) { cp->cp_reconnect_jiffies = rds_sysctl_reconnect_min_jiffies; -- cgit v0.10.2 From 11bb62f7c05240a933dd2e6b3bf3871d99464524 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 30 Jun 2016 16:11:18 -0700 Subject: RDS: Do not send a pong to an incoming ping with 0 src port RDS ping messages are sent with a non-zero src port to a zero dst port, so that the rds pong messages can be sent back to the originators src port. However if a confused/malicious sender sends a ping with a 0 src port, we'd have an infinite ping-pong loop. To avoid this, the receiver should ignore ping messages with a 0 src port. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/recv.c b/net/rds/recv.c index b58f505..fed53a6 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -226,6 +226,10 @@ void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr, cp->cp_next_rx_seq = be64_to_cpu(inc->i_hdr.h_sequence) + 1; if (rds_sysctl_ping_enable && inc->i_hdr.h_dport == 0) { + if (inc->i_hdr.h_sport == 0) { + rdsdebug("ignore ping with 0 sport from 0x%x\n", saddr); + goto out; + } rds_stats_inc(s_recv_ping); rds_send_pong(cp, inc->i_hdr.h_sport); goto out; -- cgit v0.10.2 From b30d74e42d780b2c5bc640651df61412551d628e Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Fri, 1 Jul 2016 17:34:05 +0800 Subject: MAINTAINERS: add maintainers for hns driver This patch adds maintainers for hisilicon network subsystem driver Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/MAINTAINERS b/MAINTAINERS index d8c0784..772c9ff 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5437,6 +5437,15 @@ F: include/uapi/linux/if_hippi.h F: net/802/hippi.c F: drivers/net/hippi/ +HISILICON NETWORK SUBSYSTEM DRIVER +M: Yisen Zhuang +M: Salil Mehta +L: netdev@vger.kernel.org +W: http://www.hisilicon.com +S: Maintained +F: drivers/net/ethernet/hisilicon/ +F: Documentation/devicetree/bindings/net/hisilicon*.txt + HISILICON SAS Controller M: John Garry W: http://www.hisilicon.com -- cgit v0.10.2 From d9fdb4ed00036514146497fd48b906f0e12b68f7 Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Fri, 1 Jul 2016 17:34:06 +0800 Subject: net: hns: fix code style about hns driver This patch fixes code sytle of hns driver to make it simple. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c index 8473287..611b67b 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c @@ -253,10 +253,9 @@ static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, reg_val_1 = 0x1 << port; port_rst_off = dsaf_dev->mac_cb[port]->port_rst_off; /* there is difference between V1 and V2 in register.*/ - if (AE_IS_VER1(dsaf_dev->dsaf_ver)) - reg_val_2 = 0x1041041 << port_rst_off; - else - reg_val_2 = 0x2082082 << port_rst_off; + reg_val_2 = AE_IS_VER1(dsaf_dev->dsaf_ver) ? + 0x1041041 : 0x2082082; + reg_val_2 <<= port_rst_off; if (!dereset) { dsaf_write_sub(dsaf_dev, DSAF_SUB_SC_GE_RESET_REQ1_REG, @@ -272,12 +271,11 @@ static void hns_dsaf_ge_srst_by_port(struct dsaf_device *dsaf_dev, u32 port, reg_val_1); } } else { - reg_val_1 = 0x15540 << dsaf_dev->reset_offset; + reg_val_1 = 0x15540; + reg_val_2 = AE_IS_VER1(dsaf_dev->dsaf_ver) ? 0x100 : 0x40; - if (AE_IS_VER1(dsaf_dev->dsaf_ver)) - reg_val_2 = 0x100 << dsaf_dev->reset_offset; - else - reg_val_2 = 0x40 << dsaf_dev->reset_offset; + reg_val_1 <<= dsaf_dev->reset_offset; + reg_val_2 <<= dsaf_dev->reset_offset; if (!dereset) { dsaf_write_sub(dsaf_dev, DSAF_SUB_SC_GE_RESET_REQ1_REG, -- cgit v0.10.2 From 8ec98ba711ed26f9ecf1820f28cb14af84f19d43 Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Fri, 1 Jul 2016 17:34:07 +0800 Subject: net: hns: change code style from a = a + x to a += x This patch fixes the code style in hns driver. Change it from "buff = buff + xxx" to "buff += xxx". The reveiw comments is from andy. Reviewed-by: Andriy Shevchenko Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index 67e8e13..b9d01ea 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -2540,45 +2540,45 @@ static char *hns_dsaf_get_node_stats_strings(char *data, int node, bool is_ver1 = AE_IS_VER1(dsaf_dev->dsaf_ver); snprintf(buff, ETH_GSTRING_LEN, "innod%d_pad_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_manage_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_rx_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_rx_pkt_id", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_rx_pause_frame", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_release_buf_num", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_sbm_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_crc_false_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_bp_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_lookup_rslt_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_local_rslt_fail_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_vlan_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_stp_drop_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; if ((node < DSAF_SERVICE_NW_NUM) && (!is_ver1)) { for (i = 0; i < DSAF_PRIO_NR; i++) { snprintf(buff, ETH_GSTRING_LEN, "inod%d_pfc_prio%d_pkts", node, i); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; } for (i = 0; i < DSAF_PRIO_NR; i++) { snprintf(buff, ETH_GSTRING_LEN, "onod%d_pfc_prio%d_pkts", node, i); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; } } snprintf(buff, ETH_GSTRING_LEN, "onnod%d_tx_pkts", node); - buff = buff + ETH_GSTRING_LEN; + buff += ETH_GSTRING_LEN; return buff; } -- cgit v0.10.2 From 68fa1636a4d0a206a4e80a4f2adc920bf102abc9 Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Fri, 1 Jul 2016 17:34:08 +0800 Subject: net: hns: delete redundant parenthese According to the previous review comments from Andy, this patch deletes the redundant parens in the patch. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index b9d01ea..e36ee22 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -2565,7 +2565,7 @@ static char *hns_dsaf_get_node_stats_strings(char *data, int node, buff += ETH_GSTRING_LEN; snprintf(buff, ETH_GSTRING_LEN, "innod%d_stp_drop_pkts", node); buff += ETH_GSTRING_LEN; - if ((node < DSAF_SERVICE_NW_NUM) && (!is_ver1)) { + if (node < DSAF_SERVICE_NW_NUM && !is_ver1) { for (i = 0; i < DSAF_PRIO_NR; i++) { snprintf(buff, ETH_GSTRING_LEN, "inod%d_pfc_prio%d_pkts", node, i); @@ -2604,7 +2604,7 @@ static u64 *hns_dsaf_get_node_stats(struct dsaf_device *ddev, u64 *data, p[10] = hw_stats->local_addr_false; p[11] = hw_stats->vlan_drop; p[12] = hw_stats->stp_drop; - if ((node_num < DSAF_SERVICE_NW_NUM) && (!is_ver1)) { + if (node_num < DSAF_SERVICE_NW_NUM && !is_ver1) { for (i = 0; i < DSAF_PRIO_NR; i++) { p[13 + i] = hw_stats->rx_pfc[i]; p[13 + i + DSAF_PRIO_NR] = hw_stats->tx_pfc[i]; -- cgit v0.10.2 From 6ba312eb1ce696ad8ac8c9d6e9e22663631ac740 Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Fri, 1 Jul 2016 17:34:09 +0800 Subject: net: hns: add a space before "*/" In comment line, some time miss a space before */, so this patch adds a space before */. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c index d5297ec..d7e1f8c 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c @@ -762,13 +762,13 @@ static int hns_nic_rx_poll_one(struct hns_nic_ring_data *ring_data, recv_pkts = 0, recv_bds = 0, clean_count = 0; recv: while (recv_pkts < budget && recv_bds < num) { - /* reuse or realloc buffers*/ + /* reuse or realloc buffers */ if (clean_count >= RCB_NOF_ALLOC_RX_BUFF_ONCE) { hns_nic_alloc_rx_buffers(ring_data, clean_count); clean_count = 0; } - /* poll one pkt*/ + /* poll one pkt */ err = hns_nic_poll_rx_skb(ring_data, &skb, &bnum); if (unlikely(!skb)) /* this fault cannot be repaired */ goto out; -- cgit v0.10.2 From 45fc764e3eecba6962b28a49a7c00ee4f5efff43 Mon Sep 17 00:00:00 2001 From: Daode Huang Date: Fri, 1 Jul 2016 17:34:10 +0800 Subject: net: hns: normalize two different loop There are two approaches to assign data, one does 2 loops, another does 1 loop. This patch normalize the different methods to 1 loop. Signed-off-by: Daode Huang Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index e36ee22..86ce28a 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -2567,15 +2567,15 @@ static char *hns_dsaf_get_node_stats_strings(char *data, int node, buff += ETH_GSTRING_LEN; if (node < DSAF_SERVICE_NW_NUM && !is_ver1) { for (i = 0; i < DSAF_PRIO_NR; i++) { - snprintf(buff, ETH_GSTRING_LEN, - "inod%d_pfc_prio%d_pkts", node, i); - buff += ETH_GSTRING_LEN; - } - for (i = 0; i < DSAF_PRIO_NR; i++) { - snprintf(buff, ETH_GSTRING_LEN, - "onod%d_pfc_prio%d_pkts", node, i); + snprintf(buff + 0 * ETH_GSTRING_LEN * DSAF_PRIO_NR, + ETH_GSTRING_LEN, "inod%d_pfc_prio%d_pkts", + node, i); + snprintf(buff + 1 * ETH_GSTRING_LEN * DSAF_PRIO_NR, + ETH_GSTRING_LEN, "onod%d_pfc_prio%d_pkts", + node, i); buff += ETH_GSTRING_LEN; } + buff += 1 * DSAF_PRIO_NR * ETH_GSTRING_LEN; } snprintf(buff, ETH_GSTRING_LEN, "onnod%d_tx_pkts", node); buff += ETH_GSTRING_LEN; @@ -2606,8 +2606,8 @@ static u64 *hns_dsaf_get_node_stats(struct dsaf_device *ddev, u64 *data, p[12] = hw_stats->stp_drop; if (node_num < DSAF_SERVICE_NW_NUM && !is_ver1) { for (i = 0; i < DSAF_PRIO_NR; i++) { - p[13 + i] = hw_stats->rx_pfc[i]; - p[13 + i + DSAF_PRIO_NR] = hw_stats->tx_pfc[i]; + p[13 + i + 0 * DSAF_PRIO_NR] = hw_stats->rx_pfc[i]; + p[13 + i + 1 * DSAF_PRIO_NR] = hw_stats->tx_pfc[i]; } p[29] = hw_stats->tx_pkts; return &p[30]; -- cgit v0.10.2 From 2e14b218f5f702520a99193fd7dea9888d734804 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 1 Jul 2016 17:34:11 +0800 Subject: net: hns: remove redundant hns_mac_dev_to_enet_if() The sequence of hns_mac_dev_to_enet_if() is the same as hns_get_enet_interface(), and hns_get_enet_interface() is called by initialization to get the mac mode. And the mode is not changed anywhere. Thus add hns_mac_dev_to_enet_if() function to get the mac mode is obviously redundant. Reported-by: Jinchuan Tian Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c index c526558..d2effcc 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c @@ -56,20 +56,6 @@ static const enum mac_mode g_mac_mode_1000[] = { [PHY_INTERFACE_MODE_RTBI] = MAC_MODE_RTBI_1000 }; -static enum mac_mode hns_mac_dev_to_enet_if(const struct hns_mac_cb *mac_cb) -{ - switch (mac_cb->max_speed) { - case MAC_SPEED_100: - return g_mac_mode_100[mac_cb->phy_if]; - case MAC_SPEED_1000: - return g_mac_mode_1000[mac_cb->phy_if]; - case MAC_SPEED_10000: - return MAC_MODE_XGMII_10000; - default: - return MAC_MODE_MII_100; - } -} - static enum mac_mode hns_get_enet_interface(const struct hns_mac_cb *mac_cb) { switch (mac_cb->max_speed) { @@ -134,7 +120,6 @@ void hns_mac_adjust_link(struct hns_mac_cb *mac_cb, int speed, int duplex) mac_cb->speed = speed; mac_cb->half_duplex = !duplex; - mac_ctrl_drv->mac_mode = hns_mac_dev_to_enet_if(mac_cb); if (mac_ctrl_drv->adjust_link) { ret = mac_ctrl_drv->adjust_link(mac_ctrl_drv, -- cgit v0.10.2 From 5d2525f7b8a7df810c3fbc548a91ba6e3cde578a Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 1 Jul 2016 17:34:12 +0800 Subject: net: hns: add media-type property for hns It is PORT_TP type if the service port is GE mode. It is wrong to judge the port type by using if it is service port. Adding the media type to know port type. Reported-by: Jinchuan Tian Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.h b/drivers/net/ethernet/hisilicon/hns/hnae.h index 3869322..e093cbf 100644 --- a/drivers/net/ethernet/hisilicon/hns/hnae.h +++ b/drivers/net/ethernet/hisilicon/hns/hnae.h @@ -363,6 +363,14 @@ enum hnae_port_type { HNAE_PORT_DEBUG }; +/* mac media type */ +enum hnae_media_type { + HNAE_MEDIA_TYPE_UNKNOWN = 0, + HNAE_MEDIA_TYPE_FIBER, + HNAE_MEDIA_TYPE_COPPER, + HNAE_MEDIA_TYPE_BACKPLANE, +}; + /* This struct defines the operation on the handle. * * get_handle(): (mandatory) @@ -525,6 +533,7 @@ struct hnae_handle { u32 eport_id; u32 dport_id; /* v2 tx bd should fill the dport_id */ enum hnae_port_type port_type; + enum hnae_media_type media_type; struct list_head node; /* list to hnae_ae_dev->handle_list */ struct hnae_buf_ops *bops; /* operation for the buffer */ struct hnae_queue **qs; /* array base of all queues */ diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c index 835521b..e28d960 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c @@ -134,6 +134,7 @@ struct hnae_handle *hns_ae_get_handle(struct hnae_ae_dev *dev, ae_handle->phy_dev = vf_cb->mac_cb->phy_dev; ae_handle->if_support = vf_cb->mac_cb->if_support; ae_handle->port_type = vf_cb->mac_cb->mac_type; + ae_handle->media_type = vf_cb->mac_cb->media_type; ae_handle->dport_id = port_id; return ae_handle; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c index d2effcc..3fb87e2 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c @@ -733,6 +733,18 @@ static void hns_mac_register_phy(struct hns_mac_cb *mac_cb) mac_cb->mac_id, addr); } +#define MAC_MEDIA_TYPE_MAX_LEN 16 + +static const struct { + enum hnae_media_type value; + const char *name; +} media_type_defs[] = { + {HNAE_MEDIA_TYPE_UNKNOWN, "unknown" }, + {HNAE_MEDIA_TYPE_FIBER, "fiber" }, + {HNAE_MEDIA_TYPE_COPPER, "copper" }, + {HNAE_MEDIA_TYPE_BACKPLANE, "backplane" }, +}; + /** *hns_mac_get_info - get mac information from device node *@mac_cb: mac device @@ -744,10 +756,13 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) struct device_node *np; struct regmap *syscon; struct of_phandle_args cpld_args; + const char *media_type; + u32 i; u32 ret; mac_cb->link = false; mac_cb->half_duplex = false; + mac_cb->media_type = HNAE_MEDIA_TYPE_UNKNOWN; mac_cb->speed = mac_phy_to_speed[mac_cb->phy_if]; mac_cb->max_speed = mac_cb->speed; @@ -849,6 +864,17 @@ static int hns_mac_get_info(struct hns_mac_cb *mac_cb) mac_cb->mac_id); } + if (!fwnode_property_read_string(mac_cb->fw_port, "media-type", + &media_type)) { + for (i = 0; i < ARRAY_SIZE(media_type_defs); i++) { + if (!strncmp(media_type_defs[i].name, media_type, + MAC_MEDIA_TYPE_MAX_LEN)) { + mac_cb->media_type = media_type_defs[i].value; + break; + } + } + } + return 0; } diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h index 05a6e8f..4cbdf14 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.h @@ -335,6 +335,7 @@ struct hns_mac_cb { u64 txpkt_for_led; u64 rxpkt_for_led; enum hnae_port_type mac_type; + enum hnae_media_type media_type; phy_interface_t phy_if; enum hnae_loop loop_mode; diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c index a395ca1..ab33487 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c @@ -165,13 +165,21 @@ static int hns_nic_get_settings(struct net_device *net_dev, cmd->advertising |= ADVERTISED_10000baseKR_Full; } - if (h->port_type == HNAE_PORT_SERVICE) { + switch (h->media_type) { + case HNAE_MEDIA_TYPE_FIBER: cmd->port = PORT_FIBRE; - cmd->supported |= SUPPORTED_Pause; - } else { + break; + case HNAE_MEDIA_TYPE_COPPER: cmd->port = PORT_TP; + break; + case HNAE_MEDIA_TYPE_UNKNOWN: + default: + break; } + if (!(AE_IS_VER1(priv->enet_ver) && h->port_type == HNAE_PORT_DEBUG)) + cmd->supported |= SUPPORTED_Pause; + cmd->transceiver = XCVR_EXTERNAL; cmd->mdio_support = (ETH_MDIO_SUPPORTS_C45 | ETH_MDIO_SUPPORTS_C22); hns_get_mdix_mode(net_dev, cmd); -- cgit v0.10.2 From b15dc29264d0bdad603d6b5956a4847443790078 Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 1 Jul 2016 17:34:13 +0800 Subject: net: hns: get reset registers from DT Since the registers of subctrl may be different, it is better to mv the registers from hns mdio driver routine to device tree node. Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns_mdio.c b/drivers/net/ethernet/hisilicon/hns_mdio.c index 761a32f..33f4c48 100644 --- a/drivers/net/ethernet/hisilicon/hns_mdio.c +++ b/drivers/net/ethernet/hisilicon/hns_mdio.c @@ -37,9 +37,19 @@ #define MDIO_TIMEOUT 1000000 +struct hns_mdio_sc_reg { + u16 mdio_clk_en; + u16 mdio_clk_dis; + u16 mdio_reset_req; + u16 mdio_reset_dreq; + u16 mdio_clk_st; + u16 mdio_reset_st; +}; + struct hns_mdio_device { void *vbase; /* mdio reg base address */ struct regmap *subctrl_vbase; + struct hns_mdio_sc_reg sc_reg; }; /* mdio reg */ @@ -93,7 +103,6 @@ enum mdio_c45_op_seq { #define MDIO_SC_CLK_DIS 0x33C #define MDIO_SC_RESET_REQ 0xA38 #define MDIO_SC_RESET_DREQ 0xA3C -#define MDIO_SC_CTRL 0x2010 #define MDIO_SC_CLK_ST 0x531C #define MDIO_SC_RESET_ST 0x5A1C @@ -353,6 +362,7 @@ static int hns_mdio_read(struct mii_bus *bus, int phy_id, int regnum) static int hns_mdio_reset(struct mii_bus *bus) { struct hns_mdio_device *mdio_dev = (struct hns_mdio_device *)bus->priv; + const struct hns_mdio_sc_reg *sc_reg; int ret; if (dev_of_node(bus->parent)) { @@ -361,9 +371,10 @@ static int hns_mdio_reset(struct mii_bus *bus) return -ENODEV; } + sc_reg = &mdio_dev->sc_reg; /* 1. reset req, and read reset st check */ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_RESET_REQ, 0x1, - MDIO_SC_RESET_ST, 0x1, + ret = mdio_sc_cfg_reg_write(mdio_dev, sc_reg->mdio_reset_req, + 0x1, sc_reg->mdio_reset_st, 0x1, MDIO_CHECK_SET_ST); if (ret) { dev_err(&bus->dev, "MDIO reset fail\n"); @@ -371,8 +382,8 @@ static int hns_mdio_reset(struct mii_bus *bus) } /* 2. dis clk, and read clk st check */ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_CLK_DIS, - 0x1, MDIO_SC_CLK_ST, 0x1, + ret = mdio_sc_cfg_reg_write(mdio_dev, sc_reg->mdio_clk_dis, + 0x1, sc_reg->mdio_clk_st, 0x1, MDIO_CHECK_CLR_ST); if (ret) { dev_err(&bus->dev, "MDIO dis clk fail\n"); @@ -380,8 +391,8 @@ static int hns_mdio_reset(struct mii_bus *bus) } /* 3. reset dreq, and read reset st check */ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_RESET_DREQ, 0x1, - MDIO_SC_RESET_ST, 0x1, + ret = mdio_sc_cfg_reg_write(mdio_dev, sc_reg->mdio_reset_dreq, + 0x1, sc_reg->mdio_reset_st, 0x1, MDIO_CHECK_CLR_ST); if (ret) { dev_err(&bus->dev, "MDIO dis clk fail\n"); @@ -389,8 +400,8 @@ static int hns_mdio_reset(struct mii_bus *bus) } /* 4. en clk, and read clk st check */ - ret = mdio_sc_cfg_reg_write(mdio_dev, MDIO_SC_CLK_EN, - 0x1, MDIO_SC_CLK_ST, 0x1, + ret = mdio_sc_cfg_reg_write(mdio_dev, sc_reg->mdio_clk_en, + 0x1, sc_reg->mdio_clk_st, 0x1, MDIO_CHECK_SET_ST); if (ret) dev_err(&bus->dev, "MDIO en clk fail\n"); @@ -458,13 +469,54 @@ static int hns_mdio_probe(struct platform_device *pdev) snprintf(new_bus->id, MII_BUS_ID_SIZE, "%s-%s", "Mii", dev_name(&pdev->dev)); if (dev_of_node(&pdev->dev)) { - mdio_dev->subctrl_vbase = syscon_node_to_regmap( - of_parse_phandle(pdev->dev.of_node, - "subctrl-vbase", 0)); - if (IS_ERR(mdio_dev->subctrl_vbase)) { - dev_warn(&pdev->dev, "no syscon hisilicon,peri-c-subctrl\n"); + struct of_phandle_args reg_args; + + ret = of_parse_phandle_with_fixed_args(pdev->dev.of_node, + "subctrl-vbase", + 4, + 0, + ®_args); + if (!ret) { + mdio_dev->subctrl_vbase = + syscon_node_to_regmap(reg_args.np); + if (IS_ERR(mdio_dev->subctrl_vbase)) { + dev_warn(&pdev->dev, "syscon_node_to_regmap error\n"); + mdio_dev->subctrl_vbase = NULL; + } else { + if (reg_args.args_count == 4) { + mdio_dev->sc_reg.mdio_clk_en = + (u16)reg_args.args[0]; + mdio_dev->sc_reg.mdio_clk_dis = + (u16)reg_args.args[0] + 4; + mdio_dev->sc_reg.mdio_reset_req = + (u16)reg_args.args[1]; + mdio_dev->sc_reg.mdio_reset_dreq = + (u16)reg_args.args[1] + 4; + mdio_dev->sc_reg.mdio_clk_st = + (u16)reg_args.args[2]; + mdio_dev->sc_reg.mdio_reset_st = + (u16)reg_args.args[3]; + } else { + /* for compatible */ + mdio_dev->sc_reg.mdio_clk_en = + MDIO_SC_CLK_EN; + mdio_dev->sc_reg.mdio_clk_dis = + MDIO_SC_CLK_DIS; + mdio_dev->sc_reg.mdio_reset_req = + MDIO_SC_RESET_REQ; + mdio_dev->sc_reg.mdio_reset_dreq = + MDIO_SC_RESET_DREQ; + mdio_dev->sc_reg.mdio_clk_st = + MDIO_SC_CLK_ST; + mdio_dev->sc_reg.mdio_reset_st = + MDIO_SC_RESET_ST; + } + } + } else { + dev_warn(&pdev->dev, "find syscon ret = %#x\n", ret); mdio_dev->subctrl_vbase = NULL; } + ret = of_mdiobus_register(new_bus, pdev->dev.of_node); } else if (is_acpi_node(pdev->dev.fwnode)) { /* Clear all the IRQ properties */ -- cgit v0.10.2 From 6ab36e35f11244b469cdf3c976c185363f1ce61c Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:50:54 +0300 Subject: net/mlx5: E-Switch, Add operational mode to the SRIOV e-Switch Define three modes for the SRIOV e-switch operation, none (SRIOV_NONE, none of the VF vports are enabled), legacy (SRIOV_LEGACY, the current mode) and sriov offloads (SRIOV_OFFLOADS). Currently, when in SRIOV, only the legacy mode is supported, where steering rules are of the form: destination mac --> VF vport This patch does not change any functionality. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index aebbd6c..8068dde 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -428,7 +428,7 @@ esw_fdb_set_vport_promisc_rule(struct mlx5_eswitch *esw, u32 vport) return __esw_fdb_set_vport_rule(esw, vport, true, mac_c, mac_v); } -static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) +static int esw_create_legacy_fdb_table(struct mlx5_eswitch *esw, int nvports) { int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); struct mlx5_core_dev *dev = esw->dev; @@ -479,7 +479,7 @@ static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) esw_warn(dev, "Failed to create flow group err(%d)\n", err); goto out; } - esw->fdb_table.addr_grp = g; + esw->fdb_table.legacy.addr_grp = g; /* Allmulti group : One rule that forwards any mcast traffic */ MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, @@ -494,7 +494,7 @@ static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) esw_warn(dev, "Failed to create allmulti flow group err(%d)\n", err); goto out; } - esw->fdb_table.allmulti_grp = g; + esw->fdb_table.legacy.allmulti_grp = g; /* Promiscuous group : * One rule that forward all unmatched traffic from previous groups @@ -511,17 +511,17 @@ static int esw_create_fdb_table(struct mlx5_eswitch *esw, int nvports) esw_warn(dev, "Failed to create promisc flow group err(%d)\n", err); goto out; } - esw->fdb_table.promisc_grp = g; + esw->fdb_table.legacy.promisc_grp = g; out: if (err) { - if (!IS_ERR_OR_NULL(esw->fdb_table.allmulti_grp)) { - mlx5_destroy_flow_group(esw->fdb_table.allmulti_grp); - esw->fdb_table.allmulti_grp = NULL; + if (!IS_ERR_OR_NULL(esw->fdb_table.legacy.allmulti_grp)) { + mlx5_destroy_flow_group(esw->fdb_table.legacy.allmulti_grp); + esw->fdb_table.legacy.allmulti_grp = NULL; } - if (!IS_ERR_OR_NULL(esw->fdb_table.addr_grp)) { - mlx5_destroy_flow_group(esw->fdb_table.addr_grp); - esw->fdb_table.addr_grp = NULL; + if (!IS_ERR_OR_NULL(esw->fdb_table.legacy.addr_grp)) { + mlx5_destroy_flow_group(esw->fdb_table.legacy.addr_grp); + esw->fdb_table.legacy.addr_grp = NULL; } if (!IS_ERR_OR_NULL(esw->fdb_table.fdb)) { mlx5_destroy_flow_table(esw->fdb_table.fdb); @@ -533,20 +533,20 @@ out: return err; } -static void esw_destroy_fdb_table(struct mlx5_eswitch *esw) +static void esw_destroy_legacy_fdb_table(struct mlx5_eswitch *esw) { if (!esw->fdb_table.fdb) return; esw_debug(esw->dev, "Destroy FDB Table\n"); - mlx5_destroy_flow_group(esw->fdb_table.promisc_grp); - mlx5_destroy_flow_group(esw->fdb_table.allmulti_grp); - mlx5_destroy_flow_group(esw->fdb_table.addr_grp); + mlx5_destroy_flow_group(esw->fdb_table.legacy.promisc_grp); + mlx5_destroy_flow_group(esw->fdb_table.legacy.allmulti_grp); + mlx5_destroy_flow_group(esw->fdb_table.legacy.addr_grp); mlx5_destroy_flow_table(esw->fdb_table.fdb); esw->fdb_table.fdb = NULL; - esw->fdb_table.addr_grp = NULL; - esw->fdb_table.allmulti_grp = NULL; - esw->fdb_table.promisc_grp = NULL; + esw->fdb_table.legacy.addr_grp = NULL; + esw->fdb_table.legacy.allmulti_grp = NULL; + esw->fdb_table.legacy.promisc_grp = NULL; } /* E-Switch vport UC/MC lists management */ @@ -1540,7 +1540,7 @@ static void esw_disable_vport(struct mlx5_eswitch *esw, int vport_num) } /* Public E-Switch API */ -int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs) +int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode) { int err; int i; @@ -1561,11 +1561,14 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs) if (!MLX5_CAP_ESW_EGRESS_ACL(esw->dev, ft_support)) esw_warn(esw->dev, "E-Switch engress ACL is not supported by FW\n"); - esw_info(esw->dev, "E-Switch enable SRIOV: nvfs(%d)\n", nvfs); + esw_info(esw->dev, "E-Switch enable SRIOV: nvfs(%d) mode (%d)\n", nvfs, mode); + if (mode != SRIOV_LEGACY) + return -EINVAL; + esw->mode = mode; esw_disable_vport(esw, 0); - err = esw_create_fdb_table(esw, nvfs + 1); + err = esw_create_legacy_fdb_table(esw, nvfs + 1); if (err) goto abort; @@ -1590,8 +1593,8 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) MLX5_CAP_GEN(esw->dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) return; - esw_info(esw->dev, "disable SRIOV: active vports(%d)\n", - esw->enabled_vports); + esw_info(esw->dev, "disable SRIOV: active vports(%d) mode(%d)\n", + esw->enabled_vports, esw->mode); mc_promisc = esw->mc_promisc; @@ -1601,8 +1604,9 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) if (mc_promisc && mc_promisc->uplink_rule) mlx5_del_flow_rule(mc_promisc->uplink_rule); - esw_destroy_fdb_table(esw); + esw_destroy_legacy_fdb_table(esw); + esw->mode = SRIOV_NONE; /* VPORT 0 (PF) must be enabled back with non-sriov configuration */ esw_enable_vport(esw, 0, UC_ADDR_CHANGE); } @@ -1673,6 +1677,7 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev) esw->total_vports = total_vports; esw->enabled_vports = 0; + esw->mode = SRIOV_NONE; dev->priv.eswitch = esw; esw_enable_vport(esw, 0, UC_ADDR_CHANGE); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index fd68002..544fbfe 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -134,9 +134,19 @@ struct mlx5_l2_table { struct mlx5_eswitch_fdb { void *fdb; - struct mlx5_flow_group *addr_grp; - struct mlx5_flow_group *allmulti_grp; - struct mlx5_flow_group *promisc_grp; + union { + struct legacy_fdb { + struct mlx5_flow_group *addr_grp; + struct mlx5_flow_group *allmulti_grp; + struct mlx5_flow_group *promisc_grp; + } legacy; + }; +}; + +enum { + SRIOV_NONE, + SRIOV_LEGACY, + SRIOV_OFFLOADS }; struct mlx5_eswitch { @@ -153,13 +163,14 @@ struct mlx5_eswitch { */ struct mutex state_lock; struct esw_mc_addr *mc_promisc; + int mode; }; /* E-Switch API */ int mlx5_eswitch_init(struct mlx5_core_dev *dev); void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw); void mlx5_eswitch_vport_event(struct mlx5_eswitch *esw, struct mlx5_eqe *eqe); -int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs); +int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode); void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw); int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw, int vport, u8 mac[ETH_ALEN]); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c index d6a3f41..b380a6b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c @@ -167,7 +167,7 @@ int mlx5_core_sriov_configure(struct pci_dev *pdev, int num_vfs) mlx5_core_init_vfs(dev, num_vfs); #ifdef CONFIG_MLX5_CORE_EN - mlx5_eswitch_enable_sriov(dev->priv.eswitch, num_vfs); + mlx5_eswitch_enable_sriov(dev->priv.eswitch, num_vfs, SRIOV_LEGACY); #endif return num_vfs; @@ -209,7 +209,8 @@ int mlx5_sriov_init(struct mlx5_core_dev *dev) mlx5_core_init_vfs(dev, cur_vfs); #ifdef CONFIG_MLX5_CORE_EN if (cur_vfs) - mlx5_eswitch_enable_sriov(dev->priv.eswitch, cur_vfs); + mlx5_eswitch_enable_sriov(dev->priv.eswitch, cur_vfs, + SRIOV_LEGACY); #endif enable_vfs(dev, cur_vfs); -- cgit v0.10.2 From 69697b6e2086b5860bd2d216bc4c6c49d84d73ff Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:50:55 +0300 Subject: net/mlx5: E-Switch, Add support for the sriov offloads mode Unlike the legacy mode, here, forwarding rules are not learned by the driver per events on macs set by VFs/VMs into their vports, but rather should be programmed by higher-level SW entities. Saying that, still, in the offloads mode (SRIOV_OFFLOADS), two flow groups are created by the driver for management (slow path) purposes: The first group will be used for sending packets over e-switch vports from the host OS where the e-switch management code runs, to be received by VFs. The second group will be used by a miss rule which forwards packets toward the e-switch manager. Further logic will trap these packets such that the receiving net-device as seen by the networking stack is the representor of the vport that sent the packet over the e-switch data-path. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index c4f450f..96f1826 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -5,7 +5,7 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o \ fs_counters.o rl.o -mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \ +mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o eswitch_offloads.o \ en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \ en_rx_am.o en_txrx.o en_clock.o vxlan.o en_tc.o \ en_arfs.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 8068dde..1fc4cfd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -40,17 +40,6 @@ #define UPLINK_VPORT 0xFFFF -#define MLX5_DEBUG_ESWITCH_MASK BIT(3) - -#define esw_info(dev, format, ...) \ - pr_info("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__) - -#define esw_warn(dev, format, ...) \ - pr_warn("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__) - -#define esw_debug(dev, format, ...) \ - mlx5_core_dbg_mask(dev, MLX5_DEBUG_ESWITCH_MASK, format, ##__VA_ARGS__) - enum { MLX5_ACTION_NONE = 0, MLX5_ACTION_ADD = 1, @@ -92,6 +81,9 @@ enum { MC_ADDR_CHANGE | \ PROMISC_CHANGE) +int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports); +void esw_destroy_offloads_fdb_table(struct mlx5_eswitch *esw); + static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, u16 vport, u32 events_mask) { @@ -578,7 +570,8 @@ static int esw_add_uc_addr(struct mlx5_eswitch *esw, struct vport_addr *vaddr) if (err) goto abort; - if (esw->fdb_table.fdb) /* SRIOV is enabled: Forward UC MAC to vport */ + /* SRIOV is enabled: Forward UC MAC to vport */ + if (esw->fdb_table.fdb && esw->mode == SRIOV_LEGACY) vaddr->flow_rule = esw_fdb_set_vport_rule(esw, mac, vport); esw_debug(esw->dev, "\tADDED UC MAC: vport[%d] %pM index:%d fr(%p)\n", @@ -1543,7 +1536,7 @@ static void esw_disable_vport(struct mlx5_eswitch *esw, int vport_num) int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode) { int err; - int i; + int i, enabled_events; if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager) || MLX5_CAP_GEN(esw->dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) @@ -1562,18 +1555,19 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode) esw_warn(esw->dev, "E-Switch engress ACL is not supported by FW\n"); esw_info(esw->dev, "E-Switch enable SRIOV: nvfs(%d) mode (%d)\n", nvfs, mode); - if (mode != SRIOV_LEGACY) - return -EINVAL; - esw->mode = mode; esw_disable_vport(esw, 0); - err = esw_create_legacy_fdb_table(esw, nvfs + 1); + if (mode == SRIOV_LEGACY) + err = esw_create_legacy_fdb_table(esw, nvfs + 1); + else + err = esw_create_offloads_fdb_table(esw, nvfs + 1); if (err) goto abort; + enabled_events = (mode == SRIOV_LEGACY) ? SRIOV_VPORT_EVENTS : UC_ADDR_CHANGE; for (i = 0; i <= nvfs; i++) - esw_enable_vport(esw, i, SRIOV_VPORT_EVENTS); + esw_enable_vport(esw, i, enabled_events); esw_info(esw->dev, "SRIOV enabled: active vports(%d)\n", esw->enabled_vports); @@ -1604,7 +1598,10 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) if (mc_promisc && mc_promisc->uplink_rule) mlx5_del_flow_rule(mc_promisc->uplink_rule); - esw_destroy_legacy_fdb_table(esw); + if (esw->mode == SRIOV_LEGACY) + esw_destroy_legacy_fdb_table(esw); + else + esw_destroy_offloads_fdb_table(esw); esw->mode = SRIOV_NONE; /* VPORT 0 (PF) must be enabled back with non-sriov configuration */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 544fbfe..2360180 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -140,6 +140,11 @@ struct mlx5_eswitch_fdb { struct mlx5_flow_group *allmulti_grp; struct mlx5_flow_group *promisc_grp; } legacy; + + struct offloads_fdb { + struct mlx5_flow_group *send_to_vport_grp; + struct mlx5_flow_group *miss_grp; + } offloads; }; }; @@ -188,4 +193,15 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, int vport, struct ifla_vf_stats *vf_stats); +#define MLX5_DEBUG_ESWITCH_MASK BIT(3) + +#define esw_info(dev, format, ...) \ + pr_info("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__) + +#define esw_warn(dev, format, ...) \ + pr_warn("(%s): E-Switch: " format, (dev)->priv.name, ##__VA_ARGS__) + +#define esw_debug(dev, format, ...) \ + mlx5_core_dbg_mask(dev, MLX5_DEBUG_ESWITCH_MASK, format, ##__VA_ARGS__) + #endif /* __MLX5_ESWITCH_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c new file mode 100644 index 0000000..c6b28df --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -0,0 +1,135 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include "mlx5_core.h" +#include "eswitch.h" + +#define MAX_PF_SQ 256 + +int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_core_dev *dev = esw->dev; + struct mlx5_flow_namespace *root_ns; + struct mlx5_flow_table *fdb = NULL; + struct mlx5_flow_group *g; + u32 *flow_group_in; + void *match_criteria; + int table_size, ix, err = 0; + + flow_group_in = mlx5_vzalloc(inlen); + if (!flow_group_in) + return -ENOMEM; + + root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB); + if (!root_ns) { + esw_warn(dev, "Failed to get FDB flow namespace\n"); + goto ns_err; + } + + esw_debug(dev, "Create offloads FDB table, log_max_size(%d)\n", + MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size)); + + table_size = nvports + MAX_PF_SQ + 1; + fdb = mlx5_create_flow_table(root_ns, 0, table_size, 0); + if (IS_ERR(fdb)) { + err = PTR_ERR(fdb); + esw_warn(dev, "Failed to create FDB Table err %d\n", err); + goto fdb_err; + } + esw->fdb_table.fdb = fdb; + + /* create send-to-vport group */ + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_MISC_PARAMETERS); + + match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria); + + MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_sqn); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_port); + + ix = nvports + MAX_PF_SQ; + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ix - 1); + + g = mlx5_create_flow_group(fdb, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + esw_warn(dev, "Failed to create send-to-vport flow group err(%d)\n", err); + goto send_vport_err; + } + esw->fdb_table.offloads.send_to_vport_grp = g; + + /* create miss group */ + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, 0); + + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, ix); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ix + 1); + + g = mlx5_create_flow_group(fdb, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + esw_warn(dev, "Failed to create miss flow group err(%d)\n", err); + goto miss_err; + } + esw->fdb_table.offloads.miss_grp = g; + + return 0; + +miss_err: + mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp); +send_vport_err: + mlx5_destroy_flow_table(fdb); +fdb_err: +ns_err: + kvfree(flow_group_in); + return err; +} + +void esw_destroy_offloads_fdb_table(struct mlx5_eswitch *esw) +{ + if (!esw->fdb_table.fdb) + return; + + esw_debug(esw->dev, "Destroy offloads FDB Table\n"); + mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp); + mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp); + + mlx5_destroy_flow_table(esw->fdb_table.fdb); +} -- cgit v0.10.2 From 3aa335724f0793027c87fae03ecf0a297fc04b29 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:50:56 +0300 Subject: net/mlx5: E-Switch, Add miss rule for offloads mode In the sriov offloads mode, packets that are not matched by any other rule should be sent towards the e-switch manager for further processing. Add such "miss" rule which matches ANY packet as the last rule in the e-switch FDB and programs the HW to send the packet to vport 0 where the e-switch manager runs. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 2360180..8eed33f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -144,6 +144,7 @@ struct mlx5_eswitch_fdb { struct offloads_fdb { struct mlx5_flow_group *send_to_vport_grp; struct mlx5_flow_group *miss_grp; + struct mlx5_flow_rule *miss_rule; } offloads; }; }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index c6b28df..e3d81ae 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -38,6 +38,39 @@ #include "mlx5_core.h" #include "eswitch.h" +static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) +{ + struct mlx5_flow_destination dest; + struct mlx5_flow_rule *flow_rule = NULL; + u32 *match_v, *match_c; + int err = 0; + + match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + if (!match_v || !match_c) { + esw_warn(esw->dev, "FDB: Failed to alloc match parameters\n"); + err = -ENOMEM; + goto out; + } + + dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; + dest.vport_num = 0; + + flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, 0, match_c, match_v, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); + if (IS_ERR(flow_rule)) { + err = PTR_ERR(flow_rule); + esw_warn(esw->dev, "FDB: Failed to add miss flow rule err %d\n", err); + goto out; + } + + esw->fdb_table.offloads.miss_rule = flow_rule; +out: + kfree(match_v); + kfree(match_c); + return err; +} + #define MAX_PF_SQ 256 int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports) @@ -110,8 +143,14 @@ int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports) } esw->fdb_table.offloads.miss_grp = g; + err = esw_add_fdb_miss_rule(esw); + if (err) + goto miss_rule_err; + return 0; +miss_rule_err: + mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp); miss_err: mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp); send_vport_err: @@ -128,6 +167,7 @@ void esw_destroy_offloads_fdb_table(struct mlx5_eswitch *esw) return; esw_debug(esw->dev, "Destroy offloads FDB Table\n"); + mlx5_del_flow_rule(esw->fdb_table.offloads.miss_rule); mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp); mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp); -- cgit v0.10.2 From ab22be9ba30a08482b2c2effb36ac3f0ed3df465 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:50:57 +0300 Subject: net/mlx5: E-Switch, Add API to create send-to-vport rules Add the API to create send-to-vport e-switch rules of the form packet meta-data :: send-queue-number == $SQN and source-vport == 0 --> $VPORT These rules are to be used for a send-to-vport logic which conceptually bypasses the "normal" steering rules currently present at the e-switch datapath. Such rule should apply only for packets that originate in the e-switch manager vport (0) and are sent for a given SQN which is used by a given VF representor device, and hence the matching logic. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 8eed33f..b7fabd1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -193,6 +193,8 @@ int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw, int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, int vport, struct ifla_vf_stats *vf_stats); +struct mlx5_flow_rule * +mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn); #define MLX5_DEBUG_ESWITCH_MASK BIT(3) @@ -204,5 +206,4 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, #define esw_debug(dev, format, ...) \ mlx5_core_dbg_mask(dev, MLX5_DEBUG_ESWITCH_MASK, format, ##__VA_ARGS__) - #endif /* __MLX5_ESWITCH_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index e3d81ae..8964f71 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -38,6 +38,45 @@ #include "mlx5_core.h" #include "eswitch.h" +struct mlx5_flow_rule * +mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn) +{ + struct mlx5_flow_destination dest; + struct mlx5_flow_rule *flow_rule; + int match_header = MLX5_MATCH_MISC_PARAMETERS; + u32 *match_v, *match_c; + void *misc; + + match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + if (!match_v || !match_c) { + esw_warn(esw->dev, "FDB: Failed to alloc match parameters\n"); + flow_rule = ERR_PTR(-ENOMEM); + goto out; + } + + misc = MLX5_ADDR_OF(fte_match_param, match_v, misc_parameters); + MLX5_SET(fte_match_set_misc, misc, source_sqn, sqn); + MLX5_SET(fte_match_set_misc, misc, source_port, 0x0); /* source vport is 0 */ + + misc = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); + MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_sqn); + MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + + dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; + dest.vport_num = vport; + + flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, match_header, match_c, + match_v, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + 0, &dest); + if (IS_ERR(flow_rule)) + esw_warn(esw->dev, "FDB: Failed to add send to vport rule err %ld\n", PTR_ERR(flow_rule)); +out: + kfree(match_v); + kfree(match_c); + return flow_rule; +} + static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) { struct mlx5_flow_destination dest; -- cgit v0.10.2 From acbc2004d7129a1ecf02414c1da8808bdc06d5a2 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:50:58 +0300 Subject: net/mlx5: Introduce offloads steering namespace Add a new namespace (MLX5_FLOW_NAMESPACE_OFFLOADS) to be populated with flow steering rules that deal with rules that have have to be executed before the EN NIC steering rules are matched. The namespace is located after the bypass name-space and before the kernel name-space. Therefore, it precedes the HW processing done for rules set for the kernel NIC name-space. Under SRIOV, it would allow us to match on e-switch missed packet and forward them to the relevant VF representor TIR. Signed-off-by: Or Gerlitz Signed-off-by: Amir Vadai Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index e912a3d..b040110 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -83,6 +83,11 @@ #define ANCHOR_NUM_LEVELS 1 #define ANCHOR_NUM_PRIOS 1 #define ANCHOR_MIN_LEVEL (BY_PASS_MIN_LEVEL + 1) + +#define OFFLOADS_MAX_FT 1 +#define OFFLOADS_NUM_PRIOS 1 +#define OFFLOADS_MIN_LEVEL (ANCHOR_MIN_LEVEL + 1) + struct node_caps { size_t arr_sz; long *caps; @@ -98,7 +103,7 @@ static struct init_tree_node { int num_levels; } root_fs = { .type = FS_TYPE_NAMESPACE, - .ar_size = 4, + .ar_size = 5, .children = (struct init_tree_node[]) { ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0, FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), @@ -107,6 +112,9 @@ static struct init_tree_node { FS_CAP(flow_table_properties_nic_receive.flow_table_modify)), ADD_NS(ADD_MULTIPLE_PRIO(MLX5_BY_PASS_NUM_PRIOS, BY_PASS_PRIO_NUM_LEVELS))), + ADD_PRIO(0, OFFLOADS_MIN_LEVEL, 0, {}, + ADD_NS(ADD_MULTIPLE_PRIO(OFFLOADS_NUM_PRIOS, OFFLOADS_MAX_FT))), + ADD_PRIO(0, KERNEL_MIN_LEVEL, 0, {}, ADD_NS(ADD_MULTIPLE_PRIO(1, 1), ADD_MULTIPLE_PRIO(KERNEL_NIC_NUM_PRIOS, @@ -1369,6 +1377,7 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, switch (type) { case MLX5_FLOW_NAMESPACE_BYPASS: + case MLX5_FLOW_NAMESPACE_OFFLOADS: case MLX5_FLOW_NAMESPACE_KERNEL: case MLX5_FLOW_NAMESPACE_LEFTOVERS: case MLX5_FLOW_NAMESPACE_ANCHOR: diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index 4b7a107..6ad1119 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -54,6 +54,7 @@ static inline void build_leftovers_ft_param(int *priority, enum mlx5_flow_namespace_type { MLX5_FLOW_NAMESPACE_BYPASS, + MLX5_FLOW_NAMESPACE_OFFLOADS, MLX5_FLOW_NAMESPACE_KERNEL, MLX5_FLOW_NAMESPACE_LEFTOVERS, MLX5_FLOW_NAMESPACE_ANCHOR, -- cgit v0.10.2 From c116c6eec6f72aac82ff4228ab1d277f3f9a2460 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:50:59 +0300 Subject: net/mlx5: E-Switch, Add offloads table Belongs to the NIC offloads name-space, and to be used as part of the SRIOV offloads logic to steer packets that hit the e-switch miss rule to the TIR of the relevant VF representor. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index b7fabd1..32db37a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -155,6 +155,10 @@ enum { SRIOV_OFFLOADS }; +struct mlx5_esw_offload { + struct mlx5_flow_table *ft_offloads; +}; + struct mlx5_eswitch { struct mlx5_core_dev *dev; struct mlx5_l2_table l2_table; @@ -169,6 +173,7 @@ struct mlx5_eswitch { */ struct mutex state_lock; struct esw_mc_addr *mc_promisc; + struct mlx5_esw_offload offloads; int mode; }; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 8964f71..e895c6f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -212,3 +212,34 @@ void esw_destroy_offloads_fdb_table(struct mlx5_eswitch *esw) mlx5_destroy_flow_table(esw->fdb_table.fdb); } + +static int esw_create_offloads_table(struct mlx5_eswitch *esw) +{ + struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *ft_offloads; + struct mlx5_core_dev *dev = esw->dev; + int err = 0; + + ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_OFFLOADS); + if (!ns) { + esw_warn(esw->dev, "Failed to get offloads flow namespace\n"); + return -ENOMEM; + } + + ft_offloads = mlx5_create_flow_table(ns, 0, dev->priv.sriov.num_vfs + 2, 0); + if (IS_ERR(ft_offloads)) { + err = PTR_ERR(ft_offloads); + esw_warn(esw->dev, "Failed to create offloads table, err %d\n", err); + return err; + } + + esw->offloads.ft_offloads = ft_offloads; + return 0; +} + +static void esw_destroy_offloads_table(struct mlx5_eswitch *esw) +{ + struct mlx5_esw_offload *offloads = &esw->offloads; + + mlx5_destroy_flow_table(offloads->ft_offloads); +} -- cgit v0.10.2 From fed9ce22bf8ae8f417b8f047d2d630542d152ccf Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:51:00 +0300 Subject: net/mlx5: E-Switch, Add API to create vport rx rules Add the API to create vport rx rules of the form packet meta-data :: vport == $VPORT --> $TIR where the TIR is opened by this VF representor. This logic will by used for packets that didn't match any rule in the e-switch datapath and should be received into the host OS through the netdevice that represents the VF they were sent from. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 32db37a..cf959f7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -157,6 +157,7 @@ enum { struct mlx5_esw_offload { struct mlx5_flow_table *ft_offloads; + struct mlx5_flow_group *vport_rx_group; }; struct mlx5_eswitch { @@ -201,6 +202,9 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, struct mlx5_flow_rule * mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn); +struct mlx5_flow_rule * +mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn); + #define MLX5_DEBUG_ESWITCH_MASK BIT(3) #define esw_info(dev, format, ...) \ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index e895c6f..7aad367 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -243,3 +243,88 @@ static void esw_destroy_offloads_table(struct mlx5_eswitch *esw) mlx5_destroy_flow_table(offloads->ft_offloads); } + +static int esw_create_vport_rx_group(struct mlx5_eswitch *esw) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_group *g; + struct mlx5_priv *priv = &esw->dev->priv; + u32 *flow_group_in; + void *match_criteria, *misc; + int err = 0; + int nvports = priv->sriov.num_vfs + 2; + + flow_group_in = mlx5_vzalloc(inlen); + if (!flow_group_in) + return -ENOMEM; + + /* create vport rx group */ + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_MISC_PARAMETERS); + + match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria); + misc = MLX5_ADDR_OF(fte_match_param, match_criteria, misc_parameters); + MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, nvports - 1); + + g = mlx5_create_flow_group(esw->offloads.ft_offloads, flow_group_in); + + if (IS_ERR(g)) { + err = PTR_ERR(g); + mlx5_core_warn(esw->dev, "Failed to create vport rx group err %d\n", err); + goto out; + } + + esw->offloads.vport_rx_group = g; +out: + kfree(flow_group_in); + return err; +} + +static void esw_destroy_vport_rx_group(struct mlx5_eswitch *esw) +{ + mlx5_destroy_flow_group(esw->offloads.vport_rx_group); +} + +struct mlx5_flow_rule * +mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn) +{ + struct mlx5_flow_destination dest; + struct mlx5_flow_rule *flow_rule; + int match_header = MLX5_MATCH_MISC_PARAMETERS; + u32 *match_v, *match_c; + void *misc; + + match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + if (!match_v || !match_c) { + esw_warn(esw->dev, "Failed to alloc match parameters\n"); + flow_rule = ERR_PTR(-ENOMEM); + goto out; + } + + misc = MLX5_ADDR_OF(fte_match_param, match_v, misc_parameters); + MLX5_SET(fte_match_set_misc, misc, source_port, vport); + + misc = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); + MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + + dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; + dest.tir_num = tirn; + + flow_rule = mlx5_add_flow_rule(esw->offloads.ft_offloads, match_header, match_c, + match_v, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + 0, &dest); + if (IS_ERR(flow_rule)) { + esw_warn(esw->dev, "fs offloads: Failed to add vport rx rule err %ld\n", PTR_ERR(flow_rule)); + goto out; + } + +out: + kfree(match_v); + kfree(match_c); + return flow_rule; +} -- cgit v0.10.2 From 08f4b5918b2d6b491f0403cc1886f5cdccef89bb Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:51:01 +0300 Subject: net/devlink: Add E-Switch mode control Add the commands to set and show the mode of SRIOV E-Switch, two modes are supported: * legacy: operating in the "old" L2 based mode (DMAC --> VF vport) * switchdev: the E-Switch is referred to as whitebox switch configured using standard tools such as tc, bridge, openvswitch etc. To allow working with the tools, for each VF, a VF representor netdevice is created by the E-Switch manager vendor device driver instance (e.g PF). Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/include/net/devlink.h b/include/net/devlink.h index 1d45b61..c99ffe8 100644 --- a/include/net/devlink.h +++ b/include/net/devlink.h @@ -90,6 +90,9 @@ struct devlink_ops { u16 tc_index, enum devlink_sb_pool_type pool_type, u32 *p_cur, u32 *p_max); + + int (*eswitch_mode_get)(struct devlink *devlink, u16 *p_mode); + int (*eswitch_mode_set)(struct devlink *devlink, u16 mode); }; static inline void *devlink_priv(struct devlink *devlink) diff --git a/include/uapi/linux/devlink.h b/include/uapi/linux/devlink.h index ba0073b..915bfa7 100644 --- a/include/uapi/linux/devlink.h +++ b/include/uapi/linux/devlink.h @@ -57,6 +57,8 @@ enum devlink_command { DEVLINK_CMD_SB_OCC_SNAPSHOT, DEVLINK_CMD_SB_OCC_MAX_CLEAR, + DEVLINK_CMD_ESWITCH_MODE_GET, + DEVLINK_CMD_ESWITCH_MODE_SET, /* add new commands above here */ __DEVLINK_CMD_MAX, @@ -95,6 +97,11 @@ enum devlink_sb_threshold_type { #define DEVLINK_SB_THRESHOLD_TO_ALPHA_MAX 20 +enum devlink_eswitch_mode { + DEVLINK_ESWITCH_MODE_LEGACY, + DEVLINK_ESWITCH_MODE_SWITCHDEV, +}; + enum devlink_attr { /* don't change the order or add anything between, this is ABI! */ DEVLINK_ATTR_UNSPEC, @@ -125,6 +132,7 @@ enum devlink_attr { DEVLINK_ATTR_SB_TC_INDEX, /* u16 */ DEVLINK_ATTR_SB_OCC_CUR, /* u32 */ DEVLINK_ATTR_SB_OCC_MAX, /* u32 */ + DEVLINK_ATTR_ESWITCH_MODE, /* u16 */ /* add new attributes above here, update the policy in devlink.c */ diff --git a/net/core/devlink.c b/net/core/devlink.c index 933e8d4..b2e592a 100644 --- a/net/core/devlink.c +++ b/net/core/devlink.c @@ -1394,6 +1394,78 @@ static int devlink_nl_cmd_sb_occ_max_clear_doit(struct sk_buff *skb, return -EOPNOTSUPP; } +static int devlink_eswitch_fill(struct sk_buff *msg, struct devlink *devlink, + enum devlink_command cmd, u32 portid, + u32 seq, int flags, u16 mode) +{ + void *hdr; + + hdr = genlmsg_put(msg, portid, seq, &devlink_nl_family, flags, cmd); + if (!hdr) + return -EMSGSIZE; + + if (devlink_nl_put_handle(msg, devlink)) + goto nla_put_failure; + + if (nla_put_u16(msg, DEVLINK_ATTR_ESWITCH_MODE, mode)) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); + return -EMSGSIZE; +} + +static int devlink_nl_cmd_eswitch_mode_get_doit(struct sk_buff *skb, + struct genl_info *info) +{ + struct devlink *devlink = info->user_ptr[0]; + const struct devlink_ops *ops = devlink->ops; + struct sk_buff *msg; + u16 mode; + int err; + + if (!ops || !ops->eswitch_mode_get) + return -EOPNOTSUPP; + + err = ops->eswitch_mode_get(devlink, &mode); + if (err) + return err; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + err = devlink_eswitch_fill(msg, devlink, DEVLINK_CMD_ESWITCH_MODE_GET, + info->snd_portid, info->snd_seq, 0, mode); + + if (err) { + nlmsg_free(msg); + return err; + } + + return genlmsg_reply(msg, info); +} + +static int devlink_nl_cmd_eswitch_mode_set_doit(struct sk_buff *skb, + struct genl_info *info) +{ + struct devlink *devlink = info->user_ptr[0]; + const struct devlink_ops *ops = devlink->ops; + u16 mode; + + if (!info->attrs[DEVLINK_ATTR_ESWITCH_MODE]) + return -EINVAL; + + mode = nla_get_u16(info->attrs[DEVLINK_ATTR_ESWITCH_MODE]); + + if (ops && ops->eswitch_mode_set) + return ops->eswitch_mode_set(devlink, mode); + return -EOPNOTSUPP; +} + static const struct nla_policy devlink_nl_policy[DEVLINK_ATTR_MAX + 1] = { [DEVLINK_ATTR_BUS_NAME] = { .type = NLA_NUL_STRING }, [DEVLINK_ATTR_DEV_NAME] = { .type = NLA_NUL_STRING }, @@ -1407,6 +1479,7 @@ static const struct nla_policy devlink_nl_policy[DEVLINK_ATTR_MAX + 1] = { [DEVLINK_ATTR_SB_POOL_THRESHOLD_TYPE] = { .type = NLA_U8 }, [DEVLINK_ATTR_SB_THRESHOLD] = { .type = NLA_U32 }, [DEVLINK_ATTR_SB_TC_INDEX] = { .type = NLA_U16 }, + [DEVLINK_ATTR_ESWITCH_MODE] = { .type = NLA_U16 }, }; static const struct genl_ops devlink_nl_ops[] = { @@ -1525,6 +1598,20 @@ static const struct genl_ops devlink_nl_ops[] = { DEVLINK_NL_FLAG_NEED_SB | DEVLINK_NL_FLAG_LOCK_PORTS, }, + { + .cmd = DEVLINK_CMD_ESWITCH_MODE_GET, + .doit = devlink_nl_cmd_eswitch_mode_get_doit, + .policy = devlink_nl_policy, + .flags = GENL_ADMIN_PERM, + .internal_flags = DEVLINK_NL_FLAG_NEED_DEVLINK, + }, + { + .cmd = DEVLINK_CMD_ESWITCH_MODE_SET, + .doit = devlink_nl_cmd_eswitch_mode_set_doit, + .policy = devlink_nl_policy, + .flags = GENL_ADMIN_PERM, + .internal_flags = DEVLINK_NL_FLAG_NEED_DEVLINK, + }, }; /** -- cgit v0.10.2 From feae908744d7f78b9dd06afda9de47f997f2d81a Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:51:02 +0300 Subject: net/mlx5: Add devlink interface The devlink interface is initially used to set/get the mode of the SRIOV e-switch. Currently, these are only stubs for get/set, down-stream patch will actually fill them out. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig index 1cf722e..aae4688 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig +++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig @@ -4,6 +4,7 @@ config MLX5_CORE tristate "Mellanox Technologies ConnectX-4 and Connect-IB core driver" + depends on MAY_USE_DEVLINK depends on PCI default n ---help--- diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index cf959f7..7843f98 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -35,6 +35,7 @@ #include #include +#include #include #define MLX5_MAX_UC_PER_VPORT(dev) \ @@ -205,6 +206,9 @@ mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn struct mlx5_flow_rule * mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn); +int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode); +int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode); + #define MLX5_DEBUG_ESWITCH_MASK BIT(3) #define esw_info(dev, format, ...) \ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 7aad367..e1727a9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -328,3 +328,13 @@ out: kfree(match_c); return flow_rule; } + +int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode) +{ + return -EOPNOTSUPP; +} + +int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode) +{ + return -EOPNOTSUPP; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 1f3b6d6..1fb3c68 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -51,6 +51,7 @@ #ifdef CONFIG_RFS_ACCEL #include #endif +#include #include "mlx5_core.h" #include "fs_core.h" #ifdef CONFIG_MLX5_CORE_EN @@ -1315,19 +1316,28 @@ struct mlx5_core_event_handler { void *data); }; +static const struct devlink_ops mlx5_devlink_ops = { +#ifdef CONFIG_MLX5_CORE_EN + .eswitch_mode_set = mlx5_devlink_eswitch_mode_set, + .eswitch_mode_get = mlx5_devlink_eswitch_mode_get, +#endif +}; static int init_one(struct pci_dev *pdev, const struct pci_device_id *id) { struct mlx5_core_dev *dev; + struct devlink *devlink; struct mlx5_priv *priv; int err; - dev = kzalloc(sizeof(*dev), GFP_KERNEL); - if (!dev) { + devlink = devlink_alloc(&mlx5_devlink_ops, sizeof(*dev)); + if (!devlink) { dev_err(&pdev->dev, "kzalloc failed\n"); return -ENOMEM; } + + dev = devlink_priv(devlink); priv = &dev->priv; priv->pci_dev_data = id->driver_data; @@ -1364,15 +1374,21 @@ static int init_one(struct pci_dev *pdev, goto clean_health; } + err = devlink_register(devlink, &pdev->dev); + if (err) + goto clean_load; + return 0; +clean_load: + mlx5_unload_one(dev, priv); clean_health: mlx5_health_cleanup(dev); close_pci: mlx5_pci_close(dev, priv); clean_dev: pci_set_drvdata(pdev, NULL); - kfree(dev); + devlink_free(devlink); return err; } @@ -1380,8 +1396,10 @@ clean_dev: static void remove_one(struct pci_dev *pdev) { struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + struct devlink *devlink = priv_to_devlink(dev); struct mlx5_priv *priv = &dev->priv; + devlink_unregister(devlink); if (mlx5_unload_one(dev, priv)) { dev_err(&dev->pdev->dev, "mlx5_unload_one failed\n"); mlx5_health_cleanup(dev); @@ -1390,7 +1408,7 @@ static void remove_one(struct pci_dev *pdev) mlx5_health_cleanup(dev); mlx5_pci_close(dev, priv); pci_set_drvdata(pdev, NULL); - kfree(dev); + devlink_free(devlink); } static pci_ers_result_t mlx5_pci_err_detected(struct pci_dev *pdev, -- cgit v0.10.2 From c930a3ad7453615b6707509e23afa5969095b5b7 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:51:03 +0300 Subject: net/mlx5e: Add devlink based SRIOV mode changes Implement handlers for the devlink commands to get and set the SRIOV E-Switch mode. When turning to the switchdev/offloads mode, we disable the e-switch and enable it again in the new mode, create the NIC offloads table and create VF reps. When turning to legacy mode, we remove the VF reps and the offloads table, and re-initiate the e-switch in it's legacy mode. The actual creation/removal of the VF reps is done in downstream patches. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 1fc4cfd..12f509c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -81,8 +81,8 @@ enum { MC_ADDR_CHANGE | \ PROMISC_CHANGE) -int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports); -void esw_destroy_offloads_fdb_table(struct mlx5_eswitch *esw); +int esw_offloads_init(struct mlx5_eswitch *esw, int nvports); +void esw_offloads_cleanup(struct mlx5_eswitch *esw, int nvports); static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, u16 vport, u32 events_mask) @@ -1561,7 +1561,7 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode) if (mode == SRIOV_LEGACY) err = esw_create_legacy_fdb_table(esw, nvfs + 1); else - err = esw_create_offloads_fdb_table(esw, nvfs + 1); + err = esw_offloads_init(esw, nvfs + 1); if (err) goto abort; @@ -1581,6 +1581,7 @@ abort: void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) { struct esw_mc_addr *mc_promisc; + int nvports; int i; if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager) || @@ -1591,6 +1592,7 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) esw->enabled_vports, esw->mode); mc_promisc = esw->mc_promisc; + nvports = esw->enabled_vports; for (i = 0; i < esw->total_vports; i++) esw_disable_vport(esw, i); @@ -1600,8 +1602,8 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) if (esw->mode == SRIOV_LEGACY) esw_destroy_legacy_fdb_table(esw); - else - esw_destroy_offloads_fdb_table(esw); + else if (esw->mode == SRIOV_OFFLOADS) + esw_offloads_cleanup(esw, nvports); esw->mode = SRIOV_NONE; /* VPORT 0 (PF) must be enabled back with non-sriov configuration */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index e1727a9..312b6f3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -112,7 +112,7 @@ out: #define MAX_PF_SQ 256 -int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports) +static int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports) { int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); struct mlx5_core_dev *dev = esw->dev; @@ -200,7 +200,7 @@ ns_err: return err; } -void esw_destroy_offloads_fdb_table(struct mlx5_eswitch *esw) +static void esw_destroy_offloads_fdb_table(struct mlx5_eswitch *esw) { if (!esw->fdb_table.fdb) return; @@ -329,12 +329,125 @@ out: return flow_rule; } +static int esw_offloads_start(struct mlx5_eswitch *esw) +{ + int err, num_vfs = esw->dev->priv.sriov.num_vfs; + + if (esw->mode != SRIOV_LEGACY) { + esw_warn(esw->dev, "Can't set offloads mode, SRIOV legacy not enabled\n"); + return -EINVAL; + } + + mlx5_eswitch_disable_sriov(esw); + err = mlx5_eswitch_enable_sriov(esw, num_vfs, SRIOV_OFFLOADS); + if (err) + esw_warn(esw->dev, "Failed set eswitch to offloads, err %d\n", err); + return err; +} + +int esw_offloads_init(struct mlx5_eswitch *esw, int nvports) +{ + int err; + + err = esw_create_offloads_fdb_table(esw, nvports); + if (err) + return err; + + err = esw_create_offloads_table(esw); + if (err) + goto create_ft_err; + + err = esw_create_vport_rx_group(esw); + if (err) + goto create_fg_err; + + return 0; + +create_fg_err: + esw_destroy_offloads_table(esw); + +create_ft_err: + esw_destroy_offloads_fdb_table(esw); + return err; +} + +static int esw_offloads_stop(struct mlx5_eswitch *esw) +{ + int err, num_vfs = esw->dev->priv.sriov.num_vfs; + + mlx5_eswitch_disable_sriov(esw); + err = mlx5_eswitch_enable_sriov(esw, num_vfs, SRIOV_LEGACY); + if (err) + esw_warn(esw->dev, "Failed set eswitch legacy mode. err %d\n", err); + + return err; +} + +void esw_offloads_cleanup(struct mlx5_eswitch *esw, int nvports) +{ + esw_destroy_vport_rx_group(esw); + esw_destroy_offloads_table(esw); + esw_destroy_offloads_fdb_table(esw); +} + +static int mlx5_esw_mode_from_devlink(u16 mode, u16 *mlx5_mode) +{ + switch (mode) { + case DEVLINK_ESWITCH_MODE_LEGACY: + *mlx5_mode = SRIOV_LEGACY; + break; + case DEVLINK_ESWITCH_MODE_SWITCHDEV: + *mlx5_mode = SRIOV_OFFLOADS; + break; + default: + return -EINVAL; + } + + return 0; +} + int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode) { - return -EOPNOTSUPP; + struct mlx5_core_dev *dev; + u16 cur_mlx5_mode, mlx5_mode = 0; + + dev = devlink_priv(devlink); + + if (!MLX5_CAP_GEN(dev, vport_group_manager)) + return -EOPNOTSUPP; + + cur_mlx5_mode = dev->priv.eswitch->mode; + + if (cur_mlx5_mode == SRIOV_NONE) + return -EOPNOTSUPP; + + if (mlx5_esw_mode_from_devlink(mode, &mlx5_mode)) + return -EINVAL; + + if (cur_mlx5_mode == mlx5_mode) + return 0; + + if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV) + return esw_offloads_start(dev->priv.eswitch); + else if (mode == DEVLINK_ESWITCH_MODE_LEGACY) + return esw_offloads_stop(dev->priv.eswitch); + else + return -EINVAL; } int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode) { - return -EOPNOTSUPP; + struct mlx5_core_dev *dev; + + dev = devlink_priv(devlink); + + if (!MLX5_CAP_GEN(dev, vport_group_manager)) + return -EOPNOTSUPP; + + if (dev->priv.eswitch->mode == SRIOV_NONE) + return -EOPNOTSUPP; + + *mode = dev->priv.eswitch->mode; + + return 0; } -- cgit v0.10.2 From b50d292b4399f4eb11e82d0430aacf62dd5d5365 Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Fri, 1 Jul 2016 14:51:04 +0300 Subject: net/mlx5e: Create NIC global resources only once To allow creating more than one netdev over the same PCI function, we change the driver such that global NIC resources are created once and later be shared amongst all the mlx5e netdevs running over that port. Move the CQ UAR, PD (pdn), Transport Domain (tdn), MKey resources from being kept in the mlx5e priv part to a new resources structure (mlx5e_resources) placed under the mlx5_core device. This patch doesn't add any new functionality. Signed-off-by: Hadar Hen Zion Reviewed-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 96f1826..9b14dad 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -6,8 +6,8 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ fs_counters.o rl.o mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o eswitch_offloads.o \ - en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \ - en_rx_am.o en_txrx.o en_clock.o vxlan.o en_tc.o \ - en_arfs.o + en_main.o en_common.o en_fs.o en_ethtool.o en_tx.o \ + en_rx.o en_rx_am.o en_txrx.o en_clock.o vxlan.o \ + en_tc.o en_arfs.o mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index b97511b..3226b92 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -570,10 +570,6 @@ struct mlx5e_priv { unsigned long state; struct mutex state_lock; /* Protects Interface state */ - struct mlx5_uar cq_uar; - u32 pdn; - u32 tdn; - struct mlx5_core_mkey mkey; struct mlx5_core_mkey umr_mkey; struct mlx5e_rq drop_rq; @@ -788,5 +784,7 @@ int mlx5e_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, #endif u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev); +int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev); +void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev); #endif /* __MLX5_EN_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c new file mode 100644 index 0000000..33b3732 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -0,0 +1,112 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "en.h" + +/* mlx5e global resources should be placed in this file. + * Global resources are common to all the netdevices crated on the same nic. + */ + +static int mlx5e_create_mkey(struct mlx5_core_dev *mdev, u32 pdn, + struct mlx5_core_mkey *mkey) +{ + struct mlx5_create_mkey_mbox_in *in; + int err; + + in = mlx5_vzalloc(sizeof(*in)); + if (!in) + return -ENOMEM; + + in->seg.flags = MLX5_PERM_LOCAL_WRITE | + MLX5_PERM_LOCAL_READ | + MLX5_ACCESS_MODE_PA; + in->seg.flags_pd = cpu_to_be32(pdn | MLX5_MKEY_LEN64); + in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); + + err = mlx5_core_create_mkey(mdev, mkey, in, sizeof(*in), NULL, NULL, + NULL); + + kvfree(in); + + return err; +} + +int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev) +{ + struct mlx5e_resources *res = &mdev->mlx5e_res; + int err; + + err = mlx5_alloc_map_uar(mdev, &res->cq_uar, false); + if (err) { + mlx5_core_err(mdev, "alloc_map uar failed, %d\n", err); + return err; + } + + err = mlx5_core_alloc_pd(mdev, &res->pdn); + if (err) { + mlx5_core_err(mdev, "alloc pd failed, %d\n", err); + goto err_unmap_free_uar; + } + + err = mlx5_core_alloc_transport_domain(mdev, &res->td.tdn); + if (err) { + mlx5_core_err(mdev, "alloc td failed, %d\n", err); + goto err_dealloc_pd; + } + + err = mlx5e_create_mkey(mdev, res->pdn, &res->mkey); + if (err) { + mlx5_core_err(mdev, "create mkey failed, %d\n", err); + goto err_dealloc_transport_domain; + } + + return 0; + +err_dealloc_transport_domain: + mlx5_core_dealloc_transport_domain(mdev, res->td.tdn); +err_dealloc_pd: + mlx5_core_dealloc_pd(mdev, res->pdn); +err_unmap_free_uar: + mlx5_unmap_free_uar(mdev, &res->cq_uar); + + return err; +} + +void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev) +{ + struct mlx5e_resources *res = &mdev->mlx5e_res; + + mlx5_core_destroy_mkey(mdev, &res->mkey); + mlx5_core_dealloc_transport_domain(mdev, res->td.tdn); + mlx5_core_dealloc_pd(mdev, res->pdn); + mlx5_unmap_free_uar(mdev, &res->cq_uar); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index a64ce5d..9b2e2b2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -858,7 +858,7 @@ static int mlx5e_create_cq(struct mlx5e_channel *c, mcq->comp = mlx5e_completion_event; mcq->event = mlx5e_cq_error_event; mcq->irqn = irqn; - mcq->uar = &priv->cq_uar; + mcq->uar = &mdev->mlx5e_res.cq_uar; for (i = 0; i < mlx5_cqwq_get_size(&cq->wq); i++) { struct mlx5_cqe64 *cqe = mlx5_cqwq_get_wqe(&cq->wq, i); @@ -1136,7 +1136,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->cpu = cpu; c->pdev = &priv->mdev->pdev->dev; c->netdev = priv->netdev; - c->mkey_be = cpu_to_be32(priv->mkey.key); + c->mkey_be = cpu_to_be32(priv->mdev->mlx5e_res.mkey.key); c->num_tc = priv->params.num_tc; if (priv->params.rx_am_enabled) @@ -1252,7 +1252,7 @@ static void mlx5e_build_rq_param(struct mlx5e_priv *priv, MLX5_SET(wq, wq, end_padding_mode, MLX5_WQ_END_PAD_MODE_ALIGN); MLX5_SET(wq, wq, log_wq_stride, ilog2(sizeof(struct mlx5e_rx_wqe))); MLX5_SET(wq, wq, log_wq_sz, priv->params.log_rq_size); - MLX5_SET(wq, wq, pd, priv->pdn); + MLX5_SET(wq, wq, pd, priv->mdev->mlx5e_res.pdn); MLX5_SET(rqc, rqc, counter_set_id, priv->q_counter); param->wq.buf_numa_node = dev_to_node(&priv->mdev->pdev->dev); @@ -1277,7 +1277,7 @@ static void mlx5e_build_sq_param_common(struct mlx5e_priv *priv, void *wq = MLX5_ADDR_OF(sqc, sqc, wq); MLX5_SET(wq, wq, log_wq_stride, ilog2(MLX5_SEND_WQE_BB)); - MLX5_SET(wq, wq, pd, priv->pdn); + MLX5_SET(wq, wq, pd, priv->mdev->mlx5e_res.pdn); param->wq.buf_numa_node = dev_to_node(&priv->mdev->pdev->dev); } @@ -1299,7 +1299,7 @@ static void mlx5e_build_common_cq_param(struct mlx5e_priv *priv, { void *cqc = param->cqc; - MLX5_SET(cqc, cqc, uar_page, priv->cq_uar.index); + MLX5_SET(cqc, cqc, uar_page, priv->mdev->mlx5e_res.cq_uar.index); } static void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, @@ -1920,7 +1920,7 @@ static int mlx5e_create_drop_cq(struct mlx5e_priv *priv, mcq->comp = mlx5e_completion_event; mcq->event = mlx5e_cq_error_event; mcq->irqn = irqn; - mcq->uar = &priv->cq_uar; + mcq->uar = &mdev->mlx5e_res.cq_uar; cq->priv = priv; @@ -1986,7 +1986,7 @@ static int mlx5e_create_tis(struct mlx5e_priv *priv, int tc) memset(in, 0, sizeof(in)); MLX5_SET(tisc, tisc, prio, tc << 1); - MLX5_SET(tisc, tisc, transport_domain, priv->tdn); + MLX5_SET(tisc, tisc, transport_domain, mdev->mlx5e_res.td.tdn); return mlx5_core_create_tis(mdev, in, sizeof(in), &priv->tisn[tc]); } @@ -2029,7 +2029,7 @@ static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, { void *hfso = MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_outer); - MLX5_SET(tirc, tirc, transport_domain, priv->tdn); + MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn); #define MLX5_HASH_IP (MLX5_HASH_FIELD_SEL_SRC_IP |\ MLX5_HASH_FIELD_SEL_DST_IP) @@ -2136,7 +2136,7 @@ static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, static void mlx5e_build_direct_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, u32 rqtn) { - MLX5_SET(tirc, tirc, transport_domain, priv->tdn); + MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn); mlx5e_build_tir_ctx_lro(tirc, priv); @@ -3082,31 +3082,6 @@ static void mlx5e_build_netdev(struct net_device *netdev) mlx5e_set_netdev_dev_addr(netdev); } -static int mlx5e_create_mkey(struct mlx5e_priv *priv, u32 pdn, - struct mlx5_core_mkey *mkey) -{ - struct mlx5_core_dev *mdev = priv->mdev; - struct mlx5_create_mkey_mbox_in *in; - int err; - - in = mlx5_vzalloc(sizeof(*in)); - if (!in) - return -ENOMEM; - - in->seg.flags = MLX5_PERM_LOCAL_WRITE | - MLX5_PERM_LOCAL_READ | - MLX5_ACCESS_MODE_PA; - in->seg.flags_pd = cpu_to_be32(pdn | MLX5_MKEY_LEN64); - in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); - - err = mlx5_core_create_mkey(mdev, mkey, in, sizeof(*in), NULL, NULL, - NULL); - - kvfree(in); - - return err; -} - static void mlx5e_create_q_counter(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; @@ -3149,7 +3124,7 @@ static int mlx5e_create_umr_mkey(struct mlx5e_priv *priv) MLX5_ACCESS_MODE_MTT; mkc->qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); - mkc->flags_pd = cpu_to_be32(priv->pdn); + mkc->flags_pd = cpu_to_be32(mdev->mlx5e_res.pdn); mkc->len = cpu_to_be64(npages << PAGE_SHIFT); mkc->xlt_oct_size = cpu_to_be32(mlx5e_get_mtt_octw(npages)); mkc->log2_page_size = PAGE_SHIFT; @@ -3169,9 +3144,6 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev) int nch = mlx5e_get_max_num_channels(mdev); int err; - if (mlx5e_check_required_hca_cap(mdev)) - return NULL; - netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv), nch * MLX5E_MAX_NUM_TC, nch); @@ -3191,34 +3163,10 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev) if (!priv->wq) goto err_free_netdev; - err = mlx5_alloc_map_uar(mdev, &priv->cq_uar, false); - if (err) { - mlx5_core_err(mdev, "alloc_map uar failed, %d\n", err); - goto err_destroy_wq; - } - - err = mlx5_core_alloc_pd(mdev, &priv->pdn); - if (err) { - mlx5_core_err(mdev, "alloc pd failed, %d\n", err); - goto err_unmap_free_uar; - } - - err = mlx5_core_alloc_transport_domain(mdev, &priv->tdn); - if (err) { - mlx5_core_err(mdev, "alloc td failed, %d\n", err); - goto err_dealloc_pd; - } - - err = mlx5e_create_mkey(priv, priv->pdn, &priv->mkey); - if (err) { - mlx5_core_err(mdev, "create mkey failed, %d\n", err); - goto err_dealloc_transport_domain; - } - err = mlx5e_create_umr_mkey(priv); if (err) { mlx5_core_err(mdev, "create umr mkey failed, %d\n", err); - goto err_destroy_mkey; + goto err_destroy_wq; } err = mlx5e_create_tises(priv); @@ -3304,18 +3252,6 @@ err_destroy_tises: err_destroy_umr_mkey: mlx5_core_destroy_mkey(mdev, &priv->umr_mkey); -err_destroy_mkey: - mlx5_core_destroy_mkey(mdev, &priv->mkey); - -err_dealloc_transport_domain: - mlx5_core_dealloc_transport_domain(mdev, priv->tdn); - -err_dealloc_pd: - mlx5_core_dealloc_pd(mdev, priv->pdn); - -err_unmap_free_uar: - mlx5_unmap_free_uar(mdev, &priv->cq_uar); - err_destroy_wq: destroy_workqueue(priv->wq); @@ -3325,9 +3261,27 @@ err_free_netdev: return NULL; } -static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv) +static void *mlx5e_add(struct mlx5_core_dev *mdev) +{ + void *ret; + + if (mlx5e_check_required_hca_cap(mdev)) + return NULL; + + if (mlx5e_create_mdev_resources(mdev)) + return NULL; + + ret = mlx5e_create_netdev(mdev); + if (!ret) { + mlx5e_destroy_mdev_resources(mdev); + return NULL; + } + return ret; +} + +static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, + struct mlx5e_priv *priv) { - struct mlx5e_priv *priv = vpriv; struct net_device *netdev = priv->netdev; set_bit(MLX5E_STATE_DESTROYING, &priv->state); @@ -3351,10 +3305,6 @@ static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv) mlx5e_close_drop_rq(priv); mlx5e_destroy_tises(priv); mlx5_core_destroy_mkey(priv->mdev, &priv->umr_mkey); - mlx5_core_destroy_mkey(priv->mdev, &priv->mkey); - mlx5_core_dealloc_transport_domain(priv->mdev, priv->tdn); - mlx5_core_dealloc_pd(priv->mdev, priv->pdn); - mlx5_unmap_free_uar(priv->mdev, &priv->cq_uar); cancel_delayed_work_sync(&priv->update_stats_work); destroy_workqueue(priv->wq); @@ -3362,6 +3312,14 @@ static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv) free_netdev(netdev); } +static void mlx5e_remove(struct mlx5_core_dev *mdev, void *vpriv) +{ + struct mlx5e_priv *priv = vpriv; + + mlx5e_destroy_netdev(mdev, priv); + mlx5e_destroy_mdev_resources(mdev); +} + static void *mlx5e_get_netdev(void *vpriv) { struct mlx5e_priv *priv = vpriv; @@ -3370,8 +3328,8 @@ static void *mlx5e_get_netdev(void *vpriv) } static struct mlx5_interface mlx5e_interface = { - .add = mlx5e_create_netdev, - .remove = mlx5e_destroy_netdev, + .add = mlx5e_add, + .remove = mlx5e_remove, .event = mlx5e_async_event, .protocol = MLX5_INTERFACE_PROTOCOL_ETH, .get_dev = mlx5e_get_netdev, diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 46260fd..e22b345 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -578,6 +578,18 @@ enum mlx5_pci_status { MLX5_PCI_STATUS_ENABLED, }; +struct mlx5_td { + struct list_head tirs_list; + u32 tdn; +}; + +struct mlx5e_resources { + struct mlx5_uar cq_uar; + u32 pdn; + struct mlx5_td td; + struct mlx5_core_mkey mkey; +}; + struct mlx5_core_dev { struct pci_dev *pdev; /* sync pci state */ @@ -602,6 +614,7 @@ struct mlx5_core_dev { struct mlx5_profile *profile; atomic_t num_qps; u32 issi; + struct mlx5e_resources mlx5e_res; #ifdef CONFIG_RFS_ACCEL struct cpu_rmap *rmap; #endif -- cgit v0.10.2 From 724b2aa15126d9e24b36650c5cad9cf468c20785 Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Fri, 1 Jul 2016 14:51:05 +0300 Subject: net/mlx5e: TIRs management refactoring The current refresh tirs self loopback mechanism, refreshes all the tirs belonging to the same mlx5e instance to prevent self loopback by packets sent over any ring of that instance. This mechanism relies on all the tirs/tises of an instance to be created with the same transport domain number (tdn). Change the driver to refresh all the tirs created under the same tdn regardless of which mlx5e netdev instance they belong to. This behaviour is needed for introducing new mlx5e instances which serve to represent SRIOV VFs. The representors and the PF share vport used for E-Switch management, and we want to avoid NIC level HW loopback between them, e.g when sending broadcast packets. To achieve that, both the representors and the PF NIC will share the tdn. This patch doesn't add any new functionality. Signed-off-by: Hadar Hen Zion Reviewed-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 3226b92..8dad50c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -552,9 +552,10 @@ struct mlx5e_flow_steering { struct mlx5e_arfs_tables arfs; }; -struct mlx5e_direct_tir { +struct mlx5e_tir { u32 tirn; u32 rqtn; + struct list_head list; }; enum { @@ -576,8 +577,8 @@ struct mlx5e_priv { struct mlx5e_channel **channel; u32 tisn[MLX5E_MAX_NUM_TC]; u32 indir_rqtn; - u32 indir_tirn[MLX5E_NUM_INDIR_TIRS]; - struct mlx5e_direct_tir direct_tir[MLX5E_MAX_NUM_CHANNELS]; + struct mlx5e_tir indir_tir[MLX5E_NUM_INDIR_TIRS]; + struct mlx5e_tir direct_tir[MLX5E_MAX_NUM_CHANNELS]; u32 tx_rates[MLX5E_MAX_NUM_SQS]; struct mlx5e_flow_steering fs; @@ -784,7 +785,12 @@ int mlx5e_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, #endif u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev); +int mlx5e_create_tir(struct mlx5_core_dev *mdev, + struct mlx5e_tir *tir, u32 *in, int inlen); +void mlx5e_destroy_tir(struct mlx5_core_dev *mdev, + struct mlx5e_tir *tir); int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev); void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev); +int mlx5e_refresh_tirs_self_loopback_enable(struct mlx5_core_dev *mdev); #endif /* __MLX5_EN_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c index 3515e78..10f18d4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c @@ -93,14 +93,14 @@ static enum mlx5e_traffic_types arfs_get_tt(enum arfs_type type) static int arfs_disable(struct mlx5e_priv *priv) { struct mlx5_flow_destination dest; - u32 *tirn = priv->indir_tirn; + struct mlx5e_tir *tir = priv->indir_tir; int err = 0; int tt; int i; dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; for (i = 0; i < ARFS_NUM_TYPES; i++) { - dest.tir_num = tirn[i]; + dest.tir_num = tir[i].tirn; tt = arfs_get_tt(i); /* Modify ttc rules destination to bypass the aRFS tables*/ err = mlx5_modify_rule_destination(priv->fs.ttc.rules[tt], @@ -176,7 +176,7 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, struct arfs_table *arfs_t = &priv->fs.arfs.arfs_tables[type]; struct mlx5_flow_destination dest; u8 match_criteria_enable = 0; - u32 *tirn = priv->indir_tirn; + struct mlx5e_tir *tir = priv->indir_tir; u32 *match_criteria; u32 *match_value; int err = 0; @@ -192,16 +192,16 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; switch (type) { case ARFS_IPV4_TCP: - dest.tir_num = tirn[MLX5E_TT_IPV4_TCP]; + dest.tir_num = tir[MLX5E_TT_IPV4_TCP].tirn; break; case ARFS_IPV4_UDP: - dest.tir_num = tirn[MLX5E_TT_IPV4_UDP]; + dest.tir_num = tir[MLX5E_TT_IPV4_UDP].tirn; break; case ARFS_IPV6_TCP: - dest.tir_num = tirn[MLX5E_TT_IPV6_TCP]; + dest.tir_num = tir[MLX5E_TT_IPV6_TCP].tirn; break; case ARFS_IPV6_UDP: - dest.tir_num = tirn[MLX5E_TT_IPV6_UDP]; + dest.tir_num = tir[MLX5E_TT_IPV6_UDP].tirn; break; default: err = -EINVAL; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c index 33b3732..673043c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -36,6 +36,27 @@ * Global resources are common to all the netdevices crated on the same nic. */ +int mlx5e_create_tir(struct mlx5_core_dev *mdev, + struct mlx5e_tir *tir, u32 *in, int inlen) +{ + int err; + + err = mlx5_core_create_tir(mdev, in, inlen, &tir->tirn); + if (err) + return err; + + list_add(&tir->list, &mdev->mlx5e_res.td.tirs_list); + + return 0; +} + +void mlx5e_destroy_tir(struct mlx5_core_dev *mdev, + struct mlx5e_tir *tir) +{ + mlx5_core_destroy_tir(mdev, tir->tirn); + list_del(&tir->list); +} + static int mlx5e_create_mkey(struct mlx5_core_dev *mdev, u32 pdn, struct mlx5_core_mkey *mkey) { @@ -89,6 +110,8 @@ int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev) goto err_dealloc_transport_domain; } + INIT_LIST_HEAD(&mdev->mlx5e_res.td.tirs_list); + return 0; err_dealloc_transport_domain: @@ -110,3 +133,28 @@ void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev) mlx5_core_dealloc_pd(mdev, res->pdn); mlx5_unmap_free_uar(mdev, &res->cq_uar); } + +int mlx5e_refresh_tirs_self_loopback_enable(struct mlx5_core_dev *mdev) +{ + struct mlx5e_tir *tir; + void *in; + int inlen; + int err; + + inlen = MLX5_ST_SZ_BYTES(modify_tir_in); + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + + MLX5_SET(modify_tir_in, in, bitmask.self_lb_en, 1); + + list_for_each_entry(tir, &mdev->mlx5e_res.td.tirs_list, list) { + err = mlx5_core_modify_tir(mdev, tir->tirn, in, inlen); + if (err) + return err; + } + + kvfree(in); + + return 0; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index b29684d..5b88967 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -876,7 +876,7 @@ static void mlx5e_modify_tirs_hash(struct mlx5e_priv *priv, void *in, int inlen) mlx5e_build_tir_ctx_hash(tirc, priv); for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) - mlx5_core_modify_tir(mdev, priv->indir_tirn[i], in, inlen); + mlx5_core_modify_tir(mdev, priv->indir_tir[i].tirn, in, inlen); } static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c index b327400..606e69b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c @@ -655,7 +655,7 @@ static int mlx5e_generate_ttc_table_rules(struct mlx5e_priv *priv) if (tt == MLX5E_TT_ANY) dest.tir_num = priv->direct_tir[0].tirn; else - dest.tir_num = priv->indir_tirn[tt]; + dest.tir_num = priv->indir_tir[tt].tirn; rules[tt] = mlx5e_generate_ttc_rule(priv, ft, &dest, ttc_rules[tt].etype, ttc_rules[tt].proto); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 9b2e2b2..30efa8a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -1661,7 +1661,7 @@ static int mlx5e_modify_tirs_lro(struct mlx5e_priv *priv) mlx5e_build_tir_ctx_lro(tirc, priv); for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { - err = mlx5_core_modify_tir(mdev, priv->indir_tirn[tt], in, + err = mlx5_core_modify_tir(mdev, priv->indir_tir[tt].tirn, in, inlen); if (err) goto free_in; @@ -1680,40 +1680,6 @@ free_in: return err; } -static int mlx5e_refresh_tirs_self_loopback_enable(struct mlx5e_priv *priv) -{ - void *in; - int inlen; - int err; - int i; - - inlen = MLX5_ST_SZ_BYTES(modify_tir_in); - in = mlx5_vzalloc(inlen); - if (!in) - return -ENOMEM; - - MLX5_SET(modify_tir_in, in, bitmask.self_lb_en, 1); - - for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) { - err = mlx5_core_modify_tir(priv->mdev, priv->indir_tirn[i], in, - inlen); - if (err) - return err; - } - - for (i = 0; i < priv->params.num_channels; i++) { - err = mlx5_core_modify_tir(priv->mdev, - priv->direct_tir[i].tirn, in, - inlen); - if (err) - return err; - } - - kvfree(in); - - return 0; -} - static int mlx5e_set_mtu(struct mlx5e_priv *priv, u16 mtu) { struct mlx5_core_dev *mdev = priv->mdev; @@ -1804,7 +1770,7 @@ int mlx5e_open_locked(struct net_device *netdev) goto err_clear_state_opened_flag; } - err = mlx5e_refresh_tirs_self_loopback_enable(priv); + err = mlx5e_refresh_tirs_self_loopback_enable(priv->mdev); if (err) { netdev_err(netdev, "%s: mlx5e_refresh_tirs_self_loopback_enable failed, %d\n", __func__, err); @@ -2148,9 +2114,9 @@ static void mlx5e_build_direct_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, static int mlx5e_create_tirs(struct mlx5e_priv *priv) { int nch = mlx5e_get_max_num_channels(priv->mdev); + struct mlx5e_tir *tir; void *tirc; int inlen; - u32 *tirn; int err; u32 *in; int ix; @@ -2164,10 +2130,10 @@ static int mlx5e_create_tirs(struct mlx5e_priv *priv) /* indirect tirs */ for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { memset(in, 0, inlen); - tirn = &priv->indir_tirn[tt]; + tir = &priv->indir_tir[tt]; tirc = MLX5_ADDR_OF(create_tir_in, in, ctx); mlx5e_build_indir_tir_ctx(priv, tirc, tt); - err = mlx5_core_create_tir(priv->mdev, in, inlen, tirn); + err = mlx5e_create_tir(priv->mdev, tir, in, inlen); if (err) goto err_destroy_tirs; } @@ -2175,11 +2141,11 @@ static int mlx5e_create_tirs(struct mlx5e_priv *priv) /* direct tirs */ for (ix = 0; ix < nch; ix++) { memset(in, 0, inlen); - tirn = &priv->direct_tir[ix].tirn; + tir = &priv->direct_tir[ix]; tirc = MLX5_ADDR_OF(create_tir_in, in, ctx); mlx5e_build_direct_tir_ctx(priv, tirc, priv->direct_tir[ix].rqtn); - err = mlx5_core_create_tir(priv->mdev, in, inlen, tirn); + err = mlx5e_create_tir(priv->mdev, tir, in, inlen); if (err) goto err_destroy_ch_tirs; } @@ -2190,11 +2156,11 @@ static int mlx5e_create_tirs(struct mlx5e_priv *priv) err_destroy_ch_tirs: for (ix--; ix >= 0; ix--) - mlx5_core_destroy_tir(priv->mdev, priv->direct_tir[ix].tirn); + mlx5e_destroy_tir(priv->mdev, &priv->direct_tir[ix]); err_destroy_tirs: for (tt--; tt >= 0; tt--) - mlx5_core_destroy_tir(priv->mdev, priv->indir_tirn[tt]); + mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[tt]); kvfree(in); @@ -2207,10 +2173,10 @@ static void mlx5e_destroy_tirs(struct mlx5e_priv *priv) int i; for (i = 0; i < nch; i++) - mlx5_core_destroy_tir(priv->mdev, priv->direct_tir[i].tirn); + mlx5e_destroy_tir(priv->mdev, &priv->direct_tir[i]); for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) - mlx5_core_destroy_tir(priv->mdev, priv->indir_tirn[i]); + mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[i]); } int mlx5e_modify_rqs_vsd(struct mlx5e_priv *priv, bool vsd) -- cgit v0.10.2 From 398f33511e97aad7f259e864a1596fc8ef559dc1 Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Fri, 1 Jul 2016 14:51:06 +0300 Subject: net/mlx5e: Mark enabled RQTs instances explicitly In the current driver implementation two types of receive queue tables (RQTs) are in use - direct and indirect. Change the driver to mark each new created RQT (direct or indirect) as "enabled". This behaviour is needed for introducing new mlx5e instances which serve to represent SRIOV VFs. The VF representors will have only one type of RQTs (direct). An "enabled" flag is added to each RQT to allow better handling and code sharing between the representors and the nic netdevices. This patch doesn't add any new functionality. Signed-off-by: Hadar Hen Zion Reviewed-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 8dad50c..91c6bbe 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -552,10 +552,15 @@ struct mlx5e_flow_steering { struct mlx5e_arfs_tables arfs; }; -struct mlx5e_tir { - u32 tirn; +struct mlx5e_rqt { u32 rqtn; - struct list_head list; + bool enabled; +}; + +struct mlx5e_tir { + u32 tirn; + struct mlx5e_rqt rqt; + struct list_head list; }; enum { @@ -576,7 +581,7 @@ struct mlx5e_priv { struct mlx5e_channel **channel; u32 tisn[MLX5E_MAX_NUM_TC]; - u32 indir_rqtn; + struct mlx5e_rqt indir_rqt; struct mlx5e_tir indir_tir[MLX5E_NUM_INDIR_TIRS]; struct mlx5e_tir direct_tir[MLX5E_MAX_NUM_CHANNELS]; u32 tx_rates[MLX5E_MAX_NUM_SQS]; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 5b88967..7e61ffa 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -898,7 +898,7 @@ static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir, mutex_lock(&priv->state_lock); if (indir) { - u32 rqtn = priv->indir_rqtn; + u32 rqtn = priv->indir_rqt.rqtn; memcpy(priv->params.indirection_rqt, indir, sizeof(priv->params.indirection_rqt)); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 30efa8a..7f1f1ec 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -1486,7 +1486,8 @@ static void mlx5e_fill_direct_rqt_rqn(struct mlx5e_priv *priv, void *rqtc, MLX5_SET(rqtc, rqtc, rq_num[0], rqn); } -static int mlx5e_create_rqt(struct mlx5e_priv *priv, int sz, int ix, u32 *rqtn) +static int mlx5e_create_rqt(struct mlx5e_priv *priv, int sz, + int ix, struct mlx5e_rqt *rqt) { struct mlx5_core_dev *mdev = priv->mdev; void *rqtc; @@ -1509,34 +1510,37 @@ static int mlx5e_create_rqt(struct mlx5e_priv *priv, int sz, int ix, u32 *rqtn) else mlx5e_fill_direct_rqt_rqn(priv, rqtc, ix); - err = mlx5_core_create_rqt(mdev, in, inlen, rqtn); + err = mlx5_core_create_rqt(mdev, in, inlen, &rqt->rqtn); + if (!err) + rqt->enabled = true; kvfree(in); return err; } -static void mlx5e_destroy_rqt(struct mlx5e_priv *priv, u32 rqtn) +static void mlx5e_destroy_rqt(struct mlx5e_priv *priv, struct mlx5e_rqt *rqt) { - mlx5_core_destroy_rqt(priv->mdev, rqtn); + rqt->enabled = false; + mlx5_core_destroy_rqt(priv->mdev, rqt->rqtn); } static int mlx5e_create_rqts(struct mlx5e_priv *priv) { int nch = mlx5e_get_max_num_channels(priv->mdev); - u32 *rqtn; + struct mlx5e_rqt *rqt; int err; int ix; /* Indirect RQT */ - rqtn = &priv->indir_rqtn; - err = mlx5e_create_rqt(priv, MLX5E_INDIR_RQT_SIZE, 0, rqtn); + rqt = &priv->indir_rqt; + err = mlx5e_create_rqt(priv, MLX5E_INDIR_RQT_SIZE, 0, rqt); if (err) return err; /* Direct RQTs */ for (ix = 0; ix < nch; ix++) { - rqtn = &priv->direct_tir[ix].rqtn; - err = mlx5e_create_rqt(priv, 1 /*size */, ix, rqtn); + rqt = &priv->direct_tir[ix].rqt; + err = mlx5e_create_rqt(priv, 1 /*size */, ix, rqt); if (err) goto err_destroy_rqts; } @@ -1545,9 +1549,9 @@ static int mlx5e_create_rqts(struct mlx5e_priv *priv) err_destroy_rqts: for (ix--; ix >= 0; ix--) - mlx5e_destroy_rqt(priv, priv->direct_tir[ix].rqtn); + mlx5e_destroy_rqt(priv, &priv->direct_tir[ix].rqt); - mlx5e_destroy_rqt(priv, priv->indir_rqtn); + mlx5e_destroy_rqt(priv, &priv->indir_rqt); return err; } @@ -1558,9 +1562,9 @@ static void mlx5e_destroy_rqts(struct mlx5e_priv *priv) int i; for (i = 0; i < nch; i++) - mlx5e_destroy_rqt(priv, priv->direct_tir[i].rqtn); + mlx5e_destroy_rqt(priv, &priv->direct_tir[i].rqt); - mlx5e_destroy_rqt(priv, priv->indir_rqtn); + mlx5e_destroy_rqt(priv, &priv->indir_rqt); } int mlx5e_redirect_rqt(struct mlx5e_priv *priv, u32 rqtn, int sz, int ix) @@ -1598,10 +1602,15 @@ static void mlx5e_redirect_rqts(struct mlx5e_priv *priv) u32 rqtn; int ix; - rqtn = priv->indir_rqtn; - mlx5e_redirect_rqt(priv, rqtn, MLX5E_INDIR_RQT_SIZE, 0); + if (priv->indir_rqt.enabled) { + rqtn = priv->indir_rqt.rqtn; + mlx5e_redirect_rqt(priv, rqtn, MLX5E_INDIR_RQT_SIZE, 0); + } + for (ix = 0; ix < priv->params.num_channels; ix++) { - rqtn = priv->direct_tir[ix].rqtn; + if (!priv->direct_tir[ix].rqt.enabled) + continue; + rqtn = priv->direct_tir[ix].rqt.rqtn; mlx5e_redirect_rqt(priv, rqtn, 1, ix); } } @@ -2012,7 +2021,7 @@ static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, mlx5e_build_tir_ctx_lro(tirc, priv); MLX5_SET(tirc, tirc, disp_type, MLX5_TIRC_DISP_TYPE_INDIRECT); - MLX5_SET(tirc, tirc, indirect_table, priv->indir_rqtn); + MLX5_SET(tirc, tirc, indirect_table, priv->indir_rqt.rqtn); mlx5e_build_tir_ctx_hash(tirc, priv); switch (tt) { @@ -2144,7 +2153,7 @@ static int mlx5e_create_tirs(struct mlx5e_priv *priv) tir = &priv->direct_tir[ix]; tirc = MLX5_ADDR_OF(create_tir_in, in, ctx); mlx5e_build_direct_tir_ctx(priv, tirc, - priv->direct_tir[ix].rqtn); + priv->direct_tir[ix].rqt.rqtn); err = mlx5e_create_tir(priv->mdev, tir, in, inlen); if (err) goto err_destroy_ch_tirs; -- cgit v0.10.2 From 6bfd390ba5466675f6f02f77a3e957bd4e6075ee Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Fri, 1 Jul 2016 14:51:07 +0300 Subject: net/mlx5e: Add support for multiple profiles To allow support in representor netdevices where we create more than one netdevice per NIC, add profiles to the mlx5e driver. The profiling allows for creation of mlx5e instances with different characteristics. Each profile implements its own behavior using set of function pointers defined in struct mlx5e_profile. This is done to allow for avoiding complex per profix branching in the code. Currently only the profile for the conventional NIC is implemented, which is of use when a netdev is created upon pci probe. This patch doesn't add any new functionality. Signed-off-by: Hadar Hen Zion Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 91c6bbe..edfc9be 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -568,6 +568,22 @@ enum { MLX5E_NIC_PRIO }; +struct mlx5e_profile { + void (*init)(struct mlx5_core_dev *mdev, + struct net_device *netdev, + const struct mlx5e_profile *profile); + void (*cleanup)(struct mlx5e_priv *priv); + int (*init_rx)(struct mlx5e_priv *priv); + void (*cleanup_rx)(struct mlx5e_priv *priv); + int (*init_tx)(struct mlx5e_priv *priv); + void (*cleanup_tx)(struct mlx5e_priv *priv); + void (*enable)(struct mlx5e_priv *priv); + void (*disable)(struct mlx5e_priv *priv); + void (*update_stats)(struct mlx5e_priv *priv); + int (*max_nch)(struct mlx5_core_dev *mdev); + int max_tc; +}; + struct mlx5e_priv { /* priv data path fields - start */ struct mlx5e_sq **txq_to_sq_map; @@ -601,6 +617,7 @@ struct mlx5e_priv { struct mlx5e_stats stats; struct mlx5e_tstamp tstamp; u16 q_counter; + const struct mlx5e_profile *profile; }; enum mlx5e_link_mode { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 7f1f1ec..3e22c5e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -233,7 +233,7 @@ static void mlx5e_update_stats_work(struct work_struct *work) update_stats_work); mutex_lock(&priv->state_lock); if (test_bit(MLX5E_STATE_OPENED, &priv->state)) { - mlx5e_update_stats(priv); + priv->profile->update_stats(priv); queue_delayed_work(priv->wq, dwork, msecs_to_jiffies(MLX5E_UPDATE_STATS_INTERVAL)); } @@ -1036,7 +1036,7 @@ static void mlx5e_build_channeltc_to_txq_map(struct mlx5e_priv *priv, int ix) { int i; - for (i = 0; i < MLX5E_MAX_NUM_TC; i++) + for (i = 0; i < priv->profile->max_tc; i++) priv->channeltc_to_txq_map[ix][i] = ix + i * priv->params.num_channels; } @@ -1524,21 +1524,20 @@ static void mlx5e_destroy_rqt(struct mlx5e_priv *priv, struct mlx5e_rqt *rqt) mlx5_core_destroy_rqt(priv->mdev, rqt->rqtn); } -static int mlx5e_create_rqts(struct mlx5e_priv *priv) +static int mlx5e_create_indirect_rqts(struct mlx5e_priv *priv) +{ + struct mlx5e_rqt *rqt = &priv->indir_rqt; + + return mlx5e_create_rqt(priv, MLX5E_INDIR_RQT_SIZE, 0, rqt); +} + +static int mlx5e_create_direct_rqts(struct mlx5e_priv *priv) { - int nch = mlx5e_get_max_num_channels(priv->mdev); struct mlx5e_rqt *rqt; int err; int ix; - /* Indirect RQT */ - rqt = &priv->indir_rqt; - err = mlx5e_create_rqt(priv, MLX5E_INDIR_RQT_SIZE, 0, rqt); - if (err) - return err; - - /* Direct RQTs */ - for (ix = 0; ix < nch; ix++) { + for (ix = 0; ix < priv->profile->max_nch(priv->mdev); ix++) { rqt = &priv->direct_tir[ix].rqt; err = mlx5e_create_rqt(priv, 1 /*size */, ix, rqt); if (err) @@ -1551,22 +1550,9 @@ err_destroy_rqts: for (ix--; ix >= 0; ix--) mlx5e_destroy_rqt(priv, &priv->direct_tir[ix].rqt); - mlx5e_destroy_rqt(priv, &priv->indir_rqt); - return err; } -static void mlx5e_destroy_rqts(struct mlx5e_priv *priv) -{ - int nch = mlx5e_get_max_num_channels(priv->mdev); - int i; - - for (i = 0; i < nch; i++) - mlx5e_destroy_rqt(priv, &priv->direct_tir[i].rqt); - - mlx5e_destroy_rqt(priv, &priv->indir_rqt); -} - int mlx5e_redirect_rqt(struct mlx5e_priv *priv, u32 rqtn, int sz, int ix) { struct mlx5_core_dev *mdev = priv->mdev; @@ -1676,7 +1662,7 @@ static int mlx5e_modify_tirs_lro(struct mlx5e_priv *priv) goto free_in; } - for (ix = 0; ix < mlx5e_get_max_num_channels(mdev); ix++) { + for (ix = 0; ix < priv->profile->max_nch(priv->mdev); ix++) { err = mlx5_core_modify_tir(mdev, priv->direct_tir[ix].tirn, in, inlen); if (err) @@ -1976,7 +1962,7 @@ static int mlx5e_create_tises(struct mlx5e_priv *priv) int err; int tc; - for (tc = 0; tc < MLX5E_MAX_NUM_TC; tc++) { + for (tc = 0; tc < priv->profile->max_tc; tc++) { err = mlx5e_create_tis(priv, tc); if (err) goto err_close_tises; @@ -1991,11 +1977,11 @@ err_close_tises: return err; } -static void mlx5e_destroy_tises(struct mlx5e_priv *priv) +static void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv) { int tc; - for (tc = 0; tc < MLX5E_MAX_NUM_TC; tc++) + for (tc = 0; tc < priv->profile->max_tc; tc++) mlx5e_destroy_tis(priv, tc); } @@ -2120,15 +2106,13 @@ static void mlx5e_build_direct_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, MLX5_SET(tirc, tirc, rx_hash_fn, MLX5_RX_HASH_FN_INVERTED_XOR8); } -static int mlx5e_create_tirs(struct mlx5e_priv *priv) +static int mlx5e_create_indirect_tirs(struct mlx5e_priv *priv) { - int nch = mlx5e_get_max_num_channels(priv->mdev); struct mlx5e_tir *tir; void *tirc; int inlen; int err; u32 *in; - int ix; int tt; inlen = MLX5_ST_SZ_BYTES(create_tir_in); @@ -2136,7 +2120,6 @@ static int mlx5e_create_tirs(struct mlx5e_priv *priv) if (!in) return -ENOMEM; - /* indirect tirs */ for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { memset(in, 0, inlen); tir = &priv->indir_tir[tt]; @@ -2147,7 +2130,34 @@ static int mlx5e_create_tirs(struct mlx5e_priv *priv) goto err_destroy_tirs; } - /* direct tirs */ + kvfree(in); + + return 0; + +err_destroy_tirs: + for (tt--; tt >= 0; tt--) + mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[tt]); + + kvfree(in); + + return err; +} + +static int mlx5e_create_direct_tirs(struct mlx5e_priv *priv) +{ + int nch = priv->profile->max_nch(priv->mdev); + struct mlx5e_tir *tir; + void *tirc; + int inlen; + int err; + u32 *in; + int ix; + + inlen = MLX5_ST_SZ_BYTES(create_tir_in); + in = mlx5_vzalloc(inlen); + if (!in) + return -ENOMEM; + for (ix = 0; ix < nch; ix++) { memset(in, 0, inlen); tir = &priv->direct_tir[ix]; @@ -2167,27 +2177,28 @@ err_destroy_ch_tirs: for (ix--; ix >= 0; ix--) mlx5e_destroy_tir(priv->mdev, &priv->direct_tir[ix]); -err_destroy_tirs: - for (tt--; tt >= 0; tt--) - mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[tt]); - kvfree(in); return err; } -static void mlx5e_destroy_tirs(struct mlx5e_priv *priv) +static void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv) { - int nch = mlx5e_get_max_num_channels(priv->mdev); int i; - for (i = 0; i < nch; i++) - mlx5e_destroy_tir(priv->mdev, &priv->direct_tir[i]); - for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[i]); } +static void mlx5e_destroy_direct_tirs(struct mlx5e_priv *priv) +{ + int nch = priv->profile->max_nch(priv->mdev); + int i; + + for (i = 0; i < nch; i++) + mlx5e_destroy_tir(priv->mdev, &priv->direct_tir[i]); +} + int mlx5e_modify_rqs_vsd(struct mlx5e_priv *priv, bool vsd) { int err = 0; @@ -2867,9 +2878,9 @@ void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC_FROM_CQE; } -static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, - struct net_device *netdev, - int num_channels) +static void mlx5e_build_nic_netdev_priv(struct mlx5_core_dev *mdev, + struct net_device *netdev, + const struct mlx5e_profile *profile) { struct mlx5e_priv *priv = netdev_priv(netdev); u32 link_speed = 0; @@ -2938,7 +2949,7 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, sizeof(priv->params.toeplitz_hash_key)); mlx5e_build_default_indir_rqt(mdev, priv->params.indirection_rqt, - MLX5E_INDIR_RQT_SIZE, num_channels); + MLX5E_INDIR_RQT_SIZE, profile->max_nch(mdev)); priv->params.lro_wqe_sz = MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ; @@ -2949,7 +2960,8 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev, priv->mdev = mdev; priv->netdev = netdev; - priv->params.num_channels = num_channels; + priv->params.num_channels = profile->max_nch(mdev); + priv->profile = profile; #ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_ets_init(priv); @@ -2974,7 +2986,7 @@ static void mlx5e_set_netdev_dev_addr(struct net_device *netdev) } } -static void mlx5e_build_netdev(struct net_device *netdev) +static void mlx5e_build_nic_netdev(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; @@ -3084,7 +3096,7 @@ static int mlx5e_create_umr_mkey(struct mlx5e_priv *priv) struct mlx5_mkey_seg *mkc; int inlen = sizeof(*in); u64 npages = - mlx5e_get_max_num_channels(mdev) * MLX5_CHANNEL_MAX_NUM_MTTS; + priv->profile->max_nch(mdev) * MLX5_CHANNEL_MAX_NUM_MTTS; int err; in = mlx5_vzalloc(inlen); @@ -3112,23 +3124,159 @@ static int mlx5e_create_umr_mkey(struct mlx5e_priv *priv) return err; } -static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev) +static void mlx5e_nic_init(struct mlx5_core_dev *mdev, + struct net_device *netdev, + const struct mlx5e_profile *profile) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + + mlx5e_build_nic_netdev_priv(mdev, netdev, profile); + mlx5e_build_nic_netdev(netdev); + mlx5e_vxlan_init(priv); +} + +static void mlx5e_nic_cleanup(struct mlx5e_priv *priv) +{ + mlx5e_vxlan_cleanup(priv); +} + +static int mlx5e_init_nic_rx(struct mlx5e_priv *priv) +{ + struct mlx5_core_dev *mdev = priv->mdev; + int err; + int i; + + err = mlx5e_create_indirect_rqts(priv); + if (err) { + mlx5_core_warn(mdev, "create indirect rqts failed, %d\n", err); + return err; + } + + err = mlx5e_create_direct_rqts(priv); + if (err) { + mlx5_core_warn(mdev, "create direct rqts failed, %d\n", err); + goto err_destroy_indirect_rqts; + } + + err = mlx5e_create_indirect_tirs(priv); + if (err) { + mlx5_core_warn(mdev, "create indirect tirs failed, %d\n", err); + goto err_destroy_direct_rqts; + } + + err = mlx5e_create_direct_tirs(priv); + if (err) { + mlx5_core_warn(mdev, "create direct tirs failed, %d\n", err); + goto err_destroy_indirect_tirs; + } + + err = mlx5e_create_flow_steering(priv); + if (err) { + mlx5_core_warn(mdev, "create flow steering failed, %d\n", err); + goto err_destroy_direct_tirs; + } + + err = mlx5e_tc_init(priv); + if (err) + goto err_destroy_flow_steering; + + return 0; + +err_destroy_flow_steering: + mlx5e_destroy_flow_steering(priv); +err_destroy_direct_tirs: + mlx5e_destroy_direct_tirs(priv); +err_destroy_indirect_tirs: + mlx5e_destroy_indirect_tirs(priv); +err_destroy_direct_rqts: + for (i = 0; i < priv->profile->max_nch(mdev); i++) + mlx5e_destroy_rqt(priv, &priv->direct_tir[i].rqt); +err_destroy_indirect_rqts: + mlx5e_destroy_rqt(priv, &priv->indir_rqt); + return err; +} + +static void mlx5e_cleanup_nic_rx(struct mlx5e_priv *priv) +{ + int i; + + mlx5e_tc_cleanup(priv); + mlx5e_destroy_flow_steering(priv); + mlx5e_destroy_direct_tirs(priv); + mlx5e_destroy_indirect_tirs(priv); + for (i = 0; i < priv->profile->max_nch(priv->mdev); i++) + mlx5e_destroy_rqt(priv, &priv->direct_tir[i].rqt); + mlx5e_destroy_rqt(priv, &priv->indir_rqt); +} + +static int mlx5e_init_nic_tx(struct mlx5e_priv *priv) +{ + int err; + + err = mlx5e_create_tises(priv); + if (err) { + mlx5_core_warn(priv->mdev, "create tises failed, %d\n", err); + return err; + } + +#ifdef CONFIG_MLX5_CORE_EN_DCB + mlx5e_dcbnl_ieee_setets_core(priv, &priv->params.ets); +#endif + return 0; +} + +static void mlx5e_nic_enable(struct mlx5e_priv *priv) +{ + struct net_device *netdev = priv->netdev; + struct mlx5_core_dev *mdev = priv->mdev; + + if (mlx5e_vxlan_allowed(mdev)) { + rtnl_lock(); + udp_tunnel_get_rx_info(netdev); + rtnl_unlock(); + } + + mlx5e_enable_async_events(priv); + queue_work(priv->wq, &priv->set_rx_mode_work); +} + +static void mlx5e_nic_disable(struct mlx5e_priv *priv) +{ + queue_work(priv->wq, &priv->set_rx_mode_work); + mlx5e_disable_async_events(priv); +} + +static const struct mlx5e_profile mlx5e_nic_profile = { + .init = mlx5e_nic_init, + .cleanup = mlx5e_nic_cleanup, + .init_rx = mlx5e_init_nic_rx, + .cleanup_rx = mlx5e_cleanup_nic_rx, + .init_tx = mlx5e_init_nic_tx, + .cleanup_tx = mlx5e_cleanup_nic_tx, + .enable = mlx5e_nic_enable, + .disable = mlx5e_nic_disable, + .update_stats = mlx5e_update_stats, + .max_nch = mlx5e_get_max_num_channels, + .max_tc = MLX5E_MAX_NUM_TC, +}; + +static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile) { struct net_device *netdev; struct mlx5e_priv *priv; - int nch = mlx5e_get_max_num_channels(mdev); + int nch = profile->max_nch(mdev); int err; netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv), - nch * MLX5E_MAX_NUM_TC, + nch * profile->max_tc, nch); if (!netdev) { mlx5_core_err(mdev, "alloc_etherdev_mqs() failed\n"); return NULL; } - mlx5e_build_netdev_priv(mdev, netdev, nch); - mlx5e_build_netdev(netdev); + profile->init(mdev, netdev, profile); netif_carrier_off(netdev); @@ -3144,85 +3292,44 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev) goto err_destroy_wq; } - err = mlx5e_create_tises(priv); - if (err) { - mlx5_core_warn(mdev, "create tises failed, %d\n", err); + err = profile->init_tx(priv); + if (err) goto err_destroy_umr_mkey; - } err = mlx5e_open_drop_rq(priv); if (err) { mlx5_core_err(mdev, "open drop rq failed, %d\n", err); - goto err_destroy_tises; + goto err_cleanup_tx; } - err = mlx5e_create_rqts(priv); - if (err) { - mlx5_core_warn(mdev, "create rqts failed, %d\n", err); + err = profile->init_rx(priv); + if (err) goto err_close_drop_rq; - } - - err = mlx5e_create_tirs(priv); - if (err) { - mlx5_core_warn(mdev, "create tirs failed, %d\n", err); - goto err_destroy_rqts; - } - - err = mlx5e_create_flow_steering(priv); - if (err) { - mlx5_core_warn(mdev, "create flow steering failed, %d\n", err); - goto err_destroy_tirs; - } mlx5e_create_q_counter(priv); mlx5e_init_l2_addr(priv); - mlx5e_vxlan_init(priv); - - err = mlx5e_tc_init(priv); - if (err) - goto err_dealloc_q_counters; - -#ifdef CONFIG_MLX5_CORE_EN_DCB - mlx5e_dcbnl_ieee_setets_core(priv, &priv->params.ets); -#endif - err = register_netdev(netdev); if (err) { mlx5_core_err(mdev, "register_netdev failed, %d\n", err); - goto err_tc_cleanup; - } - - if (mlx5e_vxlan_allowed(mdev)) { - rtnl_lock(); - udp_tunnel_get_rx_info(netdev); - rtnl_unlock(); + goto err_dealloc_q_counters; } - mlx5e_enable_async_events(priv); - queue_work(priv->wq, &priv->set_rx_mode_work); + if (profile->enable) + profile->enable(priv); return priv; -err_tc_cleanup: - mlx5e_tc_cleanup(priv); - err_dealloc_q_counters: mlx5e_destroy_q_counter(priv); - mlx5e_destroy_flow_steering(priv); - -err_destroy_tirs: - mlx5e_destroy_tirs(priv); - -err_destroy_rqts: - mlx5e_destroy_rqts(priv); + profile->cleanup_rx(priv); err_close_drop_rq: mlx5e_close_drop_rq(priv); -err_destroy_tises: - mlx5e_destroy_tises(priv); +err_cleanup_tx: + profile->cleanup_tx(priv); err_destroy_umr_mkey: mlx5_core_destroy_mkey(mdev, &priv->umr_mkey); @@ -3246,7 +3353,7 @@ static void *mlx5e_add(struct mlx5_core_dev *mdev) if (mlx5e_create_mdev_resources(mdev)) return NULL; - ret = mlx5e_create_netdev(mdev); + ret = mlx5e_create_netdev(mdev, &mlx5e_nic_profile); if (!ret) { mlx5e_destroy_mdev_resources(mdev); return NULL; @@ -3254,15 +3361,15 @@ static void *mlx5e_add(struct mlx5_core_dev *mdev) return ret; } -static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, - struct mlx5e_priv *priv) +static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, struct mlx5e_priv *priv) { + const struct mlx5e_profile *profile = priv->profile; struct net_device *netdev = priv->netdev; set_bit(MLX5E_STATE_DESTROYING, &priv->state); + if (profile->disable) + profile->disable(priv); - queue_work(priv->wq, &priv->set_rx_mode_work); - mlx5e_disable_async_events(priv); flush_workqueue(priv->wq); if (test_bit(MLX5_INTERFACE_STATE_SHUTDOWN, &mdev->intf_state)) { netif_device_detach(netdev); @@ -3271,17 +3378,15 @@ static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, unregister_netdev(netdev); } - mlx5e_tc_cleanup(priv); - mlx5e_vxlan_cleanup(priv); mlx5e_destroy_q_counter(priv); - mlx5e_destroy_flow_steering(priv); - mlx5e_destroy_tirs(priv); - mlx5e_destroy_rqts(priv); + profile->cleanup_rx(priv); mlx5e_close_drop_rq(priv); - mlx5e_destroy_tises(priv); + profile->cleanup_tx(priv); mlx5_core_destroy_mkey(priv->mdev, &priv->umr_mkey); cancel_delayed_work_sync(&priv->update_stats_work); destroy_workqueue(priv->wq); + if (profile->cleanup) + profile->cleanup(priv); if (!test_bit(MLX5_INTERFACE_STATE_SHUTDOWN, &mdev->intf_state)) free_netdev(netdev); -- cgit v0.10.2 From 127ea380acc9de16c2cbd57ed99475944c9917ec Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Fri, 1 Jul 2016 14:51:08 +0300 Subject: net/mlx5: Add Representors registration API Introduce E-Switch registration/unregister representors functions. Those functions are called by the mlx5e driver when the PF NIC is created upon pci probe action regardless of the E-Switch mode (NONE, LEGACY or OFFLOADS). Adding basic E-Switch database that will hold the vport represntors upon creation. This patch doesn't add any new functionality. Signed-off-by: Hadar Hen Zion Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index edfc9be..081259a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -571,7 +571,7 @@ enum { struct mlx5e_profile { void (*init)(struct mlx5_core_dev *mdev, struct net_device *netdev, - const struct mlx5e_profile *profile); + const struct mlx5e_profile *profile, void *ppriv); void (*cleanup)(struct mlx5e_priv *priv); int (*init_rx)(struct mlx5e_priv *priv); void (*cleanup_rx)(struct mlx5e_priv *priv); @@ -618,6 +618,7 @@ struct mlx5e_priv { struct mlx5e_tstamp tstamp; u16 q_counter; const struct mlx5e_profile *profile; + void *ppriv; }; enum mlx5e_link_mode { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 3e22c5e..2c9e458 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -2880,7 +2880,8 @@ void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) static void mlx5e_build_nic_netdev_priv(struct mlx5_core_dev *mdev, struct net_device *netdev, - const struct mlx5e_profile *profile) + const struct mlx5e_profile *profile, + void *ppriv) { struct mlx5e_priv *priv = netdev_priv(netdev); u32 link_speed = 0; @@ -2962,6 +2963,7 @@ static void mlx5e_build_nic_netdev_priv(struct mlx5_core_dev *mdev, priv->netdev = netdev; priv->params.num_channels = profile->max_nch(mdev); priv->profile = profile; + priv->ppriv = ppriv; #ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_ets_init(priv); @@ -3126,18 +3128,25 @@ static int mlx5e_create_umr_mkey(struct mlx5e_priv *priv) static void mlx5e_nic_init(struct mlx5_core_dev *mdev, struct net_device *netdev, - const struct mlx5e_profile *profile) + const struct mlx5e_profile *profile, + void *ppriv) { struct mlx5e_priv *priv = netdev_priv(netdev); - mlx5e_build_nic_netdev_priv(mdev, netdev, profile); + mlx5e_build_nic_netdev_priv(mdev, netdev, profile, ppriv); mlx5e_build_nic_netdev(netdev); mlx5e_vxlan_init(priv); } static void mlx5e_nic_cleanup(struct mlx5e_priv *priv) { + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_eswitch *esw = mdev->priv.eswitch; + mlx5e_vxlan_cleanup(priv); + + if (MLX5_CAP_GEN(mdev, vport_group_manager)) + mlx5_eswitch_unregister_vport_rep(esw, 0); } static int mlx5e_init_nic_rx(struct mlx5e_priv *priv) @@ -3229,6 +3238,8 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv) { struct net_device *netdev = priv->netdev; struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_eswitch *esw = mdev->priv.eswitch; + struct mlx5_eswitch_rep rep; if (mlx5e_vxlan_allowed(mdev)) { rtnl_lock(); @@ -3238,6 +3249,12 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv) mlx5e_enable_async_events(priv); queue_work(priv->wq, &priv->set_rx_mode_work); + + if (MLX5_CAP_GEN(mdev, vport_group_manager)) { + rep.vport = 0; + rep.priv_data = priv; + mlx5_eswitch_register_vport_rep(esw, &rep); + } } static void mlx5e_nic_disable(struct mlx5e_priv *priv) @@ -3261,7 +3278,7 @@ static const struct mlx5e_profile mlx5e_nic_profile = { }; static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev, - const struct mlx5e_profile *profile) + const struct mlx5e_profile *profile, void *ppriv) { struct net_device *netdev; struct mlx5e_priv *priv; @@ -3276,7 +3293,7 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev, return NULL; } - profile->init(mdev, netdev, profile); + profile->init(mdev, netdev, profile, ppriv); netif_carrier_off(netdev); @@ -3343,8 +3360,27 @@ err_free_netdev: return NULL; } +static void mlx5e_register_vport_rep(struct mlx5_core_dev *mdev) +{ + struct mlx5_eswitch *esw = mdev->priv.eswitch; + int total_vfs = MLX5_TOTAL_VPORTS(mdev); + int vport; + + if (!MLX5_CAP_GEN(mdev, vport_group_manager)) + return; + + for (vport = 1; vport < total_vfs; vport++) { + struct mlx5_eswitch_rep rep; + + rep.vport = vport; + mlx5_eswitch_register_vport_rep(esw, &rep); + } +} + static void *mlx5e_add(struct mlx5_core_dev *mdev) { + struct mlx5_eswitch *esw = mdev->priv.eswitch; + void *ppriv = NULL; void *ret; if (mlx5e_check_required_hca_cap(mdev)) @@ -3353,7 +3389,12 @@ static void *mlx5e_add(struct mlx5_core_dev *mdev) if (mlx5e_create_mdev_resources(mdev)) return NULL; - ret = mlx5e_create_netdev(mdev, &mlx5e_nic_profile); + mlx5e_register_vport_rep(mdev); + + if (MLX5_CAP_GEN(mdev, vport_group_manager)) + ppriv = &esw->offloads.vport_reps[0]; + + ret = mlx5e_create_netdev(mdev, &mlx5e_nic_profile, ppriv); if (!ret) { mlx5e_destroy_mdev_resources(mdev); return NULL; @@ -3394,9 +3435,16 @@ static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, struct mlx5e_priv * static void mlx5e_remove(struct mlx5_core_dev *mdev, void *vpriv) { + struct mlx5_eswitch *esw = mdev->priv.eswitch; + int total_vfs = MLX5_TOTAL_VPORTS(mdev); struct mlx5e_priv *priv = vpriv; + int vport; mlx5e_destroy_netdev(mdev, priv); + + for (vport = 1; vport < total_vfs; vport++) + mlx5_eswitch_unregister_vport_rep(esw, vport); + mlx5e_destroy_mdev_resources(mdev); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 12f509c..f0a9735 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -1663,6 +1663,14 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev) goto abort; } + esw->offloads.vport_reps = + kzalloc(total_vports * sizeof(struct mlx5_eswitch_rep), + GFP_KERNEL); + if (!esw->offloads.vport_reps) { + err = -ENOMEM; + goto abort; + } + mutex_init(&esw->state_lock); for (vport_num = 0; vport_num < total_vports; vport_num++) { @@ -1687,6 +1695,7 @@ abort: destroy_workqueue(esw->work_queue); kfree(esw->l2_table.bitmap); kfree(esw->vports); + kfree(esw->offloads.vport_reps); kfree(esw); return err; } @@ -1704,6 +1713,7 @@ void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) destroy_workqueue(esw->work_queue); kfree(esw->l2_table.bitmap); kfree(esw->mc_promisc); + kfree(esw->offloads.vport_reps); kfree(esw->vports); kfree(esw); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 7843f98..ffe5eab 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -156,9 +156,17 @@ enum { SRIOV_OFFLOADS }; + +struct mlx5_eswitch_rep { + u16 vport; + void *priv_data; + bool valid; +}; + struct mlx5_esw_offload { struct mlx5_flow_table *ft_offloads; struct mlx5_flow_group *vport_rx_group; + struct mlx5_eswitch_rep *vport_reps; }; struct mlx5_eswitch { @@ -208,6 +216,10 @@ mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn) int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode); int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode); +void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); +void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw, + int vport); #define MLX5_DEBUG_ESWITCH_MASK BIT(3) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 312b6f3..f84aa79 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -451,3 +451,22 @@ int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode) return 0; } + +void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep) +{ + struct mlx5_esw_offload *offloads = &esw->offloads; + + memcpy(&offloads->vport_reps[rep->vport], rep, + sizeof(struct mlx5_eswitch_rep)); + + offloads->vport_reps[rep->vport].valid = true; +} + +void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw, + int vport) +{ + struct mlx5_esw_offload *offloads = &esw->offloads; + + offloads->vport_reps[vport].valid = false; +} -- cgit v0.10.2 From cb67b832921cfa20ad79bafdc51f1745339d0557 Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Fri, 1 Jul 2016 14:51:09 +0300 Subject: net/mlx5e: Introduce SRIOV VF representors Implement the relevant profile functions to create mlx5e driver instance serving as VF representor. When SRIOV offloads mode is enabled, each VF will have a representor netdevice instance on the host. To do that, we also export set of shared service functions from en_main.c, such that they can be used by both NIC and repsresentors netdevs. The newly created representor netdevice has a basic set of net_device_ops which are the same ndo functions as the NIC netdevice and an ndo of it's own for phys port name. The profiling infrastructure allow sharing code between the NIC and the vport representor even though the representor has only a subset of the NIC functionality. The VF reps and the PF which is used in that mode to represent the uplink, expose switchdev ops. Currently the only op supposed is attr get for the port parent ID which here serves to identify net-devices belonging to the same HW E-Switch. Other than that, no offloading is implemented and hence switching functionality is achieved if one sets SW switching rules, e.g using tc, bridge or ovs. Port phys name (ndo_get_phys_port_name) is implemented to allow exporting to user-space the VF vport number and along with the switchdev port parent id (phys_switch_id) enable a udev base consistent naming scheme: SUBSYSTEM=="net", ACTION=="add", ATTR{phys_switch_id}=="", \ ATTR{phys_port_name}!="", NAME="$PF_NIC$attr{phys_port_name}" where phys_switch_id is exposed by the PF (and VF reps) and $PF_NIC is the name of the PF netdevice. Signed-off-by: Hadar Hen Zion Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 9b14dad..a574dea 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -8,6 +8,6 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o eswitch_offloads.o \ en_main.o en_common.o en_fs.o en_ethtool.o en_tx.o \ en_rx.o en_rx_am.o en_txrx.o en_clock.o vxlan.o \ - en_tc.o en_arfs.o + en_tc.o en_arfs.o en_rep.o mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 081259a..00643a1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -44,6 +44,7 @@ #include #include #include +#include #include "wq.h" #include "mlx5_core.h" #include "en_stats.h" @@ -816,4 +817,31 @@ int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev); void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev); int mlx5e_refresh_tirs_self_loopback_enable(struct mlx5_core_dev *mdev); +struct mlx5_eswitch_rep; +int mlx5e_vport_rep_load(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); +void mlx5e_vport_rep_unload(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); +int mlx5e_nic_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep); +void mlx5e_nic_rep_unload(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); +int mlx5e_add_sqs_fwd_rules(struct mlx5e_priv *priv); +void mlx5e_remove_sqs_fwd_rules(struct mlx5e_priv *priv); +int mlx5e_attr_get(struct net_device *dev, struct switchdev_attr *attr); + +int mlx5e_create_direct_rqts(struct mlx5e_priv *priv); +void mlx5e_destroy_rqt(struct mlx5e_priv *priv, struct mlx5e_rqt *rqt); +int mlx5e_create_direct_tirs(struct mlx5e_priv *priv); +void mlx5e_destroy_direct_tirs(struct mlx5e_priv *priv); +int mlx5e_create_tises(struct mlx5e_priv *priv); +void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv); +int mlx5e_close(struct net_device *netdev); +int mlx5e_open(struct net_device *netdev); +void mlx5e_update_stats_work(struct work_struct *work); +void *mlx5e_create_netdev(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile, void *ppriv); +void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, struct mlx5e_priv *priv); +struct rtnl_link_stats64 * +mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats); + #endif /* __MLX5_EN_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 2c9e458..96ec53a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -226,7 +226,7 @@ void mlx5e_update_stats(struct mlx5e_priv *priv) mlx5e_update_sw_counters(priv); } -static void mlx5e_update_stats_work(struct work_struct *work) +void mlx5e_update_stats_work(struct work_struct *work) { struct delayed_work *dwork = to_delayed_work(work); struct mlx5e_priv *priv = container_of(dwork, struct mlx5e_priv, @@ -1518,7 +1518,7 @@ static int mlx5e_create_rqt(struct mlx5e_priv *priv, int sz, return err; } -static void mlx5e_destroy_rqt(struct mlx5e_priv *priv, struct mlx5e_rqt *rqt) +void mlx5e_destroy_rqt(struct mlx5e_priv *priv, struct mlx5e_rqt *rqt) { rqt->enabled = false; mlx5_core_destroy_rqt(priv->mdev, rqt->rqtn); @@ -1531,7 +1531,7 @@ static int mlx5e_create_indirect_rqts(struct mlx5e_priv *priv) return mlx5e_create_rqt(priv, MLX5E_INDIR_RQT_SIZE, 0, rqt); } -static int mlx5e_create_direct_rqts(struct mlx5e_priv *priv) +int mlx5e_create_direct_rqts(struct mlx5e_priv *priv) { struct mlx5e_rqt *rqt; int err; @@ -1743,6 +1743,7 @@ static void mlx5e_netdev_set_tcs(struct net_device *netdev) int mlx5e_open_locked(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; int num_txqs; int err; @@ -1778,9 +1779,14 @@ int mlx5e_open_locked(struct net_device *netdev) #ifdef CONFIG_RFS_ACCEL priv->netdev->rx_cpu_rmap = priv->mdev->rmap; #endif + if (priv->profile->update_stats) + queue_delayed_work(priv->wq, &priv->update_stats_work, 0); - queue_delayed_work(priv->wq, &priv->update_stats_work, 0); - + if (MLX5_CAP_GEN(mdev, vport_group_manager)) { + err = mlx5e_add_sqs_fwd_rules(priv); + if (err) + goto err_close_channels; + } return 0; err_close_channels: @@ -1790,7 +1796,7 @@ err_clear_state_opened_flag: return err; } -static int mlx5e_open(struct net_device *netdev) +int mlx5e_open(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); int err; @@ -1805,6 +1811,7 @@ static int mlx5e_open(struct net_device *netdev) int mlx5e_close_locked(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; /* May already be CLOSED in case a previous configuration operation * (e.g RX/TX queue size change) that involves close&open failed. @@ -1814,6 +1821,9 @@ int mlx5e_close_locked(struct net_device *netdev) clear_bit(MLX5E_STATE_OPENED, &priv->state); + if (MLX5_CAP_GEN(mdev, vport_group_manager)) + mlx5e_remove_sqs_fwd_rules(priv); + mlx5e_timestamp_cleanup(priv); netif_carrier_off(priv->netdev); mlx5e_redirect_rqts(priv); @@ -1822,7 +1832,7 @@ int mlx5e_close_locked(struct net_device *netdev) return 0; } -static int mlx5e_close(struct net_device *netdev) +int mlx5e_close(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); int err; @@ -1957,7 +1967,7 @@ static void mlx5e_destroy_tis(struct mlx5e_priv *priv, int tc) mlx5_core_destroy_tis(priv->mdev, priv->tisn[tc]); } -static int mlx5e_create_tises(struct mlx5e_priv *priv) +int mlx5e_create_tises(struct mlx5e_priv *priv) { int err; int tc; @@ -1977,7 +1987,7 @@ err_close_tises: return err; } -static void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv) +void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv) { int tc; @@ -2143,7 +2153,7 @@ err_destroy_tirs: return err; } -static int mlx5e_create_direct_tirs(struct mlx5e_priv *priv) +int mlx5e_create_direct_tirs(struct mlx5e_priv *priv) { int nch = priv->profile->max_nch(priv->mdev); struct mlx5e_tir *tir; @@ -2190,7 +2200,7 @@ static void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv) mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[i]); } -static void mlx5e_destroy_direct_tirs(struct mlx5e_priv *priv) +void mlx5e_destroy_direct_tirs(struct mlx5e_priv *priv) { int nch = priv->profile->max_nch(priv->mdev); int i; @@ -2270,7 +2280,7 @@ mqprio: return mlx5e_setup_tc(dev, tc->tc); } -static struct rtnl_link_stats64 * +struct rtnl_link_stats64 * mlx5e_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats) { struct mlx5e_priv *priv = netdev_priv(dev); @@ -2988,6 +2998,10 @@ static void mlx5e_set_netdev_dev_addr(struct net_device *netdev) } } +static const struct switchdev_ops mlx5e_switchdev_ops = { + .switchdev_port_attr_get = mlx5e_attr_get, +}; + static void mlx5e_build_nic_netdev(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); @@ -3069,6 +3083,11 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) netdev->priv_flags |= IFF_UNICAST_FLT; mlx5e_set_netdev_dev_addr(netdev); + +#ifdef CONFIG_NET_SWITCHDEV + if (MLX5_CAP_GEN(mdev, vport_group_manager)) + netdev->switchdev_ops = &mlx5e_switchdev_ops; +#endif } static void mlx5e_create_q_counter(struct mlx5e_priv *priv) @@ -3251,6 +3270,8 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv) queue_work(priv->wq, &priv->set_rx_mode_work); if (MLX5_CAP_GEN(mdev, vport_group_manager)) { + rep.load = mlx5e_nic_rep_load; + rep.unload = mlx5e_nic_rep_unload; rep.vport = 0; rep.priv_data = priv; mlx5_eswitch_register_vport_rep(esw, &rep); @@ -3277,8 +3298,8 @@ static const struct mlx5e_profile mlx5e_nic_profile = { .max_tc = MLX5E_MAX_NUM_TC, }; -static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev, - const struct mlx5e_profile *profile, void *ppriv) +void *mlx5e_create_netdev(struct mlx5_core_dev *mdev, + const struct mlx5e_profile *profile, void *ppriv) { struct net_device *netdev; struct mlx5e_priv *priv; @@ -3372,6 +3393,8 @@ static void mlx5e_register_vport_rep(struct mlx5_core_dev *mdev) for (vport = 1; vport < total_vfs; vport++) { struct mlx5_eswitch_rep rep; + rep.load = mlx5e_vport_rep_load; + rep.unload = mlx5e_vport_rep_unload; rep.vport = vport; mlx5_eswitch_register_vport_rep(esw, &rep); } @@ -3402,7 +3425,7 @@ static void *mlx5e_add(struct mlx5_core_dev *mdev) return ret; } -static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, struct mlx5e_priv *priv) +void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, struct mlx5e_priv *priv) { const struct mlx5e_profile *profile = priv->profile; struct net_device *netdev = priv->netdev; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c new file mode 100644 index 0000000..5ef02f0 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -0,0 +1,394 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include + +#include "eswitch.h" +#include "en.h" + +static const char mlx5e_rep_driver_name[] = "mlx5e_rep"; + +static void mlx5e_rep_get_drvinfo(struct net_device *dev, + struct ethtool_drvinfo *drvinfo) +{ + strlcpy(drvinfo->driver, mlx5e_rep_driver_name, + sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, UTS_RELEASE, sizeof(drvinfo->version)); +} + +static const struct counter_desc sw_rep_stats_desc[] = { + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_bytes) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_packets) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_bytes) }, +}; + +#define NUM_VPORT_REP_COUNTERS ARRAY_SIZE(sw_rep_stats_desc) + +static void mlx5e_rep_get_strings(struct net_device *dev, + u32 stringset, uint8_t *data) +{ + int i; + + switch (stringset) { + case ETH_SS_STATS: + for (i = 0; i < NUM_VPORT_REP_COUNTERS; i++) + strcpy(data + (i * ETH_GSTRING_LEN), + sw_rep_stats_desc[i].format); + break; + } +} + +static void mlx5e_update_sw_rep_counters(struct mlx5e_priv *priv) +{ + struct mlx5e_sw_stats *s = &priv->stats.sw; + struct mlx5e_rq_stats *rq_stats; + struct mlx5e_sq_stats *sq_stats; + int i, j; + + memset(s, 0, sizeof(*s)); + for (i = 0; i < priv->params.num_channels; i++) { + rq_stats = &priv->channel[i]->rq.stats; + + s->rx_packets += rq_stats->packets; + s->rx_bytes += rq_stats->bytes; + + for (j = 0; j < priv->params.num_tc; j++) { + sq_stats = &priv->channel[i]->sq[j].stats; + + s->tx_packets += sq_stats->packets; + s->tx_bytes += sq_stats->bytes; + } + } +} + +static void mlx5e_rep_get_ethtool_stats(struct net_device *dev, + struct ethtool_stats *stats, u64 *data) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + int i; + + if (!data) + return; + + mutex_lock(&priv->state_lock); + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_update_sw_rep_counters(priv); + mutex_unlock(&priv->state_lock); + + for (i = 0; i < NUM_VPORT_REP_COUNTERS; i++) + data[i] = MLX5E_READ_CTR64_CPU(&priv->stats.sw, + sw_rep_stats_desc, i); +} + +static int mlx5e_rep_get_sset_count(struct net_device *dev, int sset) +{ + switch (sset) { + case ETH_SS_STATS: + return NUM_VPORT_REP_COUNTERS; + default: + return -EOPNOTSUPP; + } +} + +static const struct ethtool_ops mlx5e_rep_ethtool_ops = { + .get_drvinfo = mlx5e_rep_get_drvinfo, + .get_link = ethtool_op_get_link, + .get_strings = mlx5e_rep_get_strings, + .get_sset_count = mlx5e_rep_get_sset_count, + .get_ethtool_stats = mlx5e_rep_get_ethtool_stats, +}; + +int mlx5e_attr_get(struct net_device *dev, struct switchdev_attr *attr) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + u8 mac[ETH_ALEN]; + + if (esw->mode == SRIOV_NONE) + return -EOPNOTSUPP; + + switch (attr->id) { + case SWITCHDEV_ATTR_ID_PORT_PARENT_ID: + mlx5_query_nic_vport_mac_address(priv->mdev, 0, mac); + attr->u.ppid.id_len = ETH_ALEN; + memcpy(&attr->u.ppid.id, &mac, ETH_ALEN); + break; + default: + return -EOPNOTSUPP; + } + + return 0; +} + +int mlx5e_add_sqs_fwd_rules(struct mlx5e_priv *priv) + +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_eswitch_rep *rep = priv->ppriv; + struct mlx5e_channel *c; + int n, tc, err, num_sqs = 0; + u16 *sqs; + + sqs = kcalloc(priv->params.num_channels * priv->params.num_tc, sizeof(u16), GFP_KERNEL); + if (!sqs) + return -ENOMEM; + + for (n = 0; n < priv->params.num_channels; n++) { + c = priv->channel[n]; + for (tc = 0; tc < c->num_tc; tc++) + sqs[num_sqs++] = c->sq[tc].sqn; + } + + err = mlx5_eswitch_sqs2vport_start(esw, rep, sqs, num_sqs); + + kfree(sqs); + return err; +} + +int mlx5e_nic_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep) +{ + struct mlx5e_priv *priv = rep->priv_data; + + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + return mlx5e_add_sqs_fwd_rules(priv); + return 0; +} + +void mlx5e_remove_sqs_fwd_rules(struct mlx5e_priv *priv) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_eswitch_rep *rep = priv->ppriv; + + mlx5_eswitch_sqs2vport_stop(esw, rep); +} + +void mlx5e_nic_rep_unload(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep) +{ + struct mlx5e_priv *priv = rep->priv_data; + + if (test_bit(MLX5E_STATE_OPENED, &priv->state)) + mlx5e_remove_sqs_fwd_rules(priv); +} + +static int mlx5e_rep_get_phys_port_name(struct net_device *dev, + char *buf, size_t len) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5_eswitch_rep *rep = priv->ppriv; + int ret; + + ret = snprintf(buf, len, "%d", rep->vport - 1); + if (ret >= len) + return -EOPNOTSUPP; + + return 0; +} + +static const struct switchdev_ops mlx5e_rep_switchdev_ops = { + .switchdev_port_attr_get = mlx5e_attr_get, +}; + +static const struct net_device_ops mlx5e_netdev_ops_rep = { + .ndo_open = mlx5e_open, + .ndo_stop = mlx5e_close, + .ndo_start_xmit = mlx5e_xmit, + .ndo_get_phys_port_name = mlx5e_rep_get_phys_port_name, + .ndo_get_stats64 = mlx5e_get_stats, +}; + +static void mlx5e_build_rep_netdev_priv(struct mlx5_core_dev *mdev, + struct net_device *netdev, + const struct mlx5e_profile *profile, + void *ppriv) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + u8 cq_period_mode = MLX5_CAP_GEN(mdev, cq_period_start_from_cqe) ? + MLX5_CQ_PERIOD_MODE_START_FROM_CQE : + MLX5_CQ_PERIOD_MODE_START_FROM_EQE; + + priv->params.log_sq_size = + MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE; + priv->params.rq_wq_type = MLX5_WQ_TYPE_LINKED_LIST; + priv->params.log_rq_size = MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE; + + priv->params.min_rx_wqes = mlx5_min_rx_wqes(priv->params.rq_wq_type, + BIT(priv->params.log_rq_size)); + + priv->params.rx_am_enabled = MLX5_CAP_GEN(mdev, cq_moderation); + mlx5e_set_rx_cq_mode_params(&priv->params, cq_period_mode); + + priv->params.tx_max_inline = mlx5e_get_max_inline_cap(mdev); + priv->params.num_tc = 1; + + priv->params.lro_wqe_sz = + MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ; + + priv->mdev = mdev; + priv->netdev = netdev; + priv->params.num_channels = profile->max_nch(mdev); + priv->profile = profile; + priv->ppriv = ppriv; + + mutex_init(&priv->state_lock); + + INIT_DELAYED_WORK(&priv->update_stats_work, mlx5e_update_stats_work); +} + +static void mlx5e_build_rep_netdev(struct net_device *netdev) +{ + netdev->netdev_ops = &mlx5e_netdev_ops_rep; + + netdev->watchdog_timeo = 15 * HZ; + + netdev->ethtool_ops = &mlx5e_rep_ethtool_ops; + +#ifdef CONFIG_NET_SWITCHDEV + netdev->switchdev_ops = &mlx5e_rep_switchdev_ops; +#endif + + netdev->features |= NETIF_F_VLAN_CHALLENGED; + + eth_hw_addr_random(netdev); +} + +static void mlx5e_init_rep(struct mlx5_core_dev *mdev, + struct net_device *netdev, + const struct mlx5e_profile *profile, + void *ppriv) +{ + mlx5e_build_rep_netdev_priv(mdev, netdev, profile, ppriv); + mlx5e_build_rep_netdev(netdev); +} + +static int mlx5e_init_rep_rx(struct mlx5e_priv *priv) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_eswitch_rep *rep = priv->ppriv; + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5_flow_rule *flow_rule; + int err; + int i; + + err = mlx5e_create_direct_rqts(priv); + if (err) { + mlx5_core_warn(mdev, "create direct rqts failed, %d\n", err); + return err; + } + + err = mlx5e_create_direct_tirs(priv); + if (err) { + mlx5_core_warn(mdev, "create direct tirs failed, %d\n", err); + goto err_destroy_direct_rqts; + } + + flow_rule = mlx5_eswitch_create_vport_rx_rule(esw, + rep->vport, + priv->direct_tir[0].tirn); + if (IS_ERR(flow_rule)) { + err = PTR_ERR(flow_rule); + goto err_destroy_direct_tirs; + } + rep->vport_rx_rule = flow_rule; + + return 0; + +err_destroy_direct_tirs: + mlx5e_destroy_direct_tirs(priv); +err_destroy_direct_rqts: + for (i = 0; i < priv->params.num_channels; i++) + mlx5e_destroy_rqt(priv, &priv->direct_tir[i].rqt); + return err; +} + +static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv) +{ + struct mlx5_eswitch_rep *rep = priv->ppriv; + int i; + + mlx5_del_flow_rule(rep->vport_rx_rule); + mlx5e_destroy_direct_tirs(priv); + for (i = 0; i < priv->params.num_channels; i++) + mlx5e_destroy_rqt(priv, &priv->direct_tir[i].rqt); +} + +static int mlx5e_init_rep_tx(struct mlx5e_priv *priv) +{ + int err; + + err = mlx5e_create_tises(priv); + if (err) { + mlx5_core_warn(priv->mdev, "create tises failed, %d\n", err); + return err; + } + return 0; +} + +static int mlx5e_get_rep_max_num_channels(struct mlx5_core_dev *mdev) +{ +#define MLX5E_PORT_REPRESENTOR_NCH 1 + return MLX5E_PORT_REPRESENTOR_NCH; +} + +static struct mlx5e_profile mlx5e_rep_profile = { + .init = mlx5e_init_rep, + .init_rx = mlx5e_init_rep_rx, + .cleanup_rx = mlx5e_cleanup_rep_rx, + .init_tx = mlx5e_init_rep_tx, + .cleanup_tx = mlx5e_cleanup_nic_tx, + .update_stats = mlx5e_update_sw_rep_counters, + .max_nch = mlx5e_get_rep_max_num_channels, + .max_tc = 1, +}; + +int mlx5e_vport_rep_load(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep) +{ + rep->priv_data = mlx5e_create_netdev(esw->dev, &mlx5e_rep_profile, rep); + if (!rep->priv_data) { + pr_warn("Failed to create representor for vport %d\n", + rep->vport); + return -EINVAL; + } + return 0; +} + +void mlx5e_vport_rep_unload(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep) +{ + struct mlx5e_priv *priv = rep->priv_data; + + mlx5e_destroy_netdev(esw->dev, priv); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index ffe5eab..7b45e6a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -47,6 +47,8 @@ #define MLX5_L2_ADDR_HASH_SIZE (BIT(BITS_PER_BYTE)) #define MLX5_L2_ADDR_HASH(addr) (addr[5]) +#define FDB_UPLINK_VPORT 0xffff + /* L2 -mac address based- hash helpers */ struct l2addr_node { struct hlist_node hlist; @@ -156,10 +158,20 @@ enum { SRIOV_OFFLOADS }; +struct mlx5_esw_sq { + struct mlx5_flow_rule *send_to_vport_rule; + struct list_head list; +}; struct mlx5_eswitch_rep { + int (*load)(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); + void (*unload)(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); u16 vport; + struct mlx5_flow_rule *vport_rx_rule; void *priv_data; + struct list_head vport_sqs_list; bool valid; }; @@ -208,12 +220,16 @@ int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw, int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, int vport, struct ifla_vf_stats *vf_stats); -struct mlx5_flow_rule * -mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn); struct mlx5_flow_rule * mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn); +int mlx5_eswitch_sqs2vport_start(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep, + u16 *sqns_array, int sqns_num); +void mlx5_eswitch_sqs2vport_stop(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep); + int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode); int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode); void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index f84aa79..ed8ad98 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -38,7 +38,7 @@ #include "mlx5_core.h" #include "eswitch.h" -struct mlx5_flow_rule * +static struct mlx5_flow_rule * mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn) { struct mlx5_flow_destination dest; @@ -77,6 +77,63 @@ out: return flow_rule; } +void mlx5_eswitch_sqs2vport_stop(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep) +{ + struct mlx5_esw_sq *esw_sq, *tmp; + + if (esw->mode != SRIOV_OFFLOADS) + return; + + list_for_each_entry_safe(esw_sq, tmp, &rep->vport_sqs_list, list) { + mlx5_del_flow_rule(esw_sq->send_to_vport_rule); + list_del(&esw_sq->list); + kfree(esw_sq); + } +} + +int mlx5_eswitch_sqs2vport_start(struct mlx5_eswitch *esw, + struct mlx5_eswitch_rep *rep, + u16 *sqns_array, int sqns_num) +{ + struct mlx5_flow_rule *flow_rule; + struct mlx5_esw_sq *esw_sq; + int vport; + int err; + int i; + + if (esw->mode != SRIOV_OFFLOADS) + return 0; + + vport = rep->vport == 0 ? + FDB_UPLINK_VPORT : rep->vport; + + for (i = 0; i < sqns_num; i++) { + esw_sq = kzalloc(sizeof(*esw_sq), GFP_KERNEL); + if (!esw_sq) { + err = -ENOMEM; + goto out_err; + } + + /* Add re-inject rule to the PF/representor sqs */ + flow_rule = mlx5_eswitch_add_send_to_vport_rule(esw, + vport, + sqns_array[i]); + if (IS_ERR(flow_rule)) { + err = PTR_ERR(flow_rule); + kfree(esw_sq); + goto out_err; + } + esw_sq->send_to_vport_rule = flow_rule; + list_add(&esw_sq->list, &rep->vport_sqs_list); + } + return 0; + +out_err: + mlx5_eswitch_sqs2vport_stop(esw, rep); + return err; +} + static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) { struct mlx5_flow_destination dest; @@ -347,6 +404,8 @@ static int esw_offloads_start(struct mlx5_eswitch *esw) int esw_offloads_init(struct mlx5_eswitch *esw, int nvports) { + struct mlx5_eswitch_rep *rep; + int vport; int err; err = esw_create_offloads_fdb_table(esw, nvports); @@ -361,8 +420,26 @@ int esw_offloads_init(struct mlx5_eswitch *esw, int nvports) if (err) goto create_fg_err; + for (vport = 0; vport < nvports; vport++) { + rep = &esw->offloads.vport_reps[vport]; + if (!rep->valid) + continue; + + err = rep->load(esw, rep); + if (err) + goto err_reps; + } return 0; +err_reps: + for (vport--; vport >= 0; vport--) { + rep = &esw->offloads.vport_reps[vport]; + if (!rep->valid) + continue; + rep->unload(esw, rep); + } + esw_destroy_vport_rx_group(esw); + create_fg_err: esw_destroy_offloads_table(esw); @@ -385,6 +462,16 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw) void esw_offloads_cleanup(struct mlx5_eswitch *esw, int nvports) { + struct mlx5_eswitch_rep *rep; + int vport; + + for (vport = 0; vport < nvports; vport++) { + rep = &esw->offloads.vport_reps[vport]; + if (!rep->valid) + continue; + rep->unload(esw, rep); + } + esw_destroy_vport_rx_group(esw); esw_destroy_offloads_table(esw); esw_destroy_offloads_fdb_table(esw); @@ -460,6 +547,7 @@ void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw, memcpy(&offloads->vport_reps[rep->vport], rep, sizeof(struct mlx5_eswitch_rep)); + INIT_LIST_HEAD(&offloads->vport_reps[rep->vport].vport_sqs_list); offloads->vport_reps[rep->vport].valid = true; } @@ -467,6 +555,12 @@ void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw, int vport) { struct mlx5_esw_offload *offloads = &esw->offloads; + struct mlx5_eswitch_rep *rep; + + rep = &offloads->vport_reps[vport]; + + if (esw->mode == SRIOV_OFFLOADS && esw->vports[vport].enabled) + rep->unload(esw, rep); offloads->vport_reps[vport].valid = false; } -- cgit v0.10.2 From c332177e5ee5ef9280d346fb1cba0f00ffe4c18d Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 2 Jul 2016 00:02:34 +0200 Subject: net: ethernet: davinci_emac: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index f56d66e..76683c7 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -348,7 +348,6 @@ struct emac_priv { u32 rx_addr_type; const char *phy_id; struct device_node *phy_node; - struct phy_device *phydev; spinlock_t lock; /*platform specific members*/ void (*int_enable) (void); @@ -496,9 +495,8 @@ static void emac_get_drvinfo(struct net_device *ndev, static int emac_get_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) { - struct emac_priv *priv = netdev_priv(ndev); - if (priv->phydev) - return phy_ethtool_gset(priv->phydev, ecmd); + if (ndev->phydev) + return phy_ethtool_gset(ndev->phydev, ecmd); else return -EOPNOTSUPP; @@ -514,9 +512,8 @@ static int emac_get_settings(struct net_device *ndev, */ static int emac_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) { - struct emac_priv *priv = netdev_priv(ndev); - if (priv->phydev) - return phy_ethtool_sset(priv->phydev, ecmd); + if (ndev->phydev) + return phy_ethtool_sset(ndev->phydev, ecmd); else return -EOPNOTSUPP; @@ -651,8 +648,8 @@ static void emac_update_phystatus(struct emac_priv *priv) mac_control = emac_read(EMAC_MACCONTROL); cur_duplex = (mac_control & EMAC_MACCONTROL_FULLDUPLEXEN) ? DUPLEX_FULL : DUPLEX_HALF; - if (priv->phydev) - new_duplex = priv->phydev->duplex; + if (ndev->phydev) + new_duplex = ndev->phydev->duplex; else new_duplex = DUPLEX_FULL; @@ -1454,7 +1451,7 @@ static void emac_poll_controller(struct net_device *ndev) static void emac_adjust_link(struct net_device *ndev) { struct emac_priv *priv = netdev_priv(ndev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = ndev->phydev; unsigned long flags; int new_state = 0; @@ -1483,7 +1480,7 @@ static void emac_adjust_link(struct net_device *ndev) } if (new_state) { emac_update_phystatus(priv); - phy_print_status(priv->phydev); + phy_print_status(ndev->phydev); } spin_unlock_irqrestore(&priv->lock, flags); @@ -1505,15 +1502,13 @@ static void emac_adjust_link(struct net_device *ndev) */ static int emac_devioctl(struct net_device *ndev, struct ifreq *ifrq, int cmd) { - struct emac_priv *priv = netdev_priv(ndev); - if (!(netif_running(ndev))) return -EINVAL; /* TODO: Add phy read and write and private statistics get feature */ - if (priv->phydev) - return phy_mii_ioctl(priv->phydev, ifrq, cmd); + if (ndev->phydev) + return phy_mii_ioctl(ndev->phydev, ifrq, cmd); else return -EOPNOTSUPP; } @@ -1542,6 +1537,7 @@ static int emac_dev_open(struct net_device *ndev) int res_num = 0, irq_num = 0; int i = 0; struct emac_priv *priv = netdev_priv(ndev); + struct phy_device *phydev = NULL; ret = pm_runtime_get_sync(&priv->pdev->dev); if (ret < 0) { @@ -1607,12 +1603,10 @@ static int emac_dev_open(struct net_device *ndev) cpdma_ctlr_start(priv->dma); - priv->phydev = NULL; - if (priv->phy_node) { - priv->phydev = of_phy_connect(ndev, priv->phy_node, - &emac_adjust_link, 0, 0); - if (!priv->phydev) { + phydev = of_phy_connect(ndev, priv->phy_node, + &emac_adjust_link, 0, 0); + if (!phydev) { dev_err(emac_dev, "could not connect to phy %s\n", priv->phy_node->full_name); ret = -ENODEV; @@ -1621,7 +1615,7 @@ static int emac_dev_open(struct net_device *ndev) } /* use the first phy on the bus if pdata did not give us a phy id */ - if (!priv->phydev && !priv->phy_id) { + if (!phydev && !priv->phy_id) { struct device *phy; phy = bus_find_device(&mdio_bus_type, NULL, NULL, @@ -1630,16 +1624,15 @@ static int emac_dev_open(struct net_device *ndev) priv->phy_id = dev_name(phy); } - if (!priv->phydev && priv->phy_id && *priv->phy_id) { - priv->phydev = phy_connect(ndev, priv->phy_id, - &emac_adjust_link, - PHY_INTERFACE_MODE_MII); + if (!phydev && priv->phy_id && *priv->phy_id) { + phydev = phy_connect(ndev, priv->phy_id, + &emac_adjust_link, + PHY_INTERFACE_MODE_MII); - if (IS_ERR(priv->phydev)) { + if (IS_ERR(phydev)) { dev_err(emac_dev, "could not connect to phy %s\n", priv->phy_id); - ret = PTR_ERR(priv->phydev); - priv->phydev = NULL; + ret = PTR_ERR(phydev); goto err; } @@ -1647,10 +1640,10 @@ static int emac_dev_open(struct net_device *ndev) priv->speed = 0; priv->duplex = ~0; - phy_attached_info(priv->phydev); + phy_attached_info(phydev); } - if (!priv->phydev) { + if (!phydev) { /* No PHY , fix the link, speed and duplex settings */ dev_notice(emac_dev, "no phy, defaulting to 100/full\n"); priv->link = 1; @@ -1665,8 +1658,8 @@ static int emac_dev_open(struct net_device *ndev) if (netif_msg_drv(priv)) dev_notice(emac_dev, "DaVinci EMAC: Opened %s\n", ndev->name); - if (priv->phydev) - phy_start(priv->phydev); + if (phydev) + phy_start(phydev); return 0; @@ -1717,8 +1710,8 @@ static int emac_dev_stop(struct net_device *ndev) cpdma_ctlr_stop(priv->dma); emac_write(EMAC_SOFTRESET, 1); - if (priv->phydev) - phy_disconnect(priv->phydev); + if (ndev->phydev) + phy_disconnect(ndev->phydev); /* Free IRQ */ while ((res = platform_get_resource(priv->pdev, IORESOURCE_IRQ, i))) { -- cgit v0.10.2 From efb15c396445644aaea9d88e68300762f01fca92 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 2 Jul 2016 00:02:35 +0200 Subject: net: ethernet: davinci_emac: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index 76683c7..c6c5465 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -485,41 +485,6 @@ static void emac_get_drvinfo(struct net_device *ndev, } /** - * emac_get_settings - Get EMAC settings - * @ndev: The DaVinci EMAC network adapter - * @ecmd: ethtool command - * - * Executes ethool get command - * - */ -static int emac_get_settings(struct net_device *ndev, - struct ethtool_cmd *ecmd) -{ - if (ndev->phydev) - return phy_ethtool_gset(ndev->phydev, ecmd); - else - return -EOPNOTSUPP; - -} - -/** - * emac_set_settings - Set EMAC settings - * @ndev: The DaVinci EMAC network adapter - * @ecmd: ethtool command - * - * Executes ethool set command - * - */ -static int emac_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) -{ - if (ndev->phydev) - return phy_ethtool_sset(ndev->phydev, ecmd); - else - return -EOPNOTSUPP; - -} - -/** * emac_get_coalesce - Get interrupt coalesce settings for this device * @ndev : The DaVinci EMAC network adapter * @coal : ethtool coalesce settings structure @@ -622,12 +587,12 @@ static int emac_set_coalesce(struct net_device *ndev, */ static const struct ethtool_ops ethtool_ops = { .get_drvinfo = emac_get_drvinfo, - .get_settings = emac_get_settings, - .set_settings = emac_set_settings, .get_link = ethtool_op_get_link, .get_coalesce = emac_get_coalesce, .set_coalesce = emac_set_coalesce, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /** -- cgit v0.10.2 From 75362a3fd4e37ff8af1ef5e3d9f2d9d5ccf2f3ab Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 1 Jul 2016 18:46:19 -0400 Subject: bnxt_en: VF/NPAR should return -EOPNOTSUPP for unsupported ethtool ops. Returning 0 for doing nothing is confusing to the user. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index d7ab2d79..c63ed2f 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -961,7 +961,7 @@ static int bnxt_set_pauseparam(struct net_device *dev, struct bnxt_link_info *link_info = &bp->link_info; if (!BNXT_SINGLE_PF(bp)) - return rc; + return -EOPNOTSUPP; if (epause->autoneg) { if (!(link_info->autoneg & BNXT_AUTONEG_SPEED)) @@ -1483,7 +1483,7 @@ static int bnxt_set_eee(struct net_device *dev, struct ethtool_eee *edata) int rc = 0; if (!BNXT_SINGLE_PF(bp)) - return 0; + return -EOPNOTSUPP; if (!(bp->flags & BNXT_FLAG_EEE_CAP)) return -EOPNOTSUPP; -- cgit v0.10.2 From a58a3e68037647de78e3461194239a1104f76003 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 1 Jul 2016 18:46:20 -0400 Subject: bnxt_en: Update firmware spec. to 1.3.0. And update driver version to 1.3.0. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 927ece9..084b3f2 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -11,10 +11,10 @@ #define BNXT_H #define DRV_MODULE_NAME "bnxt_en" -#define DRV_MODULE_VERSION "1.2.0" +#define DRV_MODULE_VERSION "1.3.0" #define DRV_VER_MAJ 1 -#define DRV_VER_MIN 0 +#define DRV_VER_MIN 3 #define DRV_VER_UPD 0 struct tx_bd { @@ -359,7 +359,8 @@ struct rx_tpa_end_cmp { RX_TPA_END_CMP_FLAGS_PLACEMENT_ANY_GRO) #define TPA_END_GRO_TS(rx_tpa_end) \ - ((rx_tpa_end)->rx_tpa_end_cmp_tsdelta & cpu_to_le32(RX_TPA_END_GRO_TS)) + (!!((rx_tpa_end)->rx_tpa_end_cmp_tsdelta & \ + cpu_to_le32(RX_TPA_END_GRO_TS))) struct rx_tpa_end_cmp_ext { __le32 rx_tpa_end_cmp_dup_acks; @@ -753,8 +754,8 @@ struct bnxt_vf_info { struct bnxt_pf_info { #define BNXT_FIRST_PF_FID 1 #define BNXT_FIRST_VF_FID 128 - u32 fw_fid; - u8 port_id; + u16 fw_fid; + u16 port_id; u8 mac_addr[ETH_ALEN]; u16 max_rsscos_ctxs; u16 max_cp_rings; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h index 05e3c49..517567f 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h @@ -105,6 +105,7 @@ struct hwrm_async_event_cmpl { #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED (0x4UL << 0) #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_NOT_ALLOWED (0x5UL << 0) #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_CHANGE (0x6UL << 0) + #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PORT_PHY_CFG_CHANGE (0x7UL << 0) #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_FUNC_DRVR_UNLOAD (0x10UL << 0) #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_FUNC_DRVR_LOAD (0x11UL << 0) #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD (0x20UL << 0) @@ -484,12 +485,12 @@ struct hwrm_async_event_cmpl_hwrm_error { #define HWRM_ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA1_TIMESTAMP 0x1UL }; -/* HW Resource Manager Specification 1.2.2 */ +/* HW Resource Manager Specification 1.3.0 */ #define HWRM_VERSION_MAJOR 1 -#define HWRM_VERSION_MINOR 2 -#define HWRM_VERSION_UPDATE 2 +#define HWRM_VERSION_MINOR 3 +#define HWRM_VERSION_UPDATE 0 -#define HWRM_VERSION_STR "1.2.2" +#define HWRM_VERSION_STR "1.3.0" /* * Following is the signature for HWRM message field that indicates not * applicable (All F's). Need to cast it the size of the field if needed. @@ -611,6 +612,9 @@ struct cmd_nums { #define HWRM_FWD_RESP (0xd2UL) #define HWRM_FWD_ASYNC_EVENT_CMPL (0xd3UL) #define HWRM_TEMP_MONITOR_QUERY (0xe0UL) + #define HWRM_WOL_FILTER_ALLOC (0xf0UL) + #define HWRM_WOL_FILTER_FREE (0xf1UL) + #define HWRM_WOL_FILTER_QCFG (0xf2UL) #define HWRM_DBG_READ_DIRECT (0xff10UL) #define HWRM_DBG_READ_INDIRECT (0xff11UL) #define HWRM_DBG_WRITE_DIRECT (0xff12UL) @@ -1020,6 +1024,10 @@ struct hwrm_func_qcaps_output { #define FUNC_QCAPS_RESP_FLAGS_PUSH_MODE_SUPPORTED 0x1UL #define FUNC_QCAPS_RESP_FLAGS_GLOBAL_MSIX_AUTOMASKING 0x2UL #define FUNC_QCAPS_RESP_FLAGS_PTP_SUPPORTED 0x4UL + #define FUNC_QCAPS_RESP_FLAGS_ROCE_V1_SUPPORTED 0x8UL + #define FUNC_QCAPS_RESP_FLAGS_ROCE_V2_SUPPORTED 0x10UL + #define FUNC_QCAPS_RESP_FLAGS_WOL_MAGICPKT_SUPPORTED 0x20UL + #define FUNC_QCAPS_RESP_FLAGS_WOL_BMP_SUPPORTED 0x40UL u8 mac_address[6]; __le16 max_rsscos_ctx; __le16 max_cmpl_rings; @@ -1066,8 +1074,9 @@ struct hwrm_func_qcfg_output { __le16 fid; __le16 port_id; __le16 vlan; - u8 unused_0; - u8 unused_1; + __le16 flags; + #define FUNC_QCFG_RESP_FLAGS_OOB_WOL_MAGICPKT_ENABLED 0x1UL + #define FUNC_QCFG_RESP_FLAGS_OOB_WOL_BMP_ENABLED 0x2UL u8 mac_address[6]; __le16 pci_id; __le16 alloc_rsscos_ctx; @@ -1086,23 +1095,23 @@ struct hwrm_func_qcfg_output { #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_5 (0x3UL << 0) #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR2_0 (0x4UL << 0) #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_UNKNOWN (0xffUL << 0) - u8 unused_2; + u8 unused_0; __le16 dflt_vnic_id; - u8 unused_3; - u8 unused_4; + u8 unused_1; + u8 unused_2; __le32 min_bw; __le32 max_bw; u8 evb_mode; #define FUNC_QCFG_RESP_EVB_MODE_NO_EVB (0x0UL << 0) #define FUNC_QCFG_RESP_EVB_MODE_VEB (0x1UL << 0) #define FUNC_QCFG_RESP_EVB_MODE_VEPA (0x2UL << 0) - u8 unused_5; - __le16 unused_6; + u8 unused_3; + __le16 unused_4; __le32 alloc_mcast_filters; __le32 alloc_hw_ring_grps; + u8 unused_5; + u8 unused_6; u8 unused_7; - u8 unused_8; - u8 unused_9; u8 valid; }; @@ -1410,8 +1419,8 @@ struct hwrm_func_buf_rgtr_input { #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_4K (0xcUL << 0) #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_8K (0xdUL << 0) #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_64K (0x10UL << 0) - #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_2M (0x16UL << 0) - #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_4M (0x17UL << 0) + #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_2M (0x15UL << 0) + #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_4M (0x16UL << 0) #define FUNC_BUF_RGTR_REQ_REQ_BUF_PAGE_SIZE_1G (0x1eUL << 0) __le16 req_buf_len; __le16 resp_buf_len; @@ -1499,6 +1508,12 @@ struct hwrm_port_phy_cfg_input { #define PORT_PHY_CFG_REQ_FLAGS_EEE_DISABLE 0x20UL #define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_ENABLE 0x40UL #define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_DISABLE 0x80UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_ENABLE 0x100UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_AUTONEG_DISABLE 0x200UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_ENABLE 0x400UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE74_DISABLE 0x800UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_ENABLE 0x1000UL + #define PORT_PHY_CFG_REQ_FLAGS_FEC_CLAUSE91_DISABLE 0x2000UL __le32 enables; #define PORT_PHY_CFG_REQ_ENABLES_AUTO_MODE 0x1UL #define PORT_PHY_CFG_REQ_ENABLES_AUTO_DUPLEX 0x2UL @@ -1815,13 +1830,22 @@ struct hwrm_port_phy_qcfg_output { #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFP (0xcUL << 24) #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFPPLUS (0xdUL << 24) #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFP28 (0x11UL << 24) - __le32 unused_1; + __le16 fec_cfg; + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_NONE_SUPPORTED 0x1UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_SUPPORTED 0x2UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_AUTONEG_ENABLED 0x4UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_SUPPORTED 0x8UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE74_ENABLED 0x10UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_SUPPORTED 0x20UL + #define PORT_PHY_QCFG_RESP_FEC_CFG_FEC_CLAUSE91_ENABLED 0x40UL + u8 unused_1; + u8 unused_2; char phy_vendor_name[16]; char phy_vendor_partnumber[16]; - __le32 unused_2; - u8 unused_3; + __le32 unused_3; u8 unused_4; u8 unused_5; + u8 unused_6; u8 valid; }; @@ -1842,6 +1866,8 @@ struct hwrm_port_mac_cfg_input { #define PORT_MAC_CFG_REQ_FLAGS_PTP_RX_TS_CAPTURE_DISABLE 0x20UL #define PORT_MAC_CFG_REQ_FLAGS_PTP_TX_TS_CAPTURE_ENABLE 0x40UL #define PORT_MAC_CFG_REQ_FLAGS_PTP_TX_TS_CAPTURE_DISABLE 0x80UL + #define PORT_MAC_CFG_REQ_FLAGS_OOB_WOL_ENABLE 0x100UL + #define PORT_MAC_CFG_REQ_FLAGS_OOB_WOL_DISABLE 0x200UL __le32 enables; #define PORT_MAC_CFG_REQ_ENABLES_IPG 0x1UL #define PORT_MAC_CFG_REQ_ENABLES_LPBK 0x2UL @@ -2127,6 +2153,7 @@ struct hwrm_port_phy_i2c_read_output { u8 valid; }; +/* hwrm_queue_qportcfg */ /* Input (24 bytes) */ struct hwrm_queue_qportcfg_input { __le16 req_type; @@ -2382,7 +2409,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID0_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id0_pri_lvl; u8 queue_id0_bw_weight; u8 queue_id1; @@ -2392,7 +2419,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID1_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id1_pri_lvl; u8 queue_id1_bw_weight; u8 queue_id2; @@ -2402,7 +2429,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID2_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id2_pri_lvl; u8 queue_id2_bw_weight; u8 queue_id3; @@ -2412,7 +2439,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID3_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id3_pri_lvl; u8 queue_id3_bw_weight; u8 queue_id4; @@ -2422,7 +2449,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID4_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id4_pri_lvl; u8 queue_id4_bw_weight; u8 queue_id5; @@ -2432,7 +2459,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID5_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id5_pri_lvl; u8 queue_id5_bw_weight; u8 queue_id6; @@ -2442,7 +2469,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID6_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id6_pri_lvl; u8 queue_id6_bw_weight; u8 queue_id7; @@ -2452,7 +2479,7 @@ struct hwrm_queue_cos2bw_cfg_input { #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_SP (0x0UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_ETS (0x1UL << 0) #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_RESERVED_FIRST (0x2UL << 0) - #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_RESERVED_LAST (0xffffUL << 0) + #define QUEUE_COS2BW_CFG_REQ_QUEUE_ID7_TSA_ASSIGN_RESERVED_LAST (0xffUL << 0) u8 queue_id7_pri_lvl; u8 queue_id7_bw_weight; u8 unused_1[5]; @@ -3150,7 +3177,7 @@ struct hwrm_cfa_l2_filter_cfg_output { }; /* hwrm_cfa_l2_set_rx_mask */ -/* Input (40 bytes) */ +/* Input (56 bytes) */ struct hwrm_cfa_l2_set_rx_mask_input { __le16 req_type; __le16 cmpl_ring; @@ -3165,9 +3192,15 @@ struct hwrm_cfa_l2_set_rx_mask_input { #define CFA_L2_SET_RX_MASK_REQ_MASK_BCAST 0x8UL #define CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS 0x10UL #define CFA_L2_SET_RX_MASK_REQ_MASK_OUTERMOST 0x20UL + #define CFA_L2_SET_RX_MASK_REQ_MASK_VLANONLY 0x40UL + #define CFA_L2_SET_RX_MASK_REQ_MASK_VLAN_NONVLAN 0x80UL + #define CFA_L2_SET_RX_MASK_REQ_MASK_ANYVLAN_NONVLAN 0x100UL __le64 mc_tbl_addr; __le32 num_mc_entries; __le32 unused_0; + __le64 vlan_tag_tbl_addr; + __le32 num_vlan_tags; + __le32 unused_1; }; /* Output (16 bytes) */ -- cgit v0.10.2 From 2a5bedfa674cf81d60a20a76f456778834bd2123 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 1 Jul 2016 18:46:21 -0400 Subject: bnxt_en: Do function reset on the 1st PF open only. Calling the firmware to do function reset on the PF will kill all the VFs. To prevent that, we call function reset on the 1st PF open before any VF can be activated. On subsequent PF opens (with possibly some active VFs), a bit has been set and we'll skip the function reset. VF driver will always do function reset on every open. If there is an AER event, we will always do function reset. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 673f4d6..b489fb6 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -5295,12 +5295,19 @@ static int bnxt_open(struct net_device *dev) struct bnxt *bp = netdev_priv(dev); int rc = 0; - rc = bnxt_hwrm_func_reset(bp); - if (rc) { - netdev_err(bp->dev, "hwrm chip reset failure rc: %x\n", - rc); - rc = -1; - return rc; + if (!test_bit(BNXT_STATE_FN_RST_DONE, &bp->state)) { + rc = bnxt_hwrm_func_reset(bp); + if (rc) { + netdev_err(bp->dev, "hwrm chip reset failure rc: %x\n", + rc); + rc = -EBUSY; + return rc; + } + /* Do func_reset during the 1st PF open only to prevent killing + * the VFs when the PF is brought down and up. + */ + if (BNXT_PF(bp)) + set_bit(BNXT_STATE_FN_RST_DONE, &bp->state); } return __bnxt_open_nic(bp, true, true); } @@ -6676,6 +6683,7 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev, pci_channel_state_t state) { struct net_device *netdev = pci_get_drvdata(pdev); + struct bnxt *bp = netdev_priv(netdev); netdev_info(netdev, "PCI I/O error detected\n"); @@ -6690,6 +6698,8 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev, if (netif_running(netdev)) bnxt_close(netdev); + /* So that func_reset will be done during slot_reset */ + clear_bit(BNXT_STATE_FN_RST_DONE, &bp->state); pci_disable_device(pdev); rtnl_unlock(); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 084b3f2..1d5a3cd 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -1018,6 +1018,7 @@ struct bnxt { unsigned long state; #define BNXT_STATE_OPEN 0 #define BNXT_STATE_IN_SP_TASK 1 +#define BNXT_STATE_FN_RST_DONE 2 struct bnxt_irq *irq_tbl; u8 mac_addr[ETH_ALEN]; -- cgit v0.10.2 From a4c363471f2fa2b0f0abbd9f0563b034340585c3 Mon Sep 17 00:00:00 2001 From: Rob Swindell Date: Fri, 1 Jul 2016 18:46:22 -0400 Subject: bnxt_en: Add support for updating flash more securely To support Secure Firmware Update, we must be able to allocate a staging area in the Flash. This patch adds support for the "update" type to tell firmware to do that. Signed-off-by: Rob Swindell Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index c63ed2f..3328aa5 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -1186,7 +1186,8 @@ static int bnxt_flash_firmware_from_file(struct net_device *dev, const struct firmware *fw; int rc; - if (bnxt_dir_type_is_executable(dir_type) == false) + if (dir_type != BNX_DIR_TYPE_UPDATE && + bnxt_dir_type_is_executable(dir_type) == false) return -EINVAL; rc = request_firmware(&fw, filename, &dev->dev); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h index 40a7b0e..73f2249 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h @@ -13,6 +13,7 @@ enum bnxt_nvm_directory_type { BNX_DIR_TYPE_UNUSED = 0, BNX_DIR_TYPE_PKG_LOG = 1, + BNX_DIR_TYPE_UPDATE = 2, BNX_DIR_TYPE_CHIMP_PATCH = 3, BNX_DIR_TYPE_BOOTCODE = 4, BNX_DIR_TYPE_VPD = 5, -- cgit v0.10.2 From 08141e0bf4f6cb82d51930e34e6a8e4af46c776f Mon Sep 17 00:00:00 2001 From: Rob Swindell Date: Fri, 1 Jul 2016 18:46:23 -0400 Subject: bnxt_en: Request firmware reset after successful firwmare update Upon successful mgmt processor firmware update, request a self reset upon next PCIe reset (e.g. system reboot). Signed-off-by: Rob Swindell Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 3328aa5..12a5141 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -1059,6 +1059,8 @@ static int bnxt_firmware_reset(struct net_device *dev, case BNX_DIR_TYPE_APE_FW: case BNX_DIR_TYPE_APE_PATCH: req.embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_MGMT; + /* Self-reset APE upon next PCIe reset: */ + req.selfrst_status = FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST; break; case BNX_DIR_TYPE_KONG_FW: case BNX_DIR_TYPE_KONG_PATCH: -- cgit v0.10.2 From 93e0b4feb90cc651f7fbdfe07c257a969c51d1bb Mon Sep 17 00:00:00 2001 From: Rob Swindell Date: Fri, 1 Jul 2016 18:46:24 -0400 Subject: bnxt_en: Add support for firmware updates for additional processors. Add support to the Ethtool FLASHDEV command handler for additional firmware types to cover all the on-chip processors. Signed-off-by: Rob Swindell Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 12a5141..33b3135 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -1094,9 +1094,27 @@ static int bnxt_flash_firmware(struct net_device *dev, case BNX_DIR_TYPE_BOOTCODE_2: code_type = CODE_BOOT; break; + case BNX_DIR_TYPE_CHIMP_PATCH: + code_type = CODE_CHIMP_PATCH; + break; case BNX_DIR_TYPE_APE_FW: code_type = CODE_MCTP_PASSTHRU; break; + case BNX_DIR_TYPE_APE_PATCH: + code_type = CODE_APE_PATCH; + break; + case BNX_DIR_TYPE_KONG_FW: + code_type = CODE_KONG_FW; + break; + case BNX_DIR_TYPE_KONG_PATCH: + code_type = CODE_KONG_PATCH; + break; + case BNX_DIR_TYPE_BONO_FW: + code_type = CODE_BONO_FW; + break; + case BNX_DIR_TYPE_BONO_PATCH: + code_type = CODE_BONO_PATCH; + break; default: netdev_err(dev, "Unsupported directory entry type: %u\n", dir_type); @@ -1151,6 +1169,8 @@ static bool bnxt_dir_type_is_ape_bin_format(u16 dir_type) case BNX_DIR_TYPE_APE_PATCH: case BNX_DIR_TYPE_KONG_FW: case BNX_DIR_TYPE_KONG_PATCH: + case BNX_DIR_TYPE_BONO_FW: + case BNX_DIR_TYPE_BONO_PATCH: return true; } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h index 461675c..82bf44a 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h @@ -70,6 +70,7 @@ enum SUPPORTED_CODE { CODE_KONG_PATCH, /* 18 - KONG Patch firmware */ CODE_BONO_FW, /* 19 - BONO firmware */ CODE_BONO_PATCH, /* 20 - BONO Patch firmware */ + CODE_CHIMP_PATCH, /* 21 - ChiMP Patch firmware */ MAX_CODE_TYPE, }; -- cgit v0.10.2 From 550feebf5cb075f7576b3cfe9bcf05abc1ffb8cd Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 1 Jul 2016 18:46:25 -0400 Subject: bnxt_en: Enable MRU enables bit when configuring VNIC MRU. For correctness, the MRU enables bit must be set when passing the MRU to firmware during vnic configuration. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index b489fb6..28a5aee 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -3414,7 +3414,8 @@ static int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id) bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_CFG, -1, -1); /* Only RSS support for now TBD: COS & LB */ req.enables = cpu_to_le32(VNIC_CFG_REQ_ENABLES_DFLT_RING_GRP | - VNIC_CFG_REQ_ENABLES_RSS_RULE); + VNIC_CFG_REQ_ENABLES_RSS_RULE | + VNIC_CFG_REQ_ENABLES_MRU); req.rss_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx); req.cos_rule = cpu_to_le16(0xffff); if (vnic->flags & BNXT_VNIC_RSS_FLAG) -- cgit v0.10.2 From dc7aadb5133846f738c59da7af3261335af35ad3 Mon Sep 17 00:00:00 2001 From: Vasundhara Volam Date: Fri, 1 Jul 2016 18:46:26 -0400 Subject: bnxt_en: Increase maximum supported MTU to 9500. Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 28a5aee..9134268 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -6049,7 +6049,7 @@ static int bnxt_change_mtu(struct net_device *dev, int new_mtu) { struct bnxt *bp = netdev_priv(dev); - if (new_mtu < 60 || new_mtu > 9000) + if (new_mtu < 60 || new_mtu > 9500) return -EINVAL; if (netif_running(dev)) -- cgit v0.10.2 From 17c71ac38134c3369479e34911b2035a85566caf Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 1 Jul 2016 18:46:27 -0400 Subject: bnxt_en: Allow promiscuous mode for VF if default VLAN is enabled. With a default VLAN, the VF has its own VLAN domain and it can receive all traffic within that domain. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 9134268..2740ac3 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -4316,6 +4316,16 @@ static int bnxt_alloc_rfs_vnics(struct bnxt *bp) #endif } +/* Allow PF and VF with default VLAN to be in promiscuous mode */ +static bool bnxt_promisc_ok(struct bnxt *bp) +{ +#ifdef CONFIG_BNXT_SRIOV + if (BNXT_VF(bp) && !bp->vf.vlan) + return false; +#endif + return true; +} + static int bnxt_cfg_rx_mode(struct bnxt *); static bool bnxt_mc_list_updated(struct bnxt *, u32 *); @@ -4381,7 +4391,7 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init) vnic->rx_mask = CFA_L2_SET_RX_MASK_REQ_MASK_BCAST; - if ((bp->dev->flags & IFF_PROMISC) && BNXT_PF(bp)) + if ((bp->dev->flags & IFF_PROMISC) && bnxt_promisc_ok(bp)) vnic->rx_mask |= CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS; if (bp->dev->flags & IFF_ALLMULTI) { @@ -5528,8 +5538,7 @@ static void bnxt_set_rx_mode(struct net_device *dev) CFA_L2_SET_RX_MASK_REQ_MASK_MCAST | CFA_L2_SET_RX_MASK_REQ_MASK_ALL_MCAST); - /* Only allow PF to be in promiscuous mode */ - if ((dev->flags & IFF_PROMISC) && BNXT_PF(bp)) + if ((dev->flags & IFF_PROMISC) && bnxt_promisc_ok(bp)) mask |= CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS; uc_update = bnxt_uc_list_updated(bp); -- cgit v0.10.2 From 87027db19c30aafb8ff8d98e1c8802bc920f7b32 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 1 Jul 2016 18:46:28 -0400 Subject: bnxt_en: Assign netdev->dev_port with port ID. This is useful for multi-function devices. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 2740ac3..18be62b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -4026,6 +4026,7 @@ int bnxt_hwrm_func_qcaps(struct bnxt *bp) pf->fw_fid = le16_to_cpu(resp->fid); pf->port_id = le16_to_cpu(resp->port_id); + bp->dev->dev_port = pf->port_id; memcpy(pf->mac_addr, resp->mac_address, ETH_ALEN); memcpy(bp->dev->dev_addr, pf->mac_addr, ETH_ALEN); pf->max_rsscos_ctxs = le16_to_cpu(resp->max_rsscos_ctx); -- cgit v0.10.2 From 51f307856b60e6b10975654e15bc236aa87b53d7 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 1 Jul 2016 18:46:29 -0400 Subject: bnxt_en: Allow statistics DMA to be configurable using ethtool -C. The allowable range is 0.25 seconds to 1 second interval. Default is 1 second. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 18be62b..70b148a 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -3952,7 +3952,7 @@ static int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp) bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_STAT_CTX_ALLOC, -1, -1); - req.update_period_ms = cpu_to_le32(1000); + req.update_period_ms = cpu_to_le32(bp->stats_coal_ticks / 1000); mutex_lock(&bp->hwrm_cmd_lock); for (i = 0; i < bp->cp_nr_rings; i++) { @@ -5994,6 +5994,8 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev) bp->tx_coal_ticks_irq = 2; bp->tx_coal_bufs_irq = 2; + bp->stats_coal_ticks = BNXT_DEF_STATS_COAL_TICKS; + init_timer(&bp->timer); bp->timer.data = (unsigned long)bp; bp->timer.function = bnxt_timer; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 1d5a3cd..2313e37 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -1067,6 +1067,11 @@ struct bnxt { #define BNXT_USEC_TO_COAL_TIMER(x) ((x) * 25 / 2) + u32 stats_coal_ticks; +#define BNXT_DEF_STATS_COAL_TICKS 1000000 +#define BNXT_MIN_STATS_COAL_TICKS 250000 +#define BNXT_MAX_STATS_COAL_TICKS 1000000 + struct work_struct sp_task; unsigned long sp_event; #define BNXT_RX_MASK_SP_EVENT 0 diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 33b3135..0f7dd86 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -56,6 +56,8 @@ static int bnxt_get_coalesce(struct net_device *dev, coal->tx_coalesce_usecs_irq = bp->tx_coal_ticks_irq; coal->tx_max_coalesced_frames_irq = bp->tx_coal_bufs_irq; + coal->stats_block_coalesce_usecs = bp->stats_coal_ticks; + return 0; } @@ -63,6 +65,7 @@ static int bnxt_set_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) { struct bnxt *bp = netdev_priv(dev); + bool update_stats = false; int rc = 0; bp->rx_coal_ticks = coal->rx_coalesce_usecs; @@ -76,8 +79,26 @@ static int bnxt_set_coalesce(struct net_device *dev, bp->tx_coal_ticks_irq = coal->tx_coalesce_usecs_irq; bp->tx_coal_bufs_irq = coal->tx_max_coalesced_frames_irq; - if (netif_running(dev)) - rc = bnxt_hwrm_set_coal(bp); + if (bp->stats_coal_ticks != coal->stats_block_coalesce_usecs) { + u32 stats_ticks = coal->stats_block_coalesce_usecs; + + stats_ticks = clamp_t(u32, stats_ticks, + BNXT_MIN_STATS_COAL_TICKS, + BNXT_MAX_STATS_COAL_TICKS); + stats_ticks = rounddown(stats_ticks, BNXT_MIN_STATS_COAL_TICKS); + bp->stats_coal_ticks = stats_ticks; + update_stats = true; + } + + if (netif_running(dev)) { + if (update_stats) { + rc = bnxt_close_nic(bp, true, false); + if (!rc) + rc = bnxt_open_nic(bp, true, false); + } else { + rc = bnxt_hwrm_set_coal(bp); + } + } return rc; } -- cgit v0.10.2 From 52697a9ede4ff9ce55930342c2a4d0f985df4ebf Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:09 +0200 Subject: mlxsw: spectrum: Send untagged packets through a port netdev Port netdevs (e.g. swXpY) that are not bridged are represented in the device using a vPort with VID=PVID=1 (the PVID vPort), as untagged packets entering the switch are internally tagged with the PVID VLAN. When these packets are routed through a different port netdev they should egress untagged. This wasn't a problem until now, as non-bridged traffic only originated from the CPU, which transmits packets out of the port as-is. When a vPort is created with VID 1 mark it as egress untagged. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index a453fff..afd06dc 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -816,6 +816,7 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); struct mlxsw_sp_port *mlxsw_sp_vport; + bool untagged = vid == 1; int err; /* VLAN 0 is added to HW filter when device goes up, but it is @@ -859,7 +860,7 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, goto err_port_vid_learning_set; } - err = mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, true, false); + err = mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, true, untagged); if (err) { netdev_err(dev, "Failed to set VLAN membership for VID=%d\n", vid); -- cgit v0.10.2 From 32d863fb9308588e94fc9035a40046aaf81e2966 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:10 +0200 Subject: mlxsw: spectrum: Remove VLANs configuration via SELF flag When port isn't bridged it is still possible to invoke switchdev ops and configure the device's VLAN filters. However, this will require us to use different Router InterFaces (RIFs) for the same netdev, instead of one per-netdev as with any other configuration. Taking the above into account and the fact that this functionality is questionable with regards to the device's normal use-case, remove it and instead return an error. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index afd06dc..4f67a8c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -890,8 +890,8 @@ err_port_vp_mode_trans: return err; } -int mlxsw_sp_port_kill_vid(struct net_device *dev, - __be16 __always_unused proto, u16 vid) +static int mlxsw_sp_port_kill_vid(struct net_device *dev, + __be16 __always_unused proto, u16 vid) { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); struct mlxsw_sp_port *mlxsw_sp_vport; @@ -1845,23 +1845,6 @@ err_port_active_vlans_alloc: return err; } -static void mlxsw_sp_port_vports_fini(struct mlxsw_sp_port *mlxsw_sp_port) -{ - struct net_device *dev = mlxsw_sp_port->dev; - struct mlxsw_sp_port *mlxsw_sp_vport, *tmp; - - list_for_each_entry_safe(mlxsw_sp_vport, tmp, - &mlxsw_sp_port->vports_list, vport.list) { - u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); - - /* vPorts created for VLAN devices should already be gone - * by now, since we unregistered the port netdev. - */ - WARN_ON(is_vlan_dev(mlxsw_sp_vport->dev)); - mlxsw_sp_port_kill_vid(dev, 0, vid); - } -} - static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port) { struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp->ports[local_port]; @@ -1872,13 +1855,14 @@ static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port) mlxsw_core_port_fini(&mlxsw_sp_port->core_port); unregister_netdev(mlxsw_sp_port->dev); /* This calls ndo_stop */ mlxsw_sp_port_dcb_fini(mlxsw_sp_port); - mlxsw_sp_port_vports_fini(mlxsw_sp_port); + mlxsw_sp_port_kill_vid(mlxsw_sp_port->dev, 0, 1); mlxsw_sp_port_switchdev_fini(mlxsw_sp_port); mlxsw_sp_port_swid_set(mlxsw_sp_port, MLXSW_PORT_SWID_DISABLED_PORT); mlxsw_sp_port_module_unmap(mlxsw_sp, mlxsw_sp_port->local_port); free_percpu(mlxsw_sp_port->pcpu_stats); kfree(mlxsw_sp_port->untagged_vlans); kfree(mlxsw_sp_port->active_vlans); + WARN_ON_ONCE(!list_empty(&mlxsw_sp_port->vports_list)); free_netdev(mlxsw_sp_port->dev); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 36c9835..05d5fcc 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -377,8 +377,6 @@ int mlxsw_sp_port_vlan_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin, u16 vid_end, bool is_member, bool untagged); int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, u16 vid); -int mlxsw_sp_port_kill_vid(struct net_device *dev, - __be16 __always_unused proto, u16 vid); int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, bool set); void mlxsw_sp_port_active_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index a0c7376..927117e 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -633,25 +633,6 @@ err_port_allow_untagged_set: return err; } -static int mlxsw_sp_port_add_vids(struct net_device *dev, u16 vid_begin, - u16 vid_end) -{ - u16 vid; - int err; - - for (vid = vid_begin; vid <= vid_end; vid++) { - err = mlxsw_sp_port_add_vid(dev, 0, vid); - if (err) - goto err_port_add_vid; - } - return 0; - -err_port_add_vid: - for (vid--; vid >= vid_begin; vid--) - mlxsw_sp_port_kill_vid(dev, 0, vid); - return err; -} - static int __mlxsw_sp_port_vlans_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin, u16 vid_end, bool is_member, bool untagged) @@ -681,12 +662,8 @@ static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid, old_pvid; int err; - /* In case this is invoked with BRIDGE_FLAGS_SELF and port is - * not bridged, then packets ingressing through the port with - * the specified VIDs will be directed to CPU. - */ if (!mlxsw_sp_port->bridged) - return mlxsw_sp_port_add_vids(dev, vid_begin, vid_end); + return -EINVAL; err = mlxsw_sp_port_fid_join(mlxsw_sp_port, vid_begin, vid_end); if (err) { @@ -1019,21 +996,6 @@ static int mlxsw_sp_port_obj_add(struct net_device *dev, return err; } -static int mlxsw_sp_port_kill_vids(struct net_device *dev, u16 vid_begin, - u16 vid_end) -{ - u16 vid; - int err; - - for (vid = vid_begin; vid <= vid_end; vid++) { - err = mlxsw_sp_port_kill_vid(dev, 0, vid); - if (err) - return err; - } - - return 0; -} - static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin, u16 vid_end, bool init) { @@ -1041,12 +1003,8 @@ static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid, pvid; int err; - /* In case this is invoked with BRIDGE_FLAGS_SELF and port is - * not bridged, then prevent packets ingressing through the - * port with the specified VIDs from being trapped to CPU. - */ if (!init && !mlxsw_sp_port->bridged) - return mlxsw_sp_port_kill_vids(dev, vid_begin, vid_end); + return -EINVAL; err = __mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end, false, false); -- cgit v0.10.2 From 86bf95b334a77c388b2aca4b26d24216d9784823 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:11 +0200 Subject: mlxsw: spectrum: Sync PVID vPort LAG status When VLAN devices are created on top of LAG, their underlying vPorts are configured correctly with LAG membership. However, the PVID vPort is implicit and already present when the port netdev is put under LAG, so its LAG membership is never set. Set it correctly when joining / leaving LAG. This didn't matter until now, but we are going to introduce support for router interfaces (RIFs), which need to take into account LAG membership. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 4f67a8c..f276c45 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2795,6 +2795,32 @@ static int mlxsw_sp_port_lag_index_get(struct mlxsw_sp *mlxsw_sp, return -EBUSY; } +static void +mlxsw_sp_port_pvid_vport_lag_join(struct mlxsw_sp_port *mlxsw_sp_port, + u16 lag_id) +{ + struct mlxsw_sp_port *mlxsw_sp_vport; + + mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, 1); + if (WARN_ON(!mlxsw_sp_vport)) + return; + + mlxsw_sp_vport->lag_id = lag_id; + mlxsw_sp_vport->lagged = 1; +} + +static void +mlxsw_sp_port_pvid_vport_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp_port *mlxsw_sp_vport; + + mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, 1); + if (WARN_ON(!mlxsw_sp_vport)) + return; + + mlxsw_sp_vport->lagged = 0; +} + static int mlxsw_sp_port_lag_join(struct mlxsw_sp_port *mlxsw_sp_port, struct net_device *lag_dev) { @@ -2830,6 +2856,9 @@ static int mlxsw_sp_port_lag_join(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_port->lag_id = lag_id; mlxsw_sp_port->lagged = 1; lag->ref_count++; + + mlxsw_sp_port_pvid_vport_lag_join(mlxsw_sp_port, lag_id); + return 0; err_col_port_enable: @@ -2867,6 +2896,8 @@ static void mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port, mlxsw_sp_port->local_port); mlxsw_sp_port->lagged = 0; lag->ref_count--; + + mlxsw_sp_port_pvid_vport_lag_leave(mlxsw_sp_port); } static int mlxsw_sp_lag_dist_port_add(struct mlxsw_sp_port *mlxsw_sp_port, -- cgit v0.10.2 From 11943ff4423d1f3f84d502d65d72f6a98bd2dc91 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:12 +0200 Subject: mlxsw: spectrum: Remove RIF from PVID vPort when joining / leaving LAG We are going to assign router interfaces (RIFs) to netdevs if an IPv4 address was assigned to them. If one was assigned to a port netdev, this will translate to the PVID vPort being member in a RIF. While it's possible for a LAG slave to have an IP address, we can't have a vPort being member in two FIDs (assuming the LAG device will be put in bridge / assigned an IP address). Solve that by making the PVID vPort leave any FID it might be a member in when joining / leaving LAG. Note that the PVID vPort is the only vPort that can be present on the port when it's put under LAG. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index f276c45..30fe0d2 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2800,11 +2800,19 @@ mlxsw_sp_port_pvid_vport_lag_join(struct mlxsw_sp_port *mlxsw_sp_port, u16 lag_id) { struct mlxsw_sp_port *mlxsw_sp_vport; + struct mlxsw_sp_fid *f; mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, 1); if (WARN_ON(!mlxsw_sp_vport)) return; + /* If vPort is assigned a RIF, then leave it since it's no + * longer valid. + */ + f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + if (f) + f->leave(mlxsw_sp_vport); + mlxsw_sp_vport->lag_id = lag_id; mlxsw_sp_vport->lagged = 1; } @@ -2813,11 +2821,16 @@ static void mlxsw_sp_port_pvid_vport_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port) { struct mlxsw_sp_port *mlxsw_sp_vport; + struct mlxsw_sp_fid *f; mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, 1); if (WARN_ON(!mlxsw_sp_vport)) return; + f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + if (f) + f->leave(mlxsw_sp_vport); + mlxsw_sp_vport->lagged = 0; } -- cgit v0.10.2 From 69c407aaf902f8761a9a121012225d08efe4fbaa Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:13 +0200 Subject: mlxsw: reg: Add Router General Configuration Register Add the Router General Configuration Register (RGCR), which allows us to enable the router in the device and configure its various parameters. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 1977e7a..7f74eb7 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -1,7 +1,7 @@ /* * drivers/net/ethernet/mellanox/mlxsw/reg.h * Copyright (c) 2015 Mellanox Technologies. All rights reserved. - * Copyright (c) 2015 Ido Schimmel + * Copyright (c) 2015-2016 Ido Schimmel * Copyright (c) 2015 Elad Raz * Copyright (c) 2015 Jiri Pirko * @@ -3186,6 +3186,80 @@ static inline void mlxsw_reg_hpkt_pack(char *payload, u8 action, u16 trap_id) mlxsw_reg_hpkt_ctrl_set(payload, MLXSW_REG_HPKT_CTRL_PACKET_DEFAULT); } +/* RGCR - Router General Configuration Register + * -------------------------------------------- + * The register is used for setting up the router configuration. + */ +#define MLXSW_REG_RGCR_ID 0x8001 +#define MLXSW_REG_RGCR_LEN 0x28 + +static const struct mlxsw_reg_info mlxsw_reg_rgcr = { + .id = MLXSW_REG_RGCR_ID, + .len = MLXSW_REG_RGCR_LEN, +}; + +/* reg_rgcr_ipv4_en + * IPv4 router enable. + * Access: RW + */ +MLXSW_ITEM32(reg, rgcr, ipv4_en, 0x00, 31, 1); + +/* reg_rgcr_ipv6_en + * IPv6 router enable. + * Access: RW + */ +MLXSW_ITEM32(reg, rgcr, ipv6_en, 0x00, 30, 1); + +/* reg_rgcr_max_router_interfaces + * Defines the maximum number of active router interfaces for all virtual + * routers. + * Access: RW + */ +MLXSW_ITEM32(reg, rgcr, max_router_interfaces, 0x10, 0, 16); + +/* reg_rgcr_usp + * Update switch priority and packet color. + * 0 - Preserve the value of Switch Priority and packet color. + * 1 - Recalculate the value of Switch Priority and packet color. + * Access: RW + * + * Note: Not supported by SwitchX and SwitchX-2. + */ +MLXSW_ITEM32(reg, rgcr, usp, 0x18, 20, 1); + +/* reg_rgcr_pcp_rw + * Indicates how to handle the pcp_rewrite_en value: + * 0 - Preserve the value of pcp_rewrite_en. + * 2 - Disable PCP rewrite. + * 3 - Enable PCP rewrite. + * Access: RW + * + * Note: Not supported by SwitchX and SwitchX-2. + */ +MLXSW_ITEM32(reg, rgcr, pcp_rw, 0x18, 16, 2); + +/* reg_rgcr_activity_dis + * Activity disable: + * 0 - Activity will be set when an entry is hit (default). + * 1 - Activity will not be set when an entry is hit. + * + * Bit 0 - Disable activity bit in Router Algorithmic LPM Unicast Entry + * (RALUE). + * Bit 1 - Disable activity bit in Router Algorithmic LPM Unicast Host + * Entry (RAUHT). + * Bits 2:7 are reserved. + * Access: RW + * + * Note: Not supported by SwitchX, SwitchX-2 and Switch-IB. + */ +MLXSW_ITEM32(reg, rgcr, activity_dis, 0x20, 0, 8); + +static inline void mlxsw_reg_rgcr_pack(char *payload, bool ipv4_en) +{ + MLXSW_REG_ZERO(rgcr, payload); + mlxsw_reg_rgcr_ipv4_en_set(payload, ipv4_en); +} + /* MFCR - Management Fan Control Register * -------------------------------------- * This register controls the settings of the Fan Speed PWM mechanism. @@ -3924,6 +3998,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "HTGT"; case MLXSW_REG_HPKT_ID: return "HPKT"; + case MLXSW_REG_RGCR_ID: + return "RGCR"; case MLXSW_REG_MFCR_ID: return "MFCR"; case MLXSW_REG_MFSC_ID: -- cgit v0.10.2 From bbf2a4757b307b6ba64c0f4cce97e6993703d53d Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:14 +0200 Subject: mlxsw: spectrum: Initialize ports at the end of init sequence During ports initialization a net device is registered for each available port, which implies the port is usable. However, a port is only usable after the different parts of the device (e.g. flooding, buffers) are initialized. This is especially important now, when we must initialize the router before the ports, as otherwise the device can't be initialized. Solve that by initializing the switch ports at the end of init sequence. Also, remove an unnecessary warning about port up/down events, which would otherwise be invoked whenever removing the driver, as ports are removed before unregistering the listener for these events. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 30fe0d2..b6f4dfb 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2099,11 +2099,8 @@ static void mlxsw_sp_pude_event_func(const struct mlxsw_reg_info *reg, local_port = mlxsw_reg_pude_local_port_get(pude_pl); mlxsw_sp_port = mlxsw_sp->ports[local_port]; - if (!mlxsw_sp_port) { - dev_warn(mlxsw_sp->bus_info->dev, "Port %d: Link event received for non-existent port\n", - local_port); + if (!mlxsw_sp_port) return; - } status = mlxsw_reg_pude_oper_status_get(pude_pl); if (status == MLXSW_PORT_OPER_STATUS_UP) { @@ -2405,16 +2402,10 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, return err; } - err = mlxsw_sp_ports_create(mlxsw_sp); - if (err) { - dev_err(mlxsw_sp->bus_info->dev, "Failed to create ports\n"); - return err; - } - err = mlxsw_sp_event_register(mlxsw_sp, MLXSW_TRAP_ID_PUDE); if (err) { dev_err(mlxsw_sp->bus_info->dev, "Failed to register for PUDE events\n"); - goto err_event_register; + return err; } err = mlxsw_sp_traps_init(mlxsw_sp); @@ -2447,8 +2438,16 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, goto err_switchdev_init; } + err = mlxsw_sp_ports_create(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to create ports\n"); + goto err_ports_create; + } + return 0; +err_ports_create: + mlxsw_sp_switchdev_fini(mlxsw_sp); err_switchdev_init: err_lag_init: mlxsw_sp_buffers_fini(mlxsw_sp); @@ -2457,8 +2456,6 @@ err_flood_init: mlxsw_sp_traps_fini(mlxsw_sp); err_rx_listener_register: mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE); -err_event_register: - mlxsw_sp_ports_remove(mlxsw_sp); return err; } @@ -2466,11 +2463,11 @@ static void mlxsw_sp_fini(struct mlxsw_core *mlxsw_core) { struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + mlxsw_sp_ports_remove(mlxsw_sp); mlxsw_sp_switchdev_fini(mlxsw_sp); mlxsw_sp_buffers_fini(mlxsw_sp); mlxsw_sp_traps_fini(mlxsw_sp); mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE); - mlxsw_sp_ports_remove(mlxsw_sp); WARN_ON(!list_empty(&mlxsw_sp->fids)); } -- cgit v0.10.2 From 464dce188487bcf8c4751a41ff291e367744ef28 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:15 +0200 Subject: mlxsw: spectrum_router: Add basic ipv4 router initialization Create a skeleton router file and do basic HW initialization of router. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/Makefile b/drivers/net/ethernet/mellanox/mlxsw/Makefile index 9b5ebf8..ea05f8a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/Makefile +++ b/drivers/net/ethernet/mellanox/mlxsw/Makefile @@ -7,5 +7,5 @@ obj-$(CONFIG_MLXSW_SWITCHX2) += mlxsw_switchx2.o mlxsw_switchx2-objs := switchx2.o obj-$(CONFIG_MLXSW_SPECTRUM) += mlxsw_spectrum.o mlxsw_spectrum-objs := spectrum.o spectrum_buffers.o \ - spectrum_switchdev.o + spectrum_switchdev.o spectrum_router.o mlxsw_spectrum-$(CONFIG_MLXSW_SPECTRUM_DCB) += spectrum_dcb.o diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index b6f4dfb..d011902 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2438,6 +2438,12 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, goto err_switchdev_init; } + err = mlxsw_sp_router_init(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to initialize router\n"); + goto err_router_init; + } + err = mlxsw_sp_ports_create(mlxsw_sp); if (err) { dev_err(mlxsw_sp->bus_info->dev, "Failed to create ports\n"); @@ -2447,6 +2453,8 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, return 0; err_ports_create: + mlxsw_sp_router_fini(mlxsw_sp); +err_router_init: mlxsw_sp_switchdev_fini(mlxsw_sp); err_switchdev_init: err_lag_init: @@ -2464,6 +2472,7 @@ static void mlxsw_sp_fini(struct mlxsw_core *mlxsw_core) struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); mlxsw_sp_ports_remove(mlxsw_sp); + mlxsw_sp_router_fini(mlxsw_sp); mlxsw_sp_switchdev_fini(mlxsw_sp); mlxsw_sp_buffers_fini(mlxsw_sp); mlxsw_sp_traps_fini(mlxsw_sp); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 05d5fcc..c2ac037 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -74,6 +74,8 @@ #define MLXSW_SP_CELL_FACTOR 2 /* 2 * cell_size / (IPG + cell_size + 1) */ +#define MLXSW_SP_RIF_MAX 800 + static inline u16 mlxsw_sp_pfc_delay_get(int mtu, u16 delay) { delay = MLXSW_SP_BYTES_TO_CELLS(DIV_ROUND_UP(delay, BITS_PER_BYTE)); @@ -411,4 +413,7 @@ static inline void mlxsw_sp_port_dcb_fini(struct mlxsw_sp_port *mlxsw_sp_port) #endif +int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp); +void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp); + #endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c new file mode 100644 index 0000000..8d70496 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -0,0 +1,68 @@ +/* + * drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c + * Copyright (c) 2016 Mellanox Technologies. All rights reserved. + * Copyright (c) 2016 Jiri Pirko + * Copyright (c) 2016 Ido Schimmel + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include +#include + +#include "spectrum.h" +#include "core.h" +#include "reg.h" + +static int __mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) +{ + char rgcr_pl[MLXSW_REG_RGCR_LEN]; + + mlxsw_reg_rgcr_pack(rgcr_pl, true); + mlxsw_reg_rgcr_max_router_interfaces_set(rgcr_pl, MLXSW_SP_RIF_MAX); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rgcr), rgcr_pl); +} + +static void __mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) +{ + char rgcr_pl[MLXSW_REG_RGCR_LEN]; + + mlxsw_reg_rgcr_pack(rgcr_pl, false); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rgcr), rgcr_pl); +} + +int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) +{ + return __mlxsw_sp_router_init(mlxsw_sp); +} + +void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) +{ + __mlxsw_sp_router_fini(mlxsw_sp); +} -- cgit v0.10.2 From fa3054f5a890854c161f37ae76d3649a8eb38aa0 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:16 +0200 Subject: mlxsw: spectrum: Add router interface struct When enabling the router in the device we will represent L3 netdevs using router interfaces (RIFs). These will be specified whenever programming routes or neighbours on the netdev. Introduce the basic RIF infrastructure which allows one to lookup a RIF by its netdev. Later patches in the series will extend this, but the basic routines are needed now in order to direct traffic to CPU. Pointers to the RIF structs are stored in an array indexed by the RIF's number. This will allow us to efficiently update the kernel's neighbour table when regularly dumping the device's table. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index d011902..d15ebf3 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2470,6 +2470,7 @@ err_rx_listener_register: static void mlxsw_sp_fini(struct mlxsw_core *mlxsw_core) { struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + int i; mlxsw_sp_ports_remove(mlxsw_sp); mlxsw_sp_router_fini(mlxsw_sp); @@ -2478,6 +2479,8 @@ static void mlxsw_sp_fini(struct mlxsw_core *mlxsw_core) mlxsw_sp_traps_fini(mlxsw_sp); mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE); WARN_ON(!list_empty(&mlxsw_sp->fids)); + for (i = 0; i < MLXSW_SP_RIF_MAX; i++) + WARN_ON_ONCE(mlxsw_sp->rifs[i]); } static struct mlxsw_config_profile mlxsw_sp_config_profile = { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index c2ac037..83d5807 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -98,6 +98,11 @@ struct mlxsw_sp_fid { u16 vid; }; +struct mlxsw_sp_rif { + struct net_device *dev; + u16 rif; +}; + struct mlxsw_sp_mid { struct list_head list; unsigned char addr[ETH_ALEN]; @@ -169,6 +174,7 @@ struct mlxsw_sp { DECLARE_BITMAP(mapped, MLXSW_SP_MID_MAX); } br_mids; struct list_head fids; /* VLAN-aware bridge FIDs */ + struct mlxsw_sp_rif *rifs[MLXSW_SP_RIF_MAX]; struct mlxsw_sp_port **ports; struct mlxsw_core *core; const struct mlxsw_bus_info *bus_info; @@ -327,6 +333,19 @@ mlxsw_sp_port_vport_find_by_fid(const struct mlxsw_sp_port *mlxsw_sp_port, return NULL; } +static inline struct mlxsw_sp_rif * +mlxsw_sp_rif_find_by_dev(const struct mlxsw_sp *mlxsw_sp, + const struct net_device *dev) +{ + int i; + + for (i = 0; i < MLXSW_SP_RIF_MAX; i++) + if (mlxsw_sp->rifs[i] && mlxsw_sp->rifs[i]->dev == dev) + return mlxsw_sp->rifs[i]; + + return NULL; +} + enum mlxsw_sp_flood_table { MLXSW_SP_FLOOD_TABLE_UC, MLXSW_SP_FLOOD_TABLE_BM, -- cgit v0.10.2 From d82d8c060f682dd2474ddee22e5754a109255912 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:17 +0200 Subject: mlxsw: reg: Add FDB action to forward to router Incoming packets are directed to the router when they match an FDB entry with action forward to IP router. Add this action, which was mistakenly named "TRAP". Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 7f74eb7..ea05e83 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -386,7 +386,9 @@ enum mlxsw_reg_sfd_rec_action { /* forward and trap, trap_id is FDB_TRAP */ MLXSW_REG_SFD_REC_ACTION_MIRROR_TO_CPU = 1, /* trap and do not forward, trap_id is FDB_TRAP */ - MLXSW_REG_SFD_REC_ACTION_TRAP = 3, + MLXSW_REG_SFD_REC_ACTION_TRAP = 2, + /* forward to IP router */ + MLXSW_REG_SFD_REC_ACTION_FORWARD_IP_ROUTER = 3, MLXSW_REG_SFD_REC_ACTION_DISCARD_ERROR = 15, }; -- cgit v0.10.2 From 3dc266896d75039aa9726bf2627b7b08cfd324bf Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:18 +0200 Subject: mlxsw: reg: Add Router Interface Table Register Add the Router Interface Table Register (RITR), which allows us to create and configure router interfaces (RIFs). Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index ea05e83..5ddc1d3 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -3262,6 +3262,198 @@ static inline void mlxsw_reg_rgcr_pack(char *payload, bool ipv4_en) mlxsw_reg_rgcr_ipv4_en_set(payload, ipv4_en); } +/* RITR - Router Interface Table Register + * -------------------------------------- + * The register is used to configure the router interface table. + */ +#define MLXSW_REG_RITR_ID 0x8002 +#define MLXSW_REG_RITR_LEN 0x40 + +static const struct mlxsw_reg_info mlxsw_reg_ritr = { + .id = MLXSW_REG_RITR_ID, + .len = MLXSW_REG_RITR_LEN, +}; + +/* reg_ritr_enable + * Enables routing on the router interface. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, enable, 0x00, 31, 1); + +/* reg_ritr_ipv4 + * IPv4 routing enable. Enables routing of IPv4 traffic on the router + * interface. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, ipv4, 0x00, 29, 1); + +/* reg_ritr_ipv6 + * IPv6 routing enable. Enables routing of IPv6 traffic on the router + * interface. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, ipv6, 0x00, 28, 1); + +enum mlxsw_reg_ritr_if_type { + MLXSW_REG_RITR_VLAN_IF, + MLXSW_REG_RITR_FID_IF, + MLXSW_REG_RITR_SP_IF, +}; + +/* reg_ritr_type + * Router interface type. + * 0 - VLAN interface. + * 1 - FID interface. + * 2 - Sub-port interface. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, type, 0x00, 23, 3); + +enum { + MLXSW_REG_RITR_RIF_CREATE, + MLXSW_REG_RITR_RIF_DEL, +}; + +/* reg_ritr_op + * Opcode: + * 0 - Create or edit RIF. + * 1 - Delete RIF. + * Reserved for SwitchX-2. For Spectrum, editing of interface properties + * is not supported. An interface must be deleted and re-created in order + * to update properties. + * Access: WO + */ +MLXSW_ITEM32(reg, ritr, op, 0x00, 20, 2); + +/* reg_ritr_rif + * Router interface index. A pointer to the Router Interface Table. + * Access: Index + */ +MLXSW_ITEM32(reg, ritr, rif, 0x00, 0, 16); + +/* reg_ritr_ipv4_fe + * IPv4 Forwarding Enable. + * Enables routing of IPv4 traffic on the router interface. When disabled, + * forwarding is blocked but local traffic (traps and IP2ME) will be enabled. + * Not supported in SwitchX-2. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, ipv4_fe, 0x04, 29, 1); + +/* reg_ritr_ipv6_fe + * IPv6 Forwarding Enable. + * Enables routing of IPv6 traffic on the router interface. When disabled, + * forwarding is blocked but local traffic (traps and IP2ME) will be enabled. + * Not supported in SwitchX-2. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, ipv6_fe, 0x04, 28, 1); + +/* reg_ritr_virtual_router + * Virtual router ID associated with the router interface. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, virtual_router, 0x04, 0, 16); + +/* reg_ritr_mtu + * Router interface MTU. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, mtu, 0x34, 0, 16); + +/* reg_ritr_if_swid + * Switch partition ID. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, if_swid, 0x08, 24, 8); + +/* reg_ritr_if_mac + * Router interface MAC address. + * In Spectrum, all MAC addresses must have the same 38 MSBits. + * Access: RW + */ +MLXSW_ITEM_BUF(reg, ritr, if_mac, 0x12, 6); + +/* VLAN Interface */ + +/* reg_ritr_vlan_if_vid + * VLAN ID. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, vlan_if_vid, 0x08, 0, 12); + +/* FID Interface */ + +/* reg_ritr_fid_if_fid + * Filtering ID. Used to connect a bridge to the router. Only FIDs from + * the vFID range are supported. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, fid_if_fid, 0x08, 0, 16); + +static inline void mlxsw_reg_ritr_fid_set(char *payload, + enum mlxsw_reg_ritr_if_type rif_type, + u16 fid) +{ + if (rif_type == MLXSW_REG_RITR_FID_IF) + mlxsw_reg_ritr_fid_if_fid_set(payload, fid); + else + mlxsw_reg_ritr_vlan_if_vid_set(payload, fid); +} + +/* Sub-port Interface */ + +/* reg_ritr_sp_if_lag + * LAG indication. When this bit is set the system_port field holds the + * LAG identifier. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, sp_if_lag, 0x08, 24, 1); + +/* reg_ritr_sp_system_port + * Port unique indentifier. When lag bit is set, this field holds the + * lag_id in bits 0:9. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, sp_if_system_port, 0x08, 0, 16); + +/* reg_ritr_sp_if_vid + * VLAN ID. + * Access: RW + */ +MLXSW_ITEM32(reg, ritr, sp_if_vid, 0x18, 0, 12); + +static inline void mlxsw_reg_ritr_rif_pack(char *payload, u16 rif) +{ + MLXSW_REG_ZERO(ritr, payload); + mlxsw_reg_ritr_rif_set(payload, rif); +} + +static inline void mlxsw_reg_ritr_sp_if_pack(char *payload, bool lag, + u16 system_port, u16 vid) +{ + mlxsw_reg_ritr_sp_if_lag_set(payload, lag); + mlxsw_reg_ritr_sp_if_system_port_set(payload, system_port); + mlxsw_reg_ritr_sp_if_vid_set(payload, vid); +} + +static inline void mlxsw_reg_ritr_pack(char *payload, bool enable, + enum mlxsw_reg_ritr_if_type type, + u16 rif, u16 mtu, const char *mac) +{ + bool op = enable ? MLXSW_REG_RITR_RIF_CREATE : MLXSW_REG_RITR_RIF_DEL; + + MLXSW_REG_ZERO(ritr, payload); + mlxsw_reg_ritr_enable_set(payload, enable); + mlxsw_reg_ritr_ipv4_set(payload, 1); + mlxsw_reg_ritr_type_set(payload, type); + mlxsw_reg_ritr_op_set(payload, op); + mlxsw_reg_ritr_rif_set(payload, rif); + mlxsw_reg_ritr_ipv4_fe_set(payload, 1); + mlxsw_reg_ritr_mtu_set(payload, mtu); + mlxsw_reg_ritr_if_mac_memcpy_to(payload, mac); +} + /* MFCR - Management Fan Control Register * -------------------------------------- * This register controls the settings of the Fan Speed PWM mechanism. @@ -4002,6 +4194,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "HPKT"; case MLXSW_REG_RGCR_ID: return "RGCR"; + case MLXSW_REG_RITR_ID: + return "RITR"; case MLXSW_REG_MFCR_ID: return "MFCR"; case MLXSW_REG_MFSC_ID: -- cgit v0.10.2 From 10f00aa1c410e46febcf09d92211fcb7da9ba5e9 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 2 Jul 2016 11:00:19 +0200 Subject: mlxsw: spectrum: Use action 'discard' when removing traps When removing packet traps we should use action 'discard' instead of 'forward', as some trap IDs we'll add cannot be configured with the later. However, result is the same, as packets are not trapped to the CPU. In the future we will be able to reverse the operation properly by detaching the trap group from the CPU. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index d15ebf3..de30763 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2295,7 +2295,7 @@ err_rx_trap_set: mlxsw_sp); err_rx_listener_register: for (i--; i >= 0; i--) { - mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, + mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_DISCARD, mlxsw_sp_rx_listener[i].trap_id); mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(hpkt), hpkt_pl); @@ -2312,7 +2312,7 @@ static void mlxsw_sp_traps_fini(struct mlxsw_sp *mlxsw_sp) int i; for (i = 0; i < ARRAY_SIZE(mlxsw_sp_rx_listener); i++) { - mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, + mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_DISCARD, mlxsw_sp_rx_listener[i].trap_id); mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(hpkt), hpkt_pl); -- cgit v0.10.2 From 7b27ce7bb9cd6cff0e56d0a677d891b84ea4c665 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Sat, 2 Jul 2016 11:00:20 +0200 Subject: mlxsw: spectrum: Add traps needed for router implementation ip2me: To instruct HW to send trapped ip2me traffic to kernel, we have to add this trap. Selection ip2me traffic is introduced later on in this set. ARPs: We are going to stop flooding to CPU port when netdev isn't bridged and only get packets destined to the netdev's IP address and certain control packets. Add traps for ARP request (broadcast) and response (unicast) in order to get these to the CPU and resolve neighbours. host miss: If a packet is routed through a directly connected route and its destination IP is not in the device's neighbour table, then we need to trap it to CPU. This will cause the host to resolve the MAC of the neighbour, which will be eventually programmed to the device's table. router ingress: In order to trap packets in router part. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index de30763..f0799898 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2255,6 +2255,31 @@ static const struct mlxsw_rx_listener mlxsw_sp_rx_listener[] = { .local_port = MLXSW_PORT_DONT_CARE, .trap_id = MLXSW_TRAP_ID_IGMP_V3_REPORT, }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_ARPBC, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_ARPUC, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_IP2ME, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_RTR_INGRESS0, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_HOST_MISS_IPV4, + }, }; static int mlxsw_sp_traps_init(struct mlxsw_sp *mlxsw_sp) diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h index 53a9550..470d769 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/trap.h +++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h @@ -54,6 +54,11 @@ enum { MLXSW_TRAP_ID_IGMP_V2_REPORT = 0x32, MLXSW_TRAP_ID_IGMP_V2_LEAVE = 0x33, MLXSW_TRAP_ID_IGMP_V3_REPORT = 0x34, + MLXSW_TRAP_ID_ARPBC = 0x50, + MLXSW_TRAP_ID_ARPUC = 0x51, + MLXSW_TRAP_ID_IP2ME = 0x5F, + MLXSW_TRAP_ID_RTR_INGRESS0 = 0x70, + MLXSW_TRAP_ID_HOST_MISS_IPV4 = 0x90, MLXSW_TRAP_ID_MAX = 0x1FF }; -- cgit v0.10.2 From 8186f6e382d8719d0a4bc0ef218c4dd7cf55b496 Mon Sep 17 00:00:00 2001 From: John Crispin Date: Sat, 2 Jul 2016 08:00:50 +0200 Subject: net-next: mediatek: fix compile error inside mtk_poll_controller() commit 8067302973a1 ("net-next: mediatek: add support for IRQ grouping") failed to properly update the irq handling inside mtk_poll_controller() causing compile errors if netconsole was enabled. Fix this by updating the code to use the new separated irq handler function for RX. Signed-off-by: John Crispin Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 7056a37..fbab9b2 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -1297,7 +1297,7 @@ static void mtk_poll_controller(struct net_device *dev) u32 int_mask = MTK_TX_DONE_INT | MTK_RX_DONE_INT; mtk_irq_disable(eth, int_mask); - mtk_handle_irq(dev->irq[0], dev); + mtk_handle_irq_rx(eth->irq[2], dev); mtk_irq_enable(eth, int_mask); } #endif -- cgit v0.10.2 From c37a2dfa67f7920b14ea77dc9f9f9660f7a1f6dd Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Fri, 24 Jun 2016 13:25:22 -0700 Subject: netfilter: Convert FWINV<[foo]> macros and uses to NF_INVF netfilter uses multiple FWINV #defines with identical form that hide a specific structure variable and dereference it with a invflags member. $ git grep "#define FWINV" include/linux/netfilter_bridge/ebtables.h:#define FWINV(bool,invflg) ((bool) ^ !!(info->invflags & invflg)) net/bridge/netfilter/ebtables.c:#define FWINV2(bool, invflg) ((bool) ^ !!(e->invflags & invflg)) net/ipv4/netfilter/arp_tables.c:#define FWINV(bool, invflg) ((bool) ^ !!(arpinfo->invflags & (invflg))) net/ipv4/netfilter/ip_tables.c:#define FWINV(bool, invflg) ((bool) ^ !!(ipinfo->invflags & (invflg))) net/ipv6/netfilter/ip6_tables.c:#define FWINV(bool, invflg) ((bool) ^ !!(ip6info->invflags & (invflg))) net/netfilter/xt_tcpudp.c:#define FWINVTCP(bool, invflg) ((bool) ^ !!(tcpinfo->invflags & (invflg))) Consolidate these macros into a single NF_INVF macro. Miscellanea: o Neaten the alignment around these uses o A few lines are > 80 columns for intelligibility Signed-off-by: Joe Perches Signed-off-by: Pablo Neira Ayuso diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index dc4f58a..e94e81a 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -6,6 +6,10 @@ #include #include +/* Test a struct->invflags and a boolean for inequality */ +#define NF_INVF(ptr, flag, boolean) \ + ((boolean) ^ !!((ptr)->invflags & (flag))) + /** * struct xt_action_param - parameters for matches/targets * diff --git a/include/linux/netfilter_bridge/ebtables.h b/include/linux/netfilter_bridge/ebtables.h index 2ea517c..984b211 100644 --- a/include/linux/netfilter_bridge/ebtables.h +++ b/include/linux/netfilter_bridge/ebtables.h @@ -115,8 +115,6 @@ extern unsigned int ebt_do_table(struct sk_buff *skb, const struct nf_hook_state *state, struct ebt_table *table); -/* Used in the kernel match() functions */ -#define FWINV(bool,invflg) ((bool) ^ !!(info->invflags & invflg)) /* True if the hook mask denotes that the rule is in a base chain, * used in the check() functions */ #define BASE_CHAIN (par->hook_mask & (1 << NF_BR_NUMHOOKS)) diff --git a/net/bridge/netfilter/ebt_802_3.c b/net/bridge/netfilter/ebt_802_3.c index 2a449b7..5fc4aff 100644 --- a/net/bridge/netfilter/ebt_802_3.c +++ b/net/bridge/netfilter/ebt_802_3.c @@ -20,16 +20,16 @@ ebt_802_3_mt(const struct sk_buff *skb, struct xt_action_param *par) __be16 type = hdr->llc.ui.ctrl & IS_UI ? hdr->llc.ui.type : hdr->llc.ni.type; if (info->bitmask & EBT_802_3_SAP) { - if (FWINV(info->sap != hdr->llc.ui.ssap, EBT_802_3_SAP)) + if (NF_INVF(info, EBT_802_3_SAP, info->sap != hdr->llc.ui.ssap)) return false; - if (FWINV(info->sap != hdr->llc.ui.dsap, EBT_802_3_SAP)) + if (NF_INVF(info, EBT_802_3_SAP, info->sap != hdr->llc.ui.dsap)) return false; } if (info->bitmask & EBT_802_3_TYPE) { if (!(hdr->llc.ui.dsap == CHECK_TYPE && hdr->llc.ui.ssap == CHECK_TYPE)) return false; - if (FWINV(info->type != type, EBT_802_3_TYPE)) + if (NF_INVF(info, EBT_802_3_TYPE, info->type != type)) return false; } diff --git a/net/bridge/netfilter/ebt_arp.c b/net/bridge/netfilter/ebt_arp.c index cca0a89..2271422 100644 --- a/net/bridge/netfilter/ebt_arp.c +++ b/net/bridge/netfilter/ebt_arp.c @@ -25,14 +25,14 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) ah = skb_header_pointer(skb, 0, sizeof(_arph), &_arph); if (ah == NULL) return false; - if (info->bitmask & EBT_ARP_OPCODE && FWINV(info->opcode != - ah->ar_op, EBT_ARP_OPCODE)) + if ((info->bitmask & EBT_ARP_OPCODE) && + NF_INVF(info, EBT_ARP_OPCODE, info->opcode != ah->ar_op)) return false; - if (info->bitmask & EBT_ARP_HTYPE && FWINV(info->htype != - ah->ar_hrd, EBT_ARP_HTYPE)) + if ((info->bitmask & EBT_ARP_HTYPE) && + NF_INVF(info, EBT_ARP_HTYPE, info->htype != ah->ar_hrd)) return false; - if (info->bitmask & EBT_ARP_PTYPE && FWINV(info->ptype != - ah->ar_pro, EBT_ARP_PTYPE)) + if ((info->bitmask & EBT_ARP_PTYPE) && + NF_INVF(info, EBT_ARP_PTYPE, info->ptype != ah->ar_pro)) return false; if (info->bitmask & (EBT_ARP_SRC_IP | EBT_ARP_DST_IP | EBT_ARP_GRAT)) { @@ -51,14 +51,16 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) sizeof(daddr), &daddr); if (dap == NULL) return false; - if (info->bitmask & EBT_ARP_SRC_IP && - FWINV(info->saddr != (*sap & info->smsk), EBT_ARP_SRC_IP)) + if ((info->bitmask & EBT_ARP_SRC_IP) && + NF_INVF(info, EBT_ARP_SRC_IP, + info->saddr != (*sap & info->smsk))) return false; - if (info->bitmask & EBT_ARP_DST_IP && - FWINV(info->daddr != (*dap & info->dmsk), EBT_ARP_DST_IP)) + if ((info->bitmask & EBT_ARP_DST_IP) && + NF_INVF(info, EBT_ARP_DST_IP, + info->daddr != (*dap & info->dmsk))) return false; - if (info->bitmask & EBT_ARP_GRAT && - FWINV(*dap != *sap, EBT_ARP_GRAT)) + if ((info->bitmask & EBT_ARP_GRAT) && + NF_INVF(info, EBT_ARP_GRAT, *dap != *sap)) return false; } @@ -73,9 +75,9 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) sizeof(_mac), &_mac); if (mp == NULL) return false; - if (FWINV(!ether_addr_equal_masked(mp, info->smaddr, - info->smmsk), - EBT_ARP_SRC_MAC)) + if (NF_INVF(info, EBT_ARP_SRC_MAC, + !ether_addr_equal_masked(mp, info->smaddr, + info->smmsk))) return false; } @@ -85,9 +87,9 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) sizeof(_mac), &_mac); if (mp == NULL) return false; - if (FWINV(!ether_addr_equal_masked(mp, info->dmaddr, - info->dmmsk), - EBT_ARP_DST_MAC)) + if (NF_INVF(info, EBT_ARP_DST_MAC, + !ether_addr_equal_masked(mp, info->dmaddr, + info->dmmsk))) return false; } } diff --git a/net/bridge/netfilter/ebt_ip.c b/net/bridge/netfilter/ebt_ip.c index 23bca62..d06968b 100644 --- a/net/bridge/netfilter/ebt_ip.c +++ b/net/bridge/netfilter/ebt_ip.c @@ -36,19 +36,19 @@ ebt_ip_mt(const struct sk_buff *skb, struct xt_action_param *par) ih = skb_header_pointer(skb, 0, sizeof(_iph), &_iph); if (ih == NULL) return false; - if (info->bitmask & EBT_IP_TOS && - FWINV(info->tos != ih->tos, EBT_IP_TOS)) + if ((info->bitmask & EBT_IP_TOS) && + NF_INVF(info, EBT_IP_TOS, info->tos != ih->tos)) return false; - if (info->bitmask & EBT_IP_SOURCE && - FWINV((ih->saddr & info->smsk) != - info->saddr, EBT_IP_SOURCE)) + if ((info->bitmask & EBT_IP_SOURCE) && + NF_INVF(info, EBT_IP_SOURCE, + (ih->saddr & info->smsk) != info->saddr)) return false; if ((info->bitmask & EBT_IP_DEST) && - FWINV((ih->daddr & info->dmsk) != - info->daddr, EBT_IP_DEST)) + NF_INVF(info, EBT_IP_DEST, + (ih->daddr & info->dmsk) != info->daddr)) return false; if (info->bitmask & EBT_IP_PROTO) { - if (FWINV(info->protocol != ih->protocol, EBT_IP_PROTO)) + if (NF_INVF(info, EBT_IP_PROTO, info->protocol != ih->protocol)) return false; if (!(info->bitmask & EBT_IP_DPORT) && !(info->bitmask & EBT_IP_SPORT)) @@ -61,16 +61,16 @@ ebt_ip_mt(const struct sk_buff *skb, struct xt_action_param *par) return false; if (info->bitmask & EBT_IP_DPORT) { u32 dst = ntohs(pptr->dst); - if (FWINV(dst < info->dport[0] || - dst > info->dport[1], - EBT_IP_DPORT)) + if (NF_INVF(info, EBT_IP_DPORT, + dst < info->dport[0] || + dst > info->dport[1])) return false; } if (info->bitmask & EBT_IP_SPORT) { u32 src = ntohs(pptr->src); - if (FWINV(src < info->sport[0] || - src > info->sport[1], - EBT_IP_SPORT)) + if (NF_INVF(info, EBT_IP_SPORT, + src < info->sport[0] || + src > info->sport[1])) return false; } } diff --git a/net/bridge/netfilter/ebt_ip6.c b/net/bridge/netfilter/ebt_ip6.c index 98de6e7..4617491 100644 --- a/net/bridge/netfilter/ebt_ip6.c +++ b/net/bridge/netfilter/ebt_ip6.c @@ -45,15 +45,18 @@ ebt_ip6_mt(const struct sk_buff *skb, struct xt_action_param *par) ih6 = skb_header_pointer(skb, 0, sizeof(_ip6h), &_ip6h); if (ih6 == NULL) return false; - if (info->bitmask & EBT_IP6_TCLASS && - FWINV(info->tclass != ipv6_get_dsfield(ih6), EBT_IP6_TCLASS)) + if ((info->bitmask & EBT_IP6_TCLASS) && + NF_INVF(info, EBT_IP6_TCLASS, + info->tclass != ipv6_get_dsfield(ih6))) return false; - if ((info->bitmask & EBT_IP6_SOURCE && - FWINV(ipv6_masked_addr_cmp(&ih6->saddr, &info->smsk, - &info->saddr), EBT_IP6_SOURCE)) || - (info->bitmask & EBT_IP6_DEST && - FWINV(ipv6_masked_addr_cmp(&ih6->daddr, &info->dmsk, - &info->daddr), EBT_IP6_DEST))) + if (((info->bitmask & EBT_IP6_SOURCE) && + NF_INVF(info, EBT_IP6_SOURCE, + ipv6_masked_addr_cmp(&ih6->saddr, &info->smsk, + &info->saddr))) || + ((info->bitmask & EBT_IP6_DEST) && + NF_INVF(info, EBT_IP6_DEST, + ipv6_masked_addr_cmp(&ih6->daddr, &info->dmsk, + &info->daddr)))) return false; if (info->bitmask & EBT_IP6_PROTO) { uint8_t nexthdr = ih6->nexthdr; @@ -63,7 +66,7 @@ ebt_ip6_mt(const struct sk_buff *skb, struct xt_action_param *par) offset_ph = ipv6_skip_exthdr(skb, sizeof(_ip6h), &nexthdr, &frag_off); if (offset_ph == -1) return false; - if (FWINV(info->protocol != nexthdr, EBT_IP6_PROTO)) + if (NF_INVF(info, EBT_IP6_PROTO, info->protocol != nexthdr)) return false; if (!(info->bitmask & (EBT_IP6_DPORT | EBT_IP6_SPORT | EBT_IP6_ICMP6))) @@ -76,22 +79,24 @@ ebt_ip6_mt(const struct sk_buff *skb, struct xt_action_param *par) return false; if (info->bitmask & EBT_IP6_DPORT) { u16 dst = ntohs(pptr->tcpudphdr.dst); - if (FWINV(dst < info->dport[0] || - dst > info->dport[1], EBT_IP6_DPORT)) + if (NF_INVF(info, EBT_IP6_DPORT, + dst < info->dport[0] || + dst > info->dport[1])) return false; } if (info->bitmask & EBT_IP6_SPORT) { u16 src = ntohs(pptr->tcpudphdr.src); - if (FWINV(src < info->sport[0] || - src > info->sport[1], EBT_IP6_SPORT)) + if (NF_INVF(info, EBT_IP6_SPORT, + src < info->sport[0] || + src > info->sport[1])) return false; } if ((info->bitmask & EBT_IP6_ICMP6) && - FWINV(pptr->icmphdr.type < info->icmpv6_type[0] || - pptr->icmphdr.type > info->icmpv6_type[1] || - pptr->icmphdr.code < info->icmpv6_code[0] || - pptr->icmphdr.code > info->icmpv6_code[1], - EBT_IP6_ICMP6)) + NF_INVF(info, EBT_IP6_ICMP6, + pptr->icmphdr.type < info->icmpv6_type[0] || + pptr->icmphdr.type > info->icmpv6_type[1] || + pptr->icmphdr.code < info->icmpv6_code[0] || + pptr->icmphdr.code > info->icmpv6_code[1])) return false; } return true; diff --git a/net/bridge/netfilter/ebt_stp.c b/net/bridge/netfilter/ebt_stp.c index 45f73d5..3140eb9 100644 --- a/net/bridge/netfilter/ebt_stp.c +++ b/net/bridge/netfilter/ebt_stp.c @@ -49,66 +49,68 @@ static bool ebt_filter_config(const struct ebt_stp_info *info, c = &info->config; if ((info->bitmask & EBT_STP_FLAGS) && - FWINV(c->flags != stpc->flags, EBT_STP_FLAGS)) + NF_INVF(info, EBT_STP_FLAGS, c->flags != stpc->flags)) return false; if (info->bitmask & EBT_STP_ROOTPRIO) { v16 = NR16(stpc->root); - if (FWINV(v16 < c->root_priol || v16 > c->root_priou, - EBT_STP_ROOTPRIO)) + if (NF_INVF(info, EBT_STP_ROOTPRIO, + v16 < c->root_priol || v16 > c->root_priou)) return false; } if (info->bitmask & EBT_STP_ROOTADDR) { - if (FWINV(!ether_addr_equal_masked(&stpc->root[2], c->root_addr, - c->root_addrmsk), - EBT_STP_ROOTADDR)) + if (NF_INVF(info, EBT_STP_ROOTADDR, + !ether_addr_equal_masked(&stpc->root[2], + c->root_addr, + c->root_addrmsk))) return false; } if (info->bitmask & EBT_STP_ROOTCOST) { v32 = NR32(stpc->root_cost); - if (FWINV(v32 < c->root_costl || v32 > c->root_costu, - EBT_STP_ROOTCOST)) + if (NF_INVF(info, EBT_STP_ROOTCOST, + v32 < c->root_costl || v32 > c->root_costu)) return false; } if (info->bitmask & EBT_STP_SENDERPRIO) { v16 = NR16(stpc->sender); - if (FWINV(v16 < c->sender_priol || v16 > c->sender_priou, - EBT_STP_SENDERPRIO)) + if (NF_INVF(info, EBT_STP_SENDERPRIO, + v16 < c->sender_priol || v16 > c->sender_priou)) return false; } if (info->bitmask & EBT_STP_SENDERADDR) { - if (FWINV(!ether_addr_equal_masked(&stpc->sender[2], - c->sender_addr, - c->sender_addrmsk), - EBT_STP_SENDERADDR)) + if (NF_INVF(info, EBT_STP_SENDERADDR, + !ether_addr_equal_masked(&stpc->sender[2], + c->sender_addr, + c->sender_addrmsk))) return false; } if (info->bitmask & EBT_STP_PORT) { v16 = NR16(stpc->port); - if (FWINV(v16 < c->portl || v16 > c->portu, EBT_STP_PORT)) + if (NF_INVF(info, EBT_STP_PORT, + v16 < c->portl || v16 > c->portu)) return false; } if (info->bitmask & EBT_STP_MSGAGE) { v16 = NR16(stpc->msg_age); - if (FWINV(v16 < c->msg_agel || v16 > c->msg_ageu, - EBT_STP_MSGAGE)) + if (NF_INVF(info, EBT_STP_MSGAGE, + v16 < c->msg_agel || v16 > c->msg_ageu)) return false; } if (info->bitmask & EBT_STP_MAXAGE) { v16 = NR16(stpc->max_age); - if (FWINV(v16 < c->max_agel || v16 > c->max_ageu, - EBT_STP_MAXAGE)) + if (NF_INVF(info, EBT_STP_MAXAGE, + v16 < c->max_agel || v16 > c->max_ageu)) return false; } if (info->bitmask & EBT_STP_HELLOTIME) { v16 = NR16(stpc->hello_time); - if (FWINV(v16 < c->hello_timel || v16 > c->hello_timeu, - EBT_STP_HELLOTIME)) + if (NF_INVF(info, EBT_STP_HELLOTIME, + v16 < c->hello_timel || v16 > c->hello_timeu)) return false; } if (info->bitmask & EBT_STP_FWDD) { v16 = NR16(stpc->forward_delay); - if (FWINV(v16 < c->forward_delayl || v16 > c->forward_delayu, - EBT_STP_FWDD)) + if (NF_INVF(info, EBT_STP_FWDD, + v16 < c->forward_delayl || v16 > c->forward_delayu)) return false; } return true; @@ -130,8 +132,8 @@ ebt_stp_mt(const struct sk_buff *skb, struct xt_action_param *par) if (memcmp(sp, header, sizeof(header))) return false; - if (info->bitmask & EBT_STP_TYPE && - FWINV(info->type != sp->type, EBT_STP_TYPE)) + if ((info->bitmask & EBT_STP_TYPE) && + NF_INVF(info, EBT_STP_TYPE, info->type != sp->type)) return false; if (sp->type == BPDU_TYPE_CONFIG && diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 5721a25..cceac5b 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -121,7 +121,6 @@ ebt_dev_check(const char *entry, const struct net_device *device) return devname[i] != entry[i] && entry[i] != 1; } -#define FWINV2(bool, invflg) ((bool) ^ !!(e->invflags & invflg)) /* process standard matches */ static inline int ebt_basic_match(const struct ebt_entry *e, const struct sk_buff *skb, @@ -137,34 +136,36 @@ ebt_basic_match(const struct ebt_entry *e, const struct sk_buff *skb, ethproto = h->h_proto; if (e->bitmask & EBT_802_3) { - if (FWINV2(eth_proto_is_802_3(ethproto), EBT_IPROTO)) + if (NF_INVF(e, EBT_IPROTO, eth_proto_is_802_3(ethproto))) return 1; } else if (!(e->bitmask & EBT_NOPROTO) && - FWINV2(e->ethproto != ethproto, EBT_IPROTO)) + NF_INVF(e, EBT_IPROTO, e->ethproto != ethproto)) return 1; - if (FWINV2(ebt_dev_check(e->in, in), EBT_IIN)) + if (NF_INVF(e, EBT_IIN, ebt_dev_check(e->in, in))) return 1; - if (FWINV2(ebt_dev_check(e->out, out), EBT_IOUT)) + if (NF_INVF(e, EBT_IOUT, ebt_dev_check(e->out, out))) return 1; /* rcu_read_lock()ed by nf_hook_slow */ if (in && (p = br_port_get_rcu(in)) != NULL && - FWINV2(ebt_dev_check(e->logical_in, p->br->dev), EBT_ILOGICALIN)) + NF_INVF(e, EBT_ILOGICALIN, + ebt_dev_check(e->logical_in, p->br->dev))) return 1; if (out && (p = br_port_get_rcu(out)) != NULL && - FWINV2(ebt_dev_check(e->logical_out, p->br->dev), EBT_ILOGICALOUT)) + NF_INVF(e, EBT_ILOGICALOUT, + ebt_dev_check(e->logical_out, p->br->dev))) return 1; if (e->bitmask & EBT_SOURCEMAC) { - if (FWINV2(!ether_addr_equal_masked(h->h_source, - e->sourcemac, e->sourcemsk), - EBT_ISOURCE)) + if (NF_INVF(e, EBT_ISOURCE, + !ether_addr_equal_masked(h->h_source, e->sourcemac, + e->sourcemsk))) return 1; } if (e->bitmask & EBT_DESTMAC) { - if (FWINV2(!ether_addr_equal_masked(h->h_dest, - e->destmac, e->destmsk), - EBT_IDEST)) + if (NF_INVF(e, EBT_IDEST, + !ether_addr_equal_masked(h->h_dest, e->destmac, + e->destmsk))) return 1; } return 0; diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 2033f92..c8dd9e2 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -89,22 +89,20 @@ static inline int arp_packet_match(const struct arphdr *arphdr, __be32 src_ipaddr, tgt_ipaddr; long ret; -#define FWINV(bool, invflg) ((bool) ^ !!(arpinfo->invflags & (invflg))) - - if (FWINV((arphdr->ar_op & arpinfo->arpop_mask) != arpinfo->arpop, - ARPT_INV_ARPOP)) + if (NF_INVF(arpinfo, ARPT_INV_ARPOP, + (arphdr->ar_op & arpinfo->arpop_mask) != arpinfo->arpop)) return 0; - if (FWINV((arphdr->ar_hrd & arpinfo->arhrd_mask) != arpinfo->arhrd, - ARPT_INV_ARPHRD)) + if (NF_INVF(arpinfo, ARPT_INV_ARPHRD, + (arphdr->ar_hrd & arpinfo->arhrd_mask) != arpinfo->arhrd)) return 0; - if (FWINV((arphdr->ar_pro & arpinfo->arpro_mask) != arpinfo->arpro, - ARPT_INV_ARPPRO)) + if (NF_INVF(arpinfo, ARPT_INV_ARPPRO, + (arphdr->ar_pro & arpinfo->arpro_mask) != arpinfo->arpro)) return 0; - if (FWINV((arphdr->ar_hln & arpinfo->arhln_mask) != arpinfo->arhln, - ARPT_INV_ARPHLN)) + if (NF_INVF(arpinfo, ARPT_INV_ARPHLN, + (arphdr->ar_hln & arpinfo->arhln_mask) != arpinfo->arhln)) return 0; src_devaddr = arpptr; @@ -115,31 +113,32 @@ static inline int arp_packet_match(const struct arphdr *arphdr, arpptr += dev->addr_len; memcpy(&tgt_ipaddr, arpptr, sizeof(u32)); - if (FWINV(arp_devaddr_compare(&arpinfo->src_devaddr, src_devaddr, dev->addr_len), - ARPT_INV_SRCDEVADDR) || - FWINV(arp_devaddr_compare(&arpinfo->tgt_devaddr, tgt_devaddr, dev->addr_len), - ARPT_INV_TGTDEVADDR)) + if (NF_INVF(arpinfo, ARPT_INV_SRCDEVADDR, + arp_devaddr_compare(&arpinfo->src_devaddr, src_devaddr, + dev->addr_len)) || + NF_INVF(arpinfo, ARPT_INV_TGTDEVADDR, + arp_devaddr_compare(&arpinfo->tgt_devaddr, tgt_devaddr, + dev->addr_len))) return 0; - if (FWINV((src_ipaddr & arpinfo->smsk.s_addr) != arpinfo->src.s_addr, - ARPT_INV_SRCIP) || - FWINV(((tgt_ipaddr & arpinfo->tmsk.s_addr) != arpinfo->tgt.s_addr), - ARPT_INV_TGTIP)) + if (NF_INVF(arpinfo, ARPT_INV_SRCIP, + (src_ipaddr & arpinfo->smsk.s_addr) != arpinfo->src.s_addr) || + NF_INVF(arpinfo, ARPT_INV_TGTIP, + (tgt_ipaddr & arpinfo->tmsk.s_addr) != arpinfo->tgt.s_addr)) return 0; /* Look for ifname matches. */ ret = ifname_compare(indev, arpinfo->iniface, arpinfo->iniface_mask); - if (FWINV(ret != 0, ARPT_INV_VIA_IN)) + if (NF_INVF(arpinfo, ARPT_INV_VIA_IN, ret != 0)) return 0; ret = ifname_compare(outdev, arpinfo->outiface, arpinfo->outiface_mask); - if (FWINV(ret != 0, ARPT_INV_VIA_OUT)) + if (NF_INVF(arpinfo, ARPT_INV_VIA_OUT, ret != 0)) return 0; return 1; -#undef FWINV } static inline int arp_checkentry(const struct arpt_arp *arp) diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 54906e0..f0df66f 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -58,32 +58,31 @@ ip_packet_match(const struct iphdr *ip, { unsigned long ret; -#define FWINV(bool, invflg) ((bool) ^ !!(ipinfo->invflags & (invflg))) - - if (FWINV((ip->saddr&ipinfo->smsk.s_addr) != ipinfo->src.s_addr, - IPT_INV_SRCIP) || - FWINV((ip->daddr&ipinfo->dmsk.s_addr) != ipinfo->dst.s_addr, - IPT_INV_DSTIP)) + if (NF_INVF(ipinfo, IPT_INV_SRCIP, + (ip->saddr & ipinfo->smsk.s_addr) != ipinfo->src.s_addr) || + NF_INVF(ipinfo, IPT_INV_DSTIP, + (ip->daddr & ipinfo->dmsk.s_addr) != ipinfo->dst.s_addr)) return false; ret = ifname_compare_aligned(indev, ipinfo->iniface, ipinfo->iniface_mask); - if (FWINV(ret != 0, IPT_INV_VIA_IN)) + if (NF_INVF(ipinfo, IPT_INV_VIA_IN, ret != 0)) return false; ret = ifname_compare_aligned(outdev, ipinfo->outiface, ipinfo->outiface_mask); - if (FWINV(ret != 0, IPT_INV_VIA_OUT)) + if (NF_INVF(ipinfo, IPT_INV_VIA_OUT, ret != 0)) return false; /* Check specific protocol */ if (ipinfo->proto && - FWINV(ip->protocol != ipinfo->proto, IPT_INV_PROTO)) + NF_INVF(ipinfo, IPT_INV_PROTO, ip->protocol != ipinfo->proto)) return false; /* If we have a fragment rule but the packet is not a fragment * then we return zero */ - if (FWINV((ipinfo->flags&IPT_F_FRAG) && !isfrag, IPT_INV_FRAG)) + if (NF_INVF(ipinfo, IPT_INV_FRAG, + (ipinfo->flags & IPT_F_FRAG) && !isfrag)) return false; return true; @@ -122,7 +121,6 @@ static inline bool unconditional(const struct ipt_entry *e) return e->target_offset == sizeof(struct ipt_entry) && memcmp(&e->ip, &uncond, sizeof(uncond)) == 0; -#undef FWINV } /* for const-correctness */ diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 63e06c3..61ed950 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -73,22 +73,22 @@ ip6_packet_match(const struct sk_buff *skb, unsigned long ret; const struct ipv6hdr *ipv6 = ipv6_hdr(skb); -#define FWINV(bool, invflg) ((bool) ^ !!(ip6info->invflags & (invflg))) - - if (FWINV(ipv6_masked_addr_cmp(&ipv6->saddr, &ip6info->smsk, - &ip6info->src), IP6T_INV_SRCIP) || - FWINV(ipv6_masked_addr_cmp(&ipv6->daddr, &ip6info->dmsk, - &ip6info->dst), IP6T_INV_DSTIP)) + if (NF_INVF(ip6info, IP6T_INV_SRCIP, + ipv6_masked_addr_cmp(&ipv6->saddr, &ip6info->smsk, + &ip6info->src)) || + NF_INVF(ip6info, IP6T_INV_DSTIP, + ipv6_masked_addr_cmp(&ipv6->daddr, &ip6info->dmsk, + &ip6info->dst))) return false; ret = ifname_compare_aligned(indev, ip6info->iniface, ip6info->iniface_mask); - if (FWINV(ret != 0, IP6T_INV_VIA_IN)) + if (NF_INVF(ip6info, IP6T_INV_VIA_IN, ret != 0)) return false; ret = ifname_compare_aligned(outdev, ip6info->outiface, ip6info->outiface_mask); - if (FWINV(ret != 0, IP6T_INV_VIA_OUT)) + if (NF_INVF(ip6info, IP6T_INV_VIA_OUT, ret != 0)) return false; /* ... might want to do something with class and flowlabel here ... */ diff --git a/net/netfilter/xt_tcpudp.c b/net/netfilter/xt_tcpudp.c index c14d464..ade024c 100644 --- a/net/netfilter/xt_tcpudp.c +++ b/net/netfilter/xt_tcpudp.c @@ -83,8 +83,6 @@ static bool tcp_mt(const struct sk_buff *skb, struct xt_action_param *par) return false; } -#define FWINVTCP(bool, invflg) ((bool) ^ !!(tcpinfo->invflags & (invflg))) - th = skb_header_pointer(skb, par->thoff, sizeof(_tcph), &_tcph); if (th == NULL) { /* We've been asked to examine this packet, and we @@ -102,9 +100,8 @@ static bool tcp_mt(const struct sk_buff *skb, struct xt_action_param *par) ntohs(th->dest), !!(tcpinfo->invflags & XT_TCP_INV_DSTPT))) return false; - if (!FWINVTCP((((unsigned char *)th)[13] & tcpinfo->flg_mask) - == tcpinfo->flg_cmp, - XT_TCP_INV_FLAGS)) + if (!NF_INVF(tcpinfo, XT_TCP_INV_FLAGS, + (((unsigned char *)th)[13] & tcpinfo->flg_mask) == tcpinfo->flg_cmp)) return false; if (tcpinfo->option) { if (th->doff * 4 < sizeof(_tcph)) { -- cgit v0.10.2 From f36acc334fb86da3761583dce50faa17a4cbd4de Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Tue, 22 Mar 2016 23:17:00 +0000 Subject: NFC: set info->ram_patch to NULL when it is released When info->ram_patch is released info->otp_patch is being set to NULL rather than info->ram_patch. I believe this is a cut-n-paste bug from almost identical code proceeding it that uses the same idiom for info->otp_patch. Signed-off-by: Colin Ian King Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/fdp/fdp.c b/drivers/nfc/fdp/fdp.c index e44a7a2..d93d314 100644 --- a/drivers/nfc/fdp/fdp.c +++ b/drivers/nfc/fdp/fdp.c @@ -345,7 +345,7 @@ static void fdp_nci_release_firmware(struct nci_dev *ndev) if (info->ram_patch) { release_firmware(info->ram_patch); - info->otp_patch = NULL; + info->ram_patch = NULL; } } -- cgit v0.10.2 From f86dec94e3a86c992a637df1c301a4df25a85801 Mon Sep 17 00:00:00 2001 From: Denys Vlasenko Date: Fri, 15 Apr 2016 18:14:25 +0200 Subject: NFC: hci: delete unused nfc_llc_get_rx_head_tail_room() It used to be EXPORTed, but then EXPORT usage was cleaned up (in 2012), without noticing that the function has no users at all (and curiously, never had any users). Delete it. While at it, remove non-static "inline" hints on nearby functions: these hints don't work across compilation units anyway, and these functions are not used in their .c file, thus they are never inlined. IOW: "inline" here does not help in any way. Signed-off-by: Denys Vlasenko CC: Samuel Ortiz CC: Christophe Ricard CC: linux-wireless@vger.kernel.org CC: linux-kernel@vger.kernel.org Signed-off-by: Samuel Ortiz diff --git a/include/net/nfc/llc.h b/include/net/nfc/llc.h index c25fbde..7ecb457 100644 --- a/include/net/nfc/llc.h +++ b/include/net/nfc/llc.h @@ -37,10 +37,6 @@ struct nfc_llc *nfc_llc_allocate(const char *name, struct nfc_hci_dev *hdev, int tx_tailroom, llc_failure_t llc_failure); void nfc_llc_free(struct nfc_llc *llc); -void nfc_llc_get_rx_head_tail_room(struct nfc_llc *llc, int *rx_headroom, - int *rx_tailroom); - - int nfc_llc_start(struct nfc_llc *llc); int nfc_llc_stop(struct nfc_llc *llc); void nfc_llc_rcv_from_drv(struct nfc_llc *llc, struct sk_buff *skb); diff --git a/net/nfc/hci/llc.c b/net/nfc/hci/llc.c index 1399a03..3d699cb 100644 --- a/net/nfc/hci/llc.c +++ b/net/nfc/hci/llc.c @@ -133,36 +133,29 @@ void nfc_llc_free(struct nfc_llc *llc) kfree(llc); } -inline void nfc_llc_get_rx_head_tail_room(struct nfc_llc *llc, int *rx_headroom, - int *rx_tailroom) -{ - *rx_headroom = llc->rx_headroom; - *rx_tailroom = llc->rx_tailroom; -} - -inline int nfc_llc_start(struct nfc_llc *llc) +int nfc_llc_start(struct nfc_llc *llc) { return llc->ops->start(llc); } EXPORT_SYMBOL(nfc_llc_start); -inline int nfc_llc_stop(struct nfc_llc *llc) +int nfc_llc_stop(struct nfc_llc *llc) { return llc->ops->stop(llc); } EXPORT_SYMBOL(nfc_llc_stop); -inline void nfc_llc_rcv_from_drv(struct nfc_llc *llc, struct sk_buff *skb) +void nfc_llc_rcv_from_drv(struct nfc_llc *llc, struct sk_buff *skb) { llc->ops->rcv_from_drv(llc, skb); } -inline int nfc_llc_xmit_from_hci(struct nfc_llc *llc, struct sk_buff *skb) +int nfc_llc_xmit_from_hci(struct nfc_llc *llc, struct sk_buff *skb) { return llc->ops->xmit_from_hci(llc, skb); } -inline void *nfc_llc_get_data(struct nfc_llc *llc) +void *nfc_llc_get_data(struct nfc_llc *llc) { return llc->data; } -- cgit v0.10.2 From 6d2f70cae481a2e8e85a244d3b2d36828dadff83 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 17 May 2016 10:32:19 +0300 Subject: NFC: pn533: double free on error in probe() We can't pass devm_ allocated pointers to kfree() because they will be freed again after the drive is unloaded. Signed-off-by: Dan Carpenter Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/pn533/usb.c b/drivers/nfc/pn533/usb.c index 8ca0603..33ed78b 100644 --- a/drivers/nfc/pn533/usb.c +++ b/drivers/nfc/pn533/usb.c @@ -464,10 +464,8 @@ static int pn533_usb_probe(struct usb_interface *interface, return -ENOMEM; in_buf = kzalloc(in_buf_len, GFP_KERNEL); - if (!in_buf) { - rc = -ENOMEM; - goto out_free_phy; - } + if (!in_buf) + return -ENOMEM; phy->udev = usb_get_dev(interface_to_usbdev(interface)); phy->interface = interface; @@ -554,8 +552,7 @@ error: usb_free_urb(phy->out_urb); usb_put_dev(phy->udev); kfree(in_buf); -out_free_phy: - kfree(phy); + return rc; } -- cgit v0.10.2 From fa1ce54ea38f7f83473fce62e64fefbd7ebd170e Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 5 Jun 2016 11:17:10 +0200 Subject: NFC: fdp: Detect errors from fdp_nci_create_conn() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit drivers/nfc/fdp/fdp.c: In function ‘fdp_nci_patch_otp’: drivers/nfc/fdp/fdp.c:373: warning: comparison is always false due to limited range of data type drivers/nfc/fdp/fdp.c: In function ‘fdp_nci_patch_ram’: drivers/nfc/fdp/fdp.c:444: warning: comparison is always false due to limited range of data type fdp_nci_create_conn() may return a negative error code, which is silently ignored by assigning it to a u8. Change conn_id from u8 to int to fix this. Fixes: a06347c04c13e380 ("NFC: Add Intel Fields Peak NFC solution driver") Signed-off-by: Geert Uytterhoeven Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/fdp/fdp.c b/drivers/nfc/fdp/fdp.c index d93d314..7c1eaea 100644 --- a/drivers/nfc/fdp/fdp.c +++ b/drivers/nfc/fdp/fdp.c @@ -353,7 +353,7 @@ static int fdp_nci_patch_otp(struct nci_dev *ndev) { struct fdp_nci_info *info = nci_get_drvdata(ndev); struct device *dev = &info->phy->i2c_dev->dev; - u8 conn_id; + int conn_id; int r = 0; if (info->otp_version >= info->otp_patch_version) @@ -424,7 +424,7 @@ static int fdp_nci_patch_ram(struct nci_dev *ndev) { struct fdp_nci_info *info = nci_get_drvdata(ndev); struct device *dev = &info->phy->i2c_dev->dev; - u8 conn_id; + int conn_id; int r = 0; if (info->ram_version >= info->ram_patch_version) -- cgit v0.10.2 From 58d46f538b602d4a93fbae945e5c844a90c01f14 Mon Sep 17 00:00:00 2001 From: Geoff Lansberry Date: Mon, 18 Apr 2016 15:48:39 -0400 Subject: NFC: trf7970a: add TI recommended write of zero to Register 0x18 Signed-off-by: Geoff Lansberry Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/trf7970a.c b/drivers/nfc/trf7970a.c index 10842b7..26c9dbb 100644 --- a/drivers/nfc/trf7970a.c +++ b/drivers/nfc/trf7970a.c @@ -1048,6 +1048,10 @@ static int trf7970a_init(struct trf7970a *trf) if (ret) goto err_out; + ret = trf7970a_write(trf, TRF7970A_NFC_TARGET_LEVEL, 0); + if (ret) + goto err_out; + usleep_range(1000, 2000); trf->chip_status_ctrl &= ~TRF7970A_CHIP_STATUS_RF_ON; -- cgit v0.10.2 From a81ba50a89930a96e34862cf236b4f4461e741ae Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Tue, 7 Jun 2016 16:21:51 +0200 Subject: NFC: port100: Explicitly set NFC-F framing for NFC-DEP When setting the driver framing as NFC_DIGITAL_FRAMING_NFCF_NFC_DEP it used to be already configured as NFC_DIGITAL_FRAMING_NFCF which is the same. So this entry was empty in the in_protocols table. Now that the digital stack can handle PLS requests, it can be changed on the fly from NFC_DIGITAL_FRAMING_NFCA_NFC_DEP. This patch explicitly defines the framing configuration values for NFC_DIGITAL_FRAMING_NFCF_NFC_DEP. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/port100.c b/drivers/nfc/port100.c index 87d5099..2d4bbe3 100644 --- a/drivers/nfc/port100.c +++ b/drivers/nfc/port100.c @@ -343,7 +343,26 @@ in_protocols[][PORT100_IN_MAX_NUM_PROTOCOLS + 1] = { }, [NFC_DIGITAL_FRAMING_NFCF_NFC_DEP] = { /* nfc_digital_framing_nfcf */ - { PORT100_IN_PROT_END, 0 }, + { PORT100_IN_PROT_INITIAL_GUARD_TIME, 18 }, + { PORT100_IN_PROT_ADD_CRC, 1 }, + { PORT100_IN_PROT_CHECK_CRC, 1 }, + { PORT100_IN_PROT_MULTI_CARD, 0 }, + { PORT100_IN_PROT_ADD_PARITY, 0 }, + { PORT100_IN_PROT_CHECK_PARITY, 0 }, + { PORT100_IN_PROT_BITWISE_AC_RECV_MODE, 0 }, + { PORT100_IN_PROT_VALID_BIT_NUMBER, 8 }, + { PORT100_IN_PROT_CRYPTO1, 0 }, + { PORT100_IN_PROT_ADD_SOF, 0 }, + { PORT100_IN_PROT_CHECK_SOF, 0 }, + { PORT100_IN_PROT_ADD_EOF, 0 }, + { PORT100_IN_PROT_CHECK_EOF, 0 }, + { PORT100_IN_PROT_DEAF_TIME, 4 }, + { PORT100_IN_PROT_CRM, 0 }, + { PORT100_IN_PROT_CRM_MIN_LEN, 0 }, + { PORT100_IN_PROT_T1_TAG_FRAME, 0 }, + { PORT100_IN_PROT_RFCA, 0 }, + { PORT100_IN_PROT_GUARD_TIME_AT_INITIATOR, 6 }, + { PORT100_IN_PROT_END, 0 }, }, [NFC_DIGITAL_FRAMING_NFC_DEP_ACTIVATED] = { { PORT100_IN_PROT_END, 0 }, -- cgit v0.10.2 From 7854a44526de84142e367f08288c9f3a33c4c8ee Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Tue, 7 Jun 2016 16:21:52 +0200 Subject: NFC: digital: Add a delay between poll cycles This replaces the polling work struct with a delayed work struct and add a 10 ms delay between 2 poll cycles. This avoids to flood the device with 'switch off'/'switch on' commands. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/include/net/nfc/digital.h b/include/net/nfc/digital.h index 0ae101e..506e3f6 100644 --- a/include/net/nfc/digital.h +++ b/include/net/nfc/digital.h @@ -220,7 +220,7 @@ struct nfc_digital_dev { struct list_head cmd_queue; struct mutex cmd_lock; - struct work_struct poll_work; + struct delayed_work poll_work; u8 curr_protocol; u8 curr_rf_tech; diff --git a/net/nfc/digital_core.c b/net/nfc/digital_core.c index dd9003f..27769ac 100644 --- a/net/nfc/digital_core.c +++ b/net/nfc/digital_core.c @@ -30,6 +30,9 @@ #define DIGITAL_PROTO_ISO15693_RF_TECH NFC_PROTO_ISO15693_MASK +/* Delay between each poll frame (ms) */ +#define DIGITAL_POLL_INTERVAL 10 + struct digital_cmd { struct list_head queue; @@ -419,7 +422,8 @@ void digital_poll_next_tech(struct nfc_digital_dev *ddev) mutex_unlock(&ddev->poll_lock); - schedule_work(&ddev->poll_work); + schedule_delayed_work(&ddev->poll_work, + msecs_to_jiffies(DIGITAL_POLL_INTERVAL)); } static void digital_wq_poll(struct work_struct *work) @@ -428,7 +432,7 @@ static void digital_wq_poll(struct work_struct *work) struct digital_poll_tech *poll_tech; struct nfc_digital_dev *ddev = container_of(work, struct nfc_digital_dev, - poll_work); + poll_work.work); mutex_lock(&ddev->poll_lock); if (!ddev->poll_tech_count) { @@ -543,7 +547,7 @@ static int digital_start_poll(struct nfc_dev *nfc_dev, __u32 im_protocols, return -EINVAL; } - schedule_work(&ddev->poll_work); + schedule_delayed_work(&ddev->poll_work, 0); return 0; } @@ -564,7 +568,7 @@ static void digital_stop_poll(struct nfc_dev *nfc_dev) mutex_unlock(&ddev->poll_lock); - cancel_work_sync(&ddev->poll_work); + cancel_delayed_work_sync(&ddev->poll_work); digital_abort_cmd(ddev); } @@ -770,7 +774,7 @@ struct nfc_digital_dev *nfc_digital_allocate_device(struct nfc_digital_ops *ops, INIT_WORK(&ddev->cmd_complete_work, digital_wq_cmd_complete); mutex_init(&ddev->poll_lock); - INIT_WORK(&ddev->poll_work, digital_wq_poll); + INIT_DELAYED_WORK(&ddev->poll_work, digital_wq_poll); if (supported_protocols & NFC_PROTO_JEWEL_MASK) ddev->protocols |= NFC_PROTO_JEWEL_MASK; @@ -832,7 +836,7 @@ void nfc_digital_unregister_device(struct nfc_digital_dev *ddev) ddev->poll_tech_count = 0; mutex_unlock(&ddev->poll_lock); - cancel_work_sync(&ddev->poll_work); + cancel_delayed_work_sync(&ddev->poll_work); cancel_work_sync(&ddev->cmd_work); cancel_work_sync(&ddev->cmd_complete_work); -- cgit v0.10.2 From 806bfe31c96f77e917eac476ba87164f7bbd1366 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Tue, 7 Jun 2016 16:21:53 +0200 Subject: NFC: llcp: Use dynamic debug for hex dump LLCP skb tx and rx functions now use print_hex_dump_debug() making these verbose traces controllable using dynamic debug. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/llcp_core.c b/net/nfc/llcp_core.c index 9887627..e69786c 100644 --- a/net/nfc/llcp_core.c +++ b/net/nfc/llcp_core.c @@ -732,9 +732,8 @@ static void nfc_llcp_tx_work(struct work_struct *work) int ret; pr_debug("Sending pending skb\n"); - print_hex_dump(KERN_DEBUG, "LLCP Tx: ", - DUMP_PREFIX_OFFSET, 16, 1, - skb->data, skb->len, true); + print_hex_dump_debug("LLCP Tx: ", DUMP_PREFIX_OFFSET, + 16, 1, skb->data, skb->len, true); if (ptype == LLCP_PDU_DISC && sk != NULL && sk->sk_state == LLCP_DISCONNECTING) { @@ -1412,8 +1411,8 @@ static void nfc_llcp_rx_skb(struct nfc_llcp_local *local, struct sk_buff *skb) pr_debug("ptype 0x%x dsap 0x%x ssap 0x%x\n", ptype, dsap, ssap); if (ptype != LLCP_PDU_SYMM) - print_hex_dump(KERN_DEBUG, "LLCP Rx: ", DUMP_PREFIX_OFFSET, - 16, 1, skb->data, skb->len, true); + print_hex_dump_debug("LLCP Rx: ", DUMP_PREFIX_OFFSET, 16, 1, + skb->data, skb->len, true); switch (ptype) { case LLCP_PDU_SYMM: -- cgit v0.10.2 From 204bddcb508fe3bca5c97a9f528bafd7ba8fcec8 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Thu, 23 Jun 2016 11:20:26 +0200 Subject: NFC: nfcsim: Make use of the Digital layer With this complete rewrite, the loopback nfcsim driver now relies on the Digital layer of the nfc stack. As with the previous version, 2 nfc devices are declared when the driver is initialized. The driver supports the NFC_DEP protocol in NFC-A and NFC-F technologies. The 2 devices are using a pair of virtual links for sk_buff exchange. The out-link of one device is the in-link of the other and conversely. To receive data, a device calls nfcsim_link_recv_skb() on its in-link and waits for incoming data on a wait queue. To send data, a device calls nfcsim_link_send_skb() on its out-link which stores the passed skb and signals its wait queue. If the peer device was in the nfcsim_link_recv_skb() call, it will be signaled and will be able to pass the received sk_buff up to the Digital layer. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/nfcsim.c b/drivers/nfc/nfcsim.c index 93aaca5..40b4846 100644 --- a/drivers/nfc/nfcsim.c +++ b/drivers/nfc/nfcsim.c @@ -18,523 +18,427 @@ #include #include #include +#include -#define DEV_ERR(_dev, fmt, args...) nfc_err(&_dev->nfc_dev->dev, \ - "%s: " fmt, __func__, ## args) +#define NFCSIM_ERR(d, fmt, args...) nfc_err(&d->nfc_digital_dev->nfc_dev->dev, \ + "%s: " fmt, __func__, ## args) -#define DEV_DBG(_dev, fmt, args...) dev_dbg(&_dev->nfc_dev->dev, \ - "%s: " fmt, __func__, ## args) +#define NFCSIM_DBG(d, fmt, args...) dev_dbg(&d->nfc_digital_dev->nfc_dev->dev, \ + "%s: " fmt, __func__, ## args) -#define NFCSIM_VERSION "0.1" +#define NFCSIM_VERSION "0.2" -#define NFCSIM_POLL_NONE 0 -#define NFCSIM_POLL_INITIATOR 1 -#define NFCSIM_POLL_TARGET 2 -#define NFCSIM_POLL_DUAL (NFCSIM_POLL_INITIATOR | NFCSIM_POLL_TARGET) +#define NFCSIM_MODE_NONE 0 +#define NFCSIM_MODE_INITIATOR 1 +#define NFCSIM_MODE_TARGET 2 -#define RX_DEFAULT_DELAY 5 +#define NFCSIM_CAPABILITIES (NFC_DIGITAL_DRV_CAPS_IN_CRC | \ + NFC_DIGITAL_DRV_CAPS_TG_CRC) struct nfcsim { - struct nfc_dev *nfc_dev; + struct nfc_digital_dev *nfc_digital_dev; - struct mutex lock; - - struct delayed_work recv_work; + struct work_struct recv_work; + struct delayed_work send_work; - struct sk_buff *clone_skb; + struct nfcsim_link *link_in; + struct nfcsim_link *link_out; - struct delayed_work poll_work; - u8 polling_mode; - u8 curr_polling_mode; + bool up; + u8 mode; + u8 rf_tech; - u8 shutting_down; + u16 recv_timeout; - u8 up; + nfc_digital_cmd_complete_t cb; + void *arg; +}; - u8 initiator; +struct nfcsim_link { + struct mutex lock; - u32 rx_delay; + u8 rf_tech; + u8 mode; - data_exchange_cb_t cb; - void *cb_context; + u8 shutdown; - struct nfcsim *peer_dev; + struct sk_buff *skb; + wait_queue_head_t recv_wait; + u8 cond; }; -static struct nfcsim *dev0; -static struct nfcsim *dev1; - -static struct workqueue_struct *wq; - -static void nfcsim_cleanup_dev(struct nfcsim *dev, u8 shutdown) +static struct nfcsim_link *nfcsim_link_new(void) { - DEV_DBG(dev, "shutdown=%d\n", shutdown); - - mutex_lock(&dev->lock); + struct nfcsim_link *link; - dev->polling_mode = NFCSIM_POLL_NONE; - dev->shutting_down = shutdown; - dev->cb = NULL; - dev_kfree_skb(dev->clone_skb); - dev->clone_skb = NULL; + link = kzalloc(sizeof(struct nfcsim_link), GFP_KERNEL); + if (!link) + return NULL; - mutex_unlock(&dev->lock); + mutex_init(&link->lock); + init_waitqueue_head(&link->recv_wait); - cancel_delayed_work_sync(&dev->poll_work); - cancel_delayed_work_sync(&dev->recv_work); + return link; } -static int nfcsim_target_found(struct nfcsim *dev) +static void nfcsim_link_free(struct nfcsim_link *link) { - struct nfc_target nfc_tgt; + dev_kfree_skb(link->skb); + kfree(link); +} - DEV_DBG(dev, "\n"); +static void nfcsim_link_recv_wake(struct nfcsim_link *link) +{ + link->cond = 1; + wake_up_interruptible(&link->recv_wait); +} - memset(&nfc_tgt, 0, sizeof(struct nfc_target)); +static void nfcsim_link_set_skb(struct nfcsim_link *link, struct sk_buff *skb, + u8 rf_tech, u8 mode) +{ + mutex_lock(&link->lock); - nfc_tgt.supported_protocols = NFC_PROTO_NFC_DEP_MASK; - nfc_targets_found(dev->nfc_dev, &nfc_tgt, 1); + dev_kfree_skb(link->skb); + link->skb = skb; + link->rf_tech = rf_tech; + link->mode = mode; - return 0; + mutex_unlock(&link->lock); } -static int nfcsim_dev_up(struct nfc_dev *nfc_dev) +static void nfcsim_link_recv_cancel(struct nfcsim_link *link) { - struct nfcsim *dev = nfc_get_drvdata(nfc_dev); + mutex_lock(&link->lock); - DEV_DBG(dev, "\n"); + link->mode = NFCSIM_MODE_NONE; - mutex_lock(&dev->lock); + mutex_unlock(&link->lock); - dev->up = 1; - - mutex_unlock(&dev->lock); - - return 0; + nfcsim_link_recv_wake(link); } -static int nfcsim_dev_down(struct nfc_dev *nfc_dev) +static void nfcsim_link_shutdown(struct nfcsim_link *link) { - struct nfcsim *dev = nfc_get_drvdata(nfc_dev); - - DEV_DBG(dev, "\n"); + mutex_lock(&link->lock); - mutex_lock(&dev->lock); + link->shutdown = 1; + link->mode = NFCSIM_MODE_NONE; - dev->up = 0; + mutex_unlock(&link->lock); - mutex_unlock(&dev->lock); - - return 0; + nfcsim_link_recv_wake(link); } -static int nfcsim_dep_link_up(struct nfc_dev *nfc_dev, - struct nfc_target *target, - u8 comm_mode, u8 *gb, size_t gb_len) +static struct sk_buff *nfcsim_link_recv_skb(struct nfcsim_link *link, + int timeout, u8 rf_tech, u8 mode) { int rc; - struct nfcsim *dev = nfc_get_drvdata(nfc_dev); - struct nfcsim *peer = dev->peer_dev; - u8 *remote_gb; - size_t remote_gb_len; + struct sk_buff *skb; - DEV_DBG(dev, "target_idx: %d, comm_mode: %d\n", target->idx, comm_mode); + rc = wait_event_interruptible_timeout(link->recv_wait, + link->cond, + msecs_to_jiffies(timeout)); - mutex_lock(&peer->lock); + mutex_lock(&link->lock); - nfc_tm_activated(peer->nfc_dev, NFC_PROTO_NFC_DEP_MASK, - NFC_COMM_ACTIVE, gb, gb_len); + skb = link->skb; + link->skb = NULL; - remote_gb = nfc_get_local_general_bytes(peer->nfc_dev, &remote_gb_len); - if (!remote_gb) { - DEV_ERR(peer, "Can't get remote general bytes\n"); + if (!rc) { + rc = -ETIMEDOUT; + goto done; + } - mutex_unlock(&peer->lock); - return -EINVAL; + if (!skb || link->rf_tech != rf_tech || link->mode == mode) { + rc = -EINVAL; + goto done; } - mutex_unlock(&peer->lock); + if (link->shutdown) { + rc = -ENODEV; + goto done; + } - mutex_lock(&dev->lock); +done: + mutex_unlock(&link->lock); - rc = nfc_set_remote_general_bytes(nfc_dev, remote_gb, remote_gb_len); - if (rc) { - DEV_ERR(dev, "Can't set remote general bytes\n"); - mutex_unlock(&dev->lock); - return rc; + if (rc < 0) { + dev_kfree_skb(skb); + skb = ERR_PTR(rc); } - rc = nfc_dep_link_is_up(nfc_dev, target->idx, NFC_COMM_ACTIVE, - NFC_RF_INITIATOR); - - mutex_unlock(&dev->lock); + link->cond = 0; - return rc; + return skb; } -static int nfcsim_dep_link_down(struct nfc_dev *nfc_dev) +static void nfcsim_send_wq(struct work_struct *work) { - struct nfcsim *dev = nfc_get_drvdata(nfc_dev); + struct nfcsim *dev = container_of(work, struct nfcsim, send_work.work); - DEV_DBG(dev, "\n"); - - nfcsim_cleanup_dev(dev, 0); - - return 0; + /* + * To effectively send data, the device just wake up its link_out which + * is the link_in of the peer device. The exchanged skb has already been + * stored in the dev->link_out through nfcsim_link_set_skb(). + */ + nfcsim_link_recv_wake(dev->link_out); } -static int nfcsim_start_poll(struct nfc_dev *nfc_dev, - u32 im_protocols, u32 tm_protocols) +static void nfcsim_recv_wq(struct work_struct *work) { - struct nfcsim *dev = nfc_get_drvdata(nfc_dev); - int rc; - - mutex_lock(&dev->lock); + struct nfcsim *dev = container_of(work, struct nfcsim, recv_work); + struct sk_buff *skb; - if (dev->polling_mode != NFCSIM_POLL_NONE) { - DEV_ERR(dev, "Already in polling mode\n"); - rc = -EBUSY; - goto exit; - } + skb = nfcsim_link_recv_skb(dev->link_in, dev->recv_timeout, + dev->rf_tech, dev->mode); - if (im_protocols & NFC_PROTO_NFC_DEP_MASK) - dev->polling_mode |= NFCSIM_POLL_INITIATOR; + if (!dev->up) { + NFCSIM_ERR(dev, "Device is down\n"); - if (tm_protocols & NFC_PROTO_NFC_DEP_MASK) - dev->polling_mode |= NFCSIM_POLL_TARGET; + if (!IS_ERR(skb)) + dev_kfree_skb(skb); - if (dev->polling_mode == NFCSIM_POLL_NONE) { - DEV_ERR(dev, "Unsupported polling mode\n"); - rc = -EINVAL; - goto exit; + skb = ERR_PTR(-ENODEV); } - dev->initiator = 0; - dev->curr_polling_mode = NFCSIM_POLL_NONE; - - queue_delayed_work(wq, &dev->poll_work, 0); - - DEV_DBG(dev, "Start polling: im: 0x%X, tm: 0x%X\n", im_protocols, - tm_protocols); - - rc = 0; -exit: - mutex_unlock(&dev->lock); - - return rc; + dev->cb(dev->nfc_digital_dev, dev->arg, skb); } -static void nfcsim_stop_poll(struct nfc_dev *nfc_dev) +static int nfcsim_send(struct nfc_digital_dev *ddev, struct sk_buff *skb, + u16 timeout, nfc_digital_cmd_complete_t cb, void *arg) { - struct nfcsim *dev = nfc_get_drvdata(nfc_dev); + struct nfcsim *dev = nfc_digital_get_drvdata(ddev); + u8 delay; - DEV_DBG(dev, "Stop poll\n"); + if (!dev->up) { + NFCSIM_ERR(dev, "Device is down\n"); + return -ENODEV; + } + + dev->recv_timeout = timeout; + dev->cb = cb; + dev->arg = arg; - mutex_lock(&dev->lock); + schedule_work(&dev->recv_work); - dev->polling_mode = NFCSIM_POLL_NONE; + if (skb) { + nfcsim_link_set_skb(dev->link_out, skb, dev->rf_tech, + dev->mode); - mutex_unlock(&dev->lock); + /* Add random delay (between 3 and 10 ms) before sending data */ + get_random_bytes(&delay, 1); + delay = 3 + (delay & 0x07); - cancel_delayed_work_sync(&dev->poll_work); + schedule_delayed_work(&dev->send_work, msecs_to_jiffies(delay)); + } + + return 0; } -static int nfcsim_activate_target(struct nfc_dev *nfc_dev, - struct nfc_target *target, u32 protocol) +static void nfcsim_abort_cmd(struct nfc_digital_dev *ddev) { - struct nfcsim *dev = nfc_get_drvdata(nfc_dev); - - DEV_DBG(dev, "\n"); + struct nfcsim *dev = nfc_digital_get_drvdata(ddev); - return -ENOTSUPP; + nfcsim_link_recv_cancel(dev->link_in); } -static void nfcsim_deactivate_target(struct nfc_dev *nfc_dev, - struct nfc_target *target, u8 mode) +static int nfcsim_switch_rf(struct nfc_digital_dev *ddev, bool on) { - struct nfcsim *dev = nfc_get_drvdata(nfc_dev); + struct nfcsim *dev = nfc_digital_get_drvdata(ddev); - DEV_DBG(dev, "\n"); + dev->up = on; + + return 0; } -static void nfcsim_wq_recv(struct work_struct *work) +static int nfcsim_in_configure_hw(struct nfc_digital_dev *ddev, + int type, int param) { - struct nfcsim *dev = container_of(work, struct nfcsim, - recv_work.work); + struct nfcsim *dev = nfc_digital_get_drvdata(ddev); - mutex_lock(&dev->lock); + switch (type) { + case NFC_DIGITAL_CONFIG_RF_TECH: + dev->up = true; + dev->mode = NFCSIM_MODE_INITIATOR; + dev->rf_tech = param; + break; - if (dev->shutting_down || !dev->up || !dev->clone_skb) { - dev_kfree_skb(dev->clone_skb); - goto exit; - } + case NFC_DIGITAL_CONFIG_FRAMING: + break; - if (dev->initiator) { - if (!dev->cb) { - DEV_ERR(dev, "Null recv callback\n"); - dev_kfree_skb(dev->clone_skb); - goto exit; - } - - dev->cb(dev->cb_context, dev->clone_skb, 0); - dev->cb = NULL; - } else { - nfc_tm_data_received(dev->nfc_dev, dev->clone_skb); + default: + NFCSIM_ERR(dev, "Invalid configuration type: %d\n", type); + return -EINVAL; } -exit: - dev->clone_skb = NULL; - - mutex_unlock(&dev->lock); + return 0; } -static int nfcsim_tx(struct nfc_dev *nfc_dev, struct nfc_target *target, - struct sk_buff *skb, data_exchange_cb_t cb, - void *cb_context) +static int nfcsim_in_send_cmd(struct nfc_digital_dev *ddev, + struct sk_buff *skb, u16 timeout, + nfc_digital_cmd_complete_t cb, void *arg) { - struct nfcsim *dev = nfc_get_drvdata(nfc_dev); - struct nfcsim *peer = dev->peer_dev; - int err; - - mutex_lock(&dev->lock); - - if (dev->shutting_down || !dev->up) { - mutex_unlock(&dev->lock); - err = -ENODEV; - goto exit; - } - - dev->cb = cb; - dev->cb_context = cb_context; + return nfcsim_send(ddev, skb, timeout, cb, arg); +} - mutex_unlock(&dev->lock); +static int nfcsim_tg_configure_hw(struct nfc_digital_dev *ddev, + int type, int param) +{ + struct nfcsim *dev = nfc_digital_get_drvdata(ddev); - mutex_lock(&peer->lock); + switch (type) { + case NFC_DIGITAL_CONFIG_RF_TECH: + dev->up = true; + dev->mode = NFCSIM_MODE_TARGET; + dev->rf_tech = param; + break; - peer->clone_skb = skb_clone(skb, GFP_KERNEL); + case NFC_DIGITAL_CONFIG_FRAMING: + break; - if (!peer->clone_skb) { - DEV_ERR(dev, "skb_clone failed\n"); - mutex_unlock(&peer->lock); - err = -ENOMEM; - goto exit; + default: + NFCSIM_ERR(dev, "Invalid configuration type: %d\n", type); + return -EINVAL; } - /* This simulates an arbitrary transmission delay between the 2 devices. - * If packet transmission occurs immediately between them, we have a - * non-stop flow of several tens of thousands SYMM packets per second - * and a burning cpu. - */ - queue_delayed_work(wq, &peer->recv_work, - msecs_to_jiffies(dev->rx_delay)); - - mutex_unlock(&peer->lock); - - err = 0; -exit: - dev_kfree_skb(skb); - - return err; + return 0; } -static int nfcsim_im_transceive(struct nfc_dev *nfc_dev, - struct nfc_target *target, struct sk_buff *skb, - data_exchange_cb_t cb, void *cb_context) +static int nfcsim_tg_send_cmd(struct nfc_digital_dev *ddev, + struct sk_buff *skb, u16 timeout, + nfc_digital_cmd_complete_t cb, void *arg) { - return nfcsim_tx(nfc_dev, target, skb, cb, cb_context); + return nfcsim_send(ddev, skb, timeout, cb, arg); } -static int nfcsim_tm_send(struct nfc_dev *nfc_dev, struct sk_buff *skb) +static int nfcsim_tg_listen(struct nfc_digital_dev *ddev, u16 timeout, + nfc_digital_cmd_complete_t cb, void *arg) { - return nfcsim_tx(nfc_dev, NULL, skb, NULL, NULL); + return nfcsim_send(ddev, NULL, timeout, cb, arg); } -static struct nfc_ops nfcsim_nfc_ops = { - .dev_up = nfcsim_dev_up, - .dev_down = nfcsim_dev_down, - .dep_link_up = nfcsim_dep_link_up, - .dep_link_down = nfcsim_dep_link_down, - .start_poll = nfcsim_start_poll, - .stop_poll = nfcsim_stop_poll, - .activate_target = nfcsim_activate_target, - .deactivate_target = nfcsim_deactivate_target, - .im_transceive = nfcsim_im_transceive, - .tm_send = nfcsim_tm_send, -}; - -static void nfcsim_set_polling_mode(struct nfcsim *dev) -{ - if (dev->polling_mode == NFCSIM_POLL_NONE) { - dev->curr_polling_mode = NFCSIM_POLL_NONE; - return; - } +static struct nfc_digital_ops nfcsim_digital_ops = { + .in_configure_hw = nfcsim_in_configure_hw, + .in_send_cmd = nfcsim_in_send_cmd, - if (dev->curr_polling_mode == NFCSIM_POLL_NONE) { - if (dev->polling_mode & NFCSIM_POLL_INITIATOR) - dev->curr_polling_mode = NFCSIM_POLL_INITIATOR; - else - dev->curr_polling_mode = NFCSIM_POLL_TARGET; + .tg_listen = nfcsim_tg_listen, + .tg_configure_hw = nfcsim_tg_configure_hw, + .tg_send_cmd = nfcsim_tg_send_cmd, - return; - } - - if (dev->polling_mode == NFCSIM_POLL_DUAL) { - if (dev->curr_polling_mode == NFCSIM_POLL_TARGET) - dev->curr_polling_mode = NFCSIM_POLL_INITIATOR; - else - dev->curr_polling_mode = NFCSIM_POLL_TARGET; - } -} + .abort_cmd = nfcsim_abort_cmd, + .switch_rf = nfcsim_switch_rf, +}; -static void nfcsim_wq_poll(struct work_struct *work) +static struct nfcsim *nfcsim_device_new(struct nfcsim_link *link_in, + struct nfcsim_link *link_out) { - struct nfcsim *dev = container_of(work, struct nfcsim, poll_work.work); - struct nfcsim *peer = dev->peer_dev; + struct nfcsim *dev; + int rc; - /* These work items run on an ordered workqueue and are therefore - * serialized. So we can take both mutexes without being dead locked. - */ - mutex_lock(&dev->lock); - mutex_lock(&peer->lock); + dev = kzalloc(sizeof(struct nfcsim), GFP_KERNEL); + if (!dev) + return ERR_PTR(-ENOMEM); - nfcsim_set_polling_mode(dev); + INIT_DELAYED_WORK(&dev->send_work, nfcsim_send_wq); + INIT_WORK(&dev->recv_work, nfcsim_recv_wq); - if (dev->curr_polling_mode == NFCSIM_POLL_NONE) { - DEV_DBG(dev, "Not polling\n"); - goto unlock; + dev->nfc_digital_dev = + nfc_digital_allocate_device(&nfcsim_digital_ops, + NFC_PROTO_NFC_DEP_MASK, + NFCSIM_CAPABILITIES, + 0, 0); + if (!dev->nfc_digital_dev) { + kfree(dev); + return ERR_PTR(-ENOMEM); } - DEV_DBG(dev, "Polling as %s", - dev->curr_polling_mode == NFCSIM_POLL_INITIATOR ? - "initiator\n" : "target\n"); - - if (dev->curr_polling_mode == NFCSIM_POLL_TARGET) - goto sched_work; + nfc_digital_set_drvdata(dev->nfc_digital_dev, dev); - if (peer->curr_polling_mode == NFCSIM_POLL_TARGET) { - peer->polling_mode = NFCSIM_POLL_NONE; - dev->polling_mode = NFCSIM_POLL_NONE; + dev->link_in = link_in; + dev->link_out = link_out; - dev->initiator = 1; - - nfcsim_target_found(dev); + rc = nfc_digital_register_device(dev->nfc_digital_dev); + if (rc) { + pr_err("Could not register digital device (%d)\n", rc); + nfc_digital_free_device(dev->nfc_digital_dev); + kfree(dev); - goto unlock; + return ERR_PTR(rc); } -sched_work: - /* This defines the delay for an initiator to check if the other device - * is polling in target mode. - * If the device starts in dual mode polling, it switches between - * initiator and target at every round. - * Because the wq is ordered and only 1 work item is executed at a time, - * we'll always have one device polling as initiator and the other as - * target at some point, even if both are started in dual mode. - */ - queue_delayed_work(wq, &dev->poll_work, msecs_to_jiffies(200)); - -unlock: - mutex_unlock(&peer->lock); - mutex_unlock(&dev->lock); + return dev; } -static struct nfcsim *nfcsim_init_dev(void) +static void nfcsim_device_free(struct nfcsim *dev) { - struct nfcsim *dev; - int rc = -ENOMEM; - - dev = kzalloc(sizeof(*dev), GFP_KERNEL); - if (dev == NULL) - return ERR_PTR(-ENOMEM); - - mutex_init(&dev->lock); - - INIT_DELAYED_WORK(&dev->recv_work, nfcsim_wq_recv); - INIT_DELAYED_WORK(&dev->poll_work, nfcsim_wq_poll); - - dev->nfc_dev = nfc_allocate_device(&nfcsim_nfc_ops, - NFC_PROTO_NFC_DEP_MASK, - 0, 0); - if (!dev->nfc_dev) - goto error; + nfc_digital_unregister_device(dev->nfc_digital_dev); - nfc_set_drvdata(dev->nfc_dev, dev); + dev->up = false; - rc = nfc_register_device(dev->nfc_dev); - if (rc) - goto free_nfc_dev; + nfcsim_link_shutdown(dev->link_in); - dev->rx_delay = RX_DEFAULT_DELAY; - return dev; + cancel_delayed_work_sync(&dev->send_work); + cancel_work_sync(&dev->recv_work); -free_nfc_dev: - nfc_free_device(dev->nfc_dev); + nfc_digital_free_device(dev->nfc_digital_dev); -error: kfree(dev); - - return ERR_PTR(rc); } -static void nfcsim_free_device(struct nfcsim *dev) -{ - nfc_unregister_device(dev->nfc_dev); - - nfc_free_device(dev->nfc_dev); - - kfree(dev); -} +static struct nfcsim *dev0; +static struct nfcsim *dev1; static int __init nfcsim_init(void) { + struct nfcsim_link *link0, *link1; int rc; - /* We need an ordered wq to ensure that poll_work items are executed - * one at a time. - */ - wq = alloc_ordered_workqueue("nfcsim", 0); - if (!wq) { + link0 = nfcsim_link_new(); + link1 = nfcsim_link_new(); + if (!link0 || !link1) { rc = -ENOMEM; - goto exit; + goto exit_err; } - dev0 = nfcsim_init_dev(); + dev0 = nfcsim_device_new(link0, link1); if (IS_ERR(dev0)) { rc = PTR_ERR(dev0); - goto exit; + goto exit_err; } - dev1 = nfcsim_init_dev(); + dev1 = nfcsim_device_new(link1, link0); if (IS_ERR(dev1)) { - kfree(dev0); + nfcsim_device_free(dev0); rc = PTR_ERR(dev1); - goto exit; + goto exit_err; } - dev0->peer_dev = dev1; - dev1->peer_dev = dev0; + pr_info("nfcsim " NFCSIM_VERSION " initialized\n"); - pr_debug("NFCsim " NFCSIM_VERSION " initialized\n"); + return 0; + +exit_err: + pr_err("Failed to initialize nfcsim driver (%d)\n", rc); - rc = 0; -exit: - if (rc) - pr_err("Failed to initialize nfcsim driver (%d)\n", - rc); + nfcsim_link_free(link0); + nfcsim_link_free(link1); return rc; } static void __exit nfcsim_exit(void) { - nfcsim_cleanup_dev(dev0, 1); - nfcsim_cleanup_dev(dev1, 1); + struct nfcsim_link *link0, *link1; + + link0 = dev0->link_in; + link1 = dev0->link_out; - nfcsim_free_device(dev0); - nfcsim_free_device(dev1); + nfcsim_device_free(dev0); + nfcsim_device_free(dev1); - destroy_workqueue(wq); + nfcsim_link_free(link0); + nfcsim_link_free(link1); } module_init(nfcsim_init); -- cgit v0.10.2 From 09748a22f4ab7b0ab5a83c432f6e18f65f18e09b Mon Sep 17 00:00:00 2001 From: Matthias Schiffer Date: Mon, 9 May 2016 18:41:08 +0200 Subject: batman-adv: add generic netlink family for batman-adv debugfs is currently severely broken virtually everywhere in the kernel where files are dynamically added and removed (see http://lkml.iu.edu/hypermail/linux/kernel/1506.1/02196.html for some details). In addition to that, debugfs is not namespace-aware. Instead of adding new debugfs entries, the whole infrastructure should be moved to netlink. This will fix the long standing problem of large buffers for debug tables and hard to parse text files. Signed-off-by: Matthias Schiffer Signed-off-by: Andrew Lunn [sven.eckelmann@open-mesh.com: Strip down patch to only add genl family, add missing kerneldoc] Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/MAINTAINERS b/MAINTAINERS index 0e26025..e25091f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2291,6 +2291,7 @@ S: Maintained F: Documentation/ABI/testing/sysfs-class-net-batman-adv F: Documentation/ABI/testing/sysfs-class-net-mesh F: Documentation/networking/batman-adv.txt +F: include/uapi/linux/batman_adv.h F: net/batman-adv/ BAYCOM/HDLCDRV DRIVERS FOR AX.25 diff --git a/include/uapi/linux/batman_adv.h b/include/uapi/linux/batman_adv.h new file mode 100644 index 0000000..79f7972 --- /dev/null +++ b/include/uapi/linux/batman_adv.h @@ -0,0 +1,53 @@ +/* Copyright (C) 2016 B.A.T.M.A.N. contributors: + * + * Matthias Schiffer + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#ifndef _UAPI_LINUX_BATMAN_ADV_H_ +#define _UAPI_LINUX_BATMAN_ADV_H_ + +#define BATADV_NL_NAME "batadv" + +/** + * enum batadv_nl_attrs - batman-adv netlink attributes + * + * @BATADV_ATTR_UNSPEC: unspecified attribute to catch errors + * @__BATADV_ATTR_AFTER_LAST: internal use + * @NUM_BATADV_ATTR: total number of batadv_nl_attrs available + * @BATADV_ATTR_MAX: highest attribute number currently defined + */ +enum batadv_nl_attrs { + BATADV_ATTR_UNSPEC, + /* add attributes above here, update the policy in netlink.c */ + __BATADV_ATTR_AFTER_LAST, + NUM_BATADV_ATTR = __BATADV_ATTR_AFTER_LAST, + BATADV_ATTR_MAX = __BATADV_ATTR_AFTER_LAST - 1 +}; + +/** + * enum batadv_nl_commands - supported batman-adv netlink commands + * + * @BATADV_CMD_UNSPEC: unspecified command to catch errors + * @__BATADV_CMD_AFTER_LAST: internal use + * @BATADV_CMD_MAX: highest used command number + */ +enum batadv_nl_commands { + BATADV_CMD_UNSPEC, + /* add new commands above here */ + __BATADV_CMD_AFTER_LAST, + BATADV_CMD_MAX = __BATADV_CMD_AFTER_LAST - 1 +}; + +#endif /* _UAPI_LINUX_BATMAN_ADV_H_ */ diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile index a55f4ec..7da5901 100644 --- a/net/batman-adv/Makefile +++ b/net/batman-adv/Makefile @@ -35,6 +35,7 @@ batman-adv-y += icmp_socket.o batman-adv-$(CONFIG_BATMAN_ADV_DEBUG) += log.o batman-adv-y += main.o batman-adv-$(CONFIG_BATMAN_ADV_MCAST) += multicast.o +batman-adv-y += netlink.o batman-adv-$(CONFIG_BATMAN_ADV_NC) += network-coding.o batman-adv-y += originator.o batman-adv-y += routing.o diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index eab9d1b..275604b 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -57,6 +57,7 @@ #include "icmp_socket.h" #include "log.h" #include "multicast.h" +#include "netlink.h" #include "network-coding.h" #include "originator.h" #include "packet.h" @@ -99,6 +100,7 @@ static int __init batadv_init(void) register_netdevice_notifier(&batadv_hard_if_notifier); rtnl_link_register(&batadv_link_ops); + batadv_netlink_register(); pr_info("B.A.T.M.A.N. advanced %s (compatibility version %i) loaded\n", BATADV_SOURCE_VERSION, BATADV_COMPAT_VERSION); @@ -109,6 +111,7 @@ static int __init batadv_init(void) static void __exit batadv_exit(void) { batadv_debugfs_destroy(); + batadv_netlink_unregister(); rtnl_link_unregister(&batadv_link_ops); unregister_netdevice_notifier(&batadv_hard_if_notifier); batadv_hardif_remove_interfaces(); diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c new file mode 100644 index 0000000..9e4c865 --- /dev/null +++ b/net/batman-adv/netlink.c @@ -0,0 +1,57 @@ +/* Copyright (C) 2016 B.A.T.M.A.N. contributors: + * + * Matthias Schiffer + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#include "netlink.h" +#include "main.h" + +#include +#include +#include +#include +#include + +static struct genl_family batadv_netlink_family = { + .id = GENL_ID_GENERATE, + .hdrsize = 0, + .name = BATADV_NL_NAME, + .version = 1, + .maxattr = BATADV_ATTR_MAX, +}; + +static struct genl_ops batadv_netlink_ops[] = { +}; + +/** + * batadv_netlink_register - register batadv genl netlink family + */ +void __init batadv_netlink_register(void) +{ + int ret; + + ret = genl_register_family_with_ops(&batadv_netlink_family, + batadv_netlink_ops); + if (ret) + pr_warn("unable to register netlink family"); +} + +/** + * batadv_netlink_unregister - unregister batadv genl netlink family + */ +void batadv_netlink_unregister(void) +{ + genl_unregister_family(&batadv_netlink_family); +} diff --git a/net/batman-adv/netlink.h b/net/batman-adv/netlink.h new file mode 100644 index 0000000..39044cc --- /dev/null +++ b/net/batman-adv/netlink.h @@ -0,0 +1,26 @@ +/* Copyright (C) 2016 B.A.T.M.A.N. contributors: + * + * Matthias Schiffer + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#ifndef _NET_BATMAN_ADV_NETLINK_H_ +#define _NET_BATMAN_ADV_NETLINK_H_ + +#include "main.h" + +void batadv_netlink_register(void); +void batadv_netlink_unregister(void); + +#endif /* _NET_BATMAN_ADV_NETLINK_H_ */ -- cgit v0.10.2 From 5da0aef5e93591b373010c10f374c4161b37728c Mon Sep 17 00:00:00 2001 From: Matthias Schiffer Date: Mon, 9 May 2016 18:41:09 +0200 Subject: batman-adv: add netlink command to query generic mesh information files BATADV_CMD_GET_MESH_INFO is used to query basic information about a batman-adv softif (name, index and MAC address for both the softif and the primary hardif; routing algorithm; batman-adv version). Signed-off-by: Matthias Schiffer Signed-off-by: Andrew Lunn [sven.eckelmann@open-mesh.com: Reduce the number of changes to BATADV_CMD_GET_MESH_INFO, add missing kerneldoc, add policy for attributes] Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/include/uapi/linux/batman_adv.h b/include/uapi/linux/batman_adv.h index 79f7972..c39623c7 100644 --- a/include/uapi/linux/batman_adv.h +++ b/include/uapi/linux/batman_adv.h @@ -24,12 +24,28 @@ * enum batadv_nl_attrs - batman-adv netlink attributes * * @BATADV_ATTR_UNSPEC: unspecified attribute to catch errors + * @BATADV_ATTR_VERSION: batman-adv version string + * @BATADV_ATTR_ALGO_NAME: name of routing algorithm + * @BATADV_ATTR_MESH_IFINDEX: index of the batman-adv interface + * @BATADV_ATTR_MESH_IFNAME: name of the batman-adv interface + * @BATADV_ATTR_MESH_ADDRESS: mac address of the batman-adv interface + * @BATADV_ATTR_HARD_IFINDEX: index of the non-batman-adv interface + * @BATADV_ATTR_HARD_IFNAME: name of the non-batman-adv interface + * @BATADV_ATTR_HARD_ADDRESS: mac address of the non-batman-adv interface * @__BATADV_ATTR_AFTER_LAST: internal use * @NUM_BATADV_ATTR: total number of batadv_nl_attrs available * @BATADV_ATTR_MAX: highest attribute number currently defined */ enum batadv_nl_attrs { BATADV_ATTR_UNSPEC, + BATADV_ATTR_VERSION, + BATADV_ATTR_ALGO_NAME, + BATADV_ATTR_MESH_IFINDEX, + BATADV_ATTR_MESH_IFNAME, + BATADV_ATTR_MESH_ADDRESS, + BATADV_ATTR_HARD_IFINDEX, + BATADV_ATTR_HARD_IFNAME, + BATADV_ATTR_HARD_ADDRESS, /* add attributes above here, update the policy in netlink.c */ __BATADV_ATTR_AFTER_LAST, NUM_BATADV_ATTR = __BATADV_ATTR_AFTER_LAST, @@ -40,11 +56,13 @@ enum batadv_nl_attrs { * enum batadv_nl_commands - supported batman-adv netlink commands * * @BATADV_CMD_UNSPEC: unspecified command to catch errors + * @BATADV_CMD_GET_MESH_INFO: Query basic information about batman-adv device * @__BATADV_CMD_AFTER_LAST: internal use * @BATADV_CMD_MAX: highest used command number */ enum batadv_nl_commands { BATADV_CMD_UNSPEC, + BATADV_CMD_GET_MESH_INFO, /* add new commands above here */ __BATADV_CMD_AFTER_LAST, BATADV_CMD_MAX = __BATADV_CMD_AFTER_LAST - 1 diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c index 9e4c865..68152aa 100644 --- a/net/batman-adv/netlink.c +++ b/net/batman-adv/netlink.c @@ -18,12 +18,24 @@ #include "netlink.h" #include "main.h" +#include +#include #include +#include #include +#include +#include #include +#include #include +#include #include +#include "hard-interface.h" +#include "soft-interface.h" + +struct sk_buff; + static struct genl_family batadv_netlink_family = { .id = GENL_ID_GENERATE, .hdrsize = 0, @@ -32,7 +44,132 @@ static struct genl_family batadv_netlink_family = { .maxattr = BATADV_ATTR_MAX, }; +static struct nla_policy batadv_netlink_policy[NUM_BATADV_ATTR] = { + [BATADV_ATTR_VERSION] = { .type = NLA_STRING }, + [BATADV_ATTR_ALGO_NAME] = { .type = NLA_STRING }, + [BATADV_ATTR_MESH_IFINDEX] = { .type = NLA_U32 }, + [BATADV_ATTR_MESH_IFNAME] = { .type = NLA_STRING }, + [BATADV_ATTR_MESH_ADDRESS] = { .len = ETH_ALEN }, + [BATADV_ATTR_HARD_IFINDEX] = { .type = NLA_U32 }, + [BATADV_ATTR_HARD_IFNAME] = { .type = NLA_STRING }, + [BATADV_ATTR_HARD_ADDRESS] = { .len = ETH_ALEN }, +}; + +/** + * batadv_netlink_mesh_info_put - fill in generic information about mesh + * interface + * @msg: netlink message to be sent back + * @soft_iface: interface for which the data should be taken + * + * Return: 0 on success, < 0 on error + */ +static int +batadv_netlink_mesh_info_put(struct sk_buff *msg, struct net_device *soft_iface) +{ + struct batadv_priv *bat_priv = netdev_priv(soft_iface); + struct batadv_hard_iface *primary_if = NULL; + struct net_device *hard_iface; + int ret = -ENOBUFS; + + if (nla_put_string(msg, BATADV_ATTR_VERSION, BATADV_SOURCE_VERSION) || + nla_put_string(msg, BATADV_ATTR_ALGO_NAME, + bat_priv->bat_algo_ops->name) || + nla_put_u32(msg, BATADV_ATTR_MESH_IFINDEX, soft_iface->ifindex) || + nla_put_string(msg, BATADV_ATTR_MESH_IFNAME, soft_iface->name) || + nla_put(msg, BATADV_ATTR_MESH_ADDRESS, ETH_ALEN, + soft_iface->dev_addr)) + goto out; + + primary_if = batadv_primary_if_get_selected(bat_priv); + if (primary_if && primary_if->if_status == BATADV_IF_ACTIVE) { + hard_iface = primary_if->net_dev; + + if (nla_put_u32(msg, BATADV_ATTR_HARD_IFINDEX, + hard_iface->ifindex) || + nla_put_string(msg, BATADV_ATTR_HARD_IFNAME, + hard_iface->name) || + nla_put(msg, BATADV_ATTR_HARD_ADDRESS, ETH_ALEN, + hard_iface->dev_addr)) + goto out; + } + + ret = 0; + + out: + if (primary_if) + batadv_hardif_put(primary_if); + + return ret; +} + +/** + * batadv_netlink_get_mesh_info - handle incoming BATADV_CMD_GET_MESH_INFO + * netlink request + * @skb: received netlink message + * @info: receiver information + * + * Return: 0 on success, < 0 on error + */ +static int +batadv_netlink_get_mesh_info(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct net_device *soft_iface; + struct sk_buff *msg = NULL; + void *msg_head; + int ifindex; + int ret; + + if (!info->attrs[BATADV_ATTR_MESH_IFINDEX]) + return -EINVAL; + + ifindex = nla_get_u32(info->attrs[BATADV_ATTR_MESH_IFINDEX]); + if (!ifindex) + return -EINVAL; + + soft_iface = dev_get_by_index(net, ifindex); + if (!soft_iface || !batadv_softif_is_valid(soft_iface)) { + ret = -ENODEV; + goto out; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) { + ret = -ENOMEM; + goto out; + } + + msg_head = genlmsg_put(msg, info->snd_portid, info->snd_seq, + &batadv_netlink_family, 0, + BATADV_CMD_GET_MESH_INFO); + if (!msg_head) { + ret = -ENOBUFS; + goto out; + } + + ret = batadv_netlink_mesh_info_put(msg, soft_iface); + + out: + if (soft_iface) + dev_put(soft_iface); + + if (ret) { + if (msg) + nlmsg_free(msg); + return ret; + } + + genlmsg_end(msg, msg_head); + return genlmsg_reply(msg, info); +} + static struct genl_ops batadv_netlink_ops[] = { + { + .cmd = BATADV_CMD_GET_MESH_INFO, + .flags = GENL_ADMIN_PERM, + .policy = batadv_netlink_policy, + .doit = batadv_netlink_get_mesh_info, + }, }; /** -- cgit v0.10.2 From f50ca95a691e9fd1fce530aade58c98d621cb1fe Mon Sep 17 00:00:00 2001 From: Antonio Quartulli Date: Wed, 18 May 2016 11:38:48 +0200 Subject: batman-adv: return netdev status in the TX path Return the proper netdev TX status along the TX path so that the tp_meter can understand when the queue is full and should stop sending packets. Signed-off-by: Antonio Quartulli Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c index 9f41a0a..0934730 100644 --- a/net/batman-adv/fragmentation.c +++ b/net/batman-adv/fragmentation.c @@ -433,11 +433,12 @@ err: * @orig_node: final destination of the created fragments * @neigh_node: next-hop of the created fragments * - * Return: true on success, false otherwise. + * Return: the netdev tx status or -1 in case of error. + * When -1 is returned the skb is not consumed. */ -bool batadv_frag_send_packet(struct sk_buff *skb, - struct batadv_orig_node *orig_node, - struct batadv_neigh_node *neigh_node) +int batadv_frag_send_packet(struct sk_buff *skb, + struct batadv_orig_node *orig_node, + struct batadv_neigh_node *neigh_node) { struct batadv_priv *bat_priv; struct batadv_hard_iface *primary_if = NULL; @@ -446,7 +447,7 @@ bool batadv_frag_send_packet(struct sk_buff *skb, unsigned int mtu = neigh_node->if_incoming->net_dev->mtu; unsigned int header_size = sizeof(frag_header); unsigned int max_fragment_size, max_packet_size; - bool ret = false; + int ret = -1; /* To avoid merge and refragmentation at next-hops we never send * fragments larger than BATADV_FRAG_MAX_FRAG_SIZE @@ -457,12 +458,12 @@ bool batadv_frag_send_packet(struct sk_buff *skb, /* Don't even try to fragment, if we need more than 16 fragments */ if (skb->len > max_packet_size) - goto out_err; + goto out; bat_priv = orig_node->bat_priv; primary_if = batadv_primary_if_get_selected(bat_priv); if (!primary_if) - goto out_err; + goto out; /* Create one header to be copied to all fragments */ frag_header.packet_type = BATADV_UNICAST_FRAG; @@ -488,23 +489,33 @@ bool batadv_frag_send_packet(struct sk_buff *skb, while (skb->len > max_fragment_size) { skb_fragment = batadv_frag_create(skb, &frag_header, mtu); if (!skb_fragment) - goto out_err; + goto out; batadv_inc_counter(bat_priv, BATADV_CNT_FRAG_TX); batadv_add_counter(bat_priv, BATADV_CNT_FRAG_TX_BYTES, skb_fragment->len + ETH_HLEN); - batadv_send_unicast_skb(skb_fragment, neigh_node); + ret = batadv_send_unicast_skb(skb_fragment, neigh_node); + if (ret != NET_XMIT_SUCCESS) { + /* return -1 so that the caller can free the original + * skb + */ + ret = -1; + goto out; + } + frag_header.no++; /* The initial check in this function should cover this case */ - if (frag_header.no == BATADV_FRAG_MAX_FRAGMENTS - 1) - goto out_err; + if (frag_header.no == BATADV_FRAG_MAX_FRAGMENTS - 1) { + ret = -1; + goto out; + } } /* Make room for the fragment header. */ if (batadv_skb_head_push(skb, header_size) < 0 || pskb_expand_head(skb, header_size + ETH_HLEN, 0, GFP_ATOMIC) < 0) - goto out_err; + goto out; memcpy(skb->data, &frag_header, header_size); @@ -512,11 +523,9 @@ bool batadv_frag_send_packet(struct sk_buff *skb, batadv_inc_counter(bat_priv, BATADV_CNT_FRAG_TX); batadv_add_counter(bat_priv, BATADV_CNT_FRAG_TX_BYTES, skb->len + ETH_HLEN); - batadv_send_unicast_skb(skb, neigh_node); + ret = batadv_send_unicast_skb(skb, neigh_node); - ret = true; - -out_err: +out: if (primary_if) batadv_hardif_put(primary_if); diff --git a/net/batman-adv/fragmentation.h b/net/batman-adv/fragmentation.h index 9ff77c7..3202fe3 100644 --- a/net/batman-adv/fragmentation.h +++ b/net/batman-adv/fragmentation.h @@ -34,9 +34,9 @@ bool batadv_frag_skb_fwd(struct sk_buff *skb, struct batadv_orig_node *orig_node_src); bool batadv_frag_skb_buffer(struct sk_buff **skb, struct batadv_orig_node *orig_node); -bool batadv_frag_send_packet(struct sk_buff *skb, - struct batadv_orig_node *orig_node, - struct batadv_neigh_node *neigh_node); +int batadv_frag_send_packet(struct sk_buff *skb, + struct batadv_orig_node *orig_node, + struct batadv_neigh_node *neigh_node); /** * batadv_frag_check_entry - check if a list of fragments has timed out diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index 5833ab3..76de583 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -270,8 +270,10 @@ static int batadv_recv_my_icmp_packet(struct batadv_priv *bat_priv, icmph->ttl = BATADV_TTL; res = batadv_send_skb_to_orig(skb, orig_node, NULL); - if (res != NET_XMIT_DROP) - ret = NET_RX_SUCCESS; + if (res == -1) + goto out; + + ret = NET_RX_SUCCESS; break; default: @@ -292,7 +294,7 @@ static int batadv_recv_icmp_ttl_exceeded(struct batadv_priv *bat_priv, struct batadv_hard_iface *primary_if = NULL; struct batadv_orig_node *orig_node = NULL; struct batadv_icmp_packet *icmp_packet; - int ret = NET_RX_DROP; + int res, ret = NET_RX_DROP; icmp_packet = (struct batadv_icmp_packet *)skb->data; @@ -323,7 +325,8 @@ static int batadv_recv_icmp_ttl_exceeded(struct batadv_priv *bat_priv, icmp_packet->msg_type = BATADV_TTL_EXCEEDED; icmp_packet->ttl = BATADV_TTL; - if (batadv_send_skb_to_orig(skb, orig_node, NULL) != NET_XMIT_DROP) + res = batadv_send_skb_to_orig(skb, orig_node, NULL); + if (res != -1) ret = NET_RX_SUCCESS; out: @@ -343,7 +346,7 @@ int batadv_recv_icmp_packet(struct sk_buff *skb, struct ethhdr *ethhdr; struct batadv_orig_node *orig_node = NULL; int hdr_size = sizeof(struct batadv_icmp_header); - int ret = NET_RX_DROP; + int res, ret = NET_RX_DROP; /* drop packet if it has not necessary minimum size */ if (unlikely(!pskb_may_pull(skb, hdr_size))) @@ -409,7 +412,8 @@ int batadv_recv_icmp_packet(struct sk_buff *skb, icmph->ttl--; /* route it */ - if (batadv_send_skb_to_orig(skb, orig_node, recv_if) != NET_XMIT_DROP) + res = batadv_send_skb_to_orig(skb, orig_node, recv_if); + if (res != -1) ret = NET_RX_SUCCESS; out: @@ -646,6 +650,8 @@ static int batadv_route_unicast_packet(struct sk_buff *skb, len = skb->len; res = batadv_send_skb_to_orig(skb, orig_node, recv_if); + if (res == -1) + goto out; /* translate transmit result into receive result */ if (res == NET_XMIT_SUCCESS) { @@ -653,13 +659,10 @@ static int batadv_route_unicast_packet(struct sk_buff *skb, batadv_inc_counter(bat_priv, BATADV_CNT_FORWARD); batadv_add_counter(bat_priv, BATADV_CNT_FORWARD_BYTES, len + ETH_HLEN); - - ret = NET_RX_SUCCESS; - } else if (res == -EINPROGRESS) { - /* skb was buffered and consumed */ - ret = NET_RX_SUCCESS; } + ret = NET_RX_SUCCESS; + out: if (orig_node) batadv_orig_node_put(orig_node); diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c index 3a59df2..3a10d87 100644 --- a/net/batman-adv/send.c +++ b/net/batman-adv/send.c @@ -20,6 +20,7 @@ #include #include +#include #include #include #include @@ -72,6 +73,7 @@ int batadv_send_skb_packet(struct sk_buff *skb, { struct batadv_priv *bat_priv; struct ethhdr *ethhdr; + int ret; bat_priv = netdev_priv(hard_iface->soft_iface); @@ -109,8 +111,15 @@ int batadv_send_skb_packet(struct sk_buff *skb, /* dev_queue_xmit() returns a negative result on error. However on * congestion and traffic shaping, it drops and returns NET_XMIT_DROP * (which is > 0). This will not be treated as an error. + * + * a negative value cannot be returned because it could be interepreted + * as not consumed skb by callers of batadv_send_skb_to_orig. */ - return dev_queue_xmit(skb); + ret = dev_queue_xmit(skb); + if (ret < 0) + ret = NET_XMIT_DROP; + + return ret; send_skb_err: kfree_skb(skb); return NET_XMIT_DROP; @@ -156,8 +165,11 @@ int batadv_send_unicast_skb(struct sk_buff *skb, * host, NULL can be passed as recv_if and no interface alternating is * attempted. * - * Return: NET_XMIT_SUCCESS on success, NET_XMIT_DROP on failure, or - * -EINPROGRESS if the skb is buffered for later transmit. + * Return: -1 on failure (and the skb is not consumed), -EINPROGRESS if the + * skb is buffered for later transmit or the NET_XMIT status returned by the + * lower routine if the packet has been passed down. + * + * If the returning value is not -1 the skb has been consumed. */ int batadv_send_skb_to_orig(struct sk_buff *skb, struct batadv_orig_node *orig_node, @@ -165,7 +177,7 @@ int batadv_send_skb_to_orig(struct sk_buff *skb, { struct batadv_priv *bat_priv = orig_node->bat_priv; struct batadv_neigh_node *neigh_node; - int ret = NET_XMIT_DROP; + int ret = -1; /* batadv_find_router() increases neigh_nodes refcount if found. */ neigh_node = batadv_find_router(bat_priv, orig_node, recv_if); @@ -178,8 +190,7 @@ int batadv_send_skb_to_orig(struct sk_buff *skb, if (atomic_read(&bat_priv->fragmentation) && skb->len > neigh_node->if_incoming->net_dev->mtu) { /* Fragment and send packet. */ - if (batadv_frag_send_packet(skb, orig_node, neigh_node)) - ret = NET_XMIT_SUCCESS; + ret = batadv_frag_send_packet(skb, orig_node, neigh_node); goto out; } @@ -188,12 +199,10 @@ int batadv_send_skb_to_orig(struct sk_buff *skb, * (i.e. being forwarded). If the packet originates from this node or if * network coding fails, then send the packet as usual. */ - if (recv_if && batadv_nc_skb_forward(skb, neigh_node)) { + if (recv_if && batadv_nc_skb_forward(skb, neigh_node)) ret = -EINPROGRESS; - } else { - batadv_send_unicast_skb(skb, neigh_node); - ret = NET_XMIT_SUCCESS; - } + else + ret = batadv_send_unicast_skb(skb, neigh_node); out: if (neigh_node) @@ -319,7 +328,7 @@ int batadv_send_skb_unicast(struct batadv_priv *bat_priv, { struct batadv_unicast_packet *unicast_packet; struct ethhdr *ethhdr; - int ret = NET_XMIT_DROP; + int res, ret = NET_XMIT_DROP; if (!orig_node) goto out; @@ -356,7 +365,8 @@ int batadv_send_skb_unicast(struct batadv_priv *bat_priv, if (batadv_tt_global_client_is_roaming(bat_priv, ethhdr->h_dest, vid)) unicast_packet->ttvn = unicast_packet->ttvn - 1; - if (batadv_send_skb_to_orig(skb, orig_node, NULL) != NET_XMIT_DROP) + res = batadv_send_skb_to_orig(skb, orig_node, NULL); + if (res != -1) ret = NET_XMIT_SUCCESS; out: diff --git a/net/batman-adv/tvlv.c b/net/batman-adv/tvlv.c index 2fd542e..3d1cf0f 100644 --- a/net/batman-adv/tvlv.c +++ b/net/batman-adv/tvlv.c @@ -591,6 +591,7 @@ void batadv_tvlv_unicast_send(struct batadv_priv *bat_priv, u8 *src, unsigned char *tvlv_buff; unsigned int tvlv_len; ssize_t hdr_len = sizeof(*unicast_tvlv_packet); + int res; orig_node = batadv_orig_hash_find(bat_priv, dst); if (!orig_node) @@ -623,7 +624,8 @@ void batadv_tvlv_unicast_send(struct batadv_priv *bat_priv, u8 *src, tvlv_buff += sizeof(*tvlv_hdr); memcpy(tvlv_buff, tvlv_value, tvlv_value_len); - if (batadv_send_skb_to_orig(skb, orig_node, NULL) == NET_XMIT_DROP) + res = batadv_send_skb_to_orig(skb, orig_node, NULL); + if (res == -1) kfree_skb(skb); out: batadv_orig_node_put(orig_node); -- cgit v0.10.2 From 33a3bb4a3345bb511f9c69c913da95d4693e2a4e Mon Sep 17 00:00:00 2001 From: Antonio Quartulli Date: Thu, 5 May 2016 13:09:43 +0200 Subject: batman-adv: throughput meter implementation The throughput meter module is a simple, kernel-space replacement for throughtput measurements tool like iperf and netperf. It is intended to approximate TCP behaviour. It is invoked through batctl: the protocol is connection oriented, with cumulative acknowledgment and a dynamic-size sliding window. The test *can* be interrupted by batctl. A receiver side timeout avoids unlimited waitings for sender packets: after one second of inactivity, the receiver abort the ongoing test. Based on a prototype from Edo Monticelli Signed-off-by: Antonio Quartulli Signed-off-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/include/uapi/linux/batman_adv.h b/include/uapi/linux/batman_adv.h index c39623c7..0fbf6fd 100644 --- a/include/uapi/linux/batman_adv.h +++ b/include/uapi/linux/batman_adv.h @@ -20,6 +20,8 @@ #define BATADV_NL_NAME "batadv" +#define BATADV_NL_MCAST_GROUP_TPMETER "tpmeter" + /** * enum batadv_nl_attrs - batman-adv netlink attributes * @@ -32,6 +34,12 @@ * @BATADV_ATTR_HARD_IFINDEX: index of the non-batman-adv interface * @BATADV_ATTR_HARD_IFNAME: name of the non-batman-adv interface * @BATADV_ATTR_HARD_ADDRESS: mac address of the non-batman-adv interface + * @BATADV_ATTR_ORIG_ADDRESS: originator mac address + * @BATADV_ATTR_TPMETER_RESULT: result of run (see batadv_tp_meter_status) + * @BATADV_ATTR_TPMETER_TEST_TIME: time (msec) the run took + * @BATADV_ATTR_TPMETER_BYTES: amount of acked bytes during run + * @BATADV_ATTR_TPMETER_COOKIE: session cookie to match tp_meter session + * @BATADV_ATTR_PAD: attribute used for padding for 64-bit alignment * @__BATADV_ATTR_AFTER_LAST: internal use * @NUM_BATADV_ATTR: total number of batadv_nl_attrs available * @BATADV_ATTR_MAX: highest attribute number currently defined @@ -46,6 +54,12 @@ enum batadv_nl_attrs { BATADV_ATTR_HARD_IFINDEX, BATADV_ATTR_HARD_IFNAME, BATADV_ATTR_HARD_ADDRESS, + BATADV_ATTR_ORIG_ADDRESS, + BATADV_ATTR_TPMETER_RESULT, + BATADV_ATTR_TPMETER_TEST_TIME, + BATADV_ATTR_TPMETER_BYTES, + BATADV_ATTR_TPMETER_COOKIE, + BATADV_ATTR_PAD, /* add attributes above here, update the policy in netlink.c */ __BATADV_ATTR_AFTER_LAST, NUM_BATADV_ATTR = __BATADV_ATTR_AFTER_LAST, @@ -57,15 +71,44 @@ enum batadv_nl_attrs { * * @BATADV_CMD_UNSPEC: unspecified command to catch errors * @BATADV_CMD_GET_MESH_INFO: Query basic information about batman-adv device + * @BATADV_CMD_TP_METER: Start a tp meter session + * @BATADV_CMD_TP_METER_CANCEL: Cancel a tp meter session * @__BATADV_CMD_AFTER_LAST: internal use * @BATADV_CMD_MAX: highest used command number */ enum batadv_nl_commands { BATADV_CMD_UNSPEC, BATADV_CMD_GET_MESH_INFO, + BATADV_CMD_TP_METER, + BATADV_CMD_TP_METER_CANCEL, /* add new commands above here */ __BATADV_CMD_AFTER_LAST, BATADV_CMD_MAX = __BATADV_CMD_AFTER_LAST - 1 }; +/** + * enum batadv_tp_meter_reason - reason of a tp meter test run stop + * @BATADV_TP_REASON_COMPLETE: sender finished tp run + * @BATADV_TP_REASON_CANCEL: sender was stopped during run + * @BATADV_TP_REASON_DST_UNREACHABLE: receiver could not be reached or didn't + * answer + * @BATADV_TP_REASON_RESEND_LIMIT: (unused) sender retry reached limit + * @BATADV_TP_REASON_ALREADY_ONGOING: test to or from the same node already + * ongoing + * @BATADV_TP_REASON_MEMORY_ERROR: test was stopped due to low memory + * @BATADV_TP_REASON_CANT_SEND: failed to send via outgoing interface + * @BATADV_TP_REASON_TOO_MANY: too many ongoing sessions + */ +enum batadv_tp_meter_reason { + BATADV_TP_REASON_COMPLETE = 3, + BATADV_TP_REASON_CANCEL = 4, + /* error status >= 128 */ + BATADV_TP_REASON_DST_UNREACHABLE = 128, + BATADV_TP_REASON_RESEND_LIMIT = 129, + BATADV_TP_REASON_ALREADY_ONGOING = 130, + BATADV_TP_REASON_MEMORY_ERROR = 131, + BATADV_TP_REASON_CANT_SEND = 132, + BATADV_TP_REASON_TOO_MANY = 133, +}; + #endif /* _UAPI_LINUX_BATMAN_ADV_H_ */ diff --git a/net/batman-adv/Makefile b/net/batman-adv/Makefile index 7da5901..a83fc6c 100644 --- a/net/batman-adv/Makefile +++ b/net/batman-adv/Makefile @@ -42,5 +42,6 @@ batman-adv-y += routing.o batman-adv-y += send.o batman-adv-y += soft-interface.o batman-adv-y += sysfs.o +batman-adv-y += tp_meter.o batman-adv-y += translation-table.o batman-adv-y += tvlv.o diff --git a/net/batman-adv/log.h b/net/batman-adv/log.h index 9948e56..e0e1a88 100644 --- a/net/batman-adv/log.h +++ b/net/batman-adv/log.h @@ -51,17 +51,19 @@ static inline void batadv_debug_log_cleanup(struct batadv_priv *bat_priv) * @BATADV_DBG_DAT: ARP snooping and DAT related messages * @BATADV_DBG_NC: network coding related messages * @BATADV_DBG_MCAST: multicast related messages + * @BATADV_DBG_TP_METER: throughput meter messages * @BATADV_DBG_ALL: the union of all the above log levels */ enum batadv_dbg_level { - BATADV_DBG_BATMAN = BIT(0), - BATADV_DBG_ROUTES = BIT(1), - BATADV_DBG_TT = BIT(2), - BATADV_DBG_BLA = BIT(3), - BATADV_DBG_DAT = BIT(4), - BATADV_DBG_NC = BIT(5), - BATADV_DBG_MCAST = BIT(6), - BATADV_DBG_ALL = 127, + BATADV_DBG_BATMAN = BIT(0), + BATADV_DBG_ROUTES = BIT(1), + BATADV_DBG_TT = BIT(2), + BATADV_DBG_BLA = BIT(3), + BATADV_DBG_DAT = BIT(4), + BATADV_DBG_NC = BIT(5), + BATADV_DBG_MCAST = BIT(6), + BATADV_DBG_TP_METER = BIT(7), + BATADV_DBG_ALL = 127, }; #ifdef CONFIG_BATMAN_ADV_DEBUG diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index 275604b..fe4c5e2 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -64,6 +64,7 @@ #include "routing.h" #include "send.h" #include "soft-interface.h" +#include "tp_meter.h" #include "translation-table.h" /* List manipulations on hardif_list have to be rtnl_lock()'ed, @@ -89,6 +90,7 @@ static int __init batadv_init(void) batadv_v_init(); batadv_iv_init(); batadv_nc_init(); + batadv_tp_meter_init(); batadv_event_workqueue = create_singlethread_workqueue("bat_events"); @@ -142,6 +144,7 @@ int batadv_mesh_init(struct net_device *soft_iface) spin_lock_init(&bat_priv->tvlv.container_list_lock); spin_lock_init(&bat_priv->tvlv.handler_list_lock); spin_lock_init(&bat_priv->softif_vlan_list_lock); + spin_lock_init(&bat_priv->tp_list_lock); INIT_HLIST_HEAD(&bat_priv->forw_bat_list); INIT_HLIST_HEAD(&bat_priv->forw_bcast_list); @@ -160,6 +163,7 @@ int batadv_mesh_init(struct net_device *soft_iface) INIT_HLIST_HEAD(&bat_priv->tvlv.container_list); INIT_HLIST_HEAD(&bat_priv->tvlv.handler_list); INIT_HLIST_HEAD(&bat_priv->softif_vlan_list); + INIT_HLIST_HEAD(&bat_priv->tp_list); ret = batadv_v_mesh_init(bat_priv); if (ret < 0) diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 857fb5a..06a8608 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -100,6 +100,9 @@ #define BATADV_NUM_BCASTS_WIRELESS 3 #define BATADV_NUM_BCASTS_MAX 3 +/* length of the single packet used by the TP meter */ +#define BATADV_TP_PACKET_LEN ETH_DATA_LEN + /* msecs after which an ARP_REQUEST is sent in broadcast as fallback */ #define ARP_REQ_DELAY 250 /* numbers of originator to contact for any PUT/GET DHT operation */ @@ -131,6 +134,11 @@ #define BATADV_NC_NODE_TIMEOUT 10000 /* Milliseconds */ +/** + * BATADV_TP_MAX_NUM - maximum number of simultaneously active tp sessions + */ +#define BATADV_TP_MAX_NUM 5 + enum batadv_mesh_state { BATADV_MESH_INACTIVE, BATADV_MESH_ACTIVE, diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c index 68152aa..c25bbb8 100644 --- a/net/batman-adv/netlink.c +++ b/net/batman-adv/netlink.c @@ -27,12 +27,14 @@ #include #include #include +#include #include #include #include #include "hard-interface.h" #include "soft-interface.h" +#include "tp_meter.h" struct sk_buff; @@ -44,6 +46,15 @@ static struct genl_family batadv_netlink_family = { .maxattr = BATADV_ATTR_MAX, }; +/* multicast groups */ +enum batadv_netlink_multicast_groups { + BATADV_NL_MCGRP_TPMETER, +}; + +static struct genl_multicast_group batadv_netlink_mcgrps[] = { + [BATADV_NL_MCGRP_TPMETER] = { .name = BATADV_NL_MCAST_GROUP_TPMETER }, +}; + static struct nla_policy batadv_netlink_policy[NUM_BATADV_ATTR] = { [BATADV_ATTR_VERSION] = { .type = NLA_STRING }, [BATADV_ATTR_ALGO_NAME] = { .type = NLA_STRING }, @@ -53,6 +64,11 @@ static struct nla_policy batadv_netlink_policy[NUM_BATADV_ATTR] = { [BATADV_ATTR_HARD_IFINDEX] = { .type = NLA_U32 }, [BATADV_ATTR_HARD_IFNAME] = { .type = NLA_STRING }, [BATADV_ATTR_HARD_ADDRESS] = { .len = ETH_ALEN }, + [BATADV_ATTR_ORIG_ADDRESS] = { .len = ETH_ALEN }, + [BATADV_ATTR_TPMETER_RESULT] = { .type = NLA_U8 }, + [BATADV_ATTR_TPMETER_TEST_TIME] = { .type = NLA_U32 }, + [BATADV_ATTR_TPMETER_BYTES] = { .type = NLA_U64 }, + [BATADV_ATTR_TPMETER_COOKIE] = { .type = NLA_U32 }, }; /** @@ -163,6 +179,207 @@ batadv_netlink_get_mesh_info(struct sk_buff *skb, struct genl_info *info) return genlmsg_reply(msg, info); } +/** + * batadv_netlink_tp_meter_put - Fill information of started tp_meter session + * @msg: netlink message to be sent back + * @cookie: tp meter session cookie + * + * Return: 0 on success, < 0 on error + */ +static int +batadv_netlink_tp_meter_put(struct sk_buff *msg, u32 cookie) +{ + if (nla_put_u32(msg, BATADV_ATTR_TPMETER_COOKIE, cookie)) + return -ENOBUFS; + + return 0; +} + +/** + * batadv_netlink_tpmeter_notify - send tp_meter result via netlink to client + * @bat_priv: the bat priv with all the soft interface information + * @dst: destination of tp_meter session + * @result: reason for tp meter session stop + * @test_time: total time ot the tp_meter session + * @total_bytes: bytes acked to the receiver + * @cookie: cookie of tp_meter session + * + * Return: 0 on success, < 0 on error + */ +int batadv_netlink_tpmeter_notify(struct batadv_priv *bat_priv, const u8 *dst, + u8 result, u32 test_time, u64 total_bytes, + u32 cookie) +{ + struct sk_buff *msg; + void *hdr; + int ret; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + hdr = genlmsg_put(msg, 0, 0, &batadv_netlink_family, 0, + BATADV_CMD_TP_METER); + if (!hdr) { + ret = -ENOBUFS; + goto err_genlmsg; + } + + if (nla_put_u32(msg, BATADV_ATTR_TPMETER_COOKIE, cookie)) + goto nla_put_failure; + + if (nla_put_u32(msg, BATADV_ATTR_TPMETER_TEST_TIME, test_time)) + goto nla_put_failure; + + if (nla_put_u64_64bit(msg, BATADV_ATTR_TPMETER_BYTES, total_bytes, + BATADV_ATTR_PAD)) + goto nla_put_failure; + + if (nla_put_u8(msg, BATADV_ATTR_TPMETER_RESULT, result)) + goto nla_put_failure; + + if (nla_put(msg, BATADV_ATTR_ORIG_ADDRESS, ETH_ALEN, dst)) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + + genlmsg_multicast_netns(&batadv_netlink_family, + dev_net(bat_priv->soft_iface), msg, 0, + BATADV_NL_MCGRP_TPMETER, GFP_KERNEL); + + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); + ret = -EMSGSIZE; + +err_genlmsg: + nlmsg_free(msg); + return ret; +} + +/** + * batadv_netlink_tp_meter_start - Start a new tp_meter session + * @skb: received netlink message + * @info: receiver information + * + * Return: 0 on success, < 0 on error + */ +static int +batadv_netlink_tp_meter_start(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct net_device *soft_iface; + struct batadv_priv *bat_priv; + struct sk_buff *msg = NULL; + u32 test_length; + void *msg_head; + int ifindex; + u32 cookie; + u8 *dst; + int ret; + + if (!info->attrs[BATADV_ATTR_MESH_IFINDEX]) + return -EINVAL; + + if (!info->attrs[BATADV_ATTR_ORIG_ADDRESS]) + return -EINVAL; + + if (!info->attrs[BATADV_ATTR_TPMETER_TEST_TIME]) + return -EINVAL; + + ifindex = nla_get_u32(info->attrs[BATADV_ATTR_MESH_IFINDEX]); + if (!ifindex) + return -EINVAL; + + dst = nla_data(info->attrs[BATADV_ATTR_ORIG_ADDRESS]); + + test_length = nla_get_u32(info->attrs[BATADV_ATTR_TPMETER_TEST_TIME]); + + soft_iface = dev_get_by_index(net, ifindex); + if (!soft_iface || !batadv_softif_is_valid(soft_iface)) { + ret = -ENODEV; + goto out; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) { + ret = -ENOMEM; + goto out; + } + + msg_head = genlmsg_put(msg, info->snd_portid, info->snd_seq, + &batadv_netlink_family, 0, + BATADV_CMD_TP_METER); + if (!msg_head) { + ret = -ENOBUFS; + goto out; + } + + bat_priv = netdev_priv(soft_iface); + batadv_tp_start(bat_priv, dst, test_length, &cookie); + + ret = batadv_netlink_tp_meter_put(msg, cookie); + + out: + if (soft_iface) + dev_put(soft_iface); + + if (ret) { + if (msg) + nlmsg_free(msg); + return ret; + } + + genlmsg_end(msg, msg_head); + return genlmsg_reply(msg, info); +} + +/** + * batadv_netlink_tp_meter_start - Cancel a running tp_meter session + * @skb: received netlink message + * @info: receiver information + * + * Return: 0 on success, < 0 on error + */ +static int +batadv_netlink_tp_meter_cancel(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = genl_info_net(info); + struct net_device *soft_iface; + struct batadv_priv *bat_priv; + int ifindex; + u8 *dst; + int ret = 0; + + if (!info->attrs[BATADV_ATTR_MESH_IFINDEX]) + return -EINVAL; + + if (!info->attrs[BATADV_ATTR_ORIG_ADDRESS]) + return -EINVAL; + + ifindex = nla_get_u32(info->attrs[BATADV_ATTR_MESH_IFINDEX]); + if (!ifindex) + return -EINVAL; + + dst = nla_data(info->attrs[BATADV_ATTR_ORIG_ADDRESS]); + + soft_iface = dev_get_by_index(net, ifindex); + if (!soft_iface || !batadv_softif_is_valid(soft_iface)) { + ret = -ENODEV; + goto out; + } + + bat_priv = netdev_priv(soft_iface); + batadv_tp_stop(bat_priv, dst, BATADV_TP_REASON_CANCEL); + +out: + if (soft_iface) + dev_put(soft_iface); + + return ret; +} + static struct genl_ops batadv_netlink_ops[] = { { .cmd = BATADV_CMD_GET_MESH_INFO, @@ -170,6 +387,18 @@ static struct genl_ops batadv_netlink_ops[] = { .policy = batadv_netlink_policy, .doit = batadv_netlink_get_mesh_info, }, + { + .cmd = BATADV_CMD_TP_METER, + .flags = GENL_ADMIN_PERM, + .policy = batadv_netlink_policy, + .doit = batadv_netlink_tp_meter_start, + }, + { + .cmd = BATADV_CMD_TP_METER_CANCEL, + .flags = GENL_ADMIN_PERM, + .policy = batadv_netlink_policy, + .doit = batadv_netlink_tp_meter_cancel, + }, }; /** @@ -179,8 +408,9 @@ void __init batadv_netlink_register(void) { int ret; - ret = genl_register_family_with_ops(&batadv_netlink_family, - batadv_netlink_ops); + ret = genl_register_family_with_ops_groups(&batadv_netlink_family, + batadv_netlink_ops, + batadv_netlink_mcgrps); if (ret) pr_warn("unable to register netlink family"); } diff --git a/net/batman-adv/netlink.h b/net/batman-adv/netlink.h index 39044cc..945653a 100644 --- a/net/batman-adv/netlink.h +++ b/net/batman-adv/netlink.h @@ -20,7 +20,13 @@ #include "main.h" +#include + void batadv_netlink_register(void); void batadv_netlink_unregister(void); +int batadv_netlink_tpmeter_notify(struct batadv_priv *bat_priv, const u8 *dst, + u8 result, u32 test_time, u64 total_bytes, + u32 cookie); + #endif /* _NET_BATMAN_ADV_NETLINK_H_ */ diff --git a/net/batman-adv/packet.h b/net/batman-adv/packet.h index 7156779..6b011ff 100644 --- a/net/batman-adv/packet.h +++ b/net/batman-adv/packet.h @@ -21,6 +21,8 @@ #include #include +#define batadv_tp_is_error(n) ((u8)n > 127 ? 1 : 0) + /** * enum batadv_packettype - types for batman-adv encapsulated packets * @BATADV_IV_OGM: originator messages for B.A.T.M.A.N. IV @@ -93,6 +95,7 @@ enum batadv_icmp_packettype { BATADV_ECHO_REQUEST = 8, BATADV_TTL_EXCEEDED = 11, BATADV_PARAMETER_PROBLEM = 12, + BATADV_TP = 15, }; /** @@ -285,6 +288,16 @@ struct batadv_elp_packet { #define BATADV_ELP_HLEN sizeof(struct batadv_elp_packet) /** + * enum batadv_icmp_user_cmd_type - types for batman-adv icmp cmd modes + * @BATADV_TP_START: start a throughput meter run + * @BATADV_TP_STOP: stop a throughput meter run + */ +enum batadv_icmp_user_cmd_type { + BATADV_TP_START = 0, + BATADV_TP_STOP = 2, +}; + +/** * struct batadv_icmp_header - common members among all the ICMP packets * @packet_type: batman-adv packet type, part of the general header * @version: batman-adv protocol version, part of the genereal header @@ -334,6 +347,47 @@ struct batadv_icmp_packet { __be16 seqno; }; +/** + * struct batadv_icmp_tp_packet - ICMP TP Meter packet + * @packet_type: batman-adv packet type, part of the general header + * @version: batman-adv protocol version, part of the genereal header + * @ttl: time to live for this packet, part of the genereal header + * @msg_type: ICMP packet type + * @dst: address of the destination node + * @orig: address of the source node + * @uid: local ICMP socket identifier + * @subtype: TP packet subtype (see batadv_icmp_tp_subtype) + * @session: TP session identifier + * @seqno: the TP sequence number + * @timestamp: time when the packet has been sent. This value is filled in a + * TP_MSG and echoed back in the next TP_ACK so that the sender can compute the + * RTT. Since it is read only by the host which wrote it, there is no need to + * store it using network order + */ +struct batadv_icmp_tp_packet { + u8 packet_type; + u8 version; + u8 ttl; + u8 msg_type; /* see ICMP message types above */ + u8 dst[ETH_ALEN]; + u8 orig[ETH_ALEN]; + u8 uid; + u8 subtype; + u8 session[2]; + __be32 seqno; + __be32 timestamp; +}; + +/** + * enum batadv_icmp_tp_subtype - ICMP TP Meter packet subtypes + * @BATADV_TP_MSG: Msg from sender to receiver + * @BATADV_TP_ACK: acknowledgment from receiver to sender + */ +enum batadv_icmp_tp_subtype { + BATADV_TP_MSG = 0, + BATADV_TP_ACK, +}; + #define BATADV_RR_LEN 16 /** diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index 76de583..7b5de40 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -46,6 +46,7 @@ #include "packet.h" #include "send.h" #include "soft-interface.h" +#include "tp_meter.h" #include "translation-table.h" #include "tvlv.h" @@ -276,6 +277,13 @@ static int batadv_recv_my_icmp_packet(struct batadv_priv *bat_priv, ret = NET_RX_SUCCESS; break; + case BATADV_TP: + if (!pskb_may_pull(skb, sizeof(struct batadv_icmp_tp_packet))) + goto out; + + batadv_tp_meter_recv(bat_priv, skb); + ret = NET_RX_SUCCESS; + goto out; default: /* drop unknown type */ goto out; diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index f75631e..18b6d07 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -842,6 +842,8 @@ static int batadv_softif_init_late(struct net_device *dev) #ifdef CONFIG_BATMAN_ADV_BLA atomic_set(&bat_priv->bla.num_requests, 0); #endif + atomic_set(&bat_priv->tp_num, 0); + bat_priv->tt.last_changeset = NULL; bat_priv->tt.last_changeset_len = 0; bat_priv->isolation_mark = 0; diff --git a/net/batman-adv/tp_meter.c b/net/batman-adv/tp_meter.c new file mode 100644 index 0000000..2333777 --- /dev/null +++ b/net/batman-adv/tp_meter.c @@ -0,0 +1,1507 @@ +/* Copyright (C) 2012-2016 B.A.T.M.A.N. contributors: + * + * Edo Monticelli, Antonio Quartulli + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#include "tp_meter.h" +#include "main.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "hard-interface.h" +#include "log.h" +#include "netlink.h" +#include "originator.h" +#include "packet.h" +#include "send.h" + +/** + * BATADV_TP_DEF_TEST_LENGTH - Default test length if not specified by the user + * in milliseconds + */ +#define BATADV_TP_DEF_TEST_LENGTH 10000 + +/** + * BATADV_TP_AWND - Advertised window by the receiver (in bytes) + */ +#define BATADV_TP_AWND 0x20000000 + +/** + * BATADV_TP_RECV_TIMEOUT - Receiver activity timeout. If the receiver does not + * get anything for such amount of milliseconds, the connection is killed + */ +#define BATADV_TP_RECV_TIMEOUT 1000 + +/** + * BATADV_TP_MAX_RTO - Maximum sender timeout. If the sender RTO gets beyond + * such amound of milliseconds, the receiver is considered unreachable and the + * connection is killed + */ +#define BATADV_TP_MAX_RTO 30000 + +/** + * BATADV_TP_FIRST_SEQ - First seqno of each session. The number is rather high + * in order to immediately trigger a wrap around (test purposes) + */ +#define BATADV_TP_FIRST_SEQ ((u32)-1 - 2000) + +/** + * BATADV_TP_PLEN - length of the payload (data after the batadv_unicast header) + * to simulate + */ +#define BATADV_TP_PLEN (BATADV_TP_PACKET_LEN - ETH_HLEN - \ + sizeof(struct batadv_unicast_packet)) + +static u8 batadv_tp_prerandom[4096] __read_mostly; + +/** + * batadv_tp_session_cookie - generate session cookie based on session ids + * @session: TP session identifier + * @icmp_uid: icmp pseudo uid of the tp session + * + * Return: 32 bit tp_meter session cookie + */ +static u32 batadv_tp_session_cookie(const u8 session[2], u8 icmp_uid) +{ + u32 cookie; + + cookie = icmp_uid << 16; + cookie |= session[0] << 8; + cookie |= session[1]; + + return cookie; +} + +/** + * batadv_tp_cwnd - compute the new cwnd size + * @base: base cwnd size value + * @increment: the value to add to base to get the new size + * @min: minumim cwnd value (usually MSS) + * + * Return the new cwnd size and ensures it does not exceed the Advertised + * Receiver Window size. It is wrap around safe. + * For details refer to Section 3.1 of RFC5681 + * + * Return: new congestion window size in bytes + */ +static u32 batadv_tp_cwnd(u32 base, u32 increment, u32 min) +{ + u32 new_size = base + increment; + + /* check for wrap-around */ + if (new_size < base) + new_size = (u32)ULONG_MAX; + + new_size = min_t(u32, new_size, BATADV_TP_AWND); + + return max_t(u32, new_size, min); +} + +/** + * batadv_tp_updated_cwnd - update the Congestion Windows + * @tp_vars: the private data of the current TP meter session + * @mss: maximum segment size of transmission + * + * 1) if the session is in Slow Start, the CWND has to be increased by 1 + * MSS every unique received ACK + * 2) if the session is in Congestion Avoidance, the CWND has to be + * increased by MSS * MSS / CWND for every unique received ACK + */ +static void batadv_tp_update_cwnd(struct batadv_tp_vars *tp_vars, u32 mss) +{ + spin_lock_bh(&tp_vars->cwnd_lock); + + /* slow start... */ + if (tp_vars->cwnd <= tp_vars->ss_threshold) { + tp_vars->dec_cwnd = 0; + tp_vars->cwnd = batadv_tp_cwnd(tp_vars->cwnd, mss, mss); + spin_unlock_bh(&tp_vars->cwnd_lock); + return; + } + + /* increment CWND at least of 1 (section 3.1 of RFC5681) */ + tp_vars->dec_cwnd += max_t(u32, 1U << 3, + ((mss * mss) << 6) / (tp_vars->cwnd << 3)); + if (tp_vars->dec_cwnd < (mss << 3)) { + spin_unlock_bh(&tp_vars->cwnd_lock); + return; + } + + tp_vars->cwnd = batadv_tp_cwnd(tp_vars->cwnd, mss, mss); + tp_vars->dec_cwnd = 0; + + spin_unlock_bh(&tp_vars->cwnd_lock); +} + +/** + * batadv_tp_update_rto - calculate new retransmission timeout + * @tp_vars: the private data of the current TP meter session + * @new_rtt: new roundtrip time in msec + */ +static void batadv_tp_update_rto(struct batadv_tp_vars *tp_vars, + u32 new_rtt) +{ + long m = new_rtt; + + /* RTT update + * Details in Section 2.2 and 2.3 of RFC6298 + * + * It's tricky to understand. Don't lose hair please. + * Inspired by tcp_rtt_estimator() tcp_input.c + */ + if (tp_vars->srtt != 0) { + m -= (tp_vars->srtt >> 3); /* m is now error in rtt est */ + tp_vars->srtt += m; /* rtt = 7/8 srtt + 1/8 new */ + if (m < 0) + m = -m; + + m -= (tp_vars->rttvar >> 2); + tp_vars->rttvar += m; /* mdev ~= 3/4 rttvar + 1/4 new */ + } else { + /* first measure getting in */ + tp_vars->srtt = m << 3; /* take the measured time to be srtt */ + tp_vars->rttvar = m << 1; /* new_rtt / 2 */ + } + + /* rto = srtt + 4 * rttvar. + * rttvar is scaled by 4, therefore doesn't need to be multiplied + */ + tp_vars->rto = (tp_vars->srtt >> 3) + tp_vars->rttvar; +} + +/** + * batadv_tp_batctl_notify - send client status result to client + * @reason: reason for tp meter session stop + * @dst: destination of tp_meter session + * @bat_priv: the bat priv with all the soft interface information + * @start_time: start of transmission in jiffies + * @total_sent: bytes acked to the receiver + * @cookie: cookie of tp_meter session + */ +static void batadv_tp_batctl_notify(enum batadv_tp_meter_reason reason, + const u8 *dst, struct batadv_priv *bat_priv, + unsigned long start_time, u64 total_sent, + u32 cookie) +{ + u32 test_time; + u8 result; + u32 total_bytes; + + if (!batadv_tp_is_error(reason)) { + result = BATADV_TP_REASON_COMPLETE; + test_time = jiffies_to_msecs(jiffies - start_time); + total_bytes = total_sent; + } else { + result = reason; + test_time = 0; + total_bytes = 0; + } + + batadv_netlink_tpmeter_notify(bat_priv, dst, result, test_time, + total_bytes, cookie); +} + +/** + * batadv_tp_batctl_error_notify - send client error result to client + * @reason: reason for tp meter session stop + * @dst: destination of tp_meter session + * @bat_priv: the bat priv with all the soft interface information + * @cookie: cookie of tp_meter session + */ +static void batadv_tp_batctl_error_notify(enum batadv_tp_meter_reason reason, + const u8 *dst, + struct batadv_priv *bat_priv, + u32 cookie) +{ + batadv_tp_batctl_notify(reason, dst, bat_priv, 0, 0, cookie); +} + +/** + * batadv_tp_list_find - find a tp_vars object in the global list + * @bat_priv: the bat priv with all the soft interface information + * @dst: the other endpoint MAC address to look for + * + * Look for a tp_vars object matching dst as end_point and return it after + * having incremented the refcounter. Return NULL is not found + * + * Return: matching tp_vars or NULL when no tp_vars with @dst was found + */ +static struct batadv_tp_vars *batadv_tp_list_find(struct batadv_priv *bat_priv, + const u8 *dst) +{ + struct batadv_tp_vars *pos, *tp_vars = NULL; + + rcu_read_lock(); + hlist_for_each_entry_rcu(pos, &bat_priv->tp_list, list) { + if (!batadv_compare_eth(pos->other_end, dst)) + continue; + + /* most of the time this function is invoked during the normal + * process..it makes sens to pay more when the session is + * finished and to speed the process up during the measurement + */ + if (unlikely(!kref_get_unless_zero(&pos->refcount))) + continue; + + tp_vars = pos; + break; + } + rcu_read_unlock(); + + return tp_vars; +} + +/** + * batadv_tp_list_find_session - find tp_vars session object in the global list + * @bat_priv: the bat priv with all the soft interface information + * @dst: the other endpoint MAC address to look for + * @session: session identifier + * + * Look for a tp_vars object matching dst as end_point, session as tp meter + * session and return it after having incremented the refcounter. Return NULL + * is not found + * + * Return: matching tp_vars or NULL when no tp_vars was found + */ +static struct batadv_tp_vars * +batadv_tp_list_find_session(struct batadv_priv *bat_priv, const u8 *dst, + const u8 *session) +{ + struct batadv_tp_vars *pos, *tp_vars = NULL; + + rcu_read_lock(); + hlist_for_each_entry_rcu(pos, &bat_priv->tp_list, list) { + if (!batadv_compare_eth(pos->other_end, dst)) + continue; + + if (memcmp(pos->session, session, sizeof(pos->session)) != 0) + continue; + + /* most of the time this function is invoked during the normal + * process..it makes sense to pay more when the session is + * finished and to speed the process up during the measurement + */ + if (unlikely(!kref_get_unless_zero(&pos->refcount))) + continue; + + tp_vars = pos; + break; + } + rcu_read_unlock(); + + return tp_vars; +} + +/** + * batadv_tp_vars_release - release batadv_tp_vars from lists and queue for + * free after rcu grace period + * @ref: kref pointer of the batadv_tp_vars + */ +static void batadv_tp_vars_release(struct kref *ref) +{ + struct batadv_tp_vars *tp_vars; + struct batadv_tp_unacked *un, *safe; + + tp_vars = container_of(ref, struct batadv_tp_vars, refcount); + + /* lock should not be needed because this object is now out of any + * context! + */ + spin_lock_bh(&tp_vars->unacked_lock); + list_for_each_entry_safe(un, safe, &tp_vars->unacked_list, list) { + list_del(&un->list); + kfree(un); + } + spin_unlock_bh(&tp_vars->unacked_lock); + + kfree_rcu(tp_vars, rcu); +} + +/** + * batadv_tp_vars_put - decrement the batadv_tp_vars refcounter and possibly + * release it + * @tp_vars: the private data of the current TP meter session to be free'd + */ +static void batadv_tp_vars_put(struct batadv_tp_vars *tp_vars) +{ + kref_put(&tp_vars->refcount, batadv_tp_vars_release); +} + +/** + * batadv_tp_sender_cleanup - cleanup sender data and drop and timer + * @bat_priv: the bat priv with all the soft interface information + * @tp_vars: the private data of the current TP meter session to cleanup + */ +static void batadv_tp_sender_cleanup(struct batadv_priv *bat_priv, + struct batadv_tp_vars *tp_vars) +{ + cancel_delayed_work(&tp_vars->finish_work); + + spin_lock_bh(&tp_vars->bat_priv->tp_list_lock); + hlist_del_rcu(&tp_vars->list); + spin_unlock_bh(&tp_vars->bat_priv->tp_list_lock); + + /* drop list reference */ + batadv_tp_vars_put(tp_vars); + + atomic_dec(&tp_vars->bat_priv->tp_num); + + /* kill the timer and remove its reference */ + del_timer_sync(&tp_vars->timer); + /* the worker might have rearmed itself therefore we kill it again. Note + * that if the worker should run again before invoking the following + * del_timer(), it would not re-arm itself once again because the status + * is OFF now + */ + del_timer(&tp_vars->timer); + batadv_tp_vars_put(tp_vars); +} + +/** + * batadv_tp_sender_end - print info about ended session and inform client + * @bat_priv: the bat priv with all the soft interface information + * @tp_vars: the private data of the current TP meter session + */ +static void batadv_tp_sender_end(struct batadv_priv *bat_priv, + struct batadv_tp_vars *tp_vars) +{ + u32 session_cookie; + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Test towards %pM finished..shutting down (reason=%d)\n", + tp_vars->other_end, tp_vars->reason); + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Last timing stats: SRTT=%ums RTTVAR=%ums RTO=%ums\n", + tp_vars->srtt >> 3, tp_vars->rttvar >> 2, tp_vars->rto); + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Final values: cwnd=%u ss_threshold=%u\n", + tp_vars->cwnd, tp_vars->ss_threshold); + + session_cookie = batadv_tp_session_cookie(tp_vars->session, + tp_vars->icmp_uid); + + batadv_tp_batctl_notify(tp_vars->reason, + tp_vars->other_end, + bat_priv, + tp_vars->start_time, + atomic64_read(&tp_vars->tot_sent), + session_cookie); +} + +/** + * batadv_tp_sender_shutdown - let sender thread/timer stop gracefully + * @tp_vars: the private data of the current TP meter session + * @reason: reason for tp meter session stop + */ +static void batadv_tp_sender_shutdown(struct batadv_tp_vars *tp_vars, + enum batadv_tp_meter_reason reason) +{ + if (!atomic_dec_and_test(&tp_vars->sending)) + return; + + tp_vars->reason = reason; +} + +/** + * batadv_tp_sender_finish - stop sender session after test_length was reached + * @work: delayed work reference of the related tp_vars + */ +static void batadv_tp_sender_finish(struct work_struct *work) +{ + struct delayed_work *delayed_work; + struct batadv_tp_vars *tp_vars; + + delayed_work = to_delayed_work(work); + tp_vars = container_of(delayed_work, struct batadv_tp_vars, + finish_work); + + batadv_tp_sender_shutdown(tp_vars, BATADV_TP_REASON_COMPLETE); +} + +/** + * batadv_tp_reset_sender_timer - reschedule the sender timer + * @tp_vars: the private TP meter data for this session + * + * Reschedule the timer using tp_vars->rto as delay + */ +static void batadv_tp_reset_sender_timer(struct batadv_tp_vars *tp_vars) +{ + /* most of the time this function is invoked while normal packet + * reception... + */ + if (unlikely(atomic_read(&tp_vars->sending) == 0)) + /* timer ref will be dropped in batadv_tp_sender_cleanup */ + return; + + mod_timer(&tp_vars->timer, jiffies + msecs_to_jiffies(tp_vars->rto)); +} + +/** + * batadv_tp_sender_timeout - timer that fires in case of packet loss + * @arg: address of the related tp_vars + * + * If fired it means that there was packet loss. + * Switch to Slow Start, set the ss_threshold to half of the current cwnd and + * reset the cwnd to 3*MSS + */ +static void batadv_tp_sender_timeout(unsigned long arg) +{ + struct batadv_tp_vars *tp_vars = (struct batadv_tp_vars *)arg; + struct batadv_priv *bat_priv = tp_vars->bat_priv; + + if (atomic_read(&tp_vars->sending) == 0) + return; + + /* if the user waited long enough...shutdown the test */ + if (unlikely(tp_vars->rto >= BATADV_TP_MAX_RTO)) { + batadv_tp_sender_shutdown(tp_vars, + BATADV_TP_REASON_DST_UNREACHABLE); + return; + } + + /* RTO exponential backoff + * Details in Section 5.5 of RFC6298 + */ + tp_vars->rto <<= 1; + + spin_lock_bh(&tp_vars->cwnd_lock); + + tp_vars->ss_threshold = tp_vars->cwnd >> 1; + if (tp_vars->ss_threshold < BATADV_TP_PLEN * 2) + tp_vars->ss_threshold = BATADV_TP_PLEN * 2; + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: RTO fired during test towards %pM! cwnd=%u new ss_thr=%u, resetting last_sent to %u\n", + tp_vars->other_end, tp_vars->cwnd, tp_vars->ss_threshold, + atomic_read(&tp_vars->last_acked)); + + tp_vars->cwnd = BATADV_TP_PLEN * 3; + + spin_unlock_bh(&tp_vars->cwnd_lock); + + /* resend the non-ACKed packets.. */ + tp_vars->last_sent = atomic_read(&tp_vars->last_acked); + wake_up(&tp_vars->more_bytes); + + batadv_tp_reset_sender_timer(tp_vars); +} + +/** + * batadv_tp_fill_prerandom - Fill buffer with prefetched random bytes + * @tp_vars: the private TP meter data for this session + * @buf: Buffer to fill with bytes + * @nbytes: amount of pseudorandom bytes + */ +static void batadv_tp_fill_prerandom(struct batadv_tp_vars *tp_vars, + u8 *buf, size_t nbytes) +{ + u32 local_offset; + size_t bytes_inbuf; + size_t to_copy; + size_t pos = 0; + + spin_lock_bh(&tp_vars->prerandom_lock); + local_offset = tp_vars->prerandom_offset; + tp_vars->prerandom_offset += nbytes; + tp_vars->prerandom_offset %= sizeof(batadv_tp_prerandom); + spin_unlock_bh(&tp_vars->prerandom_lock); + + while (nbytes) { + local_offset %= sizeof(batadv_tp_prerandom); + bytes_inbuf = sizeof(batadv_tp_prerandom) - local_offset; + to_copy = min(nbytes, bytes_inbuf); + + memcpy(&buf[pos], &batadv_tp_prerandom[local_offset], to_copy); + pos += to_copy; + nbytes -= to_copy; + local_offset = 0; + } +} + +/** + * batadv_tp_send_msg - send a single message + * @tp_vars: the private TP meter data for this session + * @src: source mac address + * @orig_node: the originator of the destination + * @seqno: sequence number of this packet + * @len: length of the entire packet + * @session: session identifier + * @uid: local ICMP "socket" index + * @timestamp: timestamp in jiffies which is replied in ack + * + * Create and send a single TP Meter message. + * + * Return: 0 on success, BATADV_TP_REASON_DST_UNREACHABLE if the destination is + * not reachable, BATADV_TP_REASON_MEMORY_ERROR if the packet couldn't be + * allocated + */ +static int batadv_tp_send_msg(struct batadv_tp_vars *tp_vars, const u8 *src, + struct batadv_orig_node *orig_node, + u32 seqno, size_t len, const u8 *session, + int uid, u32 timestamp) +{ + struct batadv_icmp_tp_packet *icmp; + struct sk_buff *skb; + int r; + u8 *data; + size_t data_len; + + skb = netdev_alloc_skb_ip_align(NULL, len + ETH_HLEN); + if (unlikely(!skb)) + return BATADV_TP_REASON_MEMORY_ERROR; + + skb_reserve(skb, ETH_HLEN); + icmp = (struct batadv_icmp_tp_packet *)skb_put(skb, sizeof(*icmp)); + + /* fill the icmp header */ + ether_addr_copy(icmp->dst, orig_node->orig); + ether_addr_copy(icmp->orig, src); + icmp->version = BATADV_COMPAT_VERSION; + icmp->packet_type = BATADV_ICMP; + icmp->ttl = BATADV_TTL; + icmp->msg_type = BATADV_TP; + icmp->uid = uid; + + icmp->subtype = BATADV_TP_MSG; + memcpy(icmp->session, session, sizeof(icmp->session)); + icmp->seqno = htonl(seqno); + icmp->timestamp = htonl(timestamp); + + data_len = len - sizeof(*icmp); + data = (u8 *)skb_put(skb, data_len); + batadv_tp_fill_prerandom(tp_vars, data, data_len); + + r = batadv_send_skb_to_orig(skb, orig_node, NULL); + if (r == -1) + kfree_skb(skb); + + if (r == NET_XMIT_SUCCESS) + return 0; + + return BATADV_TP_REASON_CANT_SEND; +} + +/** + * batadv_tp_recv_ack - ACK receiving function + * @bat_priv: the bat priv with all the soft interface information + * @skb: the buffer containing the received packet + * + * Process a received TP ACK packet + */ +static void batadv_tp_recv_ack(struct batadv_priv *bat_priv, + const struct sk_buff *skb) +{ + struct batadv_hard_iface *primary_if = NULL; + struct batadv_orig_node *orig_node = NULL; + const struct batadv_icmp_tp_packet *icmp; + struct batadv_tp_vars *tp_vars; + size_t packet_len, mss; + u32 rtt, recv_ack, cwnd; + unsigned char *dev_addr; + + packet_len = BATADV_TP_PLEN; + mss = BATADV_TP_PLEN; + packet_len += sizeof(struct batadv_unicast_packet); + + icmp = (struct batadv_icmp_tp_packet *)skb->data; + + /* find the tp_vars */ + tp_vars = batadv_tp_list_find_session(bat_priv, icmp->orig, + icmp->session); + if (unlikely(!tp_vars)) + return; + + if (unlikely(atomic_read(&tp_vars->sending) == 0)) + goto out; + + /* old ACK? silently drop it.. */ + if (batadv_seq_before(ntohl(icmp->seqno), + (u32)atomic_read(&tp_vars->last_acked))) + goto out; + + primary_if = batadv_primary_if_get_selected(bat_priv); + if (unlikely(!primary_if)) + goto out; + + orig_node = batadv_orig_hash_find(bat_priv, icmp->orig); + if (unlikely(!orig_node)) + goto out; + + /* update RTO with the new sampled RTT, if any */ + rtt = jiffies_to_msecs(jiffies) - ntohl(icmp->timestamp); + if (icmp->timestamp && rtt) + batadv_tp_update_rto(tp_vars, rtt); + + /* ACK for new data... reset the timer */ + batadv_tp_reset_sender_timer(tp_vars); + + recv_ack = ntohl(icmp->seqno); + + /* check if this ACK is a duplicate */ + if (atomic_read(&tp_vars->last_acked) == recv_ack) { + atomic_inc(&tp_vars->dup_acks); + if (atomic_read(&tp_vars->dup_acks) != 3) + goto out; + + if (recv_ack >= tp_vars->recover) + goto out; + + /* if this is the third duplicate ACK do Fast Retransmit */ + batadv_tp_send_msg(tp_vars, primary_if->net_dev->dev_addr, + orig_node, recv_ack, packet_len, + icmp->session, icmp->uid, + jiffies_to_msecs(jiffies)); + + spin_lock_bh(&tp_vars->cwnd_lock); + + /* Fast Recovery */ + tp_vars->fast_recovery = true; + /* Set recover to the last outstanding seqno when Fast Recovery + * is entered. RFC6582, Section 3.2, step 1 + */ + tp_vars->recover = tp_vars->last_sent; + tp_vars->ss_threshold = tp_vars->cwnd >> 1; + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: Fast Recovery, (cur cwnd=%u) ss_thr=%u last_sent=%u recv_ack=%u\n", + tp_vars->cwnd, tp_vars->ss_threshold, + tp_vars->last_sent, recv_ack); + tp_vars->cwnd = batadv_tp_cwnd(tp_vars->ss_threshold, 3 * mss, + mss); + tp_vars->dec_cwnd = 0; + tp_vars->last_sent = recv_ack; + + spin_unlock_bh(&tp_vars->cwnd_lock); + } else { + /* count the acked data */ + atomic64_add(recv_ack - atomic_read(&tp_vars->last_acked), + &tp_vars->tot_sent); + /* reset the duplicate ACKs counter */ + atomic_set(&tp_vars->dup_acks, 0); + + if (tp_vars->fast_recovery) { + /* partial ACK */ + if (batadv_seq_before(recv_ack, tp_vars->recover)) { + /* this is another hole in the window. React + * immediately as specified by NewReno (see + * Section 3.2 of RFC6582 for details) + */ + dev_addr = primary_if->net_dev->dev_addr; + batadv_tp_send_msg(tp_vars, dev_addr, + orig_node, recv_ack, + packet_len, icmp->session, + icmp->uid, + jiffies_to_msecs(jiffies)); + tp_vars->cwnd = batadv_tp_cwnd(tp_vars->cwnd, + mss, mss); + } else { + tp_vars->fast_recovery = false; + /* set cwnd to the value of ss_threshold at the + * moment that Fast Recovery was entered. + * RFC6582, Section 3.2, step 3 + */ + cwnd = batadv_tp_cwnd(tp_vars->ss_threshold, 0, + mss); + tp_vars->cwnd = cwnd; + } + goto move_twnd; + } + + if (recv_ack - atomic_read(&tp_vars->last_acked) >= mss) + batadv_tp_update_cwnd(tp_vars, mss); +move_twnd: + /* move the Transmit Window */ + atomic_set(&tp_vars->last_acked, recv_ack); + } + + wake_up(&tp_vars->more_bytes); +out: + if (likely(primary_if)) + batadv_hardif_put(primary_if); + if (likely(orig_node)) + batadv_orig_node_put(orig_node); + if (likely(tp_vars)) + batadv_tp_vars_put(tp_vars); +} + +/** + * batadv_tp_avail - check if congestion window is not full + * @tp_vars: the private data of the current TP meter session + * @payload_len: size of the payload of a single message + * + * Return: true when congestion window is not full, false otherwise + */ +static bool batadv_tp_avail(struct batadv_tp_vars *tp_vars, + size_t payload_len) +{ + u32 win_left, win_limit; + + win_limit = atomic_read(&tp_vars->last_acked) + tp_vars->cwnd; + win_left = win_limit - tp_vars->last_sent; + + return win_left >= payload_len; +} + +/** + * batadv_tp_wait_available - wait until congestion window becomes free or + * timeout is reached + * @tp_vars: the private data of the current TP meter session + * @plen: size of the payload of a single message + * + * Return: 0 if the condition evaluated to false after the timeout elapsed, + * 1 if the condition evaluated to true after the timeout elapsed, the + * remaining jiffies (at least 1) if the condition evaluated to true before + * the timeout elapsed, or -ERESTARTSYS if it was interrupted by a signal. + */ +static int batadv_tp_wait_available(struct batadv_tp_vars *tp_vars, size_t plen) +{ + int ret; + + ret = wait_event_interruptible_timeout(tp_vars->more_bytes, + batadv_tp_avail(tp_vars, plen), + HZ / 10); + + return ret; +} + +/** + * batadv_tp_send - main sending thread of a tp meter session + * @arg: address of the related tp_vars + * + * Return: nothing, this function never returns + */ +static int batadv_tp_send(void *arg) +{ + struct batadv_tp_vars *tp_vars = arg; + struct batadv_priv *bat_priv = tp_vars->bat_priv; + struct batadv_hard_iface *primary_if = NULL; + struct batadv_orig_node *orig_node = NULL; + size_t payload_len, packet_len; + int err = 0; + + if (unlikely(tp_vars->role != BATADV_TP_SENDER)) { + err = BATADV_TP_REASON_DST_UNREACHABLE; + tp_vars->reason = err; + goto out; + } + + orig_node = batadv_orig_hash_find(bat_priv, tp_vars->other_end); + if (unlikely(!orig_node)) { + err = BATADV_TP_REASON_DST_UNREACHABLE; + tp_vars->reason = err; + goto out; + } + + primary_if = batadv_primary_if_get_selected(bat_priv); + if (unlikely(!primary_if)) { + err = BATADV_TP_REASON_DST_UNREACHABLE; + goto out; + } + + /* assume that all the hard_interfaces have a correctly + * configured MTU, so use the soft_iface MTU as MSS. + * This might not be true and in that case the fragmentation + * should be used. + * Now, try to send the packet as it is + */ + payload_len = BATADV_TP_PLEN; + BUILD_BUG_ON(sizeof(struct batadv_icmp_tp_packet) > BATADV_TP_PLEN); + + batadv_tp_reset_sender_timer(tp_vars); + + /* queue the worker in charge of terminating the test */ + queue_delayed_work(batadv_event_workqueue, &tp_vars->finish_work, + msecs_to_jiffies(tp_vars->test_length)); + + while (atomic_read(&tp_vars->sending) != 0) { + if (unlikely(!batadv_tp_avail(tp_vars, payload_len))) { + batadv_tp_wait_available(tp_vars, payload_len); + continue; + } + + /* to emulate normal unicast traffic, add to the payload len + * the size of the unicast header + */ + packet_len = payload_len + sizeof(struct batadv_unicast_packet); + + err = batadv_tp_send_msg(tp_vars, primary_if->net_dev->dev_addr, + orig_node, tp_vars->last_sent, + packet_len, + tp_vars->session, tp_vars->icmp_uid, + jiffies_to_msecs(jiffies)); + + /* something went wrong during the preparation/transmission */ + if (unlikely(err && err != BATADV_TP_REASON_CANT_SEND)) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: batadv_tp_send() cannot send packets (%d)\n", + err); + /* ensure nobody else tries to stop the thread now */ + if (atomic_dec_and_test(&tp_vars->sending)) + tp_vars->reason = err; + break; + } + + /* right-shift the TWND */ + if (!err) + tp_vars->last_sent += payload_len; + + cond_resched(); + } + +out: + if (likely(primary_if)) + batadv_hardif_put(primary_if); + if (likely(orig_node)) + batadv_orig_node_put(orig_node); + + batadv_tp_sender_end(bat_priv, tp_vars); + batadv_tp_sender_cleanup(bat_priv, tp_vars); + + batadv_tp_vars_put(tp_vars); + + do_exit(0); +} + +/** + * batadv_tp_start_kthread - start new thread which manages the tp meter sender + * @tp_vars: the private data of the current TP meter session + */ +static void batadv_tp_start_kthread(struct batadv_tp_vars *tp_vars) +{ + struct task_struct *kthread; + struct batadv_priv *bat_priv = tp_vars->bat_priv; + u32 session_cookie; + + kref_get(&tp_vars->refcount); + kthread = kthread_create(batadv_tp_send, tp_vars, "kbatadv_tp_meter"); + if (IS_ERR(kthread)) { + session_cookie = batadv_tp_session_cookie(tp_vars->session, + tp_vars->icmp_uid); + pr_err("batadv: cannot create tp meter kthread\n"); + batadv_tp_batctl_error_notify(BATADV_TP_REASON_MEMORY_ERROR, + tp_vars->other_end, + bat_priv, session_cookie); + + /* drop reserved reference for kthread */ + batadv_tp_vars_put(tp_vars); + + /* cleanup of failed tp meter variables */ + batadv_tp_sender_cleanup(bat_priv, tp_vars); + return; + } + + wake_up_process(kthread); +} + +/** + * batadv_tp_start - start a new tp meter session + * @bat_priv: the bat priv with all the soft interface information + * @dst: the receiver MAC address + * @test_length: test length in milliseconds + * @cookie: session cookie + */ +void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, + u32 test_length, u32 *cookie) +{ + struct batadv_tp_vars *tp_vars; + u8 session_id[2]; + u8 icmp_uid; + u32 session_cookie; + + get_random_bytes(session_id, sizeof(session_id)); + get_random_bytes(&icmp_uid, 1); + session_cookie = batadv_tp_session_cookie(session_id, icmp_uid); + *cookie = session_cookie; + + /* look for an already existing test towards this node */ + spin_lock_bh(&bat_priv->tp_list_lock); + tp_vars = batadv_tp_list_find(bat_priv, dst); + if (tp_vars) { + spin_unlock_bh(&bat_priv->tp_list_lock); + batadv_tp_vars_put(tp_vars); + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: test to or from the same node already ongoing, aborting\n"); + batadv_tp_batctl_error_notify(BATADV_TP_REASON_ALREADY_ONGOING, + dst, bat_priv, session_cookie); + return; + } + + if (!atomic_add_unless(&bat_priv->tp_num, 1, BATADV_TP_MAX_NUM)) { + spin_unlock_bh(&bat_priv->tp_list_lock); + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: too many ongoing sessions, aborting (SEND)\n"); + batadv_tp_batctl_error_notify(BATADV_TP_REASON_TOO_MANY, dst, + bat_priv, session_cookie); + return; + } + + tp_vars = kmalloc(sizeof(*tp_vars), GFP_ATOMIC); + if (!tp_vars) { + spin_unlock_bh(&bat_priv->tp_list_lock); + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: batadv_tp_start cannot allocate list elements\n"); + batadv_tp_batctl_error_notify(BATADV_TP_REASON_MEMORY_ERROR, + dst, bat_priv, session_cookie); + return; + } + + /* initialize tp_vars */ + ether_addr_copy(tp_vars->other_end, dst); + kref_init(&tp_vars->refcount); + tp_vars->role = BATADV_TP_SENDER; + atomic_set(&tp_vars->sending, 1); + memcpy(tp_vars->session, session_id, sizeof(session_id)); + tp_vars->icmp_uid = icmp_uid; + + tp_vars->last_sent = BATADV_TP_FIRST_SEQ; + atomic_set(&tp_vars->last_acked, BATADV_TP_FIRST_SEQ); + tp_vars->fast_recovery = false; + tp_vars->recover = BATADV_TP_FIRST_SEQ; + + /* initialise the CWND to 3*MSS (Section 3.1 in RFC5681). + * For batman-adv the MSS is the size of the payload received by the + * soft_interface, hence its MTU + */ + tp_vars->cwnd = BATADV_TP_PLEN * 3; + /* at the beginning initialise the SS threshold to the biggest possible + * window size, hence the AWND size + */ + tp_vars->ss_threshold = BATADV_TP_AWND; + + /* RTO initial value is 3 seconds. + * Details in Section 2.1 of RFC6298 + */ + tp_vars->rto = 1000; + tp_vars->srtt = 0; + tp_vars->rttvar = 0; + + atomic64_set(&tp_vars->tot_sent, 0); + + kref_get(&tp_vars->refcount); + setup_timer(&tp_vars->timer, batadv_tp_sender_timeout, + (unsigned long)tp_vars); + + tp_vars->bat_priv = bat_priv; + tp_vars->start_time = jiffies; + + init_waitqueue_head(&tp_vars->more_bytes); + + spin_lock_init(&tp_vars->unacked_lock); + INIT_LIST_HEAD(&tp_vars->unacked_list); + + spin_lock_init(&tp_vars->cwnd_lock); + + tp_vars->prerandom_offset = 0; + spin_lock_init(&tp_vars->prerandom_lock); + + kref_get(&tp_vars->refcount); + hlist_add_head_rcu(&tp_vars->list, &bat_priv->tp_list); + spin_unlock_bh(&bat_priv->tp_list_lock); + + tp_vars->test_length = test_length; + if (!tp_vars->test_length) + tp_vars->test_length = BATADV_TP_DEF_TEST_LENGTH; + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: starting throughput meter towards %pM (length=%ums)\n", + dst, test_length); + + /* init work item for finished tp tests */ + INIT_DELAYED_WORK(&tp_vars->finish_work, batadv_tp_sender_finish); + + /* start tp kthread. This way the write() call issued from userspace can + * happily return and avoid to block + */ + batadv_tp_start_kthread(tp_vars); + + /* don't return reference to new tp_vars */ + batadv_tp_vars_put(tp_vars); +} + +/** + * batadv_tp_stop - stop currently running tp meter session + * @bat_priv: the bat priv with all the soft interface information + * @dst: the receiver MAC address + * @return_value: reason for tp meter session stop + */ +void batadv_tp_stop(struct batadv_priv *bat_priv, const u8 *dst, + u8 return_value) +{ + struct batadv_orig_node *orig_node; + struct batadv_tp_vars *tp_vars; + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: stopping test towards %pM\n", dst); + + orig_node = batadv_orig_hash_find(bat_priv, dst); + if (!orig_node) + return; + + tp_vars = batadv_tp_list_find(bat_priv, orig_node->orig); + if (!tp_vars) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: trying to interrupt an already over connection\n"); + goto out; + } + + batadv_tp_sender_shutdown(tp_vars, return_value); + batadv_tp_vars_put(tp_vars); +out: + batadv_orig_node_put(orig_node); +} + +/** + * batadv_tp_reset_receiver_timer - reset the receiver shutdown timer + * @tp_vars: the private data of the current TP meter session + * + * start the receiver shutdown timer or reset it if already started + */ +static void batadv_tp_reset_receiver_timer(struct batadv_tp_vars *tp_vars) +{ + mod_timer(&tp_vars->timer, + jiffies + msecs_to_jiffies(BATADV_TP_RECV_TIMEOUT)); +} + +/** + * batadv_tp_receiver_shutdown - stop a tp meter receiver when timeout is + * reached without received ack + * @arg: address of the related tp_vars + */ +static void batadv_tp_receiver_shutdown(unsigned long arg) +{ + struct batadv_tp_vars *tp_vars = (struct batadv_tp_vars *)arg; + struct batadv_tp_unacked *un, *safe; + struct batadv_priv *bat_priv; + + bat_priv = tp_vars->bat_priv; + + /* if there is recent activity rearm the timer */ + if (!batadv_has_timed_out(tp_vars->last_recv_time, + BATADV_TP_RECV_TIMEOUT)) { + /* reset the receiver shutdown timer */ + batadv_tp_reset_receiver_timer(tp_vars); + return; + } + + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Shutting down for inactivity (more than %dms) from %pM\n", + BATADV_TP_RECV_TIMEOUT, tp_vars->other_end); + + spin_lock_bh(&tp_vars->bat_priv->tp_list_lock); + hlist_del_rcu(&tp_vars->list); + spin_unlock_bh(&tp_vars->bat_priv->tp_list_lock); + + /* drop list reference */ + batadv_tp_vars_put(tp_vars); + + atomic_dec(&bat_priv->tp_num); + + spin_lock_bh(&tp_vars->unacked_lock); + list_for_each_entry_safe(un, safe, &tp_vars->unacked_list, list) { + list_del(&un->list); + kfree(un); + } + spin_unlock_bh(&tp_vars->unacked_lock); + + /* drop reference of timer */ + batadv_tp_vars_put(tp_vars); +} + +/** + * batadv_tp_send_ack - send an ACK packet + * @bat_priv: the bat priv with all the soft interface information + * @dst: the mac address of the destination originator + * @seq: the sequence number to ACK + * @timestamp: the timestamp to echo back in the ACK + * @session: session identifier + * @socket_index: local ICMP socket identifier + * + * Return: 0 on success, a positive integer representing the reason of the + * failure otherwise + */ +static int batadv_tp_send_ack(struct batadv_priv *bat_priv, const u8 *dst, + u32 seq, __be32 timestamp, const u8 *session, + int socket_index) +{ + struct batadv_hard_iface *primary_if = NULL; + struct batadv_orig_node *orig_node; + struct batadv_icmp_tp_packet *icmp; + struct sk_buff *skb; + int r, ret; + + orig_node = batadv_orig_hash_find(bat_priv, dst); + if (unlikely(!orig_node)) { + ret = BATADV_TP_REASON_DST_UNREACHABLE; + goto out; + } + + primary_if = batadv_primary_if_get_selected(bat_priv); + if (unlikely(!primary_if)) { + ret = BATADV_TP_REASON_DST_UNREACHABLE; + goto out; + } + + skb = netdev_alloc_skb_ip_align(NULL, sizeof(*icmp) + ETH_HLEN); + if (unlikely(!skb)) { + ret = BATADV_TP_REASON_MEMORY_ERROR; + goto out; + } + + skb_reserve(skb, ETH_HLEN); + icmp = (struct batadv_icmp_tp_packet *)skb_put(skb, sizeof(*icmp)); + icmp->packet_type = BATADV_ICMP; + icmp->version = BATADV_COMPAT_VERSION; + icmp->ttl = BATADV_TTL; + icmp->msg_type = BATADV_TP; + ether_addr_copy(icmp->dst, orig_node->orig); + ether_addr_copy(icmp->orig, primary_if->net_dev->dev_addr); + icmp->uid = socket_index; + + icmp->subtype = BATADV_TP_ACK; + memcpy(icmp->session, session, sizeof(icmp->session)); + icmp->seqno = htonl(seq); + icmp->timestamp = timestamp; + + /* send the ack */ + r = batadv_send_skb_to_orig(skb, orig_node, NULL); + if (r == -1) + kfree_skb(skb); + + if (unlikely(r < 0) || (r == NET_XMIT_DROP)) { + ret = BATADV_TP_REASON_DST_UNREACHABLE; + goto out; + } + ret = 0; + +out: + if (likely(orig_node)) + batadv_orig_node_put(orig_node); + if (likely(primary_if)) + batadv_hardif_put(primary_if); + + return ret; +} + +/** + * batadv_tp_handle_out_of_order - store an out of order packet + * @tp_vars: the private data of the current TP meter session + * @skb: the buffer containing the received packet + * + * Store the out of order packet in the unacked list for late processing. This + * packets are kept in this list so that they can be ACKed at once as soon as + * all the previous packets have been received + * + * Return: true if the packed has been successfully processed, false otherwise + */ +static bool batadv_tp_handle_out_of_order(struct batadv_tp_vars *tp_vars, + const struct sk_buff *skb) +{ + const struct batadv_icmp_tp_packet *icmp; + struct batadv_tp_unacked *un, *new; + u32 payload_len; + bool added = false; + + new = kmalloc(sizeof(*new), GFP_ATOMIC); + if (unlikely(!new)) + return false; + + icmp = (struct batadv_icmp_tp_packet *)skb->data; + + new->seqno = ntohl(icmp->seqno); + payload_len = skb->len - sizeof(struct batadv_unicast_packet); + new->len = payload_len; + + spin_lock_bh(&tp_vars->unacked_lock); + /* if the list is empty immediately attach this new object */ + if (list_empty(&tp_vars->unacked_list)) { + list_add(&new->list, &tp_vars->unacked_list); + goto out; + } + + /* otherwise loop over the list and either drop the packet because this + * is a duplicate or store it at the right position. + * + * The iteration is done in the reverse way because it is likely that + * the last received packet (the one being processed now) has a bigger + * seqno than all the others already stored. + */ + list_for_each_entry_reverse(un, &tp_vars->unacked_list, list) { + /* check for duplicates */ + if (new->seqno == un->seqno) { + if (new->len > un->len) + un->len = new->len; + kfree(new); + added = true; + break; + } + + /* look for the right position */ + if (batadv_seq_before(new->seqno, un->seqno)) + continue; + + /* as soon as an entry having a bigger seqno is found, the new + * one is attached _after_ it. In this way the list is kept in + * ascending order + */ + list_add_tail(&new->list, &un->list); + added = true; + break; + } + + /* received packet with smallest seqno out of order; add it to front */ + if (!added) + list_add(&new->list, &tp_vars->unacked_list); + +out: + spin_unlock_bh(&tp_vars->unacked_lock); + + return true; +} + +/** + * batadv_tp_ack_unordered - update number received bytes in current stream + * without gaps + * @tp_vars: the private data of the current TP meter session + */ +static void batadv_tp_ack_unordered(struct batadv_tp_vars *tp_vars) +{ + struct batadv_tp_unacked *un, *safe; + u32 to_ack; + + /* go through the unacked packet list and possibly ACK them as + * well + */ + spin_lock_bh(&tp_vars->unacked_lock); + list_for_each_entry_safe(un, safe, &tp_vars->unacked_list, list) { + /* the list is ordered, therefore it is possible to stop as soon + * there is a gap between the last acked seqno and the seqno of + * the packet under inspection + */ + if (batadv_seq_before(tp_vars->last_recv, un->seqno)) + break; + + to_ack = un->seqno + un->len - tp_vars->last_recv; + + if (batadv_seq_before(tp_vars->last_recv, un->seqno + un->len)) + tp_vars->last_recv += to_ack; + + list_del(&un->list); + kfree(un); + } + spin_unlock_bh(&tp_vars->unacked_lock); +} + +/** + * batadv_tp_init_recv - return matching or create new receiver tp_vars + * @bat_priv: the bat priv with all the soft interface information + * @icmp: received icmp tp msg + * + * Return: corresponding tp_vars or NULL on errors + */ +static struct batadv_tp_vars * +batadv_tp_init_recv(struct batadv_priv *bat_priv, + const struct batadv_icmp_tp_packet *icmp) +{ + struct batadv_tp_vars *tp_vars; + + spin_lock_bh(&bat_priv->tp_list_lock); + tp_vars = batadv_tp_list_find_session(bat_priv, icmp->orig, + icmp->session); + if (tp_vars) + goto out_unlock; + + if (!atomic_add_unless(&bat_priv->tp_num, 1, BATADV_TP_MAX_NUM)) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: too many ongoing sessions, aborting (RECV)\n"); + goto out_unlock; + } + + tp_vars = kmalloc(sizeof(*tp_vars), GFP_ATOMIC); + if (!tp_vars) + goto out_unlock; + + ether_addr_copy(tp_vars->other_end, icmp->orig); + tp_vars->role = BATADV_TP_RECEIVER; + memcpy(tp_vars->session, icmp->session, sizeof(tp_vars->session)); + tp_vars->last_recv = BATADV_TP_FIRST_SEQ; + tp_vars->bat_priv = bat_priv; + kref_init(&tp_vars->refcount); + + spin_lock_init(&tp_vars->unacked_lock); + INIT_LIST_HEAD(&tp_vars->unacked_list); + + kref_get(&tp_vars->refcount); + hlist_add_head_rcu(&tp_vars->list, &bat_priv->tp_list); + + kref_get(&tp_vars->refcount); + setup_timer(&tp_vars->timer, batadv_tp_receiver_shutdown, + (unsigned long)tp_vars); + + batadv_tp_reset_receiver_timer(tp_vars); + +out_unlock: + spin_unlock_bh(&bat_priv->tp_list_lock); + + return tp_vars; +} + +/** + * batadv_tp_recv_msg - process a single data message + * @bat_priv: the bat priv with all the soft interface information + * @skb: the buffer containing the received packet + * + * Process a received TP MSG packet + */ +static void batadv_tp_recv_msg(struct batadv_priv *bat_priv, + const struct sk_buff *skb) +{ + const struct batadv_icmp_tp_packet *icmp; + struct batadv_tp_vars *tp_vars; + size_t packet_size; + u32 seqno; + + icmp = (struct batadv_icmp_tp_packet *)skb->data; + + seqno = ntohl(icmp->seqno); + /* check if this is the first seqno. This means that if the + * first packet is lost, the tp meter does not work anymore! + */ + if (seqno == BATADV_TP_FIRST_SEQ) { + tp_vars = batadv_tp_init_recv(bat_priv, icmp); + if (!tp_vars) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: seqno != BATADV_TP_FIRST_SEQ cannot initiate connection\n"); + goto out; + } + } else { + tp_vars = batadv_tp_list_find_session(bat_priv, icmp->orig, + icmp->session); + if (!tp_vars) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Unexpected packet from %pM!\n", + icmp->orig); + goto out; + } + } + + if (unlikely(tp_vars->role != BATADV_TP_RECEIVER)) { + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Meter: dropping packet: not expected (role=%u)\n", + tp_vars->role); + goto out; + } + + tp_vars->last_recv_time = jiffies; + + /* if the packet is a duplicate, it may be the case that an ACK has been + * lost. Resend the ACK + */ + if (batadv_seq_before(seqno, tp_vars->last_recv)) + goto send_ack; + + /* if the packet is out of order enqueue it */ + if (ntohl(icmp->seqno) != tp_vars->last_recv) { + /* exit immediately (and do not send any ACK) if the packet has + * not been enqueued correctly + */ + if (!batadv_tp_handle_out_of_order(tp_vars, skb)) + goto out; + + /* send a duplicate ACK */ + goto send_ack; + } + + /* if everything was fine count the ACKed bytes */ + packet_size = skb->len - sizeof(struct batadv_unicast_packet); + tp_vars->last_recv += packet_size; + + /* check if this ordered message filled a gap.... */ + batadv_tp_ack_unordered(tp_vars); + +send_ack: + /* send the ACK. If the received packet was out of order, the ACK that + * is going to be sent is a duplicate (the sender will count them and + * possibly enter Fast Retransmit as soon as it has reached 3) + */ + batadv_tp_send_ack(bat_priv, icmp->orig, tp_vars->last_recv, + icmp->timestamp, icmp->session, icmp->uid); +out: + if (likely(tp_vars)) + batadv_tp_vars_put(tp_vars); +} + +/** + * batadv_tp_meter_recv - main TP Meter receiving function + * @bat_priv: the bat priv with all the soft interface information + * @skb: the buffer containing the received packet + */ +void batadv_tp_meter_recv(struct batadv_priv *bat_priv, struct sk_buff *skb) +{ + struct batadv_icmp_tp_packet *icmp; + + icmp = (struct batadv_icmp_tp_packet *)skb->data; + + switch (icmp->subtype) { + case BATADV_TP_MSG: + batadv_tp_recv_msg(bat_priv, skb); + break; + case BATADV_TP_ACK: + batadv_tp_recv_ack(bat_priv, skb); + break; + default: + batadv_dbg(BATADV_DBG_TP_METER, bat_priv, + "Received unknown TP Metric packet type %u\n", + icmp->subtype); + } + consume_skb(skb); +} + +/** + * batadv_tp_meter_init - initialize global tp_meter structures + */ +void batadv_tp_meter_init(void) +{ + get_random_bytes(batadv_tp_prerandom, sizeof(batadv_tp_prerandom)); +} diff --git a/net/batman-adv/tp_meter.h b/net/batman-adv/tp_meter.h new file mode 100644 index 0000000..ba922c4 --- /dev/null +++ b/net/batman-adv/tp_meter.h @@ -0,0 +1,34 @@ +/* Copyright (C) 2012-2016 B.A.T.M.A.N. contributors: + * + * Edo Monticelli, Antonio Quartulli + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, see . + */ + +#ifndef _NET_BATMAN_ADV_TP_METER_H_ +#define _NET_BATMAN_ADV_TP_METER_H_ + +#include "main.h" + +#include + +struct sk_buff; + +void batadv_tp_meter_init(void); +void batadv_tp_start(struct batadv_priv *bat_priv, const u8 *dst, + u32 test_length, u32 *cookie); +void batadv_tp_stop(struct batadv_priv *bat_priv, const u8 *dst, + u8 return_value); +void batadv_tp_meter_recv(struct batadv_priv *bat_priv, struct sk_buff *skb); + +#endif /* _NET_BATMAN_ADV_TP_METER_H_ */ diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index ab863a5..a331e3a 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -33,6 +33,7 @@ #include #include #include +#include #include "packet.h" @@ -833,6 +834,111 @@ struct batadv_priv_nc { }; /** + * struct batadv_tp_unacked - unacked packet meta-information + * @seqno: seqno of the unacked packet + * @len: length of the packet + * @list: list node for batadv_tp_vars::unacked_list + * + * This struct is supposed to represent a buffer unacked packet. However, since + * the purpose of the TP meter is to count the traffic only, there is no need to + * store the entire sk_buff, the starting offset and the length are enough + */ +struct batadv_tp_unacked { + u32 seqno; + u16 len; + struct list_head list; +}; + +/** + * enum batadv_tp_meter_role - Modus in tp meter session + * @BATADV_TP_RECEIVER: Initialized as receiver + * @BATADV_TP_SENDER: Initialized as sender + */ +enum batadv_tp_meter_role { + BATADV_TP_RECEIVER, + BATADV_TP_SENDER +}; + +/** + * struct batadv_tp_vars - tp meter private variables per session + * @list: list node for bat_priv::tp_list + * @timer: timer for ack (receiver) and retry (sender) + * @bat_priv: pointer to the mesh object + * @start_time: start time in jiffies + * @other_end: mac address of remote + * @role: receiver/sender modi + * @sending: sending binary semaphore: 1 if sending, 0 is not + * @reason: reason for a stopped session + * @finish_work: work item for the finishing procedure + * @test_length: test length in milliseconds + * @session: TP session identifier + * @icmp_uid: local ICMP "socket" index + * @dec_cwnd: decimal part of the cwnd used during linear growth + * @cwnd: current size of the congestion window + * @cwnd_lock: lock do protect @cwnd & @dec_cwnd + * @ss_threshold: Slow Start threshold. Once cwnd exceeds this value the + * connection switches to the Congestion Avoidance state + * @last_acked: last acked byte + * @last_sent: last sent byte, not yet acked + * @tot_sent: amount of data sent/ACKed so far + * @dup_acks: duplicate ACKs counter + * @fast_recovery: true if in Fast Recovery mode + * @recover: last sent seqno when entering Fast Recovery + * @rto: sender timeout + * @srtt: smoothed RTT scaled by 2^3 + * @rttvar: RTT variation scaled by 2^2 + * @more_bytes: waiting queue anchor when waiting for more ack/retry timeout + * @prerandom_offset: offset inside the prerandom buffer + * @prerandom_lock: spinlock protecting access to prerandom_offset + * @last_recv: last in-order received packet + * @unacked_list: list of unacked packets (meta-info only) + * @unacked_lock: protect unacked_list + * @last_recv_time: time time (jiffies) a msg was received + * @refcount: number of context where the object is used + * @rcu: struct used for freeing in an RCU-safe manner + */ +struct batadv_tp_vars { + struct hlist_node list; + struct timer_list timer; + struct batadv_priv *bat_priv; + unsigned long start_time; + u8 other_end[ETH_ALEN]; + enum batadv_tp_meter_role role; + atomic_t sending; + enum batadv_tp_meter_reason reason; + struct delayed_work finish_work; + u32 test_length; + u8 session[2]; + u8 icmp_uid; + + /* sender variables */ + u16 dec_cwnd; + u32 cwnd; + spinlock_t cwnd_lock; /* Protects cwnd & dec_cwnd */ + u32 ss_threshold; + atomic_t last_acked; + u32 last_sent; + atomic64_t tot_sent; + atomic_t dup_acks; + bool fast_recovery; + u32 recover; + u32 rto; + u32 srtt; + u32 rttvar; + wait_queue_head_t more_bytes; + u32 prerandom_offset; + spinlock_t prerandom_lock; /* Protects prerandom_offset */ + + /* receiver variables */ + u32 last_recv; + struct list_head unacked_list; + spinlock_t unacked_lock; /* Protects unacked_list */ + unsigned long last_recv_time; + struct kref refcount; + struct rcu_head rcu; +}; + +/** * struct batadv_softif_vlan - per VLAN attributes set * @bat_priv: pointer to the mesh object * @vid: VLAN identifier @@ -900,9 +1006,12 @@ struct batadv_priv_bat_v { * @debug_dir: dentry for debugfs batman-adv subdirectory * @forw_bat_list: list of aggregated OGMs that will be forwarded * @forw_bcast_list: list of broadcast packets that will be rebroadcasted + * @tp_list: list of tp sessions + * @tp_num: number of currently active tp sessions * @orig_hash: hash table containing mesh participants (orig nodes) * @forw_bat_list_lock: lock protecting forw_bat_list * @forw_bcast_list_lock: lock protecting forw_bcast_list + * @tp_list_lock: spinlock protecting @tp_list * @orig_work: work queue callback item for orig node purging * @cleanup_work: work queue callback item for soft-interface deinit * @primary_if: one of the hard-interfaces assigned to this mesh interface @@ -956,9 +1065,12 @@ struct batadv_priv { struct dentry *debug_dir; struct hlist_head forw_bat_list; struct hlist_head forw_bcast_list; + struct hlist_head tp_list; struct batadv_hashtable *orig_hash; spinlock_t forw_bat_list_lock; /* protects forw_bat_list */ spinlock_t forw_bcast_list_lock; /* protects forw_bcast_list */ + spinlock_t tp_list_lock; /* protects tp_list */ + atomic_t tp_num; struct delayed_work orig_work; struct work_struct cleanup_work; struct batadv_hard_iface __rcu *primary_if; /* rcu protected pointer */ -- cgit v0.10.2 From 29824a55c07cd79a530d4bc1020a529c402515b6 Mon Sep 17 00:00:00 2001 From: Antonio Quartulli Date: Wed, 25 May 2016 23:27:31 +0800 Subject: batman-adv: split routing API data structure in subobjects The routing API data structure contains several function pointers that can easily be grouped together based on the component they work with. Split the API in subobjects in order to improve definition readability. At the same time, remove the "bat_" prefix from the API object and its fields names. These are batman-adv private structs and there is no need to always prepend such prefix, which only makes function invocations much much longer. Signed-off-by: Antonio Quartulli Reviewed-by: Sven Eckelmann Signed-off-by: Marek Lindner Signed-off-by: Simon Wunderlich diff --git a/net/batman-adv/bat_algo.c b/net/batman-adv/bat_algo.c index 610d4de..81dbbf5 100644 --- a/net/batman-adv/bat_algo.c +++ b/net/batman-adv/bat_algo.c @@ -65,12 +65,12 @@ int batadv_algo_register(struct batadv_algo_ops *bat_algo_ops) } /* all algorithms must implement all ops (for now) */ - if (!bat_algo_ops->bat_iface_enable || - !bat_algo_ops->bat_iface_disable || - !bat_algo_ops->bat_iface_update_mac || - !bat_algo_ops->bat_primary_iface_set || - !bat_algo_ops->bat_neigh_cmp || - !bat_algo_ops->bat_neigh_is_similar_or_better) { + if (!bat_algo_ops->iface.enable || + !bat_algo_ops->iface.disable || + !bat_algo_ops->iface.update_mac || + !bat_algo_ops->iface.primary_set || + !bat_algo_ops->neigh.cmp || + !bat_algo_ops->neigh.is_similar_or_better) { pr_info("Routing algo '%s' does not implement required ops\n", bat_algo_ops->name); return -EINVAL; @@ -90,7 +90,7 @@ int batadv_algo_select(struct batadv_priv *bat_priv, char *name) if (!bat_algo_ops) return -EINVAL; - bat_priv->bat_algo_ops = bat_algo_ops; + bat_priv->algo_ops = bat_algo_ops; return 0; } diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index e2d8848..19b0abd 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -1853,8 +1853,7 @@ static int batadv_iv_ogm_receive(struct sk_buff *skb, /* did we receive a B.A.T.M.A.N. IV OGM packet on an interface * that does not have B.A.T.M.A.N. IV enabled ? */ - if (bat_priv->bat_algo_ops->bat_iface_enable != - batadv_iv_ogm_iface_enable) + if (bat_priv->algo_ops->iface.enable != batadv_iv_ogm_iface_enable) return NET_RX_DROP; batadv_inc_counter(bat_priv, BATADV_CNT_MGMT_RX); @@ -2120,18 +2119,24 @@ static void batadv_iv_iface_activate(struct batadv_hard_iface *hard_iface) static struct batadv_algo_ops batadv_batman_iv __read_mostly = { .name = "BATMAN_IV", - .bat_iface_activate = batadv_iv_iface_activate, - .bat_iface_enable = batadv_iv_ogm_iface_enable, - .bat_iface_disable = batadv_iv_ogm_iface_disable, - .bat_iface_update_mac = batadv_iv_ogm_iface_update_mac, - .bat_primary_iface_set = batadv_iv_ogm_primary_iface_set, - .bat_neigh_cmp = batadv_iv_ogm_neigh_cmp, - .bat_neigh_is_similar_or_better = batadv_iv_ogm_neigh_is_sob, - .bat_neigh_print = batadv_iv_neigh_print, - .bat_orig_print = batadv_iv_ogm_orig_print, - .bat_orig_free = batadv_iv_ogm_orig_free, - .bat_orig_add_if = batadv_iv_ogm_orig_add_if, - .bat_orig_del_if = batadv_iv_ogm_orig_del_if, + .iface = { + .activate = batadv_iv_iface_activate, + .enable = batadv_iv_ogm_iface_enable, + .disable = batadv_iv_ogm_iface_disable, + .update_mac = batadv_iv_ogm_iface_update_mac, + .primary_set = batadv_iv_ogm_primary_iface_set, + }, + .neigh = { + .cmp = batadv_iv_ogm_neigh_cmp, + .is_similar_or_better = batadv_iv_ogm_neigh_is_sob, + .print = batadv_iv_neigh_print, + }, + .orig = { + .print = batadv_iv_ogm_orig_print, + .free = batadv_iv_ogm_orig_free, + .add_if = batadv_iv_ogm_orig_add_if, + .del_if = batadv_iv_ogm_orig_del_if, + }, }; int __init batadv_iv_init(void) diff --git a/net/batman-adv/bat_v.c b/net/batman-adv/bat_v.c index 7231440..0366cbf 100644 --- a/net/batman-adv/bat_v.c +++ b/net/batman-adv/bat_v.c @@ -322,16 +322,22 @@ err_ifinfo1: static struct batadv_algo_ops batadv_batman_v __read_mostly = { .name = "BATMAN_V", - .bat_iface_activate = batadv_v_iface_activate, - .bat_iface_enable = batadv_v_iface_enable, - .bat_iface_disable = batadv_v_iface_disable, - .bat_iface_update_mac = batadv_v_iface_update_mac, - .bat_primary_iface_set = batadv_v_primary_iface_set, - .bat_hardif_neigh_init = batadv_v_hardif_neigh_init, - .bat_orig_print = batadv_v_orig_print, - .bat_neigh_cmp = batadv_v_neigh_cmp, - .bat_neigh_is_similar_or_better = batadv_v_neigh_is_sob, - .bat_neigh_print = batadv_v_neigh_print, + .iface = { + .activate = batadv_v_iface_activate, + .enable = batadv_v_iface_enable, + .disable = batadv_v_iface_disable, + .update_mac = batadv_v_iface_update_mac, + .primary_set = batadv_v_primary_iface_set, + }, + .neigh = { + .hardif_init = batadv_v_hardif_neigh_init, + .cmp = batadv_v_neigh_cmp, + .is_similar_or_better = batadv_v_neigh_is_sob, + .print = batadv_v_neigh_print, + }, + .orig = { + .print = batadv_v_orig_print, + }, }; /** diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c index 15cf272..7d17001 100644 --- a/net/batman-adv/bat_v_elp.c +++ b/net/batman-adv/bat_v_elp.c @@ -504,7 +504,7 @@ int batadv_v_elp_packet_recv(struct sk_buff *skb, /* did we receive a B.A.T.M.A.N. V ELP packet on an interface * that does not have B.A.T.M.A.N. V ELP enabled ? */ - if (strcmp(bat_priv->bat_algo_ops->name, "BATMAN_V") != 0) + if (strcmp(bat_priv->algo_ops->name, "BATMAN_V") != 0) return NET_RX_DROP; elp_packet = (struct batadv_elp_packet *)skb->data; diff --git a/net/batman-adv/bat_v_ogm.c b/net/batman-adv/bat_v_ogm.c index 7ac9e0b..6fbba4e 100644 --- a/net/batman-adv/bat_v_ogm.c +++ b/net/batman-adv/bat_v_ogm.c @@ -754,7 +754,7 @@ int batadv_v_ogm_packet_recv(struct sk_buff *skb, /* did we receive a OGM2 packet on an interface that does not have * B.A.T.M.A.N. V enabled ? */ - if (strcmp(bat_priv->bat_algo_ops->name, "BATMAN_V") != 0) + if (strcmp(bat_priv->algo_ops->name, "BATMAN_V") != 0) return NET_RX_DROP; if (!batadv_check_management_packet(skb, if_incoming, BATADV_OGM2_HLEN)) diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index 70841c1..1f90808 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -247,7 +247,7 @@ static void batadv_primary_if_select(struct batadv_priv *bat_priv, if (!new_hard_iface) goto out; - bat_priv->bat_algo_ops->bat_primary_iface_set(new_hard_iface); + bat_priv->algo_ops->iface.primary_set(new_hard_iface); batadv_primary_if_update_addr(bat_priv, curr_hard_iface); out: @@ -394,7 +394,7 @@ batadv_hardif_activate_interface(struct batadv_hard_iface *hard_iface) bat_priv = netdev_priv(hard_iface->soft_iface); - bat_priv->bat_algo_ops->bat_iface_update_mac(hard_iface); + bat_priv->algo_ops->iface.update_mac(hard_iface); hard_iface->if_status = BATADV_IF_TO_BE_ACTIVATED; /* the first active interface becomes our primary interface or @@ -409,8 +409,8 @@ batadv_hardif_activate_interface(struct batadv_hard_iface *hard_iface) batadv_update_min_mtu(hard_iface->soft_iface); - if (bat_priv->bat_algo_ops->bat_iface_activate) - bat_priv->bat_algo_ops->bat_iface_activate(hard_iface); + if (bat_priv->algo_ops->iface.activate) + bat_priv->algo_ops->iface.activate(hard_iface); out: if (primary_if) @@ -508,7 +508,7 @@ int batadv_hardif_enable_interface(struct batadv_hard_iface *hard_iface, if (ret) goto err_dev; - ret = bat_priv->bat_algo_ops->bat_iface_enable(hard_iface); + ret = bat_priv->algo_ops->iface.enable(hard_iface); if (ret < 0) goto err_upper; @@ -517,7 +517,7 @@ int batadv_hardif_enable_interface(struct batadv_hard_iface *hard_iface, hard_iface->if_status = BATADV_IF_INACTIVE; ret = batadv_orig_hash_add_if(hard_iface, bat_priv->num_ifaces); if (ret < 0) { - bat_priv->bat_algo_ops->bat_iface_disable(hard_iface); + bat_priv->algo_ops->iface.disable(hard_iface); bat_priv->num_ifaces--; hard_iface->if_status = BATADV_IF_NOT_IN_USE; goto err_upper; @@ -598,7 +598,7 @@ void batadv_hardif_disable_interface(struct batadv_hard_iface *hard_iface, batadv_hardif_put(new_if); } - bat_priv->bat_algo_ops->bat_iface_disable(hard_iface); + bat_priv->algo_ops->iface.disable(hard_iface); hard_iface->if_status = BATADV_IF_NOT_IN_USE; /* delete all references to this hard_iface */ @@ -783,7 +783,7 @@ static int batadv_hard_if_event(struct notifier_block *this, batadv_check_known_mac_addr(hard_iface->net_dev); bat_priv = netdev_priv(hard_iface->soft_iface); - bat_priv->bat_algo_ops->bat_iface_update_mac(hard_iface); + bat_priv->algo_ops->iface.update_mac(hard_iface); primary_if = batadv_primary_if_get_selected(bat_priv); if (!primary_if) diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c index c25bbb8..231f8ea 100644 --- a/net/batman-adv/netlink.c +++ b/net/batman-adv/netlink.c @@ -89,7 +89,7 @@ batadv_netlink_mesh_info_put(struct sk_buff *msg, struct net_device *soft_iface) if (nla_put_string(msg, BATADV_ATTR_VERSION, BATADV_SOURCE_VERSION) || nla_put_string(msg, BATADV_ATTR_ALGO_NAME, - bat_priv->bat_algo_ops->name) || + bat_priv->algo_ops->name) || nla_put_u32(msg, BATADV_ATTR_MESH_IFINDEX, soft_iface->ifindex) || nla_put_string(msg, BATADV_ATTR_MESH_IFNAME, soft_iface->name) || nla_put(msg, BATADV_ATTR_MESH_ADDRESS, ETH_ALEN, diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index 8ad17ad..7d1e542 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -534,8 +534,8 @@ batadv_hardif_neigh_create(struct batadv_hard_iface *hard_iface, kref_init(&hardif_neigh->refcount); - if (bat_priv->bat_algo_ops->bat_hardif_neigh_init) - bat_priv->bat_algo_ops->bat_hardif_neigh_init(hardif_neigh); + if (bat_priv->algo_ops->neigh.hardif_init) + bat_priv->algo_ops->neigh.hardif_init(hardif_neigh); hlist_add_head(&hardif_neigh->list, &hard_iface->neigh_list); @@ -706,17 +706,17 @@ int batadv_hardif_neigh_seq_print_text(struct seq_file *seq, void *offset) seq_printf(seq, "[B.A.T.M.A.N. adv %s, MainIF/MAC: %s/%pM (%s %s)]\n", BATADV_SOURCE_VERSION, primary_if->net_dev->name, primary_if->net_dev->dev_addr, net_dev->name, - bat_priv->bat_algo_ops->name); + bat_priv->algo_ops->name); batadv_hardif_put(primary_if); - if (!bat_priv->bat_algo_ops->bat_neigh_print) { + if (!bat_priv->algo_ops->neigh.print) { seq_puts(seq, "No printing function for this routing protocol\n"); return 0; } - bat_priv->bat_algo_ops->bat_neigh_print(bat_priv, seq); + bat_priv->algo_ops->neigh.print(bat_priv, seq); return 0; } @@ -767,8 +767,8 @@ static void batadv_orig_node_free_rcu(struct rcu_head *rcu) batadv_frag_purge_orig(orig_node, NULL); - if (orig_node->bat_priv->bat_algo_ops->bat_orig_free) - orig_node->bat_priv->bat_algo_ops->bat_orig_free(orig_node); + if (orig_node->bat_priv->algo_ops->orig.free) + orig_node->bat_priv->algo_ops->orig.free(orig_node); kfree(orig_node->tt_buff); kfree(orig_node); @@ -1097,12 +1097,12 @@ batadv_find_best_neighbor(struct batadv_priv *bat_priv, struct batadv_hard_iface *if_outgoing) { struct batadv_neigh_node *best = NULL, *neigh; - struct batadv_algo_ops *bao = bat_priv->bat_algo_ops; + struct batadv_algo_ops *bao = bat_priv->algo_ops; rcu_read_lock(); hlist_for_each_entry_rcu(neigh, &orig_node->neigh_list, list) { - if (best && (bao->bat_neigh_cmp(neigh, if_outgoing, - best, if_outgoing) <= 0)) + if (best && (bao->neigh.cmp(neigh, if_outgoing, best, + if_outgoing) <= 0)) continue; if (!kref_get_unless_zero(&neigh->refcount)) @@ -1254,18 +1254,17 @@ int batadv_orig_seq_print_text(struct seq_file *seq, void *offset) seq_printf(seq, "[B.A.T.M.A.N. adv %s, MainIF/MAC: %s/%pM (%s %s)]\n", BATADV_SOURCE_VERSION, primary_if->net_dev->name, primary_if->net_dev->dev_addr, net_dev->name, - bat_priv->bat_algo_ops->name); + bat_priv->algo_ops->name); batadv_hardif_put(primary_if); - if (!bat_priv->bat_algo_ops->bat_orig_print) { + if (!bat_priv->algo_ops->orig.print) { seq_puts(seq, "No printing function for this routing protocol\n"); return 0; } - bat_priv->bat_algo_ops->bat_orig_print(bat_priv, seq, - BATADV_IF_DEFAULT); + bat_priv->algo_ops->orig.print(bat_priv, seq, BATADV_IF_DEFAULT); return 0; } @@ -1292,7 +1291,7 @@ int batadv_orig_hardif_seq_print_text(struct seq_file *seq, void *offset) } bat_priv = netdev_priv(hard_iface->soft_iface); - if (!bat_priv->bat_algo_ops->bat_orig_print) { + if (!bat_priv->algo_ops->orig.print) { seq_puts(seq, "No printing function for this routing protocol\n"); goto out; @@ -1306,9 +1305,9 @@ int batadv_orig_hardif_seq_print_text(struct seq_file *seq, void *offset) seq_printf(seq, "[B.A.T.M.A.N. adv %s, IF/MAC: %s/%pM (%s %s)]\n", BATADV_SOURCE_VERSION, hard_iface->net_dev->name, hard_iface->net_dev->dev_addr, - hard_iface->soft_iface->name, bat_priv->bat_algo_ops->name); + hard_iface->soft_iface->name, bat_priv->algo_ops->name); - bat_priv->bat_algo_ops->bat_orig_print(bat_priv, seq, hard_iface); + bat_priv->algo_ops->orig.print(bat_priv, seq, hard_iface); out: if (hard_iface) @@ -1320,7 +1319,7 @@ int batadv_orig_hash_add_if(struct batadv_hard_iface *hard_iface, int max_if_num) { struct batadv_priv *bat_priv = netdev_priv(hard_iface->soft_iface); - struct batadv_algo_ops *bao = bat_priv->bat_algo_ops; + struct batadv_algo_ops *bao = bat_priv->algo_ops; struct batadv_hashtable *hash = bat_priv->orig_hash; struct hlist_head *head; struct batadv_orig_node *orig_node; @@ -1336,9 +1335,8 @@ int batadv_orig_hash_add_if(struct batadv_hard_iface *hard_iface, rcu_read_lock(); hlist_for_each_entry_rcu(orig_node, head, hash_entry) { ret = 0; - if (bao->bat_orig_add_if) - ret = bao->bat_orig_add_if(orig_node, - max_if_num); + if (bao->orig.add_if) + ret = bao->orig.add_if(orig_node, max_if_num); if (ret == -ENOMEM) goto err; } @@ -1360,7 +1358,7 @@ int batadv_orig_hash_del_if(struct batadv_hard_iface *hard_iface, struct hlist_head *head; struct batadv_hard_iface *hard_iface_tmp; struct batadv_orig_node *orig_node; - struct batadv_algo_ops *bao = bat_priv->bat_algo_ops; + struct batadv_algo_ops *bao = bat_priv->algo_ops; u32 i; int ret; @@ -1373,10 +1371,9 @@ int batadv_orig_hash_del_if(struct batadv_hard_iface *hard_iface, rcu_read_lock(); hlist_for_each_entry_rcu(orig_node, head, hash_entry) { ret = 0; - if (bao->bat_orig_del_if) - ret = bao->bat_orig_del_if(orig_node, - max_if_num, - hard_iface->if_num); + if (bao->orig.del_if) + ret = bao->orig.del_if(orig_node, max_if_num, + hard_iface->if_num); if (ret == -ENOMEM) goto err; } diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index 7b5de40..0f8c0dd 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -482,7 +482,7 @@ batadv_find_router(struct batadv_priv *bat_priv, struct batadv_orig_node *orig_node, struct batadv_hard_iface *recv_if) { - struct batadv_algo_ops *bao = bat_priv->bat_algo_ops; + struct batadv_algo_ops *bao = bat_priv->algo_ops; struct batadv_neigh_node *first_candidate_router = NULL; struct batadv_neigh_node *next_candidate_router = NULL; struct batadv_neigh_node *router, *cand_router = NULL; @@ -536,9 +536,9 @@ batadv_find_router(struct batadv_priv *bat_priv, /* alternative candidate should be good enough to be * considered */ - if (!bao->bat_neigh_is_similar_or_better(cand_router, - cand->if_outgoing, - router, recv_if)) + if (!bao->neigh.is_similar_or_better(cand_router, + cand->if_outgoing, router, + recv_if)) goto next; /* don't use the same router twice */ diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c index 1a7942d..fe9ca94 100644 --- a/net/batman-adv/sysfs.c +++ b/net/batman-adv/sysfs.c @@ -412,7 +412,7 @@ static ssize_t batadv_show_bat_algo(struct kobject *kobj, { struct batadv_priv *bat_priv = batadv_kobj_to_batpriv(kobj); - return sprintf(buff, "%s\n", bat_priv->bat_algo_ops->name); + return sprintf(buff, "%s\n", bat_priv->algo_ops->name); } static void batadv_post_gw_reselect(struct net_device *net_dev) diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 48ce788..8bb82a3 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -1546,7 +1546,7 @@ batadv_transtable_best_orig(struct batadv_priv *bat_priv, struct batadv_tt_global_entry *tt_global_entry) { struct batadv_neigh_node *router, *best_router = NULL; - struct batadv_algo_ops *bao = bat_priv->bat_algo_ops; + struct batadv_algo_ops *bao = bat_priv->algo_ops; struct hlist_head *head; struct batadv_tt_orig_list_entry *orig_entry, *best_entry = NULL; @@ -1558,8 +1558,8 @@ batadv_transtable_best_orig(struct batadv_priv *bat_priv, continue; if (best_router && - bao->bat_neigh_cmp(router, BATADV_IF_DEFAULT, - best_router, BATADV_IF_DEFAULT) <= 0) { + bao->neigh.cmp(router, BATADV_IF_DEFAULT, best_router, + BATADV_IF_DEFAULT) <= 0) { batadv_neigh_node_put(router); continue; } diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index a331e3a..4d6a7ce 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -1016,7 +1016,7 @@ struct batadv_priv_bat_v { * @cleanup_work: work queue callback item for soft-interface deinit * @primary_if: one of the hard-interfaces assigned to this mesh interface * becomes the primary interface - * @bat_algo_ops: routing algorithm used by this mesh interface + * @algo_ops: routing algorithm used by this mesh interface * @softif_vlan_list: a list of softif_vlan structs, one per VLAN created on top * of the mesh interface represented by this object * @softif_vlan_list_lock: lock protecting softif_vlan_list @@ -1074,7 +1074,7 @@ struct batadv_priv { struct delayed_work orig_work; struct work_struct cleanup_work; struct batadv_hard_iface __rcu *primary_if; /* rcu protected pointer */ - struct batadv_algo_ops *bat_algo_ops; + struct batadv_algo_ops *algo_ops; struct hlist_head softif_vlan_list; spinlock_t softif_vlan_list_lock; /* protects softif_vlan_list */ #ifdef CONFIG_BATMAN_ADV_BLA @@ -1388,59 +1388,77 @@ struct batadv_forw_packet { }; /** + * struct batadv_algo_iface_ops - mesh algorithm callbacks (interface specific) + * @activate: start routing mechanisms when hard-interface is brought up + * @enable: init routing info when hard-interface is enabled + * @disable: de-init routing info when hard-interface is disabled + * @update_mac: (re-)init mac addresses of the protocol information + * belonging to this hard-interface + * @primary_set: called when primary interface is selected / changed + */ +struct batadv_algo_iface_ops { + void (*activate)(struct batadv_hard_iface *hard_iface); + int (*enable)(struct batadv_hard_iface *hard_iface); + void (*disable)(struct batadv_hard_iface *hard_iface); + void (*update_mac)(struct batadv_hard_iface *hard_iface); + void (*primary_set)(struct batadv_hard_iface *hard_iface); +}; + +/** + * struct batadv_algo_neigh_ops - mesh algorithm callbacks (neighbour specific) + * @hardif_init: called on creation of single hop entry + * @cmp: compare the metrics of two neighbors for their respective outgoing + * interfaces + * @is_similar_or_better: check if neigh1 is equally similar or better than + * neigh2 for their respective outgoing interface from the metric prospective + * @print: print the single hop neighbor list (optional) + */ +struct batadv_algo_neigh_ops { + void (*hardif_init)(struct batadv_hardif_neigh_node *neigh); + int (*cmp)(struct batadv_neigh_node *neigh1, + struct batadv_hard_iface *if_outgoing1, + struct batadv_neigh_node *neigh2, + struct batadv_hard_iface *if_outgoing2); + bool (*is_similar_or_better)(struct batadv_neigh_node *neigh1, + struct batadv_hard_iface *if_outgoing1, + struct batadv_neigh_node *neigh2, + struct batadv_hard_iface *if_outgoing2); + void (*print)(struct batadv_priv *priv, struct seq_file *seq); +}; + +/** + * struct batadv_algo_orig_ops - mesh algorithm callbacks (originator specific) + * @free: free the resources allocated by the routing algorithm for an orig_node + * object + * @add_if: ask the routing algorithm to apply the needed changes to the + * orig_node due to a new hard-interface being added into the mesh + * @del_if: ask the routing algorithm to apply the needed changes to the + * orig_node due to an hard-interface being removed from the mesh + * @print: print the originator table (optional) + */ +struct batadv_algo_orig_ops { + void (*free)(struct batadv_orig_node *orig_node); + int (*add_if)(struct batadv_orig_node *orig_node, int max_if_num); + int (*del_if)(struct batadv_orig_node *orig_node, int max_if_num, + int del_if_num); + void (*print)(struct batadv_priv *priv, struct seq_file *seq, + struct batadv_hard_iface *hard_iface); +}; + +/** * struct batadv_algo_ops - mesh algorithm callbacks * @list: list node for the batadv_algo_list * @name: name of the algorithm - * @bat_iface_activate: start routing mechanisms when hard-interface is brought - * up - * @bat_iface_enable: init routing info when hard-interface is enabled - * @bat_iface_disable: de-init routing info when hard-interface is disabled - * @bat_iface_update_mac: (re-)init mac addresses of the protocol information - * belonging to this hard-interface - * @bat_primary_iface_set: called when primary interface is selected / changed - * @bat_hardif_neigh_init: called on creation of single hop entry - * @bat_neigh_cmp: compare the metrics of two neighbors for their respective - * outgoing interfaces - * @bat_neigh_is_similar_or_better: check if neigh1 is equally similar or - * better than neigh2 for their respective outgoing interface from the metric - * prospective - * @bat_neigh_print: print the single hop neighbor list (optional) - * @bat_orig_print: print the originator table (optional) - * @bat_orig_free: free the resources allocated by the routing algorithm for an - * orig_node object - * @bat_orig_add_if: ask the routing algorithm to apply the needed changes to - * the orig_node due to a new hard-interface being added into the mesh - * @bat_orig_del_if: ask the routing algorithm to apply the needed changes to - * the orig_node due to an hard-interface being removed from the mesh + * @iface: callbacks related to interface handling + * @neigh: callbacks related to neighbors handling + * @orig: callbacks related to originators handling */ struct batadv_algo_ops { struct hlist_node list; char *name; - void (*bat_iface_activate)(struct batadv_hard_iface *hard_iface); - int (*bat_iface_enable)(struct batadv_hard_iface *hard_iface); - void (*bat_iface_disable)(struct batadv_hard_iface *hard_iface); - void (*bat_iface_update_mac)(struct batadv_hard_iface *hard_iface); - void (*bat_primary_iface_set)(struct batadv_hard_iface *hard_iface); - /* neigh_node handling API */ - void (*bat_hardif_neigh_init)(struct batadv_hardif_neigh_node *neigh); - int (*bat_neigh_cmp)(struct batadv_neigh_node *neigh1, - struct batadv_hard_iface *if_outgoing1, - struct batadv_neigh_node *neigh2, - struct batadv_hard_iface *if_outgoing2); - bool (*bat_neigh_is_similar_or_better) - (struct batadv_neigh_node *neigh1, - struct batadv_hard_iface *if_outgoing1, - struct batadv_neigh_node *neigh2, - struct batadv_hard_iface *if_outgoing2); - void (*bat_neigh_print)(struct batadv_priv *priv, struct seq_file *seq); - /* orig_node handling API */ - void (*bat_orig_print)(struct batadv_priv *priv, struct seq_file *seq, - struct batadv_hard_iface *hard_iface); - void (*bat_orig_free)(struct batadv_orig_node *orig_node); - int (*bat_orig_add_if)(struct batadv_orig_node *orig_node, - int max_if_num); - int (*bat_orig_del_if)(struct batadv_orig_node *orig_node, - int max_if_num, int del_if_num); + struct batadv_algo_iface_ops iface; + struct batadv_algo_neigh_ops neigh; + struct batadv_algo_orig_ops orig; }; /** -- cgit v0.10.2 From 8b10cab64c134ffbffac96edd1899d303d3afcac Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sat, 2 Jul 2016 06:43:14 -0400 Subject: net: simplify and make pkt_type_ok() available for other users Suggested-by: Daniel Borkmann Signed-off-by: Jamal Hadi Salim Acked-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index dc0fca7..638b0e0 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -37,6 +37,7 @@ #include #include #include +#include #include /* The interface for checksum offload between the stack and networking drivers @@ -881,6 +882,15 @@ static inline struct rtable *skb_rtable(const struct sk_buff *skb) return (struct rtable *)skb_dst(skb); } +/* For mangling skb->pkt_type from user space side from applications + * such as nft, tc, etc, we only allow a conservative subset of + * possible pkt_types to be set. +*/ +static inline bool skb_pkt_type_ok(u32 ptype) +{ + return ptype <= PACKET_OTHERHOST; +} + void kfree_skb(struct sk_buff *skb); void kfree_skb_list(struct sk_buff *segs); void skb_tx_error(struct sk_buff *skb); diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c index 16c50b0..03e5e33 100644 --- a/net/netfilter/nft_meta.c +++ b/net/netfilter/nft_meta.c @@ -199,13 +199,6 @@ err: } EXPORT_SYMBOL_GPL(nft_meta_get_eval); -/* don't change or set _LOOPBACK, _USER, etc. */ -static bool pkt_type_ok(u32 p) -{ - return p == PACKET_HOST || p == PACKET_BROADCAST || - p == PACKET_MULTICAST || p == PACKET_OTHERHOST; -} - void nft_meta_set_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) @@ -223,7 +216,7 @@ void nft_meta_set_eval(const struct nft_expr *expr, break; case NFT_META_PKTTYPE: if (skb->pkt_type != value && - pkt_type_ok(value) && pkt_type_ok(skb->pkt_type)) + skb_pkt_type_ok(value) && skb_pkt_type_ok(skb->pkt_type)) skb->pkt_type = value; break; case NFT_META_NFTRACE: -- cgit v0.10.2 From ff202ee1ed8f032f05b80b541664cf02e75d7080 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sat, 2 Jul 2016 06:43:15 -0400 Subject: net sched actions: skbedit add support for mod-ing skb pkt_type Extremely useful for setting packet type to host so i dont have to modify the dst mac address using pedit (which requires that i know the mac address) Example usage: tc filter add dev eth0 parent ffff: protocol ip pref 9 u32 \ match ip src 5.5.5.5/32 \ flowid 1:5 action skbedit ptype host This will tag all packets incoming from 5.5.5.5 with type PACKET_HOST Signed-off-by: Jamal Hadi Salim Acked-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/include/net/tc_act/tc_skbedit.h b/include/net/tc_act/tc_skbedit.h index b496d5a..d01a5d4 100644 --- a/include/net/tc_act/tc_skbedit.h +++ b/include/net/tc_act/tc_skbedit.h @@ -24,11 +24,11 @@ struct tcf_skbedit { struct tcf_common common; - u32 flags; - u32 priority; - u32 mark; - u16 queue_mapping; - /* XXX: 16-bit pad here? */ + u32 flags; + u32 priority; + u32 mark; + u16 queue_mapping; + u16 ptype; }; #define to_skbedit(a) \ container_of(a->priv, struct tcf_skbedit, common) diff --git a/include/uapi/linux/tc_act/tc_skbedit.h b/include/uapi/linux/tc_act/tc_skbedit.h index fecb5cc..a4d00c6 100644 --- a/include/uapi/linux/tc_act/tc_skbedit.h +++ b/include/uapi/linux/tc_act/tc_skbedit.h @@ -27,6 +27,7 @@ #define SKBEDIT_F_PRIORITY 0x1 #define SKBEDIT_F_QUEUE_MAPPING 0x2 #define SKBEDIT_F_MARK 0x4 +#define SKBEDIT_F_PTYPE 0x8 struct tc_skbedit { tc_gen; @@ -40,6 +41,7 @@ enum { TCA_SKBEDIT_QUEUE_MAPPING, TCA_SKBEDIT_MARK, TCA_SKBEDIT_PAD, + TCA_SKBEDIT_PTYPE, __TCA_SKBEDIT_MAX }; #define TCA_SKBEDIT_MAX (__TCA_SKBEDIT_MAX - 1) diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index 53d1486..1c4c924 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -47,6 +47,8 @@ static int tcf_skbedit(struct sk_buff *skb, const struct tc_action *a, skb_set_queue_mapping(skb, d->queue_mapping); if (d->flags & SKBEDIT_F_MARK) skb->mark = d->mark; + if (d->flags & SKBEDIT_F_PTYPE) + skb->pkt_type = d->ptype; spin_unlock(&d->tcf_lock); return d->tcf_action; @@ -57,6 +59,7 @@ static const struct nla_policy skbedit_policy[TCA_SKBEDIT_MAX + 1] = { [TCA_SKBEDIT_PRIORITY] = { .len = sizeof(u32) }, [TCA_SKBEDIT_QUEUE_MAPPING] = { .len = sizeof(u16) }, [TCA_SKBEDIT_MARK] = { .len = sizeof(u32) }, + [TCA_SKBEDIT_PTYPE] = { .len = sizeof(u16) }, }; static int tcf_skbedit_init(struct net *net, struct nlattr *nla, @@ -68,7 +71,7 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, struct tc_skbedit *parm; struct tcf_skbedit *d; u32 flags = 0, *priority = NULL, *mark = NULL; - u16 *queue_mapping = NULL; + u16 *queue_mapping = NULL, *ptype = NULL; bool exists = false; int ret = 0, err; @@ -92,6 +95,13 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, queue_mapping = nla_data(tb[TCA_SKBEDIT_QUEUE_MAPPING]); } + if (tb[TCA_SKBEDIT_PTYPE] != NULL) { + ptype = nla_data(tb[TCA_SKBEDIT_PTYPE]); + if (!skb_pkt_type_ok(*ptype)) + return -EINVAL; + flags |= SKBEDIT_F_PTYPE; + } + if (tb[TCA_SKBEDIT_MARK] != NULL) { flags |= SKBEDIT_F_MARK; mark = nla_data(tb[TCA_SKBEDIT_MARK]); @@ -132,6 +142,8 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, d->queue_mapping = *queue_mapping; if (flags & SKBEDIT_F_MARK) d->mark = *mark; + if (flags & SKBEDIT_F_PTYPE) + d->ptype = *ptype; d->tcf_action = parm->action; @@ -169,6 +181,10 @@ static int tcf_skbedit_dump(struct sk_buff *skb, struct tc_action *a, nla_put(skb, TCA_SKBEDIT_MARK, sizeof(d->mark), &d->mark)) goto nla_put_failure; + if ((d->flags & SKBEDIT_F_PTYPE) && + nla_put(skb, TCA_SKBEDIT_PTYPE, sizeof(d->ptype), + &d->ptype)) + goto nla_put_failure; tcf_tm_dump(&t, &d->tcf_tm); if (nla_put_64bit(skb, TCA_SKBEDIT_TM, sizeof(t), &t, TCA_SKBEDIT_PAD)) -- cgit v0.10.2 From 61cc535de36838bf4cfe08c8c4eeaad1ca4a89b1 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sat, 2 Jul 2016 06:43:16 -0400 Subject: net sched actions: skbedit convert to use more modern nla_put_xxx Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index 1c4c924..8e573c0 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -170,20 +170,16 @@ static int tcf_skbedit_dump(struct sk_buff *skb, struct tc_action *a, if (nla_put(skb, TCA_SKBEDIT_PARMS, sizeof(opt), &opt)) goto nla_put_failure; if ((d->flags & SKBEDIT_F_PRIORITY) && - nla_put(skb, TCA_SKBEDIT_PRIORITY, sizeof(d->priority), - &d->priority)) + nla_put_u32(skb, TCA_SKBEDIT_PRIORITY, d->priority)) goto nla_put_failure; if ((d->flags & SKBEDIT_F_QUEUE_MAPPING) && - nla_put(skb, TCA_SKBEDIT_QUEUE_MAPPING, - sizeof(d->queue_mapping), &d->queue_mapping)) + nla_put_u16(skb, TCA_SKBEDIT_QUEUE_MAPPING, d->queue_mapping)) goto nla_put_failure; if ((d->flags & SKBEDIT_F_MARK) && - nla_put(skb, TCA_SKBEDIT_MARK, sizeof(d->mark), - &d->mark)) + nla_put_u32(skb, TCA_SKBEDIT_MARK, d->mark)) goto nla_put_failure; if ((d->flags & SKBEDIT_F_PTYPE) && - nla_put(skb, TCA_SKBEDIT_PTYPE, sizeof(d->ptype), - &d->ptype)) + nla_put_u16(skb, TCA_SKBEDIT_PTYPE, d->ptype)) goto nla_put_failure; tcf_tm_dump(&t, &d->tcf_tm); -- cgit v0.10.2 From 5d4de16c6db6ecc51a549bfbf7ca1c11c82a4240 Mon Sep 17 00:00:00 2001 From: Christophe Jaillet Date: Sat, 2 Jul 2016 14:31:05 +0200 Subject: net/mlx4: Fix some indent inconsistancy Silent a few smatch warnings about indentation. This include the removal of a 'return' statement in 'resource_tracker.c'. This 'return' will still be performed when breaking out of the corresponding 'switch' block. Signed-off-by: Christophe JAILLET Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/intf.c b/drivers/net/ethernet/mellanox/mlx4/intf.c index dec77d6..7ae1cda 100644 --- a/drivers/net/ethernet/mellanox/mlx4/intf.c +++ b/drivers/net/ethernet/mellanox/mlx4/intf.c @@ -147,7 +147,7 @@ int mlx4_do_bond(struct mlx4_dev *dev, bool enable) if (enable) { dev->flags |= MLX4_FLAG_BONDED; } else { - ret = mlx4_virt2phy_port_map(dev, 1, 2); + ret = mlx4_virt2phy_port_map(dev, 1, 2); if (ret) { mlx4_err(dev, "Fail to reset port map\n"); return ret; diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index b673a5f..75dd2e3 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -2600,7 +2600,7 @@ static int mlx4_setup_hca(struct mlx4_dev *dev) err = mlx4_init_uar_table(dev); if (err) { mlx4_err(dev, "Failed to initialize user access region table, aborting\n"); - return err; + return err; } err = mlx4_uar_alloc(dev, &priv->driver_uar); diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c index f2d0920..94b891c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mcg.c +++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c @@ -618,8 +618,8 @@ static int remove_promisc_qp(struct mlx4_dev *dev, u8 port, err = mlx4_READ_ENTRY(dev, entry->index, mailbox); - if (err) - goto out_mailbox; + if (err) + goto out_mailbox; members_count = be32_to_cpu(mgm->members_count) & 0xffffff; @@ -657,8 +657,8 @@ static int remove_promisc_qp(struct mlx4_dev *dev, u8 port, err = mlx4_WRITE_ENTRY(dev, entry->index, mailbox); - if (err) - goto out_mailbox; + if (err) + goto out_mailbox; } } } diff --git a/drivers/net/ethernet/mellanox/mlx4/mr.c b/drivers/net/ethernet/mellanox/mlx4/mr.c index 9319519..395b546 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mr.c +++ b/drivers/net/ethernet/mellanox/mlx4/mr.c @@ -248,7 +248,7 @@ static void mlx4_free_mtt_range(struct mlx4_dev *dev, u32 offset, int order) offset, order); return; } - __mlx4_free_mtt_range(dev, offset, order); + __mlx4_free_mtt_range(dev, offset, order); } void mlx4_mtt_cleanup(struct mlx4_dev *dev, struct mlx4_mtt *mtt) diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c index cd9b2b2..8b81114 100644 --- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c +++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c @@ -2372,16 +2372,15 @@ static int mpt_free_res(struct mlx4_dev *dev, int slave, int op, int cmd, __mlx4_mpt_release(dev, index); break; case RES_OP_MAP_ICM: - index = get_param_l(&in_param); - id = index & mpt_mask(dev); - err = mr_res_start_move_to(dev, slave, id, - RES_MPT_RESERVED, &mpt); - if (err) - return err; - - __mlx4_mpt_free_icm(dev, mpt->key); - res_end_move(dev, slave, RES_MPT, id); + index = get_param_l(&in_param); + id = index & mpt_mask(dev); + err = mr_res_start_move_to(dev, slave, id, + RES_MPT_RESERVED, &mpt); + if (err) return err; + + __mlx4_mpt_free_icm(dev, mpt->key); + res_end_move(dev, slave, RES_MPT, id); break; default: err = -EINVAL; @@ -4253,9 +4252,8 @@ int mlx4_UPDATE_QP_wrapper(struct mlx4_dev *dev, int slave, (1ULL << MLX4_UPD_QP_PATH_MASK_ETH_SRC_CHECK_MC_LB)) && !(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_UPDATE_QP_SRC_CHECK_LB)) { - mlx4_warn(dev, - "Src check LB for slave %d isn't supported\n", - slave); + mlx4_warn(dev, "Src check LB for slave %d isn't supported\n", + slave); return -ENOTSUPP; } -- cgit v0.10.2 From 9e8e6e880d9346508d1eb6a632bb9e0048cc0bea Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 2 Jul 2016 23:36:59 +0200 Subject: net: ethernet: cavium: octeon: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c index 388cd79..de678e6 100644 --- a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c +++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c @@ -146,7 +146,6 @@ struct octeon_mgmt { struct device *dev; struct napi_struct napi; struct tasklet_struct tx_clean_tasklet; - struct phy_device *phydev; struct device_node *phy_np; resource_size_t mix_phys; resource_size_t mix_size; @@ -787,14 +786,12 @@ static int octeon_mgmt_ioctl_hwtstamp(struct net_device *netdev, static int octeon_mgmt_ioctl(struct net_device *netdev, struct ifreq *rq, int cmd) { - struct octeon_mgmt *p = netdev_priv(netdev); - switch (cmd) { case SIOCSHWTSTAMP: return octeon_mgmt_ioctl_hwtstamp(netdev, rq, cmd); default: - if (p->phydev) - return phy_mii_ioctl(p->phydev, rq, cmd); + if (netdev->phydev) + return phy_mii_ioctl(netdev->phydev, rq, cmd); return -EINVAL; } } @@ -836,16 +833,18 @@ static void octeon_mgmt_enable_link(struct octeon_mgmt *p) static void octeon_mgmt_update_link(struct octeon_mgmt *p) { + struct net_device *ndev = p->netdev; + struct phy_device *phydev = ndev->phydev; union cvmx_agl_gmx_prtx_cfg prtx_cfg; prtx_cfg.u64 = cvmx_read_csr(p->agl + AGL_GMX_PRT_CFG); - if (!p->phydev->link) + if (!phydev->link) prtx_cfg.s.duplex = 1; else - prtx_cfg.s.duplex = p->phydev->duplex; + prtx_cfg.s.duplex = phydev->duplex; - switch (p->phydev->speed) { + switch (phydev->speed) { case 10: prtx_cfg.s.speed = 0; prtx_cfg.s.slottime = 0; @@ -871,7 +870,7 @@ static void octeon_mgmt_update_link(struct octeon_mgmt *p) prtx_cfg.s.speed_msb = 0; /* Only matters for half-duplex */ prtx_cfg.s.slottime = 1; - prtx_cfg.s.burst = p->phydev->duplex; + prtx_cfg.s.burst = phydev->duplex; } break; case 0: /* No link */ @@ -894,9 +893,9 @@ static void octeon_mgmt_update_link(struct octeon_mgmt *p) /* MII (both speeds) and RGMII 1000 speed. */ agl_clk.s.clk_cnt = 1; if (prtx_ctl.s.mode == 0) { /* RGMII mode */ - if (p->phydev->speed == 10) + if (phydev->speed == 10) agl_clk.s.clk_cnt = 50; - else if (p->phydev->speed == 100) + else if (phydev->speed == 100) agl_clk.s.clk_cnt = 5; } cvmx_write_csr(p->agl + AGL_GMX_TX_CLK, agl_clk.u64); @@ -906,39 +905,40 @@ static void octeon_mgmt_update_link(struct octeon_mgmt *p) static void octeon_mgmt_adjust_link(struct net_device *netdev) { struct octeon_mgmt *p = netdev_priv(netdev); + struct phy_device *phydev = netdev->phydev; unsigned long flags; int link_changed = 0; - if (!p->phydev) + if (!phydev) return; spin_lock_irqsave(&p->lock, flags); - if (!p->phydev->link && p->last_link) + if (!phydev->link && p->last_link) link_changed = -1; - if (p->phydev->link - && (p->last_duplex != p->phydev->duplex - || p->last_link != p->phydev->link - || p->last_speed != p->phydev->speed)) { + if (phydev->link && + (p->last_duplex != phydev->duplex || + p->last_link != phydev->link || + p->last_speed != phydev->speed)) { octeon_mgmt_disable_link(p); link_changed = 1; octeon_mgmt_update_link(p); octeon_mgmt_enable_link(p); } - p->last_link = p->phydev->link; - p->last_speed = p->phydev->speed; - p->last_duplex = p->phydev->duplex; + p->last_link = phydev->link; + p->last_speed = phydev->speed; + p->last_duplex = phydev->duplex; spin_unlock_irqrestore(&p->lock, flags); if (link_changed != 0) { if (link_changed > 0) { pr_info("%s: Link is up - %d/%s\n", netdev->name, - p->phydev->speed, - DUPLEX_FULL == p->phydev->duplex ? + phydev->speed, + phydev->duplex == DUPLEX_FULL ? "Full" : "Half"); } else { pr_info("%s: Link is down\n", netdev->name); @@ -949,6 +949,7 @@ static void octeon_mgmt_adjust_link(struct net_device *netdev) static int octeon_mgmt_init_phy(struct net_device *netdev) { struct octeon_mgmt *p = netdev_priv(netdev); + struct phy_device *phydev = NULL; if (octeon_is_simulation() || p->phy_np == NULL) { /* No PHYs in the simulator. */ @@ -956,11 +957,11 @@ static int octeon_mgmt_init_phy(struct net_device *netdev) return 0; } - p->phydev = of_phy_connect(netdev, p->phy_np, - octeon_mgmt_adjust_link, 0, - PHY_INTERFACE_MODE_MII); + phydev = of_phy_connect(netdev, p->phy_np, + octeon_mgmt_adjust_link, 0, + PHY_INTERFACE_MODE_MII); - if (!p->phydev) + if (!phydev) return -ENODEV; return 0; @@ -1080,9 +1081,9 @@ static int octeon_mgmt_open(struct net_device *netdev) } /* Set the mode of the interface, RGMII/MII. */ - if (OCTEON_IS_MODEL(OCTEON_CN6XXX) && p->phydev) { + if (OCTEON_IS_MODEL(OCTEON_CN6XXX) && netdev->phydev) { union cvmx_agl_prtx_ctl agl_prtx_ctl; - int rgmii_mode = (p->phydev->supported & + int rgmii_mode = (netdev->phydev->supported & (SUPPORTED_1000baseT_Half | SUPPORTED_1000baseT_Full)) != 0; agl_prtx_ctl.u64 = cvmx_read_csr(p->agl_prt_ctl); @@ -1205,7 +1206,7 @@ static int octeon_mgmt_open(struct net_device *netdev) /* Configure the port duplex, speed and enables */ octeon_mgmt_disable_link(p); - if (p->phydev) + if (netdev->phydev) octeon_mgmt_update_link(p); octeon_mgmt_enable_link(p); @@ -1214,9 +1215,9 @@ static int octeon_mgmt_open(struct net_device *netdev) /* PHY is not present in simulator. The carrier is enabled * while initializing the phy for simulator, leave it enabled. */ - if (p->phydev) { + if (netdev->phydev) { netif_carrier_off(netdev); - phy_start_aneg(p->phydev); + phy_start_aneg(netdev->phydev); } netif_wake_queue(netdev); @@ -1244,9 +1245,8 @@ static int octeon_mgmt_stop(struct net_device *netdev) napi_disable(&p->napi); netif_stop_queue(netdev); - if (p->phydev) - phy_disconnect(p->phydev); - p->phydev = NULL; + if (netdev->phydev) + phy_disconnect(netdev->phydev); netif_carrier_off(netdev); @@ -1349,10 +1349,8 @@ static void octeon_mgmt_get_drvinfo(struct net_device *netdev, static int octeon_mgmt_get_settings(struct net_device *netdev, struct ethtool_cmd *cmd) { - struct octeon_mgmt *p = netdev_priv(netdev); - - if (p->phydev) - return phy_ethtool_gset(p->phydev, cmd); + if (netdev->phydev) + return phy_ethtool_gset(netdev->phydev, cmd); return -EOPNOTSUPP; } @@ -1360,26 +1358,22 @@ static int octeon_mgmt_get_settings(struct net_device *netdev, static int octeon_mgmt_set_settings(struct net_device *netdev, struct ethtool_cmd *cmd) { - struct octeon_mgmt *p = netdev_priv(netdev); - if (!capable(CAP_NET_ADMIN)) return -EPERM; - if (p->phydev) - return phy_ethtool_sset(p->phydev, cmd); + if (netdev->phydev) + return phy_ethtool_sset(netdev->phydev, cmd); return -EOPNOTSUPP; } static int octeon_mgmt_nway_reset(struct net_device *dev) { - struct octeon_mgmt *p = netdev_priv(dev); - if (!capable(CAP_NET_ADMIN)) return -EPERM; - if (p->phydev) - return phy_start_aneg(p->phydev); + if (dev->phydev) + return phy_start_aneg(dev->phydev); return -EOPNOTSUPP; } -- cgit v0.10.2 From f4400ded03b46a8288d6d5ca4bfb6bb29dbcbb94 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 2 Jul 2016 23:37:00 +0200 Subject: net: ethernet: cavium: octeon: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c index de678e6..e8bc15b 100644 --- a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c +++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c @@ -1346,27 +1346,6 @@ static void octeon_mgmt_get_drvinfo(struct net_device *netdev, strlcpy(info->bus_info, "N/A", sizeof(info->bus_info)); } -static int octeon_mgmt_get_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) -{ - if (netdev->phydev) - return phy_ethtool_gset(netdev->phydev, cmd); - - return -EOPNOTSUPP; -} - -static int octeon_mgmt_set_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) -{ - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - if (netdev->phydev) - return phy_ethtool_sset(netdev->phydev, cmd); - - return -EOPNOTSUPP; -} - static int octeon_mgmt_nway_reset(struct net_device *dev) { if (!capable(CAP_NET_ADMIN)) @@ -1380,10 +1359,10 @@ static int octeon_mgmt_nway_reset(struct net_device *dev) static const struct ethtool_ops octeon_mgmt_ethtool_ops = { .get_drvinfo = octeon_mgmt_get_drvinfo, - .get_settings = octeon_mgmt_get_settings, - .set_settings = octeon_mgmt_set_settings, .nway_reset = octeon_mgmt_nway_reset, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops octeon_mgmt_ops = { -- cgit v0.10.2 From d1e3a356f569adefde26c8f5d514cd108617bcff Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 3 Jul 2016 00:05:04 +0200 Subject: net: ethernet: lantiq_etop: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c index 0d2f8e9..ad715a4 100644 --- a/drivers/net/ethernet/lantiq_etop.c +++ b/drivers/net/ethernet/lantiq_etop.c @@ -102,7 +102,6 @@ struct ltq_etop_priv { struct resource *res; struct mii_bus *mii_bus; - struct phy_device *phydev; struct ltq_etop_chan ch[MAX_DMA_CHAN]; int tx_free[MAX_DMA_CHAN >> 1]; @@ -307,25 +306,19 @@ ltq_etop_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) static int ltq_etop_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct ltq_etop_priv *priv = netdev_priv(dev); - - return phy_ethtool_gset(priv->phydev, cmd); + return phy_ethtool_gset(dev->phydev, cmd); } static int ltq_etop_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct ltq_etop_priv *priv = netdev_priv(dev); - - return phy_ethtool_sset(priv->phydev, cmd); + return phy_ethtool_sset(dev->phydev, cmd); } static int ltq_etop_nway_reset(struct net_device *dev) { - struct ltq_etop_priv *priv = netdev_priv(dev); - - return phy_start_aneg(priv->phydev); + return phy_start_aneg(dev->phydev); } static const struct ethtool_ops ltq_etop_ethtool_ops = { @@ -401,7 +394,6 @@ ltq_etop_mdio_probe(struct net_device *dev) | SUPPORTED_TP); phydev->advertising = phydev->supported; - priv->phydev = phydev; phy_attached_info(phydev); return 0; @@ -450,7 +442,7 @@ ltq_etop_mdio_cleanup(struct net_device *dev) { struct ltq_etop_priv *priv = netdev_priv(dev); - phy_disconnect(priv->phydev); + phy_disconnect(dev->phydev); mdiobus_unregister(priv->mii_bus); mdiobus_free(priv->mii_bus); } @@ -469,7 +461,7 @@ ltq_etop_open(struct net_device *dev) ltq_dma_open(&ch->dma); napi_enable(&ch->napi); } - phy_start(priv->phydev); + phy_start(dev->phydev); netif_tx_start_all_queues(dev); return 0; } @@ -481,7 +473,7 @@ ltq_etop_stop(struct net_device *dev) int i; netif_tx_stop_all_queues(dev); - phy_stop(priv->phydev); + phy_stop(dev->phydev); for (i = 0; i < MAX_DMA_CHAN; i++) { struct ltq_etop_chan *ch = &priv->ch[i]; @@ -556,10 +548,8 @@ ltq_etop_change_mtu(struct net_device *dev, int new_mtu) static int ltq_etop_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct ltq_etop_priv *priv = netdev_priv(dev); - /* TODO: mii-toll reports "No MII transceiver present!." ?!*/ - return phy_mii_ioctl(priv->phydev, rq, cmd); + return phy_mii_ioctl(dev->phydev, rq, cmd); } static int -- cgit v0.10.2 From 5376d95fa2eccbe1d59fe91a3fb2b4c9a6c1c004 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 3 Jul 2016 00:05:05 +0200 Subject: net: ethernet: lantiq_etop: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c index ad715a4..91e09d6 100644 --- a/drivers/net/ethernet/lantiq_etop.c +++ b/drivers/net/ethernet/lantiq_etop.c @@ -304,18 +304,6 @@ ltq_etop_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) } static int -ltq_etop_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - return phy_ethtool_gset(dev->phydev, cmd); -} - -static int -ltq_etop_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - return phy_ethtool_sset(dev->phydev, cmd); -} - -static int ltq_etop_nway_reset(struct net_device *dev) { return phy_start_aneg(dev->phydev); @@ -323,9 +311,9 @@ ltq_etop_nway_reset(struct net_device *dev) static const struct ethtool_ops ltq_etop_ethtool_ops = { .get_drvinfo = ltq_etop_get_drvinfo, - .get_settings = ltq_etop_get_settings, - .set_settings = ltq_etop_set_settings, .nway_reset = ltq_etop_nway_reset, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int -- cgit v0.10.2 From f788e322b730e42419ea2090fa3fdd99abcdc6fd Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 3 Jul 2016 01:14:20 +0200 Subject: net: ethernet: smsc: smsc911x: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index b5ab5e1..6fded67 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -114,7 +114,6 @@ struct smsc911x_data { /* spinlock to ensure register accesses are serialised */ spinlock_t dev_lock; - struct phy_device *phy_dev; struct mii_bus *mii_bus; unsigned int using_extphy; int last_duplex; @@ -833,7 +832,7 @@ static int smsc911x_phy_reset(struct smsc911x_data *pdata) static int smsc911x_phy_loopbacktest(struct net_device *dev) { struct smsc911x_data *pdata = netdev_priv(dev); - struct phy_device *phy_dev = pdata->phy_dev; + struct phy_device *phy_dev = dev->phydev; int result = -EIO; unsigned int i, val; unsigned long flags; @@ -903,7 +902,8 @@ static int smsc911x_phy_loopbacktest(struct net_device *dev) static void smsc911x_phy_update_flowcontrol(struct smsc911x_data *pdata) { - struct phy_device *phy_dev = pdata->phy_dev; + struct net_device *ndev = pdata->dev; + struct phy_device *phy_dev = ndev->phydev; u32 afc = smsc911x_reg_read(pdata, AFC_CFG); u32 flow; unsigned long flags; @@ -944,7 +944,7 @@ static void smsc911x_phy_update_flowcontrol(struct smsc911x_data *pdata) static void smsc911x_phy_adjust_link(struct net_device *dev) { struct smsc911x_data *pdata = netdev_priv(dev); - struct phy_device *phy_dev = pdata->phy_dev; + struct phy_device *phy_dev = dev->phydev; unsigned long flags; int carrier; @@ -1037,7 +1037,6 @@ static int smsc911x_mii_probe(struct net_device *dev) SUPPORTED_Asym_Pause); phydev->advertising = phydev->supported; - pdata->phy_dev = phydev; pdata->last_duplex = -1; pdata->last_carrier = -1; @@ -1338,9 +1337,11 @@ static void smsc911x_rx_multicast_update_workaround(struct smsc911x_data *pdata) static int smsc911x_phy_general_power_up(struct smsc911x_data *pdata) { + struct net_device *ndev = pdata->dev; + struct phy_device *phy_dev = ndev->phydev; int rc = 0; - if (!pdata->phy_dev) + if (!phy_dev) return rc; /* If the internal PHY is in General Power-Down mode, all, except the @@ -1350,7 +1351,7 @@ static int smsc911x_phy_general_power_up(struct smsc911x_data *pdata) * In that case, clear the bit 0.11, so the PHY powers up and we can * access to the phy registers. */ - rc = phy_read(pdata->phy_dev, MII_BMCR); + rc = phy_read(phy_dev, MII_BMCR); if (rc < 0) { SMSC_WARN(pdata, drv, "Failed reading PHY control reg"); return rc; @@ -1360,7 +1361,7 @@ static int smsc911x_phy_general_power_up(struct smsc911x_data *pdata) * disable the general power down-mode. */ if (rc & BMCR_PDOWN) { - rc = phy_write(pdata->phy_dev, MII_BMCR, rc & ~BMCR_PDOWN); + rc = phy_write(phy_dev, MII_BMCR, rc & ~BMCR_PDOWN); if (rc < 0) { SMSC_WARN(pdata, drv, "Failed writing PHY control reg"); return rc; @@ -1374,12 +1375,14 @@ static int smsc911x_phy_general_power_up(struct smsc911x_data *pdata) static int smsc911x_phy_disable_energy_detect(struct smsc911x_data *pdata) { + struct net_device *ndev = pdata->dev; + struct phy_device *phy_dev = ndev->phydev; int rc = 0; - if (!pdata->phy_dev) + if (!phy_dev) return rc; - rc = phy_read(pdata->phy_dev, MII_LAN83C185_CTRL_STATUS); + rc = phy_read(phy_dev, MII_LAN83C185_CTRL_STATUS); if (rc < 0) { SMSC_WARN(pdata, drv, "Failed reading PHY control reg"); @@ -1389,7 +1392,7 @@ static int smsc911x_phy_disable_energy_detect(struct smsc911x_data *pdata) /* Only disable if energy detect mode is already enabled */ if (rc & MII_LAN83C185_EDPWRDOWN) { /* Disable energy detect mode for this SMSC Transceivers */ - rc = phy_write(pdata->phy_dev, MII_LAN83C185_CTRL_STATUS, + rc = phy_write(phy_dev, MII_LAN83C185_CTRL_STATUS, rc & (~MII_LAN83C185_EDPWRDOWN)); if (rc < 0) { @@ -1405,12 +1408,14 @@ static int smsc911x_phy_disable_energy_detect(struct smsc911x_data *pdata) static int smsc911x_phy_enable_energy_detect(struct smsc911x_data *pdata) { + struct net_device *ndev = pdata->dev; + struct phy_device *phy_dev = ndev->phydev; int rc = 0; - if (!pdata->phy_dev) + if (!phy_dev) return rc; - rc = phy_read(pdata->phy_dev, MII_LAN83C185_CTRL_STATUS); + rc = phy_read(phy_dev, MII_LAN83C185_CTRL_STATUS); if (rc < 0) { SMSC_WARN(pdata, drv, "Failed reading PHY control reg"); @@ -1420,7 +1425,7 @@ static int smsc911x_phy_enable_energy_detect(struct smsc911x_data *pdata) /* Only enable if energy detect mode is already disabled */ if (!(rc & MII_LAN83C185_EDPWRDOWN)) { /* Enable energy detect mode for this SMSC Transceivers */ - rc = phy_write(pdata->phy_dev, MII_LAN83C185_CTRL_STATUS, + rc = phy_write(phy_dev, MII_LAN83C185_CTRL_STATUS, rc | MII_LAN83C185_EDPWRDOWN); if (rc < 0) { @@ -1517,7 +1522,7 @@ static int smsc911x_open(struct net_device *dev) unsigned int intcfg; /* if the phy is not yet registered, retry later*/ - if (!pdata->phy_dev) { + if (!dev->phydev) { SMSC_WARN(pdata, hw, "phy_dev is NULL"); return -EAGAIN; } @@ -1608,7 +1613,7 @@ static int smsc911x_open(struct net_device *dev) pdata->last_carrier = -1; /* Bring the PHY up */ - phy_start(pdata->phy_dev); + phy_start(dev->phydev); temp = smsc911x_reg_read(pdata, HW_CFG); /* Preserve TX FIFO size and external PHY configuration */ @@ -1663,8 +1668,8 @@ static int smsc911x_stop(struct net_device *dev) smsc911x_tx_update_txcounters(dev); /* Bring the PHY down */ - if (pdata->phy_dev) - phy_stop(pdata->phy_dev); + if (dev->phydev) + phy_stop(dev->phydev); SMSC_TRACE(pdata, ifdown, "Interface stopped"); return 0; @@ -1904,30 +1909,24 @@ static int smsc911x_set_mac_address(struct net_device *dev, void *p) /* Standard ioctls for mii-tool */ static int smsc911x_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { - struct smsc911x_data *pdata = netdev_priv(dev); - - if (!netif_running(dev) || !pdata->phy_dev) + if (!netif_running(dev) || !dev->phydev) return -EINVAL; - return phy_mii_ioctl(pdata->phy_dev, ifr, cmd); + return phy_mii_ioctl(dev->phydev, ifr, cmd); } static int smsc911x_ethtool_getsettings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct smsc911x_data *pdata = netdev_priv(dev); - cmd->maxtxpkt = 1; cmd->maxrxpkt = 1; - return phy_ethtool_gset(pdata->phy_dev, cmd); + return phy_ethtool_gset(dev->phydev, cmd); } static int smsc911x_ethtool_setsettings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct smsc911x_data *pdata = netdev_priv(dev); - - return phy_ethtool_sset(pdata->phy_dev, cmd); + return phy_ethtool_sset(dev->phydev, cmd); } static void smsc911x_ethtool_getdrvinfo(struct net_device *dev, @@ -1941,9 +1940,7 @@ static void smsc911x_ethtool_getdrvinfo(struct net_device *dev, static int smsc911x_ethtool_nwayreset(struct net_device *dev) { - struct smsc911x_data *pdata = netdev_priv(dev); - - return phy_start_aneg(pdata->phy_dev); + return phy_start_aneg(dev->phydev); } static u32 smsc911x_ethtool_getmsglevel(struct net_device *dev) @@ -1969,7 +1966,7 @@ smsc911x_ethtool_getregs(struct net_device *dev, struct ethtool_regs *regs, void *buf) { struct smsc911x_data *pdata = netdev_priv(dev); - struct phy_device *phy_dev = pdata->phy_dev; + struct phy_device *phy_dev = dev->phydev; unsigned long flags; unsigned int i; unsigned int j = 0; @@ -2308,12 +2305,11 @@ static int smsc911x_drv_remove(struct platform_device *pdev) pdata = netdev_priv(dev); BUG_ON(!pdata); BUG_ON(!pdata->ioaddr); - BUG_ON(!pdata->phy_dev); + BUG_ON(!dev->phydev); SMSC_TRACE(pdata, ifdown, "Stopping driver"); - phy_disconnect(pdata->phy_dev); - pdata->phy_dev = NULL; + phy_disconnect(dev->phydev); mdiobus_unregister(pdata->mii_bus); mdiobus_free(pdata->mii_bus); -- cgit v0.10.2 From 2c087409eb17c9f00d50e2d8e75a57724f6763f8 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 3 Jul 2016 01:14:21 +0200 Subject: net: ethernet: smsc: smsc911x: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index 6fded67..ca31345 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1915,20 +1915,6 @@ static int smsc911x_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) return phy_mii_ioctl(dev->phydev, ifr, cmd); } -static int -smsc911x_ethtool_getsettings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - cmd->maxtxpkt = 1; - cmd->maxrxpkt = 1; - return phy_ethtool_gset(dev->phydev, cmd); -} - -static int -smsc911x_ethtool_setsettings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - return phy_ethtool_sset(dev->phydev, cmd); -} - static void smsc911x_ethtool_getdrvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { @@ -2112,8 +2098,6 @@ static int smsc911x_ethtool_set_eeprom(struct net_device *dev, } static const struct ethtool_ops smsc911x_ethtool_ops = { - .get_settings = smsc911x_ethtool_getsettings, - .set_settings = smsc911x_ethtool_setsettings, .get_link = ethtool_op_get_link, .get_drvinfo = smsc911x_ethtool_getdrvinfo, .nway_reset = smsc911x_ethtool_nwayreset, @@ -2125,6 +2109,8 @@ static const struct ethtool_ops smsc911x_ethtool_ops = { .get_eeprom = smsc911x_ethtool_get_eeprom, .set_eeprom = smsc911x_ethtool_set_eeprom, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops smsc911x_netdev_ops = { -- cgit v0.10.2 From 2a62416dc67922458f809776d67138f0cc5544cd Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 2 Jul 2016 14:06:14 +0200 Subject: net: ethernet: ixp4xx_eth: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c index 5138407..a7778d9 100644 --- a/drivers/net/ethernet/xscale/ixp4xx_eth.c +++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c @@ -171,7 +171,6 @@ struct port { struct npe *npe; struct net_device *netdev; struct napi_struct napi; - struct phy_device *phydev; struct eth_plat_info *plat; buffer_t *rx_buff_tab[RX_DESCS], *tx_buff_tab[TX_DESCS]; struct desc *desc_tab; /* coherent */ @@ -562,7 +561,7 @@ static void ixp4xx_mdio_remove(void) static void ixp4xx_adjust_link(struct net_device *dev) { struct port *port = netdev_priv(dev); - struct phy_device *phydev = port->phydev; + struct phy_device *phydev = dev->phydev; if (!phydev->link) { if (port->speed) { @@ -976,8 +975,6 @@ static void eth_set_mcast_list(struct net_device *dev) static int eth_ioctl(struct net_device *dev, struct ifreq *req, int cmd) { - struct port *port = netdev_priv(dev); - if (!netif_running(dev)) return -EINVAL; @@ -988,7 +985,7 @@ static int eth_ioctl(struct net_device *dev, struct ifreq *req, int cmd) return hwtstamp_get(dev, req); } - return phy_mii_ioctl(port->phydev, req, cmd); + return phy_mii_ioctl(dev->phydev, req, cmd); } /* ethtool support */ @@ -1007,20 +1004,17 @@ static void ixp4xx_get_drvinfo(struct net_device *dev, static int ixp4xx_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct port *port = netdev_priv(dev); - return phy_ethtool_gset(port->phydev, cmd); + return phy_ethtool_gset(dev->phydev, cmd); } static int ixp4xx_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct port *port = netdev_priv(dev); - return phy_ethtool_sset(port->phydev, cmd); + return phy_ethtool_sset(dev->phydev, cmd); } static int ixp4xx_nway_reset(struct net_device *dev) { - struct port *port = netdev_priv(dev); - return phy_start_aneg(port->phydev); + return phy_start_aneg(dev->phydev); } int ixp46x_phc_index = -1; @@ -1259,7 +1253,7 @@ static int eth_open(struct net_device *dev) } port->speed = 0; /* force "link up" message */ - phy_start(port->phydev); + phy_start(dev->phydev); for (i = 0; i < ETH_ALEN; i++) __raw_writel(dev->dev_addr[i], &port->regs->hw_addr[i]); @@ -1380,7 +1374,7 @@ static int eth_close(struct net_device *dev) printk(KERN_CRIT "%s: unable to disable loopback\n", dev->name); - phy_stop(port->phydev); + phy_stop(dev->phydev); if (!ports_open) qmgr_disable_irq(TXDONE_QUEUE); @@ -1405,6 +1399,7 @@ static int eth_init_one(struct platform_device *pdev) struct port *port; struct net_device *dev; struct eth_plat_info *plat = dev_get_platdata(&pdev->dev); + struct phy_device *phydev = NULL; u32 regs_phys; char phy_id[MII_BUS_ID_SIZE + 3]; int err; @@ -1466,14 +1461,14 @@ static int eth_init_one(struct platform_device *pdev) snprintf(phy_id, MII_BUS_ID_SIZE + 3, PHY_ID_FMT, mdio_bus->id, plat->phy); - port->phydev = phy_connect(dev, phy_id, &ixp4xx_adjust_link, - PHY_INTERFACE_MODE_MII); - if (IS_ERR(port->phydev)) { - err = PTR_ERR(port->phydev); + phydev = phy_connect(dev, phy_id, &ixp4xx_adjust_link, + PHY_INTERFACE_MODE_MII); + if (IS_ERR(phydev)) { + err = PTR_ERR(phydev); goto err_free_mem; } - port->phydev->irq = PHY_POLL; + phydev->irq = PHY_POLL; if ((err = register_netdev(dev))) goto err_phy_dis; @@ -1484,7 +1479,7 @@ static int eth_init_one(struct platform_device *pdev) return 0; err_phy_dis: - phy_disconnect(port->phydev); + phy_disconnect(phydev); err_free_mem: npe_port_tab[NPE_ID(port->id)] = NULL; release_resource(port->mem_res); @@ -1498,10 +1493,11 @@ err_free: static int eth_remove_one(struct platform_device *pdev) { struct net_device *dev = platform_get_drvdata(pdev); + struct phy_device *phydev = dev->phydev; struct port *port = netdev_priv(dev); unregister_netdev(dev); - phy_disconnect(port->phydev); + phy_disconnect(phydev); npe_port_tab[NPE_ID(port->id)] = NULL; npe_release(port->npe); release_resource(port->mem_res); -- cgit v0.10.2 From fa018484cf5ea67678143c4570b63154fa819682 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 2 Jul 2016 14:06:15 +0200 Subject: net: ethernet: ixp4xx_eth: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c index a7778d9..7f127dc 100644 --- a/drivers/net/ethernet/xscale/ixp4xx_eth.c +++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c @@ -1002,16 +1002,6 @@ static void ixp4xx_get_drvinfo(struct net_device *dev, strlcpy(info->bus_info, "internal", sizeof(info->bus_info)); } -static int ixp4xx_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - return phy_ethtool_gset(dev->phydev, cmd); -} - -static int ixp4xx_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - return phy_ethtool_sset(dev->phydev, cmd); -} - static int ixp4xx_nway_reset(struct net_device *dev) { return phy_start_aneg(dev->phydev); @@ -1048,11 +1038,11 @@ static int ixp4xx_get_ts_info(struct net_device *dev, static const struct ethtool_ops ixp4xx_ethtool_ops = { .get_drvinfo = ixp4xx_get_drvinfo, - .get_settings = ixp4xx_get_settings, - .set_settings = ixp4xx_set_settings, .nway_reset = ixp4xx_nway_reset, .get_link = ethtool_op_get_link, .get_ts_info = ixp4xx_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; -- cgit v0.10.2 From 01dea536dc4e8eccd0d7741ec0e3dacd3060055c Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 2 Jul 2016 20:06:51 +0200 Subject: net: ethernet: arc: emac: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/arc/emac.h b/drivers/net/ethernet/arc/emac.h index ca562bc..e4feb71 100644 --- a/drivers/net/ethernet/arc/emac.h +++ b/drivers/net/ethernet/arc/emac.h @@ -134,7 +134,6 @@ struct arc_emac_priv { /* Devices */ struct device *dev; - struct phy_device *phy_dev; struct mii_bus *bus; struct arc_emac_mdio_bus_data bus_data; diff --git a/drivers/net/ethernet/arc/emac_main.c b/drivers/net/ethernet/arc/emac_main.c index a3a9392..a8a1dc9 100644 --- a/drivers/net/ethernet/arc/emac_main.c +++ b/drivers/net/ethernet/arc/emac_main.c @@ -47,7 +47,7 @@ static inline int arc_emac_tx_avail(struct arc_emac_priv *priv) static void arc_emac_adjust_link(struct net_device *ndev) { struct arc_emac_priv *priv = netdev_priv(ndev); - struct phy_device *phy_dev = priv->phy_dev; + struct phy_device *phy_dev = ndev->phydev; unsigned int reg, state_changed = 0; if (priv->link != phy_dev->link) { @@ -92,9 +92,7 @@ static void arc_emac_adjust_link(struct net_device *ndev) static int arc_emac_get_settings(struct net_device *ndev, struct ethtool_cmd *cmd) { - struct arc_emac_priv *priv = netdev_priv(ndev); - - return phy_ethtool_gset(priv->phy_dev, cmd); + return phy_ethtool_gset(ndev->phydev, cmd); } /** @@ -111,12 +109,10 @@ static int arc_emac_get_settings(struct net_device *ndev, static int arc_emac_set_settings(struct net_device *ndev, struct ethtool_cmd *cmd) { - struct arc_emac_priv *priv = netdev_priv(ndev); - if (!capable(CAP_NET_ADMIN)) return -EPERM; - return phy_ethtool_sset(priv->phy_dev, cmd); + return phy_ethtool_sset(ndev->phydev, cmd); } /** @@ -403,7 +399,7 @@ static void arc_emac_poll_controller(struct net_device *dev) static int arc_emac_open(struct net_device *ndev) { struct arc_emac_priv *priv = netdev_priv(ndev); - struct phy_device *phy_dev = priv->phy_dev; + struct phy_device *phy_dev = ndev->phydev; int i; phy_dev->autoneg = AUTONEG_ENABLE; @@ -474,7 +470,7 @@ static int arc_emac_open(struct net_device *ndev) /* Enable EMAC */ arc_reg_or(priv, R_CTRL, EN_MASK); - phy_start_aneg(priv->phy_dev); + phy_start_aneg(ndev->phydev); netif_start_queue(ndev); @@ -772,6 +768,7 @@ int arc_emac_probe(struct net_device *ndev, int interface) struct device *dev = ndev->dev.parent; struct resource res_regs; struct device_node *phy_node; + struct phy_device *phydev = NULL; struct arc_emac_priv *priv; const char *mac_addr; unsigned int id, clock_frequency, irq; @@ -887,16 +884,16 @@ int arc_emac_probe(struct net_device *ndev, int interface) goto out_clken; } - priv->phy_dev = of_phy_connect(ndev, phy_node, arc_emac_adjust_link, 0, - interface); - if (!priv->phy_dev) { + phydev = of_phy_connect(ndev, phy_node, arc_emac_adjust_link, 0, + interface); + if (!phydev) { dev_err(dev, "of_phy_connect() failed\n"); err = -ENODEV; goto out_mdio; } dev_info(dev, "connected to %s phy with id 0x%x\n", - priv->phy_dev->drv->name, priv->phy_dev->phy_id); + phydev->drv->name, phydev->phy_id); netif_napi_add(ndev, &priv->napi, arc_emac_poll, ARC_EMAC_NAPI_WEIGHT); @@ -910,8 +907,7 @@ int arc_emac_probe(struct net_device *ndev, int interface) out_netif_api: netif_napi_del(&priv->napi); - phy_disconnect(priv->phy_dev); - priv->phy_dev = NULL; + phy_disconnect(phydev); out_mdio: arc_mdio_remove(priv); out_clken: @@ -925,8 +921,7 @@ int arc_emac_remove(struct net_device *ndev) { struct arc_emac_priv *priv = netdev_priv(ndev); - phy_disconnect(priv->phy_dev); - priv->phy_dev = NULL; + phy_disconnect(ndev->phydev); arc_mdio_remove(priv); unregister_netdev(ndev); netif_napi_del(&priv->napi); -- cgit v0.10.2 From 4694e6e3f40c2ece234b65f0cda776c5590c0618 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 2 Jul 2016 20:06:52 +0200 Subject: net: ethernet: arc: emac: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/arc/emac_main.c b/drivers/net/ethernet/arc/emac_main.c index a8a1dc9..586beda 100644 --- a/drivers/net/ethernet/arc/emac_main.c +++ b/drivers/net/ethernet/arc/emac_main.c @@ -80,42 +80,6 @@ static void arc_emac_adjust_link(struct net_device *ndev) } /** - * arc_emac_get_settings - Get PHY settings. - * @ndev: Pointer to net_device structure. - * @cmd: Pointer to ethtool_cmd structure. - * - * This implements ethtool command for getting PHY settings. If PHY could - * not be found, the function returns -ENODEV. This function calls the - * relevant PHY ethtool API to get the PHY settings. - * Issue "ethtool ethX" under linux prompt to execute this function. - */ -static int arc_emac_get_settings(struct net_device *ndev, - struct ethtool_cmd *cmd) -{ - return phy_ethtool_gset(ndev->phydev, cmd); -} - -/** - * arc_emac_set_settings - Set PHY settings as passed in the argument. - * @ndev: Pointer to net_device structure. - * @cmd: Pointer to ethtool_cmd structure. - * - * This implements ethtool command for setting various PHY settings. If PHY - * could not be found, the function returns -ENODEV. This function calls the - * relevant PHY ethtool API to set the PHY. - * Issue e.g. "ethtool -s ethX speed 1000" under linux prompt to execute this - * function. - */ -static int arc_emac_set_settings(struct net_device *ndev, - struct ethtool_cmd *cmd) -{ - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - return phy_ethtool_sset(ndev->phydev, cmd); -} - -/** * arc_emac_get_drvinfo - Get EMAC driver information. * @ndev: Pointer to net_device structure. * @info: Pointer to ethtool_drvinfo structure. @@ -133,10 +97,10 @@ static void arc_emac_get_drvinfo(struct net_device *ndev, } static const struct ethtool_ops arc_emac_ethtool_ops = { - .get_settings = arc_emac_get_settings, - .set_settings = arc_emac_set_settings, .get_drvinfo = arc_emac_get_drvinfo, .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; #define FIRST_OR_LAST_MASK (FIRST_MASK | LAST_MASK) -- cgit v0.10.2 From 62469c76007e11428e2ee3c6de90cbe74b588d44 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 3 Jul 2016 17:33:56 +0200 Subject: net: ethernet: bcmgenet: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 5414563..8d4f849 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -453,29 +453,25 @@ static inline void bcmgenet_rdma_ring_writel(struct bcmgenet_priv *priv, static int bcmgenet_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct bcmgenet_priv *priv = netdev_priv(dev); - if (!netif_running(dev)) return -EINVAL; - if (!priv->phydev) + if (!dev->phydev) return -ENODEV; - return phy_ethtool_gset(priv->phydev, cmd); + return phy_ethtool_gset(dev->phydev, cmd); } static int bcmgenet_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct bcmgenet_priv *priv = netdev_priv(dev); - if (!netif_running(dev)) return -EINVAL; - if (!priv->phydev) + if (!dev->phydev) return -ENODEV; - return phy_ethtool_sset(priv->phydev, cmd); + return phy_ethtool_sset(dev->phydev, cmd); } static int bcmgenet_set_rx_csum(struct net_device *dev, @@ -941,7 +937,7 @@ static int bcmgenet_get_eee(struct net_device *dev, struct ethtool_eee *e) e->eee_active = p->eee_active; e->tx_lpi_timer = bcmgenet_umac_readl(priv, UMAC_EEE_LPI_TIMER); - return phy_ethtool_get_eee(priv->phydev, e); + return phy_ethtool_get_eee(dev->phydev, e); } static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e) @@ -958,7 +954,7 @@ static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e) if (!p->eee_enabled) { bcmgenet_eee_enable_set(dev, false); } else { - ret = phy_init_eee(priv->phydev, 0); + ret = phy_init_eee(dev->phydev, 0); if (ret) { netif_err(priv, hw, dev, "EEE initialization failed\n"); return ret; @@ -968,14 +964,12 @@ static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e) bcmgenet_eee_enable_set(dev, true); } - return phy_ethtool_set_eee(priv->phydev, e); + return phy_ethtool_set_eee(dev->phydev, e); } static int bcmgenet_nway_reset(struct net_device *dev) { - struct bcmgenet_priv *priv = netdev_priv(dev); - - return genphy_restart_aneg(priv->phydev); + return genphy_restart_aneg(dev->phydev); } /* standard ethtool support functions. */ @@ -1002,12 +996,13 @@ static struct ethtool_ops bcmgenet_ethtool_ops = { static int bcmgenet_power_down(struct bcmgenet_priv *priv, enum bcmgenet_power_mode mode) { + struct net_device *ndev = priv->dev; int ret = 0; u32 reg; switch (mode) { case GENET_POWER_CABLE_SENSE: - phy_detach(priv->phydev); + phy_detach(ndev->phydev); break; case GENET_POWER_WOL_MAGIC: @@ -1068,7 +1063,6 @@ static void bcmgenet_power_up(struct bcmgenet_priv *priv, /* ioctl handle special commands that are not present in ethtool. */ static int bcmgenet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct bcmgenet_priv *priv = netdev_priv(dev); int val = 0; if (!netif_running(dev)) @@ -1078,10 +1072,10 @@ static int bcmgenet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) case SIOCGMIIPHY: case SIOCGMIIREG: case SIOCSMIIREG: - if (!priv->phydev) + if (!dev->phydev) val = -ENODEV; else - val = phy_mii_ioctl(priv->phydev, rq, cmd); + val = phy_mii_ioctl(dev->phydev, rq, cmd); break; default: @@ -2464,6 +2458,7 @@ static void bcmgenet_irq_task(struct work_struct *work) { struct bcmgenet_priv *priv = container_of( work, struct bcmgenet_priv, bcmgenet_irq_work); + struct net_device *ndev = priv->dev; netif_dbg(priv, intr, priv->dev, "%s\n", __func__); @@ -2476,7 +2471,7 @@ static void bcmgenet_irq_task(struct work_struct *work) /* Link UP/DOWN event */ if (priv->irq0_stat & UMAC_IRQ_LINK_EVENT) { - phy_mac_interrupt(priv->phydev, + phy_mac_interrupt(ndev->phydev, !!(priv->irq0_stat & UMAC_IRQ_LINK_UP)); priv->irq0_stat &= ~UMAC_IRQ_LINK_EVENT; } @@ -2838,7 +2833,7 @@ static void bcmgenet_netif_start(struct net_device *dev) /* Monitor link interrupts now */ bcmgenet_link_intr_enable(priv); - phy_start(priv->phydev); + phy_start(dev->phydev); } static int bcmgenet_open(struct net_device *dev) @@ -2937,7 +2932,7 @@ static void bcmgenet_netif_stop(struct net_device *dev) struct bcmgenet_priv *priv = netdev_priv(dev); netif_tx_stop_all_queues(dev); - phy_stop(priv->phydev); + phy_stop(dev->phydev); bcmgenet_intr_disable(priv); bcmgenet_disable_rx_napi(priv); bcmgenet_disable_tx_napi(priv); @@ -2963,7 +2958,7 @@ static int bcmgenet_close(struct net_device *dev) bcmgenet_netif_stop(dev); /* Really kill the PHY state machine and disconnect from it */ - phy_disconnect(priv->phydev); + phy_disconnect(dev->phydev); /* Disable MAC receive */ umac_enable_set(priv, CMD_RX_EN, false); @@ -3522,7 +3517,7 @@ static int bcmgenet_suspend(struct device *d) bcmgenet_netif_stop(dev); - phy_suspend(priv->phydev); + phy_suspend(dev->phydev); netif_device_detach(dev); @@ -3586,7 +3581,7 @@ static int bcmgenet_resume(struct device *d) if (priv->wolopts) clk_disable_unprepare(priv->clk_wol); - phy_init_hw(priv->phydev); + phy_init_hw(dev->phydev); /* Speed settings must be restored */ bcmgenet_mii_config(priv->dev); @@ -3619,7 +3614,7 @@ static int bcmgenet_resume(struct device *d) netif_device_attach(dev); - phy_resume(priv->phydev); + phy_resume(dev->phydev); if (priv->eee.eee_enabled) bcmgenet_eee_enable_set(dev, true); diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h index 1e2dc34..0f0868c 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h @@ -597,7 +597,6 @@ struct bcmgenet_priv { /* MDIO bus variables */ wait_queue_head_t wq; - struct phy_device *phydev; bool internal_phy; struct device_node *phy_dn; struct device_node *mdio_dn; diff --git a/drivers/net/ethernet/broadcom/genet/bcmmii.c b/drivers/net/ethernet/broadcom/genet/bcmmii.c index 457c3bc..e907acd 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmmii.c +++ b/drivers/net/ethernet/broadcom/genet/bcmmii.c @@ -86,7 +86,7 @@ static int bcmgenet_mii_write(struct mii_bus *bus, int phy_id, void bcmgenet_mii_setup(struct net_device *dev) { struct bcmgenet_priv *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; u32 reg, cmd_bits = 0; bool status_changed = false; @@ -183,9 +183,9 @@ void bcmgenet_mii_reset(struct net_device *dev) if (GENET_IS_V4(priv)) return; - if (priv->phydev) { - phy_init_hw(priv->phydev); - phy_start_aneg(priv->phydev); + if (dev->phydev) { + phy_init_hw(dev->phydev); + phy_start_aneg(dev->phydev); } } @@ -236,6 +236,7 @@ static void bcmgenet_internal_phy_setup(struct net_device *dev) static void bcmgenet_moca_phy_setup(struct bcmgenet_priv *priv) { + struct net_device *ndev = priv->dev; u32 reg; /* Speed settings are set in bcmgenet_mii_setup() */ @@ -244,14 +245,14 @@ static void bcmgenet_moca_phy_setup(struct bcmgenet_priv *priv) bcmgenet_sys_writel(priv, reg, SYS_PORT_CTRL); if (priv->hw_params->flags & GENET_HAS_MOCA_LINK_DET) - fixed_phy_set_link_update(priv->phydev, + fixed_phy_set_link_update(ndev->phydev, bcmgenet_fixed_phy_link_update); } int bcmgenet_mii_config(struct net_device *dev) { struct bcmgenet_priv *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phydev; + struct phy_device *phydev = dev->phydev; struct device *kdev = &priv->pdev->dev; const char *phy_name = NULL; u32 id_mode_dis = 0; @@ -302,7 +303,7 @@ int bcmgenet_mii_config(struct net_device *dev) * capabilities, use that knowledge to also configure the * Reverse MII interface correctly. */ - if ((priv->phydev->supported & PHY_BASIC_FEATURES) == + if ((phydev->supported & PHY_BASIC_FEATURES) == PHY_BASIC_FEATURES) port_ctrl = PORT_MODE_EXT_RVMII_25; else @@ -371,7 +372,7 @@ int bcmgenet_mii_probe(struct net_device *dev) return -ENODEV; } } else { - phydev = priv->phydev; + phydev = dev->phydev; phydev->dev_flags = phy_flags; ret = phy_connect_direct(dev, phydev, bcmgenet_mii_setup, @@ -382,8 +383,6 @@ int bcmgenet_mii_probe(struct net_device *dev) } } - priv->phydev = phydev; - /* Configure port multiplexer based on what the probed PHY device since * reading the 'max-speed' property determines the maximum supported * PHY speed which is needed for bcmgenet_mii_config() to configure @@ -391,7 +390,7 @@ int bcmgenet_mii_probe(struct net_device *dev) */ ret = bcmgenet_mii_config(dev); if (ret) { - phy_disconnect(priv->phydev); + phy_disconnect(phydev); return ret; } @@ -401,7 +400,7 @@ int bcmgenet_mii_probe(struct net_device *dev) * Ethernet MAC ISRs */ if (priv->internal_phy) - priv->phydev->irq = PHY_IGNORE_INTERRUPT; + phydev->irq = PHY_IGNORE_INTERRUPT; return 0; } @@ -606,7 +605,6 @@ static int bcmgenet_mii_pd_init(struct bcmgenet_priv *priv) } - priv->phydev = phydev; priv->phy_interface = pd->phy_interface; return 0; -- cgit v0.10.2 From 4386f5662e639553552408c33014ca3e8d612b14 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 3 Jul 2016 17:33:57 +0200 Subject: net: ethernet: bcmgenet: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 8d4f849..76ed6df 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -450,30 +450,6 @@ static inline void bcmgenet_rdma_ring_writel(struct bcmgenet_priv *priv, genet_dma_ring_regs[r]); } -static int bcmgenet_get_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - if (!netif_running(dev)) - return -EINVAL; - - if (!dev->phydev) - return -ENODEV; - - return phy_ethtool_gset(dev->phydev, cmd); -} - -static int bcmgenet_set_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - if (!netif_running(dev)) - return -EINVAL; - - if (!dev->phydev) - return -ENODEV; - - return phy_ethtool_sset(dev->phydev, cmd); -} - static int bcmgenet_set_rx_csum(struct net_device *dev, netdev_features_t wanted) { @@ -977,8 +953,6 @@ static struct ethtool_ops bcmgenet_ethtool_ops = { .get_strings = bcmgenet_get_strings, .get_sset_count = bcmgenet_get_sset_count, .get_ethtool_stats = bcmgenet_get_ethtool_stats, - .get_settings = bcmgenet_get_settings, - .set_settings = bcmgenet_set_settings, .get_drvinfo = bcmgenet_get_drvinfo, .get_link = ethtool_op_get_link, .get_msglevel = bcmgenet_get_msglevel, @@ -990,6 +964,8 @@ static struct ethtool_ops bcmgenet_ethtool_ops = { .nway_reset = bcmgenet_nway_reset, .get_coalesce = bcmgenet_get_coalesce, .set_coalesce = bcmgenet_set_coalesce, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /* Power down the unimac, based on mode. */ -- cgit v0.10.2 From 0967f2445963b63269d7dd2f5b6f234ea57dd10e Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Sat, 2 Jul 2016 14:12:54 -0700 Subject: net: pktgen: support injecting packets for qdisc testing Add another xmit_mode to pktgen to allow testing xmit functionality of qdiscs. The new mode "queue_xmit" injects packets at __dev_queue_xmit() so that qdisc is called. Signed-off-by: John Fastabend Signed-off-by: David S. Miller diff --git a/net/core/pktgen.c b/net/core/pktgen.c index f74ab9c..bbd118b 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -213,6 +213,7 @@ /* Xmit modes */ #define M_START_XMIT 0 /* Default normal TX */ #define M_NETIF_RECEIVE 1 /* Inject packets into stack */ +#define M_QUEUE_XMIT 2 /* Inject packet into qdisc */ /* If lock -- protects updating of if_list */ #define if_lock(t) spin_lock(&(t->if_lock)); @@ -626,6 +627,8 @@ static int pktgen_if_show(struct seq_file *seq, void *v) if (pkt_dev->xmit_mode == M_NETIF_RECEIVE) seq_puts(seq, " xmit_mode: netif_receive\n"); + else if (pkt_dev->xmit_mode == M_QUEUE_XMIT) + seq_puts(seq, " xmit_mode: xmit_queue\n"); seq_puts(seq, " Flags: "); @@ -1142,8 +1145,10 @@ static ssize_t pktgen_if_write(struct file *file, return len; i += len; - if ((value > 1) && (pkt_dev->xmit_mode == M_START_XMIT) && - (!(pkt_dev->odev->priv_flags & IFF_TX_SKB_SHARING))) + if ((value > 1) && + ((pkt_dev->xmit_mode == M_QUEUE_XMIT) || + ((pkt_dev->xmit_mode == M_START_XMIT) && + (!(pkt_dev->odev->priv_flags & IFF_TX_SKB_SHARING))))) return -ENOTSUPP; pkt_dev->burst = value < 1 ? 1 : value; sprintf(pg_result, "OK: burst=%d", pkt_dev->burst); @@ -1198,6 +1203,9 @@ static ssize_t pktgen_if_write(struct file *file, * at module loading time */ pkt_dev->clone_skb = 0; + } else if (strcmp(f, "queue_xmit") == 0) { + pkt_dev->xmit_mode = M_QUEUE_XMIT; + pkt_dev->last_ok = 1; } else { sprintf(pg_result, "xmit_mode -:%s:- unknown\nAvailable modes: %s", @@ -3434,6 +3442,36 @@ static void pktgen_xmit(struct pktgen_dev *pkt_dev) #endif } while (--burst > 0); goto out; /* Skips xmit_mode M_START_XMIT */ + } else if (pkt_dev->xmit_mode == M_QUEUE_XMIT) { + local_bh_disable(); + atomic_inc(&pkt_dev->skb->users); + + ret = dev_queue_xmit(pkt_dev->skb); + switch (ret) { + case NET_XMIT_SUCCESS: + pkt_dev->sofar++; + pkt_dev->seq_num++; + pkt_dev->tx_bytes += pkt_dev->last_pkt_size; + break; + case NET_XMIT_DROP: + case NET_XMIT_CN: + /* These are all valid return codes for a qdisc but + * indicate packets are being dropped or will likely + * be dropped soon. + */ + case NETDEV_TX_BUSY: + /* qdisc may call dev_hard_start_xmit directly in cases + * where no queues exist e.g. loopback device, virtual + * devices, etc. In this case we need to handle + * NETDEV_TX_ codes. + */ + default: + pkt_dev->errors++; + net_info_ratelimited("%s xmit error: %d\n", + pkt_dev->odevname, ret); + break; + } + goto out; } txq = skb_get_tx_queue(odev, pkt_dev->skb); -- cgit v0.10.2 From 6fd980ac39efee9c26b1eb256c3271fcb139bd99 Mon Sep 17 00:00:00 2001 From: John Fastabend Date: Sat, 2 Jul 2016 14:13:13 -0700 Subject: net: samples: pktgen mode samples/tests for qdisc layer This adds samples for pktgen to use with new mode to inject pkts into the qdisc layer. This also doubles as nice test cases to test any patches against qdisc layer. Signed-off-by: John Fastabend Signed-off-by: David S. Miller diff --git a/samples/pktgen/pktgen_bench_xmit_mode_queue_xmit.sh b/samples/pktgen/pktgen_bench_xmit_mode_queue_xmit.sh new file mode 100755 index 0000000..4e4e92b --- /dev/null +++ b/samples/pktgen/pktgen_bench_xmit_mode_queue_xmit.sh @@ -0,0 +1,66 @@ +#!/bin/bash +# +# Benchmark script: +# - developed for benchmarking egress qdisc path, derived (more +# like cut'n'pasted) from ingress benchmark script. +# +# Script for injecting packets into egress qdisc path of the stack +# with pktgen "xmit_mode queue_xmit". +# +basedir=`dirname $0` +source ${basedir}/functions.sh +root_check_run_with_sudo "$@" + +# Parameter parsing via include +source ${basedir}/parameters.sh +[ -z "$DEST_IP" ] && DEST_IP="198.18.0.42" +[ -z "$DST_MAC" ] && DST_MAC="90:e2:ba:ff:ff:ff" + +# Burst greater than 1 are invalid for queue_xmit mode +if [[ -n "$BURST" ]]; then + err 1 "Bursting not supported for this mode" +fi + +# Base Config +DELAY="0" # Zero means max speed +COUNT="10000000" # Zero means indefinitely + +# General cleanup everything since last run +pg_ctrl "reset" + +# Threads are specified with parameter -t value in $THREADS +for ((thread = 0; thread < $THREADS; thread++)); do + # The device name is extended with @name, using thread number to + # make then unique, but any name will do. + dev=${DEV}@${thread} + + # Add remove all other devices and add_device $dev to thread + pg_thread $thread "rem_device_all" + pg_thread $thread "add_device" $dev + + # Base config of dev + pg_set $dev "flag QUEUE_MAP_CPU" + pg_set $dev "count $COUNT" + pg_set $dev "pkt_size $PKT_SIZE" + pg_set $dev "delay $DELAY" + pg_set $dev "flag NO_TIMESTAMP" + + # Destination + pg_set $dev "dst_mac $DST_MAC" + pg_set $dev "dst $DEST_IP" + + # Inject packet into TX qdisc egress path of stack + pg_set $dev "xmit_mode queue_xmit" +done + +# start_run +echo "Running... ctrl^C to stop" >&2 +pg_ctrl "start" +echo "Done" >&2 + +# Print results +for ((thread = 0; thread < $THREADS; thread++)); do + dev=${DEV}@${thread} + echo "Device: $dev" + cat /proc/net/pktgen/$dev | grep -A2 "Result:" +done -- cgit v0.10.2 From 13c5c240f789bbd2bcacb14a23771491485ae61f Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Sun, 3 Jul 2016 01:28:47 +0200 Subject: bpf: add bpf_get_hash_recalc helper If skb_clear_hash() was invoked due to mangling of relevant headers and BPF program needs skb->hash later on, we can add a helper to trigger hash recalculation via bpf_get_hash_recalc(). The helper will return the newly retrieved hash directly, but later access can also be done via skb context again through skb->hash directly (inline) without needing to call the helper once more. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index f44504d..c14ca1c 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -348,6 +348,15 @@ enum bpf_func_id { * < 0 error */ BPF_FUNC_skb_in_cgroup, + + /** + * bpf_get_hash_recalc(skb) + * Retrieve and possibly recalculate skb->hash. + * @skb: pointer to skb + * Return: hash + */ + BPF_FUNC_get_hash_recalc, + __BPF_FUNC_MAX_ID, }; diff --git a/net/core/filter.c b/net/core/filter.c index 54071cf..10c4a2f 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1729,6 +1729,23 @@ static const struct bpf_func_proto bpf_get_route_realm_proto = { .arg1_type = ARG_PTR_TO_CTX, }; +static u64 bpf_get_hash_recalc(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + /* If skb_clear_hash() was called due to mangling, we can + * trigger SW recalculation here. Later access to hash + * can then use the inline skb->hash via context directly + * instead of calling this helper again. + */ + return skb_get_hash((struct sk_buff *) (unsigned long) r1); +} + +static const struct bpf_func_proto bpf_get_hash_recalc_proto = { + .func = bpf_get_hash_recalc, + .gpl_only = false, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, +}; + static u64 bpf_skb_vlan_push(u64 r1, u64 r2, u64 vlan_tci, u64 r4, u64 r5) { struct sk_buff *skb = (struct sk_buff *) (long) r1; @@ -2337,6 +2354,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) return &bpf_redirect_proto; case BPF_FUNC_get_route_realm: return &bpf_get_route_realm_proto; + case BPF_FUNC_get_hash_recalc: + return &bpf_get_hash_recalc_proto; case BPF_FUNC_perf_event_output: return bpf_get_event_output_proto(); case BPF_FUNC_get_smp_processor_id: -- cgit v0.10.2 From 01fb237ac0b0f8c6a5245dfd194ff9f50c308434 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Sun, 3 Jul 2016 13:56:47 -0700 Subject: liquidio: Vxlan support This patch adds support for Vxaln offloads in liquidio driver. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 03bfa97..a060586 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -106,6 +106,7 @@ static const char oct_stats_strings[][ETH_GSTRING_LEN] = { "tx_tso", "tx_tso_packets", "tx_tso_err", + "tx_vxlan", "mac_tx_total_pkts", "mac_tx_total_bytes", @@ -129,6 +130,9 @@ static const char oct_stats_strings[][ETH_GSTRING_LEN] = { "rx_err_link", "rx_err_drop", + "rx_vxlan", + "rx_vxlan_err", + "rx_lro_pkts", "rx_lro_bytes", "rx_total_lro", @@ -167,6 +171,7 @@ static const char oct_iq_stats_strings[][ETH_GSTRING_LEN] = { "fw_bytes_sent", "tso", + "vxlan", "txq_restart", }; @@ -186,6 +191,7 @@ static const char oct_droq_stats_strings[][ETH_GSTRING_LEN] = { "fw_bytes_received", "fw_dropped_nodispatch", + "vxlan", "buffer_alloc_failure", }; @@ -675,6 +681,10 @@ lio_get_ethtool_stats(struct net_device *netdev, *fw_err_tso */ data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_err_tso); + /*per_core_stats[cvmx_get_core_num()].link_stats[idx].fromhost. + *fw_tx_vxlan + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromhost.fw_tx_vxlan); /* mac tx statistics */ /*CVMX_BGXX_CMRX_TX_STAT5 */ @@ -729,6 +739,15 @@ lio_get_ethtool_stats(struct net_device *netdev, */ data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_err_drop); + /*per_core_stats[cvmx_get_core_num()].link_stats[lro_ctx->ifidx]. + *fromwire.fw_rx_vxlan + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_rx_vxlan); + /*per_core_stats[cvmx_get_core_num()].link_stats[lro_ctx->ifidx]. + *fromwire.fw_rx_vxlan_err + */ + data[i++] = CVM_CAST64(oct_dev->link_stats.fromwire.fw_rx_vxlan_err); + /* LRO */ /*per_core_stats[cvmx_get_core_num()].link_stats[ifidx].fromwire. *fw_lro_pkts @@ -822,6 +841,8 @@ lio_get_ethtool_stats(struct net_device *netdev, /*tso request*/ data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_gso); + /*vxlan request*/ + data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_vxlan); /*txq restart*/ data[i++] = CVM_CAST64(oct_dev->instr_queue[j]->stats.tx_restart); @@ -858,6 +879,9 @@ lio_get_ethtool_stats(struct net_device *netdev, CVM_CAST64(oct_dev->droq[j]->stats.bytes_received); data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.dropped_nodispatch); + + data[i++] = + CVM_CAST64(oct_dev->droq[j]->stats.rx_vxlan); data[i++] = CVM_CAST64(oct_dev->droq[j]->stats.rx_alloc_failure); } @@ -1083,6 +1107,9 @@ octnet_nic_stats_callback(struct octeon_device *oct_dev, rstats->fw_err_pko = rsp_rstats->fw_err_pko; rstats->fw_err_link = rsp_rstats->fw_err_link; rstats->fw_err_drop = rsp_rstats->fw_err_drop; + rstats->fw_rx_vxlan = rsp_rstats->fw_rx_vxlan; + rstats->fw_rx_vxlan_err = rsp_rstats->fw_rx_vxlan_err; + /* Number of packets that are LROed */ rstats->fw_lro_pkts = rsp_rstats->fw_lro_pkts; /* Number of octets that are LROed */ @@ -1127,6 +1154,8 @@ octnet_nic_stats_callback(struct octeon_device *oct_dev, tstats->fw_tso = rsp_tstats->fw_tso; tstats->fw_tso_fwd = rsp_tstats->fw_tso_fwd; tstats->fw_err_tso = rsp_tstats->fw_err_tso; + tstats->fw_tx_vxlan = rsp_tstats->fw_tx_vxlan; + resp->status = 1; } else { resp->status = -1; diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 1a584eb..4a5629f 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -37,6 +37,7 @@ #include #include #include +#include #include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" @@ -2000,14 +2001,25 @@ liquidio_push_packet(u32 octeon_id, } skb->protocol = eth_type_trans(skb, skb->dev); - if ((netdev->features & NETIF_F_RXCSUM) && - (rh->r_dh.csum_verified == CNNIC_CSUM_VERIFIED)) + (((rh->r_dh.encap_on) && + (rh->r_dh.csum_verified & CNNIC_TUN_CSUM_VERIFIED)) || + (!(rh->r_dh.encap_on) && + (rh->r_dh.csum_verified & CNNIC_CSUM_VERIFIED)))) /* checksum has already been verified */ skb->ip_summed = CHECKSUM_UNNECESSARY; else skb->ip_summed = CHECKSUM_NONE; + /* Setting Encapsulation field on basis of status received + * from the firmware + */ + if (rh->r_dh.encap_on) { + skb->encapsulation = 1; + skb->csum_level = 1; + droq->stats.rx_vxlan++; + } + /* inbound VLAN tag */ if ((netdev->features & NETIF_F_HW_VLAN_CTAG_RX) && (rh->r_dh.vlan != 0)) { @@ -2410,6 +2422,55 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr) netdev->name); break; + /* Case to handle "OCTNET_CMD_TNL_RX_CSUM_CTL" + * Command passed by NIC driver + */ + case OCTNET_CMD_TNL_RX_CSUM_CTL: + if (nctrl->ncmd.s.param1 == OCTNET_CMD_RXCSUM_ENABLE) { + netif_info(lio, probe, lio->netdev, + "%s RX Checksum Offload Enabled\n", + netdev->name); + } else if (nctrl->ncmd.s.param1 == + OCTNET_CMD_RXCSUM_DISABLE) { + netif_info(lio, probe, lio->netdev, + "%s RX Checksum Offload Disabled\n", + netdev->name); + } + break; + + /* Case to handle "OCTNET_CMD_TNL_TX_CSUM_CTL" + * Command passed by NIC driver + */ + case OCTNET_CMD_TNL_TX_CSUM_CTL: + if (nctrl->ncmd.s.param1 == OCTNET_CMD_TXCSUM_ENABLE) { + netif_info(lio, probe, lio->netdev, + "%s TX Checksum Offload Enabled\n", + netdev->name); + } else if (nctrl->ncmd.s.param1 == + OCTNET_CMD_TXCSUM_DISABLE) { + netif_info(lio, probe, lio->netdev, + "%s TX Checksum Offload Disabled\n", + netdev->name); + } + break; + + /* Case to handle "OCTNET_CMD_VXLAN_PORT_CONFIG" + * Command passed by NIC driver + */ + case OCTNET_CMD_VXLAN_PORT_CONFIG: + if (nctrl->ncmd.s.more == OCTNET_CMD_VXLAN_PORT_ADD) { + netif_info(lio, probe, lio->netdev, + "%s VxLAN Destination UDP PORT:%d ADDED\n", + netdev->name, + nctrl->ncmd.s.param1); + } else if (nctrl->ncmd.s.more == + OCTNET_CMD_VXLAN_PORT_DEL) { + netif_info(lio, probe, lio->netdev, + "%s VxLAN Destination UDP PORT:%d DELETED\n", + netdev->name, + nctrl->ncmd.s.param1); + } + break; case OCTNET_CMD_SET_FLOW_CTL: netif_info(lio, probe, lio->netdev, "Set RX/TX flow control parameters\n"); @@ -2899,9 +2960,14 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) cmdsetup.u64 = 0; cmdsetup.s.iq_no = iq_no; - if (skb->ip_summed == CHECKSUM_PARTIAL) - cmdsetup.s.transport_csum = 1; - + if (skb->ip_summed == CHECKSUM_PARTIAL) { + if (skb->encapsulation) { + cmdsetup.s.tnl_csum = 1; + stats->tx_vxlan++; + } else { + cmdsetup.s.transport_csum = 1; + } + } if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) { skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; cmdsetup.s.timestamp = 1; @@ -3124,6 +3190,72 @@ static int liquidio_vlan_rx_kill_vid(struct net_device *netdev, return ret; } +/** Sending command to enable/disable RX checksum offload + * @param netdev pointer to network device + * @param command OCTNET_CMD_TNL_RX_CSUM_CTL + * @param rx_cmd_bit OCTNET_CMD_RXCSUM_ENABLE/ + * OCTNET_CMD_RXCSUM_DISABLE + * @returns SUCCESS or FAILURE + */ +int liquidio_set_rxcsum_command(struct net_device *netdev, int command, + u8 rx_cmd) +{ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; + struct octnic_ctrl_pkt nctrl; + int ret = 0; + + nctrl.ncmd.u64 = 0; + nctrl.ncmd.s.cmd = command; + nctrl.ncmd.s.param1 = rx_cmd; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; + nctrl.wait_time = 100; + nctrl.netpndev = (u64)netdev; + nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; + + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); + if (ret < 0) { + dev_err(&oct->pci_dev->dev, + "DEVFLAGS RXCSUM change failed in core(ret:0x%x)\n", + ret); + } + return ret; +} + +/** Sending command to add/delete VxLAN UDP port to firmware + * @param netdev pointer to network device + * @param command OCTNET_CMD_VXLAN_PORT_CONFIG + * @param vxlan_port VxLAN port to be added or deleted + * @param vxlan_cmd_bit OCTNET_CMD_VXLAN_PORT_ADD, + * OCTNET_CMD_VXLAN_PORT_DEL + * @returns SUCCESS or FAILURE + */ +static int liquidio_vxlan_port_command(struct net_device *netdev, int command, + u16 vxlan_port, u8 vxlan_cmd_bit) +{ + struct lio *lio = GET_LIO(netdev); + struct octeon_device *oct = lio->oct_dev; + struct octnic_ctrl_pkt nctrl; + int ret = 0; + + nctrl.ncmd.u64 = 0; + nctrl.ncmd.s.cmd = command; + nctrl.ncmd.s.more = vxlan_cmd_bit; + nctrl.ncmd.s.param1 = vxlan_port; + nctrl.iq_no = lio->linfo.txpciq[0].s.q_no; + nctrl.wait_time = 100; + nctrl.netpndev = (u64)netdev; + nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; + + ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl); + if (ret < 0) { + dev_err(&oct->pci_dev->dev, + "VxLAN port add/delete failed in core (ret:0x%x)\n", + ret); + } + return ret; +} + int liquidio_set_feature(struct net_device *netdev, int cmd, u16 param1) { struct lio *lio = GET_LIO(netdev); @@ -3204,9 +3336,48 @@ static int liquidio_set_features(struct net_device *netdev, liquidio_set_feature(netdev, OCTNET_CMD_LRO_DISABLE, OCTNIC_LROIPV4 | OCTNIC_LROIPV6); + /* Sending command to firmware to enable/disable RX checksum + * offload settings using ethtool + */ + if (!(netdev->features & NETIF_F_RXCSUM) && + (lio->enc_dev_capability & NETIF_F_RXCSUM) && + (features & NETIF_F_RXCSUM)) + liquidio_set_rxcsum_command(netdev, + OCTNET_CMD_TNL_RX_CSUM_CTL, + OCTNET_CMD_RXCSUM_ENABLE); + else if ((netdev->features & NETIF_F_RXCSUM) && + (lio->enc_dev_capability & NETIF_F_RXCSUM) && + !(features & NETIF_F_RXCSUM)) + liquidio_set_rxcsum_command(netdev, OCTNET_CMD_TNL_RX_CSUM_CTL, + OCTNET_CMD_RXCSUM_DISABLE); + return 0; } +static void liquidio_add_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti) +{ + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + liquidio_vxlan_port_command(netdev, + OCTNET_CMD_VXLAN_PORT_CONFIG, + htons(ti->port), + OCTNET_CMD_VXLAN_PORT_ADD); +} + +static void liquidio_del_vxlan_port(struct net_device *netdev, + struct udp_tunnel_info *ti) +{ + if (ti->type != UDP_TUNNEL_TYPE_VXLAN) + return; + + liquidio_vxlan_port_command(netdev, + OCTNET_CMD_VXLAN_PORT_CONFIG, + htons(ti->port), + OCTNET_CMD_VXLAN_PORT_DEL); +} + static struct net_device_ops lionetdevops = { .ndo_open = liquidio_open, .ndo_stop = liquidio_stop, @@ -3222,6 +3393,8 @@ static struct net_device_ops lionetdevops = { .ndo_do_ioctl = liquidio_ioctl, .ndo_fix_features = liquidio_fix_features, .ndo_set_features = liquidio_set_features, + .ndo_udp_tunnel_add = liquidio_add_vxlan_port, + .ndo_udp_tunnel_del = liquidio_del_vxlan_port, }; /** \brief Entry point for the liquidio module @@ -3479,6 +3652,22 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) | NETIF_F_LRO; netif_set_gso_max_size(netdev, OCTNIC_GSO_MAX_SIZE); + /* Copy of transmit encapsulation capabilities: + * TSO, TSO6, Checksums for this device + */ + lio->enc_dev_capability = NETIF_F_IP_CSUM + | NETIF_F_IPV6_CSUM + | NETIF_F_GSO_UDP_TUNNEL + | NETIF_F_HW_CSUM | NETIF_F_SG + | NETIF_F_RXCSUM + | NETIF_F_TSO | NETIF_F_TSO6 + | NETIF_F_LRO; + + netdev->hw_enc_features = (lio->enc_dev_capability & + ~NETIF_F_LRO); + + lio->dev_capability |= NETIF_F_GSO_UDP_TUNNEL; + netdev->vlan_features = lio->dev_capability; /* Add any unchangeable hw features */ lio->dev_capability |= NETIF_F_HW_VLAN_CTAG_FILTER | @@ -3561,6 +3750,15 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) ifstate_set(lio, LIO_IFSTATE_REGISTERED); + /* Sending command to firmware to enable Rx checksum offload + * by default at the time of setup of Liquidio driver for + * this device + */ + liquidio_set_rxcsum_command(netdev, OCTNET_CMD_TNL_RX_CSUM_CTL, + OCTNET_CMD_RXCSUM_ENABLE); + liquidio_set_feature(netdev, OCTNET_CMD_TNL_TX_CSUM_CTL, + OCTNET_CMD_TXCSUM_ENABLE); + dev_dbg(&octeon_dev->pci_dev->dev, "NIC ifidx:%d Setup successful\n", i); diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 5aa01f4..c6e3d57 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -216,6 +216,13 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry, #define OCTNET_CMD_ENABLE_VLAN_FILTER 0x16 #define OCTNET_CMD_ADD_VLAN_FILTER 0x17 #define OCTNET_CMD_DEL_VLAN_FILTER 0x18 +#define OCTNET_CMD_VXLAN_PORT_CONFIG 0x19 +#define OCTNET_CMD_VXLAN_PORT_ADD 0x0 +#define OCTNET_CMD_VXLAN_PORT_DEL 0x1 +#define OCTNET_CMD_RXCSUM_ENABLE 0x0 +#define OCTNET_CMD_RXCSUM_DISABLE 0x1 +#define OCTNET_CMD_TXCSUM_ENABLE 0x0 +#define OCTNET_CMD_TXCSUM_DISABLE 0x1 /* RX(packets coming from wire) Checksum verification flags */ /* TCP/UDP csum */ @@ -533,6 +540,7 @@ union octeon_rh { u64 priority:3; u64 csum_verified:3; /** checksum verified. */ u64 has_hwtstamp:1; /** Has hardware timestamp. 1 = yes. */ + u64 encap_on:1; } r_dh; struct { u64 opcode:4; @@ -562,6 +570,7 @@ union octeon_rh { u64 opcode:4; } r; struct { + u64 encap_on:1; u64 has_hwtstamp:1; /** 1 = has hwtstamp */ u64 csum_verified:3; /** checksum verified. */ u64 priority:3; @@ -736,6 +745,8 @@ struct nic_rx_stats { u64 fw_err_pko; u64 fw_err_link; u64 fw_err_drop; + u64 fw_rx_vxlan; + u64 fw_rx_vxlan_err; /* LRO */ u64 fw_lro_pkts; /* Number of packets that are LROed */ @@ -776,6 +787,7 @@ struct nic_tx_stats { u64 fw_err_tso; u64 fw_tso; /* number of tso requests */ u64 fw_tso_fwd; /* number of packets segmented in tso */ + u64 fw_tx_vxlan; }; struct oct_link_stats { diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h index 1ca9c4f..886772c 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h @@ -121,6 +121,9 @@ struct oct_droq_stats { /** Num of Packets dropped due to receive path failures. */ u64 rx_dropped; + /** Num of vxlan packets received; */ + u64 rx_vxlan; + /** Num of failures of recv_buffer_alloc() */ u64 rx_alloc_failure; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h index caa2b4f..5ac7e66 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h @@ -66,6 +66,7 @@ struct oct_iq_stats { u64 tx_dropped;/**< Numof pkts dropped dueto xmitpath errors. */ u64 tx_tot_bytes;/**< Total count of bytes sento to network. */ u64 tx_gso; /* count of tso */ + u64 tx_vxlan; /* tunnel */ u64 tx_dmamap_fail; u64 tx_restart; /*u64 tx_timeout_count;*/ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h index b481edc..24bcbd4 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h @@ -96,6 +96,12 @@ struct lio { /** Copy of Interface capabilities: TSO, TSO6, LRO, Chescksums . */ u64 dev_capability; + /* Copy of transmit encapsulation capabilities: + * TSO, TSO6, Checksums for this device for Kernel + * 3.10.0 onwards + */ + u64 enc_dev_capability; + /** Copy of beacaon reg in phy */ u32 phy_beacon_val; -- cgit v0.10.2 From a7d5a3dcf71fff3d03e6c2dd3981882a70e78b46 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Sun, 3 Jul 2016 13:56:48 -0700 Subject: liquidio: Macro replacements This patch has minor replacements of ACCESS_ONCE macros with WRITE_ONCE and replacement of BUG_ON with polite version WARN_ON. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c index d35864a..ee40b47 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c +++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c @@ -19,26 +19,16 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include -#include -#include -#include #include -#include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" -#include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "liquidio_image.h" -#include "octeon_mem_ops.h" int lio_cn6xxx_soft_reset(struct octeon_device *oct) { @@ -547,14 +537,14 @@ static void lio_cn6xxx_get_pcie_qlmport(struct octeon_device *oct) dev_dbg(&oct->pci_dev->dev, "Using PCIE Port %d\n", oct->pcie_port); } -void +static void lio_cn6xxx_process_pcie_error_intr(struct octeon_device *oct, u64 intr64) { dev_err(&oct->pci_dev->dev, "Error Intr: 0x%016llx\n", CVM_CAST64(intr64)); } -int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct) +static int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct) { struct octeon_droq *droq; int oq_no; @@ -579,7 +569,7 @@ int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct) continue; droq = oct->droq[oq_no]; - pkt_count = octeon_droq_check_hw_for_pkts(oct, droq); + pkt_count = octeon_droq_check_hw_for_pkts(droq); if (pkt_count) { oct->droq_intr |= (1ULL << oq_no); if (droq->ops.poll_mode) { diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h index fe2932c..28c4722 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h +++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h @@ -82,8 +82,6 @@ void lio_cn6xxx_setup_iq_regs(struct octeon_device *oct, u32 iq_no); void lio_cn6xxx_setup_oq_regs(struct octeon_device *oct, u32 oq_no); void lio_cn6xxx_enable_io_queues(struct octeon_device *oct); void lio_cn6xxx_disable_io_queues(struct octeon_device *oct); -void lio_cn6xxx_process_pcie_error_intr(struct octeon_device *oct, u64 intr64); -int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct); irqreturn_t lio_cn6xxx_process_interrupt_regs(void *dev); void lio_cn6xxx_reinit_regs(struct octeon_device *oct); void lio_cn6xxx_bar1_idx_setup(struct octeon_device *oct, u64 core_addr, diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c index 8e830d0..29755bc 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c +++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c @@ -19,28 +19,17 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include -#include -#include -#include #include -#include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" -#include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" #include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" -#include "octeon_mem_ops.h" static void lio_cn68xx_set_dpi_regs(struct octeon_device *oct) { @@ -129,7 +118,7 @@ static inline void lio_cn68xx_vendor_message_fix(struct octeon_device *oct) pci_write_config_dword(oct->pci_dev, CN6XXX_PCIE_FLTMSK, val); } -int lio_is_210nv(struct octeon_device *oct) +static int lio_is_210nv(struct octeon_device *oct) { u64 mio_qlm4_cfg = lio_pci_readq(oct, CN6XXX_MIO_QLM4_CFG); diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h index d4e1c9f..ea7bdcc 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h +++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h @@ -28,6 +28,5 @@ #define __CN68XX_DEVICE_H__ int lio_setup_cn68xx_octeon_device(struct octeon_device *oct); -int lio_is_210nv(struct octeon_device *oct); #endif diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index a060586..0f29dc4 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -19,13 +19,9 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include #include #include -#include -#include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" @@ -36,9 +32,6 @@ #include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" static int octnet_get_link_stats(struct net_device *netdev); @@ -346,20 +339,18 @@ static void octnet_mdio_resp_callback(struct octeon_device *oct, u32 status, void *buf) { - struct oct_mdio_cmd_resp *mdio_cmd_rsp; struct oct_mdio_cmd_context *mdio_cmd_ctx; struct octeon_soft_command *sc = (struct octeon_soft_command *)buf; - mdio_cmd_rsp = (struct oct_mdio_cmd_resp *)sc->virtrptr; mdio_cmd_ctx = (struct oct_mdio_cmd_context *)sc->ctxptr; oct = lio_get_device(mdio_cmd_ctx->octeon_id); if (status) { dev_err(&oct->pci_dev->dev, "MIDO instruction failed. Status: %llx\n", CVM_CAST64(status)); - ACCESS_ONCE(mdio_cmd_ctx->cond) = -1; + WRITE_ONCE(mdio_cmd_ctx->cond, -1); } else { - ACCESS_ONCE(mdio_cmd_ctx->cond) = 1; + WRITE_ONCE(mdio_cmd_ctx->cond, 1); } wake_up_interruptible(&mdio_cmd_ctx->wc); } @@ -390,7 +381,7 @@ octnet_mdio45_access(struct lio *lio, int op, int loc, int *value) mdio_cmd_rsp = (struct oct_mdio_cmd_resp *)sc->virtrptr; mdio_cmd = (struct oct_mdio_cmd *)sc->virtdptr; - ACCESS_ONCE(mdio_cmd_ctx->cond) = 0; + WRITE_ONCE(mdio_cmd_ctx->cond, 0); mdio_cmd_ctx->octeon_id = lio_get_device_id(oct_dev); mdio_cmd->op = op; mdio_cmd->mdio_addr = loc; @@ -429,7 +420,7 @@ octnet_mdio45_access(struct lio *lio, int op, int loc, int *value) octeon_swap_8B_data((u64 *)(&mdio_cmd_rsp->resp), sizeof(struct oct_mdio_cmd) / 8); - if (ACCESS_ONCE(mdio_cmd_ctx->cond) == 1) { + if (READ_ONCE(mdio_cmd_ctx->cond) == 1) { if (!op) *value = mdio_cmd_rsp->resp.value1; } else { @@ -623,7 +614,8 @@ lio_get_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) static void lio_get_ethtool_stats(struct net_device *netdev, - struct ethtool_stats *stats, u64 *data) + struct ethtool_stats *stats __attribute__((unused)), + u64 *data) { struct lio *lio = GET_LIO(netdev); struct octeon_device *oct_dev = lio->oct_dev; @@ -1067,7 +1059,7 @@ static int octnet_set_intrmod_cfg(struct lio *lio, return 0; } -void +static void octnet_nic_stats_callback(struct octeon_device *oct_dev, u32 status, void *ptr) { @@ -1552,7 +1544,7 @@ static int lio_nway_reset(struct net_device *netdev) } /* Return register dump len. */ -static int lio_get_regs_len(struct net_device *dev) +static int lio_get_regs_len(struct net_device *dev __attribute__((unused))) { return OCT_ETHTOOL_REGDUMP_LEN; } diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 4a5629f..3d229c0 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -20,25 +20,12 @@ * Contact Cavium, Inc. for more information **********************************************************************/ #include -#include -#include -#include #include -#include -#include -#include -#include #include #include #include -#include #include -#include -#include -#include -#include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" @@ -49,7 +36,6 @@ #include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "cn68xx_regs.h" #include "cn68xx_device.h" #include "liquidio_image.h" @@ -252,8 +238,7 @@ static int lio_wait_for_oq_pkts(struct octeon_device *oct) for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { if (!(oct->io_qmask.oq & (1ULL << i))) continue; - pkt_cnt += octeon_droq_check_hw_for_pkts(oct, - oct->droq[i]); + pkt_cnt += octeon_droq_check_hw_for_pkts(oct->droq[i]); } if (pkt_cnt > 0) { pending_pkts += pkt_cnt; @@ -508,7 +493,8 @@ static pci_ers_result_t liquidio_pcie_error_detected(struct pci_dev *pdev, * \brief mmio handler * @param pdev Pointer to PCI device */ -static pci_ers_result_t liquidio_pcie_mmio_enabled(struct pci_dev *pdev) +static pci_ers_result_t liquidio_pcie_mmio_enabled( + struct pci_dev *pdev __attribute__((unused))) { /* We should never hit this since we never ask for a reset for a Fatal * Error. We always return DISCONNECT in io_error above. @@ -524,7 +510,8 @@ static pci_ers_result_t liquidio_pcie_mmio_enabled(struct pci_dev *pdev) * Restart the card from scratch, as if from a cold-boot. Implementation * resembles the first-half of the octeon_resume routine. */ -static pci_ers_result_t liquidio_pcie_slot_reset(struct pci_dev *pdev) +static pci_ers_result_t liquidio_pcie_slot_reset( + struct pci_dev *pdev __attribute__((unused))) { /* We should never hit this since we never ask for a reset for a Fatal * Error. We always return DISCONNECT in io_error above. @@ -541,7 +528,7 @@ static pci_ers_result_t liquidio_pcie_slot_reset(struct pci_dev *pdev) * its OK to resume normal operation. Implementation resembles the * second-half of the octeon_resume routine. */ -static void liquidio_pcie_resume(struct pci_dev *pdev) +static void liquidio_pcie_resume(struct pci_dev *pdev __attribute__((unused))) { /* Nothing to be done here. */ } @@ -552,7 +539,8 @@ static void liquidio_pcie_resume(struct pci_dev *pdev) * @param pdev Pointer to PCI device * @param state state to suspend to */ -static int liquidio_suspend(struct pci_dev *pdev, pm_message_t state) +static int liquidio_suspend(struct pci_dev *pdev __attribute__((unused)), + pm_message_t state __attribute__((unused))) { return 0; } @@ -561,7 +549,7 @@ static int liquidio_suspend(struct pci_dev *pdev, pm_message_t state) * \brief called when resuming * @param pdev Pointer to PCI device */ -static int liquidio_resume(struct pci_dev *pdev) +static int liquidio_resume(struct pci_dev *pdev __attribute__((unused))) { return 0; } @@ -1105,7 +1093,9 @@ static int octeon_setup_interrupt(struct octeon_device *oct) * @param pdev PCI device structure * @param ent unused */ -static int liquidio_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +static int +liquidio_probe(struct pci_dev *pdev, + const struct pci_device_id *ent __attribute__((unused))) { struct octeon_device *oct_dev = NULL; struct handshake *hs; @@ -1725,8 +1715,10 @@ static int liquidio_ptp_settime(struct ptp_clock_info *ptp, * @param rq request * @param on is it on */ -static int liquidio_ptp_enable(struct ptp_clock_info *ptp, - struct ptp_clock_request *rq, int on) +static int +liquidio_ptp_enable(struct ptp_clock_info *ptp __attribute__((unused)), + struct ptp_clock_request *rq __attribute__((unused)), + int on __attribute__((unused))) { return -EOPNOTSUPP; } @@ -1867,7 +1859,7 @@ static int octeon_setup_droq(struct octeon_device *oct, int q_no, int num_descs, * @param buf pointer to resp structure */ static void if_cfg_callback(struct octeon_device *oct, - u32 status, + u32 status __attribute__((unused)), void *buf) { struct octeon_soft_command *sc = (struct octeon_soft_command *)buf; @@ -1881,7 +1873,7 @@ static void if_cfg_callback(struct octeon_device *oct, if (resp->status) dev_err(&oct->pci_dev->dev, "nic if cfg instruction failed. Status: %llx\n", CVM_CAST64(resp->status)); - ACCESS_ONCE(ctx->cond) = 1; + WRITE_ONCE(ctx->cond, 1); snprintf(oct->fw_info.liquidio_firmware_version, 32, "%s", resp->cfg_info.liquidio_firmware_version); @@ -1901,7 +1893,8 @@ static void if_cfg_callback(struct octeon_device *oct, * @returns selected queue number */ static u16 select_q(struct net_device *dev, struct sk_buff *skb, - void *accel_priv, select_queue_fallback_t fallback) + void *accel_priv __attribute__((unused)), + select_queue_fallback_t fallback __attribute__((unused))) { u32 qindex = 0; struct lio *lio; @@ -1921,7 +1914,7 @@ static u16 select_q(struct net_device *dev, struct sk_buff *skb, * @param arg - farg registered in droq_ops */ static void -liquidio_push_packet(u32 octeon_id, +liquidio_push_packet(u32 octeon_id __attribute__((unused)), void *skbuff, u32 len, union octeon_rh *rh, @@ -2526,7 +2519,7 @@ static void liquidio_set_mcast_list(struct net_device *netdev) struct octnic_ctrl_pkt nctrl; struct netdev_hw_addr *ha; u64 *mc; - int ret, i; + int ret; int mc_count = min(netdev_mc_count(netdev), MAX_OCTEON_MULTICAST_ADDR); memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt)); @@ -2542,7 +2535,6 @@ static void liquidio_set_mcast_list(struct net_device *netdev) nctrl.cb_fn = liquidio_link_ctrl_cmd_completion; /* copy all the addresses into the udd */ - i = 0; mc = &nctrl.udd[0]; netdev_for_each_mc_addr(ha, netdev) { *mc = 0; @@ -2707,7 +2699,7 @@ static int liquidio_change_mtu(struct net_device *netdev, int new_mtu) * @param ifr interface request * @param cmd command */ -static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr) { struct hwtstamp_config conf; struct lio *lio = GET_LIO(netdev); @@ -2768,7 +2760,7 @@ static int liquidio_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { switch (cmd) { case SIOCSHWTSTAMP: - return hwtstamp_ioctl(netdev, ifr, cmd); + return hwtstamp_ioctl(netdev, ifr); default: return -EOPNOTSUPP; } @@ -3545,7 +3537,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) dev_dbg(&octeon_dev->pci_dev->dev, "requesting config for interface %d, iqs %d, oqs %d\n", ifidx_or_pfnum, num_iqueues, num_oqueues); - ACCESS_ONCE(ctx->cond) = 0; + WRITE_ONCE(ctx->cond, 0); ctx->octeon_id = lio_get_device_id(octeon_dev); init_waitqueue_head(&ctx->wc); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_console.c b/drivers/net/ethernet/cavium/liquidio/octeon_console.c index 466147e..bceff8a 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_console.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_console.c @@ -23,27 +23,14 @@ /** * @file octeon_console.c */ -#include -#include -#include -#include #include -#include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" -#include "octeon_network.h" -#include "cn66xx_regs.h" -#include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" #include "octeon_mem_ops.h" static void octeon_remote_lock(void); @@ -51,6 +38,8 @@ static void octeon_remote_unlock(void); static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct, const char *name, u32 flags); +static int octeon_console_read(struct octeon_device *oct, u32 console_num, + char *buffer, u32 buf_size); #define MIN(a, b) min((a), (b)) #define CAST_ULL(v) ((u64)(v)) @@ -170,8 +159,8 @@ struct octeon_pci_console_desc { offsetof(struct cvmx_bootmem_desc, field), \ SIZEOF_FIELD(struct cvmx_bootmem_desc, field)) -#define __cvmx_bootmem_lock(flags) -#define __cvmx_bootmem_unlock(flags) +#define __cvmx_bootmem_lock(flags) (flags = flags) +#define __cvmx_bootmem_unlock(flags) (flags = flags) /** * This macro returns a member of the @@ -440,8 +429,7 @@ int octeon_wait_for_bootloader(struct octeon_device *oct, } static void octeon_console_handle_result(struct octeon_device *oct, - size_t console_num, - char *buffer, s32 bytes_read) + size_t console_num) { struct octeon_console *console; @@ -505,14 +493,11 @@ static void check_console(struct work_struct *work) */ bytes_read = octeon_console_read(oct, console_num, console_buffer, - sizeof(console_buffer) - 1, 0); + sizeof(console_buffer) - 1); if (bytes_read > 0) { total_read += bytes_read; - if (console->waiting) { - octeon_console_handle_result(oct, console_num, - console_buffer, - bytes_read); - } + if (console->waiting) + octeon_console_handle_result(oct, console_num); if (octeon_console_debug_enabled(console_num)) { output_console_line(oct, console, console_num, console_buffer, bytes_read); @@ -675,8 +660,8 @@ static inline int octeon_console_avail_bytes(u32 buffer_size, octeon_console_free_bytes(buffer_size, wr_idx, rd_idx); } -int octeon_console_read(struct octeon_device *oct, u32 console_num, - char *buffer, u32 buf_size, u32 flags) +static int octeon_console_read(struct octeon_device *oct, u32 console_num, + char *buffer, u32 buf_size) { int bytes_to_read; u32 rd_idx, wr_idx; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index 3372207..046f096 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -19,27 +19,19 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include -#include -#include #include #include -#include #include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" #include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" #include "liquidio_image.h" #include "octeon_mem_ops.h" @@ -752,13 +744,11 @@ struct octeon_device *octeon_allocate_device(u32 pci_id, /* this function is only for setting up the first queue */ int octeon_setup_instr_queues(struct octeon_device *oct) { - u32 num_iqs = 0; u32 num_descs = 0; u32 iq_no = 0; union oct_txpciq txpciq; int numa_node = cpu_to_node(iq_no % num_online_cpus()); - num_iqs = 1; /* this causes queue 0 to be default queue */ if (OCTEON_CN6XXX(oct)) num_descs = @@ -793,13 +783,11 @@ int octeon_setup_instr_queues(struct octeon_device *oct) int octeon_setup_output_queues(struct octeon_device *oct) { - u32 num_oqs = 0; u32 num_descs = 0; u32 desc_size = 0; u32 oq_no = 0; int numa_node = cpu_to_node(oq_no % num_online_cpus()); - num_oqs = 1; /* this causes queue 0 to be default queue */ if (OCTEON_CN6XXX(oct)) { num_descs = @@ -1019,79 +1007,6 @@ octeon_register_dispatch_fn(struct octeon_device *oct, return 0; } -/* octeon_unregister_dispatch_fn - * Parameters: - * oct - octeon device - * opcode - driver should unregister the function for this opcode - * subcode - driver should unregister the function for this subcode - * Description: - * Unregister the function set for this opcode+subcode. - * Returns: - * Success: 0 - * Failure: 1 - * Locks: - * No locks are held. - */ -int -octeon_unregister_dispatch_fn(struct octeon_device *oct, u16 opcode, - u16 subcode) -{ - int retval = 0; - u32 idx; - struct list_head *dispatch, *dfree = NULL, *tmp2; - u16 combined_opcode = OPCODE_SUBCODE(opcode, subcode); - - idx = combined_opcode & OCTEON_OPCODE_MASK; - - spin_lock_bh(&oct->dispatch.lock); - - if (oct->dispatch.count == 0) { - spin_unlock_bh(&oct->dispatch.lock); - dev_err(&oct->pci_dev->dev, - "No dispatch functions registered for this device\n"); - return 1; - } - - if (oct->dispatch.dlist[idx].opcode == combined_opcode) { - dispatch = &oct->dispatch.dlist[idx].list; - if (dispatch->next != dispatch) { - dispatch = dispatch->next; - oct->dispatch.dlist[idx].opcode = - ((struct octeon_dispatch *)dispatch)->opcode; - oct->dispatch.dlist[idx].dispatch_fn = - ((struct octeon_dispatch *) - dispatch)->dispatch_fn; - oct->dispatch.dlist[idx].arg = - ((struct octeon_dispatch *)dispatch)->arg; - list_del(dispatch); - dfree = dispatch; - } else { - oct->dispatch.dlist[idx].opcode = 0; - oct->dispatch.dlist[idx].dispatch_fn = NULL; - oct->dispatch.dlist[idx].arg = NULL; - } - } else { - retval = 1; - list_for_each_safe(dispatch, tmp2, - &(oct->dispatch.dlist[idx]. - list)) { - if (((struct octeon_dispatch *)dispatch)->opcode == - combined_opcode) { - list_del(dispatch); - dfree = dispatch; - retval = 0; - } - } - } - - if (!retval) - oct->dispatch.count--; - - spin_unlock_bh(&oct->dispatch.lock); - vfree(dfree); - return retval; -} - int octeon_core_drv_init(struct octeon_recv_info *recv_info, void *buf) { u32 i; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h index b4e566d..579acd4 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h @@ -585,8 +585,7 @@ int octeon_add_console(struct octeon_device *oct, u32 console_num); int octeon_console_write(struct octeon_device *oct, u32 console_num, char *buffer, u32 write_request_size, u32 flags); int octeon_console_write_avail(struct octeon_device *oct, u32 console_num); -int octeon_console_read(struct octeon_device *oct, u32 console_num, - char *buffer, u32 buf_size, u32 flags); + int octeon_console_read_avail(struct octeon_device *oct, u32 console_num); /** Removes all attached consoles. */ diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c index d9bb2f7..8bfbe8c 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c @@ -19,30 +19,18 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include -#include -#include #include -#include #include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" #include "octeon_network.h" #include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" -#include "octeon_mem_ops.h" - -/* #define CAVIUM_ONLY_PERF_MODE */ #define CVM_MIN(d1, d2) (((d1) < (d2)) ? (d1) : (d2)) #define CVM_MAX(d1, d2) (((d1) > (d2)) ? (d1) : (d2)) @@ -104,8 +92,7 @@ static inline void *octeon_get_dispatch_arg(struct octeon_device *octeon_dev, return fn_arg; } -u32 octeon_droq_check_hw_for_pkts(struct octeon_device *oct, - struct octeon_droq *droq) +u32 octeon_droq_check_hw_for_pkts(struct octeon_droq *droq) { u32 pkt_count = 0; @@ -810,7 +797,7 @@ octeon_droq_process_poll_pkts(struct octeon_device *oct, total_pkts_processed += pkts_processed; - octeon_droq_check_hw_for_pkts(oct, droq); + octeon_droq_check_hw_for_pkts(droq); } spin_unlock(&droq->lock); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h index 886772c..5a6fb91 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h @@ -416,24 +416,9 @@ int octeon_register_dispatch_fn(struct octeon_device *oct, u16 subcode, octeon_dispatch_fn_t fn, void *fn_arg); -/** Remove registration for an opcode/subcode. This will delete the mapping for - * an opcode/subcode. The dispatch function will be unregistered and will no - * longer be called if a packet with the opcode/subcode arrives in the driver - * output queues. - * @param oct - the octeon device to unregister from. - * @param opcode - the opcode to be unregistered. - * @param subcode - the subcode to be unregistered. - * - * @return Success: 0; Failure: 1 - */ -int octeon_unregister_dispatch_fn(struct octeon_device *oct, - u16 opcode, - u16 subcode); - void octeon_droq_print_stats(void); -u32 octeon_droq_check_hw_for_pkts(struct octeon_device *oct, - struct octeon_droq *droq); +u32 octeon_droq_check_hw_for_pkts(struct octeon_droq *droq); int octeon_create_droq(struct octeon_device *oct, u32 q_no, u32 num_descs, u32 desc_size, void *app_ctx); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_main.h b/drivers/net/ethernet/cavium/liquidio/octeon_main.h index 0ff3efc..bc14e4c 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_main.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_main.h @@ -174,7 +174,7 @@ sleep_cond(wait_queue_head_t *wait_queue, int *condition) init_waitqueue_entry(&we, current); add_wait_queue(wait_queue, &we); - while (!(ACCESS_ONCE(*condition))) { + while (!(READ_ONCE(*condition))) { set_current_state(TASK_INTERRUPTIBLE); if (signal_pending(current)) goto out; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c index 5aecef8..95a4bbe 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c @@ -19,43 +19,29 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include -#include -#include -#include #include -#include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" -#include "octeon_main.h" -#include "octeon_network.h" -#include "cn66xx_regs.h" -#include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" -#include "octeon_mem_ops.h" #define MEMOPS_IDX MAX_BAR1_MAP_INDEX +#ifdef __BIG_ENDIAN_BITFIELD static inline void -octeon_toggle_bar1_swapmode(struct octeon_device *oct __attribute__((unused)), - u32 idx __attribute__((unused))) +octeon_toggle_bar1_swapmode(struct octeon_device *oct, u32 idx) { -#ifdef __BIG_ENDIAN_BITFIELD u32 mask; mask = oct->fn_list.bar1_idx_read(oct, idx); mask = (mask & 0x2) ? (mask & ~2) : (mask | 2); oct->fn_list.bar1_idx_write(oct, idx, mask); -#endif } +#else +#define octeon_toggle_bar1_swapmode(oct, idx) (oct = oct) +#endif static void octeon_pci_fastwrite(struct octeon_device *oct, u8 __iomem *mapped_addr, diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h index 24bcbd4..5d89f55 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h @@ -357,7 +357,7 @@ lio_map_ring_info(struct octeon_droq *droq, u32 i) dma_addr = dma_map_single(&oct->pci_dev->dev, &droq->info_list[i], OCT_DROQ_INFO_SIZE, DMA_FROM_DEVICE); - BUG_ON(dma_mapping_error(&oct->pci_dev->dev, dma_addr)); + WARN_ON(dma_mapping_error(&oct->pci_dev->dev, dma_addr)); return (u64)dma_addr; } diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c index 36f1970..166727b 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c @@ -19,14 +19,9 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include -#include -#include #include #include -#include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" @@ -34,13 +29,6 @@ #include "octeon_device.h" #include "octeon_nic.h" #include "octeon_main.h" -#include "octeon_network.h" -#include "cn66xx_regs.h" -#include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" -#include "octeon_mem_ops.h" void * octeon_alloc_soft_command_resp(struct octeon_device *oct, diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index 7eafa75..5e2211f 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -19,28 +19,17 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include -#include -#include -#include #include -#include #include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" #include "octeon_network.h" -#include "cn66xx_regs.h" #include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" #define INCR_INSTRQUEUE_PKT_COUNT(octeon_dev_ptr, iq_no, field, count) \ (octeon_dev_ptr->instr_queue[iq_no]->stats.field += count) @@ -301,40 +290,8 @@ static inline void __copy_cmd_into_iq(struct octeon_instr_queue *iq, memcpy(iqptr, cmd, cmdsize); } -static inline int -__post_command(struct octeon_device *octeon_dev __attribute__((unused)), - struct octeon_instr_queue *iq, - u32 force_db __attribute__((unused)), u8 *cmd) -{ - u32 index = -1; - - /* This ensures that the read index does not wrap around to the same - * position if queue gets full before Octeon could fetch any instr. - */ - if (atomic_read(&iq->instr_pending) >= (s32)(iq->max_count - 1)) - return -1; - - __copy_cmd_into_iq(iq, cmd); - - /* "index" is returned, host_write_index is modified. */ - index = iq->host_write_index; - INCR_INDEX_BY1(iq->host_write_index, iq->max_count); - iq->fill_cnt++; - - /* Flush the command into memory. We need to be sure the data is in - * memory before indicating that the instruction is pending. - */ - wmb(); - - atomic_inc(&iq->instr_pending); - - return index; -} - static inline struct iq_post_status -__post_command2(struct octeon_device *octeon_dev __attribute__((unused)), - struct octeon_instr_queue *iq, - u32 force_db __attribute__((unused)), u8 *cmd) +__post_command2(struct octeon_instr_queue *iq, u8 *cmd) { struct iq_post_status st; @@ -579,7 +536,7 @@ octeon_send_command(struct octeon_device *oct, u32 iq_no, */ spin_lock_bh(&iq->post_lock); - st = __post_command2(oct, iq, force_db, cmd); + st = __post_command2(iq, cmd); if (st.status != IQ_SEND_FAILED) { octeon_report_sent_bytes_to_bql(buf, reqtype); @@ -618,8 +575,8 @@ octeon_prepare_soft_command(struct octeon_device *oct, struct octeon_instr_irh *irh; struct octeon_instr_rdp *rdp; - BUG_ON(opcode > 15); - BUG_ON(subcode > 127); + WARN_ON(opcode > 15); + WARN_ON(subcode > 127); oct_cfg = octeon_get_conf(oct); @@ -661,7 +618,6 @@ int octeon_send_soft_command(struct octeon_device *oct, { struct octeon_instr_ih2 *ih2; struct octeon_instr_irh *irh; - struct octeon_instr_rdp *rdp; u32 len; ih2 = (struct octeon_instr_ih2 *)&sc->cmd.cmd2.ih2; @@ -671,12 +627,10 @@ int octeon_send_soft_command(struct octeon_device *oct, } irh = (struct octeon_instr_irh *)&sc->cmd.cmd2.irh; if (irh->rflag) { - BUG_ON(!sc->dmarptr); - BUG_ON(!sc->status_word); + WARN_ON(!sc->dmarptr); + WARN_ON(!sc->status_word); *sc->status_word = COMPLETION_WORD_INIT; - rdp = (struct octeon_instr_rdp *)&sc->cmd.cmd2.rdp; - sc->cmd.cmd2.rptr = sc->dmarptr; } len = (u32)ih2->dlengsz; @@ -748,7 +702,7 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct, struct octeon_soft_command *sc = NULL; struct list_head *tmp; - BUG_ON((offset + datasize + rdatasize + ctxsize) > + WARN_ON((offset + datasize + rdatasize + ctxsize) > SOFT_COMMAND_BUFFER_SIZE); spin_lock(&oct->sc_buf_pool.lock); @@ -795,7 +749,7 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct, offset = (offset + datasize + 127) & 0xffffff80; if (rdatasize) { - BUG_ON(rdatasize < 16); + WARN_ON(rdatasize < 16); sc->virtrptr = (u8 *)sc + offset; sc->dmarptr = dma_addr + offset; sc->rdatasize = rdatasize; diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.c b/drivers/net/ethernet/cavium/liquidio/response_manager.c index c93210f..0cc2f66 100644 --- a/drivers/net/ethernet/cavium/liquidio/response_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/response_manager.c @@ -19,28 +19,14 @@ * This file may also be available under a different license from Cavium. * Contact Cavium, Inc. for more information **********************************************************************/ -#include -#include -#include -#include -#include #include -#include #include -#include "octeon_config.h" #include "liquidio_common.h" #include "octeon_droq.h" #include "octeon_iq.h" #include "response_manager.h" #include "octeon_device.h" -#include "octeon_nic.h" #include "octeon_main.h" -#include "octeon_network.h" -#include "cn66xx_regs.h" -#include "cn66xx_device.h" -#include "cn68xx_regs.h" -#include "cn68xx_device.h" -#include "liquidio_image.h" static void oct_poll_req_completion(struct work_struct *work); -- cgit v0.10.2 From 14866ccd8fbb9b1ace953a7fad719b050763426a Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Sun, 3 Jul 2016 13:56:49 -0700 Subject: liquidio: IQ synchronization This patch tries to protect against bh preemption with sc_buf_pool. It also modifies the syncronization primitives during input queue processing. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index 5e2211f..ef0bdd8 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -360,6 +360,7 @@ lio_process_iq_request_list(struct octeon_device *oct, unsigned int pkts_compl = 0, bytes_compl = 0; struct octeon_soft_command *sc; struct octeon_instr_irh *irh; + unsigned long flags; while (old != iq->octeon_read_index) { reqtype = iq->request_list[old].reqtype; @@ -389,15 +390,19 @@ lio_process_iq_request_list(struct octeon_device *oct, * command response list because we expect * a response from Octeon. */ - spin_lock_bh(&oct->response_list - [OCTEON_ORDERED_SC_LIST].lock); + spin_lock_irqsave + (&oct->response_list + [OCTEON_ORDERED_SC_LIST].lock, + flags); atomic_inc(&oct->response_list [OCTEON_ORDERED_SC_LIST]. pending_req_count); list_add_tail(&sc->node, &oct->response_list [OCTEON_ORDERED_SC_LIST].head); - spin_unlock_bh(&oct->response_list - [OCTEON_ORDERED_SC_LIST].lock); + spin_unlock_irqrestore + (&oct->response_list + [OCTEON_ORDERED_SC_LIST].lock, + flags); } else { if (sc->callback) { sc->callback(oct, OCTEON_REQUEST_DONE, @@ -674,7 +679,7 @@ int octeon_free_sc_buffer_pool(struct octeon_device *oct) struct list_head *tmp, *tmp2; struct octeon_soft_command *sc; - spin_lock(&oct->sc_buf_pool.lock); + spin_lock_bh(&oct->sc_buf_pool.lock); list_for_each_safe(tmp, tmp2, &oct->sc_buf_pool.head) { list_del(tmp); @@ -686,7 +691,7 @@ int octeon_free_sc_buffer_pool(struct octeon_device *oct) INIT_LIST_HEAD(&oct->sc_buf_pool.head); - spin_unlock(&oct->sc_buf_pool.lock); + spin_unlock_bh(&oct->sc_buf_pool.lock); return 0; } @@ -705,10 +710,10 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct, WARN_ON((offset + datasize + rdatasize + ctxsize) > SOFT_COMMAND_BUFFER_SIZE); - spin_lock(&oct->sc_buf_pool.lock); + spin_lock_bh(&oct->sc_buf_pool.lock); if (list_empty(&oct->sc_buf_pool.head)) { - spin_unlock(&oct->sc_buf_pool.lock); + spin_unlock_bh(&oct->sc_buf_pool.lock); return NULL; } @@ -719,7 +724,7 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct, atomic_inc(&oct->sc_buf_pool.alloc_buf_count); - spin_unlock(&oct->sc_buf_pool.lock); + spin_unlock_bh(&oct->sc_buf_pool.lock); sc = (struct octeon_soft_command *)tmp; @@ -762,11 +767,11 @@ struct octeon_soft_command *octeon_alloc_soft_command(struct octeon_device *oct, void octeon_free_soft_command(struct octeon_device *oct, struct octeon_soft_command *sc) { - spin_lock(&oct->sc_buf_pool.lock); + spin_lock_bh(&oct->sc_buf_pool.lock); list_add_tail(&sc->node, &oct->sc_buf_pool.head); atomic_dec(&oct->sc_buf_pool.alloc_buf_count); - spin_unlock(&oct->sc_buf_pool.lock); + spin_unlock_bh(&oct->sc_buf_pool.lock); } -- cgit v0.10.2 From 55893a63d1495d7154f37f02754fb95390743cec Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Sun, 3 Jul 2016 13:56:50 -0700 Subject: liquidio: softcommand delay This patch updates the delay constant for softcommands in accrodance with new requirements. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 3d229c0..906d32e 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -3555,7 +3555,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) sc->callback = if_cfg_callback; sc->callback_arg = sc; - sc->wait_time = 1000; + sc->wait_time = 3000; retval = octeon_send_soft_command(octeon_dev, sc); if (retval == IQ_SEND_FAILED) { diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index ef0bdd8..3f7044c 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -523,9 +523,10 @@ static void check_db_timeout(struct work_struct *work) struct octeon_device *oct = (struct octeon_device *)wk->ctxptr; unsigned long iq_no = wk->ctxul; struct cavium_wq *db_wq = &oct->check_db_wq[iq_no]; + u32 delay = 10; __check_db_timeout(oct, iq_no); - queue_delayed_work(db_wq->wq, &db_wq->wk.work, msecs_to_jiffies(1)); + queue_delayed_work(db_wq->wq, &db_wq->wk.work, msecs_to_jiffies(delay)); } int diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.c b/drivers/net/ethernet/cavium/liquidio/response_manager.c index 0cc2f66..709049e 100644 --- a/drivers/net/ethernet/cavium/liquidio/response_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/response_manager.c @@ -52,7 +52,7 @@ int octeon_setup_response_list(struct octeon_device *oct) INIT_DELAYED_WORK(&cwq->wk.work, oct_poll_req_completion); cwq->wk.ctxptr = oct; oct->cmd_resp_state = OCT_DRV_ONLINE; - queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(100)); + queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(50)); return ret; } @@ -162,6 +162,5 @@ static void oct_poll_req_completion(struct work_struct *work) struct cavium_wq *cwq = &oct->dma_comp_wq; lio_process_ordered_list(oct, 0); - - queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(100)); + queue_delayed_work(cwq->wq, &cwq->wk.work, msecs_to_jiffies(50)); } -- cgit v0.10.2 From 3dcef2ca02f46c469bda7be8890d9f13aa0c7d69 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Sun, 3 Jul 2016 13:56:51 -0700 Subject: liquidio: iq/oq limits This patch removes the dependency of number of iq/oq's on number of cpus. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 906d32e..a759cc9 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -3488,7 +3488,6 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) struct liquidio_if_cfg_resp *resp; struct octdev_props *props; int retval, num_iqueues, num_oqueues; - int num_cpus = num_online_cpus(); union oct_nic_if_cfg if_cfg; unsigned int base_queue; unsigned int gmx_port_id; @@ -3530,10 +3529,7 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) gmx_port_id = CFG_GET_GMXID_NIC_IF(octeon_get_conf(octeon_dev), i); ifidx_or_pfnum = i; - if (num_iqueues > num_cpus) - num_iqueues = num_cpus; - if (num_oqueues > num_cpus) - num_oqueues = num_cpus; + dev_dbg(&octeon_dev->pci_dev->dev, "requesting config for interface %d, iqs %d, oqs %d\n", ifidx_or_pfnum, num_iqueues, num_oqueues); -- cgit v0.10.2 From 1e0d30fe2143ae8db342e62be67a12479c3c2a20 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Sun, 3 Jul 2016 13:56:52 -0700 Subject: liquidio: free resources during shutdown This patch fixes the issue of proper freeing of queue memory resources during free device. It also has fix for correct pcie error reporting. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c index ee40b47..6271c57 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c +++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c @@ -64,9 +64,9 @@ void lio_cn6xxx_enable_error_reporting(struct octeon_device *oct) u32 val; pci_read_config_dword(oct->pci_dev, CN6XXX_PCIE_DEVCTL, &val); - if (val & 0x000f0000) { + if (val & 0x000c0000) { dev_err(&oct->pci_dev->dev, "PCI-E Link error detected: 0x%08x\n", - val & 0x000f0000); + val & 0x000c0000); } val |= 0xf; /* Enable Link error reporting */ diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index a759cc9..e4e476f 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -3957,6 +3957,7 @@ static int octeon_device_init(struct octeon_device *octeon_dev) /* Release any previously allocated queues */ for (j = 0; j < octeon_dev->num_oqs; j++) octeon_delete_droq(octeon_dev, j); + return 1; } atomic_set(&octeon_dev->status, OCT_DEV_DROQ_INIT_DONE); @@ -3979,7 +3980,8 @@ static int octeon_device_init(struct octeon_device *octeon_dev) /* Setup the interrupt handler and record the INT SUM register address */ - octeon_setup_interrupt(octeon_dev); + if (octeon_setup_interrupt(octeon_dev)) + return 1; /* Enable Octeon device interrupts */ octeon_dev->fn_list.enable_interrupt(octeon_dev->chip); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_console.c b/drivers/net/ethernet/cavium/liquidio/octeon_console.c index bceff8a..70a8cd3 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_console.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_console.c @@ -312,6 +312,9 @@ static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct, if (name && named_size) { char *name_tmp = kmalloc(name_length + 1, GFP_KERNEL); + if (!name_tmp) + break; + CVMX_BOOTMEM_NAMED_GET_NAME(oct, named_addr, name_tmp, name_length); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index 046f096..fda93be 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -644,16 +644,16 @@ int octeon_download_firmware(struct octeon_device *oct, const u8 *data, void octeon_free_device_mem(struct octeon_device *oct) { - u32 i; + int i; for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) { - /* could check mask as well */ - vfree(oct->droq[i]); + if (oct->io_qmask.oq & (1ULL << i)) + vfree(oct->droq[i]); } for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) { - /* could check mask as well */ - vfree(oct->instr_queue[i]); + if (oct->io_qmask.iq & (1ULL << i)) + vfree(oct->instr_queue[i]); } i = oct->octeon_id; -- cgit v0.10.2 From 4c2743f9ac00e6a1c9b61ee59d9648ad9c83d1a8 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Sun, 3 Jul 2016 13:56:53 -0700 Subject: liquidio: MTU limits This patch limits the MTU between 68 bytes and 16000 bytes. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 0f29dc4..82f20c9 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -554,7 +554,7 @@ lio_ethtool_get_ringparam(struct net_device *netdev, tx_pending = CFG_GET_NUM_TX_DESCS_NIC_IF(conf6x, lio->ifidx); } - if (lio->mtu > OCTNET_DEFAULT_FRM_SIZE) { + if (lio->mtu > OCTNET_DEFAULT_FRM_SIZE - OCTNET_FRM_HEADER_SIZE) { ering->rx_pending = 0; ering->rx_max_pending = 0; ering->rx_mini_pending = 0; diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index e4e476f..c75734f 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -2365,11 +2365,14 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr) case OCTNET_CMD_CHANGE_MTU: /* If command is successful, change the MTU. */ netif_info(lio, probe, lio->netdev, " MTU Changed from %d to %d\n", - netdev->mtu, nctrl->ncmd.s.param2); + netdev->mtu, nctrl->ncmd.s.param1); dev_info(&oct->pci_dev->dev, "%s MTU Changed from %d to %d\n", netdev->name, netdev->mtu, - nctrl->ncmd.s.param2); - netdev->mtu = nctrl->ncmd.s.param2; + nctrl->ncmd.s.param1); + rtnl_lock(); + netdev->mtu = nctrl->ncmd.s.param1; + call_netdevice_notifiers(NETDEV_CHANGEMTU, netdev); + rtnl_unlock(); break; case OCTNET_CMD_GPIO_ACCESS: @@ -2657,18 +2660,16 @@ static int liquidio_change_mtu(struct net_device *netdev, int new_mtu) struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; struct octnic_ctrl_pkt nctrl; - int max_frm_size = new_mtu + OCTNET_FRM_HEADER_SIZE; int ret = 0; - /* Limit the MTU to make sure the ethernet packets are between 64 bytes - * and 65535 bytes + /* Limit the MTU to make sure the ethernet packets are between 68 bytes + * and 16000 bytes */ - if ((max_frm_size < OCTNET_MIN_FRM_SIZE) || - (max_frm_size > OCTNET_MAX_FRM_SIZE)) { + if ((new_mtu < LIO_MIN_MTU_SIZE) || + (new_mtu > LIO_MAX_MTU_SIZE)) { dev_err(&oct->pci_dev->dev, "Invalid MTU: %d\n", new_mtu); dev_err(&oct->pci_dev->dev, "Valid range %d and %d\n", - (OCTNET_MIN_FRM_SIZE - OCTNET_FRM_HEADER_SIZE), - (OCTNET_MAX_FRM_SIZE - OCTNET_FRM_HEADER_SIZE)); + LIO_MIN_MTU_SIZE, LIO_MAX_MTU_SIZE); return -EINVAL; } diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h index 5d89f55..ab41866 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h @@ -30,6 +30,9 @@ #include #include +#define LIO_MAX_MTU_SIZE (OCTNET_MAX_FRM_SIZE - OCTNET_FRM_HEADER_SIZE) +#define LIO_MIN_MTU_SIZE 68 + struct oct_nic_stats_resp { u64 rh; struct oct_link_stats stats; -- cgit v0.10.2 From a22b15ce2751bfffa7b585bf426a5738d7a4d75d Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Sun, 3 Jul 2016 13:56:54 -0700 Subject: liquidio: Droq validation This patch removes redudant droq num validation. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c index 8bfbe8c..93ab4ec 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c @@ -821,18 +821,6 @@ octeon_process_droq_poll_cmd(struct octeon_device *oct, u32 q_no, int cmd, u32 arg) { struct octeon_droq *droq; - struct octeon_config *oct_cfg = NULL; - - oct_cfg = octeon_get_conf(oct); - - if (!oct_cfg) - return -EINVAL; - - if (q_no >= CFG_GET_OQ_MAX_Q(oct_cfg)) { - dev_err(&oct->pci_dev->dev, "%s: droq id (%d) exceeds MAX (%d)\n", - __func__, q_no, (oct->num_oqs - 1)); - return -EINVAL; - } droq = oct->droq[q_no]; -- cgit v0.10.2 From a2c64b67c6226f923ed6ee0b2f611675abc06686 Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Sun, 3 Jul 2016 13:56:55 -0700 Subject: liquidio: Remove redundant code This patch removes redundant file includes and conditions. Provides some meaningful comments and code alignment. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c index 6271c57..c03d370 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c +++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c @@ -219,7 +219,7 @@ void lio_cn6xxx_setup_global_output_regs(struct octeon_device *oct) /* / Select Packet count instead of bytes for SLI_PKTi_CNTS[CNT] */ octeon_write_csr(oct, CN6XXX_SLI_PKT_OUT_BMODE, 0); - /* / Select ES,RO,NS setting from register for Output Queue Packet + /* Select ES, RO, NS setting from register for Output Queue Packet * Address */ octeon_write_csr(oct, CN6XXX_SLI_PKT_DPADDR, 0xFFFFFFFF); diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h b/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h index 38cddbd..d45a0f4 100644 --- a/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h +++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h @@ -29,7 +29,6 @@ #ifndef __CN68XX_REGS_H__ #define __CN68XX_REGS_H__ -#include "cn66xx_regs.h" /*###################### REQUEST QUEUE #########################*/ diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c index 82f20c9..289eb89 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c @@ -464,18 +464,16 @@ static int lio_set_phys_id(struct net_device *netdev, /* Configure Beacon values */ value = LIO68XX_LED_BEACON_CFGON; - ret = - octnet_mdio45_access(lio, 1, - LIO68XX_LED_BEACON_ADDR, - &value); + ret = octnet_mdio45_access(lio, 1, + LIO68XX_LED_BEACON_ADDR, + &value); if (ret) return ret; value = LIO68XX_LED_CTRL_CFGON; - ret = - octnet_mdio45_access(lio, 1, - LIO68XX_LED_CTRL_ADDR, - &value); + ret = octnet_mdio45_access(lio, 1, + LIO68XX_LED_CTRL_ADDR, + &value); if (ret) return ret; } else { @@ -961,7 +959,6 @@ static int lio_get_intr_coalesce(struct net_device *netdev, intr_coal->rx_max_coalesced_frames = CFG_GET_OQ_INTR_PKT(cn6xxx->conf); } - iq = oct->instr_queue[lio->linfo.txpciq[0].s.q_no]; intr_coal->tx_max_coalesced_frames = iq->fill_threshold; break; @@ -1688,13 +1685,12 @@ static void lio_get_regs(struct net_device *dev, int len = 0; struct octeon_device *oct = lio->oct_dev; - memset(regbuf, 0, OCT_ETHTOOL_REGDUMP_LEN); regs->version = OCT_ETHTOOL_REGSVER; switch (oct->chip_id) { - /* case OCTEON_CN73XX: Todo */ case OCTEON_CN68XX: case OCTEON_CN66XX: + memset(regbuf, 0, OCT_ETHTOOL_REGDUMP_LEN); len += cn6xxx_read_csr_reg(regbuf + len, oct); len += cn6xxx_read_config_reg(regbuf + len, oct); break; diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index c75734f..20d6942 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -1258,7 +1258,7 @@ static void octeon_destroy_resources(struct octeon_device *oct) /* Nothing to be done here either */ break; - } /* end switch(oct->status) */ + } /* end switch (oct->status) */ tasklet_kill(&oct_priv->droq_tasklet); } @@ -2125,7 +2125,7 @@ static int liquidio_napi_poll(struct napi_struct *napi, int budget) /** * \brief Setup input and output queues * @param octeon_dev octeon device - * @param net_device Net device + * @param ifidx Interface Index * * Note: Queues are with respect to the octeon device. Thus * an input queue is for egress packets, and output queues @@ -2336,7 +2336,6 @@ static int liquidio_stop(struct net_device *netdev) } dev_info(&oct->pci_dev->dev, "%s interface is stopped\n", netdev->name); - module_put(THIS_MODULE); return 0; } @@ -2347,6 +2346,7 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr) struct net_device *netdev = (struct net_device *)nctrl->netpndev; struct lio *lio = GET_LIO(netdev); struct octeon_device *oct = lio->oct_dev; + u8 *mac; switch (nctrl->ncmd.s.cmd) { case OCTNET_CMD_CHANGE_DEVFLAGS: @@ -2354,12 +2354,11 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr) break; case OCTNET_CMD_CHANGE_MACADDR: - /* If command is successful, change the MACADDR. */ - netif_info(lio, probe, lio->netdev, " MACAddr changed to 0x%llx\n", - CVM_CAST64(nctrl->udd[0])); - dev_info(&oct->pci_dev->dev, "%s MACAddr changed to 0x%llx\n", - netdev->name, CVM_CAST64(nctrl->udd[0])); - memcpy(netdev->dev_addr, ((u8 *)&nctrl->udd[0]) + 2, ETH_ALEN); + mac = ((u8 *)&nctrl->udd[0]) + 2; + netif_info(lio, probe, lio->netdev, + "%s %2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x\n", + "MACAddr changed to", mac[0], mac[1], + mac[2], mac[3], mac[4], mac[5]); break; case OCTNET_CMD_CHANGE_MTU: @@ -2940,12 +2939,12 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) /* defer sending if queue is full */ stats->tx_iq_busy++; netif_info(lio, tx_err, lio->netdev, "Transmit failed iq:%d full\n", - ndata.q_no); + lio->txq); return NETDEV_TX_BUSY; } } /* pr_info(" XMIT - valid Qs: %d, 1st Q no: %d, cpu: %d, q_no:%d\n", - * lio->linfo.num_txpciq, lio->txq, cpu, ndata.q_no ); + * lio->linfo.num_txpciq, lio->txq, cpu, ndata.q_no); */ ndata.datasize = skb->len; @@ -2969,6 +2968,7 @@ static int liquidio_xmit(struct sk_buff *skb, struct net_device *netdev) if (skb_shinfo(skb)->nr_frags == 0) { cmdsetup.s.u.datasize = skb->len; octnet_prepare_pci_cmd(oct, &ndata.cmd, &cmdsetup, tag); + /* Offload checksum calculation for TCP/UDP packets */ dptr = dma_map_single(&oct->pci_dev->dev, skb->data, @@ -3716,8 +3716,8 @@ static int setup_nic_devices(struct octeon_device *octeon_dev) octeon_dev->priv_flags = 0x0; if (netdev->features & NETIF_F_LRO) - liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE, - OCTNIC_LROIPV4 | OCTNIC_LROIPV6); + liquidio_set_feature(netdev, OCTNET_CMD_LRO_ENABLE, + OCTNIC_LROIPV4 | OCTNIC_LROIPV6); liquidio_set_feature(netdev, OCTNET_CMD_ENABLE_VLAN_FILTER, 0); diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index c6e3d57..603d205 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -34,6 +34,7 @@ #define LIQUIDIO_MICRO_VERSION ".1" #define LIQUIDIO_PACKAGE "" #define LIQUIDIO_VERSION "1.4.1" + #define CONTROL_IQ 0 /** Tag types used by Octeon cores in its work. */ enum octeon_tag_type { @@ -295,7 +296,7 @@ union octnet_cmd { #define OCTNET_CMD_SIZE (sizeof(union octnet_cmd)) -/* Instruction Header (DPI - CN23xx) - for OCTEON-III models */ +/* Instruction Header(DPI) - for OCTEON-III models */ struct octeon_instr_ih3 { #ifdef __BIG_ENDIAN_BITFIELD @@ -345,7 +346,7 @@ struct octeon_instr_ih3 { #endif }; -/* Optional PKI Instruction Header(PKI IH) - for OCTEON CN23XX models */ +/* Optional PKI Instruction Header(PKI IH) - for OCTEON-III models */ /** BIG ENDIAN format. */ struct octeon_instr_pki_ih3 { #ifdef __BIG_ENDIAN_BITFIELD diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_config.h b/drivers/net/ethernet/cavium/liquidio/octeon_config.h index 4b8c948..b3396e3 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_config.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_config.h @@ -226,7 +226,7 @@ struct octeon_oq_config { */ u64 refill_threshold:16; - /** If set, the Output queue uses info-pointer mode. (Default: 1 ) */ + /** If set, the Output queue uses info-pointer mode. (Default: 1) */ u64 info_ptr:32; /* Max number of OQs available */ @@ -236,7 +236,7 @@ struct octeon_oq_config { /* Max number of OQs available */ u64 max_oqs:8; - /** If set, the Output queue uses info-pointer mode. (Default: 1 ) */ + /** If set, the Output queue uses info-pointer mode. (Default: 1) */ u64 info_ptr:32; /** The number of buffers that were consumed during packet processing by diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_console.c b/drivers/net/ethernet/cavium/liquidio/octeon_console.c index 70a8cd3..bbb50ea 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_console.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_console.c @@ -223,7 +223,7 @@ static void CVMX_BOOTMEM_NAMED_GET_NAME(struct octeon_device *oct, u32 len) { addr += offsetof(struct cvmx_bootmem_named_block_desc, name); - octeon_pci_read_core_mem(oct, addr, str, len); + octeon_pci_read_core_mem(oct, addr, (u8 *)str, len); str[len] = 0; } @@ -375,7 +375,7 @@ static void octeon_remote_unlock(void) int octeon_console_send_cmd(struct octeon_device *oct, char *cmd_str, u32 wait_hundredths) { - u32 len = strlen(cmd_str); + u32 len = (u32)strlen(cmd_str); dev_dbg(&oct->pci_dev->dev, "sending \"%s\" to bootloader\n", cmd_str); @@ -483,7 +483,7 @@ static void check_console(struct work_struct *work) struct octeon_console *console; struct cavium_wk *wk = (struct cavium_wk *)work; struct octeon_device *oct = (struct octeon_device *)wk->ctxptr; - size_t console_num = wk->ctxul; + u32 console_num = (u32)wk->ctxul; u32 delay; console = &oct->console[console_num]; @@ -506,7 +506,7 @@ static void check_console(struct work_struct *work) console_buffer, bytes_read); } } else if (bytes_read < 0) { - dev_err(&oct->pci_dev->dev, "Error reading console %lu, ret=%d\n", + dev_err(&oct->pci_dev->dev, "Error reading console %u, ret=%d\n", console_num, bytes_read); } @@ -518,7 +518,7 @@ static void check_console(struct work_struct *work) */ if (octeon_console_debug_enabled(console_num) && (total_read == 0) && (console->leftover[0])) { - dev_info(&oct->pci_dev->dev, "%lu: %s\n", + dev_info(&oct->pci_dev->dev, "%u: %s\n", console_num, console->leftover); console->leftover[0] = '\0'; } @@ -700,7 +700,7 @@ static int octeon_console_read(struct octeon_device *oct, u32 console_num, bytes_to_read = console->buffer_size - rd_idx; octeon_pci_read_core_mem(oct, console->output_base_addr + rd_idx, - buffer, bytes_to_read); + (u8 *)buffer, bytes_to_read); octeon_write_device_mem32(oct, console->addr + offsetof(struct octeon_pci_console, output_read_index), diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c index fda93be..0eb504a 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c @@ -440,10 +440,10 @@ static struct octeon_config_ptr { }; static char oct_dev_state_str[OCT_DEV_STATES + 1][32] = { - "BEGIN", "PCI-MAP-DONE", "DISPATCH-INIT-DONE", + "BEGIN", "PCI-MAP-DONE", "DISPATCH-INIT-DONE", "IQ-INIT-DONE", "SCBUFF-POOL-INIT-DONE", "RESPLIST-INIT-DONE", "DROQ-INIT-DONE", "IO-QUEUES-INIT-DONE", "CONSOLE-INIT-DONE", - "HOST-READY", "CORE-READY", "RUNNING", "IN-RESET", + "HOST-READY", "CORE-READY", "RUNNING", "IN-RESET", "INVALID" }; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h index 579acd4..01edfb4 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h @@ -221,7 +221,7 @@ struct octeon_fn_list { /* Structure for named memory blocks * Number of descriptors - * available can be changed without affecting compatiblity, + * available can be changed without affecting compatibility, * but name length changes require a bump in the bootmem * descriptor version * Note: This structure must be naturally 64 bit aligned, as a single @@ -254,7 +254,7 @@ struct oct_fw_info { struct cavium_wk { struct delayed_work work; void *ctxptr; - size_t ctxul; + u64 ctxul; }; struct cavium_wq { diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c index 93ab4ec..e0afe4c 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c @@ -92,6 +92,11 @@ static inline void *octeon_get_dispatch_arg(struct octeon_device *octeon_dev, return fn_arg; } +/** Check for packets on Droq. This function should be called with + * lock held. + * @param droq - Droq on which count is checked. + * @return Returns packet count. + */ u32 octeon_droq_check_hw_for_pkts(struct octeon_droq *droq) { u32 pkt_count = 0; @@ -183,7 +188,6 @@ octeon_droq_setup_ring_buffers(struct octeon_device *oct, droq->recv_buf_list[i].buffer = buf; droq->recv_buf_list[i].data = get_rbd(buf); - droq->info_list[i].length = 0; /* map ring buffers into memory */ @@ -556,7 +560,9 @@ octeon_droq_dispatch_pkt(struct octeon_device *oct, droq->stats.dropped_nomem++; } } else { - dev_err(&oct->pci_dev->dev, "DROQ: No dispatch function\n"); + dev_err(&oct->pci_dev->dev, "DROQ: No dispatch function (opcode %u/%u)\n", + (unsigned int)rh->r.opcode, + (unsigned int)rh->r.subcode); droq->stats.dropped_nodispatch++; } /* else (dispatch_fn ... */ @@ -641,6 +647,7 @@ octeon_droq_fast_process_packets(struct octeon_device *oct, pg_info->page = NULL; droq->recv_buf_list[droq->read_idx].buffer = NULL; + INCR_INDEX_BY1(droq->read_idx, droq->max_count); droq->refill_count++; } else { @@ -735,7 +742,7 @@ octeon_droq_process_packets(struct octeon_device *oct, if (pkt_count > budget) pkt_count = budget; - /* Grab the lock */ + /* Grab the droq lock */ spin_lock(&droq->lock); pkts_processed = octeon_droq_fast_process_packets(oct, droq, pkt_count); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h index 5ac7e66..ff4b1d6 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h @@ -99,7 +99,7 @@ struct octeon_instr_queue { u32 rsvd:17; - /* Controls the periodic flushing of iq */ + /* Controls whether extra flushing of IQ is done on Tx */ u32 do_auto_flush:1; u32 status:8; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h index ab41866..fb820dc 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h @@ -124,7 +124,6 @@ struct lio { /* work queue for txq status */ struct cavium_wq txq_status_wq; - }; #define LIO_SIZE (sizeof(struct lio)) diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index 3f7044c..d32492f1 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -349,6 +349,7 @@ __add_to_request_list(struct octeon_instr_queue *iq, iq->request_list[idx].reqtype = reqtype; } +/* Can only run in process context */ int lio_process_iq_request_list(struct octeon_device *oct, struct octeon_instr_queue *iq, u32 napi_budget) @@ -405,6 +406,7 @@ lio_process_iq_request_list(struct octeon_device *oct, flags); } else { if (sc->callback) { + /* This callback must not sleep */ sc->callback(oct, OCTEON_REQUEST_DONE, sc->callback_arg); } @@ -521,7 +523,7 @@ static void check_db_timeout(struct work_struct *work) { struct cavium_wk *wk = (struct cavium_wk *)work; struct octeon_device *oct = (struct octeon_device *)wk->ctxptr; - unsigned long iq_no = wk->ctxul; + u64 iq_no = wk->ctxul; struct cavium_wq *db_wq = &oct->check_db_wq[iq_no]; u32 delay = 10; @@ -550,7 +552,7 @@ octeon_send_command(struct octeon_device *oct, u32 iq_no, INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, bytes_sent, datasize); INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, instr_posted, 1); - if (iq->fill_cnt >= iq->fill_threshold || force_db) + if (force_db) ring_doorbell(oct, iq); } else { INCR_INSTRQUEUE_PKT_COUNT(oct, iq_no, instr_dropped, 1); -- cgit v0.10.2 From 9fbc48f6a7e17ca34090a4be72f08afe557a6afb Mon Sep 17 00:00:00 2001 From: Raghu Vatsavayi Date: Sun, 3 Jul 2016 13:56:56 -0700 Subject: liquidio: Response header changes This patch changes response header to be able to communicate with new firmware interface. Signed-off-by: Derek Chickles Signed-off-by: Satanand Burla Signed-off-by: Felix Manlunas Signed-off-by: Raghu Vatsavayi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h index 603d205..199a8b9 100644 --- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h +++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h @@ -542,6 +542,7 @@ union octeon_rh { u64 csum_verified:3; /** checksum verified. */ u64 has_hwtstamp:1; /** Has hardware timestamp. 1 = yes. */ u64 encap_on:1; + u64 has_hash:1; /** Has hash (rth or rss). 1 = yes. */ } r_dh; struct { u64 opcode:4; @@ -551,7 +552,8 @@ union octeon_rh { u64 num_gmx_ports:8; u64 max_nic_ports:10; u64 app_cap_flags:4; - u64 app_mode:16; + u64 app_mode:8; + u64 pkind:8; } r_core_drv_init; struct { u64 opcode:4; @@ -571,6 +573,7 @@ union octeon_rh { u64 opcode:4; } r; struct { + u64 has_hash:1; /** Has hash (rth or rss). 1 = yes. */ u64 encap_on:1; u64 has_hwtstamp:1; /** 1 = has hwtstamp */ u64 csum_verified:3; /** checksum verified. */ @@ -582,7 +585,8 @@ union octeon_rh { u64 opcode:4; } r_dh; struct { - u64 app_mode:16; + u64 pkind:8; + u64 app_mode:8; u64 app_cap_flags:4; u64 max_nic_ports:10; u64 num_gmx_ports:8; @@ -640,9 +644,11 @@ union oct_link_status { u64 autoneg:1; u64 if_mode:5; u64 pause:1; - u64 reserved:16; + u64 flashing:1; + u64 reserved:15; #else - u64 reserved:16; + u64 reserved:15; + u64 flashing:1; u64 pause:1; u64 if_mode:5; u64 autoneg:1; @@ -869,9 +875,9 @@ union oct_nic_if_cfg { u64 num_iqueues:16; u64 num_oqueues:16; u64 gmx_port_id:8; - u64 reserved:8; + u64 vf_id:8; #else - u64 reserved:8; + u64 vf_id:8; u64 gmx_port_id:8; u64 num_oqueues:16; u64 num_iqueues:16; -- cgit v0.10.2 From f48cc6b2661ccb6e9d85a5cdd78c0150929a2821 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Mon, 4 Jul 2016 13:53:38 +0800 Subject: tun: fix build warnings Stephen Rothwell reports a build warnings(powerpc ppc64_defconfig) drivers/net/tun.c: In function 'tun_do_read.part.5': /home/sfr/next/next/drivers/net/tun.c:1491:6: warning: 'err' may be used uninitialized in this function [-Wmaybe-uninitialized] int err; This is because tun_ring_recv() may return an uninitialized err, fix this. Reported-by: Stephen Rothwell Signed-off-by: Jason Wang Signed-off-by: David S. Miller diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 7475215..5eadb7a 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1447,12 +1447,13 @@ static struct sk_buff *tun_ring_recv(struct tun_file *tfile, int noblock, { DECLARE_WAITQUEUE(wait, current); struct sk_buff *skb = NULL; + int error = 0; skb = skb_array_consume(&tfile->tx_array); if (skb) goto out; if (noblock) { - *err = -EAGAIN; + error = -EAGAIN; goto out; } @@ -1464,11 +1465,11 @@ static struct sk_buff *tun_ring_recv(struct tun_file *tfile, int noblock, if (skb) break; if (signal_pending(current)) { - *err = -ERESTARTSYS; + error = -ERESTARTSYS; break; } if (tfile->socket.sk->sk_shutdown & RCV_SHUTDOWN) { - *err = -EFAULT; + error = -EFAULT; break; } @@ -1479,6 +1480,7 @@ static struct sk_buff *tun_ring_recv(struct tun_file *tfile, int noblock, remove_wait_queue(&tfile->wq.wait, &wait); out: + *err = error; return skb; } -- cgit v0.10.2 From 5e9c16cc83a708a031f3999ac84f57101123d109 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 4 Jul 2016 08:23:04 +0200 Subject: mlxsw: spectrum_router: Implement private fib Shadow FIB is needed in order to hold additional information for FIB entries and keep track of used prefixes. That is needed for the LPM tree construction to be introduced later on in this set. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 83d5807..da0e072 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -43,6 +43,7 @@ #include #include #include +#include #include #include "port.h" @@ -160,6 +161,12 @@ struct mlxsw_sp_sb { } ports[MLXSW_PORT_MAX_PORTS]; }; +#define MLXSW_SP_PREFIX_COUNT (sizeof(struct in6_addr) * BITS_PER_BYTE) + +struct mlxsw_sp_prefix_usage { + DECLARE_BITMAP(b, MLXSW_SP_PREFIX_COUNT); +}; + struct mlxsw_sp { struct { struct list_head list; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 8d70496..af4a35d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -35,11 +35,131 @@ #include #include +#include +#include +#include #include "spectrum.h" #include "core.h" #include "reg.h" +static void +mlxsw_sp_prefix_usage_set(struct mlxsw_sp_prefix_usage *prefix_usage, + unsigned char prefix_len) +{ + set_bit(prefix_len, prefix_usage->b); +} + +static void +mlxsw_sp_prefix_usage_clear(struct mlxsw_sp_prefix_usage *prefix_usage, + unsigned char prefix_len) +{ + clear_bit(prefix_len, prefix_usage->b); +} + +struct mlxsw_sp_fib_key { + unsigned char addr[sizeof(struct in6_addr)]; + unsigned char prefix_len; +}; + +struct mlxsw_sp_fib_entry { + struct rhash_head ht_node; + struct mlxsw_sp_fib_key key; +}; + +struct mlxsw_sp_fib { + struct rhashtable ht; + unsigned long prefix_ref_count[MLXSW_SP_PREFIX_COUNT]; + struct mlxsw_sp_prefix_usage prefix_usage; +}; + +static const struct rhashtable_params mlxsw_sp_fib_ht_params = { + .key_offset = offsetof(struct mlxsw_sp_fib_entry, key), + .head_offset = offsetof(struct mlxsw_sp_fib_entry, ht_node), + .key_len = sizeof(struct mlxsw_sp_fib_key), + .automatic_shrinking = true, +}; + +static int mlxsw_sp_fib_entry_insert(struct mlxsw_sp_fib *fib, + struct mlxsw_sp_fib_entry *fib_entry) +{ + unsigned char prefix_len = fib_entry->key.prefix_len; + int err; + + err = rhashtable_insert_fast(&fib->ht, &fib_entry->ht_node, + mlxsw_sp_fib_ht_params); + if (err) + return err; + if (fib->prefix_ref_count[prefix_len]++ == 0) + mlxsw_sp_prefix_usage_set(&fib->prefix_usage, prefix_len); + return 0; +} + +static void mlxsw_sp_fib_entry_remove(struct mlxsw_sp_fib *fib, + struct mlxsw_sp_fib_entry *fib_entry) +{ + unsigned char prefix_len = fib_entry->key.prefix_len; + + if (--fib->prefix_ref_count[prefix_len] == 0) + mlxsw_sp_prefix_usage_clear(&fib->prefix_usage, prefix_len); + rhashtable_remove_fast(&fib->ht, &fib_entry->ht_node, + mlxsw_sp_fib_ht_params); +} + +static struct mlxsw_sp_fib_entry * +mlxsw_sp_fib_entry_create(struct mlxsw_sp_fib *fib, const void *addr, + size_t addr_len, unsigned char prefix_len) +{ + struct mlxsw_sp_fib_entry *fib_entry; + + fib_entry = kzalloc(sizeof(*fib_entry), GFP_KERNEL); + if (!fib_entry) + return NULL; + memcpy(fib_entry->key.addr, addr, addr_len); + fib_entry->key.prefix_len = prefix_len; + return fib_entry; +} + +static void mlxsw_sp_fib_entry_destroy(struct mlxsw_sp_fib_entry *fib_entry) +{ + kfree(fib_entry); +} + +static struct mlxsw_sp_fib_entry * +mlxsw_sp_fib_entry_lookup(struct mlxsw_sp_fib *fib, const void *addr, + size_t addr_len, unsigned char prefix_len) +{ + struct mlxsw_sp_fib_key key = {{ 0 } }; + + memcpy(key.addr, addr, addr_len); + key.prefix_len = prefix_len; + return rhashtable_lookup_fast(&fib->ht, &key, mlxsw_sp_fib_ht_params); +} + +static struct mlxsw_sp_fib *mlxsw_sp_fib_create(void) +{ + struct mlxsw_sp_fib *fib; + int err; + + fib = kzalloc(sizeof(*fib), GFP_KERNEL); + if (!fib) + return ERR_PTR(-ENOMEM); + err = rhashtable_init(&fib->ht, &mlxsw_sp_fib_ht_params); + if (err) + goto err_rhashtable_init; + return fib; + +err_rhashtable_init: + kfree(fib); + return ERR_PTR(err); +} + +static void mlxsw_sp_fib_destroy(struct mlxsw_sp_fib *fib) +{ + rhashtable_destroy(&fib->ht); + kfree(fib); +} + static int __mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) { char rgcr_pl[MLXSW_REG_RGCR_LEN]; -- cgit v0.10.2 From 6f9fc3cee4f1e97d19f25950521af397ec683558 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 4 Jul 2016 08:23:05 +0200 Subject: mlxsw: reg: Add Router Algorithmic LPM Tree Allocation Register definition Register serves for allocation and deallocation of LPM search tree. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 5ddc1d3..a358e1b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -3,7 +3,7 @@ * Copyright (c) 2015 Mellanox Technologies. All rights reserved. * Copyright (c) 2015-2016 Ido Schimmel * Copyright (c) 2015 Elad Raz - * Copyright (c) 2015 Jiri Pirko + * Copyright (c) 2015-2016 Jiri Pirko * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: @@ -3454,6 +3454,56 @@ static inline void mlxsw_reg_ritr_pack(char *payload, bool enable, mlxsw_reg_ritr_if_mac_memcpy_to(payload, mac); } +/* RALTA - Router Algorithmic LPM Tree Allocation Register + * ------------------------------------------------------- + * RALTA is used to allocate the LPM trees of the SHSPM method. + */ +#define MLXSW_REG_RALTA_ID 0x8010 +#define MLXSW_REG_RALTA_LEN 0x04 + +static const struct mlxsw_reg_info mlxsw_reg_ralta = { + .id = MLXSW_REG_RALTA_ID, + .len = MLXSW_REG_RALTA_LEN, +}; + +/* reg_ralta_op + * opcode (valid for Write, must be 0 on Read) + * 0 - allocate a tree + * 1 - deallocate a tree + * Access: OP + */ +MLXSW_ITEM32(reg, ralta, op, 0x00, 28, 2); + +enum mlxsw_reg_ralxx_protocol { + MLXSW_REG_RALXX_PROTOCOL_IPV4, + MLXSW_REG_RALXX_PROTOCOL_IPV6, +}; + +/* reg_ralta_protocol + * Protocol. + * Deallocation opcode: Reserved. + * Access: RW + */ +MLXSW_ITEM32(reg, ralta, protocol, 0x00, 24, 4); + +/* reg_ralta_tree_id + * An identifier (numbered from 1..cap_shspm_max_trees-1) representing + * the tree identifier (managed by software). + * Note that tree_id 0 is allocated for a default-route tree. + * Access: Index + */ +MLXSW_ITEM32(reg, ralta, tree_id, 0x00, 0, 8); + +static inline void mlxsw_reg_ralta_pack(char *payload, bool alloc, + enum mlxsw_reg_ralxx_protocol protocol, + u8 tree_id) +{ + MLXSW_REG_ZERO(ralta, payload); + mlxsw_reg_ralta_op_set(payload, !alloc); + mlxsw_reg_ralta_protocol_set(payload, protocol); + mlxsw_reg_ralta_tree_id_set(payload, tree_id); +} + /* MFCR - Management Fan Control Register * -------------------------------------- * This register controls the settings of the Fan Speed PWM mechanism. @@ -4196,6 +4246,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "RGCR"; case MLXSW_REG_RITR_ID: return "RITR"; + case MLXSW_REG_RALTA_ID: + return "RALTA"; case MLXSW_REG_MFCR_ID: return "MFCR"; case MLXSW_REG_MFSC_ID: -- cgit v0.10.2 From a9823359c6fd3f1322df8baf1f39c17b48690251 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 4 Jul 2016 08:23:06 +0200 Subject: mlxsw: reg: Add Router Algorithmic LPM Structure Tree Register definition Serves to build LPM tree structure. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index a358e1b..4ea608f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -3504,6 +3504,81 @@ static inline void mlxsw_reg_ralta_pack(char *payload, bool alloc, mlxsw_reg_ralta_tree_id_set(payload, tree_id); } +/* RALST - Router Algorithmic LPM Structure Tree Register + * ------------------------------------------------------ + * RALST is used to set and query the structure of an LPM tree. + * The structure of the tree must be sorted as a sorted binary tree, while + * each node is a bin that is tagged as the length of the prefixes the lookup + * will refer to. Therefore, bin X refers to a set of entries with prefixes + * of X bits to match with the destination address. The bin 0 indicates + * the default action, when there is no match of any prefix. + */ +#define MLXSW_REG_RALST_ID 0x8011 +#define MLXSW_REG_RALST_LEN 0x104 + +static const struct mlxsw_reg_info mlxsw_reg_ralst = { + .id = MLXSW_REG_RALST_ID, + .len = MLXSW_REG_RALST_LEN, +}; + +/* reg_ralst_root_bin + * The bin number of the root bin. + * 0 Date: Mon, 4 Jul 2016 08:23:07 +0200 Subject: mlxsw: reg: Add Router Algorithmic LPM Tree Binding Register definition This register is used to bind virtual router and protocol to an allocated LPM tree. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 4ea608f..0b7b91f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -3579,6 +3579,49 @@ static inline void mlxsw_reg_ralst_bin_pack(char *payload, u8 bin_number, right_child_bin); } +/* RALTB - Router Algorithmic LPM Tree Binding Register + * ---------------------------------------------------- + * RALTB is used to bind virtual router and protocol to an allocated LPM tree. + */ +#define MLXSW_REG_RALTB_ID 0x8012 +#define MLXSW_REG_RALTB_LEN 0x04 + +static const struct mlxsw_reg_info mlxsw_reg_raltb = { + .id = MLXSW_REG_RALTB_ID, + .len = MLXSW_REG_RALTB_LEN, +}; + +/* reg_raltb_virtual_router + * Virtual Router ID + * Range is 0..cap_max_virtual_routers-1 + * Access: Index + */ +MLXSW_ITEM32(reg, raltb, virtual_router, 0x00, 16, 16); + +/* reg_raltb_protocol + * Protocol. + * Access: Index + */ +MLXSW_ITEM32(reg, raltb, protocol, 0x00, 12, 4); + +/* reg_raltb_tree_id + * Tree to be used for the {virtual_router, protocol} + * Tree identifier numbered from 1..(cap_shspm_max_trees-1). + * By default, all Unicast IPv4 and IPv6 are bound to tree_id 0. + * Access: RW + */ +MLXSW_ITEM32(reg, raltb, tree_id, 0x00, 0, 8); + +static inline void mlxsw_reg_raltb_pack(char *payload, u16 virtual_router, + enum mlxsw_reg_ralxx_protocol protocol, + u8 tree_id) +{ + MLXSW_REG_ZERO(raltb, payload); + mlxsw_reg_raltb_virtual_router_set(payload, virtual_router); + mlxsw_reg_raltb_protocol_set(payload, protocol); + mlxsw_reg_raltb_tree_id_set(payload, tree_id); +} + /* MFCR - Management Fan Control Register * -------------------------------------- * This register controls the settings of the Fan Speed PWM mechanism. @@ -4325,6 +4368,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "RALTA"; case MLXSW_REG_RALST_ID: return "RALST"; + case MLXSW_REG_RALTB_ID: + return "RALTB"; case MLXSW_REG_MFCR_ID: return "MFCR"; case MLXSW_REG_MFSC_ID: -- cgit v0.10.2 From 53342023eed997503c497cedc18b0441b0674b6e Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 4 Jul 2016 08:23:08 +0200 Subject: mlxsw: spectrum_router: Implement LPM trees management Introduce basic LPM tree management allowing to share the trees in between tables if the used prefixes in the tables are the same. Build the tree structure according to the used prefixes. Although it is not optimal for many use cases, this initial implementation does only simple linear left-tree. More advanced structures will be introduced later on, possibly including mechanisms to change trees on the fly. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index da0e072..5db57a7 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -61,6 +61,10 @@ #define MLXSW_SP_PORTS_PER_CLUSTER_MAX 4 +#define MLXSW_SP_LPM_TREE_MIN 2 /* trees 0 and 1 are reserved */ +#define MLXSW_SP_LPM_TREE_MAX 22 +#define MLXSW_SP_LPM_TREE_COUNT (MLXSW_SP_LPM_TREE_MAX - MLXSW_SP_LPM_TREE_MIN) + #define MLXSW_SP_PORT_BASE_SPEED 25000 /* Mb/s */ #define MLXSW_SP_BYTES_PER_CELL 96 @@ -167,6 +171,22 @@ struct mlxsw_sp_prefix_usage { DECLARE_BITMAP(b, MLXSW_SP_PREFIX_COUNT); }; +enum mlxsw_sp_l3proto { + MLXSW_SP_L3_PROTO_IPV4, + MLXSW_SP_L3_PROTO_IPV6, +}; + +struct mlxsw_sp_lpm_tree { + u8 id; /* tree ID */ + unsigned int ref_count; + enum mlxsw_sp_l3proto proto; + struct mlxsw_sp_prefix_usage prefix_usage; +}; + +struct mlxsw_sp_router { + struct mlxsw_sp_lpm_tree lpm_trees[MLXSW_SP_LPM_TREE_COUNT]; +}; + struct mlxsw_sp { struct { struct list_head list; @@ -199,6 +219,7 @@ struct mlxsw_sp { struct mlxsw_sp_upper lags[MLXSW_SP_LAG_MAX]; u8 port_to_module[MLXSW_PORT_MAX_PORTS]; struct mlxsw_sp_sb sb; + struct mlxsw_sp_router router; }; static inline struct mlxsw_sp_upper * diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index af4a35d..73fd85c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -43,6 +43,16 @@ #include "core.h" #include "reg.h" +#define mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage) \ + for_each_set_bit(prefix, (prefix_usage)->b, MLXSW_SP_PREFIX_COUNT) + +static bool +mlxsw_sp_prefix_usage_eq(struct mlxsw_sp_prefix_usage *prefix_usage1, + struct mlxsw_sp_prefix_usage *prefix_usage2) +{ + return !memcmp(prefix_usage1, prefix_usage2, sizeof(*prefix_usage1)); +} + static void mlxsw_sp_prefix_usage_set(struct mlxsw_sp_prefix_usage *prefix_usage, unsigned char prefix_len) @@ -160,6 +170,143 @@ static void mlxsw_sp_fib_destroy(struct mlxsw_sp_fib *fib) kfree(fib); } +static struct mlxsw_sp_lpm_tree * +mlxsw_sp_lpm_tree_find_unused(struct mlxsw_sp *mlxsw_sp, bool one_reserved) +{ + static struct mlxsw_sp_lpm_tree *lpm_tree; + int i; + + for (i = 0; i < MLXSW_SP_LPM_TREE_COUNT; i++) { + lpm_tree = &mlxsw_sp->router.lpm_trees[i]; + if (lpm_tree->ref_count == 0) { + if (one_reserved) + one_reserved = false; + else + return lpm_tree; + } + } + return NULL; +} + +static int mlxsw_sp_lpm_tree_alloc(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_lpm_tree *lpm_tree) +{ + char ralta_pl[MLXSW_REG_RALTA_LEN]; + + mlxsw_reg_ralta_pack(ralta_pl, true, lpm_tree->proto, lpm_tree->id); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralta), ralta_pl); +} + +static int mlxsw_sp_lpm_tree_free(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_lpm_tree *lpm_tree) +{ + char ralta_pl[MLXSW_REG_RALTA_LEN]; + + mlxsw_reg_ralta_pack(ralta_pl, false, lpm_tree->proto, lpm_tree->id); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralta), ralta_pl); +} + +static int +mlxsw_sp_lpm_tree_left_struct_set(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_prefix_usage *prefix_usage, + struct mlxsw_sp_lpm_tree *lpm_tree) +{ + char ralst_pl[MLXSW_REG_RALST_LEN]; + u8 root_bin = 0; + u8 prefix; + u8 last_prefix = MLXSW_REG_RALST_BIN_NO_CHILD; + + mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage) + root_bin = prefix; + + mlxsw_reg_ralst_pack(ralst_pl, root_bin, lpm_tree->id); + mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage) { + if (prefix == 0) + continue; + mlxsw_reg_ralst_bin_pack(ralst_pl, prefix, last_prefix, + MLXSW_REG_RALST_BIN_NO_CHILD); + last_prefix = prefix; + } + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralst), ralst_pl); +} + +static struct mlxsw_sp_lpm_tree * +mlxsw_sp_lpm_tree_create(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_prefix_usage *prefix_usage, + enum mlxsw_sp_l3proto proto, bool one_reserved) +{ + struct mlxsw_sp_lpm_tree *lpm_tree; + int err; + + lpm_tree = mlxsw_sp_lpm_tree_find_unused(mlxsw_sp, one_reserved); + if (!lpm_tree) + return ERR_PTR(-EBUSY); + lpm_tree->proto = proto; + err = mlxsw_sp_lpm_tree_alloc(mlxsw_sp, lpm_tree); + if (err) + return ERR_PTR(err); + + err = mlxsw_sp_lpm_tree_left_struct_set(mlxsw_sp, prefix_usage, + lpm_tree); + if (err) + goto err_left_struct_set; + return lpm_tree; + +err_left_struct_set: + mlxsw_sp_lpm_tree_free(mlxsw_sp, lpm_tree); + return ERR_PTR(err); +} + +static int mlxsw_sp_lpm_tree_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_lpm_tree *lpm_tree) +{ + return mlxsw_sp_lpm_tree_free(mlxsw_sp, lpm_tree); +} + +static struct mlxsw_sp_lpm_tree * +mlxsw_sp_lpm_tree_get(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_prefix_usage *prefix_usage, + enum mlxsw_sp_l3proto proto, bool one_reserved) +{ + struct mlxsw_sp_lpm_tree *lpm_tree; + int i; + + for (i = 0; i < MLXSW_SP_LPM_TREE_COUNT; i++) { + lpm_tree = &mlxsw_sp->router.lpm_trees[i]; + if (lpm_tree->proto == proto && + mlxsw_sp_prefix_usage_eq(&lpm_tree->prefix_usage, + prefix_usage)) + goto inc_ref_count; + } + lpm_tree = mlxsw_sp_lpm_tree_create(mlxsw_sp, prefix_usage, + proto, one_reserved); + if (IS_ERR(lpm_tree)) + return lpm_tree; + +inc_ref_count: + lpm_tree->ref_count++; + return lpm_tree; +} + +static int mlxsw_sp_lpm_tree_put(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_lpm_tree *lpm_tree) +{ + if (--lpm_tree->ref_count == 0) + return mlxsw_sp_lpm_tree_destroy(mlxsw_sp, lpm_tree); + return 0; +} + +static void mlxsw_sp_lpm_init(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_sp_lpm_tree *lpm_tree; + int i; + + for (i = 0; i < MLXSW_SP_LPM_TREE_COUNT; i++) { + lpm_tree = &mlxsw_sp->router.lpm_trees[i]; + lpm_tree->id = i + MLXSW_SP_LPM_TREE_MIN; + } +} + static int __mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) { char rgcr_pl[MLXSW_REG_RGCR_LEN]; @@ -179,7 +326,13 @@ static void __mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) { - return __mlxsw_sp_router_init(mlxsw_sp); + int err; + + err = __mlxsw_sp_router_init(mlxsw_sp); + if (err) + return err; + mlxsw_sp_lpm_init(mlxsw_sp); + return 0; } void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) -- cgit v0.10.2 From 6b75c4807db3047b15c248026f07054951b911ce Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 4 Jul 2016 08:23:09 +0200 Subject: mlxsw: spectrum_router: Add virtual router management Virtual router is a construct used inside HW. In this implementation we map kernel tables to virtual routers one to one. Introduce management logic to create virtual routers when needed and destroy in case they are no longer in use. According to that, call into LPM tree management. Each virtual router is always bound to one LPM tree. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 5db57a7..5b40dfc 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -65,6 +65,8 @@ #define MLXSW_SP_LPM_TREE_MAX 22 #define MLXSW_SP_LPM_TREE_COUNT (MLXSW_SP_LPM_TREE_MAX - MLXSW_SP_LPM_TREE_MIN) +#define MLXSW_SP_VIRTUAL_ROUTER_MAX 256 + #define MLXSW_SP_PORT_BASE_SPEED 25000 /* Mb/s */ #define MLXSW_SP_BYTES_PER_CELL 96 @@ -183,8 +185,20 @@ struct mlxsw_sp_lpm_tree { struct mlxsw_sp_prefix_usage prefix_usage; }; +struct mlxsw_sp_fib; + +struct mlxsw_sp_vr { + u16 id; /* virtual router ID */ + bool used; + enum mlxsw_sp_l3proto proto; + u32 tb_id; /* kernel fib table id */ + struct mlxsw_sp_lpm_tree *lpm_tree; + struct mlxsw_sp_fib *fib; +}; + struct mlxsw_sp_router { struct mlxsw_sp_lpm_tree lpm_trees[MLXSW_SP_LPM_TREE_COUNT]; + struct mlxsw_sp_vr vrs[MLXSW_SP_VIRTUAL_ROUTER_MAX]; }; struct mlxsw_sp { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 73fd85c..11dab74 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -47,12 +47,46 @@ for_each_set_bit(prefix, (prefix_usage)->b, MLXSW_SP_PREFIX_COUNT) static bool +mlxsw_sp_prefix_usage_subset(struct mlxsw_sp_prefix_usage *prefix_usage1, + struct mlxsw_sp_prefix_usage *prefix_usage2) +{ + unsigned char prefix; + + mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage1) { + if (!test_bit(prefix, prefix_usage2->b)) + return false; + } + return true; +} + +static bool mlxsw_sp_prefix_usage_eq(struct mlxsw_sp_prefix_usage *prefix_usage1, struct mlxsw_sp_prefix_usage *prefix_usage2) { return !memcmp(prefix_usage1, prefix_usage2, sizeof(*prefix_usage1)); } +static bool +mlxsw_sp_prefix_usage_none(struct mlxsw_sp_prefix_usage *prefix_usage) +{ + struct mlxsw_sp_prefix_usage prefix_usage_none = {{ 0 } }; + + return mlxsw_sp_prefix_usage_eq(prefix_usage, &prefix_usage_none); +} + +static void +mlxsw_sp_prefix_usage_cpy(struct mlxsw_sp_prefix_usage *prefix_usage1, + struct mlxsw_sp_prefix_usage *prefix_usage2) +{ + memcpy(prefix_usage1, prefix_usage2, sizeof(*prefix_usage1)); +} + +static void +mlxsw_sp_prefix_usage_zero(struct mlxsw_sp_prefix_usage *prefix_usage) +{ + memset(prefix_usage, 0, sizeof(*prefix_usage)); +} + static void mlxsw_sp_prefix_usage_set(struct mlxsw_sp_prefix_usage *prefix_usage, unsigned char prefix_len) @@ -307,6 +341,199 @@ static void mlxsw_sp_lpm_init(struct mlxsw_sp *mlxsw_sp) } } +static struct mlxsw_sp_vr *mlxsw_sp_vr_find_unused(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_sp_vr *vr; + int i; + + for (i = 0; i < MLXSW_SP_VIRTUAL_ROUTER_MAX; i++) { + vr = &mlxsw_sp->router.vrs[i]; + if (!vr->used) + return vr; + } + return NULL; +} + +static int mlxsw_sp_vr_lpm_tree_bind(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_vr *vr) +{ + char raltb_pl[MLXSW_REG_RALTB_LEN]; + + mlxsw_reg_raltb_pack(raltb_pl, vr->id, vr->proto, vr->lpm_tree->id); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(raltb), raltb_pl); +} + +static int mlxsw_sp_vr_lpm_tree_unbind(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_vr *vr) +{ + char raltb_pl[MLXSW_REG_RALTB_LEN]; + + /* Bind to tree 0 which is default */ + mlxsw_reg_raltb_pack(raltb_pl, vr->id, vr->proto, 0); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(raltb), raltb_pl); +} + +static u32 mlxsw_sp_fix_tb_id(u32 tb_id) +{ + /* For our purpose, squash main and local table into one */ + if (tb_id == RT_TABLE_LOCAL) + tb_id = RT_TABLE_MAIN; + return tb_id; +} + +static struct mlxsw_sp_vr *mlxsw_sp_vr_find(struct mlxsw_sp *mlxsw_sp, + u32 tb_id, + enum mlxsw_sp_l3proto proto) +{ + struct mlxsw_sp_vr *vr; + int i; + + tb_id = mlxsw_sp_fix_tb_id(tb_id); + for (i = 0; i < MLXSW_SP_VIRTUAL_ROUTER_MAX; i++) { + vr = &mlxsw_sp->router.vrs[i]; + if (vr->used && vr->proto == proto && vr->tb_id == tb_id) + return vr; + } + return NULL; +} + +static struct mlxsw_sp_vr *mlxsw_sp_vr_create(struct mlxsw_sp *mlxsw_sp, + unsigned char prefix_len, + u32 tb_id, + enum mlxsw_sp_l3proto proto) +{ + struct mlxsw_sp_prefix_usage req_prefix_usage; + struct mlxsw_sp_lpm_tree *lpm_tree; + struct mlxsw_sp_vr *vr; + int err; + + vr = mlxsw_sp_vr_find_unused(mlxsw_sp); + if (!vr) + return ERR_PTR(-EBUSY); + vr->fib = mlxsw_sp_fib_create(); + if (IS_ERR(vr->fib)) + return ERR_CAST(vr->fib); + + vr->proto = proto; + vr->tb_id = tb_id; + mlxsw_sp_prefix_usage_zero(&req_prefix_usage); + mlxsw_sp_prefix_usage_set(&req_prefix_usage, prefix_len); + lpm_tree = mlxsw_sp_lpm_tree_get(mlxsw_sp, &req_prefix_usage, + proto, true); + if (IS_ERR(lpm_tree)) { + err = PTR_ERR(lpm_tree); + goto err_tree_get; + } + vr->lpm_tree = lpm_tree; + err = mlxsw_sp_vr_lpm_tree_bind(mlxsw_sp, vr); + if (err) + goto err_tree_bind; + + vr->used = true; + return vr; + +err_tree_bind: + mlxsw_sp_lpm_tree_put(mlxsw_sp, vr->lpm_tree); +err_tree_get: + mlxsw_sp_fib_destroy(vr->fib); + + return ERR_PTR(err); +} + +static void mlxsw_sp_vr_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_vr *vr) +{ + mlxsw_sp_vr_lpm_tree_unbind(mlxsw_sp, vr); + mlxsw_sp_lpm_tree_put(mlxsw_sp, vr->lpm_tree); + mlxsw_sp_fib_destroy(vr->fib); + vr->used = false; +} + +static int +mlxsw_sp_vr_lpm_tree_check(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_vr *vr, + struct mlxsw_sp_prefix_usage *req_prefix_usage) +{ + struct mlxsw_sp_lpm_tree *lpm_tree; + + if (mlxsw_sp_prefix_usage_eq(req_prefix_usage, + &vr->lpm_tree->prefix_usage)) + return 0; + + lpm_tree = mlxsw_sp_lpm_tree_get(mlxsw_sp, req_prefix_usage, + vr->proto, false); + if (IS_ERR(lpm_tree)) { + /* We failed to get a tree according to the required + * prefix usage. However, the current tree might be still good + * for us if our requirement is subset of the prefixes used + * in the tree. + */ + if (mlxsw_sp_prefix_usage_subset(req_prefix_usage, + &vr->lpm_tree->prefix_usage)) + return 0; + return PTR_ERR(lpm_tree); + } + + mlxsw_sp_vr_lpm_tree_unbind(mlxsw_sp, vr); + mlxsw_sp_lpm_tree_put(mlxsw_sp, vr->lpm_tree); + vr->lpm_tree = lpm_tree; + return mlxsw_sp_vr_lpm_tree_bind(mlxsw_sp, vr); +} + +static struct mlxsw_sp_vr *mlxsw_sp_vr_get(struct mlxsw_sp *mlxsw_sp, + unsigned char prefix_len, + u32 tb_id, + enum mlxsw_sp_l3proto proto) +{ + struct mlxsw_sp_vr *vr; + int err; + + tb_id = mlxsw_sp_fix_tb_id(tb_id); + vr = mlxsw_sp_vr_find(mlxsw_sp, tb_id, proto); + if (!vr) { + vr = mlxsw_sp_vr_create(mlxsw_sp, prefix_len, tb_id, proto); + if (IS_ERR(vr)) + return vr; + } else { + struct mlxsw_sp_prefix_usage req_prefix_usage; + + mlxsw_sp_prefix_usage_cpy(&req_prefix_usage, + &vr->fib->prefix_usage); + mlxsw_sp_prefix_usage_set(&req_prefix_usage, prefix_len); + /* Need to replace LPM tree in case new prefix is required. */ + err = mlxsw_sp_vr_lpm_tree_check(mlxsw_sp, vr, + &req_prefix_usage); + if (err) + return ERR_PTR(err); + } + return vr; +} + +static void mlxsw_sp_vr_put(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_vr *vr) +{ + /* Destroy virtual router entity in case the associated FIB is empty + * and allow it to be used for other tables in future. Otherwise, + * check if some prefix usage did not disappear and change tree if + * that is the case. Note that in case new, smaller tree cannot be + * allocated, the original one will be kept being used. + */ + if (mlxsw_sp_prefix_usage_none(&vr->fib->prefix_usage)) + mlxsw_sp_vr_destroy(mlxsw_sp, vr); + else + mlxsw_sp_vr_lpm_tree_check(mlxsw_sp, vr, + &vr->fib->prefix_usage); +} + +static void mlxsw_sp_vrs_init(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_sp_vr *vr; + int i; + + for (i = 0; i < MLXSW_SP_VIRTUAL_ROUTER_MAX; i++) { + vr = &mlxsw_sp->router.vrs[i]; + vr->id = i; + } +} + static int __mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) { char rgcr_pl[MLXSW_REG_RGCR_LEN]; @@ -332,6 +559,7 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) if (err) return err; mlxsw_sp_lpm_init(mlxsw_sp); + mlxsw_sp_vrs_init(mlxsw_sp); return 0; } -- cgit v0.10.2 From d5a1c749d22c6f86c3ee2c2c38b13220ae871114 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 4 Jul 2016 08:23:10 +0200 Subject: mlxsw: reg: Add Router Algorithmic LPM Unicast Entry Register definition Serves for adding, updating and removing fib entries. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 0b7b91f..9280d96 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -3622,6 +3622,268 @@ static inline void mlxsw_reg_raltb_pack(char *payload, u16 virtual_router, mlxsw_reg_raltb_tree_id_set(payload, tree_id); } +/* RALUE - Router Algorithmic LPM Unicast Entry Register + * ----------------------------------------------------- + * RALUE is used to configure and query LPM entries that serve + * the Unicast protocols. + */ +#define MLXSW_REG_RALUE_ID 0x8013 +#define MLXSW_REG_RALUE_LEN 0x38 + +static const struct mlxsw_reg_info mlxsw_reg_ralue = { + .id = MLXSW_REG_RALUE_ID, + .len = MLXSW_REG_RALUE_LEN, +}; + +/* reg_ralue_protocol + * Protocol. + * Access: Index + */ +MLXSW_ITEM32(reg, ralue, protocol, 0x00, 24, 4); + +enum mlxsw_reg_ralue_op { + /* Read operation. If entry doesn't exist, the operation fails. */ + MLXSW_REG_RALUE_OP_QUERY_READ = 0, + /* Clear on read operation. Used to read entry and + * clear Activity bit. + */ + MLXSW_REG_RALUE_OP_QUERY_CLEAR = 1, + /* Write operation. Used to write a new entry to the table. All RW + * fields are written for new entry. Activity bit is set + * for new entries. + */ + MLXSW_REG_RALUE_OP_WRITE_WRITE = 0, + /* Update operation. Used to update an existing route entry and + * only update the RW fields that are detailed in the field + * op_u_mask. If entry doesn't exist, the operation fails. + */ + MLXSW_REG_RALUE_OP_WRITE_UPDATE = 1, + /* Clear activity. The Activity bit (the field a) is cleared + * for the entry. + */ + MLXSW_REG_RALUE_OP_WRITE_CLEAR = 2, + /* Delete operation. Used to delete an existing entry. If entry + * doesn't exist, the operation fails. + */ + MLXSW_REG_RALUE_OP_WRITE_DELETE = 3, +}; + +/* reg_ralue_op + * Operation. + * Access: OP + */ +MLXSW_ITEM32(reg, ralue, op, 0x00, 20, 3); + +/* reg_ralue_a + * Activity. Set for new entries. Set if a packet lookup has hit on the + * specific entry, only if the entry is a route. To clear the a bit, use + * "clear activity" op. + * Enabled by activity_dis in RGCR + * Access: RO + */ +MLXSW_ITEM32(reg, ralue, a, 0x00, 16, 1); + +/* reg_ralue_virtual_router + * Virtual Router ID + * Range is 0..cap_max_virtual_routers-1 + * Access: Index + */ +MLXSW_ITEM32(reg, ralue, virtual_router, 0x04, 16, 16); + +#define MLXSW_REG_RALUE_OP_U_MASK_ENTRY_TYPE BIT(0) +#define MLXSW_REG_RALUE_OP_U_MASK_BMP_LEN BIT(1) +#define MLXSW_REG_RALUE_OP_U_MASK_ACTION BIT(2) + +/* reg_ralue_op_u_mask + * opcode update mask. + * On read operation, this field is reserved. + * This field is valid for update opcode, otherwise - reserved. + * This field is a bitmask of the fields that should be updated. + * Access: WO + */ +MLXSW_ITEM32(reg, ralue, op_u_mask, 0x04, 8, 3); + +/* reg_ralue_prefix_len + * Number of bits in the prefix of the LPM route. + * Note that for IPv6 prefixes, if prefix_len>64 the entry consumes + * two entries in the physical HW table. + * Access: Index + */ +MLXSW_ITEM32(reg, ralue, prefix_len, 0x08, 0, 8); + +/* reg_ralue_dip* + * The prefix of the route or of the marker that the object of the LPM + * is compared with. The most significant bits of the dip are the prefix. + * The list significant bits must be '0' if the prefix_len is smaller + * than 128 for IPv6 or smaller than 32 for IPv4. + * IPv4 address uses bits dip[31:0] and bits dip[127:32] are reserved. + * Access: Index + */ +MLXSW_ITEM32(reg, ralue, dip4, 0x18, 0, 32); + +enum mlxsw_reg_ralue_entry_type { + MLXSW_REG_RALUE_ENTRY_TYPE_MARKER_ENTRY = 1, + MLXSW_REG_RALUE_ENTRY_TYPE_ROUTE_ENTRY = 2, + MLXSW_REG_RALUE_ENTRY_TYPE_MARKER_AND_ROUTE_ENTRY = 3, +}; + +/* reg_ralue_entry_type + * Entry type. + * Note - for Marker entries, the action_type and action fields are reserved. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, entry_type, 0x1C, 30, 2); + +/* reg_ralue_bmp_len + * The best match prefix length in the case that there is no match for + * longer prefixes. + * If (entry_type != MARKER_ENTRY), bmp_len must be equal to prefix_len + * Note for any update operation with entry_type modification this + * field must be set. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, bmp_len, 0x1C, 16, 8); + +enum mlxsw_reg_ralue_action_type { + MLXSW_REG_RALUE_ACTION_TYPE_REMOTE, + MLXSW_REG_RALUE_ACTION_TYPE_LOCAL, + MLXSW_REG_RALUE_ACTION_TYPE_IP2ME, +}; + +/* reg_ralue_action_type + * Action Type + * Indicates how the IP address is connected. + * It can be connected to a local subnet through local_erif or can be + * on a remote subnet connected through a next-hop router, + * or transmitted to the CPU. + * Reserved when entry_type = MARKER_ENTRY + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, action_type, 0x1C, 0, 2); + +enum mlxsw_reg_ralue_trap_action { + MLXSW_REG_RALUE_TRAP_ACTION_NOP, + MLXSW_REG_RALUE_TRAP_ACTION_TRAP, + MLXSW_REG_RALUE_TRAP_ACTION_MIRROR_TO_CPU, + MLXSW_REG_RALUE_TRAP_ACTION_MIRROR, + MLXSW_REG_RALUE_TRAP_ACTION_DISCARD_ERROR, +}; + +/* reg_ralue_trap_action + * Trap action. + * For IP2ME action, only NOP and MIRROR are possible. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, trap_action, 0x20, 28, 4); + +/* reg_ralue_trap_id + * Trap ID to be reported to CPU. + * Trap ID is RTR_INGRESS0 or RTR_INGRESS1. + * For trap_action of NOP, MIRROR and DISCARD_ERROR, trap_id is reserved. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, trap_id, 0x20, 0, 9); + +/* reg_ralue_adjacency_index + * Points to the first entry of the group-based ECMP. + * Only relevant in case of REMOTE action. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, adjacency_index, 0x24, 0, 24); + +/* reg_ralue_ecmp_size + * Amount of sequential entries starting + * from the adjacency_index (the number of ECMPs). + * The valid range is 1-64, 512, 1024, 2048 and 4096. + * Reserved when trap_action is TRAP or DISCARD_ERROR. + * Only relevant in case of REMOTE action. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, ecmp_size, 0x28, 0, 13); + +/* reg_ralue_local_erif + * Egress Router Interface. + * Only relevant in case of LOCAL action. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, local_erif, 0x24, 0, 16); + +/* reg_ralue_v + * Valid bit for the tunnel_ptr field. + * If valid = 0 then trap to CPU as IP2ME trap ID. + * If valid = 1 and the packet format allows NVE or IPinIP tunnel + * decapsulation then tunnel decapsulation is done. + * If valid = 1 and packet format does not allow NVE or IPinIP tunnel + * decapsulation then trap as IP2ME trap ID. + * Only relevant in case of IP2ME action. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, v, 0x24, 31, 1); + +/* reg_ralue_tunnel_ptr + * Tunnel Pointer for NVE or IPinIP tunnel decapsulation. + * For Spectrum, pointer to KVD Linear. + * Only relevant in case of IP2ME action. + * Access: RW + */ +MLXSW_ITEM32(reg, ralue, tunnel_ptr, 0x24, 0, 24); + +static inline void mlxsw_reg_ralue_pack(char *payload, + enum mlxsw_reg_ralxx_protocol protocol, + enum mlxsw_reg_ralue_op op, + u16 virtual_router, u8 prefix_len) +{ + MLXSW_REG_ZERO(ralue, payload); + mlxsw_reg_ralue_protocol_set(payload, protocol); + mlxsw_reg_ralue_virtual_router_set(payload, virtual_router); + mlxsw_reg_ralue_prefix_len_set(payload, prefix_len); + mlxsw_reg_ralue_entry_type_set(payload, + MLXSW_REG_RALUE_ENTRY_TYPE_ROUTE_ENTRY); + mlxsw_reg_ralue_bmp_len_set(payload, prefix_len); +} + +static inline void mlxsw_reg_ralue_pack4(char *payload, + enum mlxsw_reg_ralxx_protocol protocol, + enum mlxsw_reg_ralue_op op, + u16 virtual_router, u8 prefix_len, + u32 dip) +{ + mlxsw_reg_ralue_pack(payload, protocol, op, virtual_router, prefix_len); + mlxsw_reg_ralue_dip4_set(payload, dip); +} + +static inline void +mlxsw_reg_ralue_act_remote_pack(char *payload, + enum mlxsw_reg_ralue_trap_action trap_action, + u16 trap_id, u32 adjacency_index, u16 ecmp_size) +{ + mlxsw_reg_ralue_action_type_set(payload, + MLXSW_REG_RALUE_ACTION_TYPE_REMOTE); + mlxsw_reg_ralue_trap_action_set(payload, trap_action); + mlxsw_reg_ralue_trap_id_set(payload, trap_id); + mlxsw_reg_ralue_adjacency_index_set(payload, adjacency_index); + mlxsw_reg_ralue_ecmp_size_set(payload, ecmp_size); +} + +static inline void +mlxsw_reg_ralue_act_local_pack(char *payload, + enum mlxsw_reg_ralue_trap_action trap_action, + u16 trap_id, u16 local_erif) +{ + mlxsw_reg_ralue_action_type_set(payload, + MLXSW_REG_RALUE_ACTION_TYPE_LOCAL); + mlxsw_reg_ralue_trap_action_set(payload, trap_action); + mlxsw_reg_ralue_trap_id_set(payload, trap_id); + mlxsw_reg_ralue_local_erif_set(payload, local_erif); +} + +static inline void +mlxsw_reg_ralue_act_ip2me_pack(char *payload) +{ + mlxsw_reg_ralue_action_type_set(payload, + MLXSW_REG_RALUE_ACTION_TYPE_IP2ME); +} + /* MFCR - Management Fan Control Register * -------------------------------------- * This register controls the settings of the Fan Speed PWM mechanism. @@ -4370,6 +4632,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "RALST"; case MLXSW_REG_RALTB_ID: return "RALTB"; + case MLXSW_REG_RALUE_ID: + return "RALUE"; case MLXSW_REG_MFCR_ID: return "MFCR"; case MLXSW_REG_MFSC_ID: -- cgit v0.10.2 From 61c503f976b5449e6d7d31f74adbc29d6ce06125 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 4 Jul 2016 08:23:11 +0200 Subject: mlxsw: spectrum_router: Implement fib4 add/del switchdev obj ops Implement ipv4 FIB entries addition and removal. Initially, we support local and broadcast routes using "ip2me" trap action. Also, unicast routes without nexthop are supported using "local" action. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 5b40dfc..877a879 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -476,5 +476,10 @@ static inline void mlxsw_sp_port_dcb_fini(struct mlxsw_sp_port *mlxsw_sp_port) int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp); void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp); +int mlxsw_sp_router_fib4_add(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4, + struct switchdev_trans *trans); +int mlxsw_sp_router_fib4_del(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4); #endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 11dab74..7e3992a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -106,9 +106,19 @@ struct mlxsw_sp_fib_key { unsigned char prefix_len; }; +enum mlxsw_sp_fib_entry_type { + MLXSW_SP_FIB_ENTRY_TYPE_REMOTE, + MLXSW_SP_FIB_ENTRY_TYPE_LOCAL, + MLXSW_SP_FIB_ENTRY_TYPE_TRAP, +}; + struct mlxsw_sp_fib_entry { struct rhash_head ht_node; struct mlxsw_sp_fib_key key; + enum mlxsw_sp_fib_entry_type type; + u8 added:1; + u16 rif; /* used for action local */ + struct mlxsw_sp_vr *vr; }; struct mlxsw_sp_fib { @@ -567,3 +577,238 @@ void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) { __mlxsw_sp_router_fini(mlxsw_sp); } + +static int mlxsw_sp_fib_entry_op4_local(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) +{ + char ralue_pl[MLXSW_REG_RALUE_LEN]; + u32 *p_dip = (u32 *) fib_entry->key.addr; + struct mlxsw_sp_vr *vr = fib_entry->vr; + + mlxsw_reg_ralue_pack4(ralue_pl, vr->proto, op, vr->id, + fib_entry->key.prefix_len, *p_dip); + mlxsw_reg_ralue_act_local_pack(ralue_pl, + MLXSW_REG_RALUE_TRAP_ACTION_NOP, 0, + fib_entry->rif); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue), ralue_pl); +} + +static int mlxsw_sp_fib_entry_op4_trap(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) +{ + char ralue_pl[MLXSW_REG_RALUE_LEN]; + u32 *p_dip = (u32 *) fib_entry->key.addr; + struct mlxsw_sp_vr *vr = fib_entry->vr; + + mlxsw_reg_ralue_pack4(ralue_pl, vr->proto, op, vr->id, + fib_entry->key.prefix_len, *p_dip); + mlxsw_reg_ralue_act_ip2me_pack(ralue_pl); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue), ralue_pl); +} + +static int mlxsw_sp_fib_entry_op4(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) +{ + switch (fib_entry->type) { + case MLXSW_SP_FIB_ENTRY_TYPE_REMOTE: + return -EINVAL; + case MLXSW_SP_FIB_ENTRY_TYPE_LOCAL: + return mlxsw_sp_fib_entry_op4_local(mlxsw_sp, fib_entry, op); + case MLXSW_SP_FIB_ENTRY_TYPE_TRAP: + return mlxsw_sp_fib_entry_op4_trap(mlxsw_sp, fib_entry, op); + } + return -EINVAL; +} + +static int mlxsw_sp_fib_entry_op(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) +{ + switch (fib_entry->vr->proto) { + case MLXSW_SP_L3_PROTO_IPV4: + return mlxsw_sp_fib_entry_op4(mlxsw_sp, fib_entry, op); + case MLXSW_SP_L3_PROTO_IPV6: + return -EINVAL; + } + return -EINVAL; +} + +static int mlxsw_sp_fib_entry_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) +{ + enum mlxsw_reg_ralue_op op; + + op = !fib_entry->added ? MLXSW_REG_RALUE_OP_WRITE_WRITE : + MLXSW_REG_RALUE_OP_WRITE_UPDATE; + return mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry, op); +} + +static int mlxsw_sp_fib_entry_del(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) +{ + return mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry, + MLXSW_REG_RALUE_OP_WRITE_DELETE); +} + +struct mlxsw_sp_router_fib4_add_info { + struct switchdev_trans_item tritem; + struct mlxsw_sp *mlxsw_sp; + struct mlxsw_sp_fib_entry *fib_entry; +}; + +static void mlxsw_sp_router_fib4_add_info_destroy(void const *data) +{ + const struct mlxsw_sp_router_fib4_add_info *info = data; + struct mlxsw_sp_fib_entry *fib_entry = info->fib_entry; + struct mlxsw_sp *mlxsw_sp = info->mlxsw_sp; + + mlxsw_sp_fib_entry_destroy(fib_entry); + mlxsw_sp_vr_put(mlxsw_sp, fib_entry->vr); + kfree(info); +} + +static int +mlxsw_sp_router_fib4_entry_init(struct mlxsw_sp *mlxsw_sp, + const struct switchdev_obj_ipv4_fib *fib4, + struct mlxsw_sp_fib_entry *fib_entry) +{ + struct fib_info *fi = fib4->fi; + + if (fib4->type == RTN_LOCAL || fib4->type == RTN_BROADCAST) { + fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP; + return 0; + } + if (fib4->type != RTN_UNICAST) + return -EINVAL; + + if (fi->fib_scope != RT_SCOPE_UNIVERSE) { + struct mlxsw_sp_rif *r; + + fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_LOCAL; + r = mlxsw_sp_rif_find_by_dev(mlxsw_sp, fi->fib_dev); + if (!r) + return -EINVAL; + fib_entry->rif = r->rif; + return 0; + } + return -EINVAL; +} + +static int +mlxsw_sp_router_fib4_add_prepare(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4, + struct switchdev_trans *trans) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_router_fib4_add_info *info; + struct mlxsw_sp_fib_entry *fib_entry; + struct mlxsw_sp_vr *vr; + int err; + + vr = mlxsw_sp_vr_get(mlxsw_sp, fib4->dst_len, fib4->tb_id, + MLXSW_SP_L3_PROTO_IPV4); + if (IS_ERR(vr)) + return PTR_ERR(vr); + + fib_entry = mlxsw_sp_fib_entry_create(vr->fib, &fib4->dst, + sizeof(fib4->dst), fib4->dst_len); + if (!fib_entry) { + err = -ENOMEM; + goto err_fib_entry_create; + } + fib_entry->vr = vr; + + err = mlxsw_sp_router_fib4_entry_init(mlxsw_sp, fib4, fib_entry); + if (err) + goto err_fib4_entry_init; + + info = kmalloc(sizeof(*info), GFP_KERNEL); + if (!info) { + err = -ENOMEM; + goto err_alloc_info; + } + info->mlxsw_sp = mlxsw_sp; + info->fib_entry = fib_entry; + switchdev_trans_item_enqueue(trans, info, + mlxsw_sp_router_fib4_add_info_destroy, + &info->tritem); + return 0; + +err_alloc_info: +err_fib4_entry_init: + mlxsw_sp_fib_entry_destroy(fib_entry); +err_fib_entry_create: + mlxsw_sp_vr_put(mlxsw_sp, vr); + return err; +} + +static int +mlxsw_sp_router_fib4_add_commit(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4, + struct switchdev_trans *trans) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_router_fib4_add_info *info; + struct mlxsw_sp_fib_entry *fib_entry; + struct mlxsw_sp_vr *vr; + int err; + + info = switchdev_trans_item_dequeue(trans); + fib_entry = info->fib_entry; + kfree(info); + + vr = fib_entry->vr; + err = mlxsw_sp_fib_entry_insert(fib_entry->vr->fib, fib_entry); + if (err) + goto err_fib_entry_insert; + err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry); + if (err) + goto err_fib_entry_add; + return 0; + +err_fib_entry_add: + mlxsw_sp_fib_entry_remove(vr->fib, fib_entry); +err_fib_entry_insert: + mlxsw_sp_fib_entry_destroy(fib_entry); + mlxsw_sp_vr_put(mlxsw_sp, vr); + return err; +} + +int mlxsw_sp_router_fib4_add(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4, + struct switchdev_trans *trans) +{ + if (switchdev_trans_ph_prepare(trans)) + return mlxsw_sp_router_fib4_add_prepare(mlxsw_sp_port, + fib4, trans); + return mlxsw_sp_router_fib4_add_commit(mlxsw_sp_port, + fib4, trans); +} + +int mlxsw_sp_router_fib4_del(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_ipv4_fib *fib4) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_fib_entry *fib_entry; + struct mlxsw_sp_vr *vr; + + vr = mlxsw_sp_vr_find(mlxsw_sp, fib4->tb_id, MLXSW_SP_L3_PROTO_IPV4); + if (!vr) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to find virtual router for FIB4 entry being removed.\n"); + return -ENOENT; + } + fib_entry = mlxsw_sp_fib_entry_lookup(vr->fib, &fib4->dst, + sizeof(fib4->dst), fib4->dst_len); + if (!fib_entry) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to find FIB4 entry being removed.\n"); + return PTR_ERR(vr); + } + mlxsw_sp_fib_entry_del(mlxsw_sp_port->mlxsw_sp, fib_entry); + mlxsw_sp_fib_entry_remove(vr->fib, fib_entry); + mlxsw_sp_fib_entry_destroy(fib_entry); + mlxsw_sp_vr_put(mlxsw_sp, vr); + return 0; +} diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 927117e..4decc30 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -978,6 +978,11 @@ static int mlxsw_sp_port_obj_add(struct net_device *dev, SWITCHDEV_OBJ_PORT_VLAN(obj), trans); break; + case SWITCHDEV_OBJ_ID_IPV4_FIB: + err = mlxsw_sp_router_fib4_add(mlxsw_sp_port, + SWITCHDEV_OBJ_IPV4_FIB(obj), + trans); + break; case SWITCHDEV_OBJ_ID_PORT_FDB: err = mlxsw_sp_port_fdb_static_add(mlxsw_sp_port, SWITCHDEV_OBJ_PORT_FDB(obj), @@ -1123,6 +1128,10 @@ static int mlxsw_sp_port_obj_del(struct net_device *dev, err = mlxsw_sp_port_vlans_del(mlxsw_sp_port, SWITCHDEV_OBJ_PORT_VLAN(obj)); break; + case SWITCHDEV_OBJ_ID_IPV4_FIB: + err = mlxsw_sp_router_fib4_del(mlxsw_sp_port, + SWITCHDEV_OBJ_IPV4_FIB(obj)); + break; case SWITCHDEV_OBJ_ID_PORT_FDB: err = mlxsw_sp_port_fdb_static_del(mlxsw_sp_port, SWITCHDEV_OBJ_PORT_FDB(obj)); -- cgit v0.10.2 From 7ce856aaaf13a5dc969ac5f998e5daaf1abe4cd2 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 4 Jul 2016 08:23:12 +0200 Subject: mlxsw: spectrum: Add couple of lower device helper functions Add functions that iterate over lower devices and find port device. As a dependency add netdev_for_each_all_lower_dev and netdev_for_each_all_lower_dev_rcu macro with netdev_all_lower_get_next and netdev_all_lower_get_next_rcu shelpers. Also, add functions to return mlxsw struct according to lower device found and mlxsw_port struct with a reference to lower device. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index f0799898..f54fd6a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2567,6 +2567,66 @@ static struct mlxsw_driver mlxsw_sp_driver = { .profile = &mlxsw_sp_config_profile, }; +static bool mlxsw_sp_port_dev_check(const struct net_device *dev) +{ + return dev->netdev_ops == &mlxsw_sp_port_netdev_ops; +} + +static struct mlxsw_sp_port *mlxsw_sp_port_dev_lower_find(struct net_device *dev) +{ + struct net_device *lower_dev; + struct list_head *iter; + + if (mlxsw_sp_port_dev_check(dev)) + return netdev_priv(dev); + + netdev_for_each_all_lower_dev(dev, lower_dev, iter) { + if (mlxsw_sp_port_dev_check(lower_dev)) + return netdev_priv(lower_dev); + } + return NULL; +} + +static struct mlxsw_sp *mlxsw_sp_lower_get(struct net_device *dev) +{ + struct mlxsw_sp_port *mlxsw_sp_port; + + mlxsw_sp_port = mlxsw_sp_port_dev_lower_find(dev); + return mlxsw_sp_port ? mlxsw_sp_port->mlxsw_sp : NULL; +} + +static struct mlxsw_sp_port *mlxsw_sp_port_dev_lower_find_rcu(struct net_device *dev) +{ + struct net_device *lower_dev; + struct list_head *iter; + + if (mlxsw_sp_port_dev_check(dev)) + return netdev_priv(dev); + + netdev_for_each_all_lower_dev_rcu(dev, lower_dev, iter) { + if (mlxsw_sp_port_dev_check(lower_dev)) + return netdev_priv(lower_dev); + } + return NULL; +} + +struct mlxsw_sp_port *mlxsw_sp_port_lower_dev_hold(struct net_device *dev) +{ + struct mlxsw_sp_port *mlxsw_sp_port; + + rcu_read_lock(); + mlxsw_sp_port = mlxsw_sp_port_dev_lower_find_rcu(dev); + if (mlxsw_sp_port) + dev_hold(mlxsw_sp_port->dev); + rcu_read_unlock(); + return mlxsw_sp_port; +} + +void mlxsw_sp_port_dev_put(struct mlxsw_sp_port *mlxsw_sp_port) +{ + dev_put(mlxsw_sp_port->dev); +} + static bool mlxsw_sp_lag_port_fid_member(struct mlxsw_sp_port *lag_port, u16 fid) { @@ -2647,11 +2707,6 @@ int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) return mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_port, fid); } -static bool mlxsw_sp_port_dev_check(const struct net_device *dev) -{ - return dev->netdev_ops == &mlxsw_sp_port_netdev_ops; -} - static bool mlxsw_sp_master_bridge_check(struct mlxsw_sp *mlxsw_sp, struct net_device *br_dev) { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 877a879..fefff25 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -292,6 +292,9 @@ struct mlxsw_sp_port { struct list_head vports_list; }; +struct mlxsw_sp_port *mlxsw_sp_port_lower_dev_hold(struct net_device *dev); +void mlxsw_sp_port_dev_put(struct mlxsw_sp_port *mlxsw_sp_port); + static inline bool mlxsw_sp_port_is_pause_en(const struct mlxsw_sp_port *mlxsw_sp_port) { diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 7dc2ec7..0c6ee2c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3804,12 +3804,30 @@ void *netdev_lower_get_next_private_rcu(struct net_device *dev, void *netdev_lower_get_next(struct net_device *dev, struct list_head **iter); + #define netdev_for_each_lower_dev(dev, ldev, iter) \ for (iter = (dev)->adj_list.lower.next, \ ldev = netdev_lower_get_next(dev, &(iter)); \ ldev; \ ldev = netdev_lower_get_next(dev, &(iter))) +struct net_device *netdev_all_lower_get_next(struct net_device *dev, + struct list_head **iter); +struct net_device *netdev_all_lower_get_next_rcu(struct net_device *dev, + struct list_head **iter); + +#define netdev_for_each_all_lower_dev(dev, ldev, iter) \ + for (iter = (dev)->all_adj_list.lower.next, \ + ldev = netdev_all_lower_get_next(dev, &(iter)); \ + ldev; \ + ldev = netdev_all_lower_get_next(dev, &(iter))) + +#define netdev_for_each_all_lower_dev_rcu(dev, ldev, iter) \ + for (iter = (dev)->all_adj_list.lower.next, \ + ldev = netdev_all_lower_get_next_rcu(dev, &(iter)); \ + ldev; \ + ldev = netdev_all_lower_get_next_rcu(dev, &(iter))) + void *netdev_adjacent_get_private(struct list_head *adj_list); void *netdev_lower_get_first_private_rcu(struct net_device *dev); struct net_device *netdev_master_upper_dev_get(struct net_device *dev); diff --git a/net/core/dev.c b/net/core/dev.c index aba10d2..a4f3b0a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5445,6 +5445,52 @@ void *netdev_lower_get_next(struct net_device *dev, struct list_head **iter) EXPORT_SYMBOL(netdev_lower_get_next); /** + * netdev_all_lower_get_next - Get the next device from all lower neighbour list + * @dev: device + * @iter: list_head ** of the current position + * + * Gets the next netdev_adjacent from the dev's all lower neighbour + * list, starting from iter position. The caller must hold RTNL lock or + * its own locking that guarantees that the neighbour all lower + * list will remain unchanged. + */ +struct net_device *netdev_all_lower_get_next(struct net_device *dev, struct list_head **iter) +{ + struct netdev_adjacent *lower; + + lower = list_entry(*iter, struct netdev_adjacent, list); + + if (&lower->list == &dev->all_adj_list.lower) + return NULL; + + *iter = lower->list.next; + + return lower->dev; +} +EXPORT_SYMBOL(netdev_all_lower_get_next); + +/** + * netdev_all_lower_get_next_rcu - Get the next device from all + * lower neighbour list, RCU variant + * @dev: device + * @iter: list_head ** of the current position + * + * Gets the next netdev_adjacent from the dev's all lower neighbour + * list, starting from iter position. The caller must hold RCU read lock. + */ +struct net_device *netdev_all_lower_get_next_rcu(struct net_device *dev, + struct list_head **iter) +{ + struct netdev_adjacent *lower; + + lower = list_first_or_null_rcu(&dev->all_adj_list.lower, + struct netdev_adjacent, list); + + return lower ? lower->dev : NULL; +} +EXPORT_SYMBOL(netdev_all_lower_get_next_rcu); + +/** * netdev_lower_get_first_private_rcu - Get the first ->private from the * lower neighbour list, RCU * variant -- cgit v0.10.2 From 6e095fd4eb8219ce03de27635ffc4563080a7f90 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 4 Jul 2016 08:23:13 +0200 Subject: mlxsw: spectrum: Edit RIF properties based on netdev events We are just about to introduce router interfaces (RIFs), but before that we need to be able update the device with the correct RIF attributes whenever they change for the netdev the RIF is backing. Two such attributes are MTU and MAC. The MAC is used both to set the source MAC of packets egressing from the RIF and also to program an FDB rule that will direct packets to the router block. Use the existing netdevice notification block and respond to CHANGEADDR and CHANGEMTU accordingly. Store both attributes in the RIF struct in case we need to revert to old attributes following a failed update. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index f54fd6a..284f6ab 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2627,6 +2627,63 @@ void mlxsw_sp_port_dev_put(struct mlxsw_sp_port *mlxsw_sp_port) dev_put(mlxsw_sp_port->dev); } +static int mlxsw_sp_rif_edit(struct mlxsw_sp *mlxsw_sp, u16 rif, + const char *mac, int mtu) +{ + char ritr_pl[MLXSW_REG_RITR_LEN]; + int err; + + mlxsw_reg_ritr_rif_pack(ritr_pl, rif); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl); + if (err) + return err; + + mlxsw_reg_ritr_mtu_set(ritr_pl, mtu); + mlxsw_reg_ritr_if_mac_memcpy_to(ritr_pl, mac); + mlxsw_reg_ritr_op_set(ritr_pl, MLXSW_REG_RITR_RIF_CREATE); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl); +} + +static int mlxsw_sp_netdevice_router_port_event(struct net_device *dev) +{ + struct mlxsw_sp *mlxsw_sp; + struct mlxsw_sp_rif *r; + int err; + + mlxsw_sp = mlxsw_sp_lower_get(dev); + if (!mlxsw_sp) + return 0; + + r = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); + if (!r) + return 0; + + err = mlxsw_sp_rif_fdb_op(mlxsw_sp, r->addr, r->f->fid, false); + if (err) + return err; + + err = mlxsw_sp_rif_edit(mlxsw_sp, r->rif, dev->dev_addr, dev->mtu); + if (err) + goto err_rif_edit; + + err = mlxsw_sp_rif_fdb_op(mlxsw_sp, dev->dev_addr, r->f->fid, true); + if (err) + goto err_rif_fdb_op; + + ether_addr_copy(r->addr, dev->dev_addr); + r->mtu = dev->mtu; + + netdev_dbg(dev, "Updated RIF=%d\n", r->rif); + + return 0; + +err_rif_fdb_op: + mlxsw_sp_rif_edit(mlxsw_sp, r->rif, r->addr, r->mtu); +err_rif_edit: + mlxsw_sp_rif_fdb_op(mlxsw_sp, r->addr, r->f->fid, true); + return err; +} + static bool mlxsw_sp_lag_port_fid_member(struct mlxsw_sp_port *lag_port, u16 fid) { @@ -3487,7 +3544,9 @@ static int mlxsw_sp_netdevice_event(struct notifier_block *unused, struct net_device *dev = netdev_notifier_info_to_dev(ptr); int err = 0; - if (mlxsw_sp_port_dev_check(dev)) + if (event == NETDEV_CHANGEADDR || event == NETDEV_CHANGEMTU) + err = mlxsw_sp_netdevice_router_port_event(dev); + else if (mlxsw_sp_port_dev_check(dev)) err = mlxsw_sp_netdevice_port_event(dev, event, ptr); else if (netif_is_lag_master(dev)) err = mlxsw_sp_netdevice_lag_event(dev, event, ptr); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index fefff25..0d3e0e3 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -107,6 +107,9 @@ struct mlxsw_sp_fid { struct mlxsw_sp_rif { struct net_device *dev; + struct mlxsw_sp_fid *f; + unsigned char addr[ETH_ALEN]; + int mtu; u16 rif; }; @@ -448,6 +451,8 @@ int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, void mlxsw_sp_port_active_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port); int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid); int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid); +int mlxsw_sp_rif_fdb_op(struct mlxsw_sp *mlxsw_sp, const char *mac, u16 fid, + bool adding); int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index, bool dwrr, u8 dwrr_weight); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 4decc30..06f433a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -753,9 +753,10 @@ static enum mlxsw_reg_sfd_op mlxsw_sp_sfd_op(bool adding) MLXSW_REG_SFD_OP_WRITE_REMOVE; } -static int mlxsw_sp_port_fdb_uc_op(struct mlxsw_sp *mlxsw_sp, u8 local_port, - const char *mac, u16 fid, bool adding, - bool dynamic) +static int __mlxsw_sp_port_fdb_uc_op(struct mlxsw_sp *mlxsw_sp, u8 local_port, + const char *mac, u16 fid, bool adding, + enum mlxsw_reg_sfd_rec_action action, + bool dynamic) { char *sfd_pl; int err; @@ -766,14 +767,29 @@ static int mlxsw_sp_port_fdb_uc_op(struct mlxsw_sp *mlxsw_sp, u8 local_port, mlxsw_reg_sfd_pack(sfd_pl, mlxsw_sp_sfd_op(adding), 0); mlxsw_reg_sfd_uc_pack(sfd_pl, 0, mlxsw_sp_sfd_rec_policy(dynamic), - mac, fid, MLXSW_REG_SFD_REC_ACTION_NOP, - local_port); + mac, fid, action, local_port); err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfd), sfd_pl); kfree(sfd_pl); return err; } +static int mlxsw_sp_port_fdb_uc_op(struct mlxsw_sp *mlxsw_sp, u8 local_port, + const char *mac, u16 fid, bool adding, + bool dynamic) +{ + return __mlxsw_sp_port_fdb_uc_op(mlxsw_sp, local_port, mac, fid, adding, + MLXSW_REG_SFD_REC_ACTION_NOP, dynamic); +} + +int mlxsw_sp_rif_fdb_op(struct mlxsw_sp *mlxsw_sp, const char *mac, u16 fid, + bool adding) +{ + return __mlxsw_sp_port_fdb_uc_op(mlxsw_sp, 0, mac, fid, adding, + MLXSW_REG_SFD_REC_ACTION_FORWARD_IP_ROUTER, + false); +} + static int mlxsw_sp_port_fdb_uc_lag_op(struct mlxsw_sp *mlxsw_sp, u16 lag_id, const char *mac, u16 fid, u16 lag_vid, bool adding, bool dynamic) -- cgit v0.10.2 From 99724c18fc6609a9ac593ce00c1e13188f6a8ddf Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 4 Jul 2016 08:23:14 +0200 Subject: mlxsw: spectrum: Introduce support for router interfaces Up until now we only supported bridged interfaces. Packets ingressing through the switch ports were either classified to FIDs (in the case of the VLAN-aware bridge) or vFIDs (in the case of VLAN-unaware bridges). The packets were then forwarded according to the FDB. Routing was done entirely in slowpath, by splitting the vFID range in two and using the lower 0.5K vFIDs as dummy bridges that simply flooded all incoming traffic to the CPU. Instead, allow packets to be routed in the device by creating router interfaces (RIFs) that will direct them to the router block. Specifically, the RIFs introduced here are Sub-port RIFs used for VLAN devices and port netdevs. Packets ingressing from the {Port / LAG ID, VID} with which the RIF was programmed with will be assigned to a special kind of FIDs called rFIDs and from there directed to the router. Create a RIF whenever the first IPv4 address was programmed on a VLAN / LAG / port netdev. Destroy it upon removal of the last IPv4 address. Receive these notifications by registering for the 'inetaddr' notification chain. A non-zero (10) priority is used for the notification block, so that RIFs will be created before routes are offloaded via FIB code. Note that another trigger for RIF destruction are CHANGEUPPER notifications causing the underlying FID's reference count to go down to zero. This can happen, for example, when a VLAN netdev with an IP address is put under bridge. While this configuration doesn't make sense it does cause the device and the kernel to get out of sync when the netdev is unbridged. We intend to address this in the future, hopefully in current cycle. Finally, Remove the lower 0.5K vFIDs, as they are deprecated by the RIFs, which will trap packets according to their DIP. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 284f6ab..bd8448a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -51,6 +51,7 @@ #include #include #include +#include #include #include @@ -210,23 +211,6 @@ static int mlxsw_sp_port_dev_addr_init(struct mlxsw_sp_port *mlxsw_sp_port) return mlxsw_sp_port_dev_addr_set(mlxsw_sp_port, addr); } -static int mlxsw_sp_port_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port, - u16 vid, enum mlxsw_reg_spms_state state) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - char *spms_pl; - int err; - - spms_pl = kmalloc(MLXSW_REG_SPMS_LEN, GFP_KERNEL); - if (!spms_pl) - return -ENOMEM; - mlxsw_reg_spms_pack(spms_pl, mlxsw_sp_port->local_port); - mlxsw_reg_spms_vid_pack(spms_pl, vid, state); - err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spms), spms_pl); - kfree(spms_pl); - return err; -} - static int mlxsw_sp_port_mtu_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; @@ -637,87 +621,6 @@ static int mlxsw_sp_port_vlan_mode_trans(struct mlxsw_sp_port *mlxsw_sp_port) return 0; } -static struct mlxsw_sp_fid * -mlxsw_sp_vfid_find(const struct mlxsw_sp *mlxsw_sp, u16 vid) -{ - struct mlxsw_sp_fid *f; - - list_for_each_entry(f, &mlxsw_sp->port_vfids.list, list) { - if (f->vid == vid) - return f; - } - - return NULL; -} - -static u16 mlxsw_sp_avail_vfid_get(const struct mlxsw_sp *mlxsw_sp) -{ - return find_first_zero_bit(mlxsw_sp->port_vfids.mapped, - MLXSW_SP_VFID_PORT_MAX); -} - -static int mlxsw_sp_vfid_op(struct mlxsw_sp *mlxsw_sp, u16 fid, bool create) -{ - char sfmr_pl[MLXSW_REG_SFMR_LEN]; - - mlxsw_reg_sfmr_pack(sfmr_pl, !create, fid, 0); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); -} - -static void mlxsw_sp_vport_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport); - -static struct mlxsw_sp_fid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, - u16 vid) -{ - struct device *dev = mlxsw_sp->bus_info->dev; - struct mlxsw_sp_fid *f; - u16 vfid, fid; - int err; - - vfid = mlxsw_sp_avail_vfid_get(mlxsw_sp); - if (vfid == MLXSW_SP_VFID_PORT_MAX) { - dev_err(dev, "No available vFIDs\n"); - return ERR_PTR(-ERANGE); - } - - fid = mlxsw_sp_vfid_to_fid(vfid); - err = mlxsw_sp_vfid_op(mlxsw_sp, fid, true); - if (err) { - dev_err(dev, "Failed to create FID=%d\n", fid); - return ERR_PTR(err); - } - - f = kzalloc(sizeof(*f), GFP_KERNEL); - if (!f) - goto err_allocate_vfid; - - f->leave = mlxsw_sp_vport_vfid_leave; - f->fid = fid; - f->vid = vid; - - list_add(&f->list, &mlxsw_sp->port_vfids.list); - set_bit(vfid, mlxsw_sp->port_vfids.mapped); - - return f; - -err_allocate_vfid: - mlxsw_sp_vfid_op(mlxsw_sp, fid, false); - return ERR_PTR(-ENOMEM); -} - -static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_fid *f) -{ - u16 vfid = mlxsw_sp_fid_to_vfid(f->fid); - - clear_bit(vfid, mlxsw_sp->port_vfids.mapped); - list_del(&f->list); - - mlxsw_sp_vfid_op(mlxsw_sp, f->fid, false); - - kfree(f); -} - static struct mlxsw_sp_port * mlxsw_sp_port_vport_create(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) { @@ -750,67 +653,6 @@ static void mlxsw_sp_port_vport_destroy(struct mlxsw_sp_port *mlxsw_sp_vport) kfree(mlxsw_sp_vport); } -static int mlxsw_sp_vport_fid_map(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, - bool valid) -{ - enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; - u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); - - return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, mt, valid, fid, - vid); -} - -static int mlxsw_sp_vport_vfid_join(struct mlxsw_sp_port *mlxsw_sp_vport) -{ - u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); - struct mlxsw_sp_fid *f; - int err; - - f = mlxsw_sp_vfid_find(mlxsw_sp_vport->mlxsw_sp, vid); - if (!f) { - f = mlxsw_sp_vfid_create(mlxsw_sp_vport->mlxsw_sp, vid); - if (IS_ERR(f)) - return PTR_ERR(f); - } - - if (!f->ref_count) { - err = mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, true); - if (err) - goto err_vport_flood_set; - } - - err = mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, true); - if (err) - goto err_vport_fid_map; - - mlxsw_sp_vport_fid_set(mlxsw_sp_vport, f); - f->ref_count++; - - return 0; - -err_vport_fid_map: - if (!f->ref_count) - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); -err_vport_flood_set: - if (!f->ref_count) - mlxsw_sp_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); - return err; -} - -static void mlxsw_sp_vport_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) -{ - struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); - - mlxsw_sp_vport_fid_set(mlxsw_sp_vport, NULL); - - mlxsw_sp_vport_fid_map(mlxsw_sp_vport, f->fid, false); - - if (--f->ref_count == 0) { - mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); - mlxsw_sp_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); - } -} - int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, u16 vid) { @@ -848,12 +690,6 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, } } - err = mlxsw_sp_vport_vfid_join(mlxsw_sp_vport); - if (err) { - netdev_err(dev, "Failed to join vFID\n"); - goto err_vport_vfid_join; - } - err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); if (err) { netdev_err(dev, "Failed to disable learning for VID=%d\n", vid); @@ -867,22 +703,11 @@ int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, goto err_port_add_vid; } - err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, - MLXSW_REG_SPMS_STATE_FORWARDING); - if (err) { - netdev_err(dev, "Failed to set STP state for VID=%d\n", vid); - goto err_port_stp_state_set; - } - return 0; -err_port_stp_state_set: - mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, false, false); err_port_add_vid: mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true); err_port_vid_learning_set: - mlxsw_sp_vport_vfid_leave(mlxsw_sp_vport); -err_vport_vfid_join: if (list_is_singular(&mlxsw_sp_port->vports_list)) mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port); err_port_vp_mode_trans: @@ -910,13 +735,6 @@ static int mlxsw_sp_port_kill_vid(struct net_device *dev, return 0; } - err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, - MLXSW_REG_SPMS_STATE_DISCARDING); - if (err) { - netdev_err(dev, "Failed to set STP state for VID=%d\n", vid); - return err; - } - err = mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, false, false); if (err) { netdev_err(dev, "Failed to set VLAN membership for VID=%d\n", @@ -2417,7 +2235,6 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->core = mlxsw_core; mlxsw_sp->bus_info = mlxsw_bus_info; INIT_LIST_HEAD(&mlxsw_sp->fids); - INIT_LIST_HEAD(&mlxsw_sp->port_vfids.list); INIT_LIST_HEAD(&mlxsw_sp->br_vfids.list); INIT_LIST_HEAD(&mlxsw_sp->br_mids.list); @@ -2627,6 +2444,311 @@ void mlxsw_sp_port_dev_put(struct mlxsw_sp_port *mlxsw_sp_port) dev_put(mlxsw_sp_port->dev); } +static bool mlxsw_sp_rif_should_config(struct mlxsw_sp_rif *r, + unsigned long event) +{ + switch (event) { + case NETDEV_UP: + if (!r) + return true; + r->ref_count++; + return false; + case NETDEV_DOWN: + if (r && --r->ref_count == 0) + return true; + /* It is possible we already removed the RIF ourselves + * if it was assigned to a netdev that is now a bridge + * or LAG slave. + */ + return false; + } + + return false; +} + +static int mlxsw_sp_avail_rif_get(struct mlxsw_sp *mlxsw_sp) +{ + int i; + + for (i = 0; i < MLXSW_SP_RIF_MAX; i++) + if (!mlxsw_sp->rifs[i]) + return i; + + return MLXSW_SP_RIF_MAX; +} + +static void mlxsw_sp_vport_rif_sp_attr_get(struct mlxsw_sp_port *mlxsw_sp_vport, + bool *p_lagged, u16 *p_system_port) +{ + u8 local_port = mlxsw_sp_vport->local_port; + + *p_lagged = mlxsw_sp_vport->lagged; + *p_system_port = *p_lagged ? mlxsw_sp_vport->lag_id : local_port; +} + +static int mlxsw_sp_vport_rif_sp_op(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *l3_dev, u16 rif, + bool create) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; + bool lagged = mlxsw_sp_vport->lagged; + char ritr_pl[MLXSW_REG_RITR_LEN]; + u16 system_port; + + mlxsw_reg_ritr_pack(ritr_pl, create, MLXSW_REG_RITR_SP_IF, rif, + l3_dev->mtu, l3_dev->dev_addr); + + mlxsw_sp_vport_rif_sp_attr_get(mlxsw_sp_vport, &lagged, &system_port); + mlxsw_reg_ritr_sp_if_pack(ritr_pl, lagged, system_port, + mlxsw_sp_vport_vid_get(mlxsw_sp_vport)); + + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl); +} + +static void mlxsw_sp_vport_rif_sp_leave(struct mlxsw_sp_port *mlxsw_sp_vport); + +static struct mlxsw_sp_fid * +mlxsw_sp_rfid_alloc(u16 fid, struct net_device *l3_dev) +{ + struct mlxsw_sp_fid *f; + + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (!f) + return NULL; + + f->leave = mlxsw_sp_vport_rif_sp_leave; + f->ref_count = 0; + f->dev = l3_dev; + f->fid = fid; + + return f; +} + +static struct mlxsw_sp_rif * +mlxsw_sp_rif_alloc(u16 rif, struct net_device *l3_dev, struct mlxsw_sp_fid *f) +{ + struct mlxsw_sp_rif *r; + + r = kzalloc(sizeof(*r), GFP_KERNEL); + if (!r) + return NULL; + + ether_addr_copy(r->addr, l3_dev->dev_addr); + r->mtu = l3_dev->mtu; + r->ref_count = 1; + r->dev = l3_dev; + r->rif = rif; + r->f = f; + + return r; +} + +static struct mlxsw_sp_rif * +mlxsw_sp_vport_rif_sp_create(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *l3_dev) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; + struct mlxsw_sp_fid *f; + struct mlxsw_sp_rif *r; + u16 fid, rif; + int err; + + rif = mlxsw_sp_avail_rif_get(mlxsw_sp); + if (rif == MLXSW_SP_RIF_MAX) + return ERR_PTR(-ERANGE); + + err = mlxsw_sp_vport_rif_sp_op(mlxsw_sp_vport, l3_dev, rif, true); + if (err) + return ERR_PTR(err); + + fid = mlxsw_sp_rif_sp_to_fid(rif); + err = mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, fid, true); + if (err) + goto err_rif_fdb_op; + + f = mlxsw_sp_rfid_alloc(fid, l3_dev); + if (!f) { + err = -ENOMEM; + goto err_rfid_alloc; + } + + r = mlxsw_sp_rif_alloc(rif, l3_dev, f); + if (!r) { + err = -ENOMEM; + goto err_rif_alloc; + } + + f->r = r; + mlxsw_sp->rifs[rif] = r; + + return r; + +err_rif_alloc: + kfree(f); +err_rfid_alloc: + mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, fid, false); +err_rif_fdb_op: + mlxsw_sp_vport_rif_sp_op(mlxsw_sp_vport, l3_dev, rif, false); + return ERR_PTR(err); +} + +static void mlxsw_sp_vport_rif_sp_destroy(struct mlxsw_sp_port *mlxsw_sp_vport, + struct mlxsw_sp_rif *r) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; + struct net_device *l3_dev = r->dev; + struct mlxsw_sp_fid *f = r->f; + u16 fid = f->fid; + u16 rif = r->rif; + + mlxsw_sp->rifs[rif] = NULL; + f->r = NULL; + + kfree(r); + + kfree(f); + + mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, fid, false); + + mlxsw_sp_vport_rif_sp_op(mlxsw_sp_vport, l3_dev, rif, false); +} + +static int mlxsw_sp_vport_rif_sp_join(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *l3_dev) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_vport->mlxsw_sp; + struct mlxsw_sp_rif *r; + + r = mlxsw_sp_rif_find_by_dev(mlxsw_sp, l3_dev); + if (!r) { + r = mlxsw_sp_vport_rif_sp_create(mlxsw_sp_vport, l3_dev); + if (IS_ERR(r)) + return PTR_ERR(r); + } + + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, r->f); + r->f->ref_count++; + + netdev_dbg(mlxsw_sp_vport->dev, "Joined FID=%d\n", r->f->fid); + + return 0; +} + +static void mlxsw_sp_vport_rif_sp_leave(struct mlxsw_sp_port *mlxsw_sp_vport) +{ + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); + + netdev_dbg(mlxsw_sp_vport->dev, "Left FID=%d\n", f->fid); + + mlxsw_sp_vport_fid_set(mlxsw_sp_vport, NULL); + if (--f->ref_count == 0) + mlxsw_sp_vport_rif_sp_destroy(mlxsw_sp_vport, f->r); +} + +static int mlxsw_sp_inetaddr_vport_event(struct net_device *l3_dev, + struct net_device *port_dev, + unsigned long event, u16 vid) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(port_dev); + struct mlxsw_sp_port *mlxsw_sp_vport; + + mlxsw_sp_vport = mlxsw_sp_port_vport_find(mlxsw_sp_port, vid); + if (WARN_ON(!mlxsw_sp_vport)) + return -EINVAL; + + switch (event) { + case NETDEV_UP: + return mlxsw_sp_vport_rif_sp_join(mlxsw_sp_vport, l3_dev); + case NETDEV_DOWN: + mlxsw_sp_vport_rif_sp_leave(mlxsw_sp_vport); + break; + } + + return 0; +} + +static int mlxsw_sp_inetaddr_port_event(struct net_device *port_dev, + unsigned long event) +{ + if (netif_is_bridge_port(port_dev) || netif_is_lag_port(port_dev)) + return 0; + + return mlxsw_sp_inetaddr_vport_event(port_dev, port_dev, event, 1); +} + +static int __mlxsw_sp_inetaddr_lag_event(struct net_device *l3_dev, + struct net_device *lag_dev, + unsigned long event, u16 vid) +{ + struct net_device *port_dev; + struct list_head *iter; + int err; + + netdev_for_each_lower_dev(lag_dev, port_dev, iter) { + if (mlxsw_sp_port_dev_check(port_dev)) { + err = mlxsw_sp_inetaddr_vport_event(l3_dev, port_dev, + event, vid); + if (err) + return err; + } + } + + return 0; +} + +static int mlxsw_sp_inetaddr_lag_event(struct net_device *lag_dev, + unsigned long event) +{ + if (netif_is_bridge_port(lag_dev)) + return 0; + + return __mlxsw_sp_inetaddr_lag_event(lag_dev, lag_dev, event, 1); +} + +static int mlxsw_sp_inetaddr_vlan_event(struct net_device *vlan_dev, + unsigned long event) +{ + struct net_device *real_dev = vlan_dev_real_dev(vlan_dev); + u16 vid = vlan_dev_vlan_id(vlan_dev); + + if (mlxsw_sp_port_dev_check(real_dev)) + return mlxsw_sp_inetaddr_vport_event(vlan_dev, real_dev, event, + vid); + else if (netif_is_lag_master(real_dev)) + return __mlxsw_sp_inetaddr_lag_event(vlan_dev, real_dev, event, + vid); + + return 0; +} + +static int mlxsw_sp_inetaddr_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct in_ifaddr *ifa = (struct in_ifaddr *) ptr; + struct net_device *dev = ifa->ifa_dev->dev; + struct mlxsw_sp *mlxsw_sp; + struct mlxsw_sp_rif *r; + int err = 0; + + mlxsw_sp = mlxsw_sp_lower_get(dev); + if (!mlxsw_sp) + goto out; + + r = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); + if (!mlxsw_sp_rif_should_config(r, event)) + goto out; + + if (mlxsw_sp_port_dev_check(dev)) + err = mlxsw_sp_inetaddr_port_event(dev, event); + else if (netif_is_lag_master(dev)) + err = mlxsw_sp_inetaddr_lag_event(dev, event); + else if (is_vlan_dev(dev)) + err = mlxsw_sp_inetaddr_vlan_event(dev, event); + +out: + return notifier_from_errno(err); +} + static int mlxsw_sp_rif_edit(struct mlxsw_sp *mlxsw_sp, u16 rif, const char *mac, int mtu) { @@ -3264,20 +3386,18 @@ mlxsw_sp_br_vfid_find(const struct mlxsw_sp *mlxsw_sp, return NULL; } -static u16 mlxsw_sp_vfid_to_br_vfid(u16 vfid) +static u16 mlxsw_sp_avail_br_vfid_get(const struct mlxsw_sp *mlxsw_sp) { - return vfid - MLXSW_SP_VFID_PORT_MAX; + return find_first_zero_bit(mlxsw_sp->br_vfids.mapped, + MLXSW_SP_VFID_MAX); } -static u16 mlxsw_sp_br_vfid_to_vfid(u16 br_vfid) +static int mlxsw_sp_vfid_op(struct mlxsw_sp *mlxsw_sp, u16 fid, bool create) { - return MLXSW_SP_VFID_PORT_MAX + br_vfid; -} + char sfmr_pl[MLXSW_REG_SFMR_LEN]; -static u16 mlxsw_sp_avail_br_vfid_get(const struct mlxsw_sp *mlxsw_sp) -{ - return find_first_zero_bit(mlxsw_sp->br_vfids.mapped, - MLXSW_SP_VFID_BR_MAX); + mlxsw_reg_sfmr_pack(sfmr_pl, !create, fid, 0); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); } static void mlxsw_sp_vport_br_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport); @@ -3290,7 +3410,7 @@ static struct mlxsw_sp_fid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, u16 vfid, fid; int err; - vfid = mlxsw_sp_br_vfid_to_vfid(mlxsw_sp_avail_br_vfid_get(mlxsw_sp)); + vfid = mlxsw_sp_avail_br_vfid_get(mlxsw_sp); if (vfid == MLXSW_SP_VFID_MAX) { dev_err(dev, "No available vFIDs\n"); return ERR_PTR(-ERANGE); @@ -3312,7 +3432,7 @@ static struct mlxsw_sp_fid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, f->dev = br_dev; list_add(&f->list, &mlxsw_sp->br_vfids.list); - set_bit(mlxsw_sp_vfid_to_br_vfid(vfid), mlxsw_sp->br_vfids.mapped); + set_bit(vfid, mlxsw_sp->br_vfids.mapped); return f; @@ -3325,9 +3445,8 @@ static void mlxsw_sp_br_vfid_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *f) { u16 vfid = mlxsw_sp_fid_to_vfid(f->fid); - u16 br_vfid = mlxsw_sp_vfid_to_br_vfid(vfid); - clear_bit(br_vfid, mlxsw_sp->br_vfids.mapped); + clear_bit(vfid, mlxsw_sp->br_vfids.mapped); list_del(&f->list); mlxsw_sp_vfid_op(mlxsw_sp, f->fid, false); @@ -3335,6 +3454,16 @@ static void mlxsw_sp_br_vfid_destroy(struct mlxsw_sp *mlxsw_sp, kfree(f); } +static int mlxsw_sp_vport_fid_map(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, + bool valid) +{ + enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; + u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); + + return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_vport, mt, valid, fid, + vid); +} + static int mlxsw_sp_vport_br_vfid_join(struct mlxsw_sp_port *mlxsw_sp_vport, struct net_device *br_dev) { @@ -3391,16 +3520,18 @@ static void mlxsw_sp_vport_br_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, struct net_device *br_dev) { + struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); u16 vid = mlxsw_sp_vport_vid_get(mlxsw_sp_vport); struct net_device *dev = mlxsw_sp_vport->dev; int err; - mlxsw_sp_vport_vfid_leave(mlxsw_sp_vport); + if (f && !WARN_ON(!f->leave)) + f->leave(mlxsw_sp_vport); err = mlxsw_sp_vport_br_vfid_join(mlxsw_sp_vport, br_dev); if (err) { netdev_err(dev, "Failed to join vFID\n"); - goto err_vport_br_vfid_join; + return err; } err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true); @@ -3418,8 +3549,6 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, err_port_vid_learning_set: mlxsw_sp_vport_br_vfid_leave(mlxsw_sp_vport); -err_vport_br_vfid_join: - mlxsw_sp_vport_vfid_join(mlxsw_sp_vport); return err; } @@ -3431,11 +3560,6 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport) mlxsw_sp_vport_br_vfid_leave(mlxsw_sp_vport); - mlxsw_sp_vport_vfid_join(mlxsw_sp_vport); - - mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid, - MLXSW_REG_SPMS_STATE_FORWARDING); - mlxsw_sp_vport->learning = 0; mlxsw_sp_vport->learning_sync = 0; mlxsw_sp_vport->uc_flood = 0; @@ -3560,11 +3684,17 @@ static struct notifier_block mlxsw_sp_netdevice_nb __read_mostly = { .notifier_call = mlxsw_sp_netdevice_event, }; +static struct notifier_block mlxsw_sp_inetaddr_nb __read_mostly = { + .notifier_call = mlxsw_sp_inetaddr_event, + .priority = 10, /* Must be called before FIB notifier block */ +}; + static int __init mlxsw_sp_module_init(void) { int err; register_netdevice_notifier(&mlxsw_sp_netdevice_nb); + register_inetaddr_notifier(&mlxsw_sp_inetaddr_nb); err = mlxsw_core_driver_register(&mlxsw_sp_driver); if (err) goto err_core_driver_register; @@ -3578,6 +3708,7 @@ err_core_driver_register: static void __exit mlxsw_sp_module_exit(void) { mlxsw_core_driver_unregister(&mlxsw_sp_driver); + unregister_inetaddr_notifier(&mlxsw_sp_inetaddr_nb); unregister_netdevice_notifier(&mlxsw_sp_netdevice_nb); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 0d3e0e3..b6eeab7 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -50,9 +50,10 @@ #include "core.h" #define MLXSW_SP_VFID_BASE VLAN_N_VID -#define MLXSW_SP_VFID_PORT_MAX 512 /* Non-bridged VLAN interfaces */ -#define MLXSW_SP_VFID_BR_MAX 6144 /* Bridged VLAN interfaces */ -#define MLXSW_SP_VFID_MAX (MLXSW_SP_VFID_PORT_MAX + MLXSW_SP_VFID_BR_MAX) +#define MLXSW_SP_VFID_MAX 6656 /* Bridged VLAN interfaces */ + +#define MLXSW_SP_RFID_BASE 15360 +#define MLXSW_SP_RIF_MAX 800 #define MLXSW_SP_LAG_MAX 64 #define MLXSW_SP_PORT_PER_LAG_MAX 16 @@ -81,8 +82,6 @@ #define MLXSW_SP_CELL_FACTOR 2 /* 2 * cell_size / (IPG + cell_size + 1) */ -#define MLXSW_SP_RIF_MAX 800 - static inline u16 mlxsw_sp_pfc_delay_get(int mtu, u16 delay) { delay = MLXSW_SP_BYTES_TO_CELLS(DIV_ROUND_UP(delay, BITS_PER_BYTE)); @@ -101,12 +100,13 @@ struct mlxsw_sp_fid { struct list_head list; unsigned int ref_count; struct net_device *dev; + struct mlxsw_sp_rif *r; u16 fid; - u16 vid; }; struct mlxsw_sp_rif { struct net_device *dev; + unsigned int ref_count; struct mlxsw_sp_fid *f; unsigned char addr[ETH_ALEN]; int mtu; @@ -133,7 +133,17 @@ static inline u16 mlxsw_sp_fid_to_vfid(u16 fid) static inline bool mlxsw_sp_fid_is_vfid(u16 fid) { - return fid >= MLXSW_SP_VFID_BASE; + return fid >= MLXSW_SP_VFID_BASE && fid < MLXSW_SP_RFID_BASE; +} + +static inline bool mlxsw_sp_fid_is_rfid(u16 fid) +{ + return fid >= MLXSW_SP_RFID_BASE; +} + +static inline u16 mlxsw_sp_rif_sp_to_fid(u16 rif) +{ + return MLXSW_SP_RFID_BASE + rif; } struct mlxsw_sp_sb_pr { @@ -207,11 +217,7 @@ struct mlxsw_sp_router { struct mlxsw_sp { struct { struct list_head list; - DECLARE_BITMAP(mapped, MLXSW_SP_VFID_PORT_MAX); - } port_vfids; - struct { - struct list_head list; - DECLARE_BITMAP(mapped, MLXSW_SP_VFID_BR_MAX); + DECLARE_BITMAP(mapped, MLXSW_SP_VFID_MAX); } br_vfids; struct { struct list_head list; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 06f433a..941acd7 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -166,11 +166,6 @@ static int mlxsw_sp_port_attr_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port, return mlxsw_sp_port_stp_state_set(mlxsw_sp_port, state); } -static bool mlxsw_sp_vfid_is_vport_br(u16 vfid) -{ - return vfid >= MLXSW_SP_VFID_PORT_MAX; -} - static int __mlxsw_sp_port_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 idx_begin, u16 idx_end, bool set, bool only_uc) @@ -182,15 +177,10 @@ static int __mlxsw_sp_port_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, char *sftr_pl; int err; - if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) { + if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID; - if (mlxsw_sp_vfid_is_vport_br(idx_begin)) - local_port = mlxsw_sp_port->local_port; - else - local_port = MLXSW_PORT_CPU_PORT; - } else { + else table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFEST; - } sftr_pl = kmalloc(MLXSW_REG_SFTR_LEN, GFP_KERNEL); if (!sftr_pl) -- cgit v0.10.2 From 3ba2ebf4a2cbcf42fe471146d9d668865caa9f21 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 4 Jul 2016 08:23:15 +0200 Subject: mlxsw: spectrum: Unsplit the vFID range Previous commit deprecated the vFIDs used to get traffic to the CPU ('port_vfids'). Thus, we now use the vFIDs as god intended and the artificial split is no longer needed. Rename functions and variables to reflect that. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index bd8448a..e987a8a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2235,7 +2235,7 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, mlxsw_sp->core = mlxsw_core; mlxsw_sp->bus_info = mlxsw_bus_info; INIT_LIST_HEAD(&mlxsw_sp->fids); - INIT_LIST_HEAD(&mlxsw_sp->br_vfids.list); + INIT_LIST_HEAD(&mlxsw_sp->vfids.list); INIT_LIST_HEAD(&mlxsw_sp->br_mids.list); err = mlxsw_sp_base_mac_get(mlxsw_sp); @@ -2320,6 +2320,7 @@ static void mlxsw_sp_fini(struct mlxsw_core *mlxsw_core) mlxsw_sp_buffers_fini(mlxsw_sp); mlxsw_sp_traps_fini(mlxsw_sp); mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE); + WARN_ON(!list_empty(&mlxsw_sp->vfids.list)); WARN_ON(!list_empty(&mlxsw_sp->fids)); for (i = 0; i < MLXSW_SP_RIF_MAX; i++) WARN_ON_ONCE(mlxsw_sp->rifs[i]); @@ -3373,12 +3374,12 @@ static int mlxsw_sp_netdevice_lag_event(struct net_device *lag_dev, } static struct mlxsw_sp_fid * -mlxsw_sp_br_vfid_find(const struct mlxsw_sp *mlxsw_sp, - const struct net_device *br_dev) +mlxsw_sp_vfid_find(const struct mlxsw_sp *mlxsw_sp, + const struct net_device *br_dev) { struct mlxsw_sp_fid *f; - list_for_each_entry(f, &mlxsw_sp->br_vfids.list, list) { + list_for_each_entry(f, &mlxsw_sp->vfids.list, list) { if (f->dev == br_dev) return f; } @@ -3386,9 +3387,9 @@ mlxsw_sp_br_vfid_find(const struct mlxsw_sp *mlxsw_sp, return NULL; } -static u16 mlxsw_sp_avail_br_vfid_get(const struct mlxsw_sp *mlxsw_sp) +static u16 mlxsw_sp_avail_vfid_get(const struct mlxsw_sp *mlxsw_sp) { - return find_first_zero_bit(mlxsw_sp->br_vfids.mapped, + return find_first_zero_bit(mlxsw_sp->vfids.mapped, MLXSW_SP_VFID_MAX); } @@ -3400,17 +3401,17 @@ static int mlxsw_sp_vfid_op(struct mlxsw_sp *mlxsw_sp, u16 fid, bool create) return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); } -static void mlxsw_sp_vport_br_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport); +static void mlxsw_sp_vport_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport); -static struct mlxsw_sp_fid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, - struct net_device *br_dev) +static struct mlxsw_sp_fid *mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev) { struct device *dev = mlxsw_sp->bus_info->dev; struct mlxsw_sp_fid *f; u16 vfid, fid; int err; - vfid = mlxsw_sp_avail_br_vfid_get(mlxsw_sp); + vfid = mlxsw_sp_avail_vfid_get(mlxsw_sp); if (vfid == MLXSW_SP_VFID_MAX) { dev_err(dev, "No available vFIDs\n"); return ERR_PTR(-ERANGE); @@ -3427,12 +3428,12 @@ static struct mlxsw_sp_fid *mlxsw_sp_br_vfid_create(struct mlxsw_sp *mlxsw_sp, if (!f) goto err_allocate_vfid; - f->leave = mlxsw_sp_vport_br_vfid_leave; + f->leave = mlxsw_sp_vport_vfid_leave; f->fid = fid; f->dev = br_dev; - list_add(&f->list, &mlxsw_sp->br_vfids.list); - set_bit(vfid, mlxsw_sp->br_vfids.mapped); + list_add(&f->list, &mlxsw_sp->vfids.list); + set_bit(vfid, mlxsw_sp->vfids.mapped); return f; @@ -3441,12 +3442,12 @@ err_allocate_vfid: return ERR_PTR(-ENOMEM); } -static void mlxsw_sp_br_vfid_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_fid *f) +static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fid *f) { u16 vfid = mlxsw_sp_fid_to_vfid(f->fid); - clear_bit(vfid, mlxsw_sp->br_vfids.mapped); + clear_bit(vfid, mlxsw_sp->vfids.mapped); list_del(&f->list); mlxsw_sp_vfid_op(mlxsw_sp, f->fid, false); @@ -3464,15 +3465,15 @@ static int mlxsw_sp_vport_fid_map(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, vid); } -static int mlxsw_sp_vport_br_vfid_join(struct mlxsw_sp_port *mlxsw_sp_vport, - struct net_device *br_dev) +static int mlxsw_sp_vport_vfid_join(struct mlxsw_sp_port *mlxsw_sp_vport, + struct net_device *br_dev) { struct mlxsw_sp_fid *f; int err; - f = mlxsw_sp_br_vfid_find(mlxsw_sp_vport->mlxsw_sp, br_dev); + f = mlxsw_sp_vfid_find(mlxsw_sp_vport->mlxsw_sp, br_dev); if (!f) { - f = mlxsw_sp_br_vfid_create(mlxsw_sp_vport->mlxsw_sp, br_dev); + f = mlxsw_sp_vfid_create(mlxsw_sp_vport->mlxsw_sp, br_dev); if (IS_ERR(f)) return PTR_ERR(f); } @@ -3496,11 +3497,11 @@ err_vport_fid_map: mlxsw_sp_vport_flood_set(mlxsw_sp_vport, f->fid, false); err_vport_flood_set: if (!f->ref_count) - mlxsw_sp_br_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); + mlxsw_sp_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); return err; } -static void mlxsw_sp_vport_br_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) +static void mlxsw_sp_vport_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) { struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); @@ -3514,7 +3515,7 @@ static void mlxsw_sp_vport_br_vfid_leave(struct mlxsw_sp_port *mlxsw_sp_vport) mlxsw_sp_vport_fid_set(mlxsw_sp_vport, NULL); if (--f->ref_count == 0) - mlxsw_sp_br_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); + mlxsw_sp_vfid_destroy(mlxsw_sp_vport->mlxsw_sp, f); } static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, @@ -3528,7 +3529,7 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, if (f && !WARN_ON(!f->leave)) f->leave(mlxsw_sp_vport); - err = mlxsw_sp_vport_br_vfid_join(mlxsw_sp_vport, br_dev); + err = mlxsw_sp_vport_vfid_join(mlxsw_sp_vport, br_dev); if (err) { netdev_err(dev, "Failed to join vFID\n"); return err; @@ -3548,7 +3549,7 @@ static int mlxsw_sp_vport_bridge_join(struct mlxsw_sp_port *mlxsw_sp_vport, return 0; err_port_vid_learning_set: - mlxsw_sp_vport_br_vfid_leave(mlxsw_sp_vport); + mlxsw_sp_vport_vfid_leave(mlxsw_sp_vport); return err; } @@ -3558,7 +3559,7 @@ static void mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport) mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false); - mlxsw_sp_vport_br_vfid_leave(mlxsw_sp_vport); + mlxsw_sp_vport_vfid_leave(mlxsw_sp_vport); mlxsw_sp_vport->learning = 0; mlxsw_sp_vport->learning_sync = 0; @@ -3574,7 +3575,7 @@ mlxsw_sp_port_master_bridge_check(const struct mlxsw_sp_port *mlxsw_sp_port, list_for_each_entry(mlxsw_sp_vport, &mlxsw_sp_port->vports_list, vport.list) { - struct net_device *dev = mlxsw_sp_vport_br_get(mlxsw_sp_vport); + struct net_device *dev = mlxsw_sp_vport_dev_get(mlxsw_sp_vport); if (dev && dev == br_dev) return false; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index b6eeab7..b15b47b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -218,7 +218,7 @@ struct mlxsw_sp { struct { struct list_head list; DECLARE_BITMAP(mapped, MLXSW_SP_VFID_MAX); - } br_vfids; + } vfids; struct { struct list_head list; DECLARE_BITMAP(mapped, MLXSW_SP_MID_MAX); @@ -349,7 +349,7 @@ mlxsw_sp_vport_fid_get(const struct mlxsw_sp_port *mlxsw_sp_vport) } static inline struct net_device * -mlxsw_sp_vport_br_get(const struct mlxsw_sp_port *mlxsw_sp_vport) +mlxsw_sp_vport_dev_get(const struct mlxsw_sp_port *mlxsw_sp_vport) { struct mlxsw_sp_fid *f = mlxsw_sp_vport_fid_get(mlxsw_sp_vport); -- cgit v0.10.2 From 701b186ebf52f872134824ffd8bd734c4c1e35df Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 4 Jul 2016 08:23:16 +0200 Subject: mlxsw: spectrum: Configure FIDs based on bridge events Before introducing support for L3 interfaces on top of the VLAN-aware bridge we need to add some missing infrastructure. Such an interface can either be the bridge device itself or a VLAN device on top of it. In the first case the router interface (RIF) is associated with FID 1, which is created whenever the first port netdev joins the bridge. We currently assume the default PVID is 1 and that it's already created, as it seems reasonable. This can be extended in the future. However, in the second case it's entirely possible we've yet to create a matching FID. This can happen if the VLAN device was configured before making any bridge port member in the VLAN. Prevent such ordering problems by using the VLAN device's CHANGEUPPER event to configure the FID. Make the VLAN device hold a reference to the FID and prevent it from being destroyed even if none of the port netdevs is using it. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index e987a8a..e49f80ba 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2887,6 +2887,17 @@ int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) return mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_port, fid); } +static void mlxsw_sp_master_bridge_gone_sync(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_sp_fid *f, *tmp; + + list_for_each_entry_safe(f, tmp, &mlxsw_sp->fids, list) + if (--f->ref_count == 0) + mlxsw_sp_fid_destroy(mlxsw_sp, f); + else + WARN_ON_ONCE(1); +} + static bool mlxsw_sp_master_bridge_check(struct mlxsw_sp *mlxsw_sp, struct net_device *br_dev) { @@ -2903,8 +2914,15 @@ static void mlxsw_sp_master_bridge_inc(struct mlxsw_sp *mlxsw_sp, static void mlxsw_sp_master_bridge_dec(struct mlxsw_sp *mlxsw_sp) { - if (--mlxsw_sp->master_bridge.ref_count == 0) + if (--mlxsw_sp->master_bridge.ref_count == 0) { mlxsw_sp->master_bridge.dev = NULL; + /* It's possible upper VLAN devices are still holding + * references to underlying FIDs. Drop the reference + * and release the resources if it was the last one. + * If it wasn't, then something bad happened. + */ + mlxsw_sp_master_bridge_gone_sync(mlxsw_sp); + } } static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port, @@ -3373,18 +3391,68 @@ static int mlxsw_sp_netdevice_lag_event(struct net_device *lag_dev, return 0; } -static struct mlxsw_sp_fid * -mlxsw_sp_vfid_find(const struct mlxsw_sp *mlxsw_sp, - const struct net_device *br_dev) +static int mlxsw_sp_master_bridge_vlan_link(struct mlxsw_sp *mlxsw_sp, + struct net_device *vlan_dev) { + u16 fid = vlan_dev_vlan_id(vlan_dev); struct mlxsw_sp_fid *f; - list_for_each_entry(f, &mlxsw_sp->vfids.list, list) { - if (f->dev == br_dev) - return f; + f = mlxsw_sp_fid_find(mlxsw_sp, fid); + if (!f) { + f = mlxsw_sp_fid_create(mlxsw_sp, fid); + if (IS_ERR(f)) + return PTR_ERR(f); } - return NULL; + f->ref_count++; + + return 0; +} + +static void mlxsw_sp_master_bridge_vlan_unlink(struct mlxsw_sp *mlxsw_sp, + struct net_device *vlan_dev) +{ + u16 fid = vlan_dev_vlan_id(vlan_dev); + struct mlxsw_sp_fid *f; + + f = mlxsw_sp_fid_find(mlxsw_sp, fid); + if (f && --f->ref_count == 0) + mlxsw_sp_fid_destroy(mlxsw_sp, f); +} + +static int mlxsw_sp_netdevice_bridge_event(struct net_device *br_dev, + unsigned long event, void *ptr) +{ + struct netdev_notifier_changeupper_info *info; + struct net_device *upper_dev; + struct mlxsw_sp *mlxsw_sp; + int err; + + mlxsw_sp = mlxsw_sp_lower_get(br_dev); + if (!mlxsw_sp) + return 0; + if (br_dev != mlxsw_sp->master_bridge.dev) + return 0; + + info = ptr; + + switch (event) { + case NETDEV_CHANGEUPPER: + upper_dev = info->upper_dev; + if (!is_vlan_dev(upper_dev)) + break; + if (info->linking) { + err = mlxsw_sp_master_bridge_vlan_link(mlxsw_sp, + upper_dev); + if (err) + return err; + } else { + mlxsw_sp_master_bridge_vlan_unlink(mlxsw_sp, upper_dev); + } + break; + } + + return 0; } static u16 mlxsw_sp_avail_vfid_get(const struct mlxsw_sp *mlxsw_sp) @@ -3675,6 +3743,8 @@ static int mlxsw_sp_netdevice_event(struct notifier_block *unused, err = mlxsw_sp_netdevice_port_event(dev, event, ptr); else if (netif_is_lag_master(dev)) err = mlxsw_sp_netdevice_lag_event(dev, event, ptr); + else if (netif_is_bridge_master(dev)) + err = mlxsw_sp_netdevice_bridge_event(dev, event, ptr); else if (is_vlan_dev(dev)) err = mlxsw_sp_netdevice_vlan_event(dev, event, ptr); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index b15b47b..17c5d3b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -387,6 +387,31 @@ mlxsw_sp_port_vport_find_by_fid(const struct mlxsw_sp_port *mlxsw_sp_port, return NULL; } +static inline struct mlxsw_sp_fid *mlxsw_sp_fid_find(struct mlxsw_sp *mlxsw_sp, + u16 fid) +{ + struct mlxsw_sp_fid *f; + + list_for_each_entry(f, &mlxsw_sp->fids, list) + if (f->fid == fid) + return f; + + return NULL; +} + +static inline struct mlxsw_sp_fid * +mlxsw_sp_vfid_find(const struct mlxsw_sp *mlxsw_sp, + const struct net_device *br_dev) +{ + struct mlxsw_sp_fid *f; + + list_for_each_entry(f, &mlxsw_sp->vfids.list, list) + if (f->dev == br_dev) + return f; + + return NULL; +} + static inline struct mlxsw_sp_rif * mlxsw_sp_rif_find_by_dev(const struct mlxsw_sp *mlxsw_sp, const struct net_device *dev) @@ -459,6 +484,8 @@ int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid); int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid); int mlxsw_sp_rif_fdb_op(struct mlxsw_sp *mlxsw_sp, const char *mac, u16 fid, bool adding); +struct mlxsw_sp_fid *mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, u16 fid); +void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *f); int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index, bool dwrr, u8 dwrr_weight); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index 941acd7..e446640 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -374,18 +374,6 @@ static int mlxsw_sp_port_attr_set(struct net_device *dev, return err; } -static struct mlxsw_sp_fid *mlxsw_sp_fid_find(struct mlxsw_sp *mlxsw_sp, - u16 fid) -{ - struct mlxsw_sp_fid *f; - - list_for_each_entry(f, &mlxsw_sp->fids, list) - if (f->fid == fid) - return f; - - return NULL; -} - static int mlxsw_sp_fid_op(struct mlxsw_sp *mlxsw_sp, u16 fid, bool create) { char sfmr_pl[MLXSW_REG_SFMR_LEN]; @@ -416,8 +404,7 @@ static struct mlxsw_sp_fid *mlxsw_sp_fid_alloc(u16 fid) return f; } -static struct mlxsw_sp_fid *mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, - u16 fid) +struct mlxsw_sp_fid *mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, u16 fid) { struct mlxsw_sp_fid *f; int err; @@ -452,8 +439,7 @@ err_fid_map: return ERR_PTR(err); } -static void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_fid *f) +void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *f) { u16 fid = f->fid; -- cgit v0.10.2 From 99f44bb3527b25198387bd56ec1bfc88a4421050 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 4 Jul 2016 08:23:17 +0200 Subject: mlxsw: spectrum: Enable L3 interfaces on top of bridge devices As with the previously introduced L3 interfaces, listen to 'inetaddr' notifications sent for bridges devices configured on top of the port netdevs and create / destroy router interfaces (RIFs) accordingly. This also includes VLAN devices configured on top of the VLAN-aware bridge. The RIFs will be destroyed either when the last IP address is removed or when the underlying FID is is destroyed. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index e49f80ba..7b2b741b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2706,10 +2706,135 @@ static int mlxsw_sp_inetaddr_lag_event(struct net_device *lag_dev, return __mlxsw_sp_inetaddr_lag_event(lag_dev, lag_dev, event, 1); } +static struct mlxsw_sp_fid *mlxsw_sp_bridge_fid_get(struct mlxsw_sp *mlxsw_sp, + struct net_device *l3_dev) +{ + u16 fid; + + if (is_vlan_dev(l3_dev)) + fid = vlan_dev_vlan_id(l3_dev); + else if (mlxsw_sp->master_bridge.dev == l3_dev) + fid = 1; + else + return mlxsw_sp_vfid_find(mlxsw_sp, l3_dev); + + return mlxsw_sp_fid_find(mlxsw_sp, fid); +} + +static enum mlxsw_reg_ritr_if_type mlxsw_sp_rif_type_get(u16 fid) +{ + if (mlxsw_sp_fid_is_vfid(fid)) + return MLXSW_REG_RITR_FID_IF; + else + return MLXSW_REG_RITR_VLAN_IF; +} + +static int mlxsw_sp_rif_bridge_op(struct mlxsw_sp *mlxsw_sp, + struct net_device *l3_dev, + u16 fid, u16 rif, + bool create) +{ + enum mlxsw_reg_ritr_if_type rif_type; + char ritr_pl[MLXSW_REG_RITR_LEN]; + + rif_type = mlxsw_sp_rif_type_get(fid); + mlxsw_reg_ritr_pack(ritr_pl, create, rif_type, rif, l3_dev->mtu, + l3_dev->dev_addr); + mlxsw_reg_ritr_fid_set(ritr_pl, rif_type, fid); + + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl); +} + +static int mlxsw_sp_rif_bridge_create(struct mlxsw_sp *mlxsw_sp, + struct net_device *l3_dev, + struct mlxsw_sp_fid *f) +{ + struct mlxsw_sp_rif *r; + u16 rif; + int err; + + rif = mlxsw_sp_avail_rif_get(mlxsw_sp); + if (rif == MLXSW_SP_RIF_MAX) + return -ERANGE; + + err = mlxsw_sp_rif_bridge_op(mlxsw_sp, l3_dev, f->fid, rif, true); + if (err) + return err; + + err = mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, f->fid, true); + if (err) + goto err_rif_fdb_op; + + r = mlxsw_sp_rif_alloc(rif, l3_dev, f); + if (!r) { + err = -ENOMEM; + goto err_rif_alloc; + } + + f->r = r; + mlxsw_sp->rifs[rif] = r; + + netdev_dbg(l3_dev, "RIF=%d created\n", rif); + + return 0; + +err_rif_alloc: + mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, f->fid, false); +err_rif_fdb_op: + mlxsw_sp_rif_bridge_op(mlxsw_sp, l3_dev, f->fid, rif, false); + return err; +} + +void mlxsw_sp_rif_bridge_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_rif *r) +{ + struct net_device *l3_dev = r->dev; + struct mlxsw_sp_fid *f = r->f; + u16 rif = r->rif; + + mlxsw_sp->rifs[rif] = NULL; + f->r = NULL; + + kfree(r); + + mlxsw_sp_rif_fdb_op(mlxsw_sp, l3_dev->dev_addr, f->fid, false); + + mlxsw_sp_rif_bridge_op(mlxsw_sp, l3_dev, f->fid, rif, false); + + netdev_dbg(l3_dev, "RIF=%d destroyed\n", rif); +} + +static int mlxsw_sp_inetaddr_bridge_event(struct net_device *l3_dev, + struct net_device *br_dev, + unsigned long event) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(l3_dev); + struct mlxsw_sp_fid *f; + + /* FID can either be an actual FID if the L3 device is the + * VLAN-aware bridge or a VLAN device on top. Otherwise, the + * L3 device is a VLAN-unaware bridge and we get a vFID. + */ + f = mlxsw_sp_bridge_fid_get(mlxsw_sp, l3_dev); + if (WARN_ON(!f)) + return -EINVAL; + + switch (event) { + case NETDEV_UP: + return mlxsw_sp_rif_bridge_create(mlxsw_sp, l3_dev, f); + case NETDEV_DOWN: + mlxsw_sp_rif_bridge_destroy(mlxsw_sp, f->r); + break; + } + + return 0; +} + static int mlxsw_sp_inetaddr_vlan_event(struct net_device *vlan_dev, unsigned long event) { struct net_device *real_dev = vlan_dev_real_dev(vlan_dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_lower_get(vlan_dev); u16 vid = vlan_dev_vlan_id(vlan_dev); if (mlxsw_sp_port_dev_check(real_dev)) @@ -2718,6 +2843,10 @@ static int mlxsw_sp_inetaddr_vlan_event(struct net_device *vlan_dev, else if (netif_is_lag_master(real_dev)) return __mlxsw_sp_inetaddr_lag_event(vlan_dev, real_dev, event, vid); + else if (netif_is_bridge_master(real_dev) && + mlxsw_sp->master_bridge.dev == real_dev) + return mlxsw_sp_inetaddr_bridge_event(vlan_dev, real_dev, + event); return 0; } @@ -2743,6 +2872,8 @@ static int mlxsw_sp_inetaddr_event(struct notifier_block *unused, err = mlxsw_sp_inetaddr_port_event(dev, event); else if (netif_is_lag_master(dev)) err = mlxsw_sp_inetaddr_lag_event(dev, event); + else if (netif_is_bridge_master(dev)) + err = mlxsw_sp_inetaddr_bridge_event(dev, dev, event); else if (is_vlan_dev(dev)) err = mlxsw_sp_inetaddr_vlan_event(dev, event); @@ -3416,6 +3547,8 @@ static void mlxsw_sp_master_bridge_vlan_unlink(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *f; f = mlxsw_sp_fid_find(mlxsw_sp, fid); + if (f && f->r) + mlxsw_sp_rif_bridge_destroy(mlxsw_sp, f->r); if (f && --f->ref_count == 0) mlxsw_sp_fid_destroy(mlxsw_sp, f); } @@ -3514,13 +3647,17 @@ static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *f) { u16 vfid = mlxsw_sp_fid_to_vfid(f->fid); + u16 fid = f->fid; clear_bit(vfid, mlxsw_sp->vfids.mapped); list_del(&f->list); - mlxsw_sp_vfid_op(mlxsw_sp, f->fid, false); + if (f->r) + mlxsw_sp_rif_bridge_destroy(mlxsw_sp, f->r); kfree(f); + + mlxsw_sp_vfid_op(mlxsw_sp, fid, false); } static int mlxsw_sp_vport_fid_map(struct mlxsw_sp_port *mlxsw_sp_vport, u16 fid, diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 17c5d3b..958e821 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -486,6 +486,8 @@ int mlxsw_sp_rif_fdb_op(struct mlxsw_sp *mlxsw_sp, const char *mac, u16 fid, bool adding); struct mlxsw_sp_fid *mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, u16 fid); void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *f); +void mlxsw_sp_rif_bridge_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_rif *r); int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index, bool dwrr, u8 dwrr_weight); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c index e446640..a1ad5e6 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -445,6 +445,9 @@ void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *f) list_del(&f->list); + if (f->r) + mlxsw_sp_rif_bridge_destroy(mlxsw_sp, f->r); + kfree(f); mlxsw_sp_fid_op(mlxsw_sp, fid, false); -- cgit v0.10.2 From d9329bc2226f1c9445bfb61b49f6a1801597f3ee Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Mon, 4 Jul 2016 10:13:21 -0300 Subject: net: lpc_eth: Remove unused 'pldat' variable Since commit f786f3564c4f02d5026 ("net: ethernet: lpc_eth: use phydev from struct net_device") the 'pldat' variable became unused, so just remove it. Reported-by: Olof's autobuilder Signed-off-by: Fabio Estevam Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c index 01b50ff..4d4ecba 100644 --- a/drivers/net/ethernet/nxp/lpc_eth.c +++ b/drivers/net/ethernet/nxp/lpc_eth.c @@ -1183,7 +1183,6 @@ static void lpc_eth_set_multicast_list(struct net_device *ndev) static int lpc_eth_ioctl(struct net_device *ndev, struct ifreq *req, int cmd) { - struct netdata_local *pldat = netdev_priv(ndev); struct phy_device *phydev = ndev->phydev; if (!netif_running(ndev)) -- cgit v0.10.2 From 019d0c99364a818eb08f52d7ee4b75ea6df07d09 Mon Sep 17 00:00:00 2001 From: John Crispin Date: Mon, 4 Jul 2016 15:37:10 +0200 Subject: net-next: mediatek: remove superfluous free_irq() call Commit 8067302973a1 ("net-next: mediatek: add support for IRQ grouping") adds handling for irq 1 and 2 to the uninit function but did not remove irq 0 which is not used since irq grouping was introduced. Fix this by removing the superfluous call to free_irq(). Reported-by: Arnd Bergmann Signed-off-by: John Crispin Acked-by: Arnd Bergmann Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index fbab9b2..682797e 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -1492,7 +1492,6 @@ static void mtk_uninit(struct net_device *dev) phy_disconnect(mac->phy_dev); mtk_mdio_cleanup(eth); mtk_irq_disable(eth, ~0); - free_irq(eth->irq[0], dev); free_irq(eth->irq[1], dev); free_irq(eth->irq[2], dev); } -- cgit v0.10.2 From c5bb17302e734967822be559cf661704b707b4ed Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 4 Jul 2016 17:23:05 +0300 Subject: net/mlx5: Refactor mlx5_add_flow_rule Reduce the set of arguments passed to mlx5_add_flow_rule by introducing flow_spec structure. Signed-off-by: Maor Gottlieb Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index b48ad85..dad63f0 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -1528,21 +1528,18 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, { struct mlx5_flow_table *ft = ft_prio->flow_table; struct mlx5_ib_flow_handler *handler; + struct mlx5_flow_spec *spec; void *ib_flow = flow_attr + 1; - u8 match_criteria_enable = 0; unsigned int spec_index; - u32 *match_c; - u32 *match_v; u32 action; int err = 0; if (!is_valid_attr(flow_attr)) return ERR_PTR(-EINVAL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + spec = mlx5_vzalloc(sizeof(*spec)); handler = kzalloc(sizeof(*handler), GFP_KERNEL); - if (!handler || !match_c || !match_v) { + if (!handler || !spec) { err = -ENOMEM; goto free; } @@ -1550,7 +1547,8 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, INIT_LIST_HEAD(&handler->list); for (spec_index = 0; spec_index < flow_attr->num_of_specs; spec_index++) { - err = parse_flow_attr(match_c, match_v, ib_flow); + err = parse_flow_attr(spec->match_criteria, + spec->match_value, ib_flow); if (err < 0) goto free; @@ -1558,11 +1556,11 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, } /* Outer header support only */ - match_criteria_enable = (!outer_header_zero(match_c)) << 0; + spec->match_criteria_enable = (!outer_header_zero(spec->match_criteria)) + << 0; action = dst ? MLX5_FLOW_CONTEXT_ACTION_FWD_DEST : MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO; - handler->rule = mlx5_add_flow_rule(ft, match_criteria_enable, - match_c, match_v, + handler->rule = mlx5_add_flow_rule(ft, spec, action, MLX5_FS_DEFAULT_FLOW_TAG, dst); @@ -1578,8 +1576,7 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, free: if (err) kfree(handler); - kfree(match_c); - kfree(match_v); + kvfree(spec); return err ? ERR_PTR(err) : handler; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c index 10f18d4..a8cb387 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c @@ -175,15 +175,12 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, { struct arfs_table *arfs_t = &priv->fs.arfs.arfs_tables[type]; struct mlx5_flow_destination dest; - u8 match_criteria_enable = 0; struct mlx5e_tir *tir = priv->indir_tir; - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); err = -ENOMEM; goto out; @@ -208,8 +205,7 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, goto out; } - arfs_t->default_rule = mlx5_add_flow_rule(arfs_t->ft.t, match_criteria_enable, - match_criteria, match_value, + arfs_t->default_rule = mlx5_add_flow_rule(arfs_t->ft.t, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); @@ -220,8 +216,7 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, __func__, type); } out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err; } @@ -475,23 +470,20 @@ static struct mlx5_flow_rule *arfs_add_rule(struct mlx5e_priv *priv, struct mlx5_flow_rule *rule = NULL; struct mlx5_flow_destination dest; struct arfs_table *arfs_table; - u8 match_criteria_enable = 0; + struct mlx5_flow_spec *spec; struct mlx5_flow_table *ft; - u32 *match_criteria; - u32 *match_value; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); err = -ENOMEM; goto out; } - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ethertype); - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ethertype, ntohs(tuple->etype)); arfs_table = arfs_get_table(arfs, tuple->ip_proto, tuple->etype); if (!arfs_table) { @@ -501,59 +493,58 @@ static struct mlx5_flow_rule *arfs_add_rule(struct mlx5e_priv *priv, ft = arfs_table->ft.t; if (tuple->ip_proto == IPPROTO_TCP) { - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.tcp_dport); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.tcp_sport); - MLX5_SET(fte_match_param, match_value, outer_headers.tcp_dport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.tcp_dport, ntohs(tuple->dst_port)); - MLX5_SET(fte_match_param, match_value, outer_headers.tcp_sport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.tcp_sport, ntohs(tuple->src_port)); } else { - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.udp_dport); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.udp_sport); - MLX5_SET(fte_match_param, match_value, outer_headers.udp_dport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.udp_dport, ntohs(tuple->dst_port)); - MLX5_SET(fte_match_param, match_value, outer_headers.udp_sport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.udp_sport, ntohs(tuple->src_port)); } if (tuple->etype == htons(ETH_P_IP)) { - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4), &tuple->src_ipv4, 4); - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4), &tuple->dst_ipv4, 4); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4); } else { - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), &tuple->src_ipv6, 16); - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), &tuple->dst_ipv6, 16); - memset(MLX5_ADDR_OF(fte_match_param, match_criteria, + memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), 0xff, 16); - memset(MLX5_ADDR_OF(fte_match_param, match_criteria, + memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), 0xff, 16); } dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; dest.tir_num = priv->direct_tir[arfs_rule->rxq].tirn; - rule = mlx5_add_flow_rule(ft, match_criteria_enable, match_criteria, - match_value, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + rule = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); if (IS_ERR(rule)) { @@ -563,8 +554,7 @@ static struct mlx5_flow_rule *arfs_add_rule(struct mlx5e_priv *priv, } out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err ? ERR_PTR(err) : rule; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c index 606e69b..2e1e863 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c @@ -156,19 +156,18 @@ enum mlx5e_vlan_rule_type { static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv, enum mlx5e_vlan_rule_type rule_type, - u16 vid, u32 *mc, u32 *mv) + u16 vid, struct mlx5_flow_spec *spec) { struct mlx5_flow_table *ft = priv->fs.vlan.ft.t; struct mlx5_flow_destination dest; - u8 match_criteria_enable = 0; struct mlx5_flow_rule **rule_p; int err = 0; dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; dest.ft = priv->fs.l2.ft.t; - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.vlan_tag); switch (rule_type) { case MLX5E_VLAN_RULE_TYPE_UNTAGGED: @@ -176,17 +175,19 @@ static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv, break; case MLX5E_VLAN_RULE_TYPE_ANY_VID: rule_p = &priv->fs.vlan.any_vlan_rule; - MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.vlan_tag, 1); break; default: /* MLX5E_VLAN_RULE_TYPE_MATCH_VID */ rule_p = &priv->fs.vlan.active_vlans_rule[vid]; - MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1); - MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.first_vid); - MLX5_SET(fte_match_param, mv, outer_headers.first_vid, vid); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.vlan_tag, 1); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, + outer_headers.first_vid); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.first_vid, + vid); break; } - *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + *rule_p = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); @@ -203,27 +204,21 @@ static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv, static int mlx5e_add_vlan_rule(struct mlx5e_priv *priv, enum mlx5e_vlan_rule_type rule_type, u16 vid) { - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); - err = -ENOMEM; - goto add_vlan_rule_out; + return -ENOMEM; } if (rule_type == MLX5E_VLAN_RULE_TYPE_MATCH_VID) mlx5e_vport_context_update_vlans(priv); - err = __mlx5e_add_vlan_rule(priv, rule_type, vid, match_criteria, - match_value); + err = __mlx5e_add_vlan_rule(priv, rule_type, vid, spec); -add_vlan_rule_out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err; } @@ -598,32 +593,27 @@ static struct mlx5_flow_rule *mlx5e_generate_ttc_rule(struct mlx5e_priv *priv, u8 proto) { struct mlx5_flow_rule *rule; - u8 match_criteria_enable = 0; - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); - err = -ENOMEM; - goto out; + return ERR_PTR(-ENOMEM); } if (proto) { - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.ip_protocol); - MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol, proto); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ip_protocol); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ip_protocol, proto); } if (etype) { - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.ethertype); - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, etype); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ethertype); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ethertype, etype); } - rule = mlx5_add_flow_rule(ft, match_criteria_enable, - match_criteria, match_value, + rule = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, dest); @@ -631,9 +621,8 @@ static struct mlx5_flow_rule *mlx5e_generate_ttc_rule(struct mlx5e_priv *priv, err = PTR_ERR(rule); netdev_err(priv->netdev, "%s: add rule failed\n", __func__); } -out: - kvfree(match_criteria); - kvfree(match_value); + + kvfree(spec); return err ? ERR_PTR(err) : rule; } @@ -792,24 +781,20 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, { struct mlx5_flow_table *ft = priv->fs.l2.ft.t; struct mlx5_flow_destination dest; - u8 match_criteria_enable = 0; - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; u8 *mc_dmac; u8 *mv_dmac; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); - err = -ENOMEM; - goto add_l2_rule_out; + return -ENOMEM; } - mc_dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, + mc_dmac = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.dmac_47_16); - mv_dmac = MLX5_ADDR_OF(fte_match_param, match_value, + mv_dmac = MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dmac_47_16); dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; @@ -817,13 +802,13 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, switch (type) { case MLX5E_FULLMATCH: - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; eth_broadcast_addr(mc_dmac); ether_addr_copy(mv_dmac, ai->addr); break; case MLX5E_ALLMULTI: - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; mc_dmac[0] = 0x01; mv_dmac[0] = 0x01; break; @@ -832,8 +817,7 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, break; } - ai->rule = mlx5_add_flow_rule(ft, match_criteria_enable, match_criteria, - match_value, + ai->rule = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); if (IS_ERR(ai->rule)) { @@ -843,9 +827,7 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, ai->rule = NULL; } -add_l2_rule_out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 704c3d3..3261e8b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -50,7 +50,7 @@ struct mlx5e_tc_flow { #define MLX5E_TC_TABLE_NUM_GROUPS 4 static struct mlx5_flow_rule *mlx5e_tc_add_flow(struct mlx5e_priv *priv, - u32 *match_c, u32 *match_v, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag) { struct mlx5_core_dev *dev = priv->mdev; @@ -88,8 +88,8 @@ static struct mlx5_flow_rule *mlx5e_tc_add_flow(struct mlx5e_priv *priv, table_created = true; } - rule = mlx5_add_flow_rule(priv->fs.tc.t, MLX5_MATCH_OUTER_HEADERS, - match_c, match_v, + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + rule = mlx5_add_flow_rule(priv->fs.tc.t, spec, action, flow_tag, &dest); @@ -126,12 +126,13 @@ static void mlx5e_tc_del_flow(struct mlx5e_priv *priv, } } -static int parse_cls_flower(struct mlx5e_priv *priv, - u32 *match_c, u32 *match_v, +static int parse_cls_flower(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec, struct tc_cls_flower_offload *f) { - void *headers_c = MLX5_ADDR_OF(fte_match_param, match_c, outer_headers); - void *headers_v = MLX5_ADDR_OF(fte_match_param, match_v, outer_headers); + void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + outer_headers); + void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers); u16 addr_type = 0; u8 ip_proto = 0; @@ -342,12 +343,11 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, struct tc_cls_flower_offload *f) { struct mlx5e_tc_table *tc = &priv->fs.tc; - u32 *match_c; - u32 *match_v; int err = 0; u32 flow_tag; u32 action; struct mlx5e_tc_flow *flow; + struct mlx5_flow_spec *spec; struct mlx5_flow_rule *old = NULL; flow = rhashtable_lookup_fast(&tc->ht, &f->cookie, @@ -357,16 +357,15 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, else flow = kzalloc(sizeof(*flow), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_c || !match_v || !flow) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec || !flow) { err = -ENOMEM; goto err_free; } flow->cookie = f->cookie; - err = parse_cls_flower(priv, match_c, match_v, f); + err = parse_cls_flower(priv, spec, f); if (err < 0) goto err_free; @@ -379,8 +378,7 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, if (err) goto err_free; - flow->rule = mlx5e_tc_add_flow(priv, match_c, match_v, action, - flow_tag); + flow->rule = mlx5e_tc_add_flow(priv, spec, action, flow_tag); if (IS_ERR(flow->rule)) { err = PTR_ERR(flow->rule); goto err_hash_del; @@ -398,8 +396,7 @@ err_free: if (!old) kfree(flow); out: - kfree(match_c); - kfree(match_v); + kvfree(spec); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index f0a9735..f6d6677 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -329,25 +329,23 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, MLX5_MATCH_OUTER_HEADERS); struct mlx5_flow_rule *flow_rule = NULL; struct mlx5_flow_destination dest; + struct mlx5_flow_spec *spec; void *mv_misc = NULL; void *mc_misc = NULL; u8 *dmac_v = NULL; u8 *dmac_c = NULL; - u32 *match_v; - u32 *match_c; if (rx_rule) match_header |= MLX5_MATCH_MISC_PARAMETERS; - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { pr_warn("FDB: Failed to alloc match parameters\n"); - goto out; + return NULL; } - - dmac_v = MLX5_ADDR_OF(fte_match_param, match_v, + dmac_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dmac_47_16); - dmac_c = MLX5_ADDR_OF(fte_match_param, match_c, + dmac_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.dmac_47_16); if (match_header & MLX5_MATCH_OUTER_HEADERS) { @@ -356,8 +354,10 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, } if (match_header & MLX5_MATCH_MISC_PARAMETERS) { - mv_misc = MLX5_ADDR_OF(fte_match_param, match_v, misc_parameters); - mc_misc = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); + mv_misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, + misc_parameters); + mc_misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + misc_parameters); MLX5_SET(fte_match_set_misc, mv_misc, source_port, UPLINK_VPORT); MLX5_SET_TO_ONES(fte_match_set_misc, mc_misc, source_port); } @@ -368,11 +368,9 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, esw_debug(esw->dev, "\tFDB add rule dmac_v(%pM) dmac_c(%pM) -> vport(%d)\n", dmac_v, dmac_c, vport); + spec->match_criteria_enable = match_header; flow_rule = - mlx5_add_flow_rule(esw->fdb_table.fdb, - match_header, - match_c, - match_v, + mlx5_add_flow_rule(esw->fdb_table.fdb, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); if (IS_ERR(flow_rule)) { @@ -381,9 +379,8 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, dmac_v, dmac_c, vport, PTR_ERR(flow_rule)); flow_rule = NULL; } -out: - kfree(match_v); - kfree(match_c); + + kvfree(spec); return flow_rule; } @@ -1293,9 +1290,8 @@ static void esw_vport_disable_ingress_acl(struct mlx5_eswitch *esw, static int esw_vport_ingress_config(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { + struct mlx5_flow_spec *spec; u8 smac[ETH_ALEN]; - u32 *match_v; - u32 *match_c; int err = 0; u8 *smac_v; @@ -1329,9 +1325,8 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, "vport[%d] configure ingress rules, vlan(%d) qos(%d)\n", vport->vport, vport->vlan, vport->qos); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { err = -ENOMEM; esw_warn(esw->dev, "vport[%d] configure ingress rules failed, err(%d)\n", vport->vport, err); @@ -1339,22 +1334,20 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, } if (vport->vlan || vport->qos) - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.vlan_tag); if (vport->spoofchk) { - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.smac_47_16); - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.smac_15_0); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.smac_47_16); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.smac_15_0); smac_v = MLX5_ADDR_OF(fte_match_param, - match_v, + spec->match_value, outer_headers.smac_47_16); ether_addr_copy(smac_v, smac); } + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; vport->ingress.allow_rule = - mlx5_add_flow_rule(vport->ingress.acl, - MLX5_MATCH_OUTER_HEADERS, - match_c, - match_v, + mlx5_add_flow_rule(vport->ingress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_ALLOW, 0, NULL); if (IS_ERR(vport->ingress.allow_rule)) { @@ -1365,13 +1358,9 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, goto out; } - memset(match_c, 0, MLX5_ST_SZ_BYTES(fte_match_param)); - memset(match_v, 0, MLX5_ST_SZ_BYTES(fte_match_param)); + memset(spec, 0, sizeof(*spec)); vport->ingress.drop_rule = - mlx5_add_flow_rule(vport->ingress.acl, - 0, - match_c, - match_v, + mlx5_add_flow_rule(vport->ingress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_DROP, 0, NULL); if (IS_ERR(vport->ingress.drop_rule)) { @@ -1385,17 +1374,14 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, out: if (err) esw_vport_cleanup_ingress_rules(esw, vport); - - kfree(match_v); - kfree(match_c); + kvfree(spec); return err; } static int esw_vport_egress_config(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { - u32 *match_v; - u32 *match_c; + struct mlx5_flow_spec *spec; int err = 0; esw_vport_cleanup_egress_rules(esw, vport); @@ -1411,9 +1397,8 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, "vport[%d] configure egress rules, vlan(%d) qos(%d)\n", vport->vport, vport->vlan, vport->qos); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { err = -ENOMEM; esw_warn(esw->dev, "vport[%d] configure egress rules failed, err(%d)\n", vport->vport, err); @@ -1421,16 +1406,14 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, } /* Allowed vlan rule */ - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.vlan_tag); - MLX5_SET_TO_ONES(fte_match_param, match_v, outer_headers.vlan_tag); - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.first_vid); - MLX5_SET(fte_match_param, match_v, outer_headers.first_vid, vport->vlan); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_value, outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.first_vid); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.first_vid, vport->vlan); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; vport->egress.allowed_vlan = - mlx5_add_flow_rule(vport->egress.acl, - MLX5_MATCH_OUTER_HEADERS, - match_c, - match_v, + mlx5_add_flow_rule(vport->egress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_ALLOW, 0, NULL); if (IS_ERR(vport->egress.allowed_vlan)) { @@ -1442,13 +1425,9 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, } /* Drop others rule (star rule) */ - memset(match_c, 0, MLX5_ST_SZ_BYTES(fte_match_param)); - memset(match_v, 0, MLX5_ST_SZ_BYTES(fte_match_param)); + memset(spec, 0, sizeof(*spec)); vport->egress.drop_rule = - mlx5_add_flow_rule(vport->egress.acl, - 0, - match_c, - match_v, + mlx5_add_flow_rule(vport->egress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_DROP, 0, NULL); if (IS_ERR(vport->egress.drop_rule)) { @@ -1458,8 +1437,7 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, vport->egress.drop_rule = NULL; } out: - kfree(match_v); - kfree(match_c); + kvfree(spec); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index ed8ad98..1842dfb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -43,37 +43,35 @@ mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn { struct mlx5_flow_destination dest; struct mlx5_flow_rule *flow_rule; - int match_header = MLX5_MATCH_MISC_PARAMETERS; - u32 *match_v, *match_c; + struct mlx5_flow_spec *spec; void *misc; - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { esw_warn(esw->dev, "FDB: Failed to alloc match parameters\n"); flow_rule = ERR_PTR(-ENOMEM); goto out; } - misc = MLX5_ADDR_OF(fte_match_param, match_v, misc_parameters); + misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters); MLX5_SET(fte_match_set_misc, misc, source_sqn, sqn); MLX5_SET(fte_match_set_misc, misc, source_port, 0x0); /* source vport is 0 */ - misc = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); + misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters); MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_sqn); MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS; dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; dest.vport_num = vport; - flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, match_header, match_c, - match_v, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, spec, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); if (IS_ERR(flow_rule)) esw_warn(esw->dev, "FDB: Failed to add send to vport rule err %ld\n", PTR_ERR(flow_rule)); out: - kfree(match_v); - kfree(match_c); + kvfree(spec); return flow_rule; } @@ -138,12 +136,11 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) { struct mlx5_flow_destination dest; struct mlx5_flow_rule *flow_rule = NULL; - u32 *match_v, *match_c; + struct mlx5_flow_spec *spec; int err = 0; - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { esw_warn(esw->dev, "FDB: Failed to alloc match parameters\n"); err = -ENOMEM; goto out; @@ -152,8 +149,9 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; dest.vport_num = 0; - flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, 0, match_c, match_v, - MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); + flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, spec, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + 0, &dest); if (IS_ERR(flow_rule)) { err = PTR_ERR(flow_rule); esw_warn(esw->dev, "FDB: Failed to add miss flow rule err %d\n", err); @@ -162,8 +160,7 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) esw->fdb_table.offloads.miss_rule = flow_rule; out: - kfree(match_v); - kfree(match_c); + kvfree(spec); return err; } @@ -351,29 +348,28 @@ mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn) { struct mlx5_flow_destination dest; struct mlx5_flow_rule *flow_rule; - int match_header = MLX5_MATCH_MISC_PARAMETERS; - u32 *match_v, *match_c; + struct mlx5_flow_spec *spec; void *misc; - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { esw_warn(esw->dev, "Failed to alloc match parameters\n"); flow_rule = ERR_PTR(-ENOMEM); goto out; } - misc = MLX5_ADDR_OF(fte_match_param, match_v, misc_parameters); + misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters); MLX5_SET(fte_match_set_misc, misc, source_port, vport); - misc = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); + misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters); MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS; dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; dest.tir_num = tirn; - flow_rule = mlx5_add_flow_rule(esw->offloads.ft_offloads, match_header, match_c, - match_v, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + flow_rule = mlx5_add_flow_rule(esw->offloads.ft_offloads, spec, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); if (IS_ERR(flow_rule)) { esw_warn(esw->dev, "fs offloads: Failed to add vport rx rule err %ld\n", PTR_ERR(flow_rule)); @@ -381,8 +377,7 @@ mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn) } out: - kfree(match_v); - kfree(match_c); + kvfree(spec); return flow_rule; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index b040110..7fcdae1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -1160,9 +1160,7 @@ static bool dest_is_valid(struct mlx5_flow_destination *dest, static struct mlx5_flow_rule * _mlx5_add_flow_rule(struct mlx5_flow_table *ft, - u8 match_criteria_enable, - u32 *match_criteria, - u32 *match_value, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag, struct mlx5_flow_destination *dest) @@ -1176,22 +1174,23 @@ _mlx5_add_flow_rule(struct mlx5_flow_table *ft, nested_lock_ref_node(&ft->node, FS_MUTEX_GRANDPARENT); fs_for_each_fg(g, ft) if (compare_match_criteria(g->mask.match_criteria_enable, - match_criteria_enable, + spec->match_criteria_enable, g->mask.match_criteria, - match_criteria)) { - rule = add_rule_fg(g, match_value, + spec->match_criteria)) { + rule = add_rule_fg(g, spec->match_value, action, flow_tag, dest); if (!IS_ERR(rule) || PTR_ERR(rule) != -ENOSPC) goto unlock; } - g = create_autogroup(ft, match_criteria_enable, match_criteria); + g = create_autogroup(ft, spec->match_criteria_enable, + spec->match_criteria); if (IS_ERR(g)) { rule = (void *)g; goto unlock; } - rule = add_rule_fg(g, match_value, + rule = add_rule_fg(g, spec->match_value, action, flow_tag, dest); if (IS_ERR(rule)) { /* Remove assumes refcount > 0 and autogroup creates a group @@ -1215,9 +1214,7 @@ static bool fwd_next_prio_supported(struct mlx5_flow_table *ft) struct mlx5_flow_rule * mlx5_add_flow_rule(struct mlx5_flow_table *ft, - u8 match_criteria_enable, - u32 *match_criteria, - u32 *match_value, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag, struct mlx5_flow_destination *dest) @@ -1248,8 +1245,7 @@ mlx5_add_flow_rule(struct mlx5_flow_table *ft, } } - rule = _mlx5_add_flow_rule(ft, match_criteria_enable, match_criteria, - match_value, action, flow_tag, dest); + rule = _mlx5_add_flow_rule(ft, spec, action, flow_tag, dest); if (sw_action == MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO) { if (!IS_ERR_OR_NULL(rule) && diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index 6ad1119..d22fe7e 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -68,6 +68,12 @@ struct mlx5_flow_group; struct mlx5_flow_rule; struct mlx5_flow_namespace; +struct mlx5_flow_spec { + u8 match_criteria_enable; + u32 match_criteria[MLX5_ST_SZ_DW(fte_match_param)]; + u32 match_value[MLX5_ST_SZ_DW(fte_match_param)]; +}; + struct mlx5_flow_destination { enum mlx5_flow_destination_type type; union { @@ -116,9 +122,7 @@ void mlx5_destroy_flow_group(struct mlx5_flow_group *fg); */ struct mlx5_flow_rule * mlx5_add_flow_rule(struct mlx5_flow_table *ft, - u8 match_criteria_enable, - u32 *match_criteria, - u32 *match_value, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag, struct mlx5_flow_destination *dest); -- cgit v0.10.2 From fba53f7b571925b8a0d59d460ad6de1fda928a3e Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 4 Jul 2016 17:23:06 +0300 Subject: net/mlx5: Introduce mlx5_flow_steering structure Instead of having all steering private name spaces and steering module fields flat in mlx5_core_priv, we wrap them in mlx5_flow_steering for better modularity and API exposure. Signed-off-by: Maor Gottlieb Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 7fcdae1..0f969cb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -1363,12 +1363,13 @@ void mlx5_destroy_flow_group(struct mlx5_flow_group *fg) struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, enum mlx5_flow_namespace_type type) { - struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns; + struct mlx5_flow_steering *steering = dev->priv.steering; + struct mlx5_flow_root_namespace *root_ns; int prio; struct fs_prio *fs_prio; struct mlx5_flow_namespace *ns; - if (!root_ns) + if (!steering) return NULL; switch (type) { @@ -1380,24 +1381,28 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, prio = type; break; case MLX5_FLOW_NAMESPACE_FDB: - if (dev->priv.fdb_root_ns) - return &dev->priv.fdb_root_ns->ns; + if (steering->fdb_root_ns) + return &steering->fdb_root_ns->ns; else return NULL; case MLX5_FLOW_NAMESPACE_ESW_EGRESS: - if (dev->priv.esw_egress_root_ns) - return &dev->priv.esw_egress_root_ns->ns; + if (steering->esw_egress_root_ns) + return &steering->esw_egress_root_ns->ns; else return NULL; case MLX5_FLOW_NAMESPACE_ESW_INGRESS: - if (dev->priv.esw_ingress_root_ns) - return &dev->priv.esw_ingress_root_ns->ns; + if (steering->esw_ingress_root_ns) + return &steering->esw_ingress_root_ns->ns; else return NULL; default: return NULL; } + root_ns = steering->root_ns; + if (!root_ns) + return NULL; + fs_prio = find_prio(&root_ns->ns, prio); if (!fs_prio) return NULL; @@ -1483,13 +1488,13 @@ static bool has_required_caps(struct mlx5_core_dev *dev, struct node_caps *caps) return true; } -static int init_root_tree_recursive(struct mlx5_core_dev *dev, +static int init_root_tree_recursive(struct mlx5_flow_steering *steering, struct init_tree_node *init_node, struct fs_node *fs_parent_node, struct init_tree_node *init_parent_node, int prio) { - int max_ft_level = MLX5_CAP_FLOWTABLE(dev, + int max_ft_level = MLX5_CAP_FLOWTABLE(steering->dev, flow_table_properties_nic_receive. max_ft_level); struct mlx5_flow_namespace *fs_ns; @@ -1500,7 +1505,7 @@ static int init_root_tree_recursive(struct mlx5_core_dev *dev, if (init_node->type == FS_TYPE_PRIO) { if ((init_node->min_ft_level > max_ft_level) || - !has_required_caps(dev, &init_node->caps)) + !has_required_caps(steering->dev, &init_node->caps)) return 0; fs_get_obj(fs_ns, fs_parent_node); @@ -1521,7 +1526,7 @@ static int init_root_tree_recursive(struct mlx5_core_dev *dev, } prio = 0; for (i = 0; i < init_node->ar_size; i++) { - err = init_root_tree_recursive(dev, &init_node->children[i], + err = init_root_tree_recursive(steering, &init_node->children[i], base, init_node, prio); if (err) return err; @@ -1534,7 +1539,7 @@ static int init_root_tree_recursive(struct mlx5_core_dev *dev, return 0; } -static int init_root_tree(struct mlx5_core_dev *dev, +static int init_root_tree(struct mlx5_flow_steering *steering, struct init_tree_node *init_node, struct fs_node *fs_parent_node) { @@ -1544,7 +1549,7 @@ static int init_root_tree(struct mlx5_core_dev *dev, fs_get_obj(fs_ns, fs_parent_node); for (i = 0; i < init_node->ar_size; i++) { - err = init_root_tree_recursive(dev, &init_node->children[i], + err = init_root_tree_recursive(steering, &init_node->children[i], &fs_ns->node, init_node, i); if (err) @@ -1553,7 +1558,7 @@ static int init_root_tree(struct mlx5_core_dev *dev, return 0; } -static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev, +static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_flow_steering *steering, enum fs_flow_table_type table_type) { @@ -1565,7 +1570,7 @@ static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev if (!root_ns) return NULL; - root_ns->dev = dev; + root_ns->dev = steering->dev; root_ns->table_type = table_type; ns = &root_ns->ns; @@ -1620,46 +1625,45 @@ static void set_prio_attrs(struct mlx5_flow_root_namespace *root_ns) #define ANCHOR_PRIO 0 #define ANCHOR_SIZE 1 #define ANCHOR_LEVEL 0 -static int create_anchor_flow_table(struct mlx5_core_dev - *dev) +static int create_anchor_flow_table(struct mlx5_flow_steering *steering) { struct mlx5_flow_namespace *ns = NULL; struct mlx5_flow_table *ft; - ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_ANCHOR); + ns = mlx5_get_flow_namespace(steering->dev, MLX5_FLOW_NAMESPACE_ANCHOR); if (!ns) return -EINVAL; ft = mlx5_create_flow_table(ns, ANCHOR_PRIO, ANCHOR_SIZE, ANCHOR_LEVEL); if (IS_ERR(ft)) { - mlx5_core_err(dev, "Failed to create last anchor flow table"); + mlx5_core_err(steering->dev, "Failed to create last anchor flow table"); return PTR_ERR(ft); } return 0; } -static int init_root_ns(struct mlx5_core_dev *dev) +static int init_root_ns(struct mlx5_flow_steering *steering) { - dev->priv.root_ns = create_root_ns(dev, FS_FT_NIC_RX); - if (IS_ERR_OR_NULL(dev->priv.root_ns)) + steering->root_ns = create_root_ns(steering, FS_FT_NIC_RX); + if (IS_ERR_OR_NULL(steering->root_ns)) goto cleanup; - if (init_root_tree(dev, &root_fs, &dev->priv.root_ns->ns.node)) + if (init_root_tree(steering, &root_fs, &steering->root_ns->ns.node)) goto cleanup; - set_prio_attrs(dev->priv.root_ns); + set_prio_attrs(steering->root_ns); - if (create_anchor_flow_table(dev)) + if (create_anchor_flow_table(steering)) goto cleanup; return 0; cleanup: - mlx5_cleanup_fs(dev); + mlx5_cleanup_fs(steering->dev); return -ENOMEM; } -static void cleanup_single_prio_root_ns(struct mlx5_core_dev *dev, +static void cleanup_single_prio_root_ns(struct mlx5_flow_steering *steering, struct mlx5_flow_root_namespace *root_ns) { struct fs_node *prio; @@ -1672,11 +1676,11 @@ static void cleanup_single_prio_root_ns(struct mlx5_core_dev *dev, struct fs_node, list); if (tree_remove_node(prio)) - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "Flow steering priority wasn't destroyed, refcount > 1\n"); } if (tree_remove_node(&root_ns->ns.node)) - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "Flow steering namespace wasn't destroyed, refcount > 1\n"); root_ns = NULL; } @@ -1690,12 +1694,12 @@ static void destroy_flow_tables(struct fs_prio *prio) mlx5_destroy_flow_table(iter); } -static void cleanup_root_ns(struct mlx5_core_dev *dev) +static void cleanup_root_ns(struct mlx5_flow_steering *steering) { - struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns; + struct mlx5_flow_root_namespace *root_ns = steering->root_ns; struct fs_prio *iter_prio; - if (!MLX5_CAP_GEN(dev, nic_flow_table)) + if (!MLX5_CAP_GEN(steering->dev, nic_flow_table)) return; if (!root_ns) @@ -1720,7 +1724,7 @@ static void cleanup_root_ns(struct mlx5_core_dev *dev) fs_get_obj(obj_iter_prio2, iter_prio2); destroy_flow_tables(obj_iter_prio2); if (tree_remove_node(iter_prio2)) { - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "Priority %d wasn't destroyed, refcount > 1\n", obj_iter_prio2->prio); return; @@ -1737,7 +1741,7 @@ static void cleanup_root_ns(struct mlx5_core_dev *dev) struct fs_node, list); if (tree_remove_node(iter_ns)) { - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "Namespace wasn't destroyed, refcount > 1\n"); return; } @@ -1754,7 +1758,7 @@ static void cleanup_root_ns(struct mlx5_core_dev *dev) fs_get_obj(obj_prio_node, prio_node); if (tree_remove_node(prio_node)) { - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "Priority %d wasn't destroyed, refcount > 1\n", obj_prio_node->prio); return; @@ -1762,70 +1766,75 @@ static void cleanup_root_ns(struct mlx5_core_dev *dev) } if (tree_remove_node(&root_ns->ns.node)) { - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "root namespace wasn't destroyed, refcount > 1\n"); return; } - dev->priv.root_ns = NULL; + steering->root_ns = NULL; } void mlx5_cleanup_fs(struct mlx5_core_dev *dev) { + struct mlx5_flow_steering *steering = dev->priv.steering; + if (MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) return; - cleanup_root_ns(dev); - cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns); - cleanup_single_prio_root_ns(dev, dev->priv.esw_egress_root_ns); - cleanup_single_prio_root_ns(dev, dev->priv.esw_ingress_root_ns); + cleanup_root_ns(steering); + cleanup_single_prio_root_ns(steering, steering->esw_egress_root_ns); + cleanup_single_prio_root_ns(steering, steering->esw_ingress_root_ns); + cleanup_single_prio_root_ns(steering, steering->fdb_root_ns); mlx5_cleanup_fc_stats(dev); + kfree(steering); } -static int init_fdb_root_ns(struct mlx5_core_dev *dev) +static int init_fdb_root_ns(struct mlx5_flow_steering *steering) { struct fs_prio *prio; - dev->priv.fdb_root_ns = create_root_ns(dev, FS_FT_FDB); - if (!dev->priv.fdb_root_ns) + steering->fdb_root_ns = create_root_ns(steering, FS_FT_FDB); + if (!steering->fdb_root_ns) return -ENOMEM; /* Create single prio */ - prio = fs_create_prio(&dev->priv.fdb_root_ns->ns, 0, 1); + prio = fs_create_prio(&steering->fdb_root_ns->ns, 0, 1); if (IS_ERR(prio)) { - cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns); + cleanup_single_prio_root_ns(steering, steering->fdb_root_ns); return PTR_ERR(prio); } else { return 0; } } -static int init_egress_acl_root_ns(struct mlx5_core_dev *dev) +static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering) { struct fs_prio *prio; - dev->priv.esw_egress_root_ns = create_root_ns(dev, FS_FT_ESW_EGRESS_ACL); - if (!dev->priv.esw_egress_root_ns) + steering->esw_egress_root_ns = create_root_ns(steering, FS_FT_ESW_EGRESS_ACL); + if (!steering->esw_egress_root_ns) return -ENOMEM; /* create 1 prio*/ - prio = fs_create_prio(&dev->priv.esw_egress_root_ns->ns, 0, MLX5_TOTAL_VPORTS(dev)); + prio = fs_create_prio(&steering->esw_egress_root_ns->ns, 0, + MLX5_TOTAL_VPORTS(steering->dev)); if (IS_ERR(prio)) return PTR_ERR(prio); else return 0; } -static int init_ingress_acl_root_ns(struct mlx5_core_dev *dev) +static int init_egress_acl_root_ns(struct mlx5_flow_steering *steering) { struct fs_prio *prio; - dev->priv.esw_ingress_root_ns = create_root_ns(dev, FS_FT_ESW_INGRESS_ACL); - if (!dev->priv.esw_ingress_root_ns) + steering->esw_ingress_root_ns = create_root_ns(steering, FS_FT_ESW_INGRESS_ACL); + if (!steering->esw_ingress_root_ns) return -ENOMEM; /* create 1 prio*/ - prio = fs_create_prio(&dev->priv.esw_ingress_root_ns->ns, 0, MLX5_TOTAL_VPORTS(dev)); + prio = fs_create_prio(&steering->esw_ingress_root_ns->ns, 0, + MLX5_TOTAL_VPORTS(steering->dev)); if (IS_ERR(prio)) return PTR_ERR(prio); else @@ -1834,6 +1843,7 @@ static int init_ingress_acl_root_ns(struct mlx5_core_dev *dev) int mlx5_init_fs(struct mlx5_core_dev *dev) { + struct mlx5_flow_steering *steering; int err = 0; if (MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) @@ -1843,26 +1853,32 @@ int mlx5_init_fs(struct mlx5_core_dev *dev) if (err) return err; + steering = kzalloc(sizeof(*steering), GFP_KERNEL); + if (!steering) + return -ENOMEM; + steering->dev = dev; + dev->priv.steering = steering; + if (MLX5_CAP_GEN(dev, nic_flow_table) && MLX5_CAP_FLOWTABLE_NIC_RX(dev, ft_support)) { - err = init_root_ns(dev); + err = init_root_ns(steering); if (err) goto err; } if (MLX5_CAP_GEN(dev, eswitch_flow_table)) { if (MLX5_CAP_ESW_FLOWTABLE_FDB(dev, ft_support)) { - err = init_fdb_root_ns(dev); + err = init_fdb_root_ns(steering); if (err) goto err; } if (MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support)) { - err = init_egress_acl_root_ns(dev); + err = init_egress_acl_root_ns(steering); if (err) goto err; } if (MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support)) { - err = init_ingress_acl_root_ns(dev); + err = init_ingress_acl_root_ns(steering); if (err) goto err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index aa41a73..d7ba91a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -55,6 +55,14 @@ enum fs_fte_status { FS_FTE_STATUS_EXISTING = 1UL << 0, }; +struct mlx5_flow_steering { + struct mlx5_core_dev *dev; + struct mlx5_flow_root_namespace *root_ns; + struct mlx5_flow_root_namespace *fdb_root_ns; + struct mlx5_flow_root_namespace *esw_egress_root_ns; + struct mlx5_flow_root_namespace *esw_ingress_root_ns; +}; + struct fs_node { struct list_head list; struct list_head children; diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index e22b345..f21c459 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -550,14 +550,10 @@ struct mlx5_priv { struct list_head ctx_list; spinlock_t ctx_lock; + struct mlx5_flow_steering *steering; struct mlx5_eswitch *eswitch; struct mlx5_core_sriov sriov; unsigned long pci_dev_data; - struct mlx5_flow_root_namespace *root_ns; - struct mlx5_flow_root_namespace *fdb_root_ns; - struct mlx5_flow_root_namespace *esw_egress_root_ns; - struct mlx5_flow_root_namespace *esw_ingress_root_ns; - struct mlx5_fc_stats fc_stats; struct mlx5_rl_table rl_table; }; -- cgit v0.10.2 From 0da2d66666d32769fa0aebb5f1d2d0a86be6c5d2 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 4 Jul 2016 17:23:07 +0300 Subject: net/mlx5: Properly remove all steering objects Instead of explicitly cleaning up the well known parts of the steering tree, we use the generic tree structure to traverse for cleanup. No functional changes. Signed-off-by: Maor Gottlieb Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 0f969cb..3e95775 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -1663,115 +1663,24 @@ cleanup: return -ENOMEM; } -static void cleanup_single_prio_root_ns(struct mlx5_flow_steering *steering, - struct mlx5_flow_root_namespace *root_ns) +static void clean_tree(struct fs_node *node) { - struct fs_node *prio; - - if (!root_ns) - return; + if (node) { + struct fs_node *iter; + struct fs_node *temp; - if (!list_empty(&root_ns->ns.node.children)) { - prio = list_first_entry(&root_ns->ns.node.children, - struct fs_node, - list); - if (tree_remove_node(prio)) - mlx5_core_warn(steering->dev, - "Flow steering priority wasn't destroyed, refcount > 1\n"); + list_for_each_entry_safe(iter, temp, &node->children, list) + clean_tree(iter); + tree_remove_node(node); } - if (tree_remove_node(&root_ns->ns.node)) - mlx5_core_warn(steering->dev, - "Flow steering namespace wasn't destroyed, refcount > 1\n"); - root_ns = NULL; -} - -static void destroy_flow_tables(struct fs_prio *prio) -{ - struct mlx5_flow_table *iter; - struct mlx5_flow_table *tmp; - - fs_for_each_ft_safe(iter, tmp, prio) - mlx5_destroy_flow_table(iter); } -static void cleanup_root_ns(struct mlx5_flow_steering *steering) +static void cleanup_root_ns(struct mlx5_flow_root_namespace *root_ns) { - struct mlx5_flow_root_namespace *root_ns = steering->root_ns; - struct fs_prio *iter_prio; - - if (!MLX5_CAP_GEN(steering->dev, nic_flow_table)) - return; - if (!root_ns) return; - /* stage 1 */ - fs_for_each_prio(iter_prio, &root_ns->ns) { - struct fs_node *node; - struct mlx5_flow_namespace *iter_ns; - - fs_for_each_ns_or_ft(node, iter_prio) { - if (node->type == FS_TYPE_FLOW_TABLE) - continue; - fs_get_obj(iter_ns, node); - while (!list_empty(&iter_ns->node.children)) { - struct fs_prio *obj_iter_prio2; - struct fs_node *iter_prio2 = - list_first_entry(&iter_ns->node.children, - struct fs_node, - list); - - fs_get_obj(obj_iter_prio2, iter_prio2); - destroy_flow_tables(obj_iter_prio2); - if (tree_remove_node(iter_prio2)) { - mlx5_core_warn(steering->dev, - "Priority %d wasn't destroyed, refcount > 1\n", - obj_iter_prio2->prio); - return; - } - } - } - } - - /* stage 2 */ - fs_for_each_prio(iter_prio, &root_ns->ns) { - while (!list_empty(&iter_prio->node.children)) { - struct fs_node *iter_ns = - list_first_entry(&iter_prio->node.children, - struct fs_node, - list); - if (tree_remove_node(iter_ns)) { - mlx5_core_warn(steering->dev, - "Namespace wasn't destroyed, refcount > 1\n"); - return; - } - } - } - - /* stage 3 */ - while (!list_empty(&root_ns->ns.node.children)) { - struct fs_prio *obj_prio_node; - struct fs_node *prio_node = - list_first_entry(&root_ns->ns.node.children, - struct fs_node, - list); - - fs_get_obj(obj_prio_node, prio_node); - if (tree_remove_node(prio_node)) { - mlx5_core_warn(steering->dev, - "Priority %d wasn't destroyed, refcount > 1\n", - obj_prio_node->prio); - return; - } - } - - if (tree_remove_node(&root_ns->ns.node)) { - mlx5_core_warn(steering->dev, - "root namespace wasn't destroyed, refcount > 1\n"); - return; - } - - steering->root_ns = NULL; + clean_tree(&root_ns->ns.node); } void mlx5_cleanup_fs(struct mlx5_core_dev *dev) @@ -1781,10 +1690,10 @@ void mlx5_cleanup_fs(struct mlx5_core_dev *dev) if (MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) return; - cleanup_root_ns(steering); - cleanup_single_prio_root_ns(steering, steering->esw_egress_root_ns); - cleanup_single_prio_root_ns(steering, steering->esw_ingress_root_ns); - cleanup_single_prio_root_ns(steering, steering->fdb_root_ns); + cleanup_root_ns(steering->root_ns); + cleanup_root_ns(steering->esw_egress_root_ns); + cleanup_root_ns(steering->esw_ingress_root_ns); + cleanup_root_ns(steering->fdb_root_ns); mlx5_cleanup_fc_stats(dev); kfree(steering); } @@ -1800,7 +1709,8 @@ static int init_fdb_root_ns(struct mlx5_flow_steering *steering) /* Create single prio */ prio = fs_create_prio(&steering->fdb_root_ns->ns, 0, 1); if (IS_ERR(prio)) { - cleanup_single_prio_root_ns(steering, steering->fdb_root_ns); + cleanup_root_ns(steering->fdb_root_ns); + steering->fdb_root_ns = NULL; return PTR_ERR(prio); } else { return 0; -- cgit v0.10.2 From 6dc6071cfcde6cf687f8d288c9cef9ee6ee24dc7 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 4 Jul 2016 17:23:08 +0300 Subject: net/mlx5e: Add ethtool flow steering support Implement etrhtool set_rxnfc callback to support ethtool flow spec direct steering. This patch adds only the support of ether flow type spec. L3/L4 flow specs support will be added in downstream patches. Signed-off-by: Maor Gottlieb Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index a574dea..05cc1ef 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -8,6 +8,6 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o eswitch_offloads.o \ en_main.o en_common.o en_fs.o en_ethtool.o en_tx.o \ en_rx.o en_rx_am.o en_txrx.o en_clock.o vxlan.o \ - en_tc.o en_arfs.o en_rep.o + en_tc.o en_arfs.o en_rep.o en_fs_ethtool.o mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 00643a1..357320e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -544,8 +544,22 @@ enum { MLX5E_ARFS_FT_LEVEL }; +struct mlx5e_ethtool_table { + struct mlx5_flow_table *ft; + int num_rules; +}; + +#define ETHTOOL_NUM_L2_FTS 4 + +struct mlx5e_ethtool_steering { + struct mlx5e_ethtool_table l2_ft[ETHTOOL_NUM_L2_FTS]; + struct list_head rules; + int tot_num_rules; +}; + struct mlx5e_flow_steering { struct mlx5_flow_namespace *ns; + struct mlx5e_ethtool_steering ethtool; struct mlx5e_tc_table tc; struct mlx5e_vlan_table vlan; struct mlx5e_l2_table l2; @@ -701,6 +715,12 @@ int mlx5e_create_flow_steering(struct mlx5e_priv *priv); void mlx5e_destroy_flow_steering(struct mlx5e_priv *priv); void mlx5e_init_l2_addr(struct mlx5e_priv *priv); void mlx5e_destroy_flow_table(struct mlx5e_flow_table *ft); +int mlx5e_ethtool_flow_replace(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs); +int mlx5e_ethtool_flow_remove(struct mlx5e_priv *priv, + int location); +void mlx5e_ethtool_init_steering(struct mlx5e_priv *priv); +void mlx5e_ethtool_cleanup_steering(struct mlx5e_priv *priv); void mlx5e_set_rx_mode_work(struct work_struct *work); void mlx5e_fill_hwstamp(struct mlx5e_tstamp *clock, u64 timestamp, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 7e61ffa..edbb665 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -1368,6 +1368,26 @@ static u32 mlx5e_get_priv_flags(struct net_device *netdev) return priv->pflags; } +static int mlx5e_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd) +{ + int err = 0; + struct mlx5e_priv *priv = netdev_priv(dev); + + switch (cmd->cmd) { + case ETHTOOL_SRXCLSRLINS: + err = mlx5e_ethtool_flow_replace(priv, &cmd->fs); + break; + case ETHTOOL_SRXCLSRLDEL: + err = mlx5e_ethtool_flow_remove(priv, cmd->fs.location); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + const struct ethtool_ops mlx5e_ethtool_ops = { .get_drvinfo = mlx5e_get_drvinfo, .get_link = ethtool_op_get_link, @@ -1387,6 +1407,7 @@ const struct ethtool_ops mlx5e_ethtool_ops = { .get_rxfh = mlx5e_get_rxfh, .set_rxfh = mlx5e_set_rxfh, .get_rxnfc = mlx5e_get_rxnfc, + .set_rxnfc = mlx5e_set_rxnfc, .get_tunable = mlx5e_get_tunable, .set_tunable = mlx5e_set_tunable, .get_pauseparam = mlx5e_get_pauseparam, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c index 2e1e863..1587a9f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c @@ -1084,6 +1084,8 @@ int mlx5e_create_flow_steering(struct mlx5e_priv *priv) goto err_destroy_l2_table; } + mlx5e_ethtool_init_steering(priv); + return 0; err_destroy_l2_table: @@ -1103,4 +1105,5 @@ void mlx5e_destroy_flow_steering(struct mlx5e_priv *priv) mlx5e_destroy_l2_table(priv); mlx5e_destroy_ttc_table(priv); mlx5e_arfs_destroy_tables(priv); + mlx5e_ethtool_cleanup_steering(priv); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c new file mode 100644 index 0000000..ee28a9f --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c @@ -0,0 +1,393 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include "en.h" + +struct mlx5e_ethtool_rule { + struct list_head list; + struct ethtool_rx_flow_spec flow_spec; + struct mlx5_flow_rule *rule; + struct mlx5e_ethtool_table *eth_ft; +}; + +static void put_flow_table(struct mlx5e_ethtool_table *eth_ft) +{ + if (!--eth_ft->num_rules) { + mlx5_destroy_flow_table(eth_ft->ft); + eth_ft->ft = NULL; + } +} + +#define MLX5E_ETHTOOL_L2_PRIO 0 +#define MLX5E_ETHTOOL_NUM_ENTRIES 64000 +#define MLX5E_ETHTOOL_NUM_GROUPS 10 +static struct mlx5e_ethtool_table *get_flow_table(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs, + int num_tuples) +{ + struct mlx5e_ethtool_table *eth_ft; + struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *ft; + int max_tuples; + int table_size; + int prio; + + switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) { + case ETHER_FLOW: + max_tuples = ETHTOOL_NUM_L2_FTS; + prio = max_tuples - num_tuples; + eth_ft = &priv->fs.ethtool.l2_ft[prio]; + prio += MLX5E_ETHTOOL_L2_PRIO; + break; + default: + return ERR_PTR(-EINVAL); + } + + eth_ft->num_rules++; + if (eth_ft->ft) + return eth_ft; + + ns = mlx5_get_flow_namespace(priv->mdev, + MLX5_FLOW_NAMESPACE_ETHTOOL); + if (!ns) + return ERR_PTR(-ENOTSUPP); + + table_size = min_t(u32, BIT(MLX5_CAP_FLOWTABLE(priv->mdev, + flow_table_properties_nic_receive.log_max_ft_size)), + MLX5E_ETHTOOL_NUM_ENTRIES); + ft = mlx5_create_auto_grouped_flow_table(ns, prio, + table_size, + MLX5E_ETHTOOL_NUM_GROUPS, 0); + if (IS_ERR(ft)) + return (void *)ft; + + eth_ft->ft = ft; + return eth_ft; +} + +static void mask_spec(u8 *mask, u8 *val, size_t size) +{ + unsigned int i; + + for (i = 0; i < size; i++, mask++, val++) + *((u8 *)val) = *((u8 *)mask) & *((u8 *)val); +} + +static int set_flow_attrs(u32 *match_c, u32 *match_v, + struct ethtool_rx_flow_spec *fs) +{ + void *outer_headers_c = MLX5_ADDR_OF(fte_match_param, match_c, + outer_headers); + void *outer_headers_v = MLX5_ADDR_OF(fte_match_param, match_v, + outer_headers); + u32 flow_type = fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT); + struct ethhdr *eth_val; + struct ethhdr *eth_mask; + + switch (flow_type) { + case ETHER_FLOW: + eth_mask = &fs->m_u.ether_spec; + eth_val = &fs->h_u.ether_spec; + + mask_spec((u8 *)eth_mask, (u8 *)eth_val, sizeof(*eth_mask)); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_c, smac_47_16), + eth_mask->h_source); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_v, smac_47_16), + eth_val->h_source); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_c, dmac_47_16), + eth_mask->h_dest); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_v, dmac_47_16), + eth_val->h_dest); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ethertype, + ntohs(eth_mask->h_proto)); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ethertype, + ntohs(eth_val->h_proto)); + break; + default: + return -EINVAL; + } + + if ((fs->flow_type & FLOW_EXT) && + (fs->m_ext.vlan_tci & cpu_to_be16(VLAN_VID_MASK))) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + vlan_tag, 1); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + vlan_tag, 1); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + first_vid, 0xfff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + first_vid, ntohs(fs->h_ext.vlan_tci)); + } + + return 0; +} + +static void add_rule_to_list(struct mlx5e_priv *priv, + struct mlx5e_ethtool_rule *rule) +{ + struct mlx5e_ethtool_rule *iter; + struct list_head *head = &priv->fs.ethtool.rules; + + list_for_each_entry(iter, &priv->fs.ethtool.rules, list) { + if (iter->flow_spec.location > rule->flow_spec.location) + break; + head = &iter->list; + } + priv->fs.ethtool.tot_num_rules++; + list_add(&rule->list, head); +} + +static bool outer_header_zero(u32 *match_criteria) +{ + int size = MLX5_ST_SZ_BYTES(fte_match_param); + char *outer_headers_c = MLX5_ADDR_OF(fte_match_param, match_criteria, + outer_headers); + + return outer_headers_c[0] == 0 && !memcmp(outer_headers_c, + outer_headers_c + 1, + size - 1); +} + +static struct mlx5_flow_rule *add_ethtool_flow_rule(struct mlx5e_priv *priv, + struct mlx5_flow_table *ft, + struct ethtool_rx_flow_spec *fs) +{ + struct mlx5_flow_destination *dst = NULL; + struct mlx5_flow_spec *spec; + struct mlx5_flow_rule *rule; + int err = 0; + u32 action; + + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) + return ERR_PTR(-ENOMEM); + err = set_flow_attrs(spec->match_criteria, spec->match_value, + fs); + if (err) + goto free; + + if (fs->ring_cookie == RX_CLS_FLOW_DISC) { + action = MLX5_FLOW_CONTEXT_ACTION_DROP; + } else { + dst = kzalloc(sizeof(*dst), GFP_KERNEL); + if (!dst) { + err = -ENOMEM; + goto free; + } + + dst->type = MLX5_FLOW_DESTINATION_TYPE_TIR; + dst->tir_num = priv->direct_tir[fs->ring_cookie].tirn; + action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + } + + spec->match_criteria_enable = (!outer_header_zero(spec->match_criteria)); + rule = mlx5_add_flow_rule(ft, spec, action, + MLX5_FS_DEFAULT_FLOW_TAG, dst); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + netdev_err(priv->netdev, "%s: failed to add ethtool steering rule: %d\n", + __func__, err); + goto free; + } +free: + kvfree(spec); + kfree(dst); + return err ? ERR_PTR(err) : rule; +} + +static void del_ethtool_rule(struct mlx5e_priv *priv, + struct mlx5e_ethtool_rule *eth_rule) +{ + if (eth_rule->rule) + mlx5_del_flow_rule(eth_rule->rule); + list_del(ð_rule->list); + priv->fs.ethtool.tot_num_rules--; + put_flow_table(eth_rule->eth_ft); + kfree(eth_rule); +} + +static struct mlx5e_ethtool_rule *find_ethtool_rule(struct mlx5e_priv *priv, + int location) +{ + struct mlx5e_ethtool_rule *iter; + + list_for_each_entry(iter, &priv->fs.ethtool.rules, list) { + if (iter->flow_spec.location == location) + return iter; + } + return NULL; +} + +static struct mlx5e_ethtool_rule *get_ethtool_rule(struct mlx5e_priv *priv, + int location) +{ + struct mlx5e_ethtool_rule *eth_rule; + + eth_rule = find_ethtool_rule(priv, location); + if (eth_rule) + del_ethtool_rule(priv, eth_rule); + + eth_rule = kzalloc(sizeof(*eth_rule), GFP_KERNEL); + if (!eth_rule) + return ERR_PTR(-ENOMEM); + + add_rule_to_list(priv, eth_rule); + return eth_rule; +} + +#define MAX_NUM_OF_ETHTOOL_RULES BIT(10) +static int validate_flow(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs) +{ + struct ethhdr *eth_mask; + int num_tuples = 0; + + if (fs->location >= MAX_NUM_OF_ETHTOOL_RULES) + return -EINVAL; + + if (fs->ring_cookie >= priv->params.num_channels && + fs->ring_cookie != RX_CLS_FLOW_DISC) + return -EINVAL; + + switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) { + case ETHER_FLOW: + eth_mask = &fs->m_u.ether_spec; + if (!is_zero_ether_addr(eth_mask->h_dest)) + num_tuples++; + if (!is_zero_ether_addr(eth_mask->h_source)) + num_tuples++; + if (eth_mask->h_proto) + num_tuples++; + break; + default: + return -EINVAL; + } + if ((fs->flow_type & FLOW_EXT)) { + if (fs->m_ext.vlan_etype || + (fs->m_ext.vlan_tci != cpu_to_be16(VLAN_VID_MASK))) + return -EINVAL; + + if (fs->m_ext.vlan_tci) { + if (be16_to_cpu(fs->h_ext.vlan_tci) >= VLAN_N_VID) + return -EINVAL; + } + num_tuples++; + } + + return num_tuples; +} + +int mlx5e_ethtool_flow_replace(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs) +{ + struct mlx5e_ethtool_table *eth_ft; + struct mlx5e_ethtool_rule *eth_rule; + struct mlx5_flow_rule *rule; + int num_tuples; + int err; + + num_tuples = validate_flow(priv, fs); + if (num_tuples <= 0) { + netdev_warn(priv->netdev, "%s: flow is not valid\n", __func__); + return -EINVAL; + } + + eth_ft = get_flow_table(priv, fs, num_tuples); + if (IS_ERR(eth_ft)) + return PTR_ERR(eth_ft); + + eth_rule = get_ethtool_rule(priv, fs->location); + if (IS_ERR(eth_rule)) { + put_flow_table(eth_ft); + return PTR_ERR(eth_rule); + } + + eth_rule->flow_spec = *fs; + eth_rule->eth_ft = eth_ft; + if (!eth_ft->ft) { + err = -EINVAL; + goto del_ethtool_rule; + } + rule = add_ethtool_flow_rule(priv, eth_ft->ft, fs); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + goto del_ethtool_rule; + } + + eth_rule->rule = rule; + + return 0; + +del_ethtool_rule: + del_ethtool_rule(priv, eth_rule); + + return err; +} + +int mlx5e_ethtool_flow_remove(struct mlx5e_priv *priv, + int location) +{ + struct mlx5e_ethtool_rule *eth_rule; + int err = 0; + + if (location >= MAX_NUM_OF_ETHTOOL_RULES) + return -ENOSPC; + + eth_rule = find_ethtool_rule(priv, location); + if (!eth_rule) { + err = -ENOENT; + goto out; + } + + del_ethtool_rule(priv, eth_rule); +out: + return err; +} + +void mlx5e_ethtool_cleanup_steering(struct mlx5e_priv *priv) +{ + struct mlx5e_ethtool_rule *iter; + struct mlx5e_ethtool_rule *temp; + + list_for_each_entry_safe(iter, temp, &priv->fs.ethtool.rules, list) + del_ethtool_rule(priv, iter); +} + +void mlx5e_ethtool_init_steering(struct mlx5e_priv *priv) +{ + INIT_LIST_HEAD(&priv->fs.ethtool.rules); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 3e95775..83fa98f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -67,13 +67,21 @@ #define FS_REQUIRED_CAPS(...) {.arr_sz = INIT_CAPS_ARRAY_SIZE(__VA_ARGS__), \ .caps = (long[]) {__VA_ARGS__} } +#define FS_CHAINING_CAPS FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), \ + FS_CAP(flow_table_properties_nic_receive.modify_root), \ + FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), \ + FS_CAP(flow_table_properties_nic_receive.flow_table_modify)) + #define LEFTOVERS_NUM_LEVELS 1 #define LEFTOVERS_NUM_PRIOS 1 #define BY_PASS_PRIO_NUM_LEVELS 1 -#define BY_PASS_MIN_LEVEL (KERNEL_MIN_LEVEL + MLX5_BY_PASS_NUM_PRIOS +\ +#define BY_PASS_MIN_LEVEL (ETHTOOL_MIN_LEVEL + MLX5_BY_PASS_NUM_PRIOS +\ LEFTOVERS_NUM_PRIOS) +#define ETHTOOL_PRIO_NUM_LEVELS 1 +#define ETHTOOL_NUM_PRIOS 4 +#define ETHTOOL_MIN_LEVEL (KERNEL_MIN_LEVEL + ETHTOOL_NUM_PRIOS) /* Vlan, mac, ttc, aRFS */ #define KERNEL_NIC_PRIO_NUM_LEVELS 4 #define KERNEL_NIC_NUM_PRIOS 1 @@ -103,27 +111,24 @@ static struct init_tree_node { int num_levels; } root_fs = { .type = FS_TYPE_NAMESPACE, - .ar_size = 5, + .ar_size = 6, .children = (struct init_tree_node[]) { ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0, - FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), - FS_CAP(flow_table_properties_nic_receive.modify_root), - FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), - FS_CAP(flow_table_properties_nic_receive.flow_table_modify)), + FS_CHAINING_CAPS, ADD_NS(ADD_MULTIPLE_PRIO(MLX5_BY_PASS_NUM_PRIOS, BY_PASS_PRIO_NUM_LEVELS))), ADD_PRIO(0, OFFLOADS_MIN_LEVEL, 0, {}, ADD_NS(ADD_MULTIPLE_PRIO(OFFLOADS_NUM_PRIOS, OFFLOADS_MAX_FT))), - + ADD_PRIO(0, ETHTOOL_MIN_LEVEL, 0, + FS_CHAINING_CAPS, + ADD_NS(ADD_MULTIPLE_PRIO(ETHTOOL_NUM_PRIOS, + ETHTOOL_PRIO_NUM_LEVELS))), ADD_PRIO(0, KERNEL_MIN_LEVEL, 0, {}, ADD_NS(ADD_MULTIPLE_PRIO(1, 1), ADD_MULTIPLE_PRIO(KERNEL_NIC_NUM_PRIOS, KERNEL_NIC_PRIO_NUM_LEVELS))), ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0, - FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), - FS_CAP(flow_table_properties_nic_receive.modify_root), - FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), - FS_CAP(flow_table_properties_nic_receive.flow_table_modify)), + FS_CHAINING_CAPS, ADD_NS(ADD_MULTIPLE_PRIO(LEFTOVERS_NUM_PRIOS, LEFTOVERS_NUM_LEVELS))), ADD_PRIO(0, ANCHOR_MIN_LEVEL, 0, {}, ADD_NS(ADD_MULTIPLE_PRIO(ANCHOR_NUM_PRIOS, ANCHOR_NUM_LEVELS))), @@ -1375,6 +1380,7 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, switch (type) { case MLX5_FLOW_NAMESPACE_BYPASS: case MLX5_FLOW_NAMESPACE_OFFLOADS: + case MLX5_FLOW_NAMESPACE_ETHTOOL: case MLX5_FLOW_NAMESPACE_KERNEL: case MLX5_FLOW_NAMESPACE_LEFTOVERS: case MLX5_FLOW_NAMESPACE_ANCHOR: diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index d22fe7e..e036d60 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -55,6 +55,7 @@ static inline void build_leftovers_ft_param(int *priority, enum mlx5_flow_namespace_type { MLX5_FLOW_NAMESPACE_BYPASS, MLX5_FLOW_NAMESPACE_OFFLOADS, + MLX5_FLOW_NAMESPACE_ETHTOOL, MLX5_FLOW_NAMESPACE_KERNEL, MLX5_FLOW_NAMESPACE_LEFTOVERS, MLX5_FLOW_NAMESPACE_ANCHOR, -- cgit v0.10.2 From 1174fce8d1410d13b665cb7693250cc789637b9a Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 4 Jul 2016 17:23:09 +0300 Subject: net/mlx5e: Support l3/l4 flow type specs in ethtool flow steering Add support to add flow steering rules with ethtool of L3/L4 flow types (ip4/tcp4/udp4). Those rules will be in higher priority than l2 flow rules, in order to prefer more specific rules. Mask is not supported for l3/l4 flow types. Signed-off-by: Maor Gottlieb Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 357320e..9842594 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -549,9 +549,11 @@ struct mlx5e_ethtool_table { int num_rules; }; +#define ETHTOOL_NUM_L3_L4_FTS 7 #define ETHTOOL_NUM_L2_FTS 4 struct mlx5e_ethtool_steering { + struct mlx5e_ethtool_table l3_l4_ft[ETHTOOL_NUM_L3_L4_FTS]; struct mlx5e_ethtool_table l2_ft[ETHTOOL_NUM_L2_FTS]; struct list_head rules; int tot_num_rules; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c index ee28a9f..830106e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c @@ -48,7 +48,8 @@ static void put_flow_table(struct mlx5e_ethtool_table *eth_ft) } } -#define MLX5E_ETHTOOL_L2_PRIO 0 +#define MLX5E_ETHTOOL_L3_L4_PRIO 0 +#define MLX5E_ETHTOOL_L2_PRIO (MLX5E_ETHTOOL_L3_L4_PRIO + ETHTOOL_NUM_L3_L4_FTS) #define MLX5E_ETHTOOL_NUM_ENTRIES 64000 #define MLX5E_ETHTOOL_NUM_GROUPS 10 static struct mlx5e_ethtool_table *get_flow_table(struct mlx5e_priv *priv, @@ -63,6 +64,17 @@ static struct mlx5e_ethtool_table *get_flow_table(struct mlx5e_priv *priv, int prio; switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) { + case TCP_V4_FLOW: + case UDP_V4_FLOW: + max_tuples = ETHTOOL_NUM_L3_L4_FTS; + prio = MLX5E_ETHTOOL_L3_L4_PRIO + (max_tuples - num_tuples); + eth_ft = &priv->fs.ethtool.l3_l4_ft[prio]; + break; + case IP_USER_FLOW: + max_tuples = ETHTOOL_NUM_L3_L4_FTS; + prio = MLX5E_ETHTOOL_L3_L4_PRIO + (max_tuples - num_tuples); + eth_ft = &priv->fs.ethtool.l3_l4_ft[prio]; + break; case ETHER_FLOW: max_tuples = ETHTOOL_NUM_L2_FTS; prio = max_tuples - num_tuples; @@ -103,6 +115,31 @@ static void mask_spec(u8 *mask, u8 *val, size_t size) *((u8 *)val) = *((u8 *)mask) & *((u8 *)val); } +static void set_ips(void *outer_headers_v, void *outer_headers_c, __be32 ip4src_m, + __be32 ip4src_v, __be32 ip4dst_m, __be32 ip4dst_v) +{ + if (ip4src_m) { + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + src_ipv4_src_ipv6.ipv4_layout.ipv4), + &ip4src_v, sizeof(ip4src_v)); + memset(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + src_ipv4_src_ipv6.ipv4_layout.ipv4), + 0xff, sizeof(ip4src_m)); + } + if (ip4dst_m) { + memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_v, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4), + &ip4dst_v, sizeof(ip4dst_v)); + memset(MLX5_ADDR_OF(fte_match_set_lyr_2_4, outer_headers_c, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4), + 0xff, sizeof(ip4dst_m)); + } + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + ethertype, ETH_P_IP); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + ethertype, 0xffff); +} + static int set_flow_attrs(u32 *match_c, u32 *match_v, struct ethtool_rx_flow_spec *fs) { @@ -111,10 +148,66 @@ static int set_flow_attrs(u32 *match_c, u32 *match_v, void *outer_headers_v = MLX5_ADDR_OF(fte_match_param, match_v, outer_headers); u32 flow_type = fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT); + struct ethtool_tcpip4_spec *l4_mask; + struct ethtool_tcpip4_spec *l4_val; + struct ethtool_usrip4_spec *l3_mask; + struct ethtool_usrip4_spec *l3_val; struct ethhdr *eth_val; struct ethhdr *eth_mask; switch (flow_type) { + case TCP_V4_FLOW: + l4_mask = &fs->m_u.tcp_ip4_spec; + l4_val = &fs->h_u.tcp_ip4_spec; + set_ips(outer_headers_v, outer_headers_c, l4_mask->ip4src, + l4_val->ip4src, l4_mask->ip4dst, l4_val->ip4dst); + + if (l4_mask->psrc) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, tcp_sport, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, tcp_sport, + ntohs(l4_val->psrc)); + } + if (l4_mask->pdst) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, tcp_dport, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, tcp_dport, + ntohs(l4_val->pdst)); + } + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ip_protocol, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ip_protocol, + IPPROTO_TCP); + break; + case UDP_V4_FLOW: + l4_mask = &fs->m_u.tcp_ip4_spec; + l4_val = &fs->h_u.tcp_ip4_spec; + set_ips(outer_headers_v, outer_headers_c, l4_mask->ip4src, + l4_val->ip4src, l4_mask->ip4dst, l4_val->ip4dst); + + if (l4_mask->psrc) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, udp_sport, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, udp_sport, + ntohs(l4_val->psrc)); + } + if (l4_mask->pdst) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, udp_dport, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, udp_dport, + ntohs(l4_val->pdst)); + } + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ip_protocol, + 0xffff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ip_protocol, + IPPROTO_UDP); + break; + case IP_USER_FLOW: + l3_mask = &fs->m_u.usr_ip4_spec; + l3_val = &fs->h_u.usr_ip4_spec; + set_ips(outer_headers_v, outer_headers_c, l3_mask->ip4src, + l3_val->ip4src, l3_mask->ip4dst, l3_val->ip4dst); + break; case ETHER_FLOW: eth_mask = &fs->m_u.ether_spec; eth_val = &fs->h_u.ether_spec; @@ -152,6 +245,15 @@ static int set_flow_attrs(u32 *match_c, u32 *match_v, MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, first_vid, ntohs(fs->h_ext.vlan_tci)); } + if (fs->flow_type & FLOW_MAC_EXT && + !is_zero_ether_addr(fs->m_ext.h_dest)) { + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_c, dmac_47_16), + fs->m_ext.h_dest); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_v, dmac_47_16), + fs->h_ext.h_dest); + } return 0; } @@ -270,9 +372,16 @@ static struct mlx5e_ethtool_rule *get_ethtool_rule(struct mlx5e_priv *priv, } #define MAX_NUM_OF_ETHTOOL_RULES BIT(10) + +#define all_ones(field) (field == (__force typeof(field))-1) +#define all_zeros_or_all_ones(field) \ + ((field) == 0 || (field) == (__force typeof(field))-1) + static int validate_flow(struct mlx5e_priv *priv, struct ethtool_rx_flow_spec *fs) { + struct ethtool_tcpip4_spec *l4_mask; + struct ethtool_usrip4_spec *l3_mask; struct ethhdr *eth_mask; int num_tuples = 0; @@ -293,6 +402,52 @@ static int validate_flow(struct mlx5e_priv *priv, if (eth_mask->h_proto) num_tuples++; break; + case TCP_V4_FLOW: + case UDP_V4_FLOW: + if (fs->m_u.tcp_ip4_spec.tos) + return -EINVAL; + l4_mask = &fs->m_u.tcp_ip4_spec; + if (l4_mask->ip4src) { + if (!all_ones(l4_mask->ip4src)) + return -EINVAL; + num_tuples++; + } + if (l4_mask->ip4dst) { + if (!all_ones(l4_mask->ip4dst)) + return -EINVAL; + num_tuples++; + } + if (l4_mask->psrc) { + if (!all_ones(l4_mask->psrc)) + return -EINVAL; + num_tuples++; + } + if (l4_mask->pdst) { + if (!all_ones(l4_mask->pdst)) + return -EINVAL; + num_tuples++; + } + /* Flow is TCP/UDP */ + num_tuples++; + break; + case IP_USER_FLOW: + l3_mask = &fs->m_u.usr_ip4_spec; + if (l3_mask->l4_4_bytes || l3_mask->tos || l3_mask->proto || + fs->h_u.usr_ip4_spec.ip_ver != ETH_RX_NFC_IP4) + return -EINVAL; + if (l3_mask->ip4src) { + if (!all_ones(l3_mask->ip4src)) + return -EINVAL; + num_tuples++; + } + if (l3_mask->ip4dst) { + if (!all_ones(l3_mask->ip4dst)) + return -EINVAL; + num_tuples++; + } + /* Flow is IPv4 */ + num_tuples++; + break; default: return -EINVAL; } @@ -308,6 +463,10 @@ static int validate_flow(struct mlx5e_priv *priv, num_tuples++; } + if (fs->flow_type & FLOW_MAC_EXT && + !is_zero_ether_addr(fs->m_ext.h_dest)) + num_tuples++; + return num_tuples; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 83fa98f..b0a1304 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -80,7 +80,7 @@ LEFTOVERS_NUM_PRIOS) #define ETHTOOL_PRIO_NUM_LEVELS 1 -#define ETHTOOL_NUM_PRIOS 4 +#define ETHTOOL_NUM_PRIOS 10 #define ETHTOOL_MIN_LEVEL (KERNEL_MIN_LEVEL + ETHTOOL_NUM_PRIOS) /* Vlan, mac, ttc, aRFS */ #define KERNEL_NIC_PRIO_NUM_LEVELS 4 -- cgit v0.10.2 From f913a72aa008777d4a92f82acafb17cce9aed4dc Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 4 Jul 2016 17:23:10 +0300 Subject: net/mlx5e: Add support to get ethtool flow rules Enhance the existing get_rxnfc callback: 1. Get flow rule of specific ID. 2. Get all flow rules. 3. Get number of rules. Signed-off-by: Maor Gottlieb Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 9842594..1365cdc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -717,6 +717,10 @@ int mlx5e_create_flow_steering(struct mlx5e_priv *priv); void mlx5e_destroy_flow_steering(struct mlx5e_priv *priv); void mlx5e_init_l2_addr(struct mlx5e_priv *priv); void mlx5e_destroy_flow_table(struct mlx5e_flow_table *ft); +int mlx5e_ethtool_get_flow(struct mlx5e_priv *priv, struct ethtool_rxnfc *info, + int location); +int mlx5e_ethtool_get_all_flows(struct mlx5e_priv *priv, + struct ethtool_rxnfc *info, u32 *rule_locs); int mlx5e_ethtool_flow_replace(struct mlx5e_priv *priv, struct ethtool_rx_flow_spec *fs); int mlx5e_ethtool_flow_remove(struct mlx5e_priv *priv, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index edbb665..d652aa9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -931,6 +931,15 @@ static int mlx5e_get_rxnfc(struct net_device *netdev, case ETHTOOL_GRXRINGS: info->data = priv->params.num_channels; break; + case ETHTOOL_GRXCLSRLCNT: + info->rule_cnt = priv->fs.ethtool.tot_num_rules; + break; + case ETHTOOL_GRXCLSRULE: + err = mlx5e_ethtool_get_flow(priv, info, info->fs.location); + break; + case ETHTOOL_GRXCLSRLALL: + err = mlx5e_ethtool_get_all_flows(priv, info, rule_locs); + break; default: err = -EOPNOTSUPP; break; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c index 830106e..d17c242 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c @@ -537,6 +537,40 @@ out: return err; } +int mlx5e_ethtool_get_flow(struct mlx5e_priv *priv, struct ethtool_rxnfc *info, + int location) +{ + struct mlx5e_ethtool_rule *eth_rule; + + if (location < 0 || location >= MAX_NUM_OF_ETHTOOL_RULES) + return -EINVAL; + + list_for_each_entry(eth_rule, &priv->fs.ethtool.rules, list) { + if (eth_rule->flow_spec.location == location) { + info->fs = eth_rule->flow_spec; + return 0; + } + } + + return -ENOENT; +} + +int mlx5e_ethtool_get_all_flows(struct mlx5e_priv *priv, struct ethtool_rxnfc *info, + u32 *rule_locs) +{ + int location = 0; + int idx = 0; + int err = 0; + + while ((!err || err == -ENOENT) && idx < info->rule_cnt) { + err = mlx5e_ethtool_get_flow(priv, info, location); + if (!err) + rule_locs[idx++] = location; + location++; + } + return err; +} + void mlx5e_ethtool_cleanup_steering(struct mlx5e_priv *priv) { struct mlx5e_ethtool_rule *iter; -- cgit v0.10.2 From fe6b9bd9ebd691f58b4568454946b62cad84fa07 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Mon, 4 Jul 2016 17:23:11 +0300 Subject: net/mlx5e: Expose RDMA VPort counters to ethtool Add the needed descriptors to expose RoCE RDMA counters. Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h index fcd490c..ae29998 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h @@ -151,6 +151,22 @@ static const struct counter_desc vport_stats_desc[] = { VPORT_COUNTER_OFF(transmitted_eth_broadcast.packets) }, { "tx_vport_broadcast_bytes", VPORT_COUNTER_OFF(transmitted_eth_broadcast.octets) }, + { "rx_vport_rdma_unicast_packets", + VPORT_COUNTER_OFF(received_ib_unicast.packets) }, + { "rx_vport_rdma_unicast_bytes", + VPORT_COUNTER_OFF(received_ib_unicast.octets) }, + { "tx_vport_rdma_unicast_packets", + VPORT_COUNTER_OFF(transmitted_ib_unicast.packets) }, + { "tx_vport_rdma_unicast_bytes", + VPORT_COUNTER_OFF(transmitted_ib_unicast.octets) }, + { "rx_vport_rdma_multicast_packets", + VPORT_COUNTER_OFF(received_ib_multicast.packets) }, + { "rx_vport_rdma_multicast_bytes", + VPORT_COUNTER_OFF(received_ib_multicast.octets) }, + { "tx_vport_rdma_multicast_packets", + VPORT_COUNTER_OFF(transmitted_ib_multicast.packets) }, + { "tx_vport_rdma_multicast_bytes", + VPORT_COUNTER_OFF(transmitted_ib_multicast.octets) }, }; #define PPORT_802_3_OFF(c) \ -- cgit v0.10.2 From e989d5a532ce6b0512bd2997c0db46fd0b5e81ec Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Mon, 4 Jul 2016 17:23:12 +0300 Subject: net/mlx5e: Expose flow control counters to ethtool Just like per prio counters, the global flow counters are queried from per priority counters register. Global flow control counters are stored in priority 0 PFC counters. Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index d652aa9..4a3757e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -139,6 +139,18 @@ static unsigned long mlx5e_query_pfc_combined(struct mlx5e_priv *priv) return err ? 0 : pfc_en_tx | pfc_en_rx; } +static bool mlx5e_query_global_pause_combined(struct mlx5e_priv *priv) +{ + struct mlx5_core_dev *mdev = priv->mdev; + u32 rx_pause; + u32 tx_pause; + int err; + + err = mlx5_query_port_pause(mdev, &rx_pause, &tx_pause); + + return err ? false : rx_pause | tx_pause; +} + #define MLX5E_NUM_Q_CNTRS(priv) (NUM_Q_COUNTERS * (!!priv->q_counter)) #define MLX5E_NUM_RQ_STATS(priv) \ (NUM_RQ_STATS * priv->params.num_channels * \ @@ -147,8 +159,8 @@ static unsigned long mlx5e_query_pfc_combined(struct mlx5e_priv *priv) (NUM_SQ_STATS * priv->params.num_channels * priv->params.num_tc * \ test_bit(MLX5E_STATE_OPENED, &priv->state)) #define MLX5E_NUM_PFC_COUNTERS(priv) \ - (hweight8(mlx5e_query_pfc_combined(priv)) * \ - NUM_PPORT_PER_PRIO_PFC_COUNTERS) + ((mlx5e_query_global_pause_combined(priv) + hweight8(mlx5e_query_pfc_combined(priv))) * \ + NUM_PPORT_PER_PRIO_PFC_COUNTERS) static int mlx5e_get_sset_count(struct net_device *dev, int sset) { @@ -210,8 +222,18 @@ static void mlx5e_fill_stats_strings(struct mlx5e_priv *priv, uint8_t *data) pfc_combined = mlx5e_query_pfc_combined(priv); for_each_set_bit(prio, &pfc_combined, NUM_PPORT_PRIO) { for (i = 0; i < NUM_PPORT_PER_PRIO_PFC_COUNTERS; i++) { + char pfc_string[ETH_GSTRING_LEN]; + + snprintf(pfc_string, sizeof(pfc_string), "prio%d", prio); sprintf(data + (idx++) * ETH_GSTRING_LEN, - pport_per_prio_pfc_stats_desc[i].format, prio); + pport_per_prio_pfc_stats_desc[i].format, pfc_string); + } + } + + if (mlx5e_query_global_pause_combined(priv)) { + for (i = 0; i < NUM_PPORT_PER_PRIO_PFC_COUNTERS; i++) { + sprintf(data + (idx++) * ETH_GSTRING_LEN, + pport_per_prio_pfc_stats_desc[i].format, "global"); } } @@ -306,6 +328,13 @@ static void mlx5e_get_ethtool_stats(struct net_device *dev, } } + if (mlx5e_query_global_pause_combined(priv)) { + for (i = 0; i < NUM_PPORT_PER_PRIO_PFC_COUNTERS; i++) { + data[idx++] = MLX5E_READ_CTR64_BE(&priv->stats.pport.per_prio_counters[0], + pport_per_prio_pfc_stats_desc, 0); + } + } + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) return; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h index ae29998..7b9d8a9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h @@ -254,11 +254,12 @@ static const struct counter_desc pport_per_prio_traffic_stats_desc[] = { }; static const struct counter_desc pport_per_prio_pfc_stats_desc[] = { - { "rx_prio%d_pause", PPORT_PER_PRIO_OFF(rx_pause) }, - { "rx_prio%d_pause_duration", PPORT_PER_PRIO_OFF(rx_pause_duration) }, - { "tx_prio%d_pause", PPORT_PER_PRIO_OFF(tx_pause) }, - { "tx_prio%d_pause_duration", PPORT_PER_PRIO_OFF(tx_pause_duration) }, - { "rx_prio%d_pause_transition", PPORT_PER_PRIO_OFF(rx_pause_transition) }, + /* %s is "global" or "prio{i}" */ + { "rx_%s_pause", PPORT_PER_PRIO_OFF(rx_pause) }, + { "rx_%s_pause_duration", PPORT_PER_PRIO_OFF(rx_pause_duration) }, + { "tx_%s_pause", PPORT_PER_PRIO_OFF(tx_pause) }, + { "tx_%s_pause_duration", PPORT_PER_PRIO_OFF(tx_pause_duration) }, + { "rx_%s_pause_transition", PPORT_PER_PRIO_OFF(rx_pause_transition) }, }; struct mlx5e_rq_stats { -- cgit v0.10.2 From 6ebd1a6d74d5fb684ce9a48ee978e94588ecf502 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Mon, 4 Jul 2016 15:24:50 +0000 Subject: dwc_eth_qos: remove unused including Remove including that don't need it. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/synopsys/dwc_eth_qos.c b/drivers/net/ethernet/synopsys/dwc_eth_qos.c index c14fa91..c34111b 100644 --- a/drivers/net/ethernet/synopsys/dwc_eth_qos.c +++ b/drivers/net/ethernet/synopsys/dwc_eth_qos.c @@ -46,7 +46,6 @@ #include #include #include -#include #include #include -- cgit v0.10.2 From 3eb415d1afd0ec22d2cba33877da036f83098105 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 4 Jul 2016 14:36:01 -0700 Subject: net: r6040: Utilize phy_print_status Instead of open coding our own version utilize the library provided function. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 7a7a395..12fc7cd 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -1001,14 +1001,8 @@ static void r6040_adjust_link(struct net_device *dev) lp->old_duplex = phydev->duplex; } - if (status_changed) { - pr_info("%s: link %s", dev->name, phydev->link ? - "UP" : "DOWN"); - if (phydev->link) - pr_cont(" - %d/%s", phydev->speed, - DUPLEX_FULL == phydev->duplex ? "full" : "half"); - pr_cont("\n"); - } + if (status_changed) + phy_print_status(phydev); } static int r6040_mii_probe(struct net_device *dev) -- cgit v0.10.2 From 7def171ddc93c504bd91ccb56b802392770621c5 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 4 Jul 2016 14:36:02 -0700 Subject: net: r6040: Increase statistics upon transmit completion r6040_xmit() is increasing transmit statistics during transmission while this may still fail, do this in r6040_tx() where we complete transmitted buffers instead. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 12fc7cd..75776ee 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -614,6 +614,11 @@ static void r6040_tx(struct net_device *dev) if (descptr->status & DSC_OWNER_MAC) break; /* Not complete */ skb_ptr = descptr->skb_ptr; + + /* Statistic Counter */ + dev->stats.tx_packets++; + dev->stats.tx_bytes += skb_ptr->len; + pci_unmap_single(priv->pdev, le32_to_cpu(descptr->buf), skb_ptr->len, PCI_DMA_TODEVICE); /* Free buffer */ @@ -821,9 +826,6 @@ static netdev_tx_t r6040_start_xmit(struct sk_buff *skb, return NETDEV_TX_BUSY; } - /* Statistic Counter */ - dev->stats.tx_packets++; - dev->stats.tx_bytes += skb->len; /* Set TX descriptor & Transmit it */ lp->tx_free_desc--; descptr = lp->tx_insert_ptr; -- cgit v0.10.2 From a546e557d9404834159b0599a4383ff4cd3321bf Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 4 Jul 2016 14:36:03 -0700 Subject: net: r6040: Utilize skb_put_padto() Pad the SKB to the minimum length of ETH_ZLEN by using skb_put_padto() and take this operation out of the critical section since there is no need to check any HW resources before doing that. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 75776ee..46ed093 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -815,6 +815,9 @@ static netdev_tx_t r6040_start_xmit(struct sk_buff *skb, void __iomem *ioaddr = lp->base; unsigned long flags; + if (skb_put_padto(skb, ETH_ZLEN) < 0) + return NETDEV_TX_OK; + /* Critical Section */ spin_lock_irqsave(&lp->lock, flags); @@ -829,11 +832,7 @@ static netdev_tx_t r6040_start_xmit(struct sk_buff *skb, /* Set TX descriptor & Transmit it */ lp->tx_free_desc--; descptr = lp->tx_insert_ptr; - if (skb->len < ETH_ZLEN) - descptr->len = ETH_ZLEN; - else - descptr->len = skb->len; - + descptr->len = skb->len; descptr->skb_ptr = skb; descptr->buf = cpu_to_le32(pci_map_single(lp->pdev, skb->data, skb->len, PCI_DMA_TODEVICE)); -- cgit v0.10.2 From 58e6b056b0daa8f0287061cff2989e9d5e9a9b7c Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 4 Jul 2016 14:36:04 -0700 Subject: net: r6040: Reclaim transmitted buffers in NAPI Instead of taking one interrupt per packet transmitted, re-use the same NAPI context to free transmitted buffers. Since we are no longer in hard IRQ context replace dev_kfree_skb_irq() by dev_kfree_skb(). Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 46ed093..4bf78f1 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -622,7 +622,7 @@ static void r6040_tx(struct net_device *dev) pci_unmap_single(priv->pdev, le32_to_cpu(descptr->buf), skb_ptr->len, PCI_DMA_TODEVICE); /* Free buffer */ - dev_kfree_skb_irq(skb_ptr); + dev_kfree_skb(skb_ptr); descptr->skb_ptr = NULL; /* To next descriptor */ descptr = descptr->vndescp; @@ -643,12 +643,15 @@ static int r6040_poll(struct napi_struct *napi, int budget) void __iomem *ioaddr = priv->base; int work_done; + r6040_tx(dev); + work_done = r6040_rx(dev, budget); if (work_done < budget) { napi_complete(napi); - /* Enable RX interrupt */ - iowrite16(ioread16(ioaddr + MIER) | RX_INTS, ioaddr + MIER); + /* Enable RX/TX interrupt */ + iowrite16(ioread16(ioaddr + MIER) | RX_INTS | TX_INTS, + ioaddr + MIER); } return work_done; } @@ -675,7 +678,7 @@ static irqreturn_t r6040_interrupt(int irq, void *dev_id) } /* RX interrupt request */ - if (status & RX_INTS) { + if (status & (RX_INTS | TX_INTS)) { if (status & RX_NO_DESC) { /* RX descriptor unavailable */ dev->stats.rx_dropped++; @@ -686,15 +689,11 @@ static irqreturn_t r6040_interrupt(int irq, void *dev_id) if (likely(napi_schedule_prep(&lp->napi))) { /* Mask off RX interrupt */ - misr &= ~RX_INTS; + misr &= ~(RX_INTS | TX_INTS); __napi_schedule(&lp->napi); } } - /* TX interrupt request */ - if (status & TX_INTS) - r6040_tx(dev); - /* Restore RDC MAC interrupt */ iowrite16(misr, ioaddr + MIER); -- cgit v0.10.2 From 9507ffc2acf04c9b1aa02e346225d268c07d79f0 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 4 Jul 2016 14:36:05 -0700 Subject: net: r6040: Check for skb->xmit_more Kick the transmission only if this is the last SKB to transmit or the queue is not already stopped. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 4bf78f1..96b2d63 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -840,7 +840,8 @@ static netdev_tx_t r6040_start_xmit(struct sk_buff *skb, skb_tx_timestamp(skb); /* Trigger the MAC to check the TX descriptor */ - iowrite16(TM2TX, ioaddr + MTPR); + if (!skb->xmit_more || netif_queue_stopped(dev)) + iowrite16(TM2TX, ioaddr + MTPR); lp->tx_insert_ptr = descptr->vndescp; /* If no tx resource, stop */ -- cgit v0.10.2 From ffb5bce0bfd7f196dcfb3263596202cef7b4b92c Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 4 Jul 2016 14:36:06 -0700 Subject: net: r6040: Utilize __napi_schedule_irqoff We are already in hard IRQ context, so we can use __napi_schedule_irqoff() to save a few operations. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 96b2d63..13ff800 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -690,7 +690,7 @@ static irqreturn_t r6040_interrupt(int irq, void *dev_id) if (likely(napi_schedule_prep(&lp->napi))) { /* Mask off RX interrupt */ misr &= ~(RX_INTS | TX_INTS); - __napi_schedule(&lp->napi); + __napi_schedule_irqoff(&lp->napi); } } -- cgit v0.10.2 From 0305efff117ddc6671f5eaf4e4fefd7418116224 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 4 Jul 2016 14:36:07 -0700 Subject: net: r6040: Utilize napi_complete_done() We maintain how much work we did in NAPI context, so provide that with napi_complete_done(). Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index 13ff800..c0256b8 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -648,7 +648,7 @@ static int r6040_poll(struct napi_struct *napi, int budget) work_done = r6040_rx(dev, budget); if (work_done < budget) { - napi_complete(napi); + napi_complete_done(napi, work_done); /* Enable RX/TX interrupt */ iowrite16(ioread16(ioaddr + MIER) | RX_INTS | TX_INTS, ioaddr + MIER); -- cgit v0.10.2 From 35566e9687af6cce72eee8466dd7173780fdd671 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 4 Jul 2016 14:36:08 -0700 Subject: net: r6040: Update my email Update my email address in the driver and MAINTAINERS file. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/MAINTAINERS b/MAINTAINERS index 67e4a65..6374be2 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -9559,7 +9559,7 @@ M: Florian Fainelli S: Maintained RDC R6040 FAST ETHERNET DRIVER -M: Florian Fainelli +M: Florian Fainelli L: netdev@vger.kernel.org S: Maintained F: drivers/net/ethernet/rdc/r6040.c diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index c0256b8..c8985cf 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -4,7 +4,7 @@ * Copyright (C) 2004 Sten Wang * Copyright (C) 2007 * Daniel Gimpelevich - * Copyright (C) 2007-2012 Florian Fainelli + * Copyright (C) 2007-2012 Florian Fainelli * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -162,7 +162,7 @@ MODULE_AUTHOR("Sten Wang ," "Daniel Gimpelevich ," - "Florian Fainelli "); + "Florian Fainelli "); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("RDC R6040 NAPI PCI FastEthernet driver"); MODULE_VERSION(DRV_VERSION " " DRV_RELDATE); -- cgit v0.10.2 From 9da280413a2a7e03b38c880cf3e7e0c16dd7371b Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 4 Jul 2016 14:36:09 -0700 Subject: net: r6040: Bump version and date Bump version to 0.28 and date to 4th of July 2016. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rdc/r6040.c b/drivers/net/ethernet/rdc/r6040.c index c8985cf..cb29ee2 100644 --- a/drivers/net/ethernet/rdc/r6040.c +++ b/drivers/net/ethernet/rdc/r6040.c @@ -48,8 +48,8 @@ #include #define DRV_NAME "r6040" -#define DRV_VERSION "0.28" -#define DRV_RELDATE "07Oct2011" +#define DRV_VERSION "0.29" +#define DRV_RELDATE "04Jul2016" /* Time in jiffies before concluding the transmitter is hung. */ #define TX_TIMEOUT (6000 * HZ / 1000) -- cgit v0.10.2 From c6ac37d8d8843fb1fdc34e4a2a41a4f027ab670c Mon Sep 17 00:00:00 2001 From: Pavel Tikhomirov Date: Fri, 1 Jul 2016 16:53:54 +0300 Subject: netfilter: nf_log: fix error on write NONE to logger choice sysctl It is hard to unbind nf-logger: echo NONE > /proc/sys/net/netfilter/nf_log/0 bash: echo: write error: No such file or directory sysctl -w net.netfilter.nf_log.0=NONE sysctl: setting key "net.netfilter.nf_log.0": No such file or directory net.netfilter.nf_log.0 = NONE You need explicitly send '\0', for instance like: echo -e "NONE\0" > /proc/sys/net/netfilter/nf_log/0 That seem to be strange, so fix it using proc_dostring. Now it works fine: modprobe nfnetlink_log echo nfnetlink_log > /proc/sys/net/netfilter/nf_log/0 cat /proc/sys/net/netfilter/nf_log/0 nfnetlink_log echo NONE > /proc/sys/net/netfilter/nf_log/0 cat /proc/sys/net/netfilter/nf_log/0 NONE v2: add missed error check for proc_dostring Signed-off-by: Pavel Tikhomirov Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index 18e325c..aa5847a 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -418,16 +418,17 @@ static int nf_log_proc_dostring(struct ctl_table *table, int write, { const struct nf_logger *logger; char buf[NFLOGGER_NAME_LEN]; - size_t size = *lenp; int r = 0; int tindex = (unsigned long)table->extra1; struct net *net = current->nsproxy->net_ns; if (write) { - if (size > sizeof(buf)) - size = sizeof(buf); - if (copy_from_user(buf, buffer, size)) - return -EFAULT; + struct ctl_table tmp = *table; + + tmp.data = buf; + r = proc_dostring(&tmp, write, buffer, lenp, ppos); + if (r) + return r; if (!strcmp(buf, "NONE")) { nf_log_unbind_pf(net, tindex); -- cgit v0.10.2 From f152bdad62083f654e22ba4840c5131865de9dca Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Fri, 1 Jul 2016 15:39:42 -0400 Subject: mwifiex: fix unconditional error return in .add_virtual_intf callback The commit 7311ea850079 ("mwifiex: fix AP start problem for newly added interface") attempted to fix an issue when a new AP interface is added. But the patch didn't check the return value of the functions doing the firmware calls and returned an error even if the functions didn't fail. This prevents the network device to be registered properly, so fix it. Fixes: 7311ea850079 ("mwifiex: fix AP start problem for newly added interface") Signed-off-by: Javier Martinez Canillas Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c index 99e8cf1..5de9f63 100644 --- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c +++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c @@ -2865,9 +2865,11 @@ struct wireless_dev *mwifiex_add_virtual_intf(struct wiphy *wiphy, ret = mwifiex_send_cmd(priv, HostCmd_CMD_SET_BSS_MODE, HostCmd_ACT_GEN_SET, 0, NULL, true); + if (ret) return ERR_PTR(ret); ret = mwifiex_sta_init_cmd(priv, false, false); + if (ret) return ERR_PTR(ret); mwifiex_setup_ht_caps(&wiphy->bands[NL80211_BAND_2GHZ]->ht_cap, priv); -- cgit v0.10.2 From 7d54bacadce17fc8042f8b32b0c930a187388ae9 Mon Sep 17 00:00:00 2001 From: Javier Martinez Canillas Date: Mon, 6 Jun 2016 13:02:36 -0400 Subject: mwifiex: add a cfg80211 .get_tx_power operation callback The mwifiex driver implements a cfg80211 .set_tx_power operation handler but doesn't have the inverse .get_tx_power callback. This not only has the effect that the Tx power can't be reported to user space tools such as iwconfig and iwlist but also that the wireless core prints a warning when a new wiphy is created due an cfg80211 operation being implemented without its counterpart. After this patch, the Tx power is properly reported to user-space tools: $ iwlist mlan0 txpower mlan0 unknown transmit-power information. Current Tx-Power=13 dBm (19 mW) and also the following warning isn't shown anymore on the driver probe: WARNING: CPU: 3 PID: 127 at net/wireless/core.c:366 wiphy_new_nm+0x66c/0x6ac Modules linked in: mwifiex_sdio mwifiex CPU: 3 PID: 127 Comm: kworker/3:1 Tainted: G W 4.7.0-rc1-next-20160531-00006-g569df5b983f3 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) Workqueue: events request_firmware_work_func [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [] (show_stack) from [] (dump_stack+0x88/0x9c) [] (dump_stack) from [] (__warn+0xe8/0x100) [] (__warn) from [] (warn_slowpath_null+0x20/0x28) [] (warn_slowpath_null) from [] (wiphy_new_nm+0x66c/0x6ac) [] (wiphy_new_nm) from [] (mwifiex_register_cfg80211+0x28/0x3f0 [mwifiex]) [] (mwifiex_register_cfg80211 [mwifiex]) from [] (mwifiex_fw_dpc+0x2b0/0x474 [mwifiex]) [] (mwifiex_fw_dpc [mwifiex]) from [] (request_firmware_work_func+0x30/0x58) [] (request_firmware_work_func) from [] (process_one_work+0x124/0x338) [] (process_one_work) from [] (worker_thread+0x38/0x4d4) [] (worker_thread) from [] (kthread+0xdc/0xf4) [] (kthread) from [] (ret_from_fork+0x14/0x3c) Signed-off-by: Javier Martinez Canillas Tested-by: Enric Balletbo i Serra Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c index 5de9f63..3044b42 100644 --- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c +++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c @@ -377,6 +377,29 @@ mwifiex_cfg80211_set_tx_power(struct wiphy *wiphy, } /* + * CFG802.11 operation handler to get Tx power. + */ +static int +mwifiex_cfg80211_get_tx_power(struct wiphy *wiphy, + struct wireless_dev *wdev, + int *dbm) +{ + struct mwifiex_adapter *adapter = mwifiex_cfg80211_get_adapter(wiphy); + struct mwifiex_private *priv = mwifiex_get_priv(adapter, + MWIFIEX_BSS_ROLE_ANY); + int ret = mwifiex_send_cmd(priv, HostCmd_CMD_RF_TX_PWR, + HostCmd_ACT_GEN_GET, 0, NULL, true); + + if (ret < 0) + return ret; + + /* tx_power_level is set in HostCmd_CMD_RF_TX_PWR command handler */ + *dbm = priv->tx_power_level; + + return 0; +} + +/* * CFG802.11 operation handler to set Power Save option. * * The timeout value, if provided, is currently ignored. @@ -3953,6 +3976,7 @@ static struct cfg80211_ops mwifiex_cfg80211_ops = { .set_default_key = mwifiex_cfg80211_set_default_key, .set_power_mgmt = mwifiex_cfg80211_set_power_mgmt, .set_tx_power = mwifiex_cfg80211_set_tx_power, + .get_tx_power = mwifiex_cfg80211_get_tx_power, .set_bitrate_mask = mwifiex_cfg80211_set_bitrate_mask, .start_ap = mwifiex_cfg80211_start_ap, .stop_ap = mwifiex_cfg80211_stop_ap, -- cgit v0.10.2 From 3ee712857958c27f697b33fa520fdc51e7ffea88 Mon Sep 17 00:00:00 2001 From: Shengzhen Li Date: Mon, 6 Jun 2016 13:02:38 -0400 Subject: mwifiex: add get_antenna support for cfg80211 Since commit de3bb771f471 ("cfg80211: add more warnings for inconsistent ops") the wireless core warns if a driver implements a cfg80211 callback but doesn't implements the inverse operation. The mwifiex driver defines a .set_antenna handler but not a .get_antenna so this not only makes the core to print a warning when creating a new wiphy but also the antenna isn't reported to user-space apps such as iw. This patch queries the antenna to the firmware so is properly reported to user-space. With this patch, the wireless core does not warn anymore and: $ iw phy phy0 info | grep Antennas Available Antennas: TX 0x3 RX 0x3 Configured Antennas: TX 0x3 RX 0x3 Signed-off-by: Shengzhen Li Signed-off-by: Amitkumar Karwar [javier: expand the commit message] Signed-off-by: Javier Martinez Canillas Tested-by: Enric Balletbo i Serra Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c index 3044b42..df5ebdf 100644 --- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c +++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c @@ -1830,6 +1830,21 @@ mwifiex_cfg80211_set_antenna(struct wiphy *wiphy, u32 tx_ant, u32 rx_ant) HostCmd_ACT_GEN_SET, 0, &ant_cfg, true); } +static int +mwifiex_cfg80211_get_antenna(struct wiphy *wiphy, u32 *tx_ant, u32 *rx_ant) +{ + struct mwifiex_adapter *adapter = mwifiex_cfg80211_get_adapter(wiphy); + struct mwifiex_private *priv = mwifiex_get_priv(adapter, + MWIFIEX_BSS_ROLE_ANY); + mwifiex_send_cmd(priv, HostCmd_CMD_RF_ANTENNA, + HostCmd_ACT_GEN_GET, 0, NULL, true); + + *tx_ant = priv->tx_ant; + *rx_ant = priv->rx_ant; + + return 0; +} + /* cfg80211 operation handler for stop ap. * Function stops BSS running at uAP interface. */ @@ -3983,6 +3998,7 @@ static struct cfg80211_ops mwifiex_cfg80211_ops = { .change_beacon = mwifiex_cfg80211_change_beacon, .set_cqm_rssi_config = mwifiex_cfg80211_set_cqm_rssi_config, .set_antenna = mwifiex_cfg80211_set_antenna, + .get_antenna = mwifiex_cfg80211_get_antenna, .del_station = mwifiex_cfg80211_del_station, .sched_scan_start = mwifiex_cfg80211_sched_scan_start, .sched_scan_stop = mwifiex_cfg80211_sched_scan_stop, diff --git a/drivers/net/wireless/marvell/mwifiex/fw.h b/drivers/net/wireless/marvell/mwifiex/fw.h index 8e4145a..cef7234 100644 --- a/drivers/net/wireless/marvell/mwifiex/fw.h +++ b/drivers/net/wireless/marvell/mwifiex/fw.h @@ -462,6 +462,9 @@ enum P2P_MODES { #define HostCmd_ACT_SET_RX 0x0001 #define HostCmd_ACT_SET_TX 0x0002 #define HostCmd_ACT_SET_BOTH 0x0003 +#define HostCmd_ACT_GET_RX 0x0004 +#define HostCmd_ACT_GET_TX 0x0008 +#define HostCmd_ACT_GET_BOTH 0x000c #define RF_ANTENNA_AUTO 0xFFFF diff --git a/drivers/net/wireless/marvell/mwifiex/init.c b/drivers/net/wireless/marvell/mwifiex/init.c index 07eac5b..1489c90 100644 --- a/drivers/net/wireless/marvell/mwifiex/init.c +++ b/drivers/net/wireless/marvell/mwifiex/init.c @@ -110,6 +110,8 @@ int mwifiex_init_priv(struct mwifiex_private *priv) priv->tx_power_level = 0; priv->max_tx_power_level = 0; priv->min_tx_power_level = 0; + priv->tx_ant = 0; + priv->rx_ant = 0; priv->tx_rate = 0; priv->rxpd_htinfo = 0; priv->rxpd_rate = 0; diff --git a/drivers/net/wireless/marvell/mwifiex/main.h b/drivers/net/wireless/marvell/mwifiex/main.h index 0d38a72..9f6bb40 100644 --- a/drivers/net/wireless/marvell/mwifiex/main.h +++ b/drivers/net/wireless/marvell/mwifiex/main.h @@ -533,6 +533,8 @@ struct mwifiex_private { u16 tx_power_level; u8 max_tx_power_level; u8 min_tx_power_level; + u32 tx_ant; + u32 rx_ant; u8 tx_rate; u8 tx_htinfo; u8 rxpd_htinfo; diff --git a/drivers/net/wireless/marvell/mwifiex/sta_cmd.c b/drivers/net/wireless/marvell/mwifiex/sta_cmd.c index e436574..8c65849 100644 --- a/drivers/net/wireless/marvell/mwifiex/sta_cmd.c +++ b/drivers/net/wireless/marvell/mwifiex/sta_cmd.c @@ -313,23 +313,41 @@ static int mwifiex_cmd_rf_antenna(struct mwifiex_private *priv, cmd->command = cpu_to_le16(HostCmd_CMD_RF_ANTENNA); - if (cmd_action != HostCmd_ACT_GEN_SET) - return 0; - - if (priv->adapter->hw_dev_mcs_support == HT_STREAM_2X2) { - cmd->size = cpu_to_le16(sizeof(struct host_cmd_ds_rf_ant_mimo) + - S_DS_GEN); - ant_mimo->action_tx = cpu_to_le16(HostCmd_ACT_SET_TX); - ant_mimo->tx_ant_mode = cpu_to_le16((u16)ant_cfg->tx_ant); - ant_mimo->action_rx = cpu_to_le16(HostCmd_ACT_SET_RX); - ant_mimo->rx_ant_mode = cpu_to_le16((u16)ant_cfg->rx_ant); - } else { - cmd->size = cpu_to_le16(sizeof(struct host_cmd_ds_rf_ant_siso) + - S_DS_GEN); - ant_siso->action = cpu_to_le16(HostCmd_ACT_SET_BOTH); - ant_siso->ant_mode = cpu_to_le16((u16)ant_cfg->tx_ant); + switch (cmd_action) { + case HostCmd_ACT_GEN_SET: + if (priv->adapter->hw_dev_mcs_support == HT_STREAM_2X2) { + cmd->size = cpu_to_le16(sizeof(struct + host_cmd_ds_rf_ant_mimo) + + S_DS_GEN); + ant_mimo->action_tx = cpu_to_le16(HostCmd_ACT_SET_TX); + ant_mimo->tx_ant_mode = cpu_to_le16((u16)ant_cfg-> + tx_ant); + ant_mimo->action_rx = cpu_to_le16(HostCmd_ACT_SET_RX); + ant_mimo->rx_ant_mode = cpu_to_le16((u16)ant_cfg-> + rx_ant); + } else { + cmd->size = cpu_to_le16(sizeof(struct + host_cmd_ds_rf_ant_siso) + + S_DS_GEN); + ant_siso->action = cpu_to_le16(HostCmd_ACT_SET_BOTH); + ant_siso->ant_mode = cpu_to_le16((u16)ant_cfg->tx_ant); + } + break; + case HostCmd_ACT_GEN_GET: + if (priv->adapter->hw_dev_mcs_support == HT_STREAM_2X2) { + cmd->size = cpu_to_le16(sizeof(struct + host_cmd_ds_rf_ant_mimo) + + S_DS_GEN); + ant_mimo->action_tx = cpu_to_le16(HostCmd_ACT_GET_TX); + ant_mimo->action_rx = cpu_to_le16(HostCmd_ACT_GET_RX); + } else { + cmd->size = cpu_to_le16(sizeof(struct + host_cmd_ds_rf_ant_siso) + + S_DS_GEN); + ant_siso->action = cpu_to_le16(HostCmd_ACT_GET_BOTH); + } + break; } - return 0; } diff --git a/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c b/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c index bcfd4b7..9050d06 100644 --- a/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c +++ b/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c @@ -469,7 +469,9 @@ static int mwifiex_ret_rf_antenna(struct mwifiex_private *priv, struct host_cmd_ds_rf_ant_siso *ant_siso = &resp->params.ant_siso; struct mwifiex_adapter *adapter = priv->adapter; - if (adapter->hw_dev_mcs_support == HT_STREAM_2X2) + if (adapter->hw_dev_mcs_support == HT_STREAM_2X2) { + priv->tx_ant = le16_to_cpu(ant_mimo->tx_ant_mode); + priv->rx_ant = le16_to_cpu(ant_mimo->rx_ant_mode); mwifiex_dbg(adapter, INFO, "RF_ANT_RESP: Tx action = 0x%x, Tx Mode = 0x%04x\t" "Rx action = 0x%x, Rx Mode = 0x%04x\n", @@ -477,12 +479,14 @@ static int mwifiex_ret_rf_antenna(struct mwifiex_private *priv, le16_to_cpu(ant_mimo->tx_ant_mode), le16_to_cpu(ant_mimo->action_rx), le16_to_cpu(ant_mimo->rx_ant_mode)); - else + } else { + priv->tx_ant = le16_to_cpu(ant_siso->ant_mode); + priv->rx_ant = le16_to_cpu(ant_siso->ant_mode); mwifiex_dbg(adapter, INFO, "RF_ANT_RESP: action = 0x%x, Mode = 0x%04x\n", le16_to_cpu(ant_siso->action), le16_to_cpu(ant_siso->ant_mode)); - + } return 0; } -- cgit v0.10.2 From bf942091678c51f53caa679d5fabebbe04a2ea66 Mon Sep 17 00:00:00 2001 From: Shengzhen Li Date: Tue, 21 Jun 2016 14:32:32 +0530 Subject: mwifiex: disable MSIx interrupt for 8997 chipset Sometimes MSIx interrupts are received out of order on multi-core system. This creates a problem when there is a race between data packet and SLEEP event from firmware. We will disable MSIx interrupt mode to solve the problem and go with MSI mode. Signed-off-by: Shengzhen Li Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.h b/drivers/net/wireless/marvell/mwifiex/pcie.h index 2592e63..9c00c7e 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.h +++ b/drivers/net/wireless/marvell/mwifiex/pcie.h @@ -258,7 +258,7 @@ static const struct mwifiex_pcie_card_reg mwifiex_reg_8997 = { .fw_dump_end = 0xcff, .fw_dump_host_ready = 0xcc, .fw_dump_read_done = 0xdd, - .msix_support = 1, + .msix_support = 0, }; static struct memory_type_mapping mem_type_mapping_tbl_w8897[] = { -- cgit v0.10.2 From 30462b514fb335a03ce8de89e37191bfd7ce8887 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Sat, 25 Jun 2016 13:37:42 -0500 Subject: rtlwifi: Remove unused parameter from rtl_ps_set_rf_state() Commit 4b9d8d67b44a ("rtlwifi: rtl8192cu: Remove unused parameter") reworked this routine. Those changes were later reverted by commit d3feae41a347 ("rtlwifi: Update power-save routines for 062814 driver"). There were two changes in commit 4b9d8d67b44a. The first of these removed a parameter from rtl_ps_set_rf_state() that was always false. This is the change that is restored in the current patch. A second change that reworked the locking is still being analyzed. In addition to removing the unused parameter, there is no need for rtl_ps_set_rf_state() to be exported. Reported-by: Pavel Andrianov Signed-off-by: Larry Finger Cc: Pavel Andrianov Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/ps.c b/drivers/net/wireless/realtek/rtlwifi/ps.c index 93579ca..9a64f9b 100644 --- a/drivers/net/wireless/realtek/rtlwifi/ps.c +++ b/drivers/net/wireless/realtek/rtlwifi/ps.c @@ -76,9 +76,9 @@ bool rtl_ps_disable_nic(struct ieee80211_hw *hw) } EXPORT_SYMBOL(rtl_ps_disable_nic); -bool rtl_ps_set_rf_state(struct ieee80211_hw *hw, - enum rf_pwrstate state_toset, - u32 changesource, bool protect_or_not) +static bool rtl_ps_set_rf_state(struct ieee80211_hw *hw, + enum rf_pwrstate state_toset, + u32 changesource) { struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_ps_ctl *ppsc = rtl_psc(rtl_priv(hw)); @@ -86,9 +86,6 @@ bool rtl_ps_set_rf_state(struct ieee80211_hw *hw, bool actionallowed = false; u16 rfwait_cnt = 0; - if (protect_or_not) - goto no_protect; - /*Only one thread can change *the RF state at one time, and others *should wait to be executed. @@ -119,7 +116,6 @@ bool rtl_ps_set_rf_state(struct ieee80211_hw *hw, } } -no_protect: rtstate = ppsc->rfpwr_state; switch (state_toset) { @@ -162,15 +158,12 @@ no_protect: if (actionallowed) rtlpriv->cfg->ops->set_rf_power_state(hw, state_toset); - if (!protect_or_not) { - spin_lock(&rtlpriv->locks.rf_ps_lock); - ppsc->rfchange_inprogress = false; - spin_unlock(&rtlpriv->locks.rf_ps_lock); - } + spin_lock(&rtlpriv->locks.rf_ps_lock); + ppsc->rfchange_inprogress = false; + spin_unlock(&rtlpriv->locks.rf_ps_lock); return actionallowed; } -EXPORT_SYMBOL(rtl_ps_set_rf_state); static void _rtl_ps_inactive_ps(struct ieee80211_hw *hw) { @@ -191,7 +184,7 @@ static void _rtl_ps_inactive_ps(struct ieee80211_hw *hw) } rtl_ps_set_rf_state(hw, ppsc->inactive_pwrstate, - RF_CHANGE_BY_IPS, false); + RF_CHANGE_BY_IPS); if (ppsc->inactive_pwrstate == ERFOFF && rtlhal->interface == INTF_PCI) { @@ -587,7 +580,7 @@ void rtl_swlps_rf_awake(struct ieee80211_hw *hw) } spin_lock_irqsave(&rtlpriv->locks.lps_lock, flag); - rtl_ps_set_rf_state(hw, ERFON, RF_CHANGE_BY_PS, false); + rtl_ps_set_rf_state(hw, ERFON, RF_CHANGE_BY_PS); spin_unlock_irqrestore(&rtlpriv->locks.lps_lock, flag); } @@ -630,7 +623,7 @@ void rtl_swlps_rf_sleep(struct ieee80211_hw *hw) spin_unlock(&rtlpriv->locks.rf_ps_lock); spin_lock_irqsave(&rtlpriv->locks.lps_lock, flag); - rtl_ps_set_rf_state(hw, ERFSLEEP, RF_CHANGE_BY_PS , false); + rtl_ps_set_rf_state(hw, ERFSLEEP, RF_CHANGE_BY_PS); spin_unlock_irqrestore(&rtlpriv->locks.lps_lock, flag); if (ppsc->reg_rfps_level & RT_RF_OFF_LEVL_ASPM && diff --git a/drivers/net/wireless/realtek/rtlwifi/ps.h b/drivers/net/wireless/realtek/rtlwifi/ps.h index 29dfc51..0df2b52 100644 --- a/drivers/net/wireless/realtek/rtlwifi/ps.h +++ b/drivers/net/wireless/realtek/rtlwifi/ps.h @@ -28,9 +28,6 @@ #define MAX_SW_LPS_SLEEP_INTV 5 -bool rtl_ps_set_rf_state(struct ieee80211_hw *hw, - enum rf_pwrstate state_toset, u32 changesource, - bool protect_or_not); bool rtl_ps_enable_nic(struct ieee80211_hw *hw); bool rtl_ps_disable_nic(struct ieee80211_hw *hw); void rtl_ips_nic_off(struct ieee80211_hw *hw); -- cgit v0.10.2 From 204e2ab22e1e2d0eb7b4cb929bc4b417226474b8 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Sat, 25 Jun 2016 13:37:43 -0500 Subject: rtlwifi: rtl8188ee: Fix potential race condition Flag rfchange_inprogress in struct rtl_ps_ctl is protected by a spinlock in most routines but not in rtl88e_dm_watchdog(), which could lead to a race condition. The necessary locking to prevent this condition is added. Reported-by: Pavel Andrianov Signed-off-by: Larry Finger Cc: Pavel Andrianov Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/dm.c index 6e7b673..f936a49 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/dm.c @@ -1790,6 +1790,7 @@ void rtl88e_dm_watchdog(struct ieee80211_hw *hw) if (ppsc->p2p_ps_info.p2p_ps_mode) fw_ps_awake = false; + spin_lock(&rtlpriv->locks.rf_ps_lock); if ((ppsc->rfpwr_state == ERFON) && ((!fw_current_inpsmode) && fw_ps_awake) && (!ppsc->rfchange_inprogress)) { @@ -1802,4 +1803,5 @@ void rtl88e_dm_watchdog(struct ieee80211_hw *hw) rtl88e_dm_check_edca_turbo(hw); rtl88e_dm_antenna_diversity(hw); } + spin_unlock(&rtlpriv->locks.rf_ps_lock); } -- cgit v0.10.2 From c3ae8ec4a26468f4b1a0b329ca9f57d19be32be8 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Sat, 25 Jun 2016 13:37:44 -0500 Subject: rtlwifi: rtl8192ee: Fix potential race condition Flag rfchange_inprogress in struct rtl_ps_ctl is protected by a spinlock in most routines but not in rtl92ee_dm_watchdog(), which could lead to a race condition. The necessary locking to prevent this condition is added. Reported-by: Pavel Andrianov Signed-off-by: Larry Finger Cc: Pavel Andrianov Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c index 459f3d0..46efba0 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c @@ -1219,6 +1219,7 @@ void rtl92ee_dm_watchdog(struct ieee80211_hw *hw) if (ppsc->p2p_ps_info.p2p_ps_mode) fw_ps_awake = false; + spin_lock(&rtlpriv->locks.rf_ps_lock); if ((ppsc->rfpwr_state == ERFON) && ((!fw_current_inpsmode) && fw_ps_awake) && (!ppsc->rfchange_inprogress)) { @@ -1233,4 +1234,5 @@ void rtl92ee_dm_watchdog(struct ieee80211_hw *hw) rtl92ee_dm_dynamic_atc_switch(hw); rtl92ee_dm_dynamic_primary_cca_ckeck(hw); } + spin_unlock(&rtlpriv->locks.rf_ps_lock); } -- cgit v0.10.2 From 31c2e76c77fba56bd52d704802dd63aafabd5320 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Sat, 25 Jun 2016 13:37:45 -0500 Subject: rtlwifi: rtl8723be: Fix potential race condition Flag rfchange_inprogress in struct rtl_ps_ctl is protected by a spinlock in most routines but not in rtl8723be_dm_watchdog(), which could lead to a race condition. The necessary locking to prevent this condition is added. Reported-by: Pavel Andrianov Signed-off-by: Larry Finger Cc: Pavel Andrianov Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/dm.c index 6f07cd6..131c0d1 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/dm.c @@ -1279,6 +1279,7 @@ void rtl8723be_dm_watchdog(struct ieee80211_hw *hw) if (ppsc->p2p_ps_info.p2p_ps_mode) fw_ps_awake = false; + spin_lock(&rtlpriv->locks.rf_ps_lock); if ((ppsc->rfpwr_state == ERFON) && ((!fw_current_inpsmode) && fw_ps_awake) && (!ppsc->rfchange_inprogress)) { @@ -1294,5 +1295,6 @@ void rtl8723be_dm_watchdog(struct ieee80211_hw *hw) rtl8723be_dm_check_txpower_tracking(hw); rtl8723be_dm_dynamic_txpower(hw); } + spin_unlock(&rtlpriv->locks.rf_ps_lock); rtlpriv->dm.dbginfo.num_qry_beacon_pkt = 0; } -- cgit v0.10.2 From 4f29b348bdc08e6a30242f0d14d6672019ffd1bb Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Sat, 25 Jun 2016 13:37:46 -0500 Subject: rtlwifi: rtl8723ae: Fix potential race condition Flag rfchange_inprogress in struct rtl_ps_ctl is protected by a spinlock in most routines but not in rtl8723e_dm_watchdog(), which could lead to a race condition. The necessary locking to prevent this condition is added. Reported-by: Pavel Andrianov Signed-off-by: Larry Finger Cc: Pavel Andrianov Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c index 4c1c96c..3900e10 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c @@ -816,6 +816,7 @@ void rtl8723e_dm_watchdog(struct ieee80211_hw *hw) if (ppsc->p2p_ps_info.p2p_ps_mode) fw_ps_awake = false; + spin_lock(&rtlpriv->locks.rf_ps_lock); if ((ppsc->rfpwr_state == ERFON) && ((!fw_current_inpsmode) && fw_ps_awake) && (!ppsc->rfchange_inprogress)) { @@ -829,6 +830,7 @@ void rtl8723e_dm_watchdog(struct ieee80211_hw *hw) rtl8723e_dm_bt_coexist(hw); rtl8723e_dm_check_edca_turbo(hw); } + spin_unlock(&rtlpriv->locks.rf_ps_lock); if (rtlpriv->btcoexist.init_set) rtl_write_byte(rtlpriv, 0x76e, 0xc); } -- cgit v0.10.2 From 300c32ca84f5006ccbbf37f5207b6311aeccc5c8 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Sat, 25 Jun 2016 13:37:47 -0500 Subject: rtlwifi: rtl8821ae: Fix potential race condition Flag rfchange_inprogress in struct rtl_ps_ctl is protected by a spinlock in most routines but not in rtl8821ae_dm_watchdog() which could lead to a race condition. The necessary locking to prevent this condition is added. Reported-by: Pavel Andrianov Signed-off-by: Larry Finger Cc: Pavel Andrianov Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c index 69de835..03e08cb 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c @@ -2949,6 +2949,7 @@ void rtl8821ae_dm_watchdog(struct ieee80211_hw *hw) if (ppsc->p2p_ps_info.p2p_ps_mode) fw_ps_awake = false; + spin_lock(&rtlpriv->locks.rf_ps_lock); if ((ppsc->rfpwr_state == ERFON) && ((!fw_current_inpsmode) && fw_ps_awake) && (!ppsc->rfchange_inprogress)) { @@ -2967,6 +2968,7 @@ void rtl8821ae_dm_watchdog(struct ieee80211_hw *hw) rtl8821ae_dm_check_txpower_tracking_thermalmeter(hw); rtl8821ae_dm_iq_calibrate(hw); } + spin_unlock(&rtlpriv->locks.rf_ps_lock); rtlpriv->dm.dbginfo.num_qry_beacon_pkt = 0; RT_TRACE(rtlpriv, COMP_DIG, DBG_DMESG, "\n"); -- cgit v0.10.2 From 503eebc265dcf5c512454fd5a6b6673ea4f1d7f2 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 Jul 2016 11:27:37 +0200 Subject: net: add dev arg to ndo_neigh_construct/destroy As the following patch will allow upper devices to follow the call down lower devices, we need to add dev here and not rely on n->dev. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c index 28b775e..f0b09b0 100644 --- a/drivers/net/ethernet/rocker/rocker_main.c +++ b/drivers/net/ethernet/rocker/rocker_main.c @@ -1996,7 +1996,8 @@ static int rocker_port_change_proto_down(struct net_device *dev, return 0; } -static void rocker_port_neigh_destroy(struct neighbour *n) +static void rocker_port_neigh_destroy(struct net_device *dev, + struct neighbour *n) { struct rocker_port *rocker_port = netdev_priv(n->dev); int err; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 0c6ee2c..91af73c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1209,8 +1209,10 @@ struct net_device_ops { netdev_features_t features); int (*ndo_set_features)(struct net_device *dev, netdev_features_t features); - int (*ndo_neigh_construct)(struct neighbour *n); - void (*ndo_neigh_destroy)(struct neighbour *n); + int (*ndo_neigh_construct)(struct net_device *dev, + struct neighbour *n); + void (*ndo_neigh_destroy)(struct net_device *dev, + struct neighbour *n); int (*ndo_fdb_add)(struct ndmsg *ndm, struct nlattr *tb[], diff --git a/net/atm/clip.c b/net/atm/clip.c index e07f551..53b4ac0 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -286,7 +286,7 @@ static const struct neigh_ops clip_neigh_ops = { .connected_output = neigh_direct_output, }; -static int clip_constructor(struct neighbour *neigh) +static int clip_constructor(struct net_device *dev, struct neighbour *neigh) { struct atmarp_entry *entry = neighbour_priv(neigh); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 510cd62..952aabb 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -473,7 +473,7 @@ struct neighbour *__neigh_create(struct neigh_table *tbl, const void *pkey, } if (dev->netdev_ops->ndo_neigh_construct) { - error = dev->netdev_ops->ndo_neigh_construct(n); + error = dev->netdev_ops->ndo_neigh_construct(dev, n); if (error < 0) { rc = ERR_PTR(error); goto out_neigh_release; @@ -701,7 +701,7 @@ void neigh_destroy(struct neighbour *neigh) neigh->arp_queue_len_bytes = 0; if (dev->netdev_ops->ndo_neigh_destroy) - dev->netdev_ops->ndo_neigh_destroy(neigh); + dev->netdev_ops->ndo_neigh_destroy(dev, neigh); dev_put(dev); neigh_parms_put(neigh->parms); diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index 8c004a0..935ab93 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -81,7 +81,7 @@ static int lowpan_stop(struct net_device *dev) return 0; } -static int lowpan_neigh_construct(struct neighbour *n) +static int lowpan_neigh_construct(struct net_device *dev, struct neighbour *n) { struct lowpan_802154_neigh *neigh = lowpan_802154_neigh(neighbour_priv(n)); -- cgit v0.10.2 From 18bfb924f0005a728caadd90ba755b2a660bf441 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 Jul 2016 11:27:38 +0200 Subject: net: introduce default neigh_construct/destroy ndo calls for L2 upper devices L2 upper device needs to propagate neigh_construct/destroy calls down to lower devices. Do this by defining default ndo functions and use them in team, bond, bridge and vlan. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 90157e2..480d73a 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4137,6 +4137,8 @@ static const struct net_device_ops bond_netdev_ops = { .ndo_add_slave = bond_enslave, .ndo_del_slave = bond_release, .ndo_fix_features = bond_fix_features, + .ndo_neigh_construct = netdev_default_l2upper_neigh_construct, + .ndo_neigh_destroy = netdev_default_l2upper_neigh_destroy, .ndo_bridge_setlink = switchdev_port_bridge_setlink, .ndo_bridge_getlink = switchdev_port_bridge_getlink, .ndo_bridge_dellink = switchdev_port_bridge_dellink, diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index f9eebea..a380649 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -2002,6 +2002,8 @@ static const struct net_device_ops team_netdev_ops = { .ndo_add_slave = team_add_slave, .ndo_del_slave = team_del_slave, .ndo_fix_features = team_fix_features, + .ndo_neigh_construct = netdev_default_l2upper_neigh_construct, + .ndo_neigh_destroy = netdev_default_l2upper_neigh_destroy, .ndo_change_carrier = team_change_carrier, .ndo_bridge_setlink = switchdev_port_bridge_setlink, .ndo_bridge_getlink = switchdev_port_bridge_getlink, diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 91af73c..49736a3 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3845,6 +3845,10 @@ void *netdev_lower_dev_get_private(struct net_device *dev, struct net_device *lower_dev); void netdev_lower_state_changed(struct net_device *lower_dev, void *lower_state_info); +int netdev_default_l2upper_neigh_construct(struct net_device *dev, + struct neighbour *n); +void netdev_default_l2upper_neigh_destroy(struct net_device *dev, + struct neighbour *n); /* RSS keys are 40 or 52 bytes long */ #define NETDEV_RSS_KEY_LEN 52 diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 86ae75b..c8f422c 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -790,6 +790,8 @@ static const struct net_device_ops vlan_netdev_ops = { .ndo_netpoll_cleanup = vlan_dev_netpoll_cleanup, #endif .ndo_fix_features = vlan_dev_fix_features, + .ndo_neigh_construct = netdev_default_l2upper_neigh_construct, + .ndo_neigh_destroy = netdev_default_l2upper_neigh_destroy, .ndo_fdb_add = switchdev_port_fdb_add, .ndo_fdb_del = switchdev_port_fdb_del, .ndo_fdb_dump = switchdev_port_fdb_dump, diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 0c39e0f..8eecd0e 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -349,6 +349,8 @@ static const struct net_device_ops br_netdev_ops = { .ndo_add_slave = br_add_slave, .ndo_del_slave = br_del_slave, .ndo_fix_features = br_fix_features, + .ndo_neigh_construct = netdev_default_l2upper_neigh_construct, + .ndo_neigh_destroy = netdev_default_l2upper_neigh_destroy, .ndo_fdb_add = br_fdb_add, .ndo_fdb_del = br_fdb_delete, .ndo_fdb_dump = br_fdb_dump, diff --git a/net/core/dev.c b/net/core/dev.c index a4f3b0a..b92d63b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6087,6 +6087,50 @@ void netdev_lower_state_changed(struct net_device *lower_dev, } EXPORT_SYMBOL(netdev_lower_state_changed); +int netdev_default_l2upper_neigh_construct(struct net_device *dev, + struct neighbour *n) +{ + struct net_device *lower_dev, *stop_dev; + struct list_head *iter; + int err; + + netdev_for_each_lower_dev(dev, lower_dev, iter) { + if (!lower_dev->netdev_ops->ndo_neigh_construct) + continue; + err = lower_dev->netdev_ops->ndo_neigh_construct(lower_dev, n); + if (err) { + stop_dev = lower_dev; + goto rollback; + } + } + return 0; + +rollback: + netdev_for_each_lower_dev(dev, lower_dev, iter) { + if (lower_dev == stop_dev) + break; + if (!lower_dev->netdev_ops->ndo_neigh_destroy) + continue; + lower_dev->netdev_ops->ndo_neigh_destroy(lower_dev, n); + } + return err; +} +EXPORT_SYMBOL_GPL(netdev_default_l2upper_neigh_construct); + +void netdev_default_l2upper_neigh_destroy(struct net_device *dev, + struct neighbour *n) +{ + struct net_device *lower_dev; + struct list_head *iter; + + netdev_for_each_lower_dev(dev, lower_dev, iter) { + if (!lower_dev->netdev_ops->ndo_neigh_destroy) + continue; + lower_dev->netdev_ops->ndo_neigh_destroy(lower_dev, n); + } +} +EXPORT_SYMBOL_GPL(netdev_default_l2upper_neigh_destroy); + static void dev_change_rx_flags(struct net_device *dev, int flags) { const struct net_device_ops *ops = dev->netdev_ops; -- cgit v0.10.2 From 6cf3c971dc84cb36579515ddb488919b9e9fb6de Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 Jul 2016 11:27:39 +0200 Subject: mlxsw: spectrum_router: Add private neigh table We need to hold some private data for every neigh entry. It would be possible to do it using neigh_priv_len/ndo_neigh_construct/ ndo_neigh_destroy however only for the port device itself. That would not work for stacked devices like bridge/team/bond. So introduce a private neigh table. Hook onto ndos neigh_construct/destroy and add/remove table entry according to that. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 7b2b741b..9bebb7a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -803,6 +803,8 @@ static const struct net_device_ops mlxsw_sp_port_netdev_ops = { .ndo_get_stats64 = mlxsw_sp_port_get_stats64, .ndo_vlan_rx_add_vid = mlxsw_sp_port_add_vid, .ndo_vlan_rx_kill_vid = mlxsw_sp_port_kill_vid, + .ndo_neigh_construct = mlxsw_sp_router_neigh_construct, + .ndo_neigh_destroy = mlxsw_sp_router_neigh_destroy, .ndo_fdb_add = switchdev_port_fdb_add, .ndo_fdb_del = switchdev_port_fdb_del, .ndo_fdb_dump = switchdev_port_fdb_dump, diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 958e821..734c5ba 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -39,6 +39,7 @@ #include #include +#include #include #include #include @@ -212,6 +213,7 @@ struct mlxsw_sp_vr { struct mlxsw_sp_router { struct mlxsw_sp_lpm_tree lpm_trees[MLXSW_SP_LPM_TREE_COUNT]; struct mlxsw_sp_vr vrs[MLXSW_SP_VIRTUAL_ROUTER_MAX]; + struct rhashtable neigh_ht; }; struct mlxsw_sp { @@ -524,5 +526,9 @@ int mlxsw_sp_router_fib4_add(struct mlxsw_sp_port *mlxsw_sp_port, struct switchdev_trans *trans); int mlxsw_sp_router_fib4_del(struct mlxsw_sp_port *mlxsw_sp_port, const struct switchdev_obj_ipv4_fib *fib4); +int mlxsw_sp_router_neigh_construct(struct net_device *dev, + struct neighbour *n); +void mlxsw_sp_router_neigh_destroy(struct net_device *dev, + struct neighbour *n); #endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 7e3992a..90d382a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -38,6 +38,8 @@ #include #include #include +#include +#include #include "spectrum.h" #include "core.h" @@ -544,6 +546,147 @@ static void mlxsw_sp_vrs_init(struct mlxsw_sp *mlxsw_sp) } } +struct mlxsw_sp_neigh_key { + unsigned char addr[sizeof(struct in6_addr)]; + struct net_device *dev; +}; + +struct mlxsw_sp_neigh_entry { + struct rhash_head ht_node; + struct mlxsw_sp_neigh_key key; + u16 rif; + struct neighbour *n; +}; + +static const struct rhashtable_params mlxsw_sp_neigh_ht_params = { + .key_offset = offsetof(struct mlxsw_sp_neigh_entry, key), + .head_offset = offsetof(struct mlxsw_sp_neigh_entry, ht_node), + .key_len = sizeof(struct mlxsw_sp_neigh_key), +}; + +static int +mlxsw_sp_neigh_entry_insert(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_neigh_entry *neigh_entry) +{ + return rhashtable_insert_fast(&mlxsw_sp->router.neigh_ht, + &neigh_entry->ht_node, + mlxsw_sp_neigh_ht_params); +} + +static void +mlxsw_sp_neigh_entry_remove(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_neigh_entry *neigh_entry) +{ + rhashtable_remove_fast(&mlxsw_sp->router.neigh_ht, + &neigh_entry->ht_node, + mlxsw_sp_neigh_ht_params); +} + +static struct mlxsw_sp_neigh_entry * +mlxsw_sp_neigh_entry_create(const void *addr, size_t addr_len, + struct net_device *dev, u16 rif, + struct neighbour *n) +{ + struct mlxsw_sp_neigh_entry *neigh_entry; + + neigh_entry = kzalloc(sizeof(*neigh_entry), GFP_ATOMIC); + if (!neigh_entry) + return NULL; + memcpy(neigh_entry->key.addr, addr, addr_len); + neigh_entry->key.dev = dev; + neigh_entry->rif = rif; + neigh_entry->n = n; + return neigh_entry; +} + +static void +mlxsw_sp_neigh_entry_destroy(struct mlxsw_sp_neigh_entry *neigh_entry) +{ + kfree(neigh_entry); +} + +static struct mlxsw_sp_neigh_entry * +mlxsw_sp_neigh_entry_lookup(struct mlxsw_sp *mlxsw_sp, const void *addr, + size_t addr_len, struct net_device *dev) +{ + struct mlxsw_sp_neigh_key key = {{ 0 } }; + + memcpy(key.addr, addr, addr_len); + key.dev = dev; + return rhashtable_lookup_fast(&mlxsw_sp->router.neigh_ht, + &key, mlxsw_sp_neigh_ht_params); +} + +int mlxsw_sp_router_neigh_construct(struct net_device *dev, + struct neighbour *n) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_neigh_entry *neigh_entry; + struct mlxsw_sp_rif *r; + u32 dip; + int err; + + if (n->tbl != &arp_tbl) + return 0; + + dip = ntohl(*((__be32 *) n->primary_key)); + neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, &dip, sizeof(dip), + n->dev); + if (neigh_entry) { + WARN_ON(neigh_entry->n != n); + return 0; + } + + r = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); + if (WARN_ON(!r)) + return -EINVAL; + + neigh_entry = mlxsw_sp_neigh_entry_create(&dip, sizeof(dip), n->dev, + r->rif, n); + if (!neigh_entry) + return -ENOMEM; + err = mlxsw_sp_neigh_entry_insert(mlxsw_sp, neigh_entry); + if (err) + goto err_neigh_entry_insert; + return 0; + +err_neigh_entry_insert: + mlxsw_sp_neigh_entry_destroy(neigh_entry); + return err; +} + +void mlxsw_sp_router_neigh_destroy(struct net_device *dev, + struct neighbour *n) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_neigh_entry *neigh_entry; + u32 dip; + + if (n->tbl != &arp_tbl) + return; + + dip = ntohl(*((__be32 *) n->primary_key)); + neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, &dip, sizeof(dip), + n->dev); + if (!neigh_entry) + return; + mlxsw_sp_neigh_entry_remove(mlxsw_sp, neigh_entry); + mlxsw_sp_neigh_entry_destroy(neigh_entry); +} + +static int mlxsw_sp_neigh_init(struct mlxsw_sp *mlxsw_sp) +{ + return rhashtable_init(&mlxsw_sp->router.neigh_ht, + &mlxsw_sp_neigh_ht_params); +} + +static void mlxsw_sp_neigh_fini(struct mlxsw_sp *mlxsw_sp) +{ + rhashtable_destroy(&mlxsw_sp->router.neigh_ht); +} + static int __mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) { char rgcr_pl[MLXSW_REG_RGCR_LEN]; @@ -570,11 +713,12 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) return err; mlxsw_sp_lpm_init(mlxsw_sp); mlxsw_sp_vrs_init(mlxsw_sp); - return 0; + return mlxsw_sp_neigh_init(mlxsw_sp); } void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) { + mlxsw_sp_neigh_fini(mlxsw_sp); __mlxsw_sp_router_fini(mlxsw_sp); } -- cgit v0.10.2 From 4457b3df3fb933aae3f4e75eca669f1be5a4dd68 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Tue, 5 Jul 2016 11:27:40 +0200 Subject: mlxsw: reg: Add Router Algorithmic LPM Unicast Host Table register The RAUHT register is used to configure and query the Unicast Host Table in devices that implement the Algorithmic LPM. In other words, it is used to configure neighbour entries in the device. Signed-off-by: Yotam Gigi Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 9280d96..cc6a0b3 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -4,6 +4,7 @@ * Copyright (c) 2015-2016 Ido Schimmel * Copyright (c) 2015 Elad Raz * Copyright (c) 2015-2016 Jiri Pirko + * Copyright (c) 2016 Yotam Gigi * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: @@ -3884,6 +3885,144 @@ mlxsw_reg_ralue_act_ip2me_pack(char *payload) MLXSW_REG_RALUE_ACTION_TYPE_IP2ME); } +/* RAUHT - Router Algorithmic LPM Unicast Host Table Register + * ---------------------------------------------------------- + * The RAUHT register is used to configure and query the Unicast Host table in + * devices that implement the Algorithmic LPM. + */ +#define MLXSW_REG_RAUHT_ID 0x8014 +#define MLXSW_REG_RAUHT_LEN 0x74 + +static const struct mlxsw_reg_info mlxsw_reg_rauht = { + .id = MLXSW_REG_RAUHT_ID, + .len = MLXSW_REG_RAUHT_LEN, +}; + +enum mlxsw_reg_rauht_type { + MLXSW_REG_RAUHT_TYPE_IPV4, + MLXSW_REG_RAUHT_TYPE_IPV6, +}; + +/* reg_rauht_type + * Access: Index + */ +MLXSW_ITEM32(reg, rauht, type, 0x00, 24, 2); + +enum mlxsw_reg_rauht_op { + MLXSW_REG_RAUHT_OP_QUERY_READ = 0, + /* Read operation */ + MLXSW_REG_RAUHT_OP_QUERY_CLEAR_ON_READ = 1, + /* Clear on read operation. Used to read entry and clear + * activity bit. + */ + MLXSW_REG_RAUHT_OP_WRITE_ADD = 0, + /* Add. Used to write a new entry to the table. All R/W fields are + * relevant for new entry. Activity bit is set for new entries. + */ + MLXSW_REG_RAUHT_OP_WRITE_UPDATE = 1, + /* Update action. Used to update an existing route entry and + * only update the following fields: + * trap_action, trap_id, mac, counter_set_type, counter_index + */ + MLXSW_REG_RAUHT_OP_WRITE_CLEAR_ACTIVITY = 2, + /* Clear activity. A bit is cleared for the entry. */ + MLXSW_REG_RAUHT_OP_WRITE_DELETE = 3, + /* Delete entry */ + MLXSW_REG_RAUHT_OP_WRITE_DELETE_ALL = 4, + /* Delete all host entries on a RIF. In this command, dip + * field is reserved. + */ +}; + +/* reg_rauht_op + * Access: OP + */ +MLXSW_ITEM32(reg, rauht, op, 0x00, 20, 3); + +/* reg_rauht_a + * Activity. Set for new entries. Set if a packet lookup has hit on + * the specific entry. + * To clear the a bit, use "clear activity" op. + * Enabled by activity_dis in RGCR + * Access: RO + */ +MLXSW_ITEM32(reg, rauht, a, 0x00, 16, 1); + +/* reg_rauht_rif + * Router Interface + * Access: Index + */ +MLXSW_ITEM32(reg, rauht, rif, 0x00, 0, 16); + +/* reg_rauht_dip* + * Destination address. + * Access: Index + */ +MLXSW_ITEM32(reg, rauht, dip4, 0x1C, 0x0, 32); + +enum mlxsw_reg_rauht_trap_action { + MLXSW_REG_RAUHT_TRAP_ACTION_NOP, + MLXSW_REG_RAUHT_TRAP_ACTION_TRAP, + MLXSW_REG_RAUHT_TRAP_ACTION_MIRROR_TO_CPU, + MLXSW_REG_RAUHT_TRAP_ACTION_MIRROR, + MLXSW_REG_RAUHT_TRAP_ACTION_DISCARD_ERRORS, +}; + +/* reg_rauht_trap_action + * Access: RW + */ +MLXSW_ITEM32(reg, rauht, trap_action, 0x60, 28, 4); + +enum mlxsw_reg_rauht_trap_id { + MLXSW_REG_RAUHT_TRAP_ID_RTR_EGRESS0, + MLXSW_REG_RAUHT_TRAP_ID_RTR_EGRESS1, +}; + +/* reg_rauht_trap_id + * Trap ID to be reported to CPU. + * Trap-ID is RTR_EGRESS0 or RTR_EGRESS1. + * For trap_action of NOP, MIRROR and DISCARD_ERROR, + * trap_id is reserved. + * Access: RW + */ +MLXSW_ITEM32(reg, rauht, trap_id, 0x60, 0, 9); + +/* reg_rauht_counter_set_type + * Counter set type for flow counters + * Access: RW + */ +MLXSW_ITEM32(reg, rauht, counter_set_type, 0x68, 24, 8); + +/* reg_rauht_counter_index + * Counter index for flow counters + * Access: RW + */ +MLXSW_ITEM32(reg, rauht, counter_index, 0x68, 0, 24); + +/* reg_rauht_mac + * MAC address. + * Access: RW + */ +MLXSW_ITEM_BUF(reg, rauht, mac, 0x6E, 6); + +static inline void mlxsw_reg_rauht_pack(char *payload, + enum mlxsw_reg_rauht_op op, u16 rif, + const char *mac) +{ + MLXSW_REG_ZERO(rauht, payload); + mlxsw_reg_rauht_op_set(payload, op); + mlxsw_reg_rauht_rif_set(payload, rif); + mlxsw_reg_rauht_mac_memcpy_to(payload, mac); +} + +static inline void mlxsw_reg_rauht_pack4(char *payload, + enum mlxsw_reg_rauht_op op, u16 rif, + const char *mac, u32 dip) +{ + mlxsw_reg_rauht_pack(payload, op, rif, mac); + mlxsw_reg_rauht_dip4_set(payload, dip); +} + /* MFCR - Management Fan Control Register * -------------------------------------- * This register controls the settings of the Fan Speed PWM mechanism. @@ -4634,6 +4773,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "RALTB"; case MLXSW_REG_RALUE_ID: return "RALUE"; + case MLXSW_REG_RAUHT_ID: + return "RAUHT"; case MLXSW_REG_MFCR_ID: return "MFCR"; case MLXSW_REG_MFSC_ID: -- cgit v0.10.2 From 7cf2c205d7035a7ae525e3cfb5b7141fb7152372 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Tue, 5 Jul 2016 11:27:41 +0200 Subject: mlxsw: reg: Add Router Algorithmic LPM Unicast Host Table Dump register The RAUHTD register allows dumping entries from the Router Unicast Host Table. Signed-off-by: Yotam Gigi Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index cc6a0b3..fcf379b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -4023,6 +4023,151 @@ static inline void mlxsw_reg_rauht_pack4(char *payload, mlxsw_reg_rauht_dip4_set(payload, dip); } +/* RAUHTD - Router Algorithmic LPM Unicast Host Table Dump Register + * ---------------------------------------------------------------- + * The RAUHTD register allows dumping entries from the Router Unicast Host + * Table. For a given session an entry is dumped no more than one time. The + * first RAUHTD access after reset is a new session. A session ends when the + * num_rec response is smaller than num_rec request or for IPv4 when the + * num_entries is smaller than 4. The clear activity affect the current session + * or the last session if a new session has not started. + */ +#define MLXSW_REG_RAUHTD_ID 0x8018 +#define MLXSW_REG_RAUHTD_BASE_LEN 0x20 +#define MLXSW_REG_RAUHTD_REC_LEN 0x20 +#define MLXSW_REG_RAUHTD_REC_MAX_NUM 32 +#define MLXSW_REG_RAUHTD_LEN (MLXSW_REG_RAUHTD_BASE_LEN + \ + MLXSW_REG_RAUHTD_REC_MAX_NUM * MLXSW_REG_RAUHTD_REC_LEN) +#define MLXSW_REG_RAUHTD_IPV4_ENT_PER_REC 4 + +static const struct mlxsw_reg_info mlxsw_reg_rauhtd = { + .id = MLXSW_REG_RAUHTD_ID, + .len = MLXSW_REG_RAUHTD_LEN, +}; + +#define MLXSW_REG_RAUHTD_FILTER_A BIT(0) +#define MLXSW_REG_RAUHTD_FILTER_RIF BIT(3) + +/* reg_rauhtd_filter_fields + * if a bit is '0' then the relevant field is ignored and dump is done + * regardless of the field value + * Bit0 - filter by activity: entry_a + * Bit3 - filter by entry rip: entry_rif + * Access: Index + */ +MLXSW_ITEM32(reg, rauhtd, filter_fields, 0x00, 0, 8); + +enum mlxsw_reg_rauhtd_op { + MLXSW_REG_RAUHTD_OP_DUMP, + MLXSW_REG_RAUHTD_OP_DUMP_AND_CLEAR, +}; + +/* reg_rauhtd_op + * Access: OP + */ +MLXSW_ITEM32(reg, rauhtd, op, 0x04, 24, 2); + +/* reg_rauhtd_num_rec + * At request: number of records requested + * At response: number of records dumped + * For IPv4, each record has 4 entries at request and up to 4 entries + * at response + * Range is 0..MLXSW_REG_RAUHTD_REC_MAX_NUM + * Access: Index + */ +MLXSW_ITEM32(reg, rauhtd, num_rec, 0x04, 0, 8); + +/* reg_rauhtd_entry_a + * Dump only if activity has value of entry_a + * Reserved if filter_fields bit0 is '0' + * Access: Index + */ +MLXSW_ITEM32(reg, rauhtd, entry_a, 0x08, 16, 1); + +enum mlxsw_reg_rauhtd_type { + MLXSW_REG_RAUHTD_TYPE_IPV4, + MLXSW_REG_RAUHTD_TYPE_IPV6, +}; + +/* reg_rauhtd_type + * Dump only if record type is: + * 0 - IPv4 + * 1 - IPv6 + * Access: Index + */ +MLXSW_ITEM32(reg, rauhtd, type, 0x08, 0, 4); + +/* reg_rauhtd_entry_rif + * Dump only if RIF has value of entry_rif + * Reserved if filter_fields bit3 is '0' + * Access: Index + */ +MLXSW_ITEM32(reg, rauhtd, entry_rif, 0x0C, 0, 16); + +static inline void mlxsw_reg_rauhtd_pack(char *payload, + enum mlxsw_reg_rauhtd_type type) +{ + MLXSW_REG_ZERO(rauhtd, payload); + mlxsw_reg_rauhtd_filter_fields_set(payload, MLXSW_REG_RAUHTD_FILTER_A); + mlxsw_reg_rauhtd_op_set(payload, MLXSW_REG_RAUHTD_OP_DUMP_AND_CLEAR); + mlxsw_reg_rauhtd_num_rec_set(payload, MLXSW_REG_RAUHTD_REC_MAX_NUM); + mlxsw_reg_rauhtd_entry_a_set(payload, 1); + mlxsw_reg_rauhtd_type_set(payload, type); +} + +/* reg_rauhtd_ipv4_rec_num_entries + * Number of valid entries in this record: + * 0 - 1 valid entry + * 1 - 2 valid entries + * 2 - 3 valid entries + * 3 - 4 valid entries + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, rauhtd, ipv4_rec_num_entries, + MLXSW_REG_RAUHTD_BASE_LEN, 28, 2, + MLXSW_REG_RAUHTD_REC_LEN, 0x00, false); + +/* reg_rauhtd_rec_type + * Record type. + * 0 - IPv4 + * 1 - IPv6 + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, rauhtd, rec_type, MLXSW_REG_RAUHTD_BASE_LEN, 24, 2, + MLXSW_REG_RAUHTD_REC_LEN, 0x00, false); + +#define MLXSW_REG_RAUHTD_IPV4_ENT_LEN 0x8 + +/* reg_rauhtd_ipv4_ent_a + * Activity. Set for new entries. Set if a packet lookup has hit on the + * specific entry. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, rauhtd, ipv4_ent_a, MLXSW_REG_RAUHTD_BASE_LEN, 16, 1, + MLXSW_REG_RAUHTD_IPV4_ENT_LEN, 0x00, false); + +/* reg_rauhtd_ipv4_ent_rif + * Router interface. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, rauhtd, ipv4_ent_rif, MLXSW_REG_RAUHTD_BASE_LEN, 0, + 16, MLXSW_REG_RAUHTD_IPV4_ENT_LEN, 0x00, false); + +/* reg_rauhtd_ipv4_ent_dip + * Destination IPv4 address. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, rauhtd, ipv4_ent_dip, MLXSW_REG_RAUHTD_BASE_LEN, 0, + 32, MLXSW_REG_RAUHTD_IPV4_ENT_LEN, 0x04, false); + +static inline void mlxsw_reg_rauhtd_ent_ipv4_unpack(char *payload, + int ent_index, u16 *p_rif, + u32 *p_dip) +{ + *p_rif = mlxsw_reg_rauhtd_ipv4_ent_rif_get(payload, ent_index); + *p_dip = mlxsw_reg_rauhtd_ipv4_ent_dip_get(payload, ent_index); +} + /* MFCR - Management Fan Control Register * -------------------------------------- * This register controls the settings of the Fan Speed PWM mechanism. @@ -4775,6 +4920,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "RALUE"; case MLXSW_REG_RAUHT_ID: return "RAUHT"; + case MLXSW_REG_RAUHTD_ID: + return "RAUHTD"; case MLXSW_REG_MFCR_ID: return "MFCR"; case MLXSW_REG_MFSC_ID: -- cgit v0.10.2 From 2a4501ae18b52fcdf553404286e6cefabd1d17ec Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 5 Jul 2016 11:27:42 +0200 Subject: neigh: Send a notification when DELAY_PROBE_TIME changes When the data plane is offloaded the traffic doesn't go through the networking stack. Therefore, after first resolving a neighbour the NUD state machine will transition it from REACHABLE to STALE until it's finally deleted by the garbage collector. To prevent such situations the offloading driver should notify the NUD state machine on any neighbours that were recently used. The driver's polling interval should be set so that the NUD state machine can function as if the traffic wasn't offloaded. Currently, there are no in-tree drivers that can report confirmation for a neighbour, but only 'used' indication. Therefore, the polling interval should be set according to DELAY_FIRST_PROBE_TIME, as a neighbour will transition from REACHABLE state to DELAY (instead of STALE) if "a packet was sent within the last DELAY_FIRST_PROBE_TIME seconds" (RFC 4861). Send a netevent whenever the DELAY_FIRST_PROBE_TIME changes - either via netlink or sysctl - so that offloading drivers can correctly set their polling interval. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/include/net/netevent.h b/include/net/netevent.h index d8bbb38..f440df1 100644 --- a/include/net/netevent.h +++ b/include/net/netevent.h @@ -24,6 +24,7 @@ struct netevent_redirect { enum netevent_notif_type { NETEVENT_NEIGH_UPDATE = 1, /* arg is struct neighbour ptr */ NETEVENT_REDIRECT, /* arg is struct netevent_redirect ptr */ + NETEVENT_DELAY_PROBE_TIME_UPDATE, /* arg is struct neigh_parms ptr */ }; int register_netevent_notifier(struct notifier_block *nb); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 952aabb..5cdc62a 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -2047,6 +2047,7 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh) case NDTPA_DELAY_PROBE_TIME: NEIGH_VAR_SET(p, DELAY_PROBE_TIME, nla_get_msecs(tbp[i])); + call_netevent_notifiers(NETEVENT_DELAY_PROBE_TIME_UPDATE, p); break; case NDTPA_RETRANS_TIME: NEIGH_VAR_SET(p, RETRANS_TIME, @@ -2930,6 +2931,7 @@ static void neigh_proc_update(struct ctl_table *ctl, int write) return; set_bit(index, p->data_state); + call_netevent_notifiers(NETEVENT_DELAY_PROBE_TIME_UPDATE, p); if (!dev) /* NULL dev means this is default value */ neigh_copy_dflt_parms(net, p, index); } -- cgit v0.10.2 From c723c735fa6bacfdb01c4697c2cfeba142990d18 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Tue, 5 Jul 2016 11:27:43 +0200 Subject: mlxsw: spectrum_router: Periodically update the kernel's neigh table As previously explained, the driver should periodically poll the device for neighbours activity according to the configured DELAY_PROBE_TIME. This will prevent active neighbours from staying in STALE state for long periods of time. During init configure the polling interval according to the DELAY_PROBE_TIME used in the default table. In addition, register a netevent notification block, so that the interval is updated whenever DELAY_PROBE_TIME changes. Using the computed interval schedule a delayed work, which will update the kernel via neigh_event_send() on any active neighbour since the last delayed work. Signed-off-by: Yotam Gigi Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 734c5ba..9c2a60f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -214,6 +214,10 @@ struct mlxsw_sp_router { struct mlxsw_sp_lpm_tree lpm_trees[MLXSW_SP_LPM_TREE_COUNT]; struct mlxsw_sp_vr vrs[MLXSW_SP_VIRTUAL_ROUTER_MAX]; struct rhashtable neigh_ht; + struct { + struct delayed_work dw; + unsigned long interval; /* ms */ + } neighs_update; }; struct mlxsw_sp { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 90d382a..db1c2c4 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -3,6 +3,7 @@ * Copyright (c) 2016 Mellanox Technologies. All rights reserved. * Copyright (c) 2016 Jiri Pirko * Copyright (c) 2016 Ido Schimmel + * Copyright (c) 2016 Yotam Gigi * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: @@ -38,6 +39,8 @@ #include #include #include +#include +#include #include #include @@ -676,14 +679,199 @@ void mlxsw_sp_router_neigh_destroy(struct net_device *dev, mlxsw_sp_neigh_entry_destroy(neigh_entry); } +static void +mlxsw_sp_router_neighs_update_interval_init(struct mlxsw_sp *mlxsw_sp) +{ + unsigned long interval = NEIGH_VAR(&arp_tbl.parms, DELAY_PROBE_TIME); + + mlxsw_sp->router.neighs_update.interval = jiffies_to_msecs(interval); +} + +static void mlxsw_sp_router_neigh_ent_ipv4_process(struct mlxsw_sp *mlxsw_sp, + char *rauhtd_pl, + int ent_index) +{ + struct net_device *dev; + struct neighbour *n; + __be32 dipn; + u32 dip; + u16 rif; + + mlxsw_reg_rauhtd_ent_ipv4_unpack(rauhtd_pl, ent_index, &rif, &dip); + + if (!mlxsw_sp->rifs[rif]) { + dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Incorrect RIF in neighbour entry\n"); + return; + } + + dipn = htonl(dip); + dev = mlxsw_sp->rifs[rif]->dev; + n = neigh_lookup(&arp_tbl, &dipn, dev); + if (!n) { + netdev_err(dev, "Failed to find matching neighbour for IP=%pI4h\n", + &dip); + return; + } + + netdev_dbg(dev, "Updating neighbour with IP=%pI4h\n", &dip); + neigh_event_send(n, NULL); + neigh_release(n); +} + +static void mlxsw_sp_router_neigh_rec_ipv4_process(struct mlxsw_sp *mlxsw_sp, + char *rauhtd_pl, + int rec_index) +{ + u8 num_entries; + int i; + + num_entries = mlxsw_reg_rauhtd_ipv4_rec_num_entries_get(rauhtd_pl, + rec_index); + /* Hardware starts counting at 0, so add 1. */ + num_entries++; + + /* Each record consists of several neighbour entries. */ + for (i = 0; i < num_entries; i++) { + int ent_index; + + ent_index = rec_index * MLXSW_REG_RAUHTD_IPV4_ENT_PER_REC + i; + mlxsw_sp_router_neigh_ent_ipv4_process(mlxsw_sp, rauhtd_pl, + ent_index); + } + +} + +static void mlxsw_sp_router_neigh_rec_process(struct mlxsw_sp *mlxsw_sp, + char *rauhtd_pl, int rec_index) +{ + switch (mlxsw_reg_rauhtd_rec_type_get(rauhtd_pl, rec_index)) { + case MLXSW_REG_RAUHTD_TYPE_IPV4: + mlxsw_sp_router_neigh_rec_ipv4_process(mlxsw_sp, rauhtd_pl, + rec_index); + break; + case MLXSW_REG_RAUHTD_TYPE_IPV6: + WARN_ON_ONCE(1); + break; + } +} + +static void +mlxsw_sp_router_neighs_update_work_schedule(struct mlxsw_sp *mlxsw_sp) +{ + unsigned long interval = mlxsw_sp->router.neighs_update.interval; + + mlxsw_core_schedule_dw(&mlxsw_sp->router.neighs_update.dw, + msecs_to_jiffies(interval)); +} + +static void mlxsw_sp_router_neighs_update_work(struct work_struct *work) +{ + struct mlxsw_sp *mlxsw_sp; + char *rauhtd_pl; + u8 num_rec; + int i, err; + + rauhtd_pl = kmalloc(MLXSW_REG_RAUHTD_LEN, GFP_KERNEL); + if (!rauhtd_pl) + return; + + mlxsw_sp = container_of(work, struct mlxsw_sp, + router.neighs_update.dw.work); + + /* Make sure the neighbour's netdev isn't removed in the + * process. + */ + rtnl_lock(); + do { + mlxsw_reg_rauhtd_pack(rauhtd_pl, MLXSW_REG_RAUHTD_TYPE_IPV4); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(rauhtd), + rauhtd_pl); + if (err) { + dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Failed to dump neighbour talbe\n"); + break; + } + num_rec = mlxsw_reg_rauhtd_num_rec_get(rauhtd_pl); + for (i = 0; i < num_rec; i++) + mlxsw_sp_router_neigh_rec_process(mlxsw_sp, rauhtd_pl, + i); + } while (num_rec); + rtnl_unlock(); + + kfree(rauhtd_pl); + mlxsw_sp_router_neighs_update_work_schedule(mlxsw_sp); +} + +static int mlxsw_sp_router_netevent_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct mlxsw_sp_port *mlxsw_sp_port; + struct mlxsw_sp *mlxsw_sp; + unsigned long interval; + struct neigh_parms *p; + + switch (event) { + case NETEVENT_DELAY_PROBE_TIME_UPDATE: + p = ptr; + + /* We don't care about changes in the default table. */ + if (!p->dev || p->tbl != &arp_tbl) + return NOTIFY_DONE; + + /* We are in atomic context and can't take RTNL mutex, + * so use RCU variant to walk the device chain. + */ + mlxsw_sp_port = mlxsw_sp_port_lower_dev_hold(p->dev); + if (!mlxsw_sp_port) + return NOTIFY_DONE; + + mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + interval = jiffies_to_msecs(NEIGH_VAR(p, DELAY_PROBE_TIME)); + mlxsw_sp->router.neighs_update.interval = interval; + + mlxsw_sp_port_dev_put(mlxsw_sp_port); + break; + } + + return NOTIFY_DONE; +} + +static struct notifier_block mlxsw_sp_router_netevent_nb __read_mostly = { + .notifier_call = mlxsw_sp_router_netevent_event, +}; + static int mlxsw_sp_neigh_init(struct mlxsw_sp *mlxsw_sp) { - return rhashtable_init(&mlxsw_sp->router.neigh_ht, - &mlxsw_sp_neigh_ht_params); + int err; + + err = rhashtable_init(&mlxsw_sp->router.neigh_ht, + &mlxsw_sp_neigh_ht_params); + if (err) + return err; + + /* Initialize the polling interval according to the default + * table. + */ + mlxsw_sp_router_neighs_update_interval_init(mlxsw_sp); + + err = register_netevent_notifier(&mlxsw_sp_router_netevent_nb); + if (err) + goto err_register_netevent_notifier; + + INIT_DELAYED_WORK(&mlxsw_sp->router.neighs_update.dw, + mlxsw_sp_router_neighs_update_work); + mlxsw_core_schedule_dw(&mlxsw_sp->router.neighs_update.dw, 0); + + return 0; + +err_register_netevent_notifier: + rhashtable_destroy(&mlxsw_sp->router.neigh_ht); + return err; } static void mlxsw_sp_neigh_fini(struct mlxsw_sp *mlxsw_sp) { + cancel_delayed_work_sync(&mlxsw_sp->router.neighs_update.dw); + unregister_netevent_notifier(&mlxsw_sp_router_netevent_nb); rhashtable_destroy(&mlxsw_sp->router.neigh_ht); } -- cgit v0.10.2 From a6bf9e933daf9609c3673e8348e7f95f3ef325c7 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Tue, 5 Jul 2016 11:27:44 +0200 Subject: mlxsw: spectrum_router: Offload neighbours based on NUD state change Listen to any NEIGH_UPDATE events sent and program the device accordingly. If NUD state is VALID and neighbour isn't yet offloaded, then program it into the device's table. Otherwise, just edit its parameters. If NUD state machine transitioned neighbour out of VALID state and it's present in the device's table, then remove it. Note that the device is programmed in delayed work, as the netevent notification chain is atomic and prevents us from going to sleep. Signed-off-by: Yotam Gigi Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index db1c2c4..ed0e6c0 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -559,6 +559,10 @@ struct mlxsw_sp_neigh_entry { struct mlxsw_sp_neigh_key key; u16 rif; struct neighbour *n; + bool offloaded; + struct delayed_work dw; + struct mlxsw_sp_port *mlxsw_sp_port; + unsigned char ha[ETH_ALEN]; }; static const struct rhashtable_params mlxsw_sp_neigh_ht_params = { @@ -585,6 +589,8 @@ mlxsw_sp_neigh_entry_remove(struct mlxsw_sp *mlxsw_sp, mlxsw_sp_neigh_ht_params); } +static void mlxsw_sp_router_neigh_update_hw(struct work_struct *work); + static struct mlxsw_sp_neigh_entry * mlxsw_sp_neigh_entry_create(const void *addr, size_t addr_len, struct net_device *dev, u16 rif, @@ -599,6 +605,7 @@ mlxsw_sp_neigh_entry_create(const void *addr, size_t addr_len, neigh_entry->key.dev = dev; neigh_entry->rif = rif; neigh_entry->n = n; + INIT_DELAYED_WORK(&neigh_entry->dw, mlxsw_sp_router_neigh_update_hw); return neigh_entry; } @@ -801,13 +808,76 @@ static void mlxsw_sp_router_neighs_update_work(struct work_struct *work) mlxsw_sp_router_neighs_update_work_schedule(mlxsw_sp); } +static void mlxsw_sp_router_neigh_update_hw(struct work_struct *work) +{ + struct mlxsw_sp_neigh_entry *neigh_entry = + container_of(work, struct mlxsw_sp_neigh_entry, dw.work); + struct neighbour *n = neigh_entry->n; + struct mlxsw_sp_port *mlxsw_sp_port = neigh_entry->mlxsw_sp_port; + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char rauht_pl[MLXSW_REG_RAUHT_LEN]; + struct net_device *dev; + bool entry_connected; + u8 nud_state; + bool updating; + bool removing; + bool adding; + u32 dip; + int err; + + read_lock_bh(&n->lock); + dip = ntohl(*((__be32 *) n->primary_key)); + memcpy(neigh_entry->ha, n->ha, sizeof(neigh_entry->ha)); + nud_state = n->nud_state; + dev = n->dev; + read_unlock_bh(&n->lock); + + entry_connected = nud_state & NUD_VALID; + adding = (!neigh_entry->offloaded) && entry_connected; + updating = neigh_entry->offloaded && entry_connected; + removing = neigh_entry->offloaded && !entry_connected; + + if (adding || updating) { + mlxsw_reg_rauht_pack4(rauht_pl, MLXSW_REG_RAUHT_OP_WRITE_ADD, + neigh_entry->rif, + neigh_entry->ha, dip); + err = mlxsw_reg_write(mlxsw_sp->core, + MLXSW_REG(rauht), rauht_pl); + if (err) { + netdev_err(dev, "Could not add neigh %pI4h\n", &dip); + neigh_entry->offloaded = false; + } else { + neigh_entry->offloaded = true; + } + } else if (removing) { + mlxsw_reg_rauht_pack4(rauht_pl, MLXSW_REG_RAUHT_OP_WRITE_DELETE, + neigh_entry->rif, + neigh_entry->ha, dip); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rauht), + rauht_pl); + if (err) { + netdev_err(dev, "Could not delete neigh %pI4h\n", &dip); + neigh_entry->offloaded = true; + } else { + neigh_entry->offloaded = false; + } + } + + neigh_release(n); + mlxsw_sp_port_dev_put(mlxsw_sp_port); +} + static int mlxsw_sp_router_netevent_event(struct notifier_block *unused, unsigned long event, void *ptr) { + struct mlxsw_sp_neigh_entry *neigh_entry; struct mlxsw_sp_port *mlxsw_sp_port; struct mlxsw_sp *mlxsw_sp; unsigned long interval; + struct net_device *dev; struct neigh_parms *p; + struct neighbour *n; + u32 dip; switch (event) { case NETEVENT_DELAY_PROBE_TIME_UPDATE: @@ -830,6 +900,39 @@ static int mlxsw_sp_router_netevent_event(struct notifier_block *unused, mlxsw_sp_port_dev_put(mlxsw_sp_port); break; + case NETEVENT_NEIGH_UPDATE: + n = ptr; + dev = n->dev; + + if (n->tbl != &arp_tbl) + return NOTIFY_DONE; + + mlxsw_sp_port = mlxsw_sp_port_lower_dev_hold(dev); + if (!mlxsw_sp_port) + return NOTIFY_DONE; + + mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + dip = ntohl(*((__be32 *) n->primary_key)); + neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, + &dip, + sizeof(__be32), + dev); + if (WARN_ON(!neigh_entry) || WARN_ON(neigh_entry->n != n)) { + mlxsw_sp_port_dev_put(mlxsw_sp_port); + return NOTIFY_DONE; + } + neigh_entry->mlxsw_sp_port = mlxsw_sp_port; + + /* Take a reference to ensure the neighbour won't be + * destructed until we drop the reference in delayed + * work. + */ + neigh_clone(n); + if (!mlxsw_core_schedule_dw(&neigh_entry->dw, 0)) { + neigh_release(n); + mlxsw_sp_port_dev_put(mlxsw_sp_port); + } + break; } return NOTIFY_DONE; -- cgit v0.10.2 From 489107bda1d1ad20a56cb19753d705006b21bb90 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 Jul 2016 11:27:45 +0200 Subject: mlxsw: Add KVD sizes configuration into profile Up until now we only used hash-based tables in the device, but we are going to use the linear table for remote routes adjacency lists. Add the configuration fields that control the size of the linear table. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/cmd.h b/drivers/net/ethernet/mellanox/mlxsw/cmd.h index cd63b82..f9cd6e3 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/cmd.h +++ b/drivers/net/ethernet/mellanox/mlxsw/cmd.h @@ -607,6 +607,24 @@ MLXSW_ITEM32(cmd_mbox, config_profile, */ MLXSW_ITEM32(cmd_mbox, config_profile, set_ar_sec, 0x0C, 15, 1); +/* cmd_mbox_config_set_kvd_linear_size + * Capability bit. Setting a bit to 1 configures the profile + * according to the mailbox contents. + */ +MLXSW_ITEM32(cmd_mbox, config_profile, set_kvd_linear_size, 0x0C, 24, 1); + +/* cmd_mbox_config_set_kvd_hash_single_size + * Capability bit. Setting a bit to 1 configures the profile + * according to the mailbox contents. + */ +MLXSW_ITEM32(cmd_mbox, config_profile, set_kvd_hash_single_size, 0x0C, 25, 1); + +/* cmd_mbox_config_set_kvd_hash_double_size + * Capability bit. Setting a bit to 1 configures the profile + * according to the mailbox contents. + */ +MLXSW_ITEM32(cmd_mbox, config_profile, set_kvd_hash_double_size, 0x0C, 26, 1); + /* cmd_mbox_config_profile_max_vepa_channels * Maximum number of VEPA channels per port (0 through 16) * 0 - multi-channel VEPA is disabled @@ -733,6 +751,31 @@ MLXSW_ITEM32(cmd_mbox, config_profile, adaptive_routing_group_cap, 0x4C, 0, 16); */ MLXSW_ITEM32(cmd_mbox, config_profile, arn, 0x50, 31, 1); +/* cmd_mbox_config_kvd_linear_size + * KVD Linear Size + * Valid for Spectrum only + * Allowed values are 128*N where N=0 or higher + */ +MLXSW_ITEM32(cmd_mbox, config_profile, kvd_linear_size, 0x54, 0, 24); + +/* cmd_mbox_config_kvd_hash_single_size + * KVD Hash single-entries size + * Valid for Spectrum only + * Allowed values are 128*N where N=0 or higher + * Must be greater or equal to cap_min_kvd_hash_single_size + * Must be smaller or equal to cap_kvd_size - kvd_linear_size + */ +MLXSW_ITEM32(cmd_mbox, config_profile, kvd_hash_single_size, 0x58, 0, 24); + +/* cmd_mbox_config_kvd_hash_double_size + * KVD Hash double-entries size (units of single-size entries) + * Valid for Spectrum only + * Allowed values are 128*N where N=0 or higher + * Must be either 0 or greater or equal to cap_min_kvd_hash_double_size + * Must be smaller or equal to cap_kvd_size - kvd_linear_size + */ +MLXSW_ITEM32(cmd_mbox, config_profile, kvd_hash_double_size, 0x5C, 0, 24); + /* cmd_mbox_config_profile_swid_config_mask * Modify Switch Partition Configuration mask. When set, the configu- * ration value for the Switch Partition are taken from the mailbox. diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h index 436bc49..2fe385c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.h +++ b/drivers/net/ethernet/mellanox/mlxsw/core.h @@ -190,7 +190,8 @@ struct mlxsw_config_profile { used_max_ib_mc:1, used_max_pkey:1, used_ar_sec:1, - used_adaptive_routing_group_cap:1; + used_adaptive_routing_group_cap:1, + used_kvd_sizes:1; u8 max_vepa_channels; u16 max_lag; u16 max_port_per_lag; @@ -211,6 +212,9 @@ struct mlxsw_config_profile { u8 ar_sec; u16 adaptive_routing_group_cap; u8 arn; + u32 kvd_linear_size; + u32 kvd_hash_single_size; + u32 kvd_hash_double_size; struct mlxsw_swid_config swid_config[MLXSW_CONFIG_PROFILE_SWID_COUNT]; }; diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c index 7f4173c..ddbc9f2 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci.c +++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c @@ -1255,6 +1255,20 @@ static int mlxsw_pci_config_profile(struct mlxsw_pci *mlxsw_pci, char *mbox, mlxsw_cmd_mbox_config_profile_adaptive_routing_group_cap_set( mbox, profile->adaptive_routing_group_cap); } + if (profile->used_kvd_sizes) { + mlxsw_cmd_mbox_config_profile_set_kvd_linear_size_set( + mbox, 1); + mlxsw_cmd_mbox_config_profile_kvd_linear_size_set( + mbox, profile->kvd_linear_size); + mlxsw_cmd_mbox_config_profile_set_kvd_hash_single_size_set( + mbox, 1); + mlxsw_cmd_mbox_config_profile_kvd_hash_single_size_set( + mbox, profile->kvd_hash_single_size); + mlxsw_cmd_mbox_config_profile_set_kvd_hash_double_size_set( + mbox, 1); + mlxsw_cmd_mbox_config_profile_kvd_hash_double_size_set( + mbox, profile->kvd_hash_double_size); + } for (i = 0; i < MLXSW_CONFIG_PROFILE_SWID_COUNT; i++) mlxsw_pci_config_profile_swid_config(mlxsw_pci, mbox, i, -- cgit v0.10.2 From c602242761dff9f698e16f3ab474e48da1b83f8f Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 Jul 2016 11:27:46 +0200 Subject: mlxsw: spectrum: Define sizes of KVD areas Override the defaults and define the area sizes ourselves. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 9bebb7a..c812513 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2356,6 +2356,10 @@ static struct mlxsw_config_profile mlxsw_sp_config_profile = { .max_ib_mc = 0, .used_max_pkey = 1, .max_pkey = 0, + .used_kvd_sizes = 1, + .kvd_linear_size = MLXSW_SP_KVD_LINEAR_SIZE, + .kvd_hash_single_size = MLXSW_SP_KVD_HASH_SINGLE_SIZE, + .kvd_hash_double_size = MLXSW_SP_KVD_HASH_DOUBLE_SIZE, .swid_config = { { .used_type = 1, diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 9c2a60f..f7d34d8 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -76,6 +76,10 @@ #define MLXSW_SP_BYTES_TO_CELLS(b) DIV_ROUND_UP(b, MLXSW_SP_BYTES_PER_CELL) #define MLXSW_SP_CELLS_TO_BYTES(c) (c * MLXSW_SP_BYTES_PER_CELL) +#define MLXSW_SP_KVD_LINEAR_SIZE 65536 /* entries */ +#define MLXSW_SP_KVD_HASH_SINGLE_SIZE 163840 /* entries */ +#define MLXSW_SP_KVD_HASH_DOUBLE_SIZE 32768 /* entries */ + /* Maximum delay buffer needed in case of PAUSE frames, in cells. * Assumes 100m cable and maximum MTU. */ -- cgit v0.10.2 From b090ef068645d6a9e69f04e571a3dcb32eb37d81 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 Jul 2016 11:27:47 +0200 Subject: mlxsw: Introduce simplistic KVD linear area manager This is a very simple manager for KVD linear area. Currently, the allocator will either allocate a single entry from pre-defined sub-area, or in case more than one entry is needed, it will allocate 32-entry chunk in other pre-defined sub-area. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/Makefile b/drivers/net/ethernet/mellanox/mlxsw/Makefile index ea05f8a..d20ae18 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/Makefile +++ b/drivers/net/ethernet/mellanox/mlxsw/Makefile @@ -7,5 +7,6 @@ obj-$(CONFIG_MLXSW_SWITCHX2) += mlxsw_switchx2.o mlxsw_switchx2-objs := switchx2.o obj-$(CONFIG_MLXSW_SPECTRUM) += mlxsw_spectrum.o mlxsw_spectrum-objs := spectrum.o spectrum_buffers.o \ - spectrum_switchdev.o spectrum_router.o + spectrum_switchdev.o spectrum_router.o \ + spectrum_kvdl.o mlxsw_spectrum-$(CONFIG_MLXSW_SPECTRUM_DCB) += spectrum_dcb.o diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index f7d34d8..e781128 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -253,6 +253,9 @@ struct mlxsw_sp { u8 port_to_module[MLXSW_PORT_MAX_PORTS]; struct mlxsw_sp_sb sb; struct mlxsw_sp_router router; + struct { + DECLARE_BITMAP(usage, MLXSW_SP_KVD_LINEAR_SIZE); + } kvdl; }; static inline struct mlxsw_sp_upper * @@ -539,4 +542,7 @@ int mlxsw_sp_router_neigh_construct(struct net_device *dev, void mlxsw_sp_router_neigh_destroy(struct net_device *dev, struct neighbour *n); +int mlxsw_sp_kvdl_alloc(struct mlxsw_sp *mlxsw_sp, unsigned int entry_count); +void mlxsw_sp_kvdl_free(struct mlxsw_sp *mlxsw_sp, int entry_index); + #endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c new file mode 100644 index 0000000..ac321e8 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c @@ -0,0 +1,91 @@ +/* + * drivers/net/ethernet/mellanox/mlxsw/spectrum_kvdl.c + * Copyright (c) 2016 Mellanox Technologies. All rights reserved. + * Copyright (c) 2016 Jiri Pirko + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include +#include + +#include "spectrum.h" + +#define MLXSW_SP_KVDL_SINGLE_BASE 0 +#define MLXSW_SP_KVDL_SINGLE_SIZE 16384 +#define MLXSW_SP_KVDL_CHUNKS_BASE \ + (MLXSW_SP_KVDL_SINGLE_BASE + MLXSW_SP_KVDL_SINGLE_SIZE) +#define MLXSW_SP_KVDL_CHUNKS_SIZE \ + (MLXSW_SP_KVD_LINEAR_SIZE - MLXSW_SP_KVDL_CHUNKS_BASE) +#define MLXSW_SP_CHUNK_MAX 32 + +int mlxsw_sp_kvdl_alloc(struct mlxsw_sp *mlxsw_sp, unsigned int entry_count) +{ + int entry_index; + int size; + int type_base; + int type_size; + int type_entries; + + if (entry_count == 0 || entry_count > MLXSW_SP_CHUNK_MAX) { + return -EINVAL; + } else if (entry_count == 1) { + type_base = MLXSW_SP_KVDL_SINGLE_BASE; + type_size = MLXSW_SP_KVDL_SINGLE_SIZE; + type_entries = 1; + } else { + type_base = MLXSW_SP_KVDL_CHUNKS_BASE; + type_size = MLXSW_SP_KVDL_CHUNKS_SIZE; + type_entries = MLXSW_SP_CHUNK_MAX; + } + + entry_index = type_base; + size = type_base + type_size; + for_each_clear_bit_from(entry_index, mlxsw_sp->kvdl.usage, size) { + int i; + + for (i = 0; i < type_entries; i++) + set_bit(entry_index + i, mlxsw_sp->kvdl.usage); + return entry_index; + } + return -ENOBUFS; +} + +void mlxsw_sp_kvdl_free(struct mlxsw_sp *mlxsw_sp, int entry_index) +{ + int type_entries; + int i; + + if (entry_index < MLXSW_SP_KVDL_CHUNKS_BASE) + type_entries = 1; + else + type_entries = MLXSW_SP_CHUNK_MAX; + for (i = 0; i < type_entries; i++) + clear_bit(entry_index + i, mlxsw_sp->kvdl.usage); +} -- cgit v0.10.2 From 089f981683f87a299f8a3b81696629ae1c7b1abb Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Tue, 5 Jul 2016 11:27:48 +0200 Subject: mlxsw: reg: Add Router Adjacency Table register The RATR register is used to configure the Router Adjacency (next-hop) Table. Signed-off-by: Yotam Gigi Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index fcf379b..a2d7870 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -3455,6 +3455,137 @@ static inline void mlxsw_reg_ritr_pack(char *payload, bool enable, mlxsw_reg_ritr_if_mac_memcpy_to(payload, mac); } +/* RATR - Router Adjacency Table Register + * -------------------------------------- + * The RATR register is used to configure the Router Adjacency (next-hop) + * Table. + */ +#define MLXSW_REG_RATR_ID 0x8008 +#define MLXSW_REG_RATR_LEN 0x2C + +static const struct mlxsw_reg_info mlxsw_reg_ratr = { + .id = MLXSW_REG_RATR_ID, + .len = MLXSW_REG_RATR_LEN, +}; + +enum mlxsw_reg_ratr_op { + /* Read */ + MLXSW_REG_RATR_OP_QUERY_READ = 0, + /* Read and clear activity */ + MLXSW_REG_RATR_OP_QUERY_READ_CLEAR = 2, + /* Write Adjacency entry */ + MLXSW_REG_RATR_OP_WRITE_WRITE_ENTRY = 1, + /* Write Adjacency entry only if the activity is cleared. + * The write may not succeed if the activity is set. There is not + * direct feedback if the write has succeeded or not, however + * the get will reveal the actual entry (SW can compare the get + * response to the set command). + */ + MLXSW_REG_RATR_OP_WRITE_WRITE_ENTRY_ON_ACTIVITY = 3, +}; + +/* reg_ratr_op + * Note that Write operation may also be used for updating + * counter_set_type and counter_index. In this case all other + * fields must not be updated. + * Access: OP + */ +MLXSW_ITEM32(reg, ratr, op, 0x00, 28, 4); + +/* reg_ratr_v + * Valid bit. Indicates if the adjacency entry is valid. + * Note: the device may need some time before reusing an invalidated + * entry. During this time the entry can not be reused. It is + * recommended to use another entry before reusing an invalidated + * entry (e.g. software can put it at the end of the list for + * reusing). Trying to access an invalidated entry not yet cleared + * by the device results with failure indicating "Try Again" status. + * When valid is '0' then egress_router_interface,trap_action, + * adjacency_parameters and counters are reserved + * Access: RW + */ +MLXSW_ITEM32(reg, ratr, v, 0x00, 24, 1); + +/* reg_ratr_a + * Activity. Set for new entries. Set if a packet lookup has hit on + * the specific entry. To clear the a bit, use "clear activity". + * Access: RO + */ +MLXSW_ITEM32(reg, ratr, a, 0x00, 16, 1); + +/* reg_ratr_adjacency_index_low + * Bits 15:0 of index into the adjacency table. + * For SwitchX and SwitchX-2, the adjacency table is linear and + * used for adjacency entries only. + * For Spectrum, the index is to the KVD linear. + * Access: Index + */ +MLXSW_ITEM32(reg, ratr, adjacency_index_low, 0x04, 0, 16); + +/* reg_ratr_egress_router_interface + * Range is 0 .. cap_max_router_interfaces - 1 + * Access: RW + */ +MLXSW_ITEM32(reg, ratr, egress_router_interface, 0x08, 0, 16); + +enum mlxsw_reg_ratr_trap_action { + MLXSW_REG_RATR_TRAP_ACTION_NOP, + MLXSW_REG_RATR_TRAP_ACTION_TRAP, + MLXSW_REG_RATR_TRAP_ACTION_MIRROR_TO_CPU, + MLXSW_REG_RATR_TRAP_ACTION_MIRROR, + MLXSW_REG_RATR_TRAP_ACTION_DISCARD_ERRORS, +}; + +/* reg_ratr_trap_action + * see mlxsw_reg_ratr_trap_action + * Access: RW + */ +MLXSW_ITEM32(reg, ratr, trap_action, 0x0C, 28, 4); + +enum mlxsw_reg_ratr_trap_id { + MLXSW_REG_RATR_TRAP_ID_RTR_EGRESS0 = 0, + MLXSW_REG_RATR_TRAP_ID_RTR_EGRESS1 = 1, +}; + +/* reg_ratr_adjacency_index_high + * Bits 23:16 of the adjacency_index. + * Access: Index + */ +MLXSW_ITEM32(reg, ratr, adjacency_index_high, 0x0C, 16, 8); + +/* reg_ratr_trap_id + * Trap ID to be reported to CPU. + * Trap-ID is RTR_EGRESS0 or RTR_EGRESS1. + * For trap_action of NOP, MIRROR and DISCARD_ERROR + * Access: RW + */ +MLXSW_ITEM32(reg, ratr, trap_id, 0x0C, 0, 8); + +/* reg_ratr_eth_destination_mac + * MAC address of the destination next-hop. + * Access: RW + */ +MLXSW_ITEM_BUF(reg, ratr, eth_destination_mac, 0x12, 6); + +static inline void +mlxsw_reg_ratr_pack(char *payload, + enum mlxsw_reg_ratr_op op, bool valid, + u32 adjacency_index, u16 egress_rif) +{ + MLXSW_REG_ZERO(ratr, payload); + mlxsw_reg_ratr_op_set(payload, op); + mlxsw_reg_ratr_v_set(payload, valid); + mlxsw_reg_ratr_adjacency_index_low_set(payload, adjacency_index); + mlxsw_reg_ratr_adjacency_index_high_set(payload, adjacency_index >> 16); + mlxsw_reg_ratr_egress_router_interface_set(payload, egress_rif); +} + +static inline void mlxsw_reg_ratr_eth_entry_pack(char *payload, + const char *dest_mac) +{ + mlxsw_reg_ratr_eth_destination_mac_memcpy_to(payload, dest_mac); +} + /* RALTA - Router Algorithmic LPM Tree Allocation Register * ------------------------------------------------------- * RALTA is used to allocate the LPM trees of the SHSPM method. @@ -4910,6 +5041,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "RGCR"; case MLXSW_REG_RITR_ID: return "RITR"; + case MLXSW_REG_RATR_ID: + return "RATR"; case MLXSW_REG_RALTA_ID: return "RALTA"; case MLXSW_REG_RALST_ID: -- cgit v0.10.2 From a59f0b312a197d695a6d99469114d2134b5befb2 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 Jul 2016 11:27:49 +0200 Subject: mlxsw: reg: Add Router Algorithmic LPM ECMP Update Register The RALEU register is used to mass update remote action adjacency index and ecmp size. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index a2d7870..0cc1485 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -4154,6 +4154,73 @@ static inline void mlxsw_reg_rauht_pack4(char *payload, mlxsw_reg_rauht_dip4_set(payload, dip); } +/* RALEU - Router Algorithmic LPM ECMP Update Register + * --------------------------------------------------- + * The register enables updating the ECMP section in the action for multiple + * LPM Unicast entries in a single operation. The update is executed to + * all entries of a {virtual router, protocol} tuple using the same ECMP group. + */ +#define MLXSW_REG_RALEU_ID 0x8015 +#define MLXSW_REG_RALEU_LEN 0x28 + +static const struct mlxsw_reg_info mlxsw_reg_raleu = { + .id = MLXSW_REG_RALEU_ID, + .len = MLXSW_REG_RALEU_LEN, +}; + +/* reg_raleu_protocol + * Protocol. + * Access: Index + */ +MLXSW_ITEM32(reg, raleu, protocol, 0x00, 24, 4); + +/* reg_raleu_virtual_router + * Virtual Router ID + * Range is 0..cap_max_virtual_routers-1 + * Access: Index + */ +MLXSW_ITEM32(reg, raleu, virtual_router, 0x00, 0, 16); + +/* reg_raleu_adjacency_index + * Adjacency Index used for matching on the existing entries. + * Access: Index + */ +MLXSW_ITEM32(reg, raleu, adjacency_index, 0x10, 0, 24); + +/* reg_raleu_ecmp_size + * ECMP Size used for matching on the existing entries. + * Access: Index + */ +MLXSW_ITEM32(reg, raleu, ecmp_size, 0x14, 0, 13); + +/* reg_raleu_new_adjacency_index + * New Adjacency Index. + * Access: WO + */ +MLXSW_ITEM32(reg, raleu, new_adjacency_index, 0x20, 0, 24); + +/* reg_raleu_new_ecmp_size + * New ECMP Size. + * Access: WO + */ +MLXSW_ITEM32(reg, raleu, new_ecmp_size, 0x24, 0, 13); + +static inline void mlxsw_reg_raleu_pack(char *payload, + enum mlxsw_reg_ralxx_protocol protocol, + u16 virtual_router, + u32 adjacency_index, u16 ecmp_size, + u32 new_adjacency_index, + u16 new_ecmp_size) +{ + MLXSW_REG_ZERO(raleu, payload); + mlxsw_reg_raleu_protocol_set(payload, protocol); + mlxsw_reg_raleu_virtual_router_set(payload, virtual_router); + mlxsw_reg_raleu_adjacency_index_set(payload, adjacency_index); + mlxsw_reg_raleu_ecmp_size_set(payload, ecmp_size); + mlxsw_reg_raleu_new_adjacency_index_set(payload, new_adjacency_index); + mlxsw_reg_raleu_new_ecmp_size_set(payload, new_ecmp_size); +} + /* RAUHTD - Router Algorithmic LPM Unicast Host Table Dump Register * ---------------------------------------------------------------- * The RAUHTD register allows dumping entries from the Router Unicast Host @@ -5053,6 +5120,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "RALUE"; case MLXSW_REG_RAUHT_ID: return "RAUHT"; + case MLXSW_REG_RALEU_ID: + return "RALEU"; case MLXSW_REG_RAUHTD_ID: return "RAUHTD"; case MLXSW_REG_MFCR_ID: -- cgit v0.10.2 From a7ff87acd995e3c024f3262bf90c8682c99b1f6b Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 Jul 2016 11:27:50 +0200 Subject: mlxsw: spectrum_router: Implement next-hop routing Implement next-hop routing offload including ECMP. To make it possible, introduce next-hop group entity. This entity keeps track of resolved neighbours and updates HW adjacency table accordingly. Note that HW next-hops are stored in this adjacency table, in form of MAC. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index e781128..0fe6051 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -222,6 +222,7 @@ struct mlxsw_sp_router { struct delayed_work dw; unsigned long interval; /* ms */ } neighs_update; + struct list_head nexthop_group_list; }; struct mlxsw_sp { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index ed0e6c0..dc13178 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -117,6 +117,8 @@ enum mlxsw_sp_fib_entry_type { MLXSW_SP_FIB_ENTRY_TYPE_TRAP, }; +struct mlxsw_sp_nexthop_group; + struct mlxsw_sp_fib_entry { struct rhash_head ht_node; struct mlxsw_sp_fib_key key; @@ -124,6 +126,8 @@ struct mlxsw_sp_fib_entry { u8 added:1; u16 rif; /* used for action local */ struct mlxsw_sp_vr *vr; + struct list_head nexthop_group_node; + struct mlxsw_sp_nexthop_group *nh_group; }; struct mlxsw_sp_fib { @@ -563,6 +567,9 @@ struct mlxsw_sp_neigh_entry { struct delayed_work dw; struct mlxsw_sp_port *mlxsw_sp_port; unsigned char ha[ETH_ALEN]; + struct list_head nexthop_list; /* list of nexthops using + * this neigh entry + */ }; static const struct rhashtable_params mlxsw_sp_neigh_ht_params = { @@ -606,6 +613,7 @@ mlxsw_sp_neigh_entry_create(const void *addr, size_t addr_len, neigh_entry->rif = rif; neigh_entry->n = n; INIT_DELAYED_WORK(&neigh_entry->dw, mlxsw_sp_router_neigh_update_hw); + INIT_LIST_HEAD(&neigh_entry->nexthop_list); return neigh_entry; } @@ -808,6 +816,11 @@ static void mlxsw_sp_router_neighs_update_work(struct work_struct *work) mlxsw_sp_router_neighs_update_work_schedule(mlxsw_sp); } +static void +mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_neigh_entry *neigh_entry, + bool removing); + static void mlxsw_sp_router_neigh_update_hw(struct work_struct *work) { struct mlxsw_sp_neigh_entry *neigh_entry = @@ -849,6 +862,7 @@ static void mlxsw_sp_router_neigh_update_hw(struct work_struct *work) } else { neigh_entry->offloaded = true; } + mlxsw_sp_nexthop_neigh_update(mlxsw_sp, neigh_entry, false); } else if (removing) { mlxsw_reg_rauht_pack4(rauht_pl, MLXSW_REG_RAUHT_OP_WRITE_DELETE, neigh_entry->rif, @@ -861,6 +875,7 @@ static void mlxsw_sp_router_neigh_update_hw(struct work_struct *work) } else { neigh_entry->offloaded = false; } + mlxsw_sp_nexthop_neigh_update(mlxsw_sp, neigh_entry, true); } neigh_release(n); @@ -978,6 +993,434 @@ static void mlxsw_sp_neigh_fini(struct mlxsw_sp *mlxsw_sp) rhashtable_destroy(&mlxsw_sp->router.neigh_ht); } +struct mlxsw_sp_nexthop { + struct list_head neigh_list_node; /* member of neigh entry list */ + struct mlxsw_sp_nexthop_group *nh_grp; /* pointer back to the group + * this belongs to + */ + u8 should_offload:1, /* set indicates this neigh is connected and + * should be put to KVD linear area of this group. + */ + offloaded:1, /* set in case the neigh is actually put into + * KVD linear area of this group. + */ + update:1; /* set indicates that MAC of this neigh should be + * updated in HW + */ + struct mlxsw_sp_neigh_entry *neigh_entry; +}; + +struct mlxsw_sp_nexthop_group { + struct list_head list; /* node in mlxsw->router.nexthop_group_list */ + struct list_head fib_list; /* list of fib entries that use this group */ + u8 adj_index_valid:1; + u32 adj_index; + u16 ecmp_size; + u16 count; + struct mlxsw_sp_nexthop nexthops[0]; +}; + +static int mlxsw_sp_adj_index_mass_update_vr(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_vr *vr, + u32 adj_index, u16 ecmp_size, + u32 new_adj_index, + u16 new_ecmp_size) +{ + char raleu_pl[MLXSW_REG_RALEU_LEN]; + + mlxsw_reg_raleu_pack(raleu_pl, vr->proto, vr->id, + adj_index, ecmp_size, + new_adj_index, new_ecmp_size); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(raleu), raleu_pl); +} + +static int mlxsw_sp_adj_index_mass_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp, + u32 old_adj_index, u16 old_ecmp_size) +{ + struct mlxsw_sp_fib_entry *fib_entry; + struct mlxsw_sp_vr *vr = NULL; + int err; + + list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) { + if (vr == fib_entry->vr) + continue; + vr = fib_entry->vr; + err = mlxsw_sp_adj_index_mass_update_vr(mlxsw_sp, vr, + old_adj_index, + old_ecmp_size, + nh_grp->adj_index, + nh_grp->ecmp_size); + if (err) + return err; + } + return 0; +} + +static int mlxsw_sp_nexthop_mac_update(struct mlxsw_sp *mlxsw_sp, u32 adj_index, + struct mlxsw_sp_nexthop *nh) +{ + struct mlxsw_sp_neigh_entry *neigh_entry = nh->neigh_entry; + char ratr_pl[MLXSW_REG_RATR_LEN]; + + mlxsw_reg_ratr_pack(ratr_pl, MLXSW_REG_RATR_OP_WRITE_WRITE_ENTRY, + true, adj_index, neigh_entry->rif); + mlxsw_reg_ratr_eth_entry_pack(ratr_pl, neigh_entry->ha); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ratr), ratr_pl); +} + +static int +mlxsw_sp_nexthop_group_mac_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + u32 adj_index = nh_grp->adj_index; /* base */ + struct mlxsw_sp_nexthop *nh; + int i; + int err; + + for (i = 0; i < nh_grp->count; i++) { + nh = &nh_grp->nexthops[i]; + + if (!nh->should_offload) { + nh->offloaded = 0; + continue; + } + + if (nh->update) { + err = mlxsw_sp_nexthop_mac_update(mlxsw_sp, + adj_index, nh); + if (err) + return err; + nh->update = 0; + nh->offloaded = 1; + } + adj_index++; + } + return 0; +} + +static int mlxsw_sp_fib_entry_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry); + +static int +mlxsw_sp_nexthop_fib_entries_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + struct mlxsw_sp_fib_entry *fib_entry; + int err; + + list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) { + err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry); + if (err) + return err; + } + return 0; +} + +static void +mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + struct mlxsw_sp_nexthop *nh; + bool offload_change = false; + u32 adj_index; + u16 ecmp_size = 0; + bool old_adj_index_valid; + u32 old_adj_index; + u16 old_ecmp_size; + int ret; + int i; + int err; + + for (i = 0; i < nh_grp->count; i++) { + nh = &nh_grp->nexthops[i]; + + if (nh->should_offload ^ nh->offloaded) { + offload_change = true; + if (nh->should_offload) + nh->update = 1; + } + if (nh->should_offload) + ecmp_size++; + } + if (!offload_change) { + /* Nothing was added or removed, so no need to reallocate. Just + * update MAC on existing adjacency indexes. + */ + err = mlxsw_sp_nexthop_group_mac_update(mlxsw_sp, nh_grp); + if (err) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to update neigh MAC in adjacency table.\n"); + goto set_trap; + } + return; + } + if (!ecmp_size) + /* No neigh of this group is connected so we just set + * the trap and let everthing flow through kernel. + */ + goto set_trap; + + ret = mlxsw_sp_kvdl_alloc(mlxsw_sp, ecmp_size); + if (ret < 0) { + /* We ran out of KVD linear space, just set the + * trap and let everything flow through kernel. + */ + dev_warn(mlxsw_sp->bus_info->dev, "Failed to allocate KVD linear area for nexthop group.\n"); + goto set_trap; + } + adj_index = ret; + old_adj_index_valid = nh_grp->adj_index_valid; + old_adj_index = nh_grp->adj_index; + old_ecmp_size = nh_grp->ecmp_size; + nh_grp->adj_index_valid = 1; + nh_grp->adj_index = adj_index; + nh_grp->ecmp_size = ecmp_size; + err = mlxsw_sp_nexthop_group_mac_update(mlxsw_sp, nh_grp); + if (err) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to update neigh MAC in adjacency table.\n"); + goto set_trap; + } + + if (!old_adj_index_valid) { + /* The trap was set for fib entries, so we have to call + * fib entry update to unset it and use adjacency index. + */ + err = mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, nh_grp); + if (err) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to add adjacency index to fib entries.\n"); + goto set_trap; + } + return; + } + + err = mlxsw_sp_adj_index_mass_update(mlxsw_sp, nh_grp, + old_adj_index, old_ecmp_size); + mlxsw_sp_kvdl_free(mlxsw_sp, old_adj_index); + if (err) { + dev_warn(mlxsw_sp->bus_info->dev, "Failed to mass-update adjacency index for nexthop group.\n"); + goto set_trap; + } + return; + +set_trap: + old_adj_index_valid = nh_grp->adj_index_valid; + nh_grp->adj_index_valid = 0; + for (i = 0; i < nh_grp->count; i++) { + nh = &nh_grp->nexthops[i]; + nh->offloaded = 0; + } + err = mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, nh_grp); + if (err) + dev_warn(mlxsw_sp->bus_info->dev, "Failed to set traps for fib entries.\n"); + if (old_adj_index_valid) + mlxsw_sp_kvdl_free(mlxsw_sp, nh_grp->adj_index); +} + +static void __mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp_nexthop *nh, + bool removing) +{ + if (!removing && !nh->should_offload) + nh->should_offload = 1; + else if (removing && nh->offloaded) + nh->should_offload = 0; + nh->update = 1; +} + +static void +mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_neigh_entry *neigh_entry, + bool removing) +{ + struct mlxsw_sp_nexthop *nh; + + /* Take RTNL mutex here to prevent lists from changes */ + rtnl_lock(); + list_for_each_entry(nh, &neigh_entry->nexthop_list, + neigh_list_node) { + __mlxsw_sp_nexthop_neigh_update(nh, removing); + mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh->nh_grp); + } + rtnl_unlock(); +} + +static int mlxsw_sp_nexthop_init(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp, + struct mlxsw_sp_nexthop *nh, + struct fib_nh *fib_nh) +{ + struct mlxsw_sp_neigh_entry *neigh_entry; + u32 gwip = ntohl(fib_nh->nh_gw); + struct net_device *dev = fib_nh->nh_dev; + struct neighbour *n; + u8 nud_state; + + neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, &gwip, + sizeof(gwip), dev); + if (!neigh_entry) { + __be32 gwipn = htonl(gwip); + + n = neigh_create(&arp_tbl, &gwipn, dev); + if (IS_ERR(n)) + return PTR_ERR(n); + neigh_event_send(n, NULL); + neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, &gwip, + sizeof(gwip), dev); + if (!neigh_entry) { + neigh_release(n); + return -EINVAL; + } + } else { + /* Take a reference of neigh here ensuring that neigh would + * not be detructed before the nexthop entry is finished. + * The second branch takes the reference in neith_create() + */ + n = neigh_entry->n; + neigh_clone(n); + } + nh->nh_grp = nh_grp; + nh->neigh_entry = neigh_entry; + list_add_tail(&nh->neigh_list_node, &neigh_entry->nexthop_list); + read_lock_bh(&n->lock); + nud_state = n->nud_state; + read_unlock_bh(&n->lock); + __mlxsw_sp_nexthop_neigh_update(nh, !(nud_state & NUD_VALID)); + + return 0; +} + +static void mlxsw_sp_nexthop_fini(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop *nh) +{ + struct mlxsw_sp_neigh_entry *neigh_entry = nh->neigh_entry; + + list_del(&nh->neigh_list_node); + neigh_release(neigh_entry->n); +} + +static struct mlxsw_sp_nexthop_group * +mlxsw_sp_nexthop_group_create(struct mlxsw_sp *mlxsw_sp, struct fib_info *fi) +{ + struct mlxsw_sp_nexthop_group *nh_grp; + struct mlxsw_sp_nexthop *nh; + struct fib_nh *fib_nh; + size_t alloc_size; + int i; + int err; + + alloc_size = sizeof(*nh_grp) + + fi->fib_nhs * sizeof(struct mlxsw_sp_nexthop); + nh_grp = kzalloc(alloc_size, GFP_KERNEL); + if (!nh_grp) + return ERR_PTR(-ENOMEM); + INIT_LIST_HEAD(&nh_grp->fib_list); + nh_grp->count = fi->fib_nhs; + for (i = 0; i < nh_grp->count; i++) { + nh = &nh_grp->nexthops[i]; + fib_nh = &fi->fib_nh[i]; + err = mlxsw_sp_nexthop_init(mlxsw_sp, nh_grp, nh, fib_nh); + if (err) + goto err_nexthop_init; + } + list_add_tail(&nh_grp->list, &mlxsw_sp->router.nexthop_group_list); + mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh_grp); + return nh_grp; + +err_nexthop_init: + for (i--; i >= 0; i--) + mlxsw_sp_nexthop_fini(mlxsw_sp, nh); + kfree(nh_grp); + return ERR_PTR(err); +} + +static void +mlxsw_sp_nexthop_group_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + struct mlxsw_sp_nexthop *nh; + int i; + + list_del(&nh_grp->list); + for (i = 0; i < nh_grp->count; i++) { + nh = &nh_grp->nexthops[i]; + mlxsw_sp_nexthop_fini(mlxsw_sp, nh); + } + kfree(nh_grp); +} + +static bool mlxsw_sp_nexthop_match(struct mlxsw_sp_nexthop *nh, + struct fib_info *fi) +{ + int i; + + for (i = 0; i < fi->fib_nhs; i++) { + struct fib_nh *fib_nh = &fi->fib_nh[i]; + u32 gwip = ntohl(fib_nh->nh_gw); + + if (memcmp(nh->neigh_entry->key.addr, + &gwip, sizeof(u32)) == 0 && + nh->neigh_entry->key.dev == fib_nh->nh_dev) + return true; + } + return false; +} + +static bool mlxsw_sp_nexthop_group_match(struct mlxsw_sp_nexthop_group *nh_grp, + struct fib_info *fi) +{ + int i; + + if (nh_grp->count != fi->fib_nhs) + return false; + for (i = 0; i < nh_grp->count; i++) { + struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i]; + + if (!mlxsw_sp_nexthop_match(nh, fi)) + return false; + } + return true; +} + +static struct mlxsw_sp_nexthop_group * +mlxsw_sp_nexthop_group_find(struct mlxsw_sp *mlxsw_sp, struct fib_info *fi) +{ + struct mlxsw_sp_nexthop_group *nh_grp; + + list_for_each_entry(nh_grp, &mlxsw_sp->router.nexthop_group_list, + list) { + if (mlxsw_sp_nexthop_group_match(nh_grp, fi)) + return nh_grp; + } + return NULL; +} + +static int mlxsw_sp_nexthop_group_get(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + struct fib_info *fi) +{ + struct mlxsw_sp_nexthop_group *nh_grp; + + nh_grp = mlxsw_sp_nexthop_group_find(mlxsw_sp, fi); + if (!nh_grp) { + nh_grp = mlxsw_sp_nexthop_group_create(mlxsw_sp, fi); + if (IS_ERR(nh_grp)) + return PTR_ERR(nh_grp); + } + list_add_tail(&fib_entry->nexthop_group_node, &nh_grp->fib_list); + fib_entry->nh_group = nh_grp; + return 0; +} + +static void mlxsw_sp_nexthop_group_put(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) +{ + struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group; + + list_del(&fib_entry->nexthop_group_node); + if (!list_empty(&nh_grp->fib_list)) + return; + mlxsw_sp_nexthop_group_destroy(mlxsw_sp, nh_grp); +} + static int __mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) { char rgcr_pl[MLXSW_REG_RGCR_LEN]; @@ -999,6 +1442,7 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) { int err; + INIT_LIST_HEAD(&mlxsw_sp->router.nexthop_group_list); err = __mlxsw_sp_router_init(mlxsw_sp); if (err) return err; @@ -1013,6 +1457,38 @@ void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) __mlxsw_sp_router_fini(mlxsw_sp); } +static int mlxsw_sp_fib_entry_op4_remote(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry, + enum mlxsw_reg_ralue_op op) +{ + char ralue_pl[MLXSW_REG_RALUE_LEN]; + u32 *p_dip = (u32 *) fib_entry->key.addr; + struct mlxsw_sp_vr *vr = fib_entry->vr; + enum mlxsw_reg_ralue_trap_action trap_action; + u16 trap_id = 0; + u32 adjacency_index = 0; + u16 ecmp_size = 0; + + /* In case the nexthop group adjacency index is valid, use it + * with provided ECMP size. Otherwise, setup trap and pass + * traffic to kernel. + */ + if (fib_entry->nh_group->adj_index_valid) { + trap_action = MLXSW_REG_RALUE_TRAP_ACTION_NOP; + adjacency_index = fib_entry->nh_group->adj_index; + ecmp_size = fib_entry->nh_group->ecmp_size; + } else { + trap_action = MLXSW_REG_RALUE_TRAP_ACTION_TRAP; + trap_id = MLXSW_TRAP_ID_RTR_INGRESS0; + } + + mlxsw_reg_ralue_pack4(ralue_pl, vr->proto, op, vr->id, + fib_entry->key.prefix_len, *p_dip); + mlxsw_reg_ralue_act_remote_pack(ralue_pl, trap_action, trap_id, + adjacency_index, ecmp_size); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue), ralue_pl); +} + static int mlxsw_sp_fib_entry_op4_local(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib_entry *fib_entry, enum mlxsw_reg_ralue_op op) @@ -1049,7 +1525,7 @@ static int mlxsw_sp_fib_entry_op4(struct mlxsw_sp *mlxsw_sp, { switch (fib_entry->type) { case MLXSW_SP_FIB_ENTRY_TYPE_REMOTE: - return -EINVAL; + return mlxsw_sp_fib_entry_op4_remote(mlxsw_sp, fib_entry, op); case MLXSW_SP_FIB_ENTRY_TYPE_LOCAL: return mlxsw_sp_fib_entry_op4_local(mlxsw_sp, fib_entry, op); case MLXSW_SP_FIB_ENTRY_TYPE_TRAP: @@ -1129,7 +1605,17 @@ mlxsw_sp_router_fib4_entry_init(struct mlxsw_sp *mlxsw_sp, fib_entry->rif = r->rif; return 0; } - return -EINVAL; + fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_REMOTE; + return mlxsw_sp_nexthop_group_get(mlxsw_sp, fib_entry, fi); +} + +static void +mlxsw_sp_router_fib4_entry_fini(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_fib_entry *fib_entry) +{ + if (fib_entry->type != MLXSW_SP_FIB_ENTRY_TYPE_REMOTE) + return; + mlxsw_sp_nexthop_group_put(mlxsw_sp, fib_entry); } static int @@ -1173,6 +1659,7 @@ mlxsw_sp_router_fib4_add_prepare(struct mlxsw_sp_port *mlxsw_sp_port, return 0; err_alloc_info: + mlxsw_sp_router_fib4_entry_fini(mlxsw_sp, fib_entry); err_fib4_entry_init: mlxsw_sp_fib_entry_destroy(fib_entry); err_fib_entry_create: @@ -1207,6 +1694,7 @@ mlxsw_sp_router_fib4_add_commit(struct mlxsw_sp_port *mlxsw_sp_port, err_fib_entry_add: mlxsw_sp_fib_entry_remove(vr->fib, fib_entry); err_fib_entry_insert: + mlxsw_sp_router_fib4_entry_fini(mlxsw_sp, fib_entry); mlxsw_sp_fib_entry_destroy(fib_entry); mlxsw_sp_vr_put(mlxsw_sp, vr); return err; @@ -1243,6 +1731,7 @@ int mlxsw_sp_router_fib4_del(struct mlxsw_sp_port *mlxsw_sp_port, } mlxsw_sp_fib_entry_del(mlxsw_sp_port->mlxsw_sp, fib_entry); mlxsw_sp_fib_entry_remove(vr->fib, fib_entry); + mlxsw_sp_router_fib4_entry_fini(mlxsw_sp, fib_entry); mlxsw_sp_fib_entry_destroy(fib_entry); mlxsw_sp_vr_put(mlxsw_sp, vr); return 0; -- cgit v0.10.2 From b2157149b0b0f2cdaad9b82ed3bb1b84f82aa7f1 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Tue, 5 Jul 2016 11:27:51 +0200 Subject: mlxsw: spectrum_router: Add the nexthop neigh activity update For nexthop neighbours we need to make kernel to think there is a traffic flowing to them preventing it from going to stale state. Otherwise kernel would stale it and eventually the neigh would be removed from HW and nexthop as well. That would reduce ECMP group in HW. Signed-off-by: Yotam Gigi Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 0fe6051..ff5b859 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -223,6 +223,7 @@ struct mlxsw_sp_router { unsigned long interval; /* ms */ } neighs_update; struct list_head nexthop_group_list; + struct list_head nexthop_neighs_list; }; struct mlxsw_sp { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index dc13178..2b20279 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -570,6 +570,7 @@ struct mlxsw_sp_neigh_entry { struct list_head nexthop_list; /* list of nexthops using * this neigh entry */ + struct list_head nexthop_neighs_list_node; }; static const struct rhashtable_params mlxsw_sp_neigh_ht_params = { @@ -770,28 +771,15 @@ static void mlxsw_sp_router_neigh_rec_process(struct mlxsw_sp *mlxsw_sp, } } -static void -mlxsw_sp_router_neighs_update_work_schedule(struct mlxsw_sp *mlxsw_sp) +static int mlxsw_sp_router_neighs_update_rauhtd(struct mlxsw_sp *mlxsw_sp) { - unsigned long interval = mlxsw_sp->router.neighs_update.interval; - - mlxsw_core_schedule_dw(&mlxsw_sp->router.neighs_update.dw, - msecs_to_jiffies(interval)); -} - -static void mlxsw_sp_router_neighs_update_work(struct work_struct *work) -{ - struct mlxsw_sp *mlxsw_sp; char *rauhtd_pl; u8 num_rec; int i, err; rauhtd_pl = kmalloc(MLXSW_REG_RAUHTD_LEN, GFP_KERNEL); if (!rauhtd_pl) - return; - - mlxsw_sp = container_of(work, struct mlxsw_sp, - router.neighs_update.dw.work); + return -ENOMEM; /* Make sure the neighbour's netdev isn't removed in the * process. @@ -813,6 +801,47 @@ static void mlxsw_sp_router_neighs_update_work(struct work_struct *work) rtnl_unlock(); kfree(rauhtd_pl); + return err; +} + +static void mlxsw_sp_router_neighs_update_nh(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_sp_neigh_entry *neigh_entry; + + /* Take RTNL mutex here to prevent lists from changes */ + rtnl_lock(); + list_for_each_entry(neigh_entry, &mlxsw_sp->router.nexthop_neighs_list, + nexthop_neighs_list_node) { + /* If this neigh have nexthops, make the kernel think this neigh + * is active regardless of the traffic. + */ + if (!list_empty(&neigh_entry->nexthop_list)) + neigh_event_send(neigh_entry->n, NULL); + } + rtnl_unlock(); +} + +static void +mlxsw_sp_router_neighs_update_work_schedule(struct mlxsw_sp *mlxsw_sp) +{ + unsigned long interval = mlxsw_sp->router.neighs_update.interval; + + mlxsw_core_schedule_dw(&mlxsw_sp->router.neighs_update.dw, + msecs_to_jiffies(interval)); +} + +static void mlxsw_sp_router_neighs_update_work(struct work_struct *work) +{ + struct mlxsw_sp *mlxsw_sp = container_of(work, struct mlxsw_sp, + router.neighs_update.dw.work); + int err; + + err = mlxsw_sp_router_neighs_update_rauhtd(mlxsw_sp); + if (err) + dev_err(mlxsw_sp->bus_info->dev, "Could not update kernel for neigh activity"); + + mlxsw_sp_router_neighs_update_nh(mlxsw_sp); + mlxsw_sp_router_neighs_update_work_schedule(mlxsw_sp); } @@ -1277,6 +1306,14 @@ static int mlxsw_sp_nexthop_init(struct mlxsw_sp *mlxsw_sp, n = neigh_entry->n; neigh_clone(n); } + + /* If that is the first nexthop connected to that neigh, add to + * nexthop_neighs_list + */ + if (list_empty(&neigh_entry->nexthop_list)) + list_add_tail(&neigh_entry->nexthop_neighs_list_node, + &mlxsw_sp->router.nexthop_neighs_list); + nh->nh_grp = nh_grp; nh->neigh_entry = neigh_entry; list_add_tail(&nh->neigh_list_node, &neigh_entry->nexthop_list); @@ -1294,6 +1331,13 @@ static void mlxsw_sp_nexthop_fini(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_neigh_entry *neigh_entry = nh->neigh_entry; list_del(&nh->neigh_list_node); + + /* If that is the last nexthop connected to that neigh, remove from + * nexthop_neighs_list + */ + if (list_empty(&nh->neigh_entry->nexthop_list)) + list_del(&nh->neigh_entry->nexthop_neighs_list_node); + neigh_release(neigh_entry->n); } @@ -1442,6 +1486,7 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp) { int err; + INIT_LIST_HEAD(&mlxsw_sp->router.nexthop_neighs_list); INIT_LIST_HEAD(&mlxsw_sp->router.nexthop_group_list); err = __mlxsw_sp_router_init(mlxsw_sp); if (err) -- cgit v0.10.2 From 0b2361d9d946558c90f0ae8b21725022bea9f2cb Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Tue, 5 Jul 2016 11:27:52 +0200 Subject: mlxsw: Add the unresolved next-hops probes Now, the driver sends arp probes for all unresolved neighbours that are currently a nexthop for some route on the system. The job is set periodically every 5 seconds. Signed-off-by: Yotam Gigi Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index ff5b859..ef4ac89 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -222,6 +222,8 @@ struct mlxsw_sp_router { struct delayed_work dw; unsigned long interval; /* ms */ } neighs_update; + struct delayed_work nexthop_probe_dw; +#define MLXSW_SP_UNRESOLVED_NH_PROBE_INTERVAL 5000 /* ms */ struct list_head nexthop_group_list; struct list_head nexthop_neighs_list; }; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 2b20279..e084ea5 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -845,6 +845,33 @@ static void mlxsw_sp_router_neighs_update_work(struct work_struct *work) mlxsw_sp_router_neighs_update_work_schedule(mlxsw_sp); } +static void mlxsw_sp_router_probe_unresolved_nexthops(struct work_struct *work) +{ + struct mlxsw_sp_neigh_entry *neigh_entry; + struct mlxsw_sp *mlxsw_sp = container_of(work, struct mlxsw_sp, + router.nexthop_probe_dw.work); + + /* Iterate over nexthop neighbours, find those who are unresolved and + * send arp on them. This solves the chicken-egg problem when + * the nexthop wouldn't get offloaded until the neighbor is resolved + * but it wouldn't get resolved ever in case traffic is flowing in HW + * using different nexthop. + * + * Take RTNL mutex here to prevent lists from changes. + */ + rtnl_lock(); + list_for_each_entry(neigh_entry, &mlxsw_sp->router.nexthop_neighs_list, + nexthop_neighs_list_node) { + if (!(neigh_entry->n->nud_state & NUD_VALID) && + !list_empty(&neigh_entry->nexthop_list)) + neigh_event_send(neigh_entry->n, NULL); + } + rtnl_unlock(); + + mlxsw_core_schedule_dw(&mlxsw_sp->router.nexthop_probe_dw, + MLXSW_SP_UNRESOLVED_NH_PROBE_INTERVAL); +} + static void mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_neigh_entry *neigh_entry, @@ -1004,10 +1031,13 @@ static int mlxsw_sp_neigh_init(struct mlxsw_sp *mlxsw_sp) if (err) goto err_register_netevent_notifier; + /* Create the delayed works for the activity_update */ INIT_DELAYED_WORK(&mlxsw_sp->router.neighs_update.dw, mlxsw_sp_router_neighs_update_work); + INIT_DELAYED_WORK(&mlxsw_sp->router.nexthop_probe_dw, + mlxsw_sp_router_probe_unresolved_nexthops); mlxsw_core_schedule_dw(&mlxsw_sp->router.neighs_update.dw, 0); - + mlxsw_core_schedule_dw(&mlxsw_sp->router.nexthop_probe_dw, 0); return 0; err_register_netevent_notifier: @@ -1018,6 +1048,7 @@ err_register_netevent_notifier: static void mlxsw_sp_neigh_fini(struct mlxsw_sp *mlxsw_sp) { cancel_delayed_work_sync(&mlxsw_sp->router.neighs_update.dw); + cancel_delayed_work_sync(&mlxsw_sp->router.nexthop_probe_dw); unregister_netevent_notifier(&mlxsw_sp_router_netevent_nb); rhashtable_destroy(&mlxsw_sp->router.neigh_ht); } -- cgit v0.10.2 From 96329a181bfbbac573ce988d7439369c5ade9249 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 5 Jul 2016 07:56:52 +0000 Subject: net: hns: fix return value check in hns_dsaf_get_cfg() In case of error, function devm_ioremap_resource() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR(). Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c index 86ce28a..2ef4277 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c @@ -114,9 +114,9 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) dsaf_dev->sc_base = devm_ioremap_resource(&pdev->dev, res); - if (!dsaf_dev->sc_base) { + if (IS_ERR(dsaf_dev->sc_base)) { dev_err(dsaf_dev->dev, "subctrl can not map!\n"); - return -ENOMEM; + return PTR_ERR(dsaf_dev->sc_base); } res = platform_get_resource(pdev, IORESOURCE_MEM, @@ -128,9 +128,9 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) dsaf_dev->sds_base = devm_ioremap_resource(&pdev->dev, res); - if (!dsaf_dev->sds_base) { + if (IS_ERR(dsaf_dev->sds_base)) { dev_err(dsaf_dev->dev, "serdes-ctrl can not map!\n"); - return -ENOMEM; + return PTR_ERR(dsaf_dev->sds_base); } } else { dsaf_dev->sub_ctrl = syscon; @@ -146,9 +146,9 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) } } dsaf_dev->ppe_base = devm_ioremap_resource(&pdev->dev, res); - if (!dsaf_dev->ppe_base) { + if (IS_ERR(dsaf_dev->ppe_base)) { dev_err(dsaf_dev->dev, "ppe-base resource can not map!\n"); - return -ENOMEM; + return PTR_ERR(dsaf_dev->ppe_base); } dsaf_dev->ppe_paddr = res->start; @@ -165,9 +165,9 @@ int hns_dsaf_get_cfg(struct dsaf_device *dsaf_dev) } } dsaf_dev->io_base = devm_ioremap_resource(&pdev->dev, res); - if (!dsaf_dev->io_base) { + if (IS_ERR(dsaf_dev->io_base)) { dev_err(dsaf_dev->dev, "dsaf-base resource can not map!\n"); - return -ENOMEM; + return PTR_ERR(dsaf_dev->io_base); } } -- cgit v0.10.2 From 8297f2d9ef6c2c31c87807f2f110ddfd0c379443 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Mon, 4 Jul 2016 17:50:58 -0400 Subject: connector: make cn_proc explicitly non-modular The Kconfig controlling build of this code is currently: drivers/connector/Kconfig:config PROC_EVENTS drivers/connector/Kconfig: bool "Report process events to userspace" ...meaning that it currently is not being built as a module by anyone. Lets remove the two modular references, so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. Cc: Evgeniy Polyakov Cc: netdev@vger.kernel.org Signed-off-by: Paul Gortmaker Signed-off-by: David S. Miller diff --git a/drivers/connector/cn_proc.c b/drivers/connector/cn_proc.c index b02f9c6..a782ce8 100644 --- a/drivers/connector/cn_proc.c +++ b/drivers/connector/cn_proc.c @@ -22,7 +22,6 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#include #include #include #include @@ -390,5 +389,4 @@ static int __init cn_proc_init(void) } return 0; } - -module_init(cn_proc_init); +device_initcall(cn_proc_init); -- cgit v0.10.2 From 73e20b761acf8678de2d55d92b90a623b8558a77 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Mon, 4 Jul 2016 18:47:41 -0700 Subject: net: vrf: Add support for PREROUTING rules on vrf device Add support for PREROUTING rules with skb->dev set to the vrf device. INPUT rules are already allowed. Provides symmetry with the output path which allows POSTROUTING rules. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index b376282..1ce7420 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -779,6 +779,25 @@ static int vrf_get_saddr(struct net_device *dev, struct flowi4 *fl4) return rc; } +static int vrf_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb) +{ + return 0; +} + +static struct sk_buff *vrf_rcv_nfhook(u8 pf, unsigned int hook, + struct sk_buff *skb, + struct net_device *dev) +{ + struct net *net = dev_net(dev); + + nf_reset(skb); + + if (NF_HOOK(pf, hook, net, NULL, skb, dev, NULL, vrf_rcv_finish) < 0) + skb = NULL; /* kfree_skb(skb) handled by nf code */ + + return skb; +} + #if IS_ENABLED(CONFIG_IPV6) /* neighbor handling is done with actual device; do not want * to flip skb->dev for those ndisc packets. This really fails @@ -899,6 +918,7 @@ static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, if (need_strict) vrf_ip6_input_dst(skb, vrf_dev, orig_iif); + skb = vrf_rcv_nfhook(NFPROTO_IPV6, NF_INET_PRE_ROUTING, skb, vrf_dev); out: return skb; } @@ -929,6 +949,7 @@ static struct sk_buff *vrf_ip_rcv(struct net_device *vrf_dev, dev_queue_xmit_nit(skb, vrf_dev); skb_pull(skb, skb->mac_len); + skb = vrf_rcv_nfhook(NFPROTO_IPV4, NF_INET_PRE_ROUTING, skb, vrf_dev); out: return skb; } -- cgit v0.10.2 From 6d48fcd900803c7ddb9b3813176e2cdb4fcd6aaa Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 28 Apr 2016 14:39:22 +0200 Subject: iwlwifi: remove useless enum values Since the values of this enum are used only internally, we can let the compiler number them. Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c index 5c5b9f2..26afa72 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c @@ -129,8 +129,8 @@ struct iwl_drv { }; enum { - DVM_OP_MODE = 0, - MVM_OP_MODE = 1, + DVM_OP_MODE, + MVM_OP_MODE, }; /* Protects the table contents, i.e. the ops pointer & drv list */ -- cgit v0.10.2 From ca221c9b946cd4a9ea67375c8d90379a0e65179d Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 28 Apr 2016 14:40:59 +0200 Subject: iwlwifi: change fw.mvm_fw to fw.type Instead of explicitly indicating the difference between just DVM and MVM with an mvm_fw boolean change this to fw.type to be more extensible and easier to understand. Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c index 26afa72..e2df544 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c @@ -795,17 +795,17 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv, case IWL_UCODE_TLV_SEC_RT: iwl_store_ucode_sec(pieces, tlv_data, IWL_UCODE_REGULAR, tlv_len); - drv->fw.mvm_fw = true; + drv->fw.type = IWL_FW_MVM; break; case IWL_UCODE_TLV_SEC_INIT: iwl_store_ucode_sec(pieces, tlv_data, IWL_UCODE_INIT, tlv_len); - drv->fw.mvm_fw = true; + drv->fw.type = IWL_FW_MVM; break; case IWL_UCODE_TLV_SEC_WOWLAN: iwl_store_ucode_sec(pieces, tlv_data, IWL_UCODE_WOWLAN, tlv_len); - drv->fw.mvm_fw = true; + drv->fw.type = IWL_FW_MVM; break; case IWL_UCODE_TLV_DEF_CALIB: if (tlv_len != sizeof(struct iwl_tlv_calib_data)) @@ -827,17 +827,17 @@ static int iwl_parse_tlv_firmware(struct iwl_drv *drv, case IWL_UCODE_TLV_SECURE_SEC_RT: iwl_store_ucode_sec(pieces, tlv_data, IWL_UCODE_REGULAR, tlv_len); - drv->fw.mvm_fw = true; + drv->fw.type = IWL_FW_MVM; break; case IWL_UCODE_TLV_SECURE_SEC_INIT: iwl_store_ucode_sec(pieces, tlv_data, IWL_UCODE_INIT, tlv_len); - drv->fw.mvm_fw = true; + drv->fw.type = IWL_FW_MVM; break; case IWL_UCODE_TLV_SECURE_SEC_WOWLAN: iwl_store_ucode_sec(pieces, tlv_data, IWL_UCODE_WOWLAN, tlv_len); - drv->fw.mvm_fw = true; + drv->fw.type = IWL_FW_MVM; break; case IWL_UCODE_TLV_NUM_OF_CPU: if (tlv_len != sizeof(u32)) @@ -1272,7 +1272,7 @@ static void iwl_req_fw_callback(const struct firmware *ucode_raw, void *context) * In mvm uCode there is no difference between data and instructions * sections. */ - if (!fw->mvm_fw && validate_sec_sizes(drv, pieces, drv->cfg)) + if (fw->type == IWL_FW_DVM && validate_sec_sizes(drv, pieces, drv->cfg)) goto try_again; /* Allocate ucode buffers for card's bus-master loading ... */ @@ -1400,10 +1400,16 @@ static void iwl_req_fw_callback(const struct firmware *ucode_raw, void *context) release_firmware(ucode_raw); mutex_lock(&iwlwifi_opmode_table_mtx); - if (fw->mvm_fw) - op = &iwlwifi_opmode_table[MVM_OP_MODE]; - else + switch (fw->type) { + case IWL_FW_DVM: op = &iwlwifi_opmode_table[DVM_OP_MODE]; + break; + default: + WARN(1, "Invalid fw type %d\n", fw->type); + case IWL_FW_MVM: + op = &iwlwifi_opmode_table[MVM_OP_MODE]; + break; + } IWL_INFO(drv, "loaded firmware version %s op_mode %s\n", drv->fw.fw_version, op->name); diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw.h index e461d63..655ec52 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-fw.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw.h @@ -231,6 +231,16 @@ struct iwl_gscan_capabilities { }; /** + * enum iwl_fw_type - iwlwifi firmware type + * @IWL_FW_DVM: DVM firmware + * @IWL_FW_MVM: MVM firmware + */ +enum iwl_fw_type { + IWL_FW_DVM, + IWL_FW_MVM, +}; + +/** * struct iwl_fw - variables associated with the firmware * * @ucode_ver: ucode version from the ucode file @@ -244,7 +254,7 @@ struct iwl_gscan_capabilities { * @inst_evtlog_ptr: event log offset for runtime ucode. * @inst_evtlog_size: event log size for runtime ucode. * @inst_errlog_ptr: error log offfset for runtime ucode. - * @mvm_fw: indicates this is MVM firmware + * @type: firmware type (&enum iwl_fw_type) * @cipher_scheme: optional external cipher scheme. * @human_readable: human readable version * @sdio_adma_addr: the default address to set for the ADMA in SDIO mode until @@ -275,7 +285,7 @@ struct iwl_fw { u8 valid_tx_ant; u8 valid_rx_ant; - bool mvm_fw; + enum iwl_fw_type type; struct ieee80211_cipher_scheme cs[IWL_UCODE_MAX_CS]; u8 human_readable[FW_VER_HUMAN_READABLE_SZ]; -- cgit v0.10.2 From 9794c64f302d6d544acbb5ab69a327d694a70fcb Mon Sep 17 00:00:00 2001 From: Liad Kaufman Date: Wed, 19 Aug 2015 17:34:28 +0300 Subject: iwlwifi: mvm: support dqa queue inactivation upon timeout Support marking queues as inactive upon a timeout expiring, and allow inactive queues to be re-assigned to other RA/TIDs if no other queue is free. This is done by keeping a timestamp of the latest frame TXed for every RA/TID, and then going over the queues currently in use when a new queue is needed, inactivating all those that are inactive. Signed-off-by: Liad Kaufman Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index 4b75b92..bf7d78e 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -687,13 +687,22 @@ struct iwl_mvm_baid_data { * This is the state of a queue that has been fully configured (including * SCD pointers, etc), has a specific RA/TID assigned to it, and can be * used to send traffic. + * @IWL_MVM_QUEUE_INACTIVE: queue is allocated but no traffic on it + * This is a state of a queue that has had traffic on it, but during the + * last %IWL_MVM_DQA_QUEUE_TIMEOUT time period there has been no traffic on + * it. In this state, when a new queue is needed to be allocated but no + * such free queue exists, an inactive queue might be freed and given to + * the new RA/TID. */ enum iwl_mvm_queue_status { IWL_MVM_QUEUE_FREE, IWL_MVM_QUEUE_RESERVED, IWL_MVM_QUEUE_READY, + IWL_MVM_QUEUE_INACTIVE, }; +#define IWL_MVM_DQA_QUEUE_TIMEOUT (5 * HZ) + struct iwl_mvm { /* for logger access */ struct device *dev; @@ -750,11 +759,15 @@ struct iwl_mvm { u32 hw_queue_to_mac80211; u8 hw_queue_refcount; u8 ra_sta_id; /* The RA this queue is mapped to, if exists */ + bool reserved; /* Is this the TXQ reserved for a STA */ u16 tid_bitmap; /* Bitmap of the TIDs mapped to this queue */ + /* Timestamp for inactivation per TID of this queue */ + unsigned long last_frame_time[IWL_MAX_TID_COUNT + 1]; enum iwl_mvm_queue_status status; } queue_info[IWL_MAX_HW_QUEUES]; spinlock_t queue_info_lock; /* For syncing queue mgmt operations */ struct work_struct add_stream_wk; /* To add streams to queues */ + atomic_t mac80211_queue_stop_count[IEEE80211_MAX_QUEUES]; const char *nvm_file_name; @@ -1618,7 +1631,7 @@ void iwl_mvm_enable_txq(struct iwl_mvm *mvm, int queue, int mac80211_queue, */ void iwl_mvm_disable_txq(struct iwl_mvm *mvm, int queue, int mac80211_queue, u8 tid, u8 flags); -int iwl_mvm_find_free_queue(struct iwl_mvm *mvm, u8 minq, u8 maxq); +int iwl_mvm_find_free_queue(struct iwl_mvm *mvm, u8 sta_id, u8 minq, u8 maxq); /* Return a bitmask with all the hw supported queues, except for the * command queue, which can't be flushed. @@ -1725,6 +1738,8 @@ void iwl_mvm_sync_rx_queues_internal(struct iwl_mvm *mvm, void iwl_mvm_reorder_timer_expired(unsigned long data); struct ieee80211_vif *iwl_mvm_get_bss_vif(struct iwl_mvm *mvm); +void iwl_mvm_inactivity_check(struct iwl_mvm *mvm); + void iwl_mvm_nic_restart(struct iwl_mvm *mvm, bool fw_error); unsigned int iwl_mvm_get_wd_timeout(struct iwl_mvm *mvm, struct ieee80211_vif *vif, diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index 64b07b1..84384a43 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -310,6 +310,112 @@ static void iwl_mvm_tdls_sta_deinit(struct iwl_mvm *mvm, iwl_mvm_disable_txq(mvm, i, i, IWL_MAX_TID_COUNT, 0); } +/* Disable aggregations for a bitmap of TIDs for a given station */ +static int iwl_mvm_invalidate_sta_queue(struct iwl_mvm *mvm, int queue, + unsigned long disable_agg_tids, + bool remove_queue) +{ + struct iwl_mvm_add_sta_cmd cmd = {}; + struct ieee80211_sta *sta; + struct iwl_mvm_sta *mvmsta; + u32 status; + u8 sta_id; + int ret; + + spin_lock_bh(&mvm->queue_info_lock); + sta_id = mvm->queue_info[queue].ra_sta_id; + spin_unlock_bh(&mvm->queue_info_lock); + + rcu_read_lock(); + + sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]); + + if (WARN_ON_ONCE(IS_ERR_OR_NULL(sta))) { + rcu_read_unlock(); + return -EINVAL; + } + + mvmsta = iwl_mvm_sta_from_mac80211(sta); + + mvmsta->tid_disable_agg |= disable_agg_tids; + + cmd.mac_id_n_color = cpu_to_le32(mvmsta->mac_id_n_color); + cmd.sta_id = mvmsta->sta_id; + cmd.add_modify = STA_MODE_MODIFY; + cmd.modify_mask = STA_MODIFY_QUEUES; + if (disable_agg_tids) + cmd.modify_mask |= STA_MODIFY_TID_DISABLE_TX; + if (remove_queue) + cmd.modify_mask |= STA_MODIFY_QUEUE_REMOVAL; + cmd.tfd_queue_msk = cpu_to_le32(mvmsta->tfd_queue_msk); + cmd.tid_disable_tx = cpu_to_le16(mvmsta->tid_disable_agg); + + rcu_read_unlock(); + + /* Notify FW of queue removal from the STA queues */ + status = ADD_STA_SUCCESS; + ret = iwl_mvm_send_cmd_pdu_status(mvm, ADD_STA, + iwl_mvm_add_sta_cmd_size(mvm), + &cmd, &status); + + return ret; +} + +/* + * Remove a queue from a station's resources. + * Note that this only marks as free. It DOESN'T delete a BA agreement, and + * doesn't disable the queue + */ +static int iwl_mvm_remove_sta_queue_marking(struct iwl_mvm *mvm, int queue) +{ + struct ieee80211_sta *sta; + struct iwl_mvm_sta *mvmsta; + unsigned long tid_bitmap; + unsigned long disable_agg_tids = 0; + u8 sta_id; + int tid; + + lockdep_assert_held(&mvm->mutex); + + spin_lock_bh(&mvm->queue_info_lock); + sta_id = mvm->queue_info[queue].ra_sta_id; + tid_bitmap = mvm->queue_info[queue].tid_bitmap; + spin_unlock_bh(&mvm->queue_info_lock); + + rcu_read_lock(); + + sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]); + + if (WARN_ON_ONCE(IS_ERR_OR_NULL(sta))) { + rcu_read_unlock(); + return 0; + } + + mvmsta = iwl_mvm_sta_from_mac80211(sta); + + spin_lock_bh(&mvmsta->lock); + for_each_set_bit(tid, &tid_bitmap, IWL_MAX_TID_COUNT + 1) { + mvmsta->tid_data[tid].txq_id = IEEE80211_INVAL_HW_QUEUE; + + if (mvmsta->tid_data[tid].state == IWL_AGG_ON) + disable_agg_tids |= BIT(tid); + } + mvmsta->tfd_queue_msk &= ~BIT(queue); /* Don't use this queue anymore */ + + spin_unlock_bh(&mvmsta->lock); + + rcu_read_unlock(); + + spin_lock(&mvm->queue_info_lock); + /* Unmap MAC queues and TIDs from this queue */ + mvm->queue_info[queue].hw_queue_to_mac80211 = 0; + mvm->queue_info[queue].hw_queue_refcount = 0; + mvm->queue_info[queue].tid_bitmap = 0; + spin_unlock(&mvm->queue_info_lock); + + return disable_agg_tids; +} + static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, struct ieee80211_sta *sta, u8 ac, int tid, struct ieee80211_hdr *hdr) @@ -325,6 +431,9 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, iwl_mvm_get_wd_timeout(mvm, mvmsta->vif, false, false); u8 mac_queue = mvmsta->vif->hw_queue[ac]; int queue = -1; + bool using_inactive_queue = false; + unsigned long disable_agg_tids = 0; + enum iwl_mvm_agg_state queue_state; int ssn; int ret; @@ -338,7 +447,8 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, */ if (!ieee80211_is_data_qos(hdr->frame_control) || ieee80211_is_qos_nullfunc(hdr->frame_control)) { - queue = iwl_mvm_find_free_queue(mvm, IWL_MVM_DQA_MIN_MGMT_QUEUE, + queue = iwl_mvm_find_free_queue(mvm, mvmsta->sta_id, + IWL_MVM_DQA_MIN_MGMT_QUEUE, IWL_MVM_DQA_MAX_MGMT_QUEUE); if (queue >= IWL_MVM_DQA_MIN_MGMT_QUEUE) IWL_DEBUG_TX_QUEUES(mvm, "Found free MGMT queue #%d\n", @@ -347,16 +457,37 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, /* If no such queue is found, we'll use a DATA queue instead */ } - if (queue < 0 && mvmsta->reserved_queue != IEEE80211_INVAL_HW_QUEUE) { + if ((queue < 0 && mvmsta->reserved_queue != IEEE80211_INVAL_HW_QUEUE) && + (mvm->queue_info[mvmsta->reserved_queue].status == + IWL_MVM_QUEUE_RESERVED || + mvm->queue_info[mvmsta->reserved_queue].status == + IWL_MVM_QUEUE_INACTIVE)) { queue = mvmsta->reserved_queue; + mvm->queue_info[queue].reserved = true; IWL_DEBUG_TX_QUEUES(mvm, "Using reserved queue #%d\n", queue); } if (queue < 0) - queue = iwl_mvm_find_free_queue(mvm, IWL_MVM_DQA_MIN_DATA_QUEUE, + queue = iwl_mvm_find_free_queue(mvm, mvmsta->sta_id, + IWL_MVM_DQA_MIN_DATA_QUEUE, IWL_MVM_DQA_MAX_DATA_QUEUE); /* + * Check if this queue is already allocated but inactive. + * In such a case, we'll need to first free this queue before enabling + * it again, so we'll mark it as reserved to make sure no new traffic + * arrives on it + */ + if (queue > 0 && + mvm->queue_info[queue].status == IWL_MVM_QUEUE_INACTIVE) { + mvm->queue_info[queue].status = IWL_MVM_QUEUE_RESERVED; + using_inactive_queue = true; + IWL_DEBUG_TX_QUEUES(mvm, + "Re-assigning TXQ %d: sta_id=%d, tid=%d\n", + queue, mvmsta->sta_id, tid); + } + + /* * Mark TXQ as ready, even though it hasn't been fully configured yet, * to make sure no one else takes it. * This will allow avoiding re-acquiring the lock at the end of the @@ -380,6 +511,38 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, cfg.aggregate = (queue >= IWL_MVM_DQA_MIN_DATA_QUEUE || queue == IWL_MVM_DQA_BSS_CLIENT_QUEUE); + /* + * If this queue was previously inactive (idle) - we need to free it + * first + */ + if (using_inactive_queue) { + struct iwl_scd_txq_cfg_cmd cmd = { + .scd_queue = queue, + .enable = 0, + }; + + disable_agg_tids = iwl_mvm_remove_sta_queue_marking(mvm, queue); + + /* Disable the queue */ + iwl_mvm_invalidate_sta_queue(mvm, queue, disable_agg_tids, + true); + iwl_trans_txq_disable(mvm->trans, queue, false); + ret = iwl_mvm_send_cmd_pdu(mvm, SCD_QUEUE_CFG, 0, sizeof(cmd), + &cmd); + if (ret) { + IWL_ERR(mvm, + "Failed to free inactive queue %d (ret=%d)\n", + queue, ret); + + /* Re-mark the inactive queue as inactive */ + spin_lock_bh(&mvm->queue_info_lock); + mvm->queue_info[queue].status = IWL_MVM_QUEUE_INACTIVE; + spin_unlock_bh(&mvm->queue_info_lock); + + return ret; + } + } + IWL_DEBUG_TX_QUEUES(mvm, "Allocating queue #%d to sta %d on tid %d\n", queue, mvmsta->sta_id, tid); @@ -389,7 +552,9 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, spin_lock_bh(&mvmsta->lock); mvmsta->tid_data[tid].txq_id = queue; + mvmsta->tid_data[tid].is_tid_active = true; mvmsta->tfd_queue_msk |= BIT(queue); + queue_state = mvmsta->tid_data[tid].state; if (mvmsta->reserved_queue == queue) mvmsta->reserved_queue = IEEE80211_INVAL_HW_QUEUE; @@ -399,7 +564,11 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, if (ret) goto out_err; - return 0; + /* If we need to re-enable aggregations... */ + if (queue_state == IWL_AGG_ON) + ret = iwl_mvm_sta_tx_agg(mvm, sta, tid, queue, true); + + return ret; out_err: iwl_mvm_disable_txq(mvm, queue, mac_queue, tid, 0); @@ -476,6 +645,9 @@ void iwl_mvm_add_new_dqa_stream_wk(struct work_struct *wk) unsigned long deferred_tid_traffic; int sta_id, tid; + /* Check inactivity of queues */ + iwl_mvm_inactivity_check(mvm); + mutex_lock(&mvm->mutex); /* Go over all stations with deferred traffic */ @@ -505,6 +677,12 @@ static int iwl_mvm_reserve_sta_stream(struct iwl_mvm *mvm, struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta); int queue; + /* + * Check for inactive queues, so we don't reach a situation where we + * can't add a STA due to a shortage in queues that doesn't really exist + */ + iwl_mvm_inactivity_check(mvm); + spin_lock_bh(&mvm->queue_info_lock); /* Make sure we have free resources for this STA */ @@ -514,7 +692,8 @@ static int iwl_mvm_reserve_sta_stream(struct iwl_mvm *mvm, IWL_MVM_QUEUE_FREE)) queue = IWL_MVM_DQA_BSS_CLIENT_QUEUE; else - queue = iwl_mvm_find_free_queue(mvm, IWL_MVM_DQA_MIN_DATA_QUEUE, + queue = iwl_mvm_find_free_queue(mvm, mvmsta->sta_id, + IWL_MVM_DQA_MIN_DATA_QUEUE, IWL_MVM_DQA_MAX_DATA_QUEUE); if (queue < 0) { spin_unlock_bh(&mvm->queue_info_lock); @@ -1403,8 +1582,8 @@ out_free: return ret; } -static int iwl_mvm_sta_tx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta, - int tid, u8 queue, bool start) +int iwl_mvm_sta_tx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta, + int tid, u8 queue, bool start) { struct iwl_mvm_sta *mvm_sta = iwl_mvm_sta_from_mac80211(sta); struct iwl_mvm_add_sta_cmd cmd = {}; @@ -1459,6 +1638,7 @@ const u8 tid_to_mac80211_ac[] = { IEEE80211_AC_VI, IEEE80211_AC_VO, IEEE80211_AC_VO, + IEEE80211_AC_VO, /* We treat MGMT as TID 8, which is set as AC_VO */ }; static const u8 tid_to_ucode_ac[] = { @@ -1513,7 +1693,8 @@ int iwl_mvm_sta_tx_agg_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif, txq_id = mvmsta->tid_data[tid].txq_id; if (!iwl_mvm_is_dqa_supported(mvm) || mvm->queue_info[txq_id].status != IWL_MVM_QUEUE_READY) { - txq_id = iwl_mvm_find_free_queue(mvm, mvm->first_agg_queue, + txq_id = iwl_mvm_find_free_queue(mvm, mvmsta->sta_id, + mvm->first_agg_queue, mvm->last_agg_queue); if (txq_id < 0) { ret = txq_id; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h index d2c58f1..1588eb6 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h @@ -321,6 +321,9 @@ enum iwl_mvm_agg_state { * Basically when next_reclaimed reaches ssn, we can tell mac80211 that * we are ready to finish the Tx AGG stop / start flow. * @tx_time: medium time consumed by this A-MPDU + * @is_tid_active: has this TID sent traffic in the last + * %IWL_MVM_DQA_QUEUE_TIMEOUT time period. If %txq_id is invalid, this + * field should be ignored. */ struct iwl_mvm_tid_data { struct sk_buff_head deferred_tx_frames; @@ -333,6 +336,7 @@ struct iwl_mvm_tid_data { u16 txq_id; u16 ssn; u16 tx_time; + bool is_tid_active; }; static inline u16 iwl_mvm_tid_queued(struct iwl_mvm_tid_data *tid_data) @@ -509,6 +513,9 @@ int iwl_mvm_sta_tx_agg_stop(struct iwl_mvm *mvm, struct ieee80211_vif *vif, int iwl_mvm_sta_tx_agg_flush(struct iwl_mvm *mvm, struct ieee80211_vif *vif, struct ieee80211_sta *sta, u16 tid); +int iwl_mvm_sta_tx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta, + int tid, u8 queue, bool start); + int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm); void iwl_mvm_del_aux_sta(struct iwl_mvm *mvm); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c index 779bafc..9943013 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c @@ -884,9 +884,11 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb, * nullfunc frames should go to the MGMT queue regardless of QOS */ tid = IWL_MAX_TID_COUNT; - txq_id = mvmsta->tid_data[tid].txq_id; } + if (iwl_mvm_is_dqa_supported(mvm)) + txq_id = mvmsta->tid_data[tid].txq_id; + /* Copy MAC header from skb into command buffer */ memcpy(tx_cmd->hdr, hdr, hdrlen); @@ -905,9 +907,12 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb, txq_id = mvmsta->tid_data[tid].txq_id; } - if (iwl_mvm_is_dqa_supported(mvm)) { - if (unlikely(mvmsta->tid_data[tid].txq_id == - IEEE80211_INVAL_HW_QUEUE)) { + /* Check if TXQ needs to be allocated or re-activated */ + if (unlikely(txq_id == IEEE80211_INVAL_HW_QUEUE || + !mvmsta->tid_data[tid].is_tid_active) && + iwl_mvm_is_dqa_supported(mvm)) { + /* If TXQ needs to be allocated... */ + if (txq_id == IEEE80211_INVAL_HW_QUEUE) { iwl_mvm_tx_add_stream(mvm, mvmsta, tid, skb); /* @@ -917,11 +922,22 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb, iwl_trans_free_tx_cmd(mvm->trans, dev_cmd); spin_unlock(&mvmsta->lock); return 0; + } - txq_id = mvmsta->tid_data[tid].txq_id; + /* If we are here - TXQ exists and needs to be re-activated */ + spin_lock(&mvm->queue_info_lock); + mvm->queue_info[txq_id].status = IWL_MVM_QUEUE_READY; + mvmsta->tid_data[tid].is_tid_active = true; + spin_unlock(&mvm->queue_info_lock); + + IWL_DEBUG_TX_QUEUES(mvm, "Re-activating queue %d for TX\n", + txq_id); } + /* Keep track of the time of the last frame for this RA/TID */ + mvm->queue_info[txq_id].last_frame_time[tid] = jiffies; + IWL_DEBUG_TX(mvm, "TX to [%d|%d] Q:%d - seq: 0x%x\n", mvmsta->sta_id, tid, txq_id, IEEE80211_SEQ_TO_SN(seq_number)); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c index 161b99e..a0cb5ca 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c @@ -579,17 +579,29 @@ void iwl_mvm_dump_nic_error_log(struct iwl_mvm *mvm) iwl_mvm_dump_umac_error_log(mvm); } -int iwl_mvm_find_free_queue(struct iwl_mvm *mvm, u8 minq, u8 maxq) +int iwl_mvm_find_free_queue(struct iwl_mvm *mvm, u8 sta_id, u8 minq, u8 maxq) { int i; lockdep_assert_held(&mvm->queue_info_lock); + /* Start by looking for a free queue */ for (i = minq; i <= maxq; i++) if (mvm->queue_info[i].hw_queue_refcount == 0 && mvm->queue_info[i].status == IWL_MVM_QUEUE_FREE) return i; + /* + * If no free queue found - settle for an inactive one to reconfigure + * Make sure that the inactive queue either already belongs to this STA, + * or that if it belongs to another one - it isn't the reserved queue + */ + for (i = minq; i <= maxq; i++) + if (mvm->queue_info[i].status == IWL_MVM_QUEUE_INACTIVE && + (sta_id == mvm->queue_info[i].ra_sta_id || + !mvm->queue_info[i].reserved)) + return i; + return -ENOSPC; } @@ -650,6 +662,7 @@ void iwl_mvm_enable_txq(struct iwl_mvm *mvm, int queue, int mac80211_queue, else mvm->queue_info[queue].ra_sta_id = cfg->sta_id; mvm->queue_info[queue].tid_bitmap |= BIT(cfg->tid); + mvm->queue_info[queue].ra_sta_id = cfg->sta_id; IWL_DEBUG_TX_QUEUES(mvm, "Enabling TXQ #%d refcount=%d (mac80211 map:0x%x)\n", @@ -752,6 +765,9 @@ void iwl_mvm_disable_txq(struct iwl_mvm *mvm, int queue, int mac80211_queue, mvm->queue_info[queue].tid_bitmap = 0; mvm->queue_info[queue].hw_queue_to_mac80211 = 0; + /* Regardless if this is a reserved TXQ for a STA - mark it as false */ + mvm->queue_info[queue].reserved = false; + spin_unlock_bh(&mvm->queue_info_lock); iwl_trans_txq_disable(mvm->trans, queue, false); @@ -1039,6 +1055,154 @@ out: ieee80211_connection_loss(vif); } +/* + * Remove inactive TIDs of a given queue. + * If all queue TIDs are inactive - mark the queue as inactive + * If only some the queue TIDs are inactive - unmap them from the queue + */ +static void iwl_mvm_remove_inactive_tids(struct iwl_mvm *mvm, + struct iwl_mvm_sta *mvmsta, int queue, + unsigned long tid_bitmap) +{ + int tid; + + lockdep_assert_held(&mvmsta->lock); + lockdep_assert_held(&mvm->queue_info_lock); + + /* Go over all non-active TIDs, incl. IWL_MAX_TID_COUNT (for mgmt) */ + for_each_set_bit(tid, &tid_bitmap, IWL_MAX_TID_COUNT + 1) { + /* If some TFDs are still queued - don't mark TID as inactive */ + if (iwl_mvm_tid_queued(&mvmsta->tid_data[tid])) + tid_bitmap &= ~BIT(tid); + } + + /* If all TIDs in the queue are inactive - mark queue as inactive. */ + if (tid_bitmap == mvm->queue_info[queue].tid_bitmap) { + mvm->queue_info[queue].status = IWL_MVM_QUEUE_INACTIVE; + + for_each_set_bit(tid, &tid_bitmap, IWL_MAX_TID_COUNT + 1) + mvmsta->tid_data[tid].is_tid_active = false; + + IWL_DEBUG_TX_QUEUES(mvm, "Queue %d marked as inactive\n", + queue); + return; + } + + /* + * If we are here, this is a shared queue and not all TIDs timed-out. + * Remove the ones that did. + */ + for_each_set_bit(tid, &tid_bitmap, IWL_MAX_TID_COUNT + 1) { + int mac_queue = mvmsta->vif->hw_queue[tid_to_mac80211_ac[tid]]; + + mvmsta->tid_data[tid].txq_id = IEEE80211_INVAL_HW_QUEUE; + mvm->queue_info[queue].hw_queue_to_mac80211 &= ~BIT(mac_queue); + mvm->queue_info[queue].hw_queue_refcount--; + mvm->queue_info[queue].tid_bitmap &= ~BIT(tid); + mvmsta->tid_data[tid].is_tid_active = false; + + IWL_DEBUG_TX_QUEUES(mvm, + "Removing inactive TID %d from shared Q:%d\n", + tid, queue); + } + + IWL_DEBUG_TX_QUEUES(mvm, + "TXQ #%d left with tid bitmap 0x%x\n", queue, + mvm->queue_info[queue].tid_bitmap); + + /* + * There may be different TIDs with the same mac queues, so make + * sure all TIDs have existing corresponding mac queues enabled + */ + tid_bitmap = mvm->queue_info[queue].tid_bitmap; + for_each_set_bit(tid, &tid_bitmap, IWL_MAX_TID_COUNT + 1) { + mvm->queue_info[queue].hw_queue_to_mac80211 |= + BIT(mvmsta->vif->hw_queue[tid_to_mac80211_ac[tid]]); + } + + /* TODO: if queue was shared - need to re-enable AGGs */ +} + +void iwl_mvm_inactivity_check(struct iwl_mvm *mvm) +{ + unsigned long timeout_queues_map = 0; + unsigned long now = jiffies; + int i; + + spin_lock_bh(&mvm->queue_info_lock); + for (i = 0; i < IWL_MAX_HW_QUEUES; i++) + if (mvm->queue_info[i].hw_queue_refcount > 0) + timeout_queues_map |= BIT(i); + spin_unlock_bh(&mvm->queue_info_lock); + + rcu_read_lock(); + + /* + * If a queue time outs - mark it as INACTIVE (don't remove right away + * if we don't have to.) This is an optimization in case traffic comes + * later, and we don't HAVE to use a currently-inactive queue + */ + for_each_set_bit(i, &timeout_queues_map, IWL_MAX_HW_QUEUES) { + struct ieee80211_sta *sta; + struct iwl_mvm_sta *mvmsta; + u8 sta_id; + int tid; + unsigned long inactive_tid_bitmap = 0; + unsigned long queue_tid_bitmap; + + spin_lock_bh(&mvm->queue_info_lock); + queue_tid_bitmap = mvm->queue_info[i].tid_bitmap; + + /* If TXQ isn't in active use anyway - nothing to do here... */ + if (mvm->queue_info[i].status != IWL_MVM_QUEUE_READY) { + spin_unlock_bh(&mvm->queue_info_lock); + continue; + } + + /* Check to see if there are inactive TIDs on this queue */ + for_each_set_bit(tid, &queue_tid_bitmap, + IWL_MAX_TID_COUNT + 1) { + if (time_after(mvm->queue_info[i].last_frame_time[tid] + + IWL_MVM_DQA_QUEUE_TIMEOUT, now)) + continue; + + inactive_tid_bitmap |= BIT(tid); + } + spin_unlock_bh(&mvm->queue_info_lock); + + /* If all TIDs are active - finish check on this queue */ + if (!inactive_tid_bitmap) + continue; + + /* + * If we are here - the queue hadn't been served recently and is + * in use + */ + + sta_id = mvm->queue_info[i].ra_sta_id; + sta = rcu_dereference(mvm->fw_id_to_mac_id[sta_id]); + + /* + * If the STA doesn't exist anymore, it isn't an error. It could + * be that it was removed since getting the queues, and in this + * case it should've inactivated its queues anyway. + */ + if (IS_ERR_OR_NULL(sta)) + continue; + + mvmsta = iwl_mvm_sta_from_mac80211(sta); + + spin_lock_bh(&mvmsta->lock); + spin_lock(&mvm->queue_info_lock); + iwl_mvm_remove_inactive_tids(mvm, mvmsta, i, + inactive_tid_bitmap); + spin_unlock(&mvm->queue_info_lock); + spin_unlock_bh(&mvmsta->lock); + } + + rcu_read_unlock(); +} + int iwl_mvm_send_lqm_cmd(struct ieee80211_vif *vif, enum iwl_lqm_cmd_operatrions operation, u32 duration, u32 timeout) -- cgit v0.10.2 From 2047fa5401918a1e3bc0ec8923de2570265d1d78 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Sun, 1 May 2016 11:40:49 +0300 Subject: iwlwifi: pcie: unify restock calls on init Currently code calls restock for mq devices during the init function, unlike sq where restock is called after init. This causes an harmless but alarming deadlock warning from lockdep, to fix this - unify the init code. Rename the restock functions while at it. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index 70e39e4..8d4a60d 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -233,10 +233,10 @@ static void iwl_pcie_rxq_check_wrptr(struct iwl_trans *trans) } /* - * iwl_pcie_rxq_mq_restock - restock implementation for multi-queue rx + * iwl_pcie_rxmq_restock - restock implementation for multi-queue rx */ -static void iwl_pcie_rxq_mq_restock(struct iwl_trans *trans, - struct iwl_rxq *rxq) +static void iwl_pcie_rxmq_restock(struct iwl_trans *trans, + struct iwl_rxq *rxq) { struct iwl_rx_mem_buffer *rxb; @@ -281,10 +281,10 @@ static void iwl_pcie_rxq_mq_restock(struct iwl_trans *trans, } /* - * iwl_pcie_rxq_sq_restock - restock implementation for single queue rx + * iwl_pcie_rxsq_restock - restock implementation for single queue rx */ -static void iwl_pcie_rxq_sq_restock(struct iwl_trans *trans, - struct iwl_rxq *rxq) +static void iwl_pcie_rxsq_restock(struct iwl_trans *trans, + struct iwl_rxq *rxq) { struct iwl_rx_mem_buffer *rxb; @@ -343,9 +343,9 @@ static void iwl_pcie_rxq_restock(struct iwl_trans *trans, struct iwl_rxq *rxq) { if (trans->cfg->mq_rx_supported) - iwl_pcie_rxq_mq_restock(trans, rxq); + iwl_pcie_rxmq_restock(trans, rxq); else - iwl_pcie_rxq_sq_restock(trans, rxq); + iwl_pcie_rxsq_restock(trans, rxq); } /* @@ -828,9 +828,6 @@ static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans) enabled |= BIT(i) | BIT(i + 16); } - /* restock default queue */ - iwl_pcie_rxq_mq_restock(trans, &trans_pcie->rxq[0]); - /* * Enable Rx DMA * Rx buffer size 4 or 8k or 12k @@ -960,12 +957,13 @@ int iwl_pcie_rx_init(struct iwl_trans *trans) } iwl_pcie_rxq_alloc_rbs(trans, GFP_KERNEL, def_rxq); - if (trans->cfg->mq_rx_supported) { + + if (trans->cfg->mq_rx_supported) iwl_pcie_rx_mq_hw_init(trans); - } else { - iwl_pcie_rxq_sq_restock(trans, def_rxq); + else iwl_pcie_rx_hw_init(trans, def_rxq); - } + + iwl_pcie_rxq_restock(trans, def_rxq); spin_lock(&def_rxq->lock); iwl_pcie_rxq_inc_wr_ptr(trans, def_rxq); -- cgit v0.10.2 From f43495fd5e37b2ef4a187974011ff439e6354053 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Wed, 4 May 2016 14:22:10 +0300 Subject: iwlwifi: mvm: fix possible division by zero Theoretically we may get only one IRQ from OS, in which case we will have only 1 queue even in MSIx mode. This will cause division by zero in the indirection table calculation. We do not need send the command in that case, as there is only one queue so all RX traffic will be directed to it anyway. Bail out early if there is only one queue. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index 7057f35..0a5490c 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -122,6 +122,9 @@ static int iwl_send_rss_cfg_cmd(struct iwl_mvm *mvm) IWL_RSS_HASH_TYPE_IPV6_PAYLOAD, }; + if (mvm->trans->num_rx_queues == 1) + return 0; + /* Do not direct RSS traffic to Q 0 which is our fallback queue */ for (i = 0; i < ARRAY_SIZE(cmd.indirection_table); i++) cmd.indirection_table[i] = -- cgit v0.10.2 From 69e046423ad71de625ac1b0f0f390d3b9727b8c9 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Tue, 3 May 2016 12:18:33 +0300 Subject: iwlwifi: mvm: change scan timeout to a delayed work Some transports may sleep when writing to registers, which is done when calling iwl_force_nmi(). So we can't call iwl_force_nmi() in a timer context. To solve that, convert the scan timeout timer to a delayed work. Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index e5cb7db..af1d266 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -1200,6 +1200,7 @@ static void iwl_mvm_mac_stop(struct ieee80211_hw *hw) flush_work(&mvm->add_stream_wk); cancel_delayed_work_sync(&mvm->fw_dump_wk); cancel_delayed_work_sync(&mvm->cs_tx_unblock_dwork); + cancel_delayed_work_sync(&mvm->scan_timeout_dwork); iwl_mvm_free_fw_dump_desc(mvm); mutex_lock(&mvm->mutex); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index bf7d78e..3775e26 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -802,7 +802,7 @@ struct iwl_mvm { struct iwl_mcast_filter_cmd *mcast_filter_cmd; enum iwl_mvm_scan_type scan_type; enum iwl_mvm_sched_scan_pass_all_states sched_scan_pass_all; - struct timer_list scan_timer; + struct delayed_work scan_timeout_dwork; /* max number of simultaneous scans the FW supports */ unsigned int max_scans; @@ -1415,7 +1415,7 @@ int iwl_mvm_scan_size(struct iwl_mvm *mvm); int iwl_mvm_scan_stop(struct iwl_mvm *mvm, int type, bool notify); int iwl_mvm_max_scan_ie_len(struct iwl_mvm *mvm); void iwl_mvm_report_scan_aborted(struct iwl_mvm *mvm); -void iwl_mvm_scan_timeout(unsigned long data); +void iwl_mvm_scan_timeout_wk(struct work_struct *work); /* Scheduled scan */ void iwl_mvm_rx_lmac_scan_complete_notif(struct iwl_mvm *mvm, diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c index 632b1dc..063ebbe 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c @@ -608,6 +608,7 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg, INIT_WORK(&mvm->d0i3_exit_work, iwl_mvm_d0i3_exit_work); INIT_DELAYED_WORK(&mvm->fw_dump_wk, iwl_mvm_fw_error_dump_wk); INIT_DELAYED_WORK(&mvm->tdls_cs.dwork, iwl_mvm_tdls_ch_switch_work); + INIT_DELAYED_WORK(&mvm->scan_timeout_dwork, iwl_mvm_scan_timeout_wk); INIT_WORK(&mvm->add_stream_wk, iwl_mvm_add_new_dqa_stream_wk); spin_lock_init(&mvm->d0i3_tx_lock); @@ -766,9 +767,6 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg, iwl_mvm_tof_init(mvm); - setup_timer(&mvm->scan_timer, iwl_mvm_scan_timeout, - (unsigned long)mvm); - return op_mode; out_unregister: @@ -822,8 +820,6 @@ static void iwl_op_mode_mvm_stop(struct iwl_op_mode *op_mode) iwl_mvm_tof_clean(mvm); - del_timer_sync(&mvm->scan_timer); - mutex_destroy(&mvm->mutex); mutex_destroy(&mvm->d0i3_suspend_mutex); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c index 6f609dd..fb25d9e 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c @@ -399,7 +399,7 @@ void iwl_mvm_rx_lmac_scan_complete_notif(struct iwl_mvm *mvm, ieee80211_scan_completed(mvm->hw, scan_notif->status == IWL_SCAN_OFFLOAD_ABORTED); iwl_mvm_unref(mvm, IWL_MVM_REF_SCAN); - del_timer(&mvm->scan_timer); + cancel_delayed_work(&mvm->scan_timeout_dwork); } else { IWL_ERR(mvm, "got scan complete notification but no scan is running\n"); @@ -1222,15 +1222,16 @@ static int iwl_mvm_check_running_scans(struct iwl_mvm *mvm, int type) return -EIO; } -#define SCAN_TIMEOUT (16 * HZ) +#define SCAN_TIMEOUT 20000 -void iwl_mvm_scan_timeout(unsigned long data) +void iwl_mvm_scan_timeout_wk(struct work_struct *work) { - struct iwl_mvm *mvm = (struct iwl_mvm *)data; + struct delayed_work *delayed_work = to_delayed_work(work); + struct iwl_mvm *mvm = container_of(delayed_work, struct iwl_mvm, + scan_timeout_dwork); IWL_ERR(mvm, "regular scan timed out\n"); - del_timer(&mvm->scan_timer); iwl_force_nmi(mvm->trans); } @@ -1313,7 +1314,8 @@ int iwl_mvm_reg_scan_start(struct iwl_mvm *mvm, struct ieee80211_vif *vif, mvm->scan_status |= IWL_MVM_SCAN_REGULAR; iwl_mvm_ref(mvm, IWL_MVM_REF_SCAN); - mod_timer(&mvm->scan_timer, jiffies + SCAN_TIMEOUT); + queue_delayed_work(system_wq, &mvm->scan_timeout_dwork, + msecs_to_jiffies(SCAN_TIMEOUT)); return 0; } @@ -1432,7 +1434,7 @@ void iwl_mvm_rx_umac_scan_complete_notif(struct iwl_mvm *mvm, if (mvm->scan_uid_status[uid] == IWL_MVM_SCAN_REGULAR) { ieee80211_scan_completed(mvm->hw, aborted); iwl_mvm_unref(mvm, IWL_MVM_REF_SCAN); - del_timer(&mvm->scan_timer); + cancel_delayed_work(&mvm->scan_timeout_dwork); } else if (mvm->scan_uid_status[uid] == IWL_MVM_SCAN_SCHED) { ieee80211_sched_scan_stopped(mvm->hw); mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_DISABLED; @@ -1628,7 +1630,7 @@ out: * to release the scan reference here. */ iwl_mvm_unref(mvm, IWL_MVM_REF_SCAN); - del_timer(&mvm->scan_timer); + cancel_delayed_work(&mvm->scan_timeout_dwork); if (notify) ieee80211_scan_completed(mvm->hw, true); } else if (notify) { -- cgit v0.10.2 From bdbc58abe7960ec1c87bd6201d31538fbda0d1ba Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 4 May 2016 09:19:13 +0300 Subject: iwlwifi: mvm: remove an unused variable We never initialize ampdu_status so it causes a static checker warning when we pass it to iwl_mvm_pass_packet_to_mac80211(). Fortunately, it's never used so we can just remove it. Signed-off-by: Dan Carpenter Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c index 6d096b6..2d5a7cc 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c @@ -101,7 +101,7 @@ static void iwl_mvm_pass_packet_to_mac80211(struct iwl_mvm *mvm, struct napi_struct *napi, struct sk_buff *skb, struct ieee80211_hdr *hdr, u16 len, - u32 ampdu_status, u8 crypt_len, + u8 crypt_len, struct iwl_rx_cmd_buffer *rxb) { unsigned int hdrlen, fraglen; @@ -268,7 +268,6 @@ void iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct napi_struct *napi, struct ieee80211_sta *sta = NULL; struct sk_buff *skb; u32 len; - u32 ampdu_status; u32 rate_n_flags; u32 rx_pkt_status; u8 crypt_len = 0; @@ -480,7 +479,7 @@ void iwl_mvm_rx_rx_mpdu(struct iwl_mvm *mvm, struct napi_struct *napi, iwl_mvm_ref(mvm, IWL_MVM_REF_RX); iwl_mvm_pass_packet_to_mac80211(mvm, sta, napi, skb, hdr, len, - ampdu_status, crypt_len, rxb); + crypt_len, rxb); if (take_ref) iwl_mvm_unref(mvm, IWL_MVM_REF_RX); -- cgit v0.10.2 From 32afd15b0f56eb9adf107fb52c6b76c308f5966c Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 4 May 2016 09:19:54 +0300 Subject: iwlwifi: mvm: silence uninitialized variable warning "max_amsdu_len" isn't set if kstrtouint() fails. Signed-off-by: Dan Carpenter Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c index 406cf1c..b344898 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/debugfs.c @@ -1020,6 +1020,8 @@ static ssize_t iwl_dbgfs_max_amsdu_len_write(struct iwl_mvm *mvm, int ret; ret = kstrtouint(buf, 0, &max_amsdu_len); + if (ret) + return ret; if (max_amsdu_len > IEEE80211_MAX_MPDU_LEN_VHT_11454) return -EINVAL; -- cgit v0.10.2 From 42db09c1b0378e118b804d948a5bab6194721506 Mon Sep 17 00:00:00 2001 From: Liad Kaufman Date: Mon, 2 May 2016 14:01:14 +0300 Subject: iwlwifi: mvm: support dqa queue sharing Support DQA queue sharing when no free queue exists for allocation to a STA that already exists. This means that a single queue will serve more than a single TID (although the RA will be the same for all TIDs served). We try to choose the lowest AC possible, to ensure the shared queues have the lowest possible combined AC requirements. The queue to share is chosen only from the same RA's DATA queues as follows (in descending priority): 1. An AC_BE queue 2. Same AC queue 3. Highest AC queue that is lower than new AC 4. Any existing AC (there always is at least 1 DATA queue) If any aggregations existed for any of the TIDs of the shared queue - they are stopped (the FW is notified), but no delBA is sent. Signed-off-by: Liad Kaufman Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h index a45be1d..57cc67f 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h @@ -583,6 +583,7 @@ struct iwl_trans_txq_scd_cfg { * configured. May sleep. * @txq_disable: de-configure a Tx queue to send AMPDUs * Must be atomic + * @txq_set_shared_mode: change Tx queue shared/unshared marking * @wait_tx_queue_empty: wait until tx queues are empty. May sleep. * @freeze_txq_timer: prevents the timer of the queue from firing until the * queue is set to awake. Must be atomic. @@ -646,6 +647,9 @@ struct iwl_trans_ops { void (*txq_disable)(struct iwl_trans *trans, int queue, bool configure_scd); + void (*txq_set_shared_mode)(struct iwl_trans *trans, u32 txq_id, + bool shared); + int (*wait_tx_queue_empty)(struct iwl_trans *trans, u32 txq_bm); void (*freeze_txq_timer)(struct iwl_trans *trans, unsigned long txqs, bool freeze); @@ -1061,6 +1065,13 @@ iwl_trans_txq_enable_cfg(struct iwl_trans *trans, int queue, u16 ssn, trans->ops->txq_enable(trans, queue, ssn, cfg, queue_wdg_timeout); } +static inline void iwl_trans_txq_set_shared_mode(struct iwl_trans *trans, + int queue, bool shared_mode) +{ + if (trans->ops->txq_set_shared_mode) + trans->ops->txq_set_shared_mode(trans, queue, shared_mode); +} + static inline void iwl_trans_txq_enable(struct iwl_trans *trans, int queue, int fifo, int sta_id, int tid, int frame_limit, u16 ssn, diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index 3775e26..6092af2 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -687,6 +687,10 @@ struct iwl_mvm_baid_data { * This is the state of a queue that has been fully configured (including * SCD pointers, etc), has a specific RA/TID assigned to it, and can be * used to send traffic. + * @IWL_MVM_QUEUE_SHARED: queue is shared, or in a process of becoming shared + * This is a state in which a single queue serves more than one TID, all of + * which are not aggregated. Note that the queue is only associated to one + * RA. * @IWL_MVM_QUEUE_INACTIVE: queue is allocated but no traffic on it * This is a state of a queue that has had traffic on it, but during the * last %IWL_MVM_DQA_QUEUE_TIMEOUT time period there has been no traffic on @@ -698,6 +702,7 @@ enum iwl_mvm_queue_status { IWL_MVM_QUEUE_FREE, IWL_MVM_QUEUE_RESERVED, IWL_MVM_QUEUE_READY, + IWL_MVM_QUEUE_SHARED, IWL_MVM_QUEUE_INACTIVE, }; @@ -760,6 +765,7 @@ struct iwl_mvm { u8 hw_queue_refcount; u8 ra_sta_id; /* The RA this queue is mapped to, if exists */ bool reserved; /* Is this the TXQ reserved for a STA */ + u8 mac80211_ac; /* The mac80211 AC this queue is mapped to */ u16 tid_bitmap; /* Bitmap of the TIDs mapped to this queue */ /* Timestamp for inactivation per TID of this queue */ unsigned long last_frame_time[IWL_MAX_TID_COUNT + 1]; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index 84384a43..14c06cb 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -361,6 +361,40 @@ static int iwl_mvm_invalidate_sta_queue(struct iwl_mvm *mvm, int queue, return ret; } +static int iwl_mvm_get_queue_agg_tids(struct iwl_mvm *mvm, int queue) +{ + struct ieee80211_sta *sta; + struct iwl_mvm_sta *mvmsta; + unsigned long tid_bitmap; + unsigned long agg_tids = 0; + s8 sta_id; + int tid; + + lockdep_assert_held(&mvm->mutex); + + spin_lock_bh(&mvm->queue_info_lock); + sta_id = mvm->queue_info[queue].ra_sta_id; + tid_bitmap = mvm->queue_info[queue].tid_bitmap; + spin_unlock_bh(&mvm->queue_info_lock); + + sta = rcu_dereference_protected(mvm->fw_id_to_mac_id[sta_id], + lockdep_is_held(&mvm->mutex)); + + if (WARN_ON_ONCE(IS_ERR_OR_NULL(sta))) + return -EINVAL; + + mvmsta = iwl_mvm_sta_from_mac80211(sta); + + spin_lock_bh(&mvmsta->lock); + for_each_set_bit(tid, &tid_bitmap, IWL_MAX_TID_COUNT + 1) { + if (mvmsta->tid_data[tid].state == IWL_AGG_ON) + agg_tids |= BIT(tid); + } + spin_unlock_bh(&mvmsta->lock); + + return agg_tids; +} + /* * Remove a queue from a station's resources. * Note that this only marks as free. It DOESN'T delete a BA agreement, and @@ -394,28 +428,91 @@ static int iwl_mvm_remove_sta_queue_marking(struct iwl_mvm *mvm, int queue) mvmsta = iwl_mvm_sta_from_mac80211(sta); spin_lock_bh(&mvmsta->lock); + /* Unmap MAC queues and TIDs from this queue */ for_each_set_bit(tid, &tid_bitmap, IWL_MAX_TID_COUNT + 1) { - mvmsta->tid_data[tid].txq_id = IEEE80211_INVAL_HW_QUEUE; - if (mvmsta->tid_data[tid].state == IWL_AGG_ON) disable_agg_tids |= BIT(tid); + mvmsta->tid_data[tid].txq_id = IEEE80211_INVAL_HW_QUEUE; } - mvmsta->tfd_queue_msk &= ~BIT(queue); /* Don't use this queue anymore */ + mvmsta->tfd_queue_msk &= ~BIT(queue); /* Don't use this queue anymore */ spin_unlock_bh(&mvmsta->lock); rcu_read_unlock(); - spin_lock(&mvm->queue_info_lock); + spin_lock_bh(&mvm->queue_info_lock); /* Unmap MAC queues and TIDs from this queue */ mvm->queue_info[queue].hw_queue_to_mac80211 = 0; mvm->queue_info[queue].hw_queue_refcount = 0; mvm->queue_info[queue].tid_bitmap = 0; - spin_unlock(&mvm->queue_info_lock); + spin_unlock_bh(&mvm->queue_info_lock); return disable_agg_tids; } +static int iwl_mvm_get_shared_queue(struct iwl_mvm *mvm, + unsigned long tfd_queue_mask, u8 ac) +{ + int queue = 0; + u8 ac_to_queue[IEEE80211_NUM_ACS]; + int i; + + lockdep_assert_held(&mvm->queue_info_lock); + + memset(&ac_to_queue, IEEE80211_INVAL_HW_QUEUE, sizeof(ac_to_queue)); + + /* See what ACs the existing queues for this STA have */ + for_each_set_bit(i, &tfd_queue_mask, IWL_MVM_DQA_MAX_DATA_QUEUE) { + /* Only DATA queues can be shared */ + if (i < IWL_MVM_DQA_MIN_DATA_QUEUE && + i != IWL_MVM_DQA_BSS_CLIENT_QUEUE) + continue; + + ac_to_queue[mvm->queue_info[i].mac80211_ac] = i; + } + + /* + * The queue to share is chosen only from DATA queues as follows (in + * descending priority): + * 1. An AC_BE queue + * 2. Same AC queue + * 3. Highest AC queue that is lower than new AC + * 4. Any existing AC (there always is at least 1 DATA queue) + */ + + /* Priority 1: An AC_BE queue */ + if (ac_to_queue[IEEE80211_AC_BE] != IEEE80211_INVAL_HW_QUEUE) + queue = ac_to_queue[IEEE80211_AC_BE]; + /* Priority 2: Same AC queue */ + else if (ac_to_queue[ac] != IEEE80211_INVAL_HW_QUEUE) + queue = ac_to_queue[ac]; + /* Priority 3a: If new AC is VO and VI exists - use VI */ + else if (ac == IEEE80211_AC_VO && + ac_to_queue[IEEE80211_AC_VI] != IEEE80211_INVAL_HW_QUEUE) + queue = ac_to_queue[IEEE80211_AC_VI]; + /* Priority 3b: No BE so only AC less than the new one is BK */ + else if (ac_to_queue[IEEE80211_AC_BK] != IEEE80211_INVAL_HW_QUEUE) + queue = ac_to_queue[IEEE80211_AC_BK]; + /* Priority 4a: No BE nor BK - use VI if exists */ + else if (ac_to_queue[IEEE80211_AC_VI] != IEEE80211_INVAL_HW_QUEUE) + queue = ac_to_queue[IEEE80211_AC_VI]; + /* Priority 4b: No BE, BK nor VI - use VO if exists */ + else if (ac_to_queue[IEEE80211_AC_VO] != IEEE80211_INVAL_HW_QUEUE) + queue = ac_to_queue[IEEE80211_AC_VO]; + + /* Make sure queue found (or not) is legal */ + if (!((queue >= IWL_MVM_DQA_MIN_MGMT_QUEUE && + queue <= IWL_MVM_DQA_MAX_MGMT_QUEUE) || + (queue >= IWL_MVM_DQA_MIN_DATA_QUEUE && + queue <= IWL_MVM_DQA_MAX_DATA_QUEUE) || + (queue == IWL_MVM_DQA_BSS_CLIENT_QUEUE))) { + IWL_ERR(mvm, "No DATA queues available to share\n"); + queue = -ENOSPC; + } + + return queue; +} + static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, struct ieee80211_sta *sta, u8 ac, int tid, struct ieee80211_hdr *hdr) @@ -434,11 +531,17 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, bool using_inactive_queue = false; unsigned long disable_agg_tids = 0; enum iwl_mvm_agg_state queue_state; + bool shared_queue = false; int ssn; + unsigned long tfd_queue_mask; int ret; lockdep_assert_held(&mvm->mutex); + spin_lock_bh(&mvmsta->lock); + tfd_queue_mask = mvmsta->tfd_queue_msk; + spin_unlock_bh(&mvmsta->lock); + spin_lock_bh(&mvm->queue_info_lock); /* @@ -487,20 +590,32 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, queue, mvmsta->sta_id, tid); } + /* No free queue - we'll have to share */ + if (queue <= 0) { + queue = iwl_mvm_get_shared_queue(mvm, tfd_queue_mask, ac); + if (queue > 0) { + shared_queue = true; + mvm->queue_info[queue].status = IWL_MVM_QUEUE_SHARED; + } + } + /* * Mark TXQ as ready, even though it hasn't been fully configured yet, * to make sure no one else takes it. * This will allow avoiding re-acquiring the lock at the end of the * configuration. On error we'll mark it back as free. */ - if (queue >= 0) + if ((queue > 0) && !shared_queue) mvm->queue_info[queue].status = IWL_MVM_QUEUE_READY; spin_unlock_bh(&mvm->queue_info_lock); - /* TODO: support shared queues for same RA */ - if (queue < 0) + /* This shouldn't happen - out of queues */ + if (WARN_ON(queue <= 0)) { + IWL_ERR(mvm, "No available queues for tid %d on sta_id %d\n", + tid, cfg.sta_id); return -ENOSPC; + } /* * Actual en/disablement of aggregations is through the ADD_STA HCMD, @@ -543,8 +658,27 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, } } - IWL_DEBUG_TX_QUEUES(mvm, "Allocating queue #%d to sta %d on tid %d\n", - queue, mvmsta->sta_id, tid); + IWL_DEBUG_TX_QUEUES(mvm, + "Allocating %squeue #%d to sta %d on tid %d\n", + shared_queue ? "shared " : "", queue, + mvmsta->sta_id, tid); + + if (shared_queue) { + /* Disable any open aggs on this queue */ + disable_agg_tids = iwl_mvm_get_queue_agg_tids(mvm, queue); + + if (disable_agg_tids) { + IWL_DEBUG_TX_QUEUES(mvm, "Disabling aggs on queue %d\n", + queue); + iwl_mvm_invalidate_sta_queue(mvm, queue, + disable_agg_tids, false); + } + + /* Mark queue as shared in transport */ + iwl_trans_txq_set_shared_mode(mvm->trans, queue, true); + + /* TODO: a redirection may be required - DQA phase 2 */ + } ssn = IEEE80211_SEQ_TO_SN(le16_to_cpu(hdr->seq_ctrl)); iwl_mvm_enable_txq(mvm, queue, mac_queue, ssn, &cfg, @@ -560,15 +694,20 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, mvmsta->reserved_queue = IEEE80211_INVAL_HW_QUEUE; spin_unlock_bh(&mvmsta->lock); - ret = iwl_mvm_sta_send_to_fw(mvm, sta, true, STA_MODIFY_QUEUES); - if (ret) - goto out_err; + if (!shared_queue) { + ret = iwl_mvm_sta_send_to_fw(mvm, sta, true, STA_MODIFY_QUEUES); + if (ret) + goto out_err; - /* If we need to re-enable aggregations... */ - if (queue_state == IWL_AGG_ON) - ret = iwl_mvm_sta_tx_agg(mvm, sta, tid, queue, true); + /* If we need to re-enable aggregations... */ + if (queue_state == IWL_AGG_ON) { + ret = iwl_mvm_sta_tx_agg(mvm, sta, tid, queue, true); + if (ret) + goto out_err; + } + } - return ret; + return 0; out_err: iwl_mvm_disable_txq(mvm, queue, mac_queue, tid, 0); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c index a0cb5ca..2fc51e7 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c @@ -655,15 +655,22 @@ void iwl_mvm_enable_txq(struct iwl_mvm *mvm, int queue, int mac80211_queue, } /* Update mappings and refcounts */ + if (mvm->queue_info[queue].hw_queue_refcount > 0) + enable_queue = false; + mvm->queue_info[queue].hw_queue_to_mac80211 |= BIT(mac80211_queue); mvm->queue_info[queue].hw_queue_refcount++; - if (mvm->queue_info[queue].hw_queue_refcount > 1) - enable_queue = false; - else - mvm->queue_info[queue].ra_sta_id = cfg->sta_id; mvm->queue_info[queue].tid_bitmap |= BIT(cfg->tid); mvm->queue_info[queue].ra_sta_id = cfg->sta_id; + if (enable_queue) { + if (cfg->tid != IWL_MAX_TID_COUNT) + mvm->queue_info[queue].mac80211_ac = + tid_to_mac80211_ac[cfg->tid]; + else + mvm->queue_info[queue].mac80211_ac = IEEE80211_AC_VO; + } + IWL_DEBUG_TX_QUEUES(mvm, "Enabling TXQ #%d refcount=%d (mac80211 map:0x%x)\n", queue, mvm->queue_info[queue].hw_queue_refcount, @@ -1154,7 +1161,8 @@ void iwl_mvm_inactivity_check(struct iwl_mvm *mvm) queue_tid_bitmap = mvm->queue_info[i].tid_bitmap; /* If TXQ isn't in active use anyway - nothing to do here... */ - if (mvm->queue_info[i].status != IWL_MVM_QUEUE_READY) { + if (mvm->queue_info[i].status != IWL_MVM_QUEUE_READY && + mvm->queue_info[i].status != IWL_MVM_QUEUE_SHARED) { spin_unlock_bh(&mvm->queue_info_lock); continue; } diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h index 482836e..8bfa915 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h @@ -471,6 +471,8 @@ void iwl_trans_pcie_txq_enable(struct iwl_trans *trans, int queue, u16 ssn, unsigned int wdg_timeout); void iwl_trans_pcie_txq_disable(struct iwl_trans *trans, int queue, bool configure_scd); +void iwl_trans_pcie_txq_set_shared_mode(struct iwl_trans *trans, u32 txq_id, + bool shared_mode); int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb, struct iwl_device_cmd *dev_cmd, int txq_id); void iwl_pcie_txq_check_wrptrs(struct iwl_trans *trans); diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index 33fd217..3badebb 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -2745,6 +2745,8 @@ static const struct iwl_trans_ops trans_ops_pcie = { .txq_disable = iwl_trans_pcie_txq_disable, .txq_enable = iwl_trans_pcie_txq_enable, + .txq_set_shared_mode = iwl_trans_pcie_txq_set_shared_mode, + .wait_tx_queue_empty = iwl_trans_pcie_wait_txq_empty, .freeze_txq_timer = iwl_trans_pcie_freeze_txq_timer, .block_txq_ptrs = iwl_trans_pcie_block_txq_ptrs, diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c index d6beac9..8901d49 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c @@ -1354,6 +1354,15 @@ void iwl_trans_pcie_txq_enable(struct iwl_trans *trans, int txq_id, u16 ssn, txq->active = true; } +void iwl_trans_pcie_txq_set_shared_mode(struct iwl_trans *trans, u32 txq_id, + bool shared_mode) +{ + struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); + struct iwl_txq *txq = &trans_pcie->txq[txq_id]; + + txq->ampdu = !shared_mode; +} + void iwl_trans_pcie_txq_disable(struct iwl_trans *trans, int txq_id, bool configure_scd) { -- cgit v0.10.2 From 93f436e2c7feacb04a21bbfb984a7afd87fb4623 Mon Sep 17 00:00:00 2001 From: Liad Kaufman Date: Mon, 31 Aug 2015 13:41:26 +0300 Subject: iwlwifi: mvm: set sta_id in SCD_QUEUE_CONFIG cmd Set the correct sta_id in the SCD_QUEUE_CONFIG command sent to the FW when enabling/disabling queues. This is needed in DQA-mode to allow the FW to associate between queue and STA. In case the queue isn't connected to a specific station but rather is a static "generic" queue - the sta_id should be set to 0x10 (max supported STA is 0x0f). Signed-off-by: Liad Kaufman Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index 14c06cb..9d4ebc9 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -635,9 +635,16 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, .scd_queue = queue, .enable = 0, }; + u8 ac; disable_agg_tids = iwl_mvm_remove_sta_queue_marking(mvm, queue); + spin_lock_bh(&mvm->queue_info_lock); + ac = mvm->queue_info[queue].mac80211_ac; + cmd.sta_id = mvm->queue_info[queue].ra_sta_id; + cmd.tx_fifo = iwl_mvm_ac_to_tx_fifo[ac]; + spin_unlock_bh(&mvm->queue_info_lock); + /* Disable the queue */ iwl_mvm_invalidate_sta_queue(mvm, queue, disable_agg_tids, true); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c index 2fc51e7..68f4e7f 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/utils.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/utils.c @@ -691,6 +691,10 @@ void iwl_mvm_enable_txq(struct iwl_mvm *mvm, int queue, int mac80211_queue, .tid = cfg->tid, }; + /* Set sta_id in the command, if it exists */ + if (iwl_mvm_is_dqa_supported(mvm)) + cmd.sta_id = cfg->sta_id; + iwl_trans_txq_enable_cfg(mvm->trans, queue, ssn, NULL, wdg_timeout); WARN(iwl_mvm_send_cmd_pdu(mvm, SCD_QUEUE_CFG, 0, sizeof(cmd), -- cgit v0.10.2 From 28d0793ed212e4714cea79aeb77e62de99b139c1 Mon Sep 17 00:00:00 2001 From: Liad Kaufman Date: Tue, 1 Sep 2015 16:36:25 +0300 Subject: iwlwifi: mvm: update aux queue in dqa mode In DQA mode the AUX queue is mapped elsewhere than in non- DQA mode. Update the code to reflect this. Signed-off-by: Liad Kaufman Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h index b06380d..a1d3d95 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h @@ -90,6 +90,7 @@ enum { * DQA queue numbers * * @IWL_MVM_DQA_CMD_QUEUE: a queue reserved for sending HCMDs to the FW + * @IWL_MVM_DQA_AUX_QUEUE: a queue reserved for aux frames * @IWL_MVM_DQA_P2P_DEVICE_QUEUE: a queue reserved for P2P device frames * @IWL_MVM_DQA_GCAST_QUEUE: a queue reserved for P2P GO/SoftAP GCAST frames * @IWL_MVM_DQA_BSS_CLIENT_QUEUE: a queue reserved for BSS activity, to ensure @@ -108,6 +109,7 @@ enum { */ enum iwl_mvm_dqa_txq { IWL_MVM_DQA_CMD_QUEUE = 0, + IWL_MVM_DQA_AUX_QUEUE = 1, IWL_MVM_DQA_P2P_DEVICE_QUEUE = 2, IWL_MVM_DQA_GCAST_QUEUE = 3, IWL_MVM_DQA_BSS_CLIENT_QUEUE = 4, diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c index 063ebbe..a08db00 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c @@ -577,18 +577,21 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg, mvm->restart_fw = iwlwifi_mod_params.restart_fw ? -1 : 0; - mvm->aux_queue = 15; if (!iwl_mvm_is_dqa_supported(mvm)) { - mvm->first_agg_queue = 16; mvm->last_agg_queue = mvm->cfg->base_params->num_of_queues - 1; + + if (mvm->cfg->base_params->num_of_queues == 16) { + mvm->aux_queue = 11; + mvm->first_agg_queue = 12; + } else { + mvm->aux_queue = 15; + mvm->first_agg_queue = 16; + } } else { + mvm->aux_queue = IWL_MVM_DQA_AUX_QUEUE; mvm->first_agg_queue = IWL_MVM_DQA_MIN_DATA_QUEUE; mvm->last_agg_queue = IWL_MVM_DQA_MAX_DATA_QUEUE; } - if (mvm->cfg->base_params->num_of_queues == 16) { - mvm->aux_queue = 11; - mvm->first_agg_queue = 12; - } mvm->sf_state = SF_UNINIT; mvm->cur_ucode = IWL_UCODE_INIT; mvm->drop_bcn_ap_mode = true; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index 9d4ebc9..cf06817 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -1309,8 +1309,9 @@ int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm) lockdep_assert_held(&mvm->mutex); /* Map Aux queue to fifo - needs to happen before adding Aux station */ - iwl_mvm_enable_ac_txq(mvm, mvm->aux_queue, mvm->aux_queue, - IWL_MVM_TX_FIFO_MCAST, 0, wdg_timeout); + if (!iwl_mvm_is_dqa_supported(mvm)) + iwl_mvm_enable_ac_txq(mvm, mvm->aux_queue, mvm->aux_queue, + IWL_MVM_TX_FIFO_MCAST, 0, wdg_timeout); /* Allocate aux station and assign to it the aux queue */ ret = iwl_mvm_allocate_int_sta(mvm, &mvm->aux_sta, BIT(mvm->aux_queue), @@ -1318,6 +1319,19 @@ int iwl_mvm_add_aux_sta(struct iwl_mvm *mvm) if (ret) return ret; + if (iwl_mvm_is_dqa_supported(mvm)) { + struct iwl_trans_txq_scd_cfg cfg = { + .fifo = IWL_MVM_TX_FIFO_MCAST, + .sta_id = mvm->aux_sta.sta_id, + .tid = IWL_MAX_TID_COUNT, + .aggregate = false, + .frame_limit = IWL_FRAME_LIMIT, + }; + + iwl_mvm_enable_txq(mvm, mvm->aux_queue, mvm->aux_queue, 0, &cfg, + wdg_timeout); + } + ret = iwl_mvm_add_int_sta_common(mvm, &mvm->aux_sta, NULL, MAC_INDEX_AUX, 0); -- cgit v0.10.2 From 97d5be7e229426ba17a305a0e51c86e08b89436a Mon Sep 17 00:00:00 2001 From: Liad Kaufman Date: Mon, 31 Aug 2015 14:33:23 +0300 Subject: iwlwifi: mvm: support dqa-enable hcmd Support sending the DQA-enablement HCMD to the FW when working in DQA mode. This HCMD will enable DQA-specific flows in the FW. Signed-off-by: Liad Kaufman Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h index a1d3d95..eea03ec 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h @@ -332,6 +332,7 @@ enum iwl_system_subcmd_ids { }; enum iwl_data_path_subcmd_ids { + DQA_ENABLE_CMD = 0x0, UPDATE_MU_GROUPS_CMD = 0x1, TRIGGER_RX_QUEUES_NOTIF_CMD = 0x2, MU_GROUP_MGMT_NOTIF = 0xFE, @@ -362,6 +363,14 @@ struct iwl_cmd_response { }; /* + * struct iwl_dqa_enable_cmd + * @cmd_queue: the TXQ number of the command queue + */ +struct iwl_dqa_enable_cmd { + __le32 cmd_queue; +} __packed; /* DQA_CONTROL_CMD_API_S_VER_1 */ + +/* * struct iwl_tx_ant_cfg_cmd * @valid: valid antenna configuration */ diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index 0a5490c..510b4c1 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -134,6 +134,23 @@ static int iwl_send_rss_cfg_cmd(struct iwl_mvm *mvm) return iwl_mvm_send_cmd_pdu(mvm, RSS_CONFIG_CMD, 0, sizeof(cmd), &cmd); } +static int iwl_mvm_send_dqa_cmd(struct iwl_mvm *mvm) +{ + struct iwl_dqa_enable_cmd dqa_cmd = { + .cmd_queue = cpu_to_le32(IWL_MVM_DQA_CMD_QUEUE), + }; + u32 cmd_id = iwl_cmd_id(DQA_ENABLE_CMD, DATA_PATH_GROUP, 0); + int ret; + + ret = iwl_mvm_send_cmd_pdu(mvm, cmd_id, 0, sizeof(dqa_cmd), &dqa_cmd); + if (ret) + IWL_ERR(mvm, "Failed to send DQA enabling command: %d\n", ret); + else + IWL_DEBUG_FW(mvm, "Working in DQA mode\n"); + + return ret; +} + void iwl_free_fw_paging(struct iwl_mvm *mvm) { int i; @@ -979,6 +996,15 @@ int iwl_mvm_up(struct iwl_mvm *mvm) /* reset quota debouncing buffer - 0xff will yield invalid data */ memset(&mvm->last_quota_cmd, 0xff, sizeof(mvm->last_quota_cmd)); + /* Enable DQA-mode if required */ + if (iwl_mvm_is_dqa_supported(mvm)) { + ret = iwl_mvm_send_dqa_cmd(mvm); + if (ret) + goto error; + } else { + IWL_DEBUG_FW(mvm, "Working in non-DQA mode\n"); + } + /* Add auxiliary station for scanning */ ret = iwl_mvm_add_aux_sta(mvm); if (ret) -- cgit v0.10.2 From 4b79deece5d45396422d469afa11f9d69ccb3d8b Mon Sep 17 00:00:00 2001 From: Oren Givon Date: Mon, 23 May 2016 09:58:17 +0300 Subject: iwlwifi: add new 8260 PCI IDs Add 3 new 8260 series PCI IDs: - (0x24F3, 0x10B0) - (0x24F3, 0xD0B0) - (0x24F3, 0xB0B0) CC: [4.1+] Signed-off-by: Oren Givon Signed-off-by: David Spinadel Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index a588b05..1cae19d 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -433,6 +433,7 @@ static const struct pci_device_id iwl_hw_card_ids[] = { /* 8000 Series */ {IWL_PCI_DEVICE(0x24F3, 0x0010, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0x1010, iwl8260_2ac_cfg)}, + {IWL_PCI_DEVICE(0x24F3, 0x10B0, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0x0130, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0x1130, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0x0132, iwl8260_2ac_cfg)}, @@ -454,6 +455,8 @@ static const struct pci_device_id iwl_hw_card_ids[] = { {IWL_PCI_DEVICE(0x24F3, 0xD010, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0xC050, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0xD050, iwl8260_2ac_cfg)}, + {IWL_PCI_DEVICE(0x24F3, 0xD0B0, iwl8260_2ac_cfg)}, + {IWL_PCI_DEVICE(0x24F3, 0xB0B0, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0x8010, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0x8110, iwl8260_2ac_cfg)}, {IWL_PCI_DEVICE(0x24F3, 0x9010, iwl8260_2ac_cfg)}, -- cgit v0.10.2 From f24bbae565d279cd90c904fe55b539a45631705e Mon Sep 17 00:00:00 2001 From: Oren Givon Date: Mon, 23 May 2016 09:58:17 +0300 Subject: iwlwifi: add new 8265 Add 6 new 8265 series PCI IDs: - (0x24FD, 0x1130) - (0x24FD, 0x0130) - (0x24FD, 0x0910) - (0x24FD, 0x0930) - (0x24FD, 0x0950) - (0x24FD, 0x0850) CC: [4.6+] Signed-off-by: Oren Givon Signed-off-by: David Spinadel Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index 1cae19d..6f020e4 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -484,6 +484,8 @@ static const struct pci_device_id iwl_hw_card_ids[] = { {IWL_PCI_DEVICE(0x24FD, 0x0010, iwl8265_2ac_cfg)}, {IWL_PCI_DEVICE(0x24FD, 0x0110, iwl8265_2ac_cfg)}, {IWL_PCI_DEVICE(0x24FD, 0x1110, iwl8265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x24FD, 0x1130, iwl8265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x24FD, 0x0130, iwl8265_2ac_cfg)}, {IWL_PCI_DEVICE(0x24FD, 0x1010, iwl8265_2ac_cfg)}, {IWL_PCI_DEVICE(0x24FD, 0x0050, iwl8265_2ac_cfg)}, {IWL_PCI_DEVICE(0x24FD, 0x0150, iwl8265_2ac_cfg)}, @@ -494,6 +496,10 @@ static const struct pci_device_id iwl_hw_card_ids[] = { {IWL_PCI_DEVICE(0x24FD, 0x0810, iwl8265_2ac_cfg)}, {IWL_PCI_DEVICE(0x24FD, 0x9110, iwl8265_2ac_cfg)}, {IWL_PCI_DEVICE(0x24FD, 0x8130, iwl8265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x24FD, 0x0910, iwl8265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x24FD, 0x0930, iwl8265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x24FD, 0x0950, iwl8265_2ac_cfg)}, + {IWL_PCI_DEVICE(0x24FD, 0x0850, iwl8265_2ac_cfg)}, /* 9000 Series */ {IWL_PCI_DEVICE(0x2526, 0x0000, iwl9260_2ac_cfg)}, -- cgit v0.10.2 From 34777b00005525850c9f624186520745c5dd40b5 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Mon, 30 May 2016 12:52:43 +0300 Subject: iwlwifi: mvm: remove unnecessary device conversion when reading the MCC We convert the mvm device to a PCI device and then back again when trying to find the handle for the device's ACPI data. This is unnecessary, so it can be removed. Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c index 25a9840..182ec20 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c @@ -66,7 +66,6 @@ *****************************************************************************/ #include #include -#include #include #include "iwl-trans.h" #include "iwl-csr.h" @@ -802,9 +801,8 @@ static int iwl_mvm_get_bios_mcc(struct iwl_mvm *mvm, char *mcc) struct acpi_buffer wrdd = {ACPI_ALLOCATE_BUFFER, NULL}; acpi_status status; u32 mcc_val; - struct pci_dev *pdev = to_pci_dev(mvm->dev); - root_handle = ACPI_HANDLE(&pdev->dev); + root_handle = ACPI_HANDLE(mvm->dev); if (!root_handle) { IWL_DEBUG_LAR(mvm, "Could not retrieve root port ACPI handle\n"); -- cgit v0.10.2 From d7fdd0e528c5184fd22153a317683549f0c58a19 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Thu, 19 May 2016 17:53:42 +0300 Subject: iwlwifi: pcie: poll RFH for RX DMA stop Somehow we ended up stopping RX using legacy RX registers even for devices that support RFH. Fix it. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h index f08cdee..73f0ed8 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h @@ -344,6 +344,32 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(unsigned int chnl) #define RFH_RBDBUF_RBD0_LSB 0xA08300 #define RFH_RBDBUF_RBD_LSB(q) (RFH_RBDBUF_RBD0_LSB + (q) * 8) +/** + * RFH Status Register + * + * Bit fields: + * + * Bit 29: RBD_FETCH_IDLE + * This status flag is set by the RFH when there is no active RBD fetch from + * DRAM. + * Once the RFH RBD controller starts fetching (or when there is a pending + * RBD read response from DRAM), this flag is immediately turned off. + * + * Bit 30: SRAM_DMA_IDLE + * This status flag is set by the RFH when there is no active transaction from + * SRAM to DRAM. + * Once the SRAM to DRAM DMA is active, this flag is immediately turned off. + * + * Bit 31: RXF_DMA_IDLE + * This status flag is set by the RFH when there is no active transaction from + * RXF to DRAM. + * Once the RXF-to-DRAM DMA is active, this flag is immediately turned off. + */ +#define RFH_GEN_STATUS 0xA09808 +#define RBD_FETCH_IDLE BIT(29) +#define SRAM_DMA_IDLE BIT(30) +#define RXF_DMA_IDLE BIT(31) + /* DMA configuration */ #define RFH_RXF_DMA_CFG 0xA09820 /* RB size */ diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index 8d4a60d..c1c3c6a 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -173,9 +173,16 @@ static void iwl_pcie_write_prph_64_no_grab(struct iwl_trans *trans, u64 ofs, */ int iwl_pcie_rx_stop(struct iwl_trans *trans) { - iwl_write_direct32(trans, FH_MEM_RCSR_CHNL0_CONFIG_REG, 0); - return iwl_poll_direct_bit(trans, FH_MEM_RSSR_RX_STATUS_REG, - FH_RSSR_CHNL0_RX_STATUS_CHNL_IDLE, 1000); + if (trans->cfg->mq_rx_supported) { + iwl_write_prph(trans, RFH_RXF_DMA_CFG, 0); + return iwl_poll_prph_bit(trans, RFH_GEN_STATUS, + RXF_DMA_IDLE, RXF_DMA_IDLE, 1000); + } else { + iwl_write_direct32(trans, FH_MEM_RCSR_CHNL0_CONFIG_REG, 0); + return iwl_poll_direct_bit(trans, FH_MEM_RSSR_RX_STATUS_REG, + FH_RSSR_CHNL0_RX_STATUS_CHNL_IDLE, + 1000); + } } /* -- cgit v0.10.2 From ff7a68d0e7a7da411d0f87f57dde7e276e01ac92 Mon Sep 17 00:00:00 2001 From: Matti Gottlieb Date: Tue, 17 May 2016 10:50:08 +0300 Subject: iwlwifi: mvm: Do not open aggregations for null data packets Currently we try to open an aggregation for every packet (given that one is not already open). This causes redundant overhead (addba/delba) for null data packets. Do not open an aggregation for null data packets. Signed-off-by: Matti Gottlieb Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c index 81dd2f6..7aecf46 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c @@ -399,7 +399,7 @@ static int iwl_hwrate_to_plcp_idx(u32 rate_n_flags) static void rs_rate_scale_perform(struct iwl_mvm *mvm, struct ieee80211_sta *sta, struct iwl_lq_sta *lq_sta, - int tid); + int tid, bool ndp); static void rs_fill_lq_cmd(struct iwl_mvm *mvm, struct ieee80211_sta *sta, struct iwl_lq_sta *lq_sta, @@ -1161,7 +1161,7 @@ static u8 rs_get_tid(struct ieee80211_hdr *hdr) } void iwl_mvm_rs_tx_status(struct iwl_mvm *mvm, struct ieee80211_sta *sta, - int tid, struct ieee80211_tx_info *info) + int tid, struct ieee80211_tx_info *info, bool ndp) { int legacy_success; int retries; @@ -1384,7 +1384,7 @@ void iwl_mvm_rs_tx_status(struct iwl_mvm *mvm, struct ieee80211_sta *sta, done: /* See if there's a better rate or modulation mode to try. */ if (sta->supp_rates[info->band]) - rs_rate_scale_perform(mvm, sta, lq_sta, tid); + rs_rate_scale_perform(mvm, sta, lq_sta, tid, ndp); } /* @@ -1407,7 +1407,8 @@ static void rs_mac80211_tx_status(void *mvm_r, info->flags & IEEE80211_TX_CTL_NO_ACK) return; - iwl_mvm_rs_tx_status(mvm, sta, rs_get_tid(hdr), info); + iwl_mvm_rs_tx_status(mvm, sta, rs_get_tid(hdr), info, + ieee80211_is_qos_nullfunc(hdr->frame_control)); } /* @@ -2213,7 +2214,7 @@ static bool rs_tpc_perform(struct iwl_mvm *mvm, static void rs_rate_scale_perform(struct iwl_mvm *mvm, struct ieee80211_sta *sta, struct iwl_lq_sta *lq_sta, - int tid) + int tid, bool ndp) { int low = IWL_RATE_INVALID; int high = IWL_RATE_INVALID; @@ -2512,7 +2513,7 @@ lq_update: (lq_sta->tx_agg_tid_en & (1 << tid)) && (tid != IWL_MAX_TID_COUNT)) { tid_data = &sta_priv->tid_data[tid]; - if (tid_data->state == IWL_AGG_OFF) { + if (tid_data->state == IWL_AGG_OFF && !ndp) { IWL_DEBUG_RATE(mvm, "try to aggregate tid %d\n", tid); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h index 90d046f..d4a7fe2 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h @@ -362,7 +362,7 @@ void iwl_mvm_rs_rate_init(struct iwl_mvm *mvm, struct ieee80211_sta *sta, /* Notify RS about Tx status */ void iwl_mvm_rs_tx_status(struct iwl_mvm *mvm, struct ieee80211_sta *sta, - int tid, struct ieee80211_tx_info *info); + int tid, struct ieee80211_tx_info *info, bool ndp); /** * iwl_rate_control_register - Register the rate control algorithm callbacks diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c index 9943013..d91edab 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c @@ -1660,7 +1660,7 @@ void iwl_mvm_rx_ba_notif(struct iwl_mvm *mvm, struct iwl_rx_cmd_buffer *rxb) iwl_mvm_tx_info_from_ba_notif(&ba_info, ba_notif, tid_data); IWL_DEBUG_TX_REPLY(mvm, "No reclaim. Update rs directly\n"); - iwl_mvm_rs_tx_status(mvm, sta, tid, &ba_info); + iwl_mvm_rs_tx_status(mvm, sta, tid, &ba_info, false); } out: -- cgit v0.10.2 From 99980ecffffe6af53361456be41c27e2fffc89cd Mon Sep 17 00:00:00 2001 From: Ayala Beker Date: Wed, 1 Jun 2016 11:46:20 +0300 Subject: iwlwifi: mvm: fix RX mpdu status enum FW sets status for each RX packet. Enum in the driver doesn't match with FW definition - fix it. Signed-off-by: Ayala Beker Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h index 1994331..acc5cd5 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-rx.h @@ -296,7 +296,7 @@ enum iwl_rx_mpdu_status { IWL_RX_MPDU_STATUS_OVERRUN_OK = BIT(1), IWL_RX_MPDU_STATUS_SRC_STA_FOUND = BIT(2), IWL_RX_MPDU_STATUS_KEY_VALID = BIT(3), - IWL_RX_MPDU_STATUS_KEY_ERROR = BIT(4), + IWL_RX_MPDU_STATUS_KEY_PARAM_OK = BIT(4), IWL_RX_MPDU_STATUS_ICV_OK = BIT(5), IWL_RX_MPDU_STATUS_MIC_OK = BIT(6), IWL_RX_MPDU_RES_STATUS_TTAK_OK = BIT(7), @@ -311,7 +311,7 @@ enum iwl_rx_mpdu_status { IWL_RX_MPDU_STATUS_WEP_MATCH = BIT(12), IWL_RX_MPDU_STATUS_EXT_IV_MATCH = BIT(13), IWL_RX_MPDU_STATUS_KEY_ID_MATCH = BIT(14), - IWL_RX_MPDU_STATUS_KEY_COLOR = BIT(15), + IWL_RX_MPDU_STATUS_ROBUST_MNG_FRAME = BIT(15), }; enum iwl_rx_mpdu_hash_filter { -- cgit v0.10.2 From 855f492f655711259becb7d34978063ffc40fe12 Mon Sep 17 00:00:00 2001 From: Gregory Greenman Date: Wed, 25 May 2016 18:31:39 +0300 Subject: iwlwifi: mvm: rs: add rate scaling support for 160MHz channels Expand TLC to support 160MHz channels. Full support for A-MSDU case will be added separately. Signed-off-by: Gregory Greenman Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c index 7aecf46..227c5ed 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.c @@ -211,6 +211,9 @@ static bool rs_sgi_allow(struct iwl_mvm *mvm, struct ieee80211_sta *sta, if (is_ht80(rate) && (vht_cap->cap & IEEE80211_VHT_CAP_SHORT_GI_80)) return true; + if (is_ht160(rate) && (vht_cap->cap & + IEEE80211_VHT_CAP_SHORT_GI_160)) + return true; return false; } @@ -445,6 +448,13 @@ static const u16 expected_tpt_siso_80MHz[4][IWL_RATE_COUNT] = { {0, 0, 0, 0, 241, 0, 475, 701, 921, 1343, 1741, 1931, 2117, 2468, 2691}, }; +static const u16 expected_tpt_siso_160MHz[4][IWL_RATE_COUNT] = { + {0, 0, 0, 0, 191, 0, 244, 288, 298, 308, 313, 318, 323, 328, 330}, + {0, 0, 0, 0, 200, 0, 251, 293, 302, 312, 317, 322, 327, 332, 334}, + {0, 0, 0, 0, 439, 0, 875, 1307, 1736, 2584, 3419, 3831, 4240, 5049, 5581}, + {0, 0, 0, 0, 488, 0, 972, 1451, 1925, 2864, 3785, 4240, 4691, 5581, 6165}, +}; + static const u16 expected_tpt_mimo2_20MHz[4][IWL_RATE_COUNT] = { {0, 0, 0, 0, 74, 0, 123, 155, 179, 213, 235, 243, 250, 261, 0}, {0, 0, 0, 0, 81, 0, 131, 164, 187, 221, 242, 250, 256, 267, 0}, @@ -466,6 +476,13 @@ static const u16 expected_tpt_mimo2_80MHz[4][IWL_RATE_COUNT] = { {0, 0, 0, 0, 474, 0, 920, 1338, 1732, 2464, 3116, 3418, 3705, 4225, 4545}, }; +static const u16 expected_tpt_mimo2_160MHz[4][IWL_RATE_COUNT] = { + {0, 0, 0, 0, 240, 0, 278, 308, 313, 319, 322, 324, 328, 330, 334}, + {0, 0, 0, 0, 247, 0, 282, 310, 315, 320, 323, 325, 329, 332, 338}, + {0, 0, 0, 0, 875, 0, 1735, 2582, 3414, 5043, 6619, 7389, 8147, 9629, 10592}, + {0, 0, 0, 0, 971, 0, 1925, 2861, 3779, 5574, 7304, 8147, 8976, 10592, 11640}, +}; + /* mbps, mcs */ static const struct iwl_rate_mcs_info iwl_rate_mcs[IWL_RATE_COUNT] = { { "1", "BPSK DSSS"}, @@ -901,7 +918,6 @@ static int rs_rate_from_ucode_rate(const u32 ucode_rate, } } - WARN_ON_ONCE(rate->bw == RATE_MCS_CHAN_WIDTH_160); WARN_ON_ONCE(rate->bw == RATE_MCS_CHAN_WIDTH_80 && !is_vht(rate)); @@ -1495,6 +1511,9 @@ static const u16 *rs_get_expected_tpt_table(struct iwl_lq_sta *lq_sta, case RATE_MCS_CHAN_WIDTH_80: ht_tbl_pointer = expected_tpt_siso_80MHz; break; + case RATE_MCS_CHAN_WIDTH_160: + ht_tbl_pointer = expected_tpt_siso_160MHz; + break; default: WARN_ON_ONCE(1); } @@ -1509,6 +1528,9 @@ static const u16 *rs_get_expected_tpt_table(struct iwl_lq_sta *lq_sta, case RATE_MCS_CHAN_WIDTH_80: ht_tbl_pointer = expected_tpt_mimo2_80MHz; break; + case RATE_MCS_CHAN_WIDTH_160: + ht_tbl_pointer = expected_tpt_mimo2_160MHz; + break; default: WARN_ON_ONCE(1); } @@ -1583,12 +1605,17 @@ static s32 rs_get_best_rate(struct iwl_mvm *mvm, static u32 rs_bw_from_sta_bw(struct ieee80211_sta *sta) { - if (sta->bandwidth >= IEEE80211_STA_RX_BW_80) + switch (sta->bandwidth) { + case IEEE80211_STA_RX_BW_160: + return RATE_MCS_CHAN_WIDTH_160; + case IEEE80211_STA_RX_BW_80: return RATE_MCS_CHAN_WIDTH_80; - else if (sta->bandwidth >= IEEE80211_STA_RX_BW_40) + case IEEE80211_STA_RX_BW_40: return RATE_MCS_CHAN_WIDTH_40; - - return RATE_MCS_CHAN_WIDTH_20; + case IEEE80211_STA_RX_BW_20: + default: + return RATE_MCS_CHAN_WIDTH_20; + } } /* @@ -2566,6 +2593,9 @@ static const struct rs_init_rate_info rs_optimal_rates_ht[] = { { S8_MIN, IWL_RATE_MCS_0_INDEX}, }; +/* MCS index 9 is not valid for 20MHz VHT channel width, + * but is ok for 40, 80 and 160MHz channels. + */ static const struct rs_init_rate_info rs_optimal_rates_vht_20mhz[] = { { -60, IWL_RATE_MCS_8_INDEX }, { -64, IWL_RATE_MCS_7_INDEX }, @@ -2578,7 +2608,7 @@ static const struct rs_init_rate_info rs_optimal_rates_vht_20mhz[] = { { S8_MIN, IWL_RATE_MCS_0_INDEX}, }; -static const struct rs_init_rate_info rs_optimal_rates_vht_40_80mhz[] = { +static const struct rs_init_rate_info rs_optimal_rates_vht[] = { { -60, IWL_RATE_MCS_9_INDEX }, { -64, IWL_RATE_MCS_8_INDEX }, { -68, IWL_RATE_MCS_7_INDEX }, @@ -2641,9 +2671,9 @@ static void rs_init_optimal_rate(struct iwl_mvm *mvm, lq_sta->optimal_nentries = ARRAY_SIZE(rs_optimal_rates_vht_20mhz); } else { - lq_sta->optimal_rates = rs_optimal_rates_vht_40_80mhz; + lq_sta->optimal_rates = rs_optimal_rates_vht; lq_sta->optimal_nentries = - ARRAY_SIZE(rs_optimal_rates_vht_40_80mhz); + ARRAY_SIZE(rs_optimal_rates_vht); } } else if (is_ht(rate)) { lq_sta->optimal_rates = rs_optimal_rates_ht; @@ -2735,23 +2765,25 @@ static void rs_get_initial_rate(struct iwl_mvm *mvm, */ if (sta->vht_cap.vht_supported && best_rssi > IWL_RS_LOW_RSSI_THRESHOLD) { - if (sta->bandwidth >= IEEE80211_STA_RX_BW_40) { - initial_rates = rs_optimal_rates_vht_40_80mhz; - nentries = ARRAY_SIZE(rs_optimal_rates_vht_40_80mhz); - if (sta->bandwidth >= IEEE80211_STA_RX_BW_80) - rate->bw = RATE_MCS_CHAN_WIDTH_80; - else - rate->bw = RATE_MCS_CHAN_WIDTH_40; - } else if (sta->bandwidth == IEEE80211_STA_RX_BW_20) { + switch (sta->bandwidth) { + case IEEE80211_STA_RX_BW_160: + case IEEE80211_STA_RX_BW_80: + case IEEE80211_STA_RX_BW_40: + initial_rates = rs_optimal_rates_vht; + nentries = ARRAY_SIZE(rs_optimal_rates_vht); + break; + case IEEE80211_STA_RX_BW_20: initial_rates = rs_optimal_rates_vht_20mhz; nentries = ARRAY_SIZE(rs_optimal_rates_vht_20mhz); - rate->bw = RATE_MCS_CHAN_WIDTH_20; - } else { + break; + default: IWL_ERR(mvm, "Invalid BW %d\n", sta->bandwidth); goto out; } + active_rate = lq_sta->active_siso_rate; rate->type = LQ_VHT_SISO; + rate->bw = rs_bw_from_sta_bw(sta); } else if (sta->ht_cap.ht_supported && best_rssi > IWL_RS_LOW_RSSI_THRESHOLD) { initial_rates = rs_optimal_rates_ht; @@ -3058,6 +3090,9 @@ void iwl_mvm_update_frame_stats(struct iwl_mvm *mvm, u32 rate, bool agg) case RATE_MCS_CHAN_WIDTH_80: mvm->drv_rx_stats.bw_80_frames++; break; + case RATE_MCS_CHAN_WIDTH_160: + mvm->drv_rx_stats.bw_160_frames++; + break; default: WARN_ONCE(1, "bad BW. rate 0x%x", rate); } @@ -3706,7 +3741,8 @@ static ssize_t rs_sta_dbgfs_scale_table_read(struct file *file, desc += sprintf(buff + desc, " %s", (is_ht20(rate)) ? "20MHz" : (is_ht40(rate)) ? "40MHz" : - (is_ht80(rate)) ? "80Mhz" : "BAD BW"); + (is_ht80(rate)) ? "80MHz" : + (is_ht160(rate)) ? "160MHz" : "BAD BW"); desc += sprintf(buff + desc, " %s %s %s %s\n", (rate->sgi) ? "SGI" : "NGI", (rate->ldpc) ? "LDPC" : "BCC", @@ -3788,9 +3824,10 @@ static ssize_t rs_sta_dbgfs_stats_table_read(struct file *file, lq_sta->active_tbl == i ? "*" : "x", rate->type, rate->sgi, - is_ht20(rate) ? "20Mhz" : - is_ht40(rate) ? "40Mhz" : - is_ht80(rate) ? "80Mhz" : "ERR", + is_ht20(rate) ? "20MHz" : + is_ht40(rate) ? "40MHz" : + is_ht80(rate) ? "80MHz" : + is_ht160(rate) ? "160MHz" : "ERR", rate->index); for (j = 0; j < IWL_RATE_COUNT; j++) { desc += sprintf(buff+desc, diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h index d4a7fe2..ee207f2 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rs.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rs.h @@ -205,6 +205,7 @@ struct rs_rate { #define is_ht20(rate) ((rate)->bw == RATE_MCS_CHAN_WIDTH_20) #define is_ht40(rate) ((rate)->bw == RATE_MCS_CHAN_WIDTH_40) #define is_ht80(rate) ((rate)->bw == RATE_MCS_CHAN_WIDTH_80) +#define is_ht160(rate) ((rate)->bw == RATE_MCS_CHAN_WIDTH_160) #define IWL_MAX_MCS_DISPLAY_SIZE 12 -- cgit v0.10.2 From 5a7d87da8d9b9f04ecdbebe7e5710a1391f85fa8 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 27 May 2016 15:07:03 +0200 Subject: iwlwifi: mvm: avoid harmless -Wmaybe-uninialized warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit gcc is apparently unablel to track the state of the local 'resp_v2' variable across the kzalloc() function, and warns about the response variable being used without an initialization: drivers/net/wireless/intel/iwlwifi/mvm/nvm.c: In function ‘iwl_mvm_update_mcc’: drivers/net/wireless/intel/iwlwifi/mvm/nvm.c:727:36: warning: ‘mcc_resp_v1’ may be used uninitialized in this function [-Wmaybe-uninitialized] resp_cp->n_channels = mcc_resp_v1->n_channels; drivers/net/wireless/intel/iwlwifi/mvm/nvm.c:721:3: warning: ‘mcc_resp’ may be used uninitialized in this function [-Wmaybe-uninitialized] memcpy(resp_cp, mcc_resp, resp_len); The warning showed up in x86 allmodconfig after my patch to unhide -Wmaybe-uninitialized warnings by default was merged, though it always existed in randconfig builds. I did not catch the warning earlier because I was testing on ARM, which never produced the warning. This rearranges the code in a way that improves readability for both humans and the compiler, and that avoids the warning. Signed-off-by: Arnd Bergmann Fixes: 6fa52430f0b3 ("iwlwifi: mvm: change mcc update API") Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c index 182ec20..7a686f6 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/nvm.c @@ -666,8 +666,7 @@ iwl_mvm_update_mcc(struct iwl_mvm *mvm, const char *alpha2, .mcc = cpu_to_le16(alpha2[0] << 8 | alpha2[1]), .source_id = (u8)src_id, }; - struct iwl_mcc_update_resp *mcc_resp, *resp_cp = NULL; - struct iwl_mcc_update_resp_v1 *mcc_resp_v1 = NULL; + struct iwl_mcc_update_resp *resp_cp; struct iwl_rx_packet *pkt; struct iwl_host_cmd cmd = { .id = MCC_UPDATE_CMD, @@ -700,34 +699,36 @@ iwl_mvm_update_mcc(struct iwl_mvm *mvm, const char *alpha2, /* Extract MCC response */ if (resp_v2) { - mcc_resp = (void *)pkt->data; + struct iwl_mcc_update_resp *mcc_resp = (void *)pkt->data; + n_channels = __le32_to_cpu(mcc_resp->n_channels); + resp_len = sizeof(struct iwl_mcc_update_resp) + + n_channels * sizeof(__le32); + resp_cp = kmemdup(mcc_resp, resp_len, GFP_KERNEL); } else { - mcc_resp_v1 = (void *)pkt->data; + struct iwl_mcc_update_resp_v1 *mcc_resp_v1 = (void *)pkt->data; + n_channels = __le32_to_cpu(mcc_resp_v1->n_channels); + resp_len = sizeof(struct iwl_mcc_update_resp) + + n_channels * sizeof(__le32); + resp_cp = kzalloc(resp_len, GFP_KERNEL); + + if (resp_cp) { + resp_cp->status = mcc_resp_v1->status; + resp_cp->mcc = mcc_resp_v1->mcc; + resp_cp->cap = mcc_resp_v1->cap; + resp_cp->source_id = mcc_resp_v1->source_id; + resp_cp->n_channels = mcc_resp_v1->n_channels; + memcpy(resp_cp->channels, mcc_resp_v1->channels, + n_channels * sizeof(__le32)); + } } - resp_len = sizeof(struct iwl_mcc_update_resp) + n_channels * - sizeof(__le32); - - resp_cp = kzalloc(resp_len, GFP_KERNEL); if (!resp_cp) { ret = -ENOMEM; goto exit; } - if (resp_v2) { - memcpy(resp_cp, mcc_resp, resp_len); - } else { - resp_cp->status = mcc_resp_v1->status; - resp_cp->mcc = mcc_resp_v1->mcc; - resp_cp->cap = mcc_resp_v1->cap; - resp_cp->source_id = mcc_resp_v1->source_id; - resp_cp->n_channels = mcc_resp_v1->n_channels; - memcpy(resp_cp->channels, mcc_resp_v1->channels, - n_channels * sizeof(__le32)); - } - status = le32_to_cpu(resp_cp->status); mcc = le16_to_cpu(resp_cp->mcc); -- cgit v0.10.2 From 2c4a247e42526d9aae8f5ce1f190b893532f2806 Mon Sep 17 00:00:00 2001 From: Oren Givon Date: Sun, 29 May 2016 14:05:50 +0300 Subject: iwlwifi: mvm: fix txq aggregation bug Fix an issue where nullfunc frames and block ack requests had the same tid as aggregation frames and were queued on a non aggregation queue. The pending frames counter included those frames but the check whether to decrement the pending frames counter relied on the tid status and not on the txq id. The result was an inconsistent state of the pending frames counter followed by a failure to remove the station. This failure triggered SYSASSERT 0x3421. In addition, fix a situation in DQA mode where the number of pending frames turned negative. This was due to the TX queue being on the IWL_EMPTYING_HW_QUEUE_DELBA state and its frames were still decremented. Even though the SYSASSERT issue is fixed when DQA is disabled, the issue is not completely solved when DQA is enabled and should still be fixed. Signed-off-by: Oren Givon Fixes: cf961e16620f ("iwlwifi: mvm: support dqa-mode agg on non-shared queue") Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c index d91edab..1f23cee 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c @@ -1329,7 +1329,15 @@ static void iwl_mvm_rx_tx_cmd_single(struct iwl_mvm *mvm, bool send_eosp_ndp = false; spin_lock_bh(&mvmsta->lock); - txq_agg = (mvmsta->tid_data[tid].state == IWL_AGG_ON); + if (iwl_mvm_is_dqa_supported(mvm)) { + enum iwl_mvm_agg_state state; + + state = mvmsta->tid_data[tid].state; + txq_agg = (state == IWL_AGG_ON || + state == IWL_EMPTYING_HW_QUEUE_DELBA); + } else { + txq_agg = txq_id >= mvm->first_agg_queue; + } if (!is_ndp) { tid_data->next_reclaimed = next_reclaimed; -- cgit v0.10.2 From e160f635abd409748908912d1c528721127b2fc1 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Fri, 3 Jun 2016 14:39:19 -0700 Subject: iwlwifi: dvm: Remove unused array 'iwlagn_loose_lookup' gcc-6 reports the following error if -Werror=unused-const-variable is enabled. drivers/net/wireless/intel/iwlwifi/dvm/lib.c:210:21: error: 'iwlagn_loose_lookup' defined but not used Signed-off-by: Guenter Roeck Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/lib.c b/drivers/net/wireless/intel/iwlwifi/dvm/lib.c index 8dda52a..6c2d6da 100644 --- a/drivers/net/wireless/intel/iwlwifi/dvm/lib.c +++ b/drivers/net/wireless/intel/iwlwifi/dvm/lib.c @@ -205,23 +205,6 @@ static const __le32 iwlagn_def_3w_lookup[IWLAGN_BT_DECISION_LUT_SIZE] = { cpu_to_le32(0xf0005000), }; - -/* Loose Coex */ -static const __le32 iwlagn_loose_lookup[IWLAGN_BT_DECISION_LUT_SIZE] = { - cpu_to_le32(0xaaaaaaaa), - cpu_to_le32(0xaaaaaaaa), - cpu_to_le32(0xaeaaaaaa), - cpu_to_le32(0xaaaaaaaa), - cpu_to_le32(0xcc00ff28), - cpu_to_le32(0x0000aaaa), - cpu_to_le32(0xcc00aaaa), - cpu_to_le32(0x0000aaaa), - cpu_to_le32(0x00000000), - cpu_to_le32(0x00000000), - cpu_to_le32(0xf0005000), - cpu_to_le32(0xf0005000), -}; - /* Full concurrency */ static const __le32 iwlagn_concurrent_lookup[IWLAGN_BT_DECISION_LUT_SIZE] = { cpu_to_le32(0xaaaaaaaa), -- cgit v0.10.2 From f65ebd888c5660e97b73fdcb94156a4f357a885d Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Sun, 22 May 2016 16:25:40 +0300 Subject: iwlwifi: add dump of RFH Add support of dumping new RFH instead of FH registers. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-io.c b/drivers/net/wireless/intel/iwlwifi/iwl-io.c index 32c8f84..d8b4306 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-io.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-io.c @@ -1,7 +1,7 @@ /****************************************************************************** * * Copyright(c) 2003 - 2014 Intel Corporation. All rights reserved. - * Copyright(c) 2015 Intel Deutschland GmbH + * Copyright(c) 2015 - 2016 Intel Deutschland GmbH * * Portions of this file are derived from the ipw3945 project. * @@ -228,9 +228,117 @@ void iwl_force_nmi(struct iwl_trans *trans) } IWL_EXPORT_SYMBOL(iwl_force_nmi); -static const char *get_fh_string(int cmd) +static const char *get_rfh_string(int cmd) { #define IWL_CMD(x) case x: return #x +#define IWL_CMD_MQ(arg, reg, q) { if (arg == reg(q)) return #reg; } + + int i; + + for (i = 0; i < IWL_MAX_RX_HW_QUEUES; i++) { + IWL_CMD_MQ(cmd, RFH_Q_FRBDCB_BA_LSB, i); + IWL_CMD_MQ(cmd, RFH_Q_FRBDCB_WIDX, i); + IWL_CMD_MQ(cmd, RFH_Q_FRBDCB_RIDX, i); + IWL_CMD_MQ(cmd, RFH_Q_URBD_STTS_WPTR_LSB, i); + }; + + switch (cmd) { + IWL_CMD(RFH_RXF_DMA_CFG); + IWL_CMD(RFH_GEN_CFG); + IWL_CMD(RFH_GEN_STATUS); + IWL_CMD(FH_TSSR_TX_STATUS_REG); + IWL_CMD(FH_TSSR_TX_ERROR_REG); + default: + return "UNKNOWN"; + } +#undef IWL_CMD_MQ +} + +struct reg { + u32 addr; + bool is64; +}; + +static int iwl_dump_rfh(struct iwl_trans *trans, char **buf) +{ + int i, q; + int num_q = trans->num_rx_queues; + static const u32 rfh_tbl[] = { + RFH_RXF_DMA_CFG, + RFH_GEN_CFG, + RFH_GEN_STATUS, + FH_TSSR_TX_STATUS_REG, + FH_TSSR_TX_ERROR_REG, + }; + static const struct reg rfh_mq_tbl[] = { + { RFH_Q0_FRBDCB_BA_LSB, true }, + { RFH_Q0_FRBDCB_WIDX, false }, + { RFH_Q0_FRBDCB_RIDX, false }, + { RFH_Q0_URBD_STTS_WPTR_LSB, true }, + }; + +#ifdef CONFIG_IWLWIFI_DEBUGFS + if (buf) { + int pos = 0; + /* + * Register (up to 34 for name + 8 blank/q for MQ): 40 chars + * Colon + space: 2 characters + * 0X%08x: 10 characters + * New line: 1 character + * Total of 53 characters + */ + size_t bufsz = ARRAY_SIZE(rfh_tbl) * 53 + + ARRAY_SIZE(rfh_mq_tbl) * 53 * num_q + 40; + + *buf = kmalloc(bufsz, GFP_KERNEL); + if (!*buf) + return -ENOMEM; + + pos += scnprintf(*buf + pos, bufsz - pos, + "RFH register values:\n"); + + for (i = 0; i < ARRAY_SIZE(rfh_tbl); i++) + pos += scnprintf(*buf + pos, bufsz - pos, + "%40s: 0X%08x\n", + get_rfh_string(rfh_tbl[i]), + iwl_read_prph(trans, rfh_tbl[i])); + + for (i = 0; i < ARRAY_SIZE(rfh_mq_tbl); i++) + for (q = 0; q < num_q; q++) { + u32 addr = rfh_mq_tbl[i].addr; + + addr += q * (rfh_mq_tbl[i].is64 ? 8 : 4); + pos += scnprintf(*buf + pos, bufsz - pos, + "%34s(q %2d): 0X%08x\n", + get_rfh_string(addr), q, + iwl_read_prph(trans, addr)); + } + + return pos; + } +#endif + + IWL_ERR(trans, "RFH register values:\n"); + for (i = 0; i < ARRAY_SIZE(rfh_tbl); i++) + IWL_ERR(trans, " %34s: 0X%08x\n", + get_rfh_string(rfh_tbl[i]), + iwl_read_prph(trans, rfh_tbl[i])); + + for (i = 0; i < ARRAY_SIZE(rfh_mq_tbl); i++) + for (q = 0; q < num_q; q++) { + u32 addr = rfh_mq_tbl[i].addr; + + addr += q * (rfh_mq_tbl[i].is64 ? 8 : 4); + IWL_ERR(trans, " %34s(q %d): 0X%08x\n", + get_rfh_string(addr), q, + iwl_read_prph(trans, addr)); + } + + return 0; +} + +static const char *get_fh_string(int cmd) +{ switch (cmd) { IWL_CMD(FH_RSCSR_CHNL0_STTS_WPTR_REG); IWL_CMD(FH_RSCSR_CHNL0_RBDCB_BASE_REG); @@ -262,6 +370,9 @@ int iwl_dump_fh(struct iwl_trans *trans, char **buf) FH_TSSR_TX_ERROR_REG }; + if (trans->cfg->mq_rx_supported) + return iwl_dump_rfh(trans, buf); + #ifdef CONFIG_IWLWIFI_DEBUGFS if (buf) { int pos = 0; -- cgit v0.10.2 From 9c07e9aafcbe4b2a98c18d9b0ee28c84d749ea84 Mon Sep 17 00:00:00 2001 From: Golan Ben-Ami Date: Thu, 19 May 2016 07:46:20 +0300 Subject: iwlwifi: Reserve iwl_fw_error_dump_type enum Reserve a single iwl_fw_error_dump_type enum for external code utilities. Signed-off-by: Golan Ben-Ami Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-error-dump.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-error-dump.h index 09b7ea2..420c31d 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-error-dump.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-error-dump.h @@ -89,6 +89,9 @@ * @IWL_FW_ERROR_PAGING: UMAC's image memory segments which were * paged to the DRAM. * @IWL_FW_ERROR_DUMP_RADIO_REG: Dump the radio registers. + * @IWL_FW_ERROR_DUMP_EXTERNAL: used only by external code utilities, and + * for that reason is not in use in any other place in the Linux Wi-Fi + * stack. */ enum iwl_fw_error_dump_type { /* 0 is deprecated */ @@ -106,6 +109,7 @@ enum iwl_fw_error_dump_type { IWL_FW_ERROR_DUMP_PAGING = 12, IWL_FW_ERROR_DUMP_RADIO_REG = 13, IWL_FW_ERROR_DUMP_INTERNAL_TXF = 14, + IWL_FW_ERROR_DUMP_EXTERNAL = 15, /* Do not move */ IWL_FW_ERROR_DUMP_MAX, }; -- cgit v0.10.2 From 2a53d166c464d7b51d6f7204d445fc0254b25f51 Mon Sep 17 00:00:00 2001 From: Ayala Beker Date: Thu, 7 Apr 2016 16:21:57 +0300 Subject: iwlwifi: mvm: add support for GCMP encryption Newer hardware supports GCMP and GCMP 256-bit ciphers. Add support for adding/setting GCMP key for TX mode. In the TX command handling GCMP-256 is handled in a different way as the key size should be up to 128-bits: Set the key value to the key index in the key table, and specify that this key should be taken form the key table instead of from the TX command. While at it - convert security control flags to an enum. Signed-off-by: Ayala Beker Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h index 38b1d04..d1c4fb8 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-sta.h @@ -141,6 +141,7 @@ enum iwl_sta_flags { * @STA_KEY_FLG_CCM: CCMP encryption algorithm * @STA_KEY_FLG_TKIP: TKIP encryption algorithm * @STA_KEY_FLG_EXT: extended cipher algorithm (depends on the FW support) + * @STA_KEY_FLG_GCMP: GCMP encryption algorithm * @STA_KEY_FLG_CMAC: CMAC encryption algorithm * @STA_KEY_FLG_ENC_UNKNOWN: unknown encryption algorithm * @STA_KEY_FLG_EN_MSK: mask for encryption algorithmi value @@ -149,6 +150,7 @@ enum iwl_sta_flags { * @STA_KEY_FLG_KEYID_MSK: the index of the key * @STA_KEY_NOT_VALID: key is invalid * @STA_KEY_FLG_WEP_13BYTES: set for 13 bytes WEP key + * @STA_KEY_FLG_KEY_32BYTES for non-wep key set for 32 bytes key * @STA_KEY_MULTICAST: set for multical key * @STA_KEY_MFP: key is used for Management Frame Protection */ @@ -158,6 +160,7 @@ enum iwl_sta_key_flag { STA_KEY_FLG_CCM = (2 << 0), STA_KEY_FLG_TKIP = (3 << 0), STA_KEY_FLG_EXT = (4 << 0), + STA_KEY_FLG_GCMP = (5 << 0), STA_KEY_FLG_CMAC = (6 << 0), STA_KEY_FLG_ENC_UNKNOWN = (7 << 0), STA_KEY_FLG_EN_MSK = (7 << 0), @@ -167,6 +170,7 @@ enum iwl_sta_key_flag { STA_KEY_FLG_KEYID_MSK = (3 << STA_KEY_FLG_KEYID_POS), STA_KEY_NOT_VALID = BIT(11), STA_KEY_FLG_WEP_13BYTES = BIT(12), + STA_KEY_FLG_KEY_32BYTES = BIT(12), STA_KEY_MULTICAST = BIT(14), STA_KEY_MFP = BIT(15), }; @@ -388,7 +392,6 @@ struct iwl_mvm_add_sta_cmd { * @key_offset: key offset in key storage * @key_flags: type %iwl_sta_key_flag * @key: key material data - * @key2: key material data * @rx_secur_seq_cnt: RX security sequence counter for the key * @tkip_rx_tsc_byte2: TSC[2] for key mix ph1 detection * @tkip_rx_ttak: 10-byte unicast TKIP TTAK for Rx @@ -397,8 +400,7 @@ struct iwl_mvm_add_sta_key_cmd { u8 sta_id; u8 key_offset; __le16 key_flags; - u8 key[16]; - u8 key2[16]; + u8 key[32]; u8 rx_secur_seq_cnt[16]; u8 tkip_rx_tsc_byte2; u8 reserved; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-tx.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-tx.h index ee59511..4144623 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-tx.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-tx.h @@ -137,17 +137,32 @@ enum iwl_tx_pm_timeouts { PM_FRAME_ASSOC = 3, }; -/* - * TX command security control - */ -#define TX_CMD_SEC_WEP 0x01 -#define TX_CMD_SEC_CCM 0x02 -#define TX_CMD_SEC_TKIP 0x03 -#define TX_CMD_SEC_EXT 0x04 #define TX_CMD_SEC_MSK 0x07 #define TX_CMD_SEC_WEP_KEY_IDX_POS 6 #define TX_CMD_SEC_WEP_KEY_IDX_MSK 0xc0 -#define TX_CMD_SEC_KEY128 0x08 + +/** + * enum iwl_tx_cmd_sec_ctrl - bitmasks for security control in TX command + * @TX_CMD_SEC_WEP: WEP encryption algorithm. + * @TX_CMD_SEC_CCM: CCM encryption algorithm. + * @TX_CMD_SEC_TKIP: TKIP encryption algorithm. + * @TX_CMD_SEC_EXT: extended cipher algorithm. + * @TX_CMD_SEC_GCMP: GCMP encryption algorithm. + * @TX_CMD_SEC_KEY128: set for 104 bits WEP key. + * @TC_CMD_SEC_KEY_FROM_TABLE: for a non-WEP key, set if the key should be taken + * from the table instead of from the TX command. + * If the key is taken from the key table its index should be given by the + * first byte of the TX command key field. + */ +enum iwl_tx_cmd_sec_ctrl { + TX_CMD_SEC_WEP = 0x01, + TX_CMD_SEC_CCM = 0x02, + TX_CMD_SEC_TKIP = 0x03, + TX_CMD_SEC_EXT = 0x04, + TX_CMD_SEC_GCMP = 0x05, + TX_CMD_SEC_KEY128 = 0x08, + TC_CMD_SEC_KEY_FROM_TABLE = 0x08, +}; /* TODO: how does these values are OK with only 16 bit variable??? */ /* diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index af1d266..6b158f0 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -465,11 +465,20 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm) hw->uapsd_queues = IWL_MVM_UAPSD_QUEUES; hw->uapsd_max_sp_len = IWL_UAPSD_MAX_SP; - BUILD_BUG_ON(ARRAY_SIZE(mvm->ciphers) < ARRAY_SIZE(mvm_ciphers) + 2); + BUILD_BUG_ON(ARRAY_SIZE(mvm->ciphers) < ARRAY_SIZE(mvm_ciphers) + 4); memcpy(mvm->ciphers, mvm_ciphers, sizeof(mvm_ciphers)); hw->wiphy->n_cipher_suites = ARRAY_SIZE(mvm_ciphers); hw->wiphy->cipher_suites = mvm->ciphers; + if (iwl_mvm_has_new_rx_api(mvm)) { + mvm->ciphers[hw->wiphy->n_cipher_suites] = + WLAN_CIPHER_SUITE_GCMP; + hw->wiphy->n_cipher_suites++; + mvm->ciphers[hw->wiphy->n_cipher_suites] = + WLAN_CIPHER_SUITE_GCMP_256; + hw->wiphy->n_cipher_suites++; + } + /* * Enable 11w if advertised by firmware and software crypto * is not enabled (as the firmware will interpret some mgmt @@ -2721,6 +2730,8 @@ static int iwl_mvm_mac_set_key(struct ieee80211_hw *hw, key->flags |= IEEE80211_KEY_FLAG_PUT_IV_SPACE; break; case WLAN_CIPHER_SUITE_CCMP: + case WLAN_CIPHER_SUITE_GCMP: + case WLAN_CIPHER_SUITE_GCMP_256: key->flags |= IEEE80211_KEY_FLAG_PUT_IV_SPACE; break; case WLAN_CIPHER_SUITE_AES_CMAC: @@ -2782,7 +2793,8 @@ static int iwl_mvm_mac_set_key(struct ieee80211_hw *hw, sta && iwl_mvm_has_new_rx_api(mvm) && key->flags & IEEE80211_KEY_FLAG_PAIRWISE && (key->cipher == WLAN_CIPHER_SUITE_CCMP || - key->cipher == WLAN_CIPHER_SUITE_GCMP)) { + key->cipher == WLAN_CIPHER_SUITE_GCMP || + key->cipher == WLAN_CIPHER_SUITE_GCMP_256)) { struct ieee80211_key_seq seq; int tid, q; @@ -2836,7 +2848,8 @@ static int iwl_mvm_mac_set_key(struct ieee80211_hw *hw, if (sta && iwl_mvm_has_new_rx_api(mvm) && key->flags & IEEE80211_KEY_FLAG_PAIRWISE && (key->cipher == WLAN_CIPHER_SUITE_CCMP || - key->cipher == WLAN_CIPHER_SUITE_GCMP)) { + key->cipher == WLAN_CIPHER_SUITE_GCMP || + key->cipher == WLAN_CIPHER_SUITE_GCMP_256)) { mvmsta = iwl_mvm_sta_from_mac80211(sta); ptk_pn = rcu_dereference_protected( mvmsta->ptk_pn[keyidx], diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index 6092af2..e240f19 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -707,6 +707,7 @@ enum iwl_mvm_queue_status { }; #define IWL_MVM_DQA_QUEUE_TIMEOUT (5 * HZ) +#define IWL_MVM_NUM_CIPHERS 8 struct iwl_mvm { /* for logger access */ @@ -1015,7 +1016,7 @@ struct iwl_mvm { struct iwl_mvm_shared_mem_cfg shared_mem_cfg; - u32 ciphers[6]; + u32 ciphers[IWL_MVM_NUM_CIPHERS]; struct iwl_mvm_tof_data tof_data; struct ieee80211_vif *nan_vif; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index cf06817..7321bb6 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -2243,6 +2243,13 @@ static int iwl_mvm_send_sta_key(struct iwl_mvm *mvm, key_flags |= cpu_to_le16(STA_KEY_FLG_WEP); memcpy(cmd.key + 3, keyconf->key, keyconf->keylen); break; + case WLAN_CIPHER_SUITE_GCMP_256: + key_flags |= cpu_to_le16(STA_KEY_FLG_KEY_32BYTES); + /* fall through */ + case WLAN_CIPHER_SUITE_GCMP: + key_flags |= cpu_to_le16(STA_KEY_FLG_GCMP); + memcpy(cmd.key, keyconf->key, keyconf->keylen); + break; default: key_flags |= cpu_to_le16(STA_KEY_FLG_EXT); memcpy(cmd.key, keyconf->key, keyconf->keylen); @@ -2363,6 +2370,8 @@ static int __iwl_mvm_set_sta_key(struct iwl_mvm *mvm, case WLAN_CIPHER_SUITE_CCMP: case WLAN_CIPHER_SUITE_WEP40: case WLAN_CIPHER_SUITE_WEP104: + case WLAN_CIPHER_SUITE_GCMP: + case WLAN_CIPHER_SUITE_GCMP_256: ret = iwl_mvm_send_sta_key(mvm, mvm_sta, keyconf, mcast, 0, NULL, 0, key_offset); break; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c index 1f23cee..2373142 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c @@ -388,6 +388,23 @@ void iwl_mvm_set_tx_cmd_rate(struct iwl_mvm *mvm, struct iwl_tx_cmd *tx_cmd, tx_cmd->rate_n_flags = cpu_to_le32((u32)rate_plcp | rate_flags); } +static inline void iwl_mvm_set_tx_cmd_pn(struct ieee80211_tx_info *info, + u8 *crypto_hdr) +{ + struct ieee80211_key_conf *keyconf = info->control.hw_key; + u64 pn; + + pn = atomic64_inc_return(&keyconf->tx_pn); + crypto_hdr[0] = pn; + crypto_hdr[2] = 0; + crypto_hdr[3] = 0x20 | (keyconf->keyidx << 6); + crypto_hdr[1] = pn >> 8; + crypto_hdr[4] = pn >> 16; + crypto_hdr[5] = pn >> 24; + crypto_hdr[6] = pn >> 32; + crypto_hdr[7] = pn >> 40; +} + /* * Sets the fields in the Tx cmd that are crypto related */ @@ -405,15 +422,7 @@ static void iwl_mvm_set_tx_cmd_crypto(struct iwl_mvm *mvm, case WLAN_CIPHER_SUITE_CCMP: case WLAN_CIPHER_SUITE_CCMP_256: iwl_mvm_set_tx_cmd_ccmp(info, tx_cmd); - pn = atomic64_inc_return(&keyconf->tx_pn); - crypto_hdr[0] = pn; - crypto_hdr[2] = 0; - crypto_hdr[3] = 0x20 | (keyconf->keyidx << 6); - crypto_hdr[1] = pn >> 8; - crypto_hdr[4] = pn >> 16; - crypto_hdr[5] = pn >> 24; - crypto_hdr[6] = pn >> 32; - crypto_hdr[7] = pn >> 40; + iwl_mvm_set_tx_cmd_pn(info, crypto_hdr); break; case WLAN_CIPHER_SUITE_TKIP: @@ -433,6 +442,18 @@ static void iwl_mvm_set_tx_cmd_crypto(struct iwl_mvm *mvm, memcpy(&tx_cmd->key[3], keyconf->key, keyconf->keylen); break; + case WLAN_CIPHER_SUITE_GCMP: + case WLAN_CIPHER_SUITE_GCMP_256: + /* TODO: Taking the key from the table might introduce a race + * when PTK rekeying is done, having an old packets with a PN + * based on the old key but the message encrypted with a new + * one. + * Need to handle this. + */ + tx_cmd->sec_ctl |= TX_CMD_SEC_GCMP | TC_CMD_SEC_KEY_FROM_TABLE; + tx_cmd->key[0] = keyconf->hw_key_idx; + iwl_mvm_set_tx_cmd_pn(info, crypto_hdr); + break; default: tx_cmd->sec_ctl |= TX_CMD_SEC_EXT; } -- cgit v0.10.2 From 988b59684d7681bfb4eb508eeed3abb2bfde7af6 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Sun, 10 Apr 2016 15:06:55 +0300 Subject: iwlwifi: mvm: support new statistics notification For 9000 family we will get extended statistics notification with averaged data for RSSI, TCM and rogue AP detection. Support it. Future patches will added the required algorithms. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-mac.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-mac.h index 95ac59d..0246506 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-mac.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-mac.h @@ -72,6 +72,9 @@ #define NUM_MAC_INDEX_DRIVER MAC_INDEX_AUX #define NUM_MAC_INDEX (MAC_INDEX_AUX + 1) +#define IWL_MVM_STATION_COUNT 16 +#define IWL_MVM_TDLS_STA_COUNT 4 + enum iwl_ac { AC_BK, AC_BE, diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-stats.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-stats.h index 438665a..4e638a4 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-stats.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-stats.h @@ -7,6 +7,7 @@ * * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved. * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH + * Copyright(c) 2016 Intel Deutschland GmbH * * This program is free software; you can redistribute it and/or modify * it under the terms of version 2 of the GNU General Public License as @@ -252,6 +253,20 @@ struct mvm_statistics_general_v8 { u8 reserved[4 - (NUM_MAC_INDEX % 4)]; } __packed; /* STATISTICS_GENERAL_API_S_VER_8 */ +/** + * struct mvm_statistics_load - RX statistics for multi-queue devices + * @air_time: accumulated air time, per mac + * @byte_count: accumulated byte count, per mac + * @pkt_count: accumulated packet count, per mac + * @avg_energy: average RSSI, per station + */ +struct mvm_statistics_load { + __le32 air_time[NUM_MAC_INDEX]; + __le32 byte_count[NUM_MAC_INDEX]; + __le32 pkt_count[NUM_MAC_INDEX]; + u8 avg_energy[IWL_MVM_STATION_COUNT]; +} __packed; /* STATISTICS_RX_MAC_STATION_S_VER_1 */ + struct mvm_statistics_rx { struct mvm_statistics_rx_phy ofdm; struct mvm_statistics_rx_phy cck; @@ -266,7 +281,6 @@ struct mvm_statistics_rx { * while associated. To disable this behavior, set DISABLE_NOTIF flag in the * STATISTICS_CMD (0x9c), below. */ - struct iwl_notif_statistics_v10 { __le32 flag; struct mvm_statistics_rx rx; @@ -274,6 +288,14 @@ struct iwl_notif_statistics_v10 { struct mvm_statistics_general_v8 general; } __packed; /* STATISTICS_NTFY_API_S_VER_10 */ +struct iwl_notif_statistics_v11 { + __le32 flag; + struct mvm_statistics_rx rx; + struct mvm_statistics_tx tx; + struct mvm_statistics_general_v8 general; + struct mvm_statistics_load load_stats; +} __packed; /* STATISTICS_NTFY_API_S_VER_11 */ + #define IWL_STATISTICS_FLG_CLEAR 0x1 #define IWL_STATISTICS_FLG_DISABLE_NOTIF 0x2 diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h index eea03ec..71076f0 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api.h @@ -129,9 +129,6 @@ enum iwl_mvm_tx_fifo { IWL_MVM_TX_FIFO_CMD = 7, }; -#define IWL_MVM_STATION_COUNT 16 - -#define IWL_MVM_TDLS_STA_COUNT 4 /* commands */ enum { diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index 6b158f0..5c044ea 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -3922,6 +3922,11 @@ static void iwl_mvm_mac_sta_statistics(struct ieee80211_hw *hw, struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif); struct iwl_mvm_sta *mvmsta = iwl_mvm_sta_from_mac80211(sta); + if (mvmsta->avg_energy) { + sinfo->signal_avg = mvmsta->avg_energy; + sinfo->filled |= BIT(NL80211_STA_INFO_SIGNAL_AVG); + } + if (fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_RADIO_BEACON_STATS)) return; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c index 2d5a7cc..0e60e38 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rx.c @@ -498,6 +498,7 @@ struct iwl_mvm_stat_data { __le32 mac_id; u8 beacon_filter_average_energy; struct mvm_statistics_general_v8 *general; + struct mvm_statistics_load *load; }; static void iwl_mvm_stat_iterator(void *_data, u8 *mac, @@ -614,13 +615,15 @@ iwl_mvm_rx_stats_check_trigger(struct iwl_mvm *mvm, struct iwl_rx_packet *pkt) void iwl_mvm_handle_rx_statistics(struct iwl_mvm *mvm, struct iwl_rx_packet *pkt) { - struct iwl_notif_statistics_v10 *stats = (void *)&pkt->data; + struct iwl_notif_statistics_v11 *stats = (void *)&pkt->data; struct iwl_mvm_stat_data data = { .mvm = mvm, }; + int expected_size = iwl_mvm_has_new_rx_api(mvm) ? sizeof(*stats) : + sizeof(struct iwl_notif_statistics_v10); u32 temperature; - if (iwl_rx_packet_payload_len(pkt) != sizeof(*stats)) + if (iwl_rx_packet_payload_len(pkt) != expected_size) goto invalid; temperature = le32_to_cpu(stats->general.radio_temperature); @@ -638,6 +641,25 @@ void iwl_mvm_handle_rx_statistics(struct iwl_mvm *mvm, le64_to_cpu(stats->general.on_time_scan); data.general = &stats->general; + if (iwl_mvm_has_new_rx_api(mvm)) { + int i; + + data.load = &stats->load_stats; + + rcu_read_lock(); + for (i = 0; i < IWL_MVM_STATION_COUNT; i++) { + struct iwl_mvm_sta *sta; + + if (!data.load->avg_energy[i]) + continue; + + sta = iwl_mvm_sta_from_staid_rcu(mvm, i); + if (!sta) + continue; + sta->avg_energy = data.load->avg_energy[i]; + } + rcu_read_unlock(); + } iwl_mvm_rx_stats_check_trigger(mvm, pkt); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h index 1588eb6..bbc1cab 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.h @@ -438,6 +438,7 @@ struct iwl_mvm_sta { bool tlc_amsdu; u8 agg_tids; u8 sleep_tx_count; + u8 avg_energy; }; static inline struct iwl_mvm_sta * -- cgit v0.10.2 From e34d975e40ff69d55cb45f968f5ca0892e114b92 Mon Sep 17 00:00:00 2001 From: Haim Dreyfuss Date: Tue, 31 May 2016 11:13:09 +0300 Subject: iwlwifi: Add a000 HW family support Add a000 family configuration to iwl-cfg struct Signed-off-by: Haim Dreyfuss Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/Makefile b/drivers/net/wireless/intel/iwlwifi/Makefile index 05828c6..6e7ed90 100644 --- a/drivers/net/wireless/intel/iwlwifi/Makefile +++ b/drivers/net/wireless/intel/iwlwifi/Makefile @@ -8,7 +8,7 @@ iwlwifi-objs += iwl-eeprom-read.o iwl-eeprom-parse.o iwlwifi-objs += iwl-phy-db.o iwl-nvm-parse.o iwlwifi-objs += pcie/drv.o pcie/rx.o pcie/tx.o pcie/trans.o iwlwifi-$(CONFIG_IWLDVM) += iwl-1000.o iwl-2000.o iwl-5000.o iwl-6000.o -iwlwifi-$(CONFIG_IWLMVM) += iwl-7000.o iwl-8000.o iwl-9000.o +iwlwifi-$(CONFIG_IWLMVM) += iwl-7000.o iwl-8000.o iwl-9000.o iwl-a000.o iwlwifi-objs += iwl-trans.o iwlwifi-objs += $(iwlwifi-m) diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-a000.c b/drivers/net/wireless/intel/iwlwifi/iwl-a000.c new file mode 100644 index 0000000..220767e --- /dev/null +++ b/drivers/net/wireless/intel/iwlwifi/iwl-a000.c @@ -0,0 +1,130 @@ +/****************************************************************************** + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * Copyright(c) 2015-2016 Intel Deutschland GmbH + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Copyright(c) 2015-2016 Intel Deutschland GmbH + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * * Neither the name Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + *****************************************************************************/ + +#include +#include +#include "iwl-config.h" +#include "iwl-agn-hw.h" + +/* Highest firmware API version supported */ +#define IWL_A000_UCODE_API_MAX 24 + +/* Lowest firmware API version supported */ +#define IWL_A000_UCODE_API_MIN 24 + +/* NVM versions */ +#define IWL_A000_NVM_VERSION 0x0a1d +#define IWL_A000_TX_POWER_VERSION 0xffff /* meaningless */ + +/* Memory offsets and lengths */ +#define IWL_A000_DCCM_OFFSET 0x800000 +#define IWL_A000_DCCM_LEN 0x18000 +#define IWL_A000_DCCM2_OFFSET 0x880000 +#define IWL_A000_DCCM2_LEN 0x8000 +#define IWL_A000_SMEM_OFFSET 0x400000 +#define IWL_A000_SMEM_LEN 0x68000 + +#define IWL_A000_FW_PRE "iwlwifi-Qu-a0-jf-b0-" +#define IWL_A000_MODULE_FIRMWARE(api) \ + IWL_A000_FW_PRE "-" __stringify(api) ".ucode" + +#define NVM_HW_SECTION_NUM_FAMILY_A000 10 + +static const struct iwl_base_params iwl_a000_base_params = { + .eeprom_size = OTP_LOW_IMAGE_SIZE_FAMILY_A000, + .num_of_queues = 31, + .shadow_ram_support = true, + .led_compensation = 57, + .wd_timeout = IWL_LONG_WD_TIMEOUT, + .max_event_log_size = 512, + .shadow_reg_enable = true, + .pcie_l1_allowed = true, +}; + +static const struct iwl_ht_params iwl_a000_ht_params = { + .stbc = true, + .ldpc = true, + .ht40_bands = BIT(NL80211_BAND_2GHZ) | BIT(NL80211_BAND_5GHZ), +}; + +#define IWL_DEVICE_A000 \ + .ucode_api_max = IWL_A000_UCODE_API_MAX, \ + .ucode_api_min = IWL_A000_UCODE_API_MIN, \ + .device_family = IWL_DEVICE_FAMILY_8000, \ + .max_inst_size = IWL60_RTC_INST_SIZE, \ + .max_data_size = IWL60_RTC_DATA_SIZE, \ + .base_params = &iwl_a000_base_params, \ + .led_mode = IWL_LED_RF_STATE, \ + .nvm_hw_section_num = NVM_HW_SECTION_NUM_FAMILY_A000, \ + .non_shared_ant = ANT_A, \ + .dccm_offset = IWL_A000_DCCM_OFFSET, \ + .dccm_len = IWL_A000_DCCM_LEN, \ + .dccm2_offset = IWL_A000_DCCM2_OFFSET, \ + .dccm2_len = IWL_A000_DCCM2_LEN, \ + .smem_offset = IWL_A000_SMEM_OFFSET, \ + .smem_len = IWL_A000_SMEM_LEN, \ + .features = IWL_TX_CSUM_NETIF_FLAGS | NETIF_F_RXCSUM, \ + .apmg_not_supported = true, \ + .mq_rx_supported = true, \ + .vht_mu_mimo_supported = true, \ + .mac_addr_from_csr = true + +const struct iwl_cfg iwla000_2ac_cfg = { + .name = "Intel(R) Dual Band Wireless AC a000", + .fw_name_pre = IWL_A000_FW_PRE, + IWL_DEVICE_A000, + .ht_params = &iwl_a000_ht_params, + .nvm_ver = IWL_A000_NVM_VERSION, + .nvm_calib_ver = IWL_A000_TX_POWER_VERSION, + .max_ht_ampdu_exponent = IEEE80211_HT_MAX_AMPDU_64K, +}; + +MODULE_FIRMWARE(IWL_A000_MODULE_FIRMWARE(IWL_A000_UCODE_API_MAX)); diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h index 57b14f4..122e895 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h @@ -261,6 +261,7 @@ struct iwl_tt_params { #define OTP_LOW_IMAGE_SIZE_FAMILY_7000 (16 * 512 * sizeof(u16)) /* 16 KB */ #define OTP_LOW_IMAGE_SIZE_FAMILY_8000 (32 * 512 * sizeof(u16)) /* 32 KB */ #define OTP_LOW_IMAGE_SIZE_FAMILY_9000 OTP_LOW_IMAGE_SIZE_FAMILY_8000 +#define OTP_LOW_IMAGE_SIZE_FAMILY_A000 OTP_LOW_IMAGE_SIZE_FAMILY_9000 struct iwl_eeprom_params { const u8 regulatory_bands[7]; @@ -450,6 +451,7 @@ extern const struct iwl_cfg iwl4165_2ac_sdio_cfg; extern const struct iwl_cfg iwl9260_2ac_cfg; extern const struct iwl_cfg iwl9260lc_2ac_cfg; extern const struct iwl_cfg iwl5165_2ac_cfg; +extern const struct iwl_cfg iwla000_2ac_cfg; #endif /* CONFIG_IWLMVM */ #endif /* __IWL_CONFIG_H__ */ diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index 6f020e4..f70c3e2 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -516,6 +516,9 @@ static const struct pci_device_id iwl_hw_card_ids[] = { {IWL_PCI_DEVICE(0x2526, 0x1420, iwl5165_2ac_cfg)}, {IWL_PCI_DEVICE(0x9DF0, 0x0710, iwl5165_2ac_cfg)}, {IWL_PCI_DEVICE(0x9DF0, 0x2A10, iwl5165_2ac_cfg)}, + +/* a000 Series */ + {IWL_PCI_DEVICE(0x2720, 0x0A10, iwla000_2ac_cfg)}, #endif /* CONFIG_IWLMVM */ {0} -- cgit v0.10.2 From 2aabdbdc17b7c53490337bfc58de3409c84d85d2 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Wed, 8 Jun 2016 23:07:31 +0300 Subject: iwlwifi: pcie: enable interrupts before releasing the NIC's CPU The NIC's CPU gets started after the firmware has been written to its memory. The first thing it does is to send an interrupt to let the driver know that it is running. In order to get that interrupt, the driver needs to make sure it is not masked. Of course, the interrupt needs to be enabled in the driver before the CPU starts to run. I mistakenly inversed those two steps leading to races which prevented the driver from getting the alive interrupt from the firmware. Fix that. Cc: [4.5+] Fixes: a6bd005fe92 ("iwlwifi: pcie: fix RF-Kill vs. firmware load race") Signed-off-by: Emmanuel Grumbach Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index 3badebb..ac623c3 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -801,6 +801,8 @@ static int iwl_pcie_load_cpu_sections_8000(struct iwl_trans *trans, *first_ucode_section = last_read_idx; + iwl_enable_interrupts(trans); + if (cpu == 1) iwl_write_direct32(trans, FH_UCODE_LOAD_STATUS, 0xFFFF); else @@ -980,6 +982,8 @@ static int iwl_pcie_load_given_ucode(struct iwl_trans *trans, iwl_pcie_apply_destination(trans); } + iwl_enable_interrupts(trans); + /* release CPU reset */ iwl_write32(trans, CSR_RESET, 0); @@ -1215,7 +1219,6 @@ static int iwl_trans_pcie_start_fw(struct iwl_trans *trans, ret = iwl_pcie_load_given_ucode_8000(trans, fw); else ret = iwl_pcie_load_given_ucode(trans, fw); - iwl_enable_interrupts(trans); /* re-check RF-Kill state since we may have missed the interrupt */ hw_rfkill = iwl_is_rfkill_set(trans); -- cgit v0.10.2 From 849a9627299100ae3f0ce573fc87d2b476f3bb59 Mon Sep 17 00:00:00 2001 From: striebit Date: Tue, 7 Jun 2016 15:05:26 +0300 Subject: iwlmvm: mvm: set correct state in smart-fifo configuration Currently the state sent in SF configuration is always FULL_ON. This commit sets the correct state (e.g. INIT_OFF when station is not associated). Fixes: commit f4a3ee493e69 ("iwlwifi: mvm: Always enable the smart FIFO") Signed-off-by: Shaul Triebitz Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sf.c b/drivers/net/wireless/intel/iwlwifi/mvm/sf.c index 443a428..101fb04 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sf.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sf.c @@ -215,7 +215,7 @@ static int iwl_mvm_sf_config(struct iwl_mvm *mvm, u8 sta_id, enum iwl_sf_state new_state) { struct iwl_sf_cfg_cmd sf_cmd = { - .state = cpu_to_le32(SF_FULL_ON), + .state = cpu_to_le32(new_state), }; struct ieee80211_sta *sta; int ret = 0; -- cgit v0.10.2 From ecf51424152bad1b2727409f42ddf1bd86f44b7d Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Wed, 8 Jun 2016 15:15:41 +0300 Subject: iwlwifi: mvm: checksum IPv6 fragmented packet Our HW does not support checksum of fragmented packets. Fix code accordingly to checksum those packets in the driver. Signed-off-by: Sara Sharon Fixes: 5e6a98dc4863 ("iwlwifi: mvm: enable TCP/UDP checksum support for 9000 family") Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c index 2373142..f3a6e95 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c @@ -138,28 +138,19 @@ static void iwl_mvm_tx_csum(struct iwl_mvm *mvm, struct sk_buff *skb, protocol = ipv6h->nexthdr; while (protocol != NEXTHDR_NONE && ipv6_ext_hdr(protocol)) { + struct ipv6_opt_hdr *hp; + /* only supported extension headers */ if (protocol != NEXTHDR_ROUTING && protocol != NEXTHDR_HOP && - protocol != NEXTHDR_DEST && - protocol != NEXTHDR_FRAGMENT) { + protocol != NEXTHDR_DEST) { skb_checksum_help(skb); return; } - if (protocol == NEXTHDR_FRAGMENT) { - struct frag_hdr *hp = - OPT_HDR(struct frag_hdr, skb, off); - - protocol = hp->nexthdr; - off += sizeof(struct frag_hdr); - } else { - struct ipv6_opt_hdr *hp = - OPT_HDR(struct ipv6_opt_hdr, skb, off); - - protocol = hp->nexthdr; - off += ipv6_optlen(hp); - } + hp = OPT_HDR(struct ipv6_opt_hdr, skb, off); + protocol = hp->nexthdr; + off += ipv6_optlen(hp); } /* if we get here - protocol now should be TCP/UDP */ #endif -- cgit v0.10.2 From 3aa4359fe1fbfe25be4b474d20094138d04a429a Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Mon, 13 Jun 2016 10:02:04 +0300 Subject: iwlwifi: mvm: cleanup the coex code We removed support for old API for coexistence, but we forgot to remove defines and variable that are not needed anymore. Signed-off-by: Emmanuel Grumbach Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-coex.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-coex.h index 2a33b69..204c1b1 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-coex.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-coex.h @@ -70,85 +70,6 @@ #define BITS(nb) (BIT(nb) - 1) -/** - * enum iwl_bt_coex_flags - flags for BT_COEX command - * @BT_COEX_MODE_POS: - * @BT_COEX_MODE_MSK: - * @BT_COEX_DISABLE_OLD: - * @BT_COEX_2W_OLD: - * @BT_COEX_3W_OLD: - * @BT_COEX_NW_OLD: - * @BT_COEX_AUTO_OLD: - * @BT_COEX_BT_OLD: Antenna is for BT (manufacuring tests) - * @BT_COEX_WIFI_OLD: Antenna is for BT (manufacuring tests) - * @BT_COEX_SYNC2SCO: - * @BT_COEX_CORUNNING: - * @BT_COEX_MPLUT: - * @BT_COEX_TTC: - * @BT_COEX_RRC: - * - * The COEX_MODE must be set for each command. Even if it is not changed. - */ -enum iwl_bt_coex_flags { - BT_COEX_MODE_POS = 3, - BT_COEX_MODE_MSK = BITS(3) << BT_COEX_MODE_POS, - BT_COEX_DISABLE_OLD = 0x0 << BT_COEX_MODE_POS, - BT_COEX_2W_OLD = 0x1 << BT_COEX_MODE_POS, - BT_COEX_3W_OLD = 0x2 << BT_COEX_MODE_POS, - BT_COEX_NW_OLD = 0x3 << BT_COEX_MODE_POS, - BT_COEX_AUTO_OLD = 0x5 << BT_COEX_MODE_POS, - BT_COEX_BT_OLD = 0x6 << BT_COEX_MODE_POS, - BT_COEX_WIFI_OLD = 0x7 << BT_COEX_MODE_POS, - BT_COEX_SYNC2SCO = BIT(7), - BT_COEX_CORUNNING = BIT(8), - BT_COEX_MPLUT = BIT(9), - BT_COEX_TTC = BIT(20), - BT_COEX_RRC = BIT(21), -}; - -/* - * indicates what has changed in the BT_COEX command. - * BT_VALID_ENABLE must be set for each command. Commands without this bit will - * discarded by the firmware - */ -enum iwl_bt_coex_valid_bit_msk { - BT_VALID_ENABLE = BIT(0), - BT_VALID_BT_PRIO_BOOST = BIT(1), - BT_VALID_MAX_KILL = BIT(2), - BT_VALID_3W_TMRS = BIT(3), - BT_VALID_KILL_ACK = BIT(4), - BT_VALID_KILL_CTS = BIT(5), - BT_VALID_REDUCED_TX_POWER = BIT(6), - BT_VALID_LUT = BIT(7), - BT_VALID_WIFI_RX_SW_PRIO_BOOST = BIT(8), - BT_VALID_WIFI_TX_SW_PRIO_BOOST = BIT(9), - BT_VALID_MULTI_PRIO_LUT = BIT(10), - BT_VALID_TRM_KICK_FILTER = BIT(11), - BT_VALID_CORUN_LUT_20 = BIT(12), - BT_VALID_CORUN_LUT_40 = BIT(13), - BT_VALID_ANT_ISOLATION = BIT(14), - BT_VALID_ANT_ISOLATION_THRS = BIT(15), - BT_VALID_TXTX_DELTA_FREQ_THRS = BIT(16), - BT_VALID_TXRX_MAX_FREQ_0 = BIT(17), - BT_VALID_SYNC_TO_SCO = BIT(18), - BT_VALID_TTC = BIT(20), - BT_VALID_RRC = BIT(21), -}; - -/** - * enum iwl_bt_reduced_tx_power - allows to reduce txpower for WiFi frames. - * @BT_REDUCED_TX_POWER_CTL: reduce Tx power for control frames - * @BT_REDUCED_TX_POWER_DATA: reduce Tx power for data frames - * - * This mechanism allows to have BT and WiFi run concurrently. Since WiFi - * reduces its Tx power, it can work along with BT, hence reducing the amount - * of WiFi frames being killed by BT. - */ -enum iwl_bt_reduced_tx_power { - BT_REDUCED_TX_POWER_CTL = BIT(0), - BT_REDUCED_TX_POWER_DATA = BIT(1), -}; - enum iwl_bt_coex_lut_type { BT_COEX_TIGHT_LUT = 0, BT_COEX_LOOSE_LUT, @@ -158,64 +79,9 @@ enum iwl_bt_coex_lut_type { BT_COEX_INVALID_LUT = 0xff, }; /* BT_COEX_DECISION_LUT_INDEX_API_E_VER_1 */ -#define BT_COEX_LUT_SIZE (12) #define BT_COEX_CORUN_LUT_SIZE (32) -#define BT_COEX_MULTI_PRIO_LUT_SIZE (2) -#define BT_COEX_BOOST_SIZE (4) #define BT_REDUCED_TX_POWER_BIT BIT(7) -/** - * struct iwl_bt_coex_cmd_old - bt coex configuration command - * @flags:&enum iwl_bt_coex_flags - * @max_kill: - * @bt_reduced_tx_power: enum %iwl_bt_reduced_tx_power - * @override_primary_lut: enum %iwl_bt_coex_lut_type: BT_COEX_INVALID_LUT - * should be set by default - * @override_secondary_lut: enum %iwl_bt_coex_lut_type: BT_COEX_INVALID_LUT - * should be set by default - * @bt4_antenna_isolation: antenna isolation - * @bt4_antenna_isolation_thr: antenna threshold value - * @bt4_tx_tx_delta_freq_thr: TxTx delta frequency - * @bt4_tx_rx_max_freq0: TxRx max frequency - * @bt_prio_boost: BT priority boost registers - * @wifi_tx_prio_boost: SW boost of wifi tx priority - * @wifi_rx_prio_boost: SW boost of wifi rx priority - * @kill_ack_msk: kill ACK mask. 1 - Tx ACK, 0 - kill Tx of ACK. - * @kill_cts_msk: kill CTS mask. 1 - Tx CTS, 0 - kill Tx of CTS. - * @decision_lut: PTA decision LUT, per Prio-Ch - * @bt4_multiprio_lut: multi priority LUT configuration - * @bt4_corun_lut20: co-running 20 MHz LUT configuration - * @bt4_corun_lut40: co-running 40 MHz LUT configuration - * @valid_bit_msk: enum %iwl_bt_coex_valid_bit_msk - * - * The structure is used for the BT_COEX command. - */ -struct iwl_bt_coex_cmd_old { - __le32 flags; - u8 max_kill; - u8 bt_reduced_tx_power; - u8 override_primary_lut; - u8 override_secondary_lut; - - u8 bt4_antenna_isolation; - u8 bt4_antenna_isolation_thr; - u8 bt4_tx_tx_delta_freq_thr; - u8 bt4_tx_rx_max_freq0; - - __le32 bt_prio_boost[BT_COEX_BOOST_SIZE]; - __le32 wifi_tx_prio_boost; - __le32 wifi_rx_prio_boost; - __le32 kill_ack_msk; - __le32 kill_cts_msk; - - __le32 decision_lut[BT_COEX_MAX_LUT][BT_COEX_LUT_SIZE]; - __le32 bt4_multiprio_lut[BT_COEX_MULTI_PRIO_LUT_SIZE]; - __le32 bt4_corun_lut20[BT_COEX_CORUN_LUT_SIZE]; - __le32 bt4_corun_lut40[BT_COEX_CORUN_LUT_SIZE]; - - __le32 valid_bit_msk; -} __packed; /* BT_COEX_CMD_API_S_VER_5 */ - enum iwl_bt_coex_mode { BT_COEX_DISABLE = 0x0, BT_COEX_NW = 0x1, @@ -385,92 +251,4 @@ struct iwl_bt_coex_profile_notif { u8 reserved[3]; } __packed; /* BT_COEX_PROFILE_NTFY_API_S_VER_4 */ -enum iwl_bt_coex_prio_table_event { - BT_COEX_PRIO_TBL_EVT_INIT_CALIB1 = 0, - BT_COEX_PRIO_TBL_EVT_INIT_CALIB2 = 1, - BT_COEX_PRIO_TBL_EVT_PERIODIC_CALIB_LOW1 = 2, - BT_COEX_PRIO_TBL_EVT_PERIODIC_CALIB_LOW2 = 3, - BT_COEX_PRIO_TBL_EVT_PERIODIC_CALIB_HIGH1 = 4, - BT_COEX_PRIO_TBL_EVT_PERIODIC_CALIB_HIGH2 = 5, - BT_COEX_PRIO_TBL_EVT_DTIM = 6, - BT_COEX_PRIO_TBL_EVT_SCAN52 = 7, - BT_COEX_PRIO_TBL_EVT_SCAN24 = 8, - BT_COEX_PRIO_TBL_EVT_IDLE = 9, - BT_COEX_PRIO_TBL_EVT_MAX = 16, -}; /* BT_COEX_PRIO_TABLE_EVENTS_API_E_VER_1 */ - -enum iwl_bt_coex_prio_table_prio { - BT_COEX_PRIO_TBL_DISABLED = 0, - BT_COEX_PRIO_TBL_PRIO_LOW = 1, - BT_COEX_PRIO_TBL_PRIO_HIGH = 2, - BT_COEX_PRIO_TBL_PRIO_BYPASS = 3, - BT_COEX_PRIO_TBL_PRIO_COEX_OFF = 4, - BT_COEX_PRIO_TBL_PRIO_COEX_ON = 5, - BT_COEX_PRIO_TBL_PRIO_COEX_IDLE = 6, - BT_COEX_PRIO_TBL_MAX = 8, -}; /* BT_COEX_PRIO_TABLE_PRIORITIES_API_E_VER_1 */ - -#define BT_COEX_PRIO_TBL_SHRD_ANT_POS (0) -#define BT_COEX_PRIO_TBL_PRIO_POS (1) -#define BT_COEX_PRIO_TBL_RESERVED_POS (4) - -/** - * struct iwl_bt_coex_prio_tbl_cmd - priority table for BT coex - * @prio_tbl: - */ -struct iwl_bt_coex_prio_tbl_cmd { - u8 prio_tbl[BT_COEX_PRIO_TBL_EVT_MAX]; -} __packed; - -/** - * struct iwl_bt_coex_ci_cmd_old - bt coex channel inhibition command - * @bt_primary_ci: - * @bt_secondary_ci: - * @co_run_bw_primary: - * @co_run_bw_secondary: - * @primary_ch_phy_id: - * @secondary_ch_phy_id: - * - * Used for BT_COEX_CI command - */ -struct iwl_bt_coex_ci_cmd_old { - __le64 bt_primary_ci; - __le64 bt_secondary_ci; - - u8 co_run_bw_primary; - u8 co_run_bw_secondary; - u8 primary_ch_phy_id; - u8 secondary_ch_phy_id; -} __packed; /* BT_CI_MSG_API_S_VER_1 */ - -/** - * struct iwl_bt_coex_profile_notif_old - notification about BT coex - * @mbox_msg: message from BT to WiFi - * @msg_idx: the index of the message - * @bt_status: 0 - off, 1 - on - * @bt_open_conn: number of BT connections open - * @bt_traffic_load: load of BT traffic - * @bt_agg_traffic_load: aggregated load of BT traffic - * @bt_ci_compliance: 0 - no CI compliance, 1 - CI compliant - * @primary_ch_lut: LUT used for primary channel - * @secondary_ch_lut: LUT used for secondary channel - * @bt_activity_grading: the activity of BT enum %iwl_bt_activity_grading - */ -struct iwl_bt_coex_profile_notif_old { - __le32 mbox_msg[4]; - __le32 msg_idx; - u8 bt_status; - u8 bt_open_conn; - u8 bt_traffic_load; - u8 bt_agg_traffic_load; - u8 bt_ci_compliance; - u8 ttc_enabled; - u8 rrc_enabled; - u8 reserved; - - __le32 primary_ch_lut; - __le32 secondary_ch_lut; - __le32 bt_activity_grading; -} __packed; /* BT_COEX_PROFILE_NTFY_API_S_VER_3 */ - #endif /* __fw_api_bt_coex_h__ */ diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index 5c044ea..e44c337 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -1020,11 +1020,7 @@ static void iwl_mvm_restart_cleanup(struct iwl_mvm *mvm) memset(mvm->sta_deferred_frames, 0, sizeof(mvm->sta_deferred_frames)); memset(mvm->tfd_drained, 0, sizeof(mvm->tfd_drained)); memset(&mvm->last_bt_notif, 0, sizeof(mvm->last_bt_notif)); - memset(&mvm->last_bt_notif_old, 0, sizeof(mvm->last_bt_notif_old)); memset(&mvm->last_bt_ci_cmd, 0, sizeof(mvm->last_bt_ci_cmd)); - memset(&mvm->last_bt_ci_cmd_old, 0, sizeof(mvm->last_bt_ci_cmd_old)); - memset(&mvm->bt_ack_kill_msk, 0, sizeof(mvm->bt_ack_kill_msk)); - memset(&mvm->bt_cts_kill_msk, 0, sizeof(mvm->bt_cts_kill_msk)); ieee80211_wake_queues(mvm->hw); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index e240f19..c440b10 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -932,11 +932,6 @@ struct iwl_mvm { wait_queue_head_t d0i3_exit_waitq; /* BT-Coex */ - u8 bt_ack_kill_msk[NUM_PHY_CTX]; - u8 bt_cts_kill_msk[NUM_PHY_CTX]; - - struct iwl_bt_coex_profile_notif_old last_bt_notif_old; - struct iwl_bt_coex_ci_cmd_old last_bt_ci_cmd_old; struct iwl_bt_coex_profile_notif last_bt_notif; struct iwl_bt_coex_ci_cmd last_bt_ci_cmd; -- cgit v0.10.2 From da2830acf15aab9cb2c376b74463d6d6ecc18716 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Mon, 30 May 2016 13:00:44 +0300 Subject: iwlwifi: mvm: read SAR BIOS table from ACPI Read the SAR BIOS table from the ACPI and parse it into the iwl_mvm_sar_table structure. If the table is enabled, send it to the firmware via REDUCE_TX_POWER_CMD. Signed-off-by: Luca Coelho Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index 510b4c1..4db1d84 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -65,6 +65,7 @@ *****************************************************************************/ #include #include +#include #include "iwl-trans.h" #include "iwl-op-mode.h" @@ -902,6 +903,174 @@ static int iwl_mvm_config_ltr(struct iwl_mvm *mvm) sizeof(cmd), &cmd); } +#define ACPI_WRDS_METHOD "WRDS" +#define ACPI_WRDS_WIFI (0x07) +#define ACPI_WRDS_TABLE_SIZE 10 + +struct iwl_mvm_sar_table { + bool enabled; + u8 values[ACPI_WRDS_TABLE_SIZE]; +}; + +#ifdef CONFIG_ACPI +static int iwl_mvm_sar_get_wrds(struct iwl_mvm *mvm, union acpi_object *wrds, + struct iwl_mvm_sar_table *sar_table) +{ + union acpi_object *data_pkg; + u32 i; + + /* We need at least two packages, one for the revision and one + * for the data itself. Also check that the revision is valid + * (i.e. it is an integer set to 0). + */ + if (wrds->type != ACPI_TYPE_PACKAGE || + wrds->package.count < 2 || + wrds->package.elements[0].type != ACPI_TYPE_INTEGER || + wrds->package.elements[0].integer.value != 0) { + IWL_DEBUG_RADIO(mvm, "Unsupported wrds structure\n"); + return -EINVAL; + } + + /* loop through all the packages to find the one for WiFi */ + for (i = 1; i < wrds->package.count; i++) { + union acpi_object *domain; + + data_pkg = &wrds->package.elements[i]; + + /* Skip anything that is not a package with the right + * amount of elements (i.e. domain_type, + * enabled/disabled plus the sar table size. + */ + if (data_pkg->type != ACPI_TYPE_PACKAGE || + data_pkg->package.count != ACPI_WRDS_TABLE_SIZE + 2) + continue; + + domain = &data_pkg->package.elements[0]; + if (domain->type == ACPI_TYPE_INTEGER && + domain->integer.value == ACPI_WRDS_WIFI) + break; + + data_pkg = NULL; + } + + if (!data_pkg) + return -ENOENT; + + if (data_pkg->package.elements[1].type != ACPI_TYPE_INTEGER) + return -EINVAL; + + sar_table->enabled = !!(data_pkg->package.elements[1].integer.value); + + for (i = 0; i < ACPI_WRDS_TABLE_SIZE; i++) { + union acpi_object *entry; + + entry = &data_pkg->package.elements[i + 2]; + if ((entry->type != ACPI_TYPE_INTEGER) || + (entry->integer.value > U8_MAX)) + return -EINVAL; + + sar_table->values[i] = entry->integer.value; + } + + return 0; +} + +static int iwl_mvm_sar_get_table(struct iwl_mvm *mvm, + struct iwl_mvm_sar_table *sar_table) +{ + acpi_handle root_handle; + acpi_handle handle; + struct acpi_buffer wrds = {ACPI_ALLOCATE_BUFFER, NULL}; + acpi_status status; + int ret; + + root_handle = ACPI_HANDLE(mvm->dev); + if (!root_handle) { + IWL_DEBUG_RADIO(mvm, + "Could not retrieve root port ACPI handle\n"); + return -ENOENT; + } + + /* Get the method's handle */ + status = acpi_get_handle(root_handle, (acpi_string)ACPI_WRDS_METHOD, + &handle); + if (ACPI_FAILURE(status)) { + IWL_DEBUG_RADIO(mvm, "WRDS method not found\n"); + return -ENOENT; + } + + /* Call WRDS with no arguments */ + status = acpi_evaluate_object(handle, NULL, NULL, &wrds); + if (ACPI_FAILURE(status)) { + IWL_DEBUG_RADIO(mvm, "WRDS invocation failed (0x%x)\n", status); + return -ENOENT; + } + + ret = iwl_mvm_sar_get_wrds(mvm, wrds.pointer, sar_table); + kfree(wrds.pointer); + + return ret; +} +#else /* CONFIG_ACPI */ +static int iwl_mvm_sar_get_table(struct iwl_mvm *mvm, + struct iwl_mvm_sar_table *sar_table) +{ + return -ENOENT; +} +#endif /* CONFIG_ACPI */ + +static int iwl_mvm_sar_init(struct iwl_mvm *mvm) +{ + struct iwl_mvm_sar_table sar_table; + struct iwl_dev_tx_power_cmd cmd = { + .v2.set_mode = cpu_to_le32(IWL_TX_POWER_MODE_SET_CHAINS), + }; + int ret, i, j, idx; + + /* we can't do anything with the table if the FW doesn't support it */ + if (!fw_has_api(&mvm->fw->ucode_capa, + IWL_UCODE_TLV_API_TX_POWER_CHAIN)) { + IWL_DEBUG_RADIO(mvm, + "FW doesn't support per-chain TX power settings.\n"); + return 0; + } + + ret = iwl_mvm_sar_get_table(mvm, &sar_table); + if (ret < 0) { + IWL_DEBUG_RADIO(mvm, + "SAR BIOS table invalid or unavailable. (%d)\n", + ret); + /* we don't fail if the table is not available */ + return 0; + } + + if (!sar_table.enabled) + return 0; + + IWL_DEBUG_RADIO(mvm, "Sending REDUCE_TX_POWER_CMD per chain\n"); + + BUILD_BUG_ON(IWL_NUM_CHAIN_LIMITS * IWL_NUM_SUB_BANDS != + ACPI_WRDS_TABLE_SIZE); + + for (i = 0; i < IWL_NUM_CHAIN_LIMITS; i++) { + IWL_DEBUG_RADIO(mvm, " Chain[%d]:\n", i); + for (j = 0; j < IWL_NUM_SUB_BANDS; j++) { + idx = (i * IWL_NUM_SUB_BANDS) + j; + cmd.per_chain_restriction[i][j] = + cpu_to_le16(sar_table.values[idx]); + IWL_DEBUG_RADIO(mvm, " Band[%d] = %d * .125dBm\n", + j, sar_table.values[idx]); + } + } + + ret = iwl_mvm_send_cmd_pdu(mvm, REDUCE_TX_POWER_CMD, 0, + sizeof(cmd), &cmd); + if (ret) + IWL_ERR(mvm, "failed to set per-chain TX power: %d\n", ret); + + return ret; +} + int iwl_mvm_up(struct iwl_mvm *mvm) { int ret, i; @@ -1077,6 +1246,10 @@ int iwl_mvm_up(struct iwl_mvm *mvm) if (!test_bit(IWL_MVM_STATUS_IN_HW_RESTART, &mvm->status)) iwl_mvm_unref(mvm, IWL_MVM_REF_UCODE_DOWN); + ret = iwl_mvm_sar_init(mvm); + if (ret) + goto error; + IWL_DEBUG_INFO(mvm, "RT uCode started.\n"); return 0; error: -- cgit v0.10.2 From 54f315cb082bf98290575e858d1e0274bb0eebe0 Mon Sep 17 00:00:00 2001 From: Ido Yariv Date: Tue, 14 Jun 2016 10:27:57 -0400 Subject: iwlwifi: pcie: Enable MSI mode when using MSI interrupts On some of the chipsets MSI & INTA interrupts are disabled by default in the HW registers, and need to be explicitly enabled to be used. In case MSI-X isn't used, make sure MSI mode is enabled by setting the relevant HW register. Signed-off-by: Ido Yariv Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h index 6c1d20d..459bf73 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-prph.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-prph.h @@ -417,5 +417,6 @@ enum { }; #define UREG_CHICK (0xA05C00) +#define UREG_CHICK_MSI_ENABLE BIT(24) #define UREG_CHICK_MSIX_ENABLE BIT(25) #endif /* __iwl_prph_h__ */ diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index ac623c3..f5ace92 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -1396,8 +1396,12 @@ static void iwl_pcie_init_msix(struct iwl_trans_pcie *trans_pcie) max_rx_vector = trans_pcie->allocated_vector - 1; - if (!trans_pcie->msix_enabled) + if (!trans_pcie->msix_enabled) { + if (trans->cfg->mq_rx_supported) + iwl_write_prph(trans, UREG_CHICK, + UREG_CHICK_MSI_ENABLE); return; + } iwl_write_prph(trans, UREG_CHICK, UREG_CHICK_MSIX_ENABLE); -- cgit v0.10.2 From d5d0689aefc59c6a5352ca25d7e6d47d03f543ce Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Thu, 9 Jun 2016 17:19:35 +0300 Subject: iwlwifi: pcie: fix access to scratch buffer This fixes a pretty ancient bug that hasn't manifested itself until now. The scratchbuf for command queue is allocated only for 32 slots but is accessed with the queue write pointer - which can be up to 256. Since the scratch buf size was 16 and there are up to 256 TFDs we never passed a page boundary when accessing the scratch buffer, but when attempting to increase the size of the scratch buffer a panic was quick to follow when trying to access the address resulted in a page boundary. Signed-off-by: Sara Sharon Fixes: 38c0f334b359 ("iwlwifi: use coherent DMA memory for command header") Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c index 8901d49..9b5858b 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c @@ -1604,9 +1604,9 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans, /* start the TFD with the scratchbuf */ scratch_size = min_t(int, copy_size, IWL_HCMD_SCRATCHBUF_SIZE); - memcpy(&txq->scratchbufs[q->write_ptr], &out_cmd->hdr, scratch_size); + memcpy(&txq->scratchbufs[idx], &out_cmd->hdr, scratch_size); iwl_pcie_txq_build_tfd(trans, txq, - iwl_pcie_get_scratchbuf_dma(txq, q->write_ptr), + iwl_pcie_get_scratchbuf_dma(txq, idx), scratch_size, true); /* map first command fragment, if any remains */ -- cgit v0.10.2 From e7c9bd1cc632e924a69bf704385484386bb10933 Mon Sep 17 00:00:00 2001 From: Golan Ben-Ami Date: Wed, 15 Jun 2016 09:16:24 +0300 Subject: iwlwifi: mvm: write the correct internal TXF index The TX fifos are arranged consecutively in the SMEM, beginning with the regular fifos, and tailed by the internal fifos. In the current code, while trying to read the internal fifos, we read the fifos beginning with the index zero. By doing this we actually re-read the regular fifos. In order to read the internal fifos, start the reading index from the number of regular fifos configured by the fw. Signed-off-by: Golan Ben-Ami Fixes: 39654cb3a6a2 ("iwlwifi: don't access a nonexistent register upon assert") Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c index 1ece19d..6e692a8 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-dbg.c @@ -288,7 +288,8 @@ static void iwl_mvm_dump_fifos(struct iwl_mvm *mvm, fifo_hdr->fifo_num = cpu_to_le32(i); /* Mark the number of TXF we're pulling now */ - iwl_trans_write_prph(mvm->trans, TXF_CPU2_NUM, i); + iwl_trans_write_prph(mvm->trans, TXF_CPU2_NUM, i + + ARRAY_SIZE(mvm->shared_mem_cfg.txfifo_size)); fifo_hdr->available_bytes = cpu_to_le32(iwl_trans_read_prph(mvm->trans, -- cgit v0.10.2 From c934bce9de4b715b74e376a904e4426def586314 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Mon, 20 Jun 2016 09:40:40 +0300 Subject: iwlwifi: mvm: fix coex related comments Those comments were wrong, fix them. Signed-off-by: Emmanuel Grumbach Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c index a63f5bb..07fc981 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c @@ -615,8 +615,8 @@ static void iwl_mvm_bt_notif_iterator(void *_data, u8 *mac, * don't reduce the Tx power if one of these is true: * we are in LOOSE * single share antenna product - * BT is active - * we are associated + * BT is inactive + * we are not associated */ if (iwl_get_coex_type(mvm, vif) == BT_COEX_LOOSE_LUT || mvm->cfg->bt_shared_single_ant || !vif->bss_conf.assoc || -- cgit v0.10.2 From ad17b1d9536594c6f4b52a99e9efbf1050bf99d9 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Wed, 22 Jun 2016 10:51:31 +0300 Subject: iwlwifi: mvm: fix the channel inhibition table for Channel 14 The value for Channel 14 was wrong. Fix it. Signed-off-by: Emmanuel Grumbach Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c index 07fc981..5bdb6c2 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/coex.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/coex.c @@ -142,7 +142,7 @@ static const __le64 iwl_ci_mask[][3] = { cpu_to_le64(0x0) }, { - cpu_to_le64(0xFFC0000000ULL), + cpu_to_le64(0xFE00000000ULL), cpu_to_le64(0x0ULL), cpu_to_le64(0x0ULL) }, -- cgit v0.10.2 From 5e7d7eb9cc553bef0d1386bb736f8d0a6152dc21 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 21 Jun 2016 12:23:35 +0200 Subject: iwlwifi: remove iwl_ht_params.smps_mode This struct member is never set, so remove it. Since this is the last thing that needs mac80211.h, also change the includes to no longer use mac80211.h Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/rxon.c b/drivers/net/wireless/intel/iwlwifi/dvm/rxon.c index b228552..087e579 100644 --- a/drivers/net/wireless/intel/iwlwifi/dvm/rxon.c +++ b/drivers/net/wireless/intel/iwlwifi/dvm/rxon.c @@ -523,11 +523,6 @@ static int iwlagn_rxon_connect(struct iwl_priv *priv, return ret; } - if (ctx->vif && ctx->vif->type == NL80211_IFTYPE_STATION && - priv->cfg->ht_params && priv->cfg->ht_params->smps_mode) - ieee80211_request_smps(ctx->vif, - priv->cfg->ht_params->smps_mode); - return 0; } diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h index 122e895..64108cc 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h @@ -66,8 +66,9 @@ #define __IWL_CONFIG_H__ #include -#include - +#include +#include +#include enum iwl_device_family { IWL_DEVICE_FAMILY_UNDEFINED, @@ -192,7 +193,6 @@ struct iwl_base_params { * @ht40_bands: bitmap of bands (using %NL80211_BAND_*) that support HT40 */ struct iwl_ht_params { - enum ieee80211_smps_mode smps_mode; u8 ht_greenfield_support:1, stbc:1, ldpc:1, -- cgit v0.10.2 From 3edbc7dabab8ce85aa75c5e290ecda7a3692ebc9 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Sun, 19 Jun 2016 20:57:02 +0300 Subject: iwlwifi: mvm: unmap the paging memory before freeing it This led to a DMA splat. Fixes: a6c4fb4441f4 ("iwlwifi: mvm: Add FW paging mechanism for the UMAC on PCI") Signed-off-by: Emmanuel Grumbach Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index 4db1d84..4c16fa7 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -160,17 +160,21 @@ void iwl_free_fw_paging(struct iwl_mvm *mvm) return; for (i = 0; i < NUM_OF_FW_PAGING_BLOCKS; i++) { - if (!mvm->fw_paging_db[i].fw_paging_block) { + struct iwl_fw_paging *paging = &mvm->fw_paging_db[i]; + + if (!paging->fw_paging_block) { IWL_DEBUG_FW(mvm, "Paging: block %d already freed, continue to next page\n", i); continue; } + dma_unmap_page(mvm->trans->dev, paging->fw_paging_phys, + paging->fw_paging_size, DMA_BIDIRECTIONAL); - __free_pages(mvm->fw_paging_db[i].fw_paging_block, - get_order(mvm->fw_paging_db[i].fw_paging_size)); - mvm->fw_paging_db[i].fw_paging_block = NULL; + __free_pages(paging->fw_paging_block, + get_order(paging->fw_paging_size)); + paging->fw_paging_block = NULL; } kfree(mvm->trans->paging_download_buf); mvm->trans->paging_download_buf = NULL; -- cgit v0.10.2 From e25d65f2670adfcbb019f7fee4302a60333f913a Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Tue, 21 Jun 2016 11:13:47 +0300 Subject: iwlwifi: pcie: don't use vid 0 In cases of hardware or DMA error, the vid read from a zeroed location will be 0, and we will access the rxb at index 0 in the global table, while it may be NULL or owned by hardware. Invalidate vid 0 in order to detect the situation and bail out. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index c1c3c6a..0296c29 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -960,7 +960,7 @@ int iwl_pcie_rx_init(struct iwl_trans *trans) else list_add(&rxb->list, &def_rxq->rx_used); trans_pcie->global_table[i] = rxb; - rxb->vid = (u16)i; + rxb->vid = (u16)(i + 1); } iwl_pcie_rxq_alloc_rbs(trans, GFP_KERNEL, def_rxq); @@ -1249,10 +1249,13 @@ restart: */ u16 vid = le32_to_cpu(rxq->used_bd[i]) & 0x0FFF; - if (WARN(vid >= ARRAY_SIZE(trans_pcie->global_table), - "Invalid rxb index from HW %u\n", (u32)vid)) + if (WARN(!vid || + vid > ARRAY_SIZE(trans_pcie->global_table), + "Invalid rxb index from HW %u\n", (u32)vid)) { + iwl_force_nmi(trans); goto out; - rxb = trans_pcie->global_table[vid]; + } + rxb = trans_pcie->global_table[vid - 1]; } else { rxb = rxq->queue[i]; rxq->queue[i] = NULL; -- cgit v0.10.2 From e3118ad74d7ef5bbf1a4a17471466e2cafacc752 Mon Sep 17 00:00:00 2001 From: Liad Kaufman Date: Sun, 5 Jun 2016 10:49:02 +0300 Subject: iwlwifi: mvm: support tdls in dqa mode Support TDLS when working in DQA mode. This is done mainly by NOT doing any special things for TDLS, as the queues are dynamically created anyway, so no need to allocate them ahead of time. Signed-off-by: Liad Kaufman Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index 7321bb6..1617de7 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -893,8 +893,11 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm, mvm_sta->tid_disable_agg = 0xffff; /* No aggs at first */ mvm_sta->tfd_queue_msk = 0; - /* allocate new queues for a TDLS station */ - if (sta->tdls) { + /* + * Allocate new queues for a TDLS station, unless we're in DQA mode, + * and then they'll be allocated dynamically + */ + if (!iwl_mvm_is_dqa_supported(mvm) && sta->tdls) { ret = iwl_mvm_tdls_sta_init(mvm, sta); if (ret) return ret; @@ -958,7 +961,8 @@ int iwl_mvm_add_sta(struct iwl_mvm *mvm, return 0; err: - iwl_mvm_tdls_sta_deinit(mvm, sta); + if (!iwl_mvm_is_dqa_supported(mvm) && sta->tdls) + iwl_mvm_tdls_sta_deinit(mvm, sta); return ret; } @@ -1164,16 +1168,19 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm, if (iwl_mvm_is_dqa_supported(mvm)) iwl_mvm_disable_sta_queues(mvm, vif, mvm_sta); - /* if we are associated - we can't remove the AP STA now */ - if (vif->bss_conf.assoc) - return ret; + if (vif->type == NL80211_IFTYPE_STATION && + mvmvif->ap_sta_id == mvm_sta->sta_id) { + /* if associated - we can't remove the AP STA now */ + if (vif->bss_conf.assoc) + return ret; - /* unassoc - go ahead - remove the AP STA now */ - mvmvif->ap_sta_id = IWL_MVM_STATION_COUNT; + /* unassoc - go ahead - remove the AP STA now */ + mvmvif->ap_sta_id = IWL_MVM_STATION_COUNT; - /* clear d0i3_ap_sta_id if no longer relevant */ - if (mvm->d0i3_ap_sta_id == mvm_sta->sta_id) - mvm->d0i3_ap_sta_id = IWL_MVM_STATION_COUNT; + /* clear d0i3_ap_sta_id if no longer relevant */ + if (mvm->d0i3_ap_sta_id == mvm_sta->sta_id) + mvm->d0i3_ap_sta_id = IWL_MVM_STATION_COUNT; + } } /* @@ -1211,7 +1218,7 @@ int iwl_mvm_rm_sta(struct iwl_mvm *mvm, } else { spin_unlock_bh(&mvm_sta->lock); - if (sta->tdls) + if (!iwl_mvm_is_dqa_supported(mvm) && sta->tdls) iwl_mvm_tdls_sta_deinit(mvm, sta); ret = iwl_mvm_rm_sta_common(mvm, mvm_sta->sta_id); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c index f3a6e95..c6585ab 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c @@ -546,6 +546,9 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb) * (this is not possible for unicast packets as a TLDS discovery * response are sent without a station entry); otherwise use the * AUX station. + * In DQA mode, if vif is of type STATION and frames are not multicast, + * they should be sent from the BSS queue. For example, TDLS setup + * frames should be sent on this queue, as they go through the AP. */ sta_id = mvm->aux_sta.sta_id; if (info.control.vif) { @@ -563,6 +566,9 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb) if (ap_sta_id != IWL_MVM_STATION_COUNT) sta_id = ap_sta_id; + } else if (iwl_mvm_is_dqa_supported(mvm) && + info.control.vif->type == NL80211_IFTYPE_STATION) { + queue = IWL_MVM_DQA_BSS_CLIENT_QUEUE; } } @@ -906,7 +912,7 @@ static int iwl_mvm_tx_mpdu(struct iwl_mvm *mvm, struct sk_buff *skb, WARN_ON_ONCE(info->flags & IEEE80211_TX_CTL_SEND_AFTER_DTIM); - if (sta->tdls) { + if (sta->tdls && !iwl_mvm_is_dqa_supported(mvm)) { /* default to TID 0 for non-QoS packets */ u8 tdls_tid = tid == IWL_MAX_TID_COUNT ? 0 : tid; -- cgit v0.10.2 From 58f2cc57dc6ac89946757b812fc5b93157978c37 Mon Sep 17 00:00:00 2001 From: Liad Kaufman Date: Wed, 30 Sep 2015 16:44:28 +0200 Subject: iwlwifi: mvm: support dqa-mode scd queue redirection Make sure that in DQA mode, the SCD's configuration of a queue is redirected to the lower AC of the streams of the queue. Make sure that this queue is redirected to the lowest AC when adding a new RA/TID to an existing queue. If it isn't - redirect the queue. Also, as redirection revealed a bug in the marking of a shared queue, this patch contains a small fix to make sure a shared queue maintains the appropriate "shared queue marking". Signed-off-by: Liad Kaufman Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index 1617de7..a6d6df9 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -513,6 +513,101 @@ static int iwl_mvm_get_shared_queue(struct iwl_mvm *mvm, return queue; } +/* + * If a given queue has a higher AC than the TID stream that is being added to + * it, the queue needs to be redirected to the lower AC. This function does that + * in such a case, otherwise - if no redirection required - it does nothing, + * unless the %force param is true. + */ +static int iwl_mvm_scd_queue_redirect(struct iwl_mvm *mvm, int queue, int tid, + int ac, int ssn, unsigned int wdg_timeout, + bool force) +{ + struct iwl_scd_txq_cfg_cmd cmd = { + .scd_queue = queue, + .enable = 0, + }; + bool shared_queue; + unsigned long mq; + int ret; + + /* + * If the AC is lower than current one - FIFO needs to be redirected to + * the lowest one of the streams in the queue. Check if this is needed + * here. + * Notice that the enum ieee80211_ac_numbers is "flipped", so BK is with + * value 3 and VO with value 0, so to check if ac X is lower than ac Y + * we need to check if the numerical value of X is LARGER than of Y. + */ + spin_lock_bh(&mvm->queue_info_lock); + if (ac <= mvm->queue_info[queue].mac80211_ac && !force) { + spin_unlock_bh(&mvm->queue_info_lock); + + IWL_DEBUG_TX_QUEUES(mvm, + "No redirection needed on TXQ #%d\n", + queue); + return 0; + } + + cmd.sta_id = mvm->queue_info[queue].ra_sta_id; + cmd.tx_fifo = iwl_mvm_ac_to_tx_fifo[mvm->queue_info[queue].mac80211_ac]; + mq = mvm->queue_info[queue].hw_queue_to_mac80211; + shared_queue = (mvm->queue_info[queue].hw_queue_refcount > 1); + spin_unlock_bh(&mvm->queue_info_lock); + + IWL_DEBUG_TX_QUEUES(mvm, "Redirecting shared TXQ #%d to FIFO #%d\n", + queue, iwl_mvm_ac_to_tx_fifo[ac]); + + /* Stop MAC queues and wait for this queue to empty */ + iwl_mvm_stop_mac_queues(mvm, mq); + ret = iwl_trans_wait_tx_queue_empty(mvm->trans, BIT(queue)); + if (ret) { + IWL_ERR(mvm, "Error draining queue %d before reconfig\n", + queue); + ret = -EIO; + goto out; + } + + /* Before redirecting the queue we need to de-activate it */ + iwl_trans_txq_disable(mvm->trans, queue, false); + ret = iwl_mvm_send_cmd_pdu(mvm, SCD_QUEUE_CFG, 0, sizeof(cmd), &cmd); + if (ret) + IWL_ERR(mvm, "Failed SCD disable TXQ %d (ret=%d)\n", queue, + ret); + + /* Make sure the SCD wrptr is correctly set before reconfiguring */ + iwl_trans_txq_enable(mvm->trans, queue, iwl_mvm_ac_to_tx_fifo[ac], + cmd.sta_id, tid, LINK_QUAL_AGG_FRAME_LIMIT_DEF, + ssn, wdg_timeout); + + /* TODO: Work-around SCD bug when moving back by multiples of 0x40 */ + + /* Redirect to lower AC */ + iwl_mvm_reconfig_scd(mvm, queue, iwl_mvm_ac_to_tx_fifo[ac], + cmd.sta_id, tid, LINK_QUAL_AGG_FRAME_LIMIT_DEF, + ssn); + + /* Update AC marking of the queue */ + spin_lock_bh(&mvm->queue_info_lock); + mvm->queue_info[queue].mac80211_ac = ac; + spin_unlock_bh(&mvm->queue_info_lock); + + /* + * Mark queue as shared in transport if shared + * Note this has to be done after queue enablement because enablement + * can also set this value, and there is no indication there to shared + * queues + */ + if (shared_queue) + iwl_trans_txq_set_shared_mode(mvm->trans, queue, true); + +out: + /* Continue using the MAC queues */ + iwl_mvm_start_mac_queues(mvm, mq); + + return ret; +} + static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, struct ieee80211_sta *sta, u8 ac, int tid, struct ieee80211_hdr *hdr) @@ -680,17 +775,21 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, iwl_mvm_invalidate_sta_queue(mvm, queue, disable_agg_tids, false); } - - /* Mark queue as shared in transport */ - iwl_trans_txq_set_shared_mode(mvm->trans, queue, true); - - /* TODO: a redirection may be required - DQA phase 2 */ } ssn = IEEE80211_SEQ_TO_SN(le16_to_cpu(hdr->seq_ctrl)); iwl_mvm_enable_txq(mvm, queue, mac_queue, ssn, &cfg, wdg_timeout); + /* + * Mark queue as shared in transport if shared + * Note this has to be done after queue enablement because enablement + * can also set this value, and there is no indication there to shared + * queues + */ + if (shared_queue) + iwl_trans_txq_set_shared_mode(mvm->trans, queue, true); + spin_lock_bh(&mvmsta->lock); mvmsta->tid_data[tid].txq_id = queue; mvmsta->tid_data[tid].is_tid_active = true; @@ -712,6 +811,12 @@ static int iwl_mvm_sta_alloc_queue(struct iwl_mvm *mvm, if (ret) goto out_err; } + } else { + /* Redirect queue, if needed */ + ret = iwl_mvm_scd_queue_redirect(mvm, queue, tid, ac, ssn, + wdg_timeout, false); + if (ret) + goto out_err; } return 0; -- cgit v0.10.2 From 35263a03115726dcd393bc96504adbdd6ef08216 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Tue, 21 Jun 2016 12:12:10 +0300 Subject: iwlwifi: mvm: add RX aggregation prints Add some prints to track BAID assignment. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c index d13397a..a1cbf48 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c @@ -476,6 +476,9 @@ void iwl_mvm_reorder_timer_expired(unsigned long data) rcu_read_lock(); sta = rcu_dereference(buf->mvm->fw_id_to_mac_id[buf->sta_id]); /* SN is set to the last expired frame + 1 */ + IWL_DEBUG_HT(buf->mvm, + "Releasing expired frames for sta %u, sn %d\n", + buf->sta_id, sn); iwl_mvm_release_frames(buf->mvm, sta, NULL, buf, sn); rcu_read_unlock(); } else if (buf->num_stored) { @@ -956,6 +959,9 @@ void iwl_mvm_rx_frame_release(struct iwl_mvm *mvm, struct napi_struct *napi, int baid = release->baid; + IWL_DEBUG_HT(mvm, "Frame release notification for BAID %u, NSSN %d\n", + release->baid, le16_to_cpu(release->nssn)); + if (WARN_ON_ONCE(baid == IWL_RX_REORDER_DATA_INVALID_BAID)) return; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index a6d6df9..c0c2ea3 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -1768,8 +1768,8 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta, switch (status & IWL_ADD_STA_STATUS_MASK) { case ADD_STA_SUCCESS: - IWL_DEBUG_INFO(mvm, "RX BA Session %sed in fw\n", - start ? "start" : "stopp"); + IWL_DEBUG_HT(mvm, "RX BA Session %sed in fw\n", + start ? "start" : "stopp"); break; case ADD_STA_IMMEDIATE_BA_FAILURE: IWL_WARN(mvm, "RX BA Session refused by fw\n"); @@ -1824,6 +1824,8 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta, * supposed to happen) and we will free the session data while * RX is being processed in parallel */ + IWL_DEBUG_HT(mvm, "Sta %d(%d) is assigned to BAID %d\n", + mvm_sta->sta_id, tid, baid); WARN_ON(rcu_access_pointer(mvm->baid_map[baid])); rcu_assign_pointer(mvm->baid_map[baid], baid_data); } else if (mvm->rx_ba_sessions > 0) { @@ -1846,6 +1848,7 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta, del_timer_sync(&baid_data->session_timer); RCU_INIT_POINTER(mvm->baid_map[baid], NULL); kfree_rcu(baid_data, rcu_head); + IWL_DEBUG_HT(mvm, "BAID %d is free\n", baid); } return 0; -- cgit v0.10.2 From 60dec5233cd8651860e8010c953d116fb0f1ba86 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Tue, 21 Jun 2016 14:14:08 +0300 Subject: iwlwifi: mvm: free RX reorder buffer on restart Restart flow zeroes the rx_ba_sessions counter. Mac80211 asks driver to tear down of the session only afterwards, and as a result driver didn't free the data. Fix it. Signed-off-by: Sara Sharon Fixes: 10b2b2019d81 ("iwlwifi: mvm: add infrastructure for tracking BA session in driver") Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c index c0c2ea3..61f16c3 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/sta.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/sta.c @@ -1828,11 +1828,12 @@ int iwl_mvm_sta_rx_agg(struct iwl_mvm *mvm, struct ieee80211_sta *sta, mvm_sta->sta_id, tid, baid); WARN_ON(rcu_access_pointer(mvm->baid_map[baid])); rcu_assign_pointer(mvm->baid_map[baid], baid_data); - } else if (mvm->rx_ba_sessions > 0) { + } else { u8 baid = mvm_sta->tid_to_baid[tid]; - /* check that restart flow didn't zero the counter */ - mvm->rx_ba_sessions--; + if (mvm->rx_ba_sessions > 0) + /* check that restart flow didn't zero the counter */ + mvm->rx_ba_sessions--; if (!iwl_mvm_has_new_rx_api(mvm)) return 0; -- cgit v0.10.2 From 24ddddf3673b43708c9e7edc576b4d14a81a43e9 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 21 Jun 2016 12:34:36 +0200 Subject: iwlwifi: store cipher scheme independent of mac80211 In order to reduce reliance on mac80211 structs in the core iwlwifi code, store the cipher schemes in the format given by the firmware and convert it later, rather than storing it in the mac80211 format. Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c index e2df544..d675bf0 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c @@ -326,8 +326,6 @@ static int iwl_store_cscheme(struct iwl_fw *fw, const u8 *data, const u32 len) int i, j; struct iwl_fw_cscheme_list *l = (struct iwl_fw_cscheme_list *)data; struct iwl_fw_cipher_scheme *fwcs; - struct ieee80211_cipher_scheme *cs; - u32 cipher; if (len < sizeof(*l) || len < sizeof(l->size) + l->size * sizeof(l->cs[0])) @@ -335,22 +333,12 @@ static int iwl_store_cscheme(struct iwl_fw *fw, const u8 *data, const u32 len) for (i = 0, j = 0; i < IWL_UCODE_MAX_CS && i < l->size; i++) { fwcs = &l->cs[j]; - cipher = le32_to_cpu(fwcs->cipher); /* we skip schemes with zero cipher suite selector */ - if (!cipher) + if (!fwcs->cipher) continue; - cs = &fw->cs[j++]; - cs->cipher = cipher; - cs->iftype = BIT(NL80211_IFTYPE_STATION); - cs->hdr_len = fwcs->hdr_len; - cs->pn_len = fwcs->pn_len; - cs->pn_off = fwcs->pn_off; - cs->key_idx_off = fwcs->key_idx_off; - cs->key_idx_mask = fwcs->key_idx_mask; - cs->key_idx_shift = fwcs->key_idx_shift; - cs->mic_len = fwcs->mic_len; + fw->cs[j++] = *fwcs; } return 0; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw.h index 655ec52..74ea68d 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-fw.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw.h @@ -67,7 +67,6 @@ #ifndef __iwl_fw_h__ #define __iwl_fw_h__ #include -#include #include "iwl-fw-file.h" #include "iwl-fw-error-dump.h" @@ -287,7 +286,7 @@ struct iwl_fw { enum iwl_fw_type type; - struct ieee80211_cipher_scheme cs[IWL_UCODE_MAX_CS]; + struct iwl_fw_cipher_scheme cs[IWL_UCODE_MAX_CS]; u8 human_readable[FW_VER_HUMAN_READABLE_SZ]; u32 sdio_adma_addr; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index e44c337..73e49d3 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -494,10 +494,23 @@ int iwl_mvm_mac_setup_register(struct iwl_mvm *mvm) /* currently FW API supports only one optional cipher scheme */ if (mvm->fw->cs[0].cipher) { + const struct iwl_fw_cipher_scheme *fwcs = &mvm->fw->cs[0]; + struct ieee80211_cipher_scheme *cs = &mvm->cs[0]; + mvm->hw->n_cipher_schemes = 1; - mvm->hw->cipher_schemes = &mvm->fw->cs[0]; - mvm->ciphers[hw->wiphy->n_cipher_suites] = - mvm->fw->cs[0].cipher; + + cs->cipher = le32_to_cpu(fwcs->cipher); + cs->iftype = BIT(NL80211_IFTYPE_STATION); + cs->hdr_len = fwcs->hdr_len; + cs->pn_len = fwcs->pn_len; + cs->pn_off = fwcs->pn_off; + cs->key_idx_off = fwcs->key_idx_off; + cs->key_idx_mask = fwcs->key_idx_mask; + cs->key_idx_shift = fwcs->key_idx_shift; + cs->mic_len = fwcs->mic_len; + + mvm->hw->cipher_schemes = mvm->cs; + mvm->ciphers[hw->wiphy->n_cipher_suites] = cs->cipher; hw->wiphy->n_cipher_suites++; } diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h index c440b10..b4fc86d 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h @@ -1012,6 +1012,7 @@ struct iwl_mvm { struct iwl_mvm_shared_mem_cfg shared_mem_cfg; u32 ciphers[IWL_MVM_NUM_CIPHERS]; + struct ieee80211_cipher_scheme cs[IWL_UCODE_MAX_CS]; struct iwl_mvm_tof_data tof_data; struct ieee80211_vif *nan_vif; -- cgit v0.10.2 From 0c4cb7314d15d63fbc96e60c57443f3ef83b9c5e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 21 Jun 2016 12:44:21 +0200 Subject: iwlwifi: tracing: decouple from mac80211 In order to be able to properly record SKBs that didn't come through mac80211, don't rely on the IEEE80211_TX_CTRL_PORT_CTRL_PROTO flag but instead check for ETH_P_PAE directly. Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h index f4d3cd0..545d14b 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace.h @@ -1,6 +1,7 @@ /****************************************************************************** * * Copyright(c) 2009 - 2014 Intel Corporation. All rights reserved. + * Copyright(C) 2016 Intel Deutschland GmbH * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License as @@ -33,11 +34,29 @@ static inline bool iwl_trace_data(struct sk_buff *skb) { struct ieee80211_hdr *hdr = (void *)skb->data; - struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); + __le16 fc = hdr->frame_control; + int offs = 24; /* start with normal header length */ - if (!ieee80211_is_data(hdr->frame_control)) + if (!ieee80211_is_data(fc)) return false; - return !(info->control.flags & IEEE80211_TX_CTRL_PORT_CTRL_PROTO); + + /* Try to determine if the frame is EAPOL. This might have false + * positives (if there's no RFC 1042 header and we compare to some + * payload instead) but since we're only doing tracing that's not + * a problem. + */ + + if (ieee80211_has_a4(fc)) + offs += 6; + if (ieee80211_is_data_qos(fc)) + offs += 2; + /* don't account for crypto - these are unencrypted */ + + /* also account for the RFC 1042 header, of course */ + offs += 6; + + return skb->len > offs + 2 && + *(__be16 *)(skb->data + offs) == cpu_to_be16(ETH_P_PAE); } static inline size_t iwl_rx_trace_len(const struct iwl_trans *trans, -- cgit v0.10.2 From 21cb3222fe569bd09a42771ffc8da9fb4666bd8a Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 21 Jun 2016 13:11:48 +0200 Subject: iwlwifi: decouple PCIe transport from mac80211 The PCIe transport needs to store two pointers in each TX SKB, and currently assumes mac80211's ieee80211_tx_info is present in the CB to do that. In order to remove that assumption, have the opmodes pass in the offset to where the pointers can be stored in the CB and use the offset in the PCIe code. To make the disentanglement complete, remove mac80211.h includes from everywhere in the generic iwlwifi code. This required adding an include of cfg80211.h in one place. Signed-off-by: Johannes Berg Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/main.c b/drivers/net/wireless/intel/iwlwifi/dvm/main.c index 5915914..b498486 100644 --- a/drivers/net/wireless/intel/iwlwifi/dvm/main.c +++ b/drivers/net/wireless/intel/iwlwifi/dvm/main.c @@ -1337,6 +1337,8 @@ static struct iwl_op_mode *iwl_op_mode_dvm_start(struct iwl_trans *trans, trans_cfg.command_groups_size = ARRAY_SIZE(iwl_dvm_groups); trans_cfg.cmd_fifo = IWLAGN_CMD_FIFO_NUM; + trans_cfg.cb_data_offs = offsetof(struct ieee80211_tx_info, + driver_data[2]); WARN_ON(sizeof(priv->transport_queue_stop) * BITS_PER_BYTE < priv->cfg->base_params->num_of_queues); diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h b/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h index 1f4e502..e04a91d 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h @@ -66,6 +66,7 @@ #include #include +#include #include "iwl-trans.h" struct iwl_nvm_data { diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h index 0379899..4d32b10 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-modparams.h @@ -66,7 +66,6 @@ #include #include #include -#include extern struct iwl_mod_params iwlwifi_mod_params; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h index 57cc67f..9ac47e0 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h @@ -392,11 +392,6 @@ static inline void iwl_free_rxb(struct iwl_rx_cmd_buffer *r) #define MAX_NO_RECLAIM_CMDS 6 -/* - * The first entry in driver_data array in ieee80211_tx_info - * that can be used by the transport. - */ -#define IWL_TRANS_FIRST_DRIVER_DATA 2 #define IWL_MASK(lo, hi) ((1 << (hi)) | ((1 << (hi)) - (1 << (lo)))) /* @@ -500,6 +495,8 @@ struct iwl_hcmd_arr { * @command_groups_size: number of command groups, to avoid illegal access * @sdio_adma_addr: the default address to set for the ADMA in SDIO mode until * we get the ALIVE from the uCode + * @cb_data_offs: offset inside skb->cb to store transport data at, must have + * space for at least two pointers */ struct iwl_trans_config { struct iwl_op_mode *op_mode; @@ -519,6 +516,8 @@ struct iwl_trans_config { int command_groups_size; u32 sdio_adma_addr; + + u8 cb_data_offs; }; struct iwl_trans_dump_data { diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c index a08db00..55d9096 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c @@ -668,6 +668,9 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_cfg *cfg, trans_cfg.cmd_fifo = IWL_MVM_TX_FIFO_CMD; trans_cfg.scd_set_active = true; + trans_cfg.cb_data_offs = offsetof(struct ieee80211_tx_info, + driver_data[2]); + trans_cfg.sdio_adma_addr = fw->sdio_adma_addr; trans_cfg.sw_csum_tx = IWL_MVM_SW_TX_CSUM_OFFLOAD; diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h index 8bfa915..f684b9d 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h @@ -385,6 +385,8 @@ struct iwl_trans_pcie { wait_queue_head_t wait_command_queue; wait_queue_head_t d0i3_waitq; + u8 page_offs, dev_cmd_offs; + u8 cmd_queue; u8 cmd_fifo; unsigned int cmd_q_wdg_timeout; diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index f5ace92..3b7a414 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -1650,6 +1650,9 @@ static void iwl_trans_pcie_configure(struct iwl_trans *trans, trans_pcie->scd_set_active = trans_cfg->scd_set_active; trans_pcie->sw_csum_tx = trans_cfg->sw_csum_tx; + trans_pcie->page_offs = trans_cfg->cb_data_offs; + trans_pcie->dev_cmd_offs = trans_cfg->cb_data_offs + sizeof(void *); + trans->command_groups = trans_cfg->command_groups; trans->command_groups_size = trans_cfg->command_groups_size; diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c index 9b5858b..2b68826 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c @@ -584,16 +584,16 @@ static int iwl_pcie_txq_init(struct iwl_trans *trans, struct iwl_txq *txq, return 0; } -static void iwl_pcie_free_tso_page(struct sk_buff *skb) +static void iwl_pcie_free_tso_page(struct iwl_trans_pcie *trans_pcie, + struct sk_buff *skb) { - struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); + struct page **page_ptr; - if (info->driver_data[IWL_TRANS_FIRST_DRIVER_DATA]) { - struct page *page = - info->driver_data[IWL_TRANS_FIRST_DRIVER_DATA]; + page_ptr = (void *)((u8 *)skb->cb + trans_pcie->page_offs); - __free_page(page); - info->driver_data[IWL_TRANS_FIRST_DRIVER_DATA] = NULL; + if (*page_ptr) { + __free_page(*page_ptr); + *page_ptr = NULL; } } @@ -639,7 +639,7 @@ static void iwl_pcie_txq_unmap(struct iwl_trans *trans, int txq_id) if (WARN_ON_ONCE(!skb)) continue; - iwl_pcie_free_tso_page(skb); + iwl_pcie_free_tso_page(trans_pcie, skb); } iwl_pcie_txq_free_tfd(trans, txq); q->read_ptr = iwl_queue_inc_wrap(q->read_ptr); @@ -1084,7 +1084,7 @@ void iwl_trans_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn, if (WARN_ON_ONCE(!skb)) continue; - iwl_pcie_free_tso_page(skb); + iwl_pcie_free_tso_page(trans_pcie, skb); __skb_queue_tail(skbs, skb); @@ -1115,17 +1115,17 @@ void iwl_trans_pcie_reclaim(struct iwl_trans *trans, int txq_id, int ssn, while (!skb_queue_empty(&overflow_skbs)) { struct sk_buff *skb = __skb_dequeue(&overflow_skbs); - struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); - u8 dev_cmd_idx = IWL_TRANS_FIRST_DRIVER_DATA + 1; - struct iwl_device_cmd *dev_cmd = - info->driver_data[dev_cmd_idx]; + struct iwl_device_cmd *dev_cmd_ptr; + + dev_cmd_ptr = *(void **)((u8 *)skb->cb + + trans_pcie->dev_cmd_offs); /* * Note that we can very well be overflowing again. * In that case, iwl_queue_space will be small again * and we won't wake mac80211's queue. */ - iwl_trans_pcie_tx(trans, skb, dev_cmd, txq_id); + iwl_trans_pcie_tx(trans, skb, dev_cmd_ptr, txq_id); } spin_lock_bh(&txq->lock); @@ -2024,7 +2024,6 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb, struct iwl_cmd_meta *out_meta, struct iwl_device_cmd *dev_cmd, u16 tb1_len) { - struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); struct iwl_trans_pcie *trans_pcie = txq->trans_pcie; struct ieee80211_hdr *hdr = (void *)skb->data; unsigned int snap_ip_tcp_hdrlen, ip_hdrlen, total_len, hdr_room; @@ -2033,6 +2032,7 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb, u16 length, iv_len, amsdu_pad; u8 *start_hdr; struct iwl_tso_hdr_page *hdr_page; + struct page **page_ptr; int ret; struct tso_t tso; @@ -2063,7 +2063,8 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb, get_page(hdr_page->page); start_hdr = hdr_page->pos; - info->driver_data[IWL_TRANS_FIRST_DRIVER_DATA] = hdr_page->page; + page_ptr = (void *)((u8 *)skb->cb + trans_pcie->page_offs); + *page_ptr = hdr_page->page; memcpy(hdr_page->pos, skb->data + hdr_len, iv_len); hdr_page->pos += iv_len; @@ -2273,10 +2274,12 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb, /* don't put the packet on the ring, if there is no room */ if (unlikely(iwl_queue_space(q) < 3)) { - struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb); + struct iwl_device_cmd **dev_cmd_ptr; + + dev_cmd_ptr = (void *)((u8 *)skb->cb + + trans_pcie->dev_cmd_offs); - info->driver_data[IWL_TRANS_FIRST_DRIVER_DATA + 1] = - dev_cmd; + *dev_cmd_ptr = dev_cmd; __skb_queue_tail(&txq->overflow_q, skb); spin_unlock(&txq->lock); -- cgit v0.10.2 From f16c3ebfa64fdf0e2dc88e6baa72da95ab70ffd7 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Mon, 13 Jun 2016 08:28:26 +0300 Subject: iwlwifi: pcie: fix a race in firmware loading flow Upon firmware load interrupt (FH_TX), the ISR re-enables the firmware load interrupt only to avoid races with other flows as described in the commit below. When the firmware is completely loaded, the thread that is loading the firmware will enable all the interrupts to make sure that the driver gets the ALIVE interrupt. The problem with that is that the thread that is loading the firmware is actually racing against the ISR and we can get to the following situation: CPU0 CPU1 iwl_pcie_load_given_ucode ... iwl_pcie_load_firmware_chunk wait_for_interrupt ISR handles CSR_INT_BIT_FH_TX ISR wakes up the thread on CPU0 /* enable all the interrupts * to get the ALIVE interrupt */ iwl_enable_interrupts ISR re-enables CSR_INT_BIT_FH_TX only /* start the firmware */ iwl_write32(trans, CSR_RESET, 0); BUG! ALIVE interrupt will never arrive since it has been masked by CPU1. In order to fix that, change the ISR to first check if STATUS_INT_ENABLED is set. If so, re-enable all the interrupts. If STATUS_INT_ENABLED is clear, then we can check what specific interrupt happened and re-enable only that specific interrupt (RFKILL or FH_TX). All the credit for the analysis goes to Kirtika who did the actual debugging work. Cc: [4.5+] Fixes: a6bd005fe92 ("iwlwifi: pcie: fix RF-Kill vs. firmware load race") Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h index f684b9d..54af3da 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h @@ -500,7 +500,7 @@ void iwl_pcie_dump_csr(struct iwl_trans *trans); /***************************************************** * Helpers ******************************************************/ -static inline void iwl_disable_interrupts(struct iwl_trans *trans) +static inline void _iwl_disable_interrupts(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); @@ -523,7 +523,16 @@ static inline void iwl_disable_interrupts(struct iwl_trans *trans) IWL_DEBUG_ISR(trans, "Disabled interrupts\n"); } -static inline void iwl_enable_interrupts(struct iwl_trans *trans) +static inline void iwl_disable_interrupts(struct iwl_trans *trans) +{ + struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); + + spin_lock(&trans_pcie->irq_lock); + _iwl_disable_interrupts(trans); + spin_unlock(&trans_pcie->irq_lock); +} + +static inline void _iwl_enable_interrupts(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); @@ -546,6 +555,14 @@ static inline void iwl_enable_interrupts(struct iwl_trans *trans) } } +static inline void iwl_enable_interrupts(struct iwl_trans *trans) +{ + struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); + + spin_lock(&trans_pcie->irq_lock); + _iwl_enable_interrupts(trans); + spin_unlock(&trans_pcie->irq_lock); +} static inline void iwl_enable_hw_int_msk_msix(struct iwl_trans *trans, u32 msk) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index 0296c29..45f1b7e 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -1535,7 +1535,7 @@ irqreturn_t iwl_pcie_irq_handler(int irq, void *dev_id) * have anything to service */ if (test_bit(STATUS_INT_ENABLED, &trans->status)) - iwl_enable_interrupts(trans); + _iwl_enable_interrupts(trans); spin_unlock(&trans_pcie->irq_lock); lock_map_release(&trans->sync_cmd_lockdep_map); return IRQ_NONE; @@ -1727,15 +1727,17 @@ irqreturn_t iwl_pcie_irq_handler(int irq, void *dev_id) inta & ~trans_pcie->inta_mask); } + spin_lock(&trans_pcie->irq_lock); + /* only Re-enable all interrupt if disabled by irq */ + if (test_bit(STATUS_INT_ENABLED, &trans->status)) + _iwl_enable_interrupts(trans); /* we are loading the firmware, enable FH_TX interrupt only */ - if (handled & CSR_INT_BIT_FH_TX) + else if (handled & CSR_INT_BIT_FH_TX) iwl_enable_fw_load_int(trans); - /* only Re-enable all interrupt if disabled by irq */ - else if (test_bit(STATUS_INT_ENABLED, &trans->status)) - iwl_enable_interrupts(trans); /* Re-enable RF_KILL if it occurred */ else if (handled & CSR_INT_BIT_RF_KILL) iwl_enable_rfkill_int(trans); + spin_unlock(&trans_pcie->irq_lock); out: lock_map_release(&trans->sync_cmd_lockdep_map); @@ -1799,7 +1801,7 @@ void iwl_pcie_reset_ict(struct iwl_trans *trans) return; spin_lock(&trans_pcie->irq_lock); - iwl_disable_interrupts(trans); + _iwl_disable_interrupts(trans); memset(trans_pcie->ict_tbl, 0, ICT_SIZE); @@ -1815,7 +1817,7 @@ void iwl_pcie_reset_ict(struct iwl_trans *trans) trans_pcie->use_ict = true; trans_pcie->ict_index = 0; iwl_write32(trans, CSR_INT, trans_pcie->inta_mask); - iwl_enable_interrupts(trans); + _iwl_enable_interrupts(trans); spin_unlock(&trans_pcie->irq_lock); } diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index 3b7a414..9e953a4 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -1037,9 +1037,7 @@ static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans, bool low_power) was_hw_rfkill = iwl_is_rfkill_set(trans); /* tell the device to stop sending interrupts */ - spin_lock(&trans_pcie->irq_lock); iwl_disable_interrupts(trans); - spin_unlock(&trans_pcie->irq_lock); /* device going down, Stop using ICT table */ iwl_pcie_disable_ict(trans); @@ -1083,9 +1081,7 @@ static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans, bool low_power) * the time, unless the interrupt is ACKed even if the interrupt * should be masked. Re-ACK all the interrupts here. */ - spin_lock(&trans_pcie->irq_lock); iwl_disable_interrupts(trans); - spin_unlock(&trans_pcie->irq_lock); /* clear all status bits */ clear_bit(STATUS_SYNC_HCMD_ACTIVE, &trans->status); @@ -1578,15 +1574,11 @@ static void iwl_trans_pcie_op_mode_leave(struct iwl_trans *trans) mutex_lock(&trans_pcie->mutex); /* disable interrupts - don't enable HW RF kill interrupt */ - spin_lock(&trans_pcie->irq_lock); iwl_disable_interrupts(trans); - spin_unlock(&trans_pcie->irq_lock); iwl_pcie_apm_stop(trans, true); - spin_lock(&trans_pcie->irq_lock); iwl_disable_interrupts(trans); - spin_unlock(&trans_pcie->irq_lock); iwl_pcie_disable_ict(trans); -- cgit v0.10.2 From b1753c62c72fe70035498b28901f9a4bdcf2b92a Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Tue, 21 Jun 2016 12:44:01 +0300 Subject: iwlwifi: pcie: track rxb status In MQ environment and new architecture in early stages we may encounter DMA issues. Track RXB status and bail out in case we receive index to an RXB that was not mapped and handed over to HW. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h index 54af3da..190f611 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h @@ -68,12 +68,14 @@ struct iwl_host_cmd; * struct iwl_rx_mem_buffer * @page_dma: bus address of rxb page * @page: driver's pointer to the rxb page + * @invalid: rxb is in driver ownership - not owned by HW * @vid: index of this rxb in the global table */ struct iwl_rx_mem_buffer { dma_addr_t page_dma; struct page *page; u16 vid; + bool invalid; struct list_head list; }; diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index 45f1b7e..153b308 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -266,7 +266,7 @@ static void iwl_pcie_rxmq_restock(struct iwl_trans *trans, rxb = list_first_entry(&rxq->rx_free, struct iwl_rx_mem_buffer, list); list_del(&rxb->list); - + rxb->invalid = false; /* 12 first bits are expected to be empty */ WARN_ON(rxb->page_dma & DMA_BIT_MASK(12)); /* Point to Rx buffer via next RBD in circular buffer */ @@ -317,6 +317,7 @@ static void iwl_pcie_rxsq_restock(struct iwl_trans *trans, rxb = list_first_entry(&rxq->rx_free, struct iwl_rx_mem_buffer, list); list_del(&rxb->list); + rxb->invalid = false; /* Point to Rx buffer via next RBD in circular buffer */ bd[rxq->write] = iwl_pcie_dma_addr2rbd_ptr(rxb->page_dma); @@ -961,6 +962,7 @@ int iwl_pcie_rx_init(struct iwl_trans *trans) list_add(&rxb->list, &def_rxq->rx_used); trans_pcie->global_table[i] = rxb; rxb->vid = (u16)(i + 1); + rxb->invalid = true; } iwl_pcie_rxq_alloc_rbs(trans, GFP_KERNEL, def_rxq); @@ -1256,6 +1258,12 @@ restart: goto out; } rxb = trans_pcie->global_table[vid - 1]; + if (WARN(rxb->invalid, + "Invalid rxb from HW %u\n", (u32)vid)) { + iwl_force_nmi(trans); + goto out; + } + rxb->invalid = true; } else { rxb = rxq->queue[i]; rxq->queue[i] = NULL; -- cgit v0.10.2 From 8de437c71eb924cc86e4265ab575289b85de2bf5 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Thu, 9 Jun 2016 17:56:38 +0300 Subject: iwlwifi: pcie: generalize and increase the size of scratchbuf Currently the scratch buffer is set to 16 bytes and indicates the size of the bi-directional DMA. However, next HW generation will perform additional offloading, and will write the result in the key location of the TX command, so the size of the bi-directional consistent memory should grow accordingly - increase it to 40. Generalize the code to get rid of now irrelevant scratch references. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h index 190f611..c3d8974 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h @@ -232,15 +232,16 @@ struct iwl_queue { #define TFD_CMD_SLOTS 32 /* - * The FH will write back to the first TB only, so we need - * to copy some data into the buffer regardless of whether - * it should be mapped or not. This indicates how big the - * first TB must be to include the scratch buffer. Since - * the scratch is 4 bytes at offset 12, it's 16 now. If we - * make it bigger then allocations will be bigger and copy - * slower, so that's probably not useful. + * The FH will write back to the first TB only, so we need to copy some data + * into the buffer regardless of whether it should be mapped or not. + * This indicates how big the first TB must be to include the scratch buffer + * and the assigned PN. + * Since PN location is 16 bytes at offset 24, it's 40 now. + * If we make it bigger then allocations will be bigger and copy slower, so + * that's probably not useful. */ -#define IWL_HCMD_SCRATCHBUF_SIZE 16 +#define IWL_FIRST_TB_SIZE 40 +#define IWL_FIRST_TB_SIZE_ALIGN ALIGN(IWL_FIRST_TB_SIZE, 64) struct iwl_pcie_txq_entry { struct iwl_device_cmd *cmd; @@ -250,20 +251,18 @@ struct iwl_pcie_txq_entry { struct iwl_cmd_meta meta; }; -struct iwl_pcie_txq_scratch_buf { - struct iwl_cmd_header hdr; - u8 buf[8]; - __le32 scratch; +struct iwl_pcie_first_tb_buf { + u8 buf[IWL_FIRST_TB_SIZE_ALIGN]; }; /** * struct iwl_txq - Tx Queue for DMA * @q: generic Rx/Tx queue descriptor * @tfds: transmit frame descriptors (DMA memory) - * @scratchbufs: start of command headers, including scratch buffers, for + * @first_tb_bufs: start of command headers, including scratch buffers, for * the writeback -- this is DMA memory and an array holding one buffer * for each command on the queue - * @scratchbufs_dma: DMA address for the scratchbufs start + * @first_tb_dma: DMA address for the first_tb_bufs start * @entries: transmit entries (driver state) * @lock: queue lock * @stuck_timer: timer that fires if queue gets stuck @@ -281,8 +280,8 @@ struct iwl_pcie_txq_scratch_buf { struct iwl_txq { struct iwl_queue q; struct iwl_tfd *tfds; - struct iwl_pcie_txq_scratch_buf *scratchbufs; - dma_addr_t scratchbufs_dma; + struct iwl_pcie_first_tb_buf *first_tb_bufs; + dma_addr_t first_tb_dma; struct iwl_pcie_txq_entry *entries; spinlock_t lock; unsigned long frozen_expiry_remainder; @@ -298,10 +297,10 @@ struct iwl_txq { }; static inline dma_addr_t -iwl_pcie_get_scratchbuf_dma(struct iwl_txq *txq, int idx) +iwl_pcie_get_first_tb_dma(struct iwl_txq *txq, int idx) { - return txq->scratchbufs_dma + - sizeof(struct iwl_pcie_txq_scratch_buf) * idx; + return txq->first_tb_dma + + sizeof(struct iwl_pcie_first_tb_buf) * idx; } struct iwl_tso_hdr_page { diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c index 2b68826..9a1c006 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c @@ -393,7 +393,7 @@ static void iwl_pcie_tfd_unmap(struct iwl_trans *trans, return; } - /* first TB is never freed - it's the scratchbuf data */ + /* first TB is never freed - it's the bidirectional DMA data */ for (i = 1; i < num_tbs; i++) { if (meta->flags & BIT(i + CMD_TB_BITMAP_POS)) @@ -491,7 +491,7 @@ static int iwl_pcie_txq_alloc(struct iwl_trans *trans, { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); size_t tfd_sz = sizeof(struct iwl_tfd) * TFD_QUEUE_SIZE_MAX; - size_t scratchbuf_sz; + size_t tb0_buf_sz; int i; if (WARN_ON(txq->entries || txq->tfds)) @@ -526,17 +526,14 @@ static int iwl_pcie_txq_alloc(struct iwl_trans *trans, if (!txq->tfds) goto error; - BUILD_BUG_ON(IWL_HCMD_SCRATCHBUF_SIZE != sizeof(*txq->scratchbufs)); - BUILD_BUG_ON(offsetof(struct iwl_pcie_txq_scratch_buf, scratch) != - sizeof(struct iwl_cmd_header) + - offsetof(struct iwl_tx_cmd, scratch)); + BUILD_BUG_ON(IWL_FIRST_TB_SIZE_ALIGN != sizeof(*txq->first_tb_bufs)); - scratchbuf_sz = sizeof(*txq->scratchbufs) * slots_num; + tb0_buf_sz = sizeof(*txq->first_tb_bufs) * slots_num; - txq->scratchbufs = dma_alloc_coherent(trans->dev, scratchbuf_sz, - &txq->scratchbufs_dma, + txq->first_tb_bufs = dma_alloc_coherent(trans->dev, tb0_buf_sz, + &txq->first_tb_dma, GFP_KERNEL); - if (!txq->scratchbufs) + if (!txq->first_tb_bufs) goto err_free_tfds; txq->q.id = txq_id; @@ -708,8 +705,8 @@ static void iwl_pcie_txq_free(struct iwl_trans *trans, int txq_id) txq->tfds = NULL; dma_free_coherent(dev, - sizeof(*txq->scratchbufs) * txq->q.n_window, - txq->scratchbufs, txq->scratchbufs_dma); + sizeof(*txq->first_tb_bufs) * txq->q.n_window, + txq->first_tb_bufs, txq->first_tb_dma); } kfree(txq->entries); @@ -1422,7 +1419,7 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans, void *dup_buf = NULL; dma_addr_t phys_addr; int idx; - u16 copy_size, cmd_size, scratch_size; + u16 copy_size, cmd_size, tb0_size; bool had_nocopy = false; u8 group_id = iwl_cmd_groupid(cmd->id); int i, ret; @@ -1453,9 +1450,9 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans, if (!cmd->len[i]) continue; - /* need at least IWL_HCMD_SCRATCHBUF_SIZE copied */ - if (copy_size < IWL_HCMD_SCRATCHBUF_SIZE) { - int copy = IWL_HCMD_SCRATCHBUF_SIZE - copy_size; + /* need at least IWL_FIRST_TB_SIZE copied */ + if (copy_size < IWL_FIRST_TB_SIZE) { + int copy = IWL_FIRST_TB_SIZE - copy_size; if (copy > cmdlen[i]) copy = cmdlen[i]; @@ -1576,8 +1573,8 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans, } /* - * Otherwise we need at least IWL_HCMD_SCRATCHBUF_SIZE copied - * in total (for the scratchbuf handling), but copy up to what + * Otherwise we need at least IWL_FIRST_TB_SIZE copied + * in total (for bi-directional DMA), but copy up to what * we can fit into the payload for debug dump purposes. */ copy = min_t(int, TFD_MAX_PAYLOAD_SIZE - cmd_pos, cmd->len[i]); @@ -1586,8 +1583,8 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans, cmd_pos += copy; /* However, treat copy_size the proper way, we need it below */ - if (copy_size < IWL_HCMD_SCRATCHBUF_SIZE) { - copy = IWL_HCMD_SCRATCHBUF_SIZE - copy_size; + if (copy_size < IWL_FIRST_TB_SIZE) { + copy = IWL_FIRST_TB_SIZE - copy_size; if (copy > cmd->len[i]) copy = cmd->len[i]; @@ -1602,18 +1599,18 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans, le16_to_cpu(out_cmd->hdr.sequence), cmd_size, q->write_ptr, idx, trans_pcie->cmd_queue); - /* start the TFD with the scratchbuf */ - scratch_size = min_t(int, copy_size, IWL_HCMD_SCRATCHBUF_SIZE); - memcpy(&txq->scratchbufs[idx], &out_cmd->hdr, scratch_size); + /* start the TFD with the minimum copy bytes */ + tb0_size = min_t(int, copy_size, IWL_FIRST_TB_SIZE); + memcpy(&txq->first_tb_bufs[idx], &out_cmd->hdr, tb0_size); iwl_pcie_txq_build_tfd(trans, txq, - iwl_pcie_get_scratchbuf_dma(txq, idx), - scratch_size, true); + iwl_pcie_get_first_tb_dma(txq, idx), + tb0_size, true); /* map first command fragment, if any remains */ - if (copy_size > scratch_size) { + if (copy_size > tb0_size) { phys_addr = dma_map_single(trans->dev, - ((u8 *)&out_cmd->hdr) + scratch_size, - copy_size - scratch_size, + ((u8 *)&out_cmd->hdr) + tb0_size, + copy_size - tb0_size, DMA_TO_DEVICE); if (dma_mapping_error(trans->dev, phys_addr)) { iwl_pcie_tfd_unmap(trans, out_meta, @@ -1623,7 +1620,7 @@ static int iwl_pcie_enqueue_hcmd(struct iwl_trans *trans, } iwl_pcie_txq_build_tfd(trans, txq, phys_addr, - copy_size - scratch_size, false); + copy_size - tb0_size, false); } /* map the remaining (adjusted) nocopy/dup fragments */ @@ -1968,7 +1965,7 @@ static int iwl_fill_data_tbs(struct iwl_trans *trans, struct sk_buff *skb, trace_iwlwifi_dev_tx(trans->dev, skb, &txq->tfds[txq->q.write_ptr], sizeof(struct iwl_tfd), - &dev_cmd->hdr, IWL_HCMD_SCRATCHBUF_SIZE + tb1_len, + &dev_cmd->hdr, IWL_FIRST_TB_SIZE + tb1_len, skb->data + hdr_len, tb2_len); trace_iwlwifi_dev_tx_data(trans->dev, skb, hdr_len, skb->len - hdr_len); @@ -2044,7 +2041,7 @@ static int iwl_fill_data_tbs_amsdu(struct iwl_trans *trans, struct sk_buff *skb, trace_iwlwifi_dev_tx(trans->dev, skb, &txq->tfds[txq->q.write_ptr], sizeof(struct iwl_tfd), - &dev_cmd->hdr, IWL_HCMD_SCRATCHBUF_SIZE + tb1_len, + &dev_cmd->hdr, IWL_FIRST_TB_SIZE + tb1_len, NULL, 0); ip_hdrlen = skb_transport_header(skb) - skb_network_header(skb); @@ -2306,7 +2303,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb, cpu_to_le16((u16)(QUEUE_TO_SEQ(txq_id) | INDEX_TO_SEQ(q->write_ptr))); - tb0_phys = iwl_pcie_get_scratchbuf_dma(txq, q->write_ptr); + tb0_phys = iwl_pcie_get_first_tb_dma(txq, q->write_ptr); scratch_phys = tb0_phys + sizeof(struct iwl_cmd_header) + offsetof(struct iwl_tx_cmd, scratch); @@ -2324,7 +2321,7 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb, * setup of the first TB) */ len = sizeof(struct iwl_tx_cmd) + sizeof(struct iwl_cmd_header) + - hdr_len - IWL_HCMD_SCRATCHBUF_SIZE; + hdr_len - IWL_FIRST_TB_SIZE; /* do not align A-MSDU to dword as the subframe header aligns it */ amsdu = ieee80211_is_data_qos(fc) && (*ieee80211_get_qos_ctl(hdr) & @@ -2338,17 +2335,17 @@ int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb, tb1_len = len; } - /* The first TB points to the scratchbuf data - min_copy bytes */ - memcpy(&txq->scratchbufs[q->write_ptr], &dev_cmd->hdr, - IWL_HCMD_SCRATCHBUF_SIZE); + /* The first TB points to bi-directional DMA data */ + memcpy(&txq->first_tb_bufs[q->write_ptr], &dev_cmd->hdr, + IWL_FIRST_TB_SIZE); iwl_pcie_txq_build_tfd(trans, txq, tb0_phys, - IWL_HCMD_SCRATCHBUF_SIZE, true); + IWL_FIRST_TB_SIZE, true); /* there must be data left over for TB1 or this code must be changed */ - BUILD_BUG_ON(sizeof(struct iwl_tx_cmd) < IWL_HCMD_SCRATCHBUF_SIZE); + BUILD_BUG_ON(sizeof(struct iwl_tx_cmd) < IWL_FIRST_TB_SIZE); /* map the data for TB1 */ - tb1_addr = ((u8 *)&dev_cmd->hdr) + IWL_HCMD_SCRATCHBUF_SIZE; + tb1_addr = ((u8 *)&dev_cmd->hdr) + IWL_FIRST_TB_SIZE; tb1_phys = dma_map_single(trans->dev, tb1_addr, tb1_len, DMA_TO_DEVICE); if (unlikely(dma_mapping_error(trans->dev, tb1_phys))) goto out_err; -- cgit v0.10.2 From 12a17458a2078342de874a0bdc5d7b3e4ac01289 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Thu, 23 Jun 2016 12:04:55 +0300 Subject: iwlwifi: centralize 64 bit HW registers write Move the write_prph_64 of pcie to be transport agnostic. Add direct write as well, as it is needed for a000 HW. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-io.h b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-io.h index 27914ee..1dccae6 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-io.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-devtrace-io.h @@ -1,6 +1,7 @@ /****************************************************************************** * * Copyright(c) 2009 - 2014 Intel Corporation. All rights reserved. + * Copyright(c) 2016 Intel Deutschland GmbH * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License as @@ -83,6 +84,23 @@ TRACE_EVENT(iwlwifi_dev_iowrite32, __get_str(dev), __entry->offs, __entry->val) ); +TRACE_EVENT(iwlwifi_dev_iowrite64, + TP_PROTO(const struct device *dev, u64 offs, u64 val), + TP_ARGS(dev, offs, val), + TP_STRUCT__entry( + DEV_ENTRY + __field(u64, offs) + __field(u64, val) + ), + TP_fast_assign( + DEV_ASSIGN; + __entry->offs = offs; + __entry->val = val; + ), + TP_printk("[%s] write io[%llu] = %llu)", + __get_str(dev), __entry->offs, __entry->val) +); + TRACE_EVENT(iwlwifi_dev_iowrite_prph32, TP_PROTO(const struct device *dev, u32 offs, u32 val), TP_ARGS(dev, offs, val), @@ -100,6 +118,23 @@ TRACE_EVENT(iwlwifi_dev_iowrite_prph32, __get_str(dev), __entry->offs, __entry->val) ); +TRACE_EVENT(iwlwifi_dev_iowrite_prph64, + TP_PROTO(const struct device *dev, u64 offs, u64 val), + TP_ARGS(dev, offs, val), + TP_STRUCT__entry( + DEV_ENTRY + __field(u64, offs) + __field(u64, val) + ), + TP_fast_assign( + DEV_ASSIGN; + __entry->offs = offs; + __entry->val = val; + ), + TP_printk("[%s] write PRPH[%llu] = %llu)", + __get_str(dev), __entry->offs, __entry->val) +); + TRACE_EVENT(iwlwifi_dev_ioread_prph32, TP_PROTO(const struct device *dev, u32 offs, u32 val), TP_ARGS(dev, offs, val), diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-io.c b/drivers/net/wireless/intel/iwlwifi/iwl-io.c index d8b4306..92c8b5f 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-io.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-io.c @@ -51,6 +51,14 @@ void iwl_write32(struct iwl_trans *trans, u32 ofs, u32 val) } IWL_EXPORT_SYMBOL(iwl_write32); +void iwl_write64(struct iwl_trans *trans, u64 ofs, u64 val) +{ + trace_iwlwifi_dev_iowrite64(trans->dev, ofs, val); + iwl_trans_write32(trans, ofs, val & 0xffffffff); + iwl_trans_write32(trans, ofs + 4, val >> 32); +} +IWL_EXPORT_SYMBOL(iwl_write64); + u32 iwl_read32(struct iwl_trans *trans, u32 ofs) { u32 val = iwl_trans_read32(trans, ofs); @@ -102,6 +110,17 @@ void iwl_write_direct32(struct iwl_trans *trans, u32 reg, u32 value) } IWL_EXPORT_SYMBOL(iwl_write_direct32); +void iwl_write_direct64(struct iwl_trans *trans, u64 reg, u64 value) +{ + unsigned long flags; + + if (iwl_trans_grab_nic_access(trans, &flags)) { + iwl_write64(trans, reg, value); + iwl_trans_release_nic_access(trans, &flags); + } +} +IWL_EXPORT_SYMBOL(iwl_write_direct64); + int iwl_poll_direct_bit(struct iwl_trans *trans, u32 addr, u32 mask, int timeout) { @@ -133,6 +152,14 @@ void iwl_write_prph_no_grab(struct iwl_trans *trans, u32 ofs, u32 val) } IWL_EXPORT_SYMBOL(iwl_write_prph_no_grab); +void iwl_write_prph64_no_grab(struct iwl_trans *trans, u64 ofs, u64 val) +{ + trace_iwlwifi_dev_iowrite_prph64(trans->dev, ofs, val); + iwl_write_prph_no_grab(trans, ofs, val & 0xffffffff); + iwl_write_prph_no_grab(trans, ofs + 4, val >> 32); +} +IWL_EXPORT_SYMBOL(iwl_write_prph64_no_grab); + u32 iwl_read_prph(struct iwl_trans *trans, u32 ofs) { unsigned long flags; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-io.h b/drivers/net/wireless/intel/iwlwifi/iwl-io.h index a9bcc78..5c8c0e1 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-io.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-io.h @@ -34,6 +34,7 @@ void iwl_write8(struct iwl_trans *trans, u32 ofs, u8 val); void iwl_write32(struct iwl_trans *trans, u32 ofs, u32 val); +void iwl_write64(struct iwl_trans *trans, u64 ofs, u64 val); u32 iwl_read32(struct iwl_trans *trans, u32 ofs); static inline void iwl_set_bit(struct iwl_trans *trans, u32 reg, u32 mask) @@ -53,11 +54,13 @@ int iwl_poll_direct_bit(struct iwl_trans *trans, u32 addr, u32 mask, u32 iwl_read_direct32(struct iwl_trans *trans, u32 reg); void iwl_write_direct32(struct iwl_trans *trans, u32 reg, u32 value); +void iwl_write_direct64(struct iwl_trans *trans, u64 reg, u64 value); u32 iwl_read_prph_no_grab(struct iwl_trans *trans, u32 ofs); u32 iwl_read_prph(struct iwl_trans *trans, u32 ofs); void iwl_write_prph_no_grab(struct iwl_trans *trans, u32 ofs, u32 val); +void iwl_write_prph64_no_grab(struct iwl_trans *trans, u64 ofs, u64 val); void iwl_write_prph(struct iwl_trans *trans, u32 ofs, u32 val); int iwl_poll_prph_bit(struct iwl_trans *trans, u32 addr, u32 bits, u32 mask, int timeout); diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c index 153b308..5c36e6d 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c @@ -161,13 +161,6 @@ static inline __le32 iwl_pcie_dma_addr2rbd_ptr(dma_addr_t dma_addr) return cpu_to_le32((u32)(dma_addr >> 8)); } -static void iwl_pcie_write_prph_64_no_grab(struct iwl_trans *trans, u64 ofs, - u64 val) -{ - iwl_write_prph_no_grab(trans, ofs, val & 0xffffffff); - iwl_write_prph_no_grab(trans, ofs + 4, val >> 32); -} - /* * iwl_pcie_rx_stop - stops the Rx DMA */ @@ -817,17 +810,17 @@ static void iwl_pcie_rx_mq_hw_init(struct iwl_trans *trans) for (i = 0; i < trans->num_rx_queues; i++) { /* Tell device where to find RBD free table in DRAM */ - iwl_pcie_write_prph_64_no_grab(trans, - RFH_Q_FRBDCB_BA_LSB(i), - trans_pcie->rxq[i].bd_dma); + iwl_write_prph64_no_grab(trans, + RFH_Q_FRBDCB_BA_LSB(i), + trans_pcie->rxq[i].bd_dma); /* Tell device where to find RBD used table in DRAM */ - iwl_pcie_write_prph_64_no_grab(trans, - RFH_Q_URBDCB_BA_LSB(i), - trans_pcie->rxq[i].used_bd_dma); + iwl_write_prph64_no_grab(trans, + RFH_Q_URBDCB_BA_LSB(i), + trans_pcie->rxq[i].used_bd_dma); /* Tell device where in DRAM to update its Rx status */ - iwl_pcie_write_prph_64_no_grab(trans, - RFH_Q_URBD_STTS_WPTR_LSB(i), - trans_pcie->rxq[i].rb_stts_dma); + iwl_write_prph64_no_grab(trans, + RFH_Q_URBD_STTS_WPTR_LSB(i), + trans_pcie->rxq[i].rb_stts_dma); /* Reset device indice tables */ iwl_write_prph_no_grab(trans, RFH_Q_FRBDCB_WIDX(i), 0); iwl_write_prph_no_grab(trans, RFH_Q_FRBDCB_RIDX(i), 0); -- cgit v0.10.2 From e22744af3b79c42478888b20a063384f5349a21d Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Wed, 22 Jun 2016 17:23:34 +0300 Subject: iwlwifi: pcie: initialize a000 device's TFD table For a000 device the FH was replaced by the TFH. This is the first patch in a series introducing the changes stemming from this change. This patch initializes the TFQ queue table with the new 64 bit register and the relevant TFH configuration registers. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-a000.c b/drivers/net/wireless/intel/iwlwifi/iwl-a000.c index 220767e..4d78232 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-a000.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-a000.c @@ -115,7 +115,8 @@ static const struct iwl_ht_params iwl_a000_ht_params = { .apmg_not_supported = true, \ .mq_rx_supported = true, \ .vht_mu_mimo_supported = true, \ - .mac_addr_from_csr = true + .mac_addr_from_csr = true, \ + .use_tfh = true const struct iwl_cfg iwla000_2ac_cfg = { .name = "Intel(R) Dual Band Wireless AC a000", diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-config.h b/drivers/net/wireless/intel/iwlwifi/iwl-config.h index 64108cc..423b233 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-config.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-config.h @@ -365,7 +365,8 @@ struct iwl_cfg { mq_rx_supported:1, vht_mu_mimo_supported:1, rf_id:1, - integrated:1; + integrated:1, + use_tfh:1; u8 valid_tx_ant; u8 valid_rx_ant; u8 non_shared_ant; diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h index 73f0ed8..3b73467 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h @@ -77,6 +77,7 @@ */ #define FH_MEM_LOWER_BOUND (0x1000) #define FH_MEM_UPPER_BOUND (0x2000) +#define TFH_MEM_LOWER_BOUND (0xA06000) /** * Keep-Warm (KW) buffer base address. @@ -118,10 +119,17 @@ #define FH_MEM_CBBC_16_19_UPPER_BOUND (FH_MEM_LOWER_BOUND + 0xC00) #define FH_MEM_CBBC_20_31_LOWER_BOUND (FH_MEM_LOWER_BOUND + 0xB20) #define FH_MEM_CBBC_20_31_UPPER_BOUND (FH_MEM_LOWER_BOUND + 0xB80) +/* a000 TFD table address, 64 bit */ +#define TFH_TFDQ_CBB_TABLE (TFH_MEM_LOWER_BOUND + 0x1C00) /* Find TFD CB base pointer for given queue */ -static inline unsigned int FH_MEM_CBBC_QUEUE(unsigned int chnl) +static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans, + unsigned int chnl) { + if (trans->cfg->use_tfh) { + WARN_ON_ONCE(chnl >= 64); + return TFH_TFDQ_CBB_TABLE + 8 * chnl; + } if (chnl < 16) return FH_MEM_CBBC_0_15_LOWER_BOUND + 4 * chnl; if (chnl < 20) @@ -130,6 +138,36 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(unsigned int chnl) return FH_MEM_CBBC_20_31_LOWER_BOUND + 4 * (chnl - 20); } +/* a000 configuration registers */ + +/* + * TFH Configuration register. + * + * BIT fields: + * + * Bits 3:0: + * Define the maximum number of pending read requests. + * Maximum configration value allowed is 0xC + * Bits 9:8: + * Define the maximum transfer size. (64 / 128 / 256) + * Bit 10: + * When bit is set and transfer size is set to 128B, the TFH will enable + * reading chunks of more than 64B only if the read address is aligned to 128B. + * In case of DRAM read address which is not aligned to 128B, the TFH will + * enable transfer size which doesn't cross 64B DRAM address boundary. +*/ +#define TFH_TRANSFER_MODE (TFH_MEM_LOWER_BOUND + 0x1F40) +#define TFH_TRANSFER_MAX_PENDING_REQ 0xc +#define TFH_CHUNK_SIZE_128 BIT(8) +#define TFH_CHUNK_SPLIT_MODE BIT(10) +/* + * Defines the offset address in dwords referring from the beginning of the + * Tx CMD which will be updated in DRAM. + * Note that the TFH offset address for Tx CMD update is always referring to + * the start of the TFD first TB. + * In case of a DRAM Tx CMD update the TFH will update PN and Key ID + */ +#define TFH_TXCMD_UPDATE_CFG (TFH_MEM_LOWER_BOUND + 0x1F48) /** * Rx SRAM Control and Status Registers (RSCSR) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c index 9a1c006..d88d64c 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c @@ -70,6 +70,7 @@ * Tx queue resumed. * ***************************************************/ + static int iwl_queue_space(const struct iwl_queue *q) { unsigned int max; @@ -575,8 +576,13 @@ static int iwl_pcie_txq_init(struct iwl_trans *trans, struct iwl_txq *txq, * Tell nic where to find circular buffer of Tx Frame Descriptors for * given Tx queue, and enable the DMA channel used for that queue. * Circular buffer (TFD queue in DRAM) physical base address */ - iwl_write_direct32(trans, FH_MEM_CBBC_QUEUE(txq_id), - txq->q.dma_addr >> 8); + if (trans->cfg->use_tfh) + iwl_write_direct64(trans, + FH_MEM_CBBC_QUEUE(trans, txq_id), + txq->q.dma_addr); + else + iwl_write_direct32(trans, FH_MEM_CBBC_QUEUE(trans, txq_id), + txq->q.dma_addr >> 8); return 0; } @@ -783,9 +789,14 @@ void iwl_trans_pcie_tx_reset(struct iwl_trans *trans) for (txq_id = 0; txq_id < trans->cfg->base_params->num_of_queues; txq_id++) { struct iwl_txq *txq = &trans_pcie->txq[txq_id]; - - iwl_write_direct32(trans, FH_MEM_CBBC_QUEUE(txq_id), - txq->q.dma_addr >> 8); + if (trans->cfg->use_tfh) + iwl_write_direct64(trans, + FH_MEM_CBBC_QUEUE(trans, txq_id), + txq->q.dma_addr); + else + iwl_write_direct32(trans, + FH_MEM_CBBC_QUEUE(trans, txq_id), + txq->q.dma_addr >> 8); iwl_pcie_txq_unmap(trans, txq_id); txq->q.read_ptr = 0; txq->q.write_ptr = 0; @@ -993,6 +1004,12 @@ int iwl_pcie_tx_init(struct iwl_trans *trans) } } + if (trans->cfg->use_tfh) + iwl_write_direct32(trans, TFH_TRANSFER_MODE, + TFH_TRANSFER_MAX_PENDING_REQ | + TFH_CHUNK_SIZE_128 | + TFH_CHUNK_SPLIT_MODE); + iwl_set_bits_prph(trans, SCD_GP_CTRL, SCD_GP_CTRL_AUTO_ACTIVE_MODE); if (trans->cfg->base_params->num_of_queues > 20) iwl_set_bits_prph(trans, SCD_GP_CTRL, -- cgit v0.10.2 From 564cdce735da0ad036051052e5c01760cc70494f Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Wed, 22 Jun 2016 19:25:46 +0300 Subject: iwlwifi: pcie: load FW chunk for a000 devices Update the firmware load flow for TFH hardware. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h index 3b73467..1d6f5d2 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-fh.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-fh.h @@ -168,6 +168,35 @@ static inline unsigned int FH_MEM_CBBC_QUEUE(struct iwl_trans *trans, * In case of a DRAM Tx CMD update the TFH will update PN and Key ID */ #define TFH_TXCMD_UPDATE_CFG (TFH_MEM_LOWER_BOUND + 0x1F48) +/* + * Controls TX DMA operation + * + * BIT fields: + * + * Bits 31:30: Enable the SRAM DMA channel. + * Turning on bit 31 will kick the SRAM2DRAM DMA. + * Note that the sram2dram may be enabled only after configuring the DRAM and + * SRAM addresses registers and the byte count register. + * Bits 25:24: Defines the interrupt target upon dram2sram transfer done. When + * set to 1 - interrupt is sent to the driver + * Bit 0: Indicates the snoop configuration +*/ +#define TFH_SRV_DMA_CHNL0_CTRL (TFH_MEM_LOWER_BOUND + 0x1F60) +#define TFH_SRV_DMA_SNOOP BIT(0) +#define TFH_SRV_DMA_TO_DRIVER BIT(24) +#define TFH_SRV_DMA_START BIT(31) + +/* Defines the DMA SRAM write start address to transfer a data block */ +#define TFH_SRV_DMA_CHNL0_SRAM_ADDR (TFH_MEM_LOWER_BOUND + 0x1F64) + +/* Defines the 64bits DRAM start address to read the DMA data block from */ +#define TFH_SRV_DMA_CHNL0_DRAM_ADDR (TFH_MEM_LOWER_BOUND + 0x1F68) + +/* + * Defines the number of bytes to transfer from DRAM to SRAM. + * Note that this register may be configured with non-dword aligned size. + */ +#define TFH_SRV_DMA_CHNL0_BC (TFH_MEM_LOWER_BOUND + 0x1F70) /** * Rx SRAM Control and Status Registers (RSCSR) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index 9e953a4..af04dad 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -608,18 +608,10 @@ static int iwl_pcie_prepare_card_hw(struct iwl_trans *trans) /* * ucode */ -static int iwl_pcie_load_firmware_chunk(struct iwl_trans *trans, u32 dst_addr, - dma_addr_t phy_addr, u32 byte_cnt) +static void iwl_pcie_load_firmware_chunk_fh(struct iwl_trans *trans, + u32 dst_addr, dma_addr_t phy_addr, + u32 byte_cnt) { - struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); - unsigned long flags; - int ret; - - trans_pcie->ucode_write_complete = false; - - if (!iwl_trans_grab_nic_access(trans, &flags)) - return -EIO; - iwl_write32(trans, FH_TCSR_CHNL_TX_CONFIG_REG(FH_SRVC_CHNL), FH_TCSR_TX_CONFIG_REG_VAL_DMA_CHNL_PAUSE); @@ -642,7 +634,50 @@ static int iwl_pcie_load_firmware_chunk(struct iwl_trans *trans, u32 dst_addr, FH_TCSR_TX_CONFIG_REG_VAL_DMA_CHNL_ENABLE | FH_TCSR_TX_CONFIG_REG_VAL_DMA_CREDIT_DISABLE | FH_TCSR_TX_CONFIG_REG_VAL_CIRQ_HOST_ENDTFD); +} + +static void iwl_pcie_load_firmware_chunk_tfh(struct iwl_trans *trans, + u32 dst_addr, dma_addr_t phy_addr, + u32 byte_cnt) +{ + /* Stop DMA channel */ + iwl_write32(trans, TFH_SRV_DMA_CHNL0_CTRL, 0); + + /* Configure SRAM address */ + iwl_write32(trans, TFH_SRV_DMA_CHNL0_SRAM_ADDR, + dst_addr); + + /* Configure DRAM address - 64 bit */ + iwl_write64(trans, TFH_SRV_DMA_CHNL0_DRAM_ADDR, phy_addr); + /* Configure byte count to transfer */ + iwl_write32(trans, TFH_SRV_DMA_CHNL0_BC, byte_cnt); + + /* Enable the DRAM2SRAM to start */ + iwl_write32(trans, TFH_SRV_DMA_CHNL0_CTRL, TFH_SRV_DMA_SNOOP | + TFH_SRV_DMA_TO_DRIVER | + TFH_SRV_DMA_START); +} + +static int iwl_pcie_load_firmware_chunk(struct iwl_trans *trans, + u32 dst_addr, dma_addr_t phy_addr, + u32 byte_cnt) +{ + struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); + unsigned long flags; + int ret; + + trans_pcie->ucode_write_complete = false; + + if (!iwl_trans_grab_nic_access(trans, &flags)) + return -EIO; + + if (trans->cfg->use_tfh) + iwl_pcie_load_firmware_chunk_tfh(trans, dst_addr, phy_addr, + byte_cnt); + else + iwl_pcie_load_firmware_chunk_fh(trans, dst_addr, phy_addr, + byte_cnt); iwl_trans_release_nic_access(trans, &flags); ret = wait_event_timeout(trans_pcie->ucode_write_waitq, -- cgit v0.10.2 From 55bfa4b9d47d905b068b255659cc0ff039af2d21 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Wed, 29 Jun 2016 00:38:40 +0300 Subject: iwlwifi: mvm: support v4 of the TX power command Add support for the v4 version of the TX power command. Just add a new version and do the same sizing tricks that were done when support for v3 was introduced. This patch doesn't support the new functionality introduced, but makes the driver work with the new size of the command. Signed-off-by: Luca Coelho Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h index eb18de8..d56b3d8 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h @@ -328,6 +328,9 @@ typedef unsigned int __bitwise__ iwl_ucode_tlv_capa_t; * @IWL_UCODE_TLV_CAPA_EXTEND_SHARED_MEM_CFG: support getting more shared * memory addresses from the firmware. * @IWL_UCODE_TLV_CAPA_LQM_SUPPORT: supports Link Quality Measurement + * @IWL_UCODE_TLV_CAPA_TX_POWER_ACK: reduced TX power API has larger + * command size (command version 4) that supports toggling ACK TX + * power reduction. * * @NUM_IWL_UCODE_TLV_CAPA: number of bits used */ @@ -368,6 +371,7 @@ enum iwl_ucode_tlv_capa { IWL_UCODE_TLV_CAPA_USNIFFER_UNIFIED = (__force iwl_ucode_tlv_capa_t)77, IWL_UCODE_TLV_CAPA_EXTEND_SHARED_MEM_CFG = (__force iwl_ucode_tlv_capa_t)80, IWL_UCODE_TLV_CAPA_LQM_SUPPORT = (__force iwl_ucode_tlv_capa_t)81, + IWL_UCODE_TLV_CAPA_TX_POWER_ACK = (__force iwl_ucode_tlv_capa_t)84, NUM_IWL_UCODE_TLV_CAPA #ifdef __CHECKER__ diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-power.h b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-power.h index 65a7c8a..404b0de 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-power.h +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw-api-power.h @@ -7,7 +7,7 @@ * * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved. * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH - * Copyright(c) 2015 Intel Deutschland GmbH + * Copyright(c) 2015 - 2016 Intel Deutschland GmbH * * This program is free software; you can redistribute it and/or modify * it under the terms of version 2 of the GNU General Public License as @@ -34,7 +34,7 @@ * * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved. * Copyright(c) 2013 - 2014 Intel Mobile Communications GmbH - * Copyright(c) 2015 Intel Deutschland GmbH + * Copyright(c) 2015 - 2016 Intel Deutschland GmbH * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -310,7 +310,8 @@ enum iwl_dev_tx_power_cmd_mode { IWL_TX_POWER_MODE_SET_MAC = 0, IWL_TX_POWER_MODE_SET_DEVICE = 1, IWL_TX_POWER_MODE_SET_CHAINS = 2, -}; /* TX_POWER_REDUCED_FLAGS_TYPE_API_E_VER_2 */; + IWL_TX_POWER_MODE_SET_ACK = 3, +}; /* TX_POWER_REDUCED_FLAGS_TYPE_API_E_VER_4 */; /** * struct iwl_dev_tx_power_cmd_v2 - TX power reduction command @@ -338,7 +339,7 @@ struct iwl_dev_tx_power_cmd_v2 { * @v2: version 2 of the command, embedded here for easier software handling * @per_chain_restriction: per chain restrictions */ -struct iwl_dev_tx_power_cmd { +struct iwl_dev_tx_power_cmd_v3 { /* v3 is just an extension of v2 - keep this here */ struct iwl_dev_tx_power_cmd_v2 v2; __le16 per_chain_restriction[IWL_NUM_CHAIN_LIMITS][IWL_NUM_SUB_BANDS]; @@ -347,6 +348,19 @@ struct iwl_dev_tx_power_cmd { #define IWL_DEV_MAX_TX_POWER 0x7FFF /** + * struct iwl_dev_tx_power_cmd - TX power reduction command + * @v3: version 3 of the command, embedded here for easier software handling + * @enable_ack_reduction: enable or disable close range ack TX power + * reduction. + */ +struct iwl_dev_tx_power_cmd { + /* v4 is just an extension of v3 - keep this here */ + struct iwl_dev_tx_power_cmd_v3 v3; + u8 enable_ack_reduction; + u8 reserved[3]; +} __packed; /* TX_REDUCED_POWER_API_S_VER_4 */ + +/** * struct iwl_beacon_filter_cmd * REPLY_BEACON_FILTERING_CMD = 0xd2 (command) * @id_and_color: MAC contex identifier diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index 4c16fa7..7e0cdbf 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -1027,9 +1027,10 @@ static int iwl_mvm_sar_init(struct iwl_mvm *mvm) { struct iwl_mvm_sar_table sar_table; struct iwl_dev_tx_power_cmd cmd = { - .v2.set_mode = cpu_to_le32(IWL_TX_POWER_MODE_SET_CHAINS), + .v3.v2.set_mode = cpu_to_le32(IWL_TX_POWER_MODE_SET_CHAINS), }; int ret, i, j, idx; + int len = sizeof(cmd); /* we can't do anything with the table if the FW doesn't support it */ if (!fw_has_api(&mvm->fw->ucode_capa, @@ -1039,6 +1040,9 @@ static int iwl_mvm_sar_init(struct iwl_mvm *mvm) return 0; } + if (!fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_TX_POWER_ACK)) + len = sizeof(cmd.v3); + ret = iwl_mvm_sar_get_table(mvm, &sar_table); if (ret < 0) { IWL_DEBUG_RADIO(mvm, @@ -1060,15 +1064,14 @@ static int iwl_mvm_sar_init(struct iwl_mvm *mvm) IWL_DEBUG_RADIO(mvm, " Chain[%d]:\n", i); for (j = 0; j < IWL_NUM_SUB_BANDS; j++) { idx = (i * IWL_NUM_SUB_BANDS) + j; - cmd.per_chain_restriction[i][j] = + cmd.v3.per_chain_restriction[i][j] = cpu_to_le16(sar_table.values[idx]); IWL_DEBUG_RADIO(mvm, " Band[%d] = %d * .125dBm\n", j, sar_table.values[idx]); } } - ret = iwl_mvm_send_cmd_pdu(mvm, REDUCE_TX_POWER_CMD, 0, - sizeof(cmd), &cmd); + ret = iwl_mvm_send_cmd_pdu(mvm, REDUCE_TX_POWER_CMD, 0, len, &cmd); if (ret) IWL_ERR(mvm, "failed to set per-chain TX power: %d\n", ret); diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c index 73e49d3..53761c4 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c @@ -1250,18 +1250,20 @@ static int iwl_mvm_set_tx_power(struct iwl_mvm *mvm, struct ieee80211_vif *vif, s16 tx_power) { struct iwl_dev_tx_power_cmd cmd = { - .v2.set_mode = cpu_to_le32(IWL_TX_POWER_MODE_SET_MAC), - .v2.mac_context_id = + .v3.v2.set_mode = cpu_to_le32(IWL_TX_POWER_MODE_SET_MAC), + .v3.v2.mac_context_id = cpu_to_le32(iwl_mvm_vif_from_mac80211(vif)->id), - .v2.pwr_restriction = cpu_to_le16(8 * tx_power), + .v3.v2.pwr_restriction = cpu_to_le16(8 * tx_power), }; int len = sizeof(cmd); if (tx_power == IWL_DEFAULT_MAX_TX_POWER) - cmd.v2.pwr_restriction = cpu_to_le16(IWL_DEV_MAX_TX_POWER); + cmd.v3.v2.pwr_restriction = cpu_to_le16(IWL_DEV_MAX_TX_POWER); + if (!fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_TX_POWER_ACK)) + len = sizeof(cmd.v3); if (!fw_has_api(&mvm->fw->ucode_capa, IWL_UCODE_TLV_API_TX_POWER_CHAIN)) - len = sizeof(cmd.v2); + len = sizeof(cmd.v3.v2); return iwl_mvm_send_cmd_pdu(mvm, REDUCE_TX_POWER_CMD, 0, len, &cmd); } -- cgit v0.10.2 From 38398efb74b63e03c41d13429bf7f8afefe1dbe2 Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Thu, 30 Jun 2016 11:48:30 +0300 Subject: iwlwifi: pcie: centralize SCD status logging Centralize the logging of SCD status. The motivation is that for a000 devices we will have new SCD HW, but this code was duplicate anyway, so it is a proper cleanup. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h index c3d8974..e3047f2 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h @@ -476,6 +476,8 @@ void iwl_trans_pcie_txq_disable(struct iwl_trans *trans, int queue, bool configure_scd); void iwl_trans_pcie_txq_set_shared_mode(struct iwl_trans *trans, u32 txq_id, bool shared_mode); +void iwl_trans_pcie_log_scd_error(struct iwl_trans *trans, + struct iwl_txq *txq); int iwl_trans_pcie_tx(struct iwl_trans *trans, struct sk_buff *skb, struct iwl_device_cmd *dev_cmd, int txq_id); void iwl_pcie_txq_check_wrptrs(struct iwl_trans *trans); diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c index af04dad..74f2f03 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c @@ -1950,6 +1950,48 @@ static void iwl_trans_pcie_block_txq_ptrs(struct iwl_trans *trans, bool block) #define IWL_FLUSH_WAIT_MS 2000 +void iwl_trans_pcie_log_scd_error(struct iwl_trans *trans, struct iwl_txq *txq) +{ + struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); + u32 scd_sram_addr; + u8 buf[16]; + int cnt; + + IWL_ERR(trans, "Current SW read_ptr %d write_ptr %d\n", + txq->q.read_ptr, txq->q.write_ptr); + + scd_sram_addr = trans_pcie->scd_base_addr + + SCD_TX_STTS_QUEUE_OFFSET(txq->q.id); + iwl_trans_read_mem_bytes(trans, scd_sram_addr, buf, sizeof(buf)); + + iwl_print_hex_error(trans, buf, sizeof(buf)); + + for (cnt = 0; cnt < FH_TCSR_CHNL_NUM; cnt++) + IWL_ERR(trans, "FH TRBs(%d) = 0x%08x\n", cnt, + iwl_read_direct32(trans, FH_TX_TRB_REG(cnt))); + + for (cnt = 0; cnt < trans->cfg->base_params->num_of_queues; cnt++) { + u32 status = iwl_read_prph(trans, SCD_QUEUE_STATUS_BITS(cnt)); + u8 fifo = (status >> SCD_QUEUE_STTS_REG_POS_TXF) & 0x7; + bool active = !!(status & BIT(SCD_QUEUE_STTS_REG_POS_ACTIVE)); + u32 tbl_dw = + iwl_trans_read_mem32(trans, trans_pcie->scd_base_addr + + SCD_TRANS_TBL_OFFSET_QUEUE(cnt)); + + if (cnt & 0x1) + tbl_dw = (tbl_dw & 0xFFFF0000) >> 16; + else + tbl_dw = tbl_dw & 0x0000FFFF; + + IWL_ERR(trans, + "Q %d is %sactive and mapped to fifo %d ra_tid 0x%04x [%d,%d]\n", + cnt, active ? "" : "in", fifo, tbl_dw, + iwl_read_prph(trans, SCD_QUEUE_RDPTR(cnt)) & + (TFD_QUEUE_SIZE_MAX - 1), + iwl_read_prph(trans, SCD_QUEUE_WRPTR(cnt))); + } +} + static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, u32 txq_bm) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); @@ -1957,8 +1999,6 @@ static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, u32 txq_bm) struct iwl_queue *q; int cnt; unsigned long now = jiffies; - u32 scd_sram_addr; - u8 buf[16]; int ret = 0; /* waiting for all the tx frames complete might take a while */ @@ -1998,42 +2038,8 @@ static int iwl_trans_pcie_wait_txq_empty(struct iwl_trans *trans, u32 txq_bm) IWL_DEBUG_TX_QUEUES(trans, "Queue %d is now empty.\n", cnt); } - if (!ret) - return 0; - - IWL_ERR(trans, "Current SW read_ptr %d write_ptr %d\n", - txq->q.read_ptr, txq->q.write_ptr); - - scd_sram_addr = trans_pcie->scd_base_addr + - SCD_TX_STTS_QUEUE_OFFSET(txq->q.id); - iwl_trans_read_mem_bytes(trans, scd_sram_addr, buf, sizeof(buf)); - - iwl_print_hex_error(trans, buf, sizeof(buf)); - - for (cnt = 0; cnt < FH_TCSR_CHNL_NUM; cnt++) - IWL_ERR(trans, "FH TRBs(%d) = 0x%08x\n", cnt, - iwl_read_direct32(trans, FH_TX_TRB_REG(cnt))); - - for (cnt = 0; cnt < trans->cfg->base_params->num_of_queues; cnt++) { - u32 status = iwl_read_prph(trans, SCD_QUEUE_STATUS_BITS(cnt)); - u8 fifo = (status >> SCD_QUEUE_STTS_REG_POS_TXF) & 0x7; - bool active = !!(status & BIT(SCD_QUEUE_STTS_REG_POS_ACTIVE)); - u32 tbl_dw = - iwl_trans_read_mem32(trans, trans_pcie->scd_base_addr + - SCD_TRANS_TBL_OFFSET_QUEUE(cnt)); - - if (cnt & 0x1) - tbl_dw = (tbl_dw & 0xFFFF0000) >> 16; - else - tbl_dw = tbl_dw & 0x0000FFFF; - - IWL_ERR(trans, - "Q %d is %sactive and mapped to fifo %d ra_tid 0x%04x [%d,%d]\n", - cnt, active ? "" : "in", fifo, tbl_dw, - iwl_read_prph(trans, SCD_QUEUE_RDPTR(cnt)) & - (TFD_QUEUE_SIZE_MAX - 1), - iwl_read_prph(trans, SCD_QUEUE_WRPTR(cnt))); - } + if (ret) + iwl_trans_pcie_log_scd_error(trans, txq); return ret; } diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c index d88d64c..18650dc 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c @@ -155,10 +155,6 @@ static void iwl_pcie_txq_stuck_timer(unsigned long data) struct iwl_txq *txq = (void *)data; struct iwl_trans_pcie *trans_pcie = txq->trans_pcie; struct iwl_trans *trans = iwl_trans_pcie_get_trans(trans_pcie); - u32 scd_sram_addr = trans_pcie->scd_base_addr + - SCD_TX_STTS_QUEUE_OFFSET(txq->q.id); - u8 buf[16]; - int i; spin_lock(&txq->lock); /* check if triggered erroneously */ @@ -170,38 +166,8 @@ static void iwl_pcie_txq_stuck_timer(unsigned long data) IWL_ERR(trans, "Queue %d stuck for %u ms.\n", txq->q.id, jiffies_to_msecs(txq->wd_timeout)); - IWL_ERR(trans, "Current SW read_ptr %d write_ptr %d\n", - txq->q.read_ptr, txq->q.write_ptr); - - iwl_trans_read_mem_bytes(trans, scd_sram_addr, buf, sizeof(buf)); - - iwl_print_hex_error(trans, buf, sizeof(buf)); - for (i = 0; i < FH_TCSR_CHNL_NUM; i++) - IWL_ERR(trans, "FH TRBs(%d) = 0x%08x\n", i, - iwl_read_direct32(trans, FH_TX_TRB_REG(i))); - - for (i = 0; i < trans->cfg->base_params->num_of_queues; i++) { - u32 status = iwl_read_prph(trans, SCD_QUEUE_STATUS_BITS(i)); - u8 fifo = (status >> SCD_QUEUE_STTS_REG_POS_TXF) & 0x7; - bool active = !!(status & BIT(SCD_QUEUE_STTS_REG_POS_ACTIVE)); - u32 tbl_dw = - iwl_trans_read_mem32(trans, - trans_pcie->scd_base_addr + - SCD_TRANS_TBL_OFFSET_QUEUE(i)); - - if (i & 0x1) - tbl_dw = (tbl_dw & 0xFFFF0000) >> 16; - else - tbl_dw = tbl_dw & 0x0000FFFF; - - IWL_ERR(trans, - "Q %d is %sactive and mapped to fifo %d ra_tid 0x%04x [%d,%d]\n", - i, active ? "" : "in", fifo, tbl_dw, - iwl_read_prph(trans, SCD_QUEUE_RDPTR(i)) & - (TFD_QUEUE_SIZE_MAX - 1), - iwl_read_prph(trans, SCD_QUEUE_WRPTR(i))); - } + iwl_trans_pcie_log_scd_error(trans, txq); iwl_force_nmi(trans); } -- cgit v0.10.2 From 6f482e37b75d4e56cdb9b56ae7057f3ce67097bc Mon Sep 17 00:00:00 2001 From: Sara Sharon Date: Tue, 5 Jul 2016 18:50:17 +0300 Subject: iwlwifi: move iwl_drv to be shared across transports All transports has this structure. By moving it to be shared, we can get rid of casting to the specific transport in probe and remove. Signed-off-by: Sara Sharon Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h index 9ac47e0..5535e22 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-trans.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-trans.h @@ -761,6 +761,7 @@ enum iwl_plat_pm_mode { * @ops - pointer to iwl_trans_ops * @op_mode - pointer to the op_mode * @cfg - pointer to the configuration + * @drv - pointer to iwl_drv * @status: a bit-mask of transport status flags * @dev - pointer to struct device * that represents the device * @max_skb_frags: maximum number of fragments an SKB can have when transmitted. @@ -804,6 +805,7 @@ struct iwl_trans { const struct iwl_trans_ops *ops; struct iwl_op_mode *op_mode; const struct iwl_cfg *cfg; + struct iwl_drv *drv; enum iwl_trans_state state; unsigned long status; diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index f70c3e2..78cf9a7 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -610,7 +610,6 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) const struct iwl_cfg *cfg_7265d __maybe_unused = NULL; const struct iwl_cfg *cfg_9260lc __maybe_unused = NULL; struct iwl_trans *iwl_trans; - struct iwl_trans_pcie *trans_pcie; int ret; iwl_trans = iwl_trans_pcie_alloc(pdev, ent, cfg); @@ -648,12 +647,10 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) #endif pci_set_drvdata(pdev, iwl_trans); + iwl_trans->drv = iwl_drv_start(iwl_trans, cfg); - trans_pcie = IWL_TRANS_GET_PCIE_TRANS(iwl_trans); - trans_pcie->drv = iwl_drv_start(iwl_trans, cfg); - - if (IS_ERR(trans_pcie->drv)) { - ret = PTR_ERR(trans_pcie->drv); + if (IS_ERR(iwl_trans->drv)) { + ret = PTR_ERR(iwl_trans->drv); goto out_free_trans; } @@ -692,7 +689,7 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return 0; out_free_drv: - iwl_drv_stop(trans_pcie->drv); + iwl_drv_stop(iwl_trans->drv); out_free_trans: iwl_trans_pcie_free(iwl_trans); return ret; @@ -701,7 +698,6 @@ out_free_trans: static void iwl_pci_remove(struct pci_dev *pdev) { struct iwl_trans *trans = pci_get_drvdata(pdev); - struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); /* if RTPM was in use, restore it to the state before probe */ if (trans->runtime_pm_mode != IWL_PLAT_PM_MODE_DISABLED) { @@ -712,7 +708,7 @@ static void iwl_pci_remove(struct pci_dev *pdev) pm_runtime_forbid(trans->dev); } - iwl_drv_stop(trans_pcie->drv); + iwl_drv_stop(trans->drv); iwl_trans_pcie_free(trans); } diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h index e3047f2..11e347d 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h @@ -314,7 +314,6 @@ struct iwl_tso_hdr_page { * @rx_pool: initial pool of iwl_rx_mem_buffer for all the queues * @global_table: table mapping received VID from hw to rxb * @rba: allocator for RX replenishing - * @drv - pointer to iwl_drv * @trans: pointer to the generic transport area * @scd_base_addr: scheduler sram base address in SRAM * @scd_bc_tbls: pointer to the byte count table of the scheduler @@ -352,7 +351,6 @@ struct iwl_trans_pcie { struct iwl_rx_mem_buffer *global_table[RX_POOL_SIZE]; struct iwl_rb_allocator rba; struct iwl_trans *trans; - struct iwl_drv *drv; struct net_device napi_dev; -- cgit v0.10.2 From de9e5aeb4f40e72fa3bb087d378c9bd4ecf65c7f Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Wed, 29 Jun 2016 10:48:22 +0200 Subject: NFC: llcp: Fix usage of llcp_add_tlv() In functions using llcp_add_tlv(), a skb pointer could be set to NULL and then reuse afterward. With this patch, the skb pointer returned by llcp_add_tlv() is ignored since it can only be the passed skb pointer or NULL when the passed TLV is NULL. There is also no need to check for the TLV pointer as this is done by llcp_add_tlv(). Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/llcp_commands.c b/net/nfc/llcp_commands.c index 3425532..4112d00 100644 --- a/net/nfc/llcp_commands.c +++ b/net/nfc/llcp_commands.c @@ -438,12 +438,9 @@ int nfc_llcp_send_connect(struct nfc_llcp_sock *sock) goto error_tlv; } - if (service_name_tlv != NULL) - skb = llcp_add_tlv(skb, service_name_tlv, - service_name_tlv_length); - - skb = llcp_add_tlv(skb, miux_tlv, miux_tlv_length); - skb = llcp_add_tlv(skb, rw_tlv, rw_tlv_length); + llcp_add_tlv(skb, service_name_tlv, service_name_tlv_length); + llcp_add_tlv(skb, miux_tlv, miux_tlv_length); + llcp_add_tlv(skb, rw_tlv, rw_tlv_length); skb_queue_tail(&local->tx_queue, skb); @@ -493,8 +490,8 @@ int nfc_llcp_send_cc(struct nfc_llcp_sock *sock) goto error_tlv; } - skb = llcp_add_tlv(skb, miux_tlv, miux_tlv_length); - skb = llcp_add_tlv(skb, rw_tlv, rw_tlv_length); + llcp_add_tlv(skb, miux_tlv, miux_tlv_length); + llcp_add_tlv(skb, rw_tlv, rw_tlv_length); skb_queue_tail(&local->tx_queue, skb); -- cgit v0.10.2 From 256f3ee3d1468660ca3b10ad3beab7e8f6cbd969 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Wed, 29 Jun 2016 10:48:23 +0200 Subject: NFC: llcp: Fix 2 memory leaks Once copied into the sk_buff data area using llcp_add_tlv(), the allocated TLVs must be freed. With this patch nfc_llcp_send_connect() and nfc_llcp_send_cc() don't return immediately on success and now free the allocated TLVs. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/llcp_commands.c b/net/nfc/llcp_commands.c index 4112d00..c5959ce 100644 --- a/net/nfc/llcp_commands.c +++ b/net/nfc/llcp_commands.c @@ -444,10 +444,11 @@ int nfc_llcp_send_connect(struct nfc_llcp_sock *sock) skb_queue_tail(&local->tx_queue, skb); - return 0; + err = 0; error_tlv: - pr_err("error %d\n", err); + if (err) + pr_err("error %d\n", err); kfree(service_name_tlv); kfree(miux_tlv); @@ -495,10 +496,11 @@ int nfc_llcp_send_cc(struct nfc_llcp_sock *sock) skb_queue_tail(&local->tx_queue, skb); - return 0; + err = 0; error_tlv: - pr_err("error %d\n", err); + if (err) + pr_err("error %d\n", err); kfree(miux_tlv); kfree(rw_tlv); -- cgit v0.10.2 From e3e0258839a01f793a8a0c0885e8ad387681cdc6 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Thu, 16 Jun 2016 20:25:20 +0200 Subject: NFC: port100: Don't send a new command if one is still pending This patch ensures that a command is not still in process before sending a new one to the device. This can happen when neard is in constant polling mode: the configure_hw command can be sent when neard restarts polling after a LLCP SYMM timeout but before the device has returned in timeout from the last DEP frame sent. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/port100.c b/drivers/nfc/port100.c index 2d4bbe3..14a3cc2 100644 --- a/drivers/nfc/port100.c +++ b/drivers/nfc/port100.c @@ -809,6 +809,12 @@ static int port100_send_cmd_async(struct port100 *dev, u8 cmd_code, PORT100_FRAME_MAX_PAYLOAD_LEN + PORT100_FRAME_TAIL_LEN; + if (dev->cmd) { + nfc_err(&dev->interface->dev, + "A command is still in process\n"); + return -EBUSY; + } + resp = alloc_skb(resp_len, GFP_KERNEL); if (!resp) return -ENOMEM; -- cgit v0.10.2 From b74584c1a6d68110d135a6ce0336aab0055f4341 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Thu, 16 Jun 2016 20:25:21 +0200 Subject: NFC: port100: Fix the command cancellation process The USB out_urb used to send commands to the device can be submitted through the standard command processing queue coming from the Digital Protocol layer but it can also be submitted from port100_abort_cmd(). To not submit the URB while already active, a mutex is now used to protect it and a cmd_cancel flag is used to not send command while canceling the previous one. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/port100.c b/drivers/nfc/port100.c index 14a3cc2..909e3df 100644 --- a/drivers/nfc/port100.c +++ b/drivers/nfc/port100.c @@ -456,6 +456,12 @@ struct port100 { struct urb *out_urb; struct urb *in_urb; + /* This mutex protects the out_urb and avoids to submit a new command + * through port100_send_frame_async() while the previous one is being + * canceled through port100_abort_cmd(). + */ + struct mutex out_urb_lock; + struct work_struct cmd_complete_work; u8 cmd_type; @@ -464,6 +470,8 @@ struct port100 { * for any queuing/locking mechanism at driver level. */ struct port100_cmd *cmd; + + bool cmd_cancel; }; struct port100_cmd { @@ -718,10 +726,22 @@ static int port100_send_ack(struct port100 *dev) { int rc; + mutex_lock(&dev->out_urb_lock); + + usb_kill_urb(dev->out_urb); + dev->out_urb->transfer_buffer = ack_frame; dev->out_urb->transfer_buffer_length = sizeof(ack_frame); rc = usb_submit_urb(dev->out_urb, GFP_KERNEL); + /* Set the cmd_cancel flag only if the URB has been successfully + * submitted. It will be reset by the out URB completion callback + * port100_send_complete(). + */ + dev->cmd_cancel = !rc; + + mutex_unlock(&dev->out_urb_lock); + return rc; } @@ -730,6 +750,16 @@ static int port100_send_frame_async(struct port100 *dev, struct sk_buff *out, { int rc; + mutex_lock(&dev->out_urb_lock); + + /* A command cancel frame as been sent through dev->out_urb. Don't try + * to submit a new one. + */ + if (dev->cmd_cancel) { + rc = -EAGAIN; + goto exit; + } + dev->out_urb->transfer_buffer = out->data; dev->out_urb->transfer_buffer_length = out->len; @@ -741,16 +771,15 @@ static int port100_send_frame_async(struct port100 *dev, struct sk_buff *out, rc = usb_submit_urb(dev->out_urb, GFP_KERNEL); if (rc) - return rc; + goto exit; rc = port100_submit_urb_for_ack(dev, GFP_KERNEL); if (rc) - goto error; + usb_unlink_urb(dev->out_urb); - return 0; +exit: + mutex_unlock(&dev->out_urb_lock); -error: - usb_unlink_urb(dev->out_urb); return rc; } @@ -892,6 +921,8 @@ static void port100_send_complete(struct urb *urb) { struct port100 *dev = urb->context; + dev->cmd_cancel = false; + switch (urb->status) { case 0: break; /* success */ @@ -1455,6 +1486,7 @@ static int port100_probe(struct usb_interface *interface, if (!dev) return -ENOMEM; + mutex_init(&dev->out_urb_lock); dev->udev = usb_get_dev(interface_to_usbdev(interface)); dev->interface = interface; usb_set_intfdata(interface, dev); -- cgit v0.10.2 From a52bd7d2753b0567c71d604c640e9c4a86221756 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Thu, 16 Jun 2016 20:25:22 +0200 Subject: NFC: port100: Make port100_abort_cmd() synchronous This patch makes the abort_cmd function synchronous. This allows the caller to immediately send a new command after abort_cmd() returns. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/port100.c b/drivers/nfc/port100.c index 909e3df..481cb43 100644 --- a/drivers/nfc/port100.c +++ b/drivers/nfc/port100.c @@ -472,6 +472,7 @@ struct port100 { struct port100_cmd *cmd; bool cmd_cancel; + struct completion cmd_cancel_done; }; struct port100_cmd { @@ -728,6 +729,8 @@ static int port100_send_ack(struct port100 *dev) mutex_lock(&dev->out_urb_lock); + init_completion(&dev->cmd_cancel_done); + usb_kill_urb(dev->out_urb); dev->out_urb->transfer_buffer = ack_frame; @@ -742,6 +745,9 @@ static int port100_send_ack(struct port100 *dev) mutex_unlock(&dev->out_urb_lock); + if (!rc) + wait_for_completion(&dev->cmd_cancel_done); + return rc; } @@ -921,7 +927,10 @@ static void port100_send_complete(struct urb *urb) { struct port100 *dev = urb->context; - dev->cmd_cancel = false; + if (dev->cmd_cancel) { + dev->cmd_cancel = false; + complete(&dev->cmd_cancel_done); + } switch (urb->status) { case 0: -- cgit v0.10.2 From 9f0c4542c49cbd041f2b6943b16644af0a3ff48f Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Thu, 16 Jun 2016 20:25:23 +0200 Subject: NFC: port100: Abort current command before switching RF off If a command is still being processed by the device, the switch RF off command will be rejected. With this patch, the port100 driver calls port100_abort_cmd() before sending the switch RF off command. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/port100.c b/drivers/nfc/port100.c index 481cb43..2b2330b 100644 --- a/drivers/nfc/port100.c +++ b/drivers/nfc/port100.c @@ -1050,6 +1050,10 @@ static int port100_switch_rf(struct nfc_digital_dev *ddev, bool on) *skb_put(skb, 1) = on ? 1 : 0; + /* Cancel the last command if the device is being switched off */ + if (!on) + port100_abort_cmd(ddev); + resp = port100_send_cmd_sync(dev, PORT100_CMD_SWITCH_RF, skb); if (IS_ERR(resp)) -- cgit v0.10.2 From 8f49bec6c36c73f0c212e07229b19fb4bba85788 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Mon, 4 Jul 2016 16:42:55 +0200 Subject: NFC: nfcsim: Fix missing dependency on NFC_DIGITAL The nfcsim driver now depends on the Digital layer. This patch adds the missing dependency on NFC_DIGITAL for NFC_SIM config. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/Kconfig b/drivers/nfc/Kconfig index ea8321a..9d23692 100644 --- a/drivers/nfc/Kconfig +++ b/drivers/nfc/Kconfig @@ -40,6 +40,7 @@ config NFC_MEI_PHY config NFC_SIM tristate "NFC hardware simulator driver" + depends on NFC_DIGITAL help This driver declares two virtual NFC devices supporting NFC-DEP protocol. An LLCP connection can be established between them and -- cgit v0.10.2 From b77693447db987e77a39afaa8774e8702cb110d5 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Thu, 16 Jun 2016 20:24:41 +0200 Subject: NFC: digital: Fix a memory leak in NFC-F listening mode When configured as a target listening for a SENSF_REQ poll command, a nfcid2 array was allocated for no reason leading to a memory leak. The nfcid2 is sent by the target in the SENSF_RES reply. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_technology.c b/net/nfc/digital_technology.c index fb58ed2d..d9080de 100644 --- a/net/nfc/digital_technology.c +++ b/net/nfc/digital_technology.c @@ -1257,21 +1257,12 @@ static int digital_tg_config_nfcf(struct nfc_digital_dev *ddev, u8 rf_tech) int digital_tg_listen_nfcf(struct nfc_digital_dev *ddev, u8 rf_tech) { int rc; - u8 *nfcid2; rc = digital_tg_config_nfcf(ddev, rf_tech); if (rc) return rc; - nfcid2 = kzalloc(NFC_NFCID2_MAXSIZE, GFP_KERNEL); - if (!nfcid2) - return -ENOMEM; - - nfcid2[0] = DIGITAL_SENSF_NFCID2_NFC_DEP_B1; - nfcid2[1] = DIGITAL_SENSF_NFCID2_NFC_DEP_B2; - get_random_bytes(nfcid2 + 2, NFC_NFCID2_MAXSIZE - 2); - - return digital_tg_listen(ddev, 300, digital_tg_recv_sensf_req, nfcid2); + return digital_tg_listen(ddev, 300, digital_tg_recv_sensf_req, NULL); } void digital_tg_recv_md_req(struct nfc_digital_dev *ddev, void *arg, -- cgit v0.10.2 From 3f89fea35fc37b326d6b3697fcc9cba235a60811 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Thu, 16 Jun 2016 20:24:42 +0200 Subject: NFC: digital: Rework error handling in DEP_RES response The Digital Protocol stack used to send a NACK frame whatever the error type it receives in digital_in_recv_dep_res(). It actually should only send a NACK frame on CRC or parity check errors or on any transmission error if a NACK frame was previously sent. Existing drivers used to send EIO error for this kind of issues so this patch limits sending of NACK frames on EIO errors. All other errors will be reported to the upper layers. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_dep.c b/net/nfc/digital_dep.c index f72be74..b62c85d 100644 --- a/net/nfc/digital_dep.c +++ b/net/nfc/digital_dep.c @@ -664,7 +664,7 @@ static void digital_in_recv_dep_res(struct nfc_digital_dev *ddev, void *arg, rc = PTR_ERR(resp); resp = NULL; - if (((rc != -ETIMEDOUT) || ddev->nack_count) && + if ((rc == -EIO || (rc == -ETIMEDOUT && ddev->nack_count)) && (ddev->nack_count++ < DIGITAL_NFC_DEP_N_RETRY_NACK)) { ddev->atn_count = 0; -- cgit v0.10.2 From 82e57952869fbbdf09d8f9e7ac284df13741e93d Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Thu, 16 Jun 2016 20:24:43 +0200 Subject: NFC: digital: Call pending command callbacks at device unregister With this patch, when freeing the command queue in the module unregister function, the callbacks of the commands still queued are called with a ENODEV error. This gives a chance to the command issuer to free any memory it could have allocate. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_core.c b/net/nfc/digital_core.c index 27769ac..6e0b255 100644 --- a/net/nfc/digital_core.c +++ b/net/nfc/digital_core.c @@ -842,6 +842,14 @@ void nfc_digital_unregister_device(struct nfc_digital_dev *ddev) list_for_each_entry_safe(cmd, n, &ddev->cmd_queue, queue) { list_del(&cmd->queue); + + /* Call the command callback if any and pass it a ENODEV error. + * This gives a chance to the command issuer to free any + * allocated buffer. + */ + if (cmd->cmd_cb) + cmd->cmd_cb(ddev, cmd->cb_context, ERR_PTR(-ENODEV)); + kfree(cmd->mdaa_params); kfree(cmd); } -- cgit v0.10.2 From af66df0f53b9120437556d8eb00d70a36e791258 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Thu, 16 Jun 2016 20:24:44 +0200 Subject: NFC: digital: Set the command pending flag There is a flag in the command structure indicating that this command is pending. It was checked before sending the command to not send the same command twice but it was actually never set. This is now fixed. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_core.c b/net/nfc/digital_core.c index 6e0b255..0146e42 100644 --- a/net/nfc/digital_core.c +++ b/net/nfc/digital_core.c @@ -176,6 +176,8 @@ static void digital_wq_cmd(struct work_struct *work) return; } + cmd->pending = 1; + mutex_unlock(&ddev->cmd_lock); if (cmd->req) -- cgit v0.10.2 From 3cc952dbf1a7176b9247da4cd2612c9ddc1d1b51 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Thu, 16 Jun 2016 20:24:45 +0200 Subject: NFC: digital: Abort last command when dep link goes down With this patch, the Digital Protocol layer abort the last issued command when the dep link goes down. That way it does not have to wait for the driver to reply with a timeout error before sending a new command (i.e. a start poll command if constant polling is on). Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_core.c b/net/nfc/digital_core.c index 0146e42..0fd5518 100644 --- a/net/nfc/digital_core.c +++ b/net/nfc/digital_core.c @@ -612,6 +612,8 @@ static int digital_dep_link_down(struct nfc_dev *nfc_dev) { struct nfc_digital_dev *ddev = nfc_get_drvdata(nfc_dev); + digital_abort_cmd(ddev); + ddev->curr_protocol = 0; return 0; -- cgit v0.10.2 From 88b99d0b7af859bbd97e76d66527f107843340a5 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 1 Jul 2016 08:27:42 +0100 Subject: rxrpc: Fix some sparse errors Fix the following sparse errors: ../net/rxrpc/conn_object.c:77:17: warning: incorrect type in assignment (different base types) ../net/rxrpc/conn_object.c:77:17: expected restricted __be32 [usertype] call_id ../net/rxrpc/conn_object.c:77:17: got unsigned int [unsigned] [usertype] call_id ../net/rxrpc/conn_object.c:84:21: warning: restricted __be32 degrades to integer ../net/rxrpc/conn_object.c:86:26: warning: restricted __be32 degrades to integer ../net/rxrpc/conn_object.c:357:15: warning: incorrect type in assignment (different base types) ../net/rxrpc/conn_object.c:357:15: expected restricted __be32 [usertype] epoch ../net/rxrpc/conn_object.c:357:15: got unsigned int [unsigned] [usertype] epoch ../net/rxrpc/conn_object.c:369:21: warning: restricted __be32 degrades to integer ../net/rxrpc/conn_object.c:371:26: warning: restricted __be32 degrades to integer ../net/rxrpc/conn_object.c:411:21: warning: restricted __be32 degrades to integer ../net/rxrpc/conn_object.c:413:26: warning: restricted __be32 degrades to integer Signed-off-by: David Howells diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 4bfad7c..c86a3cf 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -70,7 +70,7 @@ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn, { struct rxrpc_call *xcall; struct rb_node *parent, **p; - __be32 call_id; + u32 call_id; write_lock_bh(&conn->lock); @@ -347,8 +347,7 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rb_node *p, **pp; const char *new = "old"; - __be32 epoch; - u32 cid; + u32 epoch, cid; _enter(""); -- cgit v0.10.2 From 689f4c646d6a8f0730eec11e06e5909de0b5d5d2 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 30 Jun 2016 11:34:30 +0100 Subject: rxrpc: Check the source of a packet to a client conn When looking up a client connection to which to route a packet, we need to check that the packet came from the correct source so that a peer can't try to muck around with another peer's connection. Signed-off-by: David Howells diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index c86a3cf..2c2456f 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -508,7 +508,9 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *local, } } else { conn = idr_find(&rxrpc_client_conn_ids, cid >> RXRPC_CIDSHIFT); - if (conn && conn->proto.epoch == epoch) + if (conn && + conn->proto.epoch == epoch && + conn->params.peer == peer) goto found; } -- cgit v0.10.2 From a263629da519b2064588377416e067727e2cbdf9 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sun, 26 Jun 2016 14:55:24 -0700 Subject: rxrpc: Avoid using stack memory in SG lists in rxkad rxkad uses stack memory in SG lists which would not work if stacks were allocated from vmalloc memory. In fact, in most cases this isn't even necessary as the stack memory ends up getting copied over to kmalloc memory. This patch eliminates all the unnecessary stack memory uses by supplying the final destination directly to the crypto API. In two instances where a temporary buffer is actually needed we also switch use a scratch area in the rxrpc_call struct (only one DATA packet will be being secured or verified at a time). Finally there is no need to split a split-page buffer into two SG entries so code dealing with that has been removed. Signed-off-by: Herbert Xu Signed-off-by: Andy Lutomirski Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 702db72..796368d 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -141,17 +141,16 @@ struct rxrpc_security { int (*init_connection_security)(struct rxrpc_connection *); /* prime a connection's packet security */ - void (*prime_packet_security)(struct rxrpc_connection *); + int (*prime_packet_security)(struct rxrpc_connection *); /* impose security on a packet */ - int (*secure_packet)(const struct rxrpc_call *, + int (*secure_packet)(struct rxrpc_call *, struct sk_buff *, size_t, void *); /* verify the security on a received packet */ - int (*verify_packet)(const struct rxrpc_call *, struct sk_buff *, - u32 *); + int (*verify_packet)(struct rxrpc_call *, struct sk_buff *, u32 *); /* issue a challenge */ int (*issue_challenge)(struct rxrpc_connection *); @@ -399,6 +398,7 @@ struct rxrpc_call { struct sk_buff_head rx_oos_queue; /* packets received out of sequence */ struct sk_buff *tx_pending; /* Tx socket buffer being filled */ wait_queue_head_t tx_waitq; /* wait for Tx window space to become available */ + __be32 crypto_buf[2]; /* Temporary packet crypto buffer */ unsigned long user_call_ID; /* user-defined call ID */ unsigned long creation_jif; /* time of call creation */ unsigned long flags; diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index bf69715..6a3c967 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -188,7 +188,10 @@ static int rxrpc_process_event(struct rxrpc_connection *conn, if (ret < 0) return ret; - conn->security->prime_packet_security(conn); + ret = conn->security->prime_packet_security(conn); + if (ret < 0) + return ret; + read_lock_bh(&conn->lock); spin_lock(&conn->state_lock); diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 2c2456f..c2c0926 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -138,7 +138,9 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, gfp_t gfp) if (ret < 0) goto error_1; - conn->security->prime_packet_security(conn); + ret = conn->security->prime_packet_security(conn); + if (ret < 0) + goto error_2; write_lock(&rxrpc_connection_lock); list_add_tail(&conn->link, &rxrpc_connections); @@ -152,6 +154,8 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, gfp_t gfp) _leave(" = %p", conn); return conn; +error_2: + conn->security->clear(conn); error_1: rxrpc_put_client_connection_id(conn); error_0: diff --git a/net/rxrpc/insecure.c b/net/rxrpc/insecure.c index e571403..c21ad21 100644 --- a/net/rxrpc/insecure.c +++ b/net/rxrpc/insecure.c @@ -17,11 +17,12 @@ static int none_init_connection_security(struct rxrpc_connection *conn) return 0; } -static void none_prime_packet_security(struct rxrpc_connection *conn) +static int none_prime_packet_security(struct rxrpc_connection *conn) { + return 0; } -static int none_secure_packet(const struct rxrpc_call *call, +static int none_secure_packet(struct rxrpc_call *call, struct sk_buff *skb, size_t data_size, void *sechdr) @@ -29,7 +30,7 @@ static int none_secure_packet(const struct rxrpc_call *call, return 0; } -static int none_verify_packet(const struct rxrpc_call *call, +static int none_verify_packet(struct rxrpc_call *call, struct sk_buff *skb, u32 *_abort_code) { diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index 23c05ec..3acc7c1 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -103,43 +103,43 @@ error: * prime the encryption state with the invariant parts of a connection's * description */ -static void rxkad_prime_packet_security(struct rxrpc_connection *conn) +static int rxkad_prime_packet_security(struct rxrpc_connection *conn) { struct rxrpc_key_token *token; SKCIPHER_REQUEST_ON_STACK(req, conn->cipher); - struct scatterlist sg[2]; + struct scatterlist sg; struct rxrpc_crypt iv; - struct { - __be32 x[4]; - } tmpbuf __attribute__((aligned(16))); /* must all be in same page */ + __be32 *tmpbuf; + size_t tmpsize = 4 * sizeof(__be32); _enter(""); if (!conn->params.key) - return; + return 0; + + tmpbuf = kmalloc(tmpsize, GFP_KERNEL); + if (!tmpbuf) + return -ENOMEM; token = conn->params.key->payload.data[0]; memcpy(&iv, token->kad->session_key, sizeof(iv)); - tmpbuf.x[0] = htonl(conn->proto.epoch); - tmpbuf.x[1] = htonl(conn->proto.cid); - tmpbuf.x[2] = 0; - tmpbuf.x[3] = htonl(conn->security_ix); - - sg_init_one(&sg[0], &tmpbuf, sizeof(tmpbuf)); - sg_init_one(&sg[1], &tmpbuf, sizeof(tmpbuf)); + tmpbuf[0] = htonl(conn->proto.epoch); + tmpbuf[1] = htonl(conn->proto.cid); + tmpbuf[2] = 0; + tmpbuf[3] = htonl(conn->security_ix); + sg_init_one(&sg, tmpbuf, tmpsize); skcipher_request_set_tfm(req, conn->cipher); skcipher_request_set_callback(req, 0, NULL, NULL); - skcipher_request_set_crypt(req, &sg[1], &sg[0], sizeof(tmpbuf), iv.x); - + skcipher_request_set_crypt(req, &sg, &sg, tmpsize, iv.x); crypto_skcipher_encrypt(req); skcipher_request_zero(req); - memcpy(&conn->csum_iv, &tmpbuf.x[2], sizeof(conn->csum_iv)); - ASSERTCMP((u32 __force)conn->csum_iv.n[0], ==, (u32 __force)tmpbuf.x[2]); - - _leave(""); + memcpy(&conn->csum_iv, tmpbuf + 2, sizeof(conn->csum_iv)); + kfree(tmpbuf); + _leave(" = 0"); + return 0; } /* @@ -152,12 +152,9 @@ static int rxkad_secure_packet_auth(const struct rxrpc_call *call, { struct rxrpc_skb_priv *sp; SKCIPHER_REQUEST_ON_STACK(req, call->conn->cipher); + struct rxkad_level1_hdr hdr; struct rxrpc_crypt iv; - struct scatterlist sg[2]; - struct { - struct rxkad_level1_hdr hdr; - __be32 first; /* first four bytes of data and padding */ - } tmpbuf __attribute__((aligned(8))); /* must all be in same page */ + struct scatterlist sg; u16 check; sp = rxrpc_skb(skb); @@ -167,24 +164,19 @@ static int rxkad_secure_packet_auth(const struct rxrpc_call *call, check = sp->hdr.seq ^ sp->hdr.callNumber; data_size |= (u32)check << 16; - tmpbuf.hdr.data_size = htonl(data_size); - memcpy(&tmpbuf.first, sechdr + 4, sizeof(tmpbuf.first)); + hdr.data_size = htonl(data_size); + memcpy(sechdr, &hdr, sizeof(hdr)); /* start the encryption afresh */ memset(&iv, 0, sizeof(iv)); - sg_init_one(&sg[0], &tmpbuf, sizeof(tmpbuf)); - sg_init_one(&sg[1], &tmpbuf, sizeof(tmpbuf)); - + sg_init_one(&sg, sechdr, 8); skcipher_request_set_tfm(req, call->conn->cipher); skcipher_request_set_callback(req, 0, NULL, NULL); - skcipher_request_set_crypt(req, &sg[1], &sg[0], sizeof(tmpbuf), iv.x); - + skcipher_request_set_crypt(req, &sg, &sg, 8, iv.x); crypto_skcipher_encrypt(req); skcipher_request_zero(req); - memcpy(sechdr, &tmpbuf, sizeof(tmpbuf)); - _leave(" = 0"); return 0; } @@ -198,8 +190,7 @@ static int rxkad_secure_packet_encrypt(const struct rxrpc_call *call, void *sechdr) { const struct rxrpc_key_token *token; - struct rxkad_level2_hdr rxkhdr - __attribute__((aligned(8))); /* must be all on one page */ + struct rxkad_level2_hdr rxkhdr; struct rxrpc_skb_priv *sp; SKCIPHER_REQUEST_ON_STACK(req, call->conn->cipher); struct rxrpc_crypt iv; @@ -218,18 +209,16 @@ static int rxkad_secure_packet_encrypt(const struct rxrpc_call *call, rxkhdr.data_size = htonl(data_size | (u32)check << 16); rxkhdr.checksum = 0; + memcpy(sechdr, &rxkhdr, sizeof(rxkhdr)); /* encrypt from the session key */ token = call->conn->params.key->payload.data[0]; memcpy(&iv, token->kad->session_key, sizeof(iv)); sg_init_one(&sg[0], sechdr, sizeof(rxkhdr)); - sg_init_one(&sg[1], &rxkhdr, sizeof(rxkhdr)); - skcipher_request_set_tfm(req, call->conn->cipher); skcipher_request_set_callback(req, 0, NULL, NULL); - skcipher_request_set_crypt(req, &sg[1], &sg[0], sizeof(rxkhdr), iv.x); - + skcipher_request_set_crypt(req, &sg[0], &sg[0], sizeof(rxkhdr), iv.x); crypto_skcipher_encrypt(req); /* we want to encrypt the skbuff in-place */ @@ -243,9 +232,7 @@ static int rxkad_secure_packet_encrypt(const struct rxrpc_call *call, sg_init_table(sg, nsg); skb_to_sgvec(skb, sg, 0, len); - skcipher_request_set_crypt(req, sg, sg, len, iv.x); - crypto_skcipher_encrypt(req); _leave(" = 0"); @@ -259,7 +246,7 @@ out: /* * checksum an RxRPC packet header */ -static int rxkad_secure_packet(const struct rxrpc_call *call, +static int rxkad_secure_packet(struct rxrpc_call *call, struct sk_buff *skb, size_t data_size, void *sechdr) @@ -267,10 +254,7 @@ static int rxkad_secure_packet(const struct rxrpc_call *call, struct rxrpc_skb_priv *sp; SKCIPHER_REQUEST_ON_STACK(req, call->conn->cipher); struct rxrpc_crypt iv; - struct scatterlist sg[2]; - struct { - __be32 x[2]; - } tmpbuf __attribute__((aligned(8))); /* must all be in same page */ + struct scatterlist sg; u32 x, y; int ret; @@ -293,20 +277,17 @@ static int rxkad_secure_packet(const struct rxrpc_call *call, /* calculate the security checksum */ x = call->channel << (32 - RXRPC_CIDSHIFT); x |= sp->hdr.seq & 0x3fffffff; - tmpbuf.x[0] = htonl(sp->hdr.callNumber); - tmpbuf.x[1] = htonl(x); - - sg_init_one(&sg[0], &tmpbuf, sizeof(tmpbuf)); - sg_init_one(&sg[1], &tmpbuf, sizeof(tmpbuf)); + call->crypto_buf[0] = htonl(sp->hdr.callNumber); + call->crypto_buf[1] = htonl(x); + sg_init_one(&sg, call->crypto_buf, 8); skcipher_request_set_tfm(req, call->conn->cipher); skcipher_request_set_callback(req, 0, NULL, NULL); - skcipher_request_set_crypt(req, &sg[1], &sg[0], sizeof(tmpbuf), iv.x); - + skcipher_request_set_crypt(req, &sg, &sg, 8, iv.x); crypto_skcipher_encrypt(req); skcipher_request_zero(req); - y = ntohl(tmpbuf.x[1]); + y = ntohl(call->crypto_buf[1]); y = (y >> 16) & 0xffff; if (y == 0) y = 1; /* zero checksums are not permitted */ @@ -367,7 +348,6 @@ static int rxkad_verify_packet_auth(const struct rxrpc_call *call, skcipher_request_set_tfm(req, call->conn->cipher); skcipher_request_set_callback(req, 0, NULL, NULL); skcipher_request_set_crypt(req, sg, sg, 8, iv.x); - crypto_skcipher_decrypt(req); skcipher_request_zero(req); @@ -452,7 +432,6 @@ static int rxkad_verify_packet_encrypt(const struct rxrpc_call *call, skcipher_request_set_tfm(req, call->conn->cipher); skcipher_request_set_callback(req, 0, NULL, NULL); skcipher_request_set_crypt(req, sg, sg, skb->len, iv.x); - crypto_skcipher_decrypt(req); skcipher_request_zero(req); if (sg != _sg) @@ -498,17 +477,14 @@ nomem: /* * verify the security on a received packet */ -static int rxkad_verify_packet(const struct rxrpc_call *call, +static int rxkad_verify_packet(struct rxrpc_call *call, struct sk_buff *skb, u32 *_abort_code) { SKCIPHER_REQUEST_ON_STACK(req, call->conn->cipher); struct rxrpc_skb_priv *sp; struct rxrpc_crypt iv; - struct scatterlist sg[2]; - struct { - __be32 x[2]; - } tmpbuf __attribute__((aligned(8))); /* must all be in same page */ + struct scatterlist sg; u16 cksum; u32 x, y; int ret; @@ -533,20 +509,17 @@ static int rxkad_verify_packet(const struct rxrpc_call *call, /* validate the security checksum */ x = call->channel << (32 - RXRPC_CIDSHIFT); x |= sp->hdr.seq & 0x3fffffff; - tmpbuf.x[0] = htonl(call->call_id); - tmpbuf.x[1] = htonl(x); - - sg_init_one(&sg[0], &tmpbuf, sizeof(tmpbuf)); - sg_init_one(&sg[1], &tmpbuf, sizeof(tmpbuf)); + call->crypto_buf[0] = htonl(call->call_id); + call->crypto_buf[1] = htonl(x); + sg_init_one(&sg, call->crypto_buf, 8); skcipher_request_set_tfm(req, call->conn->cipher); skcipher_request_set_callback(req, 0, NULL, NULL); - skcipher_request_set_crypt(req, &sg[1], &sg[0], sizeof(tmpbuf), iv.x); - + skcipher_request_set_crypt(req, &sg, &sg, 8, iv.x); crypto_skcipher_encrypt(req); skcipher_request_zero(req); - y = ntohl(tmpbuf.x[1]); + y = ntohl(call->crypto_buf[1]); cksum = (y >> 16) & 0xffff; if (cksum == 0) cksum = 1; /* zero checksums are not permitted */ @@ -710,29 +683,6 @@ static void rxkad_calc_response_checksum(struct rxkad_response *response) } /* - * load a scatterlist with a potentially split-page buffer - */ -static void rxkad_sg_set_buf2(struct scatterlist sg[2], - void *buf, size_t buflen) -{ - int nsg = 1; - - sg_init_table(sg, 2); - - sg_set_buf(&sg[0], buf, buflen); - if (sg[0].offset + buflen > PAGE_SIZE) { - /* the buffer was split over two pages */ - sg[0].length = PAGE_SIZE - sg[0].offset; - sg_set_buf(&sg[1], buf + sg[0].length, buflen - sg[0].length); - nsg++; - } - - sg_mark_end(&sg[nsg - 1]); - - ASSERTCMP(sg[0].length + sg[1].length, ==, buflen); -} - -/* * encrypt the response packet */ static void rxkad_encrypt_response(struct rxrpc_connection *conn, @@ -741,17 +691,16 @@ static void rxkad_encrypt_response(struct rxrpc_connection *conn, { SKCIPHER_REQUEST_ON_STACK(req, conn->cipher); struct rxrpc_crypt iv; - struct scatterlist sg[2]; + struct scatterlist sg[1]; /* continue encrypting from where we left off */ memcpy(&iv, s2->session_key, sizeof(iv)); - rxkad_sg_set_buf2(sg, &resp->encrypted, sizeof(resp->encrypted)); - + sg_init_table(sg, 1); + sg_set_buf(sg, &resp->encrypted, sizeof(resp->encrypted)); skcipher_request_set_tfm(req, conn->cipher); skcipher_request_set_callback(req, 0, NULL, NULL); skcipher_request_set_crypt(req, sg, sg, sizeof(resp->encrypted), iv.x); - crypto_skcipher_encrypt(req); skcipher_request_zero(req); } @@ -887,10 +836,8 @@ static int rxkad_decrypt_ticket(struct rxrpc_connection *conn, } sg_init_one(&sg[0], ticket, ticket_len); - skcipher_request_set_callback(req, 0, NULL, NULL); skcipher_request_set_crypt(req, sg, sg, ticket_len, iv.x); - crypto_skcipher_decrypt(req); skcipher_request_free(req); @@ -1001,7 +948,7 @@ static void rxkad_decrypt_response(struct rxrpc_connection *conn, const struct rxrpc_crypt *session_key) { SKCIPHER_REQUEST_ON_STACK(req, rxkad_ci); - struct scatterlist sg[2]; + struct scatterlist sg[1]; struct rxrpc_crypt iv; _enter(",,%08x%08x", @@ -1016,12 +963,11 @@ static void rxkad_decrypt_response(struct rxrpc_connection *conn, memcpy(&iv, session_key, sizeof(iv)); - rxkad_sg_set_buf2(sg, &resp->encrypted, sizeof(resp->encrypted)); - + sg_init_table(sg, 1); + sg_set_buf(sg, &resp->encrypted, sizeof(resp->encrypted)); skcipher_request_set_tfm(req, rxkad_ci); skcipher_request_set_callback(req, 0, NULL, NULL); skcipher_request_set_crypt(req, sg, sg, sizeof(resp->encrypted), iv.x); - crypto_skcipher_decrypt(req); skcipher_request_zero(req); -- cgit v0.10.2 From 5acbee4648789ba1fe9e7942280fb1966c76bd6f Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 27 Jun 2016 10:32:02 +0100 Subject: rxrpc: Provide queuing helper functions Provide queueing helper functions so that the queueing of local and connection objects can be fixed later. The issue is that a ref on the object needs to be passed to the work queue, but the act of queueing the object may fail because the object is already queued. Testing the queuedness of an object before hand doesn't work because there can be a race with someone else trying to queue it. What will have to be done is to adjust the refcount depending on the result of the queue operation. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 796368d..45aef3e 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -35,7 +35,6 @@ struct rxrpc_crypt { queue_delayed_work(rxrpc_workqueue, (WS), (D)) #define rxrpc_queue_call(CALL) rxrpc_queue_work(&(CALL)->processor) -#define rxrpc_queue_conn(CONN) rxrpc_queue_work(&(CONN)->processor) struct rxrpc_connection; @@ -566,6 +565,12 @@ static inline void rxrpc_get_connection(struct rxrpc_connection *conn) atomic_inc(&conn->usage); } + +static inline void rxrpc_queue_conn(struct rxrpc_connection *conn) +{ + rxrpc_queue_work(&conn->processor); +} + /* * input.c */ @@ -618,6 +623,11 @@ static inline void rxrpc_put_local(struct rxrpc_local *local) __rxrpc_put_local(local); } +static inline void rxrpc_queue_local(struct rxrpc_local *local) +{ + rxrpc_queue_work(&local->processor); +} + /* * misc.c */ diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index 6a3c967..d7e183c 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -318,7 +318,7 @@ void rxrpc_reject_packet(struct rxrpc_local *local, struct sk_buff *skb) CHECK_SLAB_OKAY(&local->usage); skb_queue_tail(&local->reject_queue, skb); - rxrpc_queue_work(&local->processor); + rxrpc_queue_local(local); } /* diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 5f26cae..fe7ff33 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -595,7 +595,7 @@ static void rxrpc_post_packet_to_local(struct rxrpc_local *local, _enter("%p,%p", local, skb); skb_queue_tail(&local->event_queue, skb); - rxrpc_queue_work(&local->processor); + rxrpc_queue_local(local); } /* -- cgit v0.10.2 From bba304db34ec3ca0d13e7f48e5a4e9896536cacc Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 27 Jun 2016 10:32:02 +0100 Subject: rxrpc: Turn connection #defines into enums and put outside struct def Turn the connection event and state #define lists into enums and move outside of the struct definition. Whilst we're at it, change _SERVER to _SERVICE in those identifiers and add EV_ into the event name to distinguish them from flags and states. Also add a symbol indicating the number of states and use that in the state text array. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 45aef3e..3f0d047 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -254,6 +254,35 @@ struct rxrpc_conn_parameters { }; /* + * Bits in the connection flags. + */ +enum rxrpc_conn_flag { + RXRPC_CONN_HAS_IDR, /* Has a client conn ID assigned */ +}; + +/* + * Events that can be raised upon a connection. + */ +enum rxrpc_conn_event { + RXRPC_CONN_EV_CHALLENGE, /* Send challenge packet */ +}; + +/* + * The connection protocol state. + */ +enum rxrpc_conn_proto_state { + RXRPC_CONN_UNUSED, /* Connection not yet attempted */ + RXRPC_CONN_CLIENT, /* Client connection */ + RXRPC_CONN_SERVICE_UNSECURED, /* Service unsecured connection */ + RXRPC_CONN_SERVICE_CHALLENGING, /* Service challenging for security */ + RXRPC_CONN_SERVICE, /* Service secured connection */ + RXRPC_CONN_REMOTELY_ABORTED, /* Conn aborted by peer */ + RXRPC_CONN_LOCALLY_ABORTED, /* Conn aborted locally */ + RXRPC_CONN_NETWORK_ERROR, /* Conn terminated by network error */ + RXRPC_CONN__NR_STATES +}; + +/* * RxRPC connection definition * - matched by { local, peer, epoch, conn_id, direction } * - each connection can only handle four simultaneous calls @@ -279,23 +308,12 @@ struct rxrpc_connection { struct crypto_skcipher *cipher; /* encryption handle */ struct rxrpc_crypt csum_iv; /* packet checksum base */ unsigned long flags; -#define RXRPC_CONN_HAS_IDR 0 /* - Has a client conn ID assigned */ unsigned long events; -#define RXRPC_CONN_CHALLENGE 0 /* send challenge packet */ unsigned long put_time; /* Time at which last put */ rwlock_t lock; /* access lock */ spinlock_t state_lock; /* state-change lock */ atomic_t usage; - enum { /* current state of connection */ - RXRPC_CONN_UNUSED, /* - connection not yet attempted */ - RXRPC_CONN_CLIENT, /* - client connection */ - RXRPC_CONN_SERVER_UNSECURED, /* - server unsecured connection */ - RXRPC_CONN_SERVER_CHALLENGING, /* - server challenging for security */ - RXRPC_CONN_SERVER, /* - server secured connection */ - RXRPC_CONN_REMOTELY_ABORTED, /* - conn aborted by peer */ - RXRPC_CONN_LOCALLY_ABORTED, /* - conn aborted locally */ - RXRPC_CONN_NETWORK_ERROR, /* - conn terminated by network error */ - } state; + enum rxrpc_conn_proto_state state : 8; /* current state of connection */ u32 local_abort; /* local abort code */ u32 remote_abort; /* remote abort code */ int error; /* local error incurred */ diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index 202e053..1c0860d 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -128,12 +128,12 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, spin_lock(&call->conn->state_lock); if (sp->hdr.securityIndex > 0 && - call->conn->state == RXRPC_CONN_SERVER_UNSECURED) { + call->conn->state == RXRPC_CONN_SERVICE_UNSECURED) { _debug("await conn sec"); list_add_tail(&call->accept_link, &rx->secureq); - call->conn->state = RXRPC_CONN_SERVER_CHALLENGING; + call->conn->state = RXRPC_CONN_SERVICE_CHALLENGING; rxrpc_get_connection(call->conn); - set_bit(RXRPC_CONN_CHALLENGE, &call->conn->events); + set_bit(RXRPC_CONN_EV_CHALLENGE, &call->conn->events); rxrpc_queue_conn(call->conn); } else { _debug("conn ready"); diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index d7e183c..b9c39b8 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -195,8 +195,8 @@ static int rxrpc_process_event(struct rxrpc_connection *conn, read_lock_bh(&conn->lock); spin_lock(&conn->state_lock); - if (conn->state == RXRPC_CONN_SERVER_CHALLENGING) { - conn->state = RXRPC_CONN_SERVER; + if (conn->state == RXRPC_CONN_SERVICE_CHALLENGING) { + conn->state = RXRPC_CONN_SERVICE; for (loop = 0; loop < RXRPC_MAXCALLS; loop++) rxrpc_call_is_secure(conn->channels[loop]); } @@ -268,7 +268,7 @@ void rxrpc_process_connection(struct work_struct *work) rxrpc_get_connection(conn); - if (test_and_clear_bit(RXRPC_CONN_CHALLENGE, &conn->events)) { + if (test_and_clear_bit(RXRPC_CONN_EV_CHALLENGE, &conn->events)) { rxrpc_secure_connection(conn); rxrpc_put_connection(conn); } diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index c2c0926..0e022df 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -399,9 +399,9 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, candidate->params.service_id = sp->hdr.serviceId; candidate->security_ix = sp->hdr.securityIndex; candidate->out_clientflag = 0; - candidate->state = RXRPC_CONN_SERVER; + candidate->state = RXRPC_CONN_SERVICE; if (candidate->params.service_id) - candidate->state = RXRPC_CONN_SERVER_UNSECURED; + candidate->state = RXRPC_CONN_SERVICE_UNSECURED; write_lock_bh(&peer->conn_lock); diff --git a/net/rxrpc/proc.c b/net/rxrpc/proc.c index 500cdcd..2a25ab42 100644 --- a/net/rxrpc/proc.c +++ b/net/rxrpc/proc.c @@ -14,15 +14,15 @@ #include #include "ar-internal.h" -static const char *const rxrpc_conn_states[] = { - [RXRPC_CONN_UNUSED] = "Unused ", - [RXRPC_CONN_CLIENT] = "Client ", - [RXRPC_CONN_SERVER_UNSECURED] = "SvUnsec ", - [RXRPC_CONN_SERVER_CHALLENGING] = "SvChall ", - [RXRPC_CONN_SERVER] = "SvSecure", - [RXRPC_CONN_REMOTELY_ABORTED] = "RmtAbort", - [RXRPC_CONN_LOCALLY_ABORTED] = "LocAbort", - [RXRPC_CONN_NETWORK_ERROR] = "NetError", +static const char *const rxrpc_conn_states[RXRPC_CONN__NR_STATES] = { + [RXRPC_CONN_UNUSED] = "Unused ", + [RXRPC_CONN_CLIENT] = "Client ", + [RXRPC_CONN_SERVICE_UNSECURED] = "SvUnsec ", + [RXRPC_CONN_SERVICE_CHALLENGING] = "SvChall ", + [RXRPC_CONN_SERVICE] = "SvSecure", + [RXRPC_CONN_REMOTELY_ABORTED] = "RmtAbort", + [RXRPC_CONN_LOCALLY_ABORTED] = "LocAbort", + [RXRPC_CONN_NETWORK_ERROR] = "NetError", }; /* -- cgit v0.10.2 From eb9b9d22754d1926771a22638e81384d517c6ce5 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 27 Jun 2016 10:32:02 +0100 Subject: rxrpc: Check that the client conns cache is empty before module removal Check that the client conns cache is empty before module removal and bug if not, listing any offending connections that are still present. Unfortunately, if there are connections still around, then the transport socket is still unexpectedly open and active, so we can't just unallocate the connections. Signed-off-by: David Howells diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 5d3e795..d5073eb 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -807,8 +807,7 @@ static void __exit af_rxrpc_exit(void) _debug("synchronise RCU"); rcu_barrier(); _debug("destroy locals"); - ASSERT(idr_is_empty(&rxrpc_client_conn_ids)); - idr_destroy(&rxrpc_client_conn_ids); + rxrpc_destroy_client_conn_ids(); rxrpc_destroy_all_locals(); remove_proc_entry("rxrpc_conns", init_net.proc_net); diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 3f0d047..6583a83 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -541,6 +541,7 @@ extern struct idr rxrpc_client_conn_ids; int rxrpc_get_client_connection_id(struct rxrpc_connection *, gfp_t); void rxrpc_put_client_connection_id(struct rxrpc_connection *); +void rxrpc_destroy_client_conn_ids(void); /* * conn_event.c diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c index 82488d6..be437d5 100644 --- a/net/rxrpc/conn_client.c +++ b/net/rxrpc/conn_client.c @@ -92,3 +92,22 @@ void rxrpc_put_client_connection_id(struct rxrpc_connection *conn) spin_unlock(&rxrpc_conn_id_lock); } } + +/* + * Destroy the client connection ID tree. + */ +void rxrpc_destroy_client_conn_ids(void) +{ + struct rxrpc_connection *conn; + int id; + + if (!idr_is_empty(&rxrpc_client_conn_ids)) { + idr_for_each_entry(&rxrpc_client_conn_ids, conn, id) { + pr_err("AF_RXRPC: Leaked client conn %p {%d}\n", + conn, atomic_read(&conn->usage)); + } + BUG(); + } + + idr_destroy(&rxrpc_client_conn_ids); +} -- cgit v0.10.2 From 2c4579e4b1d5a6219522c6e970500b2fd43fe1f8 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 27 Jun 2016 10:32:03 +0100 Subject: rxrpc: Move usage count getting into rxrpc_queue_conn() Rather than calling rxrpc_get_connection() manually before calling rxrpc_queue_conn(), do it inside the queue wrapper. This allows us to do some important fixes: (1) If the usage count is 0, do nothing. This prevents connections from being reanimated once they're dead. (2) If rxrpc_queue_work() fails because the work item is already queued, retract the usage count increment which would otherwise be lost. (3) Don't take a ref on the connection in the work function. By passing the ref through the work item, this is unnecessary. Doing it in the work function is too late anyway. Previously, connection-directed packets held a ref on the connection, but that's not really the best idea. And another useful changes: (*) Don't need to take a refcount on the connection in the data_ready handler unless we invoke the connection's work item. We're using RCU there so that's otherwise redundant. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 6583a83..9fc89cd 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -584,10 +584,17 @@ static inline void rxrpc_get_connection(struct rxrpc_connection *conn) atomic_inc(&conn->usage); } +static inline +struct rxrpc_connection *rxrpc_get_connection_maybe(struct rxrpc_connection *conn) +{ + return atomic_inc_not_zero(&conn->usage) ? conn : NULL; +} static inline void rxrpc_queue_conn(struct rxrpc_connection *conn) { - rxrpc_queue_work(&conn->processor); + if (rxrpc_get_connection_maybe(conn) && + !rxrpc_queue_work(&conn->processor)) + rxrpc_put_connection(conn); } /* diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index 1c0860d..5367dbe 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -132,7 +132,6 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, _debug("await conn sec"); list_add_tail(&call->accept_link, &rx->secureq); call->conn->state = RXRPC_CONN_SERVICE_CHALLENGING; - rxrpc_get_connection(call->conn); set_bit(RXRPC_CONN_EV_CHALLENGE, &call->conn->events); rxrpc_queue_conn(call->conn); } else { diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index b9c39b8..9ceddd3 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -266,12 +266,8 @@ void rxrpc_process_connection(struct work_struct *work) _enter("{%d}", conn->debug_id); - rxrpc_get_connection(conn); - - if (test_and_clear_bit(RXRPC_CONN_EV_CHALLENGE, &conn->events)) { + if (test_and_clear_bit(RXRPC_CONN_EV_CHALLENGE, &conn->events)) rxrpc_secure_connection(conn); - rxrpc_put_connection(conn); - } /* go through the conn-level event packets, releasing the ref on this * connection that each one has when we've finished with it */ @@ -286,7 +282,6 @@ void rxrpc_process_connection(struct work_struct *work) goto requeue_and_leave; case -ECONNABORTED: default: - rxrpc_put_connection(conn); rxrpc_free_skb(skb); break; } @@ -304,7 +299,6 @@ requeue_and_leave: protocol_error: if (rxrpc_abort_connection(conn, -ret, abort_code) < 0) goto requeue_and_leave; - rxrpc_put_connection(conn); rxrpc_free_skb(skb); _leave(" [EPROTO]"); goto out; diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index fe7ff33..b993f2d 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -580,7 +580,6 @@ static void rxrpc_post_packet_to_conn(struct rxrpc_connection *conn, { _enter("%p,%p", conn, skb); - rxrpc_get_connection(conn); skb_queue_tail(&conn->rx_queue, skb); rxrpc_queue_conn(conn); } -- cgit v0.10.2 From d1e858c5a392a50c16ce36624203032bdeb3595b Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:39 +0100 Subject: rxrpc: Fix handling of connection failure in client call creation If rxrpc_connect_call() fails during the creation of a client connection, there are two bugs that we can hit that need fixing: (1) The call state should be moved to RXRPC_CALL_DEAD before the call cleanup phase is invoked. If not, this can cause an assertion failure later. (2) call->link should be reinitialised after being deleted in rxrpc_new_client_call() - which otherwise leads to a failure later when the call cleanup attempts to delete the link again. Signed-off-by: David Howells diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index ad933da..6223a7e 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -425,9 +425,10 @@ error: rxrpc_put_call(call); write_lock_bh(&rxrpc_call_lock); - list_del(&call->link); + list_del_init(&call->link); write_unlock_bh(&rxrpc_call_lock); + call->state = RXRPC_CALL_DEAD; rxrpc_put_call(call); _leave(" = %d", ret); return ERR_PTR(ret); @@ -439,6 +440,7 @@ error: */ found_user_ID_now_present: write_unlock(&rx->call_lock); + call->state = RXRPC_CALL_DEAD; rxrpc_put_call(call); _leave(" = -EEXIST [%p]", call); return ERR_PTR(-EEXIST); -- cgit v0.10.2 From e653cfe49cec540529217933e07caf6c0f25ac93 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:38 +0100 Subject: rxrpc: Release a call's connection ref on call disconnection When a call is disconnected, clear the call's pointer to the connection and release the associated ref on that connection. This means that the call no longer pins the connection and the connection can be discarded even before the call is. As the code currently stands, the call struct is effectively pinned by userspace until userspace has enacted a recvmsg() to retrieve the final call state as sk_buffs on the receive queue pin the call to which they're related because: (1) The rxrpc_call struct contains the userspace ID that recvmsg() has to include in the control message buffer to indicate which call is being referred to. This ID must remain valid until the terminal packet is completely read and must be invalidated immediately at that point as userspace is entitled to immediately reuse it. (2) The final ACK to the reply to a client call isn't sent until the last data packet is entirely read (it's probably worth altering this in future to be send the ACK as soon as all the data has been received). This change requires a bit of rearrangement to make sure that the call isn't going to try and access the connection again after protocol completion: (1) Delete the error link earlier when we're releasing the call. Possibly network errors should be distributed via connections at the cost of adding in an access to the rxrpc_connection struct. (2) Remove the call from the connection's call tree before disconnecting the call. The call tree needs to be removed anyway and incoming packets delivered by channel pointer instead. (3) The release call event should be considered last after all other events have been processed so that we don't need access to the connection again. (4) Move the channel_lock taking from rxrpc_release_call() to rxrpc_disconnect_call() where it will be required in future. Signed-off-by: David Howells diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c index 0ba8429..638d66d 100644 --- a/net/rxrpc/call_event.c +++ b/net/rxrpc/call_event.c @@ -858,11 +858,6 @@ void rxrpc_process_call(struct work_struct *work) iov[0].iov_len = sizeof(whdr); /* deal with events of a final nature */ - if (test_bit(RXRPC_CALL_EV_RELEASE, &call->events)) { - rxrpc_release_call(call); - clear_bit(RXRPC_CALL_EV_RELEASE, &call->events); - } - if (test_bit(RXRPC_CALL_EV_RCVD_ERROR, &call->events)) { enum rxrpc_skb_mark mark; int error; @@ -1144,6 +1139,11 @@ void rxrpc_process_call(struct work_struct *work) goto maybe_reschedule; } + if (test_bit(RXRPC_CALL_EV_RELEASE, &call->events)) { + rxrpc_release_call(call); + clear_bit(RXRPC_CALL_EV_RELEASE, &call->events); + } + /* other events may have been raised since we started checking */ goto maybe_reschedule; diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 6223a7e..b43d89c 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -628,6 +628,10 @@ void rxrpc_release_call(struct rxrpc_call *call) */ _debug("RELEASE CALL %p (%d CONN %p)", call, call->debug_id, conn); + spin_lock(&conn->params.peer->lock); + hlist_del_init(&call->error_link); + spin_unlock(&conn->params.peer->lock); + write_lock_bh(&rx->call_lock); if (!list_empty(&call->accept_link)) { _debug("unlinking once-pending call %p { e=%lx f=%lx }", @@ -643,25 +647,22 @@ void rxrpc_release_call(struct rxrpc_call *call) write_unlock_bh(&rx->call_lock); /* free up the channel for reuse */ - spin_lock(&conn->channel_lock); write_lock_bh(&conn->lock); write_lock(&call->state_lock); - rxrpc_disconnect_call(call); - - spin_unlock(&conn->channel_lock); - if (call->state < RXRPC_CALL_COMPLETE && call->state != RXRPC_CALL_CLIENT_FINAL_ACK) { _debug("+++ ABORTING STATE %d +++\n", call->state); call->state = RXRPC_CALL_LOCALLY_ABORTED; call->local_abort = RX_CALL_DEAD; - set_bit(RXRPC_CALL_EV_ABORT, &call->events); - rxrpc_queue_call(call); } write_unlock(&call->state_lock); + + rb_erase(&call->conn_node, &conn->calls); write_unlock_bh(&conn->lock); + rxrpc_disconnect_call(call); + /* clean up the Rx queue */ if (!skb_queue_empty(&call->rx_queue) || !skb_queue_empty(&call->rx_oos_queue)) { @@ -817,16 +818,7 @@ static void rxrpc_cleanup_call(struct rxrpc_call *call) return; } - if (call->conn) { - spin_lock(&call->conn->params.peer->lock); - hlist_del_init(&call->error_link); - spin_unlock(&call->conn->params.peer->lock); - - write_lock_bh(&call->conn->lock); - rb_erase(&call->conn_node, &call->conn->calls); - write_unlock_bh(&call->conn->lock); - rxrpc_put_connection(call->conn); - } + ASSERTCMP(call->conn, ==, NULL); /* Remove the call from the hash */ rxrpc_call_hash_del(call); diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 0e022df..99d1810 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -540,11 +540,19 @@ void rxrpc_disconnect_call(struct rxrpc_call *call) _enter("%d,%d", conn->debug_id, call->channel); + spin_lock(&conn->channel_lock); + if (conn->channels[chan] == call) { rcu_assign_pointer(conn->channels[chan], NULL); atomic_inc(&conn->avail_chans); wake_up(&conn->channel_wq); } + + spin_unlock(&conn->channel_lock); + + call->conn = NULL; + rxrpc_put_connection(conn); + _leave(""); } /* -- cgit v0.10.2 From dee46364ce6fd0815ad9da625783eda21ccf7b06 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 27 Jun 2016 17:11:19 +0100 Subject: rxrpc: Add RCU destruction for connections and calls Add RCU destruction for connections and calls as the RCU lookup from the transport socket data_ready handler is going to come along shortly. Whilst we're at it, move the cleanup workqueue flushing and RCU barrierage into the destruction code for the objects that need it (locals and connections) and add the extra RCU barrier required for connection cleanup. Signed-off-by: David Howells diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index d5073eb..d6e4e3b 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -788,26 +788,7 @@ static void __exit af_rxrpc_exit(void) proto_unregister(&rxrpc_proto); rxrpc_destroy_all_calls(); rxrpc_destroy_all_connections(); - ASSERTCMP(atomic_read(&rxrpc_n_skbs), ==, 0); - - /* We need to flush the scheduled work twice because the local endpoint - * records involve a work item in their destruction as they can only be - * destroyed from process context. However, a connection may have a - * work item outstanding - and this will pin the local endpoint record - * until the connection goes away. - * - * Peers don't pin locals and calls pin sockets - which prevents the - * module from being unloaded - so we should only need two flushes. - */ - _debug("flush scheduled work"); - flush_workqueue(rxrpc_workqueue); - _debug("flush scheduled work 2"); - flush_workqueue(rxrpc_workqueue); - _debug("synchronise RCU"); - rcu_barrier(); - _debug("destroy locals"); - rxrpc_destroy_client_conn_ids(); rxrpc_destroy_all_locals(); remove_proc_entry("rxrpc_conns", init_net.proc_net); diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 9fc89cd..b401fa9 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -292,9 +292,10 @@ struct rxrpc_connection { struct rxrpc_conn_parameters params; spinlock_t channel_lock; - struct rxrpc_call *channels[RXRPC_MAXCALLS]; /* active calls */ + struct rxrpc_call __rcu *channels[RXRPC_MAXCALLS]; /* active calls */ wait_queue_head_t channel_wq; /* queue to wait for channel to become available */ + struct rcu_head rcu; struct work_struct processor; /* connection event processor */ union { struct rb_node client_node; /* Node in local->client_conns */ @@ -398,6 +399,7 @@ enum rxrpc_call_state { * - matched by { connection, call_id } */ struct rxrpc_call { + struct rcu_head rcu; struct rxrpc_connection *conn; /* connection carrying call */ struct rxrpc_sock *socket; /* socket responsible */ struct timer_list lifetimer; /* lifetime remaining on call */ diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index b43d89c..2c6c57c 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -480,7 +480,8 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, write_lock_bh(&conn->lock); /* set the channel for this call */ - call = conn->channels[candidate->channel]; + call = rcu_dereference_protected(conn->channels[candidate->channel], + lockdep_is_held(&conn->lock)); _debug("channel[%u] is %p", candidate->channel, call); if (call && call->call_id == sp->hdr.callNumber) { /* already set; must've been a duplicate packet */ @@ -544,7 +545,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, candidate = NULL; rb_link_node(&call->conn_node, parent, p); rb_insert_color(&call->conn_node, &conn->calls); - conn->channels[call->channel] = call; + rcu_assign_pointer(conn->channels[call->channel], call); sock_hold(&rx->sk); rxrpc_get_connection(conn); write_unlock_bh(&conn->lock); @@ -795,6 +796,17 @@ void __rxrpc_put_call(struct rxrpc_call *call) } /* + * Final call destruction under RCU. + */ +static void rxrpc_rcu_destroy_call(struct rcu_head *rcu) +{ + struct rxrpc_call *call = container_of(rcu, struct rxrpc_call, rcu); + + rxrpc_purge_queue(&call->rx_queue); + kmem_cache_free(rxrpc_call_jar, call); +} + +/* * clean up a call */ static void rxrpc_cleanup_call(struct rxrpc_call *call) @@ -849,7 +861,7 @@ static void rxrpc_cleanup_call(struct rxrpc_call *call) rxrpc_purge_queue(&call->rx_queue); ASSERT(skb_queue_empty(&call->rx_oos_queue)); sock_put(&call->socket->sk); - kmem_cache_free(rxrpc_call_jar, call); + call_rcu(&call->rcu, rxrpc_rcu_destroy_call); } /* diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index 9ceddd3..f6ca8c5 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -198,7 +198,10 @@ static int rxrpc_process_event(struct rxrpc_connection *conn, if (conn->state == RXRPC_CONN_SERVICE_CHALLENGING) { conn->state = RXRPC_CONN_SERVICE; for (loop = 0; loop < RXRPC_MAXCALLS; loop++) - rxrpc_call_is_secure(conn->channels[loop]); + rxrpc_call_is_secure( + rcu_dereference_protected( + conn->channels[loop], + lockdep_is_held(&conn->lock))); } spin_unlock(&conn->state_lock); diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 99d1810..0165a62 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -542,7 +542,7 @@ void rxrpc_disconnect_call(struct rxrpc_call *call) spin_lock(&conn->channel_lock); - if (conn->channels[chan] == call) { + if (rcu_access_pointer(conn->channels[chan]) == call) { rcu_assign_pointer(conn->channels[chan], NULL); atomic_inc(&conn->avail_chans); wake_up(&conn->channel_wq); @@ -580,9 +580,12 @@ void rxrpc_put_connection(struct rxrpc_connection *conn) /* * destroy a virtual connection */ -static void rxrpc_destroy_connection(struct rxrpc_connection *conn) +static void rxrpc_destroy_connection(struct rcu_head *rcu) { - _enter("%p{%d}", conn, atomic_read(&conn->usage)); + struct rxrpc_connection *conn = + container_of(rcu, struct rxrpc_connection, rcu); + + _enter("{%d,u=%d}", conn->debug_id, atomic_read(&conn->usage)); ASSERTCMP(atomic_read(&conn->usage), ==, 0); @@ -677,7 +680,8 @@ static void rxrpc_connection_reaper(struct work_struct *work) list_del_init(&conn->link); ASSERTCMP(atomic_read(&conn->usage), ==, 0); - rxrpc_destroy_connection(conn); + skb_queue_purge(&conn->rx_queue); + call_rcu(&conn->rcu, rxrpc_destroy_connection); } _leave(""); @@ -689,11 +693,30 @@ static void rxrpc_connection_reaper(struct work_struct *work) */ void __exit rxrpc_destroy_all_connections(void) { + struct rxrpc_connection *conn, *_p; + bool leak = false; + _enter(""); rxrpc_connection_expiry = 0; cancel_delayed_work(&rxrpc_connection_reap); rxrpc_queue_delayed_work(&rxrpc_connection_reap, 0); + flush_workqueue(rxrpc_workqueue); + + write_lock(&rxrpc_connection_lock); + list_for_each_entry_safe(conn, _p, &rxrpc_connections, link) { + pr_err("AF_RXRPC: Leaked conn %p {%d}\n", + conn, atomic_read(&conn->usage)); + leak = true; + } + write_unlock(&rxrpc_connection_lock); + BUG_ON(leak); + + /* Make sure the local and peer records pinned by any dying connections + * are released. + */ + rcu_barrier(); + rxrpc_destroy_client_conn_ids(); _leave(""); } diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c index 3ab7764..a753796 100644 --- a/net/rxrpc/local_object.c +++ b/net/rxrpc/local_object.c @@ -374,14 +374,17 @@ void __exit rxrpc_destroy_all_locals(void) _enter(""); - if (list_empty(&rxrpc_local_endpoints)) - return; + flush_workqueue(rxrpc_workqueue); - mutex_lock(&rxrpc_local_mutex); - list_for_each_entry(local, &rxrpc_local_endpoints, link) { - pr_err("AF_RXRPC: Leaked local %p {%d}\n", - local, atomic_read(&local->usage)); + if (!list_empty(&rxrpc_local_endpoints)) { + mutex_lock(&rxrpc_local_mutex); + list_for_each_entry(local, &rxrpc_local_endpoints, link) { + pr_err("AF_RXRPC: Leaked local %p {%d}\n", + local, atomic_read(&local->usage)); + } + mutex_unlock(&rxrpc_local_mutex); + BUG(); } - mutex_unlock(&rxrpc_local_mutex); - BUG(); + + rcu_barrier(); } -- cgit v0.10.2 From 30b515f4d1cf31f6901c1fa61d920f651ebc07d7 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 28 Jun 2016 16:58:36 +0100 Subject: rxrpc: Access socket accept queue under right lock The socket's accept queue (socket->acceptq) should be accessed under socket->call_lock, not under the connection lock. Signed-off-by: David Howells diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c index 638d66d..fc32aa5 100644 --- a/net/rxrpc/call_event.c +++ b/net/rxrpc/call_event.c @@ -1089,7 +1089,7 @@ void rxrpc_process_call(struct work_struct *work) if (call->state == RXRPC_CALL_SERVER_SECURING) { _debug("securing"); - write_lock(&call->conn->lock); + write_lock(&call->socket->call_lock); if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) && !test_bit(RXRPC_CALL_EV_RELEASE, &call->events)) { _debug("not released"); @@ -1097,7 +1097,7 @@ void rxrpc_process_call(struct work_struct *work) list_move_tail(&call->accept_link, &call->socket->acceptq); } - write_unlock(&call->conn->lock); + write_unlock(&call->socket->call_lock); read_lock(&call->state_lock); if (call->state < RXRPC_CALL_COMPLETE) set_bit(RXRPC_CALL_EV_POST_ACCEPT, &call->events); -- cgit v0.10.2 From a1399f8bb0331a1f50c76c4cac738fe57679b9bb Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 27 Jun 2016 14:39:44 +0100 Subject: rxrpc: Call channels should have separate call number spaces Each channel on a connection has a separate, independent number space from which to allocate callNumber values. It is entirely possible, for example, to have a connection with four active calls, each with call number 1. Note that the callNumber values for any particular channel don't have to start at 1, but they are supposed to increment monotonically for that channel from a client's perspective and may not be reused once the call number is transmitted (until the epoch cycles all the way back round). Currently, however, call numbers are allocated on a per-connection basis and, further, are held in an rb-tree. The rb-tree is redundant as the four channel pointers in the rxrpc_connection struct are entirely capable of pointing to all the calls currently in progress on a connection. To this end, make the following changes: (1) Handle call number allocation independently per channel. (2) Get rid of the conn->calls rb-tree. This is overkill as a connection may have a maximum of four calls in progress at any one time. Use the pointers in the channels[] array instead, indexed by the channel number from the packet. (3) For each channel, save the result of the last call that was in progress on that channel in conn->channels[] so that the final ACK or ABORT packet can be replayed if necessary. Any call earlier than that is just ignored. If we've seen the next call number in a packet, the last one is most definitely defunct. (4) When generating a RESPONSE packet for a connection, the call number counter for each channel must be included in it. (5) When parsing a RESPONSE packet for a connection, the call number counters contained therein should be used to set the minimum expected call numbers on each channel. To do in future commits: (1) Replay terminal packets based on the last call stored in conn->channels[]. (2) Connections should be retired before the callNumber space on any channel runs out. (3) A server is expected to disregard or reject any new incoming call that has a call number less than the current call number counter. The call number counter for that channel must be advanced to the new call number. Note that the server cannot just require that the next call that it sees on a channel be exactly the call number counter + 1 because then there's a scenario that could cause a problem: The client transmits a packet to initiate a connection, the network goes out, the server sends an ACK (which gets lost), the client sends an ABORT (which also gets lost); the network then reconnects, the client then reuses the call number for the next call (it doesn't know the server already saw the call number), but the server thinks it already has the first packet of this call (it doesn't know that the client doesn't know that it saw the call number the first time). Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index b401fa9..b697654 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -292,7 +292,14 @@ struct rxrpc_connection { struct rxrpc_conn_parameters params; spinlock_t channel_lock; - struct rxrpc_call __rcu *channels[RXRPC_MAXCALLS]; /* active calls */ + + struct rxrpc_channel { + struct rxrpc_call __rcu *call; /* Active call */ + u32 call_id; /* ID of current call */ + u32 call_counter; /* Call ID counter */ + u32 last_call; /* ID of last call */ + u32 last_result; /* Result of last call (0/abort) */ + } channels[RXRPC_MAXCALLS]; wait_queue_head_t channel_wq; /* queue to wait for channel to become available */ struct rcu_head rcu; @@ -302,7 +309,6 @@ struct rxrpc_connection { struct rb_node service_node; /* Node in peer->service_conns */ }; struct list_head link; /* link in master connection list */ - struct rb_root calls; /* calls on this connection */ struct sk_buff_head rx_queue; /* received conn-level packets */ const struct rxrpc_security *security; /* applied security module */ struct key *server_key; /* security for this service */ @@ -311,7 +317,6 @@ struct rxrpc_connection { unsigned long flags; unsigned long events; unsigned long put_time; /* Time at which last put */ - rwlock_t lock; /* access lock */ spinlock_t state_lock; /* state-change lock */ atomic_t usage; enum rxrpc_conn_proto_state state : 8; /* current state of connection */ @@ -319,7 +324,6 @@ struct rxrpc_connection { u32 remote_abort; /* remote abort code */ int error; /* local error incurred */ int debug_id; /* debug ID for printks */ - unsigned int call_counter; /* call ID counter */ atomic_t serial; /* packet serial number counter */ atomic_t hi_serial; /* highest serial number received */ atomic_t avail_chans; /* number of channels available */ @@ -412,7 +416,6 @@ struct rxrpc_call { struct hlist_node error_link; /* link in error distribution list */ struct list_head accept_link; /* calls awaiting acceptance */ struct rb_node sock_node; /* node in socket call tree */ - struct rb_node conn_node; /* node in connection call tree */ struct sk_buff_head rx_queue; /* received packets */ struct sk_buff_head rx_oos_queue; /* packets received out of sequence */ struct sk_buff *tx_pending; /* Tx socket buffer being filled */ @@ -564,6 +567,7 @@ int rxrpc_connect_call(struct rxrpc_call *, struct rxrpc_conn_parameters *, struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *, struct rxrpc_peer *, struct sk_buff *); +void __rxrpc_disconnect_call(struct rxrpc_call *); void rxrpc_disconnect_call(struct rxrpc_call *); void rxrpc_put_connection(struct rxrpc_connection *); void __exit rxrpc_destroy_all_connections(void); diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 2c6c57c..3f27872 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -456,8 +456,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, { struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rxrpc_call *call, *candidate; - struct rb_node **p, *parent; - u32 call_id; + u32 call_id, chan; _enter(",%d", conn->debug_id); @@ -467,21 +466,23 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, if (!candidate) return ERR_PTR(-EBUSY); + chan = sp->hdr.cid & RXRPC_CHANNELMASK; candidate->socket = rx; candidate->conn = conn; candidate->cid = sp->hdr.cid; candidate->call_id = sp->hdr.callNumber; - candidate->channel = sp->hdr.cid & RXRPC_CHANNELMASK; + candidate->channel = chan; candidate->rx_data_post = 0; candidate->state = RXRPC_CALL_SERVER_ACCEPTING; if (conn->security_ix > 0) candidate->state = RXRPC_CALL_SERVER_SECURING; - write_lock_bh(&conn->lock); + spin_lock(&conn->channel_lock); /* set the channel for this call */ - call = rcu_dereference_protected(conn->channels[candidate->channel], - lockdep_is_held(&conn->lock)); + call = rcu_dereference_protected(conn->channels[chan].call, + lockdep_is_held(&conn->channel_lock)); + _debug("channel[%u] is %p", candidate->channel, call); if (call && call->call_id == sp->hdr.callNumber) { /* already set; must've been a duplicate packet */ @@ -510,9 +511,9 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, call->debug_id, rxrpc_call_states[call->state]); if (call->state >= RXRPC_CALL_COMPLETE) { - conn->channels[call->channel] = NULL; + __rxrpc_disconnect_call(call); } else { - write_unlock_bh(&conn->lock); + spin_unlock(&conn->channel_lock); kmem_cache_free(rxrpc_call_jar, candidate); _leave(" = -EBUSY"); return ERR_PTR(-EBUSY); @@ -522,33 +523,22 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, /* check the call number isn't duplicate */ _debug("check dup"); call_id = sp->hdr.callNumber; - p = &conn->calls.rb_node; - parent = NULL; - while (*p) { - parent = *p; - call = rb_entry(parent, struct rxrpc_call, conn_node); - - /* The tree is sorted in order of the __be32 value without - * turning it into host order. - */ - if (call_id < call->call_id) - p = &(*p)->rb_left; - else if (call_id > call->call_id) - p = &(*p)->rb_right; - else - goto old_call; - } + + /* We just ignore calls prior to the current call ID. Terminated calls + * are handled via the connection. + */ + if (call_id <= conn->channels[chan].call_counter) + goto old_call; /* TODO: Just drop packet */ /* make the call available */ _debug("new call"); call = candidate; candidate = NULL; - rb_link_node(&call->conn_node, parent, p); - rb_insert_color(&call->conn_node, &conn->calls); - rcu_assign_pointer(conn->channels[call->channel], call); + conn->channels[chan].call_counter = call_id; + rcu_assign_pointer(conn->channels[chan].call, call); sock_hold(&rx->sk); rxrpc_get_connection(conn); - write_unlock_bh(&conn->lock); + spin_unlock(&conn->channel_lock); spin_lock(&conn->params.peer->lock); hlist_add_head(&call->error_link, &conn->params.peer->error_targets); @@ -588,19 +578,19 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, return call; extant_call: - write_unlock_bh(&conn->lock); + spin_unlock(&conn->channel_lock); kmem_cache_free(rxrpc_call_jar, candidate); _leave(" = %p {%d} [extant]", call, call ? call->debug_id : -1); return call; aborted_call: - write_unlock_bh(&conn->lock); + spin_unlock(&conn->channel_lock); kmem_cache_free(rxrpc_call_jar, candidate); _leave(" = -ECONNABORTED"); return ERR_PTR(-ECONNABORTED); old_call: - write_unlock_bh(&conn->lock); + spin_unlock(&conn->channel_lock); kmem_cache_free(rxrpc_call_jar, candidate); _leave(" = -ECONNRESET [old]"); return ERR_PTR(-ECONNRESET); @@ -648,8 +638,7 @@ void rxrpc_release_call(struct rxrpc_call *call) write_unlock_bh(&rx->call_lock); /* free up the channel for reuse */ - write_lock_bh(&conn->lock); - write_lock(&call->state_lock); + write_lock_bh(&call->state_lock); if (call->state < RXRPC_CALL_COMPLETE && call->state != RXRPC_CALL_CLIENT_FINAL_ACK) { @@ -657,10 +646,7 @@ void rxrpc_release_call(struct rxrpc_call *call) call->state = RXRPC_CALL_LOCALLY_ABORTED; call->local_abort = RX_CALL_DEAD; } - write_unlock(&call->state_lock); - - rb_erase(&call->conn_node, &conn->calls); - write_unlock_bh(&conn->lock); + write_unlock_bh(&call->state_lock); rxrpc_disconnect_call(call); diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index f6ca8c5..cee0f35 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -31,15 +31,17 @@ static void rxrpc_abort_calls(struct rxrpc_connection *conn, int state, u32 abort_code) { struct rxrpc_call *call; - struct rb_node *p; + int i; _enter("{%d},%x", conn->debug_id, abort_code); - read_lock_bh(&conn->lock); + spin_lock(&conn->channel_lock); - for (p = rb_first(&conn->calls); p; p = rb_next(p)) { - call = rb_entry(p, struct rxrpc_call, conn_node); - write_lock(&call->state_lock); + for (i = 0; i < RXRPC_MAXCALLS; i++) { + call = rcu_dereference_protected( + conn->channels[i].call, + lockdep_is_held(&conn->channel_lock)); + write_lock_bh(&call->state_lock); if (call->state <= RXRPC_CALL_COMPLETE) { call->state = state; if (state == RXRPC_CALL_LOCALLY_ABORTED) { @@ -51,10 +53,10 @@ static void rxrpc_abort_calls(struct rxrpc_connection *conn, int state, } rxrpc_queue_call(call); } - write_unlock(&call->state_lock); + write_unlock_bh(&call->state_lock); } - read_unlock_bh(&conn->lock); + spin_unlock(&conn->channel_lock); _leave(""); } @@ -192,7 +194,7 @@ static int rxrpc_process_event(struct rxrpc_connection *conn, if (ret < 0) return ret; - read_lock_bh(&conn->lock); + spin_lock(&conn->channel_lock); spin_lock(&conn->state_lock); if (conn->state == RXRPC_CONN_SERVICE_CHALLENGING) { @@ -200,12 +202,12 @@ static int rxrpc_process_event(struct rxrpc_connection *conn, for (loop = 0; loop < RXRPC_MAXCALLS; loop++) rxrpc_call_is_secure( rcu_dereference_protected( - conn->channels[loop], - lockdep_is_held(&conn->lock))); + conn->channels[loop].call, + lockdep_is_held(&conn->channel_lock))); } spin_unlock(&conn->state_lock); - read_unlock_bh(&conn->lock); + spin_unlock(&conn->channel_lock); return 0; default: diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 0165a62..ce83f3e 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -46,10 +46,8 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) init_waitqueue_head(&conn->channel_wq); INIT_WORK(&conn->processor, &rxrpc_process_connection); INIT_LIST_HEAD(&conn->link); - conn->calls = RB_ROOT; skb_queue_head_init(&conn->rx_queue); conn->security = &rxrpc_no_security; - rwlock_init(&conn->lock); spin_lock_init(&conn->state_lock); atomic_set(&conn->usage, 1); conn->debug_id = atomic_inc_return(&rxrpc_debug_id); @@ -63,39 +61,6 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) } /* - * add a call to a connection's call-by-ID tree - */ -static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn, - struct rxrpc_call *call) -{ - struct rxrpc_call *xcall; - struct rb_node *parent, **p; - u32 call_id; - - write_lock_bh(&conn->lock); - - call_id = call->call_id; - p = &conn->calls.rb_node; - parent = NULL; - while (*p) { - parent = *p; - xcall = rb_entry(parent, struct rxrpc_call, conn_node); - - if (call_id < xcall->call_id) - p = &(*p)->rb_left; - else if (call_id > xcall->call_id) - p = &(*p)->rb_right; - else - BUG(); - } - - rb_link_node(&call->conn_node, parent, p); - rb_insert_color(&call->conn_node, &conn->calls); - - write_unlock_bh(&conn->lock); -} - -/* * Allocate a client connection. The caller must take care to clear any * padding bytes in *cp. */ @@ -277,12 +242,12 @@ found_channel: call->channel = chan; call->epoch = conn->proto.epoch; call->cid = conn->proto.cid | chan; - call->call_id = ++conn->call_counter; - rcu_assign_pointer(conn->channels[chan], call); + call->call_id = ++conn->channels[chan].call_counter; + conn->channels[chan].call_id = call->call_id; + rcu_assign_pointer(conn->channels[chan].call, call); _net("CONNECT call %d on conn %d", call->debug_id, conn->debug_id); - rxrpc_add_call_ID_to_conn(conn, call); spin_unlock(&conn->channel_lock); rxrpc_put_peer(cp->peer); cp->peer = NULL; @@ -326,7 +291,7 @@ found_extant_conn: spin_lock(&conn->channel_lock); for (chan = 0; chan < RXRPC_MAXCALLS; chan++) - if (!conn->channels[chan]) + if (!conn->channels[chan].call) goto found_channel; BUG(); @@ -531,28 +496,47 @@ found: /* * Disconnect a call and clear any channel it occupies when that call - * terminates. + * terminates. The caller must hold the channel_lock and must release the + * call's ref on the connection. */ -void rxrpc_disconnect_call(struct rxrpc_call *call) +void __rxrpc_disconnect_call(struct rxrpc_call *call) { struct rxrpc_connection *conn = call->conn; - unsigned chan = call->channel; + struct rxrpc_channel *chan = &conn->channels[call->channel]; _enter("%d,%d", conn->debug_id, call->channel); - spin_lock(&conn->channel_lock); + if (rcu_access_pointer(chan->call) == call) { + /* Save the result of the call so that we can repeat it if necessary + * through the channel, whilst disposing of the actual call record. + */ + chan->last_result = call->local_abort; + smp_wmb(); + chan->last_call = chan->call_id; + chan->call_id = chan->call_counter; - if (rcu_access_pointer(conn->channels[chan]) == call) { - rcu_assign_pointer(conn->channels[chan], NULL); + rcu_assign_pointer(chan->call, NULL); atomic_inc(&conn->avail_chans); wake_up(&conn->channel_wq); } + _leave(""); +} + +/* + * Disconnect a call and clear any channel it occupies when that call + * terminates. + */ +void rxrpc_disconnect_call(struct rxrpc_call *call) +{ + struct rxrpc_connection *conn = call->conn; + + spin_lock(&conn->channel_lock); + __rxrpc_disconnect_call(call); spin_unlock(&conn->channel_lock); call->conn = NULL; rxrpc_put_connection(conn); - _leave(""); } /* @@ -591,7 +575,6 @@ static void rxrpc_destroy_connection(struct rcu_head *rcu) _net("DESTROY CONN %d", conn->debug_id); - ASSERT(RB_EMPTY_ROOT(&conn->calls)); rxrpc_purge_queue(&conn->rx_queue); conn->security->clear(conn); diff --git a/net/rxrpc/proc.c b/net/rxrpc/proc.c index 2a25ab42..ced5f07 100644 --- a/net/rxrpc/proc.c +++ b/net/rxrpc/proc.c @@ -137,7 +137,7 @@ static int rxrpc_connection_seq_show(struct seq_file *seq, void *v) if (v == &rxrpc_connections) { seq_puts(seq, "Proto Local Remote " - " SvID ConnID Calls End Use State Key " + " SvID ConnID End Use State Key " " Serial ISerial\n" ); return 0; @@ -154,13 +154,12 @@ static int rxrpc_connection_seq_show(struct seq_file *seq, void *v) ntohs(conn->params.peer->srx.transport.sin.sin_port)); seq_printf(seq, - "UDP %-22.22s %-22.22s %4x %08x %08x %s %3u" + "UDP %-22.22s %-22.22s %4x %08x %s %3u" " %s %08x %08x %08x\n", lbuff, rbuff, conn->params.service_id, conn->proto.cid, - conn->call_counter, rxrpc_conn_is_service(conn) ? "Svc" : "Clt", atomic_read(&conn->usage), rxrpc_conn_states[conn->state], diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index 3acc7c1..63afa9e 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -767,14 +767,10 @@ static int rxkad_respond_to_challenge(struct rxrpc_connection *conn, resp.kvno = htonl(token->kad->kvno); resp.ticket_len = htonl(token->kad->ticket_len); - resp.encrypted.call_id[0] = - htonl(conn->channels[0] ? conn->channels[0]->call_id : 0); - resp.encrypted.call_id[1] = - htonl(conn->channels[1] ? conn->channels[1]->call_id : 0); - resp.encrypted.call_id[2] = - htonl(conn->channels[2] ? conn->channels[2]->call_id : 0); - resp.encrypted.call_id[3] = - htonl(conn->channels[3] ? conn->channels[3]->call_id : 0); + resp.encrypted.call_id[0] = htonl(conn->channels[0].call_counter); + resp.encrypted.call_id[1] = htonl(conn->channels[1].call_counter); + resp.encrypted.call_id[2] = htonl(conn->channels[2].call_counter); + resp.encrypted.call_id[3] = htonl(conn->channels[3].call_counter); /* calculate the response checksum and then do the encryption */ rxkad_calc_response_checksum(&resp); @@ -991,7 +987,7 @@ static int rxkad_verify_response(struct rxrpc_connection *conn, void *ticket; u32 abort_code, version, kvno, ticket_len, level; __be32 csum; - int ret; + int ret, i; _enter("{%d,%x}", conn->debug_id, key_serial(conn->server_key)); @@ -1054,11 +1050,26 @@ static int rxkad_verify_response(struct rxrpc_connection *conn, if (response.encrypted.checksum != csum) goto protocol_error_free; - if (ntohl(response.encrypted.call_id[0]) > INT_MAX || - ntohl(response.encrypted.call_id[1]) > INT_MAX || - ntohl(response.encrypted.call_id[2]) > INT_MAX || - ntohl(response.encrypted.call_id[3]) > INT_MAX) - goto protocol_error_free; + spin_lock(&conn->channel_lock); + for (i = 0; i < RXRPC_MAXCALLS; i++) { + struct rxrpc_call *call; + u32 call_id = ntohl(response.encrypted.call_id[i]); + + if (call_id > INT_MAX) + goto protocol_error_unlock; + + if (call_id < conn->channels[i].call_counter) + goto protocol_error_unlock; + if (call_id > conn->channels[i].call_counter) { + call = rcu_dereference_protected( + conn->channels[i].call, + lockdep_is_held(&conn->channel_lock)); + if (call && call->state < RXRPC_CALL_COMPLETE) + goto protocol_error_unlock; + conn->channels[i].call_counter = call_id; + } + } + spin_unlock(&conn->channel_lock); abort_code = RXKADOUTOFSEQUENCE; if (ntohl(response.encrypted.inc_nonce) != conn->security_nonce + 1) @@ -1083,6 +1094,8 @@ static int rxkad_verify_response(struct rxrpc_connection *conn, _leave(" = 0"); return 0; +protocol_error_unlock: + spin_unlock(&conn->channel_lock); protocol_error_free: kfree(ticket); protocol_error: -- cgit v0.10.2 From c6d2b8d764f5edd79f708bdc49d1176072ee77a1 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:40 +0100 Subject: rxrpc: Split client connection code out into its own file Split the client-specific connection code out into its own file. It will behave somewhat differently from the service-specific connection code, so it makes sense to separate them. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index b697654..021d28b 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -544,9 +544,10 @@ void __exit rxrpc_destroy_all_calls(void); */ extern struct idr rxrpc_client_conn_ids; -int rxrpc_get_client_connection_id(struct rxrpc_connection *, gfp_t); void rxrpc_put_client_connection_id(struct rxrpc_connection *); void rxrpc_destroy_client_conn_ids(void); +int rxrpc_connect_call(struct rxrpc_call *, struct rxrpc_conn_parameters *, + struct sockaddr_rxrpc *, gfp_t); /* * conn_event.c @@ -562,8 +563,7 @@ extern unsigned int rxrpc_connection_expiry; extern struct list_head rxrpc_connections; extern rwlock_t rxrpc_connection_lock; -int rxrpc_connect_call(struct rxrpc_call *, struct rxrpc_conn_parameters *, - struct sockaddr_rxrpc *, gfp_t); +struct rxrpc_connection *rxrpc_alloc_connection(gfp_t); struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *, struct rxrpc_peer *, struct sk_buff *); diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c index be437d5..9180164 100644 --- a/net/rxrpc/conn_client.c +++ b/net/rxrpc/conn_client.c @@ -33,7 +33,8 @@ static DEFINE_SPINLOCK(rxrpc_conn_id_lock); * client conns away from the current allocation point to try and keep the IDs * concentrated. We will also need to retire connections from an old epoch. */ -int rxrpc_get_client_connection_id(struct rxrpc_connection *conn, gfp_t gfp) +static int rxrpc_get_client_connection_id(struct rxrpc_connection *conn, + gfp_t gfp) { u32 epoch; int id; @@ -111,3 +112,248 @@ void rxrpc_destroy_client_conn_ids(void) idr_destroy(&rxrpc_client_conn_ids); } + +/* + * Allocate a client connection. The caller must take care to clear any + * padding bytes in *cp. + */ +static struct rxrpc_connection * +rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, gfp_t gfp) +{ + struct rxrpc_connection *conn; + int ret; + + _enter(""); + + conn = rxrpc_alloc_connection(gfp); + if (!conn) { + _leave(" = -ENOMEM"); + return ERR_PTR(-ENOMEM); + } + + conn->params = *cp; + conn->proto.local = cp->local; + conn->proto.epoch = rxrpc_epoch; + conn->proto.cid = 0; + conn->proto.in_clientflag = 0; + conn->proto.family = cp->peer->srx.transport.family; + conn->out_clientflag = RXRPC_CLIENT_INITIATED; + conn->state = RXRPC_CONN_CLIENT; + + switch (conn->proto.family) { + case AF_INET: + conn->proto.addr_size = sizeof(conn->proto.ipv4_addr); + conn->proto.ipv4_addr = cp->peer->srx.transport.sin.sin_addr; + conn->proto.port = cp->peer->srx.transport.sin.sin_port; + break; + } + + ret = rxrpc_get_client_connection_id(conn, gfp); + if (ret < 0) + goto error_0; + + ret = rxrpc_init_client_conn_security(conn); + if (ret < 0) + goto error_1; + + ret = conn->security->prime_packet_security(conn); + if (ret < 0) + goto error_2; + + write_lock(&rxrpc_connection_lock); + list_add_tail(&conn->link, &rxrpc_connections); + write_unlock(&rxrpc_connection_lock); + + /* We steal the caller's peer ref. */ + cp->peer = NULL; + rxrpc_get_local(conn->params.local); + key_get(conn->params.key); + + _leave(" = %p", conn); + return conn; + +error_2: + conn->security->clear(conn); +error_1: + rxrpc_put_client_connection_id(conn); +error_0: + kfree(conn); + _leave(" = %d", ret); + return ERR_PTR(ret); +} + +/* + * find a connection for a call + * - called in process context with IRQs enabled + */ +int rxrpc_connect_call(struct rxrpc_call *call, + struct rxrpc_conn_parameters *cp, + struct sockaddr_rxrpc *srx, + gfp_t gfp) +{ + struct rxrpc_connection *conn, *candidate = NULL; + struct rxrpc_local *local = cp->local; + struct rb_node *p, **pp, *parent; + long diff; + int chan; + + DECLARE_WAITQUEUE(myself, current); + + _enter("{%d,%lx},", call->debug_id, call->user_call_ID); + + cp->peer = rxrpc_lookup_peer(cp->local, srx, gfp); + if (!cp->peer) + return -ENOMEM; + + if (!cp->exclusive) { + /* Search for a existing client connection unless this is going + * to be a connection that's used exclusively for a single call. + */ + _debug("search 1"); + spin_lock(&local->client_conns_lock); + p = local->client_conns.rb_node; + while (p) { + conn = rb_entry(p, struct rxrpc_connection, client_node); + +#define cmp(X) ((long)conn->params.X - (long)cp->X) + diff = (cmp(peer) ?: + cmp(key) ?: + cmp(security_level)); + if (diff < 0) + p = p->rb_left; + else if (diff > 0) + p = p->rb_right; + else + goto found_extant_conn; + } + spin_unlock(&local->client_conns_lock); + } + + /* We didn't find a connection or we want an exclusive one. */ + _debug("get new conn"); + candidate = rxrpc_alloc_client_connection(cp, gfp); + if (!candidate) { + _leave(" = -ENOMEM"); + return -ENOMEM; + } + + if (cp->exclusive) { + /* Assign the call on an exclusive connection to channel 0 and + * don't add the connection to the endpoint's shareable conn + * lookup tree. + */ + _debug("exclusive chan 0"); + conn = candidate; + atomic_set(&conn->avail_chans, RXRPC_MAXCALLS - 1); + spin_lock(&conn->channel_lock); + chan = 0; + goto found_channel; + } + + /* We need to redo the search before attempting to add a new connection + * lest we race with someone else adding a conflicting instance. + */ + _debug("search 2"); + spin_lock(&local->client_conns_lock); + + pp = &local->client_conns.rb_node; + parent = NULL; + while (*pp) { + parent = *pp; + conn = rb_entry(parent, struct rxrpc_connection, client_node); + + diff = (cmp(peer) ?: + cmp(key) ?: + cmp(security_level)); + if (diff < 0) + pp = &(*pp)->rb_left; + else if (diff > 0) + pp = &(*pp)->rb_right; + else + goto found_extant_conn; + } + + /* The second search also failed; simply add the new connection with + * the new call in channel 0. Note that we need to take the channel + * lock before dropping the client conn lock. + */ + _debug("new conn"); + conn = candidate; + candidate = NULL; + + rb_link_node(&conn->client_node, parent, pp); + rb_insert_color(&conn->client_node, &local->client_conns); + + atomic_set(&conn->avail_chans, RXRPC_MAXCALLS - 1); + spin_lock(&conn->channel_lock); + spin_unlock(&local->client_conns_lock); + chan = 0; + +found_channel: + _debug("found chan"); + call->conn = conn; + call->channel = chan; + call->epoch = conn->proto.epoch; + call->cid = conn->proto.cid | chan; + call->call_id = ++conn->channels[chan].call_counter; + conn->channels[chan].call_id = call->call_id; + rcu_assign_pointer(conn->channels[chan].call, call); + + _net("CONNECT call %d on conn %d", call->debug_id, conn->debug_id); + + spin_unlock(&conn->channel_lock); + rxrpc_put_peer(cp->peer); + cp->peer = NULL; + _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); + return 0; + + /* We found a suitable connection already in existence. Discard any + * candidate we may have allocated, and try to get a channel on this + * one. + */ +found_extant_conn: + _debug("found conn"); + rxrpc_get_connection(conn); + spin_unlock(&local->client_conns_lock); + + rxrpc_put_connection(candidate); + + if (!atomic_add_unless(&conn->avail_chans, -1, 0)) { + if (!gfpflags_allow_blocking(gfp)) { + rxrpc_put_connection(conn); + _leave(" = -EAGAIN"); + return -EAGAIN; + } + + add_wait_queue(&conn->channel_wq, &myself); + for (;;) { + set_current_state(TASK_INTERRUPTIBLE); + if (atomic_add_unless(&conn->avail_chans, -1, 0)) + break; + if (signal_pending(current)) + goto interrupted; + schedule(); + } + remove_wait_queue(&conn->channel_wq, &myself); + __set_current_state(TASK_RUNNING); + } + + /* The connection allegedly now has a free channel and we can now + * attach the call to it. + */ + spin_lock(&conn->channel_lock); + + for (chan = 0; chan < RXRPC_MAXCALLS; chan++) + if (!conn->channels[chan].call) + goto found_channel; + BUG(); + +interrupted: + remove_wait_queue(&conn->channel_wq, &myself); + __set_current_state(TASK_RUNNING); + rxrpc_put_connection(conn); + rxrpc_put_peer(cp->peer); + cp->peer = NULL; + _leave(" = -ERESTARTSYS"); + return -ERESTARTSYS; +} diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index ce83f3e..ab5c8c2 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -34,7 +34,7 @@ static DECLARE_DELAYED_WORK(rxrpc_connection_reap, rxrpc_connection_reaper); /* * allocate a new connection */ -static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) +struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) { struct rxrpc_connection *conn; @@ -61,251 +61,6 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) } /* - * Allocate a client connection. The caller must take care to clear any - * padding bytes in *cp. - */ -static struct rxrpc_connection * -rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, gfp_t gfp) -{ - struct rxrpc_connection *conn; - int ret; - - _enter(""); - - conn = rxrpc_alloc_connection(gfp); - if (!conn) { - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } - - conn->params = *cp; - conn->proto.local = cp->local; - conn->proto.epoch = rxrpc_epoch; - conn->proto.cid = 0; - conn->proto.in_clientflag = 0; - conn->proto.family = cp->peer->srx.transport.family; - conn->out_clientflag = RXRPC_CLIENT_INITIATED; - conn->state = RXRPC_CONN_CLIENT; - - switch (conn->proto.family) { - case AF_INET: - conn->proto.addr_size = sizeof(conn->proto.ipv4_addr); - conn->proto.ipv4_addr = cp->peer->srx.transport.sin.sin_addr; - conn->proto.port = cp->peer->srx.transport.sin.sin_port; - break; - } - - ret = rxrpc_get_client_connection_id(conn, gfp); - if (ret < 0) - goto error_0; - - ret = rxrpc_init_client_conn_security(conn); - if (ret < 0) - goto error_1; - - ret = conn->security->prime_packet_security(conn); - if (ret < 0) - goto error_2; - - write_lock(&rxrpc_connection_lock); - list_add_tail(&conn->link, &rxrpc_connections); - write_unlock(&rxrpc_connection_lock); - - /* We steal the caller's peer ref. */ - cp->peer = NULL; - rxrpc_get_local(conn->params.local); - key_get(conn->params.key); - - _leave(" = %p", conn); - return conn; - -error_2: - conn->security->clear(conn); -error_1: - rxrpc_put_client_connection_id(conn); -error_0: - kfree(conn); - _leave(" = %d", ret); - return ERR_PTR(ret); -} - -/* - * find a connection for a call - * - called in process context with IRQs enabled - */ -int rxrpc_connect_call(struct rxrpc_call *call, - struct rxrpc_conn_parameters *cp, - struct sockaddr_rxrpc *srx, - gfp_t gfp) -{ - struct rxrpc_connection *conn, *candidate = NULL; - struct rxrpc_local *local = cp->local; - struct rb_node *p, **pp, *parent; - long diff; - int chan; - - DECLARE_WAITQUEUE(myself, current); - - _enter("{%d,%lx},", call->debug_id, call->user_call_ID); - - cp->peer = rxrpc_lookup_peer(cp->local, srx, gfp); - if (!cp->peer) - return -ENOMEM; - - if (!cp->exclusive) { - /* Search for a existing client connection unless this is going - * to be a connection that's used exclusively for a single call. - */ - _debug("search 1"); - spin_lock(&local->client_conns_lock); - p = local->client_conns.rb_node; - while (p) { - conn = rb_entry(p, struct rxrpc_connection, client_node); - -#define cmp(X) ((long)conn->params.X - (long)cp->X) - diff = (cmp(peer) ?: - cmp(key) ?: - cmp(security_level)); - if (diff < 0) - p = p->rb_left; - else if (diff > 0) - p = p->rb_right; - else - goto found_extant_conn; - } - spin_unlock(&local->client_conns_lock); - } - - /* We didn't find a connection or we want an exclusive one. */ - _debug("get new conn"); - candidate = rxrpc_alloc_client_connection(cp, gfp); - if (!candidate) { - _leave(" = -ENOMEM"); - return -ENOMEM; - } - - if (cp->exclusive) { - /* Assign the call on an exclusive connection to channel 0 and - * don't add the connection to the endpoint's shareable conn - * lookup tree. - */ - _debug("exclusive chan 0"); - conn = candidate; - atomic_set(&conn->avail_chans, RXRPC_MAXCALLS - 1); - spin_lock(&conn->channel_lock); - chan = 0; - goto found_channel; - } - - /* We need to redo the search before attempting to add a new connection - * lest we race with someone else adding a conflicting instance. - */ - _debug("search 2"); - spin_lock(&local->client_conns_lock); - - pp = &local->client_conns.rb_node; - parent = NULL; - while (*pp) { - parent = *pp; - conn = rb_entry(parent, struct rxrpc_connection, client_node); - - diff = (cmp(peer) ?: - cmp(key) ?: - cmp(security_level)); - if (diff < 0) - pp = &(*pp)->rb_left; - else if (diff > 0) - pp = &(*pp)->rb_right; - else - goto found_extant_conn; - } - - /* The second search also failed; simply add the new connection with - * the new call in channel 0. Note that we need to take the channel - * lock before dropping the client conn lock. - */ - _debug("new conn"); - conn = candidate; - candidate = NULL; - - rb_link_node(&conn->client_node, parent, pp); - rb_insert_color(&conn->client_node, &local->client_conns); - - atomic_set(&conn->avail_chans, RXRPC_MAXCALLS - 1); - spin_lock(&conn->channel_lock); - spin_unlock(&local->client_conns_lock); - chan = 0; - -found_channel: - _debug("found chan"); - call->conn = conn; - call->channel = chan; - call->epoch = conn->proto.epoch; - call->cid = conn->proto.cid | chan; - call->call_id = ++conn->channels[chan].call_counter; - conn->channels[chan].call_id = call->call_id; - rcu_assign_pointer(conn->channels[chan].call, call); - - _net("CONNECT call %d on conn %d", call->debug_id, conn->debug_id); - - spin_unlock(&conn->channel_lock); - rxrpc_put_peer(cp->peer); - cp->peer = NULL; - _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); - return 0; - - /* We found a suitable connection already in existence. Discard any - * candidate we may have allocated, and try to get a channel on this - * one. - */ -found_extant_conn: - _debug("found conn"); - rxrpc_get_connection(conn); - spin_unlock(&local->client_conns_lock); - - rxrpc_put_connection(candidate); - - if (!atomic_add_unless(&conn->avail_chans, -1, 0)) { - if (!gfpflags_allow_blocking(gfp)) { - rxrpc_put_connection(conn); - _leave(" = -EAGAIN"); - return -EAGAIN; - } - - add_wait_queue(&conn->channel_wq, &myself); - for (;;) { - set_current_state(TASK_INTERRUPTIBLE); - if (atomic_add_unless(&conn->avail_chans, -1, 0)) - break; - if (signal_pending(current)) - goto interrupted; - schedule(); - } - remove_wait_queue(&conn->channel_wq, &myself); - __set_current_state(TASK_RUNNING); - } - - /* The connection allegedly now has a free channel and we can now - * attach the call to it. - */ - spin_lock(&conn->channel_lock); - - for (chan = 0; chan < RXRPC_MAXCALLS; chan++) - if (!conn->channels[chan].call) - goto found_channel; - BUG(); - -interrupted: - remove_wait_queue(&conn->channel_wq, &myself); - __set_current_state(TASK_RUNNING); - rxrpc_put_connection(conn); - rxrpc_put_peer(cp->peer); - cp->peer = NULL; - _leave(" = -ERESTARTSYS"); - return -ERESTARTSYS; -} - -/* * get a record of an incoming connection */ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, -- cgit v0.10.2 From 7877a4a4bdf0d782276f1cba868878aee77718ee Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:40 +0100 Subject: rxrpc: Split service connection code out into its own file Split the service-specific connection code out into into its own file. The client-specific code has already been split out. This will leave just the common code in the original file. Signed-off-by: David Howells diff --git a/net/rxrpc/Makefile b/net/rxrpc/Makefile index 6522e50..10f3f48 100644 --- a/net/rxrpc/Makefile +++ b/net/rxrpc/Makefile @@ -10,6 +10,7 @@ af-rxrpc-y := \ conn_client.o \ conn_event.o \ conn_object.o \ + conn_service.o \ input.o \ insecure.o \ key.o \ diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 021d28b..e1af258 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -563,6 +563,9 @@ extern unsigned int rxrpc_connection_expiry; extern struct list_head rxrpc_connections; extern rwlock_t rxrpc_connection_lock; +void rxrpc_conn_hash_proto_key(struct rxrpc_conn_proto *); +void rxrpc_extract_conn_params(struct rxrpc_conn_proto *, + struct rxrpc_local *, struct sk_buff *); struct rxrpc_connection *rxrpc_alloc_connection(gfp_t); struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *, struct rxrpc_peer *, @@ -571,9 +574,6 @@ void __rxrpc_disconnect_call(struct rxrpc_call *); void rxrpc_disconnect_call(struct rxrpc_call *); void rxrpc_put_connection(struct rxrpc_connection *); void __exit rxrpc_destroy_all_connections(void); -struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *, - struct rxrpc_peer *, - struct sk_buff *); static inline bool rxrpc_conn_is_client(const struct rxrpc_connection *conn) { @@ -604,6 +604,13 @@ static inline void rxrpc_queue_conn(struct rxrpc_connection *conn) } /* + * conn_service.c + */ +struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *, + struct rxrpc_peer *, + struct sk_buff *); + +/* * input.c */ void rxrpc_data_ready(struct sock *); diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index ab5c8c2..8379e374 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -61,138 +61,6 @@ struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) } /* - * get a record of an incoming connection - */ -struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, - struct rxrpc_peer *peer, - struct sk_buff *skb) -{ - struct rxrpc_connection *conn, *candidate = NULL; - struct rxrpc_skb_priv *sp = rxrpc_skb(skb); - struct rb_node *p, **pp; - const char *new = "old"; - u32 epoch, cid; - - _enter(""); - - ASSERT(sp->hdr.flags & RXRPC_CLIENT_INITIATED); - - epoch = sp->hdr.epoch; - cid = sp->hdr.cid & RXRPC_CIDMASK; - - /* search the connection list first */ - read_lock_bh(&peer->conn_lock); - - p = peer->service_conns.rb_node; - while (p) { - conn = rb_entry(p, struct rxrpc_connection, service_node); - - _debug("maybe %x", conn->proto.cid); - - if (epoch < conn->proto.epoch) - p = p->rb_left; - else if (epoch > conn->proto.epoch) - p = p->rb_right; - else if (cid < conn->proto.cid) - p = p->rb_left; - else if (cid > conn->proto.cid) - p = p->rb_right; - else - goto found_extant_connection; - } - read_unlock_bh(&peer->conn_lock); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_connection(GFP_NOIO); - if (!candidate) { - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } - - candidate->proto.local = local; - candidate->proto.epoch = sp->hdr.epoch; - candidate->proto.cid = sp->hdr.cid & RXRPC_CIDMASK; - candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED; - candidate->params.local = local; - candidate->params.peer = peer; - candidate->params.service_id = sp->hdr.serviceId; - candidate->security_ix = sp->hdr.securityIndex; - candidate->out_clientflag = 0; - candidate->state = RXRPC_CONN_SERVICE; - if (candidate->params.service_id) - candidate->state = RXRPC_CONN_SERVICE_UNSECURED; - - write_lock_bh(&peer->conn_lock); - - pp = &peer->service_conns.rb_node; - p = NULL; - while (*pp) { - p = *pp; - conn = rb_entry(p, struct rxrpc_connection, service_node); - - if (epoch < conn->proto.epoch) - pp = &(*pp)->rb_left; - else if (epoch > conn->proto.epoch) - pp = &(*pp)->rb_right; - else if (cid < conn->proto.cid) - pp = &(*pp)->rb_left; - else if (cid > conn->proto.cid) - pp = &(*pp)->rb_right; - else - goto found_extant_second; - } - - /* we can now add the new candidate to the list */ - conn = candidate; - candidate = NULL; - rb_link_node(&conn->service_node, p, pp); - rb_insert_color(&conn->service_node, &peer->service_conns); - rxrpc_get_peer(peer); - rxrpc_get_local(local); - - write_unlock_bh(&peer->conn_lock); - - write_lock(&rxrpc_connection_lock); - list_add_tail(&conn->link, &rxrpc_connections); - write_unlock(&rxrpc_connection_lock); - - new = "new"; - -success: - _net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->proto.cid); - - _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); - return conn; - - /* we found the connection in the list immediately */ -found_extant_connection: - if (sp->hdr.securityIndex != conn->security_ix) { - read_unlock_bh(&peer->conn_lock); - goto security_mismatch; - } - rxrpc_get_connection(conn); - read_unlock_bh(&peer->conn_lock); - goto success; - - /* we found the connection on the second time through the list */ -found_extant_second: - if (sp->hdr.securityIndex != conn->security_ix) { - write_unlock_bh(&peer->conn_lock); - goto security_mismatch; - } - rxrpc_get_connection(conn); - write_unlock_bh(&peer->conn_lock); - kfree(candidate); - goto success; - -security_mismatch: - kfree(candidate); - _leave(" = -EKEYREJECTED"); - return ERR_PTR(-EKEYREJECTED); -} - -/* * find a connection based on transport and RxRPC connection ID for an incoming * packet */ diff --git a/net/rxrpc/conn_service.c b/net/rxrpc/conn_service.c new file mode 100644 index 0000000..cdcac50 --- /dev/null +++ b/net/rxrpc/conn_service.c @@ -0,0 +1,145 @@ +/* Service connection management + * + * Copyright (C) 2016 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#include +#include "ar-internal.h" + +/* + * get a record of an incoming connection + */ +struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, + struct rxrpc_peer *peer, + struct sk_buff *skb) +{ + struct rxrpc_connection *conn, *candidate = NULL; + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + struct rb_node *p, **pp; + const char *new = "old"; + u32 epoch, cid; + + _enter(""); + + ASSERT(sp->hdr.flags & RXRPC_CLIENT_INITIATED); + + epoch = sp->hdr.epoch; + cid = sp->hdr.cid & RXRPC_CIDMASK; + + /* search the connection list first */ + read_lock_bh(&peer->conn_lock); + + p = peer->service_conns.rb_node; + while (p) { + conn = rb_entry(p, struct rxrpc_connection, service_node); + + _debug("maybe %x", conn->proto.cid); + + if (epoch < conn->proto.epoch) + p = p->rb_left; + else if (epoch > conn->proto.epoch) + p = p->rb_right; + else if (cid < conn->proto.cid) + p = p->rb_left; + else if (cid > conn->proto.cid) + p = p->rb_right; + else + goto found_extant_connection; + } + read_unlock_bh(&peer->conn_lock); + + /* not yet present - create a candidate for a new record and then + * redo the search */ + candidate = rxrpc_alloc_connection(GFP_NOIO); + if (!candidate) { + _leave(" = -ENOMEM"); + return ERR_PTR(-ENOMEM); + } + + candidate->proto.local = local; + candidate->proto.epoch = sp->hdr.epoch; + candidate->proto.cid = sp->hdr.cid & RXRPC_CIDMASK; + candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED; + candidate->params.local = local; + candidate->params.peer = peer; + candidate->params.service_id = sp->hdr.serviceId; + candidate->security_ix = sp->hdr.securityIndex; + candidate->out_clientflag = 0; + candidate->state = RXRPC_CONN_SERVICE; + if (candidate->params.service_id) + candidate->state = RXRPC_CONN_SERVICE_UNSECURED; + + write_lock_bh(&peer->conn_lock); + + pp = &peer->service_conns.rb_node; + p = NULL; + while (*pp) { + p = *pp; + conn = rb_entry(p, struct rxrpc_connection, service_node); + + if (epoch < conn->proto.epoch) + pp = &(*pp)->rb_left; + else if (epoch > conn->proto.epoch) + pp = &(*pp)->rb_right; + else if (cid < conn->proto.cid) + pp = &(*pp)->rb_left; + else if (cid > conn->proto.cid) + pp = &(*pp)->rb_right; + else + goto found_extant_second; + } + + /* we can now add the new candidate to the list */ + conn = candidate; + candidate = NULL; + rb_link_node(&conn->service_node, p, pp); + rb_insert_color(&conn->service_node, &peer->service_conns); + rxrpc_get_peer(peer); + rxrpc_get_local(local); + + write_unlock_bh(&peer->conn_lock); + + write_lock(&rxrpc_connection_lock); + list_add_tail(&conn->link, &rxrpc_connections); + write_unlock(&rxrpc_connection_lock); + + new = "new"; + +success: + _net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->proto.cid); + + _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); + return conn; + + /* we found the connection in the list immediately */ +found_extant_connection: + if (sp->hdr.securityIndex != conn->security_ix) { + read_unlock_bh(&peer->conn_lock); + goto security_mismatch; + } + rxrpc_get_connection(conn); + read_unlock_bh(&peer->conn_lock); + goto success; + + /* we found the connection on the second time through the list */ +found_extant_second: + if (sp->hdr.securityIndex != conn->security_ix) { + write_unlock_bh(&peer->conn_lock); + goto security_mismatch; + } + rxrpc_get_connection(conn); + write_unlock_bh(&peer->conn_lock); + kfree(candidate); + goto success; + +security_mismatch: + kfree(candidate); + _leave(" = -EKEYREJECTED"); + return ERR_PTR(-EKEYREJECTED); +} -- cgit v0.10.2 From d991b4a32f65076efaf78739c4a46406ca8c7e79 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 29 Jun 2016 14:40:39 +0100 Subject: rxrpc: Move peer lookup from call-accept to new-incoming-conn Move the lookup of a peer from a call that's being accepted into the function that creates a new incoming connection. This will allow us to avoid incrementing the peer's usage count in some cases in future. Note that I haven't bother to integrate rxrpc_get_addr_from_skb() with rxrpc_extract_addr_from_skb() as I'm going to delete the former in the very near future. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index e1af258..ad48f85 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -607,7 +607,7 @@ static inline void rxrpc_queue_conn(struct rxrpc_connection *conn) * conn_service.c */ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *, - struct rxrpc_peer *, + struct sockaddr_rxrpc *, struct sk_buff *); /* @@ -773,6 +773,7 @@ static inline void rxrpc_sysctl_exit(void) {} */ void rxrpc_get_addr_from_skb(struct rxrpc_local *, const struct sk_buff *, struct sockaddr_rxrpc *); +int rxrpc_extract_addr_from_skb(struct sockaddr_rxrpc *, struct sk_buff *); /* * debug tracing diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index 5367dbe..0b28321 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -75,7 +75,6 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, { struct rxrpc_connection *conn; struct rxrpc_skb_priv *sp, *nsp; - struct rxrpc_peer *peer; struct rxrpc_call *call; struct sk_buff *notification; int ret; @@ -94,15 +93,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local, rxrpc_new_skb(notification); notification->mark = RXRPC_SKB_MARK_NEW_CALL; - peer = rxrpc_lookup_peer(local, srx, GFP_NOIO); - if (!peer) { - _debug("no peer"); - ret = -EBUSY; - goto error; - } - - conn = rxrpc_incoming_connection(local, peer, skb); - rxrpc_put_peer(peer); + conn = rxrpc_incoming_connection(local, srx, skb); if (IS_ERR(conn)) { _debug("no conn"); ret = PTR_ERR(conn); @@ -226,20 +217,8 @@ void rxrpc_accept_incoming_calls(struct rxrpc_local *local) whdr._rsvd = 0; whdr.serviceId = htons(sp->hdr.serviceId); - /* determine the remote address */ - memset(&srx, 0, sizeof(srx)); - srx.srx_family = AF_RXRPC; - srx.transport.family = local->srx.transport.family; - srx.transport_type = local->srx.transport_type; - switch (srx.transport.family) { - case AF_INET: - srx.transport_len = sizeof(struct sockaddr_in); - srx.transport.sin.sin_port = udp_hdr(skb)->source; - srx.transport.sin.sin_addr.s_addr = ip_hdr(skb)->saddr; - break; - default: - goto busy; - } + if (rxrpc_extract_addr_from_skb(&srx, skb) < 0) + goto drop; /* get the socket providing the service */ read_lock_bh(&local->services_lock); @@ -285,6 +264,10 @@ busy: rxrpc_free_skb(skb); return; +drop: + rxrpc_free_skb(skb); + return; + invalid_service: skb->priority = RX_INVALID_OPERATION; rxrpc_reject_packet(local, skb); diff --git a/net/rxrpc/conn_service.c b/net/rxrpc/conn_service.c index cdcac50..a42b210 100644 --- a/net/rxrpc/conn_service.c +++ b/net/rxrpc/conn_service.c @@ -16,17 +16,24 @@ * get a record of an incoming connection */ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, - struct rxrpc_peer *peer, + struct sockaddr_rxrpc *srx, struct sk_buff *skb) { struct rxrpc_connection *conn, *candidate = NULL; struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + struct rxrpc_peer *peer; struct rb_node *p, **pp; const char *new = "old"; u32 epoch, cid; _enter(""); + peer = rxrpc_lookup_peer(local, srx, GFP_NOIO); + if (!peer) { + _debug("no peer"); + return ERR_PTR(-EBUSY); + } + ASSERT(sp->hdr.flags & RXRPC_CLIENT_INITIATED); epoch = sp->hdr.epoch; @@ -58,6 +65,7 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, * redo the search */ candidate = rxrpc_alloc_connection(GFP_NOIO); if (!candidate) { + rxrpc_put_peer(peer); _leave(" = -ENOMEM"); return ERR_PTR(-ENOMEM); } @@ -114,6 +122,7 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, success: _net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->proto.cid); + rxrpc_put_peer(peer); _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); return conn; diff --git a/net/rxrpc/utils.c b/net/rxrpc/utils.c index f28122a..d3db02e 100644 --- a/net/rxrpc/utils.c +++ b/net/rxrpc/utils.c @@ -10,6 +10,7 @@ */ #include +#include #include #include "ar-internal.h" @@ -39,3 +40,34 @@ void rxrpc_get_addr_from_skb(struct rxrpc_local *local, BUG(); } } + +/* + * Fill out a peer address from a socket buffer containing a packet. + */ +int rxrpc_extract_addr_from_skb(struct sockaddr_rxrpc *srx, struct sk_buff *skb) +{ + memset(srx, 0, sizeof(*srx)); + + switch (ntohs(skb->protocol)) { + case ETH_P_IP: + srx->transport_type = SOCK_DGRAM; + srx->transport_len = sizeof(srx->transport.sin); + srx->transport.sin.sin_family = AF_INET; + srx->transport.sin.sin_port = udp_hdr(skb)->source; + srx->transport.sin.sin_addr.s_addr = ip_hdr(skb)->saddr; + return 0; + + case ETH_P_IPV6: + srx->transport_type = SOCK_DGRAM; + srx->transport_len = sizeof(srx->transport.sin6); + srx->transport.sin6.sin6_family = AF_INET6; + srx->transport.sin6.sin6_port = udp_hdr(skb)->source; + srx->transport.sin6.sin6_addr = ipv6_hdr(skb)->saddr; + return 0; + + default: + pr_warn_ratelimited("AF_RXRPC: Unknown eth protocol %u\n", + ntohs(skb->protocol)); + return -EAFNOSUPPORT; + } +} -- cgit v0.10.2 From 001c11224910b25e59a65ce1b49cfecdb4c631c0 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 30 Jun 2016 10:45:22 +0100 Subject: rxrpc: Maintain an extra ref on a conn for the cache list Overhaul the usage count accounting for the rxrpc_connection struct to make it easier to implement RCU access from the data_ready handler. The problem is that currently we're using a lock to prevent the garbage collector from trying to clean up a connection that we're contemplating unidling. We could just stick incoming packets on the connection we find, but we've then got a problem that we may race when dispatching a work item to process it as we need to give that a ref to prevent the rxrpc_connection struct from disappearing in the meantime. Further, incoming packets may get discarded if attached to an rxrpc_connection struct that is going away. Whilst this is not a total disaster - the client will presumably resend - it would delay processing of the call. This would affect the AFS client filesystem's service manager operation. To this end: (1) We now maintain an extra count on the connection usage count whilst it is on the connection list. This mean it is not in use when its refcount is 1. (2) When trying to reuse an old connection, we only increment the refcount if it is greater than 0. If it is 0, we replace it in the tree with a new candidate connection. (3) Two connection flags are added to indicate whether or not a connection is in the local's client connection tree (used by sendmsg) or the peer's service connection tree (used by data_ready). This makes sure that we don't try and remove a connection if it got replaced. The flags are tested under lock with the removal operation to prevent the reaper from killing the rxrpc_connection struct whilst someone else is trying to effect a replacement. This could probably be alleviated by using memory barriers between the flag set/test and the rb_tree ops. The rb_tree op would still need to be under the lock, however. (4) When trying to reap an old connection, we try to flip the usage count from 1 to 0. If it's not 1 at that point, then it must've come back to life temporarily and we ignore it. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index ad48f85..d8e4d6e 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -258,6 +258,8 @@ struct rxrpc_conn_parameters { */ enum rxrpc_conn_flag { RXRPC_CONN_HAS_IDR, /* Has a client conn ID assigned */ + RXRPC_CONN_IN_SERVICE_CONNS, /* Conn is in peer->service_conns */ + RXRPC_CONN_IN_CLIENT_CONNS, /* Conn is in local->client_conns */ }; /* @@ -544,10 +546,10 @@ void __exit rxrpc_destroy_all_calls(void); */ extern struct idr rxrpc_client_conn_ids; -void rxrpc_put_client_connection_id(struct rxrpc_connection *); void rxrpc_destroy_client_conn_ids(void); int rxrpc_connect_call(struct rxrpc_call *, struct rxrpc_conn_parameters *, struct sockaddr_rxrpc *, gfp_t); +void rxrpc_unpublish_client_conn(struct rxrpc_connection *); /* * conn_event.c @@ -609,6 +611,7 @@ static inline void rxrpc_queue_conn(struct rxrpc_connection *conn) struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *, struct sockaddr_rxrpc *, struct sk_buff *); +void rxrpc_unpublish_service_conn(struct rxrpc_connection *); /* * input.c diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c index 9180164..aa21462 100644 --- a/net/rxrpc/conn_client.c +++ b/net/rxrpc/conn_client.c @@ -84,7 +84,7 @@ error: /* * Release a connection ID for a client connection from the global pool. */ -void rxrpc_put_client_connection_id(struct rxrpc_connection *conn) +static void rxrpc_put_client_connection_id(struct rxrpc_connection *conn) { if (test_bit(RXRPC_CONN_HAS_IDR, &conn->flags)) { spin_lock(&rxrpc_conn_id_lock); @@ -278,12 +278,13 @@ int rxrpc_connect_call(struct rxrpc_call *call, * lock before dropping the client conn lock. */ _debug("new conn"); + set_bit(RXRPC_CONN_IN_CLIENT_CONNS, &candidate->flags); + rb_link_node(&candidate->client_node, parent, pp); + rb_insert_color(&candidate->client_node, &local->client_conns); +attached: conn = candidate; candidate = NULL; - rb_link_node(&conn->client_node, parent, pp); - rb_insert_color(&conn->client_node, &local->client_conns); - atomic_set(&conn->avail_chans, RXRPC_MAXCALLS - 1); spin_lock(&conn->channel_lock); spin_unlock(&local->client_conns_lock); @@ -307,13 +308,22 @@ found_channel: _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); return 0; - /* We found a suitable connection already in existence. Discard any - * candidate we may have allocated, and try to get a channel on this - * one. + /* We found a potentially suitable connection already in existence. If + * we can reuse it (ie. its usage count hasn't been reduced to 0 by the + * reaper), discard any candidate we may have allocated, and try to get + * a channel on this one, otherwise we have to replace it. */ found_extant_conn: _debug("found conn"); - rxrpc_get_connection(conn); + if (!rxrpc_get_connection_maybe(conn)) { + set_bit(RXRPC_CONN_IN_CLIENT_CONNS, &candidate->flags); + rb_replace_node(&conn->client_node, + &candidate->client_node, + &local->client_conns); + clear_bit(RXRPC_CONN_IN_CLIENT_CONNS, &conn->flags); + goto attached; + } + spin_unlock(&local->client_conns_lock); rxrpc_put_connection(candidate); @@ -357,3 +367,19 @@ interrupted: _leave(" = -ERESTARTSYS"); return -ERESTARTSYS; } + +/* + * Remove a client connection from the local endpoint's tree, thereby removing + * it as a target for reuse for new client calls. + */ +void rxrpc_unpublish_client_conn(struct rxrpc_connection *conn) +{ + struct rxrpc_local *local = conn->params.local; + + spin_lock(&local->client_conns_lock); + if (test_and_clear_bit(RXRPC_CONN_IN_CLIENT_CONNS, &conn->flags)) + rb_erase(&conn->client_node, &local->client_conns); + spin_unlock(&local->client_conns_lock); + + rxrpc_put_client_connection_id(conn); +} diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 8379e374..89bc648 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -49,7 +49,10 @@ struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) skb_queue_head_init(&conn->rx_queue); conn->security = &rxrpc_no_security; spin_lock_init(&conn->state_lock); - atomic_set(&conn->usage, 1); + /* We maintain an extra ref on the connection whilst it is + * on the rxrpc_connections list. + */ + atomic_set(&conn->usage, 2); conn->debug_id = atomic_inc_return(&rxrpc_debug_id); atomic_set(&conn->avail_chans, RXRPC_MAXCALLS); conn->size_align = 4; @@ -111,7 +114,7 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *local, return NULL; found: - rxrpc_get_connection(conn); + conn = rxrpc_get_connection_maybe(conn); read_unlock_bh(&peer->conn_lock); _leave(" = %p", conn); return conn; @@ -173,10 +176,10 @@ void rxrpc_put_connection(struct rxrpc_connection *conn) _enter("%p{u=%d,d=%d}", conn, atomic_read(&conn->usage), conn->debug_id); - ASSERTCMP(atomic_read(&conn->usage), >, 0); + ASSERTCMP(atomic_read(&conn->usage), >, 1); conn->put_time = ktime_get_seconds(); - if (atomic_dec_and_test(&conn->usage)) { + if (atomic_dec_return(&conn->usage) == 1) { _debug("zombie"); rxrpc_queue_delayed_work(&rxrpc_connection_reap, 0); } @@ -216,59 +219,41 @@ static void rxrpc_destroy_connection(struct rcu_head *rcu) static void rxrpc_connection_reaper(struct work_struct *work) { struct rxrpc_connection *conn, *_p; - struct rxrpc_peer *peer; - unsigned long now, earliest, reap_time; + unsigned long reap_older_than, earliest, put_time, now; LIST_HEAD(graveyard); _enter(""); now = ktime_get_seconds(); + reap_older_than = now - rxrpc_connection_expiry; earliest = ULONG_MAX; write_lock(&rxrpc_connection_lock); list_for_each_entry_safe(conn, _p, &rxrpc_connections, link) { - _debug("reap CONN %d { u=%d,t=%ld }", - conn->debug_id, atomic_read(&conn->usage), - (long) now - (long) conn->put_time); - - if (likely(atomic_read(&conn->usage) > 0)) + ASSERTCMP(atomic_read(&conn->usage), >, 0); + if (likely(atomic_read(&conn->usage) > 1)) continue; - if (rxrpc_conn_is_client(conn)) { - struct rxrpc_local *local = conn->params.local; - spin_lock(&local->client_conns_lock); - reap_time = conn->put_time + rxrpc_connection_expiry; - - if (atomic_read(&conn->usage) > 0) { - ; - } else if (reap_time <= now) { - list_move_tail(&conn->link, &graveyard); - rxrpc_put_client_connection_id(conn); - rb_erase(&conn->client_node, - &local->client_conns); - } else if (reap_time < earliest) { - earliest = reap_time; - } - - spin_unlock(&local->client_conns_lock); - } else { - peer = conn->params.peer; - write_lock_bh(&peer->conn_lock); - reap_time = conn->put_time + rxrpc_connection_expiry; - - if (atomic_read(&conn->usage) > 0) { - ; - } else if (reap_time <= now) { - list_move_tail(&conn->link, &graveyard); - rb_erase(&conn->service_node, - &peer->service_conns); - } else if (reap_time < earliest) { - earliest = reap_time; - } - - write_unlock_bh(&peer->conn_lock); + put_time = READ_ONCE(conn->put_time); + if (time_after(put_time, reap_older_than)) { + if (time_before(put_time, earliest)) + earliest = put_time; + continue; } + + /* The usage count sits at 1 whilst the object is unused on the + * list; we reduce that to 0 to make the object unavailable. + */ + if (atomic_cmpxchg(&conn->usage, 1, 0) != 1) + continue; + + if (rxrpc_conn_is_client(conn)) + rxrpc_unpublish_client_conn(conn); + else + rxrpc_unpublish_service_conn(conn); + + list_move_tail(&conn->link, &graveyard); } write_unlock(&rxrpc_connection_lock); @@ -279,7 +264,6 @@ static void rxrpc_connection_reaper(struct work_struct *work) (earliest - now) * HZ); } - /* then destroy all those pulled out */ while (!list_empty(&graveyard)) { conn = list_entry(graveyard.next, struct rxrpc_connection, link); diff --git a/net/rxrpc/conn_service.c b/net/rxrpc/conn_service.c index a42b210..77a509e 100644 --- a/net/rxrpc/conn_service.c +++ b/net/rxrpc/conn_service.c @@ -104,10 +104,12 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, } /* we can now add the new candidate to the list */ + set_bit(RXRPC_CONN_IN_SERVICE_CONNS, &candidate->flags); + rb_link_node(&candidate->service_node, p, pp); + rb_insert_color(&candidate->service_node, &peer->service_conns); +attached: conn = candidate; candidate = NULL; - rb_link_node(&conn->service_node, p, pp); - rb_insert_color(&conn->service_node, &peer->service_conns); rxrpc_get_peer(peer); rxrpc_get_local(local); @@ -128,11 +130,19 @@ success: /* we found the connection in the list immediately */ found_extant_connection: + if (!rxrpc_get_connection_maybe(conn)) { + set_bit(RXRPC_CONN_IN_SERVICE_CONNS, &candidate->flags); + rb_replace_node(&conn->service_node, + &candidate->service_node, + &peer->service_conns); + clear_bit(RXRPC_CONN_IN_SERVICE_CONNS, &conn->flags); + goto attached; + } + if (sp->hdr.securityIndex != conn->security_ix) { read_unlock_bh(&peer->conn_lock); - goto security_mismatch; + goto security_mismatch_put; } - rxrpc_get_connection(conn); read_unlock_bh(&peer->conn_lock); goto success; @@ -147,8 +157,24 @@ found_extant_second: kfree(candidate); goto success; +security_mismatch_put: + rxrpc_put_connection(conn); security_mismatch: kfree(candidate); _leave(" = -EKEYREJECTED"); return ERR_PTR(-EKEYREJECTED); } + +/* + * Remove the service connection from the peer's tree, thereby removing it as a + * target for incoming packets. + */ +void rxrpc_unpublish_service_conn(struct rxrpc_connection *conn) +{ + struct rxrpc_peer *peer = conn->params.peer; + + write_lock_bh(&peer->conn_lock); + if (test_and_clear_bit(RXRPC_CONN_IN_SERVICE_CONNS, &conn->flags)) + rb_erase(&conn->service_node, &peer->service_conns); + write_unlock_bh(&peer->conn_lock); +} -- cgit v0.10.2 From e8d70ce177eeb4fbd1c218c60118d2c19c2496a6 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 30 Jun 2016 12:16:21 +0100 Subject: rxrpc: Prune the contents of the rxrpc_conn_proto struct Prune the contents of the rxrpc_conn_proto struct. Most of the fields aren't used anymore. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index d8e4d6e..6fdee76 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -229,18 +229,12 @@ struct rxrpc_peer { * Keys for matching a connection. */ struct rxrpc_conn_proto { - unsigned long hash_key; - struct rxrpc_local *local; /* Representation of local endpoint */ - u32 epoch; /* epoch of this connection */ - u32 cid; /* connection ID */ - u8 in_clientflag; /* RXRPC_CLIENT_INITIATED if we are server */ - u8 addr_size; /* Size of the address */ - sa_family_t family; /* Transport protocol */ - __be16 port; /* Peer UDP/UDP6 port */ - union { /* Peer address */ - struct in_addr ipv4_addr; - struct in6_addr ipv6_addr; - u32 raw_addr[0]; + union { + struct { + u32 epoch; /* epoch of this connection */ + u32 cid; /* connection ID */ + }; + u64 index_key; }; }; @@ -584,7 +578,7 @@ static inline bool rxrpc_conn_is_client(const struct rxrpc_connection *conn) static inline bool rxrpc_conn_is_service(const struct rxrpc_connection *conn) { - return conn->proto.in_clientflag; + return !rxrpc_conn_is_client(conn); } static inline void rxrpc_get_connection(struct rxrpc_connection *conn) diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 3f27872..ebbd7dd 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -566,7 +566,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, } call->epoch = conn->proto.epoch; call->service_id = conn->params.service_id; - call->in_clientflag = conn->proto.in_clientflag; + call->in_clientflag = RXRPC_CLIENT_INITIATED; /* Add the new call to the hashtable */ rxrpc_call_hash_add(call); diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c index aa21462..917db48d 100644 --- a/net/rxrpc/conn_client.c +++ b/net/rxrpc/conn_client.c @@ -132,22 +132,11 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, gfp_t gfp) } conn->params = *cp; - conn->proto.local = cp->local; conn->proto.epoch = rxrpc_epoch; conn->proto.cid = 0; - conn->proto.in_clientflag = 0; - conn->proto.family = cp->peer->srx.transport.family; conn->out_clientflag = RXRPC_CLIENT_INITIATED; conn->state = RXRPC_CONN_CLIENT; - switch (conn->proto.family) { - case AF_INET: - conn->proto.addr_size = sizeof(conn->proto.ipv4_addr); - conn->proto.ipv4_addr = cp->peer->srx.transport.sin.sin_addr; - conn->proto.port = cp->peer->srx.transport.sin.sin_port; - break; - } - ret = rxrpc_get_client_connection_id(conn, gfp); if (ret < 0) goto error_0; diff --git a/net/rxrpc/conn_service.c b/net/rxrpc/conn_service.c index 77a509e..c6db2e8 100644 --- a/net/rxrpc/conn_service.c +++ b/net/rxrpc/conn_service.c @@ -70,10 +70,8 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, return ERR_PTR(-ENOMEM); } - candidate->proto.local = local; candidate->proto.epoch = sp->hdr.epoch; candidate->proto.cid = sp->hdr.cid & RXRPC_CIDMASK; - candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED; candidate->params.local = local; candidate->params.peer = peer; candidate->params.service_id = sp->hdr.serviceId; -- cgit v0.10.2 From 1291e9d1084506c5cba6313ce809d7516bb5868a Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 30 Jun 2016 12:02:53 +0100 Subject: rxrpc: Move data_ready peer lookup into rxrpc_find_connection() Move the peer lookup done in input.c by data_ready into rxrpc_find_connection(). Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 6fdee76..0fe63ba 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -564,7 +564,6 @@ void rxrpc_extract_conn_params(struct rxrpc_conn_proto *, struct rxrpc_local *, struct sk_buff *); struct rxrpc_connection *rxrpc_alloc_connection(gfp_t); struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *, - struct rxrpc_peer *, struct sk_buff *); void __rxrpc_disconnect_call(struct rxrpc_call *); void rxrpc_disconnect_call(struct rxrpc_call *); @@ -768,8 +767,6 @@ static inline void rxrpc_sysctl_exit(void) {} /* * utils.c */ -void rxrpc_get_addr_from_skb(struct rxrpc_local *, const struct sk_buff *, - struct sockaddr_rxrpc *); int rxrpc_extract_addr_from_skb(struct sockaddr_rxrpc *, struct sk_buff *); /* diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 89bc648..1307138 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -68,52 +68,91 @@ struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) * packet */ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *local, - struct rxrpc_peer *peer, struct sk_buff *skb) { struct rxrpc_connection *conn; + struct rxrpc_conn_proto k; struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + struct sockaddr_rxrpc srx; + struct rxrpc_peer *peer; struct rb_node *p; - u32 epoch, cid; _enter(",{%x,%x}", sp->hdr.cid, sp->hdr.flags); - read_lock_bh(&peer->conn_lock); + if (rxrpc_extract_addr_from_skb(&srx, skb) < 0) + goto not_found; - cid = sp->hdr.cid & RXRPC_CIDMASK; - epoch = sp->hdr.epoch; + /* We may have to handle mixing IPv4 and IPv6 */ + if (srx.transport.family != local->srx.transport.family) { + pr_warn_ratelimited("AF_RXRPC: Protocol mismatch %u not %u\n", + srx.transport.family, + local->srx.transport.family); + goto not_found; + } + + k.epoch = sp->hdr.epoch; + k.cid = sp->hdr.cid & RXRPC_CIDMASK; if (sp->hdr.flags & RXRPC_CLIENT_INITIATED) { + /* We need to look up service connections by the full protocol + * parameter set. We look up the peer first as an intermediate + * step and then the connection from the peer's tree. + */ + peer = rxrpc_lookup_peer_rcu(local, &srx); + if (!peer) + goto not_found; + + read_lock_bh(&peer->conn_lock); + p = peer->service_conns.rb_node; while (p) { conn = rb_entry(p, struct rxrpc_connection, service_node); _debug("maybe %x", conn->proto.cid); - if (epoch < conn->proto.epoch) + if (k.epoch < conn->proto.epoch) p = p->rb_left; - else if (epoch > conn->proto.epoch) + else if (k.epoch > conn->proto.epoch) p = p->rb_right; - else if (cid < conn->proto.cid) + else if (k.cid < conn->proto.cid) p = p->rb_left; - else if (cid > conn->proto.cid) + else if (k.cid > conn->proto.cid) p = p->rb_right; else - goto found; + goto found_service_conn; } + read_unlock_bh(&peer->conn_lock); } else { - conn = idr_find(&rxrpc_client_conn_ids, cid >> RXRPC_CIDSHIFT); - if (conn && - conn->proto.epoch == epoch && - conn->params.peer == peer) - goto found; + conn = idr_find(&rxrpc_client_conn_ids, + k.cid >> RXRPC_CIDSHIFT); + if (!conn || + conn->proto.epoch != k.epoch || + conn->params.local != local) + goto not_found; + + peer = conn->params.peer; + switch (srx.transport.family) { + case AF_INET: + if (peer->srx.transport.sin.sin_port != + srx.transport.sin.sin_port || + peer->srx.transport.sin.sin_addr.s_addr != + srx.transport.sin.sin_addr.s_addr) + goto not_found; + break; + default: + BUG(); + } + + conn = rxrpc_get_connection_maybe(conn); + _leave(" = %p", conn); + return conn; } - read_unlock_bh(&peer->conn_lock); +not_found: _leave(" = NULL"); return NULL; -found: +found_service_conn: conn = rxrpc_get_connection_maybe(conn); read_unlock_bh(&peer->conn_lock); _leave(" = %p", conn); diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index b993f2d..c243647 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -626,32 +626,6 @@ int rxrpc_extract_header(struct rxrpc_skb_priv *sp, struct sk_buff *skb) return 0; } -static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local, - struct sk_buff *skb) -{ - struct rxrpc_peer *peer; - struct rxrpc_connection *conn; - struct sockaddr_rxrpc srx; - - rxrpc_get_addr_from_skb(local, skb, &srx); - rcu_read_lock(); - peer = rxrpc_lookup_peer_rcu(local, &srx); - if (!peer) - goto cant_find_peer; - - conn = rxrpc_find_connection(local, peer, skb); - rcu_read_unlock(); - if (!conn) - goto cant_find_conn; - - return conn; - -cant_find_peer: - rcu_read_unlock(); -cant_find_conn: - return NULL; -} - /* * handle data received on the local endpoint * - may be called in interrupt context @@ -731,7 +705,9 @@ void rxrpc_data_ready(struct sock *sk) * old-fashioned way doesn't really hurt */ struct rxrpc_connection *conn; - conn = rxrpc_conn_from_local(local, skb); + rcu_read_lock(); + conn = rxrpc_find_connection(local, skb); + rcu_read_unlock(); if (!conn) goto cant_route_call; diff --git a/net/rxrpc/utils.c b/net/rxrpc/utils.c index d3db02e..b88914d 100644 --- a/net/rxrpc/utils.c +++ b/net/rxrpc/utils.c @@ -15,33 +15,6 @@ #include "ar-internal.h" /* - * Set up an RxRPC address from a socket buffer. - */ -void rxrpc_get_addr_from_skb(struct rxrpc_local *local, - const struct sk_buff *skb, - struct sockaddr_rxrpc *srx) -{ - memset(srx, 0, sizeof(*srx)); - srx->transport_type = local->srx.transport_type; - srx->transport.family = local->srx.transport.family; - - /* Can we see an ipv4 UDP packet on an ipv6 UDP socket? and vice - * versa? - */ - switch (srx->transport.family) { - case AF_INET: - srx->transport.sin.sin_port = udp_hdr(skb)->source; - srx->transport_len = sizeof(struct sockaddr_in); - memcpy(&srx->transport.sin.sin_addr, &ip_hdr(skb)->saddr, - sizeof(struct in_addr)); - break; - - default: - BUG(); - } -} - -/* * Fill out a peer address from a socket buffer containing a packet. */ int rxrpc_extract_addr_from_skb(struct sockaddr_rxrpc *srx, struct sk_buff *skb) -- cgit v0.10.2 From c1adf20052d80f776849fa2c1acb472cdeb7786c Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 1 Jul 2016 07:53:51 +0100 Subject: Introduce rb_replace_node_rcu() Implement an RCU-safe variant of rb_replace_node() and rearrange rb_replace_node() to do things in the same order. Signed-off-by: David Howells Acked-by: Peter Zijlstra (Intel) diff --git a/include/linux/rbtree.h b/include/linux/rbtree.h index b690009..e585018 100644 --- a/include/linux/rbtree.h +++ b/include/linux/rbtree.h @@ -76,6 +76,8 @@ extern struct rb_node *rb_next_postorder(const struct rb_node *); /* Fast replacement of a single node without remove/rebalance/add/rebalance */ extern void rb_replace_node(struct rb_node *victim, struct rb_node *new, struct rb_root *root); +extern void rb_replace_node_rcu(struct rb_node *victim, struct rb_node *new, + struct rb_root *root); static inline void rb_link_node(struct rb_node *node, struct rb_node *parent, struct rb_node **rb_link) diff --git a/include/linux/rbtree_augmented.h b/include/linux/rbtree_augmented.h index 14d7b83..d076183 100644 --- a/include/linux/rbtree_augmented.h +++ b/include/linux/rbtree_augmented.h @@ -130,6 +130,19 @@ __rb_change_child(struct rb_node *old, struct rb_node *new, WRITE_ONCE(root->rb_node, new); } +static inline void +__rb_change_child_rcu(struct rb_node *old, struct rb_node *new, + struct rb_node *parent, struct rb_root *root) +{ + if (parent) { + if (parent->rb_left == old) + rcu_assign_pointer(parent->rb_left, new); + else + rcu_assign_pointer(parent->rb_right, new); + } else + rcu_assign_pointer(root->rb_node, new); +} + extern void __rb_erase_color(struct rb_node *parent, struct rb_root *root, void (*augment_rotate)(struct rb_node *old, struct rb_node *new)); diff --git a/lib/rbtree.c b/lib/rbtree.c index 1356454..eb8a19f 100644 --- a/lib/rbtree.c +++ b/lib/rbtree.c @@ -539,17 +539,39 @@ void rb_replace_node(struct rb_node *victim, struct rb_node *new, { struct rb_node *parent = rb_parent(victim); + /* Copy the pointers/colour from the victim to the replacement */ + *new = *victim; + /* Set the surrounding nodes to point to the replacement */ - __rb_change_child(victim, new, parent, root); if (victim->rb_left) rb_set_parent(victim->rb_left, new); if (victim->rb_right) rb_set_parent(victim->rb_right, new); + __rb_change_child(victim, new, parent, root); +} +EXPORT_SYMBOL(rb_replace_node); + +void rb_replace_node_rcu(struct rb_node *victim, struct rb_node *new, + struct rb_root *root) +{ + struct rb_node *parent = rb_parent(victim); /* Copy the pointers/colour from the victim to the replacement */ *new = *victim; + + /* Set the surrounding nodes to point to the replacement */ + if (victim->rb_left) + rb_set_parent(victim->rb_left, new); + if (victim->rb_right) + rb_set_parent(victim->rb_right, new); + + /* Set the parent's pointer to the new node last after an RCU barrier + * so that the pointers onwards are seen to be set correctly when doing + * an RCU walk over the tree. + */ + __rb_change_child_rcu(victim, new, parent, root); } -EXPORT_SYMBOL(rb_replace_node); +EXPORT_SYMBOL(rb_replace_node_rcu); static struct rb_node *rb_left_deepest_node(const struct rb_node *node) { -- cgit v0.10.2 From 995f1405610bd8446c5be37d2ffc031a7729e406 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 1 Jul 2016 13:44:53 -0700 Subject: rcu: Suppress sparse warnings for rcu_dereference_raw() Data structures that are used both with and without RCU protection are difficult to write in a sparse-clean manner. If you mark the relevant pointers with __rcu, sparse will complain about all non-RCU uses, but if you don't mark those pointers, sparse will complain about all RCU uses. This commit therefore suppresses sparse warnings for rcu_dereference_raw(), allowing mixed-protection data structures to avoid these warnings. Reported-by: David Howells Signed-off-by: Paul E. McKenney Signed-off-by: David Howells diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 5f1533e..85830e6 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -611,6 +611,12 @@ static inline void rcu_preempt_sleep_check(void) rcu_dereference_sparse(p, space); \ ((typeof(*p) __force __kernel *)(p)); \ }) +#define rcu_dereference_raw(p) \ +({ \ + /* Dependency order vs. p above. */ \ + typeof(p) ________p1 = lockless_dereference(p); \ + ((typeof(*p) __force __kernel *)(________p1)); \ +}) /** * RCU_INITIALIZER() - statically initialize an RCU-protected global variable @@ -729,8 +735,6 @@ static inline void rcu_preempt_sleep_check(void) __rcu_dereference_check((p), (c) || rcu_read_lock_sched_held(), \ __rcu) -#define rcu_dereference_raw(p) rcu_dereference_check(p, 1) /*@@@ needed? @@@*/ - /* * The tracing infrastructure traces RCU (we want that), but unfortunately * some of the RCU checks causes tracing to lock up the system. -- cgit v0.10.2 From 8496af50eb385c1cadff9ad396fd5359e96b6c27 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 1 Jul 2016 07:51:50 +0100 Subject: rxrpc: Use RCU to access a peer's service connection tree Move to using RCU access to a peer's service connection tree when routing an incoming packet. This is done using a seqlock to trigger retrying of the tree walk if a change happened. Further, we no longer get a ref on the connection looked up in the data_ready handler unless we queue the connection's work item - and then only if the refcount > 0. Note that I'm avoiding the use of a hash table for service connections because each service connection is addressed by a 62-bit number (constructed from epoch and connection ID >> 2) that would allow the client to engage in bucket stuffing, given knowledge of the hash algorithm. Peers, however, are hashed as the network address is less controllable by the client. The total number of peers will also be limited in a future commit. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 0fe63ba..e292276 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -10,6 +10,7 @@ */ #include +#include #include #include #include @@ -206,7 +207,7 @@ struct rxrpc_peer { struct hlist_head error_targets; /* targets for net error distribution */ struct work_struct error_distributor; struct rb_root service_conns; /* Service connections */ - rwlock_t conn_lock; + seqlock_t service_conn_lock; spinlock_t lock; /* access lock */ unsigned int if_mtu; /* interface MTU for this peer */ unsigned int mtu; /* network MTU for this peer */ @@ -559,12 +560,10 @@ extern unsigned int rxrpc_connection_expiry; extern struct list_head rxrpc_connections; extern rwlock_t rxrpc_connection_lock; -void rxrpc_conn_hash_proto_key(struct rxrpc_conn_proto *); -void rxrpc_extract_conn_params(struct rxrpc_conn_proto *, - struct rxrpc_local *, struct sk_buff *); +int rxrpc_extract_addr_from_skb(struct sockaddr_rxrpc *, struct sk_buff *); struct rxrpc_connection *rxrpc_alloc_connection(gfp_t); -struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *, - struct sk_buff *); +struct rxrpc_connection *rxrpc_find_connection_rcu(struct rxrpc_local *, + struct sk_buff *); void __rxrpc_disconnect_call(struct rxrpc_call *); void rxrpc_disconnect_call(struct rxrpc_call *); void rxrpc_put_connection(struct rxrpc_connection *); @@ -591,16 +590,20 @@ struct rxrpc_connection *rxrpc_get_connection_maybe(struct rxrpc_connection *con return atomic_inc_not_zero(&conn->usage) ? conn : NULL; } -static inline void rxrpc_queue_conn(struct rxrpc_connection *conn) +static inline bool rxrpc_queue_conn(struct rxrpc_connection *conn) { - if (rxrpc_get_connection_maybe(conn) && - !rxrpc_queue_work(&conn->processor)) + if (!rxrpc_get_connection_maybe(conn)) + return false; + if (!rxrpc_queue_work(&conn->processor)) rxrpc_put_connection(conn); + return true; } /* * conn_service.c */ +struct rxrpc_connection *rxrpc_find_service_conn_rcu(struct rxrpc_peer *, + struct sk_buff *); struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *, struct sockaddr_rxrpc *, struct sk_buff *); diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c index 917db48d..9e91f27 100644 --- a/net/rxrpc/conn_client.c +++ b/net/rxrpc/conn_client.c @@ -132,8 +132,6 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, gfp_t gfp) } conn->params = *cp; - conn->proto.epoch = rxrpc_epoch; - conn->proto.cid = 0; conn->out_clientflag = RXRPC_CLIENT_INITIATED; conn->state = RXRPC_CONN_CLIENT; diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index 1307138..896d844 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -15,7 +15,6 @@ #include #include #include -#include #include #include #include "ar-internal.h" @@ -64,24 +63,30 @@ struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp) } /* - * find a connection based on transport and RxRPC connection ID for an incoming - * packet + * Look up a connection in the cache by protocol parameters. + * + * If successful, a pointer to the connection is returned, but no ref is taken. + * NULL is returned if there is no match. + * + * The caller must be holding the RCU read lock. */ -struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *local, - struct sk_buff *skb) +struct rxrpc_connection *rxrpc_find_connection_rcu(struct rxrpc_local *local, + struct sk_buff *skb) { struct rxrpc_connection *conn; struct rxrpc_conn_proto k; struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct sockaddr_rxrpc srx; struct rxrpc_peer *peer; - struct rb_node *p; - _enter(",{%x,%x}", sp->hdr.cid, sp->hdr.flags); + _enter(",%x", sp->hdr.cid & RXRPC_CIDMASK); if (rxrpc_extract_addr_from_skb(&srx, skb) < 0) goto not_found; + k.epoch = sp->hdr.epoch; + k.cid = sp->hdr.cid & RXRPC_CIDMASK; + /* We may have to handle mixing IPv4 and IPv6 */ if (srx.transport.family != local->srx.transport.family) { pr_warn_ratelimited("AF_RXRPC: Protocol mismatch %u not %u\n", @@ -101,32 +106,23 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *local, peer = rxrpc_lookup_peer_rcu(local, &srx); if (!peer) goto not_found; - - read_lock_bh(&peer->conn_lock); - - p = peer->service_conns.rb_node; - while (p) { - conn = rb_entry(p, struct rxrpc_connection, service_node); - - _debug("maybe %x", conn->proto.cid); - - if (k.epoch < conn->proto.epoch) - p = p->rb_left; - else if (k.epoch > conn->proto.epoch) - p = p->rb_right; - else if (k.cid < conn->proto.cid) - p = p->rb_left; - else if (k.cid > conn->proto.cid) - p = p->rb_right; - else - goto found_service_conn; - } - read_unlock_bh(&peer->conn_lock); + conn = rxrpc_find_service_conn_rcu(peer, skb); + if (!conn || atomic_read(&conn->usage) == 0) + goto not_found; + _leave(" = %p", conn); + return conn; } else { + /* Look up client connections by connection ID alone as their + * IDs are unique for this machine. + */ conn = idr_find(&rxrpc_client_conn_ids, - k.cid >> RXRPC_CIDSHIFT); - if (!conn || - conn->proto.epoch != k.epoch || + sp->hdr.cid >> RXRPC_CIDSHIFT); + if (!conn || atomic_read(&conn->usage) == 0) { + _debug("no conn"); + goto not_found; + } + + if (conn->proto.epoch != k.epoch || conn->params.local != local) goto not_found; @@ -143,7 +139,6 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *local, BUG(); } - conn = rxrpc_get_connection_maybe(conn); _leave(" = %p", conn); return conn; } @@ -151,12 +146,6 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *local, not_found: _leave(" = NULL"); return NULL; - -found_service_conn: - conn = rxrpc_get_connection_maybe(conn); - read_unlock_bh(&peer->conn_lock); - _leave(" = %p", conn); - return conn; } /* diff --git a/net/rxrpc/conn_service.c b/net/rxrpc/conn_service.c index c6db2e8..7cbd612 100644 --- a/net/rxrpc/conn_service.c +++ b/net/rxrpc/conn_service.c @@ -13,18 +13,122 @@ #include "ar-internal.h" /* + * Find a service connection under RCU conditions. + * + * We could use a hash table, but that is subject to bucket stuffing by an + * attacker as the client gets to pick the epoch and cid values and would know + * the hash function. So, instead, we use a hash table for the peer and from + * that an rbtree to find the service connection. Under ordinary circumstances + * it might be slower than a large hash table, but it is at least limited in + * depth. + */ +struct rxrpc_connection *rxrpc_find_service_conn_rcu(struct rxrpc_peer *peer, + struct sk_buff *skb) +{ + struct rxrpc_connection *conn = NULL; + struct rxrpc_conn_proto k; + struct rxrpc_skb_priv *sp = rxrpc_skb(skb); + struct rb_node *p; + unsigned int seq = 0; + + k.epoch = sp->hdr.epoch; + k.cid = sp->hdr.cid & RXRPC_CIDMASK; + + do { + /* Unfortunately, rbtree walking doesn't give reliable results + * under just the RCU read lock, so we have to check for + * changes. + */ + read_seqbegin_or_lock(&peer->service_conn_lock, &seq); + + p = rcu_dereference_raw(peer->service_conns.rb_node); + while (p) { + conn = rb_entry(p, struct rxrpc_connection, service_node); + + if (conn->proto.index_key < k.index_key) + p = rcu_dereference_raw(p->rb_left); + else if (conn->proto.index_key > k.index_key) + p = rcu_dereference_raw(p->rb_right); + else + goto done; + conn = NULL; + } + } while (need_seqretry(&peer->service_conn_lock, seq)); + +done: + done_seqretry(&peer->service_conn_lock, seq); + _leave(" = %d", conn ? conn->debug_id : -1); + return conn; +} + +/* + * Insert a service connection into a peer's tree, thereby making it a target + * for incoming packets. + */ +static struct rxrpc_connection * +rxrpc_publish_service_conn(struct rxrpc_peer *peer, + struct rxrpc_connection *conn) +{ + struct rxrpc_connection *cursor = NULL; + struct rxrpc_conn_proto k = conn->proto; + struct rb_node **pp, *parent; + + write_seqlock_bh(&peer->service_conn_lock); + + pp = &peer->service_conns.rb_node; + parent = NULL; + while (*pp) { + parent = *pp; + cursor = rb_entry(parent, + struct rxrpc_connection, service_node); + + if (cursor->proto.index_key < k.index_key) + pp = &(*pp)->rb_left; + else if (cursor->proto.index_key > k.index_key) + pp = &(*pp)->rb_right; + else + goto found_extant_conn; + } + + rb_link_node_rcu(&conn->service_node, parent, pp); + rb_insert_color(&conn->service_node, &peer->service_conns); +conn_published: + set_bit(RXRPC_CONN_IN_SERVICE_CONNS, &conn->flags); + write_sequnlock_bh(&peer->service_conn_lock); + _leave(" = %d [new]", conn->debug_id); + return conn; + +found_extant_conn: + if (atomic_read(&cursor->usage) == 0) + goto replace_old_connection; + write_sequnlock_bh(&peer->service_conn_lock); + /* We should not be able to get here. rxrpc_incoming_connection() is + * called in a non-reentrant context, so there can't be a race to + * insert a new connection. + */ + BUG(); + +replace_old_connection: + /* The old connection is from an outdated epoch. */ + _debug("replace conn"); + rb_replace_node_rcu(&cursor->service_node, + &conn->service_node, + &peer->service_conns); + clear_bit(RXRPC_CONN_IN_SERVICE_CONNS, &cursor->flags); + goto conn_published; +} + +/* * get a record of an incoming connection */ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, struct sockaddr_rxrpc *srx, struct sk_buff *skb) { - struct rxrpc_connection *conn, *candidate = NULL; + struct rxrpc_connection *conn; struct rxrpc_skb_priv *sp = rxrpc_skb(skb); struct rxrpc_peer *peer; - struct rb_node *p, **pp; const char *new = "old"; - u32 epoch, cid; _enter(""); @@ -36,131 +140,79 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, ASSERT(sp->hdr.flags & RXRPC_CLIENT_INITIATED); - epoch = sp->hdr.epoch; - cid = sp->hdr.cid & RXRPC_CIDMASK; - - /* search the connection list first */ - read_lock_bh(&peer->conn_lock); - - p = peer->service_conns.rb_node; - while (p) { - conn = rb_entry(p, struct rxrpc_connection, service_node); - - _debug("maybe %x", conn->proto.cid); - - if (epoch < conn->proto.epoch) - p = p->rb_left; - else if (epoch > conn->proto.epoch) - p = p->rb_right; - else if (cid < conn->proto.cid) - p = p->rb_left; - else if (cid > conn->proto.cid) - p = p->rb_right; - else - goto found_extant_connection; + rcu_read_lock(); + peer = rxrpc_lookup_peer_rcu(local, srx); + if (peer) { + conn = rxrpc_find_service_conn_rcu(peer, skb); + if (conn) { + if (sp->hdr.securityIndex != conn->security_ix) + goto security_mismatch_rcu; + if (rxrpc_get_connection_maybe(conn)) + goto found_extant_connection_rcu; + + /* The conn has expired but we can't remove it without + * the appropriate lock, so we attempt to replace it + * when we have a new candidate. + */ + } + + if (!rxrpc_get_peer_maybe(peer)) + peer = NULL; } - read_unlock_bh(&peer->conn_lock); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_connection(GFP_NOIO); - if (!candidate) { - rxrpc_put_peer(peer); - _leave(" = -ENOMEM"); - return ERR_PTR(-ENOMEM); - } - - candidate->proto.epoch = sp->hdr.epoch; - candidate->proto.cid = sp->hdr.cid & RXRPC_CIDMASK; - candidate->params.local = local; - candidate->params.peer = peer; - candidate->params.service_id = sp->hdr.serviceId; - candidate->security_ix = sp->hdr.securityIndex; - candidate->out_clientflag = 0; - candidate->state = RXRPC_CONN_SERVICE; - if (candidate->params.service_id) - candidate->state = RXRPC_CONN_SERVICE_UNSECURED; - - write_lock_bh(&peer->conn_lock); - - pp = &peer->service_conns.rb_node; - p = NULL; - while (*pp) { - p = *pp; - conn = rb_entry(p, struct rxrpc_connection, service_node); + rcu_read_unlock(); - if (epoch < conn->proto.epoch) - pp = &(*pp)->rb_left; - else if (epoch > conn->proto.epoch) - pp = &(*pp)->rb_right; - else if (cid < conn->proto.cid) - pp = &(*pp)->rb_left; - else if (cid > conn->proto.cid) - pp = &(*pp)->rb_right; - else - goto found_extant_second; + if (!peer) { + peer = rxrpc_lookup_peer(local, srx, GFP_NOIO); + if (IS_ERR(peer)) + goto enomem; } - /* we can now add the new candidate to the list */ - set_bit(RXRPC_CONN_IN_SERVICE_CONNS, &candidate->flags); - rb_link_node(&candidate->service_node, p, pp); - rb_insert_color(&candidate->service_node, &peer->service_conns); -attached: - conn = candidate; - candidate = NULL; - rxrpc_get_peer(peer); - rxrpc_get_local(local); + /* We don't have a matching record yet. */ + conn = rxrpc_alloc_connection(GFP_NOIO); + if (!conn) + goto enomem_peer; + + conn->proto.epoch = sp->hdr.epoch; + conn->proto.cid = sp->hdr.cid & RXRPC_CIDMASK; + conn->params.local = local; + conn->params.peer = peer; + conn->params.service_id = sp->hdr.serviceId; + conn->security_ix = sp->hdr.securityIndex; + conn->out_clientflag = 0; + conn->state = RXRPC_CONN_SERVICE; + if (conn->params.service_id) + conn->state = RXRPC_CONN_SERVICE_UNSECURED; - write_unlock_bh(&peer->conn_lock); + rxrpc_get_local(local); write_lock(&rxrpc_connection_lock); list_add_tail(&conn->link, &rxrpc_connections); write_unlock(&rxrpc_connection_lock); + /* Make the connection a target for incoming packets. */ + rxrpc_publish_service_conn(peer, conn); + new = "new"; success: _net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->proto.cid); - - rxrpc_put_peer(peer); _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage)); return conn; - /* we found the connection in the list immediately */ -found_extant_connection: - if (!rxrpc_get_connection_maybe(conn)) { - set_bit(RXRPC_CONN_IN_SERVICE_CONNS, &candidate->flags); - rb_replace_node(&conn->service_node, - &candidate->service_node, - &peer->service_conns); - clear_bit(RXRPC_CONN_IN_SERVICE_CONNS, &conn->flags); - goto attached; - } - - if (sp->hdr.securityIndex != conn->security_ix) { - read_unlock_bh(&peer->conn_lock); - goto security_mismatch_put; - } - read_unlock_bh(&peer->conn_lock); - goto success; - - /* we found the connection on the second time through the list */ -found_extant_second: - if (sp->hdr.securityIndex != conn->security_ix) { - write_unlock_bh(&peer->conn_lock); - goto security_mismatch; - } - rxrpc_get_connection(conn); - write_unlock_bh(&peer->conn_lock); - kfree(candidate); +found_extant_connection_rcu: + rcu_read_unlock(); goto success; -security_mismatch_put: - rxrpc_put_connection(conn); -security_mismatch: - kfree(candidate); +security_mismatch_rcu: + rcu_read_unlock(); _leave(" = -EKEYREJECTED"); return ERR_PTR(-EKEYREJECTED); + +enomem_peer: + rxrpc_put_peer(peer); +enomem: + _leave(" = -ENOMEM"); + return ERR_PTR(-ENOMEM); } /* @@ -171,8 +223,8 @@ void rxrpc_unpublish_service_conn(struct rxrpc_connection *conn) { struct rxrpc_peer *peer = conn->params.peer; - write_lock_bh(&peer->conn_lock); + write_seqlock_bh(&peer->service_conn_lock); if (test_and_clear_bit(RXRPC_CONN_IN_SERVICE_CONNS, &conn->flags)) rb_erase(&conn->service_node, &peer->service_conns); - write_unlock_bh(&peer->conn_lock); + write_sequnlock_bh(&peer->service_conn_lock); } diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index c243647..991a20d 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -575,13 +575,13 @@ done: * post connection-level events to the connection * - this includes challenges, responses and some aborts */ -static void rxrpc_post_packet_to_conn(struct rxrpc_connection *conn, +static bool rxrpc_post_packet_to_conn(struct rxrpc_connection *conn, struct sk_buff *skb) { _enter("%p,%p", conn, skb); skb_queue_tail(&conn->rx_queue, skb); - rxrpc_queue_conn(conn); + return rxrpc_queue_conn(conn); } /* @@ -636,6 +636,7 @@ int rxrpc_extract_header(struct rxrpc_skb_priv *sp, struct sk_buff *skb) */ void rxrpc_data_ready(struct sock *sk) { + struct rxrpc_connection *conn; struct rxrpc_skb_priv *sp; struct rxrpc_local *local = sk->sk_user_data; struct sk_buff *skb; @@ -699,36 +700,37 @@ void rxrpc_data_ready(struct sock *sk) (sp->hdr.callNumber == 0 || sp->hdr.seq == 0)) goto bad_message; - if (sp->hdr.callNumber == 0) { - /* This is a connection-level packet. These should be - * fairly rare, so the extra overhead of looking them up the - * old-fashioned way doesn't really hurt */ - struct rxrpc_connection *conn; - - rcu_read_lock(); - conn = rxrpc_find_connection(local, skb); - rcu_read_unlock(); - if (!conn) - goto cant_route_call; + rcu_read_lock(); + +retry_find_conn: + conn = rxrpc_find_connection_rcu(local, skb); + if (!conn) + goto cant_route_call; + if (sp->hdr.callNumber == 0) { + /* Connection-level packet */ _debug("CONN %p {%d}", conn, conn->debug_id); - rxrpc_post_packet_to_conn(conn, skb); - rxrpc_put_connection(conn); + if (!rxrpc_post_packet_to_conn(conn, skb)) + goto retry_find_conn; } else { - struct rxrpc_call *call; + /* Call-bound packets are routed by connection channel. */ + unsigned int channel = sp->hdr.cid & RXRPC_CHANNELMASK; + struct rxrpc_channel *chan = &conn->channels[channel]; + struct rxrpc_call *call = rcu_dereference(chan->call); - call = rxrpc_find_call_hash(&sp->hdr, local, - AF_INET, &ip_hdr(skb)->saddr); - if (call) - rxrpc_post_packet_to_call(call, skb); - else + if (!call || atomic_read(&call->usage) == 0) goto cant_route_call; + + rxrpc_post_packet_to_call(call, skb); } + rcu_read_unlock(); out: return; cant_route_call: + rcu_read_unlock(); + _debug("can't route call"); if (sp->hdr.flags & RXRPC_CLIENT_INITIATED && sp->hdr.type == RXRPC_PACKET_TYPE_DATA) { diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c index 01d4930..538e983 100644 --- a/net/rxrpc/peer_object.c +++ b/net/rxrpc/peer_object.c @@ -189,7 +189,7 @@ struct rxrpc_peer *rxrpc_alloc_peer(struct rxrpc_local *local, gfp_t gfp) INIT_WORK(&peer->error_distributor, &rxrpc_peer_error_distributor); peer->service_conns = RB_ROOT; - rwlock_init(&peer->conn_lock); + seqlock_init(&peer->service_conn_lock); spin_lock_init(&peer->lock); peer->debug_id = atomic_inc_return(&rxrpc_debug_id); } -- cgit v0.10.2 From d440a1ce5d2cf9d90390f6c0d8badc4c0a4f8b6b Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 5 Jul 2016 10:57:10 +0100 Subject: rxrpc: Kill off the call hash table The call hash table is now no longer used as calls are looked up directly by channel slot on the connection, so kill it off. Signed-off-by: David Howells diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index e292276..1bb9e7a 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -461,19 +461,12 @@ struct rxrpc_call { #define RXRPC_ACKR_WINDOW_ASZ DIV_ROUND_UP(RXRPC_MAXACKS, BITS_PER_LONG) unsigned long ackr_window[RXRPC_ACKR_WINDOW_ASZ + 1]; - struct hlist_node hash_node; - unsigned long hash_key; /* Full hash key */ - u8 in_clientflag; /* Copy of conn->in_clientflag for hashing */ - struct rxrpc_local *local; /* Local endpoint. Used for hashing. */ - sa_family_t family; /* Frame protocol */ + u8 in_clientflag; /* Copy of conn->in_clientflag */ + struct rxrpc_local *local; /* Local endpoint. */ u32 call_id; /* call ID on connection */ u32 cid; /* connection ID plus channel index */ u32 epoch; /* epoch of this connection */ u16 service_id; /* service ID */ - union { /* Peer IP address for hashing */ - __be32 ipv4_addr; - __u8 ipv6_addr[16]; /* Anticipates eventual IPv6 support */ - } peer_ip; }; /* @@ -521,8 +514,6 @@ extern struct kmem_cache *rxrpc_call_jar; extern struct list_head rxrpc_calls; extern rwlock_t rxrpc_call_lock; -struct rxrpc_call *rxrpc_find_call_hash(struct rxrpc_host_header *, - void *, sa_family_t, const void *); struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *, unsigned long); struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *, struct rxrpc_conn_parameters *, diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index ebbd7dd..91287c9 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -14,7 +14,6 @@ #include #include #include -#include #include #include #include @@ -61,142 +60,6 @@ static void rxrpc_dead_call_expired(unsigned long _call); static void rxrpc_ack_time_expired(unsigned long _call); static void rxrpc_resend_time_expired(unsigned long _call); -static DEFINE_SPINLOCK(rxrpc_call_hash_lock); -static DEFINE_HASHTABLE(rxrpc_call_hash, 10); - -/* - * Hash function for rxrpc_call_hash - */ -static unsigned long rxrpc_call_hashfunc( - u8 in_clientflag, - u32 cid, - u32 call_id, - u32 epoch, - u16 service_id, - sa_family_t family, - void *localptr, - unsigned int addr_size, - const u8 *peer_addr) -{ - const u16 *p; - unsigned int i; - unsigned long key; - - _enter(""); - - key = (unsigned long)localptr; - /* We just want to add up the __be32 values, so forcing the - * cast should be okay. - */ - key += epoch; - key += service_id; - key += call_id; - key += (cid & RXRPC_CIDMASK) >> RXRPC_CIDSHIFT; - key += cid & RXRPC_CHANNELMASK; - key += in_clientflag; - key += family; - /* Step through the peer address in 16-bit portions for speed */ - for (i = 0, p = (const u16 *)peer_addr; i < addr_size >> 1; i++, p++) - key += *p; - _leave(" key = 0x%lx", key); - return key; -} - -/* - * Add a call to the hashtable - */ -static void rxrpc_call_hash_add(struct rxrpc_call *call) -{ - unsigned long key; - unsigned int addr_size = 0; - - _enter(""); - switch (call->family) { - case AF_INET: - addr_size = sizeof(call->peer_ip.ipv4_addr); - break; - case AF_INET6: - addr_size = sizeof(call->peer_ip.ipv6_addr); - break; - default: - break; - } - key = rxrpc_call_hashfunc(call->in_clientflag, call->cid, - call->call_id, call->epoch, - call->service_id, call->family, - call->conn->params.local, addr_size, - call->peer_ip.ipv6_addr); - /* Store the full key in the call */ - call->hash_key = key; - spin_lock(&rxrpc_call_hash_lock); - hash_add_rcu(rxrpc_call_hash, &call->hash_node, key); - spin_unlock(&rxrpc_call_hash_lock); - _leave(""); -} - -/* - * Remove a call from the hashtable - */ -static void rxrpc_call_hash_del(struct rxrpc_call *call) -{ - _enter(""); - spin_lock(&rxrpc_call_hash_lock); - hash_del_rcu(&call->hash_node); - spin_unlock(&rxrpc_call_hash_lock); - _leave(""); -} - -/* - * Find a call in the hashtable and return it, or NULL if it - * isn't there. - */ -struct rxrpc_call *rxrpc_find_call_hash( - struct rxrpc_host_header *hdr, - void *localptr, - sa_family_t family, - const void *peer_addr) -{ - unsigned long key; - unsigned int addr_size = 0; - struct rxrpc_call *call = NULL; - struct rxrpc_call *ret = NULL; - u8 in_clientflag = hdr->flags & RXRPC_CLIENT_INITIATED; - - _enter(""); - switch (family) { - case AF_INET: - addr_size = sizeof(call->peer_ip.ipv4_addr); - break; - case AF_INET6: - addr_size = sizeof(call->peer_ip.ipv6_addr); - break; - default: - break; - } - - key = rxrpc_call_hashfunc(in_clientflag, hdr->cid, hdr->callNumber, - hdr->epoch, hdr->serviceId, - family, localptr, addr_size, - peer_addr); - hash_for_each_possible_rcu(rxrpc_call_hash, call, hash_node, key) { - if (call->hash_key == key && - call->call_id == hdr->callNumber && - call->cid == hdr->cid && - call->in_clientflag == in_clientflag && - call->service_id == hdr->serviceId && - call->family == family && - call->local == localptr && - memcmp(call->peer_ip.ipv6_addr, peer_addr, - addr_size) == 0 && - call->epoch == hdr->epoch) { - ret = call; - break; - } - } - _leave(" = %p", ret); - return ret; -} - /* * find an extant server call * - called in process context with IRQs enabled @@ -305,20 +168,7 @@ static struct rxrpc_call *rxrpc_alloc_client_call(struct rxrpc_sock *rx, call->socket = rx; call->rx_data_post = 1; - /* Record copies of information for hashtable lookup */ - call->family = rx->family; call->local = rx->local; - switch (call->family) { - case AF_INET: - call->peer_ip.ipv4_addr = srx->transport.sin.sin_addr.s_addr; - break; - case AF_INET6: - memcpy(call->peer_ip.ipv6_addr, - srx->transport.sin6.sin6_addr.in6_u.u6_addr8, - sizeof(call->peer_ip.ipv6_addr)); - break; - } - call->service_id = srx->srx_service; call->in_clientflag = 0; @@ -345,9 +195,6 @@ static int rxrpc_begin_client_call(struct rxrpc_call *call, call->state = RXRPC_CALL_CLIENT_SEND_REQUEST; - /* Add the new call to the hashtable */ - rxrpc_call_hash_add(call); - spin_lock(&call->conn->params.peer->lock); hlist_add_head(&call->error_link, &call->conn->params.peer->error_targets); spin_unlock(&call->conn->params.peer->lock); @@ -548,27 +395,10 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx, list_add_tail(&call->link, &rxrpc_calls); write_unlock_bh(&rxrpc_call_lock); - /* Record copies of information for hashtable lookup */ - call->family = rx->family; call->local = conn->params.local; - switch (call->family) { - case AF_INET: - call->peer_ip.ipv4_addr = - conn->params.peer->srx.transport.sin.sin_addr.s_addr; - break; - case AF_INET6: - memcpy(call->peer_ip.ipv6_addr, - conn->params.peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8, - sizeof(call->peer_ip.ipv6_addr)); - break; - default: - break; - } call->epoch = conn->proto.epoch; call->service_id = conn->params.service_id; call->in_clientflag = RXRPC_CLIENT_INITIATED; - /* Add the new call to the hashtable */ - rxrpc_call_hash_add(call); _net("CALL incoming %d on CONN %d", call->debug_id, call->conn->debug_id); @@ -818,9 +648,6 @@ static void rxrpc_cleanup_call(struct rxrpc_call *call) ASSERTCMP(call->conn, ==, NULL); - /* Remove the call from the hash */ - rxrpc_call_hash_del(call); - if (call->acks_window) { _debug("kill Tx window %d", CIRC_CNT(call->acks_head, call->acks_tail, -- cgit v0.10.2 From f89e07d4cf2660a2956bc350a201398dda85284e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 6 Jul 2016 14:44:14 +0200 Subject: mac80211: agg-rx: refuse ADDBA Request with timeout update The current implementation of handling ADDBA Request while a session is already active with the peer is wrong - in case the peer is using the existing session's dialog token this should be treated as update to the session, which can update the timeout value. We don't really have a good way of supporting that, so reject, but implement the required behaviour in the spec of "Even if the updated ADDBA Request frame is not accepted, the original Block ACK setup remains active." (802.11-2012 10.5.4) Signed-off-by: Johannes Berg diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c index 3a8f881..a9aff60 100644 --- a/net/mac80211/agg-rx.c +++ b/net/mac80211/agg-rx.c @@ -306,6 +306,24 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta, mutex_lock(&sta->ampdu_mlme.mtx); if (test_bit(tid, sta->ampdu_mlme.agg_session_valid)) { + tid_agg_rx = rcu_dereference_protected( + sta->ampdu_mlme.tid_rx[tid], + lockdep_is_held(&sta->ampdu_mlme.mtx)); + + if (tid_agg_rx->dialog_token == dialog_token) { + ht_dbg_ratelimited(sta->sdata, + "updated AddBA Req from %pM on tid %u\n", + sta->sta.addr, tid); + /* We have no API to update the timeout value in the + * driver so reject the timeout update. + */ + status = WLAN_STATUS_REQUEST_DECLINED; + ieee80211_send_addba_resp(sta->sdata, sta->sta.addr, + tid, dialog_token, status, + 1, buf_size, timeout); + goto end; + } + ht_dbg_ratelimited(sta->sdata, "unexpected AddBA Req from %pM on tid %u\n", sta->sta.addr, tid); -- cgit v0.10.2 From c6e6a0c8be575c830a97b1942dabeab70f423fe0 Mon Sep 17 00:00:00 2001 From: Aviya Erenfeld Date: Tue, 5 Jul 2016 15:23:08 +0300 Subject: nl80211: Add API to support VHT MU-MIMO air sniffer add API to support VHT MU-MIMO air sniffer. in MU-MIMO there are parallel frames on the air while the HW has only one RX. add the capability to sniff one of the MU-MIMO parallel frames by giving the sniffer additional information so it'll know which of the parallel frames it shall follow. Add attribute - NL80211_ATTR_MU_MIMO_GROUP_DATA - for getting a MU-MIMO groupID in order to monitor packets from that group using VHT MU-MIMO. And add attribute -NL80211_ATTR_MU_MIMO_FOLLOW_ADDR - for passing MAC address to monitor mode. that option will be used by VHT MU-MIMO air sniffer to follow a station according to it's MAC address using VHT MU-MIMO. Signed-off-by: Aviya Erenfeld Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 7bbb00d..fa4f0f7 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -330,6 +330,9 @@ struct ieee80211_supported_band { * in a separate chapter. */ +#define VHT_MUMIMO_GROUPS_DATA_LEN (WLAN_MEMBERSHIP_LEN +\ + WLAN_USER_POSITION_LEN) + /** * struct vif_params - describes virtual interface parameters * @use_4addr: use 4-address frames @@ -339,10 +342,13 @@ struct ieee80211_supported_band { * This feature is only fully supported by drivers that enable the * %NL80211_FEATURE_MAC_ON_CREATE flag. Others may support creating ** only p2p devices with specified MAC. + * @vht_mumimo_groups: MU-MIMO groupID. used for monitoring only + * packets belonging to that MU-MIMO groupID. */ struct vif_params { - int use_4addr; - u8 macaddr[ETH_ALEN]; + int use_4addr; + u8 macaddr[ETH_ALEN]; + u8 vht_mumimo_groups[VHT_MUMIMO_GROUPS_DATA_LEN]; }; /** diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 53c8278..1d7da78 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1829,6 +1829,25 @@ enum nl80211_commands { * %NL80211_ATTR_EXT_CAPA_MASK, to specify the extended capabilities per * interface type. * + * @NL80211_ATTR_MU_MIMO_GROUP_DATA: array of 24 bytes that defines a MU-MIMO + * groupID for monitor mode. + * The first 8 bytes are a mask that defines the membership in each + * group (there are 64 groups, group 0 and 63 are reserved), + * each bit represents a group and set to 1 for being a member in + * that group and 0 for not being a member. + * The remaining 16 bytes define the position in each group: 2 bits for + * each group. + * (smaller group numbers represented on most significant bits and bigger + * group numbers on least significant bits.) + * This attribute is used only if all interfaces are in monitor mode. + * Set this attribute in order to monitor packets using the given MU-MIMO + * groupID data. + * to turn off that feature set all the bits of the groupID to zero. + * @NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR: mac address for the sniffer to follow + * when using MU-MIMO air sniffer. + * to turn that feature off set an invalid mac address + * (e.g. FF:FF:FF:FF:FF:FF) + * * @NUM_NL80211_ATTR: total number of nl80211_attrs available * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use @@ -2213,6 +2232,9 @@ enum nl80211_attrs { NL80211_ATTR_IFTYPE_EXT_CAPA, + NL80211_ATTR_MU_MIMO_GROUP_DATA, + NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, @@ -4479,6 +4501,12 @@ enum nl80211_feature_flags { * %NL80211_CMD_ASSOCIATE and %NL80211_CMD_CONNECT requests, which will set * the ASSOC_REQ_USE_RRM flag in the association request even if * NL80211_FEATURE_QUIET is not advertized. + * @NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER: This device supports MU-MIMO air + * sniffer which means that it can be configured to hear packets from + * certain groups which can be configured by the + * %NL80211_ATTR_MU_MIMO_GROUP_DATA attribute, + * or can be configured to follow a station by configuring the + * %NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR attribute. * * @NUM_NL80211_EXT_FEATURES: number of extended features. * @MAX_NL80211_EXT_FEATURES: highest extended feature index. @@ -4486,6 +4514,7 @@ enum nl80211_feature_flags { enum nl80211_ext_feature_index { NL80211_EXT_FEATURE_VHT_IBSS, NL80211_EXT_FEATURE_RRM, + NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER, /* add new features before the definition below */ NUM_NL80211_EXT_FEATURES, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 244d552..447026f 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -405,6 +405,10 @@ static const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = { [NL80211_ATTR_PBSS] = { .type = NLA_FLAG }, [NL80211_ATTR_BSS_SELECT] = { .type = NLA_NESTED }, [NL80211_ATTR_STA_SUPPORT_P2P_PS] = { .type = NLA_U8 }, + [NL80211_ATTR_MU_MIMO_GROUP_DATA] = { + .len = VHT_MUMIMO_GROUPS_DATA_LEN + }, + [NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR] = { .len = ETH_ALEN }, }; /* policy for the key attributes */ @@ -2695,6 +2699,38 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info) change = true; } + if (info->attrs[NL80211_ATTR_MU_MIMO_GROUP_DATA]) { + const u8 *mumimo_groups; + u32 cap_flag = NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER; + + if (!wiphy_ext_feature_isset(&rdev->wiphy, cap_flag)) + return -EOPNOTSUPP; + + mumimo_groups = + nla_data(info->attrs[NL80211_ATTR_MU_MIMO_GROUP_DATA]); + + /* bits 0 and 63 are reserved and must be zero */ + if ((mumimo_groups[0] & BIT(7)) || + (mumimo_groups[VHT_MUMIMO_GROUPS_DATA_LEN - 1] & BIT(0))) + return -EINVAL; + + memcpy(params.vht_mumimo_groups, mumimo_groups, + VHT_MUMIMO_GROUPS_DATA_LEN); + change = true; + } + + if (info->attrs[NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR]) { + u32 cap_flag = NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER; + + if (!wiphy_ext_feature_isset(&rdev->wiphy, cap_flag)) + return -EOPNOTSUPP; + + nla_memcpy(params.macaddr, + info->attrs[NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR], + ETH_ALEN); + change = true; + } + if (flags && (*flags & MONITOR_FLAG_ACTIVE) && !(rdev->wiphy.features & NL80211_FEATURE_ACTIVE_MONITOR)) return -EOPNOTSUPP; -- cgit v0.10.2 From f7736f501f87b3f6ffbf07d173a6893dd419ac50 Mon Sep 17 00:00:00 2001 From: Ilan Peer Date: Tue, 5 Jul 2016 15:23:09 +0300 Subject: mac80211_hwsim: Add radar bandwidths to the P2P Device combination Add radar_detect_widths to the interface combination that allows concurrent P2P Device dedicated interface and AP interfaces, to enable testing of radar detection when P2P Device interface is used. Clear the radar_detect_widths in case of multi channel contexts as this is not currently supported. As radar_detect_widths are now supported in all combinations, remove the hwsim_if_dfs_limits definition since it is no longer needed. Signed-off-by: Ilan Peer Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 382109bb..97196c4 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -457,10 +457,6 @@ static const struct ieee80211_iface_limit hwsim_if_limits[] = { { .max = 1, .types = BIT(NL80211_IFTYPE_P2P_DEVICE) } }; -static const struct ieee80211_iface_limit hwsim_if_dfs_limits[] = { - { .max = 8, .types = BIT(NL80211_IFTYPE_AP) }, -}; - static const struct ieee80211_iface_combination hwsim_if_comb[] = { { .limits = hwsim_if_limits, @@ -468,18 +464,12 @@ static const struct ieee80211_iface_combination hwsim_if_comb[] = { .n_limits = ARRAY_SIZE(hwsim_if_limits) - 1, .max_interfaces = 2048, .num_different_channels = 1, - }, - { - .limits = hwsim_if_dfs_limits, - .n_limits = ARRAY_SIZE(hwsim_if_dfs_limits), - .max_interfaces = 8, - .num_different_channels = 1, .radar_detect_widths = BIT(NL80211_CHAN_WIDTH_20_NOHT) | BIT(NL80211_CHAN_WIDTH_20) | BIT(NL80211_CHAN_WIDTH_40) | BIT(NL80211_CHAN_WIDTH_80) | BIT(NL80211_CHAN_WIDTH_160), - } + }, }; static const struct ieee80211_iface_combination hwsim_if_comb_p2p_dev[] = { @@ -488,18 +478,12 @@ static const struct ieee80211_iface_combination hwsim_if_comb_p2p_dev[] = { .n_limits = ARRAY_SIZE(hwsim_if_limits), .max_interfaces = 2048, .num_different_channels = 1, - }, - { - .limits = hwsim_if_dfs_limits, - .n_limits = ARRAY_SIZE(hwsim_if_dfs_limits), - .max_interfaces = 8, - .num_different_channels = 1, .radar_detect_widths = BIT(NL80211_CHAN_WIDTH_20_NOHT) | BIT(NL80211_CHAN_WIDTH_20) | BIT(NL80211_CHAN_WIDTH_40) | BIT(NL80211_CHAN_WIDTH_80) | BIT(NL80211_CHAN_WIDTH_160), - } + }, }; static spinlock_t hwsim_radio_lock; @@ -2487,13 +2471,14 @@ static int mac80211_hwsim_new_radio(struct genl_info *info, hw->wiphy->max_scan_ssids = 255; hw->wiphy->max_scan_ie_len = IEEE80211_MAX_DATA_LEN; hw->wiphy->max_remain_on_channel_duration = 1000; - /* For channels > 1 DFS is not allowed */ - hw->wiphy->n_iface_combinations = 1; hw->wiphy->iface_combinations = &data->if_combination; if (param->p2p_device) data->if_combination = hwsim_if_comb_p2p_dev[0]; else data->if_combination = hwsim_if_comb[0]; + hw->wiphy->n_iface_combinations = 1; + /* For channels > 1 DFS is not allowed */ + data->if_combination.radar_detect_widths = 0; data->if_combination.num_different_channels = data->channels; } else if (param->p2p_device) { hw->wiphy->iface_combinations = hwsim_if_comb_p2p_dev; -- cgit v0.10.2 From f1724b0258b412144964f2074b9b6edcf0dce87c Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 6 Jul 2016 14:47:49 +0200 Subject: mac80211_hwsim: use signed net namespace ID The API expects a pointer to a signed int so we should not use an unsigned int for it. Signed-off-by: Johannes Berg diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 97196c4..337794f 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -250,7 +250,7 @@ static inline void hwsim_clear_chanctx_magic(struct ieee80211_chanctx_conf *c) cp->magic = 0; } -static unsigned int hwsim_net_id; +static int hwsim_net_id; static int hwsim_netgroup; -- cgit v0.10.2 From 1d76250bd34af86c6498fc51e50cab3bfbbeceaa Mon Sep 17 00:00:00 2001 From: Avraham Stern Date: Tue, 5 Jul 2016 17:10:13 +0300 Subject: nl80211: support beacon report scanning Beacon report radio measurement requires reporting observed BSSs on the channels specified in the beacon request. If the measurement mode is set to passive or active, it requires actually performing a scan (passive or active, accordingly), and reporting the time that the scan was started and the time each beacon/probe was received (both in terms of TSF of the BSS of the requesting AP). If the request mode is table, this information is optional. In addition, the radio measurement request specifies the channel dwell time for the measurement. In order to use scan for beacon report when the mode is active or passive, add a parameter to scan request that specifies the channel dwell time, and add scan start time and beacon received time to scan results information. Supporting beacon report is required for Multi Band Operation (MBO). Signed-off-by: Assaf Krauss Signed-off-by: David Spinadel Signed-off-by: Avraham Stern Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.c b/drivers/net/wireless/ath/ath6kl/cfg80211.c index 4e11ba0..ef5b40e 100644 --- a/drivers/net/wireless/ath/ath6kl/cfg80211.c +++ b/drivers/net/wireless/ath/ath6kl/cfg80211.c @@ -859,7 +859,11 @@ void ath6kl_cfg80211_disconnect_event(struct ath6kl_vif *vif, u8 reason, struct ath6kl *ar = vif->ar; if (vif->scan_req) { - cfg80211_scan_done(vif->scan_req, true); + struct cfg80211_scan_info info = { + .aborted = true, + }; + + cfg80211_scan_done(vif->scan_req, &info); vif->scan_req = NULL; } @@ -1069,6 +1073,9 @@ static int ath6kl_cfg80211_scan(struct wiphy *wiphy, void ath6kl_cfg80211_scan_complete_event(struct ath6kl_vif *vif, bool aborted) { struct ath6kl *ar = vif->ar; + struct cfg80211_scan_info info = { + .aborted = aborted, + }; int i; ath6kl_dbg(ATH6KL_DBG_WLAN_CFG, "%s: status%s\n", __func__, @@ -1089,7 +1096,7 @@ void ath6kl_cfg80211_scan_complete_event(struct ath6kl_vif *vif, bool aborted) } out: - cfg80211_scan_done(vif->scan_req, aborted); + cfg80211_scan_done(vif->scan_req, &info); vif->scan_req = NULL; } @@ -3614,7 +3621,11 @@ void ath6kl_cfg80211_vif_stop(struct ath6kl_vif *vif, bool wmi_ready) } if (vif->scan_req) { - cfg80211_scan_done(vif->scan_req, true); + struct cfg80211_scan_info info = { + .aborted = true, + }; + + cfg80211_scan_done(vif->scan_req, &info); vif->scan_req = NULL; } diff --git a/drivers/net/wireless/ath/wil6210/cfg80211.c b/drivers/net/wireless/ath/wil6210/cfg80211.c index 62bf933..f0e1175 100644 --- a/drivers/net/wireless/ath/wil6210/cfg80211.c +++ b/drivers/net/wireless/ath/wil6210/cfg80211.c @@ -1369,7 +1369,11 @@ static void wil_cfg80211_stop_p2p_device(struct wiphy *wiphy, mutex_lock(&wil->mutex); started = wil_p2p_stop_discovery(wil); if (started && wil->scan_request) { - cfg80211_scan_done(wil->scan_request, 1); + struct cfg80211_scan_info info = { + .aborted = true, + }; + + cfg80211_scan_done(wil->scan_request, &info); wil->scan_request = NULL; wil->radio_wdev = wil->wdev; } diff --git a/drivers/net/wireless/ath/wil6210/main.c b/drivers/net/wireless/ath/wil6210/main.c index 8e31d75..4bc92e5 100644 --- a/drivers/net/wireless/ath/wil6210/main.c +++ b/drivers/net/wireless/ath/wil6210/main.c @@ -850,10 +850,14 @@ int wil_reset(struct wil6210_priv *wil, bool load_fw) mutex_unlock(&wil->wmi_mutex); if (wil->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + wil_dbg_misc(wil, "Abort scan_request 0x%p\n", wil->scan_request); del_timer_sync(&wil->scan_timer); - cfg80211_scan_done(wil->scan_request, true); + cfg80211_scan_done(wil->scan_request, &info); wil->scan_request = NULL; } @@ -1049,10 +1053,14 @@ int __wil_down(struct wil6210_priv *wil) (void)wil_p2p_stop_discovery(wil); if (wil->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + wil_dbg_misc(wil, "Abort scan_request 0x%p\n", wil->scan_request); del_timer_sync(&wil->scan_timer); - cfg80211_scan_done(wil->scan_request, true); + cfg80211_scan_done(wil->scan_request, &info); wil->scan_request = NULL; } diff --git a/drivers/net/wireless/ath/wil6210/p2p.c b/drivers/net/wireless/ath/wil6210/p2p.c index 213b825..e0f8aa0 100644 --- a/drivers/net/wireless/ath/wil6210/p2p.c +++ b/drivers/net/wireless/ath/wil6210/p2p.c @@ -252,8 +252,12 @@ void wil_p2p_search_expired(struct work_struct *work) mutex_unlock(&wil->mutex); if (started) { + struct cfg80211_scan_info info = { + .aborted = false, + }; + mutex_lock(&wil->p2p_wdev_mutex); - cfg80211_scan_done(wil->scan_request, 0); + cfg80211_scan_done(wil->scan_request, &info); wil->scan_request = NULL; wil->radio_wdev = wil->wdev; mutex_unlock(&wil->p2p_wdev_mutex); diff --git a/drivers/net/wireless/ath/wil6210/wmi.c b/drivers/net/wireless/ath/wil6210/wmi.c index b80c5d8..4d92541 100644 --- a/drivers/net/wireless/ath/wil6210/wmi.c +++ b/drivers/net/wireless/ath/wil6210/wmi.c @@ -426,15 +426,17 @@ static void wmi_evt_scan_complete(struct wil6210_priv *wil, int id, { if (wil->scan_request) { struct wmi_scan_complete_event *data = d; - bool aborted = (data->status != WMI_SCAN_SUCCESS); + struct cfg80211_scan_info info = { + .aborted = (data->status != WMI_SCAN_SUCCESS), + }; wil_dbg_wmi(wil, "SCAN_COMPLETE(0x%08x)\n", data->status); wil_dbg_misc(wil, "Complete scan_request 0x%p aborted %d\n", - wil->scan_request, aborted); + wil->scan_request, info.aborted); del_timer_sync(&wil->scan_timer); mutex_lock(&wil->p2p_wdev_mutex); - cfg80211_scan_done(wil->scan_request, aborted); + cfg80211_scan_done(wil->scan_request, &info); wil->radio_wdev = wil->wdev; mutex_unlock(&wil->p2p_wdev_mutex); wil->scan_request = NULL; diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index 264bd63..afe2b20 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -775,9 +775,13 @@ s32 brcmf_notify_escan_complete(struct brcmf_cfg80211_info *cfg, if (!aborted) cfg80211_sched_scan_results(cfg_to_wiphy(cfg)); } else if (scan_request) { + struct cfg80211_scan_info info = { + .aborted = aborted, + }; + brcmf_dbg(SCAN, "ESCAN Completed scan: %s\n", aborted ? "Aborted" : "Done"); - cfg80211_scan_done(scan_request, aborted); + cfg80211_scan_done(scan_request, &info); } if (!test_and_clear_bit(BRCMF_SCAN_STATUS_BUSY, &cfg->scan_status)) brcmf_dbg(SCAN, "Scan complete, probably P2P scan\n"); diff --git a/drivers/net/wireless/intersil/orinoco/scan.c b/drivers/net/wireless/intersil/orinoco/scan.c index d0ceb06..6d1d084 100644 --- a/drivers/net/wireless/intersil/orinoco/scan.c +++ b/drivers/net/wireless/intersil/orinoco/scan.c @@ -237,7 +237,11 @@ void orinoco_add_hostscan_results(struct orinoco_private *priv, scan_abort: if (priv->scan_request) { - cfg80211_scan_done(priv->scan_request, abort); + struct cfg80211_scan_info info = { + .aborted = abort, + }; + + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } } @@ -245,7 +249,11 @@ void orinoco_add_hostscan_results(struct orinoco_private *priv, void orinoco_scan_done(struct orinoco_private *priv, bool abort) { if (priv->scan_request) { - cfg80211_scan_done(priv->scan_request, abort); + struct cfg80211_scan_info info = { + .aborted = abort, + }; + + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } } diff --git a/drivers/net/wireless/marvell/libertas/cfg.c b/drivers/net/wireless/marvell/libertas/cfg.c index 776b44b..ea48024 100644 --- a/drivers/net/wireless/marvell/libertas/cfg.c +++ b/drivers/net/wireless/marvell/libertas/cfg.c @@ -796,10 +796,15 @@ void lbs_scan_done(struct lbs_private *priv) { WARN_ON(!priv->scan_req); - if (priv->internal_scan) + if (priv->internal_scan) { kfree(priv->scan_req); - else - cfg80211_scan_done(priv->scan_req, false); + } else { + struct cfg80211_scan_info info = { + .aborted = false, + }; + + cfg80211_scan_done(priv->scan_req, &info); + } priv->scan_req = NULL; } diff --git a/drivers/net/wireless/marvell/mwifiex/cmdevt.c b/drivers/net/wireless/marvell/mwifiex/cmdevt.c index 6bc2011..e7a2144 100644 --- a/drivers/net/wireless/marvell/mwifiex/cmdevt.c +++ b/drivers/net/wireless/marvell/mwifiex/cmdevt.c @@ -1057,8 +1057,12 @@ mwifiex_cancel_all_pending_cmd(struct mwifiex_adapter *adapter) if (!priv) continue; if (priv->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + mwifiex_dbg(adapter, WARN, "info: aborting scan\n"); - cfg80211_scan_done(priv->scan_request, 1); + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } } @@ -1112,8 +1116,12 @@ mwifiex_cancel_pending_ioctl(struct mwifiex_adapter *adapter) if (!priv) continue; if (priv->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + mwifiex_dbg(adapter, WARN, "info: aborting scan\n"); - cfg80211_scan_done(priv->scan_request, 1); + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } } diff --git a/drivers/net/wireless/marvell/mwifiex/main.c b/drivers/net/wireless/marvell/mwifiex/main.c index 0e280f8..db4925d 100644 --- a/drivers/net/wireless/marvell/mwifiex/main.c +++ b/drivers/net/wireless/marvell/mwifiex/main.c @@ -697,9 +697,13 @@ mwifiex_close(struct net_device *dev) struct mwifiex_private *priv = mwifiex_netdev_get_priv(dev); if (priv->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + mwifiex_dbg(priv->adapter, INFO, "aborting scan on ndo_stop\n"); - cfg80211_scan_done(priv->scan_request, 1); + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; priv->scan_aborting = true; } diff --git a/drivers/net/wireless/marvell/mwifiex/scan.c b/drivers/net/wireless/marvell/mwifiex/scan.c index bc5e52c..fdd7491 100644 --- a/drivers/net/wireless/marvell/mwifiex/scan.c +++ b/drivers/net/wireless/marvell/mwifiex/scan.c @@ -1956,9 +1956,13 @@ static void mwifiex_check_next_scan_command(struct mwifiex_private *priv) mwifiex_complete_scan(priv); if (priv->scan_request) { + struct cfg80211_scan_info info = { + .aborted = false, + }; + mwifiex_dbg(adapter, INFO, "info: notifying scan done\n"); - cfg80211_scan_done(priv->scan_request, 0); + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } else { priv->scan_aborting = false; @@ -1977,9 +1981,13 @@ static void mwifiex_check_next_scan_command(struct mwifiex_private *priv) if (!adapter->active_scan_triggered) { if (priv->scan_request) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + mwifiex_dbg(adapter, INFO, "info: aborting scan\n"); - cfg80211_scan_done(priv->scan_request, 1); + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } else { priv->scan_aborting = false; diff --git a/drivers/net/wireless/rndis_wlan.c b/drivers/net/wireless/rndis_wlan.c index 569918c..603c904 100644 --- a/drivers/net/wireless/rndis_wlan.c +++ b/drivers/net/wireless/rndis_wlan.c @@ -2134,6 +2134,7 @@ static void rndis_get_scan_results(struct work_struct *work) struct rndis_wlan_private *priv = container_of(work, struct rndis_wlan_private, scan_work.work); struct usbnet *usbdev = priv->usbdev; + struct cfg80211_scan_info info = {}; int ret; netdev_dbg(usbdev->net, "get_scan_results\n"); @@ -2143,7 +2144,8 @@ static void rndis_get_scan_results(struct work_struct *work) ret = rndis_check_bssid_list(usbdev, NULL, NULL); - cfg80211_scan_done(priv->scan_request, ret < 0); + info.aborted = ret < 0; + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } @@ -3574,7 +3576,11 @@ static int rndis_wlan_stop(struct usbnet *usbdev) flush_workqueue(priv->workqueue); if (priv->scan_request) { - cfg80211_scan_done(priv->scan_request, true); + struct cfg80211_scan_info info = { + .aborted = true, + }; + + cfg80211_scan_done(priv->scan_request, &info); priv->scan_request = NULL; } diff --git a/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c b/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c index 0da559d..d0ba377 100644 --- a/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c +++ b/drivers/staging/rtl8723au/os_dep/ioctl_cfg80211.c @@ -1256,10 +1256,15 @@ void rtw_cfg80211_indicate_scan_done(struct rtw_wdev_priv *pwdev_priv, DBG_8723A("%s with scan req\n", __func__); if (pwdev_priv->scan_request->wiphy != - pwdev_priv->rtw_wdev->wiphy) + pwdev_priv->rtw_wdev->wiphy) { DBG_8723A("error wiphy compare\n"); - else - cfg80211_scan_done(pwdev_priv->scan_request, aborted); + } else { + struct cfg80211_scan_info info = { + .aborted = aborted, + }; + + cfg80211_scan_done(pwdev_priv->scan_request, &info); + } pwdev_priv->scan_request = NULL; } else { diff --git a/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c b/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c index 51aff4f..a0d8e22 100644 --- a/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c +++ b/drivers/staging/wilc1000/wilc_wfi_cfgoperations.c @@ -454,7 +454,11 @@ static void CfgScanResult(enum scan_event scan_event, mutex_lock(&priv->scan_req_lock); if (priv->pstrScanReq) { - cfg80211_scan_done(priv->pstrScanReq, false); + struct cfg80211_scan_info info = { + .aborted = false, + }; + + cfg80211_scan_done(priv->pstrScanReq, &info); priv->u32RcvdChCount = 0; priv->bCfgScanning = false; priv->pstrScanReq = NULL; @@ -464,10 +468,14 @@ static void CfgScanResult(enum scan_event scan_event, mutex_lock(&priv->scan_req_lock); if (priv->pstrScanReq) { + struct cfg80211_scan_info info = { + .aborted = false, + }; + update_scan_time(); refresh_scan(priv, 1, false); - cfg80211_scan_done(priv->pstrScanReq, false); + cfg80211_scan_done(priv->pstrScanReq, &info); priv->bCfgScanning = false; priv->pstrScanReq = NULL; } diff --git a/drivers/staging/wlan-ng/cfg80211.c b/drivers/staging/wlan-ng/cfg80211.c index a6e6fb9..f46dfe6 100644 --- a/drivers/staging/wlan-ng/cfg80211.c +++ b/drivers/staging/wlan-ng/cfg80211.c @@ -338,6 +338,8 @@ static int prism2_scan(struct wiphy *wiphy, struct p80211msg_dot11req_scan msg1; struct p80211msg_dot11req_scan_results msg2; struct cfg80211_bss *bss; + struct cfg80211_scan_info info = {}; + int result; int err = 0; int numbss = 0; @@ -440,7 +442,8 @@ static int prism2_scan(struct wiphy *wiphy, err = prism2_result2err(msg2.resultcode.data); exit: - cfg80211_scan_done(request, err ? 1 : 0); + info.aborted = !!(err); + cfg80211_scan_done(request, &info); priv->scan_request = NULL; return err; } diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index fa4f0f7..e2658e3 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -1424,6 +1424,21 @@ struct cfg80211_ssid { }; /** + * struct cfg80211_scan_info - information about completed scan + * @scan_start_tsf: scan start time in terms of the TSF of the BSS that the + * wireless device that requested the scan is connected to. If this + * information is not available, this field is left zero. + * @tsf_bssid: the BSSID according to which %scan_start_tsf is set. + * @aborted: set to true if the scan was aborted for any reason, + * userspace will be notified of that + */ +struct cfg80211_scan_info { + u64 scan_start_tsf; + u8 tsf_bssid[ETH_ALEN] __aligned(2); + bool aborted; +}; + +/** * struct cfg80211_scan_request - scan request description * * @ssids: SSIDs to scan for (active scan only) @@ -1433,12 +1448,17 @@ struct cfg80211_ssid { * @scan_width: channel width for scanning * @ie: optional information element(s) to add into Probe Request or %NULL * @ie_len: length of ie in octets + * @duration: how long to listen on each channel, in TUs. If + * %duration_mandatory is not set, this is the maximum dwell time and + * the actual dwell time may be shorter. + * @duration_mandatory: if set, the scan duration must be as specified by the + * %duration field. * @flags: bit field of flags controlling operation * @rates: bitmap of rates to advertise for each band * @wiphy: the wiphy this was for * @scan_start: time (in jiffies) when the scan started * @wdev: the wireless device to scan for - * @aborted: (internal) scan request was notified as aborted + * @info: (internal) information about completed scan * @notified: (internal) scan request was notified as done or aborted * @no_cck: used to send probe requests at non CCK rate in 2GHz band * @mac_addr: MAC address used with randomisation @@ -1454,6 +1474,8 @@ struct cfg80211_scan_request { enum nl80211_bss_scan_width scan_width; const u8 *ie; size_t ie_len; + u16 duration; + bool duration_mandatory; u32 flags; u32 rates[NUM_NL80211_BANDS]; @@ -1467,7 +1489,8 @@ struct cfg80211_scan_request { /* internal */ struct wiphy *wiphy; unsigned long scan_start; - bool aborted, notified; + struct cfg80211_scan_info info; + bool notified; bool no_cck; /* keep last */ @@ -1600,12 +1623,19 @@ enum cfg80211_signal_type { * buffered on the device) and be accurate to about 10ms. * If the frame isn't buffered, just passing the return value of * ktime_get_boot_ns() is likely appropriate. + * @parent_tsf: the time at the start of reception of the first octet of the + * timestamp field of the frame. The time is the TSF of the BSS specified + * by %parent_bssid. + * @parent_bssid: the BSS according to which %parent_tsf is set. This is set to + * the BSS that requested the scan in which the beacon/probe was received. */ struct cfg80211_inform_bss { struct ieee80211_channel *chan; enum nl80211_bss_scan_width scan_width; s32 signal; u64 boottime_ns; + u64 parent_tsf; + u8 parent_bssid[ETH_ALEN] __aligned(2); }; /** @@ -4067,10 +4097,10 @@ const char *reg_initiator_name(enum nl80211_reg_initiator initiator); * cfg80211_scan_done - notify that scan finished * * @request: the corresponding scan request - * @aborted: set to true if the scan was aborted for any reason, - * userspace will be notified of that + * @info: information about the completed scan */ -void cfg80211_scan_done(struct cfg80211_scan_request *request, bool aborted); +void cfg80211_scan_done(struct cfg80211_scan_request *request, + struct cfg80211_scan_info *info); /** * cfg80211_sched_scan_results - notify that new scan results are available diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 1d7da78..b39ccab 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1848,6 +1848,22 @@ enum nl80211_commands { * to turn that feature off set an invalid mac address * (e.g. FF:FF:FF:FF:FF:FF) * + * @NL80211_ATTR_SCAN_START_TIME_TSF: The time at which the scan was actually + * started (u64). The time is the TSF of the BSS the interface that + * requested the scan is connected to (if available, otherwise this + * attribute must not be included). + * @NL80211_ATTR_SCAN_START_TIME_TSF_BSSID: The BSS according to which + * %NL80211_ATTR_SCAN_START_TIME_TSF is set. + * @NL80211_ATTR_MEASUREMENT_DURATION: measurement duration in TUs (u16). If + * %NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY is not set, this is the + * maximum measurement duration allowed. This attribute is used with + * measurement requests. It can also be used with %NL80211_CMD_TRIGGER_SCAN + * if the scan is used for beacon report radio measurement. + * @NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY: flag attribute that indicates + * that the duration specified with %NL80211_ATTR_MEASUREMENT_DURATION is + * mandatory. If this flag is not set, the duration is the maximum duration + * and the actual measurement duration may be shorter. + * * @NUM_NL80211_ATTR: total number of nl80211_attrs available * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use @@ -2235,6 +2251,11 @@ enum nl80211_attrs { NL80211_ATTR_MU_MIMO_GROUP_DATA, NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR, + NL80211_ATTR_SCAN_START_TIME_TSF, + NL80211_ATTR_SCAN_START_TIME_TSF_BSSID, + NL80211_ATTR_MEASUREMENT_DURATION, + NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, @@ -3496,6 +3517,12 @@ enum nl80211_bss_scan_width { * was last updated by a received frame. The value is expected to be * accurate to about 10ms. (u64, nanoseconds) * @NL80211_BSS_PAD: attribute used for padding for 64-bit alignment + * @NL80211_BSS_PARENT_TSF: the time at the start of reception of the first + * octet of the timestamp field of the last beacon/probe received for + * this BSS. The time is the TSF of the BSS specified by + * @NL80211_BSS_PARENT_BSSID. (u64). + * @NL80211_BSS_PARENT_BSSID: the BSS according to which @NL80211_BSS_PARENT_TSF + * is set. * @__NL80211_BSS_AFTER_LAST: internal * @NL80211_BSS_MAX: highest BSS attribute */ @@ -3517,6 +3544,8 @@ enum nl80211_bss { NL80211_BSS_PRESP_DATA, NL80211_BSS_LAST_SEEN_BOOTTIME, NL80211_BSS_PAD, + NL80211_BSS_PARENT_TSF, + NL80211_BSS_PARENT_BSSID, /* keep last */ __NL80211_BSS_AFTER_LAST, @@ -4507,6 +4536,16 @@ enum nl80211_feature_flags { * %NL80211_ATTR_MU_MIMO_GROUP_DATA attribute, * or can be configured to follow a station by configuring the * %NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR attribute. + * @NL80211_EXT_FEATURE_SCAN_START_TIME: This driver includes the actual + * time the scan started in scan results event. The time is the TSF of + * the BSS that the interface that requested the scan is connected to + * (if available). + * @NL80211_EXT_FEATURE_BSS_PARENT_TSF: Per BSS, this driver reports the + * time the last beacon/probe was received. The time is the TSF of the + * BSS that the interface that requested the scan is connected to + * (if available). + * @NL80211_EXT_FEATURE_SET_SCAN_DWELL: This driver supports configuration of + * channel dwell time. * * @NUM_NL80211_EXT_FEATURES: number of extended features. * @MAX_NL80211_EXT_FEATURES: highest extended feature index. @@ -4515,6 +4554,9 @@ enum nl80211_ext_feature_index { NL80211_EXT_FEATURE_VHT_IBSS, NL80211_EXT_FEATURE_RRM, NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER, + NL80211_EXT_FEATURE_SCAN_START_TIME, + NL80211_EXT_FEATURE_BSS_PARENT_TSF, + NL80211_EXT_FEATURE_SET_SCAN_DWELL, /* add new features before the definition below */ NUM_NL80211_EXT_FEATURES, diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index f9648ef..4ec1c52 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -353,8 +353,13 @@ static void __ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) scan_req = rcu_dereference_protected(local->scan_req, lockdep_is_held(&local->mtx)); - if (scan_req != local->int_scan_req) - cfg80211_scan_done(scan_req, aborted); + if (scan_req != local->int_scan_req) { + struct cfg80211_scan_info info = { + .aborted = aborted, + }; + + cfg80211_scan_done(scan_req, &info); + } RCU_INIT_POINTER(local->scan_req, NULL); scan_sdata = rcu_dereference_protected(local->scan_sdata, diff --git a/net/wireless/core.c b/net/wireless/core.c index 39d9abd..7645e97 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -220,7 +220,7 @@ void cfg80211_stop_p2p_device(struct cfg80211_registered_device *rdev, if (rdev->scan_req && rdev->scan_req->wdev == wdev) { if (WARN_ON(!rdev->scan_req->notified)) - rdev->scan_req->aborted = true; + rdev->scan_req->info.aborted = true; ___cfg80211_scan_done(rdev, false); } } @@ -1087,7 +1087,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb, cfg80211_update_iface_num(rdev, wdev->iftype, -1); if (rdev->scan_req && rdev->scan_req->wdev == wdev) { if (WARN_ON(!rdev->scan_req->notified)) - rdev->scan_req->aborted = true; + rdev->scan_req->info.aborted = true; ___cfg80211_scan_done(rdev, false); } diff --git a/net/wireless/core.h b/net/wireless/core.h index a4d547f..eee9144 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -141,6 +141,18 @@ struct cfg80211_internal_bss { unsigned long refcount; atomic_t hold; + /* time at the start of the reception of the first octet of the + * timestamp field of the last beacon/probe received for this BSS. + * The time is the TSF of the BSS specified by %parent_bssid. + */ + u64 parent_tsf; + + /* the BSS according to which %parent_tsf is set. This is set to + * the BSS that the interface that requested the scan was connected to + * when the beacon/probe was received. + */ + u8 parent_bssid[ETH_ALEN] __aligned(2); + /* must be last because of priv member */ struct cfg80211_bss pub; }; diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 447026f..c53b546 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -6223,6 +6223,19 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) } } + if (info->attrs[NL80211_ATTR_MEASUREMENT_DURATION]) { + if (!wiphy_ext_feature_isset(wiphy, + NL80211_EXT_FEATURE_SET_SCAN_DWELL)) { + err = -EOPNOTSUPP; + goto out_free; + } + + request->duration = + nla_get_u16(info->attrs[NL80211_ATTR_MEASUREMENT_DURATION]); + request->duration_mandatory = + nla_get_flag(info->attrs[NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY]); + } + if (info->attrs[NL80211_ATTR_SCAN_FLAGS]) { request->flags = nla_get_u32( info->attrs[NL80211_ATTR_SCAN_FLAGS]); @@ -7056,6 +7069,13 @@ static int nl80211_send_bss(struct sk_buff *msg, struct netlink_callback *cb, jiffies_to_msecs(jiffies - intbss->ts))) goto nla_put_failure; + if (intbss->parent_tsf && + (nla_put_u64_64bit(msg, NL80211_BSS_PARENT_TSF, + intbss->parent_tsf, NL80211_BSS_PAD) || + nla_put(msg, NL80211_BSS_PARENT_BSSID, ETH_ALEN, + intbss->parent_bssid))) + goto nla_put_failure; + if (intbss->ts_boottime && nla_put_u64_64bit(msg, NL80211_BSS_LAST_SEEN_BOOTTIME, intbss->ts_boottime, NL80211_BSS_PAD)) @@ -11829,6 +11849,13 @@ static int nl80211_add_scan_req(struct sk_buff *msg, nla_put_u32(msg, NL80211_ATTR_SCAN_FLAGS, req->flags)) goto nla_put_failure; + if (req->info.scan_start_tsf && + (nla_put_u64_64bit(msg, NL80211_ATTR_SCAN_START_TIME_TSF, + req->info.scan_start_tsf, NL80211_BSS_PAD) || + nla_put(msg, NL80211_ATTR_SCAN_START_TIME_TSF_BSSID, ETH_ALEN, + req->info.tsf_bssid))) + goto nla_put_failure; + return 0; nla_put_failure: return -ENOBUFS; diff --git a/net/wireless/scan.c b/net/wireless/scan.c index ef2955c..0358e12 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -3,6 +3,7 @@ * * Copyright 2008 Johannes Berg * Copyright 2013-2014 Intel Mobile Communications GmbH + * Copyright 2016 Intel Deutschland GmbH */ #include #include @@ -194,7 +195,7 @@ void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev, if (wdev->netdev) cfg80211_sme_scan_done(wdev->netdev); - if (!request->aborted && + if (!request->info.aborted && request->flags & NL80211_SCAN_FLAG_FLUSH) { /* flush entries from previous scans */ spin_lock_bh(&rdev->bss_lock); @@ -202,10 +203,10 @@ void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev, spin_unlock_bh(&rdev->bss_lock); } - msg = nl80211_build_scan_msg(rdev, wdev, request->aborted); + msg = nl80211_build_scan_msg(rdev, wdev, request->info.aborted); #ifdef CONFIG_CFG80211_WEXT - if (wdev->netdev && !request->aborted) { + if (wdev->netdev && !request->info.aborted) { memset(&wrqu, 0, sizeof(wrqu)); wireless_send_event(wdev->netdev, SIOCGIWSCAN, &wrqu, NULL); @@ -236,12 +237,13 @@ void __cfg80211_scan_done(struct work_struct *wk) rtnl_unlock(); } -void cfg80211_scan_done(struct cfg80211_scan_request *request, bool aborted) +void cfg80211_scan_done(struct cfg80211_scan_request *request, + struct cfg80211_scan_info *info) { - trace_cfg80211_scan_done(request, aborted); + trace_cfg80211_scan_done(request, info); WARN_ON(request != wiphy_to_rdev(request->wiphy)->scan_req); - request->aborted = aborted; + request->info = *info; request->notified = true; queue_work(cfg80211_wq, &wiphy_to_rdev(request->wiphy)->scan_done_wk); } @@ -843,6 +845,8 @@ cfg80211_bss_update(struct cfg80211_registered_device *rdev, found->pub.capability = tmp->pub.capability; found->ts = tmp->ts; found->ts_boottime = tmp->ts_boottime; + found->parent_tsf = tmp->parent_tsf; + ether_addr_copy(found->parent_bssid, tmp->parent_bssid); } else { struct cfg80211_internal_bss *new; struct cfg80211_internal_bss *hidden; @@ -1086,6 +1090,8 @@ cfg80211_inform_bss_frame_data(struct wiphy *wiphy, tmp.pub.beacon_interval = le16_to_cpu(mgmt->u.probe_resp.beacon_int); tmp.pub.capability = le16_to_cpu(mgmt->u.probe_resp.capab_info); tmp.ts_boottime = data->boottime_ns; + tmp.parent_tsf = data->parent_tsf; + ether_addr_copy(tmp.parent_bssid, data->parent_bssid); signal_valid = abs(data->chan->center_freq - channel->center_freq) <= wiphy->max_adj_channel_rssi_comp; diff --git a/net/wireless/trace.h b/net/wireless/trace.h index 3c1091ae..72b5255 100644 --- a/net/wireless/trace.h +++ b/net/wireless/trace.h @@ -2642,8 +2642,9 @@ TRACE_EVENT(cfg80211_tdls_oper_request, ); TRACE_EVENT(cfg80211_scan_done, - TP_PROTO(struct cfg80211_scan_request *request, bool aborted), - TP_ARGS(request, aborted), + TP_PROTO(struct cfg80211_scan_request *request, + struct cfg80211_scan_info *info), + TP_ARGS(request, info), TP_STRUCT__entry( __field(u32, n_channels) __dynamic_array(u8, ie, request ? request->ie_len : 0) @@ -2652,6 +2653,8 @@ TRACE_EVENT(cfg80211_scan_done, MAC_ENTRY(wiphy_mac) __field(bool, no_cck) __field(bool, aborted) + __field(u64, scan_start_tsf) + MAC_ENTRY(tsf_bssid) ), TP_fast_assign( if (request) { @@ -2666,9 +2669,16 @@ TRACE_EVENT(cfg80211_scan_done, request->wiphy->perm_addr); __entry->no_cck = request->no_cck; } - __entry->aborted = aborted; + if (info) { + __entry->aborted = info->aborted; + __entry->scan_start_tsf = info->scan_start_tsf; + MAC_ASSIGN(tsf_bssid, info->tsf_bssid); + } ), - TP_printk("aborted: %s", BOOL_TO_STR(__entry->aborted)) + TP_printk("aborted: %s, scan start (TSF): %llu, tsf_bssid: " MAC_PR_FMT, + BOOL_TO_STR(__entry->aborted), + (unsigned long long)__entry->scan_start_tsf, + MAC_PR_ARG(tsf_bssid)) ); DEFINE_EVENT(wiphy_only_evt, cfg80211_sched_scan_results, @@ -2721,6 +2731,8 @@ TRACE_EVENT(cfg80211_inform_bss_frame, __dynamic_array(u8, mgmt, len) __field(s32, signal) __field(u64, ts_boottime) + __field(u64, parent_tsf) + MAC_ENTRY(parent_bssid) ), TP_fast_assign( WIPHY_ASSIGN; @@ -2730,10 +2742,15 @@ TRACE_EVENT(cfg80211_inform_bss_frame, memcpy(__get_dynamic_array(mgmt), mgmt, len); __entry->signal = data->signal; __entry->ts_boottime = data->boottime_ns; - ), - TP_printk(WIPHY_PR_FMT ", " CHAN_PR_FMT "(scan_width: %d) signal: %d, tsb:%llu", - WIPHY_PR_ARG, CHAN_PR_ARG, __entry->scan_width, - __entry->signal, (unsigned long long)__entry->ts_boottime) + __entry->parent_tsf = data->parent_tsf; + MAC_ASSIGN(parent_bssid, data->parent_bssid); + ), + TP_printk(WIPHY_PR_FMT ", " CHAN_PR_FMT + "(scan_width: %d) signal: %d, tsb:%llu, detect_tsf:%llu, tsf_bssid: " + MAC_PR_FMT, WIPHY_PR_ARG, CHAN_PR_ARG, __entry->scan_width, + __entry->signal, (unsigned long long)__entry->ts_boottime, + (unsigned long long)__entry->parent_tsf, + MAC_PR_ARG(parent_bssid)) ); DECLARE_EVENT_CLASS(cfg80211_bss_evt, -- cgit v0.10.2 From 7947d3e075cde1a18e538f2dafbc850aa356ff79 Mon Sep 17 00:00:00 2001 From: Avraham Stern Date: Tue, 5 Jul 2016 15:23:12 +0300 Subject: mac80211: Add support for beacon report radio measurement Add the following to support beacon report radio measurement with the measurement mode field set to passive or active: 1. Propagate the required scan duration to the device 2. Report the scan start time (in terms of TSF) 3. Report each BSS's detection time (also in terms of TSF) TSF times refer to the BSS that the interface that requested the scan is connected to. Signed-off-by: Assaf Krauss Signed-off-by: Avraham Stern [changed ath9k/10k, at76c59x-usb, iwlegacy, wl1251 and wlcore to match the new API] Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index d4b7a16..ebc12c5 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -3858,12 +3858,16 @@ void __ath10k_scan_finish(struct ath10k *ar) break; case ATH10K_SCAN_RUNNING: case ATH10K_SCAN_ABORTING: - if (!ar->scan.is_roc) - ieee80211_scan_completed(ar->hw, - (ar->scan.state == - ATH10K_SCAN_ABORTING)); - else if (ar->scan.roc_notify) + if (!ar->scan.is_roc) { + struct cfg80211_scan_info info = { + .aborted = (ar->scan.state == + ATH10K_SCAN_ABORTING), + }; + + ieee80211_scan_completed(ar->hw, &info); + } else if (ar->scan.roc_notify) { ieee80211_remain_on_channel_expired(ar->hw); + } /* fall through */ case ATH10K_SCAN_STARTING: ar->scan.state = ATH10K_SCAN_IDLE; diff --git a/drivers/net/wireless/ath/ath9k/channel.c b/drivers/net/wireless/ath/ath9k/channel.c index e56bafc..57e26a6 100644 --- a/drivers/net/wireless/ath/ath9k/channel.c +++ b/drivers/net/wireless/ath/ath9k/channel.c @@ -960,6 +960,9 @@ void ath_roc_complete(struct ath_softc *sc, enum ath_roc_complete_reason reason) void ath_scan_complete(struct ath_softc *sc, bool abort) { struct ath_common *common = ath9k_hw_common(sc->sc_ah); + struct cfg80211_scan_info info = { + .aborted = abort, + }; if (abort) ath_dbg(common, CHAN_CTX, "HW scan aborted\n"); @@ -969,7 +972,7 @@ void ath_scan_complete(struct ath_softc *sc, bool abort) sc->offchannel.scan_req = NULL; sc->offchannel.scan_vif = NULL; sc->offchannel.state = ATH_OFFCHANNEL_IDLE; - ieee80211_scan_completed(sc->hw, abort); + ieee80211_scan_completed(sc->hw, &info); clear_bit(ATH_OP_SCANNING, &common->op_flags); spin_lock_bh(&sc->chan_lock); if (test_bit(ATH_OP_MULTI_CHANNEL, &common->op_flags)) diff --git a/drivers/net/wireless/atmel/at76c50x-usb.c b/drivers/net/wireless/atmel/at76c50x-usb.c index 7c10804..0e18067 100644 --- a/drivers/net/wireless/atmel/at76c50x-usb.c +++ b/drivers/net/wireless/atmel/at76c50x-usb.c @@ -1922,6 +1922,9 @@ static void at76_dwork_hw_scan(struct work_struct *work) { struct at76_priv *priv = container_of(work, struct at76_priv, dwork_hw_scan.work); + struct cfg80211_scan_info info = { + .aborted = false, + }; int ret; if (priv->device_unplugged) @@ -1948,7 +1951,7 @@ static void at76_dwork_hw_scan(struct work_struct *work) mutex_unlock(&priv->mtx); - ieee80211_scan_completed(priv->hw, false); + ieee80211_scan_completed(priv->hw, &info); ieee80211_wake_queues(priv->hw); } diff --git a/drivers/net/wireless/intel/iwlegacy/common.c b/drivers/net/wireless/intel/iwlegacy/common.c index eb24b92..140b6ea 100644 --- a/drivers/net/wireless/intel/iwlegacy/common.c +++ b/drivers/net/wireless/intel/iwlegacy/common.c @@ -1305,10 +1305,14 @@ il_send_scan_abort(struct il_priv *il) static void il_complete_scan(struct il_priv *il, bool aborted) { + struct cfg80211_scan_info info = { + .aborted = aborted, + }; + /* check if scan was requested from mac80211 */ if (il->scan_request) { D_SCAN("Complete scan in mac80211\n"); - ieee80211_scan_completed(il->hw, aborted); + ieee80211_scan_completed(il->hw, &info); } il->scan_vif = NULL; diff --git a/drivers/net/wireless/intel/iwlwifi/dvm/scan.c b/drivers/net/wireless/intel/iwlwifi/dvm/scan.c index d01766f..17e6a32 100644 --- a/drivers/net/wireless/intel/iwlwifi/dvm/scan.c +++ b/drivers/net/wireless/intel/iwlwifi/dvm/scan.c @@ -94,10 +94,14 @@ static int iwl_send_scan_abort(struct iwl_priv *priv) static void iwl_complete_scan(struct iwl_priv *priv, bool aborted) { + struct cfg80211_scan_info info = { + .aborted = aborted, + }; + /* check if scan was requested from mac80211 */ if (priv->scan_request) { IWL_DEBUG_SCAN(priv, "Complete scan in mac80211\n"); - ieee80211_scan_completed(priv->hw, aborted); + ieee80211_scan_completed(priv->hw, &info); } priv->scan_type = IWL_SCAN_NORMAL; diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c index e78fc56..1cac10c 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c @@ -391,13 +391,16 @@ void iwl_mvm_rx_lmac_scan_complete_notif(struct iwl_mvm *mvm, ieee80211_sched_scan_stopped(mvm->hw); mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_DISABLED; } else if (mvm->scan_status & IWL_MVM_SCAN_REGULAR) { + struct cfg80211_scan_info info = { + .aborted = aborted, + }; + IWL_DEBUG_SCAN(mvm, "Regular scan %s, EBS status %s (FW)\n", aborted ? "aborted" : "completed", iwl_mvm_ebs_status_str(scan_notif->ebs_status)); mvm->scan_status &= ~IWL_MVM_SCAN_REGULAR; - ieee80211_scan_completed(mvm->hw, - scan_notif->status == IWL_SCAN_OFFLOAD_ABORTED); + ieee80211_scan_completed(mvm->hw, &info); iwl_mvm_unref(mvm, IWL_MVM_REF_SCAN); del_timer(&mvm->scan_timer); } else { @@ -1430,7 +1433,11 @@ void iwl_mvm_rx_umac_scan_complete_notif(struct iwl_mvm *mvm, /* if the scan is already stopping, we don't need to notify mac80211 */ if (mvm->scan_uid_status[uid] == IWL_MVM_SCAN_REGULAR) { - ieee80211_scan_completed(mvm->hw, aborted); + struct cfg80211_scan_info info = { + .aborted = aborted, + }; + + ieee80211_scan_completed(mvm->hw, &info); iwl_mvm_unref(mvm, IWL_MVM_REF_SCAN); del_timer(&mvm->scan_timer); } else if (mvm->scan_uid_status[uid] == IWL_MVM_SCAN_SCHED) { @@ -1564,7 +1571,11 @@ void iwl_mvm_report_scan_aborted(struct iwl_mvm *mvm) uid = iwl_mvm_scan_uid_by_status(mvm, IWL_MVM_SCAN_REGULAR); if (uid >= 0) { - ieee80211_scan_completed(mvm->hw, true); + struct cfg80211_scan_info info = { + .aborted = true, + }; + + ieee80211_scan_completed(mvm->hw, &info); mvm->scan_uid_status[uid] = 0; } uid = iwl_mvm_scan_uid_by_status(mvm, IWL_MVM_SCAN_SCHED); @@ -1585,8 +1596,13 @@ void iwl_mvm_report_scan_aborted(struct iwl_mvm *mvm) mvm->scan_uid_status[i] = 0; } } else { - if (mvm->scan_status & IWL_MVM_SCAN_REGULAR) - ieee80211_scan_completed(mvm->hw, true); + if (mvm->scan_status & IWL_MVM_SCAN_REGULAR) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + + ieee80211_scan_completed(mvm->hw, &info); + } /* Sched scan will be restarted by mac80211 in * restart_hw, so do not report if FW is about to be @@ -1629,8 +1645,13 @@ out: */ iwl_mvm_unref(mvm, IWL_MVM_REF_SCAN); del_timer(&mvm->scan_timer); - if (notify) - ieee80211_scan_completed(mvm->hw, true); + if (notify) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + + ieee80211_scan_completed(mvm->hw, &info); + } } else if (notify) { ieee80211_sched_scan_stopped(mvm->hw); mvm->sched_scan_pass_all = SCHED_SCAN_PASS_ALL_DISABLED; diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index 337794f..8c35ac8 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -1941,8 +1941,12 @@ static void hw_scan_work(struct work_struct *work) mutex_lock(&hwsim->mutex); if (hwsim->scan_chan_idx >= req->n_channels) { + struct cfg80211_scan_info info = { + .aborted = false, + }; + wiphy_debug(hwsim->hw->wiphy, "hw scan complete\n"); - ieee80211_scan_completed(hwsim->hw, false); + ieee80211_scan_completed(hwsim->hw, &info); hwsim->hw_scan_request = NULL; hwsim->hw_scan_vif = NULL; hwsim->tmp_chan = NULL; @@ -2027,13 +2031,16 @@ static void mac80211_hwsim_cancel_hw_scan(struct ieee80211_hw *hw, struct ieee80211_vif *vif) { struct mac80211_hwsim_data *hwsim = hw->priv; + struct cfg80211_scan_info info = { + .aborted = true, + }; wiphy_debug(hw->wiphy, "hwsim cancel_hw_scan\n"); cancel_delayed_work_sync(&hwsim->hw_scan); mutex_lock(&hwsim->mutex); - ieee80211_scan_completed(hwsim->hw, true); + ieee80211_scan_completed(hwsim->hw, &info); hwsim->tmp_chan = NULL; hwsim->hw_scan_request = NULL; hwsim->hw_scan_vif = NULL; diff --git a/drivers/net/wireless/st/cw1200/scan.c b/drivers/net/wireless/st/cw1200/scan.c index 9837881..0a0ff7e 100644 --- a/drivers/net/wireless/st/cw1200/scan.c +++ b/drivers/net/wireless/st/cw1200/scan.c @@ -167,6 +167,10 @@ void cw1200_scan_work(struct work_struct *work) } if (!priv->scan.req || (priv->scan.curr == priv->scan.end)) { + struct cfg80211_scan_info info = { + .aborted = priv->scan.status ? 1 : 0, + }; + if (priv->scan.output_power != priv->output_power) wsm_set_output_power(priv, priv->output_power * 10); if (priv->join_status == CW1200_JOIN_STATUS_STA && @@ -188,7 +192,7 @@ void cw1200_scan_work(struct work_struct *work) cw1200_scan_restart_delayed(priv); wsm_unlock_tx(priv); mutex_unlock(&priv->conf_mutex); - ieee80211_scan_completed(priv->hw, priv->scan.status ? 1 : 0); + ieee80211_scan_completed(priv->hw, &info); up(&priv->scan.lock); return; } else { diff --git a/drivers/net/wireless/ti/wl1251/event.c b/drivers/net/wireless/ti/wl1251/event.c index c986303..d0593bc 100644 --- a/drivers/net/wireless/ti/wl1251/event.c +++ b/drivers/net/wireless/ti/wl1251/event.c @@ -36,7 +36,11 @@ static int wl1251_event_scan_complete(struct wl1251 *wl, mbox->scheduled_scan_channels); if (wl->scanning) { - ieee80211_scan_completed(wl->hw, false); + struct cfg80211_scan_info info = { + .aborted = false, + }; + + ieee80211_scan_completed(wl->hw, &info); wl1251_debug(DEBUG_MAC80211, "mac80211 hw scan completed"); wl->scanning = false; if (wl->hw->conf.flags & IEEE80211_CONF_IDLE) diff --git a/drivers/net/wireless/ti/wl1251/main.c b/drivers/net/wireless/ti/wl1251/main.c index 56384a4e..bbf7604 100644 --- a/drivers/net/wireless/ti/wl1251/main.c +++ b/drivers/net/wireless/ti/wl1251/main.c @@ -448,7 +448,11 @@ static void wl1251_op_stop(struct ieee80211_hw *hw) WARN_ON(wl->state != WL1251_STATE_ON); if (wl->scanning) { - ieee80211_scan_completed(wl->hw, true); + struct cfg80211_scan_info info = { + .aborted = true, + }; + + ieee80211_scan_completed(wl->hw, &info); wl->scanning = false; } diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c index 10fd24c..69267d5 100644 --- a/drivers/net/wireless/ti/wlcore/main.c +++ b/drivers/net/wireless/ti/wlcore/main.c @@ -2615,6 +2615,10 @@ static void __wl1271_op_remove_interface(struct wl1271 *wl, if (wl->scan.state != WL1271_SCAN_STATE_IDLE && wl->scan_wlvif == wlvif) { + struct cfg80211_scan_info info = { + .aborted = true, + }; + /* * Rearm the tx watchdog just before idling scan. This * prevents just-finished scans from triggering the watchdog @@ -2625,7 +2629,7 @@ static void __wl1271_op_remove_interface(struct wl1271 *wl, memset(wl->scan.scanned_ch, 0, sizeof(wl->scan.scanned_ch)); wl->scan_wlvif = NULL; wl->scan.req = NULL; - ieee80211_scan_completed(wl->hw, true); + ieee80211_scan_completed(wl->hw, &info); } if (wl->sched_vif == wlvif) @@ -3649,6 +3653,9 @@ static void wl1271_op_cancel_hw_scan(struct ieee80211_hw *hw, { struct wl1271 *wl = hw->priv; struct wl12xx_vif *wlvif = wl12xx_vif_to_data(vif); + struct cfg80211_scan_info info = { + .aborted = true, + }; int ret; wl1271_debug(DEBUG_MAC80211, "mac80211 cancel hw scan"); @@ -3681,7 +3688,7 @@ static void wl1271_op_cancel_hw_scan(struct ieee80211_hw *hw, memset(wl->scan.scanned_ch, 0, sizeof(wl->scan.scanned_ch)); wl->scan_wlvif = NULL; wl->scan.req = NULL; - ieee80211_scan_completed(wl->hw, true); + ieee80211_scan_completed(wl->hw, &info); out_sleep: wl1271_ps_elp_sleep(wl); diff --git a/drivers/net/wireless/ti/wlcore/scan.c b/drivers/net/wireless/ti/wlcore/scan.c index 2334364..5612f59 100644 --- a/drivers/net/wireless/ti/wlcore/scan.c +++ b/drivers/net/wireless/ti/wlcore/scan.c @@ -36,6 +36,9 @@ void wl1271_scan_complete_work(struct work_struct *work) struct delayed_work *dwork; struct wl1271 *wl; struct wl12xx_vif *wlvif; + struct cfg80211_scan_info info = { + .aborted = false, + }; int ret; dwork = to_delayed_work(work); @@ -82,7 +85,7 @@ void wl1271_scan_complete_work(struct work_struct *work) wlcore_cmd_regdomain_config_locked(wl); - ieee80211_scan_completed(wl->hw, false); + ieee80211_scan_completed(wl->hw, &info); out: mutex_unlock(&wl->mutex); diff --git a/include/net/mac80211.h b/include/net/mac80211.h index a52009f..b4faadb 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -4697,9 +4697,10 @@ void ieee80211_wake_queues(struct ieee80211_hw *hw); * any context, including hardirq context. * * @hw: the hardware that finished the scan - * @aborted: set to true if scan was aborted + * @info: information about the completed scan */ -void ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted); +void ieee80211_scan_completed(struct ieee80211_hw *hw, + struct cfg80211_scan_info *info); /** * ieee80211_sched_scan_results - got results from scheduled scan diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 54edfb6..f56d342 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -1250,6 +1250,7 @@ struct ieee80211_local { int scan_channel_idx; int scan_ies_len; int hw_scan_ies_bufsize; + struct cfg80211_scan_info scan_info; struct work_struct sched_scan_stopped_work; struct ieee80211_sub_if_data __rcu *sched_scan_sdata; diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index 4ec1c52..8d4a9cd 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -7,6 +7,7 @@ * Copyright 2006-2007 Jiri Benc * Copyright 2007, Michael Wu * Copyright 2013-2015 Intel Mobile Communications GmbH + * Copyright 2016 Intel Deutschland GmbH * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -70,6 +71,7 @@ ieee80211_bss_info_update(struct ieee80211_local *local, .boottime_ns = rx_status->boottime_ns, }; bool signal_valid; + struct ieee80211_sub_if_data *scan_sdata; if (ieee80211_hw_check(&local->hw, SIGNAL_DBM)) bss_meta.signal = rx_status->signal * 100; @@ -83,6 +85,20 @@ ieee80211_bss_info_update(struct ieee80211_local *local, bss_meta.scan_width = NL80211_BSS_CHAN_WIDTH_10; bss_meta.chan = channel; + + rcu_read_lock(); + scan_sdata = rcu_dereference(local->scan_sdata); + if (scan_sdata && scan_sdata->vif.type == NL80211_IFTYPE_STATION && + scan_sdata->vif.bss_conf.assoc && + ieee80211_have_rx_timestamp(rx_status)) { + bss_meta.parent_tsf = + ieee80211_calculate_rx_timestamp(local, rx_status, + len + FCS_LEN, 24); + ether_addr_copy(bss_meta.parent_bssid, + scan_sdata->vif.bss_conf.bssid); + } + rcu_read_unlock(); + cbss = cfg80211_inform_bss_frame_data(local->hw.wiphy, &bss_meta, mgmt, len, GFP_ATOMIC); if (!cbss) @@ -345,6 +361,11 @@ static void __ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) if (rc == 0) return; + + /* HW scan failed and is going to be reported as done, so clear + * old scan info. + */ + memset(&local->scan_info, 0, sizeof(local->scan_info)); } kfree(local->hw_scan_req); @@ -354,11 +375,8 @@ static void __ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) lockdep_is_held(&local->mtx)); if (scan_req != local->int_scan_req) { - struct cfg80211_scan_info info = { - .aborted = aborted, - }; - - cfg80211_scan_done(scan_req, &info); + local->scan_info.aborted = aborted; + cfg80211_scan_done(scan_req, &local->scan_info); } RCU_INIT_POINTER(local->scan_req, NULL); @@ -396,15 +414,19 @@ static void __ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) ieee80211_start_next_roc(local); } -void ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) +void ieee80211_scan_completed(struct ieee80211_hw *hw, + struct cfg80211_scan_info *info) { struct ieee80211_local *local = hw_to_local(hw); - trace_api_scan_completed(local, aborted); + trace_api_scan_completed(local, info); set_bit(SCAN_COMPLETED, &local->scanning); - if (aborted) + if (info->aborted) set_bit(SCAN_ABORTED, &local->scanning); + + memcpy(&local->scan_info, info, sizeof(*info)); + ieee80211_queue_delayed_work(&local->hw, &local->scan_work, 0); } EXPORT_SYMBOL(ieee80211_scan_completed); @@ -571,6 +593,9 @@ static int __ieee80211_start_scan(struct ieee80211_sub_if_data *sdata, local->hw_scan_req->req.ie = ies; local->hw_scan_req->req.flags = req->flags; eth_broadcast_addr(local->hw_scan_req->req.bssid); + local->hw_scan_req->req.duration = req->duration; + local->hw_scan_req->req.duration_mandatory = + req->duration_mandatory; local->hw_scan_band = 0; @@ -1078,6 +1103,7 @@ void ieee80211_scan_cancel(struct ieee80211_local *local) */ cancel_delayed_work(&local->scan_work); /* and clean up */ + memset(&local->scan_info, 0, sizeof(local->scan_info)); __ieee80211_scan_completed(&local->hw, true); out: mutex_unlock(&local->mtx); -- cgit v0.10.2 From 7d10f6b179bc82e6633a4521a4cd69ad6846723e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 5 Jul 2016 15:23:13 +0300 Subject: mac80211: report failure to start (partial) scan as scan abort Rather than reporting the scan as having completed, report it as being aborted. Signed-off-by: Johannes Berg diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index 8d4a9cd..070b40f 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -362,10 +362,11 @@ static void __ieee80211_scan_completed(struct ieee80211_hw *hw, bool aborted) if (rc == 0) return; - /* HW scan failed and is going to be reported as done, so clear - * old scan info. + /* HW scan failed and is going to be reported as aborted, + * so clear old scan info. */ memset(&local->scan_info, 0, sizeof(local->scan_info)); + aborted = true; } kfree(local->hw_scan_req); -- cgit v0.10.2 From 92b3a28a2b4b2fb777b64f0891a4d3388f617c15 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 5 Jul 2016 15:23:14 +0300 Subject: mac80211: parse wide bandwidth channel switch IE with workaround Continuing the workaround implemented in commit 23665aaf9170 ("mac80211: Interoperability workaround for 80+80 and 160 MHz channels") use the same code to parse the Wide Bandwidth Channel Switch element by converting to VHT Operation element since the spec also just refers to that for parsing semantics, particularly with the workaround. While at it, remove some dead code - the IEEE80211_STA_DISABLE_40MHZ flag can never be set at this point since it's checked earlier and the wide_bw_chansw_ie pointer is set to NULL if it's set. Signed-off-by: Johannes Berg diff --git a/net/mac80211/spectmgmt.c b/net/mac80211/spectmgmt.c index 2ddc661..97f4c9d 100644 --- a/net/mac80211/spectmgmt.c +++ b/net/mac80211/spectmgmt.c @@ -129,42 +129,31 @@ int ieee80211_parse_ch_switch_ie(struct ieee80211_sub_if_data *sdata, } if (wide_bw_chansw_ie) { - new_vht_chandef.chan = new_chan; - new_vht_chandef.center_freq1 = - ieee80211_channel_to_frequency( + struct ieee80211_vht_operation vht_oper = { + .chan_width = + wide_bw_chansw_ie->new_channel_width, + .center_freq_seg1_idx = wide_bw_chansw_ie->new_center_freq_seg0, - new_band); - - switch (wide_bw_chansw_ie->new_channel_width) { - default: - /* hmmm, ignore VHT and use HT if present */ - case IEEE80211_VHT_CHANWIDTH_USE_HT: + .center_freq_seg2_idx = + wide_bw_chansw_ie->new_center_freq_seg1, + /* .basic_mcs_set doesn't matter */ + }; + + /* default, for the case of IEEE80211_VHT_CHANWIDTH_USE_HT, + * to the previously parsed chandef + */ + new_vht_chandef = csa_ie->chandef; + + /* ignore if parsing fails */ + if (!ieee80211_chandef_vht_oper(&vht_oper, &new_vht_chandef)) new_vht_chandef.chan = NULL; - break; - case IEEE80211_VHT_CHANWIDTH_80MHZ: - new_vht_chandef.width = NL80211_CHAN_WIDTH_80; - break; - case IEEE80211_VHT_CHANWIDTH_160MHZ: - new_vht_chandef.width = NL80211_CHAN_WIDTH_160; - break; - case IEEE80211_VHT_CHANWIDTH_80P80MHZ: - /* field is otherwise reserved */ - new_vht_chandef.center_freq2 = - ieee80211_channel_to_frequency( - wide_bw_chansw_ie->new_center_freq_seg1, - new_band); - new_vht_chandef.width = NL80211_CHAN_WIDTH_80P80; - break; - } + if (sta_flags & IEEE80211_STA_DISABLE_80P80MHZ && new_vht_chandef.width == NL80211_CHAN_WIDTH_80P80) ieee80211_chandef_downgrade(&new_vht_chandef); if (sta_flags & IEEE80211_STA_DISABLE_160MHZ && new_vht_chandef.width == NL80211_CHAN_WIDTH_160) ieee80211_chandef_downgrade(&new_vht_chandef); - if (sta_flags & IEEE80211_STA_DISABLE_40MHZ && - new_vht_chandef.width > NL80211_CHAN_WIDTH_20) - ieee80211_chandef_downgrade(&new_vht_chandef); } /* if VHT data is there validate & use it */ -- cgit v0.10.2 From 7d27a0ba7adc8ef30c2aae7592fce4c162aee4df Mon Sep 17 00:00:00 2001 From: Masashi Honma Date: Fri, 1 Jul 2016 10:19:34 +0900 Subject: cfg80211: Add mesh peer AID setting API Previously, mesh power management functionality works only with kernel MPM. Because user space MPM did not report mesh peer AID to kernel, the kernel could not identify the bit in TIM element. So this patch adds mesh peer AID setting API. Signed-off-by: Masashi Honma Signed-off-by: Johannes Berg diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index e2658e3..9c23f4d3 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -780,6 +780,7 @@ enum station_parameters_apply_mask { * (bitmask of BIT(NL80211_STA_FLAG_...)) * @listen_interval: listen interval or -1 for no change * @aid: AID or zero for no change + * @peer_aid: mesh peer AID or zero for no change * @plink_action: plink action to take * @plink_state: set the peer link state for a station * @ht_capa: HT capabilities of station @@ -811,6 +812,7 @@ struct station_parameters { u32 sta_modify_mask; int listen_interval; u16 aid; + u16 peer_aid; u8 supported_rates_len; u8 plink_action; u8 plink_state; diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index b39ccab..2206941 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -1864,6 +1864,9 @@ enum nl80211_commands { * mandatory. If this flag is not set, the duration is the maximum duration * and the actual measurement duration may be shorter. * + * @NL80211_ATTR_MESH_PEER_AID: Association ID for the mesh peer (u16). This is + * used to pull the stored data for mesh peer in power save state. + * * @NUM_NL80211_ATTR: total number of nl80211_attrs available * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use @@ -2256,6 +2259,8 @@ enum nl80211_attrs { NL80211_ATTR_MEASUREMENT_DURATION, NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY, + NL80211_ATTR_MESH_PEER_AID, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 0c12e40..47e99ab8 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -997,6 +997,7 @@ static void sta_apply_mesh_params(struct ieee80211_local *local, if (sta->mesh->plink_state != NL80211_PLINK_ESTAB) changed = mesh_plink_inc_estab_count(sdata); sta->mesh->plink_state = params->plink_state; + sta->mesh->aid = params->peer_aid; ieee80211_mps_sta_status_update(sta); changed |= ieee80211_mps_set_sta_local_pm(sta, diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index c53b546..5782f71 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -4446,6 +4446,12 @@ static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info) nla_get_u8(info->attrs[NL80211_ATTR_STA_PLINK_STATE]); if (params.plink_state >= NUM_NL80211_PLINK_STATES) return -EINVAL; + if (info->attrs[NL80211_ATTR_MESH_PEER_AID]) { + params.peer_aid = nla_get_u16( + info->attrs[NL80211_ATTR_MESH_PEER_AID]); + if (params.peer_aid > IEEE80211_MAX_AID) + return -EINVAL; + } params.sta_modify_mask |= STATION_PARAM_APPLY_PLINK_STATE; } -- cgit v0.10.2 From fcf752ae19c1a79e90b8613ffb51c845594e3692 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 6 Jul 2016 13:30:13 +0000 Subject: net: mediatek: remove .owner field for driver Remove .owner field since calls to module_platform_driver() will set it automatically. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 682797e..760f3d7 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -1910,7 +1910,6 @@ static struct platform_driver mtk_driver = { .remove = mtk_remove, .driver = { .name = "mtk_soc_eth", - .owner = THIS_MODULE, .of_match_table = of_mtk_match, }, }; -- cgit v0.10.2 From be2cef49904b34dd5f75d96bbc8cd8341bab1bc0 Mon Sep 17 00:00:00 2001 From: Michal Kubecek Date: Fri, 3 Jun 2016 17:56:50 +0200 Subject: ipvs: count pre-established TCP states as active Some users observed that "least connection" distribution algorithm doesn't handle well bursts of TCP connections from reconnecting clients after a node or network failure. This is because the algorithm counts active connection as worth 256 inactive ones where for TCP, "active" only means TCP connections in ESTABLISHED state. In case of a connection burst, new connections are handled before previous ones have finished the three way handshaking so that all are still counted as "inactive", i.e. cheap ones. The become "active" quickly but at that time, all of them are already assigned to one real server (or few), resulting in highly unbalanced distribution. Address this by counting the "pre-established" states as "active". Signed-off-by: Michal Kubecek Acked-by: Julian Anastasov Signed-off-by: Simon Horman diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index d7024b2..5117bcb 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -395,6 +395,20 @@ static const char *const tcp_state_name_table[IP_VS_TCP_S_LAST+1] = { [IP_VS_TCP_S_LAST] = "BUG!", }; +static const bool tcp_state_active_table[IP_VS_TCP_S_LAST] = { + [IP_VS_TCP_S_NONE] = false, + [IP_VS_TCP_S_ESTABLISHED] = true, + [IP_VS_TCP_S_SYN_SENT] = true, + [IP_VS_TCP_S_SYN_RECV] = true, + [IP_VS_TCP_S_FIN_WAIT] = false, + [IP_VS_TCP_S_TIME_WAIT] = false, + [IP_VS_TCP_S_CLOSE] = false, + [IP_VS_TCP_S_CLOSE_WAIT] = false, + [IP_VS_TCP_S_LAST_ACK] = false, + [IP_VS_TCP_S_LISTEN] = false, + [IP_VS_TCP_S_SYNACK] = true, +}; + #define sNO IP_VS_TCP_S_NONE #define sES IP_VS_TCP_S_ESTABLISHED #define sSS IP_VS_TCP_S_SYN_SENT @@ -418,6 +432,13 @@ static const char * tcp_state_name(int state) return tcp_state_name_table[state] ? tcp_state_name_table[state] : "?"; } +static bool tcp_state_active(int state) +{ + if (state >= IP_VS_TCP_S_LAST) + return false; + return tcp_state_active_table[state]; +} + static struct tcp_states_t tcp_states [] = { /* INPUT */ /* sNO, sES, sSS, sSR, sFW, sTW, sCL, sCW, sLA, sLI, sSA */ @@ -540,12 +561,12 @@ set_tcp_state(struct ip_vs_proto_data *pd, struct ip_vs_conn *cp, if (dest) { if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && - (new_state != IP_VS_TCP_S_ESTABLISHED)) { + !tcp_state_active(new_state)) { atomic_dec(&dest->activeconns); atomic_inc(&dest->inactconns); cp->flags |= IP_VS_CONN_F_INACTIVE; } else if ((cp->flags & IP_VS_CONN_F_INACTIVE) && - (new_state == IP_VS_TCP_S_ESTABLISHED)) { + tcp_state_active(new_state)) { atomic_inc(&dest->activeconns); atomic_dec(&dest->inactconns); cp->flags &= ~IP_VS_CONN_F_INACTIVE; -- cgit v0.10.2 From 4a49ae94a448faa71e89fac8f0b276261123387e Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Thu, 30 Jun 2016 15:23:47 +0300 Subject: ath10k: fix 10.4 extended peer stats update 10.4 'extended peer stats' will be not be appended with normal peer stats data and they shall be coming in separate chunks. Fix this by maintaining a separate linked list 'extender peer stats' for 10.4 and update rx_duration for per station statistics. Also parse through beacon filter (if enabled), to make sure we parse the extended peer stats properly. This issue was exposed when more than one client is connected and extended peer stats for 10.4 is enabled The order for the stats is as below S - standard peer stats, E- extended peer stats, B - beacon filter stats {S1, S2, S3..} -> {B1, B2, B3..}(if available) -> {E1, E2, E3..} Fixes: f9575793d44c ("ath10k: enable parsing per station rx duration for 10.4") Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h index 3da18c9..9374bcd 100644 --- a/drivers/net/wireless/ath/ath10k/core.h +++ b/drivers/net/wireless/ath/ath10k/core.h @@ -165,6 +165,13 @@ struct ath10k_fw_stats_peer { u32 rx_duration; }; +struct ath10k_fw_extd_stats_peer { + struct list_head list; + + u8 peer_macaddr[ETH_ALEN]; + u32 rx_duration; +}; + struct ath10k_fw_stats_vdev { struct list_head list; @@ -256,9 +263,11 @@ struct ath10k_fw_stats_pdev { }; struct ath10k_fw_stats { + bool extended; struct list_head pdevs; struct list_head vdevs; struct list_head peers; + struct list_head peers_extd; }; #define ATH10K_TPC_TABLE_TYPE_FLAG 1 diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c index 8fbb8f2..355e1ae 100644 --- a/drivers/net/wireless/ath/ath10k/debug.c +++ b/drivers/net/wireless/ath/ath10k/debug.c @@ -313,13 +313,25 @@ static void ath10k_fw_stats_peers_free(struct list_head *head) } } +static void ath10k_fw_extd_stats_peers_free(struct list_head *head) +{ + struct ath10k_fw_extd_stats_peer *i, *tmp; + + list_for_each_entry_safe(i, tmp, head, list) { + list_del(&i->list); + kfree(i); + } +} + static void ath10k_debug_fw_stats_reset(struct ath10k *ar) { spin_lock_bh(&ar->data_lock); ar->debug.fw_stats_done = false; + ar->debug.fw_stats.extended = false; ath10k_fw_stats_pdevs_free(&ar->debug.fw_stats.pdevs); ath10k_fw_stats_vdevs_free(&ar->debug.fw_stats.vdevs); ath10k_fw_stats_peers_free(&ar->debug.fw_stats.peers); + ath10k_fw_extd_stats_peers_free(&ar->debug.fw_stats.peers_extd); spin_unlock_bh(&ar->data_lock); } @@ -334,6 +346,7 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb) INIT_LIST_HEAD(&stats.pdevs); INIT_LIST_HEAD(&stats.vdevs); INIT_LIST_HEAD(&stats.peers); + INIT_LIST_HEAD(&stats.peers_extd); spin_lock_bh(&ar->data_lock); ret = ath10k_wmi_pull_fw_stats(ar, skb, &stats); @@ -354,7 +367,7 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb) * delivered which is treated as end-of-data and is itself discarded */ if (ath10k_peer_stats_enabled(ar)) - ath10k_sta_update_rx_duration(ar, &stats.peers); + ath10k_sta_update_rx_duration(ar, &stats); if (ar->debug.fw_stats_done) { if (!ath10k_peer_stats_enabled(ar)) @@ -396,6 +409,8 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *skb) list_splice_tail_init(&stats.peers, &ar->debug.fw_stats.peers); list_splice_tail_init(&stats.vdevs, &ar->debug.fw_stats.vdevs); + list_splice_tail_init(&stats.peers_extd, + &ar->debug.fw_stats.peers_extd); } complete(&ar->debug.fw_stats_complete); @@ -407,6 +422,7 @@ free: ath10k_fw_stats_pdevs_free(&stats.pdevs); ath10k_fw_stats_vdevs_free(&stats.vdevs); ath10k_fw_stats_peers_free(&stats.peers); + ath10k_fw_extd_stats_peers_free(&stats.peers_extd); spin_unlock_bh(&ar->data_lock); } @@ -2330,6 +2346,7 @@ int ath10k_debug_create(struct ath10k *ar) INIT_LIST_HEAD(&ar->debug.fw_stats.pdevs); INIT_LIST_HEAD(&ar->debug.fw_stats.vdevs); INIT_LIST_HEAD(&ar->debug.fw_stats.peers); + INIT_LIST_HEAD(&ar->debug.fw_stats.peers_extd); return 0; } diff --git a/drivers/net/wireless/ath/ath10k/debug.h b/drivers/net/wireless/ath/ath10k/debug.h index 75c89e3..dc549c4 100644 --- a/drivers/net/wireless/ath/ath10k/debug.h +++ b/drivers/net/wireless/ath/ath10k/debug.h @@ -154,10 +154,12 @@ ath10k_debug_get_new_fw_crash_data(struct ath10k *ar) #ifdef CONFIG_MAC80211_DEBUGFS void ath10k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_sta *sta, struct dentry *dir); -void ath10k_sta_update_rx_duration(struct ath10k *ar, struct list_head *peer); +void ath10k_sta_update_rx_duration(struct ath10k *ar, + struct ath10k_fw_stats *stats); #else -static inline void ath10k_sta_update_rx_duration(struct ath10k *ar, - struct list_head *peer) +static inline +void ath10k_sta_update_rx_duration(struct ath10k *ar, + struct ath10k_fw_stats *stats) { } #endif /* CONFIG_MAC80211_DEBUGFS */ diff --git a/drivers/net/wireless/ath/ath10k/debugfs_sta.c b/drivers/net/wireless/ath/ath10k/debugfs_sta.c index 67ef75b..0da8a57 100644 --- a/drivers/net/wireless/ath/ath10k/debugfs_sta.c +++ b/drivers/net/wireless/ath/ath10k/debugfs_sta.c @@ -18,13 +18,34 @@ #include "wmi-ops.h" #include "debug.h" -void ath10k_sta_update_rx_duration(struct ath10k *ar, struct list_head *head) -{ struct ieee80211_sta *sta; +static void ath10k_sta_update_extd_stats_rx_duration(struct ath10k *ar, + struct ath10k_fw_stats *stats) +{ + struct ath10k_fw_extd_stats_peer *peer; + struct ieee80211_sta *sta; + struct ath10k_sta *arsta; + + rcu_read_lock(); + list_for_each_entry(peer, &stats->peers_extd, list) { + sta = ieee80211_find_sta_by_ifaddr(ar->hw, peer->peer_macaddr, + NULL); + if (!sta) + continue; + arsta = (struct ath10k_sta *)sta->drv_priv; + arsta->rx_duration += (u64)peer->rx_duration; + } + rcu_read_unlock(); +} + +static void ath10k_sta_update_stats_rx_duration(struct ath10k *ar, + struct ath10k_fw_stats *stats) +{ struct ath10k_fw_stats_peer *peer; + struct ieee80211_sta *sta; struct ath10k_sta *arsta; rcu_read_lock(); - list_for_each_entry(peer, head, list) { + list_for_each_entry(peer, &stats->peers, list) { sta = ieee80211_find_sta_by_ifaddr(ar->hw, peer->peer_macaddr, NULL); if (!sta) @@ -35,6 +56,15 @@ void ath10k_sta_update_rx_duration(struct ath10k *ar, struct list_head *head) rcu_read_unlock(); } +void ath10k_sta_update_rx_duration(struct ath10k *ar, + struct ath10k_fw_stats *stats) +{ + if (stats->extended) + ath10k_sta_update_extd_stats_rx_duration(ar, stats); + else + ath10k_sta_update_stats_rx_duration(ar, stats); +} + static ssize_t ath10k_dbg_sta_read_aggr_mode(struct file *file, char __user *user_buf, size_t count, loff_t *ppos) diff --git a/drivers/net/wireless/ath/ath10k/wmi.c b/drivers/net/wireless/ath/ath10k/wmi.c index 12a4347..169cd2e7 100644 --- a/drivers/net/wireless/ath/ath10k/wmi.c +++ b/drivers/net/wireless/ath/ath10k/wmi.c @@ -2926,6 +2926,7 @@ static int ath10k_wmi_10_4_op_pull_fw_stats(struct ath10k *ar, u32 num_pdev_ext_stats; u32 num_vdev_stats; u32 num_peer_stats; + u32 num_bcnflt_stats; u32 stats_id; int i; @@ -2936,6 +2937,7 @@ static int ath10k_wmi_10_4_op_pull_fw_stats(struct ath10k *ar, num_pdev_ext_stats = __le32_to_cpu(ev->num_pdev_ext_stats); num_vdev_stats = __le32_to_cpu(ev->num_vdev_stats); num_peer_stats = __le32_to_cpu(ev->num_peer_stats); + num_bcnflt_stats = __le32_to_cpu(ev->num_bcnflt_stats); stats_id = __le32_to_cpu(ev->stats_id); for (i = 0; i < num_pdev_stats; i++) { @@ -2976,32 +2978,57 @@ static int ath10k_wmi_10_4_op_pull_fw_stats(struct ath10k *ar, /* fw doesn't implement vdev stats */ for (i = 0; i < num_peer_stats; i++) { - const struct wmi_10_4_peer_extd_stats *src; + const struct wmi_10_4_peer_stats *src; struct ath10k_fw_stats_peer *dst; - int stats_len; - bool extd_peer_stats = !!(stats_id & WMI_10_4_STAT_PEER_EXTD); - - if (extd_peer_stats) - stats_len = sizeof(struct wmi_10_4_peer_extd_stats); - else - stats_len = sizeof(struct wmi_10_4_peer_stats); src = (void *)skb->data; - if (!skb_pull(skb, stats_len)) + if (!skb_pull(skb, sizeof(*src))) return -EPROTO; dst = kzalloc(sizeof(*dst), GFP_ATOMIC); if (!dst) continue; - ath10k_wmi_10_4_pull_peer_stats(&src->common, dst); - /* FIXME: expose 10.4 specific values */ - if (extd_peer_stats) - dst->rx_duration = __le32_to_cpu(src->rx_duration); - + ath10k_wmi_10_4_pull_peer_stats(src, dst); list_add_tail(&dst->list, &stats->peers); } + for (i = 0; i < num_bcnflt_stats; i++) { + const struct wmi_10_4_bss_bcn_filter_stats *src; + + src = (void *)skb->data; + if (!skb_pull(skb, sizeof(*src))) + return -EPROTO; + + /* FIXME: expose values to userspace + * + * Note: Even though this loop seems to do nothing it is + * required to parse following sub-structures properly. + */ + } + + if ((stats_id & WMI_10_4_STAT_PEER_EXTD) == 0) + return 0; + + stats->extended = true; + + for (i = 0; i < num_peer_stats; i++) { + const struct wmi_10_4_peer_extd_stats *src; + struct ath10k_fw_extd_stats_peer *dst; + + src = (void *)skb->data; + if (!skb_pull(skb, sizeof(*src))) + return -EPROTO; + + dst = kzalloc(sizeof(*dst), GFP_ATOMIC); + if (!dst) + continue; + + ether_addr_copy(dst->peer_macaddr, src->peer_macaddr.addr); + dst->rx_duration = __le32_to_cpu(src->rx_duration); + list_add_tail(&dst->list, &stats->peers_extd); + } + return 0; } diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireless/ath/ath10k/wmi.h index 90f594e..3ef4688 100644 --- a/drivers/net/wireless/ath/ath10k/wmi.h +++ b/drivers/net/wireless/ath/ath10k/wmi.h @@ -4340,7 +4340,6 @@ struct wmi_10_4_peer_stats { } __packed; struct wmi_10_4_peer_extd_stats { - struct wmi_10_4_peer_stats common; struct wmi_mac_addr peer_macaddr; __le32 inactive_time; __le32 peer_chain_rssi; @@ -4348,6 +4347,19 @@ struct wmi_10_4_peer_extd_stats { __le32 reserved[10]; } __packed; +struct wmi_10_4_bss_bcn_stats { + __le32 vdev_id; + __le32 bss_bcns_dropped; + __le32 bss_bcn_delivered; +} __packed; + +struct wmi_10_4_bss_bcn_filter_stats { + __le32 bcns_dropped; + __le32 bcns_delivered; + __le32 active_filters; + struct wmi_10_4_bss_bcn_stats bss_stats; +} __packed; + struct wmi_10_2_pdev_ext_stats { __le32 rx_rssi_comb; __le32 rx_rssi[4]; -- cgit v0.10.2 From 120a1f02a5c4a3c91a9b082713ca8cd32f9acf76 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Thu, 30 Jun 2016 15:23:50 +0300 Subject: ath10k: add support for ath10k_sta_statistics support Enable support for 'drv_sta_statistics' callback. Export rx_duration support if available to cfg80211/nl80211 This can also act as a placeholder for any new per STA stats support Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/debug.h b/drivers/net/wireless/ath/ath10k/debug.h index dc549c4..c458fa9 100644 --- a/drivers/net/wireless/ath/ath10k/debug.h +++ b/drivers/net/wireless/ath/ath10k/debug.h @@ -156,6 +156,9 @@ void ath10k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_sta *sta, struct dentry *dir); void ath10k_sta_update_rx_duration(struct ath10k *ar, struct ath10k_fw_stats *stats); +void ath10k_sta_statistics(struct ieee80211_hw *hw, struct ieee80211_vif *vif, + struct ieee80211_sta *sta, + struct station_info *sinfo); #else static inline void ath10k_sta_update_rx_duration(struct ath10k *ar, diff --git a/drivers/net/wireless/ath/ath10k/debugfs_sta.c b/drivers/net/wireless/ath/ath10k/debugfs_sta.c index 0da8a57..60e5da0 100644 --- a/drivers/net/wireless/ath/ath10k/debugfs_sta.c +++ b/drivers/net/wireless/ath/ath10k/debugfs_sta.c @@ -65,6 +65,20 @@ void ath10k_sta_update_rx_duration(struct ath10k *ar, ath10k_sta_update_stats_rx_duration(ar, stats); } +void ath10k_sta_statistics(struct ieee80211_hw *hw, struct ieee80211_vif *vif, + struct ieee80211_sta *sta, + struct station_info *sinfo) +{ + struct ath10k_sta *arsta = (struct ath10k_sta *)sta->drv_priv; + struct ath10k *ar = arsta->arvif->ar; + + if (!ath10k_peer_stats_enabled(ar)) + return; + + sinfo->rx_duration = arsta->rx_duration; + sinfo->filled |= 1ULL << NL80211_STA_INFO_RX_DURATION; +} + static ssize_t ath10k_dbg_sta_read_aggr_mode(struct file *file, char __user *user_buf, size_t count, loff_t *ppos) diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index e9d464c..c6c4ae7 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -7407,6 +7407,7 @@ static const struct ieee80211_ops ath10k_ops = { #endif #ifdef CONFIG_MAC80211_DEBUGFS .sta_add_debugfs = ath10k_sta_add_debugfs, + .sta_statistics = ath10k_sta_statistics, #endif }; -- cgit v0.10.2 From 2ba1f3709452a1e55b9944bdda88b043b6b3fad0 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Thu, 30 Jun 2016 15:23:50 +0300 Subject: ath10k: remove debugfs support for Per STA total rx duration cfg80211/nl80211 interface changes for per STA total rx-duration and very basic 'ath10k_sta_statistics' mac80211 callback is implemented to extend support for per station statistics from the driver. Also provision in 'iw dev wlan#N station dump' to parse rx-duration is supported. So its safer to remove the debugfs entry for per STA rx-duration Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/debugfs_sta.c b/drivers/net/wireless/ath/ath10k/debugfs_sta.c index 60e5da0..9955fea 100644 --- a/drivers/net/wireless/ath/ath10k/debugfs_sta.c +++ b/drivers/net/wireless/ath/ath10k/debugfs_sta.c @@ -293,28 +293,6 @@ static const struct file_operations fops_delba = { .llseek = default_llseek, }; -static ssize_t ath10k_dbg_sta_read_rx_duration(struct file *file, - char __user *user_buf, - size_t count, loff_t *ppos) -{ - struct ieee80211_sta *sta = file->private_data; - struct ath10k_sta *arsta = (struct ath10k_sta *)sta->drv_priv; - char buf[100]; - int len = 0; - - len = scnprintf(buf, sizeof(buf), - "%llu usecs\n", arsta->rx_duration); - - return simple_read_from_buffer(user_buf, count, ppos, buf, len); -} - -static const struct file_operations fops_rx_duration = { - .read = ath10k_dbg_sta_read_rx_duration, - .open = simple_open, - .owner = THIS_MODULE, - .llseek = default_llseek, -}; - void ath10k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_sta *sta, struct dentry *dir) { @@ -323,6 +301,4 @@ void ath10k_sta_add_debugfs(struct ieee80211_hw *hw, struct ieee80211_vif *vif, debugfs_create_file("addba", S_IWUSR, dir, sta, &fops_addba); debugfs_create_file("addba_resp", S_IWUSR, dir, sta, &fops_addba_resp); debugfs_create_file("delba", S_IWUSR, dir, sta, &fops_delba); - debugfs_create_file("rx_duration", S_IRUGO, dir, sta, - &fops_rx_duration); } -- cgit v0.10.2 From de0170beaa88bc5d7210a72db7cd5738a59bc23d Mon Sep 17 00:00:00 2001 From: Ben Greear Date: Thu, 30 Jun 2016 15:23:53 +0300 Subject: ath10k: ensure txrx-compl-task is stopped when cleaning htt-tx Otherwise, the txrx-compl-task may access some bad memory? Signed-off-by: Ben Greear Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/htt_tx.c b/drivers/net/wireless/ath/ath10k/htt_tx.c index ae5b33f..7c072b6 100644 --- a/drivers/net/wireless/ath/ath10k/htt_tx.c +++ b/drivers/net/wireless/ath/ath10k/htt_tx.c @@ -390,6 +390,8 @@ void ath10k_htt_tx_free(struct ath10k_htt *htt) { int size; + tasklet_kill(&htt->txrx_compl_task); + idr_for_each(&htt->pending_tx, ath10k_htt_tx_clean_up_pending, htt->ar); idr_destroy(&htt->pending_tx); -- cgit v0.10.2 From 6d68f7900d2572372557dc8d96b6377bc8512f69 Mon Sep 17 00:00:00 2001 From: Ben Greear Date: Thu, 30 Jun 2016 15:23:53 +0300 Subject: ath10k: ensure peer_map references are cleaned up While debugging OS crashes due to firmware crashes, I enabled kasan, and it noticed that peer objects were being used-after-freed. Looks like there are two places we could be leaving stale references in the peer-map, so clean that up. Signed-off-by: Ben Greear Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index c6c4ae7..78add06 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -802,6 +802,7 @@ static void ath10k_peer_cleanup(struct ath10k *ar, u32 vdev_id) { struct ath10k_peer *peer, *tmp; int peer_id; + int i; lockdep_assert_held(&ar->conf_mutex); @@ -818,6 +819,17 @@ static void ath10k_peer_cleanup(struct ath10k *ar, u32 vdev_id) ar->peer_map[peer_id] = NULL; } + /* Double check that peer is properly un-referenced from + * the peer_map + */ + for (i = 0; i < ARRAY_SIZE(ar->peer_map); i++) { + if (ar->peer_map[i] == peer) { + ath10k_warn(ar, "removing stale peer_map entry for %pM (ptr %p idx %d)\n", + peer->addr, peer, i); + ar->peer_map[i] = NULL; + } + } + list_del(&peer->list); kfree(peer); ar->num_peers--; @@ -828,6 +840,7 @@ static void ath10k_peer_cleanup(struct ath10k *ar, u32 vdev_id) static void ath10k_peer_cleanup_all(struct ath10k *ar) { struct ath10k_peer *peer, *tmp; + int i; lockdep_assert_held(&ar->conf_mutex); @@ -836,6 +849,10 @@ static void ath10k_peer_cleanup_all(struct ath10k *ar) list_del(&peer->list); kfree(peer); } + + for (i = 0; i < ARRAY_SIZE(ar->peer_map); i++) + ar->peer_map[i] = NULL; + spin_unlock_bh(&ar->data_lock); ar->num_peers = 0; -- cgit v0.10.2 From a3dadad73324e95a461235ec016e686445bd468d Mon Sep 17 00:00:00 2001 From: Chaehyun Lim Date: Thu, 30 Jun 2016 15:23:54 +0300 Subject: ath10k: remove unused is not used anymore, so just remove the include. Signed-off-by: Chaehyun Lim Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/htc.h b/drivers/net/wireless/ath/ath10k/htc.h index cc82718..0c55cd9 100644 --- a/drivers/net/wireless/ath/ath10k/htc.h +++ b/drivers/net/wireless/ath/ath10k/htc.h @@ -22,7 +22,6 @@ #include #include #include -#include #include struct ath10k; -- cgit v0.10.2 From 34293f75586bab3bf7e8e5acb4aa9e65c81f72e6 Mon Sep 17 00:00:00 2001 From: Ashok Raj Nagarajan Date: Thu, 30 Jun 2016 15:23:55 +0300 Subject: ath10k: simplify pktlog htt event processing It is expected that all pktlog events for 10.4 firmware based solutions should come through CE8 where as in case of 10.2 firmware based solutions, it should come through one of the HTT events (HTT_T2H_MSG_TYPE_PKTLOG). But from experiments with 10.4 based solutions, it is observed that pktlog event for ATH_PKTLOG_TYPE_TX_MSDU_ID is coming through HTT pktlog event. Currently, we always parse with 10.2 pktlog header which will lead to pktlog decoding issues (payload length mismatch exceptions) For trace points, it is required to provide only the payload size. So fixing this by simplifying the payload size calculation without the use of ath10k pktlog headers. While there, remove the unused ath10k pktlog headers. Signed-off-by: Ashok Raj Nagarajan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c index 8bae023..5ec7ac1 100644 --- a/drivers/net/wireless/ath/ath10k/htt_rx.c +++ b/drivers/net/wireless/ath/ath10k/htt_rx.c @@ -2308,12 +2308,10 @@ bool ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb) ath10k_htt_rx_delba(ar, resp); break; case HTT_T2H_MSG_TYPE_PKTLOG: { - struct ath10k_pktlog_hdr *hdr = - (struct ath10k_pktlog_hdr *)resp->pktlog_msg.payload; - trace_ath10k_htt_pktlog(ar, resp->pktlog_msg.payload, - sizeof(*hdr) + - __le16_to_cpu(hdr->size)); + skb->len - + offsetof(struct htt_resp, + pktlog_msg.payload)); break; } case HTT_T2H_MSG_TYPE_RX_FLUSH: { diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h index f31d3ce..427b720 100644 --- a/drivers/net/wireless/ath/ath10k/hw.h +++ b/drivers/net/wireless/ath/ath10k/hw.h @@ -296,25 +296,6 @@ enum ath10k_mcast2ucast_mode { ATH10K_MCAST2UCAST_ENABLED = 1, }; -struct ath10k_pktlog_hdr { - __le16 flags; - __le16 missed_cnt; - __le16 log_type; - __le16 size; - __le32 timestamp; - u8 payload[0]; -} __packed; - -struct ath10k_pktlog_10_4_hdr { - __le16 flags; - __le16 missed_cnt; - __le16 log_type; - __le16 size; - __le32 timestamp; - __le32 type_specific_data; - u8 payload[0]; -} __packed; - enum ath10k_hw_rate_ofdm { ATH10K_HW_RATE_OFDM_48M = 0, ATH10K_HW_RATE_OFDM_24M, -- cgit v0.10.2 From 9802977dcce58c585086bd270401c7881fe8bca0 Mon Sep 17 00:00:00 2001 From: Eduardo Abinader Date: Thu, 30 Jun 2016 15:23:55 +0300 Subject: ath10k: remove extra space on ath10k_update_channel_list just to comply to coding style. Signed-off-by: Eduardo Abinader Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index 78add06..60a5324 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -2956,7 +2956,7 @@ static int ath10k_update_channel_list(struct ath10k *ar) if (channel->flags & IEEE80211_CHAN_DISABLED) continue; - ch->allow_ht = true; + ch->allow_ht = true; /* FIXME: when should we really allow VHT? */ ch->allow_vht = true; -- cgit v0.10.2 From c5ace87a886d8ae35a7fc75a8140c2a7b3ff79b1 Mon Sep 17 00:00:00 2001 From: Ben Greear Date: Thu, 30 Jun 2016 15:23:58 +0300 Subject: ath10k: Add WARN_ON if we over-write peer-map pointer. Not sure this can happen, but seems like a reasonable sanity check. Signed-off-by: Ben Greear Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/txrx.c b/drivers/net/wireless/ath/ath10k/txrx.c index f524ac0..b29a86a 100644 --- a/drivers/net/wireless/ath/ath10k/txrx.c +++ b/drivers/net/wireless/ath/ath10k/txrx.c @@ -217,6 +217,7 @@ void ath10k_peer_map_event(struct ath10k_htt *htt, ath10k_dbg(ar, ATH10K_DBG_HTT, "htt peer map vdev %d peer %pM id %d\n", ev->vdev_id, ev->addr, ev->peer_id); + WARN_ON(ar->peer_map[ev->peer_id] && (ar->peer_map[ev->peer_id] != peer)); ar->peer_map[ev->peer_id] = peer; set_bit(ev->peer_id, peer->peer_ids); exit: -- cgit v0.10.2 From d0eeafad118940fe445ca00f45be5624fea2ec34 Mon Sep 17 00:00:00 2001 From: Ben Greear Date: Thu, 30 Jun 2016 15:23:59 +0300 Subject: ath10k: Clean up peer when sta goes away. If WMI and/or firmware has issues removing the peer object, then we still need to clean up the peer object in the driver. Signed-off-by: Ben Greear Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index 60a5324..532c077 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -5997,9 +5997,17 @@ static int ath10k_sta_state(struct ieee80211_hw *hw, continue; if (peer->sta == sta) { - ath10k_warn(ar, "found sta peer %pM entry on vdev %i after it was supposedly removed\n", - sta->addr, arvif->vdev_id); + ath10k_warn(ar, "found sta peer %pM (ptr %p id %d) entry on vdev %i after it was supposedly removed\n", + sta->addr, peer, i, arvif->vdev_id); peer->sta = NULL; + + /* Clean up the peer object as well since we + * must have failed to do this above. + */ + list_del(&peer->list); + ar->peer_map[i] = NULL; + kfree(peer); + ar->num_peers--; } } spin_unlock_bh(&ar->data_lock); -- cgit v0.10.2 From 2225378d840cc16d13b55df466dca6bb3d10e6bc Mon Sep 17 00:00:00 2001 From: Anilkumar Kolli Date: Thu, 30 Jun 2016 15:24:00 +0300 Subject: ath10k: remove unused member in ath10k_hw_regs rtc_state_cold_reset_mask is unused in ath10k_hw_regs. instead fixed delays are used. Signed-off-by: Anilkumar Kolli Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/hw.c b/drivers/net/wireless/ath/ath10k/hw.c index bd86e7a..af3e214 100644 --- a/drivers/net/wireless/ath/ath10k/hw.c +++ b/drivers/net/wireless/ath/ath10k/hw.c @@ -19,7 +19,6 @@ #include "hw.h" const struct ath10k_hw_regs qca988x_regs = { - .rtc_state_cold_reset_mask = 0x00000400, .rtc_soc_base_address = 0x00004000, .rtc_wmac_base_address = 0x00005000, .soc_core_base_address = 0x00009000, @@ -46,7 +45,6 @@ const struct ath10k_hw_regs qca988x_regs = { }; const struct ath10k_hw_regs qca6174_regs = { - .rtc_state_cold_reset_mask = 0x00002000, .rtc_soc_base_address = 0x00000800, .rtc_wmac_base_address = 0x00001000, .soc_core_base_address = 0x0003a000, @@ -73,7 +71,6 @@ const struct ath10k_hw_regs qca6174_regs = { }; const struct ath10k_hw_regs qca99x0_regs = { - .rtc_state_cold_reset_mask = 0x00000400, .rtc_soc_base_address = 0x00080000, .rtc_wmac_base_address = 0x00000000, .soc_core_base_address = 0x00082000, diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h index 427b720..4cfea00 100644 --- a/drivers/net/wireless/ath/ath10k/hw.h +++ b/drivers/net/wireless/ath/ath10k/hw.h @@ -217,7 +217,6 @@ enum ath10k_hw_rev { }; struct ath10k_hw_regs { - u32 rtc_state_cold_reset_mask; u32 rtc_soc_base_address; u32 rtc_wmac_base_address; u32 soc_core_base_address; @@ -516,7 +515,6 @@ enum ath10k_hw_cc_wraparound_type { /* as of IP3.7.1 */ #define RTC_STATE_V_ON ar->hw_values->rtc_state_val_on -#define RTC_STATE_COLD_RESET_MASK ar->regs->rtc_state_cold_reset_mask #define RTC_STATE_V_LSB 0 #define RTC_STATE_V_MASK 0x00000007 #define RTC_STATE_ADDRESS 0x0000 -- cgit v0.10.2 From e565c3125e03bb4d3fe99d98d7e3f511e1073f06 Mon Sep 17 00:00:00 2001 From: Anilkumar Kolli Date: Thu, 30 Jun 2016 15:24:00 +0300 Subject: ath10k: enable support for QCA9888 QCA9888 shares the same configuration with QCA99X0 with NSS=2. Signed-off-by: Anilkumar Kolli Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index c6291c2..e48eefd 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -207,6 +207,28 @@ static const struct ath10k_hw_params ath10k_hw_params_list[] = { }, }, { + .id = QCA9888_HW_2_0_DEV_VERSION, + .dev_id = QCA9888_2_0_DEVICE_ID, + .name = "qca9888 hw2.0", + .patch_load_addr = QCA9888_HW_2_0_PATCH_LOAD_ADDR, + .uart_pin = 7, + .otp_exe_param = 0x00000700, + .continuous_frag_desc = true, + .channel_counters_freq_hz = 150000, + .max_probe_resp_desc_thres = 24, + .hw_4addr_pad = ATH10K_HW_4ADDR_PAD_BEFORE, + .tx_chain_mask = 3, + .rx_chain_mask = 3, + .max_spatial_stream = 2, + .cal_data_len = 12064, + .fw = { + .dir = QCA9888_HW_2_0_FW_DIR, + .board = QCA9888_HW_2_0_BOARD_DATA_FILE, + .board_size = QCA99X0_BOARD_DATA_SZ, + .board_ext_size = QCA99X0_BOARD_EXT_DATA_SZ, + }, + }, + { .id = QCA9377_HW_1_0_DEV_VERSION, .dev_id = QCA9377_1_0_DEVICE_ID, .name = "qca9377 hw1.0", @@ -2171,6 +2193,10 @@ struct ath10k *ath10k_core_create(size_t priv_size, struct device *dev, ar->regs = &qca99x0_regs; ar->hw_values = &qca99x0_values; break; + case ATH10K_HW_QCA9888: + ar->regs = &qca99x0_regs; + ar->hw_values = &qca9888_values; + break; case ATH10K_HW_QCA4019: ar->regs = &qca4019_regs; ar->hw_values = &qca4019_values; diff --git a/drivers/net/wireless/ath/ath10k/hw.c b/drivers/net/wireless/ath/ath10k/hw.c index af3e214..f1e0695 100644 --- a/drivers/net/wireless/ath/ath10k/hw.c +++ b/drivers/net/wireless/ath/ath10k/hw.c @@ -165,6 +165,15 @@ const struct ath10k_hw_values qca99x0_values = { .ce_desc_meta_data_lsb = 4, }; +const struct ath10k_hw_values qca9888_values = { + .rtc_state_val_on = 3, + .ce_count = 12, + .msi_assign_ce_max = 12, + .num_target_ce_config_wlan = 10, + .ce_desc_meta_data_mask = 0xFFF0, + .ce_desc_meta_data_lsb = 4, +}; + const struct ath10k_hw_values qca4019_values = { .ce_count = 12, .num_target_ce_config_wlan = 10, diff --git a/drivers/net/wireless/ath/ath10k/hw.h b/drivers/net/wireless/ath/ath10k/hw.h index 4cfea00..e014cd7 100644 --- a/drivers/net/wireless/ath/ath10k/hw.h +++ b/drivers/net/wireless/ath/ath10k/hw.h @@ -26,6 +26,7 @@ #define QCA6164_2_1_DEVICE_ID (0x0041) #define QCA6174_2_1_DEVICE_ID (0x003e) #define QCA99X0_2_0_DEVICE_ID (0x0040) +#define QCA9888_2_0_DEVICE_ID (0x0056) #define QCA9984_1_0_DEVICE_ID (0x0046) #define QCA9377_1_0_DEVICE_ID (0x0042) #define QCA9887_1_0_DEVICE_ID (0x0050) @@ -108,6 +109,14 @@ enum qca9377_chip_id_rev { #define QCA9984_HW_1_0_BOARD_DATA_FILE "board.bin" #define QCA9984_HW_1_0_PATCH_LOAD_ADDR 0x1234 +/* QCA9888 2.0 defines */ +#define QCA9888_HW_2_0_DEV_VERSION 0x1000000 +#define QCA9888_HW_DEV_TYPE 0xc +#define QCA9888_HW_2_0_CHIP_ID_REV 0x0 +#define QCA9888_HW_2_0_FW_DIR ATH10K_FW_DIR "/QCA9888/hw2.0" +#define QCA9888_HW_2_0_BOARD_DATA_FILE "board.bin" +#define QCA9888_HW_2_0_PATCH_LOAD_ADDR 0x1234 + /* QCA9377 1.0 definitions */ #define QCA9377_HW_1_0_FW_DIR ATH10K_FW_DIR "/QCA9377/hw1.0" #define QCA9377_HW_1_0_BOARD_DATA_FILE "board.bin" @@ -210,6 +219,7 @@ enum ath10k_hw_rev { ATH10K_HW_QCA988X, ATH10K_HW_QCA6174, ATH10K_HW_QCA99X0, + ATH10K_HW_QCA9888, ATH10K_HW_QCA9984, ATH10K_HW_QCA9377, ATH10K_HW_QCA4019, @@ -259,6 +269,7 @@ struct ath10k_hw_values { extern const struct ath10k_hw_values qca988x_values; extern const struct ath10k_hw_values qca6174_values; extern const struct ath10k_hw_values qca99x0_values; +extern const struct ath10k_hw_values qca9888_values; extern const struct ath10k_hw_values qca4019_values; void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey, @@ -268,6 +279,7 @@ void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey, #define QCA_REV_9887(ar) ((ar)->hw_rev == ATH10K_HW_QCA9887) #define QCA_REV_6174(ar) ((ar)->hw_rev == ATH10K_HW_QCA6174) #define QCA_REV_99X0(ar) ((ar)->hw_rev == ATH10K_HW_QCA99X0) +#define QCA_REV_9888(ar) ((ar)->hw_rev == ATH10K_HW_QCA9888) #define QCA_REV_9984(ar) ((ar)->hw_rev == ATH10K_HW_QCA9984) #define QCA_REV_9377(ar) ((ar)->hw_rev == ATH10K_HW_QCA9377) #define QCA_REV_40XX(ar) ((ar)->hw_rev == ATH10K_HW_QCA4019) diff --git a/drivers/net/wireless/ath/ath10k/pci.c b/drivers/net/wireless/ath/ath10k/pci.c index 20128fe..9a22c47 100644 --- a/drivers/net/wireless/ath/ath10k/pci.c +++ b/drivers/net/wireless/ath/ath10k/pci.c @@ -56,6 +56,7 @@ static const struct pci_device_id ath10k_pci_id_table[] = { { PCI_VDEVICE(ATHEROS, QCA6164_2_1_DEVICE_ID) }, /* PCI-E QCA6164 V2.1 */ { PCI_VDEVICE(ATHEROS, QCA6174_2_1_DEVICE_ID) }, /* PCI-E QCA6174 V2.1 */ { PCI_VDEVICE(ATHEROS, QCA99X0_2_0_DEVICE_ID) }, /* PCI-E QCA99X0 V2 */ + { PCI_VDEVICE(ATHEROS, QCA9888_2_0_DEVICE_ID) }, /* PCI-E QCA9888 V2 */ { PCI_VDEVICE(ATHEROS, QCA9984_1_0_DEVICE_ID) }, /* PCI-E QCA9984 V1 */ { PCI_VDEVICE(ATHEROS, QCA9377_1_0_DEVICE_ID) }, /* PCI-E QCA9377 V1 */ { PCI_VDEVICE(ATHEROS, QCA9887_1_0_DEVICE_ID) }, /* PCI-E QCA9887 */ @@ -85,6 +86,8 @@ static const struct ath10k_pci_supp_chip ath10k_pci_supp_chips[] = { { QCA9984_1_0_DEVICE_ID, QCA9984_HW_1_0_CHIP_ID_REV }, + { QCA9888_2_0_DEVICE_ID, QCA9888_HW_2_0_CHIP_ID_REV }, + { QCA9377_1_0_DEVICE_ID, QCA9377_HW_1_0_CHIP_ID_REV }, { QCA9377_1_0_DEVICE_ID, QCA9377_HW_1_1_CHIP_ID_REV }, @@ -850,6 +853,7 @@ static u32 ath10k_pci_targ_cpu_to_ce_addr(struct ath10k *ar, u32 addr) CORE_CTRL_ADDRESS) & 0x7ff) << 21; break; + case ATH10K_HW_QCA9888: case ATH10K_HW_QCA99X0: case ATH10K_HW_QCA9984: case ATH10K_HW_QCA4019: @@ -1583,6 +1587,7 @@ static void ath10k_pci_irq_msi_fw_mask(struct ath10k *ar) break; case ATH10K_HW_QCA99X0: case ATH10K_HW_QCA9984: + case ATH10K_HW_QCA9888: case ATH10K_HW_QCA4019: /* TODO: Find appropriate register configuration for QCA99X0 * to mask irq/MSI. @@ -1608,6 +1613,7 @@ static void ath10k_pci_irq_msi_fw_unmask(struct ath10k *ar) break; case ATH10K_HW_QCA99X0: case ATH10K_HW_QCA9984: + case ATH10K_HW_QCA9888: case ATH10K_HW_QCA4019: /* TODO: Find appropriate register configuration for QCA99X0 * to unmask irq/MSI. @@ -1948,6 +1954,7 @@ static int ath10k_pci_get_num_banks(struct ath10k *ar) switch (ar_pci->pdev->device) { case QCA988X_2_0_DEVICE_ID: case QCA99X0_2_0_DEVICE_ID: + case QCA9888_2_0_DEVICE_ID: case QCA9984_1_0_DEVICE_ID: case QCA9887_1_0_DEVICE_ID: return 1; @@ -3180,6 +3187,12 @@ static int ath10k_pci_probe(struct pci_dev *pdev, pci_soft_reset = ath10k_pci_qca99x0_soft_chip_reset; pci_hard_reset = ath10k_pci_qca99x0_chip_reset; break; + case QCA9888_2_0_DEVICE_ID: + hw_rev = ATH10K_HW_QCA9888; + pci_ps = false; + pci_soft_reset = ath10k_pci_qca99x0_soft_chip_reset; + pci_hard_reset = ath10k_pci_qca99x0_chip_reset; + break; case QCA9377_1_0_DEVICE_ID: hw_rev = ATH10K_HW_QCA9377; pci_ps = true; -- cgit v0.10.2 From ce30c4fe1a229bffab3fe3594306f332e120c199 Mon Sep 17 00:00:00 2001 From: Mohammed Shafi Shajakhan Date: Thu, 30 Jun 2016 15:24:01 +0300 Subject: ath10k: replace warning with an error message if HTT op version is unset Print an ath10k error message rather a call trace when HTT op version is not found from firmware META data (IE). This should be sufficient to figure out what went wrong. Signed-off-by: Mohammed Shafi Shajakhan Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c index e48eefd..fa6a0f9 100644 --- a/drivers/net/wireless/ath/ath10k/core.c +++ b/drivers/net/wireless/ath/ath10k/core.c @@ -1697,7 +1697,7 @@ static int ath10k_core_init_firmware_features(struct ath10k *ar) case ATH10K_FW_WMI_OP_VERSION_10_4: case ATH10K_FW_WMI_OP_VERSION_UNSET: case ATH10K_FW_WMI_OP_VERSION_MAX: - WARN_ON(1); + ath10k_err(ar, "htt op version not found from fw meta data"); return -EINVAL; } } -- cgit v0.10.2 From efcb32883f365cb458a5cb376132965eacaf5571 Mon Sep 17 00:00:00 2001 From: Luis de Bethencourt Date: Thu, 23 Jun 2016 14:35:52 -0400 Subject: rtl8xxxu: remove unneeded assignments reg_eac and reg_ecc are only used if candidate is bigger than 0, and in that case new values will be given to them. Removing the unused assignments. Signed-off-by: Luis de Bethencourt Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c index 65c079a..b50bf2b 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192e.c @@ -1244,11 +1244,9 @@ static void rtl8192eu_phy_iq_calibrate(struct rtl8xxxu_priv *priv) reg_e94 = result[i][0]; reg_e9c = result[i][1]; reg_ea4 = result[i][2]; - reg_eac = result[i][3]; reg_eb4 = result[i][4]; reg_ebc = result[i][5]; reg_ec4 = result[i][6]; - reg_ecc = result[i][7]; } if (candidate >= 0) { -- cgit v0.10.2 From 78383ac951816a64657ca3e17868ad8c7c1b9393 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Thu, 23 Jun 2016 14:35:53 -0400 Subject: rtl8xxxu: Reduce console noise when removing the kernel module USB urbs will return with a status != 0 when rmmod'ing the driver. No need to fill the log with messages from rtl8xxxu_int_complete() Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index 9f6dbb4..cfa5528 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -5267,7 +5267,7 @@ static void rtl8xxxu_int_complete(struct urb *urb) if (ret) usb_unanchor_urb(urb); } else { - dev_info(dev, "%s: Error %i\n", __func__, urb->status); + dev_dbg(dev, "%s: Error %i\n", __func__, urb->status); } } -- cgit v0.10.2 From 9ce221915a94c81779140fb52092b6f608c2ff33 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sat, 25 Jun 2016 15:46:45 -0700 Subject: rtlwifi: Create _rtl_dbg_trace function to reduce RT_TRACE code size This debugging macro can expand to a lot of code. Make it a function to reduce code size. (x86-64 defconfig w/ all rtlwifi drivers and allyesconfig) $ size drivers/net/wireless/realtek/rtlwifi/built-in.o* text data bss dec hex filename 900083 200499 1907 1102489 10d299 drivers/net/wireless/realtek/rtlwifi/built-in.o.defconfig.new 1113597 200499 1907 1316003 1414a3 drivers/net/wireless/realtek/rtlwifi/built-in.o.defconfig.old 1746879 453503 8512 2208894 21b47e drivers/net/wireless/realtek/rtlwifi/built-in.o.new 2051965 503311 8512 2563788 271ecc drivers/net/wireless/realtek/rtlwifi/built-in.o.old Signed-off-by: Joe Perches Acked-by: Larry Finger Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/debug.c b/drivers/net/wireless/realtek/rtlwifi/debug.c index fd25aba..33905bb 100644 --- a/drivers/net/wireless/realtek/rtlwifi/debug.c +++ b/drivers/net/wireless/realtek/rtlwifi/debug.c @@ -48,3 +48,28 @@ void rtl_dbgp_flag_init(struct ieee80211_hw *hw) /*Init Debug flag enable condition */ } EXPORT_SYMBOL_GPL(rtl_dbgp_flag_init); + +#ifdef CONFIG_RTLWIFI_DEBUG +void _rtl_dbg_trace(struct rtl_priv *rtlpriv, int comp, int level, + const char *modname, const char *fmt, ...) +{ + if (unlikely((comp & rtlpriv->dbg.global_debugcomponents) && + (level <= rtlpriv->dbg.global_debuglevel))) { + struct va_format vaf; + va_list args; + + va_start(args, fmt); + + vaf.fmt = fmt; + vaf.va = &args; + + printk(KERN_DEBUG "%s:%ps:<%lx-%x> %pV", + modname, __builtin_return_address(0), + in_interrupt(), in_atomic(), + &vaf); + + va_end(args); + } +} +EXPORT_SYMBOL_GPL(_rtl_dbg_trace); +#endif diff --git a/drivers/net/wireless/realtek/rtlwifi/debug.h b/drivers/net/wireless/realtek/rtlwifi/debug.h index fc794b3..6156a79 100644 --- a/drivers/net/wireless/realtek/rtlwifi/debug.h +++ b/drivers/net/wireless/realtek/rtlwifi/debug.h @@ -174,15 +174,16 @@ do { \ } \ } while (0) + +struct rtl_priv; + +__printf(5, 6) +void _rtl_dbg_trace(struct rtl_priv *rtlpriv, int comp, int level, + const char *modname, const char *fmt, ...); + #define RT_TRACE(rtlpriv, comp, level, fmt, ...) \ -do { \ - if (unlikely(((comp) & rtlpriv->dbg.global_debugcomponents) && \ - ((level) <= rtlpriv->dbg.global_debuglevel))) { \ - printk(KERN_DEBUG KBUILD_MODNAME ":%s():<%lx-%x> " fmt, \ - __func__, in_interrupt(), in_atomic(), \ - ##__VA_ARGS__); \ - } \ -} while (0) + _rtl_dbg_trace(rtlpriv, comp, level, \ + KBUILD_MODNAME, fmt, ##__VA_ARGS__) #define RTPRINT(rtlpriv, dbgtype, dbgflag, fmt, ...) \ do { \ -- cgit v0.10.2 From 4713bd1c74071836bfffbb07b24ff7fc40132f13 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sun, 26 Jun 2016 12:34:30 -0700 Subject: rtlwifi: Add missing newlines to RT_TRACE calls RT_TRACE does not add a newline to the end of a message and always emits at KERN_DEBUG so these are susceptible to message interleaving from other processes without the newline. Signed-off-by: Joe Perches Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/core.c b/drivers/net/wireless/realtek/rtlwifi/core.c index 3a0faa8..41f77f8 100644 --- a/drivers/net/wireless/realtek/rtlwifi/core.c +++ b/drivers/net/wireless/realtek/rtlwifi/core.c @@ -526,7 +526,7 @@ static void _rtl_add_wowlan_patterns(struct ieee80211_hw *hw, /* 3. calculate crc */ rtl_pattern.crc = _calculate_wol_pattern_crc(content, len); RT_TRACE(rtlpriv, COMP_POWER, DBG_TRACE, - "CRC_Remainder = 0x%x", rtl_pattern.crc); + "CRC_Remainder = 0x%x\n", rtl_pattern.crc); /* 4. write crc & mask_for_hw to hw */ rtlpriv->cfg->ops->add_wowlan_pattern(hw, &rtl_pattern, i); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c index e26a233..cfdf6d8 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c @@ -1846,12 +1846,12 @@ static void _rtl88ee_read_adapter_info(struct ieee80211_hw *hw) case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!"); + "RTL819X Not boot from eeprom, check it !!\n"); return; default: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "boot from neither eeprom nor efuse, check it !!"); + "boot from neither eeprom nor efuse, check it !!\n"); return; } memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c index 416a9ba..1d034a1 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c @@ -373,7 +373,7 @@ static bool _rtl88e_phy_bb8188e_config_parafile(struct ieee80211_hw *hw) rtstatus = phy_config_bb_with_headerfile(hw, BASEBAND_CONFIG_PHY_REG); if (!rtstatus) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!\n"); return false; } @@ -383,7 +383,7 @@ static bool _rtl88e_phy_bb8188e_config_parafile(struct ieee80211_hw *hw) phy_config_bb_with_pghdr(hw, BASEBAND_CONFIG_PHY_REG); } if (!rtstatus) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!\n"); return false; } rtstatus = diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/rf.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/rf.c index 40893ce..26ac4c2 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/rf.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/rf.c @@ -498,7 +498,7 @@ static bool _rtl88e_phy_rf6052_config_parafile(struct ieee80211_hw *hw) if (rtstatus != true) { RT_TRACE(rtlpriv, COMP_INIT, DBG_TRACE, - "Radio[%d] Fail!!", rfpath); + "Radio[%d] Fail!!\n", rfpath); return false; } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c index 1016ad4..3e3b886 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/trx.c @@ -540,7 +540,7 @@ void rtl88ee_tx_fill_desc(struct ieee80211_hw *hw, PCI_DMA_TODEVICE); if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, sizeof(struct tx_desc_88e)); @@ -703,7 +703,7 @@ void rtl88ee_tx_fill_cmddesc(struct ieee80211_hw *hw, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, TX_DESC_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c index 24162e0..56f2cec 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c @@ -213,7 +213,7 @@ bool _rtl92c_phy_bb8192c_config_parafile(struct ieee80211_hw *hw) rtstatus = rtlpriv->cfg->ops->config_bb_with_headerfile(hw, BASEBAND_CONFIG_PHY_REG); if (!rtstatus) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!\n"); return false; } if (rtlphy->rf_type == RF_1T2R) { @@ -226,7 +226,7 @@ bool _rtl92c_phy_bb8192c_config_parafile(struct ieee80211_hw *hw) BASEBAND_CONFIG_PHY_REG); } if (!rtstatus) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!\n"); return false; } rtstatus = rtlpriv->cfg->ops->config_bb_with_headerfile(hw, diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c index 58b7ac6..809a021 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c @@ -1692,7 +1692,7 @@ static void _rtl92ce_read_adapter_info(struct ieee80211_hw *hw) case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!"); + "RTL819X Not boot from eeprom, check it !!\n"); return; default: diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c index 93d3fba..781af1b 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/trx.c @@ -449,7 +449,7 @@ void rtl92ce_tx_fill_desc(struct ieee80211_hw *hw, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } rcu_read_lock(); @@ -615,7 +615,7 @@ void rtl92ce_tx_fill_cmddesc(struct ieee80211_hw *hw, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, TX_DESC_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/rf.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/rf.c index 5624ade..ec2ea56 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/rf.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/rf.c @@ -465,7 +465,7 @@ static bool _rtl92c_phy_rf6052_config_parafile(struct ieee80211_hw *hw) } if (!rtstatus) { RT_TRACE(rtlpriv, COMP_INIT, DBG_TRACE, - "Radio[%d] Fail!!", rfpath); + "Radio[%d] Fail!!\n", rfpath); goto phy_rf_cfg_fail; } } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/rf.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/rf.c index 6a6ac54..2f479d3 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/rf.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/rf.c @@ -601,7 +601,7 @@ bool rtl92d_phy_rf6052_config(struct ieee80211_hw *hw) } if (!rtstatus) { RT_TRACE(rtlpriv, COMP_INIT, DBG_TRACE, - "Radio[%d] Fail!!", rfpath); + "Radio[%d] Fail!!\n", rfpath); goto phy_rf_cfg_fail; } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c index 274b0e4..e998e98 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/trx.c @@ -586,7 +586,7 @@ void rtl92de_tx_fill_desc(struct ieee80211_hw *hw, PCI_DMA_TODEVICE); if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, sizeof(struct tx_desc_92d)); @@ -744,7 +744,7 @@ void rtl92de_tx_fill_cmddesc(struct ieee80211_hw *hw, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, TX_DESC_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c index 46efba0..a56a3da 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c @@ -983,7 +983,7 @@ static bool _rtl92ee_dm_ra_state_check(struct ieee80211_hw *hw, break; default: RT_TRACE(rtlpriv, COMP_RATR, DBG_DMESG, - "wrong rssi level setting %d !", *ratr_state); + "wrong rssi level setting %d !\n", *ratr_state); break; } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c index 28c260d..9d1ecba 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c @@ -2109,12 +2109,12 @@ static void _rtl92ee_read_adapter_info(struct ieee80211_hw *hw) case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!"); + "RTL819X Not boot from eeprom, check it !!\n"); return; default: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "boot from neither eeprom nor efuse, check it !!"); + "boot from neither eeprom nor efuse, check it !!\n"); return; } memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c index b7184f5..d678bf7 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c @@ -650,7 +650,7 @@ static bool _rtl92ee_phy_bb8192ee_config_parafile(struct ieee80211_hw *hw) rtstatus = phy_config_bb_with_hdr_file(hw, BASEBAND_CONFIG_PHY_REG); if (!rtstatus) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!\n"); return false; } @@ -662,7 +662,7 @@ static bool _rtl92ee_phy_bb8192ee_config_parafile(struct ieee80211_hw *hw) } _rtl92ee_phy_txpower_by_rate_configuration(hw); if (!rtstatus) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!\n"); return false; } rtstatus = phy_config_bb_with_hdr_file(hw, BASEBAND_CONFIG_AGC_TAB); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/rf.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/rf.c index c9bc33c..73716c0 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/rf.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/rf.c @@ -142,7 +142,7 @@ static bool _rtl92ee_phy_rf6052_config_parafile(struct ieee80211_hw *hw) if (!rtstatus) { RT_TRACE(rtlpriv, COMP_INIT, DBG_TRACE, - "Radio[%d] Fail!!", rfpath); + "Radio[%d] Fail!!\n", rfpath); return false; } } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c index 582b1fa..2d48ccd 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/trx.c @@ -703,7 +703,7 @@ void rtl92ee_tx_fill_desc(struct ieee80211_hw *hw, PCI_DMA_TODEVICE); if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } @@ -867,7 +867,7 @@ void rtl92ee_tx_fill_cmddesc(struct ieee80211_hw *hw, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, txdesc_len); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c index 442f2b6..ddfa0ae 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/hw.c @@ -2003,7 +2003,7 @@ static void _rtl92se_read_adapter_info(struct ieee80211_hw *hw) rtlefuse->b1ss_support = rtlefuse->b1x1_recvcombine; rtlefuse->eeprom_oemid = *&hwinfo[EEPROM_CUSTOMID]; - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "EEPROM Customer ID: 0x%2x", + RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "EEPROM Customer ID: 0x%2x\n", rtlefuse->eeprom_oemid); /* set channel paln to world wide 13 */ diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/trx.c index 125b29b..d53bbf6 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192se/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192se/trx.c @@ -360,7 +360,7 @@ void rtl92se_tx_fill_desc(struct ieee80211_hw *hw, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } if (mac->opmode == NL80211_IFTYPE_STATION) { @@ -529,7 +529,7 @@ void rtl92se_tx_fill_cmddesc(struct ieee80211_hw *hw, u8 *pdesc, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } /* Clear all status */ diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c index 3900e10..42a6fba 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/dm.c @@ -876,8 +876,8 @@ void rtl8723e_dm_bt_coexist(struct ieee80211_hw *hw) tmp_byte = rtl_read_byte(rtlpriv, 0x40); RT_TRACE(rtlpriv, COMP_BT_COEXIST, DBG_LOUD, - "[DM][BT], 0x40 is 0x%x", tmp_byte); + "[DM][BT], 0x40 is 0x%x\n", tmp_byte); RT_TRACE(rtlpriv, COMP_BT_COEXIST, DBG_DMESG, - "[DM][BT], bt_dm_coexist start"); + "[DM][BT], bt_dm_coexist start\n"); rtl8723e_dm_bt_coexist_8723(hw); } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hal_btc.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hal_btc.c index 44de695..ec9bcf3 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hal_btc.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hal_btc.c @@ -185,7 +185,7 @@ static void rtl8723e_dm_bt_set_hw_pta_mode(struct ieee80211_hw *hw, bool b_mode) struct rtl_priv *rtlpriv = rtl_priv(hw); if (BT_PTA_MODE_ON == b_mode) { - RT_TRACE(rtlpriv, COMP_BT_COEXIST, DBG_TRACE, "PTA mode on, "); + RT_TRACE(rtlpriv, COMP_BT_COEXIST, DBG_TRACE, "PTA mode on\n"); /* Enable GPIO 0/1/2/3/8 pins for bt */ rtl_write_byte(rtlpriv, 0x40, 0x20); rtlpriv->btcoexist.hw_coexist_all_off = false; @@ -1401,7 +1401,7 @@ static void rtl8723e_dm_bt_inq_page_monitor(struct ieee80211_hw *hw) (long)hal_coex_8723.bt_inq_page_start_time) / HZ) >= 10) { RT_TRACE(rtlpriv, COMP_BT_COEXIST, DBG_DMESG, - "[BTCoex], BT Inquiry/page >= 10sec!!!"); + "[BTCoex], BT Inquiry/page >= 10sec!!!\n"); hal_coex_8723.bt_inq_page_start_time = 0; rtlpriv->btcoexist.cstate &= ~BT_COEX_STATE_BT_INQ_PAGE; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c index 57a1ba8..0025e21 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c @@ -1645,7 +1645,7 @@ static void _rtl8723e_read_adapter_info(struct ieee80211_hw *hw, case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!"); + "RTL819X Not boot from eeprom, check it !!\n"); return; default: diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/phy.c index d367097..af07e00 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/phy.c @@ -213,7 +213,7 @@ static bool _rtl8723e_phy_bb8192c_config_parafile(struct ieee80211_hw *hw) rtstatus = _rtl8723e_phy_config_bb_with_headerfile(hw, BASEBAND_CONFIG_PHY_REG); if (rtstatus != true) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!\n"); return false; } @@ -227,7 +227,7 @@ static bool _rtl8723e_phy_bb8192c_config_parafile(struct ieee80211_hw *hw) BASEBAND_CONFIG_PHY_REG); } if (rtstatus != true) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!\n"); return false; } rtstatus = diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/rf.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/rf.c index 9ebc828..4227717 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/rf.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/rf.c @@ -504,7 +504,7 @@ static bool _rtl8723e_phy_rf6052_config_parafile(struct ieee80211_hw *hw) if (rtstatus != true) { RT_TRACE(rtlpriv, COMP_INIT, DBG_TRACE, - "Radio[%d] Fail!!", rfpath); + "Radio[%d] Fail!!\n", rfpath); return false; } } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.c index 7b4a9b6..e93125e 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/trx.c @@ -389,7 +389,7 @@ void rtl8723e_tx_fill_desc(struct ieee80211_hw *hw, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } if (mac->opmode == NL80211_IFTYPE_STATION) { @@ -557,7 +557,7 @@ void rtl8723e_tx_fill_cmddesc(struct ieee80211_hw *hw, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, TX_DESC_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c index 08288ac..63b0df6 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c @@ -2064,7 +2064,7 @@ static void _rtl8723be_read_adapter_info(struct ieee80211_hw *hw, case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!"); + "RTL819X Not boot from eeprom, check it !!\n"); return; default: diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c index aa2e670..b2b313a 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c @@ -467,7 +467,7 @@ static bool _rtl8723be_phy_bb8723b_config_parafile(struct ieee80211_hw *hw) rtstatus = _rtl8723be_phy_config_bb_with_headerfile(hw, BASEBAND_CONFIG_PHY_REG); if (!rtstatus) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!\n"); return false; } _rtl8723be_phy_init_tx_power_by_rate(hw); @@ -478,7 +478,7 @@ static bool _rtl8723be_phy_bb8723b_config_parafile(struct ieee80211_hw *hw) } phy_txpower_by_rate_config(hw); if (!rtstatus) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!\n"); return false; } rtstatus = _rtl8723be_phy_config_bb_with_headerfile(hw, diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/rf.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/rf.c index 97f5a03..78f4f18 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/rf.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/rf.c @@ -502,7 +502,7 @@ static bool _rtl8723be_phy_rf6052_config_parafile(struct ieee80211_hw *hw) if (!rtstatus) { RT_TRACE(rtlpriv, COMP_INIT, DBG_TRACE, - "Radio[%d] Fail!!", rfpath); + "Radio[%d] Fail!!\n", rfpath); return false; } } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c index e881ef8..2175aec 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/trx.c @@ -464,7 +464,7 @@ void rtl8723be_tx_fill_desc(struct ieee80211_hw *hw, mapping = pci_map_single(rtlpci->pdev, skb->data, skb->len, PCI_DMA_TODEVICE); if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { - RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, "DMA mapping error"); + RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, sizeof(struct tx_desc_8723be)); @@ -616,7 +616,7 @@ void rtl8723be_tx_fill_cmddesc(struct ieee80211_hw *hw, u8 *pdesc, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, TX_DESC_SIZE); diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c index 03e08cb..bdfd444 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/dm.c @@ -843,7 +843,7 @@ static void rtl8821ae_dm_dig(struct ieee80211_hw *hw) dm_digtable->rssi_val_min + offset; RT_TRACE(rtlpriv, COMP_DIG, DBG_LOUD, - "dm_digtable->rssi_val_min=0x%x,dm_digtable->rx_gain_max = 0x%x", + "dm_digtable->rssi_val_min=0x%x,dm_digtable->rx_gain_max = 0x%x\n", dm_digtable->rssi_val_min, dm_digtable->rx_gain_max); if (rtlpriv->dm.one_entry_only) { @@ -2682,9 +2682,9 @@ static void rtl8821ae_dm_check_edca_turbo(struct ieee80211_hw *hw) bool b_edca_turbo_on = false; RT_TRACE(rtlpriv, COMP_TURBO, DBG_LOUD, - "rtl8821ae_dm_check_edca_turbo=====>"); + "rtl8821ae_dm_check_edca_turbo=====>\n"); RT_TRACE(rtlpriv, COMP_TURBO, DBG_LOUD, - "Orginial BE PARAM: 0x%x\n", + "Original BE PARAM: 0x%x\n", rtl_read_dword(rtlpriv, DM_REG_EDCA_BE_11N)); if (rtlpriv->dm.dbginfo.num_non_be_pkt > 0x100) diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c index b9436df..ac8836b 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c @@ -3117,7 +3117,7 @@ static void _rtl8821ae_read_adapter_info(struct ieee80211_hw *hw, bool b_pseudo_ case EEPROM_93C46: RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!"); + "RTL819X Not boot from eeprom, check it !!\n"); return; default: @@ -3140,7 +3140,7 @@ static void _rtl8821ae_read_adapter_info(struct ieee80211_hw *hw, bool b_pseudo_ if (rtlefuse->autoload_failflag) { RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL8812AE autoload_failflag, check it !!"); + "RTL8812AE autoload_failflag, check it !!\n"); return; } @@ -3836,7 +3836,7 @@ void rtl8821ae_update_hal_rate_tbl(struct ieee80211_hw *hw, rtl8821ae_update_hal_rate_mask(hw, sta, rssi_level); else /*RT_TRACE(rtlpriv, COMP_RATR,DBG_LOUD, - "rtl8821ae_update_hal_rate_tbl() Error! 8821ae FW RA Only");*/ + "rtl8821ae_update_hal_rate_tbl() Error! 8821ae FW RA Only\n");*/ rtl8821ae_update_hal_rate_table(hw, sta); } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c index 46ea4f8..a71bfe3 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/phy.c @@ -581,7 +581,7 @@ void rtl8821ae_phy_switch_wirelessband(struct ieee80211_hw *hw, u8 band) count = 0; reg_41a = rtl_read_word(rtlpriv, REG_TXPKT_EMPTY); RT_TRACE(rtlpriv, COMP_SCAN, DBG_LOUD, - "Reg41A value %d", reg_41a); + "Reg41A value %d\n", reg_41a); reg_41a &= 0x30; while ((reg_41a != 0x30) && (count < 50)) { udelay(50); @@ -591,7 +591,7 @@ void rtl8821ae_phy_switch_wirelessband(struct ieee80211_hw *hw, u8 band) reg_41a &= 0x30; count++; RT_TRACE(rtlpriv, COMP_SCAN, DBG_LOUD, - "Reg41A value %d", reg_41a); + "Reg41A value %d\n", reg_41a); } if (count != 0) RT_TRACE(rtlpriv, COMP_MLME, DBG_LOUD, @@ -1013,7 +1013,7 @@ static void _rtl8812ae_phy_cross_reference_ht_and_vht_txpower_limit(struct ieee8 rtlphy->txpwr_limit_5g[regulation][bw][3][channel][RF90_PATH_A]; } - RT_TRACE(rtlpriv, COMP_INIT, DBG_TRACE, "use other value %d", temp_pwrlmt); + RT_TRACE(rtlpriv, COMP_INIT, DBG_TRACE, "use other value %d\n", temp_pwrlmt); } } } @@ -1482,12 +1482,12 @@ static s8 _rtl8812ae_phy_get_chnl_idx_of_txpwr_lmt(struct ieee80211_hw *hw, channel_index = i; } } else - RT_TRACE(rtlpriv, COMP_POWER, DBG_LOUD, "Invalid Band %d in %s", + RT_TRACE(rtlpriv, COMP_POWER, DBG_LOUD, "Invalid Band %d in %s\n", band, __func__); if (channel_index == -1) RT_TRACE(rtlpriv, COMP_POWER, DBG_LOUD, - "Invalid Channel %d of Band %d in %s", channel, + "Invalid Channel %d of Band %d in %s\n", channel, band, __func__); return channel_index; @@ -1665,7 +1665,7 @@ static bool _rtl8821ae_phy_bb8821a_config_parafile(struct ieee80211_hw *hw) rtstatus = _rtl8821ae_phy_config_bb_with_headerfile(hw, BASEBAND_CONFIG_PHY_REG); if (rtstatus != true) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "Write BB Reg Fail!!\n"); return false; } _rtl8821ae_phy_init_tx_power_by_rate(hw); @@ -1674,7 +1674,7 @@ static bool _rtl8821ae_phy_bb8821a_config_parafile(struct ieee80211_hw *hw) BASEBAND_CONFIG_PHY_REG); } if (rtstatus != true) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!"); + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, "BB_PG Reg Fail!!\n"); return false; } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/rf.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/rf.c index 2922538..c6ab957 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/rf.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/rf.c @@ -454,7 +454,7 @@ static bool _rtl8821ae_phy_rf6052_config_parafile(struct ieee80211_hw *hw) if (!rtstatus) { RT_TRACE(rtlpriv, COMP_INIT, DBG_TRACE, - "Radio[%d] Fail!!", rfpath); + "Radio[%d] Fail!!\n", rfpath); return false; } } diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c index 8c2a5e2..2772718 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/trx.c @@ -716,7 +716,7 @@ void rtl8821ae_tx_fill_desc(struct ieee80211_hw *hw, PCI_DMA_TODEVICE); if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, sizeof(struct tx_desc_8821ae)); @@ -857,7 +857,7 @@ void rtl8821ae_tx_fill_cmddesc(struct ieee80211_hw *hw, if (pci_dma_mapping_error(rtlpci->pdev, mapping)) { RT_TRACE(rtlpriv, COMP_SEND, DBG_TRACE, - "DMA mapping error"); + "DMA mapping error\n"); return; } CLEAR_PCI_TX_DESC_CONTENT(pdesc, TX_DESC_SIZE); -- cgit v0.10.2 From 322397b268f76082951248ccc197cefcdfcd01bb Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Mon, 27 Jun 2016 14:16:27 +0530 Subject: mwifiex: code rearrangement in suspend handler We will derive sta_priv at the beginning of suspend handler. This will be useful for next patch in this series. Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c index df5ebdf..e651455 100644 --- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c +++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c @@ -3313,6 +3313,8 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy, struct mwifiex_ds_hs_cfg hs_cfg; int i, ret = 0, retry_num = 10; struct mwifiex_private *priv; + struct mwifiex_private *sta_priv = + mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_STA); for (i = 0; i < adapter->priv_num; i++) { priv = adapter->priv[i]; @@ -3345,15 +3347,13 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy, return 0; } - priv = mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_STA); - - if (!priv->media_connected && !wowlan->nd_config) { + if (!sta_priv->media_connected && !wowlan->nd_config) { mwifiex_dbg(adapter, ERROR, "Can not configure WOWLAN in disconnected state\n"); return 0; } - ret = mwifiex_set_mef_filter(priv, wowlan); + ret = mwifiex_set_mef_filter(sta_priv, wowlan); if (ret) { mwifiex_dbg(adapter, ERROR, "Failed to set MEF filter\n"); return ret; @@ -3365,19 +3365,19 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy, if (wowlan->nd_config) { mwifiex_dbg(adapter, INFO, "Wake on net detect\n"); hs_cfg.conditions |= HS_CFG_COND_MAC_EVENT; - mwifiex_cfg80211_sched_scan_start(wiphy, priv->netdev, + mwifiex_cfg80211_sched_scan_start(wiphy, sta_priv->netdev, wowlan->nd_config); } if (wowlan->disconnect) { hs_cfg.conditions |= HS_CFG_COND_MAC_EVENT; - mwifiex_dbg(priv->adapter, INFO, "Wake on device disconnect\n"); + mwifiex_dbg(sta_priv->adapter, INFO, "Wake on device disconnect\n"); } hs_cfg.is_invoke_hostcmd = false; hs_cfg.gpio = adapter->hs_cfg.gpio; hs_cfg.gap = adapter->hs_cfg.gap; - ret = mwifiex_set_hs_params(priv, HostCmd_ACT_GEN_SET, + ret = mwifiex_set_hs_params(sta_priv, HostCmd_ACT_GEN_SET, MWIFIEX_SYNC_CMD, &hs_cfg); if (ret) { mwifiex_dbg(adapter, ERROR, -- cgit v0.10.2 From 09e672a16b92902f411eecc180397191ee5a7e05 Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Mon, 27 Jun 2016 14:16:28 +0530 Subject: mwifiex: clear scan_aborting flag The flag should be cleaned along with other scan operation variables. This was missing at some places. Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/scan.c b/drivers/net/wireless/marvell/mwifiex/scan.c index e331122..96d0d86 100644 --- a/drivers/net/wireless/marvell/mwifiex/scan.c +++ b/drivers/net/wireless/marvell/mwifiex/scan.c @@ -1960,6 +1960,7 @@ static void mwifiex_check_next_scan_command(struct mwifiex_private *priv) "info: notifying scan done\n"); cfg80211_scan_done(priv->scan_request, 0); priv->scan_request = NULL; + priv->scan_aborting = false; } else { priv->scan_aborting = false; mwifiex_dbg(adapter, INFO, @@ -1981,6 +1982,7 @@ static void mwifiex_check_next_scan_command(struct mwifiex_private *priv) "info: aborting scan\n"); cfg80211_scan_done(priv->scan_request, 1); priv->scan_request = NULL; + priv->scan_aborting = false; } else { priv->scan_aborting = false; mwifiex_dbg(adapter, INFO, @@ -2022,6 +2024,7 @@ void mwifiex_cancel_scan(struct mwifiex_adapter *adapter) "info: aborting scan\n"); cfg80211_scan_done(priv->scan_request, 1); priv->scan_request = NULL; + priv->scan_aborting = false; } } } -- cgit v0.10.2 From 16d25da94f3d6542a0bbd25a85d247c970026f8a Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Mon, 27 Jun 2016 14:16:29 +0530 Subject: mwifiex: fix NULL pointer dereference during suspend This patch fixes below NULL pointer dereference observed in suspend stress test. When scan is cancelled during system suspend, we may end up aceesing "priv->scan_request" in corner case. [ 3035.304682] BUG: KASAN: null-ptr-deref on address 0000000000000008 [ 3035.304704] Read of size 4 by task ksdioirqd/mmc2/1183 [ 3035.304744] CPU: 0 PID: 1183 Comm: ksdioirqd/mmc2 Tainted: G W 3.18.0 #1169 [ 3035.304772] Call trace: [ 3035.304825] [] dump_backtrace+0x0/0x190 [ 3035.304864] [] show_stack+0x1c/0x28 [ 3035.304901] [] dump_stack+0xa0/0xf8 [ 3035.304940] [] kasan_report+0x120/0x4fc [ 3035.304975] [] __asan_load4+0x20/0x80 [ 3035.305546] [] mwifiex_check_next_scan_command+0x1a4/0x588 [mwifiex] [ 3035.306091] [] mwifiex_handle_event_ext_scan_report+0x304/0x370 [mwifiex] [ 3035.306735] [] mwifiex_process_sta_event+0x6c0/0xf10 [mwifiex] [ 3035.307200] [] mwifiex_process_event+0x2f4/0x358 [mwifiex] [ 3035.307612] [] mwifiex_main_process+0x3cc/0x80c [mwifiex] [ 3035.307737] [] mwifiex_sdio_interrupt+0x198/0x1c0 [mwifiex_sdio] [ 3035.307785] [] process_sdio_pending_irqs+0x15c/0x1d4 [ 3035.307826] [] sdio_irq_thread+0xd8/0x288 Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c index e651455..e6befd5 100644 --- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c +++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c @@ -3316,6 +3316,7 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy, struct mwifiex_private *sta_priv = mwifiex_get_priv(adapter, MWIFIEX_BSS_ROLE_STA); + sta_priv->scan_aborting = true; for (i = 0; i < adapter->priv_num; i++) { priv = adapter->priv[i]; mwifiex_abort_cac(priv); @@ -3344,19 +3345,21 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy, if (!wowlan) { mwifiex_dbg(adapter, ERROR, "None of the WOWLAN triggers enabled\n"); - return 0; + ret = 0; + goto done; } if (!sta_priv->media_connected && !wowlan->nd_config) { mwifiex_dbg(adapter, ERROR, "Can not configure WOWLAN in disconnected state\n"); - return 0; + ret = 0; + goto done; } ret = mwifiex_set_mef_filter(sta_priv, wowlan); if (ret) { mwifiex_dbg(adapter, ERROR, "Failed to set MEF filter\n"); - return ret; + goto done; } memset(&hs_cfg, 0, sizeof(hs_cfg)); @@ -3379,12 +3382,11 @@ static int mwifiex_cfg80211_suspend(struct wiphy *wiphy, hs_cfg.gap = adapter->hs_cfg.gap; ret = mwifiex_set_hs_params(sta_priv, HostCmd_ACT_GEN_SET, MWIFIEX_SYNC_CMD, &hs_cfg); - if (ret) { - mwifiex_dbg(adapter, ERROR, - "Failed to set HS params\n"); - return ret; - } + if (ret) + mwifiex_dbg(adapter, ERROR, "Failed to set HS params\n"); +done: + sta_priv->scan_aborting = false; return ret; } diff --git a/drivers/net/wireless/marvell/mwifiex/scan.c b/drivers/net/wireless/marvell/mwifiex/scan.c index 96d0d86..87e7000 100644 --- a/drivers/net/wireless/marvell/mwifiex/scan.c +++ b/drivers/net/wireless/marvell/mwifiex/scan.c @@ -1896,7 +1896,8 @@ mwifiex_active_scan_req_for_passive_chan(struct mwifiex_private *priv) u8 id = 0; struct mwifiex_user_scan_cfg *user_scan_cfg; - if (adapter->active_scan_triggered || !priv->scan_request) { + if (adapter->active_scan_triggered || !priv->scan_request || + priv->scan_aborting) { adapter->active_scan_triggered = false; return 0; } diff --git a/drivers/net/wireless/marvell/mwifiex/sta_event.c b/drivers/net/wireless/marvell/mwifiex/sta_event.c index 0cefd40..7c01778 100644 --- a/drivers/net/wireless/marvell/mwifiex/sta_event.c +++ b/drivers/net/wireless/marvell/mwifiex/sta_event.c @@ -708,7 +708,7 @@ int mwifiex_process_sta_event(struct mwifiex_private *priv) case EVENT_EXT_SCAN_REPORT: mwifiex_dbg(adapter, EVENT, "event: EXT_SCAN Report\n"); - if (adapter->ext_scan) + if (adapter->ext_scan && !priv->scan_aborting) ret = mwifiex_handle_event_ext_scan_report(priv, adapter->event_skb->data); -- cgit v0.10.2 From 585dfe813fa56683a90350eeb173d98da8b6be5c Mon Sep 17 00:00:00 2001 From: "Machani, Yaniv" Date: Mon, 27 Jun 2016 16:37:14 +0300 Subject: wlcore: time sync : add support for 64 bit clock Changed the configuration to support 64bit instead of 32bit this in order to offload the driver from handling a wraparound. Signed-off-by: Yaniv Machani Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ti/wl18xx/event.c b/drivers/net/wireless/ti/wl18xx/event.c index ef81184..2c5df43 100644 --- a/drivers/net/wireless/ti/wl18xx/event.c +++ b/drivers/net/wireless/ti/wl18xx/event.c @@ -112,12 +112,18 @@ static int wlcore_smart_config_decode_event(struct wl1271 *wl, return 0; } -static void wlcore_event_time_sync(struct wl1271 *wl, u16 tsf_msb, u16 tsf_lsb) +static void wlcore_event_time_sync(struct wl1271 *wl, + u16 tsf_high_msb, u16 tsf_high_lsb, + u16 tsf_low_msb, u16 tsf_low_lsb) { - u32 clock; - /* convert the MSB+LSB to a u32 TSF value */ - clock = (tsf_msb << 16) | tsf_lsb; - wl1271_info("TIME_SYNC_EVENT_ID: clock %u", clock); + u32 clock_low; + u32 clock_high; + + clock_high = (tsf_high_msb << 16) | tsf_high_lsb; + clock_low = (tsf_low_msb << 16) | tsf_low_lsb; + + wl1271_info("TIME_SYNC_EVENT_ID: clock_high %u, clock low %u", + clock_high, clock_low); } int wl18xx_process_mailbox_events(struct wl1271 *wl) @@ -138,8 +144,10 @@ int wl18xx_process_mailbox_events(struct wl1271 *wl) if (vector & TIME_SYNC_EVENT_ID) wlcore_event_time_sync(wl, - mbox->time_sync_tsf_msb, - mbox->time_sync_tsf_lsb); + mbox->time_sync_tsf_high_msb, + mbox->time_sync_tsf_high_lsb, + mbox->time_sync_tsf_low_msb, + mbox->time_sync_tsf_low_lsb); if (vector & RADAR_DETECTED_EVENT_ID) { wl1271_info("radar event: channel %d type %s", @@ -187,11 +195,11 @@ int wl18xx_process_mailbox_events(struct wl1271 *wl) */ if (vector & MAX_TX_FAILURE_EVENT_ID) wlcore_event_max_tx_failure(wl, - le32_to_cpu(mbox->tx_retry_exceeded_bitmap)); + le16_to_cpu(mbox->tx_retry_exceeded_bitmap)); if (vector & INACTIVE_STA_EVENT_ID) wlcore_event_inactive_sta(wl, - le32_to_cpu(mbox->inactive_sta_bitmap)); + le16_to_cpu(mbox->inactive_sta_bitmap)); if (vector & REMAIN_ON_CHANNEL_COMPLETE_EVENT_ID) wlcore_event_roc_complete(wl); diff --git a/drivers/net/wireless/ti/wl18xx/event.h b/drivers/net/wireless/ti/wl18xx/event.h index 070de12..ce8ea9c0 100644 --- a/drivers/net/wireless/ti/wl18xx/event.h +++ b/drivers/net/wireless/ti/wl18xx/event.h @@ -74,10 +74,16 @@ struct wl18xx_event_mailbox { __le16 bss_loss_bitmap; /* bitmap of stations (by HLID) which exceeded max tx retries */ - __le32 tx_retry_exceeded_bitmap; + __le16 tx_retry_exceeded_bitmap; + + /* time sync high msb*/ + __le16 time_sync_tsf_high_msb; /* bitmap of inactive stations (by HLID) */ - __le32 inactive_sta_bitmap; + __le16 inactive_sta_bitmap; + + /* time sync high lsb*/ + __le16 time_sync_tsf_high_lsb; /* rx BA win size indicated by RX_BA_WIN_SIZE_CHANGE_EVENT_ID */ u8 rx_ba_role_id; @@ -98,14 +104,15 @@ struct wl18xx_event_mailbox { u8 sc_sync_channel; u8 sc_sync_band; - /* time sync msb*/ - u16 time_sync_tsf_msb; + /* time sync low msb*/ + __le16 time_sync_tsf_low_msb; + /* radar detect */ u8 radar_channel; u8 radar_type; - /* time sync lsb*/ - u16 time_sync_tsf_lsb; + /* time sync low lsb*/ + __le16 time_sync_tsf_low_lsb; } __packed; -- cgit v0.10.2 From a4770e1117f193c3e27f5f046cd4f8e2470f3b70 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Sun, 26 Jun 2016 14:55:23 -0700 Subject: Bluetooth: Switch SMP to crypto_cipher_encrypt_one() SMP does ECB crypto on stack buffers. This is complicated and fragile, and it will not work if the stack is virtually allocated. Switch to the crypto_cipher interface, which is simpler and safer. Signed-off-by: Andy Lutomirski Acked-by: Herbert Xu Acked-by: Johan Hedberg Tested-by: Johan Hedberg Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/smp.c b/net/bluetooth/smp.c index 50976a6..4c1a16a 100644 --- a/net/bluetooth/smp.c +++ b/net/bluetooth/smp.c @@ -22,9 +22,9 @@ #include #include +#include #include #include -#include #include #include @@ -88,7 +88,7 @@ struct smp_dev { u8 min_key_size; u8 max_key_size; - struct crypto_skcipher *tfm_aes; + struct crypto_cipher *tfm_aes; struct crypto_shash *tfm_cmac; }; @@ -127,7 +127,7 @@ struct smp_chan { u8 dhkey[32]; u8 mackey[16]; - struct crypto_skcipher *tfm_aes; + struct crypto_cipher *tfm_aes; struct crypto_shash *tfm_cmac; }; @@ -361,10 +361,8 @@ static int smp_h6(struct crypto_shash *tfm_cmac, const u8 w[16], * s1 and ah. */ -static int smp_e(struct crypto_skcipher *tfm, const u8 *k, u8 *r) +static int smp_e(struct crypto_cipher *tfm, const u8 *k, u8 *r) { - SKCIPHER_REQUEST_ON_STACK(req, tfm); - struct scatterlist sg; uint8_t tmp[16], data[16]; int err; @@ -378,7 +376,7 @@ static int smp_e(struct crypto_skcipher *tfm, const u8 *k, u8 *r) /* The most significant octet of key corresponds to k[0] */ swap_buf(k, tmp, 16); - err = crypto_skcipher_setkey(tfm, tmp, 16); + err = crypto_cipher_setkey(tfm, tmp, 16); if (err) { BT_ERR("cipher setkey failed: %d", err); return err; @@ -387,16 +385,7 @@ static int smp_e(struct crypto_skcipher *tfm, const u8 *k, u8 *r) /* Most significant octet of plaintextData corresponds to data[0] */ swap_buf(r, data, 16); - sg_init_one(&sg, data, 16); - - skcipher_request_set_tfm(req, tfm); - skcipher_request_set_callback(req, 0, NULL, NULL); - skcipher_request_set_crypt(req, &sg, &sg, 16, NULL); - - err = crypto_skcipher_encrypt(req); - skcipher_request_zero(req); - if (err) - BT_ERR("Encrypt data error %d", err); + crypto_cipher_encrypt_one(tfm, data, data); /* Most significant octet of encryptedData corresponds to data[0] */ swap_buf(data, r, 16); @@ -406,7 +395,7 @@ static int smp_e(struct crypto_skcipher *tfm, const u8 *k, u8 *r) return err; } -static int smp_c1(struct crypto_skcipher *tfm_aes, const u8 k[16], +static int smp_c1(struct crypto_cipher *tfm_aes, const u8 k[16], const u8 r[16], const u8 preq[7], const u8 pres[7], u8 _iat, const bdaddr_t *ia, u8 _rat, const bdaddr_t *ra, u8 res[16]) { @@ -455,7 +444,7 @@ static int smp_c1(struct crypto_skcipher *tfm_aes, const u8 k[16], return err; } -static int smp_s1(struct crypto_skcipher *tfm_aes, const u8 k[16], +static int smp_s1(struct crypto_cipher *tfm_aes, const u8 k[16], const u8 r1[16], const u8 r2[16], u8 _r[16]) { int err; @@ -471,7 +460,7 @@ static int smp_s1(struct crypto_skcipher *tfm_aes, const u8 k[16], return err; } -static int smp_ah(struct crypto_skcipher *tfm, const u8 irk[16], +static int smp_ah(struct crypto_cipher *tfm, const u8 irk[16], const u8 r[3], u8 res[3]) { u8 _res[16]; @@ -759,7 +748,7 @@ static void smp_chan_destroy(struct l2cap_conn *conn) kzfree(smp->slave_csrk); kzfree(smp->link_key); - crypto_free_skcipher(smp->tfm_aes); + crypto_free_cipher(smp->tfm_aes); crypto_free_shash(smp->tfm_cmac); /* Ensure that we don't leave any debug key around if debug key @@ -1359,9 +1348,9 @@ static struct smp_chan *smp_chan_create(struct l2cap_conn *conn) if (!smp) return NULL; - smp->tfm_aes = crypto_alloc_skcipher("ecb(aes)", 0, CRYPTO_ALG_ASYNC); + smp->tfm_aes = crypto_alloc_cipher("aes", 0, CRYPTO_ALG_ASYNC); if (IS_ERR(smp->tfm_aes)) { - BT_ERR("Unable to create ECB crypto context"); + BT_ERR("Unable to create AES crypto context"); kzfree(smp); return NULL; } @@ -1369,7 +1358,7 @@ static struct smp_chan *smp_chan_create(struct l2cap_conn *conn) smp->tfm_cmac = crypto_alloc_shash("cmac(aes)", 0, 0); if (IS_ERR(smp->tfm_cmac)) { BT_ERR("Unable to create CMAC crypto context"); - crypto_free_skcipher(smp->tfm_aes); + crypto_free_cipher(smp->tfm_aes); kzfree(smp); return NULL; } @@ -3120,7 +3109,7 @@ static struct l2cap_chan *smp_add_cid(struct hci_dev *hdev, u16 cid) { struct l2cap_chan *chan; struct smp_dev *smp; - struct crypto_skcipher *tfm_aes; + struct crypto_cipher *tfm_aes; struct crypto_shash *tfm_cmac; if (cid == L2CAP_CID_SMP_BREDR) { @@ -3132,9 +3121,9 @@ static struct l2cap_chan *smp_add_cid(struct hci_dev *hdev, u16 cid) if (!smp) return ERR_PTR(-ENOMEM); - tfm_aes = crypto_alloc_skcipher("ecb(aes)", 0, CRYPTO_ALG_ASYNC); + tfm_aes = crypto_alloc_cipher("aes", 0, CRYPTO_ALG_ASYNC); if (IS_ERR(tfm_aes)) { - BT_ERR("Unable to create ECB crypto context"); + BT_ERR("Unable to create AES crypto context"); kzfree(smp); return ERR_CAST(tfm_aes); } @@ -3142,7 +3131,7 @@ static struct l2cap_chan *smp_add_cid(struct hci_dev *hdev, u16 cid) tfm_cmac = crypto_alloc_shash("cmac(aes)", 0, 0); if (IS_ERR(tfm_cmac)) { BT_ERR("Unable to create CMAC crypto context"); - crypto_free_skcipher(tfm_aes); + crypto_free_cipher(tfm_aes); kzfree(smp); return ERR_CAST(tfm_cmac); } @@ -3156,7 +3145,7 @@ create_chan: chan = l2cap_chan_create(); if (!chan) { if (smp) { - crypto_free_skcipher(smp->tfm_aes); + crypto_free_cipher(smp->tfm_aes); crypto_free_shash(smp->tfm_cmac); kzfree(smp); } @@ -3203,7 +3192,7 @@ static void smp_del_chan(struct l2cap_chan *chan) smp = chan->data; if (smp) { chan->data = NULL; - crypto_free_skcipher(smp->tfm_aes); + crypto_free_cipher(smp->tfm_aes); crypto_free_shash(smp->tfm_cmac); kzfree(smp); } @@ -3440,7 +3429,7 @@ void smp_unregister(struct hci_dev *hdev) #if IS_ENABLED(CONFIG_BT_SELFTEST_SMP) -static int __init test_ah(struct crypto_skcipher *tfm_aes) +static int __init test_ah(struct crypto_cipher *tfm_aes) { const u8 irk[16] = { 0x9b, 0x7d, 0x39, 0x0a, 0xa6, 0x10, 0x10, 0x34, @@ -3460,7 +3449,7 @@ static int __init test_ah(struct crypto_skcipher *tfm_aes) return 0; } -static int __init test_c1(struct crypto_skcipher *tfm_aes) +static int __init test_c1(struct crypto_cipher *tfm_aes) { const u8 k[16] = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, @@ -3490,7 +3479,7 @@ static int __init test_c1(struct crypto_skcipher *tfm_aes) return 0; } -static int __init test_s1(struct crypto_skcipher *tfm_aes) +static int __init test_s1(struct crypto_cipher *tfm_aes) { const u8 k[16] = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, @@ -3686,7 +3675,7 @@ static const struct file_operations test_smp_fops = { .llseek = default_llseek, }; -static int __init run_selftests(struct crypto_skcipher *tfm_aes, +static int __init run_selftests(struct crypto_cipher *tfm_aes, struct crypto_shash *tfm_cmac) { ktime_t calltime, delta, rettime; @@ -3764,27 +3753,27 @@ done: int __init bt_selftest_smp(void) { - struct crypto_skcipher *tfm_aes; + struct crypto_cipher *tfm_aes; struct crypto_shash *tfm_cmac; int err; - tfm_aes = crypto_alloc_skcipher("ecb(aes)", 0, CRYPTO_ALG_ASYNC); + tfm_aes = crypto_alloc_cipher("aes", 0, CRYPTO_ALG_ASYNC); if (IS_ERR(tfm_aes)) { - BT_ERR("Unable to create ECB crypto context"); + BT_ERR("Unable to create AES crypto context"); return PTR_ERR(tfm_aes); } tfm_cmac = crypto_alloc_shash("cmac(aes)", 0, CRYPTO_ALG_ASYNC); if (IS_ERR(tfm_cmac)) { BT_ERR("Unable to create CMAC crypto context"); - crypto_free_skcipher(tfm_aes); + crypto_free_cipher(tfm_aes); return PTR_ERR(tfm_cmac); } err = run_selftests(tfm_aes, tfm_cmac); crypto_free_shash(tfm_cmac); - crypto_free_skcipher(tfm_aes); + crypto_free_cipher(tfm_aes); return err; } -- cgit v0.10.2 From 929946a471c1d5c6c595b4094f4c56bdfceee9c7 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 21 Jun 2016 10:27:18 +0200 Subject: 6lowpan: ndisc: fix double read unlock This patch removes a double unlock case to accessing neighbour private data. Reported-by: Dan Carpenter Signed-off-by: Alexander Aring Reviewed-by: Stefan Schmidt Signed-off-by: Marcel Holtmann diff --git a/net/6lowpan/ndisc.c b/net/6lowpan/ndisc.c index ae1d419..030504e 100644 --- a/net/6lowpan/ndisc.c +++ b/net/6lowpan/ndisc.c @@ -135,8 +135,9 @@ static int lowpan_ndisc_opt_addr_space(const struct net_device *dev, read_unlock_bh(&neigh->lock); addr_space += __ndisc_opt_addr_space(IEEE802154_SHORT_ADDR_LEN, 0); *ha = ha_buf; + } else { + read_unlock_bh(&neigh->lock); } - read_unlock_bh(&neigh->lock); break; case NDISC_NEIGHBOUR_ADVERTISEMENT: case NDISC_NEIGHBOUR_SOLICITATION: -- cgit v0.10.2 From 966be9e7909d616b03e644acd8a83f09bf023c5c Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Tue, 21 Jun 2016 10:27:19 +0200 Subject: 6lowpan: ndisc: add missing 802.15.4 only check This patch adds a missing check to handle short address parsing for 802.15.4 6LoWPAN only. Signed-off-by: Alexander Aring Reviewed-by: Stefan Schmidt Signed-off-by: Marcel Holtmann diff --git a/net/6lowpan/ndisc.c b/net/6lowpan/ndisc.c index 030504e..79c5fa9 100644 --- a/net/6lowpan/ndisc.c +++ b/net/6lowpan/ndisc.c @@ -47,6 +47,9 @@ static int lowpan_ndisc_parse_options(const struct net_device *dev, struct nd_opt_hdr *nd_opt, struct ndisc_options *ndopts) { + if (!lowpan_is_ll(dev, LOWPAN_LLTYPE_IEEE802154)) + return 0; + switch (nd_opt->nd_opt_type) { case ND_OPT_SOURCE_LL_ADDR: case ND_OPT_TARGET_LL_ADDR: -- cgit v0.10.2 From 439e65d3998c0b9501638cc0c2190e936c0caaf2 Mon Sep 17 00:00:00 2001 From: Tedd Ho-Jeong An Date: Mon, 20 Jun 2016 13:43:40 -0700 Subject: Bluetooth: Add support for Intel Bluetooth device 3168 [8087:0aa7] This patch adds support for Intel Bluetooth device 3168 also known as Sandy Peak (SdP). T: Bus=01 Lev=01 Prnt=01 Port=01 Cnt=01 Dev#= 4 Spd=12 MxCh= 0 D: Ver= 2.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=8087 ProdID=0aa7 Rev= 0.01 C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=81(I) Atr=03(Int.) MxPS= 64 Ivl=1ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=82(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 0 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 0 Ivl=1ms I: If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 9 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 9 Ivl=1ms I: If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 17 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 17 Ivl=1ms I: If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 25 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 25 Ivl=1ms I: If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 33 Ivl=1ms I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms Signed-off-by: Tedd Ho-Jeong An Signed-off-by: Marcel Holtmann diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index a3be65e..a26278b 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -314,6 +314,7 @@ static const struct usb_device_id blacklist_table[] = { { USB_DEVICE(0x8087, 0x07dc), .driver_info = BTUSB_INTEL }, { USB_DEVICE(0x8087, 0x0a2a), .driver_info = BTUSB_INTEL }, { USB_DEVICE(0x8087, 0x0a2b), .driver_info = BTUSB_INTEL_NEW }, + { USB_DEVICE(0x8087, 0x0aa7), .driver_info = BTUSB_INTEL }, /* Other Intel Bluetooth devices */ { USB_VENDOR_AND_INTERFACE_INFO(0x8087, 0xe0, 0x01, 0x01), -- cgit v0.10.2 From aece0c3fe1f06962a591268b8c236df0ae5d9e4e Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Sat, 18 Jun 2016 10:45:33 +0200 Subject: nl802154: move PAD to right position The PAD define should be above the experimental support. We don't care about if we break userspace in experimental stuff but PAD is part of the existing UAPI. Cc: Nicolas Dichtel Acked-by: Nicolas Dichtel Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/include/net/nl802154.h b/include/net/nl802154.h index fcab4de..7aad2fd 100644 --- a/include/net/nl802154.h +++ b/include/net/nl802154.h @@ -124,6 +124,8 @@ enum nl802154_attrs { NL802154_ATTR_ACKREQ_DEFAULT, + NL802154_ATTR_PAD, + /* add attributes here, update the policy in nl802154.c */ #ifdef CONFIG_IEEE802154_NL802154_EXPERIMENTAL @@ -138,8 +140,6 @@ enum nl802154_attrs { NL802154_ATTR_SEC_KEY, #endif /* CONFIG_IEEE802154_NL802154_EXPERIMENTAL */ - NL802154_ATTR_PAD, - __NL802154_ATTR_AFTER_LAST, NL802154_ATTR_MAX = __NL802154_ATTR_AFTER_LAST - 1 }; -- cgit v0.10.2 From 66e5c2672cd11b9310008099faf6a4ffb9dfb6d0 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Sat, 18 Jun 2016 10:45:34 +0200 Subject: ieee802154: add netns support This patch adds netns support for 802.15.4 subsystem. Most parts are copy&pasted from wireless subsystem, it has the identically userspace API. Cc: Nicolas Dichtel Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/include/net/cfg802154.h b/include/net/cfg802154.h index 171cd76..795ca40 100644 --- a/include/net/cfg802154.h +++ b/include/net/cfg802154.h @@ -219,9 +219,22 @@ struct wpan_phy { struct device dev; + /* the network namespace this phy lives in currently */ + possible_net_t _net; + char priv[0] __aligned(NETDEV_ALIGN); }; +static inline struct net *wpan_phy_net(struct wpan_phy *wpan_phy) +{ + return read_pnet(&wpan_phy->_net); +} + +static inline void wpan_phy_net_set(struct wpan_phy *wpan_phy, struct net *net) +{ + write_pnet(&wpan_phy->_net, net); +} + struct ieee802154_addr { u8 mode; __le16 pan_id; diff --git a/include/net/nl802154.h b/include/net/nl802154.h index 7aad2fd..ddcee12 100644 --- a/include/net/nl802154.h +++ b/include/net/nl802154.h @@ -54,6 +54,8 @@ enum nl802154_commands { NL802154_CMD_SET_ACKREQ_DEFAULT, + NL802154_CMD_SET_WPAN_PHY_NETNS, + /* add new commands above here */ #ifdef CONFIG_IEEE802154_NL802154_EXPERIMENTAL @@ -126,6 +128,9 @@ enum nl802154_attrs { NL802154_ATTR_PAD, + NL802154_ATTR_PID, + NL802154_ATTR_NETNS_FD, + /* add attributes here, update the policy in nl802154.c */ #ifdef CONFIG_IEEE802154_NL802154_EXPERIMENTAL diff --git a/net/ieee802154/core.c b/net/ieee802154/core.c index c35fdfa..cb7176c 100644 --- a/net/ieee802154/core.c +++ b/net/ieee802154/core.c @@ -140,6 +140,8 @@ wpan_phy_new(const struct cfg802154_ops *ops, size_t priv_size) rdev->wpan_phy.dev.class = &wpan_phy_class; rdev->wpan_phy.dev.platform_data = rdev; + wpan_phy_net_set(&rdev->wpan_phy, &init_net); + init_waitqueue_head(&rdev->dev_wait); return &rdev->wpan_phy; @@ -207,6 +209,49 @@ void wpan_phy_free(struct wpan_phy *phy) } EXPORT_SYMBOL(wpan_phy_free); +int cfg802154_switch_netns(struct cfg802154_registered_device *rdev, + struct net *net) +{ + struct wpan_dev *wpan_dev; + int err = 0; + + list_for_each_entry(wpan_dev, &rdev->wpan_dev_list, list) { + if (!wpan_dev->netdev) + continue; + wpan_dev->netdev->features &= ~NETIF_F_NETNS_LOCAL; + err = dev_change_net_namespace(wpan_dev->netdev, net, "wpan%d"); + if (err) + break; + wpan_dev->netdev->features |= NETIF_F_NETNS_LOCAL; + } + + if (err) { + /* failed -- clean up to old netns */ + net = wpan_phy_net(&rdev->wpan_phy); + + list_for_each_entry_continue_reverse(wpan_dev, + &rdev->wpan_dev_list, + list) { + if (!wpan_dev->netdev) + continue; + wpan_dev->netdev->features &= ~NETIF_F_NETNS_LOCAL; + err = dev_change_net_namespace(wpan_dev->netdev, net, + "wpan%d"); + WARN_ON(err); + wpan_dev->netdev->features |= NETIF_F_NETNS_LOCAL; + } + + return err; + } + + wpan_phy_net_set(&rdev->wpan_phy, net); + + err = device_rename(&rdev->wpan_phy.dev, dev_name(&rdev->wpan_phy.dev)); + WARN_ON(err); + + return 0; +} + void cfg802154_dev_free(struct cfg802154_registered_device *rdev) { kfree(rdev); @@ -286,14 +331,34 @@ static struct notifier_block cfg802154_netdev_notifier = { .notifier_call = cfg802154_netdev_notifier_call, }; +static void __net_exit cfg802154_pernet_exit(struct net *net) +{ + struct cfg802154_registered_device *rdev; + + rtnl_lock(); + list_for_each_entry(rdev, &cfg802154_rdev_list, list) { + if (net_eq(wpan_phy_net(&rdev->wpan_phy), net)) + WARN_ON(cfg802154_switch_netns(rdev, &init_net)); + } + rtnl_unlock(); +} + +static struct pernet_operations cfg802154_pernet_ops = { + .exit = cfg802154_pernet_exit, +}; + static int __init wpan_phy_class_init(void) { int rc; - rc = wpan_phy_sysfs_init(); + rc = register_pernet_device(&cfg802154_pernet_ops); if (rc) goto err; + rc = wpan_phy_sysfs_init(); + if (rc) + goto err_sysfs; + rc = register_netdevice_notifier(&cfg802154_netdev_notifier); if (rc) goto err_nl; @@ -315,6 +380,8 @@ err_notifier: unregister_netdevice_notifier(&cfg802154_netdev_notifier); err_nl: wpan_phy_sysfs_exit(); +err_sysfs: + unregister_pernet_device(&cfg802154_pernet_ops); err: return rc; } @@ -326,6 +393,7 @@ static void __exit wpan_phy_class_exit(void) ieee802154_nl_exit(); unregister_netdevice_notifier(&cfg802154_netdev_notifier); wpan_phy_sysfs_exit(); + unregister_pernet_device(&cfg802154_pernet_ops); } module_exit(wpan_phy_class_exit); diff --git a/net/ieee802154/core.h b/net/ieee802154/core.h index 231fade..81141f5 100644 --- a/net/ieee802154/core.h +++ b/net/ieee802154/core.h @@ -38,6 +38,8 @@ wpan_phy_to_rdev(struct wpan_phy *wpan_phy) extern struct list_head cfg802154_rdev_list; extern int cfg802154_rdev_list_generation; +int cfg802154_switch_netns(struct cfg802154_registered_device *rdev, + struct net *net); /* free object */ void cfg802154_dev_free(struct cfg802154_registered_device *rdev); struct cfg802154_registered_device * diff --git a/net/ieee802154/nl802154.c b/net/ieee802154/nl802154.c index 116187b..d90a4ed 100644 --- a/net/ieee802154/nl802154.c +++ b/net/ieee802154/nl802154.c @@ -80,7 +80,8 @@ __cfg802154_wpan_dev_from_attrs(struct net *netns, struct nlattr **attrs) list_for_each_entry(rdev, &cfg802154_rdev_list, list) { struct wpan_dev *wpan_dev; - /* TODO netns compare */ + if (wpan_phy_net(&rdev->wpan_phy) != netns) + continue; if (have_wpan_dev_id && rdev->wpan_phy_idx != wpan_phy_idx) continue; @@ -175,7 +176,8 @@ __cfg802154_rdev_from_attrs(struct net *netns, struct nlattr **attrs) if (!rdev) return ERR_PTR(-ENODEV); - /* TODO netns compare */ + if (netns != wpan_phy_net(&rdev->wpan_phy)) + return ERR_PTR(-ENODEV); return rdev; } @@ -233,6 +235,8 @@ static const struct nla_policy nl802154_policy[NL802154_ATTR_MAX+1] = { [NL802154_ATTR_ACKREQ_DEFAULT] = { .type = NLA_U8 }, + [NL802154_ATTR_PID] = { .type = NLA_U32 }, + [NL802154_ATTR_NETNS_FD] = { .type = NLA_U32 }, #ifdef CONFIG_IEEE802154_NL802154_EXPERIMENTAL [NL802154_ATTR_SEC_ENABLED] = { .type = NLA_U8, }, [NL802154_ATTR_SEC_OUT_LEVEL] = { .type = NLA_U32, }, @@ -590,7 +594,6 @@ static int nl802154_dump_wpan_phy_parse(struct sk_buff *skb, struct cfg802154_registered_device *rdev; int ifidx = nla_get_u32(tb[NL802154_ATTR_IFINDEX]); - /* TODO netns */ netdev = __dev_get_by_index(&init_net, ifidx); if (!netdev) return -ENODEV; @@ -629,7 +632,8 @@ nl802154_dump_wpan_phy(struct sk_buff *skb, struct netlink_callback *cb) } list_for_each_entry(rdev, &cfg802154_rdev_list, list) { - /* TODO net ns compare */ + if (!net_eq(wpan_phy_net(&rdev->wpan_phy), sock_net(skb->sk))) + continue; if (++idx <= state->start) continue; if (state->filter_wpan_phy != -1 && @@ -871,7 +875,8 @@ nl802154_dump_interface(struct sk_buff *skb, struct netlink_callback *cb) rtnl_lock(); list_for_each_entry(rdev, &cfg802154_rdev_list, list) { - /* TODO netns compare */ + if (!net_eq(wpan_phy_net(&rdev->wpan_phy), sock_net(skb->sk))) + continue; if (wp_idx < wp_start) { wp_idx++; continue; @@ -1271,6 +1276,37 @@ nl802154_set_ackreq_default(struct sk_buff *skb, struct genl_info *info) return rdev_set_ackreq_default(rdev, wpan_dev, ackreq); } +static int nl802154_wpan_phy_netns(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg802154_registered_device *rdev = info->user_ptr[0]; + struct net *net; + int err; + + if (info->attrs[NL802154_ATTR_PID]) { + u32 pid = nla_get_u32(info->attrs[NL802154_ATTR_PID]); + + net = get_net_ns_by_pid(pid); + } else if (info->attrs[NL802154_ATTR_NETNS_FD]) { + u32 fd = nla_get_u32(info->attrs[NL802154_ATTR_NETNS_FD]); + + net = get_net_ns_by_fd(fd); + } else { + return -EINVAL; + } + + if (IS_ERR(net)) + return PTR_ERR(net); + + err = 0; + + /* check if anything to do */ + if (!net_eq(wpan_phy_net(&rdev->wpan_phy), net)) + err = cfg802154_switch_netns(rdev, net); + + put_net(net); + return err; +} + #ifdef CONFIG_IEEE802154_NL802154_EXPERIMENTAL static const struct nla_policy nl802154_dev_addr_policy[NL802154_DEV_ADDR_ATTR_MAX + 1] = { [NL802154_DEV_ADDR_ATTR_PAN_ID] = { .type = NLA_U16 }, @@ -2262,6 +2298,14 @@ static const struct genl_ops nl802154_ops[] = { NL802154_FLAG_NEED_RTNL, }, { + .cmd = NL802154_CMD_SET_WPAN_PHY_NETNS, + .doit = nl802154_wpan_phy_netns, + .policy = nl802154_policy, + .flags = GENL_ADMIN_PERM, + .internal_flags = NL802154_FLAG_NEED_WPAN_PHY | + NL802154_FLAG_NEED_RTNL, + }, + { .cmd = NL802154_CMD_SET_PAN_ID, .doit = nl802154_set_pan_id, .policy = nl802154_policy, -- cgit v0.10.2 From 1c5bf998b3dca0599a2cce885619ffc06fc594df Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Sat, 18 Jun 2016 10:45:35 +0200 Subject: ieee802154: allow netns create of lowpan interface This patch reverts commit f9d1ce8f81eb ("ieee802154: fix netns settings"). The lowpan interface need to be created inside the net namespace where the wpan interface is available. The wpan namespace can be changed only by nl802154 before. Without this patch it's not possible to create a lowpan interface for a wpan interface which isn't inside init_net namespace. Cc: Nicolas Dichtel Reviewed-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index 935ab93..d7efbf0 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -130,8 +130,7 @@ static int lowpan_newlink(struct net *src_net, struct net_device *ldev, pr_debug("adding new link\n"); - if (!tb[IFLA_LINK] || - !net_eq(dev_net(ldev), &init_net)) + if (!tb[IFLA_LINK]) return -EINVAL; /* find and hold wpan device */ wdev = dev_get_by_index(dev_net(ldev), nla_get_u32(tb[IFLA_LINK])); -- cgit v0.10.2 From b5f34f9420b50c9b5876b9a2b68e96be6d629054 Mon Sep 17 00:00:00 2001 From: Denis Kenzior Date: Mon, 27 Jun 2016 11:01:12 -0500 Subject: Bluetooth: Fix bt_sock_recvmsg return value MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If recvmsg is called with a destination buffer that is too small to receive the contents of skb in its entirety, the return value from recvmsg was inconsistent with common SOCK_SEQPACKET or SOCK_DGRAM semantics. If destination buffer provided by userspace is too small (e.g. len < copied), then MSG_TRUNC flag is set and copied is returned. Instead, it should return the length of the message, which is consistent with how other datagram based sockets act. Quoting 'man recv': "All three calls return the length of the message on successful comple‐ tion. If a message is too long to fit in the supplied buffer, excess bytes may be discarded depending on the type of socket the message is received from." and "MSG_TRUNC (since Linux 2.2) For raw (AF_PACKET), Internet datagram (since Linux 2.4.27/2.6.8), netlink (since Linux 2.6.22), and UNIX datagram (since Linux 3.4) sockets: return the real length of the packet or datagram, even when it was longer than the passed buffer." Signed-off-by: Denis Kenzior Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index 3df7aef..ece45e0 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -215,6 +215,7 @@ int bt_sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, struct sock *sk = sock->sk; struct sk_buff *skb; size_t copied; + size_t skblen; int err; BT_DBG("sock %p sk %p len %zu", sock, sk, len); @@ -230,6 +231,7 @@ int bt_sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, return err; } + skblen = skb->len; copied = skb->len; if (len < copied) { msg->msg_flags |= MSG_TRUNC; @@ -248,6 +250,9 @@ int bt_sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, skb_free_datagram(sk, skb); + if (msg->msg_flags & MSG_TRUNC) + copied = skblen; + return err ? : copied; } EXPORT_SYMBOL(bt_sock_recvmsg); -- cgit v0.10.2 From 83871f8ccdfa8f3edab15d432aa4f3eb82953343 Mon Sep 17 00:00:00 2001 From: Denis Kenzior Date: Mon, 27 Jun 2016 11:01:13 -0500 Subject: Bluetooth: Fix hci_sock_recvmsg return value MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If recvmsg is called with a destination buffer that is too small to receive the contents of skb in its entirety, the return value from recvmsg was inconsistent with common SOCK_SEQPACKET or SOCK_DGRAM semantics. If destination buffer provided by userspace is too small (e.g. len < copied), then MSG_TRUNC flag is set and copied is returned. Instead, it should return the length of the message, which is consistent with how other datagram based sockets act. Quoting 'man recv': "All three calls return the length of the message on successful comple‐ tion. If a message is too long to fit in the supplied buffer, excess bytes may be discarded depending on the type of socket the message is received from." and "MSG_TRUNC (since Linux 2.2) For raw (AF_PACKET), Internet datagram (since Linux 2.4.27/2.6.8), netlink (since Linux 2.6.22), and UNIX datagram (since Linux 3.4) sockets: return the real length of the packet or datagram, even when it was longer than the passed buffer." Signed-off-by: Denis Kenzior Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index 1298d72..12e9294 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -1048,6 +1048,7 @@ static int hci_sock_recvmsg(struct socket *sock, struct msghdr *msg, struct sock *sk = sock->sk; struct sk_buff *skb; int copied, err; + unsigned int skblen; BT_DBG("sock %p, sk %p", sock, sk); @@ -1064,6 +1065,7 @@ static int hci_sock_recvmsg(struct socket *sock, struct msghdr *msg, if (!skb) return err; + skblen = skb->len; copied = skb->len; if (len < copied) { msg->msg_flags |= MSG_TRUNC; @@ -1089,6 +1091,9 @@ static int hci_sock_recvmsg(struct socket *sock, struct msghdr *msg, skb_free_datagram(sk, skb); + if (msg->msg_flags & MSG_TRUNC) + copied = skblen; + return err ? : copied; } -- cgit v0.10.2 From 230b04ac8f439d0797ab85fb356f069f0472306f Mon Sep 17 00:00:00 2001 From: Tedd Ho-Jeong An Date: Tue, 28 Jun 2016 08:56:39 -0700 Subject: Bluetooth: Replace constant hw_variant from Intel Bluetooth firmware filename The format of Intel Bluetooth firmware filename for bootloader product is ibt--.sfi Currently the driver uses a constant value 11 (0x0b) for hw_variant to support LnP/SfP product. But new product like WsP product has a different value such as 12 (0x0c). To support the multiple products, this patch replaces the constant value of hw_variant to the actual hw_variant value read from the device. Signed-off-by: Tedd Ho-Jeong An Signed-off-by: Marcel Holtmann diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index a26278b..f8f0288 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -2104,10 +2104,14 @@ static int btusb_setup_intel_new(struct hci_dev *hdev) /* With this Intel bootloader only the hardware variant and device * revision information are used to select the right firmware. * - * Currently this bootloader support is limited to hardware variant - * iBT 3.0 (LnP/SfP) which is identified by the value 11 (0x0b). + * The firmware filename is ibt--.sfi. + * + * Currently the supported hardware variants are: + * 11 (0x0b) for iBT3.0 (LnP/SfP) + * 12 (0x0c) for iBT3.5 (WsP) */ - snprintf(fwname, sizeof(fwname), "intel/ibt-11-%u.sfi", + snprintf(fwname, sizeof(fwname), "intel/ibt-%u-%u.sfi", + le16_to_cpu(ver.hw_variant), le16_to_cpu(params->dev_revid)); err = request_firmware(&fw, fwname, &hdev->dev); @@ -2123,7 +2127,8 @@ static int btusb_setup_intel_new(struct hci_dev *hdev) /* Save the DDC file name for later use to apply once the firmware * downloading is done. */ - snprintf(fwname, sizeof(fwname), "intel/ibt-11-%u.ddc", + snprintf(fwname, sizeof(fwname), "intel/ibt-%u-%u.ddc", + le16_to_cpu(ver.hw_variant), le16_to_cpu(params->dev_revid)); kfree_skb(skb); -- cgit v0.10.2 From d2f302409e022506640363277ea8f8034c558c6c Mon Sep 17 00:00:00 2001 From: Ricky Liang Date: Tue, 28 Jun 2016 15:06:57 +0800 Subject: Bluetooth: btmrvl: fix slab-out-of-bounds access in btmrvl_sdio Kasan reported slab-out-of-bounds access in btmrvl_sdio: [ 33.055400] ================================================================== [ 33.062585] BUG: KASAN: slab-out-of-bounds in memcpy+0x24/0x50 at addr ffffffc0d89b4a00 [ 33.070529] Read of size 256 by task btmrvl_main_ser/3576 [ 33.075885] ============================================================================= [ 33.084002] BUG kmalloc-256 (Tainted: G B ): kasan: bad access detected [ 33.091511] ----------------------------------------------------------------------------- [ 33.413498] Call trace: [ 33.415928] [] dump_backtrace+0x0/0x190 [ 33.421288] [] show_stack+0x1c/0x28 [ 33.426305] [] dump_stack+0xa0/0xf8 [ 33.431320] [] print_trailer+0x158/0x16c [ 33.436765] [] object_err+0x48/0x5c [ 33.441780] [] kasan_report+0x344/0x510 [ 33.447141] [] __asan_loadN+0x20/0x150 [ 33.452413] [] memcpy+0x20/0x50 [ 33.457084] [] swiotlb_tbl_map_single+0x2ec/0x310 [ 33.463305] [] map_single+0x24/0x30 [ 33.468320] [] swiotlb_map_sg_attrs+0xec/0x21c [ 33.474286] [] __swiotlb_map_sg_attrs+0x48/0xec [ 33.480339] [] msdc_prepare_data.isra.11+0xf0/0x11c [ 33.486733] [] msdc_ops_request+0x74/0xf0 [ 33.492266] [] __mmc_start_request+0x78/0x8c [ 33.498057] [] mmc_start_request+0x220/0x240 [ 33.503848] [] mmc_wait_for_req+0x78/0x250 [ 33.509468] [] mmc_io_rw_extended+0x2ec/0x388 [ 33.515347] [] sdio_io_rw_ext_helper+0x160/0x268 [ 33.521483] [] sdio_writesb+0x40/0x50 [ 33.526677] [] btmrvl_sdio_host_to_card+0x124/0x1bc [btmrvl_sdio] [ 33.534283] [] btmrvl_service_main_thread+0x384/0x428 [btmrvl] [ 33.541626] [] kthread+0x140/0x158 [ 33.546550] Memory state around the buggy address: [ 33.551305] ffffffc0d89b4980: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 33.558474] ffffffc0d89b4a00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 [ 33.565643] >ffffffc0d89b4a80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fc [ 33.572809] ^ [ 33.579889] ffffffc0d89b4b00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 33.587055] ffffffc0d89b4b80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 33.594221] ================================================================== The cause of this is that btmrvl_sdio_host_to_card can access memory region out of its allocated space due to: 1. the requested block size is smaller than SDIO_BLOCK_SIZE, and/or 2. the allocated memory is not BTSDIO_DMA_ALIGN-aligned. This patch fixes the issue by allocating a buffer which is big enough for SDIO_BLOCK_SIZE transfer and/or BTSDIO_DMA_ALIGN address relocation. Signed-off-by: Ricky Liang Signed-off-by: Marcel Holtmann diff --git a/drivers/bluetooth/btmrvl_sdio.c b/drivers/bluetooth/btmrvl_sdio.c index f425ddf..b7c3928 100644 --- a/drivers/bluetooth/btmrvl_sdio.c +++ b/drivers/bluetooth/btmrvl_sdio.c @@ -1071,7 +1071,6 @@ static int btmrvl_sdio_host_to_card(struct btmrvl_private *priv, { struct btmrvl_sdio_card *card = priv->btmrvl_dev.card; int ret = 0; - int buf_block_len; int blksz; int i = 0; u8 *buf = NULL; @@ -1083,9 +1082,13 @@ static int btmrvl_sdio_host_to_card(struct btmrvl_private *priv, return -EINVAL; } + blksz = DIV_ROUND_UP(nb, SDIO_BLOCK_SIZE) * SDIO_BLOCK_SIZE; + buf = payload; - if ((unsigned long) payload & (BTSDIO_DMA_ALIGN - 1)) { - tmpbufsz = ALIGN_SZ(nb, BTSDIO_DMA_ALIGN); + if ((unsigned long) payload & (BTSDIO_DMA_ALIGN - 1) || + nb < blksz) { + tmpbufsz = ALIGN_SZ(blksz, BTSDIO_DMA_ALIGN) + + BTSDIO_DMA_ALIGN; tmpbuf = kzalloc(tmpbufsz, GFP_KERNEL); if (!tmpbuf) return -ENOMEM; @@ -1093,15 +1096,12 @@ static int btmrvl_sdio_host_to_card(struct btmrvl_private *priv, memcpy(buf, payload, nb); } - blksz = SDIO_BLOCK_SIZE; - buf_block_len = DIV_ROUND_UP(nb, blksz); - sdio_claim_host(card->func); do { /* Transfer data to card */ ret = sdio_writesb(card->func, card->ioport, buf, - buf_block_len * blksz); + blksz); if (ret < 0) { i++; BT_ERR("i=%d writesb failed: %d", i, ret); -- cgit v0.10.2 From 47ca5898eb09bc49666958f2ea7e0ea5b4add0f5 Mon Sep 17 00:00:00 2001 From: Yvain THONNART Date: Sat, 11 Jun 2016 15:43:30 +0200 Subject: Bluetooth: btusb: add support for device 0489:e092 With current btusb.ko kernel module, Bluetooth pretends to be active but there is no real activity. I'm using an Acer Aspire VN7-791. Output of lsusb: Bus 003 Device 007: ID 0489:e092 Foxconn / Hon Hai On my laptop, this device is actually used as a combo with wifi chipset Atheros Qualcomm Killer N1525 Wireless-AC [168c:003e], * Fix by adding a declaration in kernel sources drivers/bluetooth/btusb.c { USB_DEVICE(0x0489, 0xe092), .driver_info = BTUSB_QCA_ROME }, * Compiled extra module /lib/modules/4.4.0-22-generic/extra/btusb.ko * Successfully tested against my phone (obex file transfer) Signed-off-by: Yvain THONNART Signed-off-by: Marcel Holtmann diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index f8f0288..880bb55 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -249,6 +249,7 @@ static const struct usb_device_id blacklist_table[] = { { USB_DEVICE(0x0cf3, 0xe007), .driver_info = BTUSB_QCA_ROME }, { USB_DEVICE(0x0cf3, 0xe300), .driver_info = BTUSB_QCA_ROME }, { USB_DEVICE(0x0cf3, 0xe360), .driver_info = BTUSB_QCA_ROME }, + { USB_DEVICE(0x0489, 0xe092), .driver_info = BTUSB_QCA_ROME }, /* Broadcom BCM2035 */ { USB_DEVICE(0x0a5c, 0x2009), .driver_info = BTUSB_BCM92035 }, -- cgit v0.10.2 From 9cc577dd25b9762df7f353658426bb2e048c480a Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 6 Jul 2016 23:32:24 +0200 Subject: ieee802154: add ieee802154_skb_dst_pan helper This patch adds ieee802154_skb_dst_pan function to get the pointer address of the destination pan id at skb mac pointer. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/include/linux/ieee802154.h b/include/linux/ieee802154.h index acedbb6..91f4665 100644 --- a/include/linux/ieee802154.h +++ b/include/linux/ieee802154.h @@ -31,6 +31,8 @@ #define IEEE802154_MIN_PSDU_LEN 9 #define IEEE802154_FCS_LEN 2 #define IEEE802154_MAX_AUTH_TAG_LEN 16 +#define IEEE802154_FC_LEN 2 +#define IEEE802154_SEQ_LEN 1 /* General MAC frame format: * 2 bytes: Frame Control @@ -221,9 +223,14 @@ enum { #define IEEE802154_FCTL_ACKREQ 0x0020 #define IEEE802154_FCTL_SECEN 0x0004 #define IEEE802154_FCTL_INTRA_PAN 0x0040 +#define IEEE802154_FCTL_DADDR 0x0c00 #define IEEE802154_FTYPE_DATA 0x0001 +#define IEEE802154_FCTL_ADDR_NONE 0x0000 +#define IEEE802154_FCTL_DADDR_SHORT 0x0800 +#define IEEE802154_FCTL_DADDR_EXTENDED 0x0c00 + /* * ieee802154_is_data - check if type is IEEE802154_FTYPE_DATA * @fc: frame control bytes in little-endian byteorder @@ -261,6 +268,15 @@ static inline bool ieee802154_is_intra_pan(__le16 fc) return fc & cpu_to_le16(IEEE802154_FCTL_INTRA_PAN); } +/* + * ieee802154_daddr_mode - get daddr mode from fc + * @fc: frame control bytes in little-endian byteorder + */ +static inline __le16 ieee802154_daddr_mode(__le16 fc) +{ + return fc & cpu_to_le16(IEEE802154_FCTL_DADDR); +} + /** * ieee802154_is_valid_psdu_len - check if psdu len is valid * available lengths: diff --git a/include/net/mac802154.h b/include/net/mac802154.h index e465c855..b3f7cd8 100644 --- a/include/net/mac802154.h +++ b/include/net/mac802154.h @@ -258,6 +258,35 @@ static inline __le16 ieee802154_get_fc_from_skb(const struct sk_buff *skb) } /** + * ieee802154_skb_dst_pan - get the pointer to destination pan field + * @fc: mac header frame control field + * @skb: skb where the destination pan pointer will be get from + */ +static inline unsigned char *ieee802154_skb_dst_pan(__le16 fc, + const struct sk_buff *skb) +{ + unsigned char *dst_pan; + + switch (ieee802154_daddr_mode(fc)) { + case cpu_to_le16(IEEE802154_FCTL_ADDR_NONE): + dst_pan = NULL; + break; + case cpu_to_le16(IEEE802154_FCTL_DADDR_SHORT): + case cpu_to_le16(IEEE802154_FCTL_DADDR_EXTENDED): + dst_pan = skb_mac_header(skb) + + IEEE802154_FC_LEN + + IEEE802154_SEQ_LEN; + break; + default: + WARN_ONCE(1, "invalid addr mode detected"); + dst_pan = NULL; + break; + } + + return dst_pan; +} + +/** * ieee802154_be64_to_le64 - copies and convert be64 to le64 * @le64_dst: le64 destination pointer * @be64_src: be64 source pointer -- cgit v0.10.2 From 19580cc1ed299c736b56b45c7576b477f185f8f5 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 6 Jul 2016 23:32:25 +0200 Subject: ieee802154: add ieee802154_skb_src_pan helper This patch adds ieee802154_skb_src_pan function to get the pointer address of the source pan id at skb mac pointer. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/include/linux/ieee802154.h b/include/linux/ieee802154.h index 91f4665..ddb8901 100644 --- a/include/linux/ieee802154.h +++ b/include/linux/ieee802154.h @@ -50,6 +50,7 @@ #define IEEE802154_EXTENDED_ADDR_LEN 8 #define IEEE802154_SHORT_ADDR_LEN 2 +#define IEEE802154_PAN_ID_LEN 2 #define IEEE802154_LIFS_PERIOD 40 #define IEEE802154_SIFS_PERIOD 12 @@ -224,12 +225,15 @@ enum { #define IEEE802154_FCTL_SECEN 0x0004 #define IEEE802154_FCTL_INTRA_PAN 0x0040 #define IEEE802154_FCTL_DADDR 0x0c00 +#define IEEE802154_FCTL_SADDR 0xc000 #define IEEE802154_FTYPE_DATA 0x0001 #define IEEE802154_FCTL_ADDR_NONE 0x0000 #define IEEE802154_FCTL_DADDR_SHORT 0x0800 #define IEEE802154_FCTL_DADDR_EXTENDED 0x0c00 +#define IEEE802154_FCTL_SADDR_SHORT 0x8000 +#define IEEE802154_FCTL_SADDR_EXTENDED 0xc000 /* * ieee802154_is_data - check if type is IEEE802154_FTYPE_DATA @@ -277,6 +281,15 @@ static inline __le16 ieee802154_daddr_mode(__le16 fc) return fc & cpu_to_le16(IEEE802154_FCTL_DADDR); } +/* + * ieee802154_saddr_mode - get saddr mode from fc + * @fc: frame control bytes in little-endian byteorder + */ +static inline __le16 ieee802154_saddr_mode(__le16 fc) +{ + return fc & cpu_to_le16(IEEE802154_FCTL_SADDR); +} + /** * ieee802154_is_valid_psdu_len - check if psdu len is valid * available lengths: diff --git a/include/net/mac802154.h b/include/net/mac802154.h index b3f7cd8..ec01b35 100644 --- a/include/net/mac802154.h +++ b/include/net/mac802154.h @@ -287,6 +287,65 @@ static inline unsigned char *ieee802154_skb_dst_pan(__le16 fc, } /** + * ieee802154_skb_src_pan - get the pointer to source pan field + * @fc: mac header frame control field + * @skb: skb where the source pan pointer will be get from + */ +static inline unsigned char *ieee802154_skb_src_pan(__le16 fc, + const struct sk_buff *skb) +{ + unsigned char *src_pan; + + switch (ieee802154_saddr_mode(fc)) { + case cpu_to_le16(IEEE802154_FCTL_ADDR_NONE): + src_pan = NULL; + break; + case cpu_to_le16(IEEE802154_FCTL_SADDR_SHORT): + case cpu_to_le16(IEEE802154_FCTL_SADDR_EXTENDED): + /* if intra-pan and source addr mode is non none, + * then source pan id is equal destination pan id. + */ + if (ieee802154_is_intra_pan(fc)) { + src_pan = ieee802154_skb_dst_pan(fc, skb); + break; + } + + switch (ieee802154_daddr_mode(fc)) { + case cpu_to_le16(IEEE802154_FCTL_ADDR_NONE): + src_pan = skb_mac_header(skb) + + IEEE802154_FC_LEN + + IEEE802154_SEQ_LEN; + break; + case cpu_to_le16(IEEE802154_FCTL_DADDR_SHORT): + src_pan = skb_mac_header(skb) + + IEEE802154_FC_LEN + + IEEE802154_SEQ_LEN + + IEEE802154_PAN_ID_LEN + + IEEE802154_SHORT_ADDR_LEN; + break; + case cpu_to_le16(IEEE802154_FCTL_DADDR_EXTENDED): + src_pan = skb_mac_header(skb) + + IEEE802154_FC_LEN + + IEEE802154_SEQ_LEN + + IEEE802154_PAN_ID_LEN + + IEEE802154_EXTENDED_ADDR_LEN; + break; + default: + WARN_ONCE(1, "invalid addr mode detected"); + src_pan = NULL; + break; + } + break; + default: + WARN_ONCE(1, "invalid addr mode detected"); + src_pan = NULL; + break; + } + + return src_pan; +} + +/** * ieee802154_be64_to_le64 - copies and convert be64 to le64 * @le64_dst: le64 destination pointer * @be64_src: be64 source pointer -- cgit v0.10.2 From 0ea0b9af9b7599ada307258dc841f4300873e8a1 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 6 Jul 2016 23:32:26 +0200 Subject: ieee802154: 6lowpan: fix intra pan id check The RIOT-OS stack does send intra-pan frames but don't set the intra pan flag inside the mac header. It seems this is valid frame addressing but inefficient. Anyway this patch adds a new function for intra pan addressing, doesn't matter if intra pan flag or source and destination are the same. The newly introduction function will be used to check on intra pan addressing for 6lowpan. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/include/net/mac802154.h b/include/net/mac802154.h index ec01b35..d757edd 100644 --- a/include/net/mac802154.h +++ b/include/net/mac802154.h @@ -346,6 +346,25 @@ static inline unsigned char *ieee802154_skb_src_pan(__le16 fc, } /** + * ieee802154_skb_is_intra_pan_addressing - checks whenever the mac addressing + * is an intra pan communication + * @fc: mac header frame control field + * @skb: skb where the source and destination pan should be get from + */ +static inline bool ieee802154_skb_is_intra_pan_addressing(__le16 fc, + const struct sk_buff *skb) +{ + unsigned char *dst_pan = ieee802154_skb_dst_pan(fc, skb), + *src_pan = ieee802154_skb_src_pan(fc, skb); + + /* if one is NULL is no intra pan addressing */ + if (!dst_pan || !src_pan) + return false; + + return !memcmp(dst_pan, src_pan, IEEE802154_PAN_ID_LEN); +} + +/** * ieee802154_be64_to_le64 - copies and convert be64 to le64 * @le64_dst: le64 destination pointer * @be64_src: be64 source pointer diff --git a/net/ieee802154/6lowpan/rx.c b/net/ieee802154/6lowpan/rx.c index ef185dd..649e7d45 100644 --- a/net/ieee802154/6lowpan/rx.c +++ b/net/ieee802154/6lowpan/rx.c @@ -262,7 +262,7 @@ static inline bool lowpan_rx_h_check(struct sk_buff *skb) /* check on ieee802154 conform 6LoWPAN header */ if (!ieee802154_is_data(fc) || - !ieee802154_is_intra_pan(fc)) + !ieee802154_skb_is_intra_pan_addressing(fc, skb)) return false; /* check if we can dereference the dispatch */ -- cgit v0.10.2 From aaa7088eb29a0ffe6cf8d8a443695df41e5a62f3 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 6 Jul 2016 23:32:27 +0200 Subject: ieee802154: fix skb get fc on big endian This patch fixes ieee802154_get_fc_from_skb function on big endian machines. The function get_unaligned_le16 converts the byte order to host byte order but we want to keep the byte order like in mac header. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/include/net/mac802154.h b/include/net/mac802154.h index d757edd..bb7bfec 100644 --- a/include/net/mac802154.h +++ b/include/net/mac802154.h @@ -247,6 +247,8 @@ struct ieee802154_ops { */ static inline __le16 ieee802154_get_fc_from_skb(const struct sk_buff *skb) { + __le16 fc; + /* check if we can fc at skb_mac_header of sk buffer */ if (unlikely(!skb_mac_header_was_set(skb) || (skb_tail_pointer(skb) - skb_mac_header(skb)) < 2)) { @@ -254,7 +256,8 @@ static inline __le16 ieee802154_get_fc_from_skb(const struct sk_buff *skb) return cpu_to_le16(0); } - return get_unaligned_le16(skb_mac_header(skb)); + memcpy(&fc, skb_mac_header(skb), IEEE802154_FC_LEN); + return fc; } /** -- cgit v0.10.2 From 9e262f5037b95a9ccc508debec2715e70559cc81 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 6 Jul 2016 23:32:28 +0200 Subject: 6lowpan: ndisc: set invalid unicast short addr to unspec When receiving neighbour information with short address option field we should check the complete range of invalid short addresses and set it to one invalid address setting which is the unspecified address. This address is also used when by creating at first a new neighbour entry to indicate no short address is set. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/net/6lowpan/ndisc.c b/net/6lowpan/ndisc.c index 79c5fa9..86450b7 100644 --- a/net/6lowpan/ndisc.c +++ b/net/6lowpan/ndisc.c @@ -97,10 +97,13 @@ static void lowpan_ndisc_802154_update(struct neighbour *n, u32 flags, } write_lock_bh(&n->lock); - if (lladdr_short) + if (lladdr_short) { ieee802154_be16_to_le16(&neigh->short_addr, lladdr_short); - else + if (!lowpan_802154_is_valid_src_short_addr(neigh->short_addr)) + neigh->short_addr = cpu_to_le16(IEEE802154_ADDR_SHORT_UNSPEC); + } else { neigh->short_addr = cpu_to_le16(IEEE802154_ADDR_SHORT_UNSPEC); + } write_unlock_bh(&n->lock); } -- cgit v0.10.2 From 38961294510ab20d24809c125c0a3738a02d3a28 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 6 Jul 2016 23:32:29 +0200 Subject: atusb: don't change csma settings while set channel This patch fixes the behaviour to not overwrite csma settings when set channel afterwards. Cc: Stefan Schmidt Acked-by: Stefan Schmidt Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/drivers/net/ieee802154/atusb.c b/drivers/net/ieee802154/atusb.c index 52c9051..1056ed1 100644 --- a/drivers/net/ieee802154/atusb.c +++ b/drivers/net/ieee802154/atusb.c @@ -366,11 +366,7 @@ static int atusb_channel(struct ieee802154_hw *hw, u8 page, u8 channel) struct atusb *atusb = hw->priv; int ret; - /* This implicitly sets the CCA (Clear Channel Assessment) mode to 0, - * "Mode 3a, Carrier sense OR energy above threshold". - * We should probably make this configurable. @@@ - */ - ret = atusb_write_reg(atusb, RG_PHY_CC_CCA, channel); + ret = atusb_write_subreg(atusb, SR_CHANNEL, channel); if (ret < 0) return ret; msleep(1); /* @@@ ugly synchronization */ -- cgit v0.10.2 From 048e7f7e66a8693fe1707997bffd19d08cde08f5 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 6 Jul 2016 23:32:30 +0200 Subject: ieee802154: cleanup WARN_ON for fc fetch This patch cleanups the WARN_ON which occurs when the sk buffer has insufficient buffer space by moving the WARN_ON into if condition. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/include/net/mac802154.h b/include/net/mac802154.h index bb7bfec..286824a 100644 --- a/include/net/mac802154.h +++ b/include/net/mac802154.h @@ -250,11 +250,10 @@ static inline __le16 ieee802154_get_fc_from_skb(const struct sk_buff *skb) __le16 fc; /* check if we can fc at skb_mac_header of sk buffer */ - if (unlikely(!skb_mac_header_was_set(skb) || - (skb_tail_pointer(skb) - skb_mac_header(skb)) < 2)) { - WARN_ON(1); + if (WARN_ON(!skb_mac_header_was_set(skb) || + (skb_tail_pointer(skb) - + skb_mac_header(skb)) < IEEE802154_FC_LEN)) return cpu_to_le16(0); - } memcpy(&fc, skb_mac_header(skb), IEEE802154_FC_LEN); return fc; -- cgit v0.10.2 From 7c2b9bff56a293d2cacede87f57c3148de442458 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 6 Jul 2016 23:32:31 +0200 Subject: fakelb: allow to run as monitor For my RIOT-OS in userspace experiments I need to create a fakelb monitor interface. The fakelb doesn't filter anything on L2 and is a purely raw interface. Because nl802154 checks on promiscuous mode which need to supported by creating monitors this patch adds some no-op promiscuous mode setting and the promiscuous flag. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann diff --git a/drivers/net/ieee802154/fakelb.c b/drivers/net/ieee802154/fakelb.c index 860d4ae..0becf0a 100644 --- a/drivers/net/ieee802154/fakelb.c +++ b/drivers/net/ieee802154/fakelb.c @@ -112,6 +112,12 @@ static void fakelb_hw_stop(struct ieee802154_hw *hw) write_unlock_bh(&fakelb_ifup_phys_lock); } +static int +fakelb_set_promiscuous_mode(struct ieee802154_hw *hw, const bool on) +{ + return 0; +} + static const struct ieee802154_ops fakelb_ops = { .owner = THIS_MODULE, .xmit_async = fakelb_hw_xmit, @@ -119,6 +125,7 @@ static const struct ieee802154_ops fakelb_ops = { .set_channel = fakelb_hw_channel, .start = fakelb_hw_start, .stop = fakelb_hw_stop, + .set_promiscuous_mode = fakelb_set_promiscuous_mode, }; /* Number of dummy devices to be set up by this module. */ @@ -174,6 +181,7 @@ static int fakelb_add_one(struct device *dev) hw->phy->current_channel = 13; phy->channel = hw->phy->current_channel; + hw->flags = IEEE802154_HW_PROMISCUOUS; hw->parent = dev; err = ieee802154_register_hw(hw); -- cgit v0.10.2 From c14ee43b8ae2fe0777335daaf278b86a9f6691a1 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:31:59 -0400 Subject: rtl8xxxu: Set all ieee80211_rx_status values in parse_rx_desc() This needs to be handled locally in the parse_rx_desc() function in order to be able to handle aggregated packets in the future. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index cfa5528..53efebe 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -5048,6 +5048,7 @@ static void rtl8xxxu_rx_urb_work(struct work_struct *work) int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb, struct ieee80211_rx_status *rx_status) { + struct ieee80211_hw *hw = priv->hw; struct rtl8xxxu_rxdesc16 *rx_desc = (struct rtl8xxxu_rxdesc16 *)skb->data; struct rtl8723au_phy_stats *phy_stats; @@ -5059,6 +5060,8 @@ int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb, for (i = 0; i < (sizeof(struct rtl8xxxu_rxdesc16) / sizeof(u32)); i++) _rx_desc[i] = le32_to_cpu(_rx_desc_le[i]); + memset(rx_status, 0, sizeof(struct ieee80211_rx_status)); + skb_pull(skb, sizeof(struct rtl8xxxu_rxdesc16)); phy_stats = (struct rtl8723au_phy_stats *)skb->data; @@ -5088,12 +5091,16 @@ int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb, rx_status->rate_idx = rx_desc->rxmcs; } + rx_status->freq = hw->conf.chandef.chan->center_freq; + rx_status->band = hw->conf.chandef.chan->band; + return RX_TYPE_DATA_PKT; } int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb, struct ieee80211_rx_status *rx_status) { + struct ieee80211_hw *hw = priv->hw; struct rtl8xxxu_rxdesc24 *rx_desc = (struct rtl8xxxu_rxdesc24 *)skb->data; struct rtl8723au_phy_stats *phy_stats; @@ -5105,6 +5112,8 @@ int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb, for (i = 0; i < (sizeof(struct rtl8xxxu_rxdesc24) / sizeof(u32)); i++) _rx_desc[i] = le32_to_cpu(_rx_desc_le[i]); + memset(rx_status, 0, sizeof(struct ieee80211_rx_status)); + skb_pull(skb, sizeof(struct rtl8xxxu_rxdesc24)); phy_stats = (struct rtl8723au_phy_stats *)skb->data; @@ -5140,6 +5149,9 @@ int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb, rx_status->rate_idx = rx_desc->rxmcs; } + rx_status->freq = hw->conf.chandef.chan->center_freq; + rx_status->band = hw->conf.chandef.chan->band; + return RX_TYPE_DATA_PKT; } @@ -5202,13 +5214,8 @@ static void rtl8xxxu_rx_complete(struct urb *urb) skb_put(skb, urb->actual_length); if (urb->status == 0) { - memset(rx_status, 0, sizeof(struct ieee80211_rx_status)); - rx_type = priv->fops->parse_rx_desc(priv, skb, rx_status); - rx_status->freq = hw->conf.chandef.chan->center_freq; - rx_status->band = hw->conf.chandef.chan->band; - if (rx_type == RX_TYPE_DATA_PKT) ieee80211_rx_irqsafe(hw, skb); else { -- cgit v0.10.2 From a635df8a41ace30d15cb6fa22a04ba5b4c364cff Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:00 -0400 Subject: rtl8xxxu: Move skb delivery into parse_tx_desc() handler This is another prepatory patch to be able to handle aggregated RX packets. In order to avoid adding a prototype, this also moves the rtl8723bu_handle_c2h() function. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index 53efebe..0fea007 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -5045,6 +5045,51 @@ static void rtl8xxxu_rx_urb_work(struct work_struct *work) } } +static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv, + struct sk_buff *skb) +{ + struct rtl8723bu_c2h *c2h = (struct rtl8723bu_c2h *)skb->data; + struct device *dev = &priv->udev->dev; + int len; + + len = skb->len - 2; + + dev_dbg(dev, "C2H ID %02x seq %02x, len %02x source %02x\n", + c2h->id, c2h->seq, len, c2h->bt_info.response_source); + + switch(c2h->id) { + case C2H_8723B_BT_INFO: + if (c2h->bt_info.response_source > + BT_INFO_SRC_8723B_BT_ACTIVE_SEND) + dev_dbg(dev, "C2H_BT_INFO WiFi only firmware\n"); + else + dev_dbg(dev, "C2H_BT_INFO BT/WiFi coexist firmware\n"); + + if (c2h->bt_info.bt_has_reset) + dev_dbg(dev, "BT has been reset\n"); + if (c2h->bt_info.tx_rx_mask) + dev_dbg(dev, "BT TRx mask\n"); + + break; + case C2H_8723B_BT_MP_INFO: + dev_dbg(dev, "C2H_MP_INFO ext ID %02x, status %02x\n", + c2h->bt_mp_info.ext_id, c2h->bt_mp_info.status); + break; + case C2H_8723B_RA_REPORT: + dev_dbg(dev, + "C2H RA RPT: rate %02x, unk %i, macid %02x, noise %i\n", + c2h->ra_report.rate, c2h->ra_report.dummy0_0, + c2h->ra_report.macid, c2h->ra_report.noisy_state); + break; + default: + dev_info(dev, "Unhandled C2H event %02x seq %02x\n", + c2h->id, c2h->seq); + print_hex_dump(KERN_INFO, "C2H content: ", DUMP_PREFIX_NONE, + 16, 1, c2h->raw.payload, len, false); + break; + } +} + int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb, struct ieee80211_rx_status *rx_status) { @@ -5094,6 +5139,7 @@ int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb, rx_status->freq = hw->conf.chandef.chan->center_freq; rx_status->band = hw->conf.chandef.chan->band; + ieee80211_rx_irqsafe(hw, skb); return RX_TYPE_DATA_PKT; } @@ -5125,6 +5171,8 @@ int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb, if (rx_desc->rpt_sel) { struct device *dev = &priv->udev->dev; dev_dbg(dev, "%s: C2H packet\n", __func__); + rtl8723bu_handle_c2h(priv, skb); + dev_kfree_skb(skb); return RX_TYPE_C2H; } @@ -5152,54 +5200,10 @@ int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb, rx_status->freq = hw->conf.chandef.chan->center_freq; rx_status->band = hw->conf.chandef.chan->band; + ieee80211_rx_irqsafe(hw, skb); return RX_TYPE_DATA_PKT; } -static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv, - struct sk_buff *skb) -{ - struct rtl8723bu_c2h *c2h = (struct rtl8723bu_c2h *)skb->data; - struct device *dev = &priv->udev->dev; - int len; - - len = skb->len - 2; - - dev_dbg(dev, "C2H ID %02x seq %02x, len %02x source %02x\n", - c2h->id, c2h->seq, len, c2h->bt_info.response_source); - - switch(c2h->id) { - case C2H_8723B_BT_INFO: - if (c2h->bt_info.response_source > - BT_INFO_SRC_8723B_BT_ACTIVE_SEND) - dev_dbg(dev, "C2H_BT_INFO WiFi only firmware\n"); - else - dev_dbg(dev, "C2H_BT_INFO BT/WiFi coexist firmware\n"); - - if (c2h->bt_info.bt_has_reset) - dev_dbg(dev, "BT has been reset\n"); - if (c2h->bt_info.tx_rx_mask) - dev_dbg(dev, "BT TRx mask\n"); - - break; - case C2H_8723B_BT_MP_INFO: - dev_dbg(dev, "C2H_MP_INFO ext ID %02x, status %02x\n", - c2h->bt_mp_info.ext_id, c2h->bt_mp_info.status); - break; - case C2H_8723B_RA_REPORT: - dev_dbg(dev, - "C2H RA RPT: rate %02x, unk %i, macid %02x, noise %i\n", - c2h->ra_report.rate, c2h->ra_report.dummy0_0, - c2h->ra_report.macid, c2h->ra_report.noisy_state); - break; - default: - dev_info(dev, "Unhandled C2H event %02x seq %02x\n", - c2h->id, c2h->seq); - print_hex_dump(KERN_INFO, "C2H content: ", DUMP_PREFIX_NONE, - 16, 1, c2h->raw.payload, len, false); - break; - } -} - static void rtl8xxxu_rx_complete(struct urb *urb) { struct rtl8xxxu_rx_urb *rx_urb = @@ -5209,19 +5213,11 @@ static void rtl8xxxu_rx_complete(struct urb *urb) struct sk_buff *skb = (struct sk_buff *)urb->context; struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); struct device *dev = &priv->udev->dev; - int rx_type; skb_put(skb, urb->actual_length); if (urb->status == 0) { - rx_type = priv->fops->parse_rx_desc(priv, skb, rx_status); - - if (rx_type == RX_TYPE_DATA_PKT) - ieee80211_rx_irqsafe(hw, skb); - else { - rtl8723bu_handle_c2h(priv, skb); - dev_kfree_skb(skb); - } + priv->fops->parse_rx_desc(priv, skb, rx_status); skb = NULL; rx_urb->urb.context = NULL; -- cgit v0.10.2 From 2db125d4af4c503564b66c615488e6e7a4583b86 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:01 -0400 Subject: rtl8xxxu: Obtain ieee80211_rx_status within parse_rx_desc() When handling aggregated packets, we'll get a new ieee80211_rx_status for each cloned skb, so passing in the pointer from the outside doesn't make sense. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h index 870c9cd..d0ccda8 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h @@ -1315,8 +1315,7 @@ struct rtl8xxxu_fileops { void (*phy_init_antenna_selection) (struct rtl8xxxu_priv *priv); void (*phy_iq_calibrate) (struct rtl8xxxu_priv *priv); void (*config_channel) (struct ieee80211_hw *hw); - int (*parse_rx_desc) (struct rtl8xxxu_priv *priv, struct sk_buff *skb, - struct ieee80211_rx_status *rx_status); + int (*parse_rx_desc) (struct rtl8xxxu_priv *priv, struct sk_buff *skb); void (*init_aggregation) (struct rtl8xxxu_priv *priv); void (*init_statistics) (struct rtl8xxxu_priv *priv); void (*enable_rf) (struct rtl8xxxu_priv *priv); @@ -1412,10 +1411,8 @@ void rtl8xxxu_gen2_report_connect(struct rtl8xxxu_priv *priv, void rtl8xxxu_gen1_enable_rf(struct rtl8xxxu_priv *priv); void rtl8xxxu_gen1_disable_rf(struct rtl8xxxu_priv *priv); void rtl8xxxu_gen2_disable_rf(struct rtl8xxxu_priv *priv); -int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb, - struct ieee80211_rx_status *rx_status); -int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb, - struct ieee80211_rx_status *rx_status); +int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb); +int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb); int rtl8xxxu_gen2_channel_to_group(int channel); bool rtl8xxxu_gen2_simularity_compare(struct rtl8xxxu_priv *priv, int result[][8], int c1, int c2); diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index 0fea007..400deea 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -5090,10 +5090,10 @@ static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv, } } -int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb, - struct ieee80211_rx_status *rx_status) +int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb) { struct ieee80211_hw *hw = priv->hw; + struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); struct rtl8xxxu_rxdesc16 *rx_desc = (struct rtl8xxxu_rxdesc16 *)skb->data; struct rtl8723au_phy_stats *phy_stats; @@ -5143,10 +5143,10 @@ int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb, return RX_TYPE_DATA_PKT; } -int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb, - struct ieee80211_rx_status *rx_status) +int rtl8xxxu_parse_rxdesc24(struct rtl8xxxu_priv *priv, struct sk_buff *skb) { struct ieee80211_hw *hw = priv->hw; + struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); struct rtl8xxxu_rxdesc24 *rx_desc = (struct rtl8xxxu_rxdesc24 *)skb->data; struct rtl8723au_phy_stats *phy_stats; @@ -5211,13 +5211,12 @@ static void rtl8xxxu_rx_complete(struct urb *urb) struct ieee80211_hw *hw = rx_urb->hw; struct rtl8xxxu_priv *priv = hw->priv; struct sk_buff *skb = (struct sk_buff *)urb->context; - struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); struct device *dev = &priv->udev->dev; skb_put(skb, urb->actual_length); if (urb->status == 0) { - priv->fops->parse_rx_desc(priv, skb, rx_status); + priv->fops->parse_rx_desc(priv, skb); skb = NULL; rx_urb->urb.context = NULL; -- cgit v0.10.2 From 41892729cf60a600fb14ed924f0d18440e2dfac9 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:02 -0400 Subject: rtl8xxxu: Correct rxdesc16 definition This corrects the definition of rxdesc16 to correctly specify pkt_cnt for aggregated packets. This is based on the code of the vendor rtl8723au driver, as opposed to the struct definitions they use. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h index d0ccda8..e8158c0 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h @@ -135,7 +135,8 @@ struct rtl8xxxu_rxdesc16 { u32 seq:12; u32 frag:4; - u32 nextpktlen:14; + u32 pkt_cnt:8; + u32 reserved:6; u32 nextind:1; u32 reserved0:1; @@ -198,7 +199,8 @@ struct rtl8xxxu_rxdesc16 { u32 reserved0:1; u32 nextind:1; - u32 nextpktlen:14; + u32 reserved:6; + u32 pkt_cnt:8; u32 frag:4; u32 seq:12; -- cgit v0.10.2 From 040b97be60567b819b97442d30533884bd266874 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:03 -0400 Subject: rtl8xxxu: Add support for aggregated RX packets on gen1 parts This implements support for demuxing aggregated RX packets on gen1 devices, using the rxdesc16 format. So far this has only been tested with rtl8723au devices. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index 400deea..8643bb3 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -5093,53 +5093,88 @@ static void rtl8723bu_handle_c2h(struct rtl8xxxu_priv *priv, int rtl8xxxu_parse_rxdesc16(struct rtl8xxxu_priv *priv, struct sk_buff *skb) { struct ieee80211_hw *hw = priv->hw; - struct ieee80211_rx_status *rx_status = IEEE80211_SKB_RXCB(skb); - struct rtl8xxxu_rxdesc16 *rx_desc = - (struct rtl8xxxu_rxdesc16 *)skb->data; + struct ieee80211_rx_status *rx_status; + struct rtl8xxxu_rxdesc16 *rx_desc; struct rtl8723au_phy_stats *phy_stats; - __le32 *_rx_desc_le = (__le32 *)skb->data; - u32 *_rx_desc = (u32 *)skb->data; + struct sk_buff *next_skb = NULL; + __le32 *_rx_desc_le; + u32 *_rx_desc; int drvinfo_sz, desc_shift; - int i; + int i, pkt_cnt, pkt_len, urb_len, pkt_offset; - for (i = 0; i < (sizeof(struct rtl8xxxu_rxdesc16) / sizeof(u32)); i++) - _rx_desc[i] = le32_to_cpu(_rx_desc_le[i]); + urb_len = skb->len; + pkt_cnt = 0; - memset(rx_status, 0, sizeof(struct ieee80211_rx_status)); + do { + rx_desc = (struct rtl8xxxu_rxdesc16 *)skb->data; + _rx_desc_le = (__le32 *)skb->data; + _rx_desc = (u32 *)skb->data; - skb_pull(skb, sizeof(struct rtl8xxxu_rxdesc16)); + for (i = 0; + i < (sizeof(struct rtl8xxxu_rxdesc16) / sizeof(u32)); i++) + _rx_desc[i] = le32_to_cpu(_rx_desc_le[i]); - phy_stats = (struct rtl8723au_phy_stats *)skb->data; + /* + * Only read pkt_cnt from the header if we're parsing the + * first packet + */ + if (!pkt_cnt) + pkt_cnt = rx_desc->pkt_cnt; + pkt_len = rx_desc->pktlen; - drvinfo_sz = rx_desc->drvinfo_sz * 8; - desc_shift = rx_desc->shift; - skb_pull(skb, drvinfo_sz + desc_shift); + drvinfo_sz = rx_desc->drvinfo_sz * 8; + desc_shift = rx_desc->shift; + pkt_offset = roundup(pkt_len + drvinfo_sz + desc_shift + + sizeof(struct rtl8xxxu_rxdesc16), 128); - if (rx_desc->phy_stats) - rtl8xxxu_rx_parse_phystats(priv, rx_status, phy_stats, - rx_desc->rxmcs); + if (pkt_cnt > 1) + next_skb = skb_clone(skb, GFP_ATOMIC); - rx_status->mactime = le32_to_cpu(rx_desc->tsfl); - rx_status->flag |= RX_FLAG_MACTIME_START; + rx_status = IEEE80211_SKB_RXCB(skb); + memset(rx_status, 0, sizeof(struct ieee80211_rx_status)); - if (!rx_desc->swdec) - rx_status->flag |= RX_FLAG_DECRYPTED; - if (rx_desc->crc32) - rx_status->flag |= RX_FLAG_FAILED_FCS_CRC; - if (rx_desc->bw) - rx_status->flag |= RX_FLAG_40MHZ; + skb_pull(skb, sizeof(struct rtl8xxxu_rxdesc16)); - if (rx_desc->rxht) { - rx_status->flag |= RX_FLAG_HT; - rx_status->rate_idx = rx_desc->rxmcs - DESC_RATE_MCS0; - } else { - rx_status->rate_idx = rx_desc->rxmcs; - } + phy_stats = (struct rtl8723au_phy_stats *)skb->data; - rx_status->freq = hw->conf.chandef.chan->center_freq; - rx_status->band = hw->conf.chandef.chan->band; + skb_pull(skb, drvinfo_sz + desc_shift); + + skb_trim(skb, pkt_len); + + if (rx_desc->phy_stats) + rtl8xxxu_rx_parse_phystats(priv, rx_status, phy_stats, + rx_desc->rxmcs); + + rx_status->mactime = le32_to_cpu(rx_desc->tsfl); + rx_status->flag |= RX_FLAG_MACTIME_START; + + if (!rx_desc->swdec) + rx_status->flag |= RX_FLAG_DECRYPTED; + if (rx_desc->crc32) + rx_status->flag |= RX_FLAG_FAILED_FCS_CRC; + if (rx_desc->bw) + rx_status->flag |= RX_FLAG_40MHZ; + + if (rx_desc->rxht) { + rx_status->flag |= RX_FLAG_HT; + rx_status->rate_idx = rx_desc->rxmcs - DESC_RATE_MCS0; + } else { + rx_status->rate_idx = rx_desc->rxmcs; + } + + rx_status->freq = hw->conf.chandef.chan->center_freq; + rx_status->band = hw->conf.chandef.chan->band; + + ieee80211_rx_irqsafe(hw, skb); + + skb = next_skb; + if (skb) + skb_pull(next_skb, pkt_offset); + + pkt_cnt--; + urb_len -= pkt_offset; + } while (skb && urb_len > 0 && pkt_cnt > 0); - ieee80211_rx_irqsafe(hw, skb); return RX_TYPE_DATA_PKT; } -- cgit v0.10.2 From 04319ae2f60974fa218f16fdf7a8576d1db6d673 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:04 -0400 Subject: rtl8xxxu: Allocate larger RX skbs when aggregation is enabled This adds support for allocating larger skbs for devices which indicate they support it. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h index e8158c0..f06e3aa 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h @@ -1247,6 +1247,7 @@ struct rtl8xxxu_priv { u32 ep_tx_normal_queue:1; u32 ep_tx_low_queue:1; u32 has_xtalk:1; + u32 rx_buf_aggregation:1; u8 xtalk; unsigned int pipe_interrupt; unsigned int pipe_in; @@ -1330,6 +1331,7 @@ struct rtl8xxxu_fileops { void (*report_connect) (struct rtl8xxxu_priv *priv, u8 macid, bool connect); int writeN_block_size; + int rx_agg_buf_size; char tx_desc_size; char rx_desc_size; char has_s0s1; diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index 8643bb3..75149e4 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -64,8 +64,6 @@ module_param_named(ht40_2g, rtl8xxxu_ht40_2g, bool, 0600); MODULE_PARM_DESC(ht40_2g, "Enable HT40 support on the 2.4GHz band"); #define USB_VENDOR_ID_REALTEK 0x0bda -/* Minimum IEEE80211_MAX_FRAME_LEN */ -#define RTL_RX_BUFFER_SIZE IEEE80211_MAX_FRAME_LEN #define RTL8XXXU_RX_URBS 32 #define RTL8XXXU_RX_URB_PENDING_WATER 8 #define RTL8XXXU_TX_URBS 64 @@ -5271,12 +5269,19 @@ cleanup: static int rtl8xxxu_submit_rx_urb(struct rtl8xxxu_priv *priv, struct rtl8xxxu_rx_urb *rx_urb) { + struct rtl8xxxu_fileops *fops = priv->fops; struct sk_buff *skb; int skb_size; int ret, rx_desc_sz; - rx_desc_sz = priv->fops->rx_desc_size; - skb_size = rx_desc_sz + RTL_RX_BUFFER_SIZE; + rx_desc_sz = fops->rx_desc_size; + + if (priv->rx_buf_aggregation && fops->rx_agg_buf_size) + skb_size = fops->rx_agg_buf_size; + else + skb_size = IEEE80211_MAX_FRAME_LEN; + skb_size += rx_desc_sz; + skb = __netdev_alloc_skb(NULL, skb_size, GFP_KERNEL); if (!skb) return -ENOMEM; -- cgit v0.10.2 From 1e5b3b3fe9e00a3afe3092dbe1a3fc1e9079bf35 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:05 -0400 Subject: rtl8xxxu: Adjust RX skb size to include space for phystats The old allocation didn't leave space for phystats in the buffer, allowing the packet to be rejected if a frame size of size IEEE80211_MAX_FRAME_LEN was received. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index 75149e4..898601a 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -5276,11 +5276,12 @@ static int rtl8xxxu_submit_rx_urb(struct rtl8xxxu_priv *priv, rx_desc_sz = fops->rx_desc_size; - if (priv->rx_buf_aggregation && fops->rx_agg_buf_size) + if (priv->rx_buf_aggregation && fops->rx_agg_buf_size) { skb_size = fops->rx_agg_buf_size; - else + skb_size += (rx_desc_sz + sizeof(struct rtl8723au_phy_stats)); + } else { skb_size = IEEE80211_MAX_FRAME_LEN; - skb_size += rx_desc_sz; + } skb = __netdev_alloc_skb(NULL, skb_size, GFP_KERNEL); if (!skb) -- cgit v0.10.2 From 91dcbb7175317da7caafc3b05b002addd42cdabd Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:06 -0400 Subject: rtl8xxxu: Enable aggregation for rtl8723au Implement rtl8xxxu_gen1_init_aggregation(). Aggregation should be the same for all gen1 parts. We may want to allow for tuning parameters in the fileopes struct. For now this is based allocating 16KB RX buffers, leaving 16000 bytes for actual packets, and the rest for the skb overhead. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h index f06e3aa..4341d56 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu.h @@ -1412,6 +1412,7 @@ void rtl8xxxu_gen1_report_connect(struct rtl8xxxu_priv *priv, u8 macid, bool connect); void rtl8xxxu_gen2_report_connect(struct rtl8xxxu_priv *priv, u8 macid, bool connect); +void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv); void rtl8xxxu_gen1_enable_rf(struct rtl8xxxu_priv *priv); void rtl8xxxu_gen1_disable_rf(struct rtl8xxxu_priv *priv); void rtl8xxxu_gen2_disable_rf(struct rtl8xxxu_priv *priv); diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c index a8e172c..686c551 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8723a.c @@ -377,6 +377,7 @@ struct rtl8xxxu_fileops rtl8723au_fops = { .phy_iq_calibrate = rtl8xxxu_gen1_phy_iq_calibrate, .config_channel = rtl8xxxu_gen1_config_channel, .parse_rx_desc = rtl8xxxu_parse_rxdesc16, + .init_aggregation = rtl8xxxu_gen1_init_aggregation, .enable_rf = rtl8xxxu_gen1_enable_rf, .disable_rf = rtl8xxxu_gen1_disable_rf, .usb_quirks = rtl8xxxu_gen1_usb_quirks, @@ -384,6 +385,7 @@ struct rtl8xxxu_fileops rtl8723au_fops = { .update_rate_mask = rtl8xxxu_update_rate_mask, .report_connect = rtl8xxxu_gen1_report_connect, .writeN_block_size = 1024, + .rx_agg_buf_size = 16000, .tx_desc_size = sizeof(struct rtl8xxxu_txdesc32), .rx_desc_size = sizeof(struct rtl8xxxu_rxdesc16), .adda_1t_init = 0x0b1b25a0, diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index 898601a..fb0e5b9 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -4405,6 +4405,40 @@ void rtl8xxxu_gen2_report_connect(struct rtl8xxxu_priv *priv, rtl8xxxu_gen2_h2c_cmd(priv, &h2c, sizeof(h2c.media_status_rpt)); } +void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) +{ + u8 agg_ctrl, usb_spec, page_thresh; + + usb_spec = rtl8xxxu_read8(priv, REG_USB_SPECIAL_OPTION); + usb_spec &= ~USB_SPEC_USB_AGG_ENABLE; + + agg_ctrl = rtl8xxxu_read8(priv, REG_TRXDMA_CTRL); + agg_ctrl &= ~TRXDMA_CTRL_RXDMA_AGG_EN; + + agg_ctrl |= TRXDMA_CTRL_RXDMA_AGG_EN; + + rtl8xxxu_write8(priv, REG_TRXDMA_CTRL, agg_ctrl); + rtl8xxxu_write8(priv, REG_USB_SPECIAL_OPTION, usb_spec); + + /* + * The number of packets we can take looks to be buffer size / 512 + * which matches the 512 byte rounding we have to do when de-muxing + * the packets. + * + * Sample numbers from the vendor driver: + * USB High-Speed mode values: + * RxAggBlockCount = 8 : 512 byte unit + * RxAggBlockTimeout = 6 + * RxAggPageCount = 48 : 128 byte unit + * RxAggPageTimeout = 4 or 6 (absolute time 34ms/(2^6)) + */ + + page_thresh = (priv->fops->rx_agg_buf_size / 512); + rtl8xxxu_write8(priv, REG_RXDMA_AGG_PG_TH, page_thresh); + rtl8xxxu_write8(priv, REG_USB_DMA_AGG_TO, 4); + priv->rx_buf_aggregation = 1; +} + static void rtl8xxxu_set_basic_rates(struct rtl8xxxu_priv *priv, u32 rate_cfg) { u32 val32; -- cgit v0.10.2 From 0a20ed8bd057c05ed138d8d39ac496d874c3ccb6 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:07 -0400 Subject: rtl8xxxu: Enable aggregation for rtl8192cu/rtl8188cu/rtl8188ru This enables aggregation on rtl8192cu and derivative parts. This uses the same parameters as for rtl8723au. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c index eefb7ca..69d1a14 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_8192c.c @@ -560,6 +560,7 @@ struct rtl8xxxu_fileops rtl8192cu_fops = { .phy_iq_calibrate = rtl8xxxu_gen1_phy_iq_calibrate, .config_channel = rtl8xxxu_gen1_config_channel, .parse_rx_desc = rtl8xxxu_parse_rxdesc16, + .init_aggregation = rtl8xxxu_gen1_init_aggregation, .enable_rf = rtl8xxxu_gen1_enable_rf, .disable_rf = rtl8xxxu_gen1_disable_rf, .usb_quirks = rtl8xxxu_gen1_usb_quirks, @@ -567,6 +568,7 @@ struct rtl8xxxu_fileops rtl8192cu_fops = { .update_rate_mask = rtl8xxxu_update_rate_mask, .report_connect = rtl8xxxu_gen1_report_connect, .writeN_block_size = 128, + .rx_agg_buf_size = 16000, .tx_desc_size = sizeof(struct rtl8xxxu_txdesc32), .rx_desc_size = sizeof(struct rtl8xxxu_rxdesc16), .adda_1t_init = 0x0b1b25a0, -- cgit v0.10.2 From 82cce22acd8ec7c75087825f23c053a642e7b672 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:08 -0400 Subject: rtl8xxxu: Make DMA aggregation optional by setting a module parameter Let the default to off until we have more data on the right default tuning values. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index fb0e5b9..33e2909 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -44,6 +44,7 @@ int rtl8xxxu_debug = RTL8XXXU_DEBUG_EFUSE; static bool rtl8xxxu_ht40_2g; +static bool rtl8xxxu_dma_aggregation; MODULE_AUTHOR("Jes Sorensen "); MODULE_DESCRIPTION("RTL8XXXu USB mac80211 Wireless LAN Driver"); @@ -62,6 +63,8 @@ module_param_named(debug, rtl8xxxu_debug, int, 0600); MODULE_PARM_DESC(debug, "Set debug mask"); module_param_named(ht40_2g, rtl8xxxu_ht40_2g, bool, 0600); MODULE_PARM_DESC(ht40_2g, "Enable HT40 support on the 2.4GHz band"); +module_param_named(dma_aggregation, rtl8xxxu_dma_aggregation, bool, 0600); +MODULE_PARM_DESC(dma_aggregation, "Enable DMA packet aggregation"); #define USB_VENDOR_ID_REALTEK 0x0bda #define RTL8XXXU_RX_URBS 32 @@ -4411,14 +4414,18 @@ void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) usb_spec = rtl8xxxu_read8(priv, REG_USB_SPECIAL_OPTION); usb_spec &= ~USB_SPEC_USB_AGG_ENABLE; + rtl8xxxu_write8(priv, REG_USB_SPECIAL_OPTION, usb_spec); agg_ctrl = rtl8xxxu_read8(priv, REG_TRXDMA_CTRL); agg_ctrl &= ~TRXDMA_CTRL_RXDMA_AGG_EN; - agg_ctrl |= TRXDMA_CTRL_RXDMA_AGG_EN; + if (!rtl8xxxu_dma_aggregation) { + rtl8xxxu_write8(priv, REG_TRXDMA_CTRL, agg_ctrl); + return; + } + agg_ctrl |= TRXDMA_CTRL_RXDMA_AGG_EN; rtl8xxxu_write8(priv, REG_TRXDMA_CTRL, agg_ctrl); - rtl8xxxu_write8(priv, REG_USB_SPECIAL_OPTION, usb_spec); /* * The number of packets we can take looks to be buffer size / 512 -- cgit v0.10.2 From 614e389f36a9d709469a74508d76c5df5141149a Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:09 -0400 Subject: rtl8xxxu: gen1: Set aggregation timeout (REG_RXDMA_AGG_PG_TH + 1) as well gen2 chips as well as 8188eu seems to use this register for setting DMA timeout threshold values, however the 8192cu is using REG_USB_DMA_AGG_TO. Set both to be on the safe side. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index 33e2909..dda8d11 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -4410,7 +4410,7 @@ void rtl8xxxu_gen2_report_connect(struct rtl8xxxu_priv *priv, void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) { - u8 agg_ctrl, usb_spec, page_thresh; + u8 agg_ctrl, usb_spec, page_thresh, timeout; usb_spec = rtl8xxxu_read8(priv, REG_USB_SPECIAL_OPTION); usb_spec &= ~USB_SPEC_USB_AGG_ENABLE; @@ -4442,7 +4442,14 @@ void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) page_thresh = (priv->fops->rx_agg_buf_size / 512); rtl8xxxu_write8(priv, REG_RXDMA_AGG_PG_TH, page_thresh); - rtl8xxxu_write8(priv, REG_USB_DMA_AGG_TO, 4); + /* + * REG_RXDMA_AGG_PG_TH + 1 seems to be the timeout register on + * gen2 chips and rtl8188eu. The rtl8723au seems unhappy if we + * don't set it, so better set both. + */ + timeout = 4; + rtl8xxxu_write8(priv, REG_RXDMA_AGG_PG_TH + 1, timeout); + rtl8xxxu_write8(priv, REG_USB_DMA_AGG_TO, timeout); priv->rx_buf_aggregation = 1; } -- cgit v0.10.2 From fd83f12278262feccd012b62c30643bd6a6c2888 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Mon, 27 Jun 2016 12:32:10 -0400 Subject: rtl8xxxu: gen1: Add module parameters to adjust DMA aggregation parameters This allows the user to specify DMA aggregation timout and block count. Blocks are presumably always 512 bytes, so the minimum block count is 6 for 802.11 packets. Signed-off-by: Jes Sorensen Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c index dda8d11..77048db 100644 --- a/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c +++ b/drivers/net/wireless/realtek/rtl8xxxu/rtl8xxxu_core.c @@ -45,6 +45,8 @@ int rtl8xxxu_debug = RTL8XXXU_DEBUG_EFUSE; static bool rtl8xxxu_ht40_2g; static bool rtl8xxxu_dma_aggregation; +static int rtl8xxxu_dma_agg_timeout = -1; +static int rtl8xxxu_dma_agg_pages = -1; MODULE_AUTHOR("Jes Sorensen "); MODULE_DESCRIPTION("RTL8XXXu USB mac80211 Wireless LAN Driver"); @@ -65,6 +67,10 @@ module_param_named(ht40_2g, rtl8xxxu_ht40_2g, bool, 0600); MODULE_PARM_DESC(ht40_2g, "Enable HT40 support on the 2.4GHz band"); module_param_named(dma_aggregation, rtl8xxxu_dma_aggregation, bool, 0600); MODULE_PARM_DESC(dma_aggregation, "Enable DMA packet aggregation"); +module_param_named(dma_agg_timeout, rtl8xxxu_dma_agg_timeout, int, 0600); +MODULE_PARM_DESC(dma_agg_timeout, "Set DMA aggregation timeout (range 1-127)"); +module_param_named(dma_agg_pages, rtl8xxxu_dma_agg_pages, int, 0600); +MODULE_PARM_DESC(dma_agg_pages, "Set DMA aggregation pages (range 1-127, 0 to disable)"); #define USB_VENDOR_ID_REALTEK 0x0bda #define RTL8XXXU_RX_URBS 32 @@ -4441,6 +4447,18 @@ void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) */ page_thresh = (priv->fops->rx_agg_buf_size / 512); + if (rtl8xxxu_dma_agg_pages >= 0) { + if (rtl8xxxu_dma_agg_pages <= page_thresh) + timeout = page_thresh; + else if (rtl8xxxu_dma_agg_pages <= 6) + dev_err(&priv->udev->dev, + "%s: dma_agg_pages=%i too small, minium is 6\n", + __func__, rtl8xxxu_dma_agg_pages); + else + dev_err(&priv->udev->dev, + "%s: dma_agg_pages=%i larger than limit %i\n", + __func__, rtl8xxxu_dma_agg_pages, page_thresh); + } rtl8xxxu_write8(priv, REG_RXDMA_AGG_PG_TH, page_thresh); /* * REG_RXDMA_AGG_PG_TH + 1 seems to be the timeout register on @@ -4448,6 +4466,16 @@ void rtl8xxxu_gen1_init_aggregation(struct rtl8xxxu_priv *priv) * don't set it, so better set both. */ timeout = 4; + + if (rtl8xxxu_dma_agg_timeout >= 0) { + if (rtl8xxxu_dma_agg_timeout <= 127) + timeout = rtl8xxxu_dma_agg_timeout; + else + dev_err(&priv->udev->dev, + "%s: Invalid dma_agg_timeout: %i\n", + __func__, rtl8xxxu_dma_agg_timeout); + } + rtl8xxxu_write8(priv, REG_RXDMA_AGG_PG_TH + 1, timeout); rtl8xxxu_write8(priv, REG_USB_DMA_AGG_TO, timeout); priv->rx_buf_aggregation = 1; -- cgit v0.10.2 From 8a1902374fa0ce51ff8421111c2cf6cdc67dfce7 Mon Sep 17 00:00:00 2001 From: Masanari Iida Date: Wed, 29 Jun 2016 12:37:19 +0900 Subject: rtlwifi: Fix typo in printk This patch fix spelling typos found in drivers/net/wireless/realtek. Signed-off-by: Masanari Iida Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c index 1d034a1..7498a12 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/phy.c @@ -1239,7 +1239,7 @@ u8 rtl88e_phy_sw_chnl(struct ieee80211_hw *hw) if (!(is_hal_stop(rtlhal)) && !(RT_CANNOT_IO(hw))) { rtl88e_phy_sw_chnl_callback(hw); RT_TRACE(rtlpriv, COMP_CHAN, DBG_LOUD, - "sw_chnl_inprogress false schdule workitem current channel %d\n", + "sw_chnl_inprogress false schedule workitem current channel %d\n", rtlphy->current_channel); rtlphy->sw_chnl_inprogress = false; } else { diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c index 56f2cec..60ab2ec 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192c/phy_common.c @@ -757,7 +757,7 @@ u8 rtl92c_phy_sw_chnl(struct ieee80211_hw *hw) if (!(is_hal_stop(rtlhal)) && !(RT_CANNOT_IO(hw))) { rtl92c_phy_sw_chnl_callback(hw); RT_TRACE(rtlpriv, COMP_CHAN, DBG_LOUD, - "sw_chnl_inprogress false schdule workitem\n"); + "sw_chnl_inprogress false schedule workitem\n"); rtlphy->sw_chnl_inprogress = false; } else { RT_TRACE(rtlpriv, COMP_CHAN, DBG_LOUD, diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c index a56a3da..e6b5786 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/dm.c @@ -496,7 +496,7 @@ static void rtl92ee_dm_find_minimum_rssi(struct ieee80211_hw *hw) rtl_dm_dig->min_undec_pwdb_for_dm = rtlpriv->dm.entry_min_undec_sm_pwdb; RT_TRACE(rtlpriv, COMP_BB_POWERSAVING, DBG_LOUD, - "AP Ext Port or disconnet PWDB = 0x%x\n", + "AP Ext Port or disconnect PWDB = 0x%x\n", rtl_dm_dig->min_undec_pwdb_for_dm); } RT_TRACE(rtlpriv, COMP_DIG, DBG_LOUD, diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c index d678bf7..beafc9a 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/phy.c @@ -1819,7 +1819,7 @@ u8 rtl92ee_phy_sw_chnl(struct ieee80211_hw *hw) if (!(is_hal_stop(rtlhal)) && !(RT_CANNOT_IO(hw))) { rtl92ee_phy_sw_chnl_callback(hw); RT_TRACE(rtlpriv, COMP_CHAN, DBG_LOUD, - "sw_chnl_inprogress false schdule workitem current channel %d\n", + "sw_chnl_inprogress false schedule workitem current channel %d\n", rtlphy->current_channel); rtlphy->sw_chnl_inprogress = false; } else { diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/phy.c index af07e00..601b78e 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/phy.c @@ -893,7 +893,7 @@ u8 rtl8723e_phy_sw_chnl(struct ieee80211_hw *hw) if (!(is_hal_stop(rtlhal)) && !(RT_CANNOT_IO(hw))) { rtl8723e_phy_sw_chnl_callback(hw); RT_TRACE(rtlpriv, COMP_CHAN, DBG_LOUD, - "sw_chnl_inprogress false schdule workitem\n"); + "sw_chnl_inprogress false schedule workitem\n"); rtlphy->sw_chnl_inprogress = false; } else { RT_TRACE(rtlpriv, COMP_CHAN, DBG_LOUD, diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c index 63b0df6..1c3e105 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c @@ -1474,7 +1474,7 @@ static enum version_8723e _rtl8723be_read_chip_version(struct ieee80211_hw *hw) value32 = rtl_read_dword(rtlpriv, REG_SYS_CFG1); if ((value32 & (CHIP_8723B)) != CHIP_8723B) - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "unkown chip version\n"); + RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "unknown chip version\n"); else version = (enum version_8723e)CHIP_8723B; diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c index b2b313a..285818d 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/phy.c @@ -1395,7 +1395,7 @@ u8 rtl8723be_phy_sw_chnl(struct ieee80211_hw *hw) if (!(is_hal_stop(rtlhal)) && !(RT_CANNOT_IO(hw))) { rtl8723be_phy_sw_chnl_callback(hw); RT_TRACE(rtlpriv, COMP_CHAN, DBG_LOUD, - "sw_chnl_inprogress false schdule workitem current channel %d\n", + "sw_chnl_inprogress false schedule workitem current channel %d\n", rtlphy->current_channel); rtlphy->sw_chnl_inprogress = false; } else { -- cgit v0.10.2 From b74d6e740be36d6d0c05f3386520fb44bae886b3 Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Wed, 29 Jun 2016 17:32:40 +0530 Subject: mwifiex: fix scan_block flag handling scan_block flag is used to block scan operation when 4 way handshake is in progress. Sometimes it doesn't get cleared due to incomplete association. An example is assoc request/response is done, but add key operation get canceled in some corner cases. As a result, further association/scan operations are blocked. This patch fixes the problem by clearing scan_block flag. Signed-off-by: Amitkumar Karwar Signed-off-by: Xinming Hu Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c index e6befd5..867ab81 100644 --- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c +++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c @@ -2250,6 +2250,9 @@ mwifiex_cfg80211_connect(struct wiphy *wiphy, struct net_device *dev, return -EALREADY; } + if (priv->scan_block) + priv->scan_block = false; + if (adapter->surprise_removed || adapter->is_cmd_timedout) { mwifiex_dbg(adapter, ERROR, "%s: Ignore connection.\t" @@ -2468,6 +2471,9 @@ mwifiex_cfg80211_scan(struct wiphy *wiphy, return -EBUSY; } + if (!priv->wdev.current_bss && priv->scan_block) + priv->scan_block = false; + if (!mwifiex_stop_bg_scan(priv)) cfg80211_sched_scan_stopped_rtnl(priv->wdev.wiphy); -- cgit v0.10.2 From a63b09872c1dc0ce0da3628647da67a112b484bf Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Wed, 29 Jun 2016 21:54:26 +0200 Subject: brcmfmac: delete interface directly in code that sent fw request MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit So far when receiving event about in-firmware-interface removal our event worker was notifying listener and afterwards it was removing Linux interface. First of all it was resulting in slightly unexpected order. The listener (del_virtual_intf callback) was (usually) returning with success before we even called unregister_netdev(ice). Please note this couldn't be simply fixed by changing order of calls in brcmf_fweh_handle_if_event as unregistering interface earlier could free struct brcmf_if. Another problem of current implementation are possible lockups. Focus on the time slot between calling event handler and removing Linux interface. During that time original caller may leave (unlocking rtnl semaphore) *and* another call to the same code may be done (locking it again). If that happens our event handler will stuck at removing Linux interface, it won't handle another event and will block process holding rtnl lock. This can be simply solved by unregistering interface in a proper callback, right after receiving confirmation event from firmware. This only required modifying worker to don't unregister on its own if there is someone waiting for the event. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c index 9da7a4c..79c081f 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c @@ -18,6 +18,7 @@ #include "brcmu_wifi.h" #include "brcmu_utils.h" +#include "cfg80211.h" #include "core.h" #include "debug.h" #include "tracepoint.h" @@ -182,8 +183,13 @@ static void brcmf_fweh_handle_if_event(struct brcmf_pub *drvr, err = brcmf_fweh_call_event_handler(ifp, emsg->event_code, emsg, data); - if (ifp && ifevent->action == BRCMF_E_IF_DEL) - brcmf_remove_interface(ifp, false); + if (ifp && ifevent->action == BRCMF_E_IF_DEL) { + bool armed = brcmf_cfg80211_vif_event_armed(drvr->config); + + /* Default handling in case no-one waits for this event */ + if (!armed) + brcmf_remove_interface(ifp, false); + } } /** diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c index f6241fd..66f942f 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/p2p.c @@ -2288,8 +2288,7 @@ int brcmf_p2p_del_vif(struct wiphy *wiphy, struct wireless_dev *wdev) else err = 0; } - if (err) - brcmf_remove_interface(vif->ifp, true); + brcmf_remove_interface(vif->ifp, true); brcmf_cfg80211_arm_vif_event(cfg, NULL); if (vif->wdev.iftype != NL80211_IFTYPE_P2P_DEVICE) -- cgit v0.10.2 From dba8fbc67ecd193590a72e960e9f5a6578b8bae6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Wed, 29 Jun 2016 21:54:27 +0200 Subject: brcmfmac: support removing AP interfaces with "interface_remove" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit New firmwares (e.g. 10.10.69.36 for BCM4366) support "interface_remove" for removing interfaces. Try to use this method on cfg80211 request. In case of older firmwares (e.g. 7.35.177.56 for BCM43602 as I tested) this will just result in firmware rejecting command and this won't change any behavior. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index a5db953..6e6066a 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -785,12 +785,48 @@ s32 brcmf_notify_escan_complete(struct brcmf_cfg80211_info *cfg, return err; } +static int brcmf_cfg80211_del_ap_iface(struct wiphy *wiphy, + struct wireless_dev *wdev) +{ + struct brcmf_cfg80211_info *cfg = wiphy_priv(wiphy); + struct net_device *ndev = wdev->netdev; + struct brcmf_if *ifp = netdev_priv(ndev); + int ret; + int err; + + brcmf_cfg80211_arm_vif_event(cfg, ifp->vif); + + err = brcmf_fil_bsscfg_data_set(ifp, "interface_remove", NULL, 0); + if (err) { + brcmf_err("interface_remove failed %d\n", err); + goto err_unarm; + } + + /* wait for firmware event */ + ret = brcmf_cfg80211_wait_vif_event(cfg, BRCMF_E_IF_DEL, + BRCMF_VIF_EVENT_TIMEOUT); + if (!ret) { + brcmf_err("timeout occurred\n"); + err = -EIO; + goto err_unarm; + } + + brcmf_remove_interface(ifp, true); + +err_unarm: + brcmf_cfg80211_arm_vif_event(cfg, NULL); + return err; +} + static int brcmf_cfg80211_del_iface(struct wiphy *wiphy, struct wireless_dev *wdev) { struct brcmf_cfg80211_info *cfg = wiphy_priv(wiphy); struct net_device *ndev = wdev->netdev; + if (ndev && ndev == cfg_to_ndev(cfg)) + return -ENOTSUPP; + /* vif event pending in firmware */ if (brcmf_cfg80211_vif_event_armed(cfg)) return -EBUSY; @@ -807,12 +843,13 @@ int brcmf_cfg80211_del_iface(struct wiphy *wiphy, struct wireless_dev *wdev) switch (wdev->iftype) { case NL80211_IFTYPE_ADHOC: case NL80211_IFTYPE_STATION: - case NL80211_IFTYPE_AP: case NL80211_IFTYPE_AP_VLAN: case NL80211_IFTYPE_WDS: case NL80211_IFTYPE_MONITOR: case NL80211_IFTYPE_MESH_POINT: return -EOPNOTSUPP; + case NL80211_IFTYPE_AP: + return brcmf_cfg80211_del_ap_iface(wiphy, wdev); case NL80211_IFTYPE_P2P_CLIENT: case NL80211_IFTYPE_P2P_GO: case NL80211_IFTYPE_P2P_DEVICE: -- cgit v0.10.2 From c18b104dd2495da60ec92f40e14559591644cc3a Mon Sep 17 00:00:00 2001 From: Prasun Maiti Date: Thu, 30 Jun 2016 14:01:23 +0530 Subject: mwifiex: Fix endianness for event TLV type TLV_BTCOEX_WL_SCANTIME The two members min_scan_time and max_scan_time of structure "mwifiex_ie_types_btcoex_scan_time" are of two bytes each. The values are assigned directtly from firmware without endian conversion handling. So, wrong datas will get saved in big-endian systems. This patch converts the values into cpu's byte order before assigning them into the local members. Signed-off-by: Prasun Maiti Acked-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/fw.h b/drivers/net/wireless/marvell/mwifiex/fw.h index cef7234..5596b6b 100644 --- a/drivers/net/wireless/marvell/mwifiex/fw.h +++ b/drivers/net/wireless/marvell/mwifiex/fw.h @@ -1961,8 +1961,8 @@ struct mwifiex_ie_types_btcoex_scan_time { struct mwifiex_ie_types_header header; u8 coex_scan; u8 reserved; - u16 min_scan_time; - u16 max_scan_time; + __le16 min_scan_time; + __le16 max_scan_time; } __packed; struct mwifiex_ie_types_btcoex_aggr_win_size { diff --git a/drivers/net/wireless/marvell/mwifiex/sta_event.c b/drivers/net/wireless/marvell/mwifiex/sta_event.c index 7c01778..a422f33 100644 --- a/drivers/net/wireless/marvell/mwifiex/sta_event.c +++ b/drivers/net/wireless/marvell/mwifiex/sta_event.c @@ -474,8 +474,8 @@ void mwifiex_bt_coex_wlan_param_update_event(struct mwifiex_private *priv, scantlv = (struct mwifiex_ie_types_btcoex_scan_time *)tlv; adapter->coex_scan = scantlv->coex_scan; - adapter->coex_min_scan_time = scantlv->min_scan_time; - adapter->coex_max_scan_time = scantlv->max_scan_time; + adapter->coex_min_scan_time = le16_to_cpu(scantlv->min_scan_time); + adapter->coex_max_scan_time = le16_to_cpu(scantlv->max_scan_time); break; default: -- cgit v0.10.2 From 473dfbfa09934cea0d08cc9023c749a5fce10cb0 Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Thu, 30 Jun 2016 22:24:57 +0530 Subject: mwifiex: Change default firmware for PCIe8997 chipset PCIe-USB8997 variant is being used in the product. Let's change default firmware from PCIe-UART to PCIe-USB. So by default PCIe-USB firmware would be downloaded if version register doesn't give any information. Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c index 1b1e266..012733c 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.c +++ b/drivers/net/wireless/marvell/mwifiex/pcie.c @@ -2841,20 +2841,20 @@ static void mwifiex_pcie_get_fw_name(struct mwifiex_adapter *adapter) version &= 0x7; switch (revision_id) { case PCIE8997_V2: - if (version == CHIP_VER_PCIEUSB) + if (version == CHIP_VER_PCIEUART) strcpy(adapter->fw_name, - PCIEUSB8997_FW_NAME_V2); + PCIEUART8997_FW_NAME_V2); else strcpy(adapter->fw_name, - PCIEUART8997_FW_NAME_V2); + PCIEUSB8997_FW_NAME_V2); break; case PCIE8997_Z: - if (version == CHIP_VER_PCIEUSB) + if (version == CHIP_VER_PCIEUART) strcpy(adapter->fw_name, - PCIEUSB8997_FW_NAME_Z); + PCIEUART8997_FW_NAME_Z); else strcpy(adapter->fw_name, - PCIEUART8997_FW_NAME_Z); + PCIEUSB8997_FW_NAME_Z); break; default: strcpy(adapter->fw_name, PCIE8997_DEFAULT_FW_NAME); diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.h b/drivers/net/wireless/marvell/mwifiex/pcie.h index 9c00c7e..f05061c 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.h +++ b/drivers/net/wireless/marvell/mwifiex/pcie.h @@ -32,7 +32,7 @@ #define PCIE8897_DEFAULT_FW_NAME "mrvl/pcie8897_uapsta.bin" #define PCIE8897_A0_FW_NAME "mrvl/pcie8897_uapsta_a0.bin" #define PCIE8897_B0_FW_NAME "mrvl/pcie8897_uapsta.bin" -#define PCIE8997_DEFAULT_FW_NAME "mrvl/pcieuart8997_combo_v2.bin" +#define PCIE8997_DEFAULT_FW_NAME "mrvl/pcieusb8997_combo_v2.bin" #define PCIEUART8997_FW_NAME_Z "mrvl/pcieuart8997_combo.bin" #define PCIEUART8997_FW_NAME_V2 "mrvl/pcieuart8997_combo_v2.bin" #define PCIEUSB8997_FW_NAME_Z "mrvl/pcieusb8997_combo.bin" @@ -48,7 +48,7 @@ #define PCIE8897_B0 0x1200 #define PCIE8997_Z 0x0 #define PCIE8997_V2 0x471 -#define CHIP_VER_PCIEUSB 0x2 +#define CHIP_VER_PCIEUART 0x3 /* Constants for Buffer Descriptor (BD) rings */ #define MWIFIEX_MAX_TXRX_BD 0x20 -- cgit v0.10.2 From d41376ca8ba74e954ba931c69271d0b29546a202 Mon Sep 17 00:00:00 2001 From: Brian Norris Date: Thu, 30 Jun 2016 15:21:02 -0700 Subject: mwifiex: mask PCIe interrupts before removal The PCIe driver didn't mask the host interrupts before trying to tear down. This causes lockups at reboot or rmmod when using MSI-X on 8997, since the MSI handler gets confused and locks up the system. Also tested on 8897, which does not support MSI-X (and wasn't experiencing this same bug). No regressions seen there. Signed-off-by: Brian Norris Tested-by: Douglas Anderson Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c index 012733c..c7f5df8 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.c +++ b/drivers/net/wireless/marvell/mwifiex/pcie.c @@ -440,6 +440,11 @@ static int mwifiex_pcie_disable_host_int(struct mwifiex_adapter *adapter) return 0; } +static void mwifiex_pcie_disable_host_int_noerr(struct mwifiex_adapter *adapter) +{ + WARN_ON(mwifiex_pcie_disable_host_int(adapter)); +} + /* * This function enables the host interrupt. * @@ -2946,6 +2951,7 @@ static struct mwifiex_if_ops pcie_ops = { .register_dev = mwifiex_register_dev, .unregister_dev = mwifiex_unregister_dev, .enable_int = mwifiex_pcie_enable_host_int, + .disable_int = mwifiex_pcie_disable_host_int_noerr, .process_int_status = mwifiex_process_int_status, .host_to_card = mwifiex_pcie_host_to_card, .wakeup = mwifiex_pm_wakeup_card, -- cgit v0.10.2 From 5781fc29dbbd3ee5e11c1bf4fa6696ae89d19840 Mon Sep 17 00:00:00 2001 From: Shengzhen Li Date: Fri, 1 Jul 2016 18:26:52 +0530 Subject: mwifiex: fix interrupt processing corner case in MSI mode As interrupt is read in interrupt handler as well as interrupt processing thread, we observed a corner case issue for MSI in which interrupt gets processed twice. This patch moves interrupt reading code for MSI mode from mwifiex_interrupt_status() to mwifiex_pcie_process_int() to avoid the issue. Signed-off-by: Shengzhen Li Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c index c7f5df8..22fe993 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.c +++ b/drivers/net/wireless/marvell/mwifiex/pcie.c @@ -2091,6 +2091,13 @@ static void mwifiex_interrupt_status(struct mwifiex_adapter *adapter, unsigned long flags; struct pcie_service_card *card = adapter->card; + if (card->msi_enable) { + spin_lock_irqsave(&adapter->int_lock, flags); + adapter->int_status = 1; + spin_unlock_irqrestore(&adapter->int_lock, flags); + return; + } + if (!mwifiex_pcie_ok_to_access_hw(adapter)) return; @@ -2192,15 +2199,44 @@ exit: static int mwifiex_process_pcie_int(struct mwifiex_adapter *adapter) { int ret; - u32 pcie_ireg; + u32 pcie_ireg = 0; unsigned long flags; + struct pcie_service_card *card = adapter->card; spin_lock_irqsave(&adapter->int_lock, flags); - /* Clear out unused interrupts */ - pcie_ireg = adapter->int_status; + if (!card->msi_enable) { + /* Clear out unused interrupts */ + pcie_ireg = adapter->int_status; + } adapter->int_status = 0; spin_unlock_irqrestore(&adapter->int_lock, flags); + if (card->msi_enable) { + if (mwifiex_pcie_ok_to_access_hw(adapter)) { + if (mwifiex_read_reg(adapter, PCIE_HOST_INT_STATUS, + &pcie_ireg)) { + mwifiex_dbg(adapter, ERROR, + "Read register failed\n"); + return -1; + } + + if ((pcie_ireg != 0xFFFFFFFF) && (pcie_ireg)) { + if (mwifiex_write_reg(adapter, + PCIE_HOST_INT_STATUS, + ~pcie_ireg)) { + mwifiex_dbg(adapter, ERROR, + "Write register failed\n"); + return -1; + } + if (!adapter->pps_uapsd_mode && + adapter->ps_state == PS_STATE_SLEEP) { + adapter->ps_state = PS_STATE_AWAKE; + adapter->pm_wakeup_fw_try = false; + del_timer(&adapter->wakeup_timer); + } + } + } + } while (pcie_ireg & HOST_INTR_MASK) { if (pcie_ireg & HOST_INTR_DNLD_DONE) { pcie_ireg &= ~HOST_INTR_DNLD_DONE; @@ -2240,6 +2276,12 @@ static int mwifiex_process_pcie_int(struct mwifiex_adapter *adapter) return ret; } + if (card->msi_enable) { + spin_lock_irqsave(&adapter->int_lock, flags); + adapter->int_status = 0; + spin_unlock_irqrestore(&adapter->int_lock, flags); + } + if (mwifiex_pcie_ok_to_access_hw(adapter)) { if (mwifiex_read_reg(adapter, PCIE_HOST_INT_STATUS, &pcie_ireg)) { @@ -2263,7 +2305,7 @@ static int mwifiex_process_pcie_int(struct mwifiex_adapter *adapter) mwifiex_dbg(adapter, INTR, "info: cmd_sent=%d data_sent=%d\n", adapter->cmd_sent, adapter->data_sent); - if (adapter->ps_state != PS_STATE_SLEEP) + if (!card->msi_enable && adapter->ps_state != PS_STATE_SLEEP) mwifiex_pcie_enable_host_int(adapter); return 0; -- cgit v0.10.2 From edb45b67a09d1be4b7b8a33922a0f3934f3e23fa Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 5 Jul 2016 10:08:06 -0500 Subject: rtlwifi: Create common routine to get hardware info All of the rtlwifi family of drivers have a similar routine that acquires the hardware info from efuse and initializes a number of variables in the driver's private area. A common routine is created for all drivers to use. Reported-by: Arnd Bergmann Signed-off-by: Larry Finger Cc: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/efuse.c b/drivers/net/wireless/realtek/rtlwifi/efuse.c index 0b4082c..7becfef 100644 --- a/drivers/net/wireless/realtek/rtlwifi/efuse.c +++ b/drivers/net/wireless/realtek/rtlwifi/efuse.c @@ -24,6 +24,7 @@ *****************************************************************************/ #include "wifi.h" #include "efuse.h" +#include "pci.h" #include static const u8 MAX_PGPKT_SIZE = 9; @@ -1243,3 +1244,80 @@ static u8 efuse_calculate_word_cnts(u8 word_en) return word_cnts; } +int rtl_get_hwinfo(struct ieee80211_hw *hw, struct rtl_priv *rtlpriv, + int max_size, u8 *hwinfo, int *params) +{ + struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); + struct rtl_pci_priv *rtlpcipriv = rtl_pcipriv(hw); + struct device *dev = &rtlpcipriv->dev.pdev->dev; + u16 eeprom_id; + u16 i, usvalue; + + switch (rtlefuse->epromtype) { + case EEPROM_BOOT_EFUSE: + rtl_efuse_shadow_map_update(hw); + break; + + case EEPROM_93C46: + RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, + "RTL8XXX did not boot from eeprom, check it !!\n"); + return 1; + + default: + dev_warn(dev, "no efuse data\n"); + return 1; + } + + memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], max_size); + + RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP", + hwinfo, max_size); + + eeprom_id = *((u16 *)&hwinfo[0]); + if (eeprom_id != params[0]) { + RT_TRACE(rtlpriv, COMP_ERR, DBG_WARNING, + "EEPROM ID(%#x) is invalid!!\n", eeprom_id); + rtlefuse->autoload_failflag = true; + } else { + RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "Autoload OK\n"); + rtlefuse->autoload_failflag = false; + } + + if (rtlefuse->autoload_failflag) + return 1; + + rtlefuse->eeprom_vid = *(u16 *)&hwinfo[params[1]]; + rtlefuse->eeprom_did = *(u16 *)&hwinfo[params[2]]; + rtlefuse->eeprom_svid = *(u16 *)&hwinfo[params[3]]; + rtlefuse->eeprom_smid = *(u16 *)&hwinfo[params[4]]; + RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, + "EEPROMId = 0x%4x\n", eeprom_id); + RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, + "EEPROM VID = 0x%4x\n", rtlefuse->eeprom_vid); + RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, + "EEPROM DID = 0x%4x\n", rtlefuse->eeprom_did); + RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, + "EEPROM SVID = 0x%4x\n", rtlefuse->eeprom_svid); + RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, + "EEPROM SMID = 0x%4x\n", rtlefuse->eeprom_smid); + + for (i = 0; i < 6; i += 2) { + usvalue = *(u16 *)&hwinfo[params[5] + i]; + *((u16 *)(&rtlefuse->dev_addr[i])) = usvalue; + } + RT_TRACE(rtlpriv, COMP_INIT, DBG_DMESG, "%pM\n", rtlefuse->dev_addr); + + rtlefuse->eeprom_channelplan = *&hwinfo[params[6]]; + rtlefuse->eeprom_version = *(u16 *)&hwinfo[params[7]]; + rtlefuse->txpwr_fromeprom = true; + rtlefuse->eeprom_oemid = *&hwinfo[params[8]]; + + RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, + "EEPROM Customer ID: 0x%2x\n", rtlefuse->eeprom_oemid); + + /* set channel plan to world wide 13 */ + rtlefuse->channel_plan = params[9]; + + return 0; +} +EXPORT_SYMBOL_GPL(rtl_get_hwinfo); diff --git a/drivers/net/wireless/realtek/rtlwifi/efuse.h b/drivers/net/wireless/realtek/rtlwifi/efuse.h index be02e78..51aa121 100644 --- a/drivers/net/wireless/realtek/rtlwifi/efuse.h +++ b/drivers/net/wireless/realtek/rtlwifi/efuse.h @@ -109,5 +109,7 @@ bool efuse_shadow_update_chk(struct ieee80211_hw *hw); void rtl_efuse_shadow_map_update(struct ieee80211_hw *hw); void efuse_force_write_vendor_Id(struct ieee80211_hw *hw); void efuse_re_pg_section(struct ieee80211_hw *hw, u8 section_idx); +int rtl_get_hwinfo(struct ieee80211_hw *hw, struct rtl_priv *rtlpriv, + int max_size, u8 *hwinfo, int *params); #endif -- cgit v0.10.2 From df5cbc697d9ba8c7304cc6e70ce36658eeef38e7 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 5 Jul 2016 10:08:07 -0500 Subject: rtlwifi: rtl8192ce: Convert driver to use common hardware info routine The driver for RTL8192CE chips is converted to use the common routine for getting the hardware information. Reported-by: Arnd Bergmann Signed-off-by: Larry Finger Cc: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c index 809a021..2446079 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ce/hw.c @@ -1680,65 +1680,18 @@ static void _rtl92ce_read_adapter_info(struct ieee80211_hw *hw) struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); - struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; - u16 i, usvalue; - u8 hwinfo[HWSET_MAX_SIZE]; - u16 eeprom_id; - - switch (rtlefuse->epromtype) { - case EEPROM_BOOT_EFUSE: - rtl_efuse_shadow_map_update(hw); - break; - - case EEPROM_93C46: - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!\n"); - return; - - default: - dev_warn(dev, "no efuse data\n"); - return; - } - - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); - - RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP", - hwinfo, HWSET_MAX_SIZE); - - eeprom_id = *((u16 *)&hwinfo[0]); - if (eeprom_id != RTL8190_EEPROM_ID) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_WARNING, - "EEPROM ID(%#x) is invalid!!\n", eeprom_id); - rtlefuse->autoload_failflag = true; - } else { - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "Autoload OK\n"); - rtlefuse->autoload_failflag = false; - } - - if (rtlefuse->autoload_failflag) + int params[] = {RTL8190_EEPROM_ID, EEPROM_VID, EEPROM_DID, + EEPROM_SVID, EEPROM_SMID, EEPROM_MAC_ADDR, + EEPROM_CHANNELPLAN, EEPROM_VERSION, EEPROM_CUSTOMER_ID, + COUNTRY_CODE_WORLD_WIDE_13}; + u8 *hwinfo; + + hwinfo = kzalloc(HWSET_MAX_SIZE, GFP_KERNEL); + if (!hwinfo) return; - rtlefuse->eeprom_vid = *(u16 *)&hwinfo[EEPROM_VID]; - rtlefuse->eeprom_did = *(u16 *)&hwinfo[EEPROM_DID]; - rtlefuse->eeprom_svid = *(u16 *)&hwinfo[EEPROM_SVID]; - rtlefuse->eeprom_smid = *(u16 *)&hwinfo[EEPROM_SMID]; - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROMId = 0x%4x\n", eeprom_id); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM VID = 0x%4x\n", rtlefuse->eeprom_vid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM DID = 0x%4x\n", rtlefuse->eeprom_did); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SVID = 0x%4x\n", rtlefuse->eeprom_svid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SMID = 0x%4x\n", rtlefuse->eeprom_smid); - - for (i = 0; i < 6; i += 2) { - usvalue = *(u16 *)&hwinfo[EEPROM_MAC_ADDR + i]; - *((u16 *) (&rtlefuse->dev_addr[i])) = usvalue; - } - - RT_TRACE(rtlpriv, COMP_INIT, DBG_DMESG, "%pM\n", rtlefuse->dev_addr); + if (rtl_get_hwinfo(hw, rtlpriv, HWSET_MAX_SIZE, hwinfo, params)) + goto exit; _rtl92ce_read_txpower_info_from_hwpg(hw, rtlefuse->autoload_failflag, @@ -1747,18 +1700,6 @@ static void _rtl92ce_read_adapter_info(struct ieee80211_hw *hw) rtl8192ce_read_bt_coexist_info_from_hwpg(hw, rtlefuse->autoload_failflag, hwinfo); - - rtlefuse->eeprom_channelplan = *&hwinfo[EEPROM_CHANNELPLAN]; - rtlefuse->eeprom_version = *(u16 *)&hwinfo[EEPROM_VERSION]; - rtlefuse->txpwr_fromeprom = true; - rtlefuse->eeprom_oemid = *&hwinfo[EEPROM_CUSTOMER_ID]; - - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM Customer ID: 0x%2x\n", rtlefuse->eeprom_oemid); - - /* set channel paln to world wide 13 */ - rtlefuse->channel_plan = COUNTRY_CODE_WORLD_WIDE_13; - if (rtlhal->oem_id == RT_CID_DEFAULT) { switch (rtlefuse->eeprom_oemid) { case EEPROM_CID_DEFAULT: @@ -1782,10 +1723,10 @@ static void _rtl92ce_read_adapter_info(struct ieee80211_hw *hw) default: rtlhal->oem_id = RT_CID_DEFAULT; break; - } } - +exit: + kfree(hwinfo); } static void _rtl92ce_hal_customized_behavior(struct ieee80211_hw *hw) -- cgit v0.10.2 From c2d9a411456c7a2c5700ace8f91134ce6081aeb6 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 5 Jul 2016 10:08:08 -0500 Subject: rtlwifi: rtl8192cu: Convert driver to use common hardware info routine The driver for RTL8192CU chips is converted to use the common routine for getting the hardware information. Reported-by: Arnd Bergmann Signed-off-by: Larry Finger Cc: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c index ae1129f..8789752 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192cu/hw.c @@ -347,56 +347,24 @@ static void _rtl92cu_read_adapter_info(struct ieee80211_hw *hw) struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); - u16 i, usvalue; - u8 hwinfo[HWSET_MAX_SIZE] = {0}; - u16 eeprom_id; - - switch (rtlefuse->epromtype) { - case EEPROM_BOOT_EFUSE: - rtl_efuse_shadow_map_update(hw); - break; - - case EEPROM_93C46: - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!\n"); + int params[] = {RTL8190_EEPROM_ID, EEPROM_VID, EEPROM_DID, + EEPROM_SVID, EEPROM_SMID, EEPROM_MAC_ADDR, + EEPROM_CHANNELPLAN, EEPROM_VERSION, EEPROM_CUSTOMER_ID, + 0}; + u8 *hwinfo; + + hwinfo = kzalloc(HWSET_MAX_SIZE, GFP_KERNEL); + if (!hwinfo) return; - default: - pr_warn("rtl92cu: no efuse data\n\n"); - return; - } - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); - RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_LOUD, "MAP", - hwinfo, HWSET_MAX_SIZE); - eeprom_id = le16_to_cpu(*((__le16 *)&hwinfo[0])); - if (eeprom_id != RTL8190_EEPROM_ID) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "EEPROM ID(%#x) is invalid!!\n", eeprom_id); - rtlefuse->autoload_failflag = true; - } else { - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "Autoload OK\n"); - rtlefuse->autoload_failflag = false; - } - if (rtlefuse->autoload_failflag) - return; - for (i = 0; i < 6; i += 2) { - usvalue = *(u16 *)&hwinfo[EEPROM_MAC_ADDR + i]; - *((u16 *) (&rtlefuse->dev_addr[i])) = usvalue; - } - pr_info("MAC address: %pM\n", rtlefuse->dev_addr); + if (rtl_get_hwinfo(hw, rtlpriv, HWSET_MAX_SIZE, hwinfo, params)) + goto exit; + _rtl92cu_read_txpower_info_from_hwpg(hw, rtlefuse->autoload_failflag, hwinfo); - rtlefuse->eeprom_vid = le16_to_cpu(*(__le16 *)&hwinfo[EEPROM_VID]); - rtlefuse->eeprom_did = le16_to_cpu(*(__le16 *)&hwinfo[EEPROM_DID]); - RT_TRACE(rtlpriv, COMP_INIT, DBG_DMESG, " VID = 0x%02x PID = 0x%02x\n", - rtlefuse->eeprom_vid, rtlefuse->eeprom_did); - rtlefuse->eeprom_channelplan = hwinfo[EEPROM_CHANNELPLAN]; - rtlefuse->eeprom_version = - le16_to_cpu(*(__le16 *)&hwinfo[EEPROM_VERSION]); + _rtl92cu_read_board_type(hw, hwinfo); + rtlefuse->txpwr_fromeprom = true; - rtlefuse->eeprom_oemid = hwinfo[EEPROM_CUSTOMER_ID]; - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "EEPROM Customer ID: 0x%2x\n", - rtlefuse->eeprom_oemid); if (rtlhal->oem_id == RT_CID_DEFAULT) { switch (rtlefuse->eeprom_oemid) { case EEPROM_CID_DEFAULT: @@ -422,7 +390,8 @@ static void _rtl92cu_read_adapter_info(struct ieee80211_hw *hw) break; } } - _rtl92cu_read_board_type(hw, hwinfo); +exit: + kfree(hwinfo); } static void _rtl92cu_hal_customized_behavior(struct ieee80211_hw *hw) -- cgit v0.10.2 From 9468792743a53bf5dd5180b23884cdd18ab1ceed Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 5 Jul 2016 10:08:09 -0500 Subject: rtlwifi: rtl8188ee: Convert driver to use common hardware info routine The driver for RTL8188EE chips is converted to use the common routine for getting the hardware information. Reported-by: Arnd Bergmann Signed-off-by: Larry Finger Cc: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c index cfdf6d8..4ab6201 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/hw.c @@ -1835,76 +1835,24 @@ static void _rtl88ee_read_adapter_info(struct ieee80211_hw *hw) struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); - u16 i, usvalue; - u8 hwinfo[HWSET_MAX_SIZE]; - u16 eeprom_id; - - switch (rtlefuse->epromtype) { - case EEPROM_BOOT_EFUSE: - rtl_efuse_shadow_map_update(hw); - break; - - case EEPROM_93C46: - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!\n"); + int params[] = {RTL8188E_EEPROM_ID, EEPROM_VID, EEPROM_DID, + EEPROM_SVID, EEPROM_SMID, EEPROM_MAC_ADDR, + EEPROM_CHANNELPLAN, EEPROM_VERSION, EEPROM_CUSTOMER_ID, + COUNTRY_CODE_WORLD_WIDE_13}; + u8 *hwinfo; + + hwinfo = kzalloc(HWSET_MAX_SIZE, GFP_KERNEL); + if (!hwinfo) return; - default: - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "boot from neither eeprom nor efuse, check it !!\n"); - return; - } - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); - - RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP\n", - hwinfo, HWSET_MAX_SIZE); - - eeprom_id = *((u16 *)&hwinfo[0]); - if (eeprom_id != RTL8188E_EEPROM_ID) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_WARNING, - "EEPROM ID(%#x) is invalid!!\n", eeprom_id); - rtlefuse->autoload_failflag = true; - } else { - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "Autoload OK\n"); - rtlefuse->autoload_failflag = false; - } + if (rtl_get_hwinfo(hw, rtlpriv, HWSET_MAX_SIZE, hwinfo, params)) + goto exit; - if (rtlefuse->autoload_failflag == true) - return; - /*VID DID SVID SDID*/ - rtlefuse->eeprom_vid = *(u16 *)&hwinfo[EEPROM_VID]; - rtlefuse->eeprom_did = *(u16 *)&hwinfo[EEPROM_DID]; - rtlefuse->eeprom_svid = *(u16 *)&hwinfo[EEPROM_SVID]; - rtlefuse->eeprom_smid = *(u16 *)&hwinfo[EEPROM_SMID]; - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROMId = 0x%4x\n", eeprom_id); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM VID = 0x%4x\n", rtlefuse->eeprom_vid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM DID = 0x%4x\n", rtlefuse->eeprom_did); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SVID = 0x%4x\n", rtlefuse->eeprom_svid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SMID = 0x%4x\n", rtlefuse->eeprom_smid); - /*customer ID*/ - rtlefuse->eeprom_oemid = hwinfo[EEPROM_CUSTOMER_ID]; if (rtlefuse->eeprom_oemid == 0xFF) - rtlefuse->eeprom_oemid = 0; + rtlefuse->eeprom_oemid = 0; RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "EEPROM Customer ID: 0x%2x\n", rtlefuse->eeprom_oemid); - /*EEPROM version*/ - rtlefuse->eeprom_version = *(u16 *)&hwinfo[EEPROM_VERSION]; - /*mac address*/ - for (i = 0; i < 6; i += 2) { - usvalue = *(u16 *)&hwinfo[EEPROM_MAC_ADDR + i]; - *((u16 *)(&rtlefuse->dev_addr[i])) = usvalue; - } - - RT_TRACE(rtlpriv, COMP_INIT, DBG_DMESG, - "dev_addr: %pM\n", rtlefuse->dev_addr); - /*channel plan */ - rtlefuse->eeprom_channelplan = hwinfo[EEPROM_CHANNELPLAN]; /* set channel plan from efuse */ rtlefuse->channel_plan = rtlefuse->eeprom_channelplan; /*tx power*/ @@ -1978,6 +1926,8 @@ static void _rtl88ee_read_adapter_info(struct ieee80211_hw *hw) } } +exit: + kfree(hwinfo); } static void _rtl88ee_hal_customized_behavior(struct ieee80211_hw *hw) -- cgit v0.10.2 From 5c392654ee81d51922704a7350ba3f0781fc103b Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 5 Jul 2016 10:08:10 -0500 Subject: rtlwifi: rtl8192ee: Convert driver to use common hardware info routine The driver for RTL8192EE chips is converted to use the common routine for getting the hardware information. Reported-by: Arnd Bergmann Signed-off-by: Larry Finger Cc: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c index 9d1ecba..b07af8d 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192ee/hw.c @@ -2098,75 +2098,24 @@ static void _rtl92ee_read_adapter_info(struct ieee80211_hw *hw) struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); - u16 i, usvalue; - u8 hwinfo[HWSET_MAX_SIZE]; - u16 eeprom_id; - - switch (rtlefuse->epromtype) { - case EEPROM_BOOT_EFUSE: - rtl_efuse_shadow_map_update(hw); - break; - - case EEPROM_93C46: - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!\n"); - return; - - default: - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "boot from neither eeprom nor efuse, check it !!\n"); + int params[] = {RTL8192E_EEPROM_ID, EEPROM_VID, EEPROM_DID, + EEPROM_SVID, EEPROM_SMID, EEPROM_MAC_ADDR, + EEPROM_CHANNELPLAN, EEPROM_VERSION, EEPROM_CUSTOMER_ID, + COUNTRY_CODE_WORLD_WIDE_13}; + u8 *hwinfo; + + hwinfo = kzalloc(HWSET_MAX_SIZE, GFP_KERNEL); + if (!hwinfo) return; - } - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); - RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP\n", - hwinfo, HWSET_MAX_SIZE); + if (rtl_get_hwinfo(hw, rtlpriv, HWSET_MAX_SIZE, hwinfo, params)) + goto exit; - eeprom_id = *((u16 *)&hwinfo[0]); - if (eeprom_id != RTL8192E_EEPROM_ID) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_WARNING, - "EEPROM ID(%#x) is invalid!!\n", eeprom_id); - rtlefuse->autoload_failflag = true; - } else { - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "Autoload OK\n"); - rtlefuse->autoload_failflag = false; - } - - if (rtlefuse->autoload_failflag) - return; - /*VID DID SVID SDID*/ - rtlefuse->eeprom_vid = *(u16 *)&hwinfo[EEPROM_VID]; - rtlefuse->eeprom_did = *(u16 *)&hwinfo[EEPROM_DID]; - rtlefuse->eeprom_svid = *(u16 *)&hwinfo[EEPROM_SVID]; - rtlefuse->eeprom_smid = *(u16 *)&hwinfo[EEPROM_SMID]; - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "EEPROMId = 0x%4x\n", eeprom_id); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM VID = 0x%4x\n", rtlefuse->eeprom_vid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM DID = 0x%4x\n", rtlefuse->eeprom_did); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SVID = 0x%4x\n", rtlefuse->eeprom_svid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SMID = 0x%4x\n", rtlefuse->eeprom_smid); - /*customer ID*/ - rtlefuse->eeprom_oemid = *(u8 *)&hwinfo[EEPROM_CUSTOMER_ID]; if (rtlefuse->eeprom_oemid == 0xFF) rtlefuse->eeprom_oemid = 0; RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "EEPROM Customer ID: 0x%2x\n", rtlefuse->eeprom_oemid); - /*EEPROM version*/ - rtlefuse->eeprom_version = *(u8 *)&hwinfo[EEPROM_VERSION]; - /*mac address*/ - for (i = 0; i < 6; i += 2) { - usvalue = *(u16 *)&hwinfo[EEPROM_MAC_ADDR + i]; - *((u16 *)(&rtlefuse->dev_addr[i])) = usvalue; - } - - RT_TRACE(rtlpriv, COMP_INIT, DBG_DMESG, - "dev_addr: %pM\n", rtlefuse->dev_addr); - /*channel plan */ - rtlefuse->eeprom_channelplan = *(u8 *)&hwinfo[EEPROM_CHANNELPLAN]; /* set channel plan from efuse */ rtlefuse->channel_plan = rtlefuse->eeprom_channelplan; /*tx power*/ @@ -2208,6 +2157,8 @@ static void _rtl92ee_read_adapter_info(struct ieee80211_hw *hw) break; } } +exit: + kfree(hwinfo); } static void _rtl92ee_hal_customized_behavior(struct ieee80211_hw *hw) -- cgit v0.10.2 From 8aaf6916de684b1013e7b1941e98982ce41cac30 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 5 Jul 2016 10:08:11 -0500 Subject: rtlwifi: rtl8723ae: Convert driver to use common hardware info routine The driver for RTL8723AE chips is converted to use the common routine for getting the hardware information. Reported-by: Arnd Bergmann Signed-off-by: Larry Finger Cc: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c index 0025e21..662c445 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c @@ -1630,67 +1630,22 @@ static void _rtl8723e_read_adapter_info(struct ieee80211_hw *hw, struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); - struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; - u16 i, usvalue; - u8 hwinfo[HWSET_MAX_SIZE]; - u16 eeprom_id; + int params[] = {RTL8190_EEPROM_ID, EEPROM_VID, EEPROM_DID, + EEPROM_SVID, EEPROM_SMID, EEPROM_MAC_ADDR, + EEPROM_CHANNELPLAN, EEPROM_VERSION, EEPROM_CUSTOMER_ID, + COUNTRY_CODE_WORLD_WIDE_13}; + u8 *hwinfo; if (b_pseudo_test) { /* need add */ return; } - switch (rtlefuse->epromtype) { - case EEPROM_BOOT_EFUSE: - rtl_efuse_shadow_map_update(hw); - - case EEPROM_93C46: - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!\n"); + hwinfo = kzalloc(HWSET_MAX_SIZE, GFP_KERNEL); + if (!hwinfo) return; - default: - dev_warn(dev, "no efuse data\n"); - } - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); - - RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP\n", - hwinfo, HWSET_MAX_SIZE); - - eeprom_id = *((u16 *)&hwinfo[0]); - if (eeprom_id != RTL8190_EEPROM_ID) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_WARNING, - "EEPROM ID(%#x) is invalid!!\n", eeprom_id); - rtlefuse->autoload_failflag = true; - } else { - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "Autoload OK\n"); - rtlefuse->autoload_failflag = false; - } - - if (rtlefuse->autoload_failflag) - return; - - rtlefuse->eeprom_vid = *(u16 *)&hwinfo[EEPROM_VID]; - rtlefuse->eeprom_did = *(u16 *)&hwinfo[EEPROM_DID]; - rtlefuse->eeprom_svid = *(u16 *)&hwinfo[EEPROM_SVID]; - rtlefuse->eeprom_smid = *(u16 *)&hwinfo[EEPROM_SMID]; - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROMId = 0x%4x\n", eeprom_id); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM VID = 0x%4x\n", rtlefuse->eeprom_vid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM DID = 0x%4x\n", rtlefuse->eeprom_did); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SVID = 0x%4x\n", rtlefuse->eeprom_svid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SMID = 0x%4x\n", rtlefuse->eeprom_smid); - - for (i = 0; i < 6; i += 2) { - usvalue = *(u16 *)&hwinfo[EEPROM_MAC_ADDR + i]; - *((u16 *)(&rtlefuse->dev_addr[i])) = usvalue; - } - - RT_TRACE(rtlpriv, COMP_INIT, DBG_DMESG, - "dev_addr: %pM\n", rtlefuse->dev_addr); + if (rtl_get_hwinfo(hw, rtlpriv, HWSET_MAX_SIZE, hwinfo, params)) + goto exit; _rtl8723e_read_txpower_info_from_hwpg(hw, rtlefuse->autoload_failflag, hwinfo); @@ -1698,17 +1653,6 @@ static void _rtl8723e_read_adapter_info(struct ieee80211_hw *hw, rtl8723e_read_bt_coexist_info_from_hwpg(hw, rtlefuse->autoload_failflag, hwinfo); - rtlefuse->eeprom_channelplan = hwinfo[EEPROM_CHANNELPLAN]; - rtlefuse->eeprom_version = *(u16 *)&hwinfo[EEPROM_VERSION]; - rtlefuse->txpwr_fromeprom = true; - rtlefuse->eeprom_oemid = hwinfo[EEPROM_CUSTOMER_ID]; - - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM Customer ID: 0x%2x\n", rtlefuse->eeprom_oemid); - - /* set channel paln to world wide 13 */ - rtlefuse->channel_plan = COUNTRY_CODE_WORLD_WIDE_13; - if (rtlhal->oem_id == RT_CID_DEFAULT) { switch (rtlefuse->eeprom_oemid) { case EEPROM_CID_DEFAULT: @@ -1836,6 +1780,8 @@ static void _rtl8723e_read_adapter_info(struct ieee80211_hw *hw, } } +exit: + kfree(hwinfo); } static void _rtl8723e_hal_customized_behavior(struct ieee80211_hw *hw) -- cgit v0.10.2 From 9e9c9c247c2126da66cef5e76818fcbc0d601f00 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 5 Jul 2016 10:08:12 -0500 Subject: rtlwifi: rtl8723be: Convert driver to use common hardware info routine The driver for RTL8723BE chips is converted to use the common routine for getting the hardware information. Reported-by: Arnd Bergmann Signed-off-by: Larry Finger Cc: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c index 1c3e105..82e4476 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723be/hw.c @@ -2026,10 +2026,12 @@ static void _rtl8723be_read_adapter_info(struct ieee80211_hw *hw, struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); - struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; - u16 i, usvalue; - u8 hwinfo[HWSET_MAX_SIZE]; - u16 eeprom_id; + int params[] = {RTL8723BE_EEPROM_ID, EEPROM_VID, EEPROM_DID, + EEPROM_SVID, EEPROM_SMID, EEPROM_MAC_ADDR, + EEPROM_CHANNELPLAN, EEPROM_VERSION, EEPROM_CUSTOMER_ID, + COUNTRY_CODE_WORLD_WIDE_13}; + u8 *hwinfo; + int i; bool is_toshiba_smid1 = false; bool is_toshiba_smid2 = false; bool is_samsung_smid = false; @@ -2057,58 +2059,12 @@ static void _rtl8723be_read_adapter_info(struct ieee80211_hw *hw, return; } - switch (rtlefuse->epromtype) { - case EEPROM_BOOT_EFUSE: - rtl_efuse_shadow_map_update(hw); - break; - - case EEPROM_93C46: - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!\n"); + hwinfo = kzalloc(HWSET_MAX_SIZE, GFP_KERNEL); + if (!hwinfo) return; - default: - dev_warn(dev, "no efuse data\n"); - return; - } - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); - RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, ("MAP\n"), - hwinfo, HWSET_MAX_SIZE); - - eeprom_id = *((u16 *)&hwinfo[0]); - if (eeprom_id != RTL8723BE_EEPROM_ID) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_WARNING, - "EEPROM ID(%#x) is invalid!!\n", eeprom_id); - rtlefuse->autoload_failflag = true; - } else { - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "Autoload OK\n"); - rtlefuse->autoload_failflag = false; - } - - if (rtlefuse->autoload_failflag) - return; - - rtlefuse->eeprom_vid = *(u16 *)&hwinfo[EEPROM_VID]; - rtlefuse->eeprom_did = *(u16 *)&hwinfo[EEPROM_DID]; - rtlefuse->eeprom_svid = *(u16 *)&hwinfo[EEPROM_SVID]; - rtlefuse->eeprom_smid = *(u16 *)&hwinfo[EEPROM_SMID]; - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROMId = 0x%4x\n", eeprom_id); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM VID = 0x%4x\n", rtlefuse->eeprom_vid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM DID = 0x%4x\n", rtlefuse->eeprom_did); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SVID = 0x%4x\n", rtlefuse->eeprom_svid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SMID = 0x%4x\n", rtlefuse->eeprom_smid); - - for (i = 0; i < 6; i += 2) { - usvalue = *(u16 *)&hwinfo[EEPROM_MAC_ADDR + i]; - *((u16 *)(&rtlefuse->dev_addr[i])) = usvalue; - } - RT_TRACE(rtlpriv, COMP_INIT, DBG_DMESG, "dev_addr: %pM\n", - rtlefuse->dev_addr); + if (rtl_get_hwinfo(hw, rtlpriv, HWSET_MAX_SIZE, hwinfo, params)) + goto exit; /*parse xtal*/ rtlefuse->crystalcap = hwinfo[EEPROM_XTAL_8723BE]; @@ -2122,14 +2078,6 @@ static void _rtl8723be_read_adapter_info(struct ieee80211_hw *hw, rtlefuse->autoload_failflag, hwinfo); - rtlefuse->eeprom_channelplan = hwinfo[EEPROM_CHANNELPLAN]; - rtlefuse->eeprom_version = *(u16 *)&hwinfo[EEPROM_VERSION]; - rtlefuse->txpwr_fromeprom = true; - rtlefuse->eeprom_oemid = hwinfo[EEPROM_CUSTOMER_ID]; - - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM Customer ID: 0x%2x\n", rtlefuse->eeprom_oemid); - /* set channel plan from efuse */ rtlefuse->channel_plan = rtlefuse->eeprom_channelplan; @@ -2240,6 +2188,8 @@ static void _rtl8723be_read_adapter_info(struct ieee80211_hw *hw, break; } } +exit: + kfree(hwinfo); } static void _rtl8723be_hal_customized_behavior(struct ieee80211_hw *hw) -- cgit v0.10.2 From 2f7b4b895365e5f6716f16ef6e1d513c1c4884f6 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 5 Jul 2016 10:08:13 -0500 Subject: rtlwifi: rtl8821ae: Convert driver to use common hardware info routine The driver for RTL8821AE chips is converted to use the common routine for getting the hardware information. Reported-by: Arnd Bergmann Signed-off-by: Larry Finger Cc: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c index ac8836b..0cddf1a 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8821ae/hw.c @@ -3101,86 +3101,22 @@ static void _rtl8821ae_read_adapter_info(struct ieee80211_hw *hw, bool b_pseudo_ struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); struct rtl_pci_priv *pcipriv = rtl_pcipriv(hw); - struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; - u16 i, usvalue; - u8 hwinfo[HWSET_MAX_SIZE]; - u16 eeprom_id; + int params[] = {RTL_EEPROM_ID, EEPROM_VID, EEPROM_DID, + EEPROM_SVID, EEPROM_SMID, EEPROM_MAC_ADDR, + EEPROM_CHANNELPLAN, EEPROM_VERSION, EEPROM_CUSTOMER_ID, + COUNTRY_CODE_WORLD_WIDE_13}; + u8 *hwinfo; if (b_pseudo_test) { ;/* need add */ } - switch (rtlefuse->epromtype) { - case EEPROM_BOOT_EFUSE: - rtl_efuse_shadow_map_update(hw); - break; - - case EEPROM_93C46: - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!\n"); - return; - - default: - dev_warn(dev, "no efuse data\n"); - } - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); - - RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP\n", - hwinfo, HWSET_MAX_SIZE); - - eeprom_id = *((u16 *)&hwinfo[0]); - if (eeprom_id != RTL_EEPROM_ID) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_WARNING, - "EEPROM ID(%#x) is invalid!!\n", eeprom_id); - rtlefuse->autoload_failflag = true; - } else { - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "Autoload OK\n"); - rtlefuse->autoload_failflag = false; - } - - if (rtlefuse->autoload_failflag) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL8812AE autoload_failflag, check it !!\n"); + hwinfo = kzalloc(HWSET_MAX_SIZE, GFP_KERNEL); + if (!hwinfo) return; - } - - rtlefuse->eeprom_version = *(u8 *)&hwinfo[EEPROM_VERSION]; - if (rtlefuse->eeprom_version == 0xff) - rtlefuse->eeprom_version = 0; - - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM version: 0x%2x\n", rtlefuse->eeprom_version); - - rtlefuse->eeprom_vid = *(u16 *)&hwinfo[EEPROM_VID]; - rtlefuse->eeprom_did = *(u16 *)&hwinfo[EEPROM_DID]; - rtlefuse->eeprom_svid = *(u16 *)&hwinfo[EEPROM_SVID]; - rtlefuse->eeprom_smid = *(u16 *)&hwinfo[EEPROM_SMID]; - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROMId = 0x%4x\n", eeprom_id); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM VID = 0x%4x\n", rtlefuse->eeprom_vid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM DID = 0x%4x\n", rtlefuse->eeprom_did); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SVID = 0x%4x\n", rtlefuse->eeprom_svid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SMID = 0x%4x\n", rtlefuse->eeprom_smid); - - /*customer ID*/ - rtlefuse->eeprom_oemid = *(u8 *)&hwinfo[EEPROM_CUSTOMER_ID]; - if (rtlefuse->eeprom_oemid == 0xFF) - rtlefuse->eeprom_oemid = 0; - - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM Customer ID: 0x%2x\n", rtlefuse->eeprom_oemid); - - for (i = 0; i < 6; i += 2) { - usvalue = *(u16 *)&hwinfo[EEPROM_MAC_ADDR + i]; - *((u16 *)(&rtlefuse->dev_addr[i])) = usvalue; - } - RT_TRACE(rtlpriv, COMP_INIT, DBG_DMESG, - "dev_addr: %pM\n", rtlefuse->dev_addr); + if (rtl_get_hwinfo(hw, rtlpriv, HWSET_MAX_SIZE, hwinfo, params)) + goto exit; _rtl8821ae_read_txpower_info_from_hwpg(hw, rtlefuse->autoload_failflag, hwinfo); @@ -3280,6 +3216,8 @@ static void _rtl8821ae_read_adapter_info(struct ieee80211_hw *hw, bool b_pseudo_ break; } } +exit: + kfree(hwinfo); } /*static void _rtl8821ae_hal_customized_behavior(struct ieee80211_hw *hw) -- cgit v0.10.2 From a8c9fb2b82ab00fad8992131c105387429e296ff Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 5 Jul 2016 10:08:14 -0500 Subject: rtlwifi: rtl8192de: Convert driver to use common hardware info routine The driver for RTL8192DE chips is converted to use the common routine for getting the hardware information. Reported-by: Arnd Bergmann Signed-off-by: Larry Finger Cc: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c index 8618c32..b0f6324 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8192de/hw.c @@ -1744,71 +1744,26 @@ static void _rtl92de_read_adapter_info(struct ieee80211_hw *hw) struct rtl_priv *rtlpriv = rtl_priv(hw); struct rtl_efuse *rtlefuse = rtl_efuse(rtl_priv(hw)); struct rtl_hal *rtlhal = rtl_hal(rtl_priv(hw)); - struct device *dev = &rtl_pcipriv(hw)->dev.pdev->dev; - u16 i, usvalue; - u8 hwinfo[HWSET_MAX_SIZE]; - u16 eeprom_id; - unsigned long flags; + int params[] = {RTL8190_EEPROM_ID, EEPROM_VID, EEPROM_DID, + EEPROM_SVID, EEPROM_SMID, EEPROM_MAC_ADDR_MAC0_92D, + EEPROM_CHANNEL_PLAN, EEPROM_VERSION, EEPROM_CUSTOMER_ID, + COUNTRY_CODE_WORLD_WIDE_13}; + int i; + u16 usvalue; + u8 *hwinfo; - switch (rtlefuse->epromtype) { - case EEPROM_BOOT_EFUSE: - spin_lock_irqsave(&globalmutex_for_power_and_efuse, flags); - rtl_efuse_shadow_map_update(hw); - _rtl92de_efuse_update_chip_version(hw); - spin_unlock_irqrestore(&globalmutex_for_power_and_efuse, flags); - break; - case EEPROM_93C46: - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!\n"); - return; - default: - dev_warn(dev, "no efuse data\n"); + hwinfo = kzalloc(HWSET_MAX_SIZE, GFP_KERNEL); + if (!hwinfo) return; - } - - memcpy(hwinfo, &rtlefuse->efuse_map[EFUSE_INIT_MAP][0], HWSET_MAX_SIZE); - RT_PRINT_DATA(rtlpriv, COMP_INIT, DBG_DMESG, "MAP", - hwinfo, HWSET_MAX_SIZE); - eeprom_id = *((u16 *)&hwinfo[0]); - if (eeprom_id != RTL8190_EEPROM_ID) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_WARNING, - "EEPROM ID(%#x) is invalid!!\n", eeprom_id); - rtlefuse->autoload_failflag = true; - } else { - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "Autoload OK\n"); - rtlefuse->autoload_failflag = false; - } - if (rtlefuse->autoload_failflag) { - RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG, - "RTL819X Not boot from eeprom, check it !!\n"); + if (rtl_get_hwinfo(hw, rtlpriv, HWSET_MAX_SIZE, hwinfo, params)) return; - } - rtlefuse->eeprom_oemid = hwinfo[EEPROM_CUSTOMER_ID]; - _rtl92de_read_macphymode_and_bandtype(hw, hwinfo); - /* VID, DID SE 0xA-D */ - rtlefuse->eeprom_vid = *(u16 *)&hwinfo[EEPROM_VID]; - rtlefuse->eeprom_did = *(u16 *)&hwinfo[EEPROM_DID]; - rtlefuse->eeprom_svid = *(u16 *)&hwinfo[EEPROM_SVID]; - rtlefuse->eeprom_smid = *(u16 *)&hwinfo[EEPROM_SMID]; - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, "EEPROMId = 0x%4x\n", eeprom_id); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM VID = 0x%4x\n", rtlefuse->eeprom_vid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM DID = 0x%4x\n", rtlefuse->eeprom_did); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SVID = 0x%4x\n", rtlefuse->eeprom_svid); - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM SMID = 0x%4x\n", rtlefuse->eeprom_smid); + _rtl92de_efuse_update_chip_version(hw); + _rtl92de_read_macphymode_and_bandtype(hw, hwinfo); - /* Read Permanent MAC address */ - if (rtlhal->interfaceindex == 0) { - for (i = 0; i < 6; i += 2) { - usvalue = *(u16 *)&hwinfo[EEPROM_MAC_ADDR_MAC0_92D + i]; - *((u16 *) (&rtlefuse->dev_addr[i])) = usvalue; - } - } else { + /* Read Permanent MAC address for 2nd interface */ + if (rtlhal->interfaceindex != 0) { for (i = 0; i < 6; i += 2) { usvalue = *(u16 *)&hwinfo[EEPROM_MAC_ADDR_MAC1_92D + i]; *((u16 *) (&rtlefuse->dev_addr[i])) = usvalue; @@ -1834,10 +1789,8 @@ static void _rtl92de_read_adapter_info(struct ieee80211_hw *hw) rtlefuse->channel_plan = COUNTRY_CODE_FCC; break; } - rtlefuse->eeprom_version = *(u16 *)&hwinfo[EEPROM_VERSION]; rtlefuse->txpwr_fromeprom = true; - RT_TRACE(rtlpriv, COMP_INIT, DBG_LOUD, - "EEPROM Customer ID: 0x%2x\n", rtlefuse->eeprom_oemid); + kfree(hwinfo); } void rtl92de_read_eeprom_info(struct ieee80211_hw *hw) -- cgit v0.10.2 From 238ad2ddf34bb3cd13f0f86c0d1b112bf9ef43ac Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Tue, 5 Jul 2016 10:08:15 -0500 Subject: rtlwifi: rtl8723ae: Clean up the hardware info routine This driver contains some complicated if ... else if ... else constructions. These are replaced by switch statements to improve readability. Signed-off-by: Larry Finger Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c index 662c445..b88c7ee 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8723ae/hw.c @@ -1653,132 +1653,135 @@ static void _rtl8723e_read_adapter_info(struct ieee80211_hw *hw, rtl8723e_read_bt_coexist_info_from_hwpg(hw, rtlefuse->autoload_failflag, hwinfo); - if (rtlhal->oem_id == RT_CID_DEFAULT) { - switch (rtlefuse->eeprom_oemid) { - case EEPROM_CID_DEFAULT: - if (rtlefuse->eeprom_did == 0x8176) { - if (CHK_SVID_SMID(0x10EC, 0x6151) || - CHK_SVID_SMID(0x10EC, 0x6152) || - CHK_SVID_SMID(0x10EC, 0x6154) || - CHK_SVID_SMID(0x10EC, 0x6155) || - CHK_SVID_SMID(0x10EC, 0x6177) || - CHK_SVID_SMID(0x10EC, 0x6178) || - CHK_SVID_SMID(0x10EC, 0x6179) || - CHK_SVID_SMID(0x10EC, 0x6180) || - CHK_SVID_SMID(0x10EC, 0x7151) || - CHK_SVID_SMID(0x10EC, 0x7152) || - CHK_SVID_SMID(0x10EC, 0x7154) || - CHK_SVID_SMID(0x10EC, 0x7155) || - CHK_SVID_SMID(0x10EC, 0x7177) || - CHK_SVID_SMID(0x10EC, 0x7178) || - CHK_SVID_SMID(0x10EC, 0x7179) || - CHK_SVID_SMID(0x10EC, 0x7180) || - CHK_SVID_SMID(0x10EC, 0x8151) || - CHK_SVID_SMID(0x10EC, 0x8152) || - CHK_SVID_SMID(0x10EC, 0x8154) || - CHK_SVID_SMID(0x10EC, 0x8155) || - CHK_SVID_SMID(0x10EC, 0x8181) || - CHK_SVID_SMID(0x10EC, 0x8182) || - CHK_SVID_SMID(0x10EC, 0x8184) || - CHK_SVID_SMID(0x10EC, 0x8185) || - CHK_SVID_SMID(0x10EC, 0x9151) || - CHK_SVID_SMID(0x10EC, 0x9152) || - CHK_SVID_SMID(0x10EC, 0x9154) || - CHK_SVID_SMID(0x10EC, 0x9155) || - CHK_SVID_SMID(0x10EC, 0x9181) || - CHK_SVID_SMID(0x10EC, 0x9182) || - CHK_SVID_SMID(0x10EC, 0x9184) || - CHK_SVID_SMID(0x10EC, 0x9185)) + if (rtlhal->oem_id != RT_CID_DEFAULT) + return; + + switch (rtlefuse->eeprom_oemid) { + case EEPROM_CID_DEFAULT: + switch (rtlefuse->eeprom_did) { + case 0x8176: + switch (rtlefuse->eeprom_svid) { + case 0x10EC: + switch (rtlefuse->eeprom_smid) { + case 0x6151 ... 0x6152: + case 0x6154 ... 0x6155: + case 0x6177 ... 0x6180: + case 0x7151 ... 0x7152: + case 0x7154 ... 0x7155: + case 0x7177 ... 0x7180: + case 0x8151 ... 0x8152: + case 0x8154 ... 0x8155: + case 0x8181 ... 0x8182: + case 0x8184 ... 0x8185: + case 0x9151 ... 0x9152: + case 0x9154 ... 0x9155: + case 0x9181 ... 0x9182: + case 0x9184 ... 0x9185: rtlhal->oem_id = RT_CID_TOSHIBA; - else if (rtlefuse->eeprom_svid == 0x1025) - rtlhal->oem_id = RT_CID_819X_ACER; - else if (CHK_SVID_SMID(0x10EC, 0x6191) || - CHK_SVID_SMID(0x10EC, 0x6192) || - CHK_SVID_SMID(0x10EC, 0x6193) || - CHK_SVID_SMID(0x10EC, 0x7191) || - CHK_SVID_SMID(0x10EC, 0x7192) || - CHK_SVID_SMID(0x10EC, 0x7193) || - CHK_SVID_SMID(0x10EC, 0x8191) || - CHK_SVID_SMID(0x10EC, 0x8192) || - CHK_SVID_SMID(0x10EC, 0x8193) || - CHK_SVID_SMID(0x10EC, 0x9191) || - CHK_SVID_SMID(0x10EC, 0x9192) || - CHK_SVID_SMID(0x10EC, 0x9193)) + break; + case 0x6191 ... 0x6193: + case 0x7191 ... 0x7193: + case 0x8191 ... 0x8193: + case 0x9191 ... 0x9193: rtlhal->oem_id = RT_CID_819X_SAMSUNG; - else if (CHK_SVID_SMID(0x10EC, 0x8195) || - CHK_SVID_SMID(0x10EC, 0x9195) || - CHK_SVID_SMID(0x10EC, 0x7194) || - CHK_SVID_SMID(0x10EC, 0x8200) || - CHK_SVID_SMID(0x10EC, 0x8201) || - CHK_SVID_SMID(0x10EC, 0x8202) || - CHK_SVID_SMID(0x10EC, 0x9200)) - rtlhal->oem_id = RT_CID_819X_LENOVO; - else if (CHK_SVID_SMID(0x10EC, 0x8197) || - CHK_SVID_SMID(0x10EC, 0x9196)) + break; + case 0x8197: + case 0x9196: rtlhal->oem_id = RT_CID_819X_CLEVO; - else if (CHK_SVID_SMID(0x1028, 0x8194) || - CHK_SVID_SMID(0x1028, 0x8198) || - CHK_SVID_SMID(0x1028, 0x9197) || - CHK_SVID_SMID(0x1028, 0x9198)) + break; + case 0x8203: + rtlhal->oem_id = RT_CID_819X_PRONETS; + break; + case 0x8195: + case 0x9195: + case 0x7194: + case 0x8200 ... 0x8202: + case 0x9200: + rtlhal->oem_id = RT_CID_819X_LENOVO; + break; + } + case 0x1025: + rtlhal->oem_id = RT_CID_819X_ACER; + break; + case 0x1028: + switch (rtlefuse->eeprom_smid) { + case 0x8194: + case 0x8198: + case 0x9197 ... 0x9198: rtlhal->oem_id = RT_CID_819X_DELL; - else if (CHK_SVID_SMID(0x103C, 0x1629)) + break; + } + break; + case 0x103C: + switch (rtlefuse->eeprom_smid) { + case 0x1629: rtlhal->oem_id = RT_CID_819X_HP; - else if (CHK_SVID_SMID(0x1A32, 0x2315)) + } + break; + case 0x1A32: + switch (rtlefuse->eeprom_smid) { + case 0x2315: rtlhal->oem_id = RT_CID_819X_QMI; - else if (CHK_SVID_SMID(0x10EC, 0x8203)) - rtlhal->oem_id = RT_CID_819X_PRONETS; - else if (CHK_SVID_SMID(0x1043, 0x84B5)) - rtlhal->oem_id = - RT_CID_819X_EDIMAX_ASUS; - else - rtlhal->oem_id = RT_CID_DEFAULT; - } else if (rtlefuse->eeprom_did == 0x8178) { - if (CHK_SVID_SMID(0x10EC, 0x6181) || - CHK_SVID_SMID(0x10EC, 0x6182) || - CHK_SVID_SMID(0x10EC, 0x6184) || - CHK_SVID_SMID(0x10EC, 0x6185) || - CHK_SVID_SMID(0x10EC, 0x7181) || - CHK_SVID_SMID(0x10EC, 0x7182) || - CHK_SVID_SMID(0x10EC, 0x7184) || - CHK_SVID_SMID(0x10EC, 0x7185) || - CHK_SVID_SMID(0x10EC, 0x8181) || - CHK_SVID_SMID(0x10EC, 0x8182) || - CHK_SVID_SMID(0x10EC, 0x8184) || - CHK_SVID_SMID(0x10EC, 0x8185) || - CHK_SVID_SMID(0x10EC, 0x9181) || - CHK_SVID_SMID(0x10EC, 0x9182) || - CHK_SVID_SMID(0x10EC, 0x9184) || - CHK_SVID_SMID(0x10EC, 0x9185)) - rtlhal->oem_id = RT_CID_TOSHIBA; - else if (rtlefuse->eeprom_svid == 0x1025) - rtlhal->oem_id = RT_CID_819X_ACER; - else if (CHK_SVID_SMID(0x10EC, 0x8186)) - rtlhal->oem_id = RT_CID_819X_PRONETS; - else if (CHK_SVID_SMID(0x1043, 0x8486)) + break; + } + break; + case 0x1043: + switch (rtlefuse->eeprom_smid) { + case 0x84B5: rtlhal->oem_id = - RT_CID_819X_EDIMAX_ASUS; - else - rtlhal->oem_id = RT_CID_DEFAULT; - } else { - rtlhal->oem_id = RT_CID_DEFAULT; + RT_CID_819X_EDIMAX_ASUS; + } + break; } break; - case EEPROM_CID_TOSHIBA: - rtlhal->oem_id = RT_CID_TOSHIBA; - break; - case EEPROM_CID_CCX: - rtlhal->oem_id = RT_CID_CCX; - break; - case EEPROM_CID_QMI: - rtlhal->oem_id = RT_CID_819X_QMI; - break; - case EEPROM_CID_WHQL: + case 0x8178: + switch (rtlefuse->eeprom_svid) { + case 0x10ec: + switch (rtlefuse->eeprom_smid) { + case 0x6181 ... 0x6182: + case 0x6184 ... 0x6185: + case 0x7181 ... 0x7182: + case 0x7184 ... 0x7185: + case 0x8181 ... 0x8182: + case 0x8184 ... 0x8185: + case 0x9181 ... 0x9182: + case 0x9184 ... 0x9185: + rtlhal->oem_id = RT_CID_TOSHIBA; + break; + case 0x8186: + rtlhal->oem_id = + RT_CID_819X_PRONETS; + break; + } break; - default: - rtlhal->oem_id = RT_CID_DEFAULT; + case 0x1025: + rtlhal->oem_id = RT_CID_819X_ACER; + break; + case 0x1043: + switch (rtlefuse->eeprom_smid) { + case 0x8486: + rtlhal->oem_id = + RT_CID_819X_EDIMAX_ASUS; + } + break; + } break; - } + break; + case EEPROM_CID_TOSHIBA: + rtlhal->oem_id = RT_CID_TOSHIBA; + break; + case EEPROM_CID_CCX: + rtlhal->oem_id = RT_CID_CCX; + break; + case EEPROM_CID_QMI: + rtlhal->oem_id = RT_CID_819X_QMI; + break; + case EEPROM_CID_WHQL: + break; + default: + rtlhal->oem_id = RT_CID_DEFAULT; + break; } exit: kfree(hwinfo); -- cgit v0.10.2 From 909aa60c057209dd422564884ea8828530b5f1c9 Mon Sep 17 00:00:00 2001 From: Bjorn Andersson Date: Sun, 19 Jun 2016 23:19:43 -0700 Subject: wcn36xx: Fold indication payload into message header Merge the two allocation instead of separately allocating room for the indication payload. Signed-off-by: Bjorn Andersson Signed-off-by: Bjorn Andersson Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wcn36xx/smd.c b/drivers/net/wireless/ath/wcn36xx/smd.c index e8b630c..1a9580d 100644 --- a/drivers/net/wireless/ath/wcn36xx/smd.c +++ b/drivers/net/wireless/ath/wcn36xx/smd.c @@ -2229,14 +2229,8 @@ static void wcn36xx_smd_rsp_process(struct wcn36xx *wcn, void *buf, size_t len) case WCN36XX_HAL_OTA_TX_COMPL_IND: case WCN36XX_HAL_MISSED_BEACON_IND: case WCN36XX_HAL_DELETE_STA_CONTEXT_IND: - msg_ind = kmalloc(sizeof(*msg_ind), GFP_KERNEL); - if (!msg_ind) - goto nomem; - msg_ind->msg_len = len; - msg_ind->msg = kmemdup(buf, len, GFP_KERNEL); - if (!msg_ind->msg) { - kfree(msg_ind); -nomem: + msg_ind = kmalloc(sizeof(*msg_ind) + len, GFP_KERNEL); + if (!msg_ind) { /* * FIXME: Do something smarter then just * printing an error. @@ -2245,6 +2239,10 @@ nomem: msg_header->msg_type); break; } + + msg_ind->msg_len = len; + memcpy(msg_ind->msg, buf, len); + mutex_lock(&wcn->hal_ind_mutex); list_add_tail(&msg_ind->list, &wcn->hal_ind_queue); queue_work(wcn->hal_ind_wq, &wcn->hal_ind_work); @@ -2295,7 +2293,6 @@ static void wcn36xx_ind_smd_work(struct work_struct *work) msg_header->msg_type); } list_del(wcn->hal_ind_queue.next); - kfree(hal_ind_msg->msg); kfree(hal_ind_msg); mutex_unlock(&wcn->hal_ind_mutex); } diff --git a/drivers/net/wireless/ath/wcn36xx/smd.h b/drivers/net/wireless/ath/wcn36xx/smd.h index d93e3fd..c4375ce 100644 --- a/drivers/net/wireless/ath/wcn36xx/smd.h +++ b/drivers/net/wireless/ath/wcn36xx/smd.h @@ -46,8 +46,8 @@ struct wcn36xx_fw_msg_status_rsp { struct wcn36xx_hal_ind_msg { struct list_head list; - u8 *msg; size_t msg_len; + u8 msg[]; }; struct wcn36xx; -- cgit v0.10.2 From 1c41fd5fb7514fa1b30992865cee4844715688bb Mon Sep 17 00:00:00 2001 From: Bjorn Andersson Date: Sun, 19 Jun 2016 23:19:44 -0700 Subject: wcn36xx: Change indication list lock to spinlock In preparation for handling incoming messages from IRQ context, change the indication list lock to a spinlock Signed-off-by: Bjorn Andersson Signed-off-by: Bjorn Andersson Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wcn36xx/smd.c b/drivers/net/wireless/ath/wcn36xx/smd.c index 1a9580d..b210807 100644 --- a/drivers/net/wireless/ath/wcn36xx/smd.c +++ b/drivers/net/wireless/ath/wcn36xx/smd.c @@ -2243,10 +2243,10 @@ static void wcn36xx_smd_rsp_process(struct wcn36xx *wcn, void *buf, size_t len) msg_ind->msg_len = len; memcpy(msg_ind->msg, buf, len); - mutex_lock(&wcn->hal_ind_mutex); + spin_lock(&wcn->hal_ind_lock); list_add_tail(&msg_ind->list, &wcn->hal_ind_queue); queue_work(wcn->hal_ind_wq, &wcn->hal_ind_work); - mutex_unlock(&wcn->hal_ind_mutex); + spin_unlock(&wcn->hal_ind_lock); wcn36xx_dbg(WCN36XX_DBG_HAL, "indication arrived\n"); break; default: @@ -2260,8 +2260,9 @@ static void wcn36xx_ind_smd_work(struct work_struct *work) container_of(work, struct wcn36xx, hal_ind_work); struct wcn36xx_hal_msg_header *msg_header; struct wcn36xx_hal_ind_msg *hal_ind_msg; + unsigned long flags; - mutex_lock(&wcn->hal_ind_mutex); + spin_lock_irqsave(&wcn->hal_ind_lock, flags); hal_ind_msg = list_first_entry(&wcn->hal_ind_queue, struct wcn36xx_hal_ind_msg, @@ -2293,8 +2294,8 @@ static void wcn36xx_ind_smd_work(struct work_struct *work) msg_header->msg_type); } list_del(wcn->hal_ind_queue.next); + spin_unlock_irqrestore(&wcn->hal_ind_lock, flags); kfree(hal_ind_msg); - mutex_unlock(&wcn->hal_ind_mutex); } int wcn36xx_smd_open(struct wcn36xx *wcn) { @@ -2307,7 +2308,7 @@ int wcn36xx_smd_open(struct wcn36xx *wcn) } INIT_WORK(&wcn->hal_ind_work, wcn36xx_ind_smd_work); INIT_LIST_HEAD(&wcn->hal_ind_queue); - mutex_init(&wcn->hal_ind_mutex); + spin_lock_init(&wcn->hal_ind_lock); ret = wcn->ctrl_ops->open(wcn, wcn36xx_smd_rsp_process); if (ret) { @@ -2327,5 +2328,4 @@ void wcn36xx_smd_close(struct wcn36xx *wcn) { wcn->ctrl_ops->close(); destroy_workqueue(wcn->hal_ind_wq); - mutex_destroy(&wcn->hal_ind_mutex); } diff --git a/drivers/net/wireless/ath/wcn36xx/wcn36xx.h b/drivers/net/wireless/ath/wcn36xx/wcn36xx.h index 7433d67..f6d8d14 100644 --- a/drivers/net/wireless/ath/wcn36xx/wcn36xx.h +++ b/drivers/net/wireless/ath/wcn36xx/wcn36xx.h @@ -215,7 +215,7 @@ struct wcn36xx { struct completion hal_rsp_compl; struct workqueue_struct *hal_ind_wq; struct work_struct hal_ind_work; - struct mutex hal_ind_mutex; + spinlock_t hal_ind_lock; struct list_head hal_ind_queue; /* DXE channels */ -- cgit v0.10.2 From 05ddce497c056307e25c7865f540751813711d2d Mon Sep 17 00:00:00 2001 From: Bjorn Andersson Date: Sun, 19 Jun 2016 23:19:45 -0700 Subject: wcn36xx: Split mmio space into explicit regions Split the wcnss mmio space into explicit regions for ccu and dxe and acquire these from the node referenced by the qcom,mmio phandle. Signed-off-by: Bjorn Andersson Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wcn36xx/dxe.c b/drivers/net/wireless/ath/wcn36xx/dxe.c index 8643801..a8ff454 100644 --- a/drivers/net/wireless/ath/wcn36xx/dxe.c +++ b/drivers/net/wireless/ath/wcn36xx/dxe.c @@ -35,26 +35,27 @@ void *wcn36xx_dxe_get_next_bd(struct wcn36xx *wcn, bool is_low) return ch->head_blk_ctl->bd_cpu_addr; } +static void wcn36xx_ccu_write_register(struct wcn36xx *wcn, int addr, int data) +{ + wcn36xx_dbg(WCN36XX_DBG_DXE, + "wcn36xx_ccu_write_register: addr=%x, data=%x\n", + addr, data); + + writel(data, wcn->ccu_base + addr); +} + static void wcn36xx_dxe_write_register(struct wcn36xx *wcn, int addr, int data) { wcn36xx_dbg(WCN36XX_DBG_DXE, "wcn36xx_dxe_write_register: addr=%x, data=%x\n", addr, data); - writel(data, wcn->mmio + addr); + writel(data, wcn->dxe_base + addr); } -#define wcn36xx_dxe_write_register_x(wcn, reg, reg_data) \ -do { \ - if (wcn->chip_version == WCN36XX_CHIP_3680) \ - wcn36xx_dxe_write_register(wcn, reg ## _3680, reg_data); \ - else \ - wcn36xx_dxe_write_register(wcn, reg ## _3660, reg_data); \ -} while (0) \ - static void wcn36xx_dxe_read_register(struct wcn36xx *wcn, int addr, int *data) { - *data = readl(wcn->mmio + addr); + *data = readl(wcn->dxe_base + addr); wcn36xx_dbg(WCN36XX_DBG_DXE, "wcn36xx_dxe_read_register: addr=%x, data=%x\n", @@ -703,7 +704,10 @@ int wcn36xx_dxe_init(struct wcn36xx *wcn) /* Setting interrupt path */ reg_data = WCN36XX_DXE_CCU_INT; - wcn36xx_dxe_write_register_x(wcn, WCN36XX_DXE_REG_CCU_INT, reg_data); + if (wcn->chip_version == WCN36XX_CHIP_3680) + wcn36xx_ccu_write_register(wcn, WCN36XX_DXE_REG_CCU_INT_3680, reg_data); + else + wcn36xx_ccu_write_register(wcn, WCN36XX_DXE_REG_CCU_INT_3660, reg_data); /***************************************/ /* Init descriptors for TX LOW channel */ diff --git a/drivers/net/wireless/ath/wcn36xx/dxe.h b/drivers/net/wireless/ath/wcn36xx/dxe.h index 3eca4f9..012b59f 100644 --- a/drivers/net/wireless/ath/wcn36xx/dxe.h +++ b/drivers/net/wireless/ath/wcn36xx/dxe.h @@ -28,11 +28,11 @@ H2H_TEST_RX_TX = DMA2 */ /* DXE registers */ -#define WCN36XX_DXE_MEM_REG 0x202000 +#define WCN36XX_DXE_MEM_REG 0 #define WCN36XX_DXE_CCU_INT 0xA0011 -#define WCN36XX_DXE_REG_CCU_INT_3660 0x200b10 -#define WCN36XX_DXE_REG_CCU_INT_3680 0x2050dc +#define WCN36XX_DXE_REG_CCU_INT_3660 0x310 +#define WCN36XX_DXE_REG_CCU_INT_3680 0x10dc /* TODO This must calculated properly but not hardcoded */ #define WCN36XX_DXE_CTRL_TX_L 0x328a44 diff --git a/drivers/net/wireless/ath/wcn36xx/main.c b/drivers/net/wireless/ath/wcn36xx/main.c index a920d70..8146eeb 100644 --- a/drivers/net/wireless/ath/wcn36xx/main.c +++ b/drivers/net/wireless/ath/wcn36xx/main.c @@ -19,6 +19,8 @@ #include #include #include +#include +#include #include "wcn36xx.h" unsigned int wcn36xx_dbg_mask; @@ -1064,7 +1066,11 @@ static int wcn36xx_init_ieee80211(struct wcn36xx *wcn) static int wcn36xx_platform_get_resources(struct wcn36xx *wcn, struct platform_device *pdev) { + struct device_node *mmio_node; struct resource *res; + int index; + int ret; + /* Set TX IRQ */ res = platform_get_resource_byname(pdev, IORESOURCE_IRQ, "wcnss_wlantx_irq"); @@ -1083,19 +1089,38 @@ static int wcn36xx_platform_get_resources(struct wcn36xx *wcn, } wcn->rx_irq = res->start; - /* Map the memory */ - res = platform_get_resource_byname(pdev, IORESOURCE_MEM, - "wcnss_mmio"); - if (!res) { - wcn36xx_err("failed to get mmio\n"); - return -ENOENT; + mmio_node = of_parse_phandle(pdev->dev.parent->of_node, "qcom,mmio", 0); + if (!mmio_node) { + wcn36xx_err("failed to acquire qcom,mmio reference\n"); + return -EINVAL; + } + + /* Map the CCU memory */ + index = of_property_match_string(mmio_node, "reg-names", "ccu"); + wcn->ccu_base = of_iomap(mmio_node, index); + if (!wcn->ccu_base) { + wcn36xx_err("failed to map ccu memory\n"); + ret = -ENOMEM; + goto put_mmio_node; } - wcn->mmio = ioremap(res->start, resource_size(res)); - if (!wcn->mmio) { - wcn36xx_err("failed to map io memory\n"); - return -ENOMEM; + + /* Map the DXE memory */ + index = of_property_match_string(mmio_node, "reg-names", "dxe"); + wcn->dxe_base = of_iomap(mmio_node, index); + if (!wcn->dxe_base) { + wcn36xx_err("failed to map dxe memory\n"); + ret = -ENOMEM; + goto unmap_ccu; } + + of_node_put(mmio_node); return 0; + +unmap_ccu: + iounmap(wcn->ccu_base); +put_mmio_node: + of_node_put(mmio_node); + return ret; } static int wcn36xx_probe(struct platform_device *pdev) @@ -1138,7 +1163,8 @@ static int wcn36xx_probe(struct platform_device *pdev) return 0; out_unmap: - iounmap(wcn->mmio); + iounmap(wcn->ccu_base); + iounmap(wcn->dxe_base); out_wq: ieee80211_free_hw(hw); out_err: @@ -1154,7 +1180,8 @@ static int wcn36xx_remove(struct platform_device *pdev) mutex_destroy(&wcn->hal_mutex); ieee80211_unregister_hw(hw); - iounmap(wcn->mmio); + iounmap(wcn->dxe_base); + iounmap(wcn->ccu_base); ieee80211_free_hw(hw); return 0; diff --git a/drivers/net/wireless/ath/wcn36xx/wcn36xx.h b/drivers/net/wireless/ath/wcn36xx/wcn36xx.h index f6d8d14..845f2446 100644 --- a/drivers/net/wireless/ath/wcn36xx/wcn36xx.h +++ b/drivers/net/wireless/ath/wcn36xx/wcn36xx.h @@ -202,7 +202,8 @@ struct wcn36xx { /* IRQs */ int tx_irq; int rx_irq; - void __iomem *mmio; + void __iomem *ccu_base; + void __iomem *dxe_base; struct wcn36xx_platform_ctrl_ops *ctrl_ops; /* -- cgit v0.10.2 From 6f10b4e1e6adaa8679a719a636e9310b65b7d993 Mon Sep 17 00:00:00 2001 From: Bjorn Andersson Date: Sun, 19 Jun 2016 23:19:46 -0700 Subject: wcn36xx: Correct DXE chip version differentiation The CCU block in WCNSS is configured for appropriate routing of interrupts from the DXE to the application cpu, this is not dependant on the iris version (wcn3660 vs wcn3680), but rather if the SoC has a riva or pronto built in. Signed-off-by: Bjorn Andersson Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wcn36xx/dxe.c b/drivers/net/wireless/ath/wcn36xx/dxe.c index a8ff454..231fd02 100644 --- a/drivers/net/wireless/ath/wcn36xx/dxe.c +++ b/drivers/net/wireless/ath/wcn36xx/dxe.c @@ -702,12 +702,13 @@ int wcn36xx_dxe_init(struct wcn36xx *wcn) reg_data = WCN36XX_DXE_REG_RESET; wcn36xx_dxe_write_register(wcn, WCN36XX_DXE_REG_CSR_RESET, reg_data); - /* Setting interrupt path */ - reg_data = WCN36XX_DXE_CCU_INT; - if (wcn->chip_version == WCN36XX_CHIP_3680) - wcn36xx_ccu_write_register(wcn, WCN36XX_DXE_REG_CCU_INT_3680, reg_data); + /* Select channels for rx avail and xfer done interrupts... */ + reg_data = (WCN36XX_DXE_INT_CH3_MASK | WCN36XX_DXE_INT_CH1_MASK) << 16 | + WCN36XX_DXE_INT_CH0_MASK | WCN36XX_DXE_INT_CH4_MASK; + if (wcn->is_pronto) + wcn36xx_ccu_write_register(wcn, WCN36XX_CCU_DXE_INT_SELECT_PRONTO, reg_data); else - wcn36xx_ccu_write_register(wcn, WCN36XX_DXE_REG_CCU_INT_3660, reg_data); + wcn36xx_ccu_write_register(wcn, WCN36XX_CCU_DXE_INT_SELECT_RIVA, reg_data); /***************************************/ /* Init descriptors for TX LOW channel */ diff --git a/drivers/net/wireless/ath/wcn36xx/dxe.h b/drivers/net/wireless/ath/wcn36xx/dxe.h index 012b59f..c012e80 100644 --- a/drivers/net/wireless/ath/wcn36xx/dxe.h +++ b/drivers/net/wireless/ath/wcn36xx/dxe.h @@ -30,9 +30,8 @@ H2H_TEST_RX_TX = DMA2 /* DXE registers */ #define WCN36XX_DXE_MEM_REG 0 -#define WCN36XX_DXE_CCU_INT 0xA0011 -#define WCN36XX_DXE_REG_CCU_INT_3660 0x310 -#define WCN36XX_DXE_REG_CCU_INT_3680 0x10dc +#define WCN36XX_CCU_DXE_INT_SELECT_RIVA 0x310 +#define WCN36XX_CCU_DXE_INT_SELECT_PRONTO 0x10dc /* TODO This must calculated properly but not hardcoded */ #define WCN36XX_DXE_CTRL_TX_L 0x328a44 diff --git a/drivers/net/wireless/ath/wcn36xx/main.c b/drivers/net/wireless/ath/wcn36xx/main.c index 8146eeb..e1d59da 100644 --- a/drivers/net/wireless/ath/wcn36xx/main.c +++ b/drivers/net/wireless/ath/wcn36xx/main.c @@ -261,17 +261,6 @@ static void wcn36xx_feat_caps_info(struct wcn36xx *wcn) } } -static void wcn36xx_detect_chip_version(struct wcn36xx *wcn) -{ - if (get_feat_caps(wcn->fw_feat_caps, DOT11AC)) { - wcn36xx_info("Chip is 3680\n"); - wcn->chip_version = WCN36XX_CHIP_3680; - } else { - wcn36xx_info("Chip is 3660\n"); - wcn->chip_version = WCN36XX_CHIP_3660; - } -} - static int wcn36xx_start(struct ieee80211_hw *hw) { struct wcn36xx *wcn = hw->priv; @@ -326,9 +315,6 @@ static int wcn36xx_start(struct ieee80211_hw *hw) wcn36xx_feat_caps_info(wcn); } - wcn36xx_detect_chip_version(wcn); - wcn36xx_smd_update_cfg(wcn, WCN36XX_HAL_CFG_ENABLE_MC_ADDR_LIST, 1); - /* DMA channel initialization */ ret = wcn36xx_dxe_init(wcn); if (ret) { @@ -1095,6 +1081,8 @@ static int wcn36xx_platform_get_resources(struct wcn36xx *wcn, return -EINVAL; } + wcn->is_pronto = !!of_device_is_compatible(mmio_node, "qcom,pronto"); + /* Map the CCU memory */ index = of_property_match_string(mmio_node, "reg-names", "ccu"); wcn->ccu_base = of_iomap(mmio_node, index); diff --git a/drivers/net/wireless/ath/wcn36xx/wcn36xx.h b/drivers/net/wireless/ath/wcn36xx/wcn36xx.h index 845f2446..22242d1 100644 --- a/drivers/net/wireless/ath/wcn36xx/wcn36xx.h +++ b/drivers/net/wireless/ath/wcn36xx/wcn36xx.h @@ -193,7 +193,7 @@ struct wcn36xx { u8 fw_minor; u8 fw_major; u32 fw_feat_caps[WCN36XX_HAL_CAPS_SIZE]; - u32 chip_version; + bool is_pronto; /* extra byte for the NULL termination */ u8 crm_version[WCN36XX_HAL_VERSION_LENGTH + 1]; @@ -242,9 +242,6 @@ struct wcn36xx { }; -#define WCN36XX_CHIP_3660 0 -#define WCN36XX_CHIP_3680 1 - static inline bool wcn36xx_is_fw_version(struct wcn36xx *wcn, u8 major, u8 minor, -- cgit v0.10.2 From 86ceae90d3f9580d8c69f814d252cde80f85b404 Mon Sep 17 00:00:00 2001 From: Bjorn Andersson Date: Sun, 19 Jun 2016 23:19:48 -0700 Subject: wcn36xx: Fix up wcn36xx_smd_update_scan_params() Fix up the wcn36xx_smd_update_scan_params() to work with non-ancient versions of the firmware and support actually specifying the list of channels. Signed-off-by: Bjorn Andersson Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wcn36xx/hal.h b/drivers/net/wireless/ath/wcn36xx/hal.h index 658bfb8..4f87ef1 100644 --- a/drivers/net/wireless/ath/wcn36xx/hal.h +++ b/drivers/net/wireless/ath/wcn36xx/hal.h @@ -4123,7 +4123,7 @@ struct wcn36xx_hal_update_scan_params_req { /* Update scan params - sent from host to PNO to be used during PNO * scanningx */ -struct update_scan_params_req_ex { +struct wcn36xx_hal_update_scan_params_req_ex { struct wcn36xx_hal_msg_header header; @@ -4151,7 +4151,7 @@ struct update_scan_params_req_ex { /* Cb State */ enum phy_chan_bond_state state; -}; +} __packed; /* Update scan params - sent from host to PNO to be used during PNO * scanningx */ diff --git a/drivers/net/wireless/ath/wcn36xx/smd.c b/drivers/net/wireless/ath/wcn36xx/smd.c index b210807..feebaae 100644 --- a/drivers/net/wireless/ath/wcn36xx/smd.c +++ b/drivers/net/wireless/ath/wcn36xx/smd.c @@ -674,22 +674,25 @@ static int wcn36xx_smd_update_scan_params_rsp(void *buf, size_t len) return 0; } -int wcn36xx_smd_update_scan_params(struct wcn36xx *wcn) +int wcn36xx_smd_update_scan_params(struct wcn36xx *wcn, + u8 *channels, size_t channel_count) { - struct wcn36xx_hal_update_scan_params_req msg_body; + struct wcn36xx_hal_update_scan_params_req_ex msg_body; int ret = 0; mutex_lock(&wcn->hal_mutex); INIT_HAL_MSG(msg_body, WCN36XX_HAL_UPDATE_SCAN_PARAM_REQ); - msg_body.dot11d_enabled = 0; - msg_body.dot11d_resolved = 0; - msg_body.channel_count = 26; + msg_body.dot11d_enabled = false; + msg_body.dot11d_resolved = true; + + msg_body.channel_count = channel_count; + memcpy(msg_body.channels, channels, channel_count); msg_body.active_min_ch_time = 60; msg_body.active_max_ch_time = 120; msg_body.passive_min_ch_time = 60; msg_body.passive_max_ch_time = 110; - msg_body.state = 0; + msg_body.state = PHY_SINGLE_CHANNEL_CENTERED; PREPARE_HAL_BUF(wcn->hal_buf, msg_body); diff --git a/drivers/net/wireless/ath/wcn36xx/smd.h b/drivers/net/wireless/ath/wcn36xx/smd.h index c4375ce..df80cbb 100644 --- a/drivers/net/wireless/ath/wcn36xx/smd.h +++ b/drivers/net/wireless/ath/wcn36xx/smd.h @@ -63,7 +63,7 @@ int wcn36xx_smd_start_scan(struct wcn36xx *wcn); int wcn36xx_smd_end_scan(struct wcn36xx *wcn); int wcn36xx_smd_finish_scan(struct wcn36xx *wcn, enum wcn36xx_hal_sys_mode mode); -int wcn36xx_smd_update_scan_params(struct wcn36xx *wcn); +int wcn36xx_smd_update_scan_params(struct wcn36xx *wcn, u8 *channels, size_t channel_count); int wcn36xx_smd_add_sta_self(struct wcn36xx *wcn, struct ieee80211_vif *vif); int wcn36xx_smd_delete_sta_self(struct wcn36xx *wcn, u8 *addr); int wcn36xx_smd_delete_sta(struct wcn36xx *wcn, u8 sta_index); -- cgit v0.10.2 From 3467f0d433016c45d1851f3587d32816b7b2ffb0 Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Thu, 23 Jun 2016 16:57:09 +0200 Subject: ath9k: Allow configuration of LED polarity in platform data. Some devices running OpenWrt need this and it makes sense to add this to ath9k_platform_data as the next patches will add a devicetree (boolean) property for it as well. Suggested-by: Vittorio Gambaletta Signed-off-by: Martin Blumenstingl Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c index 2ee8624..384929d 100644 --- a/drivers/net/wireless/ath/ath9k/init.c +++ b/drivers/net/wireless/ath/ath9k/init.c @@ -527,6 +527,9 @@ static int ath9k_init_soc_platform(struct ath_softc *sc) return ret; } + if (pdata->led_active_high) + ah->config.led_active_high = true; + if (pdata->tx_gain_buffalo) ah->config.tx_gain_buffalo = true; diff --git a/include/linux/ath9k_platform.h b/include/linux/ath9k_platform.h index e66153d..76860a4 100644 --- a/include/linux/ath9k_platform.h +++ b/include/linux/ath9k_platform.h @@ -40,6 +40,7 @@ struct ath9k_platform_data { bool tx_gain_buffalo; bool disable_2ghz; bool disable_5ghz; + bool led_active_high; int (*get_mac_revision)(void); int (*external_reset)(void); -- cgit v0.10.2 From b27301f86cc7e2352b266d386e3d77915cfe98f1 Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Thu, 23 Jun 2016 16:57:10 +0200 Subject: ath9k: remove variable which is set but never read No functional changes - this only removes a variable which is set but never read. Signed-off-by: Martin Blumenstingl Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c index 8b2895f..4f98ca0 100644 --- a/drivers/net/wireless/ath/ath9k/hw.c +++ b/drivers/net/wireless/ath/ath9k/hw.c @@ -474,15 +474,12 @@ static void ath9k_hw_init_defaults(struct ath_hw *ah) static int ath9k_hw_init_macaddr(struct ath_hw *ah) { struct ath_common *common = ath9k_hw_common(ah); - u32 sum; int i; u16 eeval; static const u32 EEP_MAC[] = { EEP_MAC_LSW, EEP_MAC_MID, EEP_MAC_MSW }; - sum = 0; for (i = 0; i < 3; i++) { eeval = ah->eep_ops->get_eeprom(ah, EEP_MAC[i]); - sum += eeval; common->macaddr[2 * i] = eeval >> 8; common->macaddr[2 * i + 1] = eeval & 0xff; } -- cgit v0.10.2 From 0cefa9749883391dd7203b059df55202863f459c Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Thu, 23 Jun 2016 16:57:11 +0200 Subject: ath9k: ath9k_hw_init_macaddr should not overwrite valid MAC addresses Currently setting the MAC address via ath9k_platform_data works only due to the order in which init.c sets common->macaddr, which is done after ath9k_hw_init_macaddr was executed. It would be better if the latter was independent of the order in which it's being called. Signed-off-by: Martin Blumenstingl Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c index 4f98ca0..4dd3aca 100644 --- a/drivers/net/wireless/ath/ath9k/hw.c +++ b/drivers/net/wireless/ath/ath9k/hw.c @@ -478,21 +478,25 @@ static int ath9k_hw_init_macaddr(struct ath_hw *ah) u16 eeval; static const u32 EEP_MAC[] = { EEP_MAC_LSW, EEP_MAC_MID, EEP_MAC_MSW }; + /* MAC address may already be loaded via ath9k_platform_data */ + if (is_valid_ether_addr(common->macaddr)) + return 0; + for (i = 0; i < 3; i++) { eeval = ah->eep_ops->get_eeprom(ah, EEP_MAC[i]); common->macaddr[2 * i] = eeval >> 8; common->macaddr[2 * i + 1] = eeval & 0xff; } - if (!is_valid_ether_addr(common->macaddr)) { - ath_err(common, - "eeprom contains invalid mac address: %pM\n", - common->macaddr); - random_ether_addr(common->macaddr); - ath_err(common, - "random mac address will be used: %pM\n", - common->macaddr); - } + if (is_valid_ether_addr(common->macaddr)) + return 0; + + ath_err(common, "eeprom contains invalid mac address: %pM\n", + common->macaddr); + + random_ether_addr(common->macaddr); + ath_err(common, "random mac address will be used: %pM\n", + common->macaddr); return 0; } -- cgit v0.10.2 From d323cb71abeffe8bb20cac0177bc045e5a59948e Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Thu, 23 Jun 2016 16:57:12 +0200 Subject: ath9k: remove return value from ath9k_hw_init_macaddr ath9k_hw_init_macaddr unconditionally returns 0 in all cases, making the return value unnecessary. Signed-off-by: Martin Blumenstingl Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c index 4dd3aca..fa59117 100644 --- a/drivers/net/wireless/ath/ath9k/hw.c +++ b/drivers/net/wireless/ath/ath9k/hw.c @@ -471,7 +471,7 @@ static void ath9k_hw_init_defaults(struct ath_hw *ah) ah->tx_trig_level = (AR_FTRIG_512B >> AR_FTRIG_S); } -static int ath9k_hw_init_macaddr(struct ath_hw *ah) +static void ath9k_hw_init_macaddr(struct ath_hw *ah) { struct ath_common *common = ath9k_hw_common(ah); int i; @@ -480,7 +480,7 @@ static int ath9k_hw_init_macaddr(struct ath_hw *ah) /* MAC address may already be loaded via ath9k_platform_data */ if (is_valid_ether_addr(common->macaddr)) - return 0; + return; for (i = 0; i < 3; i++) { eeval = ah->eep_ops->get_eeprom(ah, EEP_MAC[i]); @@ -489,7 +489,7 @@ static int ath9k_hw_init_macaddr(struct ath_hw *ah) } if (is_valid_ether_addr(common->macaddr)) - return 0; + return; ath_err(common, "eeprom contains invalid mac address: %pM\n", common->macaddr); @@ -498,7 +498,7 @@ static int ath9k_hw_init_macaddr(struct ath_hw *ah) ath_err(common, "random mac address will be used: %pM\n", common->macaddr); - return 0; + return; } static int ath9k_hw_post_init(struct ath_hw *ah) @@ -637,12 +637,7 @@ static int __ath9k_hw_init(struct ath_hw *ah) if (r) return r; - r = ath9k_hw_init_macaddr(ah); - if (r) { - ath_err(common, "Failed to initialize MAC address\n"); - return r; - } - + ath9k_hw_init_macaddr(ah); ath9k_hw_init_hang_checks(ah); common->state = ATH_HW_INITIALIZED; -- cgit v0.10.2 From 28755b8f6d94a51810f68e1dfe302309ce452719 Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Thu, 23 Jun 2016 16:57:13 +0200 Subject: ath9k: move all ath9k_platform_data initialization into one function No functional changes, this simply makes the code easier to understand because all initialization based on ath9k_platform_data is now within one function. Signed-off-by: Martin Blumenstingl Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c index 384929d..a0f4a52 100644 --- a/drivers/net/wireless/ath/ath9k/init.c +++ b/drivers/net/wireless/ath/ath9k/init.c @@ -512,15 +512,31 @@ static void ath9k_eeprom_release(struct ath_softc *sc) release_firmware(sc->sc_ah->eeprom_blob); } -static int ath9k_init_soc_platform(struct ath_softc *sc) +static int ath9k_init_platform(struct ath_softc *sc) { struct ath9k_platform_data *pdata = sc->dev->platform_data; struct ath_hw *ah = sc->sc_ah; - int ret = 0; + struct ath_common *common = ath9k_hw_common(ah); + int ret; if (!pdata) return 0; + if (!pdata->use_eeprom) { + ah->ah_flags &= ~AH_USE_EEPROM; + ah->gpio_mask = pdata->gpio_mask; + ah->gpio_val = pdata->gpio_val; + ah->led_pin = pdata->led_pin; + ah->is_clk_25mhz = pdata->is_clk_25mhz; + ah->get_mac_revision = pdata->get_mac_revision; + ah->external_reset = pdata->external_reset; + ah->disable_2ghz = pdata->disable_2ghz; + ah->disable_5ghz = pdata->disable_5ghz; + + if (!pdata->endian_check) + ah->ah_flags |= AH_NO_EEP_SWAP; + } + if (pdata->eeprom_name) { ret = ath9k_eeprom_request(sc, pdata->eeprom_name); if (ret) @@ -533,13 +549,15 @@ static int ath9k_init_soc_platform(struct ath_softc *sc) if (pdata->tx_gain_buffalo) ah->config.tx_gain_buffalo = true; - return ret; + if (pdata->macaddr) + ether_addr_copy(common->macaddr, pdata->macaddr); + + return 0; } static int ath9k_init_softc(u16 devid, struct ath_softc *sc, const struct ath_bus_ops *bus_ops) { - struct ath9k_platform_data *pdata = sc->dev->platform_data; struct ath_hw *ah = NULL; struct ath9k_hw_capabilities *pCap; struct ath_common *common; @@ -553,6 +571,8 @@ static int ath9k_init_softc(u16 devid, struct ath_softc *sc, ah->dev = sc->dev; ah->hw = sc->hw; ah->hw_version.devid = devid; + ah->ah_flags |= AH_USE_EEPROM; + ah->led_pin = -1; ah->reg_ops.read = ath9k_ioread32; ah->reg_ops.multi_read = ath9k_multi_ioread32; ah->reg_ops.write = ath9k_iowrite32; @@ -572,22 +592,6 @@ static int ath9k_init_softc(u16 devid, struct ath_softc *sc, if (!ath9k_is_chanctx_enabled()) sc->cur_chan->hw_queue_base = 0; - if (!pdata || pdata->use_eeprom) { - ah->ah_flags |= AH_USE_EEPROM; - sc->sc_ah->led_pin = -1; - } else { - sc->sc_ah->gpio_mask = pdata->gpio_mask; - sc->sc_ah->gpio_val = pdata->gpio_val; - sc->sc_ah->led_pin = pdata->led_pin; - ah->is_clk_25mhz = pdata->is_clk_25mhz; - ah->get_mac_revision = pdata->get_mac_revision; - ah->external_reset = pdata->external_reset; - ah->disable_2ghz = pdata->disable_2ghz; - ah->disable_5ghz = pdata->disable_5ghz; - if (!pdata->endian_check) - ah->ah_flags |= AH_NO_EEP_SWAP; - } - common->ops = &ah->reg_ops; common->bus_ops = bus_ops; common->ps_ops = &ath9k_ps_ops; @@ -603,7 +607,7 @@ static int ath9k_init_softc(u16 devid, struct ath_softc *sc, */ ath9k_init_pcoem_platform(sc); - ret = ath9k_init_soc_platform(sc); + ret = ath9k_init_platform(sc); if (ret) return ret; @@ -649,9 +653,6 @@ static int ath9k_init_softc(u16 devid, struct ath_softc *sc, if (ret) goto err_hw; - if (pdata && pdata->macaddr) - memcpy(common->macaddr, pdata->macaddr, ETH_ALEN); - ret = ath9k_init_queues(sc); if (ret) goto err_queues; -- cgit v0.10.2 From fa5106e1d79382fc3181524902a92b813e57a56d Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 4 Jul 2016 14:37:20 +0200 Subject: ath9k: Correct TSF adjustment to align the beacon time correctly Beacons were not send out at (timestamp % beacon_time == 0) for interfaces other than the primary one. To send out beacons with the correct timestamp according to 10.1.3.2 of the 802.11 standard the tsf_adjustment has to be set to the negative time difference instead of positive. This way the later beacons get corrected to have a lower (and similar) timestamp with regard to the beacon from slot 0. I am not aware about any issues that have been caused by this. Signed-off-by: Benjamin Berg Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/beacon.c b/drivers/net/wireless/ath/ath9k/beacon.c index 5cf0cd7..800c96b 100644 --- a/drivers/net/wireless/ath/ath9k/beacon.c +++ b/drivers/net/wireless/ath/ath9k/beacon.c @@ -279,17 +279,21 @@ static void ath9k_set_tsfadjust(struct ath_softc *sc, struct ieee80211_vif *vif) struct ath_common *common = ath9k_hw_common(sc->sc_ah); struct ath_vif *avp = (void *)vif->drv_priv; struct ath_beacon_config *cur_conf = &avp->chanctx->beacon; - u32 tsfadjust; + s64 tsfadjust; if (avp->av_bslot == 0) return; + /* tsf_adjust is added to the TSF value. We send out the beacon late, + * so need to adjust the TSF starting point to be later in time (i.e. + * the theoretical first beacon has a TSF of 0 after correction). + */ tsfadjust = cur_conf->beacon_interval * avp->av_bslot; - tsfadjust = TU_TO_USEC(tsfadjust) / ATH_BCBUF; + tsfadjust = -TU_TO_USEC(tsfadjust) / ATH_BCBUF; avp->tsf_adjust = cpu_to_le64(tsfadjust); - ath_dbg(common, CONFIG, "tsfadjust is: %llu for bslot: %d\n", - (unsigned long long)tsfadjust, avp->av_bslot); + ath_dbg(common, CONFIG, "tsfadjust is: %lld for bslot: %d\n", + (signed long long)tsfadjust, avp->av_bslot); } bool ath9k_csa_is_finished(struct ath_softc *sc, struct ieee80211_vif *vif) -- cgit v0.10.2 From 9580cb889f35c1ce98fcfec85ebc50b7979f0ecd Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 4 Jul 2016 14:37:21 +0200 Subject: ath9k: Handle channel context in get_/set_/reset_tsf The ath9k TSF handling routines need to be aware of the channel context that is being modified. With this change the TSF related values that are stored in each channel context will be correctly tracked and the harware will only be updated if the modified context is currently the active one. Without this change the TSF modifications done using these routines would for example be lost during a hardware reset as done by ath_complete_reset. Signed-off-by: Benjamin Berg Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c index 8b63988..375c2ac 100644 --- a/drivers/net/wireless/ath/ath9k/main.c +++ b/drivers/net/wireless/ath/ath9k/main.c @@ -1823,11 +1823,18 @@ static void ath9k_bss_info_changed(struct ieee80211_hw *hw, static u64 ath9k_get_tsf(struct ieee80211_hw *hw, struct ieee80211_vif *vif) { struct ath_softc *sc = hw->priv; + struct ath_vif *avp = (void *)vif->drv_priv; u64 tsf; mutex_lock(&sc->mutex); ath9k_ps_wakeup(sc); - tsf = ath9k_hw_gettsf64(sc->sc_ah); + /* Get current TSF either from HW or kernel time. */ + if (sc->cur_chan == avp->chanctx) { + tsf = ath9k_hw_gettsf64(sc->sc_ah); + } else { + tsf = sc->cur_chan->tsf_val + + ath9k_hw_get_tsf_offset(&sc->cur_chan->tsf_ts, NULL); + } ath9k_ps_restore(sc); mutex_unlock(&sc->mutex); @@ -1839,10 +1846,14 @@ static void ath9k_set_tsf(struct ieee80211_hw *hw, u64 tsf) { struct ath_softc *sc = hw->priv; + struct ath_vif *avp = (void *)vif->drv_priv; mutex_lock(&sc->mutex); ath9k_ps_wakeup(sc); - ath9k_hw_settsf64(sc->sc_ah, tsf); + getrawmonotonic(&avp->chanctx->tsf_ts); + if (sc->cur_chan == avp->chanctx) + ath9k_hw_settsf64(sc->sc_ah, tsf); + avp->chanctx->tsf_val = tsf; ath9k_ps_restore(sc); mutex_unlock(&sc->mutex); } @@ -1850,11 +1861,15 @@ static void ath9k_set_tsf(struct ieee80211_hw *hw, static void ath9k_reset_tsf(struct ieee80211_hw *hw, struct ieee80211_vif *vif) { struct ath_softc *sc = hw->priv; + struct ath_vif *avp = (void *)vif->drv_priv; mutex_lock(&sc->mutex); ath9k_ps_wakeup(sc); - ath9k_hw_reset_tsf(sc->sc_ah); + getrawmonotonic(&avp->chanctx->tsf_ts); + if (sc->cur_chan == avp->chanctx) + ath9k_hw_reset_tsf(sc->sc_ah); + avp->chanctx->tsf_val = 0; ath9k_ps_restore(sc); mutex_unlock(&sc->mutex); -- cgit v0.10.2 From bec9a94bc4e267a7d9be6f255cd90fe87a2259f6 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 4 Jul 2016 14:37:22 +0200 Subject: ath9k: Use tsf offset helper in ath9k_hw_reset These changes make ath9k_hw_reset more consistent with other places that handle the TSF value by using the same helper routine. A slight improvement is to not assume that a fixed time of 1.5ms has passed for the initval writes when compared to the first write attempt. Instead the TSF value is re-calculated which will yield a higher accuracy of the restored TSF timer. Signed-off-by: Benjamin Berg Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c index fa59117..a2aebf6 100644 --- a/drivers/net/wireless/ath/ath9k/hw.c +++ b/drivers/net/wireless/ath/ath9k/hw.c @@ -1828,8 +1828,9 @@ int ath9k_hw_reset(struct ath_hw *ah, struct ath9k_channel *chan, u32 saveLedState; u32 saveDefAntenna; u32 macStaId1; + struct timespec tsf_ts; + u32 tsf_offset; u64 tsf = 0; - s64 usec = 0; int r; bool start_mci_reset = false; bool save_fullsleep = ah->chip_fullsleep; @@ -1873,8 +1874,8 @@ int ath9k_hw_reset(struct ath_hw *ah, struct ath9k_channel *chan, macStaId1 = REG_READ(ah, AR_STA_ID1) & AR_STA_ID1_BASE_RATE_11B; /* Save TSF before chip reset, a cold reset clears it */ + getrawmonotonic(&tsf_ts); tsf = ath9k_hw_gettsf64(ah); - usec = ktime_to_us(ktime_get_raw()); saveLedState = REG_READ(ah, AR_CFG_LED) & (AR_CFG_LED_ASSOC_CTL | AR_CFG_LED_MODE_SEL | @@ -1907,8 +1908,8 @@ int ath9k_hw_reset(struct ath_hw *ah, struct ath9k_channel *chan, } /* Restore TSF */ - usec = ktime_to_us(ktime_get_raw()) - usec; - ath9k_hw_settsf64(ah, tsf + usec); + tsf_offset = ath9k_hw_get_tsf_offset(&tsf_ts, NULL); + ath9k_hw_settsf64(ah, tsf + tsf_offset); if (AR_SREV_9280_20_OR_LATER(ah)) REG_SET_BIT(ah, AR_GPIO_INPUT_EN_VAL, AR_GPIO_JTAG_DISABLE); @@ -1928,12 +1929,11 @@ int ath9k_hw_reset(struct ath_hw *ah, struct ath9k_channel *chan, /* * Some AR91xx SoC devices frequently fail to accept TSF writes * right after the chip reset. When that happens, write a new - * value after the initvals have been applied, with an offset - * based on measured time difference + * value after the initvals have been applied. */ if (AR_SREV_9100(ah) && (ath9k_hw_gettsf64(ah) < tsf)) { - tsf += 1500; - ath9k_hw_settsf64(ah, tsf); + tsf_offset = ath9k_hw_get_tsf_offset(&tsf_ts, NULL); + ath9k_hw_settsf64(ah, tsf + tsf_offset); } ath9k_hw_init_mfp(ah); -- cgit v0.10.2 From 7fde51227aed97984bc3a8572b60939d393fc2cb Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 4 Jul 2016 14:37:23 +0200 Subject: ath9k: Expose tsf_adjustment in mac80211 tsf getters and setters. The ath9k driver modifies the TSF for VIFs for the purpose of sending beacons in a staggered fashion. This patch exposes this VIF specific adjustment of the TSF value to mac80211. Without the change the TSF routines handle the hardware TSF value instead of the actual TSF value as seen on the air. Signed-off-by: Benjamin Berg Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c index 375c2ac..f2ebc85 100644 --- a/drivers/net/wireless/ath/ath9k/main.c +++ b/drivers/net/wireless/ath/ath9k/main.c @@ -1835,6 +1835,7 @@ static u64 ath9k_get_tsf(struct ieee80211_hw *hw, struct ieee80211_vif *vif) tsf = sc->cur_chan->tsf_val + ath9k_hw_get_tsf_offset(&sc->cur_chan->tsf_ts, NULL); } + tsf += le64_to_cpu(avp->tsf_adjust); ath9k_ps_restore(sc); mutex_unlock(&sc->mutex); @@ -1850,6 +1851,7 @@ static void ath9k_set_tsf(struct ieee80211_hw *hw, mutex_lock(&sc->mutex); ath9k_ps_wakeup(sc); + tsf -= le64_to_cpu(avp->tsf_adjust); getrawmonotonic(&avp->chanctx->tsf_ts); if (sc->cur_chan == avp->chanctx) ath9k_hw_settsf64(sc->sc_ah, tsf); -- cgit v0.10.2 From 11b0ac2e0c1c943c71fd89a6029a3995a0ca7e76 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 4 Jul 2016 14:37:24 +0200 Subject: ath9k: Remove some #defined constants to decrease verbosity The removed ATH9K_SLOT_TIME_X constants simply map the value in microseconds to the same integer. These constants were not used consistently, so fix the inconsistency issue by replacing all occurances with the integer equivalent. Signed-off-by: Benjamin Berg Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/beacon.c b/drivers/net/wireless/ath/ath9k/beacon.c index 800c96b..d8b4971 100644 --- a/drivers/net/wireless/ath/ath9k/beacon.c +++ b/drivers/net/wireless/ath/ath9k/beacon.c @@ -50,7 +50,7 @@ static void ath9k_beaconq_config(struct ath_softc *sc) txq = sc->tx.txq_map[IEEE80211_AC_BE]; ath9k_hw_get_txq_props(ah, txq->axq_qnum, &qi_be); qi.tqi_aifs = qi_be.tqi_aifs; - if (ah->slottime == ATH9K_SLOT_TIME_20) + if (ah->slottime == 20) qi.tqi_cwmin = 2*qi_be.tqi_cwmin; else qi.tqi_cwmin = 4*qi_be.tqi_cwmin; diff --git a/drivers/net/wireless/ath/ath9k/dynack.c b/drivers/net/wireless/ath/ath9k/dynack.c index d2ff0fc..7334c9b0 100644 --- a/drivers/net/wireless/ath/ath9k/dynack.c +++ b/drivers/net/wireless/ath/ath9k/dynack.c @@ -280,7 +280,7 @@ EXPORT_SYMBOL(ath_dynack_sample_ack_ts); void ath_dynack_node_init(struct ath_hw *ah, struct ath_node *an) { /* ackto = slottime + sifs + air delay */ - u32 ackto = ATH9K_SLOT_TIME_9 + 16 + 64; + u32 ackto = 9 + 16 + 64; struct ath_dynack *da = &ah->dynack; an->ackto = ackto; @@ -315,7 +315,7 @@ EXPORT_SYMBOL(ath_dynack_node_deinit); void ath_dynack_reset(struct ath_hw *ah) { /* ackto = slottime + sifs + air delay */ - u32 ackto = ATH9K_SLOT_TIME_9 + 16 + 64; + u32 ackto = 9 + 16 + 64; struct ath_dynack *da = &ah->dynack; da->lto = jiffies; diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_beacon.c b/drivers/net/wireless/ath/ath9k/htc_drv_beacon.c index e6bcb4c..2c0e4d2 100644 --- a/drivers/net/wireless/ath/ath9k/htc_drv_beacon.c +++ b/drivers/net/wireless/ath/ath9k/htc_drv_beacon.c @@ -45,7 +45,7 @@ void ath9k_htc_beaconq_config(struct ath9k_htc_priv *priv) * Long slot time : 2x cwmin * Short slot time : 4x cwmin */ - if (ah->slottime == ATH9K_SLOT_TIME_20) + if (ah->slottime == 20) qi.tqi_cwmin = 2*qi_be.tqi_cwmin; else qi.tqi_cwmin = 4*qi_be.tqi_cwmin; diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_init.c b/drivers/net/wireless/ath/ath9k/htc_drv_init.c index c148c6c..b65c1b6 100644 --- a/drivers/net/wireless/ath/ath9k/htc_drv_init.c +++ b/drivers/net/wireless/ath/ath9k/htc_drv_init.c @@ -678,7 +678,7 @@ static int ath9k_init_priv(struct ath9k_htc_priv *priv, for (i = 0; i < ATH9K_HTC_MAX_BCN_VIF; i++) priv->beacon.bslot[i] = NULL; - priv->beacon.slottime = ATH9K_SLOT_TIME_9; + priv->beacon.slottime = 9; ath9k_cmn_init_channels_rates(common); ath9k_cmn_init_crypto(ah); diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c index a2aebf6..d1d0c06 100644 --- a/drivers/net/wireless/ath/ath9k/hw.c +++ b/drivers/net/wireless/ath/ath9k/hw.c @@ -454,7 +454,7 @@ static void ath9k_hw_init_defaults(struct ath_hw *ah) if (AR_SREV_9100(ah)) ah->sta_id1_defaults |= AR_STA_ID1_AR9100_BA_FIX; - ah->slottime = ATH9K_SLOT_TIME_9; + ah->slottime = 9; ah->globaltxtimeout = (u32) -1; ah->power_mode = ATH9K_PM_UNDEFINED; ah->htc_reset_init = true; diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c index a0f4a52..edc74fc 100644 --- a/drivers/net/wireless/ath/ath9k/init.c +++ b/drivers/net/wireless/ath/ath9k/init.c @@ -372,7 +372,7 @@ static void ath9k_init_misc(struct ath_softc *sc) common->last_rssi = ATH_RSSI_DUMMY_MARKER; memcpy(common->bssidmask, ath_bcast_mac, ETH_ALEN); - sc->beacon.slottime = ATH9K_SLOT_TIME_9; + sc->beacon.slottime = 9; for (i = 0; i < ARRAY_SIZE(sc->beacon.bslot); i++) sc->beacon.bslot[i] = NULL; diff --git a/drivers/net/wireless/ath/ath9k/mac.h b/drivers/net/wireless/ath/ath9k/mac.h index 7fbf7f9..3bab014 100644 --- a/drivers/net/wireless/ath/ath9k/mac.h +++ b/drivers/net/wireless/ath/ath9k/mac.h @@ -65,10 +65,6 @@ #define INIT_SSH_RETRY 32 #define INIT_SLG_RETRY 32 -#define ATH9K_SLOT_TIME_6 6 -#define ATH9K_SLOT_TIME_9 9 -#define ATH9K_SLOT_TIME_20 20 - #define ATH9K_TXERR_XRETRY 0x01 #define ATH9K_TXERR_FILT 0x02 #define ATH9K_TXERR_FIFO 0x04 diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c index f2ebc85..820df45 100644 --- a/drivers/net/wireless/ath/ath9k/main.c +++ b/drivers/net/wireless/ath/ath9k/main.c @@ -926,7 +926,7 @@ static void ath9k_vif_iter(struct ath9k_vif_iter_data *iter_data, } if (!vif->bss_conf.use_short_slot) - iter_data->slottime = ATH9K_SLOT_TIME_20; + iter_data->slottime = 20; switch (vif->type) { case NL80211_IFTYPE_AP: @@ -999,7 +999,7 @@ void ath9k_calculate_iter_data(struct ath_softc *sc, */ memset(iter_data, 0, sizeof(*iter_data)); eth_broadcast_addr(iter_data->mask); - iter_data->slottime = ATH9K_SLOT_TIME_9; + iter_data->slottime = 9; list_for_each_entry(avp, &ctx->vifs, list) ath9k_vif_iter(iter_data, avp->vif->addr, avp->vif); @@ -1061,7 +1061,7 @@ static void ath9k_set_offchannel_state(struct ath_softc *sc) ah->opmode = vif->type; ah->imask &= ~ATH9K_INT_SWBA; ah->imask &= ~ATH9K_INT_TSFOOR; - ah->slottime = ATH9K_SLOT_TIME_9; + ah->slottime = 9; ath_hw_setbssidmask(common); ath9k_hw_setopmode(ah); @@ -1788,6 +1788,7 @@ static void ath9k_bss_info_changed(struct ieee80211_hw *hw, slottime = 9; else slottime = 20; + if (vif->type == NL80211_IFTYPE_AP) { /* * Defer update, so that connected stations can adjust -- cgit v0.10.2 From cfda2d8e2314d195902653bdd689e23b287e7d45 Mon Sep 17 00:00:00 2001 From: Benjamin Berg Date: Mon, 4 Jul 2016 14:37:25 +0200 Subject: ath9k: Fix beacon configuration for addition/removal of interfaces This patch fixes some issues with interface reconfiguration. It could for example happen that an AP interface in beacon slot 0 was removed leaving an IBSS station in one of the other slots. When this happens the driver never sends out the beacon as it only tries to send a beacon from slot 0. Appart from that the tracking of required changes to the beacon config is relatively complicated and prone to errors. The approach taken here is to solve reconfiguration issues is to reconfigure the beacons when any interface changes. This means that the complexity of deciding whether an interface change may modify the beacon configuration is gone. It also means that the beacon config will be reliably updated when an interface is removed. The issue that a single non-AP interface might not be in beacon slot 0 and wouldn't be send out is solved by moving it into the first slot. The TSF value in hardware is adjusted accordingly so that the timestamp of the beacons stay consistent. Signed-off-by: Benjamin Berg Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/ath9k.h b/drivers/net/wireless/ath/ath9k/ath9k.h index 93b3793..26fc8ec 100644 --- a/drivers/net/wireless/ath/ath9k/ath9k.h +++ b/drivers/net/wireless/ath/ath9k/ath9k.h @@ -637,6 +637,8 @@ struct ath9k_vif_iter_data { int nwds; /* number of WDS vifs */ int nadhocs; /* number of adhoc vifs */ int nocbs; /* number of OCB vifs */ + int nbcnvifs; /* number of beaconing vifs */ + struct ieee80211_vif *primary_beacon_vif; struct ieee80211_vif *primary_sta; }; @@ -685,10 +687,11 @@ struct ath_beacon { }; void ath9k_beacon_tasklet(unsigned long data); -void ath9k_beacon_config(struct ath_softc *sc, struct ieee80211_vif *vif, - u32 changed); +void ath9k_beacon_config(struct ath_softc *sc, struct ieee80211_vif *main_vif, + bool beacons); void ath9k_beacon_assign_slot(struct ath_softc *sc, struct ieee80211_vif *vif); void ath9k_beacon_remove_slot(struct ath_softc *sc, struct ieee80211_vif *vif); +void ath9k_beacon_ensure_primary_slot(struct ath_softc *sc); void ath9k_set_beacon(struct ath_softc *sc); bool ath9k_csa_is_finished(struct ath_softc *sc, struct ieee80211_vif *vif); void ath9k_csa_update(struct ath_softc *sc); diff --git a/drivers/net/wireless/ath/ath9k/beacon.c b/drivers/net/wireless/ath/ath9k/beacon.c index d8b4971..e36f947 100644 --- a/drivers/net/wireless/ath/ath9k/beacon.c +++ b/drivers/net/wireless/ath/ath9k/beacon.c @@ -209,7 +209,6 @@ void ath9k_beacon_assign_slot(struct ath_softc *sc, struct ieee80211_vif *vif) } sc->beacon.bslot[avp->av_bslot] = vif; - sc->nbcnvifs++; ath_dbg(common, CONFIG, "Added interface at beacon slot: %d\n", avp->av_bslot); @@ -220,15 +219,12 @@ void ath9k_beacon_remove_slot(struct ath_softc *sc, struct ieee80211_vif *vif) struct ath_common *common = ath9k_hw_common(sc->sc_ah); struct ath_vif *avp = (void *)vif->drv_priv; struct ath_buf *bf = avp->av_bcbuf; - struct ath_beacon_config *cur_conf = &sc->cur_chan->beacon; ath_dbg(common, CONFIG, "Removing interface at beacon slot: %d\n", avp->av_bslot); tasklet_disable(&sc->bcon_tasklet); - cur_conf->enable_beacon &= ~BIT(avp->av_bslot); - if (bf && bf->bf_mpdu) { struct sk_buff *skb = bf->bf_mpdu; dma_unmap_single(sc->dev, bf->bf_buf_addr, @@ -240,12 +236,73 @@ void ath9k_beacon_remove_slot(struct ath_softc *sc, struct ieee80211_vif *vif) avp->av_bcbuf = NULL; sc->beacon.bslot[avp->av_bslot] = NULL; - sc->nbcnvifs--; list_add_tail(&bf->list, &sc->beacon.bbuf); tasklet_enable(&sc->bcon_tasklet); } +void ath9k_beacon_ensure_primary_slot(struct ath_softc *sc) +{ + struct ath_common *common = ath9k_hw_common(sc->sc_ah); + struct ieee80211_vif *vif; + struct ath_vif *avp; + s64 tsfadjust; + u32 offset; + int first_slot = ATH_BCBUF; + int slot; + + tasklet_disable(&sc->bcon_tasklet); + + /* Find first taken slot. */ + for (slot = 0; slot < ATH_BCBUF; slot++) { + if (sc->beacon.bslot[slot]) { + first_slot = slot; + break; + } + } + if (first_slot == 0) + goto out; + + /* Re-enumarate all slots, moving them forward. */ + for (slot = 0; slot < ATH_BCBUF; slot++) { + if (slot + first_slot < ATH_BCBUF) { + vif = sc->beacon.bslot[slot + first_slot]; + sc->beacon.bslot[slot] = vif; + + if (vif) { + avp = (void *)vif->drv_priv; + avp->av_bslot = slot; + } + } else { + sc->beacon.bslot[slot] = NULL; + } + } + + vif = sc->beacon.bslot[0]; + if (WARN_ON(!vif)) + goto out; + + /* Get the tsf_adjust value for the new first slot. */ + avp = (void *)vif->drv_priv; + tsfadjust = le64_to_cpu(avp->tsf_adjust); + + ath_dbg(common, CONFIG, + "Adjusting global TSF after beacon slot reassignment: %lld\n", + (signed long long)tsfadjust); + + /* Modify TSF as required and update the HW. */ + avp->chanctx->tsf_val += tsfadjust; + if (sc->cur_chan == avp->chanctx) { + offset = ath9k_hw_get_tsf_offset(&avp->chanctx->tsf_ts, NULL); + ath9k_hw_settsf64(sc->sc_ah, avp->chanctx->tsf_val + offset); + } + + /* The slots tsf_adjust will be updated by ath9k_beacon_config later. */ + +out: + tasklet_enable(&sc->bcon_tasklet); +} + static int ath9k_beacon_choose_slot(struct ath_softc *sc) { struct ath_common *common = ath9k_hw_common(sc->sc_ah); @@ -274,26 +331,33 @@ static int ath9k_beacon_choose_slot(struct ath_softc *sc) return slot; } -static void ath9k_set_tsfadjust(struct ath_softc *sc, struct ieee80211_vif *vif) +static void ath9k_set_tsfadjust(struct ath_softc *sc, + struct ath_beacon_config *cur_conf) { struct ath_common *common = ath9k_hw_common(sc->sc_ah); - struct ath_vif *avp = (void *)vif->drv_priv; - struct ath_beacon_config *cur_conf = &avp->chanctx->beacon; s64 tsfadjust; + int slot; - if (avp->av_bslot == 0) - return; + for (slot = 0; slot < ATH_BCBUF; slot++) { + struct ath_vif *avp; - /* tsf_adjust is added to the TSF value. We send out the beacon late, - * so need to adjust the TSF starting point to be later in time (i.e. - * the theoretical first beacon has a TSF of 0 after correction). - */ - tsfadjust = cur_conf->beacon_interval * avp->av_bslot; - tsfadjust = -TU_TO_USEC(tsfadjust) / ATH_BCBUF; - avp->tsf_adjust = cpu_to_le64(tsfadjust); + if (!sc->beacon.bslot[slot]) + continue; + + avp = (void *)sc->beacon.bslot[slot]->drv_priv; - ath_dbg(common, CONFIG, "tsfadjust is: %lld for bslot: %d\n", - (signed long long)tsfadjust, avp->av_bslot); + /* tsf_adjust is added to the TSF value. We send out the + * beacon late, so need to adjust the TSF starting point to be + * later in time (i.e. the theoretical first beacon has a TSF + * of 0 after correction). + */ + tsfadjust = cur_conf->beacon_interval * avp->av_bslot; + tsfadjust = -TU_TO_USEC(tsfadjust) / ATH_BCBUF; + avp->tsf_adjust = cpu_to_le64(tsfadjust); + + ath_dbg(common, CONFIG, "tsfadjust is: %lld for bslot: %d\n", + (signed long long)tsfadjust, avp->av_bslot); + } } bool ath9k_csa_is_finished(struct ath_softc *sc, struct ieee80211_vif *vif) @@ -447,20 +511,28 @@ void ath9k_beacon_tasklet(unsigned long data) * Both nexttbtt and intval have to be in usecs. */ static void ath9k_beacon_init(struct ath_softc *sc, u32 nexttbtt, - u32 intval, bool reset_tsf) + u32 intval) { struct ath_hw *ah = sc->sc_ah; ath9k_hw_disable_interrupts(ah); - if (reset_tsf) - ath9k_hw_reset_tsf(ah); ath9k_beaconq_config(sc); ath9k_hw_beaconinit(ah, nexttbtt, intval); + ah->imask |= ATH9K_INT_SWBA; sc->beacon.bmisscnt = 0; ath9k_hw_set_interrupts(ah); ath9k_hw_enable_interrupts(ah); } +static void ath9k_beacon_stop(struct ath_softc *sc) +{ + ath9k_hw_disable_interrupts(sc->sc_ah); + sc->sc_ah->imask &= ~(ATH9K_INT_SWBA | ATH9K_INT_BMISS); + sc->beacon.bmisscnt = 0; + ath9k_hw_set_interrupts(sc->sc_ah); + ath9k_hw_enable_interrupts(sc->sc_ah); +} + /* * For multi-bss ap support beacons are either staggered evenly over N slots or * burst together. For the former arrange for the SWBA to be delivered for each @@ -472,7 +544,7 @@ static void ath9k_beacon_config_ap(struct ath_softc *sc, struct ath_hw *ah = sc->sc_ah; ath9k_cmn_beacon_config_ap(ah, conf, ATH_BCBUF); - ath9k_beacon_init(sc, conf->nexttbtt, conf->intval, false); + ath9k_beacon_init(sc, conf->nexttbtt, conf->intval); } static void ath9k_beacon_config_sta(struct ath_hw *ah, @@ -501,7 +573,7 @@ static void ath9k_beacon_config_adhoc(struct ath_softc *sc, ath9k_cmn_beacon_config_adhoc(ah, conf); - ath9k_beacon_init(sc, conf->nexttbtt, conf->intval, conf->ibss_creator); + ath9k_beacon_init(sc, conf->nexttbtt, conf->intval); /* * Set the global 'beacon has been configured' flag for the @@ -511,44 +583,6 @@ static void ath9k_beacon_config_adhoc(struct ath_softc *sc, set_bit(ATH_OP_BEACONS, &common->op_flags); } -static bool ath9k_allow_beacon_config(struct ath_softc *sc, - struct ieee80211_vif *vif) -{ - struct ath_common *common = ath9k_hw_common(sc->sc_ah); - struct ath_vif *avp = (void *)vif->drv_priv; - - if (ath9k_is_chanctx_enabled()) { - /* - * If the VIF is not present in the current channel context, - * then we can't do the usual opmode checks. Allow the - * beacon config for the VIF to be updated in this case and - * return immediately. - */ - if (sc->cur_chan != avp->chanctx) - return true; - } - - if (sc->sc_ah->opmode == NL80211_IFTYPE_AP) { - if (vif->type != NL80211_IFTYPE_AP) { - ath_dbg(common, CONFIG, - "An AP interface is already present !\n"); - return false; - } - } - - if (sc->sc_ah->opmode == NL80211_IFTYPE_STATION) { - if ((vif->type == NL80211_IFTYPE_STATION) && - test_bit(ATH_OP_BEACONS, &common->op_flags) && - vif != sc->cur_chan->primary_sta) { - ath_dbg(common, CONFIG, - "Beacon already configured for a station interface\n"); - return false; - } - } - - return true; -} - static void ath9k_cache_beacon_config(struct ath_softc *sc, struct ath_chanctx *ctx, struct ieee80211_bss_conf *bss_conf) @@ -584,87 +618,79 @@ static void ath9k_cache_beacon_config(struct ath_softc *sc, if (cur_conf->dtim_period == 0) cur_conf->dtim_period = 1; + ath9k_set_tsfadjust(sc, cur_conf); } -void ath9k_beacon_config(struct ath_softc *sc, struct ieee80211_vif *vif, - u32 changed) +void ath9k_beacon_config(struct ath_softc *sc, struct ieee80211_vif *main_vif, + bool beacons) { - struct ieee80211_bss_conf *bss_conf = &vif->bss_conf; - struct ath_hw *ah = sc->sc_ah; - struct ath_common *common = ath9k_hw_common(ah); - struct ath_vif *avp = (void *)vif->drv_priv; - struct ath_chanctx *ctx = avp->chanctx; + struct ath_hw *ah = sc->sc_ah; + struct ath_common *common = ath9k_hw_common(ah); + struct ath_vif *avp; + struct ath_chanctx *ctx; struct ath_beacon_config *cur_conf; unsigned long flags; + bool enabled; bool skip_beacon = false; - if (!ctx) + if (!beacons) { + clear_bit(ATH_OP_BEACONS, &common->op_flags); + ath9k_beacon_stop(sc); return; + } - cur_conf = &avp->chanctx->beacon; - if (vif->type == NL80211_IFTYPE_AP) - ath9k_set_tsfadjust(sc, vif); - - if (!ath9k_allow_beacon_config(sc, vif)) + if (WARN_ON(!main_vif)) return; - if (vif->type == NL80211_IFTYPE_STATION) { - ath9k_cache_beacon_config(sc, ctx, bss_conf); - if (ctx != sc->cur_chan) - return; + avp = (void *)main_vif->drv_priv; + ctx = avp->chanctx; + cur_conf = &ctx->beacon; + enabled = cur_conf->enable_beacon; + cur_conf->enable_beacon = beacons; + + if (sc->sc_ah->opmode == NL80211_IFTYPE_STATION) { + ath9k_cache_beacon_config(sc, ctx, &main_vif->bss_conf); ath9k_set_beacon(sc); set_bit(ATH_OP_BEACONS, &common->op_flags); return; } - /* - * Take care of multiple interfaces when - * enabling/disabling SWBA. - */ - if (changed & BSS_CHANGED_BEACON_ENABLED) { - bool enabled = cur_conf->enable_beacon; - - if (!bss_conf->enable_beacon) { - cur_conf->enable_beacon &= ~BIT(avp->av_bslot); - } else { - cur_conf->enable_beacon |= BIT(avp->av_bslot); - if (!enabled) - ath9k_cache_beacon_config(sc, ctx, bss_conf); - } - } - - if (ctx != sc->cur_chan) - return; + /* Update the beacon configuration. */ + ath9k_cache_beacon_config(sc, ctx, &main_vif->bss_conf); /* * Configure the HW beacon registers only when we have a valid * beacon interval. */ if (cur_conf->beacon_interval) { - /* - * If we are joining an existing IBSS network, start beaconing - * only after a TSF-sync has taken place. Ensure that this - * happens by setting the appropriate flags. + /* Special case to sync the TSF when joining an existing IBSS. + * This is only done if no AP interface is active. + * Note that mac80211 always resets the TSF when creating a new + * IBSS interface. */ - if ((changed & BSS_CHANGED_IBSS) && !bss_conf->ibss_creator && - bss_conf->enable_beacon) { + if (sc->sc_ah->opmode == NL80211_IFTYPE_ADHOC && + !enabled && beacons && !main_vif->bss_conf.ibss_creator) { spin_lock_irqsave(&sc->sc_pm_lock, flags); sc->ps_flags |= PS_BEACON_SYNC | PS_WAIT_FOR_BEACON; spin_unlock_irqrestore(&sc->sc_pm_lock, flags); skip_beacon = true; - } else { - ath9k_set_beacon(sc); } /* * Do not set the ATH_OP_BEACONS flag for IBSS joiner mode * here, it is done in ath9k_beacon_config_adhoc(). */ - if (cur_conf->enable_beacon && !skip_beacon) + if (beacons && !skip_beacon) { set_bit(ATH_OP_BEACONS, &common->op_flags); - else + ath9k_set_beacon(sc); + } else { clear_bit(ATH_OP_BEACONS, &common->op_flags); + ath9k_beacon_stop(sc); + } + } else { + clear_bit(ATH_OP_BEACONS, &common->op_flags); + ath9k_beacon_stop(sc); } } diff --git a/drivers/net/wireless/ath/ath9k/common.h b/drivers/net/wireless/ath/ath9k/common.h index d237373..f0ab6f9 100644 --- a/drivers/net/wireless/ath/ath9k/common.h +++ b/drivers/net/wireless/ath/ath9k/common.h @@ -50,6 +50,7 @@ #define IEEE80211_MS_TO_TU(x) (((x) * 1000) / 1024) struct ath_beacon_config { + struct ieee80211_vif *main_vif; int beacon_interval; u16 dtim_period; u16 bmiss_timeout; diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c index 820df45..7594650 100644 --- a/drivers/net/wireless/ath/ath9k/main.c +++ b/drivers/net/wireless/ath/ath9k/main.c @@ -910,6 +910,22 @@ static bool ath9k_uses_beacons(int type) } } +static void ath9k_vif_iter_set_beacon(struct ath9k_vif_iter_data *iter_data, + struct ieee80211_vif *vif) +{ + /* Use the first (configured) interface, but prefering AP interfaces. */ + if (!iter_data->primary_beacon_vif) { + iter_data->primary_beacon_vif = vif; + } else { + if (iter_data->primary_beacon_vif->type != NL80211_IFTYPE_AP && + vif->type == NL80211_IFTYPE_AP) + iter_data->primary_beacon_vif = vif; + } + + iter_data->beacons = true; + iter_data->nbcnvifs += 1; +} + static void ath9k_vif_iter(struct ath9k_vif_iter_data *iter_data, u8 *mac, struct ieee80211_vif *vif) { @@ -931,6 +947,8 @@ static void ath9k_vif_iter(struct ath9k_vif_iter_data *iter_data, switch (vif->type) { case NL80211_IFTYPE_AP: iter_data->naps++; + if (vif->bss_conf.enable_beacon) + ath9k_vif_iter_set_beacon(iter_data, vif); break; case NL80211_IFTYPE_STATION: iter_data->nstations++; @@ -943,12 +961,12 @@ static void ath9k_vif_iter(struct ath9k_vif_iter_data *iter_data, case NL80211_IFTYPE_ADHOC: iter_data->nadhocs++; if (vif->bss_conf.enable_beacon) - iter_data->beacons = true; + ath9k_vif_iter_set_beacon(iter_data, vif); break; case NL80211_IFTYPE_MESH_POINT: iter_data->nmeshes++; if (vif->bss_conf.enable_beacon) - iter_data->beacons = true; + ath9k_vif_iter_set_beacon(iter_data, vif); break; case NL80211_IFTYPE_WDS: iter_data->nwds++; @@ -1081,7 +1099,6 @@ void ath9k_calculate_summary_state(struct ath_softc *sc, struct ath_hw *ah = sc->sc_ah; struct ath_common *common = ath9k_hw_common(ah); struct ath9k_vif_iter_data iter_data; - struct ath_beacon_config *cur_conf; ath_chanctx_check_active(sc, ctx); @@ -1103,13 +1120,12 @@ void ath9k_calculate_summary_state(struct ath_softc *sc, ath_hw_setbssidmask(common); if (iter_data.naps > 0) { - cur_conf = &ctx->beacon; ath9k_hw_set_tsfadjust(ah, true); ah->opmode = NL80211_IFTYPE_AP; - if (cur_conf->enable_beacon) - iter_data.beacons = true; } else { ath9k_hw_set_tsfadjust(ah, false); + if (iter_data.beacons) + ath9k_beacon_ensure_primary_slot(sc); if (iter_data.nmeshes) ah->opmode = NL80211_IFTYPE_MESH_POINT; @@ -1134,7 +1150,6 @@ void ath9k_calculate_summary_state(struct ath_softc *sc, ctx->switch_after_beacon = true; } - ah->imask &= ~ATH9K_INT_SWBA; if (ah->opmode == NL80211_IFTYPE_STATION) { bool changed = (iter_data.primary_sta != ctx->primary_sta); @@ -1151,16 +1166,12 @@ void ath9k_calculate_summary_state(struct ath_softc *sc, if (ath9k_hw_mci_is_enabled(sc->sc_ah)) ath9k_mci_update_wlan_channels(sc, true); } - } else if (iter_data.beacons) { - ah->imask |= ATH9K_INT_SWBA; } + sc->nbcnvifs = iter_data.nbcnvifs; + ath9k_beacon_config(sc, iter_data.primary_beacon_vif, + iter_data.beacons); ath9k_hw_set_interrupts(ah); - if (iter_data.beacons) - set_bit(ATH_OP_BEACONS, &common->op_flags); - else - clear_bit(ATH_OP_BEACONS, &common->op_flags); - if (ah->slottime != iter_data.slottime) { ah->slottime = iter_data.slottime; ath9k_hw_init_global_settings(ah); @@ -1777,9 +1788,7 @@ static void ath9k_bss_info_changed(struct ieee80211_hw *hw, if ((changed & BSS_CHANGED_BEACON_ENABLED) || (changed & BSS_CHANGED_BEACON_INT) || (changed & BSS_CHANGED_BEACON_INFO)) { - ath9k_beacon_config(sc, vif, changed); - if (changed & BSS_CHANGED_BEACON_ENABLED) - ath9k_calculate_summary_state(sc, avp->chanctx); + ath9k_calculate_summary_state(sc, avp->chanctx); } if ((avp->chanctx == sc->cur_chan) && -- cgit v0.10.2 From 7860eb7537a0a308664da34e71d302cbb5130626 Mon Sep 17 00:00:00 2001 From: Bjorn Andersson Date: Thu, 7 Jul 2016 07:48:53 -0700 Subject: wcn36xx: Silence error about unsupported smd event 188 Sometimes the firmware sends a HAL_DEL_BA_IND, the prima driver silently ignore this message so let's do the same to silence the error message. Cc: Nicolas Dechesne Signed-off-by: Bjorn Andersson Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/wcn36xx/smd.c b/drivers/net/wireless/ath/wcn36xx/smd.c index feebaae..a443992 100644 --- a/drivers/net/wireless/ath/wcn36xx/smd.c +++ b/drivers/net/wireless/ath/wcn36xx/smd.c @@ -2229,6 +2229,7 @@ static void wcn36xx_smd_rsp_process(struct wcn36xx *wcn, void *buf, size_t len) case WCN36XX_HAL_COEX_IND: case WCN36XX_HAL_AVOID_FREQ_RANGE_IND: + case WCN36XX_HAL_DEL_BA_IND: case WCN36XX_HAL_OTA_TX_COMPL_IND: case WCN36XX_HAL_MISSED_BEACON_IND: case WCN36XX_HAL_DELETE_STA_CONTEXT_IND: @@ -2275,6 +2276,7 @@ static void wcn36xx_ind_smd_work(struct work_struct *work) switch (msg_header->msg_type) { case WCN36XX_HAL_COEX_IND: + case WCN36XX_HAL_DEL_BA_IND: case WCN36XX_HAL_AVOID_FREQ_RANGE_IND: break; case WCN36XX_HAL_OTA_TX_COMPL_IND: -- cgit v0.10.2 From c11e99396d0f77b5a6adc25f7c7bd84750194724 Mon Sep 17 00:00:00 2001 From: Vasanthakumar Thiagarajan Date: Fri, 1 Jul 2016 16:37:25 +0530 Subject: ath10k: fix possible wrong rx_busy time reporting in QCA4019 As hw cycle counters in QCA4019 wraparound independantly in QCA4019 it is possible cycle counter and rx clear counter would wraparound at the same time. Current logic assumes only one of the counters would wraparound at anytime. Fix this by moving 'else' part to another 'if'. Fixes: 8e100354a98 ("ath10k: fix cycle counter wraparound handling for QCA4019") Signed-off-by: Vasanthakumar Thiagarajan Reviewed-by: Julian Calaby Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/hw.c b/drivers/net/wireless/ath/ath10k/hw.c index f1e0695..f903d46 100644 --- a/drivers/net/wireless/ath/ath10k/hw.c +++ b/drivers/net/wireless/ath/ath10k/hw.c @@ -204,7 +204,8 @@ void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey, case ATH10K_HW_CC_WRAP_SHIFTED_EACH: if (cc < cc_prev) cc_fix = 0x7fffffff; - else + + if (rcc < rcc_prev) rcc_fix = 0x7fffffff; break; case ATH10K_HW_CC_WRAP_DISABLED: -- cgit v0.10.2 From bba7eb5d9b4ebccacd30331121ee29764212a29d Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Mon, 4 Jul 2016 16:22:20 +0200 Subject: hfsc: reduce hfsc_sched to 14 cachelines hfsc_sched is huge (size: 920, cachelines: 15), but we can get it to 14 cachelines by placing level after filter_cnt (covering 4 byte hole) and reducing period/nactive/flags to u32 (period is just a counter, incremented when class becomes active -- 2**32 is plenty for this purpose, also, long is only 32bit wide on 32bit platforms anyway). cl_vtperiod is exported to userspace via tc_hfsc_stats, but its period member is already u32, so no precision is lost there either. Cc: Michal Soltys Signed-off-by: Florian Westphal Signed-off-by: David S. Miller diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index dff92ea..3ddc7bd 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -115,9 +115,9 @@ struct hfsc_class { struct gnet_stats_basic_packed bstats; struct gnet_stats_queue qstats; struct gnet_stats_rate_est64 rate_est; - unsigned int level; /* class level in hierarchy */ struct tcf_proto __rcu *filter_list; /* filter list */ unsigned int filter_cnt; /* filter count */ + unsigned int level; /* class level in hierarchy */ struct hfsc_sched *sched; /* scheduler data */ struct hfsc_class *cl_parent; /* parent class */ @@ -165,10 +165,10 @@ struct hfsc_class { struct runtime_sc cl_virtual; /* virtual curve */ struct runtime_sc cl_ulimit; /* upperlimit curve */ - unsigned long cl_flags; /* which curves are valid */ - unsigned long cl_vtperiod; /* vt period sequence number */ - unsigned long cl_parentperiod;/* parent's vt period sequence number*/ - unsigned long cl_nactive; /* number of active children */ + u8 cl_flags; /* which curves are valid */ + u32 cl_vtperiod; /* vt period sequence number */ + u32 cl_parentperiod;/* parent's vt period sequence number*/ + u32 cl_nactive; /* number of active children */ }; struct hfsc_sched { -- cgit v0.10.2 From 99a50bb11cad44cd1d478256d2388e7afce982ac Mon Sep 17 00:00:00 2001 From: "K. Y. Srinivasan" Date: Tue, 5 Jul 2016 16:52:46 -0700 Subject: netvsc: Use the new in-place consumption APIs in the rx path Use the new APIs for eliminating a copy on the receive path. These new APIs also help in minimizing the number of memory barriers we end up issuing (in the ringbuffer code) since we can better control when we want to expose the ring state to the host. The patch is being resent to address earlier email issues. Signed-off-by: K. Y. Srinivasan Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c index 6909c32..20e0917 100644 --- a/drivers/net/hyperv/netvsc.c +++ b/drivers/net/hyperv/netvsc.c @@ -1128,6 +1128,39 @@ static inline void netvsc_receive_inband(struct hv_device *hdev, } } +static void netvsc_process_raw_pkt(struct hv_device *device, + struct vmbus_channel *channel, + struct netvsc_device *net_device, + struct net_device *ndev, + u64 request_id, + struct vmpacket_descriptor *desc) +{ + struct nvsp_message *nvmsg; + + nvmsg = (struct nvsp_message *)((unsigned long) + desc + (desc->offset8 << 3)); + + switch (desc->type) { + case VM_PKT_COMP: + netvsc_send_completion(net_device, channel, device, desc); + break; + + case VM_PKT_DATA_USING_XFER_PAGES: + netvsc_receive(net_device, channel, device, desc); + break; + + case VM_PKT_DATA_INBAND: + netvsc_receive_inband(device, net_device, nvmsg); + break; + + default: + netdev_err(ndev, "unhandled packet type %d, tid %llx\n", + desc->type, request_id); + break; + } +} + + void netvsc_channel_cb(void *context) { int ret; @@ -1140,7 +1173,7 @@ void netvsc_channel_cb(void *context) unsigned char *buffer; int bufferlen = NETVSC_PACKET_SIZE; struct net_device *ndev; - struct nvsp_message *nvmsg; + bool need_to_commit = false; if (channel->primary_channel != NULL) device = channel->primary_channel->device_obj; @@ -1154,39 +1187,36 @@ void netvsc_channel_cb(void *context) buffer = get_per_channel_state(channel); do { + desc = get_next_pkt_raw(channel); + if (desc != NULL) { + netvsc_process_raw_pkt(device, + channel, + net_device, + ndev, + desc->trans_id, + desc); + + put_pkt_raw(channel, desc); + need_to_commit = true; + continue; + } + if (need_to_commit) { + need_to_commit = false; + commit_rd_index(channel); + } + ret = vmbus_recvpacket_raw(channel, buffer, bufferlen, &bytes_recvd, &request_id); if (ret == 0) { if (bytes_recvd > 0) { desc = (struct vmpacket_descriptor *)buffer; - nvmsg = (struct nvsp_message *)((unsigned long) - desc + (desc->offset8 << 3)); - switch (desc->type) { - case VM_PKT_COMP: - netvsc_send_completion(net_device, - channel, - device, desc); - break; - - case VM_PKT_DATA_USING_XFER_PAGES: - netvsc_receive(net_device, channel, - device, desc); - break; - - case VM_PKT_DATA_INBAND: - netvsc_receive_inband(device, - net_device, - nvmsg); - break; - - default: - netdev_err(ndev, - "unhandled packet type %d, " - "tid %llx len %d\n", - desc->type, request_id, - bytes_recvd); - break; - } + netvsc_process_raw_pkt(device, + channel, + net_device, + ndev, + request_id, + desc); + } else { /* -- cgit v0.10.2 From 86dfb4acb378cb3a4eede3db919604c583beaa7c Mon Sep 17 00:00:00 2001 From: Craig Gallek Date: Wed, 6 Jul 2016 18:44:20 -0400 Subject: tun: Don't assume type tun in tun_device_event The referenced change added a netlink notifier for processing device queue size events. These events are fired for all devices but the registered callback assumed they only occurred for tun devices. This fix adds a check (borrowed from macvtap.c) to discard non-tun device events. For reference, this fixes the following splat: [ 71.505935] BUG: unable to handle kernel NULL pointer dereference at 0000000000000010 [ 71.513870] IP: [] tun_device_event+0x110/0x340 [ 71.519906] PGD 3f41f56067 PUD 3f264b7067 PMD 0 [ 71.524497] Oops: 0002 [#1] SMP DEBUG_PAGEALLOC [ 71.529374] gsmi: Log Shutdown Reason 0x03 [ 71.533417] Modules linked in:[ 71.533826] mlx4_en: eth1: Link Up [ 71.539616] bonding w1_therm wire cdc_acm ehci_pci ehci_hcd mlx4_en ib_uverbs mlx4_ib ib_core mlx4_core [ 71.549282] CPU: 12 PID: 7915 Comm: set.ixion-haswe Not tainted 4.7.0-dbx-DEV #8 [ 71.556586] Hardware name: Intel Grantley,Wellsburg/Ixion_IT_15, BIOS 2.58.0 05/03/2016 [ 71.564495] task: ffff887f00bb20c0 ti: ffff887f00798000 task.ti: ffff887f00798000 [ 71.571894] RIP: 0010:[] [] tun_device_event+0x110/0x340 [ 71.580327] RSP: 0018:ffff887f0079bbd8 EFLAGS: 00010202 [ 71.585576] RAX: fffffffffffffae8 RBX: ffff887ef6d03378 RCX: 0000000000000000 [ 71.592624] RDX: 0000000000000000 RSI: 0000000000000028 RDI: 0000000000000000 [ 71.599675] RBP: ffff887f0079bc48 R08: 0000000000000000 R09: 0000000000000001 [ 71.606730] R10: 0000000000000004 R11: 0000000000000000 R12: 0000000000000010 [ 71.613780] R13: 0000000000000000 R14: 0000000000000001 R15: ffff887f0079bd00 [ 71.620832] FS: 00007f5cdc581700(0000) GS:ffff883f7f700000(0000) knlGS:0000000000000000 [ 71.628826] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 71.634500] CR2: 0000000000000010 CR3: 0000003f3eb62000 CR4: 00000000001406e0 [ 71.641549] Stack: [ 71.643533] ffff887f0079bc08 0000000000000246 000000000000001e ffff887ef6d00000 [ 71.650871] ffff887f0079bd00 0000000000000000 0000000000000000 ffffffff00000000 [ 71.658210] ffff887f0079bc48 ffffffff81d24070 00000000fffffff9 ffffffff81cec7a0 [ 71.665549] Call Trace: [ 71.667975] [] notifier_call_chain+0x5d/0x80 [ 71.673823] [] ? show_tx_maxrate+0x30/0x30 [ 71.679502] [] __raw_notifier_call_chain+0xe/0x10 [ 71.685778] [] raw_notifier_call_chain+0x16/0x20 [ 71.691976] [] call_netdevice_notifiers_info+0x40/0x70 [ 71.698681] [] call_netdevice_notifiers+0x16/0x20 [ 71.704956] [] change_tx_queue_len+0x66/0x90 [ 71.710807] [] netdev_store.isra.5+0xbf/0xd0 [ 71.716658] [] tx_queue_len_store+0x50/0x60 [ 71.722431] [] dev_attr_store+0x18/0x30 [ 71.727857] [] sysfs_kf_write+0x4f/0x70 [ 71.733274] [] kernfs_fop_write+0x147/0x1d0 [ 71.739045] [] ? rcu_read_lock_sched_held+0x8f/0xa0 [ 71.745499] [] __vfs_write+0x28/0x120 [ 71.750748] [] ? percpu_down_read+0x57/0x90 [ 71.756516] [] ? __sb_start_write+0xc8/0xe0 [ 71.762278] [] ? __sb_start_write+0xc8/0xe0 [ 71.768038] [] vfs_write+0xbe/0x1b0 [ 71.773113] [] SyS_write+0x52/0xa0 [ 71.778110] [] entry_SYSCALL_64_fastpath+0x18/0xa8 [ 71.784472] Code: 45 31 f6 48 8b 93 78 33 00 00 48 81 c3 78 33 00 00 48 39 d3 48 8d 82 e8 fa ff ff 74 25 48 8d b0 40 05 00 00 49 63 d6 41 83 c6 01 <49> 89 34 d4 48 8b 90 18 05 00 00 48 39 d3 48 8d 82 e8 fa ff ff [ 71.803655] RIP [] tun_device_event+0x110/0x340 [ 71.809769] RSP [ 71.813213] CR2: 0000000000000010 [ 71.816512] ---[ end trace 4db6449606319f73 ]--- Fixes: 1576d9860599 ("tun: switch to use skb array for tx") Signed-off-by: Craig Gallek Acked-by: Jason Wang Signed-off-by: David S. Miller diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 5eadb7a..9c8b5bc 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -2503,6 +2503,9 @@ static int tun_device_event(struct notifier_block *unused, struct net_device *dev = netdev_notifier_info_to_dev(ptr); struct tun_struct *tun = netdev_priv(dev); + if (dev->rtnl_link_ops != &tun_link_ops) + return NOTIFY_DONE; + switch (event) { case NETDEV_CHANGE_TX_QUEUE_LEN: if (tun_queue_resize(tun)) -- cgit v0.10.2 From d390238c4fba7c87a3bcd859ce3373c864eb7b02 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Wed, 6 Jul 2016 20:03:54 -0400 Subject: net: dsa: initialize the routing table The routing table of every switch in a tree is currently initialized to all zeros. This is an issue since 0 is a valid port number. Add a DSA_RTABLE_NONE=-1 constant to initialize the signed values of the routing table pointing to other switches. This fixes the device mapping of the mv88e6xxx driver where the port pointing to the switch itself and to non-existent switches was wrongly configured to be 0. It is now set to the expected 0xf value. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/include/net/dsa.h b/include/net/dsa.h index 20b3087..52ab18b 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -32,6 +32,8 @@ enum dsa_tag_protocol { #define DSA_MAX_SWITCHES 4 #define DSA_MAX_PORTS 12 +#define DSA_RTABLE_NONE -1 + struct dsa_chip_data { /* * How to access the switch configuration registers. diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 766d2a5..7e68bc6 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -774,11 +774,17 @@ static int dsa_of_probe(struct device *dev) chip_index = -1; for_each_available_child_of_node(np, child) { + int i; + chip_index++; cd = &pd->chip[chip_index]; cd->of_node = child; + /* Initialize the routing table */ + for (i = 0; i < DSA_MAX_SWITCHES; ++i) + cd->rtable[i] = DSA_RTABLE_NONE; + /* When assigning the host device, increment its refcount */ cd->host_dev = get_device(&mdio_bus->dev); diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 83b95fc..78e4c01 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -595,7 +595,7 @@ static int _dsa_register_switch(struct dsa_switch *ds, struct device_node *np) struct device_node *ports = dsa_get_ports(ds, np); struct dsa_switch_tree *dst; u32 tree, index; - int err; + int i, err; err = dsa_parse_member(np, &tree, &index); if (err) @@ -622,6 +622,11 @@ static int _dsa_register_switch(struct dsa_switch *ds, struct device_node *np) ds->dst = dst; ds->index = index; + + /* Initialize the routing table */ + for (i = 0; i < DSA_MAX_SWITCHES; ++i) + ds->rtable[i] = DSA_RTABLE_NONE; + dsa_dst_add_ds(dst, ds, index); err = dsa_dst_complete(dst); -- cgit v0.10.2 From 606274c5abd8e245add01bc7145a8cbb92b69ba8 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Wed, 6 Jul 2016 22:38:36 -0700 Subject: bpf: introduce bpf_get_current_task() helper over time there were multiple requests to access different data structures and fields of task_struct current, so finally add the helper to access 'current' as-is. Tracing bpf programs will do the rest of walking the pointers via bpf_probe_read(). Note that current can be null and bpf program has to deal it with, but even dumb passing null into bpf_probe_read() is still safe. Suggested-by: Brendan Gregg Signed-off-by: Alexei Starovoitov Acked-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index c14ca1c..262a7e8 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -357,6 +357,13 @@ enum bpf_func_id { */ BPF_FUNC_get_hash_recalc, + /** + * u64 bpf_get_current_task(void) + * Returns current task_struct + * Return: current + */ + BPF_FUNC_get_current_task, + __BPF_FUNC_MAX_ID, }; diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 19c5b4a..094c716 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -312,6 +312,17 @@ const struct bpf_func_proto *bpf_get_event_output_proto(void) return &bpf_event_output_proto; } +static u64 bpf_get_current_task(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + return (long) current; +} + +static const struct bpf_func_proto bpf_get_current_task_proto = { + .func = bpf_get_current_task, + .gpl_only = true, + .ret_type = RET_INTEGER, +}; + static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id) { switch (func_id) { @@ -329,6 +340,8 @@ static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id) return &bpf_tail_call_proto; case BPF_FUNC_get_current_pid_tgid: return &bpf_get_current_pid_tgid_proto; + case BPF_FUNC_get_current_task: + return &bpf_get_current_task_proto; case BPF_FUNC_get_current_uid_gid: return &bpf_get_current_uid_gid_proto; case BPF_FUNC_get_current_comm: -- cgit v0.10.2 From 09a7636a5b151670072de60767ddf096dc7bd12e Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 7 Jul 2016 11:23:09 +0300 Subject: bnxt: fix a condition This code generates as static checker warning because htons(ETH_P_IPV6) is always true. From the context it looks like the && was intended to be !=. Fixes: 94758f8de037 ('bnxt_en: Add GRO logic for BCM5731X chips.') Signed-off-by: Dan Carpenter Acked-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 70b148a..659faa6 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -997,7 +997,7 @@ static struct sk_buff *bnxt_gro_func_5731x(struct bnxt_tpa_info *tpa_info, * correct protocol ID, it must be a loopback packet where * the offsets are off by 4. */ - if (proto != htons(ETH_P_IP) && proto && htons(ETH_P_IPV6)) + if (proto != htons(ETH_P_IP) && proto != htons(ETH_P_IPV6)) loopback = true; } if (loopback) { -- cgit v0.10.2 From f1533cce60d1f84378c1dd925f9ef1038fa93507 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 7 Jul 2016 09:39:29 -0300 Subject: sctp: fix panic when sending auth chunks When we introduced GSO support, if using auth the auth chunk was being left queued on the packet even after the final segment was generated. Later on sctp_transmit_packet it calls sctp_packet_reset, which zeroed the packet len while not accounting for this left-over. This caused more space to be used the next packet due to the chunk still being queued, but space which wasn't allocated as its size wasn't accounted. The fix is to only queue it back when we know that we are going to generate another segment. Fixes: 90017accff61 ("sctp: Add GSO support") Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/output.c b/net/sctp/output.c index 1541a91..2e9223b 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -582,9 +582,7 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) */ pkt_size -= WORD_ROUND(chunk->skb->len); - if (chunk == packet->auth && !list_empty(&packet->chunk_list)) - list_add(&chunk->list, &packet->chunk_list); - else if (!sctp_chunk_is_data(chunk)) + if (!sctp_chunk_is_data(chunk) && chunk != packet->auth) sctp_chunk_free(chunk); if (!pkt_size) @@ -605,6 +603,18 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) (struct sctp_auth_chunk *)auth, gfp); + if (packet->auth) { + if (!list_empty(&packet->chunk_list)) { + /* We will generate more packets, so re-queue + * auth chunk. + */ + list_add(&chunk->list, &packet->chunk_list); + } else { + sctp_chunk_free(packet->auth); + packet->auth = NULL; + } + } + if (!gso) break; @@ -735,6 +745,8 @@ err: } goto out; nomem: + if (packet->auth && list_empty(&packet->auth->list)) + sctp_chunk_free(packet->auth); err = -ENOMEM; goto err; } -- cgit v0.10.2 From 2a0be139868cae2465f9ed5b599203fa4f8e06ca Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Tue, 5 Jul 2016 14:30:12 +0200 Subject: Bluetooth: Remove connection link attributes The connection link attributes are not used and expose no valuable information. Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 555982a..6bfdca8 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -7,50 +7,6 @@ static struct class *bt_class; -static inline char *link_typetostr(int type) -{ - switch (type) { - case ACL_LINK: - return "ACL"; - case SCO_LINK: - return "SCO"; - case ESCO_LINK: - return "eSCO"; - case LE_LINK: - return "LE"; - default: - return "UNKNOWN"; - } -} - -static ssize_t show_link_type(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hci_conn *conn = to_hci_conn(dev); - return sprintf(buf, "%s\n", link_typetostr(conn->type)); -} - -static ssize_t show_link_address(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hci_conn *conn = to_hci_conn(dev); - return sprintf(buf, "%pMR\n", &conn->dst); -} - -#define LINK_ATTR(_name, _mode, _show, _store) \ -struct device_attribute link_attr_##_name = __ATTR(_name, _mode, _show, _store) - -static LINK_ATTR(type, S_IRUGO, show_link_type, NULL); -static LINK_ATTR(address, S_IRUGO, show_link_address, NULL); - -static struct attribute *bt_link_attrs[] = { - &link_attr_type.attr, - &link_attr_address.attr, - NULL -}; - -ATTRIBUTE_GROUPS(bt_link); - static void bt_link_release(struct device *dev) { struct hci_conn *conn = to_hci_conn(dev); @@ -59,7 +15,6 @@ static void bt_link_release(struct device *dev) static struct device_type bt_link = { .name = "link", - .groups = bt_link_groups, .release = bt_link_release, }; -- cgit v0.10.2 From e14dbe72033152135eb3bae212228728089d4dd9 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Tue, 5 Jul 2016 14:30:13 +0200 Subject: Bluetooth: Remove controller device attributes The controller device attributes are not used and expose no valuable information. Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 6bfdca8..ca7a35e 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -79,59 +79,6 @@ void hci_conn_del_sysfs(struct hci_conn *conn) hci_dev_put(hdev); } -static inline char *host_typetostr(int type) -{ - switch (type) { - case HCI_BREDR: - return "BR/EDR"; - case HCI_AMP: - return "AMP"; - default: - return "UNKNOWN"; - } -} - -static ssize_t show_type(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hci_dev *hdev = to_hci_dev(dev); - return sprintf(buf, "%s\n", host_typetostr(hdev->dev_type)); -} - -static ssize_t show_name(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hci_dev *hdev = to_hci_dev(dev); - char name[HCI_MAX_NAME_LENGTH + 1]; - int i; - - for (i = 0; i < HCI_MAX_NAME_LENGTH; i++) - name[i] = hdev->dev_name[i]; - - name[HCI_MAX_NAME_LENGTH] = '\0'; - return sprintf(buf, "%s\n", name); -} - -static ssize_t show_address(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct hci_dev *hdev = to_hci_dev(dev); - return sprintf(buf, "%pMR\n", &hdev->bdaddr); -} - -static DEVICE_ATTR(type, S_IRUGO, show_type, NULL); -static DEVICE_ATTR(name, S_IRUGO, show_name, NULL); -static DEVICE_ATTR(address, S_IRUGO, show_address, NULL); - -static struct attribute *bt_host_attrs[] = { - &dev_attr_type.attr, - &dev_attr_name.attr, - &dev_attr_address.attr, - NULL -}; - -ATTRIBUTE_GROUPS(bt_host); - static void bt_host_release(struct device *dev) { struct hci_dev *hdev = to_hci_dev(dev); @@ -141,7 +88,6 @@ static void bt_host_release(struct device *dev) static struct device_type bt_host = { .name = "host", - .groups = bt_host_groups, .release = bt_host_release, }; -- cgit v0.10.2 From ca8bee5dde1f02c2dbe8c8453dce27f2dfafb21c Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Tue, 5 Jul 2016 14:30:14 +0200 Subject: Bluetooth: Rename HCI_BREDR into HCI_PRIMARY The HCI_BREDR naming is confusing since it actually stands for Primary Bluetooth Controller. Which is a term that has been used in the latest standard. However from a legacy point of view there only really have been Basic Rate (BR) and Enhanced Data Rate (EDR). Recent versions of Bluetooth introduced Low Energy (LE) and made this terminology a little bit confused since Dual Mode Controllers include BR/EDR and LE. To simplify this the name HCI_PRIMARY stands for the Primary Controller which can be a single mode or dual mode controller. Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg diff --git a/drivers/bluetooth/btmrvl_main.c b/drivers/bluetooth/btmrvl_main.c index 7ad8d61..e6a85f0 100644 --- a/drivers/bluetooth/btmrvl_main.c +++ b/drivers/bluetooth/btmrvl_main.c @@ -138,7 +138,7 @@ int btmrvl_process_event(struct btmrvl_private *priv, struct sk_buff *skb) if (event->length > 3 && event->data[3]) priv->btmrvl_dev.dev_type = HCI_AMP; else - priv->btmrvl_dev.dev_type = HCI_BREDR; + priv->btmrvl_dev.dev_type = HCI_PRIMARY; BT_DBG("dev_type: %d", priv->btmrvl_dev.dev_type); } else if (priv->btmrvl_dev.sendcmdflag && diff --git a/drivers/bluetooth/btsdio.c b/drivers/bluetooth/btsdio.c index 2b05661..1cb958e 100644 --- a/drivers/bluetooth/btsdio.c +++ b/drivers/bluetooth/btsdio.c @@ -311,7 +311,7 @@ static int btsdio_probe(struct sdio_func *func, if (id->class == SDIO_CLASS_BT_AMP) hdev->dev_type = HCI_AMP; else - hdev->dev_type = HCI_BREDR; + hdev->dev_type = HCI_PRIMARY; data->hdev = hdev; diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 880bb55..f2e8fd7 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -2832,7 +2832,7 @@ static int btusb_probe(struct usb_interface *intf, if (id->driver_info & BTUSB_AMP) hdev->dev_type = HCI_AMP; else - hdev->dev_type = HCI_BREDR; + hdev->dev_type = HCI_PRIMARY; data->hdev = hdev; diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c index 49b3e1e..dda9739 100644 --- a/drivers/bluetooth/hci_ldisc.c +++ b/drivers/bluetooth/hci_ldisc.c @@ -609,7 +609,7 @@ static int hci_uart_register_dev(struct hci_uart *hu) if (test_bit(HCI_UART_CREATE_AMP, &hu->hdev_flags)) hdev->dev_type = HCI_AMP; else - hdev->dev_type = HCI_BREDR; + hdev->dev_type = HCI_PRIMARY; if (test_bit(HCI_UART_INIT_PENDING, &hu->hdev_flags)) return 0; diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c index aba3121..3ff229b 100644 --- a/drivers/bluetooth/hci_vhci.c +++ b/drivers/bluetooth/hci_vhci.c @@ -97,10 +97,10 @@ static int __vhci_create_device(struct vhci_data *data, __u8 opcode) if (data->hdev) return -EBADFD; - /* bits 0-1 are dev_type (BR/EDR or AMP) */ + /* bits 0-1 are dev_type (Primary or AMP) */ dev_type = opcode & 0x03; - if (dev_type != HCI_BREDR && dev_type != HCI_AMP) + if (dev_type != HCI_PRIMARY && dev_type != HCI_AMP) return -EINVAL; /* bits 2-5 are reserved (must be zero) */ @@ -316,7 +316,7 @@ static void vhci_open_timeout(struct work_struct *work) struct vhci_data *data = container_of(work, struct vhci_data, open_timeout.work); - vhci_create_device(data, amp ? HCI_AMP : HCI_BREDR); + vhci_create_device(data, amp ? HCI_AMP : HCI_PRIMARY); } static int vhci_open(struct inode *inode, struct file *file) diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index eefcf3e..a3f86de 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -65,7 +65,7 @@ #define HCI_I2C 8 /* HCI controller types */ -#define HCI_BREDR 0x00 +#define HCI_PRIMARY 0x00 #define HCI_AMP 0x01 /* First BR/EDR Controller shall have ID = 0 */ diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index bf9f8a8..3809617 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -625,7 +625,7 @@ struct hci_dev *hci_get_route(bdaddr_t *dst, bdaddr_t *src) list_for_each_entry(d, &hci_dev_list, list) { if (!test_bit(HCI_UP, &d->flags) || hci_dev_test_flag(d, HCI_USER_CHANNEL) || - d->dev_type != HCI_BREDR) + d->dev_type != HCI_PRIMARY) continue; /* Simple routing: diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 45a9fc6..98f6c37 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -260,14 +260,12 @@ static int hci_init1_req(struct hci_request *req, unsigned long opt) hci_reset_req(req, 0); switch (hdev->dev_type) { - case HCI_BREDR: + case HCI_PRIMARY: bredr_init(req); break; - case HCI_AMP: amp_init1(req); break; - default: BT_ERR("Unknown device type %d", hdev->dev_type); break; @@ -791,11 +789,11 @@ static int __hci_init(struct hci_dev *hdev) if (err < 0) return err; - /* HCI_BREDR covers both single-mode LE, BR/EDR and dual-mode + /* HCI_PRIMARY covers both single-mode LE, BR/EDR and dual-mode * BR/EDR/LE type controllers. AMP controllers only need the * first two stages of init. */ - if (hdev->dev_type != HCI_BREDR) + if (hdev->dev_type != HCI_PRIMARY) return 0; err = __hci_req_sync(hdev, hci_init3_req, 0, HCI_INIT_TIMEOUT, NULL); @@ -1202,7 +1200,7 @@ int hci_inquiry(void __user *arg) goto done; } - if (hdev->dev_type != HCI_BREDR) { + if (hdev->dev_type != HCI_PRIMARY) { err = -EOPNOTSUPP; goto done; } @@ -1307,7 +1305,7 @@ static int hci_dev_do_open(struct hci_dev *hdev) * since AMP controllers do not have an address. */ if (!hci_dev_test_flag(hdev, HCI_USER_CHANNEL) && - hdev->dev_type == HCI_BREDR && + hdev->dev_type == HCI_PRIMARY && !bacmp(&hdev->bdaddr, BDADDR_ANY) && !bacmp(&hdev->static_addr, BDADDR_ANY)) { ret = -EADDRNOTAVAIL; @@ -1402,7 +1400,7 @@ static int hci_dev_do_open(struct hci_dev *hdev) !hci_dev_test_flag(hdev, HCI_UNCONFIGURED) && !hci_dev_test_flag(hdev, HCI_USER_CHANNEL) && hci_dev_test_flag(hdev, HCI_MGMT) && - hdev->dev_type == HCI_BREDR) { + hdev->dev_type == HCI_PRIMARY) { ret = __hci_req_hci_power_on(hdev); mgmt_power_on(hdev, ret); } @@ -1563,7 +1561,7 @@ int hci_dev_do_close(struct hci_dev *hdev) auto_off = hci_dev_test_and_clear_flag(hdev, HCI_AUTO_OFF); - if (!auto_off && hdev->dev_type == HCI_BREDR && + if (!auto_off && hdev->dev_type == HCI_PRIMARY && hci_dev_test_flag(hdev, HCI_MGMT)) __mgmt_power_off(hdev); @@ -1802,7 +1800,7 @@ int hci_dev_cmd(unsigned int cmd, void __user *arg) goto done; } - if (hdev->dev_type != HCI_BREDR) { + if (hdev->dev_type != HCI_PRIMARY) { err = -EOPNOTSUPP; goto done; } @@ -2043,7 +2041,7 @@ static void hci_power_on(struct work_struct *work) */ if (hci_dev_test_flag(hdev, HCI_RFKILLED) || hci_dev_test_flag(hdev, HCI_UNCONFIGURED) || - (hdev->dev_type == HCI_BREDR && + (hdev->dev_type == HCI_PRIMARY && !bacmp(&hdev->bdaddr, BDADDR_ANY) && !bacmp(&hdev->static_addr, BDADDR_ANY))) { hci_dev_clear_flag(hdev, HCI_AUTO_OFF); @@ -3030,7 +3028,7 @@ int hci_register_dev(struct hci_dev *hdev) * so the index can be used as the AMP controller ID. */ switch (hdev->dev_type) { - case HCI_BREDR: + case HCI_PRIMARY: id = ida_simple_get(&hci_index_ida, 0, 0, GFP_KERNEL); break; case HCI_AMP: @@ -3090,7 +3088,7 @@ int hci_register_dev(struct hci_dev *hdev) hci_dev_set_flag(hdev, HCI_SETUP); hci_dev_set_flag(hdev, HCI_AUTO_OFF); - if (hdev->dev_type == HCI_BREDR) { + if (hdev->dev_type == HCI_PRIMARY) { /* Assume BR/EDR support until proven otherwise (such as * through reading supported features during init. */ @@ -3415,7 +3413,7 @@ static void hci_queue_acl(struct hci_chan *chan, struct sk_buff_head *queue, hci_skb_pkt_type(skb) = HCI_ACLDATA_PKT; switch (hdev->dev_type) { - case HCI_BREDR: + case HCI_PRIMARY: hci_add_acl_hdr(skb, conn->handle, flags); break; case HCI_AMP: @@ -3826,7 +3824,7 @@ static void hci_sched_acl(struct hci_dev *hdev) BT_DBG("%s", hdev->name); /* No ACL link over BR/EDR controller */ - if (!hci_conn_num(hdev, ACL_LINK) && hdev->dev_type == HCI_BREDR) + if (!hci_conn_num(hdev, ACL_LINK) && hdev->dev_type == HCI_PRIMARY) return; /* No AMP link over AMP controller */ diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index d4b3dd5..3fb95c4 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -3249,7 +3249,7 @@ static struct hci_conn *__hci_conn_lookup_handle(struct hci_dev *hdev, struct hci_chan *chan; switch (hdev->dev_type) { - case HCI_BREDR: + case HCI_PRIMARY: return hci_conn_hash_lookup_handle(hdev, handle); case HCI_AMP: chan = hci_chan_lookup_handle(hdev, handle); diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c index 12e9294..6ef8a01 100644 --- a/net/bluetooth/hci_sock.c +++ b/net/bluetooth/hci_sock.c @@ -676,7 +676,7 @@ static int hci_sock_bound_ioctl(struct sock *sk, unsigned int cmd, if (hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) return -EOPNOTSUPP; - if (hdev->dev_type != HCI_BREDR) + if (hdev->dev_type != HCI_PRIMARY) return -EOPNOTSUPP; switch (cmd) { diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index eb4f5f2..54ceb1f 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -7468,7 +7468,7 @@ void l2cap_recv_acldata(struct hci_conn *hcon, struct sk_buff *skb, u16 flags) int len; /* For AMP controller do not create l2cap conn */ - if (!conn && hcon->hdev->dev_type != HCI_BREDR) + if (!conn && hcon->hdev->dev_type != HCI_PRIMARY) goto drop; if (!conn) diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 9e4b931..7983ec8 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -359,7 +359,7 @@ static int read_index_list(struct sock *sk, struct hci_dev *hdev, void *data, count = 0; list_for_each_entry(d, &hci_dev_list, list) { - if (d->dev_type == HCI_BREDR && + if (d->dev_type == HCI_PRIMARY && !hci_dev_test_flag(d, HCI_UNCONFIGURED)) count++; } @@ -384,7 +384,7 @@ static int read_index_list(struct sock *sk, struct hci_dev *hdev, void *data, if (test_bit(HCI_QUIRK_RAW_DEVICE, &d->quirks)) continue; - if (d->dev_type == HCI_BREDR && + if (d->dev_type == HCI_PRIMARY && !hci_dev_test_flag(d, HCI_UNCONFIGURED)) { rp->index[count++] = cpu_to_le16(d->id); BT_DBG("Added hci%u", d->id); @@ -419,7 +419,7 @@ static int read_unconf_index_list(struct sock *sk, struct hci_dev *hdev, count = 0; list_for_each_entry(d, &hci_dev_list, list) { - if (d->dev_type == HCI_BREDR && + if (d->dev_type == HCI_PRIMARY && hci_dev_test_flag(d, HCI_UNCONFIGURED)) count++; } @@ -444,7 +444,7 @@ static int read_unconf_index_list(struct sock *sk, struct hci_dev *hdev, if (test_bit(HCI_QUIRK_RAW_DEVICE, &d->quirks)) continue; - if (d->dev_type == HCI_BREDR && + if (d->dev_type == HCI_PRIMARY && hci_dev_test_flag(d, HCI_UNCONFIGURED)) { rp->index[count++] = cpu_to_le16(d->id); BT_DBG("Added hci%u", d->id); @@ -479,7 +479,7 @@ static int read_ext_index_list(struct sock *sk, struct hci_dev *hdev, count = 0; list_for_each_entry(d, &hci_dev_list, list) { - if (d->dev_type == HCI_BREDR || d->dev_type == HCI_AMP) + if (d->dev_type == HCI_PRIMARY || d->dev_type == HCI_AMP) count++; } @@ -503,7 +503,7 @@ static int read_ext_index_list(struct sock *sk, struct hci_dev *hdev, if (test_bit(HCI_QUIRK_RAW_DEVICE, &d->quirks)) continue; - if (d->dev_type == HCI_BREDR) { + if (d->dev_type == HCI_PRIMARY) { if (hci_dev_test_flag(d, HCI_UNCONFIGURED)) rp->entry[count].type = 0x01; else @@ -6366,7 +6366,7 @@ void mgmt_index_added(struct hci_dev *hdev) return; switch (hdev->dev_type) { - case HCI_BREDR: + case HCI_PRIMARY: if (hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) { mgmt_index_event(MGMT_EV_UNCONF_INDEX_ADDED, hdev, NULL, 0, HCI_MGMT_UNCONF_INDEX_EVENTS); @@ -6399,7 +6399,7 @@ void mgmt_index_removed(struct hci_dev *hdev) return; switch (hdev->dev_type) { - case HCI_BREDR: + case HCI_PRIMARY: mgmt_pending_foreach(0, hdev, cmd_complete_rsp, &status); if (hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) { -- cgit v0.10.2 From a65056ecf4b48be0d0284a7b6a57b6dace10b843 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Wed, 6 Jul 2016 12:12:21 -0700 Subject: net: bridge: extend MLD/IGMP query stats MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As was suggested this patch adds support for the different versions of MLD and IGMP query types. Since the user visible structure is still in net-next we can augment it instead of adding netlink attributes. The distinction between the different IGMP/MLD query types is done as suggested in Section 7.1, RFC 3376 [1] and Section 8.1, RFC 3810 [2] based on query payload size and code for IGMP. Since all IGMP packets go through multicast_rcv() and it uses ip_mc_check_igmp/ipv6_mc_check_mld we can be sure that at least the ip/ipv6 header can be directly used. [1] https://tools.ietf.org/html/rfc3376#section-7 [2] https://tools.ietf.org/html/rfc3810#section-8.1 Suggested-by: Linus Lüssing Signed-off-by: Nikolay Aleksandrov Acked-by: Stephen Hemminger Signed-off-by: David S. Miller diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h index 8304fe6..c186f64 100644 --- a/include/uapi/linux/if_bridge.h +++ b/include/uapi/linux/if_bridge.h @@ -261,14 +261,17 @@ enum { /* IGMP/MLD statistics */ struct br_mcast_stats { - __u64 igmp_queries[BR_MCAST_DIR_SIZE]; + __u64 igmp_v1queries[BR_MCAST_DIR_SIZE]; + __u64 igmp_v2queries[BR_MCAST_DIR_SIZE]; + __u64 igmp_v3queries[BR_MCAST_DIR_SIZE]; __u64 igmp_leaves[BR_MCAST_DIR_SIZE]; __u64 igmp_v1reports[BR_MCAST_DIR_SIZE]; __u64 igmp_v2reports[BR_MCAST_DIR_SIZE]; __u64 igmp_v3reports[BR_MCAST_DIR_SIZE]; __u64 igmp_parse_errors; - __u64 mld_queries[BR_MCAST_DIR_SIZE]; + __u64 mld_v1queries[BR_MCAST_DIR_SIZE]; + __u64 mld_v2queries[BR_MCAST_DIR_SIZE]; __u64 mld_leaves[BR_MCAST_DIR_SIZE]; __u64 mld_v1reports[BR_MCAST_DIR_SIZE]; __u64 mld_v2reports[BR_MCAST_DIR_SIZE]; diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 6c19603..d610644 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -199,7 +199,6 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, bool unicast) { u8 igmp_type = br_multicast_igmp_type(skb); - __be16 proto = skb->protocol; struct net_bridge_port *prev; struct net_bridge_port *p; @@ -221,7 +220,7 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, if (IS_ERR(prev)) goto out; if (prev == p) - br_multicast_count(p->br, p, proto, igmp_type, + br_multicast_count(p->br, p, skb, igmp_type, BR_MCAST_DIR_TX); } @@ -266,8 +265,6 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, struct net_bridge *br = netdev_priv(dev); struct net_bridge_port *prev = NULL; struct net_bridge_port_group *p; - __be16 proto = skb->protocol; - struct hlist_node *rp; rp = rcu_dereference(hlist_first_rcu(&br->router_list)); @@ -286,7 +283,7 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, if (IS_ERR(prev)) goto out; if (prev == port) - br_multicast_count(port->br, port, proto, igmp_type, + br_multicast_count(port->br, port, skb, igmp_type, BR_MCAST_DIR_TX); if ((unsigned long)lport >= (unsigned long)port) diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 786602b..a7817e6 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -61,7 +61,7 @@ static int br_pass_frame_up(struct sk_buff *skb) if (!skb) return NET_RX_DROP; /* update the multicast stats if the packet is IGMP/MLD */ - br_multicast_count(br, NULL, skb->protocol, br_multicast_igmp_type(skb), + br_multicast_count(br, NULL, skb, br_multicast_igmp_type(skb), BR_MCAST_DIR_TX); return NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_IN, diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index e405eef..a5423a1 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -843,14 +843,14 @@ static void __br_multicast_send_query(struct net_bridge *br, if (port) { skb->dev = port->dev; - br_multicast_count(br, port, skb->protocol, igmp_type, + br_multicast_count(br, port, skb, igmp_type, BR_MCAST_DIR_TX); NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, dev_net(port->dev), NULL, skb, NULL, skb->dev, br_dev_queue_push_xmit); } else { br_multicast_select_own_querier(br, ip, skb); - br_multicast_count(br, port, skb->protocol, igmp_type, + br_multicast_count(br, port, skb, igmp_type, BR_MCAST_DIR_RX); netif_rx(skb); } @@ -1676,7 +1676,7 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, if (skb_trimmed && skb_trimmed != skb) kfree_skb(skb_trimmed); - br_multicast_count(br, port, skb->protocol, BR_INPUT_SKB_CB(skb)->igmp, + br_multicast_count(br, port, skb, BR_INPUT_SKB_CB(skb)->igmp, BR_MCAST_DIR_RX); return err; @@ -1725,7 +1725,7 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, if (skb_trimmed && skb_trimmed != skb) kfree_skb(skb_trimmed); - br_multicast_count(br, port, skb->protocol, BR_INPUT_SKB_CB(skb)->igmp, + br_multicast_count(br, port, skb, BR_INPUT_SKB_CB(skb)->igmp, BR_MCAST_DIR_RX); return err; @@ -2251,13 +2251,16 @@ unlock: EXPORT_SYMBOL_GPL(br_multicast_has_querier_adjacent); static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, - __be16 proto, u8 type, u8 dir) + const struct sk_buff *skb, u8 type, u8 dir) { struct bridge_mcast_stats *pstats = this_cpu_ptr(stats); + __be16 proto = skb->protocol; + unsigned int t_len; u64_stats_update_begin(&pstats->syncp); switch (proto) { case htons(ETH_P_IP): + t_len = ntohs(ip_hdr(skb)->tot_len) - ip_hdrlen(skb); switch (type) { case IGMP_HOST_MEMBERSHIP_REPORT: pstats->mstats.igmp_v1reports[dir]++; @@ -2269,7 +2272,21 @@ static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, pstats->mstats.igmp_v3reports[dir]++; break; case IGMP_HOST_MEMBERSHIP_QUERY: - pstats->mstats.igmp_queries[dir]++; + if (t_len != sizeof(struct igmphdr)) { + pstats->mstats.igmp_v3queries[dir]++; + } else { + unsigned int offset = skb_transport_offset(skb); + struct igmphdr *ih, _ihdr; + + ih = skb_header_pointer(skb, offset, + sizeof(_ihdr), &_ihdr); + if (!ih) + break; + if (!ih->code) + pstats->mstats.igmp_v1queries[dir]++; + else + pstats->mstats.igmp_v2queries[dir]++; + } break; case IGMP_HOST_LEAVE_MESSAGE: pstats->mstats.igmp_leaves[dir]++; @@ -2278,6 +2295,9 @@ static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, break; #if IS_ENABLED(CONFIG_IPV6) case htons(ETH_P_IPV6): + t_len = ntohs(ipv6_hdr(skb)->payload_len) + + sizeof(struct ipv6hdr); + t_len -= skb_network_header_len(skb); switch (type) { case ICMPV6_MGM_REPORT: pstats->mstats.mld_v1reports[dir]++; @@ -2286,7 +2306,10 @@ static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, pstats->mstats.mld_v2reports[dir]++; break; case ICMPV6_MGM_QUERY: - pstats->mstats.mld_queries[dir]++; + if (t_len != sizeof(struct mld_msg)) + pstats->mstats.mld_v2queries[dir]++; + else + pstats->mstats.mld_v1queries[dir]++; break; case ICMPV6_MGM_REDUCTION: pstats->mstats.mld_leaves[dir]++; @@ -2299,7 +2322,7 @@ static void br_mcast_stats_add(struct bridge_mcast_stats __percpu *stats, } void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, - __be16 proto, u8 type, u8 dir) + const struct sk_buff *skb, u8 type, u8 dir) { struct bridge_mcast_stats __percpu *stats; @@ -2314,7 +2337,7 @@ void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, if (WARN_ON(!stats)) return; - br_mcast_stats_add(stats, proto, type, dir); + br_mcast_stats_add(stats, skb, type, dir); } int br_multicast_init_stats(struct net_bridge *br) @@ -2359,14 +2382,17 @@ void br_multicast_get_stats(const struct net_bridge *br, memcpy(&temp, &cpu_stats->mstats, sizeof(temp)); } while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start)); - mcast_stats_add_dir(tdst.igmp_queries, temp.igmp_queries); + mcast_stats_add_dir(tdst.igmp_v1queries, temp.igmp_v1queries); + mcast_stats_add_dir(tdst.igmp_v2queries, temp.igmp_v2queries); + mcast_stats_add_dir(tdst.igmp_v3queries, temp.igmp_v3queries); mcast_stats_add_dir(tdst.igmp_leaves, temp.igmp_leaves); mcast_stats_add_dir(tdst.igmp_v1reports, temp.igmp_v1reports); mcast_stats_add_dir(tdst.igmp_v2reports, temp.igmp_v2reports); mcast_stats_add_dir(tdst.igmp_v3reports, temp.igmp_v3reports); tdst.igmp_parse_errors += temp.igmp_parse_errors; - mcast_stats_add_dir(tdst.mld_queries, temp.mld_queries); + mcast_stats_add_dir(tdst.mld_v1queries, temp.mld_v1queries); + mcast_stats_add_dir(tdst.mld_v2queries, temp.mld_v2queries); mcast_stats_add_dir(tdst.mld_leaves, temp.mld_leaves); mcast_stats_add_dir(tdst.mld_v1reports, temp.mld_v1reports); mcast_stats_add_dir(tdst.mld_v2reports, temp.mld_v2reports); diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 4dc8511..40f2009 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -586,7 +586,7 @@ void br_mdb_notify(struct net_device *dev, struct net_bridge_port *port, void br_rtr_notify(struct net_device *dev, struct net_bridge_port *port, int type); void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, - __be16 proto, u8 type, u8 dir); + const struct sk_buff *skb, u8 type, u8 dir); int br_multicast_init_stats(struct net_bridge *br); void br_multicast_get_stats(const struct net_bridge *br, const struct net_bridge_port *p, @@ -719,7 +719,8 @@ static inline void br_mdb_uninit(void) static inline void br_multicast_count(struct net_bridge *br, const struct net_bridge_port *p, - __be16 proto, u8 type, u8 dir) + const struct sk_buff *skb, + u8 type, u8 dir) { } -- cgit v0.10.2 From 8afe97e5d4165c9d181d42504af3f96c8427659a Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 7 Jul 2016 07:56:12 +0200 Subject: tunnels: support MPLS over IPv4 tunnels Extend tunnel support to MPLS over IPv4. The implementation extends the existing differentiation between IPIP and IPv6 over IPv4 to also cover MPLS over IPv4. Signed-off-by: Simon Horman Reviewed-by: Dinan Gunawardena Signed-off-by: David S. Miller diff --git a/net/ipv4/tunnel4.c b/net/ipv4/tunnel4.c index 0d01718..45cd425 100644 --- a/net/ipv4/tunnel4.c +++ b/net/ipv4/tunnel4.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include @@ -16,11 +17,14 @@ static struct xfrm_tunnel __rcu *tunnel4_handlers __read_mostly; static struct xfrm_tunnel __rcu *tunnel64_handlers __read_mostly; +static struct xfrm_tunnel __rcu *tunnelmpls4_handlers __read_mostly; static DEFINE_MUTEX(tunnel4_mutex); static inline struct xfrm_tunnel __rcu **fam_handlers(unsigned short family) { - return (family == AF_INET) ? &tunnel4_handlers : &tunnel64_handlers; + return (family == AF_INET) ? &tunnel4_handlers : + (family == AF_INET6) ? &tunnel64_handlers : + &tunnelmpls4_handlers; } int xfrm4_tunnel_register(struct xfrm_tunnel *handler, unsigned short family) @@ -125,6 +129,26 @@ drop: } #endif +#if IS_ENABLED(CONFIG_MPLS) +static int tunnelmpls4_rcv(struct sk_buff *skb) +{ + struct xfrm_tunnel *handler; + + if (!pskb_may_pull(skb, sizeof(struct mpls_label))) + goto drop; + + for_each_tunnel_rcu(tunnelmpls4_handlers, handler) + if (!handler->handler(skb)) + return 0; + + icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); + +drop: + kfree_skb(skb); + return 0; +} +#endif + static void tunnel4_err(struct sk_buff *skb, u32 info) { struct xfrm_tunnel *handler; @@ -145,6 +169,17 @@ static void tunnel64_err(struct sk_buff *skb, u32 info) } #endif +#if IS_ENABLED(CONFIG_MPLS) +static void tunnelmpls4_err(struct sk_buff *skb, u32 info) +{ + struct xfrm_tunnel *handler; + + for_each_tunnel_rcu(tunnelmpls4_handlers, handler) + if (!handler->err_handler(skb, info)) + break; +} +#endif + static const struct net_protocol tunnel4_protocol = { .handler = tunnel4_rcv, .err_handler = tunnel4_err, @@ -161,24 +196,46 @@ static const struct net_protocol tunnel64_protocol = { }; #endif +#if IS_ENABLED(CONFIG_MPLS) +static const struct net_protocol tunnelmpls4_protocol = { + .handler = tunnelmpls4_rcv, + .err_handler = tunnelmpls4_err, + .no_policy = 1, + .netns_ok = 1, +}; +#endif + static int __init tunnel4_init(void) { - if (inet_add_protocol(&tunnel4_protocol, IPPROTO_IPIP)) { - pr_err("%s: can't add protocol\n", __func__); - return -EAGAIN; - } + if (inet_add_protocol(&tunnel4_protocol, IPPROTO_IPIP)) + goto err_ipip; #if IS_ENABLED(CONFIG_IPV6) - if (inet_add_protocol(&tunnel64_protocol, IPPROTO_IPV6)) { - pr_err("tunnel64 init: can't add protocol\n"); - inet_del_protocol(&tunnel4_protocol, IPPROTO_IPIP); - return -EAGAIN; - } + if (inet_add_protocol(&tunnel64_protocol, IPPROTO_IPV6)) + goto err_ipv6; +#endif +#if IS_ENABLED(CONFIG_MPLS) + if (inet_add_protocol(&tunnelmpls4_protocol, IPPROTO_MPLS)) + goto err_mpls; #endif return 0; + +#if IS_ENABLED(CONFIG_IPV6) +err_mpls: + inet_del_protocol(&tunnel4_protocol, IPPROTO_IPV6); +#endif +err_ipv6: + inet_del_protocol(&tunnel4_protocol, IPPROTO_IPIP); +err_ipip: + pr_err("%s: can't add protocol\n", __func__); + return -EAGAIN; } static void __exit tunnel4_fini(void) { +#if IS_ENABLED(CONFIG_MPLS) + if (inet_del_protocol(&tunnelmpls4_protocol, IPPROTO_MPLS)) + pr_err("tunnelmpls4 close: can't remove protocol\n"); +#endif #if IS_ENABLED(CONFIG_IPV6) if (inet_del_protocol(&tunnel64_protocol, IPPROTO_IPV6)) pr_err("tunnel64 close: can't remove protocol\n"); -- cgit v0.10.2 From 49dbe7ae2168b3a933ecea1118fc0515c186bd64 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 7 Jul 2016 07:56:13 +0200 Subject: sit: support MPLS over IPv4 Extend the SIT driver to support MPLS over IPv4. This implementation extends existing support for IPv6 over IPv4 and IPv4 over IPv4. Signed-off-by: Simon Horman Reviewed-by: Dinan Gunawardena Signed-off-by: David S. Miller diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 917a5cd..182b6a9 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -688,12 +688,19 @@ out: return 0; } -static const struct tnl_ptk_info tpi = { +static const struct tnl_ptk_info ipip_tpi = { /* no tunnel info required for ipip. */ .proto = htons(ETH_P_IP), }; -static int ipip_rcv(struct sk_buff *skb) +#if IS_ENABLED(CONFIG_MPLS) +static const struct tnl_ptk_info mplsip_tpi = { + /* no tunnel info required for mplsip. */ + .proto = htons(ETH_P_MPLS_UC), +}; +#endif + +static int sit_tunnel_rcv(struct sk_buff *skb, u8 ipproto) { const struct iphdr *iph; struct ip_tunnel *tunnel; @@ -702,15 +709,23 @@ static int ipip_rcv(struct sk_buff *skb) tunnel = ipip6_tunnel_lookup(dev_net(skb->dev), skb->dev, iph->saddr, iph->daddr); if (tunnel) { - if (tunnel->parms.iph.protocol != IPPROTO_IPIP && + const struct tnl_ptk_info *tpi; + + if (tunnel->parms.iph.protocol != ipproto && tunnel->parms.iph.protocol != 0) goto drop; if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) goto drop; - if (iptunnel_pull_header(skb, 0, tpi.proto, false)) +#if IS_ENABLED(CONFIG_MPLS) + if (ipproto == IPPROTO_MPLS) + tpi = &mplsip_tpi; + else +#endif + tpi = &ipip_tpi; + if (iptunnel_pull_header(skb, 0, tpi->proto, false)) goto drop; - return ip_tunnel_rcv(tunnel, skb, &tpi, NULL, log_ecn_error); + return ip_tunnel_rcv(tunnel, skb, tpi, NULL, log_ecn_error); } return 1; @@ -720,6 +735,18 @@ drop: return 0; } +static int ipip_rcv(struct sk_buff *skb) +{ + return sit_tunnel_rcv(skb, IPPROTO_IPIP); +} + +#if IS_ENABLED(CONFIG_MPLS) +static int mplsip_rcv(struct sk_buff *skb) +{ + return sit_tunnel_rcv(skb, IPPROTO_MPLS); +} +#endif + /* * If the IPv6 address comes from 6rd / 6to4 (RFC 3056) addr space this function * stores the embedded IPv4 address in v4dst and returns true. @@ -958,7 +985,8 @@ tx_error: return NETDEV_TX_OK; } -static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t sit_tunnel_xmit__(struct sk_buff *skb, + struct net_device *dev, u8 ipproto) { struct ip_tunnel *tunnel = netdev_priv(dev); const struct iphdr *tiph = &tunnel->parms.iph; @@ -966,9 +994,9 @@ static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (iptunnel_handle_offloads(skb, SKB_GSO_IPXIP4)) goto tx_error; - skb_set_inner_ipproto(skb, IPPROTO_IPIP); + skb_set_inner_ipproto(skb, ipproto); - ip_tunnel_xmit(skb, dev, tiph, IPPROTO_IPIP); + ip_tunnel_xmit(skb, dev, tiph, ipproto); return NETDEV_TX_OK; tx_error: kfree_skb(skb); @@ -981,11 +1009,16 @@ static netdev_tx_t sit_tunnel_xmit(struct sk_buff *skb, { switch (skb->protocol) { case htons(ETH_P_IP): - ipip_tunnel_xmit(skb, dev); + sit_tunnel_xmit__(skb, dev, IPPROTO_IPIP); break; case htons(ETH_P_IPV6): ipip6_tunnel_xmit(skb, dev); break; +#if IS_ENABLED(CONFIG_MPLS) + case htons(ETH_P_MPLS_UC): + sit_tunnel_xmit__(skb, dev, IPPROTO_MPLS); + break; +#endif default: goto tx_err; } @@ -1093,6 +1126,16 @@ static int ipip6_tunnel_update_6rd(struct ip_tunnel *t, } #endif +bool ipip6_valid_ip_proto(u8 ipproto) +{ + return ipproto == IPPROTO_IPV6 || + ipproto == IPPROTO_IPIP || +#if IS_ENABLED(CONFIG_MPLS) + ipproto == IPPROTO_MPLS || +#endif + ipproto == 0; +} + static int ipip6_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { @@ -1152,9 +1195,7 @@ ipip6_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) goto done; err = -EINVAL; - if (p.iph.protocol != IPPROTO_IPV6 && - p.iph.protocol != IPPROTO_IPIP && - p.iph.protocol != 0) + if (!ipip6_valid_ip_proto(p.iph.protocol)) goto done; if (p.iph.version != 4 || p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF))) @@ -1379,9 +1420,7 @@ static int ipip6_validate(struct nlattr *tb[], struct nlattr *data[]) return 0; proto = nla_get_u8(data[IFLA_IPTUN_PROTO]); - if (proto != IPPROTO_IPV6 && - proto != IPPROTO_IPIP && - proto != 0) + if (!ipip6_valid_ip_proto(proto)) return -EINVAL; return 0; @@ -1723,6 +1762,14 @@ static struct xfrm_tunnel ipip_handler __read_mostly = { .priority = 2, }; +#if IS_ENABLED(CONFIG_MPLS) +static struct xfrm_tunnel mplsip_handler __read_mostly = { + .handler = mplsip_rcv, + .err_handler = ipip6_err, + .priority = 2, +}; +#endif + static void __net_exit sit_destroy_tunnels(struct net *net, struct list_head *head) { @@ -1818,6 +1865,9 @@ static void __exit sit_cleanup(void) rtnl_link_unregister(&sit_link_ops); xfrm4_tunnel_deregister(&sit_handler, AF_INET6); xfrm4_tunnel_deregister(&ipip_handler, AF_INET); +#if IS_ENABLED(CONFIG_MPLS) + xfrm4_tunnel_deregister(&mplsip_handler, AF_MPLS); +#endif unregister_pernet_device(&sit_net_ops); rcu_barrier(); /* Wait for completion of call_rcu()'s */ @@ -1827,7 +1877,7 @@ static int __init sit_init(void) { int err; - pr_info("IPv6 over IPv4 tunneling driver\n"); + pr_info("IPv6, IPv4 and MPLS over IPv4 tunneling driver\n"); err = register_pernet_device(&sit_net_ops); if (err < 0) @@ -1842,6 +1892,13 @@ static int __init sit_init(void) pr_info("%s: can't register ip4ip4\n", __func__); goto xfrm_tunnel4_failed; } +#if IS_ENABLED(CONFIG_MPLS) + err = xfrm4_tunnel_register(&mplsip_handler, AF_MPLS); + if (err < 0) { + pr_info("%s: can't register mplsip\n", __func__); + goto xfrm_tunnel_mpls_failed; + } +#endif err = rtnl_link_register(&sit_link_ops); if (err < 0) goto rtnl_link_failed; @@ -1850,6 +1907,10 @@ out: return err; rtnl_link_failed: +#if IS_ENABLED(CONFIG_MPLS) + xfrm4_tunnel_deregister(&mplsip_handler, AF_MPLS); +xfrm_tunnel_mpls_failed: +#endif xfrm4_tunnel_deregister(&ipip_handler, AF_INET); xfrm_tunnel4_failed: xfrm4_tunnel_deregister(&sit_handler, AF_INET6); -- cgit v0.10.2 From 1b69e7e6c4da1e84edc2496fa91db289e5e493b0 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 7 Jul 2016 07:56:14 +0200 Subject: ipip: support MPLS over IPv4 Extend the IPIP driver to support MPLS over IPv4. The implementation is an extension of existing support for IPv4 over IPv4 and is based of multiple inner-protocol support for the SIT driver. Signed-off-by: Simon Horman Reviewed-by: Dinan Gunawardena Signed-off-by: David S. Miller diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 9783701..4ae3f8e 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -148,14 +148,14 @@ static int ipip_err(struct sk_buff *skb, u32 info) if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED) { ipv4_update_pmtu(skb, dev_net(skb->dev), info, - t->parms.link, 0, IPPROTO_IPIP, 0); + t->parms.link, 0, iph->protocol, 0); err = 0; goto out; } if (type == ICMP_REDIRECT) { ipv4_redirect(skb, dev_net(skb->dev), t->parms.link, 0, - IPPROTO_IPIP, 0); + iph->protocol, 0); err = 0; goto out; } @@ -177,12 +177,19 @@ out: return err; } -static const struct tnl_ptk_info tpi = { +static const struct tnl_ptk_info ipip_tpi = { /* no tunnel info required for ipip. */ .proto = htons(ETH_P_IP), }; -static int ipip_rcv(struct sk_buff *skb) +#if IS_ENABLED(CONFIG_MPLS) +static const struct tnl_ptk_info mplsip_tpi = { + /* no tunnel info required for mplsip. */ + .proto = htons(ETH_P_MPLS_UC), +}; +#endif + +static int ipip_tunnel_rcv(struct sk_buff *skb, u8 ipproto) { struct net *net = dev_net(skb->dev); struct ip_tunnel_net *itn = net_generic(net, ipip_net_id); @@ -193,11 +200,23 @@ static int ipip_rcv(struct sk_buff *skb) tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY, iph->saddr, iph->daddr, 0); if (tunnel) { + const struct tnl_ptk_info *tpi; + + if (tunnel->parms.iph.protocol != ipproto && + tunnel->parms.iph.protocol != 0) + goto drop; + if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) goto drop; - if (iptunnel_pull_header(skb, 0, tpi.proto, false)) +#if IS_ENABLED(CONFIG_MPLS) + if (ipproto == IPPROTO_MPLS) + tpi = &mplsip_tpi; + else +#endif + tpi = &ipip_tpi; + if (iptunnel_pull_header(skb, 0, tpi->proto, false)) goto drop; - return ip_tunnel_rcv(tunnel, skb, &tpi, NULL, log_ecn_error); + return ip_tunnel_rcv(tunnel, skb, tpi, NULL, log_ecn_error); } return -1; @@ -207,24 +226,51 @@ drop: return 0; } +static int ipip_rcv(struct sk_buff *skb) +{ + return ipip_tunnel_rcv(skb, IPPROTO_IPIP); +} + +#if IS_ENABLED(CONFIG_MPLS) +static int mplsip_rcv(struct sk_buff *skb) +{ + return ipip_tunnel_rcv(skb, IPPROTO_MPLS); +} +#endif + /* * This function assumes it is being called from dev_queue_xmit() * and that skb is filled properly by that function. */ -static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, + struct net_device *dev) { struct ip_tunnel *tunnel = netdev_priv(dev); const struct iphdr *tiph = &tunnel->parms.iph; + u8 ipproto; + + switch (skb->protocol) { + case htons(ETH_P_IP): + ipproto = IPPROTO_IPIP; + break; +#if IS_ENABLED(CONFIG_MPLS) + case htons(ETH_P_MPLS_UC): + ipproto = IPPROTO_MPLS; + break; +#endif + default: + goto tx_error; + } - if (unlikely(skb->protocol != htons(ETH_P_IP))) + if (tiph->protocol != ipproto && tiph->protocol != 0) goto tx_error; if (iptunnel_handle_offloads(skb, SKB_GSO_IPXIP4)) goto tx_error; - skb_set_inner_ipproto(skb, IPPROTO_IPIP); + skb_set_inner_ipproto(skb, ipproto); - ip_tunnel_xmit(skb, dev, tiph, tiph->protocol); + ip_tunnel_xmit(skb, dev, tiph, ipproto); return NETDEV_TX_OK; tx_error: @@ -234,6 +280,20 @@ tx_error: return NETDEV_TX_OK; } +static bool ipip_tunnel_ioctl_verify_protocol(u8 ipproto) +{ + switch (ipproto) { + case 0: + case IPPROTO_IPIP: +#if IS_ENABLED(CONFIG_MPLS) + case IPPROTO_MPLS: +#endif + return true; + } + + return false; +} + static int ipip_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { @@ -244,7 +304,8 @@ ipip_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) return -EFAULT; if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) { - if (p.iph.version != 4 || p.iph.protocol != IPPROTO_IPIP || + if (p.iph.version != 4 || + !ipip_tunnel_ioctl_verify_protocol(p.iph.protocol) || p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF))) return -EINVAL; } @@ -301,10 +362,23 @@ static int ipip_tunnel_init(struct net_device *dev) tunnel->tun_hlen = 0; tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen; - tunnel->parms.iph.protocol = IPPROTO_IPIP; return ip_tunnel_init(dev); } +static int ipip_tunnel_validate(struct nlattr *tb[], struct nlattr *data[]) +{ + u8 proto; + + if (!data || !data[IFLA_IPTUN_PROTO]) + return 0; + + proto = nla_get_u8(data[IFLA_IPTUN_PROTO]); + if (proto != IPPROTO_IPIP && proto != IPPROTO_MPLS && proto != 0) + return -EINVAL; + + return 0; +} + static void ipip_netlink_parms(struct nlattr *data[], struct ip_tunnel_parm *parms) { @@ -335,6 +409,9 @@ static void ipip_netlink_parms(struct nlattr *data[], if (data[IFLA_IPTUN_TOS]) parms->iph.tos = nla_get_u8(data[IFLA_IPTUN_TOS]); + if (data[IFLA_IPTUN_PROTO]) + parms->iph.protocol = nla_get_u8(data[IFLA_IPTUN_PROTO]); + if (!data[IFLA_IPTUN_PMTUDISC] || nla_get_u8(data[IFLA_IPTUN_PMTUDISC])) parms->iph.frag_off = htons(IP_DF); } @@ -427,6 +504,8 @@ static size_t ipip_get_size(const struct net_device *dev) nla_total_size(1) + /* IFLA_IPTUN_TOS */ nla_total_size(1) + + /* IFLA_IPTUN_PROTO */ + nla_total_size(1) + /* IFLA_IPTUN_PMTUDISC */ nla_total_size(1) + /* IFLA_IPTUN_ENCAP_TYPE */ @@ -450,6 +529,7 @@ static int ipip_fill_info(struct sk_buff *skb, const struct net_device *dev) nla_put_in_addr(skb, IFLA_IPTUN_REMOTE, parm->iph.daddr) || nla_put_u8(skb, IFLA_IPTUN_TTL, parm->iph.ttl) || nla_put_u8(skb, IFLA_IPTUN_TOS, parm->iph.tos) || + nla_put_u8(skb, IFLA_IPTUN_PROTO, parm->iph.protocol) || nla_put_u8(skb, IFLA_IPTUN_PMTUDISC, !!(parm->iph.frag_off & htons(IP_DF)))) goto nla_put_failure; @@ -476,6 +556,7 @@ static const struct nla_policy ipip_policy[IFLA_IPTUN_MAX + 1] = { [IFLA_IPTUN_REMOTE] = { .type = NLA_U32 }, [IFLA_IPTUN_TTL] = { .type = NLA_U8 }, [IFLA_IPTUN_TOS] = { .type = NLA_U8 }, + [IFLA_IPTUN_PROTO] = { .type = NLA_U8 }, [IFLA_IPTUN_PMTUDISC] = { .type = NLA_U8 }, [IFLA_IPTUN_ENCAP_TYPE] = { .type = NLA_U16 }, [IFLA_IPTUN_ENCAP_FLAGS] = { .type = NLA_U16 }, @@ -489,6 +570,7 @@ static struct rtnl_link_ops ipip_link_ops __read_mostly = { .policy = ipip_policy, .priv_size = sizeof(struct ip_tunnel), .setup = ipip_tunnel_setup, + .validate = ipip_tunnel_validate, .newlink = ipip_newlink, .changelink = ipip_changelink, .dellink = ip_tunnel_dellink, @@ -503,6 +585,14 @@ static struct xfrm_tunnel ipip_handler __read_mostly = { .priority = 1, }; +#if IS_ENABLED(CONFIG_MPLS) +static struct xfrm_tunnel mplsip_handler __read_mostly = { + .handler = mplsip_rcv, + .err_handler = ipip_err, + .priority = 1, +}; +#endif + static int __net_init ipip_init_net(struct net *net) { return ip_tunnel_init_net(net, ipip_net_id, &ipip_link_ops, "tunl0"); @@ -525,7 +615,7 @@ static int __init ipip_init(void) { int err; - pr_info("ipip: IPv4 over IPv4 tunneling driver\n"); + pr_info("ipip: IPv4 and MPLS over IPv4 tunneling driver\n"); err = register_pernet_device(&ipip_net_ops); if (err < 0) @@ -533,8 +623,15 @@ static int __init ipip_init(void) err = xfrm4_tunnel_register(&ipip_handler, AF_INET); if (err < 0) { pr_info("%s: can't register tunnel\n", __func__); - goto xfrm_tunnel_failed; + goto xfrm_tunnel_ipip_failed; + } +#if IS_ENABLED(CONFIG_MPLS) + err = xfrm4_tunnel_register(&mplsip_handler, AF_MPLS); + if (err < 0) { + pr_info("%s: can't register tunnel\n", __func__); + goto xfrm_tunnel_mplsip_failed; } +#endif err = rtnl_link_register(&ipip_link_ops); if (err < 0) goto rtnl_link_failed; @@ -543,8 +640,13 @@ out: return err; rtnl_link_failed: +#if IS_ENABLED(CONFIG_MPLS) + xfrm4_tunnel_deregister(&mplsip_handler, AF_INET); +xfrm_tunnel_mplsip_failed: + +#endif xfrm4_tunnel_deregister(&ipip_handler, AF_INET); -xfrm_tunnel_failed: +xfrm_tunnel_ipip_failed: unregister_pernet_device(&ipip_net_ops); goto out; } @@ -554,7 +656,10 @@ static void __exit ipip_fini(void) rtnl_link_unregister(&ipip_link_ops); if (xfrm4_tunnel_deregister(&ipip_handler, AF_INET)) pr_info("%s: can't deregister tunnel\n", __func__); - +#if IS_ENABLED(CONFIG_MPLS) + if (xfrm4_tunnel_deregister(&mplsip_handler, AF_MPLS)) + pr_info("%s: can't deregister tunnel\n", __func__); +#endif unregister_pernet_device(&ipip_net_ops); } -- cgit v0.10.2 From 407f31be9ddfbcc51ae8054c1218db00c08b92e9 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 7 Jul 2016 07:56:15 +0200 Subject: mpls: allow routes on ipip and sit devices Allow MPLS routes on IPIP and SIT devices now that they support forwarding MPLS packets. Signed-off-by: Simon Horman Reviewed-by: Dinan Gunawardena Signed-off-by: David S. Miller diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index e9beaa5..5c161e7 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -1009,10 +1009,12 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event, unsigned int flags; if (event == NETDEV_REGISTER) { - /* For now just support Ethernet and IPGRE devices */ + /* For now just support Ethernet, IPGRE, SIT and IPIP devices */ if (dev->type == ARPHRD_ETHER || dev->type == ARPHRD_LOOPBACK || - dev->type == ARPHRD_IPGRE) { + dev->type == ARPHRD_IPGRE || + dev->type == ARPHRD_SIT || + dev->type == ARPHRD_TUNNEL) { mdev = mpls_add_dev(dev); if (IS_ERR(mdev)) return notifier_from_errno(PTR_ERR(mdev)); -- cgit v0.10.2 From b1648066b41f01057e2b802f71410bddae8dce45 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Thu, 7 Jul 2016 15:09:18 +0800 Subject: r8152: remove rtl_phy_reset function In rtl_hw_phy_work_func_t(), the flag of PHY_RESET is set in rtl_ops.hw_phy_cfg() and cleared in rtl8152_set_speed(). Therefore, the rtl_phy_reset() is never run and is unnecessary. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index b225bc2..e005ee6 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -2450,27 +2450,6 @@ static void rtl8153_runtime_enable(struct r8152 *tp, bool enable) } } -static void rtl_phy_reset(struct r8152 *tp) -{ - u16 data; - int i; - - data = r8152_mdio_read(tp, MII_BMCR); - - /* don't reset again before the previous one complete */ - if (data & BMCR_RESET) - return; - - data |= BMCR_RESET; - r8152_mdio_write(tp, MII_BMCR, data); - - for (i = 0; i < 50; i++) { - msleep(20); - if ((r8152_mdio_read(tp, MII_BMCR) & BMCR_RESET) == 0) - break; - } -} - static void r8153_teredo_off(struct r8152 *tp) { u32 ocp_data; @@ -3069,9 +3048,6 @@ static void rtl_work_func_t(struct work_struct *work) netif_carrier_ok(tp->netdev)) napi_schedule(&tp->napi); - if (test_and_clear_bit(PHY_RESET, &tp->flags)) - rtl_phy_reset(tp); - mutex_unlock(&tp->control); out1: -- cgit v0.10.2 From c79262f3f8e06e75e960ff45b48b3c12b2d6b709 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Thu, 7 Jul 2016 15:09:19 +0800 Subject: r8152: remove a netif_carrier_off in rtl8152_open function After commit 90186af404ad ("r8152: fix lockup when runtime PM is enabled"), the autoresume wouldn't start the device before rtl8152_open() is finished. Therefore, we don't have to reset the linking status before and after autoresume. That is, one of netif_carrier_off() in rtl8152_open() could be removed. Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index e005ee6..f1c5020 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -3111,8 +3111,6 @@ static int rtl8152_open(struct net_device *netdev) if (res) goto out; - netif_carrier_off(netdev); - res = usb_autopm_get_interface(tp->intf); if (res < 0) { free_all_mem(tp); -- cgit v0.10.2 From c23d86ae94b5eda449e71560e9028907abc91764 Mon Sep 17 00:00:00 2001 From: hayeswang Date: Thu, 7 Jul 2016 15:09:20 +0800 Subject: r8152: remove cancel_delayed_work_sync in rtl8152_set_speed There is no conflict between the work_queue function and rtl8152_set_speed(), so we don't have to cancel the delayed work in rtl8152_set_speed(). Signed-off-by: Hayes Wang Signed-off-by: David S. Miller diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c index f1c5020..168a8e2 100644 --- a/drivers/net/usb/r8152.c +++ b/drivers/net/usb/r8152.c @@ -2829,7 +2829,6 @@ static int rtl8152_set_speed(struct r8152 *tp, u8 autoneg, u16 speed, u8 duplex) u16 bmcr, anar, gbcr; int ret = 0; - cancel_delayed_work_sync(&tp->schedule); anar = r8152_mdio_read(tp, MII_ADVERTISE); anar &= ~(ADVERTISE_10HALF | ADVERTISE_10FULL | ADVERTISE_100HALF | ADVERTISE_100FULL); -- cgit v0.10.2 From 1db19db7f5ff4ddd3b1b6dd2092a87298ee5bd0b Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Thu, 7 Jul 2016 18:01:32 +0200 Subject: net: tracepoint napi:napi_poll add work and budget An important information for the napi_poll tracepoint is knowing the work done (packets processed) by the napi_poll() call. Add both the work done and budget, as they are related. Handle trace_napi_poll() param change in dropwatch/drop_monitor and in python perf script netdev-times.py in backward compat way, as python fortunately supports optional parameter handling. Signed-off-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller diff --git a/include/trace/events/napi.h b/include/trace/events/napi.h index 8fe1e93..118ed77 100644 --- a/include/trace/events/napi.h +++ b/include/trace/events/napi.h @@ -12,22 +12,27 @@ TRACE_EVENT(napi_poll, - TP_PROTO(struct napi_struct *napi), + TP_PROTO(struct napi_struct *napi, int work, int budget), - TP_ARGS(napi), + TP_ARGS(napi, work, budget), TP_STRUCT__entry( __field( struct napi_struct *, napi) + __field( int, work) + __field( int, budget) __string( dev_name, napi->dev ? napi->dev->name : NO_DEV) ), TP_fast_assign( __entry->napi = napi; + __entry->work = work; + __entry->budget = budget; __assign_str(dev_name, napi->dev ? napi->dev->name : NO_DEV); ), - TP_printk("napi poll on napi struct %p for device %s", - __entry->napi, __get_str(dev_name)) + TP_printk("napi poll on napi struct %p for device %s work %d budget %d", + __entry->napi, __get_str(dev_name), + __entry->work, __entry->budget) ); #undef NO_DEV diff --git a/net/core/dev.c b/net/core/dev.c index b92d63b..7894e40 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4972,7 +4972,7 @@ bool sk_busy_loop(struct sock *sk, int nonblock) if (test_bit(NAPI_STATE_SCHED, &napi->state)) { rc = napi->poll(napi, BUSY_POLL_BUDGET); - trace_napi_poll(napi); + trace_napi_poll(napi, rc, BUSY_POLL_BUDGET); if (rc == BUSY_POLL_BUDGET) { napi_complete_done(napi, rc); napi_schedule(napi); @@ -5128,7 +5128,7 @@ static int napi_poll(struct napi_struct *n, struct list_head *repoll) work = 0; if (test_bit(NAPI_STATE_SCHED, &n->state)) { work = n->poll(n, weight); - trace_napi_poll(n); + trace_napi_poll(n, work, weight); } WARN_ON_ONCE(work > weight); diff --git a/net/core/drop_monitor.c b/net/core/drop_monitor.c index 252e155..d6b3b57 100644 --- a/net/core/drop_monitor.c +++ b/net/core/drop_monitor.c @@ -187,7 +187,8 @@ static void trace_kfree_skb_hit(void *ignore, struct sk_buff *skb, void *locatio trace_drop_common(skb, location); } -static void trace_napi_poll_hit(void *ignore, struct napi_struct *napi) +static void trace_napi_poll_hit(void *ignore, struct napi_struct *napi, + int work, int budget) { struct dm_hw_stat_delta *new_stat; diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 94acfc8..53599bd 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -163,7 +163,7 @@ static void poll_one_napi(struct napi_struct *napi) */ work = napi->poll(napi, 0); WARN_ONCE(work, "%pF exceeded budget in poll\n", napi->poll); - trace_napi_poll(napi); + trace_napi_poll(napi, work, 0); clear_bit(NAPI_STATE_NPSVC, &napi->state); } diff --git a/tools/perf/scripts/python/netdev-times.py b/tools/perf/scripts/python/netdev-times.py index 4d21ef2..4c6f09a 100644 --- a/tools/perf/scripts/python/netdev-times.py +++ b/tools/perf/scripts/python/netdev-times.py @@ -252,9 +252,10 @@ def irq__irq_handler_exit(name, context, cpu, sec, nsec, pid, comm, callchain, i event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, irq, ret) all_event_list.append(event_info) -def napi__napi_poll(name, context, cpu, sec, nsec, pid, comm, callchain, napi, dev_name): +def napi__napi_poll(name, context, cpu, sec, nsec, pid, comm, callchain, napi, + dev_name, work=None, budget=None): event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, - napi, dev_name) + napi, dev_name, work, budget) all_event_list.append(event_info) def net__netif_receive_skb(name, context, cpu, sec, nsec, pid, comm, callchain, skbaddr, @@ -354,11 +355,13 @@ def handle_irq_softirq_exit(event_info): receive_hunk_list.append(rec_data) def handle_napi_poll(event_info): - (name, context, cpu, time, pid, comm, napi, dev_name) = event_info + (name, context, cpu, time, pid, comm, napi, dev_name, + work, budget) = event_info if cpu in net_rx_dic.keys(): event_list = net_rx_dic[cpu]['event_list'] rec_data = {'event_name':'napi_poll', - 'dev':dev_name, 'event_t':time} + 'dev':dev_name, 'event_t':time, + 'work':work, 'budget':budget} event_list.append(rec_data) def handle_netif_rx(event_info): -- cgit v0.10.2 From d00a8281bcc962027dfe409c2f3e3f0546be9200 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Thu, 7 Jul 2016 19:08:53 -0400 Subject: net: ethernet: bgmac: change bgmac_* prints to dev_* prints The bgmac_* print wrappers call dev_* prints with the dev pointer from the bcma core. In anticipation of removing the bcma requirement for this driver, these must be changed to not reference that struct. So, simply change all of the bgmac_* prints to their dev_* counterparts. In some cases netdev_* prints are more appropriate, so change those as well. Signed-off-by: Jon Mason Acked-by: Arnd Bergmann Reviewed-by: Florian Fainelli Tested-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index b045dc0..eb09de2 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -50,7 +50,7 @@ static bool bgmac_wait_value(struct bcma_device *core, u16 reg, u32 mask, return true; udelay(10); } - pr_err("Timeout waiting for reg 0x%X\n", reg); + dev_err(&core->dev, "Timeout waiting for reg 0x%X\n", reg); return false; } @@ -84,8 +84,8 @@ static void bgmac_dma_tx_reset(struct bgmac *bgmac, struct bgmac_dma_ring *ring) udelay(10); } if (i) - bgmac_err(bgmac, "Timeout suspending DMA TX ring 0x%X (BGMAC_DMA_TX_STAT: 0x%08X)\n", - ring->mmio_base, val); + dev_err(bgmac->dev, "Timeout suspending DMA TX ring 0x%X (BGMAC_DMA_TX_STAT: 0x%08X)\n", + ring->mmio_base, val); /* Remove SUSPEND bit */ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL, 0); @@ -93,13 +93,13 @@ static void bgmac_dma_tx_reset(struct bgmac *bgmac, struct bgmac_dma_ring *ring) ring->mmio_base + BGMAC_DMA_TX_STATUS, BGMAC_DMA_TX_STAT, BGMAC_DMA_TX_STAT_DISABLED, 10000)) { - bgmac_warn(bgmac, "DMA TX ring 0x%X wasn't disabled on time, waiting additional 300us\n", - ring->mmio_base); + dev_warn(bgmac->dev, "DMA TX ring 0x%X wasn't disabled on time, waiting additional 300us\n", + ring->mmio_base); udelay(300); val = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_STATUS); if ((val & BGMAC_DMA_TX_STAT) != BGMAC_DMA_TX_STAT_DISABLED) - bgmac_err(bgmac, "Reset of DMA TX ring 0x%X failed\n", - ring->mmio_base); + dev_err(bgmac->dev, "Reset of DMA TX ring 0x%X failed\n", + ring->mmio_base); } } @@ -161,7 +161,7 @@ static netdev_tx_t bgmac_dma_tx_add(struct bgmac *bgmac, int i; if (skb->len > BGMAC_DESC_CTL1_LEN) { - bgmac_err(bgmac, "Too long skb (%d)\n", skb->len); + netdev_err(bgmac->net_dev, "Too long skb (%d)\n", skb->len); goto err_drop; } @@ -174,7 +174,7 @@ static netdev_tx_t bgmac_dma_tx_add(struct bgmac *bgmac, * even when ring->end overflows */ if (ring->end - ring->start + nr_frags + 1 >= BGMAC_TX_RING_SLOTS) { - bgmac_err(bgmac, "TX ring is full, queue should be stopped!\n"); + netdev_err(bgmac->net_dev, "TX ring is full, queue should be stopped!\n"); netif_stop_queue(net_dev); return NETDEV_TX_BUSY; } @@ -241,8 +241,8 @@ err_dma: } err_dma_head: - bgmac_err(bgmac, "Mapping error of skb on ring 0x%X\n", - ring->mmio_base); + netdev_err(bgmac->net_dev, "Mapping error of skb on ring 0x%X\n", + ring->mmio_base); err_drop: dev_kfree_skb(skb); @@ -321,8 +321,8 @@ static void bgmac_dma_rx_reset(struct bgmac *bgmac, struct bgmac_dma_ring *ring) ring->mmio_base + BGMAC_DMA_RX_STATUS, BGMAC_DMA_RX_STAT, BGMAC_DMA_RX_STAT_DISABLED, 10000)) - bgmac_err(bgmac, "Reset of ring 0x%X RX failed\n", - ring->mmio_base); + dev_err(bgmac->dev, "Reset of ring 0x%X RX failed\n", + ring->mmio_base); } static void bgmac_dma_rx_enable(struct bgmac *bgmac, @@ -371,7 +371,7 @@ static int bgmac_dma_rx_skb_for_slot(struct bgmac *bgmac, dma_addr = dma_map_single(dma_dev, buf + BGMAC_RX_BUF_OFFSET, BGMAC_RX_BUF_SIZE, DMA_FROM_DEVICE); if (dma_mapping_error(dma_dev, dma_addr)) { - bgmac_err(bgmac, "DMA mapping error\n"); + netdev_err(bgmac->net_dev, "DMA mapping error\n"); put_page(virt_to_head_page(buf)); return -ENOMEM; } @@ -466,16 +466,16 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, /* Check for poison and drop or pass the packet */ if (len == 0xdead && flags == 0xbeef) { - bgmac_err(bgmac, "Found poisoned packet at slot %d, DMA issue!\n", - ring->start); + netdev_err(bgmac->net_dev, "Found poisoned packet at slot %d, DMA issue!\n", + ring->start); put_page(virt_to_head_page(buf)); bgmac->net_dev->stats.rx_errors++; break; } if (len > BGMAC_RX_ALLOC_SIZE) { - bgmac_err(bgmac, "Found oversized packet at slot %d, DMA issue!\n", - ring->start); + netdev_err(bgmac->net_dev, "Found oversized packet at slot %d, DMA issue!\n", + ring->start); put_page(virt_to_head_page(buf)); bgmac->net_dev->stats.rx_length_errors++; bgmac->net_dev->stats.rx_errors++; @@ -487,7 +487,7 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, skb = build_skb(buf, BGMAC_RX_ALLOC_SIZE); if (unlikely(!skb)) { - bgmac_err(bgmac, "build_skb failed\n"); + netdev_err(bgmac->net_dev, "build_skb failed\n"); put_page(virt_to_head_page(buf)); bgmac->net_dev->stats.rx_errors++; break; @@ -641,7 +641,7 @@ static int bgmac_dma_alloc(struct bgmac *bgmac) BUILD_BUG_ON(BGMAC_MAX_RX_RINGS > ARRAY_SIZE(ring_base)); if (!(bcma_aread32(bgmac->core, BCMA_IOST) & BCMA_IOST_DMA64)) { - bgmac_err(bgmac, "Core does not report 64-bit DMA\n"); + dev_err(bgmac->dev, "Core does not report 64-bit DMA\n"); return -ENOTSUPP; } @@ -655,8 +655,8 @@ static int bgmac_dma_alloc(struct bgmac *bgmac) &ring->dma_base, GFP_KERNEL); if (!ring->cpu_base) { - bgmac_err(bgmac, "Allocation of TX ring 0x%X failed\n", - ring->mmio_base); + dev_err(bgmac->dev, "Allocation of TX ring 0x%X failed\n", + ring->mmio_base); goto err_dma_free; } @@ -680,8 +680,8 @@ static int bgmac_dma_alloc(struct bgmac *bgmac) &ring->dma_base, GFP_KERNEL); if (!ring->cpu_base) { - bgmac_err(bgmac, "Allocation of RX ring 0x%X failed\n", - ring->mmio_base); + dev_err(bgmac->dev, "Allocation of RX ring 0x%X failed\n", + ring->mmio_base); err = -ENOMEM; goto err_dma_free; } @@ -800,8 +800,8 @@ static u16 bgmac_phy_read(struct bgmac *bgmac, u8 phyaddr, u8 reg) bcma_write32(core, phy_access_addr, tmp); if (!bgmac_wait_value(core, phy_access_addr, BGMAC_PA_START, 0, 1000)) { - bgmac_err(bgmac, "Reading PHY %d register 0x%X failed\n", - phyaddr, reg); + dev_err(bgmac->dev, "Reading PHY %d register 0x%X failed\n", + phyaddr, reg); return 0xffff; } @@ -833,7 +833,7 @@ static int bgmac_phy_write(struct bgmac *bgmac, u8 phyaddr, u8 reg, u16 value) bgmac_write(bgmac, BGMAC_INT_STATUS, BGMAC_IS_MDIO); if (bgmac_read(bgmac, BGMAC_INT_STATUS) & BGMAC_IS_MDIO) - bgmac_warn(bgmac, "Error setting MDIO int\n"); + dev_warn(bgmac->dev, "Error setting MDIO int\n"); tmp = BGMAC_PA_START; tmp |= BGMAC_PA_WRITE; @@ -843,8 +843,8 @@ static int bgmac_phy_write(struct bgmac *bgmac, u8 phyaddr, u8 reg, u16 value) bcma_write32(core, phy_access_addr, tmp); if (!bgmac_wait_value(core, phy_access_addr, BGMAC_PA_START, 0, 1000)) { - bgmac_err(bgmac, "Writing to PHY %d register 0x%X failed\n", - phyaddr, reg); + dev_err(bgmac->dev, "Writing to PHY %d register 0x%X failed\n", + phyaddr, reg); return -ETIMEDOUT; } @@ -897,7 +897,7 @@ static void bgmac_phy_reset(struct bgmac *bgmac) bgmac_phy_write(bgmac, bgmac->phyaddr, MII_BMCR, BMCR_RESET); udelay(100); if (bgmac_phy_read(bgmac, bgmac->phyaddr, MII_BMCR) & BMCR_RESET) - bgmac_err(bgmac, "PHY reset failed\n"); + dev_err(bgmac->dev, "PHY reset failed\n"); bgmac_phy_init(bgmac); } @@ -998,7 +998,8 @@ static void bgmac_mac_speed(struct bgmac *bgmac) set |= BGMAC_CMDCFG_ES_2500; break; default: - bgmac_err(bgmac, "Unsupported speed: %d\n", bgmac->mac_speed); + dev_err(bgmac->dev, "Unsupported speed: %d\n", + bgmac->mac_speed); } if (bgmac->mac_duplex == DUPLEX_HALF) @@ -1097,8 +1098,8 @@ static void bgmac_chip_reset(struct bgmac *bgmac) if (bcm47xx_nvram_getenv("et_swtype", buf, sizeof(buf)) > 0) { if (kstrtou8(buf, 0, &et_swtype)) - bgmac_err(bgmac, "Failed to parse et_swtype (%s)\n", - buf); + dev_err(bgmac->dev, "Failed to parse et_swtype (%s)\n", + buf); et_swtype &= 0x0f; et_swtype <<= 4; sw_type = et_swtype; @@ -1261,7 +1262,7 @@ static irqreturn_t bgmac_interrupt(int irq, void *dev_id) int_status &= ~(BGMAC_IS_TX0 | BGMAC_IS_RX); if (int_status) - bgmac_err(bgmac, "Unknown IRQs: 0x%08X\n", int_status); + dev_err(bgmac->dev, "Unknown IRQs: 0x%08X\n", int_status); /* Disable new interrupts until handling existing ones */ bgmac_chip_intrs_off(bgmac); @@ -1315,7 +1316,7 @@ static int bgmac_open(struct net_device *net_dev) err = request_irq(bgmac->core->irq, bgmac_interrupt, IRQF_SHARED, KBUILD_MODNAME, net_dev); if (err < 0) { - bgmac_err(bgmac, "IRQ request error: %d!\n", err); + dev_err(bgmac->dev, "IRQ request error: %d!\n", err); bgmac_dma_cleanup(bgmac); return err; } @@ -1580,14 +1581,14 @@ static int bgmac_fixed_phy_register(struct bgmac *bgmac) phy_dev = fixed_phy_register(PHY_POLL, &fphy_status, -1, NULL); if (!phy_dev || IS_ERR(phy_dev)) { - bgmac_err(bgmac, "Failed to register fixed PHY device\n"); + dev_err(bgmac->dev, "Failed to register fixed PHY device\n"); return -ENODEV; } err = phy_connect_direct(bgmac->net_dev, phy_dev, bgmac_adjust_link, PHY_INTERFACE_MODE_MII); if (err) { - bgmac_err(bgmac, "Connecting PHY failed\n"); + dev_err(bgmac->dev, "Connecting PHY failed\n"); return err; } @@ -1619,7 +1620,7 @@ static int bgmac_mii_register(struct bgmac *bgmac) err = mdiobus_register(mii_bus); if (err) { - bgmac_err(bgmac, "Registration of mii bus failed\n"); + dev_err(bgmac->dev, "Registration of mii bus failed\n"); goto err_free_bus; } @@ -1631,7 +1632,7 @@ static int bgmac_mii_register(struct bgmac *bgmac) phy_dev = phy_connect(bgmac->net_dev, bus_id, &bgmac_adjust_link, PHY_INTERFACE_MODE_MII); if (IS_ERR(phy_dev)) { - bgmac_err(bgmac, "PHY connection failed\n"); + dev_err(bgmac->dev, "PHY connecton failed\n"); err = PTR_ERR(phy_dev); goto err_unregister_bus; } @@ -1677,7 +1678,8 @@ static int bgmac_probe(struct bcma_device *core) mac = sprom->et2mac; break; default: - pr_err("Unsupported core_unit %d\n", core->core_unit); + dev_err(&core->dev, "Unsupported core_unit %d\n", + core->core_unit); return -ENOTSUPP; } @@ -1700,6 +1702,7 @@ static int bgmac_probe(struct bcma_device *core) net_dev->irq = core->irq; net_dev->ethtool_ops = &bgmac_ethtool_ops; bgmac = netdev_priv(net_dev); + bgmac->dev = &core->dev; bgmac->net_dev = net_dev; bgmac->core = core; bcma_set_drvdata(core, bgmac); @@ -1711,7 +1714,7 @@ static int bgmac_probe(struct bcma_device *core) /* On BCM4706 we need common core to access PHY */ if (core->id.id == BCMA_CORE_4706_MAC_GBIT && !core->bus->drv_gmac_cmn.core) { - bgmac_err(bgmac, "GMAC CMN core not found (required for BCM4706)\n"); + dev_err(bgmac->dev, "GMAC CMN core not found (required for BCM4706)\n"); err = -ENODEV; goto err_netdev_free; } @@ -1730,15 +1733,15 @@ static int bgmac_probe(struct bcma_device *core) } bgmac->phyaddr &= BGMAC_PHY_MASK; if (bgmac->phyaddr == BGMAC_PHY_MASK) { - bgmac_err(bgmac, "No PHY found\n"); + dev_err(bgmac->dev, "No PHY found\n"); err = -ENODEV; goto err_netdev_free; } - bgmac_info(bgmac, "Found PHY addr: %d%s\n", bgmac->phyaddr, - bgmac->phyaddr == BGMAC_PHY_NOREGS ? " (NOREGS)" : ""); + dev_info(bgmac->dev, "Found PHY addr: %d%s\n", bgmac->phyaddr, + bgmac->phyaddr == BGMAC_PHY_NOREGS ? " (NOREGS)" : ""); if (core->bus->hosttype == BCMA_HOSTTYPE_PCI) { - bgmac_err(bgmac, "PCI setup not implemented\n"); + dev_err(bgmac->dev, "PCI setup not implemented\n"); err = -ENOTSUPP; goto err_netdev_free; } @@ -1767,7 +1770,7 @@ static int bgmac_probe(struct bcma_device *core) err = bgmac_dma_alloc(bgmac); if (err) { - bgmac_err(bgmac, "Unable to alloc memory for DMA\n"); + dev_err(bgmac->dev, "Unable to alloc memory for DMA\n"); goto err_netdev_free; } @@ -1781,16 +1784,16 @@ static int bgmac_probe(struct bcma_device *core) bgmac->has_robosw = !!(core->bus->sprom.boardflags_lo & BGMAC_BFL_ENETROBO); if (bgmac->has_robosw) - bgmac_warn(bgmac, "Support for Roboswitch not implemented\n"); + dev_warn(bgmac->dev, "Support for Roboswitch not implemented\n"); if (core->bus->sprom.boardflags_lo & BGMAC_BFL_ENETADM) - bgmac_warn(bgmac, "Support for ADMtek ethernet switch not implemented\n"); + dev_warn(bgmac->dev, "Support for ADMtek ethernet switch not implemented\n"); netif_napi_add(net_dev, &bgmac->napi, bgmac_poll, BGMAC_WEIGHT); err = bgmac_mii_register(bgmac); if (err) { - bgmac_err(bgmac, "Cannot register MDIO\n"); + dev_err(bgmac->dev, "Cannot connect to phy\n"); goto err_dma_free; } @@ -1800,7 +1803,7 @@ static int bgmac_probe(struct bcma_device *core) err = register_netdev(bgmac->net_dev); if (err) { - bgmac_err(bgmac, "Cannot register net device\n"); + dev_err(bgmac->dev, "Cannot register net device\n"); goto err_mii_unregister; } diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h index 99beb18..abb9dd8 100644 --- a/drivers/net/ethernet/broadcom/bgmac.h +++ b/drivers/net/ethernet/broadcom/bgmac.h @@ -1,17 +1,6 @@ #ifndef _BGMAC_H #define _BGMAC_H -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#define bgmac_err(bgmac, fmt, ...) \ - dev_err(&(bgmac)->core->dev, fmt, ##__VA_ARGS__) -#define bgmac_warn(bgmac, fmt, ...) \ - dev_warn(&(bgmac)->core->dev, fmt, ##__VA_ARGS__) -#define bgmac_info(bgmac, fmt, ...) \ - dev_info(&(bgmac)->core->dev, fmt, ##__VA_ARGS__) -#define bgmac_dbg(bgmac, fmt, ...) \ - dev_dbg(&(bgmac)->core->dev, fmt, ##__VA_ARGS__) - #include #include #include @@ -438,6 +427,8 @@ struct bgmac_rx_header { struct bgmac { struct bcma_device *core; struct bcma_device *cmn; /* Reference to CMN core for BCM4706 */ + + struct device *dev; struct net_device *net_dev; struct napi_struct napi; struct mii_bus *mii_bus; @@ -489,5 +480,4 @@ static inline void bgmac_set(struct bgmac *bgmac, u16 offset, u32 set) { bgmac_maskset(bgmac, offset, ~0, set); } - #endif /* _BGMAC_H */ -- cgit v0.10.2 From a0b68486f6f680c7c0352a47c60042d7d95ffd87 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Thu, 7 Jul 2016 19:08:54 -0400 Subject: net: ethernet: bgmac: add dma_dev pointer The dma buffer allocation, etc references a dma_dev device pointer from the bcma core. In anticipation of removing the bcma requirement for this driver, these must be changed to not reference that struct. Add a dma_dev device pointer to the bgmac stuct and reference that instead. Signed-off-by: Jon Mason Acked-by: Arnd Bergmann Reviewed-by: Florian Fainelli Tested-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index eb09de2..556c87c 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -152,7 +152,7 @@ static netdev_tx_t bgmac_dma_tx_add(struct bgmac *bgmac, struct bgmac_dma_ring *ring, struct sk_buff *skb) { - struct device *dma_dev = bgmac->core->dma_dev; + struct device *dma_dev = bgmac->dma_dev; struct net_device *net_dev = bgmac->net_dev; int index = ring->end % BGMAC_TX_RING_SLOTS; struct bgmac_slot_info *slot = &ring->slots[index]; @@ -254,7 +254,7 @@ err_drop: /* Free transmitted packets */ static void bgmac_dma_tx_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring) { - struct device *dma_dev = bgmac->core->dma_dev; + struct device *dma_dev = bgmac->dma_dev; int empty_slot; bool freed = false; unsigned bytes_compl = 0, pkts_compl = 0; @@ -352,7 +352,7 @@ static void bgmac_dma_rx_enable(struct bgmac *bgmac, static int bgmac_dma_rx_skb_for_slot(struct bgmac *bgmac, struct bgmac_slot_info *slot) { - struct device *dma_dev = bgmac->core->dma_dev; + struct device *dma_dev = bgmac->dma_dev; dma_addr_t dma_addr; struct bgmac_rx_header *rx; void *buf; @@ -441,7 +441,7 @@ static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring, end_slot /= sizeof(struct bgmac_dma_desc); while (ring->start != end_slot) { - struct device *dma_dev = bgmac->core->dma_dev; + struct device *dma_dev = bgmac->dma_dev; struct bgmac_slot_info *slot = &ring->slots[ring->start]; struct bgmac_rx_header *rx = slot->buf + BGMAC_RX_BUF_OFFSET; struct sk_buff *skb; @@ -544,7 +544,7 @@ static bool bgmac_dma_unaligned(struct bgmac *bgmac, static void bgmac_dma_tx_ring_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring) { - struct device *dma_dev = bgmac->core->dma_dev; + struct device *dma_dev = bgmac->dma_dev; struct bgmac_dma_desc *dma_desc = ring->cpu_base; struct bgmac_slot_info *slot; int i; @@ -570,7 +570,7 @@ static void bgmac_dma_tx_ring_free(struct bgmac *bgmac, static void bgmac_dma_rx_ring_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring) { - struct device *dma_dev = bgmac->core->dma_dev; + struct device *dma_dev = bgmac->dma_dev; struct bgmac_slot_info *slot; int i; @@ -591,7 +591,7 @@ static void bgmac_dma_ring_desc_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring, int num_slots) { - struct device *dma_dev = bgmac->core->dma_dev; + struct device *dma_dev = bgmac->dma_dev; int size; if (!ring->cpu_base) @@ -629,7 +629,7 @@ static void bgmac_dma_free(struct bgmac *bgmac) static int bgmac_dma_alloc(struct bgmac *bgmac) { - struct device *dma_dev = bgmac->core->dma_dev; + struct device *dma_dev = bgmac->dma_dev; struct bgmac_dma_ring *ring; static const u16 ring_base[] = { BGMAC_DMA_BASE0, BGMAC_DMA_BASE1, BGMAC_DMA_BASE2, BGMAC_DMA_BASE3, }; @@ -1703,6 +1703,7 @@ static int bgmac_probe(struct bcma_device *core) net_dev->ethtool_ops = &bgmac_ethtool_ops; bgmac = netdev_priv(net_dev); bgmac->dev = &core->dev; + bgmac->dma_dev = core->dma_dev; bgmac->net_dev = net_dev; bgmac->core = core; bcma_set_drvdata(core, bgmac); diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h index abb9dd8..fd20018 100644 --- a/drivers/net/ethernet/broadcom/bgmac.h +++ b/drivers/net/ethernet/broadcom/bgmac.h @@ -429,6 +429,7 @@ struct bgmac { struct bcma_device *cmn; /* Reference to CMN core for BCM4706 */ struct device *dev; + struct device *dma_dev; struct net_device *net_dev; struct napi_struct napi; struct mii_bus *mii_bus; -- cgit v0.10.2 From 55954f3bfdacc5908515b0c306cea23e77fab740 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Thu, 7 Jul 2016 19:08:55 -0400 Subject: net: ethernet: bgmac: move BCMA MDIO Phy code into a separate file Move the BCMA MDIO phy into a separate file, as it is very tightly coupled with the BCMA bus. This will help with the upcoming BCMA removal from the bgmac driver. Optimally, this should be moved into phy drivers, but it is too tightly coupled with the bgmac driver to effectively move it without more changes to the driver. Note: the phy_reset was intentionally removed, as the mdio phy subsystem automatically resets the phy if a reset function pointer is present. In addition to the moving of the driver, this reset function is added. Signed-off-by: Jon Mason Acked-by: Arnd Bergmann Reviewed-by: Florian Fainelli Tested-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/Makefile b/drivers/net/ethernet/broadcom/Makefile index 00584d7..f559794 100644 --- a/drivers/net/ethernet/broadcom/Makefile +++ b/drivers/net/ethernet/broadcom/Makefile @@ -10,6 +10,6 @@ obj-$(CONFIG_CNIC) += cnic.o obj-$(CONFIG_BNX2X) += bnx2x/ obj-$(CONFIG_SB1250_MAC) += sb1250-mac.o obj-$(CONFIG_TIGON3) += tg3.o -obj-$(CONFIG_BGMAC) += bgmac.o +obj-$(CONFIG_BGMAC) += bgmac.o bgmac-bcma-mdio.o obj-$(CONFIG_SYSTEMPORT) += bcmsysport.o obj-$(CONFIG_BNXT) += bnxt/ diff --git a/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c b/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c new file mode 100644 index 0000000..1e65349 --- /dev/null +++ b/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c @@ -0,0 +1,264 @@ +/* + * Driver for (BCM4706)? GBit MAC core on BCMA bus. + * + * Copyright (C) 2012 Rafał Miłecki + * + * Licensed under the GNU/GPL. See COPYING for details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include "bgmac.h" + +struct bcma_mdio { + struct bcma_device *core; + u8 phyaddr; +}; + +static bool bcma_mdio_wait_value(struct bcma_device *core, u16 reg, u32 mask, + u32 value, int timeout) +{ + u32 val; + int i; + + for (i = 0; i < timeout / 10; i++) { + val = bcma_read32(core, reg); + if ((val & mask) == value) + return true; + udelay(10); + } + dev_err(&core->dev, "Timeout waiting for reg 0x%X\n", reg); + return false; +} + +/************************************************** + * PHY ops + **************************************************/ + +static u16 bcma_mdio_phy_read(struct bcma_mdio *bcma_mdio, u8 phyaddr, u8 reg) +{ + struct bcma_device *core; + u16 phy_access_addr; + u16 phy_ctl_addr; + u32 tmp; + + BUILD_BUG_ON(BGMAC_PA_DATA_MASK != BCMA_GMAC_CMN_PA_DATA_MASK); + BUILD_BUG_ON(BGMAC_PA_ADDR_MASK != BCMA_GMAC_CMN_PA_ADDR_MASK); + BUILD_BUG_ON(BGMAC_PA_ADDR_SHIFT != BCMA_GMAC_CMN_PA_ADDR_SHIFT); + BUILD_BUG_ON(BGMAC_PA_REG_MASK != BCMA_GMAC_CMN_PA_REG_MASK); + BUILD_BUG_ON(BGMAC_PA_REG_SHIFT != BCMA_GMAC_CMN_PA_REG_SHIFT); + BUILD_BUG_ON(BGMAC_PA_WRITE != BCMA_GMAC_CMN_PA_WRITE); + BUILD_BUG_ON(BGMAC_PA_START != BCMA_GMAC_CMN_PA_START); + BUILD_BUG_ON(BGMAC_PC_EPA_MASK != BCMA_GMAC_CMN_PC_EPA_MASK); + BUILD_BUG_ON(BGMAC_PC_MCT_MASK != BCMA_GMAC_CMN_PC_MCT_MASK); + BUILD_BUG_ON(BGMAC_PC_MCT_SHIFT != BCMA_GMAC_CMN_PC_MCT_SHIFT); + BUILD_BUG_ON(BGMAC_PC_MTE != BCMA_GMAC_CMN_PC_MTE); + + if (bcma_mdio->core->id.id == BCMA_CORE_4706_MAC_GBIT) { + core = bcma_mdio->core->bus->drv_gmac_cmn.core; + phy_access_addr = BCMA_GMAC_CMN_PHY_ACCESS; + phy_ctl_addr = BCMA_GMAC_CMN_PHY_CTL; + } else { + core = bcma_mdio->core; + phy_access_addr = BGMAC_PHY_ACCESS; + phy_ctl_addr = BGMAC_PHY_CNTL; + } + + tmp = bcma_read32(core, phy_ctl_addr); + tmp &= ~BGMAC_PC_EPA_MASK; + tmp |= phyaddr; + bcma_write32(core, phy_ctl_addr, tmp); + + tmp = BGMAC_PA_START; + tmp |= phyaddr << BGMAC_PA_ADDR_SHIFT; + tmp |= reg << BGMAC_PA_REG_SHIFT; + bcma_write32(core, phy_access_addr, tmp); + + if (!bcma_mdio_wait_value(core, phy_access_addr, BGMAC_PA_START, 0, + 1000)) { + dev_err(&core->dev, "Reading PHY %d register 0x%X failed\n", + phyaddr, reg); + return 0xffff; + } + + return bcma_read32(core, phy_access_addr) & BGMAC_PA_DATA_MASK; +} + +/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphywr */ +static int bcma_mdio_phy_write(struct bcma_mdio *bcma_mdio, u8 phyaddr, u8 reg, + u16 value) +{ + struct bcma_device *core; + u16 phy_access_addr; + u16 phy_ctl_addr; + u32 tmp; + + if (bcma_mdio->core->id.id == BCMA_CORE_4706_MAC_GBIT) { + core = bcma_mdio->core->bus->drv_gmac_cmn.core; + phy_access_addr = BCMA_GMAC_CMN_PHY_ACCESS; + phy_ctl_addr = BCMA_GMAC_CMN_PHY_CTL; + } else { + core = bcma_mdio->core; + phy_access_addr = BGMAC_PHY_ACCESS; + phy_ctl_addr = BGMAC_PHY_CNTL; + } + + tmp = bcma_read32(core, phy_ctl_addr); + tmp &= ~BGMAC_PC_EPA_MASK; + tmp |= phyaddr; + bcma_write32(core, phy_ctl_addr, tmp); + + bcma_write32(bcma_mdio->core, BGMAC_INT_STATUS, BGMAC_IS_MDIO); + if (bcma_read32(bcma_mdio->core, BGMAC_INT_STATUS) & BGMAC_IS_MDIO) + dev_warn(&core->dev, "Error setting MDIO int\n"); + + tmp = BGMAC_PA_START; + tmp |= BGMAC_PA_WRITE; + tmp |= phyaddr << BGMAC_PA_ADDR_SHIFT; + tmp |= reg << BGMAC_PA_REG_SHIFT; + tmp |= value; + bcma_write32(core, phy_access_addr, tmp); + + if (!bcma_mdio_wait_value(core, phy_access_addr, BGMAC_PA_START, 0, + 1000)) { + dev_err(&core->dev, "Writing to PHY %d register 0x%X failed\n", + phyaddr, reg); + return -ETIMEDOUT; + } + + return 0; +} + +/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphyinit */ +static void bcma_mdio_phy_init(struct bcma_mdio *bcma_mdio) +{ + struct bcma_chipinfo *ci = &bcma_mdio->core->bus->chipinfo; + u8 i; + + if (ci->id == BCMA_CHIP_ID_BCM5356) { + for (i = 0; i < 5; i++) { + bcma_mdio_phy_write(bcma_mdio, i, 0x1f, 0x008b); + bcma_mdio_phy_write(bcma_mdio, i, 0x15, 0x0100); + bcma_mdio_phy_write(bcma_mdio, i, 0x1f, 0x000f); + bcma_mdio_phy_write(bcma_mdio, i, 0x12, 0x2aaa); + bcma_mdio_phy_write(bcma_mdio, i, 0x1f, 0x000b); + } + } + if ((ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg != 10) || + (ci->id == BCMA_CHIP_ID_BCM4749 && ci->pkg != 10) || + (ci->id == BCMA_CHIP_ID_BCM53572 && ci->pkg != 9)) { + struct bcma_drv_cc *cc = &bcma_mdio->core->bus->drv_cc; + + bcma_chipco_chipctl_maskset(cc, 2, ~0xc0000000, 0); + bcma_chipco_chipctl_maskset(cc, 4, ~0x80000000, 0); + for (i = 0; i < 5; i++) { + bcma_mdio_phy_write(bcma_mdio, i, 0x1f, 0x000f); + bcma_mdio_phy_write(bcma_mdio, i, 0x16, 0x5284); + bcma_mdio_phy_write(bcma_mdio, i, 0x1f, 0x000b); + bcma_mdio_phy_write(bcma_mdio, i, 0x17, 0x0010); + bcma_mdio_phy_write(bcma_mdio, i, 0x1f, 0x000f); + bcma_mdio_phy_write(bcma_mdio, i, 0x16, 0x5296); + bcma_mdio_phy_write(bcma_mdio, i, 0x17, 0x1073); + bcma_mdio_phy_write(bcma_mdio, i, 0x17, 0x9073); + bcma_mdio_phy_write(bcma_mdio, i, 0x16, 0x52b6); + bcma_mdio_phy_write(bcma_mdio, i, 0x17, 0x9273); + bcma_mdio_phy_write(bcma_mdio, i, 0x1f, 0x000b); + } + } +} + +/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphyreset */ +static int bcma_mdio_phy_reset(struct mii_bus *bus) +{ + struct bcma_mdio *bcma_mdio = bus->priv; + u8 phyaddr = bcma_mdio->phyaddr; + + if (bcma_mdio->phyaddr == BGMAC_PHY_NOREGS) + return 0; + + bcma_mdio_phy_write(bcma_mdio, phyaddr, MII_BMCR, BMCR_RESET); + udelay(100); + if (bcma_mdio_phy_read(bcma_mdio, phyaddr, MII_BMCR) & BMCR_RESET) + dev_err(&bcma_mdio->core->dev, "PHY reset failed\n"); + bcma_mdio_phy_init(bcma_mdio); + + return 0; +} + +/************************************************** + * MII + **************************************************/ + +static int bcma_mdio_mii_read(struct mii_bus *bus, int mii_id, int regnum) +{ + return bcma_mdio_phy_read(bus->priv, mii_id, regnum); +} + +static int bcma_mdio_mii_write(struct mii_bus *bus, int mii_id, int regnum, + u16 value) +{ + return bcma_mdio_phy_write(bus->priv, mii_id, regnum, value); +} + +struct mii_bus *bcma_mdio_mii_register(struct bcma_device *core, u8 phyaddr) +{ + struct bcma_mdio *bcma_mdio; + struct mii_bus *mii_bus; + int err; + + bcma_mdio = kzalloc(sizeof(*bcma_mdio), GFP_KERNEL); + if (!bcma_mdio) + return ERR_PTR(-ENOMEM); + + mii_bus = mdiobus_alloc(); + if (!mii_bus) { + err = -ENOMEM; + goto err; + } + + mii_bus->name = "bcma_mdio mii bus"; + sprintf(mii_bus->id, "%s-%d-%d", "bcma_mdio", core->bus->num, + core->core_unit); + mii_bus->priv = bcma_mdio; + mii_bus->read = bcma_mdio_mii_read; + mii_bus->write = bcma_mdio_mii_write; + mii_bus->reset = bcma_mdio_phy_reset; + mii_bus->parent = &core->dev; + mii_bus->phy_mask = ~(1 << phyaddr); + + bcma_mdio->core = core; + bcma_mdio->phyaddr = phyaddr; + + err = mdiobus_register(mii_bus); + if (err) { + dev_err(&core->dev, "Registration of mii bus failed\n"); + goto err_free_bus; + } + + return mii_bus; + +err_free_bus: + mdiobus_free(mii_bus); +err: + kfree(bcma_mdio); + return ERR_PTR(err); +} + +void bcma_mdio_mii_unregister(struct mii_bus *mii_bus) +{ + struct bcma_mdio *bcma_mdio; + + if (!mii_bus) + return; + + bcma_mdio = mii_bus->priv; + + mdiobus_unregister(mii_bus); + mdiobus_free(mii_bus); + kfree(bcma_mdio); +} + +MODULE_AUTHOR("Rafał Miłecki"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index 556c87c..5117470 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -756,150 +756,6 @@ error: return err; } -/************************************************** - * PHY ops - **************************************************/ - -static u16 bgmac_phy_read(struct bgmac *bgmac, u8 phyaddr, u8 reg) -{ - struct bcma_device *core; - u16 phy_access_addr; - u16 phy_ctl_addr; - u32 tmp; - - BUILD_BUG_ON(BGMAC_PA_DATA_MASK != BCMA_GMAC_CMN_PA_DATA_MASK); - BUILD_BUG_ON(BGMAC_PA_ADDR_MASK != BCMA_GMAC_CMN_PA_ADDR_MASK); - BUILD_BUG_ON(BGMAC_PA_ADDR_SHIFT != BCMA_GMAC_CMN_PA_ADDR_SHIFT); - BUILD_BUG_ON(BGMAC_PA_REG_MASK != BCMA_GMAC_CMN_PA_REG_MASK); - BUILD_BUG_ON(BGMAC_PA_REG_SHIFT != BCMA_GMAC_CMN_PA_REG_SHIFT); - BUILD_BUG_ON(BGMAC_PA_WRITE != BCMA_GMAC_CMN_PA_WRITE); - BUILD_BUG_ON(BGMAC_PA_START != BCMA_GMAC_CMN_PA_START); - BUILD_BUG_ON(BGMAC_PC_EPA_MASK != BCMA_GMAC_CMN_PC_EPA_MASK); - BUILD_BUG_ON(BGMAC_PC_MCT_MASK != BCMA_GMAC_CMN_PC_MCT_MASK); - BUILD_BUG_ON(BGMAC_PC_MCT_SHIFT != BCMA_GMAC_CMN_PC_MCT_SHIFT); - BUILD_BUG_ON(BGMAC_PC_MTE != BCMA_GMAC_CMN_PC_MTE); - - if (bgmac->core->id.id == BCMA_CORE_4706_MAC_GBIT) { - core = bgmac->core->bus->drv_gmac_cmn.core; - phy_access_addr = BCMA_GMAC_CMN_PHY_ACCESS; - phy_ctl_addr = BCMA_GMAC_CMN_PHY_CTL; - } else { - core = bgmac->core; - phy_access_addr = BGMAC_PHY_ACCESS; - phy_ctl_addr = BGMAC_PHY_CNTL; - } - - tmp = bcma_read32(core, phy_ctl_addr); - tmp &= ~BGMAC_PC_EPA_MASK; - tmp |= phyaddr; - bcma_write32(core, phy_ctl_addr, tmp); - - tmp = BGMAC_PA_START; - tmp |= phyaddr << BGMAC_PA_ADDR_SHIFT; - tmp |= reg << BGMAC_PA_REG_SHIFT; - bcma_write32(core, phy_access_addr, tmp); - - if (!bgmac_wait_value(core, phy_access_addr, BGMAC_PA_START, 0, 1000)) { - dev_err(bgmac->dev, "Reading PHY %d register 0x%X failed\n", - phyaddr, reg); - return 0xffff; - } - - return bcma_read32(core, phy_access_addr) & BGMAC_PA_DATA_MASK; -} - -/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphywr */ -static int bgmac_phy_write(struct bgmac *bgmac, u8 phyaddr, u8 reg, u16 value) -{ - struct bcma_device *core; - u16 phy_access_addr; - u16 phy_ctl_addr; - u32 tmp; - - if (bgmac->core->id.id == BCMA_CORE_4706_MAC_GBIT) { - core = bgmac->core->bus->drv_gmac_cmn.core; - phy_access_addr = BCMA_GMAC_CMN_PHY_ACCESS; - phy_ctl_addr = BCMA_GMAC_CMN_PHY_CTL; - } else { - core = bgmac->core; - phy_access_addr = BGMAC_PHY_ACCESS; - phy_ctl_addr = BGMAC_PHY_CNTL; - } - - tmp = bcma_read32(core, phy_ctl_addr); - tmp &= ~BGMAC_PC_EPA_MASK; - tmp |= phyaddr; - bcma_write32(core, phy_ctl_addr, tmp); - - bgmac_write(bgmac, BGMAC_INT_STATUS, BGMAC_IS_MDIO); - if (bgmac_read(bgmac, BGMAC_INT_STATUS) & BGMAC_IS_MDIO) - dev_warn(bgmac->dev, "Error setting MDIO int\n"); - - tmp = BGMAC_PA_START; - tmp |= BGMAC_PA_WRITE; - tmp |= phyaddr << BGMAC_PA_ADDR_SHIFT; - tmp |= reg << BGMAC_PA_REG_SHIFT; - tmp |= value; - bcma_write32(core, phy_access_addr, tmp); - - if (!bgmac_wait_value(core, phy_access_addr, BGMAC_PA_START, 0, 1000)) { - dev_err(bgmac->dev, "Writing to PHY %d register 0x%X failed\n", - phyaddr, reg); - return -ETIMEDOUT; - } - - return 0; -} - -/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphyinit */ -static void bgmac_phy_init(struct bgmac *bgmac) -{ - struct bcma_chipinfo *ci = &bgmac->core->bus->chipinfo; - struct bcma_drv_cc *cc = &bgmac->core->bus->drv_cc; - u8 i; - - if (ci->id == BCMA_CHIP_ID_BCM5356) { - for (i = 0; i < 5; i++) { - bgmac_phy_write(bgmac, i, 0x1f, 0x008b); - bgmac_phy_write(bgmac, i, 0x15, 0x0100); - bgmac_phy_write(bgmac, i, 0x1f, 0x000f); - bgmac_phy_write(bgmac, i, 0x12, 0x2aaa); - bgmac_phy_write(bgmac, i, 0x1f, 0x000b); - } - } - if ((ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg != 10) || - (ci->id == BCMA_CHIP_ID_BCM4749 && ci->pkg != 10) || - (ci->id == BCMA_CHIP_ID_BCM53572 && ci->pkg != 9)) { - bcma_chipco_chipctl_maskset(cc, 2, ~0xc0000000, 0); - bcma_chipco_chipctl_maskset(cc, 4, ~0x80000000, 0); - for (i = 0; i < 5; i++) { - bgmac_phy_write(bgmac, i, 0x1f, 0x000f); - bgmac_phy_write(bgmac, i, 0x16, 0x5284); - bgmac_phy_write(bgmac, i, 0x1f, 0x000b); - bgmac_phy_write(bgmac, i, 0x17, 0x0010); - bgmac_phy_write(bgmac, i, 0x1f, 0x000f); - bgmac_phy_write(bgmac, i, 0x16, 0x5296); - bgmac_phy_write(bgmac, i, 0x17, 0x1073); - bgmac_phy_write(bgmac, i, 0x17, 0x9073); - bgmac_phy_write(bgmac, i, 0x16, 0x52b6); - bgmac_phy_write(bgmac, i, 0x17, 0x9273); - bgmac_phy_write(bgmac, i, 0x1f, 0x000b); - } - } -} - -/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphyreset */ -static void bgmac_phy_reset(struct bgmac *bgmac) -{ - if (bgmac->phyaddr == BGMAC_PHY_NOREGS) - return; - - bgmac_phy_write(bgmac, bgmac->phyaddr, MII_BMCR, BMCR_RESET); - udelay(100); - if (bgmac_phy_read(bgmac, bgmac->phyaddr, MII_BMCR) & BMCR_RESET) - dev_err(bgmac->dev, "PHY reset failed\n"); - bgmac_phy_init(bgmac); -} /************************************************** * Chip ops @@ -1156,7 +1012,8 @@ static void bgmac_chip_reset(struct bgmac *bgmac) else bgmac_set(bgmac, BGMAC_PHY_CNTL, BGMAC_PC_MTE); bgmac_miiconfig(bgmac); - bgmac_phy_init(bgmac); + if (bgmac->mii_bus) + bgmac->mii_bus->reset(bgmac->mii_bus); netdev_reset_queue(bgmac->net_dev); } @@ -1534,17 +1391,6 @@ static const struct ethtool_ops bgmac_ethtool_ops = { * MII **************************************************/ -static int bgmac_mii_read(struct mii_bus *bus, int mii_id, int regnum) -{ - return bgmac_phy_read(bus->priv, mii_id, regnum); -} - -static int bgmac_mii_write(struct mii_bus *bus, int mii_id, int regnum, - u16 value) -{ - return bgmac_phy_write(bus->priv, mii_id, regnum, value); -} - static void bgmac_adjust_link(struct net_device *net_dev) { struct bgmac *bgmac = netdev_priv(net_dev); @@ -1569,7 +1415,7 @@ static void bgmac_adjust_link(struct net_device *net_dev) } } -static int bgmac_fixed_phy_register(struct bgmac *bgmac) +static int bgmac_phy_connect_direct(struct bgmac *bgmac) { struct fixed_phy_status fphy_status = { .link = 1, @@ -1595,70 +1441,24 @@ static int bgmac_fixed_phy_register(struct bgmac *bgmac) return err; } -static int bgmac_mii_register(struct bgmac *bgmac) +static int bgmac_phy_connect(struct bgmac *bgmac) { - struct mii_bus *mii_bus; struct phy_device *phy_dev; char bus_id[MII_BUS_ID_SIZE + 3]; - int err = 0; - - if (bgmac_is_bcm4707_family(bgmac)) - return bgmac_fixed_phy_register(bgmac); - - mii_bus = mdiobus_alloc(); - if (!mii_bus) - return -ENOMEM; - - mii_bus->name = "bgmac mii bus"; - sprintf(mii_bus->id, "%s-%d-%d", "bgmac", bgmac->core->bus->num, - bgmac->core->core_unit); - mii_bus->priv = bgmac; - mii_bus->read = bgmac_mii_read; - mii_bus->write = bgmac_mii_write; - mii_bus->parent = &bgmac->core->dev; - mii_bus->phy_mask = ~(1 << bgmac->phyaddr); - - err = mdiobus_register(mii_bus); - if (err) { - dev_err(bgmac->dev, "Registration of mii bus failed\n"); - goto err_free_bus; - } - - bgmac->mii_bus = mii_bus; /* Connect to the PHY */ - snprintf(bus_id, sizeof(bus_id), PHY_ID_FMT, mii_bus->id, + snprintf(bus_id, sizeof(bus_id), PHY_ID_FMT, bgmac->mii_bus->id, bgmac->phyaddr); phy_dev = phy_connect(bgmac->net_dev, bus_id, &bgmac_adjust_link, PHY_INTERFACE_MODE_MII); if (IS_ERR(phy_dev)) { dev_err(bgmac->dev, "PHY connecton failed\n"); - err = PTR_ERR(phy_dev); - goto err_unregister_bus; + return PTR_ERR(phy_dev); } - return err; - -err_unregister_bus: - mdiobus_unregister(mii_bus); -err_free_bus: - mdiobus_free(mii_bus); - return err; -} - -static void bgmac_mii_unregister(struct bgmac *bgmac) -{ - struct mii_bus *mii_bus = bgmac->mii_bus; - - mdiobus_unregister(mii_bus); - mdiobus_free(mii_bus); + return 0; } -/************************************************** - * BCMA bus ops - **************************************************/ - -/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipattach */ static int bgmac_probe(struct bcma_device *core) { struct net_device *net_dev; @@ -1779,9 +1579,6 @@ static int bgmac_probe(struct bcma_device *core) if (bcm47xx_nvram_getenv("et0_no_txint", NULL, 0) == 0) bgmac->int_mask &= ~BGMAC_IS_TX_MASK; - /* TODO: reset the external phy. Specs are needed */ - bgmac_phy_reset(bgmac); - bgmac->has_robosw = !!(core->bus->sprom.boardflags_lo & BGMAC_BFL_ENETROBO); if (bgmac->has_robosw) @@ -1792,10 +1589,25 @@ static int bgmac_probe(struct bcma_device *core) netif_napi_add(net_dev, &bgmac->napi, bgmac_poll, BGMAC_WEIGHT); - err = bgmac_mii_register(bgmac); + if (!bgmac_is_bcm4707_family(bgmac)) { + struct mii_bus *mii_bus; + + mii_bus = bcma_mdio_mii_register(core, bgmac->phyaddr); + if (!IS_ERR(mii_bus)) { + err = PTR_ERR(mii_bus); + goto err_dma_free; + } + + bgmac->mii_bus = mii_bus; + } + + if (!bgmac->mii_bus) + err = bgmac_phy_connect_direct(bgmac); + else + err = bgmac_phy_connect(bgmac); if (err) { dev_err(bgmac->dev, "Cannot connect to phy\n"); - goto err_dma_free; + goto err_mii_unregister; } net_dev->features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM; @@ -1805,18 +1617,19 @@ static int bgmac_probe(struct bcma_device *core) err = register_netdev(bgmac->net_dev); if (err) { dev_err(bgmac->dev, "Cannot register net device\n"); - goto err_mii_unregister; + goto err_phy_disconnect; } netif_carrier_off(net_dev); return 0; +err_phy_disconnect: + phy_disconnect(net_dev->phydev); err_mii_unregister: - bgmac_mii_unregister(bgmac); + bcma_mdio_mii_unregister(bgmac->mii_bus); err_dma_free: bgmac_dma_free(bgmac); - err_netdev_free: bcma_set_drvdata(core, NULL); free_netdev(net_dev); @@ -1829,7 +1642,8 @@ static void bgmac_remove(struct bcma_device *core) struct bgmac *bgmac = bcma_get_drvdata(core); unregister_netdev(bgmac->net_dev); - bgmac_mii_unregister(bgmac); + phy_disconnect(bgmac->net_dev->phydev); + bcma_mdio_mii_unregister(bgmac->mii_bus); netif_napi_del(&bgmac->napi); bgmac_dma_free(bgmac); bcma_set_drvdata(core, NULL); diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h index fd20018..191e64a 100644 --- a/drivers/net/ethernet/broadcom/bgmac.h +++ b/drivers/net/ethernet/broadcom/bgmac.h @@ -456,6 +456,9 @@ struct bgmac { bool loopback; }; +struct mii_bus *bcma_mdio_mii_register(struct bcma_device *core, u8 phyaddr); +void bcma_mdio_mii_unregister(struct mii_bus *mii_bus); + static inline u32 bgmac_read(struct bgmac *bgmac, u16 offset) { return bcma_read32(bgmac->core, offset); -- cgit v0.10.2 From db791eb2970bad193b1dc95a4461b222dd22cb64 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Thu, 7 Jul 2016 19:08:56 -0400 Subject: net: ethernet: bgmac: convert to feature flags The bgmac driver is using the bcma provides device ID and revision, as well as the SoC ID and package, to determine which features are necessary to enable, reset, etc in the driver. In anticipation of removing the bcma requirement for this driver, these must be changed to not reference that struct. In place of that, each "feature" has been given a flag, and the flags are enabled for their respective device and SoC. Signed-off-by: Jon Mason Acked-by: Arnd Bergmann Reviewed-by: Florian Fainelli Tested-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index 5117470..770796a 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -109,7 +109,7 @@ static void bgmac_dma_tx_enable(struct bgmac *bgmac, u32 ctl; ctl = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL); - if (bgmac->core->id.rev >= 4) { + if (bgmac->feature_flags & BGMAC_FEAT_TX_MASK_SETUP) { ctl &= ~BGMAC_DMA_TX_BL_MASK; ctl |= BGMAC_DMA_TX_BL_128 << BGMAC_DMA_TX_BL_SHIFT; @@ -331,7 +331,7 @@ static void bgmac_dma_rx_enable(struct bgmac *bgmac, u32 ctl; ctl = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_CTL); - if (bgmac->core->id.rev >= 4) { + if (bgmac->feature_flags & BGMAC_FEAT_RX_MASK_SETUP) { ctl &= ~BGMAC_DMA_RX_BL_MASK; ctl |= BGMAC_DMA_RX_BL_128 << BGMAC_DMA_RX_BL_SHIFT; @@ -769,14 +769,20 @@ static void bgmac_cmdcfg_maskset(struct bgmac *bgmac, u32 mask, u32 set, { u32 cmdcfg = bgmac_read(bgmac, BGMAC_CMDCFG); u32 new_val = (cmdcfg & mask) | set; + u32 cmdcfg_sr; - bgmac_set(bgmac, BGMAC_CMDCFG, BGMAC_CMDCFG_SR(bgmac->core->id.rev)); + if (bgmac->feature_flags & BGMAC_FEAT_CMDCFG_SR_REV4) + cmdcfg_sr = BGMAC_CMDCFG_SR_REV4; + else + cmdcfg_sr = BGMAC_CMDCFG_SR_REV0; + + bgmac_set(bgmac, BGMAC_CMDCFG, cmdcfg_sr); udelay(2); if (new_val != cmdcfg || force) bgmac_write(bgmac, BGMAC_CMDCFG, new_val); - bgmac_mask(bgmac, BGMAC_CMDCFG, ~BGMAC_CMDCFG_SR(bgmac->core->id.rev)); + bgmac_mask(bgmac, BGMAC_CMDCFG, ~cmdcfg_sr); udelay(2); } @@ -805,7 +811,7 @@ static void bgmac_chip_stats_update(struct bgmac *bgmac) { int i; - if (bgmac->core->id.id != BCMA_CORE_4706_MAC_GBIT) { + if (!(bgmac->feature_flags & BGMAC_FEAT_NO_CLR_MIB)) { for (i = 0; i < BGMAC_NUM_MIB_TX_REGS; i++) bgmac->mib_tx_regs[i] = bgmac_read(bgmac, @@ -824,7 +830,7 @@ static void bgmac_clear_mib(struct bgmac *bgmac) { int i; - if (bgmac->core->id.id == BCMA_CORE_4706_MAC_GBIT) + if (bgmac->feature_flags & BGMAC_FEAT_NO_CLR_MIB) return; bgmac_set(bgmac, BGMAC_DEV_CTL, BGMAC_DC_MROR); @@ -867,9 +873,8 @@ static void bgmac_mac_speed(struct bgmac *bgmac) static void bgmac_miiconfig(struct bgmac *bgmac) { struct bcma_device *core = bgmac->core; - u8 imode; - if (bgmac_is_bcm4707_family(bgmac)) { + if (bgmac->feature_flags & BGMAC_FEAT_FORCE_SPEED_2500) { bcma_awrite32(core, BCMA_IOCTL, bcma_aread32(core, BCMA_IOCTL) | 0x40 | BGMAC_BCMA_IOCTL_SW_CLKEN); @@ -877,6 +882,8 @@ static void bgmac_miiconfig(struct bgmac *bgmac) bgmac->mac_duplex = DUPLEX_FULL; bgmac_mac_speed(bgmac); } else { + u8 imode; + imode = (bgmac_read(bgmac, BGMAC_DEV_STATUS) & BGMAC_DS_MM_MASK) >> BGMAC_DS_MM_SHIFT; if (imode == 0 || imode == 1) { @@ -891,9 +898,7 @@ static void bgmac_miiconfig(struct bgmac *bgmac) static void bgmac_chip_reset(struct bgmac *bgmac) { struct bcma_device *core = bgmac->core; - struct bcma_bus *bus = core->bus; - struct bcma_chipinfo *ci = &bus->chipinfo; - u32 flags; + u32 cmdcfg_sr; u32 iost; int i; @@ -916,15 +921,12 @@ static void bgmac_chip_reset(struct bgmac *bgmac) } iost = bcma_aread32(core, BCMA_IOST); - if ((ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg == BCMA_PKG_ID_BCM47186) || - (ci->id == BCMA_CHIP_ID_BCM4749 && ci->pkg == 10) || - (ci->id == BCMA_CHIP_ID_BCM53572 && ci->pkg == BCMA_PKG_ID_BCM47188)) + if (bgmac->feature_flags & BGMAC_FEAT_IOST_ATTACHED) iost &= ~BGMAC_BCMA_IOST_ATTACHED; /* 3GMAC: for BCM4707 & BCM47094, only do core reset at bgmac_probe() */ - if (ci->id != BCMA_CHIP_ID_BCM4707 && - ci->id != BCMA_CHIP_ID_BCM47094) { - flags = 0; + if (!(bgmac->feature_flags & BGMAC_FEAT_NO_RESET)) { + u32 flags = 0; if (iost & BGMAC_BCMA_IOST_ATTACHED) { flags = BGMAC_BCMA_IOCTL_SW_CLKEN; if (!bgmac->has_robosw) @@ -934,7 +936,7 @@ static void bgmac_chip_reset(struct bgmac *bgmac) } /* Request Misc PLL for corerev > 2 */ - if (core->id.rev > 2 && !bgmac_is_bcm4707_family(bgmac)) { + if (bgmac->feature_flags & BGMAC_FEAT_MISC_PLL_REQ) { bgmac_set(bgmac, BCMA_CLKCTLST, BGMAC_BCMA_CLKCTLST_MISC_PLL_REQ); bgmac_wait_value(bgmac->core, BCMA_CLKCTLST, @@ -943,9 +945,7 @@ static void bgmac_chip_reset(struct bgmac *bgmac) 1000); } - if (ci->id == BCMA_CHIP_ID_BCM5357 || - ci->id == BCMA_CHIP_ID_BCM4749 || - ci->id == BCMA_CHIP_ID_BCM53572) { + if (bgmac->feature_flags & BGMAC_FEAT_SW_TYPE_PHY) { struct bcma_drv_cc *cc = &bgmac->core->bus->drv_cc; u8 et_swtype = 0; u8 sw_type = BGMAC_CHIPCTL_1_SW_TYPE_EPHY | @@ -959,11 +959,9 @@ static void bgmac_chip_reset(struct bgmac *bgmac) et_swtype &= 0x0f; et_swtype <<= 4; sw_type = et_swtype; - } else if (ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg == BCMA_PKG_ID_BCM5358) { + } else if (bgmac->feature_flags & BGMAC_FEAT_SW_TYPE_EPHYRMII) { sw_type = BGMAC_CHIPCTL_1_SW_TYPE_EPHYRMII; - } else if ((ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg == BCMA_PKG_ID_BCM47186) || - (ci->id == BCMA_CHIP_ID_BCM4749 && ci->pkg == 10) || - (ci->id == BCMA_CHIP_ID_BCM53572 && ci->pkg == BCMA_PKG_ID_BCM47188)) { + } else if (bgmac->feature_flags & BGMAC_FEAT_SW_TYPE_RGMII) { sw_type = BGMAC_CHIPCTL_1_IF_TYPE_RGMII | BGMAC_CHIPCTL_1_SW_TYPE_RGMII; } @@ -983,6 +981,11 @@ static void bgmac_chip_reset(struct bgmac *bgmac) * BGMAC_CMDCFG is read _after_ putting chip in a reset. So it has to * be keps until taking MAC out of the reset. */ + if (bgmac->feature_flags & BGMAC_FEAT_CMDCFG_SR_REV4) + cmdcfg_sr = BGMAC_CMDCFG_SR_REV4; + else + cmdcfg_sr = BGMAC_CMDCFG_SR_REV0; + bgmac_cmdcfg_maskset(bgmac, ~(BGMAC_CMDCFG_TE | BGMAC_CMDCFG_RE | @@ -1000,13 +1003,13 @@ static void bgmac_chip_reset(struct bgmac *bgmac) BGMAC_CMDCFG_PROM | BGMAC_CMDCFG_NLC | BGMAC_CMDCFG_CFE | - BGMAC_CMDCFG_SR(core->id.rev), + cmdcfg_sr, false); bgmac->mac_speed = SPEED_UNKNOWN; bgmac->mac_duplex = DUPLEX_UNKNOWN; bgmac_clear_mib(bgmac); - if (core->id.id == BCMA_CORE_4706_MAC_GBIT) + if (bgmac->feature_flags & BGMAC_FEAT_CMN_PHY_CTL) bcma_maskset32(bgmac->cmn, BCMA_GMAC_CMN_PHY_CTL, ~0, BCMA_GMAC_CMN_PC_MTE); else @@ -1032,46 +1035,48 @@ static void bgmac_chip_intrs_off(struct bgmac *bgmac) /* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/gmac_enable */ static void bgmac_enable(struct bgmac *bgmac) { - struct bcma_chipinfo *ci = &bgmac->core->bus->chipinfo; + u32 cmdcfg_sr; u32 cmdcfg; u32 mode; - u32 rxq_ctl; - u32 fl_ctl; - u16 bp_clk; - u8 mdp; + + if (bgmac->feature_flags & BGMAC_FEAT_CMDCFG_SR_REV4) + cmdcfg_sr = BGMAC_CMDCFG_SR_REV4; + else + cmdcfg_sr = BGMAC_CMDCFG_SR_REV0; cmdcfg = bgmac_read(bgmac, BGMAC_CMDCFG); bgmac_cmdcfg_maskset(bgmac, ~(BGMAC_CMDCFG_TE | BGMAC_CMDCFG_RE), - BGMAC_CMDCFG_SR(bgmac->core->id.rev), true); + cmdcfg_sr, true); udelay(2); cmdcfg |= BGMAC_CMDCFG_TE | BGMAC_CMDCFG_RE; bgmac_write(bgmac, BGMAC_CMDCFG, cmdcfg); mode = (bgmac_read(bgmac, BGMAC_DEV_STATUS) & BGMAC_DS_MM_MASK) >> BGMAC_DS_MM_SHIFT; - if (ci->id != BCMA_CHIP_ID_BCM47162 || mode != 0) + if (bgmac->feature_flags & BGMAC_FEAT_CLKCTLST || mode != 0) bgmac_set(bgmac, BCMA_CLKCTLST, BCMA_CLKCTLST_FORCEHT); - if (ci->id == BCMA_CHIP_ID_BCM47162 && mode == 2) + if (bgmac->feature_flags & BGMAC_FEAT_CLKCTLST && mode == 2) bcma_chipco_chipctl_maskset(&bgmac->core->bus->drv_cc, 1, ~0, BGMAC_CHIPCTL_1_RXC_DLL_BYPASS); - switch (ci->id) { - case BCMA_CHIP_ID_BCM5357: - case BCMA_CHIP_ID_BCM4749: - case BCMA_CHIP_ID_BCM53572: - case BCMA_CHIP_ID_BCM4716: - case BCMA_CHIP_ID_BCM47162: - fl_ctl = 0x03cb04cb; - if (ci->id == BCMA_CHIP_ID_BCM5357 || - ci->id == BCMA_CHIP_ID_BCM4749 || - ci->id == BCMA_CHIP_ID_BCM53572) + if (bgmac->feature_flags & (BGMAC_FEAT_FLW_CTRL1 | + BGMAC_FEAT_FLW_CTRL2)) { + u32 fl_ctl; + + if (bgmac->feature_flags & BGMAC_FEAT_FLW_CTRL1) fl_ctl = 0x2300e1; + else + fl_ctl = 0x03cb04cb; + bgmac_write(bgmac, BGMAC_FLOW_CTL_THRESH, fl_ctl); bgmac_write(bgmac, BGMAC_PAUSE_CTL, 0x27fff); - break; } - if (!bgmac_is_bcm4707_family(bgmac)) { + if (bgmac->feature_flags & BGMAC_FEAT_SET_RXQ_CLK) { + u32 rxq_ctl; + u16 bp_clk; + u8 mdp; + rxq_ctl = bgmac_read(bgmac, BGMAC_RXQ_CTL); rxq_ctl &= ~BGMAC_RXQ_CTL_MDP_MASK; bp_clk = bcma_pmu_get_bus_clock(&bgmac->core->bus->drv_cc) / @@ -1587,6 +1592,74 @@ static int bgmac_probe(struct bcma_device *core) if (core->bus->sprom.boardflags_lo & BGMAC_BFL_ENETADM) dev_warn(bgmac->dev, "Support for ADMtek ethernet switch not implemented\n"); + /* Feature Flags */ + switch (core->bus->chipinfo.id) { + case BCMA_CHIP_ID_BCM5357: + bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL1; + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_PHY; + if (core->bus->chipinfo.pkg == BCMA_PKG_ID_BCM47186) { + bgmac->feature_flags |= BGMAC_FEAT_IOST_ATTACHED; + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_RGMII; + } + if (core->bus->chipinfo.pkg == BCMA_PKG_ID_BCM5358) + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_EPHYRMII; + break; + case BCMA_CHIP_ID_BCM53572: + bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL1; + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_PHY; + if (core->bus->chipinfo.pkg == BCMA_PKG_ID_BCM47188) { + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_RGMII; + bgmac->feature_flags |= BGMAC_FEAT_IOST_ATTACHED; + } + break; + case BCMA_CHIP_ID_BCM4749: + bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL1; + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_PHY; + if (core->bus->chipinfo.pkg == 10) { + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_RGMII; + bgmac->feature_flags |= BGMAC_FEAT_IOST_ATTACHED; + } + break; + case BCMA_CHIP_ID_BCM4716: + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + /* fallthrough */ + case BCMA_CHIP_ID_BCM47162: + bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL2; + bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; + break; + /* bcm4707_family */ + case BCMA_CHIP_ID_BCM4707: + case BCMA_CHIP_ID_BCM47094: + case BCMA_CHIP_ID_BCM53018: + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_NO_RESET; + bgmac->feature_flags |= BGMAC_FEAT_FORCE_SPEED_2500; + break; + default: + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; + } + + if (!bgmac_is_bcm4707_family(bgmac) && core->id.rev > 2) + bgmac->feature_flags |= BGMAC_FEAT_MISC_PLL_REQ; + + if (core->id.id == BCMA_CORE_4706_MAC_GBIT) { + bgmac->feature_flags |= BGMAC_FEAT_CMN_PHY_CTL; + bgmac->feature_flags |= BGMAC_FEAT_NO_CLR_MIB; + } + + if (core->id.rev >= 4) { + bgmac->feature_flags |= BGMAC_FEAT_CMDCFG_SR_REV4; + bgmac->feature_flags |= BGMAC_FEAT_TX_MASK_SETUP; + bgmac->feature_flags |= BGMAC_FEAT_RX_MASK_SETUP; + } + netif_napi_add(net_dev, &bgmac->napi, bgmac_poll, BGMAC_WEIGHT); if (!bgmac_is_bcm4707_family(bgmac)) { diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h index 191e64a..89dfee1 100644 --- a/drivers/net/ethernet/broadcom/bgmac.h +++ b/drivers/net/ethernet/broadcom/bgmac.h @@ -190,7 +190,6 @@ #define BGMAC_CMDCFG_HD_SHIFT 10 #define BGMAC_CMDCFG_SR_REV0 0x00000800 /* Set to reset mode, for core rev 0-3 */ #define BGMAC_CMDCFG_SR_REV4 0x00002000 /* Set to reset mode, for core rev >= 4 */ -#define BGMAC_CMDCFG_SR(rev) ((rev >= 4) ? BGMAC_CMDCFG_SR_REV4 : BGMAC_CMDCFG_SR_REV0) #define BGMAC_CMDCFG_ML 0x00008000 /* Set to activate mac loopback mode */ #define BGMAC_CMDCFG_AE 0x00400000 #define BGMAC_CMDCFG_CFE 0x00800000 @@ -376,6 +375,24 @@ #define ETHER_MAX_LEN 1518 +/* Feature Flags */ +#define BGMAC_FEAT_TX_MASK_SETUP BIT(0) +#define BGMAC_FEAT_RX_MASK_SETUP BIT(1) +#define BGMAC_FEAT_IOST_ATTACHED BIT(2) +#define BGMAC_FEAT_NO_RESET BIT(3) +#define BGMAC_FEAT_MISC_PLL_REQ BIT(4) +#define BGMAC_FEAT_SW_TYPE_PHY BIT(5) +#define BGMAC_FEAT_SW_TYPE_EPHYRMII BIT(6) +#define BGMAC_FEAT_SW_TYPE_RGMII BIT(7) +#define BGMAC_FEAT_CMN_PHY_CTL BIT(8) +#define BGMAC_FEAT_FLW_CTRL1 BIT(9) +#define BGMAC_FEAT_FLW_CTRL2 BIT(10) +#define BGMAC_FEAT_SET_RXQ_CLK BIT(11) +#define BGMAC_FEAT_CLKCTLST BIT(12) +#define BGMAC_FEAT_NO_CLR_MIB BIT(13) +#define BGMAC_FEAT_FORCE_SPEED_2500 BIT(14) +#define BGMAC_FEAT_CMDCFG_SR_REV4 BIT(15) + struct bgmac_slot_info { union { struct sk_buff *skb; @@ -430,6 +447,8 @@ struct bgmac { struct device *dev; struct device *dma_dev; + u32 feature_flags; + struct net_device *net_dev; struct napi_struct napi; struct mii_bus *mii_bus; -- cgit v0.10.2 From f6a95a24957aec5bb488c3f978c4ed508177998f Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Thu, 7 Jul 2016 19:08:57 -0400 Subject: net: ethernet: bgmac: Add platform device support The bcma portion of the driver has been split off into a bcma specific driver. This has been mirrored for the platform driver. The last references to the bcma core struct have been changed into a generic function call. These function calls are wrappers to either the original bcma code or new platform functions that access the same areas via MMIO. This necessitated adding function pointers for both platform and bcma to hide which backend is being used from the generic bgmac code. Signed-off-by: Jon Mason Acked-by: Arnd Bergmann Reviewed-by: Florian Fainelli Tested-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig index d74a92e..bd8c80c 100644 --- a/drivers/net/ethernet/broadcom/Kconfig +++ b/drivers/net/ethernet/broadcom/Kconfig @@ -140,10 +140,18 @@ config BNX2X_SRIOV allows for virtual function acceleration in virtual environments. config BGMAC - tristate "BCMA bus GBit core support" + tristate + help + This enables the integrated ethernet controller support for many + Broadcom (mostly iProc) SoCs. An appropriate bus interface driver + needs to be enabled to select this. + +config BGMAC_BCMA + tristate "Broadcom iProc GBit BCMA support" depends on BCMA && BCMA_HOST_SOC depends on HAS_DMA depends on BCM47XX || ARCH_BCM_5301X || COMPILE_TEST + select BGMAC select PHYLIB select FIXED_PHY ---help--- @@ -152,6 +160,19 @@ config BGMAC In case of using this driver on BCM4706 it's also requires to enable BCMA_DRIVER_GMAC_CMN to make it work. +config BGMAC_PLATFORM + tristate "Broadcom iProc GBit platform support" + depends on HAS_DMA + depends on ARCH_BCM_IPROC || COMPILE_TEST + depends on OF + select BGMAC + select PHYLIB + select FIXED_PHY + default ARCH_BCM_IPROC + ---help--- + Say Y here if you want to use the Broadcom iProc Gigabit Ethernet + controller through the generic platform interface + config SYSTEMPORT tristate "Broadcom SYSTEMPORT internal MAC support" depends on OF diff --git a/drivers/net/ethernet/broadcom/Makefile b/drivers/net/ethernet/broadcom/Makefile index f559794..79f2372 100644 --- a/drivers/net/ethernet/broadcom/Makefile +++ b/drivers/net/ethernet/broadcom/Makefile @@ -10,6 +10,8 @@ obj-$(CONFIG_CNIC) += cnic.o obj-$(CONFIG_BNX2X) += bnx2x/ obj-$(CONFIG_SB1250_MAC) += sb1250-mac.o obj-$(CONFIG_TIGON3) += tg3.o -obj-$(CONFIG_BGMAC) += bgmac.o bgmac-bcma-mdio.o +obj-$(CONFIG_BGMAC) += bgmac.o +obj-$(CONFIG_BGMAC_BCMA) += bgmac-bcma.o bgmac-bcma-mdio.o +obj-$(CONFIG_BGMAC_PLATFORM) += bgmac-platform.o obj-$(CONFIG_SYSTEMPORT) += bcmsysport.o obj-$(CONFIG_BNXT) += bnxt/ diff --git a/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c b/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c index 1e65349..7c19c8e 100644 --- a/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c +++ b/drivers/net/ethernet/broadcom/bgmac-bcma-mdio.c @@ -245,6 +245,7 @@ err: kfree(bcma_mdio); return ERR_PTR(err); } +EXPORT_SYMBOL_GPL(bcma_mdio_mii_register); void bcma_mdio_mii_unregister(struct mii_bus *mii_bus) { @@ -259,6 +260,7 @@ void bcma_mdio_mii_unregister(struct mii_bus *mii_bus) mdiobus_free(mii_bus); kfree(bcma_mdio); } +EXPORT_SYMBOL_GPL(bcma_mdio_mii_unregister); MODULE_AUTHOR("Rafał Miłecki"); MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/broadcom/bgmac-bcma.c b/drivers/net/ethernet/broadcom/bgmac-bcma.c new file mode 100644 index 0000000..9a9745c4 --- /dev/null +++ b/drivers/net/ethernet/broadcom/bgmac-bcma.c @@ -0,0 +1,315 @@ +/* + * Driver for (BCM4706)? GBit MAC core on BCMA bus. + * + * Copyright (C) 2012 Rafał Miłecki + * + * Licensed under the GNU/GPL. See COPYING for details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include "bgmac.h" + +static inline bool bgmac_is_bcm4707_family(struct bcma_device *core) +{ + switch (core->bus->chipinfo.id) { + case BCMA_CHIP_ID_BCM4707: + case BCMA_CHIP_ID_BCM47094: + case BCMA_CHIP_ID_BCM53018: + return true; + default: + return false; + } +} + +/************************************************** + * BCMA bus ops + **************************************************/ + +static u32 bcma_bgmac_read(struct bgmac *bgmac, u16 offset) +{ + return bcma_read32(bgmac->bcma.core, offset); +} + +static void bcma_bgmac_write(struct bgmac *bgmac, u16 offset, u32 value) +{ + bcma_write32(bgmac->bcma.core, offset, value); +} + +static u32 bcma_bgmac_idm_read(struct bgmac *bgmac, u16 offset) +{ + return bcma_aread32(bgmac->bcma.core, offset); +} + +static void bcma_bgmac_idm_write(struct bgmac *bgmac, u16 offset, u32 value) +{ + return bcma_awrite32(bgmac->bcma.core, offset, value); +} + +static bool bcma_bgmac_clk_enabled(struct bgmac *bgmac) +{ + return bcma_core_is_enabled(bgmac->bcma.core); +} + +static void bcma_bgmac_clk_enable(struct bgmac *bgmac, u32 flags) +{ + bcma_core_enable(bgmac->bcma.core, flags); +} + +static void bcma_bgmac_cco_ctl_maskset(struct bgmac *bgmac, u32 offset, + u32 mask, u32 set) +{ + struct bcma_drv_cc *cc = &bgmac->bcma.core->bus->drv_cc; + + bcma_chipco_chipctl_maskset(cc, offset, mask, set); +} + +static u32 bcma_bgmac_get_bus_clock(struct bgmac *bgmac) +{ + struct bcma_drv_cc *cc = &bgmac->bcma.core->bus->drv_cc; + + return bcma_pmu_get_bus_clock(cc); +} + +static void bcma_bgmac_cmn_maskset32(struct bgmac *bgmac, u16 offset, u32 mask, + u32 set) +{ + bcma_maskset32(bgmac->bcma.cmn, offset, mask, set); +} + +static const struct bcma_device_id bgmac_bcma_tbl[] = { + BCMA_CORE(BCMA_MANUF_BCM, BCMA_CORE_4706_MAC_GBIT, + BCMA_ANY_REV, BCMA_ANY_CLASS), + BCMA_CORE(BCMA_MANUF_BCM, BCMA_CORE_MAC_GBIT, BCMA_ANY_REV, + BCMA_ANY_CLASS), + {}, +}; +MODULE_DEVICE_TABLE(bcma, bgmac_bcma_tbl); + +/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipattach */ +static int bgmac_probe(struct bcma_device *core) +{ + struct ssb_sprom *sprom = &core->bus->sprom; + struct mii_bus *mii_bus; + struct bgmac *bgmac; + u8 *mac; + int err; + + bgmac = kzalloc(sizeof(*bgmac), GFP_KERNEL); + if (!bgmac) + return -ENOMEM; + + bgmac->bcma.core = core; + bgmac->dev = &core->dev; + bgmac->dma_dev = core->dma_dev; + bgmac->irq = core->irq; + + bcma_set_drvdata(core, bgmac); + + switch (core->core_unit) { + case 0: + mac = sprom->et0mac; + break; + case 1: + mac = sprom->et1mac; + break; + case 2: + mac = sprom->et2mac; + break; + default: + dev_err(bgmac->dev, "Unsupported core_unit %d\n", + core->core_unit); + err = -ENOTSUPP; + goto err; + } + + ether_addr_copy(bgmac->mac_addr, mac); + + /* On BCM4706 we need common core to access PHY */ + if (core->id.id == BCMA_CORE_4706_MAC_GBIT && + !core->bus->drv_gmac_cmn.core) { + dev_err(bgmac->dev, "GMAC CMN core not found (required for BCM4706)\n"); + err = -ENODEV; + goto err; + } + bgmac->bcma.cmn = core->bus->drv_gmac_cmn.core; + + switch (core->core_unit) { + case 0: + bgmac->phyaddr = sprom->et0phyaddr; + break; + case 1: + bgmac->phyaddr = sprom->et1phyaddr; + break; + case 2: + bgmac->phyaddr = sprom->et2phyaddr; + break; + } + bgmac->phyaddr &= BGMAC_PHY_MASK; + if (bgmac->phyaddr == BGMAC_PHY_MASK) { + dev_err(bgmac->dev, "No PHY found\n"); + err = -ENODEV; + goto err; + } + dev_info(bgmac->dev, "Found PHY addr: %d%s\n", bgmac->phyaddr, + bgmac->phyaddr == BGMAC_PHY_NOREGS ? " (NOREGS)" : ""); + + if (!bgmac_is_bcm4707_family(core)) { + mii_bus = bcma_mdio_mii_register(core, bgmac->phyaddr); + if (!IS_ERR(mii_bus)) { + err = PTR_ERR(mii_bus); + goto err; + } + + bgmac->mii_bus = mii_bus; + } + + if (core->bus->hosttype == BCMA_HOSTTYPE_PCI) { + dev_err(bgmac->dev, "PCI setup not implemented\n"); + err = -ENOTSUPP; + goto err1; + } + + bgmac->has_robosw = !!(core->bus->sprom.boardflags_lo & + BGMAC_BFL_ENETROBO); + if (bgmac->has_robosw) + dev_warn(bgmac->dev, "Support for Roboswitch not implemented\n"); + + if (core->bus->sprom.boardflags_lo & BGMAC_BFL_ENETADM) + dev_warn(bgmac->dev, "Support for ADMtek ethernet switch not implemented\n"); + + /* Feature Flags */ + switch (core->bus->chipinfo.id) { + case BCMA_CHIP_ID_BCM5357: + bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL1; + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_PHY; + if (core->bus->chipinfo.pkg == BCMA_PKG_ID_BCM47186) { + bgmac->feature_flags |= BGMAC_FEAT_IOST_ATTACHED; + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_RGMII; + } + if (core->bus->chipinfo.pkg == BCMA_PKG_ID_BCM5358) + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_EPHYRMII; + break; + case BCMA_CHIP_ID_BCM53572: + bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL1; + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_PHY; + if (core->bus->chipinfo.pkg == BCMA_PKG_ID_BCM47188) { + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_RGMII; + bgmac->feature_flags |= BGMAC_FEAT_IOST_ATTACHED; + } + break; + case BCMA_CHIP_ID_BCM4749: + bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL1; + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_PHY; + if (core->bus->chipinfo.pkg == 10) { + bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_RGMII; + bgmac->feature_flags |= BGMAC_FEAT_IOST_ATTACHED; + } + break; + case BCMA_CHIP_ID_BCM4716: + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + /* fallthrough */ + case BCMA_CHIP_ID_BCM47162: + bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL2; + bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; + break; + /* bcm4707_family */ + case BCMA_CHIP_ID_BCM4707: + case BCMA_CHIP_ID_BCM47094: + case BCMA_CHIP_ID_BCM53018: + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_NO_RESET; + bgmac->feature_flags |= BGMAC_FEAT_FORCE_SPEED_2500; + break; + default: + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; + } + + if (!bgmac_is_bcm4707_family(core) && core->id.rev > 2) + bgmac->feature_flags |= BGMAC_FEAT_MISC_PLL_REQ; + + if (core->id.id == BCMA_CORE_4706_MAC_GBIT) { + bgmac->feature_flags |= BGMAC_FEAT_CMN_PHY_CTL; + bgmac->feature_flags |= BGMAC_FEAT_NO_CLR_MIB; + } + + if (core->id.rev >= 4) { + bgmac->feature_flags |= BGMAC_FEAT_CMDCFG_SR_REV4; + bgmac->feature_flags |= BGMAC_FEAT_TX_MASK_SETUP; + bgmac->feature_flags |= BGMAC_FEAT_RX_MASK_SETUP; + } + + bgmac->read = bcma_bgmac_read; + bgmac->write = bcma_bgmac_write; + bgmac->idm_read = bcma_bgmac_idm_read; + bgmac->idm_write = bcma_bgmac_idm_write; + bgmac->clk_enabled = bcma_bgmac_clk_enabled; + bgmac->clk_enable = bcma_bgmac_clk_enable; + bgmac->cco_ctl_maskset = bcma_bgmac_cco_ctl_maskset; + bgmac->get_bus_clock = bcma_bgmac_get_bus_clock; + bgmac->cmn_maskset32 = bcma_bgmac_cmn_maskset32; + + err = bgmac_enet_probe(bgmac); + if (err) + goto err1; + + return 0; + +err1: + bcma_mdio_mii_unregister(bgmac->mii_bus); +err: + kfree(bgmac); + bcma_set_drvdata(core, NULL); + + return err; +} + +static void bgmac_remove(struct bcma_device *core) +{ + struct bgmac *bgmac = bcma_get_drvdata(core); + + bcma_mdio_mii_unregister(bgmac->mii_bus); + bgmac_enet_remove(bgmac); + bcma_set_drvdata(core, NULL); + kfree(bgmac); +} + +static struct bcma_driver bgmac_bcma_driver = { + .name = KBUILD_MODNAME, + .id_table = bgmac_bcma_tbl, + .probe = bgmac_probe, + .remove = bgmac_remove, +}; + +static int __init bgmac_init(void) +{ + int err; + + err = bcma_driver_register(&bgmac_bcma_driver); + if (err) + return err; + pr_info("Broadcom 47xx GBit MAC driver loaded\n"); + + return 0; +} + +static void __exit bgmac_exit(void) +{ + bcma_driver_unregister(&bgmac_bcma_driver); +} + +module_init(bgmac_init) +module_exit(bgmac_exit) + +MODULE_AUTHOR("Rafał Miłecki"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/broadcom/bgmac-platform.c b/drivers/net/ethernet/broadcom/bgmac-platform.c new file mode 100644 index 0000000..7a8f7ef --- /dev/null +++ b/drivers/net/ethernet/broadcom/bgmac-platform.c @@ -0,0 +1,189 @@ +/* + * Copyright (C) 2016 Broadcom + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation version 2. + * + * This program is distributed "as is" WITHOUT ANY WARRANTY of any + * kind, whether express or implied; without even the implied warranty + * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include "bgmac.h" + +static u32 platform_bgmac_read(struct bgmac *bgmac, u16 offset) +{ + return readl(bgmac->plat.base + offset); +} + +static void platform_bgmac_write(struct bgmac *bgmac, u16 offset, u32 value) +{ + writel(value, bgmac->plat.base + offset); +} + +static u32 platform_bgmac_idm_read(struct bgmac *bgmac, u16 offset) +{ + return readl(bgmac->plat.idm_base + offset); +} + +static void platform_bgmac_idm_write(struct bgmac *bgmac, u16 offset, u32 value) +{ + return writel(value, bgmac->plat.idm_base + offset); +} + +static bool platform_bgmac_clk_enabled(struct bgmac *bgmac) +{ + if ((bgmac_idm_read(bgmac, BCMA_IOCTL) & + (BCMA_IOCTL_CLK | BCMA_IOCTL_FGC)) != BCMA_IOCTL_CLK) + return false; + if (bgmac_idm_read(bgmac, BCMA_RESET_CTL) & BCMA_RESET_CTL_RESET) + return false; + return true; +} + +static void platform_bgmac_clk_enable(struct bgmac *bgmac, u32 flags) +{ + bgmac_idm_write(bgmac, BCMA_IOCTL, + (BCMA_IOCTL_CLK | BCMA_IOCTL_FGC | flags)); + bgmac_idm_read(bgmac, BCMA_IOCTL); + + bgmac_idm_write(bgmac, BCMA_RESET_CTL, 0); + bgmac_idm_read(bgmac, BCMA_RESET_CTL); + udelay(1); + + bgmac_idm_write(bgmac, BCMA_IOCTL, (BCMA_IOCTL_CLK | flags)); + bgmac_idm_read(bgmac, BCMA_IOCTL); + udelay(1); +} + +static void platform_bgmac_cco_ctl_maskset(struct bgmac *bgmac, u32 offset, + u32 mask, u32 set) +{ + /* This shouldn't be encountered */ + WARN_ON(1); +} + +static u32 platform_bgmac_get_bus_clock(struct bgmac *bgmac) +{ + /* This shouldn't be encountered */ + WARN_ON(1); + + return 0; +} + +static void platform_bgmac_cmn_maskset32(struct bgmac *bgmac, u16 offset, + u32 mask, u32 set) +{ + /* This shouldn't be encountered */ + WARN_ON(1); +} + +static int bgmac_probe(struct platform_device *pdev) +{ + struct device_node *np = pdev->dev.of_node; + struct bgmac *bgmac; + struct resource *regs; + const u8 *mac_addr; + + bgmac = devm_kzalloc(&pdev->dev, sizeof(*bgmac), GFP_KERNEL); + if (!bgmac) + return -ENOMEM; + + platform_set_drvdata(pdev, bgmac); + + /* Set the features of the 4707 family */ + bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; + bgmac->feature_flags |= BGMAC_FEAT_NO_RESET; + bgmac->feature_flags |= BGMAC_FEAT_FORCE_SPEED_2500; + bgmac->feature_flags |= BGMAC_FEAT_CMDCFG_SR_REV4; + bgmac->feature_flags |= BGMAC_FEAT_TX_MASK_SETUP; + bgmac->feature_flags |= BGMAC_FEAT_RX_MASK_SETUP; + + bgmac->dev = &pdev->dev; + bgmac->dma_dev = &pdev->dev; + + mac_addr = of_get_mac_address(np); + if (mac_addr) + ether_addr_copy(bgmac->mac_addr, mac_addr); + else + dev_warn(&pdev->dev, "MAC address not present in device tree\n"); + + bgmac->irq = platform_get_irq(pdev, 0); + if (bgmac->irq < 0) { + dev_err(&pdev->dev, "Unable to obtain IRQ\n"); + return bgmac->irq; + } + + regs = platform_get_resource_byname(pdev, IORESOURCE_MEM, "amac_base"); + if (!regs) { + dev_err(&pdev->dev, "Unable to obtain base resource\n"); + return -EINVAL; + } + + bgmac->plat.base = devm_ioremap_resource(&pdev->dev, regs); + if (IS_ERR(bgmac->plat.base)) { + dev_err(&pdev->dev, "Unable to map base resource\n"); + return PTR_ERR(bgmac->plat.base); + } + + regs = platform_get_resource_byname(pdev, IORESOURCE_MEM, "idm_base"); + if (!regs) { + dev_err(&pdev->dev, "Unable to obtain idm resource\n"); + return -EINVAL; + } + + bgmac->plat.idm_base = devm_ioremap_resource(&pdev->dev, regs); + if (!bgmac->plat.idm_base) { + dev_err(&pdev->dev, "Unable to map idm resource\n"); + return PTR_ERR(bgmac->plat.idm_base); + } + + bgmac->read = platform_bgmac_read; + bgmac->write = platform_bgmac_write; + bgmac->idm_read = platform_bgmac_idm_read; + bgmac->idm_write = platform_bgmac_idm_write; + bgmac->clk_enabled = platform_bgmac_clk_enabled; + bgmac->clk_enable = platform_bgmac_clk_enable; + bgmac->cco_ctl_maskset = platform_bgmac_cco_ctl_maskset; + bgmac->get_bus_clock = platform_bgmac_get_bus_clock; + bgmac->cmn_maskset32 = platform_bgmac_cmn_maskset32; + + return bgmac_enet_probe(bgmac); +} + +static int bgmac_remove(struct platform_device *pdev) +{ + struct bgmac *bgmac = platform_get_drvdata(pdev); + + bgmac_enet_remove(bgmac); + + return 0; +} + +static const struct of_device_id bgmac_of_enet_match[] = { + {.compatible = "brcm,amac",}, + {.compatible = "brcm,nsp-amac",}, + {}, +}; + +MODULE_DEVICE_TABLE(of, bgmac_of_enet_match); + +static struct platform_driver bgmac_enet_driver = { + .driver = { + .name = "bgmac-enet", + .of_match_table = bgmac_of_enet_match, + }, + .probe = bgmac_probe, + .remove = bgmac_remove, +}; + +module_platform_driver(bgmac_enet_driver); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c index 770796a..13b0725 100644 --- a/drivers/net/ethernet/broadcom/bgmac.c +++ b/drivers/net/ethernet/broadcom/bgmac.c @@ -6,51 +6,27 @@ * Licensed under the GNU/GPL. See COPYING for details. */ -#include "bgmac.h" -#include -#include -#include +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include #include -#include -#include -#include -#include -#include #include +#include "bgmac.h" -static const struct bcma_device_id bgmac_bcma_tbl[] = { - BCMA_CORE(BCMA_MANUF_BCM, BCMA_CORE_4706_MAC_GBIT, BCMA_ANY_REV, BCMA_ANY_CLASS), - BCMA_CORE(BCMA_MANUF_BCM, BCMA_CORE_MAC_GBIT, BCMA_ANY_REV, BCMA_ANY_CLASS), - {}, -}; -MODULE_DEVICE_TABLE(bcma, bgmac_bcma_tbl); - -static inline bool bgmac_is_bcm4707_family(struct bgmac *bgmac) -{ - switch (bgmac->core->bus->chipinfo.id) { - case BCMA_CHIP_ID_BCM4707: - case BCMA_CHIP_ID_BCM47094: - case BCMA_CHIP_ID_BCM53018: - return true; - default: - return false; - } -} - -static bool bgmac_wait_value(struct bcma_device *core, u16 reg, u32 mask, +static bool bgmac_wait_value(struct bgmac *bgmac, u16 reg, u32 mask, u32 value, int timeout) { u32 val; int i; for (i = 0; i < timeout / 10; i++) { - val = bcma_read32(core, reg); + val = bgmac_read(bgmac, reg); if ((val & mask) == value) return true; udelay(10); } - dev_err(&core->dev, "Timeout waiting for reg 0x%X\n", reg); + dev_err(bgmac->dev, "Timeout waiting for reg 0x%X\n", reg); return false; } @@ -89,7 +65,7 @@ static void bgmac_dma_tx_reset(struct bgmac *bgmac, struct bgmac_dma_ring *ring) /* Remove SUSPEND bit */ bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL, 0); - if (!bgmac_wait_value(bgmac->core, + if (!bgmac_wait_value(bgmac, ring->mmio_base + BGMAC_DMA_TX_STATUS, BGMAC_DMA_TX_STAT, BGMAC_DMA_TX_STAT_DISABLED, 10000)) { @@ -317,7 +293,7 @@ static void bgmac_dma_rx_reset(struct bgmac *bgmac, struct bgmac_dma_ring *ring) return; bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_CTL, 0); - if (!bgmac_wait_value(bgmac->core, + if (!bgmac_wait_value(bgmac, ring->mmio_base + BGMAC_DMA_RX_STATUS, BGMAC_DMA_RX_STAT, BGMAC_DMA_RX_STAT_DISABLED, 10000)) @@ -640,7 +616,7 @@ static int bgmac_dma_alloc(struct bgmac *bgmac) BUILD_BUG_ON(BGMAC_MAX_TX_RINGS > ARRAY_SIZE(ring_base)); BUILD_BUG_ON(BGMAC_MAX_RX_RINGS > ARRAY_SIZE(ring_base)); - if (!(bcma_aread32(bgmac->core, BCMA_IOST) & BCMA_IOST_DMA64)) { + if (!(bgmac_idm_read(bgmac, BCMA_IOST) & BCMA_IOST_DMA64)) { dev_err(bgmac->dev, "Core does not report 64-bit DMA\n"); return -ENOTSUPP; } @@ -872,12 +848,10 @@ static void bgmac_mac_speed(struct bgmac *bgmac) static void bgmac_miiconfig(struct bgmac *bgmac) { - struct bcma_device *core = bgmac->core; - if (bgmac->feature_flags & BGMAC_FEAT_FORCE_SPEED_2500) { - bcma_awrite32(core, BCMA_IOCTL, - bcma_aread32(core, BCMA_IOCTL) | 0x40 | - BGMAC_BCMA_IOCTL_SW_CLKEN); + bgmac_idm_write(bgmac, BCMA_IOCTL, + bgmac_idm_read(bgmac, BCMA_IOCTL) | 0x40 | + BGMAC_BCMA_IOCTL_SW_CLKEN); bgmac->mac_speed = SPEED_2500; bgmac->mac_duplex = DUPLEX_FULL; bgmac_mac_speed(bgmac); @@ -897,12 +871,11 @@ static void bgmac_miiconfig(struct bgmac *bgmac) /* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipreset */ static void bgmac_chip_reset(struct bgmac *bgmac) { - struct bcma_device *core = bgmac->core; u32 cmdcfg_sr; u32 iost; int i; - if (bcma_core_is_enabled(core)) { + if (bgmac_clk_enabled(bgmac)) { if (!bgmac->stats_grabbed) { /* bgmac_chip_stats_update(bgmac); */ bgmac->stats_grabbed = true; @@ -920,7 +893,7 @@ static void bgmac_chip_reset(struct bgmac *bgmac) /* TODO: Clear software multicast filter list */ } - iost = bcma_aread32(core, BCMA_IOST); + iost = bgmac_idm_read(bgmac, BCMA_IOST); if (bgmac->feature_flags & BGMAC_FEAT_IOST_ATTACHED) iost &= ~BGMAC_BCMA_IOST_ATTACHED; @@ -932,21 +905,20 @@ static void bgmac_chip_reset(struct bgmac *bgmac) if (!bgmac->has_robosw) flags |= BGMAC_BCMA_IOCTL_SW_RESET; } - bcma_core_enable(core, flags); + bgmac_clk_enable(bgmac, flags); } /* Request Misc PLL for corerev > 2 */ if (bgmac->feature_flags & BGMAC_FEAT_MISC_PLL_REQ) { bgmac_set(bgmac, BCMA_CLKCTLST, BGMAC_BCMA_CLKCTLST_MISC_PLL_REQ); - bgmac_wait_value(bgmac->core, BCMA_CLKCTLST, + bgmac_wait_value(bgmac, BCMA_CLKCTLST, BGMAC_BCMA_CLKCTLST_MISC_PLL_ST, BGMAC_BCMA_CLKCTLST_MISC_PLL_ST, 1000); } if (bgmac->feature_flags & BGMAC_FEAT_SW_TYPE_PHY) { - struct bcma_drv_cc *cc = &bgmac->core->bus->drv_cc; u8 et_swtype = 0; u8 sw_type = BGMAC_CHIPCTL_1_SW_TYPE_EPHY | BGMAC_CHIPCTL_1_IF_TYPE_MII; @@ -965,16 +937,15 @@ static void bgmac_chip_reset(struct bgmac *bgmac) sw_type = BGMAC_CHIPCTL_1_IF_TYPE_RGMII | BGMAC_CHIPCTL_1_SW_TYPE_RGMII; } - bcma_chipco_chipctl_maskset(cc, 1, - ~(BGMAC_CHIPCTL_1_IF_TYPE_MASK | - BGMAC_CHIPCTL_1_SW_TYPE_MASK), - sw_type); + bgmac_cco_ctl_maskset(bgmac, 1, ~(BGMAC_CHIPCTL_1_IF_TYPE_MASK | + BGMAC_CHIPCTL_1_SW_TYPE_MASK), + sw_type); } if (iost & BGMAC_BCMA_IOST_ATTACHED && !bgmac->has_robosw) - bcma_awrite32(core, BCMA_IOCTL, - bcma_aread32(core, BCMA_IOCTL) & - ~BGMAC_BCMA_IOCTL_SW_RESET); + bgmac_idm_write(bgmac, BCMA_IOCTL, + bgmac_idm_read(bgmac, BCMA_IOCTL) & + ~BGMAC_BCMA_IOCTL_SW_RESET); /* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/gmac_reset * Specs don't say about using BGMAC_CMDCFG_SR, but in this routine @@ -1010,8 +981,8 @@ static void bgmac_chip_reset(struct bgmac *bgmac) bgmac_clear_mib(bgmac); if (bgmac->feature_flags & BGMAC_FEAT_CMN_PHY_CTL) - bcma_maskset32(bgmac->cmn, BCMA_GMAC_CMN_PHY_CTL, ~0, - BCMA_GMAC_CMN_PC_MTE); + bgmac_cmn_maskset32(bgmac, BCMA_GMAC_CMN_PHY_CTL, ~0, + BCMA_GMAC_CMN_PC_MTE); else bgmac_set(bgmac, BGMAC_PHY_CNTL, BGMAC_PC_MTE); bgmac_miiconfig(bgmac); @@ -1056,8 +1027,8 @@ static void bgmac_enable(struct bgmac *bgmac) if (bgmac->feature_flags & BGMAC_FEAT_CLKCTLST || mode != 0) bgmac_set(bgmac, BCMA_CLKCTLST, BCMA_CLKCTLST_FORCEHT); if (bgmac->feature_flags & BGMAC_FEAT_CLKCTLST && mode == 2) - bcma_chipco_chipctl_maskset(&bgmac->core->bus->drv_cc, 1, ~0, - BGMAC_CHIPCTL_1_RXC_DLL_BYPASS); + bgmac_cco_ctl_maskset(bgmac, 1, ~0, + BGMAC_CHIPCTL_1_RXC_DLL_BYPASS); if (bgmac->feature_flags & (BGMAC_FEAT_FLW_CTRL1 | BGMAC_FEAT_FLW_CTRL2)) { @@ -1079,8 +1050,7 @@ static void bgmac_enable(struct bgmac *bgmac) rxq_ctl = bgmac_read(bgmac, BGMAC_RXQ_CTL); rxq_ctl &= ~BGMAC_RXQ_CTL_MDP_MASK; - bp_clk = bcma_pmu_get_bus_clock(&bgmac->core->bus->drv_cc) / - 1000000; + bp_clk = bgmac_get_bus_clock(bgmac) / 1000000; mdp = (bp_clk * 128 / 1000) - 3; rxq_ctl |= (mdp << BGMAC_RXQ_CTL_MDP_SHIFT); bgmac_write(bgmac, BGMAC_RXQ_CTL, rxq_ctl); @@ -1175,7 +1145,7 @@ static int bgmac_open(struct net_device *net_dev) /* Specs say about reclaiming rings here, but we do that in DMA init */ bgmac_chip_init(bgmac); - err = request_irq(bgmac->core->irq, bgmac_interrupt, IRQF_SHARED, + err = request_irq(bgmac->irq, bgmac_interrupt, IRQF_SHARED, KBUILD_MODNAME, net_dev); if (err < 0) { dev_err(bgmac->dev, "IRQ request error: %d!\n", err); @@ -1201,7 +1171,7 @@ static int bgmac_stop(struct net_device *net_dev) napi_disable(&bgmac->napi); bgmac_chip_intrs_off(bgmac); - free_irq(bgmac->core->irq, net_dev); + free_irq(bgmac->irq, net_dev); bgmac_chip_reset(bgmac); bgmac_dma_cleanup(bgmac); @@ -1380,7 +1350,7 @@ static void bgmac_get_drvinfo(struct net_device *net_dev, struct ethtool_drvinfo *info) { strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver)); - strlcpy(info->bus_info, "BCMA", sizeof(info->bus_info)); + strlcpy(info->bus_info, "AXI", sizeof(info->bus_info)); } static const struct ethtool_ops bgmac_ethtool_ops = { @@ -1464,116 +1434,41 @@ static int bgmac_phy_connect(struct bgmac *bgmac) return 0; } -static int bgmac_probe(struct bcma_device *core) +int bgmac_enet_probe(struct bgmac *info) { struct net_device *net_dev; struct bgmac *bgmac; - struct ssb_sprom *sprom = &core->bus->sprom; - u8 *mac; int err; - switch (core->core_unit) { - case 0: - mac = sprom->et0mac; - break; - case 1: - mac = sprom->et1mac; - break; - case 2: - mac = sprom->et2mac; - break; - default: - dev_err(&core->dev, "Unsupported core_unit %d\n", - core->core_unit); - return -ENOTSUPP; - } - - if (!is_valid_ether_addr(mac)) { - dev_err(&core->dev, "Invalid MAC addr: %pM\n", mac); - eth_random_addr(mac); - dev_warn(&core->dev, "Using random MAC: %pM\n", mac); - } - - /* This (reset &) enable is not preset in specs or reference driver but - * Broadcom does it in arch PCI code when enabling fake PCI device. - */ - bcma_core_enable(core, 0); - /* Allocation and references */ net_dev = alloc_etherdev(sizeof(*bgmac)); if (!net_dev) return -ENOMEM; + net_dev->netdev_ops = &bgmac_netdev_ops; - net_dev->irq = core->irq; net_dev->ethtool_ops = &bgmac_ethtool_ops; bgmac = netdev_priv(net_dev); - bgmac->dev = &core->dev; - bgmac->dma_dev = core->dma_dev; + memcpy(bgmac, info, sizeof(*bgmac)); bgmac->net_dev = net_dev; - bgmac->core = core; - bcma_set_drvdata(core, bgmac); - SET_NETDEV_DEV(net_dev, &core->dev); - - /* Defaults */ - memcpy(bgmac->net_dev->dev_addr, mac, ETH_ALEN); - - /* On BCM4706 we need common core to access PHY */ - if (core->id.id == BCMA_CORE_4706_MAC_GBIT && - !core->bus->drv_gmac_cmn.core) { - dev_err(bgmac->dev, "GMAC CMN core not found (required for BCM4706)\n"); - err = -ENODEV; - goto err_netdev_free; - } - bgmac->cmn = core->bus->drv_gmac_cmn.core; - - switch (core->core_unit) { - case 0: - bgmac->phyaddr = sprom->et0phyaddr; - break; - case 1: - bgmac->phyaddr = sprom->et1phyaddr; - break; - case 2: - bgmac->phyaddr = sprom->et2phyaddr; - break; - } - bgmac->phyaddr &= BGMAC_PHY_MASK; - if (bgmac->phyaddr == BGMAC_PHY_MASK) { - dev_err(bgmac->dev, "No PHY found\n"); - err = -ENODEV; - goto err_netdev_free; + net_dev->irq = bgmac->irq; + SET_NETDEV_DEV(net_dev, bgmac->dev); + + if (!is_valid_ether_addr(bgmac->mac_addr)) { + dev_err(bgmac->dev, "Invalid MAC addr: %pM\n", + bgmac->mac_addr); + eth_random_addr(bgmac->mac_addr); + dev_warn(bgmac->dev, "Using random MAC: %pM\n", + bgmac->mac_addr); } - dev_info(bgmac->dev, "Found PHY addr: %d%s\n", bgmac->phyaddr, - bgmac->phyaddr == BGMAC_PHY_NOREGS ? " (NOREGS)" : ""); + ether_addr_copy(net_dev->dev_addr, bgmac->mac_addr); - if (core->bus->hosttype == BCMA_HOSTTYPE_PCI) { - dev_err(bgmac->dev, "PCI setup not implemented\n"); - err = -ENOTSUPP; - goto err_netdev_free; - } + /* This (reset &) enable is not preset in specs or reference driver but + * Broadcom does it in arch PCI code when enabling fake PCI device. + */ + bgmac_clk_enable(bgmac, 0); bgmac_chip_reset(bgmac); - /* For Northstar, we have to take all GMAC core out of reset */ - if (bgmac_is_bcm4707_family(bgmac)) { - struct bcma_device *ns_core; - int ns_gmac; - - /* Northstar has 4 GMAC cores */ - for (ns_gmac = 0; ns_gmac < 4; ns_gmac++) { - /* As Northstar requirement, we have to reset all GMACs - * before accessing one. bgmac_chip_reset() call - * bcma_core_enable() for this core. Then the other - * three GMACs didn't reset. We do it here. - */ - ns_core = bcma_find_core_unit(core->bus, - BCMA_CORE_MAC_GBIT, - ns_gmac); - if (ns_core && !bcma_core_is_enabled(ns_core)) - bcma_core_enable(ns_core, 0); - } - } - err = bgmac_dma_alloc(bgmac); if (err) { dev_err(bgmac->dev, "Unable to alloc memory for DMA\n"); @@ -1584,103 +1479,15 @@ static int bgmac_probe(struct bcma_device *core) if (bcm47xx_nvram_getenv("et0_no_txint", NULL, 0) == 0) bgmac->int_mask &= ~BGMAC_IS_TX_MASK; - bgmac->has_robosw = !!(core->bus->sprom.boardflags_lo & - BGMAC_BFL_ENETROBO); - if (bgmac->has_robosw) - dev_warn(bgmac->dev, "Support for Roboswitch not implemented\n"); - - if (core->bus->sprom.boardflags_lo & BGMAC_BFL_ENETADM) - dev_warn(bgmac->dev, "Support for ADMtek ethernet switch not implemented\n"); - - /* Feature Flags */ - switch (core->bus->chipinfo.id) { - case BCMA_CHIP_ID_BCM5357: - bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; - bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; - bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL1; - bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_PHY; - if (core->bus->chipinfo.pkg == BCMA_PKG_ID_BCM47186) { - bgmac->feature_flags |= BGMAC_FEAT_IOST_ATTACHED; - bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_RGMII; - } - if (core->bus->chipinfo.pkg == BCMA_PKG_ID_BCM5358) - bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_EPHYRMII; - break; - case BCMA_CHIP_ID_BCM53572: - bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; - bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; - bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL1; - bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_PHY; - if (core->bus->chipinfo.pkg == BCMA_PKG_ID_BCM47188) { - bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_RGMII; - bgmac->feature_flags |= BGMAC_FEAT_IOST_ATTACHED; - } - break; - case BCMA_CHIP_ID_BCM4749: - bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; - bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; - bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL1; - bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_PHY; - if (core->bus->chipinfo.pkg == 10) { - bgmac->feature_flags |= BGMAC_FEAT_SW_TYPE_RGMII; - bgmac->feature_flags |= BGMAC_FEAT_IOST_ATTACHED; - } - break; - case BCMA_CHIP_ID_BCM4716: - bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; - /* fallthrough */ - case BCMA_CHIP_ID_BCM47162: - bgmac->feature_flags |= BGMAC_FEAT_FLW_CTRL2; - bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; - break; - /* bcm4707_family */ - case BCMA_CHIP_ID_BCM4707: - case BCMA_CHIP_ID_BCM47094: - case BCMA_CHIP_ID_BCM53018: - bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; - bgmac->feature_flags |= BGMAC_FEAT_NO_RESET; - bgmac->feature_flags |= BGMAC_FEAT_FORCE_SPEED_2500; - break; - default: - bgmac->feature_flags |= BGMAC_FEAT_CLKCTLST; - bgmac->feature_flags |= BGMAC_FEAT_SET_RXQ_CLK; - } - - if (!bgmac_is_bcm4707_family(bgmac) && core->id.rev > 2) - bgmac->feature_flags |= BGMAC_FEAT_MISC_PLL_REQ; - - if (core->id.id == BCMA_CORE_4706_MAC_GBIT) { - bgmac->feature_flags |= BGMAC_FEAT_CMN_PHY_CTL; - bgmac->feature_flags |= BGMAC_FEAT_NO_CLR_MIB; - } - - if (core->id.rev >= 4) { - bgmac->feature_flags |= BGMAC_FEAT_CMDCFG_SR_REV4; - bgmac->feature_flags |= BGMAC_FEAT_TX_MASK_SETUP; - bgmac->feature_flags |= BGMAC_FEAT_RX_MASK_SETUP; - } - netif_napi_add(net_dev, &bgmac->napi, bgmac_poll, BGMAC_WEIGHT); - if (!bgmac_is_bcm4707_family(bgmac)) { - struct mii_bus *mii_bus; - - mii_bus = bcma_mdio_mii_register(core, bgmac->phyaddr); - if (!IS_ERR(mii_bus)) { - err = PTR_ERR(mii_bus); - goto err_dma_free; - } - - bgmac->mii_bus = mii_bus; - } - if (!bgmac->mii_bus) err = bgmac_phy_connect_direct(bgmac); else err = bgmac_phy_connect(bgmac); if (err) { dev_err(bgmac->dev, "Cannot connect to phy\n"); - goto err_mii_unregister; + goto err_dma_free; } net_dev->features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM; @@ -1699,56 +1506,24 @@ static int bgmac_probe(struct bcma_device *core) err_phy_disconnect: phy_disconnect(net_dev->phydev); -err_mii_unregister: - bcma_mdio_mii_unregister(bgmac->mii_bus); err_dma_free: bgmac_dma_free(bgmac); err_netdev_free: - bcma_set_drvdata(core, NULL); free_netdev(net_dev); return err; } +EXPORT_SYMBOL_GPL(bgmac_enet_probe); -static void bgmac_remove(struct bcma_device *core) +void bgmac_enet_remove(struct bgmac *bgmac) { - struct bgmac *bgmac = bcma_get_drvdata(core); - unregister_netdev(bgmac->net_dev); phy_disconnect(bgmac->net_dev->phydev); - bcma_mdio_mii_unregister(bgmac->mii_bus); netif_napi_del(&bgmac->napi); bgmac_dma_free(bgmac); - bcma_set_drvdata(core, NULL); free_netdev(bgmac->net_dev); } - -static struct bcma_driver bgmac_bcma_driver = { - .name = KBUILD_MODNAME, - .id_table = bgmac_bcma_tbl, - .probe = bgmac_probe, - .remove = bgmac_remove, -}; - -static int __init bgmac_init(void) -{ - int err; - - err = bcma_driver_register(&bgmac_bcma_driver); - if (err) - return err; - pr_info("Broadcom 47xx GBit MAC driver loaded\n"); - - return 0; -} - -static void __exit bgmac_exit(void) -{ - bcma_driver_unregister(&bgmac_bcma_driver); -} - -module_init(bgmac_init) -module_exit(bgmac_exit) +EXPORT_SYMBOL_GPL(bgmac_enet_remove); MODULE_AUTHOR("Rafał Miłecki"); MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/broadcom/bgmac.h b/drivers/net/ethernet/broadcom/bgmac.h index 89dfee1..24a2502 100644 --- a/drivers/net/ethernet/broadcom/bgmac.h +++ b/drivers/net/ethernet/broadcom/bgmac.h @@ -1,8 +1,6 @@ #ifndef _BGMAC_H #define _BGMAC_H -#include -#include #include #define BGMAC_DEV_CTL 0x000 @@ -442,11 +440,21 @@ struct bgmac_rx_header { }; struct bgmac { - struct bcma_device *core; - struct bcma_device *cmn; /* Reference to CMN core for BCM4706 */ + union { + struct { + void *base; + void *idm_base; + } plat; + struct { + struct bcma_device *core; + /* Reference to CMN core for BCM4706 */ + struct bcma_device *cmn; + } bcma; + }; struct device *dev; struct device *dma_dev; + unsigned char mac_addr[ETH_ALEN]; u32 feature_flags; struct net_device *net_dev; @@ -463,6 +471,7 @@ struct bgmac { u32 mib_rx_regs[BGMAC_NUM_MIB_RX_REGS]; /* Int */ + int irq; u32 int_mask; /* Current MAC state */ @@ -473,19 +482,71 @@ struct bgmac { bool has_robosw; bool loopback; + + u32 (*read)(struct bgmac *bgmac, u16 offset); + void (*write)(struct bgmac *bgmac, u16 offset, u32 value); + u32 (*idm_read)(struct bgmac *bgmac, u16 offset); + void (*idm_write)(struct bgmac *bgmac, u16 offset, u32 value); + bool (*clk_enabled)(struct bgmac *bgmac); + void (*clk_enable)(struct bgmac *bgmac, u32 flags); + void (*cco_ctl_maskset)(struct bgmac *bgmac, u32 offset, u32 mask, + u32 set); + u32 (*get_bus_clock)(struct bgmac *bgmac); + void (*cmn_maskset32)(struct bgmac *bgmac, u16 offset, u32 mask, + u32 set); }; +int bgmac_enet_probe(struct bgmac *info); +void bgmac_enet_remove(struct bgmac *bgmac); + struct mii_bus *bcma_mdio_mii_register(struct bcma_device *core, u8 phyaddr); void bcma_mdio_mii_unregister(struct mii_bus *mii_bus); static inline u32 bgmac_read(struct bgmac *bgmac, u16 offset) { - return bcma_read32(bgmac->core, offset); + return bgmac->read(bgmac, offset); } static inline void bgmac_write(struct bgmac *bgmac, u16 offset, u32 value) { - bcma_write32(bgmac->core, offset, value); + bgmac->write(bgmac, offset, value); +} + +static inline u32 bgmac_idm_read(struct bgmac *bgmac, u16 offset) +{ + return bgmac->idm_read(bgmac, offset); +} + +static inline void bgmac_idm_write(struct bgmac *bgmac, u16 offset, u32 value) +{ + bgmac->idm_write(bgmac, offset, value); +} + +static inline bool bgmac_clk_enabled(struct bgmac *bgmac) +{ + return bgmac->clk_enabled(bgmac); +} + +static inline void bgmac_clk_enable(struct bgmac *bgmac, u32 flags) +{ + bgmac->clk_enable(bgmac, flags); +} + +static inline void bgmac_cco_ctl_maskset(struct bgmac *bgmac, u32 offset, + u32 mask, u32 set) +{ + bgmac->cco_ctl_maskset(bgmac, offset, mask, set); +} + +static inline u32 bgmac_get_bus_clock(struct bgmac *bgmac) +{ + return bgmac->get_bus_clock(bgmac); +} + +static inline void bgmac_cmn_maskset32(struct bgmac *bgmac, u16 offset, + u32 mask, u32 set) +{ + bgmac->cmn_maskset32(bgmac, offset, mask, set); } static inline void bgmac_maskset(struct bgmac *bgmac, u16 offset, u32 mask, -- cgit v0.10.2 From fa17806cde76fb1087532f07e72aa757a30e0500 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 8 Jul 2016 05:18:24 +0200 Subject: ipv4: do not abuse GFP_ATOMIC in inet_netconf_notify_devconf() inet_forward_change() runs with RTNL held. We are allowed to sleep if required. If we use __in_dev_get_rtnl() instead of __in_dev_get_rcu(), we no longer have to use GFP_ATOMIC allocations in inet_netconf_notify_devconf(), meaning we are less likely to miss notifications under memory pressure, and wont touch precious memory reserves either and risk dropping incoming packets. inet_netconf_get_devconf() can also use GFP_KERNEL allocation. Fixes: edc9e748934c ("rtnl/ipv4: use netconf msg to advertise forwarding status") Fixes: 9e5511106f99 ("rtnl/ipv4: add support of RTM_GETNETCONF") Signed-off-by: Eric Dumazet Cc: Nicolas Dichtel Acked-by: Nicolas Dichtel Signed-off-by: David S. Miller diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index e333bc8..415e117 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1834,7 +1834,7 @@ void inet_netconf_notify_devconf(struct net *net, int type, int ifindex, struct sk_buff *skb; int err = -ENOBUFS; - skb = nlmsg_new(inet_netconf_msgsize_devconf(type), GFP_ATOMIC); + skb = nlmsg_new(inet_netconf_msgsize_devconf(type), GFP_KERNEL); if (!skb) goto errout; @@ -1846,7 +1846,7 @@ void inet_netconf_notify_devconf(struct net *net, int type, int ifindex, kfree_skb(skb); goto errout; } - rtnl_notify(skb, net, 0, RTNLGRP_IPV4_NETCONF, NULL, GFP_ATOMIC); + rtnl_notify(skb, net, 0, RTNLGRP_IPV4_NETCONF, NULL, GFP_KERNEL); return; errout: if (err < 0) @@ -1903,7 +1903,7 @@ static int inet_netconf_get_devconf(struct sk_buff *in_skb, } err = -ENOBUFS; - skb = nlmsg_new(inet_netconf_msgsize_devconf(NETCONFA_ALL), GFP_ATOMIC); + skb = nlmsg_new(inet_netconf_msgsize_devconf(NETCONFA_ALL), GFP_KERNEL); if (!skb) goto errout; @@ -2027,16 +2027,16 @@ static void inet_forward_change(struct net *net) for_each_netdev(net, dev) { struct in_device *in_dev; + if (on) dev_disable_lro(dev); - rcu_read_lock(); - in_dev = __in_dev_get_rcu(dev); + + in_dev = __in_dev_get_rtnl(dev); if (in_dev) { IN_DEV_CONF_SET(in_dev, FORWARDING, on); inet_netconf_notify_devconf(net, NETCONFA_FORWARDING, dev->ifindex, &in_dev->cnf); } - rcu_read_unlock(); } } -- cgit v0.10.2 From 927265bc6cd6374c9bafc43408ece4e92311b149 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 8 Jul 2016 05:46:04 +0200 Subject: ipv6: do not abuse GFP_ATOMIC in inet6_netconf_notify_devconf() All inet6_netconf_notify_devconf() callers are in process context, so we can use GFP_KERNEL allocations if we take care of not holding a rwlock while not needed in ip6mr (we hold RTNL there) Fixes: d67b8c616b48 ("netconf: advertise mc_forwarding status") Fixes: f3a1bfb11ccb ("rtnl/ipv6: use netconf msg to advertise forwarding status") Signed-off-by: Eric Dumazet Cc: Nicolas Dichtel Acked-by: Nicolas Dichtel Signed-off-by: David S. Miller diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index a1f6b7b..24f1b08 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -547,7 +547,7 @@ void inet6_netconf_notify_devconf(struct net *net, int type, int ifindex, struct sk_buff *skb; int err = -ENOBUFS; - skb = nlmsg_new(inet6_netconf_msgsize_devconf(type), GFP_ATOMIC); + skb = nlmsg_new(inet6_netconf_msgsize_devconf(type), GFP_KERNEL); if (!skb) goto errout; @@ -559,7 +559,7 @@ void inet6_netconf_notify_devconf(struct net *net, int type, int ifindex, kfree_skb(skb); goto errout; } - rtnl_notify(skb, net, 0, RTNLGRP_IPV6_NETCONF, NULL, GFP_ATOMIC); + rtnl_notify(skb, net, 0, RTNLGRP_IPV6_NETCONF, NULL, GFP_KERNEL); return; errout: rtnl_set_sk_err(net, RTNLGRP_IPV6_NETCONF, err); diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 487ef3b..c7ca0f5 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -1592,14 +1592,15 @@ static int ip6mr_sk_init(struct mr6_table *mrt, struct sock *sk) if (likely(mrt->mroute6_sk == NULL)) { mrt->mroute6_sk = sk; net->ipv6.devconf_all->mc_forwarding++; - inet6_netconf_notify_devconf(net, NETCONFA_MC_FORWARDING, - NETCONFA_IFINDEX_ALL, - net->ipv6.devconf_all); - } - else + } else { err = -EADDRINUSE; + } write_unlock_bh(&mrt_lock); + if (!err) + inet6_netconf_notify_devconf(net, NETCONFA_MC_FORWARDING, + NETCONFA_IFINDEX_ALL, + net->ipv6.devconf_all); rtnl_unlock(); return err; @@ -1617,11 +1618,11 @@ int ip6mr_sk_done(struct sock *sk) write_lock_bh(&mrt_lock); mrt->mroute6_sk = NULL; net->ipv6.devconf_all->mc_forwarding--; + write_unlock_bh(&mrt_lock); inet6_netconf_notify_devconf(net, NETCONFA_MC_FORWARDING, NETCONFA_IFINDEX_ALL, net->ipv6.devconf_all); - write_unlock_bh(&mrt_lock); mroute_clean_tables(mrt, false); err = 0; -- cgit v0.10.2 From 1d984c2e03c1fb21539a9f50627e312788512013 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Fri, 8 Jul 2016 15:52:39 +0200 Subject: NFC: digital: Fix handling of saved PDU sk_buff pointers This patch fixes the way an I-PDU is saved in case it needs to be sent again. It is now copied using pskb_copy() and not simply referenced using skb_get() since it could be modified by the driver. digital_in_send_saved_skb() and digital_tg_send_saved_skb() still get a reference on the saved skb which is re-sent but release it if the send operation fails. That way the caller doesn't have to take care about skb ref in case of error. RTOX supervisor PDU must not be saved as this can override a previously saved I-PDU that should be re-sent later on. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/include/net/nfc/digital.h b/include/net/nfc/digital.h index 506e3f6..f9a4e47 100644 --- a/include/net/nfc/digital.h +++ b/include/net/nfc/digital.h @@ -237,7 +237,6 @@ struct nfc_digital_dev { int nack_count; struct sk_buff *saved_skb; - unsigned int saved_skb_len; u16 target_fsc; diff --git a/net/nfc/digital_dep.c b/net/nfc/digital_dep.c index b62c85d..804585c 100644 --- a/net/nfc/digital_dep.c +++ b/net/nfc/digital_dep.c @@ -524,8 +524,7 @@ static int digital_in_send_ack(struct nfc_digital_dev *ddev, ddev->skb_add_crc(skb); - ddev->saved_skb = skb_get(skb); - ddev->saved_skb_len = skb->len; + ddev->saved_skb = pskb_copy(skb, GFP_KERNEL); rc = digital_in_send_cmd(ddev, skb, 1500, digital_in_recv_dep_res, data_exch); @@ -627,16 +626,10 @@ static int digital_in_send_rtox(struct nfc_digital_dev *ddev, ddev->skb_add_crc(skb); - ddev->saved_skb = skb_get(skb); - ddev->saved_skb_len = skb->len; - rc = digital_in_send_cmd(ddev, skb, 1500, digital_in_recv_dep_res, data_exch); - if (rc) { + if (rc) kfree_skb(skb); - kfree_skb(ddev->saved_skb); - ddev->saved_skb = NULL; - } return rc; } @@ -644,11 +637,19 @@ static int digital_in_send_rtox(struct nfc_digital_dev *ddev, static int digital_in_send_saved_skb(struct nfc_digital_dev *ddev, struct digital_data_exch *data_exch) { + int rc; + + if (!ddev->saved_skb) + return -EINVAL; + skb_get(ddev->saved_skb); - skb_push(ddev->saved_skb, ddev->saved_skb_len); - return digital_in_send_cmd(ddev, ddev->saved_skb, 1500, - digital_in_recv_dep_res, data_exch); + rc = digital_in_send_cmd(ddev, ddev->saved_skb, 1500, + digital_in_recv_dep_res, data_exch); + if (rc) + kfree_skb(ddev->saved_skb); + + return rc; } static void digital_in_recv_dep_res(struct nfc_digital_dev *ddev, void *arg, @@ -812,17 +813,12 @@ static void digital_in_recv_dep_res(struct nfc_digital_dev *ddev, void *arg, case DIGITAL_NFC_DEP_PFB_SUPERVISOR_PDU: if (!DIGITAL_NFC_DEP_PFB_IS_TIMEOUT(pfb)) { /* ATN */ rc = digital_in_send_saved_skb(ddev, data_exch); - if (rc) { - kfree_skb(ddev->saved_skb); + if (rc) goto error; - } return; } - kfree_skb(ddev->saved_skb); - ddev->saved_skb = NULL; - rc = digital_in_send_rtox(ddev, data_exch, resp->data[0]); if (rc) goto error; @@ -876,8 +872,7 @@ int digital_in_send_dep_req(struct nfc_digital_dev *ddev, ddev->skb_add_crc(tmp_skb); - ddev->saved_skb = skb_get(tmp_skb); - ddev->saved_skb_len = tmp_skb->len; + ddev->saved_skb = pskb_copy(tmp_skb, GFP_KERNEL); rc = digital_in_send_cmd(ddev, tmp_skb, 1500, digital_in_recv_dep_res, data_exch); @@ -956,8 +951,7 @@ static int digital_tg_send_ack(struct nfc_digital_dev *ddev, ddev->skb_add_crc(skb); - ddev->saved_skb = skb_get(skb); - ddev->saved_skb_len = skb->len; + ddev->saved_skb = pskb_copy(skb, GFP_KERNEL); rc = digital_tg_send_cmd(ddev, skb, 1500, digital_tg_recv_dep_req, data_exch); @@ -1009,11 +1003,19 @@ static int digital_tg_send_atn(struct nfc_digital_dev *ddev) static int digital_tg_send_saved_skb(struct nfc_digital_dev *ddev) { + int rc; + + if (!ddev->saved_skb) + return -EINVAL; + skb_get(ddev->saved_skb); - skb_push(ddev->saved_skb, ddev->saved_skb_len); - return digital_tg_send_cmd(ddev, ddev->saved_skb, 1500, - digital_tg_recv_dep_req, NULL); + rc = digital_tg_send_cmd(ddev, ddev->saved_skb, 1500, + digital_tg_recv_dep_req, NULL); + if (rc) + kfree_skb(ddev->saved_skb); + + return rc; } static void digital_tg_recv_dep_req(struct nfc_digital_dev *ddev, void *arg, @@ -1163,10 +1165,8 @@ static void digital_tg_recv_dep_req(struct nfc_digital_dev *ddev, void *arg, ddev->atn_count = 0; rc = digital_tg_send_saved_skb(ddev); - if (rc) { - kfree_skb(ddev->saved_skb); + if (rc) goto exit; - } } return; @@ -1235,8 +1235,7 @@ int digital_tg_send_dep_res(struct nfc_digital_dev *ddev, struct sk_buff *skb) ddev->skb_add_crc(tmp_skb); - ddev->saved_skb = skb_get(tmp_skb); - ddev->saved_skb_len = tmp_skb->len; + ddev->saved_skb = pskb_copy(tmp_skb, GFP_KERNEL); rc = digital_tg_send_cmd(ddev, tmp_skb, 1500, digital_tg_recv_dep_req, NULL); -- cgit v0.10.2 From e8e7f4217564fc115b60a9373646afb193aa08cf Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Fri, 8 Jul 2016 15:52:40 +0200 Subject: NFC: digital: Remove useless call to skb_reserve() When allocating chained I-PDUs, there is no need to call skb_reserve() since it's already done by digital_alloc_skb() and contains enough room for the driver head and tail data. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_dep.c b/net/nfc/digital_dep.c index 804585c..ed3a529 100644 --- a/net/nfc/digital_dep.c +++ b/net/nfc/digital_dep.c @@ -190,8 +190,6 @@ digital_send_dep_data_prep(struct nfc_digital_dev *ddev, struct sk_buff *skb, return ERR_PTR(-ENOMEM); } - skb_reserve(new_skb, ddev->tx_headroom + NFC_HEADER_SIZE + - DIGITAL_NFC_DEP_REQ_RES_HEADROOM); memcpy(skb_put(new_skb, ddev->remote_payload_max), skb->data, ddev->remote_payload_max); skb_pull(skb, ddev->remote_payload_max); -- cgit v0.10.2 From f23a9868b1c45e77ec6082eb95508885111ffda1 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Fri, 8 Jul 2016 15:52:41 +0200 Subject: NFC: digital: Fix target DEP_REQ I-PDU handling after ATN PDU When the initiator sends a DEP_REQ I-PDU, the target device may not reply in a timely manner. In this case the initiator device must send an attention PDU (ATN) and if the recipient replies with an ATN PDU in return, then the last I-PDU must be sent again by the initiator. This patch fixes how the target handles I-PDU received after an ATN PDU has been received. There are 2 possible cases: - The target has received the initial DEP_REQ and sends back the DEP_RES but the initiator did not receive it. In this case, after the initiator has sent an ATN PDU and the target replied it (with an ATN as well), the initiator sends the saved skb of the initial DEP_REQ again and the target replies with the saved skb of the initial DEP_RES. - Or the target did not even received the initial DEP_REQ. In this case, after the ATN PDUs exchange, the initiator sends the saved skb and the target simply passes it up, just as usual. This behavior is controlled using the atn_count and the PNI field of the digital device structure. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_dep.c b/net/nfc/digital_dep.c index ed3a529..1778c23 100644 --- a/net/nfc/digital_dep.c +++ b/net/nfc/digital_dep.c @@ -1086,22 +1086,38 @@ static void digital_tg_recv_dep_req(struct nfc_digital_dev *ddev, void *arg, case DIGITAL_NFC_DEP_PFB_I_PDU: pr_debug("DIGITAL_NFC_DEP_PFB_I_PDU\n"); - if ((ddev->atn_count && (DIGITAL_NFC_DEP_PFB_PNI(pfb - 1) != - ddev->curr_nfc_dep_pni)) || - (DIGITAL_NFC_DEP_PFB_PNI(pfb) != ddev->curr_nfc_dep_pni)) { - PROTOCOL_ERR("14.12.3.4"); - rc = -EIO; - goto exit; - } - if (ddev->atn_count) { + /* The target has received (and replied to) at least one + * ATN DEP_REQ. + */ ddev->atn_count = 0; - rc = digital_tg_send_saved_skb(ddev); - if (rc) - goto exit; + /* pni of resp PDU equal to the target current pni - 1 + * means resp is the previous DEP_REQ PDU received from + * the initiator so the target replies with saved_skb + * which is the previous DEP_RES saved in + * digital_tg_send_dep_res(). + */ + if (DIGITAL_NFC_DEP_PFB_PNI(pfb) == + DIGITAL_NFC_DEP_PFB_PNI(ddev->curr_nfc_dep_pni - 1)) { + rc = digital_tg_send_saved_skb(ddev); + if (rc) + goto exit; - return; + goto free_resp; + } + + /* atn_count > 0 and PDU pni != curr_nfc_dep_pni - 1 + * means the target probably did not received the last + * DEP_REQ PDU sent by the initiator. The target + * fallbacks to normal processing then. + */ + } + + if (DIGITAL_NFC_DEP_PFB_PNI(pfb) != ddev->curr_nfc_dep_pni) { + PROTOCOL_ERR("14.12.3.4"); + rc = -EIO; + goto exit; } kfree_skb(ddev->saved_skb); @@ -1197,6 +1213,11 @@ exit: if (rc) kfree_skb(resp); + + return; + +free_resp: + dev_kfree_skb(resp); } int digital_tg_send_dep_res(struct nfc_digital_dev *ddev, struct sk_buff *skb) -- cgit v0.10.2 From 482333b277de181ce80c833d84f2598e2527b267 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Fri, 8 Jul 2016 15:52:42 +0200 Subject: NFC: digital: Fix ACK & NACK PDUs handling in target mode When the target receives a NACK PDU, it re-sends the last sent PDU. ACK PDUs are received by the target as a reply from the initiator to chained I-PDUs. There are 3 cases to handle: - If the target has previously received 1 or more ATN PDUs and the PNI in the ACK PDU is equal to the target PNI - 1, then it means that the initiator did not received the last issued PDU from the target. In this case it re-sends this PDU. - If the target has received 1 or more ATN PDUs but the ACK PNI is not the target PNI - 1, then this means that this ACK is the reply of the previous chained I-PDU sent by the target. The target did not received it on the first attempt and it is being re-sent by the initiator. The process continues as usual. - No ATN PDU received before this ACK PDU. This is the reply of a chained I-PDU. The target keeps on processing its chained I-PDU. The code has been refactored to avoid too many indentation levels. Also, ACK and NACK PDUs were not freed. This is now fixed. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_dep.c b/net/nfc/digital_dep.c index 1778c23..e026877 100644 --- a/net/nfc/digital_dep.c +++ b/net/nfc/digital_dep.c @@ -1141,49 +1141,64 @@ static void digital_tg_recv_dep_req(struct nfc_digital_dev *ddev, void *arg, rc = 0; break; case DIGITAL_NFC_DEP_PFB_ACK_NACK_PDU: - if (!DIGITAL_NFC_DEP_NACK_BIT_SET(pfb)) { /* ACK */ - if ((ddev->atn_count && - (DIGITAL_NFC_DEP_PFB_PNI(pfb - 1) != - ddev->curr_nfc_dep_pni)) || - (DIGITAL_NFC_DEP_PFB_PNI(pfb) != - ddev->curr_nfc_dep_pni) || - !ddev->chaining_skb || !ddev->saved_skb) { + if (DIGITAL_NFC_DEP_NACK_BIT_SET(pfb)) { /* NACK */ + if (DIGITAL_NFC_DEP_PFB_PNI(pfb + 1) != + ddev->curr_nfc_dep_pni) { rc = -EIO; goto exit; } - if (ddev->atn_count) { - ddev->atn_count = 0; + ddev->atn_count = 0; + + rc = digital_tg_send_saved_skb(ddev); + if (rc) + goto exit; + + goto free_resp; + } + + /* ACK */ + if (ddev->atn_count) { + /* The target has previously recevied one or more ATN + * PDUs. + */ + ddev->atn_count = 0; + /* If the ACK PNI is equal to the target PNI - 1 means + * that the initiator did not receive the previous PDU + * sent by the target so re-send it. + */ + if (DIGITAL_NFC_DEP_PFB_PNI(pfb + 1) == + ddev->curr_nfc_dep_pni) { rc = digital_tg_send_saved_skb(ddev); if (rc) goto exit; - return; + goto free_resp; } - kfree_skb(ddev->saved_skb); - ddev->saved_skb = NULL; + /* Otherwise, the target did not receive the previous + * ACK PDU from the initiator. Fallback to normal + * processing of chained PDU then. + */ + } - rc = digital_tg_send_dep_res(ddev, ddev->chaining_skb); - if (rc) - goto exit; - } else { /* NACK */ - if ((DIGITAL_NFC_DEP_PFB_PNI(pfb + 1) != - ddev->curr_nfc_dep_pni) || - !ddev->saved_skb) { - rc = -EIO; - goto exit; - } + /* Keep on sending chained PDU */ + if (!ddev->chaining_skb || + DIGITAL_NFC_DEP_PFB_PNI(pfb) != + ddev->curr_nfc_dep_pni) { + rc = -EIO; + goto exit; + } - ddev->atn_count = 0; + kfree_skb(ddev->saved_skb); + ddev->saved_skb = NULL; - rc = digital_tg_send_saved_skb(ddev); - if (rc) - goto exit; - } + rc = digital_tg_send_dep_res(ddev, ddev->chaining_skb); + if (rc) + goto exit; - return; + goto free_resp; case DIGITAL_NFC_DEP_PFB_SUPERVISOR_PDU: if (DIGITAL_NFC_DEP_PFB_IS_TIMEOUT(pfb)) { rc = -EINVAL; -- cgit v0.10.2 From e073eb6797191abe2fe30ca643ab0cc3d8e1e534 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Fri, 8 Jul 2016 15:52:43 +0200 Subject: NFC: digital: Rework ACK PDU handling in initiator mode With this patch, ACK PDU sk_buffs are now freed and code has been refactored for better errors handling. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_dep.c b/net/nfc/digital_dep.c index e026877..03bfc74 100644 --- a/net/nfc/digital_dep.c +++ b/net/nfc/digital_dep.c @@ -782,6 +782,12 @@ static void digital_in_recv_dep_res(struct nfc_digital_dev *ddev, void *arg, break; case DIGITAL_NFC_DEP_PFB_ACK_NACK_PDU: + if (DIGITAL_NFC_DEP_NACK_BIT_SET(pfb)) { + PROTOCOL_ERR("14.12.4.5"); + rc = -EIO; + goto exit; + } + if (DIGITAL_NFC_DEP_PFB_PNI(pfb) != ddev->curr_nfc_dep_pni) { PROTOCOL_ERR("14.12.3.3"); rc = -EIO; @@ -791,22 +797,25 @@ static void digital_in_recv_dep_res(struct nfc_digital_dev *ddev, void *arg, ddev->curr_nfc_dep_pni = DIGITAL_NFC_DEP_PFB_PNI(ddev->curr_nfc_dep_pni + 1); - if (ddev->chaining_skb && !DIGITAL_NFC_DEP_NACK_BIT_SET(pfb)) { - kfree_skb(ddev->saved_skb); - ddev->saved_skb = NULL; + if (!ddev->chaining_skb) { + PROTOCOL_ERR("14.12.4.3"); + rc = -EIO; + goto exit; + } - rc = digital_in_send_dep_req(ddev, NULL, - ddev->chaining_skb, - ddev->data_exch); - if (rc) - goto error; + /* The initiator has received a valid ACK. Free the last sent + * PDU and keep on sending chained skb. + */ + kfree_skb(ddev->saved_skb); + ddev->saved_skb = NULL; - return; - } + rc = digital_in_send_dep_req(ddev, NULL, + ddev->chaining_skb, + ddev->data_exch); + if (rc) + goto error; - pr_err("Received a ACK/NACK PDU\n"); - rc = -EINVAL; - goto exit; + goto free_resp; case DIGITAL_NFC_DEP_PFB_SUPERVISOR_PDU: if (!DIGITAL_NFC_DEP_PFB_IS_TIMEOUT(pfb)) { /* ATN */ @@ -839,6 +848,11 @@ error: if (rc) kfree_skb(resp); + + return; + +free_resp: + dev_kfree_skb(resp); } int digital_in_send_dep_req(struct nfc_digital_dev *ddev, -- cgit v0.10.2 From e200f008ace69eebac0a1432dc9e24ab5cd0d029 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Fri, 8 Jul 2016 15:52:44 +0200 Subject: NFC: digital: Free supervisor PDUs This patch frees the RTOX resp sk_buff in initiator mode. It also makes use of the free_resp exit point for ATN supervisor PDUs in both initiator and target mode. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_dep.c b/net/nfc/digital_dep.c index 03bfc74..ba52a5d 100644 --- a/net/nfc/digital_dep.c +++ b/net/nfc/digital_dep.c @@ -823,15 +823,14 @@ static void digital_in_recv_dep_res(struct nfc_digital_dev *ddev, void *arg, if (rc) goto error; - return; + goto free_resp; } rc = digital_in_send_rtox(ddev, data_exch, resp->data[0]); if (rc) goto error; - kfree_skb(resp); - return; + goto free_resp; } exit: @@ -1225,8 +1224,7 @@ static void digital_tg_recv_dep_req(struct nfc_digital_dev *ddev, void *arg, ddev->atn_count++; - kfree_skb(resp); - return; + goto free_resp; } rc = nfc_tm_data_received(ddev->nfc_dev, resp); -- cgit v0.10.2 From 1a09c56f545c8ff8d338a38c7c40d79f4165a94c Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Fri, 8 Jul 2016 15:52:45 +0200 Subject: NFC: digital: Add support for NFC DEP Response Waiting Time When sending an ATR_REQ, the initiator must wait for the ATR_RES at least 'RWT(nfcdep,activation) + dRWT(nfcdep)' and no more than 'RWT(nfcdep,activation) + dRWT(nfcdep) + dT(nfcdep,initiator)'. This gives a timeout value between 1237 ms and 1337 ms. This patch defines DIGITAL_ATR_RES_RWT to 1337 used for the timeout value of ATR_REQ command. For other DEP PDUs, the initiator must wait between 'RWT + dRWT(nfcdep)' and 'RWT + dRWT(nfcdep) + dT(nfcdep,initiator)' where RWT is given by the following formula: '(256 * 16 / f(c)) * 2^wt' where wt is the value of the TO field in the ATR_RES response and is in the range between 0 and 14. This patch declares a mapping table for wt values and gives RWT max values between 100 ms and 5049 ms. This patch also defines DIGITAL_ATR_RES_TO_WT, the maximum wt value in target mode, to 8. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/include/net/nfc/digital.h b/include/net/nfc/digital.h index f9a4e47..74fa7eb 100644 --- a/include/net/nfc/digital.h +++ b/include/net/nfc/digital.h @@ -226,6 +226,7 @@ struct nfc_digital_dev { u8 curr_rf_tech; u8 curr_nfc_dep_pni; u8 did; + u16 dep_rwt; u8 local_payload_max; u8 remote_payload_max; diff --git a/net/nfc/digital_dep.c b/net/nfc/digital_dep.c index ba52a5d..6cf2eeb 100644 --- a/net/nfc/digital_dep.c +++ b/net/nfc/digital_dep.c @@ -35,6 +35,8 @@ #define DIGITAL_ATR_REQ_MIN_SIZE 16 #define DIGITAL_ATR_REQ_MAX_SIZE 64 +#define DIGITAL_ATR_RES_TO_WT(s) ((s) & 0xF) + #define DIGITAL_DID_MAX 14 #define DIGITAL_PAYLOAD_SIZE_MAX 254 @@ -122,6 +124,37 @@ static const u8 digital_payload_bits_map[4] = { [3] = 254 }; +/* Response Waiting Time for ATR_RES PDU in ms + * + * RWT(ATR_RES) = RWT(nfcdep,activation) + dRWT(nfcdep) + dT(nfcdep,initiator) + * + * with: + * RWT(nfcdep,activation) = 4096 * 2^12 / f(c) s + * dRWT(nfcdep) = 16 / f(c) s + * dT(nfcdep,initiator) = 100 ms + * f(c) = 13560000 Hz + */ +#define DIGITAL_ATR_RES_RWT 1337 + +/* Response Waiting Time for other DEP PDUs in ms + * + * max_rwt = rwt + dRWT(nfcdep) + dT(nfcdep,initiator) + * + * with: + * rwt = (256 * 16 / f(c)) * 2^wt s + * dRWT(nfcdep) = 16 / f(c) s + * dT(nfcdep,initiator) = 100 ms + * f(c) = 13560000 Hz + * 0 <= wt <= 14 (given by the target by the TO field of ATR_RES response) + */ +#define DIGITAL_NFC_DEP_IN_MAX_WT 14 +#define DIGITAL_NFC_DEP_TG_MAX_WT 8 +static const u16 digital_rwt_map[DIGITAL_NFC_DEP_IN_MAX_WT + 1] = { + 100, 101, 101, 102, 105, + 110, 119, 139, 177, 255, + 409, 719, 1337, 2575, 5049, +}; + static u8 digital_payload_bits_to_size(u8 payload_bits) { if (payload_bits >= ARRAY_SIZE(digital_payload_bits_map)) @@ -366,8 +399,8 @@ static int digital_in_send_psl_req(struct nfc_digital_dev *ddev, ddev->skb_add_crc(skb); - rc = digital_in_send_cmd(ddev, skb, 500, digital_in_recv_psl_res, - target); + rc = digital_in_send_cmd(ddev, skb, ddev->dep_rwt, + digital_in_recv_psl_res, target); if (rc) kfree_skb(skb); @@ -380,6 +413,7 @@ static void digital_in_recv_atr_res(struct nfc_digital_dev *ddev, void *arg, struct nfc_target *target = arg; struct digital_atr_res *atr_res; u8 gb_len, payload_bits; + u8 wt; int rc; if (IS_ERR(resp)) { @@ -409,6 +443,11 @@ static void digital_in_recv_atr_res(struct nfc_digital_dev *ddev, void *arg, atr_res = (struct digital_atr_res *)resp->data; + wt = DIGITAL_ATR_RES_TO_WT(atr_res->to); + if (wt > DIGITAL_NFC_DEP_IN_MAX_WT) + wt = DIGITAL_NFC_DEP_IN_MAX_WT; + ddev->dep_rwt = digital_rwt_map[wt]; + payload_bits = DIGITAL_PAYLOAD_PP_TO_BITS(atr_res->pp); ddev->remote_payload_max = digital_payload_bits_to_size(payload_bits); @@ -490,8 +529,8 @@ int digital_in_send_atr_req(struct nfc_digital_dev *ddev, ddev->skb_add_crc(skb); - rc = digital_in_send_cmd(ddev, skb, 500, digital_in_recv_atr_res, - target); + rc = digital_in_send_cmd(ddev, skb, DIGITAL_ATR_RES_RWT, + digital_in_recv_atr_res, target); if (rc) kfree_skb(skb); @@ -524,8 +563,8 @@ static int digital_in_send_ack(struct nfc_digital_dev *ddev, ddev->saved_skb = pskb_copy(skb, GFP_KERNEL); - rc = digital_in_send_cmd(ddev, skb, 1500, digital_in_recv_dep_res, - data_exch); + rc = digital_in_send_cmd(ddev, skb, ddev->dep_rwt, + digital_in_recv_dep_res, data_exch); if (rc) { kfree_skb(skb); kfree_skb(ddev->saved_skb); @@ -559,8 +598,8 @@ static int digital_in_send_nack(struct nfc_digital_dev *ddev, ddev->skb_add_crc(skb); - rc = digital_in_send_cmd(ddev, skb, 1500, digital_in_recv_dep_res, - data_exch); + rc = digital_in_send_cmd(ddev, skb, ddev->dep_rwt, + digital_in_recv_dep_res, data_exch); if (rc) kfree_skb(skb); @@ -590,8 +629,8 @@ static int digital_in_send_atn(struct nfc_digital_dev *ddev, ddev->skb_add_crc(skb); - rc = digital_in_send_cmd(ddev, skb, 1500, digital_in_recv_dep_res, - data_exch); + rc = digital_in_send_cmd(ddev, skb, ddev->dep_rwt, + digital_in_recv_dep_res, data_exch); if (rc) kfree_skb(skb); @@ -624,8 +663,8 @@ static int digital_in_send_rtox(struct nfc_digital_dev *ddev, ddev->skb_add_crc(skb); - rc = digital_in_send_cmd(ddev, skb, 1500, digital_in_recv_dep_res, - data_exch); + rc = digital_in_send_cmd(ddev, skb, ddev->dep_rwt, + digital_in_recv_dep_res, data_exch); if (rc) kfree_skb(skb); @@ -642,7 +681,7 @@ static int digital_in_send_saved_skb(struct nfc_digital_dev *ddev, skb_get(ddev->saved_skb); - rc = digital_in_send_cmd(ddev, ddev->saved_skb, 1500, + rc = digital_in_send_cmd(ddev, ddev->saved_skb, ddev->dep_rwt, digital_in_recv_dep_res, data_exch); if (rc) kfree_skb(ddev->saved_skb); @@ -885,8 +924,8 @@ int digital_in_send_dep_req(struct nfc_digital_dev *ddev, ddev->saved_skb = pskb_copy(tmp_skb, GFP_KERNEL); - rc = digital_in_send_cmd(ddev, tmp_skb, 1500, digital_in_recv_dep_res, - data_exch); + rc = digital_in_send_cmd(ddev, tmp_skb, ddev->dep_rwt, + digital_in_recv_dep_res, data_exch); if (rc) { if (tmp_skb != skb) kfree_skb(tmp_skb); @@ -1465,7 +1504,7 @@ static int digital_tg_send_atr_res(struct nfc_digital_dev *ddev, atr_res->dir = DIGITAL_NFC_DEP_FRAME_DIR_IN; atr_res->cmd = DIGITAL_CMD_ATR_RES; memcpy(atr_res->nfcid3, atr_req->nfcid3, sizeof(atr_req->nfcid3)); - atr_res->to = 8; + atr_res->to = DIGITAL_NFC_DEP_TG_MAX_WT; ddev->local_payload_max = DIGITAL_PAYLOAD_SIZE_MAX; payload_bits = digital_payload_size_to_bits(ddev->local_payload_max); -- cgit v0.10.2 From d85a301c26621d3466956dc477c32c20c15a52ee Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Fri, 8 Jul 2016 15:52:46 +0200 Subject: NFC: digital: Fix RTOX supervisor PDU handling When the target needs more time to process the received PDU, it sends Response Timeout Extension (RTOX) PDU. When the initiator receives a RTOX PDU, it must reply with a RTOX PDU and extends the current rwt value with the formula: rwt_int = rwt * rtox This patch takes care of the rtox value passed by the target in the RTOX PDU and extends the timeout for the next response accordingly. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/net/nfc/digital_dep.c b/net/nfc/digital_dep.c index 6cf2eeb..f864ce1 100644 --- a/net/nfc/digital_dep.c +++ b/net/nfc/digital_dep.c @@ -65,6 +65,9 @@ #define DIGITAL_NFC_DEP_DID_BIT_SET(pfb) ((pfb) & DIGITAL_NFC_DEP_PFB_DID_BIT) #define DIGITAL_NFC_DEP_PFB_PNI(pfb) ((pfb) & 0x03) +#define DIGITAL_NFC_DEP_RTOX_VALUE(data) ((data) & 0x3F) +#define DIGITAL_NFC_DEP_RTOX_MAX 59 + #define DIGITAL_NFC_DEP_PFB_I_PDU 0x00 #define DIGITAL_NFC_DEP_PFB_ACK_NACK_PDU 0x40 #define DIGITAL_NFC_DEP_PFB_SUPERVISOR_PDU 0x80 @@ -643,6 +646,11 @@ static int digital_in_send_rtox(struct nfc_digital_dev *ddev, struct digital_dep_req_res *dep_req; struct sk_buff *skb; int rc; + u16 rwt_int; + + rwt_int = ddev->dep_rwt * rtox; + if (rwt_int > digital_rwt_map[DIGITAL_NFC_DEP_IN_MAX_WT]) + rwt_int = digital_rwt_map[DIGITAL_NFC_DEP_IN_MAX_WT]; skb = digital_skb_alloc(ddev, 1); if (!skb) @@ -663,7 +671,7 @@ static int digital_in_send_rtox(struct nfc_digital_dev *ddev, ddev->skb_add_crc(skb); - rc = digital_in_send_cmd(ddev, skb, ddev->dep_rwt, + rc = digital_in_send_cmd(ddev, skb, rwt_int, digital_in_recv_dep_res, data_exch); if (rc) kfree_skb(skb); @@ -697,6 +705,7 @@ static void digital_in_recv_dep_res(struct nfc_digital_dev *ddev, void *arg, u8 pfb; uint size; int rc; + u8 rtox; if (IS_ERR(resp)) { rc = PTR_ERR(resp); @@ -865,7 +874,20 @@ static void digital_in_recv_dep_res(struct nfc_digital_dev *ddev, void *arg, goto free_resp; } - rc = digital_in_send_rtox(ddev, data_exch, resp->data[0]); + if (ddev->atn_count || ddev->nack_count) { + PROTOCOL_ERR("14.12.4.4"); + rc = -EIO; + goto error; + } + + rtox = DIGITAL_NFC_DEP_RTOX_VALUE(resp->data[0]); + if (!rtox || rtox > DIGITAL_NFC_DEP_RTOX_MAX) { + PROTOCOL_ERR("14.8.4.1"); + rc = -EIO; + goto error; + } + + rc = digital_in_send_rtox(ddev, data_exch, rtox); if (rc) goto error; -- cgit v0.10.2 From fb3bbdb859891e6bc27fd1afb3a07319f82c2ee4 Mon Sep 17 00:00:00 2001 From: Tien Hock Loh Date: Thu, 7 Jul 2016 20:23:30 -0700 Subject: net: ethernet: Add TSE PCS support to dwmac-socfpga This adds support for TSE PCS that uses SGMII adapter when the phy-mode of the dwmac is set to sgmii. Signed-off-by: Tien Hock Loh Acked-by: Rob Herring Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/socfpga-dwmac.txt b/Documentation/devicetree/bindings/net/socfpga-dwmac.txt index 72d82d6..2e68a3c 100644 --- a/Documentation/devicetree/bindings/net/socfpga-dwmac.txt +++ b/Documentation/devicetree/bindings/net/socfpga-dwmac.txt @@ -17,9 +17,26 @@ Required properties: Optional properties: altr,emac-splitter: Should be the phandle to the emac splitter soft IP node if DWMAC controller is connected emac splitter. +phy-mode: The phy mode the ethernet operates in +altr,sgmii-to-sgmii-converter: phandle to the TSE SGMII converter + +This device node has additional phandle dependency, the sgmii converter: + +Required properties: + - compatible : Should be altr,gmii-to-sgmii-2.0 + - reg-names : Should be "eth_tse_control_port" Example: +gmii_to_sgmii_converter: phy@0x100000240 { + compatible = "altr,gmii-to-sgmii-2.0"; + reg = <0x00000001 0x00000240 0x00000008>, + <0x00000001 0x00000200 0x00000040>; + reg-names = "eth_tse_control_port"; + clocks = <&sgmii_1_clk_0 &emac1 1 &sgmii_clk_125 &sgmii_clk_125>; + clock-names = "tse_pcs_ref_clk_clock_connection", "tse_rx_cdr_refclk"; +}; + gmac0: ethernet@ff700000 { compatible = "altr,socfpga-stmmac", "snps,dwmac-3.70a", "snps,dwmac"; altr,sysmgr-syscon = <&sysmgr 0x60 0>; @@ -30,4 +47,6 @@ gmac0: ethernet@ff700000 { mac-address = [00 00 00 00 00 00];/* Filled in by U-Boot */ clocks = <&emac_0_clk>; clock-names = "stmmaceth"; + phy-mode = "sgmii"; + altr,gmii-to-sgmii-converter = <&gmii_to_sgmii_converter>; }; diff --git a/drivers/net/ethernet/stmicro/stmmac/Makefile b/drivers/net/ethernet/stmicro/stmmac/Makefile index 0fb362d..44b630c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Makefile +++ b/drivers/net/ethernet/stmicro/stmmac/Makefile @@ -11,11 +11,12 @@ obj-$(CONFIG_DWMAC_IPQ806X) += dwmac-ipq806x.o obj-$(CONFIG_DWMAC_LPC18XX) += dwmac-lpc18xx.o obj-$(CONFIG_DWMAC_MESON) += dwmac-meson.o obj-$(CONFIG_DWMAC_ROCKCHIP) += dwmac-rk.o -obj-$(CONFIG_DWMAC_SOCFPGA) += dwmac-socfpga.o +obj-$(CONFIG_DWMAC_SOCFPGA) += dwmac-altr-socfpga.o obj-$(CONFIG_DWMAC_STI) += dwmac-sti.o obj-$(CONFIG_DWMAC_SUNXI) += dwmac-sunxi.o obj-$(CONFIG_DWMAC_GENERIC) += dwmac-generic.o stmmac-platform-objs:= stmmac_platform.o +dwmac-altr-socfpga-objs := altr_tse_pcs.o dwmac-socfpga.o obj-$(CONFIG_STMMAC_PCI) += stmmac-pci.o stmmac-pci-objs:= stmmac_pci.o diff --git a/drivers/net/ethernet/stmicro/stmmac/altr_tse_pcs.c b/drivers/net/ethernet/stmicro/stmmac/altr_tse_pcs.c new file mode 100644 index 0000000..2920e2e --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/altr_tse_pcs.c @@ -0,0 +1,274 @@ +/* Copyright Altera Corporation (C) 2016. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + * + * Author: Tien Hock Loh + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "stmmac.h" +#include "stmmac_platform.h" +#include "altr_tse_pcs.h" + +#define SYSMGR_EMACGRP_CTRL_PHYSEL_ENUM_GMII_MII 0 +#define SYSMGR_EMACGRP_CTRL_PHYSEL_ENUM_RGMII BIT(1) +#define SYSMGR_EMACGRP_CTRL_PHYSEL_ENUM_RMII BIT(2) +#define SYSMGR_EMACGRP_CTRL_PHYSEL_WIDTH 2 +#define SYSMGR_EMACGRP_CTRL_PHYSEL_MASK GENMASK(1, 0) + +#define TSE_PCS_CONTROL_AN_EN_MASK BIT(12) +#define TSE_PCS_CONTROL_REG 0x00 +#define TSE_PCS_CONTROL_RESTART_AN_MASK BIT(9) +#define TSE_PCS_IF_MODE_REG 0x28 +#define TSE_PCS_LINK_TIMER_0_REG 0x24 +#define TSE_PCS_LINK_TIMER_1_REG 0x26 +#define TSE_PCS_SIZE 0x40 +#define TSE_PCS_STATUS_AN_COMPLETED_MASK BIT(5) +#define TSE_PCS_STATUS_LINK_MASK 0x0004 +#define TSE_PCS_STATUS_REG 0x02 +#define TSE_PCS_SGMII_SPEED_1000 BIT(3) +#define TSE_PCS_SGMII_SPEED_100 BIT(2) +#define TSE_PCS_SGMII_SPEED_10 0x0 +#define TSE_PCS_SW_RST_MASK 0x8000 +#define TSE_PCS_PARTNER_ABILITY_REG 0x0A +#define TSE_PCS_PARTNER_DUPLEX_FULL 0x1000 +#define TSE_PCS_PARTNER_DUPLEX_HALF 0x0000 +#define TSE_PCS_PARTNER_DUPLEX_MASK 0x1000 +#define TSE_PCS_PARTNER_SPEED_MASK GENMASK(11, 10) +#define TSE_PCS_PARTNER_SPEED_1000 BIT(11) +#define TSE_PCS_PARTNER_SPEED_100 BIT(10) +#define TSE_PCS_PARTNER_SPEED_10 0x0000 +#define TSE_PCS_PARTNER_SPEED_1000 BIT(11) +#define TSE_PCS_PARTNER_SPEED_100 BIT(10) +#define TSE_PCS_PARTNER_SPEED_10 0x0000 +#define TSE_PCS_SGMII_SPEED_MASK GENMASK(3, 2) +#define TSE_PCS_SGMII_LINK_TIMER_0 0x0D40 +#define TSE_PCS_SGMII_LINK_TIMER_1 0x0003 +#define TSE_PCS_SW_RESET_TIMEOUT 100 +#define TSE_PCS_USE_SGMII_AN_MASK BIT(2) +#define TSE_PCS_USE_SGMII_ENA BIT(1) + +#define SGMII_ADAPTER_CTRL_REG 0x00 +#define SGMII_ADAPTER_DISABLE 0x0001 +#define SGMII_ADAPTER_ENABLE 0x0000 + +#define AUTONEGO_LINK_TIMER 20 + +static int tse_pcs_reset(void __iomem *base, struct tse_pcs *pcs) +{ + int counter = 0; + u16 val; + + val = readw(base + TSE_PCS_CONTROL_REG); + val |= TSE_PCS_SW_RST_MASK; + writew(val, base + TSE_PCS_CONTROL_REG); + + while (counter < TSE_PCS_SW_RESET_TIMEOUT) { + val = readw(base + TSE_PCS_CONTROL_REG); + val &= TSE_PCS_SW_RST_MASK; + if (val == 0) + break; + counter++; + udelay(1); + } + if (counter >= TSE_PCS_SW_RESET_TIMEOUT) { + dev_err(pcs->dev, "PCS could not get out of sw reset\n"); + return -ETIMEDOUT; + } + + return 0; +} + +int tse_pcs_init(void __iomem *base, struct tse_pcs *pcs) +{ + int ret = 0; + + writew(TSE_PCS_USE_SGMII_ENA, base + TSE_PCS_IF_MODE_REG); + + writew(TSE_PCS_SGMII_LINK_TIMER_0, base + TSE_PCS_LINK_TIMER_0_REG); + writew(TSE_PCS_SGMII_LINK_TIMER_1, base + TSE_PCS_LINK_TIMER_1_REG); + + ret = tse_pcs_reset(base, pcs); + if (ret == 0) + writew(SGMII_ADAPTER_ENABLE, + pcs->sgmii_adapter_base + SGMII_ADAPTER_CTRL_REG); + + return ret; +} + +static void pcs_link_timer_callback(unsigned long data) +{ + u16 val = 0; + struct tse_pcs *pcs = (struct tse_pcs *)data; + void __iomem *tse_pcs_base = pcs->tse_pcs_base; + void __iomem *sgmii_adapter_base = pcs->sgmii_adapter_base; + + val = readw(tse_pcs_base + TSE_PCS_STATUS_REG); + val &= TSE_PCS_STATUS_LINK_MASK; + + if (val != 0) { + dev_dbg(pcs->dev, "Adapter: Link is established\n"); + writew(SGMII_ADAPTER_ENABLE, + sgmii_adapter_base + SGMII_ADAPTER_CTRL_REG); + } else { + mod_timer(&pcs->aneg_link_timer, jiffies + + msecs_to_jiffies(AUTONEGO_LINK_TIMER)); + } +} + +static void auto_nego_timer_callback(unsigned long data) +{ + u16 val = 0; + u16 speed = 0; + u16 duplex = 0; + struct tse_pcs *pcs = (struct tse_pcs *)data; + void __iomem *tse_pcs_base = pcs->tse_pcs_base; + void __iomem *sgmii_adapter_base = pcs->sgmii_adapter_base; + + val = readw(tse_pcs_base + TSE_PCS_STATUS_REG); + val &= TSE_PCS_STATUS_AN_COMPLETED_MASK; + + if (val != 0) { + dev_dbg(pcs->dev, "Adapter: Auto Negotiation is completed\n"); + val = readw(tse_pcs_base + TSE_PCS_PARTNER_ABILITY_REG); + speed = val & TSE_PCS_PARTNER_SPEED_MASK; + duplex = val & TSE_PCS_PARTNER_DUPLEX_MASK; + + if (speed == TSE_PCS_PARTNER_SPEED_10 && + duplex == TSE_PCS_PARTNER_DUPLEX_FULL) + dev_dbg(pcs->dev, + "Adapter: Link Partner is Up - 10/Full\n"); + else if (speed == TSE_PCS_PARTNER_SPEED_100 && + duplex == TSE_PCS_PARTNER_DUPLEX_FULL) + dev_dbg(pcs->dev, + "Adapter: Link Partner is Up - 100/Full\n"); + else if (speed == TSE_PCS_PARTNER_SPEED_1000 && + duplex == TSE_PCS_PARTNER_DUPLEX_FULL) + dev_dbg(pcs->dev, + "Adapter: Link Partner is Up - 1000/Full\n"); + else if (speed == TSE_PCS_PARTNER_SPEED_10 && + duplex == TSE_PCS_PARTNER_DUPLEX_HALF) + dev_err(pcs->dev, + "Adapter does not support Half Duplex\n"); + else if (speed == TSE_PCS_PARTNER_SPEED_100 && + duplex == TSE_PCS_PARTNER_DUPLEX_HALF) + dev_err(pcs->dev, + "Adapter does not support Half Duplex\n"); + else if (speed == TSE_PCS_PARTNER_SPEED_1000 && + duplex == TSE_PCS_PARTNER_DUPLEX_HALF) + dev_err(pcs->dev, + "Adapter does not support Half Duplex\n"); + else + dev_err(pcs->dev, + "Adapter: Invalid Partner Speed and Duplex\n"); + + if (duplex == TSE_PCS_PARTNER_DUPLEX_FULL && + (speed == TSE_PCS_PARTNER_SPEED_10 || + speed == TSE_PCS_PARTNER_SPEED_100 || + speed == TSE_PCS_PARTNER_SPEED_1000)) + writew(SGMII_ADAPTER_ENABLE, + sgmii_adapter_base + SGMII_ADAPTER_CTRL_REG); + } else { + val = readw(tse_pcs_base + TSE_PCS_CONTROL_REG); + val |= TSE_PCS_CONTROL_RESTART_AN_MASK; + writew(val, tse_pcs_base + TSE_PCS_CONTROL_REG); + + tse_pcs_reset(tse_pcs_base, pcs); + mod_timer(&pcs->aneg_link_timer, jiffies + + msecs_to_jiffies(AUTONEGO_LINK_TIMER)); + } +} + +static void aneg_link_timer_callback(unsigned long data) +{ + struct tse_pcs *pcs = (struct tse_pcs *)data; + + if (pcs->autoneg == AUTONEG_ENABLE) + auto_nego_timer_callback(data); + else if (pcs->autoneg == AUTONEG_DISABLE) + pcs_link_timer_callback(data); +} + +void tse_pcs_fix_mac_speed(struct tse_pcs *pcs, struct phy_device *phy_dev, + unsigned int speed) +{ + void __iomem *tse_pcs_base = pcs->tse_pcs_base; + void __iomem *sgmii_adapter_base = pcs->sgmii_adapter_base; + u32 val; + + writew(SGMII_ADAPTER_ENABLE, + sgmii_adapter_base + SGMII_ADAPTER_CTRL_REG); + + pcs->autoneg = phy_dev->autoneg; + + if (phy_dev->autoneg == AUTONEG_ENABLE) { + val = readw(tse_pcs_base + TSE_PCS_CONTROL_REG); + val |= TSE_PCS_CONTROL_AN_EN_MASK; + writew(val, tse_pcs_base + TSE_PCS_CONTROL_REG); + + val = readw(tse_pcs_base + TSE_PCS_IF_MODE_REG); + val |= TSE_PCS_USE_SGMII_AN_MASK; + writew(val, tse_pcs_base + TSE_PCS_IF_MODE_REG); + + val = readw(tse_pcs_base + TSE_PCS_CONTROL_REG); + val |= TSE_PCS_CONTROL_RESTART_AN_MASK; + + tse_pcs_reset(tse_pcs_base, pcs); + + setup_timer(&pcs->aneg_link_timer, + aneg_link_timer_callback, (unsigned long)pcs); + mod_timer(&pcs->aneg_link_timer, jiffies + + msecs_to_jiffies(AUTONEGO_LINK_TIMER)); + } else if (phy_dev->autoneg == AUTONEG_DISABLE) { + val = readw(tse_pcs_base + TSE_PCS_CONTROL_REG); + val &= ~TSE_PCS_CONTROL_AN_EN_MASK; + writew(val, tse_pcs_base + TSE_PCS_CONTROL_REG); + + val = readw(tse_pcs_base + TSE_PCS_IF_MODE_REG); + val &= ~TSE_PCS_USE_SGMII_AN_MASK; + writew(val, tse_pcs_base + TSE_PCS_IF_MODE_REG); + + val = readw(tse_pcs_base + TSE_PCS_IF_MODE_REG); + val &= ~TSE_PCS_SGMII_SPEED_MASK; + + switch (speed) { + case 1000: + val |= TSE_PCS_SGMII_SPEED_1000; + break; + case 100: + val |= TSE_PCS_SGMII_SPEED_100; + break; + case 10: + val |= TSE_PCS_SGMII_SPEED_10; + break; + default: + return; + } + writew(val, tse_pcs_base + TSE_PCS_IF_MODE_REG); + + tse_pcs_reset(tse_pcs_base, pcs); + + setup_timer(&pcs->aneg_link_timer, + aneg_link_timer_callback, (unsigned long)pcs); + mod_timer(&pcs->aneg_link_timer, jiffies + + msecs_to_jiffies(AUTONEGO_LINK_TIMER)); + } +} diff --git a/drivers/net/ethernet/stmicro/stmmac/altr_tse_pcs.h b/drivers/net/ethernet/stmicro/stmmac/altr_tse_pcs.h new file mode 100644 index 0000000..2f58824 --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/altr_tse_pcs.h @@ -0,0 +1,36 @@ +/* Copyright Altera Corporation (C) 2016. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + * + * Author: Tien Hock Loh + */ + +#ifndef __TSE_PCS_H__ +#define __TSE_PCS_H__ + +#include +#include + +struct tse_pcs { + struct device *dev; + void __iomem *tse_pcs_base; + void __iomem *sgmii_adapter_base; + struct timer_list aneg_link_timer; + int autoneg; +}; + +int tse_pcs_init(void __iomem *base, struct tse_pcs *pcs); +void tse_pcs_fix_mac_speed(struct tse_pcs *pcs, struct phy_device *phy_dev, + unsigned int speed); + +#endif /* __TSE_PCS_H__ */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index f13499f..4bee2f9 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -27,6 +27,11 @@ #include "stmmac.h" #include "stmmac_platform.h" +#include "altr_tse_pcs.h" + +#define SGMII_ADAPTER_CTRL_REG 0x00 +#define SGMII_ADAPTER_DISABLE 0x0001 + #define SYSMGR_EMACGRP_CTRL_PHYSEL_ENUM_GMII_MII 0x0 #define SYSMGR_EMACGRP_CTRL_PHYSEL_ENUM_RGMII 0x1 #define SYSMGR_EMACGRP_CTRL_PHYSEL_ENUM_RMII 0x2 @@ -52,35 +57,46 @@ struct socfpga_dwmac { struct reset_control *stmmac_rst; void __iomem *splitter_base; bool f2h_ptp_ref_clk; + struct tse_pcs pcs; }; static void socfpga_dwmac_fix_mac_speed(void *priv, unsigned int speed) { struct socfpga_dwmac *dwmac = (struct socfpga_dwmac *)priv; void __iomem *splitter_base = dwmac->splitter_base; + void __iomem *tse_pcs_base = dwmac->pcs.tse_pcs_base; + void __iomem *sgmii_adapter_base = dwmac->pcs.sgmii_adapter_base; + struct device *dev = dwmac->dev; + struct net_device *ndev = dev_get_drvdata(dev); + struct phy_device *phy_dev = ndev->phydev; u32 val; - if (!splitter_base) - return; - - val = readl(splitter_base + EMAC_SPLITTER_CTRL_REG); - val &= ~EMAC_SPLITTER_CTRL_SPEED_MASK; - - switch (speed) { - case 1000: - val |= EMAC_SPLITTER_CTRL_SPEED_1000; - break; - case 100: - val |= EMAC_SPLITTER_CTRL_SPEED_100; - break; - case 10: - val |= EMAC_SPLITTER_CTRL_SPEED_10; - break; - default: - return; + if ((tse_pcs_base) && (sgmii_adapter_base)) + writew(SGMII_ADAPTER_DISABLE, + sgmii_adapter_base + SGMII_ADAPTER_CTRL_REG); + + if (splitter_base) { + val = readl(splitter_base + EMAC_SPLITTER_CTRL_REG); + val &= ~EMAC_SPLITTER_CTRL_SPEED_MASK; + + switch (speed) { + case 1000: + val |= EMAC_SPLITTER_CTRL_SPEED_1000; + break; + case 100: + val |= EMAC_SPLITTER_CTRL_SPEED_100; + break; + case 10: + val |= EMAC_SPLITTER_CTRL_SPEED_10; + break; + default: + return; + } + writel(val, splitter_base + EMAC_SPLITTER_CTRL_REG); } - writel(val, splitter_base + EMAC_SPLITTER_CTRL_REG); + if (tse_pcs_base && sgmii_adapter_base) + tse_pcs_fix_mac_speed(&dwmac->pcs, phy_dev, speed); } static int socfpga_dwmac_parse_data(struct socfpga_dwmac *dwmac, struct device *dev) @@ -88,9 +104,12 @@ static int socfpga_dwmac_parse_data(struct socfpga_dwmac *dwmac, struct device * struct device_node *np = dev->of_node; struct regmap *sys_mgr_base_addr; u32 reg_offset, reg_shift; - int ret; - struct device_node *np_splitter; + int ret, index; + struct device_node *np_splitter = NULL; + struct device_node *np_sgmii_adapter = NULL; struct resource res_splitter; + struct resource res_tse_pcs; + struct resource res_sgmii_adapter; dwmac->interface = of_get_phy_mode(np); @@ -128,6 +147,77 @@ static int socfpga_dwmac_parse_data(struct socfpga_dwmac *dwmac, struct device * } } + np_sgmii_adapter = of_parse_phandle(np, + "altr,gmii-to-sgmii-converter", 0); + if (np_sgmii_adapter) { + index = of_property_match_string(np_sgmii_adapter, "reg-names", + "hps_emac_interface_splitter_avalon_slave"); + + if (index >= 0) { + if (of_address_to_resource(np_sgmii_adapter, index, + &res_splitter)) { + dev_err(dev, + "%s: ERROR: missing emac splitter address\n", + __func__); + return -EINVAL; + } + + dwmac->splitter_base = + devm_ioremap_resource(dev, &res_splitter); + + if (IS_ERR(dwmac->splitter_base)) { + dev_err(dev, + "%s: ERROR: failed mapping emac splitter\n", + __func__); + return PTR_ERR(dwmac->splitter_base); + } + } + + index = of_property_match_string(np_sgmii_adapter, "reg-names", + "gmii_to_sgmii_adapter_avalon_slave"); + + if (index >= 0) { + if (of_address_to_resource(np_sgmii_adapter, index, + &res_sgmii_adapter)) { + dev_err(dev, + "%s: ERROR: failed mapping adapter\n", + __func__); + return -EINVAL; + } + + dwmac->pcs.sgmii_adapter_base = + devm_ioremap_resource(dev, &res_sgmii_adapter); + + if (IS_ERR(dwmac->pcs.sgmii_adapter_base)) { + dev_err(dev, "%s: failed to mapping adapter\n", + __func__); + return PTR_ERR(dwmac->pcs.sgmii_adapter_base); + } + } + + index = of_property_match_string(np_sgmii_adapter, "reg-names", + "eth_tse_control_port"); + + if (index >= 0) { + if (of_address_to_resource(np_sgmii_adapter, index, + &res_tse_pcs)) { + dev_err(dev, + "%s: ERROR: failed mapping tse control port\n", + __func__); + return -EINVAL; + } + + dwmac->pcs.tse_pcs_base = + devm_ioremap_resource(dev, &res_tse_pcs); + + if (IS_ERR(dwmac->pcs.tse_pcs_base)) { + dev_err(dev, + "%s: ERROR: failed mapping tse control port\n", + __func__); + return PTR_ERR(dwmac->pcs.sgmii_adapter_base); + } + } + } dwmac->reg_offset = reg_offset; dwmac->reg_shift = reg_shift; dwmac->sys_mgr_base_addr = sys_mgr_base_addr; @@ -151,6 +241,7 @@ static int socfpga_dwmac_set_phy_mode(struct socfpga_dwmac *dwmac) break; case PHY_INTERFACE_MODE_MII: case PHY_INTERFACE_MODE_GMII: + case PHY_INTERFACE_MODE_SGMII: val = SYSMGR_EMACGRP_CTRL_PHYSEL_ENUM_GMII_MII; break; default: @@ -191,6 +282,12 @@ static int socfpga_dwmac_set_phy_mode(struct socfpga_dwmac *dwmac) */ if (dwmac->stmmac_rst) reset_control_deassert(dwmac->stmmac_rst); + if (phymode == PHY_INTERFACE_MODE_SGMII) { + if (tse_pcs_init(dwmac->pcs.tse_pcs_base, &dwmac->pcs) != 0) { + dev_err(dwmac->dev, "Unable to initialize TSE PCS"); + return -EINVAL; + } + } return 0; } @@ -225,6 +322,7 @@ static int socfpga_dwmac_probe(struct platform_device *pdev) plat_dat->fix_mac_speed = socfpga_dwmac_fix_mac_speed; ret = stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res); + if (!ret) { struct net_device *ndev = platform_get_drvdata(pdev); struct stmmac_priv *stpriv = netdev_priv(ndev); -- cgit v0.10.2 From 64b87639c9cbeb03e26bc65528416c961b1dde96 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sun, 3 Jul 2016 13:18:43 +0800 Subject: netfilter: conntrack: fix race between nf_conntrack proc read and hash resize MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When we do "cat /proc/net/nf_conntrack", and meanwhile resize the conntrack hash table via /sys/module/nf_conntrack/parameters/hashsize, race will happen, because reader can observe a newly allocated hash but the old size (or vice versa). So oops will happen like follows： BUG: unable to handle kernel NULL pointer dereference at 0000000000000017 IP: [] seq_print_acct+0x11/0x50 [nf_conntrack] Call Trace: [] ? ct_seq_show+0x14e/0x340 [nf_conntrack] [] seq_read+0x2cc/0x390 [] proc_reg_read+0x42/0x70 [] __vfs_read+0x37/0x130 [] ? security_file_permission+0xa0/0xc0 [] vfs_read+0x95/0x140 [] SyS_read+0x55/0xc0 [] entry_SYSCALL_64_fastpath+0x1a/0xa4 It is very easy to reproduce this kernel crash. 1. open one shell and input the following cmds: while : ; do echo $RANDOM > /sys/module/nf_conntrack/parameters/hashsize done 2. open more shells and input the following cmds: while : ; do cat /proc/net/nf_conntrack done 3. just wait a monent, oops will happen soon. The solution in this patch is based on Florian's Commit 5e3c61f98175 ("netfilter: conntrack: fix lookup race during hash resize"). And add a wrapper function nf_conntrack_get_ht to get hash and hsize suggested by Florian Westphal. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index 3e2f332..79d7ac5 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -51,6 +51,8 @@ bool nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, const struct nf_conntrack_l3proto *l3proto, const struct nf_conntrack_l4proto *l4proto); +void nf_conntrack_get_ht(struct hlist_nulls_head **hash, unsigned int *hsize); + /* Find a connection corresponding to a tuple. */ struct nf_conntrack_tuple_hash * nf_conntrack_find_get(struct net *net, diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c index c6f3c40..6392371 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c @@ -26,6 +26,8 @@ struct ct_iter_state { struct seq_net_private p; + struct hlist_nulls_head *hash; + unsigned int htable_size; unsigned int bucket; }; @@ -35,10 +37,10 @@ static struct hlist_nulls_node *ct_get_first(struct seq_file *seq) struct hlist_nulls_node *n; for (st->bucket = 0; - st->bucket < nf_conntrack_htable_size; + st->bucket < st->htable_size; st->bucket++) { n = rcu_dereference( - hlist_nulls_first_rcu(&nf_conntrack_hash[st->bucket])); + hlist_nulls_first_rcu(&st->hash[st->bucket])); if (!is_a_nulls(n)) return n; } @@ -53,11 +55,11 @@ static struct hlist_nulls_node *ct_get_next(struct seq_file *seq, head = rcu_dereference(hlist_nulls_next_rcu(head)); while (is_a_nulls(head)) { if (likely(get_nulls_value(head) == st->bucket)) { - if (++st->bucket >= nf_conntrack_htable_size) + if (++st->bucket >= st->htable_size) return NULL; } head = rcu_dereference( - hlist_nulls_first_rcu(&nf_conntrack_hash[st->bucket])); + hlist_nulls_first_rcu(&st->hash[st->bucket])); } return head; } @@ -75,7 +77,11 @@ static struct hlist_nulls_node *ct_get_idx(struct seq_file *seq, loff_t pos) static void *ct_seq_start(struct seq_file *seq, loff_t *pos) __acquires(RCU) { + struct ct_iter_state *st = seq->private; + rcu_read_lock(); + + nf_conntrack_get_ht(&st->hash, &st->htable_size); return ct_get_idx(seq, *pos); } diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 153e33f..1289e7e 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -460,6 +460,23 @@ nf_ct_key_equal(struct nf_conntrack_tuple_hash *h, net_eq(net, nf_ct_net(ct)); } +/* must be called with rcu read lock held */ +void nf_conntrack_get_ht(struct hlist_nulls_head **hash, unsigned int *hsize) +{ + struct hlist_nulls_head *hptr; + unsigned int sequence, hsz; + + do { + sequence = read_seqcount_begin(&nf_conntrack_generation); + hsz = nf_conntrack_htable_size; + hptr = nf_conntrack_hash; + } while (read_seqcount_retry(&nf_conntrack_generation, sequence)); + + *hash = hptr; + *hsize = hsz; +} +EXPORT_SYMBOL_GPL(nf_conntrack_get_ht); + /* * Warning : * - Caller must take a reference on returned object diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 2aaa188..958a145 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -48,6 +48,8 @@ EXPORT_SYMBOL_GPL(print_tuple); struct ct_iter_state { struct seq_net_private p; + struct hlist_nulls_head *hash; + unsigned int htable_size; unsigned int bucket; u_int64_t time_now; }; @@ -58,9 +60,10 @@ static struct hlist_nulls_node *ct_get_first(struct seq_file *seq) struct hlist_nulls_node *n; for (st->bucket = 0; - st->bucket < nf_conntrack_htable_size; + st->bucket < st->htable_size; st->bucket++) { - n = rcu_dereference(hlist_nulls_first_rcu(&nf_conntrack_hash[st->bucket])); + n = rcu_dereference( + hlist_nulls_first_rcu(&st->hash[st->bucket])); if (!is_a_nulls(n)) return n; } @@ -75,12 +78,11 @@ static struct hlist_nulls_node *ct_get_next(struct seq_file *seq, head = rcu_dereference(hlist_nulls_next_rcu(head)); while (is_a_nulls(head)) { if (likely(get_nulls_value(head) == st->bucket)) { - if (++st->bucket >= nf_conntrack_htable_size) + if (++st->bucket >= st->htable_size) return NULL; } head = rcu_dereference( - hlist_nulls_first_rcu( - &nf_conntrack_hash[st->bucket])); + hlist_nulls_first_rcu(&st->hash[st->bucket])); } return head; } @@ -102,6 +104,8 @@ static void *ct_seq_start(struct seq_file *seq, loff_t *pos) st->time_now = ktime_get_real_ns(); rcu_read_lock(); + + nf_conntrack_get_ht(&st->hash, &st->htable_size); return ct_get_idx(seq, *pos); } -- cgit v0.10.2 From 474803d37e7fb6291d22cb964014afe457ba5212 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sun, 3 Jul 2016 13:18:44 +0800 Subject: netfilter: cttimeout: unlink timeout obj again when hash resize happen Imagine such situation, nf_conntrack_htable_size now is 4096, we are doing ctnl_untimeout, and iterate on 3000# bucket. Meanwhile, another user try to reduce hash size to 2048, then all nf_conn are removed to the new hashtable. When this hash resize operation finished, we still try to itreate ct begin from 3000# bucket, find nothing to do and just return. We may miss unlinking some timeout objects. And later we will end up with invalid references to timeout object that are already gone. So when we find that hash resize happened, try to unlink timeout objects from the 0# bucket again. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index 3c84f14..4cdcd96 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -303,16 +303,24 @@ static void ctnl_untimeout(struct net *net, struct ctnl_timeout *timeout) { struct nf_conntrack_tuple_hash *h; const struct hlist_nulls_node *nn; + unsigned int last_hsize; + spinlock_t *lock; int i; local_bh_disable(); - for (i = 0; i < nf_conntrack_htable_size; i++) { - nf_conntrack_lock(&nf_conntrack_locks[i % CONNTRACK_LOCKS]); - if (i < nf_conntrack_htable_size) { - hlist_nulls_for_each_entry(h, nn, &nf_conntrack_hash[i], hnnode) - untimeout(h, timeout); +restart: + last_hsize = nf_conntrack_htable_size; + for (i = 0; i < last_hsize; i++) { + lock = &nf_conntrack_locks[i % CONNTRACK_LOCKS]; + nf_conntrack_lock(lock); + if (last_hsize != nf_conntrack_htable_size) { + spin_unlock(lock); + goto restart; } - spin_unlock(&nf_conntrack_locks[i % CONNTRACK_LOCKS]); + + hlist_nulls_for_each_entry(h, nn, &nf_conntrack_hash[i], hnnode) + untimeout(h, timeout); + spin_unlock(lock); } local_bh_enable(); } -- cgit v0.10.2 From 8786a9716d028083f56f944996883f7d1a05919e Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sun, 3 Jul 2016 13:18:45 +0800 Subject: netfilter: nf_ct_helper: unlink helper again when hash resize happen From: Liping Zhang Similar to ctnl_untimeout, when hash resize happened, we should try to do unhelp from the 0# bucket again. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 3a1a88b..a4294e9 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -409,6 +409,8 @@ void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) struct nf_conntrack_expect *exp; const struct hlist_node *next; const struct hlist_nulls_node *nn; + unsigned int last_hsize; + spinlock_t *lock; struct net *net; unsigned int i; @@ -446,13 +448,18 @@ void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) rtnl_unlock(); local_bh_disable(); - for (i = 0; i < nf_conntrack_htable_size; i++) { - nf_conntrack_lock(&nf_conntrack_locks[i % CONNTRACK_LOCKS]); - if (i < nf_conntrack_htable_size) { - hlist_nulls_for_each_entry(h, nn, &nf_conntrack_hash[i], hnnode) - unhelp(h, me); +restart: + last_hsize = nf_conntrack_htable_size; + for (i = 0; i < last_hsize; i++) { + lock = &nf_conntrack_locks[i % CONNTRACK_LOCKS]; + nf_conntrack_lock(lock); + if (last_hsize != nf_conntrack_htable_size) { + spin_unlock(lock); + goto restart; } - spin_unlock(&nf_conntrack_locks[i % CONNTRACK_LOCKS]); + hlist_nulls_for_each_entry(h, nn, &nf_conntrack_hash[i], hnnode) + unhelp(h, me); + spin_unlock(lock); } local_bh_enable(); } -- cgit v0.10.2 From 242922a027176cd260c5adce4ba6bbfa3a05190c Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sun, 3 Jul 2016 20:44:01 +0200 Subject: netfilter: conntrack: simplify early_drop We don't need to acquire the bucket lock during early drop, we can use lockless traveral just like ____nf_conntrack_find. The timer deletion serves as synchronization point, if another cpu attempts to evict same entry, only one will succeed with timer deletion. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 5d3397f..2a5133e 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -301,6 +301,7 @@ void nf_ct_tmpl_free(struct nf_conn *tmpl); #define NF_CT_STAT_INC(net, count) __this_cpu_inc((net)->ct.stat->count) #define NF_CT_STAT_INC_ATOMIC(net, count) this_cpu_inc((net)->ct.stat->count) +#define NF_CT_STAT_ADD_ATOMIC(net, count, v) this_cpu_add((net)->ct.stat->count, (v)) #define MODULE_ALIAS_NFCT_HELPER(helper) \ MODULE_ALIAS("nfct-helper-" helper) diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 1289e7e..e0e9c9a 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -834,67 +834,66 @@ EXPORT_SYMBOL_GPL(nf_conntrack_tuple_taken); /* There's a small race here where we may free a just-assured connection. Too bad: we're in trouble anyway. */ -static noinline int early_drop(struct net *net, unsigned int _hash) +static unsigned int early_drop_list(struct net *net, + struct hlist_nulls_head *head) { - /* Use oldest entry, which is roughly LRU */ struct nf_conntrack_tuple_hash *h; - struct nf_conn *tmp; struct hlist_nulls_node *n; - unsigned int i, hash, sequence; - struct nf_conn *ct = NULL; - spinlock_t *lockp; - bool ret = false; + unsigned int drops = 0; + struct nf_conn *tmp; - i = 0; + hlist_nulls_for_each_entry_rcu(h, n, head, hnnode) { + tmp = nf_ct_tuplehash_to_ctrack(h); - local_bh_disable(); -restart: - sequence = read_seqcount_begin(&nf_conntrack_generation); - for (; i < NF_CT_EVICTION_RANGE; i++) { - hash = scale_hash(_hash++); - lockp = &nf_conntrack_locks[hash % CONNTRACK_LOCKS]; - nf_conntrack_lock(lockp); - if (read_seqcount_retry(&nf_conntrack_generation, sequence)) { - spin_unlock(lockp); - goto restart; - } - hlist_nulls_for_each_entry_rcu(h, n, &nf_conntrack_hash[hash], - hnnode) { - tmp = nf_ct_tuplehash_to_ctrack(h); - - if (test_bit(IPS_ASSURED_BIT, &tmp->status) || - !net_eq(nf_ct_net(tmp), net) || - nf_ct_is_dying(tmp)) - continue; - - if (atomic_inc_not_zero(&tmp->ct_general.use)) { - ct = tmp; - break; - } - } + if (test_bit(IPS_ASSURED_BIT, &tmp->status) || + !net_eq(nf_ct_net(tmp), net) || + nf_ct_is_dying(tmp)) + continue; - spin_unlock(lockp); - if (ct) - break; + if (!atomic_inc_not_zero(&tmp->ct_general.use)) + continue; + + /* kill only if still in same netns -- might have moved due to + * SLAB_DESTROY_BY_RCU rules. + * + * We steal the timer reference. If that fails timer has + * already fired or someone else deleted it. Just drop ref + * and move to next entry. + */ + if (net_eq(nf_ct_net(tmp), net) && + nf_ct_is_confirmed(tmp) && + del_timer(&tmp->timeout) && + nf_ct_delete(tmp, 0, 0)) + drops++; + + nf_ct_put(tmp); } - local_bh_enable(); + return drops; +} - if (!ct) - return false; +static noinline int early_drop(struct net *net, unsigned int _hash) +{ + unsigned int i; - /* kill only if in same netns -- might have moved due to - * SLAB_DESTROY_BY_RCU rules - */ - if (net_eq(nf_ct_net(ct), net) && del_timer(&ct->timeout)) { - if (nf_ct_delete(ct, 0, 0)) { - NF_CT_STAT_INC_ATOMIC(net, early_drop); - ret = true; + for (i = 0; i < NF_CT_EVICTION_RANGE; i++) { + struct hlist_nulls_head *ct_hash; + unsigned hash, sequence, drops; + + do { + sequence = read_seqcount_begin(&nf_conntrack_generation); + hash = scale_hash(_hash++); + ct_hash = nf_conntrack_hash; + } while (read_seqcount_retry(&nf_conntrack_generation, sequence)); + + drops = early_drop_list(net, &ct_hash[hash]); + if (drops) { + NF_CT_STAT_ADD_ATOMIC(net, early_drop, drops); + return true; } } - nf_ct_put(ct); - return ret; + return false; } static struct nf_conn * -- cgit v0.10.2 From 7c9664351980aaa6a4b8837a314360b3a4ad382a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 5 Jul 2016 12:07:23 +0200 Subject: netfilter: move nat hlist_head to nf_conn The nat extension structure is 32bytes in size on x86_64: struct nf_conn_nat { struct hlist_node bysource; /* 0 16 */ struct nf_conn * ct; /* 16 8 */ union nf_conntrack_nat_help help; /* 24 4 */ int masq_index; /* 28 4 */ /* size: 32, cachelines: 1, members: 4 */ /* last cacheline: 32 bytes */ }; The hlist is needed to quickly check for possible tuple collisions when installing a new nat binding. Storing this in the extension area has two drawbacks: 1. We need ct backpointer to get the conntrack struct from the extension. 2. When reallocation of extension area occurs we need to fixup the bysource hash head via hlist_replace_rcu. We can avoid both by placing the hlist_head in nf_conn and place nf_conn in the bysource hash rather than the extenstion. We can also remove the ->move support; no other extension needs it. Moving the entire nat extension into nf_conn would be possible as well but then we have to add yet another callback for deletion from the bysource hash table rather than just using nat extension ->destroy hook for this. nf_conn size doesn't increase due to aligment, followup patch replaces hlist_node with single pointer. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 2a5133e..e5135d8 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -117,6 +117,9 @@ struct nf_conn { /* Extensions */ struct nf_ct_ext *ext; +#if IS_ENABLED(CONFIG_NF_NAT) + struct hlist_node nat_bysource; +#endif /* Storage reserved for other modules, must be the last member */ union nf_conntrack_proto proto; }; diff --git a/include/net/netfilter/nf_conntrack_extend.h b/include/net/netfilter/nf_conntrack_extend.h index b925395..1c3035d 100644 --- a/include/net/netfilter/nf_conntrack_extend.h +++ b/include/net/netfilter/nf_conntrack_extend.h @@ -99,9 +99,6 @@ void *__nf_ct_ext_add_length(struct nf_conn *ct, enum nf_ct_ext_id id, struct nf_ct_ext_type { /* Destroys relationships (can be NULL). */ void (*destroy)(struct nf_conn *ct); - /* Called when realloacted (can be NULL). - Contents has already been moved. */ - void (*move)(void *new, void *old); enum nf_ct_ext_id id; diff --git a/include/net/netfilter/nf_nat.h b/include/net/netfilter/nf_nat.h index 344b1ab..02515f7 100644 --- a/include/net/netfilter/nf_nat.h +++ b/include/net/netfilter/nf_nat.h @@ -29,8 +29,6 @@ struct nf_conn; /* The structure embedded in the conntrack structure. */ struct nf_conn_nat { - struct hlist_node bysource; - struct nf_conn *ct; union nf_conntrack_nat_help help; #if IS_ENABLED(CONFIG_NF_NAT_MASQUERADE_IPV4) || \ IS_ENABLED(CONFIG_NF_NAT_MASQUERADE_IPV6) diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c index 1a95459..02bcf00 100644 --- a/net/netfilter/nf_conntrack_extend.c +++ b/net/netfilter/nf_conntrack_extend.c @@ -73,7 +73,7 @@ void *__nf_ct_ext_add_length(struct nf_conn *ct, enum nf_ct_ext_id id, size_t var_alloc_len, gfp_t gfp) { struct nf_ct_ext *old, *new; - int i, newlen, newoff; + int newlen, newoff; struct nf_ct_ext_type *t; /* Conntrack must not be confirmed to avoid races on reallocation. */ @@ -99,19 +99,8 @@ void *__nf_ct_ext_add_length(struct nf_conn *ct, enum nf_ct_ext_id id, return NULL; if (new != old) { - for (i = 0; i < NF_CT_EXT_NUM; i++) { - if (!__nf_ct_ext_exist(old, i)) - continue; - - rcu_read_lock(); - t = rcu_dereference(nf_ct_ext_types[i]); - if (t && t->move) - t->move((void *)new + new->offset[i], - (void *)old + old->offset[i]); - rcu_read_unlock(); - } kfree_rcu(old, rcu); - ct->ext = new; + rcu_assign_pointer(ct->ext, new); } new->offset[id] = newoff; diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c index 6877a39..6925347 100644 --- a/net/netfilter/nf_nat_core.c +++ b/net/netfilter/nf_nat_core.c @@ -198,11 +198,9 @@ find_appropriate_src(struct net *net, const struct nf_nat_range *range) { unsigned int h = hash_by_src(net, tuple); - const struct nf_conn_nat *nat; const struct nf_conn *ct; - hlist_for_each_entry_rcu(nat, &nf_nat_bysource[h], bysource) { - ct = nat->ct; + hlist_for_each_entry_rcu(ct, &nf_nat_bysource[h], nat_bysource) { if (same_src(ct, tuple) && net_eq(net, nf_ct_net(ct)) && nf_ct_zone_equal(ct, zone, IP_CT_DIR_ORIGINAL)) { @@ -435,8 +433,7 @@ nf_nat_setup_info(struct nf_conn *ct, spin_lock_bh(&nf_nat_lock); /* nf_conntrack_alter_reply might re-allocate extension aera */ nat = nfct_nat(ct); - nat->ct = ct; - hlist_add_head_rcu(&nat->bysource, + hlist_add_head_rcu(&ct->nat_bysource, &nf_nat_bysource[srchash]); spin_unlock_bh(&nf_nat_lock); } @@ -543,7 +540,7 @@ static int nf_nat_proto_clean(struct nf_conn *ct, void *data) if (nf_nat_proto_remove(ct, data)) return 1; - if (!nat || !nat->ct) + if (!nat) return 0; /* This netns is being destroyed, and conntrack has nat null binding. @@ -556,9 +553,8 @@ static int nf_nat_proto_clean(struct nf_conn *ct, void *data) return 1; spin_lock_bh(&nf_nat_lock); - hlist_del_rcu(&nat->bysource); + hlist_del_rcu(&ct->nat_bysource); ct->status &= ~IPS_NAT_DONE_MASK; - nat->ct = NULL; spin_unlock_bh(&nf_nat_lock); add_timer(&ct->timeout); @@ -688,27 +684,13 @@ static void nf_nat_cleanup_conntrack(struct nf_conn *ct) { struct nf_conn_nat *nat = nf_ct_ext_find(ct, NF_CT_EXT_NAT); - if (nat == NULL || nat->ct == NULL) + if (!nat) return; - NF_CT_ASSERT(nat->ct->status & IPS_SRC_NAT_DONE); - - spin_lock_bh(&nf_nat_lock); - hlist_del_rcu(&nat->bysource); - spin_unlock_bh(&nf_nat_lock); -} - -static void nf_nat_move_storage(void *new, void *old) -{ - struct nf_conn_nat *new_nat = new; - struct nf_conn_nat *old_nat = old; - struct nf_conn *ct = old_nat->ct; - - if (!ct || !(ct->status & IPS_SRC_NAT_DONE)) - return; + NF_CT_ASSERT(ct->status & IPS_SRC_NAT_DONE); spin_lock_bh(&nf_nat_lock); - hlist_replace_rcu(&old_nat->bysource, &new_nat->bysource); + hlist_del_rcu(&ct->nat_bysource); spin_unlock_bh(&nf_nat_lock); } @@ -716,7 +698,6 @@ static struct nf_ct_ext_type nat_extend __read_mostly = { .len = sizeof(struct nf_conn_nat), .align = __alignof__(struct nf_conn_nat), .destroy = nf_nat_cleanup_conntrack, - .move = nf_nat_move_storage, .id = NF_CT_EXT_NAT, .flags = NF_CT_EXT_F_PREALLOC, }; -- cgit v0.10.2 From 870190a9ec9075205c0fa795a09fa931694a3ff1 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 5 Jul 2016 12:07:24 +0200 Subject: netfilter: nat: convert nat bysrc hash to rhashtable It did use a fixed-size bucket list plus single lock to protect add/del. Unlike the main conntrack table we only need to add and remove keys. Convert it to rhashtable to get table autosizing and per-bucket locking. The maximum number of entries is -- as before -- tied to the number of conntracks so we do not need another upperlimit. The change does not handle rhashtable_remove_fast error, only possible "error" is -ENOENT, and that is something that can happen legitimetely, e.g. because nat module was inserted at a later time and no src manip took place yet. Tested with http-client-benchmark + httpterm with DNAT and SNAT rules in place. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index e5135d8..a08825b 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -17,6 +17,7 @@ #include #include #include +#include #include #include @@ -118,7 +119,7 @@ struct nf_conn { struct nf_ct_ext *ext; #if IS_ENABLED(CONFIG_NF_NAT) - struct hlist_node nat_bysource; + struct rhash_head nat_bysource; #endif /* Storage reserved for other modules, must be the last member */ union nf_conntrack_proto proto; diff --git a/include/net/netfilter/nf_nat.h b/include/net/netfilter/nf_nat.h index 02515f7..c327a43 100644 --- a/include/net/netfilter/nf_nat.h +++ b/include/net/netfilter/nf_nat.h @@ -1,5 +1,6 @@ #ifndef _NF_NAT_H #define _NF_NAT_H +#include #include #include #include diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c index 6925347..de31818 100644 --- a/net/netfilter/nf_nat_core.c +++ b/net/netfilter/nf_nat_core.c @@ -30,17 +30,19 @@ #include #include -static DEFINE_SPINLOCK(nf_nat_lock); - static DEFINE_MUTEX(nf_nat_proto_mutex); static const struct nf_nat_l3proto __rcu *nf_nat_l3protos[NFPROTO_NUMPROTO] __read_mostly; static const struct nf_nat_l4proto __rcu **nf_nat_l4protos[NFPROTO_NUMPROTO] __read_mostly; -static struct hlist_head *nf_nat_bysource __read_mostly; -static unsigned int nf_nat_htable_size __read_mostly; -static unsigned int nf_nat_hash_rnd __read_mostly; +struct nf_nat_conn_key { + const struct net *net; + const struct nf_conntrack_tuple *tuple; + const struct nf_conntrack_zone *zone; +}; + +static struct rhashtable nf_nat_bysource_table; inline const struct nf_nat_l3proto * __nf_nat_l3proto_find(u8 family) @@ -119,19 +121,17 @@ int nf_xfrm_me_harder(struct net *net, struct sk_buff *skb, unsigned int family) EXPORT_SYMBOL(nf_xfrm_me_harder); #endif /* CONFIG_XFRM */ -/* We keep an extra hash for each conntrack, for fast searching. */ -static inline unsigned int -hash_by_src(const struct net *n, const struct nf_conntrack_tuple *tuple) +static u32 nf_nat_bysource_hash(const void *data, u32 len, u32 seed) { - unsigned int hash; - - get_random_once(&nf_nat_hash_rnd, sizeof(nf_nat_hash_rnd)); + const struct nf_conntrack_tuple *t; + const struct nf_conn *ct = data; + t = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; /* Original src, to ensure we map it consistently if poss. */ - hash = jhash2((u32 *)&tuple->src, sizeof(tuple->src) / sizeof(u32), - tuple->dst.protonum ^ nf_nat_hash_rnd ^ net_hash_mix(n)); - return reciprocal_scale(hash, nf_nat_htable_size); + seed ^= net_hash_mix(nf_ct_net(ct)); + return jhash2((const u32 *)&t->src, sizeof(t->src) / sizeof(u32), + t->dst.protonum ^ seed); } /* Is this tuple already taken? (not by us) */ @@ -187,6 +187,26 @@ same_src(const struct nf_conn *ct, t->src.u.all == tuple->src.u.all); } +static int nf_nat_bysource_cmp(struct rhashtable_compare_arg *arg, + const void *obj) +{ + const struct nf_nat_conn_key *key = arg->key; + const struct nf_conn *ct = obj; + + return same_src(ct, key->tuple) && + net_eq(nf_ct_net(ct), key->net) && + nf_ct_zone_equal(ct, key->zone, IP_CT_DIR_ORIGINAL); +} + +static struct rhashtable_params nf_nat_bysource_params = { + .head_offset = offsetof(struct nf_conn, nat_bysource), + .obj_hashfn = nf_nat_bysource_hash, + .obj_cmpfn = nf_nat_bysource_cmp, + .nelem_hint = 256, + .min_size = 1024, + .nulls_base = (1U << RHT_BASE_SHIFT), +}; + /* Only called for SRC manip */ static int find_appropriate_src(struct net *net, @@ -197,23 +217,23 @@ find_appropriate_src(struct net *net, struct nf_conntrack_tuple *result, const struct nf_nat_range *range) { - unsigned int h = hash_by_src(net, tuple); const struct nf_conn *ct; + struct nf_nat_conn_key key = { + .net = net, + .tuple = tuple, + .zone = zone + }; - hlist_for_each_entry_rcu(ct, &nf_nat_bysource[h], nat_bysource) { - if (same_src(ct, tuple) && - net_eq(net, nf_ct_net(ct)) && - nf_ct_zone_equal(ct, zone, IP_CT_DIR_ORIGINAL)) { - /* Copy source part from reply tuple. */ - nf_ct_invert_tuplepr(result, - &ct->tuplehash[IP_CT_DIR_REPLY].tuple); - result->dst = tuple->dst; - - if (in_range(l3proto, l4proto, result, range)) - return 1; - } - } - return 0; + ct = rhashtable_lookup_fast(&nf_nat_bysource_table, &key, + nf_nat_bysource_params); + if (!ct) + return 0; + + nf_ct_invert_tuplepr(result, + &ct->tuplehash[IP_CT_DIR_REPLY].tuple); + result->dst = tuple->dst; + + return in_range(l3proto, l4proto, result, range); } /* For [FUTURE] fragmentation handling, we want the least-used @@ -385,7 +405,6 @@ nf_nat_setup_info(struct nf_conn *ct, const struct nf_nat_range *range, enum nf_nat_manip_type maniptype) { - struct net *net = nf_ct_net(ct); struct nf_conntrack_tuple curr_tuple, new_tuple; struct nf_conn_nat *nat; @@ -426,16 +445,13 @@ nf_nat_setup_info(struct nf_conn *ct, } if (maniptype == NF_NAT_MANIP_SRC) { - unsigned int srchash; - - srchash = hash_by_src(net, - &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); - spin_lock_bh(&nf_nat_lock); - /* nf_conntrack_alter_reply might re-allocate extension aera */ - nat = nfct_nat(ct); - hlist_add_head_rcu(&ct->nat_bysource, - &nf_nat_bysource[srchash]); - spin_unlock_bh(&nf_nat_lock); + int err; + + err = rhashtable_insert_fast(&nf_nat_bysource_table, + &ct->nat_bysource, + nf_nat_bysource_params); + if (err) + return NF_DROP; } /* It's done. */ @@ -552,10 +568,10 @@ static int nf_nat_proto_clean(struct nf_conn *ct, void *data) if (!del_timer(&ct->timeout)) return 1; - spin_lock_bh(&nf_nat_lock); - hlist_del_rcu(&ct->nat_bysource); ct->status &= ~IPS_NAT_DONE_MASK; - spin_unlock_bh(&nf_nat_lock); + + rhashtable_remove_fast(&nf_nat_bysource_table, &ct->nat_bysource, + nf_nat_bysource_params); add_timer(&ct->timeout); @@ -687,11 +703,8 @@ static void nf_nat_cleanup_conntrack(struct nf_conn *ct) if (!nat) return; - NF_CT_ASSERT(ct->status & IPS_SRC_NAT_DONE); - - spin_lock_bh(&nf_nat_lock); - hlist_del_rcu(&ct->nat_bysource); - spin_unlock_bh(&nf_nat_lock); + rhashtable_remove_fast(&nf_nat_bysource_table, &ct->nat_bysource, + nf_nat_bysource_params); } static struct nf_ct_ext_type nat_extend __read_mostly = { @@ -826,16 +839,13 @@ static int __init nf_nat_init(void) { int ret; - /* Leave them the same for the moment. */ - nf_nat_htable_size = nf_conntrack_htable_size; - - nf_nat_bysource = nf_ct_alloc_hashtable(&nf_nat_htable_size, 0); - if (!nf_nat_bysource) - return -ENOMEM; + ret = rhashtable_init(&nf_nat_bysource_table, &nf_nat_bysource_params); + if (ret) + return ret; ret = nf_ct_extend_register(&nat_extend); if (ret < 0) { - nf_ct_free_hashtable(nf_nat_bysource, nf_nat_htable_size); + rhashtable_destroy(&nf_nat_bysource_table); printk(KERN_ERR "nf_nat_core: Unable to register extension\n"); return ret; } @@ -859,7 +869,7 @@ static int __init nf_nat_init(void) return 0; cleanup_extend: - nf_ct_free_hashtable(nf_nat_bysource, nf_nat_htable_size); + rhashtable_destroy(&nf_nat_bysource_table); nf_ct_extend_unregister(&nat_extend); return ret; } @@ -877,8 +887,8 @@ static void __exit nf_nat_cleanup(void) #endif for (i = 0; i < NFPROTO_NUMPROTO; i++) kfree(nf_nat_l4protos[i]); - synchronize_net(); - nf_ct_free_hashtable(nf_nat_bysource, nf_nat_htable_size); + + rhashtable_destroy(&nf_nat_bysource_table); } MODULE_LICENSE("GPL"); -- cgit v0.10.2 From 47c74456257d2e50ace8c473d358cf4b9c0a912f Mon Sep 17 00:00:00 2001 From: Hangbin Liu Date: Tue, 5 Jul 2016 20:55:36 +0800 Subject: netfilter: physdev: physdev-is-out should not work with OUTPUT chain physdev_mt() will check skb->nf_bridge first, which was alloced in br_nf_pre_routing. So if we want to use --physdev-out and physdev-is-out, we need to match it in FORWARD or POSTROUTING chain. physdev_mt_check() only checked physdev-out and missed physdev-is-out. Fix it and update the debug message to make it clearer. Signed-off-by: Hangbin Liu Reviewed-by: Marcelo R Leitner Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/xt_physdev.c b/net/netfilter/xt_physdev.c index 1caaccb..e5f1898 100644 --- a/net/netfilter/xt_physdev.c +++ b/net/netfilter/xt_physdev.c @@ -102,14 +102,14 @@ static int physdev_mt_check(const struct xt_mtchk_param *par) if (!(info->bitmask & XT_PHYSDEV_OP_MASK) || info->bitmask & ~XT_PHYSDEV_OP_MASK) return -EINVAL; - if (info->bitmask & XT_PHYSDEV_OP_OUT && + if (info->bitmask & (XT_PHYSDEV_OP_OUT | XT_PHYSDEV_OP_ISOUT) && (!(info->bitmask & XT_PHYSDEV_OP_BRIDGED) || info->invert & XT_PHYSDEV_OP_BRIDGED) && par->hook_mask & ((1 << NF_INET_LOCAL_OUT) | (1 << NF_INET_FORWARD) | (1 << NF_INET_POST_ROUTING))) { - pr_info("using --physdev-out in the OUTPUT, FORWARD and " - "POSTROUTING chains for non-bridged traffic is not " - "supported anymore.\n"); + pr_info("using --physdev-out and --physdev-is-out are only" + "supported in the FORWARD and POSTROUTING chains with" + "bridged traffic.\n"); if (par->hook_mask & (1 << NF_INET_LOCAL_OUT)) return -EINVAL; } -- cgit v0.10.2 From 3f8b61b7f9aea414d162821817d89a7a6aae41c3 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Tue, 5 Jul 2016 23:23:00 +0800 Subject: netfilter: nft_ct: make byte/packet expr more friendly If we want to use ct packets expr, and add a rule like follows: # nft add rule filter input ct packets gt 1 counter We will find that no packets will hit it, because nf_conntrack_acct is disabled by default. So It will not work until we enable it manually via "echo 1 > /proc/sys/net/netfilter/nf_conntrack_acct". This is not friendly, so like xt_connbytes do, if the user want to use ct byte/packet expr, enable nf_conntrack_acct automatically. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index 137e308..7ce8fd7 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -355,6 +355,9 @@ static int nft_ct_get_init(const struct nft_ctx *ctx, if (err < 0) return err; + if (priv->key == NFT_CT_BYTES || priv->key == NFT_CT_PKTS) + nf_ct_set_acct(ctx->net, true); + return 0; } -- cgit v0.10.2 From d51ed8367bcbbb06f4f4986d1ef7dc2480bed1ad Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Fri, 8 Jul 2016 13:08:50 +0200 Subject: netfilter: constify arg to is_dying/confirmed Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index a08825b..1e04911 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -270,12 +270,12 @@ static inline int nf_ct_is_template(const struct nf_conn *ct) } /* It's confirmed if it is, or has been in the hash table. */ -static inline int nf_ct_is_confirmed(struct nf_conn *ct) +static inline int nf_ct_is_confirmed(const struct nf_conn *ct) { return test_bit(IPS_CONFIRMED_BIT, &ct->status); } -static inline int nf_ct_is_dying(struct nf_conn *ct) +static inline int nf_ct_is_dying(const struct nf_conn *ct) { return test_bit(IPS_DYING_BIT, &ct->status); } -- cgit v0.10.2 From 42a55769132fdf4f44bac1471b371d7f80bcde35 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 8 Jul 2016 14:41:49 +0200 Subject: netfilter: nf_tables: get rid of possible_net_t from set and basechain We can pass the netns pointer as parameter to the functions that need to gain access to it. From basechains, I didn't find any client for this field anymore so let's remove this too. Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 30c1d94..f2f1339 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -236,7 +236,8 @@ struct nft_expr; * @features: features supported by the implementation */ struct nft_set_ops { - bool (*lookup)(const struct nft_set *set, + bool (*lookup)(const struct net *net, + const struct nft_set *set, const u32 *key, const struct nft_set_ext **ext); bool (*update)(struct nft_set *set, @@ -248,11 +249,14 @@ struct nft_set_ops { struct nft_regs *regs, const struct nft_set_ext **ext); - int (*insert)(const struct nft_set *set, + int (*insert)(const struct net *net, + const struct nft_set *set, const struct nft_set_elem *elem); - void (*activate)(const struct nft_set *set, + void (*activate)(const struct net *net, + const struct nft_set *set, const struct nft_set_elem *elem); - void * (*deactivate)(const struct nft_set *set, + void * (*deactivate)(const struct net *net, + const struct nft_set *set, const struct nft_set_elem *elem); void (*remove)(const struct nft_set *set, const struct nft_set_elem *elem); @@ -295,7 +299,6 @@ void nft_unregister_set(struct nft_set_ops *ops); * @udlen: user data length * @udata: user data * @ops: set ops - * @pnet: network namespace * @flags: set flags * @genmask: generation mask * @klen: key length @@ -318,7 +321,6 @@ struct nft_set { unsigned char *udata; /* runtime data below here */ const struct nft_set_ops *ops ____cacheline_aligned; - possible_net_t pnet; u16 flags:14, genmask:2; u8 klen; @@ -804,7 +806,6 @@ struct nft_stats { * struct nft_base_chain - nf_tables base chain * * @ops: netfilter hook ops - * @pnet: net namespace that this chain belongs to * @type: chain type * @policy: default policy * @stats: per-cpu chain stats @@ -813,7 +814,6 @@ struct nft_stats { */ struct nft_base_chain { struct nf_hook_ops ops[NFT_HOOK_OPS_MAX]; - possible_net_t pnet; const struct nf_chain_type *type; u8 policy; u8 flags; @@ -1009,10 +1009,11 @@ static inline bool nft_set_elem_active(const struct nft_set_ext *ext, return !(ext->genmask & genmask); } -static inline void nft_set_elem_change_active(const struct nft_set *set, +static inline void nft_set_elem_change_active(const struct net *net, + const struct nft_set *set, struct nft_set_ext *ext) { - ext->genmask ^= nft_genmask_next(read_pnet(&set->pnet)); + ext->genmask ^= nft_genmask_next(net); } /* diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 18b7f85..0211eae 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1405,7 +1405,6 @@ static int nf_tables_newchain(struct net *net, struct sock *nlsk, rcu_assign_pointer(basechain->stats, stats); } - write_pnet(&basechain->pnet, net); basechain->type = type; chain = &basechain->chain; @@ -2841,7 +2840,6 @@ static int nf_tables_newset(struct net *net, struct sock *nlsk, } INIT_LIST_HEAD(&set->bindings); - write_pnet(&set->pnet, net); set->ops = ops; set->ktype = ktype; set->klen = desc.klen; @@ -3520,7 +3518,7 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set, goto err4; ext->genmask = nft_genmask_cur(ctx->net) | NFT_SET_ELEM_BUSY_MASK; - err = set->ops->insert(set, &elem); + err = set->ops->insert(ctx->net, set, &elem); if (err < 0) goto err5; @@ -3644,7 +3642,7 @@ static int nft_del_setelem(struct nft_ctx *ctx, struct nft_set *set, goto err3; } - priv = set->ops->deactivate(set, &elem); + priv = set->ops->deactivate(ctx->net, set, &elem); if (priv == NULL) { err = -ENOENT; goto err4; @@ -4018,7 +4016,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) case NFT_MSG_NEWSETELEM: te = (struct nft_trans_elem *)trans->data; - te->set->ops->activate(te->set, &te->elem); + te->set->ops->activate(net, te->set, &te->elem); nf_tables_setelem_notify(&trans->ctx, te->set, &te->elem, NFT_MSG_NEWSETELEM, 0); @@ -4143,7 +4141,7 @@ static int nf_tables_abort(struct net *net, struct sk_buff *skb) case NFT_MSG_DELSETELEM: te = (struct nft_trans_elem *)trans->data; - te->set->ops->activate(te->set, &te->elem); + te->set->ops->activate(net, te->set, &te->elem); te->set->ndeact--; nft_trans_destroy(trans); diff --git a/net/netfilter/nft_hash.c b/net/netfilter/nft_hash.c index ea92481..564fa79 100644 --- a/net/netfilter/nft_hash.c +++ b/net/netfilter/nft_hash.c @@ -71,13 +71,13 @@ static inline int nft_hash_cmp(struct rhashtable_compare_arg *arg, return 0; } -static bool nft_hash_lookup(const struct nft_set *set, const u32 *key, - const struct nft_set_ext **ext) +static bool nft_hash_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext) { struct nft_hash *priv = nft_set_priv(set); const struct nft_hash_elem *he; struct nft_hash_cmp_arg arg = { - .genmask = nft_genmask_cur(read_pnet(&set->pnet)), + .genmask = nft_genmask_cur(net), .set = set, .key = key, }; @@ -125,13 +125,13 @@ err1: return false; } -static int nft_hash_insert(const struct nft_set *set, +static int nft_hash_insert(const struct net *net, const struct nft_set *set, const struct nft_set_elem *elem) { struct nft_hash *priv = nft_set_priv(set); struct nft_hash_elem *he = elem->priv; struct nft_hash_cmp_arg arg = { - .genmask = nft_genmask_next(read_pnet(&set->pnet)), + .genmask = nft_genmask_next(net), .set = set, .key = elem->key.val.data, }; @@ -140,20 +140,20 @@ static int nft_hash_insert(const struct nft_set *set, nft_hash_params); } -static void nft_hash_activate(const struct nft_set *set, +static void nft_hash_activate(const struct net *net, const struct nft_set *set, const struct nft_set_elem *elem) { struct nft_hash_elem *he = elem->priv; - nft_set_elem_change_active(set, &he->ext); + nft_set_elem_change_active(net, set, &he->ext); nft_set_elem_clear_busy(&he->ext); } -static void *nft_hash_deactivate(const struct nft_set *set, +static void *nft_hash_deactivate(const struct net *net, + const struct nft_set *set, const struct nft_set_elem *elem) { struct nft_hash *priv = nft_set_priv(set); - struct net *net = read_pnet(&set->pnet); struct nft_hash_elem *he; struct nft_hash_cmp_arg arg = { .genmask = nft_genmask_next(net), @@ -166,7 +166,7 @@ static void *nft_hash_deactivate(const struct nft_set *set, if (he != NULL) { if (!nft_set_elem_mark_busy(&he->ext) || !nft_is_active(net, &he->ext)) - nft_set_elem_change_active(set, &he->ext); + nft_set_elem_change_active(net, set, &he->ext); else he = NULL; } diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index b8d18f5..e164325 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -35,7 +35,7 @@ static void nft_lookup_eval(const struct nft_expr *expr, const struct nft_set_ext *ext; bool found; - found = set->ops->lookup(set, ®s->data[priv->sreg], &ext) ^ + found = set->ops->lookup(pkt->net, set, ®s->data[priv->sreg], &ext) ^ priv->invert; if (!found) { diff --git a/net/netfilter/nft_rbtree.c b/net/netfilter/nft_rbtree.c index c0f6387..6473936 100644 --- a/net/netfilter/nft_rbtree.c +++ b/net/netfilter/nft_rbtree.c @@ -41,13 +41,13 @@ static bool nft_rbtree_equal(const struct nft_set *set, const void *this, return memcmp(this, nft_set_ext_key(&interval->ext), set->klen) == 0; } -static bool nft_rbtree_lookup(const struct nft_set *set, const u32 *key, - const struct nft_set_ext **ext) +static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set, + const u32 *key, const struct nft_set_ext **ext) { const struct nft_rbtree *priv = nft_set_priv(set); const struct nft_rbtree_elem *rbe, *interval = NULL; + u8 genmask = nft_genmask_cur(net); const struct rb_node *parent; - u8 genmask = nft_genmask_cur(read_pnet(&set->pnet)); const void *this; int d; @@ -93,13 +93,13 @@ out: return false; } -static int __nft_rbtree_insert(const struct nft_set *set, +static int __nft_rbtree_insert(const struct net *net, const struct nft_set *set, struct nft_rbtree_elem *new) { struct nft_rbtree *priv = nft_set_priv(set); + u8 genmask = nft_genmask_next(net); struct nft_rbtree_elem *rbe; struct rb_node *parent, **p; - u8 genmask = nft_genmask_next(read_pnet(&set->pnet)); int d; parent = NULL; @@ -132,14 +132,14 @@ static int __nft_rbtree_insert(const struct nft_set *set, return 0; } -static int nft_rbtree_insert(const struct nft_set *set, +static int nft_rbtree_insert(const struct net *net, const struct nft_set *set, const struct nft_set_elem *elem) { struct nft_rbtree_elem *rbe = elem->priv; int err; spin_lock_bh(&nft_rbtree_lock); - err = __nft_rbtree_insert(set, rbe); + err = __nft_rbtree_insert(net, set, rbe); spin_unlock_bh(&nft_rbtree_lock); return err; @@ -156,21 +156,23 @@ static void nft_rbtree_remove(const struct nft_set *set, spin_unlock_bh(&nft_rbtree_lock); } -static void nft_rbtree_activate(const struct nft_set *set, +static void nft_rbtree_activate(const struct net *net, + const struct nft_set *set, const struct nft_set_elem *elem) { struct nft_rbtree_elem *rbe = elem->priv; - nft_set_elem_change_active(set, &rbe->ext); + nft_set_elem_change_active(net, set, &rbe->ext); } -static void *nft_rbtree_deactivate(const struct nft_set *set, +static void *nft_rbtree_deactivate(const struct net *net, + const struct nft_set *set, const struct nft_set_elem *elem) { const struct nft_rbtree *priv = nft_set_priv(set); const struct rb_node *parent = priv->root.rb_node; struct nft_rbtree_elem *rbe, *this = elem->priv; - u8 genmask = nft_genmask_next(read_pnet(&set->pnet)); + u8 genmask = nft_genmask_next(net); int d; while (parent != NULL) { @@ -196,7 +198,7 @@ static void *nft_rbtree_deactivate(const struct nft_set *set, parent = parent->rb_right; continue; } - nft_set_elem_change_active(set, &rbe->ext); + nft_set_elem_change_active(net, set, &rbe->ext); return rbe; } } -- cgit v0.10.2 From c2b9b4fee8ab86f2bb657e5ac48d803879e92765 Mon Sep 17 00:00:00 2001 From: Toby DiPasquale Date: Mon, 11 Jul 2016 11:32:45 +0100 Subject: netfilter: nf_conntrack_h323: fix off-by-one in DecodeQ931 This patch corrects an off-by-one error in the DecodeQ931 function in the nf_conntrack_h323 module. This error could result in reading off the end of a Q.931 frame. Signed-off-by: Toby DiPasquale Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_conntrack_h323_asn1.c b/net/netfilter/nf_conntrack_h323_asn1.c index bcd5ed6..89b2e46 100644 --- a/net/netfilter/nf_conntrack_h323_asn1.c +++ b/net/netfilter/nf_conntrack_h323_asn1.c @@ -846,9 +846,10 @@ int DecodeQ931(unsigned char *buf, size_t sz, Q931 *q931) sz -= len; /* Message Type */ - if (sz < 1) + if (sz < 2) return H323_ERROR_BOUND; q931->MessageType = *p++; + sz--; PRINT("MessageType = %02X\n", q931->MessageType); if (*p & 0x80) { p++; -- cgit v0.10.2 From ad38844b8e00c8d33eb1d1ffb49b945e6a985e13 Mon Sep 17 00:00:00 2001 From: Luciano Coelho Date: Mon, 11 Jul 2016 09:19:51 +0300 Subject: iwlwifi: mvm: bump MAX firmware API for mvm devices The driver is now ready to load the new firmware versions. Signed-off-by: Luca Coelho diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-7000.c b/drivers/net/wireless/intel/iwlwifi/iwl-7000.c index f4d9215..64690c14 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-7000.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-7000.c @@ -73,8 +73,8 @@ /* Highest firmware API version supported */ #define IWL7260_UCODE_API_MAX 17 #define IWL7265_UCODE_API_MAX 17 -#define IWL7265D_UCODE_API_MAX 21 -#define IWL3168_UCODE_API_MAX 21 +#define IWL7265D_UCODE_API_MAX 24 +#define IWL3168_UCODE_API_MAX 24 /* Lowest firmware API version supported */ #define IWL7260_UCODE_API_MIN 16 diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-8000.c b/drivers/net/wireless/intel/iwlwifi/iwl-8000.c index 8bf11c9..6c6725e 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-8000.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-8000.c @@ -70,8 +70,8 @@ #include "iwl-agn-hw.h" /* Highest firmware API version supported */ -#define IWL8000_UCODE_API_MAX 21 -#define IWL8265_UCODE_API_MAX 21 +#define IWL8000_UCODE_API_MAX 24 +#define IWL8265_UCODE_API_MAX 24 /* Lowest firmware API version supported */ #define IWL8000_UCODE_API_MIN 16 diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c index 5c1e71f..fbaf705 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-9000.c +++ b/drivers/net/wireless/intel/iwlwifi/iwl-9000.c @@ -55,7 +55,7 @@ #include "iwl-agn-hw.h" /* Highest firmware API version supported */ -#define IWL9000_UCODE_API_MAX 21 +#define IWL9000_UCODE_API_MAX 24 /* Lowest firmware API version supported */ #define IWL9000_UCODE_API_MIN 16 -- cgit v0.10.2 From fefae6909ead1798c39bee4d94e7e8f1f2752ef6 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 8 Jul 2016 11:39:12 -0700 Subject: net: dsa: b53: Allow SRAB driver to specify platform data For Northstart Plus SoCs, we cannot detect the switch because only the revision information is provied in the Management page, instead, rely on Device Tree to tell us the chip id, and pass it down using the b53_platform_data structure. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/b53/b53_srab.c b/drivers/net/dsa/b53/b53_srab.c index 70fd472..de2b9e7 100644 --- a/drivers/net/dsa/b53/b53_srab.c +++ b/drivers/net/dsa/b53/b53_srab.c @@ -21,6 +21,7 @@ #include #include #include +#include #include "b53_priv.h" @@ -356,12 +357,37 @@ static struct b53_io_ops b53_srab_ops = { .write64 = b53_srab_write64, }; +static const struct of_device_id b53_srab_of_match[] = { + { .compatible = "brcm,bcm53010-srab" }, + { .compatible = "brcm,bcm53011-srab" }, + { .compatible = "brcm,bcm53012-srab" }, + { .compatible = "brcm,bcm53018-srab" }, + { .compatible = "brcm,bcm53019-srab" }, + { .compatible = "brcm,bcm5301x-srab" }, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, b53_srab_of_match); + static int b53_srab_probe(struct platform_device *pdev) { + struct b53_platform_data *pdata = pdev->dev.platform_data; + struct device_node *dn = pdev->dev.of_node; + const struct of_device_id *of_id = NULL; struct b53_srab_priv *priv; struct b53_device *dev; struct resource *r; + if (dn) + of_id = of_match_node(b53_srab_of_match, dn); + + if (of_id) { + pdata = devm_kzalloc(&pdev->dev, sizeof(*pdata), GFP_KERNEL); + if (!pdata) + return -ENOMEM; + + pdata->chip_id = (u32)of_id->data; + } + priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL); if (!priv) return -ENOMEM; @@ -375,6 +401,9 @@ static int b53_srab_probe(struct platform_device *pdev) if (!dev) return -ENOMEM; + if (pdata) + dev->pdata = pdata; + platform_set_drvdata(pdev, dev); return b53_switch_register(dev); @@ -390,16 +419,6 @@ static int b53_srab_remove(struct platform_device *pdev) return 0; } -static const struct of_device_id b53_srab_of_match[] = { - { .compatible = "brcm,bcm53010-srab" }, - { .compatible = "brcm,bcm53011-srab" }, - { .compatible = "brcm,bcm53012-srab" }, - { .compatible = "brcm,bcm53018-srab" }, - { .compatible = "brcm,bcm53019-srab" }, - { .compatible = "brcm,bcm5301x-srab" }, - { /* sentinel */ }, -}; - static struct platform_driver b53_srab_driver = { .probe = b53_srab_probe, .remove = b53_srab_remove, -- cgit v0.10.2 From 991a36bb464589f169d6c9e3d01f5b99299a73fa Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 8 Jul 2016 11:39:13 -0700 Subject: net: dsa: b53: Add support for BCM585xx/586xx/88312 integrated switch Update the SRAB, core driver and binding document to support the BCM585xx/586xx/88312 integrated switch (Northstar Plus SoCs family). Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/dsa/b53.txt b/Documentation/devicetree/bindings/net/dsa/b53.txt index ca752db..d6c6e41 100644 --- a/Documentation/devicetree/bindings/net/dsa/b53.txt +++ b/Documentation/devicetree/bindings/net/dsa/b53.txt @@ -20,6 +20,15 @@ Required properties: "brcm,bcm53018-srab" "brcm,bcm53019-srab" and the mandatory "brcm,bcm5301x-srab" string + For the BCM585xx/586XX/88312 SoCs with an integrated switch, must be one of: + "brcm,bcm58522-srab" + "brcm,bcm58523-srab" + "brcm,bcm58525-srab" + "brcm,bcm58622-srab" + "brcm,bcm58623-srab" + "brcm,bcm58625-srab" + "brcm,bcm88312-srab" and the mandatory "brcm,nsp-srab string + For the BCM63xx/33xx SoCs with an integrated switch, must be one of: "brcm,bcm3384-switch" "brcm,bcm6328-switch" diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 444de66..bda37d3 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1581,6 +1581,18 @@ static const struct b53_chip_data b53_switch_chips[] = { .jumbo_pm_reg = B53_JUMBO_PORT_MASK, .jumbo_size_reg = B53_JUMBO_MAX_SIZE, }, + { + .chip_id = BCM58XX_DEVICE_ID, + .dev_name = "BCM585xx/586xx/88312", + .vlans = 4096, + .enabled_ports = 0x1ff, + .arl_entries = 4, + .cpu_port = B53_CPU_PORT_25, + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, }; static int b53_switch_init(struct b53_device *dev) diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h index 5d8c602..835a744 100644 --- a/drivers/net/dsa/b53/b53_priv.h +++ b/drivers/net/dsa/b53/b53_priv.h @@ -59,6 +59,7 @@ enum { BCM53012_DEVICE_ID = 0x53012, BCM53018_DEVICE_ID = 0x53018, BCM53019_DEVICE_ID = 0x53019, + BCM58XX_DEVICE_ID = 0x5800, }; #define B53_N_PORTS 9 diff --git a/drivers/net/dsa/b53/b53_srab.c b/drivers/net/dsa/b53/b53_srab.c index de2b9e7..2b304ea 100644 --- a/drivers/net/dsa/b53/b53_srab.c +++ b/drivers/net/dsa/b53/b53_srab.c @@ -364,6 +364,14 @@ static const struct of_device_id b53_srab_of_match[] = { { .compatible = "brcm,bcm53018-srab" }, { .compatible = "brcm,bcm53019-srab" }, { .compatible = "brcm,bcm5301x-srab" }, + { .compatible = "brcm,bcm58522-srab", .data = (void *)BCM58XX_DEVICE_ID }, + { .compatible = "brcm,bcm58525-srab", .data = (void *)BCM58XX_DEVICE_ID }, + { .compatible = "brcm,bcm58535-srab", .data = (void *)BCM58XX_DEVICE_ID }, + { .compatible = "brcm,bcm58622-srab", .data = (void *)BCM58XX_DEVICE_ID }, + { .compatible = "brcm,bcm58623-srab", .data = (void *)BCM58XX_DEVICE_ID }, + { .compatible = "brcm,bcm58625-srab", .data = (void *)BCM58XX_DEVICE_ID }, + { .compatible = "brcm,bcm88312-srab", .data = (void *)BCM58XX_DEVICE_ID }, + { .compatible = "brcm,nsp-srab", .data = (void *)BCM58XX_DEVICE_ID }, { /* sentinel */ }, }; MODULE_DEVICE_TABLE(of, b53_srab_of_match); -- cgit v0.10.2 From bac65c4b39ca1fe13b3a82fb49b71fc351305464 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sat, 9 Jul 2016 00:54:47 +0200 Subject: Revert "net: ethernet: bcmgenet: use phy_ethtool_{get|set}_link_ksettings" This reverts commit 4386f5662e63 ("net: ethernet: bcmgenet: use phy_ethtool_{get|set}_link_ksettings") This patch is wrong, the function phy_ethtool_{get|set}_link_ksettings don't check if the device is running, but the driver bcmgenet need this check. The function {get|set}_settings need to access the mdio bus, and this bus may only be used when the device is running. Otherwise, the clock is disable and a mdio access will fail. Signed-off-by: Philippe Reynes Acked-by: Florian Fainelli Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 76ed6df..8d4f849 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -450,6 +450,30 @@ static inline void bcmgenet_rdma_ring_writel(struct bcmgenet_priv *priv, genet_dma_ring_regs[r]); } +static int bcmgenet_get_settings(struct net_device *dev, + struct ethtool_cmd *cmd) +{ + if (!netif_running(dev)) + return -EINVAL; + + if (!dev->phydev) + return -ENODEV; + + return phy_ethtool_gset(dev->phydev, cmd); +} + +static int bcmgenet_set_settings(struct net_device *dev, + struct ethtool_cmd *cmd) +{ + if (!netif_running(dev)) + return -EINVAL; + + if (!dev->phydev) + return -ENODEV; + + return phy_ethtool_sset(dev->phydev, cmd); +} + static int bcmgenet_set_rx_csum(struct net_device *dev, netdev_features_t wanted) { @@ -953,6 +977,8 @@ static struct ethtool_ops bcmgenet_ethtool_ops = { .get_strings = bcmgenet_get_strings, .get_sset_count = bcmgenet_get_sset_count, .get_ethtool_stats = bcmgenet_get_ethtool_stats, + .get_settings = bcmgenet_get_settings, + .set_settings = bcmgenet_set_settings, .get_drvinfo = bcmgenet_get_drvinfo, .get_link = ethtool_op_get_link, .get_msglevel = bcmgenet_get_msglevel, @@ -964,8 +990,6 @@ static struct ethtool_ops bcmgenet_ethtool_ops = { .nway_reset = bcmgenet_nway_reset, .get_coalesce = bcmgenet_get_coalesce, .set_coalesce = bcmgenet_set_coalesce, - .get_link_ksettings = phy_ethtool_get_link_ksettings, - .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /* Power down the unimac, based on mode. */ -- cgit v0.10.2 From 31eff2670301d0367ce06645e41c3b57db4e9e78 Mon Sep 17 00:00:00 2001 From: Loic Poulain Date: Mon, 11 Jul 2016 21:55:36 +0200 Subject: Bluetooth: hci_intel: Remove LPM enabling from setup procedure LPM can be enabled via a DDC write command at specific DDC ID. As any other DDC value, this is up to the DDC config file to include (or not) the low power mode configuration. Signed-off-by: Loic Poulain Signed-off-by: Marcel Holtmann diff --git a/drivers/bluetooth/hci_intel.c b/drivers/bluetooth/hci_intel.c index f6f2b01..ed0a420 100644 --- a/drivers/bluetooth/hci_intel.c +++ b/drivers/bluetooth/hci_intel.c @@ -537,9 +537,7 @@ static int intel_setup(struct hci_uart *hu) { static const u8 reset_param[] = { 0x00, 0x01, 0x00, 0x01, 0x00, 0x08, 0x04, 0x00 }; - static const u8 lpm_param[] = { 0x03, 0x07, 0x01, 0x0b }; struct intel_data *intel = hu->priv; - struct intel_device *idev = NULL; struct hci_dev *hdev = hu->hdev; struct sk_buff *skb; struct intel_version ver; @@ -884,35 +882,23 @@ done: bt_dev_info(hdev, "Device booted in %llu usecs", duration); - /* Enable LPM if matching pdev with wakeup enabled */ + /* Enable LPM if matching pdev with wakeup enabled, set TX active + * until further LPM TX notification. + */ mutex_lock(&intel_device_list_lock); list_for_each(p, &intel_device_list) { struct intel_device *dev = list_entry(p, struct intel_device, list); if (hu->tty->dev->parent == dev->pdev->dev.parent) { - if (device_may_wakeup(&dev->pdev->dev)) - idev = dev; + if (device_may_wakeup(&dev->pdev->dev)) { + set_bit(STATE_LPM_ENABLED, &intel->flags); + set_bit(STATE_TX_ACTIVE, &intel->flags); + } break; } } mutex_unlock(&intel_device_list_lock); - if (!idev) - goto no_lpm; - - bt_dev_info(hdev, "Enabling LPM"); - - skb = __hci_cmd_sync(hdev, 0xfc8b, sizeof(lpm_param), lpm_param, - HCI_CMD_TIMEOUT); - if (IS_ERR(skb)) { - bt_dev_err(hdev, "Failed to enable LPM"); - goto no_lpm; - } - kfree_skb(skb); - - set_bit(STATE_LPM_ENABLED, &intel->flags); - -no_lpm: /* Ignore errors, device can work without DDC parameters */ btintel_load_ddc_config(hdev, fwname); -- cgit v0.10.2 From 28aa4c26fce2202db8d42ae76b639ca1d9a23d25 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 9 Jul 2016 19:47:40 +0800 Subject: sctp: add SCTP_PR_SUPPORTED on sctp sockopt According to section 4.5 of rfc7496, prsctp_enable should be per asoc. We will add prsctp_enable to both asoc and ep, and replace the places where it used net.sctp->prsctp_enable with asoc->prsctp_enable. ep->prsctp_enable will be initialized with net.sctp->prsctp_enable, and asoc->prsctp_enable will be initialized with ep->prsctp_enable. We can also modify it's value through sockopt SCTP_PR_SUPPORTED. Signed-off-by: Xin Long Signed-off-by: David S. Miller diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 83c5ec5..07115ca 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1256,7 +1256,8 @@ struct sctp_endpoint { /* SCTP-AUTH: endpoint shared keys */ struct list_head endpoint_shared_keys; __u16 active_key_id; - __u8 auth_enable; + __u8 auth_enable:1, + prsctp_enable:1; }; /* Recover the outter endpoint structure. */ @@ -1848,7 +1849,8 @@ struct sctp_association { __u16 active_key_id; __u8 need_ecne:1, /* Need to send an ECNE Chunk? */ - temp:1; /* Is it a temporary association? */ + temp:1, /* Is it a temporary association? */ + prsctp_enable:1; struct sctp_priv_assoc_stats stats; }; diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h index ce70fe6..aa08906 100644 --- a/include/uapi/linux/sctp.h +++ b/include/uapi/linux/sctp.h @@ -112,6 +112,7 @@ typedef __s32 sctp_assoc_t; #define SCTP_SOCKOPT_CONNECTX 110 /* CONNECTX requests. */ #define SCTP_SOCKOPT_CONNECTX3 111 /* CONNECTX requests (updated) */ #define SCTP_GET_ASSOC_STATS 112 /* Read only */ +#define SCTP_PR_SUPPORTED 113 /* These are bit fields for msghdr->msg_flags. See section 5.1. */ /* On user space Linux, these live in as an enum. */ diff --git a/net/sctp/associola.c b/net/sctp/associola.c index e1849f3..1c23060 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -268,6 +268,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a goto fail_init; asoc->active_key_id = ep->active_key_id; + asoc->prsctp_enable = ep->prsctp_enable; /* Save the hmacs and chunks list into this association */ if (ep->auth_hmacs_list) diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 9d494e3..1f03065 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -163,6 +163,7 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, */ ep->auth_hmacs_list = auth_hmacs; ep->auth_chunk_list = auth_chunks; + ep->prsctp_enable = net->sctp.prsctp_enable; return ep; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 56f364d..0e3045e 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -261,7 +261,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, chunksize += WORD_ROUND(SCTP_SAT_LEN(num_types)); chunksize += sizeof(ecap_param); - if (net->sctp.prsctp_enable) + if (asoc->prsctp_enable) chunksize += sizeof(prsctp_param); /* ADDIP: Section 4.2.7: @@ -355,7 +355,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, sctp_addto_param(retval, num_ext, extensions); } - if (net->sctp.prsctp_enable) + if (asoc->prsctp_enable) sctp_addto_chunk(retval, sizeof(prsctp_param), &prsctp_param); if (sp->adaptation_ind) { @@ -2024,8 +2024,8 @@ static void sctp_process_ext_param(struct sctp_association *asoc, for (i = 0; i < num_ext; i++) { switch (param.ext->chunks[i]) { case SCTP_CID_FWD_TSN: - if (net->sctp.prsctp_enable && !asoc->peer.prsctp_capable) - asoc->peer.prsctp_capable = 1; + if (asoc->prsctp_enable && !asoc->peer.prsctp_capable) + asoc->peer.prsctp_capable = 1; break; case SCTP_CID_AUTH: /* if the peer reports AUTH, assume that he @@ -2169,7 +2169,7 @@ static sctp_ierror_t sctp_verify_param(struct net *net, break; case SCTP_PARAM_FWD_TSN_SUPPORT: - if (net->sctp.prsctp_enable) + if (ep->prsctp_enable) break; goto fallthrough; @@ -2653,7 +2653,7 @@ do_addr_param: break; case SCTP_PARAM_FWD_TSN_SUPPORT: - if (net->sctp.prsctp_enable) { + if (asoc->prsctp_enable) { asoc->peer.prsctp_capable = 1; break; } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index cdabbd8..7460dde 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3661,6 +3661,39 @@ static int sctp_setsockopt_recvnxtinfo(struct sock *sk, return 0; } +static int sctp_setsockopt_pr_supported(struct sock *sk, + char __user *optval, + unsigned int optlen) +{ + struct sctp_assoc_value params; + struct sctp_association *asoc; + int retval = -EINVAL; + + if (optlen != sizeof(params)) + goto out; + + if (copy_from_user(¶ms, optval, optlen)) { + retval = -EFAULT; + goto out; + } + + asoc = sctp_id2assoc(sk, params.assoc_id); + if (asoc) { + asoc->prsctp_enable = !!params.assoc_value; + } else if (!params.assoc_id) { + struct sctp_sock *sp = sctp_sk(sk); + + sp->ep->prsctp_enable = !!params.assoc_value; + } else { + goto out; + } + + retval = 0; + +out: + return retval; +} + /* API 6.2 setsockopt(), getsockopt() * * Applications use setsockopt() and getsockopt() to set or retrieve @@ -3821,6 +3854,9 @@ static int sctp_setsockopt(struct sock *sk, int level, int optname, case SCTP_RECVNXTINFO: retval = sctp_setsockopt_recvnxtinfo(sk, optval, optlen); break; + case SCTP_PR_SUPPORTED: + retval = sctp_setsockopt_pr_supported(sk, optval, optlen); + break; default: retval = -ENOPROTOOPT; break; @@ -6166,6 +6202,47 @@ static int sctp_getsockopt_recvnxtinfo(struct sock *sk, int len, return 0; } +static int sctp_getsockopt_pr_supported(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_assoc_value params; + struct sctp_association *asoc; + int retval = -EFAULT; + + if (len < sizeof(params)) { + retval = -EINVAL; + goto out; + } + + len = sizeof(params); + if (copy_from_user(¶ms, optval, len)) + goto out; + + asoc = sctp_id2assoc(sk, params.assoc_id); + if (asoc) { + params.assoc_value = asoc->prsctp_enable; + } else if (!params.assoc_id) { + struct sctp_sock *sp = sctp_sk(sk); + + params.assoc_value = sp->ep->prsctp_enable; + } else { + retval = -EINVAL; + goto out; + } + + if (put_user(len, optlen)) + goto out; + + if (copy_to_user(optval, ¶ms, len)) + goto out; + + retval = 0; + +out: + return retval; +} + static int sctp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -6319,6 +6396,9 @@ static int sctp_getsockopt(struct sock *sk, int level, int optname, case SCTP_RECVNXTINFO: retval = sctp_getsockopt_recvnxtinfo(sk, len, optval, optlen); break; + case SCTP_PR_SUPPORTED: + retval = sctp_getsockopt_pr_supported(sk, len, optval, optlen); + break; default: retval = -ENOPROTOOPT; break; -- cgit v0.10.2 From f959fb442c35f4b61fea341401b8463dd0a1b959 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 9 Jul 2016 19:47:41 +0800 Subject: sctp: add SCTP_DEFAULT_PRINFO into sctp sockopt This patch adds SCTP_DEFAULT_PRINFO to sctp sockopt. It is used to set/get sctp Partially Reliable Policies' default params, which includes 3 policies (ttl, rtx, prio) and their values. Still, if we set policy params in sndinfo, we will use the params of sndinfo against chunks, instead of the default params. In this patch, we will use 5-8bit of sp/asoc->default_flags to store prsctp policies, and reuse asoc->default_timetolive to store their values. It means if we enable and set prsctp policy, prior ttl timeout in sctp will not work any more. Signed-off-by: Xin Long Signed-off-by: David S. Miller diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h index aa08906..984cf2e 100644 --- a/include/uapi/linux/sctp.h +++ b/include/uapi/linux/sctp.h @@ -113,6 +113,29 @@ typedef __s32 sctp_assoc_t; #define SCTP_SOCKOPT_CONNECTX3 111 /* CONNECTX requests (updated) */ #define SCTP_GET_ASSOC_STATS 112 /* Read only */ #define SCTP_PR_SUPPORTED 113 +#define SCTP_DEFAULT_PRINFO 114 + +/* PR-SCTP policies */ +#define SCTP_PR_SCTP_NONE 0x0000 +#define SCTP_PR_SCTP_TTL 0x0010 +#define SCTP_PR_SCTP_RTX 0x0020 +#define SCTP_PR_SCTP_PRIO 0x0030 +#define SCTP_PR_SCTP_MAX SCTP_PR_SCTP_PRIO +#define SCTP_PR_SCTP_MASK 0x0030 + +#define __SCTP_PR_INDEX(x) ((x >> 4) - 1) +#define SCTP_PR_INDEX(x) __SCTP_PR_INDEX(SCTP_PR_SCTP_ ## x) + +#define SCTP_PR_POLICY(x) ((x) & SCTP_PR_SCTP_MASK) +#define SCTP_PR_SET_POLICY(flags, x) \ + do { \ + flags &= ~SCTP_PR_SCTP_MASK; \ + flags |= x; \ + } while (0) + +#define SCTP_PR_TTL_ENABLED(x) (SCTP_PR_POLICY(x) == SCTP_PR_SCTP_TTL) +#define SCTP_PR_RTX_ENABLED(x) (SCTP_PR_POLICY(x) == SCTP_PR_SCTP_RTX) +#define SCTP_PR_PRIO_ENABLED(x) (SCTP_PR_POLICY(x) == SCTP_PR_SCTP_PRIO) /* These are bit fields for msghdr->msg_flags. See section 5.1. */ /* On user space Linux, these live in as an enum. */ @@ -903,4 +926,10 @@ struct sctp_paddrthlds { __u16 spt_pathpfthld; }; +struct sctp_default_prinfo { + sctp_assoc_t pr_assoc_id; + __u32 pr_value; + __u16 pr_policy; +}; + #endif /* _UAPI_SCTP_H */ diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 7460dde..c03fe1b 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3694,6 +3694,47 @@ out: return retval; } +static int sctp_setsockopt_default_prinfo(struct sock *sk, + char __user *optval, + unsigned int optlen) +{ + struct sctp_default_prinfo info; + struct sctp_association *asoc; + int retval = -EINVAL; + + if (optlen != sizeof(info)) + goto out; + + if (copy_from_user(&info, optval, sizeof(info))) { + retval = -EFAULT; + goto out; + } + + if (info.pr_policy & ~SCTP_PR_SCTP_MASK) + goto out; + + if (info.pr_policy == SCTP_PR_SCTP_NONE) + info.pr_value = 0; + + asoc = sctp_id2assoc(sk, info.pr_assoc_id); + if (asoc) { + SCTP_PR_SET_POLICY(asoc->default_flags, info.pr_policy); + asoc->default_timetolive = info.pr_value; + } else if (!info.pr_assoc_id) { + struct sctp_sock *sp = sctp_sk(sk); + + SCTP_PR_SET_POLICY(sp->default_flags, info.pr_policy); + sp->default_timetolive = info.pr_value; + } else { + goto out; + } + + retval = 0; + +out: + return retval; +} + /* API 6.2 setsockopt(), getsockopt() * * Applications use setsockopt() and getsockopt() to set or retrieve @@ -3857,6 +3898,9 @@ static int sctp_setsockopt(struct sock *sk, int level, int optname, case SCTP_PR_SUPPORTED: retval = sctp_setsockopt_pr_supported(sk, optval, optlen); break; + case SCTP_DEFAULT_PRINFO: + retval = sctp_setsockopt_default_prinfo(sk, optval, optlen); + break; default: retval = -ENOPROTOOPT; break; @@ -6243,6 +6287,49 @@ out: return retval; } +static int sctp_getsockopt_default_prinfo(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_default_prinfo info; + struct sctp_association *asoc; + int retval = -EFAULT; + + if (len < sizeof(info)) { + retval = -EINVAL; + goto out; + } + + len = sizeof(info); + if (copy_from_user(&info, optval, len)) + goto out; + + asoc = sctp_id2assoc(sk, info.pr_assoc_id); + if (asoc) { + info.pr_policy = SCTP_PR_POLICY(asoc->default_flags); + info.pr_value = asoc->default_timetolive; + } else if (!info.pr_assoc_id) { + struct sctp_sock *sp = sctp_sk(sk); + + info.pr_policy = SCTP_PR_POLICY(sp->default_flags); + info.pr_value = sp->default_timetolive; + } else { + retval = -EINVAL; + goto out; + } + + if (put_user(len, optlen)) + goto out; + + if (copy_to_user(optval, &info, len)) + goto out; + + retval = 0; + +out: + return retval; +} + static int sctp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -6399,6 +6486,10 @@ static int sctp_getsockopt(struct sock *sk, int level, int optname, case SCTP_PR_SUPPORTED: retval = sctp_getsockopt_pr_supported(sk, len, optval, optlen); break; + case SCTP_DEFAULT_PRINFO: + retval = sctp_getsockopt_default_prinfo(sk, len, optval, + optlen); + break; default: retval = -ENOPROTOOPT; break; -- cgit v0.10.2 From 826d253d57b11f69add81c8086d2e7f1dce5ec77 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 9 Jul 2016 19:47:42 +0800 Subject: sctp: add SCTP_PR_ASSOC_STATUS on sctp sockopt This patch adds SCTP_PR_ASSOC_STATUS to sctp sockopt, which is used to dump the prsctp statistics info from the asoc. The prsctp statistics includes abandoned_sent/unsent from the asoc. abandoned_sent is the count of the packets we drop packets from retransmit/transmited queue, and abandoned_unsent is the count of the packets we drop from out_queue according to the policy. Note: another option for prsctp statistics dump described in rfc is SCTP_PR_STREAM_STATUS, which is used to dump the prsctp statistics info from each stream. But by now, linux doesn't yet have per stream statistics info, it needs rfc6525 to be implemented. As the prsctp statistics for each stream has to be based on per stream statistics, we will delay it until rfc6525 is done in linux. Signed-off-by: Xin Long Signed-off-by: David S. Miller diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 07115ca..d8e464a 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1853,6 +1853,9 @@ struct sctp_association { prsctp_enable:1; struct sctp_priv_assoc_stats stats; + + __u64 abandoned_unsent[SCTP_PR_INDEX(MAX) + 1]; + __u64 abandoned_sent[SCTP_PR_INDEX(MAX) + 1]; }; diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h index 984cf2e..d304f4c 100644 --- a/include/uapi/linux/sctp.h +++ b/include/uapi/linux/sctp.h @@ -114,6 +114,7 @@ typedef __s32 sctp_assoc_t; #define SCTP_GET_ASSOC_STATS 112 /* Read only */ #define SCTP_PR_SUPPORTED 113 #define SCTP_DEFAULT_PRINFO 114 +#define SCTP_PR_ASSOC_STATUS 115 /* PR-SCTP policies */ #define SCTP_PR_SCTP_NONE 0x0000 @@ -926,6 +927,17 @@ struct sctp_paddrthlds { __u16 spt_pathpfthld; }; +/* + * Socket Option for Getting the Association/Stream-Specific PR-SCTP Status + */ +struct sctp_prstatus { + sctp_assoc_t sprstat_assoc_id; + __u16 sprstat_sid; + __u16 sprstat_policy; + __u64 sprstat_abandoned_unsent; + __u64 sprstat_abandoned_sent; +}; + struct sctp_default_prinfo { sctp_assoc_t pr_assoc_id; __u32 pr_value; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index c03fe1b..c3167c4 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -6330,6 +6330,64 @@ out: return retval; } +static int sctp_getsockopt_pr_assocstatus(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_prstatus params; + struct sctp_association *asoc; + int policy; + int retval = -EINVAL; + + if (len < sizeof(params)) + goto out; + + len = sizeof(params); + if (copy_from_user(¶ms, optval, len)) { + retval = -EFAULT; + goto out; + } + + policy = params.sprstat_policy; + if (policy & ~SCTP_PR_SCTP_MASK) + goto out; + + asoc = sctp_id2assoc(sk, params.sprstat_assoc_id); + if (!asoc) + goto out; + + if (policy == SCTP_PR_SCTP_NONE) { + params.sprstat_abandoned_unsent = 0; + params.sprstat_abandoned_sent = 0; + for (policy = 0; policy <= SCTP_PR_INDEX(MAX); policy++) { + params.sprstat_abandoned_unsent += + asoc->abandoned_unsent[policy]; + params.sprstat_abandoned_sent += + asoc->abandoned_sent[policy]; + } + } else { + params.sprstat_abandoned_unsent = + asoc->abandoned_unsent[__SCTP_PR_INDEX(policy)]; + params.sprstat_abandoned_sent = + asoc->abandoned_sent[__SCTP_PR_INDEX(policy)]; + } + + if (put_user(len, optlen)) { + retval = -EFAULT; + goto out; + } + + if (copy_to_user(optval, ¶ms, len)) { + retval = -EFAULT; + goto out; + } + + retval = 0; + +out: + return retval; +} + static int sctp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -6490,6 +6548,10 @@ static int sctp_getsockopt(struct sock *sk, int level, int optname, retval = sctp_getsockopt_default_prinfo(sk, len, optval, optlen); break; + case SCTP_PR_ASSOC_STATUS: + retval = sctp_getsockopt_pr_assocstatus(sk, len, optval, + optlen); + break; default: retval = -ENOPROTOOPT; break; -- cgit v0.10.2 From a6c2f792873aff332a4689717c3cd6104f46684c Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 9 Jul 2016 19:47:43 +0800 Subject: sctp: implement prsctp TTL policy prsctp TTL policy is a policy to abandon chunks when they expire at the specific time in local stack. It's similar with expires_at in struct sctp_datamsg. This patch uses sinfo->sinfo_timetolive to set the specific time for TTL policy. sinfo->sinfo_timetolive is also used for msg->expires_at. So if prsctp_enable or TTL policy is not enabled, msg->expires_at still works as before. Signed-off-by: Xin Long Signed-off-by: David S. Miller diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index d8e464a..6bcda71 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -602,6 +602,16 @@ struct sctp_chunk { /* This needs to be recoverable for SCTP_SEND_FAILED events. */ struct sctp_sndrcvinfo sinfo; + /* We use this field to record param for prsctp policies, + * for TTL policy, it is the time_to_drop of this chunk, + * for RTX policy, it is the max_sent_count of this chunk, + * for PRIO policy, it is the priority of this chunk. + */ + unsigned long prsctp_param; + + /* How many times this chunk have been sent, for prsctp RTX policy */ + int sent_count; + /* Which association does this belong to? */ struct sctp_association *asoc; diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c index 1eb94bf..2698d12 100644 --- a/net/sctp/chunk.c +++ b/net/sctp/chunk.c @@ -335,13 +335,27 @@ errout: /* Check whether this message has expired. */ int sctp_chunk_abandoned(struct sctp_chunk *chunk) { - struct sctp_datamsg *msg = chunk->msg; + if (!chunk->asoc->prsctp_enable || + !SCTP_PR_POLICY(chunk->sinfo.sinfo_flags)) { + struct sctp_datamsg *msg = chunk->msg; + + if (!msg->can_abandon) + return 0; + + if (time_after(jiffies, msg->expires_at)) + return 1; - if (!msg->can_abandon) return 0; + } - if (time_after(jiffies, msg->expires_at)) + if (SCTP_PR_TTL_ENABLED(chunk->sinfo.sinfo_flags) && + time_after(jiffies, chunk->prsctp_param)) { + if (chunk->sent_count) + chunk->asoc->abandoned_sent[SCTP_PR_INDEX(TTL)]++; + else + chunk->asoc->abandoned_unsent[SCTP_PR_INDEX(TTL)]++; return 1; + } return 0; } diff --git a/net/sctp/output.c b/net/sctp/output.c index 2e9223b..7425f6c 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -316,6 +316,8 @@ static sctp_xmit_t __sctp_packet_append_chunk(struct sctp_packet *packet, packet->has_data = 1; /* timestamp the chunk for rtx purposes */ chunk->sent_at = jiffies; + /* Mainly used for prsctp RTX policy */ + chunk->sent_count++; break; case SCTP_CID_COOKIE_ECHO: packet->has_cookie_echo = 1; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 0e3045e..2c431ee 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -711,6 +711,17 @@ nodata: return retval; } +static void sctp_set_prsctp_policy(struct sctp_chunk *chunk, + const struct sctp_sndrcvinfo *sinfo) +{ + if (!chunk->asoc->prsctp_enable) + return; + + if (SCTP_PR_TTL_ENABLED(sinfo->sinfo_flags)) + chunk->prsctp_param = + jiffies + msecs_to_jiffies(sinfo->sinfo_timetolive); +} + /* Make a DATA chunk for the given association from the provided * parameters. However, do not populate the data payload. */ @@ -744,6 +755,7 @@ struct sctp_chunk *sctp_make_datafrag_empty(struct sctp_association *asoc, retval->subh.data_hdr = sctp_addto_chunk(retval, sizeof(dp), &dp); memcpy(&retval->sinfo, sinfo, sizeof(struct sctp_sndrcvinfo)); + sctp_set_prsctp_policy(retval, sinfo); nodata: return retval; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index c3167c4..0861429 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -7099,7 +7099,7 @@ static int sctp_msghdr_parse(const struct msghdr *msg, sctp_cmsgs_t *cmsgs) if (cmsgs->srinfo->sinfo_flags & ~(SCTP_UNORDERED | SCTP_ADDR_OVER | - SCTP_SACK_IMMEDIATELY | + SCTP_SACK_IMMEDIATELY | SCTP_PR_SCTP_MASK | SCTP_ABORT | SCTP_EOF)) return -EINVAL; break; @@ -7123,7 +7123,7 @@ static int sctp_msghdr_parse(const struct msghdr *msg, sctp_cmsgs_t *cmsgs) if (cmsgs->sinfo->snd_flags & ~(SCTP_UNORDERED | SCTP_ADDR_OVER | - SCTP_SACK_IMMEDIATELY | + SCTP_SACK_IMMEDIATELY | SCTP_PR_SCTP_MASK | SCTP_ABORT | SCTP_EOF)) return -EINVAL; break; -- cgit v0.10.2 From 01aadb3af6e1b10970c1f7e510b5097c8f725d64 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 9 Jul 2016 19:47:44 +0800 Subject: sctp: implement prsctp RTX policy prsctp RTX policy is a policy to abandon chunks when they are retransmitted beyond the max count. This patch uses sent_count to count how many times one chunk has been sent, and prsctp_param is the max rtx count, which is from sinfo->sinfo_timetolive in sctp_set_prsctp_policy(). So similar to TTL policy, if RTX policy is enabled, msg->expire_at won't work. Then in sctp_chunk_abandoned, this patch checks if chunk->sent_count is bigger than chunk->prsctp_param to abandon this chunk. Signed-off-by: Xin Long Signed-off-by: David S. Miller diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c index 2698d12..b3692b5 100644 --- a/net/sctp/chunk.c +++ b/net/sctp/chunk.c @@ -355,6 +355,10 @@ int sctp_chunk_abandoned(struct sctp_chunk *chunk) else chunk->asoc->abandoned_unsent[SCTP_PR_INDEX(TTL)]++; return 1; + } else if (SCTP_PR_RTX_ENABLED(chunk->sinfo.sinfo_flags) && + chunk->sent_count > chunk->prsctp_param) { + chunk->asoc->abandoned_sent[SCTP_PR_INDEX(RTX)]++; + return 1; } return 0; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 2c431ee..cfde934 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -720,6 +720,8 @@ static void sctp_set_prsctp_policy(struct sctp_chunk *chunk, if (SCTP_PR_TTL_ENABLED(sinfo->sinfo_flags)) chunk->prsctp_param = jiffies + msecs_to_jiffies(sinfo->sinfo_timetolive); + else if (SCTP_PR_RTX_ENABLED(sinfo->sinfo_flags)) + chunk->prsctp_param = sinfo->sinfo_timetolive; } /* Make a DATA chunk for the given association from the provided -- cgit v0.10.2 From 8dbdf1f5b09cb22560e7c7173b52fe3c631046bd Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sat, 9 Jul 2016 19:47:45 +0800 Subject: sctp: implement prsctp PRIO policy prsctp PRIO policy is a policy to abandon lower priority chunks when asoc doesn't have enough snd buffer, so that the current chunk with higher priority can be queued successfully. Similar to TTL/RTX policy, we will set the priority of the chunk to prsctp_param with sinfo->sinfo_timetolive in sctp_set_prsctp_policy(). So if PRIO policy is enabled, msg->expire_at won't work. asoc->sent_cnt_removable will record how many chunks can be checked to remove. If priority policy is enabled, when the chunk is queued into the out_queue, we will increase sent_cnt_removable. When the chunk is moved to abandon_queue or dequeue and free, we will decrease sent_cnt_removable. In sctp_sendmsg, we will check if there is enough snd buffer for current msg and if sent_cnt_removable is not 0. Then try to abandon chunks in sctp_prune_prsctp when sendmsg from the retransmit/transmited queue, and free chunks from out_queue in right order until the abandon+free size > msg_len - sctp_wfree. For the abandon size, we have to wait until it sends FORWARD TSN, receives the sack and the chunks are really freed. Signed-off-by: Xin Long Signed-off-by: David S. Miller diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 6bcda71..8626bdd 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1084,6 +1084,8 @@ void sctp_retransmit(struct sctp_outq *, struct sctp_transport *, sctp_retransmit_reason_t); void sctp_retransmit_mark(struct sctp_outq *, struct sctp_transport *, __u8); int sctp_outq_uncork(struct sctp_outq *, gfp_t gfp); +void sctp_prsctp_prune(struct sctp_association *asoc, + struct sctp_sndrcvinfo *sinfo, int msg_len); /* Uncork and flush an outqueue. */ static inline void sctp_outq_cork(struct sctp_outq *q) { @@ -1864,6 +1866,8 @@ struct sctp_association { struct sctp_priv_assoc_stats stats; + int sent_cnt_removable; + __u64 abandoned_unsent[SCTP_PR_INDEX(MAX) + 1]; __u64 abandoned_sent[SCTP_PR_INDEX(MAX) + 1]; }; diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c index b3692b5..a55e547 100644 --- a/net/sctp/chunk.c +++ b/net/sctp/chunk.c @@ -360,6 +360,7 @@ int sctp_chunk_abandoned(struct sctp_chunk *chunk) chunk->asoc->abandoned_sent[SCTP_PR_INDEX(RTX)]++; return 1; } + /* PRIO policy is processed by sendmsg, not here */ return 0; } diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index 084718f..72e54a4 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -326,6 +326,9 @@ int sctp_outq_tail(struct sctp_outq *q, struct sctp_chunk *chunk, gfp_t gfp) sctp_chunk_hold(chunk); sctp_outq_tail_data(q, chunk); + if (chunk->asoc->prsctp_enable && + SCTP_PR_PRIO_ENABLED(chunk->sinfo.sinfo_flags)) + chunk->asoc->sent_cnt_removable++; if (chunk->chunk_hdr->flags & SCTP_DATA_UNORDERED) SCTP_INC_STATS(net, SCTP_MIB_OUTUNORDERCHUNKS); else @@ -372,6 +375,96 @@ static void sctp_insert_list(struct list_head *head, struct list_head *new) list_add_tail(new, head); } +static int sctp_prsctp_prune_sent(struct sctp_association *asoc, + struct sctp_sndrcvinfo *sinfo, + struct list_head *queue, int msg_len) +{ + struct sctp_chunk *chk, *temp; + + list_for_each_entry_safe(chk, temp, queue, transmitted_list) { + if (!SCTP_PR_PRIO_ENABLED(chk->sinfo.sinfo_flags) || + chk->prsctp_param <= sinfo->sinfo_timetolive) + continue; + + list_del_init(&chk->transmitted_list); + sctp_insert_list(&asoc->outqueue.abandoned, + &chk->transmitted_list); + + asoc->sent_cnt_removable--; + asoc->abandoned_sent[SCTP_PR_INDEX(PRIO)]++; + + if (!chk->tsn_gap_acked) { + if (chk->transport) + chk->transport->flight_size -= + sctp_data_size(chk); + asoc->outqueue.outstanding_bytes -= sctp_data_size(chk); + } + + msg_len -= SCTP_DATA_SNDSIZE(chk) + + sizeof(struct sk_buff) + + sizeof(struct sctp_chunk); + if (msg_len <= 0) + break; + } + + return msg_len; +} + +static int sctp_prsctp_prune_unsent(struct sctp_association *asoc, + struct sctp_sndrcvinfo *sinfo, + struct list_head *queue, int msg_len) +{ + struct sctp_chunk *chk, *temp; + + list_for_each_entry_safe(chk, temp, queue, list) { + if (!SCTP_PR_PRIO_ENABLED(chk->sinfo.sinfo_flags) || + chk->prsctp_param <= sinfo->sinfo_timetolive) + continue; + + list_del_init(&chk->list); + asoc->sent_cnt_removable--; + asoc->abandoned_unsent[SCTP_PR_INDEX(PRIO)]++; + + msg_len -= SCTP_DATA_SNDSIZE(chk) + + sizeof(struct sk_buff) + + sizeof(struct sctp_chunk); + sctp_chunk_free(chk); + if (msg_len <= 0) + break; + } + + return msg_len; +} + +/* Abandon the chunks according their priorities */ +void sctp_prsctp_prune(struct sctp_association *asoc, + struct sctp_sndrcvinfo *sinfo, int msg_len) +{ + struct sctp_transport *transport; + + if (!asoc->prsctp_enable || !asoc->sent_cnt_removable) + return; + + msg_len = sctp_prsctp_prune_sent(asoc, sinfo, + &asoc->outqueue.retransmit, + msg_len); + if (msg_len <= 0) + return; + + list_for_each_entry(transport, &asoc->peer.transport_addr_list, + transports) { + msg_len = sctp_prsctp_prune_sent(asoc, sinfo, + &transport->transmitted, + msg_len); + if (msg_len <= 0) + return; + } + + sctp_prsctp_prune_unsent(asoc, sinfo, + &asoc->outqueue.out_chunk_list, + msg_len); +} + /* Mark all the eligible packets on a transport for retransmission. */ void sctp_retransmit_mark(struct sctp_outq *q, struct sctp_transport *transport, @@ -962,6 +1055,9 @@ static int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout, gfp_t gfp) /* Mark as failed send. */ sctp_chunk_fail(chunk, SCTP_ERROR_INV_STRM); + if (asoc->prsctp_enable && + SCTP_PR_PRIO_ENABLED(chunk->sinfo.sinfo_flags)) + asoc->sent_cnt_removable--; sctp_chunk_free(chunk); continue; } @@ -1251,6 +1347,9 @@ int sctp_outq_sack(struct sctp_outq *q, struct sctp_chunk *chunk) tsn = ntohl(tchunk->subh.data_hdr->tsn); if (TSN_lte(tsn, ctsn)) { list_del_init(&tchunk->transmitted_list); + if (asoc->prsctp_enable && + SCTP_PR_PRIO_ENABLED(chunk->sinfo.sinfo_flags)) + asoc->sent_cnt_removable--; sctp_chunk_free(tchunk); } } diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index cfde934..1c96f47 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -720,7 +720,8 @@ static void sctp_set_prsctp_policy(struct sctp_chunk *chunk, if (SCTP_PR_TTL_ENABLED(sinfo->sinfo_flags)) chunk->prsctp_param = jiffies + msecs_to_jiffies(sinfo->sinfo_timetolive); - else if (SCTP_PR_RTX_ENABLED(sinfo->sinfo_flags)) + else if (SCTP_PR_RTX_ENABLED(sinfo->sinfo_flags) || + SCTP_PR_PRIO_ENABLED(sinfo->sinfo_flags)) chunk->prsctp_param = sinfo->sinfo_timetolive; } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 0861429..71c7dc5 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1914,6 +1914,9 @@ static int sctp_sendmsg(struct sock *sk, struct msghdr *msg, size_t msg_len) goto out_free; } + if (sctp_wspace(asoc) < msg_len) + sctp_prsctp_prune(asoc, sinfo, msg_len - sctp_wspace(asoc)); + timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT); if (!sctp_wspace(asoc)) { err = sctp_wait_for_sndbuf(asoc, &timeo, msg_len); -- cgit v0.10.2 From aa9667e7f626232cde5310435dc96bfbe796f05c Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Sun, 10 Jul 2016 10:20:11 +0900 Subject: tunnels: correct conditional build of MPLS and IPv6 Using a combination if #if conditionals and goto labels to unwind tunnel4_init seems unwieldy. This patch takes a simpler approach of directly unregistering previously registered protocols when an error occurs. This fixes a number of problems with the current implementation including the potential presence of labels when they are unused and the potential absence of unregister code when it is needed. Fixes: 8afe97e5d416 ("tunnels: support MPLS over IPv4 tunnels") Signed-off-by: Simon Horman Signed-off-by: David S. Miller diff --git a/net/ipv4/tunnel4.c b/net/ipv4/tunnel4.c index 45cd425..ec35eaa 100644 --- a/net/ipv4/tunnel4.c +++ b/net/ipv4/tunnel4.c @@ -208,24 +208,25 @@ static const struct net_protocol tunnelmpls4_protocol = { static int __init tunnel4_init(void) { if (inet_add_protocol(&tunnel4_protocol, IPPROTO_IPIP)) - goto err_ipip; + goto err; #if IS_ENABLED(CONFIG_IPV6) - if (inet_add_protocol(&tunnel64_protocol, IPPROTO_IPV6)) - goto err_ipv6; + if (inet_add_protocol(&tunnel64_protocol, IPPROTO_IPV6)) { + inet_del_protocol(&tunnel4_protocol, IPPROTO_IPIP); + goto err; + } #endif #if IS_ENABLED(CONFIG_MPLS) - if (inet_add_protocol(&tunnelmpls4_protocol, IPPROTO_MPLS)) - goto err_mpls; + if (inet_add_protocol(&tunnelmpls4_protocol, IPPROTO_MPLS)) { + inet_del_protocol(&tunnel4_protocol, IPPROTO_IPIP); +#if IS_ENABLED(CONFIG_IPV6) + inet_del_protocol(&tunnel64_protocol, IPPROTO_IPV6); +#endif + goto err; + } #endif return 0; -#if IS_ENABLED(CONFIG_IPV6) -err_mpls: - inet_del_protocol(&tunnel4_protocol, IPPROTO_IPV6); -#endif -err_ipv6: - inet_del_protocol(&tunnel4_protocol, IPPROTO_IPIP); -err_ipip: +err: pr_err("%s: can't add protocol\n", __func__); return -EAGAIN; } -- cgit v0.10.2 From 752e6d5dfafd6acb06ae4379151ad1685bb0d1e2 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Sun, 10 Jul 2016 09:42:44 +0200 Subject: MAINTAINERS: release Scott from being a rocker maintainer As requested by Scott, removing him. Signed-off-by: Scott Feldman Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/MAINTAINERS b/MAINTAINERS index 6374be2..06e8411 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -9710,7 +9710,6 @@ F: Documentation/ABI/*/sysfs-driver-hid-roccat* ROCKER DRIVER M: Jiri Pirko -M: Scott Feldman L: netdev@vger.kernel.org S: Supported F: drivers/net/ethernet/rocker/ -- cgit v0.10.2 From e5de25dce9243a3d29b5ebc131cc9d59008f39f7 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 11 Jul 2016 13:12:28 +0200 Subject: drivers/net: fixup comments after "Future-proof tunnel offload handlers" Some comments weren't updated to reflect the renaming of ndo's and the change of arguments. Signed-off-by: Sabrina Dubroca Acked-by: Alexander Duyck Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c index d00cb19..20a5bbe 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_netdev.c @@ -432,9 +432,7 @@ static void fm10k_restore_vxlan_port(struct fm10k_intfc *interface) /** * fm10k_add_vxlan_port * @netdev: network interface device structure - * @sa_family: Address family of new port - * @port: port number used for VXLAN - * @type: Enumerated value specifying udp encapsulation type + * @ti: Tunnel endpoint information * * This function is called when a new VXLAN interface has added a new port * number to the range that is currently in use for VXLAN. The new port @@ -480,9 +478,7 @@ insert_tail: /** * fm10k_del_vxlan_port * @netdev: network interface device structure - * @sa_family: Address family of freed port - * @port: port number used for VXLAN - * @type: Enumerated value specifying udp encapsulation type + * @ti: Tunnel endpoint information * * This function is called when a new VXLAN interface has freed a port * number from the range that is currently in use for VXLAN. The freed diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index fd5a761..918b94b 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -8851,9 +8851,7 @@ static int ixgbe_set_features(struct net_device *netdev, /** * ixgbe_add_vxlan_port - Get notifications about VXLAN ports that come up * @dev: The port's netdev - * @sa_family: Socket Family that VXLAN is notifiying us about - * @port: New UDP port number that VXLAN started listening to - * @type: Enumerated type specifying UDP tunnel type + * @ti: Tunnel endpoint information **/ static void ixgbe_add_vxlan_port(struct net_device *dev, struct udp_tunnel_info *ti) @@ -8888,9 +8886,7 @@ static void ixgbe_add_vxlan_port(struct net_device *dev, /** * ixgbe_del_vxlan_port - Get notifications about VXLAN ports that go away * @dev: The port's netdev - * @sa_family: Socket Family that VXLAN is notifying us about - * @port: UDP port number that VXLAN stopped listening to - * @type: Enumerated type specifying UDP tunnel type + * @ti: Tunnel endpoint information **/ static void ixgbe_del_vxlan_port(struct net_device *dev, struct udp_tunnel_info *ti) diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 5de892f..3c20e87 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -1134,9 +1134,9 @@ static struct device_type geneve_type = { .name = "geneve", }; -/* Calls the ndo_add_udp_enc_port of the caller in order to +/* Calls the ndo_udp_tunnel_add of the caller in order to * supply the listening GENEVE udp ports. Callers are expected - * to implement the ndo_add_udp_enc_port. + * to implement the ndo_udp_tunnel_add. */ static void geneve_push_rx_ports(struct net_device *dev) { diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index ae7455d..da4e3d6 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -2475,9 +2475,9 @@ static struct device_type vxlan_type = { .name = "vxlan", }; -/* Calls the ndo_add_udp_enc_port of the caller in order to +/* Calls the ndo_udp_tunnel_add of the caller in order to * supply the listening VXLAN udp ports. Callers are expected - * to implement the ndo_add_udp_enc_port. + * to implement the ndo_udp_tunnel_add. */ static void vxlan_push_rx_ports(struct net_device *dev) { -- cgit v0.10.2 From 8876d94b410849a20e78c25541ca13664e8fe5c6 Mon Sep 17 00:00:00 2001 From: Jeremy Linton Date: Mon, 11 Jul 2016 10:28:40 -0500 Subject: net: smc91x: ACPI Enable lan91x adapters Enable lan91x adapters in some ARM machines and models when booted with an ACPI kernel. Signed-off-by: Jeremy Linton Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/smsc/smc91x.c b/drivers/net/ethernet/smsc/smc91x.c index 18ac52d..726b80f 100644 --- a/drivers/net/ethernet/smsc/smc91x.c +++ b/drivers/net/ethernet/smsc/smc91x.c @@ -2195,6 +2195,12 @@ static void smc_release_datacs(struct platform_device *pdev, struct net_device * } } +static const struct acpi_device_id smc91x_acpi_match[] = { + { "LNRO0003", 0 }, + { } +}; +MODULE_DEVICE_TABLE(acpi, smc91x_acpi_match); + #if IS_BUILTIN(CONFIG_OF) static const struct of_device_id smc91x_match[] = { { .compatible = "smsc,lan91c94", }, @@ -2274,7 +2280,6 @@ static int smc_drv_probe(struct platform_device *pdev) #if IS_BUILTIN(CONFIG_OF) match = of_match_device(of_match_ptr(smc91x_match), &pdev->dev); if (match) { - struct device_node *np = pdev->dev.of_node; u32 val; /* Optional pwrdwn GPIO configured? */ @@ -2300,7 +2305,8 @@ static int smc_drv_probe(struct platform_device *pdev) usleep_range(750, 1000); /* Combination of IO widths supported, default to 16-bit */ - if (!of_property_read_u32(np, "reg-io-width", &val)) { + if (!device_property_read_u32(&pdev->dev, "reg-io-width", + &val)) { if (val & 1) lp->cfg.flags |= SMC91X_USE_8BIT; if ((val == 0) || (val & 2)) @@ -2478,7 +2484,8 @@ static struct platform_driver smc_driver = { .driver = { .name = CARDNAME, .pm = &smc_drv_pm_ops, - .of_match_table = of_match_ptr(smc91x_match), + .of_match_table = of_match_ptr(smc91x_match), + .acpi_match_table = smc91x_acpi_match, }, }; -- cgit v0.10.2 From f50cef6f77d08fced02eba7ac7c1cab4972c3786 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Mon, 11 Jul 2016 16:54:20 +0100 Subject: nfp: check idx is -ENOSPC before using it is an index idx can be returned as -ENOSPC, so we should check for this first before using it as an index into nn->vxlan_usecnt[] to avoid an out of bounds array offset read. Signed-off-by: Colin Ian King Acked-by: Jakub Kicinski Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 1e74b91..88678c1 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -2578,7 +2578,7 @@ static void nfp_net_del_vxlan_port(struct net_device *netdev, return; idx = nfp_net_find_vxlan_idx(nn, ti->port); - if (!nn->vxlan_usecnt[idx] || idx == -ENOSPC) + if (idx == -ENOSPC || !nn->vxlan_usecnt[idx]) return; if (!--nn->vxlan_usecnt[idx]) -- cgit v0.10.2 From a536a6e13ecd0d6eb0ffc36c5d56555896617282 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Mon, 11 Jul 2016 12:51:01 -0400 Subject: bpf: make inode code explicitly non-modular The Kconfig currently controlling compilation of this code is: init/Kconfig:config BPF_SYSCALL init/Kconfig: bool "Enable bpf() system call" ...meaning that it currently is not being built as a module by anyone. Lets remove the couple traces of modular infrastructure use, so that when reading the driver there is no doubt it is builtin-only. Note that MODULE_ALIAS is a no-op for non-modular code. We replace module.h with init.h since the file does use __init. Cc: Alexei Starovoitov Cc: netdev@vger.kernel.org Signed-off-by: Paul Gortmaker Acked-by: Daniel Borkmann Signed-off-by: David S. Miller diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c index 318858e..5967b87 100644 --- a/kernel/bpf/inode.c +++ b/kernel/bpf/inode.c @@ -11,7 +11,7 @@ * version 2 as published by the Free Software Foundation. */ -#include +#include #include #include #include @@ -367,8 +367,6 @@ static struct file_system_type bpf_fs_type = { .kill_sb = kill_litter_super, }; -MODULE_ALIAS_FS("bpf"); - static int __init bpf_init(void) { int ret; -- cgit v0.10.2 From cf81b2ccad03a9477cf3fd135341642420d18620 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 11 Jul 2016 14:30:52 -0700 Subject: b53: Fix build warning. drivers/net/dsa/b53/b53_srab.c: In function 'b53_srab_probe': >> drivers/net/dsa/b53/b53_srab.c:388:20: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast] pdata->chip_id = (u32)of_id->data; ^ Reported-by: kbuild test robot Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/b53/b53_srab.c b/drivers/net/dsa/b53/b53_srab.c index 2b304ea..3e2d4a5 100644 --- a/drivers/net/dsa/b53/b53_srab.c +++ b/drivers/net/dsa/b53/b53_srab.c @@ -393,7 +393,7 @@ static int b53_srab_probe(struct platform_device *pdev) if (!pdata) return -ENOMEM; - pdata->chip_id = (u32)of_id->data; + pdata->chip_id = (u32)(unsigned long)of_id->data; } priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL); -- cgit v0.10.2 From 12d868964f7352e8b18e755488f7265a93431de1 Mon Sep 17 00:00:00 2001 From: Dmitry Tunin Date: Tue, 12 Jul 2016 01:35:18 +0300 Subject: Bluetooth: Add support of 13d3:3490 AR3012 device T: Bus=01 Lev=01 Prnt=01 Port=07 Cnt=05 Dev#= 5 Spd=12 MxCh= 0 D: Ver= 1.10 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1 P: Vendor=13d3 ProdID=3490 Rev=00.01 C: #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA I: If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb I: If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb BugLink: https://bugs.launchpad.net/bugs/1600623 Signed-off-by: Dmitry Tunin Signed-off-by: Marcel Holtmann Cc: stable@vger.kernel.org diff --git a/drivers/bluetooth/ath3k.c b/drivers/bluetooth/ath3k.c index 2589468..fadba88 100644 --- a/drivers/bluetooth/ath3k.c +++ b/drivers/bluetooth/ath3k.c @@ -123,6 +123,7 @@ static const struct usb_device_id ath3k_table[] = { { USB_DEVICE(0x13d3, 0x3472) }, { USB_DEVICE(0x13d3, 0x3474) }, { USB_DEVICE(0x13d3, 0x3487) }, + { USB_DEVICE(0x13d3, 0x3490) }, /* Atheros AR5BBU12 with sflash firmware */ { USB_DEVICE(0x0489, 0xE02C) }, @@ -190,6 +191,7 @@ static const struct usb_device_id ath3k_blist_tbl[] = { { USB_DEVICE(0x13d3, 0x3472), .driver_info = BTUSB_ATH3012 }, { USB_DEVICE(0x13d3, 0x3474), .driver_info = BTUSB_ATH3012 }, { USB_DEVICE(0x13d3, 0x3487), .driver_info = BTUSB_ATH3012 }, + { USB_DEVICE(0x13d3, 0x3490), .driver_info = BTUSB_ATH3012 }, /* Atheros AR5BBU22 with sflash firmware */ { USB_DEVICE(0x0489, 0xE036), .driver_info = BTUSB_ATH3012 }, diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index f2e8fd7..811f9b9 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -237,6 +237,7 @@ static const struct usb_device_id blacklist_table[] = { { USB_DEVICE(0x13d3, 0x3472), .driver_info = BTUSB_ATH3012 }, { USB_DEVICE(0x13d3, 0x3474), .driver_info = BTUSB_ATH3012 }, { USB_DEVICE(0x13d3, 0x3487), .driver_info = BTUSB_ATH3012 }, + { USB_DEVICE(0x13d3, 0x3490), .driver_info = BTUSB_ATH3012 }, /* Atheros AR5BBU12 with sflash firmware */ { USB_DEVICE(0x0489, 0xe02c), .driver_info = BTUSB_IGNORE }, -- cgit v0.10.2 From 12c2e32f14da857b58af281b029d4549d24c3292 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 12 Jul 2016 00:17:28 +0000 Subject: net: ethernet: bgmac: Fix return value check in bgmac_probe() In case of error, the function devm_ioremap_resource() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR(). Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac-platform.c b/drivers/net/ethernet/broadcom/bgmac-platform.c index 7a8f7ef..1a2d841 100644 --- a/drivers/net/ethernet/broadcom/bgmac-platform.c +++ b/drivers/net/ethernet/broadcom/bgmac-platform.c @@ -141,7 +141,7 @@ static int bgmac_probe(struct platform_device *pdev) } bgmac->plat.idm_base = devm_ioremap_resource(&pdev->dev, regs); - if (!bgmac->plat.idm_base) { + if (IS_ERR(bgmac->plat.idm_base)) { dev_err(&pdev->dev, "Unable to map idm resource\n"); return PTR_ERR(bgmac->plat.idm_base); } -- cgit v0.10.2 From d3fc0353f7c709a6a7fea340211eeb7bbc3e4c66 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Mon, 11 Jul 2016 16:37:51 -0400 Subject: ipv4: af_inet: make it explicitly non-modular The Makefile controlling compilation of this file is obj-y, meaning that it currently is never being built as a module. Since MODULE_ALIAS is a no-op for non-modular code, we can simply remove the MODULE_ALIAS_NETPROTO variant used here. We replace module.h with kmod.h since the file does make use of request_module() in order to load other modules from here. We don't have to worry about init.h coming in via the removed module.h since the file explicitly includes init.h already. Cc: "David S. Miller" Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Cc: netdev@vger.kernel.org Signed-off-by: Paul Gortmaker Signed-off-by: David S. Miller diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index d39e9e4..55513e6 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -73,7 +73,7 @@ #include #include #include -#include +#include #include #include #include @@ -1916,6 +1916,3 @@ static int __init ipv4_proc_init(void) return 0; } #endif /* CONFIG_PROC_FS */ - -MODULE_ALIAS_NETPROTO(PF_INET); - -- cgit v0.10.2 From 3faf56437239326113a3c4b99ab8204e2b09c2e4 Mon Sep 17 00:00:00 2001 From: Walter Mack Date: Mon, 11 Jul 2016 20:02:16 -0700 Subject: mrf24j40: avoid uninitialized byte in SPI transfer to radio. isr function issues SPI read command to mrf to obtain INTSTAT. SPI transfer is 2 bytes, but value of 2nd byte is not defined. This had the effect that only the first ISR worked as intended. The second ISR read incorrect INTSTAT values. Observed on Raspberry PI B+. Signed-off-by: Walter Mack Signed-off-by: Marcel Holtmann diff --git a/drivers/net/ieee802154/mrf24j40.c b/drivers/net/ieee802154/mrf24j40.c index f446db8..7b131f8 100644 --- a/drivers/net/ieee802154/mrf24j40.c +++ b/drivers/net/ieee802154/mrf24j40.c @@ -1054,6 +1054,8 @@ static irqreturn_t mrf24j40_isr(int irq, void *data) disable_irq_nosync(irq); devrec->irq_buf[0] = MRF24J40_READSHORT(REG_INTSTAT); + devrec->irq_buf[1] = 0; + /* Read the interrupt status */ ret = spi_async(devrec->spi, &devrec->irq_msg); if (ret) { -- cgit v0.10.2 From 25f700ef0653d7644ed273f8770230e734cae726 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 11 Jul 2016 22:49:53 +0200 Subject: iwlwifi: add missing type declaration The iwl-debug.h header relies in implicit inclusion of linux/device.h and we get a lot of warnings without that: drivers/net/wireless/intel/iwlwifi/iwl-debug.h:44:23: error: 'struct device' declared inside parameter list will not be visible outside of this definition or declaration [-Werror] void __iwl_err(struct device *dev, bool rfkill_prefix, bool only_trace, ^~~~~~ In file included from drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.h:66:0, from drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.c:68: drivers/net/wireless/intel/iwlwifi/iwl-trans.h: In function 'iwl_trans_tx': drivers/net/wireless/intel/iwlwifi/iwl-trans.h:1030:348: error: passing argument 1 of '__iwl_err' from incompatible pointer type [-Werror=incompatible-pointer-types] IWL_ERR(trans, "%s bad state = %d\n", __func__, trans->state); ^ In file included from drivers/net/wireless/intel/iwlwifi/iwl-eeprom-read.c:67:0: drivers/net/wireless/intel/iwlwifi/iwl-debug.h:44:6: note: expected 'struct device *' but argument is of type 'struct device *' void __iwl_err(struct device *dev, bool rfkill_prefix, bool only_trace, ^~~~~~~~~ The easiest workaround is to just declare 'struct device' before its first use, rather than including the entire header file. Signed-off-by: Arnd Bergmann Fixes: 21cb3222fe56 ("iwlwifi: decouple PCIe transport from mac80211") Acked-by: Luca Coelho Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-debug.h b/drivers/net/wireless/intel/iwlwifi/iwl-debug.h index 1103332..cd77c69 100644 --- a/drivers/net/wireless/intel/iwlwifi/iwl-debug.h +++ b/drivers/net/wireless/intel/iwlwifi/iwl-debug.h @@ -41,6 +41,7 @@ static inline bool iwl_have_debug_level(u32 level) #endif } +struct device; void __iwl_err(struct device *dev, bool rfkill_prefix, bool only_trace, const char *fmt, ...) __printf(4, 5); void __iwl_warn(struct device *dev, const char *fmt, ...) __printf(2, 3); -- cgit v0.10.2 From 3101e0fc1f6e809d38fbb5845c6c5eb0eefeda07 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Tue, 12 Jul 2016 19:45:00 +0800 Subject: netfilter: conntrack: protect early_drop by rcu read lock User can add ct entry via nfnetlink(IPCTNL_MSG_CT_NEW), and if the total number reach the nf_conntrack_max, we will try to drop some ct entries. But in this case(the main function call path is ctnetlink_create_conntrack -> nf_conntrack_alloc -> early_drop), rcu_read_lock is not held, so race with hash resize will happen. Fixes: 242922a02717 ("netfilter: conntrack: simplify early_drop") Cc: Florian Westphal Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index e0e9c9a..2d46225 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -880,6 +880,7 @@ static noinline int early_drop(struct net *net, unsigned int _hash) struct hlist_nulls_head *ct_hash; unsigned hash, sequence, drops; + rcu_read_lock(); do { sequence = read_seqcount_begin(&nf_conntrack_generation); hash = scale_hash(_hash++); @@ -887,6 +888,8 @@ static noinline int early_drop(struct net *net, unsigned int _hash) } while (read_seqcount_retry(&nf_conntrack_generation, sequence)); drops = early_drop_list(net, &ct_hash[hash]); + rcu_read_unlock(); + if (drops) { NF_CT_STAT_ADD_ATOMIC(net, early_drop, drops); return true; -- cgit v0.10.2 From 178cd55f086629cf0bad9c66c793a7e2bcc3abb6 Mon Sep 17 00:00:00 2001 From: Haiyang Zhang Date: Mon, 11 Jul 2016 17:06:42 -0700 Subject: tools: hv: Add a script to help bonding synthetic and VF NICs This script helps to create bonding network devices based on synthetic NIC (the virtual network adapter usually provided by Hyper-V) and the matching VF NIC (SRIOV virtual function). So the synthetic NIC and VF NIC can function as one network device, and fail over to the synthetic NIC if VF is down. Mayjor distros (RHEL, Ubuntu, SLES) supported by Hyper-V are supported by this script. Signed-off-by: Haiyang Zhang Reviewed-by: K. Y. Srinivasan Signed-off-by: David S. Miller diff --git a/tools/hv/bondvf.sh b/tools/hv/bondvf.sh new file mode 100755 index 0000000..8e96023 --- /dev/null +++ b/tools/hv/bondvf.sh @@ -0,0 +1,193 @@ +#!/bin/bash + +# This example script creates bonding network devices based on synthetic NIC +# (the virtual network adapter usually provided by Hyper-V) and the matching +# VF NIC (SRIOV virtual function). So the synthetic NIC and VF NIC can +# function as one network device, and fail over to the synthetic NIC if VF is +# down. +# +# Usage: +# - After configured vSwitch and vNIC with SRIOV, start Linux virtual +# machine (VM) +# - Run this scripts on the VM. It will create configuration files in +# distro specific directory. +# - Reboot the VM, so that the bonding config are enabled. +# +# The config files are DHCP by default. You may edit them if you need to change +# to Static IP or change other settings. +# + +sysdir=/sys/class/net +netvsc_cls={f8615163-df3e-46c5-913f-f2d2f965ed0e} +bondcnt=0 + +# Detect Distro +if [ -f /etc/redhat-release ]; +then + cfgdir=/etc/sysconfig/network-scripts + distro=redhat +elif grep -q 'Ubuntu' /etc/issue +then + cfgdir=/etc/network + distro=ubuntu +elif grep -q 'SUSE' /etc/issue +then + cfgdir=/etc/sysconfig/network + distro=suse +else + echo "Unsupported Distro" + exit 1 +fi + +echo Detected Distro: $distro, or compatible + +# Get a list of ethernet names +list_eth=(`cd $sysdir && ls -d */ | cut -d/ -f1 | grep -v bond`) +eth_cnt=${#list_eth[@]} + +echo List of net devices: + +# Get the MAC addresses +for (( i=0; i < $eth_cnt; i++ )) +do + list_mac[$i]=`cat $sysdir/${list_eth[$i]}/address` + echo ${list_eth[$i]}, ${list_mac[$i]} +done + +# Find NIC with matching MAC +for (( i=0; i < $eth_cnt-1; i++ )) +do + for (( j=i+1; j < $eth_cnt; j++ )) + do + if [ "${list_mac[$i]}" = "${list_mac[$j]}" ] + then + list_match[$i]=${list_eth[$j]} + break + fi + done +done + +function create_eth_cfg_redhat { + local fn=$cfgdir/ifcfg-$1 + + rm -f $fn + echo DEVICE=$1 >>$fn + echo TYPE=Ethernet >>$fn + echo BOOTPROTO=none >>$fn + echo ONBOOT=yes >>$fn + echo NM_CONTROLLED=no >>$fn + echo PEERDNS=yes >>$fn + echo IPV6INIT=yes >>$fn + echo MASTER=$2 >>$fn + echo SLAVE=yes >>$fn +} + +function create_eth_cfg_pri_redhat { + create_eth_cfg_redhat $1 $2 +} + +function create_bond_cfg_redhat { + local fn=$cfgdir/ifcfg-$1 + + rm -f $fn + echo DEVICE=$1 >>$fn + echo TYPE=Bond >>$fn + echo BOOTPROTO=dhcp >>$fn + echo ONBOOT=yes >>$fn + echo NM_CONTROLLED=no >>$fn + echo PEERDNS=yes >>$fn + echo IPV6INIT=yes >>$fn + echo BONDING_MASTER=yes >>$fn + echo BONDING_OPTS=\"mode=active-backup miimon=100 primary=$2\" >>$fn +} + +function create_eth_cfg_ubuntu { + local fn=$cfgdir/interfaces + + echo $'\n'auto $1 >>$fn + echo iface $1 inet manual >>$fn + echo bond-master $2 >>$fn +} + +function create_eth_cfg_pri_ubuntu { + local fn=$cfgdir/interfaces + + create_eth_cfg_ubuntu $1 $2 + echo bond-primary $1 >>$fn +} + +function create_bond_cfg_ubuntu { + local fn=$cfgdir/interfaces + + echo $'\n'auto $1 >>$fn + echo iface $1 inet dhcp >>$fn + echo bond-mode active-backup >>$fn + echo bond-miimon 100 >>$fn + echo bond-slaves none >>$fn +} + +function create_eth_cfg_suse { + local fn=$cfgdir/ifcfg-$1 + + rm -f $fn + echo BOOTPROTO=none >>$fn + echo STARTMODE=auto >>$fn +} + +function create_eth_cfg_pri_suse { + create_eth_cfg_suse $1 +} + +function create_bond_cfg_suse { + local fn=$cfgdir/ifcfg-$1 + + rm -f $fn + echo BOOTPROTO=dhcp >>$fn + echo STARTMODE=auto >>$fn + echo BONDING_MASTER=yes >>$fn + echo BONDING_SLAVE_0=$2 >>$fn + echo BONDING_SLAVE_1=$3 >>$fn + echo BONDING_MODULE_OPTS=\'mode=active-backup miimon=100 primary=$2\' >>$fn +} + +function create_bond { + local bondname=bond$bondcnt + local primary + local secondary + + local class_id1=`cat $sysdir/$1/device/class_id 2>/dev/null` + local class_id2=`cat $sysdir/$2/device/class_id 2>/dev/null` + + if [ "$class_id1" = "$netvsc_cls" ] + then + primary=$2 + secondary=$1 + elif [ "$class_id2" = "$netvsc_cls" ] + then + primary=$1 + secondary=$2 + else + return 0 + fi + + echo $'\nBond name:' $bondname + + echo configuring $primary + create_eth_cfg_pri_$distro $primary $bondname + + echo configuring $secondary + create_eth_cfg_$distro $secondary $bondname + + echo creating: $bondname with primary slave: $primary + create_bond_cfg_$distro $bondname $primary $secondary + + let bondcnt=bondcnt+1 +} + +for (( i=0; i < $eth_cnt-1; i++ )) +do + if [ -n "${list_match[$i]}" ] + then + create_bond ${list_eth[$i]} ${list_match[$i]} + fi +done -- cgit v0.10.2 From 2b8fb41844ec9d6a437b0009f777039d845b1b04 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 12 Jul 2016 11:00:09 +0000 Subject: stmmac: dwmac-socfpga: fix wrong pointer passed to PTR_ERR() PTR_ERR should access the value just tested by IS_ERR, otherwise the wrong error code will be returned. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index 4bee2f9..3bc1fa2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -214,7 +214,7 @@ static int socfpga_dwmac_parse_data(struct socfpga_dwmac *dwmac, struct device * dev_err(dev, "%s: ERROR: failed mapping tse control port\n", __func__); - return PTR_ERR(dwmac->pcs.sgmii_adapter_base); + return PTR_ERR(dwmac->pcs.tse_pcs_base); } } } -- cgit v0.10.2 From 8addc0440bdba9280bf212b70f17f029c2801805 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 12 Jul 2016 11:21:17 +0000 Subject: rxrpc: Fix error handling in af_rxrpc_init() security initialized after alloc workqueue, so we should exit security before destroy workqueue in the error handing. Fixes: 648af7fca159 ("rxrpc: Absorb the rxkad security module") Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index d6e4e3b..88effad 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -766,9 +766,9 @@ error_key_type: error_sock: proto_unregister(&rxrpc_proto); error_proto: - destroy_workqueue(rxrpc_workqueue); -error_security: rxrpc_exit_security(); +error_security: + destroy_workqueue(rxrpc_workqueue); error_work_queue: kmem_cache_destroy(rxrpc_call_jar); error_call_jar: -- cgit v0.10.2 From 379672deb2dcb567e2a425a816bede8df476ae4e Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 12 Jul 2016 11:36:44 +0000 Subject: net: mediatek: fix non static symbol warnings Fixes the following sparse warnings: drivers/net/ethernet/mediatek/mtk_eth_soc.c:79:5: warning: symbol '_mtk_mdio_write' was not declared. Should it be static? drivers/net/ethernet/mediatek/mtk_eth_soc.c:98:5: warning: symbol '_mtk_mdio_read' was not declared. Should it be static? Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 760f3d7..b57ae3a 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -76,8 +76,8 @@ static int mtk_mdio_busy_wait(struct mtk_eth *eth) return -1; } -u32 _mtk_mdio_write(struct mtk_eth *eth, u32 phy_addr, - u32 phy_register, u32 write_data) +static u32 _mtk_mdio_write(struct mtk_eth *eth, u32 phy_addr, + u32 phy_register, u32 write_data) { if (mtk_mdio_busy_wait(eth)) return -1; @@ -95,7 +95,7 @@ u32 _mtk_mdio_write(struct mtk_eth *eth, u32 phy_addr, return 0; } -u32 _mtk_mdio_read(struct mtk_eth *eth, int phy_addr, int phy_reg) +static u32 _mtk_mdio_read(struct mtk_eth *eth, int phy_addr, int phy_reg) { u32 d; -- cgit v0.10.2 From a6acf689328d7d4e63a66c8296f988ad8dc15752 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 12 Jul 2016 11:43:37 +0000 Subject: dwc_eth_qos: fix missing clk_disable_unprepare() on error in dwceqos_probe() Fix missing clk_disable_unprepare() call before return from dwceqos_probe() in the error handling case of invalid fixed-link. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/synopsys/dwc_eth_qos.c b/drivers/net/ethernet/synopsys/dwc_eth_qos.c index c34111b..fc1ea80 100644 --- a/drivers/net/ethernet/synopsys/dwc_eth_qos.c +++ b/drivers/net/ethernet/synopsys/dwc_eth_qos.c @@ -2877,7 +2877,7 @@ static int dwceqos_probe(struct platform_device *pdev) ret = of_phy_register_fixed_link(lp->pdev->dev.of_node); if (ret < 0) { dev_err(&pdev->dev, "invalid fixed-link"); - goto err_out_unregister_netdev; + goto err_out_unregister_clk_notifier; } lp->phy_node = of_node_get(lp->pdev->dev.of_node); -- cgit v0.10.2 From 85c22bad56ec0d0d8d8518414f97106233132e43 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 12 Jul 2016 15:24:10 +0000 Subject: net: dsa: Fix non static symbol warning Fixes the following sparse warning: net/dsa/dsa2.c:680:6: warning: symbol '_dsa_unregister_switch' was not declared. Should it be static? Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 78e4c01..f30bad9 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -677,7 +677,7 @@ int dsa_register_switch(struct dsa_switch *ds, struct device_node *np) } EXPORT_SYMBOL_GPL(dsa_register_switch); -void _dsa_unregister_switch(struct dsa_switch *ds) +static void _dsa_unregister_switch(struct dsa_switch *ds) { struct dsa_switch_tree *dst = ds->dst; -- cgit v0.10.2 From e5224f0fe2acddbc2fa9b419d8867ced7f5381fc Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 12 Jul 2016 18:05:03 +0200 Subject: devlink: add hardware messages tracing facility Define a tracepoint and allow user to trace messages going to and from hardware associated with devlink instance. Signed-off-by: Jiri Pirko Acked-by: Steven Rostedt Signed-off-by: David S. Miller diff --git a/include/trace/events/devlink.h b/include/trace/events/devlink.h new file mode 100644 index 0000000..333c32a --- /dev/null +++ b/include/trace/events/devlink.h @@ -0,0 +1,68 @@ +#if IS_ENABLED(CONFIG_NET_DEVLINK) + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM devlink + +#if !defined(_TRACE_DEVLINK_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_DEVLINK_H + +#include +#include +#include + +/* + * Tracepoint for devlink hardware message: + */ +TRACE_EVENT(devlink_hwmsg, + TP_PROTO(const struct devlink *devlink, bool incoming, + unsigned long type, const u8 *buf, size_t len), + + TP_ARGS(devlink, incoming, type, buf, len), + + TP_STRUCT__entry( + __string(bus_name, devlink->dev->bus->name) + __string(dev_name, dev_name(devlink->dev)) + __string(owner_name, devlink->dev->driver->owner->name) + __field(bool, incoming) + __field(unsigned long, type) + __dynamic_array(u8, buf, len) + __field(size_t, len) + ), + + TP_fast_assign( + __assign_str(bus_name, devlink->dev->bus->name); + __assign_str(dev_name, dev_name(devlink->dev)); + __assign_str(owner_name, devlink->dev->driver->owner->name); + __entry->incoming = incoming; + __entry->type = type; + memcpy(__get_dynamic_array(buf), buf, len); + __entry->len = len; + ), + + TP_printk("bus_name=%s dev_name=%s owner_name=%s incoming=%d type=%lu buf=0x[%*phD] len=%lu", + __get_str(bus_name), __get_str(dev_name), + __get_str(owner_name), __entry->incoming, __entry->type, + (int) __entry->len, __get_dynamic_array(buf), __entry->len) +); + +#endif /* _TRACE_DEVLINK_H */ + +/* This part must be outside protection */ +#include + +#else /* CONFIG_NET_DEVLINK */ + +#if !defined(_TRACE_DEVLINK_H) +#define _TRACE_DEVLINK_H + +#include + +static inline void trace_devlink_hwmsg(const struct devlink *devlink, + bool incoming, unsigned long type, + const u8 *buf, size_t len) +{ +} + +#endif /* _TRACE_DEVLINK_H */ + +#endif diff --git a/net/core/devlink.c b/net/core/devlink.c index b2e592a..1b50630 100644 --- a/net/core/devlink.c +++ b/net/core/devlink.c @@ -26,6 +26,10 @@ #include #include #include +#define CREATE_TRACE_POINTS +#include + +EXPORT_TRACEPOINT_SYMBOL_GPL(devlink_hwmsg); static LIST_HEAD(devlink_list); -- cgit v0.10.2 From b38a75d2d32412477e2625211febb98b63087754 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 12 Jul 2016 18:05:04 +0200 Subject: mlxsw: core: Trace EMAD messages Trace EMAD messages going down to HW and up from HW. Devlink needs to be registered before EMAD init so the trace function can be called with valid devlink handle. Signed-off-by: Jiri Pirko v1->v2: - Use trace_devlink_hwmsg directly Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index 01ae548..480a3ba 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -58,6 +58,7 @@ #include #include #include +#include #include "core.h" #include "item.h" @@ -447,6 +448,10 @@ static int mlxsw_emad_transmit(struct mlxsw_core *mlxsw_core, if (!skb) return -ENOMEM; + trace_devlink_hwmsg(priv_to_devlink(mlxsw_core), false, 0, + skb->data + mlxsw_core->driver->txhdr_len, + skb->len - mlxsw_core->driver->txhdr_len); + atomic_set(&trans->active, 1); err = mlxsw_core_skb_transmit(mlxsw_core, skb, &trans->tx_info); if (err) { @@ -529,6 +534,9 @@ static void mlxsw_emad_rx_listener_func(struct sk_buff *skb, u8 local_port, struct mlxsw_core *mlxsw_core = priv; struct mlxsw_reg_trans *trans; + trace_devlink_hwmsg(priv_to_devlink(mlxsw_core), true, 0, + skb->data, skb->len); + if (!mlxsw_emad_is_resp(skb)) goto free_skb; @@ -1110,14 +1118,14 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info, if (err) goto err_emad_init; - err = mlxsw_hwmon_init(mlxsw_core, mlxsw_bus_info, &mlxsw_core->hwmon); - if (err) - goto err_hwmon_init; - err = devlink_register(devlink, mlxsw_bus_info->dev); if (err) goto err_devlink_register; + err = mlxsw_hwmon_init(mlxsw_core, mlxsw_bus_info, &mlxsw_core->hwmon); + if (err) + goto err_hwmon_init; + err = mlxsw_driver->init(mlxsw_core, mlxsw_bus_info); if (err) goto err_driver_init; @@ -1131,9 +1139,9 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info, err_debugfs_init: mlxsw_core->driver->fini(mlxsw_core); err_driver_init: +err_hwmon_init: devlink_unregister(devlink); err_devlink_register: -err_hwmon_init: mlxsw_emad_fini(mlxsw_core); err_emad_init: mlxsw_bus->fini(bus_priv); -- cgit v0.10.2 From 160b925163c0aabc2c2fbb7d58a75e38b7cd6a17 Mon Sep 17 00:00:00 2001 From: Szymon Janc Date: Tue, 12 Jul 2016 02:12:16 +0200 Subject: Bluetooth: Add Authentication Failed reason to Disconnected Mgmt event If link is disconnected due to Authentication Failure (PIN or Key Missing status) userspace will be notified about this with proper error code. Many LE profiles define "PIN or Key Missing" status as indication of remote lost bond so this allows userspace to take action on this. @ Device Connected: 88:63:DF:88:0E:83 (1) flags 0x0000 02 01 1a 05 03 0a 18 0d 18 0b 09 48 65 61 72 74 ...........Heart 20 52 61 74 65 Rate > HCI Event: Command Status (0x0f) plen 4 LE Read Remote Used Features (0x08|0x0016) ncmd 1 Status: Success (0x00) > ACL Data RX: Handle 3585 flags 0x02 dlen 11 ATT: Read By Group Type Request (0x10) len 6 Handle range: 0x0001-0xffff Attribute group type: Primary Service (0x2800) > HCI Event: LE Meta Event (0x3e) plen 12 LE Read Remote Used Features (0x04) Status: Success (0x00) Handle: 3585 Features: 0x01 0x00 0x00 0x00 0x00 0x00 0x00 0x00 LE Encryption < HCI Command: LE Start Encryption (0x08|0x0019) plen 28 Handle: 3585 Random number: 0x0000000000000000 Encrypted diversifier: 0x0000 Long term key: 26201cd479a0921b6f949f0b1fa8dc82 > HCI Event: Command Status (0x0f) plen 4 LE Start Encryption (0x08|0x0019) ncmd 1 Status: Success (0x00) > HCI Event: Encryption Change (0x08) plen 4 Status: PIN or Key Missing (0x06) Handle: 3585 Encryption: Disabled (0x00) < HCI Command: Disconnect (0x01|0x0006) plen 3 Handle: 3585 Reason: Authentication Failure (0x05) > HCI Event: Command Status (0x0f) plen 4 Disconnect (0x01|0x0006) ncmd 1 Status: Success (0x00) > HCI Event: Disconnect Complete (0x05) plen 4 Status: Success (0x00) Handle: 3585 Reason: Connection Terminated By Local Host (0x16) @ Device Disconnected: 88:63:DF:88:0E:83 (1) reason 4 @ Device Connected: C4:43:8F:A3:4D:83 (0) flags 0x0000 08 09 4e 65 78 75 73 20 35 ..Nexus 5 > HCI Event: Command Status (0x0f) plen 4 Authentication Requested (0x01|0x0011) ncmd 1 Status: Success (0x00) > HCI Event: Link Key Request (0x17) plen 6 Address: C4:43:8F:A3:4D:83 (LG Electronics) < HCI Command: Link Key Request Reply (0x01|0x000b) plen 22 Address: C4:43:8F:A3:4D:83 (LG Electronics) Link key: 080812e4aa97a863d11826f71f65a933 > HCI Event: Command Complete (0x0e) plen 10 Link Key Request Reply (0x01|0x000b) ncmd 1 Status: Success (0x00) Address: C4:43:8F:A3:4D:83 (LG Electronics) > HCI Event: Auth Complete (0x06) plen 3 Status: PIN or Key Missing (0x06) Handle: 75 @ Authentication Failed: C4:43:8F:A3:4D:83 (0) status 0x05 < HCI Command: Disconnect (0x01|0x0006) plen 3 Handle: 75 Reason: Remote User Terminated Connection (0x13) > HCI Event: Command Status (0x0f) plen 4 Disconnect (0x01|0x0006) ncmd 1 Status: Success (0x00) > HCI Event: Disconnect Complete (0x05) plen 4 Status: Success (0x00) Handle: 75 Reason: Connection Terminated By Local Host (0x16) @ Device Disconnected: C4:43:8F:A3:4D:83 (0) reason 4 Signed-off-by: Szymon Janc Signed-off-by: Johan Hedberg diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index a3f86de..003b252 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -445,6 +445,7 @@ enum { /* ---- HCI Error Codes ---- */ #define HCI_ERROR_UNKNOWN_CONN_ID 0x02 #define HCI_ERROR_AUTH_FAILURE 0x05 +#define HCI_ERROR_PIN_OR_KEY_MISSING 0x06 #define HCI_ERROR_MEMORY_EXCEEDED 0x07 #define HCI_ERROR_CONNECTION_TIMEOUT 0x08 #define HCI_ERROR_REJ_LIMITED_RESOURCES 0x0d diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index dc71473..77d7fe1 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -654,6 +654,7 @@ enum { HCI_CONN_PARAM_REMOVAL_PEND, HCI_CONN_NEW_LINK_KEY, HCI_CONN_SCANNING, + HCI_CONN_AUTH_FAILURE, }; static inline bool hci_conn_ssp_enabled(struct hci_conn *conn) diff --git a/include/net/bluetooth/mgmt.h b/include/net/bluetooth/mgmt.h index ea73e08..7647964 100644 --- a/include/net/bluetooth/mgmt.h +++ b/include/net/bluetooth/mgmt.h @@ -645,6 +645,7 @@ struct mgmt_ev_device_connected { #define MGMT_DEV_DISCONN_TIMEOUT 0x01 #define MGMT_DEV_DISCONN_LOCAL_HOST 0x02 #define MGMT_DEV_DISCONN_REMOTE 0x03 +#define MGMT_DEV_DISCONN_AUTH_FAILURE 0x04 #define MGMT_EV_DEVICE_DISCONNECTED 0x000C struct mgmt_ev_device_disconnected { diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 3fb95c4..e17aacb 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -2332,7 +2332,7 @@ static u8 hci_to_mgmt_reason(u8 err) static void hci_disconn_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) { struct hci_ev_disconn_complete *ev = (void *) skb->data; - u8 reason = hci_to_mgmt_reason(ev->reason); + u8 reason; struct hci_conn_params *params; struct hci_conn *conn; bool mgmt_connected; @@ -2355,6 +2355,12 @@ static void hci_disconn_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) conn->state = BT_CLOSED; mgmt_connected = test_and_clear_bit(HCI_CONN_MGMT_CONNECTED, &conn->flags); + + if (test_bit(HCI_CONN_AUTH_FAILURE, &conn->flags)) + reason = MGMT_DEV_DISCONN_AUTH_FAILURE; + else + reason = hci_to_mgmt_reason(ev->reason); + mgmt_device_disconnected(hdev, &conn->dst, conn->type, conn->dst_type, reason, mgmt_connected); @@ -2421,6 +2427,8 @@ static void hci_auth_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) goto unlock; if (!ev->status) { + clear_bit(HCI_CONN_AUTH_FAILURE, &conn->flags); + if (!hci_conn_ssp_enabled(conn) && test_bit(HCI_CONN_REAUTH_PEND, &conn->flags)) { BT_INFO("re-auth of legacy device is not possible."); @@ -2429,6 +2437,9 @@ static void hci_auth_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) conn->sec_level = conn->pending_sec_level; } } else { + if (ev->status == HCI_ERROR_PIN_OR_KEY_MISSING) + set_bit(HCI_CONN_AUTH_FAILURE, &conn->flags); + mgmt_auth_failed(conn, ev->status); } @@ -2613,6 +2624,9 @@ static void hci_encrypt_change_evt(struct hci_dev *hdev, struct sk_buff *skb) clear_bit(HCI_CONN_ENCRYPT_PEND, &conn->flags); if (ev->status && conn->state == BT_CONNECTED) { + if (ev->status == HCI_ERROR_PIN_OR_KEY_MISSING) + set_bit(HCI_CONN_AUTH_FAILURE, &conn->flags); + hci_disconnect(conn, HCI_ERROR_AUTH_FAILURE); hci_conn_drop(conn); goto unlock; -- cgit v0.10.2 From 87510973d6e137c33552b3365b5afbd5be81c5dd Mon Sep 17 00:00:00 2001 From: Johan Hedberg Date: Wed, 13 Jul 2016 10:57:18 +0300 Subject: Bluetooth: Increment management interface revision Increment the mgmt revision due to the recently added new reason code for the Disconnected event. Signed-off-by: Johan Hedberg Signed-off-by: Marcel Holtmann diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 7983ec8..7639290 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -38,7 +38,7 @@ #include "mgmt_util.h" #define MGMT_VERSION 1 -#define MGMT_REVISION 12 +#define MGMT_REVISION 13 static const u16 mgmt_commands[] = { MGMT_OP_READ_INDEX_LIST, -- cgit v0.10.2 From 6e07653765bdfc1762b874509001213860b609e4 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Tue, 12 Jul 2016 15:04:23 -0600 Subject: net: vrf: Documentation update Update vrf documentation for changes made to 4.4 - 4.8 kernels and iproute2 support for vrf keyword. Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/Documentation/networking/vrf.txt b/Documentation/networking/vrf.txt index 5da679c..11a2b99 100644 --- a/Documentation/networking/vrf.txt +++ b/Documentation/networking/vrf.txt @@ -15,9 +15,9 @@ the use of higher priority ip rules (Policy Based Routing, PBR) to take precedence over the VRF device rules directing specific traffic as desired. In addition, VRF devices allow VRFs to be nested within namespaces. For -example network namespaces provide separation of network interfaces at L1 -(Layer 1 separation), VLANs on the interfaces within a namespace provide -L2 separation and then VRF devices provide L3 separation. +example network namespaces provide separation of network interfaces at the +device layer, VLANs on the interfaces within a namespace provide L2 separation +and then VRF devices provide L3 separation. Design ------ @@ -37,21 +37,22 @@ are then enslaved to a VRF device: +------+ +------+ Packets received on an enslaved device and are switched to the VRF device -using an rx_handler which gives the impression that packets flow through -the VRF device. Similarly on egress routing rules are used to send packets -to the VRF device driver before getting sent out the actual interface. This -allows tcpdump on a VRF device to capture all packets into and out of the -VRF as a whole.[1] Similarly, netfilter [2] and tc rules can be applied -using the VRF device to specify rules that apply to the VRF domain as a whole. +in the IPv4 and IPv6 processing stacks giving the impression that packets +flow through the VRF device. Similarly on egress routing rules are used to +send packets to the VRF device driver before getting sent out the actual +interface. This allows tcpdump on a VRF device to capture all packets into +and out of the VRF as a whole.[1] Similarly, netfilter[2] and tc rules can be +applied using the VRF device to specify rules that apply to the VRF domain +as a whole. [1] Packets in the forwarded state do not flow through the device, so those packets are not seen by tcpdump. Will revisit this limitation in a future release. -[2] Iptables on ingress is limited to NF_INET_PRE_ROUTING only with skb->dev - set to real ingress device and egress is limited to NF_INET_POST_ROUTING. - Will revisit this limitation in a future release. - +[2] Iptables on ingress supports PREROUTING with skb->dev set to the real + ingress device and both INPUT and PREROUTING rules with skb->dev set to + the VRF device. For egress POSTROUTING and OUTPUT rules can be written + using either the VRF device or real egress device. Setup ----- @@ -59,23 +60,33 @@ Setup e.g, ip link add vrf-blue type vrf table 10 ip link set dev vrf-blue up -2. Rules are added that send lookups to the associated FIB table when the - iif or oif is the VRF device. e.g., +2. An l3mdev FIB rule directs lookups to the table associated with the device. + A single l3mdev rule is sufficient for all VRFs. The VRF device adds the + l3mdev rule for IPv4 and IPv6 when the first device is created with a + default preference of 1000. Users may delete the rule if desired and add + with a different priority or install per-VRF rules. + + Prior to the v4.8 kernel iif and oif rules are needed for each VRF device: ip ru add oif vrf-blue table 10 ip ru add iif vrf-blue table 10 - Set the default route for the table (and hence default route for the VRF). - e.g, ip route add table 10 prohibit default +3. Set the default route for the table (and hence default route for the VRF). + ip route add table 10 unreachable default -3. Enslave L3 interfaces to a VRF device. - e.g, ip link set dev eth1 master vrf-blue +4. Enslave L3 interfaces to a VRF device. + ip link set dev eth1 master vrf-blue Local and connected routes for enslaved devices are automatically moved to the table associated with VRF device. Any additional routes depending on - the enslaved device will need to be reinserted following the enslavement. + the enslaved device are dropped and will need to be reinserted to the VRF + FIB table following the enslavement. + + The IPv6 sysctl option keep_addr_on_down can be enabled to keep IPv6 global + addresses as VRF enslavement changes. + sysctl -w net.ipv6.conf.all.keep_addr_on_down=1 -4. Additional VRF routes are added to associated table. - e.g., ip route add table 10 ... +5. Additional VRF routes are added to associated table. + ip route add table 10 ... Applications @@ -87,39 +98,34 @@ VRF device: or to specify the output device using cmsg and IP_PKTINFO. +TCP services running in the default VRF context (ie., not bound to any VRF +device) can work across all VRF domains by enabling the tcp_l3mdev_accept +sysctl option: + sysctl -w net.ipv4.tcp_l3mdev_accept=1 -Limitations ------------ -Index of original ingress interface is not available via cmsg. Will address -soon. +netfilter rules on the VRF device can be used to limit access to services +running in the default VRF context as well. + +The default VRF does not have limited scope with respect to port bindings. +That is, if a process does a wildcard bind to a port in the default VRF it +owns the port across all VRF domains within the network namespace. ################################################################################ Using iproute2 for VRFs ======================= -VRF devices do *not* have to start with 'vrf-'. That is a convention used here -for emphasis of the device type, similar to use of 'br' in bridge names. +iproute2 supports the vrf keyword as of v4.7. For backwards compatibility this +section lists both commands where appropriate -- with the vrf keyword and the +older form without it. 1. Create a VRF To instantiate a VRF device and associate it with a table: $ ip link add dev NAME type vrf table ID - Remember to add the ip rules as well: - $ ip ru add oif NAME table 10 - $ ip ru add iif NAME table 10 - $ ip -6 ru add oif NAME table 10 - $ ip -6 ru add iif NAME table 10 - - Without the rules route lookups are not directed to the table. - - For example: - $ ip link add dev vrf-blue type vrf table 10 - $ ip ru add pref 200 oif vrf-blue table 10 - $ ip ru add pref 200 iif vrf-blue table 10 - $ ip -6 ru add pref 200 oif vrf-blue table 10 - $ ip -6 ru add pref 200 iif vrf-blue table 10 - + As of v4.8 the kernel supports the l3mdev FIB rule where a single rule + covers all VRFs. The l3mdev rule is created for IPv4 and IPv6 on first + device create. 2. List VRFs @@ -129,16 +135,16 @@ for emphasis of the device type, similar to use of 'br' in bridge names. For example: $ ip -d link show type vrf - 11: vrf-mgmt: mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000 + 11: mgmt: mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000 link/ether 72:b3:ba:91:e2:24 brd ff:ff:ff:ff:ff:ff promiscuity 0 vrf table 1 addrgenmode eui64 - 12: vrf-red: mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000 + 12: red: mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000 link/ether b6:6f:6e:f6:da:73 brd ff:ff:ff:ff:ff:ff promiscuity 0 vrf table 10 addrgenmode eui64 - 13: vrf-blue: mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000 + 13: blue: mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000 link/ether 36:62:e8:7d:bb:8c brd ff:ff:ff:ff:ff:ff promiscuity 0 vrf table 66 addrgenmode eui64 - 14: vrf-green: mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000 + 14: green: mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000 link/ether e6:28:b8:63:70:bb brd ff:ff:ff:ff:ff:ff promiscuity 0 vrf table 81 addrgenmode eui64 @@ -146,43 +152,44 @@ for emphasis of the device type, similar to use of 'br' in bridge names. Or in brief output: $ ip -br link show type vrf - vrf-mgmt UP 72:b3:ba:91:e2:24 - vrf-red UP b6:6f:6e:f6:da:73 - vrf-blue UP 36:62:e8:7d:bb:8c - vrf-green UP e6:28:b8:63:70:bb + mgmt UP 72:b3:ba:91:e2:24 + red UP b6:6f:6e:f6:da:73 + blue UP 36:62:e8:7d:bb:8c + green UP e6:28:b8:63:70:bb 3. Assign a Network Interface to a VRF Network interfaces are assigned to a VRF by enslaving the netdevice to a VRF device: - $ ip link set dev NAME master VRF-NAME + $ ip link set dev NAME master NAME On enslavement connected and local routes are automatically moved to the table associated with the VRF device. For example: - $ ip link set dev eth0 master vrf-mgmt + $ ip link set dev eth0 master mgmt 4. Show Devices Assigned to a VRF To show devices that have been assigned to a specific VRF add the master option to the ip command: - $ ip link show master VRF-NAME + $ ip link show vrf NAME + $ ip link show master NAME For example: - $ ip link show master vrf-red - 3: eth1: mtu 1500 qdisc pfifo_fast master vrf-red state UP mode DEFAULT group default qlen 1000 + $ ip link show vrf red + 3: eth1: mtu 1500 qdisc pfifo_fast master red state UP mode DEFAULT group default qlen 1000 link/ether 02:00:00:00:02:02 brd ff:ff:ff:ff:ff:ff - 4: eth2: mtu 1500 qdisc pfifo_fast master vrf-red state UP mode DEFAULT group default qlen 1000 + 4: eth2: mtu 1500 qdisc pfifo_fast master red state UP mode DEFAULT group default qlen 1000 link/ether 02:00:00:00:02:03 brd ff:ff:ff:ff:ff:ff - 7: eth5: mtu 1500 qdisc noop master vrf-red state DOWN mode DEFAULT group default qlen 1000 + 7: eth5: mtu 1500 qdisc noop master red state DOWN mode DEFAULT group default qlen 1000 link/ether 02:00:00:00:02:06 brd ff:ff:ff:ff:ff:ff Or using the brief output: - $ ip -br link show master vrf-red + $ ip -br link show master red eth1 UP 02:00:00:00:02:02 eth2 UP 02:00:00:00:02:03 eth5 DOWN 02:00:00:00:02:06 @@ -192,14 +199,15 @@ for emphasis of the device type, similar to use of 'br' in bridge names. To list neighbor entries associated with devices enslaved to a VRF device add the master option to the ip command: - $ ip [-6] neigh show master VRF-NAME + $ ip [-6] neigh show vrf NAME + $ ip [-6] neigh show master NAME For example: - $ ip neigh show master vrf-red + $ ip neigh show vrf red 10.2.1.254 dev eth1 lladdr a6:d9:c7:4f:06:23 REACHABLE 10.2.2.254 dev eth2 lladdr 5e:54:01:6a:ee:80 REACHABLE - $ ip -6 neigh show master vrf-red + $ ip -6 neigh show vrf red 2002:1::64 dev eth1 lladdr a6:d9:c7:4f:06:23 REACHABLE @@ -207,11 +215,12 @@ for emphasis of the device type, similar to use of 'br' in bridge names. To show addresses for interfaces associated with a VRF add the master option to the ip command: - $ ip addr show master VRF-NAME + $ ip addr show vrf NAME + $ ip addr show master NAME For example: - $ ip addr show master vrf-red - 3: eth1: mtu 1500 qdisc pfifo_fast master vrf-red state UP group default qlen 1000 + $ ip addr show vrf red + 3: eth1: mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000 link/ether 02:00:00:00:02:02 brd ff:ff:ff:ff:ff:ff inet 10.2.1.2/24 brd 10.2.1.255 scope global eth1 valid_lft forever preferred_lft forever @@ -219,7 +228,7 @@ for emphasis of the device type, similar to use of 'br' in bridge names. valid_lft forever preferred_lft forever inet6 fe80::ff:fe00:202/64 scope link valid_lft forever preferred_lft forever - 4: eth2: mtu 1500 qdisc pfifo_fast master vrf-red state UP group default qlen 1000 + 4: eth2: mtu 1500 qdisc pfifo_fast master red state UP group default qlen 1000 link/ether 02:00:00:00:02:03 brd ff:ff:ff:ff:ff:ff inet 10.2.2.2/24 brd 10.2.2.255 scope global eth2 valid_lft forever preferred_lft forever @@ -227,11 +236,11 @@ for emphasis of the device type, similar to use of 'br' in bridge names. valid_lft forever preferred_lft forever inet6 fe80::ff:fe00:203/64 scope link valid_lft forever preferred_lft forever - 7: eth5: mtu 1500 qdisc noop master vrf-red state DOWN group default qlen 1000 + 7: eth5: mtu 1500 qdisc noop master red state DOWN group default qlen 1000 link/ether 02:00:00:00:02:06 brd ff:ff:ff:ff:ff:ff Or in brief format: - $ ip -br addr show master vrf-red + $ ip -br addr show vrf red eth1 UP 10.2.1.2/24 2002:1::2/120 fe80::ff:fe00:202/64 eth2 UP 10.2.2.2/24 2002:2::2/120 fe80::ff:fe00:203/64 eth5 DOWN @@ -241,10 +250,11 @@ for emphasis of the device type, similar to use of 'br' in bridge names. To show routes for a VRF use the ip command to display the table associated with the VRF device: + $ ip [-6] route show vrf NAME $ ip [-6] route show table ID For example: - $ ip route show table vrf-red + $ ip route show vrf red prohibit default broadcast 10.2.1.0 dev eth1 proto kernel scope link src 10.2.1.2 10.2.1.0/24 dev eth1 proto kernel scope link src 10.2.1.2 @@ -255,7 +265,7 @@ for emphasis of the device type, similar to use of 'br' in bridge names. local 10.2.2.2 dev eth2 proto kernel scope host src 10.2.2.2 broadcast 10.2.2.255 dev eth2 proto kernel scope link src 10.2.2.2 - $ ip -6 route show table vrf-red + $ ip -6 route show vrf red local 2002:1:: dev lo proto none metric 0 pref medium local 2002:1::2 dev lo proto none metric 0 pref medium 2002:1::/120 dev eth1 proto kernel metric 256 pref medium @@ -268,23 +278,24 @@ for emphasis of the device type, similar to use of 'br' in bridge names. local fe80::ff:fe00:203 dev lo proto none metric 0 pref medium fe80::/64 dev eth1 proto kernel metric 256 pref medium fe80::/64 dev eth2 proto kernel metric 256 pref medium - ff00::/8 dev vrf-red metric 256 pref medium + ff00::/8 dev red metric 256 pref medium ff00::/8 dev eth1 metric 256 pref medium ff00::/8 dev eth2 metric 256 pref medium 8. Route Lookup for a VRF - A test route lookup can be done for a VRF by adding the oif option to ip: - $ ip [-6] route get oif VRF-NAME ADDRESS + A test route lookup can be done for a VRF: + $ ip [-6] route get vrf NAME ADDRESS + $ ip [-6] route get oif NAME ADDRESS For example: - $ ip route get 10.2.1.40 oif vrf-red - 10.2.1.40 dev eth1 table vrf-red src 10.2.1.2 + $ ip route get 10.2.1.40 vrf red + 10.2.1.40 dev eth1 table red src 10.2.1.2 cache - $ ip -6 route get 2002:1::32 oif vrf-red - 2002:1::32 from :: dev eth1 table vrf-red proto kernel src 2002:1::2 metric 256 pref medium + $ ip -6 route get 2002:1::32 vrf red + 2002:1::32 from :: dev eth1 table red proto kernel src 2002:1::2 metric 256 pref medium 9. Removing Network Interface from a VRF @@ -303,46 +314,40 @@ for emphasis of the device type, similar to use of 'br' in bridge names. Commands used in this example: -cat >> /etc/iproute2/rt_tables <> /etc/iproute2/rt_tables.d/vrf.conf < Date: Wed, 13 Jul 2016 18:28:16 -0600 Subject: net: vrf: Address comments from last documentation update Comments from Frank Kellerman on last doc update: - extra whitespace in front of a neigh show command - convert the brief link example to 'vrf red' Signed-off-by: David Ahern Signed-off-by: David S. Miller diff --git a/Documentation/networking/vrf.txt b/Documentation/networking/vrf.txt index 11a2b99..755dab8 100644 --- a/Documentation/networking/vrf.txt +++ b/Documentation/networking/vrf.txt @@ -189,7 +189,7 @@ older form without it. Or using the brief output: - $ ip -br link show master red + $ ip -br link show vrf red eth1 UP 02:00:00:00:02:02 eth2 UP 02:00:00:00:02:03 eth5 DOWN 02:00:00:00:02:06 @@ -207,8 +207,8 @@ older form without it. 10.2.1.254 dev eth1 lladdr a6:d9:c7:4f:06:23 REACHABLE 10.2.2.254 dev eth2 lladdr 5e:54:01:6a:ee:80 REACHABLE - $ ip -6 neigh show vrf red - 2002:1::64 dev eth1 lladdr a6:d9:c7:4f:06:23 REACHABLE + $ ip -6 neigh show vrf red + 2002:1::64 dev eth1 lladdr a6:d9:c7:4f:06:23 REACHABLE 6. Show Addresses for a VRF -- cgit v0.10.2 From 3f30849f1f18b852f60ed04c7136967ec3a314a7 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 13 Jul 2016 12:46:40 +0000 Subject: stmmac: dwmac-socfpga: remove redundant dev_err call in socfpga_dwmac_parse_data() There is a error message within devm_ioremap_resource already, so remove the dev_err call to avoid redundant error message. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c index 3bc1fa2..edd20c3 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c @@ -165,12 +165,8 @@ static int socfpga_dwmac_parse_data(struct socfpga_dwmac *dwmac, struct device * dwmac->splitter_base = devm_ioremap_resource(dev, &res_splitter); - if (IS_ERR(dwmac->splitter_base)) { - dev_err(dev, - "%s: ERROR: failed mapping emac splitter\n", - __func__); + if (IS_ERR(dwmac->splitter_base)) return PTR_ERR(dwmac->splitter_base); - } } index = of_property_match_string(np_sgmii_adapter, "reg-names", @@ -188,11 +184,8 @@ static int socfpga_dwmac_parse_data(struct socfpga_dwmac *dwmac, struct device * dwmac->pcs.sgmii_adapter_base = devm_ioremap_resource(dev, &res_sgmii_adapter); - if (IS_ERR(dwmac->pcs.sgmii_adapter_base)) { - dev_err(dev, "%s: failed to mapping adapter\n", - __func__); + if (IS_ERR(dwmac->pcs.sgmii_adapter_base)) return PTR_ERR(dwmac->pcs.sgmii_adapter_base); - } } index = of_property_match_string(np_sgmii_adapter, "reg-names", @@ -210,12 +203,8 @@ static int socfpga_dwmac_parse_data(struct socfpga_dwmac *dwmac, struct device * dwmac->pcs.tse_pcs_base = devm_ioremap_resource(dev, &res_tse_pcs); - if (IS_ERR(dwmac->pcs.tse_pcs_base)) { - dev_err(dev, - "%s: ERROR: failed mapping tse control port\n", - __func__); + if (IS_ERR(dwmac->pcs.tse_pcs_base)) return PTR_ERR(dwmac->pcs.tse_pcs_base); - } } } dwmac->reg_offset = reg_offset; -- cgit v0.10.2 From ce3a380dddd0cb16cb3d8d947b69657d7646c121 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 13 Jul 2016 12:46:57 +0000 Subject: net: ethernet: bgmac: Remove redundant dev_err call in bgmac_probe() There is a error message within devm_ioremap_resource already, so remove the dev_err call to avoid redundant error message. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bgmac-platform.c b/drivers/net/ethernet/broadcom/bgmac-platform.c index 1a2d841..be52f27 100644 --- a/drivers/net/ethernet/broadcom/bgmac-platform.c +++ b/drivers/net/ethernet/broadcom/bgmac-platform.c @@ -129,10 +129,8 @@ static int bgmac_probe(struct platform_device *pdev) } bgmac->plat.base = devm_ioremap_resource(&pdev->dev, regs); - if (IS_ERR(bgmac->plat.base)) { - dev_err(&pdev->dev, "Unable to map base resource\n"); + if (IS_ERR(bgmac->plat.base)) return PTR_ERR(bgmac->plat.base); - } regs = platform_get_resource_byname(pdev, IORESOURCE_MEM, "idm_base"); if (!regs) { @@ -141,10 +139,8 @@ static int bgmac_probe(struct platform_device *pdev) } bgmac->plat.idm_base = devm_ioremap_resource(&pdev->dev, regs); - if (IS_ERR(bgmac->plat.idm_base)) { - dev_err(&pdev->dev, "Unable to map idm resource\n"); + if (IS_ERR(bgmac->plat.idm_base)) return PTR_ERR(bgmac->plat.idm_base); - } bgmac->read = platform_bgmac_read; bgmac->write = platform_bgmac_write; -- cgit v0.10.2 From 9e238323799fb8c2add2b1de9a22edd4d4e51e30 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Wed, 13 Jul 2016 15:08:55 -0300 Subject: sctp: allow others to use sctp_input_cb We process input path in other files too and having access to it is nice, so move it to a header where it's shared. Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 8626bdd..966c3a4 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -59,6 +59,7 @@ #include /* We need tq_struct. */ #include /* We need sctp* header structs. */ #include /* We need auth specific structs */ +#include /* For inet_skb_parm */ /* A convenience structure for handling sockaddr structures. * We should wean ourselves off this. @@ -1092,6 +1093,20 @@ static inline void sctp_outq_cork(struct sctp_outq *q) q->cork = 1; } +/* SCTP skb control block. + * sctp_input_cb is currently used on rx and sock rx queue + */ +struct sctp_input_cb { + union { + struct inet_skb_parm h4; +#if IS_ENABLED(CONFIG_IPV6) + struct inet6_skb_parm h6; +#endif + } header; + struct sctp_chunk *chunk; +}; +#define SCTP_INPUT_CB(__skb) ((struct sctp_input_cb *)&((__skb)->cb[0])) + /* These bind address data fields common between endpoints and associations */ struct sctp_bind_addr { diff --git a/net/sctp/input.c b/net/sctp/input.c index 6f8e676..7a327ff 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -90,17 +90,6 @@ static inline int sctp_rcv_checksum(struct net *net, struct sk_buff *skb) return 0; } -struct sctp_input_cb { - union { - struct inet_skb_parm h4; -#if IS_ENABLED(CONFIG_IPV6) - struct inet6_skb_parm h6; -#endif - } header; - struct sctp_chunk *chunk; -}; -#define SCTP_INPUT_CB(__skb) ((struct sctp_input_cb *)&((__skb)->cb[0])) - /* * This is the routine which IP calls when receiving an SCTP packet. */ -- cgit v0.10.2 From f5d258e60722142e88cb6f0f337d78bca67cf973 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Wed, 13 Jul 2016 15:08:56 -0300 Subject: sctp: reorder sctp_ulpevent and shrink msg_flags The next patch needs 8 bytes in there. sctp_ulpevent has a hole due to bad alignment; msg_flags is using 4 bytes while it actually uses only 2, so we shrink it, and iif member (4 bytes) which can be easily fetched from another place once the next patch is there, so we remove it and thus creating space for 8 bytes. Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/include/net/sctp/ulpevent.h b/include/net/sctp/ulpevent.h index cccdcfd..aa34264 100644 --- a/include/net/sctp/ulpevent.h +++ b/include/net/sctp/ulpevent.h @@ -48,15 +48,15 @@ */ struct sctp_ulpevent { struct sctp_association *asoc; - __u16 stream; - __u16 ssn; - __u16 flags; + unsigned int rmem_len; __u32 ppid; __u32 tsn; __u32 cumtsn; - int msg_flags; int iif; - unsigned int rmem_len; + __u16 stream; + __u16 ssn; + __u16 flags; + __u16 msg_flags; }; /* Retrieve the skb this event sits inside of. */ diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c index d1e3830..706f5bc 100644 --- a/net/sctp/ulpevent.c +++ b/net/sctp/ulpevent.c @@ -51,7 +51,7 @@ static void sctp_ulpevent_release_frag_data(struct sctp_ulpevent *event); /* Initialize an ULP event from an given skb. */ static void sctp_ulpevent_init(struct sctp_ulpevent *event, - int msg_flags, + __u16 msg_flags, unsigned int len) { memset(event, 0, sizeof(struct sctp_ulpevent)); @@ -60,7 +60,7 @@ static void sctp_ulpevent_init(struct sctp_ulpevent *event, } /* Create a new sctp_ulpevent. */ -static struct sctp_ulpevent *sctp_ulpevent_new(int size, int msg_flags, +static struct sctp_ulpevent *sctp_ulpevent_new(int size, __u16 msg_flags, gfp_t gfp) { struct sctp_ulpevent *event; -- cgit v0.10.2 From 1f45f78f8e511203f03138f2ccde3d2cf90d2cbf Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Wed, 13 Jul 2016 15:08:57 -0300 Subject: sctp: allow GSO frags to access the chunk too SCTP will try to access original IP headers on sctp_recvmsg in order to copy the addresses used. There are also other places that do similar access to IP or even SCTP headers. But after 90017accff61 ("sctp: Add GSO support") they aren't always there because they are only present in the header skb. SCTP handles the queueing of incoming data by cloning the incoming skb and limiting to only the relevant payload. This clone has its cb updated to something different and it's then queued on socket rx queue. Thus we need to fix this in two moments. For rx path, not related to socket queue yet, this patch uses a partially copied sctp_input_cb to such GSO frags. This restores the ability to access the headers for this part of the code. Regarding the socket rx queue, it removes iif member from sctp_event and also add a chunk pointer on it. With these changes we're always able to reach the headers again. The biggest change here is that now the sctp_chunk struct and the original skb are only freed after the application consumed the buffer. Note however that the original payload was already like this due to the skb cloning. For iif, SCTP's IPv4 code doesn't use it, so no change is necessary. IPv6 now can fetch it directly from original's IPv6 CB as the original skb is still accessible. In the future we probably can simplify sctp_v*_skb_iif() stuff, as sctp_v4_skb_iif() was called but it's return value not used, and now it's not even called, but such cleanup is out of scope for this change. Fixes: 90017accff61 ("sctp: Add GSO support") Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 966c3a4..f6f201d 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1107,6 +1107,13 @@ struct sctp_input_cb { }; #define SCTP_INPUT_CB(__skb) ((struct sctp_input_cb *)&((__skb)->cb[0])) +static inline const struct sk_buff *sctp_gso_headskb(const struct sk_buff *skb) +{ + const struct sctp_chunk *chunk = SCTP_INPUT_CB(skb)->chunk; + + return chunk->head_skb ? : skb; +} + /* These bind address data fields common between endpoints and associations */ struct sctp_bind_addr { diff --git a/include/net/sctp/ulpevent.h b/include/net/sctp/ulpevent.h index aa34264..2c098cd 100644 --- a/include/net/sctp/ulpevent.h +++ b/include/net/sctp/ulpevent.h @@ -48,11 +48,11 @@ */ struct sctp_ulpevent { struct sctp_association *asoc; + struct sctp_chunk *chunk; unsigned int rmem_len; __u32 ppid; __u32 tsn; __u32 cumtsn; - int iif; __u16 stream; __u16 ssn; __u16 flags; diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index edabbbd..147d975 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -218,6 +218,13 @@ new_skb: chunk->has_asconf = 0; chunk->end_of_packet = 0; chunk->ecn_ce_done = 0; + if (chunk->head_skb) { + struct sctp_input_cb + *cb = SCTP_INPUT_CB(chunk->skb), + *head_cb = SCTP_INPUT_CB(chunk->head_skb); + + cb->chunk = head_cb->chunk; + } } chunk->chunk_hdr = ch; diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 0657d18..ae6f1a2 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -420,6 +420,7 @@ static void sctp_v6_from_skb(union sctp_addr *addr, struct sk_buff *skb, addr->v6.sin6_flowinfo = 0; /* FIXME */ addr->v6.sin6_scope_id = ((struct inet6_skb_parm *)skb->cb)->iif; + /* Always called on head skb, so this is safe */ sh = sctp_hdr(skb); if (is_saddr) { *port = sh->source; @@ -710,8 +711,7 @@ static int sctp_v6_addr_to_user(struct sctp_sock *sp, union sctp_addr *addr) /* Where did this skb come from? */ static int sctp_v6_skb_iif(const struct sk_buff *skb) { - struct inet6_skb_parm *opt = (struct inet6_skb_parm *) skb->cb; - return opt->iif; + return IP6CB(skb)->iif; } /* Was this packet marked by Explicit Congestion Notification? */ @@ -780,15 +780,14 @@ static void sctp_inet6_skb_msgname(struct sk_buff *skb, char *msgname, if (ip_hdr(skb)->version == 4) { addr->v4.sin_family = AF_INET; addr->v4.sin_port = sh->source; - addr->v4.sin_addr.s_addr = ip_hdr(skb)->saddr; + addr->v4.sin_addr.s_addr = ip_hdr(skb)->saddr; } else { addr->v6.sin6_family = AF_INET6; addr->v6.sin6_flowinfo = 0; addr->v6.sin6_port = sh->source; addr->v6.sin6_addr = ipv6_hdr(skb)->saddr; if (ipv6_addr_type(&addr->v6.sin6_addr) & IPV6_ADDR_LINKLOCAL) { - struct sctp_ulpevent *ev = sctp_skb2event(skb); - addr->v6.sin6_scope_id = ev->iif; + addr->v6.sin6_scope_id = sctp_v6_skb_iif(skb); } } diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 3b56ae5..1adb927 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -240,6 +240,7 @@ static void sctp_v4_from_skb(union sctp_addr *addr, struct sk_buff *skb, port = &addr->v4.sin_port; addr->v4.sin_family = AF_INET; + /* Always called on head skb, so this is safe */ sh = sctp_hdr(skb); if (is_saddr) { *port = sh->source; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index f1f08c8..5aabf42 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -6125,7 +6125,8 @@ static int sctp_eat_data(const struct sctp_association *asoc, af = sctp_get_af_specific( ipver2af(ip_hdr(chunk->skb)->version)); - if (af && af->is_ce(chunk->skb) && asoc->peer.ecn_capable) { + if (af && af->is_ce(sctp_gso_headskb(chunk->skb)) && + asoc->peer.ecn_capable) { /* Do real work as sideffect. */ sctp_add_cmd_sf(commands, SCTP_CMD_ECN_CE, SCTP_U32(tsn)); diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 71c7dc5..52fdd54 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -2066,7 +2066,7 @@ static int sctp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, { struct sctp_ulpevent *event = NULL; struct sctp_sock *sp = sctp_sk(sk); - struct sk_buff *skb; + struct sk_buff *skb, *head_skb; int copied; int err = 0; int skb_len; @@ -2102,12 +2102,16 @@ static int sctp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, if (err) goto out_free; - sock_recv_ts_and_drops(msg, sk, skb); + if (event->chunk && event->chunk->head_skb) + head_skb = event->chunk->head_skb; + else + head_skb = skb; + sock_recv_ts_and_drops(msg, sk, head_skb); if (sctp_ulpevent_is_notification(event)) { msg->msg_flags |= MSG_NOTIFICATION; sp->pf->event_msgname(event, msg->msg_name, addr_len); } else { - sp->pf->skb_msgname(skb, msg->msg_name, addr_len); + sp->pf->skb_msgname(head_skb, msg->msg_name, addr_len); } /* Check if we allow SCTP_NXTINFO. */ diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c index 706f5bc..f6219b1 100644 --- a/net/sctp/ulpevent.c +++ b/net/sctp/ulpevent.c @@ -701,6 +701,12 @@ struct sctp_ulpevent *sctp_ulpevent_make_rcvmsg(struct sctp_association *asoc, sctp_ulpevent_receive_data(event, asoc); + /* And hold the chunk as we need it for getting the IP headers + * later in recvmsg + */ + sctp_chunk_hold(chunk); + event->chunk = chunk; + event->stream = ntohs(chunk->subh.data_hdr->stream); event->ssn = ntohs(chunk->subh.data_hdr->ssn); event->ppid = chunk->subh.data_hdr->ppid; @@ -710,11 +716,11 @@ struct sctp_ulpevent *sctp_ulpevent_make_rcvmsg(struct sctp_association *asoc, } event->tsn = ntohl(chunk->subh.data_hdr->tsn); event->msg_flags |= chunk->chunk_hdr->flags; - event->iif = sctp_chunk_iif(chunk); return event; fail_mark: + sctp_chunk_put(chunk); kfree_skb(skb); fail: return NULL; @@ -1007,6 +1013,7 @@ static void sctp_ulpevent_release_data(struct sctp_ulpevent *event) done: sctp_assoc_rwnd_increase(event->asoc, len); + sctp_chunk_put(event->chunk); sctp_ulpevent_release_owner(event); } @@ -1029,6 +1036,7 @@ static void sctp_ulpevent_release_frag_data(struct sctp_ulpevent *event) } done: + sctp_chunk_put(event->chunk); sctp_ulpevent_release_owner(event); } -- cgit v0.10.2 From e7487c86dc5c4a528a7dbd9dc14f453a0de61a84 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Wed, 13 Jul 2016 15:08:58 -0300 Subject: sctp: avoid identifying address family many times for a chunk Identifying address family operations during rx path is not something expensive but it's ugly to the eye to have it done multiple times, specially when we already validated it during initial rx processing. This patch takes advantage of the now shared sctp_input_cb and make the pointer to the operations readily available. Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index f6f201d..ce93c4b 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1104,6 +1104,7 @@ struct sctp_input_cb { #endif } header; struct sctp_chunk *chunk; + struct sctp_af *af; }; #define SCTP_INPUT_CB(__skb) ((struct sctp_input_cb *)&((__skb)->cb[0])) diff --git a/net/sctp/input.c b/net/sctp/input.c index 7a327ff..30d72f7 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -140,6 +140,7 @@ int sctp_rcv(struct sk_buff *skb) af = sctp_get_af_specific(family); if (unlikely(!af)) goto discard_it; + SCTP_INPUT_CB(skb)->af = af; /* Initialize local addresses for lookups. */ af->from_skb(&src, skb, 1); diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index 147d975..8fc773f 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -224,6 +224,7 @@ new_skb: *head_cb = SCTP_INPUT_CB(chunk->head_skb); cb->chunk = head_cb->chunk; + cb->af = head_cb->af; } } diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 1c96f47..8c77b87 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -108,14 +108,9 @@ static void sctp_control_set_owner_w(struct sctp_chunk *chunk) /* What was the inbound interface for this chunk? */ int sctp_chunk_iif(const struct sctp_chunk *chunk) { - struct sctp_af *af; - int iif = 0; - - af = sctp_get_af_specific(ipver2af(ip_hdr(chunk->skb)->version)); - if (af) - iif = af->skb_iif(chunk->skb); + struct sk_buff *skb = chunk->skb; - return iif; + return SCTP_INPUT_CB(skb)->af->skb_iif(skb); } /* RFC 2960 3.3.2 Initiation (INIT) (1) @@ -1600,7 +1595,6 @@ struct sctp_association *sctp_make_temp_asoc(const struct sctp_endpoint *ep, struct sctp_association *asoc; struct sk_buff *skb; sctp_scope_t scope; - struct sctp_af *af; /* Create the bare association. */ scope = sctp_scope(sctp_source(chunk)); @@ -1610,16 +1604,10 @@ struct sctp_association *sctp_make_temp_asoc(const struct sctp_endpoint *ep, asoc->temp = 1; skb = chunk->skb; /* Create an entry for the source address of the packet. */ - af = sctp_get_af_specific(ipver2af(ip_hdr(skb)->version)); - if (unlikely(!af)) - goto fail; - af->from_skb(&asoc->c.peer_addr, skb, 1); + SCTP_INPUT_CB(skb)->af->from_skb(&asoc->c.peer_addr, skb, 1); + nodata: return asoc; - -fail: - sctp_association_free(asoc); - return NULL; } /* Build a cookie representing asoc. diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 5aabf42..b7c1f7f 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -6119,13 +6119,10 @@ static int sctp_eat_data(const struct sctp_association *asoc, */ if (!chunk->ecn_ce_done) { - struct sctp_af *af; + struct sctp_af *af = SCTP_INPUT_CB(chunk->skb)->af; chunk->ecn_ce_done = 1; - af = sctp_get_af_specific( - ipver2af(ip_hdr(chunk->skb)->version)); - - if (af && af->is_ce(sctp_gso_headskb(chunk->skb)) && + if (af->is_ce(sctp_gso_headskb(chunk->skb)) && asoc->peer.ecn_capable) { /* Do real work as sideffect. */ sctp_add_cmd_sf(commands, SCTP_CMD_ECN_CE, -- cgit v0.10.2 From d9cef42529402f9fce10376b6e427a5137d90c3d Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Wed, 13 Jul 2016 15:08:59 -0300 Subject: sctp: do not clear chunk->ecn_ce_done flag We should not clear that flag when switching to a new skb from a GSO skb because it would cause ECN processing to happen multiple times per GSO skb, which is not wanted. Instead, let it be processed once per chunk. That is, in other words, once per IP header available. Fixes: 90017accff61 ("sctp: Add GSO support") Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index 8fc773f..9427706 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -217,7 +217,6 @@ new_skb: chunk->auth = 0; chunk->has_asconf = 0; chunk->end_of_packet = 0; - chunk->ecn_ce_done = 0; if (chunk->head_skb) { struct sctp_input_cb *cb = SCTP_INPUT_CB(chunk->skb), -- cgit v0.10.2 From 2d47fd120d23390fea38c3c7cc5ee05a5b95c49f Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Wed, 13 Jul 2016 15:09:00 -0300 Subject: sctp: only check for ECN if peer is using it Currently only read-only checks are performed up to the point on where we check if peer is ECN capable, checks which we can avoid otherwise. The flag ecn_ce_done is only used to perform this check once per incoming packet, and nothing more. Thus this patch moves the peer check up. Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index b7c1f7f..d88bb2b 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -6118,12 +6118,11 @@ static int sctp_eat_data(const struct sctp_association *asoc, * chunk later. */ - if (!chunk->ecn_ce_done) { + if (asoc->peer.ecn_capable && !chunk->ecn_ce_done) { struct sctp_af *af = SCTP_INPUT_CB(chunk->skb)->af; chunk->ecn_ce_done = 1; - if (af->is_ce(sctp_gso_headskb(chunk->skb)) && - asoc->peer.ecn_capable) { + if (af->is_ce(sctp_gso_headskb(chunk->skb))) { /* Do real work as sideffect. */ sctp_add_cmd_sf(commands, SCTP_CMD_ECN_CE, SCTP_U32(tsn)); -- cgit v0.10.2 From ddbff3e8dde8b6a266eccd4312c6b16ae8aa1d2d Mon Sep 17 00:00:00 2001 From: Elad Kanfi Date: Wed, 13 Jul 2016 16:58:06 +0300 Subject: net: nps_enet: fix coding style issues Fix following coding style problems : ERROR: else should follow close brace '}' + } + else { /* !dst_is_aligned */ WARNING: Missing a blank line after declarations + u32 buf = nps_enet_reg_get(priv, NPS_ENET_REG_RX_BUF); + put_unaligned_be32(buf, reg); WARNING: Missing a blank line after declarations + u32 buf; + ioread32_rep(priv->regs_base + NPS_ENET_REG_RX_BUF, &buf, 1); CHECK: Blank lines aren't necessary before a close brace '}' + + } total: 1 errors, 2 warnings, 1 checks, 683 lines checked Signed-off-by: Elad Kanfi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ezchip/nps_enet.c b/drivers/net/ethernet/ezchip/nps_enet.c index 06f0317..b182e2a 100644 --- a/drivers/net/ethernet/ezchip/nps_enet.c +++ b/drivers/net/ethernet/ezchip/nps_enet.c @@ -46,16 +46,17 @@ static void nps_enet_read_rx_fifo(struct net_device *ndev, if (dst_is_aligned) { ioread32_rep(priv->regs_base + NPS_ENET_REG_RX_BUF, reg, len); reg += len; - } - else { /* !dst_is_aligned */ + } else { /* !dst_is_aligned */ for (i = 0; i < len; i++, reg++) { u32 buf = nps_enet_reg_get(priv, NPS_ENET_REG_RX_BUF); + put_unaligned_be32(buf, reg); } } /* copy last bytes (if any) */ if (last) { u32 buf; + ioread32_rep(priv->regs_base + NPS_ENET_REG_RX_BUF, &buf, 1); memcpy((u8 *)reg, &buf, last); } @@ -459,7 +460,6 @@ static void nps_enet_set_rx_mode(struct net_device *ndev) | NPS_ENET_ENABLE << CFG_2_DISK_DA_SHIFT; ge_mac_cfg_2_value = (ge_mac_cfg_2_value & ~CFG_2_DISK_MC_MASK) | NPS_ENET_ENABLE << CFG_2_DISK_MC_SHIFT; - } nps_enet_reg_set(priv, NPS_ENET_REG_GE_MAC_CFG_2, ge_mac_cfg_2_value); -- cgit v0.10.2 From 094f57aafa3c73329f0c887111d2e596b1aee8ba Mon Sep 17 00:00:00 2001 From: Elad Kanfi Date: Wed, 13 Jul 2016 16:58:07 +0300 Subject: net: nps_enet: code reuse Add inline function that checks if there is a pending tx packet. Signed-off-by: Elad Kanfi Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ezchip/nps_enet.c b/drivers/net/ethernet/ezchip/nps_enet.c index b182e2a..25faa3d 100644 --- a/drivers/net/ethernet/ezchip/nps_enet.c +++ b/drivers/net/ethernet/ezchip/nps_enet.c @@ -24,6 +24,14 @@ #define DRV_NAME "nps_mgt_enet" +static inline bool nps_enet_is_tx_pending(struct nps_enet_priv *priv) +{ + u32 tx_ctrl_value = nps_enet_reg_get(priv, NPS_ENET_REG_TX_CTL); + u32 tx_ctrl_ct = (tx_ctrl_value & TX_CTL_CT_MASK) >> TX_CTL_CT_SHIFT; + + return (!tx_ctrl_ct && priv->tx_skb); +} + static void nps_enet_clean_rx_fifo(struct net_device *ndev, u32 frame_len) { struct nps_enet_priv *priv = netdev_priv(ndev); @@ -141,12 +149,11 @@ static void nps_enet_tx_handler(struct net_device *ndev) { struct nps_enet_priv *priv = netdev_priv(ndev); u32 tx_ctrl_value = nps_enet_reg_get(priv, NPS_ENET_REG_TX_CTL); - u32 tx_ctrl_ct = (tx_ctrl_value & TX_CTL_CT_MASK) >> TX_CTL_CT_SHIFT; u32 tx_ctrl_et = (tx_ctrl_value & TX_CTL_ET_MASK) >> TX_CTL_ET_SHIFT; u32 tx_ctrl_nt = (tx_ctrl_value & TX_CTL_NT_MASK) >> TX_CTL_NT_SHIFT; /* Check if we got TX */ - if (!priv->tx_skb || tx_ctrl_ct) + if (!nps_enet_is_tx_pending(priv)) return; /* Ack Tx ctrl register */ @@ -184,9 +191,6 @@ static int nps_enet_poll(struct napi_struct *napi, int budget) work_done = nps_enet_rx_handler(ndev); if (work_done < budget) { u32 buf_int_enable_value = 0; - u32 tx_ctrl_value = nps_enet_reg_get(priv, NPS_ENET_REG_TX_CTL); - u32 tx_ctrl_ct = - (tx_ctrl_value & TX_CTL_CT_MASK) >> TX_CTL_CT_SHIFT; napi_complete(napi); @@ -205,8 +209,7 @@ static int nps_enet_poll(struct napi_struct *napi, int budget) * the two code lines below will solve this situation by * re-adding ourselves to the poll list. */ - - if (priv->tx_skb && !tx_ctrl_ct) { + if (nps_enet_is_tx_pending(priv)) { nps_enet_reg_set(priv, NPS_ENET_REG_BUF_INT_ENABLE, 0); napi_reschedule(napi); } @@ -231,11 +234,9 @@ static irqreturn_t nps_enet_irq_handler(s32 irq, void *dev_instance) struct net_device *ndev = dev_instance; struct nps_enet_priv *priv = netdev_priv(ndev); u32 rx_ctrl_value = nps_enet_reg_get(priv, NPS_ENET_REG_RX_CTL); - u32 tx_ctrl_value = nps_enet_reg_get(priv, NPS_ENET_REG_TX_CTL); - u32 tx_ctrl_ct = (tx_ctrl_value & TX_CTL_CT_MASK) >> TX_CTL_CT_SHIFT; u32 rx_ctrl_cr = (rx_ctrl_value & RX_CTL_CR_MASK) >> RX_CTL_CR_SHIFT; - if ((!tx_ctrl_ct && priv->tx_skb) || rx_ctrl_cr) + if (nps_enet_is_tx_pending(priv) || rx_ctrl_cr) if (likely(napi_schedule_prep(&priv->napi))) { nps_enet_reg_set(priv, NPS_ENET_REG_BUF_INT_ENABLE, 0); __napi_schedule(&priv->napi); -- cgit v0.10.2 From 29cc6679076a00a6ce193004dcf2d14ae7c428a5 Mon Sep 17 00:00:00 2001 From: Amir Vadai Date: Thu, 14 Jul 2016 10:32:37 +0300 Subject: net/mlx5: Store counters in rbtree instead of list In order to use bulk counters, we need to have counters sorted by id. Signed-off-by: Amir Vadai Reviewed-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index d7ba91a..9cffb6a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -111,6 +111,7 @@ struct mlx5_fc_cache { }; struct mlx5_fc { + struct rb_node node; struct list_head list; /* last{packets,bytes} members are used when calculating the delta since diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c index 164dc37..aaf8fd1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c @@ -32,6 +32,7 @@ #include #include +#include #include "mlx5_core.h" #include "fs_core.h" #include "fs_cmd.h" @@ -68,6 +69,27 @@ * elapsed, the thread will actually query the hardware. */ +static void mlx5_fc_stats_insert(struct rb_root *root, struct mlx5_fc *counter) +{ + struct rb_node **new = &root->rb_node; + struct rb_node *parent = NULL; + + while (*new) { + struct mlx5_fc *this = container_of(*new, struct mlx5_fc, node); + int result = counter->id - this->id; + + parent = *new; + if (result < 0) + new = &((*new)->rb_left); + else + new = &((*new)->rb_right); + } + + /* Add new node and rebalance tree. */ + rb_link_node(&counter->node, parent, new); + rb_insert_color(&counter->node, root); +} + static void mlx5_fc_stats_work(struct work_struct *work) { struct mlx5_core_dev *dev = container_of(work, struct mlx5_core_dev, @@ -75,25 +97,35 @@ static void mlx5_fc_stats_work(struct work_struct *work) struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats; unsigned long now = jiffies; struct mlx5_fc *counter; - struct mlx5_fc *tmp; + struct rb_node *node; + LIST_HEAD(tmplist); int err = 0; spin_lock(&fc_stats->addlist_lock); - list_splice_tail_init(&fc_stats->addlist, &fc_stats->list); + list_splice_tail_init(&fc_stats->addlist, &tmplist); - if (!list_empty(&fc_stats->list)) + if (!list_empty(&tmplist) || !RB_EMPTY_ROOT(&fc_stats->counters)) queue_delayed_work(fc_stats->wq, &fc_stats->work, MLX5_FC_STATS_PERIOD); spin_unlock(&fc_stats->addlist_lock); - list_for_each_entry_safe(counter, tmp, &fc_stats->list, list) { - struct mlx5_fc_cache *c = &counter->cache; + list_for_each_entry(counter, &tmplist, list) + mlx5_fc_stats_insert(&fc_stats->counters, counter); + + node = rb_first(&fc_stats->counters); + while (node) { + struct mlx5_fc_cache *c; u64 packets; u64 bytes; + counter = rb_entry(node, struct mlx5_fc, node); + c = &counter->cache; + + node = rb_next(node); + if (counter->deleted) { - list_del(&counter->list); + rb_erase(&counter->node, &fc_stats->counters); mlx5_cmd_fc_free(dev, counter->id); @@ -176,7 +208,7 @@ int mlx5_init_fc_stats(struct mlx5_core_dev *dev) { struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats; - INIT_LIST_HEAD(&fc_stats->list); + fc_stats->counters = RB_ROOT; INIT_LIST_HEAD(&fc_stats->addlist); spin_lock_init(&fc_stats->addlist_lock); @@ -194,20 +226,32 @@ void mlx5_cleanup_fc_stats(struct mlx5_core_dev *dev) struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats; struct mlx5_fc *counter; struct mlx5_fc *tmp; + struct rb_node *node; cancel_delayed_work_sync(&dev->priv.fc_stats.work); destroy_workqueue(dev->priv.fc_stats.wq); dev->priv.fc_stats.wq = NULL; - list_splice_tail_init(&fc_stats->addlist, &fc_stats->list); - - list_for_each_entry_safe(counter, tmp, &fc_stats->list, list) { + list_for_each_entry_safe(counter, tmp, &fc_stats->addlist, list) { list_del(&counter->list); mlx5_cmd_fc_free(dev, counter->id); kfree(counter); } + + node = rb_first(&fc_stats->counters); + while (node) { + counter = rb_entry(node, struct mlx5_fc, node); + + node = rb_next(node); + + rb_erase(&counter->node, &fc_stats->counters); + + mlx5_cmd_fc_free(dev, counter->id); + + kfree(counter); + } } void mlx5_fc_query_cached(struct mlx5_fc *counter, diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 81e8396..a041b99 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -469,7 +469,7 @@ struct mlx5_irq_info { }; struct mlx5_fc_stats { - struct list_head list; + struct rb_root counters; struct list_head addlist; /* protect addlist add/splice operations */ spinlock_t addlist_lock; -- cgit v0.10.2 From a351a1b03bf169f77891060be30036ef71cbe618 Mon Sep 17 00:00:00 2001 From: Amir Vadai Date: Thu, 14 Jul 2016 10:32:38 +0300 Subject: net/mlx5: Introduce bulk reading of flow counters This commit utilize the ability of ConnectX-4 to bulk read flow counters. Few bulk counter queries could be done instead of issuing thousands of firmware commands per second to get statistics of all flows set to HW, such as those programmed when we offload tc filters. Counters are stored sorted by hardware id, and queried in blocks (id + number of counters). Due to hardware requirement, start of block and number of counters in a block must be four aligned. Reviewed-by: Or Gerlitz Signed-off-by: Amir Vadai Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c index a5bb6b6..9134010 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c @@ -413,3 +413,70 @@ int mlx5_cmd_fc_query(struct mlx5_core_dev *dev, u16 id, return 0; } + +struct mlx5_cmd_fc_bulk { + u16 id; + int num; + int outlen; + u32 out[0]; +}; + +struct mlx5_cmd_fc_bulk * +mlx5_cmd_fc_bulk_alloc(struct mlx5_core_dev *dev, u16 id, int num) +{ + struct mlx5_cmd_fc_bulk *b; + int outlen = sizeof(*b) + + MLX5_ST_SZ_BYTES(query_flow_counter_out) + + MLX5_ST_SZ_BYTES(traffic_counter) * num; + + b = kzalloc(outlen, GFP_KERNEL); + if (!b) + return NULL; + + b->id = id; + b->num = num; + b->outlen = outlen; + + return b; +} + +void mlx5_cmd_fc_bulk_free(struct mlx5_cmd_fc_bulk *b) +{ + kfree(b); +} + +int +mlx5_cmd_fc_bulk_query(struct mlx5_core_dev *dev, struct mlx5_cmd_fc_bulk *b) +{ + u32 in[MLX5_ST_SZ_DW(query_flow_counter_in)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(query_flow_counter_in, in, opcode, + MLX5_CMD_OP_QUERY_FLOW_COUNTER); + MLX5_SET(query_flow_counter_in, in, op_mod, 0); + MLX5_SET(query_flow_counter_in, in, flow_counter_id, b->id); + MLX5_SET(query_flow_counter_in, in, num_of_counters, b->num); + + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), + b->out, b->outlen); +} + +void mlx5_cmd_fc_bulk_get(struct mlx5_core_dev *dev, + struct mlx5_cmd_fc_bulk *b, u16 id, + u64 *packets, u64 *bytes) +{ + int index = id - b->id; + void *stats; + + if (index < 0 || index >= b->num) { + mlx5_core_warn(dev, "Flow counter id (0x%x) out of range (0x%x..0x%x). Counter ignored.\n", + id, b->id, b->id + b->num - 1); + return; + } + + stats = MLX5_ADDR_OF(query_flow_counter_out, b->out, + flow_statistics[index]); + *packets = MLX5_GET64(traffic_counter, stats, packets); + *bytes = MLX5_GET64(traffic_counter, stats, octets); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h index fc4f7b8..158844c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h @@ -76,4 +76,16 @@ int mlx5_cmd_fc_alloc(struct mlx5_core_dev *dev, u16 *id); int mlx5_cmd_fc_free(struct mlx5_core_dev *dev, u16 id); int mlx5_cmd_fc_query(struct mlx5_core_dev *dev, u16 id, u64 *packets, u64 *bytes); + +struct mlx5_cmd_fc_bulk; + +struct mlx5_cmd_fc_bulk * +mlx5_cmd_fc_bulk_alloc(struct mlx5_core_dev *dev, u16 id, int num); +void mlx5_cmd_fc_bulk_free(struct mlx5_cmd_fc_bulk *b); +int +mlx5_cmd_fc_bulk_query(struct mlx5_core_dev *dev, struct mlx5_cmd_fc_bulk *b); +void mlx5_cmd_fc_bulk_get(struct mlx5_core_dev *dev, + struct mlx5_cmd_fc_bulk *b, u16 id, + u64 *packets, u64 *bytes); + #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c index aaf8fd1..c2877e9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c @@ -90,16 +90,66 @@ static void mlx5_fc_stats_insert(struct rb_root *root, struct mlx5_fc *counter) rb_insert_color(&counter->node, root); } +static struct rb_node *mlx5_fc_stats_query(struct mlx5_core_dev *dev, + struct mlx5_fc *first, + u16 last_id) +{ + struct mlx5_cmd_fc_bulk *b; + struct rb_node *node = NULL; + u16 afirst_id; + int num; + int err; + int max_bulk = 1 << MLX5_CAP_GEN(dev, log_max_flow_counter_bulk); + + /* first id must be aligned to 4 when using bulk query */ + afirst_id = first->id & ~0x3; + + /* number of counters to query inc. the last counter */ + num = ALIGN(last_id - afirst_id + 1, 4); + if (num > max_bulk) { + num = max_bulk; + last_id = afirst_id + num - 1; + } + + b = mlx5_cmd_fc_bulk_alloc(dev, afirst_id, num); + if (!b) { + mlx5_core_err(dev, "Error allocating resources for bulk query\n"); + return NULL; + } + + err = mlx5_cmd_fc_bulk_query(dev, b); + if (err) { + mlx5_core_err(dev, "Error doing bulk query: %d\n", err); + goto out; + } + + for (node = &first->node; node; node = rb_next(node)) { + struct mlx5_fc *counter = rb_entry(node, struct mlx5_fc, node); + struct mlx5_fc_cache *c = &counter->cache; + + if (counter->id > last_id) + break; + + mlx5_cmd_fc_bulk_get(dev, b, + counter->id, &c->packets, &c->bytes); + } + +out: + mlx5_cmd_fc_bulk_free(b); + + return node; +} + static void mlx5_fc_stats_work(struct work_struct *work) { struct mlx5_core_dev *dev = container_of(work, struct mlx5_core_dev, priv.fc_stats.work.work); struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats; unsigned long now = jiffies; - struct mlx5_fc *counter; + struct mlx5_fc *counter = NULL; + struct mlx5_fc *last = NULL; struct rb_node *node; LIST_HEAD(tmplist); - int err = 0; spin_lock(&fc_stats->addlist_lock); @@ -115,12 +165,7 @@ static void mlx5_fc_stats_work(struct work_struct *work) node = rb_first(&fc_stats->counters); while (node) { - struct mlx5_fc_cache *c; - u64 packets; - u64 bytes; - counter = rb_entry(node, struct mlx5_fc, node); - c = &counter->cache; node = rb_next(node); @@ -133,26 +178,20 @@ static void mlx5_fc_stats_work(struct work_struct *work) continue; } - if (time_before(now, fc_stats->next_query)) - continue; + last = counter; + } - err = mlx5_cmd_fc_query(dev, counter->id, &packets, &bytes); - if (err) { - pr_err("Error querying stats for counter id %d\n", - counter->id); - continue; - } + if (time_before(now, fc_stats->next_query) || !last) + return; - if (packets == c->packets) - continue; + node = rb_first(&fc_stats->counters); + while (node) { + counter = rb_entry(node, struct mlx5_fc, node); - c->lastuse = jiffies; - c->packets = packets; - c->bytes = bytes; + node = mlx5_fc_stats_query(dev, counter, last->id); } - if (time_after_eq(now, fc_stats->next_query)) - fc_stats->next_query = now + MLX5_FC_STATS_PERIOD; + fc_stats->next_query = now + MLX5_FC_STATS_PERIOD; } struct mlx5_fc *mlx5_fc_create(struct mlx5_core_dev *dev, bool aging) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 152421c..d671e4e 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -893,7 +893,10 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 reserved_at_330[0xb]; u8 log_max_xrcd[0x5]; - u8 reserved_at_340[0x20]; + u8 reserved_at_340[0x8]; + u8 log_max_flow_counter_bulk[0x8]; + u8 max_flow_counter[0x10]; + u8 reserved_at_360[0x3]; u8 log_max_rq[0x5]; @@ -980,7 +983,8 @@ struct mlx5_ifc_dest_format_struct_bits { }; struct mlx5_ifc_flow_counter_list_bits { - u8 reserved_at_0[0x10]; + u8 clear[0x1]; + u8 num_of_counters[0xf]; u8 flow_counter_id[0x10]; u8 reserved_at_20[0x20]; -- cgit v0.10.2 From 55130287875c96b1e4669ee9713621e7d7f055b2 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 14 Jul 2016 10:32:39 +0300 Subject: net/mlx5e: Offload TC flow counters only when supported Currenly, the code that programs the flow actions into the firmware doesn't check if was actually asked to offload the statistics, fix that. Fixes: aad7e08d39bd ('net/mlx5e: Hardware offloaded flower filter statistics support') Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 3261e8b..cd58fc8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -62,7 +62,7 @@ static struct mlx5_flow_rule *mlx5e_tc_add_flow(struct mlx5e_priv *priv, if (action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) { dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; dest.ft = priv->fs.vlan.ft.t; - } else { + } else if (action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { counter = mlx5_fc_create(dev, true); if (IS_ERR(counter)) return ERR_CAST(counter); -- cgit v0.10.2 From 1033665e63b6d98e91c1b938bad2dc624a72c137 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 14 Jul 2016 10:32:40 +0300 Subject: net/mlx5: E-Switch, Use two priorities for SRIOV offloads mode In the offloads mode, some slow path rules are added by the driver (e.g send-to-vport), while offloaded rules are to be added from upper layers. The slow path rules have lower priority and we don't want matching on offloaded rules to suffer from extra steering hops related to the slow path rules. We use two priorities, one for offloaded rules (fast path), and one for the control rules (slow path). To allow for that, we enable two priorities for the FDB namespace in the FS core code. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 7b45e6a..035e536 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -145,6 +145,7 @@ struct mlx5_eswitch_fdb { } legacy; struct offloads_fdb { + struct mlx5_flow_table *fdb; struct mlx5_flow_group *send_to_vport_grp; struct mlx5_flow_group *miss_grp; struct mlx5_flow_rule *miss_rule; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 1842dfb..27122c0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -38,6 +38,11 @@ #include "mlx5_core.h" #include "eswitch.h" +enum { + FDB_FAST_PATH = 0, + FDB_SLOW_PATH +}; + static struct mlx5_flow_rule * mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn) { @@ -149,7 +154,7 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; dest.vport_num = 0; - flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, spec, + flow_rule = mlx5_add_flow_rule(esw->fdb_table.offloads.fdb, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); if (IS_ERR(flow_rule)) { @@ -165,6 +170,8 @@ out: } #define MAX_PF_SQ 256 +#define ESW_OFFLOADS_NUM_ENTRIES (1 << 13) /* 8K */ +#define ESW_OFFLOADS_NUM_GROUPS 4 static int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports) { @@ -190,15 +197,25 @@ static int esw_create_offloads_fdb_table(struct mlx5_eswitch *esw, int nvports) esw_debug(dev, "Create offloads FDB table, log_max_size(%d)\n", MLX5_CAP_ESW_FLOWTABLE_FDB(dev, log_max_ft_size)); - table_size = nvports + MAX_PF_SQ + 1; - fdb = mlx5_create_flow_table(root_ns, 0, table_size, 0); + fdb = mlx5_create_auto_grouped_flow_table(root_ns, FDB_FAST_PATH, + ESW_OFFLOADS_NUM_ENTRIES, + ESW_OFFLOADS_NUM_GROUPS, 0); if (IS_ERR(fdb)) { err = PTR_ERR(fdb); - esw_warn(dev, "Failed to create FDB Table err %d\n", err); - goto fdb_err; + esw_warn(dev, "Failed to create Fast path FDB Table err %d\n", err); + goto fast_fdb_err; } esw->fdb_table.fdb = fdb; + table_size = nvports + MAX_PF_SQ + 1; + fdb = mlx5_create_flow_table(root_ns, FDB_SLOW_PATH, table_size, 0); + if (IS_ERR(fdb)) { + err = PTR_ERR(fdb); + esw_warn(dev, "Failed to create slow path FDB Table err %d\n", err); + goto slow_fdb_err; + } + esw->fdb_table.offloads.fdb = fdb; + /* create send-to-vport group */ memset(flow_group_in, 0, inlen); MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, @@ -247,8 +264,10 @@ miss_rule_err: miss_err: mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp); send_vport_err: - mlx5_destroy_flow_table(fdb); -fdb_err: + mlx5_destroy_flow_table(esw->fdb_table.offloads.fdb); +slow_fdb_err: + mlx5_destroy_flow_table(esw->fdb_table.fdb); +fast_fdb_err: ns_err: kvfree(flow_group_in); return err; @@ -264,6 +283,7 @@ static void esw_destroy_offloads_fdb_table(struct mlx5_eswitch *esw) mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp); mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp); + mlx5_destroy_flow_table(esw->fdb_table.offloads.fdb); mlx5_destroy_flow_table(esw->fdb_table.fdb); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index b0a1304..1a377b4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -1712,15 +1712,21 @@ static int init_fdb_root_ns(struct mlx5_flow_steering *steering) if (!steering->fdb_root_ns) return -ENOMEM; - /* Create single prio */ prio = fs_create_prio(&steering->fdb_root_ns->ns, 0, 1); - if (IS_ERR(prio)) { - cleanup_root_ns(steering->fdb_root_ns); - steering->fdb_root_ns = NULL; - return PTR_ERR(prio); - } else { - return 0; - } + if (IS_ERR(prio)) + goto out_err; + + prio = fs_create_prio(&steering->fdb_root_ns->ns, 1, 1); + if (IS_ERR(prio)) + goto out_err; + + set_prio_attrs(steering->fdb_root_ns); + return 0; + +out_err: + cleanup_root_ns(steering->fdb_root_ns); + steering->fdb_root_ns = NULL; + return PTR_ERR(prio); } static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering) -- cgit v0.10.2 From 3d80d1a2f59ab350f452e1dbc085ba68e5fd8169 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 14 Jul 2016 10:32:41 +0300 Subject: net/mlx5: E-Switch, Add API to configure rules for the offloaded mode This allows for upper levels in the driver, e.g the TC offload code to add e-switch offloaded steering rules. The caller provides the rule spec for matching, action, source and destination vports. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 035e536..c0b0560 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -222,6 +222,12 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, int vport, struct ifla_vf_stats *vf_stats); +struct mlx5_flow_spec; + +struct mlx5_flow_rule * +mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, + struct mlx5_flow_spec *spec, + u32 action, u32 src_vport, u32 dst_vport); struct mlx5_flow_rule * mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 27122c0..a357e8e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -43,6 +43,49 @@ enum { FDB_SLOW_PATH }; +struct mlx5_flow_rule * +mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, + struct mlx5_flow_spec *spec, + u32 action, u32 src_vport, u32 dst_vport) +{ + struct mlx5_flow_destination dest = { 0 }; + struct mlx5_fc *counter = NULL; + struct mlx5_flow_rule *rule; + void *misc; + + if (esw->mode != SRIOV_OFFLOADS) + return ERR_PTR(-EOPNOTSUPP); + + if (action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) { + dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; + dest.vport_num = dst_vport; + action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + } else if (action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { + counter = mlx5_fc_create(esw->dev, true); + if (IS_ERR(counter)) + return ERR_CAST(counter); + dest.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + dest.counter = counter; + } + + misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters); + MLX5_SET(fte_match_set_misc, misc, source_port, src_vport); + + misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters); + MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS | + MLX5_MATCH_MISC_PARAMETERS; + + rule = mlx5_add_flow_rule((struct mlx5_flow_table *)esw->fdb_table.fdb, + spec, action, 0, &dest); + + if (IS_ERR(rule)) + mlx5_fc_destroy(esw->dev, counter); + + return rule; +} + static struct mlx5_flow_rule * mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn) { -- cgit v0.10.2 From 5c40348c69f33c4c14c051181088e8c71e38be7d Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 14 Jul 2016 10:32:42 +0300 Subject: net/mlx5e: Adjustments in the TC offload code towards reuse for SRIOV Towards reusing the TC offloads code for an SRIOV use-case, change some of the helper functions to have _nic in their names so it's clear what's NIC unique and what's general. Also group together the NIC related helpers so we can easily branch per the use-case in downstream patch. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index cd58fc8..57b76f7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -49,9 +49,9 @@ struct mlx5e_tc_flow { #define MLX5E_TC_TABLE_NUM_ENTRIES 1024 #define MLX5E_TC_TABLE_NUM_GROUPS 4 -static struct mlx5_flow_rule *mlx5e_tc_add_flow(struct mlx5e_priv *priv, - struct mlx5_flow_spec *spec, - u32 action, u32 flow_tag) +static struct mlx5_flow_rule *mlx5e_tc_add_nic_flow(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + u32 action, u32 flow_tag) { struct mlx5_core_dev *dev = priv->mdev; struct mlx5_flow_destination dest = { 0 }; @@ -120,7 +120,7 @@ static void mlx5e_tc_del_flow(struct mlx5e_priv *priv, mlx5_fc_destroy(priv->mdev, counter); - if (!mlx5e_tc_num_filters(priv)) { + if (!mlx5e_tc_num_filters(priv) && (priv->fs.tc.t)) { mlx5_destroy_flow_table(priv->fs.tc.t); priv->fs.tc.t = NULL; } @@ -295,8 +295,8 @@ static int parse_cls_flower(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec return 0; } -static int parse_tc_actions(struct mlx5e_priv *priv, struct tcf_exts *exts, - u32 *action, u32 *flow_tag) +static int parse_tc_nic_actions(struct mlx5e_priv *priv, struct tcf_exts *exts, + u32 *action, u32 *flow_tag) { const struct tc_action *a; @@ -369,28 +369,28 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, if (err < 0) goto err_free; - err = parse_tc_actions(priv, f->exts, &action, &flow_tag); + err = parse_tc_nic_actions(priv, f->exts, &action, &flow_tag); if (err < 0) goto err_free; - err = rhashtable_insert_fast(&tc->ht, &flow->node, - tc->ht_params); - if (err) - goto err_free; - - flow->rule = mlx5e_tc_add_flow(priv, spec, action, flow_tag); + flow->rule = mlx5e_tc_add_nic_flow(priv, spec, action, flow_tag); if (IS_ERR(flow->rule)) { err = PTR_ERR(flow->rule); - goto err_hash_del; + goto err_free; } + err = rhashtable_insert_fast(&tc->ht, &flow->node, + tc->ht_params); + if (err) + goto err_del_rule; + if (old) mlx5e_tc_del_flow(priv, old); goto out; -err_hash_del: - rhashtable_remove_fast(&tc->ht, &flow->node, tc->ht_params); +err_del_rule: + mlx5_del_flow_rule(flow->rule); err_free: if (!old) -- cgit v0.10.2 From 8438884d4ab423161b974854ebb90c08219dd678 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 14 Jul 2016 10:32:43 +0300 Subject: net/switchdev: Export the same parent ID service function This helper serves to know if two switchdev port netdevices belong to the same HW ASIC, e.g to figure out if forwarding offload is possible between them. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/include/net/switchdev.h b/include/net/switchdev.h index 985619a..9023e3e 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -227,6 +227,8 @@ void switchdev_port_fwd_mark_set(struct net_device *dev, struct net_device *group_dev, bool joining); +bool switchdev_port_same_parent_id(struct net_device *a, + struct net_device *b); #else static inline void switchdev_deferred_process(void) @@ -351,6 +353,12 @@ static inline void switchdev_port_fwd_mark_set(struct net_device *dev, { } +static inline bool switchdev_port_same_parent_id(struct net_device *a, + struct net_device *b) +{ + return false; +} + #endif #endif /* _LINUX_SWITCHDEV_H_ */ diff --git a/net/switchdev/switchdev.c b/net/switchdev/switchdev.c index 59658b2..a5fc9dd 100644 --- a/net/switchdev/switchdev.c +++ b/net/switchdev/switchdev.c @@ -1286,8 +1286,8 @@ void switchdev_fib_ipv4_abort(struct fib_info *fi) } EXPORT_SYMBOL_GPL(switchdev_fib_ipv4_abort); -static bool switchdev_port_same_parent_id(struct net_device *a, - struct net_device *b) +bool switchdev_port_same_parent_id(struct net_device *a, + struct net_device *b) { struct switchdev_attr a_attr = { .orig_dev = a, @@ -1323,6 +1323,7 @@ static u32 switchdev_port_fwd_mark_get(struct net_device *dev, return dev->ifindex; } +EXPORT_SYMBOL_GPL(switchdev_port_same_parent_id); static void switchdev_port_fwd_mark_reset(struct net_device *group_dev, u32 old_mark, u32 *reset_mark) -- cgit v0.10.2 From 03a9d11e6eeb30363971a9e372b4d61b6e817882 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 14 Jul 2016 10:32:44 +0300 Subject: net/mlx5e: Add TC drop and mirred/redirect action parsing for SRIOV offloads Add the setup code that parses the TC actions needed to support offloading drop and mirred/redirect for SRIOV e-switch. We can redirect between two devices if they belong to the same HW switch, compare the switchdev HW ID attribute to enforce that. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 57b76f7..9a66441 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -37,8 +37,11 @@ #include #include #include +#include +#include #include "en.h" #include "en_tc.h" +#include "eswitch.h" struct mlx5e_tc_flow { struct rhash_head node; @@ -339,6 +342,56 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv, struct tcf_exts *exts, return 0; } +static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts, + u32 *action, u32 *dest_vport) +{ + const struct tc_action *a; + + if (tc_no_actions(exts)) + return -EINVAL; + + *action = 0; + + tc_for_each_action(a, exts) { + /* Only support a single action per rule */ + if (*action) + return -EINVAL; + + if (is_tcf_gact_shot(a)) { + *action = MLX5_FLOW_CONTEXT_ACTION_DROP | + MLX5_FLOW_CONTEXT_ACTION_COUNT; + continue; + } + + if (is_tcf_mirred_redirect(a)) { + int ifindex = tcf_mirred_ifindex(a); + struct net_device *out_dev; + struct mlx5e_priv *out_priv; + struct mlx5_eswitch_rep *out_rep; + + out_dev = __dev_get_by_index(dev_net(priv->netdev), ifindex); + + if (!switchdev_port_same_parent_id(priv->netdev, out_dev)) { + pr_err("devices %s %s not on same switch HW, can't offload forwarding\n", + priv->netdev->name, out_dev->name); + return -EINVAL; + } + + out_priv = netdev_priv(out_dev); + out_rep = out_priv->ppriv; + if (out_rep->vport == 0) + *dest_vport = FDB_UPLINK_VPORT; + else + *dest_vport = out_rep->vport; + *action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + continue; + } + + return -EINVAL; + } + return 0; +} + int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, struct tc_cls_flower_offload *f) { -- cgit v0.10.2 From adb4c123f88dfa7a9c3a320731c765f07a125503 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 14 Jul 2016 10:32:45 +0300 Subject: net/mlx5e: Add TC HW support for FDB (SRIOV e-switch) offloads Enhance the TC offload code such that when the eswitch exists and it's mode being SRIOV offloads, we do TC actions parsing and setup targeted for eswitch. Next, we add the offloaded flow to the HW e-switch (fdb). Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 5ef02f0..fdaf2fa 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -36,6 +36,7 @@ #include "eswitch.h" #include "en.h" +#include "en_tc.h" static const char mlx5e_rep_driver_name[] = "mlx5e_rep"; @@ -201,6 +202,10 @@ void mlx5e_nic_rep_unload(struct mlx5_eswitch *esw, if (test_bit(MLX5E_STATE_OPENED, &priv->state)) mlx5e_remove_sqs_fwd_rules(priv); + + /* clean (and re-init) existing uplink offloaded TC rules */ + mlx5e_tc_cleanup(priv); + mlx5e_tc_init(priv); } static int mlx5e_rep_get_phys_port_name(struct net_device *dev, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 9a66441..0f19b01 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -112,6 +112,22 @@ err_create_ft: return rule; } +static struct mlx5_flow_rule *mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + u32 action, u32 dst_vport) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_eswitch_rep *rep = priv->ppriv; + u32 src_vport; + + if (rep->vport) /* set source vport for the flow */ + src_vport = rep->vport; + else + src_vport = FDB_UPLINK_VPORT; + + return mlx5_eswitch_add_offloaded_rule(esw, spec, action, src_vport, dst_vport); +} + static void mlx5e_tc_del_flow(struct mlx5e_priv *priv, struct mlx5_flow_rule *rule) { @@ -397,11 +413,11 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, { struct mlx5e_tc_table *tc = &priv->fs.tc; int err = 0; - u32 flow_tag; - u32 action; + u32 flow_tag, action, dest_vport = 0; struct mlx5e_tc_flow *flow; struct mlx5_flow_spec *spec; struct mlx5_flow_rule *old = NULL; + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; flow = rhashtable_lookup_fast(&tc->ht, &f->cookie, tc->ht_params); @@ -422,11 +438,18 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, if (err < 0) goto err_free; - err = parse_tc_nic_actions(priv, f->exts, &action, &flow_tag); - if (err < 0) - goto err_free; + if (esw && esw->mode == SRIOV_OFFLOADS) { + err = parse_tc_fdb_actions(priv, f->exts, &action, &dest_vport); + if (err < 0) + goto err_free; + flow->rule = mlx5e_tc_add_fdb_flow(priv, spec, action, dest_vport); + } else { + err = parse_tc_nic_actions(priv, f->exts, &action, &flow_tag); + if (err < 0) + goto err_free; + flow->rule = mlx5e_tc_add_nic_flow(priv, spec, action, flow_tag); + } - flow->rule = mlx5e_tc_add_nic_flow(priv, spec, action, flow_tag); if (IS_ERR(flow->rule)) { err = PTR_ERR(flow->rule); goto err_free; -- cgit v0.10.2 From d957b4e38388a7ff6911db70930436ce8a2e0b2c Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Thu, 14 Jul 2016 10:32:46 +0300 Subject: net/mlx5e: Add TC offload support for the VF representors netdevice The VF representors support only TC filter/action offloads (not mqprio) and this is enabled for them by default. Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index fdaf2fa..1c7d8b8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -33,6 +33,7 @@ #include #include #include +#include #include "eswitch.h" #include "en.h" @@ -222,6 +223,29 @@ static int mlx5e_rep_get_phys_port_name(struct net_device *dev, return 0; } +static int mlx5e_rep_ndo_setup_tc(struct net_device *dev, u32 handle, + __be16 proto, struct tc_to_netdev *tc) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + + if (TC_H_MAJ(handle) != TC_H_MAJ(TC_H_INGRESS)) + return -EOPNOTSUPP; + + switch (tc->type) { + case TC_SETUP_CLSFLOWER: + switch (tc->cls_flower->command) { + case TC_CLSFLOWER_REPLACE: + return mlx5e_configure_flower(priv, proto, tc->cls_flower); + case TC_CLSFLOWER_DESTROY: + return mlx5e_delete_flower(priv, tc->cls_flower); + case TC_CLSFLOWER_STATS: + return mlx5e_stats_flower(priv, tc->cls_flower); + } + default: + return -EOPNOTSUPP; + } +} + static const struct switchdev_ops mlx5e_rep_switchdev_ops = { .switchdev_port_attr_get = mlx5e_attr_get, }; @@ -231,6 +255,7 @@ static const struct net_device_ops mlx5e_netdev_ops_rep = { .ndo_stop = mlx5e_close, .ndo_start_xmit = mlx5e_xmit, .ndo_get_phys_port_name = mlx5e_rep_get_phys_port_name, + .ndo_setup_tc = mlx5e_rep_ndo_setup_tc, .ndo_get_stats64 = mlx5e_get_stats, }; @@ -284,7 +309,8 @@ static void mlx5e_build_rep_netdev(struct net_device *netdev) netdev->switchdev_ops = &mlx5e_rep_switchdev_ops; #endif - netdev->features |= NETIF_F_VLAN_CHALLENGED; + netdev->features |= NETIF_F_VLAN_CHALLENGED | NETIF_F_HW_TC; + netdev->hw_features |= NETIF_F_HW_TC; eth_hw_addr_random(netdev); } @@ -328,8 +354,14 @@ static int mlx5e_init_rep_rx(struct mlx5e_priv *priv) } rep->vport_rx_rule = flow_rule; + err = mlx5e_tc_init(priv); + if (err) + goto err_del_flow_rule; + return 0; +err_del_flow_rule: + mlx5_del_flow_rule(rep->vport_rx_rule); err_destroy_direct_tirs: mlx5e_destroy_direct_tirs(priv); err_destroy_direct_rqts: @@ -343,6 +375,7 @@ static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv) struct mlx5_eswitch_rep *rep = priv->ppriv; int i; + mlx5e_tc_cleanup(priv); mlx5_del_flow_rule(rep->vport_rx_rule); mlx5e_destroy_direct_tirs(priv); for (i = 0; i < priv->params.num_channels; i++) -- cgit v0.10.2 From 15f2cbbde4cff41904f5e87504ff45b36796b8d2 Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Wed, 13 Jul 2016 22:06:04 +0200 Subject: pktgen: add sample script pktgen_sample04_many_flows.sh Adding a pktgen sample script that demonstrates how to use pktgen for simulating flows. Script will generate a certain number of concurrent flows ($FLOWS) and each flow will contain $FLOWLEN packets, which will be send back-to-back, before switching to a new flow, due to flag FLOW_SEQ. This script obsoletes the old sample script 'pktgen.conf-1-1-flows', which is removed. Signed-off-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller diff --git a/samples/pktgen/pktgen.conf-1-1-flows b/samples/pktgen/pktgen.conf-1-1-flows deleted file mode 100755 index 081749c..0000000 --- a/samples/pktgen/pktgen.conf-1-1-flows +++ /dev/null @@ -1,67 +0,0 @@ -#!/bin/bash - -#modprobe pktgen - - -function pgset() { - local result - - echo $1 > $PGDEV - - result=`cat $PGDEV | fgrep "Result: OK:"` - if [ "$result" = "" ]; then - cat $PGDEV | fgrep Result: - fi -} - -# Config Start Here ----------------------------------------------------------- - - -# thread config -# Each CPU has its own thread. One CPU example. We add eth1. - -PGDEV=/proc/net/pktgen/kpktgend_0 - echo "Removing all devices" - pgset "rem_device_all" - echo "Adding eth1" - pgset "add_device eth1" - - -# device config -# delay 0 -# We need to do alloc for every skb since we cannot clone here. - -CLONE_SKB="clone_skb 0" -# NIC adds 4 bytes CRC -PKT_SIZE="pkt_size 60" - -# COUNT 0 means forever -#COUNT="count 0" -COUNT="count 10000000" -DELAY="delay 0" - -PGDEV=/proc/net/pktgen/eth1 - echo "Configuring $PGDEV" - pgset "$COUNT" - pgset "$CLONE_SKB" - pgset "$PKT_SIZE" - pgset "$DELAY" - # Random address with in the min-max range - pgset "flag IPDST_RND" - pgset "dst_min 10.0.0.0" - pgset "dst_max 10.255.255.255" - - # 8k Concurrent flows at 4 pkts - pgset "flows 8192" - pgset "flowlen 4" - - pgset "dst_mac 00:04:23:08:91:dc" - -# Time to run -PGDEV=/proc/net/pktgen/pgctrl - - echo "Running... ctrl^C to stop" - trap true INT - pgset "start" - echo "Done" - cat /proc/net/pktgen/eth1 diff --git a/samples/pktgen/pktgen_sample04_many_flows.sh b/samples/pktgen/pktgen_sample04_many_flows.sh new file mode 100755 index 0000000..f60412e --- /dev/null +++ b/samples/pktgen/pktgen_sample04_many_flows.sh @@ -0,0 +1,93 @@ +#!/bin/bash +# +# Script example for many flows testing +# +# Number of simultaneous flows limited by variable $FLOWS +# and number of packets per flow controlled by variable $FLOWLEN +# +basedir=`dirname $0` +source ${basedir}/functions.sh +root_check_run_with_sudo "$@" + +# Parameter parsing via include +source ${basedir}/parameters.sh +# Set some default params, if they didn't get set +[ -z "$DEST_IP" ] && DEST_IP="198.18.0.42" +[ -z "$DST_MAC" ] && DST_MAC="90:e2:ba:ff:ff:ff" +[ -z "$CLONE_SKB" ] && CLONE_SKB="0" + +# NOTICE: Script specific settings +# ======= +# Limiting the number of concurrent flows ($FLOWS) +# and also set how many packets each flow contains ($FLOWLEN) +# +[ -z "$FLOWS" ] && FLOWS="8000" +[ -z "$FLOWLEN" ] && FLOWLEN="10" + +# Base Config +DELAY="0" # Zero means max speed +COUNT="0" # Zero means indefinitely + +if [[ -n "$BURST" ]]; then + err 1 "Bursting not supported for this mode" +fi + +# General cleanup everything since last run +pg_ctrl "reset" + +# Threads are specified with parameter -t value in $THREADS +for ((thread = 0; thread < $THREADS; thread++)); do + dev=${DEV}@${thread} + + # Add remove all other devices and add_device $dev to thread + pg_thread $thread "rem_device_all" + pg_thread $thread "add_device" $dev + + # Base config + pg_set $dev "flag QUEUE_MAP_CPU" + pg_set $dev "count $COUNT" + pg_set $dev "clone_skb $CLONE_SKB" + pg_set $dev "pkt_size $PKT_SIZE" + pg_set $dev "delay $DELAY" + pg_set $dev "flag NO_TIMESTAMP" + + # Single destination + pg_set $dev "dst_mac $DST_MAC" + pg_set $dev "dst $DEST_IP" + + # Randomize source IP-addresses + pg_set $dev "flag IPSRC_RND" + pg_set $dev "src_min 198.18.0.0" + pg_set $dev "src_max 198.19.255.255" + + # Limit number of flows (max 65535) + pg_set $dev "flows $FLOWS" + # + # How many packets a flow will send, before flow "entry" is + # re-generated/setup. + pg_set $dev "flowlen $FLOWLEN" + # + # Flag FLOW_SEQ will cause $FLOWLEN packets from the same flow + # being send back-to-back, before next flow is selected + # incrementally. This helps lookup caches, and is more realistic. + # + pg_set $dev "flag FLOW_SEQ" + +done + +# Run if user hits control-c +function print_result() { + # Print results + for ((thread = 0; thread < $THREADS; thread++)); do + dev=${DEV}@${thread} + echo "Device: $dev" + cat /proc/net/pktgen/$dev | grep -A2 "Result:" + done +} +# trap keyboard interrupt (Ctrl-C) +trap true SIGINT + +echo "Running... ctrl^C to stop" >&2 +pg_ctrl "start" + +print_result -- cgit v0.10.2 From d25692e4b74573e2d9418bca56f9210adba27972 Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Wed, 13 Jul 2016 22:06:10 +0200 Subject: pktgen: add sample script pktgen_sample05_flow_per_thread.sh This pktgen sample script is useful for scalability testing a receiver. The script will simply generate one flow per thread (option -t N) using the thread number as part of the source IP-address. The single flow sample (pktgen_sample03_burst_single_flow.sh) have become quite popular, but it is important that developers also make sure to benchmark scalability of multiple receive queues. Signed-off-by: Jesper Dangaard Brouer Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/samples/pktgen/pktgen_sample05_flow_per_thread.sh b/samples/pktgen/pktgen_sample05_flow_per_thread.sh new file mode 100755 index 0000000..32ad818 --- /dev/null +++ b/samples/pktgen/pktgen_sample05_flow_per_thread.sh @@ -0,0 +1,81 @@ +#!/bin/bash +# +# Script will generate one flow per thread (-t N) +# - Same destination IP +# - Fake source IPs for each flow (fixed based on thread number) +# +# Useful for scale testing on receiver, to see whether silo'ing flows +# works and scales. For optimal scalability (on receiver) each +# separate-flow should not access shared variables/data. This script +# helps magnify any of these scaling issues by overloading the receiver. +# +basedir=`dirname $0` +source ${basedir}/functions.sh +root_check_run_with_sudo "$@" + +# Parameter parsing via include +source ${basedir}/parameters.sh +# Set some default params, if they didn't get set +[ -z "$DEST_IP" ] && DEST_IP="198.18.0.42" +[ -z "$DST_MAC" ] && DST_MAC="90:e2:ba:ff:ff:ff" +[ -z "$CLONE_SKB" ] && CLONE_SKB="0" +[ -z "$BURST" ] && BURST=32 + + +# Base Config +DELAY="0" # Zero means max speed +COUNT="0" # Zero means indefinitely + +# General cleanup everything since last run +pg_ctrl "reset" + +# Threads are specified with parameter -t value in $THREADS +for ((thread = 0; thread < $THREADS; thread++)); do + dev=${DEV}@${thread} + + # Add remove all other devices and add_device $dev to thread + pg_thread $thread "rem_device_all" + pg_thread $thread "add_device" $dev + + # Base config + pg_set $dev "flag QUEUE_MAP_CPU" + pg_set $dev "count $COUNT" + pg_set $dev "clone_skb $CLONE_SKB" + pg_set $dev "pkt_size $PKT_SIZE" + pg_set $dev "delay $DELAY" + pg_set $dev "flag NO_TIMESTAMP" + + # Single destination + pg_set $dev "dst_mac $DST_MAC" + pg_set $dev "dst $DEST_IP" + + # Setup source IP-addresses based on thread number + pg_set $dev "src_min 198.18.$((thread+1)).1" + pg_set $dev "src_max 198.18.$((thread+1)).1" + + # Setup burst, for easy testing -b 0 disable bursting + # (internally in pktgen default and minimum burst=1) + if [[ ${BURST} -ne 0 ]]; then + pg_set $dev "burst $BURST" + else + info "$dev: Not using burst" + fi + +done + +# Run if user hits control-c +function print_result() { + # Print results + for ((thread = 0; thread < $THREADS; thread++)); do + dev=${DEV}@${thread} + echo "Device: $dev" + cat /proc/net/pktgen/$dev | grep -A2 "Result:" + done +} +# trap keyboard interrupt (Ctrl-C) +trap true SIGINT + +echo "Running... ctrl^C to stop" >&2 +pg_ctrl "start" + +print_result -- cgit v0.10.2 From d3c937bb4c69796d390f9bc8d920d1b60905f305 Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Wed, 13 Jul 2016 22:06:15 +0200 Subject: pktgen: remove sample script pktgen.conf-1-1-rdos Removing the pktgen sample script pktgen.conf-1-1-rdos, because it does not contain anything that is not covered by the other and newer style sample scripts. Signed-off-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller diff --git a/samples/pktgen/pktgen.conf-1-1-rdos b/samples/pktgen/pktgen.conf-1-1-rdos deleted file mode 100755 index c7553be..0000000 --- a/samples/pktgen/pktgen.conf-1-1-rdos +++ /dev/null @@ -1,64 +0,0 @@ -#!/bin/bash - -#modprobe pktgen - - -function pgset() { - local result - - echo $1 > $PGDEV - - result=`cat $PGDEV | fgrep "Result: OK:"` - if [ "$result" = "" ]; then - cat $PGDEV | fgrep Result: - fi -} - -# Config Start Here ----------------------------------------------------------- - - -# thread config -# Each CPU has its own thread. One CPU example. We add eth1. - -PGDEV=/proc/net/pktgen/kpktgend_0 - echo "Removing all devices" - pgset "rem_device_all" - echo "Adding eth1" - pgset "add_device eth1" - - -# device config -# delay 0 - -# We need to do alloc for every skb since we cannot clone here. - -CLONE_SKB="clone_skb 0" -# NIC adds 4 bytes CRC -PKT_SIZE="pkt_size 60" - -# COUNT 0 means forever -#COUNT="count 0" -COUNT="count 10000000" -DELAY="delay 0" - -PGDEV=/proc/net/pktgen/eth1 - echo "Configuring $PGDEV" - pgset "$COUNT" - pgset "$CLONE_SKB" - pgset "$PKT_SIZE" - pgset "$DELAY" - # Random address with in the min-max range - pgset "flag IPDST_RND" - pgset "dst_min 10.0.0.0" - pgset "dst_max 10.255.255.255" - - pgset "dst_mac 00:04:23:08:91:dc" - -# Time to run -PGDEV=/proc/net/pktgen/pgctrl - - echo "Running... ctrl^C to stop" - trap true INT - pgset "start" - echo "Done" - cat /proc/net/pktgen/eth1 -- cgit v0.10.2 From 31abbe34e0f07e8c92f68865cb1d6dde0fe1d52d Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 14 Jul 2016 01:48:51 +0200 Subject: net: ethernet: ll_temac: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/xilinx/ll_temac.h b/drivers/net/ethernet/xilinx/ll_temac.h index 902457e..7d06e3e 100644 --- a/drivers/net/ethernet/xilinx/ll_temac.h +++ b/drivers/net/ethernet/xilinx/ll_temac.h @@ -332,7 +332,6 @@ struct temac_local { struct device *dev; /* Connection to PHY device */ - struct phy_device *phy_dev; /* Pointer to PHY device */ struct device_node *phy_node; /* MDIO bus data */ diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index 7397087..8d6a178 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -590,7 +590,7 @@ static void temac_device_reset(struct net_device *ndev) static void temac_adjust_link(struct net_device *ndev) { struct temac_local *lp = netdev_priv(ndev); - struct phy_device *phy = lp->phy_dev; + struct phy_device *phy = ndev->phydev; u32 mii_speed; int link_state; @@ -843,19 +843,20 @@ static irqreturn_t ll_temac_rx_irq(int irq, void *_ndev) static int temac_open(struct net_device *ndev) { struct temac_local *lp = netdev_priv(ndev); + struct phy_device *phydev = NULL; int rc; dev_dbg(&ndev->dev, "temac_open()\n"); if (lp->phy_node) { - lp->phy_dev = of_phy_connect(lp->ndev, lp->phy_node, - temac_adjust_link, 0, 0); - if (!lp->phy_dev) { + phydev = of_phy_connect(lp->ndev, lp->phy_node, + temac_adjust_link, 0, 0); + if (!phydev) { dev_err(lp->dev, "of_phy_connect() failed\n"); return -ENODEV; } - phy_start(lp->phy_dev); + phy_start(phydev); } temac_device_reset(ndev); @@ -872,9 +873,8 @@ static int temac_open(struct net_device *ndev) err_rx_irq: free_irq(lp->tx_irq, ndev); err_tx_irq: - if (lp->phy_dev) - phy_disconnect(lp->phy_dev); - lp->phy_dev = NULL; + if (phydev) + phy_disconnect(phydev); dev_err(lp->dev, "request_irq() failed\n"); return rc; } @@ -882,15 +882,15 @@ static int temac_open(struct net_device *ndev) static int temac_stop(struct net_device *ndev) { struct temac_local *lp = netdev_priv(ndev); + struct phy_device *phydev = ndev->phydev; dev_dbg(&ndev->dev, "temac_close()\n"); free_irq(lp->tx_irq, ndev); free_irq(lp->rx_irq, ndev); - if (lp->phy_dev) - phy_disconnect(lp->phy_dev); - lp->phy_dev = NULL; + if (phydev) + phy_disconnect(phydev); temac_dma_bd_release(ndev); @@ -916,15 +916,13 @@ temac_poll_controller(struct net_device *ndev) static int temac_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd) { - struct temac_local *lp = netdev_priv(ndev); - if (!netif_running(ndev)) return -EINVAL; - if (!lp->phy_dev) + if (!ndev->phydev) return -EINVAL; - return phy_mii_ioctl(lp->phy_dev, rq, cmd); + return phy_mii_ioctl(ndev->phydev, rq, cmd); } static const struct net_device_ops temac_netdev_ops = { @@ -971,20 +969,17 @@ static const struct attribute_group temac_attr_group = { /* ethtool support */ static int temac_get_settings(struct net_device *ndev, struct ethtool_cmd *cmd) { - struct temac_local *lp = netdev_priv(ndev); - return phy_ethtool_gset(lp->phy_dev, cmd); + return phy_ethtool_gset(ndev->phydev, cmd); } static int temac_set_settings(struct net_device *ndev, struct ethtool_cmd *cmd) { - struct temac_local *lp = netdev_priv(ndev); - return phy_ethtool_sset(lp->phy_dev, cmd); + return phy_ethtool_sset(ndev->phydev, cmd); } static int temac_nway_reset(struct net_device *ndev) { - struct temac_local *lp = netdev_priv(ndev); - return phy_start_aneg(lp->phy_dev); + return phy_start_aneg(ndev->phydev); } static const struct ethtool_ops temac_ethtool_ops = { -- cgit v0.10.2 From e6dab9021f13ad961aadb24770931e30e90e8aa7 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 14 Jul 2016 01:48:52 +0200 Subject: net: ethernet: ll_temac: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index 8d6a178..a9bd665 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -967,27 +967,17 @@ static const struct attribute_group temac_attr_group = { }; /* ethtool support */ -static int temac_get_settings(struct net_device *ndev, struct ethtool_cmd *cmd) -{ - return phy_ethtool_gset(ndev->phydev, cmd); -} - -static int temac_set_settings(struct net_device *ndev, struct ethtool_cmd *cmd) -{ - return phy_ethtool_sset(ndev->phydev, cmd); -} - static int temac_nway_reset(struct net_device *ndev) { return phy_start_aneg(ndev->phydev); } static const struct ethtool_ops temac_ethtool_ops = { - .get_settings = temac_get_settings, - .set_settings = temac_set_settings, .nway_reset = temac_nway_reset, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int temac_of_probe(struct platform_device *op) -- cgit v0.10.2 From a1e3e7372c54fb74994a9cb72965d6e9b4c2efda Mon Sep 17 00:00:00 2001 From: Christophe Jaillet Date: Thu, 14 Jul 2016 08:18:45 +0200 Subject: mlxsw: spectrum_router: Return -ENOENT in case of error 'vr' should be a valid pointer here, so returning 'PTR_ERR(vr)' is wrong. Return an explicit error code (-ENOENT) instead. Fixes: 61c503f976 ("mlxsw: spectrum_router: Implement fib4 add/del switchdev obj ops") Signed-off-by: Christophe JAILLET Acked-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index e084ea5..81418d6 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -1803,7 +1803,7 @@ int mlxsw_sp_router_fib4_del(struct mlxsw_sp_port *mlxsw_sp_port, sizeof(fib4->dst), fib4->dst_len); if (!fib_entry) { dev_warn(mlxsw_sp->bus_info->dev, "Failed to find FIB4 entry being removed.\n"); - return PTR_ERR(vr); + return -ENOENT; } mlxsw_sp_fib_entry_del(mlxsw_sp_port->mlxsw_sp, fib_entry); mlxsw_sp_fib_entry_remove(vr->fib, fib_entry); -- cgit v0.10.2 From 0e1824c98a0ffd7fd9ffb2a3da01ec49ff1348a2 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 14 Jul 2016 11:37:28 +0200 Subject: tracing: change owner name to driver name for devlink hwmsg tracepoint Turned on that driver->owner which is struct module is not available when modules are disabled. Better to depend on a driver name which is always available. Reported-by: Randy Dunlap Fixes: e5224f0fe2 ("devlink: add hardware messages tracing facility") Signed-off-by: Jiri Pirko Acked-by: Randy Dunlap Acked-by: Steven Rostedt Signed-off-by: David S. Miller diff --git a/include/trace/events/devlink.h b/include/trace/events/devlink.h index 333c32a..77dce71 100644 --- a/include/trace/events/devlink.h +++ b/include/trace/events/devlink.h @@ -22,7 +22,7 @@ TRACE_EVENT(devlink_hwmsg, TP_STRUCT__entry( __string(bus_name, devlink->dev->bus->name) __string(dev_name, dev_name(devlink->dev)) - __string(owner_name, devlink->dev->driver->owner->name) + __string(driver_name, devlink->dev->driver->name) __field(bool, incoming) __field(unsigned long, type) __dynamic_array(u8, buf, len) @@ -32,16 +32,16 @@ TRACE_EVENT(devlink_hwmsg, TP_fast_assign( __assign_str(bus_name, devlink->dev->bus->name); __assign_str(dev_name, dev_name(devlink->dev)); - __assign_str(owner_name, devlink->dev->driver->owner->name); + __assign_str(driver_name, devlink->dev->driver->name); __entry->incoming = incoming; __entry->type = type; memcpy(__get_dynamic_array(buf), buf, len); __entry->len = len; ), - TP_printk("bus_name=%s dev_name=%s owner_name=%s incoming=%d type=%lu buf=0x[%*phD] len=%lu", + TP_printk("bus_name=%s dev_name=%s driver_name=%s incoming=%d type=%lu buf=0x[%*phD] len=%lu", __get_str(bus_name), __get_str(dev_name), - __get_str(owner_name), __entry->incoming, __entry->type, + __get_str(driver_name), __entry->incoming, __entry->type, (int) __entry->len, __get_dynamic_array(buf), __entry->len) ); -- cgit v0.10.2 From caeccd5180930eb8586771bb1935f4f2e456a8e8 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 14 Jul 2016 11:37:29 +0200 Subject: devlink: fix trace format string Including devlink.h on ARM and probably other 32-bit architectures results in a harmless warning: In file included from ../include/trace/define_trace.h:95:0, from ../include/trace/events/devlink.h:51, from ../net/core/devlink.c:30: include/trace/events/devlink.h: In function 'trace_raw_output_devlink_hwmsg': include/trace/events/devlink.h:42:12: error: format '%lu' expects argument of type 'long unsigned int', but argument 10 has type 'size_t {aka unsigned int}' [-Werror=format=] The correct format string for 'size_t' is %zu, not %lu, this works on all architectures. Signed-off-by: Arnd Bergmann Fixes: e5224f0fe2ac ("devlink: add hardware messages tracing facility") Signed-off-by: Jiri Pirko Acked-by: Randy Dunlap Signed-off-by: David S. Miller diff --git a/include/trace/events/devlink.h b/include/trace/events/devlink.h index 77dce71..09f1df2 100644 --- a/include/trace/events/devlink.h +++ b/include/trace/events/devlink.h @@ -39,7 +39,7 @@ TRACE_EVENT(devlink_hwmsg, __entry->len = len; ), - TP_printk("bus_name=%s dev_name=%s driver_name=%s incoming=%d type=%lu buf=0x[%*phD] len=%lu", + TP_printk("bus_name=%s dev_name=%s driver_name=%s incoming=%d type=%lu buf=0x[%*phD] len=%zu", __get_str(bus_name), __get_str(dev_name), __get_str(driver_name), __entry->incoming, __entry->type, (int) __entry->len, __get_dynamic_array(buf), __entry->len) -- cgit v0.10.2 From a93d01f5777e99f24b5b3948e06673ada148337c Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 14 Jul 2016 03:51:01 -0700 Subject: RDS: TCP: avoid bad page reference in rds_tcp_listen_data_ready As the existing comments in rds_tcp_listen_data_ready() indicate, it is possible under some race-windows to get to this function with the accept() socket. If that happens, we could run into a sequence whereby thread 1 thread 2 rds_tcp_accept_one() thread sets up new_sock via ->accept(). The sk_user_data is now sock_def_readable data comes in for new_sock, ->sk_data_ready is called, and we land in rds_tcp_listen_data_ready rds_tcp_set_callbacks() takes the sk_callback_lock and sets up sk_user_data to be the cp read_lock sk_callback_lock ready = cp unlock sk_callback_lock page fault on ready In the above sequence, we end up with a panic on a bad page reference when trying to execute (*ready)(). Instead we need to call sock_def_readable() safely, which is what this patch achieves. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/tcp.c b/net/rds/tcp.c index d24f6c1..b411bb7 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -551,6 +551,13 @@ static void rds_tcp_kill_sock(struct net *net) } } +void *rds_tcp_listen_sock_def_readable(struct net *net) +{ + struct rds_tcp_net *rtn = net_generic(net, rds_tcp_netid); + + return rtn->rds_tcp_listen_sock->sk->sk_user_data; +} + static int rds_tcp_dev_event(struct notifier_block *this, unsigned long event, void *ptr) { diff --git a/net/rds/tcp.h b/net/rds/tcp.h index 1c3160f..9a1cc89 100644 --- a/net/rds/tcp.h +++ b/net/rds/tcp.h @@ -70,6 +70,7 @@ void rds_tcp_listen_stop(struct socket *); void rds_tcp_listen_data_ready(struct sock *sk); int rds_tcp_accept_one(struct socket *sock); int rds_tcp_keepalive(struct socket *sock); +void *rds_tcp_listen_sock_def_readable(struct net *net); /* tcp_recv.c */ int rds_tcp_recv_init(void); diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index ca975a2..73040e3 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -183,6 +183,8 @@ void rds_tcp_listen_data_ready(struct sock *sk) */ if (sk->sk_state == TCP_LISTEN) rds_tcp_accept_work(sk); + else + ready = rds_tcp_listen_sock_def_readable(sock_net(sk)); out: read_unlock_bh(&sk->sk_callback_lock); -- cgit v0.10.2 From ac3615e7f3cffe2a1a6b25172dfd09e138593d82 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 14 Jul 2016 03:51:02 -0700 Subject: RDS: TCP: Reduce code duplication in rds_tcp_reset_callbacks() Some code duplication in rds_tcp_reset_callbacks() can be avoided by having the function call rds_tcp_restore_callbacks() and rds_tcp_set_callbacks(). Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/net/rds/tcp.c b/net/rds/tcp.c index b411bb7..0a683cf 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -168,35 +168,21 @@ void rds_tcp_reset_callbacks(struct socket *sock, wait_event(cp->cp_waitq, !test_bit(RDS_IN_XMIT, &cp->cp_flags)); lock_sock(osock->sk); /* reset receive side state for rds_tcp_data_recv() for osock */ + cancel_delayed_work_sync(&cp->cp_send_w); + cancel_delayed_work_sync(&cp->cp_recv_w); if (tc->t_tinc) { rds_inc_put(&tc->t_tinc->ti_inc); tc->t_tinc = NULL; } tc->t_tinc_hdr_rem = sizeof(struct rds_header); tc->t_tinc_data_rem = 0; - tc->t_sock = NULL; - - write_lock_bh(&osock->sk->sk_callback_lock); - - osock->sk->sk_user_data = NULL; - osock->sk->sk_data_ready = tc->t_orig_data_ready; - osock->sk->sk_write_space = tc->t_orig_write_space; - osock->sk->sk_state_change = tc->t_orig_state_change; - write_unlock_bh(&osock->sk->sk_callback_lock); + rds_tcp_restore_callbacks(osock, tc); release_sock(osock->sk); sock_release(osock); newsock: rds_send_path_reset(cp); lock_sock(sock->sk); - write_lock_bh(&sock->sk->sk_callback_lock); - tc->t_sock = sock; - tc->t_cpath = cp; - sock->sk->sk_user_data = cp; - sock->sk->sk_data_ready = rds_tcp_data_ready; - sock->sk->sk_write_space = rds_tcp_write_space; - sock->sk->sk_state_change = rds_tcp_state_change; - - write_unlock_bh(&sock->sk->sk_callback_lock); + rds_tcp_set_callbacks(sock, cp); release_sock(sock->sk); } -- cgit v0.10.2 From 5916e2c1554f3e36f770401c989c3c7fadf619ca Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 14 Jul 2016 03:51:03 -0700 Subject: RDS: TCP: Enable multipath RDS for TCP Use RDS probe-ping to compute how many paths may be used with the peer, and to synchronously start the multiple paths. If mprds is supported, hash outgoing traffic to one of multiple paths in rds_sendmsg() when multipath RDS is supported by the transport. CC: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Acked-by: Santosh Shilimkar Signed-off-by: David S. Miller diff --git a/net/rds/bind.c b/net/rds/bind.c index b22ea95..095f6ce 100644 --- a/net/rds/bind.c +++ b/net/rds/bind.c @@ -81,6 +81,8 @@ static int rds_add_bound(struct rds_sock *rs, __be32 addr, __be16 *port) if (*port != 0) { rover = be16_to_cpu(*port); + if (rover == RDS_FLAG_PROBE_PORT) + return -EINVAL; last = rover; } else { rover = max_t(u16, prandom_u32(), 2); @@ -91,12 +93,16 @@ static int rds_add_bound(struct rds_sock *rs, __be32 addr, __be16 *port) if (rover == 0) rover++; + if (rover == RDS_FLAG_PROBE_PORT) + continue; key = ((u64)addr << 32) | cpu_to_be16(rover); if (rhashtable_lookup_fast(&bind_hash_table, &key, ht_parms)) continue; rs->rs_bound_key = key; rs->rs_bound_addr = addr; + net_get_random_once(&rs->rs_hash_initval, + sizeof(rs->rs_hash_initval)); rs->rs_bound_port = cpu_to_be16(rover); rs->rs_bound_node.next = NULL; rds_sock_addref(rs); diff --git a/net/rds/connection.c b/net/rds/connection.c index 19a4fee..f505855 100644 --- a/net/rds/connection.c +++ b/net/rds/connection.c @@ -155,7 +155,7 @@ static struct rds_connection *__rds_conn_create(struct net *net, struct hlist_head *head = rds_conn_bucket(laddr, faddr); struct rds_transport *loop_trans; unsigned long flags; - int ret; + int ret, i; rcu_read_lock(); conn = rds_conn_lookup(net, head, laddr, faddr, trans); @@ -211,6 +211,12 @@ static struct rds_connection *__rds_conn_create(struct net *net, conn->c_trans = trans; + init_waitqueue_head(&conn->c_hs_waitq); + for (i = 0; i < RDS_MPATH_WORKERS; i++) { + __rds_conn_path_init(conn, &conn->c_path[i], + is_outgoing); + conn->c_path[i].cp_index = i; + } ret = trans->conn_alloc(conn, gfp); if (ret) { kmem_cache_free(rds_conn_slab, conn); @@ -263,14 +269,6 @@ static struct rds_connection *__rds_conn_create(struct net *net, kmem_cache_free(rds_conn_slab, conn); conn = found; } else { - int i; - - for (i = 0; i < RDS_MPATH_WORKERS; i++) { - __rds_conn_path_init(conn, &conn->c_path[i], - is_outgoing); - conn->c_path[i].cp_index = i; - } - hlist_add_head_rcu(&conn->c_hash_node, head); rds_cong_add_conn(conn); rds_conn_count++; @@ -668,6 +666,7 @@ EXPORT_SYMBOL_GPL(rds_conn_path_drop); void rds_conn_drop(struct rds_connection *conn) { + WARN_ON(conn->c_trans->t_mp_capable); rds_conn_path_drop(&conn->c_path[0]); } EXPORT_SYMBOL_GPL(rds_conn_drop); diff --git a/net/rds/message.c b/net/rds/message.c index 756c737..6cb9106 100644 --- a/net/rds/message.c +++ b/net/rds/message.c @@ -41,6 +41,7 @@ static unsigned int rds_exthdr_size[__RDS_EXTHDR_MAX] = { [RDS_EXTHDR_VERSION] = sizeof(struct rds_ext_header_version), [RDS_EXTHDR_RDMA] = sizeof(struct rds_ext_header_rdma), [RDS_EXTHDR_RDMA_DEST] = sizeof(struct rds_ext_header_rdma_dest), +[RDS_EXTHDR_NPATHS] = sizeof(u16), }; diff --git a/net/rds/rds.h b/net/rds/rds.h index 6ef07bd..b2d17f0 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -85,7 +85,9 @@ enum { #define RDS_RECV_REFILL 3 /* Max number of multipaths per RDS connection. Must be a power of 2 */ -#define RDS_MPATH_WORKERS 1 +#define RDS_MPATH_WORKERS 8 +#define RDS_MPATH_HASH(rs, n) (jhash_1word((rs)->rs_bound_port, \ + (rs)->rs_hash_initval) & ((n) - 1)) /* Per mpath connection state */ struct rds_conn_path { @@ -131,7 +133,8 @@ struct rds_connection { __be32 c_laddr; __be32 c_faddr; unsigned int c_loopback:1, - c_pad_to_32:31; + c_ping_triggered:1, + c_pad_to_32:30; int c_npaths; struct rds_connection *c_passive; struct rds_transport *c_trans; @@ -147,6 +150,7 @@ struct rds_connection { unsigned long c_map_queued; struct rds_conn_path c_path[RDS_MPATH_WORKERS]; + wait_queue_head_t c_hs_waitq; /* handshake waitq */ }; static inline @@ -166,6 +170,17 @@ void rds_conn_net_set(struct rds_connection *conn, struct net *net) #define RDS_FLAG_RETRANSMITTED 0x04 #define RDS_MAX_ADV_CREDIT 255 +/* RDS_FLAG_PROBE_PORT is the reserved sport used for sending a ping + * probe to exchange control information before establishing a connection. + * Currently the control information that is exchanged is the number of + * supported paths. If the peer is a legacy (older kernel revision) peer, + * it would return a pong message without additional control information + * that would then alert the sender that the peer was an older rev. + */ +#define RDS_FLAG_PROBE_PORT 1 +#define RDS_HS_PROBE(sport, dport) \ + ((sport == RDS_FLAG_PROBE_PORT && dport == 0) || \ + (sport == 0 && dport == RDS_FLAG_PROBE_PORT)) /* * Maximum space available for extension headers. */ @@ -225,6 +240,11 @@ struct rds_ext_header_rdma_dest { __be32 h_rdma_offset; }; +/* Extension header announcing number of paths. + * Implicit length = 2 bytes. + */ +#define RDS_EXTHDR_NPATHS 4 + #define __RDS_EXTHDR_MAX 16 /* for now */ struct rds_incoming { @@ -545,6 +565,7 @@ struct rds_sock { /* Socket options - in case there will be more */ unsigned char rs_recverr, rs_cong_monitor; + u32 rs_hash_initval; }; static inline struct rds_sock *rds_sk_to_rs(const struct sock *sk) diff --git a/net/rds/recv.c b/net/rds/recv.c index fed53a6..cbfabdf 100644 --- a/net/rds/recv.c +++ b/net/rds/recv.c @@ -156,6 +156,67 @@ static void rds_recv_incoming_exthdrs(struct rds_incoming *inc, struct rds_sock } } +static void rds_recv_hs_exthdrs(struct rds_header *hdr, + struct rds_connection *conn) +{ + unsigned int pos = 0, type, len; + union { + struct rds_ext_header_version version; + u16 rds_npaths; + } buffer; + + while (1) { + len = sizeof(buffer); + type = rds_message_next_extension(hdr, &pos, &buffer, &len); + if (type == RDS_EXTHDR_NONE) + break; + /* Process extension header here */ + switch (type) { + case RDS_EXTHDR_NPATHS: + conn->c_npaths = min_t(int, RDS_MPATH_WORKERS, + buffer.rds_npaths); + break; + default: + pr_warn_ratelimited("ignoring unknown exthdr type " + "0x%x\n", type); + } + } + /* if RDS_EXTHDR_NPATHS was not found, default to a single-path */ + conn->c_npaths = max_t(int, conn->c_npaths, 1); +} + +/* rds_start_mprds() will synchronously start multiple paths when appropriate. + * The scheme is based on the following rules: + * + * 1. rds_sendmsg on first connect attempt sends the probe ping, with the + * sender's npaths (s_npaths) + * 2. rcvr of probe-ping knows the mprds_paths = min(s_npaths, r_npaths). It + * sends back a probe-pong with r_npaths. After that, if rcvr is the + * smaller ip addr, it starts rds_conn_path_connect_if_down on all + * mprds_paths. + * 3. sender gets woken up, and can move to rds_conn_path_connect_if_down. + * If it is the smaller ipaddr, rds_conn_path_connect_if_down can be + * called after reception of the probe-pong on all mprds_paths. + * Otherwise (sender of probe-ping is not the smaller ip addr): just call + * rds_conn_path_connect_if_down on the hashed path. (see rule 4) + * 4. when cp_index > 0, rds_connect_worker must only trigger + * a connection if laddr < faddr. + * 5. sender may end up queuing the packet on the cp. will get sent out later. + * when connection is completed. + */ +static void rds_start_mprds(struct rds_connection *conn) +{ + int i; + struct rds_conn_path *cp; + + if (conn->c_npaths > 1 && conn->c_laddr < conn->c_faddr) { + for (i = 1; i < conn->c_npaths; i++) { + cp = &conn->c_path[i]; + rds_conn_path_connect_if_down(cp); + } + } +} + /* * The transport must make sure that this is serialized against other * rx and conn reset on this specific conn. @@ -232,6 +293,20 @@ void rds_recv_incoming(struct rds_connection *conn, __be32 saddr, __be32 daddr, } rds_stats_inc(s_recv_ping); rds_send_pong(cp, inc->i_hdr.h_sport); + /* if this is a handshake ping, start multipath if necessary */ + if (RDS_HS_PROBE(inc->i_hdr.h_sport, inc->i_hdr.h_dport)) { + rds_recv_hs_exthdrs(&inc->i_hdr, cp->cp_conn); + rds_start_mprds(cp->cp_conn); + } + goto out; + } + + if (inc->i_hdr.h_dport == RDS_FLAG_PROBE_PORT && + inc->i_hdr.h_sport == 0) { + rds_recv_hs_exthdrs(&inc->i_hdr, cp->cp_conn); + /* if this is a handshake pong, start multipath if necessary */ + rds_start_mprds(cp->cp_conn); + wake_up(&cp->cp_conn->c_hs_waitq); goto out; } diff --git a/net/rds/send.c b/net/rds/send.c index 5a9caf1..896626b 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -963,6 +963,29 @@ static int rds_cmsg_send(struct rds_sock *rs, struct rds_message *rm, return ret; } +static void rds_send_ping(struct rds_connection *conn); + +static int rds_send_mprds_hash(struct rds_sock *rs, struct rds_connection *conn) +{ + int hash; + + if (conn->c_npaths == 0) + hash = RDS_MPATH_HASH(rs, RDS_MPATH_WORKERS); + else + hash = RDS_MPATH_HASH(rs, conn->c_npaths); + if (conn->c_npaths == 0 && hash != 0) { + rds_send_ping(conn); + + if (conn->c_npaths == 0) { + wait_event_interruptible(conn->c_hs_waitq, + (conn->c_npaths != 0)); + } + if (conn->c_npaths == 1) + hash = 0; + } + return hash; +} + int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) { struct sock *sk = sock->sk; @@ -1075,7 +1098,10 @@ int rds_sendmsg(struct socket *sock, struct msghdr *msg, size_t payload_len) goto out; } - cpath = &conn->c_path[0]; + if (conn->c_trans->t_mp_capable) + cpath = &conn->c_path[rds_send_mprds_hash(rs, conn)]; + else + cpath = &conn->c_path[0]; rds_conn_path_connect_if_down(cpath); @@ -1135,10 +1161,16 @@ out: } /* - * Reply to a ping packet. + * send out a probe. Can be shared by rds_send_ping, + * rds_send_pong, rds_send_hb. + * rds_send_hb should use h_flags + * RDS_FLAG_HB_PING|RDS_FLAG_ACK_REQUIRED + * or + * RDS_FLAG_HB_PONG|RDS_FLAG_ACK_REQUIRED */ int -rds_send_pong(struct rds_conn_path *cp, __be16 dport) +rds_send_probe(struct rds_conn_path *cp, __be16 sport, + __be16 dport, u8 h_flags) { struct rds_message *rm; unsigned long flags; @@ -1166,9 +1198,18 @@ rds_send_pong(struct rds_conn_path *cp, __be16 dport) rm->m_inc.i_conn = cp->cp_conn; rm->m_inc.i_conn_path = cp; - rds_message_populate_header(&rm->m_inc.i_hdr, 0, dport, + rds_message_populate_header(&rm->m_inc.i_hdr, sport, dport, cp->cp_next_tx_seq); + rm->m_inc.i_hdr.h_flags |= h_flags; cp->cp_next_tx_seq++; + + if (RDS_HS_PROBE(sport, dport) && cp->cp_conn->c_trans->t_mp_capable) { + u16 npaths = RDS_MPATH_WORKERS; + + rds_message_add_extension(&rm->m_inc.i_hdr, + RDS_EXTHDR_NPATHS, &npaths, + sizeof(npaths)); + } spin_unlock_irqrestore(&cp->cp_lock, flags); rds_stats_inc(s_send_queued); @@ -1185,3 +1226,25 @@ out: rds_message_put(rm); return ret; } + +int +rds_send_pong(struct rds_conn_path *cp, __be16 dport) +{ + return rds_send_probe(cp, 0, dport, 0); +} + +void +rds_send_ping(struct rds_connection *conn) +{ + unsigned long flags; + struct rds_conn_path *cp = &conn->c_path[0]; + + spin_lock_irqsave(&cp->cp_lock, flags); + if (conn->c_ping_triggered) { + spin_unlock_irqrestore(&cp->cp_lock, flags); + return; + } + conn->c_ping_triggered = 1; + spin_unlock_irqrestore(&cp->cp_lock, flags); + rds_send_probe(&conn->c_path[0], RDS_FLAG_PROBE_PORT, 0, 0); +} diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 0a683cf..fcddacc 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -38,7 +38,6 @@ #include #include -#include "rds_single_path.h" #include "rds.h" #include "tcp.h" @@ -358,6 +357,7 @@ struct rds_transport rds_tcp_transport = { .t_name = "tcp", .t_type = RDS_TRANS_TCP, .t_prefer_loopback = 1, + .t_mp_capable = 1, }; static int rds_tcp_netid; diff --git a/net/rds/tcp_connect.c b/net/rds/tcp_connect.c index c916715..05f61c5 100644 --- a/net/rds/tcp_connect.c +++ b/net/rds/tcp_connect.c @@ -34,7 +34,6 @@ #include #include -#include "rds_single_path.h" #include "rds.h" #include "tcp.h" @@ -82,6 +81,12 @@ int rds_tcp_conn_path_connect(struct rds_conn_path *cp) struct rds_connection *conn = cp->cp_conn; struct rds_tcp_connection *tc = cp->cp_transport_data; + /* for multipath rds,we only trigger the connection after + * the handshake probe has determined the number of paths. + */ + if (cp->cp_index > 0 && cp->cp_conn->c_npaths < 2) + return -EAGAIN; + mutex_lock(&tc->t_conn_path_lock); if (rds_conn_path_up(cp)) { diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index 73040e3..e0b23fb 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -35,7 +35,6 @@ #include #include -#include "rds_single_path.h" #include "rds.h" #include "tcp.h" @@ -71,6 +70,52 @@ bail: return ret; } +/* rds_tcp_accept_one_path(): if accepting on cp_index > 0, make sure the + * client's ipaddr < server's ipaddr. Otherwise, close the accepted + * socket and force a reconneect from smaller -> larger ip addr. The reason + * we special case cp_index 0 is to allow the rds probe ping itself to itself + * get through efficiently. + * Since reconnects are only initiated from the node with the numerically + * smaller ip address, we recycle conns in RDS_CONN_ERROR on the passive side + * by moving them to CONNECTING in this function. + */ +struct rds_tcp_connection *rds_tcp_accept_one_path(struct rds_connection *conn) +{ + int i; + bool peer_is_smaller = (conn->c_faddr < conn->c_laddr); + int npaths = conn->c_npaths; + + if (npaths <= 1) { + struct rds_conn_path *cp = &conn->c_path[0]; + int ret; + + ret = rds_conn_path_transition(cp, RDS_CONN_DOWN, + RDS_CONN_CONNECTING); + if (!ret) + rds_conn_path_transition(cp, RDS_CONN_ERROR, + RDS_CONN_CONNECTING); + return cp->cp_transport_data; + } + + /* for mprds, paths with cp_index > 0 MUST be initiated by the peer + * with the smaller address. + */ + if (!peer_is_smaller) + return NULL; + + for (i = 1; i < npaths; i++) { + struct rds_conn_path *cp = &conn->c_path[i]; + + if (rds_conn_path_transition(cp, RDS_CONN_DOWN, + RDS_CONN_CONNECTING) || + rds_conn_path_transition(cp, RDS_CONN_ERROR, + RDS_CONN_CONNECTING)) { + return cp->cp_transport_data; + } + } + return NULL; +} + int rds_tcp_accept_one(struct socket *sock) { struct socket *new_sock = NULL; @@ -120,12 +165,14 @@ int rds_tcp_accept_one(struct socket *sock) * If the client reboots, this conn will need to be cleaned up. * rds_tcp_state_change() will do that cleanup */ - rs_tcp = (struct rds_tcp_connection *)conn->c_transport_data; - cp = &conn->c_path[0]; - rds_conn_transition(conn, RDS_CONN_DOWN, RDS_CONN_CONNECTING); + rs_tcp = rds_tcp_accept_one_path(conn); + if (!rs_tcp) + goto rst_nsk; mutex_lock(&rs_tcp->t_conn_path_lock); - conn_state = rds_conn_state(conn); - if (conn_state != RDS_CONN_CONNECTING && conn_state != RDS_CONN_UP) + cp = rs_tcp->t_cpath; + conn_state = rds_conn_path_state(cp); + if (conn_state != RDS_CONN_CONNECTING && conn_state != RDS_CONN_UP && + conn_state != RDS_CONN_ERROR) goto rst_nsk; if (rs_tcp->t_sock) { /* Need to resolve a duelling SYN between peers. @@ -135,11 +182,11 @@ int rds_tcp_accept_one(struct socket *sock) * c_transport_data. */ if (ntohl(inet->inet_saddr) < ntohl(inet->inet_daddr) || - !conn->c_path[0].cp_outgoing) { + !cp->cp_outgoing) { goto rst_nsk; } else { rds_tcp_reset_callbacks(new_sock, cp); - conn->c_path[0].cp_outgoing = 0; + cp->cp_outgoing = 0; /* rds_connect_path_complete() marks RDS_CONN_UP */ rds_connect_path_complete(cp, RDS_CONN_RESETTING); } diff --git a/net/rds/tcp_send.c b/net/rds/tcp_send.c index 57e0f58..89d09b4 100644 --- a/net/rds/tcp_send.c +++ b/net/rds/tcp_send.c @@ -81,7 +81,8 @@ static int rds_tcp_sendmsg(struct socket *sock, void *data, unsigned int len) int rds_tcp_xmit(struct rds_connection *conn, struct rds_message *rm, unsigned int hdr_off, unsigned int sg, unsigned int off) { - struct rds_tcp_connection *tc = conn->c_transport_data; + struct rds_conn_path *cp = rm->m_inc.i_conn_path; + struct rds_tcp_connection *tc = cp->cp_transport_data; int done = 0; int ret = 0; int more; @@ -150,10 +151,17 @@ out: rds_tcp_stats_inc(s_tcp_sndbuf_full); ret = 0; } else { - printk(KERN_WARNING "RDS/tcp: send to %pI4 " - "returned %d, disconnecting and reconnecting\n", - &conn->c_faddr, ret); - rds_conn_drop(conn); + /* No need to disconnect/reconnect if path_drop + * has already been triggered, because, e.g., of + * an incoming RST. + */ + if (rds_conn_path_up(cp)) { + pr_warn("RDS/tcp: send to %pI4 on cp [%d]" + "returned %d, " + "disconnecting and reconnecting\n", + &conn->c_faddr, cp->cp_index, ret); + rds_conn_path_drop(cp); + } } } if (done == 0) diff --git a/net/rds/threads.c b/net/rds/threads.c index bc97d67..e42df11b 100644 --- a/net/rds/threads.c +++ b/net/rds/threads.c @@ -156,6 +156,8 @@ void rds_connect_worker(struct work_struct *work) struct rds_connection *conn = cp->cp_conn; int ret; + if (cp->cp_index > 1 && cp->cp_conn->c_laddr > cp->cp_conn->c_faddr) + return; clear_bit(RDS_RECONNECT_PENDING, &cp->cp_flags); ret = rds_conn_path_transition(cp, RDS_CONN_DOWN, RDS_CONN_CONNECTING); if (ret) { -- cgit v0.10.2 From d67214a29bc8be3216849215aa4fdca057bd8119 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 14 Jul 2016 03:51:04 -0700 Subject: Documentation: RDS: updates for SO_RDS_TRANSPORT socket option Update the documentation to describe the changes added by commit 8ba38460f363 ("net/rds Add getsockopt support for SO_RDS_TRANSPORT") Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/Documentation/networking/rds.txt b/Documentation/networking/rds.txt index 9d219d8..1366e11 100644 --- a/Documentation/networking/rds.txt +++ b/Documentation/networking/rds.txt @@ -85,7 +85,8 @@ Socket Interface bind(fd, &sockaddr_in, ...) This binds the socket to a local IP address and port, and a - transport. + transport, if one has not already been selected via the + SO_RDS_TRANSPORT socket option sendmsg(fd, ...) Sends a message to the indicated recipient. The kernel will @@ -146,6 +147,20 @@ Socket Interface operation. In this case, it would use RDS_CANCEL_SENT_TO to nuke any pending messages. + setsockopt(fd, SOL_RDS, SO_RDS_TRANSPORT, (int *)&transport ..) + getsockopt(fd, SOL_RDS, SO_RDS_TRANSPORT, (int *)&transport ..) + Set or read an integer defining the underlying + encapsulating transport to be used for RDS packets on the + socket. When setting the option, integer argument may be + one of RDS_TRANS_TCP or RDS_TRANS_IB. When retrieving the + value, RDS_TRANS_NONE will be returned on an unbound socket. + This socket option may only be set exactly once on the socket, + prior to binding it via the bind(2) system call. Attempts to + set SO_RDS_TRANSPORT on a socket for which the transport has + been previously attached explicitly (by SO_RDS_TRANSPORT) or + implicitly (via bind(2)) will return an error of EOPNOTSUPP. + An attempt to set SO_RDS_TRANSPPORT to RDS_TRANS_NONE will + always return EINVAL. RDMA for RDS ============ -- cgit v0.10.2 From 09204a6cda4b2189775cb6ee78d7ddd179b0647d Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 14 Jul 2016 03:51:05 -0700 Subject: Documentation: RDS: Document Multipath RDS (mprds) Document the design of mprds, covering a brief description of the motivation, data-structures and modifications to the RDS control plane. Acked-by: Santosh Shilimkar Signed-off-by: Sowmini Varadhan Signed-off-by: David S. Miller diff --git a/Documentation/networking/rds.txt b/Documentation/networking/rds.txt index 1366e11..0235ae6 100644 --- a/Documentation/networking/rds.txt +++ b/Documentation/networking/rds.txt @@ -365,4 +365,59 @@ The recv path handle CMSGs return to application +Multipath RDS (mprds) +===================== + Mprds is multipathed-RDS, primarily intended for RDS-over-TCP + (though the concept can be extended to other transports). The classical + implementation of RDS-over-TCP is implemented by demultiplexing multiple + PF_RDS sockets between any 2 endpoints (where endpoint == [IP address, + port]) over a single TCP socket between the 2 IP addresses involved. This + has the limitation that it ends up funneling multiple RDS flows over a + single TCP flow, thus it is + (a) upper-bounded to the single-flow bandwidth, + (b) suffers from head-of-line blocking for all the RDS sockets. + + Better throughput (for a fixed small packet size, MTU) can be achieved + by having multiple TCP/IP flows per rds/tcp connection, i.e., multipathed + RDS (mprds). Each such TCP/IP flow constitutes a path for the rds/tcp + connection. RDS sockets will be attached to a path based on some hash + (e.g., of local address and RDS port number) and packets for that RDS + socket will be sent over the attached path using TCP to segment/reassemble + RDS datagrams on that path. + + Multipathed RDS is implemented by splitting the struct rds_connection into + a common (to all paths) part, and a per-path struct rds_conn_path. All + I/O workqs and reconnect threads are driven from the rds_conn_path. + Transports such as TCP that are multipath capable may then set up a + TPC socket per rds_conn_path, and this is managed by the transport via + the transport privatee cp_transport_data pointer. + + Transports announce themselves as multipath capable by setting the + t_mp_capable bit during registration with the rds core module. When the + transport is multipath-capable, rds_sendmsg() hashes outgoing traffic + across multiple paths. The outgoing hash is computed based on the + local address and port that the PF_RDS socket is bound to. + + Additionally, even if the transport is MP capable, we may be + peering with some node that does not support mprds, or supports + a different number of paths. As a result, the peering nodes need + to agree on the number of paths to be used for the connection. + This is done by sending out a control packet exchange before the + first data packet. The control packet exchange must have completed + prior to outgoing hash completion in rds_sendmsg() when the transport + is mutlipath capable. + + The control packet is an RDS ping packet (i.e., packet to rds dest + port 0) with the ping packet having a rds extension header option of + type RDS_EXTHDR_NPATHS, length 2 bytes, and the value is the + number of paths supported by the sender. The "probe" ping packet will + get sent from some reserved port, RDS_FLAG_PROBE_PORT (in ) + The receiver of a ping from RDS_FLAG_PROBE_PORT will thus immediately + be able to compute the min(sender_paths, rcvr_paths). The pong + sent in response to a probe-ping should contain the rcvr's npaths + when the rcvr is mprds-capable. + + If the rcvr is not mprds-capable, the exthdr in the ping will be + ignored. In this case the pong will not have any exthdrs, so the sender + of the probe-ping can default to single-path mprds. -- cgit v0.10.2 From 2f43b9beeac8e063353452b4bcf8712f6c2de27b Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 14 Jul 2016 14:16:53 +0300 Subject: wan/fsl_ucc_hdlc: info leak in uhdlc_ioctl() There is a 2 byte struct whole after line.loopback so we need to clear that out to avoid disclosing stack information. Fixes: c19b6d246a35 ('drivers/net: support hdlc function for QE-UCC') Signed-off-by: Dan Carpenter Signed-off-by: David S. Miller diff --git a/drivers/net/wan/fsl_ucc_hdlc.c b/drivers/net/wan/fsl_ucc_hdlc.c index 19174ac..6edd48a 100644 --- a/drivers/net/wan/fsl_ucc_hdlc.c +++ b/drivers/net/wan/fsl_ucc_hdlc.c @@ -635,9 +635,8 @@ static int uhdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) ifr->ifr_settings.size = size; /* data size wanted */ return -ENOBUFS; } + memset(&line, 0, sizeof(line)); line.clock_type = priv->clocking; - line.clock_rate = 0; - line.loopback = 0; if (copy_to_user(ifr->ifr_settings.ifs_ifsu.sync, &line, size)) return -EFAULT; -- cgit v0.10.2 From 77501a79cec40eac65c59ee7af3f786c703ead9c Mon Sep 17 00:00:00 2001 From: Philipp Zabel Date: Thu, 14 Jul 2016 16:29:43 +0200 Subject: net: phy: micrel: Add KSZ8041FTL fiber mode support We can't detect the FXEN (fiber mode) bootstrap pin, so configure it via a boolean device tree property "micrel,fiber-mode". If it is enabled, auto-negotiation is not supported. The only available modes are 100base-fx (full duplex and half duplex). Signed-off-by: Philipp Zabel Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/micrel.txt b/Documentation/devicetree/bindings/net/micrel.txt index 87496a8..8d157f0 100644 --- a/Documentation/devicetree/bindings/net/micrel.txt +++ b/Documentation/devicetree/bindings/net/micrel.txt @@ -35,3 +35,13 @@ Optional properties: supported clocks: - KSZ8021, KSZ8031, KSZ8081, KSZ8091: "rmii-ref": The RMII reference input clock. Used to determine the XI input clock. + + - micrel,fiber-mode: If present the PHY is configured to operate in fiber mode + + Some PHYs, such as the KSZ8041FTL variant, support fiber mode, enabled + by the FXEN boot strapping pin. It can't be determined from the PHY + registers whether the PHY is in fiber mode, so this boolean device tree + property can be used to describe it. + + In fiber mode, auto-negotiation is disabled and the PHY can only work in + 100base-fx (full and half duplex) modes. diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 5a8fefc..059f13b 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -311,6 +311,36 @@ static int kszphy_config_init(struct phy_device *phydev) return 0; } +static int ksz8041_config_init(struct phy_device *phydev) +{ + struct device_node *of_node = phydev->mdio.dev.of_node; + + /* Limit supported and advertised modes in fiber mode */ + if (of_property_read_bool(of_node, "micrel,fiber-mode")) { + phydev->dev_flags |= MICREL_PHY_FXEN; + phydev->supported &= SUPPORTED_FIBRE | + SUPPORTED_100baseT_Full | + SUPPORTED_100baseT_Half; + phydev->advertising &= ADVERTISED_FIBRE | + ADVERTISED_100baseT_Full | + ADVERTISED_100baseT_Half; + phydev->autoneg = AUTONEG_DISABLE; + } + + return kszphy_config_init(phydev); +} + +static int ksz8041_config_aneg(struct phy_device *phydev) +{ + /* Skip auto-negotiation in fiber mode */ + if (phydev->dev_flags & MICREL_PHY_FXEN) { + phydev->speed = SPEED_100; + return 0; + } + + return genphy_config_aneg(phydev); +} + static int ksz9021_load_values_from_of(struct phy_device *phydev, const struct device_node *of_node, u16 reg, @@ -788,8 +818,8 @@ static struct phy_driver ksphy_driver[] = { .flags = PHY_HAS_MAGICANEG | PHY_HAS_INTERRUPT, .driver_data = &ksz8041_type, .probe = kszphy_probe, - .config_init = kszphy_config_init, - .config_aneg = genphy_config_aneg, + .config_init = ksz8041_config_init, + .config_aneg = ksz8041_config_aneg, .read_status = genphy_read_status, .ack_interrupt = kszphy_ack_interrupt, .config_intr = kszphy_config_intr, diff --git a/include/linux/micrel_phy.h b/include/linux/micrel_phy.h index 2e5b194..257173e 100644 --- a/include/linux/micrel_phy.h +++ b/include/linux/micrel_phy.h @@ -37,6 +37,7 @@ /* struct phy_device dev_flags definitions */ #define MICREL_PHY_50MHZ_CLK 0x00000001 +#define MICREL_PHY_FXEN 0x00000002 #define MICREL_KSZ9021_EXTREG_CTRL 0xB #define MICREL_KSZ9021_EXTREG_DATA_WRITE 0xC -- cgit v0.10.2 From 7acef60455c4814a52afb8629d166a3b4dfa0ebb Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 14 Jul 2016 15:47:01 +0100 Subject: rxrpc: checking for IS_ERR() instead of NULL The rxrpc_lookup_peer() function returns NULL on error, it never returns error pointers. Fixes: 8496af50eb38 ('rxrpc: Use RCU to access a peer's service connection tree') Signed-off-by: Dan Carpenter Signed-off-by: David Howells Signed-off-by: David S. Miller diff --git a/net/rxrpc/conn_service.c b/net/rxrpc/conn_service.c index 7cbd612..fd9027c 100644 --- a/net/rxrpc/conn_service.c +++ b/net/rxrpc/conn_service.c @@ -163,7 +163,7 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local, if (!peer) { peer = rxrpc_lookup_peer(local, srx, GFP_NOIO); - if (IS_ERR(peer)) + if (!peer) goto enomem; } -- cgit v0.10.2 From 7e3f977edd0bd9ea6104156feba95bb5ae9bdd38 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 14 Jul 2016 18:08:03 +0200 Subject: perf, events: add non-linear data support for raw records This patch adds support for non-linear data on raw records. It extends raw records to have one or multiple fragments that will be written linearly into the ring slot, where each fragment can optionally have a custom callback handler to walk and extract complex, possibly non-linear data. If a callback handler is provided for a fragment, then the new __output_custom() will be used instead of __output_copy() for the perf_output_sample() part. perf_prepare_sample() does all the size calculation only once, so perf_output_sample() doesn't need to redo the same work anymore, meaning real_size and padding will be cached in the raw record. The raw record becomes 32 bytes in size without holes; to not increase it further and to avoid doing unnecessary recalculations in fast-path, we can reuse next pointer of the last fragment, idea here is borrowed from ZERO_OR_NULL_PTR(), which should keep the perf_output_sample() path for PERF_SAMPLE_RAW minimal. This facility is needed for BPF's event output helper as a first user that will, in a follow-up, add an additional perf_raw_frag to its perf_raw_record in order to be able to more efficiently dump skb context after a linear head meta data related to it. skbs can be non-linear and thus need a custom output function to dump buffers. Currently, the skb data needs to be copied twice; with the help of __output_custom() this work only needs to be done once. Future users could be things like XDP/BPF programs that work on different context though and would thus also have a different callback function. The few users of raw records are adapted to initialize their frag data from the raw record itself, no change in behavior for them. The code is based upon a PoC diff provided by Peter Zijlstra [1]. [1] http://thread.gmane.org/gmane.linux.network/421294 Suggested-by: Peter Zijlstra Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c index a8e8321..92619cc 100644 --- a/arch/s390/kernel/perf_cpum_sf.c +++ b/arch/s390/kernel/perf_cpum_sf.c @@ -979,12 +979,15 @@ static int perf_push_sample(struct perf_event *event, struct sf_raw_sample *sfr) struct pt_regs regs; struct perf_sf_sde_regs *sde_regs; struct perf_sample_data data; - struct perf_raw_record raw; + struct perf_raw_record raw = { + .frag = { + .size = sfr->size, + .data = sfr, + }, + }; /* Setup perf sample */ perf_sample_data_init(&data, 0, event->hw.last_period); - raw.size = sfr->size; - raw.data = sfr; data.raw = &raw; /* Setup pt_regs to look like an CPU-measurement external interrupt diff --git a/arch/x86/events/amd/ibs.c b/arch/x86/events/amd/ibs.c index feb90f6..72dea2f 100644 --- a/arch/x86/events/amd/ibs.c +++ b/arch/x86/events/amd/ibs.c @@ -655,8 +655,12 @@ fail: } if (event->attr.sample_type & PERF_SAMPLE_RAW) { - raw.size = sizeof(u32) + ibs_data.size; - raw.data = ibs_data.data; + raw = (struct perf_raw_record){ + .frag = { + .size = sizeof(u32) + ibs_data.size, + .data = ibs_data.data, + }, + }; data.raw = &raw; } diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 1a827ce..e79e6c6 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -69,9 +69,22 @@ struct perf_callchain_entry_ctx { bool contexts_maxed; }; +typedef unsigned long (*perf_copy_f)(void *dst, const void *src, + unsigned long len); + +struct perf_raw_frag { + union { + struct perf_raw_frag *next; + unsigned long pad; + }; + perf_copy_f copy; + void *data; + u32 size; +} __packed; + struct perf_raw_record { + struct perf_raw_frag frag; u32 size; - void *data; }; /* @@ -1283,6 +1296,11 @@ extern void perf_restore_debug_store(void); static inline void perf_restore_debug_store(void) { } #endif +static __always_inline bool perf_raw_frag_last(const struct perf_raw_frag *frag) +{ + return frag->pad < sizeof(u64); +} + #define perf_output_put(handle, x) perf_output_copy((handle), &(x), sizeof(x)) /* diff --git a/kernel/events/core.c b/kernel/events/core.c index 9c51ec3..b1891b6 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5553,16 +5553,26 @@ void perf_output_sample(struct perf_output_handle *handle, } if (sample_type & PERF_SAMPLE_RAW) { - if (data->raw) { - u32 raw_size = data->raw->size; - u32 real_size = round_up(raw_size + sizeof(u32), - sizeof(u64)) - sizeof(u32); - u64 zero = 0; - - perf_output_put(handle, real_size); - __output_copy(handle, data->raw->data, raw_size); - if (real_size - raw_size) - __output_copy(handle, &zero, real_size - raw_size); + struct perf_raw_record *raw = data->raw; + + if (raw) { + struct perf_raw_frag *frag = &raw->frag; + + perf_output_put(handle, raw->size); + do { + if (frag->copy) { + __output_custom(handle, frag->copy, + frag->data, frag->size); + } else { + __output_copy(handle, frag->data, + frag->size); + } + if (perf_raw_frag_last(frag)) + break; + frag = frag->next; + } while (1); + if (frag->pad) + __output_skip(handle, NULL, frag->pad); } else { struct { u32 size; @@ -5687,14 +5697,28 @@ void perf_prepare_sample(struct perf_event_header *header, } if (sample_type & PERF_SAMPLE_RAW) { - int size = sizeof(u32); - - if (data->raw) - size += data->raw->size; - else - size += sizeof(u32); + struct perf_raw_record *raw = data->raw; + int size; + + if (raw) { + struct perf_raw_frag *frag = &raw->frag; + u32 sum = 0; + + do { + sum += frag->size; + if (perf_raw_frag_last(frag)) + break; + frag = frag->next; + } while (1); + + size = round_up(sum + sizeof(u32), sizeof(u64)); + raw->size = size - sizeof(u32); + frag->pad = raw->size - sum; + } else { + size = sizeof(u64); + } - header->size += round_up(size, sizeof(u64)); + header->size += size; } if (sample_type & PERF_SAMPLE_BRANCH_STACK) { @@ -7331,7 +7355,7 @@ static struct pmu perf_swevent = { static int perf_tp_filter_match(struct perf_event *event, struct perf_sample_data *data) { - void *record = data->raw->data; + void *record = data->raw->frag.data; /* only top level events have filters set */ if (event->parent) @@ -7387,8 +7411,10 @@ void perf_tp_event(u16 event_type, u64 count, void *record, int entry_size, struct perf_event *event; struct perf_raw_record raw = { - .size = entry_size, - .data = record, + .frag = { + .size = entry_size, + .data = record, + }, }; perf_sample_data_init(&data, 0, 0); diff --git a/kernel/events/internal.h b/kernel/events/internal.h index 05f9f6d..2417eb5 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -123,10 +123,7 @@ static inline unsigned long perf_aux_size(struct ring_buffer *rb) return rb->aux_nr_pages << PAGE_SHIFT; } -#define DEFINE_OUTPUT_COPY(func_name, memcpy_func) \ -static inline unsigned long \ -func_name(struct perf_output_handle *handle, \ - const void *buf, unsigned long len) \ +#define __DEFINE_OUTPUT_COPY_BODY(memcpy_func) \ { \ unsigned long size, written; \ \ @@ -152,6 +149,17 @@ func_name(struct perf_output_handle *handle, \ return len; \ } +#define DEFINE_OUTPUT_COPY(func_name, memcpy_func) \ +static inline unsigned long \ +func_name(struct perf_output_handle *handle, \ + const void *buf, unsigned long len) \ +__DEFINE_OUTPUT_COPY_BODY(memcpy_func) + +static inline unsigned long +__output_custom(struct perf_output_handle *handle, perf_copy_f copy_func, + const void *buf, unsigned long len) +__DEFINE_OUTPUT_COPY_BODY(copy_func) + static inline unsigned long memcpy_common(void *dst, const void *src, unsigned long n) { diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 094c716..35ab1b2 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -245,8 +245,10 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) struct bpf_event_entry *ee; struct perf_event *event; struct perf_raw_record raw = { - .size = size, - .data = data, + .frag = { + .size = size, + .data = data, + }, }; if (unlikely(flags & ~(BPF_F_INDEX_MASK))) -- cgit v0.10.2 From 8e7a3920ac277dd4e690c0e70c9750176e3acb83 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 14 Jul 2016 18:08:04 +0200 Subject: bpf, perf: split bpf_perf_event_output Split the bpf_perf_event_output() helper as a preparation into two parts. The new bpf_perf_event_output() will prepare the raw record itself and test for unknown flags from BPF trace context, where the __bpf_perf_event_output() does the core work. The latter will be reused later on from bpf_event_output() directly. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 35ab1b2..c35883a 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -233,26 +233,17 @@ static const struct bpf_func_proto bpf_perf_event_read_proto = { .arg2_type = ARG_ANYTHING, }; -static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) +static __always_inline u64 +__bpf_perf_event_output(struct pt_regs *regs, struct bpf_map *map, + u64 flags, struct perf_raw_record *raw) { - struct pt_regs *regs = (struct pt_regs *) (long) r1; - struct bpf_map *map = (struct bpf_map *) (long) r2; struct bpf_array *array = container_of(map, struct bpf_array, map); unsigned int cpu = smp_processor_id(); u64 index = flags & BPF_F_INDEX_MASK; - void *data = (void *) (long) r4; struct perf_sample_data sample_data; struct bpf_event_entry *ee; struct perf_event *event; - struct perf_raw_record raw = { - .frag = { - .size = size, - .data = data, - }, - }; - if (unlikely(flags & ~(BPF_F_INDEX_MASK))) - return -EINVAL; if (index == BPF_F_CURRENT_CPU) index = cpu; if (unlikely(index >= array->map.max_entries)) @@ -271,11 +262,29 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) return -EOPNOTSUPP; perf_sample_data_init(&sample_data, 0, 0); - sample_data.raw = &raw; + sample_data.raw = raw; perf_event_output(event, &sample_data, regs); return 0; } +static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) +{ + struct pt_regs *regs = (struct pt_regs *)(long) r1; + struct bpf_map *map = (struct bpf_map *)(long) r2; + void *data = (void *)(long) r4; + struct perf_raw_record raw = { + .frag = { + .size = size, + .data = data, + }, + }; + + if (unlikely(flags & ~(BPF_F_INDEX_MASK))) + return -EINVAL; + + return __bpf_perf_event_output(regs, map, flags, &raw); +} + static const struct bpf_func_proto bpf_perf_event_output_proto = { .func = bpf_perf_event_output, .gpl_only = true, -- cgit v0.10.2 From 555c8a8623a3a87b3c990ba30b7fd2e5914e41d2 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 14 Jul 2016 18:08:05 +0200 Subject: bpf: avoid stack copy and use skb ctx for event output This work addresses a couple of issues bpf_skb_event_output() helper currently has: i) We need two copies instead of just a single one for the skb data when it should be part of a sample. The data can be non-linear and thus needs to be extracted via bpf_skb_load_bytes() helper first, and then copied once again into the ring buffer slot. ii) Since bpf_skb_load_bytes() currently needs to be used first, the helper needs to see a constant size on the passed stack buffer to make sure BPF verifier can do sanity checks on it during verification time. Thus, just passing skb->len (or any other non-constant value) wouldn't work, but changing bpf_skb_load_bytes() is also not the proper solution, since the two copies are generally still needed. iii) bpf_skb_load_bytes() is just for rather small buffers like headers, since they need to sit on the limited BPF stack anyway. Instead of working around in bpf_skb_load_bytes(), this work improves the bpf_skb_event_output() helper to address all 3 at once. We can make use of the passed in skb context that we have in the helper anyway, and use some of the reserved flag bits as a length argument. The helper will use the new __output_custom() facility from perf side with bpf_skb_copy() as callback helper to walk and extract the data. It will pass the data for setup to bpf_event_output(), which generates and pushes the raw record with an additional frag part. The linear data used in the first frag of the record serves as programmatically defined meta data passed along with the appended sample. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/bpf.h b/include/linux/bpf.h index b3336b4..c13e92b 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -209,7 +209,12 @@ u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp); const struct bpf_func_proto *bpf_get_trace_printk_proto(void); -const struct bpf_func_proto *bpf_get_event_output_proto(void); + +typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src, + unsigned long len); + +u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, + void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy); #ifdef CONFIG_BPF_SYSCALL DECLARE_PER_CPU(int, bpf_prog_active); diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 262a7e8..c4d9224 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -401,6 +401,8 @@ enum bpf_func_id { /* BPF_FUNC_perf_event_output and BPF_FUNC_perf_event_read flags. */ #define BPF_F_INDEX_MASK 0xffffffffULL #define BPF_F_CURRENT_CPU BPF_F_INDEX_MASK +/* BPF_FUNC_perf_event_output for sk_buff input context. */ +#define BPF_F_CTXLEN_MASK (0xfffffULL << 32) /* user accessible mirror of in-kernel sk_buff. * new fields can only be added to the end of this structure diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index d638062..03fd23d 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -1054,9 +1054,11 @@ const struct bpf_func_proto * __weak bpf_get_trace_printk_proto(void) return NULL; } -const struct bpf_func_proto * __weak bpf_get_event_output_proto(void) +u64 __weak +bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, + void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy) { - return NULL; + return -ENOTSUPP; } /* Always built-in helper functions. */ diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index c35883a..ebfbb7d 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -298,29 +298,26 @@ static const struct bpf_func_proto bpf_perf_event_output_proto = { static DEFINE_PER_CPU(struct pt_regs, bpf_pt_regs); -static u64 bpf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) +u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, + void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy) { struct pt_regs *regs = this_cpu_ptr(&bpf_pt_regs); + struct perf_raw_frag frag = { + .copy = ctx_copy, + .size = ctx_size, + .data = ctx, + }; + struct perf_raw_record raw = { + .frag = { + .next = ctx_size ? &frag : NULL, + .size = meta_size, + .data = meta, + }, + }; perf_fetch_caller_regs(regs); - return bpf_perf_event_output((long)regs, r2, flags, r4, size); -} - -static const struct bpf_func_proto bpf_event_output_proto = { - .func = bpf_event_output, - .gpl_only = true, - .ret_type = RET_INTEGER, - .arg1_type = ARG_PTR_TO_CTX, - .arg2_type = ARG_CONST_MAP_PTR, - .arg3_type = ARG_ANYTHING, - .arg4_type = ARG_PTR_TO_STACK, - .arg5_type = ARG_CONST_STACK_SIZE, -}; - -const struct bpf_func_proto *bpf_get_event_output_proto(void) -{ - return &bpf_event_output_proto; + return __bpf_perf_event_output(regs, map, flags, &raw); } static u64 bpf_get_current_task(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) diff --git a/net/core/filter.c b/net/core/filter.c index 10c4a2f..22e3992 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2025,6 +2025,47 @@ bool bpf_helper_changes_skb_data(void *func) return false; } +static unsigned long bpf_skb_copy(void *dst_buff, const void *skb, + unsigned long len) +{ + void *ptr = skb_header_pointer(skb, 0, len, dst_buff); + + if (unlikely(!ptr)) + return len; + if (ptr != dst_buff) + memcpy(dst_buff, ptr, len); + + return 0; +} + +static u64 bpf_skb_event_output(u64 r1, u64 r2, u64 flags, u64 r4, + u64 meta_size) +{ + struct sk_buff *skb = (struct sk_buff *)(long) r1; + struct bpf_map *map = (struct bpf_map *)(long) r2; + u64 skb_size = (flags & BPF_F_CTXLEN_MASK) >> 32; + void *meta = (void *)(long) r4; + + if (unlikely(flags & ~(BPF_F_CTXLEN_MASK | BPF_F_INDEX_MASK))) + return -EINVAL; + if (unlikely(skb_size > skb->len)) + return -EFAULT; + + return bpf_event_output(map, flags, meta, meta_size, skb, skb_size, + bpf_skb_copy); +} + +static const struct bpf_func_proto bpf_skb_event_output_proto = { + .func = bpf_skb_event_output, + .gpl_only = true, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_CONST_MAP_PTR, + .arg3_type = ARG_ANYTHING, + .arg4_type = ARG_PTR_TO_STACK, + .arg5_type = ARG_CONST_STACK_SIZE, +}; + static unsigned short bpf_tunnel_key_af(u64 flags) { return flags & BPF_F_TUNINFO_IPV6 ? AF_INET6 : AF_INET; @@ -2357,7 +2398,7 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) case BPF_FUNC_get_hash_recalc: return &bpf_get_hash_recalc_proto; case BPF_FUNC_perf_event_output: - return bpf_get_event_output_proto(); + return &bpf_skb_event_output_proto; case BPF_FUNC_get_smp_processor_id: return &bpf_get_smp_processor_id_proto; #ifdef CONFIG_SOCK_CGROUP_DATA -- cgit v0.10.2 From 1b16bf42d154c8fbbab2cccc419e2ba47d700849 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Fri, 15 Jul 2016 03:46:30 -0400 Subject: macvtap: avoid hash calculating for single queue We decide the rxq through calculating its hash which is not necessary if we only have one rx queue. So this patch skip this and just return queue 0. Test shows 22% improving on guest rx pps. Before: 1201504 pkts/s After: 1472731 pkts/s Signed-off-by: Jason Wang Signed-off-by: David S. Miller diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index 95a1332..2476923 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -299,6 +299,9 @@ static struct macvtap_queue *macvtap_get_queue(struct net_device *dev, if (!numvtaps) goto out; + if (numvtaps == 1) + goto single; + /* Check if we can use flow to select a queue */ rxq = skb_get_hash(skb); if (rxq) { @@ -316,6 +319,7 @@ static struct macvtap_queue *macvtap_get_queue(struct net_device *dev, goto out; } +single: tap = rcu_dereference(vlan->taps[0]); out: return tap; -- cgit v0.10.2 From 362899b8725b35e32802882c67f99cbf42bce2af Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Fri, 15 Jul 2016 03:46:31 -0400 Subject: macvtap: switch to use skb array This patch switch to use skb array instead of sk_receive_queue to avoid spinlock contentions. Tests shows about 21% improvements for guest rx pps: Before: 1472731 pkts/s After: 1786289 pkts/s Signed-off-by: Jason Wang Signed-off-by: David S. Miller diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index 2476923..9204d19 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -21,6 +21,7 @@ #include #include #include +#include /* * A macvtap queue is the central object of this driver, it connects @@ -43,6 +44,7 @@ struct macvtap_queue { u16 queue_index; bool enabled; struct list_head next; + struct skb_array skb_array; }; #define MACVTAP_FEATURES (IFF_VNET_HDR | IFF_MULTI_QUEUE) @@ -273,6 +275,7 @@ static void macvtap_put_queue(struct macvtap_queue *q) rtnl_unlock(); synchronize_rcu(); + skb_array_cleanup(&q->skb_array); sock_put(&q->sk); } @@ -366,7 +369,7 @@ static rx_handler_result_t macvtap_handle_frame(struct sk_buff **pskb) if (!q) return RX_HANDLER_PASS; - if (skb_queue_len(&q->sk.sk_receive_queue) >= dev->tx_queue_len) + if (__skb_array_full(&q->skb_array)) goto drop; skb_push(skb, ETH_HLEN); @@ -384,7 +387,8 @@ static rx_handler_result_t macvtap_handle_frame(struct sk_buff **pskb) goto drop; if (!segs) { - skb_queue_tail(&q->sk.sk_receive_queue, skb); + if (skb_array_produce(&q->skb_array, skb)) + goto drop; goto wake_up; } @@ -393,7 +397,11 @@ static rx_handler_result_t macvtap_handle_frame(struct sk_buff **pskb) struct sk_buff *nskb = segs->next; segs->next = NULL; - skb_queue_tail(&q->sk.sk_receive_queue, segs); + if (skb_array_produce(&q->skb_array, segs)) { + kfree_skb(segs); + kfree_skb_list(nskb); + break; + } segs = nskb; } } else { @@ -406,7 +414,8 @@ static rx_handler_result_t macvtap_handle_frame(struct sk_buff **pskb) !(features & NETIF_F_CSUM_MASK) && skb_checksum_help(skb)) goto drop; - skb_queue_tail(&q->sk.sk_receive_queue, skb); + if (skb_array_produce(&q->skb_array, skb)) + goto drop; } wake_up: @@ -523,7 +532,11 @@ static void macvtap_sock_write_space(struct sock *sk) static void macvtap_sock_destruct(struct sock *sk) { - skb_queue_purge(&sk->sk_receive_queue); + struct macvtap_queue *q = container_of(sk, struct macvtap_queue, sk); + struct sk_buff *skb; + + while ((skb = skb_array_consume(&q->skb_array)) != NULL) + kfree(skb); } static int macvtap_open(struct inode *inode, struct file *file) @@ -536,13 +549,13 @@ static int macvtap_open(struct inode *inode, struct file *file) rtnl_lock(); dev = dev_get_by_macvtap_minor(iminor(inode)); if (!dev) - goto out; + goto err; err = -ENOMEM; q = (struct macvtap_queue *)sk_alloc(net, AF_UNSPEC, GFP_KERNEL, &macvtap_proto, 0); if (!q) - goto out; + goto err; RCU_INIT_POINTER(q->sock.wq, &q->wq); init_waitqueue_head(&q->wq.wait); @@ -566,11 +579,24 @@ static int macvtap_open(struct inode *inode, struct file *file) if ((dev->features & NETIF_F_HIGHDMA) && (dev->features & NETIF_F_SG)) sock_set_flag(&q->sk, SOCK_ZEROCOPY); + err = -ENOMEM; + if (skb_array_init(&q->skb_array, dev->tx_queue_len, GFP_KERNEL)) + goto err_array; + err = macvtap_set_queue(dev, file, q); if (err) - sock_put(&q->sk); + goto err_queue; -out: + dev_put(dev); + + rtnl_unlock(); + return err; + +err_queue: + skb_array_cleanup(&q->skb_array); +err_array: + sock_put(&q->sk); +err: if (dev) dev_put(dev); @@ -596,7 +622,7 @@ static unsigned int macvtap_poll(struct file *file, poll_table * wait) mask = 0; poll_wait(file, &q->wq.wait, wait); - if (!skb_queue_empty(&q->sk.sk_receive_queue)) + if (!skb_array_empty(&q->skb_array)) mask |= POLLIN | POLLRDNORM; if (sock_writeable(&q->sk) || @@ -856,7 +882,7 @@ static ssize_t macvtap_do_read(struct macvtap_queue *q, TASK_INTERRUPTIBLE); /* Read frames from the queue */ - skb = skb_dequeue(&q->sk.sk_receive_queue); + skb = skb_array_consume(&q->skb_array); if (skb) break; if (noblock) { @@ -1180,10 +1206,18 @@ static int macvtap_recvmsg(struct socket *sock, struct msghdr *m, return ret; } +static int macvtap_peek_len(struct socket *sock) +{ + struct macvtap_queue *q = container_of(sock, struct macvtap_queue, + sock); + return skb_array_peek_len(&q->skb_array); +} + /* Ops structure to mimic raw sockets with tun */ static const struct proto_ops macvtap_socket_ops = { .sendmsg = macvtap_sendmsg, .recvmsg = macvtap_recvmsg, + .peek_len = macvtap_peek_len, }; /* Get an underlying socket object from tun file. Returns error unless file is @@ -1202,6 +1236,28 @@ struct socket *macvtap_get_socket(struct file *file) } EXPORT_SYMBOL_GPL(macvtap_get_socket); +static int macvtap_queue_resize(struct macvlan_dev *vlan) +{ + struct net_device *dev = vlan->dev; + struct macvtap_queue *q; + struct skb_array **arrays; + int n = vlan->numqueues; + int ret, i = 0; + + arrays = kmalloc(sizeof *arrays * n, GFP_KERNEL); + if (!arrays) + return -ENOMEM; + + list_for_each_entry(q, &vlan->queue_list, next) + arrays[i++] = &q->skb_array; + + ret = skb_array_resize_multiple(arrays, n, + dev->tx_queue_len, GFP_KERNEL); + + kfree(arrays); + return ret; +} + static int macvtap_device_event(struct notifier_block *unused, unsigned long event, void *ptr) { @@ -1249,6 +1305,10 @@ static int macvtap_device_event(struct notifier_block *unused, device_destroy(&macvtap_class, devt); macvtap_free_minor(vlan); break; + case NETDEV_CHANGE_TX_QUEUE_LEN: + if (macvtap_queue_resize(vlan)) + return NOTIFY_BAD; + break; } return NOTIFY_DONE; -- cgit v0.10.2 From 02a198777e646a12a8aabae5639f1d33d81d79ef Mon Sep 17 00:00:00 2001 From: Jesper Dangaard Brouer Date: Fri, 15 Jul 2016 23:55:20 +0200 Subject: net: fixup for tracepoint napi:napi_poll The recent change to tracepoint napi:napi_poll changed the order of the parameters that perf scripts sees, the printk was correct. The problem was that the new parameters (work and budget) were pushed in front of dev_name. The new parameters obviously need to be appended to keep backward compatible. Fixes: 1db19db7f5ff ("net: tracepoint napi:napi_poll add work and budget") Signed-off-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller diff --git a/include/trace/events/napi.h b/include/trace/events/napi.h index 118ed77..0b9e513 100644 --- a/include/trace/events/napi.h +++ b/include/trace/events/napi.h @@ -18,16 +18,16 @@ TRACE_EVENT(napi_poll, TP_STRUCT__entry( __field( struct napi_struct *, napi) + __string( dev_name, napi->dev ? napi->dev->name : NO_DEV) __field( int, work) __field( int, budget) - __string( dev_name, napi->dev ? napi->dev->name : NO_DEV) ), TP_fast_assign( __entry->napi = napi; + __assign_str(dev_name, napi->dev ? napi->dev->name : NO_DEV); __entry->work = work; __entry->budget = budget; - __assign_str(dev_name, napi->dev ? napi->dev->name : NO_DEV); ), TP_printk("napi poll on napi struct %p for device %s work %d budget %d", -- cgit v0.10.2 From a4fc549af8addaf0d99a85b4dd1b34a54d12d05e Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 14 Jul 2016 15:20:46 +0200 Subject: net: ethernet: tc35815: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c index 74e6719..cdf8d58 100644 --- a/drivers/net/ethernet/toshiba/tc35815.c +++ b/drivers/net/ethernet/toshiba/tc35815.c @@ -405,7 +405,6 @@ struct tc35815_local { spinlock_t rx_lock; struct mii_bus *mii_bus; - struct phy_device *phy_dev; int duplex; int speed; int link; @@ -539,7 +538,7 @@ static int tc_mdio_write(struct mii_bus *bus, int mii_id, int regnum, u16 val) static void tc_handle_link_change(struct net_device *dev) { struct tc35815_local *lp = netdev_priv(dev); - struct phy_device *phydev = lp->phy_dev; + struct phy_device *phydev = dev->phydev; unsigned long flags; int status_change = 0; @@ -645,7 +644,6 @@ static int tc_mii_probe(struct net_device *dev) lp->link = 0; lp->speed = 0; lp->duplex = -1; - lp->phy_dev = phydev; return 0; } @@ -853,7 +851,7 @@ static void tc35815_remove_one(struct pci_dev *pdev) struct net_device *dev = pci_get_drvdata(pdev); struct tc35815_local *lp = netdev_priv(dev); - phy_disconnect(lp->phy_dev); + phy_disconnect(dev->phydev); mdiobus_unregister(lp->mii_bus); mdiobus_free(lp->mii_bus); unregister_netdev(dev); @@ -1143,8 +1141,8 @@ static void tc35815_restart(struct net_device *dev) struct tc35815_local *lp = netdev_priv(dev); int ret; - if (lp->phy_dev) { - ret = phy_init_hw(lp->phy_dev); + if (dev->phydev) { + ret = phy_init_hw(dev->phydev); if (ret) printk(KERN_ERR "%s: PHY init failed.\n", dev->name); } @@ -1236,7 +1234,7 @@ tc35815_open(struct net_device *dev) netif_carrier_off(dev); /* schedule a link state check */ - phy_start(lp->phy_dev); + phy_start(dev->phydev); /* We are now ready to accept transmit requeusts from * the queueing layer of the networking. @@ -1819,8 +1817,8 @@ tc35815_close(struct net_device *dev) netif_stop_queue(dev); napi_disable(&lp->napi); - if (lp->phy_dev) - phy_stop(lp->phy_dev); + if (dev->phydev) + phy_stop(dev->phydev); cancel_work_sync(&lp->restart_work); /* Flush the Tx and disable Rx here. */ @@ -1948,20 +1946,16 @@ static void tc35815_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo * static int tc35815_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct tc35815_local *lp = netdev_priv(dev); - - if (!lp->phy_dev) + if (!dev->phydev) return -ENODEV; - return phy_ethtool_gset(lp->phy_dev, cmd); + return phy_ethtool_gset(dev->phydev, cmd); } static int tc35815_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct tc35815_local *lp = netdev_priv(dev); - - if (!lp->phy_dev) + if (!dev->phydev) return -ENODEV; - return phy_ethtool_sset(lp->phy_dev, cmd); + return phy_ethtool_sset(dev->phydev, cmd); } static u32 tc35815_get_msglevel(struct net_device *dev) @@ -2025,13 +2019,11 @@ static const struct ethtool_ops tc35815_ethtool_ops = { static int tc35815_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct tc35815_local *lp = netdev_priv(dev); - if (!netif_running(dev)) return -EINVAL; - if (!lp->phy_dev) + if (!dev->phydev) return -ENODEV; - return phy_mii_ioctl(lp->phy_dev, rq, cmd); + return phy_mii_ioctl(dev->phydev, rq, cmd); } static void tc35815_chip_reset(struct net_device *dev) @@ -2116,7 +2108,7 @@ static void tc35815_chip_init(struct net_device *dev) if (lp->chiptype == TC35815_TX4939) txctl &= ~Tx_EnLCarr; /* WORKAROUND: ignore LostCrS in full duplex operation */ - if (!lp->phy_dev || !lp->link || lp->duplex == DUPLEX_FULL) + if (!dev->phydev || !lp->link || lp->duplex == DUPLEX_FULL) txctl &= ~Tx_EnLCarr; tc_writel(txctl, &tr->Tx_Ctl); } @@ -2132,8 +2124,8 @@ static int tc35815_suspend(struct pci_dev *pdev, pm_message_t state) if (!netif_running(dev)) return 0; netif_device_detach(dev); - if (lp->phy_dev) - phy_stop(lp->phy_dev); + if (dev->phydev) + phy_stop(dev->phydev); spin_lock_irqsave(&lp->lock, flags); tc35815_chip_reset(dev); spin_unlock_irqrestore(&lp->lock, flags); @@ -2144,7 +2136,6 @@ static int tc35815_suspend(struct pci_dev *pdev, pm_message_t state) static int tc35815_resume(struct pci_dev *pdev) { struct net_device *dev = pci_get_drvdata(pdev); - struct tc35815_local *lp = netdev_priv(dev); pci_restore_state(pdev); if (!netif_running(dev)) @@ -2152,8 +2143,8 @@ static int tc35815_resume(struct pci_dev *pdev) pci_set_power_state(pdev, PCI_D0); tc35815_restart(dev); netif_carrier_off(dev); - if (lp->phy_dev) - phy_start(lp->phy_dev); + if (dev->phydev) + phy_start(dev->phydev); netif_device_attach(dev); return 0; } -- cgit v0.10.2 From 3a11d9ef65e788fffe348b071c965186af551368 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 14 Jul 2016 15:20:47 +0200 Subject: net: ethernet: tc35815: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/toshiba/tc35815.c b/drivers/net/ethernet/toshiba/tc35815.c index cdf8d58..5b01b3f 100644 --- a/drivers/net/ethernet/toshiba/tc35815.c +++ b/drivers/net/ethernet/toshiba/tc35815.c @@ -1944,20 +1944,6 @@ static void tc35815_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo * strlcpy(info->bus_info, pci_name(lp->pci_dev), sizeof(info->bus_info)); } -static int tc35815_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - if (!dev->phydev) - return -ENODEV; - return phy_ethtool_gset(dev->phydev, cmd); -} - -static int tc35815_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - if (!dev->phydev) - return -ENODEV; - return phy_ethtool_sset(dev->phydev, cmd); -} - static u32 tc35815_get_msglevel(struct net_device *dev) { struct tc35815_local *lp = netdev_priv(dev); @@ -2007,14 +1993,14 @@ static void tc35815_get_strings(struct net_device *dev, u32 stringset, u8 *data) static const struct ethtool_ops tc35815_ethtool_ops = { .get_drvinfo = tc35815_get_drvinfo, - .get_settings = tc35815_get_settings, - .set_settings = tc35815_set_settings, .get_link = ethtool_op_get_link, .get_msglevel = tc35815_get_msglevel, .set_msglevel = tc35815_set_msglevel, .get_strings = tc35815_get_strings, .get_sset_count = tc35815_get_sset_count, .get_ethtool_stats = tc35815_get_ethtool_stats, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int tc35815_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) -- cgit v0.10.2 From b1b7dcffed3281908322120f6f7abc879022dc68 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 14 Jul 2016 19:45:57 +0200 Subject: net: ethernet: xilinx: axienet: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet.h b/drivers/net/ethernet/xilinx/xilinx_axienet.h index 9ead4e2..af27f7d 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet.h +++ b/drivers/net/ethernet/xilinx/xilinx_axienet.h @@ -382,7 +382,6 @@ struct axidma_bd { * struct axienet_local - axienet private per device data * @ndev: Pointer for net_device to which it will be attached. * @dev: Pointer to device structure - * @phy_dev: Pointer to PHY device structure attached to the axienet_local * @phy_node: Pointer to device node structure * @mii_bus: Pointer to MII bus structure * @regs: Base address for the axienet_local device address space @@ -420,7 +419,6 @@ struct axienet_local { struct device *dev; /* Connection to PHY device */ - struct phy_device *phy_dev; /* Pointer to PHY device */ struct device_node *phy_node; /* MDIO bus data */ diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c index 8c7f5be..71abd00 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c @@ -525,7 +525,7 @@ static void axienet_adjust_link(struct net_device *ndev) u32 link_state; u32 setspeed = 1; struct axienet_local *lp = netdev_priv(ndev); - struct phy_device *phy = lp->phy_dev; + struct phy_device *phy = ndev->phydev; link_state = phy->speed | (phy->duplex << 1) | phy->link; if (lp->last_link != link_state) { @@ -911,6 +911,7 @@ static int axienet_open(struct net_device *ndev) { int ret, mdio_mcreg; struct axienet_local *lp = netdev_priv(ndev); + struct phy_device *phydev = NULL; dev_dbg(&ndev->dev, "axienet_open()\n"); @@ -934,19 +935,19 @@ static int axienet_open(struct net_device *ndev) if (lp->phy_node) { if (lp->phy_type == XAE_PHY_TYPE_GMII) { - lp->phy_dev = of_phy_connect(lp->ndev, lp->phy_node, - axienet_adjust_link, 0, - PHY_INTERFACE_MODE_GMII); + phydev = of_phy_connect(lp->ndev, lp->phy_node, + axienet_adjust_link, 0, + PHY_INTERFACE_MODE_GMII); } else if (lp->phy_type == XAE_PHY_TYPE_RGMII_2_0) { - lp->phy_dev = of_phy_connect(lp->ndev, lp->phy_node, - axienet_adjust_link, 0, - PHY_INTERFACE_MODE_RGMII_ID); + phydev = of_phy_connect(lp->ndev, lp->phy_node, + axienet_adjust_link, 0, + PHY_INTERFACE_MODE_RGMII_ID); } - if (!lp->phy_dev) + if (!phydev) dev_err(lp->dev, "of_phy_connect() failed\n"); else - phy_start(lp->phy_dev); + phy_start(phydev); } /* Enable tasklets for Axi DMA error handling */ @@ -967,9 +968,8 @@ static int axienet_open(struct net_device *ndev) err_rx_irq: free_irq(lp->tx_irq, ndev); err_tx_irq: - if (lp->phy_dev) - phy_disconnect(lp->phy_dev); - lp->phy_dev = NULL; + if (phydev) + phy_disconnect(phydev); tasklet_kill(&lp->dma_err_tasklet); dev_err(lp->dev, "request_irq() failed\n"); return ret; @@ -1006,9 +1006,8 @@ static int axienet_stop(struct net_device *ndev) free_irq(lp->tx_irq, ndev); free_irq(lp->rx_irq, ndev); - if (lp->phy_dev) - phy_disconnect(lp->phy_dev); - lp->phy_dev = NULL; + if (ndev->phydev) + phy_disconnect(ndev->phydev); axienet_dma_bd_release(ndev); return 0; @@ -1092,8 +1091,7 @@ static const struct net_device_ops axienet_netdev_ops = { static int axienet_ethtools_get_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) { - struct axienet_local *lp = netdev_priv(ndev); - struct phy_device *phydev = lp->phy_dev; + struct phy_device *phydev = ndev->phydev; if (!phydev) return -ENODEV; return phy_ethtool_gset(phydev, ecmd); @@ -1115,8 +1113,7 @@ static int axienet_ethtools_get_settings(struct net_device *ndev, static int axienet_ethtools_set_settings(struct net_device *ndev, struct ethtool_cmd *ecmd) { - struct axienet_local *lp = netdev_priv(ndev); - struct phy_device *phydev = lp->phy_dev; + struct phy_device *phydev = ndev->phydev; if (!phydev) return -ENODEV; return phy_ethtool_sset(phydev, ecmd); -- cgit v0.10.2 From 6e3848404d418f4200790c3bdbab05e58de7976b Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 14 Jul 2016 19:45:58 +0200 Subject: net: ethernet: xilinx: axienet: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c index 71abd00..9c82993 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c @@ -1077,49 +1077,6 @@ static const struct net_device_ops axienet_netdev_ops = { }; /** - * axienet_ethtools_get_settings - Get Axi Ethernet settings related to PHY. - * @ndev: Pointer to net_device structure - * @ecmd: Pointer to ethtool_cmd structure - * - * This implements ethtool command for getting PHY settings. If PHY could - * not be found, the function returns -ENODEV. This function calls the - * relevant PHY ethtool API to get the PHY settings. - * Issue "ethtool ethX" under linux prompt to execute this function. - * - * Return: 0 on success, -ENODEV if PHY doesn't exist - */ -static int axienet_ethtools_get_settings(struct net_device *ndev, - struct ethtool_cmd *ecmd) -{ - struct phy_device *phydev = ndev->phydev; - if (!phydev) - return -ENODEV; - return phy_ethtool_gset(phydev, ecmd); -} - -/** - * axienet_ethtools_set_settings - Set PHY settings as passed in the argument. - * @ndev: Pointer to net_device structure - * @ecmd: Pointer to ethtool_cmd structure - * - * This implements ethtool command for setting various PHY settings. If PHY - * could not be found, the function returns -ENODEV. This function calls the - * relevant PHY ethtool API to set the PHY. - * Issue e.g. "ethtool -s ethX speed 1000" under linux prompt to execute this - * function. - * - * Return: 0 on success, -ENODEV if PHY doesn't exist - */ -static int axienet_ethtools_set_settings(struct net_device *ndev, - struct ethtool_cmd *ecmd) -{ - struct phy_device *phydev = ndev->phydev; - if (!phydev) - return -ENODEV; - return phy_ethtool_sset(phydev, ecmd); -} - -/** * axienet_ethtools_get_drvinfo - Get various Axi Ethernet driver information. * @ndev: Pointer to net_device structure * @ed: Pointer to ethtool_drvinfo structure @@ -1341,8 +1298,6 @@ static int axienet_ethtools_set_coalesce(struct net_device *ndev, } static struct ethtool_ops axienet_ethtool_ops = { - .get_settings = axienet_ethtools_get_settings, - .set_settings = axienet_ethtools_set_settings, .get_drvinfo = axienet_ethtools_get_drvinfo, .get_regs_len = axienet_ethtools_get_regs_len, .get_regs = axienet_ethtools_get_regs, @@ -1351,6 +1306,8 @@ static struct ethtool_ops axienet_ethtool_ops = { .set_pauseparam = axienet_ethtools_set_pauseparam, .get_coalesce = axienet_ethtools_get_coalesce, .set_coalesce = axienet_ethtools_set_coalesce, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /** -- cgit v0.10.2 From 80721e7fa68f914a21b2685a01399b3aa80ac116 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 14 Jul 2016 23:44:52 +0200 Subject: net: ethernet: pasemi_mac: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/pasemi/pasemi_mac.c b/drivers/net/ethernet/pasemi/pasemi_mac.c index af54df5..2f4a837 100644 --- a/drivers/net/ethernet/pasemi/pasemi_mac.c +++ b/drivers/net/ethernet/pasemi/pasemi_mac.c @@ -989,7 +989,7 @@ static void pasemi_adjust_link(struct net_device *dev) unsigned int flags; unsigned int new_flags; - if (!mac->phydev->link) { + if (!dev->phydev->link) { /* If no link, MAC speed settings don't matter. Just report * link down and return. */ @@ -1010,10 +1010,10 @@ static void pasemi_adjust_link(struct net_device *dev) new_flags = flags & ~(PAS_MAC_CFG_PCFG_HD | PAS_MAC_CFG_PCFG_SPD_M | PAS_MAC_CFG_PCFG_TSR_M); - if (!mac->phydev->duplex) + if (!dev->phydev->duplex) new_flags |= PAS_MAC_CFG_PCFG_HD; - switch (mac->phydev->speed) { + switch (dev->phydev->speed) { case 1000: new_flags |= PAS_MAC_CFG_PCFG_SPD_1G | PAS_MAC_CFG_PCFG_TSR_1G; @@ -1027,15 +1027,15 @@ static void pasemi_adjust_link(struct net_device *dev) PAS_MAC_CFG_PCFG_TSR_10M; break; default: - printk("Unsupported speed %d\n", mac->phydev->speed); + printk("Unsupported speed %d\n", dev->phydev->speed); } /* Print on link or speed/duplex change */ - msg = mac->link != mac->phydev->link || flags != new_flags; + msg = mac->link != dev->phydev->link || flags != new_flags; - mac->duplex = mac->phydev->duplex; - mac->speed = mac->phydev->speed; - mac->link = mac->phydev->link; + mac->duplex = dev->phydev->duplex; + mac->speed = dev->phydev->speed; + mac->link = dev->phydev->link; if (new_flags != flags) write_mac_reg(mac, PAS_MAC_CFG_PCFG, new_flags); @@ -1067,8 +1067,6 @@ static int pasemi_mac_phy_init(struct net_device *dev) return -ENODEV; } - mac->phydev = phydev; - return 0; } @@ -1198,8 +1196,8 @@ static int pasemi_mac_open(struct net_device *dev) goto out_rx_int; } - if (mac->phydev) - phy_start(mac->phydev); + if (dev->phydev) + phy_start(dev->phydev); setup_timer(&mac->tx->clean_timer, pasemi_mac_tx_timer, (unsigned long)mac->tx); @@ -1293,9 +1291,9 @@ static int pasemi_mac_close(struct net_device *dev) rxch = rx_ring(mac)->chan.chno; txch = tx_ring(mac)->chan.chno; - if (mac->phydev) { - phy_stop(mac->phydev); - phy_disconnect(mac->phydev); + if (dev->phydev) { + phy_stop(dev->phydev); + phy_disconnect(dev->phydev); } del_timer_sync(&mac->tx->clean_timer); diff --git a/drivers/net/ethernet/pasemi/pasemi_mac.h b/drivers/net/ethernet/pasemi/pasemi_mac.h index 161c99a..7c47e26 100644 --- a/drivers/net/ethernet/pasemi/pasemi_mac.h +++ b/drivers/net/ethernet/pasemi/pasemi_mac.h @@ -70,7 +70,6 @@ struct pasemi_mac { struct pci_dev *pdev; struct pci_dev *dma_pdev; struct pci_dev *iob_pdev; - struct phy_device *phydev; struct napi_struct napi; int bufsz; /* RX ring buffer size */ diff --git a/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c b/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c index f046bfc..e99cf42 100644 --- a/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c +++ b/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c @@ -66,8 +66,7 @@ static int pasemi_mac_ethtool_get_settings(struct net_device *netdev, struct ethtool_cmd *cmd) { - struct pasemi_mac *mac = netdev_priv(netdev); - struct phy_device *phydev = mac->phydev; + struct phy_device *phydev = netdev->phydev; if (!phydev) return -EOPNOTSUPP; @@ -79,8 +78,7 @@ static int pasemi_mac_ethtool_set_settings(struct net_device *netdev, struct ethtool_cmd *cmd) { - struct pasemi_mac *mac = netdev_priv(netdev); - struct phy_device *phydev = mac->phydev; + struct phy_device *phydev = netdev->phydev; if (!phydev) return -EOPNOTSUPP; -- cgit v0.10.2 From 6cf285de0231e53057726aea1fb87ab772765cb7 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Thu, 14 Jul 2016 23:44:53 +0200 Subject: net: ethernet: pasemi_mac: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c b/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c index e99cf42..d0afc2b 100644 --- a/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c +++ b/drivers/net/ethernet/pasemi/pasemi_mac_ethtool.c @@ -62,30 +62,6 @@ static struct { { "tx-1024-1518-byte-packets" }, }; -static int -pasemi_mac_ethtool_get_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = netdev->phydev; - - if (!phydev) - return -EOPNOTSUPP; - - return phy_ethtool_gset(phydev, cmd); -} - -static int -pasemi_mac_ethtool_set_settings(struct net_device *netdev, - struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = netdev->phydev; - - if (!phydev) - return -EOPNOTSUPP; - - return phy_ethtool_sset(phydev, cmd); -} - static u32 pasemi_mac_ethtool_get_msglevel(struct net_device *netdev) { @@ -143,8 +119,6 @@ static void pasemi_mac_get_strings(struct net_device *netdev, u32 stringset, } const struct ethtool_ops pasemi_mac_ethtool_ops = { - .get_settings = pasemi_mac_ethtool_get_settings, - .set_settings = pasemi_mac_ethtool_set_settings, .get_msglevel = pasemi_mac_ethtool_get_msglevel, .set_msglevel = pasemi_mac_ethtool_set_msglevel, .get_link = ethtool_op_get_link, @@ -152,5 +126,7 @@ const struct ethtool_ops pasemi_mac_ethtool_ops = { .get_strings = pasemi_mac_get_strings, .get_sset_count = pasemi_mac_get_sset_count, .get_ethtool_stats = pasemi_mac_get_ethtool_stats, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; -- cgit v0.10.2 From 11331fc294da76649a18a79c122efb04835fcad1 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 15 Jul 2016 09:59:11 +0200 Subject: net: ethernet: ethoc: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Reviewed-by: Tobias Klauser Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index 4edb98c..317168a 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -192,7 +192,6 @@ MODULE_PARM_DESC(buffer_size, "DMA buffer allocation size"); * @napi: NAPI structure * @msg_enable: device state flags * @lock: device lock - * @phy: attached PHY * @mdio: MDIO bus for PHY access * @phy_id: address of attached PHY */ @@ -219,7 +218,6 @@ struct ethoc { spinlock_t lock; - struct phy_device *phy; struct mii_bus *mdio; struct clk *clk; s8 phy_id; @@ -694,7 +692,6 @@ static int ethoc_mdio_probe(struct net_device *dev) return err; } - priv->phy = phy; phy->advertising &= ~(ADVERTISED_1000baseT_Full | ADVERTISED_1000baseT_Half); phy->supported &= ~(SUPPORTED_1000baseT_Full | @@ -724,7 +721,7 @@ static int ethoc_open(struct net_device *dev) netif_start_queue(dev); } - phy_start(priv->phy); + phy_start(dev->phydev); napi_enable(&priv->napi); if (netif_msg_ifup(priv)) { @@ -741,8 +738,8 @@ static int ethoc_stop(struct net_device *dev) napi_disable(&priv->napi); - if (priv->phy) - phy_stop(priv->phy); + if (dev->phydev) + phy_stop(dev->phydev); ethoc_disable_rx_and_tx(priv); free_irq(dev->irq, dev); @@ -770,7 +767,7 @@ static int ethoc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) if (!phy) return -ENODEV; } else { - phy = priv->phy; + phy = dev->phydev; } return phy_mii_ioctl(phy, ifr, cmd); @@ -899,8 +896,7 @@ out: static int ethoc_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct ethoc *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phy; + struct phy_device *phydev = dev->phydev; if (!phydev) return -EOPNOTSUPP; @@ -910,8 +906,7 @@ static int ethoc_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) static int ethoc_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct ethoc *priv = netdev_priv(dev); - struct phy_device *phydev = priv->phy; + struct phy_device *phydev = dev->phydev; if (!phydev) return -EOPNOTSUPP; @@ -1261,8 +1256,7 @@ static int ethoc_remove(struct platform_device *pdev) if (netdev) { netif_napi_del(&priv->napi); - phy_disconnect(priv->phy); - priv->phy = NULL; + phy_disconnect(netdev->phydev); if (priv->mdio) { mdiobus_unregister(priv->mdio); -- cgit v0.10.2 From 87e544bfd52214bab515343d93510a2ad296edff Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 15 Jul 2016 09:59:12 +0200 Subject: net: ethernet: ethoc: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Reviewed-by: Tobias Klauser Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c index 317168a..37a39b4 100644 --- a/drivers/net/ethernet/ethoc.c +++ b/drivers/net/ethernet/ethoc.c @@ -894,26 +894,6 @@ out: return NETDEV_TX_OK; } -static int ethoc_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = dev->phydev; - - if (!phydev) - return -EOPNOTSUPP; - - return phy_ethtool_gset(phydev, cmd); -} - -static int ethoc_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - struct phy_device *phydev = dev->phydev; - - if (!phydev) - return -EOPNOTSUPP; - - return phy_ethtool_sset(phydev, cmd); -} - static int ethoc_get_regs_len(struct net_device *netdev) { return ETH_END; @@ -978,14 +958,14 @@ static int ethoc_set_ringparam(struct net_device *dev, } const struct ethtool_ops ethoc_ethtool_ops = { - .get_settings = ethoc_get_settings, - .set_settings = ethoc_set_settings, .get_regs_len = ethoc_get_regs_len, .get_regs = ethoc_get_regs, .get_link = ethtool_op_get_link, .get_ringparam = ethoc_get_ringparam, .set_ringparam = ethoc_set_ringparam, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops ethoc_netdev_ops = { -- cgit v0.10.2 From 5d872c506fbb90c9ee43ba90ddc56aebec124aa9 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 15 Jul 2016 10:36:20 +0200 Subject: net: ethernet: smsc9420: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c index 8594b9e..3e5f533 100644 --- a/drivers/net/ethernet/smsc/smsc9420.c +++ b/drivers/net/ethernet/smsc/smsc9420.c @@ -76,7 +76,6 @@ struct smsc9420_pdata { bool rx_csum; u32 msg_enable; - struct phy_device *phy_dev; struct mii_bus *mii_bus; int last_duplex; int last_carrier; @@ -226,36 +225,30 @@ static int smsc9420_eeprom_reload(struct smsc9420_pdata *pd) /* Standard ioctls for mii-tool */ static int smsc9420_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { - struct smsc9420_pdata *pd = netdev_priv(dev); - - if (!netif_running(dev) || !pd->phy_dev) + if (!netif_running(dev) || !dev->phydev) return -EINVAL; - return phy_mii_ioctl(pd->phy_dev, ifr, cmd); + return phy_mii_ioctl(dev->phydev, ifr, cmd); } static int smsc9420_ethtool_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct smsc9420_pdata *pd = netdev_priv(dev); - - if (!pd->phy_dev) + if (!dev->phydev) return -ENODEV; cmd->maxtxpkt = 1; cmd->maxrxpkt = 1; - return phy_ethtool_gset(pd->phy_dev, cmd); + return phy_ethtool_gset(dev->phydev, cmd); } static int smsc9420_ethtool_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct smsc9420_pdata *pd = netdev_priv(dev); - - if (!pd->phy_dev) + if (!dev->phydev) return -ENODEV; - return phy_ethtool_sset(pd->phy_dev, cmd); + return phy_ethtool_sset(dev->phydev, cmd); } static void smsc9420_ethtool_get_drvinfo(struct net_device *netdev, @@ -283,12 +276,10 @@ static void smsc9420_ethtool_set_msglevel(struct net_device *netdev, u32 data) static int smsc9420_ethtool_nway_reset(struct net_device *netdev) { - struct smsc9420_pdata *pd = netdev_priv(netdev); - - if (!pd->phy_dev) + if (!netdev->phydev) return -ENODEV; - return phy_start_aneg(pd->phy_dev); + return phy_start_aneg(netdev->phydev); } static int smsc9420_ethtool_getregslen(struct net_device *dev) @@ -302,7 +293,7 @@ smsc9420_ethtool_getregs(struct net_device *dev, struct ethtool_regs *regs, void *buf) { struct smsc9420_pdata *pd = netdev_priv(dev); - struct phy_device *phy_dev = pd->phy_dev; + struct phy_device *phy_dev = dev->phydev; unsigned int i, j = 0; u32 *data = buf; @@ -736,7 +727,7 @@ static int smsc9420_stop(struct net_device *dev) ulong flags; BUG_ON(!pd); - BUG_ON(!pd->phy_dev); + BUG_ON(!dev->phydev); /* disable master interrupt */ spin_lock_irqsave(&pd->int_lock, flags); @@ -757,10 +748,9 @@ static int smsc9420_stop(struct net_device *dev) smsc9420_dmac_soft_reset(pd); - phy_stop(pd->phy_dev); + phy_stop(dev->phydev); - phy_disconnect(pd->phy_dev); - pd->phy_dev = NULL; + phy_disconnect(dev->phydev); mdiobus_unregister(pd->mii_bus); mdiobus_free(pd->mii_bus); @@ -1093,7 +1083,8 @@ static void smsc9420_set_multicast_list(struct net_device *dev) static void smsc9420_phy_update_flowcontrol(struct smsc9420_pdata *pd) { - struct phy_device *phy_dev = pd->phy_dev; + struct net_device *dev = pd->dev; + struct phy_device *phy_dev = dev->phydev; u32 flow; if (phy_dev->duplex == DUPLEX_FULL) { @@ -1122,7 +1113,7 @@ static void smsc9420_phy_update_flowcontrol(struct smsc9420_pdata *pd) static void smsc9420_phy_adjust_link(struct net_device *dev) { struct smsc9420_pdata *pd = netdev_priv(dev); - struct phy_device *phy_dev = pd->phy_dev; + struct phy_device *phy_dev = dev->phydev; int carrier; if (phy_dev->duplex != pd->last_duplex) { @@ -1155,7 +1146,7 @@ static int smsc9420_mii_probe(struct net_device *dev) struct smsc9420_pdata *pd = netdev_priv(dev); struct phy_device *phydev = NULL; - BUG_ON(pd->phy_dev); + BUG_ON(dev->phydev); /* Device only supports internal PHY at address 1 */ phydev = mdiobus_get_phy(pd->mii_bus, 1); @@ -1179,7 +1170,6 @@ static int smsc9420_mii_probe(struct net_device *dev) phy_attached_info(phydev); - pd->phy_dev = phydev; pd->last_duplex = -1; pd->last_carrier = -1; @@ -1440,7 +1430,7 @@ static int smsc9420_open(struct net_device *dev) } /* Bring the PHY up */ - phy_start(pd->phy_dev); + phy_start(dev->phydev); napi_enable(&pd->napi); -- cgit v0.10.2 From a1b198b71686ea7deaf070cbea4ef371937e5bdf Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 15 Jul 2016 10:36:21 +0200 Subject: net: ethernet: smsc9420: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c index 3e5f533..b7bfed4b 100644 --- a/drivers/net/ethernet/smsc/smsc9420.c +++ b/drivers/net/ethernet/smsc/smsc9420.c @@ -231,26 +231,6 @@ static int smsc9420_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) return phy_mii_ioctl(dev->phydev, ifr, cmd); } -static int smsc9420_ethtool_get_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - if (!dev->phydev) - return -ENODEV; - - cmd->maxtxpkt = 1; - cmd->maxrxpkt = 1; - return phy_ethtool_gset(dev->phydev, cmd); -} - -static int smsc9420_ethtool_set_settings(struct net_device *dev, - struct ethtool_cmd *cmd) -{ - if (!dev->phydev) - return -ENODEV; - - return phy_ethtool_sset(dev->phydev, cmd); -} - static void smsc9420_ethtool_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { @@ -434,8 +414,6 @@ static int smsc9420_ethtool_set_eeprom(struct net_device *dev, } static const struct ethtool_ops smsc9420_ethtool_ops = { - .get_settings = smsc9420_ethtool_get_settings, - .set_settings = smsc9420_ethtool_set_settings, .get_drvinfo = smsc9420_ethtool_get_drvinfo, .get_msglevel = smsc9420_ethtool_get_msglevel, .set_msglevel = smsc9420_ethtool_set_msglevel, @@ -447,6 +425,8 @@ static const struct ethtool_ops smsc9420_ethtool_ops = { .get_regs_len = smsc9420_ethtool_getregslen, .get_regs = smsc9420_ethtool_getregs, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /* Sets the device MAC address to dev_addr */ -- cgit v0.10.2 From 1e8b73896f15b5a83dfb18924169f23450bae064 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 15 Jul 2016 12:05:11 +0200 Subject: net: ethernet: amd: au1000_eth: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c index 20760e1..18fe333 100644 --- a/drivers/net/ethernet/amd/au1000_eth.c +++ b/drivers/net/ethernet/amd/au1000_eth.c @@ -412,13 +412,13 @@ static void au1000_adjust_link(struct net_device *dev) { struct au1000_private *aup = netdev_priv(dev); - struct phy_device *phydev = aup->phy_dev; + struct phy_device *phydev = dev->phydev; unsigned long flags; u32 reg; int status_change = 0; - BUG_ON(!aup->phy_dev); + BUG_ON(!phydev); spin_lock_irqsave(&aup->lock, flags); @@ -579,7 +579,6 @@ static int au1000_mii_probe(struct net_device *dev) aup->old_link = 0; aup->old_speed = 0; aup->old_duplex = -1; - aup->phy_dev = phydev; phy_attached_info(phydev); @@ -680,23 +679,19 @@ au1000_setup_hw_rings(struct au1000_private *aup, void __iomem *tx_base) static int au1000_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct au1000_private *aup = netdev_priv(dev); - - if (aup->phy_dev) - return phy_ethtool_gset(aup->phy_dev, cmd); + if (dev->phydev) + return phy_ethtool_gset(dev->phydev, cmd); return -EINVAL; } static int au1000_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct au1000_private *aup = netdev_priv(dev); - if (!capable(CAP_NET_ADMIN)) return -EPERM; - if (aup->phy_dev) - return phy_ethtool_sset(aup->phy_dev, cmd); + if (dev->phydev) + return phy_ethtool_sset(dev->phydev, cmd); return -EINVAL; } @@ -778,8 +773,8 @@ static int au1000_init(struct net_device *dev) #ifndef CONFIG_CPU_LITTLE_ENDIAN control |= MAC_BIG_ENDIAN; #endif - if (aup->phy_dev) { - if (aup->phy_dev->link && (DUPLEX_FULL == aup->phy_dev->duplex)) + if (dev->phydev) { + if (dev->phydev->link && (DUPLEX_FULL == dev->phydev->duplex)) control |= MAC_FULL_DUPLEX; else control |= MAC_DISABLE_RX_OWN; @@ -891,11 +886,10 @@ static int au1000_rx(struct net_device *dev) static void au1000_update_tx_stats(struct net_device *dev, u32 status) { - struct au1000_private *aup = netdev_priv(dev); struct net_device_stats *ps = &dev->stats; if (status & TX_FRAME_ABORTED) { - if (!aup->phy_dev || (DUPLEX_FULL == aup->phy_dev->duplex)) { + if (!dev->phydev || (DUPLEX_FULL == dev->phydev->duplex)) { if (status & (TX_JAB_TIMEOUT | TX_UNDERRUN)) { /* any other tx errors are only valid * in half duplex mode @@ -975,10 +969,10 @@ static int au1000_open(struct net_device *dev) return retval; } - if (aup->phy_dev) { + if (dev->phydev) { /* cause the PHY state machine to schedule a link state check */ - aup->phy_dev->state = PHY_CHANGELINK; - phy_start(aup->phy_dev); + dev->phydev->state = PHY_CHANGELINK; + phy_start(dev->phydev); } netif_start_queue(dev); @@ -995,8 +989,8 @@ static int au1000_close(struct net_device *dev) netif_dbg(aup, drv, dev, "close: dev=%p\n", dev); - if (aup->phy_dev) - phy_stop(aup->phy_dev); + if (dev->phydev) + phy_stop(dev->phydev); spin_lock_irqsave(&aup->lock, flags); @@ -1110,15 +1104,13 @@ static void au1000_multicast_list(struct net_device *dev) static int au1000_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - struct au1000_private *aup = netdev_priv(dev); - if (!netif_running(dev)) return -EINVAL; - if (!aup->phy_dev) + if (!dev->phydev) return -EINVAL; /* PHY not controllable */ - return phy_mii_ioctl(aup->phy_dev, rq, cmd); + return phy_mii_ioctl(dev->phydev, rq, cmd); } static const struct net_device_ops au1000_netdev_ops = { diff --git a/drivers/net/ethernet/amd/au1000_eth.h b/drivers/net/ethernet/amd/au1000_eth.h index ca53024..4c47c23 100644 --- a/drivers/net/ethernet/amd/au1000_eth.h +++ b/drivers/net/ethernet/amd/au1000_eth.h @@ -106,7 +106,6 @@ struct au1000_private { int old_speed; int old_duplex; - struct phy_device *phy_dev; struct mii_bus *mii_bus; /* PHY configuration */ -- cgit v0.10.2 From b4cafd8c4909187cb17b358be60ec3b86e91af00 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 15 Jul 2016 12:05:12 +0200 Subject: net: ethernet: amd: au1000_eth: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. There was a check on CAP_NET_ADMIN in au1000_set_settings, but this check is already done in dev_ethtool, so no need to repeat it before calling the generic function. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/amd/au1000_eth.c b/drivers/net/ethernet/amd/au1000_eth.c index 18fe333..df66418 100644 --- a/drivers/net/ethernet/amd/au1000_eth.c +++ b/drivers/net/ethernet/amd/au1000_eth.c @@ -677,25 +677,6 @@ au1000_setup_hw_rings(struct au1000_private *aup, void __iomem *tx_base) * ethtool operations */ -static int au1000_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - if (dev->phydev) - return phy_ethtool_gset(dev->phydev, cmd); - - return -EINVAL; -} - -static int au1000_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - if (dev->phydev) - return phy_ethtool_sset(dev->phydev, cmd); - - return -EINVAL; -} - static void au1000_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { @@ -720,12 +701,12 @@ static u32 au1000_get_msglevel(struct net_device *dev) } static const struct ethtool_ops au1000_ethtool_ops = { - .get_settings = au1000_get_settings, - .set_settings = au1000_set_settings, .get_drvinfo = au1000_get_drvinfo, .get_link = ethtool_op_get_link, .get_msglevel = au1000_get_msglevel, .set_msglevel = au1000_set_msglevel, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; -- cgit v0.10.2 From b401a9bce7f3e73ebbe10639e0986f7f3b01e5ca Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 15 Jul 2016 12:39:01 +0200 Subject: net: ethernet: ti: cpmac: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phy in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpmac.c b/drivers/net/ethernet/ti/cpmac.c index 7eef45e..f48e43b 100644 --- a/drivers/net/ethernet/ti/cpmac.c +++ b/drivers/net/ethernet/ti/cpmac.c @@ -205,7 +205,6 @@ struct cpmac_priv { dma_addr_t dma_ring; void __iomem *regs; struct mii_bus *mii_bus; - struct phy_device *phy; char phy_name[MII_BUS_ID_SIZE + 3]; int oldlink, oldspeed, oldduplex; u32 msg_enable; @@ -830,35 +829,29 @@ static void cpmac_tx_timeout(struct net_device *dev) static int cpmac_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { - struct cpmac_priv *priv = netdev_priv(dev); - if (!(netif_running(dev))) return -EINVAL; - if (!priv->phy) + if (!dev->phydev) return -EINVAL; - return phy_mii_ioctl(priv->phy, ifr, cmd); + return phy_mii_ioctl(dev->phydev, ifr, cmd); } static int cpmac_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct cpmac_priv *priv = netdev_priv(dev); - - if (priv->phy) - return phy_ethtool_gset(priv->phy, cmd); + if (dev->phydev) + return phy_ethtool_gset(dev->phydev, cmd); return -EINVAL; } static int cpmac_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct cpmac_priv *priv = netdev_priv(dev); - if (!capable(CAP_NET_ADMIN)) return -EPERM; - if (priv->phy) - return phy_ethtool_sset(priv->phy, cmd); + if (dev->phydev) + return phy_ethtool_sset(dev->phydev, cmd); return -EINVAL; } @@ -914,16 +907,16 @@ static void cpmac_adjust_link(struct net_device *dev) int new_state = 0; spin_lock(&priv->lock); - if (priv->phy->link) { + if (dev->phydev->link) { netif_tx_start_all_queues(dev); - if (priv->phy->duplex != priv->oldduplex) { + if (dev->phydev->duplex != priv->oldduplex) { new_state = 1; - priv->oldduplex = priv->phy->duplex; + priv->oldduplex = dev->phydev->duplex; } - if (priv->phy->speed != priv->oldspeed) { + if (dev->phydev->speed != priv->oldspeed) { new_state = 1; - priv->oldspeed = priv->phy->speed; + priv->oldspeed = dev->phydev->speed; } if (!priv->oldlink) { @@ -938,7 +931,7 @@ static void cpmac_adjust_link(struct net_device *dev) } if (new_state && netif_msg_link(priv) && net_ratelimit()) - phy_print_status(priv->phy); + phy_print_status(dev->phydev); spin_unlock(&priv->lock); } @@ -1016,8 +1009,8 @@ static int cpmac_open(struct net_device *dev) cpmac_hw_start(dev); napi_enable(&priv->napi); - priv->phy->state = PHY_CHANGELINK; - phy_start(priv->phy); + dev->phydev->state = PHY_CHANGELINK; + phy_start(dev->phydev); return 0; @@ -1053,7 +1046,7 @@ static int cpmac_stop(struct net_device *dev) cancel_work_sync(&priv->reset_work); napi_disable(&priv->napi); - phy_stop(priv->phy); + phy_stop(dev->phydev); cpmac_hw_stop(dev); @@ -1106,6 +1099,7 @@ static int cpmac_probe(struct platform_device *pdev) struct cpmac_priv *priv; struct net_device *dev; struct plat_cpmac_data *pdata; + struct phy_device *phydev = NULL; pdata = dev_get_platdata(&pdev->dev); @@ -1162,14 +1156,14 @@ static int cpmac_probe(struct platform_device *pdev) snprintf(priv->phy_name, MII_BUS_ID_SIZE, PHY_ID_FMT, mdio_bus_id, phy_id); - priv->phy = phy_connect(dev, priv->phy_name, cpmac_adjust_link, - PHY_INTERFACE_MODE_MII); + phydev = phy_connect(dev, priv->phy_name, cpmac_adjust_link, + PHY_INTERFACE_MODE_MII); - if (IS_ERR(priv->phy)) { + if (IS_ERR(phydev)) { if (netif_msg_drv(priv)) dev_err(&pdev->dev, "Could not attach to PHY\n"); - rc = PTR_ERR(priv->phy); + rc = PTR_ERR(phydev); goto out; } -- cgit v0.10.2 From 7dc099345268154a6d5997743da4b08ff89bdca3 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 15 Jul 2016 12:39:02 +0200 Subject: net: ethernet: ti: cpmac: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. There was a check on CAP_NET_ADMIN in cpmac_set_settings, but this check is already done in dev_ethtool, so no need to repeat it before calling the generic function. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpmac.c b/drivers/net/ethernet/ti/cpmac.c index f48e43b..f86497c 100644 --- a/drivers/net/ethernet/ti/cpmac.c +++ b/drivers/net/ethernet/ti/cpmac.c @@ -837,25 +837,6 @@ static int cpmac_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) return phy_mii_ioctl(dev->phydev, ifr, cmd); } -static int cpmac_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - if (dev->phydev) - return phy_ethtool_gset(dev->phydev, cmd); - - return -EINVAL; -} - -static int cpmac_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - if (dev->phydev) - return phy_ethtool_sset(dev->phydev, cmd); - - return -EINVAL; -} - static void cpmac_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ring) { @@ -893,12 +874,12 @@ static void cpmac_get_drvinfo(struct net_device *dev, } static const struct ethtool_ops cpmac_ethtool_ops = { - .get_settings = cpmac_get_settings, - .set_settings = cpmac_set_settings, .get_drvinfo = cpmac_get_drvinfo, .get_link = ethtool_op_get_link, .get_ringparam = cpmac_get_ringparam, .set_ringparam = cpmac_set_ringparam, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static void cpmac_adjust_link(struct net_device *dev) -- cgit v0.10.2 From c380d37e97e783e36a924279fbd2f6837508546a Mon Sep 17 00:00:00 2001 From: Richard Sailer Date: Sat, 16 Jul 2016 04:04:34 +0200 Subject: tcp_timer.c: Add kernel-doc function descriptions This adds kernel-doc style descriptions for 6 functions and fixes 1 typo. Signed-off-by: Richard Sailer Signed-off-by: David S. Miller diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index debdd8b..d84930b 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -24,6 +24,13 @@ int sysctl_tcp_thin_linear_timeouts __read_mostly; +/** + * tcp_write_err() - close socket and save error info + * @sk: The socket the error has appeared on. + * + * Returns: Nothing (void) + */ + static void tcp_write_err(struct sock *sk) { sk->sk_err = sk->sk_err_soft ? : ETIMEDOUT; @@ -33,16 +40,21 @@ static void tcp_write_err(struct sock *sk) __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONTIMEOUT); } -/* Do not allow orphaned sockets to eat all our resources. - * This is direct violation of TCP specs, but it is required - * to prevent DoS attacks. It is called when a retransmission timeout - * or zero probe timeout occurs on orphaned socket. +/** + * tcp_out_of_resources() - Close socket if out of resources + * @sk: pointer to current socket + * @do_reset: send a last packet with reset flag * - * Criteria is still not confirmed experimentally and may change. - * We kill the socket, if: - * 1. If number of orphaned sockets exceeds an administratively configured - * limit. - * 2. If we have strong memory pressure. + * Do not allow orphaned sockets to eat all our resources. + * This is direct violation of TCP specs, but it is required + * to prevent DoS attacks. It is called when a retransmission timeout + * or zero probe timeout occurs on orphaned socket. + * + * Criteria is still not confirmed experimentally and may change. + * We kill the socket, if: + * 1. If number of orphaned sockets exceeds an administratively configured + * limit. + * 2. If we have strong memory pressure. */ static int tcp_out_of_resources(struct sock *sk, bool do_reset) { @@ -74,7 +86,11 @@ static int tcp_out_of_resources(struct sock *sk, bool do_reset) return 0; } -/* Calculate maximal number or retries on an orphaned socket. */ +/** + * tcp_orphan_retries() - Returns maximal number of retries on an orphaned socket + * @sk: Pointer to the current socket. + * @alive: bool, socket alive state + */ static int tcp_orphan_retries(struct sock *sk, bool alive) { int retries = sock_net(sk)->ipv4.sysctl_tcp_orphan_retries; /* May be zero. */ @@ -115,10 +131,22 @@ static void tcp_mtu_probing(struct inet_connection_sock *icsk, struct sock *sk) } } -/* This function calculates a "timeout" which is equivalent to the timeout of a - * TCP connection after "boundary" unsuccessful, exponentially backed-off + +/** + * retransmits_timed_out() - returns true if this connection has timed out + * @sk: The current socket + * @boundary: max number of retransmissions + * @timeout: A custom timeout value. + * If set to 0 the default timeout is calculated and used. + * Using TCP_RTO_MIN and the number of unsuccessful retransmits. + * @syn_set: true if the SYN Bit was set. + * + * The default "timeout" value this function can calculate and use + * is equivalent to the timeout of a TCP Connection + * after "boundary" unsuccessful, exponentially backed-off * retransmissions with an initial RTO of TCP_RTO_MIN or TCP_TIMEOUT_INIT if * syn_set flag is set. + * */ static bool retransmits_timed_out(struct sock *sk, unsigned int boundary, @@ -257,6 +285,16 @@ out: sk_mem_reclaim(sk); } + +/** + * tcp_delack_timer() - The TCP delayed ACK timeout handler + * @data: Pointer to the current socket. (gets casted to struct sock *) + * + * This function gets (indirectly) called when the kernel timer for a TCP packet + * of this socket expires. Calls tcp_delack_timer_handler() to do the actual work. + * + * Returns: Nothing (void) + */ static void tcp_delack_timer(unsigned long data) { struct sock *sk = (struct sock *)data; @@ -350,10 +388,18 @@ static void tcp_fastopen_synack_timer(struct sock *sk) TCP_TIMEOUT_INIT << req->num_timeout, TCP_RTO_MAX); } -/* - * The TCP retransmit timer. - */ +/** + * tcp_retransmit_timer() - The TCP retransmit timeout handler + * @sk: Pointer to the current socket. + * + * This function gets called when the kernel timer for a TCP packet + * of this socket expires. + * + * It handles retransmission, timer adjustment and other necesarry measures. + * + * Returns: Nothing (void) + */ void tcp_retransmit_timer(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); @@ -494,7 +540,8 @@ out_reset_timer: out:; } -/* Called with BH disabled */ +/* Called with bottom-half processing disabled. + Called by tcp_write_timer() */ void tcp_write_timer_handler(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); @@ -539,7 +586,7 @@ static void tcp_write_timer(unsigned long data) if (!sock_owned_by_user(sk)) { tcp_write_timer_handler(sk); } else { - /* deleguate our work to tcp_release_cb() */ + /* delegate our work to tcp_release_cb() */ if (!test_and_set_bit(TCP_WRITE_TIMER_DEFERRED, &tcp_sk(sk)->tsq_flags)) sock_hold(sk); } -- cgit v0.10.2 From 46c0772d85306f2edec03c8fa40a6efa6af915bc Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 14 Jul 2016 06:09:59 +0300 Subject: net: bridge: minor style adjustments in br_handle_frame_finish Trivial style changes in br_handle_frame_finish. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index a7817e6..0b6d326 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -131,11 +131,11 @@ static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br, /* note: already called with rcu_read_lock */ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb) { - const unsigned char *dest = eth_hdr(skb)->h_dest; struct net_bridge_port *p = br_port_get_rcu(skb->dev); - struct net_bridge *br; - struct net_bridge_fdb_entry *dst; + const unsigned char *dest = eth_hdr(skb)->h_dest; + struct net_bridge_fdb_entry *dst = NULL; struct net_bridge_mdb_entry *mdst; + struct net_bridge *br; struct sk_buff *skb2; bool unicast = true; u16 vid = 0; @@ -166,8 +166,6 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb if (br->dev->flags & IFF_PROMISC) skb2 = skb; - dst = NULL; - if (IS_ENABLED(CONFIG_INET) && skb->protocol == htons(ETH_P_ARP)) br_do_proxy_arp(skb, br, vid, p); @@ -185,13 +183,12 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb skb = NULL; if (!skb2) goto out; - } else + } else { skb2 = skb; - + } unicast = false; br->dev->stats.multicast++; - } else if ((dst = __br_fdb_get(br, dest, vid)) && - dst->is_local) { + } else if ((dst = __br_fdb_get(br, dest, vid)) && dst->is_local) { skb2 = skb; /* Do not forward the packet since it's local. */ skb = NULL; @@ -201,8 +198,9 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb if (dst) { dst->used = jiffies; br_forward(dst->dst, skb, skb2); - } else + } else { br_flood_forward(br, skb, skb2, unicast); + } } if (skb2) -- cgit v0.10.2 From e151aab9b5b3fae96b0fcd6cbe3a7f952d6cb8f8 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 14 Jul 2016 06:10:00 +0300 Subject: net: bridge: rearrange flood vs unicast receive paths This patch removes one conditional from the unicast path by using the fact that skb is NULL only when the packet is multicast or is local. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 0b6d326..c20c5be 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -134,10 +134,10 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb struct net_bridge_port *p = br_port_get_rcu(skb->dev); const unsigned char *dest = eth_hdr(skb)->h_dest; struct net_bridge_fdb_entry *dst = NULL; + bool mcast_hit = false, unicast = true; struct net_bridge_mdb_entry *mdst; struct net_bridge *br; struct sk_buff *skb2; - bool unicast = true; u16 vid = 0; if (!p || p->state == BR_STATE_DISABLED) @@ -177,30 +177,29 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb if ((mdst || BR_INPUT_SKB_CB_MROUTERS_ONLY(skb)) && br_multicast_querier_exists(br, eth_hdr(skb))) { if ((mdst && mdst->mglist) || - br_multicast_is_router(br)) + br_multicast_is_router(br)) { skb2 = skb; - br_multicast_forward(mdst, skb, skb2); - skb = NULL; - if (!skb2) - goto out; + br->dev->stats.multicast++; + } + mcast_hit = true; } else { skb2 = skb; + br->dev->stats.multicast++; } unicast = false; - br->dev->stats.multicast++; } else if ((dst = __br_fdb_get(br, dest, vid)) && dst->is_local) { - skb2 = skb; /* Do not forward the packet since it's local. */ - skb = NULL; + return br_pass_frame_up(skb); } - if (skb) { - if (dst) { - dst->used = jiffies; - br_forward(dst->dst, skb, skb2); - } else { + if (dst) { + dst->used = jiffies; + br_forward(dst->dst, skb, skb2); + } else { + if (!mcast_hit) br_flood_forward(br, skb, skb2, unicast); - } + else + br_multicast_forward(mdst, skb, skb2); } if (skb2) -- cgit v0.10.2 From b35c5f632b630183396a2ea2e2247ff8bbf2c94f Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 14 Jul 2016 06:10:01 +0300 Subject: net: bridge: drop skb2/skb0 variables and use a local_rcv boolean Currently if the packet is going to be received locally we set skb0 or sometimes called skb2 variables to the original skb. This can get confusing and also we can avoid one conditional on the fast path by simply using a boolean and passing it around. Thanks to Roopa for the name suggestion. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index d610644..204f993 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -138,17 +138,18 @@ void br_deliver(const struct net_bridge_port *to, struct sk_buff *skb) EXPORT_SYMBOL_GPL(br_deliver); /* called with rcu_read_lock */ -void br_forward(const struct net_bridge_port *to, struct sk_buff *skb, struct sk_buff *skb0) +void br_forward(const struct net_bridge_port *to, struct sk_buff *skb, + bool local_rcv) { if (to && should_deliver(to, skb)) { - if (skb0) + if (local_rcv) deliver_clone(to, skb, __br_forward); else __br_forward(to, skb); return; } - if (!skb0) + if (!local_rcv) kfree_skb(skb); } @@ -193,10 +194,9 @@ out: /* called under bridge lock */ static void br_flood(struct net_bridge *br, struct sk_buff *skb, - struct sk_buff *skb0, void (*__packet_hook)(const struct net_bridge_port *p, struct sk_buff *skb), - bool unicast) + bool local_rcv, bool unicast) { u8 igmp_type = br_multicast_igmp_type(skb); struct net_bridge_port *prev; @@ -227,14 +227,14 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, if (!prev) goto out; - if (skb0) + if (local_rcv) deliver_clone(prev, skb, __packet_hook); else __packet_hook(prev, skb); return; out: - if (!skb0) + if (!local_rcv) kfree_skb(skb); } @@ -242,23 +242,24 @@ out: /* called with rcu_read_lock */ void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb, bool unicast) { - br_flood(br, skb, NULL, __br_deliver, unicast); + br_flood(br, skb, __br_deliver, false, unicast); } /* called under bridge lock */ void br_flood_forward(struct net_bridge *br, struct sk_buff *skb, - struct sk_buff *skb2, bool unicast) + bool local_rcv, bool unicast) { - br_flood(br, skb, skb2, __br_forward, unicast); + br_flood(br, skb, __br_forward, local_rcv, unicast); } #ifdef CONFIG_BRIDGE_IGMP_SNOOPING /* called with rcu_read_lock */ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, - struct sk_buff *skb, struct sk_buff *skb0, + struct sk_buff *skb, void (*__packet_hook)( const struct net_bridge_port *p, - struct sk_buff *skb)) + struct sk_buff *skb), + bool local_rcv) { struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev; u8 igmp_type = br_multicast_igmp_type(skb); @@ -295,14 +296,14 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, if (!prev) goto out; - if (skb0) + if (local_rcv) deliver_clone(prev, skb, __packet_hook); else __packet_hook(prev, skb); return; out: - if (!skb0) + if (!local_rcv) kfree_skb(skb); } @@ -310,13 +311,13 @@ out: void br_multicast_deliver(struct net_bridge_mdb_entry *mdst, struct sk_buff *skb) { - br_multicast_flood(mdst, skb, NULL, __br_deliver); + br_multicast_flood(mdst, skb, __br_deliver, false); } /* called with rcu_read_lock */ void br_multicast_forward(struct net_bridge_mdb_entry *mdst, - struct sk_buff *skb, struct sk_buff *skb2) + struct sk_buff *skb, bool local_rcv) { - br_multicast_flood(mdst, skb, skb2, __br_forward); + br_multicast_flood(mdst, skb, __br_forward, local_rcv); } #endif diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index c20c5be..dd8885d 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -131,13 +131,12 @@ static void br_do_proxy_arp(struct sk_buff *skb, struct net_bridge *br, /* note: already called with rcu_read_lock */ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb) { + bool local_rcv = false, mcast_hit = false, unicast = true; struct net_bridge_port *p = br_port_get_rcu(skb->dev); const unsigned char *dest = eth_hdr(skb)->h_dest; struct net_bridge_fdb_entry *dst = NULL; - bool mcast_hit = false, unicast = true; struct net_bridge_mdb_entry *mdst; struct net_bridge *br; - struct sk_buff *skb2; u16 vid = 0; if (!p || p->state == BR_STATE_DISABLED) @@ -160,17 +159,13 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb BR_INPUT_SKB_CB(skb)->brdev = br->dev; - /* The packet skb2 goes to the local host (NULL to skip). */ - skb2 = NULL; - - if (br->dev->flags & IFF_PROMISC) - skb2 = skb; + local_rcv = !!(br->dev->flags & IFF_PROMISC); if (IS_ENABLED(CONFIG_INET) && skb->protocol == htons(ETH_P_ARP)) br_do_proxy_arp(skb, br, vid, p); if (is_broadcast_ether_addr(dest)) { - skb2 = skb; + local_rcv = true; unicast = false; } else if (is_multicast_ether_addr(dest)) { mdst = br_mdb_get(br, skb, vid); @@ -178,12 +173,12 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb br_multicast_querier_exists(br, eth_hdr(skb))) { if ((mdst && mdst->mglist) || br_multicast_is_router(br)) { - skb2 = skb; + local_rcv = true; br->dev->stats.multicast++; } mcast_hit = true; } else { - skb2 = skb; + local_rcv = true; br->dev->stats.multicast++; } unicast = false; @@ -194,16 +189,16 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb if (dst) { dst->used = jiffies; - br_forward(dst->dst, skb, skb2); + br_forward(dst->dst, skb, local_rcv); } else { if (!mcast_hit) - br_flood_forward(br, skb, skb2, unicast); + br_flood_forward(br, skb, local_rcv, unicast); else - br_multicast_forward(mdst, skb, skb2); + br_multicast_forward(mdst, skb, local_rcv); } - if (skb2) - return br_pass_frame_up(skb2); + if (local_rcv) + return br_pass_frame_up(skb); out: return 0; diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 40f2009..4d6cdf4 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -507,12 +507,12 @@ int br_fdb_external_learn_del(struct net_bridge *br, struct net_bridge_port *p, /* br_forward.c */ void br_deliver(const struct net_bridge_port *to, struct sk_buff *skb); int br_dev_queue_push_xmit(struct net *net, struct sock *sk, struct sk_buff *skb); -void br_forward(const struct net_bridge_port *to, - struct sk_buff *skb, struct sk_buff *skb0); +void br_forward(const struct net_bridge_port *to, struct sk_buff *skb, + bool local_rcv); int br_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb); void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb, bool unicast); void br_flood_forward(struct net_bridge *br, struct sk_buff *skb, - struct sk_buff *skb2, bool unicast); + bool local_rcv, bool unicast); /* br_if.c */ void br_port_carrier_check(struct net_bridge_port *p); @@ -563,7 +563,7 @@ void br_multicast_dev_del(struct net_bridge *br); void br_multicast_deliver(struct net_bridge_mdb_entry *mdst, struct sk_buff *skb); void br_multicast_forward(struct net_bridge_mdb_entry *mdst, - struct sk_buff *skb, struct sk_buff *skb2); + struct sk_buff *skb, bool local_rcv); int br_multicast_set_router(struct net_bridge *br, unsigned long val); int br_multicast_set_port_router(struct net_bridge_port *p, unsigned long val); int br_multicast_toggle(struct net_bridge *br, unsigned long val); @@ -698,7 +698,7 @@ static inline void br_multicast_deliver(struct net_bridge_mdb_entry *mdst, static inline void br_multicast_forward(struct net_bridge_mdb_entry *mdst, struct sk_buff *skb, - struct sk_buff *skb2) + bool local_rcv) { } static inline bool br_multicast_is_router(struct net_bridge *br) -- cgit v0.10.2 From 37b090e6be2dc98ccb55bb663931546282abf2e8 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 14 Jul 2016 06:10:02 +0300 Subject: net: bridge: remove _deliver functions and consolidate forward code Before this patch we had two flavors of most forwarding functions - _forward and _deliver, the difference being that the latter are used when the packets are locally originated. Instead of all this function pointer passing and code duplication, we can just pass a boolean noting that the packet was locally originated and use that to perform the necessary checks in __br_forward. This gives a minor performance improvement but more importantly consolidates the forwarding paths. Also add a kernel doc comment to explain the exported br_forward()'s arguments. Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 8eecd0e..09f2694 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -61,11 +61,11 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) if (!br_allowed_ingress(br, br_vlan_group_rcu(br), skb, &vid)) goto out; - if (is_broadcast_ether_addr(dest)) - br_flood_deliver(br, skb, false); - else if (is_multicast_ether_addr(dest)) { + if (is_broadcast_ether_addr(dest)) { + br_flood(br, skb, false, false, true); + } else if (is_multicast_ether_addr(dest)) { if (unlikely(netpoll_tx_running(dev))) { - br_flood_deliver(br, skb, false); + br_flood(br, skb, false, false, true); goto out; } if (br_multicast_rcv(br, NULL, skb, vid)) { @@ -76,14 +76,14 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) mdst = br_mdb_get(br, skb, vid); if ((mdst || BR_INPUT_SKB_CB_MROUTERS_ONLY(skb)) && br_multicast_querier_exists(br, eth_hdr(skb))) - br_multicast_deliver(mdst, skb); + br_multicast_flood(mdst, skb, false, true); else - br_flood_deliver(br, skb, false); - } else if ((dst = __br_fdb_get(br, dest, vid)) != NULL) - br_deliver(dst->dst, skb); - else - br_flood_deliver(br, skb, true); - + br_flood(br, skb, false, false, true); + } else if ((dst = __br_fdb_get(br, dest, vid)) != NULL) { + br_forward(dst->dst, skb, false, true); + } else { + br_flood(br, skb, true, false, true); + } out: rcu_read_unlock(); return NETDEV_TX_OK; diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 204f993..63a83d8 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -21,11 +21,6 @@ #include #include "br_private.h" -static int deliver_clone(const struct net_bridge_port *prev, - struct sk_buff *skb, - void (*__packet_hook)(const struct net_bridge_port *p, - struct sk_buff *skb)); - /* Don't forward packets to originating port or forwarding disabled */ static inline int should_deliver(const struct net_bridge_port *p, const struct sk_buff *skb) @@ -75,106 +70,92 @@ int br_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb) } EXPORT_SYMBOL_GPL(br_forward_finish); -static void __br_deliver(const struct net_bridge_port *to, struct sk_buff *skb) +static void __br_forward(const struct net_bridge_port *to, + struct sk_buff *skb, bool local_orig) { struct net_bridge_vlan_group *vg; + struct net_device *indev; + struct net *net; + int br_hook; vg = nbp_vlan_group_rcu(to); skb = br_handle_vlan(to->br, vg, skb); if (!skb) return; + indev = skb->dev; skb->dev = to->dev; - - if (unlikely(netpoll_tx_running(to->br->dev))) { - if (!is_skb_forwardable(skb->dev, skb)) + if (!local_orig) { + if (skb_warn_if_lro(skb)) { kfree_skb(skb); - else { - skb_push(skb, ETH_HLEN); - br_netpoll_send_skb(to, skb); + return; } - return; + br_hook = NF_BR_FORWARD; + skb_forward_csum(skb); + net = dev_net(indev); + } else { + if (unlikely(netpoll_tx_running(to->br->dev))) { + if (!is_skb_forwardable(skb->dev, skb)) { + kfree_skb(skb); + } else { + skb_push(skb, ETH_HLEN); + br_netpoll_send_skb(to, skb); + } + return; + } + br_hook = NF_BR_LOCAL_OUT; + net = dev_net(skb->dev); + indev = NULL; } - NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, - dev_net(skb->dev), NULL, skb,NULL, skb->dev, + NF_HOOK(NFPROTO_BRIDGE, br_hook, + net, NULL, skb, indev, skb->dev, br_forward_finish); } -static void __br_forward(const struct net_bridge_port *to, struct sk_buff *skb) +static int deliver_clone(const struct net_bridge_port *prev, + struct sk_buff *skb, bool local_orig) { - struct net_bridge_vlan_group *vg; - struct net_device *indev; - - if (skb_warn_if_lro(skb)) { - kfree_skb(skb); - return; - } - - vg = nbp_vlan_group_rcu(to); - skb = br_handle_vlan(to->br, vg, skb); - if (!skb) - return; - - indev = skb->dev; - skb->dev = to->dev; - skb_forward_csum(skb); - - NF_HOOK(NFPROTO_BRIDGE, NF_BR_FORWARD, - dev_net(indev), NULL, skb, indev, skb->dev, - br_forward_finish); -} + struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev; -/* called with rcu_read_lock */ -void br_deliver(const struct net_bridge_port *to, struct sk_buff *skb) -{ - if (to && should_deliver(to, skb)) { - __br_deliver(to, skb); - return; + skb = skb_clone(skb, GFP_ATOMIC); + if (!skb) { + dev->stats.tx_dropped++; + return -ENOMEM; } - kfree_skb(skb); + __br_forward(prev, skb, local_orig); + return 0; } -EXPORT_SYMBOL_GPL(br_deliver); -/* called with rcu_read_lock */ -void br_forward(const struct net_bridge_port *to, struct sk_buff *skb, - bool local_rcv) +/** + * br_forward - forward a packet to a specific port + * @to: destination port + * @skb: packet being forwarded + * @local_rcv: packet will be received locally after forwarding + * @local_orig: packet is locally originated + * + * Should be called with rcu_read_lock. + */ +void br_forward(const struct net_bridge_port *to, + struct sk_buff *skb, bool local_rcv, bool local_orig) { if (to && should_deliver(to, skb)) { if (local_rcv) - deliver_clone(to, skb, __br_forward); + deliver_clone(to, skb, local_orig); else - __br_forward(to, skb); + __br_forward(to, skb, local_orig); return; } if (!local_rcv) kfree_skb(skb); } - -static int deliver_clone(const struct net_bridge_port *prev, - struct sk_buff *skb, - void (*__packet_hook)(const struct net_bridge_port *p, - struct sk_buff *skb)) -{ - struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev; - - skb = skb_clone(skb, GFP_ATOMIC); - if (!skb) { - dev->stats.tx_dropped++; - return -ENOMEM; - } - - __packet_hook(prev, skb); - return 0; -} +EXPORT_SYMBOL_GPL(br_forward); static struct net_bridge_port *maybe_deliver( struct net_bridge_port *prev, struct net_bridge_port *p, - struct sk_buff *skb, - void (*__packet_hook)(const struct net_bridge_port *p, - struct sk_buff *skb)) + struct sk_buff *skb, bool local_orig) { int err; @@ -184,7 +165,7 @@ static struct net_bridge_port *maybe_deliver( if (!prev) goto out; - err = deliver_clone(prev, skb, __packet_hook); + err = deliver_clone(prev, skb, local_orig); if (err) return ERR_PTR(err); @@ -192,18 +173,14 @@ out: return p; } -/* called under bridge lock */ -static void br_flood(struct net_bridge *br, struct sk_buff *skb, - void (*__packet_hook)(const struct net_bridge_port *p, - struct sk_buff *skb), - bool local_rcv, bool unicast) +/* called under rcu_read_lock */ +void br_flood(struct net_bridge *br, struct sk_buff *skb, + bool unicast, bool local_rcv, bool local_orig) { u8 igmp_type = br_multicast_igmp_type(skb); - struct net_bridge_port *prev; + struct net_bridge_port *prev = NULL; struct net_bridge_port *p; - prev = NULL; - list_for_each_entry_rcu(p, &br->port_list, list) { /* Do not flood unicast traffic to ports that turn it off */ if (unicast && !(p->flags & BR_FLOOD)) @@ -216,7 +193,7 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, BR_INPUT_SKB_CB(skb)->proxyarp_replied) continue; - prev = maybe_deliver(prev, p, skb, __packet_hook); + prev = maybe_deliver(prev, p, skb, local_orig); if (IS_ERR(prev)) goto out; if (prev == p) @@ -228,9 +205,9 @@ static void br_flood(struct net_bridge *br, struct sk_buff *skb, goto out; if (local_rcv) - deliver_clone(prev, skb, __packet_hook); + deliver_clone(prev, skb, local_orig); else - __packet_hook(prev, skb); + __br_forward(prev, skb, local_orig); return; out: @@ -238,28 +215,11 @@ out: kfree_skb(skb); } - -/* called with rcu_read_lock */ -void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb, bool unicast) -{ - br_flood(br, skb, __br_deliver, false, unicast); -} - -/* called under bridge lock */ -void br_flood_forward(struct net_bridge *br, struct sk_buff *skb, - bool local_rcv, bool unicast) -{ - br_flood(br, skb, __br_forward, local_rcv, unicast); -} - #ifdef CONFIG_BRIDGE_IGMP_SNOOPING /* called with rcu_read_lock */ -static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, - struct sk_buff *skb, - void (*__packet_hook)( - const struct net_bridge_port *p, - struct sk_buff *skb), - bool local_rcv) +void br_multicast_flood(struct net_bridge_mdb_entry *mdst, + struct sk_buff *skb, + bool local_rcv, bool local_orig) { struct net_device *dev = BR_INPUT_SKB_CB(skb)->brdev; u8 igmp_type = br_multicast_igmp_type(skb); @@ -280,7 +240,7 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, port = (unsigned long)lport > (unsigned long)rport ? lport : rport; - prev = maybe_deliver(prev, port, skb, __packet_hook); + prev = maybe_deliver(prev, port, skb, local_orig); if (IS_ERR(prev)) goto out; if (prev == port) @@ -297,27 +257,13 @@ static void br_multicast_flood(struct net_bridge_mdb_entry *mdst, goto out; if (local_rcv) - deliver_clone(prev, skb, __packet_hook); + deliver_clone(prev, skb, local_orig); else - __packet_hook(prev, skb); + __br_forward(prev, skb, local_orig); return; out: if (!local_rcv) kfree_skb(skb); } - -/* called with rcu_read_lock */ -void br_multicast_deliver(struct net_bridge_mdb_entry *mdst, - struct sk_buff *skb) -{ - br_multicast_flood(mdst, skb, __br_deliver, false); -} - -/* called with rcu_read_lock */ -void br_multicast_forward(struct net_bridge_mdb_entry *mdst, - struct sk_buff *skb, bool local_rcv) -{ - br_multicast_flood(mdst, skb, __br_forward, local_rcv); -} #endif diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index dd8885d..8b08eec 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -189,12 +189,12 @@ int br_handle_frame_finish(struct net *net, struct sock *sk, struct sk_buff *skb if (dst) { dst->used = jiffies; - br_forward(dst->dst, skb, local_rcv); + br_forward(dst->dst, skb, local_rcv, false); } else { if (!mcast_hit) - br_flood_forward(br, skb, local_rcv, unicast); + br_flood(br, skb, unicast, local_rcv, false); else - br_multicast_forward(mdst, skb, local_rcv); + br_multicast_flood(mdst, skb, local_rcv, false); } if (local_rcv) diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 4d6cdf4..b308826 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -505,14 +505,12 @@ int br_fdb_external_learn_del(struct net_bridge *br, struct net_bridge_port *p, const unsigned char *addr, u16 vid); /* br_forward.c */ -void br_deliver(const struct net_bridge_port *to, struct sk_buff *skb); int br_dev_queue_push_xmit(struct net *net, struct sock *sk, struct sk_buff *skb); void br_forward(const struct net_bridge_port *to, struct sk_buff *skb, - bool local_rcv); + bool local_rcv, bool local_orig); int br_forward_finish(struct net *net, struct sock *sk, struct sk_buff *skb); -void br_flood_deliver(struct net_bridge *br, struct sk_buff *skb, bool unicast); -void br_flood_forward(struct net_bridge *br, struct sk_buff *skb, - bool local_rcv, bool unicast); +void br_flood(struct net_bridge *br, struct sk_buff *skb, + bool unicast, bool local_rcv, bool local_orig); /* br_if.c */ void br_port_carrier_check(struct net_bridge_port *p); @@ -560,10 +558,8 @@ void br_multicast_init(struct net_bridge *br); void br_multicast_open(struct net_bridge *br); void br_multicast_stop(struct net_bridge *br); void br_multicast_dev_del(struct net_bridge *br); -void br_multicast_deliver(struct net_bridge_mdb_entry *mdst, - struct sk_buff *skb); -void br_multicast_forward(struct net_bridge_mdb_entry *mdst, - struct sk_buff *skb, bool local_rcv); +void br_multicast_flood(struct net_bridge_mdb_entry *mdst, + struct sk_buff *skb, bool local_rcv, bool local_orig); int br_multicast_set_router(struct net_bridge *br, unsigned long val); int br_multicast_set_port_router(struct net_bridge_port *p, unsigned long val); int br_multicast_toggle(struct net_bridge *br, unsigned long val); @@ -691,28 +687,27 @@ static inline void br_multicast_dev_del(struct net_bridge *br) { } -static inline void br_multicast_deliver(struct net_bridge_mdb_entry *mdst, - struct sk_buff *skb) +static inline void br_multicast_flood(struct net_bridge_mdb_entry *mdst, + struct sk_buff *skb, + bool local_rcv, bool local_orig) { } -static inline void br_multicast_forward(struct net_bridge_mdb_entry *mdst, - struct sk_buff *skb, - bool local_rcv) -{ -} static inline bool br_multicast_is_router(struct net_bridge *br) { return 0; } + static inline bool br_multicast_querier_exists(struct net_bridge *br, struct ethhdr *eth) { return false; } + static inline void br_mdb_init(void) { } + static inline void br_mdb_uninit(void) { } diff --git a/net/bridge/netfilter/nft_reject_bridge.c b/net/bridge/netfilter/nft_reject_bridge.c index 77f7e7a..0b77ffb 100644 --- a/net/bridge/netfilter/nft_reject_bridge.c +++ b/net/bridge/netfilter/nft_reject_bridge.c @@ -72,7 +72,7 @@ static void nft_reject_br_send_v4_tcp_reset(struct net *net, nft_reject_br_push_etherhdr(oldskb, nskb); - br_deliver(br_port_get_rcu(dev), nskb); + br_forward(br_port_get_rcu(dev), nskb, false, true); } static void nft_reject_br_send_v4_unreach(struct net *net, @@ -140,7 +140,7 @@ static void nft_reject_br_send_v4_unreach(struct net *net, nft_reject_br_push_etherhdr(oldskb, nskb); - br_deliver(br_port_get_rcu(dev), nskb); + br_forward(br_port_get_rcu(dev), nskb, false, true); } static void nft_reject_br_send_v6_tcp_reset(struct net *net, @@ -174,7 +174,7 @@ static void nft_reject_br_send_v6_tcp_reset(struct net *net, nft_reject_br_push_etherhdr(oldskb, nskb); - br_deliver(br_port_get_rcu(dev), nskb); + br_forward(br_port_get_rcu(dev), nskb, false, true); } static bool reject6_br_csum_ok(struct sk_buff *skb, int hook) @@ -255,7 +255,7 @@ static void nft_reject_br_send_v6_unreach(struct net *net, nft_reject_br_push_etherhdr(oldskb, nskb); - br_deliver(br_port_get_rcu(dev), nskb); + br_forward(br_port_get_rcu(dev), nskb, false, true); } static void nft_reject_bridge_eval(const struct nft_expr *expr, -- cgit v0.10.2 From a5a18bdf7453d505783e40e47ebb84bfdd35f93b Mon Sep 17 00:00:00 2001 From: Kristian Evensen Date: Thu, 14 Jul 2016 10:23:03 +0200 Subject: rndis_host: Set valid random MAC on buggy devices MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Some devices of the same type all export the same, random MAC address. This behavior has been seen on the ZTE MF910, MF823 and MF831, and there are probably more devices out there. Fix this by generating a valid random MAC address if we read a random MAC from device. Also, changed the memcpy() to ether_addr_copy(), as pointed out by checkpatch. Suggested-by: Bjørn Mork Signed-off-by: Kristian Evensen Signed-off-by: David S. Miller diff --git a/drivers/net/usb/rndis_host.c b/drivers/net/usb/rndis_host.c index 524a47a281..4f4f71b 100644 --- a/drivers/net/usb/rndis_host.c +++ b/drivers/net/usb/rndis_host.c @@ -428,7 +428,11 @@ generic_rndis_bind(struct usbnet *dev, struct usb_interface *intf, int flags) dev_err(&intf->dev, "rndis get ethaddr, %d\n", retval); goto halt_fail_and_release; } - memcpy(net->dev_addr, bp, ETH_ALEN); + + if (bp[0] & 0x02) + eth_hw_addr_random(net); + else + ether_addr_copy(net->dev_addr, bp); /* set a nonzero filter to enable data transfers */ memset(u.set, 0, sizeof *u.set); -- cgit v0.10.2 From 43b9e127406079d187794a5140a2411fbc6df2df Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 14 Jul 2016 19:28:27 +0300 Subject: net: ipmr/ip6mr: add support for keeping an entry age In preparation for hardware offloading of ipmr/ip6mr we need an interface that allows to check (and later update) the age of entries. Relying on stats alone can show activity but not actual age of the entry, furthermore when there're tens of thousands of entries a lot of the hardware implementations only support "hit" bits which are cleared on read to denote that the entry was active and shouldn't be aged out, these can then be naturally translated into age timestamp and will be compatible with the software forwarding age. Using a lastuse entry doesn't affect performance because the members in that cache line are written to along with the age. Since all new users are encouraged to use ipmr via netlink, this is exported via the RTA_EXPIRES attribute. Also do a minor local variable declaration style adjustment - arrange them longest to shortest. Signed-off-by: Nikolay Aleksandrov CC: Roopa Prabhu CC: Shrijeet Mukherjee CC: Satish Ashok CC: Donald Sharp CC: David S. Miller CC: Alexey Kuznetsov CC: James Morris CC: Hideaki YOSHIFUJI CC: Patrick McHardy Signed-off-by: David S. Miller diff --git a/include/linux/mroute.h b/include/linux/mroute.h index bf9b322..d351fd3 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -104,6 +104,7 @@ struct mfc_cache { unsigned long bytes; unsigned long pkt; unsigned long wrong_if; + unsigned long lastuse; unsigned char ttls[MAXVIFS]; /* TTL thresholds */ } res; } mfc_un; diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h index 66982e7..3987b64 100644 --- a/include/linux/mroute6.h +++ b/include/linux/mroute6.h @@ -92,6 +92,7 @@ struct mfc6_cache { unsigned long bytes; unsigned long pkt; unsigned long wrong_if; + unsigned long lastuse; unsigned char ttls[MAXMIFS]; /* TTL thresholds */ } res; } mfc_un; diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 5ad48ec..e0d76f5 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1150,6 +1150,7 @@ static int ipmr_mfc_add(struct net *net, struct mr_table *mrt, c->mfc_origin = mfc->mfcc_origin.s_addr; c->mfc_mcastgrp = mfc->mfcc_mcastgrp.s_addr; c->mfc_parent = mfc->mfcc_parent; + c->mfc_un.res.lastuse = jiffies; ipmr_update_thresholds(mrt, c, mfc->mfcc_ttls); if (!mrtsock) c->mfc_flags |= MFC_STATIC; @@ -1792,6 +1793,7 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt, vif = cache->mfc_parent; cache->mfc_un.res.pkt++; cache->mfc_un.res.bytes += skb->len; + cache->mfc_un.res.lastuse = jiffies; if (cache->mfc_origin == htonl(INADDR_ANY) && true_vifi >= 0) { struct mfc_cache *cache_proxy; @@ -2071,10 +2073,10 @@ drop: static int __ipmr_fill_mroute(struct mr_table *mrt, struct sk_buff *skb, struct mfc_cache *c, struct rtmsg *rtm) { - int ct; - struct rtnexthop *nhp; - struct nlattr *mp_attr; struct rta_mfc_stats mfcs; + struct nlattr *mp_attr; + struct rtnexthop *nhp; + int ct; /* If cache is unresolved, don't try to parse IIF and OIF */ if (c->mfc_parent >= MAXVIFS) @@ -2106,7 +2108,10 @@ static int __ipmr_fill_mroute(struct mr_table *mrt, struct sk_buff *skb, mfcs.mfcs_packets = c->mfc_un.res.pkt; mfcs.mfcs_bytes = c->mfc_un.res.bytes; mfcs.mfcs_wrong_if = c->mfc_un.res.wrong_if; - if (nla_put_64bit(skb, RTA_MFC_STATS, sizeof(mfcs), &mfcs, RTA_PAD) < 0) + if (nla_put_64bit(skb, RTA_MFC_STATS, sizeof(mfcs), &mfcs, RTA_PAD) || + nla_put_u64_64bit(skb, RTA_EXPIRES, + jiffies_to_clock_t(c->mfc_un.res.lastuse), + RTA_PAD)) return -EMSGSIZE; rtm->rtm_type = RTN_MULTICAST; diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index c7ca0f5..7adce13 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -1500,6 +1500,7 @@ static int ip6mr_mfc_add(struct net *net, struct mr6_table *mrt, c->mf6c_origin = mfc->mf6cc_origin.sin6_addr; c->mf6c_mcastgrp = mfc->mf6cc_mcastgrp.sin6_addr; c->mf6c_parent = mfc->mf6cc_parent; + c->mfc_un.res.lastuse = jiffies; ip6mr_update_thresholds(mrt, c, ttls); if (!mrtsock) c->mfc_flags |= MFC_STATIC; @@ -2092,6 +2093,7 @@ static void ip6_mr_forward(struct net *net, struct mr6_table *mrt, vif = cache->mf6c_parent; cache->mfc_un.res.pkt++; cache->mfc_un.res.bytes += skb->len; + cache->mfc_un.res.lastuse = jiffies; if (ipv6_addr_any(&cache->mf6c_origin) && true_vifi >= 0) { struct mfc6_cache *cache_proxy; @@ -2234,10 +2236,10 @@ int ip6_mr_input(struct sk_buff *skb) static int __ip6mr_fill_mroute(struct mr6_table *mrt, struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm) { - int ct; - struct rtnexthop *nhp; - struct nlattr *mp_attr; struct rta_mfc_stats mfcs; + struct nlattr *mp_attr; + struct rtnexthop *nhp; + int ct; /* If cache is unresolved, don't try to parse IIF and OIF */ if (c->mf6c_parent >= MAXMIFS) @@ -2270,7 +2272,10 @@ static int __ip6mr_fill_mroute(struct mr6_table *mrt, struct sk_buff *skb, mfcs.mfcs_packets = c->mfc_un.res.pkt; mfcs.mfcs_bytes = c->mfc_un.res.bytes; mfcs.mfcs_wrong_if = c->mfc_un.res.wrong_if; - if (nla_put_64bit(skb, RTA_MFC_STATS, sizeof(mfcs), &mfcs, RTA_PAD) < 0) + if (nla_put_64bit(skb, RTA_MFC_STATS, sizeof(mfcs), &mfcs, RTA_PAD) || + nla_put_u64_64bit(skb, RTA_EXPIRES, + jiffies_to_clock_t(c->mfc_un.res.lastuse), + RTA_PAD)) return -EMSGSIZE; rtm->rtm_type = RTN_MULTICAST; -- cgit v0.10.2 From 18210510cfe418b374e364654743f738ddb75196 Mon Sep 17 00:00:00 2001 From: Zhao Qiang Date: Fri, 15 Jul 2016 10:38:24 +0800 Subject: wan/fsl_ucc_hdlc: remove reduplicative freed memory 'uhdlc_priv' 'uhdlc_priv' has freed twice, drop the first one. Signed-off-by: Zhao Qiang Signed-off-by: David S. Miller diff --git a/drivers/net/wan/fsl_ucc_hdlc.c b/drivers/net/wan/fsl_ucc_hdlc.c index 6edd48a..040a45c 100644 --- a/drivers/net/wan/fsl_ucc_hdlc.c +++ b/drivers/net/wan/fsl_ucc_hdlc.c @@ -1126,7 +1126,6 @@ static int ucc_hdlc_probe(struct platform_device *pdev) err_hdlc_init: err_miss_tsa_property: - kfree(uhdlc_priv); if (uhdlc_priv->tsa) kfree(utdm); err_alloc_utdm: -- cgit v0.10.2 From 1efb597d8bf56cbe9faa79e245472f1451e04d64 Mon Sep 17 00:00:00 2001 From: Zhao Qiang Date: Fri, 15 Jul 2016 10:38:25 +0800 Subject: wan/fsl_ucc_hdlc: rewrite error handling to make it clearer It was used err_xxx for labeled statement, it is not easy to understand, now use free_xxx for labeled statement. Signed-off-by: Zhao Qiang Signed-off-by: David S. Miller diff --git a/drivers/net/wan/fsl_ucc_hdlc.c b/drivers/net/wan/fsl_ucc_hdlc.c index 040a45c..b3861bf 100644 --- a/drivers/net/wan/fsl_ucc_hdlc.c +++ b/drivers/net/wan/fsl_ucc_hdlc.c @@ -143,7 +143,7 @@ static int uhdlc_init(struct ucc_hdlc_private *priv) if (!priv->rx_bd_base) { dev_err(priv->dev, "Cannot allocate MURAM memory for RxBDs\n"); ret = -ENOMEM; - goto rxbd_alloc_error; + goto free_uccf; } /* Alloc Tx BD */ @@ -154,7 +154,7 @@ static int uhdlc_init(struct ucc_hdlc_private *priv) if (!priv->tx_bd_base) { dev_err(priv->dev, "Cannot allocate MURAM memory for TxBDs\n"); ret = -ENOMEM; - goto txbd_alloc_error; + goto free_rx_bd; } /* Alloc parameter ram for ucc hdlc */ @@ -164,18 +164,18 @@ static int uhdlc_init(struct ucc_hdlc_private *priv) if (priv->ucc_pram_offset < 0) { dev_err(priv->dev, "Can not allocate MURAM for hdlc prameter.\n"); ret = -ENOMEM; - goto pram_alloc_error; + goto free_tx_bd; } priv->rx_skbuff = kzalloc(priv->rx_ring_size * sizeof(*priv->rx_skbuff), GFP_KERNEL); if (!priv->rx_skbuff) - goto rx_skb_alloc_error; + goto free_ucc_pram; priv->tx_skbuff = kzalloc(priv->tx_ring_size * sizeof(*priv->tx_skbuff), GFP_KERNEL); if (!priv->tx_skbuff) - goto tx_skb_alloc_error; + goto free_rx_skbuff; priv->skb_curtx = 0; priv->skb_dirtytx = 0; @@ -200,14 +200,14 @@ static int uhdlc_init(struct ucc_hdlc_private *priv) if (riptr < 0) { dev_err(priv->dev, "Cannot allocate MURAM mem for Receive internal temp data pointer\n"); ret = -ENOMEM; - goto riptr_alloc_error; + goto free_tx_skbuff; } tiptr = qe_muram_alloc(32, 32); if (tiptr < 0) { dev_err(priv->dev, "Cannot allocate MURAM mem for Transmit internal temp data pointer\n"); ret = -ENOMEM; - goto tiptr_alloc_error; + goto free_riptr; } /* Set RIPTR, TIPTR */ @@ -247,7 +247,7 @@ static int uhdlc_init(struct ucc_hdlc_private *priv) if (!bd_buffer) { dev_err(priv->dev, "Could not allocate buffer descriptors\n"); ret = -ENOMEM; - goto bd_alloc_error; + goto free_tiptr; } memset(bd_buffer, 0, (RX_BD_RING_LEN + TX_BD_RING_LEN) @@ -283,25 +283,25 @@ static int uhdlc_init(struct ucc_hdlc_private *priv) return 0; -bd_alloc_error: +free_tiptr: qe_muram_free(tiptr); -tiptr_alloc_error: +free_riptr: qe_muram_free(riptr); -riptr_alloc_error: +free_tx_skbuff: kfree(priv->tx_skbuff); -tx_skb_alloc_error: +free_rx_skbuff: kfree(priv->rx_skbuff); -rx_skb_alloc_error: +free_ucc_pram: qe_muram_free(priv->ucc_pram_offset); -pram_alloc_error: +free_tx_bd: dma_free_coherent(priv->dev, TX_BD_RING_LEN * sizeof(struct qe_bd), priv->tx_bd_base, priv->dma_tx_bd); -txbd_alloc_error: +free_rx_bd: dma_free_coherent(priv->dev, RX_BD_RING_LEN * sizeof(struct qe_bd), priv->rx_bd_base, priv->dma_rx_bd); -rxbd_alloc_error: +free_uccf: ucc_fast_free(priv->uccf); return ret; @@ -1067,9 +1067,7 @@ static int ucc_hdlc_probe(struct platform_device *pdev) uhdlc_priv = kzalloc(sizeof(*uhdlc_priv), GFP_KERNEL); if (!uhdlc_priv) { - ret = -ENOMEM; - dev_err(&pdev->dev, "No mem to alloc hdlc private data\n"); - goto err_alloc_priv; + return -ENOMEM; } dev_set_drvdata(&pdev->dev, uhdlc_priv); @@ -1087,25 +1085,25 @@ static int ucc_hdlc_probe(struct platform_device *pdev) if (!utdm) { ret = -ENOMEM; dev_err(&pdev->dev, "No mem to alloc ucc tdm data\n"); - goto err_alloc_utdm; + goto free_uhdlc_priv; } uhdlc_priv->utdm = utdm; ret = ucc_of_parse_tdm(np, utdm, ut_info); if (ret) - goto err_miss_tsa_property; + goto free_utdm; } ret = uhdlc_init(uhdlc_priv); if (ret) { dev_err(&pdev->dev, "Failed to init uhdlc\n"); - goto err_hdlc_init; + goto free_utdm; } dev = alloc_hdlcdev(uhdlc_priv); if (!dev) { ret = -ENOMEM; pr_err("ucc_hdlc: unable to allocate memory\n"); - goto err_hdlc_init; + goto undo_uhdlc_init; } uhdlc_priv->ndev = dev; @@ -1119,18 +1117,19 @@ static int ucc_hdlc_probe(struct platform_device *pdev) ret = -ENOBUFS; pr_err("ucc_hdlc: unable to register hdlc device\n"); free_netdev(dev); - goto err_hdlc_init; + goto free_dev; } return 0; -err_hdlc_init: -err_miss_tsa_property: +free_dev: + free_netdev(dev); +undo_uhdlc_init: +free_utdm: if (uhdlc_priv->tsa) kfree(utdm); -err_alloc_utdm: +free_uhdlc_priv: kfree(uhdlc_priv); -err_alloc_priv: return ret; } -- cgit v0.10.2 From 8d8836d4dc74ae13acfa604f0355440eaaf23ee0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Fri, 15 Jul 2016 10:12:15 +0200 Subject: net: cpsw: make TI_CPSW_PHY_SEL invisible MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit TI_CPSW_PHY_SEL depended on TI_CPSW and was selected by the latter. So there is no reason to have this symbol visible. A further optimisation would be to put the code for both symbols into a single module which would allow to not export at least cpsw_phy_sel() and simplify the module load process. Signed-off-by: Uwe Kleine-König Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/Kconfig b/drivers/net/ethernet/ti/Kconfig index e7f0b7d..9904d74 100644 --- a/drivers/net/ethernet/ti/Kconfig +++ b/drivers/net/ethernet/ti/Kconfig @@ -48,8 +48,7 @@ config TI_DAVINCI_CPDMA will be called davinci_cpdma. This is recommended. config TI_CPSW_PHY_SEL - bool "TI CPSW Switch Phy sel Support" - depends on TI_CPSW + bool ---help--- This driver supports configuring of the phy mode connected to the CPSW. -- cgit v0.10.2 From 4960e4b1e69908ee70c716755a9415079e0d554f Mon Sep 17 00:00:00 2001 From: Dongpo Li Date: Fri, 15 Jul 2016 16:26:33 +0800 Subject: net: Add MDIO bus driver for the Hisilicon FEMAC This patch adds a separate driver for the MDIO interface of the Hisilicon Fast Ethernet MAC. Signed-off-by: Dongpo Li Reviewed-by: Jiancheng Xue Acked-by: Rob Herring Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/hisilicon-femac-mdio.txt b/Documentation/devicetree/bindings/net/hisilicon-femac-mdio.txt new file mode 100644 index 0000000..23a39a3 --- /dev/null +++ b/Documentation/devicetree/bindings/net/hisilicon-femac-mdio.txt @@ -0,0 +1,22 @@ +Hisilicon Fast Ethernet MDIO Controller interface + +Required properties: +- compatible: should be "hisilicon,hisi-femac-mdio". +- reg: address and length of the register set for the device. +- clocks: A phandle to the reference clock for this device. + +- PHY subnode: inherits from phy binding [1] +[1] Documentation/devicetree/bindings/net/phy.txt + +Example: +mdio: mdio@10091100 { + compatible = "hisilicon,hisi-femac-mdio"; + reg = <0x10091100 0x10>; + clocks = <&crg HI3516CV300_MDIO_CLK>; + #address-cells = <1>; + #size-cells = <0>; + + phy0: phy@1 { + reg = <1>; + }; +}; diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig index f968294..1d7b208 100644 --- a/drivers/net/phy/Kconfig +++ b/drivers/net/phy/Kconfig @@ -294,6 +294,13 @@ config INTEL_XWAY_PHY PEF 7061, PEF 7071 and PEF 7072 or integrated into the Intel SoCs xRX200, xRX300, xRX330, xRX350 and xRX550. +config MDIO_HISI_FEMAC + tristate "Hisilicon FEMAC MDIO bus controller" + depends on HAS_IOMEM && OF_MDIO + help + This module provides a driver for the MDIO busses found in the + Hisilicon SoC that have an Fast Ethernet MAC. + endif # PHYLIB config MICREL_KS8995MA diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile index 7158274..19e38a9 100644 --- a/drivers/net/phy/Makefile +++ b/drivers/net/phy/Makefile @@ -47,3 +47,4 @@ obj-$(CONFIG_MDIO_BCM_UNIMAC) += mdio-bcm-unimac.o obj-$(CONFIG_MICROCHIP_PHY) += microchip.o obj-$(CONFIG_MDIO_BCM_IPROC) += mdio-bcm-iproc.o obj-$(CONFIG_INTEL_XWAY_PHY) += intel-xway.o +obj-$(CONFIG_MDIO_HISI_FEMAC) += mdio-hisi-femac.o diff --git a/drivers/net/phy/mdio-hisi-femac.c b/drivers/net/phy/mdio-hisi-femac.c new file mode 100644 index 0000000..b03fedd --- /dev/null +++ b/drivers/net/phy/mdio-hisi-femac.c @@ -0,0 +1,166 @@ +/* + * Hisilicon Fast Ethernet MDIO Bus Driver + * + * Copyright (c) 2016 HiSilicon Technologies Co., Ltd. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +#include +#include +#include +#include +#include +#include +#include + +#define MDIO_RWCTRL 0x00 +#define MDIO_RO_DATA 0x04 +#define MDIO_WRITE BIT(13) +#define MDIO_RW_FINISH BIT(15) +#define BIT_PHY_ADDR_OFFSET 8 +#define BIT_WR_DATA_OFFSET 16 + +struct hisi_femac_mdio_data { + struct clk *clk; + void __iomem *membase; +}; + +static int hisi_femac_mdio_wait_ready(struct hisi_femac_mdio_data *data) +{ + u32 val; + + return readl_poll_timeout(data->membase + MDIO_RWCTRL, + val, val & MDIO_RW_FINISH, 20, 10000); +} + +static int hisi_femac_mdio_read(struct mii_bus *bus, int mii_id, int regnum) +{ + struct hisi_femac_mdio_data *data = bus->priv; + int ret; + + ret = hisi_femac_mdio_wait_ready(data); + if (ret) + return ret; + + writel((mii_id << BIT_PHY_ADDR_OFFSET) | regnum, + data->membase + MDIO_RWCTRL); + + ret = hisi_femac_mdio_wait_ready(data); + if (ret) + return ret; + + return readl(data->membase + MDIO_RO_DATA) & 0xFFFF; +} + +static int hisi_femac_mdio_write(struct mii_bus *bus, int mii_id, int regnum, + u16 value) +{ + struct hisi_femac_mdio_data *data = bus->priv; + int ret; + + ret = hisi_femac_mdio_wait_ready(data); + if (ret) + return ret; + + writel(MDIO_WRITE | (value << BIT_WR_DATA_OFFSET) | + (mii_id << BIT_PHY_ADDR_OFFSET) | regnum, + data->membase + MDIO_RWCTRL); + + return hisi_femac_mdio_wait_ready(data); +} + +static int hisi_femac_mdio_probe(struct platform_device *pdev) +{ + struct device_node *np = pdev->dev.of_node; + struct mii_bus *bus; + struct hisi_femac_mdio_data *data; + struct resource *res; + int ret; + + bus = mdiobus_alloc_size(sizeof(*data)); + if (!bus) + return -ENOMEM; + + bus->name = "hisi_femac_mii_bus"; + bus->read = &hisi_femac_mdio_read; + bus->write = &hisi_femac_mdio_write; + snprintf(bus->id, MII_BUS_ID_SIZE, "%s", pdev->name); + bus->parent = &pdev->dev; + + data = bus->priv; + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + data->membase = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(data->membase)) { + ret = PTR_ERR(data->membase); + goto err_out_free_mdiobus; + } + + data->clk = devm_clk_get(&pdev->dev, NULL); + if (IS_ERR(data->clk)) { + ret = PTR_ERR(data->clk); + goto err_out_free_mdiobus; + } + + ret = clk_prepare_enable(data->clk); + if (ret) + goto err_out_free_mdiobus; + + ret = of_mdiobus_register(bus, np); + if (ret) + goto err_out_disable_clk; + + platform_set_drvdata(pdev, bus); + + return 0; + +err_out_disable_clk: + clk_disable_unprepare(data->clk); +err_out_free_mdiobus: + mdiobus_free(bus); + return ret; +} + +static int hisi_femac_mdio_remove(struct platform_device *pdev) +{ + struct mii_bus *bus = platform_get_drvdata(pdev); + struct hisi_femac_mdio_data *data = bus->priv; + + mdiobus_unregister(bus); + clk_disable_unprepare(data->clk); + mdiobus_free(bus); + + return 0; +} + +static const struct of_device_id hisi_femac_mdio_dt_ids[] = { + { .compatible = "hisilicon,hisi-femac-mdio" }, + { } +}; +MODULE_DEVICE_TABLE(of, hisi_femac_mdio_dt_ids); + +static struct platform_driver hisi_femac_mdio_driver = { + .probe = hisi_femac_mdio_probe, + .remove = hisi_femac_mdio_remove, + .driver = { + .name = "hisi-femac-mdio", + .of_match_table = hisi_femac_mdio_dt_ids, + }, +}; + +module_platform_driver(hisi_femac_mdio_driver); + +MODULE_DESCRIPTION("Hisilicon Fast Ethernet MAC MDIO interface driver"); +MODULE_AUTHOR("Dongpo Li "); +MODULE_LICENSE("GPL v2"); -- cgit v0.10.2 From b786241253041c13d94309ca4dace301833f63d1 Mon Sep 17 00:00:00 2001 From: Dongpo Li Date: Fri, 15 Jul 2016 16:26:34 +0800 Subject: of_mdio: Abstract a general interface for phy connect Abstract a general interface "of_phy_get_and_connect" for PHY connect. User will have no bother with getting "phy-mode" and "phy-handle" any more. Suggested-by: Arnd Bergmann Signed-off-by: Dongpo Li Reviewed-by: Jiancheng Xue Signed-off-by: David S. Miller diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index e2b50bc..b470f7e 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -19,6 +19,7 @@ #include #include #include +#include #include MODULE_AUTHOR("Grant Likely "); @@ -332,6 +333,41 @@ struct phy_device *of_phy_connect(struct net_device *dev, EXPORT_SYMBOL(of_phy_connect); /** + * of_phy_get_and_connect + * - Get phy node and connect to the phy described in the device tree + * @dev: pointer to net_device claiming the phy + * @np: Pointer to device tree node for the net_device claiming the phy + * @hndlr: Link state callback for the network device + * + * If successful, returns a pointer to the phy_device with the embedded + * struct device refcount incremented by one, or NULL on failure. The + * refcount must be dropped by calling phy_disconnect() or phy_detach(). + */ +struct phy_device *of_phy_get_and_connect(struct net_device *dev, + struct device_node *np, + void (*hndlr)(struct net_device *)) +{ + phy_interface_t iface; + struct device_node *phy_np; + struct phy_device *phy; + + iface = of_get_phy_mode(np); + if (iface < 0) + return NULL; + + phy_np = of_parse_phandle(np, "phy-handle", 0); + if (!phy_np) + return NULL; + + phy = of_phy_connect(dev, phy_np, hndlr, 0, iface); + + of_node_put(phy_np); + + return phy; +} +EXPORT_SYMBOL(of_phy_get_and_connect); + +/** * of_phy_attach - Attach to a PHY without starting the state machine * @dev: pointer to net_device claiming the phy * @phy_np: Node pointer for the PHY diff --git a/include/linux/of_mdio.h b/include/linux/of_mdio.h index 4b04587..2ab2336 100644 --- a/include/linux/of_mdio.h +++ b/include/linux/of_mdio.h @@ -19,6 +19,9 @@ extern struct phy_device *of_phy_connect(struct net_device *dev, struct device_node *phy_np, void (*hndlr)(struct net_device *), u32 flags, phy_interface_t iface); +extern struct phy_device * +of_phy_get_and_connect(struct net_device *dev, struct device_node *np, + void (*hndlr)(struct net_device *)); struct phy_device *of_phy_attach(struct net_device *dev, struct device_node *phy_np, u32 flags, phy_interface_t iface); @@ -52,6 +55,13 @@ static inline struct phy_device *of_phy_connect(struct net_device *dev, return NULL; } +static inline struct phy_device * +of_phy_get_and_connect(struct net_device *dev, struct device_node *np, + void (*hndlr)(struct net_device *)) +{ + return NULL; +} + static inline struct phy_device *of_phy_attach(struct net_device *dev, struct device_node *phy_np, u32 flags, phy_interface_t iface) -- cgit v0.10.2 From 542ae60af24f02e130e62cb3b7c23163a2350056 Mon Sep 17 00:00:00 2001 From: Dongpo Li Date: Fri, 15 Jul 2016 16:26:35 +0800 Subject: net: hisilicon: Add Fast Ethernet MAC driver This patch adds the Hisilicon Fast Ethernet MAC(FEMAC) driver. The FEMAC supports max speed 100Mbps and has been used in many Hisilicon SoC. Signed-off-by: Dongpo Li Reviewed-by: Jiancheng Xue Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/hisilicon-femac.txt b/Documentation/devicetree/bindings/net/hisilicon-femac.txt new file mode 100644 index 0000000..d11af5e --- /dev/null +++ b/Documentation/devicetree/bindings/net/hisilicon-femac.txt @@ -0,0 +1,39 @@ +Hisilicon Fast Ethernet MAC controller + +Required properties: +- compatible: should contain one of the following version strings: + * "hisilicon,hisi-femac-v1" + * "hisilicon,hisi-femac-v2" + and the soc string "hisilicon,hi3516cv300-femac". +- reg: specifies base physical address(s) and size of the device registers. + The first region is the MAC core register base and size. + The second region is the global MAC control register. +- interrupts: should contain the MAC interrupt. +- clocks: A phandle to the MAC main clock. +- resets: should contain the phandle to the MAC reset signal(required) and + the PHY reset signal(optional). +- reset-names: should contain the reset signal name "mac"(required) + and "phy"(optional). +- mac-address: see ethernet.txt [1]. +- phy-mode: see ethernet.txt [1]. +- phy-handle: see ethernet.txt [1]. +- hisilicon,phy-reset-delays-us: triplet of delays if PHY reset signal given. + The 1st cell is reset pre-delay in micro seconds. + The 2nd cell is reset pulse in micro seconds. + The 3rd cell is reset post-delay in micro seconds. + +[1] Documentation/devicetree/bindings/net/ethernet.txt + +Example: + hisi_femac: ethernet@10090000 { + compatible = "hisilicon,hi3516cv300-femac","hisilicon,hisi-femac-v2"; + reg = <0x10090000 0x1000>,<0x10091300 0x200>; + interrupts = <12>; + clocks = <&crg HI3518EV200_ETH_CLK>; + resets = <&crg 0xec 0>,<&crg 0xec 3>; + reset-names = "mac","phy"; + mac-address = [00 00 00 00 00 00]; + phy-mode = "mii"; + phy-handle = <&phy0>; + hisilicon,phy-reset-delays-us = <10000 20000 20000>; + }; diff --git a/drivers/net/ethernet/hisilicon/Kconfig b/drivers/net/ethernet/hisilicon/Kconfig index 2e25662..d11287e 100644 --- a/drivers/net/ethernet/hisilicon/Kconfig +++ b/drivers/net/ethernet/hisilicon/Kconfig @@ -23,6 +23,18 @@ config HIX5HD2_GMAC help This selects the hix5hd2 mac family network device. +config HISI_FEMAC + tristate "Hisilicon Fast Ethernet MAC device support" + depends on HAS_IOMEM + select PHYLIB + select RESET_CONTROLLER + help + This selects the Hisilicon Fast Ethernet MAC device(FEMAC). + The FEMAC receives and transmits data over Ethernet + ports at 10/100 Mbps in full-duplex or half-duplex mode. + The FEMAC exchanges data with the CPU, and supports + the energy efficient Ethernet (EEE). + config HIP04_ETH tristate "HISILICON P04 Ethernet support" depends on HAS_IOMEM # For MFD_SYSCON diff --git a/drivers/net/ethernet/hisilicon/Makefile b/drivers/net/ethernet/hisilicon/Makefile index 390b71f..8661695 100644 --- a/drivers/net/ethernet/hisilicon/Makefile +++ b/drivers/net/ethernet/hisilicon/Makefile @@ -6,3 +6,4 @@ obj-$(CONFIG_HIX5HD2_GMAC) += hix5hd2_gmac.o obj-$(CONFIG_HIP04_ETH) += hip04_eth.o obj-$(CONFIG_HNS_MDIO) += hns_mdio.o obj-$(CONFIG_HNS) += hns/ +obj-$(CONFIG_HISI_FEMAC) += hisi_femac.o diff --git a/drivers/net/ethernet/hisilicon/hisi_femac.c b/drivers/net/ethernet/hisilicon/hisi_femac.c new file mode 100644 index 0000000..b5d7ad0 --- /dev/null +++ b/drivers/net/ethernet/hisilicon/hisi_femac.c @@ -0,0 +1,1007 @@ +/* + * Hisilicon Fast Ethernet MAC Driver + * + * Copyright (c) 2016 HiSilicon Technologies Co., Ltd. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* MAC control register list */ +#define MAC_PORTSEL 0x0200 +#define MAC_PORTSEL_STAT_CPU BIT(0) +#define MAC_PORTSEL_RMII BIT(1) +#define MAC_PORTSET 0x0208 +#define MAC_PORTSET_DUPLEX_FULL BIT(0) +#define MAC_PORTSET_LINKED BIT(1) +#define MAC_PORTSET_SPEED_100M BIT(2) +#define MAC_SET 0x0210 +#define MAX_FRAME_SIZE 1600 +#define MAX_FRAME_SIZE_MASK GENMASK(10, 0) +#define BIT_PAUSE_EN BIT(18) +#define RX_COALESCE_SET 0x0340 +#define RX_COALESCED_FRAME_OFFSET 24 +#define RX_COALESCED_FRAMES 8 +#define RX_COALESCED_TIMER 0x74 +#define QLEN_SET 0x0344 +#define RX_DEPTH_OFFSET 8 +#define MAX_HW_FIFO_DEPTH 64 +#define HW_TX_FIFO_DEPTH 12 +#define HW_RX_FIFO_DEPTH (MAX_HW_FIFO_DEPTH - HW_TX_FIFO_DEPTH) +#define IQFRM_DES 0x0354 +#define RX_FRAME_LEN_MASK GENMASK(11, 0) +#define IQ_ADDR 0x0358 +#define EQ_ADDR 0x0360 +#define EQFRM_LEN 0x0364 +#define ADDRQ_STAT 0x036C +#define TX_CNT_INUSE_MASK GENMASK(5, 0) +#define BIT_TX_READY BIT(24) +#define BIT_RX_READY BIT(25) +/* global control register list */ +#define GLB_HOSTMAC_L32 0x0000 +#define GLB_HOSTMAC_H16 0x0004 +#define GLB_SOFT_RESET 0x0008 +#define SOFT_RESET_ALL BIT(0) +#define GLB_FWCTRL 0x0010 +#define FWCTRL_VLAN_ENABLE BIT(0) +#define FWCTRL_FW2CPU_ENA BIT(5) +#define FWCTRL_FWALL2CPU BIT(7) +#define GLB_MACTCTRL 0x0014 +#define MACTCTRL_UNI2CPU BIT(1) +#define MACTCTRL_MULTI2CPU BIT(3) +#define MACTCTRL_BROAD2CPU BIT(5) +#define MACTCTRL_MACT_ENA BIT(7) +#define GLB_IRQ_STAT 0x0030 +#define GLB_IRQ_ENA 0x0034 +#define IRQ_ENA_PORT0_MASK GENMASK(7, 0) +#define IRQ_ENA_PORT0 BIT(18) +#define IRQ_ENA_ALL BIT(19) +#define GLB_IRQ_RAW 0x0038 +#define IRQ_INT_RX_RDY BIT(0) +#define IRQ_INT_TX_PER_PACKET BIT(1) +#define IRQ_INT_TX_FIFO_EMPTY BIT(6) +#define IRQ_INT_MULTI_RXRDY BIT(7) +#define DEF_INT_MASK (IRQ_INT_MULTI_RXRDY | \ + IRQ_INT_TX_PER_PACKET | \ + IRQ_INT_TX_FIFO_EMPTY) +#define GLB_MAC_L32_BASE 0x0100 +#define GLB_MAC_H16_BASE 0x0104 +#define MACFLT_HI16_MASK GENMASK(15, 0) +#define BIT_MACFLT_ENA BIT(17) +#define BIT_MACFLT_FW2CPU BIT(21) +#define GLB_MAC_H16(reg) (GLB_MAC_H16_BASE + ((reg) * 0x8)) +#define GLB_MAC_L32(reg) (GLB_MAC_L32_BASE + ((reg) * 0x8)) +#define MAX_MAC_FILTER_NUM 8 +#define MAX_UNICAST_ADDRESSES 2 +#define MAX_MULTICAST_ADDRESSES (MAX_MAC_FILTER_NUM - \ + MAX_UNICAST_ADDRESSES) +/* software tx and rx queue number, should be power of 2 */ +#define TXQ_NUM 64 +#define RXQ_NUM 128 +#define FEMAC_POLL_WEIGHT 16 + +#define PHY_RESET_DELAYS_PROPERTY "hisilicon,phy-reset-delays-us" + +enum phy_reset_delays { + PRE_DELAY, + PULSE, + POST_DELAY, + DELAYS_NUM, +}; + +struct hisi_femac_queue { + struct sk_buff **skb; + dma_addr_t *dma_phys; + int num; + unsigned int head; + unsigned int tail; +}; + +struct hisi_femac_priv { + void __iomem *port_base; + void __iomem *glb_base; + struct clk *clk; + struct reset_control *mac_rst; + struct reset_control *phy_rst; + u32 phy_reset_delays[DELAYS_NUM]; + u32 link_status; + + struct device *dev; + struct net_device *ndev; + + struct hisi_femac_queue txq; + struct hisi_femac_queue rxq; + u32 tx_fifo_used_cnt; + struct napi_struct napi; +}; + +static void hisi_femac_irq_enable(struct hisi_femac_priv *priv, int irqs) +{ + u32 val; + + val = readl(priv->glb_base + GLB_IRQ_ENA); + writel(val | irqs, priv->glb_base + GLB_IRQ_ENA); +} + +static void hisi_femac_irq_disable(struct hisi_femac_priv *priv, int irqs) +{ + u32 val; + + val = readl(priv->glb_base + GLB_IRQ_ENA); + writel(val & (~irqs), priv->glb_base + GLB_IRQ_ENA); +} + +static void hisi_femac_tx_dma_unmap(struct hisi_femac_priv *priv, + struct sk_buff *skb, unsigned int pos) +{ + dma_addr_t dma_addr; + + dma_addr = priv->txq.dma_phys[pos]; + dma_unmap_single(priv->dev, dma_addr, skb->len, DMA_TO_DEVICE); +} + +static void hisi_femac_xmit_reclaim(struct net_device *dev) +{ + struct sk_buff *skb; + struct hisi_femac_priv *priv = netdev_priv(dev); + struct hisi_femac_queue *txq = &priv->txq; + unsigned int bytes_compl = 0, pkts_compl = 0; + u32 val; + + netif_tx_lock(dev); + + val = readl(priv->port_base + ADDRQ_STAT) & TX_CNT_INUSE_MASK; + while (val < priv->tx_fifo_used_cnt) { + skb = txq->skb[txq->tail]; + if (unlikely(!skb)) { + netdev_err(dev, "xmitq_cnt_inuse=%d, tx_fifo_used=%d\n", + val, priv->tx_fifo_used_cnt); + break; + } + hisi_femac_tx_dma_unmap(priv, skb, txq->tail); + pkts_compl++; + bytes_compl += skb->len; + dev_kfree_skb_any(skb); + + priv->tx_fifo_used_cnt--; + + val = readl(priv->port_base + ADDRQ_STAT) & TX_CNT_INUSE_MASK; + txq->skb[txq->tail] = NULL; + txq->tail = (txq->tail + 1) % txq->num; + } + + netdev_completed_queue(dev, pkts_compl, bytes_compl); + + if (unlikely(netif_queue_stopped(dev)) && pkts_compl) + netif_wake_queue(dev); + + netif_tx_unlock(dev); +} + +static void hisi_femac_adjust_link(struct net_device *dev) +{ + struct hisi_femac_priv *priv = netdev_priv(dev); + struct phy_device *phy = dev->phydev; + u32 status = 0; + + if (phy->link) + status |= MAC_PORTSET_LINKED; + if (phy->duplex == DUPLEX_FULL) + status |= MAC_PORTSET_DUPLEX_FULL; + if (phy->speed == SPEED_100) + status |= MAC_PORTSET_SPEED_100M; + + if ((status != priv->link_status) && + ((status | priv->link_status) & MAC_PORTSET_LINKED)) { + writel(status, priv->port_base + MAC_PORTSET); + priv->link_status = status; + phy_print_status(phy); + } +} + +static void hisi_femac_rx_refill(struct hisi_femac_priv *priv) +{ + struct hisi_femac_queue *rxq = &priv->rxq; + struct sk_buff *skb; + u32 pos; + u32 len = MAX_FRAME_SIZE; + dma_addr_t addr; + + pos = rxq->head; + while (readl(priv->port_base + ADDRQ_STAT) & BIT_RX_READY) { + if (!CIRC_SPACE(pos, rxq->tail, rxq->num)) + break; + if (unlikely(rxq->skb[pos])) { + netdev_err(priv->ndev, "err skb[%d]=%p\n", + pos, rxq->skb[pos]); + break; + } + skb = netdev_alloc_skb_ip_align(priv->ndev, len); + if (unlikely(!skb)) + break; + + addr = dma_map_single(priv->dev, skb->data, len, + DMA_FROM_DEVICE); + if (dma_mapping_error(priv->dev, addr)) { + dev_kfree_skb_any(skb); + break; + } + rxq->dma_phys[pos] = addr; + rxq->skb[pos] = skb; + writel(addr, priv->port_base + IQ_ADDR); + pos = (pos + 1) % rxq->num; + } + rxq->head = pos; +} + +static int hisi_femac_rx(struct net_device *dev, int limit) +{ + struct hisi_femac_priv *priv = netdev_priv(dev); + struct hisi_femac_queue *rxq = &priv->rxq; + struct sk_buff *skb; + dma_addr_t addr; + u32 rx_pkt_info, pos, len, rx_pkts_num = 0; + + pos = rxq->tail; + while (readl(priv->glb_base + GLB_IRQ_RAW) & IRQ_INT_RX_RDY) { + rx_pkt_info = readl(priv->port_base + IQFRM_DES); + len = rx_pkt_info & RX_FRAME_LEN_MASK; + len -= ETH_FCS_LEN; + + /* tell hardware we will deal with this packet */ + writel(IRQ_INT_RX_RDY, priv->glb_base + GLB_IRQ_RAW); + + rx_pkts_num++; + + skb = rxq->skb[pos]; + if (unlikely(!skb)) { + netdev_err(dev, "rx skb NULL. pos=%d\n", pos); + break; + } + rxq->skb[pos] = NULL; + + addr = rxq->dma_phys[pos]; + dma_unmap_single(priv->dev, addr, MAX_FRAME_SIZE, + DMA_FROM_DEVICE); + skb_put(skb, len); + if (unlikely(skb->len > MAX_FRAME_SIZE)) { + netdev_err(dev, "rcv len err, len = %d\n", skb->len); + dev->stats.rx_errors++; + dev->stats.rx_length_errors++; + dev_kfree_skb_any(skb); + goto next; + } + + skb->protocol = eth_type_trans(skb, dev); + napi_gro_receive(&priv->napi, skb); + dev->stats.rx_packets++; + dev->stats.rx_bytes += skb->len; +next: + pos = (pos + 1) % rxq->num; + if (rx_pkts_num >= limit) + break; + } + rxq->tail = pos; + + hisi_femac_rx_refill(priv); + + return rx_pkts_num; +} + +static int hisi_femac_poll(struct napi_struct *napi, int budget) +{ + struct hisi_femac_priv *priv = container_of(napi, + struct hisi_femac_priv, napi); + struct net_device *dev = priv->ndev; + int work_done = 0, task = budget; + int ints, num; + + do { + hisi_femac_xmit_reclaim(dev); + num = hisi_femac_rx(dev, task); + work_done += num; + task -= num; + if (work_done >= budget) + break; + + ints = readl(priv->glb_base + GLB_IRQ_RAW); + writel(ints & DEF_INT_MASK, + priv->glb_base + GLB_IRQ_RAW); + } while (ints & DEF_INT_MASK); + + if (work_done < budget) { + napi_complete(napi); + hisi_femac_irq_enable(priv, DEF_INT_MASK & + (~IRQ_INT_TX_PER_PACKET)); + } + + return work_done; +} + +static irqreturn_t hisi_femac_interrupt(int irq, void *dev_id) +{ + int ints; + struct net_device *dev = (struct net_device *)dev_id; + struct hisi_femac_priv *priv = netdev_priv(dev); + + ints = readl(priv->glb_base + GLB_IRQ_RAW); + + if (likely(ints & DEF_INT_MASK)) { + writel(ints & DEF_INT_MASK, + priv->glb_base + GLB_IRQ_RAW); + hisi_femac_irq_disable(priv, DEF_INT_MASK); + napi_schedule(&priv->napi); + } + + return IRQ_HANDLED; +} + +static int hisi_femac_init_queue(struct device *dev, + struct hisi_femac_queue *queue, + unsigned int num) +{ + queue->skb = devm_kcalloc(dev, num, sizeof(struct sk_buff *), + GFP_KERNEL); + if (!queue->skb) + return -ENOMEM; + + queue->dma_phys = devm_kcalloc(dev, num, sizeof(dma_addr_t), + GFP_KERNEL); + if (!queue->dma_phys) + return -ENOMEM; + + queue->num = num; + queue->head = 0; + queue->tail = 0; + + return 0; +} + +static int hisi_femac_init_tx_and_rx_queues(struct hisi_femac_priv *priv) +{ + int ret; + + ret = hisi_femac_init_queue(priv->dev, &priv->txq, TXQ_NUM); + if (ret) + return ret; + + ret = hisi_femac_init_queue(priv->dev, &priv->rxq, RXQ_NUM); + if (ret) + return ret; + + priv->tx_fifo_used_cnt = 0; + + return 0; +} + +static void hisi_femac_free_skb_rings(struct hisi_femac_priv *priv) +{ + struct hisi_femac_queue *txq = &priv->txq; + struct hisi_femac_queue *rxq = &priv->rxq; + struct sk_buff *skb; + dma_addr_t dma_addr; + u32 pos; + + pos = rxq->tail; + while (pos != rxq->head) { + skb = rxq->skb[pos]; + if (unlikely(!skb)) { + netdev_err(priv->ndev, "NULL rx skb. pos=%d, head=%d\n", + pos, rxq->head); + continue; + } + + dma_addr = rxq->dma_phys[pos]; + dma_unmap_single(priv->dev, dma_addr, MAX_FRAME_SIZE, + DMA_FROM_DEVICE); + + dev_kfree_skb_any(skb); + rxq->skb[pos] = NULL; + pos = (pos + 1) % rxq->num; + } + rxq->tail = pos; + + pos = txq->tail; + while (pos != txq->head) { + skb = txq->skb[pos]; + if (unlikely(!skb)) { + netdev_err(priv->ndev, "NULL tx skb. pos=%d, head=%d\n", + pos, txq->head); + continue; + } + hisi_femac_tx_dma_unmap(priv, skb, pos); + dev_kfree_skb_any(skb); + txq->skb[pos] = NULL; + pos = (pos + 1) % txq->num; + } + txq->tail = pos; + priv->tx_fifo_used_cnt = 0; +} + +static int hisi_femac_set_hw_mac_addr(struct hisi_femac_priv *priv, + unsigned char *mac) +{ + u32 reg; + + reg = mac[1] | (mac[0] << 8); + writel(reg, priv->glb_base + GLB_HOSTMAC_H16); + + reg = mac[5] | (mac[4] << 8) | (mac[3] << 16) | (mac[2] << 24); + writel(reg, priv->glb_base + GLB_HOSTMAC_L32); + + return 0; +} + +static int hisi_femac_port_reset(struct hisi_femac_priv *priv) +{ + u32 val; + + val = readl(priv->glb_base + GLB_SOFT_RESET); + val |= SOFT_RESET_ALL; + writel(val, priv->glb_base + GLB_SOFT_RESET); + + usleep_range(500, 800); + + val &= ~SOFT_RESET_ALL; + writel(val, priv->glb_base + GLB_SOFT_RESET); + + return 0; +} + +static int hisi_femac_net_open(struct net_device *dev) +{ + struct hisi_femac_priv *priv = netdev_priv(dev); + + hisi_femac_port_reset(priv); + hisi_femac_set_hw_mac_addr(priv, dev->dev_addr); + hisi_femac_rx_refill(priv); + + netif_carrier_off(dev); + netdev_reset_queue(dev); + netif_start_queue(dev); + napi_enable(&priv->napi); + + priv->link_status = 0; + if (dev->phydev) + phy_start(dev->phydev); + + writel(IRQ_ENA_PORT0_MASK, priv->glb_base + GLB_IRQ_RAW); + hisi_femac_irq_enable(priv, IRQ_ENA_ALL | IRQ_ENA_PORT0 | DEF_INT_MASK); + + return 0; +} + +static int hisi_femac_net_close(struct net_device *dev) +{ + struct hisi_femac_priv *priv = netdev_priv(dev); + + hisi_femac_irq_disable(priv, IRQ_ENA_PORT0); + + if (dev->phydev) + phy_stop(dev->phydev); + + netif_stop_queue(dev); + napi_disable(&priv->napi); + + hisi_femac_free_skb_rings(priv); + + return 0; +} + +static netdev_tx_t hisi_femac_net_xmit(struct sk_buff *skb, + struct net_device *dev) +{ + struct hisi_femac_priv *priv = netdev_priv(dev); + struct hisi_femac_queue *txq = &priv->txq; + dma_addr_t addr; + u32 val; + + val = readl(priv->port_base + ADDRQ_STAT); + val &= BIT_TX_READY; + if (!val) { + hisi_femac_irq_enable(priv, IRQ_INT_TX_PER_PACKET); + dev->stats.tx_dropped++; + dev->stats.tx_fifo_errors++; + netif_stop_queue(dev); + return NETDEV_TX_BUSY; + } + + if (unlikely(!CIRC_SPACE(txq->head, txq->tail, + txq->num))) { + hisi_femac_irq_enable(priv, IRQ_INT_TX_PER_PACKET); + dev->stats.tx_dropped++; + dev->stats.tx_fifo_errors++; + netif_stop_queue(dev); + return NETDEV_TX_BUSY; + } + + addr = dma_map_single(priv->dev, skb->data, + skb->len, DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(priv->dev, addr))) { + dev_kfree_skb_any(skb); + dev->stats.tx_dropped++; + return NETDEV_TX_OK; + } + txq->dma_phys[txq->head] = addr; + + txq->skb[txq->head] = skb; + txq->head = (txq->head + 1) % txq->num; + + writel(addr, priv->port_base + EQ_ADDR); + writel(skb->len + ETH_FCS_LEN, priv->port_base + EQFRM_LEN); + + priv->tx_fifo_used_cnt++; + + dev->stats.tx_packets++; + dev->stats.tx_bytes += skb->len; + netdev_sent_queue(dev, skb->len); + + return NETDEV_TX_OK; +} + +static int hisi_femac_set_mac_address(struct net_device *dev, void *p) +{ + struct hisi_femac_priv *priv = netdev_priv(dev); + struct sockaddr *skaddr = p; + + if (!is_valid_ether_addr(skaddr->sa_data)) + return -EADDRNOTAVAIL; + + memcpy(dev->dev_addr, skaddr->sa_data, dev->addr_len); + dev->addr_assign_type &= ~NET_ADDR_RANDOM; + + hisi_femac_set_hw_mac_addr(priv, dev->dev_addr); + + return 0; +} + +static void hisi_femac_enable_hw_addr_filter(struct hisi_femac_priv *priv, + unsigned int reg_n, bool enable) +{ + u32 val; + + val = readl(priv->glb_base + GLB_MAC_H16(reg_n)); + if (enable) + val |= BIT_MACFLT_ENA; + else + val &= ~BIT_MACFLT_ENA; + writel(val, priv->glb_base + GLB_MAC_H16(reg_n)); +} + +static void hisi_femac_set_hw_addr_filter(struct hisi_femac_priv *priv, + unsigned char *addr, + unsigned int reg_n) +{ + unsigned int high, low; + u32 val; + + high = GLB_MAC_H16(reg_n); + low = GLB_MAC_L32(reg_n); + + val = (addr[2] << 24) | (addr[3] << 16) | (addr[4] << 8) | addr[5]; + writel(val, priv->glb_base + low); + + val = readl(priv->glb_base + high); + val &= ~MACFLT_HI16_MASK; + val |= ((addr[0] << 8) | addr[1]); + val |= (BIT_MACFLT_ENA | BIT_MACFLT_FW2CPU); + writel(val, priv->glb_base + high); +} + +static void hisi_femac_set_promisc_mode(struct hisi_femac_priv *priv, + bool promisc_mode) +{ + u32 val; + + val = readl(priv->glb_base + GLB_FWCTRL); + if (promisc_mode) + val |= FWCTRL_FWALL2CPU; + else + val &= ~FWCTRL_FWALL2CPU; + writel(val, priv->glb_base + GLB_FWCTRL); +} + +/* Handle multiple multicast addresses (perfect filtering)*/ +static void hisi_femac_set_mc_addr_filter(struct hisi_femac_priv *priv) +{ + struct net_device *dev = priv->ndev; + u32 val; + + val = readl(priv->glb_base + GLB_MACTCTRL); + if ((netdev_mc_count(dev) > MAX_MULTICAST_ADDRESSES) || + (dev->flags & IFF_ALLMULTI)) { + val |= MACTCTRL_MULTI2CPU; + } else { + int reg = MAX_UNICAST_ADDRESSES; + int i; + struct netdev_hw_addr *ha; + + for (i = reg; i < MAX_MAC_FILTER_NUM; i++) + hisi_femac_enable_hw_addr_filter(priv, i, false); + + netdev_for_each_mc_addr(ha, dev) { + hisi_femac_set_hw_addr_filter(priv, ha->addr, reg); + reg++; + } + val &= ~MACTCTRL_MULTI2CPU; + } + writel(val, priv->glb_base + GLB_MACTCTRL); +} + +/* Handle multiple unicast addresses (perfect filtering)*/ +static void hisi_femac_set_uc_addr_filter(struct hisi_femac_priv *priv) +{ + struct net_device *dev = priv->ndev; + u32 val; + + val = readl(priv->glb_base + GLB_MACTCTRL); + if (netdev_uc_count(dev) > MAX_UNICAST_ADDRESSES) { + val |= MACTCTRL_UNI2CPU; + } else { + int reg = 0; + int i; + struct netdev_hw_addr *ha; + + for (i = reg; i < MAX_UNICAST_ADDRESSES; i++) + hisi_femac_enable_hw_addr_filter(priv, i, false); + + netdev_for_each_uc_addr(ha, dev) { + hisi_femac_set_hw_addr_filter(priv, ha->addr, reg); + reg++; + } + val &= ~MACTCTRL_UNI2CPU; + } + writel(val, priv->glb_base + GLB_MACTCTRL); +} + +static void hisi_femac_net_set_rx_mode(struct net_device *dev) +{ + struct hisi_femac_priv *priv = netdev_priv(dev); + + if (dev->flags & IFF_PROMISC) { + hisi_femac_set_promisc_mode(priv, true); + } else { + hisi_femac_set_promisc_mode(priv, false); + hisi_femac_set_mc_addr_filter(priv); + hisi_femac_set_uc_addr_filter(priv); + } +} + +static int hisi_femac_net_ioctl(struct net_device *dev, + struct ifreq *ifreq, int cmd) +{ + if (!netif_running(dev)) + return -EINVAL; + + if (!dev->phydev) + return -EINVAL; + + return phy_mii_ioctl(dev->phydev, ifreq, cmd); +} + +static struct ethtool_ops hisi_femac_ethtools_ops = { + .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, +}; + +static const struct net_device_ops hisi_femac_netdev_ops = { + .ndo_open = hisi_femac_net_open, + .ndo_stop = hisi_femac_net_close, + .ndo_start_xmit = hisi_femac_net_xmit, + .ndo_do_ioctl = hisi_femac_net_ioctl, + .ndo_set_mac_address = hisi_femac_set_mac_address, + .ndo_set_rx_mode = hisi_femac_net_set_rx_mode, + .ndo_change_mtu = eth_change_mtu, +}; + +static void hisi_femac_core_reset(struct hisi_femac_priv *priv) +{ + reset_control_assert(priv->mac_rst); + reset_control_deassert(priv->mac_rst); +} + +static void hisi_femac_sleep_us(u32 time_us) +{ + u32 time_ms; + + if (!time_us) + return; + + time_ms = DIV_ROUND_UP(time_us, 1000); + if (time_ms < 20) + usleep_range(time_us, time_us + 500); + else + msleep(time_ms); +} + +static void hisi_femac_phy_reset(struct hisi_femac_priv *priv) +{ + /* To make sure PHY hardware reset success, + * we must keep PHY in deassert state first and + * then complete the hardware reset operation + */ + reset_control_deassert(priv->phy_rst); + hisi_femac_sleep_us(priv->phy_reset_delays[PRE_DELAY]); + + reset_control_assert(priv->phy_rst); + /* delay some time to ensure reset ok, + * this depends on PHY hardware feature + */ + hisi_femac_sleep_us(priv->phy_reset_delays[PULSE]); + reset_control_deassert(priv->phy_rst); + /* delay some time to ensure later MDIO access */ + hisi_femac_sleep_us(priv->phy_reset_delays[POST_DELAY]); +} + +static void hisi_femac_port_init(struct hisi_femac_priv *priv) +{ + u32 val; + + /* MAC gets link status info and phy mode by software config */ + val = MAC_PORTSEL_STAT_CPU; + if (priv->ndev->phydev->interface == PHY_INTERFACE_MODE_RMII) + val |= MAC_PORTSEL_RMII; + writel(val, priv->port_base + MAC_PORTSEL); + + /*clear all interrupt status */ + writel(IRQ_ENA_PORT0_MASK, priv->glb_base + GLB_IRQ_RAW); + hisi_femac_irq_disable(priv, IRQ_ENA_PORT0_MASK | IRQ_ENA_PORT0); + + val = readl(priv->glb_base + GLB_FWCTRL); + val &= ~(FWCTRL_VLAN_ENABLE | FWCTRL_FWALL2CPU); + val |= FWCTRL_FW2CPU_ENA; + writel(val, priv->glb_base + GLB_FWCTRL); + + val = readl(priv->glb_base + GLB_MACTCTRL); + val |= (MACTCTRL_BROAD2CPU | MACTCTRL_MACT_ENA); + writel(val, priv->glb_base + GLB_MACTCTRL); + + val = readl(priv->port_base + MAC_SET); + val &= ~MAX_FRAME_SIZE_MASK; + val |= MAX_FRAME_SIZE; + writel(val, priv->port_base + MAC_SET); + + val = RX_COALESCED_TIMER | + (RX_COALESCED_FRAMES << RX_COALESCED_FRAME_OFFSET); + writel(val, priv->port_base + RX_COALESCE_SET); + + val = (HW_RX_FIFO_DEPTH << RX_DEPTH_OFFSET) | HW_TX_FIFO_DEPTH; + writel(val, priv->port_base + QLEN_SET); +} + +static int hisi_femac_drv_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct device_node *node = dev->of_node; + struct resource *res; + struct net_device *ndev; + struct hisi_femac_priv *priv; + struct phy_device *phy; + const char *mac_addr; + int ret; + + ndev = alloc_etherdev(sizeof(*priv)); + if (!ndev) + return -ENOMEM; + + platform_set_drvdata(pdev, ndev); + + priv = netdev_priv(ndev); + priv->dev = dev; + priv->ndev = ndev; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + priv->port_base = devm_ioremap_resource(dev, res); + if (IS_ERR(priv->port_base)) { + ret = PTR_ERR(priv->port_base); + goto out_free_netdev; + } + + res = platform_get_resource(pdev, IORESOURCE_MEM, 1); + priv->glb_base = devm_ioremap_resource(dev, res); + if (IS_ERR(priv->glb_base)) { + ret = PTR_ERR(priv->glb_base); + goto out_free_netdev; + } + + priv->clk = devm_clk_get(&pdev->dev, NULL); + if (IS_ERR(priv->clk)) { + dev_err(dev, "failed to get clk\n"); + ret = -ENODEV; + goto out_free_netdev; + } + + ret = clk_prepare_enable(priv->clk); + if (ret) { + dev_err(dev, "failed to enable clk %d\n", ret); + goto out_free_netdev; + } + + priv->mac_rst = devm_reset_control_get(dev, "mac"); + if (IS_ERR(priv->mac_rst)) { + ret = PTR_ERR(priv->mac_rst); + goto out_disable_clk; + } + hisi_femac_core_reset(priv); + + priv->phy_rst = devm_reset_control_get(dev, "phy"); + if (IS_ERR(priv->phy_rst)) { + priv->phy_rst = NULL; + } else { + ret = of_property_read_u32_array(node, + PHY_RESET_DELAYS_PROPERTY, + priv->phy_reset_delays, + DELAYS_NUM); + if (ret) + goto out_disable_clk; + hisi_femac_phy_reset(priv); + } + + phy = of_phy_get_and_connect(ndev, node, hisi_femac_adjust_link); + if (!phy) { + dev_err(dev, "connect to PHY failed!\n"); + ret = -ENODEV; + goto out_disable_clk; + } + + phy_attached_print(phy, "phy_id=0x%.8lx, phy_mode=%s\n", + (unsigned long)phy->phy_id, + phy_modes(phy->interface)); + + mac_addr = of_get_mac_address(node); + if (mac_addr) + ether_addr_copy(ndev->dev_addr, mac_addr); + if (!is_valid_ether_addr(ndev->dev_addr)) { + eth_hw_addr_random(ndev); + dev_warn(dev, "using random MAC address %pM\n", + ndev->dev_addr); + } + + ndev->watchdog_timeo = 6 * HZ; + ndev->priv_flags |= IFF_UNICAST_FLT; + ndev->netdev_ops = &hisi_femac_netdev_ops; + ndev->ethtool_ops = &hisi_femac_ethtools_ops; + netif_napi_add(ndev, &priv->napi, hisi_femac_poll, FEMAC_POLL_WEIGHT); + SET_NETDEV_DEV(ndev, &pdev->dev); + + hisi_femac_port_init(priv); + + ret = hisi_femac_init_tx_and_rx_queues(priv); + if (ret) + goto out_disconnect_phy; + + ndev->irq = platform_get_irq(pdev, 0); + if (ndev->irq <= 0) { + dev_err(dev, "No irq resource\n"); + ret = -ENODEV; + goto out_disconnect_phy; + } + + ret = devm_request_irq(dev, ndev->irq, hisi_femac_interrupt, + IRQF_SHARED, pdev->name, ndev); + if (ret) { + dev_err(dev, "devm_request_irq %d failed!\n", ndev->irq); + goto out_disconnect_phy; + } + + ret = register_netdev(ndev); + if (ret) { + dev_err(dev, "register_netdev failed!\n"); + goto out_disconnect_phy; + } + + return ret; + +out_disconnect_phy: + netif_napi_del(&priv->napi); + phy_disconnect(phy); +out_disable_clk: + clk_disable_unprepare(priv->clk); +out_free_netdev: + free_netdev(ndev); + + return ret; +} + +static int hisi_femac_drv_remove(struct platform_device *pdev) +{ + struct net_device *ndev = platform_get_drvdata(pdev); + struct hisi_femac_priv *priv = netdev_priv(ndev); + + netif_napi_del(&priv->napi); + unregister_netdev(ndev); + + phy_disconnect(ndev->phydev); + clk_disable_unprepare(priv->clk); + free_netdev(ndev); + + return 0; +} + +#ifdef CONFIG_PM +int hisi_femac_drv_suspend(struct platform_device *pdev, + pm_message_t state) +{ + struct net_device *ndev = platform_get_drvdata(pdev); + struct hisi_femac_priv *priv = netdev_priv(ndev); + + disable_irq(ndev->irq); + if (netif_running(ndev)) { + hisi_femac_net_close(ndev); + netif_device_detach(ndev); + } + + clk_disable_unprepare(priv->clk); + + return 0; +} + +int hisi_femac_drv_resume(struct platform_device *pdev) +{ + struct net_device *ndev = platform_get_drvdata(pdev); + struct hisi_femac_priv *priv = netdev_priv(ndev); + + clk_prepare_enable(priv->clk); + if (priv->phy_rst) + hisi_femac_phy_reset(priv); + + if (netif_running(ndev)) { + hisi_femac_port_init(priv); + hisi_femac_net_open(ndev); + netif_device_attach(ndev); + } + enable_irq(ndev->irq); + + return 0; +} +#endif + +static const struct of_device_id hisi_femac_match[] = { + {.compatible = "hisilicon,hisi-femac-v1",}, + {.compatible = "hisilicon,hisi-femac-v2",}, + {.compatible = "hisilicon,hi3516cv300-femac",}, + {}, +}; + +MODULE_DEVICE_TABLE(of, hisi_femac_match); + +static struct platform_driver hisi_femac_driver = { + .driver = { + .name = "hisi-femac", + .of_match_table = hisi_femac_match, + }, + .probe = hisi_femac_drv_probe, + .remove = hisi_femac_drv_remove, +#ifdef CONFIG_PM + .suspend = hisi_femac_drv_suspend, + .resume = hisi_femac_drv_resume, +#endif +}; + +module_platform_driver(hisi_femac_driver); + +MODULE_DESCRIPTION("Hisilicon Fast Ethernet MAC driver"); +MODULE_AUTHOR("Dongpo Li "); +MODULE_LICENSE("GPL v2"); +MODULE_ALIAS("platform:hisi-femac"); -- cgit v0.10.2 From b3f2cf8feecfa89b3e94a61c0ca29ea59e36d146 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Fri, 15 Jul 2016 15:25:36 +0200 Subject: net: usb: ax88172x: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/usb/ax88172a.c b/drivers/net/usb/ax88172a.c index cf77f2d..163a2c5 100644 --- a/drivers/net/usb/ax88172a.c +++ b/drivers/net/usb/ax88172a.c @@ -149,24 +149,6 @@ static const struct net_device_ops ax88172a_netdev_ops = { .ndo_set_rx_mode = asix_set_multicast, }; -static int ax88172a_get_settings(struct net_device *net, - struct ethtool_cmd *cmd) -{ - if (!net->phydev) - return -ENODEV; - - return phy_ethtool_gset(net->phydev, cmd); -} - -static int ax88172a_set_settings(struct net_device *net, - struct ethtool_cmd *cmd) -{ - if (!net->phydev) - return -ENODEV; - - return phy_ethtool_sset(net->phydev, cmd); -} - static int ax88172a_nway_reset(struct net_device *net) { if (!net->phydev) @@ -185,9 +167,9 @@ static const struct ethtool_ops ax88172a_ethtool_ops = { .get_eeprom_len = asix_get_eeprom_len, .get_eeprom = asix_get_eeprom, .set_eeprom = asix_set_eeprom, - .get_settings = ax88172a_get_settings, - .set_settings = ax88172a_set_settings, .nway_reset = ax88172a_nway_reset, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static int ax88172a_reset_phy(struct usbnet *dev, int embd_phy) -- cgit v0.10.2 From e5b13f3444dfe4f014343e83aa7948b83ef58168 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Fri, 15 Jul 2016 16:38:19 -0300 Subject: sctp: recvmsg should be able to run even if sock is in closing state Commit d46e416c11c8 missed to update some other places which checked for the socket being TCP-style AND Established state, as Closing state has some overlapping with the previous understanding of Established. Without this fix, one of the effects is that some already queued rx messages may not be readable anymore depending on how the association teared down, and sending may also not be possible if peer initiated the shutdown. Also merge two if() blocks into one condition on sctp_sendmsg(). Cc: Xin Long Fixes: d46e416c11c8 ("sctp: sctp should change socket state when shutdown is received") Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 52fdd54..d2681cb 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -202,7 +202,7 @@ struct sctp_association *sctp_id2assoc(struct sock *sk, sctp_assoc_t id) * could be a TCP-style listening socket or a socket which * hasn't yet called connect() to establish an association. */ - if (!sctp_sstate(sk, ESTABLISHED)) + if (!sctp_sstate(sk, ESTABLISHED) && !sctp_sstate(sk, CLOSING)) return NULL; /* Get the first and the only association from the list. */ @@ -1068,7 +1068,7 @@ static int __sctp_connect(struct sock *sk, * is already connected. * It cannot be done even on a TCP-style listening socket. */ - if (sctp_sstate(sk, ESTABLISHED) || + if (sctp_sstate(sk, ESTABLISHED) || sctp_sstate(sk, CLOSING) || (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING))) { err = -EISCONN; goto out_free; @@ -1705,18 +1705,19 @@ static int sctp_sendmsg(struct sock *sk, struct msghdr *msg, size_t msg_len) if (msg_name) { /* Look for a matching association on the endpoint. */ asoc = sctp_endpoint_lookup_assoc(ep, &to, &transport); - if (!asoc) { - /* If we could not find a matching association on the - * endpoint, make sure that it is not a TCP-style - * socket that already has an association or there is - * no peeled-off association on another socket. - */ - if ((sctp_style(sk, TCP) && - sctp_sstate(sk, ESTABLISHED)) || - sctp_endpoint_is_peeled_off(ep, &to)) { - err = -EADDRNOTAVAIL; - goto out_unlock; - } + + /* If we could not find a matching association on the + * endpoint, make sure that it is not a TCP-style + * socket that already has an association or there is + * no peeled-off association on another socket. + */ + if (!asoc && + ((sctp_style(sk, TCP) && + (sctp_sstate(sk, ESTABLISHED) || + sctp_sstate(sk, CLOSING))) || + sctp_endpoint_is_peeled_off(ep, &to))) { + err = -EADDRNOTAVAIL; + goto out_unlock; } } else { asoc = sctp_id2assoc(sk, associd); @@ -2077,7 +2078,8 @@ static int sctp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, lock_sock(sk); - if (sctp_style(sk, TCP) && !sctp_sstate(sk, ESTABLISHED)) { + if (sctp_style(sk, TCP) && !sctp_sstate(sk, ESTABLISHED) && + !sctp_sstate(sk, CLOSING)) { err = -ENOTCONN; goto out; } -- cgit v0.10.2 From c5c4e45c4b79acb23f07e43dac1348e67b4ddf91 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Fri, 15 Jul 2016 16:40:02 -0300 Subject: sctp: fix GSO for IPv6 commit 90017accff61 ("sctp: Add GSO support") didn't register SCTP GSO offloading for IPv6 and yet didn't put any restrictions on generating GSO packets while in IPv6, which causes all IPv6 GSO'ed packets to be silently dropped. The fix is to properly register the offload this time. Fixes: 90017accff61 ("sctp: Add GSO support") Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/offload.c b/net/sctp/offload.c index a37887b..7e869d0 100644 --- a/net/sctp/offload.c +++ b/net/sctp/offload.c @@ -92,7 +92,28 @@ static const struct net_offload sctp_offload = { }, }; +static const struct net_offload sctp6_offload = { + .callbacks = { + .gso_segment = sctp_gso_segment, + }, +}; + int __init sctp_offload_init(void) { - return inet_add_offload(&sctp_offload, IPPROTO_SCTP); + int ret; + + ret = inet_add_offload(&sctp_offload, IPPROTO_SCTP); + if (ret) + goto out; + + ret = inet6_add_offload(&sctp6_offload, IPPROTO_SCTP); + if (ret) + goto ipv4; + + return ret; + +ipv4: + inet_del_offload(&sctp_offload, IPPROTO_SCTP); +out: + return ret; } -- cgit v0.10.2 From 858d68f10238fdd1ebdd0096f912f063e97c6766 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Sat, 16 Jul 2016 01:15:55 +0200 Subject: bpf: bpf_event_entry_gen's alloc needs to be in atomic context Should have been obvious, only called from bpf() syscall via map_update_elem() that calls bpf_fd_array_map_update_elem() under RCU read lock and thus this must also be in GFP_ATOMIC, of course. Fixes: 3b1efb196eee ("bpf, maps: flush own entries on perf map release") Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index db1a743..633a650 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -430,7 +430,7 @@ static struct bpf_event_entry *bpf_event_entry_gen(struct file *perf_file, { struct bpf_event_entry *ee; - ee = kzalloc(sizeof(*ee), GFP_KERNEL); + ee = kzalloc(sizeof(*ee), GFP_ATOMIC); if (ee) { ee->event = perf_file->private_data; ee->perf_file = perf_file; -- cgit v0.10.2 From 4360fa22ad5b48a1d1e10e31ffb383ed8c977435 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 6 Jun 2016 11:02:03 +0200 Subject: drivers: misc: ti-st: Use int instead of fuzzy char for callback status On mips and parisc: drivers/bluetooth/btwilink.c: In function 'ti_st_open': drivers/bluetooth/btwilink.c:174:21: warning: overflow in implicit constant conversion [-Woverflow] hst->reg_status = -EINPROGRESS; drivers/nfc/nfcwilink.c: In function 'nfcwilink_open': drivers/nfc/nfcwilink.c:396:31: warning: overflow in implicit constant conversion [-Woverflow] drv->st_register_cb_status = -EINPROGRESS; There are actually two issues: 1. Whether "char" is signed or unsigned depends on the architecture. As the completion callback data is used to pass a (negative) error code, it should always be signed. 2. EINPROGRESS is 150 on mips, 245 on parisc. Hence -EINPROGRESS doesn't fit in a signed 8-bit number. Change the callback status from "char" to "int" to fix these. Signed-off-by: Geert Uytterhoeven Acked-by: Mauro Carvalho Chehab Acked-by: Samuel Ortiz Signed-off-by: Marcel Holtmann diff --git a/drivers/bluetooth/btwilink.c b/drivers/bluetooth/btwilink.c index 24a652f..485281b 100644 --- a/drivers/bluetooth/btwilink.c +++ b/drivers/bluetooth/btwilink.c @@ -51,7 +51,7 @@ */ struct ti_st { struct hci_dev *hdev; - char reg_status; + int reg_status; long (*st_write) (struct sk_buff *); struct completion wait_reg_completion; }; @@ -83,7 +83,7 @@ static inline void ti_st_tx_complete(struct ti_st *hst, int pkt_type) * status.ti_st_open() function will wait for signal from this * API when st_register() function returns ST_PENDING. */ -static void st_reg_completion_cb(void *priv_data, char data) +static void st_reg_completion_cb(void *priv_data, int data) { struct ti_st *lhst = priv_data; diff --git a/drivers/media/radio/wl128x/fmdrv_common.c b/drivers/media/radio/wl128x/fmdrv_common.c index 3f9e6df..642b89c 100644 --- a/drivers/media/radio/wl128x/fmdrv_common.c +++ b/drivers/media/radio/wl128x/fmdrv_common.c @@ -1472,7 +1472,7 @@ static long fm_st_receive(void *arg, struct sk_buff *skb) * Called by ST layer to indicate protocol registration completion * status. */ -static void fm_st_reg_comp_cb(void *arg, char data) +static void fm_st_reg_comp_cb(void *arg, int data) { struct fmdev *fmdev; diff --git a/drivers/misc/ti-st/st_core.c b/drivers/misc/ti-st/st_core.c index dcdbd58..0005159 100644 --- a/drivers/misc/ti-st/st_core.c +++ b/drivers/misc/ti-st/st_core.c @@ -141,7 +141,7 @@ static void st_send_frame(unsigned char chnl_id, struct st_data_s *st_gdata) * This function is being called with spin lock held, protocol drivers are * only expected to complete their waits and do nothing more than that. */ -static void st_reg_complete(struct st_data_s *st_gdata, char err) +static void st_reg_complete(struct st_data_s *st_gdata, int err) { unsigned char i = 0; pr_info(" %s ", __func__); diff --git a/drivers/nfc/nfcwilink.c b/drivers/nfc/nfcwilink.c index f81e500..3fbd18b 100644 --- a/drivers/nfc/nfcwilink.c +++ b/drivers/nfc/nfcwilink.c @@ -94,7 +94,7 @@ struct nfcwilink { struct nci_dev *ndev; unsigned long flags; - char st_register_cb_status; + int st_register_cb_status; long (*st_write) (struct sk_buff *); struct completion completed; @@ -320,7 +320,7 @@ exit: } /* Called by ST when registration is complete */ -static void nfcwilink_register_complete(void *priv_data, char data) +static void nfcwilink_register_complete(void *priv_data, int data) { struct nfcwilink *drv = priv_data; diff --git a/include/linux/ti_wilink_st.h b/include/linux/ti_wilink_st.h index 0a0d568..f229302 100644 --- a/include/linux/ti_wilink_st.h +++ b/include/linux/ti_wilink_st.h @@ -71,7 +71,7 @@ struct st_proto_s { enum proto_type type; long (*recv) (void *, struct sk_buff *); unsigned char (*match_packet) (const unsigned char *data); - void (*reg_complete_cb) (void *, char data); + void (*reg_complete_cb) (void *, int data); long (*write) (struct sk_buff *skb); void *priv_data; -- cgit v0.10.2 From 23bc6ab0a0912146fd674a0becc758c3162baabc Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Amadeusz=20S=C5=82awi=C5=84ski?= Date: Thu, 14 Jul 2016 10:50:23 +0200 Subject: Bluetooth: Fix l2cap_sock_setsockopt() with optname BT_RCVMTU MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When we retrieve imtu value from userspace we should use 16 bit pointer cast instead of 32 as it's defined that way in headers. Fixes setsockopt calls on big-endian platforms. Signed-off-by: Amadeusz Sławiński Signed-off-by: Marcel Holtmann Cc: stable@vger.kernel.org diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index 388ee8b..1842141 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -927,7 +927,7 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname, break; } - if (get_user(opt, (u32 __user *) optval)) { + if (get_user(opt, (u16 __user *) optval)) { err = -EFAULT; break; } -- cgit v0.10.2 From bd25997d77730d0c6679883a899899a939c704b3 Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Sat, 16 Jul 2016 13:53:28 +0530 Subject: dwc_eth_qos: Remove deprecated create_singlethread_workqueue alloc_workqueue replaces deprecated create_singlethread_workqueue(). A dedicated workqueue has been used since the workitem viz lp->txtimeout_reinit is involved in reinitialization if a TX timeout occurs, which is necessary to guarantee forward progress in packet processing. As a network device can be used during memory reclaim, the workqueue needs forward progress guarantee under memory pressure. WQ_MEM_RECLAIM has been set to ensure this. Since there is only a single work item, explicit concurrency limit is unnecessary here. Signed-off-by: Bhaktipriya Shridhar Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/synopsys/dwc_eth_qos.c b/drivers/net/ethernet/synopsys/dwc_eth_qos.c index fc1ea80..9f159a7 100644 --- a/drivers/net/ethernet/synopsys/dwc_eth_qos.c +++ b/drivers/net/ethernet/synopsys/dwc_eth_qos.c @@ -2910,7 +2910,8 @@ static int dwceqos_probe(struct platform_device *pdev) (unsigned long)ndev); tasklet_disable(&lp->tx_bdreclaim_tasklet); - lp->txtimeout_handler_wq = create_singlethread_workqueue(DRIVER_NAME); + lp->txtimeout_handler_wq = alloc_workqueue(DRIVER_NAME, + WQ_MEM_RECLAIM, 0); INIT_WORK(&lp->txtimeout_reinit, dwceqos_reinit_for_txtimeout); platform_set_drvdata(pdev, ndev); -- cgit v0.10.2 From ec87485dbecd95af06ca8b681681c433679db974 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 17 Jul 2016 01:10:14 +0200 Subject: net: ethernet: adi: bfin_mac: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/adi/bfin_mac.c b/drivers/net/ethernet/adi/bfin_mac.c index 3d2245f..32af9d2 100644 --- a/drivers/net/ethernet/adi/bfin_mac.c +++ b/drivers/net/ethernet/adi/bfin_mac.c @@ -310,7 +310,7 @@ static int bfin_mdiobus_write(struct mii_bus *bus, int phy_addr, int regnum, static void bfin_mac_adjust_link(struct net_device *dev) { struct bfin_mac_local *lp = netdev_priv(dev); - struct phy_device *phydev = lp->phydev; + struct phy_device *phydev = dev->phydev; unsigned long flags; int new_state = 0; @@ -430,7 +430,6 @@ static int mii_probe(struct net_device *dev, int phy_mode) lp->old_link = 0; lp->old_speed = 0; lp->old_duplex = -1; - lp->phydev = phydev; phy_attached_print(phydev, "mdc_clk=%dHz(mdc_div=%d)@sclk=%dMHz)\n", MDC_CLK, mdc_div, sclk / 1000000); @@ -453,10 +452,8 @@ static irqreturn_t bfin_mac_wake_interrupt(int irq, void *dev_id) static int bfin_mac_ethtool_getsettings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct bfin_mac_local *lp = netdev_priv(dev); - - if (lp->phydev) - return phy_ethtool_gset(lp->phydev, cmd); + if (dev->phydev) + return phy_ethtool_gset(dev->phydev, cmd); return -EINVAL; } @@ -464,13 +461,11 @@ bfin_mac_ethtool_getsettings(struct net_device *dev, struct ethtool_cmd *cmd) static int bfin_mac_ethtool_setsettings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct bfin_mac_local *lp = netdev_priv(dev); - if (!capable(CAP_NET_ADMIN)) return -EPERM; - if (lp->phydev) - return phy_ethtool_sset(lp->phydev, cmd); + if (dev->phydev) + return phy_ethtool_sset(dev->phydev, cmd); return -EINVAL; } @@ -1427,7 +1422,7 @@ static void bfin_mac_timeout(struct net_device *dev) if (netif_queue_stopped(dev)) netif_wake_queue(dev); - bfin_mac_enable(lp->phydev); + bfin_mac_enable(dev->phydev); /* We can accept TX packets again */ netif_trans_update(dev); /* prevent tx timeout */ @@ -1491,8 +1486,6 @@ static void bfin_mac_set_multicast_list(struct net_device *dev) static int bfin_mac_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { - struct bfin_mac_local *lp = netdev_priv(netdev); - if (!netif_running(netdev)) return -EINVAL; @@ -1502,8 +1495,8 @@ static int bfin_mac_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) case SIOCGHWTSTAMP: return bfin_mac_hwtstamp_get(netdev, ifr); default: - if (lp->phydev) - return phy_mii_ioctl(lp->phydev, ifr, cmd); + if (netdev->phydev) + return phy_mii_ioctl(netdev->phydev, ifr, cmd); else return -EOPNOTSUPP; } @@ -1547,12 +1540,12 @@ static int bfin_mac_open(struct net_device *dev) if (ret) return ret; - phy_start(lp->phydev); + phy_start(dev->phydev); setup_system_regs(dev); setup_mac_addr(dev->dev_addr); bfin_mac_disable(); - ret = bfin_mac_enable(lp->phydev); + ret = bfin_mac_enable(dev->phydev); if (ret) return ret; pr_debug("hardware init finished\n"); @@ -1578,8 +1571,8 @@ static int bfin_mac_close(struct net_device *dev) napi_disable(&lp->napi); netif_carrier_off(dev); - phy_stop(lp->phydev); - phy_write(lp->phydev, MII_BMCR, BMCR_PDOWN); + phy_stop(dev->phydev); + phy_write(dev->phydev, MII_BMCR, BMCR_PDOWN); /* clear everything */ bfin_mac_shutdown(dev); diff --git a/drivers/net/ethernet/adi/bfin_mac.h b/drivers/net/ethernet/adi/bfin_mac.h index d1217db..8c3b561 100644 --- a/drivers/net/ethernet/adi/bfin_mac.h +++ b/drivers/net/ethernet/adi/bfin_mac.h @@ -92,7 +92,6 @@ struct bfin_mac_local { int old_speed; int old_duplex; - struct phy_device *phydev; struct mii_bus *mii_bus; #if defined(CONFIG_BFIN_MAC_USE_HWSTAMP) -- cgit v0.10.2 From fe7c11b2cf41e2dfee30a46fcc027c0e84d0f3cb Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 17 Jul 2016 01:10:15 +0200 Subject: net: ethernet: adi: bfin_mac: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. There was a check on CAP_NET_ADMIN in bfin_mac_ethtool_setsettings, but this check is already done in dev_ethtool, so no need to repeat it before calling the generic function. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/adi/bfin_mac.c b/drivers/net/ethernet/adi/bfin_mac.c index 32af9d2..38eaea1 100644 --- a/drivers/net/ethernet/adi/bfin_mac.c +++ b/drivers/net/ethernet/adi/bfin_mac.c @@ -449,27 +449,6 @@ static irqreturn_t bfin_mac_wake_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -static int -bfin_mac_ethtool_getsettings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - if (dev->phydev) - return phy_ethtool_gset(dev->phydev, cmd); - - return -EINVAL; -} - -static int -bfin_mac_ethtool_setsettings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - if (dev->phydev) - return phy_ethtool_sset(dev->phydev, cmd); - - return -EINVAL; -} - static void bfin_mac_ethtool_getdrvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { @@ -547,8 +526,6 @@ static int bfin_mac_ethtool_get_ts_info(struct net_device *dev, #endif static const struct ethtool_ops bfin_mac_ethtool_ops = { - .get_settings = bfin_mac_ethtool_getsettings, - .set_settings = bfin_mac_ethtool_setsettings, .get_link = ethtool_op_get_link, .get_drvinfo = bfin_mac_ethtool_getdrvinfo, .get_wol = bfin_mac_ethtool_getwol, @@ -556,6 +533,8 @@ static const struct ethtool_ops bfin_mac_ethtool_ops = { #ifdef CONFIG_BFIN_MAC_USE_HWSTAMP .get_ts_info = bfin_mac_ethtool_get_ts_info, #endif + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; /**************************************************************************/ -- cgit v0.10.2 From 7d32184513baca1b66383825d57e24be62af40d1 Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 17 Jul 2016 23:30:45 +0200 Subject: net: ethernet: marvell: pxa168_eth: use phydev from struct net_device The private structure contain a pointer to phydev, but the structure net_device already contain such pointer. So we can remove the pointer phydev in the private structure, and update the driver to use the one contained in struct net_device. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c index 54d5154..d466326 100644 --- a/drivers/net/ethernet/marvell/pxa168_eth.c +++ b/drivers/net/ethernet/marvell/pxa168_eth.c @@ -247,7 +247,6 @@ struct pxa168_eth_private { */ struct timer_list timeout; struct mii_bus *smi_bus; - struct phy_device *phy; /* clock */ struct clk *clk; @@ -644,7 +643,7 @@ static void eth_port_start(struct net_device *dev) struct pxa168_eth_private *pep = netdev_priv(dev); int tx_curr_desc, rx_curr_desc; - phy_start(pep->phy); + phy_start(dev->phydev); /* Assignment of Tx CTRP of given queue */ tx_curr_desc = pep->tx_curr_desc_q; @@ -700,7 +699,7 @@ static void eth_port_reset(struct net_device *dev) val &= ~PCR_EN; wrl(pep, PORT_CONFIG, val); - phy_stop(pep->phy); + phy_stop(dev->phydev); } /* @@ -943,7 +942,7 @@ static int set_port_config_ext(struct pxa168_eth_private *pep) static void pxa168_eth_adjust_link(struct net_device *dev) { struct pxa168_eth_private *pep = netdev_priv(dev); - struct phy_device *phy = pep->phy; + struct phy_device *phy = dev->phydev; u32 cfg, cfg_o = rdl(pep, PORT_CONFIG); u32 cfgext, cfgext_o = rdl(pep, PORT_CONFIG_EXT); @@ -973,16 +972,17 @@ static int pxa168_init_phy(struct net_device *dev) { struct pxa168_eth_private *pep = netdev_priv(dev); struct ethtool_cmd cmd; + struct phy_device *phy = NULL; int err; - if (pep->phy) + if (dev->phydev) return 0; - pep->phy = mdiobus_scan(pep->smi_bus, pep->phy_addr); - if (IS_ERR(pep->phy)) - return PTR_ERR(pep->phy); + phy = mdiobus_scan(pep->smi_bus, pep->phy_addr); + if (IS_ERR(phy)) + return PTR_ERR(phy); - err = phy_connect_direct(dev, pep->phy, pxa168_eth_adjust_link, + err = phy_connect_direct(dev, phy, pxa168_eth_adjust_link, pep->phy_intf); if (err) return err; @@ -1366,30 +1366,26 @@ static int pxa168_smi_write(struct mii_bus *bus, int phy_addr, int regnum, static int pxa168_eth_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { - struct pxa168_eth_private *pep = netdev_priv(dev); - if (pep->phy != NULL) - return phy_mii_ioctl(pep->phy, ifr, cmd); + if (dev->phydev != NULL) + return phy_mii_ioctl(dev->phydev, ifr, cmd); return -EOPNOTSUPP; } static int pxa168_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct pxa168_eth_private *pep = netdev_priv(dev); int err; - err = phy_read_status(pep->phy); + err = phy_read_status(dev->phydev); if (err == 0) - err = phy_ethtool_gset(pep->phy, cmd); + err = phy_ethtool_gset(dev->phydev, cmd); return err; } static int pxa168_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct pxa168_eth_private *pep = netdev_priv(dev); - - return phy_ethtool_sset(pep->phy, cmd); + return phy_ethtool_sset(dev->phydev, cmd); } static void pxa168_get_drvinfo(struct net_device *dev, @@ -1569,8 +1565,8 @@ static int pxa168_eth_remove(struct platform_device *pdev) pep->htpr, pep->htpr_dma); pep->htpr = NULL; } - if (pep->phy) - phy_disconnect(pep->phy); + if (dev->phydev) + phy_disconnect(dev->phydev); if (pep->clk) { clk_disable_unprepare(pep->clk); } -- cgit v0.10.2 From 2186f6eec2739ecd3944f9278e59edf0474f207c Mon Sep 17 00:00:00 2001 From: Philippe Reynes Date: Sun, 17 Jul 2016 23:30:46 +0200 Subject: net: ethernet: marvell: pxa168_eth: use phy_ethtool_{get|set}_link_ksettings There are two generics functions phy_ethtool_{get|set}_link_ksettings, so we can use them instead of defining the same code in the driver. Signed-off-by: Philippe Reynes Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c index d466326..aeeb2e7 100644 --- a/drivers/net/ethernet/marvell/pxa168_eth.c +++ b/drivers/net/ethernet/marvell/pxa168_eth.c @@ -274,8 +274,8 @@ enum hash_table_entry { HASH_ENTRY_RECEIVE_DISCARD_BIT = 2 }; -static int pxa168_get_settings(struct net_device *dev, struct ethtool_cmd *cmd); -static int pxa168_set_settings(struct net_device *dev, struct ethtool_cmd *cmd); +static int pxa168_get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings *cmd); static int pxa168_init_hw(struct pxa168_eth_private *pep); static int pxa168_init_phy(struct net_device *dev); static void eth_port_reset(struct net_device *dev); @@ -971,7 +971,7 @@ static void pxa168_eth_adjust_link(struct net_device *dev) static int pxa168_init_phy(struct net_device *dev) { struct pxa168_eth_private *pep = netdev_priv(dev); - struct ethtool_cmd cmd; + struct ethtool_link_ksettings cmd; struct phy_device *phy = NULL; int err; @@ -987,20 +987,21 @@ static int pxa168_init_phy(struct net_device *dev) if (err) return err; - err = pxa168_get_settings(dev, &cmd); + err = pxa168_get_link_ksettings(dev, &cmd); if (err) return err; - cmd.phy_address = pep->phy_addr; - cmd.speed = pep->phy_speed; - cmd.duplex = pep->phy_duplex; - cmd.advertising = PHY_BASIC_FEATURES; - cmd.autoneg = AUTONEG_ENABLE; + cmd.base.phy_address = pep->phy_addr; + cmd.base.speed = pep->phy_speed; + cmd.base.duplex = pep->phy_duplex; + ethtool_convert_legacy_u32_to_link_mode(cmd.link_modes.advertising, + PHY_BASIC_FEATURES); + cmd.base.autoneg = AUTONEG_ENABLE; - if (cmd.speed != 0) - cmd.autoneg = AUTONEG_DISABLE; + if (cmd.base.speed != 0) + cmd.base.autoneg = AUTONEG_DISABLE; - return pxa168_set_settings(dev, &cmd); + return phy_ethtool_set_link_ksettings(dev, &cmd); } static int pxa168_init_hw(struct pxa168_eth_private *pep) @@ -1372,22 +1373,18 @@ static int pxa168_eth_do_ioctl(struct net_device *dev, struct ifreq *ifr, return -EOPNOTSUPP; } -static int pxa168_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) +static int pxa168_get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings *cmd) { int err; err = phy_read_status(dev->phydev); if (err == 0) - err = phy_ethtool_gset(dev->phydev, cmd); + err = phy_ethtool_ksettings_get(dev->phydev, cmd); return err; } -static int pxa168_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - return phy_ethtool_sset(dev->phydev, cmd); -} - static void pxa168_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { @@ -1398,11 +1395,11 @@ static void pxa168_get_drvinfo(struct net_device *dev, } static const struct ethtool_ops pxa168_ethtool_ops = { - .get_settings = pxa168_get_settings, - .set_settings = pxa168_set_settings, .get_drvinfo = pxa168_get_drvinfo, .get_link = ethtool_op_get_link, .get_ts_info = ethtool_op_get_ts_info, + .get_link_ksettings = pxa168_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, }; static const struct net_device_ops pxa168_eth_netdev_ops = { -- cgit v0.10.2 From f962fe32f2f85769cd835ddcecbff8c1d34cf561 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 17 Jul 2016 19:55:15 +0200 Subject: Bluetooth: Move hci_recv_frame and hci_recv_diag prototypes The protoypes for hci_recv_frame and hci_recv_diag are in the wrong location in the header file. Move them close to all the other hci_dev related exported functions. Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 77d7fe1..84d0273 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -1022,6 +1022,8 @@ void hci_unregister_dev(struct hci_dev *hdev); int hci_suspend_dev(struct hci_dev *hdev); int hci_resume_dev(struct hci_dev *hdev); int hci_reset_dev(struct hci_dev *hdev); +int hci_recv_frame(struct hci_dev *hdev, struct sk_buff *skb); +int hci_recv_diag(struct hci_dev *hdev, struct sk_buff *skb); int hci_dev_open(__u16 dev); int hci_dev_close(__u16 dev); int hci_dev_do_close(struct hci_dev *hdev); @@ -1098,9 +1100,6 @@ int hci_remove_adv_instance(struct hci_dev *hdev, u8 instance); void hci_event_packet(struct hci_dev *hdev, struct sk_buff *skb); -int hci_recv_frame(struct hci_dev *hdev, struct sk_buff *skb); -int hci_recv_diag(struct hci_dev *hdev, struct sk_buff *skb); - void hci_init_sysfs(struct hci_dev *hdev); void hci_conn_init_sysfs(struct hci_conn *conn); void hci_conn_add_sysfs(struct hci_conn *conn); -- cgit v0.10.2 From 5177a83827cd0b8cf6ce0391b00dd4417352d2f1 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 17 Jul 2016 19:55:16 +0200 Subject: Bluetooth: Add debugfs fields for hardware and firmware info Some Bluetooth controllers allow for reading hardware and firmware related vendor specific infos. If they are available, then they can be exposed via debugfs now. Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 84d0273..ee7fc47 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -372,6 +372,8 @@ struct hci_dev { atomic_t promisc; + const char *hw_info; + const char *fw_info; struct dentry *debugfs; struct device dev; @@ -1024,6 +1026,8 @@ int hci_resume_dev(struct hci_dev *hdev); int hci_reset_dev(struct hci_dev *hdev); int hci_recv_frame(struct hci_dev *hdev, struct sk_buff *skb); int hci_recv_diag(struct hci_dev *hdev, struct sk_buff *skb); +void hci_set_hw_info(struct hci_dev *hdev, const char *fmt, ...); +void hci_set_fw_info(struct hci_dev *hdev, const char *fmt, ...); int hci_dev_open(__u16 dev); int hci_dev_close(__u16 dev); int hci_dev_do_close(struct hci_dev *hdev); diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 98f6c37..ddf8432 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -3163,6 +3163,8 @@ void hci_unregister_dev(struct hci_dev *hdev) device_del(&hdev->dev); debugfs_remove_recursive(hdev->debugfs); + kfree_const(hdev->hw_info); + kfree_const(hdev->fw_info); destroy_workqueue(hdev->workqueue); destroy_workqueue(hdev->req_workqueue); @@ -3266,6 +3268,28 @@ int hci_recv_diag(struct hci_dev *hdev, struct sk_buff *skb) } EXPORT_SYMBOL(hci_recv_diag); +void hci_set_hw_info(struct hci_dev *hdev, const char *fmt, ...) +{ + va_list vargs; + + va_start(vargs, fmt); + kfree_const(hdev->hw_info); + hdev->hw_info = kvasprintf_const(GFP_KERNEL, fmt, vargs); + va_end(vargs); +} +EXPORT_SYMBOL(hci_set_hw_info); + +void hci_set_fw_info(struct hci_dev *hdev, const char *fmt, ...) +{ + va_list vargs; + + va_start(vargs, fmt); + kfree_const(hdev->fw_info); + hdev->fw_info = kvasprintf_const(GFP_KERNEL, fmt, vargs); + va_end(vargs); +} +EXPORT_SYMBOL(hci_set_fw_info); + /* ---- Interface to upper protocols ---- */ int hci_register_cb(struct hci_cb *cb) diff --git a/net/bluetooth/hci_debugfs.c b/net/bluetooth/hci_debugfs.c index 7db4220..63df63e 100644 --- a/net/bluetooth/hci_debugfs.c +++ b/net/bluetooth/hci_debugfs.c @@ -76,6 +76,30 @@ static const struct file_operations __name ## _fops = { \ .llseek = default_llseek, \ } \ +#define DEFINE_INFO_ATTRIBUTE(__name, __field) \ +static int __name ## _show(struct seq_file *f, void *ptr) \ +{ \ + struct hci_dev *hdev = f->private; \ + \ + hci_dev_lock(hdev); \ + seq_printf(f, "%s\n", hdev->__field ? : ""); \ + hci_dev_unlock(hdev); \ + \ + return 0; \ +} \ + \ +static int __name ## _open(struct inode *inode, struct file *file) \ +{ \ + return single_open(file, __name ## _show, inode->i_private); \ +} \ + \ +static const struct file_operations __name ## _fops = { \ + .open = __name ## _open, \ + .read = seq_read, \ + .llseek = seq_lseek, \ + .release = single_release, \ +} \ + static int features_show(struct seq_file *f, void *ptr) { struct hci_dev *hdev = f->private; @@ -349,6 +373,9 @@ static const struct file_operations sc_only_mode_fops = { .llseek = default_llseek, }; +DEFINE_INFO_ATTRIBUTE(hardware_info, hw_info); +DEFINE_INFO_ATTRIBUTE(firmware_info, fw_info); + void hci_debugfs_create_common(struct hci_dev *hdev) { debugfs_create_file("features", 0444, hdev->debugfs, hdev, @@ -382,6 +409,14 @@ void hci_debugfs_create_common(struct hci_dev *hdev) if (lmp_sc_capable(hdev) || lmp_le_capable(hdev)) debugfs_create_file("sc_only_mode", 0444, hdev->debugfs, hdev, &sc_only_mode_fops); + + if (hdev->hw_info) + debugfs_create_file("hardware_info", 0444, hdev->debugfs, + hdev, &hardware_info_fops); + + if (hdev->fw_info) + debugfs_create_file("firmware_info", 0444, hdev->debugfs, + hdev, &firmware_info_fops); } static int inquiry_cache_show(struct seq_file *f, void *p) -- cgit v0.10.2 From b2999c195bb5cec1a1c713f17c7f382bbcdb8b39 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Sun, 17 Jul 2016 19:55:17 +0200 Subject: Bluetooth: bpa10x: Add support for hci_set_fw_info The Digianswer sniffer devices allow for reading a firmware specific information string. If it is available, then inform the Bluetooth core about it via hci_set_fw_info. That exposes it via debugfs like this: # cat /sys/kernel/debug/bluetooth/hci0/firmware_info SNIF_102,BB930,02/01/18,10:37:56 Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg diff --git a/drivers/bluetooth/bpa10x.c b/drivers/bluetooth/bpa10x.c index fd6b53e..a9932fe 100644 --- a/drivers/bluetooth/bpa10x.c +++ b/drivers/bluetooth/bpa10x.c @@ -274,6 +274,8 @@ static int bpa10x_setup(struct hci_dev *hdev) BT_INFO("%s: %s", hdev->name, (char *)(skb->data + 1)); + hci_set_fw_info(hdev, "%s", skb->data + 1); + kfree_skb(skb); return 0; } -- cgit v0.10.2 From e7acf43024e95faa7460e77fe7e0fec2a3de582a Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Fri, 15 Jul 2016 16:20:52 +0530 Subject: Bluetooth: btmrvl: reset is_suspending flag in failure path is_suspending flag remains on when host sleep fails to enable. Data path is unnecessarily blocked after this. This patch ensures to reset the flag in failure path. Signed-off-by: Amitkumar Karwar Signed-off-by: Marcel Holtmann diff --git a/drivers/bluetooth/btmrvl_sdio.c b/drivers/bluetooth/btmrvl_sdio.c index b7c3928..d02f2c1 100644 --- a/drivers/bluetooth/btmrvl_sdio.c +++ b/drivers/bluetooth/btmrvl_sdio.c @@ -1625,6 +1625,7 @@ static int btmrvl_sdio_suspend(struct device *dev) if (priv->adapter->hs_state != HS_ACTIVATED) { if (btmrvl_enable_hs(priv)) { BT_ERR("HS not actived, suspend failed!"); + priv->adapter->is_suspending = false; return -EBUSY; } } -- cgit v0.10.2 From 8cfb86003dbfbe7341574fef02c96b97cbd63997 Mon Sep 17 00:00:00 2001 From: Prasun Maiti Date: Mon, 27 Jun 2016 15:43:22 +0530 Subject: mwifiex: Reduce endian conversion for REG Host Commands For multiple REG Host Commands (e.g HostCmd_CMD_802_11_EEPROM_ACCESS, HostCmd_CMD_MAC_REG_ACCESS etc.) "cpu_to_leX"-converted values are saved to driver. So, "leX_to_cpu" conversion is required too many times afterwards in driver. This patch reduces the endian: conversion without saving "cpu_to_leX" converted values in driver. This will convert endianness in prepare command and command response path. Signed-off-by: Prasun Maiti Acked-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/ioctl.h b/drivers/net/wireless/marvell/mwifiex/ioctl.h index f5b8fd1..7042981 100644 --- a/drivers/net/wireless/marvell/mwifiex/ioctl.h +++ b/drivers/net/wireless/marvell/mwifiex/ioctl.h @@ -343,16 +343,16 @@ enum { }; struct mwifiex_ds_reg_rw { - __le32 type; - __le32 offset; - __le32 value; + u32 type; + u32 offset; + u32 value; }; #define MAX_EEPROM_DATA 256 struct mwifiex_ds_read_eeprom { - __le16 offset; - __le16 byte_count; + u16 offset; + u16 byte_count; u8 value[MAX_EEPROM_DATA]; }; diff --git a/drivers/net/wireless/marvell/mwifiex/sta_cmd.c b/drivers/net/wireless/marvell/mwifiex/sta_cmd.c index 8c65849..7897037 100644 --- a/drivers/net/wireless/marvell/mwifiex/sta_cmd.c +++ b/drivers/net/wireless/marvell/mwifiex/sta_cmd.c @@ -1148,9 +1148,8 @@ static int mwifiex_cmd_reg_access(struct host_cmd_ds_command *cmd, cmd->size = cpu_to_le16(sizeof(*mac_reg) + S_DS_GEN); mac_reg = &cmd->params.mac_reg; mac_reg->action = cpu_to_le16(cmd_action); - mac_reg->offset = - cpu_to_le16((u16) le32_to_cpu(reg_rw->offset)); - mac_reg->value = reg_rw->value; + mac_reg->offset = cpu_to_le16((u16) reg_rw->offset); + mac_reg->value = cpu_to_le32(reg_rw->value); break; } case HostCmd_CMD_BBP_REG_ACCESS: @@ -1160,9 +1159,8 @@ static int mwifiex_cmd_reg_access(struct host_cmd_ds_command *cmd, cmd->size = cpu_to_le16(sizeof(*bbp_reg) + S_DS_GEN); bbp_reg = &cmd->params.bbp_reg; bbp_reg->action = cpu_to_le16(cmd_action); - bbp_reg->offset = - cpu_to_le16((u16) le32_to_cpu(reg_rw->offset)); - bbp_reg->value = (u8) le32_to_cpu(reg_rw->value); + bbp_reg->offset = cpu_to_le16((u16) reg_rw->offset); + bbp_reg->value = (u8) reg_rw->value; break; } case HostCmd_CMD_RF_REG_ACCESS: @@ -1172,8 +1170,8 @@ static int mwifiex_cmd_reg_access(struct host_cmd_ds_command *cmd, cmd->size = cpu_to_le16(sizeof(*rf_reg) + S_DS_GEN); rf_reg = &cmd->params.rf_reg; rf_reg->action = cpu_to_le16(cmd_action); - rf_reg->offset = cpu_to_le16((u16) le32_to_cpu(reg_rw->offset)); - rf_reg->value = (u8) le32_to_cpu(reg_rw->value); + rf_reg->offset = cpu_to_le16((u16) reg_rw->offset); + rf_reg->value = (u8) reg_rw->value; break; } case HostCmd_CMD_PMIC_REG_ACCESS: @@ -1183,9 +1181,8 @@ static int mwifiex_cmd_reg_access(struct host_cmd_ds_command *cmd, cmd->size = cpu_to_le16(sizeof(*pmic_reg) + S_DS_GEN); pmic_reg = &cmd->params.pmic_reg; pmic_reg->action = cpu_to_le16(cmd_action); - pmic_reg->offset = - cpu_to_le16((u16) le32_to_cpu(reg_rw->offset)); - pmic_reg->value = (u8) le32_to_cpu(reg_rw->value); + pmic_reg->offset = cpu_to_le16((u16) reg_rw->offset); + pmic_reg->value = (u8) reg_rw->value; break; } case HostCmd_CMD_CAU_REG_ACCESS: @@ -1195,9 +1192,8 @@ static int mwifiex_cmd_reg_access(struct host_cmd_ds_command *cmd, cmd->size = cpu_to_le16(sizeof(*cau_reg) + S_DS_GEN); cau_reg = &cmd->params.rf_reg; cau_reg->action = cpu_to_le16(cmd_action); - cau_reg->offset = - cpu_to_le16((u16) le32_to_cpu(reg_rw->offset)); - cau_reg->value = (u8) le32_to_cpu(reg_rw->value); + cau_reg->offset = cpu_to_le16((u16) reg_rw->offset); + cau_reg->value = (u8) reg_rw->value; break; } case HostCmd_CMD_802_11_EEPROM_ACCESS: @@ -1208,8 +1204,8 @@ static int mwifiex_cmd_reg_access(struct host_cmd_ds_command *cmd, cmd->size = cpu_to_le16(sizeof(*cmd_eeprom) + S_DS_GEN); cmd_eeprom->action = cpu_to_le16(cmd_action); - cmd_eeprom->offset = rd_eeprom->offset; - cmd_eeprom->byte_count = rd_eeprom->byte_count; + cmd_eeprom->offset = cpu_to_le16(rd_eeprom->offset); + cmd_eeprom->byte_count = cpu_to_le16(rd_eeprom->byte_count); cmd_eeprom->value = 0; break; } diff --git a/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c b/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c index 9050d06..ccf54932 100644 --- a/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c +++ b/drivers/net/wireless/marvell/mwifiex/sta_cmdresp.c @@ -786,45 +786,44 @@ static int mwifiex_ret_reg_access(u16 type, struct host_cmd_ds_command *resp, switch (type) { case HostCmd_CMD_MAC_REG_ACCESS: r.mac = &resp->params.mac_reg; - reg_rw->offset = cpu_to_le32((u32) le16_to_cpu(r.mac->offset)); - reg_rw->value = r.mac->value; + reg_rw->offset = (u32) le16_to_cpu(r.mac->offset); + reg_rw->value = le32_to_cpu(r.mac->value); break; case HostCmd_CMD_BBP_REG_ACCESS: r.bbp = &resp->params.bbp_reg; - reg_rw->offset = cpu_to_le32((u32) le16_to_cpu(r.bbp->offset)); - reg_rw->value = cpu_to_le32((u32) r.bbp->value); + reg_rw->offset = (u32) le16_to_cpu(r.bbp->offset); + reg_rw->value = (u32) r.bbp->value; break; case HostCmd_CMD_RF_REG_ACCESS: r.rf = &resp->params.rf_reg; - reg_rw->offset = cpu_to_le32((u32) le16_to_cpu(r.rf->offset)); - reg_rw->value = cpu_to_le32((u32) r.bbp->value); + reg_rw->offset = (u32) le16_to_cpu(r.rf->offset); + reg_rw->value = (u32) r.bbp->value; break; case HostCmd_CMD_PMIC_REG_ACCESS: r.pmic = &resp->params.pmic_reg; - reg_rw->offset = cpu_to_le32((u32) le16_to_cpu(r.pmic->offset)); - reg_rw->value = cpu_to_le32((u32) r.pmic->value); + reg_rw->offset = (u32) le16_to_cpu(r.pmic->offset); + reg_rw->value = (u32) r.pmic->value; break; case HostCmd_CMD_CAU_REG_ACCESS: r.rf = &resp->params.rf_reg; - reg_rw->offset = cpu_to_le32((u32) le16_to_cpu(r.rf->offset)); - reg_rw->value = cpu_to_le32((u32) r.rf->value); + reg_rw->offset = (u32) le16_to_cpu(r.rf->offset); + reg_rw->value = (u32) r.rf->value; break; case HostCmd_CMD_802_11_EEPROM_ACCESS: r.eeprom = &resp->params.eeprom; - pr_debug("info: EEPROM read len=%x\n", r.eeprom->byte_count); - if (le16_to_cpu(eeprom->byte_count) < - le16_to_cpu(r.eeprom->byte_count)) { - eeprom->byte_count = cpu_to_le16(0); + pr_debug("info: EEPROM read len=%x\n", + le16_to_cpu(r.eeprom->byte_count)); + if (eeprom->byte_count < le16_to_cpu(r.eeprom->byte_count)) { + eeprom->byte_count = 0; pr_debug("info: EEPROM read length is too big\n"); return -1; } - eeprom->offset = r.eeprom->offset; - eeprom->byte_count = r.eeprom->byte_count; - if (le16_to_cpu(eeprom->byte_count) > 0) + eeprom->offset = le16_to_cpu(r.eeprom->offset); + eeprom->byte_count = le16_to_cpu(r.eeprom->byte_count); + if (eeprom->byte_count > 0) memcpy(&eeprom->value, &r.eeprom->value, - le16_to_cpu(r.eeprom->byte_count)); - + min((u16)MAX_EEPROM_DATA, eeprom->byte_count)); break; default: return -1; diff --git a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c index 2ba5397..e06647a 100644 --- a/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c +++ b/drivers/net/wireless/marvell/mwifiex/sta_ioctl.c @@ -1251,7 +1251,7 @@ static int mwifiex_reg_mem_ioctl_reg_rw(struct mwifiex_private *priv, { u16 cmd_no; - switch (le32_to_cpu(reg_rw->type)) { + switch (reg_rw->type) { case MWIFIEX_REG_MAC: cmd_no = HostCmd_CMD_MAC_REG_ACCESS; break; @@ -1286,9 +1286,9 @@ mwifiex_reg_write(struct mwifiex_private *priv, u32 reg_type, { struct mwifiex_ds_reg_rw reg_rw; - reg_rw.type = cpu_to_le32(reg_type); - reg_rw.offset = cpu_to_le32(reg_offset); - reg_rw.value = cpu_to_le32(reg_value); + reg_rw.type = reg_type; + reg_rw.offset = reg_offset; + reg_rw.value = reg_value; return mwifiex_reg_mem_ioctl_reg_rw(priv, ®_rw, HostCmd_ACT_GEN_SET); } @@ -1306,14 +1306,14 @@ mwifiex_reg_read(struct mwifiex_private *priv, u32 reg_type, int ret; struct mwifiex_ds_reg_rw reg_rw; - reg_rw.type = cpu_to_le32(reg_type); - reg_rw.offset = cpu_to_le32(reg_offset); + reg_rw.type = reg_type; + reg_rw.offset = reg_offset; ret = mwifiex_reg_mem_ioctl_reg_rw(priv, ®_rw, HostCmd_ACT_GEN_GET); if (ret) goto done; - *value = le32_to_cpu(reg_rw.value); + *value = reg_rw.value; done: return ret; @@ -1332,15 +1332,16 @@ mwifiex_eeprom_read(struct mwifiex_private *priv, u16 offset, u16 bytes, int ret; struct mwifiex_ds_read_eeprom rd_eeprom; - rd_eeprom.offset = cpu_to_le16((u16) offset); - rd_eeprom.byte_count = cpu_to_le16((u16) bytes); + rd_eeprom.offset = offset; + rd_eeprom.byte_count = bytes; /* Send request to firmware */ ret = mwifiex_send_cmd(priv, HostCmd_CMD_802_11_EEPROM_ACCESS, HostCmd_ACT_GEN_GET, 0, &rd_eeprom, true); if (!ret) - memcpy(value, rd_eeprom.value, MAX_EEPROM_DATA); + memcpy(value, rd_eeprom.value, min((u16)MAX_EEPROM_DATA, + rd_eeprom.byte_count)); return ret; } -- cgit v0.10.2 From c0174ee28003b22dba0edc160ff6f16c27d3dff1 Mon Sep 17 00:00:00 2001 From: Maital Hahn Date: Tue, 28 Jun 2016 13:41:35 +0300 Subject: wlcore/wl18xx: mesh: added initial mesh support for wl8 1. Added support for interface and role of mesh type. 2. Enabled enable/start of mesh-point role, and opening and closing a connection with a mesh peer. 3. Added multirole combination of mesh and ap under the same limits of dual ap mode. 4. Add support for 'sta_rc_update' opcode for mesh IF. The 'sta_rc_update' opcode is being used in mesh_plink.c. Add support in wlcore to handle this opcode correctly for mesh (as opposed to current implementation that handles STA only). 5. Bumped the firmware version to support new Mesh functionality Signed-off-by: Maital Hahn Signed-off-by: Yaniv Machani Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ti/wl18xx/main.c b/drivers/net/wireless/ti/wl18xx/main.c index ae47c79..4811b74 100644 --- a/drivers/net/wireless/ti/wl18xx/main.c +++ b/drivers/net/wireless/ti/wl18xx/main.c @@ -1821,9 +1821,12 @@ static const struct ieee80211_iface_limit wl18xx_iface_limits[] = { }, { .max = 1, - .types = BIT(NL80211_IFTYPE_AP) | - BIT(NL80211_IFTYPE_P2P_GO) | - BIT(NL80211_IFTYPE_P2P_CLIENT), + .types = BIT(NL80211_IFTYPE_AP) + | BIT(NL80211_IFTYPE_P2P_GO) + | BIT(NL80211_IFTYPE_P2P_CLIENT) +#ifdef CONFIG_MAC80211_MESH + | BIT(NL80211_IFTYPE_MESH_POINT) +#endif }, { .max = 1, @@ -1836,6 +1839,12 @@ static const struct ieee80211_iface_limit wl18xx_iface_ap_limits[] = { .max = 2, .types = BIT(NL80211_IFTYPE_AP), }, +#ifdef CONFIG_MAC80211_MESH + { + .max = 1, + .types = BIT(NL80211_IFTYPE_MESH_POINT), + }, +#endif { .max = 1, .types = BIT(NL80211_IFTYPE_P2P_DEVICE), diff --git a/drivers/net/wireless/ti/wl18xx/wl18xx.h b/drivers/net/wireless/ti/wl18xx/wl18xx.h index 71e9e38..d65cc6d 100644 --- a/drivers/net/wireless/ti/wl18xx/wl18xx.h +++ b/drivers/net/wireless/ti/wl18xx/wl18xx.h @@ -29,7 +29,7 @@ #define WL18XX_IFTYPE_VER 9 #define WL18XX_MAJOR_VER WLCORE_FW_VER_IGNORE #define WL18XX_SUBTYPE_VER WLCORE_FW_VER_IGNORE -#define WL18XX_MINOR_VER 11 +#define WL18XX_MINOR_VER 58 #define WL18XX_CMD_MAX_SIZE 740 diff --git a/drivers/net/wireless/ti/wlcore/acx.h b/drivers/net/wireless/ti/wlcore/acx.h index 0d61fae..6321ed4 100644 --- a/drivers/net/wireless/ti/wlcore/acx.h +++ b/drivers/net/wireless/ti/wlcore/acx.h @@ -105,6 +105,7 @@ enum wl12xx_role { WL1271_ROLE_DEVICE, WL1271_ROLE_P2P_CL, WL1271_ROLE_P2P_GO, + WL1271_ROLE_MESH_POINT, WL12XX_INVALID_ROLE_TYPE = 0xff }; diff --git a/drivers/net/wireless/ti/wlcore/boot.c b/drivers/net/wireless/ti/wlcore/boot.c index 19b7ec7..f75d304 100644 --- a/drivers/net/wireless/ti/wlcore/boot.c +++ b/drivers/net/wireless/ti/wlcore/boot.c @@ -130,7 +130,7 @@ fail: wl1271_error("Your WiFi FW version (%u.%u.%u.%u.%u) is invalid.\n" "Please use at least FW %s\n" "You can get the latest firmwares at:\n" - "git://github.com/TI-OpenLink/firmwares.git", + "git://git.ti.com/wilink8-wlan/wl18xx_fw.git", fw_ver[FW_VER_CHIP], fw_ver[FW_VER_IF_TYPE], fw_ver[FW_VER_MAJOR], fw_ver[FW_VER_SUBTYPE], fw_ver[FW_VER_MINOR], min_fw_str); diff --git a/drivers/net/wireless/ti/wlcore/cmd.c b/drivers/net/wireless/ti/wlcore/cmd.c index 5f360ce..7f4da72 100644 --- a/drivers/net/wireless/ti/wlcore/cmd.c +++ b/drivers/net/wireless/ti/wlcore/cmd.c @@ -629,11 +629,14 @@ int wl12xx_cmd_role_start_ap(struct wl1271 *wl, struct wl12xx_vif *wlvif) wl1271_debug(DEBUG_CMD, "cmd role start ap %d", wlvif->role_id); - /* trying to use hidden SSID with an old hostapd version */ - if (wlvif->ssid_len == 0 && !bss_conf->hidden_ssid) { - wl1271_error("got a null SSID from beacon/bss"); - ret = -EINVAL; - goto out; + /* If MESH --> ssid_len is always 0 */ + if (!ieee80211_vif_is_mesh(vif)) { + /* trying to use hidden SSID with an old hostapd version */ + if (wlvif->ssid_len == 0 && !bss_conf->hidden_ssid) { + wl1271_error("got a null SSID from beacon/bss"); + ret = -EINVAL; + goto out; + } } cmd = kzalloc(sizeof(*cmd), GFP_KERNEL); diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c index 9abc152..4652636 100644 --- a/drivers/net/wireless/ti/wlcore/main.c +++ b/drivers/net/wireless/ti/wlcore/main.c @@ -221,6 +221,7 @@ static void wlcore_rc_update_work(struct work_struct *work) struct wl12xx_vif *wlvif = container_of(work, struct wl12xx_vif, rc_update_work); struct wl1271 *wl = wlvif->wl; + struct ieee80211_vif *vif = wl12xx_wlvif_to_vif(wlvif); mutex_lock(&wl->mutex); @@ -231,8 +232,16 @@ static void wlcore_rc_update_work(struct work_struct *work) if (ret < 0) goto out; - wlcore_hw_sta_rc_update(wl, wlvif); + if (ieee80211_vif_is_mesh(vif)) { + ret = wl1271_acx_set_ht_capabilities(wl, &wlvif->rc_ht_cap, + true, wlvif->sta.hlid); + if (ret < 0) + goto out_sleep; + } else { + wlcore_hw_sta_rc_update(wl, wlvif); + } +out_sleep: wl1271_ps_elp_sleep(wl); out: mutex_unlock(&wl->mutex); @@ -2153,10 +2162,14 @@ static void wlcore_free_klv_template(struct wl1271 *wl, u8 *idx) static u8 wl12xx_get_role_type(struct wl1271 *wl, struct wl12xx_vif *wlvif) { + struct ieee80211_vif *vif = wl12xx_wlvif_to_vif(wlvif); + switch (wlvif->bss_type) { case BSS_TYPE_AP_BSS: if (wlvif->p2p) return WL1271_ROLE_P2P_GO; + else if (ieee80211_vif_is_mesh(vif)) + return WL1271_ROLE_MESH_POINT; else return WL1271_ROLE_AP; @@ -2198,6 +2211,7 @@ static int wl12xx_init_vif_data(struct wl1271 *wl, struct ieee80211_vif *vif) wlvif->p2p = 1; /* fall-through */ case NL80211_IFTYPE_AP: + case NL80211_IFTYPE_MESH_POINT: wlvif->bss_type = BSS_TYPE_AP_BSS; break; default: @@ -4131,9 +4145,14 @@ static void wl1271_bss_info_changed_ap(struct wl1271 *wl, if (ret < 0) goto out; - ret = wl1271_ap_set_probe_resp_tmpl(wl, wlvif->basic_rate, vif); - if (ret < 0) - goto out; + /* No need to set probe resp template for mesh */ + if (!ieee80211_vif_is_mesh(vif)) { + ret = wl1271_ap_set_probe_resp_tmpl(wl, + wlvif->basic_rate, + vif); + if (ret < 0) + goto out; + } ret = wlcore_set_beacon_template(wl, vif, true); if (ret < 0) @@ -5641,6 +5660,7 @@ static void wlcore_op_sta_rc_update(struct ieee80211_hw *hw, /* this callback is atomic, so schedule a new work */ wlvif->rc_update_bw = sta->bandwidth; + memcpy(&wlvif->rc_ht_cap, &sta->ht_cap, sizeof(sta->ht_cap)); ieee80211_queue_work(hw, &wlvif->rc_update_work); } @@ -6062,7 +6082,11 @@ static int wl1271_init_ieee80211(struct wl1271 *wl) BIT(NL80211_IFTYPE_AP) | BIT(NL80211_IFTYPE_P2P_DEVICE) | BIT(NL80211_IFTYPE_P2P_CLIENT) | +#ifdef CONFIG_MAC80211_MESH + BIT(NL80211_IFTYPE_MESH_POINT) | +#endif BIT(NL80211_IFTYPE_P2P_GO); + wl->hw->wiphy->max_scan_ssids = 1; wl->hw->wiphy->max_sched_scan_ssids = 16; wl->hw->wiphy->max_match_sets = 16; diff --git a/drivers/net/wireless/ti/wlcore/wlcore_i.h b/drivers/net/wireless/ti/wlcore/wlcore_i.h index 5c4199f..f5f910b 100644 --- a/drivers/net/wireless/ti/wlcore/wlcore_i.h +++ b/drivers/net/wireless/ti/wlcore/wlcore_i.h @@ -472,6 +472,7 @@ struct wl12xx_vif { /* update rate conrol */ enum ieee80211_sta_rx_bandwidth rc_update_bw; + struct ieee80211_sta_ht_cap rc_ht_cap; struct work_struct rc_update_work; /* -- cgit v0.10.2 From c940de10d45efc5664ee993a6da281f45c804e59 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Wed, 6 Jul 2016 12:22:54 +0200 Subject: brcmfmac: respect hidden_ssid for AP interfaces MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This was succesfully tested with 4366B1. A small workaround is needed for the main interface otherwise it would stuck at the hidden state. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c index b8f7dec..2628d5e 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c @@ -4666,6 +4666,15 @@ brcmf_cfg80211_start_ap(struct wiphy *wiphy, struct net_device *ndev, brcmf_err("SET SSID error (%d)\n", err); goto exit; } + + if (settings->hidden_ssid) { + err = brcmf_fil_iovar_int_set(ifp, "closednet", 1); + if (err) { + brcmf_err("closednet error (%d)\n", err); + goto exit; + } + } + brcmf_dbg(TRACE, "AP mode configuration complete\n"); } else if (dev_role == NL80211_IFTYPE_P2P_GO) { err = brcmf_fil_iovar_int_set(ifp, "chanspec", chanspec); @@ -4724,6 +4733,10 @@ static int brcmf_cfg80211_stop_ap(struct wiphy *wiphy, struct net_device *ndev) return err; } + /* First BSS doesn't get a full reset */ + if (ifp->bsscfgidx == 0) + brcmf_fil_iovar_int_set(ifp, "closednet", 0); + memset(&join_params, 0, sizeof(join_params)); err = brcmf_fil_cmd_data_set(ifp, BRCMF_C_SET_SSID, &join_params, sizeof(join_params)); -- cgit v0.10.2 From f4dc77713f8016d2e8a3295e1c9c53a21f296def Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 14 Jul 2016 17:51:26 +0200 Subject: netfilter: x_tables: speed up jump target validation The dummy ruleset I used to test the original validation change was broken, most rules were unreachable and were not tested by mark_source_chains(). In some cases rulesets that used to load in a few seconds now require several minutes. sample ruleset that shows the behaviour: echo "*filter" for i in $(seq 0 100000);do printf ":chain_%06x - [0:0]\n" $i done for i in $(seq 0 100000);do printf -- "-A INPUT -j chain_%06x\n" $i printf -- "-A INPUT -j chain_%06x\n" $i printf -- "-A INPUT -j chain_%06x\n" $i done echo COMMIT [ pipe result into iptables-restore ] This ruleset will be about 74mbyte in size, with ~500k searches though all 500k[1] rule entries. iptables-restore will take forever (gave up after 10 minutes) Instead of always searching the entire blob for a match, fill an array with the start offsets of every single ipt_entry struct, then do a binary search to check if the jump target is present or not. After this change ruleset restore times get again close to what one gets when reverting 36472341017529e (~3 seconds on my workstation). [1] every user-defined rule gets an implicit RETURN, so we get 300k jumps + 100k userchains + 100k returns -> 500k rule entries Fixes: 36472341017529e ("netfilter: x_tables: validate targets of jumps") Reported-by: Jeff Wu Tested-by: Jeff Wu Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index e94e81a..2ad1a2b 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -250,6 +250,10 @@ int xt_check_entry_offsets(const void *base, const char *elems, unsigned int target_offset, unsigned int next_offset); +unsigned int *xt_alloc_entry_offsets(unsigned int size); +bool xt_find_jump_offset(const unsigned int *offsets, + unsigned int target, unsigned int size); + int xt_check_match(struct xt_mtchk_param *, unsigned int size, u_int8_t proto, bool inv_proto); int xt_check_target(struct xt_tgchk_param *, unsigned int size, u_int8_t proto, diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index c8dd9e2..b31df59 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -299,23 +299,12 @@ static inline bool unconditional(const struct arpt_entry *e) memcmp(&e->arp, &uncond, sizeof(uncond)) == 0; } -static bool find_jump_target(const struct xt_table_info *t, - const struct arpt_entry *target) -{ - struct arpt_entry *iter; - - xt_entry_foreach(iter, t->entries, t->size) { - if (iter == target) - return true; - } - return false; -} - /* Figures out from what hook each rule can be called: returns 0 if * there are loops. Puts hook bitmask in comefrom. */ static int mark_source_chains(const struct xt_table_info *newinfo, - unsigned int valid_hooks, void *entry0) + unsigned int valid_hooks, void *entry0, + unsigned int *offsets) { unsigned int hook; @@ -388,10 +377,11 @@ static int mark_source_chains(const struct xt_table_info *newinfo, XT_STANDARD_TARGET) == 0 && newpos >= 0) { /* This a jump; chase it. */ + if (!xt_find_jump_offset(offsets, newpos, + newinfo->number)) + return 0; e = (struct arpt_entry *) (entry0 + newpos); - if (!find_jump_target(newinfo, e)) - return 0; } else { /* ... this is a fallthru */ newpos = pos + e->next_offset; @@ -543,6 +533,7 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, const struct arpt_replace *repl) { struct arpt_entry *iter; + unsigned int *offsets; unsigned int i; int ret = 0; @@ -555,6 +546,9 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, newinfo->underflow[i] = 0xFFFFFFFF; } + offsets = xt_alloc_entry_offsets(newinfo->number); + if (!offsets) + return -ENOMEM; i = 0; /* Walk through entries, checking offsets. */ @@ -565,17 +559,20 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, repl->underflow, repl->valid_hooks); if (ret != 0) - break; + goto out_free; + if (i < repl->num_entries) + offsets[i] = (void *)iter - entry0; ++i; if (strcmp(arpt_get_target(iter)->u.user.name, XT_ERROR_TARGET) == 0) ++newinfo->stacksize; } if (ret != 0) - return ret; + goto out_free; + ret = -EINVAL; if (i != repl->num_entries) - return -EINVAL; + goto out_free; /* Check hooks all assigned */ for (i = 0; i < NF_ARP_NUMHOOKS; i++) { @@ -583,13 +580,16 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, if (!(repl->valid_hooks & (1 << i))) continue; if (newinfo->hook_entry[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; if (newinfo->underflow[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; } - if (!mark_source_chains(newinfo, repl->valid_hooks, entry0)) - return -ELOOP; + if (!mark_source_chains(newinfo, repl->valid_hooks, entry0, offsets)) { + ret = -ELOOP; + goto out_free; + } + kvfree(offsets); /* Finally, each sanity check must pass */ i = 0; @@ -610,6 +610,9 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, } return ret; + out_free: + kvfree(offsets); + return ret; } static void get_counters(const struct xt_table_info *t, diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index f0df66f..f993545 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -373,23 +373,12 @@ ipt_do_table(struct sk_buff *skb, else return verdict; } -static bool find_jump_target(const struct xt_table_info *t, - const struct ipt_entry *target) -{ - struct ipt_entry *iter; - - xt_entry_foreach(iter, t->entries, t->size) { - if (iter == target) - return true; - } - return false; -} - /* Figures out from what hook each rule can be called: returns 0 if there are loops. Puts hook bitmask in comefrom. */ static int mark_source_chains(const struct xt_table_info *newinfo, - unsigned int valid_hooks, void *entry0) + unsigned int valid_hooks, void *entry0, + unsigned int *offsets) { unsigned int hook; @@ -458,10 +447,11 @@ mark_source_chains(const struct xt_table_info *newinfo, XT_STANDARD_TARGET) == 0 && newpos >= 0) { /* This a jump; chase it. */ + if (!xt_find_jump_offset(offsets, newpos, + newinfo->number)) + return 0; e = (struct ipt_entry *) (entry0 + newpos); - if (!find_jump_target(newinfo, e)) - return 0; } else { /* ... this is a fallthru */ newpos = pos + e->next_offset; @@ -694,6 +684,7 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, const struct ipt_replace *repl) { struct ipt_entry *iter; + unsigned int *offsets; unsigned int i; int ret = 0; @@ -706,6 +697,9 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, newinfo->underflow[i] = 0xFFFFFFFF; } + offsets = xt_alloc_entry_offsets(newinfo->number); + if (!offsets) + return -ENOMEM; i = 0; /* Walk through entries, checking offsets. */ xt_entry_foreach(iter, entry0, newinfo->size) { @@ -715,15 +709,18 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, repl->underflow, repl->valid_hooks); if (ret != 0) - return ret; + goto out_free; + if (i < repl->num_entries) + offsets[i] = (void *)iter - entry0; ++i; if (strcmp(ipt_get_target(iter)->u.user.name, XT_ERROR_TARGET) == 0) ++newinfo->stacksize; } + ret = -EINVAL; if (i != repl->num_entries) - return -EINVAL; + goto out_free; /* Check hooks all assigned */ for (i = 0; i < NF_INET_NUMHOOKS; i++) { @@ -731,13 +728,16 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, if (!(repl->valid_hooks & (1 << i))) continue; if (newinfo->hook_entry[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; if (newinfo->underflow[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; } - if (!mark_source_chains(newinfo, repl->valid_hooks, entry0)) - return -ELOOP; + if (!mark_source_chains(newinfo, repl->valid_hooks, entry0, offsets)) { + ret = -ELOOP; + goto out_free; + } + kvfree(offsets); /* Finally, each sanity check must pass */ i = 0; @@ -758,6 +758,9 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, } return ret; + out_free: + kvfree(offsets); + return ret; } static void diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 61ed950..552fac2 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -402,23 +402,12 @@ ip6t_do_table(struct sk_buff *skb, else return verdict; } -static bool find_jump_target(const struct xt_table_info *t, - const struct ip6t_entry *target) -{ - struct ip6t_entry *iter; - - xt_entry_foreach(iter, t->entries, t->size) { - if (iter == target) - return true; - } - return false; -} - /* Figures out from what hook each rule can be called: returns 0 if there are loops. Puts hook bitmask in comefrom. */ static int mark_source_chains(const struct xt_table_info *newinfo, - unsigned int valid_hooks, void *entry0) + unsigned int valid_hooks, void *entry0, + unsigned int *offsets) { unsigned int hook; @@ -487,10 +476,11 @@ mark_source_chains(const struct xt_table_info *newinfo, XT_STANDARD_TARGET) == 0 && newpos >= 0) { /* This a jump; chase it. */ + if (!xt_find_jump_offset(offsets, newpos, + newinfo->number)) + return 0; e = (struct ip6t_entry *) (entry0 + newpos); - if (!find_jump_target(newinfo, e)) - return 0; } else { /* ... this is a fallthru */ newpos = pos + e->next_offset; @@ -724,6 +714,7 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, const struct ip6t_replace *repl) { struct ip6t_entry *iter; + unsigned int *offsets; unsigned int i; int ret = 0; @@ -736,6 +727,9 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, newinfo->underflow[i] = 0xFFFFFFFF; } + offsets = xt_alloc_entry_offsets(newinfo->number); + if (!offsets) + return -ENOMEM; i = 0; /* Walk through entries, checking offsets. */ xt_entry_foreach(iter, entry0, newinfo->size) { @@ -745,15 +739,18 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, repl->underflow, repl->valid_hooks); if (ret != 0) - return ret; + goto out_free; + if (i < repl->num_entries) + offsets[i] = (void *)iter - entry0; ++i; if (strcmp(ip6t_get_target(iter)->u.user.name, XT_ERROR_TARGET) == 0) ++newinfo->stacksize; } + ret = -EINVAL; if (i != repl->num_entries) - return -EINVAL; + goto out_free; /* Check hooks all assigned */ for (i = 0; i < NF_INET_NUMHOOKS; i++) { @@ -761,13 +758,16 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, if (!(repl->valid_hooks & (1 << i))) continue; if (newinfo->hook_entry[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; if (newinfo->underflow[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; } - if (!mark_source_chains(newinfo, repl->valid_hooks, entry0)) - return -ELOOP; + if (!mark_source_chains(newinfo, repl->valid_hooks, entry0, offsets)) { + ret = -ELOOP; + goto out_free; + } + kvfree(offsets); /* Finally, each sanity check must pass */ i = 0; @@ -788,6 +788,9 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, } return ret; + out_free: + kvfree(offsets); + return ret; } static void diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index fe0e2db..e0aa7c1 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -702,6 +702,56 @@ int xt_check_entry_offsets(const void *base, } EXPORT_SYMBOL(xt_check_entry_offsets); +/** + * xt_alloc_entry_offsets - allocate array to store rule head offsets + * + * @size: number of entries + * + * Return: NULL or kmalloc'd or vmalloc'd array + */ +unsigned int *xt_alloc_entry_offsets(unsigned int size) +{ + unsigned int *off; + + off = kcalloc(size, sizeof(unsigned int), GFP_KERNEL | __GFP_NOWARN); + + if (off) + return off; + + if (size < (SIZE_MAX / sizeof(unsigned int))) + off = vmalloc(size * sizeof(unsigned int)); + + return off; +} +EXPORT_SYMBOL(xt_alloc_entry_offsets); + +/** + * xt_find_jump_offset - check if target is a valid jump offset + * + * @offsets: array containing all valid rule start offsets of a rule blob + * @target: the jump target to search for + * @size: entries in @offset + */ +bool xt_find_jump_offset(const unsigned int *offsets, + unsigned int target, unsigned int size) +{ + int m, low = 0, hi = size; + + while (hi > low) { + m = (low + hi) / 2u; + + if (offsets[m] > target) + hi = m; + else if (offsets[m] < target) + low = m + 1; + else + return true; + } + + return false; +} +EXPORT_SYMBOL(xt_find_jump_offset); + int xt_check_target(struct xt_tgchk_param *par, unsigned int size, u_int8_t proto, bool inv_proto) { -- cgit v0.10.2 From 88f07e70d1be6bac4e105a0b690a64ba84fe867e Mon Sep 17 00:00:00 2001 From: Maxim Altshul Date: Mon, 11 Jul 2016 17:22:32 +0300 Subject: wlcore/wl18xx: Add functionality to accept TX rate per link FW will provide a TX rate per link for each FW status, and wlcore will be able to store the information for the use of the mesh hwmp module. This is used mainly in mesh. Rates are reported when a mesh interface is up. Signed-off-by: Maxim Altshul Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ti/wl18xx/main.c b/drivers/net/wireless/ti/wl18xx/main.c index 4811b74..00a04df 100644 --- a/drivers/net/wireless/ti/wl18xx/main.c +++ b/drivers/net/wireless/ti/wl18xx/main.c @@ -1214,6 +1214,10 @@ static void wl18xx_convert_fw_status(struct wl1271 *wl, void *raw_fw_status, int_fw_status->counters.tx_voice_released_blks; fw_status->counters.tx_last_rate = int_fw_status->counters.tx_last_rate; + fw_status->counters.tx_last_rate_mbps = + int_fw_status->counters.tx_last_rate_mbps; + fw_status->counters.hlid = + int_fw_status->counters.hlid; fw_status->log_start_addr = le32_to_cpu(int_fw_status->log_start_addr); diff --git a/drivers/net/wireless/ti/wl18xx/tx.c b/drivers/net/wireless/ti/wl18xx/tx.c index ebaf66e..876aef1 100644 --- a/drivers/net/wireless/ti/wl18xx/tx.c +++ b/drivers/net/wireless/ti/wl18xx/tx.c @@ -30,9 +30,9 @@ static void wl18xx_get_last_tx_rate(struct wl1271 *wl, struct ieee80211_vif *vif, - u8 band, struct ieee80211_tx_rate *rate) + u8 band, struct ieee80211_tx_rate *rate, u8 hlid) { - u8 fw_rate = wl->fw_status->counters.tx_last_rate; + u8 fw_rate = wl->links[hlid].fw_rate_idx; if (fw_rate > CONF_HW_RATE_INDEX_MAX) { wl1271_error("last Tx rate invalid: %d", fw_rate); @@ -79,6 +79,7 @@ static void wl18xx_tx_complete_packet(struct wl1271 *wl, u8 tx_stat_byte) struct sk_buff *skb; int id = tx_stat_byte & WL18XX_TX_STATUS_DESC_ID_MASK; bool tx_success; + struct wl1271_tx_hw_descr *tx_desc; /* check for id legality */ if (unlikely(id >= wl->num_tx_desc || wl->tx_frames[id] == NULL)) { @@ -91,6 +92,7 @@ static void wl18xx_tx_complete_packet(struct wl1271 *wl, u8 tx_stat_byte) skb = wl->tx_frames[id]; info = IEEE80211_SKB_CB(skb); + tx_desc = (struct wl1271_tx_hw_descr *)skb->data; if (wl12xx_is_dummy_packet(wl, skb)) { wl1271_free_tx_id(wl, id); @@ -105,7 +107,9 @@ static void wl18xx_tx_complete_packet(struct wl1271 *wl, u8 tx_stat_byte) * the info->status structures */ wl18xx_get_last_tx_rate(wl, info->control.vif, - info->band, &info->status.rates[0]); + info->band, + &info->status.rates[0], + tx_desc->hlid); info->status.rates[0].count = 1; /* no data about retries */ info->status.ack_signal = -1; @@ -144,12 +148,22 @@ void wl18xx_tx_immediate_complete(struct wl1271 *wl) struct wl18xx_fw_status_priv *status_priv = (struct wl18xx_fw_status_priv *)wl->fw_status->priv; struct wl18xx_priv *priv = wl->priv; - u8 i; + u8 i, hlid; /* nothing to do here */ if (priv->last_fw_rls_idx == status_priv->fw_release_idx) return; + /* update rates per link */ + hlid = wl->fw_status->counters.hlid; + + if (hlid < WLCORE_MAX_LINKS) { + wl->links[hlid].fw_rate_idx = + wl->fw_status->counters.tx_last_rate; + wl->links[hlid].fw_rate_mbps = + wl->fw_status->counters.tx_last_rate_mbps; + } + /* freed Tx descriptors */ wl1271_debug(DEBUG_TX, "last released desc = %d, current idx = %d", priv->last_fw_rls_idx, status_priv->fw_release_idx); diff --git a/drivers/net/wireless/ti/wl18xx/wl18xx.h b/drivers/net/wireless/ti/wl18xx/wl18xx.h index d65cc6d..5371cbd 100644 --- a/drivers/net/wireless/ti/wl18xx/wl18xx.h +++ b/drivers/net/wireless/ti/wl18xx/wl18xx.h @@ -125,7 +125,11 @@ struct wl18xx_fw_packet_counters { /* Tx rate of the last transmitted packet */ u8 tx_last_rate; - u8 padding[2]; + /* Tx rate or Tx rate estimate pre-calculated by fw in mbps units */ + u8 tx_last_rate_mbps; + + /* hlid for which the rates were reported */ + u8 hlid; } __packed; /* FW status registers */ diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c index 4652636..45aab10 100644 --- a/drivers/net/wireless/ti/wlcore/main.c +++ b/drivers/net/wireless/ti/wlcore/main.c @@ -4986,6 +4986,7 @@ static int wl12xx_sta_add(struct wl1271 *wl, return ret; wl_sta = (struct wl1271_station *)sta->drv_priv; + wl_sta->wl = wl; hlid = wl_sta->hlid; ret = wl12xx_cmd_add_peer(wl, wlvif, sta, hlid); diff --git a/drivers/net/wireless/ti/wlcore/rx.c b/drivers/net/wireless/ti/wlcore/rx.c index c9bd294..b9e1404 100644 --- a/drivers/net/wireless/ti/wlcore/rx.c +++ b/drivers/net/wireless/ti/wlcore/rx.c @@ -222,6 +222,13 @@ int wlcore_rx(struct wl1271 *wl, struct wl_fw_status *status) enum wl_rx_buf_align rx_align; int ret = 0; + /* update rates per link */ + hlid = status->counters.hlid; + + if (hlid < WLCORE_MAX_LINKS) + wl->links[hlid].fw_rate_mbps = + status->counters.tx_last_rate_mbps; + while (drv_rx_counter != fw_rx_counter) { buf_size = 0; rx_counter = drv_rx_counter; diff --git a/drivers/net/wireless/ti/wlcore/wlcore_i.h b/drivers/net/wireless/ti/wlcore/wlcore_i.h index f5f910b..242b4e3 100644 --- a/drivers/net/wireless/ti/wlcore/wlcore_i.h +++ b/drivers/net/wireless/ti/wlcore/wlcore_i.h @@ -171,6 +171,12 @@ struct wl_fw_status { /* Tx rate of the last transmitted packet */ u8 tx_last_rate; + + /* Tx rate or Tx rate estimate pre calculated by fw in mbps */ + u8 tx_last_rate_mbps; + + /* hlid for which the rates were reported */ + u8 hlid; } counters; u32 log_start_addr; @@ -273,6 +279,12 @@ struct wl1271_link { /* bitmap of TIDs where RX BA sessions are active for this link */ u8 ba_bitmap; + /* the last fw rate index we used for this link */ + u8 fw_rate_idx; + + /* the last fw rate [Mbps] we used for this link */ + u8 fw_rate_mbps; + /* The wlvif this link belongs to. Might be null for global links */ struct wl12xx_vif *wlvif; @@ -335,6 +347,7 @@ struct wl1271_station { * Used in both AP and STA mode. */ u64 total_freed_pkts; + struct wl1271 *wl; }; struct wl12xx_vif { -- cgit v0.10.2 From 5f6d4ca3c196814bef0cbbb195acd9ecc178588b Mon Sep 17 00:00:00 2001 From: Maxim Altshul Date: Mon, 11 Jul 2016 17:22:33 +0300 Subject: wlcore: Add support for get_expected_throughput opcode Adding this opcode, allows the TI wireless driver, to report throughput directly from FW to mac80211. This is used mainly for mesh metric calculation. Signed-off-by: Maxim Altshul [kvalo@codeaurora.org: fix indentation] Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c index 45aab10..1d68916 100644 --- a/drivers/net/wireless/ti/wlcore/main.c +++ b/drivers/net/wireless/ti/wlcore/main.c @@ -5700,6 +5700,16 @@ out: mutex_unlock(&wl->mutex); } +static u32 wlcore_op_get_expected_throughput(struct ieee80211_sta *sta) +{ + struct wl1271_station *wl_sta = (struct wl1271_station *)sta->drv_priv; + struct wl1271 *wl = wl_sta->wl; + u8 hlid = wl_sta->hlid; + + /* return in units of Kbps */ + return (wl->links[hlid].fw_rate_mbps * 1000); +} + static bool wl1271_tx_frames_pending(struct ieee80211_hw *hw) { struct wl1271 *wl = hw->priv; @@ -5900,6 +5910,7 @@ static const struct ieee80211_ops wl1271_ops = { .switch_vif_chanctx = wlcore_op_switch_vif_chanctx, .sta_rc_update = wlcore_op_sta_rc_update, .sta_statistics = wlcore_op_sta_statistics, + .get_expected_throughput = wlcore_op_get_expected_throughput, CFG80211_TESTMODE_CMD(wl1271_tm_cmd) }; -- cgit v0.10.2 From 1bea0512c3394965de28a152149b90afd686fae5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Mon, 11 Jul 2016 23:01:36 +0200 Subject: bcma: add PCI ID for Foxconn's BCM43142 device MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After discovering there are 2 very different 14e4:4365 PCI devices we made ID tables less generic. Back then we believed there are only 2 such devices: 1) 14e4:4365 1028:0016 with SoftMAC BCM43142 chipset 2) 14e4:4365 14e4:4365 with FullMAC BCM4366 chipset >From the recent report it appears there is also 14e4:4365 105b:e092 which should be claimed by bcma. Add back support for it. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=121881 Fixes: 515b399c9a20 ("bcma: claim only 14e4:4365 PCI Dell card with SoftMAC BCM43142") Reported-by: Igor Mammedov Signed-off-by: Rafał Miłecki Cc: Stable [4.6+] Tested-by: Igor Mammedov Signed-off-by: Kalle Valo diff --git a/drivers/bcma/host_pci.c b/drivers/bcma/host_pci.c index cae5385..bd46569 100644 --- a/drivers/bcma/host_pci.c +++ b/drivers/bcma/host_pci.c @@ -295,6 +295,7 @@ static const struct pci_device_id bcma_pci_bridge_tbl[] = { { PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4359) }, { PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x4360) }, { PCI_DEVICE_SUB(PCI_VENDOR_ID_BROADCOM, 0x4365, PCI_VENDOR_ID_DELL, 0x0016) }, + { PCI_DEVICE_SUB(PCI_VENDOR_ID_BROADCOM, 0x4365, PCI_VENDOR_ID_FOXCONN, 0xe092) }, { PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x43a0) }, { PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x43a9) }, { PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, 0x43aa) }, -- cgit v0.10.2 From 83d58d53e062aad0f9dfbb68780bfaebf4e8bafe Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 12 Jul 2016 11:32:19 +0000 Subject: libertas: fix non static symbol warning Fixes the following sparse warning: drivers/net/wireless/marvell/libertas/cfg.c:2047:5: warning: symbol 'lbs_set_power_mgmt' was not declared. Should it be static? Signed-off-by: Wei Yongjun Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/libertas/cfg.c b/drivers/net/wireless/marvell/libertas/cfg.c index ea48024..7ff2efa 100644 --- a/drivers/net/wireless/marvell/libertas/cfg.c +++ b/drivers/net/wireless/marvell/libertas/cfg.c @@ -2044,8 +2044,8 @@ static int lbs_leave_ibss(struct wiphy *wiphy, struct net_device *dev) -int lbs_set_power_mgmt(struct wiphy *wiphy, struct net_device *dev, - bool enabled, int timeout) +static int lbs_set_power_mgmt(struct wiphy *wiphy, struct net_device *dev, + bool enabled, int timeout) { struct lbs_private *priv = wiphy_priv(wiphy); -- cgit v0.10.2 From 4028a514eaa6649693be1a7cb8d2b84a8fc8cc8c Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 12 Jul 2016 11:43:18 +0000 Subject: mwifiex: fix possible memory leak in mwifiex_cfg80211_start_ap() memory is malloced in mwifiex_cfg80211_start_ap() and should be freed before leaving from the error handling cases, otherwise it will cause memory leak. Signed-off-by: Wei Yongjun Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/cfg80211.c b/drivers/net/wireless/marvell/mwifiex/cfg80211.c index 867ab81..a8ff969 100644 --- a/drivers/net/wireless/marvell/mwifiex/cfg80211.c +++ b/drivers/net/wireless/marvell/mwifiex/cfg80211.c @@ -1936,10 +1936,9 @@ static int mwifiex_cfg80211_start_ap(struct wiphy *wiphy, mwifiex_set_uap_rates(bss_cfg, params); if (mwifiex_set_secure_params(priv, bss_cfg, params)) { - kfree(bss_cfg); mwifiex_dbg(priv->adapter, ERROR, "Failed to parse secuirty parameters!\n"); - return -1; + goto out; } mwifiex_set_ht_params(priv, bss_cfg, params); @@ -1968,7 +1967,7 @@ static int mwifiex_cfg80211_start_ap(struct wiphy *wiphy, if (mwifiex_11h_activate(priv, false)) { mwifiex_dbg(priv->adapter, ERROR, "Failed to disable 11h extensions!!"); - return -1; + goto out; } priv->state_11h.is_11h_active = false; } @@ -1976,12 +1975,11 @@ static int mwifiex_cfg80211_start_ap(struct wiphy *wiphy, if (mwifiex_config_start_uap(priv, bss_cfg)) { mwifiex_dbg(priv->adapter, ERROR, "Failed to start AP\n"); - kfree(bss_cfg); - return -1; + goto out; } if (mwifiex_set_mgmt_ies(priv, ¶ms->beacon)) - return -1; + goto out; if (!netif_carrier_ok(priv->netdev)) netif_carrier_on(priv->netdev); @@ -1990,6 +1988,10 @@ static int mwifiex_cfg80211_start_ap(struct wiphy *wiphy, memcpy(&priv->bss_cfg, bss_cfg, sizeof(priv->bss_cfg)); kfree(bss_cfg); return 0; + +out: + kfree(bss_cfg); + return -1; } /* -- cgit v0.10.2 From 82bc9ab6a8f577d2174a736c33f3d4ecf7d9ef47 Mon Sep 17 00:00:00 2001 From: Arend Van Spriel Date: Fri, 15 Jul 2016 12:16:12 +0200 Subject: brcmfmac: restore stopping netdev queue when bus clogs up MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When the host-interface bus has hard time handling transmit packets it informs higher layer about this and it would stop the netdev queue when needed. However, since commit 9cd18359d31e ("brcmfmac: Make FWS queueing configurable.") this was broken. With this patch the behaviour is restored. Cc: stable@vger.kernel.org # v4.5, v4.6, v4.7 Fixes: 9cd18359d31e ("brcmfmac: Make FWS queueing configurable.") Tested-by: Per Förlin Reviewed-by: Hante Meuleman Reviewed-by: Pieter-Paul Giesberts Reviewed-by: Franky Lin Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c index cd221ab..9f9024a 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fwsignal.c @@ -2469,10 +2469,22 @@ void brcmf_fws_bustxfail(struct brcmf_fws_info *fws, struct sk_buff *skb) void brcmf_fws_bus_blocked(struct brcmf_pub *drvr, bool flow_blocked) { struct brcmf_fws_info *fws = drvr->fws; + struct brcmf_if *ifp; + int i; - fws->bus_flow_blocked = flow_blocked; - if (!flow_blocked) - brcmf_fws_schedule_deq(fws); - else - fws->stats.bus_flow_block++; + if (fws->avoid_queueing) { + for (i = 0; i < BRCMF_MAX_IFS; i++) { + ifp = drvr->iflist[i]; + if (!ifp || !ifp->ndev) + continue; + brcmf_txflowblock_if(ifp, BRCMF_NETIF_STOP_REASON_FLOW, + flow_blocked); + } + } else { + fws->bus_flow_blocked = flow_blocked; + if (!flow_blocked) + brcmf_fws_schedule_deq(fws); + else + fws->stats.bus_flow_block++; + } } -- cgit v0.10.2 From fd3ed33f51c2a586412d35b4f64803f019ab589f Mon Sep 17 00:00:00 2001 From: Arend Van Spriel Date: Fri, 15 Jul 2016 12:39:13 +0200 Subject: brcmfmac: defer DPC processing during probe The sdio dpc starts processing when in SDIOD_STATE_DATA. This state was entered right after firmware download. This patch moves that transition just before enabling sdio interrupt handling thus avoiding watchdog expiry which would put the bus to sleep while probing. Reviewed-by: Hante Meuleman Reviewed-by: Pieter-Paul Giesberts Reviewed-by: Franky Lin Signed-off-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c index 5fb8b91..68ab3ac 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c @@ -3305,10 +3305,6 @@ static int brcmf_sdio_download_firmware(struct brcmf_sdio *bus, goto err; } - /* Allow full data communication using DPC from now on. */ - brcmf_sdiod_change_state(bus->sdiodev, BRCMF_SDIOD_DATA); - bcmerror = 0; - err: brcmf_sdio_clkctl(bus, CLK_SDONLY, false); sdio_release_host(bus->sdiodev->func[1]); @@ -4046,6 +4042,9 @@ static void brcmf_sdio_firmware_callback(struct device *dev, } if (err == 0) { + /* Allow full data communication using DPC from now on. */ + brcmf_sdiod_change_state(bus->sdiodev, BRCMF_SDIOD_DATA); + err = brcmf_sdiod_intr_register(sdiodev); if (err != 0) brcmf_err("intr register failed:%d\n", err); -- cgit v0.10.2 From 2af86f9d954ee86e9b1c492e025de37a1b6d2db8 Mon Sep 17 00:00:00 2001 From: Karthik D A Date: Fri, 15 Jul 2016 18:18:39 +0530 Subject: mwifiex: Fix request_irq() failure handling It's been observed that request_irq() failure leads to a system crash due to a bug in mwifiex driver. When this failure happens, mwifiex_add_card() already takes care of clearing and freeing adapter->card pointer. This patch removes the redundant cleanup code causing crash. Signed-off-by: Karthik D A Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c index 22fe993..0faf4a2 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.c +++ b/drivers/net/wireless/marvell/mwifiex/pcie.c @@ -202,7 +202,6 @@ static int mwifiex_pcie_probe(struct pci_dev *pdev, if (mwifiex_add_card(card, &add_remove_card_sem, &pcie_ops, MWIFIEX_PCIE)) { pr_err("%s failed\n", __func__); - kfree(card); return -1; } @@ -2843,7 +2842,6 @@ static int mwifiex_pcie_request_irq(struct mwifiex_adapter *adapter) "MRVL_PCIE", &card->share_irq_ctx); if (ret) { pr_err("request_irq failed: ret=%d\n", ret); - adapter->card = NULL; return -1; } -- cgit v0.10.2 From 514952889544fcb069bbedfec4aa624fc9e26c4b Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Fri, 15 Jul 2016 19:07:04 +0530 Subject: mwifiex: fix PCIe legacy interrupt problem In corner case, we may end up processing same interrupt twice. We have a logic to read pending interrupts at the end of interrupt processing routine. It has a race with interrupts read in interrupt handler. This patch solves the problem by ORing the interrupt bitmap in this case. The symptom for this bug is below messages in dmesg log. [ 11.522123] mwifiex_pcie 0000:01:00.0: CMD_RESP: invalid cmd resp [ 11.680412] mwifiex_pcie 0000:01:00.0: There is no command but got cmdrsp Link: https://bugzilla.kernel.org/show_bug.cgi?id=109681 Signed-off-by: Amitkumar Karwar Signed-off-by: Cathy Luo Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c index 0faf4a2..12d0ee1 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.c +++ b/drivers/net/wireless/marvell/mwifiex/pcie.c @@ -2300,6 +2300,12 @@ static int mwifiex_process_pcie_int(struct mwifiex_adapter *adapter) } } + if (!card->msi_enable) { + spin_lock_irqsave(&adapter->int_lock, flags); + pcie_ireg |= adapter->int_status; + adapter->int_status = 0; + spin_unlock_irqrestore(&adapter->int_lock, flags); + } } mwifiex_dbg(adapter, INTR, "info: cmd_sent=%d data_sent=%d\n", -- cgit v0.10.2 From 2fd40d2d349f0faf49460b4eca3f49dfe9094619 Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Fri, 15 Jul 2016 19:07:05 +0530 Subject: mwifiex: update command response skb length correctly Same skb is being reused for storing command response from firmware in PCIe chipsets. There was a bug while updating the skb length. This patch ensures skb length correctly gets updated based on rx_len. Signed-off-by: Amitkumar Karwar Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/marvell/mwifiex/pcie.c b/drivers/net/wireless/marvell/mwifiex/pcie.c index 12d0ee1..453ab6a 100644 --- a/drivers/net/wireless/marvell/mwifiex/pcie.c +++ b/drivers/net/wireless/marvell/mwifiex/pcie.c @@ -1616,6 +1616,7 @@ static int mwifiex_pcie_process_cmd_complete(struct mwifiex_adapter *adapter) pkt_len = *((__le16 *)skb->data); rx_len = le16_to_cpu(pkt_len); + skb_put(skb, MWIFIEX_UPLD_SIZE - skb->len); skb_trim(skb, rx_len); skb_pull(skb, INTF_HEADER_LEN); -- cgit v0.10.2 From 57d8f7dd2132df3ac21044e93a8ecdc9744b4459 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Mon, 18 Jul 2016 12:34:14 +0200 Subject: bcma: allow enabling serial flash support on non-MIPS SoCs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit So far we had only MIPS devices with serial flash connected to the SoC's ChipCommon. ARM devices got a separated SPI controller and weere using standard SPI drivers. This has changed with the wireless SoC BCM47189B0. It's ARM based but has serial flash attached just like older devices. This allows using existing driver with these devices. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/bcma/Kconfig b/drivers/bcma/Kconfig index efdc2ae..b5c48a8 100644 --- a/drivers/bcma/Kconfig +++ b/drivers/bcma/Kconfig @@ -76,9 +76,16 @@ config BCMA_PFLASH default y config BCMA_SFLASH - bool - depends on BCMA_DRIVER_MIPS + bool "ChipCommon-attached serial flash support" + depends on BCMA_HOST_SOC default y + help + Some cheap devices have serial flash connected to the ChipCommon + instead of independent SPI controller. It requires using a separated + driver that implements ChipCommon specific interface communication. + + Enabling this symbol will let bcma recognize serial flash and register + it as platform device. config BCMA_NFLASH bool -- cgit v0.10.2 From 0d7eacbe637952fc737a968bc16db1f2ccbbe71c Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Tue, 19 Jul 2016 11:02:59 +0800 Subject: macvtap: correctly free skb during socket destruction We should use kfree_skb() instead of kfree() to free an skb. Fixes: 362899b8725b ("macvtap: switch to use skb array") Reported-by: Dan Carpenter Signed-off-by: Jason Wang Signed-off-by: David S. Miller diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index 9204d19..a38c0da 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -536,7 +536,7 @@ static void macvtap_sock_destruct(struct sock *sk) struct sk_buff *skb; while ((skb = skb_array_consume(&q->skb_array)) != NULL) - kfree(skb); + kfree_skb(skb); } static int macvtap_open(struct inode *inode, struct file *file) -- cgit v0.10.2 From 4ca1807815aa6801aaced7fdefa9edacc2521767 Mon Sep 17 00:00:00 2001 From: Michal Kazior Date: Mon, 18 Jul 2016 23:22:18 +0300 Subject: ath10k: disable wake_tx_queue for older devices Ideally wake_tx_queue should be used regardless as it is a requirement for reducing bufferbloat and implementing airtime fairness in the future. However some setups (typically low-end platforms hosting QCA988X) suffer performance regressions with the current wake_tx_queue implementation. Therefore disable it unless it is really beneficial with current codebase (which is when firmware supports smart pull-push tx scheduling). Signed-off-by: Michal Kazior Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h index 9374bcd..30ae5bf 100644 --- a/drivers/net/wireless/ath/ath10k/core.h +++ b/drivers/net/wireless/ath/ath10k/core.h @@ -676,6 +676,7 @@ struct ath10k_fw_components { struct ath10k { struct ath_common ath_common; struct ieee80211_hw *hw; + struct ieee80211_ops *ops; struct device *dev; u8 mac_addr[ETH_ALEN]; diff --git a/drivers/net/wireless/ath/ath10k/mac.c b/drivers/net/wireless/ath/ath10k/mac.c index 55c823f..fb8e38d 100644 --- a/drivers/net/wireless/ath/ath10k/mac.c +++ b/drivers/net/wireless/ath/ath10k/mac.c @@ -7506,21 +7506,32 @@ static const struct ieee80211_channel ath10k_5ghz_channels[] = { struct ath10k *ath10k_mac_create(size_t priv_size) { struct ieee80211_hw *hw; + struct ieee80211_ops *ops; struct ath10k *ar; - hw = ieee80211_alloc_hw(sizeof(struct ath10k) + priv_size, &ath10k_ops); - if (!hw) + ops = kmemdup(&ath10k_ops, sizeof(ath10k_ops), GFP_KERNEL); + if (!ops) return NULL; + hw = ieee80211_alloc_hw(sizeof(struct ath10k) + priv_size, ops); + if (!hw) { + kfree(ops); + return NULL; + } + ar = hw->priv; ar->hw = hw; + ar->ops = ops; return ar; } void ath10k_mac_destroy(struct ath10k *ar) { + struct ieee80211_ops *ops = ar->ops; + ieee80211_free_hw(ar->hw); + kfree(ops); } static const struct ieee80211_iface_limit ath10k_if_limits[] = { @@ -7954,6 +7965,15 @@ int ath10k_mac_register(struct ath10k *ar) ath10k_warn(ar, "failed to initialise DFS pattern detector\n"); } + /* Current wake_tx_queue implementation imposes a significant + * performance penalty in some setups. The tx scheduling code needs + * more work anyway so disable the wake_tx_queue unless firmware + * supports the pull-push mechanism. + */ + if (!test_bit(ATH10K_FW_FEATURE_PEER_FLOW_CONTROL, + ar->running_fw->fw_file.fw_features)) + ar->ops->wake_tx_queue = NULL; + ret = ath_regd_init(&ar->ath_common.regulatory, ar->hw->wiphy, ath10k_reg_notifier); if (ret) { -- cgit v0.10.2 From 270df8f80d67f257d5b3a94d10b899870f39606a Mon Sep 17 00:00:00 2001 From: Pierre Le Magourou Date: Mon, 18 Jul 2016 23:22:19 +0300 Subject: ath6kl: Fix WLAN tethering authentication problem. When enabling WLAN tethering, a new AP is visible and a STA could connect to it. When the STA tries to authenticate to the newly created AP, the WPA authentication mechanism is stuck in the 1/4 msg of 4-Way Handshake. In ath6kl_rx(), the ath6kl_find_sta() function is looking for the h_source field of the Ethernet frame header received by the STA. The datap pointer that points to the Ethernet frame header is incorrect, and was pointing at the wrong offset in the buffer. This commit adds a pad_before_data_start offset to set the datap pointer to the Ethernet frame header. datap->h_source parameter is now really pointing to the source ethernet address and the authentication process can continue. Signed-off-by: Pierre Le Magourou Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath6kl/txrx.c b/drivers/net/wireless/ath/ath6kl/txrx.c index 40432fe..9df41d5 100644 --- a/drivers/net/wireless/ath/ath6kl/txrx.c +++ b/drivers/net/wireless/ath/ath6kl/txrx.c @@ -1401,6 +1401,10 @@ void ath6kl_rx(struct htc_target *target, struct htc_packet *packet) return; } + pad_before_data_start = + (le16_to_cpu(dhdr->info3) >> WMI_DATA_HDR_PAD_BEFORE_DATA_SHIFT) + & WMI_DATA_HDR_PAD_BEFORE_DATA_MASK; + /* Get the Power save state of the STA */ if (vif->nw_type == AP_NETWORK) { meta_type = wmi_data_hdr_get_meta(dhdr); @@ -1408,7 +1412,7 @@ void ath6kl_rx(struct htc_target *target, struct htc_packet *packet) ps_state = !!((dhdr->info >> WMI_DATA_HDR_PS_SHIFT) & WMI_DATA_HDR_PS_MASK); - offset = sizeof(struct wmi_data_hdr); + offset = sizeof(struct wmi_data_hdr) + pad_before_data_start; trig_state = !!(le16_to_cpu(dhdr->info3) & WMI_DATA_HDR_TRIG); switch (meta_type) { @@ -1523,9 +1527,6 @@ void ath6kl_rx(struct htc_target *target, struct htc_packet *packet) seq_no = wmi_data_hdr_get_seqno(dhdr); meta_type = wmi_data_hdr_get_meta(dhdr); dot11_hdr = wmi_data_hdr_get_dot11(dhdr); - pad_before_data_start = - (le16_to_cpu(dhdr->info3) >> WMI_DATA_HDR_PAD_BEFORE_DATA_SHIFT) - & WMI_DATA_HDR_PAD_BEFORE_DATA_MASK; skb_pull(skb, sizeof(struct wmi_data_hdr)); -- cgit v0.10.2 From f3651bae564497250bfde375761554d2e6bb0772 Mon Sep 17 00:00:00 2001 From: Pierre Le Magourou Date: Mon, 18 Jul 2016 23:22:19 +0300 Subject: ath6kl: Fix wrong regulatory domain disconnection. One minute after a successful connection, the kernel checks if the frequency and the channel width are well configured for the country we are in. (regulatory domain) ath6kl driver was setting the NL80211_CHAN_HT20 channel without checking for the HT capabilities. (we should have NL80211_CHAN_NO_HT in our case because the firmware did not support HT) This patch adds a check on ht_cap.ht_supported in order to create the channel corresponding to the firmware capabilities. Signed-off-by: Pierre Le Magourou Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.c b/drivers/net/wireless/ath/ath6kl/cfg80211.c index ef5b40e..07dcb82 100644 --- a/drivers/net/wireless/ath/ath6kl/cfg80211.c +++ b/drivers/net/wireless/ath/ath6kl/cfg80211.c @@ -1111,7 +1111,8 @@ void ath6kl_cfg80211_ch_switch_notify(struct ath6kl_vif *vif, int freq, cfg80211_chandef_create(&chandef, ieee80211_get_channel(vif->ar->wiphy, freq), - (mode == WMI_11G_HT20) ? + (mode == WMI_11G_HT20 && + ath6kl_band_2ghz.ht_cap.ht_supported) ? NL80211_CHAN_HT20 : NL80211_CHAN_NO_HT); mutex_lock(&vif->wdev.mtx); -- cgit v0.10.2 From 76b817f6c9a901c4d1e7b9e4e4c7b69ce8330353 Mon Sep 17 00:00:00 2001 From: Pierre Le Magourou Date: Mon, 18 Jul 2016 23:22:20 +0300 Subject: ath6kl: Unset IFF_LOWER_UP flag on AP mode leave. When disabling tethering after having enabled it, the ath6kl driver does not set the IFF_LOWER_UP flag to 0 (carrier off) on the wlan interface. The upper layers (eg. connman) are not notified of the tethering mode status change. So, tethering can not be activated anymore. This patch adds a netif_carrier_off() call when stopping AP mode to fix the problem. Signed-off-by: Pierre Le Magourou Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.c b/drivers/net/wireless/ath/ath6kl/cfg80211.c index 07dcb82..62354ec 100644 --- a/drivers/net/wireless/ath/ath6kl/cfg80211.c +++ b/drivers/net/wireless/ath/ath6kl/cfg80211.c @@ -2979,6 +2979,7 @@ static int ath6kl_stop_ap(struct wiphy *wiphy, struct net_device *dev) ath6kl_wmi_disconnect_cmd(ar->wmi, vif->fw_vif_idx); clear_bit(CONNECTED, &vif->flags); + netif_carrier_off(vif->ndev); /* Restore ht setting in firmware */ return ath6kl_restore_htcap(vif); -- cgit v0.10.2 From 590025a27fe0603e855a054c4ad57d966bd8af07 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sat, 16 Jul 2016 14:27:21 +0800 Subject: netfilter: nft_ct: fix unpaired nf_connlabels_get/put call We only get nf_connlabels if the user add ct label set expr successfully, but we will also put nf_connlabels if the user delete ct lable get expr. This is mismathced, and will cause ct label expr cannot work properly. Also, if we init something fail, we should put nf_connlabels back. Otherwise, we may waste to alloc the memory that will never be used. Signed-off-by: Liping Zhang Acked-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index 7ce8fd7..d9e44ca 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -366,6 +366,7 @@ static int nft_ct_set_init(const struct nft_ctx *ctx, const struct nlattr * const tb[]) { struct nft_ct *priv = nft_expr_priv(expr); + bool label_got = false; unsigned int len; int err; @@ -384,6 +385,7 @@ static int nft_ct_set_init(const struct nft_ctx *ctx, err = nf_connlabels_get(ctx->net, (len * BITS_PER_BYTE) - 1); if (err) return err; + label_got = true; break; #endif default: @@ -393,17 +395,28 @@ static int nft_ct_set_init(const struct nft_ctx *ctx, priv->sreg = nft_parse_register(tb[NFTA_CT_SREG]); err = nft_validate_register_load(priv->sreg, len); if (err < 0) - return err; + goto err1; err = nft_ct_l3proto_try_module_get(ctx->afi->family); if (err < 0) - return err; + goto err1; return 0; + +err1: + if (label_got) + nf_connlabels_put(ctx->net); + return err; +} + +static void nft_ct_get_destroy(const struct nft_ctx *ctx, + const struct nft_expr *expr) +{ + nft_ct_l3proto_module_put(ctx->afi->family); } -static void nft_ct_destroy(const struct nft_ctx *ctx, - const struct nft_expr *expr) +static void nft_ct_set_destroy(const struct nft_ctx *ctx, + const struct nft_expr *expr) { struct nft_ct *priv = nft_expr_priv(expr); @@ -475,7 +488,7 @@ static const struct nft_expr_ops nft_ct_get_ops = { .size = NFT_EXPR_SIZE(sizeof(struct nft_ct)), .eval = nft_ct_get_eval, .init = nft_ct_get_init, - .destroy = nft_ct_destroy, + .destroy = nft_ct_get_destroy, .dump = nft_ct_get_dump, }; @@ -484,7 +497,7 @@ static const struct nft_expr_ops nft_ct_set_ops = { .size = NFT_EXPR_SIZE(sizeof(struct nft_ct)), .eval = nft_ct_set_eval, .init = nft_ct_set_init, - .destroy = nft_ct_destroy, + .destroy = nft_ct_set_destroy, .dump = nft_ct_set_dump, }; -- cgit v0.10.2 From 42e54152e7070f53294fc41d7f12749ba8bf9aab Mon Sep 17 00:00:00 2001 From: Dan Kephart Date: Fri, 8 Jul 2016 21:04:19 -0400 Subject: ath6kl: sme_state shortcut to SME_DISCONNECTED removed When an NL80211_DISCONNECT is sent to cfg80211, the driver's cfg80211 disconnect function sets the sme_state to SME_DISCONNECTED before receiving a WMI_DISCONNECT_EVENT from the firmware. This caused cfg80211 to not know that the connection is disconnected because the driver fails to notify cfg80211 upon receiving WMI_DISCONNECT_EVENT from the firmware believing it is already disconnected. Signed-off-by: Dan Kephart Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath6kl/cfg80211.c b/drivers/net/wireless/ath/ath6kl/cfg80211.c index 62354ec..4ad6284 100644 --- a/drivers/net/wireless/ath/ath6kl/cfg80211.c +++ b/drivers/net/wireless/ath/ath6kl/cfg80211.c @@ -847,8 +847,6 @@ static int ath6kl_cfg80211_disconnect(struct wiphy *wiphy, up(&ar->sem); - vif->sme_state = SME_DISCONNECTED; - return 0; } -- cgit v0.10.2 From 56398519969365351e70de5c1226f32ebaaadbd3 Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Sun, 10 Jul 2016 01:09:55 +0200 Subject: ath9k: simplify the code-paths when not using the built-in EEPROM There were two paths in the code for "external" eeprom sources. The code in eeprom.c only handled the cases where the eeprom data was loaded via request_firmware. ahb.c and pci.c on the other hand had some duplicate code which was only used when the eeprom data was passed via ath9k_platform_data. With this change all eeprom data handling is now unified in eeprom.c. Signed-off-by: Martin Blumenstingl Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/ahb.c b/drivers/net/wireless/ath/ath9k/ahb.c index bd4a1a6..bea6186 100644 --- a/drivers/net/wireless/ath/ath9k/ahb.c +++ b/drivers/net/wireless/ath/ath9k/ahb.c @@ -18,7 +18,6 @@ #include #include -#include #include #include "ath9k.h" @@ -58,20 +57,9 @@ static void ath_ahb_read_cachesize(struct ath_common *common, int *csz) static bool ath_ahb_eeprom_read(struct ath_common *common, u32 off, u16 *data) { - struct ath_softc *sc = (struct ath_softc *)common->priv; - struct platform_device *pdev = to_platform_device(sc->dev); - struct ath9k_platform_data *pdata; - - pdata = dev_get_platdata(&pdev->dev); - if (off >= (ARRAY_SIZE(pdata->eeprom_data))) { - ath_err(common, - "%s: flash read failed, offset %08x is out of range\n", - __func__, off); - return false; - } - - *data = pdata->eeprom_data[off]; - return true; + ath_err(common, "%s: eeprom data has to be provided externally\n", + __func__); + return false; } static struct ath_bus_ops ath_ahb_bus_ops = { diff --git a/drivers/net/wireless/ath/ath9k/eeprom.c b/drivers/net/wireless/ath/ath9k/eeprom.c index a794157..a449588 100644 --- a/drivers/net/wireless/ath/ath9k/eeprom.c +++ b/drivers/net/wireless/ath/ath9k/eeprom.c @@ -15,6 +15,7 @@ */ #include "hw.h" +#include void ath9k_hw_analog_shift_regwrite(struct ath_hw *ah, u32 reg, u32 val) { @@ -108,26 +109,42 @@ void ath9k_hw_usb_gen_fill_eeprom(struct ath_hw *ah, u16 *eep_data, } } -static bool ath9k_hw_nvram_read_blob(struct ath_hw *ah, u32 off, - u16 *data) +static bool ath9k_hw_nvram_read_array(u16 *blob, size_t blob_size, + off_t offset, u16 *data) { - u16 *blob_data; - - if (off * sizeof(u16) > ah->eeprom_blob->size) + if (offset > blob_size) return false; - blob_data = (u16 *)ah->eeprom_blob->data; - *data = blob_data[off]; + *data = blob[offset]; return true; } +static bool ath9k_hw_nvram_read_pdata(struct ath9k_platform_data *pdata, + off_t offset, u16 *data) +{ + return ath9k_hw_nvram_read_array(pdata->eeprom_data, + ARRAY_SIZE(pdata->eeprom_data), + offset, data); +} + +static bool ath9k_hw_nvram_read_firmware(const struct firmware *eeprom_blob, + off_t offset, u16 *data) +{ + return ath9k_hw_nvram_read_array((u16 *) eeprom_blob->data, + eeprom_blob->size / sizeof(u16), + offset, data); +} + bool ath9k_hw_nvram_read(struct ath_hw *ah, u32 off, u16 *data) { struct ath_common *common = ath9k_hw_common(ah); + struct ath9k_platform_data *pdata = ah->dev->platform_data; bool ret; if (ah->eeprom_blob) - ret = ath9k_hw_nvram_read_blob(ah, off, data); + ret = ath9k_hw_nvram_read_firmware(ah->eeprom_blob, off, data); + else if (pdata && !pdata->use_eeprom && pdata->eeprom_data) + ret = ath9k_hw_nvram_read_pdata(pdata, off, data); else ret = common->bus_ops->eeprom_read(common, off, data); diff --git a/drivers/net/wireless/ath/ath9k/pci.c b/drivers/net/wireless/ath/ath9k/pci.c index aa04b13..0dd454a 100644 --- a/drivers/net/wireless/ath/ath9k/pci.c +++ b/drivers/net/wireless/ath/ath9k/pci.c @@ -19,7 +19,6 @@ #include #include #include -#include #include #include "ath9k.h" @@ -786,37 +785,21 @@ static void ath_pci_read_cachesize(struct ath_common *common, int *csz) static bool ath_pci_eeprom_read(struct ath_common *common, u32 off, u16 *data) { - struct ath_softc *sc = (struct ath_softc *) common->priv; - struct ath9k_platform_data *pdata = sc->dev->platform_data; - - if (pdata && !pdata->use_eeprom) { - if (off >= (ARRAY_SIZE(pdata->eeprom_data))) { - ath_err(common, - "%s: eeprom read failed, offset %08x is out of range\n", - __func__, off); - - return false; - } - - *data = pdata->eeprom_data[off]; - } else { - struct ath_hw *ah = (struct ath_hw *) common->ah; - - common->ops->read(ah, AR5416_EEPROM_OFFSET + - (off << AR5416_EEPROM_S)); - - if (!ath9k_hw_wait(ah, - AR_EEPROM_STATUS_DATA, - AR_EEPROM_STATUS_DATA_BUSY | - AR_EEPROM_STATUS_DATA_PROT_ACCESS, 0, - AH_WAIT_TIMEOUT)) { - return false; - } - - *data = MS(common->ops->read(ah, AR_EEPROM_STATUS_DATA), - AR_EEPROM_STATUS_DATA_VAL); + struct ath_hw *ah = (struct ath_hw *) common->ah; + + common->ops->read(ah, AR5416_EEPROM_OFFSET + (off << AR5416_EEPROM_S)); + + if (!ath9k_hw_wait(ah, + AR_EEPROM_STATUS_DATA, + AR_EEPROM_STATUS_DATA_BUSY | + AR_EEPROM_STATUS_DATA_PROT_ACCESS, 0, + AH_WAIT_TIMEOUT)) { + return false; } + *data = MS(common->ops->read(ah, AR_EEPROM_STATUS_DATA), + AR_EEPROM_STATUS_DATA_VAL); + return true; } -- cgit v0.10.2 From 7d6c2d1e34ef9a5675a7a4a8811cecbb2dd273dd Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Mon, 11 Jul 2016 10:35:39 +0200 Subject: ath9k_hw: fix spectral scan on AR9285 and newer The register layout of AR_PHY_SPECTRAL_SCAN has changed, only AR9280 uses the old layout Signed-off-by: Felix Fietkau Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/ar9002_phy.c b/drivers/net/wireless/ath/ath9k/ar9002_phy.c index 53d7445..61a9b85 100644 --- a/drivers/net/wireless/ath/ath9k/ar9002_phy.c +++ b/drivers/net/wireless/ath/ath9k/ar9002_phy.c @@ -476,6 +476,7 @@ static void ar9002_hw_set_bt_ant_diversity(struct ath_hw *ah, bool enable) static void ar9002_hw_spectral_scan_config(struct ath_hw *ah, struct ath_spec_scan *param) { + u32 repeat_bit; u8 count; if (!param->enabled) { @@ -486,12 +487,15 @@ static void ar9002_hw_spectral_scan_config(struct ath_hw *ah, REG_SET_BIT(ah, AR_PHY_RADAR_0, AR_PHY_RADAR_0_FFT_ENA); REG_SET_BIT(ah, AR_PHY_SPECTRAL_SCAN, AR_PHY_SPECTRAL_SCAN_ENABLE); + if (AR_SREV_9280(ah)) + repeat_bit = AR_PHY_SPECTRAL_SCAN_SHORT_REPEAT; + else + repeat_bit = AR_PHY_SPECTRAL_SCAN_SHORT_REPEAT_KIWI; + if (param->short_repeat) - REG_SET_BIT(ah, AR_PHY_SPECTRAL_SCAN, - AR_PHY_SPECTRAL_SCAN_SHORT_REPEAT); + REG_SET_BIT(ah, AR_PHY_SPECTRAL_SCAN, repeat_bit); else - REG_CLR_BIT(ah, AR_PHY_SPECTRAL_SCAN, - AR_PHY_SPECTRAL_SCAN_SHORT_REPEAT); + REG_CLR_BIT(ah, AR_PHY_SPECTRAL_SCAN, repeat_bit); /* on AR92xx, the highest bit of count will make the the chip send * spectral samples endlessly. Check if this really was intended, @@ -499,15 +503,25 @@ static void ar9002_hw_spectral_scan_config(struct ath_hw *ah, */ count = param->count; if (param->endless) { - if (AR_SREV_9271(ah)) - count = 0; - else + if (AR_SREV_9280(ah)) count = 0x80; + else + count = 0; } else if (count & 0x80) count = 0x7f; + else if (!count) + count = 1; + + if (AR_SREV_9280(ah)) { + REG_RMW_FIELD(ah, AR_PHY_SPECTRAL_SCAN, + AR_PHY_SPECTRAL_SCAN_COUNT, count); + } else { + REG_RMW_FIELD(ah, AR_PHY_SPECTRAL_SCAN, + AR_PHY_SPECTRAL_SCAN_COUNT_KIWI, count); + REG_SET_BIT(ah, AR_PHY_SPECTRAL_SCAN, + AR_PHY_SPECTRAL_SCAN_PHYERR_MASK_SELECT); + } - REG_RMW_FIELD(ah, AR_PHY_SPECTRAL_SCAN, - AR_PHY_SPECTRAL_SCAN_COUNT, count); REG_RMW_FIELD(ah, AR_PHY_SPECTRAL_SCAN, AR_PHY_SPECTRAL_SCAN_PERIOD, param->period); REG_RMW_FIELD(ah, AR_PHY_SPECTRAL_SCAN, diff --git a/drivers/net/wireless/ath/ath9k/ar9002_phy.h b/drivers/net/wireless/ath/ath9k/ar9002_phy.h index 9d17a53..2b58245 100644 --- a/drivers/net/wireless/ath/ath9k/ar9002_phy.h +++ b/drivers/net/wireless/ath/ath9k/ar9002_phy.h @@ -177,8 +177,11 @@ #define AR_PHY_SPECTRAL_SCAN_PERIOD_S 8 #define AR_PHY_SPECTRAL_SCAN_COUNT 0x00FF0000 /* Number of reports, reg 68, bits 16-23*/ #define AR_PHY_SPECTRAL_SCAN_COUNT_S 16 +#define AR_PHY_SPECTRAL_SCAN_COUNT_KIWI 0x0FFF0000 /* Number of reports, reg 68, bits 16-27*/ +#define AR_PHY_SPECTRAL_SCAN_COUNT_KIWI_S 16 #define AR_PHY_SPECTRAL_SCAN_SHORT_REPEAT 0x01000000 /* Short repeat, reg 68, bit 24*/ -#define AR_PHY_SPECTRAL_SCAN_SHORT_REPEAT_S 24 /* Short repeat, reg 68, bit 24*/ +#define AR_PHY_SPECTRAL_SCAN_SHORT_REPEAT_KIWI 0x10000000 /* Short repeat, reg 68, bit 28*/ +#define AR_PHY_SPECTRAL_SCAN_PHYERR_MASK_SELECT 0x40000000 #define AR_PHY_RX_DELAY 0x9914 #define AR_PHY_SEARCH_START_DELAY 0x9918 -- cgit v0.10.2 From 3ff25093cbdd04add5c23f1f71f6a2a4ea6b9ec0 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Mon, 11 Jul 2016 12:02:45 +0200 Subject: ath9k_hw: fix duplicate (and partially wrong) definition of AR_CH0_THERM AR_PHY_65NM_CH0_THERM and AR_CH0_THERM were supposed to refer to the same register, however they had different SREV checks. Remove the duplicate and use the checks. Since there were other SREV checks present in the only place that uses this, this will probaby not affect runtime behavior. Signed-off-by: Felix Fietkau Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/ar9003_phy.h b/drivers/net/wireless/ath/ath9k/ar9003_phy.h index 566da78..a171dbb 100644 --- a/drivers/net/wireless/ath/ath9k/ar9003_phy.h +++ b/drivers/net/wireless/ath/ath9k/ar9003_phy.h @@ -689,13 +689,6 @@ #define AR_CH0_TOP_XPABIASLVL (AR_SREV_9550(ah) ? 0x3c0 : 0x300) #define AR_CH0_TOP_XPABIASLVL_S (AR_SREV_9550(ah) ? 6 : 8) -#define AR_CH0_THERM (AR_SREV_9300(ah) ? 0x16290 : \ - ((AR_SREV_9485(ah) ? 0x1628c : 0x16294))) -#define AR_CH0_THERM_XPABIASLVL_MSB 0x3 -#define AR_CH0_THERM_XPABIASLVL_MSB_S 0 -#define AR_CH0_THERM_XPASHORT2GND 0x4 -#define AR_CH0_THERM_XPASHORT2GND_S 2 - #define AR_SWITCH_TABLE_COM_ALL (0xffff) #define AR_SWITCH_TABLE_COM_ALL_S (0) #define AR_SWITCH_TABLE_COM_AR9462_ALL (0xffffff) @@ -712,15 +705,17 @@ #define AR_SWITCH_TABLE_ALL (0xfff) #define AR_SWITCH_TABLE_ALL_S (0) -#define AR_PHY_65NM_CH0_THERM (AR_SREV_9300(ah) ? 0x16290 :\ - ((AR_SREV_9462(ah) || AR_SREV_9565(ah)) ? 0x16294 : 0x1628c)) +#define AR_CH0_THERM (AR_SREV_9300(ah) ? 0x16290 :\ + ((AR_SREV_9462(ah) || AR_SREV_9565(ah)) ? 0x16294 : 0x1628c)) +#define AR_CH0_THERM_XPABIASLVL_MSB 0x3 +#define AR_CH0_THERM_XPABIASLVL_MSB_S 0 +#define AR_CH0_THERM_XPASHORT2GND 0x4 +#define AR_CH0_THERM_XPASHORT2GND_S 2 -#define AR_PHY_65NM_CH0_THERM_LOCAL 0x80000000 -#define AR_PHY_65NM_CH0_THERM_LOCAL_S 31 -#define AR_PHY_65NM_CH0_THERM_START 0x20000000 -#define AR_PHY_65NM_CH0_THERM_START_S 29 -#define AR_PHY_65NM_CH0_THERM_SAR_ADC_OUT 0x0000ff00 -#define AR_PHY_65NM_CH0_THERM_SAR_ADC_OUT_S 8 +#define AR_CH0_THERM_LOCAL 0x80000000 +#define AR_CH0_THERM_START 0x20000000 +#define AR_CH0_THERM_SAR_ADC_OUT 0x0000ff00 +#define AR_CH0_THERM_SAR_ADC_OUT_S 8 #define AR_CH0_TOP2 (AR_SREV_9300(ah) ? 0x1628c : \ (AR_SREV_9462(ah) ? 0x16290 : 0x16284)) -- cgit v0.10.2 From 8f778c72ac8f2664c065623fde0d01cbcf99205c Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Mon, 11 Jul 2016 12:02:46 +0200 Subject: ath9k_hw: simplify ar9003_hw_per_calibration Reduce indentation, use a variable to save a few pointer dereferences Signed-off-by: Felix Fietkau Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/ar9003_calib.c b/drivers/net/wireless/ath/ath9k/ar9003_calib.c index 518e649..b9ef3d3 100644 --- a/drivers/net/wireless/ath/ath9k/ar9003_calib.c +++ b/drivers/net/wireless/ath/ath9k/ar9003_calib.c @@ -75,50 +75,49 @@ static bool ar9003_hw_per_calibration(struct ath_hw *ah, struct ath9k_cal_list *currCal) { struct ath9k_hw_cal_data *caldata = ah->caldata; - /* Cal is assumed not done until explicitly set below */ - bool iscaldone = false; + const struct ath9k_percal_data *cur_caldata = currCal->calData; /* Calibration in progress. */ if (currCal->calState == CAL_RUNNING) { /* Check to see if it has finished. */ - if (!(REG_READ(ah, AR_PHY_TIMING4) & AR_PHY_TIMING4_DO_CAL)) { + if (REG_READ(ah, AR_PHY_TIMING4) & AR_PHY_TIMING4_DO_CAL) + return false; + + /* + * Accumulate cal measures for active chains + */ + cur_caldata->calCollect(ah); + ah->cal_samples++; + + if (ah->cal_samples >= cur_caldata->calNumSamples) { + unsigned int i, numChains = 0; + for (i = 0; i < AR9300_MAX_CHAINS; i++) { + if (rxchainmask & (1 << i)) + numChains++; + } + /* - * Accumulate cal measures for active chains + * Process accumulated data */ - currCal->calData->calCollect(ah); - ah->cal_samples++; - - if (ah->cal_samples >= - currCal->calData->calNumSamples) { - unsigned int i, numChains = 0; - for (i = 0; i < AR9300_MAX_CHAINS; i++) { - if (rxchainmask & (1 << i)) - numChains++; - } + cur_caldata->calPostProc(ah, numChains); - /* - * Process accumulated data - */ - currCal->calData->calPostProc(ah, numChains); - - /* Calibration has finished. */ - caldata->CalValid |= currCal->calData->calType; - currCal->calState = CAL_DONE; - iscaldone = true; - } else { + /* Calibration has finished. */ + caldata->CalValid |= cur_caldata->calType; + currCal->calState = CAL_DONE; + return true; + } else { /* * Set-up collection of another sub-sample until we * get desired number */ ar9003_hw_setup_calibration(ah, currCal); - } } - } else if (!(caldata->CalValid & currCal->calData->calType)) { + } else if (!(caldata->CalValid & cur_caldata->calType)) { /* If current cal is marked invalid in channel, kick it off */ ath9k_hw_reset_calibration(ah, currCal); } - return iscaldone; + return false; } static int ar9003_hw_calibrate(struct ath_hw *ah, struct ath9k_channel *chan, -- cgit v0.10.2 From feaacb1748927c831773a159feb9e5edc1002cb4 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Mon, 11 Jul 2016 12:02:47 +0200 Subject: ath9k_hw: get rid of some duplicate code in calibration init Remove a misleading debug message as well Signed-off-by: Felix Fietkau Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/ar9003_calib.c b/drivers/net/wireless/ath/ath9k/ar9003_calib.c index b9ef3d3..7e27a06 100644 --- a/drivers/net/wireless/ath/ath9k/ar9003_calib.c +++ b/drivers/net/wireless/ath/ath9k/ar9003_calib.c @@ -1373,6 +1373,26 @@ static void ar9003_hw_cl_cal_post_proc(struct ath_hw *ah, bool is_reusable) } } +static void ar9003_hw_init_cal_common(struct ath_hw *ah) +{ + struct ath9k_hw_cal_data *caldata = ah->caldata; + + /* Initialize list pointers */ + ah->cal_list = ah->cal_list_last = ah->cal_list_curr = NULL; + + INIT_CAL(&ah->iq_caldata); + INSERT_CAL(ah, &ah->iq_caldata); + + /* Initialize current pointer to first element in list */ + ah->cal_list_curr = ah->cal_list; + + if (ah->cal_list_curr) + ath9k_hw_reset_calibration(ah, ah->cal_list_curr); + + if (caldata) + caldata->CalValid = 0; +} + static bool ar9003_hw_init_cal_pcoem(struct ath_hw *ah, struct ath9k_channel *chan) { @@ -1532,21 +1552,7 @@ skip_tx_iqcal: /* Revert chainmask to runtime parameters */ ar9003_hw_set_chain_masks(ah, ah->rxchainmask, ah->txchainmask); - /* Initialize list pointers */ - ah->cal_list = ah->cal_list_last = ah->cal_list_curr = NULL; - - INIT_CAL(&ah->iq_caldata); - INSERT_CAL(ah, &ah->iq_caldata); - ath_dbg(common, CALIBRATE, "enabling IQ Calibration\n"); - - /* Initialize current pointer to first element in list */ - ah->cal_list_curr = ah->cal_list; - - if (ah->cal_list_curr) - ath9k_hw_reset_calibration(ah, ah->cal_list_curr); - - if (caldata) - caldata->CalValid = 0; + ar9003_hw_init_cal_common(ah); return true; } @@ -1577,8 +1583,6 @@ static bool do_ar9003_agc_cal(struct ath_hw *ah) static bool ar9003_hw_init_cal_soc(struct ath_hw *ah, struct ath9k_channel *chan) { - struct ath_common *common = ath9k_hw_common(ah); - struct ath9k_hw_cal_data *caldata = ah->caldata; bool txiqcal_done = false; bool status = true; bool run_agc_cal = false, sep_iq_cal = false; @@ -1676,21 +1680,7 @@ skip_tx_iqcal: /* Revert chainmask to runtime parameters */ ar9003_hw_set_chain_masks(ah, ah->rxchainmask, ah->txchainmask); - /* Initialize list pointers */ - ah->cal_list = ah->cal_list_last = ah->cal_list_curr = NULL; - - INIT_CAL(&ah->iq_caldata); - INSERT_CAL(ah, &ah->iq_caldata); - ath_dbg(common, CALIBRATE, "enabling IQ Calibration\n"); - - /* Initialize current pointer to first element in list */ - ah->cal_list_curr = ah->cal_list; - - if (ah->cal_list_curr) - ath9k_hw_reset_calibration(ah, ah->cal_list_curr); - - if (caldata) - caldata->CalValid = 0; + ar9003_hw_init_cal_common(ah); return true; } -- cgit v0.10.2 From 171f6402e4aa5cd3b8407f82501f7ea21fa54ccc Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Mon, 11 Jul 2016 12:02:48 +0200 Subject: ath9k_hw: implement temperature compensation support for AR9003+ Signed-off-by: Felix Fietkau Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ath/ath9k/ar9003_calib.c b/drivers/net/wireless/ath/ath9k/ar9003_calib.c index 7e27a06..b6f064a 100644 --- a/drivers/net/wireless/ath/ath9k/ar9003_calib.c +++ b/drivers/net/wireless/ath/ath9k/ar9003_calib.c @@ -33,6 +33,7 @@ struct coeff { enum ar9003_cal_types { IQ_MISMATCH_CAL = BIT(0), + TEMP_COMP_CAL = BIT(1), }; static void ar9003_hw_setup_calibration(struct ath_hw *ah, @@ -58,6 +59,12 @@ static void ar9003_hw_setup_calibration(struct ath_hw *ah, /* Kick-off cal */ REG_SET_BIT(ah, AR_PHY_TIMING4, AR_PHY_TIMING4_DO_CAL); break; + case TEMP_COMP_CAL: + ath_dbg(common, CALIBRATE, + "starting Temperature Compensation Calibration\n"); + REG_SET_BIT(ah, AR_CH0_THERM, AR_CH0_THERM_LOCAL); + REG_SET_BIT(ah, AR_CH0_THERM, AR_CH0_THERM_START); + break; default: ath_err(common, "Invalid calibration type\n"); break; @@ -86,7 +93,8 @@ static bool ar9003_hw_per_calibration(struct ath_hw *ah, /* * Accumulate cal measures for active chains */ - cur_caldata->calCollect(ah); + if (cur_caldata->calCollect) + cur_caldata->calCollect(ah); ah->cal_samples++; if (ah->cal_samples >= cur_caldata->calNumSamples) { @@ -99,7 +107,8 @@ static bool ar9003_hw_per_calibration(struct ath_hw *ah, /* * Process accumulated data */ - cur_caldata->calPostProc(ah, numChains); + if (cur_caldata->calPostProc) + cur_caldata->calPostProc(ah, numChains); /* Calibration has finished. */ caldata->CalValid |= cur_caldata->calType; @@ -314,9 +323,16 @@ static const struct ath9k_percal_data iq_cal_single_sample = { ar9003_hw_iqcalibrate }; +static const struct ath9k_percal_data temp_cal_single_sample = { + TEMP_COMP_CAL, + MIN_CAL_SAMPLES, + PER_MAX_LOG_COUNT, +}; + static void ar9003_hw_init_cal_settings(struct ath_hw *ah) { ah->iq_caldata.calData = &iq_cal_single_sample; + ah->temp_caldata.calData = &temp_cal_single_sample; if (AR_SREV_9300_20_OR_LATER(ah)) { ah->enabled_cals |= TX_IQ_CAL; @@ -324,7 +340,7 @@ static void ar9003_hw_init_cal_settings(struct ath_hw *ah) ah->enabled_cals |= TX_IQ_ON_AGC_CAL; } - ah->supp_cals = IQ_MISMATCH_CAL; + ah->supp_cals = IQ_MISMATCH_CAL | TEMP_COMP_CAL; } #define OFF_UPPER_LT 24 @@ -1383,6 +1399,9 @@ static void ar9003_hw_init_cal_common(struct ath_hw *ah) INIT_CAL(&ah->iq_caldata); INSERT_CAL(ah, &ah->iq_caldata); + INIT_CAL(&ah->temp_caldata); + INSERT_CAL(ah, &ah->temp_caldata); + /* Initialize current pointer to first element in list */ ah->cal_list_curr = ah->cal_list; diff --git a/drivers/net/wireless/ath/ath9k/hw.h b/drivers/net/wireless/ath/ath9k/hw.h index 9cbca12..2a5d3ad 100644 --- a/drivers/net/wireless/ath/ath9k/hw.h +++ b/drivers/net/wireless/ath/ath9k/hw.h @@ -830,6 +830,7 @@ struct ath_hw { /* Calibration */ u32 supp_cals; struct ath9k_cal_list iq_caldata; + struct ath9k_cal_list temp_caldata; struct ath9k_cal_list adcgain_caldata; struct ath9k_cal_list adcdc_caldata; struct ath9k_cal_list *cal_list; -- cgit v0.10.2 From cc2d1de06f0572a51437d1f31633d81afea5eb47 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Fri, 8 Jul 2016 17:14:18 +0200 Subject: bcma: define ChipCommon B MII registers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We don't have access to datasheets to document all the bits but we can name these registers at least. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo diff --git a/drivers/bcma/driver_chipcommon_b.c b/drivers/bcma/driver_chipcommon_b.c index c20b5f4..57f10b5 100644 --- a/drivers/bcma/driver_chipcommon_b.c +++ b/drivers/bcma/driver_chipcommon_b.c @@ -33,11 +33,12 @@ static bool bcma_wait_reg(struct bcma_bus *bus, void __iomem *addr, u32 mask, void bcma_chipco_b_mii_write(struct bcma_drv_cc_b *ccb, u32 offset, u32 value) { struct bcma_bus *bus = ccb->core->bus; + void __iomem *mii = ccb->mii; - writel(offset, ccb->mii + 0x00); - bcma_wait_reg(bus, ccb->mii + 0x00, 0x0100, 0x0000, 100); - writel(value, ccb->mii + 0x04); - bcma_wait_reg(bus, ccb->mii + 0x00, 0x0100, 0x0000, 100); + writel(offset, mii + BCMA_CCB_MII_MNG_CTL); + bcma_wait_reg(bus, mii + BCMA_CCB_MII_MNG_CTL, 0x0100, 0x0000, 100); + writel(value, mii + BCMA_CCB_MII_MNG_CMD_DATA); + bcma_wait_reg(bus, mii + BCMA_CCB_MII_MNG_CTL, 0x0100, 0x0000, 100); } EXPORT_SYMBOL_GPL(bcma_chipco_b_mii_write); diff --git a/include/linux/bcma/bcma_driver_chipcommon.h b/include/linux/bcma/bcma_driver_chipcommon.h index a5ac2ca..b20e3d5 100644 --- a/include/linux/bcma/bcma_driver_chipcommon.h +++ b/include/linux/bcma/bcma_driver_chipcommon.h @@ -504,6 +504,9 @@ #define BCMA_CC_PMU1_PLL0_PC2_NDIV_INT_MASK 0x1ff00000 #define BCMA_CC_PMU1_PLL0_PC2_NDIV_INT_SHIFT 20 +#define BCMA_CCB_MII_MNG_CTL 0x0000 +#define BCMA_CCB_MII_MNG_CMD_DATA 0x0004 + /* BCM4331 ChipControl numbers. */ #define BCMA_CHIPCTL_4331_BT_COEXIST BIT(0) /* 0 disable */ #define BCMA_CHIPCTL_4331_SECI BIT(1) /* 0 SECI is disabled (JATG functional) */ -- cgit v0.10.2 From 3bdae810721b33061d2e541bd78a70f86ca42af3 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 18 Jul 2016 16:24:34 -0700 Subject: brcmfmac: Fix glob_skb leak in brcmf_sdiod_recv_chain In case brcmf_sdiod_recv_chain() cannot complete a succeful call to brcmf_sdiod_buffrw, we would be leaking glom_skb and not free it as we should, fix this. Reported-by: coverity (CID 1164856) Fixes: a413e39a38573 ("brcmfmac: fix brcmf_sdcard_recv_chain() for host without sg support") Signed-off-by: Florian Fainelli Acked-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c index c4b89d2..f549c25 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c @@ -726,8 +726,10 @@ int brcmf_sdiod_recv_chain(struct brcmf_sdio_dev *sdiodev, return -ENOMEM; err = brcmf_sdiod_buffrw(sdiodev, SDIO_FUNC_2, false, addr, glom_skb); - if (err) + if (err) { + brcmu_pkt_buf_free_skb(glom_skb); goto done; + } skb_queue_walk(pktq, skb) { memcpy(skb->data, glom_skb->data, skb->len); -- cgit v0.10.2 From 5c5fa1f464ac954982df1d96b9f9a5103d21aedd Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 18 Jul 2016 16:24:35 -0700 Subject: brcmsmac: Free packet if dma_mapping_error() fails in dma_rxfill In case dma_mapping_error() returns an error in dma_rxfill, we would be leaking a packet that we allocated with brcmu_pkt_buf_get_skb(). Reported-by: coverity (CID 1081819) Fixes: 67d0cf50bd32 ("brcmsmac: Fix WARNING caused by lack of calls to dma_mapping_error()") Signed-off-by: Florian Fainelli Acked-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/dma.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/dma.c index 796f5f9..b7df576 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/dma.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/dma.c @@ -1079,8 +1079,10 @@ bool dma_rxfill(struct dma_pub *pub) pa = dma_map_single(di->dmadev, p->data, di->rxbufsize, DMA_FROM_DEVICE); - if (dma_mapping_error(di->dmadev, pa)) + if (dma_mapping_error(di->dmadev, pa)) { + brcmu_pkt_buf_free_skb(p); return false; + } /* save the free packet pointer */ di->rxp[rxout] = p; -- cgit v0.10.2 From f823a2aa8f4674c095a5413b9e3ba12d82df06f2 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 18 Jul 2016 16:24:37 -0700 Subject: brcmsmac: Initialize power in brcms_c_stf_ss_algo_channel_get() wlc_phy_txpower_get_current() does a logical OR of power->flags, which presumes that power.flags was initiliazed earlier by the caller, unfortunately, this is not the case, so make sure we zero out the struct tx_power before calling into wlc_phy_txpower_get_current(). Reported-by: coverity (CID 146011) Fixes: 5b435de0d7868 ("net: wireless: add brcm80211 drivers") Signed-off-by: Florian Fainelli Acked-by: Arend van Spriel Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/stf.c b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/stf.c index dd91627..0ab865d 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmsmac/stf.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmsmac/stf.c @@ -87,7 +87,7 @@ void brcms_c_stf_ss_algo_channel_get(struct brcms_c_info *wlc, u16 *ss_algo_channel, u16 chanspec) { - struct tx_power power; + struct tx_power power = { }; u8 siso_mcs_id, cdd_mcs_id, stbc_mcs_id; /* Clear previous settings */ -- cgit v0.10.2 From 01efe65aba6585c78c6fb90d63c3d10773b6f868 Mon Sep 17 00:00:00 2001 From: Eyal Reizer Date: Tue, 19 Jul 2016 09:58:02 +0300 Subject: wlcore: spi: add wl18xx support Add support for using with both wl12xx and wl18xx. - all wilink family needs special init command for entering wspi mode. extra clock cycles should be sent after the spi init command while the cs pin is high. - Use inverted chip select for sending a dummy 4 bytes command that completes the init stage. Signed-off-by: Eyal Reizer Acked-by: Rob Herring Signed-off-by: Kalle Valo diff --git a/Documentation/devicetree/bindings/net/wireless/ti,wlcore,spi.txt b/Documentation/devicetree/bindings/net/wireless/ti,wlcore,spi.txt index 9180724..8f9ced0 100644 --- a/Documentation/devicetree/bindings/net/wireless/ti,wlcore,spi.txt +++ b/Documentation/devicetree/bindings/net/wireless/ti,wlcore,spi.txt @@ -1,19 +1,30 @@ -* Texas Instruments wl1271 wireless lan controller +* Texas Instruments wl12xx/wl18xx wireless lan controller -The wl1271 chip can be connected via SPI or via SDIO. This +The wl12xx/wl18xx chips can be connected via SPI or via SDIO. This document describes the binding for the SPI connected chip. Required properties: -- compatible : Should be "ti,wl1271" +- compatible : Should be one of the following: + * "ti,wl1271" + * "ti,wl1273" + * "ti,wl1281" + * "ti,wl1283" + * "ti,wl1801" + * "ti,wl1805" + * "ti,wl1807" + * "ti,wl1831" + * "ti,wl1835" + * "ti,wl1837" - reg : Chip select address of device - spi-max-frequency : Maximum SPI clocking speed of device in Hz -- ref-clock-frequency : Reference clock frequency - interrupt-parent, interrupts : Should contain parameters for 1 interrupt line. Interrupt parameters: parent, line number, type. -- vwlan-supply : Point the node of the regulator that powers/enable the wl1271 chip +- vwlan-supply : Point the node of the regulator that powers/enable the + wl12xx/wl18xx chip Optional properties: +- ref-clock-frequency : Reference clock frequency (should be set for wl12xx) - clock-xtal : boolean, clock is generated from XTAL - Please consult Documentation/devicetree/bindings/spi/spi-bus.txt @@ -21,16 +32,28 @@ Optional properties: Examples: +For wl12xx family: &spi1 { - wl1271@1 { + wlcore: wlcore@1 { compatible = "ti,wl1271"; - reg = <1>; spi-max-frequency = <48000000>; - clock-xtal; - ref-clock-frequency = <38400000>; interrupt-parent = <&gpio3>; interrupts = <8 IRQ_TYPE_LEVEL_HIGH>; vwlan-supply = <&vwlan_fixed>; + clock-xtal; + ref-clock-frequency = <38400000>; + }; +}; + +For wl18xx family: +&spi0 { + wlcore: wlcore@0 { + compatible = "ti,wl1835"; + reg = <0>; + spi-max-frequency = <48000000>; + interrupt-parent = <&gpio0>; + interrupts = <27 IRQ_TYPE_EDGE_RISING>; + vwlan-supply = <&vwlan_fixed>; }; }; diff --git a/drivers/net/wireless/ti/wlcore/spi.c b/drivers/net/wireless/ti/wlcore/spi.c index cea9443..73fbcf1 100644 --- a/drivers/net/wireless/ti/wlcore/spi.c +++ b/drivers/net/wireless/ti/wlcore/spi.c @@ -70,16 +70,30 @@ #define WSPI_MAX_CHUNK_SIZE 4092 /* - * only support SPI for 12xx - this code should be reworked when 18xx - * support is introduced + * wl18xx driver aggregation buffer size is (13 * PAGE_SIZE) compared to + * (4 * PAGE_SIZE) for wl12xx, so use the larger buffer needed for wl18xx */ -#define SPI_AGGR_BUFFER_SIZE (4 * PAGE_SIZE) +#define SPI_AGGR_BUFFER_SIZE (13 * PAGE_SIZE) /* Maximum number of SPI write chunks */ #define WSPI_MAX_NUM_OF_CHUNKS \ ((SPI_AGGR_BUFFER_SIZE / WSPI_MAX_CHUNK_SIZE) + 1) +struct wilink_familiy_data { + char name[8]; +}; + +const struct wilink_familiy_data *wilink_data; + +static const struct wilink_familiy_data wl18xx_data = { + .name = "wl18xx", +}; + +static const struct wilink_familiy_data wl12xx_data = { + .name = "wl12xx", +}; + struct wl12xx_spi_glue { struct device *dev; struct platform_device *core; @@ -119,6 +133,7 @@ static void wl12xx_spi_init(struct device *child) struct wl12xx_spi_glue *glue = dev_get_drvdata(child->parent); struct spi_transfer t; struct spi_message m; + struct spi_device *spi = to_spi_device(glue->dev); u8 *cmd = kzalloc(WSPI_INIT_CMD_LEN, GFP_KERNEL); if (!cmd) { @@ -151,6 +166,7 @@ static void wl12xx_spi_init(struct device *child) cmd[6] |= WSPI_INIT_CMD_EN_FIXEDBUSY; cmd[7] = crc7_be(0, cmd+2, WSPI_INIT_CMD_CRC_LEN) | WSPI_INIT_CMD_END; + /* * The above is the logical order; it must actually be stored * in the buffer byte-swapped. @@ -163,6 +179,28 @@ static void wl12xx_spi_init(struct device *child) spi_message_add_tail(&t, &m); spi_sync(to_spi_device(glue->dev), &m); + + /* Send extra clocks with inverted CS (high). this is required + * by the wilink family in order to successfully enter WSPI mode. + */ + spi->mode ^= SPI_CS_HIGH; + memset(&m, 0, sizeof(m)); + spi_message_init(&m); + + cmd[0] = 0xff; + cmd[1] = 0xff; + cmd[2] = 0xff; + cmd[3] = 0xff; + __swab32s((u32 *)cmd); + + t.tx_buf = cmd; + t.len = 4; + spi_message_add_tail(&t, &m); + + spi_sync(to_spi_device(glue->dev), &m); + + /* Restore chip select configration to normal */ + spi->mode ^= SPI_CS_HIGH; kfree(cmd); } @@ -270,22 +308,25 @@ static int __must_check wl12xx_spi_raw_read(struct device *child, int addr, return 0; } -static int __must_check wl12xx_spi_raw_write(struct device *child, int addr, - void *buf, size_t len, bool fixed) +static int __wl12xx_spi_raw_write(struct device *child, int addr, + void *buf, size_t len, bool fixed) { struct wl12xx_spi_glue *glue = dev_get_drvdata(child->parent); - /* SPI write buffers - 2 for each chunk */ - struct spi_transfer t[2 * WSPI_MAX_NUM_OF_CHUNKS]; + struct spi_transfer *t; struct spi_message m; u32 commands[WSPI_MAX_NUM_OF_CHUNKS]; /* 1 command per chunk */ u32 *cmd; u32 chunk_len; int i; + /* SPI write buffers - 2 for each chunk */ + t = kzalloc(sizeof(*t) * 2 * WSPI_MAX_NUM_OF_CHUNKS, GFP_KERNEL); + if (!t) + return -ENOMEM; + WARN_ON(len > SPI_AGGR_BUFFER_SIZE); spi_message_init(&m); - memset(t, 0, sizeof(t)); cmd = &commands[0]; i = 0; @@ -318,9 +359,26 @@ static int __must_check wl12xx_spi_raw_write(struct device *child, int addr, spi_sync(to_spi_device(glue->dev), &m); + kfree(t); return 0; } +static int __must_check wl12xx_spi_raw_write(struct device *child, int addr, + void *buf, size_t len, bool fixed) +{ + int ret; + + /* The ELP wakeup write may fail the first time due to internal + * hardware latency. It is safer to send the wakeup command twice to + * avoid unexpected failures. + */ + if (addr == HW_ACCESS_ELP_CTRL_REG) + ret = __wl12xx_spi_raw_write(child, addr, buf, len, fixed); + ret = __wl12xx_spi_raw_write(child, addr, buf, len, fixed); + + return ret; +} + /** * wl12xx_spi_set_power - power on/off the wl12xx unit * @child: wl12xx device handle. @@ -349,17 +407,38 @@ static int wl12xx_spi_set_power(struct device *child, bool enable) return ret; } +/** + * wl12xx_spi_set_block_size + * + * This function is not needed for spi mode, but need to be present. + * Without it defined the wlcore fallback to use the wrong packet + * allignment on tx. + */ +static void wl12xx_spi_set_block_size(struct device *child, + unsigned int blksz) +{ +} + static struct wl1271_if_operations spi_ops = { .read = wl12xx_spi_raw_read, .write = wl12xx_spi_raw_write, .reset = wl12xx_spi_reset, .init = wl12xx_spi_init, .power = wl12xx_spi_set_power, - .set_block_size = NULL, + .set_block_size = wl12xx_spi_set_block_size, }; static const struct of_device_id wlcore_spi_of_match_table[] = { - { .compatible = "ti,wl1271" }, + { .compatible = "ti,wl1271", .data = &wl12xx_data}, + { .compatible = "ti,wl1273", .data = &wl12xx_data}, + { .compatible = "ti,wl1281", .data = &wl12xx_data}, + { .compatible = "ti,wl1283", .data = &wl12xx_data}, + { .compatible = "ti,wl1801", .data = &wl18xx_data}, + { .compatible = "ti,wl1805", .data = &wl18xx_data}, + { .compatible = "ti,wl1807", .data = &wl18xx_data}, + { .compatible = "ti,wl1831", .data = &wl18xx_data}, + { .compatible = "ti,wl1835", .data = &wl18xx_data}, + { .compatible = "ti,wl1837", .data = &wl18xx_data}, { } }; MODULE_DEVICE_TABLE(of, wlcore_spi_of_match_table); @@ -376,17 +455,24 @@ static int wlcore_probe_of(struct spi_device *spi, struct wl12xx_spi_glue *glue, { struct device_node *dt_node = spi->dev.of_node; int ret; + const struct of_device_id *of_id; + + of_id = of_match_node(wlcore_spi_of_match_table, dt_node); + if (!of_id) + return -ENODEV; + + wilink_data = of_id->data; + dev_info(&spi->dev, "selected chip familiy is %s\n", + wilink_data->name); if (of_find_property(dt_node, "clock-xtal", NULL)) pdev_data->ref_clock_xtal = true; - ret = of_property_read_u32(dt_node, "ref-clock-frequency", - &pdev_data->ref_clock_freq); - if (ret) { - dev_err(glue->dev, - "can't get reference clock frequency (%d)\n", ret); - return ret; - } + /* optional clock frequency params */ + of_property_read_u32(dt_node, "ref-clock-frequency", + &pdev_data->ref_clock_freq); + of_property_read_u32(dt_node, "tcxo-clock-frequency", + &pdev_data->tcxo_clock_freq); return 0; } @@ -437,7 +523,8 @@ static int wl1271_probe(struct spi_device *spi) return ret; } - glue->core = platform_device_alloc("wl12xx", PLATFORM_DEVID_AUTO); + glue->core = platform_device_alloc(wilink_data->name, + PLATFORM_DEVID_AUTO); if (!glue->core) { dev_err(glue->dev, "can't allocate platform_device\n"); return -ENOMEM; -- cgit v0.10.2 From efacc699139e13f9d3ed8b47df92acb88ff8479f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= Date: Tue, 19 Jul 2016 09:08:32 +0200 Subject: mtd: add arch dependency for MTD_BCM47XXSFLASH symbol MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We dropped strict MIPS dependency for bcm47xxsflash driver in: commit 5651d6aaf489 ("mtd: bcm47xxsflash: use ioremap_cache() instead of KSEG0ADDR()") but using ioremap_cache still limits building it to few selected architectures only. A recent commit 57d8f7dd2132 ("bcma: allow enabling serial flash support on non-MIPS SoCs") automatically dropped MIPS dependency for MTD_BCM47XXSFLASH which broke building e.g. on powerpc and cris. The bcma change is alright as it doesn't break building bcma code in any way. MTD_BCM47XXSFLASH on the other hand should be limited to archs which need it and can build it (by providing ioremap_cache). Fixes: 57d8f7dd2132 ("bcma: allow enabling serial flash support on non-MIPS SoCs") Signed-off-by: Rafał Miłecki Cc: Brian Norris Acked-by: Brian Norris Signed-off-by: Kalle Valo diff --git a/drivers/mtd/devices/Kconfig b/drivers/mtd/devices/Kconfig index f73c416..64a2485 100644 --- a/drivers/mtd/devices/Kconfig +++ b/drivers/mtd/devices/Kconfig @@ -114,7 +114,7 @@ config MTD_SST25L config MTD_BCM47XXSFLASH tristate "R/O support for serial flash on BCMA bus" - depends on BCMA_SFLASH + depends on BCMA_SFLASH && (MIPS || ARM) help BCMA bus can have various flash memories attached, they are registered by bcma as platform devices. This enables driver for -- cgit v0.10.2 From 944c07a7aac95fb3e85d8d250cae6491197f63bb Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 19 Jul 2016 17:33:00 +0200 Subject: rtlwifi: don't add include path for rtl8188ee For rtl8188ee, we pass -Idrivers/net/wireless/rtlwifi/ to gcc, however that directy no longer exists, so evidently this option is no longer required here and can be removed to avoid a warning when building with 'make W=1' or 'gcc -Wmissing-include-dirs' Signed-off-by: Arnd Bergmann Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/Makefile b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/Makefile index a85419a..676e7de 100644 --- a/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/Makefile +++ b/drivers/net/wireless/realtek/rtlwifi/rtl8188ee/Makefile @@ -12,4 +12,4 @@ rtl8188ee-objs := \ obj-$(CONFIG_RTL8188EE) += rtl8188ee.o -ccflags-y += -Idrivers/net/wireless/rtlwifi -D__CHECK_ENDIAN__ +ccflags-y += -D__CHECK_ENDIAN__ -- cgit v0.10.2 From 731e6f007aa2d07fc1a420bffeb7c9e8226e89e6 Mon Sep 17 00:00:00 2001 From: Christophe Jaillet Date: Sun, 17 Jul 2016 08:15:50 +0200 Subject: net: ti: cpmac: Use the correct function to free some resources. In 'cpmac_open', 'dma_alloc_coherent' has been used to allocate some resources, so we need to free them using 'dma_free_coherent' instead of 'kfree'. Also, we don't need to free these resources if the allocation has failed. So I have slighly modified the goto label in this case. Signed-off-by: Christophe JAILLET Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpmac.c b/drivers/net/ethernet/ti/cpmac.c index f86497c..29f381b 100644 --- a/drivers/net/ethernet/ti/cpmac.c +++ b/drivers/net/ethernet/ti/cpmac.c @@ -1006,8 +1006,10 @@ fail_desc: kfree_skb(priv->rx_head[i].skb); } } + dma_free_coherent(&dev->dev, sizeof(struct cpmac_desc) * size, + priv->desc_ring, priv->dma_ring); + fail_alloc: - kfree(priv->desc_ring); iounmap(priv->regs); fail_remap: -- cgit v0.10.2 From eab814661945656d35d4a0e5870e932cd7139977 Mon Sep 17 00:00:00 2001 From: Christophe Jaillet Date: Sun, 17 Jul 2016 09:03:24 +0200 Subject: drivers: atm: nicstar: Use the correct function to free some resources In 'get_scq', 'dma_alloc_coherent' has been used to allocate some resources, so we need to free them using 'dma_free_coherent' instead of 'kfree'. Signed-off-by: Christophe JAILLET Signed-off-by: David S. Miller diff --git a/drivers/atm/nicstar.c b/drivers/atm/nicstar.c index ddc4ceb..700ed15 100644 --- a/drivers/atm/nicstar.c +++ b/drivers/atm/nicstar.c @@ -874,7 +874,8 @@ static scq_info *get_scq(ns_dev *card, int size, u32 scd) scq->skb = kmalloc(sizeof(struct sk_buff *) * (size / NS_SCQE_SIZE), GFP_KERNEL); if (!scq->skb) { - kfree(scq->org); + dma_free_coherent(&card->pcidev->dev, + 2 * size, scq->org, scq->dma); kfree(scq); return NULL; } -- cgit v0.10.2 From f9ac6273e5b8fa45e7cd2086e1bbc91af9af7f19 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Tue, 19 Jul 2016 11:58:16 +0200 Subject: NFC: nfcsim: Add support for sysfs control entry The idea is to have a way to control and/or modify the behavior of the nfcsim virtual devices. This patch creates a folder tree in the debug filesystem. The debugfs is usually mounted into /sys/kernel/debug and the nfcsim entries are located in DEBUGFS/nfcsim/nfcX/ where X is either 0 or 1 depending on the device you want to address. These folders are empty for now and control entries will be added by upcoming commits. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/nfcsim.c b/drivers/nfc/nfcsim.c index 40b4846..97067a5 100644 --- a/drivers/nfc/nfcsim.c +++ b/drivers/nfc/nfcsim.c @@ -16,6 +16,8 @@ #include #include #include +#include +#include #include #include #include @@ -329,6 +331,49 @@ static struct nfc_digital_ops nfcsim_digital_ops = { .switch_rf = nfcsim_switch_rf, }; +static struct dentry *nfcsim_debugfs_root; + +static void nfcsim_debugfs_init(void) +{ + nfcsim_debugfs_root = debugfs_create_dir("nfcsim", NULL); + + if (!nfcsim_debugfs_root) + pr_err("Could not create debugfs entry\n"); + +} + +static void nfcsim_debugfs_remove(void) +{ + debugfs_remove_recursive(nfcsim_debugfs_root); +} + +static void nfcsim_debugfs_init_dev(struct nfcsim *dev) +{ + struct dentry *dev_dir; + char devname[5]; /* nfcX\0 */ + u32 idx; + int n; + + if (!nfcsim_debugfs_root) { + NFCSIM_ERR(dev, "nfcsim debugfs not initialized\n"); + return; + } + + idx = dev->nfc_digital_dev->nfc_dev->idx; + n = snprintf(devname, sizeof(devname), "nfc%d", idx); + if (n >= sizeof(devname)) { + NFCSIM_ERR(dev, "Could not compute dev name for dev %d\n", idx); + return; + } + + dev_dir = debugfs_create_dir(devname, nfcsim_debugfs_root); + if (!dev_dir) { + NFCSIM_ERR(dev, "Could not create debugfs entries for nfc%d\n", + idx); + return; + } +} + static struct nfcsim *nfcsim_device_new(struct nfcsim_link *link_in, struct nfcsim_link *link_out) { @@ -366,6 +411,8 @@ static struct nfcsim *nfcsim_device_new(struct nfcsim_link *link_in, return ERR_PTR(rc); } + nfcsim_debugfs_init_dev(dev); + return dev; } @@ -400,6 +447,8 @@ static int __init nfcsim_init(void) goto exit_err; } + nfcsim_debugfs_init(); + dev0 = nfcsim_device_new(link0, link1); if (IS_ERR(dev0)) { rc = PTR_ERR(dev0); @@ -439,6 +488,8 @@ static void __exit nfcsim_exit(void) nfcsim_link_free(link0); nfcsim_link_free(link1); + + nfcsim_debugfs_remove(); } module_init(nfcsim_init); -- cgit v0.10.2 From 2a0fe4fe5bf2a6e2277354e7e8f369a20d881891 Mon Sep 17 00:00:00 2001 From: Thierry Escande Date: Tue, 19 Jul 2016 11:58:17 +0200 Subject: NFC: nfcsim: Simulate lost frames through debugfs entry This patch allows to simulate the lost of frames exchanged between the 2 nfcsim devices through a control entry in the debugfs and is used as follow: echo n > /sys/kernel/debug/nfcsim/nfcX/dropframe Where n specifies the number of frames to be dropped between 0 and 255 and nfcX is either nfc0 or nfc1, one of the two nfcsim devices. In the following example, the next frame that should be sent by the nfc0 device will be dropped and thus not received by the nfc1 device: echo 1 > /sys/kernel/debug/nfcsim/nfc0/dropframe The value of 0 can be used to reset the dropframe counter. Signed-off-by: Thierry Escande Signed-off-by: Samuel Ortiz diff --git a/drivers/nfc/nfcsim.c b/drivers/nfc/nfcsim.c index 97067a5..a466e79 100644 --- a/drivers/nfc/nfcsim.c +++ b/drivers/nfc/nfcsim.c @@ -54,6 +54,8 @@ struct nfcsim { nfc_digital_cmd_complete_t cb; void *arg; + + u8 dropframe; }; struct nfcsim_link { @@ -223,6 +225,14 @@ static int nfcsim_send(struct nfc_digital_dev *ddev, struct sk_buff *skb, schedule_work(&dev->recv_work); + if (dev->dropframe) { + NFCSIM_DBG(dev, "dropping frame (out of %d)\n", dev->dropframe); + dev_kfree_skb(skb); + dev->dropframe--; + + return 0; + } + if (skb) { nfcsim_link_set_skb(dev->link_out, skb, dev->rf_tech, dev->mode); @@ -372,6 +382,8 @@ static void nfcsim_debugfs_init_dev(struct nfcsim *dev) idx); return; } + + debugfs_create_u8("dropframe", 0664, dev_dir, &dev->dropframe); } static struct nfcsim *nfcsim_device_new(struct nfcsim_link *link_in, -- cgit v0.10.2 From 9345b24f9f257708201dfab2d36ad2a411a281b4 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sun, 17 Jul 2016 16:08:43 +0300 Subject: ravb: fix DMA channel misreporting Currently 'ifconfig' for the Ethernet devices handled by this driver shows "DMA chan: ff" while the driver doesn't use any DMA channels. Not assigning a value to 'net_device::dma' causes 'ifconfig' to correctly not report a DMA channel. Signed-off-by: Sergei Shtylyov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 5349284..8377d02 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -1908,7 +1908,6 @@ static int ravb_probe(struct platform_device *pdev) /* The Ether-specific entries in the device structure. */ ndev->base_addr = res->start; - ndev->dma = -1; chip_id = (enum ravb_chip_id)of_device_get_match_data(&pdev->dev); -- cgit v0.10.2 From 3c78658e26b3bbb29f646416f52f4b4d875de471 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sun, 17 Jul 2016 16:09:19 +0300 Subject: sh_eth: fix DMA channel misreporting Currently 'ifconfig' for the Ethernet devices handled by this driver shows "DMA chan: ff" while the driver doesn't use any DMA channels. Not assigning a value to 'net_device::dma' causes 'ifconfig' to correctly not report a DMA channel. Signed-off-by: Sergei Shtylyov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index 04cd39f..7bd910c 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -2996,7 +2996,6 @@ static int sh_eth_drv_probe(struct platform_device *pdev) if (devno < 0) devno = 0; - ndev->dma = -1; ret = platform_get_irq(pdev, 0); if (ret < 0) goto out_release; -- cgit v0.10.2 From 6cfb3bcc0641109951a124019cd2e0623107d18d Mon Sep 17 00:00:00 2001 From: Charles-Antoine Couret Date: Tue, 19 Jul 2016 11:13:10 +0200 Subject: Marvell phy: check link status in case of fiber link. For concerned phy, the fiber link is checked before the copper link. According to datasheet, the link which is up is enabled. If both links are down, copper link would be used. To detect fiber link status, we used the real time status because of troubles with the copper method. Tested with Marvell 88E1512. Reviewed-by: Andrew Lunn Signed-off-by: Charles-Antoine Couret Signed-off-by: David S. Miller diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index ec2c1ee..75b9211 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -138,6 +138,20 @@ #define MII_88E1510_GEN_CTRL_REG_1_MODE_SGMII 0x1 /* SGMII to copper */ #define MII_88E1510_GEN_CTRL_REG_1_RESET 0x8000 /* Soft reset */ +#define LPA_FIBER_1000HALF 0x40 +#define LPA_FIBER_1000FULL 0x20 + +#define LPA_PAUSE_FIBER 0x180 +#define LPA_PAUSE_ASYM_FIBER 0x100 + +#define ADVERTISE_FIBER_1000HALF 0x40 +#define ADVERTISE_FIBER_1000FULL 0x20 + +#define ADVERTISE_PAUSE_FIBER 0x180 +#define ADVERTISE_PAUSE_ASYM_FIBER 0x100 + +#define REGISTER_LINK_STATUS 0x400 + MODULE_DESCRIPTION("Marvell PHY driver"); MODULE_AUTHOR("Andy Fleming"); MODULE_LICENSE("GPL"); @@ -890,26 +904,79 @@ static int m88e1145_config_init(struct phy_device *phydev) return 0; } -/* marvell_read_status +/** + * fiber_lpa_to_ethtool_lpa_t + * @lpa: value of the MII_LPA register for fiber link + * + * A small helper function that translates MII_LPA + * bits to ethtool LP advertisement settings. + */ +static u32 fiber_lpa_to_ethtool_lpa_t(u32 lpa) +{ + u32 result = 0; + + if (lpa & LPA_FIBER_1000HALF) + result |= ADVERTISED_1000baseT_Half; + if (lpa & LPA_FIBER_1000FULL) + result |= ADVERTISED_1000baseT_Full; + + return result; +} + +/** + * marvell_update_link - update link status in real time in @phydev + * @phydev: target phy_device struct + * + * Description: Update the value in phydev->link to reflect the + * current link value. + */ +static int marvell_update_link(struct phy_device *phydev, int fiber) +{ + int status; + + /* Use the generic register for copper link, or specific + * register for fiber case */ + if (fiber) { + status = phy_read(phydev, MII_M1011_PHY_STATUS); + if (status < 0) + return status; + + if ((status & REGISTER_LINK_STATUS) == 0) + phydev->link = 0; + else + phydev->link = 1; + } else { + return genphy_update_link(phydev); + } + + return 0; +} + +/* marvell_read_status_page * - * Generic status code does not detect Fiber correctly! * Description: * Check the link, then figure out the current state * by comparing what we advertise with what the link partner * advertises. Start by checking the gigabit possibilities, * then move on to 10/100. */ -static int marvell_read_status(struct phy_device *phydev) +static int marvell_read_status_page(struct phy_device *phydev, int page) { int adv; int err; int lpa; int lpagb; int status = 0; + int fiber; - /* Update the link, but return if there + /* Detect and update the link, but return if there * was an error */ - err = genphy_update_link(phydev); + if (page == MII_M1111_FIBER) + fiber = 1; + else + fiber = 0; + + err = marvell_update_link(phydev, fiber); if (err) return err; @@ -930,9 +997,6 @@ static int marvell_read_status(struct phy_device *phydev) if (adv < 0) return adv; - phydev->lp_advertising = mii_stat1000_to_ethtool_lpa_t(lpagb) | - mii_lpa_to_ethtool_lpa_t(lpa); - lpa &= adv; if (status & MII_M1011_PHY_STATUS_FULLDUPLEX) @@ -957,9 +1021,30 @@ static int marvell_read_status(struct phy_device *phydev) break; } - if (phydev->duplex == DUPLEX_FULL) { - phydev->pause = lpa & LPA_PAUSE_CAP ? 1 : 0; - phydev->asym_pause = lpa & LPA_PAUSE_ASYM ? 1 : 0; + if (!fiber) { + phydev->lp_advertising = mii_stat1000_to_ethtool_lpa_t(lpagb) | + mii_lpa_to_ethtool_lpa_t(lpa); + + if (phydev->duplex == DUPLEX_FULL) { + phydev->pause = lpa & LPA_PAUSE_CAP ? 1 : 0; + phydev->asym_pause = lpa & LPA_PAUSE_ASYM ? 1 : 0; + } + } else { + /* The fiber link is only 1000M capable */ + phydev->lp_advertising = fiber_lpa_to_ethtool_lpa_t(lpa); + + if (phydev->duplex == DUPLEX_FULL) { + if (!(lpa & LPA_PAUSE_FIBER)) { + phydev->pause = 0; + phydev->asym_pause = 0; + } else if ((lpa & LPA_PAUSE_ASYM_FIBER)) { + phydev->pause = 1; + phydev->asym_pause = 1; + } else { + phydev->pause = 1; + phydev->asym_pause = 0; + } + } } } else { int bmcr = phy_read(phydev, MII_BMCR); @@ -986,6 +1071,50 @@ static int marvell_read_status(struct phy_device *phydev) return 0; } +/* marvell_read_status + * + * Some Marvell's phys have two modes: fiber and copper. + * Both need status checked. + * Description: + * First, check the fiber link and status. + * If the fiber link is down, check the copper link and status which + * will be the default value if both link are down. + */ +static int marvell_read_status(struct phy_device *phydev) +{ + int err; + + /* Check the fiber mode first */ + if (phydev->supported & SUPPORTED_FIBRE) { + err = phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_FIBER); + if (err < 0) + goto error; + + err = marvell_read_status_page(phydev, MII_M1111_FIBER); + if (err < 0) + goto error; + + /* If the fiber link is up, it is the selected and used link. + * In this case, we need to stay in the fiber page. + * Please to be careful about that, avoid to restore Copper page + * in other functions which could break the behaviour + * for some fiber phy like 88E1512. + * */ + if (phydev->link) + return 0; + + /* If fiber link is down, check and save copper mode state */ + err = phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER); + if (err < 0) + goto error; + } + + return marvell_read_status_page(phydev, MII_M1111_COPPER); + +error: + phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER); + return err; +} static int marvell_aneg_done(struct phy_device *phydev) { int retval = phy_read(phydev, MII_M1011_PHY_STATUS); @@ -1361,7 +1490,7 @@ static struct phy_driver marvell_drivers[] = { .phy_id = MARVELL_PHY_ID_88E1510, .phy_id_mask = MARVELL_PHY_ID_MASK, .name = "Marvell 88E1510", - .features = PHY_GBIT_FEATURES, + .features = PHY_GBIT_FEATURES | SUPPORTED_FIBRE, .flags = PHY_HAS_INTERRUPT, .probe = marvell_probe, .config_init = &m88e1510_config_init, -- cgit v0.10.2 From 2170fef78a400ca3c891da5db73db5e70388aba7 Mon Sep 17 00:00:00 2001 From: Charles-Antoine Couret Date: Tue, 19 Jul 2016 11:13:11 +0200 Subject: Marvell phy: add field to get errors from fiber link. Add support for the fiber receiver error counter in the statistics. Rename the current counter which is for copper errors to phy_receive_errors_copper, so it is easy to distinguish copper from fiber. Reviewed-by: Andrew Lunn Signed-off-by: Charles-Antoine Couret Signed-off-by: David S. Miller diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index 75b9211..0dde7ca 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -151,6 +151,7 @@ #define ADVERTISE_PAUSE_ASYM_FIBER 0x100 #define REGISTER_LINK_STATUS 0x400 +#define NB_FIBER_STATS 1 MODULE_DESCRIPTION("Marvell PHY driver"); MODULE_AUTHOR("Andy Fleming"); @@ -164,8 +165,9 @@ struct marvell_hw_stat { }; static struct marvell_hw_stat marvell_hw_stats[] = { - { "phy_receive_errors", 0, 21, 16}, + { "phy_receive_errors_copper", 0, 21, 16}, { "phy_idle_errors", 0, 10, 8 }, + { "phy_receive_errors_fiber", 1, 21, 16}, }; struct marvell_priv { @@ -1236,7 +1238,10 @@ static int m88e1318_set_wol(struct phy_device *phydev, struct ethtool_wolinfo *w static int marvell_get_sset_count(struct phy_device *phydev) { - return ARRAY_SIZE(marvell_hw_stats); + if (phydev->supported & SUPPORTED_FIBRE) + return ARRAY_SIZE(marvell_hw_stats); + else + return ARRAY_SIZE(marvell_hw_stats) - NB_FIBER_STATS; } static void marvell_get_strings(struct phy_device *phydev, u8 *data) -- cgit v0.10.2 From 78301ebe9b5a2645d05aa7f791d068b727e7500f Mon Sep 17 00:00:00 2001 From: Charles-Antoine Couret Date: Tue, 19 Jul 2016 11:13:12 +0200 Subject: Marvell phy: add configuration of autonegociation for fiber link. To be correctly initilized, the fiber interface needs to be configured via autonegociation registers which use some customs options or registers. Reviewed-by: Andrew Lunn Signed-off-by: Charles-Antoine Couret Signed-off-by: David S. Miller diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index 0dde7ca..fb4d402 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -493,15 +493,122 @@ static int m88e1318_config_aneg(struct phy_device *phydev) return m88e1121_config_aneg(phydev); } +/** + * ethtool_adv_to_fiber_adv_t + * @ethadv: the ethtool advertisement settings + * + * A small helper function that translates ethtool advertisement + * settings to phy autonegotiation advertisements for the + * MII_ADV register for fiber link. + */ +static inline u32 ethtool_adv_to_fiber_adv_t(u32 ethadv) +{ + u32 result = 0; + + if (ethadv & ADVERTISED_1000baseT_Half) + result |= ADVERTISE_FIBER_1000HALF; + if (ethadv & ADVERTISED_1000baseT_Full) + result |= ADVERTISE_FIBER_1000FULL; + + if ((ethadv & ADVERTISE_PAUSE_ASYM) && (ethadv & ADVERTISE_PAUSE_CAP)) + result |= LPA_PAUSE_ASYM_FIBER; + else if (ethadv & ADVERTISE_PAUSE_CAP) + result |= (ADVERTISE_PAUSE_FIBER + & (~ADVERTISE_PAUSE_ASYM_FIBER)); + + return result; +} + +/** + * marvell_config_aneg_fiber - restart auto-negotiation or write BMCR + * @phydev: target phy_device struct + * + * Description: If auto-negotiation is enabled, we configure the + * advertising, and then restart auto-negotiation. If it is not + * enabled, then we write the BMCR. Adapted for fiber link in + * some Marvell's devices. + */ +static int marvell_config_aneg_fiber(struct phy_device *phydev) +{ + int changed = 0; + int err; + int adv, oldadv; + u32 advertise; + + if (phydev->autoneg != AUTONEG_ENABLE) + return genphy_setup_forced(phydev); + + /* Only allow advertising what this PHY supports */ + phydev->advertising &= phydev->supported; + advertise = phydev->advertising; + + /* Setup fiber advertisement */ + adv = phy_read(phydev, MII_ADVERTISE); + if (adv < 0) + return adv; + + oldadv = adv; + adv &= ~(ADVERTISE_FIBER_1000HALF | ADVERTISE_FIBER_1000FULL + | LPA_PAUSE_FIBER); + adv |= ethtool_adv_to_fiber_adv_t(advertise); + + if (adv != oldadv) { + err = phy_write(phydev, MII_ADVERTISE, adv); + if (err < 0) + return err; + + changed = 1; + } + + if (changed == 0) { + /* Advertisement hasn't changed, but maybe aneg was never on to + * begin with? Or maybe phy was isolated? + */ + int ctl = phy_read(phydev, MII_BMCR); + + if (ctl < 0) + return ctl; + + if (!(ctl & BMCR_ANENABLE) || (ctl & BMCR_ISOLATE)) + changed = 1; /* do restart aneg */ + } + + /* Only restart aneg if we are advertising something different + * than we were before. + */ + if (changed > 0) + changed = genphy_restart_aneg(phydev); + + return changed; +} + static int m88e1510_config_aneg(struct phy_device *phydev) { int err; + err = phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER); + if (err < 0) + goto error; + + /* Configure the copper link first */ err = m88e1318_config_aneg(phydev); if (err < 0) - return err; + goto error; - return 0; + /* Then the fiber link */ + err = phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_FIBER); + if (err < 0) + goto error; + + err = marvell_config_aneg_fiber(phydev); + if (err < 0) + goto error; + + return phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER); + +error: + phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER); + return err; } static int marvell_config_init(struct phy_device *phydev) -- cgit v0.10.2 From 3758be3dc162b56eaafaa914b98845f914106ef2 Mon Sep 17 00:00:00 2001 From: Charles-Antoine Couret Date: Tue, 19 Jul 2016 11:13:13 +0200 Subject: Marvell phy: add functions to suspend and resume both interfaces: fiber and copper links. These functions used standards registers in a different page for both interfaces: copper and fiber. Reviewed-by: Andrew Lunn Signed-off-by: Charles-Antoine Couret Signed-off-by: David S. Miller diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c index fb4d402..c2dcf02 100644 --- a/drivers/net/phy/marvell.c +++ b/drivers/net/phy/marvell.c @@ -1224,6 +1224,75 @@ error: phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER); return err; } + +/* marvell_suspend + * + * Some Marvell's phys have two modes: fiber and copper. + * Both need to be suspended + */ +static int marvell_suspend(struct phy_device *phydev) +{ + int err; + + /* Suspend the fiber mode first */ + if (!(phydev->supported & SUPPORTED_FIBRE)) { + err = phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_FIBER); + if (err < 0) + goto error; + + /* With the page set, use the generic suspend */ + err = genphy_suspend(phydev); + if (err < 0) + goto error; + + /* Then, the copper link */ + err = phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER); + if (err < 0) + goto error; + } + + /* With the page set, use the generic suspend */ + return genphy_suspend(phydev); + +error: + phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER); + return err; +} + +/* marvell_resume + * + * Some Marvell's phys have two modes: fiber and copper. + * Both need to be resumed + */ +static int marvell_resume(struct phy_device *phydev) +{ + int err; + + /* Resume the fiber mode first */ + if (!(phydev->supported & SUPPORTED_FIBRE)) { + err = phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_FIBER); + if (err < 0) + goto error; + + /* With the page set, use the generic resume */ + err = genphy_resume(phydev); + if (err < 0) + goto error; + + /* Then, the copper link */ + err = phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER); + if (err < 0) + goto error; + } + + /* With the page set, use the generic resume */ + return genphy_resume(phydev); + +error: + phy_write(phydev, MII_MARVELL_PHY_PAGE, MII_M1111_COPPER); + return err; +} + static int marvell_aneg_done(struct phy_device *phydev) { int retval = phy_read(phydev, MII_M1011_PHY_STATUS); @@ -1611,8 +1680,8 @@ static struct phy_driver marvell_drivers[] = { .ack_interrupt = &marvell_ack_interrupt, .config_intr = &marvell_config_intr, .did_interrupt = &m88e1121_did_interrupt, - .resume = &genphy_resume, - .suspend = &genphy_suspend, + .resume = &marvell_resume, + .suspend = &marvell_suspend, .get_sset_count = marvell_get_sset_count, .get_strings = marvell_get_strings, .get_stats = marvell_get_stats, -- cgit v0.10.2 From 3e8060fa837630f6fb4acbf59ba588c6df5b2f50 Mon Sep 17 00:00:00 2001 From: Prashant Sreedharan Date: Mon, 18 Jul 2016 07:15:20 -0400 Subject: bnxt_en: Add basic support for Nitro in North Star 2. Nitro is the embedded version of the ethernet controller in the North Star 2 SoC. Add basic code to recognize the chip ID and disable the features (ntuple, TPA, ring and port statistics) not supported on Nitro A0. Signed-off-by: Prashant Sreedharan Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 659faa6..4970e72 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -2661,7 +2661,7 @@ static int bnxt_alloc_stats(struct bnxt *bp) cpr->hw_stats_ctx_id = INVALID_STATS_CTX_ID; } - if (BNXT_PF(bp)) { + if (BNXT_PF(bp) && bp->chip_num != CHIP_NUM_58700) { bp->hw_port_stats_size = sizeof(struct rx_port_stats) + sizeof(struct tx_port_stats) + 1024; @@ -3922,6 +3922,9 @@ static int bnxt_hwrm_stat_ctx_free(struct bnxt *bp) if (!bp->bnapi) return 0; + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) + return 0; + bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_STAT_CTX_FREE, -1, -1); mutex_lock(&bp->hwrm_cmd_lock); @@ -3950,6 +3953,9 @@ static int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp) struct hwrm_stat_ctx_alloc_input req = {0}; struct hwrm_stat_ctx_alloc_output *resp = bp->hwrm_cmd_resp_addr; + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) + return 0; + bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_STAT_CTX_ALLOC, -1, -1); req.update_period_ms = cpu_to_le32(bp->stats_coal_ticks / 1000); @@ -4163,6 +4169,9 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp) bp->hwrm_max_req_len = le16_to_cpu(resp->max_req_win_len); bp->chip_num = le16_to_cpu(resp->chip_num); + if (bp->chip_num == CHIP_NUM_58700 && !resp->chip_rev && + !resp->chip_metal) + bp->flags |= BNXT_FLAG_CHIP_NITRO_A0; hwrm_ver_get_exit: mutex_unlock(&bp->hwrm_cmd_lock); @@ -5681,7 +5690,7 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features) bool update_tpa = false; flags &= ~BNXT_FLAG_ALL_CONFIG_FEATS; - if ((features & NETIF_F_GRO) && (bp->pdev->revision > 0)) + if ((features & NETIF_F_GRO) && !BNXT_CHIP_TYPE_NITRO_A0(bp)) flags |= BNXT_FLAG_GRO; if (features & NETIF_F_LRO) flags |= BNXT_FLAG_LRO; @@ -6576,13 +6585,25 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) pci_set_drvdata(pdev, dev); + rc = bnxt_alloc_hwrm_resources(bp); + if (rc) + goto init_err; + + mutex_init(&bp->hwrm_cmd_lock); + rc = bnxt_hwrm_ver_get(bp); + if (rc) + goto init_err; + dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_GRE | NETIF_F_GSO_IPXIP4 | NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_PARTIAL | NETIF_F_RXHASH | - NETIF_F_RXCSUM | NETIF_F_LRO | NETIF_F_GRO; + NETIF_F_RXCSUM | NETIF_F_GRO; + + if (!BNXT_CHIP_TYPE_NITRO_A0(bp)) + dev->hw_features |= NETIF_F_LRO; dev->hw_enc_features = NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SG | @@ -6601,15 +6622,6 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) #ifdef CONFIG_BNXT_SRIOV init_waitqueue_head(&bp->sriov_cfg_wait); #endif - rc = bnxt_alloc_hwrm_resources(bp); - if (rc) - goto init_err; - - mutex_init(&bp->hwrm_cmd_lock); - rc = bnxt_hwrm_ver_get(bp); - if (rc) - goto init_err; - bp->gro_func = bnxt_gro_func_5730x; if (BNXT_CHIP_NUM_57X1X(bp->chip_num)) bp->gro_func = bnxt_gro_func_5731x; @@ -6647,7 +6659,7 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) #endif bnxt_set_dflt_rings(bp); - if (BNXT_PF(bp)) { + if (BNXT_PF(bp) && !BNXT_CHIP_TYPE_NITRO_A0(bp)) { dev->hw_features |= NETIF_F_NTUPLE; if (bnxt_rfs_capable(bp)) { bp->flags |= BNXT_FLAG_RFS; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 2313e37..e667150 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -893,6 +893,7 @@ struct bnxt { #define CHIP_NUM_57301 0x16c8 #define CHIP_NUM_57302 0x16c9 #define CHIP_NUM_57304 0x16ca +#define CHIP_NUM_58700 0x16cd #define CHIP_NUM_57402 0x16d0 #define CHIP_NUM_57404 0x16d1 #define CHIP_NUM_57406 0x16d2 @@ -954,6 +955,7 @@ struct bnxt { #define BNXT_FLAG_SHARED_RINGS 0x200 #define BNXT_FLAG_PORT_STATS 0x400 #define BNXT_FLAG_EEE_CAP 0x1000 + #define BNXT_FLAG_CHIP_NITRO_A0 0x1000000 #define BNXT_FLAG_ALL_CONFIG_FEATS (BNXT_FLAG_TPA | \ BNXT_FLAG_RFS | \ @@ -963,6 +965,7 @@ struct bnxt { #define BNXT_VF(bp) ((bp)->flags & BNXT_FLAG_VF) #define BNXT_NPAR(bp) ((bp)->port_partition_type) #define BNXT_SINGLE_PF(bp) (BNXT_PF(bp) && !BNXT_NPAR(bp)) +#define BNXT_CHIP_TYPE_NITRO_A0(bp) ((bp)->flags & BNXT_FLAG_CHIP_NITRO_A0) struct bnxt_napi **bnapi; -- cgit v0.10.2 From 94ce9caa0f75b0d56e69550e84d7a1653f0ef3b0 Mon Sep 17 00:00:00 2001 From: Prashant Sreedharan Date: Mon, 18 Jul 2016 07:15:21 -0400 Subject: bnxt_en: Workaround Nitro A0 hardware RX bug (part 1). Nitro A0 has a hardware bug in the rx path. The workaround is to create a special COS context as a path for non-RSS (non-IP) packets. Without this workaround, the chip may stall when receiving RSS and non-RSS packets. Add infrastructure to allow 2 contexts (RSS and CoS) per VNIC. Allocate and configure the CoS context for Nitro A0. Signed-off-by: Prashant Sreedharan Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 4970e72..de74012 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -2357,7 +2357,8 @@ static void bnxt_init_vnics(struct bnxt *bp) struct bnxt_vnic_info *vnic = &bp->vnic_info[i]; vnic->fw_vnic_id = INVALID_HW_RING_ID; - vnic->fw_rss_cos_lb_ctx = INVALID_HW_RING_ID; + vnic->fw_rss_cos_lb_ctx[0] = INVALID_HW_RING_ID; + vnic->fw_rss_cos_lb_ctx[1] = INVALID_HW_RING_ID; vnic->fw_l2_ctx_id = INVALID_HW_RING_ID; if (bp->vnic_info[i].rss_hash_key) { @@ -3308,7 +3309,7 @@ static int bnxt_hwrm_vnic_set_rss(struct bnxt *bp, u16 vnic_id, bool set_rss) struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id]; struct hwrm_vnic_rss_cfg_input req = {0}; - if (vnic->fw_rss_cos_lb_ctx == INVALID_HW_RING_ID) + if (vnic->fw_rss_cos_lb_ctx[0] == INVALID_HW_RING_ID) return 0; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_RSS_CFG, -1, -1); @@ -3336,7 +3337,7 @@ static int bnxt_hwrm_vnic_set_rss(struct bnxt *bp, u16 vnic_id, bool set_rss) req.hash_key_tbl_addr = cpu_to_le64(vnic->rss_hash_key_dma_addr); } - req.rss_ctx_idx = cpu_to_le16(vnic->fw_rss_cos_lb_ctx); + req.rss_ctx_idx = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]); return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } @@ -3359,32 +3360,35 @@ static int bnxt_hwrm_vnic_set_hds(struct bnxt *bp, u16 vnic_id) return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); } -static void bnxt_hwrm_vnic_ctx_free_one(struct bnxt *bp, u16 vnic_id) +static void bnxt_hwrm_vnic_ctx_free_one(struct bnxt *bp, u16 vnic_id, + u16 ctx_idx) { struct hwrm_vnic_rss_cos_lb_ctx_free_input req = {0}; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_RSS_COS_LB_CTX_FREE, -1, -1); req.rss_cos_lb_ctx_id = - cpu_to_le16(bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx); + cpu_to_le16(bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx[ctx_idx]); hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx = INVALID_HW_RING_ID; + bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx[ctx_idx] = INVALID_HW_RING_ID; } static void bnxt_hwrm_vnic_ctx_free(struct bnxt *bp) { - int i; + int i, j; for (i = 0; i < bp->nr_vnics; i++) { struct bnxt_vnic_info *vnic = &bp->vnic_info[i]; - if (vnic->fw_rss_cos_lb_ctx != INVALID_HW_RING_ID) - bnxt_hwrm_vnic_ctx_free_one(bp, i); + for (j = 0; j < BNXT_MAX_CTX_PER_VNIC; j++) { + if (vnic->fw_rss_cos_lb_ctx[j] != INVALID_HW_RING_ID) + bnxt_hwrm_vnic_ctx_free_one(bp, i, j); + } } bp->rsscos_nr_ctxs = 0; } -static int bnxt_hwrm_vnic_ctx_alloc(struct bnxt *bp, u16 vnic_id) +static int bnxt_hwrm_vnic_ctx_alloc(struct bnxt *bp, u16 vnic_id, u16 ctx_idx) { int rc; struct hwrm_vnic_rss_cos_lb_ctx_alloc_input req = {0}; @@ -3397,7 +3401,7 @@ static int bnxt_hwrm_vnic_ctx_alloc(struct bnxt *bp, u16 vnic_id) mutex_lock(&bp->hwrm_cmd_lock); rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); if (!rc) - bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx = + bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx[ctx_idx] = le16_to_cpu(resp->rss_cos_lb_ctx_id); mutex_unlock(&bp->hwrm_cmd_lock); @@ -3416,8 +3420,15 @@ static int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id) req.enables = cpu_to_le32(VNIC_CFG_REQ_ENABLES_DFLT_RING_GRP | VNIC_CFG_REQ_ENABLES_RSS_RULE | VNIC_CFG_REQ_ENABLES_MRU); - req.rss_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx); - req.cos_rule = cpu_to_le16(0xffff); + req.rss_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]); + + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) { + req.cos_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[1]); + req.enables |= cpu_to_le32(VNIC_CFG_REQ_ENABLES_COS_RULE); + } else { + req.cos_rule = cpu_to_le16(0xffff); + } + if (vnic->flags & BNXT_VNIC_RSS_FLAG) ring = 0; else if (vnic->flags & BNXT_VNIC_RFS_FLAG) @@ -3489,7 +3500,8 @@ static int bnxt_hwrm_vnic_alloc(struct bnxt *bp, u16 vnic_id, bp->grp_info[grp_idx].fw_grp_id; } - bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx = INVALID_HW_RING_ID; + bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx[0] = INVALID_HW_RING_ID; + bp->vnic_info[vnic_id].fw_rss_cos_lb_ctx[1] = INVALID_HW_RING_ID; if (vnic_id == 0) req.flags = cpu_to_le32(VNIC_ALLOC_REQ_FLAGS_DEFAULT); @@ -4261,7 +4273,7 @@ static int bnxt_setup_vnic(struct bnxt *bp, u16 vnic_id) int rc; /* allocate context for vnic */ - rc = bnxt_hwrm_vnic_ctx_alloc(bp, vnic_id); + rc = bnxt_hwrm_vnic_ctx_alloc(bp, vnic_id, 0); if (rc) { netdev_err(bp->dev, "hwrm vnic %d alloc failure rc: %x\n", vnic_id, rc); @@ -4269,6 +4281,16 @@ static int bnxt_setup_vnic(struct bnxt *bp, u16 vnic_id) } bp->rsscos_nr_ctxs++; + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) { + rc = bnxt_hwrm_vnic_ctx_alloc(bp, vnic_id, 1); + if (rc) { + netdev_err(bp->dev, "hwrm vnic %d cos ctx alloc failure rc: %x\n", + vnic_id, rc); + goto vnic_setup_err; + } + bp->rsscos_nr_ctxs++; + } + /* configure default vnic, ring grp */ rc = bnxt_hwrm_vnic_cfg(bp, vnic_id); if (rc) { diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index e667150..5307a2e 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -695,7 +695,8 @@ struct bnxt_ring_grp_info { struct bnxt_vnic_info { u16 fw_vnic_id; /* returned by Chimp during alloc */ - u16 fw_rss_cos_lb_ctx; +#define BNXT_MAX_CTX_PER_VNIC 2 + u16 fw_rss_cos_lb_ctx[BNXT_MAX_CTX_PER_VNIC]; u16 fw_l2_ctx_id; #define BNXT_MAX_UC_ADDRS 4 __le64 fw_l2_filter_id[BNXT_MAX_UC_ADDRS]; -- cgit v0.10.2 From 765951938e2fe2e30571ef4a7de6a46659ce4c68 Mon Sep 17 00:00:00 2001 From: Prashant Sreedharan Date: Mon, 18 Jul 2016 07:15:22 -0400 Subject: bnxt_en: Workaround Nitro A0 hardware RX bug (part 2). The hardware is unable to drop rx packets not matching the RX filters. To workaround it, we create a special VNIC and configure the hardware to direct all packets not matching the filters to it. We then setup the driver to drop packets received on this VNIC. This patch creates the infrastructure for this VNIC, reserves a completion ring, and rx rings. Only shared completion ring mode is supported. The next 2 patches add a NAPI to handle packets from this VNIC and the setup of the VNIC. Signed-off-by: Prashant Sreedharan Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index de74012..75a8b30 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -3433,6 +3433,8 @@ static int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id) ring = 0; else if (vnic->flags & BNXT_VNIC_RFS_FLAG) ring = vnic_id - 1; + else if ((vnic_id == 1) && BNXT_CHIP_TYPE_NITRO_A0(bp)) + ring = bp->rx_nr_rings - 1; grp_idx = bp->rx_ring[ring].bnapi->index; req.vnic_id = cpu_to_le16(vnic->fw_vnic_id); @@ -4365,6 +4367,7 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init) { struct bnxt_vnic_info *vnic = &bp->vnic_info[0]; int rc = 0; + unsigned int rx_nr_rings = bp->rx_nr_rings; if (irq_re_init) { rc = bnxt_hwrm_stat_ctx_alloc(bp); @@ -4387,8 +4390,11 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init) goto err_out; } + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) + rx_nr_rings--; + /* default vnic 0 */ - rc = bnxt_hwrm_vnic_alloc(bp, 0, 0, bp->rx_nr_rings); + rc = bnxt_hwrm_vnic_alloc(bp, 0, 0, rx_nr_rings); if (rc) { netdev_err(bp->dev, "hwrm vnic alloc failure rc: %x\n", rc); goto err_out; @@ -6519,7 +6525,10 @@ static void _bnxt_get_max_rings(struct bnxt *bp, int *max_rx, int *max_tx, *max_cp = min_t(int, *max_cp, bp->pf.max_stat_ctxs); max_ring_grps = bp->pf.max_hw_ring_grps; } - + if (BNXT_CHIP_TYPE_NITRO_A0(bp) && BNXT_PF(bp)) { + *max_cp -= 1; + *max_rx -= 2; + } if (bp->flags & BNXT_FLAG_AGG_RINGS) *max_rx >>= 1; *max_rx = min_t(int, *max_rx, max_ring_grps); @@ -6555,6 +6564,10 @@ static int bnxt_set_dflt_rings(struct bnxt *bp) bp->cp_nr_rings = sh ? max_t(int, bp->tx_nr_rings, bp->rx_nr_rings) : bp->tx_nr_rings + bp->rx_nr_rings; bp->num_stat_ctxs = bp->cp_nr_rings; + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) { + bp->rx_nr_rings++; + bp->cp_nr_rings++; + } return rc; } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 0f7dd86..492c06b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -362,9 +362,13 @@ static void bnxt_get_channels(struct net_device *dev, channel->max_other = 0; if (bp->flags & BNXT_FLAG_SHARED_RINGS) { channel->combined_count = bp->rx_nr_rings; + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) + channel->combined_count--; } else { - channel->rx_count = bp->rx_nr_rings; - channel->tx_count = bp->tx_nr_rings_per_tc; + if (!BNXT_CHIP_TYPE_NITRO_A0(bp)) { + channel->rx_count = bp->rx_nr_rings; + channel->tx_count = bp->tx_nr_rings_per_tc; + } } } @@ -387,6 +391,10 @@ static int bnxt_set_channels(struct net_device *dev, (channel->rx_count || channel->tx_count)) return -EINVAL; + if (BNXT_CHIP_TYPE_NITRO_A0(bp) && (channel->rx_count || + channel->tx_count)) + return -EINVAL; + if (channel->combined_count) sh = true; -- cgit v0.10.2 From 10bbdaf5e4879fd7fc51f25c84d7b10de16cbe0e Mon Sep 17 00:00:00 2001 From: Prashant Sreedharan Date: Mon, 18 Jul 2016 07:15:23 -0400 Subject: bnxt_en: Workaround Nitro A0 hardware RX bug (part 3). Allocate napi for special vnic, packets arriving on this napi will simply be dropped and the buffers will be replenished back to the HW. Signed-off-by: Prashant Sreedharan Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 75a8b30..f0aa582 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -1668,6 +1668,76 @@ static int bnxt_poll_work(struct bnxt *bp, struct bnxt_napi *bnapi, int budget) return rx_pkts; } +static int bnxt_poll_nitroa0(struct napi_struct *napi, int budget) +{ + struct bnxt_napi *bnapi = container_of(napi, struct bnxt_napi, napi); + struct bnxt *bp = bnapi->bp; + struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring; + struct bnxt_rx_ring_info *rxr = bnapi->rx_ring; + struct tx_cmp *txcmp; + struct rx_cmp_ext *rxcmp1; + u32 cp_cons, tmp_raw_cons; + u32 raw_cons = cpr->cp_raw_cons; + u32 rx_pkts = 0; + bool agg_event = false; + + while (1) { + int rc; + + cp_cons = RING_CMP(raw_cons); + txcmp = &cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)]; + + if (!TX_CMP_VALID(txcmp, raw_cons)) + break; + + if ((TX_CMP_TYPE(txcmp) & 0x30) == 0x10) { + tmp_raw_cons = NEXT_RAW_CMP(raw_cons); + cp_cons = RING_CMP(tmp_raw_cons); + rxcmp1 = (struct rx_cmp_ext *) + &cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)]; + + if (!RX_CMP_VALID(rxcmp1, tmp_raw_cons)) + break; + + /* force an error to recycle the buffer */ + rxcmp1->rx_cmp_cfa_code_errors_v2 |= + cpu_to_le32(RX_CMPL_ERRORS_CRC_ERROR); + + rc = bnxt_rx_pkt(bp, bnapi, &raw_cons, &agg_event); + if (likely(rc == -EIO)) + rx_pkts++; + else if (rc == -EBUSY) /* partial completion */ + break; + } else if (unlikely(TX_CMP_TYPE(txcmp) == + CMPL_BASE_TYPE_HWRM_DONE)) { + bnxt_hwrm_handler(bp, txcmp); + } else { + netdev_err(bp->dev, + "Invalid completion received on special ring\n"); + } + raw_cons = NEXT_RAW_CMP(raw_cons); + + if (rx_pkts == budget) + break; + } + + cpr->cp_raw_cons = raw_cons; + BNXT_CP_DB(cpr->cp_doorbell, cpr->cp_raw_cons); + writel(DB_KEY_RX | rxr->rx_prod, rxr->rx_doorbell); + writel(DB_KEY_RX | rxr->rx_prod, rxr->rx_doorbell); + + if (agg_event) { + writel(DB_KEY_RX | rxr->rx_agg_prod, rxr->rx_agg_doorbell); + writel(DB_KEY_RX | rxr->rx_agg_prod, rxr->rx_agg_doorbell); + } + + if (!bnxt_has_work(bp, cpr) && rx_pkts < budget) { + napi_complete(napi); + BNXT_CP_DB_REARM(cpr->cp_doorbell, cpr->cp_raw_cons); + } + return rx_pkts; +} + static int bnxt_poll(struct napi_struct *napi, int budget) { struct bnxt_napi *bnapi = container_of(napi, struct bnxt_napi, napi); @@ -4758,14 +4828,23 @@ static void bnxt_del_napi(struct bnxt *bp) static void bnxt_init_napi(struct bnxt *bp) { int i; + unsigned int cp_nr_rings = bp->cp_nr_rings; struct bnxt_napi *bnapi; if (bp->flags & BNXT_FLAG_USING_MSIX) { - for (i = 0; i < bp->cp_nr_rings; i++) { + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) + cp_nr_rings--; + for (i = 0; i < cp_nr_rings; i++) { bnapi = bp->bnapi[i]; netif_napi_add(bp->dev, &bnapi->napi, bnxt_poll, 64); } + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) { + bnapi = bp->bnapi[cp_nr_rings]; + netif_napi_add(bp->dev, &bnapi->napi, + bnxt_poll_nitroa0, 64); + napi_hash_add(&bnapi->napi); + } } else { bnapi = bp->bnapi[0]; netif_napi_add(bp->dev, &bnapi->napi, bnxt_poll, 64); -- cgit v0.10.2 From dc52c6c70e0066e9cef886907f820411bebe8e07 Mon Sep 17 00:00:00 2001 From: Prashant Sreedharan Date: Mon, 18 Jul 2016 07:15:24 -0400 Subject: bnxt_en: Workaround Nitro A0 RX hardware bug (part 4). Allocate special vnic for dropping packets not matching the RX filters. First vnic is for normal RX packets and the driver will drop all packets on the 2nd vnic. Signed-off-by: Prashant Sreedharan Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index f0aa582..fd6a174 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -2410,6 +2410,9 @@ static int bnxt_alloc_vnics(struct bnxt *bp) num_vnics += bp->rx_nr_rings; #endif + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) + num_vnics++; + bp->vnic_info = kcalloc(num_vnics, sizeof(struct bnxt_vnic_info), GFP_KERNEL); if (!bp->vnic_info) @@ -3271,8 +3274,10 @@ static int bnxt_hwrm_set_vnic_filter(struct bnxt *bp, u16 vnic_id, u16 idx, struct hwrm_cfa_l2_filter_alloc_output *resp = bp->hwrm_cmd_resp_addr; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_CFA_L2_FILTER_ALLOC, -1, -1); - req.flags = cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_RX | - CFA_L2_FILTER_ALLOC_REQ_FLAGS_OUTERMOST); + req.flags = cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_RX); + if (!BNXT_CHIP_TYPE_NITRO_A0(bp)) + req.flags |= + cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_FLAGS_OUTERMOST); req.dst_id = cpu_to_le16(bp->vnic_info[vnic_id].fw_vnic_id); req.enables = cpu_to_le32(CFA_L2_FILTER_ALLOC_REQ_ENABLES_L2_ADDR | @@ -3391,10 +3396,14 @@ static int bnxt_hwrm_vnic_set_rss(struct bnxt *bp, u16 vnic_id, bool set_rss) req.hash_type = cpu_to_le32(vnic->hash_type); - if (vnic->flags & BNXT_VNIC_RSS_FLAG) - max_rings = bp->rx_nr_rings; - else + if (vnic->flags & BNXT_VNIC_RSS_FLAG) { + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) + max_rings = bp->rx_nr_rings - 1; + else + max_rings = bp->rx_nr_rings; + } else { max_rings = 1; + } /* Fill the RSS indirection table with ring group ids */ for (i = 0, j = 0; i < HW_HASH_INDEX_SIZE; i++, j++) { @@ -3486,13 +3495,19 @@ static int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id) u16 def_vlan = 0; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VNIC_CFG, -1, -1); + + req.enables = cpu_to_le32(VNIC_CFG_REQ_ENABLES_DFLT_RING_GRP); /* Only RSS support for now TBD: COS & LB */ - req.enables = cpu_to_le32(VNIC_CFG_REQ_ENABLES_DFLT_RING_GRP | - VNIC_CFG_REQ_ENABLES_RSS_RULE | - VNIC_CFG_REQ_ENABLES_MRU); - req.rss_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]); + if (vnic->fw_rss_cos_lb_ctx[0] != INVALID_HW_RING_ID) { + req.rss_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[0]); + req.enables |= cpu_to_le32(VNIC_CFG_REQ_ENABLES_RSS_RULE | + VNIC_CFG_REQ_ENABLES_MRU); + } else { + req.rss_rule = cpu_to_le16(0xffff); + } - if (BNXT_CHIP_TYPE_NITRO_A0(bp)) { + if (BNXT_CHIP_TYPE_NITRO_A0(bp) && + (vnic->fw_rss_cos_lb_ctx[0] != INVALID_HW_RING_ID)) { req.cos_rule = cpu_to_le16(vnic->fw_rss_cos_lb_ctx[1]); req.enables |= cpu_to_le32(VNIC_CFG_REQ_ENABLES_COS_RULE); } else { @@ -4430,6 +4445,26 @@ static bool bnxt_promisc_ok(struct bnxt *bp) return true; } +static int bnxt_setup_nitroa0_vnic(struct bnxt *bp) +{ + unsigned int rc = 0; + + rc = bnxt_hwrm_vnic_alloc(bp, 1, bp->rx_nr_rings - 1, 1); + if (rc) { + netdev_err(bp->dev, "Cannot allocate special vnic for NS2 A0: %x\n", + rc); + return rc; + } + + rc = bnxt_hwrm_vnic_cfg(bp, 1); + if (rc) { + netdev_err(bp->dev, "Cannot allocate special vnic for NS2 A0: %x\n", + rc); + return rc; + } + return rc; +} + static int bnxt_cfg_rx_mode(struct bnxt *); static bool bnxt_mc_list_updated(struct bnxt *, u32 *); @@ -4519,7 +4554,14 @@ static int bnxt_init_chip(struct bnxt *bp, bool irq_re_init) rc = bnxt_hwrm_set_coal(bp); if (rc) netdev_warn(bp->dev, "HWRM set coalescing failure rc: %x\n", - rc); + rc); + + if (BNXT_CHIP_TYPE_NITRO_A0(bp)) { + rc = bnxt_setup_nitroa0_vnic(bp); + if (rc) + netdev_err(bp->dev, "Special vnic setup failure for NS2 A0 rc: %x\n", + rc); + } if (BNXT_VF(bp)) { bnxt_hwrm_func_qcfg(bp); -- cgit v0.10.2 From fa853dda19a1878d2a586de19f02bc9fed052425 Mon Sep 17 00:00:00 2001 From: Prashant Sreedharan Date: Mon, 18 Jul 2016 07:15:25 -0400 Subject: bnxt_en: Add BCM58700 PCI device ID for NS2 Nitro. A bridge device in NS2 has the same device ID as the ethernet controller. Add check to avoid probing the bridge device. Signed-off-by: Prashant Sreedharan Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index fd6a174..b55eef2 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -73,6 +73,7 @@ enum board_idx { BCM57301, BCM57302, BCM57304, + BCM58700, BCM57311, BCM57312, BCM57402, @@ -98,6 +99,7 @@ static const struct { { "Broadcom BCM57301 NetXtreme-C Single-port 10Gb Ethernet" }, { "Broadcom BCM57302 NetXtreme-C Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57304 NetXtreme-C Dual-port 10Gb/25Gb/40Gb/50Gb Ethernet" }, + { "Broadcom BCM58700 Nitro 4-port 1Gb/2.5Gb/10Gb Ethernet" }, { "Broadcom BCM57311 NetXtreme-C Single-port 10Gb Ethernet" }, { "Broadcom BCM57312 NetXtreme-C Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57402 NetXtreme-E Dual-port 10Gb Ethernet" }, @@ -120,6 +122,7 @@ static const struct pci_device_id bnxt_pci_tbl[] = { { PCI_VDEVICE(BROADCOM, 0x16c8), .driver_data = BCM57301 }, { PCI_VDEVICE(BROADCOM, 0x16c9), .driver_data = BCM57302 }, { PCI_VDEVICE(BROADCOM, 0x16ca), .driver_data = BCM57304 }, + { PCI_VDEVICE(BROADCOM, 0x16cd), .driver_data = BCM58700 }, { PCI_VDEVICE(BROADCOM, 0x16ce), .driver_data = BCM57311 }, { PCI_VDEVICE(BROADCOM, 0x16cf), .driver_data = BCM57312 }, { PCI_VDEVICE(BROADCOM, 0x16d0), .driver_data = BCM57402 }, @@ -6715,6 +6718,9 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) struct bnxt *bp; int rc, max_irqs; + if (pdev->device == 0x16cd && pci_is_bridge(pdev)) + return -ENODEV; + if (version_printed++ == 0) pr_info("%s", version); -- cgit v0.10.2 From 359ebda25aa06fe3a1d028f7e338a849165e661b Mon Sep 17 00:00:00 2001 From: Shmulik Ladkani Date: Mon, 18 Jul 2016 14:49:33 +0300 Subject: net/ipv4: Introduce IPSKB_FRAG_SEGS bit to inet_skb_parm.flags This flag indicates whether fragmentation of segments is allowed. Formerly this policy was hardcoded according to IPSKB_FORWARDED (set by either ip_forward or ipmr_forward). Cc: Hannes Frederic Sowa Cc: Florian Westphal Signed-off-by: Shmulik Ladkani Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/include/net/ip.h b/include/net/ip.h index 08f36cd..9742b92 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -47,6 +47,7 @@ struct inet_skb_parm { #define IPSKB_REROUTED BIT(4) #define IPSKB_DOREDIRECT BIT(5) #define IPSKB_FRAG_PMTU BIT(6) +#define IPSKB_FRAG_SEGS BIT(7) u16 frag_max_size; }; diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index 9f0a7b9..8b4ffd2 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -117,7 +117,7 @@ int ip_forward(struct sk_buff *skb) if (opt->is_strictroute && rt->rt_uses_gateway) goto sr_failed; - IPCB(skb)->flags |= IPSKB_FORWARDED; + IPCB(skb)->flags |= IPSKB_FORWARDED | IPSKB_FRAG_SEGS; mtu = ip_dst_mtu_maybe_forward(&rt->dst, true); if (ip_exceeds_mtu(skb, mtu)) { IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index e23f141..dde37fb 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -223,8 +223,10 @@ static int ip_finish_output_gso(struct net *net, struct sock *sk, struct sk_buff *segs; int ret = 0; - /* common case: locally created skb or seglen is <= mtu */ - if (((IPCB(skb)->flags & IPSKB_FORWARDED) == 0) || + /* common case: fragmentation of segments is not allowed, + * or seglen is <= mtu + */ + if (((IPCB(skb)->flags & IPSKB_FRAG_SEGS) == 0) || skb_gso_validate_mtu(skb, mtu)) return ip_finish_output2(net, sk, skb); diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index e0d76f5..eec2341 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1749,7 +1749,7 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, vif->dev->stats.tx_bytes += skb->len; } - IPCB(skb)->flags |= IPSKB_FORWARDED; + IPCB(skb)->flags |= IPSKB_FORWARDED | IPSKB_FRAG_SEGS; /* RFC1584 teaches, that DVMRP/PIM router must deliver packets locally * not only before forwarding, but after forwarding on all output -- cgit v0.10.2 From b8247f095eddfbfdba0fcecd1e3525a6cdb4b585 Mon Sep 17 00:00:00 2001 From: Shmulik Ladkani Date: Mon, 18 Jul 2016 14:49:34 +0300 Subject: net: ip_finish_output_gso: If skb_gso_network_seglen exceeds MTU, allow segmentation for local udp tunneled skbs Given: - tap0 and vxlan0 are bridged - vxlan0 stacked on eth0, eth0 having small mtu (e.g. 1400) Assume GSO skbs arriving from tap0 having a gso_size as determined by user-provided virtio_net_hdr (e.g. 1460 corresponding to VM mtu of 1500). After encapsulation these skbs have skb_gso_network_seglen that exceed eth0's ip_skb_dst_mtu. These skbs are accidentally passed to ip_finish_output2 AS IS. Alas, each final segment (segmented either by validate_xmit_skb or by hardware UFO) would be larger than eth0 mtu. As a result, those above-mtu segments get dropped on certain networks. This behavior is not aligned with the NON-GSO case: Assume a non-gso 1500-sized IP packet arrives from tap0. After encapsulation, the vxlan datagram is fragmented normally at the ip_finish_output-->ip_fragment code path. The expected behavior for the GSO case would be segmenting the "gso-oversized" skb first, then fragmenting each segment according to dst mtu, and finally passing the resulting fragments to ip_finish_output2. 'ip_finish_output_gso' already supports this "Slowpath" behavior, according to the IPSKB_FRAG_SEGS flag, which is only set during ipv4 forwarding (not set in the bridged case). In order to support the bridged case, we'll mark skbs arriving from an ingress interface that get udp-encaspulated as "allowed to be fragmented", causing their network_seglen to be validated by 'ip_finish_output_gso' (and fragment if needed). Note the TUNNEL_DONT_FRAGMENT tun_flag is still honoured (both in the gso and non-gso cases), which serves users wishing to forbid fragmentation at the udp tunnel endpoint. Cc: Hannes Frederic Sowa Cc: Florian Westphal Signed-off-by: Shmulik Ladkani Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index afd6b59..9d847c3 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -63,6 +63,7 @@ void iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, int pkt_len = skb->len - skb_inner_network_offset(skb); struct net *net = dev_net(rt->dst.dev); struct net_device *dev = skb->dev; + int skb_iif = skb->skb_iif; struct iphdr *iph; int err; @@ -72,6 +73,14 @@ void iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, skb_dst_set(skb, &rt->dst); memset(IPCB(skb), 0, sizeof(*IPCB(skb))); + if (skb_iif && proto == IPPROTO_UDP) { + /* Arrived from an ingress interface and got udp encapuslated. + * The encapsulated network segment length may exceed dst mtu. + * Allow IP Fragmentation of segments. + */ + IPCB(skb)->flags |= IPSKB_FRAG_SEGS; + } + /* Push down and install the IP header. */ skb_push(skb, sizeof(struct iphdr)); skb_reset_network_header(skb); -- cgit v0.10.2 From cbce91cad4ee39070bf3c7873767194e4be88e16 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Mon, 18 Jul 2016 13:02:47 -0700 Subject: bnxt_en: Remove locking around txr->dev_state txr->dev_state was not consistently manipulated with the acquisition of the per-queue lock, after further inspection the lock does not seem necessary, either the value is read as BNXT_DEV_STATE_CLOSING or 0. Reported-by: coverity (CID 1339583) Fixes: c0c050c58d840 ("bnxt_en: New Broadcom ethernet driver.") Signed-off-by: Florian Fainelli Acked-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index b55eef2..8a0165b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -4930,9 +4930,7 @@ static void bnxt_tx_disable(struct bnxt *bp) for (i = 0; i < bp->tx_nr_rings; i++) { txr = &bp->tx_ring[i]; txq = netdev_get_tx_queue(bp->dev, i); - __netif_tx_lock(txq, smp_processor_id()); txr->dev_state = BNXT_DEV_STATE_CLOSING; - __netif_tx_unlock(txq); } } /* Stop all TX queues */ -- cgit v0.10.2 From a725ee3e44e39dab1ec82cc745899a785d2a555e Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Mon, 18 Jul 2016 15:34:49 -0700 Subject: virtio-net: Remove more stack DMA VLAN and MQ control was doing DMA from the stack. Fix it. Cc: Michael S. Tsirkin Cc: "netdev@vger.kernel.org" Signed-off-by: Andy Lutomirski Acked-by: Michael S. Tsirkin Signed-off-by: David S. Miller diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 1dd08d4..1b5f531 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -144,8 +144,10 @@ struct virtnet_info { /* Control VQ buffers: protected by the rtnl lock */ struct virtio_net_ctrl_hdr ctrl_hdr; virtio_net_ctrl_ack ctrl_status; + struct virtio_net_ctrl_mq ctrl_mq; u8 ctrl_promisc; u8 ctrl_allmulti; + u16 ctrl_vid; /* Ethtool settings */ u8 duplex; @@ -1058,14 +1060,13 @@ static void virtnet_ack_link_announce(struct virtnet_info *vi) static int virtnet_set_queues(struct virtnet_info *vi, u16 queue_pairs) { struct scatterlist sg; - struct virtio_net_ctrl_mq s; struct net_device *dev = vi->dev; if (!vi->has_cvq || !virtio_has_feature(vi->vdev, VIRTIO_NET_F_MQ)) return 0; - s.virtqueue_pairs = cpu_to_virtio16(vi->vdev, queue_pairs); - sg_init_one(&sg, &s, sizeof(s)); + vi->ctrl_mq.virtqueue_pairs = cpu_to_virtio16(vi->vdev, queue_pairs); + sg_init_one(&sg, &vi->ctrl_mq, sizeof(vi->ctrl_mq)); if (!virtnet_send_command(vi, VIRTIO_NET_CTRL_MQ, VIRTIO_NET_CTRL_MQ_VQ_PAIRS_SET, &sg)) { @@ -1172,7 +1173,8 @@ static int virtnet_vlan_rx_add_vid(struct net_device *dev, struct virtnet_info *vi = netdev_priv(dev); struct scatterlist sg; - sg_init_one(&sg, &vid, sizeof(vid)); + vi->ctrl_vid = vid; + sg_init_one(&sg, &vi->ctrl_vid, sizeof(vi->ctrl_vid)); if (!virtnet_send_command(vi, VIRTIO_NET_CTRL_VLAN, VIRTIO_NET_CTRL_VLAN_ADD, &sg)) @@ -1186,7 +1188,8 @@ static int virtnet_vlan_rx_kill_vid(struct net_device *dev, struct virtnet_info *vi = netdev_priv(dev); struct scatterlist sg; - sg_init_one(&sg, &vid, sizeof(vid)); + vi->ctrl_vid = vid; + sg_init_one(&sg, &vi->ctrl_vid, sizeof(vi->ctrl_vid)); if (!virtnet_send_command(vi, VIRTIO_NET_CTRL_VLAN, VIRTIO_NET_CTRL_VLAN_DEL, &sg)) -- cgit v0.10.2 From 183fc1537ec39be242dc8b619f71fc11b393d295 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 18 Jul 2016 15:50:58 -0700 Subject: kernel/trace/bpf_trace.c: work around gcc-4.4.4 anon union initialization bug kernel/trace/bpf_trace.c: In function 'bpf_event_output': kernel/trace/bpf_trace.c:312: error: unknown field 'next' specified in initializer kernel/trace/bpf_trace.c:312: warning: missing braces around initializer kernel/trace/bpf_trace.c:312: warning: (near initialization for 'raw.frag.') Fixes: 555c8a8623a3a87 ("bpf: avoid stack copy and use skb ctx for event output") Acked-by: Daniel Borkmann Cc: Alexei Starovoitov Cc: David S. Miller Signed-off-by: Andrew Morton Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index ebfbb7d..a12bbd32 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -309,7 +309,9 @@ u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, }; struct perf_raw_record raw = { .frag = { - .next = ctx_size ? &frag : NULL, + { + .next = ctx_size ? &frag : NULL, + }, .size = meta_size, .data = meta, }, -- cgit v0.10.2 From d51c542b782fac35a9f37a23391f3bee884b7401 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:29 -0400 Subject: net: dsa: mv88e6xxx: remove basic function flags All 88E6xxx Marvell switches (even the old not supported yet 88E6060) have at least an ATU, per-port STP states and VLAN map, to run basic switch functions such as Spanning Tree and port based VLANs. Get rid of the related MV88E6XXX_FLAG_{ATU,PORTSTATE,VLANTABLE} flags, as they are defaults to every chip. This enables STP on 6185 and removes many inconsistencies on others. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 5cb06f7..3feb842 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -1460,9 +1460,6 @@ static void mv88e6xxx_port_stp_state_set(struct dsa_switch *ds, int port, int stp_state; int err; - if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_PORTSTATE)) - return; - switch (state) { case BR_STATE_DISABLED: stp_state = PORT_CONTROL_STATE_DISABLED; @@ -2398,11 +2395,6 @@ static int mv88e6xxx_port_fdb_prepare(struct dsa_switch *ds, int port, const struct switchdev_obj_port_fdb *fdb, struct switchdev_trans *trans) { - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - - if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_ATU)) - return -EOPNOTSUPP; - /* We don't need any dynamic resource from the kernel (yet), * so skip the prepare phase. */ @@ -2418,9 +2410,6 @@ static void mv88e6xxx_port_fdb_add(struct dsa_switch *ds, int port, GLOBAL_ATU_DATA_STATE_UC_STATIC; struct mv88e6xxx_chip *chip = ds_to_priv(ds); - if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_ATU)) - return; - mutex_lock(&chip->reg_lock); if (_mv88e6xxx_port_fdb_load(chip, port, fdb->addr, fdb->vid, state)) netdev_err(ds->ports[port].netdev, @@ -2434,9 +2423,6 @@ static int mv88e6xxx_port_fdb_del(struct dsa_switch *ds, int port, struct mv88e6xxx_chip *chip = ds_to_priv(ds); int ret; - if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_ATU)) - return -EOPNOTSUPP; - mutex_lock(&chip->reg_lock); ret = _mv88e6xxx_port_fdb_load(chip, port, fdb->addr, fdb->vid, GLOBAL_ATU_DATA_STATE_UNUSED); @@ -2542,9 +2528,6 @@ static int mv88e6xxx_port_fdb_dump(struct dsa_switch *ds, int port, u16 fid; int err; - if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_ATU)) - return -EOPNOTSUPP; - mutex_lock(&chip->reg_lock); /* Dump port's default Filtering Information Database (VLAN ID 0) */ @@ -2587,9 +2570,6 @@ static int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port, struct mv88e6xxx_chip *chip = ds_to_priv(ds); int i, err = 0; - if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_VLANTABLE)) - return -EOPNOTSUPP; - mutex_lock(&chip->reg_lock); /* Assign the bridge and remap each port's VLANTable */ @@ -2614,9 +2594,6 @@ static void mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port) struct net_device *bridge = chip->ports[port].bridge_dev; int i; - if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_VLANTABLE)) - return; - mutex_lock(&chip->reg_lock); /* Unassign the bridge and remap each port's VLANTable */ diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 83f0662..2ff62f4 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -374,11 +374,6 @@ enum mv88e6xxx_family { }; enum mv88e6xxx_cap { - /* Address Translation Unit. - * The ATU is used to lookup and learn MAC addresses. See GLOBAL_ATU_OP. - */ - MV88E6XXX_CAP_ATU, - /* Energy Efficient Ethernet. */ MV88E6XXX_CAP_EEE, @@ -394,11 +389,6 @@ enum mv88e6xxx_cap { */ MV88E6XXX_CAP_MULTI_CHIP, - /* Port State Filtering for 802.1D Spanning Tree. - * See PORT_CONTROL_STATE_* values in the PORT_CONTROL register. - */ - MV88E6XXX_CAP_PORTSTATE, - /* PHY Polling Unit. * See GLOBAL_CONTROL_PPU_ENABLE and GLOBAL_STATUS_PPU_POLLING. */ @@ -430,12 +420,6 @@ enum mv88e6xxx_cap { MV88E6XXX_CAP_TEMP, MV88E6XXX_CAP_TEMP_LIMIT, - /* In-chip Port Based VLANs. - * Each port VLANTable register (see PORT_BASE_VLAN) is used to restrict - * the output (or egress) ports to which it is allowed to send frames. - */ - MV88E6XXX_CAP_VLANTABLE, - /* VLAN Table Unit. * The VTU is used to program 802.1Q VLANs. See GLOBAL_VTU_OP. */ @@ -443,11 +427,9 @@ enum mv88e6xxx_cap { }; /* Bitmask of capabilities */ -#define MV88E6XXX_FLAG_ATU BIT(MV88E6XXX_CAP_ATU) #define MV88E6XXX_FLAG_EEE BIT(MV88E6XXX_CAP_EEE) #define MV88E6XXX_FLAG_EEPROM BIT(MV88E6XXX_CAP_EEPROM) #define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) -#define MV88E6XXX_FLAG_PORTSTATE BIT(MV88E6XXX_CAP_PORTSTATE) #define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) #define MV88E6XXX_FLAG_PPU_ACTIVE BIT(MV88E6XXX_CAP_PPU_ACTIVE) #define MV88E6XXX_FLAG_SMI_PHY BIT(MV88E6XXX_CAP_SMI_PHY) @@ -455,22 +437,17 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_SWITCH_MAC BIT(MV88E6XXX_CAP_SWITCH_MAC_WOL_WOF) #define MV88E6XXX_FLAG_TEMP BIT(MV88E6XXX_CAP_TEMP) #define MV88E6XXX_FLAG_TEMP_LIMIT BIT(MV88E6XXX_CAP_TEMP_LIMIT) -#define MV88E6XXX_FLAG_VLANTABLE BIT(MV88E6XXX_CAP_VLANTABLE) #define MV88E6XXX_FLAG_VTU BIT(MV88E6XXX_CAP_VTU) #define MV88E6XXX_FLAGS_FAMILY_6095 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ + (MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ - MV88E6XXX_FLAG_VLANTABLE | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6097 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ + (MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_STU | \ - MV88E6XXX_FLAG_VLANTABLE | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6165 \ @@ -481,51 +458,40 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6185 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ + (MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ - MV88E6XXX_FLAG_VLANTABLE | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6320 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_EEE | \ + (MV88E6XXX_FLAG_EEE | \ MV88E6XXX_FLAG_EEPROM | \ MV88E6XXX_FLAG_MULTI_CHIP | \ - MV88E6XXX_FLAG_PORTSTATE | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_SWITCH_MAC | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ - MV88E6XXX_FLAG_VLANTABLE | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6351 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_MULTI_CHIP | \ - MV88E6XXX_FLAG_PORTSTATE | \ + (MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_SWITCH_MAC | \ MV88E6XXX_FLAG_TEMP | \ - MV88E6XXX_FLAG_VLANTABLE | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6352 \ - (MV88E6XXX_FLAG_ATU | \ - MV88E6XXX_FLAG_EEE | \ + (MV88E6XXX_FLAG_EEE | \ MV88E6XXX_FLAG_EEPROM | \ MV88E6XXX_FLAG_MULTI_CHIP | \ - MV88E6XXX_FLAG_PORTSTATE | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_SWITCH_MAC | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ - MV88E6XXX_FLAG_VLANTABLE | \ MV88E6XXX_FLAG_VTU) struct mv88e6xxx_info { -- cgit v0.10.2 From 9729934c4ff6893456e9a885718b75856e1a42ae Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:30 -0400 Subject: net: dsa: mv88e6xxx: split setup of Global 1 and 2 Separate the setup of Global 1 and Global 2 internal SMI devices and add a flag to describe the presence of this second registers set. Also rearrange the G1 setup in the registers order. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 3feb842..1e39fa6 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -2993,13 +2993,12 @@ static int mv88e6xxx_setup_port(struct mv88e6xxx_chip *chip, int port) return 0; } -static int mv88e6xxx_setup_global(struct mv88e6xxx_chip *chip) +static int mv88e6xxx_g1_setup(struct mv88e6xxx_chip *chip) { struct dsa_switch *ds = chip->ds; u32 upstream_port = dsa_upstream_port(ds); u16 reg; int err; - int i; /* Enable the PHY Polling Unit if present, don't discard any packets, * and mask all interrupt sources. @@ -3040,6 +3039,16 @@ static int mv88e6xxx_setup_global(struct mv88e6xxx_chip *chip) if (err) return err; + /* Clear all the VTU and STU entries */ + err = _mv88e6xxx_vtu_stu_flush(chip); + if (err < 0) + return err; + + /* Clear all ATU entries */ + err = _mv88e6xxx_atu_flush(chip, 0, true); + if (err) + return err; + /* Configure the IP ToS mapping registers. */ err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_IP_PRI_0, 0x0000); if (err) @@ -3071,6 +3080,26 @@ static int mv88e6xxx_setup_global(struct mv88e6xxx_chip *chip) if (err) return err; + /* Clear the statistics counters for all ports */ + err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_STATS_OP, + GLOBAL_STATS_OP_FLUSH_ALL); + if (err) + return err; + + /* Wait for the flush to complete. */ + err = _mv88e6xxx_stats_wait(chip); + if (err) + return err; + + return 0; +} + +static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) +{ + struct dsa_switch *ds = chip->ds; + int err; + int i; + /* Send all frames with destination addresses matching * 01:80:c2:00:00:0x to the CPU port. */ @@ -3174,28 +3203,7 @@ static int mv88e6xxx_setup_global(struct mv88e6xxx_chip *chip) } } - /* Clear the statistics counters for all ports */ - err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_STATS_OP, - GLOBAL_STATS_OP_FLUSH_ALL); - if (err) - return err; - - /* Wait for the flush to complete. */ - err = _mv88e6xxx_stats_wait(chip); - if (err) - return err; - - /* Clear all ATU entries */ - err = _mv88e6xxx_atu_flush(chip, 0, true); - if (err) - return err; - - /* Clear all the VTU and STU entries */ - err = _mv88e6xxx_vtu_stu_flush(chip); - if (err < 0) - return err; - - return err; + return 0; } static int mv88e6xxx_setup(struct dsa_switch *ds) @@ -3216,12 +3224,21 @@ static int mv88e6xxx_setup(struct dsa_switch *ds) if (err) goto unlock; - err = mv88e6xxx_setup_global(chip); + /* Setup Switch Port Registers */ + for (i = 0; i < chip->info->num_ports; i++) { + err = mv88e6xxx_setup_port(chip, i); + if (err) + goto unlock; + } + + /* Setup Switch Global 1 Registers */ + err = mv88e6xxx_g1_setup(chip); if (err) goto unlock; - for (i = 0; i < chip->info->num_ports; i++) { - err = mv88e6xxx_setup_port(chip, i); + /* Setup Switch Global 2 Registers */ + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_GLOBAL2)) { + err = mv88e6xxx_g2_setup(chip); if (err) goto unlock; } diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 2ff62f4..390dac5 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -383,6 +383,11 @@ enum mv88e6xxx_cap { */ MV88E6XXX_CAP_EEPROM, + /* Switch Global 2 Registers. + * The device contains a second set of global 16-bit registers. + */ + MV88E6XXX_CAP_GLOBAL2, + /* Multi-chip Addressing Mode. * Some chips require an indirect SMI access when their SMI device * address is not zero. See SMI_CMD and SMI_DATA. @@ -429,6 +434,7 @@ enum mv88e6xxx_cap { /* Bitmask of capabilities */ #define MV88E6XXX_FLAG_EEE BIT(MV88E6XXX_CAP_EEE) #define MV88E6XXX_FLAG_EEPROM BIT(MV88E6XXX_CAP_EEPROM) +#define MV88E6XXX_FLAG_GLOBAL2 BIT(MV88E6XXX_CAP_GLOBAL2) #define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) #define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) #define MV88E6XXX_FLAG_PPU_ACTIVE BIT(MV88E6XXX_CAP_PPU_ACTIVE) @@ -440,31 +446,36 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_VTU BIT(MV88E6XXX_CAP_VTU) #define MV88E6XXX_FLAGS_FAMILY_6095 \ - (MV88E6XXX_FLAG_MULTI_CHIP | \ + (MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6097 \ - (MV88E6XXX_FLAG_MULTI_CHIP | \ + (MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6165 \ - (MV88E6XXX_FLAG_MULTI_CHIP | \ + (MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_SWITCH_MAC | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6185 \ - (MV88E6XXX_FLAG_MULTI_CHIP | \ + (MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6320 \ (MV88E6XXX_FLAG_EEE | \ MV88E6XXX_FLAG_EEPROM | \ + MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ @@ -474,7 +485,8 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6351 \ - (MV88E6XXX_FLAG_MULTI_CHIP | \ + (MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_STU | \ @@ -485,6 +497,7 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAGS_FAMILY_6352 \ (MV88E6XXX_FLAG_EEE | \ MV88E6XXX_FLAG_EEPROM | \ + MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ -- cgit v0.10.2 From f22ab64123da18b96bf8b3d0801c802c73797a9f Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:31 -0400 Subject: net: dsa: mv88e6xxx: extract device mapping The Device Mapping register is an indirect table access. Provide helpers to access this table and explicit the checking of the new DSA_RTABLE_NONE routing table value. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 1e39fa6..9b2525a 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -216,6 +216,32 @@ static int mv88e6xxx_write(struct mv88e6xxx_chip *chip, return 0; } +/* Indirect write to single pointer-data register with an Update bit */ +static int mv88e6xxx_update(struct mv88e6xxx_chip *chip, int addr, int reg, + u16 update) +{ + u16 val; + int i, err; + + /* Wait until the previous operation is completed */ + for (i = 0; i < 16; ++i) { + err = mv88e6xxx_read(chip, addr, reg, &val); + if (err) + return err; + + if (!(val & BIT(15))) + break; + } + + if (i == 16) + return -ETIMEDOUT; + + /* Set the Update bit to trigger a write operation */ + val = BIT(15) | update; + + return mv88e6xxx_write(chip, addr, reg, val); +} + static int _mv88e6xxx_reg_read(struct mv88e6xxx_chip *chip, int addr, int reg) { u16 val; @@ -3094,9 +3120,39 @@ static int mv88e6xxx_g1_setup(struct mv88e6xxx_chip *chip) return 0; } +static int mv88e6xxx_g2_device_mapping_write(struct mv88e6xxx_chip *chip, + int target, int port) +{ + u16 val = (target << 8) | (port & 0xf); + + return mv88e6xxx_update(chip, REG_GLOBAL2, GLOBAL2_DEVICE_MAPPING, val); +} + +static int mv88e6xxx_g2_set_device_mapping(struct mv88e6xxx_chip *chip) +{ + int target, port; + int err; + + /* Initialize the routing port to the 32 possible target devices */ + for (target = 0; target < 32; ++target) { + port = 0xf; + + if (target < DSA_MAX_SWITCHES) { + port = chip->ds->rtable[target]; + if (port == DSA_RTABLE_NONE) + port = 0xf; + } + + err = mv88e6xxx_g2_device_mapping_write(chip, target, port); + if (err) + break; + } + + return err; +} + static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) { - struct dsa_switch *ds = chip->ds; int err; int i; @@ -3120,20 +3176,9 @@ static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) return err; /* Program the DSA routing table. */ - for (i = 0; i < 32; i++) { - int nexthop = 0x1f; - - if (i != ds->index && i < DSA_MAX_SWITCHES) - nexthop = ds->rtable[i] & 0x1f; - - err = _mv88e6xxx_reg_write( - chip, REG_GLOBAL2, - GLOBAL2_DEVICE_MAPPING, - GLOBAL2_DEVICE_MAPPING_UPDATE | - (i << GLOBAL2_DEVICE_MAPPING_TARGET_SHIFT) | nexthop); - if (err) - return err; - } + err = mv88e6xxx_g2_set_device_mapping(chip); + if (err) + return err; /* Clear all trunk masks. */ for (i = 0; i < 8; i++) { -- cgit v0.10.2 From 5154041fa717fd8e4ef8c8144c6eaba9392bdaec Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:32 -0400 Subject: net: dsa: mv88e6xxx: extract trunk mapping The Trunk Mask and Trunk Mapping registers are two Global 2 indirect accesses to trunking configuration. Add helpers for these tables and simplify the Global 2 setup. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 9b2525a..d18e5c8 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -3151,6 +3151,49 @@ static int mv88e6xxx_g2_set_device_mapping(struct mv88e6xxx_chip *chip) return err; } +static int mv88e6xxx_g2_trunk_mask_write(struct mv88e6xxx_chip *chip, int num, + bool hask, u16 mask) +{ + const u16 port_mask = BIT(chip->info->num_ports) - 1; + u16 val = (num << 12) | (mask & port_mask); + + if (hask) + val |= GLOBAL2_TRUNK_MASK_HASK; + + return mv88e6xxx_update(chip, REG_GLOBAL2, GLOBAL2_TRUNK_MASK, val); +} + +static int mv88e6xxx_g2_trunk_mapping_write(struct mv88e6xxx_chip *chip, int id, + u16 map) +{ + const u16 port_mask = BIT(chip->info->num_ports) - 1; + u16 val = (id << 11) | (map & port_mask); + + return mv88e6xxx_update(chip, REG_GLOBAL2, GLOBAL2_TRUNK_MAPPING, val); +} + +static int mv88e6xxx_g2_clear_trunk(struct mv88e6xxx_chip *chip) +{ + const u16 port_mask = BIT(chip->info->num_ports) - 1; + int i, err; + + /* Clear all eight possible Trunk Mask vectors */ + for (i = 0; i < 8; ++i) { + err = mv88e6xxx_g2_trunk_mask_write(chip, i, false, port_mask); + if (err) + return err; + } + + /* Clear all sixteen possible Trunk ID routing vectors */ + for (i = 0; i < 16; ++i) { + err = mv88e6xxx_g2_trunk_mapping_write(chip, i, 0); + if (err) + return err; + } + + return 0; +} + static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) { int err; @@ -3180,27 +3223,10 @@ static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) if (err) return err; - /* Clear all trunk masks. */ - for (i = 0; i < 8; i++) { - err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, - GLOBAL2_TRUNK_MASK, - 0x8000 | - (i << GLOBAL2_TRUNK_MASK_NUM_SHIFT) | - ((1 << chip->info->num_ports) - 1)); - if (err) - return err; - } - - /* Clear all trunk mappings. */ - for (i = 0; i < 16; i++) { - err = _mv88e6xxx_reg_write( - chip, REG_GLOBAL2, - GLOBAL2_TRUNK_MAPPING, - GLOBAL2_TRUNK_MAPPING_UPDATE | - (i << GLOBAL2_TRUNK_MAPPING_ID_SHIFT)); - if (err) - return err; - } + /* Clear all trunk masks and mapping. */ + err = mv88e6xxx_g2_clear_trunk(chip); + if (err) + return err; if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 390dac5..876d9ea 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -294,6 +294,7 @@ #define GLOBAL2_TRUNK_MASK 0x07 #define GLOBAL2_TRUNK_MASK_UPDATE BIT(15) #define GLOBAL2_TRUNK_MASK_NUM_SHIFT 12 +#define GLOBAL2_TRUNK_MASK_HASK BIT(11) #define GLOBAL2_TRUNK_MAPPING 0x08 #define GLOBAL2_TRUNK_MAPPING_UPDATE BIT(15) #define GLOBAL2_TRUNK_MAPPING_ID_SHIFT 11 -- cgit v0.10.2 From 47395ed28056a7ac7fbd9e7ff06bbbd66d01e256 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:33 -0400 Subject: net: dsa: mv88e6xxx: add cap for MGMT Enables bits Some switches provide a Rsvd2CPU mechanism used to choose which of the 16 reserved multicast destination addresses matching 01:80:c2:00:00:0x should be considered as MGMT and thus forwarded to the CPU port. Other switches extend this mechanism to also configure as MGMT the additional 16 reserved multicast addresses matching 01:80:c2:00:00:2x. This mechanism is exposed via two registers in Global 2, and an Rsvd2CPU enable bit in the management register. Newer chip (such as 88E6390) has replaced these registers with a new indirect MGMT mechanism in Global 1. The patch adds two MV88E6XXX_FLAG_G2_MGMT_EN_{0,2}X flags to describe the presence of these Global 2 registers. If 88E6390 support is added, a MV88E6XXX_FLAG_G1_MGMT_CTRL flag will be needed to setup Rsvd2CPU. Note: all switches still support in parallel the ATU Load operation with an MGMT Entry State to forward such frames in a less convenient way. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index d18e5c8..cf98884 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -3196,25 +3196,40 @@ static int mv88e6xxx_g2_clear_trunk(struct mv88e6xxx_chip *chip) static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) { + u16 reg; int err; int i; - /* Send all frames with destination addresses matching - * 01:80:c2:00:00:0x to the CPU port. - */ - err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_MGMT_EN_0X, - 0xffff); - if (err) - return err; + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_G2_MGMT_EN_2X)) { + /* Consider the frames with reserved multicast destination + * addresses matching 01:80:c2:00:00:2x as MGMT. + */ + err = mv88e6xxx_write(chip, REG_GLOBAL2, GLOBAL2_MGMT_EN_2X, + 0xffff); + if (err) + return err; + } + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_G2_MGMT_EN_0X)) { + /* Consider the frames with reserved multicast destination + * addresses matching 01:80:c2:00:00:0x as MGMT. + */ + err = mv88e6xxx_write(chip, REG_GLOBAL2, GLOBAL2_MGMT_EN_0X, + 0xffff); + if (err) + return err; + } /* Ignore removed tag data on doubly tagged packets, disable * flow control messages, force flow control priority to the * highest, and send all special multicast frames to the CPU * port at the highest priority. */ - err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_SWITCH_MGMT, - 0x7 | GLOBAL2_SWITCH_MGMT_RSVD2CPU | 0x70 | - GLOBAL2_SWITCH_MGMT_FORCE_FLOW_CTRL_PRI); + reg = GLOBAL2_SWITCH_MGMT_FORCE_FLOW_CTRL_PRI | (0x7 << 4); + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_G2_MGMT_EN_0X) || + mv88e6xxx_has(chip, MV88E6XXX_FLAG_G2_MGMT_EN_2X)) + reg |= GLOBAL2_SWITCH_MGMT_RSVD2CPU | 0x7; + err = mv88e6xxx_write(chip, REG_GLOBAL2, GLOBAL2_SWITCH_MGMT, reg); if (err) return err; @@ -3231,14 +3246,6 @@ static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || mv88e6xxx_6320_family(chip)) { - /* Send all frames with destination addresses matching - * 01:80:c2:00:00:2x to the CPU port. - */ - err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, - GLOBAL2_MGMT_EN_2X, 0xffff); - if (err) - return err; - /* Initialise cross-chip port VLAN table to reset * defaults. */ diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 876d9ea..d13b0b5 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -388,6 +388,8 @@ enum mv88e6xxx_cap { * The device contains a second set of global 16-bit registers. */ MV88E6XXX_CAP_GLOBAL2, + MV88E6XXX_CAP_G2_MGMT_EN_2X, /* (0x02) MGMT Enable Register 2x */ + MV88E6XXX_CAP_G2_MGMT_EN_0X, /* (0x03) MGMT Enable Register 0x */ /* Multi-chip Addressing Mode. * Some chips require an indirect SMI access when their SMI device @@ -436,6 +438,8 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_EEE BIT(MV88E6XXX_CAP_EEE) #define MV88E6XXX_FLAG_EEPROM BIT(MV88E6XXX_CAP_EEPROM) #define MV88E6XXX_FLAG_GLOBAL2 BIT(MV88E6XXX_CAP_GLOBAL2) +#define MV88E6XXX_FLAG_G2_MGMT_EN_2X BIT(MV88E6XXX_CAP_G2_MGMT_EN_2X) +#define MV88E6XXX_FLAG_G2_MGMT_EN_0X BIT(MV88E6XXX_CAP_G2_MGMT_EN_0X) #define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) #define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) #define MV88E6XXX_FLAG_PPU_ACTIVE BIT(MV88E6XXX_CAP_PPU_ACTIVE) @@ -448,12 +452,15 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAGS_FAMILY_6095 \ (MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_VTU) #define MV88E6XXX_FLAGS_FAMILY_6097 \ (MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ + MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_STU | \ @@ -461,6 +468,8 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAGS_FAMILY_6165 \ (MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ + MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_SWITCH_MAC | \ @@ -469,6 +478,7 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAGS_FAMILY_6185 \ (MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_VTU) @@ -477,6 +487,8 @@ enum mv88e6xxx_cap { (MV88E6XXX_FLAG_EEE | \ MV88E6XXX_FLAG_EEPROM | \ MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ + MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ @@ -487,6 +499,8 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAGS_FAMILY_6351 \ (MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ + MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ @@ -499,6 +513,8 @@ enum mv88e6xxx_cap { (MV88E6XXX_FLAG_EEE | \ MV88E6XXX_FLAG_EEPROM | \ MV88E6XXX_FLAG_GLOBAL2 | \ + MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ + MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ -- cgit v0.10.2 From 3b4caa1b1ce95eaa405c8a27443fe7ddafb563a0 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:34 -0400 Subject: net: dsa: mv88e6xxx: rework Switch MAC setter Switches such as 88E6185 as 3 Switch MAC registers in Global 1. Newer chips such as 88E6352 have freed these registers in favor of an indirect access in a Switch MAC/WoL/WoF register in Global 2. Explicit this difference with G1 and G2 helpers and flags. Also, note that this indirect access is a single-register which doesn't require to wait for the operation to complete (like Switch MAC, Trunk Mapping, etc.), in contrary to multi-registers indirect accesses with several operations and a busy bit. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index cf98884..d1b4a7a 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -283,68 +283,6 @@ static int mv88e6xxx_reg_write(struct mv88e6xxx_chip *chip, int addr, return ret; } -static int mv88e6xxx_set_addr_direct(struct dsa_switch *ds, u8 *addr) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int err; - - err = mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_MAC_01, - (addr[0] << 8) | addr[1]); - if (err) - return err; - - err = mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_MAC_23, - (addr[2] << 8) | addr[3]); - if (err) - return err; - - return mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_MAC_45, - (addr[4] << 8) | addr[5]); -} - -static int mv88e6xxx_set_addr_indirect(struct dsa_switch *ds, u8 *addr) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int ret; - int i; - - for (i = 0; i < 6; i++) { - int j; - - /* Write the MAC address byte. */ - ret = mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_SWITCH_MAC, - GLOBAL2_SWITCH_MAC_BUSY | - (i << 8) | addr[i]); - if (ret) - return ret; - - /* Wait for the write to complete. */ - for (j = 0; j < 16; j++) { - ret = mv88e6xxx_reg_read(chip, REG_GLOBAL2, - GLOBAL2_SWITCH_MAC); - if (ret < 0) - return ret; - - if ((ret & GLOBAL2_SWITCH_MAC_BUSY) == 0) - break; - } - if (j == 16) - return -ETIMEDOUT; - } - - return 0; -} - -static int mv88e6xxx_set_addr(struct dsa_switch *ds, u8 *addr) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - - if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_SWITCH_MAC)) - return mv88e6xxx_set_addr_indirect(ds, addr); - else - return mv88e6xxx_set_addr_direct(ds, addr); -} - static int mv88e6xxx_mdio_read_direct(struct mv88e6xxx_chip *chip, int addr, int regnum) { @@ -3019,6 +2957,24 @@ static int mv88e6xxx_setup_port(struct mv88e6xxx_chip *chip, int port) return 0; } +static int mv88e6xxx_g1_set_switch_mac(struct mv88e6xxx_chip *chip, u8 *addr) +{ + int err; + + err = mv88e6xxx_write(chip, REG_GLOBAL, GLOBAL_MAC_01, + (addr[0] << 8) | addr[1]); + if (err) + return err; + + err = mv88e6xxx_write(chip, REG_GLOBAL, GLOBAL_MAC_23, + (addr[2] << 8) | addr[3]); + if (err) + return err; + + return mv88e6xxx_write(chip, REG_GLOBAL, GLOBAL_MAC_45, + (addr[4] << 8) | addr[5]); +} + static int mv88e6xxx_g1_setup(struct mv88e6xxx_chip *chip) { struct dsa_switch *ds = chip->ds; @@ -3194,6 +3150,28 @@ static int mv88e6xxx_g2_clear_trunk(struct mv88e6xxx_chip *chip) return 0; } +/* Indirect write to the Switch MAC/WoL/WoF register */ +static int mv88e6xxx_g2_switch_mac_write(struct mv88e6xxx_chip *chip, + unsigned int pointer, u8 data) +{ + u16 val = (pointer << 8) | data; + + return mv88e6xxx_update(chip, REG_GLOBAL2, GLOBAL2_SWITCH_MAC, val); +} + +static int mv88e6xxx_g2_set_switch_mac(struct mv88e6xxx_chip *chip, u8 *addr) +{ + int i, err; + + for (i = 0; i < 6; i++) { + err = mv88e6xxx_g2_switch_mac_write(chip, i, addr[i]); + if (err) + break; + } + + return err; +} + static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) { u16 reg; @@ -3327,6 +3305,24 @@ unlock: return err; } +static int mv88e6xxx_set_addr(struct dsa_switch *ds, u8 *addr) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int err; + + mutex_lock(&chip->reg_lock); + + /* Has an indirect Switch MAC/WoL/WoF register in Global 2? */ + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_G2_SWITCH_MAC)) + err = mv88e6xxx_g2_set_switch_mac(chip, addr); + else + err = mv88e6xxx_g1_set_switch_mac(chip, addr); + + mutex_unlock(&chip->reg_lock); + + return err; +} + static int mv88e6xxx_mdio_page_read(struct dsa_switch *ds, int port, int page, int reg) { diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index d13b0b5..da61db4 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -303,7 +303,6 @@ #define GLOBAL2_PVT_ADDR 0x0b #define GLOBAL2_PVT_DATA 0x0c #define GLOBAL2_SWITCH_MAC 0x0d -#define GLOBAL2_SWITCH_MAC_BUSY BIT(15) #define GLOBAL2_ATU_STATS 0x0e #define GLOBAL2_PRIO_OVERRIDE 0x0f #define GLOBAL2_PRIO_OVERRIDE_FORCE_SNOOP BIT(7) @@ -390,6 +389,7 @@ enum mv88e6xxx_cap { MV88E6XXX_CAP_GLOBAL2, MV88E6XXX_CAP_G2_MGMT_EN_2X, /* (0x02) MGMT Enable Register 2x */ MV88E6XXX_CAP_G2_MGMT_EN_0X, /* (0x03) MGMT Enable Register 0x */ + MV88E6XXX_CAP_G2_SWITCH_MAC, /* (0x0d) Switch MAC/WoL/WoF */ /* Multi-chip Addressing Mode. * Some chips require an indirect SMI access when their SMI device @@ -415,13 +415,6 @@ enum mv88e6xxx_cap { */ MV88E6XXX_CAP_STU, - /* Switch MAC/WoL/WoF register. - * This requires an indirect access to set the switch MAC address - * through GLOBAL2_SWITCH_MAC, otherwise GLOBAL_MAC_01, GLOBAL_MAC_23, - * and GLOBAL_MAC_45 are used with a direct access. - */ - MV88E6XXX_CAP_SWITCH_MAC_WOL_WOF, - /* Internal temperature sensor. * Available from any enabled port's PHY register 26, page 6. */ @@ -440,12 +433,12 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_GLOBAL2 BIT(MV88E6XXX_CAP_GLOBAL2) #define MV88E6XXX_FLAG_G2_MGMT_EN_2X BIT(MV88E6XXX_CAP_G2_MGMT_EN_2X) #define MV88E6XXX_FLAG_G2_MGMT_EN_0X BIT(MV88E6XXX_CAP_G2_MGMT_EN_0X) +#define MV88E6XXX_FLAG_G2_SWITCH_MAC BIT(MV88E6XXX_CAP_G2_SWITCH_MAC) #define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) #define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) #define MV88E6XXX_FLAG_PPU_ACTIVE BIT(MV88E6XXX_CAP_PPU_ACTIVE) #define MV88E6XXX_FLAG_SMI_PHY BIT(MV88E6XXX_CAP_SMI_PHY) #define MV88E6XXX_FLAG_STU BIT(MV88E6XXX_CAP_STU) -#define MV88E6XXX_FLAG_SWITCH_MAC BIT(MV88E6XXX_CAP_SWITCH_MAC_WOL_WOF) #define MV88E6XXX_FLAG_TEMP BIT(MV88E6XXX_CAP_TEMP) #define MV88E6XXX_FLAG_TEMP_LIMIT BIT(MV88E6XXX_CAP_TEMP_LIMIT) #define MV88E6XXX_FLAG_VTU BIT(MV88E6XXX_CAP_VTU) @@ -470,9 +463,9 @@ enum mv88e6xxx_cap { (MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ + MV88E6XXX_FLAG_G2_SWITCH_MAC | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_STU | \ - MV88E6XXX_FLAG_SWITCH_MAC | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_VTU) @@ -489,10 +482,10 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ + MV88E6XXX_FLAG_G2_SWITCH_MAC | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ - MV88E6XXX_FLAG_SWITCH_MAC | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ MV88E6XXX_FLAG_VTU) @@ -501,11 +494,11 @@ enum mv88e6xxx_cap { (MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ + MV88E6XXX_FLAG_G2_SWITCH_MAC | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_STU | \ - MV88E6XXX_FLAG_SWITCH_MAC | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_VTU) @@ -515,11 +508,11 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ + MV88E6XXX_FLAG_G2_SWITCH_MAC | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_STU | \ - MV88E6XXX_FLAG_SWITCH_MAC | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ MV88E6XXX_FLAG_VTU) -- cgit v0.10.2 From 63ed880dea23ddbe4a98c8f7c13ea5ad42635987 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:35 -0400 Subject: net: dsa: mv88e6xxx: add cap for PVT Add flags to describe the presence of Cross-chip Port VLAN Table (PVT) related registers and simplify the setup of Global 2. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index d1b4a7a..f346ad7 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -3221,17 +3221,17 @@ static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) if (err) return err; - if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || - mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || - mv88e6xxx_6320_family(chip)) { - /* Initialise cross-chip port VLAN table to reset - * defaults. - */ - err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, - GLOBAL2_PVT_ADDR, 0x9000); + if (mv88e6xxx_has(chip, MV88E6XXX_FLAGS_PVT)) { + /* Initialize Cross-chip Port VLAN Table to reset defaults */ + err = mv88e6xxx_write(chip, REG_GLOBAL2, GLOBAL2_PVT_ADDR, + GLOBAL2_PVT_ADDR_OP_INIT_ONES); if (err) return err; + } + if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || + mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || + mv88e6xxx_6320_family(chip)) { /* Clear the priority override table. */ for (i = 0; i < 16; i++) { err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index da61db4..4e03650 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -301,6 +301,10 @@ #define GLOBAL2_INGRESS_OP 0x09 #define GLOBAL2_INGRESS_DATA 0x0a #define GLOBAL2_PVT_ADDR 0x0b +#define GLOBAL2_PVT_ADDR_BUSY BIT(15) +#define GLOBAL2_PVT_ADDR_OP_INIT_ONES ((0x01 << 12) | GLOBAL2_PVT_ADDR_BUSY) +#define GLOBAL2_PVT_ADDR_OP_WRITE_PVLAN ((0x03 << 12) | GLOBAL2_PVT_ADDR_BUSY) +#define GLOBAL2_PVT_ADDR_OP_READ ((0x04 << 12) | GLOBAL2_PVT_ADDR_BUSY) #define GLOBAL2_PVT_DATA 0x0c #define GLOBAL2_SWITCH_MAC 0x0d #define GLOBAL2_ATU_STATS 0x0e @@ -389,6 +393,8 @@ enum mv88e6xxx_cap { MV88E6XXX_CAP_GLOBAL2, MV88E6XXX_CAP_G2_MGMT_EN_2X, /* (0x02) MGMT Enable Register 2x */ MV88E6XXX_CAP_G2_MGMT_EN_0X, /* (0x03) MGMT Enable Register 0x */ + MV88E6XXX_CAP_G2_PVT_ADDR, /* (0x0b) Cross Chip Port VLAN Addr */ + MV88E6XXX_CAP_G2_PVT_DATA, /* (0x0c) Cross Chip Port VLAN Data */ MV88E6XXX_CAP_G2_SWITCH_MAC, /* (0x0d) Switch MAC/WoL/WoF */ /* Multi-chip Addressing Mode. @@ -433,6 +439,8 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_GLOBAL2 BIT(MV88E6XXX_CAP_GLOBAL2) #define MV88E6XXX_FLAG_G2_MGMT_EN_2X BIT(MV88E6XXX_CAP_G2_MGMT_EN_2X) #define MV88E6XXX_FLAG_G2_MGMT_EN_0X BIT(MV88E6XXX_CAP_G2_MGMT_EN_0X) +#define MV88E6XXX_FLAG_G2_PVT_ADDR BIT(MV88E6XXX_CAP_G2_PVT_ADDR) +#define MV88E6XXX_FLAG_G2_PVT_DATA BIT(MV88E6XXX_CAP_G2_PVT_DATA) #define MV88E6XXX_FLAG_G2_SWITCH_MAC BIT(MV88E6XXX_CAP_G2_SWITCH_MAC) #define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) #define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) @@ -443,6 +451,11 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_TEMP_LIMIT BIT(MV88E6XXX_CAP_TEMP_LIMIT) #define MV88E6XXX_FLAG_VTU BIT(MV88E6XXX_CAP_VTU) +/* Cross-chip Port VLAN Table */ +#define MV88E6XXX_FLAGS_PVT \ + (MV88E6XXX_FLAG_G2_PVT_ADDR | \ + MV88E6XXX_FLAG_G2_PVT_DATA) + #define MV88E6XXX_FLAGS_FAMILY_6095 \ (MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ @@ -457,7 +470,8 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_STU | \ - MV88E6XXX_FLAG_VTU) + MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_PVT) #define MV88E6XXX_FLAGS_FAMILY_6165 \ (MV88E6XXX_FLAG_GLOBAL2 | \ @@ -467,7 +481,8 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_TEMP | \ - MV88E6XXX_FLAG_VTU) + MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_PVT) #define MV88E6XXX_FLAGS_FAMILY_6185 \ (MV88E6XXX_FLAG_GLOBAL2 | \ @@ -488,7 +503,8 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ - MV88E6XXX_FLAG_VTU) + MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_PVT) #define MV88E6XXX_FLAGS_FAMILY_6351 \ (MV88E6XXX_FLAG_GLOBAL2 | \ @@ -500,7 +516,8 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_SMI_PHY | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_TEMP | \ - MV88E6XXX_FLAG_VTU) + MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_PVT) #define MV88E6XXX_FLAGS_FAMILY_6352 \ (MV88E6XXX_FLAG_EEE | \ @@ -515,7 +532,8 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ - MV88E6XXX_FLAG_VTU) + MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_PVT) struct mv88e6xxx_info { enum mv88e6xxx_family family; -- cgit v0.10.2 From 9bda889faed61c2b2f7c2356b6a9e577c9812cf0 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:36 -0400 Subject: net: dsa: mv88e6xxx: add cap for Priority Override Add flags and helpers to describe the presence of Priority Override Table (POT) related registers and simplify the setup of Global 2. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index f346ad7..3422792 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -3172,6 +3172,28 @@ static int mv88e6xxx_g2_set_switch_mac(struct mv88e6xxx_chip *chip, u8 *addr) return err; } +static int mv88e6xxx_g2_pot_write(struct mv88e6xxx_chip *chip, int pointer, + u8 data) +{ + u16 val = (pointer << 8) | (data & 0x7); + + return mv88e6xxx_update(chip, REG_GLOBAL2, GLOBAL2_PRIO_OVERRIDE, val); +} + +static int mv88e6xxx_g2_clear_pot(struct mv88e6xxx_chip *chip) +{ + int i, err; + + /* Clear all sixteen possible Priority Override entries */ + for (i = 0; i < 16; i++) { + err = mv88e6xxx_g2_pot_write(chip, i, 0); + if (err) + break; + } + + return err; +} + static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) { u16 reg; @@ -3229,17 +3251,11 @@ static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) return err; } - if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || - mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || - mv88e6xxx_6320_family(chip)) { + if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_G2_POT)) { /* Clear the priority override table. */ - for (i = 0; i < 16; i++) { - err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, - GLOBAL2_PRIO_OVERRIDE, - 0x8000 | (i << 8)); - if (err) - return err; - } + err = mv88e6xxx_g2_clear_pot(chip); + if (err) + return err; } if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 4e03650..06b11fb 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -396,6 +396,7 @@ enum mv88e6xxx_cap { MV88E6XXX_CAP_G2_PVT_ADDR, /* (0x0b) Cross Chip Port VLAN Addr */ MV88E6XXX_CAP_G2_PVT_DATA, /* (0x0c) Cross Chip Port VLAN Data */ MV88E6XXX_CAP_G2_SWITCH_MAC, /* (0x0d) Switch MAC/WoL/WoF */ + MV88E6XXX_CAP_G2_POT, /* (0x0f) Priority Override Table */ /* Multi-chip Addressing Mode. * Some chips require an indirect SMI access when their SMI device @@ -442,6 +443,7 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_G2_PVT_ADDR BIT(MV88E6XXX_CAP_G2_PVT_ADDR) #define MV88E6XXX_FLAG_G2_PVT_DATA BIT(MV88E6XXX_CAP_G2_PVT_DATA) #define MV88E6XXX_FLAG_G2_SWITCH_MAC BIT(MV88E6XXX_CAP_G2_SWITCH_MAC) +#define MV88E6XXX_FLAG_G2_POT BIT(MV88E6XXX_CAP_G2_POT) #define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) #define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) #define MV88E6XXX_FLAG_PPU_ACTIVE BIT(MV88E6XXX_CAP_PPU_ACTIVE) @@ -467,6 +469,7 @@ enum mv88e6xxx_cap { (MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ + MV88E6XXX_FLAG_G2_POT | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_STU | \ @@ -478,6 +481,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_G2_SWITCH_MAC | \ + MV88E6XXX_FLAG_G2_POT | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_TEMP | \ @@ -498,6 +502,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_G2_SWITCH_MAC | \ + MV88E6XXX_FLAG_G2_POT | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ @@ -511,6 +516,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_G2_SWITCH_MAC | \ + MV88E6XXX_FLAG_G2_POT | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ @@ -526,6 +532,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ MV88E6XXX_FLAG_G2_SWITCH_MAC | \ + MV88E6XXX_FLAG_G2_POT | \ MV88E6XXX_FLAG_MULTI_CHIP | \ MV88E6XXX_FLAG_PPU_ACTIVE | \ MV88E6XXX_FLAG_SMI_PHY | \ -- cgit v0.10.2 From 8ec61c7f7c6c75943e8c785729bd8de3440d158c Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:37 -0400 Subject: net: dsa: mv88e6xxx: add cap for IRL Add capability flags to describe the presence of Ingress Rate Limit unit registers and an helper function to clear it. In the meantime, fix a few harmless issues: - 6185 and 6095 don't have such registers (reserved) - the previous code didn't wait for the IRL operation to complete Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 3422792..bc26b4f 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -3150,6 +3150,29 @@ static int mv88e6xxx_g2_clear_trunk(struct mv88e6xxx_chip *chip) return 0; } +static int mv88e6xxx_g2_clear_irl(struct mv88e6xxx_chip *chip) +{ + int port, err; + + /* Init all Ingress Rate Limit resources of all ports */ + for (port = 0; port < chip->info->num_ports; ++port) { + /* XXX newer chips (like 88E6390) have different 2-bit ops */ + err = mv88e6xxx_write(chip, REG_GLOBAL2, GLOBAL2_IRL_CMD, + GLOBAL2_IRL_CMD_OP_INIT_ALL | + (port << 8)); + if (err) + break; + + /* Wait for the operation to complete */ + err = _mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_IRL_CMD, + GLOBAL2_IRL_CMD_BUSY); + if (err) + break; + } + + return err; +} + /* Indirect write to the Switch MAC/WoL/WoF register */ static int mv88e6xxx_g2_switch_mac_write(struct mv88e6xxx_chip *chip, unsigned int pointer, u8 data) @@ -3198,7 +3221,6 @@ static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) { u16 reg; int err; - int i; if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_G2_MGMT_EN_2X)) { /* Consider the frames with reserved multicast destination @@ -3243,6 +3265,15 @@ static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) if (err) return err; + if (mv88e6xxx_has(chip, MV88E6XXX_FLAGS_IRL)) { + /* Disable ingress rate limiting by resetting all per port + * ingress rate limit resources to their initial state. + */ + err = mv88e6xxx_g2_clear_irl(chip); + if (err) + return err; + } + if (mv88e6xxx_has(chip, MV88E6XXX_FLAGS_PVT)) { /* Initialize Cross-chip Port VLAN Table to reset defaults */ err = mv88e6xxx_write(chip, REG_GLOBAL2, GLOBAL2_PVT_ADDR, @@ -3258,23 +3289,6 @@ static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) return err; } - if (mv88e6xxx_6352_family(chip) || mv88e6xxx_6351_family(chip) || - mv88e6xxx_6165_family(chip) || mv88e6xxx_6097_family(chip) || - mv88e6xxx_6185_family(chip) || mv88e6xxx_6095_family(chip) || - mv88e6xxx_6320_family(chip)) { - /* Disable ingress rate limiting by resetting all - * ingress rate limit registers to their initial - * state. - */ - for (i = 0; i < chip->info->num_ports; i++) { - err = _mv88e6xxx_reg_write(chip, REG_GLOBAL2, - GLOBAL2_INGRESS_OP, - 0x9000 | (i << 8)); - if (err) - return err; - } - } - return 0; } diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 06b11fb..9ea5363 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -298,8 +298,13 @@ #define GLOBAL2_TRUNK_MAPPING 0x08 #define GLOBAL2_TRUNK_MAPPING_UPDATE BIT(15) #define GLOBAL2_TRUNK_MAPPING_ID_SHIFT 11 -#define GLOBAL2_INGRESS_OP 0x09 -#define GLOBAL2_INGRESS_DATA 0x0a +#define GLOBAL2_IRL_CMD 0x09 +#define GLOBAL2_IRL_CMD_BUSY BIT(15) +#define GLOBAL2_IRL_CMD_OP_INIT_ALL ((0x001 << 12) | GLOBAL2_IRL_CMD_BUSY) +#define GLOBAL2_IRL_CMD_OP_INIT_SEL ((0x010 << 12) | GLOBAL2_IRL_CMD_BUSY) +#define GLOBAL2_IRL_CMD_OP_WRITE_SEL ((0x011 << 12) | GLOBAL2_IRL_CMD_BUSY) +#define GLOBAL2_IRL_CMD_OP_READ_SEL ((0x100 << 12) | GLOBAL2_IRL_CMD_BUSY) +#define GLOBAL2_IRL_DATA 0x0a #define GLOBAL2_PVT_ADDR 0x0b #define GLOBAL2_PVT_ADDR_BUSY BIT(15) #define GLOBAL2_PVT_ADDR_OP_INIT_ONES ((0x01 << 12) | GLOBAL2_PVT_ADDR_BUSY) @@ -393,6 +398,8 @@ enum mv88e6xxx_cap { MV88E6XXX_CAP_GLOBAL2, MV88E6XXX_CAP_G2_MGMT_EN_2X, /* (0x02) MGMT Enable Register 2x */ MV88E6XXX_CAP_G2_MGMT_EN_0X, /* (0x03) MGMT Enable Register 0x */ + MV88E6XXX_CAP_G2_IRL_CMD, /* (0x09) Ingress Rate Command */ + MV88E6XXX_CAP_G2_IRL_DATA, /* (0x0a) Ingress Rate Data */ MV88E6XXX_CAP_G2_PVT_ADDR, /* (0x0b) Cross Chip Port VLAN Addr */ MV88E6XXX_CAP_G2_PVT_DATA, /* (0x0c) Cross Chip Port VLAN Data */ MV88E6XXX_CAP_G2_SWITCH_MAC, /* (0x0d) Switch MAC/WoL/WoF */ @@ -440,6 +447,8 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_GLOBAL2 BIT(MV88E6XXX_CAP_GLOBAL2) #define MV88E6XXX_FLAG_G2_MGMT_EN_2X BIT(MV88E6XXX_CAP_G2_MGMT_EN_2X) #define MV88E6XXX_FLAG_G2_MGMT_EN_0X BIT(MV88E6XXX_CAP_G2_MGMT_EN_0X) +#define MV88E6XXX_FLAG_G2_IRL_CMD BIT(MV88E6XXX_CAP_G2_IRL_CMD) +#define MV88E6XXX_FLAG_G2_IRL_DATA BIT(MV88E6XXX_CAP_G2_IRL_DATA) #define MV88E6XXX_FLAG_G2_PVT_ADDR BIT(MV88E6XXX_CAP_G2_PVT_ADDR) #define MV88E6XXX_FLAG_G2_PVT_DATA BIT(MV88E6XXX_CAP_G2_PVT_DATA) #define MV88E6XXX_FLAG_G2_SWITCH_MAC BIT(MV88E6XXX_CAP_G2_SWITCH_MAC) @@ -453,6 +462,11 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_TEMP_LIMIT BIT(MV88E6XXX_CAP_TEMP_LIMIT) #define MV88E6XXX_FLAG_VTU BIT(MV88E6XXX_CAP_VTU) +/* Ingress Rate Limit unit */ +#define MV88E6XXX_FLAGS_IRL \ + (MV88E6XXX_FLAG_G2_IRL_CMD | \ + MV88E6XXX_FLAG_G2_IRL_DATA) + /* Cross-chip Port VLAN Table */ #define MV88E6XXX_FLAGS_PVT \ (MV88E6XXX_FLAG_G2_PVT_ADDR | \ @@ -474,6 +488,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_PPU | \ MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_IRL | \ MV88E6XXX_FLAGS_PVT) #define MV88E6XXX_FLAGS_FAMILY_6165 \ @@ -486,6 +501,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_IRL | \ MV88E6XXX_FLAGS_PVT) #define MV88E6XXX_FLAGS_FAMILY_6185 \ @@ -509,6 +525,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_IRL | \ MV88E6XXX_FLAGS_PVT) #define MV88E6XXX_FLAGS_FAMILY_6351 \ @@ -523,6 +540,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_STU | \ MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_IRL | \ MV88E6XXX_FLAGS_PVT) #define MV88E6XXX_FLAGS_FAMILY_6352 \ @@ -540,6 +558,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_IRL | \ MV88E6XXX_FLAGS_PVT) struct mv88e6xxx_info { -- cgit v0.10.2 From 34a79f63bbe49c888f95e75dd759685a238556b6 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:38 -0400 Subject: net: dsa: support switchdev ageing time attr Add a new function for DSA drivers to handle the switchdev SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME attribute. The ageing time is passed as milliseconds. Also because we can have multiple logical bridges on top of a physical switch and ageing time are switch-wide, call the driver function with the fastest ageing time in use on the chip instead of the requested one. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/include/net/dsa.h b/include/net/dsa.h index 52ab18b..2217a3f 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -141,6 +141,7 @@ struct dsa_switch_tree { struct dsa_port { struct net_device *netdev; struct device_node *dn; + unsigned int ageing_time; }; struct dsa_switch { @@ -329,6 +330,7 @@ struct dsa_switch_driver { /* * Bridge integration */ + int (*set_ageing_time)(struct dsa_switch *ds, unsigned int msecs); int (*port_bridge_join)(struct dsa_switch *ds, int port, struct net_device *bridge); void (*port_bridge_leave)(struct dsa_switch *ds, int port); diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 7236eb2..fc91967 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -333,6 +333,44 @@ static int dsa_slave_vlan_filtering(struct net_device *dev, return 0; } +static int dsa_fastest_ageing_time(struct dsa_switch *ds, + unsigned int ageing_time) +{ + int i; + + for (i = 0; i < DSA_MAX_PORTS; ++i) { + struct dsa_port *dp = &ds->ports[i]; + + if (dp && dp->ageing_time && dp->ageing_time < ageing_time) + ageing_time = dp->ageing_time; + } + + return ageing_time; +} + +static int dsa_slave_ageing_time(struct net_device *dev, + const struct switchdev_attr *attr, + struct switchdev_trans *trans) +{ + struct dsa_slave_priv *p = netdev_priv(dev); + struct dsa_switch *ds = p->parent; + unsigned long ageing_jiffies = clock_t_to_jiffies(attr->u.ageing_time); + unsigned int ageing_time = jiffies_to_msecs(ageing_jiffies); + + /* bridge skips -EOPNOTSUPP, so skip the prepare phase */ + if (switchdev_trans_ph_prepare(trans)) + return 0; + + /* Keep the fastest ageing time in case of multiple bridges */ + ds->ports[p->port].ageing_time = ageing_time; + ageing_time = dsa_fastest_ageing_time(ds, ageing_time); + + if (ds->drv->set_ageing_time) + return ds->drv->set_ageing_time(ds, ageing_time); + + return 0; +} + static int dsa_slave_port_attr_set(struct net_device *dev, const struct switchdev_attr *attr, struct switchdev_trans *trans) @@ -346,6 +384,9 @@ static int dsa_slave_port_attr_set(struct net_device *dev, case SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING: ret = dsa_slave_vlan_filtering(dev, attr, trans); break; + case SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME: + ret = dsa_slave_ageing_time(dev, attr, trans); + break; default: ret = -EOPNOTSUPP; break; -- cgit v0.10.2 From acddbd21477efc3da9ab414b2bc72cd580cf6fdb Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:39 -0400 Subject: net: dsa: mv88e6xxx: add G1 helper for ageing time All Marvell switch chips from (88E6060 to 88E6390) have a ATU Control register containing bits 11:4 to configure an ATU Age Time quotient. However the coefficient used to calculate the ATU Age Time vary with the models. E.g. 88E6060, 88E6352 and 88E6390 use respectively 16, 15 and 3.75 seconds. Add a age_time_coeff to the info structure to handle this and a Global 1 helper to set the default age time of 5 minutes in the setup code. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index bc26b4f..8e24c65 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -2975,6 +2975,33 @@ static int mv88e6xxx_g1_set_switch_mac(struct mv88e6xxx_chip *chip, u8 *addr) (addr[4] << 8) | addr[5]); } +static int mv88e6xxx_g1_set_age_time(struct mv88e6xxx_chip *chip, + unsigned int msecs) +{ + const unsigned int coeff = chip->info->age_time_coeff; + const unsigned int min = 0x01 * coeff; + const unsigned int max = 0xff * coeff; + u8 age_time; + u16 val; + int err; + + if (msecs < min || msecs > max) + return -ERANGE; + + /* Round to nearest multiple of coeff */ + age_time = (msecs + coeff / 2) / coeff; + + err = mv88e6xxx_read(chip, REG_GLOBAL, GLOBAL_ATU_CONTROL, &val); + if (err) + return err; + + /* AgeTime is 11:4 bits */ + val &= ~0xff0; + val |= age_time << 4; + + return mv88e6xxx_write(chip, REG_GLOBAL, GLOBAL_ATU_CONTROL, val); +} + static int mv88e6xxx_g1_setup(struct mv88e6xxx_chip *chip) { struct dsa_switch *ds = chip->ds; @@ -3012,18 +3039,22 @@ static int mv88e6xxx_g1_setup(struct mv88e6xxx_chip *chip) if (err) return err; + /* Clear all the VTU and STU entries */ + err = _mv88e6xxx_vtu_stu_flush(chip); + if (err < 0) + return err; + /* Set the default address aging time to 5 minutes, and * enable address learn messages to be sent to all message * ports. */ - err = _mv88e6xxx_reg_write(chip, REG_GLOBAL, GLOBAL_ATU_CONTROL, - 0x0140 | GLOBAL_ATU_CONTROL_LEARN2ALL); + err = mv88e6xxx_write(chip, REG_GLOBAL, GLOBAL_ATU_CONTROL, + GLOBAL_ATU_CONTROL_LEARN2ALL); if (err) return err; - /* Clear all the VTU and STU entries */ - err = _mv88e6xxx_vtu_stu_flush(chip); - if (err < 0) + err = mv88e6xxx_g1_set_age_time(chip, 300000); + if (err) return err; /* Clear all ATU entries */ @@ -3634,6 +3665,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 10, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6097, }, @@ -3644,6 +3676,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 256, .num_ports = 11, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6095, }, @@ -3654,6 +3687,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 3, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6165, }, @@ -3664,6 +3698,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 256, .num_ports = 8, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6185, }, @@ -3674,6 +3709,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 6, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6165, }, @@ -3684,6 +3720,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 6, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6165, }, @@ -3694,6 +3731,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 7, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6351, }, @@ -3704,6 +3742,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 7, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6352, }, @@ -3714,6 +3753,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 7, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6351, }, @@ -3724,6 +3764,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 7, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6352, }, @@ -3734,6 +3775,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 256, .num_ports = 10, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6185, }, @@ -3744,6 +3786,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 7, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6352, }, @@ -3754,6 +3797,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 7, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6320, }, @@ -3764,6 +3808,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 7, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6320, }, @@ -3774,6 +3819,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 7, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6351, }, @@ -3784,6 +3830,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 7, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6351, }, @@ -3794,6 +3841,7 @@ static const struct mv88e6xxx_info mv88e6xxx_table[] = { .num_databases = 4096, .num_ports = 7, .port_base_addr = 0x10, + .age_time_coeff = 15000, .flags = MV88E6XXX_FLAGS_FAMILY_6352, }, }; diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 9ea5363..899ca1d 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -568,6 +568,7 @@ struct mv88e6xxx_info { unsigned int num_databases; unsigned int num_ports; unsigned int port_base_addr; + unsigned int age_time_coeff; unsigned long flags; }; -- cgit v0.10.2 From 2cfcd9641618e71a1b823324aa4737e18662c25e Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 18 Jul 2016 20:45:40 -0400 Subject: net: dsa: mv88e6xxx: add support for DSA ageing time Implement the DSA driver function to configure the bridge ageing time. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 8e24c65..9ba2173 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -3002,6 +3002,19 @@ static int mv88e6xxx_g1_set_age_time(struct mv88e6xxx_chip *chip, return mv88e6xxx_write(chip, REG_GLOBAL, GLOBAL_ATU_CONTROL, val); } +static int mv88e6xxx_set_ageing_time(struct dsa_switch *ds, + unsigned int ageing_time) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int err; + + mutex_lock(&chip->reg_lock); + err = mv88e6xxx_g1_set_age_time(chip, ageing_time); + mutex_unlock(&chip->reg_lock); + + return err; +} + static int mv88e6xxx_g1_setup(struct mv88e6xxx_chip *chip) { struct dsa_switch *ds = chip->ds; @@ -3980,6 +3993,7 @@ static struct dsa_switch_driver mv88e6xxx_switch_driver = { .set_eeprom = mv88e6xxx_set_eeprom, .get_regs_len = mv88e6xxx_get_regs_len, .get_regs = mv88e6xxx_get_regs, + .set_ageing_time = mv88e6xxx_set_ageing_time, .port_bridge_join = mv88e6xxx_port_bridge_join, .port_bridge_leave = mv88e6xxx_port_bridge_leave, .port_stp_state_set = mv88e6xxx_port_stp_state_set, -- cgit v0.10.2 From 2d283bdd079c0ad4da020bbc9e9c2a4280823098 Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:16 +1000 Subject: net/ncsi: Resource management NCSI spec (DSP0222) defines several objects: package, channel, mode, filter, version and statistics etc. This introduces the data structs to represent those objects and implement functions to manage them. Also, this introduces CONFIG_NET_NCSI for the newly implemented NCSI stack. * The user (e.g. netdev driver) dereference NCSI device by "struct ncsi_dev", which is embedded to "struct ncsi_dev_priv". The later one is used by NCSI stack internally. * Every NCSI device can have multiple packages simultaneously, up to 8 packages. It's represented by "struct ncsi_package" and identified by 3-bits ID. * Every NCSI package can have multiple channels, up to 32. It's represented by "struct ncsi_channel" and identified by 5-bits ID. * Every NCSI channel has version, statistics, various modes and filters. They are represented by "struct ncsi_channel_version", "struct ncsi_channel_stats", "struct ncsi_channel_mode" and "struct ncsi_channel_filter" separately. * Apart from AEN (Asynchronous Event Notification), the NCSI stack works in terms of command and response. This introduces "struct ncsi_req" to represent a complete NCSI transaction made of NCSI request and response. link: https://www.dmtf.org/sites/default/files/standards/documents/DSP0222_1.1.0.pdf Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller diff --git a/include/net/ncsi.h b/include/net/ncsi.h new file mode 100644 index 0000000..70d14ee --- /dev/null +++ b/include/net/ncsi.h @@ -0,0 +1,46 @@ +#ifndef __NET_NCSI_H +#define __NET_NCSI_H + +/* + * The NCSI device states seen from external. More NCSI device states are + * only visible internally (in net/ncsi/internal.h). When the NCSI device + * is registered, it's in ncsi_dev_state_registered state. The state + * ncsi_dev_state_start is used to drive to choose active package and + * channel. After that, its state is changed to ncsi_dev_state_functional. + * + * The state ncsi_dev_state_stop helps to shut down the currently active + * package and channel while ncsi_dev_state_config helps to reconfigure + * them. + */ +enum { + ncsi_dev_state_registered = 0x0000, + ncsi_dev_state_functional = 0x0100, + ncsi_dev_state_probe = 0x0200, + ncsi_dev_state_config = 0x0300, + ncsi_dev_state_suspend = 0x0400, +}; + +struct ncsi_dev { + int state; + int link_up; + struct net_device *dev; + void (*handler)(struct ncsi_dev *ndev); +}; + +#ifdef CONFIG_NET_NCSI +struct ncsi_dev *ncsi_register_dev(struct net_device *dev, + void (*notifier)(struct ncsi_dev *nd)); +void ncsi_unregister_dev(struct ncsi_dev *nd); +#else /* !CONFIG_NET_NCSI */ +static inline struct ncsi_dev *ncsi_register_dev(struct net_device *dev, + void (*notifier)(struct ncsi_dev *nd)) +{ + return NULL; +} + +static inline void ncsi_unregister_dev(struct ncsi_dev *nd) +{ +} +#endif /* CONFIG_NET_NCSI */ + +#endif /* __NET_NCSI_H */ diff --git a/net/Kconfig b/net/Kconfig index ff40562..c2cdbce 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -237,6 +237,7 @@ source "net/hsr/Kconfig" source "net/switchdev/Kconfig" source "net/l3mdev/Kconfig" source "net/qrtr/Kconfig" +source "net/ncsi/Kconfig" config RPS bool diff --git a/net/Makefile b/net/Makefile index bdd1455..9bd20bb 100644 --- a/net/Makefile +++ b/net/Makefile @@ -79,3 +79,4 @@ ifneq ($(CONFIG_NET_L3_MASTER_DEV),) obj-y += l3mdev/ endif obj-$(CONFIG_QRTR) += qrtr/ +obj-$(CONFIG_NET_NCSI) += ncsi/ diff --git a/net/ncsi/Kconfig b/net/ncsi/Kconfig new file mode 100644 index 0000000..08a8a60 --- /dev/null +++ b/net/ncsi/Kconfig @@ -0,0 +1,12 @@ +# +# Configuration for NCSI support +# + +config NET_NCSI + bool "NCSI interface support" + depends on INET + ---help--- + This module provides NCSI (Network Controller Sideband Interface) + support. Enable this only if your system connects to a network + device via NCSI and the ethernet driver you're using supports + the protocol explicitly. diff --git a/net/ncsi/Makefile b/net/ncsi/Makefile new file mode 100644 index 0000000..07b5625 --- /dev/null +++ b/net/ncsi/Makefile @@ -0,0 +1,4 @@ +# +# Makefile for NCSI API +# +obj-$(CONFIG_NET_NCSI) += ncsi-manage.o diff --git a/net/ncsi/internal.h b/net/ncsi/internal.h new file mode 100644 index 0000000..89028e1 --- /dev/null +++ b/net/ncsi/internal.h @@ -0,0 +1,256 @@ +/* + * Copyright Gavin Shan, IBM Corporation 2016. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __NCSI_INTERNAL_H__ +#define __NCSI_INTERNAL_H__ + +enum { + NCSI_CAP_BASE = 0, + NCSI_CAP_GENERIC = 0, + NCSI_CAP_BC, + NCSI_CAP_MC, + NCSI_CAP_BUFFER, + NCSI_CAP_AEN, + NCSI_CAP_VLAN, + NCSI_CAP_MAX +}; + +enum { + NCSI_CAP_GENERIC_HWA = 0x01, /* HW arbitration */ + NCSI_CAP_GENERIC_HDS = 0x02, /* HNC driver status change */ + NCSI_CAP_GENERIC_FC = 0x04, /* HNC to MC flow control */ + NCSI_CAP_GENERIC_FC1 = 0x08, /* MC to HNC flow control */ + NCSI_CAP_GENERIC_MC = 0x10, /* Global MC filtering */ + NCSI_CAP_GENERIC_HWA_UNKNOWN = 0x00, /* Unknown HW arbitration */ + NCSI_CAP_GENERIC_HWA_SUPPORT = 0x20, /* Supported HW arbitration */ + NCSI_CAP_GENERIC_HWA_NOT_SUPPORT = 0x40, /* No HW arbitration */ + NCSI_CAP_GENERIC_HWA_RESERVED = 0x60, /* Reserved HW arbitration */ + NCSI_CAP_GENERIC_HWA_MASK = 0x60, /* Mask for HW arbitration */ + NCSI_CAP_GENERIC_MASK = 0x7f, + NCSI_CAP_BC_ARP = 0x01, /* ARP packet filtering */ + NCSI_CAP_BC_DHCPC = 0x02, /* DHCP client filtering */ + NCSI_CAP_BC_DHCPS = 0x04, /* DHCP server filtering */ + NCSI_CAP_BC_NETBIOS = 0x08, /* NetBIOS packet filtering */ + NCSI_CAP_BC_MASK = 0x0f, + NCSI_CAP_MC_IPV6_NEIGHBOR = 0x01, /* IPv6 neighbor filtering */ + NCSI_CAP_MC_IPV6_ROUTER = 0x02, /* IPv6 router filering */ + NCSI_CAP_MC_DHCPV6_RELAY = 0x04, /* DHCPv6 relay / server MC */ + NCSI_CAP_MC_DHCPV6_WELL_KNOWN = 0x08, /* DHCPv6 well-known MC */ + NCSI_CAP_MC_IPV6_MLD = 0x10, /* IPv6 MLD filtering */ + NCSI_CAP_MC_IPV6_NEIGHBOR_S = 0x20, /* IPv6 neighbour filtering */ + NCSI_CAP_MC_MASK = 0x3f, + NCSI_CAP_AEN_LSC = 0x01, /* Link status change */ + NCSI_CAP_AEN_CR = 0x02, /* Configuration required */ + NCSI_CAP_AEN_HDS = 0x04, /* HNC driver status */ + NCSI_CAP_AEN_MASK = 0x07, + NCSI_CAP_VLAN_ONLY = 0x01, /* Filter VLAN packet only */ + NCSI_CAP_VLAN_NO = 0x02, /* Filter VLAN and non-VLAN */ + NCSI_CAP_VLAN_ANY = 0x04, /* Filter Any-and-non-VLAN */ + NCSI_CAP_VLAN_MASK = 0x07 +}; + +enum { + NCSI_MODE_BASE = 0, + NCSI_MODE_ENABLE = 0, + NCSI_MODE_TX_ENABLE, + NCSI_MODE_LINK, + NCSI_MODE_VLAN, + NCSI_MODE_BC, + NCSI_MODE_MC, + NCSI_MODE_AEN, + NCSI_MODE_FC, + NCSI_MODE_MAX +}; + +enum { + NCSI_FILTER_BASE = 0, + NCSI_FILTER_VLAN = 0, + NCSI_FILTER_UC, + NCSI_FILTER_MC, + NCSI_FILTER_MIXED, + NCSI_FILTER_MAX +}; + +struct ncsi_channel_version { + u32 version; /* Supported BCD encoded NCSI version */ + u32 alpha2; /* Supported BCD encoded NCSI version */ + u8 fw_name[12]; /* Firware name string */ + u32 fw_version; /* Firmware version */ + u16 pci_ids[4]; /* PCI identification */ + u32 mf_id; /* Manufacture ID */ +}; + +struct ncsi_channel_cap { + u32 index; /* Index of channel capabilities */ + u32 cap; /* NCSI channel capability */ +}; + +struct ncsi_channel_mode { + u32 index; /* Index of channel modes */ + u32 enable; /* Enabled or disabled */ + u32 size; /* Valid entries in ncm_data[] */ + u32 data[8]; /* Data entries */ +}; + +struct ncsi_channel_filter { + u32 index; /* Index of channel filters */ + u32 total; /* Total entries in the filter table */ + u64 bitmap; /* Bitmap of valid entries */ + u32 data[]; /* Data for the valid entries */ +}; + +struct ncsi_channel_stats { + u32 hnc_cnt_hi; /* Counter cleared */ + u32 hnc_cnt_lo; /* Counter cleared */ + u32 hnc_rx_bytes; /* Rx bytes */ + u32 hnc_tx_bytes; /* Tx bytes */ + u32 hnc_rx_uc_pkts; /* Rx UC packets */ + u32 hnc_rx_mc_pkts; /* Rx MC packets */ + u32 hnc_rx_bc_pkts; /* Rx BC packets */ + u32 hnc_tx_uc_pkts; /* Tx UC packets */ + u32 hnc_tx_mc_pkts; /* Tx MC packets */ + u32 hnc_tx_bc_pkts; /* Tx BC packets */ + u32 hnc_fcs_err; /* FCS errors */ + u32 hnc_align_err; /* Alignment errors */ + u32 hnc_false_carrier; /* False carrier detection */ + u32 hnc_runt_pkts; /* Rx runt packets */ + u32 hnc_jabber_pkts; /* Rx jabber packets */ + u32 hnc_rx_pause_xon; /* Rx pause XON frames */ + u32 hnc_rx_pause_xoff; /* Rx XOFF frames */ + u32 hnc_tx_pause_xon; /* Tx XON frames */ + u32 hnc_tx_pause_xoff; /* Tx XOFF frames */ + u32 hnc_tx_s_collision; /* Single collision frames */ + u32 hnc_tx_m_collision; /* Multiple collision frames */ + u32 hnc_l_collision; /* Late collision frames */ + u32 hnc_e_collision; /* Excessive collision frames */ + u32 hnc_rx_ctl_frames; /* Rx control frames */ + u32 hnc_rx_64_frames; /* Rx 64-bytes frames */ + u32 hnc_rx_127_frames; /* Rx 65-127 bytes frames */ + u32 hnc_rx_255_frames; /* Rx 128-255 bytes frames */ + u32 hnc_rx_511_frames; /* Rx 256-511 bytes frames */ + u32 hnc_rx_1023_frames; /* Rx 512-1023 bytes frames */ + u32 hnc_rx_1522_frames; /* Rx 1024-1522 bytes frames */ + u32 hnc_rx_9022_frames; /* Rx 1523-9022 bytes frames */ + u32 hnc_tx_64_frames; /* Tx 64-bytes frames */ + u32 hnc_tx_127_frames; /* Tx 65-127 bytes frames */ + u32 hnc_tx_255_frames; /* Tx 128-255 bytes frames */ + u32 hnc_tx_511_frames; /* Tx 256-511 bytes frames */ + u32 hnc_tx_1023_frames; /* Tx 512-1023 bytes frames */ + u32 hnc_tx_1522_frames; /* Tx 1024-1522 bytes frames */ + u32 hnc_tx_9022_frames; /* Tx 1523-9022 bytes frames */ + u32 hnc_rx_valid_bytes; /* Rx valid bytes */ + u32 hnc_rx_runt_pkts; /* Rx error runt packets */ + u32 hnc_rx_jabber_pkts; /* Rx error jabber packets */ + u32 ncsi_rx_cmds; /* Rx NCSI commands */ + u32 ncsi_dropped_cmds; /* Dropped commands */ + u32 ncsi_cmd_type_errs; /* Command type errors */ + u32 ncsi_cmd_csum_errs; /* Command checksum errors */ + u32 ncsi_rx_pkts; /* Rx NCSI packets */ + u32 ncsi_tx_pkts; /* Tx NCSI packets */ + u32 ncsi_tx_aen_pkts; /* Tx AEN packets */ + u32 pt_tx_pkts; /* Tx packets */ + u32 pt_tx_dropped; /* Tx dropped packets */ + u32 pt_tx_channel_err; /* Tx channel errors */ + u32 pt_tx_us_err; /* Tx undersize errors */ + u32 pt_rx_pkts; /* Rx packets */ + u32 pt_rx_dropped; /* Rx dropped packets */ + u32 pt_rx_channel_err; /* Rx channel errors */ + u32 pt_rx_us_err; /* Rx undersize errors */ + u32 pt_rx_os_err; /* Rx oversize errors */ +}; + +struct ncsi_dev_priv; +struct ncsi_package; + +#define NCSI_PACKAGE_SHIFT 5 +#define NCSI_PACKAGE_INDEX(c) (((c) >> NCSI_PACKAGE_SHIFT) & 0x7) +#define NCSI_CHANNEL_INDEX(c) ((c) & ((1 << NCSI_PACKAGE_SHIFT) - 1)) +#define NCSI_TO_CHANNEL(p, c) (((p) << NCSI_PACKAGE_SHIFT) | (c)) + +struct ncsi_channel { + unsigned char id; + int state; +#define NCSI_CHANNEL_INACTIVE 1 +#define NCSI_CHANNEL_ACTIVE 2 + spinlock_t lock; /* Protect filters etc */ + struct ncsi_package *package; + struct ncsi_channel_version version; + struct ncsi_channel_cap caps[NCSI_CAP_MAX]; + struct ncsi_channel_mode modes[NCSI_MODE_MAX]; + struct ncsi_channel_filter *filters[NCSI_FILTER_MAX]; + struct ncsi_channel_stats stats; + struct list_head node; +}; + +struct ncsi_package { + unsigned char id; /* NCSI 3-bits package ID */ + unsigned char uuid[16]; /* UUID */ + struct ncsi_dev_priv *ndp; /* NCSI device */ + spinlock_t lock; /* Protect the package */ + unsigned int channel_num; /* Number of channels */ + struct list_head channels; /* List of chanels */ + struct list_head node; /* Form list of packages */ +}; + +struct ncsi_request { + unsigned char id; /* Request ID - 0 to 255 */ + bool used; /* Request that has been assigned */ + bool driven; /* Drive state machine */ + struct ncsi_dev_priv *ndp; /* Associated NCSI device */ + struct sk_buff *cmd; /* Associated NCSI command packet */ + struct sk_buff *rsp; /* Associated NCSI response packet */ + struct timer_list timer; /* Timer on waiting for response */ + bool enabled; /* Time has been enabled or not */ +}; + +struct ncsi_dev_priv { + struct ncsi_dev ndev; /* Associated NCSI device */ + unsigned int flags; /* NCSI device flags */ + spinlock_t lock; /* Protect the NCSI device */ + unsigned int package_num; /* Number of packages */ + struct list_head packages; /* List of packages */ + struct ncsi_request requests[256]; /* Request table */ + unsigned int request_id; /* Last used request ID */ + struct list_head node; /* Form NCSI device list */ +}; + +extern struct list_head ncsi_dev_list; +extern spinlock_t ncsi_dev_lock; + +#define TO_NCSI_DEV_PRIV(nd) \ + container_of(nd, struct ncsi_dev_priv, ndev) +#define NCSI_FOR_EACH_DEV(ndp) \ + list_for_each_entry_rcu(ndp, &ncsi_dev_list, node) +#define NCSI_FOR_EACH_PACKAGE(ndp, np) \ + list_for_each_entry_rcu(np, &ndp->packages, node) +#define NCSI_FOR_EACH_CHANNEL(np, nc) \ + list_for_each_entry_rcu(nc, &np->channels, node) + +/* Resources */ +int ncsi_find_filter(struct ncsi_channel *nc, int table, void *data); +int ncsi_add_filter(struct ncsi_channel *nc, int table, void *data); +int ncsi_remove_filter(struct ncsi_channel *nc, int table, int index); +struct ncsi_channel *ncsi_find_channel(struct ncsi_package *np, + unsigned char id); +struct ncsi_channel *ncsi_add_channel(struct ncsi_package *np, + unsigned char id); +struct ncsi_package *ncsi_find_package(struct ncsi_dev_priv *ndp, + unsigned char id); +struct ncsi_package *ncsi_add_package(struct ncsi_dev_priv *ndp, + unsigned char id); +void ncsi_remove_package(struct ncsi_package *np); +void ncsi_find_package_and_channel(struct ncsi_dev_priv *ndp, + unsigned char id, + struct ncsi_package **np, + struct ncsi_channel **nc); +struct ncsi_request *ncsi_alloc_request(struct ncsi_dev_priv *ndp, bool driven); +void ncsi_free_request(struct ncsi_request *nr); +struct ncsi_dev *ncsi_find_dev(struct net_device *dev); + +#endif /* __NCSI_INTERNAL_H__ */ diff --git a/net/ncsi/ncsi-manage.c b/net/ncsi/ncsi-manage.c new file mode 100644 index 0000000..0e28ed8 --- /dev/null +++ b/net/ncsi/ncsi-manage.c @@ -0,0 +1,436 @@ +/* + * Copyright Gavin Shan, IBM Corporation 2016. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "internal.h" + +LIST_HEAD(ncsi_dev_list); +DEFINE_SPINLOCK(ncsi_dev_lock); + +static inline int ncsi_filter_size(int table) +{ + int sizes[] = { 2, 6, 6, 6 }; + + BUILD_BUG_ON(ARRAY_SIZE(sizes) != NCSI_FILTER_MAX); + if (table < NCSI_FILTER_BASE || table >= NCSI_FILTER_MAX) + return -EINVAL; + + return sizes[table]; +} + +int ncsi_find_filter(struct ncsi_channel *nc, int table, void *data) +{ + struct ncsi_channel_filter *ncf; + void *bitmap; + int index, size; + unsigned long flags; + + ncf = nc->filters[table]; + if (!ncf) + return -ENXIO; + + size = ncsi_filter_size(table); + if (size < 0) + return size; + + spin_lock_irqsave(&nc->lock, flags); + bitmap = (void *)&ncf->bitmap; + index = -1; + while ((index = find_next_bit(bitmap, ncf->total, index + 1)) + < ncf->total) { + if (!memcmp(ncf->data + size * index, data, size)) { + spin_unlock_irqrestore(&nc->lock, flags); + return index; + } + } + spin_unlock_irqrestore(&nc->lock, flags); + + return -ENOENT; +} + +int ncsi_add_filter(struct ncsi_channel *nc, int table, void *data) +{ + struct ncsi_channel_filter *ncf; + int index, size; + void *bitmap; + unsigned long flags; + + size = ncsi_filter_size(table); + if (size < 0) + return size; + + index = ncsi_find_filter(nc, table, data); + if (index >= 0) + return index; + + ncf = nc->filters[table]; + if (!ncf) + return -ENODEV; + + spin_lock_irqsave(&nc->lock, flags); + bitmap = (void *)&ncf->bitmap; + do { + index = find_next_zero_bit(bitmap, ncf->total, 0); + if (index >= ncf->total) { + spin_unlock_irqrestore(&nc->lock, flags); + return -ENOSPC; + } + } while (test_and_set_bit(index, bitmap)); + + memcpy(ncf->data + size * index, data, size); + spin_unlock_irqrestore(&nc->lock, flags); + + return index; +} + +int ncsi_remove_filter(struct ncsi_channel *nc, int table, int index) +{ + struct ncsi_channel_filter *ncf; + int size; + void *bitmap; + unsigned long flags; + + size = ncsi_filter_size(table); + if (size < 0) + return size; + + ncf = nc->filters[table]; + if (!ncf || index >= ncf->total) + return -ENODEV; + + spin_lock_irqsave(&nc->lock, flags); + bitmap = (void *)&ncf->bitmap; + if (test_and_clear_bit(index, bitmap)) + memset(ncf->data + size * index, 0, size); + spin_unlock_irqrestore(&nc->lock, flags); + + return 0; +} + +struct ncsi_channel *ncsi_find_channel(struct ncsi_package *np, + unsigned char id) +{ + struct ncsi_channel *nc; + + NCSI_FOR_EACH_CHANNEL(np, nc) { + if (nc->id == id) + return nc; + } + + return NULL; +} + +struct ncsi_channel *ncsi_add_channel(struct ncsi_package *np, unsigned char id) +{ + struct ncsi_channel *nc, *tmp; + int index; + unsigned long flags; + + nc = kzalloc(sizeof(*nc), GFP_ATOMIC); + if (!nc) + return NULL; + + nc->id = id; + nc->package = np; + nc->state = NCSI_CHANNEL_INACTIVE; + spin_lock_init(&nc->lock); + for (index = 0; index < NCSI_CAP_MAX; index++) + nc->caps[index].index = index; + for (index = 0; index < NCSI_MODE_MAX; index++) + nc->modes[index].index = index; + + spin_lock_irqsave(&np->lock, flags); + tmp = ncsi_find_channel(np, id); + if (tmp) { + spin_unlock_irqrestore(&np->lock, flags); + kfree(nc); + return tmp; + } + + list_add_tail_rcu(&nc->node, &np->channels); + np->channel_num++; + spin_unlock_irqrestore(&np->lock, flags); + + return nc; +} + +static void ncsi_remove_channel(struct ncsi_channel *nc) +{ + struct ncsi_package *np = nc->package; + struct ncsi_channel_filter *ncf; + unsigned long flags; + int i; + + /* Release filters */ + spin_lock_irqsave(&nc->lock, flags); + for (i = 0; i < NCSI_FILTER_MAX; i++) { + ncf = nc->filters[i]; + if (!ncf) + continue; + + nc->filters[i] = NULL; + kfree(ncf); + } + + nc->state = NCSI_CHANNEL_INACTIVE; + spin_unlock_irqrestore(&nc->lock, flags); + + /* Remove and free channel */ + spin_lock_irqsave(&np->lock, flags); + list_del_rcu(&nc->node); + np->channel_num--; + spin_unlock_irqrestore(&np->lock, flags); + + kfree(nc); +} + +struct ncsi_package *ncsi_find_package(struct ncsi_dev_priv *ndp, + unsigned char id) +{ + struct ncsi_package *np; + + NCSI_FOR_EACH_PACKAGE(ndp, np) { + if (np->id == id) + return np; + } + + return NULL; +} + +struct ncsi_package *ncsi_add_package(struct ncsi_dev_priv *ndp, + unsigned char id) +{ + struct ncsi_package *np, *tmp; + unsigned long flags; + + np = kzalloc(sizeof(*np), GFP_ATOMIC); + if (!np) + return NULL; + + np->id = id; + np->ndp = ndp; + spin_lock_init(&np->lock); + INIT_LIST_HEAD(&np->channels); + + spin_lock_irqsave(&ndp->lock, flags); + tmp = ncsi_find_package(ndp, id); + if (tmp) { + spin_unlock_irqrestore(&ndp->lock, flags); + kfree(np); + return tmp; + } + + list_add_tail_rcu(&np->node, &ndp->packages); + ndp->package_num++; + spin_unlock_irqrestore(&ndp->lock, flags); + + return np; +} + +void ncsi_remove_package(struct ncsi_package *np) +{ + struct ncsi_dev_priv *ndp = np->ndp; + struct ncsi_channel *nc, *tmp; + unsigned long flags; + + /* Release all child channels */ + list_for_each_entry_safe(nc, tmp, &np->channels, node) + ncsi_remove_channel(nc); + + /* Remove and free package */ + spin_lock_irqsave(&ndp->lock, flags); + list_del_rcu(&np->node); + ndp->package_num--; + spin_unlock_irqrestore(&ndp->lock, flags); + + kfree(np); +} + +void ncsi_find_package_and_channel(struct ncsi_dev_priv *ndp, + unsigned char id, + struct ncsi_package **np, + struct ncsi_channel **nc) +{ + struct ncsi_package *p; + struct ncsi_channel *c; + + p = ncsi_find_package(ndp, NCSI_PACKAGE_INDEX(id)); + c = p ? ncsi_find_channel(p, NCSI_CHANNEL_INDEX(id)) : NULL; + + if (np) + *np = p; + if (nc) + *nc = c; +} + +/* For two consecutive NCSI commands, the packet IDs shouldn't + * be same. Otherwise, the bogus response might be replied. So + * the available IDs are allocated in round-robin fashion. + */ +struct ncsi_request *ncsi_alloc_request(struct ncsi_dev_priv *ndp, bool driven) +{ + struct ncsi_request *nr = NULL; + int i, limit = ARRAY_SIZE(ndp->requests); + unsigned long flags; + + /* Check if there is one available request until the ceiling */ + spin_lock_irqsave(&ndp->lock, flags); + for (i = ndp->request_id; !nr && i < limit; i++) { + if (ndp->requests[i].used) + continue; + + nr = &ndp->requests[i]; + nr->used = true; + nr->driven = driven; + if (++ndp->request_id >= limit) + ndp->request_id = 0; + } + + /* Fail back to check from the starting cursor */ + for (i = 0; !nr && i < ndp->request_id; i++) { + if (ndp->requests[i].used) + continue; + + nr = &ndp->requests[i]; + nr->used = true; + nr->driven = driven; + if (++ndp->request_id >= limit) + ndp->request_id = 0; + } + spin_unlock_irqrestore(&ndp->lock, flags); + + return nr; +} + +void ncsi_free_request(struct ncsi_request *nr) +{ + struct ncsi_dev_priv *ndp = nr->ndp; + struct sk_buff *cmd, *rsp; + unsigned long flags; + + if (nr->enabled) { + nr->enabled = false; + del_timer_sync(&nr->timer); + } + + spin_lock_irqsave(&ndp->lock, flags); + cmd = nr->cmd; + rsp = nr->rsp; + nr->cmd = NULL; + nr->rsp = NULL; + nr->used = false; + spin_unlock_irqrestore(&ndp->lock, flags); + + /* Release command and response */ + consume_skb(cmd); + consume_skb(rsp); +} + +struct ncsi_dev *ncsi_find_dev(struct net_device *dev) +{ + struct ncsi_dev_priv *ndp; + + NCSI_FOR_EACH_DEV(ndp) { + if (ndp->ndev.dev == dev) + return &ndp->ndev; + } + + return NULL; +} + +static void ncsi_request_timeout(unsigned long data) +{ + struct ncsi_request *nr = (struct ncsi_request *)data; + struct ncsi_dev_priv *ndp = nr->ndp; + unsigned long flags; + + /* If the request already had associated response, + * let the response handler to release it. + */ + spin_lock_irqsave(&ndp->lock, flags); + nr->enabled = false; + if (nr->rsp || !nr->cmd) { + spin_unlock_irqrestore(&ndp->lock, flags); + return; + } + spin_unlock_irqrestore(&ndp->lock, flags); + + /* Release the request */ + ncsi_free_request(nr); +} + +struct ncsi_dev *ncsi_register_dev(struct net_device *dev, + void (*handler)(struct ncsi_dev *ndev)) +{ + struct ncsi_dev_priv *ndp; + struct ncsi_dev *nd; + unsigned long flags; + int i; + + /* Check if the device has been registered or not */ + nd = ncsi_find_dev(dev); + if (nd) + return nd; + + /* Create NCSI device */ + ndp = kzalloc(sizeof(*ndp), GFP_ATOMIC); + if (!ndp) + return NULL; + + nd = &ndp->ndev; + nd->state = ncsi_dev_state_registered; + nd->dev = dev; + nd->handler = handler; + + /* Initialize private NCSI device */ + spin_lock_init(&ndp->lock); + INIT_LIST_HEAD(&ndp->packages); + ndp->request_id = 0; + for (i = 0; i < ARRAY_SIZE(ndp->requests); i++) { + ndp->requests[i].id = i; + ndp->requests[i].ndp = ndp; + setup_timer(&ndp->requests[i].timer, + ncsi_request_timeout, + (unsigned long)&ndp->requests[i]); + } + + spin_lock_irqsave(&ncsi_dev_lock, flags); + list_add_tail_rcu(&ndp->node, &ncsi_dev_list); + spin_unlock_irqrestore(&ncsi_dev_lock, flags); + + return nd; +} +EXPORT_SYMBOL_GPL(ncsi_register_dev); + +void ncsi_unregister_dev(struct ncsi_dev *nd) +{ + struct ncsi_dev_priv *ndp = TO_NCSI_DEV_PRIV(nd); + struct ncsi_package *np, *tmp; + unsigned long flags; + + list_for_each_entry_safe(np, tmp, &ndp->packages, node) + ncsi_remove_package(np); + + spin_lock_irqsave(&ncsi_dev_lock, flags); + list_del_rcu(&ndp->node); + spin_unlock_irqrestore(&ncsi_dev_lock, flags); + + kfree(ndp); +} +EXPORT_SYMBOL_GPL(ncsi_unregister_dev); -- cgit v0.10.2 From 6389eaa7fa9c3ee6c7d39f6087b86660d17236ac Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:17 +1000 Subject: net/ncsi: NCSI command packet handler The NCSI command packets are sent from MC (Management Controller) to remote end. They are used for multiple purposes: probe existing NCSI package/channel, retrieve NCSI channel's capability, configure NCSI channel etc. This defines struct to represent NCSI command packets and introduces function ncsi_xmit_cmd(), which will be used to transmit NCSI command packet according to the request. The request is represented by struct ncsi_cmd_arg. Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller diff --git a/include/uapi/linux/if_ether.h b/include/uapi/linux/if_ether.h index cec849a..117d02e 100644 --- a/include/uapi/linux/if_ether.h +++ b/include/uapi/linux/if_ether.h @@ -87,6 +87,7 @@ #define ETH_P_8021AH 0x88E7 /* 802.1ah Backbone Service Tag */ #define ETH_P_MVRP 0x88F5 /* 802.1Q MVRP */ #define ETH_P_1588 0x88F7 /* IEEE 1588 Timesync */ +#define ETH_P_NCSI 0x88F8 /* NCSI protocol */ #define ETH_P_PRP 0x88FB /* IEC 62439-3 PRP/HSRv0 */ #define ETH_P_FCOE 0x8906 /* Fibre Channel over Ethernet */ #define ETH_P_TDLS 0x890D /* TDLS */ diff --git a/net/ncsi/Makefile b/net/ncsi/Makefile index 07b5625..abc4046 100644 --- a/net/ncsi/Makefile +++ b/net/ncsi/Makefile @@ -1,4 +1,4 @@ # # Makefile for NCSI API # -obj-$(CONFIG_NET_NCSI) += ncsi-manage.o +obj-$(CONFIG_NET_NCSI) += ncsi-cmd.o ncsi-manage.o diff --git a/net/ncsi/internal.h b/net/ncsi/internal.h index 89028e1..3d81697 100644 --- a/net/ncsi/internal.h +++ b/net/ncsi/internal.h @@ -220,6 +220,21 @@ struct ncsi_dev_priv { struct list_head node; /* Form NCSI device list */ }; +struct ncsi_cmd_arg { + struct ncsi_dev_priv *ndp; /* Associated NCSI device */ + unsigned char type; /* Command in the NCSI packet */ + unsigned char id; /* Request ID (sequence number) */ + unsigned char package; /* Destination package ID */ + unsigned char channel; /* Detination channel ID or 0x1f */ + unsigned short payload; /* Command packet payload length */ + bool driven; /* Drive the state machine? */ + union { + unsigned char bytes[16]; /* Command packet specific data */ + unsigned short words[8]; + unsigned int dwords[4]; + }; +}; + extern struct list_head ncsi_dev_list; extern spinlock_t ncsi_dev_lock; @@ -253,4 +268,8 @@ struct ncsi_request *ncsi_alloc_request(struct ncsi_dev_priv *ndp, bool driven); void ncsi_free_request(struct ncsi_request *nr); struct ncsi_dev *ncsi_find_dev(struct net_device *dev); +/* Packet handlers */ +u32 ncsi_calculate_checksum(unsigned char *data, int len); +int ncsi_xmit_cmd(struct ncsi_cmd_arg *nca); + #endif /* __NCSI_INTERNAL_H__ */ diff --git a/net/ncsi/ncsi-cmd.c b/net/ncsi/ncsi-cmd.c new file mode 100644 index 0000000..21057a8 --- /dev/null +++ b/net/ncsi/ncsi-cmd.c @@ -0,0 +1,367 @@ +/* + * Copyright Gavin Shan, IBM Corporation 2016. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "internal.h" +#include "ncsi-pkt.h" + +u32 ncsi_calculate_checksum(unsigned char *data, int len) +{ + u32 checksum = 0; + int i; + + for (i = 0; i < len; i += 2) + checksum += (((u32)data[i] << 8) | data[i + 1]); + + checksum = (~checksum + 1); + return checksum; +} + +/* This function should be called after the data area has been + * populated completely. + */ +static void ncsi_cmd_build_header(struct ncsi_pkt_hdr *h, + struct ncsi_cmd_arg *nca) +{ + u32 checksum; + __be32 *pchecksum; + + h->mc_id = 0; + h->revision = NCSI_PKT_REVISION; + h->reserved = 0; + h->id = nca->id; + h->type = nca->type; + h->channel = NCSI_TO_CHANNEL(nca->package, + nca->channel); + h->length = htons(nca->payload); + h->reserved1[0] = 0; + h->reserved1[1] = 0; + + /* Fill with calculated checksum */ + checksum = ncsi_calculate_checksum((unsigned char *)h, + sizeof(*h) + nca->payload); + pchecksum = (__be32 *)((void *)h + sizeof(struct ncsi_pkt_hdr) + + nca->payload); + *pchecksum = htonl(checksum); +} + +static int ncsi_cmd_handler_default(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_pkt *cmd; + + cmd = (struct ncsi_cmd_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_sp(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_sp_pkt *cmd; + + cmd = (struct ncsi_cmd_sp_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->hw_arbitration = nca->bytes[0]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_dc(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_dc_pkt *cmd; + + cmd = (struct ncsi_cmd_dc_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->ald = nca->bytes[0]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_rc(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_rc_pkt *cmd; + + cmd = (struct ncsi_cmd_rc_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_ae(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_ae_pkt *cmd; + + cmd = (struct ncsi_cmd_ae_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mc_id = nca->bytes[0]; + cmd->mode = htonl(nca->dwords[1]); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_sl(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_sl_pkt *cmd; + + cmd = (struct ncsi_cmd_sl_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mode = htonl(nca->dwords[0]); + cmd->oem_mode = htonl(nca->dwords[1]); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_svf(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_svf_pkt *cmd; + + cmd = (struct ncsi_cmd_svf_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->vlan = htons(nca->words[0]); + cmd->index = nca->bytes[2]; + cmd->enable = nca->bytes[3]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_ev(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_ev_pkt *cmd; + + cmd = (struct ncsi_cmd_ev_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mode = nca->bytes[0]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_sma(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_sma_pkt *cmd; + int i; + + cmd = (struct ncsi_cmd_sma_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + for (i = 0; i < 6; i++) + cmd->mac[i] = nca->bytes[i]; + cmd->index = nca->bytes[6]; + cmd->at_e = nca->bytes[7]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_ebf(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_ebf_pkt *cmd; + + cmd = (struct ncsi_cmd_ebf_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mode = htonl(nca->dwords[0]); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_egmf(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_egmf_pkt *cmd; + + cmd = (struct ncsi_cmd_egmf_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mode = htonl(nca->dwords[0]); + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static int ncsi_cmd_handler_snfc(struct sk_buff *skb, + struct ncsi_cmd_arg *nca) +{ + struct ncsi_cmd_snfc_pkt *cmd; + + cmd = (struct ncsi_cmd_snfc_pkt *)skb_put(skb, sizeof(*cmd)); + memset(cmd, 0, sizeof(*cmd)); + cmd->mode = nca->bytes[0]; + ncsi_cmd_build_header(&cmd->cmd.common, nca); + + return 0; +} + +static struct ncsi_cmd_handler { + unsigned char type; + int payload; + int (*handler)(struct sk_buff *skb, + struct ncsi_cmd_arg *nca); +} ncsi_cmd_handlers[] = { + { NCSI_PKT_CMD_CIS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_SP, 4, ncsi_cmd_handler_sp }, + { NCSI_PKT_CMD_DP, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_EC, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_DC, 4, ncsi_cmd_handler_dc }, + { NCSI_PKT_CMD_RC, 4, ncsi_cmd_handler_rc }, + { NCSI_PKT_CMD_ECNT, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_DCNT, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_AE, 8, ncsi_cmd_handler_ae }, + { NCSI_PKT_CMD_SL, 8, ncsi_cmd_handler_sl }, + { NCSI_PKT_CMD_GLS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_SVF, 4, ncsi_cmd_handler_svf }, + { NCSI_PKT_CMD_EV, 4, ncsi_cmd_handler_ev }, + { NCSI_PKT_CMD_DV, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_SMA, 8, ncsi_cmd_handler_sma }, + { NCSI_PKT_CMD_EBF, 4, ncsi_cmd_handler_ebf }, + { NCSI_PKT_CMD_DBF, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_EGMF, 4, ncsi_cmd_handler_egmf }, + { NCSI_PKT_CMD_DGMF, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_SNFC, 4, ncsi_cmd_handler_snfc }, + { NCSI_PKT_CMD_GVI, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GC, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GP, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GCPS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GNS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GNPTS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_GPS, 0, ncsi_cmd_handler_default }, + { NCSI_PKT_CMD_OEM, 0, NULL }, + { NCSI_PKT_CMD_PLDM, 0, NULL }, + { NCSI_PKT_CMD_GPUUID, 0, ncsi_cmd_handler_default } +}; + +static struct ncsi_request *ncsi_alloc_command(struct ncsi_cmd_arg *nca) +{ + struct ncsi_dev_priv *ndp = nca->ndp; + struct ncsi_dev *nd = &ndp->ndev; + struct net_device *dev = nd->dev; + int hlen = LL_RESERVED_SPACE(dev); + int tlen = dev->needed_tailroom; + int len = hlen + tlen; + struct sk_buff *skb; + struct ncsi_request *nr; + + nr = ncsi_alloc_request(ndp, nca->driven); + if (!nr) + return NULL; + + /* NCSI command packet has 16-bytes header, payload, 4 bytes checksum. + * The packet needs padding if its payload is less than 26 bytes to + * meet 64 bytes minimal ethernet frame length. + */ + len += sizeof(struct ncsi_cmd_pkt_hdr) + 4; + if (nca->payload < 26) + len += 26; + else + len += nca->payload; + + /* Allocate skb */ + skb = alloc_skb(len, GFP_ATOMIC); + if (!skb) { + ncsi_free_request(nr); + return NULL; + } + + nr->cmd = skb; + skb_reserve(skb, hlen); + skb_reset_network_header(skb); + + skb->dev = dev; + skb->protocol = htons(ETH_P_NCSI); + + return nr; +} + +int ncsi_xmit_cmd(struct ncsi_cmd_arg *nca) +{ + struct ncsi_request *nr; + struct ethhdr *eh; + struct ncsi_cmd_handler *nch = NULL; + int i, ret; + + /* Search for the handler */ + for (i = 0; i < ARRAY_SIZE(ncsi_cmd_handlers); i++) { + if (ncsi_cmd_handlers[i].type == nca->type) { + if (ncsi_cmd_handlers[i].handler) + nch = &ncsi_cmd_handlers[i]; + else + nch = NULL; + + break; + } + } + + if (!nch) { + netdev_err(nca->ndp->ndev.dev, + "Cannot send packet with type 0x%02x\n", nca->type); + return -ENOENT; + } + + /* Get packet payload length and allocate the request */ + nca->payload = nch->payload; + nr = ncsi_alloc_command(nca); + if (!nr) + return -ENOMEM; + + /* Prepare the packet */ + nca->id = nr->id; + ret = nch->handler(nr->cmd, nca); + if (ret) { + ncsi_free_request(nr); + return ret; + } + + /* Fill the ethernet header */ + eh = (struct ethhdr *)skb_push(nr->cmd, sizeof(*eh)); + eh->h_proto = htons(ETH_P_NCSI); + eth_broadcast_addr(eh->h_dest); + eth_broadcast_addr(eh->h_source); + + /* Start the timer for the request that might not have + * corresponding response. Given NCSI is an internal + * connection a 1 second delay should be sufficient. + */ + nr->enabled = true; + mod_timer(&nr->timer, jiffies + 1 * HZ); + + /* Send NCSI packet */ + skb_get(nr->cmd); + ret = dev_queue_xmit(nr->cmd); + if (ret < 0) { + ncsi_free_request(nr); + return ret; + } + + return 0; +} diff --git a/net/ncsi/ncsi-pkt.h b/net/ncsi/ncsi-pkt.h new file mode 100644 index 0000000..5481458 --- /dev/null +++ b/net/ncsi/ncsi-pkt.h @@ -0,0 +1,171 @@ +/* + * Copyright Gavin Shan, IBM Corporation 2016. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __NCSI_PKT_H__ +#define __NCSI_PKT_H__ + +struct ncsi_pkt_hdr { + unsigned char mc_id; /* Management controller ID */ + unsigned char revision; /* NCSI version - 0x01 */ + unsigned char reserved; /* Reserved */ + unsigned char id; /* Packet sequence number */ + unsigned char type; /* Packet type */ + unsigned char channel; /* Network controller ID */ + __be16 length; /* Payload length */ + __be32 reserved1[2]; /* Reserved */ +}; + +struct ncsi_cmd_pkt_hdr { + struct ncsi_pkt_hdr common; /* Common NCSI packet header */ +}; + +/* NCSI common command packet */ +struct ncsi_cmd_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be32 checksum; /* Checksum */ + unsigned char pad[26]; +}; + +/* Select Package */ +struct ncsi_cmd_sp_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char reserved[3]; /* Reserved */ + unsigned char hw_arbitration; /* HW arbitration */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* Disable Channel */ +struct ncsi_cmd_dc_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char reserved[3]; /* Reserved */ + unsigned char ald; /* Allow link down */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* Reset Channel */ +struct ncsi_cmd_rc_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be32 reserved; /* Reserved */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* AEN Enable */ +struct ncsi_cmd_ae_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char reserved[3]; /* Reserved */ + unsigned char mc_id; /* MC ID */ + __be32 mode; /* AEN working mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[18]; +}; + +/* Set Link */ +struct ncsi_cmd_sl_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be32 mode; /* Link working mode */ + __be32 oem_mode; /* OEM link mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[18]; +}; + +/* Set VLAN Filter */ +struct ncsi_cmd_svf_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be16 reserved; /* Reserved */ + __be16 vlan; /* VLAN ID */ + __be16 reserved1; /* Reserved */ + unsigned char index; /* VLAN table index */ + unsigned char enable; /* Enable or disable */ + __be32 checksum; /* Checksum */ + unsigned char pad[14]; +}; + +/* Enable VLAN */ +struct ncsi_cmd_ev_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char reserved[3]; /* Reserved */ + unsigned char mode; /* VLAN filter mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* Set MAC Address */ +struct ncsi_cmd_sma_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char mac[6]; /* MAC address */ + unsigned char index; /* MAC table index */ + unsigned char at_e; /* Addr type and operation */ + __be32 checksum; /* Checksum */ + unsigned char pad[18]; +}; + +/* Enable Broadcast Filter */ +struct ncsi_cmd_ebf_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be32 mode; /* Filter mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* Enable Global Multicast Filter */ +struct ncsi_cmd_egmf_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + __be32 mode; /* Global MC mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* Set NCSI Flow Control */ +struct ncsi_cmd_snfc_pkt { + struct ncsi_cmd_pkt_hdr cmd; /* Command header */ + unsigned char reserved[3]; /* Reserved */ + unsigned char mode; /* Flow control mode */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* NCSI packet revision */ +#define NCSI_PKT_REVISION 0x01 + +/* NCSI packet commands */ +#define NCSI_PKT_CMD_CIS 0x00 /* Clear Initial State */ +#define NCSI_PKT_CMD_SP 0x01 /* Select Package */ +#define NCSI_PKT_CMD_DP 0x02 /* Deselect Package */ +#define NCSI_PKT_CMD_EC 0x03 /* Enable Channel */ +#define NCSI_PKT_CMD_DC 0x04 /* Disable Channel */ +#define NCSI_PKT_CMD_RC 0x05 /* Reset Channel */ +#define NCSI_PKT_CMD_ECNT 0x06 /* Enable Channel Network Tx */ +#define NCSI_PKT_CMD_DCNT 0x07 /* Disable Channel Network Tx */ +#define NCSI_PKT_CMD_AE 0x08 /* AEN Enable */ +#define NCSI_PKT_CMD_SL 0x09 /* Set Link */ +#define NCSI_PKT_CMD_GLS 0x0a /* Get Link */ +#define NCSI_PKT_CMD_SVF 0x0b /* Set VLAN Filter */ +#define NCSI_PKT_CMD_EV 0x0c /* Enable VLAN */ +#define NCSI_PKT_CMD_DV 0x0d /* Disable VLAN */ +#define NCSI_PKT_CMD_SMA 0x0e /* Set MAC address */ +#define NCSI_PKT_CMD_EBF 0x10 /* Enable Broadcast Filter */ +#define NCSI_PKT_CMD_DBF 0x11 /* Disable Broadcast Filter */ +#define NCSI_PKT_CMD_EGMF 0x12 /* Enable Global Multicast Filter */ +#define NCSI_PKT_CMD_DGMF 0x13 /* Disable Global Multicast Filter */ +#define NCSI_PKT_CMD_SNFC 0x14 /* Set NCSI Flow Control */ +#define NCSI_PKT_CMD_GVI 0x15 /* Get Version ID */ +#define NCSI_PKT_CMD_GC 0x16 /* Get Capabilities */ +#define NCSI_PKT_CMD_GP 0x17 /* Get Parameters */ +#define NCSI_PKT_CMD_GCPS 0x18 /* Get Controller Packet Statistics */ +#define NCSI_PKT_CMD_GNS 0x19 /* Get NCSI Statistics */ +#define NCSI_PKT_CMD_GNPTS 0x1a /* Get NCSI Pass-throu Statistics */ +#define NCSI_PKT_CMD_GPS 0x1b /* Get package status */ +#define NCSI_PKT_CMD_OEM 0x50 /* OEM */ +#define NCSI_PKT_CMD_PLDM 0x51 /* PLDM request over NCSI over RBT */ +#define NCSI_PKT_CMD_GPUUID 0x52 /* Get package UUID */ + +#endif /* __NCSI_PKT_H__ */ -- cgit v0.10.2 From 138635cc27c9737f940c3aa80912ff7a61c825af Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:18 +1000 Subject: net/ncsi: NCSI response packet handler The NCSI response packets are sent to MC (Management Controller) from the remote end. They are responses of NCSI command packets for multiple purposes: completion status of NCSI command packets, return NCSI channel's capability or configuration etc. This defines struct to represent NCSI response packets and introduces function ncsi_rcv_rsp() which will be used to receive NCSI response packets and parse them. Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller diff --git a/net/ncsi/Makefile b/net/ncsi/Makefile index abc4046..4751819 100644 --- a/net/ncsi/Makefile +++ b/net/ncsi/Makefile @@ -1,4 +1,4 @@ # # Makefile for NCSI API # -obj-$(CONFIG_NET_NCSI) += ncsi-cmd.o ncsi-manage.o +obj-$(CONFIG_NET_NCSI) += ncsi-cmd.o ncsi-rsp.o ncsi-manage.o diff --git a/net/ncsi/internal.h b/net/ncsi/internal.h index 3d81697..bd000c9 100644 --- a/net/ncsi/internal.h +++ b/net/ncsi/internal.h @@ -271,5 +271,7 @@ struct ncsi_dev *ncsi_find_dev(struct net_device *dev); /* Packet handlers */ u32 ncsi_calculate_checksum(unsigned char *data, int len); int ncsi_xmit_cmd(struct ncsi_cmd_arg *nca); +int ncsi_rcv_rsp(struct sk_buff *skb, struct net_device *dev, + struct packet_type *pt, struct net_device *orig_dev); #endif /* __NCSI_INTERNAL_H__ */ diff --git a/net/ncsi/ncsi-pkt.h b/net/ncsi/ncsi-pkt.h index 5481458..4bdefd9 100644 --- a/net/ncsi/ncsi-pkt.h +++ b/net/ncsi/ncsi-pkt.h @@ -25,6 +25,12 @@ struct ncsi_cmd_pkt_hdr { struct ncsi_pkt_hdr common; /* Common NCSI packet header */ }; +struct ncsi_rsp_pkt_hdr { + struct ncsi_pkt_hdr common; /* Common NCSI packet header */ + __be16 code; /* Response code */ + __be16 reason; /* Response reason */ +}; + /* NCSI common command packet */ struct ncsi_cmd_pkt { struct ncsi_cmd_pkt_hdr cmd; /* Command header */ @@ -32,6 +38,12 @@ struct ncsi_cmd_pkt { unsigned char pad[26]; }; +struct ncsi_rsp_pkt { + struct ncsi_rsp_pkt_hdr rsp; /* Response header */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + /* Select Package */ struct ncsi_cmd_sp_pkt { struct ncsi_cmd_pkt_hdr cmd; /* Command header */ @@ -133,6 +145,157 @@ struct ncsi_cmd_snfc_pkt { unsigned char pad[22]; }; +/* Get Link Status */ +struct ncsi_rsp_gls_pkt { + struct ncsi_rsp_pkt_hdr rsp; /* Response header */ + __be32 status; /* Link status */ + __be32 other; /* Other indications */ + __be32 oem_status; /* OEM link status */ + __be32 checksum; + unsigned char pad[10]; +}; + +/* Get Version ID */ +struct ncsi_rsp_gvi_pkt { + struct ncsi_rsp_pkt_hdr rsp; /* Response header */ + __be32 ncsi_version; /* NCSI version */ + unsigned char reserved[3]; /* Reserved */ + unsigned char alpha2; /* NCSI version */ + unsigned char fw_name[12]; /* f/w name string */ + __be32 fw_version; /* f/w version */ + __be16 pci_ids[4]; /* PCI IDs */ + __be32 mf_id; /* Manufacture ID */ + __be32 checksum; +}; + +/* Get Capabilities */ +struct ncsi_rsp_gc_pkt { + struct ncsi_rsp_pkt_hdr rsp; /* Response header */ + __be32 cap; /* Capabilities */ + __be32 bc_cap; /* Broadcast cap */ + __be32 mc_cap; /* Multicast cap */ + __be32 buf_cap; /* Buffering cap */ + __be32 aen_cap; /* AEN cap */ + unsigned char vlan_cnt; /* VLAN filter count */ + unsigned char mixed_cnt; /* Mix filter count */ + unsigned char mc_cnt; /* MC filter count */ + unsigned char uc_cnt; /* UC filter count */ + unsigned char reserved[2]; /* Reserved */ + unsigned char vlan_mode; /* VLAN mode */ + unsigned char channel_cnt; /* Channel count */ + __be32 checksum; /* Checksum */ +}; + +/* Get Parameters */ +struct ncsi_rsp_gp_pkt { + struct ncsi_rsp_pkt_hdr rsp; /* Response header */ + unsigned char mac_cnt; /* Number of MAC addr */ + unsigned char reserved[2]; /* Reserved */ + unsigned char mac_enable; /* MAC addr enable flags */ + unsigned char vlan_cnt; /* VLAN tag count */ + unsigned char reserved1; /* Reserved */ + __be16 vlan_enable; /* VLAN tag enable flags */ + __be32 link_mode; /* Link setting */ + __be32 bc_mode; /* BC filter mode */ + __be32 valid_modes; /* Valid mode parameters */ + unsigned char vlan_mode; /* VLAN mode */ + unsigned char fc_mode; /* Flow control mode */ + unsigned char reserved2[2]; /* Reserved */ + __be32 aen_mode; /* AEN mode */ + unsigned char mac[6]; /* Supported MAC addr */ + __be16 vlan; /* Supported VLAN tags */ + __be32 checksum; /* Checksum */ +}; + +/* Get Controller Packet Statistics */ +struct ncsi_rsp_gcps_pkt { + struct ncsi_rsp_pkt_hdr rsp; /* Response header */ + __be32 cnt_hi; /* Counter cleared */ + __be32 cnt_lo; /* Counter cleared */ + __be32 rx_bytes; /* Rx bytes */ + __be32 tx_bytes; /* Tx bytes */ + __be32 rx_uc_pkts; /* Rx UC packets */ + __be32 rx_mc_pkts; /* Rx MC packets */ + __be32 rx_bc_pkts; /* Rx BC packets */ + __be32 tx_uc_pkts; /* Tx UC packets */ + __be32 tx_mc_pkts; /* Tx MC packets */ + __be32 tx_bc_pkts; /* Tx BC packets */ + __be32 fcs_err; /* FCS errors */ + __be32 align_err; /* Alignment errors */ + __be32 false_carrier; /* False carrier detection */ + __be32 runt_pkts; /* Rx runt packets */ + __be32 jabber_pkts; /* Rx jabber packets */ + __be32 rx_pause_xon; /* Rx pause XON frames */ + __be32 rx_pause_xoff; /* Rx XOFF frames */ + __be32 tx_pause_xon; /* Tx XON frames */ + __be32 tx_pause_xoff; /* Tx XOFF frames */ + __be32 tx_s_collision; /* Single collision frames */ + __be32 tx_m_collision; /* Multiple collision frames */ + __be32 l_collision; /* Late collision frames */ + __be32 e_collision; /* Excessive collision frames */ + __be32 rx_ctl_frames; /* Rx control frames */ + __be32 rx_64_frames; /* Rx 64-bytes frames */ + __be32 rx_127_frames; /* Rx 65-127 bytes frames */ + __be32 rx_255_frames; /* Rx 128-255 bytes frames */ + __be32 rx_511_frames; /* Rx 256-511 bytes frames */ + __be32 rx_1023_frames; /* Rx 512-1023 bytes frames */ + __be32 rx_1522_frames; /* Rx 1024-1522 bytes frames */ + __be32 rx_9022_frames; /* Rx 1523-9022 bytes frames */ + __be32 tx_64_frames; /* Tx 64-bytes frames */ + __be32 tx_127_frames; /* Tx 65-127 bytes frames */ + __be32 tx_255_frames; /* Tx 128-255 bytes frames */ + __be32 tx_511_frames; /* Tx 256-511 bytes frames */ + __be32 tx_1023_frames; /* Tx 512-1023 bytes frames */ + __be32 tx_1522_frames; /* Tx 1024-1522 bytes frames */ + __be32 tx_9022_frames; /* Tx 1523-9022 bytes frames */ + __be32 rx_valid_bytes; /* Rx valid bytes */ + __be32 rx_runt_pkts; /* Rx error runt packets */ + __be32 rx_jabber_pkts; /* Rx error jabber packets */ + __be32 checksum; /* Checksum */ +}; + +/* Get NCSI Statistics */ +struct ncsi_rsp_gns_pkt { + struct ncsi_rsp_pkt_hdr rsp; /* Response header */ + __be32 rx_cmds; /* Rx NCSI commands */ + __be32 dropped_cmds; /* Dropped commands */ + __be32 cmd_type_errs; /* Command type errors */ + __be32 cmd_csum_errs; /* Command checksum errors */ + __be32 rx_pkts; /* Rx NCSI packets */ + __be32 tx_pkts; /* Tx NCSI packets */ + __be32 tx_aen_pkts; /* Tx AEN packets */ + __be32 checksum; /* Checksum */ +}; + +/* Get NCSI Pass-through Statistics */ +struct ncsi_rsp_gnpts_pkt { + struct ncsi_rsp_pkt_hdr rsp; /* Response header */ + __be32 tx_pkts; /* Tx packets */ + __be32 tx_dropped; /* Tx dropped packets */ + __be32 tx_channel_err; /* Tx channel errors */ + __be32 tx_us_err; /* Tx undersize errors */ + __be32 rx_pkts; /* Rx packets */ + __be32 rx_dropped; /* Rx dropped packets */ + __be32 rx_channel_err; /* Rx channel errors */ + __be32 rx_us_err; /* Rx undersize errors */ + __be32 rx_os_err; /* Rx oversize errors */ + __be32 checksum; /* Checksum */ +}; + +/* Get package status */ +struct ncsi_rsp_gps_pkt { + struct ncsi_rsp_pkt_hdr rsp; /* Response header */ + __be32 status; /* Hardware arbitration status */ + __be32 checksum; +}; + +/* Get package UUID */ +struct ncsi_rsp_gpuuid_pkt { + struct ncsi_rsp_pkt_hdr rsp; /* Response header */ + unsigned char uuid[16]; /* UUID */ + __be32 checksum; +}; + /* NCSI packet revision */ #define NCSI_PKT_REVISION 0x01 @@ -168,4 +331,49 @@ struct ncsi_cmd_snfc_pkt { #define NCSI_PKT_CMD_PLDM 0x51 /* PLDM request over NCSI over RBT */ #define NCSI_PKT_CMD_GPUUID 0x52 /* Get package UUID */ +/* NCSI packet responses */ +#define NCSI_PKT_RSP_CIS (NCSI_PKT_CMD_CIS + 0x80) +#define NCSI_PKT_RSP_SP (NCSI_PKT_CMD_SP + 0x80) +#define NCSI_PKT_RSP_DP (NCSI_PKT_CMD_DP + 0x80) +#define NCSI_PKT_RSP_EC (NCSI_PKT_CMD_EC + 0x80) +#define NCSI_PKT_RSP_DC (NCSI_PKT_CMD_DC + 0x80) +#define NCSI_PKT_RSP_RC (NCSI_PKT_CMD_RC + 0x80) +#define NCSI_PKT_RSP_ECNT (NCSI_PKT_CMD_ECNT + 0x80) +#define NCSI_PKT_RSP_DCNT (NCSI_PKT_CMD_DCNT + 0x80) +#define NCSI_PKT_RSP_AE (NCSI_PKT_CMD_AE + 0x80) +#define NCSI_PKT_RSP_SL (NCSI_PKT_CMD_SL + 0x80) +#define NCSI_PKT_RSP_GLS (NCSI_PKT_CMD_GLS + 0x80) +#define NCSI_PKT_RSP_SVF (NCSI_PKT_CMD_SVF + 0x80) +#define NCSI_PKT_RSP_EV (NCSI_PKT_CMD_EV + 0x80) +#define NCSI_PKT_RSP_DV (NCSI_PKT_CMD_DV + 0x80) +#define NCSI_PKT_RSP_SMA (NCSI_PKT_CMD_SMA + 0x80) +#define NCSI_PKT_RSP_EBF (NCSI_PKT_CMD_EBF + 0x80) +#define NCSI_PKT_RSP_DBF (NCSI_PKT_CMD_DBF + 0x80) +#define NCSI_PKT_RSP_EGMF (NCSI_PKT_CMD_EGMF + 0x80) +#define NCSI_PKT_RSP_DGMF (NCSI_PKT_CMD_DGMF + 0x80) +#define NCSI_PKT_RSP_SNFC (NCSI_PKT_CMD_SNFC + 0x80) +#define NCSI_PKT_RSP_GVI (NCSI_PKT_CMD_GVI + 0x80) +#define NCSI_PKT_RSP_GC (NCSI_PKT_CMD_GC + 0x80) +#define NCSI_PKT_RSP_GP (NCSI_PKT_CMD_GP + 0x80) +#define NCSI_PKT_RSP_GCPS (NCSI_PKT_CMD_GCPS + 0x80) +#define NCSI_PKT_RSP_GNS (NCSI_PKT_CMD_GNS + 0x80) +#define NCSI_PKT_RSP_GNPTS (NCSI_PKT_CMD_GNPTS + 0x80) +#define NCSI_PKT_RSP_GPS (NCSI_PKT_CMD_GPS + 0x80) +#define NCSI_PKT_RSP_OEM (NCSI_PKT_CMD_OEM + 0x80) +#define NCSI_PKT_RSP_PLDM (NCSI_PKT_CMD_PLDM + 0x80) +#define NCSI_PKT_RSP_GPUUID (NCSI_PKT_CMD_GPUUID + 0x80) + +/* NCSI response code/reason */ +#define NCSI_PKT_RSP_C_COMPLETED 0x0000 /* Command Completed */ +#define NCSI_PKT_RSP_C_FAILED 0x0001 /* Command Failed */ +#define NCSI_PKT_RSP_C_UNAVAILABLE 0x0002 /* Command Unavailable */ +#define NCSI_PKT_RSP_C_UNSUPPORTED 0x0003 /* Command Unsupported */ +#define NCSI_PKT_RSP_R_NO_ERROR 0x0000 /* No Error */ +#define NCSI_PKT_RSP_R_INTERFACE 0x0001 /* Interface not ready */ +#define NCSI_PKT_RSP_R_PARAM 0x0002 /* Invalid Parameter */ +#define NCSI_PKT_RSP_R_CHANNEL 0x0003 /* Channel not Ready */ +#define NCSI_PKT_RSP_R_PACKAGE 0x0004 /* Package not Ready */ +#define NCSI_PKT_RSP_R_LENGTH 0x0005 /* Invalid payload length */ +#define NCSI_PKT_RSP_R_UNKNOWN 0x7fff /* Command type unsupported */ + #endif /* __NCSI_PKT_H__ */ diff --git a/net/ncsi/ncsi-rsp.c b/net/ncsi/ncsi-rsp.c new file mode 100644 index 0000000..6ec25cb --- /dev/null +++ b/net/ncsi/ncsi-rsp.c @@ -0,0 +1,1016 @@ +/* + * Copyright Gavin Shan, IBM Corporation 2016. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "internal.h" +#include "ncsi-pkt.h" + +static int ncsi_validate_rsp_pkt(struct ncsi_request *nr, + unsigned short payload) +{ + struct ncsi_rsp_pkt_hdr *h; + u32 checksum; + __be32 *pchecksum; + + /* Check NCSI packet header. We don't need validate + * the packet type, which should have been checked + * before calling this function. + */ + h = (struct ncsi_rsp_pkt_hdr *)skb_network_header(nr->rsp); + if (h->common.revision != NCSI_PKT_REVISION) + return -EINVAL; + if (ntohs(h->common.length) != payload) + return -EINVAL; + + /* Check on code and reason */ + if (ntohs(h->code) != NCSI_PKT_RSP_C_COMPLETED || + ntohs(h->reason) != NCSI_PKT_RSP_R_NO_ERROR) + return -EINVAL; + + /* Validate checksum, which might be zeroes if the + * sender doesn't support checksum according to NCSI + * specification. + */ + pchecksum = (__be32 *)((void *)(h + 1) + payload - 4); + if (ntohl(*pchecksum) == 0) + return 0; + + checksum = ncsi_calculate_checksum((unsigned char *)h, + sizeof(*h) + payload - 4); + if (*pchecksum != htonl(checksum)) + return -EINVAL; + + return 0; +} + +static int ncsi_rsp_handler_cis(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_package *np; + struct ncsi_channel *nc; + unsigned char id; + + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, &np, &nc); + if (!nc) { + id = NCSI_CHANNEL_INDEX(rsp->rsp.common.channel); + nc = ncsi_add_channel(np, id); + } + + return nc ? 0 : -ENODEV; +} + +static int ncsi_rsp_handler_sp(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_package *np; + unsigned char id; + + /* Add the package if it's not existing. Otherwise, + * to change the state of its child channels. + */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + &np, NULL); + if (!np) { + id = NCSI_PACKAGE_INDEX(rsp->rsp.common.channel); + np = ncsi_add_package(ndp, id); + if (!np) + return -ENODEV; + } + + return 0; +} + +static int ncsi_rsp_handler_dp(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_package *np; + struct ncsi_channel *nc; + unsigned long flags; + + /* Find the package */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + &np, NULL); + if (!np) + return -ENODEV; + + /* Change state of all channels attached to the package */ + NCSI_FOR_EACH_CHANNEL(np, nc) { + spin_lock_irqsave(&nc->lock, flags); + nc->state = NCSI_CHANNEL_INACTIVE; + spin_unlock_irqrestore(&nc->lock, flags); + } + + return 0; +} + +static int ncsi_rsp_handler_ec(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + ncm = &nc->modes[NCSI_MODE_ENABLE]; + if (ncm->enable) + return -EBUSY; + + ncm->enable = 1; + return 0; +} + +static int ncsi_rsp_handler_dc(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + int ret; + + ret = ncsi_validate_rsp_pkt(nr, 4); + if (ret) + return ret; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + ncm = &nc->modes[NCSI_MODE_ENABLE]; + if (!ncm->enable) + return -EBUSY; + + ncm->enable = 0; + return 0; +} + +static int ncsi_rsp_handler_rc(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + unsigned long flags; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Update state for the specified channel */ + spin_lock_irqsave(&nc->lock, flags); + nc->state = NCSI_CHANNEL_INACTIVE; + spin_unlock_irqrestore(&nc->lock, flags); + + return 0; +} + +static int ncsi_rsp_handler_ecnt(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + ncm = &nc->modes[NCSI_MODE_TX_ENABLE]; + if (ncm->enable) + return -EBUSY; + + ncm->enable = 1; + return 0; +} + +static int ncsi_rsp_handler_dcnt(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + ncm = &nc->modes[NCSI_MODE_TX_ENABLE]; + if (!ncm->enable) + return -EBUSY; + + ncm->enable = 1; + return 0; +} + +static int ncsi_rsp_handler_ae(struct ncsi_request *nr) +{ + struct ncsi_cmd_ae_pkt *cmd; + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Check if the AEN has been enabled */ + ncm = &nc->modes[NCSI_MODE_AEN]; + if (ncm->enable) + return -EBUSY; + + /* Update to AEN configuration */ + cmd = (struct ncsi_cmd_ae_pkt *)skb_network_header(nr->cmd); + ncm->enable = 1; + ncm->data[0] = cmd->mc_id; + ncm->data[1] = ntohl(cmd->mode); + + return 0; +} + +static int ncsi_rsp_handler_sl(struct ncsi_request *nr) +{ + struct ncsi_cmd_sl_pkt *cmd; + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + cmd = (struct ncsi_cmd_sl_pkt *)skb_network_header(nr->cmd); + ncm = &nc->modes[NCSI_MODE_LINK]; + ncm->data[0] = ntohl(cmd->mode); + ncm->data[1] = ntohl(cmd->oem_mode); + + return 0; +} + +static int ncsi_rsp_handler_gls(struct ncsi_request *nr) +{ + struct ncsi_rsp_gls_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_gls_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + ncm = &nc->modes[NCSI_MODE_LINK]; + ncm->data[2] = ntohl(rsp->status); + ncm->data[3] = ntohl(rsp->other); + ncm->data[4] = ntohl(rsp->oem_status); + + return 0; +} + +static int ncsi_rsp_handler_svf(struct ncsi_request *nr) +{ + struct ncsi_cmd_svf_pkt *cmd; + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_filter *ncf; + unsigned short vlan; + int ret; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + cmd = (struct ncsi_cmd_svf_pkt *)skb_network_header(nr->cmd); + ncf = nc->filters[NCSI_FILTER_VLAN]; + if (!ncf) + return -ENOENT; + if (cmd->index >= ncf->total) + return -ERANGE; + + /* Add or remove the VLAN filter */ + if (!(cmd->enable & 0x1)) { + ret = ncsi_remove_filter(nc, NCSI_FILTER_VLAN, cmd->index); + } else { + vlan = ntohs(cmd->vlan); + ret = ncsi_add_filter(nc, NCSI_FILTER_VLAN, &vlan); + } + + return ret; +} + +static int ncsi_rsp_handler_ev(struct ncsi_request *nr) +{ + struct ncsi_cmd_ev_pkt *cmd; + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Check if VLAN mode has been enabled */ + ncm = &nc->modes[NCSI_MODE_VLAN]; + if (ncm->enable) + return -EBUSY; + + /* Update to VLAN mode */ + cmd = (struct ncsi_cmd_ev_pkt *)skb_network_header(nr->cmd); + ncm->enable = 1; + ncm->data[0] = ntohl(cmd->mode); + + return 0; +} + +static int ncsi_rsp_handler_dv(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Check if VLAN mode has been enabled */ + ncm = &nc->modes[NCSI_MODE_VLAN]; + if (!ncm->enable) + return -EBUSY; + + /* Update to VLAN mode */ + ncm->enable = 0; + return 0; +} + +static int ncsi_rsp_handler_sma(struct ncsi_request *nr) +{ + struct ncsi_cmd_sma_pkt *cmd; + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_filter *ncf; + void *bitmap; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* According to NCSI spec 1.01, the mixed filter table + * isn't supported yet. + */ + cmd = (struct ncsi_cmd_sma_pkt *)skb_network_header(nr->cmd); + switch (cmd->at_e >> 5) { + case 0x0: /* UC address */ + ncf = nc->filters[NCSI_FILTER_UC]; + break; + case 0x1: /* MC address */ + ncf = nc->filters[NCSI_FILTER_MC]; + break; + default: + return -EINVAL; + } + + /* Sanity check on the filter */ + if (!ncf) + return -ENOENT; + else if (cmd->index >= ncf->total) + return -ERANGE; + + bitmap = &ncf->bitmap; + if (cmd->at_e & 0x1) { + if (test_and_set_bit(cmd->index, bitmap)) + return -EBUSY; + memcpy(ncf->data + 6 * cmd->index, cmd->mac, 6); + } else { + if (!test_and_clear_bit(cmd->index, bitmap)) + return -EBUSY; + + memset(ncf->data + 6 * cmd->index, 0, 6); + } + + return 0; +} + +static int ncsi_rsp_handler_ebf(struct ncsi_request *nr) +{ + struct ncsi_cmd_ebf_pkt *cmd; + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the package and channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, NULL, &nc); + if (!nc) + return -ENODEV; + + /* Check if broadcast filter has been enabled */ + ncm = &nc->modes[NCSI_MODE_BC]; + if (ncm->enable) + return -EBUSY; + + /* Update to broadcast filter mode */ + cmd = (struct ncsi_cmd_ebf_pkt *)skb_network_header(nr->cmd); + ncm->enable = 1; + ncm->data[0] = ntohl(cmd->mode); + + return 0; +} + +static int ncsi_rsp_handler_dbf(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Check if broadcast filter isn't enabled */ + ncm = &nc->modes[NCSI_MODE_BC]; + if (!ncm->enable) + return -EBUSY; + + /* Update to broadcast filter mode */ + ncm->enable = 0; + ncm->data[0] = 0; + + return 0; +} + +static int ncsi_rsp_handler_egmf(struct ncsi_request *nr) +{ + struct ncsi_cmd_egmf_pkt *cmd; + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Check if multicast filter has been enabled */ + ncm = &nc->modes[NCSI_MODE_MC]; + if (ncm->enable) + return -EBUSY; + + /* Update to multicast filter mode */ + cmd = (struct ncsi_cmd_egmf_pkt *)skb_network_header(nr->cmd); + ncm->enable = 1; + ncm->data[0] = ntohl(cmd->mode); + + return 0; +} + +static int ncsi_rsp_handler_dgmf(struct ncsi_request *nr) +{ + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Check if multicast filter has been enabled */ + ncm = &nc->modes[NCSI_MODE_MC]; + if (!ncm->enable) + return -EBUSY; + + /* Update to multicast filter mode */ + ncm->enable = 0; + ncm->data[0] = 0; + + return 0; +} + +static int ncsi_rsp_handler_snfc(struct ncsi_request *nr) +{ + struct ncsi_cmd_snfc_pkt *cmd; + struct ncsi_rsp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + + /* Find the channel */ + rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Check if flow control has been enabled */ + ncm = &nc->modes[NCSI_MODE_FC]; + if (ncm->enable) + return -EBUSY; + + /* Update to flow control mode */ + cmd = (struct ncsi_cmd_snfc_pkt *)skb_network_header(nr->cmd); + ncm->enable = 1; + ncm->data[0] = cmd->mode; + + return 0; +} + +static int ncsi_rsp_handler_gvi(struct ncsi_request *nr) +{ + struct ncsi_rsp_gvi_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_version *ncv; + int i; + + /* Find the channel */ + rsp = (struct ncsi_rsp_gvi_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Update to channel's version info */ + ncv = &nc->version; + ncv->version = ntohl(rsp->ncsi_version); + ncv->alpha2 = rsp->alpha2; + memcpy(ncv->fw_name, rsp->fw_name, 12); + ncv->fw_version = ntohl(rsp->fw_version); + for (i = 0; i < ARRAY_SIZE(ncv->pci_ids); i++) + ncv->pci_ids[i] = ntohs(rsp->pci_ids[i]); + ncv->mf_id = ntohl(rsp->mf_id); + + return 0; +} + +static int ncsi_rsp_handler_gc(struct ncsi_request *nr) +{ + struct ncsi_rsp_gc_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_filter *ncf; + size_t size, entry_size; + int cnt, i; + + /* Find the channel */ + rsp = (struct ncsi_rsp_gc_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Update channel's capabilities */ + nc->caps[NCSI_CAP_GENERIC].cap = ntohl(rsp->cap) & + NCSI_CAP_GENERIC_MASK; + nc->caps[NCSI_CAP_BC].cap = ntohl(rsp->bc_cap) & + NCSI_CAP_BC_MASK; + nc->caps[NCSI_CAP_MC].cap = ntohl(rsp->mc_cap) & + NCSI_CAP_MC_MASK; + nc->caps[NCSI_CAP_BUFFER].cap = ntohl(rsp->buf_cap); + nc->caps[NCSI_CAP_AEN].cap = ntohl(rsp->aen_cap) & + NCSI_CAP_AEN_MASK; + nc->caps[NCSI_CAP_VLAN].cap = rsp->vlan_mode & + NCSI_CAP_VLAN_MASK; + + /* Build filters */ + for (i = 0; i < NCSI_FILTER_MAX; i++) { + switch (i) { + case NCSI_FILTER_VLAN: + cnt = rsp->vlan_cnt; + entry_size = 2; + break; + case NCSI_FILTER_MIXED: + cnt = rsp->mixed_cnt; + entry_size = 6; + break; + case NCSI_FILTER_MC: + cnt = rsp->mc_cnt; + entry_size = 6; + break; + case NCSI_FILTER_UC: + cnt = rsp->uc_cnt; + entry_size = 6; + break; + default: + continue; + } + + if (!cnt || nc->filters[i]) + continue; + + size = sizeof(*ncf) + cnt * entry_size; + ncf = kzalloc(size, GFP_ATOMIC); + if (!ncf) { + pr_warn("%s: Cannot alloc filter table (%d)\n", + __func__, i); + return -ENOMEM; + } + + ncf->index = i; + ncf->total = cnt; + ncf->bitmap = 0x0ul; + nc->filters[i] = ncf; + } + + return 0; +} + +static int ncsi_rsp_handler_gp(struct ncsi_request *nr) +{ + struct ncsi_rsp_gp_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + unsigned short enable, vlan; + unsigned char *pdata; + int table, i; + + /* Find the channel */ + rsp = (struct ncsi_rsp_gp_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Modes with explicit enabled indications */ + if (ntohl(rsp->valid_modes) & 0x1) { /* BC filter mode */ + nc->modes[NCSI_MODE_BC].enable = 1; + nc->modes[NCSI_MODE_BC].data[0] = ntohl(rsp->bc_mode); + } + if (ntohl(rsp->valid_modes) & 0x2) /* Channel enabled */ + nc->modes[NCSI_MODE_ENABLE].enable = 1; + if (ntohl(rsp->valid_modes) & 0x4) /* Channel Tx enabled */ + nc->modes[NCSI_MODE_TX_ENABLE].enable = 1; + if (ntohl(rsp->valid_modes) & 0x8) /* MC filter mode */ + nc->modes[NCSI_MODE_MC].enable = 1; + + /* Modes without explicit enabled indications */ + nc->modes[NCSI_MODE_LINK].enable = 1; + nc->modes[NCSI_MODE_LINK].data[0] = ntohl(rsp->link_mode); + nc->modes[NCSI_MODE_VLAN].enable = 1; + nc->modes[NCSI_MODE_VLAN].data[0] = rsp->vlan_mode; + nc->modes[NCSI_MODE_FC].enable = 1; + nc->modes[NCSI_MODE_FC].data[0] = rsp->fc_mode; + nc->modes[NCSI_MODE_AEN].enable = 1; + nc->modes[NCSI_MODE_AEN].data[0] = ntohl(rsp->aen_mode); + + /* MAC addresses filter table */ + pdata = (unsigned char *)rsp + 48; + enable = rsp->mac_enable; + for (i = 0; i < rsp->mac_cnt; i++, pdata += 6) { + if (i >= (nc->filters[NCSI_FILTER_UC]->total + + nc->filters[NCSI_FILTER_MC]->total)) + table = NCSI_FILTER_MIXED; + else if (i >= nc->filters[NCSI_FILTER_UC]->total) + table = NCSI_FILTER_MC; + else + table = NCSI_FILTER_UC; + + if (!(enable & (0x1 << i))) + continue; + + if (ncsi_find_filter(nc, table, pdata) >= 0) + continue; + + ncsi_add_filter(nc, table, pdata); + } + + /* VLAN filter table */ + enable = ntohs(rsp->vlan_enable); + for (i = 0; i < rsp->vlan_cnt; i++, pdata += 2) { + if (!(enable & (0x1 << i))) + continue; + + vlan = ntohs(*(__be16 *)pdata); + if (ncsi_find_filter(nc, NCSI_FILTER_VLAN, &vlan) >= 0) + continue; + + ncsi_add_filter(nc, NCSI_FILTER_VLAN, &vlan); + } + + return 0; +} + +static int ncsi_rsp_handler_gcps(struct ncsi_request *nr) +{ + struct ncsi_rsp_gcps_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_stats *ncs; + + /* Find the channel */ + rsp = (struct ncsi_rsp_gcps_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Update HNC's statistics */ + ncs = &nc->stats; + ncs->hnc_cnt_hi = ntohl(rsp->cnt_hi); + ncs->hnc_cnt_lo = ntohl(rsp->cnt_lo); + ncs->hnc_rx_bytes = ntohl(rsp->rx_bytes); + ncs->hnc_tx_bytes = ntohl(rsp->tx_bytes); + ncs->hnc_rx_uc_pkts = ntohl(rsp->rx_uc_pkts); + ncs->hnc_rx_mc_pkts = ntohl(rsp->rx_mc_pkts); + ncs->hnc_rx_bc_pkts = ntohl(rsp->rx_bc_pkts); + ncs->hnc_tx_uc_pkts = ntohl(rsp->tx_uc_pkts); + ncs->hnc_tx_mc_pkts = ntohl(rsp->tx_mc_pkts); + ncs->hnc_tx_bc_pkts = ntohl(rsp->tx_bc_pkts); + ncs->hnc_fcs_err = ntohl(rsp->fcs_err); + ncs->hnc_align_err = ntohl(rsp->align_err); + ncs->hnc_false_carrier = ntohl(rsp->false_carrier); + ncs->hnc_runt_pkts = ntohl(rsp->runt_pkts); + ncs->hnc_jabber_pkts = ntohl(rsp->jabber_pkts); + ncs->hnc_rx_pause_xon = ntohl(rsp->rx_pause_xon); + ncs->hnc_rx_pause_xoff = ntohl(rsp->rx_pause_xoff); + ncs->hnc_tx_pause_xon = ntohl(rsp->tx_pause_xon); + ncs->hnc_tx_pause_xoff = ntohl(rsp->tx_pause_xoff); + ncs->hnc_tx_s_collision = ntohl(rsp->tx_s_collision); + ncs->hnc_tx_m_collision = ntohl(rsp->tx_m_collision); + ncs->hnc_l_collision = ntohl(rsp->l_collision); + ncs->hnc_e_collision = ntohl(rsp->e_collision); + ncs->hnc_rx_ctl_frames = ntohl(rsp->rx_ctl_frames); + ncs->hnc_rx_64_frames = ntohl(rsp->rx_64_frames); + ncs->hnc_rx_127_frames = ntohl(rsp->rx_127_frames); + ncs->hnc_rx_255_frames = ntohl(rsp->rx_255_frames); + ncs->hnc_rx_511_frames = ntohl(rsp->rx_511_frames); + ncs->hnc_rx_1023_frames = ntohl(rsp->rx_1023_frames); + ncs->hnc_rx_1522_frames = ntohl(rsp->rx_1522_frames); + ncs->hnc_rx_9022_frames = ntohl(rsp->rx_9022_frames); + ncs->hnc_tx_64_frames = ntohl(rsp->tx_64_frames); + ncs->hnc_tx_127_frames = ntohl(rsp->tx_127_frames); + ncs->hnc_tx_255_frames = ntohl(rsp->tx_255_frames); + ncs->hnc_tx_511_frames = ntohl(rsp->tx_511_frames); + ncs->hnc_tx_1023_frames = ntohl(rsp->tx_1023_frames); + ncs->hnc_tx_1522_frames = ntohl(rsp->tx_1522_frames); + ncs->hnc_tx_9022_frames = ntohl(rsp->tx_9022_frames); + ncs->hnc_rx_valid_bytes = ntohl(rsp->rx_valid_bytes); + ncs->hnc_rx_runt_pkts = ntohl(rsp->rx_runt_pkts); + ncs->hnc_rx_jabber_pkts = ntohl(rsp->rx_jabber_pkts); + + return 0; +} + +static int ncsi_rsp_handler_gns(struct ncsi_request *nr) +{ + struct ncsi_rsp_gns_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_stats *ncs; + + /* Find the channel */ + rsp = (struct ncsi_rsp_gns_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Update HNC's statistics */ + ncs = &nc->stats; + ncs->ncsi_rx_cmds = ntohl(rsp->rx_cmds); + ncs->ncsi_dropped_cmds = ntohl(rsp->dropped_cmds); + ncs->ncsi_cmd_type_errs = ntohl(rsp->cmd_type_errs); + ncs->ncsi_cmd_csum_errs = ntohl(rsp->cmd_csum_errs); + ncs->ncsi_rx_pkts = ntohl(rsp->rx_pkts); + ncs->ncsi_tx_pkts = ntohl(rsp->tx_pkts); + ncs->ncsi_tx_aen_pkts = ntohl(rsp->tx_aen_pkts); + + return 0; +} + +static int ncsi_rsp_handler_gnpts(struct ncsi_request *nr) +{ + struct ncsi_rsp_gnpts_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_channel *nc; + struct ncsi_channel_stats *ncs; + + /* Find the channel */ + rsp = (struct ncsi_rsp_gnpts_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + NULL, &nc); + if (!nc) + return -ENODEV; + + /* Update HNC's statistics */ + ncs = &nc->stats; + ncs->pt_tx_pkts = ntohl(rsp->tx_pkts); + ncs->pt_tx_dropped = ntohl(rsp->tx_dropped); + ncs->pt_tx_channel_err = ntohl(rsp->tx_channel_err); + ncs->pt_tx_us_err = ntohl(rsp->tx_us_err); + ncs->pt_rx_pkts = ntohl(rsp->rx_pkts); + ncs->pt_rx_dropped = ntohl(rsp->rx_dropped); + ncs->pt_rx_channel_err = ntohl(rsp->rx_channel_err); + ncs->pt_rx_us_err = ntohl(rsp->rx_us_err); + ncs->pt_rx_os_err = ntohl(rsp->rx_os_err); + + return 0; +} + +static int ncsi_rsp_handler_gps(struct ncsi_request *nr) +{ + struct ncsi_rsp_gps_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_package *np; + + /* Find the package */ + rsp = (struct ncsi_rsp_gps_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + &np, NULL); + if (!np) + return -ENODEV; + + return 0; +} + +static int ncsi_rsp_handler_gpuuid(struct ncsi_request *nr) +{ + struct ncsi_rsp_gpuuid_pkt *rsp; + struct ncsi_dev_priv *ndp = nr->ndp; + struct ncsi_package *np; + + /* Find the package */ + rsp = (struct ncsi_rsp_gpuuid_pkt *)skb_network_header(nr->rsp); + ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, + &np, NULL); + if (!np) + return -ENODEV; + + memcpy(np->uuid, rsp->uuid, sizeof(rsp->uuid)); + + return 0; +} + +static struct ncsi_rsp_handler { + unsigned char type; + int payload; + int (*handler)(struct ncsi_request *nr); +} ncsi_rsp_handlers[] = { + { NCSI_PKT_RSP_CIS, 4, ncsi_rsp_handler_cis }, + { NCSI_PKT_RSP_SP, 4, ncsi_rsp_handler_sp }, + { NCSI_PKT_RSP_DP, 4, ncsi_rsp_handler_dp }, + { NCSI_PKT_RSP_EC, 4, ncsi_rsp_handler_ec }, + { NCSI_PKT_RSP_DC, 4, ncsi_rsp_handler_dc }, + { NCSI_PKT_RSP_RC, 4, ncsi_rsp_handler_rc }, + { NCSI_PKT_RSP_ECNT, 4, ncsi_rsp_handler_ecnt }, + { NCSI_PKT_RSP_DCNT, 4, ncsi_rsp_handler_dcnt }, + { NCSI_PKT_RSP_AE, 4, ncsi_rsp_handler_ae }, + { NCSI_PKT_RSP_SL, 4, ncsi_rsp_handler_sl }, + { NCSI_PKT_RSP_GLS, 16, ncsi_rsp_handler_gls }, + { NCSI_PKT_RSP_SVF, 4, ncsi_rsp_handler_svf }, + { NCSI_PKT_RSP_EV, 4, ncsi_rsp_handler_ev }, + { NCSI_PKT_RSP_DV, 4, ncsi_rsp_handler_dv }, + { NCSI_PKT_RSP_SMA, 4, ncsi_rsp_handler_sma }, + { NCSI_PKT_RSP_EBF, 4, ncsi_rsp_handler_ebf }, + { NCSI_PKT_RSP_DBF, 4, ncsi_rsp_handler_dbf }, + { NCSI_PKT_RSP_EGMF, 4, ncsi_rsp_handler_egmf }, + { NCSI_PKT_RSP_DGMF, 4, ncsi_rsp_handler_dgmf }, + { NCSI_PKT_RSP_SNFC, 4, ncsi_rsp_handler_snfc }, + { NCSI_PKT_RSP_GVI, 36, ncsi_rsp_handler_gvi }, + { NCSI_PKT_RSP_GC, 32, ncsi_rsp_handler_gc }, + { NCSI_PKT_RSP_GP, -1, ncsi_rsp_handler_gp }, + { NCSI_PKT_RSP_GCPS, 172, ncsi_rsp_handler_gcps }, + { NCSI_PKT_RSP_GNS, 172, ncsi_rsp_handler_gns }, + { NCSI_PKT_RSP_GNPTS, 172, ncsi_rsp_handler_gnpts }, + { NCSI_PKT_RSP_GPS, 8, ncsi_rsp_handler_gps }, + { NCSI_PKT_RSP_OEM, 0, NULL }, + { NCSI_PKT_RSP_PLDM, 0, NULL }, + { NCSI_PKT_RSP_GPUUID, 20, ncsi_rsp_handler_gpuuid } +}; + +int ncsi_rcv_rsp(struct sk_buff *skb, struct net_device *dev, + struct packet_type *pt, struct net_device *orig_dev) +{ + struct ncsi_rsp_handler *nrh = NULL; + struct ncsi_dev *nd; + struct ncsi_dev_priv *ndp; + struct ncsi_request *nr; + struct ncsi_pkt_hdr *hdr; + unsigned long flags; + int payload, i, ret; + + /* Find the NCSI device */ + nd = ncsi_find_dev(dev); + ndp = nd ? TO_NCSI_DEV_PRIV(nd) : NULL; + if (!ndp) + return -ENODEV; + + /* Find the handler */ + hdr = (struct ncsi_pkt_hdr *)skb_network_header(skb); + for (i = 0; i < ARRAY_SIZE(ncsi_rsp_handlers); i++) { + if (ncsi_rsp_handlers[i].type == hdr->type) { + if (ncsi_rsp_handlers[i].handler) + nrh = &ncsi_rsp_handlers[i]; + else + nrh = NULL; + + break; + } + } + + if (!nrh) { + netdev_err(nd->dev, "Received unrecognized packet (0x%x)\n", + hdr->type); + return -ENOENT; + } + + /* Associate with the request */ + spin_lock_irqsave(&ndp->lock, flags); + nr = &ndp->requests[hdr->id]; + if (!nr->used) { + spin_unlock_irqrestore(&ndp->lock, flags); + return -ENODEV; + } + + nr->rsp = skb; + if (!nr->enabled) { + spin_unlock_irqrestore(&ndp->lock, flags); + ret = -ENOENT; + goto out; + } + + /* Validate the packet */ + spin_unlock_irqrestore(&ndp->lock, flags); + payload = nrh->payload; + if (payload < 0) + payload = ntohs(hdr->length); + ret = ncsi_validate_rsp_pkt(nr, payload); + if (ret) + goto out; + + /* Process the packet */ + ret = nrh->handler(nr); +out: + ncsi_free_request(nr); + return ret; +} -- cgit v0.10.2 From e6f44ed6d04d3185dcd8e8e98af8742d87bdffcc Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:19 +1000 Subject: net/ncsi: Package and channel management This manages NCSI packages and channels: * The available packages and channels are enumerated in the first time of calling ncsi_start_dev(). The channels' capabilities are probed in the meanwhile. The NCSI network topology won't change until the NCSI device is destroyed. * There in a queue in every NCSI device. The element in the queue, channel, is waiting for configuration (bringup) or suspending (teardown). The channel's state (inactive/active) indicates the futher action (configuration or suspending) will be applied on the channel. Another channel's state (invisible) means the requested action is being applied. * The hardware arbitration will be enabled if all available packages and channels support it. All available channels try to provide service when hardware arbitration is enabled. Otherwise, one channel is selected as the active one at once. * When channel is in active state, meaning it's providing service, a timer started to retrieve the channe's link status. If the channel's link status fails to be updated in the determined period, the channel is going to be reconfigured. It's the error handling implementation as defined in NCSI spec. Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller diff --git a/include/net/ncsi.h b/include/net/ncsi.h index 70d14ee..1dbf42f 100644 --- a/include/net/ncsi.h +++ b/include/net/ncsi.h @@ -30,6 +30,7 @@ struct ncsi_dev { #ifdef CONFIG_NET_NCSI struct ncsi_dev *ncsi_register_dev(struct net_device *dev, void (*notifier)(struct ncsi_dev *nd)); +int ncsi_start_dev(struct ncsi_dev *nd); void ncsi_unregister_dev(struct ncsi_dev *nd); #else /* !CONFIG_NET_NCSI */ static inline struct ncsi_dev *ncsi_register_dev(struct net_device *dev, @@ -38,6 +39,11 @@ static inline struct ncsi_dev *ncsi_register_dev(struct net_device *dev, return NULL; } +static inline int ncsi_start_dev(struct ncsi_dev *nd) +{ + return -ENOTTY; +} + static inline void ncsi_unregister_dev(struct ncsi_dev *nd) { } diff --git a/net/ncsi/internal.h b/net/ncsi/internal.h index bd000c9..38fc95a 100644 --- a/net/ncsi/internal.h +++ b/net/ncsi/internal.h @@ -178,6 +178,7 @@ struct ncsi_channel { int state; #define NCSI_CHANNEL_INACTIVE 1 #define NCSI_CHANNEL_ACTIVE 2 +#define NCSI_CHANNEL_INVISIBLE 3 spinlock_t lock; /* Protect filters etc */ struct ncsi_package *package; struct ncsi_channel_version version; @@ -185,7 +186,11 @@ struct ncsi_channel { struct ncsi_channel_mode modes[NCSI_MODE_MAX]; struct ncsi_channel_filter *filters[NCSI_FILTER_MAX]; struct ncsi_channel_stats stats; + struct timer_list timer; /* Link monitor timer */ + bool enabled; /* Timer is enabled */ + unsigned int timeout; /* Times of timeout */ struct list_head node; + struct list_head link; }; struct ncsi_package { @@ -209,14 +214,56 @@ struct ncsi_request { bool enabled; /* Time has been enabled or not */ }; +enum { + ncsi_dev_state_major = 0xff00, + ncsi_dev_state_minor = 0x00ff, + ncsi_dev_state_probe_deselect = 0x0201, + ncsi_dev_state_probe_package, + ncsi_dev_state_probe_channel, + ncsi_dev_state_probe_cis, + ncsi_dev_state_probe_gvi, + ncsi_dev_state_probe_gc, + ncsi_dev_state_probe_gls, + ncsi_dev_state_probe_dp, + ncsi_dev_state_config_sp = 0x0301, + ncsi_dev_state_config_cis, + ncsi_dev_state_config_sma, + ncsi_dev_state_config_ebf, +#if IS_ENABLED(CONFIG_IPV6) + ncsi_dev_state_config_egmf, +#endif + ncsi_dev_state_config_ecnt, + ncsi_dev_state_config_ec, + ncsi_dev_state_config_ae, + ncsi_dev_state_config_gls, + ncsi_dev_state_config_done, + ncsi_dev_state_suspend_select = 0x0401, + ncsi_dev_state_suspend_dcnt, + ncsi_dev_state_suspend_dc, + ncsi_dev_state_suspend_deselect, + ncsi_dev_state_suspend_done +}; + struct ncsi_dev_priv { struct ncsi_dev ndev; /* Associated NCSI device */ unsigned int flags; /* NCSI device flags */ +#define NCSI_DEV_PROBED 1 /* Finalized NCSI topology */ +#define NCSI_DEV_HWA 2 /* Enabled HW arbitration */ +#define NCSI_DEV_RESHUFFLE 4 spinlock_t lock; /* Protect the NCSI device */ +#if IS_ENABLED(CONFIG_IPV6) + unsigned int inet6_addr_num; /* Number of IPv6 addresses */ +#endif unsigned int package_num; /* Number of packages */ struct list_head packages; /* List of packages */ struct ncsi_request requests[256]; /* Request table */ unsigned int request_id; /* Last used request ID */ + unsigned int pending_req_num; /* Number of pending requests */ + struct ncsi_package *active_package; /* Currently handled package */ + struct ncsi_channel *active_channel; /* Currently handled channel */ + struct list_head channel_queue; /* Config queue of channels */ + struct work_struct work; /* For channel management */ + struct packet_type ptype; /* NCSI packet Rx handler */ struct list_head node; /* Form NCSI device list */ }; @@ -251,6 +298,8 @@ extern spinlock_t ncsi_dev_lock; int ncsi_find_filter(struct ncsi_channel *nc, int table, void *data); int ncsi_add_filter(struct ncsi_channel *nc, int table, void *data); int ncsi_remove_filter(struct ncsi_channel *nc, int table, int index); +void ncsi_start_channel_monitor(struct ncsi_channel *nc); +void ncsi_stop_channel_monitor(struct ncsi_channel *nc); struct ncsi_channel *ncsi_find_channel(struct ncsi_package *np, unsigned char id); struct ncsi_channel *ncsi_add_channel(struct ncsi_package *np, @@ -267,6 +316,7 @@ void ncsi_find_package_and_channel(struct ncsi_dev_priv *ndp, struct ncsi_request *ncsi_alloc_request(struct ncsi_dev_priv *ndp, bool driven); void ncsi_free_request(struct ncsi_request *nr); struct ncsi_dev *ncsi_find_dev(struct net_device *dev); +int ncsi_process_next_channel(struct ncsi_dev_priv *ndp); /* Packet handlers */ u32 ncsi_calculate_checksum(unsigned char *data, int len); diff --git a/net/ncsi/ncsi-manage.c b/net/ncsi/ncsi-manage.c index 0e28ed8..d627a39 100644 --- a/net/ncsi/ncsi-manage.c +++ b/net/ncsi/ncsi-manage.c @@ -17,8 +17,12 @@ #include #include #include +#include +#include +#include #include "internal.h" +#include "ncsi-pkt.h" LIST_HEAD(ncsi_dev_list); DEFINE_SPINLOCK(ncsi_dev_lock); @@ -123,6 +127,120 @@ int ncsi_remove_filter(struct ncsi_channel *nc, int table, int index) return 0; } +static void ncsi_report_link(struct ncsi_dev_priv *ndp, bool force_down) +{ + struct ncsi_dev *nd = &ndp->ndev; + struct ncsi_package *np; + struct ncsi_channel *nc; + + nd->state = ncsi_dev_state_functional; + if (force_down) { + nd->link_up = 0; + goto report; + } + + nd->link_up = 0; + NCSI_FOR_EACH_PACKAGE(ndp, np) { + NCSI_FOR_EACH_CHANNEL(np, nc) { + if (!list_empty(&nc->link) || + nc->state != NCSI_CHANNEL_ACTIVE) + continue; + + if (nc->modes[NCSI_MODE_LINK].data[2] & 0x1) { + nd->link_up = 1; + goto report; + } + } + } + +report: + nd->handler(nd); +} + +static void ncsi_channel_monitor(unsigned long data) +{ + struct ncsi_channel *nc = (struct ncsi_channel *)data; + struct ncsi_package *np = nc->package; + struct ncsi_dev_priv *ndp = np->ndp; + struct ncsi_cmd_arg nca; + bool enabled; + unsigned int timeout; + unsigned long flags; + int ret; + + spin_lock_irqsave(&nc->lock, flags); + timeout = nc->timeout; + enabled = nc->enabled; + spin_unlock_irqrestore(&nc->lock, flags); + + if (!enabled || !list_empty(&nc->link)) + return; + if (nc->state != NCSI_CHANNEL_INACTIVE && + nc->state != NCSI_CHANNEL_ACTIVE) + return; + + if (!(timeout % 2)) { + nca.ndp = ndp; + nca.package = np->id; + nca.channel = nc->id; + nca.type = NCSI_PKT_CMD_GLS; + nca.driven = false; + ret = ncsi_xmit_cmd(&nca); + if (ret) { + netdev_err(ndp->ndev.dev, "Error %d sending GLS\n", + ret); + return; + } + } + + if (timeout + 1 >= 3) { + if (!(ndp->flags & NCSI_DEV_HWA) && + nc->state == NCSI_CHANNEL_ACTIVE) + ncsi_report_link(ndp, true); + + spin_lock_irqsave(&ndp->lock, flags); + xchg(&nc->state, NCSI_CHANNEL_INACTIVE); + list_add_tail_rcu(&nc->link, &ndp->channel_queue); + spin_unlock_irqrestore(&ndp->lock, flags); + ncsi_process_next_channel(ndp); + return; + } + + spin_lock_irqsave(&nc->lock, flags); + nc->timeout = timeout + 1; + nc->enabled = true; + spin_unlock_irqrestore(&nc->lock, flags); + mod_timer(&nc->timer, jiffies + HZ * (1 << (nc->timeout / 2))); +} + +void ncsi_start_channel_monitor(struct ncsi_channel *nc) +{ + unsigned long flags; + + spin_lock_irqsave(&nc->lock, flags); + WARN_ON_ONCE(nc->enabled); + nc->timeout = 0; + nc->enabled = true; + spin_unlock_irqrestore(&nc->lock, flags); + + mod_timer(&nc->timer, jiffies + HZ * (1 << (nc->timeout / 2))); +} + +void ncsi_stop_channel_monitor(struct ncsi_channel *nc) +{ + unsigned long flags; + + spin_lock_irqsave(&nc->lock, flags); + if (!nc->enabled) { + spin_unlock_irqrestore(&nc->lock, flags); + return; + } + nc->enabled = false; + spin_unlock_irqrestore(&nc->lock, flags); + + del_timer_sync(&nc->timer); +} + struct ncsi_channel *ncsi_find_channel(struct ncsi_package *np, unsigned char id) { @@ -149,7 +267,10 @@ struct ncsi_channel *ncsi_add_channel(struct ncsi_package *np, unsigned char id) nc->id = id; nc->package = np; nc->state = NCSI_CHANNEL_INACTIVE; + nc->enabled = false; + setup_timer(&nc->timer, ncsi_channel_monitor, (unsigned long)nc); spin_lock_init(&nc->lock); + INIT_LIST_HEAD(&nc->link); for (index = 0; index < NCSI_CAP_MAX; index++) nc->caps[index].index = index; for (index = 0; index < NCSI_MODE_MAX; index++) @@ -190,6 +311,7 @@ static void ncsi_remove_channel(struct ncsi_channel *nc) nc->state = NCSI_CHANNEL_INACTIVE; spin_unlock_irqrestore(&nc->lock, flags); + ncsi_stop_channel_monitor(nc); /* Remove and free channel */ spin_lock_irqsave(&np->lock, flags); @@ -323,6 +445,7 @@ void ncsi_free_request(struct ncsi_request *nr) struct ncsi_dev_priv *ndp = nr->ndp; struct sk_buff *cmd, *rsp; unsigned long flags; + bool driven; if (nr->enabled) { nr->enabled = false; @@ -335,8 +458,12 @@ void ncsi_free_request(struct ncsi_request *nr) nr->cmd = NULL; nr->rsp = NULL; nr->used = false; + driven = nr->driven; spin_unlock_irqrestore(&ndp->lock, flags); + if (driven && cmd && --ndp->pending_req_num == 0) + schedule_work(&ndp->work); + /* Release command and response */ consume_skb(cmd); consume_skb(rsp); @@ -375,6 +502,587 @@ static void ncsi_request_timeout(unsigned long data) ncsi_free_request(nr); } +static void ncsi_suspend_channel(struct ncsi_dev_priv *ndp) +{ + struct ncsi_dev *nd = &ndp->ndev; + struct ncsi_package *np = ndp->active_package; + struct ncsi_channel *nc = ndp->active_channel; + struct ncsi_cmd_arg nca; + int ret; + + nca.ndp = ndp; + nca.driven = true; + switch (nd->state) { + case ncsi_dev_state_suspend: + nd->state = ncsi_dev_state_suspend_select; + /* Fall through */ + case ncsi_dev_state_suspend_select: + case ncsi_dev_state_suspend_dcnt: + case ncsi_dev_state_suspend_dc: + case ncsi_dev_state_suspend_deselect: + ndp->pending_req_num = 1; + + np = ndp->active_package; + nc = ndp->active_channel; + nca.package = np->id; + if (nd->state == ncsi_dev_state_suspend_select) { + nca.type = NCSI_PKT_CMD_SP; + nca.channel = 0x1f; + if (ndp->flags & NCSI_DEV_HWA) + nca.bytes[0] = 0; + else + nca.bytes[0] = 1; + nd->state = ncsi_dev_state_suspend_dcnt; + } else if (nd->state == ncsi_dev_state_suspend_dcnt) { + nca.type = NCSI_PKT_CMD_DCNT; + nca.channel = nc->id; + nd->state = ncsi_dev_state_suspend_dc; + } else if (nd->state == ncsi_dev_state_suspend_dc) { + nca.type = NCSI_PKT_CMD_DC; + nca.channel = nc->id; + nca.bytes[0] = 1; + nd->state = ncsi_dev_state_suspend_deselect; + } else if (nd->state == ncsi_dev_state_suspend_deselect) { + nca.type = NCSI_PKT_CMD_DP; + nca.channel = 0x1f; + nd->state = ncsi_dev_state_suspend_done; + } + + ret = ncsi_xmit_cmd(&nca); + if (ret) { + nd->state = ncsi_dev_state_functional; + return; + } + + break; + case ncsi_dev_state_suspend_done: + xchg(&nc->state, NCSI_CHANNEL_INACTIVE); + ncsi_process_next_channel(ndp); + + break; + default: + netdev_warn(nd->dev, "Wrong NCSI state 0x%x in suspend\n", + nd->state); + } +} + +static void ncsi_configure_channel(struct ncsi_dev_priv *ndp) +{ + struct ncsi_dev *nd = &ndp->ndev; + struct net_device *dev = nd->dev; + struct ncsi_package *np = ndp->active_package; + struct ncsi_channel *nc = ndp->active_channel; + struct ncsi_cmd_arg nca; + unsigned char index; + int ret; + + nca.ndp = ndp; + nca.driven = true; + switch (nd->state) { + case ncsi_dev_state_config: + case ncsi_dev_state_config_sp: + ndp->pending_req_num = 1; + + /* Select the specific package */ + nca.type = NCSI_PKT_CMD_SP; + if (ndp->flags & NCSI_DEV_HWA) + nca.bytes[0] = 0; + else + nca.bytes[0] = 1; + nca.package = np->id; + nca.channel = 0x1f; + ret = ncsi_xmit_cmd(&nca); + if (ret) + goto error; + + nd->state = ncsi_dev_state_config_cis; + break; + case ncsi_dev_state_config_cis: + ndp->pending_req_num = 1; + + /* Clear initial state */ + nca.type = NCSI_PKT_CMD_CIS; + nca.package = np->id; + nca.channel = nc->id; + ret = ncsi_xmit_cmd(&nca); + if (ret) + goto error; + + nd->state = ncsi_dev_state_config_sma; + break; + case ncsi_dev_state_config_sma: + case ncsi_dev_state_config_ebf: +#if IS_ENABLED(CONFIG_IPV6) + case ncsi_dev_state_config_egmf: +#endif + case ncsi_dev_state_config_ecnt: + case ncsi_dev_state_config_ec: + case ncsi_dev_state_config_ae: + case ncsi_dev_state_config_gls: + ndp->pending_req_num = 1; + + nca.package = np->id; + nca.channel = nc->id; + + /* Use first entry in unicast filter table. Note that + * the MAC filter table starts from entry 1 instead of + * 0. + */ + if (nd->state == ncsi_dev_state_config_sma) { + nca.type = NCSI_PKT_CMD_SMA; + for (index = 0; index < 6; index++) + nca.bytes[index] = dev->dev_addr[index]; + nca.bytes[6] = 0x1; + nca.bytes[7] = 0x1; + nd->state = ncsi_dev_state_config_ebf; + } else if (nd->state == ncsi_dev_state_config_ebf) { + nca.type = NCSI_PKT_CMD_EBF; + nca.dwords[0] = nc->caps[NCSI_CAP_BC].cap; + nd->state = ncsi_dev_state_config_ecnt; +#if IS_ENABLED(CONFIG_IPV6) + if (ndp->inet6_addr_num > 0 && + (nc->caps[NCSI_CAP_GENERIC].cap & + NCSI_CAP_GENERIC_MC)) + nd->state = ncsi_dev_state_config_egmf; + else + nd->state = ncsi_dev_state_config_ecnt; + } else if (nd->state == ncsi_dev_state_config_egmf) { + nca.type = NCSI_PKT_CMD_EGMF; + nca.dwords[0] = nc->caps[NCSI_CAP_MC].cap; + nd->state = ncsi_dev_state_config_ecnt; +#endif /* CONFIG_IPV6 */ + } else if (nd->state == ncsi_dev_state_config_ecnt) { + nca.type = NCSI_PKT_CMD_ECNT; + nd->state = ncsi_dev_state_config_ec; + } else if (nd->state == ncsi_dev_state_config_ec) { + /* Enable AEN if it's supported */ + nca.type = NCSI_PKT_CMD_EC; + nd->state = ncsi_dev_state_config_ae; + if (!(nc->caps[NCSI_CAP_AEN].cap & NCSI_CAP_AEN_MASK)) + nd->state = ncsi_dev_state_config_gls; + } else if (nd->state == ncsi_dev_state_config_ae) { + nca.type = NCSI_PKT_CMD_AE; + nca.bytes[0] = 0; + nca.dwords[1] = nc->caps[NCSI_CAP_AEN].cap; + nd->state = ncsi_dev_state_config_gls; + } else if (nd->state == ncsi_dev_state_config_gls) { + nca.type = NCSI_PKT_CMD_GLS; + nd->state = ncsi_dev_state_config_done; + } + + ret = ncsi_xmit_cmd(&nca); + if (ret) + goto error; + break; + case ncsi_dev_state_config_done: + if (nc->modes[NCSI_MODE_LINK].data[2] & 0x1) + xchg(&nc->state, NCSI_CHANNEL_ACTIVE); + else + xchg(&nc->state, NCSI_CHANNEL_INACTIVE); + + ncsi_start_channel_monitor(nc); + ncsi_process_next_channel(ndp); + break; + default: + netdev_warn(dev, "Wrong NCSI state 0x%x in config\n", + nd->state); + } + + return; + +error: + ncsi_report_link(ndp, true); +} + +static int ncsi_choose_active_channel(struct ncsi_dev_priv *ndp) +{ + struct ncsi_package *np; + struct ncsi_channel *nc, *found; + struct ncsi_channel_mode *ncm; + unsigned long flags; + + /* The search is done once an inactive channel with up + * link is found. + */ + found = NULL; + NCSI_FOR_EACH_PACKAGE(ndp, np) { + NCSI_FOR_EACH_CHANNEL(np, nc) { + if (!list_empty(&nc->link) || + nc->state != NCSI_CHANNEL_INACTIVE) + continue; + + if (!found) + found = nc; + + ncm = &nc->modes[NCSI_MODE_LINK]; + if (ncm->data[2] & 0x1) { + found = nc; + goto out; + } + } + } + + if (!found) { + ncsi_report_link(ndp, true); + return -ENODEV; + } + +out: + spin_lock_irqsave(&ndp->lock, flags); + list_add_tail_rcu(&found->link, &ndp->channel_queue); + spin_unlock_irqrestore(&ndp->lock, flags); + + return ncsi_process_next_channel(ndp); +} + +static bool ncsi_check_hwa(struct ncsi_dev_priv *ndp) +{ + struct ncsi_package *np; + struct ncsi_channel *nc; + unsigned int cap; + + /* The hardware arbitration is disabled if any one channel + * doesn't support explicitly. + */ + NCSI_FOR_EACH_PACKAGE(ndp, np) { + NCSI_FOR_EACH_CHANNEL(np, nc) { + cap = nc->caps[NCSI_CAP_GENERIC].cap; + if (!(cap & NCSI_CAP_GENERIC_HWA) || + (cap & NCSI_CAP_GENERIC_HWA_MASK) != + NCSI_CAP_GENERIC_HWA_SUPPORT) { + ndp->flags &= ~NCSI_DEV_HWA; + return false; + } + } + } + + ndp->flags |= NCSI_DEV_HWA; + return true; +} + +static int ncsi_enable_hwa(struct ncsi_dev_priv *ndp) +{ + struct ncsi_package *np; + struct ncsi_channel *nc; + unsigned long flags; + + /* Move all available channels to processing queue */ + spin_lock_irqsave(&ndp->lock, flags); + NCSI_FOR_EACH_PACKAGE(ndp, np) { + NCSI_FOR_EACH_CHANNEL(np, nc) { + WARN_ON_ONCE(nc->state != NCSI_CHANNEL_INACTIVE || + !list_empty(&nc->link)); + ncsi_stop_channel_monitor(nc); + list_add_tail_rcu(&nc->link, &ndp->channel_queue); + } + } + spin_unlock_irqrestore(&ndp->lock, flags); + + /* We can have no channels in extremely case */ + if (list_empty(&ndp->channel_queue)) { + ncsi_report_link(ndp, false); + return -ENOENT; + } + + return ncsi_process_next_channel(ndp); +} + +static void ncsi_probe_channel(struct ncsi_dev_priv *ndp) +{ + struct ncsi_dev *nd = &ndp->ndev; + struct ncsi_package *np; + struct ncsi_channel *nc; + struct ncsi_cmd_arg nca; + unsigned char index; + int ret; + + nca.ndp = ndp; + nca.driven = true; + switch (nd->state) { + case ncsi_dev_state_probe: + nd->state = ncsi_dev_state_probe_deselect; + /* Fall through */ + case ncsi_dev_state_probe_deselect: + ndp->pending_req_num = 8; + + /* Deselect all possible packages */ + nca.type = NCSI_PKT_CMD_DP; + nca.channel = 0x1f; + for (index = 0; index < 8; index++) { + nca.package = index; + ret = ncsi_xmit_cmd(&nca); + if (ret) + goto error; + } + + nd->state = ncsi_dev_state_probe_package; + break; + case ncsi_dev_state_probe_package: + ndp->pending_req_num = 16; + + /* Select all possible packages */ + nca.type = NCSI_PKT_CMD_SP; + nca.bytes[0] = 1; + nca.channel = 0x1f; + for (index = 0; index < 8; index++) { + nca.package = index; + ret = ncsi_xmit_cmd(&nca); + if (ret) + goto error; + } + + /* Disable all possible packages */ + nca.type = NCSI_PKT_CMD_DP; + for (index = 0; index < 8; index++) { + nca.package = index; + ret = ncsi_xmit_cmd(&nca); + if (ret) + goto error; + } + + nd->state = ncsi_dev_state_probe_channel; + break; + case ncsi_dev_state_probe_channel: + if (!ndp->active_package) + ndp->active_package = list_first_or_null_rcu( + &ndp->packages, struct ncsi_package, node); + else if (list_is_last(&ndp->active_package->node, + &ndp->packages)) + ndp->active_package = NULL; + else + ndp->active_package = list_next_entry( + ndp->active_package, node); + + /* All available packages and channels are enumerated. The + * enumeration happens for once when the NCSI interface is + * started. So we need continue to start the interface after + * the enumeration. + * + * We have to choose an active channel before configuring it. + * Note that we possibly don't have active channel in extreme + * situation. + */ + if (!ndp->active_package) { + ndp->flags |= NCSI_DEV_PROBED; + if (ncsi_check_hwa(ndp)) + ncsi_enable_hwa(ndp); + else + ncsi_choose_active_channel(ndp); + return; + } + + /* Select the active package */ + ndp->pending_req_num = 1; + nca.type = NCSI_PKT_CMD_SP; + nca.bytes[0] = 1; + nca.package = ndp->active_package->id; + nca.channel = 0x1f; + ret = ncsi_xmit_cmd(&nca); + if (ret) + goto error; + + nd->state = ncsi_dev_state_probe_cis; + break; + case ncsi_dev_state_probe_cis: + ndp->pending_req_num = 32; + + /* Clear initial state */ + nca.type = NCSI_PKT_CMD_CIS; + nca.package = ndp->active_package->id; + for (index = 0; index < 0x20; index++) { + nca.channel = index; + ret = ncsi_xmit_cmd(&nca); + if (ret) + goto error; + } + + nd->state = ncsi_dev_state_probe_gvi; + break; + case ncsi_dev_state_probe_gvi: + case ncsi_dev_state_probe_gc: + case ncsi_dev_state_probe_gls: + np = ndp->active_package; + ndp->pending_req_num = np->channel_num; + + /* Retrieve version, capability or link status */ + if (nd->state == ncsi_dev_state_probe_gvi) + nca.type = NCSI_PKT_CMD_GVI; + else if (nd->state == ncsi_dev_state_probe_gc) + nca.type = NCSI_PKT_CMD_GC; + else + nca.type = NCSI_PKT_CMD_GLS; + + nca.package = np->id; + NCSI_FOR_EACH_CHANNEL(np, nc) { + nca.channel = nc->id; + ret = ncsi_xmit_cmd(&nca); + if (ret) + goto error; + } + + if (nd->state == ncsi_dev_state_probe_gvi) + nd->state = ncsi_dev_state_probe_gc; + else if (nd->state == ncsi_dev_state_probe_gc) + nd->state = ncsi_dev_state_probe_gls; + else + nd->state = ncsi_dev_state_probe_dp; + break; + case ncsi_dev_state_probe_dp: + ndp->pending_req_num = 1; + + /* Deselect the active package */ + nca.type = NCSI_PKT_CMD_DP; + nca.package = ndp->active_package->id; + nca.channel = 0x1f; + ret = ncsi_xmit_cmd(&nca); + if (ret) + goto error; + + /* Scan channels in next package */ + nd->state = ncsi_dev_state_probe_channel; + break; + default: + netdev_warn(nd->dev, "Wrong NCSI state 0x%0x in enumeration\n", + nd->state); + } + + return; +error: + ncsi_report_link(ndp, true); +} + +static void ncsi_dev_work(struct work_struct *work) +{ + struct ncsi_dev_priv *ndp = container_of(work, + struct ncsi_dev_priv, work); + struct ncsi_dev *nd = &ndp->ndev; + + switch (nd->state & ncsi_dev_state_major) { + case ncsi_dev_state_probe: + ncsi_probe_channel(ndp); + break; + case ncsi_dev_state_suspend: + ncsi_suspend_channel(ndp); + break; + case ncsi_dev_state_config: + ncsi_configure_channel(ndp); + break; + default: + netdev_warn(nd->dev, "Wrong NCSI state 0x%x in workqueue\n", + nd->state); + } +} + +int ncsi_process_next_channel(struct ncsi_dev_priv *ndp) +{ + struct ncsi_channel *nc; + int old_state; + unsigned long flags; + + spin_lock_irqsave(&ndp->lock, flags); + nc = list_first_or_null_rcu(&ndp->channel_queue, + struct ncsi_channel, link); + if (nc) { + old_state = xchg(&nc->state, NCSI_CHANNEL_INVISIBLE); + list_del_init(&nc->link); + } + spin_unlock_irqrestore(&ndp->lock, flags); + + ndp->active_channel = nc; + ndp->active_package = nc ? nc->package : NULL; + if (!nc) { + if (ndp->flags & NCSI_DEV_RESHUFFLE) { + ndp->flags &= ~NCSI_DEV_RESHUFFLE; + return ncsi_choose_active_channel(ndp); + } + + ncsi_report_link(ndp, false); + return -ENODEV; + } + + switch (old_state) { + case NCSI_CHANNEL_INACTIVE: + ndp->ndev.state = ncsi_dev_state_config; + ncsi_configure_channel(ndp); + break; + case NCSI_CHANNEL_ACTIVE: + ndp->ndev.state = ncsi_dev_state_suspend; + ncsi_suspend_channel(ndp); + break; + default: + netdev_err(ndp->ndev.dev, "Invalid state 0x%x on %d:%d\n", + nc->state, nc->package->id, nc->id); + ncsi_report_link(ndp, false); + return -EINVAL; + } + + return 0; +} + +#if IS_ENABLED(CONFIG_IPV6) +static int ncsi_inet6addr_event(struct notifier_block *this, + unsigned long event, void *data) +{ + struct inet6_ifaddr *ifa = data; + struct net_device *dev = ifa->idev->dev; + struct ncsi_dev *nd = ncsi_find_dev(dev); + struct ncsi_dev_priv *ndp = nd ? TO_NCSI_DEV_PRIV(nd) : NULL; + struct ncsi_package *np; + struct ncsi_channel *nc; + struct ncsi_cmd_arg nca; + bool action; + int ret; + + if (!ndp || (ipv6_addr_type(&ifa->addr) & + (IPV6_ADDR_LINKLOCAL | IPV6_ADDR_LOOPBACK))) + return NOTIFY_OK; + + switch (event) { + case NETDEV_UP: + action = (++ndp->inet6_addr_num) == 1; + nca.type = NCSI_PKT_CMD_EGMF; + break; + case NETDEV_DOWN: + action = (--ndp->inet6_addr_num == 0); + nca.type = NCSI_PKT_CMD_DGMF; + break; + default: + return NOTIFY_OK; + } + + /* We might not have active channel or packages. The IPv6 + * required multicast will be enabled when active channel + * or packages are chosen. + */ + np = ndp->active_package; + nc = ndp->active_channel; + if (!action || !np || !nc) + return NOTIFY_OK; + + /* We needn't enable or disable it if the function isn't supported */ + if (!(nc->caps[NCSI_CAP_GENERIC].cap & NCSI_CAP_GENERIC_MC)) + return NOTIFY_OK; + + nca.ndp = ndp; + nca.driven = false; + nca.package = np->id; + nca.channel = nc->id; + nca.dwords[0] = nc->caps[NCSI_CAP_MC].cap; + ret = ncsi_xmit_cmd(&nca); + if (ret) { + netdev_warn(dev, "Fail to %s global multicast filter (%d)\n", + (event == NETDEV_UP) ? "enable" : "disable", ret); + return NOTIFY_DONE; + } + + return NOTIFY_OK; +} + +static struct notifier_block ncsi_inet6addr_notifier = { + .notifier_call = ncsi_inet6addr_event, +}; +#endif /* CONFIG_IPV6 */ + struct ncsi_dev *ncsi_register_dev(struct net_device *dev, void (*handler)(struct ncsi_dev *ndev)) { @@ -397,6 +1105,9 @@ struct ncsi_dev *ncsi_register_dev(struct net_device *dev, nd->state = ncsi_dev_state_registered; nd->dev = dev; nd->handler = handler; + ndp->pending_req_num = 0; + INIT_LIST_HEAD(&ndp->channel_queue); + INIT_WORK(&ndp->work, ncsi_dev_work); /* Initialize private NCSI device */ spin_lock_init(&ndp->lock); @@ -411,24 +1122,76 @@ struct ncsi_dev *ncsi_register_dev(struct net_device *dev, } spin_lock_irqsave(&ncsi_dev_lock, flags); +#if IS_ENABLED(CONFIG_IPV6) + ndp->inet6_addr_num = 0; + if (list_empty(&ncsi_dev_list)) + register_inet6addr_notifier(&ncsi_inet6addr_notifier); +#endif list_add_tail_rcu(&ndp->node, &ncsi_dev_list); spin_unlock_irqrestore(&ncsi_dev_lock, flags); + /* Register NCSI packet Rx handler */ + ndp->ptype.type = cpu_to_be16(ETH_P_NCSI); + ndp->ptype.func = ncsi_rcv_rsp; + ndp->ptype.dev = dev; + dev_add_pack(&ndp->ptype); + return nd; } EXPORT_SYMBOL_GPL(ncsi_register_dev); +int ncsi_start_dev(struct ncsi_dev *nd) +{ + struct ncsi_dev_priv *ndp = TO_NCSI_DEV_PRIV(nd); + struct ncsi_package *np; + struct ncsi_channel *nc; + int old_state, ret; + + if (nd->state != ncsi_dev_state_registered && + nd->state != ncsi_dev_state_functional) + return -ENOTTY; + + if (!(ndp->flags & NCSI_DEV_PROBED)) { + nd->state = ncsi_dev_state_probe; + schedule_work(&ndp->work); + return 0; + } + + /* Reset channel's state and start over */ + NCSI_FOR_EACH_PACKAGE(ndp, np) { + NCSI_FOR_EACH_CHANNEL(np, nc) { + old_state = xchg(&nc->state, NCSI_CHANNEL_INACTIVE); + WARN_ON_ONCE(!list_empty(&nc->link) || + old_state == NCSI_CHANNEL_INVISIBLE); + } + } + + if (ndp->flags & NCSI_DEV_HWA) + ret = ncsi_enable_hwa(ndp); + else + ret = ncsi_choose_active_channel(ndp); + + return ret; +} +EXPORT_SYMBOL_GPL(ncsi_start_dev); + void ncsi_unregister_dev(struct ncsi_dev *nd) { struct ncsi_dev_priv *ndp = TO_NCSI_DEV_PRIV(nd); struct ncsi_package *np, *tmp; unsigned long flags; + dev_remove_pack(&ndp->ptype); + list_for_each_entry_safe(np, tmp, &ndp->packages, node) ncsi_remove_package(np); spin_lock_irqsave(&ncsi_dev_lock, flags); list_del_rcu(&ndp->node); +#if IS_ENABLED(CONFIG_IPV6) + if (list_empty(&ncsi_dev_list)) + unregister_inet6addr_notifier(&ncsi_inet6addr_notifier); +#endif spin_unlock_irqrestore(&ncsi_dev_lock, flags); kfree(ndp); diff --git a/net/ncsi/ncsi-rsp.c b/net/ncsi/ncsi-rsp.c index 6ec25cb..a21af88 100644 --- a/net/ncsi/ncsi-rsp.c +++ b/net/ncsi/ncsi-rsp.c @@ -69,6 +69,9 @@ static int ncsi_rsp_handler_cis(struct ncsi_request *nr) rsp = (struct ncsi_rsp_pkt *)skb_network_header(nr->rsp); ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, &np, &nc); if (!nc) { + if (ndp->flags & NCSI_DEV_PROBED) + return -ENXIO; + id = NCSI_CHANNEL_INDEX(rsp->rsp.common.channel); nc = ncsi_add_channel(np, id); } @@ -90,6 +93,9 @@ static int ncsi_rsp_handler_sp(struct ncsi_request *nr) ncsi_find_package_and_channel(ndp, rsp->rsp.common.channel, &np, NULL); if (!np) { + if (ndp->flags & NCSI_DEV_PROBED) + return -ENXIO; + id = NCSI_PACKAGE_INDEX(rsp->rsp.common.channel); np = ncsi_add_package(ndp, id); if (!np) @@ -297,6 +303,7 @@ static int ncsi_rsp_handler_gls(struct ncsi_request *nr) struct ncsi_dev_priv *ndp = nr->ndp; struct ncsi_channel *nc; struct ncsi_channel_mode *ncm; + unsigned long flags; /* Find the package and channel */ rsp = (struct ncsi_rsp_gls_pkt *)skb_network_header(nr->rsp); @@ -310,6 +317,14 @@ static int ncsi_rsp_handler_gls(struct ncsi_request *nr) ncm->data[3] = ntohl(rsp->other); ncm->data[4] = ntohl(rsp->oem_status); + if (nr->driven) + return 0; + + /* Reset the channel monitor if it has been enabled */ + spin_lock_irqsave(&nc->lock, flags); + nc->timeout = 0; + spin_unlock_irqrestore(&nc->lock, flags); + return 0; } -- cgit v0.10.2 From 7a82ecf4cfb854955198945340ae13558b64e1af Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:20 +1000 Subject: net/ncsi: NCSI AEN packet handler This introduces NCSI AEN packet handlers that result in (A) the currently active channel is reconfigured; (B) Currently active channel is deconfigured and disabled, another channel is chosen as active one and configured. Case (B) won't happen if hardware arbitration has been enabled, the channel that was in active state is suspended simply. Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller diff --git a/net/ncsi/Makefile b/net/ncsi/Makefile index 4751819..dd12b56 100644 --- a/net/ncsi/Makefile +++ b/net/ncsi/Makefile @@ -1,4 +1,4 @@ # # Makefile for NCSI API # -obj-$(CONFIG_NET_NCSI) += ncsi-cmd.o ncsi-rsp.o ncsi-manage.o +obj-$(CONFIG_NET_NCSI) += ncsi-cmd.o ncsi-rsp.o ncsi-aen.o ncsi-manage.o diff --git a/net/ncsi/internal.h b/net/ncsi/internal.h index 38fc95a..33738c0 100644 --- a/net/ncsi/internal.h +++ b/net/ncsi/internal.h @@ -323,5 +323,6 @@ u32 ncsi_calculate_checksum(unsigned char *data, int len); int ncsi_xmit_cmd(struct ncsi_cmd_arg *nca); int ncsi_rcv_rsp(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev); +int ncsi_aen_handler(struct ncsi_dev_priv *ndp, struct sk_buff *skb); #endif /* __NCSI_INTERNAL_H__ */ diff --git a/net/ncsi/ncsi-aen.c b/net/ncsi/ncsi-aen.c new file mode 100644 index 0000000..d463468 --- /dev/null +++ b/net/ncsi/ncsi-aen.c @@ -0,0 +1,193 @@ +/* + * Copyright Gavin Shan, IBM Corporation 2016. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "internal.h" +#include "ncsi-pkt.h" + +static int ncsi_validate_aen_pkt(struct ncsi_aen_pkt_hdr *h, + const unsigned short payload) +{ + u32 checksum; + __be32 *pchecksum; + + if (h->common.revision != NCSI_PKT_REVISION) + return -EINVAL; + if (ntohs(h->common.length) != payload) + return -EINVAL; + + /* Validate checksum, which might be zeroes if the + * sender doesn't support checksum according to NCSI + * specification. + */ + pchecksum = (__be32 *)((void *)(h + 1) + payload - 4); + if (ntohl(*pchecksum) == 0) + return 0; + + checksum = ncsi_calculate_checksum((unsigned char *)h, + sizeof(*h) + payload - 4); + if (*pchecksum != htonl(checksum)) + return -EINVAL; + + return 0; +} + +static int ncsi_aen_handler_lsc(struct ncsi_dev_priv *ndp, + struct ncsi_aen_pkt_hdr *h) +{ + struct ncsi_aen_lsc_pkt *lsc; + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + unsigned long old_data; + unsigned long flags; + + /* Find the NCSI channel */ + ncsi_find_package_and_channel(ndp, h->common.channel, NULL, &nc); + if (!nc) + return -ENODEV; + + /* Update the link status */ + ncm = &nc->modes[NCSI_MODE_LINK]; + lsc = (struct ncsi_aen_lsc_pkt *)h; + old_data = ncm->data[2]; + ncm->data[2] = ntohl(lsc->status); + ncm->data[4] = ntohl(lsc->oem_status); + if (!((old_data ^ ncm->data[2]) & 0x1) || + !list_empty(&nc->link)) + return 0; + if (!(nc->state == NCSI_CHANNEL_INACTIVE && (ncm->data[2] & 0x1)) && + !(nc->state == NCSI_CHANNEL_ACTIVE && !(ncm->data[2] & 0x1))) + return 0; + + if (!(ndp->flags & NCSI_DEV_HWA) && + nc->state == NCSI_CHANNEL_ACTIVE) + ndp->flags |= NCSI_DEV_RESHUFFLE; + + ncsi_stop_channel_monitor(nc); + spin_lock_irqsave(&ndp->lock, flags); + list_add_tail_rcu(&nc->link, &ndp->channel_queue); + spin_unlock_irqrestore(&ndp->lock, flags); + + return ncsi_process_next_channel(ndp); +} + +static int ncsi_aen_handler_cr(struct ncsi_dev_priv *ndp, + struct ncsi_aen_pkt_hdr *h) +{ + struct ncsi_channel *nc; + unsigned long flags; + + /* Find the NCSI channel */ + ncsi_find_package_and_channel(ndp, h->common.channel, NULL, &nc); + if (!nc) + return -ENODEV; + + if (!list_empty(&nc->link) || + nc->state != NCSI_CHANNEL_ACTIVE) + return 0; + + ncsi_stop_channel_monitor(nc); + spin_lock_irqsave(&ndp->lock, flags); + xchg(&nc->state, NCSI_CHANNEL_INACTIVE); + list_add_tail_rcu(&nc->link, &ndp->channel_queue); + spin_unlock_irqrestore(&ndp->lock, flags); + + return ncsi_process_next_channel(ndp); +} + +static int ncsi_aen_handler_hncdsc(struct ncsi_dev_priv *ndp, + struct ncsi_aen_pkt_hdr *h) +{ + struct ncsi_channel *nc; + struct ncsi_channel_mode *ncm; + struct ncsi_aen_hncdsc_pkt *hncdsc; + unsigned long flags; + + /* Find the NCSI channel */ + ncsi_find_package_and_channel(ndp, h->common.channel, NULL, &nc); + if (!nc) + return -ENODEV; + + /* If the channel is active one, we need reconfigure it */ + ncm = &nc->modes[NCSI_MODE_LINK]; + hncdsc = (struct ncsi_aen_hncdsc_pkt *)h; + ncm->data[3] = ntohl(hncdsc->status); + if (!list_empty(&nc->link) || + nc->state != NCSI_CHANNEL_ACTIVE || + (ncm->data[3] & 0x1)) + return 0; + + if (ndp->flags & NCSI_DEV_HWA) + ndp->flags |= NCSI_DEV_RESHUFFLE; + + /* If this channel is the active one and the link doesn't + * work, we have to choose another channel to be active one. + * The logic here is exactly similar to what we do when link + * is down on the active channel. + */ + ncsi_stop_channel_monitor(nc); + spin_lock_irqsave(&ndp->lock, flags); + list_add_tail_rcu(&nc->link, &ndp->channel_queue); + spin_unlock_irqrestore(&ndp->lock, flags); + + ncsi_process_next_channel(ndp); + + return 0; +} + +static struct ncsi_aen_handler { + unsigned char type; + int payload; + int (*handler)(struct ncsi_dev_priv *ndp, + struct ncsi_aen_pkt_hdr *h); +} ncsi_aen_handlers[] = { + { NCSI_PKT_AEN_LSC, 12, ncsi_aen_handler_lsc }, + { NCSI_PKT_AEN_CR, 4, ncsi_aen_handler_cr }, + { NCSI_PKT_AEN_HNCDSC, 4, ncsi_aen_handler_hncdsc } +}; + +int ncsi_aen_handler(struct ncsi_dev_priv *ndp, struct sk_buff *skb) +{ + struct ncsi_aen_pkt_hdr *h; + struct ncsi_aen_handler *nah = NULL; + int i, ret; + + /* Find the handler */ + h = (struct ncsi_aen_pkt_hdr *)skb_network_header(skb); + for (i = 0; i < ARRAY_SIZE(ncsi_aen_handlers); i++) { + if (ncsi_aen_handlers[i].type == h->type) { + nah = &ncsi_aen_handlers[i]; + break; + } + } + + if (!nah) { + netdev_warn(ndp->ndev.dev, "Invalid AEN (0x%x) received\n", + h->type); + return -ENOENT; + } + + ret = ncsi_validate_aen_pkt(h, nah->payload); + if (ret) + goto out; + + ret = nah->handler(ndp, h); +out: + consume_skb(skb); + return ret; +} diff --git a/net/ncsi/ncsi-pkt.h b/net/ncsi/ncsi-pkt.h index 4bdefd9..3ea49ed 100644 --- a/net/ncsi/ncsi-pkt.h +++ b/net/ncsi/ncsi-pkt.h @@ -31,6 +31,12 @@ struct ncsi_rsp_pkt_hdr { __be16 reason; /* Response reason */ }; +struct ncsi_aen_pkt_hdr { + struct ncsi_pkt_hdr common; /* Common NCSI packet header */ + unsigned char reserved2[3]; /* Reserved */ + unsigned char type; /* AEN packet type */ +}; + /* NCSI common command packet */ struct ncsi_cmd_pkt { struct ncsi_cmd_pkt_hdr cmd; /* Command header */ @@ -296,6 +302,30 @@ struct ncsi_rsp_gpuuid_pkt { __be32 checksum; }; +/* AEN: Link State Change */ +struct ncsi_aen_lsc_pkt { + struct ncsi_aen_pkt_hdr aen; /* AEN header */ + __be32 status; /* Link status */ + __be32 oem_status; /* OEM link status */ + __be32 checksum; /* Checksum */ + unsigned char pad[14]; +}; + +/* AEN: Configuration Required */ +struct ncsi_aen_cr_pkt { + struct ncsi_aen_pkt_hdr aen; /* AEN header */ + __be32 checksum; /* Checksum */ + unsigned char pad[22]; +}; + +/* AEN: Host Network Controller Driver Status Change */ +struct ncsi_aen_hncdsc_pkt { + struct ncsi_aen_pkt_hdr aen; /* AEN header */ + __be32 status; /* Status */ + __be32 checksum; /* Checksum */ + unsigned char pad[18]; +}; + /* NCSI packet revision */ #define NCSI_PKT_REVISION 0x01 @@ -376,4 +406,10 @@ struct ncsi_rsp_gpuuid_pkt { #define NCSI_PKT_RSP_R_LENGTH 0x0005 /* Invalid payload length */ #define NCSI_PKT_RSP_R_UNKNOWN 0x7fff /* Command type unsupported */ +/* NCSI AEN packet type */ +#define NCSI_PKT_AEN 0xFF /* AEN Packet */ +#define NCSI_PKT_AEN_LSC 0x00 /* Link status change */ +#define NCSI_PKT_AEN_CR 0x01 /* Configuration required */ +#define NCSI_PKT_AEN_HNCDSC 0x02 /* HNC driver status change */ + #endif /* __NCSI_PKT_H__ */ diff --git a/net/ncsi/ncsi-rsp.c b/net/ncsi/ncsi-rsp.c index a21af88..af84389 100644 --- a/net/ncsi/ncsi-rsp.c +++ b/net/ncsi/ncsi-rsp.c @@ -980,8 +980,12 @@ int ncsi_rcv_rsp(struct sk_buff *skb, struct net_device *dev, if (!ndp) return -ENODEV; - /* Find the handler */ + /* Check if it is AEN packet */ hdr = (struct ncsi_pkt_hdr *)skb_network_header(skb); + if (hdr->type == NCSI_PKT_AEN) + return ncsi_aen_handler(ndp, skb); + + /* Find the handler */ for (i = 0; i < ARRAY_SIZE(ncsi_rsp_handlers); i++) { if (ncsi_rsp_handlers[i].type == hdr->type) { if (ncsi_rsp_handlers[i].handler) -- cgit v0.10.2 From eb4181849f58f31d8d68762b6d87c6f06b86dbbd Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:21 +1000 Subject: net/faraday: Helper functions to create or destroy MDIO interface This introduces two helper functions to create or destroy MDIO interface. No logical changes introduced except the proper MDIO names are given when having more than one MDIO bus. Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index e7cf313..9b09493 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -1146,6 +1146,60 @@ static const struct net_device_ops ftgmac100_netdev_ops = { .ndo_do_ioctl = ftgmac100_do_ioctl, }; +static int ftgmac100_setup_mdio(struct net_device *netdev) +{ + struct ftgmac100 *priv = netdev_priv(netdev); + struct platform_device *pdev = to_platform_device(priv->dev); + int i, err = 0; + + /* initialize mdio bus */ + priv->mii_bus = mdiobus_alloc(); + if (!priv->mii_bus) + return -EIO; + + priv->mii_bus->name = "ftgmac100_mdio"; + snprintf(priv->mii_bus->id, MII_BUS_ID_SIZE, "%s-%d", + pdev->name, pdev->id); + priv->mii_bus->priv = priv->netdev; + priv->mii_bus->read = ftgmac100_mdiobus_read; + priv->mii_bus->write = ftgmac100_mdiobus_write; + + for (i = 0; i < PHY_MAX_ADDR; i++) + priv->mii_bus->irq[i] = PHY_POLL; + + err = mdiobus_register(priv->mii_bus); + if (err) { + dev_err(priv->dev, "Cannot register MDIO bus!\n"); + goto err_register_mdiobus; + } + + err = ftgmac100_mii_probe(priv); + if (err) { + dev_err(priv->dev, "MII Probe failed!\n"); + goto err_mii_probe; + } + + return 0; + +err_mii_probe: + mdiobus_unregister(priv->mii_bus); +err_register_mdiobus: + mdiobus_free(priv->mii_bus); + return err; +} + +static void ftgmac100_destroy_mdio(struct net_device *netdev) +{ + struct ftgmac100 *priv = netdev_priv(netdev); + + if (!netdev->phydev) + return; + + phy_disconnect(netdev->phydev); + mdiobus_unregister(priv->mii_bus); + mdiobus_free(priv->mii_bus); +} + /****************************************************************************** * struct platform_driver functions *****************************************************************************/ @@ -1211,31 +1265,9 @@ static int ftgmac100_probe(struct platform_device *pdev) priv->irq = irq; - /* initialize mdio bus */ - priv->mii_bus = mdiobus_alloc(); - if (!priv->mii_bus) { - err = -EIO; - goto err_alloc_mdiobus; - } - - priv->mii_bus->name = "ftgmac100_mdio"; - snprintf(priv->mii_bus->id, MII_BUS_ID_SIZE, "ftgmac100_mii"); - - priv->mii_bus->priv = netdev; - priv->mii_bus->read = ftgmac100_mdiobus_read; - priv->mii_bus->write = ftgmac100_mdiobus_write; - - err = mdiobus_register(priv->mii_bus); - if (err) { - dev_err(&pdev->dev, "Cannot register MDIO bus!\n"); - goto err_register_mdiobus; - } - - err = ftgmac100_mii_probe(priv); - if (err) { - dev_err(&pdev->dev, "MII Probe failed!\n"); - goto err_mii_probe; - } + err = ftgmac100_setup_mdio(netdev); + if (err) + goto err_setup_mdio; /* register network device */ err = register_netdev(netdev); @@ -1255,12 +1287,8 @@ static int ftgmac100_probe(struct platform_device *pdev) return 0; err_register_netdev: - phy_disconnect(netdev->phydev); -err_mii_probe: - mdiobus_unregister(priv->mii_bus); -err_register_mdiobus: - mdiobus_free(priv->mii_bus); -err_alloc_mdiobus: + ftgmac100_destroy_mdio(netdev); +err_setup_mdio: iounmap(priv->base); err_ioremap: release_resource(priv->res); @@ -1280,10 +1308,7 @@ static int __exit ftgmac100_remove(struct platform_device *pdev) priv = netdev_priv(netdev); unregister_netdev(netdev); - - phy_disconnect(netdev->phydev); - mdiobus_unregister(priv->mii_bus); - mdiobus_free(priv->mii_bus); + ftgmac100_destroy_mdio(netdev); iounmap(priv->base); release_resource(priv->res); -- cgit v0.10.2 From 113ce107afe979902c003900bfaed7878d8a5968 Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:22 +1000 Subject: net/faraday: Read MAC address from chip The device is assigned with random MAC address. It isn't reasonable. An valid MAC address might have been provided by (uboot) firmware by device-tree or in chip. It's reasonable to use it to maintain consistency. This uses the MAC address from device-tree or that in the chip if it's valid. Otherwise, a random MAC address is given as before. Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index 9b09493..2c3f656 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -141,6 +141,64 @@ static void ftgmac100_set_mac(struct ftgmac100 *priv, const unsigned char *mac) iowrite32(laddr, priv->base + FTGMAC100_OFFSET_MAC_LADR); } +static void ftgmac100_setup_mac(struct ftgmac100 *priv) +{ + u8 mac[ETH_ALEN]; + unsigned int m; + unsigned int l; + void *addr; + + addr = device_get_mac_address(priv->dev, mac, ETH_ALEN); + if (addr) { + ether_addr_copy(priv->netdev->dev_addr, mac); + dev_info(priv->dev, "Read MAC address %pM from device tree\n", + mac); + return; + } + + m = ioread32(priv->base + FTGMAC100_OFFSET_MAC_MADR); + l = ioread32(priv->base + FTGMAC100_OFFSET_MAC_LADR); + + mac[0] = (m >> 8) & 0xff; + mac[1] = m & 0xff; + mac[2] = (l >> 24) & 0xff; + mac[3] = (l >> 16) & 0xff; + mac[4] = (l >> 8) & 0xff; + mac[5] = l & 0xff; + + if (!is_valid_ether_addr(mac)) { + mac[5] = (m >> 8) & 0xff; + mac[4] = m & 0xff; + mac[3] = (l >> 24) & 0xff; + mac[2] = (l >> 16) & 0xff; + mac[1] = (l >> 8) & 0xff; + mac[0] = l & 0xff; + } + + if (is_valid_ether_addr(mac)) { + ether_addr_copy(priv->netdev->dev_addr, mac); + dev_info(priv->dev, "Read MAC address %pM from chip\n", mac); + } else { + eth_hw_addr_random(priv->netdev); + dev_info(priv->dev, "Generated random MAC address %pM\n", + priv->netdev->dev_addr); + } +} + +static int ftgmac100_set_mac_addr(struct net_device *dev, void *p) +{ + int ret; + + ret = eth_prepare_mac_addr_change(dev, p); + if (ret < 0) + return ret; + + eth_commit_mac_addr_change(dev, p); + ftgmac100_set_mac(netdev_priv(dev), dev->dev_addr); + + return 0; +} + static void ftgmac100_init_hw(struct ftgmac100 *priv) { /* setup ring buffer base registers */ @@ -1141,7 +1199,7 @@ static const struct net_device_ops ftgmac100_netdev_ops = { .ndo_open = ftgmac100_open, .ndo_stop = ftgmac100_stop, .ndo_start_xmit = ftgmac100_hard_start_xmit, - .ndo_set_mac_address = eth_mac_addr, + .ndo_set_mac_address = ftgmac100_set_mac_addr, .ndo_validate_addr = eth_validate_addr, .ndo_do_ioctl = ftgmac100_do_ioctl, }; @@ -1265,6 +1323,9 @@ static int ftgmac100_probe(struct platform_device *pdev) priv->irq = irq; + /* MAC address from chip or random one */ + ftgmac100_setup_mac(priv); + err = ftgmac100_setup_mdio(netdev); if (err) goto err_setup_mdio; @@ -1278,12 +1339,6 @@ static int ftgmac100_probe(struct platform_device *pdev) netdev_info(netdev, "irq %d, mapped at %p\n", priv->irq, priv->base); - if (!is_valid_ether_addr(netdev->dev_addr)) { - eth_hw_addr_random(netdev); - netdev_info(netdev, "generated random MAC address %pM\n", - netdev->dev_addr); - } - return 0; err_register_netdev: -- cgit v0.10.2 From bd466c3fb5a4ff862f805213d7821d8c6f92c382 Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:23 +1000 Subject: net/faraday: Support NCSI mode This makes ftgmac100 driver support NCSI mode. The NCSI is enabled on the interface if property "use-nc-si" or "use-ncsi" is found from the device node in device tree. * No PHY device is used when NCSI mode is enabled. * The NCSI device (struct ncsi_dev) is created when probing the device while it's enabled/started when the interface is brought up. * Hardware IP checksum dosn't work when NCSI mode is enabled. It is disabled on enabled NCSI. Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index 2c3f656..1cd4975 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -31,6 +31,7 @@ #include #include #include +#include #include "ftgmac100.h" @@ -68,10 +69,13 @@ struct ftgmac100 { struct net_device *netdev; struct device *dev; + struct ncsi_dev *ndev; struct napi_struct napi; struct mii_bus *mii_bus; int old_speed; + bool use_ncsi; + bool enabled; }; static int ftgmac100_alloc_rx_page(struct ftgmac100 *priv, @@ -1010,7 +1014,10 @@ static irqreturn_t ftgmac100_interrupt(int irq, void *dev_id) struct net_device *netdev = dev_id; struct ftgmac100 *priv = netdev_priv(netdev); - if (likely(netif_running(netdev))) { + /* When running in NCSI mode, the interface should be ready for + * receiving or transmitting NCSI packets before it's opened. + */ + if (likely(priv->use_ncsi || netif_running(netdev))) { /* Disable interrupts for polling */ iowrite32(0, priv->base + FTGMAC100_OFFSET_IER); napi_schedule(&priv->napi); @@ -1123,17 +1130,33 @@ static int ftgmac100_open(struct net_device *netdev) goto err_hw; ftgmac100_init_hw(priv); - ftgmac100_start_hw(priv, 10); - - phy_start(netdev->phydev); + ftgmac100_start_hw(priv, priv->use_ncsi ? 100 : 10); + if (netdev->phydev) + phy_start(netdev->phydev); + else if (priv->use_ncsi) + netif_carrier_on(netdev); napi_enable(&priv->napi); netif_start_queue(netdev); /* enable all interrupts */ iowrite32(INT_MASK_ALL_ENABLED, priv->base + FTGMAC100_OFFSET_IER); + + /* Start the NCSI device */ + if (priv->use_ncsi) { + err = ncsi_start_dev(priv->ndev); + if (err) + goto err_ncsi; + } + + priv->enabled = true; + return 0; +err_ncsi: + napi_disable(&priv->napi); + netif_stop_queue(netdev); + iowrite32(0, priv->base + FTGMAC100_OFFSET_IER); err_hw: free_irq(priv->irq, netdev); err_irq: @@ -1146,12 +1169,17 @@ static int ftgmac100_stop(struct net_device *netdev) { struct ftgmac100 *priv = netdev_priv(netdev); + if (!priv->enabled) + return 0; + /* disable all interrupts */ + priv->enabled = false; iowrite32(0, priv->base + FTGMAC100_OFFSET_IER); netif_stop_queue(netdev); napi_disable(&priv->napi); - phy_stop(netdev->phydev); + if (netdev->phydev) + phy_stop(netdev->phydev); ftgmac100_stop_hw(priv); free_irq(priv->irq, netdev); @@ -1192,6 +1220,9 @@ static int ftgmac100_hard_start_xmit(struct sk_buff *skb, /* optional */ static int ftgmac100_do_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) { + if (!netdev->phydev) + return -ENXIO; + return phy_mii_ioctl(netdev->phydev, ifr, cmd); } @@ -1258,6 +1289,15 @@ static void ftgmac100_destroy_mdio(struct net_device *netdev) mdiobus_free(priv->mii_bus); } +static void ftgmac100_ncsi_handler(struct ncsi_dev *nd) +{ + if (unlikely(nd->state != ncsi_dev_state_functional)) + return; + + netdev_info(nd->dev, "NCSI interface %s\n", + nd->link_up ? "up" : "down"); +} + /****************************************************************************** * struct platform_driver functions *****************************************************************************/ @@ -1267,7 +1307,7 @@ static int ftgmac100_probe(struct platform_device *pdev) int irq; struct net_device *netdev; struct ftgmac100 *priv; - int err; + int err = 0; if (!pdev) return -ENODEV; @@ -1291,7 +1331,6 @@ static int ftgmac100_probe(struct platform_device *pdev) netdev->ethtool_ops = &ftgmac100_ethtool_ops; netdev->netdev_ops = &ftgmac100_netdev_ops; - netdev->features = NETIF_F_IP_CSUM | NETIF_F_GRO; platform_set_drvdata(pdev, netdev); @@ -1326,9 +1365,34 @@ static int ftgmac100_probe(struct platform_device *pdev) /* MAC address from chip or random one */ ftgmac100_setup_mac(priv); - err = ftgmac100_setup_mdio(netdev); - if (err) - goto err_setup_mdio; + if (pdev->dev.of_node && + of_get_property(pdev->dev.of_node, "use-ncsi", NULL)) { + if (!IS_ENABLED(CONFIG_NET_NCSI)) { + dev_err(&pdev->dev, "NCSI stack not enabled\n"); + goto err_ncsi_dev; + } + + dev_info(&pdev->dev, "Using NCSI interface\n"); + priv->use_ncsi = true; + priv->ndev = ncsi_register_dev(netdev, ftgmac100_ncsi_handler); + if (!priv->ndev) + goto err_ncsi_dev; + } else { + priv->use_ncsi = false; + err = ftgmac100_setup_mdio(netdev); + if (err) + goto err_setup_mdio; + } + + /* We have to disable on-chip IP checksum functionality + * when NCSI is enabled on the interface. It doesn't work + * in that case. + */ + netdev->features = NETIF_F_IP_CSUM | NETIF_F_GRO; + if (priv->use_ncsi && + of_get_property(pdev->dev.of_node, "no-hw-checksum", NULL)) + netdev->features &= ~NETIF_F_IP_CSUM; + /* register network device */ err = register_netdev(netdev); @@ -1341,6 +1405,7 @@ static int ftgmac100_probe(struct platform_device *pdev) return 0; +err_ncsi_dev: err_register_netdev: ftgmac100_destroy_mdio(netdev); err_setup_mdio: -- cgit v0.10.2 From bb168e2e9e512e6b2cc3ebf6f2ca3fcb07180370 Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:24 +1000 Subject: net/faraday: Match driver according to compatible property This matches the driver with devices compatible with "faraday,ftgmac100" declared in the device tree. Originally, device's name from device tree for it. Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index 1cd4975..d8afa2d 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -1438,14 +1438,20 @@ static int __exit ftgmac100_remove(struct platform_device *pdev) return 0; } +static const struct of_device_id ftgmac100_of_match[] = { + { .compatible = "faraday,ftgmac100" }, + { } +}; +MODULE_DEVICE_TABLE(of, ftgmac100_of_match); + static struct platform_driver ftgmac100_driver = { - .probe = ftgmac100_probe, - .remove = __exit_p(ftgmac100_remove), - .driver = { - .name = DRV_NAME, + .probe = ftgmac100_probe, + .remove = __exit_p(ftgmac100_remove), + .driver = { + .name = DRV_NAME, + .of_match_table = ftgmac100_of_match, }, }; - module_platform_driver(ftgmac100_driver); MODULE_AUTHOR("Po-Yu Chuang "); -- cgit v0.10.2 From fc6061cf93524c3e1066185922ae3ac3f41b9746 Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Tue, 19 Jul 2016 11:54:25 +1000 Subject: net/faraday: Mask PHY interrupt with NCSI mode Bogus PHY interrupts are observed. This masks the PHY interrupt when the interface works in NCSI mode as there is no attached PHY under the circumstance. Signed-off-by: Gavin Shan Acked-by: Joel Stanley Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index d8afa2d..2d4c7ea 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -74,6 +74,7 @@ struct ftgmac100 { struct mii_bus *mii_bus; int old_speed; + int int_mask_all; bool use_ncsi; bool enabled; }; @@ -84,14 +85,6 @@ static int ftgmac100_alloc_rx_page(struct ftgmac100 *priv, /****************************************************************************** * internal functions (hardware register access) *****************************************************************************/ -#define INT_MASK_ALL_ENABLED (FTGMAC100_INT_RPKT_LOST | \ - FTGMAC100_INT_XPKT_ETH | \ - FTGMAC100_INT_XPKT_LOST | \ - FTGMAC100_INT_AHB_ERR | \ - FTGMAC100_INT_PHYSTS_CHG | \ - FTGMAC100_INT_RPKT_BUF | \ - FTGMAC100_INT_NO_RXBUF) - static void ftgmac100_set_rx_ring_base(struct ftgmac100 *priv, dma_addr_t addr) { iowrite32(addr, priv->base + FTGMAC100_OFFSET_RXR_BADR); @@ -1070,8 +1063,9 @@ static int ftgmac100_poll(struct napi_struct *napi, int budget) ftgmac100_tx_complete(priv); } - if (status & (FTGMAC100_INT_NO_RXBUF | FTGMAC100_INT_RPKT_LOST | - FTGMAC100_INT_AHB_ERR | FTGMAC100_INT_PHYSTS_CHG)) { + if (status & priv->int_mask_all & (FTGMAC100_INT_NO_RXBUF | + FTGMAC100_INT_RPKT_LOST | FTGMAC100_INT_AHB_ERR | + FTGMAC100_INT_PHYSTS_CHG)) { if (net_ratelimit()) netdev_info(netdev, "[ISR] = 0x%x: %s%s%s%s\n", status, status & FTGMAC100_INT_NO_RXBUF ? "NO_RXBUF " : "", @@ -1094,7 +1088,8 @@ static int ftgmac100_poll(struct napi_struct *napi, int budget) napi_complete(napi); /* enable all interrupts */ - iowrite32(INT_MASK_ALL_ENABLED, priv->base + FTGMAC100_OFFSET_IER); + iowrite32(priv->int_mask_all, + priv->base + FTGMAC100_OFFSET_IER); } return rx; @@ -1140,7 +1135,7 @@ static int ftgmac100_open(struct net_device *netdev) netif_start_queue(netdev); /* enable all interrupts */ - iowrite32(INT_MASK_ALL_ENABLED, priv->base + FTGMAC100_OFFSET_IER); + iowrite32(priv->int_mask_all, priv->base + FTGMAC100_OFFSET_IER); /* Start the NCSI device */ if (priv->use_ncsi) { @@ -1365,6 +1360,13 @@ static int ftgmac100_probe(struct platform_device *pdev) /* MAC address from chip or random one */ ftgmac100_setup_mac(priv); + priv->int_mask_all = (FTGMAC100_INT_RPKT_LOST | + FTGMAC100_INT_XPKT_ETH | + FTGMAC100_INT_XPKT_LOST | + FTGMAC100_INT_AHB_ERR | + FTGMAC100_INT_PHYSTS_CHG | + FTGMAC100_INT_RPKT_BUF | + FTGMAC100_INT_NO_RXBUF); if (pdev->dev.of_node && of_get_property(pdev->dev.of_node, "use-ncsi", NULL)) { if (!IS_ENABLED(CONFIG_NET_NCSI)) { @@ -1374,6 +1376,7 @@ static int ftgmac100_probe(struct platform_device *pdev) dev_info(&pdev->dev, "Using NCSI interface\n"); priv->use_ncsi = true; + priv->int_mask_all &= ~FTGMAC100_INT_PHYSTS_CHG; priv->ndev = ncsi_register_dev(netdev, ftgmac100_ncsi_handler); if (!priv->ndev) goto err_ncsi_dev; -- cgit v0.10.2 From 59d3656d5bf504f771fc44fdbc7a9a8590795f22 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:46 -0700 Subject: bpf: add bpf_prog_add api for bulk prog refcnt A subsystem may need to store many copies of a bpf program, each deserving its own reference. Rather than requiring the caller to loop one by one (with possible mid-loop failure), add a bulk bpf_prog_add api. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/bpf.h b/include/linux/bpf.h index c13e92b..75a5ae6 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -224,6 +224,7 @@ void bpf_register_map_type(struct bpf_map_type_list *tl); struct bpf_prog *bpf_prog_get(u32 ufd); struct bpf_prog *bpf_prog_get_type(u32 ufd, enum bpf_prog_type type); +struct bpf_prog *bpf_prog_add(struct bpf_prog *prog, int i); struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog); void bpf_prog_put(struct bpf_prog *prog); diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 96d938a..228f962 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -670,14 +670,20 @@ static struct bpf_prog *____bpf_prog_get(struct fd f) return f.file->private_data; } -struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog) +struct bpf_prog *bpf_prog_add(struct bpf_prog *prog, int i) { - if (atomic_inc_return(&prog->aux->refcnt) > BPF_MAX_REFCNT) { - atomic_dec(&prog->aux->refcnt); + if (atomic_add_return(i, &prog->aux->refcnt) > BPF_MAX_REFCNT) { + atomic_sub(i, &prog->aux->refcnt); return ERR_PTR(-EBUSY); } return prog; } +EXPORT_SYMBOL_GPL(bpf_prog_add); + +struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog) +{ + return bpf_prog_add(prog, 1); +} static struct bpf_prog *__bpf_prog_get(u32 ufd, enum bpf_prog_type *type) { -- cgit v0.10.2 From 6a773a15a1e8874e5eccd2f29190c31085912c95 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:47 -0700 Subject: bpf: add XDP prog type for early driver filter Add a new bpf prog type that is intended to run in early stages of the packet rx path. Only minimal packet metadata will be available, hence a new context type, struct xdp_md, is exposed to userspace. So far only expose the packet start and end pointers, and only in read mode. An XDP program must return one of the well known enum values, all other return codes are reserved for future use. Unfortunately, this restriction is hard to enforce at verification time, so take the approach of warning at runtime when such programs are encountered. Out of bounds return codes should alias to XDP_ABORTED. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/filter.h b/include/linux/filter.h index 6fc31ef..15d816a 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -368,6 +368,11 @@ struct bpf_skb_data_end { void *data_end; }; +struct xdp_buff { + void *data; + void *data_end; +}; + /* compute the linear packet data range [data, data_end) which * will be accessed by cls_bpf and act_bpf programs */ @@ -429,6 +434,18 @@ static inline u32 bpf_prog_run_clear_cb(const struct bpf_prog *prog, return BPF_PROG_RUN(prog, skb); } +static inline u32 bpf_prog_run_xdp(const struct bpf_prog *prog, + struct xdp_buff *xdp) +{ + u32 ret; + + rcu_read_lock(); + ret = BPF_PROG_RUN(prog, (void *)xdp); + rcu_read_unlock(); + + return ret; +} + static inline unsigned int bpf_prog_size(unsigned int proglen) { return max(sizeof(struct bpf_prog), @@ -509,6 +526,7 @@ bool bpf_helper_changes_skb_data(void *func); struct bpf_prog *bpf_patch_insn_single(struct bpf_prog *prog, u32 off, const struct bpf_insn *patch, u32 len); +void bpf_warn_invalid_xdp_action(u32 act); #ifdef CONFIG_BPF_JIT extern int bpf_jit_enable; diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index c4d9224..a517865 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -94,6 +94,7 @@ enum bpf_prog_type { BPF_PROG_TYPE_SCHED_CLS, BPF_PROG_TYPE_SCHED_ACT, BPF_PROG_TYPE_TRACEPOINT, + BPF_PROG_TYPE_XDP, }; #define BPF_PSEUDO_MAP_FD 1 @@ -439,4 +440,23 @@ struct bpf_tunnel_key { __u32 tunnel_label; }; +/* User return codes for XDP prog type. + * A valid XDP program must return one of these defined values. All other + * return codes are reserved for future use. Unknown return codes will result + * in packet drop. + */ +enum xdp_action { + XDP_ABORTED = 0, + XDP_DROP, + XDP_PASS, +}; + +/* user accessible metadata for XDP packet hook + * new fields must be added to the end of this structure + */ +struct xdp_md { + __u32 data; + __u32 data_end; +}; + #endif /* _UAPI__LINUX_BPF_H__ */ diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index e206c21..a8d67d0 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -713,6 +713,7 @@ static int check_ptr_alignment(struct verifier_env *env, struct reg_state *reg, switch (env->prog->type) { case BPF_PROG_TYPE_SCHED_CLS: case BPF_PROG_TYPE_SCHED_ACT: + case BPF_PROG_TYPE_XDP: break; default: verbose("verifier is misconfigured\n"); diff --git a/net/core/filter.c b/net/core/filter.c index 22e3992..6c627bc 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2410,6 +2410,12 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) } } +static const struct bpf_func_proto * +xdp_func_proto(enum bpf_func_id func_id) +{ + return sk_filter_func_proto(func_id); +} + static bool __is_valid_access(int off, int size, enum bpf_access_type type) { if (off < 0 || off >= sizeof(struct __sk_buff)) @@ -2477,6 +2483,44 @@ static bool tc_cls_act_is_valid_access(int off, int size, return __is_valid_access(off, size, type); } +static bool __is_valid_xdp_access(int off, int size, + enum bpf_access_type type) +{ + if (off < 0 || off >= sizeof(struct xdp_md)) + return false; + if (off % size != 0) + return false; + if (size != 4) + return false; + + return true; +} + +static bool xdp_is_valid_access(int off, int size, + enum bpf_access_type type, + enum bpf_reg_type *reg_type) +{ + if (type == BPF_WRITE) + return false; + + switch (off) { + case offsetof(struct xdp_md, data): + *reg_type = PTR_TO_PACKET; + break; + case offsetof(struct xdp_md, data_end): + *reg_type = PTR_TO_PACKET_END; + break; + } + + return __is_valid_xdp_access(off, size, type); +} + +void bpf_warn_invalid_xdp_action(u32 act) +{ + WARN_ONCE(1, "Illegal XDP return value %u, expect packet loss\n", act); +} +EXPORT_SYMBOL_GPL(bpf_warn_invalid_xdp_action); + static u32 bpf_net_convert_ctx_access(enum bpf_access_type type, int dst_reg, int src_reg, int ctx_off, struct bpf_insn *insn_buf, @@ -2628,6 +2672,29 @@ static u32 bpf_net_convert_ctx_access(enum bpf_access_type type, int dst_reg, return insn - insn_buf; } +static u32 xdp_convert_ctx_access(enum bpf_access_type type, int dst_reg, + int src_reg, int ctx_off, + struct bpf_insn *insn_buf, + struct bpf_prog *prog) +{ + struct bpf_insn *insn = insn_buf; + + switch (ctx_off) { + case offsetof(struct xdp_md, data): + *insn++ = BPF_LDX_MEM(bytes_to_bpf_size(FIELD_SIZEOF(struct xdp_buff, data)), + dst_reg, src_reg, + offsetof(struct xdp_buff, data)); + break; + case offsetof(struct xdp_md, data_end): + *insn++ = BPF_LDX_MEM(bytes_to_bpf_size(FIELD_SIZEOF(struct xdp_buff, data_end)), + dst_reg, src_reg, + offsetof(struct xdp_buff, data_end)); + break; + } + + return insn - insn_buf; +} + static const struct bpf_verifier_ops sk_filter_ops = { .get_func_proto = sk_filter_func_proto, .is_valid_access = sk_filter_is_valid_access, @@ -2640,6 +2707,12 @@ static const struct bpf_verifier_ops tc_cls_act_ops = { .convert_ctx_access = bpf_net_convert_ctx_access, }; +static const struct bpf_verifier_ops xdp_ops = { + .get_func_proto = xdp_func_proto, + .is_valid_access = xdp_is_valid_access, + .convert_ctx_access = xdp_convert_ctx_access, +}; + static struct bpf_prog_type_list sk_filter_type __read_mostly = { .ops = &sk_filter_ops, .type = BPF_PROG_TYPE_SOCKET_FILTER, @@ -2655,11 +2728,17 @@ static struct bpf_prog_type_list sched_act_type __read_mostly = { .type = BPF_PROG_TYPE_SCHED_ACT, }; +static struct bpf_prog_type_list xdp_type __read_mostly = { + .ops = &xdp_ops, + .type = BPF_PROG_TYPE_XDP, +}; + static int __init register_sk_filter_ops(void) { bpf_register_prog_type(&sk_filter_type); bpf_register_prog_type(&sched_cls_type); bpf_register_prog_type(&sched_act_type); + bpf_register_prog_type(&xdp_type); return 0; } -- cgit v0.10.2 From a7862b45849fe2f8610a2bec89235580f55d337f Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:48 -0700 Subject: net: add ndo to setup/query xdp prog in adapter rx Add one new netdev op for drivers implementing the BPF_PROG_TYPE_XDP filter. The single op is used for both setup/query of the xdp program, modelled after ndo_setup_tc. Signed-off-by: Brenden Blanco Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 49736a3..fab9a1c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -63,6 +63,7 @@ struct wpan_dev; struct mpls_dev; /* UDP Tunnel offloads */ struct udp_tunnel_info; +struct bpf_prog; void netdev_set_default_ethtool_ops(struct net_device *dev, const struct ethtool_ops *ops); @@ -799,6 +800,33 @@ struct tc_to_netdev { }; }; +/* These structures hold the attributes of xdp state that are being passed + * to the netdevice through the xdp op. + */ +enum xdp_netdev_command { + /* Set or clear a bpf program used in the earliest stages of packet + * rx. The prog will have been loaded as BPF_PROG_TYPE_XDP. The callee + * is responsible for calling bpf_prog_put on any old progs that are + * stored. In case of error, the callee need not release the new prog + * reference, but on success it takes ownership and must bpf_prog_put + * when it is no longer used. + */ + XDP_SETUP_PROG, + /* Check if a bpf program is set on the device. The callee should + * return true if a program is currently attached and running. + */ + XDP_QUERY_PROG, +}; + +struct netdev_xdp { + enum xdp_netdev_command command; + union { + /* XDP_SETUP_PROG */ + struct bpf_prog *prog; + /* XDP_QUERY_PROG */ + bool prog_attached; + }; +}; /* * This structure defines the management hooks for network devices. @@ -1087,6 +1115,9 @@ struct tc_to_netdev { * appropriate rx headroom value allows avoiding skb head copy on * forward. Setting a negative value resets the rx headroom to the * default value. + * int (*ndo_xdp)(struct net_device *dev, struct netdev_xdp *xdp); + * This function is used to set or query state related to XDP on the + * netdevice. See definition of enum xdp_netdev_command for details. * */ struct net_device_ops { @@ -1271,6 +1302,8 @@ struct net_device_ops { struct sk_buff *skb); void (*ndo_set_rx_headroom)(struct net_device *dev, int needed_headroom); + int (*ndo_xdp)(struct net_device *dev, + struct netdev_xdp *xdp); }; /** @@ -3257,6 +3290,7 @@ int dev_get_phys_port_id(struct net_device *dev, int dev_get_phys_port_name(struct net_device *dev, char *name, size_t len); int dev_change_proto_down(struct net_device *dev, bool proto_down); +int dev_change_xdp_fd(struct net_device *dev, int fd); struct sk_buff *validate_xmit_skb_list(struct sk_buff *skb, struct net_device *dev); struct sk_buff *dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, struct netdev_queue *txq, int *ret); diff --git a/net/core/dev.c b/net/core/dev.c index 7894e40..2a9c39f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -94,6 +94,7 @@ #include #include #include +#include #include #include #include @@ -6615,6 +6616,38 @@ int dev_change_proto_down(struct net_device *dev, bool proto_down) EXPORT_SYMBOL(dev_change_proto_down); /** + * dev_change_xdp_fd - set or clear a bpf program for a device rx path + * @dev: device + * @fd: new program fd or negative value to clear + * + * Set or clear a bpf program for a device + */ +int dev_change_xdp_fd(struct net_device *dev, int fd) +{ + const struct net_device_ops *ops = dev->netdev_ops; + struct bpf_prog *prog = NULL; + struct netdev_xdp xdp = {}; + int err; + + if (!ops->ndo_xdp) + return -EOPNOTSUPP; + if (fd >= 0) { + prog = bpf_prog_get_type(fd, BPF_PROG_TYPE_XDP); + if (IS_ERR(prog)) + return PTR_ERR(prog); + } + + xdp.command = XDP_SETUP_PROG; + xdp.prog = prog; + err = ops->ndo_xdp(dev, &xdp); + if (err < 0 && prog) + bpf_prog_put(prog); + + return err; +} +EXPORT_SYMBOL(dev_change_xdp_fd); + +/** * dev_new_index - allocate an ifindex * @net: the applicable net namespace * -- cgit v0.10.2 From d1fdd9138682e0f272beee0cb08b6328c5478b26 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:49 -0700 Subject: rtnl: add option for setting link xdp prog Sets the bpf program represented by fd as an early filter in the rx path of the netdev. The fd must have been created as BPF_PROG_TYPE_XDP. Providing a negative value as fd clears the program. Getting the fd back via rtnl is not possible, therefore reading of this value merely provides a bool whether the program is valid on the link or not. Signed-off-by: Brenden Blanco Signed-off-by: David S. Miller diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 4285ac3..a1b5202 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -156,6 +156,7 @@ enum { IFLA_GSO_MAX_SEGS, IFLA_GSO_MAX_SIZE, IFLA_PAD, + IFLA_XDP, __IFLA_MAX }; @@ -843,4 +844,15 @@ enum { }; #define LINK_XSTATS_TYPE_MAX (__LINK_XSTATS_TYPE_MAX - 1) +/* XDP section */ + +enum { + IFLA_XDP_UNSPEC, + IFLA_XDP_FD, + IFLA_XDP_ATTACHED, + __IFLA_XDP_MAX, +}; + +#define IFLA_XDP_MAX (__IFLA_XDP_MAX - 1) + #endif /* _UAPI_LINUX_IF_LINK_H */ diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index a9e3805..eba2b82 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -891,6 +891,16 @@ static size_t rtnl_port_size(const struct net_device *dev, return port_self_size; } +static size_t rtnl_xdp_size(const struct net_device *dev) +{ + size_t xdp_size = nla_total_size(1); /* XDP_ATTACHED */ + + if (!dev->netdev_ops->ndo_xdp) + return 0; + else + return xdp_size; +} + static noinline size_t if_nlmsg_size(const struct net_device *dev, u32 ext_filter_mask) { @@ -927,6 +937,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev, + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_PORT_ID */ + nla_total_size(MAX_PHYS_ITEM_ID_LEN) /* IFLA_PHYS_SWITCH_ID */ + nla_total_size(IFNAMSIZ) /* IFLA_PHYS_PORT_NAME */ + + rtnl_xdp_size(dev) /* IFLA_XDP */ + nla_total_size(1); /* IFLA_PROTO_DOWN */ } @@ -1211,6 +1222,33 @@ static int rtnl_fill_link_ifmap(struct sk_buff *skb, struct net_device *dev) return 0; } +static int rtnl_xdp_fill(struct sk_buff *skb, struct net_device *dev) +{ + struct netdev_xdp xdp_op = {}; + struct nlattr *xdp; + int err; + + if (!dev->netdev_ops->ndo_xdp) + return 0; + xdp = nla_nest_start(skb, IFLA_XDP); + if (!xdp) + return -EMSGSIZE; + xdp_op.command = XDP_QUERY_PROG; + err = dev->netdev_ops->ndo_xdp(dev, &xdp_op); + if (err) + goto err_cancel; + err = nla_put_u8(skb, IFLA_XDP_ATTACHED, xdp_op.prog_attached); + if (err) + goto err_cancel; + + nla_nest_end(skb, xdp); + return 0; + +err_cancel: + nla_nest_cancel(skb, xdp); + return err; +} + static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, int type, u32 pid, u32 seq, u32 change, unsigned int flags, u32 ext_filter_mask) @@ -1307,6 +1345,9 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, if (rtnl_port_fill(skb, dev, ext_filter_mask)) goto nla_put_failure; + if (rtnl_xdp_fill(skb, dev)) + goto nla_put_failure; + if (dev->rtnl_link_ops || rtnl_have_link_slave_info(dev)) { if (rtnl_link_fill(skb, dev) < 0) goto nla_put_failure; @@ -1392,6 +1433,7 @@ static const struct nla_policy ifla_policy[IFLA_MAX+1] = { [IFLA_PHYS_SWITCH_ID] = { .type = NLA_BINARY, .len = MAX_PHYS_ITEM_ID_LEN }, [IFLA_LINK_NETNSID] = { .type = NLA_S32 }, [IFLA_PROTO_DOWN] = { .type = NLA_U8 }, + [IFLA_XDP] = { .type = NLA_NESTED }, }; static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = { @@ -1429,6 +1471,11 @@ static const struct nla_policy ifla_port_policy[IFLA_PORT_MAX+1] = { [IFLA_PORT_RESPONSE] = { .type = NLA_U16, }, }; +static const struct nla_policy ifla_xdp_policy[IFLA_XDP_MAX + 1] = { + [IFLA_XDP_FD] = { .type = NLA_S32 }, + [IFLA_XDP_ATTACHED] = { .type = NLA_U8 }, +}; + static const struct rtnl_link_ops *linkinfo_to_kind_ops(const struct nlattr *nla) { const struct rtnl_link_ops *ops = NULL; @@ -2054,6 +2101,23 @@ static int do_setlink(const struct sk_buff *skb, status |= DO_SETLINK_NOTIFY; } + if (tb[IFLA_XDP]) { + struct nlattr *xdp[IFLA_XDP_MAX + 1]; + + err = nla_parse_nested(xdp, IFLA_XDP_MAX, tb[IFLA_XDP], + ifla_xdp_policy); + if (err < 0) + goto errout; + + if (xdp[IFLA_XDP_FD]) { + err = dev_change_xdp_fd(dev, + nla_get_s32(xdp[IFLA_XDP_FD])); + if (err) + goto errout; + status |= DO_SETLINK_NOTIFY; + } + } + errout: if (status & DO_SETLINK_MODIFIED) { if (status & DO_SETLINK_NOTIFY) -- cgit v0.10.2 From 47a38e155037f417c5740e24ccae6482aedf4b68 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:50 -0700 Subject: net/mlx4_en: add support for fast rx drop bpf program Add support for the BPF_PROG_TYPE_XDP hook in mlx4 driver. In tc/socket bpf programs, helpers linearize skb fragments as needed when the program touches the packet data. However, in the pursuit of speed, XDP programs will not be allowed to use these slower functions, especially if it involves allocating an skb. Therefore, disallow MTU settings that would produce a multi-fragment packet that XDP programs would fail to access. Future enhancements could be done to increase the allowable MTU. The xdp program is present as a per-ring data structure, but as of yet it is not possible to set at that granularity through any ndo. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 6083775..c34a33d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -31,6 +31,7 @@ * */ +#include #include #include #include @@ -2112,6 +2113,11 @@ static int mlx4_en_change_mtu(struct net_device *dev, int new_mtu) en_err(priv, "Bad MTU size:%d.\n", new_mtu); return -EPERM; } + if (priv->xdp_ring_num && MLX4_EN_EFF_MTU(new_mtu) > FRAG_SZ0) { + en_err(priv, "MTU size:%d requires frags but XDP running\n", + new_mtu); + return -EOPNOTSUPP; + } dev->mtu = new_mtu; if (netif_running(dev)) { @@ -2520,6 +2526,58 @@ static int mlx4_en_set_tx_maxrate(struct net_device *dev, int queue_index, u32 m return err; } +static int mlx4_xdp_set(struct net_device *dev, struct bpf_prog *prog) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + struct bpf_prog *old_prog; + int xdp_ring_num; + int i; + + xdp_ring_num = prog ? ALIGN(priv->rx_ring_num, MLX4_EN_NUM_UP) : 0; + + if (priv->num_frags > 1) { + en_err(priv, "Cannot set XDP if MTU requires multiple frags\n"); + return -EOPNOTSUPP; + } + + if (prog) { + prog = bpf_prog_add(prog, priv->rx_ring_num - 1); + if (IS_ERR(prog)) + return PTR_ERR(prog); + } + + priv->xdp_ring_num = xdp_ring_num; + + /* This xchg is paired with READ_ONCE in the fast path */ + for (i = 0; i < priv->rx_ring_num; i++) { + old_prog = xchg(&priv->rx_ring[i]->xdp_prog, prog); + if (old_prog) + bpf_prog_put(old_prog); + } + + return 0; +} + +static bool mlx4_xdp_attached(struct net_device *dev) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + + return !!priv->xdp_ring_num; +} + +static int mlx4_xdp(struct net_device *dev, struct netdev_xdp *xdp) +{ + switch (xdp->command) { + case XDP_SETUP_PROG: + return mlx4_xdp_set(dev, xdp->prog); + case XDP_QUERY_PROG: + xdp->prog_attached = mlx4_xdp_attached(dev); + return 0; + default: + return -EINVAL; + } +} + static const struct net_device_ops mlx4_netdev_ops = { .ndo_open = mlx4_en_open, .ndo_stop = mlx4_en_close, @@ -2548,6 +2606,7 @@ static const struct net_device_ops mlx4_netdev_ops = { .ndo_udp_tunnel_del = mlx4_en_del_vxlan_port, .ndo_features_check = mlx4_en_features_check, .ndo_set_tx_maxrate = mlx4_en_set_tx_maxrate, + .ndo_xdp = mlx4_xdp, }; static const struct net_device_ops mlx4_netdev_ops_master = { @@ -2584,6 +2643,7 @@ static const struct net_device_ops mlx4_netdev_ops_master = { .ndo_udp_tunnel_del = mlx4_en_del_vxlan_port, .ndo_features_check = mlx4_en_features_check, .ndo_set_tx_maxrate = mlx4_en_set_tx_maxrate, + .ndo_xdp = mlx4_xdp, }; struct mlx4_en_bond { diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index c1b3a9c..6729545 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -32,6 +32,7 @@ */ #include +#include #include #include #include @@ -509,6 +510,8 @@ void mlx4_en_destroy_rx_ring(struct mlx4_en_priv *priv, struct mlx4_en_dev *mdev = priv->mdev; struct mlx4_en_rx_ring *ring = *pring; + if (ring->xdp_prog) + bpf_prog_put(ring->xdp_prog); mlx4_free_hwq_res(mdev->dev, &ring->wqres, size * stride + TXBB_SIZE); vfree(ring->rx_info); ring->rx_info = NULL; @@ -743,6 +746,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud struct mlx4_en_rx_ring *ring = priv->rx_ring[cq->ring]; struct mlx4_en_rx_alloc *frags; struct mlx4_en_rx_desc *rx_desc; + struct bpf_prog *xdp_prog; struct sk_buff *skb; int index; int nr; @@ -759,6 +763,8 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud if (budget <= 0) return polled; + xdp_prog = READ_ONCE(ring->xdp_prog); + /* We assume a 1:1 mapping between CQEs and Rx descriptors, so Rx * descriptor offset can be deduced from the CQE index instead of * reading 'cqe->index' */ @@ -835,6 +841,35 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud l2_tunnel = (dev->hw_enc_features & NETIF_F_RXCSUM) && (cqe->vlan_my_qpn & cpu_to_be32(MLX4_CQE_L2_TUNNEL)); + /* A bpf program gets first chance to drop the packet. It may + * read bytes but not past the end of the frag. + */ + if (xdp_prog) { + struct xdp_buff xdp; + dma_addr_t dma; + u32 act; + + dma = be64_to_cpu(rx_desc->data[0].addr); + dma_sync_single_for_cpu(priv->ddev, dma, + priv->frag_info[0].frag_size, + DMA_FROM_DEVICE); + + xdp.data = page_address(frags[0].page) + + frags[0].page_offset; + xdp.data_end = xdp.data + length; + + act = bpf_prog_run_xdp(xdp_prog, &xdp); + switch (act) { + case XDP_PASS: + break; + default: + bpf_warn_invalid_xdp_action(act); + case XDP_ABORTED: + case XDP_DROP: + goto next; + } + } + if (likely(dev->features & NETIF_F_RXCSUM)) { if (cqe->status & cpu_to_be16(MLX4_CQE_STATUS_TCP | MLX4_CQE_STATUS_UDP)) { @@ -1062,10 +1097,7 @@ static const int frag_sizes[] = { void mlx4_en_calc_rx_buf(struct net_device *dev) { struct mlx4_en_priv *priv = netdev_priv(dev); - /* VLAN_HLEN is added twice,to support skb vlan tagged with multiple - * headers. (For example: ETH_P_8021Q and ETH_P_8021AD). - */ - int eff_mtu = dev->mtu + ETH_HLEN + (2 * VLAN_HLEN); + int eff_mtu = MLX4_EN_EFF_MTU(dev->mtu); int buf_size = 0; int i = 0; diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index d39bf59..eb1238d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -164,6 +164,10 @@ enum { #define MLX4_LOOPBACK_TEST_PAYLOAD (HEADER_COPY_SIZE - ETH_HLEN) #define MLX4_EN_MIN_MTU 46 +/* VLAN_HLEN is added twice,to support skb vlan tagged with multiple + * headers. (For example: ETH_P_8021Q and ETH_P_8021AD). + */ +#define MLX4_EN_EFF_MTU(mtu) ((mtu) + ETH_HLEN + (2 * VLAN_HLEN)) #define ETH_BCAST 0xffffffffffffULL #define MLX4_EN_LOOPBACK_RETRIES 5 @@ -319,6 +323,7 @@ struct mlx4_en_rx_ring { u8 fcs_del; void *buf; void *rx_info; + struct bpf_prog *xdp_prog; unsigned long bytes; unsigned long packets; unsigned long csum_ok; @@ -558,6 +563,7 @@ struct mlx4_en_priv { struct mlx4_en_frag_info frag_info[MLX4_EN_MAX_RX_FRAGS]; u16 num_frags; u16 log_rx_info; + int xdp_ring_num; struct mlx4_en_tx_ring **tx_ring; struct mlx4_en_rx_ring *rx_ring[MAX_RX_RINGS]; -- cgit v0.10.2 From 86af8b4191d20bb17e868d3167f4cf52ca9331d0 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:51 -0700 Subject: Add sample for adding simple drop program to link Add a sample program that only drops packets at the BPF_PROG_TYPE_XDP_RX hook of a link. With the drop-only program, observed single core rate is ~20Mpps. Other tests were run, for instance without the dropcnt increment or without reading from the packet header, the packet rate was mostly unchanged. $ perf record -a samples/bpf/xdp1 $( Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile index a98b780..0e4ab3a 100644 --- a/samples/bpf/Makefile +++ b/samples/bpf/Makefile @@ -21,6 +21,7 @@ hostprogs-y += spintest hostprogs-y += map_perf_test hostprogs-y += test_overhead hostprogs-y += test_cgrp2_array_pin +hostprogs-y += xdp1 test_verifier-objs := test_verifier.o libbpf.o test_maps-objs := test_maps.o libbpf.o @@ -42,6 +43,7 @@ spintest-objs := bpf_load.o libbpf.o spintest_user.o map_perf_test-objs := bpf_load.o libbpf.o map_perf_test_user.o test_overhead-objs := bpf_load.o libbpf.o test_overhead_user.o test_cgrp2_array_pin-objs := libbpf.o test_cgrp2_array_pin.o +xdp1-objs := bpf_load.o libbpf.o xdp1_user.o # Tell kbuild to always build the programs always := $(hostprogs-y) @@ -64,6 +66,7 @@ always += test_overhead_tp_kern.o always += test_overhead_kprobe_kern.o always += parse_varlen.o parse_simple.o parse_ldabs.o always += test_cgrp2_tc_kern.o +always += xdp1_kern.o HOSTCFLAGS += -I$(objtree)/usr/include @@ -84,6 +87,7 @@ HOSTLOADLIBES_offwaketime += -lelf HOSTLOADLIBES_spintest += -lelf HOSTLOADLIBES_map_perf_test += -lelf -lrt HOSTLOADLIBES_test_overhead += -lelf -lrt +HOSTLOADLIBES_xdp1 += -lelf # Allows pointing LLC/CLANG to a LLVM backend with bpf support, redefine on cmdline: # make samples/bpf/ LLC=~/git/llvm/build/bin/llc CLANG=~/git/llvm/build/bin/clang diff --git a/samples/bpf/bpf_load.c b/samples/bpf/bpf_load.c index 022af71..0cfda23 100644 --- a/samples/bpf/bpf_load.c +++ b/samples/bpf/bpf_load.c @@ -50,6 +50,7 @@ static int load_and_attach(const char *event, struct bpf_insn *prog, int size) bool is_kprobe = strncmp(event, "kprobe/", 7) == 0; bool is_kretprobe = strncmp(event, "kretprobe/", 10) == 0; bool is_tracepoint = strncmp(event, "tracepoint/", 11) == 0; + bool is_xdp = strncmp(event, "xdp", 3) == 0; enum bpf_prog_type prog_type; char buf[256]; int fd, efd, err, id; @@ -66,6 +67,8 @@ static int load_and_attach(const char *event, struct bpf_insn *prog, int size) prog_type = BPF_PROG_TYPE_KPROBE; } else if (is_tracepoint) { prog_type = BPF_PROG_TYPE_TRACEPOINT; + } else if (is_xdp) { + prog_type = BPF_PROG_TYPE_XDP; } else { printf("Unknown event '%s'\n", event); return -1; @@ -79,6 +82,9 @@ static int load_and_attach(const char *event, struct bpf_insn *prog, int size) prog_fd[prog_cnt++] = fd; + if (is_xdp) + return 0; + if (is_socket) { event += 6; if (*event != '/') @@ -319,6 +325,7 @@ int load_bpf_file(char *path) if (memcmp(shname_prog, "kprobe/", 7) == 0 || memcmp(shname_prog, "kretprobe/", 10) == 0 || memcmp(shname_prog, "tracepoint/", 11) == 0 || + memcmp(shname_prog, "xdp", 3) == 0 || memcmp(shname_prog, "socket", 6) == 0) load_and_attach(shname_prog, insns, data_prog->d_size); } @@ -336,6 +343,7 @@ int load_bpf_file(char *path) if (memcmp(shname, "kprobe/", 7) == 0 || memcmp(shname, "kretprobe/", 10) == 0 || memcmp(shname, "tracepoint/", 11) == 0 || + memcmp(shname, "xdp", 3) == 0 || memcmp(shname, "socket", 6) == 0) load_and_attach(shname, data->d_buf, data->d_size); } diff --git a/samples/bpf/xdp1_kern.c b/samples/bpf/xdp1_kern.c new file mode 100644 index 0000000..e7dd8ac --- /dev/null +++ b/samples/bpf/xdp1_kern.c @@ -0,0 +1,93 @@ +/* Copyright (c) 2016 PLUMgrid + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ +#define KBUILD_MODNAME "foo" +#include +#include +#include +#include +#include +#include +#include +#include "bpf_helpers.h" + +struct bpf_map_def SEC("maps") dropcnt = { + .type = BPF_MAP_TYPE_PERCPU_ARRAY, + .key_size = sizeof(u32), + .value_size = sizeof(long), + .max_entries = 256, +}; + +static int parse_ipv4(void *data, u64 nh_off, void *data_end) +{ + struct iphdr *iph = data + nh_off; + + if (iph + 1 > data_end) + return 0; + return iph->protocol; +} + +static int parse_ipv6(void *data, u64 nh_off, void *data_end) +{ + struct ipv6hdr *ip6h = data + nh_off; + + if (ip6h + 1 > data_end) + return 0; + return ip6h->nexthdr; +} + +SEC("xdp1") +int xdp_prog1(struct xdp_md *ctx) +{ + void *data_end = (void *)(long)ctx->data_end; + void *data = (void *)(long)ctx->data; + struct ethhdr *eth = data; + int rc = XDP_DROP; + long *value; + u16 h_proto; + u64 nh_off; + u32 index; + + nh_off = sizeof(*eth); + if (data + nh_off > data_end) + return rc; + + h_proto = eth->h_proto; + + if (h_proto == htons(ETH_P_8021Q) || h_proto == htons(ETH_P_8021AD)) { + struct vlan_hdr *vhdr; + + vhdr = data + nh_off; + nh_off += sizeof(struct vlan_hdr); + if (data + nh_off > data_end) + return rc; + h_proto = vhdr->h_vlan_encapsulated_proto; + } + if (h_proto == htons(ETH_P_8021Q) || h_proto == htons(ETH_P_8021AD)) { + struct vlan_hdr *vhdr; + + vhdr = data + nh_off; + nh_off += sizeof(struct vlan_hdr); + if (data + nh_off > data_end) + return rc; + h_proto = vhdr->h_vlan_encapsulated_proto; + } + + if (h_proto == htons(ETH_P_IP)) + index = parse_ipv4(data, nh_off, data_end); + else if (h_proto == htons(ETH_P_IPV6)) + index = parse_ipv6(data, nh_off, data_end); + else + index = 0; + + value = bpf_map_lookup_elem(&dropcnt, &index); + if (value) + *value += 1; + + return rc; +} + +char _license[] SEC("license") = "GPL"; diff --git a/samples/bpf/xdp1_user.c b/samples/bpf/xdp1_user.c new file mode 100644 index 0000000..a5e109e --- /dev/null +++ b/samples/bpf/xdp1_user.c @@ -0,0 +1,181 @@ +/* Copyright (c) 2016 PLUMgrid + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "bpf_load.h" +#include "libbpf.h" + +static int set_link_xdp_fd(int ifindex, int fd) +{ + struct sockaddr_nl sa; + int sock, seq = 0, len, ret = -1; + char buf[4096]; + struct nlattr *nla, *nla_xdp; + struct { + struct nlmsghdr nh; + struct ifinfomsg ifinfo; + char attrbuf[64]; + } req; + struct nlmsghdr *nh; + struct nlmsgerr *err; + + memset(&sa, 0, sizeof(sa)); + sa.nl_family = AF_NETLINK; + + sock = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE); + if (sock < 0) { + printf("open netlink socket: %s\n", strerror(errno)); + return -1; + } + + if (bind(sock, (struct sockaddr *)&sa, sizeof(sa)) < 0) { + printf("bind to netlink: %s\n", strerror(errno)); + goto cleanup; + } + + memset(&req, 0, sizeof(req)); + req.nh.nlmsg_len = NLMSG_LENGTH(sizeof(struct ifinfomsg)); + req.nh.nlmsg_flags = NLM_F_REQUEST | NLM_F_ACK; + req.nh.nlmsg_type = RTM_SETLINK; + req.nh.nlmsg_pid = 0; + req.nh.nlmsg_seq = ++seq; + req.ifinfo.ifi_family = AF_UNSPEC; + req.ifinfo.ifi_index = ifindex; + nla = (struct nlattr *)(((char *)&req) + + NLMSG_ALIGN(req.nh.nlmsg_len)); + nla->nla_type = NLA_F_NESTED | 43/*IFLA_XDP*/; + + nla_xdp = (struct nlattr *)((char *)nla + NLA_HDRLEN); + nla_xdp->nla_type = 1/*IFLA_XDP_FD*/; + nla_xdp->nla_len = NLA_HDRLEN + sizeof(int); + memcpy((char *)nla_xdp + NLA_HDRLEN, &fd, sizeof(fd)); + nla->nla_len = NLA_HDRLEN + nla_xdp->nla_len; + + req.nh.nlmsg_len += NLA_ALIGN(nla->nla_len); + + if (send(sock, &req, req.nh.nlmsg_len, 0) < 0) { + printf("send to netlink: %s\n", strerror(errno)); + goto cleanup; + } + + len = recv(sock, buf, sizeof(buf), 0); + if (len < 0) { + printf("recv from netlink: %s\n", strerror(errno)); + goto cleanup; + } + + for (nh = (struct nlmsghdr *)buf; NLMSG_OK(nh, len); + nh = NLMSG_NEXT(nh, len)) { + if (nh->nlmsg_pid != getpid()) { + printf("Wrong pid %d, expected %d\n", + nh->nlmsg_pid, getpid()); + goto cleanup; + } + if (nh->nlmsg_seq != seq) { + printf("Wrong seq %d, expected %d\n", + nh->nlmsg_seq, seq); + goto cleanup; + } + switch (nh->nlmsg_type) { + case NLMSG_ERROR: + err = (struct nlmsgerr *)NLMSG_DATA(nh); + if (!err->error) + continue; + printf("nlmsg error %s\n", strerror(-err->error)); + goto cleanup; + case NLMSG_DONE: + break; + } + } + + ret = 0; + +cleanup: + close(sock); + return ret; +} + +static int ifindex; + +static void int_exit(int sig) +{ + set_link_xdp_fd(ifindex, -1); + exit(0); +} + +/* simple per-protocol drop counter + */ +static void poll_stats(int interval) +{ + unsigned int nr_cpus = sysconf(_SC_NPROCESSORS_CONF); + const unsigned int nr_keys = 256; + __u64 values[nr_cpus], prev[nr_keys][nr_cpus]; + __u32 key; + int i; + + memset(prev, 0, sizeof(prev)); + + while (1) { + sleep(interval); + + for (key = 0; key < nr_keys; key++) { + __u64 sum = 0; + + assert(bpf_lookup_elem(map_fd[0], &key, values) == 0); + for (i = 0; i < nr_cpus; i++) + sum += (values[i] - prev[key][i]); + if (sum) + printf("proto %u: %10llu pkt/s\n", + key, sum / interval); + memcpy(prev[key], values, sizeof(values)); + } + } +} + +int main(int ac, char **argv) +{ + char filename[256]; + + snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]); + + if (ac != 2) { + printf("usage: %s IFINDEX\n", argv[0]); + return 1; + } + + ifindex = strtoul(argv[1], NULL, 0); + + if (load_bpf_file(filename)) { + printf("%s", bpf_log_buf); + return 1; + } + + if (!prog_fd[0]) { + printf("load_bpf_file: %s\n", strerror(errno)); + return 1; + } + + signal(SIGINT, int_exit); + + if (set_link_xdp_fd(ifindex, prog_fd[0]) < 0) { + printf("link set xdp fd failed\n"); + return 1; + } + + poll_stats(2); + + return 0; +} -- cgit v0.10.2 From d576acf0a22890cf3f8f7a9b035f1558077f6770 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:52 -0700 Subject: net/mlx4_en: add page recycle to prepare rx ring for tx support The mlx4 driver by default allocates order-3 pages for the ring to consume in multiple fragments. When the device has an xdp program, this behavior will prevent tx actions since the page must be re-mapped in TODEVICE mode, which cannot be done if the page is still shared. Start by making the allocator configurable based on whether xdp is running, such that order-0 pages are always used and never shared. Since this will stress the page allocator, add a simple page cache to each rx ring. Pages in the cache are left dma-mapped, and in drop-only stress tests the page allocator is eliminated from the perf report. Note that setting an xdp program will now require the rings to be reconfigured. Before: 26.91% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq 17.88% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags 6.00% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag 4.49% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist 3.21% swapper [kernel.vmlinux] [k] intel_idle 2.73% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem 2.57% swapper [mlx4_en] [k] mlx4_en_process_rx_cq After: 31.72% swapper [kernel.vmlinux] [k] intel_idle 8.79% swapper [mlx4_en] [k] mlx4_en_process_rx_cq 7.54% swapper [kernel.vmlinux] [k] poll_idle 6.36% swapper [mlx4_core] [k] mlx4_eq_int 4.21% swapper [kernel.vmlinux] [k] tasklet_action 4.03% swapper [kernel.vmlinux] [k] cpuidle_enter_state 3.43% swapper [mlx4_en] [k] mlx4_en_prepare_rx_desc 2.18% swapper [kernel.vmlinux] [k] native_irq_return_iret 1.37% swapper [kernel.vmlinux] [k] menu_select 1.09% swapper [kernel.vmlinux] [k] bpf_map_lookup_elem Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index c34a33d..47ae2a2 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -2529,12 +2529,33 @@ static int mlx4_en_set_tx_maxrate(struct net_device *dev, int queue_index, u32 m static int mlx4_xdp_set(struct net_device *dev, struct bpf_prog *prog) { struct mlx4_en_priv *priv = netdev_priv(dev); + struct mlx4_en_dev *mdev = priv->mdev; struct bpf_prog *old_prog; int xdp_ring_num; + int port_up = 0; + int err; int i; xdp_ring_num = prog ? ALIGN(priv->rx_ring_num, MLX4_EN_NUM_UP) : 0; + /* No need to reconfigure buffers when simply swapping the + * program for a new one. + */ + if (priv->xdp_ring_num == xdp_ring_num) { + if (prog) { + prog = bpf_prog_add(prog, priv->rx_ring_num - 1); + if (IS_ERR(prog)) + return PTR_ERR(prog); + } + for (i = 0; i < priv->rx_ring_num; i++) { + /* This xchg is paired with READ_ONCE in the fastpath */ + old_prog = xchg(&priv->rx_ring[i]->xdp_prog, prog); + if (old_prog) + bpf_prog_put(old_prog); + } + return 0; + } + if (priv->num_frags > 1) { en_err(priv, "Cannot set XDP if MTU requires multiple frags\n"); return -EOPNOTSUPP; @@ -2546,15 +2567,30 @@ static int mlx4_xdp_set(struct net_device *dev, struct bpf_prog *prog) return PTR_ERR(prog); } + mutex_lock(&mdev->state_lock); + if (priv->port_up) { + port_up = 1; + mlx4_en_stop_port(dev, 1); + } + priv->xdp_ring_num = xdp_ring_num; - /* This xchg is paired with READ_ONCE in the fast path */ for (i = 0; i < priv->rx_ring_num; i++) { old_prog = xchg(&priv->rx_ring[i]->xdp_prog, prog); if (old_prog) bpf_prog_put(old_prog); } + if (port_up) { + err = mlx4_en_start_port(dev); + if (err) { + en_err(priv, "Failed starting port %d for XDP change\n", + priv->port); + queue_work(mdev->workqueue, &priv->watchdog_task); + } + } + + mutex_unlock(&mdev->state_lock); return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index 6729545..9dd5dc1 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -58,7 +58,7 @@ static int mlx4_alloc_pages(struct mlx4_en_priv *priv, struct page *page; dma_addr_t dma; - for (order = MLX4_EN_ALLOC_PREFER_ORDER; ;) { + for (order = frag_info->order; ;) { gfp_t gfp = _gfp; if (order) @@ -71,7 +71,7 @@ static int mlx4_alloc_pages(struct mlx4_en_priv *priv, return -ENOMEM; } dma = dma_map_page(priv->ddev, page, 0, PAGE_SIZE << order, - PCI_DMA_FROMDEVICE); + frag_info->dma_dir); if (dma_mapping_error(priv->ddev, dma)) { put_page(page); return -ENOMEM; @@ -125,7 +125,8 @@ out: while (i--) { if (page_alloc[i].page != ring_alloc[i].page) { dma_unmap_page(priv->ddev, page_alloc[i].dma, - page_alloc[i].page_size, PCI_DMA_FROMDEVICE); + page_alloc[i].page_size, + priv->frag_info[i].dma_dir); page = page_alloc[i].page; /* Revert changes done by mlx4_alloc_pages */ page_ref_sub(page, page_alloc[i].page_size / @@ -146,7 +147,7 @@ static void mlx4_en_free_frag(struct mlx4_en_priv *priv, if (next_frag_end > frags[i].page_size) dma_unmap_page(priv->ddev, frags[i].dma, frags[i].page_size, - PCI_DMA_FROMDEVICE); + frag_info->dma_dir); if (frags[i].page) put_page(frags[i].page); @@ -177,7 +178,8 @@ out: page_alloc = &ring->page_alloc[i]; dma_unmap_page(priv->ddev, page_alloc->dma, - page_alloc->page_size, PCI_DMA_FROMDEVICE); + page_alloc->page_size, + priv->frag_info[i].dma_dir); page = page_alloc->page; /* Revert changes done by mlx4_alloc_pages */ page_ref_sub(page, page_alloc->page_size / @@ -202,7 +204,7 @@ static void mlx4_en_destroy_allocator(struct mlx4_en_priv *priv, i, page_count(page_alloc->page)); dma_unmap_page(priv->ddev, page_alloc->dma, - page_alloc->page_size, PCI_DMA_FROMDEVICE); + page_alloc->page_size, frag_info->dma_dir); while (page_alloc->page_offset + frag_info->frag_stride < page_alloc->page_size) { put_page(page_alloc->page); @@ -245,6 +247,12 @@ static int mlx4_en_prepare_rx_desc(struct mlx4_en_priv *priv, struct mlx4_en_rx_alloc *frags = ring->rx_info + (index << priv->log_rx_info); + if (ring->page_cache.index > 0) { + frags[0] = ring->page_cache.buf[--ring->page_cache.index]; + rx_desc->data[0].addr = cpu_to_be64(frags[0].dma); + return 0; + } + return mlx4_en_alloc_frags(priv, rx_desc, frags, ring->page_alloc, gfp); } @@ -503,6 +511,24 @@ void mlx4_en_recover_from_oom(struct mlx4_en_priv *priv) } } +/* When the rx ring is running in page-per-packet mode, a released frame can go + * directly into a small cache, to avoid unmapping or touching the page + * allocator. In bpf prog performance scenarios, buffers are either forwarded + * or dropped, never converted to skbs, so every page can come directly from + * this cache when it is sized to be a multiple of the napi budget. + */ +bool mlx4_en_rx_recycle(struct mlx4_en_rx_ring *ring, + struct mlx4_en_rx_alloc *frame) +{ + struct mlx4_en_page_cache *cache = &ring->page_cache; + + if (cache->index >= MLX4_EN_CACHE_SIZE) + return false; + + cache->buf[cache->index++] = *frame; + return true; +} + void mlx4_en_destroy_rx_ring(struct mlx4_en_priv *priv, struct mlx4_en_rx_ring **pring, u32 size, u16 stride) @@ -525,6 +551,16 @@ void mlx4_en_destroy_rx_ring(struct mlx4_en_priv *priv, void mlx4_en_deactivate_rx_ring(struct mlx4_en_priv *priv, struct mlx4_en_rx_ring *ring) { + int i; + + for (i = 0; i < ring->page_cache.index; i++) { + struct mlx4_en_rx_alloc *frame = &ring->page_cache.buf[i]; + + dma_unmap_page(priv->ddev, frame->dma, frame->page_size, + priv->frag_info[0].dma_dir); + put_page(frame->page); + } + ring->page_cache.index = 0; mlx4_en_free_rx_buf(priv, ring); if (ring->stride <= TXBB_SIZE) ring->buf -= TXBB_SIZE; @@ -866,6 +902,8 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud bpf_warn_invalid_xdp_action(act); case XDP_ABORTED: case XDP_DROP: + if (mlx4_en_rx_recycle(ring, frags)) + goto consumed; goto next; } } @@ -1021,6 +1059,7 @@ next: for (nr = 0; nr < priv->num_frags; nr++) mlx4_en_free_frag(priv, frags, nr); +consumed: ++cq->mcq.cons_index; index = (cq->mcq.cons_index) & ring->size_mask; cqe = mlx4_en_get_cqe(cq->buf, index, priv->cqe_size) + factor; @@ -1096,19 +1135,34 @@ static const int frag_sizes[] = { void mlx4_en_calc_rx_buf(struct net_device *dev) { + enum dma_data_direction dma_dir = PCI_DMA_FROMDEVICE; struct mlx4_en_priv *priv = netdev_priv(dev); int eff_mtu = MLX4_EN_EFF_MTU(dev->mtu); + int order = MLX4_EN_ALLOC_PREFER_ORDER; + u32 align = SMP_CACHE_BYTES; int buf_size = 0; int i = 0; + /* bpf requires buffers to be set up as 1 packet per page. + * This only works when num_frags == 1. + */ + if (priv->xdp_ring_num) { + /* This will gain efficient xdp frame recycling at the expense + * of more costly truesize accounting + */ + align = PAGE_SIZE; + order = 0; + } + while (buf_size < eff_mtu) { + priv->frag_info[i].order = order; priv->frag_info[i].frag_size = (eff_mtu > buf_size + frag_sizes[i]) ? frag_sizes[i] : eff_mtu - buf_size; priv->frag_info[i].frag_prefix_size = buf_size; priv->frag_info[i].frag_stride = - ALIGN(priv->frag_info[i].frag_size, - SMP_CACHE_BYTES); + ALIGN(priv->frag_info[i].frag_size, align); + priv->frag_info[i].dma_dir = dma_dir; buf_size += priv->frag_info[i].frag_size; i++; } diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index eb1238d..eff4be0 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -259,6 +259,12 @@ struct mlx4_en_rx_alloc { u32 page_size; }; +#define MLX4_EN_CACHE_SIZE (2 * NAPI_POLL_WEIGHT) +struct mlx4_en_page_cache { + u32 index; + struct mlx4_en_rx_alloc buf[MLX4_EN_CACHE_SIZE]; +}; + struct mlx4_en_tx_ring { /* cache line used and dirtied in tx completion * (mlx4_en_free_tx_buf()) @@ -324,6 +330,7 @@ struct mlx4_en_rx_ring { void *buf; void *rx_info; struct bpf_prog *xdp_prog; + struct mlx4_en_page_cache page_cache; unsigned long bytes; unsigned long packets; unsigned long csum_ok; @@ -443,7 +450,9 @@ struct mlx4_en_mc_list { struct mlx4_en_frag_info { u16 frag_size; u16 frag_prefix_size; - u16 frag_stride; + u32 frag_stride; + enum dma_data_direction dma_dir; + int order; }; #ifdef CONFIG_MLX4_EN_DCB -- cgit v0.10.2 From 6ce96ca348a9e949f8c43f4d3e98db367d93cffd Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:53 -0700 Subject: bpf: add XDP_TX xdp_action for direct forwarding XDP enabled drivers must transmit received packets back out on the same port they were received on when a program returns this action. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index a517865..2b7076f 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -449,6 +449,7 @@ enum xdp_action { XDP_ABORTED = 0, XDP_DROP, XDP_PASS, + XDP_TX, }; /* user accessible metadata for XDP packet hook -- cgit v0.10.2 From 224e92e02a769b8028ca2450443586af8b4f1715 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:54 -0700 Subject: net/mlx4_en: break out tx_desc write into separate function In preparation for writing the tx descriptor from multiple functions, create a helper for both normal and blueflame access. Signed-off-by: Brenden Blanco Signed-off-by: David S. Miller diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index 8db8405..768085f 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -232,7 +232,7 @@ static void stamp_send_wqe(struct mlx4_ib_qp *qp, int n, int size) } } else { ctrl = buf = get_send_wqe(qp, n & (qp->sq.wqe_cnt - 1)); - s = (ctrl->fence_size & 0x3f) << 4; + s = (ctrl->qpn_vlan.fence_size & 0x3f) << 4; for (i = 64; i < s; i += 64) { wqe = buf + i; *wqe = cpu_to_be32(0xffffffff); @@ -264,7 +264,7 @@ static void post_nop_wqe(struct mlx4_ib_qp *qp, int n, int size) inl->byte_count = cpu_to_be32(1 << 31 | (size - s - sizeof *inl)); } ctrl->srcrb_flags = 0; - ctrl->fence_size = size / 16; + ctrl->qpn_vlan.fence_size = size / 16; /* * Make sure descriptor is fully written before setting ownership bit * (because HW can start executing as soon as we do). @@ -1992,7 +1992,8 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp, ctrl = get_send_wqe(qp, i); ctrl->owner_opcode = cpu_to_be32(1 << 31); if (qp->sq_max_wqes_per_wr == 1) - ctrl->fence_size = 1 << (qp->sq.wqe_shift - 4); + ctrl->qpn_vlan.fence_size = + 1 << (qp->sq.wqe_shift - 4); stamp_send_wqe(qp, i, 1 << qp->sq.wqe_shift); } @@ -3169,8 +3170,8 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, wmb(); *lso_wqe = lso_hdr_sz; - ctrl->fence_size = (wr->send_flags & IB_SEND_FENCE ? - MLX4_WQE_CTRL_FENCE : 0) | size; + ctrl->qpn_vlan.fence_size = (wr->send_flags & IB_SEND_FENCE ? + MLX4_WQE_CTRL_FENCE : 0) | size; /* * Make sure descriptor is fully written before diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c index 76aa4d2..2f56018 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c @@ -631,8 +631,7 @@ static int get_real_size(const struct sk_buff *skb, static void build_inline_wqe(struct mlx4_en_tx_desc *tx_desc, const struct sk_buff *skb, const struct skb_shared_info *shinfo, - int real_size, u16 *vlan_tag, - int tx_ind, void *fragptr) + void *fragptr) { struct mlx4_wqe_inline_seg *inl = &tx_desc->inl; int spc = MLX4_INLINE_ALIGN - CTRL_SIZE - sizeof *inl; @@ -700,10 +699,66 @@ static void mlx4_bf_copy(void __iomem *dst, const void *src, __iowrite64_copy(dst, src, bytecnt / 8); } +void mlx4_en_xmit_doorbell(struct mlx4_en_tx_ring *ring) +{ + wmb(); + /* Since there is no iowrite*_native() that writes the + * value as is, without byteswapping - using the one + * the doesn't do byteswapping in the relevant arch + * endianness. + */ +#if defined(__LITTLE_ENDIAN) + iowrite32( +#else + iowrite32be( +#endif + ring->doorbell_qpn, + ring->bf.uar->map + MLX4_SEND_DOORBELL); +} + +static void mlx4_en_tx_write_desc(struct mlx4_en_tx_ring *ring, + struct mlx4_en_tx_desc *tx_desc, + union mlx4_wqe_qpn_vlan qpn_vlan, + int desc_size, int bf_index, + __be32 op_own, bool bf_ok, + bool send_doorbell) +{ + tx_desc->ctrl.qpn_vlan = qpn_vlan; + + if (bf_ok) { + op_own |= htonl((bf_index & 0xffff) << 8); + /* Ensure new descriptor hits memory + * before setting ownership of this descriptor to HW + */ + dma_wmb(); + tx_desc->ctrl.owner_opcode = op_own; + + wmb(); + + mlx4_bf_copy(ring->bf.reg + ring->bf.offset, &tx_desc->ctrl, + desc_size); + + wmb(); + + ring->bf.offset ^= ring->bf.buf_size; + } else { + /* Ensure new descriptor hits memory + * before setting ownership of this descriptor to HW + */ + dma_wmb(); + tx_desc->ctrl.owner_opcode = op_own; + if (send_doorbell) + mlx4_en_xmit_doorbell(ring); + else + ring->xmit_more++; + } +} + netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) { struct skb_shared_info *shinfo = skb_shinfo(skb); struct mlx4_en_priv *priv = netdev_priv(dev); + union mlx4_wqe_qpn_vlan qpn_vlan = {}; struct device *ddev = priv->ddev; struct mlx4_en_tx_ring *ring; struct mlx4_en_tx_desc *tx_desc; @@ -715,7 +770,6 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) int real_size; u32 index, bf_index; __be32 op_own; - u16 vlan_tag = 0; u16 vlan_proto = 0; int i_frag; int lso_header_size; @@ -725,6 +779,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) bool stop_queue; bool inline_ok; u32 ring_cons; + bool bf_ok; tx_ind = skb_get_queue_mapping(skb); ring = priv->tx_ring[tx_ind]; @@ -749,9 +804,17 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) goto tx_drop; } + bf_ok = ring->bf_enabled; if (skb_vlan_tag_present(skb)) { - vlan_tag = skb_vlan_tag_get(skb); + qpn_vlan.vlan_tag = cpu_to_be16(skb_vlan_tag_get(skb)); vlan_proto = be16_to_cpu(skb->vlan_proto); + if (vlan_proto == ETH_P_8021AD) + qpn_vlan.ins_vlan = MLX4_WQE_CTRL_INS_SVLAN; + else if (vlan_proto == ETH_P_8021Q) + qpn_vlan.ins_vlan = MLX4_WQE_CTRL_INS_CVLAN; + else + qpn_vlan.ins_vlan = 0; + bf_ok = false; } netdev_txq_bql_enqueue_prefetchw(ring->tx_queue); @@ -771,6 +834,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) else { tx_desc = (struct mlx4_en_tx_desc *) ring->bounce_buf; bounce = true; + bf_ok = false; } /* Save skb in tx_info ring */ @@ -907,8 +971,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) AVG_PERF_COUNTER(priv->pstats.tx_pktsz_avg, skb->len); if (tx_info->inl) - build_inline_wqe(tx_desc, skb, shinfo, real_size, &vlan_tag, - tx_ind, fragptr); + build_inline_wqe(tx_desc, skb, shinfo, fragptr); if (skb->encapsulation) { union { @@ -946,60 +1009,15 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) real_size = (real_size / 16) & 0x3f; - if (ring->bf_enabled && desc_size <= MAX_BF && !bounce && - !skb_vlan_tag_present(skb) && send_doorbell) { - tx_desc->ctrl.bf_qpn = ring->doorbell_qpn | - cpu_to_be32(real_size); - - op_own |= htonl((bf_index & 0xffff) << 8); - /* Ensure new descriptor hits memory - * before setting ownership of this descriptor to HW - */ - dma_wmb(); - tx_desc->ctrl.owner_opcode = op_own; - - wmb(); - - mlx4_bf_copy(ring->bf.reg + ring->bf.offset, &tx_desc->ctrl, - desc_size); - - wmb(); - - ring->bf.offset ^= ring->bf.buf_size; - } else { - tx_desc->ctrl.vlan_tag = cpu_to_be16(vlan_tag); - if (vlan_proto == ETH_P_8021AD) - tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_SVLAN; - else if (vlan_proto == ETH_P_8021Q) - tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_CVLAN; - else - tx_desc->ctrl.ins_vlan = 0; + bf_ok &= desc_size <= MAX_BF && send_doorbell; - tx_desc->ctrl.fence_size = real_size; + if (bf_ok) + qpn_vlan.bf_qpn = ring->doorbell_qpn | cpu_to_be32(real_size); + else + qpn_vlan.fence_size = real_size; - /* Ensure new descriptor hits memory - * before setting ownership of this descriptor to HW - */ - dma_wmb(); - tx_desc->ctrl.owner_opcode = op_own; - if (send_doorbell) { - wmb(); - /* Since there is no iowrite*_native() that writes the - * value as is, without byteswapping - using the one - * the doesn't do byteswapping in the relevant arch - * endianness. - */ -#if defined(__LITTLE_ENDIAN) - iowrite32( -#else - iowrite32be( -#endif - ring->doorbell_qpn, - ring->bf.uar->map + MLX4_SEND_DOORBELL); - } else { - ring->xmit_more++; - } - } + mlx4_en_tx_write_desc(ring, tx_desc, qpn_vlan, desc_size, bf_index, + op_own, bf_ok, send_doorbell); if (unlikely(stop_queue)) { /* If queue was emptied after the if (stop_queue) , and before diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h index 587cdf9..deaa221 100644 --- a/include/linux/mlx4/qp.h +++ b/include/linux/mlx4/qp.h @@ -291,16 +291,18 @@ enum { MLX4_WQE_CTRL_FORCE_LOOPBACK = 1 << 0, }; +union mlx4_wqe_qpn_vlan { + struct { + __be16 vlan_tag; + u8 ins_vlan; + u8 fence_size; + }; + __be32 bf_qpn; +}; + struct mlx4_wqe_ctrl_seg { __be32 owner_opcode; - union { - struct { - __be16 vlan_tag; - u8 ins_vlan; - u8 fence_size; - }; - __be32 bf_qpn; - }; + union mlx4_wqe_qpn_vlan qpn_vlan; /* * High 24 bits are SRC remote buffer; low 8 bits are flags: * [7] SO (strong ordering) -- cgit v0.10.2 From 9ecc2d86171adf23796133c89610987a14624875 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:55 -0700 Subject: net/mlx4_en: add xdp forwarding and data write support A user will now be able to loop packets back out of the same port using a bpf program attached to xdp hook. Updates to the packet contents from the bpf program is also supported. For the packet write feature to work, the rx buffers are now mapped as bidirectional when the page is allocated. This occurs only when the xdp hook is active. When the program returns a TX action, enqueue the packet directly to a dedicated tx ring, so as to avoid completely any locking. This requires the tx ring to be allocated 1:1 for each rx ring, as well as the tx completion running in the same softirq. Upon tx completion, this dedicated tx ring recycles pages without unmapping directly back to the original rx ring. In steady state tx/drop workload, effectively 0 page allocs/frees will occur. In order to separate out the paths between free and recycle, a free_tx_desc func pointer is introduced that is optionally updated whenever recycle_ring is activated. By default the original free function is always initialized. Signed-off-by: Brenden Blanco Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index 51a2e82..f32e272 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -1722,6 +1722,12 @@ static int mlx4_en_set_channels(struct net_device *dev, !channel->tx_count || !channel->rx_count) return -EINVAL; + if (channel->tx_count * MLX4_EN_NUM_UP <= priv->xdp_ring_num) { + en_err(priv, "Minimum %d tx channels required with XDP on\n", + priv->xdp_ring_num / MLX4_EN_NUM_UP + 1); + return -EINVAL; + } + mutex_lock(&mdev->state_lock); if (priv->port_up) { port_up = 1; @@ -1740,7 +1746,8 @@ static int mlx4_en_set_channels(struct net_device *dev, goto out; } - netif_set_real_num_tx_queues(dev, priv->tx_ring_num); + netif_set_real_num_tx_queues(dev, priv->tx_ring_num - + priv->xdp_ring_num); netif_set_real_num_rx_queues(dev, priv->rx_ring_num); if (dev->num_tc) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 47ae2a2..9abbba6 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -1522,6 +1522,24 @@ static void mlx4_en_free_affinity_hint(struct mlx4_en_priv *priv, int ring_idx) free_cpumask_var(priv->rx_ring[ring_idx]->affinity_mask); } +static void mlx4_en_init_recycle_ring(struct mlx4_en_priv *priv, + int tx_ring_idx) +{ + struct mlx4_en_tx_ring *tx_ring = priv->tx_ring[tx_ring_idx]; + int rr_index; + + rr_index = (priv->xdp_ring_num - priv->tx_ring_num) + tx_ring_idx; + if (rr_index >= 0) { + tx_ring->free_tx_desc = mlx4_en_recycle_tx_desc; + tx_ring->recycle_ring = priv->rx_ring[rr_index]; + en_dbg(DRV, priv, + "Set tx_ring[%d]->recycle_ring = rx_ring[%d]\n", + tx_ring_idx, rr_index); + } else { + tx_ring->recycle_ring = NULL; + } +} + int mlx4_en_start_port(struct net_device *dev) { struct mlx4_en_priv *priv = netdev_priv(dev); @@ -1644,6 +1662,8 @@ int mlx4_en_start_port(struct net_device *dev) } tx_ring->tx_queue = netdev_get_tx_queue(dev, i); + mlx4_en_init_recycle_ring(priv, i); + /* Arm CQ for TX completions */ mlx4_en_arm_cq(priv, cq); @@ -2561,6 +2581,13 @@ static int mlx4_xdp_set(struct net_device *dev, struct bpf_prog *prog) return -EOPNOTSUPP; } + if (priv->tx_ring_num < xdp_ring_num + MLX4_EN_NUM_UP) { + en_err(priv, + "Minimum %d tx channels required to run XDP\n", + (xdp_ring_num + MLX4_EN_NUM_UP) / MLX4_EN_NUM_UP); + return -EINVAL; + } + if (prog) { prog = bpf_prog_add(prog, priv->rx_ring_num - 1); if (IS_ERR(prog)) @@ -2574,6 +2601,8 @@ static int mlx4_xdp_set(struct net_device *dev, struct bpf_prog *prog) } priv->xdp_ring_num = xdp_ring_num; + netif_set_real_num_tx_queues(dev, priv->tx_ring_num - + priv->xdp_ring_num); for (i = 0; i < priv->rx_ring_num; i++) { old_prog = xchg(&priv->rx_ring[i]->xdp_prog, prog); diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index 9dd5dc1..11d88c8 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -783,7 +783,9 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud struct mlx4_en_rx_alloc *frags; struct mlx4_en_rx_desc *rx_desc; struct bpf_prog *xdp_prog; + int doorbell_pending; struct sk_buff *skb; + int tx_index; int index; int nr; unsigned int length; @@ -800,6 +802,8 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud return polled; xdp_prog = READ_ONCE(ring->xdp_prog); + doorbell_pending = 0; + tx_index = (priv->tx_ring_num - priv->xdp_ring_num) + cq->ring; /* We assume a 1:1 mapping between CQEs and Rx descriptors, so Rx * descriptor offset can be deduced from the CQE index instead of @@ -898,6 +902,12 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud switch (act) { case XDP_PASS: break; + case XDP_TX: + if (!mlx4_en_xmit_frame(frags, dev, + length, tx_index, + &doorbell_pending)) + goto consumed; + break; default: bpf_warn_invalid_xdp_action(act); case XDP_ABORTED: @@ -1068,6 +1078,9 @@ consumed: } out: + if (doorbell_pending) + mlx4_en_xmit_doorbell(priv->tx_ring[tx_index]); + AVG_PERF_COUNTER(priv->pstats.rx_coal_avg, polled); mlx4_cq_set_ci(&cq->mcq); wmb(); /* ensure HW sees CQ consumer before we post new buffers */ @@ -1147,6 +1160,7 @@ void mlx4_en_calc_rx_buf(struct net_device *dev) * This only works when num_frags == 1. */ if (priv->xdp_ring_num) { + dma_dir = PCI_DMA_BIDIRECTIONAL; /* This will gain efficient xdp frame recycling at the expense * of more costly truesize accounting */ diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c index 2f56018..9df87ca 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c @@ -196,6 +196,7 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv, ring->last_nr_txbb = 1; memset(ring->tx_info, 0, ring->size * sizeof(struct mlx4_en_tx_info)); memset(ring->buf, 0, ring->buf_size); + ring->free_tx_desc = mlx4_en_free_tx_desc; ring->qp_state = MLX4_QP_STATE_RST; ring->doorbell_qpn = cpu_to_be32(ring->qp.qpn << 8); @@ -265,10 +266,10 @@ static void mlx4_en_stamp_wqe(struct mlx4_en_priv *priv, } -static u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv, - struct mlx4_en_tx_ring *ring, - int index, u8 owner, u64 timestamp, - int napi_mode) +u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv, + struct mlx4_en_tx_ring *ring, + int index, u8 owner, u64 timestamp, + int napi_mode) { struct mlx4_en_tx_info *tx_info = &ring->tx_info[index]; struct mlx4_en_tx_desc *tx_desc = ring->buf + index * TXBB_SIZE; @@ -344,6 +345,27 @@ static u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv, return tx_info->nr_txbb; } +u32 mlx4_en_recycle_tx_desc(struct mlx4_en_priv *priv, + struct mlx4_en_tx_ring *ring, + int index, u8 owner, u64 timestamp, + int napi_mode) +{ + struct mlx4_en_tx_info *tx_info = &ring->tx_info[index]; + struct mlx4_en_rx_alloc frame = { + .page = tx_info->page, + .dma = tx_info->map0_dma, + .page_offset = 0, + .page_size = PAGE_SIZE, + }; + + if (!mlx4_en_rx_recycle(ring->recycle_ring, &frame)) { + dma_unmap_page(priv->ddev, tx_info->map0_dma, + PAGE_SIZE, priv->frag_info[0].dma_dir); + put_page(tx_info->page); + } + + return tx_info->nr_txbb; +} int mlx4_en_free_tx_buf(struct net_device *dev, struct mlx4_en_tx_ring *ring) { @@ -362,7 +384,7 @@ int mlx4_en_free_tx_buf(struct net_device *dev, struct mlx4_en_tx_ring *ring) } while (ring->cons != ring->prod) { - ring->last_nr_txbb = mlx4_en_free_tx_desc(priv, ring, + ring->last_nr_txbb = ring->free_tx_desc(priv, ring, ring->cons & ring->size_mask, !!(ring->cons & ring->size), 0, 0 /* Non-NAPI caller */); @@ -444,7 +466,7 @@ static bool mlx4_en_process_tx_cq(struct net_device *dev, timestamp = mlx4_en_get_cqe_ts(cqe); /* free next descriptor */ - last_nr_txbb = mlx4_en_free_tx_desc( + last_nr_txbb = ring->free_tx_desc( priv, ring, ring_index, !!((ring_cons + txbbs_skipped) & ring->size), timestamp, napi_budget); @@ -476,6 +498,9 @@ static bool mlx4_en_process_tx_cq(struct net_device *dev, ACCESS_ONCE(ring->last_nr_txbb) = last_nr_txbb; ACCESS_ONCE(ring->cons) = ring_cons + txbbs_skipped; + if (ring->free_tx_desc == mlx4_en_recycle_tx_desc) + return done < budget; + netdev_tx_completed_queue(ring->tx_queue, packets, bytes); /* Wakeup Tx queue if this stopped, and ring is not full. @@ -1052,3 +1077,106 @@ tx_drop: return NETDEV_TX_OK; } +netdev_tx_t mlx4_en_xmit_frame(struct mlx4_en_rx_alloc *frame, + struct net_device *dev, unsigned int length, + int tx_ind, int *doorbell_pending) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + union mlx4_wqe_qpn_vlan qpn_vlan = {}; + struct mlx4_en_tx_ring *ring; + struct mlx4_en_tx_desc *tx_desc; + struct mlx4_wqe_data_seg *data; + struct mlx4_en_tx_info *tx_info; + int index, bf_index; + bool send_doorbell; + int nr_txbb = 1; + bool stop_queue; + dma_addr_t dma; + int real_size; + __be32 op_own; + u32 ring_cons; + bool bf_ok; + + BUILD_BUG_ON_MSG(ALIGN(CTRL_SIZE + DS_SIZE, TXBB_SIZE) != TXBB_SIZE, + "mlx4_en_xmit_frame requires minimum size tx desc"); + + ring = priv->tx_ring[tx_ind]; + + if (!priv->port_up) + goto tx_drop; + + if (mlx4_en_is_tx_ring_full(ring)) + goto tx_drop; + + /* fetch ring->cons far ahead before needing it to avoid stall */ + ring_cons = READ_ONCE(ring->cons); + + index = ring->prod & ring->size_mask; + tx_info = &ring->tx_info[index]; + + bf_ok = ring->bf_enabled; + + /* Track current inflight packets for performance analysis */ + AVG_PERF_COUNTER(priv->pstats.inflight_avg, + (u32)(ring->prod - ring_cons - 1)); + + bf_index = ring->prod; + tx_desc = ring->buf + index * TXBB_SIZE; + data = &tx_desc->data; + + dma = frame->dma; + + tx_info->page = frame->page; + frame->page = NULL; + tx_info->map0_dma = dma; + tx_info->map0_byte_count = length; + tx_info->nr_txbb = nr_txbb; + tx_info->nr_bytes = max_t(unsigned int, length, ETH_ZLEN); + tx_info->data_offset = (void *)data - (void *)tx_desc; + tx_info->ts_requested = 0; + tx_info->nr_maps = 1; + tx_info->linear = 1; + tx_info->inl = 0; + + dma_sync_single_for_device(priv->ddev, dma, length, PCI_DMA_TODEVICE); + + data->addr = cpu_to_be64(dma); + data->lkey = ring->mr_key; + dma_wmb(); + data->byte_count = cpu_to_be32(length); + + /* tx completion can avoid cache line miss for common cases */ + tx_desc->ctrl.srcrb_flags = priv->ctrl_flags; + + op_own = cpu_to_be32(MLX4_OPCODE_SEND) | + ((ring->prod & ring->size) ? + cpu_to_be32(MLX4_EN_BIT_DESC_OWN) : 0); + + ring->packets++; + ring->bytes += tx_info->nr_bytes; + AVG_PERF_COUNTER(priv->pstats.tx_pktsz_avg, length); + + ring->prod += nr_txbb; + + stop_queue = mlx4_en_is_tx_ring_full(ring); + send_doorbell = stop_queue || + *doorbell_pending > MLX4_EN_DOORBELL_BUDGET; + bf_ok &= send_doorbell; + + real_size = ((CTRL_SIZE + nr_txbb * DS_SIZE) / 16) & 0x3f; + + if (bf_ok) + qpn_vlan.bf_qpn = ring->doorbell_qpn | cpu_to_be32(real_size); + else + qpn_vlan.fence_size = real_size; + + mlx4_en_tx_write_desc(ring, tx_desc, qpn_vlan, TXBB_SIZE, bf_index, + op_own, bf_ok, send_doorbell); + *doorbell_pending = send_doorbell ? 0 : *doorbell_pending + 1; + + return NETDEV_TX_OK; + +tx_drop: + ring->tx_dropped++; + return NETDEV_TX_BUSY; +} diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index eff4be0..29c81d2 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -132,6 +132,7 @@ enum { MLX4_EN_NUM_UP) #define MLX4_EN_DEFAULT_TX_WORK 256 +#define MLX4_EN_DOORBELL_BUDGET 8 /* Target number of packets to coalesce with interrupt moderation */ #define MLX4_EN_RX_COAL_TARGET 44 @@ -219,7 +220,10 @@ enum cq_type { struct mlx4_en_tx_info { - struct sk_buff *skb; + union { + struct sk_buff *skb; + struct page *page; + }; dma_addr_t map0_dma; u32 map0_byte_count; u32 nr_txbb; @@ -265,6 +269,8 @@ struct mlx4_en_page_cache { struct mlx4_en_rx_alloc buf[MLX4_EN_CACHE_SIZE]; }; +struct mlx4_en_priv; + struct mlx4_en_tx_ring { /* cache line used and dirtied in tx completion * (mlx4_en_free_tx_buf()) @@ -298,6 +304,11 @@ struct mlx4_en_tx_ring { __be32 mr_key; void *buf; struct mlx4_en_tx_info *tx_info; + struct mlx4_en_rx_ring *recycle_ring; + u32 (*free_tx_desc)(struct mlx4_en_priv *priv, + struct mlx4_en_tx_ring *ring, + int index, u8 owner, + u64 timestamp, int napi_mode); u8 *bounce_buf; struct mlx4_qp_context context; int qpn; @@ -678,6 +689,12 @@ void mlx4_en_tx_irq(struct mlx4_cq *mcq); u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb, void *accel_priv, select_queue_fallback_t fallback); netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev); +netdev_tx_t mlx4_en_xmit_frame(struct mlx4_en_rx_alloc *frame, + struct net_device *dev, unsigned int length, + int tx_ind, int *doorbell_pending); +void mlx4_en_xmit_doorbell(struct mlx4_en_tx_ring *ring); +bool mlx4_en_rx_recycle(struct mlx4_en_rx_ring *ring, + struct mlx4_en_rx_alloc *frame); int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring **pring, @@ -706,6 +723,14 @@ int mlx4_en_process_rx_cq(struct net_device *dev, int budget); int mlx4_en_poll_rx_cq(struct napi_struct *napi, int budget); int mlx4_en_poll_tx_cq(struct napi_struct *napi, int budget); +u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv, + struct mlx4_en_tx_ring *ring, + int index, u8 owner, u64 timestamp, + int napi_mode); +u32 mlx4_en_recycle_tx_desc(struct mlx4_en_priv *priv, + struct mlx4_en_tx_ring *ring, + int index, u8 owner, u64 timestamp, + int napi_mode); void mlx4_en_fill_qp_context(struct mlx4_en_priv *priv, int size, int stride, int is_tx, int rss, int qpn, int cqn, int user_prio, struct mlx4_qp_context *context); -- cgit v0.10.2 From 4acf6c0b84c91243c705303cd9ff16421914150d Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:56 -0700 Subject: bpf: enable direct packet data write for xdp progs For forwarding to be effective, XDP programs should be allowed to rewrite packet data. This requires that the drivers supporting XDP must all map the packet memory as TODEVICE or BIDIRECTIONAL before invoking the program. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index a8d67d0..f72f23b 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -653,6 +653,16 @@ static int check_map_access(struct verifier_env *env, u32 regno, int off, #define MAX_PACKET_OFF 0xffff +static bool may_write_pkt_data(enum bpf_prog_type type) +{ + switch (type) { + case BPF_PROG_TYPE_XDP: + return true; + default: + return false; + } +} + static int check_packet_access(struct verifier_env *env, u32 regno, int off, int size) { @@ -806,10 +816,15 @@ static int check_mem_access(struct verifier_env *env, u32 regno, int off, err = check_stack_read(state, off, size, value_regno); } } else if (state->regs[regno].type == PTR_TO_PACKET) { - if (t == BPF_WRITE) { + if (t == BPF_WRITE && !may_write_pkt_data(env->prog->type)) { verbose("cannot write into packet\n"); return -EACCES; } + if (t == BPF_WRITE && value_regno >= 0 && + is_pointer_value(env, value_regno)) { + verbose("R%d leaks addr into packet\n", value_regno); + return -EACCES; + } err = check_packet_access(env, regno, off, size); if (!err && t == BPF_READ && value_regno >= 0) mark_reg_unknown_value(state->regs, value_regno); -- cgit v0.10.2 From 764cbccef8c9cb95e869ba2bb8371c42685c934a Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:57 -0700 Subject: bpf: add sample for xdp forwarding and rewrite Add a sample that rewrites and forwards packets out on the same interface. Observed single core forwarding performance of ~10Mpps. Since the mlx4 driver under test recycles every single packet page, the perf output shows almost exclusively just the ring management and bpf program work. Slowdowns are likely occurring due to cache misses. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile index 0e4ab3a..d2d2b35 100644 --- a/samples/bpf/Makefile +++ b/samples/bpf/Makefile @@ -22,6 +22,7 @@ hostprogs-y += map_perf_test hostprogs-y += test_overhead hostprogs-y += test_cgrp2_array_pin hostprogs-y += xdp1 +hostprogs-y += xdp2 test_verifier-objs := test_verifier.o libbpf.o test_maps-objs := test_maps.o libbpf.o @@ -44,6 +45,8 @@ map_perf_test-objs := bpf_load.o libbpf.o map_perf_test_user.o test_overhead-objs := bpf_load.o libbpf.o test_overhead_user.o test_cgrp2_array_pin-objs := libbpf.o test_cgrp2_array_pin.o xdp1-objs := bpf_load.o libbpf.o xdp1_user.o +# reuse xdp1 source intentionally +xdp2-objs := bpf_load.o libbpf.o xdp1_user.o # Tell kbuild to always build the programs always := $(hostprogs-y) @@ -67,6 +70,7 @@ always += test_overhead_kprobe_kern.o always += parse_varlen.o parse_simple.o parse_ldabs.o always += test_cgrp2_tc_kern.o always += xdp1_kern.o +always += xdp2_kern.o HOSTCFLAGS += -I$(objtree)/usr/include @@ -88,6 +92,7 @@ HOSTLOADLIBES_spintest += -lelf HOSTLOADLIBES_map_perf_test += -lelf -lrt HOSTLOADLIBES_test_overhead += -lelf -lrt HOSTLOADLIBES_xdp1 += -lelf +HOSTLOADLIBES_xdp2 += -lelf # Allows pointing LLC/CLANG to a LLVM backend with bpf support, redefine on cmdline: # make samples/bpf/ LLC=~/git/llvm/build/bin/llc CLANG=~/git/llvm/build/bin/clang diff --git a/samples/bpf/xdp2_kern.c b/samples/bpf/xdp2_kern.c new file mode 100644 index 0000000..38fe7e1 --- /dev/null +++ b/samples/bpf/xdp2_kern.c @@ -0,0 +1,114 @@ +/* Copyright (c) 2016 PLUMgrid + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ +#define KBUILD_MODNAME "foo" +#include +#include +#include +#include +#include +#include +#include +#include "bpf_helpers.h" + +struct bpf_map_def SEC("maps") dropcnt = { + .type = BPF_MAP_TYPE_PERCPU_ARRAY, + .key_size = sizeof(u32), + .value_size = sizeof(long), + .max_entries = 256, +}; + +static void swap_src_dst_mac(void *data) +{ + unsigned short *p = data; + unsigned short dst[3]; + + dst[0] = p[0]; + dst[1] = p[1]; + dst[2] = p[2]; + p[0] = p[3]; + p[1] = p[4]; + p[2] = p[5]; + p[3] = dst[0]; + p[4] = dst[1]; + p[5] = dst[2]; +} + +static int parse_ipv4(void *data, u64 nh_off, void *data_end) +{ + struct iphdr *iph = data + nh_off; + + if (iph + 1 > data_end) + return 0; + return iph->protocol; +} + +static int parse_ipv6(void *data, u64 nh_off, void *data_end) +{ + struct ipv6hdr *ip6h = data + nh_off; + + if (ip6h + 1 > data_end) + return 0; + return ip6h->nexthdr; +} + +SEC("xdp1") +int xdp_prog1(struct xdp_md *ctx) +{ + void *data_end = (void *)(long)ctx->data_end; + void *data = (void *)(long)ctx->data; + struct ethhdr *eth = data; + int rc = XDP_DROP; + long *value; + u16 h_proto; + u64 nh_off; + u32 index; + + nh_off = sizeof(*eth); + if (data + nh_off > data_end) + return rc; + + h_proto = eth->h_proto; + + if (h_proto == htons(ETH_P_8021Q) || h_proto == htons(ETH_P_8021AD)) { + struct vlan_hdr *vhdr; + + vhdr = data + nh_off; + nh_off += sizeof(struct vlan_hdr); + if (data + nh_off > data_end) + return rc; + h_proto = vhdr->h_vlan_encapsulated_proto; + } + if (h_proto == htons(ETH_P_8021Q) || h_proto == htons(ETH_P_8021AD)) { + struct vlan_hdr *vhdr; + + vhdr = data + nh_off; + nh_off += sizeof(struct vlan_hdr); + if (data + nh_off > data_end) + return rc; + h_proto = vhdr->h_vlan_encapsulated_proto; + } + + if (h_proto == htons(ETH_P_IP)) + index = parse_ipv4(data, nh_off, data_end); + else if (h_proto == htons(ETH_P_IPV6)) + index = parse_ipv6(data, nh_off, data_end); + else + index = 0; + + value = bpf_map_lookup_elem(&dropcnt, &index); + if (value) + *value += 1; + + if (index == 17) { + swap_src_dst_mac(data); + rc = XDP_TX; + } + + return rc; +} + +char _license[] SEC("license") = "GPL"; -- cgit v0.10.2 From cb6a115188500a448709df1f2d7698a4e1b7a099 Mon Sep 17 00:00:00 2001 From: "Reizer, Eyal" Date: Wed, 20 Jul 2016 07:30:20 +0000 Subject: wlcore: spi: fix build warning caused by redundant variable The ret variable is unused in wlcore_probe_of() Remove it for fixing build warning. Fixes: 01efe65aba65 ("wlcore: spi: add wl18xx support") Signed-off-by: Eyal Reizer Signed-off-by: Kalle Valo diff --git a/drivers/net/wireless/ti/wlcore/spi.c b/drivers/net/wireless/ti/wlcore/spi.c index 73fbcf1..6d24040 100644 --- a/drivers/net/wireless/ti/wlcore/spi.c +++ b/drivers/net/wireless/ti/wlcore/spi.c @@ -454,7 +454,6 @@ static int wlcore_probe_of(struct spi_device *spi, struct wl12xx_spi_glue *glue, struct wlcore_platdev_data *pdev_data) { struct device_node *dt_node = spi->dev.of_node; - int ret; const struct of_device_id *of_id; of_id = of_match_node(wlcore_spi_of_match_table, dt_node); -- cgit v0.10.2 From cc2e0b3fbcdd9667d7b7ecdf36d7b4d3647681d6 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Wed, 20 Jul 2016 07:55:52 -0700 Subject: bpf: fix implicit declaration of bpf_prog_add For the ifndef case of CONFIG_BPF_SYSCALL, an inline version of bpf_prog_add needs to exist otherwise the build breaks on some configs. drivers/net/ethernet/mellanox/mlx4/en_netdev.c:2544:10: error: implicit declaration of function 'bpf_prog_add' prog = bpf_prog_add(prog, priv->rx_ring_num - 1); The function is introduced in 59d3656d5bf50 ("bpf: add bpf_prog_add api for bulk prog refcnt") and first used in 47f1afdba2b87 ("net/mlx4_en: add support for fast rx drop bpf program"). Fixes: 47f1afdba2b87 ("net/mlx4_en: add support for fast rx drop bpf program") Reported-by: kbuild test robot Reported-by: Tariq Toukan Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 75a5ae6..36da074 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -289,6 +289,10 @@ static inline struct bpf_prog *bpf_prog_get_type(u32 ufd, { return ERR_PTR(-EOPNOTSUPP); } +static inline struct bpf_prog *bpf_prog_add(struct bpf_prog *prog, int i) +{ + return ERR_PTR(-EOPNOTSUPP); +} static inline void bpf_prog_put(struct bpf_prog *prog) { -- cgit v0.10.2 From b02b94b331be5097d248d49242bd1fc0649a8092 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 20 Jul 2016 20:17:47 +0200 Subject: bpf, elf: add official ELF machine define for eBPF Add the official BPF ELF e_machine value that was assigned recently [1,2] and will be propagated to glibc, et al. LLVM is switching to it in 3.9 release. [1] https://github.com/llvm-mirror/llvm/commit/36b9c09330bfb5e771914cfe307588f30d5510d2 [2] http://lists.iovisor.org/pipermail/iovisor-dev/2016-June/000266.html Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/uapi/linux/elf-em.h b/include/uapi/linux/elf-em.h index c3fdfe7..cb5d1a5 100644 --- a/include/uapi/linux/elf-em.h +++ b/include/uapi/linux/elf-em.h @@ -40,6 +40,7 @@ #define EM_TILEPRO 188 /* Tilera TILEPro */ #define EM_MICROBLAZE 189 /* Xilinx MicroBlaze */ #define EM_TILEGX 191 /* Tilera TILE-Gx */ +#define EM_BPF 247 /* Linux BPF - in-kernel virtual machine */ #define EM_FRV 0x5441 /* Fujitsu FR-V */ #define EM_AVR32 0x18ad /* Atmel AVR32 */ -- cgit v0.10.2 From 3ad7b1477ef9b01988ac052b02be9cd410c95157 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 19 Jul 2016 11:23:24 +0000 Subject: net: axienet: Fix return value check in axienet_probe() In case of error, the function of_parse_phandle() returns NULL pointer not ERR_PTR(). The IS_ERR() test in the return value check should be replaced with NULL test. Fixes: 46aa27df8853 ('net: axienet: Use devm_* calls') Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c index 9c82993..36ee7ab 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c @@ -1541,9 +1541,9 @@ static int axienet_probe(struct platform_device *pdev) /* Find the DMA node, map the DMA registers, and decode the DMA IRQs */ np = of_parse_phandle(pdev->dev.of_node, "axistream-connected", 0); - if (IS_ERR(np)) { + if (!np) { dev_err(&pdev->dev, "could not find DMA node\n"); - ret = PTR_ERR(np); + ret = -ENODEV; goto free_netdev; } ret = of_address_to_resource(np, 0, &dmares); -- cgit v0.10.2 From 9d5658e65c4dd33e87dcff979a7b68ecab600d5d Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 19 Jul 2016 11:25:03 +0000 Subject: wan/fsl_ucc_hdlc: remove .owner field for driver Remove .owner field if calls are used which set it automatically. Generated by: scripts/coccinelle/api/platform_no_drv_owner.cocci Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/wan/fsl_ucc_hdlc.c b/drivers/net/wan/fsl_ucc_hdlc.c index b3861bf..10ca497 100644 --- a/drivers/net/wan/fsl_ucc_hdlc.c +++ b/drivers/net/wan/fsl_ucc_hdlc.c @@ -1168,7 +1168,6 @@ static struct platform_driver ucc_hdlc_driver = { .probe = ucc_hdlc_probe, .remove = ucc_hdlc_remove, .driver = { - .owner = THIS_MODULE, .name = DRV_NAME, .pm = HDLC_PM_OPS, .of_match_table = fsl_ucc_hdlc_of_match, -- cgit v0.10.2 From 459421cc81eb4cdca96ef2c2b532359c6daa9002 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 19 Jul 2016 11:25:16 +0000 Subject: wan/fsl_ucc_hdlc: use module_platform_driver to simplify the code module_platform_driver() makes the code simpler by eliminating boilerplate code. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/wan/fsl_ucc_hdlc.c b/drivers/net/wan/fsl_ucc_hdlc.c index 10ca497..2fc50ec 100644 --- a/drivers/net/wan/fsl_ucc_hdlc.c +++ b/drivers/net/wan/fsl_ucc_hdlc.c @@ -1174,15 +1174,4 @@ static struct platform_driver ucc_hdlc_driver = { }, }; -static int __init ucc_hdlc_init(void) -{ - return platform_driver_register(&ucc_hdlc_driver); -} - -static void __exit ucc_hdlc_exit(void) -{ - platform_driver_unregister(&ucc_hdlc_driver); -} - -module_init(ucc_hdlc_init); -module_exit(ucc_hdlc_exit); +module_platform_driver(ucc_hdlc_driver); -- cgit v0.10.2 From 9a7bae8a12140c164082100a917ed5f856afccfc Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 19 Jul 2016 11:33:10 +0000 Subject: net: ethernet: nb8800: fix error handling of nb8800_probe() In ops->reset() error handling case, clk_disable_unprepare() is missed before return from this function. Signed-off-by: Wei Yongjun Acked-by: Mans Rullgard Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/aurora/nb8800.c b/drivers/net/ethernet/aurora/nb8800.c index dc2c35d..0d3aa1d 100644 --- a/drivers/net/ethernet/aurora/nb8800.c +++ b/drivers/net/ethernet/aurora/nb8800.c @@ -1418,7 +1418,7 @@ static int nb8800_probe(struct platform_device *pdev) if (ops && ops->reset) { ret = ops->reset(dev); if (ret) - goto err_free_dev; + goto err_disable_clk; } bus = devm_mdiobus_alloc(&pdev->dev); -- cgit v0.10.2 From 44fafdaa757cf251aade6c071f772ddb4e8a9885 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 19 Jul 2016 11:35:46 +0000 Subject: net/mlx5: Use PTR_ERR_OR_ZERO() to simplify the code Use PTR_ERR_OR_ZERO rather than if(IS_ERR(...)) + PTR_ERR. Generated by: scripts/coccinelle/api/ptr_ret.cocci Signed-off-by: Wei Yongjun Acked-by: Leon Romanovsky Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 1a377b4..75bb8c8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -1740,10 +1740,7 @@ static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering) /* create 1 prio*/ prio = fs_create_prio(&steering->esw_egress_root_ns->ns, 0, MLX5_TOTAL_VPORTS(steering->dev)); - if (IS_ERR(prio)) - return PTR_ERR(prio); - else - return 0; + return PTR_ERR_OR_ZERO(prio); } static int init_egress_acl_root_ns(struct mlx5_flow_steering *steering) @@ -1757,10 +1754,7 @@ static int init_egress_acl_root_ns(struct mlx5_flow_steering *steering) /* create 1 prio*/ prio = fs_create_prio(&steering->esw_ingress_root_ns->ns, 0, MLX5_TOTAL_VPORTS(steering->dev)); - if (IS_ERR(prio)) - return PTR_ERR(prio); - else - return 0; + return PTR_ERR_OR_ZERO(prio); } int mlx5_init_fs(struct mlx5_core_dev *dev) -- cgit v0.10.2 From 09714275b1ce7877f318bbd6321da1def95697d7 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 19 Jul 2016 12:37:53 +0000 Subject: net: cpmac: fix error handling of cpmac_probe() Add the missing free_netdev() before return from function cpmac_probe() in the error handling case. This patch revert commit 0465be8f4f1d ("net: cpmac: fix in releasing resources"), which changed to only free_netdev while register_netdev failed. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/cpmac.c b/drivers/net/ethernet/ti/cpmac.c index 29f381b..d300d53 100644 --- a/drivers/net/ethernet/ti/cpmac.c +++ b/drivers/net/ethernet/ti/cpmac.c @@ -1119,7 +1119,7 @@ static int cpmac_probe(struct platform_device *pdev) mem = platform_get_resource_byname(pdev, IORESOURCE_MEM, "regs"); if (!mem) { rc = -ENODEV; - goto out; + goto fail; } dev->irq = platform_get_irq_byname(pdev, "irq"); @@ -1147,7 +1147,7 @@ static int cpmac_probe(struct platform_device *pdev) dev_err(&pdev->dev, "Could not attach to PHY\n"); rc = PTR_ERR(phydev); - goto out; + goto fail; } rc = register_netdev(dev); @@ -1166,7 +1166,6 @@ static int cpmac_probe(struct platform_device *pdev) fail: free_netdev(dev); -out: return rc; } -- cgit v0.10.2 From 7ed674bc3cb10d5cd06a52365a6d3f54827cc4cb Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 19 Jul 2016 15:35:53 +0200 Subject: mlxsw: spectrum: Expose per-priority counters via ethtool Expose per-priority bytes / packets / PFC packets counters via ethtool. These counters are very useful when debugging QoS functionality and provide a better insight into the device's forwarding plane. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index c812513..784a2cd 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -899,7 +899,7 @@ struct mlxsw_sp_port_hw_stats { u64 (*getter)(char *payload); }; -static const struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_stats[] = { +static struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_stats[] = { { .str = "a_frames_transmitted_ok", .getter = mlxsw_reg_ppcnt_a_frames_transmitted_ok_get, @@ -980,6 +980,58 @@ static const struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_stats[] = { #define MLXSW_SP_PORT_HW_STATS_LEN ARRAY_SIZE(mlxsw_sp_port_hw_stats) +static struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_prio_stats[] = { + { + .str = "rx_octets_prio", + .getter = mlxsw_reg_ppcnt_rx_octets_get, + }, + { + .str = "rx_frames_prio", + .getter = mlxsw_reg_ppcnt_rx_frames_get, + }, + { + .str = "tx_octets_prio", + .getter = mlxsw_reg_ppcnt_tx_octets_get, + }, + { + .str = "tx_frames_prio", + .getter = mlxsw_reg_ppcnt_tx_frames_get, + }, + { + .str = "rx_pause_prio", + .getter = mlxsw_reg_ppcnt_rx_pause_get, + }, + { + .str = "rx_pause_duration_prio", + .getter = mlxsw_reg_ppcnt_rx_pause_duration_get, + }, + { + .str = "tx_pause_prio", + .getter = mlxsw_reg_ppcnt_tx_pause_get, + }, + { + .str = "tx_pause_duration_prio", + .getter = mlxsw_reg_ppcnt_tx_pause_duration_get, + }, +}; + +#define MLXSW_SP_PORT_HW_PRIO_STATS_LEN ARRAY_SIZE(mlxsw_sp_port_hw_prio_stats) + +#define MLXSW_SP_PORT_ETHTOOL_STATS_LEN (MLXSW_SP_PORT_HW_STATS_LEN + \ + MLXSW_SP_PORT_HW_PRIO_STATS_LEN * \ + IEEE_8021QAZ_MAX_TCS) + +static void mlxsw_sp_port_get_prio_strings(u8 **p, int prio) +{ + int i; + + for (i = 0; i < MLXSW_SP_PORT_HW_PRIO_STATS_LEN; i++) { + snprintf(*p, ETH_GSTRING_LEN, "%s_%d", + mlxsw_sp_port_hw_prio_stats[i].str, prio); + *p += ETH_GSTRING_LEN; + } +} + static void mlxsw_sp_port_get_strings(struct net_device *dev, u32 stringset, u8 *data) { @@ -993,6 +1045,10 @@ static void mlxsw_sp_port_get_strings(struct net_device *dev, ETH_GSTRING_LEN); p += ETH_GSTRING_LEN; } + + for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) + mlxsw_sp_port_get_prio_strings(&p, i); + break; } } @@ -1020,27 +1076,69 @@ static int mlxsw_sp_port_set_phys_id(struct net_device *dev, return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mlcr), mlcr_pl); } -static void mlxsw_sp_port_get_stats(struct net_device *dev, - struct ethtool_stats *stats, u64 *data) +static int +mlxsw_sp_get_hw_stats_by_group(struct mlxsw_sp_port_hw_stats **p_hw_stats, + int *p_len, enum mlxsw_reg_ppcnt_grp grp) +{ + switch (grp) { + case MLXSW_REG_PPCNT_IEEE_8023_CNT: + *p_hw_stats = mlxsw_sp_port_hw_stats; + *p_len = MLXSW_SP_PORT_HW_STATS_LEN; + break; + case MLXSW_REG_PPCNT_PRIO_CNT: + *p_hw_stats = mlxsw_sp_port_hw_prio_stats; + *p_len = MLXSW_SP_PORT_HW_PRIO_STATS_LEN; + break; + default: + WARN_ON(1); + return -ENOTSUPP; + } + return 0; +} + +static void __mlxsw_sp_port_get_stats(struct net_device *dev, + enum mlxsw_reg_ppcnt_grp grp, int prio, + u64 *data, int data_index) { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_port_hw_stats *hw_stats; char ppcnt_pl[MLXSW_REG_PPCNT_LEN]; - int i; + int i, len; int err; - mlxsw_reg_ppcnt_pack(ppcnt_pl, mlxsw_sp_port->local_port, - MLXSW_REG_PPCNT_IEEE_8023_CNT, 0); + err = mlxsw_sp_get_hw_stats_by_group(&hw_stats, &len, grp); + if (err) + return; + mlxsw_reg_ppcnt_pack(ppcnt_pl, mlxsw_sp_port->local_port, grp, prio); err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ppcnt), ppcnt_pl); - for (i = 0; i < MLXSW_SP_PORT_HW_STATS_LEN; i++) - data[i] = !err ? mlxsw_sp_port_hw_stats[i].getter(ppcnt_pl) : 0; + for (i = 0; i < len; i++) + data[data_index + i] = !err ? hw_stats[i].getter(ppcnt_pl) : 0; +} + +static void mlxsw_sp_port_get_stats(struct net_device *dev, + struct ethtool_stats *stats, u64 *data) +{ + int i, data_index = 0; + + /* IEEE 802.3 Counters */ + __mlxsw_sp_port_get_stats(dev, MLXSW_REG_PPCNT_IEEE_8023_CNT, 0, + data, data_index); + data_index = MLXSW_SP_PORT_HW_STATS_LEN; + + /* Per-Priority Counters */ + for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { + __mlxsw_sp_port_get_stats(dev, MLXSW_REG_PPCNT_PRIO_CNT, i, + data, data_index); + data_index += MLXSW_SP_PORT_HW_PRIO_STATS_LEN; + } } static int mlxsw_sp_port_get_sset_count(struct net_device *dev, int sset) { switch (sset) { case ETH_SS_STATS: - return MLXSW_SP_PORT_HW_STATS_LEN; + return MLXSW_SP_PORT_ETHTOOL_STATS_LEN; default: return -EOPNOTSUPP; } -- cgit v0.10.2 From df4750e84e95acfbe243c174a1e3eff54377b506 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 19 Jul 2016 15:35:54 +0200 Subject: mlxsw: spectrum: Expose per-tc counters via ethtool Expose the transmit queue length of each traffic class and the amount of unicast packets discarded due to insufficient room in the shared buffer. The first counter allows us to debug user priority to traffic class mapping, whereas the drop counter is useful when determining shared buffer configuration. Signed-off-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 0cc1485..b669b04 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -2503,6 +2503,7 @@ MLXSW_ITEM32(reg, ppcnt, pnat, 0x00, 14, 2); enum mlxsw_reg_ppcnt_grp { MLXSW_REG_PPCNT_IEEE_8023_CNT = 0x0, MLXSW_REG_PPCNT_PRIO_CNT = 0x10, + MLXSW_REG_PPCNT_TC_CNT = 0x11, }; /* reg_ppcnt_grp @@ -2703,6 +2704,23 @@ MLXSW_ITEM64(reg, ppcnt, tx_pause_duration, 0x08 + 0x68, 0, 64); */ MLXSW_ITEM64(reg, ppcnt, tx_pause_transition, 0x08 + 0x70, 0, 64); +/* Ethernet Per Traffic Group Counters */ + +/* reg_ppcnt_tc_transmit_queue + * Contains the transmit queue depth in cells of traffic class + * selected by prio_tc and the port selected by local_port. + * The field cannot be cleared. + * Access: RO + */ +MLXSW_ITEM64(reg, ppcnt, tc_transmit_queue, 0x08 + 0x00, 0, 64); + +/* reg_ppcnt_tc_no_buffer_discard_uc + * The number of unicast packets dropped due to lack of shared + * buffer resources. + * Access: RO + */ +MLXSW_ITEM64(reg, ppcnt, tc_no_buffer_discard_uc, 0x08 + 0x08, 0, 64); + static inline void mlxsw_reg_ppcnt_pack(char *payload, u8 local_port, enum mlxsw_reg_ppcnt_grp grp, u8 prio_tc) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 784a2cd..3e7920b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -1017,8 +1017,29 @@ static struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_prio_stats[] = { #define MLXSW_SP_PORT_HW_PRIO_STATS_LEN ARRAY_SIZE(mlxsw_sp_port_hw_prio_stats) +static u64 mlxsw_reg_ppcnt_tc_transmit_queue_bytes_get(char *ppcnt_pl) +{ + u64 transmit_queue = mlxsw_reg_ppcnt_tc_transmit_queue_get(ppcnt_pl); + + return MLXSW_SP_CELLS_TO_BYTES(transmit_queue); +} + +static struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_tc_stats[] = { + { + .str = "tc_transmit_queue_tc", + .getter = mlxsw_reg_ppcnt_tc_transmit_queue_bytes_get, + }, + { + .str = "tc_no_buffer_discard_uc_tc", + .getter = mlxsw_reg_ppcnt_tc_no_buffer_discard_uc_get, + }, +}; + +#define MLXSW_SP_PORT_HW_TC_STATS_LEN ARRAY_SIZE(mlxsw_sp_port_hw_tc_stats) + #define MLXSW_SP_PORT_ETHTOOL_STATS_LEN (MLXSW_SP_PORT_HW_STATS_LEN + \ - MLXSW_SP_PORT_HW_PRIO_STATS_LEN * \ + (MLXSW_SP_PORT_HW_PRIO_STATS_LEN + \ + MLXSW_SP_PORT_HW_TC_STATS_LEN) * \ IEEE_8021QAZ_MAX_TCS) static void mlxsw_sp_port_get_prio_strings(u8 **p, int prio) @@ -1032,6 +1053,17 @@ static void mlxsw_sp_port_get_prio_strings(u8 **p, int prio) } } +static void mlxsw_sp_port_get_tc_strings(u8 **p, int tc) +{ + int i; + + for (i = 0; i < MLXSW_SP_PORT_HW_TC_STATS_LEN; i++) { + snprintf(*p, ETH_GSTRING_LEN, "%s_%d", + mlxsw_sp_port_hw_tc_stats[i].str, tc); + *p += ETH_GSTRING_LEN; + } +} + static void mlxsw_sp_port_get_strings(struct net_device *dev, u32 stringset, u8 *data) { @@ -1049,6 +1081,9 @@ static void mlxsw_sp_port_get_strings(struct net_device *dev, for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) mlxsw_sp_port_get_prio_strings(&p, i); + for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) + mlxsw_sp_port_get_tc_strings(&p, i); + break; } } @@ -1089,6 +1124,10 @@ mlxsw_sp_get_hw_stats_by_group(struct mlxsw_sp_port_hw_stats **p_hw_stats, *p_hw_stats = mlxsw_sp_port_hw_prio_stats; *p_len = MLXSW_SP_PORT_HW_PRIO_STATS_LEN; break; + case MLXSW_REG_PPCNT_TC_CNT: + *p_hw_stats = mlxsw_sp_port_hw_tc_stats; + *p_len = MLXSW_SP_PORT_HW_TC_STATS_LEN; + break; default: WARN_ON(1); return -ENOTSUPP; @@ -1132,6 +1171,13 @@ static void mlxsw_sp_port_get_stats(struct net_device *dev, data, data_index); data_index += MLXSW_SP_PORT_HW_PRIO_STATS_LEN; } + + /* Per-TC Counters */ + for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) { + __mlxsw_sp_port_get_stats(dev, MLXSW_REG_PPCNT_TC_CNT, i, + data, data_index); + data_index += MLXSW_SP_PORT_HW_TC_STATS_LEN; + } } static int mlxsw_sp_port_get_sset_count(struct net_device *dev, int sset) -- cgit v0.10.2 From 1b00c6c064302354fce71d7c363945fe9e967f7c Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Fri, 3 Jun 2016 15:42:11 -0700 Subject: fm10k: no need to continue in fm10k_down if __FM10K_DOWN already set Return early from fm10k_down() when we are already down, since that means another thread is either already finished or has started going down, so shouldn't conflict with them. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index e05aca9..610c313 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -1601,7 +1601,8 @@ void fm10k_down(struct fm10k_intfc *interface) int err; /* signal that we are down to the interrupt handler and service task */ - set_bit(__FM10K_DOWN, &interface->state); + if (test_and_set_bit(__FM10K_DOWN, &interface->state)) + return; /* call carrier off first to avoid false dev_watchdog timeouts */ netif_carrier_off(netdev); -- cgit v0.10.2 From b624714bc90064eeefd9ba7564e90865eef00421 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Fri, 3 Jun 2016 15:42:12 -0700 Subject: fm10k: avoid possible null pointer dereference in fm10k_update_stats It's currently possible for fm10k_update_stats to be called during the window when we go down and the rings are removed. This can result in a null pointer dereference. In fm10k_get_stats64 we work around this by using ACCESS_ONCE and a null pointer check inside the loop. Use this same flow in the fm10k_update_stats to avoid the potential null pointer. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index 610c313..be0b7de 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -377,7 +377,10 @@ void fm10k_update_stats(struct fm10k_intfc *interface) /* gather some stats to the interface struct that are per queue */ for (bytes = 0, pkts = 0, i = 0; i < interface->num_tx_queues; i++) { - struct fm10k_ring *tx_ring = interface->tx_ring[i]; + struct fm10k_ring *tx_ring = READ_ONCE(interface->tx_ring[i]); + + if (!tx_ring) + continue; restart_queue += tx_ring->tx_stats.restart_queue; tx_busy += tx_ring->tx_stats.tx_busy; @@ -396,7 +399,10 @@ void fm10k_update_stats(struct fm10k_intfc *interface) /* gather some stats to the interface struct that are per queue */ for (bytes = 0, pkts = 0, i = 0; i < interface->num_rx_queues; i++) { - struct fm10k_ring *rx_ring = interface->rx_ring[i]; + struct fm10k_ring *rx_ring = READ_ONCE(interface->rx_ring[i]); + + if (!rx_ring) + continue; bytes += rx_ring->stats.bytes; pkts += rx_ring->stats.packets; -- cgit v0.10.2 From 9d73edee5900baff47887b36e9b97f45a7d13b6d Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:45 -0700 Subject: fm10k: prevent multiple threads updating statistics Also prevent updating stats while the interface is down. If we're already updating stats, just return doing nothing. When we take the device down, block stat updates until we come back up. This ensures that we avoid tearing down rings when we're updating statistics, and prevents updating statistics until we're up. We can't re-use the __FM10K_DOWN for this because it wouldn't prevent multiple threads from accessing statistics. Neither does it prevent the case where we start updating stats and then start going down in another thread. The fm10k_get_stats64 is except from this, because it has a completely different flow which does not suffer from the same issues as fm10k_update_stats might. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k.h b/drivers/net/ethernet/intel/fm10k/fm10k.h index e98b86b..c8d0817 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k.h +++ b/drivers/net/ethernet/intel/fm10k/fm10k.h @@ -362,6 +362,7 @@ enum fm10k_state_t { __FM10K_SERVICE_DISABLE, __FM10K_MBX_LOCK, __FM10K_LINK_DOWN, + __FM10K_UPDATING_STATS, }; static inline void fm10k_mbx_lock(struct fm10k_intfc *interface) diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index be0b7de..4439376 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -372,6 +372,10 @@ void fm10k_update_stats(struct fm10k_intfc *interface) u64 bytes, pkts; int i; + /* ensure only one thread updates stats at a time */ + if (test_and_set_bit(__FM10K_UPDATING_STATS, &interface->state)) + return; + /* do not allow stats update via service task for next second */ interface->next_stats_update = jiffies + HZ; @@ -449,6 +453,8 @@ void fm10k_update_stats(struct fm10k_intfc *interface) /* Fill out the OS statistics structure */ net_stats->rx_errors = rx_errors; net_stats->rx_dropped = interface->stats.nodesc_drop.count; + + clear_bit(__FM10K_UPDATING_STATS, &interface->state); } /** @@ -1572,6 +1578,9 @@ void fm10k_up(struct fm10k_intfc *interface) /* configure interrupts */ hw->mac.ops.update_int_moderator(hw); + /* enable statistics capture again */ + clear_bit(__FM10K_UPDATING_STATS, &interface->state); + /* clear down bit to indicate we are ready to go */ clear_bit(__FM10K_DOWN, &interface->state); @@ -1629,6 +1638,10 @@ void fm10k_down(struct fm10k_intfc *interface) /* capture stats one last time before stopping interface */ fm10k_update_stats(interface); + /* prevent updating statistics while we're down */ + while (test_and_set_bit(__FM10K_UPDATING_STATS, &interface->state)) + usleep_range(1000, 2000); + /* Disable DMA engine for Tx/Rx */ err = hw->mac.ops.stop_hw(hw); if (err) @@ -1757,6 +1770,7 @@ static int fm10k_sw_init(struct fm10k_intfc *interface, /* Start off interface as being down */ set_bit(__FM10K_DOWN, &interface->state); + set_bit(__FM10K_UPDATING_STATS, &interface->state); return 0; } -- cgit v0.10.2 From 5e93cbadd3e9db342e48582178b7d241097f51c4 Mon Sep 17 00:00:00 2001 From: Ngai-Mint Kwan Date: Tue, 7 Jun 2016 16:08:46 -0700 Subject: fm10k: Reset mailbox global interrupts When a data path reset is initiated, write control to the PCIE_GMBX is yanked from the switch manager. The switch manager writes to this register to clear mailbox global interrupt bits as part of its mailbox interrupt handling routine. When the device recovers from the data path reset and these bits are not cleared, it will prevent future mailbox global interrupts from being triggered. Upon confirming that the device has exited from a data path reset, clear these bits to ensure the proper functioning of the mailbox global interrupt. Signed-off-by: Ngai-Mint Kwan Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h b/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h index b7dbc8a..35c1dba 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h +++ b/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h @@ -41,6 +41,8 @@ struct fm10k_mbx_info; #define FM10K_MBX_ACK_INTERRUPT 0x00000010 #define FM10K_MBX_INTERRUPT_ENABLE 0x00000020 #define FM10K_MBX_INTERRUPT_DISABLE 0x00000040 +#define FM10K_MBX_GLOBAL_REQ_INTERRUPT 0x00000200 +#define FM10K_MBX_GLOBAL_ACK_INTERRUPT 0x00000400 #define FM10K_MBICR(_n) ((_n) + 0x18840) #define FM10K_GMBX 0x18842 diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pf.c b/drivers/net/ethernet/intel/fm10k/fm10k_pf.c index dc75507..69e2c82 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pf.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pf.c @@ -77,6 +77,10 @@ static s32 fm10k_reset_hw_pf(struct fm10k_hw *hw) if (!(reg & FM10K_IP_NOTINRESET)) err = FM10K_ERR_RESET_FAILED; + /* Reset mailbox global interrupts */ + reg = FM10K_MBX_GLOBAL_REQ_INTERRUPT | FM10K_MBX_GLOBAL_ACK_INTERRUPT; + fm10k_write_reg(hw, FM10K_GMBX, reg); + out: return err; } -- cgit v0.10.2 From ce33624f37f43267e5fa0810a058807d268142fb Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:47 -0700 Subject: fm10k: don't stop reset due to FM10K_ERR_REQUESTS_PENDING Don't report FM10K_ERR_REQUESTS_PENDING when we fail to disable queues within the timeout. This can occur due to a hardware Tx hang, or when the switch ethernet fabric is resetting while we are transmitting traffic. It can sometimes take up to 500ms before the Tx DMA engine gives up. Instead, just skip the DMA engine check and perform a data-path reset anyways. Add a statistic counter to keep track of the number of resets occurring while we have pending DMA on the rings. In order to prevent having to re-assign err to 0, re-order the last few items of the reset_hw_pf function so that we don't perform "return err" at the end. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c index 9b51954..c04cbe9 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c @@ -76,6 +76,8 @@ static const struct fm10k_stats fm10k_gstrings_global_stats[] = { FM10K_STAT("mac_rules_used", hw.swapi.mac.used), FM10K_STAT("mac_rules_avail", hw.swapi.mac.avail), + FM10K_STAT("reset_while_pending", hw.mac.reset_while_pending), + FM10K_STAT("tx_hang_count", tx_timeout_count), }; diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pf.c b/drivers/net/ethernet/intel/fm10k/fm10k_pf.c index 69e2c82..7fbd94b 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pf.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pf.c @@ -51,8 +51,12 @@ static s32 fm10k_reset_hw_pf(struct fm10k_hw *hw) /* shut down all rings */ err = fm10k_disable_queues_generic(hw, FM10K_MAX_QUEUES); - if (err) + if (err == FM10K_ERR_REQUESTS_PENDING) { + hw->mac.reset_while_pending++; + goto force_reset; + } else if (err) { return err; + } /* Verify that DMA is no longer active */ reg = fm10k_read_reg(hw, FM10K_DMA_CTRL); @@ -62,27 +66,27 @@ static s32 fm10k_reset_hw_pf(struct fm10k_hw *hw) /* verify the switch is ready for reset */ reg = fm10k_read_reg(hw, FM10K_DMA_CTRL2); if (!(reg & FM10K_DMA_CTRL2_SWITCH_READY)) - goto out; + return FM10K_ERR_DMA_PENDING; +force_reset: /* Inititate data path reset */ - reg |= FM10K_DMA_CTRL_DATAPATH_RESET; + reg = FM10K_DMA_CTRL_DATAPATH_RESET; fm10k_write_reg(hw, FM10K_DMA_CTRL, reg); /* Flush write and allow 100us for reset to complete */ fm10k_write_flush(hw); udelay(FM10K_RESET_TIMEOUT); - /* Verify we made it out of reset */ - reg = fm10k_read_reg(hw, FM10K_IP); - if (!(reg & FM10K_IP_NOTINRESET)) - err = FM10K_ERR_RESET_FAILED; - /* Reset mailbox global interrupts */ reg = FM10K_MBX_GLOBAL_REQ_INTERRUPT | FM10K_MBX_GLOBAL_ACK_INTERRUPT; fm10k_write_reg(hw, FM10K_GMBX, reg); -out: - return err; + /* Verify we made it out of reset */ + reg = fm10k_read_reg(hw, FM10K_IP); + if (!(reg & FM10K_IP_NOTINRESET)) + return FM10K_ERR_RESET_FAILED; + + return 0; } /** diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_type.h b/drivers/net/ethernet/intel/fm10k/fm10k_type.h index b8bc061..1d65ad8 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_type.h +++ b/drivers/net/ethernet/intel/fm10k/fm10k_type.h @@ -562,6 +562,7 @@ struct fm10k_mac_info { bool tx_ready; u32 dglort_map; u8 itr_scale; + u64 reset_while_pending; }; struct fm10k_swapi_table_info { diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_vf.c b/drivers/net/ethernet/intel/fm10k/fm10k_vf.c index 3b06685e..337ba65 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_vf.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_vf.c @@ -34,7 +34,7 @@ static s32 fm10k_stop_hw_vf(struct fm10k_hw *hw) /* we need to disable the queues before taking further steps */ err = fm10k_stop_hw_generic(hw); - if (err) + if (err && err != FM10K_ERR_REQUESTS_PENDING) return err; /* If permanent address is set then we need to restore it */ @@ -67,7 +67,7 @@ static s32 fm10k_stop_hw_vf(struct fm10k_hw *hw) fm10k_write_reg(hw, FM10K_TDLEN(i), tdlen); } - return 0; + return err; } /** @@ -83,7 +83,9 @@ static s32 fm10k_reset_hw_vf(struct fm10k_hw *hw) /* shut down queues we own and reset DMA configuration */ err = fm10k_stop_hw_vf(hw); - if (err) + if (err == FM10K_ERR_REQUESTS_PENDING) + hw->mac.reset_while_pending++; + else if (err) return err; /* Inititate VF reset */ @@ -96,9 +98,9 @@ static s32 fm10k_reset_hw_vf(struct fm10k_hw *hw) /* Clear reset bit and verify it was cleared */ fm10k_write_reg(hw, FM10K_VFCTRL, 0); if (fm10k_read_reg(hw, FM10K_VFCTRL) & FM10K_VFCTRL_RST) - err = FM10K_ERR_RESET_FAILED; + return FM10K_ERR_RESET_FAILED; - return err; + return 0; } /** -- cgit v0.10.2 From 892c9e0872bea23ac2dc774ea950e6526f157bd5 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:48 -0700 Subject: fm10k: perform data path reset even when switch is not ready A while ago, an additional check for the switch being ready was added to reset_hw. A recent refactor accidentally made this check return an error code on failure which caused fm10k_probe to fail when the switch wasn't brought up first. The original reasoning for the check was to prevent additional data path reset when the fabric wasn't ready yet. However, there isn't a compelling reason to keep the check, as the data path reset will restore hardware to a known good state. Remove the check and perform the data path reset regardless of the switch manager state. An alternative fix is to return FM10K_SUCCESS instead, and bypass the actual data path reset. This should be fine as we will perform a reset_hw once the switch is active. However, since data path reset will reset many parts of the hardware it seems better to just perform the reset regardless of switch state. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pf.c b/drivers/net/ethernet/intel/fm10k/fm10k_pf.c index 7fbd94b..23f3566 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pf.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pf.c @@ -63,11 +63,6 @@ static s32 fm10k_reset_hw_pf(struct fm10k_hw *hw) if (reg & (FM10K_DMA_CTRL_TX_ACTIVE | FM10K_DMA_CTRL_RX_ACTIVE)) return FM10K_ERR_DMA_PENDING; - /* verify the switch is ready for reset */ - reg = fm10k_read_reg(hw, FM10K_DMA_CTRL2); - if (!(reg & FM10K_DMA_CTRL2_SWITCH_READY)) - return FM10K_ERR_DMA_PENDING; - force_reset: /* Inititate data path reset */ reg = FM10K_DMA_CTRL_DATAPATH_RESET; -- cgit v0.10.2 From 34bad71c7c25840cd78dff12d75497d3aab5544d Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:49 -0700 Subject: fm10k: use actual hardware registers when checking for pending Tx Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c index a9ccc1e..c6a4645 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c @@ -1130,9 +1130,11 @@ static u64 fm10k_get_tx_completed(struct fm10k_ring *ring) static u64 fm10k_get_tx_pending(struct fm10k_ring *ring) { - /* use SW head and tail until we have real hardware */ - u32 head = ring->next_to_clean; - u32 tail = ring->next_to_use; + struct fm10k_intfc *interface = ring->q_vector->interface; + struct fm10k_hw *hw = &interface->hw; + + u32 head = fm10k_read_reg(hw, FM10K_TDH(ring->reg_idx)); + u32 tail = fm10k_read_reg(hw, FM10K_TDT(ring->reg_idx)); return ((head <= tail) ? tail : tail + ring->count) - head; } -- cgit v0.10.2 From 106ca42356b49a1ae6199e6630ec40df82ff7421 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:50 -0700 Subject: fm10k: only warn when stop_hw fails with FM10K_ERR_REQUESTS_PENDING When stop_hw() routine fails with FM10K_ERR_REQUESTS_PENDING, this indicates that the Tx or Rx queues did not shutdown within the time limit. Print a more suitable message at the dev_info level instead of dev_err. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index 4439376..4dfd128 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -1644,7 +1644,10 @@ void fm10k_down(struct fm10k_intfc *interface) /* Disable DMA engine for Tx/Rx */ err = hw->mac.ops.stop_hw(hw); - if (err) + if (err == FM10K_ERR_REQUESTS_PENDING) + dev_info(&interface->pdev->dev, + "due to pending requests hw was not shut down gracefully\n"); + else if (err) dev_err(&interface->pdev->dev, "stop_hw failed: %d\n", err); /* free any buffers still on the rings */ -- cgit v0.10.2 From 94877768cfaa99f7b3757f29632faa5945f18872 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:51 -0700 Subject: fm10k: wait for queues to drain if stop_hw() fails once It turns out that sometimes during a reset the Tx queues will be temporarily stuck longer than .stop_hw() expects. Work around this issue by attempting to .stop_hw() first. If it tails, wait a number of attempts until the Tx queues appear to be drained. After this, attempt stop_hw() again. This ensures that we avoid waiting if we don't need to, such as during the first initialization of a VF, and give the proper amount of time necessary to recover from most situations. It is possible that the hardware is actually stuck. For PFs, this is usually fixed by a datapath reset. Unfortunately the VF cannot request a similar reset for itself. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k.h b/drivers/net/ethernet/intel/fm10k/fm10k.h index c8d0817..c4cf08d 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k.h +++ b/drivers/net/ethernet/intel/fm10k/fm10k.h @@ -458,6 +458,7 @@ __be16 fm10k_tx_encap_offload(struct sk_buff *skb); netdev_tx_t fm10k_xmit_frame_ring(struct sk_buff *skb, struct fm10k_ring *tx_ring); void fm10k_tx_timeout_reset(struct fm10k_intfc *interface); +u64 fm10k_get_tx_pending(struct fm10k_ring *ring); bool fm10k_check_tx_hang(struct fm10k_ring *tx_ring); void fm10k_alloc_rx_buffers(struct fm10k_ring *rx_ring, u16 cleaned_count); diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c index c6a4645..c85fc9894 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c @@ -1128,7 +1128,7 @@ static u64 fm10k_get_tx_completed(struct fm10k_ring *ring) return ring->stats.packets; } -static u64 fm10k_get_tx_pending(struct fm10k_ring *ring) +u64 fm10k_get_tx_pending(struct fm10k_ring *ring) { struct fm10k_intfc *interface = ring->q_vector->interface; struct fm10k_hw *hw = &interface->hw; diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index 4dfd128..7c9b20c 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -1613,7 +1613,7 @@ void fm10k_down(struct fm10k_intfc *interface) { struct net_device *netdev = interface->netdev; struct fm10k_hw *hw = &interface->hw; - int err; + int err, i = 0, count = 0; /* signal that we are down to the interrupt handler and service task */ if (test_and_set_bit(__FM10K_DOWN, &interface->state)) @@ -1629,9 +1629,6 @@ void fm10k_down(struct fm10k_intfc *interface) /* reset Rx filters */ fm10k_reset_rx_state(interface); - /* allow 10ms for device to quiesce */ - usleep_range(10000, 20000); - /* disable polling routines */ fm10k_napi_disable_all(interface); @@ -1642,11 +1639,46 @@ void fm10k_down(struct fm10k_intfc *interface) while (test_and_set_bit(__FM10K_UPDATING_STATS, &interface->state)) usleep_range(1000, 2000); + /* skip waiting for TX DMA if we lost PCIe link */ + if (FM10K_REMOVED(hw->hw_addr)) + goto skip_tx_dma_drain; + + /* In some rare circumstances it can take a while for Tx queues to + * quiesce and be fully disabled. Attempt to .stop_hw() first, and + * then if we get ERR_REQUESTS_PENDING, go ahead and wait in a loop + * until the Tx queues have emptied, or until a number of retries. If + * we fail to clear within the retry loop, we will issue a warning + * indicating that Tx DMA is probably hung. Note this means we call + * .stop_hw() twice but this shouldn't cause any problems. + */ + err = hw->mac.ops.stop_hw(hw); + if (err != FM10K_ERR_REQUESTS_PENDING) + goto skip_tx_dma_drain; + +#define TX_DMA_DRAIN_RETRIES 25 + for (count = 0; count < TX_DMA_DRAIN_RETRIES; count++) { + usleep_range(10000, 20000); + + /* start checking at the last ring to have pending Tx */ + for (; i < interface->num_tx_queues; i++) + if (fm10k_get_tx_pending(interface->tx_ring[i])) + break; + + /* if all the queues are drained, we can break now */ + if (i == interface->num_tx_queues) + break; + } + + if (count >= TX_DMA_DRAIN_RETRIES) + dev_err(&interface->pdev->dev, + "Tx queues failed to drain after %d tries. Tx DMA is probably hung.\n", + count); +skip_tx_dma_drain: /* Disable DMA engine for Tx/Rx */ err = hw->mac.ops.stop_hw(hw); if (err == FM10K_ERR_REQUESTS_PENDING) - dev_info(&interface->pdev->dev, - "due to pending requests hw was not shut down gracefully\n"); + dev_err(&interface->pdev->dev, + "due to pending requests hw was not shut down gracefully\n"); else if (err) dev_err(&interface->pdev->dev, "stop_hw failed: %d\n", err); -- cgit v0.10.2 From 40de1fad417cb7c6958c214ec1f4c8c29101cb7f Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:52 -0700 Subject: fm10k: split fm10k_reinit into two functions There are several flows in the driver which perform the similar function of tearing down software and restoring software to recover from certain errors or PCIe events, including: * fm10k_reinit * fm10k_suspend/resume * fm10k_io_error_detected/fm10k_io_resume In addition, we want to implement a .reset_notify() handler as well which will also perform similar function. Rework how the driver codes reset and resume flows by separating out the reinit logic into two functions "fm10k_prepare_for_reset" and "fm10k_handle_reset". This first step will allow us to re-use this functionality in the similar blocks of code instead of re-coding the same sequence of events slightly different. The end result should be more maintainable and correct, fixing several inconsistencies with the work flow. The new functions expect to take the rtnl_lock() themselves, and it does have the unfortunate side effect of having the reinit flow take then release then take the rtnl_lock. However, this minor downside is out weighted by the benefits of code reduction and reducing needless difference between these flows. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index 7c9b20c..2963b41 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -136,11 +136,9 @@ static void fm10k_detach_subtask(struct fm10k_intfc *interface) rtnl_unlock(); } -static void fm10k_reinit(struct fm10k_intfc *interface) +static void fm10k_prepare_for_reset(struct fm10k_intfc *interface) { struct net_device *netdev = interface->netdev; - struct fm10k_hw *hw = &interface->hw; - int err; WARN_ON(in_interrupt()); @@ -165,6 +163,17 @@ static void fm10k_reinit(struct fm10k_intfc *interface) /* delay any future reset requests */ interface->last_reset = jiffies + (10 * HZ); + rtnl_unlock(); +} + +static int fm10k_handle_reset(struct fm10k_intfc *interface) +{ + struct net_device *netdev = interface->netdev; + struct fm10k_hw *hw = &interface->hw; + int err; + + rtnl_lock(); + /* reset and initialize the hardware so it is in a known state */ err = hw->mac.ops.reset_hw(hw); if (err) { @@ -185,7 +194,7 @@ static void fm10k_reinit(struct fm10k_intfc *interface) goto reinit_err; } - /* reassociate interrupts */ + /* re-associate interrupts */ err = fm10k_mbx_request_irq(interface); if (err) goto err_mbx_irq; @@ -219,7 +228,7 @@ static void fm10k_reinit(struct fm10k_intfc *interface) clear_bit(__FM10K_RESETTING, &interface->state); - return; + return err; err_open: fm10k_mbx_free_irq(interface); err_mbx_irq: @@ -230,6 +239,20 @@ reinit_err: rtnl_unlock(); clear_bit(__FM10K_RESETTING, &interface->state); + + return err; +} + +static void fm10k_reinit(struct fm10k_intfc *interface) +{ + int err; + + fm10k_prepare_for_reset(interface); + + err = fm10k_handle_reset(interface); + if (err) + dev_err(&interface->pdev->dev, + "fm10k_handle_reset failed: %d\n", err); } static void fm10k_reset_subtask(struct fm10k_intfc *interface) -- cgit v0.10.2 From dc4b76c0fe1753940b15413ae5fb9829552824ec Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:53 -0700 Subject: fm10k: implement prepare_suspend and handle_resume Implement fm10k_prepare_suspend and fm10k_handle_resume functions which abstract around the now existing fm10k_prepare_for_reset and fm10k_handle_reset. The new functions also handle stopping the service task, which is something that the original re-init flow does not need. Every other location that does a suspend/resume type flow is expected to use these functions, because otherwise they may have conflicts with the running watchdog routines. This also has the effect of preventing possible surprise remove events during handling of FLR events and PCIe errors. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index 2963b41..a6ee046 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -2112,6 +2112,44 @@ static void fm10k_remove(struct pci_dev *pdev) pci_disable_device(pdev); } +static void fm10k_prepare_suspend(struct fm10k_intfc *interface) +{ + /* the watchdog task reads from registers, which might appear like + * a surprise remove if the PCIe device is disabled while we're + * stopped. We stop the watchdog task until after we resume software + * activity. + */ + set_bit(__FM10K_SERVICE_DISABLE, &interface->state); + cancel_work_sync(&interface->service_task); + + fm10k_prepare_for_reset(interface); +} + +static int fm10k_handle_resume(struct fm10k_intfc *interface) +{ + struct fm10k_hw *hw = &interface->hw; + int err; + + /* reset statistics starting values */ + hw->mac.ops.rebind_hw_stats(hw, &interface->stats); + + err = fm10k_handle_reset(interface); + if (err) + return err; + + /* assume host is not ready, to prevent race with watchdog in case we + * actually don't have connection to the switch + */ + interface->host_ready = false; + fm10k_watchdog_host_not_ready(interface); + + /* clear the service task disable bit to allow service task to start */ + clear_bit(__FM10K_SERVICE_DISABLE, &interface->state); + fm10k_service_event_schedule(interface); + + return err; +} + #ifdef CONFIG_PM /** * fm10k_resume - Restore device to pre-sleep state -- cgit v0.10.2 From 820c91aa9c39e1923fbe4a6593b3d2dfd7ad593f Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:54 -0700 Subject: fm10k: use common reset flow when handling io errors from PCI stack Now that we have extracted the necessary steps for a split suspend/resume flow, re-use these functions instead of using the current open coded flow. This ensures that we don't miss any steps. It also ensures that we have the correct driver states set. Since we'll be handling all of the reset flow ourselves, we no longer need to request a reset in the io_slot_reset() function. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index a6ee046..716a5c8 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -2312,17 +2312,7 @@ static pci_ers_result_t fm10k_io_error_detected(struct pci_dev *pdev, if (state == pci_channel_io_perm_failure) return PCI_ERS_RESULT_DISCONNECT; - rtnl_lock(); - - if (netif_running(netdev)) - fm10k_close(netdev); - - fm10k_mbx_free_irq(interface); - - /* free interrupts */ - fm10k_clear_queueing_scheme(interface); - - rtnl_unlock(); + fm10k_prepare_suspend(interface); /* Request a slot reset. */ return PCI_ERS_RESULT_NEED_RESET; @@ -2336,7 +2326,6 @@ static pci_ers_result_t fm10k_io_error_detected(struct pci_dev *pdev, */ static pci_ers_result_t fm10k_io_slot_reset(struct pci_dev *pdev) { - struct fm10k_intfc *interface = pci_get_drvdata(pdev); pci_ers_result_t result; if (pci_enable_device_mem(pdev)) { @@ -2354,12 +2343,6 @@ static pci_ers_result_t fm10k_io_slot_reset(struct pci_dev *pdev) pci_wake_from_d3(pdev, false); - /* refresh hw_addr in case it was dropped */ - interface->hw.hw_addr = interface->uc_addr; - - interface->flags |= FM10K_FLAG_RESET_REQUESTED; - fm10k_service_event_schedule(interface); - result = PCI_ERS_RESULT_RECOVERED; } @@ -2379,44 +2362,15 @@ static void fm10k_io_resume(struct pci_dev *pdev) { struct fm10k_intfc *interface = pci_get_drvdata(pdev); struct net_device *netdev = interface->netdev; - struct fm10k_hw *hw = &interface->hw; - int err = 0; - - /* reset hardware to known state */ - err = hw->mac.ops.init_hw(&interface->hw); - if (err) { - dev_err(&pdev->dev, "init_hw failed: %d\n", err); - return; - } - - /* reset statistics starting values */ - hw->mac.ops.rebind_hw_stats(hw, &interface->stats); - - rtnl_lock(); - - err = fm10k_init_queueing_scheme(interface); - if (err) { - dev_err(&interface->pdev->dev, - "init_queueing_scheme failed: %d\n", err); - goto unlock; - } - - /* reassociate interrupts */ - fm10k_mbx_request_irq(interface); - - rtnl_lock(); - if (netif_running(netdev)) - err = fm10k_open(netdev); - rtnl_unlock(); + int err; - /* final check of hardware state before registering the interface */ - err = err ? : fm10k_hw_ready(interface); + err = fm10k_handle_resume(interface); - if (!err) + if (err) + dev_warn(&pdev->dev, + "fm10k_io_resume failed: %d\n", err); + else netif_device_attach(netdev); - -unlock: - rtnl_unlock(); } static const struct pci_error_handlers fm10k_err_handler = { -- cgit v0.10.2 From 0593186a17d8427804be162a724a1576367e4d4c Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:55 -0700 Subject: fm10k: implement reset_notify handler for PCIe FLR events When a function level PCI reset is triggered using sysfs, it calls the driver's .reset_notify error handler. Implement a handler based on the now split fm10k_prepare_for_reset and fm10k_handle_reset functions, so that we fully reset the driver when the PCI function level reset occurs. This also ensures the reset is handled in a clean way by first disabling all the driver bits first and then restoring them after the function reset. Previously the stack simply performed a blind function reset and our driver didn't take any part in the process. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index 716a5c8..c60c3b0 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -2373,10 +2373,43 @@ static void fm10k_io_resume(struct pci_dev *pdev) netif_device_attach(netdev); } +/** + * fm10k_io_reset_notify - called when PCI function is reset + * @pdev: Pointer to PCI device + * + * This callback is called when the PCI function is reset such as from + * /sys/class/net//device/reset or similar. When prepare is true, it + * means we should prepare for a function reset. If prepare is false, it means + * the function reset just occurred. + */ +static void fm10k_io_reset_notify(struct pci_dev *pdev, bool prepare) +{ + struct fm10k_intfc *interface = pci_get_drvdata(pdev); + int err = 0; + + if (prepare) { + /* warn incase we have any active VF devices */ + if (pci_num_vf(pdev)) + dev_warn(&pdev->dev, + "PCIe FLR may cause issues for any active VF devices\n"); + + fm10k_prepare_suspend(interface); + } else { + err = fm10k_handle_resume(interface); + } + + if (err) { + dev_warn(&pdev->dev, + "fm10k_io_reset_notify failed: %d\n", err); + netif_device_detach(interface->netdev); + } +} + static const struct pci_error_handlers fm10k_err_handler = { .error_detected = fm10k_io_error_detected, .slot_reset = fm10k_io_slot_reset, .resume = fm10k_io_resume, + .reset_notify = fm10k_io_reset_notify, }; static struct pci_driver fm10k_driver = { -- cgit v0.10.2 From 7756c08b871ac637562bebfbe896a57e804b59bb Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:56 -0700 Subject: fm10k: use common flow for suspend and resume Continuing the effort to commonize the similar suspend/resume flows, finish up by using the new fm10k_handle_suspand and fm10k_handle_resume functions for the standard suspend/resume flow. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index c60c3b0..b02361c 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -2186,60 +2186,13 @@ static int fm10k_resume(struct pci_dev *pdev) /* refresh hw_addr in case it was dropped */ hw->hw_addr = interface->uc_addr; - /* reset hardware to known state */ - err = hw->mac.ops.init_hw(&interface->hw); - if (err) { - dev_err(&pdev->dev, "init_hw failed: %d\n", err); - return err; - } - - /* reset statistics starting values */ - hw->mac.ops.rebind_hw_stats(hw, &interface->stats); - - rtnl_lock(); - - err = fm10k_init_queueing_scheme(interface); - if (err) - goto err_queueing_scheme; - - err = fm10k_mbx_request_irq(interface); - if (err) - goto err_mbx_irq; - - err = fm10k_hw_ready(interface); - if (err) - goto err_open; - - err = netif_running(netdev) ? fm10k_open(netdev) : 0; + err = fm10k_handle_resume(interface); if (err) - goto err_open; - - rtnl_unlock(); - - /* assume host is not ready, to prevent race with watchdog in case we - * actually don't have connection to the switch - */ - interface->host_ready = false; - fm10k_watchdog_host_not_ready(interface); - - /* clear the service task disable bit to allow service task to start */ - clear_bit(__FM10K_SERVICE_DISABLE, &interface->state); - fm10k_service_event_schedule(interface); - - /* restore SR-IOV interface */ - fm10k_iov_resume(pdev); + return err; netif_device_attach(netdev); return 0; -err_open: - fm10k_mbx_free_irq(interface); -err_mbx_irq: - fm10k_clear_queueing_scheme(interface); -err_queueing_scheme: - rtnl_unlock(); - - return err; } /** @@ -2259,27 +2212,7 @@ static int fm10k_suspend(struct pci_dev *pdev, netif_device_detach(netdev); - fm10k_iov_suspend(pdev); - - /* the watchdog tasks may read registers, which will appear like a - * surprise-remove event once the PCI device is disabled. This will - * cause us to close the netdevice, so we don't retain the open/closed - * state post-resume. Prevent this by disabling the service task while - * suspended, until we actually resume. - */ - set_bit(__FM10K_SERVICE_DISABLE, &interface->state); - cancel_work_sync(&interface->service_task); - - rtnl_lock(); - - if (netif_running(netdev)) - fm10k_close(netdev); - - fm10k_mbx_free_irq(interface); - - fm10k_clear_queueing_scheme(interface); - - rtnl_unlock(); + fm10k_prepare_suspend(interface); err = pci_save_state(pdev); if (err) -- cgit v0.10.2 From 0d63a8f50e7a7810b0631cd8ea66d44dafb215e6 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:57 -0700 Subject: fm10k: enable bus master after every reset If an FLR occurs, VF devices will be knocked out of bus master mode, and the driver will be unable to recover from the reset properly, resulting in malicious driver events and an infinite reset loop. In the normal case, the bus master mode will already be enabled and this call will essentially be a no-op. Since we're doing this every reset, it is possible we could remove the other calls to pci_set_master() but it seems not harmful to just leave them in place. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index b02361c..5e40460 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -174,6 +174,8 @@ static int fm10k_handle_reset(struct fm10k_intfc *interface) rtnl_lock(); + pci_set_master(interface->pdev); + /* reset and initialize the hardware so it is in a known state */ err = hw->mac.ops.reset_hw(hw); if (err) { -- cgit v0.10.2 From 9d7dbf0604d37f8cae0c6d184a646037636a8b30 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:58 -0700 Subject: fm10k: check if PCIe link is restored Sometimes, a VF driver will lose PCIe address access, such as due to a PF FLR event. In fm10k_detach_subtask, poll and check whether the PCIe register space is active again and restore the device when it has. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index 5e40460..d4ccb2a 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -123,11 +123,24 @@ static void fm10k_service_timer(unsigned long data) static void fm10k_detach_subtask(struct fm10k_intfc *interface) { struct net_device *netdev = interface->netdev; + u32 __iomem *hw_addr; + u32 value; /* do nothing if device is still present or hw_addr is set */ if (netif_device_present(netdev) || interface->hw.hw_addr) return; + /* check the real address space to see if we've recovered */ + hw_addr = READ_ONCE(interface->uc_addr); + value = readl(hw_addr); + if ((~value)) { + interface->hw.hw_addr = interface->uc_addr; + netif_device_attach(netdev); + interface->flags |= FM10K_FLAG_RESET_REQUESTED; + netdev_warn(netdev, "PCIe link restored, device now attached\n"); + return; + } + rtnl_lock(); if (netif_running(netdev)) -- cgit v0.10.2 From 0afd20e5573c4aa976f1b935b6df73592b46ded5 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:08:59 -0700 Subject: fm10k: implement request_lport_map pointer If the fm10k interface is brought up, but the switch manager software is not running, the driver will continuously request the lport map every few seconds in the base driver watchdog routine. Eventually after several minutes the switch mailbox Tx fifo will fill up and the mailbox will timeout, resulting in a reset. This reset will appear as if for no reason, and occurs regularly every few minutes until the switch manager software is loaded. Prevent this from happening by only requesting the lport map after we've verified the switch mailbox is tx_ready. In order to simplify code logic and reduce code duplication, implement this as a new function pointer "mac.ops.request_lport_map" which the VF will not implement. Otherwise, we have to duplicate the tx_ready check outside of fm10k_get_host_state_generic, or re-implement most of fm10k_get_host_state_generic in the pf version. The resulting code is simpler and easier to understand, and prevents the PF from continuously requesting lport map and filling the Tx fifo of a switch mailbox that isn't ready. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_common.c b/drivers/net/ethernet/intel/fm10k/fm10k_common.c index 5bbf19c..d6baaea 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_common.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_common.c @@ -519,8 +519,12 @@ s32 fm10k_get_host_state_generic(struct fm10k_hw *hw, bool *host_ready) goto out; /* interface cannot receive traffic without logical ports */ - if (mac->dglort_map == FM10K_DGLORTMAP_NONE) + if (mac->dglort_map == FM10K_DGLORTMAP_NONE) { + if (hw->mac.ops.request_lport_map) + ret_val = hw->mac.ops.request_lport_map(hw); + goto out; + } /* if we passed all the tests above then the switch is ready and we no * longer need to check for link diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pf.c b/drivers/net/ethernet/intel/fm10k/fm10k_pf.c index 23f3566..682299d 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pf.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pf.c @@ -1622,25 +1622,15 @@ static s32 fm10k_request_lport_map_pf(struct fm10k_hw *hw) **/ static s32 fm10k_get_host_state_pf(struct fm10k_hw *hw, bool *switch_ready) { - s32 ret_val = 0; u32 dma_ctrl2; /* verify the switch is ready for interaction */ dma_ctrl2 = fm10k_read_reg(hw, FM10K_DMA_CTRL2); if (!(dma_ctrl2 & FM10K_DMA_CTRL2_SWITCH_READY)) - goto out; + return 0; /* retrieve generic host state info */ - ret_val = fm10k_get_host_state_generic(hw, switch_ready); - if (ret_val) - goto out; - - /* interface cannot receive traffic without logical ports */ - if (hw->mac.dglort_map == FM10K_DGLORTMAP_NONE) - ret_val = fm10k_request_lport_map_pf(hw); - -out: - return ret_val; + return fm10k_get_host_state_generic(hw, switch_ready); } /* This structure defines the attibutes to be parsed below */ @@ -1816,6 +1806,7 @@ static const struct fm10k_mac_ops mac_ops_pf = { .set_dma_mask = fm10k_set_dma_mask_pf, .get_fault = fm10k_get_fault_pf, .get_host_state = fm10k_get_host_state_pf, + .request_lport_map = fm10k_request_lport_map_pf, }; static const struct fm10k_iov_ops iov_ops_pf = { diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_type.h b/drivers/net/ethernet/intel/fm10k/fm10k_type.h index 1d65ad8..f4e75c4 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_type.h +++ b/drivers/net/ethernet/intel/fm10k/fm10k_type.h @@ -526,6 +526,7 @@ struct fm10k_mac_ops { s32 (*stop_hw)(struct fm10k_hw *); s32 (*get_bus_info)(struct fm10k_hw *); s32 (*get_host_state)(struct fm10k_hw *, bool *); + s32 (*request_lport_map)(struct fm10k_hw *); s32 (*update_vlan)(struct fm10k_hw *, u32, u8, bool); s32 (*read_mac_addr)(struct fm10k_hw *); s32 (*update_uc_addr)(struct fm10k_hw *, u16, const u8 *, -- cgit v0.10.2 From 0356b23bcc347d4020f2883ad083ab54e573e214 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:09:00 -0700 Subject: fm10k: force link to remain down for at least a second on resume events When we resume from an AER recovery with many active VFs, the PF sees many spurious link up and link down events. Prevent this by delaying link down for at least one second after the resume event. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c index d4ccb2a..b8245c7 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c @@ -2158,6 +2158,10 @@ static int fm10k_handle_resume(struct fm10k_intfc *interface) interface->host_ready = false; fm10k_watchdog_host_not_ready(interface); + /* force link to stay down for a second to prevent link flutter */ + interface->link_down_event = jiffies + (HZ); + set_bit(__FM10K_LINK_DOWN, &interface->state); + /* clear the service task disable bit to allow service task to start */ clear_bit(__FM10K_SERVICE_DISABLE, &interface->state); fm10k_service_event_schedule(interface); -- cgit v0.10.2 From 30e23b711c7b92dea664e61c3eb4b2cdc4b262e9 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:09:01 -0700 Subject: fm10k: return proper error code when pci_enable_msix_range fails The pci_enable_msix_range() function returns a positive value of the number of allocated vectors if it succeeds. On failure it returns a negative error code. Return this code properly so that the error message printed by the driver will show the actual error code instead of being masked by -ENOMEM. Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c index c85fc9894..f72d1ca 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c @@ -1858,7 +1858,7 @@ static int fm10k_init_msix_capability(struct fm10k_intfc *interface) if (v_budget < 0) { kfree(interface->msix_entries); interface->msix_entries = NULL; - return -ENOMEM; + return v_budget; } /* record the number of queues available for q_vectors */ -- cgit v0.10.2 From 5264cc63ba10ebfa0e54e3e641cce2656c7a60e8 Mon Sep 17 00:00:00 2001 From: Jacob Keller Date: Tue, 7 Jun 2016 16:09:02 -0700 Subject: fm10k: bump version number Signed-off-by: Jacob Keller Tested-by: Krishneil Singh Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c index f72d1ca..e9767b6 100644 --- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c +++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c @@ -28,7 +28,7 @@ #include "fm10k.h" -#define DRV_VERSION "0.19.3-k" +#define DRV_VERSION "0.21.2-k" #define DRV_SUMMARY "Intel(R) Ethernet Switch Host Interface Driver" const char fm10k_driver_version[] = DRV_VERSION; char fm10k_driver_name[] = "fm10k"; -- cgit v0.10.2 From 82de0be6862cdca2e6802267bda57cfc8844d3a7 Mon Sep 17 00:00:00 2001 From: Gao Feng Date: Mon, 18 Jul 2016 11:39:23 +0800 Subject: netfilter: Add helper array register/unregister functions Add nf_ct_helper_init(), nf_conntrack_helpers_register() and nf_conntrack_helpers_unregister() functions to avoid repetitive opencoded initialization in helpers. This patch keeps an id parameter for nf_ct_helper_init() not to break helper matching by name that has been inconsistently exposed to userspace through ports, eg. ftp-2121, and through an incremental id, eg. tftp-1. Signed-off-by: Gao Feng Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h index 6cf614bc..1eaac1f 100644 --- a/include/net/netfilter/nf_conntrack_helper.h +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -58,10 +58,25 @@ struct nf_conntrack_helper *__nf_conntrack_helper_find(const char *name, struct nf_conntrack_helper *nf_conntrack_helper_try_module_get(const char *name, u16 l3num, u8 protonum); +void nf_ct_helper_init(struct nf_conntrack_helper *helper, + u16 l3num, u16 protonum, const char *name, + u16 default_port, u16 spec_port, u32 id, + const struct nf_conntrack_expect_policy *exp_pol, + u32 expect_class_max, u32 data_len, + int (*help)(struct sk_buff *skb, unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo), + int (*from_nlattr)(struct nlattr *attr, + struct nf_conn *ct), + struct module *module); int nf_conntrack_helper_register(struct nf_conntrack_helper *); void nf_conntrack_helper_unregister(struct nf_conntrack_helper *); +int nf_conntrack_helpers_register(struct nf_conntrack_helper *, unsigned int); +void nf_conntrack_helpers_unregister(struct nf_conntrack_helper *, + unsigned int); + struct nf_conn_help *nf_ct_helper_ext_add(struct nf_conn *ct, struct nf_conntrack_helper *helper, gfp_t gfp); diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index 19efeba..4314700 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -572,7 +572,7 @@ static int nf_ct_ftp_from_nlattr(struct nlattr *attr, struct nf_conn *ct) return 0; } -static struct nf_conntrack_helper ftp[MAX_PORTS][2] __read_mostly; +static struct nf_conntrack_helper ftp[MAX_PORTS * 2] __read_mostly; static const struct nf_conntrack_expect_policy ftp_exp_policy = { .max_expected = 1, @@ -582,24 +582,13 @@ static const struct nf_conntrack_expect_policy ftp_exp_policy = { /* don't make this __exit, since it's called from __init ! */ static void nf_conntrack_ftp_fini(void) { - int i, j; - for (i = 0; i < ports_c; i++) { - for (j = 0; j < 2; j++) { - if (ftp[i][j].me == NULL) - continue; - - pr_debug("unregistering helper for pf: %d port: %d\n", - ftp[i][j].tuple.src.l3num, ports[i]); - nf_conntrack_helper_unregister(&ftp[i][j]); - } - } - + nf_conntrack_helpers_unregister(ftp, ports_c * 2); kfree(ftp_buffer); } static int __init nf_conntrack_ftp_init(void) { - int i, j = -1, ret = 0; + int i, ret = 0; ftp_buffer = kmalloc(65536, GFP_KERNEL); if (!ftp_buffer) @@ -611,32 +600,21 @@ static int __init nf_conntrack_ftp_init(void) /* FIXME should be configurable whether IPv4 and IPv6 FTP connections are tracked or not - YK */ for (i = 0; i < ports_c; i++) { - ftp[i][0].tuple.src.l3num = PF_INET; - ftp[i][1].tuple.src.l3num = PF_INET6; - for (j = 0; j < 2; j++) { - ftp[i][j].data_len = sizeof(struct nf_ct_ftp_master); - ftp[i][j].tuple.src.u.tcp.port = htons(ports[i]); - ftp[i][j].tuple.dst.protonum = IPPROTO_TCP; - ftp[i][j].expect_policy = &ftp_exp_policy; - ftp[i][j].me = THIS_MODULE; - ftp[i][j].help = help; - ftp[i][j].from_nlattr = nf_ct_ftp_from_nlattr; - if (ports[i] == FTP_PORT) - sprintf(ftp[i][j].name, "ftp"); - else - sprintf(ftp[i][j].name, "ftp-%d", ports[i]); - - pr_debug("registering helper for pf: %d port: %d\n", - ftp[i][j].tuple.src.l3num, ports[i]); - ret = nf_conntrack_helper_register(&ftp[i][j]); - if (ret) { - pr_err("failed to register helper for pf: %d port: %d\n", - ftp[i][j].tuple.src.l3num, ports[i]); - ports_c = i; - nf_conntrack_ftp_fini(); - return ret; - } - } + nf_ct_helper_init(&ftp[2 * i], AF_INET, IPPROTO_TCP, "ftp", + FTP_PORT, ports[i], ports[i], &ftp_exp_policy, + 0, sizeof(struct nf_ct_ftp_master), help, + nf_ct_ftp_from_nlattr, THIS_MODULE); + nf_ct_helper_init(&ftp[2 * i + 1], AF_INET6, IPPROTO_TCP, "ftp", + FTP_PORT, ports[i], ports[i], &ftp_exp_policy, + 0, sizeof(struct nf_ct_ftp_master), help, + nf_ct_ftp_from_nlattr, THIS_MODULE); + } + + ret = nf_conntrack_helpers_register(ftp, ports_c * 2); + if (ret < 0) { + pr_err("failed to register helpers\n"); + kfree(ftp_buffer); + return ret; } return 0; diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index a4294e9..b989b81 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -465,6 +465,63 @@ restart: } EXPORT_SYMBOL_GPL(nf_conntrack_helper_unregister); +void nf_ct_helper_init(struct nf_conntrack_helper *helper, + u16 l3num, u16 protonum, const char *name, + u16 default_port, u16 spec_port, u32 id, + const struct nf_conntrack_expect_policy *exp_pol, + u32 expect_class_max, u32 data_len, + int (*help)(struct sk_buff *skb, unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo), + int (*from_nlattr)(struct nlattr *attr, + struct nf_conn *ct), + struct module *module) +{ + helper->tuple.src.l3num = l3num; + helper->tuple.dst.protonum = protonum; + helper->tuple.src.u.all = htons(spec_port); + helper->expect_policy = exp_pol; + helper->expect_class_max = expect_class_max; + helper->data_len = data_len; + helper->help = help; + helper->from_nlattr = from_nlattr; + helper->me = module; + + if (spec_port == default_port) + snprintf(helper->name, sizeof(helper->name), "%s", name); + else + snprintf(helper->name, sizeof(helper->name), "%s-%u", name, id); +} +EXPORT_SYMBOL_GPL(nf_ct_helper_init); + +int nf_conntrack_helpers_register(struct nf_conntrack_helper *helper, + unsigned int n) +{ + unsigned int i; + int err = 0; + + for (i = 0; i < n; i++) { + err = nf_conntrack_helper_register(&helper[i]); + if (err < 0) + goto err; + } + + return err; +err: + if (i > 0) + nf_conntrack_helpers_unregister(helper, i); + return err; +} +EXPORT_SYMBOL_GPL(nf_conntrack_helpers_register); + +void nf_conntrack_helpers_unregister(struct nf_conntrack_helper *helper, + unsigned int n) +{ + while (n-- > 0) + nf_conntrack_helper_unregister(&helper[n]); +} +EXPORT_SYMBOL_GPL(nf_conntrack_helpers_unregister); + static struct nf_ct_ext_type helper_extend __read_mostly = { .len = sizeof(struct nf_conn_help), .align = __alignof__(struct nf_conn_help), diff --git a/net/netfilter/nf_conntrack_irc.c b/net/netfilter/nf_conntrack_irc.c index f97ac61..1972a14 100644 --- a/net/netfilter/nf_conntrack_irc.c +++ b/net/netfilter/nf_conntrack_irc.c @@ -255,27 +255,18 @@ static int __init nf_conntrack_irc_init(void) ports[ports_c++] = IRC_PORT; for (i = 0; i < ports_c; i++) { - irc[i].tuple.src.l3num = AF_INET; - irc[i].tuple.src.u.tcp.port = htons(ports[i]); - irc[i].tuple.dst.protonum = IPPROTO_TCP; - irc[i].expect_policy = &irc_exp_policy; - irc[i].me = THIS_MODULE; - irc[i].help = help; - - if (ports[i] == IRC_PORT) - sprintf(irc[i].name, "irc"); - else - sprintf(irc[i].name, "irc-%u", i); - - ret = nf_conntrack_helper_register(&irc[i]); - if (ret) { - pr_err("failed to register helper for pf: %u port: %u\n", - irc[i].tuple.src.l3num, ports[i]); - ports_c = i; - nf_conntrack_irc_fini(); - return ret; - } + nf_ct_helper_init(&irc[i], AF_INET, IPPROTO_TCP, "irc", + IRC_PORT, ports[i], i, &irc_exp_policy, + 0, 0, help, NULL, THIS_MODULE); + } + + ret = nf_conntrack_helpers_register(&irc[0], ports_c); + if (ret) { + pr_err("failed to register helpers\n"); + kfree(irc_buffer); + return ret; } + return 0; } @@ -283,10 +274,7 @@ static int __init nf_conntrack_irc_init(void) * it is needed by the init function */ static void nf_conntrack_irc_fini(void) { - int i; - - for (i = 0; i < ports_c; i++) - nf_conntrack_helper_unregister(&irc[i]); + nf_conntrack_helpers_unregister(irc, ports_c); kfree(irc_buffer); } diff --git a/net/netfilter/nf_conntrack_sane.c b/net/netfilter/nf_conntrack_sane.c index 3fcbaab..9dcb9ee 100644 --- a/net/netfilter/nf_conntrack_sane.c +++ b/net/netfilter/nf_conntrack_sane.c @@ -166,7 +166,7 @@ out: return ret; } -static struct nf_conntrack_helper sane[MAX_PORTS][2] __read_mostly; +static struct nf_conntrack_helper sane[MAX_PORTS * 2] __read_mostly; static const struct nf_conntrack_expect_policy sane_exp_policy = { .max_expected = 1, @@ -176,22 +176,13 @@ static const struct nf_conntrack_expect_policy sane_exp_policy = { /* don't make this __exit, since it's called from __init ! */ static void nf_conntrack_sane_fini(void) { - int i, j; - - for (i = 0; i < ports_c; i++) { - for (j = 0; j < 2; j++) { - pr_debug("unregistering helper for pf: %d port: %d\n", - sane[i][j].tuple.src.l3num, ports[i]); - nf_conntrack_helper_unregister(&sane[i][j]); - } - } - + nf_conntrack_helpers_unregister(sane, ports_c * 2); kfree(sane_buffer); } static int __init nf_conntrack_sane_init(void) { - int i, j = -1, ret = 0; + int i, ret = 0; sane_buffer = kmalloc(65536, GFP_KERNEL); if (!sane_buffer) @@ -203,31 +194,23 @@ static int __init nf_conntrack_sane_init(void) /* FIXME should be configurable whether IPv4 and IPv6 connections are tracked or not - YK */ for (i = 0; i < ports_c; i++) { - sane[i][0].tuple.src.l3num = PF_INET; - sane[i][1].tuple.src.l3num = PF_INET6; - for (j = 0; j < 2; j++) { - sane[i][j].data_len = sizeof(struct nf_ct_sane_master); - sane[i][j].tuple.src.u.tcp.port = htons(ports[i]); - sane[i][j].tuple.dst.protonum = IPPROTO_TCP; - sane[i][j].expect_policy = &sane_exp_policy; - sane[i][j].me = THIS_MODULE; - sane[i][j].help = help; - if (ports[i] == SANE_PORT) - sprintf(sane[i][j].name, "sane"); - else - sprintf(sane[i][j].name, "sane-%d", ports[i]); - - pr_debug("registering helper for pf: %d port: %d\n", - sane[i][j].tuple.src.l3num, ports[i]); - ret = nf_conntrack_helper_register(&sane[i][j]); - if (ret) { - pr_err("failed to register helper for pf: %d port: %d\n", - sane[i][j].tuple.src.l3num, ports[i]); - ports_c = i; - nf_conntrack_sane_fini(); - return ret; - } - } + nf_ct_helper_init(&sane[2 * i], AF_INET, IPPROTO_TCP, "sane", + SANE_PORT, ports[i], ports[i], + &sane_exp_policy, 0, + sizeof(struct nf_ct_sane_master), help, NULL, + THIS_MODULE); + nf_ct_helper_init(&sane[2 * i + 1], AF_INET6, IPPROTO_TCP, "sane", + SANE_PORT, ports[i], ports[i], + &sane_exp_policy, 0, + sizeof(struct nf_ct_sane_master), help, NULL, + THIS_MODULE); + } + + ret = nf_conntrack_helpers_register(sane, ports_c * 2); + if (ret < 0) { + pr_err("failed to register helpers\n"); + kfree(sane_buffer); + return ret; } return 0; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index f72ba55..8d9db9d 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -1589,7 +1589,7 @@ static int sip_help_udp(struct sk_buff *skb, unsigned int protoff, return process_sip_msg(skb, ct, protoff, dataoff, &dptr, &datalen); } -static struct nf_conntrack_helper sip[MAX_PORTS][4] __read_mostly; +static struct nf_conntrack_helper sip[MAX_PORTS * 4] __read_mostly; static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1] = { [SIP_EXPECT_SIGNALLING] = { @@ -1616,20 +1616,12 @@ static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1 static void nf_conntrack_sip_fini(void) { - int i, j; - - for (i = 0; i < ports_c; i++) { - for (j = 0; j < ARRAY_SIZE(sip[i]); j++) { - if (sip[i][j].me == NULL) - continue; - nf_conntrack_helper_unregister(&sip[i][j]); - } - } + nf_conntrack_helpers_unregister(sip, ports_c * 4); } static int __init nf_conntrack_sip_init(void) { - int i, j, ret; + int i, ret; if (ports_c == 0) ports[ports_c++] = SIP_PORT; @@ -1637,43 +1629,32 @@ static int __init nf_conntrack_sip_init(void) for (i = 0; i < ports_c; i++) { memset(&sip[i], 0, sizeof(sip[i])); - sip[i][0].tuple.src.l3num = AF_INET; - sip[i][0].tuple.dst.protonum = IPPROTO_UDP; - sip[i][0].help = sip_help_udp; - sip[i][1].tuple.src.l3num = AF_INET; - sip[i][1].tuple.dst.protonum = IPPROTO_TCP; - sip[i][1].help = sip_help_tcp; - - sip[i][2].tuple.src.l3num = AF_INET6; - sip[i][2].tuple.dst.protonum = IPPROTO_UDP; - sip[i][2].help = sip_help_udp; - sip[i][3].tuple.src.l3num = AF_INET6; - sip[i][3].tuple.dst.protonum = IPPROTO_TCP; - sip[i][3].help = sip_help_tcp; - - for (j = 0; j < ARRAY_SIZE(sip[i]); j++) { - sip[i][j].data_len = sizeof(struct nf_ct_sip_master); - sip[i][j].tuple.src.u.udp.port = htons(ports[i]); - sip[i][j].expect_policy = sip_exp_policy; - sip[i][j].expect_class_max = SIP_EXPECT_MAX; - sip[i][j].me = THIS_MODULE; - - if (ports[i] == SIP_PORT) - sprintf(sip[i][j].name, "sip"); - else - sprintf(sip[i][j].name, "sip-%u", i); - - pr_debug("port #%u: %u\n", i, ports[i]); + nf_ct_helper_init(&sip[4 * i], AF_INET, IPPROTO_UDP, "sip", + SIP_PORT, ports[i], i, sip_exp_policy, + SIP_EXPECT_MAX, + sizeof(struct nf_ct_sip_master), sip_help_udp, + NULL, THIS_MODULE); + nf_ct_helper_init(&sip[4 * i + 1], AF_INET, IPPROTO_TCP, "sip", + SIP_PORT, ports[i], i, sip_exp_policy, + SIP_EXPECT_MAX, + sizeof(struct nf_ct_sip_master), sip_help_tcp, + NULL, THIS_MODULE); + nf_ct_helper_init(&sip[4 * i + 2], AF_INET6, IPPROTO_UDP, "sip", + SIP_PORT, ports[i], i, sip_exp_policy, + SIP_EXPECT_MAX, + sizeof(struct nf_ct_sip_master), sip_help_udp, + NULL, THIS_MODULE); + nf_ct_helper_init(&sip[4 * i + 3], AF_INET6, IPPROTO_TCP, "sip", + SIP_PORT, ports[i], i, sip_exp_policy, + SIP_EXPECT_MAX, + sizeof(struct nf_ct_sip_master), sip_help_tcp, + NULL, THIS_MODULE); + } - ret = nf_conntrack_helper_register(&sip[i][j]); - if (ret) { - pr_err("failed to register helper for pf: %u port: %u\n", - sip[i][j].tuple.src.l3num, ports[i]); - ports_c = i; - nf_conntrack_sip_fini(); - return ret; - } - } + ret = nf_conntrack_helpers_register(sip, ports_c * 4); + if (ret < 0) { + pr_err("failed to register helpers\n"); + return ret; } return 0; } diff --git a/net/netfilter/nf_conntrack_tftp.c b/net/netfilter/nf_conntrack_tftp.c index 2e65b543..b1227dc 100644 --- a/net/netfilter/nf_conntrack_tftp.c +++ b/net/netfilter/nf_conntrack_tftp.c @@ -97,7 +97,7 @@ static int tftp_help(struct sk_buff *skb, return ret; } -static struct nf_conntrack_helper tftp[MAX_PORTS][2] __read_mostly; +static struct nf_conntrack_helper tftp[MAX_PORTS * 2] __read_mostly; static const struct nf_conntrack_expect_policy tftp_exp_policy = { .max_expected = 1, @@ -106,47 +106,29 @@ static const struct nf_conntrack_expect_policy tftp_exp_policy = { static void nf_conntrack_tftp_fini(void) { - int i, j; - - for (i = 0; i < ports_c; i++) { - for (j = 0; j < 2; j++) - nf_conntrack_helper_unregister(&tftp[i][j]); - } + nf_conntrack_helpers_unregister(tftp, ports_c * 2); } static int __init nf_conntrack_tftp_init(void) { - int i, j, ret; + int i, ret; if (ports_c == 0) ports[ports_c++] = TFTP_PORT; for (i = 0; i < ports_c; i++) { - memset(&tftp[i], 0, sizeof(tftp[i])); - - tftp[i][0].tuple.src.l3num = AF_INET; - tftp[i][1].tuple.src.l3num = AF_INET6; - for (j = 0; j < 2; j++) { - tftp[i][j].tuple.dst.protonum = IPPROTO_UDP; - tftp[i][j].tuple.src.u.udp.port = htons(ports[i]); - tftp[i][j].expect_policy = &tftp_exp_policy; - tftp[i][j].me = THIS_MODULE; - tftp[i][j].help = tftp_help; - - if (ports[i] == TFTP_PORT) - sprintf(tftp[i][j].name, "tftp"); - else - sprintf(tftp[i][j].name, "tftp-%u", i); - - ret = nf_conntrack_helper_register(&tftp[i][j]); - if (ret) { - pr_err("failed to register helper for pf: %u port: %u\n", - tftp[i][j].tuple.src.l3num, ports[i]); - ports_c = i; - nf_conntrack_tftp_fini(); - return ret; - } - } + nf_ct_helper_init(&tftp[2 * i], AF_INET, IPPROTO_UDP, "tftp", + TFTP_PORT, ports[i], i, &tftp_exp_policy, + 0, 0, tftp_help, NULL, THIS_MODULE); + nf_ct_helper_init(&tftp[2 * i + 1], AF_INET6, IPPROTO_UDP, "tftp", + TFTP_PORT, ports[i], i, &tftp_exp_policy, + 0, 0, tftp_help, NULL, THIS_MODULE); + } + + ret = nf_conntrack_helpers_register(tftp, ports_c * 2); + if (ret < 0) { + pr_err("failed to register helpers\n"); + return ret; } return 0; } -- cgit v0.10.2 From c2d9a4293ced88d7dad7c35c893a31f49f8b64f5 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 18 Jul 2016 20:44:15 +0800 Subject: netfilter: nft_log: fix possible memory leak if log expr init fail Suppose that we specify the NFTA_LOG_PREFIX, then NFTA_LOG_LEVEL and NFTA_LOG_GROUP are specified together or nf_logger_find_get call returns fail, i.e. expr init fail, memory leak will happen. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nft_log.c b/net/netfilter/nft_log.c index 713d668..e1b34ff 100644 --- a/net/netfilter/nft_log.c +++ b/net/netfilter/nft_log.c @@ -52,6 +52,14 @@ static int nft_log_init(const struct nft_ctx *ctx, struct nft_log *priv = nft_expr_priv(expr); struct nf_loginfo *li = &priv->loginfo; const struct nlattr *nla; + int err; + + li->type = NF_LOG_TYPE_LOG; + if (tb[NFTA_LOG_LEVEL] != NULL && + tb[NFTA_LOG_GROUP] != NULL) + return -EINVAL; + if (tb[NFTA_LOG_GROUP] != NULL) + li->type = NF_LOG_TYPE_ULOG; nla = tb[NFTA_LOG_PREFIX]; if (nla != NULL) { @@ -63,13 +71,6 @@ static int nft_log_init(const struct nft_ctx *ctx, priv->prefix = (char *)nft_log_null_prefix; } - li->type = NF_LOG_TYPE_LOG; - if (tb[NFTA_LOG_LEVEL] != NULL && - tb[NFTA_LOG_GROUP] != NULL) - return -EINVAL; - if (tb[NFTA_LOG_GROUP] != NULL) - li->type = NF_LOG_TYPE_ULOG; - switch (li->type) { case NF_LOG_TYPE_LOG: if (tb[NFTA_LOG_LEVEL] != NULL) { @@ -96,7 +97,16 @@ static int nft_log_init(const struct nft_ctx *ctx, break; } - return nf_logger_find_get(ctx->afi->family, li->type); + err = nf_logger_find_get(ctx->afi->family, li->type); + if (err < 0) + goto err1; + + return 0; + +err1: + if (priv->prefix != nft_log_null_prefix) + kfree(priv->prefix); + return err; } static void nft_log_destroy(const struct nft_ctx *ctx, -- cgit v0.10.2 From 1bc4e0136cb32282d7968e11cfabc40763fdb03c Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 18 Jul 2016 20:44:16 +0800 Subject: netfilter: nft_log: check the validity of log level User can specify the log level larger than 7(debug level) via nfnetlink, this is invalid. So in this case, we should report EINVAL to the userspace. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nft_log.c b/net/netfilter/nft_log.c index e1b34ff..5f6f088 100644 --- a/net/netfilter/nft_log.c +++ b/net/netfilter/nft_log.c @@ -79,6 +79,11 @@ static int nft_log_init(const struct nft_ctx *ctx, } else { li->u.log.level = LOGLEVEL_WARNING; } + if (li->u.log.level > LOGLEVEL_DEBUG) { + err = -EINVAL; + goto err1; + } + if (tb[NFTA_LOG_FLAGS] != NULL) { li->u.log.logflags = ntohl(nla_get_be32(tb[NFTA_LOG_FLAGS])); -- cgit v0.10.2 From cc37c1ad42ba6bc07c3d7a999f898e11d69a2580 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Mon, 18 Jul 2016 20:44:17 +0800 Subject: netfilter: nft_log: fix snaplen does not truncate packets There's a similar problem in xt_NFLOG, and was fixed by commit 7643507fe8b5 ("netfilter: xt_NFLOG: nflog-range does not truncate packets"). Only set copy_len here does not work, so we should enable NF_LOG_F_COPY_LEN also. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nft_log.c b/net/netfilter/nft_log.c index 5f6f088..24a73bb 100644 --- a/net/netfilter/nft_log.c +++ b/net/netfilter/nft_log.c @@ -92,6 +92,7 @@ static int nft_log_init(const struct nft_ctx *ctx, case NF_LOG_TYPE_ULOG: li->u.ulog.group = ntohs(nla_get_be16(tb[NFTA_LOG_GROUP])); if (tb[NFTA_LOG_SNAPLEN] != NULL) { + li->u.ulog.flags |= NF_LOG_F_COPY_LEN; li->u.ulog.copy_len = ntohl(nla_get_be32(tb[NFTA_LOG_SNAPLEN])); } @@ -149,7 +150,7 @@ static int nft_log_dump(struct sk_buff *skb, const struct nft_expr *expr) if (nla_put_be16(skb, NFTA_LOG_GROUP, htons(li->u.ulog.group))) goto nla_put_failure; - if (li->u.ulog.copy_len) { + if (li->u.ulog.flags & NF_LOG_F_COPY_LEN) { if (nla_put_be32(skb, NFTA_LOG_SNAPLEN, htonl(li->u.ulog.copy_len))) goto nla_put_failure; -- cgit v0.10.2 From 6e1f760e13c75eb0c21c75c6eed918e25b54cd07 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 19 Jul 2016 12:20:45 +0200 Subject: netfilter: nf_tables: allow to filter out rules by table and chain If the table and/or chain attributes are set in a rule dump request, we filter out the rules based on this selection. Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 0211eae..13d50e7 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -1857,10 +1857,16 @@ err: return err; } +struct nft_rule_dump_ctx { + char table[NFT_TABLE_MAXNAMELEN]; + char chain[NFT_CHAIN_MAXNAMELEN]; +}; + static int nf_tables_dump_rules(struct sk_buff *skb, struct netlink_callback *cb) { const struct nfgenmsg *nfmsg = nlmsg_data(cb->nlh); + const struct nft_rule_dump_ctx *ctx = cb->data; const struct nft_af_info *afi; const struct nft_table *table; const struct nft_chain *chain; @@ -1877,7 +1883,15 @@ static int nf_tables_dump_rules(struct sk_buff *skb, continue; list_for_each_entry_rcu(table, &afi->tables, list) { + if (ctx && ctx->table[0] && + strcmp(ctx->table, table->name) != 0) + continue; + list_for_each_entry_rcu(chain, &table->chains, list) { + if (ctx && ctx->chain[0] && + strcmp(ctx->chain, chain->name) != 0) + continue; + list_for_each_entry_rcu(rule, &chain->rules, list) { if (!nft_is_active(net, rule)) goto cont; @@ -1907,6 +1921,12 @@ done: return skb->len; } +static int nf_tables_dump_rules_done(struct netlink_callback *cb) +{ + kfree(cb->data); + return 0; +} + static int nf_tables_getrule(struct net *net, struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -1924,7 +1944,25 @@ static int nf_tables_getrule(struct net *net, struct sock *nlsk, if (nlh->nlmsg_flags & NLM_F_DUMP) { struct netlink_dump_control c = { .dump = nf_tables_dump_rules, + .done = nf_tables_dump_rules_done, }; + + if (nla[NFTA_RULE_TABLE] || nla[NFTA_RULE_CHAIN]) { + struct nft_rule_dump_ctx *ctx; + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + if (nla[NFTA_RULE_TABLE]) + nla_strlcpy(ctx->table, nla[NFTA_RULE_TABLE], + sizeof(ctx->table)); + if (nla[NFTA_RULE_CHAIN]) + nla_strlcpy(ctx->chain, nla[NFTA_RULE_CHAIN], + sizeof(ctx->chain)); + c.data = ctx; + } + return netlink_dump_start(nlsk, skb, nlh, &c); } -- cgit v0.10.2 From e6c044f5f68e0071e94829c62dbb2549e44a09e9 Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Thu, 21 Jul 2016 11:42:54 +1000 Subject: net/faraday: Disallow using reversed MAC address from hardware The initial MAC address is retrieved from hardware if it's not provided by device-tree. The reserved MAC address from hardware will be used if non-reserved MAC address is invalid. It will cause mismatched MAC address seen by hardware and software. This disallows using the reserved hardware MAC address to avoid the mismatched MAC address seen by hardware and software. Fixes: 113ce107afe9 ("net/faraday: Read MAC address from chip") Suggested-by: David Laight Suggested-by: Benjamin Herrenschmidt Signed-off-by: Gavin Shan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/faraday/ftgmac100.c b/drivers/net/ethernet/faraday/ftgmac100.c index 2d4c7ea..36361f8 100644 --- a/drivers/net/ethernet/faraday/ftgmac100.c +++ b/drivers/net/ethernet/faraday/ftgmac100.c @@ -163,15 +163,6 @@ static void ftgmac100_setup_mac(struct ftgmac100 *priv) mac[4] = (l >> 8) & 0xff; mac[5] = l & 0xff; - if (!is_valid_ether_addr(mac)) { - mac[5] = (m >> 8) & 0xff; - mac[4] = m & 0xff; - mac[3] = (l >> 24) & 0xff; - mac[2] = (l >> 16) & 0xff; - mac[1] = (l >> 8) & 0xff; - mac[0] = l & 0xff; - } - if (is_valid_ether_addr(mac)) { ether_addr_copy(priv->netdev->dev_addr, mac); dev_info(priv->dev, "Read MAC address %pM from chip\n", mac); -- cgit v0.10.2 From d4673339ce22f1d65ebe05bbdaadabf4efb13d18 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Wed, 20 Jul 2016 18:18:34 -0400 Subject: net: dsa: mv88e6xxx: remove unused phy_mutex Only reg_lock is necessary now and phy_mutex is dead. Remove it. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 899ca1d..99de41f 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -629,12 +629,6 @@ struct mv88e6xxx_chip { */ struct mutex stats_mutex; - /* This mutex serializes phy access for chips with - * indirect phy addressing. It is unused for chips - * with direct phy access. - */ - struct mutex phy_mutex; - /* This mutex serializes eeprom access for chips with * eeprom support. */ -- cgit v0.10.2 From 855b193290e6a07b343ee3c059ea6b4bd6cdc2bf Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Wed, 20 Jul 2016 18:18:35 -0400 Subject: net: dsa: mv88e6xxx: rework EEPROM access The 6352 family of switches and compatibles provide a 8-bit address and 16-bit data access to an optional EEPROM. Newer chip such as the 6390 family slightly changed the access to 16-bit address and 8-bit data. This commit cleans up the EEPROM access code for 16-bit access and makes it easy to eventually introduce future support for 8-bit access. Here's a list of notable changes brought by this patch: - provide Global2 unlocked helpers for EEPROM commands - remove eeprom_mutex, only reg_lock is necessary for driver functions - eeprom_len is 0 for chip without EEPROM, so return it directly - the Running bit must be 0 before r/w, so wait for Busy *and* Running - remove now unused mv88e6xxx_wait and mv88e6xxx_reg_write - other than that, the logic (in _{get,set}_eeprom16) didn't change Chips with an 8-bit EEPROM access will require to implement the 8-suffixed variant of G2 helpers and the related flag: #define MV88E6XXX_FLAGS_EEPROM8 \ (MV88E6XXX_FLAG_G2_EEPROM_CMD | \ MV88E6XXX_FLAG_G2_EEPROM_ADDR) Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 9ba2173..f95f1d4 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -271,18 +271,6 @@ static int _mv88e6xxx_reg_write(struct mv88e6xxx_chip *chip, int addr, return mv88e6xxx_write(chip, addr, reg, val); } -static int mv88e6xxx_reg_write(struct mv88e6xxx_chip *chip, int addr, - int reg, u16 val) -{ - int ret; - - mutex_lock(&chip->reg_lock); - ret = _mv88e6xxx_reg_write(chip, addr, reg, val); - mutex_unlock(&chip->reg_lock); - - return ret; -} - static int mv88e6xxx_mdio_read_direct(struct mv88e6xxx_chip *chip, int addr, int regnum) { @@ -861,259 +849,12 @@ static int _mv88e6xxx_wait(struct mv88e6xxx_chip *chip, int reg, int offset, return -ETIMEDOUT; } -static int mv88e6xxx_wait(struct mv88e6xxx_chip *chip, int reg, - int offset, u16 mask) -{ - int ret; - - mutex_lock(&chip->reg_lock); - ret = _mv88e6xxx_wait(chip, reg, offset, mask); - mutex_unlock(&chip->reg_lock); - - return ret; -} - static int mv88e6xxx_mdio_wait(struct mv88e6xxx_chip *chip) { return _mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_SMI_OP, GLOBAL2_SMI_OP_BUSY); } -static int mv88e6xxx_eeprom_load_wait(struct dsa_switch *ds) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - - return mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_LOAD); -} - -static int mv88e6xxx_eeprom_busy_wait(struct dsa_switch *ds) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - - return mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_BUSY); -} - -static int mv88e6xxx_read_eeprom_word(struct dsa_switch *ds, int addr) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int ret; - - mutex_lock(&chip->eeprom_mutex); - - ret = mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_READ | - (addr & GLOBAL2_EEPROM_OP_ADDR_MASK)); - if (ret < 0) - goto error; - - ret = mv88e6xxx_eeprom_busy_wait(ds); - if (ret < 0) - goto error; - - ret = mv88e6xxx_reg_read(chip, REG_GLOBAL2, GLOBAL2_EEPROM_DATA); -error: - mutex_unlock(&chip->eeprom_mutex); - return ret; -} - -static int mv88e6xxx_get_eeprom_len(struct dsa_switch *ds) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - - if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEPROM)) - return chip->eeprom_len; - - return 0; -} - -static int mv88e6xxx_get_eeprom(struct dsa_switch *ds, - struct ethtool_eeprom *eeprom, u8 *data) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int offset; - int len; - int ret; - - if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEPROM)) - return -EOPNOTSUPP; - - offset = eeprom->offset; - len = eeprom->len; - eeprom->len = 0; - - eeprom->magic = 0xc3ec4951; - - ret = mv88e6xxx_eeprom_load_wait(ds); - if (ret < 0) - return ret; - - if (offset & 1) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - *data++ = (word >> 8) & 0xff; - - offset++; - len--; - eeprom->len++; - } - - while (len >= 2) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - *data++ = word & 0xff; - *data++ = (word >> 8) & 0xff; - - offset += 2; - len -= 2; - eeprom->len += 2; - } - - if (len) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - *data++ = word & 0xff; - - offset++; - len--; - eeprom->len++; - } - - return 0; -} - -static int mv88e6xxx_eeprom_is_readonly(struct dsa_switch *ds) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int ret; - - ret = mv88e6xxx_reg_read(chip, REG_GLOBAL2, GLOBAL2_EEPROM_OP); - if (ret < 0) - return ret; - - if (!(ret & GLOBAL2_EEPROM_OP_WRITE_EN)) - return -EROFS; - - return 0; -} - -static int mv88e6xxx_write_eeprom_word(struct dsa_switch *ds, int addr, - u16 data) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int ret; - - mutex_lock(&chip->eeprom_mutex); - - ret = mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_EEPROM_DATA, data); - if (ret < 0) - goto error; - - ret = mv88e6xxx_reg_write(chip, REG_GLOBAL2, GLOBAL2_EEPROM_OP, - GLOBAL2_EEPROM_OP_WRITE | - (addr & GLOBAL2_EEPROM_OP_ADDR_MASK)); - if (ret < 0) - goto error; - - ret = mv88e6xxx_eeprom_busy_wait(ds); -error: - mutex_unlock(&chip->eeprom_mutex); - return ret; -} - -static int mv88e6xxx_set_eeprom(struct dsa_switch *ds, - struct ethtool_eeprom *eeprom, u8 *data) -{ - struct mv88e6xxx_chip *chip = ds_to_priv(ds); - int offset; - int ret; - int len; - - if (!mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEPROM)) - return -EOPNOTSUPP; - - if (eeprom->magic != 0xc3ec4951) - return -EINVAL; - - ret = mv88e6xxx_eeprom_is_readonly(ds); - if (ret) - return ret; - - offset = eeprom->offset; - len = eeprom->len; - eeprom->len = 0; - - ret = mv88e6xxx_eeprom_load_wait(ds); - if (ret < 0) - return ret; - - if (offset & 1) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - word = (*data++ << 8) | (word & 0xff); - - ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); - if (ret < 0) - return ret; - - offset++; - len--; - eeprom->len++; - } - - while (len >= 2) { - int word; - - word = *data++; - word |= *data++ << 8; - - ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); - if (ret < 0) - return ret; - - offset += 2; - len -= 2; - eeprom->len += 2; - } - - if (len) { - int word; - - word = mv88e6xxx_read_eeprom_word(ds, offset >> 1); - if (word < 0) - return word; - - word = (word & 0xff00) | *data++; - - ret = mv88e6xxx_write_eeprom_word(ds, offset >> 1, word); - if (ret < 0) - return ret; - - offset++; - len--; - eeprom->len++; - } - - return 0; -} - static int _mv88e6xxx_atu_wait(struct mv88e6xxx_chip *chip) { return _mv88e6xxx_wait(chip, REG_GLOBAL, GLOBAL_ATU_OP, @@ -3261,6 +3002,58 @@ static int mv88e6xxx_g2_clear_pot(struct mv88e6xxx_chip *chip) return err; } +static int mv88e6xxx_g2_eeprom_wait(struct mv88e6xxx_chip *chip) +{ + return _mv88e6xxx_wait(chip, REG_GLOBAL2, GLOBAL2_EEPROM_CMD, + GLOBAL2_EEPROM_CMD_BUSY | + GLOBAL2_EEPROM_CMD_RUNNING); +} + +static int mv88e6xxx_g2_eeprom_cmd(struct mv88e6xxx_chip *chip, u16 cmd) +{ + int err; + + err = mv88e6xxx_write(chip, REG_GLOBAL2, GLOBAL2_EEPROM_CMD, cmd); + if (err) + return err; + + return mv88e6xxx_g2_eeprom_wait(chip); +} + +static int mv88e6xxx_g2_eeprom_read16(struct mv88e6xxx_chip *chip, + u8 addr, u16 *data) +{ + u16 cmd = GLOBAL2_EEPROM_CMD_OP_READ | addr; + int err; + + err = mv88e6xxx_g2_eeprom_wait(chip); + if (err) + return err; + + err = mv88e6xxx_g2_eeprom_cmd(chip, cmd); + if (err) + return err; + + return mv88e6xxx_read(chip, REG_GLOBAL2, GLOBAL2_EEPROM_DATA, data); +} + +static int mv88e6xxx_g2_eeprom_write16(struct mv88e6xxx_chip *chip, + u8 addr, u16 data) +{ + u16 cmd = GLOBAL2_EEPROM_CMD_OP_WRITE | addr; + int err; + + err = mv88e6xxx_g2_eeprom_wait(chip); + if (err) + return err; + + err = mv88e6xxx_write(chip, REG_GLOBAL2, GLOBAL2_EEPROM_DATA, data); + if (err) + return err; + + return mv88e6xxx_g2_eeprom_cmd(chip, cmd); +} + static int mv88e6xxx_g2_setup(struct mv88e6xxx_chip *chip) { u16 reg; @@ -3345,9 +3138,6 @@ static int mv88e6xxx_setup(struct dsa_switch *ds) chip->ds = ds; ds->slave_mii_bus = chip->mdio_bus; - if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEPROM)) - mutex_init(&chip->eeprom_mutex); - mutex_lock(&chip->reg_lock); err = mv88e6xxx_switch_reset(chip); @@ -3670,6 +3460,173 @@ static int mv88e6xxx_get_temp_alarm(struct dsa_switch *ds, bool *alarm) } #endif /* CONFIG_NET_DSA_HWMON */ +static int mv88e6xxx_get_eeprom_len(struct dsa_switch *ds) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + + return chip->eeprom_len; +} + +static int mv88e6xxx_get_eeprom16(struct mv88e6xxx_chip *chip, + struct ethtool_eeprom *eeprom, u8 *data) +{ + unsigned int offset = eeprom->offset; + unsigned int len = eeprom->len; + u16 val; + int err; + + eeprom->len = 0; + + if (offset & 1) { + err = mv88e6xxx_g2_eeprom_read16(chip, offset >> 1, &val); + if (err) + return err; + + *data++ = (val >> 8) & 0xff; + + offset++; + len--; + eeprom->len++; + } + + while (len >= 2) { + err = mv88e6xxx_g2_eeprom_read16(chip, offset >> 1, &val); + if (err) + return err; + + *data++ = val & 0xff; + *data++ = (val >> 8) & 0xff; + + offset += 2; + len -= 2; + eeprom->len += 2; + } + + if (len) { + err = mv88e6xxx_g2_eeprom_read16(chip, offset >> 1, &val); + if (err) + return err; + + *data++ = val & 0xff; + + offset++; + len--; + eeprom->len++; + } + + return 0; +} + +static int mv88e6xxx_get_eeprom(struct dsa_switch *ds, + struct ethtool_eeprom *eeprom, u8 *data) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int err; + + mutex_lock(&chip->reg_lock); + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAGS_EEPROM16)) + err = mv88e6xxx_get_eeprom16(chip, eeprom, data); + else + err = -EOPNOTSUPP; + + mutex_unlock(&chip->reg_lock); + + if (err) + return err; + + eeprom->magic = 0xc3ec4951; + + return 0; +} + +static int mv88e6xxx_set_eeprom16(struct mv88e6xxx_chip *chip, + struct ethtool_eeprom *eeprom, u8 *data) +{ + unsigned int offset = eeprom->offset; + unsigned int len = eeprom->len; + u16 val; + int err; + + /* Ensure the RO WriteEn bit is set */ + err = mv88e6xxx_read(chip, REG_GLOBAL2, GLOBAL2_EEPROM_CMD, &val); + if (err) + return err; + + if (!(val & GLOBAL2_EEPROM_CMD_WRITE_EN)) + return -EROFS; + + eeprom->len = 0; + + if (offset & 1) { + err = mv88e6xxx_g2_eeprom_read16(chip, offset >> 1, &val); + if (err) + return err; + + val = (*data++ << 8) | (val & 0xff); + + err = mv88e6xxx_g2_eeprom_write16(chip, offset >> 1, val); + if (err) + return err; + + offset++; + len--; + eeprom->len++; + } + + while (len >= 2) { + val = *data++; + val |= *data++ << 8; + + err = mv88e6xxx_g2_eeprom_write16(chip, offset >> 1, val); + if (err) + return err; + + offset += 2; + len -= 2; + eeprom->len += 2; + } + + if (len) { + err = mv88e6xxx_g2_eeprom_read16(chip, offset >> 1, &val); + if (err) + return err; + + val = (val & 0xff00) | *data++; + + err = mv88e6xxx_g2_eeprom_write16(chip, offset >> 1, val); + if (err) + return err; + + offset++; + len--; + eeprom->len++; + } + + return 0; +} + +static int mv88e6xxx_set_eeprom(struct dsa_switch *ds, + struct ethtool_eeprom *eeprom, u8 *data) +{ + struct mv88e6xxx_chip *chip = ds_to_priv(ds); + int err; + + if (eeprom->magic != 0xc3ec4951) + return -EINVAL; + + mutex_lock(&chip->reg_lock); + + if (mv88e6xxx_has(chip, MV88E6XXX_FLAGS_EEPROM16)) + err = mv88e6xxx_set_eeprom16(chip, eeprom, data); + else + err = -EOPNOTSUPP; + + mutex_unlock(&chip->reg_lock); + + return err; +} + static const struct mv88e6xxx_info mv88e6xxx_table[] = { [MV88E6085] = { .prod_num = PORT_SWITCH_ID_PROD_NUM_6085, @@ -4063,7 +4020,7 @@ static int mv88e6xxx_probe(struct mdio_device *mdiodev) if (IS_ERR(chip->reset)) return PTR_ERR(chip->reset); - if (mv88e6xxx_has(chip, MV88E6XXX_FLAG_EEPROM) && + if (mv88e6xxx_has(chip, MV88E6XXX_FLAGS_EEPROM16) && !of_property_read_u32(np, "eeprom-length", &eeprom_len)) chip->eeprom_len = eeprom_len; diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h index 99de41f..48d6ea7 100644 --- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h +++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h @@ -318,13 +318,14 @@ #define GLOBAL2_PRIO_OVERRIDE_SNOOP_SHIFT 4 #define GLOBAL2_PRIO_OVERRIDE_FORCE_ARP BIT(3) #define GLOBAL2_PRIO_OVERRIDE_ARP_SHIFT 0 -#define GLOBAL2_EEPROM_OP 0x14 -#define GLOBAL2_EEPROM_OP_BUSY BIT(15) -#define GLOBAL2_EEPROM_OP_WRITE ((3 << 12) | GLOBAL2_EEPROM_OP_BUSY) -#define GLOBAL2_EEPROM_OP_READ ((4 << 12) | GLOBAL2_EEPROM_OP_BUSY) -#define GLOBAL2_EEPROM_OP_LOAD BIT(11) -#define GLOBAL2_EEPROM_OP_WRITE_EN BIT(10) -#define GLOBAL2_EEPROM_OP_ADDR_MASK 0xff +#define GLOBAL2_EEPROM_CMD 0x14 +#define GLOBAL2_EEPROM_CMD_BUSY BIT(15) +#define GLOBAL2_EEPROM_CMD_OP_WRITE ((0x3 << 12) | GLOBAL2_EEPROM_CMD_BUSY) +#define GLOBAL2_EEPROM_CMD_OP_READ ((0x4 << 12) | GLOBAL2_EEPROM_CMD_BUSY) +#define GLOBAL2_EEPROM_CMD_OP_LOAD ((0x6 << 12) | GLOBAL2_EEPROM_CMD_BUSY) +#define GLOBAL2_EEPROM_CMD_RUNNING BIT(11) +#define GLOBAL2_EEPROM_CMD_WRITE_EN BIT(10) +#define GLOBAL2_EEPROM_CMD_ADDR_MASK 0xff #define GLOBAL2_EEPROM_DATA 0x15 #define GLOBAL2_PTP_AVB_OP 0x16 #define GLOBAL2_PTP_AVB_DATA 0x17 @@ -387,11 +388,6 @@ enum mv88e6xxx_cap { */ MV88E6XXX_CAP_EEE, - /* EEPROM Command and Data registers. - * See GLOBAL2_EEPROM_OP and GLOBAL2_EEPROM_DATA. - */ - MV88E6XXX_CAP_EEPROM, - /* Switch Global 2 Registers. * The device contains a second set of global 16-bit registers. */ @@ -404,6 +400,8 @@ enum mv88e6xxx_cap { MV88E6XXX_CAP_G2_PVT_DATA, /* (0x0c) Cross Chip Port VLAN Data */ MV88E6XXX_CAP_G2_SWITCH_MAC, /* (0x0d) Switch MAC/WoL/WoF */ MV88E6XXX_CAP_G2_POT, /* (0x0f) Priority Override Table */ + MV88E6XXX_CAP_G2_EEPROM_CMD, /* (0x14) EEPROM Command */ + MV88E6XXX_CAP_G2_EEPROM_DATA, /* (0x15) EEPROM Data */ /* Multi-chip Addressing Mode. * Some chips require an indirect SMI access when their SMI device @@ -443,7 +441,6 @@ enum mv88e6xxx_cap { /* Bitmask of capabilities */ #define MV88E6XXX_FLAG_EEE BIT(MV88E6XXX_CAP_EEE) -#define MV88E6XXX_FLAG_EEPROM BIT(MV88E6XXX_CAP_EEPROM) #define MV88E6XXX_FLAG_GLOBAL2 BIT(MV88E6XXX_CAP_GLOBAL2) #define MV88E6XXX_FLAG_G2_MGMT_EN_2X BIT(MV88E6XXX_CAP_G2_MGMT_EN_2X) #define MV88E6XXX_FLAG_G2_MGMT_EN_0X BIT(MV88E6XXX_CAP_G2_MGMT_EN_0X) @@ -453,6 +450,8 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_G2_PVT_DATA BIT(MV88E6XXX_CAP_G2_PVT_DATA) #define MV88E6XXX_FLAG_G2_SWITCH_MAC BIT(MV88E6XXX_CAP_G2_SWITCH_MAC) #define MV88E6XXX_FLAG_G2_POT BIT(MV88E6XXX_CAP_G2_POT) +#define MV88E6XXX_FLAG_G2_EEPROM_CMD BIT(MV88E6XXX_CAP_G2_EEPROM_CMD) +#define MV88E6XXX_FLAG_G2_EEPROM_DATA BIT(MV88E6XXX_CAP_G2_EEPROM_DATA) #define MV88E6XXX_FLAG_MULTI_CHIP BIT(MV88E6XXX_CAP_MULTI_CHIP) #define MV88E6XXX_FLAG_PPU BIT(MV88E6XXX_CAP_PPU) #define MV88E6XXX_FLAG_PPU_ACTIVE BIT(MV88E6XXX_CAP_PPU_ACTIVE) @@ -462,6 +461,11 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAG_TEMP_LIMIT BIT(MV88E6XXX_CAP_TEMP_LIMIT) #define MV88E6XXX_FLAG_VTU BIT(MV88E6XXX_CAP_VTU) +/* EEPROM Programming via Global2 with 16-bit data */ +#define MV88E6XXX_FLAGS_EEPROM16 \ + (MV88E6XXX_FLAG_G2_EEPROM_CMD | \ + MV88E6XXX_FLAG_G2_EEPROM_DATA) + /* Ingress Rate Limit unit */ #define MV88E6XXX_FLAGS_IRL \ (MV88E6XXX_FLAG_G2_IRL_CMD | \ @@ -513,7 +517,6 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAGS_FAMILY_6320 \ (MV88E6XXX_FLAG_EEE | \ - MV88E6XXX_FLAG_EEPROM | \ MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ @@ -525,6 +528,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_EEPROM16 | \ MV88E6XXX_FLAGS_IRL | \ MV88E6XXX_FLAGS_PVT) @@ -545,7 +549,6 @@ enum mv88e6xxx_cap { #define MV88E6XXX_FLAGS_FAMILY_6352 \ (MV88E6XXX_FLAG_EEE | \ - MV88E6XXX_FLAG_EEPROM | \ MV88E6XXX_FLAG_GLOBAL2 | \ MV88E6XXX_FLAG_G2_MGMT_EN_2X | \ MV88E6XXX_FLAG_G2_MGMT_EN_0X | \ @@ -558,6 +561,7 @@ enum mv88e6xxx_cap { MV88E6XXX_FLAG_TEMP | \ MV88E6XXX_FLAG_TEMP_LIMIT | \ MV88E6XXX_FLAG_VTU | \ + MV88E6XXX_FLAGS_EEPROM16 | \ MV88E6XXX_FLAGS_IRL | \ MV88E6XXX_FLAGS_PVT) @@ -629,11 +633,6 @@ struct mv88e6xxx_chip { */ struct mutex stats_mutex; - /* This mutex serializes eeprom access for chips with - * eeprom support. - */ - struct mutex eeprom_mutex; - struct mv88e6xxx_priv_port ports[DSA_MAX_PORTS]; /* A switch may have a GPIO line tied to its reset pin. Parse -- cgit v0.10.2 From 8f6345b2483be3eec9faac39d173b98790724011 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Wed, 20 Jul 2016 18:18:36 -0400 Subject: net: dsa: mv88e6xxx: kill last locked reg_read Get rid of the last usage of the locked mv88e6xxx_reg_read function with a new mv88e6xxx_port_read helper, useful later for chips with different port registers base address. Signed-off-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index f95f1d4..d36aedd 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -254,17 +254,6 @@ static int _mv88e6xxx_reg_read(struct mv88e6xxx_chip *chip, int addr, int reg) return val; } -static int mv88e6xxx_reg_read(struct mv88e6xxx_chip *chip, int addr, int reg) -{ - int ret; - - mutex_lock(&chip->reg_lock); - ret = _mv88e6xxx_reg_read(chip, addr, reg); - mutex_unlock(&chip->reg_lock); - - return ret; -} - static int _mv88e6xxx_reg_write(struct mv88e6xxx_chip *chip, int addr, int reg, u16 val) { @@ -2426,6 +2415,17 @@ static int mv88e6xxx_power_on_serdes(struct mv88e6xxx_chip *chip) return ret; } +static int mv88e6xxx_port_read(struct mv88e6xxx_chip *chip, int port, + int reg, u16 *val) +{ + int addr = chip->info->port_base_addr + port; + + if (port >= chip->info->num_ports) + return -EINVAL; + + return mv88e6xxx_read(chip, addr, reg, val); +} + static int mv88e6xxx_setup_port(struct mv88e6xxx_chip *chip, int port) { struct dsa_switch *ds = chip->ds; @@ -3830,12 +3830,15 @@ static const struct mv88e6xxx_info *mv88e6xxx_lookup_info(unsigned int prod_num) static int mv88e6xxx_detect(struct mv88e6xxx_chip *chip) { const struct mv88e6xxx_info *info; - int id, prod_num, rev; + unsigned int prod_num, rev; + u16 id; + int err; - id = mv88e6xxx_reg_read(chip, chip->info->port_base_addr, - PORT_SWITCH_ID); - if (id < 0) - return id; + mutex_lock(&chip->reg_lock); + err = mv88e6xxx_port_read(chip, 0, PORT_SWITCH_ID, &id); + mutex_unlock(&chip->reg_lock); + if (err) + return err; prod_num = (id & 0xfff0) >> 4; rev = id & 0x000f; -- cgit v0.10.2 From cb7386d37eae20141cbc292a883e5e358cb5a929 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Wed, 20 Jul 2016 17:22:33 -0700 Subject: net/mlx4_en: use READ_ONCE when freeing xdp_prog For consistency, and in order to hint at the synchronous nature of the xdp_prog field, use READ_ONCE in the destroy path of the ring. All occurrences should now use either READ_ONCE or xchg. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index 11d88c8..a02dec6 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -535,9 +535,11 @@ void mlx4_en_destroy_rx_ring(struct mlx4_en_priv *priv, { struct mlx4_en_dev *mdev = priv->mdev; struct mlx4_en_rx_ring *ring = *pring; + struct bpf_prog *old_prog; - if (ring->xdp_prog) - bpf_prog_put(ring->xdp_prog); + old_prog = READ_ONCE(ring->xdp_prog); + if (old_prog) + bpf_prog_put(old_prog); mlx4_free_hwq_res(mdev->dev, &ring->wqres, size * stride + TXBB_SIZE); vfree(ring->rx_info); ring->rx_info = NULL; -- cgit v0.10.2 From 262d8625045e0c81b7859ecd192e9811710f19da Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Wed, 20 Jul 2016 17:22:34 -0700 Subject: rtnl: protect do_setlink from IFLA_XDP_ATTACHED The IFLA_XDP_ATTACHED nested attribute is meant for read-only, and while do_setlink properly ignores it, it should be more paranoid and reject commands that try to set it. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index eba2b82..189cc78 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -2109,6 +2109,10 @@ static int do_setlink(const struct sk_buff *skb, if (err < 0) goto errout; + if (xdp[IFLA_XDP_ATTACHED]) { + err = -EINVAL; + goto errout; + } if (xdp[IFLA_XDP_FD]) { err = dev_change_xdp_fd(dev, nla_get_s32(xdp[IFLA_XDP_FD])); -- cgit v0.10.2 From d9094bda5c985d1f9da66e9e3fd6323b49dee44d Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Wed, 20 Jul 2016 17:22:35 -0700 Subject: bpf: make xdp sample variable names more meaningful The naming choice of index is not terribly descriptive, and dropcnt is in fact incorrect for xdp2. Pick better names for these: ipproto and rxcnt. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/samples/bpf/xdp1_kern.c b/samples/bpf/xdp1_kern.c index e7dd8ac..2197421 100644 --- a/samples/bpf/xdp1_kern.c +++ b/samples/bpf/xdp1_kern.c @@ -14,7 +14,7 @@ #include #include "bpf_helpers.h" -struct bpf_map_def SEC("maps") dropcnt = { +struct bpf_map_def SEC("maps") rxcnt = { .type = BPF_MAP_TYPE_PERCPU_ARRAY, .key_size = sizeof(u32), .value_size = sizeof(long), @@ -49,7 +49,7 @@ int xdp_prog1(struct xdp_md *ctx) long *value; u16 h_proto; u64 nh_off; - u32 index; + u32 ipproto; nh_off = sizeof(*eth); if (data + nh_off > data_end) @@ -77,13 +77,13 @@ int xdp_prog1(struct xdp_md *ctx) } if (h_proto == htons(ETH_P_IP)) - index = parse_ipv4(data, nh_off, data_end); + ipproto = parse_ipv4(data, nh_off, data_end); else if (h_proto == htons(ETH_P_IPV6)) - index = parse_ipv6(data, nh_off, data_end); + ipproto = parse_ipv6(data, nh_off, data_end); else - index = 0; + ipproto = 0; - value = bpf_map_lookup_elem(&dropcnt, &index); + value = bpf_map_lookup_elem(&rxcnt, &ipproto); if (value) *value += 1; diff --git a/samples/bpf/xdp2_kern.c b/samples/bpf/xdp2_kern.c index 38fe7e1..e012888 100644 --- a/samples/bpf/xdp2_kern.c +++ b/samples/bpf/xdp2_kern.c @@ -14,7 +14,7 @@ #include #include "bpf_helpers.h" -struct bpf_map_def SEC("maps") dropcnt = { +struct bpf_map_def SEC("maps") rxcnt = { .type = BPF_MAP_TYPE_PERCPU_ARRAY, .key_size = sizeof(u32), .value_size = sizeof(long), @@ -65,7 +65,7 @@ int xdp_prog1(struct xdp_md *ctx) long *value; u16 h_proto; u64 nh_off; - u32 index; + u32 ipproto; nh_off = sizeof(*eth); if (data + nh_off > data_end) @@ -93,17 +93,17 @@ int xdp_prog1(struct xdp_md *ctx) } if (h_proto == htons(ETH_P_IP)) - index = parse_ipv4(data, nh_off, data_end); + ipproto = parse_ipv4(data, nh_off, data_end); else if (h_proto == htons(ETH_P_IPV6)) - index = parse_ipv6(data, nh_off, data_end); + ipproto = parse_ipv6(data, nh_off, data_end); else - index = 0; + ipproto = 0; - value = bpf_map_lookup_elem(&dropcnt, &index); + value = bpf_map_lookup_elem(&rxcnt, &ipproto); if (value) *value += 1; - if (index == 17) { + if (ipproto == IPPROTO_UDP) { swap_src_dst_mac(data); rc = XDP_TX; } -- cgit v0.10.2 From 0f06a6787e0516352117f0720e3052f46bc13523 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Wed, 20 Jul 2016 15:48:43 -0700 Subject: samples: Add an IPv6 '-6' option to the pktgen scripts Add a '-6' option to the sample pktgen scripts for sending out IPv6 packets. [root@kerneldev010.prn1 ~/pktgen]# ./pktgen_sample03_burst_single_flow.sh -i eth0 -s 64 -d fe80::f652:14ff:fec2:a14c -m f4:52:14:c2:a1:4c -b 32 -6 [root@kerneldev011.prn1 ~]# tcpdump -i eth0 -nn -c3 port 9 tcpdump: WARNING: eth0: no IPv4 address assigned tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eth0, link-type EN10MB (Ethernet), capture size 65535 bytes 14:38:51.815297 IP6 fe80::f652:14ff:fec2:2ad2.9 > fe80::f652:14ff:fec2:a14c.9: UDP, length 16 14:38:51.815311 IP6 fe80::f652:14ff:fec2:2ad2.9 > fe80::f652:14ff:fec2:a14c.9: UDP, length 16 14:38:51.815313 IP6 fe80::f652:14ff:fec2:2ad2.9 > fe80::f652:14ff:fec2:a14c.9: UDP, length 16 Signed-off-by: Martin KaFai Lau Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/samples/pktgen/parameters.sh b/samples/pktgen/parameters.sh index 33b70fd..f70ea7d 100644 --- a/samples/pktgen/parameters.sh +++ b/samples/pktgen/parameters.sh @@ -14,12 +14,13 @@ function usage() { echo " -b : (\$BURST) HW level bursting of SKBs" echo " -v : (\$VERBOSE) verbose" echo " -x : (\$DEBUG) debug" + echo " -6 : (\$IP6) IPv6" echo "" } ## --- Parse command line arguments / parameters --- ## echo "Commandline options:" -while getopts "s:i:d:m:t:c:b:vxh" option; do +while getopts "s:i:d:m:t:c:b:vxh6" option; do case $option in i) # interface export DEV=$OPTARG @@ -59,6 +60,10 @@ while getopts "s:i:d:m:t:c:b:vxh" option; do export DEBUG=yes info "Debug mode: DEBUG=$DEBUG" ;; + 6) + export IP6=6 + info "IP6: IP6=$IP6" + ;; h|?|*) usage; err 2 "[ERROR] Unknown parameters!!!" diff --git a/samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh b/samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh index cb15903..f3e1bed 100755 --- a/samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh +++ b/samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh @@ -34,7 +34,9 @@ root_check_run_with_sudo "$@" source ${basedir}/parameters.sh # Using invalid DST_MAC will cause the packets to get dropped in # ip_rcv() which is part of the test -[ -z "$DEST_IP" ] && DEST_IP="198.18.0.42" +if [ -z "$DEST_IP" ]; then + [ -z "$IP6" ] && DEST_IP="198.18.0.42" || DEST_IP="FD00::1" +fi [ -z "$DST_MAC" ] && DST_MAC="90:e2:ba:ff:ff:ff" [ -z "$BURST" ] && BURST=1024 @@ -64,7 +66,7 @@ for ((thread = 0; thread < $THREADS; thread++)); do # Destination pg_set $dev "dst_mac $DST_MAC" - pg_set $dev "dst $DEST_IP" + pg_set $dev "dst$IP6 $DEST_IP" # Inject packet into RX path of stack pg_set $dev "xmit_mode netif_receive" diff --git a/samples/pktgen/pktgen_bench_xmit_mode_queue_xmit.sh b/samples/pktgen/pktgen_bench_xmit_mode_queue_xmit.sh index 4e4e92b..cc102e9 100755 --- a/samples/pktgen/pktgen_bench_xmit_mode_queue_xmit.sh +++ b/samples/pktgen/pktgen_bench_xmit_mode_queue_xmit.sh @@ -13,7 +13,9 @@ root_check_run_with_sudo "$@" # Parameter parsing via include source ${basedir}/parameters.sh -[ -z "$DEST_IP" ] && DEST_IP="198.18.0.42" +if [ -z "$DEST_IP" ]; then + [ -z "$IP6" ] && DEST_IP="198.18.0.42" || DEST_IP="FD00::1" +fi [ -z "$DST_MAC" ] && DST_MAC="90:e2:ba:ff:ff:ff" # Burst greater than 1 are invalid for queue_xmit mode @@ -47,7 +49,7 @@ for ((thread = 0; thread < $THREADS; thread++)); do # Destination pg_set $dev "dst_mac $DST_MAC" - pg_set $dev "dst $DEST_IP" + pg_set $dev "dst$IP6 $DEST_IP" # Inject packet into TX qdisc egress path of stack pg_set $dev "xmit_mode queue_xmit" diff --git a/samples/pktgen/pktgen_sample01_simple.sh b/samples/pktgen/pktgen_sample01_simple.sh index 8c9d318..29ef4ba 100755 --- a/samples/pktgen/pktgen_sample01_simple.sh +++ b/samples/pktgen/pktgen_sample01_simple.sh @@ -14,7 +14,9 @@ root_check_run_with_sudo "$@" source ${basedir}/parameters.sh # # Set some default params, if they didn't get set -[ -z "$DEST_IP" ] && DEST_IP="198.18.0.42" +if [ -z "$DEST_IP" ]; then + [ -z "$IP6" ] && DEST_IP="198.18.0.42" || DEST_IP="FD00::1" +fi [ -z "$CLONE_SKB" ] && CLONE_SKB="0" # Example enforce param "-m" for dst_mac [ -z "$DST_MAC" ] && usage && err 2 "Must specify -m dst_mac" @@ -54,7 +56,7 @@ pg_set $DEV "flag NO_TIMESTAMP" # Destination pg_set $DEV "dst_mac $DST_MAC" -pg_set $DEV "dst $DEST_IP" +pg_set $DEV "dst$IP6 $DEST_IP" # Setup random UDP port src range pg_set $DEV "flag UDPSRC_RND" diff --git a/samples/pktgen/pktgen_sample02_multiqueue.sh b/samples/pktgen/pktgen_sample02_multiqueue.sh index 32467ae..c88a161 100755 --- a/samples/pktgen/pktgen_sample02_multiqueue.sh +++ b/samples/pktgen/pktgen_sample02_multiqueue.sh @@ -23,7 +23,9 @@ UDP_MIN=9 UDP_MAX=109 # (example of setting default params in your script) -[ -z "$DEST_IP" ] && DEST_IP="198.18.0.42" +if [ -z "$DEST_IP" ]; then + [ -z "$IP6" ] && DEST_IP="198.18.0.42" || DEST_IP="FD00::1" +fi [ -z "$DST_MAC" ] && DST_MAC="90:e2:ba:ff:ff:ff" # General cleanup everything since last run @@ -54,7 +56,7 @@ for ((thread = 0; thread < $THREADS; thread++)); do # Destination pg_set $dev "dst_mac $DST_MAC" - pg_set $dev "dst $DEST_IP" + pg_set $dev "dst$IP6 $DEST_IP" # Setup random UDP port src range pg_set $dev "flag UDPSRC_RND" diff --git a/samples/pktgen/pktgen_sample03_burst_single_flow.sh b/samples/pktgen/pktgen_sample03_burst_single_flow.sh index 775f5d0..80cf8f5 100755 --- a/samples/pktgen/pktgen_sample03_burst_single_flow.sh +++ b/samples/pktgen/pktgen_sample03_burst_single_flow.sh @@ -25,7 +25,9 @@ root_check_run_with_sudo "$@" # Parameter parsing via include source ${basedir}/parameters.sh # Set some default params, if they didn't get set -[ -z "$DEST_IP" ] && DEST_IP="198.18.0.42" +if [ -z "$DEST_IP" ]; then + [ -z "$IP6" ] && DEST_IP="198.18.0.42" || DEST_IP="FD00::1" +fi [ -z "$DST_MAC" ] && DST_MAC="90:e2:ba:ff:ff:ff" [ -z "$BURST" ] && BURST=32 [ -z "$CLONE_SKB" ] && CLONE_SKB="100000" @@ -55,7 +57,7 @@ for ((thread = 0; thread < $THREADS; thread++)); do # Destination pg_set $dev "dst_mac $DST_MAC" - pg_set $dev "dst $DEST_IP" + pg_set $dev "dst$IP6 $DEST_IP" # Setup burst, for easy testing -b 0 disable bursting # (internally in pktgen default and minimum burst=1) -- cgit v0.10.2 From 5f652bb2eb3eb38f97194ce5c41da1fa12e914b8 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Wed, 20 Jul 2016 18:11:31 +0200 Subject: gro_cells: gro_cells_receive now return error code so that the caller can update stats accordingly, if needed Signed-off-by: Paolo Abeni Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/include/net/gro_cells.h b/include/net/gro_cells.h index cf6c745..d15214d 100644 --- a/include/net/gro_cells.h +++ b/include/net/gro_cells.h @@ -14,27 +14,26 @@ struct gro_cells { struct gro_cell __percpu *cells; }; -static inline void gro_cells_receive(struct gro_cells *gcells, struct sk_buff *skb) +static inline int gro_cells_receive(struct gro_cells *gcells, struct sk_buff *skb) { struct gro_cell *cell; struct net_device *dev = skb->dev; - if (!gcells->cells || skb_cloned(skb) || !(dev->features & NETIF_F_GRO)) { - netif_rx(skb); - return; - } + if (!gcells->cells || skb_cloned(skb) || !(dev->features & NETIF_F_GRO)) + return netif_rx(skb); cell = this_cpu_ptr(gcells->cells); if (skb_queue_len(&cell->napi_skbs) > netdev_max_backlog) { atomic_long_inc(&dev->rx_dropped); kfree_skb(skb); - return; + return NET_RX_DROP; } __skb_queue_tail(&cell->napi_skbs, skb); if (skb_queue_len(&cell->napi_skbs) == 1) napi_schedule(&cell->napi); + return NET_RX_SUCCESS; } /* called under BH context */ -- cgit v0.10.2 From 5491e7c6b1a95df39b917e16f2eeddc84f9e8491 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Wed, 20 Jul 2016 18:11:32 +0200 Subject: macsec: enable GRO and RPS on macsec devices Use gro_gells to trigger GRO and allow RPS on macsec traffic after decryption. Also, be sure to avoid clearing software offload features in macsec_fix_features(). Overall this increase TCP tput by 30% on recent h/w. Signed-off-by: Paolo Abeni Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 8bcd78f..0cbb935 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -18,6 +18,7 @@ #include #include #include +#include #include @@ -268,6 +269,7 @@ struct macsec_dev { struct net_device *real_dev; struct pcpu_secy_stats __percpu *stats; struct list_head secys; + struct gro_cells gro_cells; }; /** @@ -879,7 +881,7 @@ static void macsec_decrypt_done(struct crypto_async_request *base, int err) macsec_reset_skb(skb, macsec->secy.netdev); len = skb->len; - ret = netif_rx(skb); + ret = gro_cells_receive(&macsec->gro_cells, skb); if (ret == NET_RX_SUCCESS) count_rx(dev, len); else @@ -1052,6 +1054,7 @@ static rx_handler_result_t macsec_handle_frame(struct sk_buff **pskb) struct pcpu_rx_sc_stats *rxsc_stats; struct pcpu_secy_stats *secy_stats; bool pulled_sci; + int ret; if (skb_headroom(skb) < ETH_HLEN) goto drop_direct; @@ -1193,12 +1196,17 @@ deliver: if (rx_sa) macsec_rxsa_put(rx_sa); - count_rx(dev, skb->len); + + ret = gro_cells_receive(&macsec->gro_cells, skb); + if (ret == NET_RX_SUCCESS) + count_rx(dev, skb->len); + else + macsec->secy.netdev->stats.rx_dropped++; rcu_read_unlock(); - *pskb = skb; - return RX_HANDLER_ANOTHER; + *pskb = NULL; + return RX_HANDLER_CONSUMED; drop: macsec_rxsa_put(rx_sa); @@ -1218,7 +1226,6 @@ nosci: list_for_each_entry_rcu(macsec, &rxd->secys, secys) { struct sk_buff *nskb; - int ret; secy_stats = this_cpu_ptr(macsec->stats); @@ -2675,11 +2682,18 @@ static int macsec_dev_init(struct net_device *dev) { struct macsec_dev *macsec = macsec_priv(dev); struct net_device *real_dev = macsec->real_dev; + int err; dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); if (!dev->tstats) return -ENOMEM; + err = gro_cells_init(&macsec->gro_cells, dev); + if (err) { + free_percpu(dev->tstats); + return err; + } + dev->features = real_dev->features & MACSEC_FEATURES; dev->features |= NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE; @@ -2698,6 +2712,9 @@ static int macsec_dev_init(struct net_device *dev) static void macsec_dev_uninit(struct net_device *dev) { + struct macsec_dev *macsec = macsec_priv(dev); + + gro_cells_destroy(&macsec->gro_cells); free_percpu(dev->tstats); } @@ -2707,8 +2724,9 @@ static netdev_features_t macsec_fix_features(struct net_device *dev, struct macsec_dev *macsec = macsec_priv(dev); struct net_device *real_dev = macsec->real_dev; - features &= real_dev->features & MACSEC_FEATURES; - features |= NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE; + features &= (real_dev->features & MACSEC_FEATURES) | + NETIF_F_GSO_SOFTWARE | NETIF_F_SOFT_FEATURES; + features |= NETIF_F_LLTX; return features; } -- cgit v0.10.2 From cd956722167ba4fdba9c1ce3eed251b04ea2e10f Mon Sep 17 00:00:00 2001 From: Heinrich Schuchardt Date: Tue, 17 May 2016 22:41:33 +0200 Subject: i40e: avoid null pointer dereference In function i40e_debug_aq parameter desc is assumed to be possibly NULL. Do not dereference it before checking the value. Fixes: f905dd62be88 ("i40e/i40evf: add max buf len to aq debug print helper") Signed-off-by: Heinrich Schuchardt Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c index e447dc4..bb63df8 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_common.c +++ b/drivers/net/ethernet/intel/i40e/i40e_common.c @@ -297,13 +297,15 @@ void i40e_debug_aq(struct i40e_hw *hw, enum i40e_debug_mask mask, void *desc, void *buffer, u16 buf_len) { struct i40e_aq_desc *aq_desc = (struct i40e_aq_desc *)desc; - u16 len = le16_to_cpu(aq_desc->datalen); + u16 len; u8 *buf = (u8 *)buffer; u16 i = 0; if ((!(mask & hw->debug_mask)) || (desc == NULL)) return; + len = le16_to_cpu(aq_desc->datalen); + i40e_debug(hw, mask, "AQ CMD: opcode 0x%04X, flags 0x%04X, datalen 0x%04X, retval 0x%04X\n", le16_to_cpu(aq_desc->opcode), -- cgit v0.10.2 From 2853704f112836387b7dab319b1e92b576666056 Mon Sep 17 00:00:00 2001 From: Avinash Dayanand Date: Mon, 20 Jun 2016 09:10:33 -0700 Subject: i40e: Fix to show correct Advertised Link Modes when link is down When link is down, Advertised Link Modes was wrongly displaying full supported link modes instead of Advertised link mode. Added conditional checks in order to make sure correct Advertised link modes are displayed when the link is down. Change-ID: I8a61413f9ee174149c7a33157b5f0b0a8da9842d Signed-off-by: Avinash Dayanand Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 4962e85..0cfde30 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -272,15 +272,16 @@ static void i40e_phy_type_to_ethtool(struct i40e_pf *pf, u32 *supported, u32 *advertising) { enum i40e_aq_capabilities_phy_type phy_types = pf->hw.phy.phy_types; - + struct i40e_link_status *hw_link_info = &pf->hw.phy.link_info; *supported = 0x0; *advertising = 0x0; if (phy_types & I40E_CAP_PHY_TYPE_SGMII) { *supported |= SUPPORTED_Autoneg | SUPPORTED_1000baseT_Full; - *advertising |= ADVERTISED_Autoneg | - ADVERTISED_1000baseT_Full; + *advertising |= ADVERTISED_Autoneg; + if (hw_link_info->requested_speeds & I40E_LINK_SPEED_1GB) + *advertising |= ADVERTISED_1000baseT_Full; if (pf->flags & I40E_FLAG_100M_SGMII_CAPABLE) { *supported |= SUPPORTED_100baseT_Full; *advertising |= ADVERTISED_100baseT_Full; @@ -299,8 +300,9 @@ static void i40e_phy_type_to_ethtool(struct i40e_pf *pf, u32 *supported, phy_types & I40E_CAP_PHY_TYPE_10GBASE_LR) { *supported |= SUPPORTED_Autoneg | SUPPORTED_10000baseT_Full; - *advertising |= ADVERTISED_Autoneg | - ADVERTISED_10000baseT_Full; + *advertising |= ADVERTISED_Autoneg; + if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB) + *advertising |= ADVERTISED_10000baseT_Full; } if (phy_types & I40E_CAP_PHY_TYPE_XLAUI || phy_types & I40E_CAP_PHY_TYPE_XLPPI || @@ -310,14 +312,16 @@ static void i40e_phy_type_to_ethtool(struct i40e_pf *pf, u32 *supported, phy_types & I40E_CAP_PHY_TYPE_40GBASE_CR4) { *supported |= SUPPORTED_Autoneg | SUPPORTED_40000baseCR4_Full; - *advertising |= ADVERTISED_Autoneg | - ADVERTISED_40000baseCR4_Full; + *advertising |= ADVERTISED_Autoneg; + if (hw_link_info->requested_speeds & I40E_LINK_SPEED_40GB) + *advertising |= ADVERTISED_40000baseCR4_Full; } if (phy_types & I40E_CAP_PHY_TYPE_100BASE_TX) { *supported |= SUPPORTED_Autoneg | SUPPORTED_100baseT_Full; - *advertising |= ADVERTISED_Autoneg | - ADVERTISED_100baseT_Full; + *advertising |= ADVERTISED_Autoneg; + if (hw_link_info->requested_speeds & I40E_LINK_SPEED_100MB) + *advertising |= ADVERTISED_100baseT_Full; } if (phy_types & I40E_CAP_PHY_TYPE_1000BASE_T || phy_types & I40E_CAP_PHY_TYPE_1000BASE_SX || @@ -325,8 +329,9 @@ static void i40e_phy_type_to_ethtool(struct i40e_pf *pf, u32 *supported, phy_types & I40E_CAP_PHY_TYPE_1000BASE_T_OPTICAL) { *supported |= SUPPORTED_Autoneg | SUPPORTED_1000baseT_Full; - *advertising |= ADVERTISED_Autoneg | - ADVERTISED_1000baseT_Full; + *advertising |= ADVERTISED_Autoneg; + if (hw_link_info->requested_speeds & I40E_LINK_SPEED_1GB) + *advertising |= ADVERTISED_1000baseT_Full; } if (phy_types & I40E_CAP_PHY_TYPE_40GBASE_SR4) *supported |= SUPPORTED_40000baseSR4_Full; @@ -341,26 +346,30 @@ static void i40e_phy_type_to_ethtool(struct i40e_pf *pf, u32 *supported, if (phy_types & I40E_CAP_PHY_TYPE_20GBASE_KR2) { *supported |= SUPPORTED_20000baseKR2_Full | SUPPORTED_Autoneg; - *advertising |= ADVERTISED_20000baseKR2_Full | - ADVERTISED_Autoneg; + *advertising |= ADVERTISED_Autoneg; + if (hw_link_info->requested_speeds & I40E_LINK_SPEED_20GB) + *advertising |= ADVERTISED_20000baseKR2_Full; } if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_KR) { *supported |= SUPPORTED_10000baseKR_Full | SUPPORTED_Autoneg; - *advertising |= ADVERTISED_10000baseKR_Full | - ADVERTISED_Autoneg; + *advertising |= ADVERTISED_Autoneg; + if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB) + *advertising |= ADVERTISED_10000baseKR_Full; } if (phy_types & I40E_CAP_PHY_TYPE_10GBASE_KX4) { *supported |= SUPPORTED_10000baseKX4_Full | SUPPORTED_Autoneg; - *advertising |= ADVERTISED_10000baseKX4_Full | - ADVERTISED_Autoneg; + *advertising |= ADVERTISED_Autoneg; + if (hw_link_info->requested_speeds & I40E_LINK_SPEED_10GB) + *advertising |= ADVERTISED_10000baseKX4_Full; } if (phy_types & I40E_CAP_PHY_TYPE_1000BASE_KX) { *supported |= SUPPORTED_1000baseKX_Full | SUPPORTED_Autoneg; - *advertising |= ADVERTISED_1000baseKX_Full | - ADVERTISED_Autoneg; + *advertising |= ADVERTISED_Autoneg; + if (hw_link_info->requested_speeds & I40E_LINK_SPEED_1GB) + *advertising |= ADVERTISED_1000baseKX_Full; } } -- cgit v0.10.2 From 88dc9e6fed5b6cda0049849abf76929df3178158 Mon Sep 17 00:00:00 2001 From: Mitch Williams Date: Mon, 20 Jun 2016 09:10:35 -0700 Subject: i40e/i40evf: remove useless initializer This initializer isn't needed because the variable is assigned right away. Change-ID: I6ce3edb3f4e0364db248a7a0bcc62ca95c01d941 Signed-off-by: Mitch Williams Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c index 55f151f..2def562 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c @@ -740,14 +740,12 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi, tx_ring->q_vector->tx.total_packets += total_packets; if (tx_ring->flags & I40E_TXR_FLAGS_WB_ON_ITR) { - unsigned int j = 0; - /* check to see if there are < 4 descriptors * waiting to be written back, then kick the hardware to force * them to be written back in case we stay in NAPI. * In this mode on X722 we do not enable Interrupt. */ - j = i40e_get_tx_pending(tx_ring, false); + unsigned int j = i40e_get_tx_pending(tx_ring, false); if (budget && ((j / (WB_STRIDE + 1)) == 0) && (j != 0) && diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c index be99189..40f224f 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c @@ -259,13 +259,12 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi, tx_ring->q_vector->tx.total_packets += total_packets; if (tx_ring->flags & I40E_TXR_FLAGS_WB_ON_ITR) { - unsigned int j = 0; /* check to see if there are < 4 descriptors * waiting to be written back, then kick the hardware to force * them to be written back in case we stay in NAPI. * In this mode on X722 we do not enable Interrupt. */ - j = i40evf_get_tx_pending(tx_ring, false); + unsigned int j = i40evf_get_tx_pending(tx_ring, false); if (budget && ((j / (WB_STRIDE + 1)) == 0) && (j > 0) && -- cgit v0.10.2 From 8b60151029a16f6e59d2c927d7a846a2b8e9cc88 Mon Sep 17 00:00:00 2001 From: Catherine Sullivan Date: Mon, 20 Jun 2016 09:10:36 -0700 Subject: i40e: Remove device ID 0x37D4 This device ID is not needed, so take it out. Change-ID: I148d29f68a1f58b03980ecd83047a1b440f4f74d Signed-off-by: Catherine Sullivan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c index bb63df8..2154a34 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_common.c +++ b/drivers/net/ethernet/intel/i40e/i40e_common.c @@ -61,7 +61,6 @@ static i40e_status i40e_set_mac_type(struct i40e_hw *hw) case I40E_DEV_ID_1G_BASE_T_X722: case I40E_DEV_ID_10G_BASE_T_X722: case I40E_DEV_ID_SFP_I_X722: - case I40E_DEV_ID_QSFP_I_X722: hw->mac.type = I40E_MAC_X722; break; default: diff --git a/drivers/net/ethernet/intel/i40e/i40e_devids.h b/drivers/net/ethernet/intel/i40e/i40e_devids.h index d701861..dd4457d 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_devids.h +++ b/drivers/net/ethernet/intel/i40e/i40e_devids.h @@ -45,7 +45,6 @@ #define I40E_DEV_ID_1G_BASE_T_X722 0x37D1 #define I40E_DEV_ID_10G_BASE_T_X722 0x37D2 #define I40E_DEV_ID_SFP_I_X722 0x37D3 -#define I40E_DEV_ID_QSFP_I_X722 0x37D4 #define i40e_is_40G_device(d) ((d) == I40E_DEV_ID_QSFP_A || \ (d) == I40E_DEV_ID_QSFP_B || \ diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 2b11405..28206da 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -86,7 +86,6 @@ static const struct pci_device_id i40e_pci_tbl[] = { {PCI_VDEVICE(INTEL, I40E_DEV_ID_1G_BASE_T_X722), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T_X722), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_I_X722), 0}, - {PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_I_X722), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2_A), 0}, /* required last entry */ diff --git a/drivers/net/ethernet/intel/i40evf/i40e_common.c b/drivers/net/ethernet/intel/i40evf/i40e_common.c index 8f64204..4db0c03 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_common.c +++ b/drivers/net/ethernet/intel/i40evf/i40e_common.c @@ -59,7 +59,6 @@ i40e_status i40e_set_mac_type(struct i40e_hw *hw) case I40E_DEV_ID_1G_BASE_T_X722: case I40E_DEV_ID_10G_BASE_T_X722: case I40E_DEV_ID_SFP_I_X722: - case I40E_DEV_ID_QSFP_I_X722: hw->mac.type = I40E_MAC_X722; break; case I40E_DEV_ID_X722_VF: diff --git a/drivers/net/ethernet/intel/i40evf/i40e_devids.h b/drivers/net/ethernet/intel/i40evf/i40e_devids.h index d34972b..7023570 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_devids.h +++ b/drivers/net/ethernet/intel/i40evf/i40e_devids.h @@ -45,7 +45,6 @@ #define I40E_DEV_ID_1G_BASE_T_X722 0x37D1 #define I40E_DEV_ID_10G_BASE_T_X722 0x37D2 #define I40E_DEV_ID_SFP_I_X722 0x37D3 -#define I40E_DEV_ID_QSFP_I_X722 0x37D4 #define I40E_DEV_ID_X722_VF 0x37CD #define I40E_DEV_ID_X722_VF_HV 0x37D9 -- cgit v0.10.2 From 9287141292aa5a4a80859436f45f828bb10ea0f9 Mon Sep 17 00:00:00 2001 From: Joshua Hay Date: Mon, 20 Jun 2016 09:10:37 -0700 Subject: i40evf: add hyperv dev ids This patch adds the Hyper-V specific VF device ids. Change-ID: I9c4fe6d8dfd34f7f68ebc9fdae225c8768439c89 Signed-off-by: Joshua Hay Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c index eac057b..b12edba 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c @@ -57,7 +57,9 @@ static const char i40evf_copyright[] = */ static const struct pci_device_id i40evf_pci_tbl[] = { {PCI_VDEVICE(INTEL, I40E_DEV_ID_VF), 0}, + {PCI_VDEVICE(INTEL, I40E_DEV_ID_VF_HV), 0}, {PCI_VDEVICE(INTEL, I40E_DEV_ID_X722_VF), 0}, + {PCI_VDEVICE(INTEL, I40E_DEV_ID_X722_VF_HV), 0}, /* required last entry */ {0, } }; -- cgit v0.10.2 From c3c7ea27bec070f46dc95dbfafc032eb1b549c6b Mon Sep 17 00:00:00 2001 From: Mitch Williams Date: Mon, 20 Jun 2016 09:10:38 -0700 Subject: i40e: refactor Rx filter handling Properly track filter adds and deletes so the driver doesn't lose filters during resets and up/down cycles. Add a tracking mechanism so that the driver knows when to enter and leave promiscuous mode. Implement a simple state machine so the driver can track the status of each filter throughout its lifecycle. Properly manage the overflow promiscuous state for the each VSI, and provide a way for the driver to detect when to exit overflow promiscuous mode. Remove all possible default MAC filters that the firmware may have set up so that the driver can manage these correctly, particularly when VLANs come into play. Remove the LAA flag for filters; instead just send whatever we get through set_mac to the firmware as the LAA for wakeup purposes. Finally, add the state of each filter to debugfs output so we can see what's going on inside the driver's pointy little head. Change-ID: I97c5e366fac2254fa01eaff4f65c0af61dcf2e1f Signed-off-by: Mitch Williams Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h index e83fc8a..2a88291 100644 --- a/drivers/net/ethernet/intel/i40e/i40e.h +++ b/drivers/net/ethernet/intel/i40e/i40e.h @@ -448,6 +448,14 @@ struct i40e_pf { u16 phy_led_val; }; +enum i40e_filter_state { + I40E_FILTER_INVALID = 0, /* Invalid state */ + I40E_FILTER_NEW, /* New, not sent to FW yet */ + I40E_FILTER_ACTIVE, /* Added to switch by FW */ + I40E_FILTER_FAILED, /* Rejected by FW */ + I40E_FILTER_REMOVE, /* To be removed */ +/* There is no 'removed' state; the filter struct is freed */ +}; struct i40e_mac_filter { struct list_head list; u8 macaddr[ETH_ALEN]; @@ -456,8 +464,7 @@ struct i40e_mac_filter { u8 counter; /* number of instances of this filter */ bool is_vf; /* filter belongs to a VF */ bool is_netdev; /* filter belongs to a netdev */ - bool changed; /* filter needs to be sync'd to the HW */ - bool is_laa; /* filter is a Locally Administered Address */ + enum i40e_filter_state state; }; struct i40e_veb { @@ -523,6 +530,9 @@ struct i40e_vsi { struct i40e_ring **rx_rings; struct i40e_ring **tx_rings; + u32 active_filters; + u32 promisc_threshold; + u16 work_limit; u16 int_rate_limit; /* value in usecs */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c index e6af8c8..05cf9a7 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c +++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c @@ -116,6 +116,14 @@ static ssize_t i40e_dbg_command_read(struct file *filp, char __user *buffer, return len; } +static char *i40e_filter_state_string[] = { + "INVALID", + "NEW", + "ACTIVE", + "FAILED", + "REMOVE", +}; + /** * i40e_dbg_dump_vsi_seid - handles dump vsi seid write into command datum * @pf: the i40e_pf created in command write @@ -160,10 +168,14 @@ static void i40e_dbg_dump_vsi_seid(struct i40e_pf *pf, int seid) pf->hw.mac.port_addr); list_for_each_entry(f, &vsi->mac_filter_list, list) { dev_info(&pf->pdev->dev, - " mac_filter_list: %pM vid=%d, is_netdev=%d is_vf=%d counter=%d\n", + " mac_filter_list: %pM vid=%d, is_netdev=%d is_vf=%d counter=%d, state %s\n", f->macaddr, f->vlan, f->is_netdev, f->is_vf, - f->counter); + f->counter, i40e_filter_state_string[f->state]); } + dev_info(&pf->pdev->dev, " active_filters %d, promisc_threshold %d, overflow promisc %s\n", + vsi->active_filters, vsi->promisc_threshold, + (test_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state) ? + "ON" : "OFF")); nstat = i40e_get_vsi_stats_struct(vsi); dev_info(&pf->pdev->dev, " net_stats: rx_packets = %lu, rx_bytes = %lu, rx_errors = %lu, rx_dropped = %lu\n", diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 28206da..bdaebb6 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -1274,8 +1274,9 @@ int i40e_del_mac_all_vlan(struct i40e_vsi *vsi, u8 *macaddr, (is_vf == f->is_vf) && (is_netdev == f->is_netdev)) { f->counter--; - f->changed = true; changed = 1; + if (f->counter == 0) + f->state = I40E_FILTER_REMOVE; } } if (changed) { @@ -1291,29 +1292,32 @@ int i40e_del_mac_all_vlan(struct i40e_vsi *vsi, u8 *macaddr, * @vsi: the PF Main VSI - inappropriate for any other VSI * @macaddr: the MAC address * - * Some older firmware configurations set up a default promiscuous VLAN - * filter that needs to be removed. + * Remove whatever filter the firmware set up so the driver can manage + * its own filtering intelligently. **/ -static int i40e_rm_default_mac_filter(struct i40e_vsi *vsi, u8 *macaddr) +static void i40e_rm_default_mac_filter(struct i40e_vsi *vsi, u8 *macaddr) { struct i40e_aqc_remove_macvlan_element_data element; struct i40e_pf *pf = vsi->back; - i40e_status ret; /* Only appropriate for the PF main VSI */ if (vsi->type != I40E_VSI_MAIN) - return -EINVAL; + return; memset(&element, 0, sizeof(element)); ether_addr_copy(element.mac_addr, macaddr); element.vlan_tag = 0; - element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH | - I40E_AQC_MACVLAN_DEL_IGNORE_VLAN; - ret = i40e_aq_remove_macvlan(&pf->hw, vsi->seid, &element, 1, NULL); - if (ret) - return -ENOENT; + /* Ignore error returns, some firmware does it this way... */ + element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH; + i40e_aq_remove_macvlan(&pf->hw, vsi->seid, &element, 1, NULL); - return 0; + memset(&element, 0, sizeof(element)); + ether_addr_copy(element.mac_addr, macaddr); + element.vlan_tag = 0; + /* ...and some firmware does it this way. */ + element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH | + I40E_AQC_MACVLAN_ADD_IGNORE_VLAN; + i40e_aq_remove_macvlan(&pf->hw, vsi->seid, &element, 1, NULL); } /** @@ -1334,6 +1338,7 @@ struct i40e_mac_filter *i40e_add_filter(struct i40e_vsi *vsi, bool is_vf, bool is_netdev) { struct i40e_mac_filter *f; + int changed = false; if (!vsi || !macaddr) return NULL; @@ -1346,8 +1351,15 @@ struct i40e_mac_filter *i40e_add_filter(struct i40e_vsi *vsi, ether_addr_copy(f->macaddr, macaddr); f->vlan = vlan; - f->changed = true; - + /* If we're in overflow promisc mode, set the state directly + * to failed, so we don't bother to try sending the filter + * to the hardware. + */ + if (test_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state)) + f->state = I40E_FILTER_FAILED; + else + f->state = I40E_FILTER_NEW; + changed = true; INIT_LIST_HEAD(&f->list); list_add_tail(&f->list, &vsi->mac_filter_list); } @@ -1367,10 +1379,7 @@ struct i40e_mac_filter *i40e_add_filter(struct i40e_vsi *vsi, f->counter++; } - /* changed tells sync_filters_subtask to - * push the filter down to the firmware - */ - if (f->changed) { + if (changed) { vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED; vsi->back->flags |= I40E_FLAG_FILTER_SYNC; } @@ -1389,6 +1398,9 @@ add_filter_out: * * NOTE: This function is expected to be called with mac_filter_list_lock * being held. + * ANOTHER NOTE: This function MUST be called from within the context of + * the "safe" variants of any list iterators, e.g. list_for_each_entry_safe() + * instead of list_for_each_entry(). **/ void i40e_del_filter(struct i40e_vsi *vsi, u8 *macaddr, s16 vlan, @@ -1428,9 +1440,18 @@ void i40e_del_filter(struct i40e_vsi *vsi, * remove the filter from the firmware's list */ if (f->counter == 0) { - f->changed = true; - vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED; - vsi->back->flags |= I40E_FLAG_FILTER_SYNC; + if ((f->state == I40E_FILTER_FAILED) || + (f->state == I40E_FILTER_NEW)) { + /* this one never got added by the FW. Just remove it, + * no need to sync anything. + */ + list_del(&f->list); + kfree(f); + } else { + f->state = I40E_FILTER_REMOVE; + vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED; + vsi->back->flags |= I40E_FLAG_FILTER_SYNC; + } } } @@ -1452,7 +1473,6 @@ static int i40e_set_mac(struct net_device *netdev, void *p) struct i40e_pf *pf = vsi->back; struct i40e_hw *hw = &pf->hw; struct sockaddr *addr = p; - struct i40e_mac_filter *f; if (!is_valid_ether_addr(addr->sa_data)) return -EADDRNOTAVAIL; @@ -1473,52 +1493,23 @@ static int i40e_set_mac(struct net_device *netdev, void *p) else netdev_info(netdev, "set new mac address %pM\n", addr->sa_data); + spin_lock_bh(&vsi->mac_filter_list_lock); + i40e_del_mac_all_vlan(vsi, netdev->dev_addr, false, true); + i40e_put_mac_in_vlan(vsi, addr->sa_data, false, true); + spin_unlock_bh(&vsi->mac_filter_list_lock); + ether_addr_copy(netdev->dev_addr, addr->sa_data); if (vsi->type == I40E_VSI_MAIN) { i40e_status ret; ret = i40e_aq_mac_address_write(&vsi->back->hw, I40E_AQC_WRITE_TYPE_LAA_WOL, addr->sa_data, NULL); - if (ret) { - netdev_info(netdev, - "Addr change for Main VSI failed: %d\n", - ret); - return -EADDRNOTAVAIL; - } - } - - if (ether_addr_equal(netdev->dev_addr, hw->mac.addr)) { - struct i40e_aqc_remove_macvlan_element_data element; - - memset(&element, 0, sizeof(element)); - ether_addr_copy(element.mac_addr, netdev->dev_addr); - element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH; - i40e_aq_remove_macvlan(&pf->hw, vsi->seid, &element, 1, NULL); - } else { - spin_lock_bh(&vsi->mac_filter_list_lock); - i40e_del_filter(vsi, netdev->dev_addr, I40E_VLAN_ANY, - false, false); - spin_unlock_bh(&vsi->mac_filter_list_lock); - } - - if (ether_addr_equal(addr->sa_data, hw->mac.addr)) { - struct i40e_aqc_add_macvlan_element_data element; - - memset(&element, 0, sizeof(element)); - ether_addr_copy(element.mac_addr, hw->mac.addr); - element.flags = cpu_to_le16(I40E_AQC_MACVLAN_ADD_PERFECT_MATCH); - i40e_aq_add_macvlan(&pf->hw, vsi->seid, &element, 1, NULL); - } else { - spin_lock_bh(&vsi->mac_filter_list_lock); - f = i40e_add_filter(vsi, addr->sa_data, I40E_VLAN_ANY, - false, false); - if (f) - f->is_laa = true; - spin_unlock_bh(&vsi->mac_filter_list_lock); + if (ret) + netdev_info(netdev, "Ignoring error from firmware on LAA update, status %s, AQ ret %s\n", + i40e_stat_str(hw, ret), + i40e_aq_str(hw, hw->aq.asq_last_status)); } - ether_addr_copy(netdev->dev_addr, addr->sa_data); - /* schedule our worker thread which will take care of * applying the new filter changes */ @@ -1749,28 +1740,6 @@ bottom_of_search_loop: } /** - * i40e_mac_filter_entry_clone - Clones a MAC filter entry - * @src: source MAC filter entry to be clones - * - * Returns the pointer to newly cloned MAC filter entry or NULL - * in case of error - **/ -static struct i40e_mac_filter *i40e_mac_filter_entry_clone( - struct i40e_mac_filter *src) -{ - struct i40e_mac_filter *f; - - f = kzalloc(sizeof(*f), GFP_ATOMIC); - if (!f) - return NULL; - *f = *src; - - INIT_LIST_HEAD(&f->list); - - return f; -} - -/** * i40e_undo_del_filter_entries - Undo the changes made to MAC filter entries * @vsi: pointer to vsi struct * @from: Pointer to list which contains MAC filter entries - changes to @@ -1784,41 +1753,61 @@ static void i40e_undo_del_filter_entries(struct i40e_vsi *vsi, struct i40e_mac_filter *f, *ftmp; list_for_each_entry_safe(f, ftmp, from, list) { - f->changed = true; /* Move the element back into MAC filter list*/ list_move_tail(&f->list, &vsi->mac_filter_list); } } /** - * i40e_undo_add_filter_entries - Undo the changes made to MAC filter entries - * @vsi: pointer to vsi struct + * i40e_update_filter_state - Update filter state based on return data + * from firmware + * @count: Number of filters added + * @add_list: return data from fw + * @head: pointer to first filter in current batch + * @aq_err: status from fw * - * MAC filter entries from list were slated to be added from device. + * MAC filter entries from list were slated to be added to device. Returns + * number of successful filters. Note that 0 does NOT mean success! **/ -static void i40e_undo_add_filter_entries(struct i40e_vsi *vsi) +static int +i40e_update_filter_state(int count, + struct i40e_aqc_add_macvlan_element_data *add_list, + struct i40e_mac_filter *add_head, int aq_err) { - struct i40e_mac_filter *f, *ftmp; - - list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { - if (!f->changed && f->counter) - f->changed = true; - } -} + int retval = 0; + int i; -/** - * i40e_cleanup_add_list - Deletes the element from add list and release - * memory - * @add_list: Pointer to list which contains MAC filter entries - **/ -static void i40e_cleanup_add_list(struct list_head *add_list) -{ - struct i40e_mac_filter *f, *ftmp; - list_for_each_entry_safe(f, ftmp, add_list, list) { - list_del(&f->list); - kfree(f); + if (!aq_err) { + retval = count; + /* Everything's good, mark all filters active. */ + for (i = 0; i < count ; i++) { + add_head->state = I40E_FILTER_ACTIVE; + add_head = list_next_entry(add_head, list); + } + } else if (aq_err == I40E_AQ_RC_ENOSPC) { + /* Device ran out of filter space. Check the return value + * for each filter to see which ones are active. + */ + for (i = 0; i < count ; i++) { + if (add_list[i].match_method == + I40E_AQC_MM_ERR_NO_RES) { + add_head->state = I40E_FILTER_FAILED; + } else { + add_head->state = I40E_FILTER_ACTIVE; + retval++; + } + add_head = list_next_entry(add_head, list); + } + } else { + /* Some other horrible thing happened, fail all filters */ + retval = 0; + for (i = 0; i < count ; i++) { + add_head->state = I40E_FILTER_FAILED; + add_head = list_next_entry(add_head, list); + } } + return retval; } /** @@ -1831,22 +1820,22 @@ static void i40e_cleanup_add_list(struct list_head *add_list) **/ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) { - struct list_head tmp_del_list, tmp_add_list; - struct i40e_mac_filter *f, *ftmp, *fclone; + struct i40e_mac_filter *f, *ftmp, *add_head = NULL; + struct list_head tmp_add_list, tmp_del_list; struct i40e_hw *hw = &vsi->back->hw; - bool promisc_forced_on = false; - bool add_happened = false; + bool promisc_changed = false; char vsi_name[16] = "PF"; int filter_list_len = 0; u32 changed_flags = 0; i40e_status aq_ret = 0; - bool err_cond = false; int retval = 0; struct i40e_pf *pf; int num_add = 0; int num_del = 0; int aq_err = 0; u16 cmd_flags; + int list_size; + int fcnt; /* empty array typed pointers, kcalloc later */ struct i40e_aqc_add_macvlan_element_data *add_list; @@ -1861,8 +1850,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) vsi->current_netdev_flags = vsi->netdev->flags; } - INIT_LIST_HEAD(&tmp_del_list); INIT_LIST_HEAD(&tmp_add_list); + INIT_LIST_HEAD(&tmp_del_list); if (vsi->type == I40E_VSI_SRIOV) snprintf(vsi_name, sizeof(vsi_name) - 1, "VF %d", vsi->vf_id); @@ -1873,65 +1862,34 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) vsi->flags &= ~I40E_VSI_FLAG_FILTER_CHANGED; spin_lock_bh(&vsi->mac_filter_list_lock); + /* Create a list of filters to delete. */ list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { - if (!f->changed) - continue; - - if (f->counter != 0) - continue; - f->changed = false; - - /* Move the element into temporary del_list */ - list_move_tail(&f->list, &tmp_del_list); - } - - list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { - if (!f->changed) - continue; - - if (f->counter == 0) - continue; - f->changed = false; - - /* Clone MAC filter entry and add into temporary list */ - fclone = i40e_mac_filter_entry_clone(f); - if (!fclone) { - err_cond = true; - break; + if (f->state == I40E_FILTER_REMOVE) { + WARN_ON(f->counter != 0); + /* Move the element into temporary del_list */ + list_move_tail(&f->list, &tmp_del_list); + vsi->active_filters--; + } + if (f->state == I40E_FILTER_NEW) { + WARN_ON(f->counter == 0); + /* Move the element into temporary add_list */ + list_move_tail(&f->list, &tmp_add_list); } - list_add_tail(&fclone->list, &tmp_add_list); - } - - /* if failed to clone MAC filter entry - undo */ - if (err_cond) { - i40e_undo_del_filter_entries(vsi, &tmp_del_list); - i40e_undo_add_filter_entries(vsi); } spin_unlock_bh(&vsi->mac_filter_list_lock); - - if (err_cond) { - i40e_cleanup_add_list(&tmp_add_list); - retval = -ENOMEM; - goto out; - } } /* Now process 'del_list' outside the lock */ if (!list_empty(&tmp_del_list)) { - int del_list_size; - filter_list_len = hw->aq.asq_buf_size / sizeof(struct i40e_aqc_remove_macvlan_element_data); - del_list_size = filter_list_len * + list_size = filter_list_len * sizeof(struct i40e_aqc_remove_macvlan_element_data); - del_list = kzalloc(del_list_size, GFP_ATOMIC); + del_list = kzalloc(list_size, GFP_ATOMIC); if (!del_list) { - i40e_cleanup_add_list(&tmp_add_list); - /* Undo VSI's MAC filter entry element updates */ spin_lock_bh(&vsi->mac_filter_list_lock); i40e_undo_del_filter_entries(vsi, &tmp_del_list); - i40e_undo_add_filter_entries(vsi); spin_unlock_bh(&vsi->mac_filter_list_lock); retval = -ENOMEM; goto out; @@ -1942,9 +1900,13 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) /* add to delete list */ ether_addr_copy(del_list[num_del].mac_addr, f->macaddr); - del_list[num_del].vlan_tag = - cpu_to_le16((u16)(f->vlan == - I40E_VLAN_ANY ? 0 : f->vlan)); + if (f->vlan == I40E_VLAN_ANY) { + del_list[num_del].vlan_tag = 0; + cmd_flags |= I40E_AQC_MACVLAN_ADD_IGNORE_VLAN; + } else { + del_list[num_del].vlan_tag = + cpu_to_le16((u16)(f->vlan)); + } cmd_flags |= I40E_AQC_MACVLAN_DEL_PERFECT_MATCH; del_list[num_del].flags = cmd_flags; @@ -1952,18 +1914,20 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) /* flush a full buffer */ if (num_del == filter_list_len) { - aq_ret = - i40e_aq_remove_macvlan(hw, vsi->seid, - del_list, - num_del, NULL); + aq_ret = i40e_aq_remove_macvlan(hw, vsi->seid, + del_list, + num_del, NULL); aq_err = hw->aq.asq_last_status; num_del = 0; - memset(del_list, 0, del_list_size); + memset(del_list, 0, list_size); - if (aq_ret && aq_err != I40E_AQ_RC_ENOENT) { + /* Explicitly ignore and do not report when + * firmware returns ENOENT. + */ + if (aq_ret && !(aq_err == I40E_AQ_RC_ENOENT)) { retval = -EIO; - dev_err(&pf->pdev->dev, - "ignoring delete macvlan error on %s, err %s, aq_err %s while flushing a full buffer\n", + dev_info(&pf->pdev->dev, + "ignoring delete macvlan error on %s, err %s, aq_err %s\n", vsi_name, i40e_stat_str(hw, aq_ret), i40e_aq_str(hw, aq_err)); @@ -1982,12 +1946,17 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) aq_err = hw->aq.asq_last_status; num_del = 0; - if (aq_ret && aq_err != I40E_AQ_RC_ENOENT) + /* Explicitly ignore and do not report when firmware + * returns ENOENT. + */ + if (aq_ret && !(aq_err == I40E_AQ_RC_ENOENT)) { + retval = -EIO; dev_info(&pf->pdev->dev, "ignoring delete macvlan error on %s, err %s aq_err %s\n", vsi_name, i40e_stat_str(hw, aq_ret), i40e_aq_str(hw, aq_err)); + } } kfree(del_list); @@ -1995,38 +1964,36 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) } if (!list_empty(&tmp_add_list)) { - int add_list_size; - - /* do all the adds now */ + /* Do all the adds now. */ filter_list_len = hw->aq.asq_buf_size / - sizeof(struct i40e_aqc_add_macvlan_element_data), - add_list_size = filter_list_len * sizeof(struct i40e_aqc_add_macvlan_element_data); - add_list = kzalloc(add_list_size, GFP_ATOMIC); + list_size = filter_list_len * + sizeof(struct i40e_aqc_add_macvlan_element_data); + add_list = kzalloc(list_size, GFP_ATOMIC); if (!add_list) { - /* Purge element from temporary lists */ - i40e_cleanup_add_list(&tmp_add_list); - - /* Undo add filter entries from VSI MAC filter list */ - spin_lock_bh(&vsi->mac_filter_list_lock); - i40e_undo_add_filter_entries(vsi); - spin_unlock_bh(&vsi->mac_filter_list_lock); retval = -ENOMEM; goto out; } - - list_for_each_entry_safe(f, ftmp, &tmp_add_list, list) { - - add_happened = true; - cmd_flags = 0; - + num_add = 0; + list_for_each_entry(f, &tmp_add_list, list) { + if (test_bit(__I40E_FILTER_OVERFLOW_PROMISC, + &vsi->state)) { + f->state = I40E_FILTER_FAILED; + continue; + } /* add to add array */ + if (num_add == 0) + add_head = f; + cmd_flags = 0; ether_addr_copy(add_list[num_add].mac_addr, f->macaddr); - add_list[num_add].vlan_tag = - cpu_to_le16( - (u16)(f->vlan == I40E_VLAN_ANY ? 0 : f->vlan)); + if (f->vlan == I40E_VLAN_ANY) { + add_list[num_add].vlan_tag = 0; + cmd_flags |= I40E_AQC_MACVLAN_ADD_IGNORE_VLAN; + } else { + add_list[num_add].vlan_tag = + cpu_to_le16((u16)(f->vlan)); + } add_list[num_add].queue_number = 0; - cmd_flags |= I40E_AQC_MACVLAN_ADD_PERFECT_MATCH; add_list[num_add].flags = cpu_to_le16(cmd_flags); num_add++; @@ -2037,44 +2004,77 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) add_list, num_add, NULL); aq_err = hw->aq.asq_last_status; + fcnt = i40e_update_filter_state(num_add, + add_list, + add_head, + aq_ret); + vsi->active_filters += fcnt; + + if (fcnt != num_add) { + promisc_changed = true; + set_bit(__I40E_FILTER_OVERFLOW_PROMISC, + &vsi->state); + vsi->promisc_threshold = + (vsi->active_filters * 3) / 4; + dev_warn(&pf->pdev->dev, + "Error %s adding RX filters on %s, promiscuous mode forced on\n", + i40e_aq_str(hw, aq_err), + vsi_name); + } + memset(add_list, 0, list_size); num_add = 0; - - if (aq_ret) - break; - memset(add_list, 0, add_list_size); } - /* Entries from tmp_add_list were cloned from MAC - * filter list, hence clean those cloned entries - */ - list_del(&f->list); - kfree(f); } - if (num_add) { aq_ret = i40e_aq_add_macvlan(hw, vsi->seid, add_list, num_add, NULL); aq_err = hw->aq.asq_last_status; - num_add = 0; + fcnt = i40e_update_filter_state(num_add, add_list, + add_head, aq_ret); + vsi->active_filters += fcnt; + if (fcnt != num_add) { + promisc_changed = true; + set_bit(__I40E_FILTER_OVERFLOW_PROMISC, + &vsi->state); + vsi->promisc_threshold = + (vsi->active_filters * 3) / 4; + dev_warn(&pf->pdev->dev, + "Error %s adding RX filters on %s, promiscuous mode forced on\n", + i40e_aq_str(hw, aq_err), vsi_name); + } } + /* Now move all of the filters from the temp add list back to + * the VSI's list. + */ + spin_lock_bh(&vsi->mac_filter_list_lock); + list_for_each_entry_safe(f, ftmp, &tmp_add_list, list) { + list_move_tail(&f->list, &vsi->mac_filter_list); + } + spin_unlock_bh(&vsi->mac_filter_list_lock); kfree(add_list); add_list = NULL; + } - if (add_happened && aq_ret && aq_err != I40E_AQ_RC_EINVAL) { - retval = i40e_aq_rc_to_posix(aq_ret, aq_err); + /* Check to see if we can drop out of overflow promiscuous mode. */ + if (test_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state) && + (vsi->active_filters < vsi->promisc_threshold)) { + int failed_count = 0; + /* See if we have any failed filters. We can't drop out of + * promiscuous until these have all been deleted. + */ + spin_lock_bh(&vsi->mac_filter_list_lock); + list_for_each_entry(f, &vsi->mac_filter_list, list) { + if (f->state == I40E_FILTER_FAILED) + failed_count++; + } + spin_unlock_bh(&vsi->mac_filter_list_lock); + if (!failed_count) { dev_info(&pf->pdev->dev, - "add filter failed on %s, err %s aq_err %s\n", - vsi_name, - i40e_stat_str(hw, aq_ret), - i40e_aq_str(hw, aq_err)); - if ((hw->aq.asq_last_status == I40E_AQ_RC_ENOSPC) && - !test_bit(__I40E_FILTER_OVERFLOW_PROMISC, - &vsi->state)) { - promisc_forced_on = true; - set_bit(__I40E_FILTER_OVERFLOW_PROMISC, - &vsi->state); - dev_info(&pf->pdev->dev, "promiscuous mode forced on %s\n", - vsi_name); - } + "filter logjam cleared on %s, leaving overflow promiscuous mode\n", + vsi_name); + clear_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state); + promisc_changed = true; + vsi->promisc_threshold = 0; } } @@ -2103,7 +2103,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi) i40e_aq_str(hw, hw->aq.asq_last_status)); } } - if ((changed_flags & IFF_PROMISC) || promisc_forced_on) { + if ((changed_flags & IFF_PROMISC) || + (promisc_changed && + test_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state))) { bool cur_promisc; cur_promisc = (!!(vsi->current_netdev_flags & IFF_PROMISC) || @@ -2352,7 +2354,7 @@ static void i40e_vlan_rx_register(struct net_device *netdev, u32 features) **/ int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid) { - struct i40e_mac_filter *f, *add_f; + struct i40e_mac_filter *f, *ftmp, *add_f; bool is_netdev, is_vf; is_vf = (vsi->type == I40E_VSI_SRIOV); @@ -2373,7 +2375,7 @@ int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid) } } - list_for_each_entry(f, &vsi->mac_filter_list, list) { + list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { add_f = i40e_add_filter(vsi, f->macaddr, vid, is_vf, is_netdev); if (!add_f) { dev_info(&vsi->back->pdev->dev, @@ -2387,7 +2389,7 @@ int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid) /* Now if we add a vlan tag, make sure to check if it is the first * tag (i.e. a "tag" -1 does exist) and if so replace the -1 "tag" * with 0, so we now accept untagged and specified tagged traffic - * (and not any taged and untagged) + * (and not all tags along with untagged) */ if (vid > 0) { if (is_netdev && i40e_find_filter(vsi, vsi->netdev->dev_addr, @@ -2409,7 +2411,7 @@ int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid) /* Do not assume that I40E_VLAN_ANY should be reset to VLAN 0 */ if (vid > 0 && !vsi->info.pvid) { - list_for_each_entry(f, &vsi->mac_filter_list, list) { + list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { if (!i40e_find_filter(vsi, f->macaddr, I40E_VLAN_ANY, is_vf, is_netdev)) continue; @@ -2446,7 +2448,7 @@ int i40e_vsi_add_vlan(struct i40e_vsi *vsi, s16 vid) int i40e_vsi_kill_vlan(struct i40e_vsi *vsi, s16 vid) { struct net_device *netdev = vsi->netdev; - struct i40e_mac_filter *f, *add_f; + struct i40e_mac_filter *f, *ftmp, *add_f; bool is_vf, is_netdev; int filter_count = 0; @@ -2459,7 +2461,7 @@ int i40e_vsi_kill_vlan(struct i40e_vsi *vsi, s16 vid) if (is_netdev) i40e_del_filter(vsi, netdev->dev_addr, vid, is_vf, is_netdev); - list_for_each_entry(f, &vsi->mac_filter_list, list) + list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) i40e_del_filter(vsi, f->macaddr, vid, is_vf, is_netdev); /* go through all the filters for this VSI and if there is only @@ -2492,7 +2494,7 @@ int i40e_vsi_kill_vlan(struct i40e_vsi *vsi, s16 vid) } if (!filter_count) { - list_for_each_entry(f, &vsi->mac_filter_list, list) { + list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { i40e_del_filter(vsi, f->macaddr, 0, is_vf, is_netdev); add_f = i40e_add_filter(vsi, f->macaddr, I40E_VLAN_ANY, is_vf, is_netdev); @@ -2537,8 +2539,6 @@ static int i40e_vlan_rx_add_vid(struct net_device *netdev, if (vid > 4095) return -EINVAL; - netdev_info(netdev, "adding %pM vid=%d\n", netdev->dev_addr, vid); - /* If the network stack called us with vid = 0 then * it is asking to receive priority tagged packets with * vlan id 0. Our HW receives them by default when configured @@ -2572,8 +2572,6 @@ static int i40e_vlan_rx_kill_vid(struct net_device *netdev, struct i40e_netdev_priv *np = netdev_priv(netdev); struct i40e_vsi *vsi = np->vsi; - netdev_info(netdev, "removing %pM vid=%d\n", netdev->dev_addr, vid); - /* return code is ignored as there is nothing a user * can do about failure to remove and a log message was * already printed from the other function @@ -9005,7 +9003,6 @@ static const struct net_device_ops i40e_netdev_ops = { **/ static int i40e_config_netdev(struct i40e_vsi *vsi) { - u8 brdcast[ETH_ALEN] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; struct i40e_pf *pf = vsi->back; struct i40e_hw *hw = &pf->hw; struct i40e_netdev_priv *np; @@ -9069,18 +9066,10 @@ static int i40e_config_netdev(struct i40e_vsi *vsi) * default a MAC-VLAN filter that accepts any tagged packet * which must be replaced by a normal filter. */ - if (!i40e_rm_default_mac_filter(vsi, mac_addr)) { - spin_lock_bh(&vsi->mac_filter_list_lock); - i40e_add_filter(vsi, mac_addr, - I40E_VLAN_ANY, false, true); - spin_unlock_bh(&vsi->mac_filter_list_lock); - } - } else if ((pf->hw.aq.api_maj_ver > 1) || - ((pf->hw.aq.api_maj_ver == 1) && - (pf->hw.aq.api_min_ver > 4))) { - /* Supported in FW API version higher than 1.4 */ - pf->flags |= I40E_FLAG_GENEVE_OFFLOAD_CAPABLE; - pf->auto_disable_flags = I40E_FLAG_HW_ATR_EVICT_CAPABLE; + i40e_rm_default_mac_filter(vsi, mac_addr); + spin_lock_bh(&vsi->mac_filter_list_lock); + i40e_add_filter(vsi, mac_addr, I40E_VLAN_ANY, false, true); + spin_unlock_bh(&vsi->mac_filter_list_lock); } else { /* relate the VSI_VMDQ name to the VSI_MAIN name */ snprintf(netdev->name, IFNAMSIZ, "%sv%%d", @@ -9092,10 +9081,6 @@ static int i40e_config_netdev(struct i40e_vsi *vsi) spin_unlock_bh(&vsi->mac_filter_list_lock); } - spin_lock_bh(&vsi->mac_filter_list_lock); - i40e_add_filter(vsi, brdcast, I40E_VLAN_ANY, false, false); - spin_unlock_bh(&vsi->mac_filter_list_lock); - ether_addr_copy(netdev->dev_addr, mac_addr); ether_addr_copy(netdev->perm_addr, mac_addr); @@ -9173,8 +9158,6 @@ int i40e_is_vsi_uplink_mode_veb(struct i40e_vsi *vsi) static int i40e_add_vsi(struct i40e_vsi *vsi) { int ret = -ENODEV; - u8 laa_macaddr[ETH_ALEN]; - bool found_laa_mac_filter = false; struct i40e_pf *pf = vsi->back; struct i40e_hw *hw = &pf->hw; struct i40e_vsi_context ctxt; @@ -9363,41 +9346,16 @@ static int i40e_add_vsi(struct i40e_vsi *vsi) vsi->id = ctxt.vsi_number; } + vsi->active_filters = 0; + clear_bit(__I40E_FILTER_OVERFLOW_PROMISC, &vsi->state); spin_lock_bh(&vsi->mac_filter_list_lock); /* If macvlan filters already exist, force them to get loaded */ list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) { - f->changed = true; + f->state = I40E_FILTER_NEW; f_count++; - - /* Expected to have only one MAC filter entry for LAA in list */ - if (f->is_laa && vsi->type == I40E_VSI_MAIN) { - ether_addr_copy(laa_macaddr, f->macaddr); - found_laa_mac_filter = true; - } } spin_unlock_bh(&vsi->mac_filter_list_lock); - if (found_laa_mac_filter) { - struct i40e_aqc_remove_macvlan_element_data element; - - memset(&element, 0, sizeof(element)); - ether_addr_copy(element.mac_addr, laa_macaddr); - element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH; - ret = i40e_aq_remove_macvlan(hw, vsi->seid, - &element, 1, NULL); - if (ret) { - /* some older FW has a different default */ - element.flags |= - I40E_AQC_MACVLAN_DEL_IGNORE_VLAN; - i40e_aq_remove_macvlan(hw, vsi->seid, - &element, 1, NULL); - } - - i40e_aq_mac_address_write(hw, - I40E_AQC_WRITE_TYPE_LAA_WOL, - laa_macaddr, NULL); - } - if (f_count) { vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED; pf->flags |= I40E_FLAG_FILTER_SYNC; @@ -9608,6 +9566,8 @@ static struct i40e_vsi *i40e_vsi_reinit_setup(struct i40e_vsi *vsi) pf->vsi[pf->lan_vsi]->tc_config.enabled_tc = 0; pf->vsi[pf->lan_vsi]->seid = pf->main_vsi_seid; i40e_vsi_config_tc(pf->vsi[pf->lan_vsi], enabled_tc); + if (vsi->type == I40E_VSI_MAIN) + i40e_rm_default_mac_filter(vsi, pf->hw.mac.perm_addr); /* assign it some queues */ ret = i40e_alloc_rings(vsi); -- cgit v0.10.2 From ae33256c55d2fefcad8712e750b846461994a1af Mon Sep 17 00:00:00 2001 From: Bimmy Pujari Date: Mon, 20 Jun 2016 09:10:39 -0700 Subject: i40e/i40evf-bump version to 1.6.11 Signed-off-by: Bimmy Pujari Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index bdaebb6..2351a08 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -41,7 +41,7 @@ static const char i40e_driver_string[] = #define DRV_VERSION_MAJOR 1 #define DRV_VERSION_MINOR 6 -#define DRV_VERSION_BUILD 4 +#define DRV_VERSION_BUILD 11 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ __stringify(DRV_VERSION_MINOR) "." \ __stringify(DRV_VERSION_BUILD) DRV_KERN diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c index b12edba..600fb9c 100644 --- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c +++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c @@ -38,7 +38,7 @@ static const char i40evf_driver_string[] = #define DRV_VERSION_MAJOR 1 #define DRV_VERSION_MINOR 6 -#define DRV_VERSION_BUILD 4 +#define DRV_VERSION_BUILD 11 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \ __stringify(DRV_VERSION_MINOR) "." \ __stringify(DRV_VERSION_BUILD) \ -- cgit v0.10.2 From 4dec7d045d6d20afe794adbda5b6da2fe5d63a35 Mon Sep 17 00:00:00 2001 From: Amitoj Kaur Chawla Date: Thu, 23 Jun 2016 10:30:55 +0530 Subject: i40e: Remove redundant memset Remove redundant call to memset before a call to memcpy. The Coccinelle semantic patch used to make this change is as follows: @@ expression e1,e2,e3,e4; @@ - memset(e1,e2,e3); memcpy(e1,e4,e3); Signed-off-by: Amitoj Kaur Chawla Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 2351a08..753c3b1 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -7916,7 +7916,6 @@ static int i40e_config_rss_aq(struct i40e_vsi *vsi, const u8 *seed, u8 *rss_lut; int ret, i; - memset(&rss_key, 0, sizeof(rss_key)); memcpy(&rss_key, seed, sizeof(rss_key)); rss_lut = kzalloc(pf->rss_table_size, GFP_KERNEL); -- cgit v0.10.2 From cd3be169a5ffe6a289255e11123abe933e16f112 Mon Sep 17 00:00:00 2001 From: Catherine Sullivan Date: Thu, 23 Jun 2016 14:08:46 -0700 Subject: i40e: Move the mutex lock in i40e_client_unregister We need to lock the client list around the i40e_client_release call to prevent the release from interrupting the client instances while they are being added. Change-Id: I99993f20179aaf8730207833e7d0869d2ccffa1d Signed-off-by: Catherine Sullivan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_client.c b/drivers/net/ethernet/intel/i40e/i40e_client.c index 0e6ac84..e1370c5 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_client.c +++ b/drivers/net/ethernet/intel/i40e/i40e_client.c @@ -980,13 +980,13 @@ int i40e_unregister_client(struct i40e_client *client) * a close for each of the client instances that were opened. * client_release function is called to handle this. */ + mutex_lock(&i40e_client_mutex); if (!client || i40e_client_release(client)) { ret = -EIO; goto out; } /* TODO: check if device is in reset, or if that matters? */ - mutex_lock(&i40e_client_mutex); if (!i40e_client_is_registered(client)) { pr_info("i40e: Client %s has not been registered\n", client->name); @@ -1005,8 +1005,8 @@ int i40e_unregister_client(struct i40e_client *client) client->name); } - mutex_unlock(&i40e_client_mutex); out: + mutex_unlock(&i40e_client_mutex); return ret; } EXPORT_SYMBOL(i40e_unregister_client); -- cgit v0.10.2 From 9169827b7ad7d4dcae635a051d32cc7421ed8202 Mon Sep 17 00:00:00 2001 From: Stefan Assmann Date: Mon, 27 Jun 2016 15:03:43 +0200 Subject: i40e: add missing link advertise setting Adding the missing link advertise for some X710 NICs. This can be observed by simply calling ethtool on the interface. root@rhel7:~ # ethtool eth0 Settings for eth0: Supported ports: [ FIBRE ] Supported link modes: 10000baseT/Full Supported pause frame use: Symmetric Supports auto-negotiation: No Advertised link modes: Not reported [...] With fix applied. root@rhel7:~ # ethtool eth0 Settings for eth0: Supported ports: [ FIBRE ] Supported link modes: 10000baseT/Full Supported pause frame use: Symmetric Supports auto-negotiation: No Advertised link modes: 10000baseT/Full [...] Signed-off-by: Stefan Assmann Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c index 0cfde30..c912e04 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c +++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c @@ -461,6 +461,7 @@ static void i40e_get_settings_link_up(struct i40e_hw *hw, case I40E_PHY_TYPE_10GBASE_SFPP_CU: case I40E_PHY_TYPE_10GBASE_AOC: ecmd->supported = SUPPORTED_10000baseT_Full; + ecmd->advertising = SUPPORTED_10000baseT_Full; break; case I40E_PHY_TYPE_SGMII: ecmd->supported = SUPPORTED_Autoneg | -- cgit v0.10.2 From b1b15df5923232114a908fe1fd8b701c36259259 Mon Sep 17 00:00:00 2001 From: Tushar Dave Date: Fri, 1 Jul 2016 10:11:20 -0700 Subject: i40e: Explicitly write platform-specific mac address after PF reset i40e PF reset clears mac filters. If platform-specific mac address is used, driver has to explicitly write default mac address to mac filters otherwise all incoming traffic destined to default mac address will be dropped after reset. This issue was found on SPARC while toggling i40e ntuple via ethtool. Signed-off-by: Tushar Dave Acked-by: Sowmini Varadhan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 753c3b1..1ee45cf 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -2584,6 +2584,44 @@ static int i40e_vlan_rx_kill_vid(struct net_device *netdev, } /** + * i40e_macaddr_init - explicitly write the mac address filters + * + * @vsi: pointer to the vsi + * @macaddr: the MAC address + * + * This is needed when the macaddr has been obtained by other + * means than the default, e.g., from Open Firmware or IDPROM. + * Returns 0 on success, negative on failure + **/ +static int i40e_macaddr_init(struct i40e_vsi *vsi, u8 *macaddr) +{ + int ret; + struct i40e_aqc_add_macvlan_element_data element; + + ret = i40e_aq_mac_address_write(&vsi->back->hw, + I40E_AQC_WRITE_TYPE_LAA_WOL, + macaddr, NULL); + if (ret) { + dev_info(&vsi->back->pdev->dev, + "Addr change for VSI failed: %d\n", ret); + return -EADDRNOTAVAIL; + } + + memset(&element, 0, sizeof(element)); + ether_addr_copy(element.mac_addr, macaddr); + element.flags = cpu_to_le16(I40E_AQC_MACVLAN_ADD_PERFECT_MATCH); + ret = i40e_aq_add_macvlan(&vsi->back->hw, vsi->seid, &element, 1, NULL); + if (ret) { + dev_info(&vsi->back->pdev->dev, + "add filter failed err %s aq_err %s\n", + i40e_stat_str(&vsi->back->hw, ret), + i40e_aq_str(&vsi->back->hw, + vsi->back->hw.aq.asq_last_status)); + } + return ret; +} + +/** * i40e_restore_vlan - Reinstate vlans when vsi/netdev comes back up * @vsi: the vsi being brought back up **/ @@ -3029,8 +3067,19 @@ static void i40e_vsi_config_dcb_rings(struct i40e_vsi *vsi) **/ static void i40e_set_vsi_rx_mode(struct i40e_vsi *vsi) { + struct i40e_pf *pf = vsi->back; + int err; + if (vsi->netdev) i40e_set_rx_mode(vsi->netdev); + + if (!!(pf->flags & I40E_FLAG_PF_MAC)) { + err = i40e_macaddr_init(vsi, pf->hw.mac.addr); + if (err) { + dev_warn(&pf->pdev->dev, + "could not set up macaddr; err %d\n", err); + } + } } /** @@ -9592,44 +9641,6 @@ err_vsi: } /** - * i40e_macaddr_init - explicitly write the mac address filters. - * - * @vsi: pointer to the vsi. - * @macaddr: the MAC address - * - * This is needed when the macaddr has been obtained by other - * means than the default, e.g., from Open Firmware or IDPROM. - * Returns 0 on success, negative on failure - **/ -static int i40e_macaddr_init(struct i40e_vsi *vsi, u8 *macaddr) -{ - int ret; - struct i40e_aqc_add_macvlan_element_data element; - - ret = i40e_aq_mac_address_write(&vsi->back->hw, - I40E_AQC_WRITE_TYPE_LAA_WOL, - macaddr, NULL); - if (ret) { - dev_info(&vsi->back->pdev->dev, - "Addr change for VSI failed: %d\n", ret); - return -EADDRNOTAVAIL; - } - - memset(&element, 0, sizeof(element)); - ether_addr_copy(element.mac_addr, macaddr); - element.flags = cpu_to_le16(I40E_AQC_MACVLAN_ADD_PERFECT_MATCH); - ret = i40e_aq_add_macvlan(&vsi->back->hw, vsi->seid, &element, 1, NULL); - if (ret) { - dev_info(&vsi->back->pdev->dev, - "add filter failed err %s aq_err %s\n", - i40e_stat_str(&vsi->back->hw, ret), - i40e_aq_str(&vsi->back->hw, - vsi->back->hw.aq.asq_last_status)); - } - return ret; -} - -/** * i40e_vsi_setup - Set up a VSI by a given type * @pf: board private structure * @type: VSI type -- cgit v0.10.2 From fc355e078a830c40f6245a55ecfff6481a4bcf57 Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Thu, 9 Jun 2016 10:45:34 -0700 Subject: ixgbevf: fix NACK check in ixgbevf_set_uc_addr_vf() Fix the NACK check in ixgbevf_set_uc_addr_vf() for instances where index != 0. Signed-off-by: Emil Tantilov Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbevf/vf.c b/drivers/net/ethernet/intel/ixgbevf/vf.c index e670d3b..aa2b1e8 100644 --- a/drivers/net/ethernet/intel/ixgbevf/vf.c +++ b/drivers/net/ethernet/intel/ixgbevf/vf.c @@ -256,7 +256,7 @@ static s32 ixgbevf_get_mac_addr_vf(struct ixgbe_hw *hw, u8 *mac_addr) static s32 ixgbevf_set_uc_addr_vf(struct ixgbe_hw *hw, u32 index, u8 *addr) { struct ixgbe_mbx_info *mbx = &hw->mbx; - u32 msgbuf[3]; + u32 msgbuf[3], msgbuf_chk; u8 *msg_addr = (u8 *)(&msgbuf[1]); s32 ret_val; @@ -268,6 +268,8 @@ static s32 ixgbevf_set_uc_addr_vf(struct ixgbe_hw *hw, u32 index, u8 *addr) */ msgbuf[0] |= index << IXGBE_VT_MSGINFO_SHIFT; msgbuf[0] |= IXGBE_VF_SET_MACVLAN; + msgbuf_chk = msgbuf[0]; + if (addr) ether_addr_copy(msg_addr, addr); ret_val = mbx->ops.write_posted(hw, msgbuf, 3); @@ -275,12 +277,12 @@ static s32 ixgbevf_set_uc_addr_vf(struct ixgbe_hw *hw, u32 index, u8 *addr) if (!ret_val) ret_val = mbx->ops.read_posted(hw, msgbuf, 3); - msgbuf[0] &= ~IXGBE_VT_MSGTYPE_CTS; + if (!ret_val) { + msgbuf[0] &= ~IXGBE_VT_MSGTYPE_CTS; - if (!ret_val) - if (msgbuf[0] == - (IXGBE_VF_SET_MACVLAN | IXGBE_VT_MSGTYPE_NACK)) - ret_val = -ENOMEM; + if (msgbuf[0] == (msgbuf_chk | IXGBE_VT_MSGTYPE_NACK)) + return -ENOMEM; + } return ret_val; } -- cgit v0.10.2 From 003287e0f0875d0ba5f4ee3d7741ec9992766d71 Mon Sep 17 00:00:00 2001 From: Don Skidmore Date: Fri, 17 Jun 2016 17:10:13 -0400 Subject: ixgbevf: Correct parameter sent to LED function The second parameter of these functions is the index to the led we are interested in affecting. However we were mistakenly passing the offset in the register. This patch corrects that and adds some bonds checking which would hopefully make bugs like this more noticeable in the future. Signed-off-by: Don Skidmore Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c index 902d206..0ffba44 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c @@ -763,6 +763,9 @@ s32 ixgbe_led_on_generic(struct ixgbe_hw *hw, u32 index) { u32 led_reg = IXGBE_READ_REG(hw, IXGBE_LEDCTL); + if (index > 3) + return IXGBE_ERR_PARAM; + /* To turn on the LED, set mode to ON. */ led_reg &= ~IXGBE_LED_MODE_MASK(index); led_reg |= IXGBE_LED_ON << IXGBE_LED_MODE_SHIFT(index); @@ -781,6 +784,9 @@ s32 ixgbe_led_off_generic(struct ixgbe_hw *hw, u32 index) { u32 led_reg = IXGBE_READ_REG(hw, IXGBE_LEDCTL); + if (index > 3) + return IXGBE_ERR_PARAM; + /* To turn off the LED, set mode to OFF. */ led_reg &= ~IXGBE_LED_MODE_MASK(index); led_reg |= IXGBE_LED_OFF << IXGBE_LED_MODE_SHIFT(index); @@ -2698,6 +2704,9 @@ s32 ixgbe_blink_led_start_generic(struct ixgbe_hw *hw, u32 index) bool locked = false; s32 ret_val; + if (index > 3) + return IXGBE_ERR_PARAM; + /* * Link must be up to auto-blink the LEDs; * Force it if link is down. @@ -2741,6 +2750,9 @@ s32 ixgbe_blink_led_stop_generic(struct ixgbe_hw *hw, u32 index) bool locked = false; s32 ret_val; + if (index > 3) + return IXGBE_ERR_PARAM; + ret_val = hw->mac.ops.prot_autoc_read(hw, &locked, &autoc_reg); if (ret_val) return ret_val; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c index 8a84507..0d7209e 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -2204,11 +2204,11 @@ static int ixgbe_set_phys_id(struct net_device *netdev, return 2; case ETHTOOL_ID_ON: - hw->mac.ops.led_on(hw, IXGBE_LED_ON); + hw->mac.ops.led_on(hw, hw->bus.func); break; case ETHTOOL_ID_OFF: - hw->mac.ops.led_off(hw, IXGBE_LED_ON); + hw->mac.ops.led_off(hw, hw->bus.func); break; case ETHTOOL_ID_INACTIVE: -- cgit v0.10.2 From 9f8fe731b8e810ef0324ebd8fc5973deecb6fb0b Mon Sep 17 00:00:00 2001 From: Don Skidmore Date: Mon, 6 Jun 2016 20:23:24 -0400 Subject: ixgbevf: bump version number Bump the version number to more closely match the function included in the driver. Signed-off-by: Don Skidmore Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index acc2401..a1c83c1 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -56,7 +56,7 @@ const char ixgbevf_driver_name[] = "ixgbevf"; static const char ixgbevf_driver_string[] = "Intel(R) 10 Gigabit PCI Express Virtual Function Network Driver"; -#define DRV_VERSION "2.12.1-k" +#define DRV_VERSION "3.2.2-k" const char ixgbevf_driver_version[] = DRV_VERSION; static char ixgbevf_copyright[] = "Copyright (c) 2009 - 2015 Intel Corporation."; -- cgit v0.10.2 From 310ea1236c2a8492d3c0f1453e5995e08b24c70e Mon Sep 17 00:00:00 2001 From: Don Skidmore Date: Tue, 14 Jun 2016 14:26:28 -0400 Subject: ixgbe: Change register variable to unsigned I noticed this variable used for register reads wasn't an unsigned so this patch corrects that. I don't believe this was causing any issue as is but this is more consistent with the rest of the driver. Signed-off-by: Don Skidmore Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c index 0ffba44..ce881a7 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c @@ -2663,7 +2663,7 @@ s32 ixgbe_disable_rx_buff_generic(struct ixgbe_hw *hw) **/ s32 ixgbe_enable_rx_buff_generic(struct ixgbe_hw *hw) { - int secrxreg; + u32 secrxreg; secrxreg = IXGBE_READ_REG(hw, IXGBE_SECRXCTRL); secrxreg &= ~IXGBE_SECRXCTRL_RX_DIS; -- cgit v0.10.2 From 90c6f87786e65e12cab06e637db8c7ef2f22de95 Mon Sep 17 00:00:00 2001 From: Babu Moger Date: Sat, 18 Jun 2016 17:40:47 -0700 Subject: ixgbe: Fix minor typo while freeing irq The array subscript increments after the execution of the statement. So there is no issue here. However it helps to read the code better. Signed-off-by: Babu Moger Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 918b94b..f2837ea 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -3084,7 +3084,7 @@ static void ixgbe_free_irq(struct ixgbe_adapter *adapter) free_irq(entry->vector, q_vector); } - free_irq(adapter->msix_entries[vector++].vector, adapter); + free_irq(adapter->msix_entries[vector].vector, adapter); } /** -- cgit v0.10.2 From 1d94f987f53cb53798dbcc7e7f1dfb00f9269efb Mon Sep 17 00:00:00 2001 From: Don Skidmore Date: Wed, 29 Jun 2016 19:32:24 -0400 Subject: ixgbevf: add VF support for new hardware This patch add VF support for the new X553 hardware. Signed-off-by: Don Skidmore Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbevf/defines.h b/drivers/net/ethernet/intel/ixgbevf/defines.h index ae09d60..8617cae 100644 --- a/drivers/net/ethernet/intel/ixgbevf/defines.h +++ b/drivers/net/ethernet/intel/ixgbevf/defines.h @@ -32,6 +32,7 @@ #define IXGBE_DEV_ID_X540_VF 0x1515 #define IXGBE_DEV_ID_X550_VF 0x1565 #define IXGBE_DEV_ID_X550EM_X_VF 0x15A8 +#define IXGBE_DEV_ID_X550EM_A_VF 0x15C5 #define IXGBE_DEV_ID_82599_VF_HV 0x152E #define IXGBE_DEV_ID_X540_VF_HV 0x1530 diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h index d5944c3..be52f59 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h @@ -457,6 +457,7 @@ enum ixgbevf_boards { board_X550_vf_hv, board_X550EM_x_vf, board_X550EM_x_vf_hv, + board_x550em_a_vf, }; enum ixgbevf_xcast_modes { @@ -470,6 +471,7 @@ extern const struct ixgbevf_info ixgbevf_X540_vf_info; extern const struct ixgbevf_info ixgbevf_X550_vf_info; extern const struct ixgbevf_info ixgbevf_X550EM_x_vf_info; extern const struct ixgbe_mbx_operations ixgbevf_mbx_ops; +extern const struct ixgbevf_info ixgbevf_x550em_a_vf_info; extern const struct ixgbevf_info ixgbevf_82599_vf_hv_info; extern const struct ixgbevf_info ixgbevf_X540_vf_hv_info; diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index a1c83c1..1bffac3 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -70,6 +70,7 @@ static const struct ixgbevf_info *ixgbevf_info_tbl[] = { [board_X550_vf_hv] = &ixgbevf_X550_vf_hv_info, [board_X550EM_x_vf] = &ixgbevf_X550EM_x_vf_info, [board_X550EM_x_vf_hv] = &ixgbevf_X550EM_x_vf_hv_info, + [board_x550em_a_vf] = &ixgbevf_x550em_a_vf_info, }; /* ixgbevf_pci_tbl - PCI Device ID Table @@ -89,6 +90,7 @@ static const struct pci_device_id ixgbevf_pci_tbl[] = { {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550_VF_HV), board_X550_vf_hv }, {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550EM_X_VF), board_X550EM_x_vf }, {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550EM_X_VF_HV), board_X550EM_x_vf_hv}, + {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550EM_A_VF), board_x550em_a_vf }, /* required last entry */ {0, } }; diff --git a/drivers/net/ethernet/intel/ixgbevf/vf.c b/drivers/net/ethernet/intel/ixgbevf/vf.c index aa2b1e8..cb409b3 100644 --- a/drivers/net/ethernet/intel/ixgbevf/vf.c +++ b/drivers/net/ethernet/intel/ixgbevf/vf.c @@ -1007,3 +1007,8 @@ const struct ixgbevf_info ixgbevf_X550EM_x_vf_hv_info = { .mac = ixgbe_mac_X550EM_x_vf, .mac_ops = &ixgbevf_hv_mac_ops, }; + +const struct ixgbevf_info ixgbevf_x550em_a_vf_info = { + .mac = ixgbe_mac_x550em_a_vf, + .mac_ops = &ixgbevf_mac_ops, +}; diff --git a/drivers/net/ethernet/intel/ixgbevf/vf.h b/drivers/net/ethernet/intel/ixgbevf/vf.h index 2cac610..e3a0dea 100644 --- a/drivers/net/ethernet/intel/ixgbevf/vf.h +++ b/drivers/net/ethernet/intel/ixgbevf/vf.h @@ -78,6 +78,7 @@ enum ixgbe_mac_type { ixgbe_mac_X540_vf, ixgbe_mac_X550_vf, ixgbe_mac_X550EM_x_vf, + ixgbe_mac_x550em_a_vf, ixgbe_num_macs }; -- cgit v0.10.2 From abf76d76c5e716a0129180b8f464b1b293823adf Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Thu, 7 Jul 2016 17:18:38 -0700 Subject: ixgbe: fix setup_fc for x550em mac->ops.setup_fc can be null for backplanes which can cause the driver to crash on load. Reported-by: Patrick McLean Signed-off-by: Emil Tantilov Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c index 19b75cd..4716ca4 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c @@ -1618,6 +1618,8 @@ static void ixgbe_init_mac_link_ops_X550em(struct ixgbe_hw *hw) { struct ixgbe_mac_info *mac = &hw->mac; + mac->ops.setup_fc = ixgbe_setup_fc_x550em; + switch (mac->ops.get_media_type(hw)) { case ixgbe_media_type_fiber: /* CS4227 does not support autoneg, so disable the laser control @@ -1627,7 +1629,6 @@ static void ixgbe_init_mac_link_ops_X550em(struct ixgbe_hw *hw) mac->ops.enable_tx_laser = NULL; mac->ops.flap_tx_laser = NULL; mac->ops.setup_link = ixgbe_setup_mac_link_multispeed_fiber; - mac->ops.setup_fc = ixgbe_setup_fc_x550em; switch (hw->device_id) { case IXGBE_DEV_ID_X550EM_A_SFP_N: mac->ops.setup_mac_link = ixgbe_setup_mac_link_sfp_n; @@ -1655,7 +1656,6 @@ static void ixgbe_init_mac_link_ops_X550em(struct ixgbe_hw *hw) mac->ops.setup_link = ixgbe_setup_sgmii; break; default: - mac->ops.setup_fc = ixgbe_setup_fc_x550em; break; } } -- cgit v0.10.2 From 8e8247ab98315e096b812a68381ca812f0b0b5e3 Mon Sep 17 00:00:00 2001 From: Don Skidmore Date: Mon, 11 Jul 2016 21:29:56 -0400 Subject: ixgbevf: Add lock around ixgbevf_reinit_locked call The function ixgbevf_reinit_locked() assumes you have the rtnl lock however we didn't when calling from the service task. This patch corrects that. Signed-off-by: Don Skidmore Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index 1bffac3..7dc4245 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -2779,7 +2779,9 @@ static void ixgbevf_reset_subtask(struct ixgbevf_adapter *adapter) adapter->tx_timeout_count++; + rtnl_lock(); ixgbevf_reinit_locked(adapter); + rtnl_unlock(); } /** -- cgit v0.10.2 From 6b8368798772a4fabfec690be3b5f390c4bda600 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 12 Jul 2016 15:17:02 +0000 Subject: ixgbe: Add missing destroy_workqueue() on error in ixgbe_init_module() Add the missing destroy_workqueue() before return from ixgbe_init_module() in the error handling case. Signed-off-by: Wei Yongjun Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index f2837ea..b0039c0 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -10112,6 +10112,7 @@ static int __init ixgbe_init_module(void) ret = pci_register_driver(&ixgbe_driver); if (ret) { + destroy_workqueue(ixgbe_wq); ixgbe_dbg_exit(); return ret; } -- cgit v0.10.2 From 6e469ed03ed9b21b0c2dd46f77113a85f3e2bea6 Mon Sep 17 00:00:00 2001 From: Don Skidmore Date: Tue, 12 Jul 2016 18:47:38 -0400 Subject: ixgbevf: Protect ixgbevf_reset_subtask from remove event In ixgbevf_reset_subtask We weren't verifying that the port haven't been removed, we are with this patch. Signed-off-by: Don Skidmore Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index 7dc4245..201e482 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -2774,6 +2774,7 @@ static void ixgbevf_reset_subtask(struct ixgbevf_adapter *adapter) /* If we're already down or resetting, just bail */ if (test_bit(__IXGBEVF_DOWN, &adapter->state) || + test_bit(__IXGBEVF_REMOVING, &adapter->state) || test_bit(__IXGBEVF_RESETTING, &adapter->state)) return; -- cgit v0.10.2 From 6a11e52b6995b07a83a7d50e6301025ca35501be Mon Sep 17 00:00:00 2001 From: Tony Nguyen Date: Wed, 13 Jul 2016 10:33:16 -0700 Subject: ixgbevf: Add range checking for setting MTU Currently when setting the VF's MTU, the PF can return a NACK but this isn't passed on to the VF. Propagate the results from the PF to the VF so errors can be reported. In ixgbevf_change_mtu, return an error and reject the change. For ixgbevf_configure_rx, log the error for debugging purposes since the function is buried in a series of Rx config routines that are void. Signed-off-by: Tony Nguyen Signed-off-by: Emil Tantilov Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index 201e482..d9d6616 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -1802,16 +1802,19 @@ static void ixgbevf_configure_rx_ring(struct ixgbevf_adapter *adapter, **/ static void ixgbevf_configure_rx(struct ixgbevf_adapter *adapter) { - int i; struct ixgbe_hw *hw = &adapter->hw; struct net_device *netdev = adapter->netdev; + int i, ret; ixgbevf_setup_psrtype(adapter); if (hw->mac.type >= ixgbe_mac_X550_vf) ixgbevf_setup_vfmrqc(adapter); /* notify the PF of our intent to use this size of frame */ - hw->mac.ops.set_rlpml(hw, netdev->mtu + ETH_HLEN + ETH_FCS_LEN); + ret = hw->mac.ops.set_rlpml(hw, netdev->mtu + ETH_HLEN + ETH_FCS_LEN); + if (ret) + dev_err(&adapter->pdev->dev, + "Failed to set MTU at %d\n", netdev->mtu); /* Setup the HW Rx Head and Tail Descriptor Pointers and * the Base and Length of the Rx Descriptor Ring @@ -3737,6 +3740,7 @@ static int ixgbevf_change_mtu(struct net_device *netdev, int new_mtu) struct ixgbe_hw *hw = &adapter->hw; int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN; int max_possible_frame = MAXIMUM_ETHERNET_VLAN_SIZE; + int ret; switch (adapter->hw.api_version) { case ixgbe_mbox_api_11: @@ -3753,14 +3757,17 @@ static int ixgbevf_change_mtu(struct net_device *netdev, int new_mtu) if ((new_mtu < 68) || (max_frame > max_possible_frame)) return -EINVAL; + /* notify the PF of our intent to use this size of frame */ + ret = hw->mac.ops.set_rlpml(hw, max_frame); + if (ret) + return -EINVAL; + hw_dbg(hw, "changing MTU from %d to %d\n", netdev->mtu, new_mtu); + /* must set new MTU before calling down or up */ netdev->mtu = new_mtu; - /* notify the PF of our intent to use this size of frame */ - hw->mac.ops.set_rlpml(hw, max_frame); - return 0; } diff --git a/drivers/net/ethernet/intel/ixgbevf/vf.c b/drivers/net/ethernet/intel/ixgbevf/vf.c index cb409b3..4862bc2 100644 --- a/drivers/net/ethernet/intel/ixgbevf/vf.c +++ b/drivers/net/ethernet/intel/ixgbevf/vf.c @@ -33,6 +33,18 @@ */ #define IXGBE_HV_RESET_OFFSET 0x201 +static inline s32 ixgbevf_write_msg_read_ack(struct ixgbe_hw *hw, u32 *msg, + u32 *retmsg, u16 size) +{ + struct ixgbe_mbx_info *mbx = &hw->mbx; + s32 retval = mbx->ops.write_posted(hw, msg, size); + + if (retval) + return retval; + + return mbx->ops.read_posted(hw, retmsg, size); +} + /** * ixgbevf_start_hw_vf - Prepare hardware for Tx/Rx * @hw: pointer to hardware structure @@ -470,17 +482,6 @@ static s32 ixgbevf_hv_set_rar_vf(struct ixgbe_hw *hw, u32 index, u8 *addr, return -EOPNOTSUPP; } -static void ixgbevf_write_msg_read_ack(struct ixgbe_hw *hw, - u32 *msg, u16 size) -{ - struct ixgbe_mbx_info *mbx = &hw->mbx; - u32 retmsg[IXGBE_VFMAILBOX_SIZE]; - s32 retval = mbx->ops.write_posted(hw, msg, size); - - if (!retval) - mbx->ops.read_posted(hw, retmsg, size); -} - /** * ixgbevf_update_mc_addr_list_vf - Update Multicast addresses * @hw: pointer to the HW structure @@ -521,7 +522,7 @@ static s32 ixgbevf_update_mc_addr_list_vf(struct ixgbe_hw *hw, vector_list[i++] = ixgbevf_mta_vector(hw, ha->addr); } - ixgbevf_write_msg_read_ack(hw, msgbuf, IXGBE_VFMAILBOX_SIZE); + ixgbevf_write_msg_read_ack(hw, msgbuf, msgbuf, IXGBE_VFMAILBOX_SIZE); return 0; } @@ -799,13 +800,22 @@ out: * @hw: pointer to the HW structure * @max_size: value to assign to max frame size **/ -static void ixgbevf_set_rlpml_vf(struct ixgbe_hw *hw, u16 max_size) +static s32 ixgbevf_set_rlpml_vf(struct ixgbe_hw *hw, u16 max_size) { u32 msgbuf[2]; + s32 ret_val; msgbuf[0] = IXGBE_VF_SET_LPE; msgbuf[1] = max_size; - ixgbevf_write_msg_read_ack(hw, msgbuf, 2); + + ret_val = ixgbevf_write_msg_read_ack(hw, msgbuf, msgbuf, 2); + if (ret_val) + return ret_val; + if ((msgbuf[0] & IXGBE_VF_SET_LPE) && + (msgbuf[0] & IXGBE_VT_MSGTYPE_NACK)) + return IXGBE_ERR_MBX; + + return 0; } /** @@ -814,7 +824,7 @@ static void ixgbevf_set_rlpml_vf(struct ixgbe_hw *hw, u16 max_size) * @max_size: value to assign to max frame size * Hyper-V variant. **/ -static void ixgbevf_hv_set_rlpml_vf(struct ixgbe_hw *hw, u16 max_size) +static s32 ixgbevf_hv_set_rlpml_vf(struct ixgbe_hw *hw, u16 max_size) { u32 reg; @@ -825,6 +835,8 @@ static void ixgbevf_hv_set_rlpml_vf(struct ixgbe_hw *hw, u16 max_size) /* CRC == 4 */ reg |= ((max_size + 4) | IXGBE_RXDCTL_RLPML_EN); IXGBE_WRITE_REG(hw, IXGBE_VFRXDCTL(0), reg); + + return 0; } /** diff --git a/drivers/net/ethernet/intel/ixgbevf/vf.h b/drivers/net/ethernet/intel/ixgbevf/vf.h index e3a0dea..04d8d4e 100644 --- a/drivers/net/ethernet/intel/ixgbevf/vf.h +++ b/drivers/net/ethernet/intel/ixgbevf/vf.h @@ -69,7 +69,7 @@ struct ixgbe_mac_operations { s32 (*disable_mc)(struct ixgbe_hw *); s32 (*clear_vfta)(struct ixgbe_hw *); s32 (*set_vfta)(struct ixgbe_hw *, u32, u32, bool); - void (*set_rlpml)(struct ixgbe_hw *, u16); + s32 (*set_rlpml)(struct ixgbe_hw *, u16); }; enum ixgbe_mac_type { -- cgit v0.10.2 From 221c556acbd76154c34015b7cbbb3621ae7dbc7c Mon Sep 17 00:00:00 2001 From: Tony Nguyen Date: Wed, 13 Jul 2016 10:33:34 -0700 Subject: ixgbevf: Commonize mailbox write/read With changes to ixgbevf_write_msg_read_ack(), other functions are performing the same operations done here; change those functions to utilize ixgbevf_write_msg_read_ack(). Signed-off-by: Tony Nguyen Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbevf/vf.c b/drivers/net/ethernet/intel/ixgbevf/vf.c index 4862bc2..a52f70e 100644 --- a/drivers/net/ethernet/intel/ixgbevf/vf.c +++ b/drivers/net/ethernet/intel/ixgbevf/vf.c @@ -267,7 +267,6 @@ static s32 ixgbevf_get_mac_addr_vf(struct ixgbe_hw *hw, u8 *mac_addr) static s32 ixgbevf_set_uc_addr_vf(struct ixgbe_hw *hw, u32 index, u8 *addr) { - struct ixgbe_mbx_info *mbx = &hw->mbx; u32 msgbuf[3], msgbuf_chk; u8 *msg_addr = (u8 *)(&msgbuf[1]); s32 ret_val; @@ -284,11 +283,8 @@ static s32 ixgbevf_set_uc_addr_vf(struct ixgbe_hw *hw, u32 index, u8 *addr) if (addr) ether_addr_copy(msg_addr, addr); - ret_val = mbx->ops.write_posted(hw, msgbuf, 3); - - if (!ret_val) - ret_val = mbx->ops.read_posted(hw, msgbuf, 3); + ret_val = ixgbevf_write_msg_read_ack(hw, msgbuf, msgbuf, 3); if (!ret_val) { msgbuf[0] &= ~IXGBE_VT_MSGTYPE_CTS; @@ -437,7 +433,6 @@ int ixgbevf_get_rss_key_locked(struct ixgbe_hw *hw, u8 *rss_key) static s32 ixgbevf_set_rar_vf(struct ixgbe_hw *hw, u32 index, u8 *addr, u32 vmdq) { - struct ixgbe_mbx_info *mbx = &hw->mbx; u32 msgbuf[3]; u8 *msg_addr = (u8 *)(&msgbuf[1]); s32 ret_val; @@ -445,10 +440,8 @@ static s32 ixgbevf_set_rar_vf(struct ixgbe_hw *hw, u32 index, u8 *addr, memset(msgbuf, 0, sizeof(msgbuf)); msgbuf[0] = IXGBE_VF_SET_MAC_ADDR; ether_addr_copy(msg_addr, addr); - ret_val = mbx->ops.write_posted(hw, msgbuf, 3); - if (!ret_val) - ret_val = mbx->ops.read_posted(hw, msgbuf, 3); + ret_val = ixgbevf_write_msg_read_ack(hw, msgbuf, msgbuf, 2); msgbuf[0] &= ~IXGBE_VT_MSGTYPE_CTS; @@ -545,7 +538,6 @@ static s32 ixgbevf_hv_update_mc_addr_list_vf(struct ixgbe_hw *hw, **/ static s32 ixgbevf_update_xcast_mode(struct ixgbe_hw *hw, int xcast_mode) { - struct ixgbe_mbx_info *mbx = &hw->mbx; u32 msgbuf[2]; s32 err; @@ -559,11 +551,7 @@ static s32 ixgbevf_update_xcast_mode(struct ixgbe_hw *hw, int xcast_mode) msgbuf[0] = IXGBE_VF_UPDATE_XCAST_MODE; msgbuf[1] = xcast_mode; - err = mbx->ops.write_posted(hw, msgbuf, 2); - if (err) - return err; - - err = mbx->ops.read_posted(hw, msgbuf, 2); + err = ixgbevf_write_msg_read_ack(hw, msgbuf, msgbuf, 2); if (err) return err; @@ -592,7 +580,6 @@ static s32 ixgbevf_hv_update_xcast_mode(struct ixgbe_hw *hw, int xcast_mode) static s32 ixgbevf_set_vfta_vf(struct ixgbe_hw *hw, u32 vlan, u32 vind, bool vlan_on) { - struct ixgbe_mbx_info *mbx = &hw->mbx; u32 msgbuf[2]; s32 err; @@ -601,11 +588,7 @@ static s32 ixgbevf_set_vfta_vf(struct ixgbe_hw *hw, u32 vlan, u32 vind, /* Setting the 8 bit field MSG INFO to TRUE indicates "add" */ msgbuf[0] |= vlan_on << IXGBE_VT_MSGINFO_SHIFT; - err = mbx->ops.write_posted(hw, msgbuf, 2); - if (err) - goto mbx_err; - - err = mbx->ops.read_posted(hw, msgbuf, 2); + err = ixgbevf_write_msg_read_ack(hw, msgbuf, msgbuf, 2); if (err) goto mbx_err; @@ -853,11 +836,8 @@ static int ixgbevf_negotiate_api_version_vf(struct ixgbe_hw *hw, int api) msg[0] = IXGBE_VF_API_NEGOTIATE; msg[1] = api; msg[2] = 0; - err = hw->mbx.ops.write_posted(hw, msg, 3); - - if (!err) - err = hw->mbx.ops.read_posted(hw, msg, 3); + err = ixgbevf_write_msg_read_ack(hw, msg, msg, 3); if (!err) { msg[0] &= ~IXGBE_VT_MSGTYPE_CTS; @@ -906,11 +886,8 @@ int ixgbevf_get_queues(struct ixgbe_hw *hw, unsigned int *num_tcs, /* Fetch queue configuration from the PF */ msg[0] = IXGBE_VF_GET_QUEUE; msg[1] = msg[2] = msg[3] = msg[4] = 0; - err = hw->mbx.ops.write_posted(hw, msg, 5); - - if (!err) - err = hw->mbx.ops.read_posted(hw, msg, 5); + err = ixgbevf_write_msg_read_ack(hw, msg, msg, 5); if (!err) { msg[0] &= ~IXGBE_VT_MSGTYPE_CTS; -- cgit v0.10.2 From fdb359ee445f83400bad762d54b2964b39d42d10 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Thu, 14 Jul 2016 12:47:09 +0100 Subject: ixgbe: remove redundant check on ret_val The last check on ret_val is redundant since ret_val has not changed since the previous check, so remove it as it is extraneous. Signed-off-by: Colin Ian King Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c index 47afed7..63b2500 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c @@ -1813,9 +1813,6 @@ static s32 ixgbe_start_hw_82599(struct ixgbe_hw *hw) /* We need to run link autotry after the driver loads */ hw->mac.autotry_restart = true; - if (ret_val) - return ret_val; - return ixgbe_verify_fw_version_82599(hw); } -- cgit v0.10.2 From aac9e053f1044bf21ac068eeb0e8518d080f4a66 Mon Sep 17 00:00:00 2001 From: Don Skidmore Date: Tue, 19 Jul 2016 19:43:28 -0400 Subject: ixgbe: cleanup crosstalk fix This patch address a few issues with the initial crosstalk fix. Most important of which is the SDP that indicates the presents of a SFP+ module changes between HW types. With this change that is taken in to consideration It also moves the check closer to the base code that checks link. This makes it so we only need to do the check in one spot. Signed-off-by: Don Skidmore Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h index 9f2db18..9475ff9 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h @@ -804,8 +804,6 @@ struct ixgbe_adapter { #define IXGBE_RSS_KEY_SIZE 40 /* size of RSS Hash Key in bytes */ u32 rss_key[IXGBE_RSS_KEY_SIZE / sizeof(u32)]; - - bool need_crosstalk_fix; }; static inline u8 ixgbe_max_rss_indices(struct ixgbe_adapter *adapter) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c index ce881a7..b4217f3 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c @@ -277,6 +277,7 @@ s32 ixgbe_start_hw_generic(struct ixgbe_hw *hw) { s32 ret_val; u32 ctrl_ext; + u16 device_caps; /* Set the media type */ hw->phy.media_type = hw->mac.ops.get_media_type(hw); @@ -301,6 +302,22 @@ s32 ixgbe_start_hw_generic(struct ixgbe_hw *hw) if (ret_val) return ret_val; + /* Cashe bit indicating need for crosstalk fix */ + switch (hw->mac.type) { + case ixgbe_mac_82599EB: + case ixgbe_mac_X550EM_x: + case ixgbe_mac_x550em_a: + hw->mac.ops.get_device_caps(hw, &device_caps); + if (device_caps & IXGBE_DEVICE_CAPS_NO_CROSSTALK_WR) + hw->need_crosstalk_fix = false; + else + hw->need_crosstalk_fix = true; + break; + default: + hw->need_crosstalk_fix = false; + break; + } + /* Clear adapter stopped flag */ hw->adapter_stopped = false; @@ -3200,6 +3217,31 @@ s32 ixgbe_clear_vfta_generic(struct ixgbe_hw *hw) } /** + * ixgbe_need_crosstalk_fix - Determine if we need to do cross talk fix + * @hw: pointer to hardware structure + * + * Contains the logic to identify if we need to verify link for the + * crosstalk fix + **/ +static bool ixgbe_need_crosstalk_fix(struct ixgbe_hw *hw) +{ + /* Does FW say we need the fix */ + if (!hw->need_crosstalk_fix) + return false; + + /* Only consider SFP+ PHYs i.e. media type fiber */ + switch (hw->mac.ops.get_media_type(hw)) { + case ixgbe_media_type_fiber: + case ixgbe_media_type_fiber_qsfp: + break; + default: + return false; + } + + return true; +} + +/** * ixgbe_check_mac_link_generic - Determine link and speed status * @hw: pointer to hardware structure * @speed: pointer to link speed @@ -3214,6 +3256,35 @@ s32 ixgbe_check_mac_link_generic(struct ixgbe_hw *hw, ixgbe_link_speed *speed, u32 links_reg, links_orig; u32 i; + /* If Crosstalk fix enabled do the sanity check of making sure + * the SFP+ cage is full. + */ + if (ixgbe_need_crosstalk_fix(hw)) { + u32 sfp_cage_full; + + switch (hw->mac.type) { + case ixgbe_mac_82599EB: + sfp_cage_full = IXGBE_READ_REG(hw, IXGBE_ESDP) & + IXGBE_ESDP_SDP2; + break; + case ixgbe_mac_X550EM_x: + case ixgbe_mac_x550em_a: + sfp_cage_full = IXGBE_READ_REG(hw, IXGBE_ESDP) & + IXGBE_ESDP_SDP0; + break; + default: + /* sanity check - No SFP+ devices here */ + sfp_cage_full = false; + break; + } + + if (!sfp_cage_full) { + *link_up = false; + *speed = IXGBE_LINK_SPEED_UNKNOWN; + return 0; + } + } + /* clear the old state */ links_orig = IXGBE_READ_REG(hw, IXGBE_LINKS); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index b0039c0..4845c9f 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -5625,7 +5625,6 @@ static int ixgbe_sw_init(struct ixgbe_adapter *adapter) struct pci_dev *pdev = adapter->pdev; unsigned int rss, fdir; u32 fwsm; - u16 device_caps; int i; /* PCI config space info */ @@ -5771,22 +5770,6 @@ static int ixgbe_sw_init(struct ixgbe_adapter *adapter) adapter->tx_ring_count = IXGBE_DEFAULT_TXD; adapter->rx_ring_count = IXGBE_DEFAULT_RXD; - /* Cache bit indicating need for crosstalk fix */ - switch (hw->mac.type) { - case ixgbe_mac_82599EB: - case ixgbe_mac_X550EM_x: - case ixgbe_mac_x550em_a: - hw->mac.ops.get_device_caps(hw, &device_caps); - if (device_caps & IXGBE_DEVICE_CAPS_NO_CROSSTALK_WR) - adapter->need_crosstalk_fix = false; - else - adapter->need_crosstalk_fix = true; - break; - default: - adapter->need_crosstalk_fix = false; - break; - } - /* set default work limits */ adapter->tx_work_limit = IXGBE_DEFAULT_TX_WORK; @@ -6707,18 +6690,6 @@ static void ixgbe_watchdog_update_link(struct ixgbe_adapter *adapter) link_up = true; } - /* If Crosstalk fix enabled do the sanity check of making sure - * the SFP+ cage is empty. - */ - if (adapter->need_crosstalk_fix) { - u32 sfp_cage_full; - - sfp_cage_full = IXGBE_READ_REG(hw, IXGBE_ESDP) & - IXGBE_ESDP_SDP2; - if (ixgbe_is_sfp(hw) && link_up && !sfp_cage_full) - link_up = false; - } - if (adapter->ixgbe_ieee_pfc) pfc_en |= !!(adapter->ixgbe_ieee_pfc->pfc_en); @@ -7065,16 +7036,6 @@ static void ixgbe_sfp_detection_subtask(struct ixgbe_adapter *adapter) struct ixgbe_hw *hw = &adapter->hw; s32 err; - /* If crosstalk fix enabled verify the SFP+ cage is full */ - if (adapter->need_crosstalk_fix) { - u32 sfp_cage_full; - - sfp_cage_full = IXGBE_READ_REG(hw, IXGBE_ESDP) & - IXGBE_ESDP_SDP2; - if (!sfp_cage_full) - return; - } - /* not searching for SFP so there is nothing to do here */ if (!(adapter->flags2 & IXGBE_FLAG2_SEARCH_FOR_SFP) && !(adapter->flags2 & IXGBE_FLAG2_SFP_NEEDS_RESET)) diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h index da3d835..1248a99 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h @@ -3525,6 +3525,7 @@ struct ixgbe_hw { bool force_full_reset; bool allow_unsupported_sfp; bool wol_enabled; + bool need_crosstalk_fix; }; struct ixgbe_info { -- cgit v0.10.2 From 23014011ba4209a086931ff402eac1c41abbe456 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 21 Jul 2016 12:51:16 +0200 Subject: netfilter: conntrack: support a fixed size of 128 distinct labels The conntrack label extension is currently variable-sized, e.g. if only 2 labels are used by iptables rules then the labels->bits[] array will only contain one element. We track size of each label storage area in the 'words' member. But in nftables and openvswitch we always have to ask for worst-case since we don't know what bit will be used at configuration time. As most arches are 64bit we need to allocate 24 bytes in this case: struct nf_conn_labels { u8 words; /* 0 1 */ /* XXX 7 bytes hole, try to pack */ long unsigned bits[2]; /* 8 24 */ Make bits a fixed size and drop the words member, it simplifies the code and only increases memory requirements on x86 when less than 64bit labels are required. We still only allocate the extension if its needed. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_conntrack_labels.h b/include/net/netfilter/nf_conntrack_labels.h index c5f8fc73..0fd4989 100644 --- a/include/net/netfilter/nf_conntrack_labels.h +++ b/include/net/netfilter/nf_conntrack_labels.h @@ -10,8 +10,7 @@ #define NF_CT_LABELS_MAX_SIZE ((XT_CONNLABEL_MAXBIT + 1) / BITS_PER_BYTE) struct nf_conn_labels { - u8 words; - unsigned long bits[]; + unsigned long bits[NF_CT_LABELS_MAX_SIZE / sizeof(long)]; }; static inline struct nf_conn_labels *nf_ct_labels_find(const struct nf_conn *ct) @@ -26,20 +25,13 @@ static inline struct nf_conn_labels *nf_ct_labels_find(const struct nf_conn *ct) static inline struct nf_conn_labels *nf_ct_labels_ext_add(struct nf_conn *ct) { #ifdef CONFIG_NF_CONNTRACK_LABELS - struct nf_conn_labels *cl_ext; struct net *net = nf_ct_net(ct); - u8 words; - words = ACCESS_ONCE(net->ct.label_words); - if (words == 0) + if (net->ct.labels_used == 0) return NULL; - cl_ext = nf_ct_ext_add_length(ct, NF_CT_EXT_LABELS, - words * sizeof(long), GFP_ATOMIC); - if (cl_ext != NULL) - cl_ext->words = words; - - return cl_ext; + return nf_ct_ext_add_length(ct, NF_CT_EXT_LABELS, + sizeof(struct nf_conn_labels), GFP_ATOMIC); #else return NULL; #endif diff --git a/net/netfilter/nf_conntrack_labels.c b/net/netfilter/nf_conntrack_labels.c index 252e6a7..7686200 100644 --- a/net/netfilter/nf_conntrack_labels.c +++ b/net/netfilter/nf_conntrack_labels.c @@ -20,7 +20,7 @@ int nf_connlabel_set(struct nf_conn *ct, u16 bit) { struct nf_conn_labels *labels = nf_ct_labels_find(ct); - if (!labels || BIT_WORD(bit) >= labels->words) + if (!labels) return -ENOSPC; if (test_bit(bit, labels->bits)) @@ -60,7 +60,7 @@ int nf_connlabels_replace(struct nf_conn *ct, if (!labels) return -ENOSPC; - size = labels->words * sizeof(long); + size = sizeof(labels->bits); if (size < (words32 * sizeof(u32))) words32 = size / sizeof(u32); @@ -80,16 +80,11 @@ EXPORT_SYMBOL_GPL(nf_connlabels_replace); int nf_connlabels_get(struct net *net, unsigned int bits) { - size_t words; - - words = BIT_WORD(bits) + 1; - if (words > NF_CT_LABELS_MAX_SIZE / sizeof(long)) + if (BIT_WORD(bits) >= NF_CT_LABELS_MAX_SIZE / sizeof(long)) return -ERANGE; spin_lock(&nf_connlabels_lock); net->ct.labels_used++; - if (words > net->ct.label_words) - net->ct.label_words = words; spin_unlock(&nf_connlabels_lock); return 0; @@ -100,8 +95,6 @@ void nf_connlabels_put(struct net *net) { spin_lock(&nf_connlabels_lock); net->ct.labels_used--; - if (net->ct.labels_used == 0) - net->ct.label_words = 0; spin_unlock(&nf_connlabels_lock); } EXPORT_SYMBOL_GPL(nf_connlabels_put); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index a18d1ce..050bb34 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -346,25 +346,25 @@ static inline int ctnetlink_label_size(const struct nf_conn *ct) if (!labels) return 0; - return nla_total_size(labels->words * sizeof(long)); + return nla_total_size(sizeof(labels->bits)); } static int ctnetlink_dump_labels(struct sk_buff *skb, const struct nf_conn *ct) { struct nf_conn_labels *labels = nf_ct_labels_find(ct); - unsigned int len, i; + unsigned int i; if (!labels) return 0; - len = labels->words * sizeof(long); i = 0; do { if (labels->bits[i] != 0) - return nla_put(skb, CTA_LABELS, len, labels->bits); + return nla_put(skb, CTA_LABELS, sizeof(labels->bits), + labels->bits); i++; - } while (i < labels->words); + } while (i < ARRAY_SIZE(labels->bits)); return 0; } diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index d9e44ca..2f47d5d 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -113,18 +113,11 @@ static void nft_ct_get_eval(const struct nft_expr *expr, #ifdef CONFIG_NF_CONNTRACK_LABELS case NFT_CT_LABELS: { struct nf_conn_labels *labels = nf_ct_labels_find(ct); - unsigned int size; - if (!labels) { + if (labels) + memcpy(dest, labels->bits, NF_CT_LABELS_MAX_SIZE); + else memset(dest, 0, NF_CT_LABELS_MAX_SIZE); - return; - } - - size = labels->words * sizeof(long); - memcpy(dest, labels->bits, size); - if (size < NF_CT_LABELS_MAX_SIZE) - memset(((char *) dest) + size, 0, - NF_CT_LABELS_MAX_SIZE - size); return; } #endif diff --git a/net/netfilter/xt_connlabel.c b/net/netfilter/xt_connlabel.c index a79af25..c9fba8a 100644 --- a/net/netfilter/xt_connlabel.c +++ b/net/netfilter/xt_connlabel.c @@ -25,7 +25,7 @@ static bool connlabel_match(const struct nf_conn *ct, u16 bit) if (!labels) return false; - return BIT_WORD(bit) < labels->words && test_bit(bit, labels->bits); + return test_bit(bit, labels->bits); } static bool diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index b4069a9..c644c78 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -135,7 +135,7 @@ static void ovs_ct_get_labels(const struct nf_conn *ct, struct nf_conn_labels *cl = ct ? nf_ct_labels_find(ct) : NULL; if (cl) { - size_t len = cl->words * sizeof(long); + size_t len = sizeof(cl->bits); if (len > OVS_CT_LABELS_LEN) len = OVS_CT_LABELS_LEN; @@ -274,7 +274,7 @@ static int ovs_ct_set_labels(struct sk_buff *skb, struct sw_flow_key *key, nf_ct_labels_ext_add(ct); cl = nf_ct_labels_find(ct); } - if (!cl || cl->words * sizeof(long) < OVS_CT_LABELS_LEN) + if (!cl || sizeof(cl->bits) < OVS_CT_LABELS_LEN) return -ENOSPC; err = nf_connlabels_replace(ct, (u32 *)labels, (u32 *)mask, -- cgit v0.10.2 From 857ed310c013fe0d0059f955048dab589fa7a57a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 21 Jul 2016 12:51:17 +0200 Subject: netfilter: connlabels: move set helper to xt_connlabel xt_connlabel is the only user so move it. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso diff --git a/include/net/netfilter/nf_conntrack_labels.h b/include/net/netfilter/nf_conntrack_labels.h index 0fd4989..4988146 100644 --- a/include/net/netfilter/nf_conntrack_labels.h +++ b/include/net/netfilter/nf_conntrack_labels.h @@ -37,8 +37,6 @@ static inline struct nf_conn_labels *nf_ct_labels_ext_add(struct nf_conn *ct) #endif } -int nf_connlabel_set(struct nf_conn *ct, u16 bit); - int nf_connlabels_replace(struct nf_conn *ct, const u32 *data, const u32 *mask, unsigned int words); diff --git a/net/netfilter/nf_conntrack_labels.c b/net/netfilter/nf_conntrack_labels.c index 7686200..bcab8bd 100644 --- a/net/netfilter/nf_conntrack_labels.c +++ b/net/netfilter/nf_conntrack_labels.c @@ -16,23 +16,6 @@ static spinlock_t nf_connlabels_lock; -int nf_connlabel_set(struct nf_conn *ct, u16 bit) -{ - struct nf_conn_labels *labels = nf_ct_labels_find(ct); - - if (!labels) - return -ENOSPC; - - if (test_bit(bit, labels->bits)) - return 0; - - if (!test_and_set_bit(bit, labels->bits)) - nf_conntrack_event_cache(IPCT_LABEL, ct); - - return 0; -} -EXPORT_SYMBOL_GPL(nf_connlabel_set); - static int replace_u32(u32 *address, u32 mask, u32 new) { u32 old, tmp; diff --git a/net/netfilter/xt_connlabel.c b/net/netfilter/xt_connlabel.c index c9fba8a..03d66f1 100644 --- a/net/netfilter/xt_connlabel.c +++ b/net/netfilter/xt_connlabel.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include @@ -18,21 +19,12 @@ MODULE_DESCRIPTION("Xtables: add/match connection trackling labels"); MODULE_ALIAS("ipt_connlabel"); MODULE_ALIAS("ip6t_connlabel"); -static bool connlabel_match(const struct nf_conn *ct, u16 bit) -{ - struct nf_conn_labels *labels = nf_ct_labels_find(ct); - - if (!labels) - return false; - - return test_bit(bit, labels->bits); -} - static bool connlabel_mt(const struct sk_buff *skb, struct xt_action_param *par) { const struct xt_connlabel_mtinfo *info = par->matchinfo; enum ip_conntrack_info ctinfo; + struct nf_conn_labels *labels; struct nf_conn *ct; bool invert = info->options & XT_CONNLABEL_OP_INVERT; @@ -40,10 +32,21 @@ connlabel_mt(const struct sk_buff *skb, struct xt_action_param *par) if (ct == NULL || nf_ct_is_untracked(ct)) return invert; - if (info->options & XT_CONNLABEL_OP_SET) - return (nf_connlabel_set(ct, info->bit) == 0) ^ invert; + labels = nf_ct_labels_find(ct); + if (!labels) + return invert; + + if (test_bit(info->bit, labels->bits)) + return !invert; + + if (info->options & XT_CONNLABEL_OP_SET) { + if (!test_and_set_bit(info->bit, labels->bits)) + nf_conntrack_event_cache(IPCT_LABEL, ct); + + return !invert; + } - return connlabel_match(ct, info->bit) ^ invert; + return invert; } static int connlabel_mt_check(const struct xt_mtchk_param *par) -- cgit v0.10.2 From 96d1327ac2e3dc3ac4204fe3656dad0043fc0efd Mon Sep 17 00:00:00 2001 From: Gao Feng Date: Fri, 22 Jul 2016 12:59:15 +0800 Subject: netfilter: h323: Use mod_timer instead of set_expect_timeout Simplify the code without any side effect. The set_expect_timeout is used to modify the timer expired time. It tries to delete timer, and add it again. So we could use mod_timer directly. Signed-off-by: Gao Feng Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c index 9511af0..bb77a97 100644 --- a/net/netfilter/nf_conntrack_h323_main.c +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -1273,19 +1273,6 @@ static struct nf_conntrack_expect *find_expect(struct nf_conn *ct, } /****************************************************************************/ -static int set_expect_timeout(struct nf_conntrack_expect *exp, - unsigned int timeout) -{ - if (!exp || !del_timer(&exp->timeout)) - return 0; - - exp->timeout.expires = jiffies + timeout * HZ; - add_timer(&exp->timeout); - - return 1; -} - -/****************************************************************************/ static int expect_q931(struct sk_buff *skb, struct nf_conn *ct, enum ip_conntrack_info ctinfo, unsigned int protoff, unsigned char **data, @@ -1486,7 +1473,7 @@ static int process_rcf(struct sk_buff *skb, struct nf_conn *ct, "timeout to %u seconds for", info->timeout); nf_ct_dump_tuple(&exp->tuple); - set_expect_timeout(exp, info->timeout); + mod_timer(&exp->timeout, jiffies + info->timeout * HZ); } spin_unlock_bh(&nf_conntrack_expect_lock); } -- cgit v0.10.2 From 2bf4fade54de995f84d319ae02003609dca450f3 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sat, 23 Jul 2016 16:00:31 +0800 Subject: netfilter: nft_compat: put back match/target module if init fail If the user specify the invalid NFTA_MATCH_INFO/NFTA_TARGET_INFO attr or memory alloc fail, we should call module_put to the related match or target. Otherwise, we cannot remove the module even nobody use it. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c index 6228c42..2e07cec 100644 --- a/net/netfilter/nft_compat.c +++ b/net/netfilter/nft_compat.c @@ -634,6 +634,7 @@ nft_match_select_ops(const struct nft_ctx *ctx, struct xt_match *match; char *mt_name; u32 rev, family; + int err; if (tb[NFTA_MATCH_NAME] == NULL || tb[NFTA_MATCH_REV] == NULL || @@ -660,13 +661,17 @@ nft_match_select_ops(const struct nft_ctx *ctx, if (IS_ERR(match)) return ERR_PTR(-ENOENT); - if (match->matchsize > nla_len(tb[NFTA_MATCH_INFO])) - return ERR_PTR(-EINVAL); + if (match->matchsize > nla_len(tb[NFTA_MATCH_INFO])) { + err = -EINVAL; + goto err; + } /* This is the first time we use this match, allocate operations */ nft_match = kzalloc(sizeof(struct nft_xt), GFP_KERNEL); - if (nft_match == NULL) - return ERR_PTR(-ENOMEM); + if (nft_match == NULL) { + err = -ENOMEM; + goto err; + } nft_match->ops.type = &nft_match_type; nft_match->ops.size = NFT_EXPR_SIZE(XT_ALIGN(match->matchsize)); @@ -680,6 +685,9 @@ nft_match_select_ops(const struct nft_ctx *ctx, list_add(&nft_match->head, &nft_match_list); return &nft_match->ops; +err: + module_put(match->me); + return ERR_PTR(err); } static void nft_match_release(void) @@ -717,6 +725,7 @@ nft_target_select_ops(const struct nft_ctx *ctx, struct xt_target *target; char *tg_name; u32 rev, family; + int err; if (tb[NFTA_TARGET_NAME] == NULL || tb[NFTA_TARGET_REV] == NULL || @@ -743,13 +752,17 @@ nft_target_select_ops(const struct nft_ctx *ctx, if (IS_ERR(target)) return ERR_PTR(-ENOENT); - if (target->targetsize > nla_len(tb[NFTA_TARGET_INFO])) - return ERR_PTR(-EINVAL); + if (target->targetsize > nla_len(tb[NFTA_TARGET_INFO])) { + err = -EINVAL; + goto err; + } /* This is the first time we use this target, allocate operations */ nft_target = kzalloc(sizeof(struct nft_xt), GFP_KERNEL); - if (nft_target == NULL) - return ERR_PTR(-ENOMEM); + if (nft_target == NULL) { + err = -ENOMEM; + goto err; + } nft_target->ops.type = &nft_target_type; nft_target->ops.size = NFT_EXPR_SIZE(XT_ALIGN(target->targetsize)); @@ -767,6 +780,9 @@ nft_target_select_ops(const struct nft_ctx *ctx, list_add(&nft_target->head, &nft_target_list); return &nft_target->ops; +err: + module_put(target->me); + return ERR_PTR(err); } static void nft_target_release(void) -- cgit v0.10.2 From 4b512e1c1f8de6b9ceb796ecef8658e0a083cab7 Mon Sep 17 00:00:00 2001 From: Liping Zhang Date: Sat, 23 Jul 2016 16:00:32 +0800 Subject: netfilter: nft_compat: fix crash when related match/target module is removed We "cache" the loaded match/target modules and reuse them, but when the modules are removed, we still point to them. Then we may end up with invalid memory references when using iptables-compat to add rules later. Input the following commands will reproduce the kernel crash: # iptables-compat -A INPUT -j LOG # iptables-compat -D INPUT -j LOG # rmmod xt_LOG # iptables-compat -A INPUT -j LOG BUG: unable to handle kernel paging request at ffffffffa05a9010 IP: [] strcmp+0xe/0x30 Call Trace: [] nft_target_select_ops+0x83/0x1f0 [nft_compat] [] nf_tables_expr_parse+0x147/0x1f0 [nf_tables] [] nf_tables_newrule+0x301/0x810 [nf_tables] [] ? nla_parse+0x20/0x100 [] nfnetlink_rcv+0x33f/0x53d [nfnetlink] [] ? nfnetlink_rcv+0x1fb/0x53d [nfnetlink] [] netlink_unicast+0x178/0x220 [] netlink_sendmsg+0x2fb/0x3a0 [] sock_sendmsg+0x38/0x50 [] ___sys_sendmsg+0x28e/0x2a0 [] ? release_sock+0x1e/0xb0 [] ? _raw_spin_unlock_bh+0x35/0x40 [] ? release_sock+0x82/0xb0 [] __sys_sendmsg+0x54/0x90 [] SyS_sendmsg+0x12/0x20 [] entry_SYSCALL_64_fastpath+0x1a/0xa9 So when nobody use the related match/target module, there's no need to "cache" it. And nft_[match|target]_release are useless anymore, remove them. Signed-off-by: Liping Zhang Signed-off-by: Pablo Neira Ayuso diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c index 2e07cec..c21e7eb 100644 --- a/net/netfilter/nft_compat.c +++ b/net/netfilter/nft_compat.c @@ -23,6 +23,20 @@ #include #include +struct nft_xt { + struct list_head head; + struct nft_expr_ops ops; + unsigned int refcnt; +}; + +static void nft_xt_put(struct nft_xt *xt) +{ + if (--xt->refcnt == 0) { + list_del(&xt->head); + kfree(xt); + } +} + static int nft_compat_chain_validate_dependency(const char *tablename, const struct nft_chain *chain) { @@ -260,6 +274,7 @@ nft_target_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr) if (par.target->destroy != NULL) par.target->destroy(&par); + nft_xt_put(container_of(expr->ops, struct nft_xt, ops)); module_put(target->me); } @@ -442,6 +457,7 @@ nft_match_destroy(const struct nft_ctx *ctx, const struct nft_expr *expr) if (par.match->destroy != NULL) par.match->destroy(&par); + nft_xt_put(container_of(expr->ops, struct nft_xt, ops)); module_put(match->me); } @@ -612,11 +628,6 @@ static const struct nfnetlink_subsystem nfnl_compat_subsys = { static LIST_HEAD(nft_match_list); -struct nft_xt { - struct list_head head; - struct nft_expr_ops ops; -}; - static struct nft_expr_type nft_match_type; static bool nft_match_cmp(const struct xt_match *match, @@ -653,6 +664,7 @@ nft_match_select_ops(const struct nft_ctx *ctx, if (!try_module_get(match->me)) return ERR_PTR(-ENOENT); + nft_match->refcnt++; return &nft_match->ops; } } @@ -673,6 +685,7 @@ nft_match_select_ops(const struct nft_ctx *ctx, goto err; } + nft_match->refcnt = 1; nft_match->ops.type = &nft_match_type; nft_match->ops.size = NFT_EXPR_SIZE(XT_ALIGN(match->matchsize)); nft_match->ops.eval = nft_match_eval; @@ -690,14 +703,6 @@ err: return ERR_PTR(err); } -static void nft_match_release(void) -{ - struct nft_xt *nft_match, *tmp; - - list_for_each_entry_safe(nft_match, tmp, &nft_match_list, head) - kfree(nft_match); -} - static struct nft_expr_type nft_match_type __read_mostly = { .name = "match", .select_ops = nft_match_select_ops, @@ -744,6 +749,7 @@ nft_target_select_ops(const struct nft_ctx *ctx, if (!try_module_get(target->me)) return ERR_PTR(-ENOENT); + nft_target->refcnt++; return &nft_target->ops; } } @@ -764,6 +770,7 @@ nft_target_select_ops(const struct nft_ctx *ctx, goto err; } + nft_target->refcnt = 1; nft_target->ops.type = &nft_target_type; nft_target->ops.size = NFT_EXPR_SIZE(XT_ALIGN(target->targetsize)); nft_target->ops.init = nft_target_init; @@ -785,14 +792,6 @@ err: return ERR_PTR(err); } -static void nft_target_release(void) -{ - struct nft_xt *nft_target, *tmp; - - list_for_each_entry_safe(nft_target, tmp, &nft_target_list, head) - kfree(nft_target); -} - static struct nft_expr_type nft_target_type __read_mostly = { .name = "target", .select_ops = nft_target_select_ops, @@ -835,8 +834,6 @@ static void __exit nft_compat_module_exit(void) nfnetlink_subsys_unregister(&nfnl_compat_subsys); nft_unregister_expr(&nft_target_type); nft_unregister_expr(&nft_match_type); - nft_match_release(); - nft_target_release(); } MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_NFT_COMPAT); -- cgit v0.10.2 From bfe9b9d2df669a57a95d641ed46eb018e204c6ce Mon Sep 17 00:00:00 2001 From: Kristian Evensen Date: Thu, 21 Jul 2016 11:10:06 +0200 Subject: cdc_ether: Improve ZTE MF823/831/910 handling The firmware in several ZTE devices (at least the MF823/831/910 modems/mifis) use OS fingerprinting to determine which type of device to export. In addition, these devices export a REST API which can be used to control the type of device. So far, on Linux, the devices have been seen as RNDIS or CDC Ether. When CDC Ether is used, devices of the same type are, as with RNDIS, exported with the same, bogus random MAC address. In addition, the devices (at least on all firmware revisions I have found) use the bogus MAC when sending traffic routed from external networks. And as a final feature, the devices sometimes export the link state incorrectly. There are also references online to several other ZTE devices displaying this behavior, with several different PIDs and MAC addresses. This patch tries to improve the handling of ZTE devices by doing the following: * Create a new driver_info-struct that is used by ZTE devices that do not have an explicit entry in the product table. This struct is the same as the default cdc_ether driver info, but a new bind- and an rx_fixup-function have been added. * In the new bind function, we check if we have read a random MAC from the device. If we have, then we generate a new random MAC address. This will ensure that all devices get a unique MAC. * The rx_fixup-function replaces the destination MAC address in the skb with that of the device. I have not seen a revision of these devices that behaves correctly (i.e., sets the right destination MAC), so I chose not to do any comparison with for example the known, bogus addresses. * The MF823/MF832/MF910 sometimes export cdc carrier on twice on connect (the correct behavior is off then on). Work around this by manually setting carrier to off if an on-notification is received and the NOCARRIER-bit is not set. This change will affect all devices, but it should take care of similar mistakes made by other manufacturers. I tried to think of/look/test for problems/regressions that could be introduced by this behavior, but could not find any. However, my familiarity with this code path is not that great, so there could be something I have overlooked. I have tested this patch with multiple revisions of all three devices, and they behave as expected. In other words, they all got a valid, random MAC, the correct operational state and I can receive/sent traffic without problems. I also tested with some other cdc_ether devices I have and did not find any problems/regressions caused by the two general changes. v3->v4: * Forgot to remove unused variables, sorry about that (thanks David Miller). v2->v3: * I had forgot to remove the random MAC generation from usbnet_cdc_bind() (thanks Oliver). * Rework logic in the ZTE bind-function a bit. v1->v2: * Only generate random MAC for ZTE devices (thanks Oliver Neukum). * Set random MAC and do RX fixup for all ZTE devices that do not have a product-entry, as the bogus MAC have been seen on devices with several different PIDs/MAC addresses. In other words, it seems to be the default behavior of ZTE CDC Ether devices (thanks Lars Melin). Signed-off-by: Kristian Evensen Acked-by: Oliver Neukum Signed-off-by: David S. Miller diff --git a/drivers/net/usb/cdc_ether.c b/drivers/net/usb/cdc_ether.c index 7cba2c3..c47ec0a 100644 --- a/drivers/net/usb/cdc_ether.c +++ b/drivers/net/usb/cdc_ether.c @@ -388,6 +388,12 @@ void usbnet_cdc_status(struct usbnet *dev, struct urb *urb) case USB_CDC_NOTIFY_NETWORK_CONNECTION: netif_dbg(dev, timer, dev->net, "CDC: carrier %s\n", event->wValue ? "on" : "off"); + + /* Work-around for devices with broken off-notifications */ + if (event->wValue && + !test_bit(__LINK_STATE_NOCARRIER, &dev->net->state)) + usbnet_link_change(dev, 0, 0); + usbnet_link_change(dev, !!event->wValue, 0); break; case USB_CDC_NOTIFY_SPEED_CHANGE: /* tx/rx rates */ @@ -432,6 +438,34 @@ int usbnet_cdc_bind(struct usbnet *dev, struct usb_interface *intf) } EXPORT_SYMBOL_GPL(usbnet_cdc_bind); +static int usbnet_cdc_zte_bind(struct usbnet *dev, struct usb_interface *intf) +{ + int status = usbnet_cdc_bind(dev, intf); + + if (!status && (dev->net->dev_addr[0] & 0x02)) + eth_hw_addr_random(dev->net); + + return status; +} + +/* Make sure packets have correct destination MAC address + * + * A firmware bug observed on some devices (ZTE MF823/831/910) is that the + * device sends packets with a static, bogus, random MAC address (event if + * device MAC address has been updated). Always set MAC address to that of the + * device. + */ +static int usbnet_cdc_zte_rx_fixup(struct usbnet *dev, struct sk_buff *skb) +{ + if (skb->len < ETH_HLEN || !(skb->data[0] & 0x02)) + return 1; + + skb_reset_mac_header(skb); + ether_addr_copy(eth_hdr(skb)->h_dest, dev->net->dev_addr); + + return 1; +} + static const struct driver_info cdc_info = { .description = "CDC Ethernet Device", .flags = FLAG_ETHER | FLAG_POINTTOPOINT, @@ -442,6 +476,17 @@ static const struct driver_info cdc_info = { .manage_power = usbnet_manage_power, }; +static const struct driver_info zte_cdc_info = { + .description = "ZTE CDC Ethernet Device", + .flags = FLAG_ETHER | FLAG_POINTTOPOINT, + .bind = usbnet_cdc_zte_bind, + .unbind = usbnet_cdc_unbind, + .status = usbnet_cdc_status, + .set_rx_mode = usbnet_cdc_update_filter, + .manage_power = usbnet_manage_power, + .rx_fixup = usbnet_cdc_zte_rx_fixup, +}; + static const struct driver_info wwan_info = { .description = "Mobile Broadband Network Device", .flags = FLAG_WWAN, @@ -707,6 +752,12 @@ static const struct usb_device_id products[] = { USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE), .driver_info = (kernel_ulong_t)&wwan_info, }, { + /* ZTE modules */ + USB_VENDOR_AND_INTERFACE_INFO(ZTE_VENDOR_ID, USB_CLASS_COMM, + USB_CDC_SUBCLASS_ETHERNET, + USB_CDC_PROTO_NONE), + .driver_info = (unsigned long)&zte_cdc_info, +}, { USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE), .driver_info = (unsigned long) &cdc_info, -- cgit v0.10.2 From 57d316ba2017d824ed0fd95fc98b020c834023a1 Mon Sep 17 00:00:00 2001 From: Nogah Frankel Date: Thu, 21 Jul 2016 12:03:09 +0200 Subject: mlxsw: pci: Add resources query implementation. Add resources query implementation. If exists, query the HW for its builtin resources instead of having them as consts in the code. Signed-off-by: Nogah Frankel Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/cmd.h b/drivers/net/ethernet/mellanox/mlxsw/cmd.h index f9cd6e3..28271be 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/cmd.h +++ b/drivers/net/ethernet/mellanox/mlxsw/cmd.h @@ -105,6 +105,7 @@ enum mlxsw_cmd_opcode { MLXSW_CMD_OPCODE_SW2HW_EQ = 0x013, MLXSW_CMD_OPCODE_HW2SW_EQ = 0x014, MLXSW_CMD_OPCODE_QUERY_EQ = 0x015, + MLXSW_CMD_OPCODE_QUERY_RESOURCES = 0x101, }; static inline const char *mlxsw_cmd_opcode_str(u16 opcode) @@ -144,6 +145,8 @@ static inline const char *mlxsw_cmd_opcode_str(u16 opcode) return "HW2SW_EQ"; case MLXSW_CMD_OPCODE_QUERY_EQ: return "QUERY_EQ"; + case MLXSW_CMD_OPCODE_QUERY_RESOURCES: + return "QUERY_RESOURCES"; default: return "*UNKNOWN*"; } @@ -500,6 +503,35 @@ static inline int mlxsw_cmd_unmap_fa(struct mlxsw_core *mlxsw_core) return mlxsw_cmd_exec_none(mlxsw_core, MLXSW_CMD_OPCODE_UNMAP_FA, 0, 0); } +/* QUERY_RESOURCES - Query chip resources + * -------------------------------------- + * OpMod == 0 (N/A) , INMmod is index + * ---------------------------------- + * The QUERY_RESOURCES command retrieves information related to chip resources + * by resource ID. Every command returns 32 entries. INmod is being use as base. + * for example, index 1 will return entries 32-63. When the tables end and there + * are no more sources in the table, will return resource id 0xFFF to indicate + * it. + */ +static inline int mlxsw_cmd_query_resources(struct mlxsw_core *mlxsw_core, + char *out_mbox, int index) +{ + return mlxsw_cmd_exec_out(mlxsw_core, MLXSW_CMD_OPCODE_QUERY_RESOURCES, + 0, index, false, out_mbox, + MLXSW_CMD_MBOX_SIZE); +} + +/* cmd_mbox_query_resource_id + * The resource id. 0xFFFF indicates table's end. + */ +MLXSW_ITEM32_INDEXED(cmd_mbox, query_resource, id, 0x00, 16, 16, 0x8, 0, false); + +/* cmd_mbox_query_resource_data + * The resource + */ +MLXSW_ITEM64_INDEXED(cmd_mbox, query_resource, data, + 0x00, 0, 40, 0x8, 0, false); + /* CONFIG_PROFILE (Set) - Configure Switch Profile * ------------------------------ * OpMod == 1 (Set), INMmod == 0 (N/A) diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index 480a3ba..068ee65a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -111,6 +111,7 @@ struct mlxsw_core { struct { u8 *mapping; /* lag_id+port_index to local_port mapping */ } lag; + struct mlxsw_resources resources; struct mlxsw_hwmon *hwmon; unsigned long driver_priv[0]; /* driver_priv has to be always the last item */ @@ -1110,7 +1111,8 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info, } } - err = mlxsw_bus->init(bus_priv, mlxsw_core, mlxsw_driver->profile); + err = mlxsw_bus->init(bus_priv, mlxsw_core, mlxsw_driver->profile, + &mlxsw_core->resources); if (err) goto err_bus_init; @@ -1652,6 +1654,12 @@ void mlxsw_core_lag_mapping_clear(struct mlxsw_core *mlxsw_core, } EXPORT_SYMBOL(mlxsw_core_lag_mapping_clear); +struct mlxsw_resources *mlxsw_core_resources_get(struct mlxsw_core *mlxsw_core) +{ + return &mlxsw_core->resources; +} +EXPORT_SYMBOL(mlxsw_core_resources_get); + int mlxsw_core_port_init(struct mlxsw_core *mlxsw_core, struct mlxsw_core_port *mlxsw_core_port, u8 local_port, struct net_device *dev, bool split, u32 split_group) diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h index 2fe385c..d57ad0d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.h +++ b/drivers/net/ethernet/mellanox/mlxsw/core.h @@ -215,6 +215,7 @@ struct mlxsw_config_profile { u32 kvd_linear_size; u32 kvd_hash_single_size; u32 kvd_hash_double_size; + u8 resource_query_enable; struct mlxsw_swid_config swid_config[MLXSW_CONFIG_PROFILE_SWID_COUNT]; }; @@ -266,10 +267,16 @@ struct mlxsw_driver { const struct mlxsw_config_profile *profile; }; +struct mlxsw_resources { +}; + +struct mlxsw_resources *mlxsw_core_resources_get(struct mlxsw_core *mlxsw_core); + struct mlxsw_bus { const char *kind; int (*init)(void *bus_priv, struct mlxsw_core *mlxsw_core, - const struct mlxsw_config_profile *profile); + const struct mlxsw_config_profile *profile, + struct mlxsw_resources *resources); void (*fini)(void *bus_priv); bool (*skb_transmit_busy)(void *bus_priv, const struct mlxsw_tx_info *tx_info); diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c index ddbc9f2..a724b66 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci.c +++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c @@ -1154,6 +1154,56 @@ mlxsw_pci_config_profile_swid_config(struct mlxsw_pci *mlxsw_pci, mlxsw_cmd_mbox_config_profile_swid_config_mask_set(mbox, index, mask); } +#define MLXSW_RESOURCES_TABLE_END_ID 0xffff +#define MLXSW_RESOURCES_QUERY_MAX_QUERIES 100 +#define MLXSW_RESOURCES_PER_QUERY 32 + +static void mlxsw_pci_resources_query_parse(int id, u64 val, + struct mlxsw_resources *resources) +{ + switch (id) { + default: + break; + } +} + +static int mlxsw_pci_resources_query(struct mlxsw_pci *mlxsw_pci, char *mbox, + struct mlxsw_resources *resources, + u8 query_enabled) +{ + int index, i; + u64 data; + u16 id; + int err; + + /* Not all the versions support resources query */ + if (!query_enabled) + return 0; + + mlxsw_cmd_mbox_zero(mbox); + + for (index = 0; index < MLXSW_RESOURCES_QUERY_MAX_QUERIES; index++) { + err = mlxsw_cmd_query_resources(mlxsw_pci->core, mbox, index); + if (err) + return err; + + for (i = 0; i < MLXSW_RESOURCES_PER_QUERY; i++) { + id = mlxsw_cmd_mbox_query_resource_id_get(mbox, i); + data = mlxsw_cmd_mbox_query_resource_data_get(mbox, i); + + if (id == MLXSW_RESOURCES_TABLE_END_ID) + return 0; + + mlxsw_pci_resources_query_parse(id, data, resources); + } + } + + /* If after MLXSW_RESOURCES_QUERY_MAX_QUERIES we still didn't get + * MLXSW_RESOURCES_TABLE_END_ID, something went bad in the FW. + */ + return -EIO; +} + static int mlxsw_pci_config_profile(struct mlxsw_pci *mlxsw_pci, char *mbox, const struct mlxsw_config_profile *profile) { @@ -1404,7 +1454,8 @@ static void mlxsw_pci_mbox_free(struct mlxsw_pci *mlxsw_pci, } static int mlxsw_pci_init(void *bus_priv, struct mlxsw_core *mlxsw_core, - const struct mlxsw_config_profile *profile) + const struct mlxsw_config_profile *profile, + struct mlxsw_resources *resources) { struct mlxsw_pci *mlxsw_pci = bus_priv; struct pci_dev *pdev = mlxsw_pci->pdev; @@ -1463,6 +1514,11 @@ static int mlxsw_pci_init(void *bus_priv, struct mlxsw_core *mlxsw_core, if (err) goto err_boardinfo; + err = mlxsw_pci_resources_query(mlxsw_pci, mbox, resources, + profile->resource_query_enable); + if (err) + goto err_query_resources; + err = mlxsw_pci_config_profile(mlxsw_pci, mbox, profile); if (err) goto err_config_profile; @@ -1485,6 +1541,7 @@ err_request_eq_irq: mlxsw_pci_aqs_fini(mlxsw_pci); err_aqs_init: err_config_profile: +err_query_resources: err_boardinfo: mlxsw_pci_fw_area_fini(mlxsw_pci); err_fw_area_init: diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 2ba8cc4..80172fc 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2488,6 +2488,7 @@ static struct mlxsw_config_profile mlxsw_sp_config_profile = { .type = MLXSW_PORT_SWID_TYPE_ETH, } }, + .resource_query_enable = 1, }; static struct mlxsw_driver mlxsw_sp_driver = { diff --git a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c index 25f658b..377daa4 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c +++ b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c @@ -1541,6 +1541,7 @@ static struct mlxsw_config_profile mlxsw_sx_config_profile = { .type = MLXSW_PORT_SWID_TYPE_ETH, } }, + .resource_query_enable = 0, }; static struct mlxsw_driver mlxsw_sx_driver = { -- cgit v0.10.2 From ded821c8d3793efe00195dd7ab633f6412dd8ae0 Mon Sep 17 00:00:00 2001 From: Nogah Frankel Date: Thu, 21 Jul 2016 12:03:10 +0200 Subject: mlxsw: pci: Add max span resources to resources query Add max span resources to resources query. Signed-off-by: Nogah Frankel Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h index d57ad0d..d3476ea 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.h +++ b/drivers/net/ethernet/mellanox/mlxsw/core.h @@ -268,6 +268,8 @@ struct mlxsw_driver { }; struct mlxsw_resources { + u8 max_span_valid:1; + u8 max_span; }; struct mlxsw_resources *mlxsw_core_resources_get(struct mlxsw_core *mlxsw_core); diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c index a724b66..1d1360c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci.c +++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c @@ -1155,6 +1155,7 @@ mlxsw_pci_config_profile_swid_config(struct mlxsw_pci *mlxsw_pci, } #define MLXSW_RESOURCES_TABLE_END_ID 0xffff +#define MLXSW_MAX_SPAN_ID 0x2420 #define MLXSW_RESOURCES_QUERY_MAX_QUERIES 100 #define MLXSW_RESOURCES_PER_QUERY 32 @@ -1162,6 +1163,10 @@ static void mlxsw_pci_resources_query_parse(int id, u64 val, struct mlxsw_resources *resources) { switch (id) { + case MLXSW_MAX_SPAN_ID: + resources->max_span = val; + resources->max_span_valid = 1; + break; default: break; } -- cgit v0.10.2 From bf3994d2ed310813da28362d87bfe9f0e1c3e37f Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 21 Jul 2016 12:03:11 +0200 Subject: net/sched: introduce Match-all classifier The matchall classifier matches every packet and allows the user to apply actions on it. This filter is very useful in usecases where every packet should be matched, for example, packet mirroring (SPAN) can be setup very easily using that filter. Signed-off-by: Jiri Pirko Signed-off-by: Yotam Gigi Signed-off-by: David S. Miller diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h index 5702e93..a324948 100644 --- a/include/uapi/linux/pkt_cls.h +++ b/include/uapi/linux/pkt_cls.h @@ -433,6 +433,17 @@ enum { #define TCA_FLOWER_MAX (__TCA_FLOWER_MAX - 1) +/* Match-all classifier */ + +enum { + TCA_MATCHALL_UNSPEC, + TCA_MATCHALL_CLASSID, + TCA_MATCHALL_ACT, + __TCA_MATCHALL_MAX, +}; + +#define TCA_MATCHALL_MAX (__TCA_MATCHALL_MAX - 1) + /* Extended Matches */ struct tcf_ematch_tree_hdr { diff --git a/net/sched/Kconfig b/net/sched/Kconfig index b148302..ccf931b 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -494,6 +494,16 @@ config NET_CLS_FLOWER To compile this code as a module, choose M here: the module will be called cls_flower. +config NET_CLS_MATCHALL + tristate "Match-all classifier" + select NET_CLS + ---help--- + If you say Y here, you will be able to classify packets based on + nothing. Every packet will match. + + To compile this code as a module, choose M here: the module will + be called cls_matchall. + config NET_EMATCH bool "Extended Matches" select NET_CLS diff --git a/net/sched/Makefile b/net/sched/Makefile index 84bddb3..ae088a5 100644 --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -60,6 +60,7 @@ obj-$(CONFIG_NET_CLS_FLOW) += cls_flow.o obj-$(CONFIG_NET_CLS_CGROUP) += cls_cgroup.o obj-$(CONFIG_NET_CLS_BPF) += cls_bpf.o obj-$(CONFIG_NET_CLS_FLOWER) += cls_flower.o +obj-$(CONFIG_NET_CLS_MATCHALL) += cls_matchall.o obj-$(CONFIG_NET_EMATCH) += ematch.o obj-$(CONFIG_NET_EMATCH_CMP) += em_cmp.o obj-$(CONFIG_NET_EMATCH_NBYTE) += em_nbyte.o diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c new file mode 100644 index 0000000..8a6b4de --- /dev/null +++ b/net/sched/cls_matchall.c @@ -0,0 +1,248 @@ +/* + * net/sched/cls_matchll.c Match-all classifier + * + * Copyright (c) 2016 Jiri Pirko + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include + +#include +#include + +struct cls_mall_filter { + struct tcf_exts exts; + struct tcf_result res; + u32 handle; + struct rcu_head rcu; +}; + +struct cls_mall_head { + struct cls_mall_filter *filter; + struct rcu_head rcu; +}; + +static int mall_classify(struct sk_buff *skb, const struct tcf_proto *tp, + struct tcf_result *res) +{ + struct cls_mall_head *head = rcu_dereference_bh(tp->root); + struct cls_mall_filter *f = head->filter; + + return tcf_exts_exec(skb, &f->exts, res); +} + +static int mall_init(struct tcf_proto *tp) +{ + struct cls_mall_head *head; + + head = kzalloc(sizeof(*head), GFP_KERNEL); + if (!head) + return -ENOBUFS; + + rcu_assign_pointer(tp->root, head); + + return 0; +} + +static void mall_destroy_filter(struct rcu_head *head) +{ + struct cls_mall_filter *f = container_of(head, struct cls_mall_filter, rcu); + + tcf_exts_destroy(&f->exts); + kfree(f); +} + +static bool mall_destroy(struct tcf_proto *tp, bool force) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + + if (!force && head->filter) + return false; + + if (head->filter) + call_rcu(&head->filter->rcu, mall_destroy_filter); + RCU_INIT_POINTER(tp->root, NULL); + kfree_rcu(head, rcu); + return true; +} + +static unsigned long mall_get(struct tcf_proto *tp, u32 handle) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + struct cls_mall_filter *f = head->filter; + + if (f && f->handle == handle) + return (unsigned long) f; + return 0; +} + +static const struct nla_policy mall_policy[TCA_MATCHALL_MAX + 1] = { + [TCA_MATCHALL_UNSPEC] = { .type = NLA_UNSPEC }, + [TCA_MATCHALL_CLASSID] = { .type = NLA_U32 }, +}; + +static int mall_set_parms(struct net *net, struct tcf_proto *tp, + struct cls_mall_filter *f, + unsigned long base, struct nlattr **tb, + struct nlattr *est, bool ovr) +{ + struct tcf_exts e; + int err; + + tcf_exts_init(&e, TCA_MATCHALL_ACT, 0); + err = tcf_exts_validate(net, tp, tb, est, &e, ovr); + if (err < 0) + return err; + + if (tb[TCA_MATCHALL_CLASSID]) { + f->res.classid = nla_get_u32(tb[TCA_MATCHALL_CLASSID]); + tcf_bind_filter(tp, &f->res, base); + } + + tcf_exts_change(tp, &f->exts, &e); + + return 0; +} + +static int mall_change(struct net *net, struct sk_buff *in_skb, + struct tcf_proto *tp, unsigned long base, + u32 handle, struct nlattr **tca, + unsigned long *arg, bool ovr) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + struct cls_mall_filter *fold = (struct cls_mall_filter *) *arg; + struct cls_mall_filter *f; + struct nlattr *tb[TCA_MATCHALL_MAX + 1]; + int err; + + if (!tca[TCA_OPTIONS]) + return -EINVAL; + + if (head->filter) + return -EBUSY; + + if (fold) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_MATCHALL_MAX, + tca[TCA_OPTIONS], mall_policy); + if (err < 0) + return err; + + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (!f) + return -ENOBUFS; + + tcf_exts_init(&f->exts, TCA_MATCHALL_ACT, 0); + + if (!handle) + handle = 1; + f->handle = handle; + + err = mall_set_parms(net, tp, f, base, tb, tca[TCA_RATE], ovr); + if (err) + goto errout; + + *arg = (unsigned long) f; + rcu_assign_pointer(head->filter, f); + + return 0; + +errout: + kfree(f); + return err; +} + +static int mall_delete(struct tcf_proto *tp, unsigned long arg) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + struct cls_mall_filter *f = (struct cls_mall_filter *) arg; + + RCU_INIT_POINTER(head->filter, NULL); + tcf_unbind_filter(tp, &f->res); + call_rcu(&f->rcu, mall_destroy_filter); + return 0; +} + +static void mall_walk(struct tcf_proto *tp, struct tcf_walker *arg) +{ + struct cls_mall_head *head = rtnl_dereference(tp->root); + struct cls_mall_filter *f = head->filter; + + if (arg->count < arg->skip) + goto skip; + if (arg->fn(tp, (unsigned long) f, arg) < 0) + arg->stop = 1; +skip: + arg->count++; +} + +static int mall_dump(struct net *net, struct tcf_proto *tp, unsigned long fh, + struct sk_buff *skb, struct tcmsg *t) +{ + struct cls_mall_filter *f = (struct cls_mall_filter *) fh; + struct nlattr *nest; + + if (!f) + return skb->len; + + t->tcm_handle = f->handle; + + nest = nla_nest_start(skb, TCA_OPTIONS); + if (!nest) + goto nla_put_failure; + + if (f->res.classid && + nla_put_u32(skb, TCA_MATCHALL_CLASSID, f->res.classid)) + goto nla_put_failure; + + if (tcf_exts_dump(skb, &f->exts)) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + if (tcf_exts_dump_stats(skb, &f->exts) < 0) + goto nla_put_failure; + + return skb->len; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + +static struct tcf_proto_ops cls_mall_ops __read_mostly = { + .kind = "matchall", + .classify = mall_classify, + .init = mall_init, + .destroy = mall_destroy, + .get = mall_get, + .change = mall_change, + .delete = mall_delete, + .walk = mall_walk, + .dump = mall_dump, + .owner = THIS_MODULE, +}; + +static int __init cls_mall_init(void) +{ + return register_tcf_proto_ops(&cls_mall_ops); +} + +static void __exit cls_mall_exit(void) +{ + unregister_tcf_proto_ops(&cls_mall_ops); +} + +module_init(cls_mall_init); +module_exit(cls_mall_exit); + +MODULE_AUTHOR("Jiri Pirko "); +MODULE_DESCRIPTION("Match-all classifier"); +MODULE_LICENSE("GPL v2"); -- cgit v0.10.2 From b87f7936a93246804cf70e7e2e0568799c948bb1 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Thu, 21 Jul 2016 12:03:12 +0200 Subject: net/sched: Add match-all classifier hw offloading. Following the work that have been done on offloading classifiers like u32 and flower, now the match-all classifier hw offloading is possible. if the interface supports tc offloading. To control the offloading, two tc flags have been introduced: skip_sw and skip_hw. Typical usage: tc filter add dev eth25 parent ffff: \ matchall skip_sw \ action mirred egress mirror \ dev eth27 Signed-off-by: Yotam Gigi Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 43c749b..076df53 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -787,6 +787,7 @@ enum { TC_SETUP_MQPRIO, TC_SETUP_CLSU32, TC_SETUP_CLSFLOWER, + TC_SETUP_MATCHALL, }; struct tc_cls_u32_offload; @@ -797,6 +798,7 @@ struct tc_to_netdev { u8 tc; struct tc_cls_u32_offload *cls_u32; struct tc_cls_flower_offload *cls_flower; + struct tc_cls_matchall_offload *cls_mall; }; }; diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 3722dda..6f8d653 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -442,4 +442,15 @@ struct tc_cls_flower_offload { struct tcf_exts *exts; }; +enum tc_matchall_command { + TC_CLSMATCHALL_REPLACE, + TC_CLSMATCHALL_DESTROY, +}; + +struct tc_cls_matchall_offload { + enum tc_matchall_command command; + struct tcf_exts *exts; + unsigned long cookie; +}; + #endif diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h index a324948..d1c1cca 100644 --- a/include/uapi/linux/pkt_cls.h +++ b/include/uapi/linux/pkt_cls.h @@ -439,6 +439,7 @@ enum { TCA_MATCHALL_UNSPEC, TCA_MATCHALL_CLASSID, TCA_MATCHALL_ACT, + TCA_MATCHALL_FLAGS, __TCA_MATCHALL_MAX, }; diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c index 8a6b4de..25927b6 100644 --- a/net/sched/cls_matchall.c +++ b/net/sched/cls_matchall.c @@ -21,6 +21,7 @@ struct cls_mall_filter { struct tcf_result res; u32 handle; struct rcu_head rcu; + u32 flags; }; struct cls_mall_head { @@ -34,6 +35,9 @@ static int mall_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct cls_mall_head *head = rcu_dereference_bh(tp->root); struct cls_mall_filter *f = head->filter; + if (tc_skip_sw(f->flags)) + return -1; + return tcf_exts_exec(skb, &f->exts, res); } @@ -55,18 +59,61 @@ static void mall_destroy_filter(struct rcu_head *head) struct cls_mall_filter *f = container_of(head, struct cls_mall_filter, rcu); tcf_exts_destroy(&f->exts); + kfree(f); } +static int mall_replace_hw_filter(struct tcf_proto *tp, + struct cls_mall_filter *f, + unsigned long cookie) +{ + struct net_device *dev = tp->q->dev_queue->dev; + struct tc_to_netdev offload; + struct tc_cls_matchall_offload mall_offload = {0}; + + offload.type = TC_SETUP_MATCHALL; + offload.cls_mall = &mall_offload; + offload.cls_mall->command = TC_CLSMATCHALL_REPLACE; + offload.cls_mall->exts = &f->exts; + offload.cls_mall->cookie = cookie; + + return dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, + &offload); +} + +static void mall_destroy_hw_filter(struct tcf_proto *tp, + struct cls_mall_filter *f, + unsigned long cookie) +{ + struct net_device *dev = tp->q->dev_queue->dev; + struct tc_to_netdev offload; + struct tc_cls_matchall_offload mall_offload = {0}; + + offload.type = TC_SETUP_MATCHALL; + offload.cls_mall = &mall_offload; + offload.cls_mall->command = TC_CLSMATCHALL_DESTROY; + offload.cls_mall->exts = NULL; + offload.cls_mall->cookie = cookie; + + dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, + &offload); +} + static bool mall_destroy(struct tcf_proto *tp, bool force) { struct cls_mall_head *head = rtnl_dereference(tp->root); + struct net_device *dev = tp->q->dev_queue->dev; + struct cls_mall_filter *f = head->filter; - if (!force && head->filter) + if (!force && f) return false; - if (head->filter) - call_rcu(&head->filter->rcu, mall_destroy_filter); + if (f) { + if (tc_should_offload(dev, tp, f->flags)) + mall_destroy_hw_filter(tp, f, (unsigned long) f); + + call_rcu(&f->rcu, mall_destroy_filter); + } RCU_INIT_POINTER(tp->root, NULL); kfree_rcu(head, rcu); return true; @@ -117,8 +164,10 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, { struct cls_mall_head *head = rtnl_dereference(tp->root); struct cls_mall_filter *fold = (struct cls_mall_filter *) *arg; + struct net_device *dev = tp->q->dev_queue->dev; struct cls_mall_filter *f; struct nlattr *tb[TCA_MATCHALL_MAX + 1]; + u32 flags = 0; int err; if (!tca[TCA_OPTIONS]) @@ -135,6 +184,12 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, if (err < 0) return err; + if (tb[TCA_MATCHALL_FLAGS]) { + flags = nla_get_u32(tb[TCA_MATCHALL_FLAGS]); + if (!tc_flags_valid(flags)) + return -EINVAL; + } + f = kzalloc(sizeof(*f), GFP_KERNEL); if (!f) return -ENOBUFS; @@ -144,11 +199,22 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, if (!handle) handle = 1; f->handle = handle; + f->flags = flags; err = mall_set_parms(net, tp, f, base, tb, tca[TCA_RATE], ovr); if (err) goto errout; + if (tc_should_offload(dev, tp, flags)) { + err = mall_replace_hw_filter(tp, f, (unsigned long) f); + if (err) { + if (tc_skip_sw(flags)) + goto errout; + else + err = 0; + } + } + *arg = (unsigned long) f; rcu_assign_pointer(head->filter, f); @@ -163,6 +229,10 @@ static int mall_delete(struct tcf_proto *tp, unsigned long arg) { struct cls_mall_head *head = rtnl_dereference(tp->root); struct cls_mall_filter *f = (struct cls_mall_filter *) arg; + struct net_device *dev = tp->q->dev_queue->dev; + + if (tc_should_offload(dev, tp, f->flags)) + mall_destroy_hw_filter(tp, f, (unsigned long) f); RCU_INIT_POINTER(head->filter, NULL); tcf_unbind_filter(tp, &f->res); -- cgit v0.10.2 From 51ae8cc66244f19f77bc251a102b1f414586f069 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Thu, 21 Jul 2016 12:03:13 +0200 Subject: mlxsw: reg: Add Shared Buffer Internal Buffer register The SBIB register configures per port buffer for internal use. This register is used to configure an egress mirror buffer on the egress port which does the mirroring. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 5b2a0b9..70bb970 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -5062,6 +5062,45 @@ static inline void mlxsw_reg_sbsr_rec_unpack(char *payload, int rec_index, mlxsw_reg_sbsr_rec_max_buff_occupancy_get(payload, rec_index); } +/* SBIB - Shared Buffer Internal Buffer Register + * --------------------------------------------- + * The SBIB register configures per port buffers for internal use. The internal + * buffers consume memory on the port buffers (note that the port buffers are + * used also by PBMC). + * + * For Spectrum this is used for egress mirroring. + */ +#define MLXSW_REG_SBIB_ID 0xB006 +#define MLXSW_REG_SBIB_LEN 0x10 + +static const struct mlxsw_reg_info mlxsw_reg_sbib = { + .id = MLXSW_REG_SBIB_ID, + .len = MLXSW_REG_SBIB_LEN, +}; + +/* reg_sbib_local_port + * Local port number + * Not supported for CPU port and router port + * Access: Index + */ +MLXSW_ITEM32(reg, sbib, local_port, 0x00, 16, 8); + +/* reg_sbib_buff_size + * Units represented in cells + * Allowed range is 0 to (cap_max_headroom_size - 1) + * Default is 0 + * Access: RW + */ +MLXSW_ITEM32(reg, sbib, buff_size, 0x08, 0, 24); + +static inline void mlxsw_reg_sbib_pack(char *payload, u8 local_port, + u32 buff_size) +{ + MLXSW_REG_ZERO(sbib, payload); + mlxsw_reg_sbib_local_port_set(payload, local_port); + mlxsw_reg_sbib_buff_size_set(payload, buff_size); +} + static inline const char *mlxsw_reg_id_str(u16 reg_id) { switch (reg_id) { @@ -5179,6 +5218,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "SBMM"; case MLXSW_REG_SBSR_ID: return "SBSR"; + case MLXSW_REG_SBIB_ID: + return "SBIB"; default: return "*UNKNOWN*"; } -- cgit v0.10.2 From 43a468562030e428728be14dac583ae19046b22e Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Thu, 21 Jul 2016 12:03:14 +0200 Subject: mlxsw: reg: Add Monitoring Port Analyzer Table register The MPAT register is used to query and configure the Switch Port Analyzer (SPAN) table. This register is used to configure a port as a mirror output port, while after that a mirrored input port can be bound using MPAR register. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 70bb970..d0b97a6 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -4633,6 +4633,58 @@ static inline void mlxsw_reg_mtmp_unpack(char *payload, unsigned int *p_temp, mlxsw_reg_mtmp_sensor_name_memcpy_from(payload, sensor_name); } +/* MPAT - Monitoring Port Analyzer Table + * ------------------------------------- + * MPAT Register is used to query and configure the Switch PortAnalyzer Table. + * For an enabled analyzer, all fields except e (enable) cannot be modified. + */ +#define MLXSW_REG_MPAT_ID 0x901A +#define MLXSW_REG_MPAT_LEN 0x78 + +static const struct mlxsw_reg_info mlxsw_reg_mpat = { + .id = MLXSW_REG_MPAT_ID, + .len = MLXSW_REG_MPAT_LEN, +}; + +/* reg_mpat_pa_id + * Port Analyzer ID. + * Access: Index + */ +MLXSW_ITEM32(reg, mpat, pa_id, 0x00, 28, 4); + +/* reg_mpat_system_port + * A unique port identifier for the final destination of the packet. + * Access: RW + */ +MLXSW_ITEM32(reg, mpat, system_port, 0x00, 0, 16); + +/* reg_mpat_e + * Enable. Indicating the Port Analyzer is enabled. + * Access: RW + */ +MLXSW_ITEM32(reg, mpat, e, 0x04, 31, 1); + +/* reg_mpat_qos + * Quality Of Service Mode. + * 0: CONFIGURED - QoS parameters (Switch Priority, and encapsulation + * PCP, DEI, DSCP or VL) are configured. + * 1: MAINTAIN - QoS parameters (Switch Priority, Color) are the + * same as in the original packet that has triggered the mirroring. For + * SPAN also the pcp,dei are maintained. + * Access: RW + */ +MLXSW_ITEM32(reg, mpat, qos, 0x04, 26, 1); + +static inline void mlxsw_reg_mpat_pack(char *payload, u8 pa_id, + u16 system_port, bool e) +{ + MLXSW_REG_ZERO(mpat, payload); + mlxsw_reg_mpat_pa_id_set(payload, pa_id); + mlxsw_reg_mpat_system_port_set(payload, system_port); + mlxsw_reg_mpat_e_set(payload, e); + mlxsw_reg_mpat_qos_set(payload, 1); +} + /* MLCR - Management LED Control Register * -------------------------------------- * Controls the system LEDs. @@ -5204,6 +5256,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "MFSM"; case MLXSW_REG_MTCAP_ID: return "MTCAP"; + case MLXSW_REG_MPAT_ID: + return "MPAT"; case MLXSW_REG_MTMP_ID: return "MTMP"; case MLXSW_REG_MLCR_ID: -- cgit v0.10.2 From 230190548b7529d02d38f450748baf0721332f26 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Thu, 21 Jul 2016 12:03:15 +0200 Subject: mlxsw: reg: Add the Monitoring Port Analyzer register The MPAR register is used to bind ports to a SPAN entry (which was created using MPAT register) and thus mirror their traffic (ingress / egress) to a different port. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index d0b97a6..7ca9201 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -4675,6 +4675,16 @@ MLXSW_ITEM32(reg, mpat, e, 0x04, 31, 1); */ MLXSW_ITEM32(reg, mpat, qos, 0x04, 26, 1); +/* reg_mpat_be + * Best effort mode. Indicates mirroring traffic should not cause packet + * drop or back pressure, but will discard the mirrored packets. Mirrored + * packets will be forwarded on a best effort manner. + * 0: Do not discard mirrored packets + * 1: Discard mirrored packets if causing congestion + * Access: RW + */ +MLXSW_ITEM32(reg, mpat, be, 0x04, 25, 1); + static inline void mlxsw_reg_mpat_pack(char *payload, u8 pa_id, u16 system_port, bool e) { @@ -4683,6 +4693,61 @@ static inline void mlxsw_reg_mpat_pack(char *payload, u8 pa_id, mlxsw_reg_mpat_system_port_set(payload, system_port); mlxsw_reg_mpat_e_set(payload, e); mlxsw_reg_mpat_qos_set(payload, 1); + mlxsw_reg_mpat_be_set(payload, 1); +} + +/* MPAR - Monitoring Port Analyzer Register + * ---------------------------------------- + * MPAR register is used to query and configure the port analyzer port mirroring + * properties. + */ +#define MLXSW_REG_MPAR_ID 0x901B +#define MLXSW_REG_MPAR_LEN 0x08 + +static const struct mlxsw_reg_info mlxsw_reg_mpar = { + .id = MLXSW_REG_MPAR_ID, + .len = MLXSW_REG_MPAR_LEN, +}; + +/* reg_mpar_local_port + * The local port to mirror the packets from. + * Access: Index + */ +MLXSW_ITEM32(reg, mpar, local_port, 0x00, 16, 8); + +enum mlxsw_reg_mpar_i_e { + MLXSW_REG_MPAR_TYPE_EGRESS, + MLXSW_REG_MPAR_TYPE_INGRESS, +}; + +/* reg_mpar_i_e + * Ingress/Egress + * Access: Index + */ +MLXSW_ITEM32(reg, mpar, i_e, 0x00, 0, 4); + +/* reg_mpar_enable + * Enable mirroring + * By default, port mirroring is disabled for all ports. + * Access: RW + */ +MLXSW_ITEM32(reg, mpar, enable, 0x04, 31, 1); + +/* reg_mpar_pa_id + * Port Analyzer ID. + * Access: RW + */ +MLXSW_ITEM32(reg, mpar, pa_id, 0x04, 0, 4); + +static inline void mlxsw_reg_mpar_pack(char *payload, u8 local_port, + enum mlxsw_reg_mpar_i_e i_e, + bool enable, u8 pa_id) +{ + MLXSW_REG_ZERO(mpar, payload); + mlxsw_reg_mpar_local_port_set(payload, local_port); + mlxsw_reg_mpar_enable_set(payload, enable); + mlxsw_reg_mpar_i_e_set(payload, i_e); + mlxsw_reg_mpar_pa_id_set(payload, pa_id); } /* MLCR - Management LED Control Register @@ -5258,6 +5323,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "MTCAP"; case MLXSW_REG_MPAT_ID: return "MPAT"; + case MLXSW_REG_MPAR_ID: + return "MPAR"; case MLXSW_REG_MTMP_ID: return "MTMP"; case MLXSW_REG_MLCR_ID: -- cgit v0.10.2 From 56a20680f70393d199fc5e8ecc7859549ca5a0c0 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Thu, 21 Jul 2016 12:03:16 +0200 Subject: net/sched: act_mirred: Add helper inlines to access tcf_mirred info. The helper function is_tcf_mirred_mirror helps finding whether an action struct is of type mirred and is configured to be of type mirror. Signed-off-by: Yotam Gigi Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/include/net/tc_act/tc_mirred.h b/include/net/tc_act/tc_mirred.h index e891835..6a13a7c 100644 --- a/include/net/tc_act/tc_mirred.h +++ b/include/net/tc_act/tc_mirred.h @@ -24,6 +24,15 @@ static inline bool is_tcf_mirred_redirect(const struct tc_action *a) return false; } +static inline bool is_tcf_mirred_mirror(const struct tc_action *a) +{ +#ifdef CONFIG_NET_CLS_ACT + if (a->ops && a->ops->type == TCA_ACT_MIRRED) + return to_mirred(a)->tcfm_eaction == TCA_EGRESS_MIRROR; +#endif + return false; +} + static inline int tcf_mirred_ifindex(const struct tc_action *a) { return to_mirred(a)->tcfm_ifindex; -- cgit v0.10.2 From 763b4b70afcd3289fedfba6fb681e192a2291a95 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Thu, 21 Jul 2016 12:03:17 +0200 Subject: mlxsw: spectrum: Add support in matchall mirror TC offloading This patch offloads port mirroring directives to hw using the matchall TC with action mirror. It includes both the implementation of the ndo_setup_tc function for the spectrum driver and the spectrum hardware offload configuration code. The hardware offload code is basically two new functions which are capable of adding and removing a new mirror ports pair. It is done using the MPAT, MPAR and SBIB registers: - A new Switch-Port Analyzer (SPAN) entry is added using MPAT to the 'to' port. - The 'to' port is bound to the SPAN entry using MPAR register. - In case of egress SPAN, the 'to' port gets a new internal shared buffer using SBIB register. In addition, a new database was added to the mlxsw_sp struct to store all the SPAN entries and their bound ports list. The number of supported SPAN entries is determined by resource query. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 80172fc..552636b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -54,6 +54,8 @@ #include #include #include +#include +#include #include "spectrum.h" #include "core.h" @@ -133,6 +135,8 @@ MLXSW_ITEM32(tx, hdr, fid, 0x08, 0, 16); */ MLXSW_ITEM32(tx, hdr, type, 0x0C, 0, 4); +static bool mlxsw_sp_port_dev_check(const struct net_device *dev); + static void mlxsw_sp_txhdr_construct(struct sk_buff *skb, const struct mlxsw_tx_info *tx_info) { @@ -161,6 +165,303 @@ static int mlxsw_sp_base_mac_get(struct mlxsw_sp *mlxsw_sp) return 0; } +static int mlxsw_sp_span_init(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_resources *resources; + int i; + + resources = mlxsw_core_resources_get(mlxsw_sp->core); + if (!resources->max_span_valid) + return -EIO; + + mlxsw_sp->span.entries_count = resources->max_span; + mlxsw_sp->span.entries = kcalloc(mlxsw_sp->span.entries_count, + sizeof(struct mlxsw_sp_span_entry), + GFP_KERNEL); + if (!mlxsw_sp->span.entries) + return -ENOMEM; + + for (i = 0; i < mlxsw_sp->span.entries_count; i++) + INIT_LIST_HEAD(&mlxsw_sp->span.entries[i].bound_ports_list); + + return 0; +} + +static void mlxsw_sp_span_fini(struct mlxsw_sp *mlxsw_sp) +{ + int i; + + for (i = 0; i < mlxsw_sp->span.entries_count; i++) { + struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span.entries[i]; + + WARN_ON_ONCE(!list_empty(&curr->bound_ports_list)); + } + kfree(mlxsw_sp->span.entries); +} + +static struct mlxsw_sp_span_entry * +mlxsw_sp_span_entry_create(struct mlxsw_sp_port *port) +{ + struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; + struct mlxsw_sp_span_entry *span_entry; + char mpat_pl[MLXSW_REG_MPAT_LEN]; + u8 local_port = port->local_port; + int index; + int i; + int err; + + /* find a free entry to use */ + index = -1; + for (i = 0; i < mlxsw_sp->span.entries_count; i++) { + if (!mlxsw_sp->span.entries[i].used) { + index = i; + span_entry = &mlxsw_sp->span.entries[i]; + break; + } + } + if (index < 0) + return NULL; + + /* create a new port analayzer entry for local_port */ + mlxsw_reg_mpat_pack(mpat_pl, index, local_port, true); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mpat), mpat_pl); + if (err) + return NULL; + + span_entry->used = true; + span_entry->id = index; + span_entry->ref_count = 0; + span_entry->local_port = local_port; + return span_entry; +} + +static void mlxsw_sp_span_entry_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_span_entry *span_entry) +{ + u8 local_port = span_entry->local_port; + char mpat_pl[MLXSW_REG_MPAT_LEN]; + int pa_id = span_entry->id; + + mlxsw_reg_mpat_pack(mpat_pl, pa_id, local_port, false); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mpat), mpat_pl); + span_entry->used = false; +} + +struct mlxsw_sp_span_entry *mlxsw_sp_span_entry_find(struct mlxsw_sp_port *port) +{ + struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; + int i; + + for (i = 0; i < mlxsw_sp->span.entries_count; i++) { + struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span.entries[i]; + + if (curr->used && curr->local_port == port->local_port) + return curr; + } + return NULL; +} + +struct mlxsw_sp_span_entry *mlxsw_sp_span_entry_get(struct mlxsw_sp_port *port) +{ + struct mlxsw_sp_span_entry *span_entry; + + span_entry = mlxsw_sp_span_entry_find(port); + if (span_entry) { + span_entry->ref_count++; + return span_entry; + } + + return mlxsw_sp_span_entry_create(port); +} + +static int mlxsw_sp_span_entry_put(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_span_entry *span_entry) +{ + if (--span_entry->ref_count == 0) + mlxsw_sp_span_entry_destroy(mlxsw_sp, span_entry); + return 0; +} + +static bool mlxsw_sp_span_is_egress_mirror(struct mlxsw_sp_port *port) +{ + struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; + struct mlxsw_sp_span_inspected_port *p; + int i; + + for (i = 0; i < mlxsw_sp->span.entries_count; i++) { + struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span.entries[i]; + + list_for_each_entry(p, &curr->bound_ports_list, list) + if (p->local_port == port->local_port && + p->type == MLXSW_SP_SPAN_EGRESS) + return true; + } + + return false; +} + +static int mlxsw_sp_span_mtu_to_buffsize(int mtu) +{ + return MLXSW_SP_BYTES_TO_CELLS(mtu * 5 / 2) + 1; +} + +static int mlxsw_sp_span_port_mtu_update(struct mlxsw_sp_port *port, u16 mtu) +{ + struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; + char sbib_pl[MLXSW_REG_SBIB_LEN]; + int err; + + /* If port is egress mirrored, the shared buffer size should be + * updated according to the mtu value + */ + if (mlxsw_sp_span_is_egress_mirror(port)) { + mlxsw_reg_sbib_pack(sbib_pl, port->local_port, + mlxsw_sp_span_mtu_to_buffsize(mtu)); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); + if (err) { + netdev_err(port->dev, "Could not update shared buffer for mirroring\n"); + return err; + } + } + + return 0; +} + +static struct mlxsw_sp_span_inspected_port * +mlxsw_sp_span_entry_bound_port_find(struct mlxsw_sp_port *port, + struct mlxsw_sp_span_entry *span_entry) +{ + struct mlxsw_sp_span_inspected_port *p; + + list_for_each_entry(p, &span_entry->bound_ports_list, list) + if (port->local_port == p->local_port) + return p; + return NULL; +} + +static int +mlxsw_sp_span_inspected_port_bind(struct mlxsw_sp_port *port, + struct mlxsw_sp_span_entry *span_entry, + enum mlxsw_sp_span_type type) +{ + struct mlxsw_sp_span_inspected_port *inspected_port; + struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; + char mpar_pl[MLXSW_REG_MPAR_LEN]; + char sbib_pl[MLXSW_REG_SBIB_LEN]; + int pa_id = span_entry->id; + int err; + + /* if it is an egress SPAN, bind a shared buffer to it */ + if (type == MLXSW_SP_SPAN_EGRESS) { + mlxsw_reg_sbib_pack(sbib_pl, port->local_port, + mlxsw_sp_span_mtu_to_buffsize(port->dev->mtu)); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); + if (err) { + netdev_err(port->dev, "Could not create shared buffer for mirroring\n"); + return err; + } + } + + /* bind the port to the SPAN entry */ + mlxsw_reg_mpar_pack(mpar_pl, port->local_port, type, true, pa_id); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mpar), mpar_pl); + if (err) + goto err_mpar_reg_write; + + inspected_port = kzalloc(sizeof(*inspected_port), GFP_KERNEL); + if (!inspected_port) { + err = -ENOMEM; + goto err_inspected_port_alloc; + } + inspected_port->local_port = port->local_port; + inspected_port->type = type; + list_add_tail(&inspected_port->list, &span_entry->bound_ports_list); + + return 0; + +err_mpar_reg_write: +err_inspected_port_alloc: + if (type == MLXSW_SP_SPAN_EGRESS) { + mlxsw_reg_sbib_pack(sbib_pl, port->local_port, 0); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); + } + return err; +} + +static void +mlxsw_sp_span_inspected_port_unbind(struct mlxsw_sp_port *port, + struct mlxsw_sp_span_entry *span_entry, + enum mlxsw_sp_span_type type) +{ + struct mlxsw_sp_span_inspected_port *inspected_port; + struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; + char mpar_pl[MLXSW_REG_MPAR_LEN]; + char sbib_pl[MLXSW_REG_SBIB_LEN]; + int pa_id = span_entry->id; + + inspected_port = mlxsw_sp_span_entry_bound_port_find(port, span_entry); + if (!inspected_port) + return; + + /* remove the inspected port */ + mlxsw_reg_mpar_pack(mpar_pl, port->local_port, type, false, pa_id); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mpar), mpar_pl); + + /* remove the SBIB buffer if it was egress SPAN */ + if (type == MLXSW_SP_SPAN_EGRESS) { + mlxsw_reg_sbib_pack(sbib_pl, port->local_port, 0); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); + } + + mlxsw_sp_span_entry_put(mlxsw_sp, span_entry); + + list_del(&inspected_port->list); + kfree(inspected_port); +} + +static int mlxsw_sp_span_mirror_add(struct mlxsw_sp_port *from, + struct mlxsw_sp_port *to, + enum mlxsw_sp_span_type type) +{ + struct mlxsw_sp *mlxsw_sp = from->mlxsw_sp; + struct mlxsw_sp_span_entry *span_entry; + int err; + + span_entry = mlxsw_sp_span_entry_get(to); + if (!span_entry) + return -ENOENT; + + netdev_dbg(from->dev, "Adding inspected port to SPAN entry %d\n", + span_entry->id); + + err = mlxsw_sp_span_inspected_port_bind(from, span_entry, type); + if (err) + goto err_port_bind; + + return 0; + +err_port_bind: + mlxsw_sp_span_entry_put(mlxsw_sp, span_entry); + return err; +} + +static void mlxsw_sp_span_mirror_remove(struct mlxsw_sp_port *from, + struct mlxsw_sp_port *to, + enum mlxsw_sp_span_type type) +{ + struct mlxsw_sp_span_entry *span_entry; + + span_entry = mlxsw_sp_span_entry_find(to); + if (!span_entry) { + netdev_err(from->dev, "no span entry found\n"); + return; + } + + netdev_dbg(from->dev, "removing inspected port from SPAN entry %d\n", + span_entry->id); + mlxsw_sp_span_inspected_port_unbind(from, span_entry, type); +} + static int mlxsw_sp_port_admin_status_set(struct mlxsw_sp_port *mlxsw_sp_port, bool is_up) { @@ -493,6 +794,9 @@ static int mlxsw_sp_port_change_mtu(struct net_device *dev, int mtu) err = mlxsw_sp_port_headroom_set(mlxsw_sp_port, mtu, pause_en); if (err) return err; + err = mlxsw_sp_span_port_mtu_update(mlxsw_sp_port, mtu); + if (err) + goto err_span_port_mtu_update; err = mlxsw_sp_port_mtu_set(mlxsw_sp_port, mtu); if (err) goto err_port_mtu_set; @@ -500,6 +804,8 @@ static int mlxsw_sp_port_change_mtu(struct net_device *dev, int mtu) return 0; err_port_mtu_set: + mlxsw_sp_span_port_mtu_update(mlxsw_sp_port, dev->mtu); +err_span_port_mtu_update: mlxsw_sp_port_headroom_set(mlxsw_sp_port, dev->mtu, pause_en); return err; } @@ -776,10 +1082,155 @@ static int mlxsw_sp_port_get_phys_port_name(struct net_device *dev, char *name, return 0; } +static struct mlxsw_sp_port_mall_tc_entry * +mlxsw_sp_port_mirror_entry_find(struct mlxsw_sp_port *port, + unsigned long cookie) { + struct mlxsw_sp_port_mall_tc_entry *mall_tc_entry; + + list_for_each_entry(mall_tc_entry, &port->mall_tc_list, list) + if (mall_tc_entry->cookie == cookie) + return mall_tc_entry; + + return NULL; +} + +static int +mlxsw_sp_port_add_cls_matchall_mirror(struct mlxsw_sp_port *mlxsw_sp_port, + struct tc_cls_matchall_offload *cls, + const struct tc_action *a, + bool ingress) +{ + struct mlxsw_sp_port_mall_tc_entry *mall_tc_entry; + struct net *net = dev_net(mlxsw_sp_port->dev); + enum mlxsw_sp_span_type span_type; + struct mlxsw_sp_port *to_port; + struct net_device *to_dev; + int ifindex; + int err; + + ifindex = tcf_mirred_ifindex(a); + to_dev = __dev_get_by_index(net, ifindex); + if (!to_dev) { + netdev_err(mlxsw_sp_port->dev, "Could not find requested device\n"); + return -EINVAL; + } + + if (!mlxsw_sp_port_dev_check(to_dev)) { + netdev_err(mlxsw_sp_port->dev, "Cannot mirror to a non-spectrum port"); + return -ENOTSUPP; + } + to_port = netdev_priv(to_dev); + + mall_tc_entry = kzalloc(sizeof(*mall_tc_entry), GFP_KERNEL); + if (!mall_tc_entry) + return -ENOMEM; + + mall_tc_entry->cookie = cls->cookie; + mall_tc_entry->type = MLXSW_SP_PORT_MALL_MIRROR; + mall_tc_entry->mirror.to_local_port = to_port->local_port; + mall_tc_entry->mirror.ingress = ingress; + list_add_tail(&mall_tc_entry->list, &mlxsw_sp_port->mall_tc_list); + + span_type = ingress ? MLXSW_SP_SPAN_INGRESS : MLXSW_SP_SPAN_EGRESS; + err = mlxsw_sp_span_mirror_add(mlxsw_sp_port, to_port, span_type); + if (err) + goto err_mirror_add; + return 0; + +err_mirror_add: + list_del(&mall_tc_entry->list); + kfree(mall_tc_entry); + return err; +} + +static int mlxsw_sp_port_add_cls_matchall(struct mlxsw_sp_port *mlxsw_sp_port, + __be16 protocol, + struct tc_cls_matchall_offload *cls, + bool ingress) +{ + struct tcf_exts *exts = cls->exts; + const struct tc_action *a; + int err; + + if (!list_is_singular(&exts->actions)) { + netdev_err(mlxsw_sp_port->dev, "only singular actions are supported\n"); + return -ENOTSUPP; + } + + a = list_first_entry(&exts->actions, struct tc_action, list); + if (is_tcf_mirred_mirror(a) && protocol == htons(ETH_P_ALL)) { + err = mlxsw_sp_port_add_cls_matchall_mirror(mlxsw_sp_port, cls, + a, ingress); + if (err) + return err; + } else { + return -ENOTSUPP; + } + + return 0; +} + +static void mlxsw_sp_port_del_cls_matchall(struct mlxsw_sp_port *mlxsw_sp_port, + struct tc_cls_matchall_offload *cls) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_port_mall_tc_entry *mall_tc_entry; + enum mlxsw_sp_span_type span_type; + struct mlxsw_sp_port *to_port; + + mall_tc_entry = mlxsw_sp_port_mirror_entry_find(mlxsw_sp_port, + cls->cookie); + if (!mall_tc_entry) { + netdev_dbg(mlxsw_sp_port->dev, "tc entry not found on port\n"); + return; + } + + switch (mall_tc_entry->type) { + case MLXSW_SP_PORT_MALL_MIRROR: + to_port = mlxsw_sp->ports[mall_tc_entry->mirror.to_local_port]; + span_type = mall_tc_entry->mirror.ingress ? + MLXSW_SP_SPAN_INGRESS : MLXSW_SP_SPAN_EGRESS; + + mlxsw_sp_span_mirror_remove(mlxsw_sp_port, to_port, span_type); + break; + default: + WARN_ON(1); + } + + list_del(&mall_tc_entry->list); + kfree(mall_tc_entry); +} + +static int mlxsw_sp_setup_tc(struct net_device *dev, u32 handle, + __be16 proto, struct tc_to_netdev *tc) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + bool ingress = TC_H_MAJ(handle) == TC_H_MAJ(TC_H_INGRESS); + + if (tc->type == TC_SETUP_MATCHALL) { + switch (tc->cls_mall->command) { + case TC_CLSMATCHALL_REPLACE: + return mlxsw_sp_port_add_cls_matchall(mlxsw_sp_port, + proto, + tc->cls_mall, + ingress); + case TC_CLSMATCHALL_DESTROY: + mlxsw_sp_port_del_cls_matchall(mlxsw_sp_port, + tc->cls_mall); + return 0; + default: + return -EINVAL; + } + } + + return -ENOTSUPP; +} + static const struct net_device_ops mlxsw_sp_port_netdev_ops = { .ndo_open = mlxsw_sp_port_open, .ndo_stop = mlxsw_sp_port_stop, .ndo_start_xmit = mlxsw_sp_port_xmit, + .ndo_setup_tc = mlxsw_sp_setup_tc, .ndo_set_rx_mode = mlxsw_sp_set_rx_mode, .ndo_set_mac_address = mlxsw_sp_port_set_mac_address, .ndo_change_mtu = mlxsw_sp_port_change_mtu, @@ -1657,6 +2108,7 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, goto err_port_untagged_vlans_alloc; } INIT_LIST_HEAD(&mlxsw_sp_port->vports_list); + INIT_LIST_HEAD(&mlxsw_sp_port->mall_tc_list); mlxsw_sp_port->pcpu_stats = netdev_alloc_pcpu_stats(struct mlxsw_sp_port_pcpu_stats); @@ -1678,7 +2130,8 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, netif_carrier_off(dev); dev->features |= NETIF_F_NETNS_LOCAL | NETIF_F_LLTX | NETIF_F_SG | - NETIF_F_HW_VLAN_CTAG_FILTER; + NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_TC; + dev->hw_features |= NETIF_F_HW_TC; /* Each packet needs to have a Tx header (metadata) on top all other * headers. @@ -2410,6 +2863,12 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, goto err_router_init; } + err = mlxsw_sp_span_init(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to init span system\n"); + goto err_span_init; + } + err = mlxsw_sp_ports_create(mlxsw_sp); if (err) { dev_err(mlxsw_sp->bus_info->dev, "Failed to create ports\n"); @@ -2419,6 +2878,8 @@ static int mlxsw_sp_init(struct mlxsw_core *mlxsw_core, return 0; err_ports_create: + mlxsw_sp_span_fini(mlxsw_sp); +err_span_init: mlxsw_sp_router_fini(mlxsw_sp); err_router_init: mlxsw_sp_switchdev_fini(mlxsw_sp); @@ -2439,6 +2900,7 @@ static void mlxsw_sp_fini(struct mlxsw_core *mlxsw_core) int i; mlxsw_sp_ports_remove(mlxsw_sp); + mlxsw_sp_span_fini(mlxsw_sp); mlxsw_sp_router_fini(mlxsw_sp); mlxsw_sp_switchdev_fini(mlxsw_sp); mlxsw_sp_buffers_fini(mlxsw_sp); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index ef4ac89..f69aa37 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -214,6 +214,43 @@ struct mlxsw_sp_vr { struct mlxsw_sp_fib *fib; }; +enum mlxsw_sp_span_type { + MLXSW_SP_SPAN_EGRESS, + MLXSW_SP_SPAN_INGRESS +}; + +struct mlxsw_sp_span_inspected_port { + struct list_head list; + enum mlxsw_sp_span_type type; + u8 local_port; +}; + +struct mlxsw_sp_span_entry { + u8 local_port; + bool used; + struct list_head bound_ports_list; + int ref_count; + int id; +}; + +enum mlxsw_sp_port_mall_action_type { + MLXSW_SP_PORT_MALL_MIRROR, +}; + +struct mlxsw_sp_port_mall_mirror_tc_entry { + u8 to_local_port; + bool ingress; +}; + +struct mlxsw_sp_port_mall_tc_entry { + struct list_head list; + unsigned long cookie; + enum mlxsw_sp_port_mall_action_type type; + union { + struct mlxsw_sp_port_mall_mirror_tc_entry mirror; + }; +}; + struct mlxsw_sp_router { struct mlxsw_sp_lpm_tree lpm_trees[MLXSW_SP_LPM_TREE_COUNT]; struct mlxsw_sp_vr vrs[MLXSW_SP_VIRTUAL_ROUTER_MAX]; @@ -260,6 +297,11 @@ struct mlxsw_sp { struct { DECLARE_BITMAP(usage, MLXSW_SP_KVD_LINEAR_SIZE); } kvdl; + + struct { + struct mlxsw_sp_span_entry *entries; + int entries_count; + } span; }; static inline struct mlxsw_sp_upper * @@ -316,6 +358,8 @@ struct mlxsw_sp_port { unsigned long *untagged_vlans; /* VLAN interfaces */ struct list_head vports_list; + /* TC handles */ + struct list_head mall_tc_list; }; struct mlxsw_sp_port *mlxsw_sp_port_lower_dev_hold(struct net_device *dev); -- cgit v0.10.2 From 1533e77315220dc1d5ec3bd6d9fe32e2aa0a74c0 Mon Sep 17 00:00:00 2001 From: Mark Bloch Date: Thu, 21 Jul 2016 11:52:55 +0300 Subject: net/bonding: Enforce active-backup policy for IPoIB bonds When using an IPoIB bond currently only active-backup mode is a valid use case and this commit strengthens it. Since commit 2ab82852a270 ("net/bonding: Enable bonding to enslave netdevices not supporting set_mac_address()") was introduced till 4.7-rc1, IPoIB didn't support the set_mac_address ndo, and hence the fail over mac policy always applied to IPoIB bonds. With the introduction of commit 492a7e67ff83 ("IB/IPoIB: Allow setting the device address"), that doesn't hold and practically IPoIB bonds are broken as of that. To fix it, lets go to fail over mac if the device doesn't support the ndo OR this is IPoIB device. As a by-product, this commit also prevents a stack corruption which occurred when trying to copy 20 bytes (IPoIB) device address to a sockaddr struct that has only 16 bytes of storage. Signed-off-by: Mark Bloch Signed-off-by: Or Gerlitz Signed-off-by: Saeed Mahameed Acked-by: Andy Gospodarek Signed-off-by: Jay Vosburgh Signed-off-by: David S. Miller diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index b571ed9..1f276fa 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1422,7 +1422,16 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) return -EINVAL; } - if (slave_ops->ndo_set_mac_address == NULL) { + if (slave_dev->type == ARPHRD_INFINIBAND && + BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP) { + netdev_warn(bond_dev, "Type (%d) supports only active-backup mode\n", + slave_dev->type); + res = -EOPNOTSUPP; + goto err_undo_flags; + } + + if (!slave_ops->ndo_set_mac_address || + slave_dev->type == ARPHRD_INFINIBAND) { netdev_warn(bond_dev, "The slave device specified does not support setting the MAC address\n"); if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP && bond->params.fail_over_mac != BOND_FOM_ACTIVE) { -- cgit v0.10.2 From eb97ad99f9edb827e34e32b117284265ef2bfc33 Mon Sep 17 00:00:00 2001 From: Ganesh Goudar Date: Thu, 21 Jul 2016 20:19:18 +0530 Subject: cxgb4/cxgb4vf: Add link mode mask API to cxgb4 and cxgb4vf Based on original work by Casey Leedom Signed-off-by: Ganesh Goudar Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h index b4fceb9..2e2aa9f 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h @@ -418,6 +418,7 @@ struct trace_params { struct link_config { unsigned short supported; /* link capabilities */ unsigned short advertising; /* advertised capabilities */ + unsigned short lp_advertising; /* peer advertised capabilities */ unsigned short requested_speed; /* speed user has requested */ unsigned short speed; /* actual link speed */ unsigned char requested_fc; /* flow control user has requested */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c index 7a0b92b..02f80fe 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c @@ -480,178 +480,293 @@ static int identify_port(struct net_device *dev, return t4_identify_port(adap, adap->pf, netdev2pinfo(dev)->viid, val); } -static unsigned int from_fw_linkcaps(enum fw_port_type type, unsigned int caps) +/** + * from_fw_port_mod_type - translate Firmware Port/Module type to Ethtool + * @port_type: Firmware Port Type + * @mod_type: Firmware Module Type + * + * Translate Firmware Port/Module type to Ethtool Port Type. + */ +static int from_fw_port_mod_type(enum fw_port_type port_type, + enum fw_port_module_type mod_type) { - unsigned int v = 0; - - if (type == FW_PORT_TYPE_BT_SGMII || type == FW_PORT_TYPE_BT_XFI || - type == FW_PORT_TYPE_BT_XAUI) { - v |= SUPPORTED_TP; - if (caps & FW_PORT_CAP_SPEED_100M) - v |= SUPPORTED_100baseT_Full; - if (caps & FW_PORT_CAP_SPEED_1G) - v |= SUPPORTED_1000baseT_Full; - if (caps & FW_PORT_CAP_SPEED_10G) - v |= SUPPORTED_10000baseT_Full; - } else if (type == FW_PORT_TYPE_KX4 || type == FW_PORT_TYPE_KX) { - v |= SUPPORTED_Backplane; - if (caps & FW_PORT_CAP_SPEED_1G) - v |= SUPPORTED_1000baseKX_Full; - if (caps & FW_PORT_CAP_SPEED_10G) - v |= SUPPORTED_10000baseKX4_Full; - } else if (type == FW_PORT_TYPE_KR) { - v |= SUPPORTED_Backplane | SUPPORTED_10000baseKR_Full; - } else if (type == FW_PORT_TYPE_BP_AP) { - v |= SUPPORTED_Backplane | SUPPORTED_10000baseR_FEC | - SUPPORTED_10000baseKR_Full | SUPPORTED_1000baseKX_Full; - } else if (type == FW_PORT_TYPE_BP4_AP) { - v |= SUPPORTED_Backplane | SUPPORTED_10000baseR_FEC | - SUPPORTED_10000baseKR_Full | SUPPORTED_1000baseKX_Full | - SUPPORTED_10000baseKX4_Full; - } else if (type == FW_PORT_TYPE_FIBER_XFI || - type == FW_PORT_TYPE_FIBER_XAUI || - type == FW_PORT_TYPE_SFP || - type == FW_PORT_TYPE_QSFP_10G || - type == FW_PORT_TYPE_QSA) { - v |= SUPPORTED_FIBRE; - if (caps & FW_PORT_CAP_SPEED_1G) - v |= SUPPORTED_1000baseT_Full; - if (caps & FW_PORT_CAP_SPEED_10G) - v |= SUPPORTED_10000baseT_Full; - } else if (type == FW_PORT_TYPE_BP40_BA || - type == FW_PORT_TYPE_QSFP) { - v |= SUPPORTED_40000baseSR4_Full; - v |= SUPPORTED_FIBRE; + if (port_type == FW_PORT_TYPE_BT_SGMII || + port_type == FW_PORT_TYPE_BT_XFI || + port_type == FW_PORT_TYPE_BT_XAUI) { + return PORT_TP; + } else if (port_type == FW_PORT_TYPE_FIBER_XFI || + port_type == FW_PORT_TYPE_FIBER_XAUI) { + return PORT_FIBRE; + } else if (port_type == FW_PORT_TYPE_SFP || + port_type == FW_PORT_TYPE_QSFP_10G || + port_type == FW_PORT_TYPE_QSA || + port_type == FW_PORT_TYPE_QSFP) { + if (mod_type == FW_PORT_MOD_TYPE_LR || + mod_type == FW_PORT_MOD_TYPE_SR || + mod_type == FW_PORT_MOD_TYPE_ER || + mod_type == FW_PORT_MOD_TYPE_LRM) + return PORT_FIBRE; + else if (mod_type == FW_PORT_MOD_TYPE_TWINAX_PASSIVE || + mod_type == FW_PORT_MOD_TYPE_TWINAX_ACTIVE) + return PORT_DA; + else + return PORT_OTHER; } - if (caps & FW_PORT_CAP_ANEG) - v |= SUPPORTED_Autoneg; - return v; + return PORT_OTHER; } -static unsigned int to_fw_linkcaps(unsigned int caps) +/** + * speed_to_fw_caps - translate Port Speed to Firmware Port Capabilities + * @speed: speed in Kb/s + * + * Translates a specific Port Speed into a Firmware Port Capabilities + * value. + */ +static unsigned int speed_to_fw_caps(int speed) { - unsigned int v = 0; - - if (caps & ADVERTISED_100baseT_Full) - v |= FW_PORT_CAP_SPEED_100M; - if (caps & ADVERTISED_1000baseT_Full) - v |= FW_PORT_CAP_SPEED_1G; - if (caps & ADVERTISED_10000baseT_Full) - v |= FW_PORT_CAP_SPEED_10G; - if (caps & ADVERTISED_40000baseSR4_Full) - v |= FW_PORT_CAP_SPEED_40G; - return v; + if (speed == 100) + return FW_PORT_CAP_SPEED_100M; + if (speed == 1000) + return FW_PORT_CAP_SPEED_1G; + if (speed == 10000) + return FW_PORT_CAP_SPEED_10G; + if (speed == 25000) + return FW_PORT_CAP_SPEED_25G; + if (speed == 40000) + return FW_PORT_CAP_SPEED_40G; + if (speed == 100000) + return FW_PORT_CAP_SPEED_100G; + return 0; } -static int get_settings(struct net_device *dev, struct ethtool_cmd *cmd) +/** + * fw_caps_to_lmm - translate Firmware to ethtool Link Mode Mask + * @port_type: Firmware Port Type + * @fw_caps: Firmware Port Capabilities + * @link_mode_mask: ethtool Link Mode Mask + * + * Translate a Firmware Port Capabilities specification to an ethtool + * Link Mode Mask. + */ +static void fw_caps_to_lmm(enum fw_port_type port_type, + unsigned int fw_caps, + unsigned long *link_mode_mask) { - const struct port_info *p = netdev_priv(dev); - - if (p->port_type == FW_PORT_TYPE_BT_SGMII || - p->port_type == FW_PORT_TYPE_BT_XFI || - p->port_type == FW_PORT_TYPE_BT_XAUI) { - cmd->port = PORT_TP; - } else if (p->port_type == FW_PORT_TYPE_FIBER_XFI || - p->port_type == FW_PORT_TYPE_FIBER_XAUI) { - cmd->port = PORT_FIBRE; - } else if (p->port_type == FW_PORT_TYPE_SFP || - p->port_type == FW_PORT_TYPE_QSFP_10G || - p->port_type == FW_PORT_TYPE_QSA || - p->port_type == FW_PORT_TYPE_QSFP) { - if (p->mod_type == FW_PORT_MOD_TYPE_LR || - p->mod_type == FW_PORT_MOD_TYPE_SR || - p->mod_type == FW_PORT_MOD_TYPE_ER || - p->mod_type == FW_PORT_MOD_TYPE_LRM) - cmd->port = PORT_FIBRE; - else if (p->mod_type == FW_PORT_MOD_TYPE_TWINAX_PASSIVE || - p->mod_type == FW_PORT_MOD_TYPE_TWINAX_ACTIVE) - cmd->port = PORT_DA; - else - cmd->port = PORT_OTHER; + #define SET_LMM(__lmm_name) __set_bit(ETHTOOL_LINK_MODE_ ## __lmm_name \ + ## _BIT, link_mode_mask) + + #define FW_CAPS_TO_LMM(__fw_name, __lmm_name) \ + do { \ + if (fw_caps & FW_PORT_CAP_ ## __fw_name) \ + SET_LMM(__lmm_name); \ + } while (0) + + switch (port_type) { + case FW_PORT_TYPE_BT_SGMII: + case FW_PORT_TYPE_BT_XFI: + case FW_PORT_TYPE_BT_XAUI: + SET_LMM(TP); + FW_CAPS_TO_LMM(SPEED_100M, 100baseT_Full); + FW_CAPS_TO_LMM(SPEED_1G, 1000baseT_Full); + FW_CAPS_TO_LMM(SPEED_10G, 10000baseT_Full); + break; + + case FW_PORT_TYPE_KX4: + case FW_PORT_TYPE_KX: + SET_LMM(Backplane); + FW_CAPS_TO_LMM(SPEED_1G, 1000baseKX_Full); + FW_CAPS_TO_LMM(SPEED_10G, 10000baseKX4_Full); + break; + + case FW_PORT_TYPE_KR: + SET_LMM(Backplane); + SET_LMM(10000baseKR_Full); + break; + + case FW_PORT_TYPE_BP_AP: + SET_LMM(Backplane); + SET_LMM(10000baseR_FEC); + SET_LMM(10000baseKR_Full); + SET_LMM(1000baseKX_Full); + break; + + case FW_PORT_TYPE_BP4_AP: + SET_LMM(Backplane); + SET_LMM(10000baseR_FEC); + SET_LMM(10000baseKR_Full); + SET_LMM(1000baseKX_Full); + SET_LMM(10000baseKX4_Full); + break; + + case FW_PORT_TYPE_FIBER_XFI: + case FW_PORT_TYPE_FIBER_XAUI: + case FW_PORT_TYPE_SFP: + case FW_PORT_TYPE_QSFP_10G: + case FW_PORT_TYPE_QSA: + SET_LMM(FIBRE); + FW_CAPS_TO_LMM(SPEED_1G, 1000baseT_Full); + FW_CAPS_TO_LMM(SPEED_10G, 10000baseT_Full); + break; + + case FW_PORT_TYPE_BP40_BA: + case FW_PORT_TYPE_QSFP: + SET_LMM(FIBRE); + SET_LMM(40000baseSR4_Full); + break; + + case FW_PORT_TYPE_CR_QSFP: + case FW_PORT_TYPE_SFP28: + SET_LMM(FIBRE); + SET_LMM(25000baseCR_Full); + break; + + case FW_PORT_TYPE_KR4_100G: + case FW_PORT_TYPE_CR4_QSFP: + SET_LMM(FIBRE); + SET_LMM(100000baseCR4_Full); + break; + + default: + break; + } + + FW_CAPS_TO_LMM(ANEG, Autoneg); + FW_CAPS_TO_LMM(802_3_PAUSE, Pause); + FW_CAPS_TO_LMM(802_3_ASM_DIR, Asym_Pause); + + #undef FW_CAPS_TO_LMM + #undef SET_LMM +} + +/** + * lmm_to_fw_caps - translate ethtool Link Mode Mask to Firmware + * capabilities + * + * @link_mode_mask: ethtool Link Mode Mask + * + * Translate ethtool Link Mode Mask into a Firmware Port capabilities + * value. + */ +static unsigned int lmm_to_fw_caps(const unsigned long *link_mode_mask) +{ + unsigned int fw_caps = 0; + + #define LMM_TO_FW_CAPS(__lmm_name, __fw_name) \ + do { \ + if (test_bit(ETHTOOL_LINK_MODE_ ## __lmm_name ## _BIT, \ + link_mode_mask)) \ + fw_caps |= FW_PORT_CAP_ ## __fw_name; \ + } while (0) + + LMM_TO_FW_CAPS(100baseT_Full, SPEED_100M); + LMM_TO_FW_CAPS(1000baseT_Full, SPEED_1G); + LMM_TO_FW_CAPS(10000baseT_Full, SPEED_10G); + LMM_TO_FW_CAPS(40000baseSR4_Full, SPEED_40G); + LMM_TO_FW_CAPS(25000baseCR_Full, SPEED_25G); + LMM_TO_FW_CAPS(100000baseCR4_Full, SPEED_100G); + + #undef LMM_TO_FW_CAPS + + return fw_caps; +} + +static int get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings *link_ksettings) +{ + const struct port_info *pi = netdev_priv(dev); + struct ethtool_link_settings *base = &link_ksettings->base; + + ethtool_link_ksettings_zero_link_mode(link_ksettings, supported); + ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising); + ethtool_link_ksettings_zero_link_mode(link_ksettings, lp_advertising); + + base->port = from_fw_port_mod_type(pi->port_type, pi->mod_type); + + if (pi->mdio_addr >= 0) { + base->phy_address = pi->mdio_addr; + base->mdio_support = (pi->port_type == FW_PORT_TYPE_BT_SGMII + ? ETH_MDIO_SUPPORTS_C22 + : ETH_MDIO_SUPPORTS_C45); } else { - cmd->port = PORT_OTHER; + base->phy_address = 255; + base->mdio_support = 0; } - if (p->mdio_addr >= 0) { - cmd->phy_address = p->mdio_addr; - cmd->transceiver = XCVR_EXTERNAL; - cmd->mdio_support = p->port_type == FW_PORT_TYPE_BT_SGMII ? - MDIO_SUPPORTS_C22 : MDIO_SUPPORTS_C45; + fw_caps_to_lmm(pi->port_type, pi->link_cfg.supported, + link_ksettings->link_modes.supported); + fw_caps_to_lmm(pi->port_type, pi->link_cfg.advertising, + link_ksettings->link_modes.advertising); + fw_caps_to_lmm(pi->port_type, pi->link_cfg.lp_advertising, + link_ksettings->link_modes.lp_advertising); + + if (netif_carrier_ok(dev)) { + base->speed = pi->link_cfg.speed; + base->duplex = DUPLEX_FULL; } else { - cmd->phy_address = 0; /* not really, but no better option */ - cmd->transceiver = XCVR_INTERNAL; - cmd->mdio_support = 0; + base->speed = SPEED_UNKNOWN; + base->duplex = DUPLEX_UNKNOWN; } - cmd->supported = from_fw_linkcaps(p->port_type, p->link_cfg.supported); - cmd->advertising = from_fw_linkcaps(p->port_type, - p->link_cfg.advertising); - ethtool_cmd_speed_set(cmd, - netif_carrier_ok(dev) ? p->link_cfg.speed : 0); - cmd->duplex = DUPLEX_FULL; - cmd->autoneg = p->link_cfg.autoneg; - cmd->maxtxpkt = 0; - cmd->maxrxpkt = 0; - return 0; -} + base->autoneg = pi->link_cfg.autoneg; + if (pi->link_cfg.supported & FW_PORT_CAP_ANEG) + ethtool_link_ksettings_add_link_mode(link_ksettings, + supported, Autoneg); + if (pi->link_cfg.autoneg) + ethtool_link_ksettings_add_link_mode(link_ksettings, + advertising, Autoneg); -static unsigned int speed_to_caps(int speed) -{ - if (speed == 100) - return FW_PORT_CAP_SPEED_100M; - if (speed == 1000) - return FW_PORT_CAP_SPEED_1G; - if (speed == 10000) - return FW_PORT_CAP_SPEED_10G; - if (speed == 40000) - return FW_PORT_CAP_SPEED_40G; return 0; } -static int set_settings(struct net_device *dev, struct ethtool_cmd *cmd) +static int set_link_ksettings(struct net_device *dev, + const struct ethtool_link_ksettings + *link_ksettings) { - unsigned int cap; - struct port_info *p = netdev_priv(dev); - struct link_config *lc = &p->link_cfg; - u32 speed = ethtool_cmd_speed(cmd); + struct port_info *pi = netdev_priv(dev); + struct link_config *lc = &pi->link_cfg; + const struct ethtool_link_settings *base = &link_ksettings->base; struct link_config old_lc; - int ret; + unsigned int fw_caps; + int ret = 0; - if (cmd->duplex != DUPLEX_FULL) /* only full-duplex supported */ + /* only full-duplex supported */ + if (base->duplex != DUPLEX_FULL) return -EINVAL; if (!(lc->supported & FW_PORT_CAP_ANEG)) { /* PHY offers a single speed. See if that's what's * being requested. */ - if (cmd->autoneg == AUTONEG_DISABLE && - (lc->supported & speed_to_caps(speed))) + if (base->autoneg == AUTONEG_DISABLE && + (lc->supported & speed_to_fw_caps(base->speed))) return 0; return -EINVAL; } old_lc = *lc; - if (cmd->autoneg == AUTONEG_DISABLE) { - cap = speed_to_caps(speed); + if (base->autoneg == AUTONEG_DISABLE) { + fw_caps = speed_to_fw_caps(base->speed); - if (!(lc->supported & cap)) + if (!(lc->supported & fw_caps)) return -EINVAL; - lc->requested_speed = cap; + lc->requested_speed = fw_caps; lc->advertising = 0; } else { - cap = to_fw_linkcaps(cmd->advertising); - if (!(lc->supported & cap)) + fw_caps = + lmm_to_fw_caps(link_ksettings->link_modes.advertising); + + if (!(lc->supported & fw_caps)) return -EINVAL; lc->requested_speed = 0; - lc->advertising = cap | FW_PORT_CAP_ANEG; + lc->advertising = fw_caps | FW_PORT_CAP_ANEG; } - lc->autoneg = cmd->autoneg; + lc->autoneg = base->autoneg; /* If the firmware rejects the Link Configuration request, back out * the changes and report the error. */ - ret = t4_link_l1cfg(p->adapter, p->adapter->mbox, p->tx_chan, lc); + ret = t4_link_l1cfg(pi->adapter, pi->adapter->mbox, pi->tx_chan, lc); if (ret) *lc = old_lc; @@ -1093,8 +1208,8 @@ static int get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *info, } static const struct ethtool_ops cxgb_ethtool_ops = { - .get_settings = get_settings, - .set_settings = set_settings, + .get_link_ksettings = get_link_ksettings, + .set_link_ksettings = set_link_ksettings, .get_drvinfo = get_drvinfo, .get_msglevel = get_msglevel, .set_msglevel = set_msglevel, diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c index a63addb..dc92c80 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c @@ -7219,6 +7219,7 @@ void t4_handle_get_port_info(struct port_info *pi, const __be64 *rpl) lc->speed = speed; lc->fc = fc; lc->supported = be16_to_cpu(p->u.info.pcap); + lc->lp_advertising = be16_to_cpu(p->u.info.lpacap); t4_os_link_changed(adap, pi->port_id, link_ok); } } @@ -7284,6 +7285,7 @@ static void get_pci_mode(struct adapter *adapter, struct pci_params *p) static void init_link_config(struct link_config *lc, unsigned int caps) { lc->supported = caps; + lc->lp_advertising = 0; lc->requested_speed = 0; lc->speed = 0; lc->requested_fc = lc->fc = PAUSE_RX | PAUSE_TX; diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h index 392d664..a89b307 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h @@ -2249,20 +2249,20 @@ struct fw_acl_vlan_cmd { enum fw_port_cap { FW_PORT_CAP_SPEED_100M = 0x0001, FW_PORT_CAP_SPEED_1G = 0x0002, - FW_PORT_CAP_SPEED_2_5G = 0x0004, + FW_PORT_CAP_SPEED_25G = 0x0004, FW_PORT_CAP_SPEED_10G = 0x0008, FW_PORT_CAP_SPEED_40G = 0x0010, FW_PORT_CAP_SPEED_100G = 0x0020, FW_PORT_CAP_FC_RX = 0x0040, FW_PORT_CAP_FC_TX = 0x0080, FW_PORT_CAP_ANEG = 0x0100, - FW_PORT_CAP_MDI_0 = 0x0200, - FW_PORT_CAP_MDI_1 = 0x0400, - FW_PORT_CAP_BEAN = 0x0800, - FW_PORT_CAP_PMA_LPBK = 0x1000, - FW_PORT_CAP_PCS_LPBK = 0x2000, - FW_PORT_CAP_PHYXS_LPBK = 0x4000, - FW_PORT_CAP_FAR_END_LPBK = 0x8000, + FW_PORT_CAP_MDIX = 0x0200, + FW_PORT_CAP_MDIAUTO = 0x0400, + FW_PORT_CAP_FEC = 0x0800, + FW_PORT_CAP_TECHKR = 0x1000, + FW_PORT_CAP_TECHKX4 = 0x2000, + FW_PORT_CAP_802_3_PAUSE = 0x4000, + FW_PORT_CAP_802_3_ASM_DIR = 0x8000, }; enum fw_port_mdi { @@ -2376,7 +2376,8 @@ struct fw_port_cmd { __u8 cbllen; __u8 auxlinfo; __u8 dcbxdis_pkd; - __u8 r8_lo[3]; + __u8 r8_lo; + __be16 lpacap; __be64 r9; } info; struct fw_port_diags { @@ -2555,6 +2556,11 @@ enum fw_port_type { FW_PORT_TYPE_QSA, FW_PORT_TYPE_QSFP, FW_PORT_TYPE_BP40_BA, + FW_PORT_TYPE_KR4_100G, + FW_PORT_TYPE_CR4_QSFP, + FW_PORT_TYPE_CR_QSFP, + FW_PORT_TYPE_CR2_QSFP, + FW_PORT_TYPE_SFP28, FW_PORT_TYPE_NONE = FW_PORT_CMD_PTYPE_M }; diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index 9f55264..e116bb8 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -1201,105 +1201,187 @@ static void cxgb4vf_poll_controller(struct net_device *dev) * state of the port to which we're linked. */ -static unsigned int t4vf_from_fw_linkcaps(enum fw_port_type type, - unsigned int caps) -{ - unsigned int v = 0; - - if (type == FW_PORT_TYPE_BT_SGMII || type == FW_PORT_TYPE_BT_XFI || - type == FW_PORT_TYPE_BT_XAUI) { - v |= SUPPORTED_TP; - if (caps & FW_PORT_CAP_SPEED_100M) - v |= SUPPORTED_100baseT_Full; - if (caps & FW_PORT_CAP_SPEED_1G) - v |= SUPPORTED_1000baseT_Full; - if (caps & FW_PORT_CAP_SPEED_10G) - v |= SUPPORTED_10000baseT_Full; - } else if (type == FW_PORT_TYPE_KX4 || type == FW_PORT_TYPE_KX) { - v |= SUPPORTED_Backplane; - if (caps & FW_PORT_CAP_SPEED_1G) - v |= SUPPORTED_1000baseKX_Full; - if (caps & FW_PORT_CAP_SPEED_10G) - v |= SUPPORTED_10000baseKX4_Full; - } else if (type == FW_PORT_TYPE_KR) - v |= SUPPORTED_Backplane | SUPPORTED_10000baseKR_Full; - else if (type == FW_PORT_TYPE_BP_AP) - v |= SUPPORTED_Backplane | SUPPORTED_10000baseR_FEC | - SUPPORTED_10000baseKR_Full | SUPPORTED_1000baseKX_Full; - else if (type == FW_PORT_TYPE_BP4_AP) - v |= SUPPORTED_Backplane | SUPPORTED_10000baseR_FEC | - SUPPORTED_10000baseKR_Full | SUPPORTED_1000baseKX_Full | - SUPPORTED_10000baseKX4_Full; - else if (type == FW_PORT_TYPE_FIBER_XFI || - type == FW_PORT_TYPE_FIBER_XAUI || - type == FW_PORT_TYPE_SFP || - type == FW_PORT_TYPE_QSFP_10G || - type == FW_PORT_TYPE_QSA) { - v |= SUPPORTED_FIBRE; - if (caps & FW_PORT_CAP_SPEED_1G) - v |= SUPPORTED_1000baseT_Full; - if (caps & FW_PORT_CAP_SPEED_10G) - v |= SUPPORTED_10000baseT_Full; - } else if (type == FW_PORT_TYPE_BP40_BA || - type == FW_PORT_TYPE_QSFP) { - v |= SUPPORTED_40000baseSR4_Full; - v |= SUPPORTED_FIBRE; - } - - if (caps & FW_PORT_CAP_ANEG) - v |= SUPPORTED_Autoneg; - return v; -} - -static int cxgb4vf_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) -{ - const struct port_info *p = netdev_priv(dev); - - if (p->port_type == FW_PORT_TYPE_BT_SGMII || - p->port_type == FW_PORT_TYPE_BT_XFI || - p->port_type == FW_PORT_TYPE_BT_XAUI) - cmd->port = PORT_TP; - else if (p->port_type == FW_PORT_TYPE_FIBER_XFI || - p->port_type == FW_PORT_TYPE_FIBER_XAUI) - cmd->port = PORT_FIBRE; - else if (p->port_type == FW_PORT_TYPE_SFP || - p->port_type == FW_PORT_TYPE_QSFP_10G || - p->port_type == FW_PORT_TYPE_QSA || - p->port_type == FW_PORT_TYPE_QSFP) { - if (p->mod_type == FW_PORT_MOD_TYPE_LR || - p->mod_type == FW_PORT_MOD_TYPE_SR || - p->mod_type == FW_PORT_MOD_TYPE_ER || - p->mod_type == FW_PORT_MOD_TYPE_LRM) - cmd->port = PORT_FIBRE; - else if (p->mod_type == FW_PORT_MOD_TYPE_TWINAX_PASSIVE || - p->mod_type == FW_PORT_MOD_TYPE_TWINAX_ACTIVE) - cmd->port = PORT_DA; +/** + * from_fw_port_mod_type - translate Firmware Port/Module type to Ethtool + * @port_type: Firmware Port Type + * @mod_type: Firmware Module Type + * + * Translate Firmware Port/Module type to Ethtool Port Type. + */ +static int from_fw_port_mod_type(enum fw_port_type port_type, + enum fw_port_module_type mod_type) +{ + if (port_type == FW_PORT_TYPE_BT_SGMII || + port_type == FW_PORT_TYPE_BT_XFI || + port_type == FW_PORT_TYPE_BT_XAUI) { + return PORT_TP; + } else if (port_type == FW_PORT_TYPE_FIBER_XFI || + port_type == FW_PORT_TYPE_FIBER_XAUI) { + return PORT_FIBRE; + } else if (port_type == FW_PORT_TYPE_SFP || + port_type == FW_PORT_TYPE_QSFP_10G || + port_type == FW_PORT_TYPE_QSA || + port_type == FW_PORT_TYPE_QSFP) { + if (mod_type == FW_PORT_MOD_TYPE_LR || + mod_type == FW_PORT_MOD_TYPE_SR || + mod_type == FW_PORT_MOD_TYPE_ER || + mod_type == FW_PORT_MOD_TYPE_LRM) + return PORT_FIBRE; + else if (mod_type == FW_PORT_MOD_TYPE_TWINAX_PASSIVE || + mod_type == FW_PORT_MOD_TYPE_TWINAX_ACTIVE) + return PORT_DA; else - cmd->port = PORT_OTHER; - } else - cmd->port = PORT_OTHER; + return PORT_OTHER; + } + + return PORT_OTHER; +} + +/** + * fw_caps_to_lmm - translate Firmware to ethtool Link Mode Mask + * @port_type: Firmware Port Type + * @fw_caps: Firmware Port Capabilities + * @link_mode_mask: ethtool Link Mode Mask + * + * Translate a Firmware Port Capabilities specification to an ethtool + * Link Mode Mask. + */ +static void fw_caps_to_lmm(enum fw_port_type port_type, + unsigned int fw_caps, + unsigned long *link_mode_mask) +{ + #define SET_LMM(__lmm_name) __set_bit(ETHTOOL_LINK_MODE_ ## __lmm_name\ + ## _BIT, link_mode_mask) + + #define FW_CAPS_TO_LMM(__fw_name, __lmm_name) \ + do { \ + if (fw_caps & FW_PORT_CAP_ ## __fw_name) \ + SET_LMM(__lmm_name); \ + } while (0) + + switch (port_type) { + case FW_PORT_TYPE_BT_SGMII: + case FW_PORT_TYPE_BT_XFI: + case FW_PORT_TYPE_BT_XAUI: + SET_LMM(TP); + FW_CAPS_TO_LMM(SPEED_100M, 100baseT_Full); + FW_CAPS_TO_LMM(SPEED_1G, 1000baseT_Full); + FW_CAPS_TO_LMM(SPEED_10G, 10000baseT_Full); + break; + + case FW_PORT_TYPE_KX4: + case FW_PORT_TYPE_KX: + SET_LMM(Backplane); + FW_CAPS_TO_LMM(SPEED_1G, 1000baseKX_Full); + FW_CAPS_TO_LMM(SPEED_10G, 10000baseKX4_Full); + break; + + case FW_PORT_TYPE_KR: + SET_LMM(Backplane); + SET_LMM(10000baseKR_Full); + break; + + case FW_PORT_TYPE_BP_AP: + SET_LMM(Backplane); + SET_LMM(10000baseR_FEC); + SET_LMM(10000baseKR_Full); + SET_LMM(1000baseKX_Full); + break; + + case FW_PORT_TYPE_BP4_AP: + SET_LMM(Backplane); + SET_LMM(10000baseR_FEC); + SET_LMM(10000baseKR_Full); + SET_LMM(1000baseKX_Full); + SET_LMM(10000baseKX4_Full); + break; + + case FW_PORT_TYPE_FIBER_XFI: + case FW_PORT_TYPE_FIBER_XAUI: + case FW_PORT_TYPE_SFP: + case FW_PORT_TYPE_QSFP_10G: + case FW_PORT_TYPE_QSA: + SET_LMM(FIBRE); + FW_CAPS_TO_LMM(SPEED_1G, 1000baseT_Full); + FW_CAPS_TO_LMM(SPEED_10G, 10000baseT_Full); + break; - if (p->mdio_addr >= 0) { - cmd->phy_address = p->mdio_addr; - cmd->transceiver = XCVR_EXTERNAL; - cmd->mdio_support = p->port_type == FW_PORT_TYPE_BT_SGMII ? - MDIO_SUPPORTS_C22 : MDIO_SUPPORTS_C45; + case FW_PORT_TYPE_BP40_BA: + case FW_PORT_TYPE_QSFP: + SET_LMM(FIBRE); + SET_LMM(40000baseSR4_Full); + break; + + case FW_PORT_TYPE_CR_QSFP: + case FW_PORT_TYPE_SFP28: + SET_LMM(FIBRE); + SET_LMM(25000baseCR_Full); + break; + + case FW_PORT_TYPE_KR4_100G: + case FW_PORT_TYPE_CR4_QSFP: + SET_LMM(FIBRE); + SET_LMM(100000baseCR4_Full); + break; + + default: + break; + } + + FW_CAPS_TO_LMM(ANEG, Autoneg); + FW_CAPS_TO_LMM(802_3_PAUSE, Pause); + FW_CAPS_TO_LMM(802_3_ASM_DIR, Asym_Pause); + + #undef FW_CAPS_TO_LMM + #undef SET_LMM +} + +static int cxgb4vf_get_link_ksettings(struct net_device *dev, + struct ethtool_link_ksettings + *link_ksettings) +{ + const struct port_info *pi = netdev_priv(dev); + struct ethtool_link_settings *base = &link_ksettings->base; + + ethtool_link_ksettings_zero_link_mode(link_ksettings, supported); + ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising); + ethtool_link_ksettings_zero_link_mode(link_ksettings, lp_advertising); + + base->port = from_fw_port_mod_type(pi->port_type, pi->mod_type); + + if (pi->mdio_addr >= 0) { + base->phy_address = pi->mdio_addr; + base->mdio_support = (pi->port_type == FW_PORT_TYPE_BT_SGMII + ? ETH_MDIO_SUPPORTS_C22 + : ETH_MDIO_SUPPORTS_C45); + } else { + base->phy_address = 255; + base->mdio_support = 0; + } + + fw_caps_to_lmm(pi->port_type, pi->link_cfg.supported, + link_ksettings->link_modes.supported); + fw_caps_to_lmm(pi->port_type, pi->link_cfg.advertising, + link_ksettings->link_modes.advertising); + fw_caps_to_lmm(pi->port_type, pi->link_cfg.lp_advertising, + link_ksettings->link_modes.lp_advertising); + + if (netif_carrier_ok(dev)) { + base->speed = pi->link_cfg.speed; + base->duplex = DUPLEX_FULL; } else { - cmd->phy_address = 0; /* not really, but no better option */ - cmd->transceiver = XCVR_INTERNAL; - cmd->mdio_support = 0; - } - - cmd->supported = t4vf_from_fw_linkcaps(p->port_type, - p->link_cfg.supported); - cmd->advertising = t4vf_from_fw_linkcaps(p->port_type, - p->link_cfg.advertising); - ethtool_cmd_speed_set(cmd, - netif_carrier_ok(dev) ? p->link_cfg.speed : 0); - cmd->duplex = DUPLEX_FULL; - cmd->autoneg = p->link_cfg.autoneg; - cmd->maxtxpkt = 0; - cmd->maxrxpkt = 0; + base->speed = SPEED_UNKNOWN; + base->duplex = DUPLEX_UNKNOWN; + } + + base->autoneg = pi->link_cfg.autoneg; + if (pi->link_cfg.supported & FW_PORT_CAP_ANEG) + ethtool_link_ksettings_add_link_mode(link_ksettings, + supported, Autoneg); + if (pi->link_cfg.autoneg) + ethtool_link_ksettings_add_link_mode(link_ksettings, + advertising, Autoneg); + return 0; } @@ -1675,7 +1757,7 @@ static void cxgb4vf_get_wol(struct net_device *dev, #define TSO_FLAGS (NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_TSO_ECN) static const struct ethtool_ops cxgb4vf_ethtool_ops = { - .get_settings = cxgb4vf_get_settings, + .get_link_ksettings = cxgb4vf_get_link_ksettings, .get_drvinfo = cxgb4vf_get_drvinfo, .get_msglevel = cxgb4vf_get_msglevel, .set_msglevel = cxgb4vf_set_msglevel, diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h index 438374a..8ee5414 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h +++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_common.h @@ -107,6 +107,7 @@ struct t4vf_port_stats { struct link_config { unsigned int supported; /* link capabilities */ unsigned int advertising; /* advertised capabilities */ + unsigned short lp_advertising; /* peer advertised capabilities */ unsigned short requested_speed; /* speed user has requested */ unsigned short speed; /* actual link speed */ unsigned char requested_fc; /* flow control user has requested */ diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c index 61bfe86..427bfa7 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c @@ -328,6 +328,7 @@ int t4vf_wr_mbox_core(struct adapter *adapter, const void *cmd, int size, static void init_link_config(struct link_config *lc, unsigned int caps) { lc->supported = caps; + lc->lp_advertising = 0; lc->requested_speed = 0; lc->speed = 0; lc->requested_fc = lc->fc = PAUSE_RX | PAUSE_TX; @@ -1743,6 +1744,8 @@ int t4vf_handle_fw_rpl(struct adapter *adapter, const __be64 *rpl) lc->fc = fc; lc->supported = be16_to_cpu(port_cmd->u.info.pcap); + lc->lp_advertising = + be16_to_cpu(port_cmd->u.info.lpacap); t4vf_os_link_changed(adapter, pidx, link_ok); } } -- cgit v0.10.2 From dba479f3d60a9d1f90458a4c6eb0754ecd22348c Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Thu, 21 Jul 2016 12:42:10 -0400 Subject: net: bridge: fix br_stp_enable_bridge comment br_stp_enable_bridge() does take the br->lock spinlock. Fix its wrongly pasted comment and use the same as br_stp_disable_bridge(). Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/net/bridge/br_stp_if.c b/net/bridge/br_stp_if.c index 984d462..341caa0 100644 --- a/net/bridge/br_stp_if.c +++ b/net/bridge/br_stp_if.c @@ -55,7 +55,7 @@ void br_init_port(struct net_bridge_port *p) netdev_err(p->dev, "failed to set HW ageing time\n"); } -/* called under bridge lock */ +/* NO locks held */ void br_stp_enable_bridge(struct net_bridge *br) { struct net_bridge_port *p; -- cgit v0.10.2 From 9e0b27fe5ada7752577f3e1260eec44e79476142 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Thu, 21 Jul 2016 12:42:19 -0400 Subject: net: bridge: br_set_ageing_time takes a clock_t Change the ageing_time type in br_set_ageing_time() from u32 to what it is expected to be, i.e. a clock_t. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index b308826..aac2a6e 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -975,7 +975,7 @@ void __br_set_forward_delay(struct net_bridge *br, unsigned long t); int br_set_forward_delay(struct net_bridge *br, unsigned long x); int br_set_hello_time(struct net_bridge *br, unsigned long x); int br_set_max_age(struct net_bridge *br, unsigned long x); -int br_set_ageing_time(struct net_bridge *br, u32 ageing_time); +int br_set_ageing_time(struct net_bridge *br, clock_t ageing_time); /* br_stp_if.c */ diff --git a/net/bridge/br_stp.c b/net/bridge/br_stp.c index 9cb7044..9258b8e 100644 --- a/net/bridge/br_stp.c +++ b/net/bridge/br_stp.c @@ -570,7 +570,7 @@ int br_set_max_age(struct net_bridge *br, unsigned long val) * * Offloaded switch entries maybe more restrictive */ -int br_set_ageing_time(struct net_bridge *br, u32 ageing_time) +int br_set_ageing_time(struct net_bridge *br, clock_t ageing_time) { struct switchdev_attr attr = { .orig_dev = br->dev, -- cgit v0.10.2 From b8b9d81b36a4adada3ef3504c252e2518393884d Mon Sep 17 00:00:00 2001 From: Varun Prakash Date: Thu, 21 Jul 2016 22:57:14 +0530 Subject: libcxgb: add library module for Chelsio drivers Add common library module(libcxgb.ko) for Chelsio drivers to remove duplicate code. Code for iSCSI DDP Page Pod Manager is moved from cxgb4.ko to libcxgb.ko. Earlier only cxgbit.ko was using this code, now cxgb3i and cxgb4i will also use common Page Pod manager code. In future this module will have common connection management and hardware specific code that can be shared by multiple Chelsio drivers. Signed-off-by: Varun Prakash Reviewed-by: Steve Wise Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/chelsio/Kconfig b/drivers/net/ethernet/chelsio/Kconfig index 4686a85..5713e83 100644 --- a/drivers/net/ethernet/chelsio/Kconfig +++ b/drivers/net/ethernet/chelsio/Kconfig @@ -96,17 +96,6 @@ config CHELSIO_T4_DCB If unsure, say N. -config CHELSIO_T4_UWIRE - bool "Unified Wire Support for Chelsio T5 cards" - default n - depends on CHELSIO_T4 - ---help--- - Enable unified-wire offload features. - Say Y here if you want to enable unified-wire over Ethernet - in the driver. - - If unsure, say N. - config CHELSIO_T4_FCOE bool "Fibre Channel over Ethernet (FCoE) Support for Chelsio T5 cards" default n @@ -137,4 +126,9 @@ config CHELSIO_T4VF To compile this driver as a module choose M here; the module will be called cxgb4vf. +config CHELSIO_LIB + tristate + ---help--- + Common library for Chelsio drivers. + endif # NET_VENDOR_CHELSIO diff --git a/drivers/net/ethernet/chelsio/Makefile b/drivers/net/ethernet/chelsio/Makefile index 390510b..b6a5eec 100644 --- a/drivers/net/ethernet/chelsio/Makefile +++ b/drivers/net/ethernet/chelsio/Makefile @@ -6,3 +6,4 @@ obj-$(CONFIG_CHELSIO_T1) += cxgb/ obj-$(CONFIG_CHELSIO_T3) += cxgb3/ obj-$(CONFIG_CHELSIO_T4) += cxgb4/ obj-$(CONFIG_CHELSIO_T4VF) += cxgb4vf/ +obj-$(CONFIG_CHELSIO_LIB) += libcxgb/ diff --git a/drivers/net/ethernet/chelsio/cxgb4/Makefile b/drivers/net/ethernet/chelsio/cxgb4/Makefile index 85c9282..ace0ab9 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/Makefile +++ b/drivers/net/ethernet/chelsio/cxgb4/Makefile @@ -7,5 +7,4 @@ obj-$(CONFIG_CHELSIO_T4) += cxgb4.o cxgb4-objs := cxgb4_main.o l2t.o t4_hw.o sge.o clip_tbl.o cxgb4_ethtool.o cxgb4-$(CONFIG_CHELSIO_T4_DCB) += cxgb4_dcb.o cxgb4-$(CONFIG_CHELSIO_T4_FCOE) += cxgb4_fcoe.o -cxgb4-$(CONFIG_CHELSIO_T4_UWIRE) += cxgb4_ppm.o cxgb4-$(CONFIG_DEBUG_FS) += cxgb4_debugfs.o diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ppm.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ppm.c deleted file mode 100644 index d88a7a7..0000000 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ppm.c +++ /dev/null @@ -1,464 +0,0 @@ -/* - * cxgb4_ppm.c: Chelsio common library for T4/T5 iSCSI PagePod Manager - * - * Copyright (c) 2016 Chelsio Communications, Inc. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - * Written by: Karen Xie (kxie@chelsio.com) - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include "cxgb4_ppm.h" - -/* Direct Data Placement - - * Directly place the iSCSI Data-In or Data-Out PDU's payload into - * pre-posted final destination host-memory buffers based on the - * Initiator Task Tag (ITT) in Data-In or Target Task Tag (TTT) - * in Data-Out PDUs. The host memory address is programmed into - * h/w in the format of pagepod entries. The location of the - * pagepod entry is encoded into ddp tag which is used as the base - * for ITT/TTT. - */ - -/* Direct-Data Placement page size adjustment - */ -int cxgbi_ppm_find_page_index(struct cxgbi_ppm *ppm, unsigned long pgsz) -{ - struct cxgbi_tag_format *tformat = &ppm->tformat; - int i; - - for (i = 0; i < DDP_PGIDX_MAX; i++) { - if (pgsz == 1UL << (DDP_PGSZ_BASE_SHIFT + - tformat->pgsz_order[i])) { - pr_debug("%s: %s ppm, pgsz %lu -> idx %d.\n", - __func__, ppm->ndev->name, pgsz, i); - return i; - } - } - pr_info("ippm: ddp page size %lu not supported.\n", pgsz); - return DDP_PGIDX_MAX; -} - -/* DDP setup & teardown - */ -static int ppm_find_unused_entries(unsigned long *bmap, - unsigned int max_ppods, - unsigned int start, - unsigned int nr, - unsigned int align_mask) -{ - unsigned long i; - - i = bitmap_find_next_zero_area(bmap, max_ppods, start, nr, align_mask); - - if (unlikely(i >= max_ppods) && (start > nr)) - i = bitmap_find_next_zero_area(bmap, max_ppods, 0, start - 1, - align_mask); - if (unlikely(i >= max_ppods)) - return -ENOSPC; - - bitmap_set(bmap, i, nr); - return (int)i; -} - -static void ppm_mark_entries(struct cxgbi_ppm *ppm, int i, int count, - unsigned long caller_data) -{ - struct cxgbi_ppod_data *pdata = ppm->ppod_data + i; - - pdata->caller_data = caller_data; - pdata->npods = count; - - if (pdata->color == ((1 << PPOD_IDX_SHIFT) - 1)) - pdata->color = 0; - else - pdata->color++; -} - -static int ppm_get_cpu_entries(struct cxgbi_ppm *ppm, unsigned int count, - unsigned long caller_data) -{ - struct cxgbi_ppm_pool *pool; - unsigned int cpu; - int i; - - cpu = get_cpu(); - pool = per_cpu_ptr(ppm->pool, cpu); - spin_lock_bh(&pool->lock); - put_cpu(); - - i = ppm_find_unused_entries(pool->bmap, ppm->pool_index_max, - pool->next, count, 0); - if (i < 0) { - pool->next = 0; - spin_unlock_bh(&pool->lock); - return -ENOSPC; - } - - pool->next = i + count; - if (pool->next >= ppm->pool_index_max) - pool->next = 0; - - spin_unlock_bh(&pool->lock); - - pr_debug("%s: cpu %u, idx %d + %d (%d), next %u.\n", - __func__, cpu, i, count, i + cpu * ppm->pool_index_max, - pool->next); - - i += cpu * ppm->pool_index_max; - ppm_mark_entries(ppm, i, count, caller_data); - - return i; -} - -static int ppm_get_entries(struct cxgbi_ppm *ppm, unsigned int count, - unsigned long caller_data) -{ - int i; - - spin_lock_bh(&ppm->map_lock); - i = ppm_find_unused_entries(ppm->ppod_bmap, ppm->bmap_index_max, - ppm->next, count, 0); - if (i < 0) { - ppm->next = 0; - spin_unlock_bh(&ppm->map_lock); - pr_debug("ippm: NO suitable entries %u available.\n", - count); - return -ENOSPC; - } - - ppm->next = i + count; - if (ppm->next >= ppm->bmap_index_max) - ppm->next = 0; - - spin_unlock_bh(&ppm->map_lock); - - pr_debug("%s: idx %d + %d (%d), next %u, caller_data 0x%lx.\n", - __func__, i, count, i + ppm->pool_rsvd, ppm->next, - caller_data); - - i += ppm->pool_rsvd; - ppm_mark_entries(ppm, i, count, caller_data); - - return i; -} - -static void ppm_unmark_entries(struct cxgbi_ppm *ppm, int i, int count) -{ - pr_debug("%s: idx %d + %d.\n", __func__, i, count); - - if (i < ppm->pool_rsvd) { - unsigned int cpu; - struct cxgbi_ppm_pool *pool; - - cpu = i / ppm->pool_index_max; - i %= ppm->pool_index_max; - - pool = per_cpu_ptr(ppm->pool, cpu); - spin_lock_bh(&pool->lock); - bitmap_clear(pool->bmap, i, count); - - if (i < pool->next) - pool->next = i; - spin_unlock_bh(&pool->lock); - - pr_debug("%s: cpu %u, idx %d, next %u.\n", - __func__, cpu, i, pool->next); - } else { - spin_lock_bh(&ppm->map_lock); - - i -= ppm->pool_rsvd; - bitmap_clear(ppm->ppod_bmap, i, count); - - if (i < ppm->next) - ppm->next = i; - spin_unlock_bh(&ppm->map_lock); - - pr_debug("%s: idx %d, next %u.\n", __func__, i, ppm->next); - } -} - -void cxgbi_ppm_ppod_release(struct cxgbi_ppm *ppm, u32 idx) -{ - struct cxgbi_ppod_data *pdata; - - if (idx >= ppm->ppmax) { - pr_warn("ippm: idx too big %u > %u.\n", idx, ppm->ppmax); - return; - } - - pdata = ppm->ppod_data + idx; - if (!pdata->npods) { - pr_warn("ippm: idx %u, npods 0.\n", idx); - return; - } - - pr_debug("release idx %u, npods %u.\n", idx, pdata->npods); - ppm_unmark_entries(ppm, idx, pdata->npods); -} -EXPORT_SYMBOL(cxgbi_ppm_ppod_release); - -int cxgbi_ppm_ppods_reserve(struct cxgbi_ppm *ppm, unsigned short nr_pages, - u32 per_tag_pg_idx, u32 *ppod_idx, - u32 *ddp_tag, unsigned long caller_data) -{ - struct cxgbi_ppod_data *pdata; - unsigned int npods; - int idx = -1; - unsigned int hwidx; - u32 tag; - - npods = (nr_pages + PPOD_PAGES_MAX - 1) >> PPOD_PAGES_SHIFT; - if (!npods) { - pr_warn("%s: pages %u -> npods %u, full.\n", - __func__, nr_pages, npods); - return -EINVAL; - } - - /* grab from cpu pool first */ - idx = ppm_get_cpu_entries(ppm, npods, caller_data); - /* try the general pool */ - if (idx < 0) - idx = ppm_get_entries(ppm, npods, caller_data); - if (idx < 0) { - pr_debug("ippm: pages %u, nospc %u, nxt %u, 0x%lx.\n", - nr_pages, npods, ppm->next, caller_data); - return idx; - } - - pdata = ppm->ppod_data + idx; - hwidx = ppm->base_idx + idx; - - tag = cxgbi_ppm_make_ddp_tag(hwidx, pdata->color); - - if (per_tag_pg_idx) - tag |= (per_tag_pg_idx << 30) & 0xC0000000; - - *ppod_idx = idx; - *ddp_tag = tag; - - pr_debug("ippm: sg %u, tag 0x%x(%u,%u), data 0x%lx.\n", - nr_pages, tag, idx, npods, caller_data); - - return npods; -} -EXPORT_SYMBOL(cxgbi_ppm_ppods_reserve); - -void cxgbi_ppm_make_ppod_hdr(struct cxgbi_ppm *ppm, u32 tag, - unsigned int tid, unsigned int offset, - unsigned int length, - struct cxgbi_pagepod_hdr *hdr) -{ - /* The ddp tag in pagepod should be with bit 31:30 set to 0. - * The ddp Tag on the wire should be with non-zero 31:30 to the peer - */ - tag &= 0x3FFFFFFF; - - hdr->vld_tid = htonl(PPOD_VALID_FLAG | PPOD_TID(tid)); - - hdr->rsvd = 0; - hdr->pgsz_tag_clr = htonl(tag & ppm->tformat.idx_clr_mask); - hdr->max_offset = htonl(length); - hdr->page_offset = htonl(offset); - - pr_debug("ippm: tag 0x%x, tid 0x%x, xfer %u, off %u.\n", - tag, tid, length, offset); -} -EXPORT_SYMBOL(cxgbi_ppm_make_ppod_hdr); - -static void ppm_free(struct cxgbi_ppm *ppm) -{ - vfree(ppm); -} - -static void ppm_destroy(struct kref *kref) -{ - struct cxgbi_ppm *ppm = container_of(kref, - struct cxgbi_ppm, - refcnt); - pr_info("ippm: kref 0, destroy %s ppm 0x%p.\n", - ppm->ndev->name, ppm); - - *ppm->ppm_pp = NULL; - - free_percpu(ppm->pool); - ppm_free(ppm); -} - -int cxgbi_ppm_release(struct cxgbi_ppm *ppm) -{ - if (ppm) { - int rv; - - rv = kref_put(&ppm->refcnt, ppm_destroy); - return rv; - } - return 1; -} - -static struct cxgbi_ppm_pool *ppm_alloc_cpu_pool(unsigned int *total, - unsigned int *pcpu_ppmax) -{ - struct cxgbi_ppm_pool *pools; - unsigned int ppmax = (*total) / num_possible_cpus(); - unsigned int max = (PCPU_MIN_UNIT_SIZE - sizeof(*pools)) << 3; - unsigned int bmap; - unsigned int alloc_sz; - unsigned int count = 0; - unsigned int cpu; - - /* make sure per cpu pool fits into PCPU_MIN_UNIT_SIZE */ - if (ppmax > max) - ppmax = max; - - /* pool size must be multiple of unsigned long */ - bmap = BITS_TO_LONGS(ppmax); - ppmax = (bmap * sizeof(unsigned long)) << 3; - - alloc_sz = sizeof(*pools) + sizeof(unsigned long) * bmap; - pools = __alloc_percpu(alloc_sz, __alignof__(struct cxgbi_ppm_pool)); - - if (!pools) - return NULL; - - for_each_possible_cpu(cpu) { - struct cxgbi_ppm_pool *ppool = per_cpu_ptr(pools, cpu); - - memset(ppool, 0, alloc_sz); - spin_lock_init(&ppool->lock); - count += ppmax; - } - - *total = count; - *pcpu_ppmax = ppmax; - - return pools; -} - -int cxgbi_ppm_init(void **ppm_pp, struct net_device *ndev, - struct pci_dev *pdev, void *lldev, - struct cxgbi_tag_format *tformat, - unsigned int ppmax, - unsigned int llimit, - unsigned int start, - unsigned int reserve_factor) -{ - struct cxgbi_ppm *ppm = (struct cxgbi_ppm *)(*ppm_pp); - struct cxgbi_ppm_pool *pool = NULL; - unsigned int ppmax_pool = 0; - unsigned int pool_index_max = 0; - unsigned int alloc_sz; - unsigned int ppod_bmap_size; - - if (ppm) { - pr_info("ippm: %s, ppm 0x%p,0x%p already initialized, %u/%u.\n", - ndev->name, ppm_pp, ppm, ppm->ppmax, ppmax); - kref_get(&ppm->refcnt); - return 1; - } - - if (reserve_factor) { - ppmax_pool = ppmax / reserve_factor; - pool = ppm_alloc_cpu_pool(&ppmax_pool, &pool_index_max); - - pr_debug("%s: ppmax %u, cpu total %u, per cpu %u.\n", - ndev->name, ppmax, ppmax_pool, pool_index_max); - } - - ppod_bmap_size = BITS_TO_LONGS(ppmax - ppmax_pool); - alloc_sz = sizeof(struct cxgbi_ppm) + - ppmax * (sizeof(struct cxgbi_ppod_data)) + - ppod_bmap_size * sizeof(unsigned long); - - ppm = vmalloc(alloc_sz); - if (!ppm) - goto release_ppm_pool; - - memset(ppm, 0, alloc_sz); - - ppm->ppod_bmap = (unsigned long *)(&ppm->ppod_data[ppmax]); - - if ((ppod_bmap_size >> 3) > (ppmax - ppmax_pool)) { - unsigned int start = ppmax - ppmax_pool; - unsigned int end = ppod_bmap_size >> 3; - - bitmap_set(ppm->ppod_bmap, ppmax, end - start); - pr_info("%s: %u - %u < %u * 8, mask extra bits %u, %u.\n", - __func__, ppmax, ppmax_pool, ppod_bmap_size, start, - end); - } - - spin_lock_init(&ppm->map_lock); - kref_init(&ppm->refcnt); - - memcpy(&ppm->tformat, tformat, sizeof(struct cxgbi_tag_format)); - - ppm->ppm_pp = ppm_pp; - ppm->ndev = ndev; - ppm->pdev = pdev; - ppm->lldev = lldev; - ppm->ppmax = ppmax; - ppm->next = 0; - ppm->llimit = llimit; - ppm->base_idx = start > llimit ? - (start - llimit + 1) >> PPOD_SIZE_SHIFT : 0; - ppm->bmap_index_max = ppmax - ppmax_pool; - - ppm->pool = pool; - ppm->pool_rsvd = ppmax_pool; - ppm->pool_index_max = pool_index_max; - - /* check one more time */ - if (*ppm_pp) { - ppm_free(ppm); - ppm = (struct cxgbi_ppm *)(*ppm_pp); - - pr_info("ippm: %s, ppm 0x%p,0x%p already initialized, %u/%u.\n", - ndev->name, ppm_pp, *ppm_pp, ppm->ppmax, ppmax); - - kref_get(&ppm->refcnt); - return 1; - } - *ppm_pp = ppm; - - ppm->tformat.pgsz_idx_dflt = cxgbi_ppm_find_page_index(ppm, PAGE_SIZE); - - pr_info("ippm %s: ppm 0x%p, 0x%p, base %u/%u, pg %lu,%u, rsvd %u,%u.\n", - ndev->name, ppm_pp, ppm, ppm->base_idx, ppm->ppmax, PAGE_SIZE, - ppm->tformat.pgsz_idx_dflt, ppm->pool_rsvd, - ppm->pool_index_max); - - return 0; - -release_ppm_pool: - free_percpu(pool); - return -ENOMEM; -} -EXPORT_SYMBOL(cxgbi_ppm_init); - -unsigned int cxgbi_tagmask_set(unsigned int ppmax) -{ - unsigned int bits = fls(ppmax); - - if (bits > PPOD_IDX_MAX_SIZE) - bits = PPOD_IDX_MAX_SIZE; - - pr_info("ippm: ppmax %u/0x%x -> bits %u, tagmask 0x%x.\n", - ppmax, ppmax, bits, 1 << (bits + PPOD_IDX_SHIFT)); - - return 1 << (bits + PPOD_IDX_SHIFT); -} diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ppm.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ppm.h deleted file mode 100644 index d487326..0000000 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ppm.h +++ /dev/null @@ -1,310 +0,0 @@ -/* - * cxgb4_ppm.h: Chelsio common library for T4/T5 iSCSI ddp operation - * - * Copyright (c) 2016 Chelsio Communications, Inc. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - * Written by: Karen Xie (kxie@chelsio.com) - */ - -#ifndef __CXGB4PPM_H__ -#define __CXGB4PPM_H__ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -struct cxgbi_pagepod_hdr { - u32 vld_tid; - u32 pgsz_tag_clr; - u32 max_offset; - u32 page_offset; - u64 rsvd; -}; - -#define PPOD_PAGES_MAX 4 -struct cxgbi_pagepod { - struct cxgbi_pagepod_hdr hdr; - u64 addr[PPOD_PAGES_MAX + 1]; -}; - -/* ddp tag format - * for a 32-bit tag: - * bit # - * 31 ..... ..... 0 - * X Y...Y Z...Z, where - * ^ ^^^^^ ^^^^ - * | | |____ when ddp bit = 0: color bits - * | | - * | |____ when ddp bit = 0: idx into the ddp memory region - * | - * |____ ddp bit: 0 - ddp tag, 1 - non-ddp tag - * - * [page selector:2] [sw/free bits] [0] [idx] [color:6] - */ - -#define DDP_PGIDX_MAX 4 -#define DDP_PGSZ_BASE_SHIFT 12 /* base page 4K */ - -struct cxgbi_task_tag_info { - unsigned char flags; -#define CXGBI_PPOD_INFO_FLAG_VALID 0x1 -#define CXGBI_PPOD_INFO_FLAG_MAPPED 0x2 - unsigned char cid; - unsigned short pg_shift; - unsigned int npods; - unsigned int idx; - unsigned int tag; - struct cxgbi_pagepod_hdr hdr; - int nents; - int nr_pages; - struct scatterlist *sgl; -}; - -struct cxgbi_tag_format { - unsigned char pgsz_order[DDP_PGIDX_MAX]; - unsigned char pgsz_idx_dflt; - unsigned char free_bits:4; - unsigned char color_bits:4; - unsigned char idx_bits; - unsigned char rsvd_bits; - unsigned int no_ddp_mask; - unsigned int idx_mask; - unsigned int color_mask; - unsigned int idx_clr_mask; - unsigned int rsvd_mask; -}; - -struct cxgbi_ppod_data { - unsigned char pg_idx:2; - unsigned char color:6; - unsigned char chan_id; - unsigned short npods; - unsigned long caller_data; -}; - -/* per cpu ppm pool */ -struct cxgbi_ppm_pool { - unsigned int base; /* base index */ - unsigned int next; /* next possible free index */ - spinlock_t lock; /* ppm pool lock */ - unsigned long bmap[0]; -} ____cacheline_aligned_in_smp; - -struct cxgbi_ppm { - struct kref refcnt; - struct net_device *ndev; /* net_device, 1st port */ - struct pci_dev *pdev; - void *lldev; - void **ppm_pp; - struct cxgbi_tag_format tformat; - unsigned int ppmax; - unsigned int llimit; - unsigned int base_idx; - - unsigned int pool_rsvd; - unsigned int pool_index_max; - struct cxgbi_ppm_pool __percpu *pool; - /* map lock */ - spinlock_t map_lock; /* ppm map lock */ - unsigned int bmap_index_max; - unsigned int next; - unsigned long *ppod_bmap; - struct cxgbi_ppod_data ppod_data[0]; -}; - -#define DDP_THRESHOLD 512 - -#define PPOD_PAGES_SHIFT 2 /* 4 pages per pod */ - -#define IPPOD_SIZE sizeof(struct cxgbi_pagepod) /* 64 */ -#define PPOD_SIZE_SHIFT 6 - -/* page pods are allocated in groups of this size (must be power of 2) */ -#define PPOD_CLUSTER_SIZE 16U - -#define ULPMEM_DSGL_MAX_NPPODS 16 /* 1024/PPOD_SIZE */ -#define ULPMEM_IDATA_MAX_NPPODS 3 /* (PPOD_SIZE * 3 + ulptx hdr) < 256B */ -#define PCIE_MEMWIN_MAX_NPPODS 16 /* 1024/PPOD_SIZE */ - -#define PPOD_COLOR_SHIFT 0 -#define PPOD_COLOR(x) ((x) << PPOD_COLOR_SHIFT) - -#define PPOD_IDX_SHIFT 6 -#define PPOD_IDX_MAX_SIZE 24 - -#define PPOD_TID_SHIFT 0 -#define PPOD_TID(x) ((x) << PPOD_TID_SHIFT) - -#define PPOD_TAG_SHIFT 6 -#define PPOD_TAG(x) ((x) << PPOD_TAG_SHIFT) - -#define PPOD_VALID_SHIFT 24 -#define PPOD_VALID(x) ((x) << PPOD_VALID_SHIFT) -#define PPOD_VALID_FLAG PPOD_VALID(1U) - -#define PPOD_PI_EXTRACT_CTL_SHIFT 31 -#define PPOD_PI_EXTRACT_CTL(x) ((x) << PPOD_PI_EXTRACT_CTL_SHIFT) -#define PPOD_PI_EXTRACT_CTL_FLAG V_PPOD_PI_EXTRACT_CTL(1U) - -#define PPOD_PI_TYPE_SHIFT 29 -#define PPOD_PI_TYPE_MASK 0x3 -#define PPOD_PI_TYPE(x) ((x) << PPOD_PI_TYPE_SHIFT) - -#define PPOD_PI_CHECK_CTL_SHIFT 27 -#define PPOD_PI_CHECK_CTL_MASK 0x3 -#define PPOD_PI_CHECK_CTL(x) ((x) << PPOD_PI_CHECK_CTL_SHIFT) - -#define PPOD_PI_REPORT_CTL_SHIFT 25 -#define PPOD_PI_REPORT_CTL_MASK 0x3 -#define PPOD_PI_REPORT_CTL(x) ((x) << PPOD_PI_REPORT_CTL_SHIFT) - -static inline int cxgbi_ppm_is_ddp_tag(struct cxgbi_ppm *ppm, u32 tag) -{ - return !(tag & ppm->tformat.no_ddp_mask); -} - -static inline int cxgbi_ppm_sw_tag_is_usable(struct cxgbi_ppm *ppm, - u32 tag) -{ - /* the sw tag must be using <= 31 bits */ - return !(tag & 0x80000000U); -} - -static inline int cxgbi_ppm_make_non_ddp_tag(struct cxgbi_ppm *ppm, - u32 sw_tag, - u32 *final_tag) -{ - struct cxgbi_tag_format *tformat = &ppm->tformat; - - if (!cxgbi_ppm_sw_tag_is_usable(ppm, sw_tag)) { - pr_info("sw_tag 0x%x NOT usable.\n", sw_tag); - return -EINVAL; - } - - if (!sw_tag) { - *final_tag = tformat->no_ddp_mask; - } else { - unsigned int shift = tformat->idx_bits + tformat->color_bits; - u32 lower = sw_tag & tformat->idx_clr_mask; - u32 upper = (sw_tag >> shift) << (shift + 1); - - *final_tag = upper | tformat->no_ddp_mask | lower; - } - return 0; -} - -static inline u32 cxgbi_ppm_decode_non_ddp_tag(struct cxgbi_ppm *ppm, - u32 tag) -{ - struct cxgbi_tag_format *tformat = &ppm->tformat; - unsigned int shift = tformat->idx_bits + tformat->color_bits; - u32 lower = tag & tformat->idx_clr_mask; - u32 upper = (tag >> tformat->rsvd_bits) << shift; - - return upper | lower; -} - -static inline u32 cxgbi_ppm_ddp_tag_get_idx(struct cxgbi_ppm *ppm, - u32 ddp_tag) -{ - u32 hw_idx = (ddp_tag >> PPOD_IDX_SHIFT) & - ppm->tformat.idx_mask; - - return hw_idx - ppm->base_idx; -} - -static inline u32 cxgbi_ppm_make_ddp_tag(unsigned int hw_idx, - unsigned char color) -{ - return (hw_idx << PPOD_IDX_SHIFT) | ((u32)color); -} - -static inline unsigned long -cxgbi_ppm_get_tag_caller_data(struct cxgbi_ppm *ppm, - u32 ddp_tag) -{ - u32 idx = cxgbi_ppm_ddp_tag_get_idx(ppm, ddp_tag); - - return ppm->ppod_data[idx].caller_data; -} - -/* sw bits are the free bits */ -static inline int cxgbi_ppm_ddp_tag_update_sw_bits(struct cxgbi_ppm *ppm, - u32 val, u32 orig_tag, - u32 *final_tag) -{ - struct cxgbi_tag_format *tformat = &ppm->tformat; - u32 v = val >> tformat->free_bits; - - if (v) { - pr_info("sw_bits 0x%x too large, avail bits %u.\n", - val, tformat->free_bits); - return -EINVAL; - } - if (!cxgbi_ppm_is_ddp_tag(ppm, orig_tag)) - return -EINVAL; - - *final_tag = (val << tformat->rsvd_bits) | - (orig_tag & ppm->tformat.rsvd_mask); - return 0; -} - -static inline void cxgbi_ppm_ppod_clear(struct cxgbi_pagepod *ppod) -{ - ppod->hdr.vld_tid = 0U; -} - -static inline void cxgbi_tagmask_check(unsigned int tagmask, - struct cxgbi_tag_format *tformat) -{ - unsigned int bits = fls(tagmask); - - /* reserve top most 2 bits for page selector */ - tformat->free_bits = 32 - 2 - bits; - tformat->rsvd_bits = bits; - tformat->color_bits = PPOD_IDX_SHIFT; - tformat->idx_bits = bits - 1 - PPOD_IDX_SHIFT; - tformat->no_ddp_mask = 1 << (bits - 1); - tformat->idx_mask = (1 << tformat->idx_bits) - 1; - tformat->color_mask = (1 << PPOD_IDX_SHIFT) - 1; - tformat->idx_clr_mask = (1 << (bits - 1)) - 1; - tformat->rsvd_mask = (1 << bits) - 1; - - pr_info("ippm: tagmask 0x%x, rsvd %u=%u+%u+1, mask 0x%x,0x%x, " - "pg %u,%u,%u,%u.\n", - tagmask, tformat->rsvd_bits, tformat->idx_bits, - tformat->color_bits, tformat->no_ddp_mask, tformat->rsvd_mask, - tformat->pgsz_order[0], tformat->pgsz_order[1], - tformat->pgsz_order[2], tformat->pgsz_order[3]); -} - -int cxgbi_ppm_find_page_index(struct cxgbi_ppm *ppm, unsigned long pgsz); -void cxgbi_ppm_make_ppod_hdr(struct cxgbi_ppm *ppm, u32 tag, - unsigned int tid, unsigned int offset, - unsigned int length, - struct cxgbi_pagepod_hdr *hdr); -void cxgbi_ppm_ppod_release(struct cxgbi_ppm *, u32 idx); -int cxgbi_ppm_ppods_reserve(struct cxgbi_ppm *, unsigned short nr_pages, - u32 per_tag_pg_idx, u32 *ppod_idx, u32 *ddp_tag, - unsigned long caller_data); -int cxgbi_ppm_init(void **ppm_pp, struct net_device *, struct pci_dev *, - void *lldev, struct cxgbi_tag_format *, - unsigned int ppmax, unsigned int llimit, - unsigned int start, - unsigned int reserve_factor); -int cxgbi_ppm_release(struct cxgbi_ppm *ppm); -void cxgbi_tagmask_check(unsigned int tagmask, struct cxgbi_tag_format *); -unsigned int cxgbi_tagmask_set(unsigned int ppmax); - -#endif /*__CXGB4PPM_H__*/ diff --git a/drivers/net/ethernet/chelsio/libcxgb/Makefile b/drivers/net/ethernet/chelsio/libcxgb/Makefile new file mode 100644 index 0000000..2362230 --- /dev/null +++ b/drivers/net/ethernet/chelsio/libcxgb/Makefile @@ -0,0 +1,3 @@ +obj-$(CONFIG_CHELSIO_LIB) += libcxgb.o + +libcxgb-y := libcxgb_ppm.o diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c new file mode 100644 index 0000000..01a4329 --- /dev/null +++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c @@ -0,0 +1,497 @@ +/* + * libcxgb_ppm.c: Chelsio common library for T3/T4/T5 iSCSI PagePod Manager + * + * Copyright (c) 2016 Chelsio Communications, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * Written by: Karen Xie (kxie@chelsio.com) + */ + +#define DRV_NAME "libcxgb" +#define DRV_VERSION "1.0.0-ko" +#define pr_fmt(fmt) DRV_NAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "libcxgb_ppm.h" + +/* Direct Data Placement - + * Directly place the iSCSI Data-In or Data-Out PDU's payload into + * pre-posted final destination host-memory buffers based on the + * Initiator Task Tag (ITT) in Data-In or Target Task Tag (TTT) + * in Data-Out PDUs. The host memory address is programmed into + * h/w in the format of pagepod entries. The location of the + * pagepod entry is encoded into ddp tag which is used as the base + * for ITT/TTT. + */ + +/* Direct-Data Placement page size adjustment + */ +int cxgbi_ppm_find_page_index(struct cxgbi_ppm *ppm, unsigned long pgsz) +{ + struct cxgbi_tag_format *tformat = &ppm->tformat; + int i; + + for (i = 0; i < DDP_PGIDX_MAX; i++) { + if (pgsz == 1UL << (DDP_PGSZ_BASE_SHIFT + + tformat->pgsz_order[i])) { + pr_debug("%s: %s ppm, pgsz %lu -> idx %d.\n", + __func__, ppm->ndev->name, pgsz, i); + return i; + } + } + pr_info("ippm: ddp page size %lu not supported.\n", pgsz); + return DDP_PGIDX_MAX; +} + +/* DDP setup & teardown + */ +static int ppm_find_unused_entries(unsigned long *bmap, + unsigned int max_ppods, + unsigned int start, + unsigned int nr, + unsigned int align_mask) +{ + unsigned long i; + + i = bitmap_find_next_zero_area(bmap, max_ppods, start, nr, align_mask); + + if (unlikely(i >= max_ppods) && (start > nr)) + i = bitmap_find_next_zero_area(bmap, max_ppods, 0, start - 1, + align_mask); + if (unlikely(i >= max_ppods)) + return -ENOSPC; + + bitmap_set(bmap, i, nr); + return (int)i; +} + +static void ppm_mark_entries(struct cxgbi_ppm *ppm, int i, int count, + unsigned long caller_data) +{ + struct cxgbi_ppod_data *pdata = ppm->ppod_data + i; + + pdata->caller_data = caller_data; + pdata->npods = count; + + if (pdata->color == ((1 << PPOD_IDX_SHIFT) - 1)) + pdata->color = 0; + else + pdata->color++; +} + +static int ppm_get_cpu_entries(struct cxgbi_ppm *ppm, unsigned int count, + unsigned long caller_data) +{ + struct cxgbi_ppm_pool *pool; + unsigned int cpu; + int i; + + cpu = get_cpu(); + pool = per_cpu_ptr(ppm->pool, cpu); + spin_lock_bh(&pool->lock); + put_cpu(); + + i = ppm_find_unused_entries(pool->bmap, ppm->pool_index_max, + pool->next, count, 0); + if (i < 0) { + pool->next = 0; + spin_unlock_bh(&pool->lock); + return -ENOSPC; + } + + pool->next = i + count; + if (pool->next >= ppm->pool_index_max) + pool->next = 0; + + spin_unlock_bh(&pool->lock); + + pr_debug("%s: cpu %u, idx %d + %d (%d), next %u.\n", + __func__, cpu, i, count, i + cpu * ppm->pool_index_max, + pool->next); + + i += cpu * ppm->pool_index_max; + ppm_mark_entries(ppm, i, count, caller_data); + + return i; +} + +static int ppm_get_entries(struct cxgbi_ppm *ppm, unsigned int count, + unsigned long caller_data) +{ + int i; + + spin_lock_bh(&ppm->map_lock); + i = ppm_find_unused_entries(ppm->ppod_bmap, ppm->bmap_index_max, + ppm->next, count, 0); + if (i < 0) { + ppm->next = 0; + spin_unlock_bh(&ppm->map_lock); + pr_debug("ippm: NO suitable entries %u available.\n", + count); + return -ENOSPC; + } + + ppm->next = i + count; + if (ppm->next >= ppm->bmap_index_max) + ppm->next = 0; + + spin_unlock_bh(&ppm->map_lock); + + pr_debug("%s: idx %d + %d (%d), next %u, caller_data 0x%lx.\n", + __func__, i, count, i + ppm->pool_rsvd, ppm->next, + caller_data); + + i += ppm->pool_rsvd; + ppm_mark_entries(ppm, i, count, caller_data); + + return i; +} + +static void ppm_unmark_entries(struct cxgbi_ppm *ppm, int i, int count) +{ + pr_debug("%s: idx %d + %d.\n", __func__, i, count); + + if (i < ppm->pool_rsvd) { + unsigned int cpu; + struct cxgbi_ppm_pool *pool; + + cpu = i / ppm->pool_index_max; + i %= ppm->pool_index_max; + + pool = per_cpu_ptr(ppm->pool, cpu); + spin_lock_bh(&pool->lock); + bitmap_clear(pool->bmap, i, count); + + if (i < pool->next) + pool->next = i; + spin_unlock_bh(&pool->lock); + + pr_debug("%s: cpu %u, idx %d, next %u.\n", + __func__, cpu, i, pool->next); + } else { + spin_lock_bh(&ppm->map_lock); + + i -= ppm->pool_rsvd; + bitmap_clear(ppm->ppod_bmap, i, count); + + if (i < ppm->next) + ppm->next = i; + spin_unlock_bh(&ppm->map_lock); + + pr_debug("%s: idx %d, next %u.\n", __func__, i, ppm->next); + } +} + +void cxgbi_ppm_ppod_release(struct cxgbi_ppm *ppm, u32 idx) +{ + struct cxgbi_ppod_data *pdata; + + if (idx >= ppm->ppmax) { + pr_warn("ippm: idx too big %u > %u.\n", idx, ppm->ppmax); + return; + } + + pdata = ppm->ppod_data + idx; + if (!pdata->npods) { + pr_warn("ippm: idx %u, npods 0.\n", idx); + return; + } + + pr_debug("release idx %u, npods %u.\n", idx, pdata->npods); + ppm_unmark_entries(ppm, idx, pdata->npods); +} +EXPORT_SYMBOL(cxgbi_ppm_ppod_release); + +int cxgbi_ppm_ppods_reserve(struct cxgbi_ppm *ppm, unsigned short nr_pages, + u32 per_tag_pg_idx, u32 *ppod_idx, + u32 *ddp_tag, unsigned long caller_data) +{ + struct cxgbi_ppod_data *pdata; + unsigned int npods; + int idx = -1; + unsigned int hwidx; + u32 tag; + + npods = (nr_pages + PPOD_PAGES_MAX - 1) >> PPOD_PAGES_SHIFT; + if (!npods) { + pr_warn("%s: pages %u -> npods %u, full.\n", + __func__, nr_pages, npods); + return -EINVAL; + } + + /* grab from cpu pool first */ + idx = ppm_get_cpu_entries(ppm, npods, caller_data); + /* try the general pool */ + if (idx < 0) + idx = ppm_get_entries(ppm, npods, caller_data); + if (idx < 0) { + pr_debug("ippm: pages %u, nospc %u, nxt %u, 0x%lx.\n", + nr_pages, npods, ppm->next, caller_data); + return idx; + } + + pdata = ppm->ppod_data + idx; + hwidx = ppm->base_idx + idx; + + tag = cxgbi_ppm_make_ddp_tag(hwidx, pdata->color); + + if (per_tag_pg_idx) + tag |= (per_tag_pg_idx << 30) & 0xC0000000; + + *ppod_idx = idx; + *ddp_tag = tag; + + pr_debug("ippm: sg %u, tag 0x%x(%u,%u), data 0x%lx.\n", + nr_pages, tag, idx, npods, caller_data); + + return npods; +} +EXPORT_SYMBOL(cxgbi_ppm_ppods_reserve); + +void cxgbi_ppm_make_ppod_hdr(struct cxgbi_ppm *ppm, u32 tag, + unsigned int tid, unsigned int offset, + unsigned int length, + struct cxgbi_pagepod_hdr *hdr) +{ + /* The ddp tag in pagepod should be with bit 31:30 set to 0. + * The ddp Tag on the wire should be with non-zero 31:30 to the peer + */ + tag &= 0x3FFFFFFF; + + hdr->vld_tid = htonl(PPOD_VALID_FLAG | PPOD_TID(tid)); + + hdr->rsvd = 0; + hdr->pgsz_tag_clr = htonl(tag & ppm->tformat.idx_clr_mask); + hdr->max_offset = htonl(length); + hdr->page_offset = htonl(offset); + + pr_debug("ippm: tag 0x%x, tid 0x%x, xfer %u, off %u.\n", + tag, tid, length, offset); +} +EXPORT_SYMBOL(cxgbi_ppm_make_ppod_hdr); + +static void ppm_free(struct cxgbi_ppm *ppm) +{ + vfree(ppm); +} + +static void ppm_destroy(struct kref *kref) +{ + struct cxgbi_ppm *ppm = container_of(kref, + struct cxgbi_ppm, + refcnt); + pr_info("ippm: kref 0, destroy %s ppm 0x%p.\n", + ppm->ndev->name, ppm); + + *ppm->ppm_pp = NULL; + + free_percpu(ppm->pool); + ppm_free(ppm); +} + +int cxgbi_ppm_release(struct cxgbi_ppm *ppm) +{ + if (ppm) { + int rv; + + rv = kref_put(&ppm->refcnt, ppm_destroy); + return rv; + } + return 1; +} + +static struct cxgbi_ppm_pool *ppm_alloc_cpu_pool(unsigned int *total, + unsigned int *pcpu_ppmax) +{ + struct cxgbi_ppm_pool *pools; + unsigned int ppmax = (*total) / num_possible_cpus(); + unsigned int max = (PCPU_MIN_UNIT_SIZE - sizeof(*pools)) << 3; + unsigned int bmap; + unsigned int alloc_sz; + unsigned int count = 0; + unsigned int cpu; + + /* make sure per cpu pool fits into PCPU_MIN_UNIT_SIZE */ + if (ppmax > max) + ppmax = max; + + /* pool size must be multiple of unsigned long */ + bmap = BITS_TO_LONGS(ppmax); + ppmax = (bmap * sizeof(unsigned long)) << 3; + + alloc_sz = sizeof(*pools) + sizeof(unsigned long) * bmap; + pools = __alloc_percpu(alloc_sz, __alignof__(struct cxgbi_ppm_pool)); + + if (!pools) + return NULL; + + for_each_possible_cpu(cpu) { + struct cxgbi_ppm_pool *ppool = per_cpu_ptr(pools, cpu); + + memset(ppool, 0, alloc_sz); + spin_lock_init(&ppool->lock); + count += ppmax; + } + + *total = count; + *pcpu_ppmax = ppmax; + + return pools; +} + +int cxgbi_ppm_init(void **ppm_pp, struct net_device *ndev, + struct pci_dev *pdev, void *lldev, + struct cxgbi_tag_format *tformat, + unsigned int ppmax, + unsigned int llimit, + unsigned int start, + unsigned int reserve_factor) +{ + struct cxgbi_ppm *ppm = (struct cxgbi_ppm *)(*ppm_pp); + struct cxgbi_ppm_pool *pool = NULL; + unsigned int ppmax_pool = 0; + unsigned int pool_index_max = 0; + unsigned int alloc_sz; + unsigned int ppod_bmap_size; + + if (ppm) { + pr_info("ippm: %s, ppm 0x%p,0x%p already initialized, %u/%u.\n", + ndev->name, ppm_pp, ppm, ppm->ppmax, ppmax); + kref_get(&ppm->refcnt); + return 1; + } + + if (reserve_factor) { + ppmax_pool = ppmax / reserve_factor; + pool = ppm_alloc_cpu_pool(&ppmax_pool, &pool_index_max); + + pr_debug("%s: ppmax %u, cpu total %u, per cpu %u.\n", + ndev->name, ppmax, ppmax_pool, pool_index_max); + } + + ppod_bmap_size = BITS_TO_LONGS(ppmax - ppmax_pool); + alloc_sz = sizeof(struct cxgbi_ppm) + + ppmax * (sizeof(struct cxgbi_ppod_data)) + + ppod_bmap_size * sizeof(unsigned long); + + ppm = vmalloc(alloc_sz); + if (!ppm) + goto release_ppm_pool; + + memset(ppm, 0, alloc_sz); + + ppm->ppod_bmap = (unsigned long *)(&ppm->ppod_data[ppmax]); + + if ((ppod_bmap_size >> 3) > (ppmax - ppmax_pool)) { + unsigned int start = ppmax - ppmax_pool; + unsigned int end = ppod_bmap_size >> 3; + + bitmap_set(ppm->ppod_bmap, ppmax, end - start); + pr_info("%s: %u - %u < %u * 8, mask extra bits %u, %u.\n", + __func__, ppmax, ppmax_pool, ppod_bmap_size, start, + end); + } + + spin_lock_init(&ppm->map_lock); + kref_init(&ppm->refcnt); + + memcpy(&ppm->tformat, tformat, sizeof(struct cxgbi_tag_format)); + + ppm->ppm_pp = ppm_pp; + ppm->ndev = ndev; + ppm->pdev = pdev; + ppm->lldev = lldev; + ppm->ppmax = ppmax; + ppm->next = 0; + ppm->llimit = llimit; + ppm->base_idx = start > llimit ? + (start - llimit + 1) >> PPOD_SIZE_SHIFT : 0; + ppm->bmap_index_max = ppmax - ppmax_pool; + + ppm->pool = pool; + ppm->pool_rsvd = ppmax_pool; + ppm->pool_index_max = pool_index_max; + + /* check one more time */ + if (*ppm_pp) { + ppm_free(ppm); + ppm = (struct cxgbi_ppm *)(*ppm_pp); + + pr_info("ippm: %s, ppm 0x%p,0x%p already initialized, %u/%u.\n", + ndev->name, ppm_pp, *ppm_pp, ppm->ppmax, ppmax); + + kref_get(&ppm->refcnt); + return 1; + } + *ppm_pp = ppm; + + ppm->tformat.pgsz_idx_dflt = cxgbi_ppm_find_page_index(ppm, PAGE_SIZE); + + pr_info("ippm %s: ppm 0x%p, 0x%p, base %u/%u, pg %lu,%u, rsvd %u,%u.\n", + ndev->name, ppm_pp, ppm, ppm->base_idx, ppm->ppmax, PAGE_SIZE, + ppm->tformat.pgsz_idx_dflt, ppm->pool_rsvd, + ppm->pool_index_max); + + return 0; + +release_ppm_pool: + free_percpu(pool); + return -ENOMEM; +} +EXPORT_SYMBOL(cxgbi_ppm_init); + +unsigned int cxgbi_tagmask_set(unsigned int ppmax) +{ + unsigned int bits = fls(ppmax); + + if (bits > PPOD_IDX_MAX_SIZE) + bits = PPOD_IDX_MAX_SIZE; + + pr_info("ippm: ppmax %u/0x%x -> bits %u, tagmask 0x%x.\n", + ppmax, ppmax, bits, 1 << (bits + PPOD_IDX_SHIFT)); + + return 1 << (bits + PPOD_IDX_SHIFT); +} + +MODULE_AUTHOR("Chelsio Communications"); +MODULE_DESCRIPTION("Chelsio common library"); +MODULE_VERSION(DRV_VERSION); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h new file mode 100644 index 0000000..e995a1a --- /dev/null +++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.h @@ -0,0 +1,334 @@ +/* + * libcxgb_ppm.h: Chelsio common library for T3/T4/T5 iSCSI ddp operation + * + * Copyright (c) 2016 Chelsio Communications, Inc. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + * + * Written by: Karen Xie (kxie@chelsio.com) + */ + +#ifndef __LIBCXGB_PPM_H__ +#define __LIBCXGB_PPM_H__ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +struct cxgbi_pagepod_hdr { + u32 vld_tid; + u32 pgsz_tag_clr; + u32 max_offset; + u32 page_offset; + u64 rsvd; +}; + +#define PPOD_PAGES_MAX 4 +struct cxgbi_pagepod { + struct cxgbi_pagepod_hdr hdr; + u64 addr[PPOD_PAGES_MAX + 1]; +}; + +/* ddp tag format + * for a 32-bit tag: + * bit # + * 31 ..... ..... 0 + * X Y...Y Z...Z, where + * ^ ^^^^^ ^^^^ + * | | |____ when ddp bit = 0: color bits + * | | + * | |____ when ddp bit = 0: idx into the ddp memory region + * | + * |____ ddp bit: 0 - ddp tag, 1 - non-ddp tag + * + * [page selector:2] [sw/free bits] [0] [idx] [color:6] + */ + +#define DDP_PGIDX_MAX 4 +#define DDP_PGSZ_BASE_SHIFT 12 /* base page 4K */ + +struct cxgbi_task_tag_info { + unsigned char flags; +#define CXGBI_PPOD_INFO_FLAG_VALID 0x1 +#define CXGBI_PPOD_INFO_FLAG_MAPPED 0x2 + unsigned char cid; + unsigned short pg_shift; + unsigned int npods; + unsigned int idx; + unsigned int tag; + struct cxgbi_pagepod_hdr hdr; + int nents; + int nr_pages; + struct scatterlist *sgl; +}; + +struct cxgbi_tag_format { + unsigned char pgsz_order[DDP_PGIDX_MAX]; + unsigned char pgsz_idx_dflt; + unsigned char free_bits:4; + unsigned char color_bits:4; + unsigned char idx_bits; + unsigned char rsvd_bits; + unsigned int no_ddp_mask; + unsigned int idx_mask; + unsigned int color_mask; + unsigned int idx_clr_mask; + unsigned int rsvd_mask; +}; + +struct cxgbi_ppod_data { + unsigned char pg_idx:2; + unsigned char color:6; + unsigned char chan_id; + unsigned short npods; + unsigned long caller_data; +}; + +/* per cpu ppm pool */ +struct cxgbi_ppm_pool { + unsigned int base; /* base index */ + unsigned int next; /* next possible free index */ + spinlock_t lock; /* ppm pool lock */ + unsigned long bmap[0]; +} ____cacheline_aligned_in_smp; + +struct cxgbi_ppm { + struct kref refcnt; + struct net_device *ndev; /* net_device, 1st port */ + struct pci_dev *pdev; + void *lldev; + void **ppm_pp; + struct cxgbi_tag_format tformat; + unsigned int ppmax; + unsigned int llimit; + unsigned int base_idx; + + unsigned int pool_rsvd; + unsigned int pool_index_max; + struct cxgbi_ppm_pool __percpu *pool; + /* map lock */ + spinlock_t map_lock; /* ppm map lock */ + unsigned int bmap_index_max; + unsigned int next; + unsigned long *ppod_bmap; + struct cxgbi_ppod_data ppod_data[0]; +}; + +#define DDP_THRESHOLD 512 + +#define PPOD_PAGES_SHIFT 2 /* 4 pages per pod */ + +#define IPPOD_SIZE sizeof(struct cxgbi_pagepod) /* 64 */ +#define PPOD_SIZE_SHIFT 6 + +/* page pods are allocated in groups of this size (must be power of 2) */ +#define PPOD_CLUSTER_SIZE 16U + +#define ULPMEM_DSGL_MAX_NPPODS 16 /* 1024/PPOD_SIZE */ +#define ULPMEM_IDATA_MAX_NPPODS 3 /* (PPOD_SIZE * 3 + ulptx hdr) < 256B */ +#define PCIE_MEMWIN_MAX_NPPODS 16 /* 1024/PPOD_SIZE */ + +#define PPOD_COLOR_SHIFT 0 +#define PPOD_COLOR(x) ((x) << PPOD_COLOR_SHIFT) + +#define PPOD_IDX_SHIFT 6 +#define PPOD_IDX_MAX_SIZE 24 + +#define PPOD_TID_SHIFT 0 +#define PPOD_TID(x) ((x) << PPOD_TID_SHIFT) + +#define PPOD_TAG_SHIFT 6 +#define PPOD_TAG(x) ((x) << PPOD_TAG_SHIFT) + +#define PPOD_VALID_SHIFT 24 +#define PPOD_VALID(x) ((x) << PPOD_VALID_SHIFT) +#define PPOD_VALID_FLAG PPOD_VALID(1U) + +#define PPOD_PI_EXTRACT_CTL_SHIFT 31 +#define PPOD_PI_EXTRACT_CTL(x) ((x) << PPOD_PI_EXTRACT_CTL_SHIFT) +#define PPOD_PI_EXTRACT_CTL_FLAG V_PPOD_PI_EXTRACT_CTL(1U) + +#define PPOD_PI_TYPE_SHIFT 29 +#define PPOD_PI_TYPE_MASK 0x3 +#define PPOD_PI_TYPE(x) ((x) << PPOD_PI_TYPE_SHIFT) + +#define PPOD_PI_CHECK_CTL_SHIFT 27 +#define PPOD_PI_CHECK_CTL_MASK 0x3 +#define PPOD_PI_CHECK_CTL(x) ((x) << PPOD_PI_CHECK_CTL_SHIFT) + +#define PPOD_PI_REPORT_CTL_SHIFT 25 +#define PPOD_PI_REPORT_CTL_MASK 0x3 +#define PPOD_PI_REPORT_CTL(x) ((x) << PPOD_PI_REPORT_CTL_SHIFT) + +static inline int cxgbi_ppm_is_ddp_tag(struct cxgbi_ppm *ppm, u32 tag) +{ + return !(tag & ppm->tformat.no_ddp_mask); +} + +static inline int cxgbi_ppm_sw_tag_is_usable(struct cxgbi_ppm *ppm, + u32 tag) +{ + /* the sw tag must be using <= 31 bits */ + return !(tag & 0x80000000U); +} + +static inline int cxgbi_ppm_make_non_ddp_tag(struct cxgbi_ppm *ppm, + u32 sw_tag, + u32 *final_tag) +{ + struct cxgbi_tag_format *tformat = &ppm->tformat; + + if (!cxgbi_ppm_sw_tag_is_usable(ppm, sw_tag)) { + pr_info("sw_tag 0x%x NOT usable.\n", sw_tag); + return -EINVAL; + } + + if (!sw_tag) { + *final_tag = tformat->no_ddp_mask; + } else { + unsigned int shift = tformat->idx_bits + tformat->color_bits; + u32 lower = sw_tag & tformat->idx_clr_mask; + u32 upper = (sw_tag >> shift) << (shift + 1); + + *final_tag = upper | tformat->no_ddp_mask | lower; + } + return 0; +} + +static inline u32 cxgbi_ppm_decode_non_ddp_tag(struct cxgbi_ppm *ppm, + u32 tag) +{ + struct cxgbi_tag_format *tformat = &ppm->tformat; + unsigned int shift = tformat->idx_bits + tformat->color_bits; + u32 lower = tag & tformat->idx_clr_mask; + u32 upper = (tag >> tformat->rsvd_bits) << shift; + + return upper | lower; +} + +static inline u32 cxgbi_ppm_ddp_tag_get_idx(struct cxgbi_ppm *ppm, + u32 ddp_tag) +{ + u32 hw_idx = (ddp_tag >> PPOD_IDX_SHIFT) & + ppm->tformat.idx_mask; + + return hw_idx - ppm->base_idx; +} + +static inline u32 cxgbi_ppm_make_ddp_tag(unsigned int hw_idx, + unsigned char color) +{ + return (hw_idx << PPOD_IDX_SHIFT) | ((u32)color); +} + +static inline unsigned long +cxgbi_ppm_get_tag_caller_data(struct cxgbi_ppm *ppm, + u32 ddp_tag) +{ + u32 idx = cxgbi_ppm_ddp_tag_get_idx(ppm, ddp_tag); + + return ppm->ppod_data[idx].caller_data; +} + +/* sw bits are the free bits */ +static inline int cxgbi_ppm_ddp_tag_update_sw_bits(struct cxgbi_ppm *ppm, + u32 val, u32 orig_tag, + u32 *final_tag) +{ + struct cxgbi_tag_format *tformat = &ppm->tformat; + u32 v = val >> tformat->free_bits; + + if (v) { + pr_info("sw_bits 0x%x too large, avail bits %u.\n", + val, tformat->free_bits); + return -EINVAL; + } + if (!cxgbi_ppm_is_ddp_tag(ppm, orig_tag)) + return -EINVAL; + + *final_tag = (val << tformat->rsvd_bits) | + (orig_tag & ppm->tformat.rsvd_mask); + return 0; +} + +static inline void cxgbi_ppm_ppod_clear(struct cxgbi_pagepod *ppod) +{ + ppod->hdr.vld_tid = 0U; +} + +static inline void cxgbi_tagmask_check(unsigned int tagmask, + struct cxgbi_tag_format *tformat) +{ + unsigned int bits = fls(tagmask); + + /* reserve top most 2 bits for page selector */ + tformat->free_bits = 32 - 2 - bits; + tformat->rsvd_bits = bits; + tformat->color_bits = PPOD_IDX_SHIFT; + tformat->idx_bits = bits - 1 - PPOD_IDX_SHIFT; + tformat->no_ddp_mask = 1 << (bits - 1); + tformat->idx_mask = (1 << tformat->idx_bits) - 1; + tformat->color_mask = (1 << PPOD_IDX_SHIFT) - 1; + tformat->idx_clr_mask = (1 << (bits - 1)) - 1; + tformat->rsvd_mask = (1 << bits) - 1; + + pr_info("ippm: tagmask 0x%x, rsvd %u=%u+%u+1, mask 0x%x,0x%x, " + "pg %u,%u,%u,%u.\n", + tagmask, tformat->rsvd_bits, tformat->idx_bits, + tformat->color_bits, tformat->no_ddp_mask, tformat->rsvd_mask, + tformat->pgsz_order[0], tformat->pgsz_order[1], + tformat->pgsz_order[2], tformat->pgsz_order[3]); +} + +int cxgbi_ppm_find_page_index(struct cxgbi_ppm *ppm, unsigned long pgsz); +void cxgbi_ppm_make_ppod_hdr(struct cxgbi_ppm *ppm, u32 tag, + unsigned int tid, unsigned int offset, + unsigned int length, + struct cxgbi_pagepod_hdr *hdr); +void cxgbi_ppm_ppod_release(struct cxgbi_ppm *, u32 idx); +int cxgbi_ppm_ppods_reserve(struct cxgbi_ppm *, unsigned short nr_pages, + u32 per_tag_pg_idx, u32 *ppod_idx, u32 *ddp_tag, + unsigned long caller_data); +int cxgbi_ppm_init(void **ppm_pp, struct net_device *, struct pci_dev *, + void *lldev, struct cxgbi_tag_format *, + unsigned int ppmax, unsigned int llimit, + unsigned int start, + unsigned int reserve_factor); +int cxgbi_ppm_release(struct cxgbi_ppm *ppm); +void cxgbi_tagmask_check(unsigned int tagmask, struct cxgbi_tag_format *); +unsigned int cxgbi_tagmask_set(unsigned int ppmax); + +#endif /*__LIBCXGB_PPM_H__*/ diff --git a/drivers/target/iscsi/cxgbit/Kconfig b/drivers/target/iscsi/cxgbit/Kconfig index c9b6a3c..bc6c1d5 100644 --- a/drivers/target/iscsi/cxgbit/Kconfig +++ b/drivers/target/iscsi/cxgbit/Kconfig @@ -1,7 +1,7 @@ config ISCSI_TARGET_CXGB4 tristate "Chelsio iSCSI target offload driver" depends on ISCSI_TARGET && CHELSIO_T4 && INET - select CHELSIO_T4_UWIRE + select CHELSIO_LIB ---help--- To compile this driver as module, choose M here: the module will be called cxgbit. diff --git a/drivers/target/iscsi/cxgbit/Makefile b/drivers/target/iscsi/cxgbit/Makefile index bd56c07..4893ec2 100644 --- a/drivers/target/iscsi/cxgbit/Makefile +++ b/drivers/target/iscsi/cxgbit/Makefile @@ -1,4 +1,5 @@ ccflags-y := -Idrivers/net/ethernet/chelsio/cxgb4 +ccflags-y += -Idrivers/net/ethernet/chelsio/libcxgb ccflags-y += -Idrivers/target/iscsi obj-$(CONFIG_ISCSI_TARGET_CXGB4) += cxgbit.o diff --git a/drivers/target/iscsi/cxgbit/cxgbit.h b/drivers/target/iscsi/cxgbit/cxgbit.h index 625c7f6..9038869 100644 --- a/drivers/target/iscsi/cxgbit/cxgbit.h +++ b/drivers/target/iscsi/cxgbit/cxgbit.h @@ -37,7 +37,7 @@ #include "cxgb4.h" #include "cxgb4_uld.h" #include "l2t.h" -#include "cxgb4_ppm.h" +#include "libcxgb_ppm.h" #include "cxgbit_lro.h" extern struct mutex cdev_list_lock; -- cgit v0.10.2 From 5999299f1ce9e8610cb8263953f5767d4f840a3e Mon Sep 17 00:00:00 2001 From: Varun Prakash Date: Thu, 21 Jul 2016 22:57:15 +0530 Subject: cxgb3i,cxgb4i,libcxgbi: remove iSCSI DDP support Remove old ddp code from cxgb3i,cxgb4i,libcxgbi. Next two commits adds DDP support using common iSCSI DDP Page Pod Manager. Signed-off-by: Varun Prakash Signed-off-by: David S. Miller diff --git a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c index e22a268..fda0234 100644 --- a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c +++ b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c @@ -1076,65 +1076,6 @@ static inline void ulp_mem_io_set_hdr(struct sk_buff *skb, unsigned int addr) req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_BYPASS)); req->cmd_lock_addr = htonl(V_ULP_MEMIO_ADDR(addr >> 5) | V_ULPTX_CMD(ULP_MEM_WRITE)); - req->len = htonl(V_ULP_MEMIO_DATA_LEN(PPOD_SIZE >> 5) | - V_ULPTX_NFLITS((PPOD_SIZE >> 3) + 1)); -} - -static int ddp_set_map(struct cxgbi_sock *csk, struct cxgbi_pagepod_hdr *hdr, - unsigned int idx, unsigned int npods, - struct cxgbi_gather_list *gl) -{ - struct cxgbi_device *cdev = csk->cdev; - struct cxgbi_ddp_info *ddp = cdev->ddp; - unsigned int pm_addr = (idx << PPOD_SIZE_SHIFT) + ddp->llimit; - int i; - - log_debug(1 << CXGBI_DBG_DDP, - "csk 0x%p, idx %u, npods %u, gl 0x%p.\n", - csk, idx, npods, gl); - - for (i = 0; i < npods; i++, idx++, pm_addr += PPOD_SIZE) { - struct sk_buff *skb = alloc_wr(sizeof(struct ulp_mem_io) + - PPOD_SIZE, 0, GFP_ATOMIC); - - if (!skb) - return -ENOMEM; - - ulp_mem_io_set_hdr(skb, pm_addr); - cxgbi_ddp_ppod_set((struct cxgbi_pagepod *)(skb->head + - sizeof(struct ulp_mem_io)), - hdr, gl, i * PPOD_PAGES_MAX); - skb->priority = CPL_PRIORITY_CONTROL; - cxgb3_ofld_send(cdev->lldev, skb); - } - return 0; -} - -static void ddp_clear_map(struct cxgbi_hba *chba, unsigned int tag, - unsigned int idx, unsigned int npods) -{ - struct cxgbi_device *cdev = chba->cdev; - struct cxgbi_ddp_info *ddp = cdev->ddp; - unsigned int pm_addr = (idx << PPOD_SIZE_SHIFT) + ddp->llimit; - int i; - - log_debug(1 << CXGBI_DBG_DDP, - "cdev 0x%p, idx %u, npods %u, tag 0x%x.\n", - cdev, idx, npods, tag); - - for (i = 0; i < npods; i++, idx++, pm_addr += PPOD_SIZE) { - struct sk_buff *skb = alloc_wr(sizeof(struct ulp_mem_io) + - PPOD_SIZE, 0, GFP_ATOMIC); - - if (!skb) { - pr_err("tag 0x%x, 0x%x, %d/%u, skb OOM.\n", - tag, idx, i, npods); - continue; - } - ulp_mem_io_set_hdr(skb, pm_addr); - skb->priority = CPL_PRIORITY_CONTROL; - cxgb3_ofld_send(cdev->lldev, skb); - } } static int ddp_setup_conn_pgidx(struct cxgbi_sock *csk, @@ -1203,82 +1144,14 @@ static int ddp_setup_conn_digest(struct cxgbi_sock *csk, unsigned int tid, } /** - * t3_ddp_cleanup - release the cxgb3 adapter's ddp resource - * @cdev: cxgb3i adapter - * release all the resource held by the ddp pagepod manager for a given - * adapter if needed - */ - -static void t3_ddp_cleanup(struct cxgbi_device *cdev) -{ - struct t3cdev *tdev = (struct t3cdev *)cdev->lldev; - - if (cxgbi_ddp_cleanup(cdev)) { - pr_info("t3dev 0x%p, ulp_iscsi no more user.\n", tdev); - tdev->ulp_iscsi = NULL; - } -} - -/** * ddp_init - initialize the cxgb3 adapter's ddp resource * @cdev: cxgb3i adapter * initialize the ddp pagepod manager for a given adapter */ static int cxgb3i_ddp_init(struct cxgbi_device *cdev) { - struct t3cdev *tdev = (struct t3cdev *)cdev->lldev; - struct cxgbi_ddp_info *ddp = tdev->ulp_iscsi; - struct ulp_iscsi_info uinfo; - unsigned int pgsz_factor[4]; - int i, err; - - if (ddp) { - kref_get(&ddp->refcnt); - pr_warn("t3dev 0x%p, ddp 0x%p already set up.\n", - tdev, tdev->ulp_iscsi); - cdev->ddp = ddp; - return -EALREADY; - } - - err = tdev->ctl(tdev, ULP_ISCSI_GET_PARAMS, &uinfo); - if (err < 0) { - pr_err("%s, failed to get iscsi param err=%d.\n", - tdev->name, err); - return err; - } - - err = cxgbi_ddp_init(cdev, uinfo.llimit, uinfo.ulimit, - uinfo.max_txsz, uinfo.max_rxsz); - if (err < 0) - return err; - - ddp = cdev->ddp; - - uinfo.tagmask = ddp->idx_mask << PPOD_IDX_SHIFT; - cxgbi_ddp_page_size_factor(pgsz_factor); - for (i = 0; i < 4; i++) - uinfo.pgsz_factor[i] = pgsz_factor[i]; - uinfo.ulimit = uinfo.llimit + (ddp->nppods << PPOD_SIZE_SHIFT); - - err = tdev->ctl(tdev, ULP_ISCSI_SET_PARAMS, &uinfo); - if (err < 0) { - pr_warn("%s unable to set iscsi param err=%d, ddp disabled.\n", - tdev->name, err); - cxgbi_ddp_cleanup(cdev); - return err; - } - tdev->ulp_iscsi = ddp; - cdev->csk_ddp_setup_digest = ddp_setup_conn_digest; cdev->csk_ddp_setup_pgidx = ddp_setup_conn_pgidx; - cdev->csk_ddp_set = ddp_set_map; - cdev->csk_ddp_clear = ddp_clear_map; - - pr_info("tdev 0x%p, nppods %u, bits %u, mask 0x%x,0x%x pkt %u/%u, " - "%u/%u.\n", - tdev, ddp->nppods, ddp->idx_bits, ddp->idx_mask, - ddp->rsvd_tag_mask, ddp->max_txsz, uinfo.max_txsz, - ddp->max_rxsz, uinfo.max_rxsz); return 0; } @@ -1325,7 +1198,6 @@ static void cxgb3i_dev_open(struct t3cdev *t3dev) cdev->rx_credit_thres = cxgb3i_rx_credit_thres; cdev->skb_tx_rsvd = CXGB3I_TX_HEADER_LEN; cdev->skb_rx_extra = sizeof(struct cpl_iscsi_hdr_norss); - cdev->dev_ddp_cleanup = t3_ddp_cleanup; cdev->itp = &cxgb3i_iscsi_transport; err = cxgb3i_ddp_init(cdev); diff --git a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c index 339f6b7..2911214 100644 --- a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c +++ b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c @@ -1543,110 +1543,6 @@ int cxgb4i_ofld_init(struct cxgbi_device *cdev) return 0; } -/* - * functions to program the pagepod in h/w - */ -#define ULPMEM_IDATA_MAX_NPPODS 4 /* 256/PPOD_SIZE */ -static inline void ulp_mem_io_set_hdr(struct cxgb4_lld_info *lldi, - struct ulp_mem_io *req, - unsigned int wr_len, unsigned int dlen, - unsigned int pm_addr) -{ - struct ulptx_idata *idata = (struct ulptx_idata *)(req + 1); - - INIT_ULPTX_WR(req, wr_len, 0, 0); - if (is_t4(lldi->adapter_type)) - req->cmd = htonl(ULPTX_CMD_V(ULP_TX_MEM_WRITE) | - (ULP_MEMIO_ORDER_F)); - else - req->cmd = htonl(ULPTX_CMD_V(ULP_TX_MEM_WRITE) | - (T5_ULP_MEMIO_IMM_F)); - req->dlen = htonl(ULP_MEMIO_DATA_LEN_V(dlen >> 5)); - req->lock_addr = htonl(ULP_MEMIO_ADDR_V(pm_addr >> 5)); - req->len16 = htonl(DIV_ROUND_UP(wr_len - sizeof(req->wr), 16)); - - idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM)); - idata->len = htonl(dlen); -} - -static int ddp_ppod_write_idata(struct cxgbi_device *cdev, unsigned int port_id, - struct cxgbi_pagepod_hdr *hdr, unsigned int idx, - unsigned int npods, - struct cxgbi_gather_list *gl, - unsigned int gl_pidx) -{ - struct cxgbi_ddp_info *ddp = cdev->ddp; - struct cxgb4_lld_info *lldi = cxgbi_cdev_priv(cdev); - struct sk_buff *skb; - struct ulp_mem_io *req; - struct ulptx_idata *idata; - struct cxgbi_pagepod *ppod; - unsigned int pm_addr = idx * PPOD_SIZE + ddp->llimit; - unsigned int dlen = PPOD_SIZE * npods; - unsigned int wr_len = roundup(sizeof(struct ulp_mem_io) + - sizeof(struct ulptx_idata) + dlen, 16); - unsigned int i; - - skb = alloc_wr(wr_len, 0, GFP_ATOMIC); - if (!skb) { - pr_err("cdev 0x%p, idx %u, npods %u, OOM.\n", - cdev, idx, npods); - return -ENOMEM; - } - req = (struct ulp_mem_io *)skb->head; - set_wr_txq(skb, CPL_PRIORITY_CONTROL, 0); - - ulp_mem_io_set_hdr(lldi, req, wr_len, dlen, pm_addr); - idata = (struct ulptx_idata *)(req + 1); - ppod = (struct cxgbi_pagepod *)(idata + 1); - - for (i = 0; i < npods; i++, ppod++, gl_pidx += PPOD_PAGES_MAX) { - if (!hdr && !gl) - cxgbi_ddp_ppod_clear(ppod); - else - cxgbi_ddp_ppod_set(ppod, hdr, gl, gl_pidx); - } - - cxgb4_ofld_send(cdev->ports[port_id], skb); - return 0; -} - -static int ddp_set_map(struct cxgbi_sock *csk, struct cxgbi_pagepod_hdr *hdr, - unsigned int idx, unsigned int npods, - struct cxgbi_gather_list *gl) -{ - unsigned int i, cnt; - int err = 0; - - for (i = 0; i < npods; i += cnt, idx += cnt) { - cnt = npods - i; - if (cnt > ULPMEM_IDATA_MAX_NPPODS) - cnt = ULPMEM_IDATA_MAX_NPPODS; - err = ddp_ppod_write_idata(csk->cdev, csk->port_id, hdr, - idx, cnt, gl, 4 * i); - if (err < 0) - break; - } - return err; -} - -static void ddp_clear_map(struct cxgbi_hba *chba, unsigned int tag, - unsigned int idx, unsigned int npods) -{ - unsigned int i, cnt; - int err; - - for (i = 0; i < npods; i += cnt, idx += cnt) { - cnt = npods - i; - if (cnt > ULPMEM_IDATA_MAX_NPPODS) - cnt = ULPMEM_IDATA_MAX_NPPODS; - err = ddp_ppod_write_idata(chba->cdev, chba->port_id, NULL, - idx, cnt, NULL, 0); - if (err < 0) - break; - } -} - static int ddp_setup_conn_pgidx(struct cxgbi_sock *csk, unsigned int tid, int pg_idx, bool reply) { @@ -1712,46 +1608,8 @@ static int ddp_setup_conn_digest(struct cxgbi_sock *csk, unsigned int tid, static int cxgb4i_ddp_init(struct cxgbi_device *cdev) { - struct cxgb4_lld_info *lldi = cxgbi_cdev_priv(cdev); - struct cxgbi_ddp_info *ddp = cdev->ddp; - unsigned int tagmask, pgsz_factor[4]; - int err; - - if (ddp) { - kref_get(&ddp->refcnt); - pr_warn("cdev 0x%p, ddp 0x%p already set up.\n", - cdev, cdev->ddp); - return -EALREADY; - } - - err = cxgbi_ddp_init(cdev, lldi->vr->iscsi.start, - lldi->vr->iscsi.start + lldi->vr->iscsi.size - 1, - lldi->iscsi_iolen, lldi->iscsi_iolen); - if (err < 0) - return err; - - ddp = cdev->ddp; - - tagmask = ddp->idx_mask << PPOD_IDX_SHIFT; - cxgbi_ddp_page_size_factor(pgsz_factor); - cxgb4_iscsi_init(lldi->ports[0], tagmask, pgsz_factor); - cdev->csk_ddp_setup_digest = ddp_setup_conn_digest; cdev->csk_ddp_setup_pgidx = ddp_setup_conn_pgidx; - cdev->csk_ddp_set = ddp_set_map; - cdev->csk_ddp_clear = ddp_clear_map; - - pr_info("cxgb4i 0x%p tag: sw %u, rsvd %u,%u, mask 0x%x.\n", - cdev, cdev->tag_format.sw_bits, cdev->tag_format.rsvd_bits, - cdev->tag_format.rsvd_shift, cdev->tag_format.rsvd_mask); - pr_info("cxgb4i 0x%p, nppods %u, bits %u, mask 0x%x,0x%x pkt %u/%u, " - " %u/%u.\n", - cdev, ddp->nppods, ddp->idx_bits, ddp->idx_mask, - ddp->rsvd_tag_mask, ddp->max_txsz, lldi->iscsi_iolen, - ddp->max_rxsz, lldi->iscsi_iolen); - pr_info("cxgb4i 0x%p max payload size: %u/%u, %u/%u.\n", - cdev, cdev->tx_max_size, ddp->max_txsz, cdev->rx_max_size, - ddp->max_rxsz); return 0; } diff --git a/drivers/scsi/cxgbi/libcxgbi.c b/drivers/scsi/cxgbi/libcxgbi.c index ead83a2..f389d34 100644 --- a/drivers/scsi/cxgbi/libcxgbi.c +++ b/drivers/scsi/cxgbi/libcxgbi.c @@ -113,12 +113,6 @@ static inline void cxgbi_device_destroy(struct cxgbi_device *cdev) "cdev 0x%p, p# %u.\n", cdev, cdev->nports); cxgbi_hbas_remove(cdev); cxgbi_device_portmap_cleanup(cdev); - if (cdev->dev_ddp_cleanup) - cdev->dev_ddp_cleanup(cdev); - else - cxgbi_ddp_cleanup(cdev); - if (cdev->ddp) - cxgbi_ddp_cleanup(cdev); if (cdev->pmap.max_connect) cxgbi_free_big_mem(cdev->pmap.port_csk); kfree(cdev); @@ -1183,504 +1177,6 @@ out_err: } /* - * Direct Data Placement - - * Directly place the iSCSI Data-In or Data-Out PDU's payload into pre-posted - * final destination host-memory buffers based on the Initiator Task Tag (ITT) - * in Data-In or Target Task Tag (TTT) in Data-Out PDUs. - * The host memory address is programmed into h/w in the format of pagepod - * entries. - * The location of the pagepod entry is encoded into ddp tag which is used as - * the base for ITT/TTT. - */ - -static unsigned char ddp_page_order[DDP_PGIDX_MAX] = {0, 1, 2, 4}; -static unsigned char ddp_page_shift[DDP_PGIDX_MAX] = {12, 13, 14, 16}; -static unsigned char page_idx = DDP_PGIDX_MAX; - -static unsigned char sw_tag_idx_bits; -static unsigned char sw_tag_age_bits; - -/* - * Direct-Data Placement page size adjustment - */ -static int ddp_adjust_page_table(void) -{ - int i; - unsigned int base_order, order; - - if (PAGE_SIZE < (1UL << ddp_page_shift[0])) { - pr_info("PAGE_SIZE 0x%lx too small, min 0x%lx\n", - PAGE_SIZE, 1UL << ddp_page_shift[0]); - return -EINVAL; - } - - base_order = get_order(1UL << ddp_page_shift[0]); - order = get_order(1UL << PAGE_SHIFT); - - for (i = 0; i < DDP_PGIDX_MAX; i++) { - /* first is the kernel page size, then just doubling */ - ddp_page_order[i] = order - base_order + i; - ddp_page_shift[i] = PAGE_SHIFT + i; - } - return 0; -} - -static int ddp_find_page_index(unsigned long pgsz) -{ - int i; - - for (i = 0; i < DDP_PGIDX_MAX; i++) { - if (pgsz == (1UL << ddp_page_shift[i])) - return i; - } - pr_info("ddp page size %lu not supported.\n", pgsz); - return DDP_PGIDX_MAX; -} - -static void ddp_setup_host_page_size(void) -{ - if (page_idx == DDP_PGIDX_MAX) { - page_idx = ddp_find_page_index(PAGE_SIZE); - - if (page_idx == DDP_PGIDX_MAX) { - pr_info("system PAGE %lu, update hw.\n", PAGE_SIZE); - if (ddp_adjust_page_table() < 0) { - pr_info("PAGE %lu, disable ddp.\n", PAGE_SIZE); - return; - } - page_idx = ddp_find_page_index(PAGE_SIZE); - } - pr_info("system PAGE %lu, ddp idx %u.\n", PAGE_SIZE, page_idx); - } -} - -void cxgbi_ddp_page_size_factor(int *pgsz_factor) -{ - int i; - - for (i = 0; i < DDP_PGIDX_MAX; i++) - pgsz_factor[i] = ddp_page_order[i]; -} -EXPORT_SYMBOL_GPL(cxgbi_ddp_page_size_factor); - -/* - * DDP setup & teardown - */ - -void cxgbi_ddp_ppod_set(struct cxgbi_pagepod *ppod, - struct cxgbi_pagepod_hdr *hdr, - struct cxgbi_gather_list *gl, unsigned int gidx) -{ - int i; - - memcpy(ppod, hdr, sizeof(*hdr)); - for (i = 0; i < (PPOD_PAGES_MAX + 1); i++, gidx++) { - ppod->addr[i] = gidx < gl->nelem ? - cpu_to_be64(gl->phys_addr[gidx]) : 0ULL; - } -} -EXPORT_SYMBOL_GPL(cxgbi_ddp_ppod_set); - -void cxgbi_ddp_ppod_clear(struct cxgbi_pagepod *ppod) -{ - memset(ppod, 0, sizeof(*ppod)); -} -EXPORT_SYMBOL_GPL(cxgbi_ddp_ppod_clear); - -static inline int ddp_find_unused_entries(struct cxgbi_ddp_info *ddp, - unsigned int start, unsigned int max, - unsigned int count, - struct cxgbi_gather_list *gl) -{ - unsigned int i, j, k; - - /* not enough entries */ - if ((max - start) < count) { - log_debug(1 << CXGBI_DBG_DDP, - "NOT enough entries %u+%u < %u.\n", start, count, max); - return -EBUSY; - } - - max -= count; - spin_lock(&ddp->map_lock); - for (i = start; i < max;) { - for (j = 0, k = i; j < count; j++, k++) { - if (ddp->gl_map[k]) - break; - } - if (j == count) { - for (j = 0, k = i; j < count; j++, k++) - ddp->gl_map[k] = gl; - spin_unlock(&ddp->map_lock); - return i; - } - i += j + 1; - } - spin_unlock(&ddp->map_lock); - log_debug(1 << CXGBI_DBG_DDP, - "NO suitable entries %u available.\n", count); - return -EBUSY; -} - -static inline void ddp_unmark_entries(struct cxgbi_ddp_info *ddp, - int start, int count) -{ - spin_lock(&ddp->map_lock); - memset(&ddp->gl_map[start], 0, - count * sizeof(struct cxgbi_gather_list *)); - spin_unlock(&ddp->map_lock); -} - -static inline void ddp_gl_unmap(struct pci_dev *pdev, - struct cxgbi_gather_list *gl) -{ - int i; - - for (i = 0; i < gl->nelem; i++) - dma_unmap_page(&pdev->dev, gl->phys_addr[i], PAGE_SIZE, - PCI_DMA_FROMDEVICE); -} - -static inline int ddp_gl_map(struct pci_dev *pdev, - struct cxgbi_gather_list *gl) -{ - int i; - - for (i = 0; i < gl->nelem; i++) { - gl->phys_addr[i] = dma_map_page(&pdev->dev, gl->pages[i], 0, - PAGE_SIZE, - PCI_DMA_FROMDEVICE); - if (unlikely(dma_mapping_error(&pdev->dev, gl->phys_addr[i]))) { - log_debug(1 << CXGBI_DBG_DDP, - "page %d 0x%p, 0x%p dma mapping err.\n", - i, gl->pages[i], pdev); - goto unmap; - } - } - return i; -unmap: - if (i) { - unsigned int nelem = gl->nelem; - - gl->nelem = i; - ddp_gl_unmap(pdev, gl); - gl->nelem = nelem; - } - return -EINVAL; -} - -static void ddp_release_gl(struct cxgbi_gather_list *gl, - struct pci_dev *pdev) -{ - ddp_gl_unmap(pdev, gl); - kfree(gl); -} - -static struct cxgbi_gather_list *ddp_make_gl(unsigned int xferlen, - struct scatterlist *sgl, - unsigned int sgcnt, - struct pci_dev *pdev, - gfp_t gfp) -{ - struct cxgbi_gather_list *gl; - struct scatterlist *sg = sgl; - struct page *sgpage = sg_page(sg); - unsigned int sglen = sg->length; - unsigned int sgoffset = sg->offset; - unsigned int npages = (xferlen + sgoffset + PAGE_SIZE - 1) >> - PAGE_SHIFT; - int i = 1, j = 0; - - if (xferlen < DDP_THRESHOLD) { - log_debug(1 << CXGBI_DBG_DDP, - "xfer %u < threshold %u, no ddp.\n", - xferlen, DDP_THRESHOLD); - return NULL; - } - - gl = kzalloc(sizeof(struct cxgbi_gather_list) + - npages * (sizeof(dma_addr_t) + - sizeof(struct page *)), gfp); - if (!gl) { - log_debug(1 << CXGBI_DBG_DDP, - "xfer %u, %u pages, OOM.\n", xferlen, npages); - return NULL; - } - - log_debug(1 << CXGBI_DBG_DDP, - "xfer %u, sgl %u, gl max %u.\n", xferlen, sgcnt, npages); - - gl->pages = (struct page **)&gl->phys_addr[npages]; - gl->nelem = npages; - gl->length = xferlen; - gl->offset = sgoffset; - gl->pages[0] = sgpage; - - for (i = 1, sg = sg_next(sgl), j = 0; i < sgcnt; - i++, sg = sg_next(sg)) { - struct page *page = sg_page(sg); - - if (sgpage == page && sg->offset == sgoffset + sglen) - sglen += sg->length; - else { - /* make sure the sgl is fit for ddp: - * each has the same page size, and - * all of the middle pages are used completely - */ - if ((j && sgoffset) || ((i != sgcnt - 1) && - ((sglen + sgoffset) & ~PAGE_MASK))) { - log_debug(1 << CXGBI_DBG_DDP, - "page %d/%u, %u + %u.\n", - i, sgcnt, sgoffset, sglen); - goto error_out; - } - - j++; - if (j == gl->nelem || sg->offset) { - log_debug(1 << CXGBI_DBG_DDP, - "page %d/%u, offset %u.\n", - j, gl->nelem, sg->offset); - goto error_out; - } - gl->pages[j] = page; - sglen = sg->length; - sgoffset = sg->offset; - sgpage = page; - } - } - gl->nelem = ++j; - - if (ddp_gl_map(pdev, gl) < 0) - goto error_out; - - return gl; - -error_out: - kfree(gl); - return NULL; -} - -static void ddp_tag_release(struct cxgbi_hba *chba, u32 tag) -{ - struct cxgbi_device *cdev = chba->cdev; - struct cxgbi_ddp_info *ddp = cdev->ddp; - u32 idx; - - idx = (tag >> PPOD_IDX_SHIFT) & ddp->idx_mask; - if (idx < ddp->nppods) { - struct cxgbi_gather_list *gl = ddp->gl_map[idx]; - unsigned int npods; - - if (!gl || !gl->nelem) { - pr_warn("tag 0x%x, idx %u, gl 0x%p, %u.\n", - tag, idx, gl, gl ? gl->nelem : 0); - return; - } - npods = (gl->nelem + PPOD_PAGES_MAX - 1) >> PPOD_PAGES_SHIFT; - log_debug(1 << CXGBI_DBG_DDP, - "tag 0x%x, release idx %u, npods %u.\n", - tag, idx, npods); - cdev->csk_ddp_clear(chba, tag, idx, npods); - ddp_unmark_entries(ddp, idx, npods); - ddp_release_gl(gl, ddp->pdev); - } else - pr_warn("tag 0x%x, idx %u > max %u.\n", tag, idx, ddp->nppods); -} - -static int ddp_tag_reserve(struct cxgbi_sock *csk, unsigned int tid, - u32 sw_tag, u32 *tagp, struct cxgbi_gather_list *gl, - gfp_t gfp) -{ - struct cxgbi_device *cdev = csk->cdev; - struct cxgbi_ddp_info *ddp = cdev->ddp; - struct cxgbi_tag_format *tformat = &cdev->tag_format; - struct cxgbi_pagepod_hdr hdr; - unsigned int npods; - int idx = -1; - int err = -ENOMEM; - u32 tag; - - npods = (gl->nelem + PPOD_PAGES_MAX - 1) >> PPOD_PAGES_SHIFT; - if (ddp->idx_last == ddp->nppods) - idx = ddp_find_unused_entries(ddp, 0, ddp->nppods, - npods, gl); - else { - idx = ddp_find_unused_entries(ddp, ddp->idx_last + 1, - ddp->nppods, npods, - gl); - if (idx < 0 && ddp->idx_last >= npods) { - idx = ddp_find_unused_entries(ddp, 0, - min(ddp->idx_last + npods, ddp->nppods), - npods, gl); - } - } - if (idx < 0) { - log_debug(1 << CXGBI_DBG_DDP, - "xferlen %u, gl %u, npods %u NO DDP.\n", - gl->length, gl->nelem, npods); - return idx; - } - - tag = cxgbi_ddp_tag_base(tformat, sw_tag); - tag |= idx << PPOD_IDX_SHIFT; - - hdr.rsvd = 0; - hdr.vld_tid = htonl(PPOD_VALID_FLAG | PPOD_TID(tid)); - hdr.pgsz_tag_clr = htonl(tag & ddp->rsvd_tag_mask); - hdr.max_offset = htonl(gl->length); - hdr.page_offset = htonl(gl->offset); - - err = cdev->csk_ddp_set(csk, &hdr, idx, npods, gl); - if (err < 0) - goto unmark_entries; - - ddp->idx_last = idx; - log_debug(1 << CXGBI_DBG_DDP, - "xfer %u, gl %u,%u, tid 0x%x, tag 0x%x->0x%x(%u,%u).\n", - gl->length, gl->nelem, gl->offset, tid, sw_tag, tag, idx, - npods); - *tagp = tag; - return 0; - -unmark_entries: - ddp_unmark_entries(ddp, idx, npods); - return err; -} - -int cxgbi_ddp_reserve(struct cxgbi_sock *csk, unsigned int *tagp, - unsigned int sw_tag, unsigned int xferlen, - struct scatterlist *sgl, unsigned int sgcnt, gfp_t gfp) -{ - struct cxgbi_device *cdev = csk->cdev; - struct cxgbi_tag_format *tformat = &cdev->tag_format; - struct cxgbi_gather_list *gl; - int err; - - if (page_idx >= DDP_PGIDX_MAX || !cdev->ddp || - xferlen < DDP_THRESHOLD) { - log_debug(1 << CXGBI_DBG_DDP, - "pgidx %u, xfer %u, NO ddp.\n", page_idx, xferlen); - return -EINVAL; - } - - if (!cxgbi_sw_tag_usable(tformat, sw_tag)) { - log_debug(1 << CXGBI_DBG_DDP, - "sw_tag 0x%x NOT usable.\n", sw_tag); - return -EINVAL; - } - - gl = ddp_make_gl(xferlen, sgl, sgcnt, cdev->pdev, gfp); - if (!gl) - return -ENOMEM; - - err = ddp_tag_reserve(csk, csk->tid, sw_tag, tagp, gl, gfp); - if (err < 0) - ddp_release_gl(gl, cdev->pdev); - - return err; -} - -static void ddp_destroy(struct kref *kref) -{ - struct cxgbi_ddp_info *ddp = container_of(kref, - struct cxgbi_ddp_info, - refcnt); - struct cxgbi_device *cdev = ddp->cdev; - int i = 0; - - pr_info("kref 0, destroy ddp 0x%p, cdev 0x%p.\n", ddp, cdev); - - while (i < ddp->nppods) { - struct cxgbi_gather_list *gl = ddp->gl_map[i]; - - if (gl) { - int npods = (gl->nelem + PPOD_PAGES_MAX - 1) - >> PPOD_PAGES_SHIFT; - pr_info("cdev 0x%p, ddp %d + %d.\n", cdev, i, npods); - kfree(gl); - i += npods; - } else - i++; - } - cxgbi_free_big_mem(ddp); -} - -int cxgbi_ddp_cleanup(struct cxgbi_device *cdev) -{ - struct cxgbi_ddp_info *ddp = cdev->ddp; - - log_debug(1 << CXGBI_DBG_DDP, - "cdev 0x%p, release ddp 0x%p.\n", cdev, ddp); - cdev->ddp = NULL; - if (ddp) - return kref_put(&ddp->refcnt, ddp_destroy); - return 0; -} -EXPORT_SYMBOL_GPL(cxgbi_ddp_cleanup); - -int cxgbi_ddp_init(struct cxgbi_device *cdev, - unsigned int llimit, unsigned int ulimit, - unsigned int max_txsz, unsigned int max_rxsz) -{ - struct cxgbi_ddp_info *ddp; - unsigned int ppmax, bits; - - ppmax = (ulimit - llimit + 1) >> PPOD_SIZE_SHIFT; - bits = __ilog2_u32(ppmax) + 1; - if (bits > PPOD_IDX_MAX_SIZE) - bits = PPOD_IDX_MAX_SIZE; - ppmax = (1 << (bits - 1)) - 1; - - ddp = cxgbi_alloc_big_mem(sizeof(struct cxgbi_ddp_info) + - ppmax * (sizeof(struct cxgbi_gather_list *) + - sizeof(struct sk_buff *)), - GFP_KERNEL); - if (!ddp) { - pr_warn("cdev 0x%p, ddp ppmax %u OOM.\n", cdev, ppmax); - return -ENOMEM; - } - ddp->gl_map = (struct cxgbi_gather_list **)(ddp + 1); - cdev->ddp = ddp; - - spin_lock_init(&ddp->map_lock); - kref_init(&ddp->refcnt); - - ddp->cdev = cdev; - ddp->pdev = cdev->pdev; - ddp->llimit = llimit; - ddp->ulimit = ulimit; - ddp->max_txsz = min_t(unsigned int, max_txsz, ULP2_MAX_PKT_SIZE); - ddp->max_rxsz = min_t(unsigned int, max_rxsz, ULP2_MAX_PKT_SIZE); - ddp->nppods = ppmax; - ddp->idx_last = ppmax; - ddp->idx_bits = bits; - ddp->idx_mask = (1 << bits) - 1; - ddp->rsvd_tag_mask = (1 << (bits + PPOD_IDX_SHIFT)) - 1; - - cdev->tag_format.sw_bits = sw_tag_idx_bits + sw_tag_age_bits; - cdev->tag_format.rsvd_bits = ddp->idx_bits; - cdev->tag_format.rsvd_shift = PPOD_IDX_SHIFT; - cdev->tag_format.rsvd_mask = (1 << cdev->tag_format.rsvd_bits) - 1; - - pr_info("%s tag format, sw %u, rsvd %u,%u, mask 0x%x.\n", - cdev->ports[0]->name, cdev->tag_format.sw_bits, - cdev->tag_format.rsvd_bits, cdev->tag_format.rsvd_shift, - cdev->tag_format.rsvd_mask); - - cdev->tx_max_size = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD, - ddp->max_txsz - ISCSI_PDU_NONPAYLOAD_LEN); - cdev->rx_max_size = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD, - ddp->max_rxsz - ISCSI_PDU_NONPAYLOAD_LEN); - - log_debug(1 << CXGBI_DBG_DDP, - "%s max payload size: %u/%u, %u/%u.\n", - cdev->ports[0]->name, cdev->tx_max_size, ddp->max_txsz, - cdev->rx_max_size, ddp->max_rxsz); - return 0; -} -EXPORT_SYMBOL_GPL(cxgbi_ddp_init); - -/* * APIs interacting with open-iscsi libraries */ @@ -1688,77 +1184,15 @@ static unsigned char padding[4]; static void task_release_itt(struct iscsi_task *task, itt_t hdr_itt) { - struct scsi_cmnd *sc = task->sc; - struct iscsi_tcp_conn *tcp_conn = task->conn->dd_data; - struct cxgbi_conn *cconn = tcp_conn->dd_data; - struct cxgbi_hba *chba = cconn->chba; - struct cxgbi_tag_format *tformat = &chba->cdev->tag_format; - u32 tag = ntohl((__force u32)hdr_itt); - - log_debug(1 << CXGBI_DBG_DDP, - "cdev 0x%p, release tag 0x%x.\n", chba->cdev, tag); - if (sc && - (scsi_bidi_cmnd(sc) || sc->sc_data_direction == DMA_FROM_DEVICE) && - cxgbi_is_ddp_tag(tformat, tag)) - ddp_tag_release(chba, tag); } static int task_reserve_itt(struct iscsi_task *task, itt_t *hdr_itt) { - struct scsi_cmnd *sc = task->sc; - struct iscsi_conn *conn = task->conn; - struct iscsi_session *sess = conn->session; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct cxgbi_conn *cconn = tcp_conn->dd_data; - struct cxgbi_hba *chba = cconn->chba; - struct cxgbi_tag_format *tformat = &chba->cdev->tag_format; - u32 sw_tag = (sess->age << cconn->task_idx_bits) | task->itt; - u32 tag = 0; - int err = -EINVAL; - - if (sc && - (scsi_bidi_cmnd(sc) || sc->sc_data_direction == DMA_FROM_DEVICE)) { - err = cxgbi_ddp_reserve(cconn->cep->csk, &tag, sw_tag, - scsi_in(sc)->length, - scsi_in(sc)->table.sgl, - scsi_in(sc)->table.nents, - GFP_ATOMIC); - if (err < 0) - log_debug(1 << CXGBI_DBG_DDP, - "csk 0x%p, R task 0x%p, %u,%u, no ddp.\n", - cconn->cep->csk, task, scsi_in(sc)->length, - scsi_in(sc)->table.nents); - } - - if (err < 0) - tag = cxgbi_set_non_ddp_tag(tformat, sw_tag); - /* the itt need to sent in big-endian order */ - *hdr_itt = (__force itt_t)htonl(tag); - - log_debug(1 << CXGBI_DBG_DDP, - "cdev 0x%p, task 0x%p, 0x%x(0x%x,0x%x)->0x%x/0x%x.\n", - chba->cdev, task, sw_tag, task->itt, sess->age, tag, *hdr_itt); return 0; } void cxgbi_parse_pdu_itt(struct iscsi_conn *conn, itt_t itt, int *idx, int *age) { - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct cxgbi_conn *cconn = tcp_conn->dd_data; - struct cxgbi_device *cdev = cconn->chba->cdev; - u32 tag = ntohl((__force u32) itt); - u32 sw_bits; - - sw_bits = cxgbi_tag_nonrsvd_bits(&cdev->tag_format, tag); - if (idx) - *idx = sw_bits & ((1 << cconn->task_idx_bits) - 1); - if (age) - *age = (sw_bits >> cconn->task_idx_bits) & ISCSI_AGE_MASK; - - log_debug(1 << CXGBI_DBG_DDP, - "cdev 0x%p, tag 0x%x/0x%x, -> 0x%x(0x%x,0x%x).\n", - cdev, tag, itt, sw_bits, idx ? *idx : 0xFFFFF, - age ? *age : 0xFF); } EXPORT_SYMBOL_GPL(cxgbi_parse_pdu_itt); @@ -2540,9 +1974,6 @@ int cxgbi_bind_conn(struct iscsi_cls_session *cls_session, /* setup ddp pagesize */ cep = ep->dd_data; csk = cep->csk; - err = csk->cdev->csk_ddp_setup_pgidx(csk, csk->tid, page_idx, 0); - if (err < 0) - return err; err = iscsi_conn_bind(cls_session, cls_conn, is_leading); if (err) @@ -2915,16 +2346,7 @@ EXPORT_SYMBOL_GPL(cxgbi_attr_is_visible); static int __init libcxgbi_init_module(void) { - sw_tag_idx_bits = (__ilog2_u32(ISCSI_ITT_MASK)) + 1; - sw_tag_age_bits = (__ilog2_u32(ISCSI_AGE_MASK)) + 1; - pr_info("%s", version); - - pr_info("tag itt 0x%x, %u bits, age 0x%x, %u bits.\n", - ISCSI_ITT_MASK, sw_tag_idx_bits, - ISCSI_AGE_MASK, sw_tag_age_bits); - - ddp_setup_host_page_size(); return 0; } diff --git a/drivers/scsi/cxgbi/libcxgbi.h b/drivers/scsi/cxgbi/libcxgbi.h index 9842301..827afa7 100644 --- a/drivers/scsi/cxgbi/libcxgbi.h +++ b/drivers/scsi/cxgbi/libcxgbi.h @@ -84,91 +84,12 @@ static inline unsigned int cxgbi_ulp_extra_len(int submode) return ulp2_extra_len[submode & 3]; } -/* - * struct pagepod_hdr, pagepod - pagepod format - */ - #define CPL_RX_DDP_STATUS_DDP_SHIFT 16 /* ddp'able */ #define CPL_RX_DDP_STATUS_PAD_SHIFT 19 /* pad error */ #define CPL_RX_DDP_STATUS_HCRC_SHIFT 20 /* hcrc error */ #define CPL_RX_DDP_STATUS_DCRC_SHIFT 21 /* dcrc error */ -struct cxgbi_pagepod_hdr { - u32 vld_tid; - u32 pgsz_tag_clr; - u32 max_offset; - u32 page_offset; - u64 rsvd; -}; - -#define PPOD_PAGES_MAX 4 -struct cxgbi_pagepod { - struct cxgbi_pagepod_hdr hdr; - u64 addr[PPOD_PAGES_MAX + 1]; -}; - -struct cxgbi_tag_format { - unsigned char sw_bits; - unsigned char rsvd_bits; - unsigned char rsvd_shift; - unsigned char filler[1]; - u32 rsvd_mask; -}; - -struct cxgbi_gather_list { - unsigned int tag; - unsigned int length; - unsigned int offset; - unsigned int nelem; - struct page **pages; - dma_addr_t phys_addr[0]; -}; - -struct cxgbi_ddp_info { - struct kref refcnt; - struct cxgbi_device *cdev; - struct pci_dev *pdev; - unsigned int max_txsz; - unsigned int max_rxsz; - unsigned int llimit; - unsigned int ulimit; - unsigned int nppods; - unsigned int idx_last; - unsigned char idx_bits; - unsigned char filler[3]; - unsigned int idx_mask; - unsigned int rsvd_tag_mask; - spinlock_t map_lock; - struct cxgbi_gather_list **gl_map; -}; - #define DDP_PGIDX_MAX 4 -#define DDP_THRESHOLD 2048 - -#define PPOD_PAGES_SHIFT 2 /* 4 pages per pod */ - -#define PPOD_SIZE sizeof(struct cxgbi_pagepod) /* 64 */ -#define PPOD_SIZE_SHIFT 6 - -#define ULPMEM_DSGL_MAX_NPPODS 16 /* 1024/PPOD_SIZE */ -#define ULPMEM_IDATA_MAX_NPPODS 4 /* 256/PPOD_SIZE */ -#define PCIE_MEMWIN_MAX_NPPODS 16 /* 1024/PPOD_SIZE */ - -#define PPOD_COLOR_SHIFT 0 -#define PPOD_COLOR(x) ((x) << PPOD_COLOR_SHIFT) - -#define PPOD_IDX_SHIFT 6 -#define PPOD_IDX_MAX_SIZE 24 - -#define PPOD_TID_SHIFT 0 -#define PPOD_TID(x) ((x) << PPOD_TID_SHIFT) - -#define PPOD_TAG_SHIFT 6 -#define PPOD_TAG(x) ((x) << PPOD_TAG_SHIFT) - -#define PPOD_VALID_SHIFT 24 -#define PPOD_VALID(x) ((x) << PPOD_VALID_SHIFT) -#define PPOD_VALID_FLAG PPOD_VALID(1U) /* * sge_opaque_hdr - @@ -548,15 +469,8 @@ struct cxgbi_device { unsigned int tx_max_size; unsigned int rx_max_size; struct cxgbi_ports_map pmap; - struct cxgbi_tag_format tag_format; - struct cxgbi_ddp_info *ddp; void (*dev_ddp_cleanup)(struct cxgbi_device *); - int (*csk_ddp_set)(struct cxgbi_sock *, struct cxgbi_pagepod_hdr *, - unsigned int, unsigned int, - struct cxgbi_gather_list *); - void (*csk_ddp_clear)(struct cxgbi_hba *, - unsigned int, unsigned int, unsigned int); int (*csk_ddp_setup_digest)(struct cxgbi_sock *, unsigned int, int, int, int); int (*csk_ddp_setup_pgidx)(struct cxgbi_sock *, @@ -600,78 +514,6 @@ struct cxgbi_task_data { #define iscsi_task_cxgbi_data(task) \ ((task)->dd_data + sizeof(struct iscsi_tcp_task)) -static inline int cxgbi_is_ddp_tag(struct cxgbi_tag_format *tformat, u32 tag) -{ - return !(tag & (1 << (tformat->rsvd_bits + tformat->rsvd_shift - 1))); -} - -static inline int cxgbi_sw_tag_usable(struct cxgbi_tag_format *tformat, - u32 sw_tag) -{ - sw_tag >>= (32 - tformat->rsvd_bits); - return !sw_tag; -} - -static inline u32 cxgbi_set_non_ddp_tag(struct cxgbi_tag_format *tformat, - u32 sw_tag) -{ - unsigned char shift = tformat->rsvd_bits + tformat->rsvd_shift - 1; - u32 mask = (1 << shift) - 1; - - if (sw_tag && (sw_tag & ~mask)) { - u32 v1 = sw_tag & ((1 << shift) - 1); - u32 v2 = (sw_tag >> (shift - 1)) << shift; - - return v2 | v1 | 1 << shift; - } - - return sw_tag | 1 << shift; -} - -static inline u32 cxgbi_ddp_tag_base(struct cxgbi_tag_format *tformat, - u32 sw_tag) -{ - u32 mask = (1 << tformat->rsvd_shift) - 1; - - if (sw_tag && (sw_tag & ~mask)) { - u32 v1 = sw_tag & mask; - u32 v2 = sw_tag >> tformat->rsvd_shift; - - v2 <<= tformat->rsvd_bits + tformat->rsvd_shift; - - return v2 | v1; - } - - return sw_tag; -} - -static inline u32 cxgbi_tag_rsvd_bits(struct cxgbi_tag_format *tformat, - u32 tag) -{ - if (cxgbi_is_ddp_tag(tformat, tag)) - return (tag >> tformat->rsvd_shift) & tformat->rsvd_mask; - - return 0; -} - -static inline u32 cxgbi_tag_nonrsvd_bits(struct cxgbi_tag_format *tformat, - u32 tag) -{ - unsigned char shift = tformat->rsvd_bits + tformat->rsvd_shift - 1; - u32 v1, v2; - - if (cxgbi_is_ddp_tag(tformat, tag)) { - v1 = tag & ((1 << tformat->rsvd_shift) - 1); - v2 = (tag >> (shift + 1)) << tformat->rsvd_shift; - } else { - u32 mask = (1 << shift) - 1; - tag &= ~(1 << shift); - v1 = tag & mask; - v2 = (tag >> 1) & ~mask; - } - return v1 | v2; -} - static inline void *cxgbi_alloc_big_mem(unsigned int size, gfp_t gfp) { @@ -749,7 +591,4 @@ int cxgbi_ddp_init(struct cxgbi_device *, unsigned int, unsigned int, unsigned int, unsigned int); int cxgbi_ddp_cleanup(struct cxgbi_device *); void cxgbi_ddp_page_size_factor(int *); -void cxgbi_ddp_ppod_clear(struct cxgbi_pagepod *); -void cxgbi_ddp_ppod_set(struct cxgbi_pagepod *, struct cxgbi_pagepod_hdr *, - struct cxgbi_gather_list *, unsigned int); #endif /*__LIBCXGBI_H__*/ -- cgit v0.10.2 From 71f7a00bd13d5b0a21a77c438007210bf0a06c91 Mon Sep 17 00:00:00 2001 From: Varun Prakash Date: Thu, 21 Jul 2016 22:57:16 +0530 Subject: cxgb4i,libcxgbi: add iSCSI DDP support Add iSCSI DDP support in cxgb4i driver using common iSCSI DDP Page Pod Manager. Signed-off-by: Varun Prakash Signed-off-by: David S. Miller diff --git a/drivers/scsi/cxgbi/Makefile b/drivers/scsi/cxgbi/Makefile index 86007e3..a73781a 100644 --- a/drivers/scsi/cxgbi/Makefile +++ b/drivers/scsi/cxgbi/Makefile @@ -1,2 +1,4 @@ +ccflags-y += -Idrivers/net/ethernet/chelsio/libcxgb + obj-$(CONFIG_SCSI_CXGB3_ISCSI) += libcxgbi.o cxgb3i/ obj-$(CONFIG_SCSI_CXGB4_ISCSI) += libcxgbi.o cxgb4i/ diff --git a/drivers/scsi/cxgbi/cxgb3i/Kbuild b/drivers/scsi/cxgbi/cxgb3i/Kbuild index 961a12f..663c52e 100644 --- a/drivers/scsi/cxgbi/cxgb3i/Kbuild +++ b/drivers/scsi/cxgbi/cxgb3i/Kbuild @@ -1,3 +1,4 @@ ccflags-y += -I$(srctree)/drivers/net/ethernet/chelsio/cxgb3 +ccflags-y += -I$(srctree)/drivers/net/ethernet/chelsio/libcxgb obj-$(CONFIG_SCSI_CXGB3_ISCSI) += cxgb3i.o diff --git a/drivers/scsi/cxgbi/cxgb3i/Kconfig b/drivers/scsi/cxgbi/cxgb3i/Kconfig index e460398..f68c871 100644 --- a/drivers/scsi/cxgbi/cxgb3i/Kconfig +++ b/drivers/scsi/cxgbi/cxgb3i/Kconfig @@ -5,6 +5,7 @@ config SCSI_CXGB3_ISCSI select ETHERNET select NET_VENDOR_CHELSIO select CHELSIO_T3 + select CHELSIO_LIB select SCSI_ISCSI_ATTRS ---help--- This driver supports iSCSI offload for the Chelsio T3 devices. diff --git a/drivers/scsi/cxgbi/cxgb4i/Kbuild b/drivers/scsi/cxgbi/cxgb4i/Kbuild index 3745864..38e03c2 100644 --- a/drivers/scsi/cxgbi/cxgb4i/Kbuild +++ b/drivers/scsi/cxgbi/cxgb4i/Kbuild @@ -1,3 +1,4 @@ ccflags-y += -I$(srctree)/drivers/net/ethernet/chelsio/cxgb4 +ccflags-y += -I$(srctree)/drivers/net/ethernet/chelsio/libcxgb obj-$(CONFIG_SCSI_CXGB4_ISCSI) += cxgb4i.o diff --git a/drivers/scsi/cxgbi/cxgb4i/Kconfig b/drivers/scsi/cxgbi/cxgb4i/Kconfig index 8c4e423..594f593 100644 --- a/drivers/scsi/cxgbi/cxgb4i/Kconfig +++ b/drivers/scsi/cxgbi/cxgb4i/Kconfig @@ -5,6 +5,7 @@ config SCSI_CXGB4_ISCSI select ETHERNET select NET_VENDOR_CHELSIO select CHELSIO_T4 + select CHELSIO_LIB select SCSI_ISCSI_ATTRS ---help--- This driver supports iSCSI offload for the Chelsio T4 devices. diff --git a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c index 2911214..521f9e4 100644 --- a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c +++ b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c @@ -1543,6 +1543,115 @@ int cxgb4i_ofld_init(struct cxgbi_device *cdev) return 0; } +static inline void +ulp_mem_io_set_hdr(struct cxgbi_device *cdev, + struct ulp_mem_io *req, + unsigned int wr_len, unsigned int dlen, + unsigned int pm_addr, + int tid) +{ + struct cxgb4_lld_info *lldi = cxgbi_cdev_priv(cdev); + struct ulptx_idata *idata = (struct ulptx_idata *)(req + 1); + + INIT_ULPTX_WR(req, wr_len, 0, tid); + req->wr.wr_hi = htonl(FW_WR_OP_V(FW_ULPTX_WR) | + FW_WR_ATOMIC_V(0)); + req->cmd = htonl(ULPTX_CMD_V(ULP_TX_MEM_WRITE) | + ULP_MEMIO_ORDER_V(is_t4(lldi->adapter_type)) | + T5_ULP_MEMIO_IMM_V(!is_t4(lldi->adapter_type))); + req->dlen = htonl(ULP_MEMIO_DATA_LEN_V(dlen >> 5)); + req->lock_addr = htonl(ULP_MEMIO_ADDR_V(pm_addr >> 5)); + req->len16 = htonl(DIV_ROUND_UP(wr_len - sizeof(req->wr), 16)); + + idata->cmd_more = htonl(ULPTX_CMD_V(ULP_TX_SC_IMM)); + idata->len = htonl(dlen); +} + +static struct sk_buff * +ddp_ppod_init_idata(struct cxgbi_device *cdev, + struct cxgbi_ppm *ppm, + unsigned int idx, unsigned int npods, + unsigned int tid) +{ + unsigned int pm_addr = (idx << PPOD_SIZE_SHIFT) + ppm->llimit; + unsigned int dlen = npods << PPOD_SIZE_SHIFT; + unsigned int wr_len = roundup(sizeof(struct ulp_mem_io) + + sizeof(struct ulptx_idata) + dlen, 16); + struct sk_buff *skb = alloc_wr(wr_len, 0, GFP_ATOMIC); + + if (!skb) { + pr_err("%s: %s idx %u, npods %u, OOM.\n", + __func__, ppm->ndev->name, idx, npods); + return NULL; + } + + ulp_mem_io_set_hdr(cdev, (struct ulp_mem_io *)skb->head, wr_len, dlen, + pm_addr, tid); + + return skb; +} + +static int ddp_ppod_write_idata(struct cxgbi_ppm *ppm, struct cxgbi_sock *csk, + struct cxgbi_task_tag_info *ttinfo, + unsigned int idx, unsigned int npods, + struct scatterlist **sg_pp, + unsigned int *sg_off) +{ + struct cxgbi_device *cdev = csk->cdev; + struct sk_buff *skb = ddp_ppod_init_idata(cdev, ppm, idx, npods, + csk->tid); + struct ulp_mem_io *req; + struct ulptx_idata *idata; + struct cxgbi_pagepod *ppod; + int i; + + if (!skb) + return -ENOMEM; + + req = (struct ulp_mem_io *)skb->head; + idata = (struct ulptx_idata *)(req + 1); + ppod = (struct cxgbi_pagepod *)(idata + 1); + + for (i = 0; i < npods; i++, ppod++) + cxgbi_ddp_set_one_ppod(ppod, ttinfo, sg_pp, sg_off); + + cxgbi_skcb_set_flag(skb, SKCBF_TX_MEM_WRITE); + cxgbi_skcb_set_flag(skb, SKCBF_TX_FLAG_COMPL); + set_wr_txq(skb, CPL_PRIORITY_DATA, csk->port_id); + + spin_lock_bh(&csk->lock); + cxgbi_sock_skb_entail(csk, skb); + spin_unlock_bh(&csk->lock); + + return 0; +} + +static int ddp_set_map(struct cxgbi_ppm *ppm, struct cxgbi_sock *csk, + struct cxgbi_task_tag_info *ttinfo) +{ + unsigned int pidx = ttinfo->idx; + unsigned int npods = ttinfo->npods; + unsigned int i, cnt; + int err = 0; + struct scatterlist *sg = ttinfo->sgl; + unsigned int offset = 0; + + ttinfo->cid = csk->port_id; + + for (i = 0; i < npods; i += cnt, pidx += cnt) { + cnt = npods - i; + + if (cnt > ULPMEM_IDATA_MAX_NPPODS) + cnt = ULPMEM_IDATA_MAX_NPPODS; + err = ddp_ppod_write_idata(ppm, csk, ttinfo, pidx, cnt, + &sg, &offset); + if (err < 0) + break; + } + + return err; +} + static int ddp_setup_conn_pgidx(struct cxgbi_sock *csk, unsigned int tid, int pg_idx, bool reply) { @@ -1606,10 +1715,46 @@ static int ddp_setup_conn_digest(struct cxgbi_sock *csk, unsigned int tid, return 0; } +static struct cxgbi_ppm *cdev2ppm(struct cxgbi_device *cdev) +{ + return (struct cxgbi_ppm *)(*((struct cxgb4_lld_info *) + (cxgbi_cdev_priv(cdev)))->iscsi_ppm); +} + static int cxgb4i_ddp_init(struct cxgbi_device *cdev) { + struct cxgb4_lld_info *lldi = cxgbi_cdev_priv(cdev); + struct net_device *ndev = cdev->ports[0]; + struct cxgbi_tag_format tformat; + unsigned int ppmax; + int i; + + if (!lldi->vr->iscsi.size) { + pr_warn("%s, iscsi NOT enabled, check config!\n", ndev->name); + return -EACCES; + } + + cdev->flags |= CXGBI_FLAG_USE_PPOD_OFLDQ; + ppmax = lldi->vr->iscsi.size >> PPOD_SIZE_SHIFT; + + memset(&tformat, 0, sizeof(struct cxgbi_tag_format)); + for (i = 0; i < 4; i++) + tformat.pgsz_order[i] = (lldi->iscsi_pgsz_order >> (i << 3)) + & 0xF; + cxgbi_tagmask_check(lldi->iscsi_tagmask, &tformat); + + cxgbi_ddp_ppm_setup(lldi->iscsi_ppm, cdev, &tformat, ppmax, + lldi->iscsi_llimit, lldi->vr->iscsi.start, 2); + cdev->csk_ddp_setup_digest = ddp_setup_conn_digest; cdev->csk_ddp_setup_pgidx = ddp_setup_conn_pgidx; + cdev->csk_ddp_set_map = ddp_set_map; + cdev->tx_max_size = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD, + lldi->iscsi_iolen - ISCSI_PDU_NONPAYLOAD_LEN); + cdev->rx_max_size = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD, + lldi->iscsi_iolen - ISCSI_PDU_NONPAYLOAD_LEN); + cdev->cdev2ppm = cdev2ppm; + return 0; } diff --git a/drivers/scsi/cxgbi/libcxgbi.c b/drivers/scsi/cxgbi/libcxgbi.c index f389d34..9d425a7 100644 --- a/drivers/scsi/cxgbi/libcxgbi.c +++ b/drivers/scsi/cxgbi/libcxgbi.c @@ -64,6 +64,14 @@ static DEFINE_MUTEX(cdev_mutex); static LIST_HEAD(cdev_rcu_list); static DEFINE_SPINLOCK(cdev_rcu_lock); +static inline void cxgbi_decode_sw_tag(u32 sw_tag, int *idx, int *age) +{ + if (age) + *age = sw_tag & 0x7FFF; + if (idx) + *idx = (sw_tag >> 16) & 0x7FFF; +} + int cxgbi_device_portmap_create(struct cxgbi_device *cdev, unsigned int base, unsigned int max_conn) { @@ -1176,23 +1184,315 @@ out_err: goto done; } +static inline void +scmd_get_params(struct scsi_cmnd *sc, struct scatterlist **sgl, + unsigned int *sgcnt, unsigned int *dlen, + unsigned int prot) +{ + struct scsi_data_buffer *sdb = prot ? scsi_prot(sc) : scsi_out(sc); + + *sgl = sdb->table.sgl; + *sgcnt = sdb->table.nents; + *dlen = sdb->length; + /* Caution: for protection sdb, sdb->length is invalid */ +} + +void cxgbi_ddp_set_one_ppod(struct cxgbi_pagepod *ppod, + struct cxgbi_task_tag_info *ttinfo, + struct scatterlist **sg_pp, unsigned int *sg_off) +{ + struct scatterlist *sg = sg_pp ? *sg_pp : NULL; + unsigned int offset = sg_off ? *sg_off : 0; + dma_addr_t addr = 0UL; + unsigned int len = 0; + int i; + + memcpy(ppod, &ttinfo->hdr, sizeof(struct cxgbi_pagepod_hdr)); + + if (sg) { + addr = sg_dma_address(sg); + len = sg_dma_len(sg); + } + + for (i = 0; i < PPOD_PAGES_MAX; i++) { + if (sg) { + ppod->addr[i] = cpu_to_be64(addr + offset); + offset += PAGE_SIZE; + if (offset == (len + sg->offset)) { + offset = 0; + sg = sg_next(sg); + if (sg) { + addr = sg_dma_address(sg); + len = sg_dma_len(sg); + } + } + } else { + ppod->addr[i] = 0ULL; + } + } + + /* + * the fifth address needs to be repeated in the next ppod, so do + * not move sg + */ + if (sg_pp) { + *sg_pp = sg; + *sg_off = offset; + } + + if (offset == len) { + offset = 0; + sg = sg_next(sg); + if (sg) { + addr = sg_dma_address(sg); + len = sg_dma_len(sg); + } + } + ppod->addr[i] = sg ? cpu_to_be64(addr + offset) : 0ULL; +} +EXPORT_SYMBOL_GPL(cxgbi_ddp_set_one_ppod); + /* * APIs interacting with open-iscsi libraries */ static unsigned char padding[4]; +void cxgbi_ddp_ppm_setup(void **ppm_pp, struct cxgbi_device *cdev, + struct cxgbi_tag_format *tformat, unsigned int ppmax, + unsigned int llimit, unsigned int start, + unsigned int rsvd_factor) +{ + int err = cxgbi_ppm_init(ppm_pp, cdev->ports[0], cdev->pdev, + cdev->lldev, tformat, ppmax, llimit, start, + rsvd_factor); + + if (err >= 0) { + struct cxgbi_ppm *ppm = (struct cxgbi_ppm *)(*ppm_pp); + + if (ppm->ppmax < 1024 || + ppm->tformat.pgsz_idx_dflt >= DDP_PGIDX_MAX) + cdev->flags |= CXGBI_FLAG_DDP_OFF; + err = 0; + } else { + cdev->flags |= CXGBI_FLAG_DDP_OFF; + } +} +EXPORT_SYMBOL_GPL(cxgbi_ddp_ppm_setup); + +static int cxgbi_ddp_sgl_check(struct scatterlist *sgl, int nents) +{ + int i; + int last_sgidx = nents - 1; + struct scatterlist *sg = sgl; + + for (i = 0; i < nents; i++, sg = sg_next(sg)) { + unsigned int len = sg->length + sg->offset; + + if ((sg->offset & 0x3) || (i && sg->offset) || + ((i != last_sgidx) && len != PAGE_SIZE)) { + log_debug(1 << CXGBI_DBG_DDP, + "sg %u/%u, %u,%u, not aligned.\n", + i, nents, sg->offset, sg->length); + goto err_out; + } + } + return 0; +err_out: + return -EINVAL; +} + +static int cxgbi_ddp_reserve(struct cxgbi_conn *cconn, + struct cxgbi_task_data *tdata, u32 sw_tag, + unsigned int xferlen) +{ + struct cxgbi_sock *csk = cconn->cep->csk; + struct cxgbi_device *cdev = csk->cdev; + struct cxgbi_ppm *ppm = cdev->cdev2ppm(cdev); + struct cxgbi_task_tag_info *ttinfo = &tdata->ttinfo; + struct scatterlist *sgl = ttinfo->sgl; + unsigned int sgcnt = ttinfo->nents; + unsigned int sg_offset = sgl->offset; + int err; + + if (cdev->flags & CXGBI_FLAG_DDP_OFF) { + log_debug(1 << CXGBI_DBG_DDP, + "cdev 0x%p DDP off.\n", cdev); + return -EINVAL; + } + + if (!ppm || xferlen < DDP_THRESHOLD || !sgcnt || + ppm->tformat.pgsz_idx_dflt >= DDP_PGIDX_MAX) { + log_debug(1 << CXGBI_DBG_DDP, + "ppm 0x%p, pgidx %u, xfer %u, sgcnt %u, NO ddp.\n", + ppm, ppm ? ppm->tformat.pgsz_idx_dflt : DDP_PGIDX_MAX, + xferlen, ttinfo->nents); + return -EINVAL; + } + + /* make sure the buffer is suitable for ddp */ + if (cxgbi_ddp_sgl_check(sgl, sgcnt) < 0) + return -EINVAL; + + ttinfo->nr_pages = (xferlen + sgl->offset + (1 << PAGE_SHIFT) - 1) >> + PAGE_SHIFT; + + /* + * the ddp tag will be used for the itt in the outgoing pdu, + * the itt genrated by libiscsi is saved in the ppm and can be + * retrieved via the ddp tag + */ + err = cxgbi_ppm_ppods_reserve(ppm, ttinfo->nr_pages, 0, &ttinfo->idx, + &ttinfo->tag, (unsigned long)sw_tag); + if (err < 0) { + cconn->ddp_full++; + return err; + } + ttinfo->npods = err; + + /* setup dma from scsi command sgl */ + sgl->offset = 0; + err = dma_map_sg(&ppm->pdev->dev, sgl, sgcnt, DMA_FROM_DEVICE); + sgl->offset = sg_offset; + if (err == 0) { + pr_info("%s: 0x%x, xfer %u, sgl %u dma mapping err.\n", + __func__, sw_tag, xferlen, sgcnt); + goto rel_ppods; + } + if (err != ttinfo->nr_pages) { + log_debug(1 << CXGBI_DBG_DDP, + "%s: sw tag 0x%x, xfer %u, sgl %u, dma count %d.\n", + __func__, sw_tag, xferlen, sgcnt, err); + } + + ttinfo->flags |= CXGBI_PPOD_INFO_FLAG_MAPPED; + ttinfo->cid = csk->port_id; + + cxgbi_ppm_make_ppod_hdr(ppm, ttinfo->tag, csk->tid, sgl->offset, + xferlen, &ttinfo->hdr); + + if (cdev->flags & CXGBI_FLAG_USE_PPOD_OFLDQ) { + /* write ppod from xmit_pdu (of iscsi_scsi_command pdu) */ + ttinfo->flags |= CXGBI_PPOD_INFO_FLAG_VALID; + } else { + /* write ppod from control queue now */ + err = cdev->csk_ddp_set_map(ppm, csk, ttinfo); + if (err < 0) + goto rel_ppods; + } + + return 0; + +rel_ppods: + cxgbi_ppm_ppod_release(ppm, ttinfo->idx); + + if (ttinfo->flags & CXGBI_PPOD_INFO_FLAG_MAPPED) { + ttinfo->flags &= ~CXGBI_PPOD_INFO_FLAG_MAPPED; + dma_unmap_sg(&ppm->pdev->dev, sgl, sgcnt, DMA_FROM_DEVICE); + } + return -EINVAL; +} + static void task_release_itt(struct iscsi_task *task, itt_t hdr_itt) { + struct scsi_cmnd *sc = task->sc; + struct iscsi_tcp_conn *tcp_conn = task->conn->dd_data; + struct cxgbi_conn *cconn = tcp_conn->dd_data; + struct cxgbi_device *cdev = cconn->chba->cdev; + struct cxgbi_ppm *ppm = cdev->cdev2ppm(cdev); + u32 tag = ntohl((__force u32)hdr_itt); + + log_debug(1 << CXGBI_DBG_DDP, + "cdev 0x%p, task 0x%p, release tag 0x%x.\n", + cdev, task, tag); + if (sc && + (scsi_bidi_cmnd(sc) || sc->sc_data_direction == DMA_FROM_DEVICE) && + cxgbi_ppm_is_ddp_tag(ppm, tag)) { + struct cxgbi_task_data *tdata = iscsi_task_cxgbi_data(task); + struct cxgbi_task_tag_info *ttinfo = &tdata->ttinfo; + + if (!(cdev->flags & CXGBI_FLAG_USE_PPOD_OFLDQ)) + cdev->csk_ddp_clear_map(cdev, ppm, ttinfo); + cxgbi_ppm_ppod_release(ppm, ttinfo->idx); + dma_unmap_sg(&ppm->pdev->dev, ttinfo->sgl, ttinfo->nents, + DMA_FROM_DEVICE); + } +} + +static inline u32 cxgbi_build_sw_tag(u32 idx, u32 age) +{ + /* assume idx and age both are < 0x7FFF (32767) */ + return (idx << 16) | age; } static int task_reserve_itt(struct iscsi_task *task, itt_t *hdr_itt) { + struct scsi_cmnd *sc = task->sc; + struct iscsi_conn *conn = task->conn; + struct iscsi_session *sess = conn->session; + struct iscsi_tcp_conn *tcp_conn = conn->dd_data; + struct cxgbi_conn *cconn = tcp_conn->dd_data; + struct cxgbi_device *cdev = cconn->chba->cdev; + struct cxgbi_ppm *ppm = cdev->cdev2ppm(cdev); + u32 sw_tag = cxgbi_build_sw_tag(task->itt, sess->age); + u32 tag = 0; + int err = -EINVAL; + + if (sc && + (scsi_bidi_cmnd(sc) || sc->sc_data_direction == DMA_FROM_DEVICE) + ) { + struct cxgbi_task_data *tdata = iscsi_task_cxgbi_data(task); + struct cxgbi_task_tag_info *ttinfo = &tdata->ttinfo; + + scmd_get_params(sc, &ttinfo->sgl, &ttinfo->nents, + &tdata->dlen, 0); + err = cxgbi_ddp_reserve(cconn, tdata, sw_tag, tdata->dlen); + if (!err) + tag = ttinfo->tag; + else + log_debug(1 << CXGBI_DBG_DDP, + "csk 0x%p, R task 0x%p, %u,%u, no ddp.\n", + cconn->cep->csk, task, tdata->dlen, + ttinfo->nents); + } + + if (err < 0) { + err = cxgbi_ppm_make_non_ddp_tag(ppm, sw_tag, &tag); + if (err < 0) + return err; + } + /* the itt need to sent in big-endian order */ + *hdr_itt = (__force itt_t)htonl(tag); + + log_debug(1 << CXGBI_DBG_DDP, + "cdev 0x%p, task 0x%p, 0x%x(0x%x,0x%x)->0x%x/0x%x.\n", + cdev, task, sw_tag, task->itt, sess->age, tag, *hdr_itt); return 0; } void cxgbi_parse_pdu_itt(struct iscsi_conn *conn, itt_t itt, int *idx, int *age) { + struct iscsi_tcp_conn *tcp_conn = conn->dd_data; + struct cxgbi_conn *cconn = tcp_conn->dd_data; + struct cxgbi_device *cdev = cconn->chba->cdev; + struct cxgbi_ppm *ppm = cdev->cdev2ppm(cdev); + u32 tag = ntohl((__force u32)itt); + u32 sw_bits; + + if (ppm) { + if (cxgbi_ppm_is_ddp_tag(ppm, tag)) + sw_bits = cxgbi_ppm_get_tag_caller_data(ppm, tag); + else + sw_bits = cxgbi_ppm_decode_non_ddp_tag(ppm, tag); + } else { + sw_bits = tag; + } + + cxgbi_decode_sw_tag(sw_bits, idx, age); + log_debug(1 << CXGBI_DBG_DDP, + "cdev 0x%p, tag 0x%x/0x%x, -> 0x%x(0x%x,0x%x).\n", + cdev, tag, itt, sw_bits, idx ? *idx : 0xFFFFF, + age ? *age : 0xFF); } EXPORT_SYMBOL_GPL(cxgbi_parse_pdu_itt); @@ -1694,7 +1994,9 @@ int cxgbi_conn_xmit_pdu(struct iscsi_task *task) struct iscsi_tcp_conn *tcp_conn = task->conn->dd_data; struct cxgbi_conn *cconn = tcp_conn->dd_data; struct cxgbi_task_data *tdata = iscsi_task_cxgbi_data(task); + struct cxgbi_task_tag_info *ttinfo = &tdata->ttinfo; struct sk_buff *skb = tdata->skb; + struct cxgbi_sock *csk = NULL; unsigned int datalen; int err; @@ -1704,8 +2006,28 @@ int cxgbi_conn_xmit_pdu(struct iscsi_task *task) return 0; } + if (cconn && cconn->cep) + csk = cconn->cep->csk; + if (!csk) { + log_debug(1 << CXGBI_DBG_ISCSI | 1 << CXGBI_DBG_PDU_TX, + "task 0x%p, csk gone.\n", task); + return -EPIPE; + } + datalen = skb->data_len; tdata->skb = NULL; + + /* write ppod first if using ofldq to write ppod */ + if (ttinfo->flags & CXGBI_PPOD_INFO_FLAG_VALID) { + struct cxgbi_ppm *ppm = csk->cdev->cdev2ppm(csk->cdev); + + ttinfo->flags &= ~CXGBI_PPOD_INFO_FLAG_VALID; + if (csk->cdev->csk_ddp_set_map(ppm, csk, ttinfo) < 0) + pr_err("task 0x%p, ppod writing using ofldq failed.\n", + task); + /* continue. Let fl get the data */ + } + err = cxgbi_sock_send_pdus(cconn->cep->csk, skb); if (err > 0) { int pdulen = err; @@ -1747,12 +2069,14 @@ EXPORT_SYMBOL_GPL(cxgbi_conn_xmit_pdu); void cxgbi_cleanup_task(struct iscsi_task *task) { + struct iscsi_tcp_task *tcp_task = task->dd_data; struct cxgbi_task_data *tdata = iscsi_task_cxgbi_data(task); log_debug(1 << CXGBI_DBG_ISCSI, "task 0x%p, skb 0x%p, itt 0x%x.\n", task, tdata->skb, task->hdr_itt); + tcp_task->dd_data = NULL; /* never reached the xmit task callout */ if (tdata->skb) __kfree_skb(tdata->skb); @@ -1962,6 +2286,7 @@ int cxgbi_bind_conn(struct iscsi_cls_session *cls_session, struct iscsi_conn *conn = cls_conn->dd_data; struct iscsi_tcp_conn *tcp_conn = conn->dd_data; struct cxgbi_conn *cconn = tcp_conn->dd_data; + struct cxgbi_ppm *ppm; struct iscsi_endpoint *ep; struct cxgbi_endpoint *cep; struct cxgbi_sock *csk; @@ -1975,6 +2300,12 @@ int cxgbi_bind_conn(struct iscsi_cls_session *cls_session, cep = ep->dd_data; csk = cep->csk; + ppm = csk->cdev->cdev2ppm(csk->cdev); + err = csk->cdev->csk_ddp_setup_pgidx(csk, csk->tid, + ppm->tformat.pgsz_idx_dflt, 0); + if (err < 0) + return err; + err = iscsi_conn_bind(cls_session, cls_conn, is_leading); if (err) return -EINVAL; diff --git a/drivers/scsi/cxgbi/libcxgbi.h b/drivers/scsi/cxgbi/libcxgbi.h index 827afa7..e780273 100644 --- a/drivers/scsi/cxgbi/libcxgbi.h +++ b/drivers/scsi/cxgbi/libcxgbi.h @@ -24,9 +24,12 @@ #include #include #include +#include #include #include +#include + enum cxgbi_dbg_flag { CXGBI_DBG_ISCSI, CXGBI_DBG_DDP, @@ -89,8 +92,6 @@ static inline unsigned int cxgbi_ulp_extra_len(int submode) #define CPL_RX_DDP_STATUS_HCRC_SHIFT 20 /* hcrc error */ #define CPL_RX_DDP_STATUS_DCRC_SHIFT 21 /* dcrc error */ -#define DDP_PGIDX_MAX 4 - /* * sge_opaque_hdr - * Opaque version of structure the SGE stores at skb->head of TX_DATA packets @@ -200,6 +201,8 @@ struct cxgbi_skb_tx_cb { enum cxgbi_skcb_flags { SKCBF_TX_NEED_HDR, /* packet needs a header */ + SKCBF_TX_MEM_WRITE, /* memory write */ + SKCBF_TX_FLAG_COMPL, /* wr completion flag */ SKCBF_RX_COALESCED, /* received whole pdu */ SKCBF_RX_HDR, /* received pdu header */ SKCBF_RX_DATA, /* received pdu payload */ @@ -448,6 +451,9 @@ struct cxgbi_ports_map { #define CXGBI_FLAG_DEV_T4 0x2 #define CXGBI_FLAG_ADAPTER_RESET 0x4 #define CXGBI_FLAG_IPV4_SET 0x10 +#define CXGBI_FLAG_USE_PPOD_OFLDQ 0x40 +#define CXGBI_FLAG_DDP_OFF 0x100 + struct cxgbi_device { struct list_head list_head; struct list_head rcu_node; @@ -471,6 +477,12 @@ struct cxgbi_device { struct cxgbi_ports_map pmap; void (*dev_ddp_cleanup)(struct cxgbi_device *); + struct cxgbi_ppm* (*cdev2ppm)(struct cxgbi_device *); + int (*csk_ddp_set_map)(struct cxgbi_ppm *, struct cxgbi_sock *, + struct cxgbi_task_tag_info *); + void (*csk_ddp_clear_map)(struct cxgbi_device *cdev, + struct cxgbi_ppm *, + struct cxgbi_task_tag_info *); int (*csk_ddp_setup_digest)(struct cxgbi_sock *, unsigned int, int, int, int); int (*csk_ddp_setup_pgidx)(struct cxgbi_sock *, @@ -494,6 +506,8 @@ struct cxgbi_conn { struct iscsi_conn *iconn; struct cxgbi_hba *chba; u32 task_idx_bits; + unsigned int ddp_full; + unsigned int ddp_tag_full; }; struct cxgbi_endpoint { @@ -507,9 +521,11 @@ struct cxgbi_task_data { unsigned short nr_frags; struct page_frag frags[MAX_PDU_FRAGS]; struct sk_buff *skb; + unsigned int dlen; unsigned int offset; unsigned int count; unsigned int sgoffset; + struct cxgbi_task_tag_info ttinfo; }; #define iscsi_task_cxgbi_data(task) \ ((task)->dd_data + sizeof(struct iscsi_tcp_task)) @@ -591,4 +607,11 @@ int cxgbi_ddp_init(struct cxgbi_device *, unsigned int, unsigned int, unsigned int, unsigned int); int cxgbi_ddp_cleanup(struct cxgbi_device *); void cxgbi_ddp_page_size_factor(int *); +void cxgbi_ddp_set_one_ppod(struct cxgbi_pagepod *, + struct cxgbi_task_tag_info *, + struct scatterlist **sg_pp, unsigned int *sg_off); +void cxgbi_ddp_ppm_setup(void **ppm_pp, struct cxgbi_device *, + struct cxgbi_tag_format *, unsigned int ppmax, + unsigned int llimit, unsigned int start, + unsigned int rsvd_factor); #endif /*__LIBCXGBI_H__*/ -- cgit v0.10.2 From b75113b12c36606906538a74ae755a655ae583b7 Mon Sep 17 00:00:00 2001 From: Varun Prakash Date: Thu, 21 Jul 2016 22:57:17 +0530 Subject: cxgb3i: add iSCSI DDP support Add iSCSI DDP support in cxgb3i driver using common iSCSI DDP Page Pod Manager. Signed-off-by: Varun Prakash Reviewed-by: Steve Wise Signed-off-by: David S. Miller diff --git a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c index fda0234..bb25ebb 100644 --- a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c +++ b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c @@ -1076,6 +1076,70 @@ static inline void ulp_mem_io_set_hdr(struct sk_buff *skb, unsigned int addr) req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_BYPASS)); req->cmd_lock_addr = htonl(V_ULP_MEMIO_ADDR(addr >> 5) | V_ULPTX_CMD(ULP_MEM_WRITE)); + req->len = htonl(V_ULP_MEMIO_DATA_LEN(IPPOD_SIZE >> 5) | + V_ULPTX_NFLITS((IPPOD_SIZE >> 3) + 1)); +} + +static struct cxgbi_ppm *cdev2ppm(struct cxgbi_device *cdev) +{ + return ((struct t3cdev *)cdev->lldev)->ulp_iscsi; +} + +static int ddp_set_map(struct cxgbi_ppm *ppm, struct cxgbi_sock *csk, + struct cxgbi_task_tag_info *ttinfo) +{ + unsigned int idx = ttinfo->idx; + unsigned int npods = ttinfo->npods; + struct scatterlist *sg = ttinfo->sgl; + struct cxgbi_pagepod *ppod; + struct ulp_mem_io *req; + unsigned int sg_off; + unsigned int pm_addr = (idx << PPOD_SIZE_SHIFT) + ppm->llimit; + int i; + + for (i = 0; i < npods; i++, idx++, pm_addr += IPPOD_SIZE) { + struct sk_buff *skb = alloc_wr(sizeof(struct ulp_mem_io) + + IPPOD_SIZE, 0, GFP_ATOMIC); + + if (!skb) + return -ENOMEM; + ulp_mem_io_set_hdr(skb, pm_addr); + req = (struct ulp_mem_io *)skb->head; + ppod = (struct cxgbi_pagepod *)(req + 1); + sg_off = i * PPOD_PAGES_MAX; + cxgbi_ddp_set_one_ppod(ppod, ttinfo, &sg, + &sg_off); + skb->priority = CPL_PRIORITY_CONTROL; + cxgb3_ofld_send(ppm->lldev, skb); + } + return 0; +} + +static void ddp_clear_map(struct cxgbi_device *cdev, struct cxgbi_ppm *ppm, + struct cxgbi_task_tag_info *ttinfo) +{ + unsigned int idx = ttinfo->idx; + unsigned int pm_addr = (idx << PPOD_SIZE_SHIFT) + ppm->llimit; + unsigned int npods = ttinfo->npods; + int i; + + log_debug(1 << CXGBI_DBG_DDP, + "cdev 0x%p, clear idx %u, npods %u.\n", + cdev, idx, npods); + + for (i = 0; i < npods; i++, idx++, pm_addr += IPPOD_SIZE) { + struct sk_buff *skb = alloc_wr(sizeof(struct ulp_mem_io) + + IPPOD_SIZE, 0, GFP_ATOMIC); + + if (!skb) { + pr_err("cdev 0x%p, clear ddp, %u,%d/%u, skb OOM.\n", + cdev, idx, i, npods); + continue; + } + ulp_mem_io_set_hdr(skb, pm_addr); + skb->priority = CPL_PRIORITY_CONTROL; + cxgb3_ofld_send(ppm->lldev, skb); + } } static int ddp_setup_conn_pgidx(struct cxgbi_sock *csk, @@ -1144,14 +1208,67 @@ static int ddp_setup_conn_digest(struct cxgbi_sock *csk, unsigned int tid, } /** - * ddp_init - initialize the cxgb3 adapter's ddp resource + * cxgb3i_ddp_init - initialize the cxgb3 adapter's ddp resource * @cdev: cxgb3i adapter * initialize the ddp pagepod manager for a given adapter */ static int cxgb3i_ddp_init(struct cxgbi_device *cdev) { + struct t3cdev *tdev = (struct t3cdev *)cdev->lldev; + struct net_device *ndev = cdev->ports[0]; + struct cxgbi_tag_format tformat; + unsigned int ppmax, tagmask = 0; + struct ulp_iscsi_info uinfo; + int i, err; + + err = tdev->ctl(tdev, ULP_ISCSI_GET_PARAMS, &uinfo); + if (err < 0) { + pr_err("%s, failed to get iscsi param %d.\n", + ndev->name, err); + return err; + } + if (uinfo.llimit >= uinfo.ulimit) { + pr_warn("T3 %s, iscsi NOT enabled %u ~ %u!\n", + ndev->name, uinfo.llimit, uinfo.ulimit); + return -EACCES; + } + + ppmax = (uinfo.ulimit - uinfo.llimit + 1) >> PPOD_SIZE_SHIFT; + + pr_info("T3 %s: 0x%x~0x%x, 0x%x, tagmask 0x%x -> 0x%x.\n", + ndev->name, uinfo.llimit, uinfo.ulimit, ppmax, uinfo.tagmask, + tagmask); + + memset(&tformat, 0, sizeof(struct cxgbi_tag_format)); + for (i = 0; i < 4; i++) + tformat.pgsz_order[i] = uinfo.pgsz_factor[i]; + cxgbi_tagmask_check(tagmask, &tformat); + + cxgbi_ddp_ppm_setup(&tdev->ulp_iscsi, cdev, &tformat, ppmax, + uinfo.llimit, uinfo.llimit, 0); + if (!(cdev->flags & CXGBI_FLAG_DDP_OFF)) { + uinfo.tagmask = tagmask; + uinfo.ulimit = uinfo.llimit + (ppmax << PPOD_SIZE_SHIFT); + + err = tdev->ctl(tdev, ULP_ISCSI_SET_PARAMS, &uinfo); + if (err < 0) { + pr_err("T3 %s fail to set iscsi param %d.\n", + ndev->name, err); + cdev->flags |= CXGBI_FLAG_DDP_OFF; + } + err = 0; + } + cdev->csk_ddp_setup_digest = ddp_setup_conn_digest; cdev->csk_ddp_setup_pgidx = ddp_setup_conn_pgidx; + cdev->csk_ddp_set_map = ddp_set_map; + cdev->csk_ddp_clear_map = ddp_clear_map; + cdev->cdev2ppm = cdev2ppm; + cdev->tx_max_size = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD, + uinfo.max_txsz - ISCSI_PDU_NONPAYLOAD_LEN); + cdev->rx_max_size = min_t(unsigned int, ULP2_MAX_PDU_PAYLOAD, + uinfo.max_rxsz - ISCSI_PDU_NONPAYLOAD_LEN); + return 0; } -- cgit v0.10.2 From 9d5c44b7c4f4345341bf96b16fdeb6debc437172 Mon Sep 17 00:00:00 2001 From: Varun Prakash Date: Thu, 21 Jul 2016 22:57:18 +0530 Subject: libcxgb: export ppm release and tagmask set api Export cxgbi_ppm_release() to release ppod manager and cxgbi_tagmask_set() to set tag mask, they are used by cxgb3i, cxgb4i and cxgbit. Signed-off-by: Varun Prakash Reviewed-by: Steve Wise Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c index 01a4329..551a365 100644 --- a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c +++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c @@ -337,6 +337,7 @@ int cxgbi_ppm_release(struct cxgbi_ppm *ppm) } return 1; } +EXPORT_SYMBOL(cxgbi_ppm_release); static struct cxgbi_ppm_pool *ppm_alloc_cpu_pool(unsigned int *total, unsigned int *pcpu_ppmax) @@ -490,6 +491,7 @@ unsigned int cxgbi_tagmask_set(unsigned int ppmax) return 1 << (bits + PPOD_IDX_SHIFT); } +EXPORT_SYMBOL(cxgbi_tagmask_set); MODULE_AUTHOR("Chelsio Communications"); MODULE_DESCRIPTION("Chelsio common library"); diff --git a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c index bb25ebb..61f16a2 100644 --- a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c +++ b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c @@ -1234,6 +1234,7 @@ static int cxgb3i_ddp_init(struct cxgbi_device *cdev) } ppmax = (uinfo.ulimit - uinfo.llimit + 1) >> PPOD_SIZE_SHIFT; + tagmask = cxgbi_tagmask_set(ppmax); pr_info("T3 %s: 0x%x~0x%x, 0x%x, tagmask 0x%x -> 0x%x.\n", ndev->name, uinfo.llimit, uinfo.ulimit, ppmax, uinfo.tagmask, diff --git a/drivers/scsi/cxgbi/libcxgbi.c b/drivers/scsi/cxgbi/libcxgbi.c index 9d425a7..d142113 100644 --- a/drivers/scsi/cxgbi/libcxgbi.c +++ b/drivers/scsi/cxgbi/libcxgbi.c @@ -121,6 +121,7 @@ static inline void cxgbi_device_destroy(struct cxgbi_device *cdev) "cdev 0x%p, p# %u.\n", cdev, cdev->nports); cxgbi_hbas_remove(cdev); cxgbi_device_portmap_cleanup(cdev); + cxgbi_ppm_release(cdev->cdev2ppm(cdev)); if (cdev->pmap.max_connect) cxgbi_free_big_mem(cdev->pmap.port_csk); kfree(cdev); diff --git a/drivers/target/iscsi/cxgbit/cxgbit_main.c b/drivers/target/iscsi/cxgbit/cxgbit_main.c index 60dccd0..27dd11a 100644 --- a/drivers/target/iscsi/cxgbit/cxgbit_main.c +++ b/drivers/target/iscsi/cxgbit/cxgbit_main.c @@ -26,6 +26,8 @@ void _cxgbit_free_cdev(struct kref *kref) struct cxgbit_device *cdev; cdev = container_of(kref, struct cxgbit_device, kref); + + cxgbi_ppm_release(cdev2ppm(cdev)); kfree(cdev); } -- cgit v0.10.2 From 4665bdd53047e70281648d9625b645e3f1740320 Mon Sep 17 00:00:00 2001 From: Varun Prakash Date: Thu, 21 Jul 2016 22:57:19 +0530 Subject: cxgb3i, cxgb4i: fix symbol not declared sparse warning Fix following sparse warnings warning: symbol 'cxgb3i_ofld_init' was not declared. Should it be static? warning: symbol 'cxgb4i_cplhandlers' was not declared. Should it be static? warning: symbol 'cxgb4i_ofld_init' was not declared. Should it be static? Signed-off-by: Varun Prakash Reviewed-by: Steve Wise Signed-off-by: David S. Miller diff --git a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c index 61f16a2..33e8346 100644 --- a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c +++ b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c @@ -1028,7 +1028,7 @@ cxgb3_cpl_handler_func cxgb3i_cpl_handlers[NUM_CPL_CMDS] = { * cxgb3i_ofld_init - allocate and initialize resources for each adapter found * @cdev: cxgbi adapter */ -int cxgb3i_ofld_init(struct cxgbi_device *cdev) +static int cxgb3i_ofld_init(struct cxgbi_device *cdev) { struct t3cdev *t3dev = (struct t3cdev *)cdev->lldev; struct adap_ports port; diff --git a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c index 521f9e4..e4ba2d2 100644 --- a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c +++ b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c @@ -1503,7 +1503,7 @@ rel_resource_without_clip: return -EINVAL; } -cxgb4i_cplhandler_func cxgb4i_cplhandlers[NUM_CPL_CMDS] = { +static cxgb4i_cplhandler_func cxgb4i_cplhandlers[NUM_CPL_CMDS] = { [CPL_ACT_ESTABLISH] = do_act_establish, [CPL_ACT_OPEN_RPL] = do_act_open_rpl, [CPL_PEER_CLOSE] = do_peer_close, @@ -1519,7 +1519,7 @@ cxgb4i_cplhandler_func cxgb4i_cplhandlers[NUM_CPL_CMDS] = { [CPL_RX_DATA] = do_rx_data, }; -int cxgb4i_ofld_init(struct cxgbi_device *cdev) +static int cxgb4i_ofld_init(struct cxgbi_device *cdev) { int rc; -- cgit v0.10.2 From a1b43eddaec5a3fea55e1581caf217abda2d3147 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 21 Jul 2016 21:28:34 +0200 Subject: net/ncsi: avoid maybe-uninitialized warning gcc-4.9 and higher warn about the newly added NSCI code: net/ncsi/ncsi-manage.c: In function 'ncsi_process_next_channel': net/ncsi/ncsi-manage.c:1003:2: error: 'old_state' may be used uninitialized in this function [-Werror=maybe-uninitialized] The warning is a false positive and therefore harmless, but it would be good to avoid it anyway. I have determined that the barrier in the spin_unlock_irqsave() is what confuses gcc to the point that it cannot track whether the variable was unused or not. This rearranges the code in a way that makes it obvious to gcc that old_state is always initialized at the time of use, functionally this should not change anything. Signed-off-by: Arnd Bergmann Acked-by: Gavin Shan Signed-off-by: David S. Miller diff --git a/net/ncsi/ncsi-manage.c b/net/ncsi/ncsi-manage.c index d627a39..ef017b8 100644 --- a/net/ncsi/ncsi-manage.c +++ b/net/ncsi/ncsi-manage.c @@ -982,23 +982,18 @@ int ncsi_process_next_channel(struct ncsi_dev_priv *ndp) spin_lock_irqsave(&ndp->lock, flags); nc = list_first_or_null_rcu(&ndp->channel_queue, struct ncsi_channel, link); - if (nc) { - old_state = xchg(&nc->state, NCSI_CHANNEL_INVISIBLE); - list_del_init(&nc->link); + if (!nc) { + spin_unlock_irqrestore(&ndp->lock, flags); + goto out; } + + old_state = xchg(&nc->state, NCSI_CHANNEL_INVISIBLE); + list_del_init(&nc->link); + spin_unlock_irqrestore(&ndp->lock, flags); ndp->active_channel = nc; - ndp->active_package = nc ? nc->package : NULL; - if (!nc) { - if (ndp->flags & NCSI_DEV_RESHUFFLE) { - ndp->flags &= ~NCSI_DEV_RESHUFFLE; - return ncsi_choose_active_channel(ndp); - } - - ncsi_report_link(ndp, false); - return -ENODEV; - } + ndp->active_package = nc->package; switch (old_state) { case NCSI_CHANNEL_INACTIVE: @@ -1017,6 +1012,17 @@ int ncsi_process_next_channel(struct ncsi_dev_priv *ndp) } return 0; + +out: + ndp->active_channel = NULL; + ndp->active_package = NULL; + if (ndp->flags & NCSI_DEV_RESHUFFLE) { + ndp->flags &= ~NCSI_DEV_RESHUFFLE; + return ncsi_choose_active_channel(ndp); + } + + ncsi_report_link(ndp, false); + return -ENODEV; } #if IS_ENABLED(CONFIG_IPV6) -- cgit v0.10.2 From aa7145c16d6bf086538ad7eb20c807513bfa5efc Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 22 Jul 2016 01:19:42 +0200 Subject: bpf, events: fix offset in skb copy handler This patch fixes the __output_custom() routine we currently use with bpf_skb_copy(). I missed that when len is larger than the size of the current handle, we can issue multiple invocations of copy_func, and __output_custom() advances destination but also source buffer by the written amount of bytes. When we have __output_custom(), this is actually wrong since in that case the source buffer points to a non-linear object, in our case an skb, which the copy_func helper is supposed to walk. Therefore, since this is non-linear we thus need to pass the offset into the helper, so that copy_func can use it for extracting the data from the source object. Therefore, adjust the callback signatures properly and pass offset into the skb_header_pointer() invoked from bpf_skb_copy() callback. The __DEFINE_OUTPUT_COPY_BODY() is adjusted to accommodate for two things: i) to pass in whether we should advance source buffer or not; this is a compile-time constant condition, ii) to pass in the offset for __output_custom(), which we do with help of __VA_ARGS__, so everything can stay inlined as is currently. Both changes allow for adapting the __output_* fast-path helpers w/o extra overhead. Fixes: 555c8a8623a3 ("bpf: avoid stack copy and use skb ctx for event output") Fixes: 7e3f977edd0b ("perf, events: add non-linear data support for raw records") Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 36da074..1113423 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -211,7 +211,7 @@ bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *f const struct bpf_func_proto *bpf_get_trace_printk_proto(void); typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src, - unsigned long len); + unsigned long off, unsigned long len); u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy); diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index e79e6c6..15e55b7 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -70,7 +70,7 @@ struct perf_callchain_entry_ctx { }; typedef unsigned long (*perf_copy_f)(void *dst, const void *src, - unsigned long len); + unsigned long off, unsigned long len); struct perf_raw_frag { union { diff --git a/kernel/events/internal.h b/kernel/events/internal.h index 2417eb5..486fd78 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -123,18 +123,19 @@ static inline unsigned long perf_aux_size(struct ring_buffer *rb) return rb->aux_nr_pages << PAGE_SHIFT; } -#define __DEFINE_OUTPUT_COPY_BODY(memcpy_func) \ +#define __DEFINE_OUTPUT_COPY_BODY(advance_buf, memcpy_func, ...) \ { \ unsigned long size, written; \ \ do { \ size = min(handle->size, len); \ - written = memcpy_func(handle->addr, buf, size); \ + written = memcpy_func(__VA_ARGS__); \ written = size - written; \ \ len -= written; \ handle->addr += written; \ - buf += written; \ + if (advance_buf) \ + buf += written; \ handle->size -= written; \ if (!handle->size) { \ struct ring_buffer *rb = handle->rb; \ @@ -153,12 +154,16 @@ static inline unsigned long perf_aux_size(struct ring_buffer *rb) static inline unsigned long \ func_name(struct perf_output_handle *handle, \ const void *buf, unsigned long len) \ -__DEFINE_OUTPUT_COPY_BODY(memcpy_func) +__DEFINE_OUTPUT_COPY_BODY(true, memcpy_func, handle->addr, buf, size) static inline unsigned long __output_custom(struct perf_output_handle *handle, perf_copy_f copy_func, const void *buf, unsigned long len) -__DEFINE_OUTPUT_COPY_BODY(copy_func) +{ + unsigned long orig_len = len; + __DEFINE_OUTPUT_COPY_BODY(false, copy_func, handle->addr, buf, + orig_len - len, size) +} static inline unsigned long memcpy_common(void *dst, const void *src, unsigned long n) diff --git a/net/core/filter.c b/net/core/filter.c index 0b521353..5708999 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2026,9 +2026,9 @@ bool bpf_helper_changes_skb_data(void *func) } static unsigned long bpf_skb_copy(void *dst_buff, const void *skb, - unsigned long len) + unsigned long off, unsigned long len) { - void *ptr = skb_header_pointer(skb, 0, len, dst_buff); + void *ptr = skb_header_pointer(skb, off, len, dst_buff); if (unlikely(!ptr)) return len; -- cgit v0.10.2 From 9b97420228881e839b76c8a4506da3cb187bf004 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Fri, 22 Jul 2016 17:38:51 +0800 Subject: sctp: support ipv6 nonlocal bind This patch makes sctp support ipv6 nonlocal bind by adding sp->inet.freebind and net->ipv6.sysctl.ip_nonlocal_bind check in sctp_v6_available as what sctp did to support ipv4 nonlocal bind (commit cdac4e077489). Reported-by: Shijoe George Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index ae6f1a2..660c4a4 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -560,6 +560,7 @@ static int sctp_v6_is_any(const union sctp_addr *addr) static int sctp_v6_available(union sctp_addr *addr, struct sctp_sock *sp) { int type; + struct net *net = sock_net(&sp->inet.sk); const struct in6_addr *in6 = (const struct in6_addr *)&addr->v6.sin6_addr; type = ipv6_addr_type(in6); @@ -574,7 +575,8 @@ static int sctp_v6_available(union sctp_addr *addr, struct sctp_sock *sp) if (!(type & IPV6_ADDR_UNICAST)) return 0; - return ipv6_chk_addr(sock_net(&sp->inet.sk), in6, NULL, 0); + return sp->inet.freebind || net->ipv6.sysctl.ip_nonlocal_bind || + ipv6_chk_addr(net, in6, NULL, 0); } /* This function checks if the address is a valid address to be used for -- cgit v0.10.2 From baedbe55884c003819f5c8c063ec3d2569414296 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Fri, 22 Jul 2016 14:56:20 +0300 Subject: bridge: Fix incorrect re-injection of LLDP packets Commit 8626c56c8279 ("bridge: fix potential use-after-free when hook returns QUEUE or STOLEN verdict") caused LLDP packets arriving through a bridge port to be re-injected to the Rx path with skb->dev set to the bridge device, but this breaks the lldpad daemon. The lldpad daemon opens a packet socket with protocol set to ETH_P_LLDP for any valid device on the system, which doesn't not include soft devices such as bridge and VLAN. Since packet sockets (ptype_base) are processed in the Rx path after the Rx handler, LLDP packets with skb->dev set to the bridge device never reach the lldpad daemon. Fix this by making the bridge's Rx handler re-inject LLDP packets with RX_HANDLER_PASS, which effectively restores the behaviour prior to the mentioned commit. This means netfilter will never receive LLDP packets coming through a bridge port, as I don't see a way in which we can have okfn() consume the packet without breaking existing behaviour. I've already carried out a similar fix for STP packets in commit 56fae404fb2c ("bridge: Fix incorrect re-injection of STP packets"). Fixes: 8626c56c8279 ("bridge: fix potential use-after-free when hook returns QUEUE or STOLEN verdict") Signed-off-by: Ido Schimmel Reviewed-by: Jiri Pirko Cc: Florian Westphal Cc: John Fastabend Signed-off-by: David S. Miller diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 8b08eec..8e48620 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -283,6 +283,14 @@ rx_handler_result_t br_handle_frame(struct sk_buff **pskb) case 0x01: /* IEEE MAC (Pause) */ goto drop; + case 0x0E: /* 802.1AB LLDP */ + fwd_mask |= p->br->group_fwd_mask; + if (fwd_mask & (1u << dest[5])) + goto forward; + *pskb = skb; + __br_handle_local_finish(skb); + return RX_HANDLER_PASS; + default: /* Allow selective forwarding for most other protocols */ fwd_mask |= p->br->group_fwd_mask; -- cgit v0.10.2 From 2ccbe2cb79f2f74ab739252299b6f9ff27586f2c Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Fri, 22 Jul 2016 15:07:56 +0200 Subject: macsec: limit ICV length to 16 octets IEEE 802.1AE-2006 standard recommends that the ICV element in a MACsec frame should not exceed 16 octets: add MACSEC_STD_ICV_LEN in uapi definitions accordingly, and avoid accepting configurations where the ICV length exceeds the standard value. Leave definition of MACSEC_MAX_ICV_LEN unchanged for backwards compatibility with userspace programs. Fixes: dece8d2b78d1 ("uapi: add MACsec bits") Signed-off-by: Davide Caratti Signed-off-by: David S. Miller diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 0cbb935..18cfb46 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -510,7 +510,7 @@ static bool macsec_validate_skb(struct sk_buff *skb, u16 icv_len) } #define MACSEC_NEEDED_HEADROOM (macsec_extra_len(true)) -#define MACSEC_NEEDED_TAILROOM MACSEC_MAX_ICV_LEN +#define MACSEC_NEEDED_TAILROOM MACSEC_STD_ICV_LEN static void macsec_fill_iv(unsigned char *iv, sci_t sci, u32 pn) { @@ -3217,7 +3217,7 @@ static int macsec_validate_attr(struct nlattr *tb[], struct nlattr *data[]) case MACSEC_DEFAULT_CIPHER_ID: case MACSEC_DEFAULT_CIPHER_ALT: if (icv_len < MACSEC_MIN_ICV_LEN || - icv_len > MACSEC_MAX_ICV_LEN) + icv_len > MACSEC_STD_ICV_LEN) return -EINVAL; break; default: diff --git a/include/uapi/linux/if_macsec.h b/include/uapi/linux/if_macsec.h index f7d4831..02fc49c 100644 --- a/include/uapi/linux/if_macsec.h +++ b/include/uapi/linux/if_macsec.h @@ -26,6 +26,8 @@ #define MACSEC_MIN_ICV_LEN 8 #define MACSEC_MAX_ICV_LEN 32 +/* upper limit for ICV length as recommended by IEEE802.1AE-2006 */ +#define MACSEC_STD_ICV_LEN 16 enum macsec_attrs { MACSEC_ATTR_UNSPEC, -- cgit v0.10.2 From 34aedfee22967236adc3d9147c8b47b7f5bad26c Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Fri, 22 Jul 2016 15:07:57 +0200 Subject: macsec: fix error codes when a SA is created preserve the return value of AEAD functions that are called when a SA is created, to avoid inappropriate display of "RTNETLINK answers: Cannot allocate memory" message. Signed-off-by: Davide Caratti Signed-off-by: David S. Miller diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 18cfb46..0045108 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -1270,22 +1270,22 @@ static struct crypto_aead *macsec_alloc_tfm(char *key, int key_len, int icv_len) int ret; tfm = crypto_alloc_aead("gcm(aes)", 0, 0); - if (!tfm || IS_ERR(tfm)) - return NULL; + + if (IS_ERR(tfm)) + return tfm; ret = crypto_aead_setkey(tfm, key, key_len); - if (ret < 0) { - crypto_free_aead(tfm); - return NULL; - } + if (ret < 0) + goto fail; ret = crypto_aead_setauthsize(tfm, icv_len); - if (ret < 0) { - crypto_free_aead(tfm); - return NULL; - } + if (ret < 0) + goto fail; return tfm; +fail: + crypto_free_aead(tfm); + return ERR_PTR(ret); } static int init_rx_sa(struct macsec_rx_sa *rx_sa, char *sak, int key_len, @@ -1293,12 +1293,12 @@ static int init_rx_sa(struct macsec_rx_sa *rx_sa, char *sak, int key_len, { rx_sa->stats = alloc_percpu(struct macsec_rx_sa_stats); if (!rx_sa->stats) - return -1; + return -ENOMEM; rx_sa->key.tfm = macsec_alloc_tfm(sak, key_len, icv_len); - if (!rx_sa->key.tfm) { + if (IS_ERR(rx_sa->key.tfm)) { free_percpu(rx_sa->stats); - return -1; + return PTR_ERR(rx_sa->key.tfm); } rx_sa->active = false; @@ -1391,12 +1391,12 @@ static int init_tx_sa(struct macsec_tx_sa *tx_sa, char *sak, int key_len, { tx_sa->stats = alloc_percpu(struct macsec_tx_sa_stats); if (!tx_sa->stats) - return -1; + return -ENOMEM; tx_sa->key.tfm = macsec_alloc_tfm(sak, key_len, icv_len); - if (!tx_sa->key.tfm) { + if (IS_ERR(tx_sa->key.tfm)) { free_percpu(tx_sa->stats); - return -1; + return PTR_ERR(tx_sa->key.tfm); } tx_sa->active = false; @@ -1629,6 +1629,7 @@ static int macsec_add_rxsa(struct sk_buff *skb, struct genl_info *info) unsigned char assoc_num; struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1]; struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1]; + int err; if (!attrs[MACSEC_ATTR_IFINDEX]) return -EINVAL; @@ -1665,13 +1666,19 @@ static int macsec_add_rxsa(struct sk_buff *skb, struct genl_info *info) } rx_sa = kmalloc(sizeof(*rx_sa), GFP_KERNEL); - if (!rx_sa || init_rx_sa(rx_sa, nla_data(tb_sa[MACSEC_SA_ATTR_KEY]), - secy->key_len, secy->icv_len)) { - kfree(rx_sa); + if (!rx_sa) { rtnl_unlock(); return -ENOMEM; } + err = init_rx_sa(rx_sa, nla_data(tb_sa[MACSEC_SA_ATTR_KEY]), + secy->key_len, secy->icv_len); + if (err < 0) { + kfree(rx_sa); + rtnl_unlock(); + return err; + } + if (tb_sa[MACSEC_SA_ATTR_PN]) { spin_lock_bh(&rx_sa->lock); rx_sa->next_pn = nla_get_u32(tb_sa[MACSEC_SA_ATTR_PN]); @@ -1777,6 +1784,7 @@ static int macsec_add_txsa(struct sk_buff *skb, struct genl_info *info) struct macsec_tx_sa *tx_sa; unsigned char assoc_num; struct nlattr *tb_sa[MACSEC_SA_ATTR_MAX + 1]; + int err; if (!attrs[MACSEC_ATTR_IFINDEX]) return -EINVAL; @@ -1813,13 +1821,19 @@ static int macsec_add_txsa(struct sk_buff *skb, struct genl_info *info) } tx_sa = kmalloc(sizeof(*tx_sa), GFP_KERNEL); - if (!tx_sa || init_tx_sa(tx_sa, nla_data(tb_sa[MACSEC_SA_ATTR_KEY]), - secy->key_len, secy->icv_len)) { - kfree(tx_sa); + if (!tx_sa) { rtnl_unlock(); return -ENOMEM; } + err = init_tx_sa(tx_sa, nla_data(tb_sa[MACSEC_SA_ATTR_KEY]), + secy->key_len, secy->icv_len); + if (err < 0) { + kfree(tx_sa); + rtnl_unlock(); + return err; + } + nla_memcpy(tx_sa->key.id, tb_sa[MACSEC_SA_ATTR_KEYID], MACSEC_KEYID_LEN); spin_lock_bh(&tx_sa->lock); -- cgit v0.10.2 From f04c392d2dd97a985878f4380a1b054791301acf Mon Sep 17 00:00:00 2001 From: Davide Caratti Date: Fri, 22 Jul 2016 15:07:58 +0200 Subject: macsec: validate ICV length on link creation Test the cipher suite initialization in case ICV length has a value different than its default. If this test fails, creation of a new macsec link will also fail. This avoids situations where further security associations can't be added due to failures of crypto_aead_setauthsize(), caused by unsupported user-provided values of the ICV length. Signed-off-by: Davide Caratti Signed-off-by: David S. Miller diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 0045108..d8b2b49 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -3224,8 +3224,20 @@ static int macsec_validate_attr(struct nlattr *tb[], struct nlattr *data[]) if (data[IFLA_MACSEC_CIPHER_SUITE]) csid = nla_get_u64(data[IFLA_MACSEC_CIPHER_SUITE]); - if (data[IFLA_MACSEC_ICV_LEN]) + if (data[IFLA_MACSEC_ICV_LEN]) { icv_len = nla_get_u8(data[IFLA_MACSEC_ICV_LEN]); + if (icv_len != DEFAULT_ICV_LEN) { + char dummy_key[DEFAULT_SAK_LEN] = { 0 }; + struct crypto_aead *dummy_tfm; + + dummy_tfm = macsec_alloc_tfm(dummy_key, + DEFAULT_SAK_LEN, + icv_len); + if (IS_ERR(dummy_tfm)) + return PTR_ERR(dummy_tfm); + crypto_free_aead(dummy_tfm); + } + } switch (csid) { case MACSEC_DEFAULT_CIPHER_ID: -- cgit v0.10.2 From fd2d180a28cb5075163945d0b229926ec9782ab0 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Fri, 22 Jul 2016 21:25:42 +0800 Subject: sctp: use inet_recvmsg to support sctp RFS well Commit 486bdee0134c ("sctp: add support for RPS and RFS") saves skb->hash into sk->sk_rxhash so that the inet_* can record it to flow table. But sctp uses sock_common_recvmsg as .recvmsg instead of inet_recvmsg, sock_common_recvmsg doesn't invoke sock_rps_record_flow to record the flow. It may cause that the receiver has no chances to record the flow if it doesn't send msg or poll the socket. So this patch fixes it by using inet_recvmsg as .recvmsg in sctp. Fixes: 486bdee0134c ("sctp: add support for RPS and RFS") Signed-off-by: Xin Long Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 660c4a4..f473779 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -956,7 +956,7 @@ static const struct proto_ops inet6_seqpacket_ops = { .setsockopt = sock_common_setsockopt, .getsockopt = sock_common_getsockopt, .sendmsg = inet_sendmsg, - .recvmsg = sock_common_recvmsg, + .recvmsg = inet_recvmsg, .mmap = sock_no_mmap, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_sock_common_setsockopt, diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 1adb927..7b523e3 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1028,7 +1028,7 @@ static const struct proto_ops inet_seqpacket_ops = { .setsockopt = sock_common_setsockopt, /* IP_SOL IP_OPTION is a problem */ .getsockopt = sock_common_getsockopt, .sendmsg = inet_sendmsg, - .recvmsg = sock_common_recvmsg, + .recvmsg = inet_recvmsg, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage, #ifdef CONFIG_COMPAT -- cgit v0.10.2 From 15657841bd5bb23dac7120c4e2e181b390a5ec12 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Fri, 22 Jul 2016 14:05:32 +0000 Subject: libcxgb: remove unused including Remove including that don't need it. Signed-off-by: Wei Yongjun Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c index 551a365..0ed1616 100644 --- a/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c +++ b/drivers/net/ethernet/chelsio/libcxgb/libcxgb_ppm.c @@ -39,7 +39,6 @@ #define pr_fmt(fmt) DRV_NAME ": " fmt #include -#include #include #include #include -- cgit v0.10.2 From ea06f7176413e2538d13bb85b65387d0917943d9 Mon Sep 17 00:00:00 2001 From: Mike Manning Date: Fri, 22 Jul 2016 18:32:11 +0100 Subject: net: ipv6: Always leave anycast and multicast groups on link down Default kernel behavior is to delete IPv6 addresses on link down, which entails deletion of the multicast and the subnet-router anycast addresses. These deletions do not happen with sysctl setting to keep global IPv6 addresses on link down, so every link down/up causes an increment of the anycast and multicast refcounts. These bogus refcounts may stop these addrs from being removed on subsequent calls to delete them. The solution is to leave the groups for the multicast and subnet anycast on link down for the callflow when global IPv6 addresses are kept. Fixes: f1705ec197e7 ("net: ipv6: Make address flushing on ifdown optional") Signed-off-by: Mike Manning Acked-by: David Ahern Signed-off-by: David S. Miller diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 24f1b08..6287a8b 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3636,6 +3636,10 @@ restart: if (state != INET6_IFADDR_STATE_DEAD) { __ipv6_ifa_notify(RTM_DELADDR, ifa); inet6addr_notifier_call_chain(NETDEV_DOWN, ifa); + } else { + if (idev->cnf.forwarding) + addrconf_leave_anycast(ifa); + addrconf_leave_solict(ifa->idev, &ifa->addr); } write_lock_bh(&idev->lock); -- cgit v0.10.2 From 0a58f474928cbace609fb563295ecb32491b1c4a Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Fri, 22 Jul 2016 19:04:12 +0100 Subject: kcm: remove redundant -ve error check and return path The check for a -ve error is redundant, remove it and just immediately return the return value from the call to seq_open_net. Signed-off-by: Colin Ian King Signed-off-by: David S. Miller diff --git a/net/kcm/kcmproc.c b/net/kcm/kcmproc.c index fda7f47..16c2e03 100644 --- a/net/kcm/kcmproc.c +++ b/net/kcm/kcmproc.c @@ -88,13 +88,9 @@ struct kcm_proc_mux_state { static int kcm_seq_open(struct inode *inode, struct file *file) { struct kcm_seq_muxinfo *muxinfo = PDE_DATA(inode); - int err; - err = seq_open_net(inode, file, &muxinfo->seq_ops, + return seq_open_net(inode, file, &muxinfo->seq_ops, sizeof(struct kcm_proc_mux_state)); - if (err < 0) - return err; - return err; } static void kcm_format_mux_header(struct seq_file *seq) -- cgit v0.10.2 From e2b9f1f7af1dfe20df8e68849ebb4bbafed5727a Mon Sep 17 00:00:00 2001 From: Haiyang Zhang Date: Fri, 22 Jul 2016 18:14:50 -0700 Subject: hv_netvsc: Fix VF register on bonding devices Added a condition to avoid bonding devices with same MAC registering as VF. Signed-off-by: Haiyang Zhang Reviewed-by: K. Y. Srinivasan Signed-off-by: David S. Miller diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 787a202..41bd952 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -1494,8 +1494,8 @@ static int netvsc_netdev_event(struct notifier_block *this, { struct net_device *event_dev = netdev_notifier_info_to_dev(ptr); - /* Avoid Vlan dev with same MAC registering as VF */ - if (event_dev->priv_flags & IFF_802_1Q_VLAN) + /* Avoid Vlan, Bonding dev with same MAC registering as VF */ + if (event_dev->priv_flags & (IFF_802_1Q_VLAN | IFF_BONDING)) return NOTIFY_DONE; switch (event) { -- cgit v0.10.2 From eefc1b1d105ee4d2ce907833ce675f1e9599b5e3 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Sat, 23 Jul 2016 00:32:48 -0300 Subject: sctp: fix BH handling on socket backlog Now that the backlog processing is called with BH enabled, we have to disable BH before taking the socket lock via bh_lock_sock() otherwise it may dead lock: sctp_backlog_rcv() bh_lock_sock(sk); if (sock_owned_by_user(sk)) { if (sk_add_backlog(sk, skb, sk->sk_rcvbuf)) sctp_chunk_free(chunk); else backloged = 1; } else sctp_inq_push(inqueue, chunk); bh_unlock_sock(sk); while sctp_inq_push() was disabling/enabling BH, but enabling BH triggers pending softirq, which then may try to re-lock the socket in sctp_rcv(). [ 219.187215] [ 219.187217] [] _raw_spin_lock+0x20/0x30 [ 219.187223] [] sctp_rcv+0x48c/0xba0 [sctp] [ 219.187225] [] ? nf_iterate+0x62/0x80 [ 219.187226] [] ip_local_deliver_finish+0x94/0x1e0 [ 219.187228] [] ip_local_deliver+0x6f/0xf0 [ 219.187229] [] ? ip_rcv_finish+0x3b0/0x3b0 [ 219.187230] [] ip_rcv_finish+0xd8/0x3b0 [ 219.187232] [] ip_rcv+0x282/0x3a0 [ 219.187233] [] ? update_curr+0x66/0x180 [ 219.187235] [] __netif_receive_skb_core+0x524/0xa90 [ 219.187236] [] ? update_cfs_shares+0x30/0xf0 [ 219.187237] [] ? __enqueue_entity+0x6c/0x70 [ 219.187239] [] ? enqueue_entity+0x204/0xdf0 [ 219.187240] [] __netif_receive_skb+0x18/0x60 [ 219.187242] [] process_backlog+0x9e/0x140 [ 219.187243] [] net_rx_action+0x22c/0x370 [ 219.187245] [] __do_softirq+0x112/0x2e7 [ 219.187247] [] do_softirq_own_stack+0x1c/0x30 [ 219.187247] [ 219.187248] [] do_softirq.part.14+0x38/0x40 [ 219.187249] [] __local_bh_enable_ip+0x7d/0x80 [ 219.187254] [] sctp_inq_push+0x68/0x80 [sctp] [ 219.187258] [] sctp_backlog_rcv+0x151/0x1c0 [sctp] [ 219.187260] [] __release_sock+0x87/0xf0 [ 219.187261] [] release_sock+0x30/0xa0 [ 219.187265] [] sctp_accept+0x17d/0x210 [sctp] [ 219.187266] [] ? prepare_to_wait_event+0xf0/0xf0 [ 219.187268] [] inet_accept+0x3c/0x130 [ 219.187269] [] SYSC_accept4+0x103/0x210 [ 219.187271] [] ? _raw_spin_unlock_bh+0x1a/0x20 [ 219.187272] [] ? release_sock+0x8c/0xa0 [ 219.187276] [] ? sctp_inet_listen+0x62/0x1b0 [sctp] [ 219.187277] [] SyS_accept+0x10/0x20 Fixes: 860fbbc343bf ("sctp: prepare for socket backlog behavior change") Cc: Eric Dumazet Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/input.c b/net/sctp/input.c index 30d72f7..c182db7 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -321,6 +321,7 @@ int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb) */ sk = rcvr->sk; + local_bh_disable(); bh_lock_sock(sk); if (sock_owned_by_user(sk)) { @@ -332,6 +333,7 @@ int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb) sctp_inq_push(inqueue, chunk); bh_unlock_sock(sk); + local_bh_enable(); /* If the chunk was backloged again, don't drop refs */ if (backloged) diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index 9427706..c30ddb0 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -89,12 +89,10 @@ void sctp_inq_push(struct sctp_inq *q, struct sctp_chunk *chunk) * Eventually, we should clean up inqueue to not rely * on the BH related data structures. */ - local_bh_disable(); list_add_tail(&chunk->list, &q->in_chunk_list); if (chunk->asoc) chunk->asoc->stats.ipackets++; q->immediate.func(&q->immediate); - local_bh_enable(); } /* Peek at the next chunk on the inqeue. */ -- cgit v0.10.2 From 52253db924d1480bf2543afbb9551de31381aab9 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Sat, 23 Jul 2016 00:33:44 -0300 Subject: sctp: also point GSO head_skb to the sk when it's available The head skb for GSO packets won't travel through the inner depths of SCTP stack as it doesn't contain any chunks on it. That means skb->sk doesn't get set and then when sctp_recvmsg() calls sctp_inet6_skb_msgname() on the head_skb it panics, as this last needs to check flags at the socket (sp->v4mapped). The fix is to initialize skb->sk for th head skb once we are able to do it. That is, when the first chunk is processed. Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c index f6219b1..1bc4f71 100644 --- a/net/sctp/ulpevent.c +++ b/net/sctp/ulpevent.c @@ -91,6 +91,7 @@ int sctp_ulpevent_is_notification(const struct sctp_ulpevent *event) static inline void sctp_ulpevent_set_owner(struct sctp_ulpevent *event, const struct sctp_association *asoc) { + struct sctp_chunk *chunk = event->chunk; struct sk_buff *skb; /* Cast away the const, as we are just wanting to @@ -101,6 +102,8 @@ static inline void sctp_ulpevent_set_owner(struct sctp_ulpevent *event, event->asoc = (struct sctp_association *)asoc; atomic_add(event->rmem_len, &event->asoc->rmem_alloc); sctp_skb_set_owner_r(skb, asoc->base.sk); + if (chunk && chunk->head_skb && !chunk->head_skb->sk) + chunk->head_skb->sk = asoc->base.sk; } /* A simple destructor to give up the reference to the association. */ -- cgit v0.10.2 From d3e6952cfb7ba5f4bfa29d4803ba91f96ce1204d Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Sat, 23 Jul 2016 07:43:50 +0200 Subject: net/irda: fix NULL pointer dereference on memory allocation failure I ran into this: kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 2 PID: 2012 Comm: trinity-c3 Not tainted 4.7.0-rc7+ #19 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Ubuntu-1.8.2-1ubuntu1 04/01/2014 task: ffff8800b745f2c0 ti: ffff880111740000 task.ti: ffff880111740000 RIP: 0010:[] [] irttp_connect_request+0x36/0x710 RSP: 0018:ffff880111747bb8 EFLAGS: 00010286 RAX: dffffc0000000000 RBX: 0000000000000000 RCX: 0000000069dd8358 RDX: 0000000000000009 RSI: 0000000000000027 RDI: 0000000000000048 RBP: ffff880111747c00 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000069dd8358 R11: 1ffffffff0759723 R12: 0000000000000000 R13: ffff88011a7e4780 R14: 0000000000000027 R15: 0000000000000000 FS: 00007fc738404700(0000) GS:ffff88011af00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fc737fdfb10 CR3: 0000000118087000 CR4: 00000000000006e0 Stack: 0000000000000200 ffff880111747bd8 ffffffff810ee611 ffff880119f1f220 ffff880119f1f4f8 ffff880119f1f4f0 ffff88011a7e4780 ffff880119f1f232 ffff880119f1f220 ffff880111747d58 ffffffff82bca542 0000000000000000 Call Trace: [] irda_connect+0x562/0x1190 [] SYSC_connect+0x202/0x2a0 [] SyS_connect+0x9/0x10 [] do_syscall_64+0x19c/0x410 [] entry_SYSCALL64_slow_path+0x25/0x25 Code: 41 89 ca 48 89 e5 41 57 41 56 41 55 41 54 41 89 d7 53 48 89 fb 48 83 c7 48 48 89 fa 41 89 f6 48 c1 ea 03 48 83 ec 20 4c 8b 65 10 <0f> b6 04 02 84 c0 74 08 84 c0 0f 8e 4c 04 00 00 80 7b 48 00 74 RIP [] irttp_connect_request+0x36/0x710 RSP ---[ end trace 4cda2588bc055b30 ]--- The problem is that irda_open_tsap() can fail and leave self->tsap = NULL, and then irttp_connect_request() almost immediately dereferences it. Cc: stable@vger.kernel.org Signed-off-by: Vegard Nossum Signed-off-by: David S. Miller diff --git a/net/irda/af_irda.c b/net/irda/af_irda.c index 923abd6..8d2f7c9 100644 --- a/net/irda/af_irda.c +++ b/net/irda/af_irda.c @@ -1024,8 +1024,11 @@ static int irda_connect(struct socket *sock, struct sockaddr *uaddr, } /* Check if we have opened a local TSAP */ - if (!self->tsap) - irda_open_tsap(self, LSAP_ANY, addr->sir_name); + if (!self->tsap) { + err = irda_open_tsap(self, LSAP_ANY, addr->sir_name); + if (err) + goto out; + } /* Move to connecting socket, start sending Connect Requests */ sock->state = SS_CONNECTING; -- cgit v0.10.2 From 5fc382d87517707ad77ea4c9c12e2a3fde2c838a Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Sat, 23 Jul 2016 09:42:35 +0200 Subject: net/sctp: terminate rhashtable walk correctly I was seeing a lot of these: BUG: sleeping function called from invalid context at mm/slab.h:388 in_atomic(): 0, irqs_disabled(): 0, pid: 14971, name: trinity-c2 Preemption disabled at:[] rhashtable_walk_start+0x46/0x150 [] preempt_count_add+0x1fb/0x280 [] _raw_spin_lock+0x12/0x40 [] console_unlock+0x2f7/0x930 [] vprintk_emit+0x2fb/0x520 [] vprintk_default+0x1a/0x20 [] printk+0x94/0xb0 [] print_stack_trace+0xe0/0x170 [] ___might_sleep+0x3be/0x460 [] __might_sleep+0x90/0x1a0 [] kmem_cache_alloc+0x153/0x1e0 [] rhashtable_walk_init+0xfe/0x2d0 [] sctp_transport_walk_start+0x1e/0x60 [] sctp_transport_seq_start+0x4d/0x150 [] seq_read+0x27b/0x1180 [] proc_reg_read+0xbc/0x180 [] __vfs_read+0xdb/0x610 [] vfs_read+0xea/0x2d0 [] SyS_pread64+0x11b/0x150 [] do_syscall_64+0x19c/0x410 [] return_from_SYSCALL_64+0x0/0x6a [] 0xffffffffffffffff Apparently we always need to call rhashtable_walk_stop(), even when rhashtable_walk_start() fails: * rhashtable_walk_start - Start a hash table walk * @iter: Hash table iterator * * Start a hash table walk. Note that we take the RCU lock in all * cases including when we return an error. So you must always call * rhashtable_walk_stop to clean up. otherwise we never call rcu_read_unlock() and we get the splat above. Fixes: 53fa1036 ("sctp: fix some rhashtable functions using in sctp proc/diag") See-also: 53fa1036 ("sctp: fix some rhashtable functions using in sctp proc/diag") See-also: f2dba9c6 ("rhashtable: Introduce rhashtable_walk_*") Cc: Xin Long Cc: Herbert Xu Cc: stable@vger.kernel.org Signed-off-by: Vegard Nossum Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller diff --git a/net/sctp/socket.c b/net/sctp/socket.c index d2681cb..8812e1b 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4393,6 +4393,7 @@ int sctp_transport_walk_start(struct rhashtable_iter *iter) err = rhashtable_walk_start(iter); if (err && err != -EAGAIN) { + rhashtable_walk_stop(iter); rhashtable_walk_exit(iter); return err; } -- cgit v0.10.2 From ae76715d153e33c249b6850361e4d8d775388b5a Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Sun, 24 Jul 2016 16:12:39 +0300 Subject: net/mlx5e: Check the minimum inline header mode before xmit Each send queue (SQ) has inline mode that defines the minimal required inline headers in the SQ WQE. Before sending each packet check that the minimum required headers on the WQE are copied. Signed-off-by: Hadar Hen Zion Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 4cbd452..2c20c7b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -398,6 +398,7 @@ struct mlx5e_sq { u32 sqn; u16 bf_buf_size; u16 max_inline; + u8 min_inline_mode; u16 edge; struct device *pdev; struct mlx5e_tstamp *tstamp; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 5740b46..e073bf59 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -128,6 +128,50 @@ u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, return priv->channeltc_to_txq_map[channel_ix][up]; } +static inline int mlx5e_skb_l2_header_offset(struct sk_buff *skb) +{ +#define MLX5E_MIN_INLINE (ETH_HLEN + VLAN_HLEN) + + return max(skb_network_offset(skb), MLX5E_MIN_INLINE); +} + +static inline int mlx5e_skb_l3_header_offset(struct sk_buff *skb) +{ + struct flow_keys keys; + + if (skb_transport_header_was_set(skb)) + return skb_transport_offset(skb); + else if (skb_flow_dissect_flow_keys(skb, &keys, 0)) + return keys.control.thoff; + else + return mlx5e_skb_l2_header_offset(skb); +} + +static inline unsigned int mlx5e_calc_min_inline(enum mlx5_inline_modes mode, + struct sk_buff *skb) +{ + int hlen; + + switch (mode) { + case MLX5_INLINE_MODE_TCP_UDP: + hlen = eth_get_headlen(skb->data, skb_headlen(skb)); + if (hlen == ETH_HLEN && !skb_vlan_tag_present(skb)) + hlen += VLAN_HLEN; + return hlen; + case MLX5_INLINE_MODE_IP: + /* When transport header is set to zero, it means no transport + * header. When transport header is set to 0xff's, it means + * transport header wasn't set. + */ + if (skb_transport_offset(skb)) + return mlx5e_skb_l3_header_offset(skb); + /* fall through */ + case MLX5_INLINE_MODE_L2: + default: + return mlx5e_skb_l2_header_offset(skb); + } +} + static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq, struct sk_buff *skb, bool bf) { @@ -135,8 +179,6 @@ static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq, * headers and occur before the data gather. * Therefore these headers must be copied into the WQE */ -#define MLX5E_MIN_INLINE (ETH_HLEN + VLAN_HLEN) - if (bf) { u16 ihs = skb_headlen(skb); @@ -146,8 +188,7 @@ static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq, if (ihs <= sq->max_inline) return skb_headlen(skb); } - - return max(skb_network_offset(skb), MLX5E_MIN_INLINE); + return mlx5e_calc_min_inline(sq->min_inline_mode, skb); } static inline void mlx5e_tx_skb_pull_inline(unsigned char **skb_data, diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index e0a3ed7..0b6d15c 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -129,6 +129,13 @@ __mlx5_mask(typ, fld)) tmp; \ }) +enum mlx5_inline_modes { + MLX5_INLINE_MODE_NONE, + MLX5_INLINE_MODE_L2, + MLX5_INLINE_MODE_IP, + MLX5_INLINE_MODE_TCP_UDP, +}; + enum { MLX5_MAX_COMMANDS = 32, MLX5_CMD_DATA_BLOCK_SIZE = 512, -- cgit v0.10.2 From cff92d7c7ebd7ceddd4def6b39e0302585b1eb14 Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Sun, 24 Jul 2016 16:12:40 +0300 Subject: net/mlx5e: Query minimum required header copy during xmit Add support for query the minimum inline mode from the Firmware. It is required for correct TX steering according to L3/L4 packet headers. Each send queue (SQ) has inline mode that defines the minimal required headers that needs to be copied into the SQ WQE. The driver asks the Firmware for the wqe_inline_mode device capability value. In case the device capability defined as "vport context" the driver must check the reported min inline mode from the vport context before creating its SQs. Signed-off-by: Hadar Hen Zion Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 2c20c7b..1b495ef 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -129,6 +129,12 @@ static inline int mlx5_max_log_rq_size(int wq_type) } } +enum { + MLX5E_INLINE_MODE_L2, + MLX5E_INLINE_MODE_VPORT_CONTEXT, + MLX5_INLINE_MODE_NOT_REQUIRED, +}; + struct mlx5e_tx_wqe { struct mlx5_wqe_ctrl_seg ctrl; struct mlx5_wqe_eth_seg eth; @@ -188,6 +194,7 @@ struct mlx5e_params { bool lro_en; u32 lro_wqe_sz; u16 tx_max_inline; + u8 tx_min_inline_mode; u8 rss_hfunc; u8 toeplitz_hash_key[40]; u32 indirection_rqt[MLX5E_INDIR_RQT_SIZE]; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index bdcb699..870bea3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -56,6 +56,7 @@ struct mlx5e_sq_param { u32 sqc[MLX5_ST_SZ_DW(sqc)]; struct mlx5_wq_param wq; u16 max_inline; + u8 min_inline_mode; bool icosq; }; @@ -649,6 +650,9 @@ static int mlx5e_create_sq(struct mlx5e_channel *c, } sq->bf_buf_size = (1 << MLX5_CAP_GEN(mdev, log_bf_reg_size)) / 2; sq->max_inline = param->max_inline; + sq->min_inline_mode = + MLX5_CAP_ETH(mdev, wqe_inline_mode) == MLX5E_INLINE_MODE_VPORT_CONTEXT ? + param->min_inline_mode : 0; err = mlx5e_alloc_sq_db(sq, cpu_to_node(c->cpu)); if (err) @@ -731,6 +735,7 @@ static int mlx5e_enable_sq(struct mlx5e_sq *sq, struct mlx5e_sq_param *param) MLX5_SET(sqc, sqc, tis_num_0, param->icosq ? 0 : priv->tisn[sq->tc]); MLX5_SET(sqc, sqc, cqn, sq->cq.mcq.cqn); + MLX5_SET(sqc, sqc, min_wqe_inline_mode, sq->min_inline_mode); MLX5_SET(sqc, sqc, state, MLX5_SQC_STATE_RST); MLX5_SET(sqc, sqc, tis_lst_sz, param->icosq ? 0 : 1); MLX5_SET(sqc, sqc, flush_in_error_en, 1); @@ -1343,6 +1348,7 @@ static void mlx5e_build_sq_param(struct mlx5e_priv *priv, MLX5_SET(wq, wq, log_wq_sz, priv->params.log_sq_size); param->max_inline = priv->params.tx_max_inline; + param->min_inline_mode = priv->params.tx_min_inline_mode; } static void mlx5e_build_common_cq_param(struct mlx5e_priv *priv, @@ -2978,6 +2984,23 @@ void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC_FROM_CQE; } +static void mlx5e_query_min_inline(struct mlx5_core_dev *mdev, + u8 *min_inline_mode) +{ + switch (MLX5_CAP_ETH(mdev, wqe_inline_mode)) { + case MLX5E_INLINE_MODE_L2: + *min_inline_mode = MLX5_INLINE_MODE_L2; + break; + case MLX5E_INLINE_MODE_VPORT_CONTEXT: + mlx5_query_nic_vport_min_inline(mdev, + min_inline_mode); + break; + case MLX5_INLINE_MODE_NOT_REQUIRED: + *min_inline_mode = MLX5_INLINE_MODE_NONE; + break; + } +} + static void mlx5e_build_nic_netdev_priv(struct mlx5_core_dev *mdev, struct net_device *netdev, const struct mlx5e_profile *profile, @@ -3043,6 +3066,7 @@ static void mlx5e_build_nic_netdev_priv(struct mlx5_core_dev *mdev, priv->params.tx_cq_moderation.pkts = MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_PKTS; priv->params.tx_max_inline = mlx5e_get_max_inline_cap(mdev); + mlx5e_query_min_inline(mdev, &priv->params.tx_min_inline_mode); priv->params.num_tc = 1; priv->params.rss_hfunc = ETH_RSS_HASH_XOR; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index 91846df..21365d0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -135,6 +135,18 @@ static int mlx5_modify_nic_vport_context(struct mlx5_core_dev *mdev, void *in, return mlx5_cmd_exec_check_status(mdev, in, inlen, out, sizeof(out)); } +void mlx5_query_nic_vport_min_inline(struct mlx5_core_dev *mdev, + u8 *min_inline_mode) +{ + u32 out[MLX5_ST_SZ_DW(query_nic_vport_context_out)] = {0}; + + mlx5_query_nic_vport_context(mdev, 0, out, sizeof(out)); + + *min_inline_mode = MLX5_GET(query_nic_vport_context_out, out, + nic_vport_context.min_wqe_inline_mode); +} +EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_min_inline); + int mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, u16 vport, u8 *addr) { diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index d671e4e..21bc455 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -536,7 +536,8 @@ struct mlx5_ifc_per_protocol_networking_offload_caps_bits { u8 self_lb_en_modifiable[0x1]; u8 reserved_at_9[0x2]; u8 max_lso_cap[0x5]; - u8 reserved_at_10[0x4]; + u8 reserved_at_10[0x2]; + u8 wqe_inline_mode[0x2]; u8 rss_ind_tbl_cap[0x4]; u8 reg_umr_sq[0x1]; u8 scatter_fcs[0x1]; @@ -2270,7 +2271,8 @@ struct mlx5_ifc_sqc_bits { u8 cd_master[0x1]; u8 fre[0x1]; u8 flush_in_error_en[0x1]; - u8 reserved_at_4[0x4]; + u8 reserved_at_4[0x1]; + u8 min_wqe_inline_mode[0x3]; u8 state[0x4]; u8 reg_umr[0x1]; u8 reserved_at_d[0x13]; @@ -2367,7 +2369,9 @@ struct mlx5_ifc_rmpc_bits { }; struct mlx5_ifc_nic_vport_context_bits { - u8 reserved_at_0[0x1f]; + u8 reserved_at_0[0x5]; + u8 min_wqe_inline_mode[0x3]; + u8 reserved_at_8[0x17]; u8 roce_en[0x1]; u8 arm_change_event[0x1]; diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h index 6c16c19..e087b7d 100644 --- a/include/linux/mlx5/vport.h +++ b/include/linux/mlx5/vport.h @@ -43,6 +43,8 @@ int mlx5_modify_vport_admin_state(struct mlx5_core_dev *mdev, u8 opmod, u16 vport, u8 state); int mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, u16 vport, u8 *addr); +void mlx5_query_nic_vport_min_inline(struct mlx5_core_dev *mdev, + u8 *min_inline); int mlx5_modify_nic_vport_mac_address(struct mlx5_core_dev *dev, u16 vport, u8 *addr); int mlx5_query_nic_vport_mtu(struct mlx5_core_dev *mdev, u16 *mtu); -- cgit v0.10.2 From 9b8ac4f9dd60ac7375fb0e221dbf596db4c4e622 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Sun, 24 Jul 2016 19:24:09 +0100 Subject: gtp: #define #define _GTP_H_ and not #define _GTP_H Fix clang build warning: ./include/net/gtp.h:1:9: warning: '_GTP_H_' is used as a header guard here, followed by #define of a different macro [-Wheader-guard] fix by defining _GTP_H_ and not _GTP_H Signed-off-by: Colin Ian King Signed-off-by: David S. Miller diff --git a/include/net/gtp.h b/include/net/gtp.h index 894a37b..6398891 100644 --- a/include/net/gtp.h +++ b/include/net/gtp.h @@ -1,5 +1,5 @@ #ifndef _GTP_H_ -#define _GTP_H +#define _GTP_H_ /* General GTP protocol related definitions. */ -- cgit v0.10.2 From 3568bdf0419fcaeebc5bba7cb2e034436b3e4125 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Mon, 25 Jul 2016 11:54:45 +0200 Subject: net: davinci_cpdma: remove excessive dump of register values to kernel log MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Such a big dump of register values is hardly useful on a production system. Another downside of the now removed functions is that calling emac_dump_regs resulted in at least 87 calls to dev_info while holding a spinlock and having irqs off which is a big source of latency. Signed-off-by: Uwe Kleine-König Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c index 1c653ca..73638f7 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.c +++ b/drivers/net/ethernet/ti/davinci_cpdma.c @@ -97,8 +97,6 @@ enum cpdma_state { CPDMA_STATE_TEARDOWN, }; -static const char *cpdma_state_str[] = { "idle", "active", "teardown" }; - struct cpdma_ctlr { enum cpdma_state state; struct cpdma_params params; @@ -357,77 +355,6 @@ int cpdma_ctlr_stop(struct cpdma_ctlr *ctlr) } EXPORT_SYMBOL_GPL(cpdma_ctlr_stop); -int cpdma_ctlr_dump(struct cpdma_ctlr *ctlr) -{ - struct device *dev = ctlr->dev; - unsigned long flags; - int i; - - spin_lock_irqsave(&ctlr->lock, flags); - - dev_info(dev, "CPDMA: state: %s", cpdma_state_str[ctlr->state]); - - dev_info(dev, "CPDMA: txidver: %x", - dma_reg_read(ctlr, CPDMA_TXIDVER)); - dev_info(dev, "CPDMA: txcontrol: %x", - dma_reg_read(ctlr, CPDMA_TXCONTROL)); - dev_info(dev, "CPDMA: txteardown: %x", - dma_reg_read(ctlr, CPDMA_TXTEARDOWN)); - dev_info(dev, "CPDMA: rxidver: %x", - dma_reg_read(ctlr, CPDMA_RXIDVER)); - dev_info(dev, "CPDMA: rxcontrol: %x", - dma_reg_read(ctlr, CPDMA_RXCONTROL)); - dev_info(dev, "CPDMA: softreset: %x", - dma_reg_read(ctlr, CPDMA_SOFTRESET)); - dev_info(dev, "CPDMA: rxteardown: %x", - dma_reg_read(ctlr, CPDMA_RXTEARDOWN)); - dev_info(dev, "CPDMA: txintstatraw: %x", - dma_reg_read(ctlr, CPDMA_TXINTSTATRAW)); - dev_info(dev, "CPDMA: txintstatmasked: %x", - dma_reg_read(ctlr, CPDMA_TXINTSTATMASKED)); - dev_info(dev, "CPDMA: txintmaskset: %x", - dma_reg_read(ctlr, CPDMA_TXINTMASKSET)); - dev_info(dev, "CPDMA: txintmaskclear: %x", - dma_reg_read(ctlr, CPDMA_TXINTMASKCLEAR)); - dev_info(dev, "CPDMA: macinvector: %x", - dma_reg_read(ctlr, CPDMA_MACINVECTOR)); - dev_info(dev, "CPDMA: maceoivector: %x", - dma_reg_read(ctlr, CPDMA_MACEOIVECTOR)); - dev_info(dev, "CPDMA: rxintstatraw: %x", - dma_reg_read(ctlr, CPDMA_RXINTSTATRAW)); - dev_info(dev, "CPDMA: rxintstatmasked: %x", - dma_reg_read(ctlr, CPDMA_RXINTSTATMASKED)); - dev_info(dev, "CPDMA: rxintmaskset: %x", - dma_reg_read(ctlr, CPDMA_RXINTMASKSET)); - dev_info(dev, "CPDMA: rxintmaskclear: %x", - dma_reg_read(ctlr, CPDMA_RXINTMASKCLEAR)); - dev_info(dev, "CPDMA: dmaintstatraw: %x", - dma_reg_read(ctlr, CPDMA_DMAINTSTATRAW)); - dev_info(dev, "CPDMA: dmaintstatmasked: %x", - dma_reg_read(ctlr, CPDMA_DMAINTSTATMASKED)); - dev_info(dev, "CPDMA: dmaintmaskset: %x", - dma_reg_read(ctlr, CPDMA_DMAINTMASKSET)); - dev_info(dev, "CPDMA: dmaintmaskclear: %x", - dma_reg_read(ctlr, CPDMA_DMAINTMASKCLEAR)); - - if (!ctlr->params.has_ext_regs) { - dev_info(dev, "CPDMA: dmacontrol: %x", - dma_reg_read(ctlr, CPDMA_DMACONTROL)); - dev_info(dev, "CPDMA: dmastatus: %x", - dma_reg_read(ctlr, CPDMA_DMASTATUS)); - dev_info(dev, "CPDMA: rxbuffofs: %x", - dma_reg_read(ctlr, CPDMA_RXBUFFOFS)); - } - - for (i = 0; i < ARRAY_SIZE(ctlr->channels); i++) - if (ctlr->channels[i]) - cpdma_chan_dump(ctlr->channels[i]); - - spin_unlock_irqrestore(&ctlr->lock, flags); - return 0; -} -EXPORT_SYMBOL_GPL(cpdma_ctlr_dump); - int cpdma_ctlr_destroy(struct cpdma_ctlr *ctlr) { unsigned long flags; @@ -569,54 +496,6 @@ int cpdma_chan_get_stats(struct cpdma_chan *chan, } EXPORT_SYMBOL_GPL(cpdma_chan_get_stats); -int cpdma_chan_dump(struct cpdma_chan *chan) -{ - unsigned long flags; - struct device *dev = chan->ctlr->dev; - - spin_lock_irqsave(&chan->lock, flags); - - dev_info(dev, "channel %d (%s %d) state %s", - chan->chan_num, is_rx_chan(chan) ? "rx" : "tx", - chan_linear(chan), cpdma_state_str[chan->state]); - dev_info(dev, "\thdp: %x\n", chan_read(chan, hdp)); - dev_info(dev, "\tcp: %x\n", chan_read(chan, cp)); - if (chan->rxfree) { - dev_info(dev, "\trxfree: %x\n", - chan_read(chan, rxfree)); - } - - dev_info(dev, "\tstats head_enqueue: %d\n", - chan->stats.head_enqueue); - dev_info(dev, "\tstats tail_enqueue: %d\n", - chan->stats.tail_enqueue); - dev_info(dev, "\tstats pad_enqueue: %d\n", - chan->stats.pad_enqueue); - dev_info(dev, "\tstats misqueued: %d\n", - chan->stats.misqueued); - dev_info(dev, "\tstats desc_alloc_fail: %d\n", - chan->stats.desc_alloc_fail); - dev_info(dev, "\tstats pad_alloc_fail: %d\n", - chan->stats.pad_alloc_fail); - dev_info(dev, "\tstats runt_receive_buff: %d\n", - chan->stats.runt_receive_buff); - dev_info(dev, "\tstats runt_transmit_buff: %d\n", - chan->stats.runt_transmit_buff); - dev_info(dev, "\tstats empty_dequeue: %d\n", - chan->stats.empty_dequeue); - dev_info(dev, "\tstats busy_dequeue: %d\n", - chan->stats.busy_dequeue); - dev_info(dev, "\tstats good_dequeue: %d\n", - chan->stats.good_dequeue); - dev_info(dev, "\tstats requeue: %d\n", - chan->stats.requeue); - dev_info(dev, "\tstats teardown_dequeue: %d\n", - chan->stats.teardown_dequeue); - - spin_unlock_irqrestore(&chan->lock, flags); - return 0; -} - static void __cpdma_chan_submit(struct cpdma_chan *chan, struct cpdma_desc __iomem *desc) { diff --git a/drivers/net/ethernet/ti/davinci_cpdma.h b/drivers/net/ethernet/ti/davinci_cpdma.h index 80c015c..4b46cd6 100644 --- a/drivers/net/ethernet/ti/davinci_cpdma.h +++ b/drivers/net/ethernet/ti/davinci_cpdma.h @@ -77,7 +77,6 @@ struct cpdma_ctlr *cpdma_ctlr_create(struct cpdma_params *params); int cpdma_ctlr_destroy(struct cpdma_ctlr *ctlr); int cpdma_ctlr_start(struct cpdma_ctlr *ctlr); int cpdma_ctlr_stop(struct cpdma_ctlr *ctlr); -int cpdma_ctlr_dump(struct cpdma_ctlr *ctlr); struct cpdma_chan *cpdma_chan_create(struct cpdma_ctlr *ctlr, int chan_num, cpdma_handler_fn handler); @@ -85,7 +84,6 @@ int cpdma_chan_get_rx_buf_num(struct cpdma_ctlr *ctlr); int cpdma_chan_destroy(struct cpdma_chan *chan); int cpdma_chan_start(struct cpdma_chan *chan); int cpdma_chan_stop(struct cpdma_chan *chan); -int cpdma_chan_dump(struct cpdma_chan *chan); int cpdma_chan_get_stats(struct cpdma_chan *chan, struct cpdma_chan_stats *stats); diff --git a/drivers/net/ethernet/ti/davinci_emac.c b/drivers/net/ethernet/ti/davinci_emac.c index c6c5465..6e305a8 100644 --- a/drivers/net/ethernet/ti/davinci_emac.c +++ b/drivers/net/ethernet/ti/davinci_emac.c @@ -379,97 +379,6 @@ static char *emac_rxhost_errcodes[16] = { #define emac_ctrl_write(reg, val) iowrite32(val, (priv->ctrl_base + (reg))) /** - * emac_dump_regs - Dump important EMAC registers to debug terminal - * @priv: The DaVinci EMAC private adapter structure - * - * Executes ethtool set cmd & sets phy mode - * - */ -static void emac_dump_regs(struct emac_priv *priv) -{ - struct device *emac_dev = &priv->ndev->dev; - - /* Print important registers in EMAC */ - dev_info(emac_dev, "EMAC Basic registers\n"); - if (priv->version == EMAC_VERSION_1) { - dev_info(emac_dev, "EMAC: EWCTL: %08X, EWINTTCNT: %08X\n", - emac_ctrl_read(EMAC_CTRL_EWCTL), - emac_ctrl_read(EMAC_CTRL_EWINTTCNT)); - } - dev_info(emac_dev, "EMAC: EmuControl:%08X, FifoControl: %08X\n", - emac_read(EMAC_EMCONTROL), emac_read(EMAC_FIFOCONTROL)); - dev_info(emac_dev, "EMAC: MBPEnable:%08X, RXUnicastSet: %08X, "\ - "RXMaxLen=%08X\n", emac_read(EMAC_RXMBPENABLE), - emac_read(EMAC_RXUNICASTSET), emac_read(EMAC_RXMAXLEN)); - dev_info(emac_dev, "EMAC: MacControl:%08X, MacStatus: %08X, "\ - "MacConfig=%08X\n", emac_read(EMAC_MACCONTROL), - emac_read(EMAC_MACSTATUS), emac_read(EMAC_MACCONFIG)); - dev_info(emac_dev, "EMAC Statistics\n"); - dev_info(emac_dev, "EMAC: rx_good_frames:%d\n", - emac_read(EMAC_RXGOODFRAMES)); - dev_info(emac_dev, "EMAC: rx_broadcast_frames:%d\n", - emac_read(EMAC_RXBCASTFRAMES)); - dev_info(emac_dev, "EMAC: rx_multicast_frames:%d\n", - emac_read(EMAC_RXMCASTFRAMES)); - dev_info(emac_dev, "EMAC: rx_pause_frames:%d\n", - emac_read(EMAC_RXPAUSEFRAMES)); - dev_info(emac_dev, "EMAC: rx_crcerrors:%d\n", - emac_read(EMAC_RXCRCERRORS)); - dev_info(emac_dev, "EMAC: rx_align_code_errors:%d\n", - emac_read(EMAC_RXALIGNCODEERRORS)); - dev_info(emac_dev, "EMAC: rx_oversized_frames:%d\n", - emac_read(EMAC_RXOVERSIZED)); - dev_info(emac_dev, "EMAC: rx_jabber_frames:%d\n", - emac_read(EMAC_RXJABBER)); - dev_info(emac_dev, "EMAC: rx_undersized_frames:%d\n", - emac_read(EMAC_RXUNDERSIZED)); - dev_info(emac_dev, "EMAC: rx_fragments:%d\n", - emac_read(EMAC_RXFRAGMENTS)); - dev_info(emac_dev, "EMAC: rx_filtered_frames:%d\n", - emac_read(EMAC_RXFILTERED)); - dev_info(emac_dev, "EMAC: rx_qos_filtered_frames:%d\n", - emac_read(EMAC_RXQOSFILTERED)); - dev_info(emac_dev, "EMAC: rx_octets:%d\n", - emac_read(EMAC_RXOCTETS)); - dev_info(emac_dev, "EMAC: tx_goodframes:%d\n", - emac_read(EMAC_TXGOODFRAMES)); - dev_info(emac_dev, "EMAC: tx_bcastframes:%d\n", - emac_read(EMAC_TXBCASTFRAMES)); - dev_info(emac_dev, "EMAC: tx_mcastframes:%d\n", - emac_read(EMAC_TXMCASTFRAMES)); - dev_info(emac_dev, "EMAC: tx_pause_frames:%d\n", - emac_read(EMAC_TXPAUSEFRAMES)); - dev_info(emac_dev, "EMAC: tx_deferred_frames:%d\n", - emac_read(EMAC_TXDEFERRED)); - dev_info(emac_dev, "EMAC: tx_collision_frames:%d\n", - emac_read(EMAC_TXCOLLISION)); - dev_info(emac_dev, "EMAC: tx_single_coll_frames:%d\n", - emac_read(EMAC_TXSINGLECOLL)); - dev_info(emac_dev, "EMAC: tx_mult_coll_frames:%d\n", - emac_read(EMAC_TXMULTICOLL)); - dev_info(emac_dev, "EMAC: tx_excessive_collisions:%d\n", - emac_read(EMAC_TXEXCESSIVECOLL)); - dev_info(emac_dev, "EMAC: tx_late_collisions:%d\n", - emac_read(EMAC_TXLATECOLL)); - dev_info(emac_dev, "EMAC: tx_underrun:%d\n", - emac_read(EMAC_TXUNDERRUN)); - dev_info(emac_dev, "EMAC: tx_carrier_sense_errors:%d\n", - emac_read(EMAC_TXCARRIERSENSE)); - dev_info(emac_dev, "EMAC: tx_octets:%d\n", - emac_read(EMAC_TXOCTETS)); - dev_info(emac_dev, "EMAC: net_octets:%d\n", - emac_read(EMAC_NETOCTETS)); - dev_info(emac_dev, "EMAC: rx_sof_overruns:%d\n", - emac_read(EMAC_RXSOFOVERRUNS)); - dev_info(emac_dev, "EMAC: rx_mof_overruns:%d\n", - emac_read(EMAC_RXMOFOVERRUNS)); - dev_info(emac_dev, "EMAC: rx_dma_overruns:%d\n", - emac_read(EMAC_RXDMAOVERRUNS)); - - cpdma_ctlr_dump(priv->dma); -} - -/** * emac_get_drvinfo - Get EMAC driver information * @ndev: The DaVinci EMAC network adapter * @info: ethtool info structure containing name and version @@ -1096,8 +1005,6 @@ static void emac_dev_tx_timeout(struct net_device *ndev) if (netif_msg_tx_err(priv)) dev_err(emac_dev, "DaVinci EMAC: xmit timeout, restarting TX"); - emac_dump_regs(priv); - ndev->stats.tx_errors++; emac_int_disable(priv); cpdma_chan_stop(priv->txchan); @@ -1617,9 +1524,6 @@ static int emac_dev_open(struct net_device *ndev) emac_update_phystatus(priv); } - if (!netif_running(ndev)) /* debug only - to avoid compiler warning */ - emac_dump_regs(priv); - if (netif_msg_drv(priv)) dev_notice(emac_dev, "DaVinci EMAC: Opened %s\n", ndev->name); -- cgit v0.10.2 From 86cb13e4ec5060d94069a8418fd4f3ccb38edee2 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 25 Jul 2016 13:12:33 +0300 Subject: mlxsw: spectrum: Fix compilation error when CLS_ACT isn't set When CONFIG_NET_CLS_ACT isn't set 'struct tcf_exts' has no member named 'actions' and we therefore must not access it. Otherwise compilation fails. Fix this by introducing a new macro similar to tc_no_actions(), which always returns 'false' if CONFIG_NET_CLS_ACT isn't set. Fixes: 763b4b70afcd ("mlxsw: spectrum: Add support in matchall mirror TC offloading") Reported-by: kbuild test robot Signed-off-by: Jiri Pirko Signed-off-by: Ido Schimmel Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 552636b..c3e6150 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -1148,23 +1148,22 @@ static int mlxsw_sp_port_add_cls_matchall(struct mlxsw_sp_port *mlxsw_sp_port, struct tc_cls_matchall_offload *cls, bool ingress) { - struct tcf_exts *exts = cls->exts; const struct tc_action *a; int err; - if (!list_is_singular(&exts->actions)) { + if (!tc_single_action(cls->exts)) { netdev_err(mlxsw_sp_port->dev, "only singular actions are supported\n"); return -ENOTSUPP; } - a = list_first_entry(&exts->actions, struct tc_action, list); - if (is_tcf_mirred_mirror(a) && protocol == htons(ETH_P_ALL)) { + tc_for_each_action(a, cls->exts) { + if (!is_tcf_mirred_mirror(a) || protocol != htons(ETH_P_ALL)) + return -ENOTSUPP; + err = mlxsw_sp_port_add_cls_matchall_mirror(mlxsw_sp_port, cls, a, ingress); if (err) return err; - } else { - return -ENOTSUPP; } return 0; diff --git a/include/net/act_api.h b/include/net/act_api.h index fb82b5b..0bb2106 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -192,6 +192,9 @@ int tcf_action_copy_stats(struct sk_buff *, struct tc_action *, int); #define tc_for_each_action(_a, _exts) \ list_for_each_entry(a, &(_exts)->actions, list) +#define tc_single_action(_exts) \ + (list_is_singular(&(_exts)->actions)) + static inline void tcf_action_stats_update(struct tc_action *a, u64 bytes, u64 packets, u64 lastuse) { @@ -205,6 +208,7 @@ static inline void tcf_action_stats_update(struct tc_action *a, u64 bytes, #define tc_no_actions(_exts) true #define tc_for_each_action(_a, _exts) while ((void)(_a), 0) +#define tc_single_action(_exts) false #define tcf_action_stats_update(a, bytes, packets, lastuse) #endif /* CONFIG_NET_CLS_ACT */ -- cgit v0.10.2 From 9b022a6e0f26af108b9105b16b310393c898d9bd Mon Sep 17 00:00:00 2001 From: Alex Vesker Date: Mon, 25 Jul 2016 15:42:13 +0300 Subject: net/mlx4_core: Check device state before unregistering it Verify that the device state is registered before un-registering it. This check is required to prevent an OOPS on flows that do re-registration of the device and its previous state was unregistered. Fixes: 225c7b1feef1 ("IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters") Signed-off-by: Alex Vesker Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx4/intf.c b/drivers/net/ethernet/mellanox/mlx4/intf.c index 7ae1cda..0e8b7c4 100644 --- a/drivers/net/ethernet/mellanox/mlx4/intf.c +++ b/drivers/net/ethernet/mellanox/mlx4/intf.c @@ -218,6 +218,9 @@ void mlx4_unregister_device(struct mlx4_dev *dev) struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_interface *intf; + if (!(dev->persist->interface_state & MLX4_INTERFACE_STATE_UP)) + return; + mlx4_stop_catas_poll(dev); mutex_lock(&intf_mutex); -- cgit v0.10.2 From 96ae52279594470622ff0585621a13e96b700600 Mon Sep 17 00:00:00 2001 From: Sargun Dhillon Date: Mon, 25 Jul 2016 05:54:46 -0700 Subject: bpf: Add bpf_probe_write_user BPF helper to be called in tracers This allows user memory to be written to during the course of a kprobe. It shouldn't be used to implement any kind of security mechanism because of TOC-TOU attacks, but rather to debug, divert, and manipulate execution of semi-cooperative processes. Although it uses probe_kernel_write, we limit the address space the probe can write into by checking the space with access_ok. We do this as opposed to calling copy_to_user directly, in order to avoid sleeping. In addition we ensure the threads's current fs / segment is USER_DS and the thread isn't exiting nor a kernel thread. Given this feature is meant for experiments, and it has a risk of crashing the system, and running programs, we print a warning on when a proglet that attempts to use this helper is installed, along with the pid and process name. Signed-off-by: Sargun Dhillon Cc: Alexei Starovoitov Cc: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 2b7076f..da218fe 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -365,6 +365,16 @@ enum bpf_func_id { */ BPF_FUNC_get_current_task, + /** + * bpf_probe_write_user(void *dst, void *src, int len) + * safely attempt to write to a location + * @dst: destination address in userspace + * @src: source address on stack + * @len: number of bytes to copy + * Return: 0 on success or negative error + */ + BPF_FUNC_probe_write_user, + __BPF_FUNC_MAX_ID, }; diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index a12bbd32..b20438f 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -81,6 +81,49 @@ static const struct bpf_func_proto bpf_probe_read_proto = { .arg3_type = ARG_ANYTHING, }; +static u64 bpf_probe_write_user(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) +{ + void *unsafe_ptr = (void *) (long) r1; + void *src = (void *) (long) r2; + int size = (int) r3; + + /* + * Ensure we're in user context which is safe for the helper to + * run. This helper has no business in a kthread. + * + * access_ok() should prevent writing to non-user memory, but in + * some situations (nommu, temporary switch, etc) access_ok() does + * not provide enough validation, hence the check on KERNEL_DS. + */ + + if (unlikely(in_interrupt() || + current->flags & (PF_KTHREAD | PF_EXITING))) + return -EPERM; + if (unlikely(segment_eq(get_fs(), KERNEL_DS))) + return -EPERM; + if (!access_ok(VERIFY_WRITE, unsafe_ptr, size)) + return -EPERM; + + return probe_kernel_write(unsafe_ptr, src, size); +} + +static const struct bpf_func_proto bpf_probe_write_user_proto = { + .func = bpf_probe_write_user, + .gpl_only = true, + .ret_type = RET_INTEGER, + .arg1_type = ARG_ANYTHING, + .arg2_type = ARG_PTR_TO_STACK, + .arg3_type = ARG_CONST_STACK_SIZE, +}; + +static const struct bpf_func_proto *bpf_get_probe_write_proto(void) +{ + pr_warn_ratelimited("%s[%d] is installing a program with bpf_probe_write_user helper that may corrupt user memory!", + current->comm, task_pid_nr(current)); + + return &bpf_probe_write_user_proto; +} + /* * limited trace_printk() * only %d %u %x %ld %lu %lx %lld %llu %llx %p %s conversion specifiers allowed @@ -362,6 +405,8 @@ static const struct bpf_func_proto *tracing_func_proto(enum bpf_func_id func_id) return &bpf_get_smp_processor_id_proto; case BPF_FUNC_perf_event_read: return &bpf_perf_event_read_proto; + case BPF_FUNC_probe_write_user: + return bpf_get_probe_write_proto(); default: return NULL; } diff --git a/samples/bpf/bpf_helpers.h b/samples/bpf/bpf_helpers.h index 84e3fd9..217c8d5 100644 --- a/samples/bpf/bpf_helpers.h +++ b/samples/bpf/bpf_helpers.h @@ -41,6 +41,8 @@ static int (*bpf_perf_event_output)(void *ctx, void *map, int index, void *data, (void *) BPF_FUNC_perf_event_output; static int (*bpf_get_stackid)(void *ctx, void *map, int flags) = (void *) BPF_FUNC_get_stackid; +static int (*bpf_probe_write_user)(void *dst, void *src, int size) = + (void *) BPF_FUNC_probe_write_user; /* llvm builtin functions that eBPF C program may use to * emit BPF_LD_ABS and BPF_LD_IND instructions -- cgit v0.10.2 From cf9b1199de27ece1eafacf165933df10f0314232 Mon Sep 17 00:00:00 2001 From: Sargun Dhillon Date: Mon, 25 Jul 2016 05:55:02 -0700 Subject: samples/bpf: Add test/example of using bpf_probe_write_user bpf helper This example shows using a kprobe to act as a dnat mechanism to divert traffic for arbitrary endpoints. It rewrite the arguments to a syscall while they're still in userspace, and before the syscall has a chance to copy the argument into kernel space. Although this is an example, it also acts as a test because the mapped address is 255.255.255.255:555 -> real address, and that's not a legal address to connect to. If the helper is broken, the example will fail on the intermediate steps, as well as the final step to verify the rewrite of userspace memory succeeded. Signed-off-by: Sargun Dhillon Cc: Alexei Starovoitov Cc: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile index d2d2b35..90ebf7d 100644 --- a/samples/bpf/Makefile +++ b/samples/bpf/Makefile @@ -14,6 +14,7 @@ hostprogs-y += tracex3 hostprogs-y += tracex4 hostprogs-y += tracex5 hostprogs-y += tracex6 +hostprogs-y += test_probe_write_user hostprogs-y += trace_output hostprogs-y += lathist hostprogs-y += offwaketime @@ -37,6 +38,7 @@ tracex3-objs := bpf_load.o libbpf.o tracex3_user.o tracex4-objs := bpf_load.o libbpf.o tracex4_user.o tracex5-objs := bpf_load.o libbpf.o tracex5_user.o tracex6-objs := bpf_load.o libbpf.o tracex6_user.o +test_probe_write_user-objs := bpf_load.o libbpf.o test_probe_write_user_user.o trace_output-objs := bpf_load.o libbpf.o trace_output_user.o lathist-objs := bpf_load.o libbpf.o lathist_user.o offwaketime-objs := bpf_load.o libbpf.o offwaketime_user.o @@ -59,6 +61,7 @@ always += tracex3_kern.o always += tracex4_kern.o always += tracex5_kern.o always += tracex6_kern.o +always += test_probe_write_user_kern.o always += trace_output_kern.o always += tcbpf1_kern.o always += lathist_kern.o @@ -85,6 +88,7 @@ HOSTLOADLIBES_tracex3 += -lelf HOSTLOADLIBES_tracex4 += -lelf -lrt HOSTLOADLIBES_tracex5 += -lelf HOSTLOADLIBES_tracex6 += -lelf +HOSTLOADLIBES_test_probe_write_user += -lelf HOSTLOADLIBES_trace_output += -lelf -lrt HOSTLOADLIBES_lathist += -lelf HOSTLOADLIBES_offwaketime += -lelf diff --git a/samples/bpf/test_probe_write_user_kern.c b/samples/bpf/test_probe_write_user_kern.c new file mode 100644 index 0000000..3a677c8 --- /dev/null +++ b/samples/bpf/test_probe_write_user_kern.c @@ -0,0 +1,52 @@ +/* Copyright (c) 2016 Sargun Dhillon + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include "bpf_helpers.h" + +struct bpf_map_def SEC("maps") dnat_map = { + .type = BPF_MAP_TYPE_HASH, + .key_size = sizeof(struct sockaddr_in), + .value_size = sizeof(struct sockaddr_in), + .max_entries = 256, +}; + +/* kprobe is NOT a stable ABI + * kernel functions can be removed, renamed or completely change semantics. + * Number of arguments and their positions can change, etc. + * In such case this bpf+kprobe example will no longer be meaningful + * + * This example sits on a syscall, and the syscall ABI is relatively stable + * of course, across platforms, and over time, the ABI may change. + */ +SEC("kprobe/sys_connect") +int bpf_prog1(struct pt_regs *ctx) +{ + struct sockaddr_in new_addr, orig_addr = {}; + struct sockaddr_in *mapped_addr; + void *sockaddr_arg = (void *)PT_REGS_PARM2(ctx); + int sockaddr_len = (int)PT_REGS_PARM3(ctx); + + if (sockaddr_len > sizeof(orig_addr)) + return 0; + + if (bpf_probe_read(&orig_addr, sizeof(orig_addr), sockaddr_arg) != 0) + return 0; + + mapped_addr = bpf_map_lookup_elem(&dnat_map, &orig_addr); + if (mapped_addr != NULL) { + memcpy(&new_addr, mapped_addr, sizeof(new_addr)); + bpf_probe_write_user(sockaddr_arg, &new_addr, + sizeof(new_addr)); + } + return 0; +} + +char _license[] SEC("license") = "GPL"; +u32 _version SEC("version") = LINUX_VERSION_CODE; diff --git a/samples/bpf/test_probe_write_user_user.c b/samples/bpf/test_probe_write_user_user.c new file mode 100644 index 0000000..a44bf34 --- /dev/null +++ b/samples/bpf/test_probe_write_user_user.c @@ -0,0 +1,78 @@ +#include +#include +#include +#include +#include "libbpf.h" +#include "bpf_load.h" +#include +#include +#include +#include + +int main(int ac, char **argv) +{ + int serverfd, serverconnfd, clientfd; + socklen_t sockaddr_len; + struct sockaddr serv_addr, mapped_addr, tmp_addr; + struct sockaddr_in *serv_addr_in, *mapped_addr_in, *tmp_addr_in; + char filename[256]; + char *ip; + + serv_addr_in = (struct sockaddr_in *)&serv_addr; + mapped_addr_in = (struct sockaddr_in *)&mapped_addr; + tmp_addr_in = (struct sockaddr_in *)&tmp_addr; + + snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]); + + if (load_bpf_file(filename)) { + printf("%s", bpf_log_buf); + return 1; + } + + assert((serverfd = socket(AF_INET, SOCK_STREAM, 0)) > 0); + assert((clientfd = socket(AF_INET, SOCK_STREAM, 0)) > 0); + + /* Bind server to ephemeral port on lo */ + memset(&serv_addr, 0, sizeof(serv_addr)); + serv_addr_in->sin_family = AF_INET; + serv_addr_in->sin_port = 0; + serv_addr_in->sin_addr.s_addr = htonl(INADDR_LOOPBACK); + + assert(bind(serverfd, &serv_addr, sizeof(serv_addr)) == 0); + + sockaddr_len = sizeof(serv_addr); + assert(getsockname(serverfd, &serv_addr, &sockaddr_len) == 0); + ip = inet_ntoa(serv_addr_in->sin_addr); + printf("Server bound to: %s:%d\n", ip, ntohs(serv_addr_in->sin_port)); + + memset(&mapped_addr, 0, sizeof(mapped_addr)); + mapped_addr_in->sin_family = AF_INET; + mapped_addr_in->sin_port = htons(5555); + mapped_addr_in->sin_addr.s_addr = inet_addr("255.255.255.255"); + + assert(!bpf_update_elem(map_fd[0], &mapped_addr, &serv_addr, BPF_ANY)); + + assert(listen(serverfd, 5) == 0); + + ip = inet_ntoa(mapped_addr_in->sin_addr); + printf("Client connecting to: %s:%d\n", + ip, ntohs(mapped_addr_in->sin_port)); + assert(connect(clientfd, &mapped_addr, sizeof(mapped_addr)) == 0); + + sockaddr_len = sizeof(tmp_addr); + ip = inet_ntoa(tmp_addr_in->sin_addr); + assert((serverconnfd = accept(serverfd, &tmp_addr, &sockaddr_len)) > 0); + printf("Server received connection from: %s:%d\n", + ip, ntohs(tmp_addr_in->sin_port)); + + sockaddr_len = sizeof(tmp_addr); + assert(getpeername(clientfd, &tmp_addr, &sockaddr_len) == 0); + ip = inet_ntoa(tmp_addr_in->sin_addr); + printf("Client's peer address: %s:%d\n", + ip, ntohs(tmp_addr_in->sin_port)); + + /* Is the server's getsockname = the socket getpeername */ + assert(memcmp(&serv_addr, &tmp_addr, sizeof(struct sockaddr_in)) == 0); + + return 0; +} -- cgit v0.10.2 From deb1f45a2f8479ffa5a791e2c024a640a08f729e Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Mon, 25 Jul 2016 18:40:57 +0530 Subject: caif-hsi: Remove deprecated create_singlethread_workqueue alloc_workqueue replaces deprecated create_singlethread_workqueue(). A dedicated workqueue has been used since the workitems are being used on a packet tx/rx path. Hence, WQ_MEM_RECLAIM has been set to guarantee forward progress under memory pressure. An ordered workqueue has been used since workitems &cfhsi->wake_up_work and &cfhsi->wake_down_work cannot be run concurrently. Calls to flush_workqueue() before destroy_workqueue() have been dropped since destroy_workqueue() itself calls drain_workqueue() which flushes repeatedly till the workqueue becomes empty. Signed-off-by: Bhaktipriya Shridhar Acked-by: Tejun Heo Signed-off-by: David S. Miller diff --git a/drivers/net/caif/caif_hsi.c b/drivers/net/caif/caif_hsi.c index 615c65d..ddabce7 100644 --- a/drivers/net/caif/caif_hsi.c +++ b/drivers/net/caif/caif_hsi.c @@ -1201,7 +1201,7 @@ static int cfhsi_open(struct net_device *ndev) clear_bit(CFHSI_AWAKE, &cfhsi->bits); /* Create work thread. */ - cfhsi->wq = create_singlethread_workqueue(cfhsi->ndev->name); + cfhsi->wq = alloc_ordered_workqueue(cfhsi->ndev->name, WQ_MEM_RECLAIM); if (!cfhsi->wq) { netdev_err(cfhsi->ndev, "%s: Failed to create work queue.\n", __func__); @@ -1267,9 +1267,6 @@ static int cfhsi_close(struct net_device *ndev) /* going to shutdown driver */ set_bit(CFHSI_SHUTDOWN, &cfhsi->bits); - /* Flush workqueue */ - flush_workqueue(cfhsi->wq); - /* Delete timers if pending */ del_timer_sync(&cfhsi->inactivity_timer); del_timer_sync(&cfhsi->rx_slowpath_timer); -- cgit v0.10.2 From ba66bbe5480a012108958a71cff88b23dce84956 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Mon, 25 Jul 2016 18:06:12 +0200 Subject: udp: use sk_filter_trim_cap for udp{,6}_queue_rcv_skb After a612769774a3 ("udp: prevent bugcheck if filter truncates packet too much"), there followed various other fixes for similar cases such as f4979fcea7fd ("rose: limit sk_filter trim to payload"). Latter introduced a new helper sk_filter_trim_cap(), where we can pass the trim limit directly to the socket filter handling. Make use of it here as well with sizeof(struct udphdr) as lower cap limit and drop the extra skb->len test in UDP's input path. Signed-off-by: Daniel Borkmann Cc: Willem de Bruijn Acked-by: Willem de Bruijn Signed-off-by: David S. Miller diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 4aed8fc..e61f7cd 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1581,9 +1581,7 @@ int udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) udp_lib_checksum_complete(skb)) goto csum_error; - if (sk_filter(sk, skb)) - goto drop; - if (unlikely(skb->len < sizeof(struct udphdr))) + if (sk_filter_trim_cap(sk, skb, sizeof(struct udphdr))) goto drop; udp_csum_pull_header(skb); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index ad5292b..81e2f98 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -618,9 +618,7 @@ int udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) udp_lib_checksum_complete(skb)) goto csum_error; - if (sk_filter(sk, skb)) - goto drop; - if (unlikely(skb->len < sizeof(struct udphdr))) + if (sk_filter_trim_cap(sk, skb, sizeof(struct udphdr))) goto drop; udp_csum_pull_header(skb); -- cgit v0.10.2 From 59d3f1ceb69b54569685d0c34dff16a1e0816b19 Mon Sep 17 00:00:00 2001 From: Manish Chopra Date: Mon, 25 Jul 2016 19:07:46 +0300 Subject: qed: Fix setting/clearing bit in completion bitmap Slowpath completion handling is incorrectly changing SPQ_RING_SIZE bits instead of a single one. Fixes: 76a9a3642a0b ("qed: fix handling of concurrent ramrods") Signed-off-by: Manish Chopra Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index 97ffeae..d73456e 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -815,13 +815,12 @@ int qed_spq_completion(struct qed_hwfn *p_hwfn, * in a bitmap and increasing the chain consumer only * for the first successive completed entries. */ - bitmap_set(p_spq->p_comp_bitmap, pos, SPQ_RING_SIZE); + __set_bit(pos, p_spq->p_comp_bitmap); while (test_bit(p_spq->comp_bitmap_idx, p_spq->p_comp_bitmap)) { - bitmap_clear(p_spq->p_comp_bitmap, - p_spq->comp_bitmap_idx, - SPQ_RING_SIZE); + __clear_bit(p_spq->comp_bitmap_idx, + p_spq->p_comp_bitmap); p_spq->comp_bitmap_idx++; qed_chain_return_produced(&p_spq->chain); } -- cgit v0.10.2 From a54c4d74989b769014b359e5b66f3e571d903d25 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 25 Jul 2016 12:33:35 -0400 Subject: bnxt_en: Improve ntuple filters by checking destination MAC address. Include the destination MAC address in the ntuple filter structure. The current code assumes that the destination MAC address is always the MAC address of the NIC. This may not be true if there are macvlans, for example. Add destination MAC address checking and configure the filter correctly using the correct index for the destination MAC address. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 8a0165b..7de7d7a 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -3240,7 +3240,7 @@ static int bnxt_hwrm_cfa_ntuple_filter_alloc(struct bnxt *bp, struct bnxt_vnic_info *vnic = &bp->vnic_info[fltr->rxq + 1]; bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_CFA_NTUPLE_FILTER_ALLOC, -1, -1); - req.l2_filter_id = bp->vnic_info[0].fw_l2_filter_id[0]; + req.l2_filter_id = bp->vnic_info[0].fw_l2_filter_id[fltr->l2_fltr_idx]; req.enables = cpu_to_le32(BNXT_NTP_FLTR_FLAGS); @@ -6299,7 +6299,8 @@ static bool bnxt_fltr_match(struct bnxt_ntuple_filter *f1, keys1->ports.ports == keys2->ports.ports && keys1->basic.ip_proto == keys2->basic.ip_proto && keys1->basic.n_proto == keys2->basic.n_proto && - ether_addr_equal(f1->src_mac_addr, f2->src_mac_addr)) + ether_addr_equal(f1->src_mac_addr, f2->src_mac_addr) && + ether_addr_equal(f1->dst_mac_addr, f2->dst_mac_addr)) return true; return false; @@ -6312,12 +6313,28 @@ static int bnxt_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, struct bnxt_ntuple_filter *fltr, *new_fltr; struct flow_keys *fkeys; struct ethhdr *eth = (struct ethhdr *)skb_mac_header(skb); - int rc = 0, idx, bit_id; + int rc = 0, idx, bit_id, l2_idx = 0; struct hlist_head *head; if (skb->encapsulation) return -EPROTONOSUPPORT; + if (!ether_addr_equal(dev->dev_addr, eth->h_dest)) { + struct bnxt_vnic_info *vnic = &bp->vnic_info[0]; + int off = 0, j; + + netif_addr_lock_bh(dev); + for (j = 0; j < vnic->uc_filter_count; j++, off += ETH_ALEN) { + if (ether_addr_equal(eth->h_dest, + vnic->uc_list + off)) { + l2_idx = j + 1; + break; + } + } + netif_addr_unlock_bh(dev); + if (!l2_idx) + return -EINVAL; + } new_fltr = kzalloc(sizeof(*new_fltr), GFP_ATOMIC); if (!new_fltr) return -ENOMEM; @@ -6335,6 +6352,7 @@ static int bnxt_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, goto err_free; } + memcpy(new_fltr->dst_mac_addr, eth->h_dest, ETH_ALEN); memcpy(new_fltr->src_mac_addr, eth->h_source, ETH_ALEN); idx = skb_get_hash_raw(skb) & BNXT_NTP_FLTR_HASH_MASK; @@ -6360,6 +6378,7 @@ static int bnxt_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, new_fltr->sw_id = (u16)bit_id; new_fltr->flow_id = flow_id; + new_fltr->l2_fltr_idx = l2_idx; new_fltr->rxq = rxq_index; hlist_add_head_rcu(&new_fltr->hash, head); bp->ntp_fltr_count++; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 5307a2e..23e04a6 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -785,10 +785,12 @@ struct bnxt_pf_info { struct bnxt_ntuple_filter { struct hlist_node hash; + u8 dst_mac_addr[ETH_ALEN]; u8 src_mac_addr[ETH_ALEN]; struct flow_keys fkeys; __le64 filter_id; u16 sw_id; + u8 l2_fltr_idx; u16 rxq; u32 flow_id; unsigned long state; -- cgit v0.10.2 From a23049091d57f4bdc47f16fce01c371647d15dd7 Mon Sep 17 00:00:00 2001 From: Vasundhara Volam Date: Mon, 25 Jul 2016 12:33:36 -0400 Subject: bnxt_en: Log a message, if enabling NTUPLE filtering fails. If there are not enough resources to enable ntuple filtering, log a warning message. v2: Use single message and add missing newline. Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 7de7d7a..eac0f2b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -5790,8 +5790,12 @@ static bool bnxt_rfs_capable(struct bnxt *bp) return false; vnics = 1 + bp->rx_nr_rings; - if (vnics > pf->max_rsscos_ctxs || vnics > pf->max_vnics) + if (vnics > pf->max_rsscos_ctxs || vnics > pf->max_vnics) { + netdev_warn(bp->dev, + "Not enough resources to support NTUPLE filters, enough resources for up to %d rx rings\n", + min(pf->max_rsscos_ctxs - 1, pf->max_vnics - 1)); return false; + } return true; #else @@ -5804,7 +5808,7 @@ static netdev_features_t bnxt_fix_features(struct net_device *dev, { struct bnxt *bp = netdev_priv(dev); - if (!bnxt_rfs_capable(bp)) + if ((features & NETIF_F_NTUPLE) && !bnxt_rfs_capable(bp)) features &= ~NETIF_F_NTUPLE; /* Both CTAG and STAG VLAN accelaration on the RX side have to be -- cgit v0.10.2 From 1f681688aaf1126df981615064a68a0dced458ef Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 25 Jul 2016 12:33:37 -0400 Subject: bnxt_en: Add new NPAR and dual media device IDs. Add 5741X/5731X NPAR device IDs and dual media SFP/10GBase-T device IDs. Signed-off-by: Michael Chan Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index eac0f2b..2cf7910 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -73,19 +73,28 @@ enum board_idx { BCM57301, BCM57302, BCM57304, + BCM57417_NPAR, BCM58700, BCM57311, BCM57312, BCM57402, BCM57404, BCM57406, - BCM57404_NPAR, + BCM57402_NPAR, + BCM57407, BCM57412, BCM57414, BCM57416, BCM57417, - BCM57414_NPAR, + BCM57412_NPAR, BCM57314, + BCM57417_SFP, + BCM57416_SFP, + BCM57404_NPAR, + BCM57406_NPAR, + BCM57407_SFP, + BCM57414_NPAR, + BCM57416_NPAR, BCM57304_VF, BCM57404_VF, BCM57414_VF, @@ -99,19 +108,28 @@ static const struct { { "Broadcom BCM57301 NetXtreme-C Single-port 10Gb Ethernet" }, { "Broadcom BCM57302 NetXtreme-C Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57304 NetXtreme-C Dual-port 10Gb/25Gb/40Gb/50Gb Ethernet" }, + { "Broadcom BCM57417 NetXtreme-E Ethernet Partition" }, { "Broadcom BCM58700 Nitro 4-port 1Gb/2.5Gb/10Gb Ethernet" }, { "Broadcom BCM57311 NetXtreme-C Single-port 10Gb Ethernet" }, { "Broadcom BCM57312 NetXtreme-C Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57402 NetXtreme-E Dual-port 10Gb Ethernet" }, { "Broadcom BCM57404 NetXtreme-E Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57406 NetXtreme-E Dual-port 10GBase-T Ethernet" }, - { "Broadcom BCM57404 NetXtreme-E Ethernet Partition" }, + { "Broadcom BCM57402 NetXtreme-E Ethernet Partition" }, + { "Broadcom BCM57407 NetXtreme-E Dual-port 10GBase-T Ethernet" }, { "Broadcom BCM57412 NetXtreme-E Dual-port 10Gb Ethernet" }, { "Broadcom BCM57414 NetXtreme-E Dual-port 10Gb/25Gb Ethernet" }, { "Broadcom BCM57416 NetXtreme-E Dual-port 10GBase-T Ethernet" }, { "Broadcom BCM57417 NetXtreme-E Dual-port 10GBase-T Ethernet" }, - { "Broadcom BCM57414 NetXtreme-E Ethernet Partition" }, + { "Broadcom BCM57412 NetXtreme-E Ethernet Partition" }, { "Broadcom BCM57314 NetXtreme-C Dual-port 10Gb/25Gb/40Gb/50Gb Ethernet" }, + { "Broadcom BCM57417 NetXtreme-E Dual-port 10Gb/25Gb Ethernet" }, + { "Broadcom BCM57416 NetXtreme-E Dual-port 10Gb Ethernet" }, + { "Broadcom BCM57404 NetXtreme-E Ethernet Partition" }, + { "Broadcom BCM57406 NetXtreme-E Ethernet Partition" }, + { "Broadcom BCM57407 NetXtreme-E Dual-port 25Gb Ethernet" }, + { "Broadcom BCM57414 NetXtreme-E Ethernet Partition" }, + { "Broadcom BCM57416 NetXtreme-E Ethernet Partition" }, { "Broadcom BCM57304 NetXtreme-C Ethernet Virtual Function" }, { "Broadcom BCM57404 NetXtreme-E Ethernet Virtual Function" }, { "Broadcom BCM57414 NetXtreme-E Ethernet Virtual Function" }, @@ -122,19 +140,28 @@ static const struct pci_device_id bnxt_pci_tbl[] = { { PCI_VDEVICE(BROADCOM, 0x16c8), .driver_data = BCM57301 }, { PCI_VDEVICE(BROADCOM, 0x16c9), .driver_data = BCM57302 }, { PCI_VDEVICE(BROADCOM, 0x16ca), .driver_data = BCM57304 }, + { PCI_VDEVICE(BROADCOM, 0x16cc), .driver_data = BCM57417_NPAR }, { PCI_VDEVICE(BROADCOM, 0x16cd), .driver_data = BCM58700 }, { PCI_VDEVICE(BROADCOM, 0x16ce), .driver_data = BCM57311 }, { PCI_VDEVICE(BROADCOM, 0x16cf), .driver_data = BCM57312 }, { PCI_VDEVICE(BROADCOM, 0x16d0), .driver_data = BCM57402 }, { PCI_VDEVICE(BROADCOM, 0x16d1), .driver_data = BCM57404 }, { PCI_VDEVICE(BROADCOM, 0x16d2), .driver_data = BCM57406 }, - { PCI_VDEVICE(BROADCOM, 0x16d4), .driver_data = BCM57404_NPAR }, + { PCI_VDEVICE(BROADCOM, 0x16d4), .driver_data = BCM57402_NPAR }, + { PCI_VDEVICE(BROADCOM, 0x16d5), .driver_data = BCM57407 }, { PCI_VDEVICE(BROADCOM, 0x16d6), .driver_data = BCM57412 }, { PCI_VDEVICE(BROADCOM, 0x16d7), .driver_data = BCM57414 }, { PCI_VDEVICE(BROADCOM, 0x16d8), .driver_data = BCM57416 }, { PCI_VDEVICE(BROADCOM, 0x16d9), .driver_data = BCM57417 }, - { PCI_VDEVICE(BROADCOM, 0x16de), .driver_data = BCM57414_NPAR }, + { PCI_VDEVICE(BROADCOM, 0x16de), .driver_data = BCM57412_NPAR }, { PCI_VDEVICE(BROADCOM, 0x16df), .driver_data = BCM57314 }, + { PCI_VDEVICE(BROADCOM, 0x16e2), .driver_data = BCM57417_SFP }, + { PCI_VDEVICE(BROADCOM, 0x16e3), .driver_data = BCM57416_SFP }, + { PCI_VDEVICE(BROADCOM, 0x16e7), .driver_data = BCM57404_NPAR }, + { PCI_VDEVICE(BROADCOM, 0x16e8), .driver_data = BCM57406_NPAR }, + { PCI_VDEVICE(BROADCOM, 0x16e9), .driver_data = BCM57407_SFP }, + { PCI_VDEVICE(BROADCOM, 0x16ec), .driver_data = BCM57414_NPAR }, + { PCI_VDEVICE(BROADCOM, 0x16ee), .driver_data = BCM57416_NPAR }, #ifdef CONFIG_BNXT_SRIOV { PCI_VDEVICE(BROADCOM, 0x16cb), .driver_data = BCM57304_VF }, { PCI_VDEVICE(BROADCOM, 0x16d3), .driver_data = BCM57404_VF }, -- cgit v0.10.2 From b93dd49c1a35884864027abd707889b795637f7a Mon Sep 17 00:00:00 2001 From: Mahesh Bandewar Date: Mon, 25 Jul 2016 14:38:16 -0700 Subject: ipvlan: Scrub skb before crossing the namespace boundry The earlier patch c3aaa06d5a63 (ipvlan: scrub skb before routing in L3 mode.) did this but only for TX path in L3 mode. This patch extends it for both the modes for TX/RX path. Signed-off-by: Mahesh Bandewar Signed-off-by: David S. Miller diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c index d6d0524..b5f9511 100644 --- a/drivers/net/ipvlan/ipvlan_core.c +++ b/drivers/net/ipvlan/ipvlan_core.c @@ -254,6 +254,18 @@ acct: } } +static void ipvlan_skb_crossing_ns(struct sk_buff *skb, struct net_device *dev) +{ + bool xnet = true; + + if (dev) + xnet = !net_eq(dev_net(skb->dev), dev_net(dev)); + + skb_scrub_packet(skb, xnet); + if (dev) + skb->dev = dev; +} + static int ipvlan_rcv_frame(struct ipvl_addr *addr, struct sk_buff **pskb, bool local) { @@ -280,7 +292,7 @@ static int ipvlan_rcv_frame(struct ipvl_addr *addr, struct sk_buff **pskb, *pskb = skb; } - skb->dev = dev; + ipvlan_skb_crossing_ns(skb, dev); if (local) { skb->pkt_type = PACKET_HOST; @@ -347,7 +359,7 @@ static struct ipvl_addr *ipvlan_addr_lookup(struct ipvl_port *port, return addr; } -static int ipvlan_process_v4_outbound(struct sk_buff *skb, bool xnet) +static int ipvlan_process_v4_outbound(struct sk_buff *skb) { const struct iphdr *ip4h = ip_hdr(skb); struct net_device *dev = skb->dev; @@ -370,7 +382,6 @@ static int ipvlan_process_v4_outbound(struct sk_buff *skb, bool xnet) ip_rt_put(rt); goto err; } - skb_scrub_packet(skb, xnet); skb_dst_set(skb, &rt->dst); err = ip_local_out(net, skb->sk, skb); if (unlikely(net_xmit_eval(err))) @@ -385,7 +396,7 @@ out: return ret; } -static int ipvlan_process_v6_outbound(struct sk_buff *skb, bool xnet) +static int ipvlan_process_v6_outbound(struct sk_buff *skb) { const struct ipv6hdr *ip6h = ipv6_hdr(skb); struct net_device *dev = skb->dev; @@ -408,7 +419,6 @@ static int ipvlan_process_v6_outbound(struct sk_buff *skb, bool xnet) dst_release(dst); goto err; } - skb_scrub_packet(skb, xnet); skb_dst_set(skb, dst); err = ip6_local_out(net, skb->sk, skb); if (unlikely(net_xmit_eval(err))) @@ -423,7 +433,7 @@ out: return ret; } -static int ipvlan_process_outbound(struct sk_buff *skb, bool xnet) +static int ipvlan_process_outbound(struct sk_buff *skb) { struct ethhdr *ethh = eth_hdr(skb); int ret = NET_XMIT_DROP; @@ -447,9 +457,9 @@ static int ipvlan_process_outbound(struct sk_buff *skb, bool xnet) } if (skb->protocol == htons(ETH_P_IPV6)) - ret = ipvlan_process_v6_outbound(skb, xnet); + ret = ipvlan_process_v6_outbound(skb); else if (skb->protocol == htons(ETH_P_IP)) - ret = ipvlan_process_v4_outbound(skb, xnet); + ret = ipvlan_process_v4_outbound(skb); else { pr_warn_ratelimited("Dropped outbound packet type=%x\n", ntohs(skb->protocol)); @@ -485,7 +495,6 @@ static int ipvlan_xmit_mode_l3(struct sk_buff *skb, struct net_device *dev) void *lyr3h; struct ipvl_addr *addr; int addr_type; - bool xnet; lyr3h = ipvlan_get_L3_hdr(skb, &addr_type); if (!lyr3h) @@ -496,9 +505,8 @@ static int ipvlan_xmit_mode_l3(struct sk_buff *skb, struct net_device *dev) return ipvlan_rcv_frame(addr, &skb, true); out: - xnet = !net_eq(dev_net(skb->dev), dev_net(ipvlan->phy_dev)); - skb->dev = ipvlan->phy_dev; - return ipvlan_process_outbound(skb, xnet); + ipvlan_skb_crossing_ns(skb, ipvlan->phy_dev); + return ipvlan_process_outbound(skb); } static int ipvlan_xmit_mode_l2(struct sk_buff *skb, struct net_device *dev) @@ -528,11 +536,12 @@ static int ipvlan_xmit_mode_l2(struct sk_buff *skb, struct net_device *dev) return dev_forward_skb(ipvlan->phy_dev, skb); } else if (is_multicast_ether_addr(eth->h_dest)) { + ipvlan_skb_crossing_ns(skb, NULL); ipvlan_multicast_enqueue(ipvlan->port, skb); return NET_XMIT_SUCCESS; } - skb->dev = ipvlan->phy_dev; + ipvlan_skb_crossing_ns(skb, ipvlan->phy_dev); return dev_queue_xmit(skb); } @@ -622,8 +631,10 @@ static rx_handler_result_t ipvlan_handle_mode_l2(struct sk_buff **pskb, * when work-queue processes this frame. This is * achieved by returning RX_HANDLER_PASS. */ - if (nskb) + if (nskb) { + ipvlan_skb_crossing_ns(nskb, NULL); ipvlan_multicast_enqueue(port, nskb); + } } } else { struct ipvl_addr *addr; -- cgit v0.10.2 From a85a970af265f156740977168b542234511b28a8 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Mon, 25 Jul 2016 16:09:41 -0700 Subject: net_sched: move tc_action into tcf_common struct tc_action is confusing, currently we use it for two purposes: 1) Pass in arguments and carry out results from helper functions 2) A generic representation for tc actions The first one is error-prone, since we need to make sure we don't miss anything. This patch aims to get rid of this use, by moving tc_action into tcf_common, so that they are allocated together in hashtable and can be cast'ed easily. And together with the following patch, we could really make tc_action a generic representation for all tc actions and each type of action can inherit from it. Cc: Jamal Hadi Salim Signed-off-by: Cong Wang Signed-off-by: David S. Miller diff --git a/include/net/act_api.h b/include/net/act_api.h index 0bb2106..8b19909 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -10,7 +10,26 @@ #include #include + +struct tcf_hashinfo { + struct hlist_head *htab; + unsigned int hmask; + spinlock_t lock; + u32 index; +}; + +struct tc_action_ops; + +struct tc_action { + const struct tc_action_ops *ops; + __u32 type; /* for backward compat(TCA_OLD_COMPAT) */ + __u32 order; + struct list_head list; + struct tcf_hashinfo *hinfo; +}; + struct tcf_common { + struct tc_action tcfc_act; struct hlist_node tcfc_head; u32 tcfc_index; int tcfc_refcnt; @@ -26,6 +45,7 @@ struct tcf_common { struct gnet_stats_basic_cpu __percpu *cpu_bstats; struct gnet_stats_queue __percpu *cpu_qstats; }; +#define tcf_act common.tcfc_act #define tcf_head common.tcfc_head #define tcf_index common.tcfc_index #define tcf_refcnt common.tcfc_refcnt @@ -39,13 +59,6 @@ struct tcf_common { #define tcf_lock common.tcfc_lock #define tcf_rcu common.tcfc_rcu -struct tcf_hashinfo { - struct hlist_head *htab; - unsigned int hmask; - spinlock_t lock; - u32 index; -}; - static inline unsigned int tcf_hash(u32 index, unsigned int hmask) { return index & hmask; @@ -88,15 +101,6 @@ static inline void tcf_tm_dump(struct tcf_t *dtm, const struct tcf_t *stm) dtm->expires = jiffies_to_clock_t(stm->expires); } -struct tc_action { - void *priv; - const struct tc_action_ops *ops; - __u32 type; /* for backward compat(TCA_OLD_COMPAT) */ - __u32 order; - struct list_head list; - struct tcf_hashinfo *hinfo; -}; - #ifdef CONFIG_NET_CLS_ACT #define ACT_P_CREATED 1 @@ -106,17 +110,18 @@ struct tc_action_ops { struct list_head head; char kind[IFNAMSIZ]; __u32 type; /* TBD to match kind */ + size_t size; struct module *owner; int (*act)(struct sk_buff *, const struct tc_action *, struct tcf_result *); int (*dump)(struct sk_buff *, struct tc_action *, int, int); void (*cleanup)(struct tc_action *, int bind); - int (*lookup)(struct net *, struct tc_action *, u32); + int (*lookup)(struct net *, struct tc_action **, u32); int (*init)(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *act, int ovr, + struct nlattr *est, struct tc_action **act, int ovr, int bind); int (*walk)(struct net *, struct sk_buff *, - struct netlink_callback *, int, struct tc_action *); + struct netlink_callback *, int, const struct tc_action_ops *); void (*stats_update)(struct tc_action *, u64, u32, u64); }; @@ -152,13 +157,14 @@ static inline void tc_action_net_exit(struct tc_action_net *tn) int tcf_generic_walker(struct tc_action_net *tn, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a); -int tcf_hash_search(struct tc_action_net *tn, struct tc_action *a, u32 index); + const struct tc_action_ops *ops); +int tcf_hash_search(struct tc_action_net *tn, struct tc_action **a, u32 index); u32 tcf_hash_new_index(struct tc_action_net *tn); -bool tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action *a, +bool tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action **a, int bind); int tcf_hash_create(struct tc_action_net *tn, u32 index, struct nlattr *est, - struct tc_action *a, int size, int bind, bool cpustats); + struct tc_action **a, const struct tc_action_ops *ops, int bind, + bool cpustats); void tcf_hash_cleanup(struct tc_action *a, struct nlattr *est); void tcf_hash_insert(struct tc_action_net *tn, struct tc_action *a); diff --git a/include/net/tc_act/tc_bpf.h b/include/net/tc_act/tc_bpf.h index 958d69c..80a4d6f 100644 --- a/include/net/tc_act/tc_bpf.h +++ b/include/net/tc_act/tc_bpf.h @@ -23,7 +23,6 @@ struct tcf_bpf { struct sock_filter *bpf_ops; const char *bpf_name; }; -#define to_bpf(a) \ - container_of(a->priv, struct tcf_bpf, common) +#define to_bpf(a) ((struct tcf_bpf *)a) #endif /* __NET_TC_BPF_H */ diff --git a/include/net/tc_act/tc_connmark.h b/include/net/tc_act/tc_connmark.h index 02caa40..8a66113 100644 --- a/include/net/tc_act/tc_connmark.h +++ b/include/net/tc_act/tc_connmark.h @@ -9,7 +9,6 @@ struct tcf_connmark_info { u16 zone; }; -#define to_connmark(a) \ - container_of(a->priv, struct tcf_connmark_info, common) +#define to_connmark(a) ((struct tcf_connmark_info *)a) #endif /* __NET_TC_CONNMARK_H */ diff --git a/include/net/tc_act/tc_csum.h b/include/net/tc_act/tc_csum.h index fa8f5fa..1a9ef15 100644 --- a/include/net/tc_act/tc_csum.h +++ b/include/net/tc_act/tc_csum.h @@ -9,7 +9,6 @@ struct tcf_csum { u32 update_flags; }; -#define to_tcf_csum(a) \ - container_of(a->priv,struct tcf_csum,common) +#define to_tcf_csum(a) ((struct tcf_csum *)a) #endif /* __NET_TC_CSUM_H */ diff --git a/include/net/tc_act/tc_defact.h b/include/net/tc_act/tc_defact.h index ab9b5d6..e25b4eb 100644 --- a/include/net/tc_act/tc_defact.h +++ b/include/net/tc_act/tc_defact.h @@ -8,7 +8,6 @@ struct tcf_defact { u32 tcfd_datalen; void *tcfd_defdata; }; -#define to_defact(a) \ - container_of(a->priv, struct tcf_defact, common) +#define to_defact(a) ((struct tcf_defact *)a) #endif /* __NET_TC_DEF_H */ diff --git a/include/net/tc_act/tc_gact.h b/include/net/tc_act/tc_gact.h index 93c520b..119cdb4 100644 --- a/include/net/tc_act/tc_gact.h +++ b/include/net/tc_act/tc_gact.h @@ -13,8 +13,7 @@ struct tcf_gact { atomic_t packets; #endif }; -#define to_gact(a) \ - container_of(a->priv, struct tcf_gact, common) +#define to_gact(a) ((struct tcf_gact *)a) static inline bool is_tcf_gact_shot(const struct tc_action *a) { @@ -24,7 +23,7 @@ static inline bool is_tcf_gact_shot(const struct tc_action *a) if (a->ops && a->ops->type != TCA_ACT_GACT) return false; - gact = a->priv; + gact = to_gact(a); if (gact->tcf_action == TC_ACT_SHOT) return true; diff --git a/include/net/tc_act/tc_ife.h b/include/net/tc_act/tc_ife.h index c55facd..7921abe 100644 --- a/include/net/tc_act/tc_ife.h +++ b/include/net/tc_act/tc_ife.h @@ -16,8 +16,7 @@ struct tcf_ife_info { /* list of metaids allowed */ struct list_head metalist; }; -#define to_ife(a) \ - container_of(a->priv, struct tcf_ife_info, common) +#define to_ife(a) ((struct tcf_ife_info *)a) struct tcf_meta_info { const struct tcf_meta_ops *ops; diff --git a/include/net/tc_act/tc_ipt.h b/include/net/tc_act/tc_ipt.h index c0f4193..c22ae7a 100644 --- a/include/net/tc_act/tc_ipt.h +++ b/include/net/tc_act/tc_ipt.h @@ -11,7 +11,6 @@ struct tcf_ipt { char *tcfi_tname; struct xt_entry_target *tcfi_t; }; -#define to_ipt(a) \ - container_of(a->priv, struct tcf_ipt, common) +#define to_ipt(a) ((struct tcf_ipt *)a) #endif /* __NET_TC_IPT_H */ diff --git a/include/net/tc_act/tc_mirred.h b/include/net/tc_act/tc_mirred.h index 6a13a7c..89aebd2 100644 --- a/include/net/tc_act/tc_mirred.h +++ b/include/net/tc_act/tc_mirred.h @@ -12,8 +12,7 @@ struct tcf_mirred { struct net_device __rcu *tcfm_dev; struct list_head tcfm_list; }; -#define to_mirred(a) \ - container_of(a->priv, struct tcf_mirred, common) +#define to_mirred(a) ((struct tcf_mirred *)a) static inline bool is_tcf_mirred_redirect(const struct tc_action *a) { diff --git a/include/net/tc_act/tc_nat.h b/include/net/tc_act/tc_nat.h index 63d8e9c..a91ad3a 100644 --- a/include/net/tc_act/tc_nat.h +++ b/include/net/tc_act/tc_nat.h @@ -13,9 +13,6 @@ struct tcf_nat { u32 flags; }; -static inline struct tcf_nat *to_tcf_nat(struct tc_action *a) -{ - return container_of(a->priv, struct tcf_nat, common); -} +#define to_tcf_nat(a) ((struct tcf_nat *)a) #endif /* __NET_TC_NAT_H */ diff --git a/include/net/tc_act/tc_pedit.h b/include/net/tc_act/tc_pedit.h index 5b80998..2cccfba 100644 --- a/include/net/tc_act/tc_pedit.h +++ b/include/net/tc_act/tc_pedit.h @@ -9,7 +9,6 @@ struct tcf_pedit { unsigned char tcfp_flags; struct tc_pedit_key *tcfp_keys; }; -#define to_pedit(a) \ - container_of(a->priv, struct tcf_pedit, common) +#define to_pedit(a) ((struct tcf_pedit *)a) #endif /* __NET_TC_PED_H */ diff --git a/include/net/tc_act/tc_skbedit.h b/include/net/tc_act/tc_skbedit.h index d01a5d4..9e05489 100644 --- a/include/net/tc_act/tc_skbedit.h +++ b/include/net/tc_act/tc_skbedit.h @@ -30,8 +30,7 @@ struct tcf_skbedit { u16 queue_mapping; u16 ptype; }; -#define to_skbedit(a) \ - container_of(a->priv, struct tcf_skbedit, common) +#define to_skbedit(a) ((struct tcf_skbedit *)a) /* Return true iff action is mark */ static inline bool is_tcf_skbedit_mark(const struct tc_action *a) diff --git a/include/net/tc_act/tc_vlan.h b/include/net/tc_act/tc_vlan.h index 93b70ad..584b807 100644 --- a/include/net/tc_act/tc_vlan.h +++ b/include/net/tc_act/tc_vlan.h @@ -21,7 +21,6 @@ struct tcf_vlan { u16 tcfv_push_vid; __be16 tcfv_push_proto; }; -#define to_vlan(a) \ - container_of(a->priv, struct tcf_vlan, common) +#define to_vlan(a) ((struct tcf_vlan *)a) #endif /* __NET_TC_VLAN_H */ diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 47ec230..d97419f 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -38,7 +38,7 @@ static void free_tcf(struct rcu_head *head) static void tcf_hash_destroy(struct tcf_hashinfo *hinfo, struct tc_action *a) { - struct tcf_common *p = a->priv; + struct tcf_common *p = (struct tcf_common *)a; spin_lock_bh(&hinfo->lock); hlist_del(&p->tcfc_head); @@ -54,7 +54,7 @@ static void tcf_hash_destroy(struct tcf_hashinfo *hinfo, struct tc_action *a) int __tcf_hash_release(struct tc_action *a, bool bind, bool strict) { - struct tcf_common *p = a->priv; + struct tcf_common *p = (struct tcf_common *)a; int ret = 0; if (p) { @@ -67,6 +67,7 @@ int __tcf_hash_release(struct tc_action *a, bool bind, bool strict) if (p->tcfc_bindcnt <= 0 && p->tcfc_refcnt <= 0) { if (a->ops->cleanup) a->ops->cleanup(a, bind); + list_del(&a->list); tcf_hash_destroy(a->hinfo, a); ret = ACT_P_DELETED; } @@ -77,10 +78,8 @@ int __tcf_hash_release(struct tc_action *a, bool bind, bool strict) EXPORT_SYMBOL(__tcf_hash_release); static int tcf_dump_walker(struct tcf_hashinfo *hinfo, struct sk_buff *skb, - struct netlink_callback *cb, struct tc_action *a) + struct netlink_callback *cb) { - struct hlist_head *head; - struct tcf_common *p; int err = 0, index = -1, i = 0, s_i = 0, n_i = 0; struct nlattr *nest; @@ -89,19 +88,20 @@ static int tcf_dump_walker(struct tcf_hashinfo *hinfo, struct sk_buff *skb, s_i = cb->args[0]; for (i = 0; i < (hinfo->hmask + 1); i++) { + struct hlist_head *head; + struct tcf_common *p; + head = &hinfo->htab[tcf_hash(i, hinfo->hmask)]; hlist_for_each_entry_rcu(p, head, tcfc_head) { index++; if (index < s_i) continue; - a->priv = p; - a->order = n_i; - nest = nla_nest_start(skb, a->order); + nest = nla_nest_start(skb, n_i); if (nest == NULL) goto nla_put_failure; - err = tcf_action_dump_1(skb, a, 0, 0); + err = tcf_action_dump_1(skb, (struct tc_action *)p, 0, 0); if (err < 0) { index--; nlmsg_trim(skb, nest); @@ -125,27 +125,27 @@ nla_put_failure: } static int tcf_del_walker(struct tcf_hashinfo *hinfo, struct sk_buff *skb, - struct tc_action *a) + const struct tc_action_ops *ops) { - struct hlist_head *head; - struct hlist_node *n; - struct tcf_common *p; struct nlattr *nest; int i = 0, n_i = 0; int ret = -EINVAL; - nest = nla_nest_start(skb, a->order); + nest = nla_nest_start(skb, 0); if (nest == NULL) goto nla_put_failure; - if (nla_put_string(skb, TCA_KIND, a->ops->kind)) + if (nla_put_string(skb, TCA_KIND, ops->kind)) goto nla_put_failure; for (i = 0; i < (hinfo->hmask + 1); i++) { + struct hlist_head *head; + struct hlist_node *n; + struct tcf_common *p; + head = &hinfo->htab[tcf_hash(i, hinfo->hmask)]; hlist_for_each_entry_safe(p, n, head, tcfc_head) { - a->priv = p; - ret = __tcf_hash_release(a, false, true); + ret = __tcf_hash_release((struct tc_action *)p, false, true); if (ret == ACT_P_DELETED) { - module_put(a->ops->owner); + module_put(p->tcfc_act.ops->owner); n_i++; } else if (ret < 0) goto nla_put_failure; @@ -163,16 +163,14 @@ nla_put_failure: int tcf_generic_walker(struct tc_action_net *tn, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tcf_hashinfo *hinfo = tn->hinfo; - a->hinfo = hinfo; - if (type == RTM_DELACTION) { - return tcf_del_walker(hinfo, skb, a); + return tcf_del_walker(hinfo, skb, ops); } else if (type == RTM_GETACTION) { - return tcf_dump_walker(hinfo, skb, cb, a); + return tcf_dump_walker(hinfo, skb, cb); } else { WARN(1, "tcf_generic_walker: unknown action %d\n", type); return -EINVAL; @@ -210,21 +208,20 @@ u32 tcf_hash_new_index(struct tc_action_net *tn) } EXPORT_SYMBOL(tcf_hash_new_index); -int tcf_hash_search(struct tc_action_net *tn, struct tc_action *a, u32 index) +int tcf_hash_search(struct tc_action_net *tn, struct tc_action **a, u32 index) { struct tcf_hashinfo *hinfo = tn->hinfo; struct tcf_common *p = tcf_hash_lookup(index, hinfo); if (p) { - a->priv = p; - a->hinfo = hinfo; + *a = &p->tcfc_act; return 1; } return 0; } EXPORT_SYMBOL(tcf_hash_search); -bool tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action *a, +bool tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action **a, int bind) { struct tcf_hashinfo *hinfo = tn->hinfo; @@ -233,8 +230,7 @@ bool tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action *a, if (bind) p->tcfc_bindcnt++; p->tcfc_refcnt++; - a->priv = p; - a->hinfo = hinfo; + *a = &p->tcfc_act; return true; } return false; @@ -243,7 +239,7 @@ EXPORT_SYMBOL(tcf_hash_check); void tcf_hash_cleanup(struct tc_action *a, struct nlattr *est) { - struct tcf_common *pc = a->priv; + struct tcf_common *pc = (struct tcf_common *)a; if (est) gen_kill_estimator(&pc->tcfc_bstats, &pc->tcfc_rate_est); @@ -252,9 +248,10 @@ void tcf_hash_cleanup(struct tc_action *a, struct nlattr *est) EXPORT_SYMBOL(tcf_hash_cleanup); int tcf_hash_create(struct tc_action_net *tn, u32 index, struct nlattr *est, - struct tc_action *a, int size, int bind, bool cpustats) + struct tc_action **a, const struct tc_action_ops *ops, + int bind, bool cpustats) { - struct tcf_common *p = kzalloc(size, GFP_KERNEL); + struct tcf_common *p = kzalloc(ops->size, GFP_KERNEL); struct tcf_hashinfo *hinfo = tn->hinfo; int err = -ENOMEM; @@ -294,15 +291,17 @@ err2: } } - a->priv = (void *) p; - a->hinfo = hinfo; + p->tcfc_act.hinfo = hinfo; + p->tcfc_act.ops = ops; + INIT_LIST_HEAD(&p->tcfc_act.list); + *a = &p->tcfc_act; return 0; } EXPORT_SYMBOL(tcf_hash_create); void tcf_hash_insert(struct tc_action_net *tn, struct tc_action *a) { - struct tcf_common *p = a->priv; + struct tcf_common *p = (struct tcf_common *)a; struct tcf_hashinfo *hinfo = tn->hinfo; unsigned int h = tcf_hash(p->tcfc_index, hinfo->hmask); @@ -315,10 +314,6 @@ EXPORT_SYMBOL(tcf_hash_insert); void tcf_hashinfo_destroy(const struct tc_action_ops *ops, struct tcf_hashinfo *hinfo) { - struct tc_action a = { - .ops = ops, - .hinfo = hinfo, - }; int i; for (i = 0; i < hinfo->hmask + 1; i++) { @@ -328,8 +323,7 @@ void tcf_hashinfo_destroy(const struct tc_action_ops *ops, hlist_for_each_entry_safe(p, n, &hinfo->htab[i], tcfc_head) { int ret; - a.priv = p; - ret = __tcf_hash_release(&a, false, true); + ret = __tcf_hash_release((struct tc_action *)p, false, true); if (ret == ACT_P_DELETED) module_put(ops->owner); else if (ret < 0) @@ -466,8 +460,6 @@ int tcf_action_destroy(struct list_head *actions, int bind) module_put(a->ops->owner); else if (ret < 0) return ret; - list_del(&a->list); - kfree(a); } return ret; } @@ -581,20 +573,13 @@ struct tc_action *tcf_action_init_1(struct net *net, struct nlattr *nla, goto err_out; } - err = -ENOMEM; - a = kzalloc(sizeof(*a), GFP_KERNEL); - if (a == NULL) - goto err_mod; - - a->ops = a_o; - INIT_LIST_HEAD(&a->list); /* backward compatibility for policer */ if (name == NULL) - err = a_o->init(net, tb[TCA_ACT_OPTIONS], est, a, ovr, bind); + err = a_o->init(net, tb[TCA_ACT_OPTIONS], est, &a, ovr, bind); else - err = a_o->init(net, nla, est, a, ovr, bind); + err = a_o->init(net, nla, est, &a, ovr, bind); if (err < 0) - goto err_free; + goto err_mod; /* module count goes up only when brand new policy is created * if it exists and is only bound to in a_o->init() then @@ -605,8 +590,6 @@ struct tc_action *tcf_action_init_1(struct net *net, struct nlattr *nla, return a; -err_free: - kfree(a); err_mod: module_put(a_o->owner); err_out: @@ -647,7 +630,7 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a, { int err = 0; struct gnet_dump d; - struct tcf_common *p = a->priv; + struct tcf_common *p = (struct tcf_common *)a; if (p == NULL) goto errout; @@ -740,24 +723,11 @@ act_get_notify(struct net *net, u32 portid, struct nlmsghdr *n, return rtnl_unicast(skb, net, portid); } -static struct tc_action *create_a(int i) -{ - struct tc_action *act; - - act = kzalloc(sizeof(*act), GFP_KERNEL); - if (act == NULL) { - pr_debug("create_a: failed to alloc!\n"); - return NULL; - } - act->order = i; - INIT_LIST_HEAD(&act->list); - return act; -} - static struct tc_action *tcf_action_get_1(struct net *net, struct nlattr *nla, struct nlmsghdr *n, u32 portid) { struct nlattr *tb[TCA_ACT_MAX + 1]; + const struct tc_action_ops *ops; struct tc_action *a; int index; int err; @@ -772,26 +742,19 @@ static struct tc_action *tcf_action_get_1(struct net *net, struct nlattr *nla, goto err_out; index = nla_get_u32(tb[TCA_ACT_INDEX]); - err = -ENOMEM; - a = create_a(0); - if (a == NULL) - goto err_out; - err = -EINVAL; - a->ops = tc_lookup_action(tb[TCA_ACT_KIND]); - if (a->ops == NULL) /* could happen in batch of actions */ - goto err_free; + ops = tc_lookup_action(tb[TCA_ACT_KIND]); + if (!ops) /* could happen in batch of actions */ + goto err_out; err = -ENOENT; - if (a->ops->lookup(net, a, index) == 0) + if (ops->lookup(net, &a, index) == 0) goto err_mod; - module_put(a->ops->owner); + module_put(ops->owner); return a; err_mod: - module_put(a->ops->owner); -err_free: - kfree(a); + module_put(ops->owner); err_out: return ERR_PTR(err); } @@ -816,8 +779,8 @@ static int tca_action_flush(struct net *net, struct nlattr *nla, struct netlink_callback dcb; struct nlattr *nest; struct nlattr *tb[TCA_ACT_MAX + 1]; + const struct tc_action_ops *ops; struct nlattr *kind; - struct tc_action a; int err = -ENOMEM; skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); @@ -834,10 +797,8 @@ static int tca_action_flush(struct net *net, struct nlattr *nla, err = -EINVAL; kind = tb[TCA_ACT_KIND]; - memset(&a, 0, sizeof(struct tc_action)); - INIT_LIST_HEAD(&a.list); - a.ops = tc_lookup_action(kind); - if (a.ops == NULL) /*some idjot trying to flush unknown action */ + ops = tc_lookup_action(kind); + if (!ops) /*some idjot trying to flush unknown action */ goto err_out; nlh = nlmsg_put(skb, portid, n->nlmsg_seq, RTM_DELACTION, @@ -853,7 +814,7 @@ static int tca_action_flush(struct net *net, struct nlattr *nla, if (nest == NULL) goto out_module_put; - err = a.ops->walk(net, skb, &dcb, RTM_DELACTION, &a); + err = ops->walk(net, skb, &dcb, RTM_DELACTION, ops); if (err < 0) goto out_module_put; if (err == 0) @@ -863,7 +824,7 @@ static int tca_action_flush(struct net *net, struct nlattr *nla, nlh->nlmsg_len = skb_tail_pointer(skb) - b; nlh->nlmsg_flags |= NLM_F_ROOT; - module_put(a.ops->owner); + module_put(ops->owner); err = rtnetlink_send(skb, net, portid, RTNLGRP_TC, n->nlmsg_flags & NLM_F_ECHO); if (err > 0) @@ -872,7 +833,7 @@ static int tca_action_flush(struct net *net, struct nlattr *nla, return err; out_module_put: - module_put(a.ops->owner); + module_put(ops->owner); err_out: noflush_out: kfree_skb(skb); @@ -1084,7 +1045,6 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) unsigned char *b = skb_tail_pointer(skb); struct nlattr *nest; struct tc_action_ops *a_o; - struct tc_action a; int ret = 0; struct tcamsg *t = (struct tcamsg *) nlmsg_data(cb->nlh); struct nlattr *kind = find_dump_kind(cb->nlh); @@ -1098,9 +1058,6 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) if (a_o == NULL) return 0; - memset(&a, 0, sizeof(struct tc_action)); - a.ops = a_o; - nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, cb->nlh->nlmsg_type, sizeof(*t), 0); if (!nlh) @@ -1114,7 +1071,7 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) if (nest == NULL) goto out_module_put; - ret = a_o->walk(net, skb, cb, RTM_GETACTION, &a); + ret = a_o->walk(net, skb, cb, RTM_GETACTION, a_o); if (ret < 0) goto out_module_put; diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index ef74bff..bfa8707 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -34,11 +34,12 @@ struct tcf_bpf_cfg { }; static int bpf_net_id; +static struct tc_action_ops act_bpf_ops; static int tcf_bpf(struct sk_buff *skb, const struct tc_action *act, struct tcf_result *res) { - struct tcf_bpf *prog = act->priv; + struct tcf_bpf *prog = to_bpf(act); struct bpf_prog *filter; int action, filter_res; bool at_ingress = G_TC_AT(skb->tc_verd) & AT_INGRESS; @@ -134,7 +135,7 @@ static int tcf_bpf_dump(struct sk_buff *skb, struct tc_action *act, int bind, int ref) { unsigned char *tp = skb_tail_pointer(skb); - struct tcf_bpf *prog = act->priv; + struct tcf_bpf *prog = to_bpf(act); struct tc_act_bpf opt = { .index = prog->tcf_index, .refcnt = prog->tcf_refcnt - ref, @@ -270,7 +271,7 @@ static void tcf_bpf_prog_fill_cfg(const struct tcf_bpf *prog, } static int tcf_bpf_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *act, + struct nlattr *est, struct tc_action **act, int replace, int bind) { struct tc_action_net *tn = net_generic(net, bpf_net_id); @@ -295,7 +296,7 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, if (!tcf_hash_check(tn, parm->index, act, bind)) { ret = tcf_hash_create(tn, parm->index, est, act, - sizeof(*prog), bind, true); + &act_bpf_ops, bind, true); if (ret < 0) return ret; @@ -305,7 +306,7 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, if (bind) return 0; - tcf_hash_release(act, bind); + tcf_hash_release(*act, bind); if (!replace) return -EEXIST; } @@ -325,7 +326,7 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, if (ret < 0) goto out; - prog = to_bpf(act); + prog = to_bpf(*act); ASSERT_RTNL(); if (res != ACT_P_CREATED) @@ -343,7 +344,7 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, rcu_assign_pointer(prog->filter, cfg.filter); if (res == ACT_P_CREATED) { - tcf_hash_insert(tn, act); + tcf_hash_insert(tn, *act); } else { /* make sure the program being replaced is no longer executing */ synchronize_rcu(); @@ -353,7 +354,7 @@ static int tcf_bpf_init(struct net *net, struct nlattr *nla, return res; out: if (res == ACT_P_CREATED) - tcf_hash_cleanup(act, est); + tcf_hash_cleanup(*act, est); return ret; } @@ -362,20 +363,20 @@ static void tcf_bpf_cleanup(struct tc_action *act, int bind) { struct tcf_bpf_cfg tmp; - tcf_bpf_prog_fill_cfg(act->priv, &tmp); + tcf_bpf_prog_fill_cfg(to_bpf(act), &tmp); tcf_bpf_cfg_cleanup(&tmp); } static int tcf_bpf_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, bpf_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_bpf_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_bpf_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, bpf_net_id); @@ -392,6 +393,7 @@ static struct tc_action_ops act_bpf_ops __read_mostly = { .init = tcf_bpf_init, .walk = tcf_bpf_walker, .lookup = tcf_bpf_search, + .size = sizeof(struct tcf_bpf), }; static __net_init int bpf_init_net(struct net *net) diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c index 35a5270..eae07a2 100644 --- a/net/sched/act_connmark.c +++ b/net/sched/act_connmark.c @@ -31,6 +31,7 @@ #define CONNMARK_TAB_MASK 3 static int connmark_net_id; +static struct tc_action_ops act_connmark_ops; static int tcf_connmark(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) @@ -38,7 +39,7 @@ static int tcf_connmark(struct sk_buff *skb, const struct tc_action *a, const struct nf_conntrack_tuple_hash *thash; struct nf_conntrack_tuple tuple; enum ip_conntrack_info ctinfo; - struct tcf_connmark_info *ca = a->priv; + struct tcf_connmark_info *ca = to_connmark(a); struct nf_conntrack_zone zone; struct nf_conn *c; int proto; @@ -96,7 +97,7 @@ static const struct nla_policy connmark_policy[TCA_CONNMARK_MAX + 1] = { }; static int tcf_connmark_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, connmark_net_id); @@ -116,22 +117,22 @@ static int tcf_connmark_init(struct net *net, struct nlattr *nla, if (!tcf_hash_check(tn, parm->index, a, bind)) { ret = tcf_hash_create(tn, parm->index, est, a, - sizeof(*ci), bind, false); + &act_connmark_ops, bind, false); if (ret) return ret; - ci = to_connmark(a); + ci = to_connmark(*a); ci->tcf_action = parm->action; ci->net = net; ci->zone = parm->zone; - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); ret = ACT_P_CREATED; } else { - ci = to_connmark(a); + ci = to_connmark(*a); if (bind) return 0; - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; /* replacing action and zone */ @@ -146,7 +147,7 @@ static inline int tcf_connmark_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_connmark_info *ci = a->priv; + struct tcf_connmark_info *ci = to_connmark(a); struct tc_connmark opt = { .index = ci->tcf_index, @@ -173,14 +174,14 @@ nla_put_failure: static int tcf_connmark_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, connmark_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_connmark_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_connmark_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, connmark_net_id); @@ -196,6 +197,7 @@ static struct tc_action_ops act_connmark_ops = { .init = tcf_connmark_init, .walk = tcf_connmark_walker, .lookup = tcf_connmark_search, + .size = sizeof(struct tcf_connmark_info), }; static __net_init int connmark_init_net(struct net *net) diff --git a/net/sched/act_csum.c b/net/sched/act_csum.c index dcd9aba..b5dbf63 100644 --- a/net/sched/act_csum.c +++ b/net/sched/act_csum.c @@ -43,9 +43,10 @@ static const struct nla_policy csum_policy[TCA_CSUM_MAX + 1] = { }; static int csum_net_id; +static struct tc_action_ops act_csum_ops; static int tcf_csum_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, int ovr, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, csum_net_id); @@ -67,26 +68,26 @@ static int tcf_csum_init(struct net *net, struct nlattr *nla, if (!tcf_hash_check(tn, parm->index, a, bind)) { ret = tcf_hash_create(tn, parm->index, est, a, - sizeof(*p), bind, false); + &act_csum_ops, bind, false); if (ret) return ret; ret = ACT_P_CREATED; } else { if (bind)/* dont override defaults */ return 0; - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; } - p = to_tcf_csum(a); + p = to_tcf_csum(*a); spin_lock_bh(&p->tcf_lock); p->tcf_action = parm->action; p->update_flags = parm->update_flags; spin_unlock_bh(&p->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); return ret; } @@ -496,7 +497,7 @@ fail: static int tcf_csum(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_csum *p = a->priv; + struct tcf_csum *p = to_tcf_csum(a); int action; u32 update_flags; @@ -534,7 +535,7 @@ static int tcf_csum_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_csum *p = a->priv; + struct tcf_csum *p = to_tcf_csum(a); struct tc_csum opt = { .update_flags = p->update_flags, .index = p->tcf_index, @@ -560,14 +561,14 @@ nla_put_failure: static int tcf_csum_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, csum_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_csum_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_csum_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, csum_net_id); @@ -583,6 +584,7 @@ static struct tc_action_ops act_csum_ops = { .init = tcf_csum_init, .walk = tcf_csum_walker, .lookup = tcf_csum_search, + .size = sizeof(struct tcf_csum), }; static __net_init int csum_init_net(struct net *net) diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c index 19058a7..e24a409 100644 --- a/net/sched/act_gact.c +++ b/net/sched/act_gact.c @@ -26,6 +26,7 @@ #define GACT_TAB_MASK 15 static int gact_net_id; +static struct tc_action_ops act_gact_ops; #ifdef CONFIG_GACT_PROB static int gact_net_rand(struct tcf_gact *gact) @@ -56,7 +57,7 @@ static const struct nla_policy gact_policy[TCA_GACT_MAX + 1] = { }; static int tcf_gact_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, gact_net_id); @@ -93,19 +94,19 @@ static int tcf_gact_init(struct net *net, struct nlattr *nla, if (!tcf_hash_check(tn, parm->index, a, bind)) { ret = tcf_hash_create(tn, parm->index, est, a, - sizeof(*gact), bind, true); + &act_gact_ops, bind, true); if (ret) return ret; ret = ACT_P_CREATED; } else { if (bind)/* dont override defaults */ return 0; - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; } - gact = to_gact(a); + gact = to_gact(*a); ASSERT_RTNL(); gact->tcf_action = parm->action; @@ -121,14 +122,14 @@ static int tcf_gact_init(struct net *net, struct nlattr *nla, } #endif if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); return ret; } static int tcf_gact(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_gact *gact = a->priv; + struct tcf_gact *gact = to_gact(a); int action = READ_ONCE(gact->tcf_action); #ifdef CONFIG_GACT_PROB @@ -151,7 +152,7 @@ static int tcf_gact(struct sk_buff *skb, const struct tc_action *a, static void tcf_gact_stats_update(struct tc_action *a, u64 bytes, u32 packets, u64 lastuse) { - struct tcf_gact *gact = a->priv; + struct tcf_gact *gact = to_gact(a); int action = READ_ONCE(gact->tcf_action); struct tcf_t *tm = &gact->tcf_tm; @@ -166,7 +167,7 @@ static int tcf_gact_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_gact *gact = a->priv; + struct tcf_gact *gact = to_gact(a); struct tc_gact opt = { .index = gact->tcf_index, .refcnt = gact->tcf_refcnt - ref, @@ -201,14 +202,14 @@ nla_put_failure: static int tcf_gact_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, gact_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_gact_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_gact_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, gact_net_id); @@ -225,6 +226,7 @@ static struct tc_action_ops act_gact_ops = { .init = tcf_gact_init, .walk = tcf_gact_walker, .lookup = tcf_gact_search, + .size = sizeof(struct tcf_gact), }; static __net_init int gact_init_net(struct net *net) diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c index 845ab51..141a06e 100644 --- a/net/sched/act_ife.c +++ b/net/sched/act_ife.c @@ -37,6 +37,7 @@ static int ife_net_id; static int max_metacnt = IFE_META_MAX + 1; +static struct tc_action_ops act_ife_ops; static const struct nla_policy ife_policy[TCA_IFE_MAX + 1] = { [TCA_IFE_PARMS] = { .len = sizeof(struct tc_ife)}, @@ -364,7 +365,7 @@ out_nlmsg_trim: /* under ife->tcf_lock */ static void _tcf_ife_cleanup(struct tc_action *a, int bind) { - struct tcf_ife_info *ife = a->priv; + struct tcf_ife_info *ife = to_ife(a); struct tcf_meta_info *e, *n; list_for_each_entry_safe(e, n, &ife->metalist, metalist) { @@ -382,7 +383,7 @@ static void _tcf_ife_cleanup(struct tc_action *a, int bind) static void tcf_ife_cleanup(struct tc_action *a, int bind) { - struct tcf_ife_info *ife = a->priv; + struct tcf_ife_info *ife = to_ife(a); spin_lock_bh(&ife->tcf_lock); _tcf_ife_cleanup(a, bind); @@ -417,7 +418,7 @@ static int populate_metalist(struct tcf_ife_info *ife, struct nlattr **tb, } static int tcf_ife_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, ife_net_id); @@ -451,25 +452,25 @@ static int tcf_ife_init(struct net *net, struct nlattr *nla, **/ if (!tb[TCA_IFE_TYPE]) { if (exists) - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); pr_info("You MUST pass etherype for encoding\n"); return -EINVAL; } } if (!exists) { - ret = tcf_hash_create(tn, parm->index, est, a, sizeof(*ife), + ret = tcf_hash_create(tn, parm->index, est, a, &act_ife_ops, bind, false); if (ret) return ret; ret = ACT_P_CREATED; } else { - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; } - ife = to_ife(a); + ife = to_ife(*a); ife->flags = parm->flags; if (parm->flags & IFE_ENCODE) { @@ -507,9 +508,9 @@ static int tcf_ife_init(struct net *net, struct nlattr *nla, if (err) { metadata_parse_err: if (exists) - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (ret == ACT_P_CREATED) - _tcf_ife_cleanup(a, bind); + _tcf_ife_cleanup(*a, bind); if (exists) spin_unlock_bh(&ife->tcf_lock); @@ -529,7 +530,7 @@ metadata_parse_err: err = use_all_metadata(ife); if (err) { if (ret == ACT_P_CREATED) - _tcf_ife_cleanup(a, bind); + _tcf_ife_cleanup(*a, bind); if (exists) spin_unlock_bh(&ife->tcf_lock); @@ -541,7 +542,7 @@ metadata_parse_err: spin_unlock_bh(&ife->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); return ret; } @@ -550,7 +551,7 @@ static int tcf_ife_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_ife_info *ife = a->priv; + struct tcf_ife_info *ife = to_ife(a); struct tc_ife opt = { .index = ife->tcf_index, .refcnt = ife->tcf_refcnt - ref, @@ -623,7 +624,7 @@ struct meta_tlvhdr { static int tcf_ife_decode(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_ife_info *ife = a->priv; + struct tcf_ife_info *ife = to_ife(a); int action = ife->tcf_action; struct ifeheadr *ifehdr = (struct ifeheadr *)skb->data; u16 ifehdrln = ifehdr->metalen; @@ -695,7 +696,7 @@ static int ife_get_sz(struct sk_buff *skb, struct tcf_ife_info *ife) static int tcf_ife_encode(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_ife_info *ife = a->priv; + struct tcf_ife_info *ife = to_ife(a); int action = ife->tcf_action; struct ethhdr *oethh; /* outer ether header */ struct ethhdr *iethh; /* inner eth header */ @@ -799,7 +800,7 @@ static int tcf_ife_encode(struct sk_buff *skb, const struct tc_action *a, static int tcf_ife_act(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_ife_info *ife = a->priv; + struct tcf_ife_info *ife = to_ife(a); if (ife->flags & IFE_ENCODE) return tcf_ife_encode(skb, a, res); @@ -819,14 +820,14 @@ static int tcf_ife_act(struct sk_buff *skb, const struct tc_action *a, static int tcf_ife_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, ife_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_ife_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_ife_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, ife_net_id); @@ -843,6 +844,7 @@ static struct tc_action_ops act_ife_ops = { .init = tcf_ife_init, .walk = tcf_ife_walker, .lookup = tcf_ife_search, + .size = sizeof(struct tcf_ife_info), }; static __net_init int ife_init_net(struct net *net) diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index b8c5060..378c1c9 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -31,8 +31,10 @@ #define IPT_TAB_MASK 15 static int ipt_net_id; +static struct tc_action_ops act_ipt_ops; static int xt_net_id; +static struct tc_action_ops act_xt_ops; static int ipt_init_target(struct xt_entry_target *t, char *table, unsigned int hook) @@ -90,8 +92,8 @@ static const struct nla_policy ipt_policy[TCA_IPT_MAX + 1] = { }; static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, int ovr, - int bind) + struct nlattr *est, struct tc_action **a, + const struct tc_action_ops *ops, int ovr, int bind) { struct nlattr *tb[TCA_IPT_MAX + 1]; struct tcf_ipt *ipt; @@ -118,19 +120,19 @@ static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla, if (tb[TCA_IPT_HOOK] == NULL || tb[TCA_IPT_TARG] == NULL) { if (exists) - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); return -EINVAL; } td = (struct xt_entry_target *)nla_data(tb[TCA_IPT_TARG]); if (nla_len(tb[TCA_IPT_TARG]) < td->u.target_size) { if (exists) - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); return -EINVAL; } if (!exists) { - ret = tcf_hash_create(tn, index, est, a, sizeof(*ipt), bind, + ret = tcf_hash_create(tn, index, est, a, ops, bind, false); if (ret) return ret; @@ -138,13 +140,11 @@ static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla, } else { if (bind)/* dont override defaults */ return 0; - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; } - ipt = to_ipt(a); - hook = nla_get_u32(tb[TCA_IPT_HOOK]); err = -ENOMEM; @@ -163,6 +163,8 @@ static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla, if (err < 0) goto err3; + ipt = to_ipt(*a); + spin_lock_bh(&ipt->tcf_lock); if (ret != ACT_P_CREATED) { ipt_destroy_target(ipt->tcfi_t); @@ -174,7 +176,7 @@ static int __tcf_ipt_init(struct tc_action_net *tn, struct nlattr *nla, ipt->tcfi_hook = hook; spin_unlock_bh(&ipt->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); return ret; err3: @@ -183,33 +185,33 @@ err2: kfree(tname); err1: if (ret == ACT_P_CREATED) - tcf_hash_cleanup(a, est); + tcf_hash_cleanup(*a, est); return err; } static int tcf_ipt_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, int ovr, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, ipt_net_id); - return __tcf_ipt_init(tn, nla, est, a, ovr, bind); + return __tcf_ipt_init(tn, nla, est, a, &act_ipt_ops, ovr, bind); } static int tcf_xt_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, int ovr, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, xt_net_id); - return __tcf_ipt_init(tn, nla, est, a, ovr, bind); + return __tcf_ipt_init(tn, nla, est, a, &act_xt_ops, ovr, bind); } static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { int ret = 0, result = 0; - struct tcf_ipt *ipt = a->priv; + struct tcf_ipt *ipt = to_ipt(a); struct xt_action_param par; if (skb_unclone(skb, GFP_ATOMIC)) @@ -259,7 +261,7 @@ static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_ipt *ipt = a->priv; + struct tcf_ipt *ipt = to_ipt(a); struct xt_entry_target *t; struct tcf_t tm; struct tc_cnt c; @@ -299,14 +301,14 @@ nla_put_failure: static int tcf_ipt_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, ipt_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_ipt_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_ipt_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, ipt_net_id); @@ -323,6 +325,7 @@ static struct tc_action_ops act_ipt_ops = { .init = tcf_ipt_init, .walk = tcf_ipt_walker, .lookup = tcf_ipt_search, + .size = sizeof(struct tcf_ipt), }; static __net_init int ipt_init_net(struct net *net) @@ -348,14 +351,14 @@ static struct pernet_operations ipt_net_ops = { static int tcf_xt_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, xt_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_xt_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_xt_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, xt_net_id); @@ -372,6 +375,7 @@ static struct tc_action_ops act_xt_ops = { .init = tcf_xt_init, .walk = tcf_xt_walker, .lookup = tcf_xt_search, + .size = sizeof(struct tcf_ipt), }; static __net_init int xt_init_net(struct net *net) diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 70cfbbf..6038c85 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -52,9 +52,10 @@ static const struct nla_policy mirred_policy[TCA_MIRRED_MAX + 1] = { }; static int mirred_net_id; +static struct tc_action_ops act_mirred_ops; static int tcf_mirred_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, int ovr, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, mirred_net_id); @@ -84,14 +85,14 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla, break; default: if (exists) - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); return -EINVAL; } if (parm->ifindex) { dev = __dev_get_by_index(net, parm->ifindex); if (dev == NULL) { if (exists) - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); return -ENODEV; } switch (dev->type) { @@ -115,16 +116,16 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla, if (dev == NULL) return -EINVAL; ret = tcf_hash_create(tn, parm->index, est, a, - sizeof(*m), bind, true); + &act_mirred_ops, bind, true); if (ret) return ret; ret = ACT_P_CREATED; } else { - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; } - m = to_mirred(a); + m = to_mirred(*a); ASSERT_RTNL(); m->tcf_action = parm->action; @@ -142,7 +143,7 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla, spin_lock_bh(&mirred_list_lock); list_add(&m->tcfm_list, &mirred_list); spin_unlock_bh(&mirred_list_lock); - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); } return ret; @@ -151,7 +152,7 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla, static int tcf_mirred(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_mirred *m = a->priv; + struct tcf_mirred *m = to_mirred(a); struct net_device *dev; struct sk_buff *skb2; int retval, err; @@ -206,7 +207,7 @@ out: static int tcf_mirred_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_mirred *m = a->priv; + struct tcf_mirred *m = to_mirred(a); struct tc_mirred opt = { .index = m->tcf_index, .action = m->tcf_action, @@ -232,14 +233,14 @@ nla_put_failure: static int tcf_mirred_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, mirred_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_mirred_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_mirred_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, mirred_net_id); @@ -284,6 +285,7 @@ static struct tc_action_ops act_mirred_ops = { .init = tcf_mirred_init, .walk = tcf_mirred_walker, .lookup = tcf_mirred_search, + .size = sizeof(struct tcf_mirred), }; static __net_init int mirred_init_net(struct net *net) diff --git a/net/sched/act_nat.c b/net/sched/act_nat.c index 06ccb03..8e8b0cc 100644 --- a/net/sched/act_nat.c +++ b/net/sched/act_nat.c @@ -32,13 +32,14 @@ #define NAT_TAB_MASK 15 static int nat_net_id; +static struct tc_action_ops act_nat_ops; static const struct nla_policy nat_policy[TCA_NAT_MAX + 1] = { [TCA_NAT_PARMS] = { .len = sizeof(struct tc_nat) }, }; static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est, - struct tc_action *a, int ovr, int bind) + struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, nat_net_id); struct nlattr *tb[TCA_NAT_MAX + 1]; @@ -59,18 +60,18 @@ static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est, if (!tcf_hash_check(tn, parm->index, a, bind)) { ret = tcf_hash_create(tn, parm->index, est, a, - sizeof(*p), bind, false); + &act_nat_ops, bind, false); if (ret) return ret; ret = ACT_P_CREATED; } else { if (bind) return 0; - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; } - p = to_tcf_nat(a); + p = to_tcf_nat(*a); spin_lock_bh(&p->tcf_lock); p->old_addr = parm->old_addr; @@ -82,7 +83,7 @@ static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est, spin_unlock_bh(&p->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); return ret; } @@ -90,7 +91,7 @@ static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est, static int tcf_nat(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_nat *p = a->priv; + struct tcf_nat *p = to_tcf_nat(a); struct iphdr *iph; __be32 old_addr; __be32 new_addr; @@ -248,7 +249,7 @@ static int tcf_nat_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_nat *p = a->priv; + struct tcf_nat *p = to_tcf_nat(a); struct tc_nat opt = { .old_addr = p->old_addr, .new_addr = p->new_addr, @@ -278,14 +279,14 @@ nla_put_failure: static int tcf_nat_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, nat_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_nat_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_nat_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, nat_net_id); @@ -301,6 +302,7 @@ static struct tc_action_ops act_nat_ops = { .init = tcf_nat_init, .walk = tcf_nat_walker, .lookup = tcf_nat_search, + .size = sizeof(struct tcf_nat), }; static __net_init int nat_init_net(struct net *net) diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index 82d3c14..b54d56d 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -26,13 +26,14 @@ #define PEDIT_TAB_MASK 15 static int pedit_net_id; +static struct tc_action_ops act_pedit_ops; static const struct nla_policy pedit_policy[TCA_PEDIT_MAX + 1] = { [TCA_PEDIT_PARMS] = { .len = sizeof(struct tc_pedit) }, }; static int tcf_pedit_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, pedit_net_id); @@ -61,23 +62,23 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, if (!parm->nkeys) return -EINVAL; ret = tcf_hash_create(tn, parm->index, est, a, - sizeof(*p), bind, false); + &act_pedit_ops, bind, false); if (ret) return ret; - p = to_pedit(a); + p = to_pedit(*a); keys = kmalloc(ksize, GFP_KERNEL); if (keys == NULL) { - tcf_hash_cleanup(a, est); + tcf_hash_cleanup(*a, est); return -ENOMEM; } ret = ACT_P_CREATED; } else { if (bind) return 0; - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; - p = to_pedit(a); + p = to_pedit(*a); if (p->tcfp_nkeys && p->tcfp_nkeys != parm->nkeys) { keys = kmalloc(ksize, GFP_KERNEL); if (keys == NULL) @@ -96,13 +97,13 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, memcpy(p->tcfp_keys, parm->keys, ksize); spin_unlock_bh(&p->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); return ret; } static void tcf_pedit_cleanup(struct tc_action *a, int bind) { - struct tcf_pedit *p = a->priv; + struct tcf_pedit *p = to_pedit(a); struct tc_pedit_key *keys = p->tcfp_keys; kfree(keys); } @@ -110,7 +111,7 @@ static void tcf_pedit_cleanup(struct tc_action *a, int bind) static int tcf_pedit(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_pedit *p = a->priv; + struct tcf_pedit *p = to_pedit(a); int i; unsigned int off; @@ -177,7 +178,7 @@ static int tcf_pedit_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_pedit *p = a->priv; + struct tcf_pedit *p = to_pedit(a); struct tc_pedit *opt; struct tcf_t t; int s; @@ -216,14 +217,14 @@ nla_put_failure: static int tcf_pedit_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, pedit_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_pedit_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_pedit_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, pedit_net_id); @@ -240,6 +241,7 @@ static struct tc_action_ops act_pedit_ops = { .init = tcf_pedit_init, .walk = tcf_pedit_walker, .lookup = tcf_pedit_search, + .size = sizeof(struct tcf_pedit), }; static __net_init int pedit_init_net(struct net *net) diff --git a/net/sched/act_police.c b/net/sched/act_police.c index 1e8ede3..123794a 100644 --- a/net/sched/act_police.c +++ b/net/sched/act_police.c @@ -37,8 +37,8 @@ struct tcf_police { struct psched_ratecfg peak; bool peak_present; }; -#define to_police(pc) \ - container_of(pc->priv, struct tcf_police, common) + +#define to_police(pc) ((struct tcf_police *)pc) #define POL_TAB_MASK 15 @@ -56,15 +56,14 @@ struct tc_police_compat { /* Each policer is serialized by its individual spinlock */ static int police_net_id; +static struct tc_action_ops act_police_ops; static int tcf_act_police_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, police_net_id); struct tcf_hashinfo *hinfo = tn->hinfo; - struct hlist_head *head; - struct tcf_common *p; int err = 0, index = -1, i = 0, s_i = 0, n_i = 0; struct nlattr *nest; @@ -73,21 +72,22 @@ static int tcf_act_police_walker(struct net *net, struct sk_buff *skb, s_i = cb->args[0]; for (i = 0; i < (POL_TAB_MASK + 1); i++) { + struct hlist_head *head; + struct tcf_common *p; + head = &hinfo->htab[tcf_hash(i, POL_TAB_MASK)]; hlist_for_each_entry_rcu(p, head, tcfc_head) { index++; if (index < s_i) continue; - a->priv = p; - a->order = index; - nest = nla_nest_start(skb, a->order); + nest = nla_nest_start(skb, index); if (nest == NULL) goto nla_put_failure; if (type == RTM_DELACTION) - err = tcf_action_dump_1(skb, a, 0, 1); + err = tcf_action_dump_1(skb, (struct tc_action *)p, 0, 1); else - err = tcf_action_dump_1(skb, a, 0, 0); + err = tcf_action_dump_1(skb, (struct tc_action *)p, 0, 0); if (err < 0) { index--; nla_nest_cancel(skb, nest); @@ -116,7 +116,7 @@ static const struct nla_policy police_policy[TCA_POLICE_MAX + 1] = { }; static int tcf_act_police_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { int ret = 0, err; @@ -142,13 +142,7 @@ static int tcf_act_police_init(struct net *net, struct nlattr *nla, parm = nla_data(tb[TCA_POLICE_TBF]); if (parm->index) { - if (tcf_hash_search(tn, a, parm->index)) { - police = to_police(a); - if (bind) { - police->tcf_bindcnt += 1; - police->tcf_refcnt += 1; - return 0; - } + if (tcf_hash_check(tn, parm->index, a, bind)) { if (ovr) goto override; /* not replacing */ @@ -156,14 +150,14 @@ static int tcf_act_police_init(struct net *net, struct nlattr *nla, } } else { ret = tcf_hash_create(tn, parm->index, NULL, a, - sizeof(*police), bind, false); + &act_police_ops, bind, false); if (ret) return ret; ret = ACT_P_CREATED; } - police = to_police(a); override: + police = to_police(*a); if (parm->rate.rate) { err = -ENOMEM; R_tab = qdisc_get_rtab(&parm->rate, tb[TCA_POLICE_RATE]); @@ -235,7 +229,7 @@ override: return ret; police->tcfp_t_c = ktime_get_ns(); - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); return ret; @@ -245,14 +239,14 @@ failure: qdisc_put_rtab(P_tab); qdisc_put_rtab(R_tab); if (ret == ACT_P_CREATED) - tcf_hash_cleanup(a, est); + tcf_hash_cleanup(*a, est); return err; } static int tcf_act_police(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_police *police = a->priv; + struct tcf_police *police = to_police(a); s64 now; s64 toks; s64 ptoks = 0; @@ -311,7 +305,7 @@ static int tcf_act_police_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_police *police = a->priv; + struct tcf_police *police = to_police(a); struct tc_police opt = { .index = police->tcf_index, .action = police->tcf_action, @@ -349,7 +343,7 @@ nla_put_failure: return -1; } -static int tcf_police_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_police_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, police_net_id); @@ -369,6 +363,7 @@ static struct tc_action_ops act_police_ops = { .init = tcf_act_police_init, .walk = tcf_act_police_walker, .lookup = tcf_police_search, + .size = sizeof(struct tcf_police), }; static __net_init int police_init_net(struct net *net) diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c index 318328d3..289af6f 100644 --- a/net/sched/act_simple.c +++ b/net/sched/act_simple.c @@ -27,12 +27,13 @@ #define SIMP_TAB_MASK 7 static int simp_net_id; +static struct tc_action_ops act_simp_ops; #define SIMP_MAX_DATA 32 static int tcf_simp(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_defact *d = a->priv; + struct tcf_defact *d = to_defact(a); spin_lock(&d->tcf_lock); tcf_lastuse_update(&d->tcf_tm); @@ -79,7 +80,7 @@ static const struct nla_policy simple_policy[TCA_DEF_MAX + 1] = { }; static int tcf_simp_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, simp_net_id); @@ -100,7 +101,6 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla, if (tb[TCA_DEF_PARMS] == NULL) return -EINVAL; - parm = nla_data(tb[TCA_DEF_PARMS]); exists = tcf_hash_check(tn, parm->index, a, bind); if (exists && bind) @@ -108,7 +108,7 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla, if (tb[TCA_DEF_DATA] == NULL) { if (exists) - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); return -EINVAL; } @@ -116,22 +116,22 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla, if (!exists) { ret = tcf_hash_create(tn, parm->index, est, a, - sizeof(*d), bind, false); + &act_simp_ops, bind, false); if (ret) return ret; - d = to_defact(a); + d = to_defact(*a); ret = alloc_defdata(d, defdata); if (ret < 0) { - tcf_hash_cleanup(a, est); + tcf_hash_cleanup(*a, est); return ret; } d->tcf_action = parm->action; ret = ACT_P_CREATED; } else { - d = to_defact(a); + d = to_defact(*a); - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; @@ -139,7 +139,7 @@ static int tcf_simp_init(struct net *net, struct nlattr *nla, } if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); return ret; } @@ -147,7 +147,7 @@ static int tcf_simp_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_defact *d = a->priv; + struct tcf_defact *d = to_defact(a); struct tc_defact opt = { .index = d->tcf_index, .refcnt = d->tcf_refcnt - ref, @@ -172,14 +172,14 @@ nla_put_failure: static int tcf_simp_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, simp_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_simp_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_simp_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, simp_net_id); @@ -196,6 +196,7 @@ static struct tc_action_ops act_simp_ops = { .init = tcf_simp_init, .walk = tcf_simp_walker, .lookup = tcf_simp_search, + .size = sizeof(struct tcf_defact), }; static __net_init int simp_init_net(struct net *net) diff --git a/net/sched/act_skbedit.c b/net/sched/act_skbedit.c index 8e573c0..a133dcb 100644 --- a/net/sched/act_skbedit.c +++ b/net/sched/act_skbedit.c @@ -30,11 +30,12 @@ #define SKBEDIT_TAB_MASK 15 static int skbedit_net_id; +static struct tc_action_ops act_skbedit_ops; static int tcf_skbedit(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_skbedit *d = a->priv; + struct tcf_skbedit *d = to_skbedit(a); spin_lock(&d->tcf_lock); tcf_lastuse_update(&d->tcf_tm); @@ -63,7 +64,7 @@ static const struct nla_policy skbedit_policy[TCA_SKBEDIT_MAX + 1] = { }; static int tcf_skbedit_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, skbedit_net_id); @@ -114,21 +115,21 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, return 0; if (!flags) { - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); return -EINVAL; } if (!exists) { ret = tcf_hash_create(tn, parm->index, est, a, - sizeof(*d), bind, false); + &act_skbedit_ops, bind, false); if (ret) return ret; - d = to_skbedit(a); + d = to_skbedit(*a); ret = ACT_P_CREATED; } else { - d = to_skbedit(a); - tcf_hash_release(a, bind); + d = to_skbedit(*a); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; } @@ -150,7 +151,7 @@ static int tcf_skbedit_init(struct net *net, struct nlattr *nla, spin_unlock_bh(&d->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); return ret; } @@ -158,7 +159,7 @@ static int tcf_skbedit_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_skbedit *d = a->priv; + struct tcf_skbedit *d = to_skbedit(a); struct tc_skbedit opt = { .index = d->tcf_index, .refcnt = d->tcf_refcnt - ref, @@ -194,14 +195,14 @@ nla_put_failure: static int tcf_skbedit_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, skbedit_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_skbedit_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_skbedit_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, skbedit_net_id); @@ -217,6 +218,7 @@ static struct tc_action_ops act_skbedit_ops = { .init = tcf_skbedit_init, .walk = tcf_skbedit_walker, .lookup = tcf_skbedit_search, + .size = sizeof(struct tcf_skbedit), }; static __net_init int skbedit_init_net(struct net *net) diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index db9b7ed..691409d 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -22,11 +22,12 @@ #define VLAN_TAB_MASK 15 static int vlan_net_id; +static struct tc_action_ops act_vlan_ops; static int tcf_vlan(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { - struct tcf_vlan *v = a->priv; + struct tcf_vlan *v = to_vlan(a); int action; int err; @@ -67,7 +68,7 @@ static const struct nla_policy vlan_policy[TCA_VLAN_MAX + 1] = { }; static int tcf_vlan_init(struct net *net, struct nlattr *nla, - struct nlattr *est, struct tc_action *a, + struct nlattr *est, struct tc_action **a, int ovr, int bind) { struct tc_action_net *tn = net_generic(net, vlan_net_id); @@ -100,13 +101,13 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla, case TCA_VLAN_ACT_PUSH: if (!tb[TCA_VLAN_PUSH_VLAN_ID]) { if (exists) - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); return -EINVAL; } push_vid = nla_get_u16(tb[TCA_VLAN_PUSH_VLAN_ID]); if (push_vid >= VLAN_VID_MASK) { if (exists) - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); return -ERANGE; } @@ -125,25 +126,25 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla, break; default: if (exists) - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); return -EINVAL; } action = parm->v_action; if (!exists) { ret = tcf_hash_create(tn, parm->index, est, a, - sizeof(*v), bind, false); + &act_vlan_ops, bind, false); if (ret) return ret; ret = ACT_P_CREATED; } else { - tcf_hash_release(a, bind); + tcf_hash_release(*a, bind); if (!ovr) return -EEXIST; } - v = to_vlan(a); + v = to_vlan(*a); spin_lock_bh(&v->tcf_lock); @@ -156,7 +157,7 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla, spin_unlock_bh(&v->tcf_lock); if (ret == ACT_P_CREATED) - tcf_hash_insert(tn, a); + tcf_hash_insert(tn, *a); return ret; } @@ -164,7 +165,7 @@ static int tcf_vlan_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); - struct tcf_vlan *v = a->priv; + struct tcf_vlan *v = to_vlan(a); struct tc_vlan opt = { .index = v->tcf_index, .refcnt = v->tcf_refcnt - ref, @@ -195,14 +196,14 @@ nla_put_failure: static int tcf_vlan_walker(struct net *net, struct sk_buff *skb, struct netlink_callback *cb, int type, - struct tc_action *a) + const struct tc_action_ops *ops) { struct tc_action_net *tn = net_generic(net, vlan_net_id); - return tcf_generic_walker(tn, skb, cb, type, a); + return tcf_generic_walker(tn, skb, cb, type, ops); } -static int tcf_vlan_search(struct net *net, struct tc_action *a, u32 index) +static int tcf_vlan_search(struct net *net, struct tc_action **a, u32 index) { struct tc_action_net *tn = net_generic(net, vlan_net_id); @@ -218,6 +219,7 @@ static struct tc_action_ops act_vlan_ops = { .init = tcf_vlan_init, .walk = tcf_vlan_walker, .lookup = tcf_vlan_search, + .size = sizeof(struct tcf_vlan), }; static __net_init int vlan_init_net(struct net *net) -- cgit v0.10.2 From ec0595cc4495be579309b4bfd5e997af0f2ae6f9 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Mon, 25 Jul 2016 16:09:42 -0700 Subject: net_sched: get rid of struct tcf_common After the previous patch, struct tc_action should be enough to represent the generic tc action, tcf_common is not necessary any more. This patch gets rid of it to make tc action code more readable. Cc: Jamal Hadi Salim Signed-off-by: Cong Wang Signed-off-by: David S. Miller diff --git a/include/net/act_api.h b/include/net/act_api.h index 8b19909..41e6a24 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -22,42 +22,39 @@ struct tc_action_ops; struct tc_action { const struct tc_action_ops *ops; - __u32 type; /* for backward compat(TCA_OLD_COMPAT) */ - __u32 order; - struct list_head list; - struct tcf_hashinfo *hinfo; -}; - -struct tcf_common { - struct tc_action tcfc_act; - struct hlist_node tcfc_head; - u32 tcfc_index; - int tcfc_refcnt; - int tcfc_bindcnt; - u32 tcfc_capab; - int tcfc_action; - struct tcf_t tcfc_tm; - struct gnet_stats_basic_packed tcfc_bstats; - struct gnet_stats_queue tcfc_qstats; - struct gnet_stats_rate_est64 tcfc_rate_est; - spinlock_t tcfc_lock; - struct rcu_head tcfc_rcu; + __u32 type; /* for backward compat(TCA_OLD_COMPAT) */ + __u32 order; + struct list_head list; + struct tcf_hashinfo *hinfo; + + struct hlist_node tcfa_head; + u32 tcfa_index; + int tcfa_refcnt; + int tcfa_bindcnt; + u32 tcfa_capab; + int tcfa_action; + struct tcf_t tcfa_tm; + struct gnet_stats_basic_packed tcfa_bstats; + struct gnet_stats_queue tcfa_qstats; + struct gnet_stats_rate_est64 tcfa_rate_est; + spinlock_t tcfa_lock; + struct rcu_head tcfa_rcu; struct gnet_stats_basic_cpu __percpu *cpu_bstats; struct gnet_stats_queue __percpu *cpu_qstats; }; -#define tcf_act common.tcfc_act -#define tcf_head common.tcfc_head -#define tcf_index common.tcfc_index -#define tcf_refcnt common.tcfc_refcnt -#define tcf_bindcnt common.tcfc_bindcnt -#define tcf_capab common.tcfc_capab -#define tcf_action common.tcfc_action -#define tcf_tm common.tcfc_tm -#define tcf_bstats common.tcfc_bstats -#define tcf_qstats common.tcfc_qstats -#define tcf_rate_est common.tcfc_rate_est -#define tcf_lock common.tcfc_lock -#define tcf_rcu common.tcfc_rcu +#define tcf_act common.tcfa_act +#define tcf_head common.tcfa_head +#define tcf_index common.tcfa_index +#define tcf_refcnt common.tcfa_refcnt +#define tcf_bindcnt common.tcfa_bindcnt +#define tcf_capab common.tcfa_capab +#define tcf_action common.tcfa_action +#define tcf_tm common.tcfa_tm +#define tcf_bstats common.tcfa_bstats +#define tcf_qstats common.tcfa_qstats +#define tcf_rate_est common.tcfa_rate_est +#define tcf_lock common.tcfa_lock +#define tcf_rcu common.tcfa_rcu static inline unsigned int tcf_hash(u32 index, unsigned int hmask) { diff --git a/include/net/tc_act/tc_bpf.h b/include/net/tc_act/tc_bpf.h index 80a4d6f..2b94673 100644 --- a/include/net/tc_act/tc_bpf.h +++ b/include/net/tc_act/tc_bpf.h @@ -14,7 +14,7 @@ #include struct tcf_bpf { - struct tcf_common common; + struct tc_action common; struct bpf_prog __rcu *filter; union { u32 bpf_fd; diff --git a/include/net/tc_act/tc_connmark.h b/include/net/tc_act/tc_connmark.h index 8a66113..59b515d 100644 --- a/include/net/tc_act/tc_connmark.h +++ b/include/net/tc_act/tc_connmark.h @@ -4,7 +4,7 @@ #include struct tcf_connmark_info { - struct tcf_common common; + struct tc_action common; struct net *net; u16 zone; }; diff --git a/include/net/tc_act/tc_csum.h b/include/net/tc_act/tc_csum.h index 1a9ef15..f31fb63 100644 --- a/include/net/tc_act/tc_csum.h +++ b/include/net/tc_act/tc_csum.h @@ -5,7 +5,7 @@ #include struct tcf_csum { - struct tcf_common common; + struct tc_action common; u32 update_flags; }; diff --git a/include/net/tc_act/tc_defact.h b/include/net/tc_act/tc_defact.h index e25b4eb..d47f040 100644 --- a/include/net/tc_act/tc_defact.h +++ b/include/net/tc_act/tc_defact.h @@ -4,7 +4,7 @@ #include struct tcf_defact { - struct tcf_common common; + struct tc_action common; u32 tcfd_datalen; void *tcfd_defdata; }; diff --git a/include/net/tc_act/tc_gact.h b/include/net/tc_act/tc_gact.h index 119cdb4..b6f1739 100644 --- a/include/net/tc_act/tc_gact.h +++ b/include/net/tc_act/tc_gact.h @@ -5,7 +5,7 @@ #include struct tcf_gact { - struct tcf_common common; + struct tc_action common; #ifdef CONFIG_GACT_PROB u16 tcfg_ptype; u16 tcfg_pval; diff --git a/include/net/tc_act/tc_ife.h b/include/net/tc_act/tc_ife.h index 7921abe..5164bd7 100644 --- a/include/net/tc_act/tc_ife.h +++ b/include/net/tc_act/tc_ife.h @@ -8,7 +8,7 @@ #define IFE_METAHDRLEN 2 struct tcf_ife_info { - struct tcf_common common; + struct tc_action common; u8 eth_dst[ETH_ALEN]; u8 eth_src[ETH_ALEN]; u16 eth_type; diff --git a/include/net/tc_act/tc_ipt.h b/include/net/tc_act/tc_ipt.h index c22ae7a..3130976 100644 --- a/include/net/tc_act/tc_ipt.h +++ b/include/net/tc_act/tc_ipt.h @@ -6,7 +6,7 @@ struct xt_entry_target; struct tcf_ipt { - struct tcf_common common; + struct tc_action common; u32 tcfi_hook; char *tcfi_tname; struct xt_entry_target *tcfi_t; diff --git a/include/net/tc_act/tc_mirred.h b/include/net/tc_act/tc_mirred.h index 89aebd2..62770ad 100644 --- a/include/net/tc_act/tc_mirred.h +++ b/include/net/tc_act/tc_mirred.h @@ -5,7 +5,7 @@ #include struct tcf_mirred { - struct tcf_common common; + struct tc_action common; int tcfm_eaction; int tcfm_ifindex; int tcfm_ok_push; diff --git a/include/net/tc_act/tc_nat.h b/include/net/tc_act/tc_nat.h index a91ad3a..56681a3 100644 --- a/include/net/tc_act/tc_nat.h +++ b/include/net/tc_act/tc_nat.h @@ -5,7 +5,7 @@ #include struct tcf_nat { - struct tcf_common common; + struct tc_action common; __be32 old_addr; __be32 new_addr; diff --git a/include/net/tc_act/tc_pedit.h b/include/net/tc_act/tc_pedit.h index 2cccfba..29e38d6 100644 --- a/include/net/tc_act/tc_pedit.h +++ b/include/net/tc_act/tc_pedit.h @@ -4,7 +4,7 @@ #include struct tcf_pedit { - struct tcf_common common; + struct tc_action common; unsigned char tcfp_nkeys; unsigned char tcfp_flags; struct tc_pedit_key *tcfp_keys; diff --git a/include/net/tc_act/tc_skbedit.h b/include/net/tc_act/tc_skbedit.h index 9e05489..5767e9d 100644 --- a/include/net/tc_act/tc_skbedit.h +++ b/include/net/tc_act/tc_skbedit.h @@ -23,7 +23,7 @@ #include struct tcf_skbedit { - struct tcf_common common; + struct tc_action common; u32 flags; u32 priority; u32 mark; diff --git a/include/net/tc_act/tc_vlan.h b/include/net/tc_act/tc_vlan.h index 584b807..e29f52e 100644 --- a/include/net/tc_act/tc_vlan.h +++ b/include/net/tc_act/tc_vlan.h @@ -16,7 +16,7 @@ #define VLAN_F_PUSH 0x2 struct tcf_vlan { - struct tcf_common common; + struct tc_action common; int tcfv_action; u16 tcfv_push_vid; __be16 tcfv_push_proto; diff --git a/net/sched/act_api.c b/net/sched/act_api.c index d97419f..e4a5f26 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -29,46 +29,43 @@ static void free_tcf(struct rcu_head *head) { - struct tcf_common *p = container_of(head, struct tcf_common, tcfc_rcu); + struct tc_action *p = container_of(head, struct tc_action, tcfa_rcu); free_percpu(p->cpu_bstats); free_percpu(p->cpu_qstats); kfree(p); } -static void tcf_hash_destroy(struct tcf_hashinfo *hinfo, struct tc_action *a) +static void tcf_hash_destroy(struct tcf_hashinfo *hinfo, struct tc_action *p) { - struct tcf_common *p = (struct tcf_common *)a; - spin_lock_bh(&hinfo->lock); - hlist_del(&p->tcfc_head); + hlist_del(&p->tcfa_head); spin_unlock_bh(&hinfo->lock); - gen_kill_estimator(&p->tcfc_bstats, - &p->tcfc_rate_est); + gen_kill_estimator(&p->tcfa_bstats, + &p->tcfa_rate_est); /* - * gen_estimator est_timer() might access p->tcfc_lock + * gen_estimator est_timer() might access p->tcfa_lock * or bstats, wait a RCU grace period before freeing p */ - call_rcu(&p->tcfc_rcu, free_tcf); + call_rcu(&p->tcfa_rcu, free_tcf); } -int __tcf_hash_release(struct tc_action *a, bool bind, bool strict) +int __tcf_hash_release(struct tc_action *p, bool bind, bool strict) { - struct tcf_common *p = (struct tcf_common *)a; int ret = 0; if (p) { if (bind) - p->tcfc_bindcnt--; - else if (strict && p->tcfc_bindcnt > 0) + p->tcfa_bindcnt--; + else if (strict && p->tcfa_bindcnt > 0) return -EPERM; - p->tcfc_refcnt--; - if (p->tcfc_bindcnt <= 0 && p->tcfc_refcnt <= 0) { - if (a->ops->cleanup) - a->ops->cleanup(a, bind); - list_del(&a->list); - tcf_hash_destroy(a->hinfo, a); + p->tcfa_refcnt--; + if (p->tcfa_bindcnt <= 0 && p->tcfa_refcnt <= 0) { + if (p->ops->cleanup) + p->ops->cleanup(p, bind); + list_del(&p->list); + tcf_hash_destroy(p->hinfo, p); ret = ACT_P_DELETED; } } @@ -89,11 +86,11 @@ static int tcf_dump_walker(struct tcf_hashinfo *hinfo, struct sk_buff *skb, for (i = 0; i < (hinfo->hmask + 1); i++) { struct hlist_head *head; - struct tcf_common *p; + struct tc_action *p; head = &hinfo->htab[tcf_hash(i, hinfo->hmask)]; - hlist_for_each_entry_rcu(p, head, tcfc_head) { + hlist_for_each_entry_rcu(p, head, tcfa_head) { index++; if (index < s_i) continue; @@ -101,7 +98,7 @@ static int tcf_dump_walker(struct tcf_hashinfo *hinfo, struct sk_buff *skb, nest = nla_nest_start(skb, n_i); if (nest == NULL) goto nla_put_failure; - err = tcf_action_dump_1(skb, (struct tc_action *)p, 0, 0); + err = tcf_action_dump_1(skb, p, 0, 0); if (err < 0) { index--; nlmsg_trim(skb, nest); @@ -139,13 +136,13 @@ static int tcf_del_walker(struct tcf_hashinfo *hinfo, struct sk_buff *skb, for (i = 0; i < (hinfo->hmask + 1); i++) { struct hlist_head *head; struct hlist_node *n; - struct tcf_common *p; + struct tc_action *p; head = &hinfo->htab[tcf_hash(i, hinfo->hmask)]; - hlist_for_each_entry_safe(p, n, head, tcfc_head) { - ret = __tcf_hash_release((struct tc_action *)p, false, true); + hlist_for_each_entry_safe(p, n, head, tcfa_head) { + ret = __tcf_hash_release(p, false, true); if (ret == ACT_P_DELETED) { - module_put(p->tcfc_act.ops->owner); + module_put(p->ops->owner); n_i++; } else if (ret < 0) goto nla_put_failure; @@ -178,15 +175,15 @@ int tcf_generic_walker(struct tc_action_net *tn, struct sk_buff *skb, } EXPORT_SYMBOL(tcf_generic_walker); -static struct tcf_common *tcf_hash_lookup(u32 index, struct tcf_hashinfo *hinfo) +static struct tc_action *tcf_hash_lookup(u32 index, struct tcf_hashinfo *hinfo) { - struct tcf_common *p = NULL; + struct tc_action *p = NULL; struct hlist_head *head; spin_lock_bh(&hinfo->lock); head = &hinfo->htab[tcf_hash(index, hinfo->hmask)]; - hlist_for_each_entry_rcu(p, head, tcfc_head) - if (p->tcfc_index == index) + hlist_for_each_entry_rcu(p, head, tcfa_head) + if (p->tcfa_index == index) break; spin_unlock_bh(&hinfo->lock); @@ -211,10 +208,10 @@ EXPORT_SYMBOL(tcf_hash_new_index); int tcf_hash_search(struct tc_action_net *tn, struct tc_action **a, u32 index) { struct tcf_hashinfo *hinfo = tn->hinfo; - struct tcf_common *p = tcf_hash_lookup(index, hinfo); + struct tc_action *p = tcf_hash_lookup(index, hinfo); if (p) { - *a = &p->tcfc_act; + *a = p; return 1; } return 0; @@ -225,12 +222,13 @@ bool tcf_hash_check(struct tc_action_net *tn, u32 index, struct tc_action **a, int bind) { struct tcf_hashinfo *hinfo = tn->hinfo; - struct tcf_common *p = NULL; + struct tc_action *p = NULL; + if (index && (p = tcf_hash_lookup(index, hinfo)) != NULL) { if (bind) - p->tcfc_bindcnt++; - p->tcfc_refcnt++; - *a = &p->tcfc_act; + p->tcfa_bindcnt++; + p->tcfa_refcnt++; + *a = p; return true; } return false; @@ -239,11 +237,10 @@ EXPORT_SYMBOL(tcf_hash_check); void tcf_hash_cleanup(struct tc_action *a, struct nlattr *est) { - struct tcf_common *pc = (struct tcf_common *)a; if (est) - gen_kill_estimator(&pc->tcfc_bstats, - &pc->tcfc_rate_est); - call_rcu(&pc->tcfc_rcu, free_tcf); + gen_kill_estimator(&a->tcfa_bstats, + &a->tcfa_rate_est); + call_rcu(&a->tcfa_rcu, free_tcf); } EXPORT_SYMBOL(tcf_hash_cleanup); @@ -251,15 +248,15 @@ int tcf_hash_create(struct tc_action_net *tn, u32 index, struct nlattr *est, struct tc_action **a, const struct tc_action_ops *ops, int bind, bool cpustats) { - struct tcf_common *p = kzalloc(ops->size, GFP_KERNEL); + struct tc_action *p = kzalloc(ops->size, GFP_KERNEL); struct tcf_hashinfo *hinfo = tn->hinfo; int err = -ENOMEM; if (unlikely(!p)) return -ENOMEM; - p->tcfc_refcnt = 1; + p->tcfa_refcnt = 1; if (bind) - p->tcfc_bindcnt = 1; + p->tcfa_bindcnt = 1; if (cpustats) { p->cpu_bstats = netdev_alloc_pcpu_stats(struct gnet_stats_basic_cpu); @@ -275,38 +272,37 @@ err2: goto err1; } } - spin_lock_init(&p->tcfc_lock); - INIT_HLIST_NODE(&p->tcfc_head); - p->tcfc_index = index ? index : tcf_hash_new_index(tn); - p->tcfc_tm.install = jiffies; - p->tcfc_tm.lastuse = jiffies; - p->tcfc_tm.firstuse = 0; + spin_lock_init(&p->tcfa_lock); + INIT_HLIST_NODE(&p->tcfa_head); + p->tcfa_index = index ? index : tcf_hash_new_index(tn); + p->tcfa_tm.install = jiffies; + p->tcfa_tm.lastuse = jiffies; + p->tcfa_tm.firstuse = 0; if (est) { - err = gen_new_estimator(&p->tcfc_bstats, p->cpu_bstats, - &p->tcfc_rate_est, - &p->tcfc_lock, NULL, est); + err = gen_new_estimator(&p->tcfa_bstats, p->cpu_bstats, + &p->tcfa_rate_est, + &p->tcfa_lock, NULL, est); if (err) { free_percpu(p->cpu_qstats); goto err2; } } - p->tcfc_act.hinfo = hinfo; - p->tcfc_act.ops = ops; - INIT_LIST_HEAD(&p->tcfc_act.list); - *a = &p->tcfc_act; + p->hinfo = hinfo; + p->ops = ops; + INIT_LIST_HEAD(&p->list); + *a = p; return 0; } EXPORT_SYMBOL(tcf_hash_create); void tcf_hash_insert(struct tc_action_net *tn, struct tc_action *a) { - struct tcf_common *p = (struct tcf_common *)a; struct tcf_hashinfo *hinfo = tn->hinfo; - unsigned int h = tcf_hash(p->tcfc_index, hinfo->hmask); + unsigned int h = tcf_hash(a->tcfa_index, hinfo->hmask); spin_lock_bh(&hinfo->lock); - hlist_add_head(&p->tcfc_head, &hinfo->htab[h]); + hlist_add_head(&a->tcfa_head, &hinfo->htab[h]); spin_unlock_bh(&hinfo->lock); } EXPORT_SYMBOL(tcf_hash_insert); @@ -317,13 +313,13 @@ void tcf_hashinfo_destroy(const struct tc_action_ops *ops, int i; for (i = 0; i < hinfo->hmask + 1; i++) { - struct tcf_common *p; + struct tc_action *p; struct hlist_node *n; - hlist_for_each_entry_safe(p, n, &hinfo->htab[i], tcfc_head) { + hlist_for_each_entry_safe(p, n, &hinfo->htab[i], tcfa_head) { int ret; - ret = __tcf_hash_release((struct tc_action *)p, false, true); + ret = __tcf_hash_release(p, false, true); if (ret == ACT_P_DELETED) module_put(ops->owner); else if (ret < 0) @@ -625,12 +621,11 @@ err: return err; } -int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a, +int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *p, int compat_mode) { int err = 0; struct gnet_dump d; - struct tcf_common *p = (struct tcf_common *)a; if (p == NULL) goto errout; @@ -639,27 +634,27 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a, * to add additional backward compatibility statistic TLVs. */ if (compat_mode) { - if (a->type == TCA_OLD_COMPAT) + if (p->type == TCA_OLD_COMPAT) err = gnet_stats_start_copy_compat(skb, 0, TCA_STATS, TCA_XSTATS, - &p->tcfc_lock, &d, + &p->tcfa_lock, &d, TCA_PAD); else return 0; } else err = gnet_stats_start_copy(skb, TCA_ACT_STATS, - &p->tcfc_lock, &d, TCA_ACT_PAD); + &p->tcfa_lock, &d, TCA_ACT_PAD); if (err < 0) goto errout; - if (gnet_stats_copy_basic(NULL, &d, p->cpu_bstats, &p->tcfc_bstats) < 0 || - gnet_stats_copy_rate_est(&d, &p->tcfc_bstats, - &p->tcfc_rate_est) < 0 || + if (gnet_stats_copy_basic(NULL, &d, p->cpu_bstats, &p->tcfa_bstats) < 0 || + gnet_stats_copy_rate_est(&d, &p->tcfa_bstats, + &p->tcfa_rate_est) < 0 || gnet_stats_copy_queue(&d, p->cpu_qstats, - &p->tcfc_qstats, - p->tcfc_qstats.qlen) < 0) + &p->tcfa_qstats, + p->tcfa_qstats.qlen) < 0) goto errout; if (gnet_stats_finish_copy(&d) < 0) diff --git a/net/sched/act_police.c b/net/sched/act_police.c index 123794a..b3c7e97 100644 --- a/net/sched/act_police.c +++ b/net/sched/act_police.c @@ -23,7 +23,7 @@ #include struct tcf_police { - struct tcf_common common; + struct tc_action common; int tcfp_result; u32 tcfp_ewma_rate; s64 tcfp_burst; @@ -73,11 +73,11 @@ static int tcf_act_police_walker(struct net *net, struct sk_buff *skb, for (i = 0; i < (POL_TAB_MASK + 1); i++) { struct hlist_head *head; - struct tcf_common *p; + struct tc_action *p; head = &hinfo->htab[tcf_hash(i, POL_TAB_MASK)]; - hlist_for_each_entry_rcu(p, head, tcfc_head) { + hlist_for_each_entry_rcu(p, head, tcfa_head) { index++; if (index < s_i) continue; @@ -85,9 +85,9 @@ static int tcf_act_police_walker(struct net *net, struct sk_buff *skb, if (nest == NULL) goto nla_put_failure; if (type == RTM_DELACTION) - err = tcf_action_dump_1(skb, (struct tc_action *)p, 0, 1); + err = tcf_action_dump_1(skb, p, 0, 1); else - err = tcf_action_dump_1(skb, (struct tc_action *)p, 0, 0); + err = tcf_action_dump_1(skb, p, 0, 0); if (err < 0) { index--; nla_nest_cancel(skb, nest); -- cgit v0.10.2 From 9a8c5ddedd9805cf52744ef6bdf591326684f88c Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:36 -0700 Subject: drivers: net: xgene: Separate set_speed from mac_init Since mac_init is too heavy to be called when the link changes, moved the speed_set configuration to a new function and added mac_ops->set_speed function pointer. This function will be called from adjust_link callback. Added cases for 10/100 support for SGMII based 1G interface. Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index 2f5638f..725109b 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -512,14 +512,11 @@ static void xgene_enet_configure_clock(struct xgene_enet_pdata *pdata) #endif } -static void xgene_gmac_init(struct xgene_enet_pdata *pdata) +static void xgene_gmac_set_speed(struct xgene_enet_pdata *pdata) { struct device *dev = &pdata->pdev->dev; - u32 value, mc2; - u32 intf_ctl, rgmii; - u32 icm0, icm2; - - xgene_gmac_reset(pdata); + u32 icm0, icm2, mc2; + u32 intf_ctl, rgmii, value; xgene_enet_rd_mcx_csr(pdata, ICM_CONFIG0_REG_0_ADDR, &icm0); xgene_enet_rd_mcx_csr(pdata, ICM_CONFIG2_REG_0_ADDR, &icm2); @@ -564,7 +561,19 @@ static void xgene_gmac_init(struct xgene_enet_pdata *pdata) mc2 |= FULL_DUPLEX2 | PAD_CRC; xgene_enet_wr_mcx_mac(pdata, MAC_CONFIG_2_ADDR, mc2); xgene_enet_wr_mcx_mac(pdata, INTERFACE_CONTROL_ADDR, intf_ctl); + xgene_enet_wr_csr(pdata, RGMII_REG_0_ADDR, rgmii); + xgene_enet_configure_clock(pdata); + + xgene_enet_wr_mcx_csr(pdata, ICM_CONFIG0_REG_0_ADDR, icm0); + xgene_enet_wr_mcx_csr(pdata, ICM_CONFIG2_REG_0_ADDR, icm2); +} +static void xgene_gmac_init(struct xgene_enet_pdata *pdata) +{ + u32 value; + + xgene_gmac_reset(pdata); + xgene_gmac_set_speed(pdata); xgene_gmac_set_mac_addr(pdata); /* Adjust MDC clock frequency */ @@ -579,15 +588,10 @@ static void xgene_gmac_init(struct xgene_enet_pdata *pdata) /* Rtype should be copied from FP */ xgene_enet_wr_csr(pdata, RSIF_RAM_DBG_REG0_ADDR, 0); - xgene_enet_wr_csr(pdata, RGMII_REG_0_ADDR, rgmii); - xgene_enet_configure_clock(pdata); /* Rx-Tx traffic resume */ xgene_enet_wr_csr(pdata, CFG_LINK_AGGR_RESUME_0_ADDR, TX_PORT0); - xgene_enet_wr_mcx_csr(pdata, ICM_CONFIG0_REG_0_ADDR, icm0); - xgene_enet_wr_mcx_csr(pdata, ICM_CONFIG2_REG_0_ADDR, icm2); - xgene_enet_rd_mcx_csr(pdata, RX_DV_GATE_REG_0_ADDR, &value); value &= ~TX_DV_GATE_EN0; value &= ~RX_DV_GATE_EN0; @@ -724,12 +728,13 @@ static int xgene_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, static void xgene_enet_adjust_link(struct net_device *ndev) { struct xgene_enet_pdata *pdata = netdev_priv(ndev); + const struct xgene_mac_ops *mac_ops = pdata->mac_ops; struct phy_device *phydev = pdata->phy_dev; if (phydev->link) { if (pdata->phy_speed != phydev->speed) { pdata->phy_speed = phydev->speed; - xgene_gmac_init(pdata); + mac_ops->set_speed(pdata); xgene_gmac_rx_enable(pdata); xgene_gmac_tx_enable(pdata); phy_print_status(phydev); @@ -890,6 +895,7 @@ const struct xgene_mac_ops xgene_gmac_ops = { .tx_enable = xgene_gmac_tx_enable, .rx_disable = xgene_gmac_rx_disable, .tx_disable = xgene_gmac_tx_disable, + .set_speed = xgene_gmac_set_speed, .set_mac_addr = xgene_gmac_set_mac_addr, }; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h index 45220be..e840f96 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h @@ -104,6 +104,8 @@ enum xgene_enet_rm { #define RECOMBBUF BIT(27) #define MAC_OFFSET 0x30 +#define OFFSET_4 0x04 +#define OFFSET_8 0x08 #define BLOCK_ETH_CSR_OFFSET 0x2000 #define BLOCK_ETH_CLE_CSR_OFFSET 0x6000 diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index d208b17..8da3860 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -730,8 +730,10 @@ static int xgene_enet_open(struct net_device *ndev) if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) phy_start(pdata->phy_dev); - else + else { schedule_delayed_work(&pdata->link_work, PHY_POLL_LINK_OFF); + netif_carrier_off(ndev); + } netif_start_queue(ndev); @@ -761,7 +763,6 @@ static int xgene_enet_close(struct net_device *ndev) return 0; } - static void xgene_enet_delete_ring(struct xgene_enet_desc_ring *ring) { struct xgene_enet_pdata *pdata; @@ -1447,6 +1448,7 @@ static int xgene_enet_init_hw(struct xgene_enet_pdata *pdata) pdata->port_ops->cle_bypass(pdata, dst_ring_num, buf_pool->id); } + pdata->phy_speed = SPEED_UNKNOWN; pdata->mac_ops->init(pdata); return ret; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h index 092fbec..aed9f43 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h @@ -140,6 +140,7 @@ struct xgene_mac_ops { void (*rx_enable)(struct xgene_enet_pdata *pdata); void (*tx_disable)(struct xgene_enet_pdata *pdata); void (*rx_disable)(struct xgene_enet_pdata *pdata); + void (*set_speed)(struct xgene_enet_pdata *pdata); void (*set_mac_addr)(struct xgene_enet_pdata *pdata); void (*set_mss)(struct xgene_enet_pdata *pdata); void (*link_state)(struct work_struct *work); diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c index 7847551..a3063fd 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c @@ -93,6 +93,11 @@ static u32 xgene_enet_rd_diag_csr(struct xgene_enet_pdata *p, u32 offset) return ioread32(p->eth_diag_csr_addr + offset); } +static u32 xgene_enet_rd_mcx_csr(struct xgene_enet_pdata *p, u32 offset) +{ + return ioread32(p->mcx_mac_csr_addr + offset); +} + static u32 xgene_enet_rd_indirect(struct xgene_indirect_ctl *ctl, u32 rd_addr) { u32 rd_data; @@ -230,21 +235,105 @@ static u32 xgene_enet_link_status(struct xgene_enet_pdata *p) data = xgene_mii_phy_read(p, INT_PHY_ADDR, SGMII_BASE_PAGE_ABILITY_ADDR >> 2); + if (LINK_SPEED(data) == PHY_SPEED_1000) + p->phy_speed = SPEED_1000; + else if (LINK_SPEED(data) == PHY_SPEED_100) + p->phy_speed = SPEED_100; + else + p->phy_speed = SPEED_10; + return data & LINK_UP; } -static void xgene_sgmac_init(struct xgene_enet_pdata *p) +static void xgene_sgmii_configure(struct xgene_enet_pdata *p) { - u32 data, loop = 10; - u32 offset = p->port_id * 4; - u32 enet_spare_cfg_reg, rsif_config_reg; - u32 cfg_bypass_reg, rx_dv_gate_reg; - - xgene_sgmac_reset(p); + xgene_mii_phy_write(p, INT_PHY_ADDR, SGMII_TBI_CONTROL_ADDR >> 2, + 0x8000); + xgene_mii_phy_write(p, INT_PHY_ADDR, SGMII_CONTROL_ADDR >> 2, 0x9000); + xgene_mii_phy_write(p, INT_PHY_ADDR, SGMII_TBI_CONTROL_ADDR >> 2, 0); +} - /* Enable auto-negotiation */ - xgene_mii_phy_write(p, INT_PHY_ADDR, SGMII_CONTROL_ADDR >> 2, 0x1000); +static void xgene_sgmii_tbi_control_reset(struct xgene_enet_pdata *p) +{ + xgene_mii_phy_write(p, INT_PHY_ADDR, SGMII_TBI_CONTROL_ADDR >> 2, + 0x8000); xgene_mii_phy_write(p, INT_PHY_ADDR, SGMII_TBI_CONTROL_ADDR >> 2, 0); +} + +static void xgene_sgmii_reset(struct xgene_enet_pdata *p) +{ + u32 value; + + if (p->phy_speed == SPEED_UNKNOWN) + return; + + value = xgene_mii_phy_read(p, INT_PHY_ADDR, + SGMII_BASE_PAGE_ABILITY_ADDR >> 2); + if (!(value & LINK_UP)) + xgene_sgmii_tbi_control_reset(p); +} + +static void xgene_sgmac_set_speed(struct xgene_enet_pdata *p) +{ + u32 icm0_addr, icm2_addr, debug_addr; + u32 icm0, icm2, intf_ctl; + u32 mc2, value; + + xgene_sgmii_reset(p); + + if (p->enet_id == XGENE_ENET1) { + icm0_addr = ICM_CONFIG0_REG_0_ADDR + p->port_id * OFFSET_8; + icm2_addr = ICM_CONFIG2_REG_0_ADDR + p->port_id * OFFSET_4; + debug_addr = DEBUG_REG_ADDR; + } else { + icm0_addr = XG_MCX_ICM_CONFIG0_REG_0_ADDR; + icm2_addr = XG_MCX_ICM_CONFIG2_REG_0_ADDR; + debug_addr = XG_DEBUG_REG_ADDR; + } + + icm0 = xgene_enet_rd_mcx_csr(p, icm0_addr); + icm2 = xgene_enet_rd_mcx_csr(p, icm2_addr); + mc2 = xgene_enet_rd_mac(p, MAC_CONFIG_2_ADDR); + intf_ctl = xgene_enet_rd_mac(p, INTERFACE_CONTROL_ADDR); + + switch (p->phy_speed) { + case SPEED_10: + ENET_INTERFACE_MODE2_SET(&mc2, 1); + intf_ctl &= ~(ENET_LHD_MODE | ENET_GHD_MODE); + CFG_MACMODE_SET(&icm0, 0); + CFG_WAITASYNCRD_SET(&icm2, 500); + break; + case SPEED_100: + ENET_INTERFACE_MODE2_SET(&mc2, 1); + intf_ctl &= ~ENET_GHD_MODE; + intf_ctl |= ENET_LHD_MODE; + CFG_MACMODE_SET(&icm0, 1); + CFG_WAITASYNCRD_SET(&icm2, 80); + break; + default: + ENET_INTERFACE_MODE2_SET(&mc2, 2); + intf_ctl &= ~ENET_LHD_MODE; + intf_ctl |= ENET_GHD_MODE; + CFG_MACMODE_SET(&icm0, 2); + CFG_WAITASYNCRD_SET(&icm2, 16); + value = xgene_enet_rd_csr(p, debug_addr); + value |= CFG_BYPASS_UNISEC_TX | CFG_BYPASS_UNISEC_RX; + xgene_enet_wr_csr(p, debug_addr, value); + break; + } + + mc2 |= FULL_DUPLEX2 | PAD_CRC; + xgene_enet_wr_mac(p, MAC_CONFIG_2_ADDR, mc2); + xgene_enet_wr_mac(p, INTERFACE_CONTROL_ADDR, intf_ctl); + xgene_enet_wr_mcx_csr(p, icm0_addr, icm0); + xgene_enet_wr_mcx_csr(p, icm2_addr, icm2); +} + +static void xgene_sgmii_enable_autoneg(struct xgene_enet_pdata *p) +{ + u32 data, loop = 10; + + xgene_sgmii_configure(p); while (loop--) { data = xgene_mii_phy_read(p, INT_PHY_ADDR, @@ -255,17 +344,25 @@ static void xgene_sgmac_init(struct xgene_enet_pdata *p) } if (!(data & AUTO_NEG_COMPLETE) || !(data & LINK_STATUS)) netdev_err(p->ndev, "Auto-negotiation failed\n"); +} - data = xgene_enet_rd_mac(p, MAC_CONFIG_2_ADDR); - ENET_INTERFACE_MODE2_SET(&data, 2); - xgene_enet_wr_mac(p, MAC_CONFIG_2_ADDR, data | FULL_DUPLEX2); - xgene_enet_wr_mac(p, INTERFACE_CONTROL_ADDR, ENET_GHD_MODE); +static void xgene_sgmac_init(struct xgene_enet_pdata *p) +{ + u32 enet_spare_cfg_reg, rsif_config_reg; + u32 cfg_bypass_reg, rx_dv_gate_reg; + u32 data, offset; + + xgene_sgmac_reset(p); + xgene_sgmii_enable_autoneg(p); + xgene_sgmac_set_speed(p); + xgene_sgmac_set_mac_addr(p); if (p->enet_id == XGENE_ENET1) { enet_spare_cfg_reg = ENET_SPARE_CFG_REG_ADDR; rsif_config_reg = RSIF_CONFIG_REG_ADDR; cfg_bypass_reg = CFG_BYPASS_ADDR; - rx_dv_gate_reg = SG_RX_DV_GATE_REG_0_ADDR; + offset = p->port_id * OFFSET_4; + rx_dv_gate_reg = SG_RX_DV_GATE_REG_0_ADDR + offset; } else { enet_spare_cfg_reg = XG_ENET_SPARE_CFG_REG_ADDR; rsif_config_reg = XG_RSIF_CONFIG_REG_ADDR; @@ -277,8 +374,6 @@ static void xgene_sgmac_init(struct xgene_enet_pdata *p) data |= MPA_IDLE_WITH_QMI_EMPTY; xgene_enet_wr_csr(p, enet_spare_cfg_reg, data); - xgene_sgmac_set_mac_addr(p); - /* Adjust MDC clock frequency */ data = xgene_enet_rd_mac(p, MII_MGMT_CONFIG_ADDR); MGMT_CLOCK_SEL_SET(&data, 7); @@ -292,7 +387,7 @@ static void xgene_sgmac_init(struct xgene_enet_pdata *p) /* Bypass traffic gating */ xgene_enet_wr_csr(p, XG_ENET_SPARE_CFG_REG_1_ADDR, 0x84); xgene_enet_wr_csr(p, cfg_bypass_reg, RESUME_TX); - xgene_enet_wr_mcx_csr(p, rx_dv_gate_reg + offset, RESUME_RX0); + xgene_enet_wr_mcx_csr(p, rx_dv_gate_reg, RESUME_RX0); } static void xgene_sgmac_rxtx(struct xgene_enet_pdata *p, u32 bits, bool set) @@ -386,10 +481,11 @@ static void xgene_enet_link_state(struct work_struct *work) if (link) { if (!netif_carrier_ok(ndev)) { netif_carrier_on(ndev); - xgene_sgmac_init(p); + xgene_sgmac_set_speed(p); xgene_sgmac_rx_enable(p); xgene_sgmac_tx_enable(p); - netdev_info(ndev, "Link is Up - 1Gbps\n"); + netdev_info(ndev, "Link is Up - %dMbps\n", + p->phy_speed); } poll_interval = PHY_POLL_LINK_ON; } else { @@ -412,6 +508,7 @@ const struct xgene_mac_ops xgene_sgmac_ops = { .tx_enable = xgene_sgmac_tx_enable, .rx_disable = xgene_sgmac_rx_disable, .tx_disable = xgene_sgmac_tx_disable, + .set_speed = xgene_sgmac_set_speed, .set_mac_addr = xgene_sgmac_set_mac_addr, .link_state = xgene_enet_link_state }; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h index 002df5a..32f140a 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h @@ -24,6 +24,7 @@ #define PHY_ADDR(src) (((src)<<8) & GENMASK(12, 8)) #define REG_ADDR(src) ((src) & GENMASK(4, 0)) #define PHY_CONTROL(src) ((src) & GENMASK(15, 0)) +#define LINK_SPEED(src) (((src) & GENMASK(11, 10)) >> 10) #define INT_PHY_ADDR 0x1e #define SGMII_TBI_CONTROL_ADDR 0x44 #define SGMII_CONTROL_ADDR 0x00 @@ -35,6 +36,12 @@ #define MPA_IDLE_WITH_QMI_EMPTY BIT(12) #define SG_RX_DV_GATE_REG_0_ADDR 0x05fc +enum xgene_phy_speed { + PHY_SPEED_10, + PHY_SPEED_100, + PHY_SPEED_1000 +}; + extern const struct xgene_mac_ops xgene_sgmac_ops; extern const struct xgene_port_ops xgene_sgport_ops; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h index 0a2dca8..f1ea485 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.h @@ -65,9 +65,12 @@ #define XG_CFG_LINK_AGGR_RESUME_0_ADDR 0x0214 #define XG_LINK_STATUS_ADDR 0x0228 #define XG_TSIF_MSS_REG0_ADDR 0x02a4 +#define XG_DEBUG_REG_ADDR 0x0400 #define XG_ENET_SPARE_CFG_REG_ADDR 0x040c #define XG_ENET_SPARE_CFG_REG_1_ADDR 0x0410 #define XGENET_RX_DV_GATE_REG_0_ADDR 0x0804 +#define XG_MCX_ICM_CONFIG0_REG_0_ADDR 0x00e0 +#define XG_MCX_ICM_CONFIG2_REG_0_ADDR 0x00e8 extern const struct xgene_mac_ops xgene_xgmac_ops; extern const struct xgene_port_ops xgene_xgport_ops; -- cgit v0.10.2 From cb11c062f9052c6bde6a5fa18cab1f41d81131b3 Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:37 -0700 Subject: drivers: net: xgene: Fix module unload crash - hw resource cleanup When the driver is configured as kernel module and when it gets unloaded and reloaded, kernel crash was observed. This patch address the hardware resource cleanups by doing the following, - Added mac_ops->clear() to do prefetch buffer clean up - Fixed delete freepool buffers logic - Reordered mac_enable and mac_disable - Added Tx completion ring free - Moved down delete_desc_rings after ring cleanup Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index 725109b..009fb8e 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -697,8 +697,48 @@ static int xgene_enet_reset(struct xgene_enet_pdata *pdata) return 0; } +static void xgene_enet_clear(struct xgene_enet_pdata *pdata, + struct xgene_enet_desc_ring *ring) +{ + u32 addr, val, data; + + val = xgene_enet_ring_bufnum(ring->id); + + if (xgene_enet_is_bufpool(ring->id)) { + addr = ENET_CFGSSQMIFPRESET_ADDR; + data = BIT(val - 0x20); + } else { + addr = ENET_CFGSSQMIWQRESET_ADDR; + data = BIT(val); + } + + xgene_enet_wr_ring_if(pdata, addr, data); +} + static void xgene_gport_shutdown(struct xgene_enet_pdata *pdata) { + struct xgene_enet_desc_ring *ring; + u32 pb, val; + int i; + + pb = 0; + for (i = 0; i < pdata->rxq_cnt; i++) { + ring = pdata->rx_ring[i]->buf_pool; + + val = xgene_enet_ring_bufnum(ring->id); + pb |= BIT(val - 0x20); + } + xgene_enet_wr_ring_if(pdata, ENET_CFGSSQMIFPRESET_ADDR, pb); + + pb = 0; + for (i = 0; i < pdata->txq_cnt; i++) { + ring = pdata->tx_ring[i]; + + val = xgene_enet_ring_bufnum(ring->id); + pb |= BIT(val); + } + xgene_enet_wr_ring_if(pdata, ENET_CFGSSQMIWQRESET_ADDR, pb); + if (!IS_ERR(pdata->clk)) clk_disable_unprepare(pdata->clk); } @@ -901,6 +941,7 @@ const struct xgene_mac_ops xgene_gmac_ops = { const struct xgene_port_ops xgene_gport_ops = { .reset = xgene_enet_reset, + .clear = xgene_enet_clear, .cle_bypass = xgene_enet_cle_bypass, .shutdown = xgene_gport_shutdown, }; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h index e840f96..eec55c1 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h @@ -167,6 +167,8 @@ enum xgene_enet_rm { #define TX_DV_GATE_EN0 BIT(2) #define RX_DV_GATE_EN0 BIT(1) #define RESUME_RX0 BIT(0) +#define ENET_CFGSSQMIFPRESET_ADDR 0x14 +#define ENET_CFGSSQMIWQRESET_ADDR 0x1c #define ENET_CFGSSQMIWQASSOC_ADDR 0xe0 #define ENET_CFGSSQMIFPQASSOC_ADDR 0xdc #define ENET_CFGSSQMIQMLITEFPQASSOC_ADDR 0xf0 diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index 8da3860..f79950a 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -102,25 +102,13 @@ static u8 xgene_enet_hdr_len(const void *data) static void xgene_enet_delete_bufpool(struct xgene_enet_desc_ring *buf_pool) { - struct xgene_enet_pdata *pdata = netdev_priv(buf_pool->ndev); - struct xgene_enet_raw_desc16 *raw_desc; - u32 slots = buf_pool->slots - 1; - u32 tail = buf_pool->tail; - u32 userinfo; - int i, len; - - len = pdata->ring_ops->len(buf_pool); - for (i = 0; i < len; i++) { - tail = (tail - 1) & slots; - raw_desc = &buf_pool->raw_desc16[tail]; + int i; - /* Hardware stores descriptor in little endian format */ - userinfo = GET_VAL(USERINFO, le64_to_cpu(raw_desc->m0)); - dev_kfree_skb_any(buf_pool->rx_skb[userinfo]); + /* Free up the buffers held by hardware */ + for (i = 0; i < buf_pool->slots; i++) { + if (buf_pool->rx_skb[i]) + dev_kfree_skb_any(buf_pool->rx_skb[i]); } - - pdata->ring_ops->wr_cmd(buf_pool, -len); - buf_pool->tail = tail; } static irqreturn_t xgene_enet_rx_irq(const int irq, void *data) @@ -481,6 +469,7 @@ static int xgene_enet_rx_frame(struct xgene_enet_desc_ring *rx_ring, XGENE_ENET_MAX_MTU, DMA_FROM_DEVICE); skb_index = GET_VAL(USERINFO, le64_to_cpu(raw_desc->m0)); skb = buf_pool->rx_skb[skb_index]; + buf_pool->rx_skb[skb_index] = NULL; /* checking for error */ status = (GET_VAL(ELERR, le64_to_cpu(raw_desc->m0)) << LERR_LEN) || @@ -720,9 +709,6 @@ static int xgene_enet_open(struct net_device *ndev) if (ret) return ret; - mac_ops->tx_enable(pdata); - mac_ops->rx_enable(pdata); - xgene_enet_napi_enable(pdata); ret = xgene_enet_register_irq(ndev); if (ret) @@ -735,6 +721,8 @@ static int xgene_enet_open(struct net_device *ndev) netif_carrier_off(ndev); } + mac_ops->tx_enable(pdata); + mac_ops->rx_enable(pdata); netif_start_queue(ndev); return ret; @@ -747,15 +735,14 @@ static int xgene_enet_close(struct net_device *ndev) int i; netif_stop_queue(ndev); + mac_ops->tx_disable(pdata); + mac_ops->rx_disable(pdata); if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) phy_stop(pdata->phy_dev); else cancel_delayed_work_sync(&pdata->link_work); - mac_ops->tx_disable(pdata); - mac_ops->rx_disable(pdata); - xgene_enet_free_irq(ndev); xgene_enet_napi_disable(pdata); for (i = 0; i < pdata->rxq_cnt; i++) @@ -785,6 +772,9 @@ static void xgene_enet_delete_desc_rings(struct xgene_enet_pdata *pdata) ring = pdata->tx_ring[i]; if (ring) { xgene_enet_delete_ring(ring); + pdata->port_ops->clear(pdata, ring); + if (pdata->cq_cnt) + xgene_enet_delete_ring(ring->cp_ring); pdata->tx_ring[i] = NULL; } } @@ -795,6 +785,7 @@ static void xgene_enet_delete_desc_rings(struct xgene_enet_pdata *pdata) buf_pool = ring->buf_pool; xgene_enet_delete_bufpool(buf_pool); xgene_enet_delete_ring(buf_pool); + pdata->port_ops->clear(pdata, buf_pool); xgene_enet_delete_ring(ring); pdata->rx_ring[i] = NULL; } @@ -1682,8 +1673,8 @@ static int xgene_enet_remove(struct platform_device *pdev) if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) xgene_enet_mdio_remove(pdata); unregister_netdev(ndev); - xgene_enet_delete_desc_rings(pdata); pdata->port_ops->shutdown(pdata); + xgene_enet_delete_desc_rings(pdata); free_netdev(ndev); return 0; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h index aed9f43..681a473 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h @@ -148,6 +148,8 @@ struct xgene_mac_ops { struct xgene_port_ops { int (*reset)(struct xgene_enet_pdata *pdata); + void (*clear)(struct xgene_enet_pdata *pdata, + struct xgene_enet_desc_ring *ring); void (*cle_bypass)(struct xgene_enet_pdata *pdata, u32 dst_ring_num, u16 bufpool_id); void (*shutdown)(struct xgene_enet_pdata *pdata); diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c index a3063fd..f1477d2 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c @@ -137,9 +137,17 @@ static u32 xgene_enet_rd_mac(struct xgene_enet_pdata *p, u32 rd_addr) static int xgene_enet_ecc_init(struct xgene_enet_pdata *p) { struct net_device *ndev = p->ndev; - u32 data; + u32 data, shutdown; int i = 0; + shutdown = xgene_enet_rd_diag_csr(p, ENET_CFG_MEM_RAM_SHUTDOWN_ADDR); + data = xgene_enet_rd_diag_csr(p, ENET_BLOCK_MEM_RDY_ADDR); + + if (!shutdown && data == ~0U) { + netdev_dbg(ndev, "+ ecc_init done, skipping\n"); + return 0; + } + xgene_enet_wr_diag_csr(p, ENET_CFG_MEM_RAM_SHUTDOWN_ADDR, 0); do { usleep_range(100, 110); @@ -464,10 +472,47 @@ static void xgene_enet_cle_bypass(struct xgene_enet_pdata *p, xgene_enet_wr_csr(p, cle_bypass_reg1 + offset, data); } +static void xgene_enet_clear(struct xgene_enet_pdata *pdata, + struct xgene_enet_desc_ring *ring) +{ + u32 addr, val, data; + + val = xgene_enet_ring_bufnum(ring->id); + + if (xgene_enet_is_bufpool(ring->id)) { + addr = ENET_CFGSSQMIFPRESET_ADDR; + data = BIT(val - 0x20); + } else { + addr = ENET_CFGSSQMIWQRESET_ADDR; + data = BIT(val); + } + + xgene_enet_wr_ring_if(pdata, addr, data); +} + static void xgene_enet_shutdown(struct xgene_enet_pdata *p) { - if (!IS_ERR(p->clk)) - clk_disable_unprepare(p->clk); + struct xgene_enet_desc_ring *ring; + u32 pb, val; + int i; + + pb = 0; + for (i = 0; i < p->rxq_cnt; i++) { + ring = p->rx_ring[i]->buf_pool; + + val = xgene_enet_ring_bufnum(ring->id); + pb |= BIT(val - 0x20); + } + xgene_enet_wr_ring_if(p, ENET_CFGSSQMIFPRESET_ADDR, pb); + + pb = 0; + for (i = 0; i < p->txq_cnt; i++) { + ring = p->tx_ring[i]; + + val = xgene_enet_ring_bufnum(ring->id); + pb |= BIT(val); + } + xgene_enet_wr_ring_if(p, ENET_CFGSSQMIWQRESET_ADDR, pb); } static void xgene_enet_link_state(struct work_struct *work) @@ -515,6 +560,7 @@ const struct xgene_mac_ops xgene_sgmac_ops = { const struct xgene_port_ops xgene_sgport_ops = { .reset = xgene_enet_reset, + .clear = xgene_enet_clear, .cle_bypass = xgene_enet_cle_bypass, .shutdown = xgene_enet_shutdown }; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c index ba030dc..d0b4419 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c @@ -292,8 +292,45 @@ static void xgene_enet_xgcle_bypass(struct xgene_enet_pdata *pdata, static void xgene_enet_shutdown(struct xgene_enet_pdata *pdata) { - if (!IS_ERR(pdata->clk)) - clk_disable_unprepare(pdata->clk); + struct xgene_enet_desc_ring *ring; + u32 pb, val; + int i; + + pb = 0; + for (i = 0; i < pdata->rxq_cnt; i++) { + ring = pdata->rx_ring[i]->buf_pool; + + val = xgene_enet_ring_bufnum(ring->id); + pb |= BIT(val - 0x20); + } + xgene_enet_wr_ring_if(pdata, ENET_CFGSSQMIFPRESET_ADDR, pb); + + pb = 0; + for (i = 0; i < pdata->txq_cnt; i++) { + ring = pdata->tx_ring[i]; + + val = xgene_enet_ring_bufnum(ring->id); + pb |= BIT(val); + } + xgene_enet_wr_ring_if(pdata, ENET_CFGSSQMIWQRESET_ADDR, pb); +} + +static void xgene_enet_clear(struct xgene_enet_pdata *pdata, + struct xgene_enet_desc_ring *ring) +{ + u32 addr, val, data; + + val = xgene_enet_ring_bufnum(ring->id); + + if (xgene_enet_is_bufpool(ring->id)) { + addr = ENET_CFGSSQMIFPRESET_ADDR; + data = BIT(val - 0x20); + } else { + addr = ENET_CFGSSQMIWQRESET_ADDR; + data = BIT(val); + } + + xgene_enet_wr_ring_if(pdata, addr, data); } static void xgene_enet_link_state(struct work_struct *work) @@ -340,6 +377,7 @@ const struct xgene_mac_ops xgene_xgmac_ops = { const struct xgene_port_ops xgene_xgport_ops = { .reset = xgene_enet_reset, + .clear = xgene_enet_clear, .cle_bypass = xgene_enet_xgcle_bypass, .shutdown = xgene_enet_shutdown, }; -- cgit v0.10.2 From cb0366b7c16427a25923350b69f53a5b1345a34b Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:38 -0700 Subject: drivers: net: xgene: Fix module unload crash - change sw sequence When the driver is configured as kernel module and when it gets unloaded and reloaded, kernel crash was observed. This patch addresses the software cleanup by doing the following, - Moved register_netdev call after hardware is ready - Since ndev is not ready, added set_irq_name to set irq name - Since ndev is not ready, changed mdio_bus->parent to pdev->dev - Replaced netif_start(stop)_queue by netif_tx_start(stop)_queues - Removed napi_del call since it's called by free_netdev - Added dev_close call, within remove - Added shutdown callback - Changed to use dmam_ APIs Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index 009fb8e..4f98749 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -901,7 +901,7 @@ int xgene_enet_mdio_config(struct xgene_enet_pdata *pdata) ndev->name); mdio_bus->priv = pdata; - mdio_bus->parent = &ndev->dev; + mdio_bus->parent = &pdata->pdev->dev; ret = xgene_mdiobus_register(pdata, mdio_bus); if (ret) { diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index f79950a..87e5929 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -608,6 +608,30 @@ static void xgene_enet_timeout(struct net_device *ndev) } } +static void xgene_enet_set_irq_name(struct net_device *ndev) +{ + struct xgene_enet_pdata *pdata = netdev_priv(ndev); + struct xgene_enet_desc_ring *ring; + int i; + + for (i = 0; i < pdata->rxq_cnt; i++) { + ring = pdata->rx_ring[i]; + if (!pdata->cq_cnt) { + snprintf(ring->irq_name, IRQ_ID_SIZE, "%s-rx-txc", + ndev->name); + } else { + snprintf(ring->irq_name, IRQ_ID_SIZE, "%s-rx-%d", + ndev->name, i); + } + } + + for (i = 0; i < pdata->cq_cnt; i++) { + ring = pdata->tx_ring[i]->cp_ring; + snprintf(ring->irq_name, IRQ_ID_SIZE, "%s-txc-%d", + ndev->name, i); + } +} + static int xgene_enet_register_irq(struct net_device *ndev) { struct xgene_enet_pdata *pdata = netdev_priv(ndev); @@ -615,6 +639,7 @@ static int xgene_enet_register_irq(struct net_device *ndev) struct xgene_enet_desc_ring *ring; int ret = 0, i; + xgene_enet_set_irq_name(ndev); for (i = 0; i < pdata->rxq_cnt; i++) { ring = pdata->rx_ring[i]; irq_set_status_flags(ring->irq, IRQ_DISABLE_UNLAZY); @@ -723,7 +748,7 @@ static int xgene_enet_open(struct net_device *ndev) mac_ops->tx_enable(pdata); mac_ops->rx_enable(pdata); - netif_start_queue(ndev); + netif_tx_start_all_queues(ndev); return ret; } @@ -734,7 +759,7 @@ static int xgene_enet_close(struct net_device *ndev) const struct xgene_mac_ops *mac_ops = pdata->mac_ops; int i; - netif_stop_queue(ndev); + netif_tx_stop_all_queues(ndev); mac_ops->tx_disable(pdata); mac_ops->rx_disable(pdata); @@ -759,7 +784,7 @@ static void xgene_enet_delete_ring(struct xgene_enet_desc_ring *ring) dev = ndev_to_dev(ring->ndev); pdata->ring_ops->clear(ring); - dma_free_coherent(dev, ring->size, ring->desc_addr, ring->dma); + dmam_free_coherent(dev, ring->size, ring->desc_addr, ring->dma); } static void xgene_enet_delete_desc_rings(struct xgene_enet_pdata *pdata) @@ -834,7 +859,7 @@ static void xgene_enet_free_desc_ring(struct xgene_enet_desc_ring *ring) if (ring->desc_addr) { pdata->ring_ops->clear(ring); - dma_free_coherent(dev, ring->size, ring->desc_addr, ring->dma); + dmam_free_coherent(dev, ring->size, ring->desc_addr, ring->dma); } devm_kfree(dev, ring); } @@ -892,9 +917,10 @@ static struct xgene_enet_desc_ring *xgene_enet_create_desc_ring( struct net_device *ndev, u32 ring_num, enum xgene_enet_ring_cfgsize cfgsize, u32 ring_id) { - struct xgene_enet_desc_ring *ring; struct xgene_enet_pdata *pdata = netdev_priv(ndev); struct device *dev = ndev_to_dev(ndev); + struct xgene_enet_desc_ring *ring; + void *irq_mbox_addr; int size; size = xgene_enet_get_ring_size(dev, cfgsize); @@ -911,8 +937,8 @@ static struct xgene_enet_desc_ring *xgene_enet_create_desc_ring( ring->cfgsize = cfgsize; ring->id = ring_id; - ring->desc_addr = dma_zalloc_coherent(dev, size, &ring->dma, - GFP_KERNEL); + ring->desc_addr = dmam_alloc_coherent(dev, size, &ring->dma, + GFP_KERNEL | __GFP_ZERO); if (!ring->desc_addr) { devm_kfree(dev, ring); return NULL; @@ -920,14 +946,16 @@ static struct xgene_enet_desc_ring *xgene_enet_create_desc_ring( ring->size = size; if (is_irq_mbox_required(pdata, ring)) { - ring->irq_mbox_addr = dma_zalloc_coherent(dev, INTR_MBOX_SIZE, - &ring->irq_mbox_dma, GFP_KERNEL); - if (!ring->irq_mbox_addr) { - dma_free_coherent(dev, size, ring->desc_addr, - ring->dma); + irq_mbox_addr = dmam_alloc_coherent(dev, INTR_MBOX_SIZE, + &ring->irq_mbox_dma, + GFP_KERNEL | __GFP_ZERO); + if (!irq_mbox_addr) { + dmam_free_coherent(dev, size, ring->desc_addr, + ring->dma); devm_kfree(dev, ring); return NULL; } + ring->irq_mbox_addr = irq_mbox_addr; } ring->cmd_base = xgene_enet_ring_cmd_base(pdata, ring); @@ -988,6 +1016,7 @@ static int xgene_enet_create_desc_rings(struct net_device *ndev) u8 eth_bufnum = pdata->eth_bufnum; u8 bp_bufnum = pdata->bp_bufnum; u16 ring_num = pdata->ring_num; + __le64 *exp_bufs; u16 ring_id; int i, ret, size; @@ -1019,13 +1048,6 @@ static int xgene_enet_create_desc_rings(struct net_device *ndev) rx_ring->nbufpool = NUM_BUFPOOL; rx_ring->buf_pool = buf_pool; rx_ring->irq = pdata->irqs[i]; - if (!pdata->cq_cnt) { - snprintf(rx_ring->irq_name, IRQ_ID_SIZE, "%s-rx-txc", - ndev->name); - } else { - snprintf(rx_ring->irq_name, IRQ_ID_SIZE, "%s-rx%d", - ndev->name, i); - } buf_pool->rx_skb = devm_kcalloc(dev, buf_pool->slots, sizeof(struct sk_buff *), GFP_KERNEL); @@ -1052,13 +1074,13 @@ static int xgene_enet_create_desc_rings(struct net_device *ndev) } size = (tx_ring->slots / 2) * sizeof(__le64) * MAX_EXP_BUFFS; - tx_ring->exp_bufs = dma_zalloc_coherent(dev, size, - &dma_exp_bufs, - GFP_KERNEL); - if (!tx_ring->exp_bufs) { + exp_bufs = dmam_alloc_coherent(dev, size, &dma_exp_bufs, + GFP_KERNEL | __GFP_ZERO); + if (!exp_bufs) { ret = -ENOMEM; goto err; } + tx_ring->exp_bufs = exp_bufs; pdata->tx_ring[i] = tx_ring; @@ -1078,8 +1100,6 @@ static int xgene_enet_create_desc_rings(struct net_device *ndev) cp_ring->irq = pdata->irqs[pdata->rxq_cnt + i]; cp_ring->index = i; - snprintf(cp_ring->irq_name, IRQ_ID_SIZE, "%s-txc%d", - ndev->name, i); } cp_ring->cp_skb = devm_kcalloc(dev, tx_ring->slots, @@ -1549,22 +1569,6 @@ static void xgene_enet_napi_add(struct xgene_enet_pdata *pdata) } } -static void xgene_enet_napi_del(struct xgene_enet_pdata *pdata) -{ - struct napi_struct *napi; - int i; - - for (i = 0; i < pdata->rxq_cnt; i++) { - napi = &pdata->rx_ring[i]->napi; - netif_napi_del(napi); - } - - for (i = 0; i < pdata->cq_cnt; i++) { - napi = &pdata->tx_ring[i]->cp_ring->napi; - netif_napi_del(napi); - } -} - static int xgene_enet_probe(struct platform_device *pdev) { struct net_device *ndev; @@ -1628,12 +1632,6 @@ static int xgene_enet_probe(struct platform_device *pdev) goto err; } - ret = register_netdev(ndev); - if (ret) { - netdev_err(ndev, "Failed to register netdev\n"); - goto err; - } - ret = xgene_enet_init_hw(pdata); if (ret) goto err_netdev; @@ -1648,7 +1646,14 @@ static int xgene_enet_probe(struct platform_device *pdev) } xgene_enet_napi_add(pdata); + ret = register_netdev(ndev); + if (ret) { + netdev_err(ndev, "Failed to register netdev\n"); + goto err; + } + return 0; + err_netdev: unregister_netdev(ndev); err: @@ -1666,10 +1671,14 @@ static int xgene_enet_remove(struct platform_device *pdev) mac_ops = pdata->mac_ops; ndev = pdata->ndev; + rtnl_lock(); + if (netif_running(ndev)) + dev_close(ndev); + rtnl_unlock(); + mac_ops->rx_disable(pdata); mac_ops->tx_disable(pdata); - xgene_enet_napi_del(pdata); if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) xgene_enet_mdio_remove(pdata); unregister_netdev(ndev); @@ -1680,6 +1689,20 @@ static int xgene_enet_remove(struct platform_device *pdev) return 0; } +static void xgene_enet_shutdown(struct platform_device *pdev) +{ + struct xgene_enet_pdata *pdata; + + pdata = platform_get_drvdata(pdev); + if (!pdata) + return; + + if (!pdata->ndev) + return; + + xgene_enet_remove(pdev); +} + #ifdef CONFIG_ACPI static const struct acpi_device_id xgene_enet_acpi_match[] = { { "APMC0D05", XGENE_ENET1}, @@ -1714,6 +1737,7 @@ static struct platform_driver xgene_enet_driver = { }, .probe = xgene_enet_probe, .remove = xgene_enet_remove, + .shutdown = xgene_enet_shutdown, }; module_platform_driver(xgene_enet_driver); -- cgit v0.10.2 From bc61167ac816621c94f722177d2ae718c103005f Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:39 -0700 Subject: drivers: net: xgene: Fix module unload crash - clkrst sequence This patch fixes clock reset sequence. - Added clock reset sequence for ACPI - Added delay in clock reset sequence to make sure pulse is generated - Added clk_unprepare_disable() in port shutdown to make sure clock increment/decrement counts are matching - Removed MII_MGMT_CONFIG programming, since it is not required - Fixed programming XGENET_CONFIG_REG to enable SGMII mode Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index 4f98749..91a67a0 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -675,24 +675,33 @@ bool xgene_ring_mgr_init(struct xgene_enet_pdata *p) static int xgene_enet_reset(struct xgene_enet_pdata *pdata) { - u32 val; + struct device *dev = &pdata->pdev->dev; if (!xgene_ring_mgr_init(pdata)) return -ENODEV; - if (!IS_ERR(pdata->clk)) { + if (dev->of_node) { clk_prepare_enable(pdata->clk); + udelay(5); clk_disable_unprepare(pdata->clk); + udelay(5); clk_prepare_enable(pdata->clk); - xgene_enet_ecc_init(pdata); + udelay(5); + } else { +#ifdef CONFIG_ACPI + if (acpi_has_method(ACPI_HANDLE(&pdata->pdev->dev), "_RST")) { + acpi_evaluate_object(ACPI_HANDLE(&pdata->pdev->dev), + "_RST", NULL, NULL); + } else if (acpi_has_method(ACPI_HANDLE(&pdata->pdev->dev), + "_INI")) { + acpi_evaluate_object(ACPI_HANDLE(&pdata->pdev->dev), + "_INI", NULL, NULL); + } +#endif } - xgene_enet_config_ring_if_assoc(pdata); - /* Enable auto-incr for scanning */ - xgene_enet_rd_mcx_mac(pdata, MII_MGMT_CONFIG_ADDR, &val); - val |= SCAN_AUTO_INCR; - MGMT_CLOCK_SEL_SET(&val, 1); - xgene_enet_wr_mcx_mac(pdata, MII_MGMT_CONFIG_ADDR, val); + xgene_enet_ecc_init(pdata); + xgene_enet_config_ring_if_assoc(pdata); return 0; } @@ -717,6 +726,7 @@ static void xgene_enet_clear(struct xgene_enet_pdata *pdata, static void xgene_gport_shutdown(struct xgene_enet_pdata *pdata) { + struct device *dev = &pdata->pdev->dev; struct xgene_enet_desc_ring *ring; u32 pb, val; int i; @@ -739,8 +749,10 @@ static void xgene_gport_shutdown(struct xgene_enet_pdata *pdata) } xgene_enet_wr_ring_if(pdata, ENET_CFGSSQMIWQRESET_ADDR, pb); - if (!IS_ERR(pdata->clk)) - clk_disable_unprepare(pdata->clk); + if (dev->of_node) { + if (!IS_ERR(pdata->clk)) + clk_disable_unprepare(pdata->clk); + } } static int xgene_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c index f1477d2..03a70c5 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c @@ -28,6 +28,12 @@ static void xgene_enet_wr_csr(struct xgene_enet_pdata *p, u32 offset, u32 val) iowrite32(val, p->eth_csr_addr + offset); } +static void xgene_enet_wr_clkrst_csr(struct xgene_enet_pdata *p, u32 offset, + u32 val) +{ + iowrite32(val, p->base_addr + offset); +} + static void xgene_enet_wr_ring_if(struct xgene_enet_pdata *p, u32 offset, u32 val) { @@ -434,17 +440,38 @@ static void xgene_sgmac_tx_disable(struct xgene_enet_pdata *p) static int xgene_enet_reset(struct xgene_enet_pdata *p) { + struct device *dev = &p->pdev->dev; + if (!xgene_ring_mgr_init(p)) return -ENODEV; - if (!IS_ERR(p->clk)) { - clk_prepare_enable(p->clk); - clk_disable_unprepare(p->clk); - clk_prepare_enable(p->clk); + if (p->enet_id == XGENE_ENET2) + xgene_enet_wr_clkrst_csr(p, XGENET_CONFIG_REG_ADDR, SGMII_EN); + + if (dev->of_node) { + if (!IS_ERR(p->clk)) { + clk_prepare_enable(p->clk); + udelay(5); + clk_disable_unprepare(p->clk); + udelay(5); + clk_prepare_enable(p->clk); + udelay(5); + } + } else { +#ifdef CONFIG_ACPI + if (acpi_has_method(ACPI_HANDLE(&p->pdev->dev), "_RST")) + acpi_evaluate_object(ACPI_HANDLE(&p->pdev->dev), + "_RST", NULL, NULL); + else if (acpi_has_method(ACPI_HANDLE(&p->pdev->dev), "_INI")) + acpi_evaluate_object(ACPI_HANDLE(&p->pdev->dev), + "_INI", NULL, NULL); +#endif } - xgene_enet_ecc_init(p); - xgene_enet_config_ring_if_assoc(p); + if (!p->port_id) { + xgene_enet_ecc_init(p); + xgene_enet_config_ring_if_assoc(p); + } return 0; } @@ -492,6 +519,7 @@ static void xgene_enet_clear(struct xgene_enet_pdata *pdata, static void xgene_enet_shutdown(struct xgene_enet_pdata *p) { + struct device *dev = &p->pdev->dev; struct xgene_enet_desc_ring *ring; u32 pb, val; int i; @@ -513,6 +541,11 @@ static void xgene_enet_shutdown(struct xgene_enet_pdata *p) pb |= BIT(val); } xgene_enet_wr_ring_if(p, ENET_CFGSSQMIWQRESET_ADDR, pb); + + if (dev->of_node) { + if (!IS_ERR(p->clk)) + clk_disable_unprepare(p->clk); + } } static void xgene_enet_link_state(struct work_struct *work) diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h index 32f140a..3d0ba37 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.h @@ -35,6 +35,7 @@ #define LINK_UP BIT(15) #define MPA_IDLE_WITH_QMI_EMPTY BIT(12) #define SG_RX_DV_GATE_REG_0_ADDR 0x05fc +#define SGMII_EN 0x1 enum xgene_phy_speed { PHY_SPEED_10, diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c index d0b4419..9c6ad0d 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c @@ -258,13 +258,29 @@ static void xgene_xgmac_tx_disable(struct xgene_enet_pdata *pdata) static int xgene_enet_reset(struct xgene_enet_pdata *pdata) { + struct device *dev = &pdata->pdev->dev; + if (!xgene_ring_mgr_init(pdata)) return -ENODEV; - if (!IS_ERR(pdata->clk)) { + if (dev->of_node) { clk_prepare_enable(pdata->clk); + udelay(5); clk_disable_unprepare(pdata->clk); + udelay(5); clk_prepare_enable(pdata->clk); + udelay(5); + } else { +#ifdef CONFIG_ACPI + if (acpi_has_method(ACPI_HANDLE(&pdata->pdev->dev), "_RST")) { + acpi_evaluate_object(ACPI_HANDLE(&pdata->pdev->dev), + "_RST", NULL, NULL); + } else if (acpi_has_method(ACPI_HANDLE(&pdata->pdev->dev), + "_INI")) { + acpi_evaluate_object(ACPI_HANDLE(&pdata->pdev->dev), + "_INI", NULL, NULL); + } +#endif } xgene_enet_ecc_init(pdata); @@ -292,6 +308,7 @@ static void xgene_enet_xgcle_bypass(struct xgene_enet_pdata *pdata, static void xgene_enet_shutdown(struct xgene_enet_pdata *pdata) { + struct device *dev = &pdata->pdev->dev; struct xgene_enet_desc_ring *ring; u32 pb, val; int i; @@ -313,6 +330,11 @@ static void xgene_enet_shutdown(struct xgene_enet_pdata *pdata) pb |= BIT(val); } xgene_enet_wr_ring_if(pdata, ENET_CFGSSQMIWQRESET_ADDR, pb); + + if (dev->of_node) { + if (!IS_ERR(pdata->clk)) + clk_disable_unprepare(pdata->clk); + } } static void xgene_enet_clear(struct xgene_enet_pdata *pdata, -- cgit v0.10.2 From 43b3cf6634a4ae2eac3b6f08019db8f19a114811 Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:40 -0700 Subject: drivers: net: phy: xgene: Add MDIO driver Currently, SGMII based 1G rely on the hardware registers for link state and sometimes it's not reliable. To get most accurate link state, this interface has to use the MDIO bus to poll the PHY. In X-Gene SoC, MDIO bus is shared across RGMII and SGMII based 1G interfaces, so adding this driver to manage MDIO bus. This driver registers the mdio bus and registers the PHYs connected to it. Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Signed-off-by: David S. Miller diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig index 1d7b208..47a6434 100644 --- a/drivers/net/phy/Kconfig +++ b/drivers/net/phy/Kconfig @@ -301,6 +301,12 @@ config MDIO_HISI_FEMAC This module provides a driver for the MDIO busses found in the Hisilicon SoC that have an Fast Ethernet MAC. +config MDIO_XGENE + tristate "APM X-Gene SoC MDIO bus controller" + help + This module provides a driver for the MDIO busses found in the + APM X-Gene SoC's. + endif # PHYLIB config MICREL_KS8995MA diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile index 19e38a9..534dfa7 100644 --- a/drivers/net/phy/Makefile +++ b/drivers/net/phy/Makefile @@ -48,3 +48,4 @@ obj-$(CONFIG_MICROCHIP_PHY) += microchip.o obj-$(CONFIG_MDIO_BCM_IPROC) += mdio-bcm-iproc.o obj-$(CONFIG_INTEL_XWAY_PHY) += intel-xway.o obj-$(CONFIG_MDIO_HISI_FEMAC) += mdio-hisi-femac.o +obj-$(CONFIG_MDIO_XGENE) += mdio-xgene.o diff --git a/drivers/net/phy/mdio-xgene.c b/drivers/net/phy/mdio-xgene.c new file mode 100644 index 0000000..d94a978 --- /dev/null +++ b/drivers/net/phy/mdio-xgene.c @@ -0,0 +1,477 @@ +/* Applied Micro X-Gene SoC MDIO Driver + * + * Copyright (c) 2016, Applied Micro Circuits Corporation + * Author: Iyappan Subramanian + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "mdio-xgene.h" + +static bool xgene_mdio_status; + +static u32 xgene_enet_rd_mac(void __iomem *base_addr, u32 rd_addr) +{ + void __iomem *addr, *rd, *cmd, *cmd_done; + u32 done, rd_data = BUSY_MASK; + u8 wait = 10; + + addr = base_addr + MAC_ADDR_REG_OFFSET; + rd = base_addr + MAC_READ_REG_OFFSET; + cmd = base_addr + MAC_COMMAND_REG_OFFSET; + cmd_done = base_addr + MAC_COMMAND_DONE_REG_OFFSET; + + iowrite32(rd_addr, addr); + iowrite32(XGENE_ENET_RD_CMD, cmd); + + while (wait--) { + done = ioread32(cmd_done); + if (done) + break; + udelay(1); + } + + if (!done) + return rd_data; + + rd_data = ioread32(rd); + iowrite32(0, cmd); + + return rd_data; +} + +static void xgene_enet_wr_mac(void __iomem *base_addr, u32 wr_addr, u32 wr_data) +{ + void __iomem *addr, *wr, *cmd, *cmd_done; + u8 wait = 10; + u32 done; + + addr = base_addr + MAC_ADDR_REG_OFFSET; + wr = base_addr + MAC_WRITE_REG_OFFSET; + cmd = base_addr + MAC_COMMAND_REG_OFFSET; + cmd_done = base_addr + MAC_COMMAND_DONE_REG_OFFSET; + + iowrite32(wr_addr, addr); + iowrite32(wr_data, wr); + iowrite32(XGENE_ENET_WR_CMD, cmd); + + while (wait--) { + done = ioread32(cmd_done); + if (done) + break; + udelay(1); + } + + if (!done) + pr_err("MCX mac write failed, addr: 0x%04x\n", wr_addr); + + iowrite32(0, cmd); +} + +int xgene_mdio_rgmii_read(struct mii_bus *bus, int phy_id, int reg) +{ + void __iomem *addr = (void __iomem *)bus->priv; + u32 data, done; + u8 wait = 10; + + data = SET_VAL(PHY_ADDR, phy_id) | SET_VAL(REG_ADDR, reg); + xgene_enet_wr_mac(addr, MII_MGMT_ADDRESS_ADDR, data); + xgene_enet_wr_mac(addr, MII_MGMT_COMMAND_ADDR, READ_CYCLE_MASK); + do { + usleep_range(5, 10); + done = xgene_enet_rd_mac(addr, MII_MGMT_INDICATORS_ADDR); + } while ((done & BUSY_MASK) && wait--); + + if (done & BUSY_MASK) { + dev_err(&bus->dev, "MII_MGMT read failed\n"); + return -EBUSY; + } + + data = xgene_enet_rd_mac(addr, MII_MGMT_STATUS_ADDR); + xgene_enet_wr_mac(addr, MII_MGMT_COMMAND_ADDR, 0); + + return data; +} +EXPORT_SYMBOL(xgene_mdio_rgmii_read); + +int xgene_mdio_rgmii_write(struct mii_bus *bus, int phy_id, int reg, u16 data) +{ + void __iomem *addr = (void __iomem *)bus->priv; + u32 val, done; + u8 wait = 10; + + val = SET_VAL(PHY_ADDR, phy_id) | SET_VAL(REG_ADDR, reg); + xgene_enet_wr_mac(addr, MII_MGMT_ADDRESS_ADDR, val); + + xgene_enet_wr_mac(addr, MII_MGMT_CONTROL_ADDR, data); + do { + usleep_range(5, 10); + done = xgene_enet_rd_mac(addr, MII_MGMT_INDICATORS_ADDR); + } while ((done & BUSY_MASK) && wait--); + + if (done & BUSY_MASK) { + dev_err(&bus->dev, "MII_MGMT write failed\n"); + return -EBUSY; + } + + return 0; +} +EXPORT_SYMBOL(xgene_mdio_rgmii_write); + +static u32 xgene_menet_rd_diag_csr(struct xgene_mdio_pdata *pdata, u32 offset) +{ + return ioread32(pdata->diag_csr_addr + offset); +} + +static void xgene_menet_wr_diag_csr(struct xgene_mdio_pdata *pdata, + u32 offset, u32 val) +{ + iowrite32(val, pdata->diag_csr_addr + offset); +} + +static int xgene_enet_ecc_init(struct xgene_mdio_pdata *pdata) +{ + u32 data; + u8 wait = 10; + + xgene_menet_wr_diag_csr(pdata, MENET_CFG_MEM_RAM_SHUTDOWN_ADDR, 0x0); + do { + usleep_range(100, 110); + data = xgene_menet_rd_diag_csr(pdata, MENET_BLOCK_MEM_RDY_ADDR); + } while ((data != 0xffffffff) && wait--); + + if (data != 0xffffffff) { + dev_err(pdata->dev, "Failed to release memory from shutdown\n"); + return -ENODEV; + } + + return 0; +} + +static void xgene_gmac_reset(struct xgene_mdio_pdata *pdata) +{ + xgene_enet_wr_mac(pdata->mac_csr_addr, MAC_CONFIG_1_ADDR, SOFT_RESET); + xgene_enet_wr_mac(pdata->mac_csr_addr, MAC_CONFIG_1_ADDR, 0); +} + +static int xgene_mdio_reset(struct xgene_mdio_pdata *pdata) +{ + int ret; + + if (pdata->dev->of_node) { + clk_prepare_enable(pdata->clk); + udelay(5); + clk_disable_unprepare(pdata->clk); + udelay(5); + clk_prepare_enable(pdata->clk); + udelay(5); + } else { +#ifdef CONFIG_ACPI + acpi_evaluate_object(ACPI_HANDLE(pdata->dev), + "_RST", NULL, NULL); +#endif + } + + ret = xgene_enet_ecc_init(pdata); + if (ret) + return ret; + xgene_gmac_reset(pdata); + + return 0; +} + +static void xgene_enet_rd_mdio_csr(void __iomem *base_addr, + u32 offset, u32 *val) +{ + void __iomem *addr = base_addr + offset; + + *val = ioread32(addr); +} + +static void xgene_enet_wr_mdio_csr(void __iomem *base_addr, + u32 offset, u32 val) +{ + void __iomem *addr = base_addr + offset; + + iowrite32(val, addr); +} + +static int xgene_xfi_mdio_write(struct mii_bus *bus, int phy_id, + int reg, u16 data) +{ + void __iomem *addr = (void __iomem *)bus->priv; + int timeout = 100; + u32 status, val; + + val = SET_VAL(HSTPHYADX, phy_id) | SET_VAL(HSTREGADX, reg) | + SET_VAL(HSTMIIMWRDAT, data); + xgene_enet_wr_mdio_csr(addr, MIIM_FIELD_ADDR, data); + + val = HSTLDCMD | SET_VAL(HSTMIIMCMD, MIIM_CMD_LEGACY_WRITE); + xgene_enet_wr_mdio_csr(addr, MIIM_COMMAND_ADDR, val); + + do { + usleep_range(5, 10); + xgene_enet_rd_mdio_csr(addr, MIIM_INDICATOR_ADDR, &status); + } while ((status & BUSY_MASK) && timeout--); + + xgene_enet_wr_mdio_csr(addr, MIIM_COMMAND_ADDR, 0); + + return 0; +} + +static int xgene_xfi_mdio_read(struct mii_bus *bus, int phy_id, int reg) +{ + void __iomem *addr = (void __iomem *)bus->priv; + u32 data, status, val; + int timeout = 100; + + val = SET_VAL(HSTPHYADX, phy_id) | SET_VAL(HSTREGADX, reg); + xgene_enet_wr_mdio_csr(addr, MIIM_FIELD_ADDR, val); + + val = HSTLDCMD | SET_VAL(HSTMIIMCMD, MIIM_CMD_LEGACY_READ); + xgene_enet_wr_mdio_csr(addr, MIIM_COMMAND_ADDR, val); + + do { + usleep_range(5, 10); + xgene_enet_rd_mdio_csr(addr, MIIM_INDICATOR_ADDR, &status); + } while ((status & BUSY_MASK) && timeout--); + + if (status & BUSY_MASK) { + pr_err("XGENET_MII_MGMT write failed\n"); + return -EBUSY; + } + + xgene_enet_rd_mdio_csr(addr, MIIMRD_FIELD_ADDR, &data); + xgene_enet_wr_mdio_csr(addr, MIIM_COMMAND_ADDR, 0); + + return data; +} + +struct phy_device *xgene_enet_phy_register(struct mii_bus *bus, int phy_addr) +{ + struct phy_device *phy_dev; + + phy_dev = get_phy_device(bus, phy_addr, false); + if (!phy_dev || IS_ERR(phy_dev)) + return NULL; + + if (phy_device_register(phy_dev)) + phy_device_free(phy_dev); + + return phy_dev; +} +EXPORT_SYMBOL(xgene_enet_phy_register); + +#ifdef CONFIG_ACPI +static acpi_status acpi_register_phy(acpi_handle handle, u32 lvl, + void *context, void **ret) +{ + struct mii_bus *mdio = context; + struct acpi_device *adev; + struct phy_device *phy_dev; + const union acpi_object *obj; + u32 phy_addr; + + if (acpi_bus_get_device(handle, &adev)) + return AE_OK; + + if (acpi_dev_get_property(adev, "phy-channel", ACPI_TYPE_INTEGER, &obj)) + return AE_OK; + phy_addr = obj->integer.value; + + phy_dev = xgene_enet_phy_register(mdio, phy_addr); + adev->driver_data = phy_dev; + + return AE_OK; +} +#endif + +static int xgene_mdio_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct mii_bus *mdio_bus; + const struct of_device_id *of_id; + struct resource *res; + struct xgene_mdio_pdata *pdata; + void __iomem *csr_base; + int mdio_id = 0, ret = 0; + + of_id = of_match_device(xgene_mdio_of_match, &pdev->dev); + if (of_id) { + mdio_id = (enum xgene_mdio_id)of_id->data; + } else { +#ifdef CONFIG_ACPI + const struct acpi_device_id *acpi_id; + + acpi_id = acpi_match_device(xgene_mdio_acpi_match, &pdev->dev); + if (acpi_id) + mdio_id = (enum xgene_mdio_id)acpi_id->driver_data; +#endif + } + + if (!mdio_id) + return -ENODEV; + + pdata = devm_kzalloc(dev, sizeof(struct xgene_mdio_pdata), GFP_KERNEL); + if (!pdata) + return -ENOMEM; + pdata->mdio_id = mdio_id; + pdata->dev = dev; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + csr_base = devm_ioremap_resource(dev, res); + if (IS_ERR(csr_base)) { + dev_err(dev, "Unable to retrieve mac CSR region\n"); + return PTR_ERR(csr_base); + } + pdata->mac_csr_addr = csr_base; + pdata->mdio_csr_addr = csr_base + BLOCK_XG_MDIO_CSR_OFFSET; + pdata->diag_csr_addr = csr_base + BLOCK_DIAG_CSR_OFFSET; + + if (dev->of_node) { + pdata->clk = devm_clk_get(dev, NULL); + if (IS_ERR(pdata->clk)) { + dev_err(dev, "Unable to retrieve clk\n"); + return PTR_ERR(pdata->clk); + } + } + + ret = xgene_mdio_reset(pdata); + if (ret) + return ret; + + mdio_bus = mdiobus_alloc(); + if (!mdio_bus) + return -ENOMEM; + + mdio_bus->name = "APM X-Gene MDIO bus"; + + if (mdio_id == XGENE_MDIO_RGMII) { + mdio_bus->read = xgene_mdio_rgmii_read; + mdio_bus->write = xgene_mdio_rgmii_write; + mdio_bus->priv = (void __force *)pdata->mac_csr_addr; + snprintf(mdio_bus->id, MII_BUS_ID_SIZE, "%s", + "xgene-mii-rgmii"); + } else { + mdio_bus->read = xgene_xfi_mdio_read; + mdio_bus->write = xgene_xfi_mdio_write; + mdio_bus->priv = (void __force *)pdata->mdio_csr_addr; + snprintf(mdio_bus->id, MII_BUS_ID_SIZE, "%s", + "xgene-mii-xfi"); + } + + mdio_bus->parent = dev; + platform_set_drvdata(pdev, pdata); + + if (dev->of_node) { + ret = of_mdiobus_register(mdio_bus, dev->of_node); + } else { +#ifdef CONFIG_ACPI + /* Mask out all PHYs from auto probing. */ + mdio_bus->phy_mask = ~0; + ret = mdiobus_register(mdio_bus); + if (ret) + goto out; + + acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_HANDLE(dev), 1, + acpi_register_phy, NULL, mdio_bus, NULL); +#endif + } + + if (ret) + goto out; + + pdata->mdio_bus = mdio_bus; + xgene_mdio_status = true; + + return 0; + +out: + mdiobus_free(mdio_bus); + + return ret; +} + +static int xgene_mdio_remove(struct platform_device *pdev) +{ + struct xgene_mdio_pdata *pdata = platform_get_drvdata(pdev); + struct mii_bus *mdio_bus = pdata->mdio_bus; + struct device *dev = &pdev->dev; + + mdiobus_unregister(mdio_bus); + mdiobus_free(mdio_bus); + + if (dev->of_node) { + if (IS_ERR(pdata->clk)) + clk_disable_unprepare(pdata->clk); + } + + return 0; +} + +#ifdef CONFIG_OF +static const struct of_device_id xgene_mdio_of_match[] = { + { + .compatible = "apm,xgene-mdio-rgmii", + .data = (void *)XGENE_MDIO_RGMII + }, + { + .compatible = "apm,xgene-mdio-xfi", + .data = (void *)XGENE_MDIO_XFI + }, + {}, +}; + +MODULE_DEVICE_TABLE(of, xgene_mdio_of_match); +#endif + +#ifdef CONFIG_ACPI +static const struct acpi_device_id xgene_mdio_acpi_match[] = { + { "APMC0D65", XGENE_MDIO_RGMII }, + { "APMC0D66", XGENE_MDIO_XFI }, + { } +}; + +MODULE_DEVICE_TABLE(acpi, xgene_mdio_acpi_match); +#endif + +static struct platform_driver xgene_mdio_driver = { + .driver = { + .name = "xgene-mdio", + .of_match_table = of_match_ptr(xgene_mdio_of_match), + .acpi_match_table = ACPI_PTR(xgene_mdio_acpi_match), + }, + .probe = xgene_mdio_probe, + .remove = xgene_mdio_remove, +}; + +module_platform_driver(xgene_mdio_driver); + +MODULE_DESCRIPTION("APM X-Gene SoC MDIO driver"); +MODULE_AUTHOR("Iyappan Subramanian "); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/phy/mdio-xgene.h b/drivers/net/phy/mdio-xgene.h new file mode 100644 index 0000000..354241b --- /dev/null +++ b/drivers/net/phy/mdio-xgene.h @@ -0,0 +1,143 @@ +/* Applied Micro X-Gene SoC MDIO Driver + * + * Copyright (c) 2016, Applied Micro Circuits Corporation + * Author: Iyappan Subramanian + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +#ifndef __MDIO_XGENE_H__ +#define __MDIO_XGENE_H__ + +#define BLOCK_XG_MDIO_CSR_OFFSET 0x5000 +#define BLOCK_DIAG_CSR_OFFSET 0xd000 +#define XGENET_CONFIG_REG_ADDR 0x20 + +#define MAC_ADDR_REG_OFFSET 0x00 +#define MAC_COMMAND_REG_OFFSET 0x04 +#define MAC_WRITE_REG_OFFSET 0x08 +#define MAC_READ_REG_OFFSET 0x0c +#define MAC_COMMAND_DONE_REG_OFFSET 0x10 + +#define CLKEN_OFFSET 0x08 +#define SRST_OFFSET 0x00 + +#define MENET_CFG_MEM_RAM_SHUTDOWN_ADDR 0x70 +#define MENET_BLOCK_MEM_RDY_ADDR 0x74 + +#define MAC_CONFIG_1_ADDR 0x00 +#define MII_MGMT_COMMAND_ADDR 0x24 +#define MII_MGMT_ADDRESS_ADDR 0x28 +#define MII_MGMT_CONTROL_ADDR 0x2c +#define MII_MGMT_STATUS_ADDR 0x30 +#define MII_MGMT_INDICATORS_ADDR 0x34 +#define SOFT_RESET BIT(31) + +#define MII_MGMT_CONFIG_ADDR 0x20 +#define MII_MGMT_COMMAND_ADDR 0x24 +#define MII_MGMT_ADDRESS_ADDR 0x28 +#define MII_MGMT_CONTROL_ADDR 0x2c +#define MII_MGMT_STATUS_ADDR 0x30 +#define MII_MGMT_INDICATORS_ADDR 0x34 + +#define MIIM_COMMAND_ADDR 0x20 +#define MIIM_FIELD_ADDR 0x24 +#define MIIM_CONFIGURATION_ADDR 0x28 +#define MIIM_LINKFAILVECTOR_ADDR 0x2c +#define MIIM_INDICATOR_ADDR 0x30 +#define MIIMRD_FIELD_ADDR 0x34 + +#define MDIO_CSR_OFFSET 0x5000 + +#define REG_ADDR_POS 0 +#define REG_ADDR_LEN 5 +#define PHY_ADDR_POS 8 +#define PHY_ADDR_LEN 5 + +#define HSTMIIMWRDAT_POS 0 +#define HSTMIIMWRDAT_LEN 16 +#define HSTPHYADX_POS 23 +#define HSTPHYADX_LEN 5 +#define HSTREGADX_POS 18 +#define HSTREGADX_LEN 5 +#define HSTLDCMD BIT(3) +#define HSTMIIMCMD_POS 0 +#define HSTMIIMCMD_LEN 3 + +#define BUSY_MASK BIT(0) +#define READ_CYCLE_MASK BIT(0) + +enum xgene_enet_cmd { + XGENE_ENET_WR_CMD = BIT(31), + XGENE_ENET_RD_CMD = BIT(30) +}; + +enum { + MIIM_CMD_IDLE, + MIIM_CMD_LEGACY_WRITE, + MIIM_CMD_LEGACY_READ, +}; + +enum xgene_mdio_id { + XGENE_MDIO_RGMII = 1, + XGENE_MDIO_XFI +}; + +struct xgene_mdio_pdata { + struct clk *clk; + struct device *dev; + void __iomem *mac_csr_addr; + void __iomem *diag_csr_addr; + void __iomem *mdio_csr_addr; + struct mii_bus *mdio_bus; + int mdio_id; +}; + +/* Set the specified value into a bit-field defined by its starting position + * and length within a single u64. + */ +static inline u64 xgene_enet_set_field_value(int pos, int len, u64 val) +{ + return (val & ((1ULL << len) - 1)) << pos; +} + +#define SET_VAL(field, val) \ + xgene_enet_set_field_value(field ## _POS, field ## _LEN, val) + +#define SET_BIT(field) \ + xgene_enet_set_field_value(field ## _POS, 1, 1) + +/* Get the value from a bit-field defined by its starting position + * and length within the specified u64. + */ +static inline u64 xgene_enet_get_field_value(int pos, int len, u64 src) +{ + return (src >> pos) & ((1ULL << len) - 1); +} + +#define GET_VAL(field, src) \ + xgene_enet_get_field_value(field ## _POS, field ## _LEN, src) + +#define GET_BIT(field, src) \ + xgene_enet_get_field_value(field ## _POS, 1, src) + +static const struct of_device_id xgene_mdio_of_match[]; +#ifdef CONFIG_ACPI +static const struct acpi_device_id xgene_mdio_acpi_match[]; +#endif +int xgene_mdio_rgmii_read(struct mii_bus *bus, int phy_id, int reg); +int xgene_mdio_rgmii_write(struct mii_bus *bus, int phy_id, int reg, u16 data); +struct phy_device *xgene_enet_phy_register(struct mii_bus *bus, int phy_addr); + +#endif /* __MDIO_XGENE_H__ */ -- cgit v0.10.2 From 8089a96f601bdfe3e1b41d14bb703aafaf1b8f34 Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:41 -0700 Subject: drivers: net: xgene: Add backward compatibility This patch adds xgene_enet_check_phy_hanlde() function that checks whether MDIO driver is probed successfully and sets pdata->mdio_driver to true. If MDIO driver is not probed, ethernet driver falls back to backward compatibility mode. Since enum xgene_enet_cmd is used by MDIO driver, removing this from ethernet driver. Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index 91a67a0..b8b643f 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -572,7 +572,9 @@ static void xgene_gmac_init(struct xgene_enet_pdata *pdata) { u32 value; - xgene_gmac_reset(pdata); + if (!pdata->mdio_driver) + xgene_gmac_reset(pdata); + xgene_gmac_set_speed(pdata); xgene_gmac_set_mac_addr(pdata); @@ -680,6 +682,11 @@ static int xgene_enet_reset(struct xgene_enet_pdata *pdata) if (!xgene_ring_mgr_init(pdata)) return -ENODEV; + if (pdata->mdio_driver) { + xgene_enet_config_ring_if_assoc(pdata); + return 0; + } + if (dev->of_node) { clk_prepare_enable(pdata->clk); udelay(5); @@ -799,21 +806,47 @@ static void xgene_enet_adjust_link(struct net_device *ndev) } } -static int xgene_enet_phy_connect(struct net_device *ndev) +#ifdef CONFIG_ACPI +static struct acpi_device *acpi_phy_find_device(struct device *dev) +{ + struct acpi_reference_args args; + struct fwnode_handle *fw_node; + int status; + + fw_node = acpi_fwnode_handle(ACPI_COMPANION(dev)); + status = acpi_node_get_property_reference(fw_node, "phy-handle", 0, + &args); + if (ACPI_FAILURE(status)) { + dev_dbg(dev, "No matching phy in ACPI table\n"); + return NULL; + } + + return args.adev; +} +#endif + +int xgene_enet_phy_connect(struct net_device *ndev) { struct xgene_enet_pdata *pdata = netdev_priv(ndev); - struct device_node *phy_np; + struct device_node *np; struct phy_device *phy_dev; struct device *dev = &pdata->pdev->dev; + struct acpi_device *adev; + int i; if (dev->of_node) { - phy_np = of_parse_phandle(dev->of_node, "phy-handle", 0); - if (!phy_np) { + for (i = 0 ; i < 2; i++) { + np = of_parse_phandle(dev->of_node, "phy-handle", i); + if (np) + break; + } + + if (!np) { netdev_dbg(ndev, "No phy-handle found in DT\n"); return -ENODEV; } - phy_dev = of_phy_connect(ndev, phy_np, &xgene_enet_adjust_link, + phy_dev = of_phy_connect(ndev, np, &xgene_enet_adjust_link, 0, pdata->phy_mode); if (!phy_dev) { netdev_err(ndev, "Could not connect to PHY\n"); @@ -822,6 +855,11 @@ static int xgene_enet_phy_connect(struct net_device *ndev) pdata->phy_dev = phy_dev; } else { +#ifdef CONFIG_ACPI + adev = acpi_phy_find_device(dev); + if (adev) + pdata->phy_dev = adev->driver_data; + phy_dev = pdata->phy_dev; if (!phy_dev || @@ -830,6 +868,7 @@ static int xgene_enet_phy_connect(struct net_device *ndev) netdev_err(ndev, "Could not connect to PHY\n"); return -ENODEV; } +#endif } pdata->phy_speed = SPEED_UNKNOWN; @@ -930,6 +969,12 @@ int xgene_enet_mdio_config(struct xgene_enet_pdata *pdata) return ret; } +void xgene_enet_phy_disconnect(struct xgene_enet_pdata *pdata) +{ + if (pdata->phy_dev) + phy_disconnect(pdata->phy_dev); +} + void xgene_enet_mdio_remove(struct xgene_enet_pdata *pdata) { if (pdata->phy_dev) diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h index eec55c1..179a44d 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.h @@ -301,11 +301,6 @@ enum xgene_enet_ring_bufnum { RING_BUFNUM_INVALID }; -enum xgene_enet_cmd { - XGENE_ENET_WR_CMD = BIT(31), - XGENE_ENET_RD_CMD = BIT(30) -}; - enum xgene_enet_err_code { HBF_READ_DATA = 3, HBF_LL_READ = 4, @@ -351,6 +346,8 @@ void xgene_enet_parse_error(struct xgene_enet_desc_ring *ring, int xgene_enet_mdio_config(struct xgene_enet_pdata *pdata); void xgene_enet_mdio_remove(struct xgene_enet_pdata *pdata); bool xgene_ring_mgr_init(struct xgene_enet_pdata *p); +int xgene_enet_phy_connect(struct net_device *ndev); +void xgene_enet_phy_disconnect(struct xgene_enet_pdata *pdata); extern const struct xgene_mac_ops xgene_gmac_ops; extern const struct xgene_port_ops xgene_gport_ops; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index 87e5929..6398337 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -763,7 +763,7 @@ static int xgene_enet_close(struct net_device *ndev) mac_ops->tx_disable(pdata); mac_ops->rx_disable(pdata); - if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) + if (pdata->phy_dev) phy_stop(pdata->phy_dev); else cancel_delayed_work_sync(&pdata->link_work); @@ -1295,6 +1295,23 @@ static int xgene_enet_get_irqs(struct xgene_enet_pdata *pdata) return 0; } +static int xgene_enet_check_phy_handle(struct xgene_enet_pdata *pdata) +{ + int ret; + + if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) + return 0; + + if (!IS_ENABLED(CONFIG_MDIO_XGENE)) + return 0; + + ret = xgene_enet_phy_connect(pdata->ndev); + if (!ret) + pdata->mdio_driver = true; + + return 0; +} + static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata) { struct platform_device *pdev; @@ -1380,6 +1397,10 @@ static int xgene_enet_get_resources(struct xgene_enet_pdata *pdata) if (ret) return ret; + ret = xgene_enet_check_phy_handle(pdata); + if (ret) + return ret; + pdata->clk = devm_clk_get(&pdev->dev, NULL); if (IS_ERR(pdata->clk)) { /* Firmware may have set up the clock already. */ @@ -1574,7 +1595,7 @@ static int xgene_enet_probe(struct platform_device *pdev) struct net_device *ndev; struct xgene_enet_pdata *pdata; struct device *dev = &pdev->dev; - const struct xgene_mac_ops *mac_ops; + void (*link_state)(struct work_struct *); const struct of_device_id *of_id; int ret; @@ -1636,14 +1657,17 @@ static int xgene_enet_probe(struct platform_device *pdev) if (ret) goto err_netdev; - mac_ops = pdata->mac_ops; - if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) { - ret = xgene_enet_mdio_config(pdata); - if (ret) - goto err_netdev; - } else { - INIT_DELAYED_WORK(&pdata->link_work, mac_ops->link_state); + link_state = pdata->mac_ops->link_state; + if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) { + INIT_DELAYED_WORK(&pdata->link_work, link_state); + } else if (!pdata->mdio_driver) { + if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) + ret = xgene_enet_mdio_config(pdata); + else + INIT_DELAYED_WORK(&pdata->link_work, link_state); } + if (ret) + goto err; xgene_enet_napi_add(pdata); ret = register_netdev(ndev); @@ -1676,11 +1700,11 @@ static int xgene_enet_remove(struct platform_device *pdev) dev_close(ndev); rtnl_unlock(); - mac_ops->rx_disable(pdata); - mac_ops->tx_disable(pdata); - - if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) + if (pdata->mdio_driver) + xgene_enet_phy_disconnect(pdata); + else if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) xgene_enet_mdio_remove(pdata); + unregister_netdev(ndev); pdata->port_ops->shutdown(pdata); xgene_enet_delete_desc_rings(pdata); diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h index 681a473..217546e 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h @@ -38,6 +38,7 @@ #include "xgene_enet_hw.h" #include "xgene_enet_cle.h" #include "xgene_enet_ring2.h" +#include "../../../phy/mdio-xgene.h" #define XGENE_DRV_VERSION "v1.0" #define XGENE_ENET_MAX_MTU 1536 @@ -214,6 +215,7 @@ struct xgene_enet_pdata { u32 mss; u8 tx_delay; u8 rx_delay; + bool mdio_driver; }; struct xgene_indirect_ctl { @@ -223,34 +225,6 @@ struct xgene_indirect_ctl { void __iomem *cmd_done; }; -/* Set the specified value into a bit-field defined by its starting position - * and length within a single u64. - */ -static inline u64 xgene_enet_set_field_value(int pos, int len, u64 val) -{ - return (val & ((1ULL << len) - 1)) << pos; -} - -#define SET_VAL(field, val) \ - xgene_enet_set_field_value(field ## _POS, field ## _LEN, val) - -#define SET_BIT(field) \ - xgene_enet_set_field_value(field ## _POS, 1, 1) - -/* Get the value from a bit-field defined by its starting position - * and length within the specified u64. - */ -static inline u64 xgene_enet_get_field_value(int pos, int len, u64 src) -{ - return (src >> pos) & ((1ULL << len) - 1); -} - -#define GET_VAL(field, src) \ - xgene_enet_get_field_value(field ## _POS, field ## _LEN, src) - -#define GET_BIT(field, src) \ - xgene_enet_get_field_value(field ## _POS, 1, src) - static inline struct device *ndev_to_dev(struct net_device *ndev) { return ndev->dev.parent; diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c index 03a70c5..d12e9cb 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c @@ -366,7 +366,9 @@ static void xgene_sgmac_init(struct xgene_enet_pdata *p) u32 cfg_bypass_reg, rx_dv_gate_reg; u32 data, offset; - xgene_sgmac_reset(p); + if (!(p->enet_id == XGENE_ENET2 && p->mdio_driver)) + xgene_sgmac_reset(p); + xgene_sgmii_enable_autoneg(p); xgene_sgmac_set_speed(p); xgene_sgmac_set_mac_addr(p); @@ -445,6 +447,11 @@ static int xgene_enet_reset(struct xgene_enet_pdata *p) if (!xgene_ring_mgr_init(p)) return -ENODEV; + if (p->mdio_driver && p->enet_id == XGENE_ENET2) { + xgene_enet_config_ring_if_assoc(p); + return 0; + } + if (p->enet_id == XGENE_ENET2) xgene_enet_wr_clkrst_csr(p, XGENET_CONFIG_REG_ADDR, SGMII_EN); -- cgit v0.10.2 From 47c62b6d5f9466fbb81ee45273d4a9929200a273 Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:42 -0700 Subject: drivers: net: xgene: Enable MDIO driver This patch enables MDIO driver by, - Selecting MDIO_XGENE - Changed open and close to use phy_start and phy_stop - Changed to use mac_ops->tx(rx)_enable and tx(rx)_disable Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/apm/xgene/Kconfig b/drivers/net/ethernet/apm/xgene/Kconfig index 19e38af..300e3b5 100644 --- a/drivers/net/ethernet/apm/xgene/Kconfig +++ b/drivers/net/ethernet/apm/xgene/Kconfig @@ -3,6 +3,7 @@ config NET_XGENE depends on HAS_DMA depends on ARCH_XGENE || COMPILE_TEST select PHYLIB + select MDIO_XGENE help This is the Ethernet driver for the on-chip ethernet interface on the APM X-Gene SoC. diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index b8b643f..27a7a1c 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -794,13 +794,13 @@ static void xgene_enet_adjust_link(struct net_device *ndev) if (pdata->phy_speed != phydev->speed) { pdata->phy_speed = phydev->speed; mac_ops->set_speed(pdata); - xgene_gmac_rx_enable(pdata); - xgene_gmac_tx_enable(pdata); + mac_ops->rx_enable(pdata); + mac_ops->tx_enable(pdata); phy_print_status(phydev); } } else { - xgene_gmac_rx_disable(pdata); - xgene_gmac_tx_disable(pdata); + mac_ops->rx_disable(pdata); + mac_ops->tx_disable(pdata); pdata->phy_speed = SPEED_UNKNOWN; phy_print_status(phydev); } diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index 6398337..d1d6b5e 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -739,9 +739,9 @@ static int xgene_enet_open(struct net_device *ndev) if (ret) return ret; - if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) + if (pdata->phy_dev) { phy_start(pdata->phy_dev); - else { + } else { schedule_delayed_work(&pdata->link_work, PHY_POLL_LINK_OFF); netif_carrier_off(ndev); } -- cgit v0.10.2 From 8c1519632fa6cb34a5ff4dddffd4a182c4fae1b6 Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:43 -0700 Subject: drivers: net: xgene: Use exported functions This patch reuses the mdio read/write and phy_register functions and removed the local definitions. Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index 27a7a1c..8a2a221 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -381,59 +381,6 @@ static void xgene_enet_rd_mcx_mac(struct xgene_enet_pdata *pdata, rd_addr); } -static int xgene_mii_phy_write(struct xgene_enet_pdata *pdata, int phy_id, - u32 reg, u16 data) -{ - u32 addr = 0, wr_data = 0; - u32 done; - u8 wait = 10; - - PHY_ADDR_SET(&addr, phy_id); - REG_ADDR_SET(&addr, reg); - xgene_enet_wr_mcx_mac(pdata, MII_MGMT_ADDRESS_ADDR, addr); - - PHY_CONTROL_SET(&wr_data, data); - xgene_enet_wr_mcx_mac(pdata, MII_MGMT_CONTROL_ADDR, wr_data); - do { - usleep_range(5, 10); - xgene_enet_rd_mcx_mac(pdata, MII_MGMT_INDICATORS_ADDR, &done); - } while ((done & BUSY_MASK) && wait--); - - if (done & BUSY_MASK) { - netdev_err(pdata->ndev, "MII_MGMT write failed\n"); - return -EBUSY; - } - - return 0; -} - -static int xgene_mii_phy_read(struct xgene_enet_pdata *pdata, - u8 phy_id, u32 reg) -{ - u32 addr = 0; - u32 data, done; - u8 wait = 10; - - PHY_ADDR_SET(&addr, phy_id); - REG_ADDR_SET(&addr, reg); - xgene_enet_wr_mcx_mac(pdata, MII_MGMT_ADDRESS_ADDR, addr); - xgene_enet_wr_mcx_mac(pdata, MII_MGMT_COMMAND_ADDR, READ_CYCLE_MASK); - do { - usleep_range(5, 10); - xgene_enet_rd_mcx_mac(pdata, MII_MGMT_INDICATORS_ADDR, &done); - } while ((done & BUSY_MASK) && wait--); - - if (done & BUSY_MASK) { - netdev_err(pdata->ndev, "MII_MGMT read failed\n"); - return -EBUSY; - } - - xgene_enet_rd_mcx_mac(pdata, MII_MGMT_STATUS_ADDR, &data); - xgene_enet_wr_mcx_mac(pdata, MII_MGMT_COMMAND_ADDR, 0); - - return data; -} - static void xgene_gmac_set_mac_addr(struct xgene_enet_pdata *pdata) { u32 addr0, addr1; @@ -762,28 +709,6 @@ static void xgene_gport_shutdown(struct xgene_enet_pdata *pdata) } } -static int xgene_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) -{ - struct xgene_enet_pdata *pdata = bus->priv; - u32 val; - - val = xgene_mii_phy_read(pdata, mii_id, regnum); - netdev_dbg(pdata->ndev, "mdio_rd: bus=%d reg=%d val=%x\n", - mii_id, regnum, val); - - return val; -} - -static int xgene_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, - u16 val) -{ - struct xgene_enet_pdata *pdata = bus->priv; - - netdev_dbg(pdata->ndev, "mdio_wr: bus=%d reg=%d val=%x\n", - mii_id, regnum, val); - return xgene_mii_phy_write(pdata, mii_id, regnum, val); -} - static void xgene_enet_adjust_link(struct net_device *ndev) { struct xgene_enet_pdata *pdata = netdev_priv(ndev); @@ -888,8 +813,8 @@ static int xgene_mdiobus_register(struct xgene_enet_pdata *pdata, struct phy_device *phy; struct device_node *child_np; struct device_node *mdio_np = NULL; + u32 phy_addr; int ret; - u32 phy_id; if (dev->of_node) { for_each_child_of_node(dev->of_node, child_np) { @@ -916,21 +841,17 @@ static int xgene_mdiobus_register(struct xgene_enet_pdata *pdata, if (ret) return ret; - ret = device_property_read_u32(dev, "phy-channel", &phy_id); + ret = device_property_read_u32(dev, "phy-channel", &phy_addr); if (ret) - ret = device_property_read_u32(dev, "phy-addr", &phy_id); + ret = device_property_read_u32(dev, "phy-addr", &phy_addr); if (ret) return -EINVAL; - phy = get_phy_device(mdio, phy_id, false); - if (IS_ERR(phy)) + phy = xgene_enet_phy_register(mdio, phy_addr); + if (!phy) return -EIO; - ret = phy_device_register(phy); - if (ret) - phy_device_free(phy); - else - pdata->phy_dev = phy; + pdata->phy_dev = phy; return ret; } @@ -946,12 +867,12 @@ int xgene_enet_mdio_config(struct xgene_enet_pdata *pdata) return -ENOMEM; mdio_bus->name = "APM X-Gene MDIO bus"; - mdio_bus->read = xgene_enet_mdio_read; - mdio_bus->write = xgene_enet_mdio_write; + mdio_bus->read = xgene_mdio_rgmii_read; + mdio_bus->write = xgene_mdio_rgmii_write; snprintf(mdio_bus->id, MII_BUS_ID_SIZE, "%s-%s", "xgene-mii", ndev->name); - mdio_bus->priv = pdata; + mdio_bus->priv = (void __force *)pdata->mcx_mac_addr; mdio_bus->parent = &pdata->pdev->dev; ret = xgene_mdiobus_register(pdata, mdio_bus); -- cgit v0.10.2 From 52d1fd9983182e9e92507a18a997b3614864e0ee Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:44 -0700 Subject: drivers: net: xgene: ethtool: Use phy_ethtool_gset and sset Changed SGMII 1G get_settings to use phy_ethtool_gset. Changed SGMII 1G set_settings to use phy_ethtool_sset. Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c b/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c index 416d6eb..22a7b26 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c @@ -65,8 +65,15 @@ static int xgene_get_settings(struct net_device *ndev, struct ethtool_cmd *cmd) return phy_ethtool_gset(phydev, cmd); } else if (pdata->phy_mode == PHY_INTERFACE_MODE_SGMII) { - cmd->supported = SUPPORTED_1000baseT_Full | - SUPPORTED_Autoneg | SUPPORTED_MII; + if (pdata->mdio_driver) { + if (!phydev) + return -ENODEV; + + return phy_ethtool_gset(phydev, cmd); + } + + cmd->supported = SUPPORTED_1000baseT_Full | SUPPORTED_Autoneg | + SUPPORTED_MII; cmd->advertising = cmd->supported; ethtool_cmd_speed_set(cmd, SPEED_1000); cmd->duplex = DUPLEX_FULL; @@ -92,12 +99,21 @@ static int xgene_set_settings(struct net_device *ndev, struct ethtool_cmd *cmd) struct phy_device *phydev = pdata->phy_dev; if (pdata->phy_mode == PHY_INTERFACE_MODE_RGMII) { - if (phydev == NULL) + if (!phydev) return -ENODEV; return phy_ethtool_sset(phydev, cmd); } + if (pdata->phy_mode == PHY_INTERFACE_MODE_SGMII) { + if (pdata->mdio_driver) { + if (!phydev) + return -ENODEV; + + return phy_ethtool_sset(phydev, cmd); + } + } + return -EINVAL; } -- cgit v0.10.2 From 8e694cd2762c36b4104bbddaaf72740e59fdee5e Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:45 -0700 Subject: dtb: xgene: Add MDIO node Added mdio node for mdio driver. Also added phy-handle reference to the ethernet nodes. Removed unused clock node from storm sgenet1. Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Signed-off-by: David S. Miller diff --git a/arch/arm64/boot/dts/apm/apm-merlin.dts b/arch/arm64/boot/dts/apm/apm-merlin.dts index 387c6a8..b0f6441 100644 --- a/arch/arm64/boot/dts/apm/apm-merlin.dts +++ b/arch/arm64/boot/dts/apm/apm-merlin.dts @@ -83,3 +83,9 @@ status = "ok"; }; }; + +&mdio { + sgenet0phy: phy@0 { + reg = <0x0>; + }; +}; diff --git a/arch/arm64/boot/dts/apm/apm-mustang.dts b/arch/arm64/boot/dts/apm/apm-mustang.dts index 44db32e..b7fb5d9 100644 --- a/arch/arm64/boot/dts/apm/apm-mustang.dts +++ b/arch/arm64/boot/dts/apm/apm-mustang.dts @@ -79,3 +79,15 @@ &mmc0 { status = "ok"; }; + +&mdio { + menet0phy: phy@3 { + reg = <0x3>; + }; + sgenet0phy: phy@4 { + reg = <0x4>; + }; + sgenet1phy: phy@5 { + reg = <0x5>; + }; +}; diff --git a/arch/arm64/boot/dts/apm/apm-shadowcat.dtsi b/arch/arm64/boot/dts/apm/apm-shadowcat.dtsi index c569f76..2e1e5da 100644 --- a/arch/arm64/boot/dts/apm/apm-shadowcat.dtsi +++ b/arch/arm64/boot/dts/apm/apm-shadowcat.dtsi @@ -625,10 +625,18 @@ apm,irq-start = <8>; }; + mdio: mdio@1f610000 { + compatible = "apm,xgene-mdio-xfi"; + #address-cells = <1>; + #size-cells = <0>; + reg = <0x0 0x1f610000 0x0 0xd100>; + clocks = <&xge0clk 0>; + }; + sgenet0: ethernet@1f610000 { compatible = "apm,xgene2-sgenet"; status = "disabled"; - reg = <0x0 0x1f610000 0x0 0x10000>, + reg = <0x0 0x1f610000 0x0 0xd100>, <0x0 0x1f600000 0x0 0Xd100>, <0x0 0x20000000 0x0 0X20000>; interrupts = <0 96 4>, @@ -637,6 +645,7 @@ clocks = <&xge0clk 0>; local-mac-address = [00 01 73 00 00 01]; phy-connection-type = "sgmii"; + phy-handle = <&sgenet0phy>; }; xgenet1: ethernet@1f620000 { diff --git a/arch/arm64/boot/dts/apm/apm-storm.dtsi b/arch/arm64/boot/dts/apm/apm-storm.dtsi index 5147d76..6bf7cbe 100644 --- a/arch/arm64/boot/dts/apm/apm-storm.dtsi +++ b/arch/arm64/boot/dts/apm/apm-storm.dtsi @@ -237,20 +237,11 @@ clocks = <&socplldiv2 0>; reg = <0x0 0x1f21c000 0x0 0x1000>; reg-names = "csr-reg"; - csr-mask = <0x3>; + csr-mask = <0xa>; + enable-mask = <0xf>; clock-output-names = "sge0clk"; }; - sge1clk: sge1clk@1f21c000 { - compatible = "apm,xgene-device-clock"; - #clock-cells = <1>; - clocks = <&socplldiv2 0>; - reg = <0x0 0x1f21c000 0x0 0x1000>; - reg-names = "csr-reg"; - csr-mask = <0xc>; - clock-output-names = "sge1clk"; - }; - xge0clk: xge0clk@1f61c000 { compatible = "apm,xgene-device-clock"; #clock-cells = <1>; @@ -921,6 +912,14 @@ clocks = <&rtcclk 0>; }; + mdio: mdio@17020000 { + compatible = "apm,xgene-mdio-rgmii"; + #address-cells = <1>; + #size-cells = <0>; + reg = <0x0 0x17020000 0x0 0xd100>; + clocks = <&menetclk 0>; + }; + menet: ethernet@17020000 { compatible = "apm,xgene-enet"; status = "disabled"; @@ -934,7 +933,7 @@ /* mac address will be overwritten by the bootloader */ local-mac-address = [00 00 00 00 00 00]; phy-connection-type = "rgmii"; - phy-handle = <&menetphy>; + phy-handle = <&menet0phy>,<&menetphy>; mdio { compatible = "apm,xgene-mdio"; #address-cells = <1>; @@ -960,6 +959,7 @@ clocks = <&sge0clk 0>; local-mac-address = [00 00 00 00 00 00]; phy-connection-type = "sgmii"; + phy-handle = <&sgenet0phy>; }; sgenet1: ethernet@1f210030 { @@ -973,9 +973,9 @@ <0x0 0xAD 0x4>; port-id = <1>; dma-coherent; - clocks = <&sge1clk 0>; local-mac-address = [00 00 00 00 00 00]; phy-connection-type = "sgmii"; + phy-handle = <&sgenet1phy>; }; xgenet: ethernet@1f610000 { -- cgit v0.10.2 From 792a1107fe713347ed230bc79a53e875b6abe7c7 Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:46 -0700 Subject: Documentation: dtb: xgene: Add MDIO node Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen Tested-by: Toan Le Acked-by: Rob Herring Signed-off-by: David S. Miller diff --git a/Documentation/devicetree/bindings/net/apm-xgene-mdio.txt b/Documentation/devicetree/bindings/net/apm-xgene-mdio.txt new file mode 100644 index 0000000..78722d7 --- /dev/null +++ b/Documentation/devicetree/bindings/net/apm-xgene-mdio.txt @@ -0,0 +1,37 @@ +APM X-Gene SoC MDIO node + +MDIO node is defined to describe on-chip MDIO controller. + +Required properties: + - compatible: Must be "apm,xgene-mdio-rgmii" or "apm,xgene-mdio-xfi" + - #address-cells: Must be <1>. + - #size-cells: Must be <0>. + - reg: Address and length of the register set + - clocks: Reference to the clock entry + +For the phys on the mdio bus, there must be a node with the following fields: + - compatible: PHY identifier. Please refer ./phy.txt for the format. + - reg: The ID number for the phy. + +Example: + + mdio: mdio@17020000 { + compatible = "apm,xgene-mdio-rgmii"; + #address-cells = <1>; + #size-cells = <0>; + reg = <0x0 0x17020000 0x0 0xd100>; + clocks = <&menetclk 0>; + }; + + /* Board-specific peripheral configurations */ + &mdio { + menetphy: phy@3 { + reg = <0x3>; + }; + sgenet0phy: phy@4 { + reg = <0x4>; + }; + sgenet1phy: phy@5 { + reg = <0x5>; + }; + }; -- cgit v0.10.2 From 2efccc601c1c139cb88e3e5d0a6f4402fd040c46 Mon Sep 17 00:00:00 2001 From: Iyappan Subramanian Date: Mon, 25 Jul 2016 17:12:47 -0700 Subject: MAINTAINERS: xgene: Add driver and documentation path Added path to the MDIO driver and Documentation file. Signed-off-by: Iyappan Subramanian Signed-off-by: David S. Miller diff --git a/MAINTAINERS b/MAINTAINERS index dc3481d..a9bdba0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -839,7 +839,9 @@ M: Iyappan Subramanian M: Keyur Chudgar S: Supported F: drivers/net/ethernet/apm/xgene/ +F: drivers/net/phy/mdio-xgene.c F: Documentation/devicetree/bindings/net/apm-xgene-enet.txt +F: Documentation/devicetree/bindings/net/apm-xgene-mdio.txt APTINA CAMERA SENSOR PLL M: Laurent Pinchart -- cgit v0.10.2 From d1c2b5010d07e967d7cbcc232a86b2308d824ca3 Mon Sep 17 00:00:00 2001 From: He Chunhui Date: Tue, 26 Jul 2016 06:16:52 +0000 Subject: net: neigh: disallow transition to NUD_STALE if lladdr is unchanged in neigh_update() NUD_STALE is used when the caller(e.g. arp_process()) can't guarantee neighbour reachability. If the entry was NUD_VALID and lladdr is unchanged, the entry state should not be changed. Currently the code puts an extra "NUD_CONNECTED" condition. So if old state was NUD_DELAY or NUD_PROBE (they are NUD_VALID but not NUD_CONNECTED), the state can be changed to NUD_STALE. This may cause problem. Because NUD_STALE lladdr doesn't guarantee reachability, when we send traffic, the state will be changed to NUD_DELAY. In normal case, if we get no confirmation (by dst_confirm()), we will change the state to NUD_PROBE and send probe traffic. But now the state may be reset to NUD_STALE again(e.g. by broadcast ARP packets), so the probe traffic will not be sent. This situation may happen again and again, and packets will be sent to an non-reachable lladdr forever. The fix is to remove the "NUD_CONNECTED" condition. After that the "NEIGH_UPDATE_F_WEAK_OVERRIDE" condition (used by IPv6) in that branch will be redundant, so remove it. This change may increase probe traffic, but it's essential since NUD_STALE lladdr is unreliable. To ensure correctness, we prefer to resolve lladdr, when we can't get confirmation, even while remote packets try to set NUD_STALE state. Signed-off-by: Chunhui He Signed-off-by: Julian Anastasov Reviewed-by: Hannes Frederic Sowa Signed-off-by: David S. Miller diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 5cdc62a..cf26e04c4 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1060,8 +1060,6 @@ static void neigh_update_hhs(struct neighbour *neigh) NEIGH_UPDATE_F_WEAK_OVERRIDE will suspect existing "connected" lladdr instead of overriding it if it is different. - It also allows to retain current state - if lladdr is unchanged. NEIGH_UPDATE_F_ADMIN means that the change is administrative. NEIGH_UPDATE_F_OVERRIDE_ISROUTER allows to override existing @@ -1150,10 +1148,7 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new, } else goto out; } else { - if (lladdr == neigh->ha && new == NUD_STALE && - ((flags & NEIGH_UPDATE_F_WEAK_OVERRIDE) || - (old & NUD_CONNECTED)) - ) + if (lladdr == neigh->ha && new == NUD_STALE) new = old; } } -- cgit v0.10.2 From 9ff26e9fabaf52f28fb5e875c0b9ffc2d1512039 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 26 Jul 2016 08:47:18 +0200 Subject: tipc: introduce constants for tipc address validation In this commit, we introduce defines for tipc address size, offset and mask specification for Zone.Cluster.Node. There is no functional change in this commit. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller diff --git a/include/uapi/linux/tipc.h b/include/uapi/linux/tipc.h index 6f71b9b..bf049e8 100644 --- a/include/uapi/linux/tipc.h +++ b/include/uapi/linux/tipc.h @@ -60,26 +60,48 @@ struct tipc_name_seq { __u32 upper; }; +/* TIPC Address Size, Offset, Mask specification for Z.C.N + */ +#define TIPC_NODE_BITS 12 +#define TIPC_CLUSTER_BITS 12 +#define TIPC_ZONE_BITS 8 + +#define TIPC_NODE_OFFSET 0 +#define TIPC_CLUSTER_OFFSET TIPC_NODE_BITS +#define TIPC_ZONE_OFFSET (TIPC_CLUSTER_OFFSET + TIPC_CLUSTER_BITS) + +#define TIPC_NODE_SIZE ((1UL << TIPC_NODE_BITS) - 1) +#define TIPC_CLUSTER_SIZE ((1UL << TIPC_CLUSTER_BITS) - 1) +#define TIPC_ZONE_SIZE ((1UL << TIPC_ZONE_BITS) - 1) + +#define TIPC_NODE_MASK (TIPC_NODE_SIZE << TIPC_NODE_OFFSET) +#define TIPC_CLUSTER_MASK (TIPC_CLUSTER_SIZE << TIPC_CLUSTER_OFFSET) +#define TIPC_ZONE_MASK (TIPC_ZONE_SIZE << TIPC_ZONE_OFFSET) + +#define TIPC_ZONE_CLUSTER_MASK (TIPC_ZONE_MASK | TIPC_CLUSTER_MASK) + static inline __u32 tipc_addr(unsigned int zone, unsigned int cluster, unsigned int node) { - return (zone << 24) | (cluster << 12) | node; + return (zone << TIPC_ZONE_OFFSET) | + (cluster << TIPC_CLUSTER_OFFSET) | + node; } static inline unsigned int tipc_zone(__u32 addr) { - return addr >> 24; + return addr >> TIPC_ZONE_OFFSET; } static inline unsigned int tipc_cluster(__u32 addr) { - return (addr >> 12) & 0xfff; + return (addr & TIPC_CLUSTER_MASK) >> TIPC_CLUSTER_OFFSET; } static inline unsigned int tipc_node(__u32 addr) { - return addr & 0xfff; + return addr & TIPC_NODE_MASK; } /* diff --git a/net/tipc/addr.h b/net/tipc/addr.h index 64f4004..bebb347 100644 --- a/net/tipc/addr.h +++ b/net/tipc/addr.h @@ -43,9 +43,6 @@ #include #include "core.h" -#define TIPC_ZONE_MASK 0xff000000u -#define TIPC_CLUSTER_MASK 0xfffff000u - static inline u32 tipc_own_addr(struct net *net) { struct tipc_net *tn = net_generic(net, tipc_net_id); @@ -60,7 +57,7 @@ static inline u32 tipc_zone_mask(u32 addr) static inline u32 tipc_cluster_mask(u32 addr) { - return addr & TIPC_CLUSTER_MASK; + return addr & TIPC_ZONE_CLUSTER_MASK; } u32 tipc_own_addr(struct net *net); diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 4131d5a..65b0998 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -225,7 +225,7 @@ static int tipc_enable_bearer(struct net *net, const char *name, if (tipc_addr_domain_valid(disc_domain) && (disc_domain != tn->own_addr)) { if (tipc_in_scope(disc_domain, tn->own_addr)) { - disc_domain = tn->own_addr & TIPC_CLUSTER_MASK; + disc_domain = tn->own_addr & TIPC_ZONE_CLUSTER_MASK; res = 0; /* accept any node in own cluster */ } else if (in_own_cluster_exact(net, disc_domain)) res = 0; /* accept specified node in own cluster */ @@ -832,7 +832,7 @@ int tipc_nl_bearer_enable(struct sk_buff *skb, struct genl_info *info) u32 prio; prio = TIPC_MEDIA_LINK_PRI; - domain = tn->own_addr & TIPC_CLUSTER_MASK; + domain = tn->own_addr & TIPC_ZONE_CLUSTER_MASK; if (!info->attrs[TIPC_NLA_BEARER]) return -EINVAL; -- cgit v0.10.2 From 7b3f52296493656015f0c0deddb6e90e36b9cda2 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 26 Jul 2016 08:47:19 +0200 Subject: tipc: make cluster size threshold for monitoring configurable In this commit, we introduce support to configure the minimum threshold to activate the new link monitoring algorithm. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller diff --git a/include/uapi/linux/tipc_netlink.h b/include/uapi/linux/tipc_netlink.h index d4c8f14..d387b65 100644 --- a/include/uapi/linux/tipc_netlink.h +++ b/include/uapi/linux/tipc_netlink.h @@ -56,6 +56,7 @@ enum { TIPC_NL_NET_GET, TIPC_NL_NET_SET, TIPC_NL_NAME_TABLE_GET, + TIPC_NL_MON_SET, __TIPC_NL_CMD_MAX, TIPC_NL_CMD_MAX = __TIPC_NL_CMD_MAX - 1 @@ -72,6 +73,7 @@ enum { TIPC_NLA_NODE, /* nest */ TIPC_NLA_NET, /* nest */ TIPC_NLA_NAME_TABLE, /* nest */ + TIPC_NLA_MON, /* nest */ __TIPC_NLA_MAX, TIPC_NLA_MAX = __TIPC_NLA_MAX - 1 @@ -166,6 +168,15 @@ enum { TIPC_NLA_NAME_TABLE_MAX = __TIPC_NLA_NAME_TABLE_MAX - 1 }; +/* Monitor info */ +enum { + TIPC_NLA_MON_UNSPEC, + TIPC_NLA_MON_ACTIVATION_THRESHOLD, /* u32 */ + + __TIPC_NLA_MON_MAX, + TIPC_NLA_MON_MAX = __TIPC_NLA_MON_MAX - 1 +}; + /* Publication info */ enum { TIPC_NLA_PUBL_UNSPEC, diff --git a/net/tipc/monitor.c b/net/tipc/monitor.c index 0d489e8..3892d05 100644 --- a/net/tipc/monitor.c +++ b/net/tipc/monitor.c @@ -649,3 +649,15 @@ void tipc_mon_delete(struct net *net, int bearer_id) kfree(self); kfree(mon); } + +int tipc_nl_monitor_set_threshold(struct net *net, u32 cluster_size) +{ + struct tipc_net *tn = tipc_net(net); + + if (cluster_size > TIPC_CLUSTER_SIZE) + return -EINVAL; + + tn->mon_threshold = cluster_size; + + return 0; +} diff --git a/net/tipc/monitor.h b/net/tipc/monitor.h index 598459c..91f5dd09 100644 --- a/net/tipc/monitor.h +++ b/net/tipc/monitor.h @@ -69,5 +69,6 @@ void tipc_mon_get_state(struct net *net, u32 addr, int bearer_id); void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id); +int tipc_nl_monitor_set_threshold(struct net *net, u32 cluster_size); extern const int tipc_max_domain_size; #endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 56935df2..1e43ac0 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -52,7 +52,8 @@ static const struct nla_policy tipc_nl_policy[TIPC_NLA_MAX + 1] = { [TIPC_NLA_MEDIA] = { .type = NLA_NESTED, }, [TIPC_NLA_NODE] = { .type = NLA_NESTED, }, [TIPC_NLA_NET] = { .type = NLA_NESTED, }, - [TIPC_NLA_NAME_TABLE] = { .type = NLA_NESTED, } + [TIPC_NLA_NAME_TABLE] = { .type = NLA_NESTED, }, + [TIPC_NLA_MON] = { .type = NLA_NESTED, }, }; const struct nla_policy @@ -61,6 +62,11 @@ tipc_nl_name_table_policy[TIPC_NLA_NAME_TABLE_MAX + 1] = { [TIPC_NLA_NAME_TABLE_PUBL] = { .type = NLA_NESTED } }; +const struct nla_policy tipc_nl_monitor_policy[TIPC_NLA_MON_MAX + 1] = { + [TIPC_NLA_MON_UNSPEC] = { .type = NLA_UNSPEC }, + [TIPC_NLA_MON_ACTIVATION_THRESHOLD] = { .type = NLA_U32 }, +}; + const struct nla_policy tipc_nl_sock_policy[TIPC_NLA_SOCK_MAX + 1] = { [TIPC_NLA_SOCK_UNSPEC] = { .type = NLA_UNSPEC }, [TIPC_NLA_SOCK_ADDR] = { .type = NLA_U32 }, @@ -214,7 +220,12 @@ static const struct genl_ops tipc_genl_v2_ops[] = { .cmd = TIPC_NL_NAME_TABLE_GET, .dumpit = tipc_nl_name_table_dump, .policy = tipc_nl_policy, - } + }, + { + .cmd = TIPC_NL_MON_SET, + .doit = tipc_nl_node_set_monitor, + .policy = tipc_nl_policy, + }, }; int tipc_nlmsg_parse(const struct nlmsghdr *nlh, struct nlattr ***attr) diff --git a/net/tipc/netlink.h b/net/tipc/netlink.h index ed1dbcb..4ba0ad4 100644 --- a/net/tipc/netlink.h +++ b/net/tipc/netlink.h @@ -55,6 +55,7 @@ extern const struct nla_policy tipc_nl_prop_policy[]; extern const struct nla_policy tipc_nl_bearer_policy[]; extern const struct nla_policy tipc_nl_media_policy[]; extern const struct nla_policy tipc_nl_udp_policy[]; +extern const struct nla_policy tipc_nl_monitor_policy[]; int tipc_netlink_start(void); int tipc_netlink_compat_start(void); diff --git a/net/tipc/node.c b/net/tipc/node.c index 95cc78b..0fc531d 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1928,3 +1928,30 @@ out: return skb->len; } + +int tipc_nl_node_set_monitor(struct sk_buff *skb, struct genl_info *info) +{ + struct nlattr *attrs[TIPC_NLA_MON_MAX + 1]; + struct net *net = sock_net(skb->sk); + int err; + + if (!info->attrs[TIPC_NLA_MON]) + return -EINVAL; + + err = nla_parse_nested(attrs, TIPC_NLA_MON_MAX, + info->attrs[TIPC_NLA_MON], + tipc_nl_monitor_policy); + if (err) + return err; + + if (attrs[TIPC_NLA_MON_ACTIVATION_THRESHOLD]) { + u32 val; + + val = nla_get_u32(attrs[TIPC_NLA_MON_ACTIVATION_THRESHOLD]); + err = tipc_nl_monitor_set_threshold(net, val); + if (err) + return err; + } + + return 0; +} diff --git a/net/tipc/node.h b/net/tipc/node.h index 8264b3d..65aa12ed 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -78,4 +78,5 @@ int tipc_nl_node_reset_link_stats(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_get_link(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info); +int tipc_nl_node_set_monitor(struct sk_buff *skb, struct genl_info *info); #endif -- cgit v0.10.2 From bf1035b2ff5296c7c49e262152253ce29d87e82d Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 26 Jul 2016 08:47:20 +0200 Subject: tipc: get monitor threshold for the cluster In this commit, we add support to fetch the configured cluster monitoring threshold. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller diff --git a/include/uapi/linux/tipc_netlink.h b/include/uapi/linux/tipc_netlink.h index d387b65..d07c6ec 100644 --- a/include/uapi/linux/tipc_netlink.h +++ b/include/uapi/linux/tipc_netlink.h @@ -57,6 +57,7 @@ enum { TIPC_NL_NET_SET, TIPC_NL_NAME_TABLE_GET, TIPC_NL_MON_SET, + TIPC_NL_MON_GET, __TIPC_NL_CMD_MAX, TIPC_NL_CMD_MAX = __TIPC_NL_CMD_MAX - 1 diff --git a/net/tipc/monitor.c b/net/tipc/monitor.c index 3892d05..3579126 100644 --- a/net/tipc/monitor.c +++ b/net/tipc/monitor.c @@ -661,3 +661,10 @@ int tipc_nl_monitor_set_threshold(struct net *net, u32 cluster_size) return 0; } + +int tipc_nl_monitor_get_threshold(struct net *net) +{ + struct tipc_net *tn = tipc_net(net); + + return tn->mon_threshold; +} diff --git a/net/tipc/monitor.h b/net/tipc/monitor.h index 91f5dd09..aedf62c 100644 --- a/net/tipc/monitor.h +++ b/net/tipc/monitor.h @@ -70,5 +70,7 @@ void tipc_mon_get_state(struct net *net, u32 addr, void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id); int tipc_nl_monitor_set_threshold(struct net *net, u32 cluster_size); +int tipc_nl_monitor_get_threshold(struct net *net); + extern const int tipc_max_domain_size; #endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 1e43ac0..2cfc5f7c 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -226,6 +226,11 @@ static const struct genl_ops tipc_genl_v2_ops[] = { .doit = tipc_nl_node_set_monitor, .policy = tipc_nl_policy, }, + { + .cmd = TIPC_NL_MON_GET, + .doit = tipc_nl_node_get_monitor, + .policy = tipc_nl_policy, + }, }; int tipc_nlmsg_parse(const struct nlmsghdr *nlh, struct nlattr ***attr) diff --git a/net/tipc/node.c b/net/tipc/node.c index 0fc531d..2a7e747 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1955,3 +1955,55 @@ int tipc_nl_node_set_monitor(struct sk_buff *skb, struct genl_info *info) return 0; } + +static int __tipc_nl_add_monitor_prop(struct net *net, struct tipc_nl_msg *msg) +{ + struct nlattr *attrs; + void *hdr; + u32 val; + + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, + 0, TIPC_NL_MON_GET); + if (!hdr) + return -EMSGSIZE; + + attrs = nla_nest_start(msg->skb, TIPC_NLA_MON); + if (!attrs) + goto msg_full; + + val = tipc_nl_monitor_get_threshold(net); + + if (nla_put_u32(msg->skb, TIPC_NLA_MON_ACTIVATION_THRESHOLD, val)) + goto attr_msg_full; + + nla_nest_end(msg->skb, attrs); + genlmsg_end(msg->skb, hdr); + + return 0; + +attr_msg_full: + nla_nest_cancel(msg->skb, attrs); +msg_full: + genlmsg_cancel(msg->skb, hdr); + + return -EMSGSIZE; +} + +int tipc_nl_node_get_monitor(struct sk_buff *skb, struct genl_info *info) +{ + struct net *net = sock_net(skb->sk); + struct tipc_nl_msg msg; + int err; + + msg.skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + msg.portid = info->snd_portid; + msg.seq = info->snd_seq; + + err = __tipc_nl_add_monitor_prop(net, &msg); + if (err) { + nlmsg_free(msg.skb); + return err; + } + + return genlmsg_reply(msg.skb, info); +} diff --git a/net/tipc/node.h b/net/tipc/node.h index 65aa12ed..216f053 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -79,4 +79,5 @@ int tipc_nl_node_get_link(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_set_monitor(struct sk_buff *skb, struct genl_info *info); +int tipc_nl_node_get_monitor(struct sk_buff *skb, struct genl_info *info); #endif -- cgit v0.10.2 From ff0d3e78a67a8edd09688f073361de9ed8abf9dc Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 26 Jul 2016 08:47:21 +0200 Subject: tipc: add a function to get the bearer name Introduce a new function to get the bearer name from its id. This is used in subsequent commit. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 65b0998..65b1bbf 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -171,6 +171,27 @@ struct tipc_bearer *tipc_bearer_find(struct net *net, const char *name) return NULL; } +/* tipc_bearer_get_name - get the bearer name from its id. + * @net: network namespace + * @name: a pointer to the buffer where the name will be stored. + * @bearer_id: the id to get the name from. + */ +int tipc_bearer_get_name(struct net *net, char *name, u32 bearer_id) +{ + struct tipc_net *tn = tipc_net(net); + struct tipc_bearer *b; + + if (bearer_id >= MAX_BEARERS) + return -EINVAL; + + b = rtnl_dereference(tn->bearer_list[bearer_id]); + if (!b) + return -EINVAL; + + strcpy(name, b->name); + return 0; +} + void tipc_bearer_add_dest(struct net *net, u32 bearer_id, u32 dest) { struct tipc_net *tn = net_generic(net, tipc_net_id); diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index f1e6db5..43757f1 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -197,6 +197,7 @@ int tipc_l2_send_msg(struct net *net, struct sk_buff *buf, void tipc_bearer_add_dest(struct net *net, u32 bearer_id, u32 dest); void tipc_bearer_remove_dest(struct net *net, u32 bearer_id, u32 dest); struct tipc_bearer *tipc_bearer_find(struct net *net, const char *name); +int tipc_bearer_get_name(struct net *net, char *name, u32 bearer_id); struct tipc_media *tipc_media_find(const char *name); void tipc_bearer_reset_all(struct net *net); int tipc_bearer_setup(void); -- cgit v0.10.2 From cf6f7e1d51090772d5ff7355aaf0fcff17f20d1a Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 26 Jul 2016 08:47:22 +0200 Subject: tipc: dump monitor attributes In this commit, we dump the monitor attributes when queried. The link monitor attributes are separated into two kinds: 1. general attributes per bearer 2. specific attributes per node/peer This style resembles the socket attributes and the nametable publications per socket. Reviewed-by: Jon Maloy Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: David S. Miller diff --git a/include/uapi/linux/tipc_netlink.h b/include/uapi/linux/tipc_netlink.h index d07c6ec..5f3f6d0 100644 --- a/include/uapi/linux/tipc_netlink.h +++ b/include/uapi/linux/tipc_netlink.h @@ -58,6 +58,7 @@ enum { TIPC_NL_NAME_TABLE_GET, TIPC_NL_MON_SET, TIPC_NL_MON_GET, + TIPC_NL_MON_PEER_GET, __TIPC_NL_CMD_MAX, TIPC_NL_CMD_MAX = __TIPC_NL_CMD_MAX - 1 @@ -75,6 +76,7 @@ enum { TIPC_NLA_NET, /* nest */ TIPC_NLA_NAME_TABLE, /* nest */ TIPC_NLA_MON, /* nest */ + TIPC_NLA_MON_PEER, /* nest */ __TIPC_NLA_MAX, TIPC_NLA_MAX = __TIPC_NLA_MAX - 1 @@ -173,6 +175,11 @@ enum { enum { TIPC_NLA_MON_UNSPEC, TIPC_NLA_MON_ACTIVATION_THRESHOLD, /* u32 */ + TIPC_NLA_MON_REF, /* u32 */ + TIPC_NLA_MON_ACTIVE, /* flag */ + TIPC_NLA_MON_BEARER_NAME, /* string */ + TIPC_NLA_MON_PEERCNT, /* u32 */ + TIPC_NLA_MON_LISTGEN, /* u32 */ __TIPC_NLA_MON_MAX, TIPC_NLA_MON_MAX = __TIPC_NLA_MON_MAX - 1 @@ -194,6 +201,24 @@ enum { TIPC_NLA_PUBL_MAX = __TIPC_NLA_PUBL_MAX - 1 }; +/* Monitor peer info */ +enum { + TIPC_NLA_MON_PEER_UNSPEC, + + TIPC_NLA_MON_PEER_ADDR, /* u32 */ + TIPC_NLA_MON_PEER_DOMGEN, /* u32 */ + TIPC_NLA_MON_PEER_APPLIED, /* u32 */ + TIPC_NLA_MON_PEER_UPMAP, /* u64 */ + TIPC_NLA_MON_PEER_MEMBERS, /* tlv */ + TIPC_NLA_MON_PEER_UP, /* flag */ + TIPC_NLA_MON_PEER_HEAD, /* flag */ + TIPC_NLA_MON_PEER_LOCAL, /* flag */ + TIPC_NLA_MON_PEER_PAD, /* flag */ + + __TIPC_NLA_MON_PEER_MAX, + TIPC_NLA_MON_PEER_MAX = __TIPC_NLA_MON_PEER_MAX - 1 +}; + /* Nest, connection info */ enum { TIPC_NLA_CON_UNSPEC, diff --git a/net/tipc/monitor.c b/net/tipc/monitor.c index 3579126..be70a57 100644 --- a/net/tipc/monitor.c +++ b/net/tipc/monitor.c @@ -33,9 +33,11 @@ * POSSIBILITY OF SUCH DAMAGE. */ +#include #include "core.h" #include "addr.h" #include "monitor.h" +#include "bearer.h" #define MAX_MON_DOMAIN 64 #define MON_TIMEOUT 120000 @@ -668,3 +670,134 @@ int tipc_nl_monitor_get_threshold(struct net *net) return tn->mon_threshold; } + +int __tipc_nl_add_monitor_peer(struct tipc_peer *peer, struct tipc_nl_msg *msg) +{ + struct tipc_mon_domain *dom = peer->domain; + struct nlattr *attrs; + void *hdr; + + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, + NLM_F_MULTI, TIPC_NL_MON_PEER_GET); + if (!hdr) + return -EMSGSIZE; + + attrs = nla_nest_start(msg->skb, TIPC_NLA_MON_PEER); + if (!attrs) + goto msg_full; + + if (nla_put_u32(msg->skb, TIPC_NLA_MON_PEER_ADDR, peer->addr)) + goto attr_msg_full; + if (nla_put_u32(msg->skb, TIPC_NLA_MON_PEER_APPLIED, peer->applied)) + goto attr_msg_full; + + if (peer->is_up) + if (nla_put_flag(msg->skb, TIPC_NLA_MON_PEER_UP)) + goto attr_msg_full; + if (peer->is_local) + if (nla_put_flag(msg->skb, TIPC_NLA_MON_PEER_LOCAL)) + goto attr_msg_full; + if (peer->is_head) + if (nla_put_flag(msg->skb, TIPC_NLA_MON_PEER_HEAD)) + goto attr_msg_full; + + if (dom) { + if (nla_put_u32(msg->skb, TIPC_NLA_MON_PEER_DOMGEN, dom->gen)) + goto attr_msg_full; + if (nla_put_u64_64bit(msg->skb, TIPC_NLA_MON_PEER_UPMAP, + dom->up_map, TIPC_NLA_MON_PEER_PAD)) + goto attr_msg_full; + if (nla_put(msg->skb, TIPC_NLA_MON_PEER_MEMBERS, + dom->member_cnt * sizeof(u32), &dom->members)) + goto attr_msg_full; + } + + nla_nest_end(msg->skb, attrs); + genlmsg_end(msg->skb, hdr); + return 0; + +attr_msg_full: + nla_nest_cancel(msg->skb, attrs); +msg_full: + genlmsg_cancel(msg->skb, hdr); + + return -EMSGSIZE; +} + +int tipc_nl_add_monitor_peer(struct net *net, struct tipc_nl_msg *msg, + u32 bearer_id, u32 *prev_node) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + struct tipc_peer *peer = mon->self; + + if (!mon) + return -EINVAL; + + read_lock_bh(&mon->lock); + do { + if (*prev_node) { + if (peer->addr == *prev_node) + *prev_node = 0; + else + continue; + } + if (__tipc_nl_add_monitor_peer(peer, msg)) { + *prev_node = peer->addr; + read_unlock_bh(&mon->lock); + return -EMSGSIZE; + } + } while ((peer = peer_nxt(peer)) != mon->self); + read_unlock_bh(&mon->lock); + + return 0; +} + +int __tipc_nl_add_monitor(struct net *net, struct tipc_nl_msg *msg, + u32 bearer_id) +{ + struct tipc_monitor *mon = tipc_monitor(net, bearer_id); + char bearer_name[TIPC_MAX_BEARER_NAME]; + struct nlattr *attrs; + void *hdr; + int ret; + + ret = tipc_bearer_get_name(net, bearer_name, bearer_id); + if (ret || !mon) + return -EINVAL; + + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, + NLM_F_MULTI, TIPC_NL_MON_GET); + if (!hdr) + return -EMSGSIZE; + + attrs = nla_nest_start(msg->skb, TIPC_NLA_MON); + if (!attrs) + goto msg_full; + + read_lock_bh(&mon->lock); + if (nla_put_u32(msg->skb, TIPC_NLA_MON_REF, bearer_id)) + goto attr_msg_full; + if (tipc_mon_is_active(net, mon)) + if (nla_put_flag(msg->skb, TIPC_NLA_MON_ACTIVE)) + goto attr_msg_full; + if (nla_put_string(msg->skb, TIPC_NLA_MON_BEARER_NAME, bearer_name)) + goto attr_msg_full; + if (nla_put_u32(msg->skb, TIPC_NLA_MON_PEERCNT, mon->peer_cnt)) + goto attr_msg_full; + if (nla_put_u32(msg->skb, TIPC_NLA_MON_LISTGEN, mon->list_gen)) + goto attr_msg_full; + + read_unlock_bh(&mon->lock); + nla_nest_end(msg->skb, attrs); + genlmsg_end(msg->skb, hdr); + + return 0; + +attr_msg_full: + nla_nest_cancel(msg->skb, attrs); +msg_full: + genlmsg_cancel(msg->skb, hdr); + read_unlock_bh(&mon->lock); + + return -EMSGSIZE; +} diff --git a/net/tipc/monitor.h b/net/tipc/monitor.h index aedf62c..2a21b93 100644 --- a/net/tipc/monitor.h +++ b/net/tipc/monitor.h @@ -36,6 +36,8 @@ #ifndef _TIPC_MONITOR_H #define _TIPC_MONITOR_H +#include "netlink.h" + /* struct tipc_mon_state: link instance's cache of monitor list and domain state * @list_gen: current generation of this node's monitor list * @gen: current generation of this node's local domain @@ -71,6 +73,10 @@ void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id); int tipc_nl_monitor_set_threshold(struct net *net, u32 cluster_size); int tipc_nl_monitor_get_threshold(struct net *net); +int __tipc_nl_add_monitor(struct net *net, struct tipc_nl_msg *msg, + u32 bearer_id); +int tipc_nl_add_monitor_peer(struct net *net, struct tipc_nl_msg *msg, + u32 bearer_id, u32 *prev_node); extern const int tipc_max_domain_size; #endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 2cfc5f7c..a84daec 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -64,6 +64,7 @@ tipc_nl_name_table_policy[TIPC_NLA_NAME_TABLE_MAX + 1] = { const struct nla_policy tipc_nl_monitor_policy[TIPC_NLA_MON_MAX + 1] = { [TIPC_NLA_MON_UNSPEC] = { .type = NLA_UNSPEC }, + [TIPC_NLA_MON_REF] = { .type = NLA_U32 }, [TIPC_NLA_MON_ACTIVATION_THRESHOLD] = { .type = NLA_U32 }, }; @@ -229,6 +230,12 @@ static const struct genl_ops tipc_genl_v2_ops[] = { { .cmd = TIPC_NL_MON_GET, .doit = tipc_nl_node_get_monitor, + .dumpit = tipc_nl_node_dump_monitor, + .policy = tipc_nl_policy, + }, + { + .cmd = TIPC_NL_MON_PEER_GET, + .dumpit = tipc_nl_node_dump_monitor_peer, .policy = tipc_nl_policy, }, }; diff --git a/net/tipc/node.c b/net/tipc/node.c index 2a7e747..2197419 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -2007,3 +2007,89 @@ int tipc_nl_node_get_monitor(struct sk_buff *skb, struct genl_info *info) return genlmsg_reply(msg.skb, info); } + +int tipc_nl_node_dump_monitor(struct sk_buff *skb, struct netlink_callback *cb) +{ + struct net *net = sock_net(skb->sk); + u32 prev_bearer = cb->args[0]; + struct tipc_nl_msg msg; + int err; + int i; + + if (prev_bearer == MAX_BEARERS) + return 0; + + msg.skb = skb; + msg.portid = NETLINK_CB(cb->skb).portid; + msg.seq = cb->nlh->nlmsg_seq; + + rtnl_lock(); + for (i = prev_bearer; i < MAX_BEARERS; i++) { + prev_bearer = i; + err = __tipc_nl_add_monitor(net, &msg, prev_bearer); + if (err) + goto out; + } + +out: + rtnl_unlock(); + cb->args[0] = prev_bearer; + + return skb->len; +} + +int tipc_nl_node_dump_monitor_peer(struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct net *net = sock_net(skb->sk); + u32 prev_node = cb->args[1]; + u32 bearer_id = cb->args[2]; + int done = cb->args[0]; + struct tipc_nl_msg msg; + int err; + + if (!prev_node) { + struct nlattr **attrs; + struct nlattr *mon[TIPC_NLA_MON_MAX + 1]; + + err = tipc_nlmsg_parse(cb->nlh, &attrs); + if (err) + return err; + + if (!attrs[TIPC_NLA_MON]) + return -EINVAL; + + err = nla_parse_nested(mon, TIPC_NLA_MON_MAX, + attrs[TIPC_NLA_MON], + tipc_nl_monitor_policy); + if (err) + return err; + + if (!mon[TIPC_NLA_MON_REF]) + return -EINVAL; + + bearer_id = nla_get_u32(mon[TIPC_NLA_MON_REF]); + + if (bearer_id >= MAX_BEARERS) + return -EINVAL; + } + + if (done) + return 0; + + msg.skb = skb; + msg.portid = NETLINK_CB(cb->skb).portid; + msg.seq = cb->nlh->nlmsg_seq; + + rtnl_lock(); + err = tipc_nl_add_monitor_peer(net, &msg, bearer_id, &prev_node); + if (!err) + done = 1; + + rtnl_unlock(); + cb->args[0] = done; + cb->args[1] = prev_node; + cb->args[2] = bearer_id; + + return skb->len; +} diff --git a/net/tipc/node.h b/net/tipc/node.h index 216f053..d69fdfc 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -80,4 +80,7 @@ int tipc_nl_node_set_link(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_set_monitor(struct sk_buff *skb, struct genl_info *info); int tipc_nl_node_get_monitor(struct sk_buff *skb, struct genl_info *info); +int tipc_nl_node_dump_monitor(struct sk_buff *skb, struct netlink_callback *cb); +int tipc_nl_node_dump_monitor_peer(struct sk_buff *skb, + struct netlink_callback *cb); #endif -- cgit v0.10.2 From e3a3b626010a14fe067f163c2c43409d5afcd2a9 Mon Sep 17 00:00:00 2001 From: Beniamino Galvani Date: Tue, 26 Jul 2016 12:24:53 +0200 Subject: macsec: ensure rx_sa is set when validation is disabled macsec_decrypt() is not called when validation is disabled and so macsec_skb_cb(skb)->rx_sa is not set; but it is used later in macsec_post_decrypt(), ensure that it's always initialized. Fixes: c09440f7dcb3 ("macsec: introduce IEEE 802.1AE driver") Signed-off-by: Beniamino Galvani Acked-by: Sabrina Dubroca Signed-off-by: David S. Miller diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index d8b2b49..2d0beb1 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -944,7 +944,6 @@ static struct sk_buff *macsec_decrypt(struct sk_buff *skb, } macsec_skb_cb(skb)->req = req; - macsec_skb_cb(skb)->rx_sa = rx_sa; skb->dev = dev; aead_request_set_callback(req, 0, macsec_decrypt_done, skb); @@ -1172,6 +1171,8 @@ static rx_handler_result_t macsec_handle_frame(struct sk_buff **pskb) } } + macsec_skb_cb(skb)->rx_sa = rx_sa; + /* Disabled && !changed text => skip validation */ if (hdr->tci_an & MACSEC_TCI_C || secy->validate_frames != MACSEC_VALIDATE_DISABLED) -- cgit v0.10.2 From 90b5ca1766ae7806a711d66df056af1290faa2c0 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Tue, 26 Jul 2016 18:54:52 +0200 Subject: net: ipmr/ip6mr: update lastuse on entry change Currently lastuse is updated on entry creation and cache hit, but it should also be updated on entry change. Since both on add and update the ttl array is updated we can simply update the lastuse in ipmr_update_thresholds. Signed-off-by: Nikolay Aleksandrov CC: Roopa Prabhu CC: Donald Sharp CC: David S. Miller Signed-off-by: David S. Miller diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index eec2341..2625332 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -722,6 +722,7 @@ static void ipmr_update_thresholds(struct mr_table *mrt, struct mfc_cache *cache cache->mfc_un.res.maxvif = vifi + 1; } } + cache->mfc_un.res.lastuse = jiffies; } static int vif_add(struct net *net, struct mr_table *mrt, @@ -1150,7 +1151,6 @@ static int ipmr_mfc_add(struct net *net, struct mr_table *mrt, c->mfc_origin = mfc->mfcc_origin.s_addr; c->mfc_mcastgrp = mfc->mfcc_mcastgrp.s_addr; c->mfc_parent = mfc->mfcc_parent; - c->mfc_un.res.lastuse = jiffies; ipmr_update_thresholds(mrt, c, mfc->mfcc_ttls); if (!mrtsock) c->mfc_flags |= MFC_STATIC; diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 7adce13..6122f9c 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -921,6 +921,7 @@ static void ip6mr_update_thresholds(struct mr6_table *mrt, struct mfc6_cache *ca cache->mfc_un.res.maxvif = vifi + 1; } } + cache->mfc_un.res.lastuse = jiffies; } static int mif6_add(struct net *net, struct mr6_table *mrt, @@ -1500,7 +1501,6 @@ static int ip6mr_mfc_add(struct net *net, struct mr6_table *mrt, c->mf6c_origin = mfc->mf6cc_origin.sin6_addr; c->mf6c_mcastgrp = mfc->mf6cc_mcastgrp.sin6_addr; c->mf6c_parent = mfc->mf6cc_parent; - c->mfc_un.res.lastuse = jiffies; ip6mr_update_thresholds(mrt, c, ttls); if (!mrtsock) c->mfc_flags |= MFC_STATIC; -- cgit v0.10.2 From 0a91605cda883f7a25b1e7fa8cdc3f633ae8f8e0 Mon Sep 17 00:00:00 2001 From: Bhaktipriya Shridhar Date: Tue, 26 Jul 2016 22:38:24 +0530 Subject: net/mlx5_core/health: Remove deprecated create_singlethread_workqueue The workqueue health->wq was used as per device private health thread. This was done to perform delayed work. The workqueue has a single workitem(&health->work) and hence doesn't require ordering. It is involved in handling the health of the device and is not being used on a memory reclaim path. Hence, the singlethreaded workqueue has been replaced with the use of system_wq. Work item has been flushed in mlx5_health_cleanup() to ensure that there are no pending tasks while disconnecting the driver. Signed-off-by: Bhaktipriya Shridhar Acked-by: Leon Romanovsky Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c index 96a5946..1a05fb9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/health.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c @@ -272,7 +272,7 @@ static void poll_health(unsigned long data) if (in_fatal(dev) && !health->sick) { health->sick = true; print_health_info(dev); - queue_work(health->wq, &health->work); + schedule_work(&health->work); } } @@ -301,7 +301,7 @@ void mlx5_health_cleanup(struct mlx5_core_dev *dev) { struct mlx5_core_health *health = &dev->priv.health; - destroy_workqueue(health->wq); + flush_work(&health->work); } int mlx5_health_init(struct mlx5_core_dev *dev) @@ -316,10 +316,7 @@ int mlx5_health_init(struct mlx5_core_dev *dev) strcpy(name, "mlx5_health"); strcat(name, dev_name(&dev->pdev->dev)); - health->wq = create_singlethread_workqueue(name); kfree(name); - if (!health->wq) - return -ENOMEM; INIT_WORK(&health->work, health_care); -- cgit v0.10.2 From 4ac36a4adaf80013a60013d6f829f5863d5d0e05 Mon Sep 17 00:00:00 2001 From: "phil.turnbull@oracle.com" Date: Tue, 26 Jul 2016 15:14:35 -0400 Subject: l2tp: Correctly return -EBADF from pppol2tp_getname. If 'tunnel' is NULL we should return -EBADF but the 'end_put_sess' path unconditionally sets 'error' back to zero. Rework the error path so it more closely matches pppol2tp_sendmsg. Fixes: fd558d186df2 ("l2tp: Split pppol2tp patch into separate l2tp and ppp parts") Signed-off-by: Phil Turnbull Signed-off-by: David S. Miller diff --git a/net/l2tp/l2tp_ppp.c b/net/l2tp/l2tp_ppp.c index 652c250..d9560aa 100644 --- a/net/l2tp/l2tp_ppp.c +++ b/net/l2tp/l2tp_ppp.c @@ -866,10 +866,8 @@ static int pppol2tp_getname(struct socket *sock, struct sockaddr *uaddr, pls = l2tp_session_priv(session); tunnel = l2tp_sock_to_tunnel(pls->tunnel_sock); - if (tunnel == NULL) { - error = -EBADF; + if (tunnel == NULL) goto end_put_sess; - } inet = inet_sk(tunnel->sock); if ((tunnel->version == 2) && (tunnel->sock->sk_family == AF_INET)) { @@ -947,12 +945,11 @@ static int pppol2tp_getname(struct socket *sock, struct sockaddr *uaddr, } *usockaddr_len = len; + error = 0; sock_put(pls->tunnel_sock); end_put_sess: sock_put(sk); - error = 0; - end: return error; } -- cgit v0.10.2 From d3480615cf00c6f615cdb61a9d03386574b93342 Mon Sep 17 00:00:00 2001 From: "Guilherme G. Piccoli" Date: Tue, 26 Jul 2016 17:39:42 -0300 Subject: be2net: perform temperature query in adapter regardless of its interface state The be2net driver performs fw temperature queries on be_worker() routine, which is executed each second for each be_adapter. There is a frequency threshold to avoid fw query to happens at each call to be_worker(); instead, currently a fw query occurs once in 64 runs of the procedure. Nevertheless, this fw temperature query is invoked only for adapters which interface is up, so we can see I/O errors on read of hwmon counters from userspace (from tools like lm-sensors) in case we have adapters' functions which interface is down. This patch moves the fw query code to be invoked even if interface is down. No functional changes were introduced. Signed-off-by: Guilherme G. Piccoli Acked-by: Sathya Perla Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 1f16e73..874c753 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -5077,6 +5077,10 @@ static void be_worker(struct work_struct *work) struct be_rx_obj *rxo; int i; + if (be_physfn(adapter) && + MODULO(adapter->work_counter, adapter->be_get_temp_freq) == 0) + be_cmd_get_die_temperature(adapter); + /* when interrupts are not yet enabled, just reap any pending * mcc completions */ @@ -5095,10 +5099,6 @@ static void be_worker(struct work_struct *work) be_cmd_get_stats(adapter, &adapter->stats_cmd); } - if (be_physfn(adapter) && - MODULO(adapter->work_counter, adapter->be_get_temp_freq) == 0) - be_cmd_get_die_temperature(adapter); - for_all_rx_queues(adapter, rxo, i) { /* Replenish RX-queues starved due to memory * allocation failures. -- cgit v0.10.2 From 36232012344b8db67052432742deaf17f82e70e6 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 26 Jul 2016 23:19:29 -0700 Subject: xgene: Fix build warning with ACPI disabled. drivers/net/ethernet/apm/xgene/xgene_enet_hw.c: In function 'xgene_enet_phy_connect': drivers/net/ethernet/apm/xgene/xgene_enet_hw.c:759:22: warning: unused variable 'adev' [-Wunused-variable] Fixes: 8089a96f601b ("drivers: net: xgene: Add backward compatibility") Reported-by: Stephen Rothwell Signed-off-by: David S. Miller diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c index 8a2a221..7714b7d 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c @@ -756,7 +756,6 @@ int xgene_enet_phy_connect(struct net_device *ndev) struct device_node *np; struct phy_device *phy_dev; struct device *dev = &pdata->pdev->dev; - struct acpi_device *adev; int i; if (dev->of_node) { @@ -781,7 +780,7 @@ int xgene_enet_phy_connect(struct net_device *ndev) pdata->phy_dev = phy_dev; } else { #ifdef CONFIG_ACPI - adev = acpi_phy_find_device(dev); + struct acpi_device *adev = acpi_phy_find_device(dev); if (adev) pdata->phy_dev = adev->driver_data; -- cgit v0.10.2